diff options
1091 files changed, 26070 insertions, 13488 deletions
@@ -121,6 +121,8 @@ Ben Widawsky <bwidawsk@kernel.org> <benjamin.widawsky@intel.com> Benjamin Poirier <benjamin.poirier@gmail.com> <bpoirier@suse.de> Benjamin Tissoires <bentiss@kernel.org> <benjamin.tissoires@gmail.com> Benjamin Tissoires <bentiss@kernel.org> <benjamin.tissoires@redhat.com> +Bingwu Zhang <xtex@aosc.io> <xtexchooser@duck.com> +Bingwu Zhang <xtex@aosc.io> <xtex@xtexx.eu.org> Bjorn Andersson <andersson@kernel.org> <bjorn@kryo.se> Bjorn Andersson <andersson@kernel.org> <bjorn.andersson@linaro.org> Bjorn Andersson <andersson@kernel.org> <bjorn.andersson@sonymobile.com> @@ -200,6 +202,7 @@ Elliot Berman <quic_eberman@quicinc.com> <eberman@codeaurora.org> Enric Balletbo i Serra <eballetbo@kernel.org> <enric.balletbo@collabora.com> Enric Balletbo i Serra <eballetbo@kernel.org> <eballetbo@iseebcn.com> Erik Kaneda <erik.kaneda@intel.com> <erik.schmauss@intel.com> +Ethan Carter Edwards <ethan@ethancedwards.com> Ethan Edwards <ethancarteredwards@gmail.com> Eugen Hristev <eugen.hristev@linaro.org> <eugen.hristev@microchip.com> Eugen Hristev <eugen.hristev@linaro.org> <eugen.hristev@collabora.com> Evgeniy Polyakov <johnpol@2ka.mipt.ru> @@ -435,7 +438,7 @@ Martin Kepplinger <martink@posteo.de> <martin.kepplinger@ginzinger.com> Martin Kepplinger <martink@posteo.de> <martin.kepplinger@puri.sm> Martin Kepplinger <martink@posteo.de> <martin.kepplinger@theobroma-systems.com> Martyna Szapar-Mudlaw <martyna.szapar-mudlaw@linux.intel.com> <martyna.szapar-mudlaw@intel.com> -Mathieu Othacehe <m.othacehe@gmail.com> <othacehe@gnu.org> +Mathieu Othacehe <othacehe@gnu.org> <m.othacehe@gmail.com> Mat Martineau <martineau@kernel.org> <mathew.j.martineau@linux.intel.com> Mat Martineau <martineau@kernel.org> <mathewm@codeaurora.org> Matthew Wilcox <willy@infradead.org> <matthew.r.wilcox@intel.com> @@ -20,6 +20,10 @@ N: Thomas Abraham E: thomas.ab@samsung.com D: Samsung pin controller driver +N: Jose Abreu +E: jose.abreu@synopsys.com +D: Synopsys DesignWare XPCS MDIO/PCS driver. + N: Dragos Acostachioaie E: dragos@iname.com W: http://www.arbornet.org/~dragos @@ -1428,6 +1432,10 @@ S: 8124 Constitution Apt. 7 S: Sterling Heights, Michigan 48313 S: USA +N: Andy Gospodarek +E: andy@greyhouse.net +D: Maintenance and contributions to the network interface bonding driver. + N: Wolfgang Grandegger E: wg@grandegger.com D: Controller Area Network (device drivers) @@ -1812,6 +1820,10 @@ D: Author/maintainer of most DRM drivers (especially ATI, MGA) D: Core DRM templates, general DRM and 3D-related hacking S: No fixed address +N: Woojung Huh +E: woojung.huh@microchip.com +D: Microchip LAN78XX USB Ethernet driver + N: Kenn Humborg E: kenn@wombat.ie D: Mods to loop device to support sparse backing files diff --git a/Documentation/admin-guide/laptops/thinkpad-acpi.rst b/Documentation/admin-guide/laptops/thinkpad-acpi.rst index 7f674a6cfa8a..4ab0fef7d440 100644 --- a/Documentation/admin-guide/laptops/thinkpad-acpi.rst +++ b/Documentation/admin-guide/laptops/thinkpad-acpi.rst @@ -445,8 +445,10 @@ event code Key Notes 0x1008 0x07 FN+F8 IBM: toggle screen expand Lenovo: configure UltraNav, or toggle screen expand. - On newer platforms (2024+) - replaced by 0x131f (see below) + On 2024 platforms replaced by + 0x131f (see below) and on newer + platforms (2025 +) keycode is + replaced by 0x1401 (see below). 0x1009 0x08 FN+F9 - @@ -506,9 +508,11 @@ event code Key Notes 0x1019 0x18 unknown -0x131f ... FN+F8 Platform Mode change. +0x131f ... FN+F8 Platform Mode change (2024 systems). Implemented in driver. +0x1401 ... FN+F8 Platform Mode change (2025 + systems). + Implemented in driver. ... ... ... 0x1020 0x1F unknown diff --git a/Documentation/admin-guide/mm/transhuge.rst b/Documentation/admin-guide/mm/transhuge.rst index 5034915f4e8e..8872203df088 100644 --- a/Documentation/admin-guide/mm/transhuge.rst +++ b/Documentation/admin-guide/mm/transhuge.rst @@ -436,7 +436,7 @@ AnonHugePmdMapped). The number of file transparent huge pages mapped to userspace is available by reading ShmemPmdMapped and ShmemHugePages fields in ``/proc/meminfo``. To identify what applications are mapping file transparent huge pages, it -is necessary to read ``/proc/PID/smaps`` and count the FileHugeMapped fields +is necessary to read ``/proc/PID/smaps`` and count the FilePmdMapped fields for each mapping. Note that reading the smaps file is expensive and reading it diff --git a/Documentation/admin-guide/pm/cpuidle.rst b/Documentation/admin-guide/pm/cpuidle.rst index 19754beb5a4e..eb58d7a5affd 100644 --- a/Documentation/admin-guide/pm/cpuidle.rst +++ b/Documentation/admin-guide/pm/cpuidle.rst @@ -269,27 +269,7 @@ Namely, when invoked to select an idle state for a CPU (i.e. an idle state that the CPU will ask the processor hardware to enter), it attempts to predict the idle duration and uses the predicted value for idle state selection. -It first obtains the time until the closest timer event with the assumption -that the scheduler tick will be stopped. That time, referred to as the *sleep -length* in what follows, is the upper bound on the time before the next CPU -wakeup. It is used to determine the sleep length range, which in turn is needed -to get the sleep length correction factor. - -The ``menu`` governor maintains two arrays of sleep length correction factors. -One of them is used when tasks previously running on the given CPU are waiting -for some I/O operations to complete and the other one is used when that is not -the case. Each array contains several correction factor values that correspond -to different sleep length ranges organized so that each range represented in the -array is approximately 10 times wider than the previous one. - -The correction factor for the given sleep length range (determined before -selecting the idle state for the CPU) is updated after the CPU has been woken -up and the closer the sleep length is to the observed idle duration, the closer -to 1 the correction factor becomes (it must fall between 0 and 1 inclusive). -The sleep length is multiplied by the correction factor for the range that it -falls into to obtain the first approximation of the predicted idle duration. - -Next, the governor uses a simple pattern recognition algorithm to refine its +It first uses a simple pattern recognition algorithm to obtain a preliminary idle duration prediction. Namely, it saves the last 8 observed idle duration values and, when predicting the idle duration next time, it computes the average and variance of them. If the variance is small (smaller than 400 square @@ -301,29 +281,39 @@ Again, if the variance of them is small (in the above sense), the average is taken as the "typical interval" value and so on, until either the "typical interval" is determined or too many data points are disregarded, in which case the "typical interval" is assumed to equal "infinity" (the maximum unsigned -integer value). The "typical interval" computed this way is compared with the -sleep length multiplied by the correction factor and the minimum of the two is -taken as the predicted idle duration. - -Then, the governor computes an extra latency limit to help "interactive" -workloads. It uses the observation that if the exit latency of the selected -idle state is comparable with the predicted idle duration, the total time spent -in that state probably will be very short and the amount of energy to save by -entering it will be relatively small, so likely it is better to avoid the -overhead related to entering that state and exiting it. Thus selecting a -shallower state is likely to be a better option then. The first approximation -of the extra latency limit is the predicted idle duration itself which -additionally is divided by a value depending on the number of tasks that -previously ran on the given CPU and now they are waiting for I/O operations to -complete. The result of that division is compared with the latency limit coming -from the power management quality of service, or `PM QoS <cpu-pm-qos_>`_, -framework and the minimum of the two is taken as the limit for the idle states' -exit latency. +integer value). + +If the "typical interval" computed this way is long enough, the governor obtains +the time until the closest timer event with the assumption that the scheduler +tick will be stopped. That time, referred to as the *sleep length* in what follows, +is the upper bound on the time before the next CPU wakeup. It is used to determine +the sleep length range, which in turn is needed to get the sleep length correction +factor. + +The ``menu`` governor maintains an array containing several correction factor +values that correspond to different sleep length ranges organized so that each +range represented in the array is approximately 10 times wider than the previous +one. + +The correction factor for the given sleep length range (determined before +selecting the idle state for the CPU) is updated after the CPU has been woken +up and the closer the sleep length is to the observed idle duration, the closer +to 1 the correction factor becomes (it must fall between 0 and 1 inclusive). +The sleep length is multiplied by the correction factor for the range that it +falls into to obtain an approximation of the predicted idle duration that is +compared to the "typical interval" determined previously and the minimum of +the two is taken as the idle duration prediction. + +If the "typical interval" value is small, which means that the CPU is likely +to be woken up soon enough, the sleep length computation is skipped as it may +be costly and the idle duration is simply predicted to equal the "typical +interval" value. Now, the governor is ready to walk the list of idle states and choose one of them. For this purpose, it compares the target residency of each state with -the predicted idle duration and the exit latency of it with the computed latency -limit. It selects the state with the target residency closest to the predicted +the predicted idle duration and the exit latency of it with the with the latency +limit coming from the power management quality of service, or `PM QoS <cpu-pm-qos_>`_, +framework. It selects the state with the target residency closest to the predicted idle duration, but still below it, and exit latency that does not exceed the limit. diff --git a/Documentation/devicetree/bindings/display/bridge/adi,adv7533.yaml b/Documentation/devicetree/bindings/display/bridge/adi,adv7533.yaml index df20a3c9c744..ec89115c74e4 100644 --- a/Documentation/devicetree/bindings/display/bridge/adi,adv7533.yaml +++ b/Documentation/devicetree/bindings/display/bridge/adi,adv7533.yaml @@ -90,7 +90,7 @@ properties: adi,dsi-lanes: description: Number of DSI data lanes connected to the DSI host. $ref: /schemas/types.yaml#/definitions/uint32 - enum: [ 1, 2, 3, 4 ] + enum: [ 2, 3, 4 ] "#sound-dai-cells": const: 0 diff --git a/Documentation/devicetree/bindings/display/mediatek/mediatek,dp.yaml b/Documentation/devicetree/bindings/display/mediatek/mediatek,dp.yaml index 2aef1eb32e11..75ce92f4a5fd 100644 --- a/Documentation/devicetree/bindings/display/mediatek/mediatek,dp.yaml +++ b/Documentation/devicetree/bindings/display/mediatek/mediatek,dp.yaml @@ -42,6 +42,9 @@ properties: interrupts: maxItems: 1 + '#sound-dai-cells': + const: 0 + ports: $ref: /schemas/graph.yaml#/properties/ports properties: @@ -85,7 +88,21 @@ required: - ports - max-linkrate-mhz -additionalProperties: false +allOf: + - $ref: /schemas/sound/dai-common.yaml# + - if: + not: + properties: + compatible: + contains: + enum: + - mediatek,mt8188-dp-tx + - mediatek,mt8195-dp-tx + then: + properties: + '#sound-dai-cells': false + +unevaluatedProperties: false examples: - | diff --git a/Documentation/devicetree/bindings/iio/st,st-sensors.yaml b/Documentation/devicetree/bindings/iio/st,st-sensors.yaml index 71c1ee33a393..e955eb8e8797 100644 --- a/Documentation/devicetree/bindings/iio/st,st-sensors.yaml +++ b/Documentation/devicetree/bindings/iio/st,st-sensors.yaml @@ -65,6 +65,7 @@ properties: - st,lsm9ds0-gyro - description: STMicroelectronics Magnetometers enum: + - st,iis2mdc - st,lis2mdl - st,lis3mdl-magn - st,lsm303agr-magn diff --git a/Documentation/devicetree/bindings/net/pse-pd/pse-controller.yaml b/Documentation/devicetree/bindings/net/pse-pd/pse-controller.yaml index a12cda8aa764..cd09560e0aea 100644 --- a/Documentation/devicetree/bindings/net/pse-pd/pse-controller.yaml +++ b/Documentation/devicetree/bindings/net/pse-pd/pse-controller.yaml @@ -81,7 +81,7 @@ properties: List of phandles, each pointing to the power supply for the corresponding pairset named in 'pairset-names'. This property aligns with IEEE 802.3-2022, Section 33.2.3 and 145.2.4. - PSE Pinout Alternatives (as per IEEE 802.3-2022 Table 145\u20133) + PSE Pinout Alternatives (as per IEEE 802.3-2022 Table 145-3) |-----------|---------------|---------------|---------------|---------------| | Conductor | Alternative A | Alternative A | Alternative B | Alternative B | | | (MDI-X) | (MDI) | (X) | (S) | diff --git a/Documentation/devicetree/bindings/sound/realtek,rt5645.yaml b/Documentation/devicetree/bindings/sound/realtek,rt5645.yaml index 13f09f1bc800..0a698798c22b 100644 --- a/Documentation/devicetree/bindings/sound/realtek,rt5645.yaml +++ b/Documentation/devicetree/bindings/sound/realtek,rt5645.yaml @@ -51,7 +51,7 @@ properties: description: Power supply for AVDD, providing 1.8V. cpvdd-supply: - description: Power supply for CPVDD, providing 3.5V. + description: Power supply for CPVDD, providing 1.8V. hp-detect-gpios: description: diff --git a/Documentation/filesystems/bcachefs/CodingStyle.rst b/Documentation/filesystems/bcachefs/CodingStyle.rst index 01de555e21d8..b29562a6bf55 100644 --- a/Documentation/filesystems/bcachefs/CodingStyle.rst +++ b/Documentation/filesystems/bcachefs/CodingStyle.rst @@ -183,4 +183,4 @@ even better as a code comment. A good code comment is wonderful, but even better is the comment that didn't need to exist because the code was so straightforward as to be obvious; organized into small clean and tidy modules, with clear and descriptive names -for functions and variable, where every line of code has a clear purpose. +for functions and variables, where every line of code has a clear purpose. diff --git a/Documentation/filesystems/fiemap.rst b/Documentation/filesystems/fiemap.rst index 93fc96f760aa..23b3ed229e49 100644 --- a/Documentation/filesystems/fiemap.rst +++ b/Documentation/filesystems/fiemap.rst @@ -12,21 +12,10 @@ returns a list of extents. Request Basics -------------- -A fiemap request is encoded within struct fiemap:: - - struct fiemap { - __u64 fm_start; /* logical offset (inclusive) at - * which to start mapping (in) */ - __u64 fm_length; /* logical length of mapping which - * userspace cares about (in) */ - __u32 fm_flags; /* FIEMAP_FLAG_* flags for request (in/out) */ - __u32 fm_mapped_extents; /* number of extents that were - * mapped (out) */ - __u32 fm_extent_count; /* size of fm_extents array (in) */ - __u32 fm_reserved; - struct fiemap_extent fm_extents[0]; /* array of mapped extents (out) */ - }; +A fiemap request is encoded within struct fiemap: +.. kernel-doc:: include/uapi/linux/fiemap.h + :identifiers: fiemap fm_start, and fm_length specify the logical range within the file which the process would like mappings for. Extents returned mirror @@ -60,6 +49,8 @@ FIEMAP_FLAG_XATTR If this flag is set, the extents returned will describe the inodes extended attribute lookup tree, instead of its data tree. +FIEMAP_FLAG_CACHE + This flag requests caching of the extents. Extent Mapping -------------- @@ -77,18 +68,10 @@ complete the requested range and will not have the FIEMAP_EXTENT_LAST flag set (see the next section on extent flags). Each extent is described by a single fiemap_extent structure as -returned in fm_extents:: - - struct fiemap_extent { - __u64 fe_logical; /* logical offset in bytes for the start of - * the extent */ - __u64 fe_physical; /* physical offset in bytes for the start - * of the extent */ - __u64 fe_length; /* length in bytes for the extent */ - __u64 fe_reserved64[2]; - __u32 fe_flags; /* FIEMAP_EXTENT_* flags for this extent */ - __u32 fe_reserved[3]; - }; +returned in fm_extents: + +.. kernel-doc:: include/uapi/linux/fiemap.h + :identifiers: fiemap_extent All offsets and lengths are in bytes and mirror those on disk. It is valid for an extents logical offset to start before the request or its logical @@ -175,6 +158,8 @@ FIEMAP_EXTENT_MERGED userspace would be highly inefficient, the kernel will try to merge most adjacent blocks into 'extents'. +FIEMAP_EXTENT_SHARED + This flag is set to request that space be shared with other files. VFS -> File System Implementation --------------------------------- @@ -191,14 +176,10 @@ each discovered extent:: u64 len); ->fiemap is passed struct fiemap_extent_info which describes the -fiemap request:: - - struct fiemap_extent_info { - unsigned int fi_flags; /* Flags as passed from user */ - unsigned int fi_extents_mapped; /* Number of mapped extents */ - unsigned int fi_extents_max; /* Size of fiemap_extent array */ - struct fiemap_extent *fi_extents_start; /* Start of fiemap_extent array */ - }; +fiemap request: + +.. kernel-doc:: include/linux/fiemap.h + :identifiers: fiemap_extent_info It is intended that the file system should not need to access any of this structure directly. Filesystem handlers should be tolerant to signals and return diff --git a/Documentation/netlink/specs/mptcp_pm.yaml b/Documentation/netlink/specs/mptcp_pm.yaml index dc190bf838fe..dfd017780d2f 100644 --- a/Documentation/netlink/specs/mptcp_pm.yaml +++ b/Documentation/netlink/specs/mptcp_pm.yaml @@ -22,65 +22,67 @@ definitions: doc: unused event - name: created - doc: - token, family, saddr4 | saddr6, daddr4 | daddr6, sport, dport + doc: >- A new MPTCP connection has been created. It is the good time to allocate memory and send ADD_ADDR if needed. Depending on the traffic-patterns it can take a long time until the MPTCP_EVENT_ESTABLISHED is sent. + Attributes: token, family, saddr4 | saddr6, daddr4 | daddr6, sport, + dport, server-side. - name: established - doc: - token, family, saddr4 | saddr6, daddr4 | daddr6, sport, dport + doc: >- A MPTCP connection is established (can start new subflows). + Attributes: token, family, saddr4 | saddr6, daddr4 | daddr6, sport, + dport, server-side. - name: closed - doc: - token + doc: >- A MPTCP connection has stopped. + Attribute: token. - name: announced value: 6 - doc: - token, rem_id, family, daddr4 | daddr6 [, dport] + doc: >- A new address has been announced by the peer. + Attributes: token, rem_id, family, daddr4 | daddr6 [, dport]. - name: removed - doc: - token, rem_id + doc: >- An address has been lost by the peer. + Attributes: token, rem_id. - name: sub-established value: 10 - doc: - token, family, loc_id, rem_id, saddr4 | saddr6, daddr4 | daddr6, sport, - dport, backup, if_idx [, error] + doc: >- A new subflow has been established. 'error' should not be set. + Attributes: token, family, loc_id, rem_id, saddr4 | saddr6, daddr4 | + daddr6, sport, dport, backup, if_idx [, error]. - name: sub-closed - doc: - token, family, loc_id, rem_id, saddr4 | saddr6, daddr4 | daddr6, sport, - dport, backup, if_idx [, error] + doc: >- A subflow has been closed. An error (copy of sk_err) could be set if an error has been detected for this subflow. + Attributes: token, family, loc_id, rem_id, saddr4 | saddr6, daddr4 | + daddr6, sport, dport, backup, if_idx [, error]. - name: sub-priority value: 13 - doc: - token, family, loc_id, rem_id, saddr4 | saddr6, daddr4 | daddr6, sport, - dport, backup, if_idx [, error] + doc: >- The priority of a subflow has changed. 'error' should not be set. + Attributes: token, family, loc_id, rem_id, saddr4 | saddr6, daddr4 | + daddr6, sport, dport, backup, if_idx [, error]. - name: listener-created value: 15 - doc: - family, sport, saddr4 | saddr6 + doc: >- A new PM listener is created. + Attributes: family, sport, saddr4 | saddr6. - name: listener-closed - doc: - family, sport, saddr4 | saddr6 + doc: >- A PM listener is closed. + Attributes: family, sport, saddr4 | saddr6. attribute-sets: - @@ -306,8 +308,8 @@ operations: attributes: - addr - - name: flush-addrs - doc: flush addresses + name: flush-addrs + doc: Flush addresses attribute-set: endpoint dont-validate: [ strict ] flags: [ uns-admin-perm ] @@ -351,7 +353,7 @@ operations: - addr-remote - name: announce - doc: announce new sf + doc: Announce new address attribute-set: attr dont-validate: [ strict ] flags: [ uns-admin-perm ] @@ -362,7 +364,7 @@ operations: - token - name: remove - doc: announce removal + doc: Announce removal attribute-set: attr dont-validate: [ strict ] flags: [ uns-admin-perm ] @@ -373,7 +375,7 @@ operations: - loc-id - name: subflow-create - doc: todo + doc: Create subflow attribute-set: attr dont-validate: [ strict ] flags: [ uns-admin-perm ] @@ -385,7 +387,7 @@ operations: - addr-remote - name: subflow-destroy - doc: todo + doc: Destroy subflow attribute-set: attr dont-validate: [ strict ] flags: [ uns-admin-perm ] diff --git a/Documentation/security/credentials.rst b/Documentation/security/credentials.rst index 357328d566c8..2aa0791bcefe 100644 --- a/Documentation/security/credentials.rst +++ b/Documentation/security/credentials.rst @@ -527,11 +527,6 @@ There are some functions to help manage credentials: This gets a reference on a live set of credentials, returning a pointer to that set of credentials. - - ``struct cred *get_new_cred(struct cred *cred);`` - - This gets a reference on a set of credentials that is under construction - and is thus still mutable, returning a pointer to that set of credentials. - Open File Credentials ===================== diff --git a/Documentation/sound/codecs/cs35l56.rst b/Documentation/sound/codecs/cs35l56.rst new file mode 100644 index 000000000000..98c6f6c74394 --- /dev/null +++ b/Documentation/sound/codecs/cs35l56.rst @@ -0,0 +1,292 @@ +.. SPDX-License-Identifier: GPL-2.0-only + +===================================================================== +Audio drivers for Cirrus Logic CS35L54/56/57 Boosted Smart Amplifiers +===================================================================== +:Copyright: 2025 Cirrus Logic, Inc. and + Cirrus Logic International Semiconductor Ltd. + +Contact: patches@opensource.cirrus.com + +Summary +======= + +The high-level summary of this document is: + +**If you have a laptop that uses CS35L54/56/57 amplifiers but audio is not +working, DO NOT ATTEMPT TO USE FIRMWARE AND SETTINGS FROM ANOTHER LAPTOP, +EVEN IF THAT LAPTOP SEEMS SIMILAR.** + +The CS35L54/56/57 amplifiers must be correctly configured for the power +supply voltage, speaker impedance, maximum speaker voltage/current, and +other external hardware connections. + +The amplifiers feature advanced boost technology that increases the voltage +used to drive the speakers, while proprietary speaker protection algorithms +allow these boosted amplifiers to push the limits of the speakers without +causing damage. These **must** be configured correctly. + +Supported Cirrus Logic amplifiers +--------------------------------- + +The cs35l56 drivers support: + +* CS35L54 +* CS35L56 +* CS35L57 + +There are two drivers in the kernel + +*For systems using SoundWire*: sound/soc/codecs/cs35l56.c and associated files + +*For systems using HDA*: sound/pci/hda/cs35l56_hda.c + +Firmware +======== + +The amplifier is controlled and managed by firmware running on the internal +DSP. Firmware files are essential to enable the full capabilities of the +amplifier. + +Firmware is distributed in the linux-firmware repository: +https://gitlab.com/kernel-firmware/linux-firmware.git + +On most SoundWire systems the amplifier has a default minimum capability to +produce audio. However this will be + +* at low volume, to protect the speakers, since the speaker specifications + and power supply voltages are unknown. +* a mono mix of left and right channels. + +On some SoundWire systems that have both CS42L43 and CS35L56/57 the CS35L56/57 +receive their audio from the CS42L43 instead of directly from the host +SoundWire interface. These systems can be identified by the CS42L43 showing +in dmesg as a SoundWire device, but the CS35L56/57 as SPI. On these systems +the firmware is *mandatory* to enable receiving the audio from the CS42L43. + +On HDA systems the firmware is *mandatory* to enable HDA bridge mode. There +will not be any audio from the amplifiers without firmware. + +Cirrus Logic firmware files +--------------------------- + +Each amplifier requires two firmware files. One file has a .wmfw suffix, the +other has a .bin suffix. + +The firmware is customized by the OEM to match the hardware of each laptop, +and the firmware is specific to that laptop. Because of this, there are many +firmware files in linux-firmware for these amplifiers. Firmware files are +**not interchangeable between laptops**. + +Cirrus Logic submits files for known laptops to the upstream linux-firmware +repository. Providing Cirrus Logic is aware of a particular laptop and has +permission from the manufacturer to publish the firmware, it will be pushed +to linux-firmware. You may need to upgrade to a newer release of +linux-firmware to obtain the firmware for your laptop. + +**Important:** the Makefile for linux-firmware creates symlinks that are listed +in the WHENCE file. These symlinks are required for the CS35L56 driver to be +able to load the firmware. + +How do I know which firmware file I should have? +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ +All firmware file names are qualified with a unique "system ID". On normal +x86 PCs with PCI audio this is the Vendor Subsystem ID (SSID) of the host +PCI audio interface. + +The SSID can be viewed using the lspci tool:: + + lspci -v -nn | grep -A2 -i audio + 0000:00:1f.3 Audio device [0403]: Intel Corporation Meteor Lake-P HD Audio Controller [8086:7e28] + Subsystem: Dell Meteor Lake-P HD Audio Controller [1028:0c63] + +In this example the SSID is 10280c63. + +The format of the firmware file names is: + + cs35lxx-b0-dsp1-misc-SSID[-spkidX]-ampN + +Where: + + * cs35lxx-b0 is the amplifier model and silicon revision. This information + is logged by the driver during initialization. + * SSID is the 8-digit hexadecimal SSID value. + * ampN is the amplifier number (for example amp1). This is the same as + the prefix on the ALSA control names except that it is always lower-case + in the file name. + * spkidX is an optional part, used for laptops that have firmware + configurations for different makes and models of internal speakers. + +Sound Open Firmware and ALSA topology files +------------------------------------------- + +All SoundWire systems will require a Sound Open Firmware (SOF) for the +host CPU audio DSP, together with an ALSA topology file (.tplg). + +The SOF firmware will usually be provided by the manufacturer of the host +CPU (i.e. Intel or AMD). The .tplg file is normally part of the SOF firmware +release. + +SOF binary builds are available from: https://github.com/thesofproject/sof-bin/releases + +The main SOF source is here: https://github.com/thesofproject + +ALSA-ucm configurations +----------------------- +Typically an appropriate ALSA-ucm configuration file is needed for +use-case managers and audio servers such as PipeWire. + +Configuration files are available from the alsa-ucm-conf repository: +https://git.alsa-project.org/?p=alsa-ucm-conf.git + +Kernel log messages +=================== + +SoundWire +--------- +A successful initialization will look like this (this will be repeated for +each amplifier):: + + [ 7.568374] cs35l56 sdw:0:0:01fa:3556:01:0: supply VDD_P not found, using dummy regulator + [ 7.605208] cs35l56 sdw:0:0:01fa:3556:01:0: supply VDD_IO not found, using dummy regulator + [ 7.605313] cs35l56 sdw:0:0:01fa:3556:01:0: supply VDD_A not found, using dummy regulator + [ 7.939279] cs35l56 sdw:0:0:01fa:3556:01:0: Cirrus Logic CS35L56 Rev B0 OTP3 fw:3.4.4 (patched=0) + [ 7.947844] cs35l56 sdw:0:0:01fa:3556:01:0: Slave 4 state check1: UNATTACHED, status was 1 + [ 8.740280] cs35l56 sdw:0:0:01fa:3556:01:0: supply VDD_B not found, using dummy regulator + [ 8.740552] cs35l56 sdw:0:0:01fa:3556:01:0: supply VDD_AMP not found, using dummy regulator + [ 9.242164] cs35l56 sdw:0:0:01fa:3556:01:0: DSP1: cirrus/cs35l56-b0-dsp1-misc-xxxxxxxx.wmfw: format 3 timestamp 0x66b2b872 + [ 9.242173] cs35l56 sdw:0:0:01fa:3556:01:0: DSP1: cirrus/cs35l56-b0-dsp1-misc-xxxxxxxx.wmfw: Tue 05 Dec 2023 21:37:21 GMT Standard Time + [ 9.991709] cs35l56 sdw:0:0:01fa:3556:01:0: DSP1: Firmware: 1a00d6 vendor: 0x2 v3.11.23, 41 algorithms + [10.039098] cs35l56 sdw:0:0:01fa:3556:01:0: DSP1: cirrus/cs35l56-b0-dsp1-misc-xxxxxxxx-amp1.bin: v3.11.23 + [10.879235] cs35l56 sdw:0:0:01fa:3556:01:0: Slave 4 state check1: UNATTACHED, status was 1 + [11.401536] cs35l56 sdw:0:0:01fa:3556:01:0: Calibration applied + +HDA +--- +A successful initialization will look like this (this will be repeated for +each amplifier):: + + [ 6.306475] cs35l56-hda i2c-CSC3556:00-cs35l56-hda.0: Cirrus Logic CS35L56 Rev B0 OTP3 fw:3.4.4 (patched=0) + [ 6.613892] cs35l56-hda i2c-CSC3556:00-cs35l56-hda.0: DSP system name: 'xxxxxxxx', amp name: 'AMP1' + [ 8.266660] snd_hda_codec_cs8409 ehdaudio0D0: bound i2c-CSC3556:00-cs35l56-hda.0 (ops cs35l56_hda_comp_ops [snd_hda_scodec_cs35l56]) + [ 8.287525] cs35l56-hda i2c-CSC3556:00-cs35l56-hda.0: DSP1: cirrus/cs35l56-b0-dsp1-misc-xxxxxxxx.wmfw: format 3 timestamp 0x66b2b872 + [ 8.287528] cs35l56-hda i2c-CSC3556:00-cs35l56-hda.0: DSP1: cirrus/cs35l56-b0-dsp1-misc-xxxxxxxx.wmfw: Tue 05 Dec 2023 21:37:21 GMT Standard Time + [ 9.984335] cs35l56-hda i2c-CSC3556:00-cs35l56-hda.0: DSP1: Firmware: 1a00d6 vendor: 0x2 v3.11.23, 41 algorithms + [10.085797] cs35l56-hda i2c-CSC3556:00-cs35l56-hda.0: DSP1: cirrus/cs35l56-b0-dsp1-misc-xxxxxxxx-amp1.bin: v3.11.23 + [10.655237] cs35l56-hda i2c-CSC3556:00-cs35l56-hda.0: Calibration applied + +Important messages +~~~~~~~~~~~~~~~~~~ +Cirrus Logic CS35L56 Rev B0 OTP3 fw:3.4.4 (patched=0) + Shows that the driver has been able to read device ID registers from the + amplifier. + + * The actual amplifier type and silicon revision (CS35L56 B0 in this + example) is shown, as read from the amplifier identification registers. + * (patched=0) is normal, and indicates that the amplifier has been hard + reset and is running default ROM firmware. + * (patched=1) means that something has previously downloaded firmware + to the amplifier and the driver does not have control of the RESET + signal to be able to replace this preloaded firmware. This is normal + for systems where the BIOS downloads firmware to the amplifiers + before OS boot. + This status can also be seen if the cs35l56 kernel module is unloaded + and reloaded on a system where the driver does not have control of + RESET. SoundWire systems typically do not give the driver control of + RESET and only a BIOS (re)boot can reset the amplifiers. + +DSP1: cirrus/cs35l56-b0-dsp1-misc-xxxxxxxx.wmfw + Shows that a .wmfw firmware file was found and downloaded. + +DSP1: cirrus/cs35l56-b0-dsp1-misc-xxxxxxxx-amp1.bin + Shows that a .bin firmware file was found and downloaded. + +Calibration applied + Factory calibration data in EFI was written to the amplifier. + +Error messages +============== +This section explains some of the error messages that the driver can log. + +Algorithm coefficient version %d.%d.%d but expected %d.%d.%d + The version of the .bin file content does not match the loaded firmware. + Caused by mismatched .wmfw and .bin file, or .bin file was found but + .wmfw was not. + +No %s for algorithm %x + The version of the .bin file content does not match the loaded firmware. + Caused by mismatched .wmfw and .bin file, or .bin file was found but + .wmfw was not. + +.bin file required but not found + HDA driver did not find a .bin file that matches this hardware. + +Calibration disabled due to missing firmware controls + Driver was not able to write EFI calibration data to firmware registers. + This typically means that either: + + * The driver did not find a suitable wmfw for this hardware, or + * The amplifier has already been patched with firmware by something + previously, and the driver does not have control of a hard RESET line + to be able to reset the amplifier and download the firmware files it + found. This situation is indicated by the device identification + string in the kernel log shows "(patched=1)" + +Failed to write calibration + Same meaning and cause as "Calibration disabled due to missing firmware + controls" + +Failed to read calibration data from EFI + Factory calibration data in EFI is missing, empty or corrupt. + This is most likely to be cause by accidentally deleting the file from + the EFI filesystem. + +No calibration for silicon ID + The factory calibration data in EFI does not match this hardware. + The most likely cause is that an amplifier has been replaced on the + motherboard without going through manufacturer calibration process to + generate calibration data for the new amplifier. + +Did not find any buses for CSCxxxx + Only on HDA systems. The HDA codec driver found an ACPI entry for + Cirrus Logic companion amps, but could not enumerate the ACPI entries for + the I2C/SPI buses. The most likely cause of this is that: + + * The relevant bus driver (I2C or SPI) is not part of the kernel. + * The HDA codec driver was built-in to the kernel but the I2C/SPI + bus driver is a module and so the HDA codec driver cannot call the + bus driver functions. + +init_completion timed out + The SoundWire bus controller (host end) did not enumerate the amplifier. + In other words, the ACPI says there is an amplifier but for some reason + it was not detected on the bus. + +No AF01 node + Indicates an error in ACPI. A SoundWire system should have a Device() + node named "AF01" but it was not found. + +Failed to get spk-id-gpios + ACPI says that the driver should request a GPIO but the driver was not + able to get that GPIO. The most likely cause is that the kernel does not + include the correct GPIO or PINCTRL driver for this system. + +Failed to read spk-id + ACPI says that the driver should request a GPIO but the driver was not + able to read that GPIO. + +Unexpected spk-id element count + AF01 contains more speaker ID GPIO entries than the driver supports + +Overtemp error + Amplifier overheat protection was triggered and the amplifier shut down + to protect itself. + +Amp short error + Amplifier detected a short-circuit on the speaker output pins and shut + down for protection. This would normally indicate a damaged speaker. + +Hibernate wake failed + The driver tried to wake the amplifier from its power-saving state but + did not see the expected responses from the amplifier. This can be caused + by using firmware that does not match the hardware. diff --git a/Documentation/sound/codecs/index.rst b/Documentation/sound/codecs/index.rst new file mode 100644 index 000000000000..2cb95d87bbef --- /dev/null +++ b/Documentation/sound/codecs/index.rst @@ -0,0 +1,9 @@ +.. SPDX-License-Identifier: GPL-2.0 + +Codec-Specific Information +========================== + +.. toctree:: + :maxdepth: 2 + + cs35l56 diff --git a/Documentation/sound/index.rst b/Documentation/sound/index.rst index c437f2a4bc85..51cd736f65b5 100644 --- a/Documentation/sound/index.rst +++ b/Documentation/sound/index.rst @@ -13,6 +13,7 @@ Sound Subsystem Documentation alsa-configuration hd-audio/index cards/index + codecs/index utimers .. only:: subproject and html diff --git a/Documentation/trace/ftrace.rst b/Documentation/trace/ftrace.rst index 272464bb7c60..2b74f96d09d5 100644 --- a/Documentation/trace/ftrace.rst +++ b/Documentation/trace/ftrace.rst @@ -810,6 +810,12 @@ Here is the list of current tracers that may be configured. to draw a graph of function calls similar to C code source. + Note that the function graph calculates the timings of when the + function starts and returns internally and for each instance. If + there are two instances that run function graph tracer and traces + the same functions, the length of the timings may be slightly off as + each read the timestamp separately and not at the same time. + "blk" The block tracer. The tracer used by the blktrace user diff --git a/Documentation/virt/kvm/api.rst b/Documentation/virt/kvm/api.rst index 454c2aaa155e..f15b61317aad 100644 --- a/Documentation/virt/kvm/api.rst +++ b/Documentation/virt/kvm/api.rst @@ -1914,6 +1914,9 @@ No flags are specified so far, the corresponding field must be set to zero. #define KVM_IRQ_ROUTING_HV_SINT 4 #define KVM_IRQ_ROUTING_XEN_EVTCHN 5 +On s390, adding a KVM_IRQ_ROUTING_S390_ADAPTER is rejected on ucontrol VMs with +error -EINVAL. + flags: - KVM_MSI_VALID_DEVID: used along with KVM_IRQ_ROUTING_MSI routing entry diff --git a/Documentation/virt/kvm/devices/s390_flic.rst b/Documentation/virt/kvm/devices/s390_flic.rst index ea96559ba501..b784f8016748 100644 --- a/Documentation/virt/kvm/devices/s390_flic.rst +++ b/Documentation/virt/kvm/devices/s390_flic.rst @@ -58,11 +58,15 @@ Groups: Enables async page faults for the guest. So in case of a major page fault the host is allowed to handle this async and continues the guest. + -EINVAL is returned when called on the FLIC of a ucontrol VM. + KVM_DEV_FLIC_APF_DISABLE_WAIT Disables async page faults for the guest and waits until already pending async page faults are done. This is necessary to trigger a completion interrupt for every init interrupt before migrating the interrupt list. + -EINVAL is returned when called on the FLIC of a ucontrol VM. + KVM_DEV_FLIC_ADAPTER_REGISTER Register an I/O adapter interrupt source. Takes a kvm_s390_io_adapter describing the adapter to register:: diff --git a/MAINTAINERS b/MAINTAINERS index 910305c11e8a..b3d202e3a090 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -949,7 +949,6 @@ AMAZON ETHERNET DRIVERS M: Shay Agroskin <shayagr@amazon.com> M: Arthur Kiyanovski <akiyano@amazon.com> R: David Arinzon <darinzon@amazon.com> -R: Noam Dagan <ndagan@amazon.com> R: Saeed Bishara <saeedb@amazon.com> L: netdev@vger.kernel.org S: Supported @@ -1797,7 +1796,6 @@ F: include/uapi/linux/if_arcnet.h ARM AND ARM64 SoC SUB-ARCHITECTURES (COMMON PARTS) M: Arnd Bergmann <arnd@arndb.de> -M: Olof Johansson <olof@lixom.net> L: linux-arm-kernel@lists.infradead.org (moderated for non-subscribers) L: soc@lists.linux.dev S: Maintained @@ -2691,7 +2689,6 @@ N: at91 N: atmel ARM/Microchip Sparx5 SoC support -M: Lars Povlsen <lars.povlsen@microchip.com> M: Steen Hegelund <Steen.Hegelund@microchip.com> M: Daniel Machon <daniel.machon@microchip.com> M: UNGLinuxDriver@microchip.com @@ -3608,6 +3605,7 @@ F: drivers/phy/qualcomm/phy-ath79-usb.c ATHEROS ATH GENERIC UTILITIES M: Kalle Valo <kvalo@kernel.org> +M: Jeff Johnson <jjohnson@kernel.org> L: linux-wireless@vger.kernel.org S: Supported F: drivers/net/wireless/ath/* @@ -4058,7 +4056,6 @@ F: net/bluetooth/ BONDING DRIVER M: Jay Vosburgh <jv@jvosburgh.net> -M: Andy Gospodarek <andy@greyhouse.net> L: netdev@vger.kernel.org S: Maintained F: Documentation/networking/bonding.rst @@ -4131,7 +4128,6 @@ S: Odd Fixes F: drivers/net/ethernet/netronome/nfp/bpf/ BPF JIT for POWERPC (32-BIT AND 64-BIT) -M: Michael Ellerman <mpe@ellerman.id.au> M: Hari Bathini <hbathini@linux.ibm.com> M: Christophe Leroy <christophe.leroy@csgroup.eu> R: Naveen N Rao <naveen@kernel.org> @@ -5469,6 +5465,7 @@ L: linux-sound@vger.kernel.org L: patches@opensource.cirrus.com S: Maintained F: Documentation/devicetree/bindings/sound/cirrus,cs* +F: Documentation/sound/codecs/cs* F: drivers/mfd/cs42l43* F: drivers/pinctrl/cirrus/pinctrl-cs42l43* F: drivers/spi/spi-cs42l43* @@ -8551,8 +8548,8 @@ F: rust/kernel/net/phy.rs F: rust/kernel/net/phy/reg.rs EXEC & BINFMT API, ELF +M: Kees Cook <kees@kernel.org> R: Eric Biederman <ebiederm@xmission.com> -R: Kees Cook <kees@kernel.org> L: linux-mm@kvack.org S: Supported T: git git://git.kernel.org/pub/scm/linux/kernel/git/kees/linux.git for-next/execve @@ -8564,6 +8561,7 @@ F: fs/tests/binfmt_*_kunit.c F: fs/tests/exec_kunit.c F: include/linux/binfmts.h F: include/linux/elf.h +F: include/uapi/linux/auxvec.h F: include/uapi/linux/binfmts.h F: include/uapi/linux/elf.h F: tools/testing/selftests/exec/ @@ -12387,6 +12385,13 @@ F: Documentation/kbuild/kconfig* F: scripts/Kconfig.include F: scripts/kconfig/ +KCORE +M: Omar Sandoval <osandov@osandov.com> +L: linux-debuggers@vger.kernel.org +S: Maintained +F: fs/proc/kcore.c +F: include/linux/kcore.h + KCOV R: Dmitry Vyukov <dvyukov@google.com> R: Andrey Konovalov <andreyknvl@gmail.com> @@ -12632,7 +12637,7 @@ F: arch/mips/include/uapi/asm/kvm* F: arch/mips/kvm/ KERNEL VIRTUAL MACHINE FOR POWERPC (KVM/powerpc) -M: Michael Ellerman <mpe@ellerman.id.au> +M: Madhavan Srinivasan <maddy@linux.ibm.com> R: Nicholas Piggin <npiggin@gmail.com> L: linuxppc-dev@lists.ozlabs.org L: kvm@vger.kernel.org @@ -13211,11 +13216,11 @@ X: drivers/macintosh/adb-iop.c X: drivers/macintosh/via-macii.c LINUX FOR POWERPC (32-BIT AND 64-BIT) +M: Madhavan Srinivasan <maddy@linux.ibm.com> M: Michael Ellerman <mpe@ellerman.id.au> R: Nicholas Piggin <npiggin@gmail.com> R: Christophe Leroy <christophe.leroy@csgroup.eu> R: Naveen N Rao <naveen@kernel.org> -M: Madhavan Srinivasan <maddy@linux.ibm.com> L: linuxppc-dev@lists.ozlabs.org S: Supported W: https://github.com/linuxppc/wiki/wiki @@ -14566,7 +14571,6 @@ F: drivers/dma/mediatek/ MEDIATEK ETHERNET DRIVER M: Felix Fietkau <nbd@nbd.name> M: Sean Wang <sean.wang@mediatek.com> -M: Mark Lee <Mark-MC.Lee@mediatek.com> M: Lorenzo Bianconi <lorenzo@kernel.org> L: netdev@vger.kernel.org S: Maintained @@ -14756,7 +14760,7 @@ F: drivers/memory/mtk-smi.c F: include/soc/mediatek/smi.h MEDIATEK SWITCH DRIVER -M: Arınç ÃœNAL <arinc.unal@arinc9.com> +M: Chester A. Unal <chester.a.unal@arinc9.com> M: Daniel Golle <daniel@makrotopia.org> M: DENG Qingfang <dqfext@gmail.com> M: Sean Wang <sean.wang@mediatek.com> @@ -18460,7 +18464,7 @@ F: Documentation/devicetree/bindings/pinctrl/mediatek,mt8183-pinctrl.yaml F: drivers/pinctrl/mediatek/ PIN CONTROLLER - MEDIATEK MIPS -M: Arınç ÃœNAL <arinc.unal@arinc9.com> +M: Chester A. Unal <chester.a.unal@arinc9.com> M: Sergio Paracuellos <sergio.paracuellos@gmail.com> L: linux-mediatek@lists.infradead.org (moderated for non-subscribers) L: linux-mips@vger.kernel.org @@ -19504,7 +19508,7 @@ S: Maintained F: arch/mips/ralink RALINK MT7621 MIPS ARCHITECTURE -M: Arınç ÃœNAL <arinc.unal@arinc9.com> +M: Chester A. Unal <chester.a.unal@arinc9.com> M: Sergio Paracuellos <sergio.paracuellos@gmail.com> L: linux-mips@vger.kernel.org S: Maintained @@ -20907,6 +20911,8 @@ F: kernel/sched/ SCHEDULER - SCHED_EXT R: Tejun Heo <tj@kernel.org> R: David Vernet <void@manifault.com> +R: Andrea Righi <arighi@nvidia.com> +R: Changwoo Min <changwoo@igalia.com> L: linux-kernel@vger.kernel.org S: Maintained W: https://github.com/sched-ext/scx @@ -21991,6 +21997,7 @@ W: https://github.com/thesofproject/linux/ F: sound/soc/sof/ SOUND - GENERIC SOUND CARD (Simple-Audio-Card, Audio-Graph-Card) +M: Mark Brown <broonie@kernel.org> M: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> S: Supported L: linux-sound@vger.kernel.org @@ -22501,11 +22508,8 @@ F: Documentation/devicetree/bindings/phy/st,stm32mp25-combophy.yaml F: drivers/phy/st/phy-stm32-combophy.c STMMAC ETHERNET DRIVER -M: Alexandre Torgue <alexandre.torgue@foss.st.com> -M: Jose Abreu <joabreu@synopsys.com> L: netdev@vger.kernel.org -S: Supported -W: http://www.stlinux.com +S: Orphan F: Documentation/networking/device_drivers/ethernet/stmicro/ F: drivers/net/ethernet/stmicro/stmmac/ @@ -22737,9 +22741,8 @@ S: Supported F: drivers/net/ethernet/synopsys/ SYNOPSYS DESIGNWARE ETHERNET XPCS DRIVER -M: Jose Abreu <Jose.Abreu@synopsys.com> L: netdev@vger.kernel.org -S: Supported +S: Orphan F: drivers/net/pcs/pcs-xpcs.c F: drivers/net/pcs/pcs-xpcs.h F: include/linux/pcs/pcs-xpcs.h @@ -23647,7 +23650,6 @@ F: tools/testing/selftests/timers/ TIPC NETWORK LAYER M: Jon Maloy <jmaloy@redhat.com> -M: Ying Xue <ying.xue@windriver.com> L: netdev@vger.kernel.org (core kernel code) L: tipc-discussion@lists.sourceforge.net (user apps, general discussion) S: Maintained @@ -24253,7 +24255,8 @@ F: Documentation/devicetree/bindings/usb/nxp,isp1760.yaml F: drivers/usb/isp1760/* USB LAN78XX ETHERNET DRIVER -M: Woojung Huh <woojung.huh@microchip.com> +M: Thangaraj Samynathan <Thangaraj.S@microchip.com> +M: Rengarajan Sundararajan <Rengarajan.S@microchip.com> M: UNGLinuxDriver@microchip.com L: netdev@vger.kernel.org S: Maintained @@ -2,7 +2,7 @@ VERSION = 6 PATCHLEVEL = 13 SUBLEVEL = 0 -EXTRAVERSION = -rc4 +EXTRAVERSION = NAME = Baby Opossum Posse # *DOCUMENTATION* diff --git a/arch/arm/boot/dts/nxp/imx/imxrt1050.dtsi b/arch/arm/boot/dts/nxp/imx/imxrt1050.dtsi index dd714d235d5f..b0bad0d1ba36 100644 --- a/arch/arm/boot/dts/nxp/imx/imxrt1050.dtsi +++ b/arch/arm/boot/dts/nxp/imx/imxrt1050.dtsi @@ -87,7 +87,7 @@ reg = <0x402c0000 0x4000>; interrupts = <110>; clocks = <&clks IMXRT1050_CLK_IPG_PDOF>, - <&clks IMXRT1050_CLK_OSC>, + <&clks IMXRT1050_CLK_AHB_PODF>, <&clks IMXRT1050_CLK_USDHC1>; clock-names = "ipg", "ahb", "per"; bus-width = <4>; diff --git a/arch/arm/configs/imx_v6_v7_defconfig b/arch/arm/configs/imx_v6_v7_defconfig index 0beecdde55f5..f25eadcba5e6 100644 --- a/arch/arm/configs/imx_v6_v7_defconfig +++ b/arch/arm/configs/imx_v6_v7_defconfig @@ -323,6 +323,7 @@ CONFIG_SND_SOC_IMX_SGTL5000=y CONFIG_SND_SOC_FSL_ASOC_CARD=y CONFIG_SND_SOC_AC97_CODEC=y CONFIG_SND_SOC_CS42XX8_I2C=y +CONFIG_SND_SOC_SPDIF=y CONFIG_SND_SOC_TLV320AIC3X_I2C=y CONFIG_SND_SOC_WM8960=y CONFIG_SND_SOC_WM8962=y diff --git a/arch/arm/mach-imx/Kconfig b/arch/arm/mach-imx/Kconfig index e4fe059cd861..dc47b2312127 100644 --- a/arch/arm/mach-imx/Kconfig +++ b/arch/arm/mach-imx/Kconfig @@ -6,6 +6,7 @@ menuconfig ARCH_MXC select CLKSRC_IMX_GPT select GENERIC_IRQ_CHIP select GPIOLIB + select PINCTRL select PM_OPP if PM select SOC_BUS select SRAM diff --git a/arch/arm64/boot/dts/freescale/imx8-ss-audio.dtsi b/arch/arm64/boot/dts/freescale/imx8-ss-audio.dtsi index a60ebb718789..c32a6947ae9c 100644 --- a/arch/arm64/boot/dts/freescale/imx8-ss-audio.dtsi +++ b/arch/arm64/boot/dts/freescale/imx8-ss-audio.dtsi @@ -165,7 +165,7 @@ audio_subsys: bus@59000000 { }; esai0: esai@59010000 { - compatible = "fsl,imx8qm-esai"; + compatible = "fsl,imx8qm-esai", "fsl,imx6ull-esai"; reg = <0x59010000 0x10000>; interrupts = <GIC_SPI 409 IRQ_TYPE_LEVEL_HIGH>; clocks = <&esai0_lpcg IMX_LPCG_CLK_4>, diff --git a/arch/arm64/boot/dts/freescale/imx8qm-ss-audio.dtsi b/arch/arm64/boot/dts/freescale/imx8qm-ss-audio.dtsi index e24e639b98ee..c9b55f02497a 100644 --- a/arch/arm64/boot/dts/freescale/imx8qm-ss-audio.dtsi +++ b/arch/arm64/boot/dts/freescale/imx8qm-ss-audio.dtsi @@ -134,7 +134,7 @@ }; esai1: esai@59810000 { - compatible = "fsl,imx8qm-esai"; + compatible = "fsl,imx8qm-esai", "fsl,imx6ull-esai"; reg = <0x59810000 0x10000>; interrupts = <GIC_SPI 411 IRQ_TYPE_LEVEL_HIGH>; clocks = <&esai1_lpcg IMX_LPCG_CLK_0>, diff --git a/arch/arm64/boot/dts/freescale/imx95.dtsi b/arch/arm64/boot/dts/freescale/imx95.dtsi index d10f62eacfe0..e9c7a8265d71 100644 --- a/arch/arm64/boot/dts/freescale/imx95.dtsi +++ b/arch/arm64/boot/dts/freescale/imx95.dtsi @@ -1673,7 +1673,7 @@ netcmix_blk_ctrl: syscon@4c810000 { compatible = "nxp,imx95-netcmix-blk-ctrl", "syscon"; - reg = <0x0 0x4c810000 0x0 0x10000>; + reg = <0x0 0x4c810000 0x0 0x8>; #clock-cells = <1>; clocks = <&scmi_clk IMX95_CLK_BUSNETCMIX>; assigned-clocks = <&scmi_clk IMX95_CLK_BUSNETCMIX>; diff --git a/arch/arm64/boot/dts/qcom/sa8775p.dtsi b/arch/arm64/boot/dts/qcom/sa8775p.dtsi index 9f315a51a7c1..9da62d7c4d27 100644 --- a/arch/arm64/boot/dts/qcom/sa8775p.dtsi +++ b/arch/arm64/boot/dts/qcom/sa8775p.dtsi @@ -2440,6 +2440,7 @@ qcom,cmb-element-bits = <32>; qcom,cmb-msrs-num = <32>; + status = "disabled"; out-ports { port { @@ -6092,7 +6093,7 @@ <0x0 0x40000000 0x0 0xf20>, <0x0 0x40000f20 0x0 0xa8>, <0x0 0x40001000 0x0 0x4000>, - <0x0 0x40200000 0x0 0x100000>, + <0x0 0x40200000 0x0 0x1fe00000>, <0x0 0x01c03000 0x0 0x1000>, <0x0 0x40005000 0x0 0x2000>; reg-names = "parf", "dbi", "elbi", "atu", "addr_space", @@ -6250,7 +6251,7 @@ <0x0 0x60000000 0x0 0xf20>, <0x0 0x60000f20 0x0 0xa8>, <0x0 0x60001000 0x0 0x4000>, - <0x0 0x60200000 0x0 0x100000>, + <0x0 0x60200000 0x0 0x1fe00000>, <0x0 0x01c13000 0x0 0x1000>, <0x0 0x60005000 0x0 0x2000>; reg-names = "parf", "dbi", "elbi", "atu", "addr_space", diff --git a/arch/arm64/boot/dts/qcom/x1e78100-lenovo-thinkpad-t14s.dts b/arch/arm64/boot/dts/qcom/x1e78100-lenovo-thinkpad-t14s.dts index 975550139e10..66513fc8e67a 100644 --- a/arch/arm64/boot/dts/qcom/x1e78100-lenovo-thinkpad-t14s.dts +++ b/arch/arm64/boot/dts/qcom/x1e78100-lenovo-thinkpad-t14s.dts @@ -773,6 +773,10 @@ status = "okay"; }; +&usb_1_ss0_dwc3 { + dr_mode = "host"; +}; + &usb_1_ss0_dwc3_hs { remote-endpoint = <&pmic_glink_ss0_hs_in>; }; @@ -801,6 +805,10 @@ status = "okay"; }; +&usb_1_ss1_dwc3 { + dr_mode = "host"; +}; + &usb_1_ss1_dwc3_hs { remote-endpoint = <&pmic_glink_ss1_hs_in>; }; diff --git a/arch/arm64/boot/dts/qcom/x1e80100-crd.dts b/arch/arm64/boot/dts/qcom/x1e80100-crd.dts index 39f9d9cdc10d..d51a9bdcf67f 100644 --- a/arch/arm64/boot/dts/qcom/x1e80100-crd.dts +++ b/arch/arm64/boot/dts/qcom/x1e80100-crd.dts @@ -1197,6 +1197,10 @@ status = "okay"; }; +&usb_1_ss0_dwc3 { + dr_mode = "host"; +}; + &usb_1_ss0_dwc3_hs { remote-endpoint = <&pmic_glink_ss0_hs_in>; }; @@ -1225,6 +1229,10 @@ status = "okay"; }; +&usb_1_ss1_dwc3 { + dr_mode = "host"; +}; + &usb_1_ss1_dwc3_hs { remote-endpoint = <&pmic_glink_ss1_hs_in>; }; @@ -1253,6 +1261,10 @@ status = "okay"; }; +&usb_1_ss2_dwc3 { + dr_mode = "host"; +}; + &usb_1_ss2_dwc3_hs { remote-endpoint = <&pmic_glink_ss2_hs_in>; }; diff --git a/arch/arm64/boot/dts/qcom/x1e80100.dtsi b/arch/arm64/boot/dts/qcom/x1e80100.dtsi index 88805629ed2b..7e4f46ad8edd 100644 --- a/arch/arm64/boot/dts/qcom/x1e80100.dtsi +++ b/arch/arm64/boot/dts/qcom/x1e80100.dtsi @@ -2924,7 +2924,7 @@ #address-cells = <3>; #size-cells = <2>; ranges = <0x01000000 0x0 0x00000000 0x0 0x70200000 0x0 0x100000>, - <0x02000000 0x0 0x70300000 0x0 0x70300000 0x0 0x1d00000>; + <0x02000000 0x0 0x70300000 0x0 0x70300000 0x0 0x3d00000>; bus-range = <0x00 0xff>; dma-coherent; @@ -4066,8 +4066,6 @@ dma-coherent; - usb-role-switch; - ports { #address-cells = <1>; #size-cells = <0>; @@ -4321,8 +4319,6 @@ dma-coherent; - usb-role-switch; - ports { #address-cells = <1>; #size-cells = <0>; @@ -4421,8 +4417,6 @@ dma-coherent; - usb-role-switch; - ports { #address-cells = <1>; #size-cells = <0>; diff --git a/arch/arm64/boot/dts/rockchip/rk3328.dtsi b/arch/arm64/boot/dts/rockchip/rk3328.dtsi index 0597de415fe0..7d992c3c01ce 100644 --- a/arch/arm64/boot/dts/rockchip/rk3328.dtsi +++ b/arch/arm64/boot/dts/rockchip/rk3328.dtsi @@ -333,6 +333,7 @@ power-domain@RK3328_PD_HEVC { reg = <RK3328_PD_HEVC>; + clocks = <&cru SCLK_VENC_CORE>; #power-domain-cells = <0>; }; power-domain@RK3328_PD_VIDEO { diff --git a/arch/arm64/boot/dts/rockchip/rk3568.dtsi b/arch/arm64/boot/dts/rockchip/rk3568.dtsi index ecaefe208e3e..695cccbdab0f 100644 --- a/arch/arm64/boot/dts/rockchip/rk3568.dtsi +++ b/arch/arm64/boot/dts/rockchip/rk3568.dtsi @@ -350,6 +350,7 @@ assigned-clocks = <&pmucru CLK_PCIEPHY0_REF>; assigned-clock-rates = <100000000>; resets = <&cru SRST_PIPEPHY0>; + reset-names = "phy"; rockchip,pipe-grf = <&pipegrf>; rockchip,pipe-phy-grf = <&pipe_phy_grf0>; #phy-cells = <1>; diff --git a/arch/arm64/boot/dts/rockchip/rk356x-base.dtsi b/arch/arm64/boot/dts/rockchip/rk356x-base.dtsi index 62be06f3b863..e55390629114 100644 --- a/arch/arm64/boot/dts/rockchip/rk356x-base.dtsi +++ b/arch/arm64/boot/dts/rockchip/rk356x-base.dtsi @@ -1681,6 +1681,7 @@ assigned-clocks = <&pmucru CLK_PCIEPHY1_REF>; assigned-clock-rates = <100000000>; resets = <&cru SRST_PIPEPHY1>; + reset-names = "phy"; rockchip,pipe-grf = <&pipegrf>; rockchip,pipe-phy-grf = <&pipe_phy_grf1>; #phy-cells = <1>; @@ -1697,6 +1698,7 @@ assigned-clocks = <&pmucru CLK_PCIEPHY2_REF>; assigned-clock-rates = <100000000>; resets = <&cru SRST_PIPEPHY2>; + reset-names = "phy"; rockchip,pipe-grf = <&pipegrf>; rockchip,pipe-phy-grf = <&pipe_phy_grf2>; #phy-cells = <1>; diff --git a/arch/arm64/boot/dts/rockchip/rk3588-rock-5b.dts b/arch/arm64/boot/dts/rockchip/rk3588-rock-5b.dts index c44d001da169..d597112f1d5b 100644 --- a/arch/arm64/boot/dts/rockchip/rk3588-rock-5b.dts +++ b/arch/arm64/boot/dts/rockchip/rk3588-rock-5b.dts @@ -72,7 +72,7 @@ rfkill { compatible = "rfkill-gpio"; - label = "rfkill-pcie-wlan"; + label = "rfkill-m2-wlan"; radio-type = "wlan"; shutdown-gpios = <&gpio4 RK_PA2 GPIO_ACTIVE_HIGH>; }; diff --git a/arch/arm64/boot/dts/rockchip/rk3588s-nanopi-r6.dtsi b/arch/arm64/boot/dts/rockchip/rk3588s-nanopi-r6.dtsi index 76a6e8e517e9..c9749cb50076 100644 --- a/arch/arm64/boot/dts/rockchip/rk3588s-nanopi-r6.dtsi +++ b/arch/arm64/boot/dts/rockchip/rk3588s-nanopi-r6.dtsi @@ -434,6 +434,7 @@ &sdmmc { bus-width = <4>; cap-sd-highspeed; + cd-gpios = <&gpio0 RK_PA4 GPIO_ACTIVE_LOW>; disable-wp; max-frequency = <150000000>; no-mmc; diff --git a/arch/arm64/kvm/hyp/nvhe/mem_protect.c b/arch/arm64/kvm/hyp/nvhe/mem_protect.c index caba3e4bd09e..e75374d682f4 100644 --- a/arch/arm64/kvm/hyp/nvhe/mem_protect.c +++ b/arch/arm64/kvm/hyp/nvhe/mem_protect.c @@ -783,9 +783,6 @@ static int hyp_ack_unshare(u64 addr, const struct pkvm_mem_transition *tx) if (tx->initiator.id == PKVM_ID_HOST && hyp_page_count((void *)addr)) return -EBUSY; - if (__hyp_ack_skip_pgtable_check(tx)) - return 0; - return __hyp_check_page_state_range(addr, size, PKVM_PAGE_SHARED_BORROWED); } diff --git a/arch/arm64/kvm/pmu-emul.c b/arch/arm64/kvm/pmu-emul.c index 456102bc0b55..6c5950b9ceac 100644 --- a/arch/arm64/kvm/pmu-emul.c +++ b/arch/arm64/kvm/pmu-emul.c @@ -24,6 +24,7 @@ static DEFINE_MUTEX(arm_pmus_lock); static void kvm_pmu_create_perf_event(struct kvm_pmc *pmc); static void kvm_pmu_release_perf_event(struct kvm_pmc *pmc); +static bool kvm_pmu_counter_is_enabled(struct kvm_pmc *pmc); static struct kvm_vcpu *kvm_pmc_to_vcpu(const struct kvm_pmc *pmc) { @@ -327,48 +328,25 @@ u64 kvm_pmu_implemented_counter_mask(struct kvm_vcpu *vcpu) return GENMASK(val - 1, 0) | BIT(ARMV8_PMU_CYCLE_IDX); } -/** - * kvm_pmu_enable_counter_mask - enable selected PMU counters - * @vcpu: The vcpu pointer - * @val: the value guest writes to PMCNTENSET register - * - * Call perf_event_enable to start counting the perf event - */ -void kvm_pmu_enable_counter_mask(struct kvm_vcpu *vcpu, u64 val) +static void kvm_pmc_enable_perf_event(struct kvm_pmc *pmc) { - int i; - if (!kvm_vcpu_has_pmu(vcpu)) - return; - - if (!(kvm_vcpu_read_pmcr(vcpu) & ARMV8_PMU_PMCR_E) || !val) + if (!pmc->perf_event) { + kvm_pmu_create_perf_event(pmc); return; + } - for (i = 0; i < KVM_ARMV8_PMU_MAX_COUNTERS; i++) { - struct kvm_pmc *pmc; - - if (!(val & BIT(i))) - continue; - - pmc = kvm_vcpu_idx_to_pmc(vcpu, i); + perf_event_enable(pmc->perf_event); + if (pmc->perf_event->state != PERF_EVENT_STATE_ACTIVE) + kvm_debug("fail to enable perf event\n"); +} - if (!pmc->perf_event) { - kvm_pmu_create_perf_event(pmc); - } else { - perf_event_enable(pmc->perf_event); - if (pmc->perf_event->state != PERF_EVENT_STATE_ACTIVE) - kvm_debug("fail to enable perf event\n"); - } - } +static void kvm_pmc_disable_perf_event(struct kvm_pmc *pmc) +{ + if (pmc->perf_event) + perf_event_disable(pmc->perf_event); } -/** - * kvm_pmu_disable_counter_mask - disable selected PMU counters - * @vcpu: The vcpu pointer - * @val: the value guest writes to PMCNTENCLR register - * - * Call perf_event_disable to stop counting the perf event - */ -void kvm_pmu_disable_counter_mask(struct kvm_vcpu *vcpu, u64 val) +void kvm_pmu_reprogram_counter_mask(struct kvm_vcpu *vcpu, u64 val) { int i; @@ -376,16 +354,18 @@ void kvm_pmu_disable_counter_mask(struct kvm_vcpu *vcpu, u64 val) return; for (i = 0; i < KVM_ARMV8_PMU_MAX_COUNTERS; i++) { - struct kvm_pmc *pmc; + struct kvm_pmc *pmc = kvm_vcpu_idx_to_pmc(vcpu, i); if (!(val & BIT(i))) continue; - pmc = kvm_vcpu_idx_to_pmc(vcpu, i); - - if (pmc->perf_event) - perf_event_disable(pmc->perf_event); + if (kvm_pmu_counter_is_enabled(pmc)) + kvm_pmc_enable_perf_event(pmc); + else + kvm_pmc_disable_perf_event(pmc); } + + kvm_vcpu_pmu_restore_guest(vcpu); } /* @@ -626,27 +606,28 @@ void kvm_pmu_handle_pmcr(struct kvm_vcpu *vcpu, u64 val) if (!kvm_has_feat(vcpu->kvm, ID_AA64DFR0_EL1, PMUVer, V3P5)) val &= ~ARMV8_PMU_PMCR_LP; + /* Request a reload of the PMU to enable/disable affected counters */ + if ((__vcpu_sys_reg(vcpu, PMCR_EL0) ^ val) & ARMV8_PMU_PMCR_E) + kvm_make_request(KVM_REQ_RELOAD_PMU, vcpu); + /* The reset bits don't indicate any state, and shouldn't be saved. */ __vcpu_sys_reg(vcpu, PMCR_EL0) = val & ~(ARMV8_PMU_PMCR_C | ARMV8_PMU_PMCR_P); - if (val & ARMV8_PMU_PMCR_E) { - kvm_pmu_enable_counter_mask(vcpu, - __vcpu_sys_reg(vcpu, PMCNTENSET_EL0)); - } else { - kvm_pmu_disable_counter_mask(vcpu, - __vcpu_sys_reg(vcpu, PMCNTENSET_EL0)); - } - if (val & ARMV8_PMU_PMCR_C) kvm_pmu_set_counter_value(vcpu, ARMV8_PMU_CYCLE_IDX, 0); if (val & ARMV8_PMU_PMCR_P) { - unsigned long mask = kvm_pmu_accessible_counter_mask(vcpu); - mask &= ~BIT(ARMV8_PMU_CYCLE_IDX); + /* + * Unlike other PMU sysregs, the controls in PMCR_EL0 always apply + * to the 'guest' range of counters and never the 'hyp' range. + */ + unsigned long mask = kvm_pmu_implemented_counter_mask(vcpu) & + ~kvm_pmu_hyp_counter_mask(vcpu) & + ~BIT(ARMV8_PMU_CYCLE_IDX); + for_each_set_bit(i, &mask, 32) kvm_pmu_set_pmc_value(kvm_vcpu_idx_to_pmc(vcpu, i), 0, true); } - kvm_vcpu_pmu_restore_guest(vcpu); } static bool kvm_pmu_counter_is_enabled(struct kvm_pmc *pmc) @@ -910,11 +891,11 @@ void kvm_vcpu_reload_pmu(struct kvm_vcpu *vcpu) { u64 mask = kvm_pmu_implemented_counter_mask(vcpu); - kvm_pmu_handle_pmcr(vcpu, kvm_vcpu_read_pmcr(vcpu)); - __vcpu_sys_reg(vcpu, PMOVSSET_EL0) &= mask; __vcpu_sys_reg(vcpu, PMINTENSET_EL1) &= mask; __vcpu_sys_reg(vcpu, PMCNTENSET_EL0) &= mask; + + kvm_pmu_reprogram_counter_mask(vcpu, mask); } int kvm_arm_pmu_v3_enable(struct kvm_vcpu *vcpu) diff --git a/arch/arm64/kvm/sys_regs.c b/arch/arm64/kvm/sys_regs.c index e2a5c2918d9e..634ff18a59a1 100644 --- a/arch/arm64/kvm/sys_regs.c +++ b/arch/arm64/kvm/sys_regs.c @@ -1208,16 +1208,14 @@ static bool access_pmcnten(struct kvm_vcpu *vcpu, struct sys_reg_params *p, mask = kvm_pmu_accessible_counter_mask(vcpu); if (p->is_write) { val = p->regval & mask; - if (r->Op2 & 0x1) { + if (r->Op2 & 0x1) /* accessing PMCNTENSET_EL0 */ __vcpu_sys_reg(vcpu, PMCNTENSET_EL0) |= val; - kvm_pmu_enable_counter_mask(vcpu, val); - kvm_vcpu_pmu_restore_guest(vcpu); - } else { + else /* accessing PMCNTENCLR_EL0 */ __vcpu_sys_reg(vcpu, PMCNTENSET_EL0) &= ~val; - kvm_pmu_disable_counter_mask(vcpu, val); - } + + kvm_pmu_reprogram_counter_mask(vcpu, val); } else { p->regval = __vcpu_sys_reg(vcpu, PMCNTENSET_EL0); } @@ -2450,6 +2448,26 @@ static unsigned int s1pie_el2_visibility(const struct kvm_vcpu *vcpu, return __el2_visibility(vcpu, rd, s1pie_visibility); } +static bool access_mdcr(struct kvm_vcpu *vcpu, + struct sys_reg_params *p, + const struct sys_reg_desc *r) +{ + u64 old = __vcpu_sys_reg(vcpu, MDCR_EL2); + + if (!access_rw(vcpu, p, r)) + return false; + + /* + * Request a reload of the PMU to enable/disable the counters affected + * by HPME. + */ + if ((old ^ __vcpu_sys_reg(vcpu, MDCR_EL2)) & MDCR_EL2_HPME) + kvm_make_request(KVM_REQ_RELOAD_PMU, vcpu); + + return true; +} + + /* * Architected system registers. * Important: Must be sorted ascending by Op0, Op1, CRn, CRm, Op2 @@ -2983,7 +3001,7 @@ static const struct sys_reg_desc sys_reg_descs[] = { EL2_REG(SCTLR_EL2, access_rw, reset_val, SCTLR_EL2_RES1), EL2_REG(ACTLR_EL2, access_rw, reset_val, 0), EL2_REG_VNCR(HCR_EL2, reset_hcr, 0), - EL2_REG(MDCR_EL2, access_rw, reset_val, 0), + EL2_REG(MDCR_EL2, access_mdcr, reset_val, 0), EL2_REG(CPTR_EL2, access_rw, reset_val, CPTR_NVHE_EL2_RES1), EL2_REG_VNCR(HSTR_EL2, reset_val, 0), EL2_REG_VNCR(HFGRTR_EL2, reset_val, 0), diff --git a/arch/nios2/kernel/cpuinfo.c b/arch/nios2/kernel/cpuinfo.c index 338849c430a5..7b1e8f9128e9 100644 --- a/arch/nios2/kernel/cpuinfo.c +++ b/arch/nios2/kernel/cpuinfo.c @@ -143,11 +143,11 @@ static int show_cpuinfo(struct seq_file *m, void *v) " DIV:\t\t%s\n" " BMX:\t\t%s\n" " CDX:\t\t%s\n", - cpuinfo.has_mul ? "yes" : "no", - cpuinfo.has_mulx ? "yes" : "no", - cpuinfo.has_div ? "yes" : "no", - cpuinfo.has_bmx ? "yes" : "no", - cpuinfo.has_cdx ? "yes" : "no"); + str_yes_no(cpuinfo.has_mul), + str_yes_no(cpuinfo.has_mulx), + str_yes_no(cpuinfo.has_div), + str_yes_no(cpuinfo.has_bmx), + str_yes_no(cpuinfo.has_cdx)); seq_printf(m, "Icache:\t\t%ukB, line length: %u\n", diff --git a/arch/powerpc/kvm/e500.h b/arch/powerpc/kvm/e500.h index 6d0d329cbb35..f9acf866c709 100644 --- a/arch/powerpc/kvm/e500.h +++ b/arch/powerpc/kvm/e500.h @@ -34,6 +34,8 @@ enum vcpu_ftr { #define E500_TLB_BITMAP (1 << 30) /* TLB1 entry is mapped by host TLB0 */ #define E500_TLB_TLB0 (1 << 29) +/* entry is writable on the host */ +#define E500_TLB_WRITABLE (1 << 28) /* bits [6-5] MAS2_X1 and MAS2_X0 and [4-0] bits for WIMGE */ #define E500_TLB_MAS2_ATTR (0x7f) diff --git a/arch/powerpc/kvm/e500_mmu_host.c b/arch/powerpc/kvm/e500_mmu_host.c index e5a145b578a4..06caf8bbbe2b 100644 --- a/arch/powerpc/kvm/e500_mmu_host.c +++ b/arch/powerpc/kvm/e500_mmu_host.c @@ -45,11 +45,14 @@ static inline unsigned int tlb1_max_shadow_size(void) return host_tlb_params[1].entries - tlbcam_index - 1; } -static inline u32 e500_shadow_mas3_attrib(u32 mas3, int usermode) +static inline u32 e500_shadow_mas3_attrib(u32 mas3, bool writable, int usermode) { /* Mask off reserved bits. */ mas3 &= MAS3_ATTRIB_MASK; + if (!writable) + mas3 &= ~(MAS3_UW|MAS3_SW); + #ifndef CONFIG_KVM_BOOKE_HV if (!usermode) { /* Guest is in supervisor mode, @@ -242,17 +245,18 @@ static inline int tlbe_is_writable(struct kvm_book3e_206_tlb_entry *tlbe) return tlbe->mas7_3 & (MAS3_SW|MAS3_UW); } -static inline bool kvmppc_e500_ref_setup(struct tlbe_ref *ref, +static inline void kvmppc_e500_ref_setup(struct tlbe_ref *ref, struct kvm_book3e_206_tlb_entry *gtlbe, - kvm_pfn_t pfn, unsigned int wimg) + kvm_pfn_t pfn, unsigned int wimg, + bool writable) { ref->pfn = pfn; ref->flags = E500_TLB_VALID; + if (writable) + ref->flags |= E500_TLB_WRITABLE; /* Use guest supplied MAS2_G and MAS2_E */ ref->flags |= (gtlbe->mas2 & MAS2_ATTRIB_MASK) | wimg; - - return tlbe_is_writable(gtlbe); } static inline void kvmppc_e500_ref_release(struct tlbe_ref *ref) @@ -305,6 +309,7 @@ static void kvmppc_e500_setup_stlbe( { kvm_pfn_t pfn = ref->pfn; u32 pr = vcpu->arch.shared->msr & MSR_PR; + bool writable = !!(ref->flags & E500_TLB_WRITABLE); BUG_ON(!(ref->flags & E500_TLB_VALID)); @@ -312,7 +317,7 @@ static void kvmppc_e500_setup_stlbe( stlbe->mas1 = MAS1_TSIZE(tsize) | get_tlb_sts(gtlbe) | MAS1_VALID; stlbe->mas2 = (gvaddr & MAS2_EPN) | (ref->flags & E500_TLB_MAS2_ATTR); stlbe->mas7_3 = ((u64)pfn << PAGE_SHIFT) | - e500_shadow_mas3_attrib(gtlbe->mas7_3, pr); + e500_shadow_mas3_attrib(gtlbe->mas7_3, writable, pr); } static inline int kvmppc_e500_shadow_map(struct kvmppc_vcpu_e500 *vcpu_e500, @@ -321,15 +326,14 @@ static inline int kvmppc_e500_shadow_map(struct kvmppc_vcpu_e500 *vcpu_e500, struct tlbe_ref *ref) { struct kvm_memory_slot *slot; - unsigned long pfn = 0; /* silence GCC warning */ + unsigned int psize; + unsigned long pfn; struct page *page = NULL; unsigned long hva; - int pfnmap = 0; int tsize = BOOK3E_PAGESZ_4K; int ret = 0; unsigned long mmu_seq; struct kvm *kvm = vcpu_e500->vcpu.kvm; - unsigned long tsize_pages = 0; pte_t *ptep; unsigned int wimg = 0; pgd_t *pgdir; @@ -351,110 +355,12 @@ static inline int kvmppc_e500_shadow_map(struct kvmppc_vcpu_e500 *vcpu_e500, slot = gfn_to_memslot(vcpu_e500->vcpu.kvm, gfn); hva = gfn_to_hva_memslot(slot, gfn); - if (tlbsel == 1) { - struct vm_area_struct *vma; - mmap_read_lock(kvm->mm); - - vma = find_vma(kvm->mm, hva); - if (vma && hva >= vma->vm_start && - (vma->vm_flags & VM_PFNMAP)) { - /* - * This VMA is a physically contiguous region (e.g. - * /dev/mem) that bypasses normal Linux page - * management. Find the overlap between the - * vma and the memslot. - */ - - unsigned long start, end; - unsigned long slot_start, slot_end; - - pfnmap = 1; - - start = vma->vm_pgoff; - end = start + - vma_pages(vma); - - pfn = start + ((hva - vma->vm_start) >> PAGE_SHIFT); - - slot_start = pfn - (gfn - slot->base_gfn); - slot_end = slot_start + slot->npages; - - if (start < slot_start) - start = slot_start; - if (end > slot_end) - end = slot_end; - - tsize = (gtlbe->mas1 & MAS1_TSIZE_MASK) >> - MAS1_TSIZE_SHIFT; - - /* - * e500 doesn't implement the lowest tsize bit, - * or 1K pages. - */ - tsize = max(BOOK3E_PAGESZ_4K, tsize & ~1); - - /* - * Now find the largest tsize (up to what the guest - * requested) that will cover gfn, stay within the - * range, and for which gfn and pfn are mutually - * aligned. - */ - - for (; tsize > BOOK3E_PAGESZ_4K; tsize -= 2) { - unsigned long gfn_start, gfn_end; - tsize_pages = 1UL << (tsize - 2); - - gfn_start = gfn & ~(tsize_pages - 1); - gfn_end = gfn_start + tsize_pages; - - if (gfn_start + pfn - gfn < start) - continue; - if (gfn_end + pfn - gfn > end) - continue; - if ((gfn & (tsize_pages - 1)) != - (pfn & (tsize_pages - 1))) - continue; - - gvaddr &= ~((tsize_pages << PAGE_SHIFT) - 1); - pfn &= ~(tsize_pages - 1); - break; - } - } else if (vma && hva >= vma->vm_start && - is_vm_hugetlb_page(vma)) { - unsigned long psize = vma_kernel_pagesize(vma); - - tsize = (gtlbe->mas1 & MAS1_TSIZE_MASK) >> - MAS1_TSIZE_SHIFT; - - /* - * Take the largest page size that satisfies both host - * and guest mapping - */ - tsize = min(__ilog2(psize) - 10, tsize); - - /* - * e500 doesn't implement the lowest tsize bit, - * or 1K pages. - */ - tsize = max(BOOK3E_PAGESZ_4K, tsize & ~1); - } - - mmap_read_unlock(kvm->mm); - } - - if (likely(!pfnmap)) { - tsize_pages = 1UL << (tsize + 10 - PAGE_SHIFT); - pfn = __kvm_faultin_pfn(slot, gfn, FOLL_WRITE, NULL, &page); - if (is_error_noslot_pfn(pfn)) { - if (printk_ratelimit()) - pr_err("%s: real page not found for gfn %lx\n", - __func__, (long)gfn); - return -EINVAL; - } - - /* Align guest and physical address to page map boundaries */ - pfn &= ~(tsize_pages - 1); - gvaddr &= ~((tsize_pages << PAGE_SHIFT) - 1); + pfn = __kvm_faultin_pfn(slot, gfn, FOLL_WRITE, &writable, &page); + if (is_error_noslot_pfn(pfn)) { + if (printk_ratelimit()) + pr_err("%s: real page not found for gfn %lx\n", + __func__, (long)gfn); + return -EINVAL; } spin_lock(&kvm->mmu_lock); @@ -472,14 +378,13 @@ static inline int kvmppc_e500_shadow_map(struct kvmppc_vcpu_e500 *vcpu_e500, * can't run hence pfn won't change. */ local_irq_save(flags); - ptep = find_linux_pte(pgdir, hva, NULL, NULL); + ptep = find_linux_pte(pgdir, hva, NULL, &psize); if (ptep) { pte_t pte = READ_ONCE(*ptep); if (pte_present(pte)) { wimg = (pte_val(pte) >> PTE_WIMGE_SHIFT) & MAS2_WIMGE_MASK; - local_irq_restore(flags); } else { local_irq_restore(flags); pr_err_ratelimited("%s: pte not present: gfn %lx,pfn %lx\n", @@ -488,10 +393,72 @@ static inline int kvmppc_e500_shadow_map(struct kvmppc_vcpu_e500 *vcpu_e500, goto out; } } - writable = kvmppc_e500_ref_setup(ref, gtlbe, pfn, wimg); + local_irq_restore(flags); + + if (psize && tlbsel == 1) { + unsigned long psize_pages, tsize_pages; + unsigned long start, end; + unsigned long slot_start, slot_end; + + psize_pages = 1UL << (psize - PAGE_SHIFT); + start = pfn & ~(psize_pages - 1); + end = start + psize_pages; + + slot_start = pfn - (gfn - slot->base_gfn); + slot_end = slot_start + slot->npages; + + if (start < slot_start) + start = slot_start; + if (end > slot_end) + end = slot_end; + + tsize = (gtlbe->mas1 & MAS1_TSIZE_MASK) >> + MAS1_TSIZE_SHIFT; + + /* + * Any page size that doesn't satisfy the host mapping + * will fail the start and end tests. + */ + tsize = min(psize - PAGE_SHIFT + BOOK3E_PAGESZ_4K, tsize); + + /* + * e500 doesn't implement the lowest tsize bit, + * or 1K pages. + */ + tsize = max(BOOK3E_PAGESZ_4K, tsize & ~1); + + /* + * Now find the largest tsize (up to what the guest + * requested) that will cover gfn, stay within the + * range, and for which gfn and pfn are mutually + * aligned. + */ + + for (; tsize > BOOK3E_PAGESZ_4K; tsize -= 2) { + unsigned long gfn_start, gfn_end; + tsize_pages = 1UL << (tsize - 2); + + gfn_start = gfn & ~(tsize_pages - 1); + gfn_end = gfn_start + tsize_pages; + + if (gfn_start + pfn - gfn < start) + continue; + if (gfn_end + pfn - gfn > end) + continue; + if ((gfn & (tsize_pages - 1)) != + (pfn & (tsize_pages - 1))) + continue; + + gvaddr &= ~((tsize_pages << PAGE_SHIFT) - 1); + pfn &= ~(tsize_pages - 1); + break; + } + } + kvmppc_e500_ref_setup(ref, gtlbe, pfn, wimg, writable); kvmppc_e500_setup_stlbe(&vcpu_e500->vcpu, gtlbe, tsize, ref, gvaddr, stlbe); + writable = tlbe_is_writable(stlbe); /* Clear i-cache for new pages */ kvmppc_mmu_flush_icache(pfn); diff --git a/arch/powerpc/platforms/book3s/vas-api.c b/arch/powerpc/platforms/book3s/vas-api.c index f381b177ea06..0b6365d85d11 100644 --- a/arch/powerpc/platforms/book3s/vas-api.c +++ b/arch/powerpc/platforms/book3s/vas-api.c @@ -464,7 +464,43 @@ static vm_fault_t vas_mmap_fault(struct vm_fault *vmf) return VM_FAULT_SIGBUS; } +/* + * During mmap() paste address, mapping VMA is saved in VAS window + * struct which is used to unmap during migration if the window is + * still open. But the user space can remove this mapping with + * munmap() before closing the window and the VMA address will + * be invalid. Set VAS window VMA to NULL in this function which + * is called before VMA free. + */ +static void vas_mmap_close(struct vm_area_struct *vma) +{ + struct file *fp = vma->vm_file; + struct coproc_instance *cp_inst = fp->private_data; + struct vas_window *txwin; + + /* Should not happen */ + if (!cp_inst || !cp_inst->txwin) { + pr_err("No attached VAS window for the paste address mmap\n"); + return; + } + + txwin = cp_inst->txwin; + /* + * task_ref.vma is set in coproc_mmap() during mmap paste + * address. So it has to be the same VMA that is getting freed. + */ + if (WARN_ON(txwin->task_ref.vma != vma)) { + pr_err("Invalid paste address mmaping\n"); + return; + } + + mutex_lock(&txwin->task_ref.mmap_mutex); + txwin->task_ref.vma = NULL; + mutex_unlock(&txwin->task_ref.mmap_mutex); +} + static const struct vm_operations_struct vas_vm_ops = { + .close = vas_mmap_close, .fault = vas_mmap_fault, }; diff --git a/arch/riscv/include/asm/page.h b/arch/riscv/include/asm/page.h index 71aabc5c6713..125f5ecd9565 100644 --- a/arch/riscv/include/asm/page.h +++ b/arch/riscv/include/asm/page.h @@ -122,6 +122,7 @@ struct kernel_mapping { extern struct kernel_mapping kernel_map; extern phys_addr_t phys_ram_base; +extern unsigned long vmemmap_start_pfn; #define is_kernel_mapping(x) \ ((x) >= kernel_map.virt_addr && (x) < (kernel_map.virt_addr + kernel_map.size)) diff --git a/arch/riscv/include/asm/pgtable.h b/arch/riscv/include/asm/pgtable.h index d4e99eef90ac..050fdc49b5ad 100644 --- a/arch/riscv/include/asm/pgtable.h +++ b/arch/riscv/include/asm/pgtable.h @@ -87,7 +87,7 @@ * Define vmemmap for pfn_to_page & page_to_pfn calls. Needed if kernel * is configured with CONFIG_SPARSEMEM_VMEMMAP enabled. */ -#define vmemmap ((struct page *)VMEMMAP_START - (phys_ram_base >> PAGE_SHIFT)) +#define vmemmap ((struct page *)VMEMMAP_START - vmemmap_start_pfn) #define PCI_IO_SIZE SZ_16M #define PCI_IO_END VMEMMAP_START diff --git a/arch/riscv/include/asm/sbi.h b/arch/riscv/include/asm/sbi.h index 6c82318065cf..3d250824178b 100644 --- a/arch/riscv/include/asm/sbi.h +++ b/arch/riscv/include/asm/sbi.h @@ -159,6 +159,7 @@ struct riscv_pmu_snapshot_data { }; #define RISCV_PMU_RAW_EVENT_MASK GENMASK_ULL(47, 0) +#define RISCV_PMU_PLAT_FW_EVENT_MASK GENMASK_ULL(61, 0) #define RISCV_PMU_RAW_EVENT_IDX 0x20000 #define RISCV_PLAT_FW_EVENT 0xFFFF diff --git a/arch/riscv/include/asm/spinlock.h b/arch/riscv/include/asm/spinlock.h index e5121b89acea..52f11bfd0079 100644 --- a/arch/riscv/include/asm/spinlock.h +++ b/arch/riscv/include/asm/spinlock.h @@ -3,8 +3,11 @@ #ifndef __ASM_RISCV_SPINLOCK_H #define __ASM_RISCV_SPINLOCK_H -#ifdef CONFIG_RISCV_COMBO_SPINLOCKS +#ifdef CONFIG_QUEUED_SPINLOCKS #define _Q_PENDING_LOOPS (1 << 9) +#endif + +#ifdef CONFIG_RISCV_COMBO_SPINLOCKS #define __no_arch_spinlock_redefine #include <asm/ticket_spinlock.h> diff --git a/arch/riscv/kernel/entry.S b/arch/riscv/kernel/entry.S index c200d329d4bd..33a5a9f2a0d4 100644 --- a/arch/riscv/kernel/entry.S +++ b/arch/riscv/kernel/entry.S @@ -23,21 +23,21 @@ REG_S a0, TASK_TI_A0(tp) csrr a0, CSR_CAUSE /* Exclude IRQs */ - blt a0, zero, _new_vmalloc_restore_context_a0 + blt a0, zero, .Lnew_vmalloc_restore_context_a0 REG_S a1, TASK_TI_A1(tp) /* Only check new_vmalloc if we are in page/protection fault */ li a1, EXC_LOAD_PAGE_FAULT - beq a0, a1, _new_vmalloc_kernel_address + beq a0, a1, .Lnew_vmalloc_kernel_address li a1, EXC_STORE_PAGE_FAULT - beq a0, a1, _new_vmalloc_kernel_address + beq a0, a1, .Lnew_vmalloc_kernel_address li a1, EXC_INST_PAGE_FAULT - bne a0, a1, _new_vmalloc_restore_context_a1 + bne a0, a1, .Lnew_vmalloc_restore_context_a1 -_new_vmalloc_kernel_address: +.Lnew_vmalloc_kernel_address: /* Is it a kernel address? */ csrr a0, CSR_TVAL - bge a0, zero, _new_vmalloc_restore_context_a1 + bge a0, zero, .Lnew_vmalloc_restore_context_a1 /* Check if a new vmalloc mapping appeared that could explain the trap */ REG_S a2, TASK_TI_A2(tp) @@ -69,7 +69,7 @@ _new_vmalloc_kernel_address: /* Check the value of new_vmalloc for this cpu */ REG_L a2, 0(a0) and a2, a2, a1 - beq a2, zero, _new_vmalloc_restore_context + beq a2, zero, .Lnew_vmalloc_restore_context /* Atomically reset the current cpu bit in new_vmalloc */ amoxor.d a0, a1, (a0) @@ -83,11 +83,11 @@ _new_vmalloc_kernel_address: csrw CSR_SCRATCH, x0 sret -_new_vmalloc_restore_context: +.Lnew_vmalloc_restore_context: REG_L a2, TASK_TI_A2(tp) -_new_vmalloc_restore_context_a1: +.Lnew_vmalloc_restore_context_a1: REG_L a1, TASK_TI_A1(tp) -_new_vmalloc_restore_context_a0: +.Lnew_vmalloc_restore_context_a0: REG_L a0, TASK_TI_A0(tp) .endm @@ -278,6 +278,7 @@ SYM_CODE_START_NOALIGN(ret_from_exception) #else sret #endif +SYM_INNER_LABEL(ret_from_exception_end, SYM_L_GLOBAL) SYM_CODE_END(ret_from_exception) ASM_NOKPROBE(ret_from_exception) diff --git a/arch/riscv/kernel/module.c b/arch/riscv/kernel/module.c index 1cd461f3d872..47d0ebeec93c 100644 --- a/arch/riscv/kernel/module.c +++ b/arch/riscv/kernel/module.c @@ -23,7 +23,7 @@ struct used_bucket { struct relocation_head { struct hlist_node node; - struct list_head *rel_entry; + struct list_head rel_entry; void *location; }; @@ -634,7 +634,7 @@ process_accumulated_relocations(struct module *me, location = rel_head_iter->location; list_for_each_entry_safe(rel_entry_iter, rel_entry_iter_tmp, - rel_head_iter->rel_entry, + &rel_head_iter->rel_entry, head) { curr_type = rel_entry_iter->type; reloc_handlers[curr_type].reloc_handler( @@ -704,16 +704,7 @@ static int add_relocation_to_accumulate(struct module *me, int type, return -ENOMEM; } - rel_head->rel_entry = - kmalloc(sizeof(struct list_head), GFP_KERNEL); - - if (!rel_head->rel_entry) { - kfree(entry); - kfree(rel_head); - return -ENOMEM; - } - - INIT_LIST_HEAD(rel_head->rel_entry); + INIT_LIST_HEAD(&rel_head->rel_entry); rel_head->location = location; INIT_HLIST_NODE(&rel_head->node); if (!current_head->first) { @@ -722,7 +713,6 @@ static int add_relocation_to_accumulate(struct module *me, int type, if (!bucket) { kfree(entry); - kfree(rel_head->rel_entry); kfree(rel_head); return -ENOMEM; } @@ -735,7 +725,7 @@ static int add_relocation_to_accumulate(struct module *me, int type, } /* Add relocation to head of discovered rel_head */ - list_add_tail(&entry->head, rel_head->rel_entry); + list_add_tail(&entry->head, &rel_head->rel_entry); return 0; } diff --git a/arch/riscv/kernel/probes/kprobes.c b/arch/riscv/kernel/probes/kprobes.c index 380a0e8cecc0..c0738d6c6498 100644 --- a/arch/riscv/kernel/probes/kprobes.c +++ b/arch/riscv/kernel/probes/kprobes.c @@ -30,7 +30,7 @@ static void __kprobes arch_prepare_ss_slot(struct kprobe *p) p->ainsn.api.restore = (unsigned long)p->addr + len; patch_text_nosync(p->ainsn.api.insn, &p->opcode, len); - patch_text_nosync(p->ainsn.api.insn + len, &insn, GET_INSN_LENGTH(insn)); + patch_text_nosync((void *)p->ainsn.api.insn + len, &insn, GET_INSN_LENGTH(insn)); } static void __kprobes arch_prepare_simulate(struct kprobe *p) diff --git a/arch/riscv/kernel/stacktrace.c b/arch/riscv/kernel/stacktrace.c index 153a2db4c5fa..d4355c770c36 100644 --- a/arch/riscv/kernel/stacktrace.c +++ b/arch/riscv/kernel/stacktrace.c @@ -17,6 +17,7 @@ #ifdef CONFIG_FRAME_POINTER extern asmlinkage void handle_exception(void); +extern unsigned long ret_from_exception_end; static inline int fp_is_valid(unsigned long fp, unsigned long sp) { @@ -71,7 +72,8 @@ void notrace walk_stackframe(struct task_struct *task, struct pt_regs *regs, fp = frame->fp; pc = ftrace_graph_ret_addr(current, &graph_idx, frame->ra, &frame->ra); - if (pc == (unsigned long)handle_exception) { + if (pc >= (unsigned long)handle_exception && + pc < (unsigned long)&ret_from_exception_end) { if (unlikely(!__kernel_text_address(pc) || !fn(arg, pc))) break; diff --git a/arch/riscv/kernel/traps.c b/arch/riscv/kernel/traps.c index 51ebfd23e007..8ff8e8b36524 100644 --- a/arch/riscv/kernel/traps.c +++ b/arch/riscv/kernel/traps.c @@ -35,7 +35,7 @@ int show_unhandled_signals = 1; -static DEFINE_SPINLOCK(die_lock); +static DEFINE_RAW_SPINLOCK(die_lock); static int copy_code(struct pt_regs *regs, u16 *val, const u16 *insns) { @@ -81,7 +81,7 @@ void die(struct pt_regs *regs, const char *str) oops_enter(); - spin_lock_irqsave(&die_lock, flags); + raw_spin_lock_irqsave(&die_lock, flags); console_verbose(); bust_spinlocks(1); @@ -100,7 +100,7 @@ void die(struct pt_regs *regs, const char *str) bust_spinlocks(0); add_taint(TAINT_DIE, LOCKDEP_NOW_UNRELIABLE); - spin_unlock_irqrestore(&die_lock, flags); + raw_spin_unlock_irqrestore(&die_lock, flags); oops_exit(); if (in_interrupt()) diff --git a/arch/riscv/mm/init.c b/arch/riscv/mm/init.c index fc53ce748c80..8d167e09f1fe 100644 --- a/arch/riscv/mm/init.c +++ b/arch/riscv/mm/init.c @@ -33,6 +33,7 @@ #include <asm/pgtable.h> #include <asm/sections.h> #include <asm/soc.h> +#include <asm/sparsemem.h> #include <asm/tlbflush.h> #include "../kernel/head.h" @@ -62,6 +63,13 @@ EXPORT_SYMBOL(pgtable_l5_enabled); phys_addr_t phys_ram_base __ro_after_init; EXPORT_SYMBOL(phys_ram_base); +#ifdef CONFIG_SPARSEMEM_VMEMMAP +#define VMEMMAP_ADDR_ALIGN (1ULL << SECTION_SIZE_BITS) + +unsigned long vmemmap_start_pfn __ro_after_init; +EXPORT_SYMBOL(vmemmap_start_pfn); +#endif + unsigned long empty_zero_page[PAGE_SIZE / sizeof(unsigned long)] __page_aligned_bss; EXPORT_SYMBOL(empty_zero_page); @@ -240,8 +248,12 @@ static void __init setup_bootmem(void) * Make sure we align the start of the memory on a PMD boundary so that * at worst, we map the linear mapping with PMD mappings. */ - if (!IS_ENABLED(CONFIG_XIP_KERNEL)) + if (!IS_ENABLED(CONFIG_XIP_KERNEL)) { phys_ram_base = memblock_start_of_DRAM() & PMD_MASK; +#ifdef CONFIG_SPARSEMEM_VMEMMAP + vmemmap_start_pfn = round_down(phys_ram_base, VMEMMAP_ADDR_ALIGN) >> PAGE_SHIFT; +#endif + } /* * In 64-bit, any use of __va/__pa before this point is wrong as we @@ -1101,6 +1113,9 @@ asmlinkage void __init setup_vm(uintptr_t dtb_pa) kernel_map.xiprom_sz = (uintptr_t)(&_exiprom) - (uintptr_t)(&_xiprom); phys_ram_base = CONFIG_PHYS_RAM_BASE; +#ifdef CONFIG_SPARSEMEM_VMEMMAP + vmemmap_start_pfn = round_down(phys_ram_base, VMEMMAP_ADDR_ALIGN) >> PAGE_SHIFT; +#endif kernel_map.phys_addr = (uintptr_t)CONFIG_PHYS_RAM_BASE; kernel_map.size = (uintptr_t)(&_end) - (uintptr_t)(&_start); diff --git a/arch/s390/kvm/interrupt.c b/arch/s390/kvm/interrupt.c index ea8dce299954..d4f031e086fc 100644 --- a/arch/s390/kvm/interrupt.c +++ b/arch/s390/kvm/interrupt.c @@ -2678,9 +2678,13 @@ static int flic_set_attr(struct kvm_device *dev, struct kvm_device_attr *attr) kvm_s390_clear_float_irqs(dev->kvm); break; case KVM_DEV_FLIC_APF_ENABLE: + if (kvm_is_ucontrol(dev->kvm)) + return -EINVAL; dev->kvm->arch.gmap->pfault_enabled = 1; break; case KVM_DEV_FLIC_APF_DISABLE_WAIT: + if (kvm_is_ucontrol(dev->kvm)) + return -EINVAL; dev->kvm->arch.gmap->pfault_enabled = 0; /* * Make sure no async faults are in transition when @@ -2894,6 +2898,8 @@ int kvm_set_routing_entry(struct kvm *kvm, switch (ue->type) { /* we store the userspace addresses instead of the guest addresses */ case KVM_IRQ_ROUTING_S390_ADAPTER: + if (kvm_is_ucontrol(kvm)) + return -EINVAL; e->set = set_adapter_int; uaddr = gmap_translate(kvm->arch.gmap, ue->u.adapter.summary_addr); if (uaddr == -EFAULT) diff --git a/arch/s390/kvm/vsie.c b/arch/s390/kvm/vsie.c index 150b9387860a..a687695d8f68 100644 --- a/arch/s390/kvm/vsie.c +++ b/arch/s390/kvm/vsie.c @@ -854,7 +854,7 @@ unpin: static void unpin_scb(struct kvm_vcpu *vcpu, struct vsie_page *vsie_page, gpa_t gpa) { - hpa_t hpa = (hpa_t) vsie_page->scb_o; + hpa_t hpa = virt_to_phys(vsie_page->scb_o); if (hpa) unpin_guest_page(vcpu->kvm, gpa, hpa); diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index 9d7bd0ae48c4..ef6cfea9df73 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -83,7 +83,6 @@ config X86 select ARCH_HAS_DMA_OPS if GART_IOMMU || XEN select ARCH_HAS_EARLY_DEBUG if KGDB select ARCH_HAS_ELF_RANDOMIZE - select ARCH_HAS_EXECMEM_ROX if X86_64 select ARCH_HAS_FAST_MULTIPLIER select ARCH_HAS_FORTIFY_SOURCE select ARCH_HAS_GCOV_PROFILE_ALL diff --git a/arch/x86/events/intel/core.c b/arch/x86/events/intel/core.c index 2e1e26846050..99c590da0ae2 100644 --- a/arch/x86/events/intel/core.c +++ b/arch/x86/events/intel/core.c @@ -429,6 +429,16 @@ static struct event_constraint intel_lnc_event_constraints[] = { EVENT_CONSTRAINT_END }; +static struct extra_reg intel_lnc_extra_regs[] __read_mostly = { + INTEL_UEVENT_EXTRA_REG(0x012a, MSR_OFFCORE_RSP_0, 0xfffffffffffull, RSP_0), + INTEL_UEVENT_EXTRA_REG(0x012b, MSR_OFFCORE_RSP_1, 0xfffffffffffull, RSP_1), + INTEL_UEVENT_PEBS_LDLAT_EXTRA_REG(0x01cd), + INTEL_UEVENT_EXTRA_REG(0x02c6, MSR_PEBS_FRONTEND, 0x9, FE), + INTEL_UEVENT_EXTRA_REG(0x03c6, MSR_PEBS_FRONTEND, 0x7fff1f, FE), + INTEL_UEVENT_EXTRA_REG(0x40ad, MSR_PEBS_FRONTEND, 0xf, FE), + INTEL_UEVENT_EXTRA_REG(0x04c2, MSR_PEBS_FRONTEND, 0x8, FE), + EVENT_EXTRA_END +}; EVENT_ATTR_STR(mem-loads, mem_ld_nhm, "event=0x0b,umask=0x10,ldlat=3"); EVENT_ATTR_STR(mem-loads, mem_ld_snb, "event=0xcd,umask=0x1,ldlat=3"); @@ -6422,7 +6432,7 @@ static __always_inline void intel_pmu_init_lnc(struct pmu *pmu) intel_pmu_init_glc(pmu); hybrid(pmu, event_constraints) = intel_lnc_event_constraints; hybrid(pmu, pebs_constraints) = intel_lnc_pebs_event_constraints; - hybrid(pmu, extra_regs) = intel_rwc_extra_regs; + hybrid(pmu, extra_regs) = intel_lnc_extra_regs; } static __always_inline void intel_pmu_init_skt(struct pmu *pmu) diff --git a/arch/x86/events/intel/ds.c b/arch/x86/events/intel/ds.c index 1a4b326ca2ce..6ba6549f26fa 100644 --- a/arch/x86/events/intel/ds.c +++ b/arch/x86/events/intel/ds.c @@ -2517,6 +2517,7 @@ void __init intel_ds_init(void) x86_pmu.large_pebs_flags |= PERF_SAMPLE_TIME; break; + case 6: case 5: x86_pmu.pebs_ept = 1; fallthrough; diff --git a/arch/x86/events/intel/uncore.c b/arch/x86/events/intel/uncore.c index d98fac567684..e7aba7349231 100644 --- a/arch/x86/events/intel/uncore.c +++ b/arch/x86/events/intel/uncore.c @@ -1910,6 +1910,7 @@ static const struct x86_cpu_id intel_uncore_match[] __initconst = { X86_MATCH_VFM(INTEL_ATOM_GRACEMONT, &adl_uncore_init), X86_MATCH_VFM(INTEL_ATOM_CRESTMONT_X, &gnr_uncore_init), X86_MATCH_VFM(INTEL_ATOM_CRESTMONT, &gnr_uncore_init), + X86_MATCH_VFM(INTEL_ATOM_DARKMONT_X, &gnr_uncore_init), {}, }; MODULE_DEVICE_TABLE(x86cpu, intel_uncore_match); diff --git a/arch/x86/include/asm/smap.h b/arch/x86/include/asm/smap.h index bab490379c65..2de1e5a75c57 100644 --- a/arch/x86/include/asm/smap.h +++ b/arch/x86/include/asm/smap.h @@ -13,30 +13,26 @@ #include <asm/cpufeatures.h> #include <asm/alternative.h> -/* "Raw" instruction opcodes */ -#define __ASM_CLAC ".byte 0x0f,0x01,0xca" -#define __ASM_STAC ".byte 0x0f,0x01,0xcb" - #ifdef __ASSEMBLY__ #define ASM_CLAC \ - ALTERNATIVE "", __ASM_CLAC, X86_FEATURE_SMAP + ALTERNATIVE "", "clac", X86_FEATURE_SMAP #define ASM_STAC \ - ALTERNATIVE "", __ASM_STAC, X86_FEATURE_SMAP + ALTERNATIVE "", "stac", X86_FEATURE_SMAP #else /* __ASSEMBLY__ */ static __always_inline void clac(void) { /* Note: a barrier is implicit in alternative() */ - alternative("", __ASM_CLAC, X86_FEATURE_SMAP); + alternative("", "clac", X86_FEATURE_SMAP); } static __always_inline void stac(void) { /* Note: a barrier is implicit in alternative() */ - alternative("", __ASM_STAC, X86_FEATURE_SMAP); + alternative("", "stac", X86_FEATURE_SMAP); } static __always_inline unsigned long smap_save(void) @@ -44,7 +40,7 @@ static __always_inline unsigned long smap_save(void) unsigned long flags; asm volatile ("# smap_save\n\t" - ALTERNATIVE("", "pushf; pop %0; " __ASM_CLAC "\n\t", + ALTERNATIVE("", "pushf; pop %0; " "clac" "\n\t", X86_FEATURE_SMAP) : "=rm" (flags) : : "memory", "cc"); @@ -61,9 +57,9 @@ static __always_inline void smap_restore(unsigned long flags) /* These macros can be used in asm() statements */ #define ASM_CLAC \ - ALTERNATIVE("", __ASM_CLAC, X86_FEATURE_SMAP) + ALTERNATIVE("", "clac", X86_FEATURE_SMAP) #define ASM_STAC \ - ALTERNATIVE("", __ASM_STAC, X86_FEATURE_SMAP) + ALTERNATIVE("", "stac", X86_FEATURE_SMAP) #endif /* __ASSEMBLY__ */ diff --git a/arch/x86/include/asm/special_insns.h b/arch/x86/include/asm/special_insns.h index aec6e2d3aa1d..98bfc097389c 100644 --- a/arch/x86/include/asm/special_insns.h +++ b/arch/x86/include/asm/special_insns.h @@ -217,7 +217,7 @@ fail: #define nop() asm volatile ("nop") -static inline void serialize(void) +static __always_inline void serialize(void) { /* Instruction opcode for SERIALIZE; supported in binutils >= 2.35. */ asm volatile(".byte 0xf, 0x1, 0xe8" ::: "memory"); diff --git a/arch/x86/include/asm/uaccess_64.h b/arch/x86/include/asm/uaccess_64.h index b0a887209400..c52f0133425b 100644 --- a/arch/x86/include/asm/uaccess_64.h +++ b/arch/x86/include/asm/uaccess_64.h @@ -63,13 +63,13 @@ static inline unsigned long __untagged_addr_remote(struct mm_struct *mm, */ static inline void __user *mask_user_address(const void __user *ptr) { - unsigned long mask; + void __user *ret; asm("cmp %1,%0\n\t" - "sbb %0,%0" - :"=r" (mask) - :"r" (ptr), - "0" (runtime_const_ptr(USER_PTR_MAX))); - return (__force void __user *)(mask | (__force unsigned long)ptr); + "cmova %1,%0" + :"=r" (ret) + :"r" (runtime_const_ptr(USER_PTR_MAX)), + "0" (ptr)); + return ret; } #define masked_user_access_begin(x) ({ \ __auto_type __masked_ptr = (x); \ diff --git a/arch/x86/kernel/cet.c b/arch/x86/kernel/cet.c index d2c732a34e5d..303bf74d175b 100644 --- a/arch/x86/kernel/cet.c +++ b/arch/x86/kernel/cet.c @@ -81,6 +81,34 @@ static void do_user_cp_fault(struct pt_regs *regs, unsigned long error_code) static __ro_after_init bool ibt_fatal = true; +/* + * By definition, all missing-ENDBRANCH #CPs are a result of WFE && !ENDBR. + * + * For the kernel IBT no ENDBR selftest where #CPs are deliberately triggered, + * the WFE state of the interrupted context needs to be cleared to let execution + * continue. Otherwise when the CPU resumes from the instruction that just + * caused the previous #CP, another missing-ENDBRANCH #CP is raised and the CPU + * enters a dead loop. + * + * This is not a problem with IDT because it doesn't preserve WFE and IRET doesn't + * set WFE. But FRED provides space on the entry stack (in an expanded CS area) + * to save and restore the WFE state, thus the WFE state is no longer clobbered, + * so software must clear it. + */ +static void ibt_clear_fred_wfe(struct pt_regs *regs) +{ + /* + * No need to do any FRED checks. + * + * For IDT event delivery, the high-order 48 bits of CS are pushed + * as 0s into the stack, and later IRET ignores these bits. + * + * For FRED, a test to check if fred_cs.wfe is set would be dropped + * by compilers. + */ + regs->fred_cs.wfe = 0; +} + static void do_kernel_cp_fault(struct pt_regs *regs, unsigned long error_code) { if ((error_code & CP_EC) != CP_ENDBR) { @@ -90,6 +118,7 @@ static void do_kernel_cp_fault(struct pt_regs *regs, unsigned long error_code) if (unlikely(regs->ip == (unsigned long)&ibt_selftest_noendbr)) { regs->ax = 0; + ibt_clear_fred_wfe(regs); return; } @@ -97,6 +126,7 @@ static void do_kernel_cp_fault(struct pt_regs *regs, unsigned long error_code) if (!ibt_fatal) { printk(KERN_DEFAULT CUT_HERE); __warn(__FILE__, __LINE__, (void *)regs->ip, TAINT_WARN, regs, NULL); + ibt_clear_fred_wfe(regs); return; } BUG(); diff --git a/arch/x86/kernel/fpu/regset.c b/arch/x86/kernel/fpu/regset.c index 6bc1eb2a21bd..887b0b8e21e3 100644 --- a/arch/x86/kernel/fpu/regset.c +++ b/arch/x86/kernel/fpu/regset.c @@ -190,7 +190,8 @@ int ssp_get(struct task_struct *target, const struct user_regset *regset, struct fpu *fpu = &target->thread.fpu; struct cet_user_state *cetregs; - if (!cpu_feature_enabled(X86_FEATURE_USER_SHSTK)) + if (!cpu_feature_enabled(X86_FEATURE_USER_SHSTK) || + !ssp_active(target, regset)) return -ENODEV; sync_fpstate(fpu); diff --git a/arch/x86/kernel/fred.c b/arch/x86/kernel/fred.c index 8d32c3f48abc..5e2cd1004980 100644 --- a/arch/x86/kernel/fred.c +++ b/arch/x86/kernel/fred.c @@ -50,7 +50,13 @@ void cpu_init_fred_exceptions(void) FRED_CONFIG_ENTRYPOINT(asm_fred_entrypoint_user)); wrmsrl(MSR_IA32_FRED_STKLVLS, 0); - wrmsrl(MSR_IA32_FRED_RSP0, 0); + + /* + * Ater a CPU offline/online cycle, the FRED RSP0 MSR should be + * resynchronized with its per-CPU cache. + */ + wrmsrl(MSR_IA32_FRED_RSP0, __this_cpu_read(fred_rsp0)); + wrmsrl(MSR_IA32_FRED_RSP1, 0); wrmsrl(MSR_IA32_FRED_RSP2, 0); wrmsrl(MSR_IA32_FRED_RSP3, 0); diff --git a/arch/x86/kernel/static_call.c b/arch/x86/kernel/static_call.c index 9eed0c144dad..9e51242ed125 100644 --- a/arch/x86/kernel/static_call.c +++ b/arch/x86/kernel/static_call.c @@ -175,7 +175,6 @@ EXPORT_SYMBOL_GPL(arch_static_call_transform); noinstr void __static_call_update_early(void *tramp, void *func) { BUG_ON(system_state != SYSTEM_BOOTING); - BUG_ON(!early_boot_irqs_disabled); BUG_ON(static_call_initialized); __text_gen_insn(tramp, JMP32_INSN_OPCODE, tramp, func, JMP32_INSN_SIZE); sync_core(); diff --git a/arch/x86/lib/getuser.S b/arch/x86/lib/getuser.S index 4357ec2a0bfc..89ecd57c9d42 100644 --- a/arch/x86/lib/getuser.S +++ b/arch/x86/lib/getuser.S @@ -44,9 +44,8 @@ .pushsection runtime_ptr_USER_PTR_MAX,"a" .long 1b - 8 - . .popsection - cmp %rax, %rdx - sbb %rdx, %rdx - or %rdx, %rax + cmp %rdx, %rax + cmova %rdx, %rax .else cmp $TASK_SIZE_MAX-\size+1, %eax jae .Lbad_get_user diff --git a/arch/x86/mm/init.c b/arch/x86/mm/init.c index c6d29f283001..62aa4d66a032 100644 --- a/arch/x86/mm/init.c +++ b/arch/x86/mm/init.c @@ -1080,7 +1080,8 @@ struct execmem_info __init *execmem_arch_setup(void) start = MODULES_VADDR + offset; - if (IS_ENABLED(CONFIG_ARCH_HAS_EXECMEM_ROX)) { + if (IS_ENABLED(CONFIG_ARCH_HAS_EXECMEM_ROX) && + cpu_feature_enabled(X86_FEATURE_PSE)) { pgprot = PAGE_KERNEL_ROX; flags = EXECMEM_KASAN_SHADOW | EXECMEM_ROX_CACHE; } else { diff --git a/block/bfq-iosched.c b/block/bfq-iosched.c index 068c63e95738..167542201603 100644 --- a/block/bfq-iosched.c +++ b/block/bfq-iosched.c @@ -6844,16 +6844,24 @@ static struct bfq_queue *bfq_waker_bfqq(struct bfq_queue *bfqq) if (new_bfqq == waker_bfqq) { /* * If waker_bfqq is in the merge chain, and current - * is the only procress. + * is the only process, waker_bfqq can be freed. */ if (bfqq_process_refs(waker_bfqq) == 1) return NULL; - break; + + return waker_bfqq; } new_bfqq = new_bfqq->new_bfqq; } + /* + * If waker_bfqq is not in the merge chain, and it's procress reference + * is 0, waker_bfqq can be freed. + */ + if (bfqq_process_refs(waker_bfqq) == 0) + return NULL; + return waker_bfqq; } diff --git a/drivers/acpi/acpi_video.c b/drivers/acpi/acpi_video.c index 8274a17872ed..a972831dbd66 100644 --- a/drivers/acpi/acpi_video.c +++ b/drivers/acpi/acpi_video.c @@ -610,16 +610,28 @@ acpi_video_device_lcd_get_level_current(struct acpi_video_device *device, return 0; } +/** + * acpi_video_device_EDID() - Get EDID from ACPI _DDC + * @device: video output device (LCD, CRT, ..) + * @edid: address for returned EDID pointer + * @length: _DDC length to request (must be a multiple of 128) + * + * Get EDID from ACPI _DDC. On success, a pointer to the EDID data is written + * to the @edid address, and the length of the EDID is returned. The caller is + * responsible for freeing the edid pointer. + * + * Return the length of EDID (positive value) on success or error (negative + * value). + */ static int -acpi_video_device_EDID(struct acpi_video_device *device, - union acpi_object **edid, int length) +acpi_video_device_EDID(struct acpi_video_device *device, void **edid, int length) { - int status; + acpi_status status; struct acpi_buffer buffer = { ACPI_ALLOCATE_BUFFER, NULL }; union acpi_object *obj; union acpi_object arg0 = { ACPI_TYPE_INTEGER }; struct acpi_object_list args = { 1, &arg0 }; - + int ret; *edid = NULL; @@ -636,16 +648,17 @@ acpi_video_device_EDID(struct acpi_video_device *device, obj = buffer.pointer; - if (obj && obj->type == ACPI_TYPE_BUFFER) - *edid = obj; - else { + if (obj && obj->type == ACPI_TYPE_BUFFER) { + *edid = kmemdup(obj->buffer.pointer, obj->buffer.length, GFP_KERNEL); + ret = *edid ? obj->buffer.length : -ENOMEM; + } else { acpi_handle_debug(device->dev->handle, "Invalid _DDC data for length %d\n", length); - status = -EFAULT; - kfree(obj); + ret = -EFAULT; } - return status; + kfree(obj); + return ret; } /* bus */ @@ -1435,9 +1448,7 @@ int acpi_video_get_edid(struct acpi_device *device, int type, int device_id, { struct acpi_video_bus *video; struct acpi_video_device *video_device; - union acpi_object *buffer = NULL; - acpi_status status; - int i, length; + int i, length, ret; if (!device || !acpi_driver_data(device)) return -EINVAL; @@ -1477,16 +1488,10 @@ int acpi_video_get_edid(struct acpi_device *device, int type, int device_id, } for (length = 512; length > 0; length -= 128) { - status = acpi_video_device_EDID(video_device, &buffer, - length); - if (ACPI_SUCCESS(status)) - break; + ret = acpi_video_device_EDID(video_device, edid, length); + if (ret > 0) + return ret; } - if (!length) - continue; - - *edid = buffer->buffer.pointer; - return length; } return -ENODEV; diff --git a/drivers/acpi/resource.c b/drivers/acpi/resource.c index 821867de43be..90aaec923889 100644 --- a/drivers/acpi/resource.c +++ b/drivers/acpi/resource.c @@ -441,6 +441,13 @@ static const struct dmi_system_id irq1_level_low_skip_override[] = { }, }, { + /* Asus Vivobook X1504VAP */ + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."), + DMI_MATCH(DMI_BOARD_NAME, "X1504VAP"), + }, + }, + { /* Asus Vivobook X1704VAP */ .matches = { DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."), @@ -646,6 +653,17 @@ static const struct dmi_system_id irq1_edge_low_force_override[] = { DMI_MATCH(DMI_BOARD_NAME, "GMxHGxx"), }, }, + { + /* + * TongFang GM5HG0A in case of the SKIKK Vanaheim relabel the + * board-name is changed, so check OEM strings instead. Note + * OEM string matches are always exact matches. + * https://bugzilla.kernel.org/show_bug.cgi?id=219614 + */ + .matches = { + DMI_EXACT_MATCH(DMI_OEM_STRING, "GM5HG0A"), + }, + }, { } }; @@ -671,11 +689,11 @@ static bool acpi_dev_irq_override(u32 gsi, u8 triggering, u8 polarity, for (i = 0; i < ARRAY_SIZE(override_table); i++) { const struct irq_override_cmp *entry = &override_table[i]; - if (dmi_check_system(entry->system) && - entry->irq == gsi && + if (entry->irq == gsi && entry->triggering == triggering && entry->polarity == polarity && - entry->shareable == shareable) + entry->shareable == shareable && + dmi_check_system(entry->system)) return entry->override; } diff --git a/drivers/base/topology.c b/drivers/base/topology.c index cf160dd2c27b..b962da263eee 100644 --- a/drivers/base/topology.c +++ b/drivers/base/topology.c @@ -27,9 +27,17 @@ static ssize_t name##_read(struct file *file, struct kobject *kobj, \ loff_t off, size_t count) \ { \ struct device *dev = kobj_to_dev(kobj); \ + cpumask_var_t mask; \ + ssize_t n; \ \ - return cpumap_print_bitmask_to_buf(buf, topology_##mask(dev->id), \ - off, count); \ + if (!alloc_cpumask_var(&mask, GFP_KERNEL)) \ + return -ENOMEM; \ + \ + cpumask_copy(mask, topology_##mask(dev->id)); \ + n = cpumap_print_bitmask_to_buf(buf, mask, off, count); \ + free_cpumask_var(mask); \ + \ + return n; \ } \ \ static ssize_t name##_list_read(struct file *file, struct kobject *kobj, \ @@ -37,9 +45,17 @@ static ssize_t name##_list_read(struct file *file, struct kobject *kobj, \ loff_t off, size_t count) \ { \ struct device *dev = kobj_to_dev(kobj); \ + cpumask_var_t mask; \ + ssize_t n; \ + \ + if (!alloc_cpumask_var(&mask, GFP_KERNEL)) \ + return -ENOMEM; \ + \ + cpumask_copy(mask, topology_##mask(dev->id)); \ + n = cpumap_print_list_to_buf(buf, mask, off, count); \ + free_cpumask_var(mask); \ \ - return cpumap_print_list_to_buf(buf, topology_##mask(dev->id), \ - off, count); \ + return n; \ } define_id_show_func(physical_package_id, "%d"); diff --git a/drivers/block/ublk_drv.c b/drivers/block/ublk_drv.c index 6c16cb798fdd..529085181f35 100644 --- a/drivers/block/ublk_drv.c +++ b/drivers/block/ublk_drv.c @@ -1618,6 +1618,21 @@ static void ublk_unquiesce_dev(struct ublk_device *ub) blk_mq_kick_requeue_list(ub->ub_disk->queue); } +static struct gendisk *ublk_detach_disk(struct ublk_device *ub) +{ + struct gendisk *disk; + + /* Sync with ublk_abort_queue() by holding the lock */ + spin_lock(&ub->lock); + disk = ub->ub_disk; + ub->dev_info.state = UBLK_S_DEV_DEAD; + ub->dev_info.ublksrv_pid = -1; + ub->ub_disk = NULL; + spin_unlock(&ub->lock); + + return disk; +} + static void ublk_stop_dev(struct ublk_device *ub) { struct gendisk *disk; @@ -1631,14 +1646,7 @@ static void ublk_stop_dev(struct ublk_device *ub) ublk_unquiesce_dev(ub); } del_gendisk(ub->ub_disk); - - /* Sync with ublk_abort_queue() by holding the lock */ - spin_lock(&ub->lock); - disk = ub->ub_disk; - ub->dev_info.state = UBLK_S_DEV_DEAD; - ub->dev_info.ublksrv_pid = -1; - ub->ub_disk = NULL; - spin_unlock(&ub->lock); + disk = ublk_detach_disk(ub); put_disk(disk); unlock: mutex_unlock(&ub->mutex); @@ -2335,7 +2343,7 @@ static int ublk_ctrl_start_dev(struct ublk_device *ub, struct io_uring_cmd *cmd) out_put_cdev: if (ret) { - ub->dev_info.state = UBLK_S_DEV_DEAD; + ublk_detach_disk(ub); ublk_put_device(ub); } if (ret) diff --git a/drivers/block/zram/zram_drv.c b/drivers/block/zram/zram_drv.c index 45df5eeabc5e..7903a4da40ac 100644 --- a/drivers/block/zram/zram_drv.c +++ b/drivers/block/zram/zram_drv.c @@ -1468,6 +1468,7 @@ static bool zram_meta_alloc(struct zram *zram, u64 disksize) zram->mem_pool = zs_create_pool(zram->disk->disk_name); if (!zram->mem_pool) { vfree(zram->table); + zram->table = NULL; return false; } diff --git a/drivers/bluetooth/btmtk.c b/drivers/bluetooth/btmtk.c index 7fd9d5ddce02..224eafc27dbe 100644 --- a/drivers/bluetooth/btmtk.c +++ b/drivers/bluetooth/btmtk.c @@ -1472,10 +1472,15 @@ EXPORT_SYMBOL_GPL(btmtk_usb_setup); int btmtk_usb_shutdown(struct hci_dev *hdev) { + struct btmtk_data *data = hci_get_priv(hdev); struct btmtk_hci_wmt_params wmt_params; u8 param = 0; int err; + err = usb_autopm_get_interface(data->intf); + if (err < 0) + return err; + /* Disable the device */ wmt_params.op = BTMTK_WMT_FUNC_CTRL; wmt_params.flag = 0; @@ -1486,9 +1491,11 @@ int btmtk_usb_shutdown(struct hci_dev *hdev) err = btmtk_usb_hci_wmt_sync(hdev, &wmt_params); if (err < 0) { bt_dev_err(hdev, "Failed to send wmt func ctrl (%d)", err); + usb_autopm_put_interface(data->intf); return err; } + usb_autopm_put_interface(data->intf); return 0; } EXPORT_SYMBOL_GPL(btmtk_usb_shutdown); diff --git a/drivers/bluetooth/btnxpuart.c b/drivers/bluetooth/btnxpuart.c index 569f5b7d6e46..1230045d78a5 100644 --- a/drivers/bluetooth/btnxpuart.c +++ b/drivers/bluetooth/btnxpuart.c @@ -1381,6 +1381,7 @@ static void btnxpuart_tx_work(struct work_struct *work) while ((skb = nxp_dequeue(nxpdev))) { len = serdev_device_write_buf(serdev, skb->data, skb->len); + serdev_device_wait_until_sent(serdev, 0); hdev->stat.byte_tx += len; skb_pull(skb, len); diff --git a/drivers/bus/mhi/host/pci_generic.c b/drivers/bus/mhi/host/pci_generic.c index 07645ce2119a..56ba4192c89c 100644 --- a/drivers/bus/mhi/host/pci_generic.c +++ b/drivers/bus/mhi/host/pci_generic.c @@ -917,7 +917,7 @@ static int mhi_pci_claim(struct mhi_controller *mhi_cntrl, return err; } - mhi_cntrl->regs = pcim_iomap_region(pdev, 1 << bar_num, pci_name(pdev)); + mhi_cntrl->regs = pcim_iomap_region(pdev, bar_num, pci_name(pdev)); if (IS_ERR(mhi_cntrl->regs)) { err = PTR_ERR(mhi_cntrl->regs); dev_err(&pdev->dev, "failed to map pci region: %d\n", err); diff --git a/drivers/cdrom/cdrom.c b/drivers/cdrom/cdrom.c index 6a99a459b80b..51745ed1bbab 100644 --- a/drivers/cdrom/cdrom.c +++ b/drivers/cdrom/cdrom.c @@ -1106,7 +1106,7 @@ int open_for_data(struct cdrom_device_info *cdi) } } - cd_dbg(CD_OPEN, "all seems well, opening the devicen"); + cd_dbg(CD_OPEN, "all seems well, opening the device\n"); /* all seems well, we can open the device */ ret = cdo->open(cdi, 0); /* open for data */ diff --git a/drivers/clk/imx/clk-imx8mp-audiomix.c b/drivers/clk/imx/clk-imx8mp-audiomix.c index b2cb157703c5..c409fc7e0618 100644 --- a/drivers/clk/imx/clk-imx8mp-audiomix.c +++ b/drivers/clk/imx/clk-imx8mp-audiomix.c @@ -278,7 +278,8 @@ static int clk_imx8mp_audiomix_reset_controller_register(struct device *dev, #else /* !CONFIG_RESET_CONTROLLER */ -static int clk_imx8mp_audiomix_reset_controller_register(struct clk_imx8mp_audiomix_priv *priv) +static int clk_imx8mp_audiomix_reset_controller_register(struct device *dev, + struct clk_imx8mp_audiomix_priv *priv) { return 0; } diff --git a/drivers/clk/thead/clk-th1520-ap.c b/drivers/clk/thead/clk-th1520-ap.c index 17e32ae08720..1015fab95251 100644 --- a/drivers/clk/thead/clk-th1520-ap.c +++ b/drivers/clk/thead/clk-th1520-ap.c @@ -779,6 +779,13 @@ static struct ccu_div dpu1_clk = { }, }; +static CLK_FIXED_FACTOR_HW(emmc_sdio_ref_clk, "emmc-sdio-ref", + &video_pll_clk.common.hw, 4, 1, 0); + +static const struct clk_parent_data emmc_sdio_ref_clk_pd[] = { + { .hw = &emmc_sdio_ref_clk.hw }, +}; + static CCU_GATE(CLK_BROM, brom_clk, "brom", ahb2_cpusys_hclk_pd, 0x100, BIT(4), 0); static CCU_GATE(CLK_BMU, bmu_clk, "bmu", axi4_cpusys2_aclk_pd, 0x100, BIT(5), 0); static CCU_GATE(CLK_AON2CPU_A2X, aon2cpu_a2x_clk, "aon2cpu-a2x", axi4_cpusys2_aclk_pd, @@ -798,7 +805,7 @@ static CCU_GATE(CLK_PERISYS_APB4_HCLK, perisys_apb4_hclk, "perisys-apb4-hclk", p 0x150, BIT(12), 0); static CCU_GATE(CLK_NPU_AXI, npu_axi_clk, "npu-axi", axi_aclk_pd, 0x1c8, BIT(5), 0); static CCU_GATE(CLK_CPU2VP, cpu2vp_clk, "cpu2vp", axi_aclk_pd, 0x1e0, BIT(13), 0); -static CCU_GATE(CLK_EMMC_SDIO, emmc_sdio_clk, "emmc-sdio", video_pll_clk_pd, 0x204, BIT(30), 0); +static CCU_GATE(CLK_EMMC_SDIO, emmc_sdio_clk, "emmc-sdio", emmc_sdio_ref_clk_pd, 0x204, BIT(30), 0); static CCU_GATE(CLK_GMAC1, gmac1_clk, "gmac1", gmac_pll_clk_pd, 0x204, BIT(26), 0); static CCU_GATE(CLK_PADCTRL1, padctrl1_clk, "padctrl1", perisys_apb_pclk_pd, 0x204, BIT(24), 0); static CCU_GATE(CLK_DSMART, dsmart_clk, "dsmart", perisys_apb_pclk_pd, 0x204, BIT(23), 0); @@ -1059,6 +1066,10 @@ static int th1520_clk_probe(struct platform_device *pdev) return ret; priv->hws[CLK_PLL_GMAC_100M] = &gmac_pll_clk_100m.hw; + ret = devm_clk_hw_register(dev, &emmc_sdio_ref_clk.hw); + if (ret) + return ret; + ret = devm_of_clk_add_hw_provider(dev, of_clk_hw_onecell_get, priv); if (ret) return ret; diff --git a/drivers/cpufreq/Kconfig b/drivers/cpufreq/Kconfig index 92a83a9bb2e1..26e98fea991a 100644 --- a/drivers/cpufreq/Kconfig +++ b/drivers/cpufreq/Kconfig @@ -325,8 +325,6 @@ config QORIQ_CPUFREQ This adds the CPUFreq driver support for Freescale QorIQ SoCs which are capable of changing the CPU's frequency dynamically. -endif - config ACPI_CPPC_CPUFREQ tristate "CPUFreq driver based on the ACPI CPPC spec" depends on ACPI_PROCESSOR @@ -355,4 +353,6 @@ config ACPI_CPPC_CPUFREQ_FIE If in doubt, say N. +endif + endmenu diff --git a/drivers/cpuidle/cpuidle-riscv-sbi.c b/drivers/cpuidle/cpuidle-riscv-sbi.c index 14462c092039..0c92a628bbd4 100644 --- a/drivers/cpuidle/cpuidle-riscv-sbi.c +++ b/drivers/cpuidle/cpuidle-riscv-sbi.c @@ -504,12 +504,12 @@ static int sbi_cpuidle_probe(struct platform_device *pdev) int cpu, ret; struct cpuidle_driver *drv; struct cpuidle_device *dev; - struct device_node *np, *pds_node; + struct device_node *pds_node; /* Detect OSI support based on CPU DT nodes */ sbi_cpuidle_use_osi = true; for_each_possible_cpu(cpu) { - np = of_cpu_device_node_get(cpu); + struct device_node *np __free(device_node) = of_cpu_device_node_get(cpu); if (np && of_property_present(np, "power-domains") && of_property_present(np, "power-domain-names")) { diff --git a/drivers/cpuidle/governors/teo.c b/drivers/cpuidle/governors/teo.c index f2992f92d8db..173ddcac540a 100644 --- a/drivers/cpuidle/governors/teo.c +++ b/drivers/cpuidle/governors/teo.c @@ -10,25 +10,27 @@ * DOC: teo-description * * The idea of this governor is based on the observation that on many systems - * timer events are two or more orders of magnitude more frequent than any - * other interrupts, so they are likely to be the most significant cause of CPU - * wakeups from idle states. Moreover, information about what happened in the - * (relatively recent) past can be used to estimate whether or not the deepest - * idle state with target residency within the (known) time till the closest - * timer event, referred to as the sleep length, is likely to be suitable for - * the upcoming CPU idle period and, if not, then which of the shallower idle - * states to choose instead of it. + * timer interrupts are two or more orders of magnitude more frequent than any + * other interrupt types, so they are likely to dominate CPU wakeup patterns. + * Moreover, in principle, the time when the next timer event is going to occur + * can be determined at the idle state selection time, although doing that may + * be costly, so it can be regarded as the most reliable source of information + * for idle state selection. * - * Of course, non-timer wakeup sources are more important in some use cases - * which can be covered by taking a few most recent idle time intervals of the - * CPU into account. However, even in that context it is not necessary to - * consider idle duration values greater than the sleep length, because the - * closest timer will ultimately wake up the CPU anyway unless it is woken up - * earlier. + * Of course, non-timer wakeup sources are more important in some use cases, + * but even then it is generally unnecessary to consider idle duration values + * greater than the time time till the next timer event, referred as the sleep + * length in what follows, because the closest timer will ultimately wake up the + * CPU anyway unless it is woken up earlier. * - * Thus this governor estimates whether or not the prospective idle duration of - * a CPU is likely to be significantly shorter than the sleep length and selects - * an idle state for it accordingly. + * However, since obtaining the sleep length may be costly, the governor first + * checks if it can select a shallow idle state using wakeup pattern information + * from recent times, in which case it can do without knowing the sleep length + * at all. For this purpose, it counts CPU wakeup events and looks for an idle + * state whose target residency has not exceeded the idle duration (measured + * after wakeup) in the majority of relevant recent cases. If the target + * residency of that state is small enough, it may be used right away and the + * sleep length need not be determined. * * The computations carried out by this governor are based on using bins whose * boundaries are aligned with the target residency parameter values of the CPU @@ -39,7 +41,11 @@ * idle state 2, the third bin spans from the target residency of idle state 2 * up to, but not including, the target residency of idle state 3 and so on. * The last bin spans from the target residency of the deepest idle state - * supplied by the driver to infinity. + * supplied by the driver to the scheduler tick period length or to infinity if + * the tick period length is less than the target residency of that state. In + * the latter case, the governor also counts events with the measured idle + * duration between the tick period length and the target residency of the + * deepest idle state. * * Two metrics called "hits" and "intercepts" are associated with each bin. * They are updated every time before selecting an idle state for the given CPU @@ -49,47 +55,46 @@ * sleep length and the idle duration measured after CPU wakeup fall into the * same bin (that is, the CPU appears to wake up "on time" relative to the sleep * length). In turn, the "intercepts" metric reflects the relative frequency of - * situations in which the measured idle duration is so much shorter than the - * sleep length that the bin it falls into corresponds to an idle state - * shallower than the one whose bin is fallen into by the sleep length (these - * situations are referred to as "intercepts" below). + * non-timer wakeup events for which the measured idle duration falls into a bin + * that corresponds to an idle state shallower than the one whose bin is fallen + * into by the sleep length (these events are also referred to as "intercepts" + * below). * * In order to select an idle state for a CPU, the governor takes the following * steps (modulo the possible latency constraint that must be taken into account * too): * - * 1. Find the deepest CPU idle state whose target residency does not exceed - * the current sleep length (the candidate idle state) and compute 2 sums as - * follows: + * 1. Find the deepest enabled CPU idle state (the candidate idle state) and + * compute 2 sums as follows: * - * - The sum of the "hits" and "intercepts" metrics for the candidate state - * and all of the deeper idle states (it represents the cases in which the - * CPU was idle long enough to avoid being intercepted if the sleep length - * had been equal to the current one). + * - The sum of the "hits" metric for all of the idle states shallower than + * the candidate one (it represents the cases in which the CPU was likely + * woken up by a timer). * - * - The sum of the "intercepts" metrics for all of the idle states shallower - * than the candidate one (it represents the cases in which the CPU was not - * idle long enough to avoid being intercepted if the sleep length had been - * equal to the current one). + * - The sum of the "intercepts" metric for all of the idle states shallower + * than the candidate one (it represents the cases in which the CPU was + * likely woken up by a non-timer wakeup source). * - * 2. If the second sum is greater than the first one the CPU is likely to wake - * up early, so look for an alternative idle state to select. + * 2. If the second sum computed in step 1 is greater than a half of the sum of + * both metrics for the candidate state bin and all subsequent bins(if any), + * a shallower idle state is likely to be more suitable, so look for it. * - * - Traverse the idle states shallower than the candidate one in the + * - Traverse the enabled idle states shallower than the candidate one in the * descending order. * * - For each of them compute the sum of the "intercepts" metrics over all * of the idle states between it and the candidate one (including the * former and excluding the latter). * - * - If each of these sums that needs to be taken into account (because the - * check related to it has indicated that the CPU is likely to wake up - * early) is greater than a half of the corresponding sum computed in step - * 1 (which means that the target residency of the state in question had - * not exceeded the idle duration in over a half of the relevant cases), - * select the given idle state instead of the candidate one. + * - If this sum is greater than a half of the second sum computed in step 1, + * use the given idle state as the new candidate one. * - * 3. By default, select the candidate state. + * 3. If the current candidate state is state 0 or its target residency is short + * enough, return it and prevent the scheduler tick from being stopped. + * + * 4. Obtain the sleep length value and check if it is below the target + * residency of the current candidate state, in which case a new shallower + * candidate state needs to be found, so look for it. */ #include <linux/cpuidle.h> diff --git a/drivers/crypto/ccp/sev-dev.c b/drivers/crypto/ccp/sev-dev.c index af018afd9cd7..2e87ca0e292a 100644 --- a/drivers/crypto/ccp/sev-dev.c +++ b/drivers/crypto/ccp/sev-dev.c @@ -249,7 +249,7 @@ static struct file *open_file_as_root(const char *filename, int flags, umode_t m fp = file_open_root(&root, filename, flags, mode); path_put(&root); - revert_creds(old_cred); + put_cred(revert_creds(old_cred)); return fp; } diff --git a/drivers/dma/amd/qdma/qdma.c b/drivers/dma/amd/qdma/qdma.c index 6d9079458fe9..66f00ad67351 100644 --- a/drivers/dma/amd/qdma/qdma.c +++ b/drivers/dma/amd/qdma/qdma.c @@ -7,9 +7,9 @@ #include <linux/bitfield.h> #include <linux/bitops.h> #include <linux/dmaengine.h> +#include <linux/dma-mapping.h> #include <linux/module.h> #include <linux/mod_devicetable.h> -#include <linux/dma-map-ops.h> #include <linux/platform_device.h> #include <linux/platform_data/amd_qdma.h> #include <linux/regmap.h> @@ -492,18 +492,9 @@ static int qdma_device_verify(struct qdma_device *qdev) static int qdma_device_setup(struct qdma_device *qdev) { - struct device *dev = &qdev->pdev->dev; u32 ring_sz = QDMA_DEFAULT_RING_SIZE; int ret = 0; - while (dev && get_dma_ops(dev)) - dev = dev->parent; - if (!dev) { - qdma_err(qdev, "dma device not found"); - return -EINVAL; - } - set_dma_ops(&qdev->pdev->dev, get_dma_ops(dev)); - ret = qdma_setup_fmap_context(qdev); if (ret) { qdma_err(qdev, "Failed setup fmap context"); @@ -548,11 +539,12 @@ static void qdma_free_queue_resources(struct dma_chan *chan) { struct qdma_queue *queue = to_qdma_queue(chan); struct qdma_device *qdev = queue->qdev; - struct device *dev = qdev->dma_dev.dev; + struct qdma_platdata *pdata; qdma_clear_queue_context(queue); vchan_free_chan_resources(&queue->vchan); - dma_free_coherent(dev, queue->ring_size * QDMA_MM_DESC_SIZE, + pdata = dev_get_platdata(&qdev->pdev->dev); + dma_free_coherent(pdata->dma_dev, queue->ring_size * QDMA_MM_DESC_SIZE, queue->desc_base, queue->dma_desc_base); } @@ -565,6 +557,7 @@ static int qdma_alloc_queue_resources(struct dma_chan *chan) struct qdma_queue *queue = to_qdma_queue(chan); struct qdma_device *qdev = queue->qdev; struct qdma_ctxt_sw_desc desc; + struct qdma_platdata *pdata; size_t size; int ret; @@ -572,8 +565,9 @@ static int qdma_alloc_queue_resources(struct dma_chan *chan) if (ret) return ret; + pdata = dev_get_platdata(&qdev->pdev->dev); size = queue->ring_size * QDMA_MM_DESC_SIZE; - queue->desc_base = dma_alloc_coherent(qdev->dma_dev.dev, size, + queue->desc_base = dma_alloc_coherent(pdata->dma_dev, size, &queue->dma_desc_base, GFP_KERNEL); if (!queue->desc_base) { @@ -588,7 +582,7 @@ static int qdma_alloc_queue_resources(struct dma_chan *chan) if (ret) { qdma_err(qdev, "Failed to setup SW desc ctxt for %s", chan->name); - dma_free_coherent(qdev->dma_dev.dev, size, queue->desc_base, + dma_free_coherent(pdata->dma_dev, size, queue->desc_base, queue->dma_desc_base); return ret; } @@ -948,8 +942,9 @@ static int qdma_init_error_irq(struct qdma_device *qdev) static int qdmam_alloc_qintr_rings(struct qdma_device *qdev) { - u32 ctxt[QDMA_CTXT_REGMAP_LEN]; + struct qdma_platdata *pdata = dev_get_platdata(&qdev->pdev->dev); struct device *dev = &qdev->pdev->dev; + u32 ctxt[QDMA_CTXT_REGMAP_LEN]; struct qdma_intr_ring *ring; struct qdma_ctxt_intr intr_ctxt; u32 vector; @@ -969,7 +964,8 @@ static int qdmam_alloc_qintr_rings(struct qdma_device *qdev) ring->msix_id = qdev->err_irq_idx + i + 1; ring->ridx = i; ring->color = 1; - ring->base = dmam_alloc_coherent(dev, QDMA_INTR_RING_SIZE, + ring->base = dmam_alloc_coherent(pdata->dma_dev, + QDMA_INTR_RING_SIZE, &ring->dev_base, GFP_KERNEL); if (!ring->base) { qdma_err(qdev, "Failed to alloc intr ring %d", i); diff --git a/drivers/dma/apple-admac.c b/drivers/dma/apple-admac.c index c499173d80b2..bd49f0374291 100644 --- a/drivers/dma/apple-admac.c +++ b/drivers/dma/apple-admac.c @@ -153,6 +153,8 @@ static int admac_alloc_sram_carveout(struct admac_data *ad, { struct admac_sram *sram; int i, ret = 0, nblocks; + ad->txcache.size = readl_relaxed(ad->base + REG_TX_SRAM_SIZE); + ad->rxcache.size = readl_relaxed(ad->base + REG_RX_SRAM_SIZE); if (dir == DMA_MEM_TO_DEV) sram = &ad->txcache; @@ -912,12 +914,7 @@ static int admac_probe(struct platform_device *pdev) goto free_irq; } - ad->txcache.size = readl_relaxed(ad->base + REG_TX_SRAM_SIZE); - ad->rxcache.size = readl_relaxed(ad->base + REG_RX_SRAM_SIZE); - dev_info(&pdev->dev, "Audio DMA Controller\n"); - dev_info(&pdev->dev, "imprint %x TX cache %u RX cache %u\n", - readl_relaxed(ad->base + REG_IMPRINT), ad->txcache.size, ad->rxcache.size); return 0; diff --git a/drivers/dma/at_xdmac.c b/drivers/dma/at_xdmac.c index 9c7b40220004..ba25c23164e7 100644 --- a/drivers/dma/at_xdmac.c +++ b/drivers/dma/at_xdmac.c @@ -1363,6 +1363,8 @@ at_xdmac_prep_dma_memset(struct dma_chan *chan, dma_addr_t dest, int value, return NULL; desc = at_xdmac_memset_create_desc(chan, atchan, dest, len, value); + if (!desc) + return NULL; list_add_tail(&desc->desc_node, &desc->descs_list); desc->tx_dma_desc.cookie = -EBUSY; diff --git a/drivers/dma/dw/acpi.c b/drivers/dma/dw/acpi.c index c510c109d2c3..b6452fffa657 100644 --- a/drivers/dma/dw/acpi.c +++ b/drivers/dma/dw/acpi.c @@ -8,13 +8,15 @@ static bool dw_dma_acpi_filter(struct dma_chan *chan, void *param) { + struct dw_dma *dw = to_dw_dma(chan->device); + struct dw_dma_chip_pdata *data = dev_get_drvdata(dw->dma.dev); struct acpi_dma_spec *dma_spec = param; struct dw_dma_slave slave = { .dma_dev = dma_spec->dev, .src_id = dma_spec->slave_id, .dst_id = dma_spec->slave_id, - .m_master = 0, - .p_master = 1, + .m_master = data->m_master, + .p_master = data->p_master, }; return dw_dma_filter(chan, &slave); diff --git a/drivers/dma/dw/internal.h b/drivers/dma/dw/internal.h index 563ce73488db..f1bd06a20cd6 100644 --- a/drivers/dma/dw/internal.h +++ b/drivers/dma/dw/internal.h @@ -51,11 +51,15 @@ struct dw_dma_chip_pdata { int (*probe)(struct dw_dma_chip *chip); int (*remove)(struct dw_dma_chip *chip); struct dw_dma_chip *chip; + u8 m_master; + u8 p_master; }; static __maybe_unused const struct dw_dma_chip_pdata dw_dma_chip_pdata = { .probe = dw_dma_probe, .remove = dw_dma_remove, + .m_master = 0, + .p_master = 1, }; static const struct dw_dma_platform_data idma32_pdata = { @@ -72,6 +76,8 @@ static __maybe_unused const struct dw_dma_chip_pdata idma32_chip_pdata = { .pdata = &idma32_pdata, .probe = idma32_dma_probe, .remove = idma32_dma_remove, + .m_master = 0, + .p_master = 0, }; static const struct dw_dma_platform_data xbar_pdata = { @@ -88,6 +94,8 @@ static __maybe_unused const struct dw_dma_chip_pdata xbar_chip_pdata = { .pdata = &xbar_pdata, .probe = idma32_dma_probe, .remove = idma32_dma_remove, + .m_master = 0, + .p_master = 0, }; #endif /* _DMA_DW_INTERNAL_H */ diff --git a/drivers/dma/dw/pci.c b/drivers/dma/dw/pci.c index ad2d4d012cf7..e8a0eb81726a 100644 --- a/drivers/dma/dw/pci.c +++ b/drivers/dma/dw/pci.c @@ -56,10 +56,10 @@ static int dw_pci_probe(struct pci_dev *pdev, const struct pci_device_id *pid) if (ret) return ret; - dw_dma_acpi_controller_register(chip->dw); - pci_set_drvdata(pdev, data); + dw_dma_acpi_controller_register(chip->dw); + return 0; } diff --git a/drivers/dma/fsl-edma-common.h b/drivers/dma/fsl-edma-common.h index ce37e1ee9c46..fe8f103d4a63 100644 --- a/drivers/dma/fsl-edma-common.h +++ b/drivers/dma/fsl-edma-common.h @@ -166,6 +166,7 @@ struct fsl_edma_chan { struct work_struct issue_worker; struct platform_device *pdev; struct device *pd_dev; + struct device_link *pd_dev_link; u32 srcid; struct clk *clk; int priority; diff --git a/drivers/dma/fsl-edma-main.c b/drivers/dma/fsl-edma-main.c index 60de1003193a..1a613236b3e4 100644 --- a/drivers/dma/fsl-edma-main.c +++ b/drivers/dma/fsl-edma-main.c @@ -417,10 +417,33 @@ static const struct of_device_id fsl_edma_dt_ids[] = { }; MODULE_DEVICE_TABLE(of, fsl_edma_dt_ids); +static void fsl_edma3_detach_pd(struct fsl_edma_engine *fsl_edma) +{ + struct fsl_edma_chan *fsl_chan; + int i; + + for (i = 0; i < fsl_edma->n_chans; i++) { + if (fsl_edma->chan_masked & BIT(i)) + continue; + fsl_chan = &fsl_edma->chans[i]; + if (fsl_chan->pd_dev_link) + device_link_del(fsl_chan->pd_dev_link); + if (fsl_chan->pd_dev) { + dev_pm_domain_detach(fsl_chan->pd_dev, false); + pm_runtime_dont_use_autosuspend(fsl_chan->pd_dev); + pm_runtime_set_suspended(fsl_chan->pd_dev); + } + } +} + +static void devm_fsl_edma3_detach_pd(void *data) +{ + fsl_edma3_detach_pd(data); +} + static int fsl_edma3_attach_pd(struct platform_device *pdev, struct fsl_edma_engine *fsl_edma) { struct fsl_edma_chan *fsl_chan; - struct device_link *link; struct device *pd_chan; struct device *dev; int i; @@ -436,15 +459,16 @@ static int fsl_edma3_attach_pd(struct platform_device *pdev, struct fsl_edma_eng pd_chan = dev_pm_domain_attach_by_id(dev, i); if (IS_ERR_OR_NULL(pd_chan)) { dev_err(dev, "Failed attach pd %d\n", i); - return -EINVAL; + goto detach; } - link = device_link_add(dev, pd_chan, DL_FLAG_STATELESS | + fsl_chan->pd_dev_link = device_link_add(dev, pd_chan, DL_FLAG_STATELESS | DL_FLAG_PM_RUNTIME | DL_FLAG_RPM_ACTIVE); - if (!link) { + if (!fsl_chan->pd_dev_link) { dev_err(dev, "Failed to add device_link to %d\n", i); - return -EINVAL; + dev_pm_domain_detach(pd_chan, false); + goto detach; } fsl_chan->pd_dev = pd_chan; @@ -455,6 +479,10 @@ static int fsl_edma3_attach_pd(struct platform_device *pdev, struct fsl_edma_eng } return 0; + +detach: + fsl_edma3_detach_pd(fsl_edma); + return -EINVAL; } static int fsl_edma_probe(struct platform_device *pdev) @@ -544,6 +572,9 @@ static int fsl_edma_probe(struct platform_device *pdev) ret = fsl_edma3_attach_pd(pdev, fsl_edma); if (ret) return ret; + ret = devm_add_action_or_reset(&pdev->dev, devm_fsl_edma3_detach_pd, fsl_edma); + if (ret) + return ret; } if (drvdata->flags & FSL_EDMA_DRV_TCD64) diff --git a/drivers/dma/loongson2-apb-dma.c b/drivers/dma/loongson2-apb-dma.c index 367ed34ce4da..c528f02b9f84 100644 --- a/drivers/dma/loongson2-apb-dma.c +++ b/drivers/dma/loongson2-apb-dma.c @@ -31,7 +31,7 @@ #define LDMA_ASK_VALID BIT(2) #define LDMA_START BIT(3) /* DMA start operation */ #define LDMA_STOP BIT(4) /* DMA stop operation */ -#define LDMA_CONFIG_MASK GENMASK(4, 0) /* DMA controller config bits mask */ +#define LDMA_CONFIG_MASK GENMASK_ULL(4, 0) /* DMA controller config bits mask */ /* Bitfields in ndesc_addr field of HW descriptor */ #define LDMA_DESC_EN BIT(0) /*1: The next descriptor is valid */ diff --git a/drivers/dma/mv_xor.c b/drivers/dma/mv_xor.c index 43efce77bb57..40b76b40bc30 100644 --- a/drivers/dma/mv_xor.c +++ b/drivers/dma/mv_xor.c @@ -1388,6 +1388,7 @@ static int mv_xor_probe(struct platform_device *pdev) irq = irq_of_parse_and_map(np, 0); if (!irq) { ret = -ENODEV; + of_node_put(np); goto err_channel_add; } @@ -1396,6 +1397,7 @@ static int mv_xor_probe(struct platform_device *pdev) if (IS_ERR(chan)) { ret = PTR_ERR(chan); irq_dispose_mapping(irq); + of_node_put(np); goto err_channel_add; } diff --git a/drivers/dma/tegra186-gpc-dma.c b/drivers/dma/tegra186-gpc-dma.c index cacf3757adc2..4d6fe0efa76e 100644 --- a/drivers/dma/tegra186-gpc-dma.c +++ b/drivers/dma/tegra186-gpc-dma.c @@ -231,6 +231,7 @@ struct tegra_dma_channel { bool config_init; char name[30]; enum dma_transfer_direction sid_dir; + enum dma_status status; int id; int irq; int slave_id; @@ -393,6 +394,8 @@ static int tegra_dma_pause(struct tegra_dma_channel *tdc) tegra_dma_dump_chan_regs(tdc); } + tdc->status = DMA_PAUSED; + return ret; } @@ -419,6 +422,8 @@ static void tegra_dma_resume(struct tegra_dma_channel *tdc) val = tdc_read(tdc, TEGRA_GPCDMA_CHAN_CSRE); val &= ~TEGRA_GPCDMA_CHAN_CSRE_PAUSE; tdc_write(tdc, TEGRA_GPCDMA_CHAN_CSRE, val); + + tdc->status = DMA_IN_PROGRESS; } static int tegra_dma_device_resume(struct dma_chan *dc) @@ -544,6 +549,7 @@ static void tegra_dma_xfer_complete(struct tegra_dma_channel *tdc) tegra_dma_sid_free(tdc); tdc->dma_desc = NULL; + tdc->status = DMA_COMPLETE; } static void tegra_dma_chan_decode_error(struct tegra_dma_channel *tdc, @@ -716,6 +722,7 @@ static int tegra_dma_terminate_all(struct dma_chan *dc) tdc->dma_desc = NULL; } + tdc->status = DMA_COMPLETE; tegra_dma_sid_free(tdc); vchan_get_all_descriptors(&tdc->vc, &head); spin_unlock_irqrestore(&tdc->vc.lock, flags); @@ -769,6 +776,9 @@ static enum dma_status tegra_dma_tx_status(struct dma_chan *dc, if (ret == DMA_COMPLETE) return ret; + if (tdc->status == DMA_PAUSED) + ret = DMA_PAUSED; + spin_lock_irqsave(&tdc->vc.lock, flags); vd = vchan_find_desc(&tdc->vc, cookie); if (vd) { diff --git a/drivers/gpio/gpio-loongson-64bit.c b/drivers/gpio/gpio-loongson-64bit.c index 6749d4dd6d64..7f4d78fd800e 100644 --- a/drivers/gpio/gpio-loongson-64bit.c +++ b/drivers/gpio/gpio-loongson-64bit.c @@ -237,9 +237,9 @@ static const struct loongson_gpio_chip_data loongson_gpio_ls2k2000_data1 = { static const struct loongson_gpio_chip_data loongson_gpio_ls2k2000_data2 = { .label = "ls2k2000_gpio", .mode = BIT_CTRL_MODE, - .conf_offset = 0x84, - .in_offset = 0x88, - .out_offset = 0x80, + .conf_offset = 0x4, + .in_offset = 0x8, + .out_offset = 0x0, }; static const struct loongson_gpio_chip_data loongson_gpio_ls3a5000_data = { diff --git a/drivers/gpio/gpio-sim.c b/drivers/gpio/gpio-sim.c index f387dad81f29..686ae3d11ba3 100644 --- a/drivers/gpio/gpio-sim.c +++ b/drivers/gpio/gpio-sim.c @@ -1027,6 +1027,30 @@ static void gpio_sim_device_deactivate(struct gpio_sim_device *dev) dev->pdev = NULL; } +static void +gpio_sim_device_lockup_configfs(struct gpio_sim_device *dev, bool lock) +{ + struct configfs_subsystem *subsys = dev->group.cg_subsys; + struct gpio_sim_bank *bank; + struct gpio_sim_line *line; + + /* + * The device only needs to depend on leaf line entries. This is + * sufficient to lock up all the configfs entries that the + * instantiated, alive device depends on. + */ + list_for_each_entry(bank, &dev->bank_list, siblings) { + list_for_each_entry(line, &bank->line_list, siblings) { + if (lock) + WARN_ON(configfs_depend_item_unlocked( + subsys, &line->group.cg_item)); + else + configfs_undepend_item_unlocked( + &line->group.cg_item); + } + } +} + static ssize_t gpio_sim_device_config_live_store(struct config_item *item, const char *page, size_t count) @@ -1039,14 +1063,24 @@ gpio_sim_device_config_live_store(struct config_item *item, if (ret) return ret; - guard(mutex)(&dev->lock); + if (live) + gpio_sim_device_lockup_configfs(dev, true); - if (live == gpio_sim_device_is_live(dev)) - ret = -EPERM; - else if (live) - ret = gpio_sim_device_activate(dev); - else - gpio_sim_device_deactivate(dev); + scoped_guard(mutex, &dev->lock) { + if (live == gpio_sim_device_is_live(dev)) + ret = -EPERM; + else if (live) + ret = gpio_sim_device_activate(dev); + else + gpio_sim_device_deactivate(dev); + } + + /* + * Undepend is required only if device disablement (live == 0) + * succeeds or if device enablement (live == 1) fails. + */ + if (live == !!ret) + gpio_sim_device_lockup_configfs(dev, false); return ret ?: count; } diff --git a/drivers/gpio/gpio-virtuser.c b/drivers/gpio/gpio-virtuser.c index 91b6352c957c..e89f299f2140 100644 --- a/drivers/gpio/gpio-virtuser.c +++ b/drivers/gpio/gpio-virtuser.c @@ -1410,7 +1410,7 @@ gpio_virtuser_make_lookup_table(struct gpio_virtuser_device *dev) size_t num_entries = gpio_virtuser_get_lookup_count(dev); struct gpio_virtuser_lookup_entry *entry; struct gpio_virtuser_lookup *lookup; - unsigned int i = 0; + unsigned int i = 0, idx; lockdep_assert_held(&dev->lock); @@ -1424,12 +1424,12 @@ gpio_virtuser_make_lookup_table(struct gpio_virtuser_device *dev) return -ENOMEM; list_for_each_entry(lookup, &dev->lookup_list, siblings) { + idx = 0; list_for_each_entry(entry, &lookup->entry_list, siblings) { - table->table[i] = + table->table[i++] = GPIO_LOOKUP_IDX(entry->key, entry->offset < 0 ? U16_MAX : entry->offset, - lookup->con_id, i, entry->flags); - i++; + lookup->con_id, idx++, entry->flags); } } @@ -1439,6 +1439,15 @@ gpio_virtuser_make_lookup_table(struct gpio_virtuser_device *dev) return 0; } +static void +gpio_virtuser_remove_lookup_table(struct gpio_virtuser_device *dev) +{ + gpiod_remove_lookup_table(dev->lookup_table); + kfree(dev->lookup_table->dev_id); + kfree(dev->lookup_table); + dev->lookup_table = NULL; +} + static struct fwnode_handle * gpio_virtuser_make_device_swnode(struct gpio_virtuser_device *dev) { @@ -1487,10 +1496,8 @@ gpio_virtuser_device_activate(struct gpio_virtuser_device *dev) pdevinfo.fwnode = swnode; ret = gpio_virtuser_make_lookup_table(dev); - if (ret) { - fwnode_remove_software_node(swnode); - return ret; - } + if (ret) + goto err_remove_swnode; reinit_completion(&dev->probe_completion); dev->driver_bound = false; @@ -1498,23 +1505,31 @@ gpio_virtuser_device_activate(struct gpio_virtuser_device *dev) pdev = platform_device_register_full(&pdevinfo); if (IS_ERR(pdev)) { + ret = PTR_ERR(pdev); bus_unregister_notifier(&platform_bus_type, &dev->bus_notifier); - fwnode_remove_software_node(swnode); - return PTR_ERR(pdev); + goto err_remove_lookup_table; } wait_for_completion(&dev->probe_completion); bus_unregister_notifier(&platform_bus_type, &dev->bus_notifier); if (!dev->driver_bound) { - platform_device_unregister(pdev); - fwnode_remove_software_node(swnode); - return -ENXIO; + ret = -ENXIO; + goto err_unregister_pdev; } dev->pdev = pdev; return 0; + +err_unregister_pdev: + platform_device_unregister(pdev); +err_remove_lookup_table: + gpio_virtuser_remove_lookup_table(dev); +err_remove_swnode: + fwnode_remove_software_node(swnode); + + return ret; } static void @@ -1526,10 +1541,33 @@ gpio_virtuser_device_deactivate(struct gpio_virtuser_device *dev) swnode = dev_fwnode(&dev->pdev->dev); platform_device_unregister(dev->pdev); + gpio_virtuser_remove_lookup_table(dev); fwnode_remove_software_node(swnode); dev->pdev = NULL; - gpiod_remove_lookup_table(dev->lookup_table); - kfree(dev->lookup_table); +} + +static void +gpio_virtuser_device_lockup_configfs(struct gpio_virtuser_device *dev, bool lock) +{ + struct configfs_subsystem *subsys = dev->group.cg_subsys; + struct gpio_virtuser_lookup_entry *entry; + struct gpio_virtuser_lookup *lookup; + + /* + * The device only needs to depend on leaf lookup entries. This is + * sufficient to lock up all the configfs entries that the + * instantiated, alive device depends on. + */ + list_for_each_entry(lookup, &dev->lookup_list, siblings) { + list_for_each_entry(entry, &lookup->entry_list, siblings) { + if (lock) + WARN_ON(configfs_depend_item_unlocked( + subsys, &entry->group.cg_item)); + else + configfs_undepend_item_unlocked( + &entry->group.cg_item); + } + } } static ssize_t @@ -1544,15 +1582,24 @@ gpio_virtuser_device_config_live_store(struct config_item *item, if (ret) return ret; - guard(mutex)(&dev->lock); + if (live) + gpio_virtuser_device_lockup_configfs(dev, true); - if (live == gpio_virtuser_device_is_live(dev)) - return -EPERM; + scoped_guard(mutex, &dev->lock) { + if (live == gpio_virtuser_device_is_live(dev)) + ret = -EPERM; + else if (live) + ret = gpio_virtuser_device_activate(dev); + else + gpio_virtuser_device_deactivate(dev); + } - if (live) - ret = gpio_virtuser_device_activate(dev); - else - gpio_virtuser_device_deactivate(dev); + /* + * Undepend is required only if device disablement (live == 0) + * succeeds or if device enablement (live == 1) fails. + */ + if (live == !!ret) + gpio_virtuser_device_lockup_configfs(dev, false); return ret ?: count; } diff --git a/drivers/gpio/gpio-xilinx.c b/drivers/gpio/gpio-xilinx.c index c6a8f2c82680..792d94c49077 100644 --- a/drivers/gpio/gpio-xilinx.c +++ b/drivers/gpio/gpio-xilinx.c @@ -65,7 +65,7 @@ struct xgpio_instance { DECLARE_BITMAP(state, 64); DECLARE_BITMAP(last_irq_read, 64); DECLARE_BITMAP(dir, 64); - spinlock_t gpio_lock; /* For serializing operations */ + raw_spinlock_t gpio_lock; /* For serializing operations */ int irq; DECLARE_BITMAP(enable, 64); DECLARE_BITMAP(rising_edge, 64); @@ -179,14 +179,14 @@ static void xgpio_set(struct gpio_chip *gc, unsigned int gpio, int val) struct xgpio_instance *chip = gpiochip_get_data(gc); int bit = xgpio_to_bit(chip, gpio); - spin_lock_irqsave(&chip->gpio_lock, flags); + raw_spin_lock_irqsave(&chip->gpio_lock, flags); /* Write to GPIO signal and set its direction to output */ __assign_bit(bit, chip->state, val); xgpio_write_ch(chip, XGPIO_DATA_OFFSET, bit, chip->state); - spin_unlock_irqrestore(&chip->gpio_lock, flags); + raw_spin_unlock_irqrestore(&chip->gpio_lock, flags); } /** @@ -210,7 +210,7 @@ static void xgpio_set_multiple(struct gpio_chip *gc, unsigned long *mask, bitmap_remap(hw_mask, mask, chip->sw_map, chip->hw_map, 64); bitmap_remap(hw_bits, bits, chip->sw_map, chip->hw_map, 64); - spin_lock_irqsave(&chip->gpio_lock, flags); + raw_spin_lock_irqsave(&chip->gpio_lock, flags); bitmap_replace(state, chip->state, hw_bits, hw_mask, 64); @@ -218,7 +218,7 @@ static void xgpio_set_multiple(struct gpio_chip *gc, unsigned long *mask, bitmap_copy(chip->state, state, 64); - spin_unlock_irqrestore(&chip->gpio_lock, flags); + raw_spin_unlock_irqrestore(&chip->gpio_lock, flags); } /** @@ -236,13 +236,13 @@ static int xgpio_dir_in(struct gpio_chip *gc, unsigned int gpio) struct xgpio_instance *chip = gpiochip_get_data(gc); int bit = xgpio_to_bit(chip, gpio); - spin_lock_irqsave(&chip->gpio_lock, flags); + raw_spin_lock_irqsave(&chip->gpio_lock, flags); /* Set the GPIO bit in shadow register and set direction as input */ __set_bit(bit, chip->dir); xgpio_write_ch(chip, XGPIO_TRI_OFFSET, bit, chip->dir); - spin_unlock_irqrestore(&chip->gpio_lock, flags); + raw_spin_unlock_irqrestore(&chip->gpio_lock, flags); return 0; } @@ -265,7 +265,7 @@ static int xgpio_dir_out(struct gpio_chip *gc, unsigned int gpio, int val) struct xgpio_instance *chip = gpiochip_get_data(gc); int bit = xgpio_to_bit(chip, gpio); - spin_lock_irqsave(&chip->gpio_lock, flags); + raw_spin_lock_irqsave(&chip->gpio_lock, flags); /* Write state of GPIO signal */ __assign_bit(bit, chip->state, val); @@ -275,7 +275,7 @@ static int xgpio_dir_out(struct gpio_chip *gc, unsigned int gpio, int val) __clear_bit(bit, chip->dir); xgpio_write_ch(chip, XGPIO_TRI_OFFSET, bit, chip->dir); - spin_unlock_irqrestore(&chip->gpio_lock, flags); + raw_spin_unlock_irqrestore(&chip->gpio_lock, flags); return 0; } @@ -398,7 +398,7 @@ static void xgpio_irq_mask(struct irq_data *irq_data) int bit = xgpio_to_bit(chip, irq_offset); u32 mask = BIT(bit / 32), temp; - spin_lock_irqsave(&chip->gpio_lock, flags); + raw_spin_lock_irqsave(&chip->gpio_lock, flags); __clear_bit(bit, chip->enable); @@ -408,7 +408,7 @@ static void xgpio_irq_mask(struct irq_data *irq_data) temp &= ~mask; xgpio_writereg(chip->regs + XGPIO_IPIER_OFFSET, temp); } - spin_unlock_irqrestore(&chip->gpio_lock, flags); + raw_spin_unlock_irqrestore(&chip->gpio_lock, flags); gpiochip_disable_irq(&chip->gc, irq_offset); } @@ -428,7 +428,7 @@ static void xgpio_irq_unmask(struct irq_data *irq_data) gpiochip_enable_irq(&chip->gc, irq_offset); - spin_lock_irqsave(&chip->gpio_lock, flags); + raw_spin_lock_irqsave(&chip->gpio_lock, flags); __set_bit(bit, chip->enable); @@ -447,7 +447,7 @@ static void xgpio_irq_unmask(struct irq_data *irq_data) xgpio_writereg(chip->regs + XGPIO_IPIER_OFFSET, val); } - spin_unlock_irqrestore(&chip->gpio_lock, flags); + raw_spin_unlock_irqrestore(&chip->gpio_lock, flags); } /** @@ -512,7 +512,7 @@ static void xgpio_irqhandler(struct irq_desc *desc) chained_irq_enter(irqchip, desc); - spin_lock(&chip->gpio_lock); + raw_spin_lock(&chip->gpio_lock); xgpio_read_ch_all(chip, XGPIO_DATA_OFFSET, all); @@ -529,7 +529,7 @@ static void xgpio_irqhandler(struct irq_desc *desc) bitmap_copy(chip->last_irq_read, all, 64); bitmap_or(all, rising, falling, 64); - spin_unlock(&chip->gpio_lock); + raw_spin_unlock(&chip->gpio_lock); dev_dbg(gc->parent, "IRQ rising %*pb falling %*pb\n", 64, rising, 64, falling); @@ -620,7 +620,7 @@ static int xgpio_probe(struct platform_device *pdev) bitmap_set(chip->hw_map, 0, width[0]); bitmap_set(chip->hw_map, 32, width[1]); - spin_lock_init(&chip->gpio_lock); + raw_spin_lock_init(&chip->gpio_lock); chip->gc.base = -1; chip->gc.ngpio = bitmap_weight(chip->hw_map, 64); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c index 3afcd1e8aa54..c4e733c2e75e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c @@ -715,8 +715,9 @@ err: void amdgpu_amdkfd_set_compute_idle(struct amdgpu_device *adev, bool idle) { enum amd_powergating_state state = idle ? AMD_PG_STATE_GATE : AMD_PG_STATE_UNGATE; - if (IP_VERSION_MAJ(amdgpu_ip_version(adev, GC_HWIP, 0)) == 11 && - ((adev->mes.kiq_version & AMDGPU_MES_VERSION_MASK) <= 64)) { + if ((IP_VERSION_MAJ(amdgpu_ip_version(adev, GC_HWIP, 0)) == 11 && + ((adev->mes.kiq_version & AMDGPU_MES_VERSION_MASK) <= 64)) || + (IP_VERSION_MAJ(amdgpu_ip_version(adev, GC_HWIP, 0)) == 12)) { pr_debug("GFXOFF is %s\n", idle ? "enabled" : "disabled"); amdgpu_gfx_off_ctrl(adev, idle); } else if ((IP_VERSION_MAJ(amdgpu_ip_version(adev, GC_HWIP, 0)) == 9) && diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fw_attestation.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fw_attestation.c index 2d4b67175b55..328a1b963548 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fw_attestation.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fw_attestation.c @@ -122,6 +122,10 @@ static int amdgpu_is_fw_attestation_supported(struct amdgpu_device *adev) if (adev->flags & AMD_IS_APU) return 0; + if (amdgpu_ip_version(adev, MP0_HWIP, 0) == IP_VERSION(14, 0, 2) || + amdgpu_ip_version(adev, MP0_HWIP, 0) == IP_VERSION(14, 0, 3)) + return 0; + if (adev->asic_type >= CHIP_SIENNA_CICHLID) return 1; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c index 69a6b6dba0a5..1d155463d044 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c @@ -1989,6 +1989,7 @@ void amdgpu_gfx_enforce_isolation_ring_begin_use(struct amdgpu_ring *ring) { struct amdgpu_device *adev = ring->adev; u32 idx; + bool sched_work = false; if (!adev->gfx.enable_cleaner_shader) return; @@ -2007,15 +2008,19 @@ void amdgpu_gfx_enforce_isolation_ring_begin_use(struct amdgpu_ring *ring) mutex_lock(&adev->enforce_isolation_mutex); if (adev->enforce_isolation[idx]) { if (adev->kfd.init_complete) - amdgpu_gfx_kfd_sch_ctrl(adev, idx, false); + sched_work = true; } mutex_unlock(&adev->enforce_isolation_mutex); + + if (sched_work) + amdgpu_gfx_kfd_sch_ctrl(adev, idx, false); } void amdgpu_gfx_enforce_isolation_ring_end_use(struct amdgpu_ring *ring) { struct amdgpu_device *adev = ring->adev; u32 idx; + bool sched_work = false; if (!adev->gfx.enable_cleaner_shader) return; @@ -2031,9 +2036,12 @@ void amdgpu_gfx_enforce_isolation_ring_end_use(struct amdgpu_ring *ring) mutex_lock(&adev->enforce_isolation_mutex); if (adev->enforce_isolation[idx]) { if (adev->kfd.init_complete) - amdgpu_gfx_kfd_sch_ctrl(adev, idx, true); + sched_work = true; } mutex_unlock(&adev->enforce_isolation_mutex); + + if (sched_work) + amdgpu_gfx_kfd_sch_ctrl(adev, idx, true); } /* diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c index 8b512dc28df8..071f187f5e28 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c @@ -193,8 +193,8 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned int num_ibs, need_ctx_switch = ring->current_ctx != fence_ctx; if (ring->funcs->emit_pipeline_sync && job && ((tmp = amdgpu_sync_get_fence(&job->explicit_sync)) || - (amdgpu_sriov_vf(adev) && need_ctx_switch) || - amdgpu_vm_need_pipeline_sync(ring, job))) { + need_ctx_switch || amdgpu_vm_need_pipeline_sync(ring, job))) { + need_pipe_sync = true; if (tmp) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c index 7d26a962f811..ff5e52025266 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c @@ -567,7 +567,6 @@ static int amdgpu_vram_mgr_new(struct ttm_resource_manager *man, else remaining_size -= size; } - mutex_unlock(&mgr->lock); if (bo->flags & AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS && adjust_dcc_size) { struct drm_buddy_block *dcc_block; @@ -584,6 +583,7 @@ static int amdgpu_vram_mgr_new(struct ttm_resource_manager *man, (u64)vres->base.size, &vres->blocks); } + mutex_unlock(&mgr->lock); vres->base.start = 0; size = max_t(u64, amdgpu_vram_mgr_blocks_size(&vres->blocks), diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_debug.c b/drivers/gpu/drm/amd/amdkfd/kfd_debug.c index 312dfa84f29f..a8abc3091801 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_debug.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_debug.c @@ -350,10 +350,27 @@ int kfd_dbg_set_mes_debug_mode(struct kfd_process_device *pdd, bool sq_trap_en) { uint32_t spi_dbg_cntl = pdd->spi_dbg_override | pdd->spi_dbg_launch_mode; uint32_t flags = pdd->process->dbg_flags; + struct amdgpu_device *adev = pdd->dev->adev; + int r; if (!kfd_dbg_is_per_vmid_supported(pdd->dev)) return 0; + if (!pdd->proc_ctx_cpu_ptr) { + r = amdgpu_amdkfd_alloc_gtt_mem(adev, + AMDGPU_MES_PROC_CTX_SIZE, + &pdd->proc_ctx_bo, + &pdd->proc_ctx_gpu_addr, + &pdd->proc_ctx_cpu_ptr, + false); + if (r) { + dev_err(adev->dev, + "failed to allocate process context bo\n"); + return r; + } + memset(pdd->proc_ctx_cpu_ptr, 0, AMDGPU_MES_PROC_CTX_SIZE); + } + return amdgpu_mes_set_shader_debugger(pdd->dev->adev, pdd->proc_ctx_gpu_addr, spi_dbg_cntl, pdd->watch_points, flags, sq_trap_en); } diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c b/drivers/gpu/drm/amd/amdkfd/kfd_process.c index d0ee173acf82..edfe0b4788f4 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_process.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_process.c @@ -1160,7 +1160,8 @@ static void kfd_process_wq_release(struct work_struct *work) */ synchronize_rcu(); ef = rcu_access_pointer(p->ef); - dma_fence_signal(ef); + if (ef) + dma_fence_signal(ef); kfd_process_remove_sysfs(p); diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c index 48be917e7bc5..5f216d626cbb 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c @@ -8400,16 +8400,6 @@ static void manage_dm_interrupts(struct amdgpu_device *adev, struct amdgpu_crtc *acrtc, struct dm_crtc_state *acrtc_state) { - /* - * We have no guarantee that the frontend index maps to the same - * backend index - some even map to more than one. - * - * TODO: Use a different interrupt or check DC itself for the mapping. - */ - int irq_type = - amdgpu_display_crtc_idx_to_irq_type( - adev, - acrtc->crtc_id); struct drm_vblank_crtc_config config = {0}; struct dc_crtc_timing *timing; int offdelay; @@ -8435,28 +8425,7 @@ static void manage_dm_interrupts(struct amdgpu_device *adev, drm_crtc_vblank_on_config(&acrtc->base, &config); - - amdgpu_irq_get( - adev, - &adev->pageflip_irq, - irq_type); -#if defined(CONFIG_DRM_AMD_SECURE_DISPLAY) - amdgpu_irq_get( - adev, - &adev->vline0_irq, - irq_type); -#endif } else { -#if defined(CONFIG_DRM_AMD_SECURE_DISPLAY) - amdgpu_irq_put( - adev, - &adev->vline0_irq, - irq_type); -#endif - amdgpu_irq_put( - adev, - &adev->pageflip_irq, - irq_type); drm_crtc_vblank_off(&acrtc->base); } } @@ -8927,6 +8896,7 @@ static void amdgpu_dm_enable_self_refresh(struct amdgpu_crtc *acrtc_attach, struct replay_settings *pr = &acrtc_state->stream->link->replay_settings; struct amdgpu_dm_connector *aconn = (struct amdgpu_dm_connector *)acrtc_state->stream->dm_stream_context; + bool vrr_active = amdgpu_dm_crtc_vrr_active(acrtc_state); if (acrtc_state->update_type > UPDATE_TYPE_FAST) { if (pr->config.replay_supported && !pr->replay_feature_enabled) @@ -8953,14 +8923,15 @@ static void amdgpu_dm_enable_self_refresh(struct amdgpu_crtc *acrtc_attach, * adequate number of fast atomic commits to notify KMD * of update events. See `vblank_control_worker()`. */ - if (acrtc_attach->dm_irq_params.allow_sr_entry && + if (!vrr_active && + acrtc_attach->dm_irq_params.allow_sr_entry && #ifdef CONFIG_DRM_AMD_SECURE_DISPLAY !amdgpu_dm_crc_window_is_activated(acrtc_state->base.crtc) && #endif (current_ts - psr->psr_dirty_rects_change_timestamp_ns) > 500000000) { if (pr->replay_feature_enabled && !pr->replay_allow_active) amdgpu_dm_replay_enable(acrtc_state->stream, true); - if (psr->psr_version >= DC_PSR_VERSION_SU_1 && + if (psr->psr_version == DC_PSR_VERSION_SU_1 && !psr->psr_allow_active && !aconn->disallow_edp_enter_psr) amdgpu_dm_psr_enable(acrtc_state->stream); } @@ -9131,7 +9102,7 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state, acrtc_state->stream->link->psr_settings.psr_dirty_rects_change_timestamp_ns = timestamp_ns; if (acrtc_state->stream->link->psr_settings.psr_allow_active) - amdgpu_dm_psr_disable(acrtc_state->stream); + amdgpu_dm_psr_disable(acrtc_state->stream, true); mutex_unlock(&dm->dc_lock); } } @@ -9297,11 +9268,11 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state, bundle->stream_update.abm_level = &acrtc_state->abm_level; mutex_lock(&dm->dc_lock); - if (acrtc_state->update_type > UPDATE_TYPE_FAST) { + if ((acrtc_state->update_type > UPDATE_TYPE_FAST) || vrr_active) { if (acrtc_state->stream->link->replay_settings.replay_allow_active) amdgpu_dm_replay_disable(acrtc_state->stream); if (acrtc_state->stream->link->psr_settings.psr_allow_active) - amdgpu_dm_psr_disable(acrtc_state->stream); + amdgpu_dm_psr_disable(acrtc_state->stream, true); } mutex_unlock(&dm->dc_lock); @@ -11155,8 +11126,8 @@ dm_get_plane_scale(struct drm_plane_state *plane_state, int plane_src_w, plane_src_h; dm_get_oriented_plane_size(plane_state, &plane_src_w, &plane_src_h); - *out_plane_scale_w = plane_state->crtc_w * 1000 / plane_src_w; - *out_plane_scale_h = plane_state->crtc_h * 1000 / plane_src_h; + *out_plane_scale_w = plane_src_w ? plane_state->crtc_w * 1000 / plane_src_w : 0; + *out_plane_scale_h = plane_src_h ? plane_state->crtc_h * 1000 / plane_src_h : 0; } /* @@ -11410,6 +11381,25 @@ static int dm_crtc_get_cursor_mode(struct amdgpu_device *adev, return 0; } +static bool amdgpu_dm_crtc_mem_type_changed(struct drm_device *dev, + struct drm_atomic_state *state, + struct drm_crtc_state *crtc_state) +{ + struct drm_plane *plane; + struct drm_plane_state *new_plane_state, *old_plane_state; + + drm_for_each_plane_mask(plane, dev, crtc_state->plane_mask) { + new_plane_state = drm_atomic_get_plane_state(state, plane); + old_plane_state = drm_atomic_get_plane_state(state, plane); + + if (old_plane_state->fb && new_plane_state->fb && + get_mem_type(old_plane_state->fb) != get_mem_type(new_plane_state->fb)) + return true; + } + + return false; +} + /** * amdgpu_dm_atomic_check() - Atomic check implementation for AMDgpu DM. * @@ -11607,10 +11597,6 @@ static int amdgpu_dm_atomic_check(struct drm_device *dev, /* Remove exiting planes if they are modified */ for_each_oldnew_plane_in_descending_zpos(state, plane, old_plane_state, new_plane_state) { - if (old_plane_state->fb && new_plane_state->fb && - get_mem_type(old_plane_state->fb) != - get_mem_type(new_plane_state->fb)) - lock_and_validation_needed = true; ret = dm_update_plane_state(dc, state, plane, old_plane_state, @@ -11905,9 +11891,11 @@ static int amdgpu_dm_atomic_check(struct drm_device *dev, /* * Only allow async flips for fast updates that don't change - * the FB pitch, the DCC state, rotation, etc. + * the FB pitch, the DCC state, rotation, mem_type, etc. */ - if (new_crtc_state->async_flip && lock_and_validation_needed) { + if (new_crtc_state->async_flip && + (lock_and_validation_needed || + amdgpu_dm_crtc_mem_type_changed(dev, state, new_crtc_state))) { drm_dbg_atomic(crtc->dev, "[CRTC:%d:%s] async flips are only supported for fast updates\n", crtc->base.id, crtc->name); diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h index 6464a8378387..2227cd8e4a89 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h @@ -697,6 +697,8 @@ struct amdgpu_dm_connector { struct drm_dp_mst_port *mst_output_port; struct amdgpu_dm_connector *mst_root; struct drm_dp_aux *dsc_aux; + uint32_t mst_local_bw; + uint16_t vc_full_pbn; struct mutex handle_mst_msg_ready; /* TODO see if we can merge with ddc_bus or make a dm_connector */ diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.c index f936a35fa9eb..0f6ba7b1575d 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.c @@ -30,6 +30,7 @@ #include "amdgpu_dm.h" #include "dc.h" #include "amdgpu_securedisplay.h" +#include "amdgpu_dm_psr.h" static const char *const pipe_crc_sources[] = { "none", @@ -224,6 +225,10 @@ int amdgpu_dm_crtc_configure_crc_source(struct drm_crtc *crtc, mutex_lock(&adev->dm.dc_lock); + /* For PSR1, check that the panel has exited PSR */ + if (stream_state->link->psr_settings.psr_version < DC_PSR_VERSION_SU_1) + amdgpu_dm_psr_wait_disable(stream_state); + /* Enable or disable CRTC CRC generation */ if (dm_is_crc_source_crtc(source) || source == AMDGPU_DM_PIPE_CRC_SOURCE_NONE) { if (!dc_stream_configure_crc(stream_state->ctx->dc, @@ -357,6 +362,17 @@ int amdgpu_dm_crtc_set_crc_source(struct drm_crtc *crtc, const char *src_name) } + /* + * Reading the CRC requires the vblank interrupt handler to be + * enabled. Keep a reference until CRC capture stops. + */ + enabled = amdgpu_dm_is_valid_crc_source(cur_crc_src); + if (!enabled && enable) { + ret = drm_crtc_vblank_get(crtc); + if (ret) + goto cleanup; + } + #if defined(CONFIG_DRM_AMD_SECURE_DISPLAY) /* Reset secure_display when we change crc source from debugfs */ amdgpu_dm_set_crc_window_default(crtc, crtc_state->stream); @@ -367,16 +383,7 @@ int amdgpu_dm_crtc_set_crc_source(struct drm_crtc *crtc, const char *src_name) goto cleanup; } - /* - * Reading the CRC requires the vblank interrupt handler to be - * enabled. Keep a reference until CRC capture stops. - */ - enabled = amdgpu_dm_is_valid_crc_source(cur_crc_src); if (!enabled && enable) { - ret = drm_crtc_vblank_get(crtc); - if (ret) - goto cleanup; - if (dm_is_crc_source_dprx(source)) { if (drm_dp_start_crc(aux, crtc)) { DRM_DEBUG_DRIVER("dp start crc failed\n"); diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c index 64a041c2af05..36a830a7440f 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c @@ -93,7 +93,7 @@ int amdgpu_dm_crtc_set_vupdate_irq(struct drm_crtc *crtc, bool enable) return rc; } -bool amdgpu_dm_crtc_vrr_active(struct dm_crtc_state *dm_state) +bool amdgpu_dm_crtc_vrr_active(const struct dm_crtc_state *dm_state) { return dm_state->freesync_config.state == VRR_STATE_ACTIVE_VARIABLE || dm_state->freesync_config.state == VRR_STATE_ACTIVE_FIXED; @@ -142,7 +142,7 @@ static void amdgpu_dm_crtc_set_panel_sr_feature( amdgpu_dm_replay_enable(vblank_work->stream, true); } else if (vblank_enabled) { if (link->psr_settings.psr_version < DC_PSR_VERSION_SU_1 && is_sr_active) - amdgpu_dm_psr_disable(vblank_work->stream); + amdgpu_dm_psr_disable(vblank_work->stream, false); } else if (link->psr_settings.psr_feature_enabled && allow_sr_entry && !is_sr_active && !is_crc_window_active) { diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.h index 17e948753f59..c1212947a77b 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.h +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.h @@ -37,7 +37,7 @@ int amdgpu_dm_crtc_set_vupdate_irq(struct drm_crtc *crtc, bool enable); bool amdgpu_dm_crtc_vrr_active_irq(struct amdgpu_crtc *acrtc); -bool amdgpu_dm_crtc_vrr_active(struct dm_crtc_state *dm_state); +bool amdgpu_dm_crtc_vrr_active(const struct dm_crtc_state *dm_state); int amdgpu_dm_crtc_enable_vblank(struct drm_crtc *crtc); diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c index 6a97bb2d9160..8b7c1a428f64 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c @@ -3606,7 +3606,7 @@ static int crc_win_update_set(void *data, u64 val) /* PSR may write to OTG CRC window control register, * so close it before starting secure_display. */ - amdgpu_dm_psr_disable(acrtc->dm_irq_params.stream); + amdgpu_dm_psr_disable(acrtc->dm_irq_params.stream, true); spin_lock_irq(&adev_to_drm(adev)->event_lock); diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c index 6e4359490613..1080075ccb17 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c @@ -155,6 +155,17 @@ amdgpu_dm_mst_connector_late_register(struct drm_connector *connector) return 0; } + +static inline void +amdgpu_dm_mst_reset_mst_connector_setting(struct amdgpu_dm_connector *aconnector) +{ + aconnector->drm_edid = NULL; + aconnector->dsc_aux = NULL; + aconnector->mst_output_port->passthrough_aux = NULL; + aconnector->mst_local_bw = 0; + aconnector->vc_full_pbn = 0; +} + static void amdgpu_dm_mst_connector_early_unregister(struct drm_connector *connector) { @@ -182,9 +193,7 @@ amdgpu_dm_mst_connector_early_unregister(struct drm_connector *connector) dc_sink_release(dc_sink); aconnector->dc_sink = NULL; - aconnector->drm_edid = NULL; - aconnector->dsc_aux = NULL; - port->passthrough_aux = NULL; + amdgpu_dm_mst_reset_mst_connector_setting(aconnector); } aconnector->mst_status = MST_STATUS_DEFAULT; @@ -504,9 +513,7 @@ dm_dp_mst_detect(struct drm_connector *connector, dc_sink_release(aconnector->dc_sink); aconnector->dc_sink = NULL; - aconnector->drm_edid = NULL; - aconnector->dsc_aux = NULL; - port->passthrough_aux = NULL; + amdgpu_dm_mst_reset_mst_connector_setting(aconnector); amdgpu_dm_set_mst_status(&aconnector->mst_status, MST_REMOTE_EDID | MST_ALLOCATE_NEW_PAYLOAD | MST_CLEAR_ALLOCATED_PAYLOAD, @@ -1819,9 +1826,18 @@ enum dc_status dm_dp_mst_is_port_support_mode( struct drm_dp_mst_port *immediate_upstream_port = NULL; uint32_t end_link_bw = 0; - /*Get last DP link BW capability*/ - if (dp_get_link_current_set_bw(&aconnector->mst_output_port->aux, &end_link_bw)) { - if (stream_kbps > end_link_bw) { + /*Get last DP link BW capability. Mode shall be supported by Legacy peer*/ + if (aconnector->mst_output_port->pdt != DP_PEER_DEVICE_DP_LEGACY_CONV && + aconnector->mst_output_port->pdt != DP_PEER_DEVICE_NONE) { + if (aconnector->vc_full_pbn != aconnector->mst_output_port->full_pbn) { + dp_get_link_current_set_bw(&aconnector->mst_output_port->aux, &end_link_bw); + aconnector->vc_full_pbn = aconnector->mst_output_port->full_pbn; + aconnector->mst_local_bw = end_link_bw; + } else { + end_link_bw = aconnector->mst_local_bw; + } + + if (end_link_bw > 0 && stream_kbps > end_link_bw) { DRM_DEBUG_DRIVER("MST_DSC dsc decode at last link." "Mode required bw can't fit into last link\n"); return DC_FAIL_BANDWIDTH_VALIDATE; @@ -1835,11 +1851,15 @@ enum dc_status dm_dp_mst_is_port_support_mode( if (immediate_upstream_port) { virtual_channel_bw_in_kbps = kbps_from_pbn(immediate_upstream_port->full_pbn); virtual_channel_bw_in_kbps = min(root_link_bw_in_kbps, virtual_channel_bw_in_kbps); - if (bw_range.min_kbps > virtual_channel_bw_in_kbps) { - DRM_DEBUG_DRIVER("MST_DSC dsc decode at last link." - "Max dsc compression can't fit into MST available bw\n"); - return DC_FAIL_BANDWIDTH_VALIDATE; - } + } else { + /* For topology LCT 1 case - only one mstb*/ + virtual_channel_bw_in_kbps = root_link_bw_in_kbps; + } + + if (bw_range.min_kbps > virtual_channel_bw_in_kbps) { + DRM_DEBUG_DRIVER("MST_DSC dsc decode at last link." + "Max dsc compression can't fit into MST available bw\n"); + return DC_FAIL_BANDWIDTH_VALIDATE; } } diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.c index f40240aafe98..45858bf1523d 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.c @@ -201,14 +201,13 @@ void amdgpu_dm_psr_enable(struct dc_stream_state *stream) * * Return: true if success */ -bool amdgpu_dm_psr_disable(struct dc_stream_state *stream) +bool amdgpu_dm_psr_disable(struct dc_stream_state *stream, bool wait) { - unsigned int power_opt = 0; bool psr_enable = false; DRM_DEBUG_DRIVER("Disabling psr...\n"); - return dc_link_set_psr_allow_active(stream->link, &psr_enable, true, false, &power_opt); + return dc_link_set_psr_allow_active(stream->link, &psr_enable, wait, false, NULL); } /* @@ -251,3 +250,33 @@ bool amdgpu_dm_psr_is_active_allowed(struct amdgpu_display_manager *dm) return allow_active; } + +/** + * amdgpu_dm_psr_wait_disable() - Wait for eDP panel to exit PSR + * @stream: stream state attached to the eDP link + * + * Waits for a max of 500ms for the eDP panel to exit PSR. + * + * Return: true if panel exited PSR, false otherwise. + */ +bool amdgpu_dm_psr_wait_disable(struct dc_stream_state *stream) +{ + enum dc_psr_state psr_state = PSR_STATE0; + struct dc_link *link = stream->link; + int retry_count; + + if (link == NULL) + return false; + + for (retry_count = 0; retry_count <= 1000; retry_count++) { + dc_link_get_psr_state(link, &psr_state); + if (psr_state == PSR_STATE0) + break; + udelay(500); + } + + if (retry_count == 1000) + return false; + + return true; +} diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.h index cd2d45c2b5ef..e2366321a3c1 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.h +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.h @@ -34,8 +34,9 @@ void amdgpu_dm_set_psr_caps(struct dc_link *link); void amdgpu_dm_psr_enable(struct dc_stream_state *stream); bool amdgpu_dm_link_setup_psr(struct dc_stream_state *stream); -bool amdgpu_dm_psr_disable(struct dc_stream_state *stream); +bool amdgpu_dm_psr_disable(struct dc_stream_state *stream, bool wait); bool amdgpu_dm_psr_disable_all(struct amdgpu_display_manager *dm); bool amdgpu_dm_psr_is_active_allowed(struct amdgpu_display_manager *dm); +bool amdgpu_dm_psr_wait_disable(struct dc_stream_state *stream); #endif /* AMDGPU_DM_AMDGPU_DM_PSR_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c index 49fe7dcf9372..2723558049d6 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc.c @@ -4510,7 +4510,7 @@ static bool commit_minimal_transition_based_on_current_context(struct dc *dc, struct pipe_split_policy_backup policy; struct dc_state *intermediate_context; struct dc_state *old_current_state = dc->current_state; - struct dc_surface_update srf_updates[MAX_SURFACE_NUM] = {0}; + struct dc_surface_update srf_updates[MAX_SURFACES] = {0}; int surface_count; /* diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_state.c b/drivers/gpu/drm/amd/display/dc/core/dc_state.c index e006f816ff2f..1b2cce127981 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_state.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_state.c @@ -483,9 +483,9 @@ bool dc_state_add_plane( if (stream_status == NULL) { dm_error("Existing stream not found; failed to attach surface!\n"); goto out; - } else if (stream_status->plane_count == MAX_SURFACE_NUM) { + } else if (stream_status->plane_count == MAX_SURFACES) { dm_error("Surface: can not attach plane_state %p! Maximum is: %d\n", - plane_state, MAX_SURFACE_NUM); + plane_state, MAX_SURFACES); goto out; } else if (!otg_master_pipe) { goto out; @@ -600,7 +600,7 @@ bool dc_state_rem_all_planes_for_stream( { int i, old_plane_count; struct dc_stream_status *stream_status = NULL; - struct dc_plane_state *del_planes[MAX_SURFACE_NUM] = { 0 }; + struct dc_plane_state *del_planes[MAX_SURFACES] = { 0 }; for (i = 0; i < state->stream_count; i++) if (state->streams[i] == stream) { @@ -875,7 +875,7 @@ bool dc_state_rem_all_phantom_planes_for_stream( { int i, old_plane_count; struct dc_stream_status *stream_status = NULL; - struct dc_plane_state *del_planes[MAX_SURFACE_NUM] = { 0 }; + struct dc_plane_state *del_planes[MAX_SURFACES] = { 0 }; for (i = 0; i < state->stream_count; i++) if (state->streams[i] == phantom_stream) { diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h index e9b9126c0401..08c5a315b3a6 100644 --- a/drivers/gpu/drm/amd/display/dc/dc.h +++ b/drivers/gpu/drm/amd/display/dc/dc.h @@ -57,7 +57,7 @@ struct dmub_notification; #define DC_VER "3.2.310" -#define MAX_SURFACES 3 +#define MAX_SURFACES 4 #define MAX_PLANES 6 #define MAX_STREAMS 6 #define MIN_VIEWPORT_SIZE 12 @@ -1398,7 +1398,7 @@ struct dc_scratch_space { * store current value in plane states so we can still recover * a valid current state during dc update. */ - struct dc_plane_state plane_states[MAX_SURFACE_NUM]; + struct dc_plane_state plane_states[MAX_SURFACES]; struct dc_stream_state stream_state; }; diff --git a/drivers/gpu/drm/amd/display/dc/dc_stream.h b/drivers/gpu/drm/amd/display/dc/dc_stream.h index 413970588a26..860506c6bda4 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_stream.h +++ b/drivers/gpu/drm/amd/display/dc/dc_stream.h @@ -56,7 +56,7 @@ struct dc_stream_status { int plane_count; int audio_inst; struct timing_sync_info timing_sync_info; - struct dc_plane_state *plane_states[MAX_SURFACE_NUM]; + struct dc_plane_state *plane_states[MAX_SURFACES]; bool is_abm_supported; struct mall_stream_config mall_stream_config; bool fpo_in_use; diff --git a/drivers/gpu/drm/amd/display/dc/dc_types.h b/drivers/gpu/drm/amd/display/dc/dc_types.h index edf4df1d03b5..9466b63644d5 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_types.h +++ b/drivers/gpu/drm/amd/display/dc/dc_types.h @@ -76,7 +76,6 @@ struct dc_perf_trace { unsigned long last_entry_write; }; -#define MAX_SURFACE_NUM 6 #define NUM_PIXEL_FORMATS 10 enum tiling_mode { diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_hw_lock_mgr.c b/drivers/gpu/drm/amd/display/dc/dce/dmub_hw_lock_mgr.c index bf636b28e3e1..5bb8b78bf250 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dmub_hw_lock_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_hw_lock_mgr.c @@ -63,7 +63,8 @@ void dmub_hw_lock_mgr_inbox0_cmd(struct dc_dmub_srv *dmub_srv, bool should_use_dmub_lock(struct dc_link *link) { - if (link->psr_settings.psr_version == DC_PSR_VERSION_SU_1) + if (link->psr_settings.psr_version == DC_PSR_VERSION_SU_1 || + link->psr_settings.psr_version == DC_PSR_VERSION_1) return true; if (link->replay_settings.replay_feature_enabled) diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn35/dcn35_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn35/dcn35_fpu.c index beed7adbbd43..47d785204f29 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn35/dcn35_fpu.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn35/dcn35_fpu.c @@ -195,9 +195,9 @@ struct _vcs_dpi_soc_bounding_box_st dcn3_5_soc = { .dcn_downspread_percent = 0.5, .gpuvm_min_page_size_bytes = 4096, .hostvm_min_page_size_bytes = 4096, - .do_urgent_latency_adjustment = 1, + .do_urgent_latency_adjustment = 0, .urgent_latency_adjustment_fabric_clock_component_us = 0, - .urgent_latency_adjustment_fabric_clock_reference_mhz = 3000, + .urgent_latency_adjustment_fabric_clock_reference_mhz = 0, }; void dcn35_build_wm_range_table_fpu(struct clk_mgr *clk_mgr) diff --git a/drivers/gpu/drm/amd/display/dc/dml/dml_inline_defs.h b/drivers/gpu/drm/amd/display/dc/dml/dml_inline_defs.h index 072bd0539605..6b2ab4ec2b5f 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dml_inline_defs.h +++ b/drivers/gpu/drm/amd/display/dc/dml/dml_inline_defs.h @@ -66,11 +66,15 @@ static inline double dml_max5(double a, double b, double c, double d, double e) static inline double dml_ceil(double a, double granularity) { + if (granularity == 0) + return 0; return (double) dcn_bw_ceil2(a, granularity); } static inline double dml_floor(double a, double granularity) { + if (granularity == 0) + return 0; return (double) dcn_bw_floor2(a, granularity); } @@ -114,11 +118,15 @@ static inline double dml_ceil_2(double f) static inline double dml_ceil_ex(double x, double granularity) { + if (granularity == 0) + return 0; return (double) dcn_bw_ceil2(x, granularity); } static inline double dml_floor_ex(double x, double granularity) { + if (granularity == 0) + return 0; return (double) dcn_bw_floor2(x, granularity); } diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_dcn4_calcs.c b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_dcn4_calcs.c index 601320b1be81..ecfa3c898e09 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_dcn4_calcs.c +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_dcn4_calcs.c @@ -579,8 +579,8 @@ static void CalculateBytePerPixelAndBlockSizes( { *BytePerPixelDETY = 0; *BytePerPixelDETC = 0; - *BytePerPixelY = 0; - *BytePerPixelC = 0; + *BytePerPixelY = 1; + *BytePerPixelC = 1; if (SourcePixelFormat == dml2_444_64) { *BytePerPixelDETY = 8; diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml2_mall_phantom.c b/drivers/gpu/drm/amd/display/dc/dml2/dml2_mall_phantom.c index 3d29169dd6bb..6b3b8803e0ae 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml2_mall_phantom.c +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml2_mall_phantom.c @@ -813,7 +813,7 @@ static bool remove_all_phantom_planes_for_stream(struct dml2_context *ctx, struc { int i, old_plane_count; struct dc_stream_status *stream_status = NULL; - struct dc_plane_state *del_planes[MAX_SURFACE_NUM] = { 0 }; + struct dc_plane_state *del_planes[MAX_SURFACES] = { 0 }; for (i = 0; i < context->stream_count; i++) if (context->streams[i] == stream) { diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.c index fa11f075d1f9..d9136d252ee1 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.c @@ -1398,12 +1398,12 @@ void dcn32_disable_link_output(struct dc_link *link, link_hwss->disable_link_output(link, link_res, signal); link->phy_state.symclk_state = SYMCLK_OFF_TX_OFF; - - if (signal == SIGNAL_TYPE_EDP && - link->dc->hwss.edp_power_control && - !link->skip_implict_edp_power_control) - link->dc->hwss.edp_power_control(link, false); - else if (dmcu != NULL && dmcu->funcs->unlock_phy) + /* + * Add the logic to extract BOTH power up and power down sequences + * from enable/disable link output and only call edp panel control + * in enable_link_dp and disable_link_dp once. + */ + if (dmcu != NULL && dmcu->funcs->unlock_phy) dmcu->funcs->unlock_phy(dmcu); dc->link_srv->dp_trace_source_sequence(link, DPCD_SOURCE_SEQ_AFTER_DISABLE_LINK_PHY); diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h index ae3563d71fa0..356d9422b411 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h @@ -303,5 +303,7 @@ int smu_v13_0_set_wbrf_exclusion_ranges(struct smu_context *smu, int smu_v13_0_get_boot_freq_by_index(struct smu_context *smu, enum smu_clk_type clk_type, uint32_t *value); + +void smu_v13_0_interrupt_work(struct smu_context *smu); #endif #endif diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c index 2bfea740dace..2d1e7ebd1bac 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c @@ -1320,11 +1320,11 @@ static int smu_v13_0_set_irq_state(struct amdgpu_device *adev, return 0; } -static int smu_v13_0_ack_ac_dc_interrupt(struct smu_context *smu) +void smu_v13_0_interrupt_work(struct smu_context *smu) { - return smu_cmn_send_smc_msg(smu, - SMU_MSG_ReenableAcDcInterrupt, - NULL); + smu_cmn_send_smc_msg(smu, + SMU_MSG_ReenableAcDcInterrupt, + NULL); } #define THM_11_0__SRCID__THM_DIG_THERM_L2H 0 /* ASIC_TEMP > CG_THERMAL_INT.DIG_THERM_INTH */ @@ -1377,12 +1377,12 @@ static int smu_v13_0_irq_process(struct amdgpu_device *adev, switch (ctxid) { case SMU_IH_INTERRUPT_CONTEXT_ID_AC: dev_dbg(adev->dev, "Switched to AC mode!\n"); - smu_v13_0_ack_ac_dc_interrupt(smu); + schedule_work(&smu->interrupt_work); adev->pm.ac_power = true; break; case SMU_IH_INTERRUPT_CONTEXT_ID_DC: dev_dbg(adev->dev, "Switched to DC mode!\n"); - smu_v13_0_ack_ac_dc_interrupt(smu); + schedule_work(&smu->interrupt_work); adev->pm.ac_power = false; break; case SMU_IH_INTERRUPT_CONTEXT_ID_THERMAL_THROTTLING: diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c index 3aa705aae4c0..0551a3311217 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c @@ -2643,11 +2643,12 @@ static int smu_v13_0_0_set_power_profile_mode(struct smu_context *smu, &backend_workload_mask); /* Add optimizations for SMU13.0.0/10. Reuse the power saving profile */ - if ((amdgpu_ip_version(smu->adev, MP1_HWIP, 0) == IP_VERSION(13, 0, 0) && - ((smu->adev->pm.fw_version == 0x004e6601) || - (smu->adev->pm.fw_version >= 0x004e7300))) || - (amdgpu_ip_version(smu->adev, MP1_HWIP, 0) == IP_VERSION(13, 0, 10) && - smu->adev->pm.fw_version >= 0x00504500)) { + if ((workload_mask & (1 << PP_SMC_POWER_PROFILE_COMPUTE)) && + ((amdgpu_ip_version(smu->adev, MP1_HWIP, 0) == IP_VERSION(13, 0, 0) && + ((smu->adev->pm.fw_version == 0x004e6601) || + (smu->adev->pm.fw_version >= 0x004e7300))) || + (amdgpu_ip_version(smu->adev, MP1_HWIP, 0) == IP_VERSION(13, 0, 10) && + smu->adev->pm.fw_version >= 0x00504500))) { workload_type = smu_cmn_to_asic_specific_index(smu, CMN2ASIC_MAPPING_WORKLOAD, PP_SMC_POWER_PROFILE_POWERSAVING); @@ -3219,6 +3220,7 @@ static const struct pptable_funcs smu_v13_0_0_ppt_funcs = { .is_asic_wbrf_supported = smu_v13_0_0_wbrf_support_check, .enable_uclk_shadow = smu_v13_0_enable_uclk_shadow, .set_wbrf_exclusion_ranges = smu_v13_0_set_wbrf_exclusion_ranges, + .interrupt_work = smu_v13_0_interrupt_work, }; void smu_v13_0_0_set_ppt_funcs(struct smu_context *smu) diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c index aabb94796005..55ef18517b0f 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c @@ -2797,6 +2797,7 @@ static const struct pptable_funcs smu_v13_0_7_ppt_funcs = { .is_asic_wbrf_supported = smu_v13_0_7_wbrf_support_check, .enable_uclk_shadow = smu_v13_0_enable_uclk_shadow, .set_wbrf_exclusion_ranges = smu_v13_0_set_wbrf_exclusion_ranges, + .interrupt_work = smu_v13_0_interrupt_work, }; void smu_v13_0_7_set_ppt_funcs(struct smu_context *smu) diff --git a/drivers/gpu/drm/bridge/adv7511/adv7511_audio.c b/drivers/gpu/drm/bridge/adv7511/adv7511_audio.c index 61f4a38e7d2b..8f786592143b 100644 --- a/drivers/gpu/drm/bridge/adv7511/adv7511_audio.c +++ b/drivers/gpu/drm/bridge/adv7511/adv7511_audio.c @@ -153,7 +153,16 @@ static int adv7511_hdmi_hw_params(struct device *dev, void *data, ADV7511_AUDIO_CFG3_LEN_MASK, len); regmap_update_bits(adv7511->regmap, ADV7511_REG_I2C_FREQ_ID_CFG, ADV7511_I2C_FREQ_ID_CFG_RATE_MASK, rate << 4); - regmap_write(adv7511->regmap, 0x73, 0x1); + + /* send current Audio infoframe values while updating */ + regmap_update_bits(adv7511->regmap, ADV7511_REG_INFOFRAME_UPDATE, + BIT(5), BIT(5)); + + regmap_write(adv7511->regmap, ADV7511_REG_AUDIO_INFOFRAME(0), 0x1); + + /* use Audio infoframe updated info */ + regmap_update_bits(adv7511->regmap, ADV7511_REG_INFOFRAME_UPDATE, + BIT(5), 0); return 0; } @@ -184,8 +193,9 @@ static int audio_startup(struct device *dev, void *data) regmap_update_bits(adv7511->regmap, ADV7511_REG_GC(0), BIT(7) | BIT(6), BIT(7)); /* use Audio infoframe updated info */ - regmap_update_bits(adv7511->regmap, ADV7511_REG_GC(1), + regmap_update_bits(adv7511->regmap, ADV7511_REG_INFOFRAME_UPDATE, BIT(5), 0); + /* enable SPDIF receiver */ if (adv7511->audio_source == ADV7511_AUDIO_SOURCE_SPDIF) regmap_update_bits(adv7511->regmap, ADV7511_REG_AUDIO_CONFIG, diff --git a/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c b/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c index eb5919b38263..a13b3d8ab6ac 100644 --- a/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c +++ b/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c @@ -1241,8 +1241,10 @@ static int adv7511_probe(struct i2c_client *i2c) return ret; ret = adv7511_init_regulators(adv7511); - if (ret) - return dev_err_probe(dev, ret, "failed to init regulators\n"); + if (ret) { + dev_err_probe(dev, ret, "failed to init regulators\n"); + goto err_of_node_put; + } /* * The power down GPIO is optional. If present, toggle it from active to @@ -1363,6 +1365,8 @@ err_i2c_unregister_edid: i2c_unregister_device(adv7511->i2c_edid); uninit_regulators: adv7511_uninit_regulators(adv7511); +err_of_node_put: + of_node_put(adv7511->host_node); return ret; } @@ -1371,6 +1375,8 @@ static void adv7511_remove(struct i2c_client *i2c) { struct adv7511 *adv7511 = i2c_get_clientdata(i2c); + of_node_put(adv7511->host_node); + adv7511_uninit_regulators(adv7511); drm_bridge_remove(&adv7511->bridge); diff --git a/drivers/gpu/drm/bridge/adv7511/adv7533.c b/drivers/gpu/drm/bridge/adv7511/adv7533.c index 4481489aaf5e..122ad91e8a32 100644 --- a/drivers/gpu/drm/bridge/adv7511/adv7533.c +++ b/drivers/gpu/drm/bridge/adv7511/adv7533.c @@ -172,7 +172,7 @@ int adv7533_parse_dt(struct device_node *np, struct adv7511 *adv) of_property_read_u32(np, "adi,dsi-lanes", &num_lanes); - if (num_lanes < 1 || num_lanes > 4) + if (num_lanes < 2 || num_lanes > 4) return -EINVAL; adv->num_dsi_lanes = num_lanes; @@ -181,8 +181,6 @@ int adv7533_parse_dt(struct device_node *np, struct adv7511 *adv) if (!adv->host_node) return -ENODEV; - of_node_put(adv->host_node); - adv->use_timing_gen = !of_property_read_bool(np, "adi,disable-timing-generator"); diff --git a/drivers/gpu/drm/bridge/ite-it6263.c b/drivers/gpu/drm/bridge/ite-it6263.c index cbabd4e20d3e..5f138a5692c7 100644 --- a/drivers/gpu/drm/bridge/ite-it6263.c +++ b/drivers/gpu/drm/bridge/ite-it6263.c @@ -845,8 +845,8 @@ static int it6263_probe(struct i2c_client *client) it->lvds_i2c = devm_i2c_new_dummy_device(dev, client->adapter, LVDS_INPUT_CTRL_I2C_ADDR); if (IS_ERR(it->lvds_i2c)) - dev_err_probe(it->dev, PTR_ERR(it->lvds_i2c), - "failed to allocate I2C device for LVDS\n"); + return dev_err_probe(it->dev, PTR_ERR(it->lvds_i2c), + "failed to allocate I2C device for LVDS\n"); it->lvds_regmap = devm_regmap_init_i2c(it->lvds_i2c, &it6263_lvds_regmap_config); diff --git a/drivers/gpu/drm/display/drm_bridge_connector.c b/drivers/gpu/drm/display/drm_bridge_connector.c index 320c297008aa..3a5a4f92c979 100644 --- a/drivers/gpu/drm/display/drm_bridge_connector.c +++ b/drivers/gpu/drm/display/drm_bridge_connector.c @@ -459,7 +459,10 @@ struct drm_connector *drm_bridge_connector_init(struct drm_device *drm, if (connector_type == DRM_MODE_CONNECTOR_Unknown) return ERR_PTR(-EINVAL); - if (bridge_connector->bridge_hdmi) + if (bridge_connector->bridge_hdmi) { + if (!connector->ycbcr_420_allowed) + supported_formats &= ~BIT(HDMI_COLORSPACE_YUV420); + ret = drmm_connector_hdmi_init(drm, connector, bridge_connector->bridge_hdmi->vendor, bridge_connector->bridge_hdmi->product, @@ -468,10 +471,11 @@ struct drm_connector *drm_bridge_connector_init(struct drm_device *drm, connector_type, ddc, supported_formats, max_bpc); - else + } else { ret = drmm_connector_init(drm, connector, &drm_bridge_connector_funcs, connector_type, ddc); + } if (ret) return ERR_PTR(ret); diff --git a/drivers/gpu/drm/drm_bridge.c b/drivers/gpu/drm/drm_bridge.c index c6af46dd02bf..241a384ebce3 100644 --- a/drivers/gpu/drm/drm_bridge.c +++ b/drivers/gpu/drm/drm_bridge.c @@ -207,6 +207,10 @@ void drm_bridge_add(struct drm_bridge *bridge) { mutex_init(&bridge->hpd_mutex); + if (bridge->ops & DRM_BRIDGE_OP_HDMI) + bridge->ycbcr_420_allowed = !!(bridge->supported_formats & + BIT(HDMI_COLORSPACE_YUV420)); + mutex_lock(&bridge_lock); list_add_tail(&bridge->list, &bridge_list); mutex_unlock(&bridge_lock); diff --git a/drivers/gpu/drm/drm_connector.c b/drivers/gpu/drm/drm_connector.c index fc35f47e2849..ca7f43c8d6f1 100644 --- a/drivers/gpu/drm/drm_connector.c +++ b/drivers/gpu/drm/drm_connector.c @@ -507,6 +507,9 @@ int drmm_connector_hdmi_init(struct drm_device *dev, if (!supported_formats || !(supported_formats & BIT(HDMI_COLORSPACE_RGB))) return -EINVAL; + if (connector->ycbcr_420_allowed != !!(supported_formats & BIT(HDMI_COLORSPACE_YUV420))) + return -EINVAL; + if (!(max_bpc == 8 || max_bpc == 10 || max_bpc == 12)) return -EINVAL; diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy.c b/drivers/gpu/drm/i915/display/intel_cx0_phy.c index 71dc659228ab..0c7aee13495a 100644 --- a/drivers/gpu/drm/i915/display/intel_cx0_phy.c +++ b/drivers/gpu/drm/i915/display/intel_cx0_phy.c @@ -2115,14 +2115,6 @@ static void intel_c10_pll_program(struct intel_display *display, 0, C10_VDR_CTRL_MSGBUS_ACCESS, MB_WRITE_COMMITTED); - /* Custom width needs to be programmed to 0 for both the phy lanes */ - intel_cx0_rmw(encoder, INTEL_CX0_BOTH_LANES, PHY_C10_VDR_CUSTOM_WIDTH, - C10_VDR_CUSTOM_WIDTH_MASK, C10_VDR_CUSTOM_WIDTH_8_10, - MB_WRITE_COMMITTED); - intel_cx0_rmw(encoder, INTEL_CX0_BOTH_LANES, PHY_C10_VDR_CONTROL(1), - 0, C10_VDR_CTRL_UPDATE_CFG, - MB_WRITE_COMMITTED); - /* Program the pll values only for the master lane */ for (i = 0; i < ARRAY_SIZE(pll_state->pll); i++) intel_cx0_write(encoder, INTEL_CX0_LANE0, PHY_C10_VDR_PLL(i), @@ -2132,6 +2124,10 @@ static void intel_c10_pll_program(struct intel_display *display, intel_cx0_write(encoder, INTEL_CX0_LANE0, PHY_C10_VDR_CMN(0), pll_state->cmn, MB_WRITE_COMMITTED); intel_cx0_write(encoder, INTEL_CX0_LANE0, PHY_C10_VDR_TX(0), pll_state->tx, MB_WRITE_COMMITTED); + /* Custom width needs to be programmed to 0 for both the phy lanes */ + intel_cx0_rmw(encoder, INTEL_CX0_BOTH_LANES, PHY_C10_VDR_CUSTOM_WIDTH, + C10_VDR_CUSTOM_WIDTH_MASK, C10_VDR_CUSTOM_WIDTH_8_10, + MB_WRITE_COMMITTED); intel_cx0_rmw(encoder, INTEL_CX0_LANE0, PHY_C10_VDR_CONTROL(1), 0, C10_VDR_CTRL_MASTER_LANE | C10_VDR_CTRL_UPDATE_CFG, MB_WRITE_COMMITTED); diff --git a/drivers/gpu/drm/i915/display/intel_fb.c b/drivers/gpu/drm/i915/display/intel_fb.c index 6a7060889f40..223c4218c019 100644 --- a/drivers/gpu/drm/i915/display/intel_fb.c +++ b/drivers/gpu/drm/i915/display/intel_fb.c @@ -1694,7 +1694,7 @@ int intel_fill_fb_info(struct drm_i915_private *i915, struct intel_framebuffer * * arithmetic related to alignment and offset calculation. */ if (is_gen12_ccs_cc_plane(&fb->base, i)) { - if (IS_ALIGNED(fb->base.offsets[i], PAGE_SIZE)) + if (IS_ALIGNED(fb->base.offsets[i], 64)) continue; else return -EINVAL; diff --git a/drivers/gpu/drm/i915/display/intel_hdcp.c b/drivers/gpu/drm/i915/display/intel_hdcp.c index f57e4dba2873..8fee26d791f4 100644 --- a/drivers/gpu/drm/i915/display/intel_hdcp.c +++ b/drivers/gpu/drm/i915/display/intel_hdcp.c @@ -1158,9 +1158,15 @@ static int intel_hdcp_check_link(struct intel_connector *connector) goto out; } - intel_hdcp_update_value(connector, - DRM_MODE_CONTENT_PROTECTION_DESIRED, - true); + ret = intel_hdcp1_enable(connector); + if (ret) { + drm_err(display->drm, "Failed to enable hdcp (%d)\n", ret); + intel_hdcp_update_value(connector, + DRM_MODE_CONTENT_PROTECTION_DESIRED, + true); + goto out; + } + out: mutex_unlock(&dig_port->hdcp_mutex); mutex_unlock(&hdcp->mutex); diff --git a/drivers/gpu/drm/i915/gt/intel_rc6.c b/drivers/gpu/drm/i915/gt/intel_rc6.c index c864d101faf9..9378d5901c49 100644 --- a/drivers/gpu/drm/i915/gt/intel_rc6.c +++ b/drivers/gpu/drm/i915/gt/intel_rc6.c @@ -133,7 +133,7 @@ static void gen11_rc6_enable(struct intel_rc6 *rc6) GEN9_MEDIA_PG_ENABLE | GEN11_MEDIA_SAMPLER_PG_ENABLE; - if (GRAPHICS_VER(gt->i915) >= 12) { + if (GRAPHICS_VER(gt->i915) >= 12 && !IS_DG1(gt->i915)) { for (i = 0; i < I915_MAX_VCS; i++) if (HAS_ENGINE(gt, _VCS(i))) pg_enable |= (VDN_HCP_POWERGATE_ENABLE(i) | diff --git a/drivers/gpu/drm/mediatek/Kconfig b/drivers/gpu/drm/mediatek/Kconfig index f496e6cfdfe0..e47debd60619 100644 --- a/drivers/gpu/drm/mediatek/Kconfig +++ b/drivers/gpu/drm/mediatek/Kconfig @@ -14,9 +14,6 @@ config DRM_MEDIATEK select DRM_BRIDGE_CONNECTOR select DRM_MIPI_DSI select DRM_PANEL - select MEMORY - select MTK_SMI - select PHY_MTK_MIPI_DSI select VIDEOMODE_HELPERS help Choose this option if you have a Mediatek SoCs. @@ -27,7 +24,6 @@ config DRM_MEDIATEK config DRM_MEDIATEK_DP tristate "DRM DPTX Support for MediaTek SoCs" depends on DRM_MEDIATEK - select PHY_MTK_DP select DRM_DISPLAY_HELPER select DRM_DISPLAY_DP_HELPER select DRM_DISPLAY_DP_AUX_BUS @@ -38,6 +34,5 @@ config DRM_MEDIATEK_HDMI tristate "DRM HDMI Support for Mediatek SoCs" depends on DRM_MEDIATEK select SND_SOC_HDMI_CODEC if SND_SOC - select PHY_MTK_HDMI help DRM/KMS HDMI driver for Mediatek SoCs diff --git a/drivers/gpu/drm/mediatek/mtk_crtc.c b/drivers/gpu/drm/mediatek/mtk_crtc.c index eb0e1233ad04..5674f5707cca 100644 --- a/drivers/gpu/drm/mediatek/mtk_crtc.c +++ b/drivers/gpu/drm/mediatek/mtk_crtc.c @@ -112,6 +112,11 @@ static void mtk_drm_finish_page_flip(struct mtk_crtc *mtk_crtc) drm_crtc_handle_vblank(&mtk_crtc->base); +#if IS_REACHABLE(CONFIG_MTK_CMDQ) + if (mtk_crtc->cmdq_client.chan) + return; +#endif + spin_lock_irqsave(&mtk_crtc->config_lock, flags); if (!mtk_crtc->config_updating && mtk_crtc->pending_needs_vblank) { mtk_crtc_finish_page_flip(mtk_crtc); @@ -284,10 +289,8 @@ static void ddp_cmdq_cb(struct mbox_client *cl, void *mssg) state = to_mtk_crtc_state(mtk_crtc->base.state); spin_lock_irqsave(&mtk_crtc->config_lock, flags); - if (mtk_crtc->config_updating) { - spin_unlock_irqrestore(&mtk_crtc->config_lock, flags); + if (mtk_crtc->config_updating) goto ddp_cmdq_cb_out; - } state->pending_config = false; @@ -315,10 +318,15 @@ static void ddp_cmdq_cb(struct mbox_client *cl, void *mssg) mtk_crtc->pending_async_planes = false; } - spin_unlock_irqrestore(&mtk_crtc->config_lock, flags); - ddp_cmdq_cb_out: + if (mtk_crtc->pending_needs_vblank) { + mtk_crtc_finish_page_flip(mtk_crtc); + mtk_crtc->pending_needs_vblank = false; + } + + spin_unlock_irqrestore(&mtk_crtc->config_lock, flags); + mtk_crtc->cmdq_vblank_cnt = 0; wake_up(&mtk_crtc->cb_blocking_queue); } @@ -606,13 +614,18 @@ static void mtk_crtc_update_config(struct mtk_crtc *mtk_crtc, bool needs_vblank) */ mtk_crtc->cmdq_vblank_cnt = 3; + spin_lock_irqsave(&mtk_crtc->config_lock, flags); + mtk_crtc->config_updating = false; + spin_unlock_irqrestore(&mtk_crtc->config_lock, flags); + mbox_send_message(mtk_crtc->cmdq_client.chan, cmdq_handle); mbox_client_txdone(mtk_crtc->cmdq_client.chan, 0); } -#endif +#else spin_lock_irqsave(&mtk_crtc->config_lock, flags); mtk_crtc->config_updating = false; spin_unlock_irqrestore(&mtk_crtc->config_lock, flags); +#endif mutex_unlock(&mtk_crtc->hw_lock); } diff --git a/drivers/gpu/drm/mediatek/mtk_disp_ovl.c b/drivers/gpu/drm/mediatek/mtk_disp_ovl.c index f731d4fbe8b6..df82cea4bb79 100644 --- a/drivers/gpu/drm/mediatek/mtk_disp_ovl.c +++ b/drivers/gpu/drm/mediatek/mtk_disp_ovl.c @@ -460,6 +460,29 @@ static unsigned int mtk_ovl_fmt_convert(struct mtk_disp_ovl *ovl, } } +static void mtk_ovl_afbc_layer_config(struct mtk_disp_ovl *ovl, + unsigned int idx, + struct mtk_plane_pending_state *pending, + struct cmdq_pkt *cmdq_pkt) +{ + unsigned int pitch_msb = pending->pitch >> 16; + unsigned int hdr_pitch = pending->hdr_pitch; + unsigned int hdr_addr = pending->hdr_addr; + + if (pending->modifier != DRM_FORMAT_MOD_LINEAR) { + mtk_ddp_write_relaxed(cmdq_pkt, hdr_addr, &ovl->cmdq_reg, ovl->regs, + DISP_REG_OVL_HDR_ADDR(ovl, idx)); + mtk_ddp_write_relaxed(cmdq_pkt, + OVL_PITCH_MSB_2ND_SUBBUF | pitch_msb, + &ovl->cmdq_reg, ovl->regs, DISP_REG_OVL_PITCH_MSB(idx)); + mtk_ddp_write_relaxed(cmdq_pkt, hdr_pitch, &ovl->cmdq_reg, ovl->regs, + DISP_REG_OVL_HDR_PITCH(ovl, idx)); + } else { + mtk_ddp_write_relaxed(cmdq_pkt, pitch_msb, + &ovl->cmdq_reg, ovl->regs, DISP_REG_OVL_PITCH_MSB(idx)); + } +} + void mtk_ovl_layer_config(struct device *dev, unsigned int idx, struct mtk_plane_state *state, struct cmdq_pkt *cmdq_pkt) @@ -467,25 +490,14 @@ void mtk_ovl_layer_config(struct device *dev, unsigned int idx, struct mtk_disp_ovl *ovl = dev_get_drvdata(dev); struct mtk_plane_pending_state *pending = &state->pending; unsigned int addr = pending->addr; - unsigned int hdr_addr = pending->hdr_addr; - unsigned int pitch = pending->pitch; - unsigned int hdr_pitch = pending->hdr_pitch; + unsigned int pitch_lsb = pending->pitch & GENMASK(15, 0); unsigned int fmt = pending->format; + unsigned int rotation = pending->rotation; unsigned int offset = (pending->y << 16) | pending->x; unsigned int src_size = (pending->height << 16) | pending->width; unsigned int blend_mode = state->base.pixel_blend_mode; unsigned int ignore_pixel_alpha = 0; unsigned int con; - bool is_afbc = pending->modifier != DRM_FORMAT_MOD_LINEAR; - union overlay_pitch { - struct split_pitch { - u16 lsb; - u16 msb; - } split_pitch; - u32 pitch; - } overlay_pitch; - - overlay_pitch.pitch = pitch; if (!pending->enable) { mtk_ovl_layer_off(dev, idx, cmdq_pkt); @@ -513,22 +525,30 @@ void mtk_ovl_layer_config(struct device *dev, unsigned int idx, ignore_pixel_alpha = OVL_CONST_BLEND; } - if (pending->rotation & DRM_MODE_REFLECT_Y) { + /* + * Treat rotate 180 as flip x + flip y, and XOR the original rotation value + * to flip x + flip y to support both in the same time. + */ + if (rotation & DRM_MODE_ROTATE_180) + rotation ^= DRM_MODE_REFLECT_X | DRM_MODE_REFLECT_Y; + + if (rotation & DRM_MODE_REFLECT_Y) { con |= OVL_CON_VIRT_FLIP; addr += (pending->height - 1) * pending->pitch; } - if (pending->rotation & DRM_MODE_REFLECT_X) { + if (rotation & DRM_MODE_REFLECT_X) { con |= OVL_CON_HORZ_FLIP; addr += pending->pitch - 1; } if (ovl->data->supports_afbc) - mtk_ovl_set_afbc(ovl, cmdq_pkt, idx, is_afbc); + mtk_ovl_set_afbc(ovl, cmdq_pkt, idx, + pending->modifier != DRM_FORMAT_MOD_LINEAR); mtk_ddp_write_relaxed(cmdq_pkt, con, &ovl->cmdq_reg, ovl->regs, DISP_REG_OVL_CON(idx)); - mtk_ddp_write_relaxed(cmdq_pkt, overlay_pitch.split_pitch.lsb | ignore_pixel_alpha, + mtk_ddp_write_relaxed(cmdq_pkt, pitch_lsb | ignore_pixel_alpha, &ovl->cmdq_reg, ovl->regs, DISP_REG_OVL_PITCH(idx)); mtk_ddp_write_relaxed(cmdq_pkt, src_size, &ovl->cmdq_reg, ovl->regs, DISP_REG_OVL_SRC_SIZE(idx)); @@ -537,19 +557,8 @@ void mtk_ovl_layer_config(struct device *dev, unsigned int idx, mtk_ddp_write_relaxed(cmdq_pkt, addr, &ovl->cmdq_reg, ovl->regs, DISP_REG_OVL_ADDR(ovl, idx)); - if (is_afbc) { - mtk_ddp_write_relaxed(cmdq_pkt, hdr_addr, &ovl->cmdq_reg, ovl->regs, - DISP_REG_OVL_HDR_ADDR(ovl, idx)); - mtk_ddp_write_relaxed(cmdq_pkt, - OVL_PITCH_MSB_2ND_SUBBUF | overlay_pitch.split_pitch.msb, - &ovl->cmdq_reg, ovl->regs, DISP_REG_OVL_PITCH_MSB(idx)); - mtk_ddp_write_relaxed(cmdq_pkt, hdr_pitch, &ovl->cmdq_reg, ovl->regs, - DISP_REG_OVL_HDR_PITCH(ovl, idx)); - } else { - mtk_ddp_write_relaxed(cmdq_pkt, - overlay_pitch.split_pitch.msb, - &ovl->cmdq_reg, ovl->regs, DISP_REG_OVL_PITCH_MSB(idx)); - } + if (ovl->data->supports_afbc) + mtk_ovl_afbc_layer_config(ovl, idx, pending, cmdq_pkt); mtk_ovl_set_bit_depth(dev, idx, fmt, cmdq_pkt); mtk_ovl_layer_on(dev, idx, cmdq_pkt); diff --git a/drivers/gpu/drm/mediatek/mtk_dp.c b/drivers/gpu/drm/mediatek/mtk_dp.c index 36713c176cfc..3b0993abfdc3 100644 --- a/drivers/gpu/drm/mediatek/mtk_dp.c +++ b/drivers/gpu/drm/mediatek/mtk_dp.c @@ -543,18 +543,16 @@ static int mtk_dp_set_color_format(struct mtk_dp *mtk_dp, enum dp_pixelformat color_format) { u32 val; - - /* update MISC0 */ - mtk_dp_update_bits(mtk_dp, MTK_DP_ENC0_P0_3034, - color_format << DP_TEST_COLOR_FORMAT_SHIFT, - DP_TEST_COLOR_FORMAT_MASK); + u32 misc0_color; switch (color_format) { case DP_PIXELFORMAT_YUV422: val = PIXEL_ENCODE_FORMAT_DP_ENC0_P0_YCBCR422; + misc0_color = DP_COLOR_FORMAT_YCbCr422; break; case DP_PIXELFORMAT_RGB: val = PIXEL_ENCODE_FORMAT_DP_ENC0_P0_RGB; + misc0_color = DP_COLOR_FORMAT_RGB; break; default: drm_warn(mtk_dp->drm_dev, "Unsupported color format: %d\n", @@ -562,6 +560,11 @@ static int mtk_dp_set_color_format(struct mtk_dp *mtk_dp, return -EINVAL; } + /* update MISC0 */ + mtk_dp_update_bits(mtk_dp, MTK_DP_ENC0_P0_3034, + misc0_color, + DP_TEST_COLOR_FORMAT_MASK); + mtk_dp_update_bits(mtk_dp, MTK_DP_ENC0_P0_303C, val, PIXEL_ENCODE_FORMAT_DP_ENC0_P0_MASK); return 0; @@ -2100,7 +2103,6 @@ static enum drm_connector_status mtk_dp_bdg_detect(struct drm_bridge *bridge) struct mtk_dp *mtk_dp = mtk_dp_from_bridge(bridge); enum drm_connector_status ret = connector_status_disconnected; bool enabled = mtk_dp->enabled; - u8 sink_count = 0; if (!mtk_dp->train_info.cable_plugged_in) return ret; @@ -2115,8 +2117,8 @@ static enum drm_connector_status mtk_dp_bdg_detect(struct drm_bridge *bridge) * function, we just need to check the HPD connection to check * whether we connect to a sink device. */ - drm_dp_dpcd_readb(&mtk_dp->aux, DP_SINK_COUNT, &sink_count); - if (DP_GET_SINK_COUNT(sink_count)) + + if (drm_dp_read_sink_count(&mtk_dp->aux) > 0) ret = connector_status_connected; if (!enabled) @@ -2408,12 +2410,19 @@ mtk_dp_bridge_mode_valid(struct drm_bridge *bridge, { struct mtk_dp *mtk_dp = mtk_dp_from_bridge(bridge); u32 bpp = info->color_formats & DRM_COLOR_FORMAT_YCBCR422 ? 16 : 24; - u32 rate = min_t(u32, drm_dp_max_link_rate(mtk_dp->rx_cap) * - drm_dp_max_lane_count(mtk_dp->rx_cap), - drm_dp_bw_code_to_link_rate(mtk_dp->max_linkrate) * - mtk_dp->max_lanes); + u32 lane_count_min = mtk_dp->train_info.lane_count; + u32 rate = drm_dp_bw_code_to_link_rate(mtk_dp->train_info.link_rate) * + lane_count_min; - if (rate < mode->clock * bpp / 8) + /* + *FEC overhead is approximately 2.4% from DP 1.4a spec 2.2.1.4.2. + *The down-spread amplitude shall either be disabled (0.0%) or up + *to 0.5% from 1.4a 3.5.2.6. Add up to approximately 3% total overhead. + * + *Because rate is already divided by 10, + *mode->clock does not need to be multiplied by 10 + */ + if ((rate * 97 / 100) < (mode->clock * bpp / 8)) return MODE_CLOCK_HIGH; return MODE_OK; @@ -2454,10 +2463,9 @@ static u32 *mtk_dp_bridge_atomic_get_input_bus_fmts(struct drm_bridge *bridge, struct drm_display_mode *mode = &crtc_state->adjusted_mode; struct drm_display_info *display_info = &conn_state->connector->display_info; - u32 rate = min_t(u32, drm_dp_max_link_rate(mtk_dp->rx_cap) * - drm_dp_max_lane_count(mtk_dp->rx_cap), - drm_dp_bw_code_to_link_rate(mtk_dp->max_linkrate) * - mtk_dp->max_lanes); + u32 lane_count_min = mtk_dp->train_info.lane_count; + u32 rate = drm_dp_bw_code_to_link_rate(mtk_dp->train_info.link_rate) * + lane_count_min; *num_input_fmts = 0; @@ -2466,8 +2474,8 @@ static u32 *mtk_dp_bridge_atomic_get_input_bus_fmts(struct drm_bridge *bridge, * datarate of YUV422 and sink device supports YUV422, we output YUV422 * format. Use this condition, we can support more resolution. */ - if ((rate < (mode->clock * 24 / 8)) && - (rate > (mode->clock * 16 / 8)) && + if (((rate * 97 / 100) < (mode->clock * 24 / 8)) && + ((rate * 97 / 100) > (mode->clock * 16 / 8)) && (display_info->color_formats & DRM_COLOR_FORMAT_YCBCR422)) { input_fmts = kcalloc(1, sizeof(*input_fmts), GFP_KERNEL); if (!input_fmts) diff --git a/drivers/gpu/drm/mediatek/mtk_drm_drv.c b/drivers/gpu/drm/mediatek/mtk_drm_drv.c index 0829ceb9967c..b037616d3342 100644 --- a/drivers/gpu/drm/mediatek/mtk_drm_drv.c +++ b/drivers/gpu/drm/mediatek/mtk_drm_drv.c @@ -373,11 +373,12 @@ static bool mtk_drm_get_all_drm_priv(struct device *dev) struct mtk_drm_private *temp_drm_priv; struct device_node *phandle = dev->parent->of_node; const struct of_device_id *of_id; + struct device_node *node; struct device *drm_dev; unsigned int cnt = 0; int i, j; - for_each_child_of_node_scoped(phandle->parent, node) { + for_each_child_of_node(phandle->parent, node) { struct platform_device *pdev; of_id = of_match_node(mtk_drm_of_ids, node); @@ -406,8 +407,10 @@ static bool mtk_drm_get_all_drm_priv(struct device *dev) if (temp_drm_priv->mtk_drm_bound) cnt++; - if (cnt == MAX_CRTC) + if (cnt == MAX_CRTC) { + of_node_put(node); break; + } } if (drm_priv->data->mmsys_dev_num == cnt) { @@ -673,6 +676,8 @@ err_deinit: err_free: private->drm = NULL; drm_dev_put(drm); + for (i = 0; i < private->data->mmsys_dev_num; i++) + private->all_drm_private[i]->drm = NULL; return ret; } @@ -900,7 +905,7 @@ static int mtk_drm_of_ddp_path_build_one(struct device *dev, enum mtk_crtc_path const unsigned int **out_path, unsigned int *out_path_len) { - struct device_node *next, *prev, *vdo = dev->parent->of_node; + struct device_node *next = NULL, *prev, *vdo = dev->parent->of_node; unsigned int temp_path[DDP_COMPONENT_DRM_ID_MAX] = { 0 }; unsigned int *final_ddp_path; unsigned short int idx = 0; @@ -1089,7 +1094,7 @@ static int mtk_drm_probe(struct platform_device *pdev) /* No devicetree graphs support: go with hardcoded paths if present */ dev_dbg(dev, "Using hardcoded paths for MMSYS %u\n", mtk_drm_data->mmsys_id); private->data = mtk_drm_data; - }; + } private->all_drm_private = devm_kmalloc_array(dev, private->data->mmsys_dev_num, sizeof(*private->all_drm_private), diff --git a/drivers/gpu/drm/mediatek/mtk_dsi.c b/drivers/gpu/drm/mediatek/mtk_dsi.c index e61b9bc68e9a..40752f232054 100644 --- a/drivers/gpu/drm/mediatek/mtk_dsi.c +++ b/drivers/gpu/drm/mediatek/mtk_dsi.c @@ -139,11 +139,11 @@ #define CLK_HS_POST GENMASK(15, 8) #define CLK_HS_EXIT GENMASK(23, 16) -#define DSI_VM_CMD_CON 0x130 +/* DSI_VM_CMD_CON */ #define VM_CMD_EN BIT(0) #define TS_VFP_EN BIT(5) -#define DSI_SHADOW_DEBUG 0x190U +/* DSI_SHADOW_DEBUG */ #define FORCE_COMMIT BIT(0) #define BYPASS_SHADOW BIT(1) @@ -187,6 +187,8 @@ struct phy; struct mtk_dsi_driver_data { const u32 reg_cmdq_off; + const u32 reg_vm_cmd_off; + const u32 reg_shadow_dbg_off; bool has_shadow_ctl; bool has_size_ctl; bool cmdq_long_packet_ctl; @@ -246,23 +248,22 @@ static void mtk_dsi_phy_timconfig(struct mtk_dsi *dsi) u32 data_rate_mhz = DIV_ROUND_UP(dsi->data_rate, HZ_PER_MHZ); struct mtk_phy_timing *timing = &dsi->phy_timing; - timing->lpx = (80 * data_rate_mhz / (8 * 1000)) + 1; - timing->da_hs_prepare = (59 * data_rate_mhz + 4 * 1000) / 8000 + 1; - timing->da_hs_zero = (163 * data_rate_mhz + 11 * 1000) / 8000 + 1 - + timing->lpx = (60 * data_rate_mhz / (8 * 1000)) + 1; + timing->da_hs_prepare = (80 * data_rate_mhz + 4 * 1000) / 8000; + timing->da_hs_zero = (170 * data_rate_mhz + 10 * 1000) / 8000 + 1 - timing->da_hs_prepare; - timing->da_hs_trail = (78 * data_rate_mhz + 7 * 1000) / 8000 + 1; + timing->da_hs_trail = timing->da_hs_prepare + 1; - timing->ta_go = 4 * timing->lpx; - timing->ta_sure = 3 * timing->lpx / 2; - timing->ta_get = 5 * timing->lpx; - timing->da_hs_exit = (118 * data_rate_mhz / (8 * 1000)) + 1; + timing->ta_go = 4 * timing->lpx - 2; + timing->ta_sure = timing->lpx + 2; + timing->ta_get = 4 * timing->lpx; + timing->da_hs_exit = 2 * timing->lpx + 1; - timing->clk_hs_prepare = (57 * data_rate_mhz / (8 * 1000)) + 1; - timing->clk_hs_post = (65 * data_rate_mhz + 53 * 1000) / 8000 + 1; - timing->clk_hs_trail = (78 * data_rate_mhz + 7 * 1000) / 8000 + 1; - timing->clk_hs_zero = (330 * data_rate_mhz / (8 * 1000)) + 1 - - timing->clk_hs_prepare; - timing->clk_hs_exit = (118 * data_rate_mhz / (8 * 1000)) + 1; + timing->clk_hs_prepare = 70 * data_rate_mhz / (8 * 1000); + timing->clk_hs_post = timing->clk_hs_prepare + 8; + timing->clk_hs_trail = timing->clk_hs_prepare; + timing->clk_hs_zero = timing->clk_hs_trail * 4; + timing->clk_hs_exit = 2 * timing->clk_hs_trail; timcon0 = FIELD_PREP(LPX, timing->lpx) | FIELD_PREP(HS_PREP, timing->da_hs_prepare) | @@ -367,8 +368,8 @@ static void mtk_dsi_set_mode(struct mtk_dsi *dsi) static void mtk_dsi_set_vm_cmd(struct mtk_dsi *dsi) { - mtk_dsi_mask(dsi, DSI_VM_CMD_CON, VM_CMD_EN, VM_CMD_EN); - mtk_dsi_mask(dsi, DSI_VM_CMD_CON, TS_VFP_EN, TS_VFP_EN); + mtk_dsi_mask(dsi, dsi->driver_data->reg_vm_cmd_off, VM_CMD_EN, VM_CMD_EN); + mtk_dsi_mask(dsi, dsi->driver_data->reg_vm_cmd_off, TS_VFP_EN, TS_VFP_EN); } static void mtk_dsi_rxtx_control(struct mtk_dsi *dsi) @@ -714,7 +715,7 @@ static int mtk_dsi_poweron(struct mtk_dsi *dsi) if (dsi->driver_data->has_shadow_ctl) writel(FORCE_COMMIT | BYPASS_SHADOW, - dsi->regs + DSI_SHADOW_DEBUG); + dsi->regs + dsi->driver_data->reg_shadow_dbg_off); mtk_dsi_reset_engine(dsi); mtk_dsi_phy_timconfig(dsi); @@ -1263,26 +1264,36 @@ static void mtk_dsi_remove(struct platform_device *pdev) static const struct mtk_dsi_driver_data mt8173_dsi_driver_data = { .reg_cmdq_off = 0x200, + .reg_vm_cmd_off = 0x130, + .reg_shadow_dbg_off = 0x190 }; static const struct mtk_dsi_driver_data mt2701_dsi_driver_data = { .reg_cmdq_off = 0x180, + .reg_vm_cmd_off = 0x130, + .reg_shadow_dbg_off = 0x190 }; static const struct mtk_dsi_driver_data mt8183_dsi_driver_data = { .reg_cmdq_off = 0x200, + .reg_vm_cmd_off = 0x130, + .reg_shadow_dbg_off = 0x190, .has_shadow_ctl = true, .has_size_ctl = true, }; static const struct mtk_dsi_driver_data mt8186_dsi_driver_data = { .reg_cmdq_off = 0xd00, + .reg_vm_cmd_off = 0x200, + .reg_shadow_dbg_off = 0xc00, .has_shadow_ctl = true, .has_size_ctl = true, }; static const struct mtk_dsi_driver_data mt8188_dsi_driver_data = { .reg_cmdq_off = 0xd00, + .reg_vm_cmd_off = 0x200, + .reg_shadow_dbg_off = 0xc00, .has_shadow_ctl = true, .has_size_ctl = true, .cmdq_long_packet_ctl = true, diff --git a/drivers/gpu/drm/nouveau/nouveau_acpi.c b/drivers/gpu/drm/nouveau/nouveau_acpi.c index 8f0c69aad248..21b56cc7605c 100644 --- a/drivers/gpu/drm/nouveau/nouveau_acpi.c +++ b/drivers/gpu/drm/nouveau/nouveau_acpi.c @@ -384,7 +384,7 @@ nouveau_acpi_edid(struct drm_device *dev, struct drm_connector *connector) if (ret < 0) return NULL; - return kmemdup(edid, EDID_LENGTH, GFP_KERNEL); + return edid; } bool nouveau_acpi_video_backlight_use_native(void) diff --git a/drivers/gpu/drm/nouveau/nouveau_fence.c b/drivers/gpu/drm/nouveau/nouveau_fence.c index 09686d038d60..7cc84472cece 100644 --- a/drivers/gpu/drm/nouveau/nouveau_fence.c +++ b/drivers/gpu/drm/nouveau/nouveau_fence.c @@ -387,11 +387,13 @@ nouveau_fence_sync(struct nouveau_bo *nvbo, struct nouveau_channel *chan, if (f) { struct nouveau_channel *prev; bool must_wait = true; + bool local; rcu_read_lock(); prev = rcu_dereference(f->channel); - if (prev && (prev == chan || - fctx->sync(f, prev, chan) == 0)) + local = prev && prev->cli->drm == chan->cli->drm; + if (local && (prev == chan || + fctx->sync(f, prev, chan) == 0)) must_wait = false; rcu_read_unlock(); if (!must_wait) diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/mcp77.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/mcp77.c index 841e3b69fcaf..5a0c9b8a79f3 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/mcp77.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/mcp77.c @@ -31,6 +31,7 @@ mcp77_sor = { .state = g94_sor_state, .power = nv50_sor_power, .clock = nv50_sor_clock, + .bl = &nv50_sor_bl, .hdmi = &g84_sor_hdmi, .dp = &g94_sor_dp, }; diff --git a/drivers/gpu/drm/tests/drm_connector_test.c b/drivers/gpu/drm/tests/drm_connector_test.c index 6bba97d0be88..3d48d676e001 100644 --- a/drivers/gpu/drm/tests/drm_connector_test.c +++ b/drivers/gpu/drm/tests/drm_connector_test.c @@ -635,6 +635,64 @@ static void drm_test_connector_hdmi_init_formats_no_rgb(struct kunit *test) KUNIT_EXPECT_LT(test, ret, 0); } +struct drm_connector_hdmi_init_formats_yuv420_allowed_test { + unsigned long supported_formats; + bool yuv420_allowed; + int expected_result; +}; + +#define YUV420_ALLOWED_TEST(_formats, _allowed, _result) \ + { \ + .supported_formats = BIT(HDMI_COLORSPACE_RGB) | (_formats), \ + .yuv420_allowed = _allowed, \ + .expected_result = _result, \ + } + +static const struct drm_connector_hdmi_init_formats_yuv420_allowed_test +drm_connector_hdmi_init_formats_yuv420_allowed_tests[] = { + YUV420_ALLOWED_TEST(BIT(HDMI_COLORSPACE_YUV420), true, 0), + YUV420_ALLOWED_TEST(BIT(HDMI_COLORSPACE_YUV420), false, -EINVAL), + YUV420_ALLOWED_TEST(BIT(HDMI_COLORSPACE_YUV422), true, -EINVAL), + YUV420_ALLOWED_TEST(BIT(HDMI_COLORSPACE_YUV422), false, 0), +}; + +static void +drm_connector_hdmi_init_formats_yuv420_allowed_desc(const struct drm_connector_hdmi_init_formats_yuv420_allowed_test *t, + char *desc) +{ + sprintf(desc, "supported_formats=0x%lx yuv420_allowed=%d", + t->supported_formats, t->yuv420_allowed); +} + +KUNIT_ARRAY_PARAM(drm_connector_hdmi_init_formats_yuv420_allowed, + drm_connector_hdmi_init_formats_yuv420_allowed_tests, + drm_connector_hdmi_init_formats_yuv420_allowed_desc); + +/* + * Test that the registration of an HDMI connector succeeds only when + * the presence of YUV420 in the supported formats matches the value + * of the ycbcr_420_allowed flag. + */ +static void drm_test_connector_hdmi_init_formats_yuv420_allowed(struct kunit *test) +{ + const struct drm_connector_hdmi_init_formats_yuv420_allowed_test *params; + struct drm_connector_init_priv *priv = test->priv; + int ret; + + params = test->param_value; + priv->connector.ycbcr_420_allowed = params->yuv420_allowed; + + ret = drmm_connector_hdmi_init(&priv->drm, &priv->connector, + "Vendor", "Product", + &dummy_funcs, + &dummy_hdmi_funcs, + DRM_MODE_CONNECTOR_HDMIA, + &priv->ddc, + params->supported_formats, + 8); + KUNIT_EXPECT_EQ(test, ret, params->expected_result); +} + /* * Test that the registration of an HDMI connector with an HDMI * connector type succeeds. @@ -726,6 +784,8 @@ static struct kunit_case drmm_connector_hdmi_init_tests[] = { KUNIT_CASE(drm_test_connector_hdmi_init_bpc_null), KUNIT_CASE(drm_test_connector_hdmi_init_formats_empty), KUNIT_CASE(drm_test_connector_hdmi_init_formats_no_rgb), + KUNIT_CASE_PARAM(drm_test_connector_hdmi_init_formats_yuv420_allowed, + drm_connector_hdmi_init_formats_yuv420_allowed_gen_params), KUNIT_CASE(drm_test_connector_hdmi_init_null_ddc), KUNIT_CASE(drm_test_connector_hdmi_init_null_product), KUNIT_CASE(drm_test_connector_hdmi_init_null_vendor), diff --git a/drivers/gpu/drm/tests/drm_kunit_helpers.c b/drivers/gpu/drm/tests/drm_kunit_helpers.c index 04a6b8cc62ac..3c0b7824c0be 100644 --- a/drivers/gpu/drm/tests/drm_kunit_helpers.c +++ b/drivers/gpu/drm/tests/drm_kunit_helpers.c @@ -320,8 +320,7 @@ static void kunit_action_drm_mode_destroy(void *ptr) } /** - * drm_kunit_display_mode_from_cea_vic() - return a mode for CEA VIC - for a KUnit test + * drm_kunit_display_mode_from_cea_vic() - return a mode for CEA VIC for a KUnit test * @test: The test context object * @dev: DRM device * @video_code: CEA VIC of the mode diff --git a/drivers/gpu/drm/v3d/v3d_irq.c b/drivers/gpu/drm/v3d/v3d_irq.c index 20bf33702c3c..da203045df9b 100644 --- a/drivers/gpu/drm/v3d/v3d_irq.c +++ b/drivers/gpu/drm/v3d/v3d_irq.c @@ -108,6 +108,7 @@ v3d_irq(int irq, void *arg) v3d_job_update_stats(&v3d->bin_job->base, V3D_BIN); trace_v3d_bcl_irq(&v3d->drm, fence->seqno); dma_fence_signal(&fence->base); + v3d->bin_job = NULL; status = IRQ_HANDLED; } @@ -118,6 +119,7 @@ v3d_irq(int irq, void *arg) v3d_job_update_stats(&v3d->render_job->base, V3D_RENDER); trace_v3d_rcl_irq(&v3d->drm, fence->seqno); dma_fence_signal(&fence->base); + v3d->render_job = NULL; status = IRQ_HANDLED; } @@ -128,6 +130,7 @@ v3d_irq(int irq, void *arg) v3d_job_update_stats(&v3d->csd_job->base, V3D_CSD); trace_v3d_csd_irq(&v3d->drm, fence->seqno); dma_fence_signal(&fence->base); + v3d->csd_job = NULL; status = IRQ_HANDLED; } @@ -165,6 +168,7 @@ v3d_hub_irq(int irq, void *arg) v3d_job_update_stats(&v3d->tfu_job->base, V3D_TFU); trace_v3d_tfu_irq(&v3d->drm, fence->seqno); dma_fence_signal(&fence->base); + v3d->tfu_job = NULL; status = IRQ_HANDLED; } diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_bo.c b/drivers/gpu/drm/vmwgfx/vmwgfx_bo.c index a0e433fbcba6..9b5b8c1f063b 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_bo.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_bo.c @@ -228,7 +228,6 @@ int vmw_bo_pin_in_start_of_vram(struct vmw_private *dev_priv, VMW_BO_DOMAIN_VRAM, VMW_BO_DOMAIN_VRAM); buf->places[0].lpfn = PFN_UP(bo->resource->size); - buf->busy_places[0].lpfn = PFN_UP(bo->resource->size); ret = ttm_bo_validate(bo, &buf->placement, &ctx); /* For some reason we didn't end up at the start of vram */ @@ -443,7 +442,8 @@ static int vmw_bo_init(struct vmw_private *dev_priv, if (params->pin) ttm_bo_pin(&vmw_bo->tbo); - ttm_bo_unreserve(&vmw_bo->tbo); + if (!params->keep_resv) + ttm_bo_unreserve(&vmw_bo->tbo); return 0; } diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_bo.h b/drivers/gpu/drm/vmwgfx/vmwgfx_bo.h index 43b5439ec9f7..11e330c7c7f5 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_bo.h +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_bo.h @@ -56,8 +56,9 @@ struct vmw_bo_params { u32 domain; u32 busy_domain; enum ttm_bo_type bo_type; - size_t size; bool pin; + bool keep_resv; + size_t size; struct dma_resv *resv; struct sg_table *sg; }; @@ -83,7 +84,6 @@ struct vmw_bo { struct ttm_placement placement; struct ttm_place places[5]; - struct ttm_place busy_places[5]; /* Protected by reservation */ struct ttm_bo_kmap_obj map; diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c index 2c46897876dd..b19a062592b0 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c @@ -403,7 +403,8 @@ static int vmw_dummy_query_bo_create(struct vmw_private *dev_priv) .busy_domain = VMW_BO_DOMAIN_SYS, .bo_type = ttm_bo_type_kernel, .size = PAGE_SIZE, - .pin = true + .pin = true, + .keep_resv = true, }; /* @@ -415,10 +416,6 @@ static int vmw_dummy_query_bo_create(struct vmw_private *dev_priv) if (unlikely(ret != 0)) return ret; - ret = ttm_bo_reserve(&vbo->tbo, false, true, NULL); - BUG_ON(ret != 0); - vmw_bo_pin_reserved(vbo, true); - ret = ttm_bo_kmap(&vbo->tbo, 0, 1, &map); if (likely(ret == 0)) { result = ttm_kmap_obj_virtual(&map, &dummy); diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_gem.c b/drivers/gpu/drm/vmwgfx/vmwgfx_gem.c index b9857f37ca1a..ed5015ced392 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_gem.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_gem.c @@ -206,6 +206,7 @@ struct drm_gem_object *vmw_prime_import_sg_table(struct drm_device *dev, .bo_type = ttm_bo_type_sg, .size = attach->dmabuf->size, .pin = false, + .keep_resv = true, .resv = attach->dmabuf->resv, .sg = table, diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c index 8db38927729b..800a79e035ed 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c @@ -750,6 +750,7 @@ vmw_du_cursor_plane_atomic_update(struct drm_plane *plane, struct vmw_plane_state *old_vps = vmw_plane_state_to_vps(old_state); struct vmw_bo *old_bo = NULL; struct vmw_bo *new_bo = NULL; + struct ww_acquire_ctx ctx; s32 hotspot_x, hotspot_y; int ret; @@ -769,9 +770,11 @@ vmw_du_cursor_plane_atomic_update(struct drm_plane *plane, if (du->cursor_surface) du->cursor_age = du->cursor_surface->snooper.age; + ww_acquire_init(&ctx, &reservation_ww_class); + if (!vmw_user_object_is_null(&old_vps->uo)) { old_bo = vmw_user_object_buffer(&old_vps->uo); - ret = ttm_bo_reserve(&old_bo->tbo, false, false, NULL); + ret = ttm_bo_reserve(&old_bo->tbo, false, false, &ctx); if (ret != 0) return; } @@ -779,9 +782,14 @@ vmw_du_cursor_plane_atomic_update(struct drm_plane *plane, if (!vmw_user_object_is_null(&vps->uo)) { new_bo = vmw_user_object_buffer(&vps->uo); if (old_bo != new_bo) { - ret = ttm_bo_reserve(&new_bo->tbo, false, false, NULL); - if (ret != 0) + ret = ttm_bo_reserve(&new_bo->tbo, false, false, &ctx); + if (ret != 0) { + if (old_bo) { + ttm_bo_unreserve(&old_bo->tbo); + ww_acquire_fini(&ctx); + } return; + } } else { new_bo = NULL; } @@ -803,10 +811,12 @@ vmw_du_cursor_plane_atomic_update(struct drm_plane *plane, hotspot_x, hotspot_y); } - if (old_bo) - ttm_bo_unreserve(&old_bo->tbo); if (new_bo) ttm_bo_unreserve(&new_bo->tbo); + if (old_bo) + ttm_bo_unreserve(&old_bo->tbo); + + ww_acquire_fini(&ctx); du->cursor_x = new_state->crtc_x + du->set_gui_x; du->cursor_y = new_state->crtc_y + du->set_gui_y; diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_shader.c b/drivers/gpu/drm/vmwgfx/vmwgfx_shader.c index a01ca3226d0a..7fb1c88bcc47 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_shader.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_shader.c @@ -896,7 +896,8 @@ int vmw_compat_shader_add(struct vmw_private *dev_priv, .busy_domain = VMW_BO_DOMAIN_SYS, .bo_type = ttm_bo_type_device, .size = size, - .pin = true + .pin = true, + .keep_resv = true, }; if (!vmw_shader_id_ok(user_key, shader_type)) @@ -906,10 +907,6 @@ int vmw_compat_shader_add(struct vmw_private *dev_priv, if (unlikely(ret != 0)) goto out; - ret = ttm_bo_reserve(&buf->tbo, false, true, NULL); - if (unlikely(ret != 0)) - goto no_reserve; - /* Map and copy shader bytecode. */ ret = ttm_bo_kmap(&buf->tbo, 0, PFN_UP(size), &map); if (unlikely(ret != 0)) { diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c b/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c index 621d98b376bb..5553892d7c3e 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c @@ -572,15 +572,14 @@ int vmw_bo_create_and_populate(struct vmw_private *dev_priv, .busy_domain = domain, .bo_type = ttm_bo_type_kernel, .size = bo_size, - .pin = true + .pin = true, + .keep_resv = true, }; ret = vmw_bo_create(dev_priv, &bo_params, &vbo); if (unlikely(ret != 0)) return ret; - ret = ttm_bo_reserve(&vbo->tbo, false, true, NULL); - BUG_ON(ret != 0); ret = vmw_ttm_populate(vbo->tbo.bdev, vbo->tbo.ttm, &ctx); if (likely(ret == 0)) { struct vmw_ttm_tt *vmw_tt = diff --git a/drivers/gpu/drm/xe/tests/xe_bo.c b/drivers/gpu/drm/xe/tests/xe_bo.c index 3e0ae40ebbd2..812046f5d741 100644 --- a/drivers/gpu/drm/xe/tests/xe_bo.c +++ b/drivers/gpu/drm/xe/tests/xe_bo.c @@ -257,10 +257,9 @@ static int evict_test_run_tile(struct xe_device *xe, struct xe_tile *tile, struc * however seems quite fragile not to also restart the GT. Try * to do that here by triggering a GT reset. */ - for_each_gt(__gt, xe, id) { - xe_gt_reset_async(__gt); - flush_work(&__gt->reset.worker); - } + for_each_gt(__gt, xe, id) + xe_gt_reset(__gt); + if (err) { KUNIT_FAIL(test, "restore kernel err=%pe\n", ERR_PTR(err)); diff --git a/drivers/gpu/drm/xe/tests/xe_mocs.c b/drivers/gpu/drm/xe/tests/xe_mocs.c index 6f9b7a266b41..2698e433c6fa 100644 --- a/drivers/gpu/drm/xe/tests/xe_mocs.c +++ b/drivers/gpu/drm/xe/tests/xe_mocs.c @@ -162,8 +162,7 @@ static int mocs_reset_test_run_device(struct xe_device *xe) if (flags & HAS_LNCF_MOCS) read_l3cc_table(gt, &mocs.table); - xe_gt_reset_async(gt); - flush_work(>->reset.worker); + xe_gt_reset(gt); kunit_info(test, "mocs_reset_test after reset\n"); if (flags & HAS_GLOBAL_MOCS) diff --git a/drivers/gpu/drm/xe/xe_bo.c b/drivers/gpu/drm/xe/xe_bo.c index ae6b337cdc54..f61a8ef38094 100644 --- a/drivers/gpu/drm/xe/xe_bo.c +++ b/drivers/gpu/drm/xe/xe_bo.c @@ -724,7 +724,7 @@ static int xe_bo_move(struct ttm_buffer_object *ttm_bo, bool evict, new_mem->mem_type == XE_PL_SYSTEM) { long timeout = dma_resv_wait_timeout(ttm_bo->base.resv, DMA_RESV_USAGE_BOOKKEEP, - true, + false, MAX_SCHEDULE_TIMEOUT); if (timeout < 0) { ret = timeout; @@ -848,8 +848,16 @@ static int xe_bo_move(struct ttm_buffer_object *ttm_bo, bool evict, out: if ((!ttm_bo->resource || ttm_bo->resource->mem_type == XE_PL_SYSTEM) && - ttm_bo->ttm) + ttm_bo->ttm) { + long timeout = dma_resv_wait_timeout(ttm_bo->base.resv, + DMA_RESV_USAGE_KERNEL, + false, + MAX_SCHEDULE_TIMEOUT); + if (timeout < 0) + ret = timeout; + xe_tt_unmap_sg(ttm_bo->ttm); + } return ret; } diff --git a/drivers/gpu/drm/xe/xe_devcoredump.c b/drivers/gpu/drm/xe/xe_devcoredump.c index f8947e7e917e..21a50d539426 100644 --- a/drivers/gpu/drm/xe/xe_devcoredump.c +++ b/drivers/gpu/drm/xe/xe_devcoredump.c @@ -109,7 +109,11 @@ static ssize_t __xe_devcoredump_read(char *buffer, size_t count, drm_puts(&p, "\n**** GuC CT ****\n"); xe_guc_ct_snapshot_print(ss->guc.ct, &p); - drm_puts(&p, "\n**** Contexts ****\n"); + /* + * Don't add a new section header here because the mesa debug decoder + * tool expects the context information to be in the 'GuC CT' section. + */ + /* drm_puts(&p, "\n**** Contexts ****\n"); */ xe_guc_exec_queue_snapshot_print(ss->ge, &p); drm_puts(&p, "\n**** Job ****\n"); @@ -363,6 +367,15 @@ void xe_print_blob_ascii85(struct drm_printer *p, const char *prefix, char buff[ASCII85_BUFSZ], *line_buff; size_t line_pos = 0; + /* + * Splitting blobs across multiple lines is not compatible with the mesa + * debug decoder tool. Note that even dropping the explicit '\n' below + * doesn't help because the GuC log is so big some underlying implementation + * still splits the lines at 512K characters. So just bail completely for + * the moment. + */ + return; + #define DMESG_MAX_LINE_LEN 800 #define MIN_SPACE (ASCII85_BUFSZ + 2) /* 85 + "\n\0" */ diff --git a/drivers/gpu/drm/xe/xe_exec_queue.c b/drivers/gpu/drm/xe/xe_exec_queue.c index fd0f3b3c9101..268cd3123be9 100644 --- a/drivers/gpu/drm/xe/xe_exec_queue.c +++ b/drivers/gpu/drm/xe/xe_exec_queue.c @@ -8,6 +8,7 @@ #include <linux/nospec.h> #include <drm/drm_device.h> +#include <drm/drm_drv.h> #include <drm/drm_file.h> #include <uapi/drm/xe_drm.h> @@ -762,9 +763,11 @@ bool xe_exec_queue_is_idle(struct xe_exec_queue *q) */ void xe_exec_queue_update_run_ticks(struct xe_exec_queue *q) { + struct xe_device *xe = gt_to_xe(q->gt); struct xe_file *xef; struct xe_lrc *lrc; u32 old_ts, new_ts; + int idx; /* * Jobs that are run during driver load may use an exec_queue, but are @@ -774,6 +777,10 @@ void xe_exec_queue_update_run_ticks(struct xe_exec_queue *q) if (!q->vm || !q->vm->xef) return; + /* Synchronize with unbind while holding the xe file open */ + if (!drm_dev_enter(&xe->drm, &idx)) + return; + xef = q->vm->xef; /* @@ -787,6 +794,8 @@ void xe_exec_queue_update_run_ticks(struct xe_exec_queue *q) lrc = q->lrc[0]; new_ts = xe_lrc_update_timestamp(lrc, &old_ts); xef->run_ticks[q->class] += (new_ts - old_ts) * q->width; + + drm_dev_exit(idx); } /** diff --git a/drivers/gpu/drm/xe/xe_gt.c b/drivers/gpu/drm/xe/xe_gt.c index d6744be01a68..94d468d01253 100644 --- a/drivers/gpu/drm/xe/xe_gt.c +++ b/drivers/gpu/drm/xe/xe_gt.c @@ -387,6 +387,10 @@ int xe_gt_init_early(struct xe_gt *gt) xe_force_wake_init_gt(gt, gt_to_fw(gt)); spin_lock_init(>->global_invl_lock); + err = xe_gt_tlb_invalidation_init_early(gt); + if (err) + return err; + return 0; } @@ -588,10 +592,6 @@ int xe_gt_init(struct xe_gt *gt) xe_hw_fence_irq_init(>->fence_irq[i]); } - err = xe_gt_tlb_invalidation_init(gt); - if (err) - return err; - err = xe_gt_pagefault_init(gt); if (err) return err; diff --git a/drivers/gpu/drm/xe/xe_gt.h b/drivers/gpu/drm/xe/xe_gt.h index 82b9b7f82fca..eb2946963420 100644 --- a/drivers/gpu/drm/xe/xe_gt.h +++ b/drivers/gpu/drm/xe/xe_gt.h @@ -57,6 +57,31 @@ int xe_gt_sanitize_freq(struct xe_gt *gt); void xe_gt_remove(struct xe_gt *gt); /** + * xe_gt_wait_for_reset - wait for gt's async reset to finalize. + * @gt: GT structure + * Return: + * %true if it waited for the work to finish execution, + * %false if there was no scheduled reset or it was done. + */ +static inline bool xe_gt_wait_for_reset(struct xe_gt *gt) +{ + return flush_work(>->reset.worker); +} + +/** + * xe_gt_reset - perform synchronous reset + * @gt: GT structure + * Return: + * %true if it waited for the reset to finish, + * %false if there was no scheduled reset. + */ +static inline bool xe_gt_reset(struct xe_gt *gt) +{ + xe_gt_reset_async(gt); + return xe_gt_wait_for_reset(gt); +} + +/** * xe_gt_any_hw_engine_by_reset_domain - scan the list of engines and return the * first that matches the same reset domain as @class * @gt: GT structure diff --git a/drivers/gpu/drm/xe/xe_gt_ccs_mode.c b/drivers/gpu/drm/xe/xe_gt_ccs_mode.c index b6adfb9f2030..50fffc9ebf62 100644 --- a/drivers/gpu/drm/xe/xe_gt_ccs_mode.c +++ b/drivers/gpu/drm/xe/xe_gt_ccs_mode.c @@ -150,7 +150,7 @@ ccs_mode_store(struct device *kdev, struct device_attribute *attr, xe_gt_info(gt, "Setting compute mode to %d\n", num_engines); gt->ccs_mode = num_engines; xe_gt_record_user_engines(gt); - xe_gt_reset_async(gt); + xe_gt_reset(gt); } mutex_unlock(&xe->drm.filelist_mutex); diff --git a/drivers/gpu/drm/xe/xe_gt_debugfs.c b/drivers/gpu/drm/xe/xe_gt_debugfs.c index 3e8c351a0eab..e7792858b1e4 100644 --- a/drivers/gpu/drm/xe/xe_gt_debugfs.c +++ b/drivers/gpu/drm/xe/xe_gt_debugfs.c @@ -132,11 +132,9 @@ static int force_reset(struct xe_gt *gt, struct drm_printer *p) static int force_reset_sync(struct xe_gt *gt, struct drm_printer *p) { xe_pm_runtime_get(gt_to_xe(gt)); - xe_gt_reset_async(gt); + xe_gt_reset(gt); xe_pm_runtime_put(gt_to_xe(gt)); - flush_work(>->reset.worker); - return 0; } diff --git a/drivers/gpu/drm/xe/xe_gt_idle.c b/drivers/gpu/drm/xe/xe_gt_idle.c index fd80afeef56a..ffd3ba7f6656 100644 --- a/drivers/gpu/drm/xe/xe_gt_idle.c +++ b/drivers/gpu/drm/xe/xe_gt_idle.c @@ -122,10 +122,12 @@ void xe_gt_idle_enable_pg(struct xe_gt *gt) if (!xe_gt_is_media_type(gt)) gtidle->powergate_enable |= RENDER_POWERGATE_ENABLE; - for (i = XE_HW_ENGINE_VCS0, j = 0; i <= XE_HW_ENGINE_VCS7; ++i, ++j) { - if ((gt->info.engine_mask & BIT(i))) - gtidle->powergate_enable |= (VDN_HCP_POWERGATE_ENABLE(j) | - VDN_MFXVDENC_POWERGATE_ENABLE(j)); + if (xe->info.platform != XE_DG1) { + for (i = XE_HW_ENGINE_VCS0, j = 0; i <= XE_HW_ENGINE_VCS7; ++i, ++j) { + if ((gt->info.engine_mask & BIT(i))) + gtidle->powergate_enable |= (VDN_HCP_POWERGATE_ENABLE(j) | + VDN_MFXVDENC_POWERGATE_ENABLE(j)); + } } fw_ref = xe_force_wake_get(gt_to_fw(gt), XE_FW_GT); diff --git a/drivers/gpu/drm/xe/xe_gt_mcr.c b/drivers/gpu/drm/xe/xe_gt_mcr.c index 5013d674e17d..d929a78c26a9 100644 --- a/drivers/gpu/drm/xe/xe_gt_mcr.c +++ b/drivers/gpu/drm/xe/xe_gt_mcr.c @@ -550,9 +550,9 @@ void xe_gt_mcr_set_implicit_defaults(struct xe_gt *gt) * Returns true if the caller should steer to the @group/@instance values * returned. Returns false if the caller need not perform any steering */ -static bool xe_gt_mcr_get_nonterminated_steering(struct xe_gt *gt, - struct xe_reg_mcr reg_mcr, - u8 *group, u8 *instance) +bool xe_gt_mcr_get_nonterminated_steering(struct xe_gt *gt, + struct xe_reg_mcr reg_mcr, + u8 *group, u8 *instance) { const struct xe_reg reg = to_xe_reg(reg_mcr); const struct xe_mmio_range *implicit_ranges; diff --git a/drivers/gpu/drm/xe/xe_gt_mcr.h b/drivers/gpu/drm/xe/xe_gt_mcr.h index c0cd36021c24..bc06520befab 100644 --- a/drivers/gpu/drm/xe/xe_gt_mcr.h +++ b/drivers/gpu/drm/xe/xe_gt_mcr.h @@ -26,6 +26,10 @@ void xe_gt_mcr_unicast_write(struct xe_gt *gt, struct xe_reg_mcr mcr_reg, void xe_gt_mcr_multicast_write(struct xe_gt *gt, struct xe_reg_mcr mcr_reg, u32 value); +bool xe_gt_mcr_get_nonterminated_steering(struct xe_gt *gt, + struct xe_reg_mcr reg_mcr, + u8 *group, u8 *instance); + void xe_gt_mcr_steering_dump(struct xe_gt *gt, struct drm_printer *p); void xe_gt_mcr_get_dss_steering(struct xe_gt *gt, unsigned int dss, u16 *group, u16 *instance); u32 xe_gt_mcr_steering_info_to_dss_id(struct xe_gt *gt, u16 group, u16 instance); diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c b/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c index 192643d63d22..ca49860168f6 100644 --- a/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c +++ b/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c @@ -2046,7 +2046,7 @@ static int pf_validate_vf_config(struct xe_gt *gt, unsigned int vfid) valid_any = valid_any || (valid_ggtt && is_primary); if (IS_DGFX(xe)) { - bool valid_lmem = pf_get_vf_config_ggtt(primary_gt, vfid); + bool valid_lmem = pf_get_vf_config_lmem(primary_gt, vfid); valid_any = valid_any || (valid_lmem && is_primary); valid_all = valid_all && valid_lmem; diff --git a/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.c b/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.c index 6146d1776bda..0a0af413770e 100644 --- a/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.c +++ b/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.c @@ -106,7 +106,7 @@ static void xe_gt_tlb_fence_timeout(struct work_struct *work) } /** - * xe_gt_tlb_invalidation_init - Initialize GT TLB invalidation state + * xe_gt_tlb_invalidation_init_early - Initialize GT TLB invalidation state * @gt: graphics tile * * Initialize GT TLB invalidation state, purely software initialization, should @@ -114,7 +114,7 @@ static void xe_gt_tlb_fence_timeout(struct work_struct *work) * * Return: 0 on success, negative error code on error. */ -int xe_gt_tlb_invalidation_init(struct xe_gt *gt) +int xe_gt_tlb_invalidation_init_early(struct xe_gt *gt) { gt->tlb_invalidation.seqno = 1; INIT_LIST_HEAD(>->tlb_invalidation.pending_fences); diff --git a/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.h b/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.h index 00b1c6c01e8d..672acfcdf0d7 100644 --- a/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.h +++ b/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.h @@ -14,7 +14,8 @@ struct xe_gt; struct xe_guc; struct xe_vma; -int xe_gt_tlb_invalidation_init(struct xe_gt *gt); +int xe_gt_tlb_invalidation_init_early(struct xe_gt *gt); + void xe_gt_tlb_invalidation_reset(struct xe_gt *gt); int xe_gt_tlb_invalidation_ggtt(struct xe_gt *gt); int xe_gt_tlb_invalidation_vma(struct xe_gt *gt, diff --git a/drivers/gpu/drm/xe/xe_guc_ads.c b/drivers/gpu/drm/xe/xe_guc_ads.c index 4e746ae98888..dd7f9d0effa9 100644 --- a/drivers/gpu/drm/xe/xe_guc_ads.c +++ b/drivers/gpu/drm/xe/xe_guc_ads.c @@ -29,6 +29,7 @@ #include "xe_platform_types.h" #include "xe_uc_fw.h" #include "xe_wa.h" +#include "xe_gt_mcr.h" /* Slack of a few additional entries per engine */ #define ADS_REGSET_EXTRA_MAX 8 @@ -701,6 +702,20 @@ static void guc_mmio_regset_write_one(struct xe_guc_ads *ads, .flags = reg.masked ? GUC_REGSET_MASKED : 0, }; + if (reg.mcr) { + struct xe_reg_mcr mcr_reg = XE_REG_MCR(reg.addr); + u8 group, instance; + + bool steer = xe_gt_mcr_get_nonterminated_steering(ads_to_gt(ads), mcr_reg, + &group, &instance); + + if (steer) { + entry.flags |= FIELD_PREP(GUC_REGSET_STEERING_GROUP, group); + entry.flags |= FIELD_PREP(GUC_REGSET_STEERING_INSTANCE, instance); + entry.flags |= GUC_REGSET_STEERING_NEEDED; + } + } + xe_map_memcpy_to(ads_to_xe(ads), regset_map, n_entry * sizeof(entry), &entry, sizeof(entry)); } diff --git a/drivers/gpu/drm/xe/xe_hw_engine.c b/drivers/gpu/drm/xe/xe_hw_engine.c index 1557acee3523..eeeb98f66982 100644 --- a/drivers/gpu/drm/xe/xe_hw_engine.c +++ b/drivers/gpu/drm/xe/xe_hw_engine.c @@ -419,7 +419,7 @@ hw_engine_setup_default_state(struct xe_hw_engine *hwe) * Bspec: 72161 */ const u8 mocs_write_idx = gt->mocs.uc_index; - const u8 mocs_read_idx = hwe->class == XE_ENGINE_CLASS_COMPUTE && + const u8 mocs_read_idx = hwe->class == XE_ENGINE_CLASS_COMPUTE && IS_DGFX(xe) && (GRAPHICS_VER(xe) >= 20 || xe->info.platform == XE_PVC) ? gt->mocs.wb_index : gt->mocs.uc_index; u32 ring_cmd_cctl_val = REG_FIELD_PREP(CMD_CCTL_WRITE_OVERRIDE_MASK, mocs_write_idx) | diff --git a/drivers/gpu/drm/xe/xe_oa.c b/drivers/gpu/drm/xe/xe_oa.c index 8dd55798ab31..678fa40e4cea 100644 --- a/drivers/gpu/drm/xe/xe_oa.c +++ b/drivers/gpu/drm/xe/xe_oa.c @@ -74,12 +74,6 @@ struct xe_oa_config { struct rcu_head rcu; }; -struct flex { - struct xe_reg reg; - u32 offset; - u32 value; -}; - struct xe_oa_open_param { struct xe_file *xef; u32 oa_unit_id; @@ -596,19 +590,38 @@ static __poll_t xe_oa_poll(struct file *file, poll_table *wait) return ret; } +static void xe_oa_lock_vma(struct xe_exec_queue *q) +{ + if (q->vm) { + down_read(&q->vm->lock); + xe_vm_lock(q->vm, false); + } +} + +static void xe_oa_unlock_vma(struct xe_exec_queue *q) +{ + if (q->vm) { + xe_vm_unlock(q->vm); + up_read(&q->vm->lock); + } +} + static struct dma_fence *xe_oa_submit_bb(struct xe_oa_stream *stream, enum xe_oa_submit_deps deps, struct xe_bb *bb) { + struct xe_exec_queue *q = stream->exec_q ?: stream->k_exec_q; struct xe_sched_job *job; struct dma_fence *fence; int err = 0; - /* Kernel configuration is issued on stream->k_exec_q, not stream->exec_q */ - job = xe_bb_create_job(stream->k_exec_q, bb); + xe_oa_lock_vma(q); + + job = xe_bb_create_job(q, bb); if (IS_ERR(job)) { err = PTR_ERR(job); goto exit; } + job->ggtt = true; if (deps == XE_OA_SUBMIT_ADD_DEPS) { for (int i = 0; i < stream->num_syncs && !err; i++) @@ -623,10 +636,13 @@ static struct dma_fence *xe_oa_submit_bb(struct xe_oa_stream *stream, enum xe_oa fence = dma_fence_get(&job->drm.s_fence->finished); xe_sched_job_push(job); + xe_oa_unlock_vma(q); + return fence; err_put_job: xe_sched_job_put(job); exit: + xe_oa_unlock_vma(q); return ERR_PTR(err); } @@ -675,63 +691,19 @@ static void xe_oa_free_configs(struct xe_oa_stream *stream) dma_fence_put(stream->last_fence); } -static void xe_oa_store_flex(struct xe_oa_stream *stream, struct xe_lrc *lrc, - struct xe_bb *bb, const struct flex *flex, u32 count) -{ - u32 offset = xe_bo_ggtt_addr(lrc->bo); - - do { - bb->cs[bb->len++] = MI_STORE_DATA_IMM | MI_SDI_GGTT | MI_SDI_NUM_DW(1); - bb->cs[bb->len++] = offset + flex->offset * sizeof(u32); - bb->cs[bb->len++] = 0; - bb->cs[bb->len++] = flex->value; - - } while (flex++, --count); -} - -static int xe_oa_modify_ctx_image(struct xe_oa_stream *stream, struct xe_lrc *lrc, - const struct flex *flex, u32 count) -{ - struct dma_fence *fence; - struct xe_bb *bb; - int err; - - bb = xe_bb_new(stream->gt, 4 * count, false); - if (IS_ERR(bb)) { - err = PTR_ERR(bb); - goto exit; - } - - xe_oa_store_flex(stream, lrc, bb, flex, count); - - fence = xe_oa_submit_bb(stream, XE_OA_SUBMIT_NO_DEPS, bb); - if (IS_ERR(fence)) { - err = PTR_ERR(fence); - goto free_bb; - } - xe_bb_free(bb, fence); - dma_fence_put(fence); - - return 0; -free_bb: - xe_bb_free(bb, NULL); -exit: - return err; -} - -static int xe_oa_load_with_lri(struct xe_oa_stream *stream, struct xe_oa_reg *reg_lri) +static int xe_oa_load_with_lri(struct xe_oa_stream *stream, struct xe_oa_reg *reg_lri, u32 count) { struct dma_fence *fence; struct xe_bb *bb; int err; - bb = xe_bb_new(stream->gt, 3, false); + bb = xe_bb_new(stream->gt, 2 * count + 1, false); if (IS_ERR(bb)) { err = PTR_ERR(bb); goto exit; } - write_cs_mi_lri(bb, reg_lri, 1); + write_cs_mi_lri(bb, reg_lri, count); fence = xe_oa_submit_bb(stream, XE_OA_SUBMIT_NO_DEPS, bb); if (IS_ERR(fence)) { @@ -751,71 +723,55 @@ exit: static int xe_oa_configure_oar_context(struct xe_oa_stream *stream, bool enable) { const struct xe_oa_format *format = stream->oa_buffer.format; - struct xe_lrc *lrc = stream->exec_q->lrc[0]; - u32 regs_offset = xe_lrc_regs_offset(lrc) / sizeof(u32); u32 oacontrol = __format_to_oactrl(format, OAR_OACONTROL_COUNTER_SEL_MASK) | (enable ? OAR_OACONTROL_COUNTER_ENABLE : 0); - struct flex regs_context[] = { + struct xe_oa_reg reg_lri[] = { { OACTXCONTROL(stream->hwe->mmio_base), - stream->oa->ctx_oactxctrl_offset[stream->hwe->class] + 1, enable ? OA_COUNTER_RESUME : 0, }, { + OAR_OACONTROL, + oacontrol, + }, + { RING_CONTEXT_CONTROL(stream->hwe->mmio_base), - regs_offset + CTX_CONTEXT_CONTROL, - _MASKED_BIT_ENABLE(CTX_CTRL_OAC_CONTEXT_ENABLE), + _MASKED_FIELD(CTX_CTRL_OAC_CONTEXT_ENABLE, + enable ? CTX_CTRL_OAC_CONTEXT_ENABLE : 0) }, }; - struct xe_oa_reg reg_lri = { OAR_OACONTROL, oacontrol }; - int err; - - /* Modify stream hwe context image with regs_context */ - err = xe_oa_modify_ctx_image(stream, stream->exec_q->lrc[0], - regs_context, ARRAY_SIZE(regs_context)); - if (err) - return err; - /* Apply reg_lri using LRI */ - return xe_oa_load_with_lri(stream, ®_lri); + return xe_oa_load_with_lri(stream, reg_lri, ARRAY_SIZE(reg_lri)); } static int xe_oa_configure_oac_context(struct xe_oa_stream *stream, bool enable) { const struct xe_oa_format *format = stream->oa_buffer.format; - struct xe_lrc *lrc = stream->exec_q->lrc[0]; - u32 regs_offset = xe_lrc_regs_offset(lrc) / sizeof(u32); u32 oacontrol = __format_to_oactrl(format, OAR_OACONTROL_COUNTER_SEL_MASK) | (enable ? OAR_OACONTROL_COUNTER_ENABLE : 0); - struct flex regs_context[] = { + struct xe_oa_reg reg_lri[] = { { OACTXCONTROL(stream->hwe->mmio_base), - stream->oa->ctx_oactxctrl_offset[stream->hwe->class] + 1, enable ? OA_COUNTER_RESUME : 0, }, { + OAC_OACONTROL, + oacontrol + }, + { RING_CONTEXT_CONTROL(stream->hwe->mmio_base), - regs_offset + CTX_CONTEXT_CONTROL, - _MASKED_BIT_ENABLE(CTX_CTRL_OAC_CONTEXT_ENABLE) | + _MASKED_FIELD(CTX_CTRL_OAC_CONTEXT_ENABLE, + enable ? CTX_CTRL_OAC_CONTEXT_ENABLE : 0) | _MASKED_FIELD(CTX_CTRL_RUN_ALONE, enable ? CTX_CTRL_RUN_ALONE : 0), }, }; - struct xe_oa_reg reg_lri = { OAC_OACONTROL, oacontrol }; - int err; /* Set ccs select to enable programming of OAC_OACONTROL */ xe_mmio_write32(&stream->gt->mmio, __oa_regs(stream)->oa_ctrl, __oa_ccs_select(stream)); - /* Modify stream hwe context image with regs_context */ - err = xe_oa_modify_ctx_image(stream, stream->exec_q->lrc[0], - regs_context, ARRAY_SIZE(regs_context)); - if (err) - return err; - - /* Apply reg_lri using LRI */ - return xe_oa_load_with_lri(stream, ®_lri); + return xe_oa_load_with_lri(stream, reg_lri, ARRAY_SIZE(reg_lri)); } static int xe_oa_configure_oa_context(struct xe_oa_stream *stream, bool enable) @@ -2066,8 +2022,8 @@ int xe_oa_stream_open_ioctl(struct drm_device *dev, u64 data, struct drm_file *f if (XE_IOCTL_DBG(oa->xe, !param.exec_q)) return -ENOENT; - if (param.exec_q->width > 1) - drm_dbg(&oa->xe->drm, "exec_q->width > 1, programming only exec_q->lrc[0]\n"); + if (XE_IOCTL_DBG(oa->xe, param.exec_q->width > 1)) + return -EOPNOTSUPP; } /* @@ -2242,6 +2198,7 @@ static const struct xe_mmio_range xe2_oa_mux_regs[] = { { .start = 0x5194, .end = 0x5194 }, /* SYS_MEM_LAT_MEASURE_MERTF_GRP_3D */ { .start = 0x8704, .end = 0x8704 }, /* LMEM_LAT_MEASURE_MCFG_GRP */ { .start = 0xB1BC, .end = 0xB1BC }, /* L3_BANK_LAT_MEASURE_LBCF_GFX */ + { .start = 0xD0E0, .end = 0xD0F4 }, /* VISACTL */ { .start = 0xE18C, .end = 0xE18C }, /* SAMPLER_MODE */ { .start = 0xE590, .end = 0xE590 }, /* TDL_LSC_LAT_MEASURE_TDL_GFX */ { .start = 0x13000, .end = 0x137FC }, /* PES_0_PESL0 - PES_63_UPPER_PESL3 */ diff --git a/drivers/gpu/drm/xe/xe_ring_ops.c b/drivers/gpu/drm/xe/xe_ring_ops.c index 0be4f489d3e1..9f327f27c072 100644 --- a/drivers/gpu/drm/xe/xe_ring_ops.c +++ b/drivers/gpu/drm/xe/xe_ring_ops.c @@ -221,7 +221,10 @@ static int emit_pipe_imm_ggtt(u32 addr, u32 value, bool stall_only, u32 *dw, static u32 get_ppgtt_flag(struct xe_sched_job *job) { - return job->q->vm ? BIT(8) : 0; + if (job->q->vm && !job->ggtt) + return BIT(8); + + return 0; } static int emit_copy_timestamp(struct xe_lrc *lrc, u32 *dw, int i) diff --git a/drivers/gpu/drm/xe/xe_sched_job_types.h b/drivers/gpu/drm/xe/xe_sched_job_types.h index f13f333f00be..d942b20a9f29 100644 --- a/drivers/gpu/drm/xe/xe_sched_job_types.h +++ b/drivers/gpu/drm/xe/xe_sched_job_types.h @@ -56,6 +56,8 @@ struct xe_sched_job { u32 migrate_flush_flags; /** @ring_ops_flush_tlb: The ring ops need to flush TLB before payload. */ bool ring_ops_flush_tlb; + /** @ggtt: mapped in ggtt. */ + bool ggtt; /** @ptrs: per instance pointers. */ struct xe_job_ptrs ptrs[]; }; diff --git a/drivers/hwmon/acpi_power_meter.c b/drivers/hwmon/acpi_power_meter.c index 2f1c9d97ad21..3db1b9d09c32 100644 --- a/drivers/hwmon/acpi_power_meter.c +++ b/drivers/hwmon/acpi_power_meter.c @@ -682,7 +682,7 @@ static int setup_attrs(struct acpi_power_meter_resource *resource) /* _PMD method is optional. */ res = read_domain_devices(resource); - if (res != -ENODEV) + if (res && res != -ENODEV) return res; if (resource->caps.flags & POWER_METER_CAN_MEASURE) { diff --git a/drivers/hwmon/drivetemp.c b/drivers/hwmon/drivetemp.c index 6bdd21aa005a..291d91f68646 100644 --- a/drivers/hwmon/drivetemp.c +++ b/drivers/hwmon/drivetemp.c @@ -165,6 +165,7 @@ static int drivetemp_scsi_command(struct drivetemp_data *st, { u8 scsi_cmd[MAX_COMMAND_SIZE]; enum req_op op; + int err; memset(scsi_cmd, 0, sizeof(scsi_cmd)); scsi_cmd[0] = ATA_16; @@ -192,8 +193,11 @@ static int drivetemp_scsi_command(struct drivetemp_data *st, scsi_cmd[12] = lba_high; scsi_cmd[14] = ata_command; - return scsi_execute_cmd(st->sdev, scsi_cmd, op, st->smartdata, - ATA_SECT_SIZE, HZ, 5, NULL); + err = scsi_execute_cmd(st->sdev, scsi_cmd, op, st->smartdata, + ATA_SECT_SIZE, 10 * HZ, 5, NULL); + if (err > 0) + err = -EIO; + return err; } static int drivetemp_ata_command(struct drivetemp_data *st, u8 feature, diff --git a/drivers/hwmon/ltc2991.c b/drivers/hwmon/ltc2991.c index 7ca139e4b6af..6d5d4cb846da 100644 --- a/drivers/hwmon/ltc2991.c +++ b/drivers/hwmon/ltc2991.c @@ -125,7 +125,7 @@ static int ltc2991_get_curr(struct ltc2991_state *st, u32 reg, int channel, /* Vx-Vy, 19.075uV/LSB */ *val = DIV_ROUND_CLOSEST(sign_extend32(reg_val, 14) * 19075, - st->r_sense_uohm[channel]); + (s32)st->r_sense_uohm[channel]); return 0; } diff --git a/drivers/hwmon/tmp513.c b/drivers/hwmon/tmp513.c index 1c2cb12071b8..5acbfd7d088d 100644 --- a/drivers/hwmon/tmp513.c +++ b/drivers/hwmon/tmp513.c @@ -207,7 +207,8 @@ static int tmp51x_get_value(struct tmp51x_data *data, u8 reg, u8 pos, *val = sign_extend32(regval, reg == TMP51X_SHUNT_CURRENT_RESULT ? 16 - tmp51x_get_pga_shift(data) : 15); - *val = DIV_ROUND_CLOSEST(*val * 10 * MILLI, data->shunt_uohms); + *val = DIV_ROUND_CLOSEST(*val * 10 * (long)MILLI, (long)data->shunt_uohms); + break; case TMP51X_BUS_VOLTAGE_RESULT: case TMP51X_BUS_VOLTAGE_H_LIMIT: @@ -223,7 +224,7 @@ static int tmp51x_get_value(struct tmp51x_data *data, u8 reg, u8 pos, case TMP51X_BUS_CURRENT_RESULT: // Current = (ShuntVoltage * CalibrationRegister) / 4096 *val = sign_extend32(regval, 15) * (long)data->curr_lsb_ua; - *val = DIV_ROUND_CLOSEST(*val, MILLI); + *val = DIV_ROUND_CLOSEST(*val, (long)MILLI); break; case TMP51X_LOCAL_TEMP_RESULT: case TMP51X_REMOTE_TEMP_RESULT_1: @@ -263,7 +264,7 @@ static int tmp51x_set_value(struct tmp51x_data *data, u8 reg, long val) * The user enter current value and we convert it to * voltage. 1lsb = 10uV */ - val = DIV_ROUND_CLOSEST(val * data->shunt_uohms, 10 * MILLI); + val = DIV_ROUND_CLOSEST(val * (long)data->shunt_uohms, 10 * (long)MILLI); max_val = U16_MAX >> tmp51x_get_pga_shift(data); regval = clamp_val(val, -max_val, max_val); break; diff --git a/drivers/i2c/busses/i2c-imx.c b/drivers/i2c/busses/i2c-imx.c index f751d231ded8..5c9a8dfbc4a0 100644 --- a/drivers/i2c/busses/i2c-imx.c +++ b/drivers/i2c/busses/i2c-imx.c @@ -335,6 +335,7 @@ static const struct of_device_id i2c_imx_dt_ids[] = { { .compatible = "fsl,imx6sll-i2c", .data = &imx6_i2c_hwdata, }, { .compatible = "fsl,imx6sx-i2c", .data = &imx6_i2c_hwdata, }, { .compatible = "fsl,imx6ul-i2c", .data = &imx6_i2c_hwdata, }, + { .compatible = "fsl,imx7d-i2c", .data = &imx6_i2c_hwdata, }, { .compatible = "fsl,imx7s-i2c", .data = &imx6_i2c_hwdata, }, { .compatible = "fsl,imx8mm-i2c", .data = &imx6_i2c_hwdata, }, { .compatible = "fsl,imx8mn-i2c", .data = &imx6_i2c_hwdata, }, @@ -532,22 +533,20 @@ static void i2c_imx_dma_free(struct imx_i2c_struct *i2c_imx) static int i2c_imx_bus_busy(struct imx_i2c_struct *i2c_imx, int for_busy, bool atomic) { + bool multi_master = i2c_imx->multi_master; unsigned long orig_jiffies = jiffies; unsigned int temp; - if (!i2c_imx->multi_master) - return 0; - while (1) { temp = imx_i2c_read_reg(i2c_imx, IMX_I2C_I2SR); /* check for arbitration lost */ - if (temp & I2SR_IAL) { + if (multi_master && (temp & I2SR_IAL)) { i2c_imx_clear_irq(i2c_imx, I2SR_IAL); return -EAGAIN; } - if (for_busy && (temp & I2SR_IBB)) { + if (for_busy && (!multi_master || (temp & I2SR_IBB))) { i2c_imx->stopped = 0; break; } diff --git a/drivers/i2c/busses/i2c-microchip-corei2c.c b/drivers/i2c/busses/i2c-microchip-corei2c.c index d1543e7d8380..5db73429125c 100644 --- a/drivers/i2c/busses/i2c-microchip-corei2c.c +++ b/drivers/i2c/busses/i2c-microchip-corei2c.c @@ -93,27 +93,35 @@ * @base: pointer to register struct * @dev: device reference * @i2c_clk: clock reference for i2c input clock + * @msg_queue: pointer to the messages requiring sending * @buf: pointer to msg buffer for easier use * @msg_complete: xfer completion object * @adapter: core i2c abstraction * @msg_err: error code for completed message * @bus_clk_rate: current i2c bus clock rate * @isr_status: cached copy of local ISR status + * @total_num: total number of messages to be sent/received + * @current_num: index of the current message being sent/received * @msg_len: number of bytes transferred in msg * @addr: address of the current slave + * @restart_needed: whether or not a repeated start is required after current message */ struct mchp_corei2c_dev { void __iomem *base; struct device *dev; struct clk *i2c_clk; + struct i2c_msg *msg_queue; u8 *buf; struct completion msg_complete; struct i2c_adapter adapter; int msg_err; + int total_num; + int current_num; u32 bus_clk_rate; u32 isr_status; u16 msg_len; u8 addr; + bool restart_needed; }; static void mchp_corei2c_core_disable(struct mchp_corei2c_dev *idev) @@ -222,6 +230,47 @@ static int mchp_corei2c_fill_tx(struct mchp_corei2c_dev *idev) return 0; } +static void mchp_corei2c_next_msg(struct mchp_corei2c_dev *idev) +{ + struct i2c_msg *this_msg; + u8 ctrl; + + if (idev->current_num >= idev->total_num) { + complete(&idev->msg_complete); + return; + } + + /* + * If there's been an error, the isr needs to return control + * to the "main" part of the driver, so as not to keep sending + * messages once it completes and clears the SI bit. + */ + if (idev->msg_err) { + complete(&idev->msg_complete); + return; + } + + this_msg = idev->msg_queue++; + + if (idev->current_num < (idev->total_num - 1)) { + struct i2c_msg *next_msg = idev->msg_queue; + + idev->restart_needed = next_msg->flags & I2C_M_RD; + } else { + idev->restart_needed = false; + } + + idev->addr = i2c_8bit_addr_from_msg(this_msg); + idev->msg_len = this_msg->len; + idev->buf = this_msg->buf; + + ctrl = readb(idev->base + CORE_I2C_CTRL); + ctrl |= CTRL_STA; + writeb(ctrl, idev->base + CORE_I2C_CTRL); + + idev->current_num++; +} + static irqreturn_t mchp_corei2c_handle_isr(struct mchp_corei2c_dev *idev) { u32 status = idev->isr_status; @@ -238,8 +287,6 @@ static irqreturn_t mchp_corei2c_handle_isr(struct mchp_corei2c_dev *idev) ctrl &= ~CTRL_STA; writeb(idev->addr, idev->base + CORE_I2C_DATA); writeb(ctrl, idev->base + CORE_I2C_CTRL); - if (idev->msg_len == 0) - finished = true; break; case STATUS_M_ARB_LOST: idev->msg_err = -EAGAIN; @@ -247,10 +294,14 @@ static irqreturn_t mchp_corei2c_handle_isr(struct mchp_corei2c_dev *idev) break; case STATUS_M_SLAW_ACK: case STATUS_M_TX_DATA_ACK: - if (idev->msg_len > 0) + if (idev->msg_len > 0) { mchp_corei2c_fill_tx(idev); - else - last_byte = true; + } else { + if (idev->restart_needed) + finished = true; + else + last_byte = true; + } break; case STATUS_M_TX_DATA_NACK: case STATUS_M_SLAR_NACK: @@ -287,7 +338,7 @@ static irqreturn_t mchp_corei2c_handle_isr(struct mchp_corei2c_dev *idev) mchp_corei2c_stop(idev); if (last_byte || finished) - complete(&idev->msg_complete); + mchp_corei2c_next_msg(idev); return IRQ_HANDLED; } @@ -311,21 +362,48 @@ static irqreturn_t mchp_corei2c_isr(int irq, void *_dev) return ret; } -static int mchp_corei2c_xfer_msg(struct mchp_corei2c_dev *idev, - struct i2c_msg *msg) +static int mchp_corei2c_xfer(struct i2c_adapter *adap, struct i2c_msg *msgs, + int num) { - u8 ctrl; + struct mchp_corei2c_dev *idev = i2c_get_adapdata(adap); + struct i2c_msg *this_msg = msgs; unsigned long time_left; + u8 ctrl; + + mchp_corei2c_core_enable(idev); + + /* + * The isr controls the flow of a transfer, this info needs to be saved + * to a location that it can access the queue information from. + */ + idev->restart_needed = false; + idev->msg_queue = msgs; + idev->total_num = num; + idev->current_num = 0; - idev->addr = i2c_8bit_addr_from_msg(msg); - idev->msg_len = msg->len; - idev->buf = msg->buf; + /* + * But the first entry to the isr is triggered by the start in this + * function, so the first message needs to be "dequeued". + */ + idev->addr = i2c_8bit_addr_from_msg(this_msg); + idev->msg_len = this_msg->len; + idev->buf = this_msg->buf; idev->msg_err = 0; - reinit_completion(&idev->msg_complete); + if (idev->total_num > 1) { + struct i2c_msg *next_msg = msgs + 1; - mchp_corei2c_core_enable(idev); + idev->restart_needed = next_msg->flags & I2C_M_RD; + } + idev->current_num++; + idev->msg_queue++; + + reinit_completion(&idev->msg_complete); + + /* + * Send the first start to pass control to the isr + */ ctrl = readb(idev->base + CORE_I2C_CTRL); ctrl |= CTRL_STA; writeb(ctrl, idev->base + CORE_I2C_CTRL); @@ -335,20 +413,8 @@ static int mchp_corei2c_xfer_msg(struct mchp_corei2c_dev *idev, if (!time_left) return -ETIMEDOUT; - return idev->msg_err; -} - -static int mchp_corei2c_xfer(struct i2c_adapter *adap, struct i2c_msg *msgs, - int num) -{ - struct mchp_corei2c_dev *idev = i2c_get_adapdata(adap); - int i, ret; - - for (i = 0; i < num; i++) { - ret = mchp_corei2c_xfer_msg(idev, msgs++); - if (ret) - return ret; - } + if (idev->msg_err) + return idev->msg_err; return num; } diff --git a/drivers/i2c/busses/i2c-rcar.c b/drivers/i2c/busses/i2c-rcar.c index a7b77d14ee86..5693a38da7b5 100644 --- a/drivers/i2c/busses/i2c-rcar.c +++ b/drivers/i2c/busses/i2c-rcar.c @@ -130,6 +130,8 @@ #define ID_P_PM_BLOCKED BIT(31) #define ID_P_MASK GENMASK(31, 27) +#define ID_SLAVE_NACK BIT(0) + enum rcar_i2c_type { I2C_RCAR_GEN1, I2C_RCAR_GEN2, @@ -166,6 +168,7 @@ struct rcar_i2c_priv { int irq; struct i2c_client *host_notify_client; + u8 slave_flags; }; #define rcar_i2c_priv_to_dev(p) ((p)->adap.dev.parent) @@ -655,6 +658,7 @@ static bool rcar_i2c_slave_irq(struct rcar_i2c_priv *priv) { u32 ssr_raw, ssr_filtered; u8 value; + int ret; ssr_raw = rcar_i2c_read(priv, ICSSR) & 0xff; ssr_filtered = ssr_raw & rcar_i2c_read(priv, ICSIER); @@ -670,7 +674,10 @@ static bool rcar_i2c_slave_irq(struct rcar_i2c_priv *priv) rcar_i2c_write(priv, ICRXTX, value); rcar_i2c_write(priv, ICSIER, SDE | SSR | SAR); } else { - i2c_slave_event(priv->slave, I2C_SLAVE_WRITE_REQUESTED, &value); + ret = i2c_slave_event(priv->slave, I2C_SLAVE_WRITE_REQUESTED, &value); + if (ret) + priv->slave_flags |= ID_SLAVE_NACK; + rcar_i2c_read(priv, ICRXTX); /* dummy read */ rcar_i2c_write(priv, ICSIER, SDR | SSR | SAR); } @@ -683,18 +690,21 @@ static bool rcar_i2c_slave_irq(struct rcar_i2c_priv *priv) if (ssr_filtered & SSR) { i2c_slave_event(priv->slave, I2C_SLAVE_STOP, &value); rcar_i2c_write(priv, ICSCR, SIE | SDBS); /* clear our NACK */ + priv->slave_flags &= ~ID_SLAVE_NACK; rcar_i2c_write(priv, ICSIER, SAR); rcar_i2c_write(priv, ICSSR, ~SSR & 0xff); } /* master wants to write to us */ if (ssr_filtered & SDR) { - int ret; - value = rcar_i2c_read(priv, ICRXTX); ret = i2c_slave_event(priv->slave, I2C_SLAVE_WRITE_RECEIVED, &value); - /* Send NACK in case of error */ - rcar_i2c_write(priv, ICSCR, SIE | SDBS | (ret < 0 ? FNA : 0)); + if (ret) + priv->slave_flags |= ID_SLAVE_NACK; + + /* Send NACK in case of error, but it will come 1 byte late :( */ + rcar_i2c_write(priv, ICSCR, SIE | SDBS | + (priv->slave_flags & ID_SLAVE_NACK ? FNA : 0)); rcar_i2c_write(priv, ICSSR, ~SDR & 0xff); } diff --git a/drivers/i2c/i2c-atr.c b/drivers/i2c/i2c-atr.c index b7c10ced5a43..8fe9ddff8e96 100644 --- a/drivers/i2c/i2c-atr.c +++ b/drivers/i2c/i2c-atr.c @@ -412,7 +412,7 @@ static int i2c_atr_bus_notifier_call(struct notifier_block *nb, dev_name(dev), ret); break; - case BUS_NOTIFY_DEL_DEVICE: + case BUS_NOTIFY_REMOVED_DEVICE: i2c_atr_detach_client(client->adapter, client); break; diff --git a/drivers/i2c/i2c-core-base.c b/drivers/i2c/i2c-core-base.c index 7c810893bfa3..75d30861ffe2 100644 --- a/drivers/i2c/i2c-core-base.c +++ b/drivers/i2c/i2c-core-base.c @@ -1562,6 +1562,7 @@ static int i2c_register_adapter(struct i2c_adapter *adap) res = device_add(&adap->dev); if (res) { pr_err("adapter '%s': can't register device (%d)\n", adap->name, res); + put_device(&adap->dev); goto out_list; } diff --git a/drivers/i2c/i2c-slave-testunit.c b/drivers/i2c/i2c-slave-testunit.c index 0d6fbaa48248..6de4307050dd 100644 --- a/drivers/i2c/i2c-slave-testunit.c +++ b/drivers/i2c/i2c-slave-testunit.c @@ -38,6 +38,7 @@ enum testunit_regs { enum testunit_flags { TU_FLAG_IN_PROCESS, + TU_FLAG_NACK, }; struct testunit_data { @@ -90,8 +91,10 @@ static int i2c_slave_testunit_slave_cb(struct i2c_client *client, switch (event) { case I2C_SLAVE_WRITE_REQUESTED: - if (test_bit(TU_FLAG_IN_PROCESS, &tu->flags)) - return -EBUSY; + if (test_bit(TU_FLAG_IN_PROCESS | TU_FLAG_NACK, &tu->flags)) { + ret = -EBUSY; + break; + } memset(tu->regs, 0, TU_NUM_REGS); tu->reg_idx = 0; @@ -99,8 +102,10 @@ static int i2c_slave_testunit_slave_cb(struct i2c_client *client, break; case I2C_SLAVE_WRITE_RECEIVED: - if (test_bit(TU_FLAG_IN_PROCESS, &tu->flags)) - return -EBUSY; + if (test_bit(TU_FLAG_IN_PROCESS | TU_FLAG_NACK, &tu->flags)) { + ret = -EBUSY; + break; + } if (tu->reg_idx < TU_NUM_REGS) tu->regs[tu->reg_idx] = *val; @@ -129,6 +134,8 @@ static int i2c_slave_testunit_slave_cb(struct i2c_client *client, * here because we still need them in the workqueue! */ tu->reg_idx = 0; + + clear_bit(TU_FLAG_NACK, &tu->flags); break; case I2C_SLAVE_READ_PROCESSED: @@ -151,6 +158,10 @@ static int i2c_slave_testunit_slave_cb(struct i2c_client *client, break; } + /* If an error occurred somewhen, we NACK everything until next STOP */ + if (ret) + set_bit(TU_FLAG_NACK, &tu->flags); + return ret; } diff --git a/drivers/i2c/muxes/i2c-demux-pinctrl.c b/drivers/i2c/muxes/i2c-demux-pinctrl.c index dce18f763a09..77a740561fd7 100644 --- a/drivers/i2c/muxes/i2c-demux-pinctrl.c +++ b/drivers/i2c/muxes/i2c-demux-pinctrl.c @@ -68,7 +68,7 @@ static int i2c_demux_activate_master(struct i2c_demux_pinctrl_priv *priv, u32 ne } /* - * Check if there are pinctrl states at all. Note: we cant' use + * Check if there are pinctrl states at all. Note: we can't use * devm_pinctrl_get_select() because we need to distinguish between * the -ENODEV from devm_pinctrl_get() and pinctrl_lookup_state(). */ @@ -261,7 +261,9 @@ static int i2c_demux_pinctrl_probe(struct platform_device *pdev) pm_runtime_no_callbacks(&pdev->dev); /* switch to first parent as active master */ - i2c_demux_activate_master(priv, 0); + err = i2c_demux_activate_master(priv, 0); + if (err) + goto err_rollback; err = device_create_file(&pdev->dev, &dev_attr_available_masters); if (err) diff --git a/drivers/iio/adc/ad4695.c b/drivers/iio/adc/ad4695.c index 595ec4158e73..0146aed9069f 100644 --- a/drivers/iio/adc/ad4695.c +++ b/drivers/iio/adc/ad4695.c @@ -91,6 +91,7 @@ #define AD4695_T_WAKEUP_SW_MS 3 #define AD4695_T_REFBUF_MS 100 #define AD4695_T_REGCONFIG_NS 20 +#define AD4695_T_SCK_CNV_DELAY_NS 80 #define AD4695_REG_ACCESS_SCLK_HZ (10 * MEGA) /* Max number of voltage input channels. */ @@ -132,8 +133,13 @@ struct ad4695_state { unsigned int vref_mv; /* Common mode input pin voltage. */ unsigned int com_mv; - /* 1 per voltage and temperature chan plus 1 xfer to trigger 1st CNV */ - struct spi_transfer buf_read_xfer[AD4695_MAX_CHANNELS + 2]; + /* + * 2 per voltage and temperature chan plus 1 xfer to trigger 1st + * CNV. Excluding the trigger xfer, every 2nd xfer only serves + * to control CS and add a delay between the last SCLK and next + * CNV rising edges. + */ + struct spi_transfer buf_read_xfer[AD4695_MAX_CHANNELS * 2 + 3]; struct spi_message buf_read_msg; /* Raw conversion data received. */ u8 buf[ALIGN((AD4695_MAX_CHANNELS + 2) * AD4695_MAX_CHANNEL_SIZE, @@ -423,7 +429,7 @@ static int ad4695_buffer_preenable(struct iio_dev *indio_dev) u8 temp_chan_bit = st->chip_info->num_voltage_inputs; u32 bit, num_xfer, num_slots; u32 temp_en = 0; - int ret; + int ret, rx_buf_offset = 0; /* * We are using the advanced sequencer since it is the only way to read @@ -449,11 +455,9 @@ static int ad4695_buffer_preenable(struct iio_dev *indio_dev) iio_for_each_active_channel(indio_dev, bit) { xfer = &st->buf_read_xfer[num_xfer]; xfer->bits_per_word = 16; - xfer->rx_buf = &st->buf[(num_xfer - 1) * 2]; + xfer->rx_buf = &st->buf[rx_buf_offset]; xfer->len = 2; - xfer->cs_change = 1; - xfer->cs_change_delay.value = AD4695_T_CONVERT_NS; - xfer->cs_change_delay.unit = SPI_DELAY_UNIT_NSECS; + rx_buf_offset += xfer->len; if (bit == temp_chan_bit) { temp_en = 1; @@ -468,21 +472,44 @@ static int ad4695_buffer_preenable(struct iio_dev *indio_dev) } num_xfer++; + + /* + * We need to add a blank xfer in data reads, to meet the timing + * requirement of a minimum delay between the last SCLK rising + * edge and the CS deassert. + */ + xfer = &st->buf_read_xfer[num_xfer]; + xfer->delay.value = AD4695_T_SCK_CNV_DELAY_NS; + xfer->delay.unit = SPI_DELAY_UNIT_NSECS; + xfer->cs_change = 1; + xfer->cs_change_delay.value = AD4695_T_CONVERT_NS; + xfer->cs_change_delay.unit = SPI_DELAY_UNIT_NSECS; + + num_xfer++; } /* * The advanced sequencer requires that at least 2 slots are enabled. * Since slot 0 is always used for other purposes, we need only 1 - * enabled voltage channel to meet this requirement. If the temperature - * channel is the only enabled channel, we need to add one more slot - * in the sequence but not read from it. + * enabled voltage channel to meet this requirement. If the temperature + * channel is the only enabled channel, we need to add one more slot in + * the sequence but not read from it. This is because the temperature + * sensor is sampled at the end of the channel sequence in advanced + * sequencer mode (see datasheet page 38). + * + * From the iio_for_each_active_channel() block above, we now have an + * xfer with data followed by a blank xfer to allow us to meet the + * timing spec, so move both of those up before adding an extra to + * handle the temperature-only case. */ if (num_slots < 2) { - /* move last xfer so we can insert one more xfer before it */ - st->buf_read_xfer[num_xfer] = *xfer; + /* Move last two xfers */ + st->buf_read_xfer[num_xfer] = st->buf_read_xfer[num_xfer - 1]; + st->buf_read_xfer[num_xfer - 1] = st->buf_read_xfer[num_xfer - 2]; num_xfer++; - /* modify 2nd to last xfer for extra slot */ + /* Modify inserted xfer for extra slot. */ + xfer = &st->buf_read_xfer[num_xfer - 3]; memset(xfer, 0, sizeof(*xfer)); xfer->cs_change = 1; xfer->delay.value = st->chip_info->t_acq_ns; @@ -499,6 +526,12 @@ static int ad4695_buffer_preenable(struct iio_dev *indio_dev) return ret; num_slots++; + + /* + * We still want to point at the last xfer when finished, so + * update the pointer. + */ + xfer = &st->buf_read_xfer[num_xfer - 1]; } /* @@ -583,8 +616,20 @@ out: */ static int ad4695_read_one_sample(struct ad4695_state *st, unsigned int address) { - struct spi_transfer xfer[2] = { }; - int ret, i = 0; + struct spi_transfer xfers[2] = { + { + .speed_hz = AD4695_REG_ACCESS_SCLK_HZ, + .bits_per_word = 16, + .tx_buf = &st->cnv_cmd, + .len = 2, + }, + { + /* Required delay between last SCLK and CNV/CS */ + .delay.value = AD4695_T_SCK_CNV_DELAY_NS, + .delay.unit = SPI_DELAY_UNIT_NSECS, + } + }; + int ret; ret = ad4695_set_single_cycle_mode(st, address); if (ret) @@ -592,29 +637,22 @@ static int ad4695_read_one_sample(struct ad4695_state *st, unsigned int address) /* * Setting the first channel to the temperature channel isn't supported - * in single-cycle mode, so we have to do an extra xfer to read the - * temperature. + * in single-cycle mode, so we have to do an extra conversion to read + * the temperature. */ if (address == AD4695_CMD_TEMP_CHAN) { - /* We aren't reading, so we can make this a short xfer. */ - st->cnv_cmd2 = AD4695_CMD_TEMP_CHAN << 3; - xfer[0].tx_buf = &st->cnv_cmd2; - xfer[0].len = 1; - xfer[0].cs_change = 1; - xfer[0].cs_change_delay.value = AD4695_T_CONVERT_NS; - xfer[0].cs_change_delay.unit = SPI_DELAY_UNIT_NSECS; - - i = 1; + st->cnv_cmd = AD4695_CMD_TEMP_CHAN << 11; + + ret = spi_sync_transfer(st->spi, xfers, ARRAY_SIZE(xfers)); + if (ret) + return ret; } /* Then read the result and exit conversion mode. */ st->cnv_cmd = AD4695_CMD_EXIT_CNV_MODE << 11; - xfer[i].bits_per_word = 16; - xfer[i].tx_buf = &st->cnv_cmd; - xfer[i].rx_buf = &st->raw_data; - xfer[i].len = 2; + xfers[0].rx_buf = &st->raw_data; - return spi_sync_transfer(st->spi, xfer, i + 1); + return spi_sync_transfer(st->spi, xfers, ARRAY_SIZE(xfers)); } static int ad4695_read_raw(struct iio_dev *indio_dev, diff --git a/drivers/iio/adc/ad7124.c b/drivers/iio/adc/ad7124.c index 7314fb32bdec..3d678c420cbf 100644 --- a/drivers/iio/adc/ad7124.c +++ b/drivers/iio/adc/ad7124.c @@ -917,6 +917,9 @@ static int ad7124_setup(struct ad7124_state *st) * set all channels to this default value. */ ad7124_set_channel_odr(st, i, 10); + + /* Disable all channels to prevent unintended conversions. */ + ad_sd_write_reg(&st->sd, AD7124_CHANNEL(i), 2, 0); } ret = ad_sd_write_reg(&st->sd, AD7124_ADC_CONTROL, 2, st->adc_control); diff --git a/drivers/iio/adc/ad7173.c b/drivers/iio/adc/ad7173.c index 8a0c931ca83a..8b03c1e5567e 100644 --- a/drivers/iio/adc/ad7173.c +++ b/drivers/iio/adc/ad7173.c @@ -200,6 +200,7 @@ struct ad7173_channel { struct ad7173_state { struct ad_sigma_delta sd; + struct ad_sigma_delta_info sigma_delta_info; const struct ad7173_device_info *info; struct ad7173_channel *channels; struct regulator_bulk_data regulators[3]; @@ -753,7 +754,7 @@ static int ad7173_disable_one(struct ad_sigma_delta *sd, unsigned int chan) return ad_sd_write_reg(sd, AD7173_REG_CH(chan), 2, 0); } -static struct ad_sigma_delta_info ad7173_sigma_delta_info = { +static const struct ad_sigma_delta_info ad7173_sigma_delta_info = { .set_channel = ad7173_set_channel, .append_status = ad7173_append_status, .disable_all = ad7173_disable_all, @@ -1403,7 +1404,7 @@ static int ad7173_fw_parse_device_config(struct iio_dev *indio_dev) if (ret < 0) return dev_err_probe(dev, ret, "Interrupt 'rdy' is required\n"); - ad7173_sigma_delta_info.irq_line = ret; + st->sigma_delta_info.irq_line = ret; return ad7173_fw_parse_channel_config(indio_dev); } @@ -1436,8 +1437,9 @@ static int ad7173_probe(struct spi_device *spi) spi->mode = SPI_MODE_3; spi_setup(spi); - ad7173_sigma_delta_info.num_slots = st->info->num_configs; - ret = ad_sd_init(&st->sd, indio_dev, spi, &ad7173_sigma_delta_info); + st->sigma_delta_info = ad7173_sigma_delta_info; + st->sigma_delta_info.num_slots = st->info->num_configs; + ret = ad_sd_init(&st->sd, indio_dev, spi, &st->sigma_delta_info); if (ret) return ret; diff --git a/drivers/iio/adc/ad9467.c b/drivers/iio/adc/ad9467.c index d358958ab310..f30119b42ba0 100644 --- a/drivers/iio/adc/ad9467.c +++ b/drivers/iio/adc/ad9467.c @@ -895,7 +895,7 @@ static int ad9467_update_scan_mode(struct iio_dev *indio_dev, return 0; } -static struct iio_info ad9467_info = { +static const struct iio_info ad9467_info = { .read_raw = ad9467_read_raw, .write_raw = ad9467_write_raw, .update_scan_mode = ad9467_update_scan_mode, @@ -903,6 +903,14 @@ static struct iio_info ad9467_info = { .read_avail = ad9467_read_avail, }; +/* Same as above, but without .read_avail */ +static const struct iio_info ad9467_info_no_read_avail = { + .read_raw = ad9467_read_raw, + .write_raw = ad9467_write_raw, + .update_scan_mode = ad9467_update_scan_mode, + .debugfs_reg_access = ad9467_reg_access, +}; + static int ad9467_scale_fill(struct ad9467_state *st) { const struct ad9467_chip_info *info = st->info; @@ -1214,11 +1222,12 @@ static int ad9467_probe(struct spi_device *spi) } if (st->info->num_scales > 1) - ad9467_info.read_avail = ad9467_read_avail; + indio_dev->info = &ad9467_info; + else + indio_dev->info = &ad9467_info_no_read_avail; indio_dev->name = st->info->name; indio_dev->channels = st->info->channels; indio_dev->num_channels = st->info->num_channels; - indio_dev->info = &ad9467_info; ret = ad9467_iio_backend_get(st); if (ret) diff --git a/drivers/iio/adc/at91_adc.c b/drivers/iio/adc/at91_adc.c index a3f0a2321666..5927756b749a 100644 --- a/drivers/iio/adc/at91_adc.c +++ b/drivers/iio/adc/at91_adc.c @@ -979,7 +979,7 @@ static int at91_ts_register(struct iio_dev *idev, return ret; err: - input_free_device(st->ts_input); + input_free_device(input); return ret; } diff --git a/drivers/iio/adc/rockchip_saradc.c b/drivers/iio/adc/rockchip_saradc.c index 240cfa391674..dfd47a6e1f4a 100644 --- a/drivers/iio/adc/rockchip_saradc.c +++ b/drivers/iio/adc/rockchip_saradc.c @@ -368,6 +368,8 @@ static irqreturn_t rockchip_saradc_trigger_handler(int irq, void *p) int ret; int i, j = 0; + memset(&data, 0, sizeof(data)); + mutex_lock(&info->lock); iio_for_each_active_channel(i_dev, i) { diff --git a/drivers/iio/adc/stm32-dfsdm-adc.c b/drivers/iio/adc/stm32-dfsdm-adc.c index 1f9eca2fb2bf..fe11b0d8eab3 100644 --- a/drivers/iio/adc/stm32-dfsdm-adc.c +++ b/drivers/iio/adc/stm32-dfsdm-adc.c @@ -691,11 +691,14 @@ static int stm32_dfsdm_generic_channel_parse_of(struct stm32_dfsdm *dfsdm, return -EINVAL; } - ret = fwnode_property_read_string(node, "label", &ch->datasheet_name); - if (ret < 0) { - dev_err(&indio_dev->dev, - " Error parsing 'label' for idx %d\n", ch->channel); - return ret; + if (fwnode_property_present(node, "label")) { + /* label is optional */ + ret = fwnode_property_read_string(node, "label", &ch->datasheet_name); + if (ret < 0) { + dev_err(&indio_dev->dev, + " Error parsing 'label' for idx %d\n", ch->channel); + return ret; + } } df_ch = &dfsdm->ch_list[ch->channel]; diff --git a/drivers/iio/adc/ti-ads1119.c b/drivers/iio/adc/ti-ads1119.c index e9d9d4d46d38..c268e27eec12 100644 --- a/drivers/iio/adc/ti-ads1119.c +++ b/drivers/iio/adc/ti-ads1119.c @@ -500,12 +500,14 @@ static irqreturn_t ads1119_trigger_handler(int irq, void *private) struct iio_dev *indio_dev = pf->indio_dev; struct ads1119_state *st = iio_priv(indio_dev); struct { - unsigned int sample; + s16 sample; s64 timestamp __aligned(8); } scan; unsigned int index; int ret; + memset(&scan, 0, sizeof(scan)); + if (!iio_trigger_using_own(indio_dev)) { index = find_first_bit(indio_dev->active_scan_mask, iio_get_masklength(indio_dev)); diff --git a/drivers/iio/adc/ti-ads124s08.c b/drivers/iio/adc/ti-ads124s08.c index 425b48d8986f..f452f57f11c9 100644 --- a/drivers/iio/adc/ti-ads124s08.c +++ b/drivers/iio/adc/ti-ads124s08.c @@ -183,9 +183,9 @@ static int ads124s_reset(struct iio_dev *indio_dev) struct ads124s_private *priv = iio_priv(indio_dev); if (priv->reset_gpio) { - gpiod_set_value(priv->reset_gpio, 0); + gpiod_set_value_cansleep(priv->reset_gpio, 0); udelay(200); - gpiod_set_value(priv->reset_gpio, 1); + gpiod_set_value_cansleep(priv->reset_gpio, 1); } else { return ads124s_write_cmd(indio_dev, ADS124S08_CMD_RESET); } diff --git a/drivers/iio/adc/ti-ads1298.c b/drivers/iio/adc/ti-ads1298.c index 36d43495f603..03f762415fa5 100644 --- a/drivers/iio/adc/ti-ads1298.c +++ b/drivers/iio/adc/ti-ads1298.c @@ -613,6 +613,8 @@ static int ads1298_init(struct iio_dev *indio_dev) } indio_dev->name = devm_kasprintf(dev, GFP_KERNEL, "ads129%u%s", indio_dev->num_channels, suffix); + if (!indio_dev->name) + return -ENOMEM; /* Enable internal test signal, double amplitude, double frequency */ ret = regmap_write(priv->regmap, ADS1298_REG_CONFIG2, diff --git a/drivers/iio/adc/ti-ads8688.c b/drivers/iio/adc/ti-ads8688.c index 9b1814f1965a..a31658b760a4 100644 --- a/drivers/iio/adc/ti-ads8688.c +++ b/drivers/iio/adc/ti-ads8688.c @@ -381,7 +381,7 @@ static irqreturn_t ads8688_trigger_handler(int irq, void *p) struct iio_poll_func *pf = p; struct iio_dev *indio_dev = pf->indio_dev; /* Ensure naturally aligned timestamp */ - u16 buffer[ADS8688_MAX_CHANNELS + sizeof(s64)/sizeof(u16)] __aligned(8); + u16 buffer[ADS8688_MAX_CHANNELS + sizeof(s64)/sizeof(u16)] __aligned(8) = { }; int i, j = 0; iio_for_each_active_channel(indio_dev, i) { diff --git a/drivers/iio/dummy/iio_simple_dummy_buffer.c b/drivers/iio/dummy/iio_simple_dummy_buffer.c index 4ca3f1aaff99..288880346707 100644 --- a/drivers/iio/dummy/iio_simple_dummy_buffer.c +++ b/drivers/iio/dummy/iio_simple_dummy_buffer.c @@ -48,7 +48,7 @@ static irqreturn_t iio_simple_dummy_trigger_h(int irq, void *p) int i = 0, j; u16 *data; - data = kmalloc(indio_dev->scan_bytes, GFP_KERNEL); + data = kzalloc(indio_dev->scan_bytes, GFP_KERNEL); if (!data) goto done; diff --git a/drivers/iio/gyro/fxas21002c_core.c b/drivers/iio/gyro/fxas21002c_core.c index 0391c78c2f18..754c8a564ba4 100644 --- a/drivers/iio/gyro/fxas21002c_core.c +++ b/drivers/iio/gyro/fxas21002c_core.c @@ -730,14 +730,21 @@ static irqreturn_t fxas21002c_trigger_handler(int irq, void *p) int ret; mutex_lock(&data->lock); + ret = fxas21002c_pm_get(data); + if (ret < 0) + goto out_unlock; + ret = regmap_bulk_read(data->regmap, FXAS21002C_REG_OUT_X_MSB, data->buffer, CHANNEL_SCAN_MAX * sizeof(s16)); if (ret < 0) - goto out_unlock; + goto out_pm_put; iio_push_to_buffers_with_timestamp(indio_dev, data->buffer, data->timestamp); +out_pm_put: + fxas21002c_pm_put(data); + out_unlock: mutex_unlock(&data->lock); diff --git a/drivers/iio/imu/inv_icm42600/inv_icm42600.h b/drivers/iio/imu/inv_icm42600/inv_icm42600.h index 3a07e43e4cf1..18787a43477b 100644 --- a/drivers/iio/imu/inv_icm42600/inv_icm42600.h +++ b/drivers/iio/imu/inv_icm42600/inv_icm42600.h @@ -403,6 +403,7 @@ struct inv_icm42600_sensor_state { typedef int (*inv_icm42600_bus_setup)(struct inv_icm42600_state *); extern const struct regmap_config inv_icm42600_regmap_config; +extern const struct regmap_config inv_icm42600_spi_regmap_config; extern const struct dev_pm_ops inv_icm42600_pm_ops; const struct iio_mount_matrix * diff --git a/drivers/iio/imu/inv_icm42600/inv_icm42600_core.c b/drivers/iio/imu/inv_icm42600/inv_icm42600_core.c index 561d245c1d64..ef9875d3b79d 100644 --- a/drivers/iio/imu/inv_icm42600/inv_icm42600_core.c +++ b/drivers/iio/imu/inv_icm42600/inv_icm42600_core.c @@ -87,6 +87,21 @@ const struct regmap_config inv_icm42600_regmap_config = { }; EXPORT_SYMBOL_NS_GPL(inv_icm42600_regmap_config, "IIO_ICM42600"); +/* define specific regmap for SPI not supporting burst write */ +const struct regmap_config inv_icm42600_spi_regmap_config = { + .name = "inv_icm42600", + .reg_bits = 8, + .val_bits = 8, + .max_register = 0x4FFF, + .ranges = inv_icm42600_regmap_ranges, + .num_ranges = ARRAY_SIZE(inv_icm42600_regmap_ranges), + .volatile_table = inv_icm42600_regmap_volatile_accesses, + .rd_noinc_table = inv_icm42600_regmap_rd_noinc_accesses, + .cache_type = REGCACHE_RBTREE, + .use_single_write = true, +}; +EXPORT_SYMBOL_NS_GPL(inv_icm42600_spi_regmap_config, "IIO_ICM42600"); + struct inv_icm42600_hw { uint8_t whoami; const char *name; @@ -814,6 +829,8 @@ out_unlock: static int inv_icm42600_resume(struct device *dev) { struct inv_icm42600_state *st = dev_get_drvdata(dev); + struct inv_icm42600_sensor_state *gyro_st = iio_priv(st->indio_gyro); + struct inv_icm42600_sensor_state *accel_st = iio_priv(st->indio_accel); int ret; mutex_lock(&st->lock); @@ -834,9 +851,12 @@ static int inv_icm42600_resume(struct device *dev) goto out_unlock; /* restore FIFO data streaming */ - if (st->fifo.on) + if (st->fifo.on) { + inv_sensors_timestamp_reset(&gyro_st->ts); + inv_sensors_timestamp_reset(&accel_st->ts); ret = regmap_write(st->map, INV_ICM42600_REG_FIFO_CONFIG, INV_ICM42600_FIFO_CONFIG_STREAM); + } out_unlock: mutex_unlock(&st->lock); diff --git a/drivers/iio/imu/inv_icm42600/inv_icm42600_spi.c b/drivers/iio/imu/inv_icm42600/inv_icm42600_spi.c index c55d8e672183..2bd2c4c8e50c 100644 --- a/drivers/iio/imu/inv_icm42600/inv_icm42600_spi.c +++ b/drivers/iio/imu/inv_icm42600/inv_icm42600_spi.c @@ -59,7 +59,8 @@ static int inv_icm42600_probe(struct spi_device *spi) return -EINVAL; chip = (uintptr_t)match; - regmap = devm_regmap_init_spi(spi, &inv_icm42600_regmap_config); + /* use SPI specific regmap */ + regmap = devm_regmap_init_spi(spi, &inv_icm42600_spi_regmap_config); if (IS_ERR(regmap)) return PTR_ERR(regmap); diff --git a/drivers/iio/imu/kmx61.c b/drivers/iio/imu/kmx61.c index 324c38764656..e19c5d3137c6 100644 --- a/drivers/iio/imu/kmx61.c +++ b/drivers/iio/imu/kmx61.c @@ -1193,7 +1193,7 @@ static irqreturn_t kmx61_trigger_handler(int irq, void *p) struct kmx61_data *data = kmx61_get_data(indio_dev); int bit, ret, i = 0; u8 base; - s16 buffer[8]; + s16 buffer[8] = { }; if (indio_dev == data->acc_indio_dev) base = KMX61_ACC_XOUT_L; diff --git a/drivers/iio/inkern.c b/drivers/iio/inkern.c index 136b225b6bc8..9050a59129e6 100644 --- a/drivers/iio/inkern.c +++ b/drivers/iio/inkern.c @@ -500,7 +500,7 @@ struct iio_channel *iio_channel_get_all(struct device *dev) return_ptr(chans); error_free_chans: - for (i = 0; i < nummaps; i++) + for (i = 0; i < mapind; i++) iio_device_put(chans[i].indio_dev); return ERR_PTR(ret); } diff --git a/drivers/iio/light/bh1745.c b/drivers/iio/light/bh1745.c index 56e32689bb97..63bf729df517 100644 --- a/drivers/iio/light/bh1745.c +++ b/drivers/iio/light/bh1745.c @@ -746,6 +746,8 @@ static irqreturn_t bh1745_trigger_handler(int interrupt, void *p) int i; int j = 0; + memset(&scan, 0, sizeof(scan)); + iio_for_each_active_channel(indio_dev, i) { ret = regmap_bulk_read(data->regmap, BH1745_RED_LSB + 2 * i, &value, 2); diff --git a/drivers/iio/light/vcnl4035.c b/drivers/iio/light/vcnl4035.c index 337a1332c2c6..67c94be02018 100644 --- a/drivers/iio/light/vcnl4035.c +++ b/drivers/iio/light/vcnl4035.c @@ -105,7 +105,7 @@ static irqreturn_t vcnl4035_trigger_consumer_handler(int irq, void *p) struct iio_dev *indio_dev = pf->indio_dev; struct vcnl4035_data *data = iio_priv(indio_dev); /* Ensure naturally aligned timestamp */ - u8 buffer[ALIGN(sizeof(u16), sizeof(s64)) + sizeof(s64)] __aligned(8); + u8 buffer[ALIGN(sizeof(u16), sizeof(s64)) + sizeof(s64)] __aligned(8) = { }; int ret; ret = regmap_read(data->regmap, VCNL4035_ALS_DATA, (int *)buffer); diff --git a/drivers/iio/pressure/zpa2326.c b/drivers/iio/pressure/zpa2326.c index 2adea84f5b4d..9db1c94dfc18 100644 --- a/drivers/iio/pressure/zpa2326.c +++ b/drivers/iio/pressure/zpa2326.c @@ -586,6 +586,8 @@ static int zpa2326_fill_sample_buffer(struct iio_dev *indio_dev, } sample; int err; + memset(&sample, 0, sizeof(sample)); + if (test_bit(0, indio_dev->active_scan_mask)) { /* Get current pressure from hardware FIFO. */ err = zpa2326_dequeue_pressure(indio_dev, &sample.pressure); diff --git a/drivers/iio/temperature/tmp006.c b/drivers/iio/temperature/tmp006.c index 0c844137d7aa..02b27f471baa 100644 --- a/drivers/iio/temperature/tmp006.c +++ b/drivers/iio/temperature/tmp006.c @@ -252,6 +252,8 @@ static irqreturn_t tmp006_trigger_handler(int irq, void *p) } scan; s32 ret; + memset(&scan, 0, sizeof(scan)); + ret = i2c_smbus_read_word_data(data->client, TMP006_VOBJECT); if (ret < 0) goto err; diff --git a/drivers/iio/test/Kconfig b/drivers/iio/test/Kconfig index 33cca49c8058..7a181cac3cc9 100644 --- a/drivers/iio/test/Kconfig +++ b/drivers/iio/test/Kconfig @@ -5,7 +5,7 @@ # Keep in alphabetical order config IIO_GTS_KUNIT_TEST - tristate "Test IIO formatting functions" if !KUNIT_ALL_TESTS + tristate "Test IIO gain-time-scale helpers" if !KUNIT_ALL_TESTS depends on KUNIT select IIO_GTS_HELPER select TEST_KUNIT_DEVICE_HELPERS diff --git a/drivers/iio/test/iio-test-rescale.c b/drivers/iio/test/iio-test-rescale.c index cbf13337ed1f..bbc6a2e1c2c1 100644 --- a/drivers/iio/test/iio-test-rescale.c +++ b/drivers/iio/test/iio-test-rescale.c @@ -652,6 +652,8 @@ static void iio_rescale_test_scale(struct kunit *test) int rel_ppm; int ret; + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, buff); + rescale.numerator = t->numerator; rescale.denominator = t->denominator; rescale.offset = t->offset; @@ -681,6 +683,8 @@ static void iio_rescale_test_offset(struct kunit *test) int values[2]; int ret; + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, buff_off); + rescale.numerator = t->numerator; rescale.denominator = t->denominator; rescale.offset = t->offset; diff --git a/drivers/infiniband/core/cma.c b/drivers/infiniband/core/cma.c index 64ace0b968f0..91db10515d74 100644 --- a/drivers/infiniband/core/cma.c +++ b/drivers/infiniband/core/cma.c @@ -690,6 +690,7 @@ cma_validate_port(struct ib_device *device, u32 port, int bound_if_index = dev_addr->bound_dev_if; int dev_type = dev_addr->dev_type; struct net_device *ndev = NULL; + struct net_device *pdev = NULL; if (!rdma_dev_access_netns(device, id_priv->id.route.addr.dev_addr.net)) goto out; @@ -714,6 +715,21 @@ cma_validate_port(struct ib_device *device, u32 port, rcu_read_lock(); ndev = rcu_dereference(sgid_attr->ndev); + if (ndev->ifindex != bound_if_index) { + pdev = dev_get_by_index_rcu(dev_addr->net, bound_if_index); + if (pdev) { + if (is_vlan_dev(pdev)) { + pdev = vlan_dev_real_dev(pdev); + if (ndev->ifindex == pdev->ifindex) + bound_if_index = pdev->ifindex; + } + if (is_vlan_dev(ndev)) { + pdev = vlan_dev_real_dev(ndev); + if (bound_if_index == pdev->ifindex) + bound_if_index = ndev->ifindex; + } + } + } if (!net_eq(dev_net(ndev), dev_addr->net) || ndev->ifindex != bound_if_index) { rdma_put_gid_attr(sgid_attr); diff --git a/drivers/infiniband/core/nldev.c b/drivers/infiniband/core/nldev.c index ff121e59b9c0..cb987ab0177c 100644 --- a/drivers/infiniband/core/nldev.c +++ b/drivers/infiniband/core/nldev.c @@ -2833,8 +2833,8 @@ int rdma_nl_notify_event(struct ib_device *device, u32 port_num, enum rdma_nl_notify_event_type type) { struct sk_buff *skb; + int ret = -EMSGSIZE; struct net *net; - int ret = 0; void *nlh; net = read_pnet(&device->coredev.rdma_net); diff --git a/drivers/infiniband/core/uverbs_cmd.c b/drivers/infiniband/core/uverbs_cmd.c index 66b02fbf077a..5ad14c39d48c 100644 --- a/drivers/infiniband/core/uverbs_cmd.c +++ b/drivers/infiniband/core/uverbs_cmd.c @@ -161,7 +161,7 @@ static const void __user *uverbs_request_next_ptr(struct uverbs_req_iter *iter, { const void __user *res = iter->cur; - if (iter->cur + len > iter->end) + if (len > iter->end - iter->cur) return (void __force __user *)ERR_PTR(-ENOSPC); iter->cur += len; return res; @@ -2008,11 +2008,13 @@ static int ib_uverbs_post_send(struct uverbs_attr_bundle *attrs) ret = uverbs_request_start(attrs, &iter, &cmd, sizeof(cmd)); if (ret) return ret; - wqes = uverbs_request_next_ptr(&iter, cmd.wqe_size * cmd.wr_count); + wqes = uverbs_request_next_ptr(&iter, size_mul(cmd.wqe_size, + cmd.wr_count)); if (IS_ERR(wqes)) return PTR_ERR(wqes); - sgls = uverbs_request_next_ptr( - &iter, cmd.sge_count * sizeof(struct ib_uverbs_sge)); + sgls = uverbs_request_next_ptr(&iter, + size_mul(cmd.sge_count, + sizeof(struct ib_uverbs_sge))); if (IS_ERR(sgls)) return PTR_ERR(sgls); ret = uverbs_request_finish(&iter); @@ -2198,11 +2200,11 @@ ib_uverbs_unmarshall_recv(struct uverbs_req_iter *iter, u32 wr_count, if (wqe_size < sizeof(struct ib_uverbs_recv_wr)) return ERR_PTR(-EINVAL); - wqes = uverbs_request_next_ptr(iter, wqe_size * wr_count); + wqes = uverbs_request_next_ptr(iter, size_mul(wqe_size, wr_count)); if (IS_ERR(wqes)) return ERR_CAST(wqes); - sgls = uverbs_request_next_ptr( - iter, sge_count * sizeof(struct ib_uverbs_sge)); + sgls = uverbs_request_next_ptr(iter, size_mul(sge_count, + sizeof(struct ib_uverbs_sge))); if (IS_ERR(sgls)) return ERR_CAST(sgls); ret = uverbs_request_finish(iter); diff --git a/drivers/infiniband/hw/bnxt_re/ib_verbs.c b/drivers/infiniband/hw/bnxt_re/ib_verbs.c index 82023394e330..e3d26bd6de05 100644 --- a/drivers/infiniband/hw/bnxt_re/ib_verbs.c +++ b/drivers/infiniband/hw/bnxt_re/ib_verbs.c @@ -199,7 +199,7 @@ int bnxt_re_query_device(struct ib_device *ibdev, ib_attr->vendor_id = rdev->en_dev->pdev->vendor; ib_attr->vendor_part_id = rdev->en_dev->pdev->device; - ib_attr->hw_ver = rdev->en_dev->pdev->subsystem_device; + ib_attr->hw_ver = rdev->en_dev->pdev->revision; ib_attr->max_qp = dev_attr->max_qp; ib_attr->max_qp_wr = dev_attr->max_qp_wqes; ib_attr->device_cap_flags = @@ -967,13 +967,13 @@ int bnxt_re_destroy_qp(struct ib_qp *ib_qp, struct ib_udata *udata) unsigned int flags; int rc; + bnxt_re_debug_rem_qpinfo(rdev, qp); + bnxt_qplib_flush_cqn_wq(&qp->qplib_qp); rc = bnxt_qplib_destroy_qp(&rdev->qplib_res, &qp->qplib_qp); - if (rc) { + if (rc) ibdev_err(&rdev->ibdev, "Failed to destroy HW QP"); - return rc; - } if (rdma_is_kernel_res(&qp->ib_qp.res)) { flags = bnxt_re_lock_cqs(qp); @@ -983,11 +983,8 @@ int bnxt_re_destroy_qp(struct ib_qp *ib_qp, struct ib_udata *udata) bnxt_qplib_free_qp_res(&rdev->qplib_res, &qp->qplib_qp); - if (ib_qp->qp_type == IB_QPT_GSI && rdev->gsi_ctx.gsi_sqp) { - rc = bnxt_re_destroy_gsi_sqp(qp); - if (rc) - return rc; - } + if (ib_qp->qp_type == IB_QPT_GSI && rdev->gsi_ctx.gsi_sqp) + bnxt_re_destroy_gsi_sqp(qp); mutex_lock(&rdev->qp_lock); list_del(&qp->list); @@ -998,8 +995,6 @@ int bnxt_re_destroy_qp(struct ib_qp *ib_qp, struct ib_udata *udata) else if (qp->qplib_qp.type == CMDQ_CREATE_QP_TYPE_UD) atomic_dec(&rdev->stats.res.ud_qp_count); - bnxt_re_debug_rem_qpinfo(rdev, qp); - ib_umem_release(qp->rumem); ib_umem_release(qp->sumem); @@ -2167,18 +2162,20 @@ int bnxt_re_modify_qp(struct ib_qp *ib_qp, struct ib_qp_attr *qp_attr, } } - if (qp_attr_mask & IB_QP_PATH_MTU) { - qp->qplib_qp.modify_flags |= - CMDQ_MODIFY_QP_MODIFY_MASK_PATH_MTU; - qp->qplib_qp.path_mtu = __from_ib_mtu(qp_attr->path_mtu); - qp->qplib_qp.mtu = ib_mtu_enum_to_int(qp_attr->path_mtu); - } else if (qp_attr->qp_state == IB_QPS_RTR) { - qp->qplib_qp.modify_flags |= - CMDQ_MODIFY_QP_MODIFY_MASK_PATH_MTU; - qp->qplib_qp.path_mtu = - __from_ib_mtu(iboe_get_mtu(rdev->netdev->mtu)); - qp->qplib_qp.mtu = - ib_mtu_enum_to_int(iboe_get_mtu(rdev->netdev->mtu)); + if (qp_attr->qp_state == IB_QPS_RTR) { + enum ib_mtu qpmtu; + + qpmtu = iboe_get_mtu(rdev->netdev->mtu); + if (qp_attr_mask & IB_QP_PATH_MTU) { + if (ib_mtu_enum_to_int(qp_attr->path_mtu) > + ib_mtu_enum_to_int(qpmtu)) + return -EINVAL; + qpmtu = qp_attr->path_mtu; + } + + qp->qplib_qp.modify_flags |= CMDQ_MODIFY_QP_MODIFY_MASK_PATH_MTU; + qp->qplib_qp.path_mtu = __from_ib_mtu(qpmtu); + qp->qplib_qp.mtu = ib_mtu_enum_to_int(qpmtu); } if (qp_attr_mask & IB_QP_TIMEOUT) { @@ -2328,6 +2325,7 @@ int bnxt_re_query_qp(struct ib_qp *ib_qp, struct ib_qp_attr *qp_attr, qp_attr->retry_cnt = qplib_qp->retry_cnt; qp_attr->rnr_retry = qplib_qp->rnr_retry; qp_attr->min_rnr_timer = qplib_qp->min_rnr_timer; + qp_attr->port_num = __to_ib_port_num(qplib_qp->port_id); qp_attr->rq_psn = qplib_qp->rq.psn; qp_attr->max_rd_atomic = qplib_qp->max_rd_atomic; qp_attr->sq_psn = qplib_qp->sq.psn; @@ -2824,7 +2822,8 @@ bad: wr = wr->next; } bnxt_qplib_post_send_db(&qp->qplib_qp); - bnxt_ud_qp_hw_stall_workaround(qp); + if (!bnxt_qplib_is_chip_gen_p5_p7(qp->rdev->chip_ctx)) + bnxt_ud_qp_hw_stall_workaround(qp); spin_unlock_irqrestore(&qp->sq_lock, flags); return rc; } @@ -2936,7 +2935,8 @@ bad: wr = wr->next; } bnxt_qplib_post_send_db(&qp->qplib_qp); - bnxt_ud_qp_hw_stall_workaround(qp); + if (!bnxt_qplib_is_chip_gen_p5_p7(qp->rdev->chip_ctx)) + bnxt_ud_qp_hw_stall_workaround(qp); spin_unlock_irqrestore(&qp->sq_lock, flags); return rc; diff --git a/drivers/infiniband/hw/bnxt_re/ib_verbs.h b/drivers/infiniband/hw/bnxt_re/ib_verbs.h index ac59f1d73b15..fbb16a411d6a 100644 --- a/drivers/infiniband/hw/bnxt_re/ib_verbs.h +++ b/drivers/infiniband/hw/bnxt_re/ib_verbs.h @@ -268,6 +268,10 @@ void bnxt_re_dealloc_ucontext(struct ib_ucontext *context); int bnxt_re_mmap(struct ib_ucontext *context, struct vm_area_struct *vma); void bnxt_re_mmap_free(struct rdma_user_mmap_entry *rdma_entry); +static inline u32 __to_ib_port_num(u16 port_id) +{ + return (u32)port_id + 1; +} unsigned long bnxt_re_lock_cqs(struct bnxt_re_qp *qp); void bnxt_re_unlock_cqs(struct bnxt_re_qp *qp, unsigned long flags); diff --git a/drivers/infiniband/hw/bnxt_re/main.c b/drivers/infiniband/hw/bnxt_re/main.c index b7af0d5ff3b6..c143f273b759 100644 --- a/drivers/infiniband/hw/bnxt_re/main.c +++ b/drivers/infiniband/hw/bnxt_re/main.c @@ -1715,11 +1715,8 @@ static bool bnxt_re_is_qp1_or_shadow_qp(struct bnxt_re_dev *rdev, static void bnxt_re_dev_stop(struct bnxt_re_dev *rdev) { - int mask = IB_QP_STATE; - struct ib_qp_attr qp_attr; struct bnxt_re_qp *qp; - qp_attr.qp_state = IB_QPS_ERR; mutex_lock(&rdev->qp_lock); list_for_each_entry(qp, &rdev->qp_list, list) { /* Modify the state of all QPs except QP1/Shadow QP */ @@ -1727,12 +1724,9 @@ static void bnxt_re_dev_stop(struct bnxt_re_dev *rdev) if (qp->qplib_qp.state != CMDQ_MODIFY_QP_NEW_STATE_RESET && qp->qplib_qp.state != - CMDQ_MODIFY_QP_NEW_STATE_ERR) { + CMDQ_MODIFY_QP_NEW_STATE_ERR) bnxt_re_dispatch_event(&rdev->ibdev, &qp->ib_qp, 1, IB_EVENT_QP_FATAL); - bnxt_re_modify_qp(&qp->ib_qp, &qp_attr, mask, - NULL); - } } } mutex_unlock(&rdev->qp_lock); diff --git a/drivers/infiniband/hw/bnxt_re/qplib_fp.c b/drivers/infiniband/hw/bnxt_re/qplib_fp.c index e42abf5be6c0..5336f74297f8 100644 --- a/drivers/infiniband/hw/bnxt_re/qplib_fp.c +++ b/drivers/infiniband/hw/bnxt_re/qplib_fp.c @@ -659,13 +659,6 @@ int bnxt_qplib_create_srq(struct bnxt_qplib_res *res, rc = bnxt_qplib_alloc_init_hwq(&srq->hwq, &hwq_attr); if (rc) return rc; - - srq->swq = kcalloc(srq->hwq.max_elements, sizeof(*srq->swq), - GFP_KERNEL); - if (!srq->swq) { - rc = -ENOMEM; - goto fail; - } srq->dbinfo.flags = 0; bnxt_qplib_rcfw_cmd_prep((struct cmdq_base *)&req, CMDQ_BASE_OPCODE_CREATE_SRQ, @@ -694,9 +687,17 @@ int bnxt_qplib_create_srq(struct bnxt_qplib_res *res, spin_lock_init(&srq->lock); srq->start_idx = 0; srq->last_idx = srq->hwq.max_elements - 1; - for (idx = 0; idx < srq->hwq.max_elements; idx++) - srq->swq[idx].next_idx = idx + 1; - srq->swq[srq->last_idx].next_idx = -1; + if (!srq->hwq.is_user) { + srq->swq = kcalloc(srq->hwq.max_elements, sizeof(*srq->swq), + GFP_KERNEL); + if (!srq->swq) { + rc = -ENOMEM; + goto fail; + } + for (idx = 0; idx < srq->hwq.max_elements; idx++) + srq->swq[idx].next_idx = idx + 1; + srq->swq[srq->last_idx].next_idx = -1; + } srq->id = le32_to_cpu(resp.xid); srq->dbinfo.hwq = &srq->hwq; @@ -1000,9 +1001,7 @@ int bnxt_qplib_create_qp(struct bnxt_qplib_res *res, struct bnxt_qplib_qp *qp) u32 tbl_indx; u16 nsge; - if (res->dattr) - qp->is_host_msn_tbl = _is_host_msn_table(res->dattr->dev_cap_flags2); - + qp->is_host_msn_tbl = _is_host_msn_table(res->dattr->dev_cap_flags2); sq->dbinfo.flags = 0; bnxt_qplib_rcfw_cmd_prep((struct cmdq_base *)&req, CMDQ_BASE_OPCODE_CREATE_QP, @@ -1034,7 +1033,12 @@ int bnxt_qplib_create_qp(struct bnxt_qplib_res *res, struct bnxt_qplib_qp *qp) : 0; /* Update msn tbl size */ if (qp->is_host_msn_tbl && psn_sz) { - hwq_attr.aux_depth = roundup_pow_of_two(bnxt_qplib_set_sq_size(sq, qp->wqe_mode)); + if (qp->wqe_mode == BNXT_QPLIB_WQE_MODE_STATIC) + hwq_attr.aux_depth = + roundup_pow_of_two(bnxt_qplib_set_sq_size(sq, qp->wqe_mode)); + else + hwq_attr.aux_depth = + roundup_pow_of_two(bnxt_qplib_set_sq_size(sq, qp->wqe_mode)) / 2; qp->msn_tbl_sz = hwq_attr.aux_depth; qp->msn = 0; } @@ -1044,13 +1048,14 @@ int bnxt_qplib_create_qp(struct bnxt_qplib_res *res, struct bnxt_qplib_qp *qp) if (rc) return rc; - rc = bnxt_qplib_alloc_init_swq(sq); - if (rc) - goto fail_sq; - - if (psn_sz) - bnxt_qplib_init_psn_ptr(qp, psn_sz); + if (!sq->hwq.is_user) { + rc = bnxt_qplib_alloc_init_swq(sq); + if (rc) + goto fail_sq; + if (psn_sz) + bnxt_qplib_init_psn_ptr(qp, psn_sz); + } req.sq_size = cpu_to_le32(bnxt_qplib_set_sq_size(sq, qp->wqe_mode)); pbl = &sq->hwq.pbl[PBL_LVL_0]; req.sq_pbl = cpu_to_le64(pbl->pg_map_arr[0]); @@ -1076,9 +1081,11 @@ int bnxt_qplib_create_qp(struct bnxt_qplib_res *res, struct bnxt_qplib_qp *qp) rc = bnxt_qplib_alloc_init_hwq(&rq->hwq, &hwq_attr); if (rc) goto sq_swq; - rc = bnxt_qplib_alloc_init_swq(rq); - if (rc) - goto fail_rq; + if (!rq->hwq.is_user) { + rc = bnxt_qplib_alloc_init_swq(rq); + if (rc) + goto fail_rq; + } req.rq_size = cpu_to_le32(rq->max_wqe); pbl = &rq->hwq.pbl[PBL_LVL_0]; @@ -1174,9 +1181,11 @@ int bnxt_qplib_create_qp(struct bnxt_qplib_res *res, struct bnxt_qplib_qp *qp) rq->dbinfo.db = qp->dpi->dbr; rq->dbinfo.max_slot = bnxt_qplib_set_rq_max_slot(rq->wqe_size); } + spin_lock_bh(&rcfw->tbl_lock); tbl_indx = map_qp_id_to_tbl_indx(qp->id, rcfw); rcfw->qp_tbl[tbl_indx].qp_id = qp->id; rcfw->qp_tbl[tbl_indx].qp_handle = (void *)qp; + spin_unlock_bh(&rcfw->tbl_lock); return 0; fail: @@ -1283,7 +1292,8 @@ static void __filter_modify_flags(struct bnxt_qplib_qp *qp) } } -static void bnxt_set_mandatory_attributes(struct bnxt_qplib_qp *qp, +static void bnxt_set_mandatory_attributes(struct bnxt_qplib_res *res, + struct bnxt_qplib_qp *qp, struct cmdq_modify_qp *req) { u32 mandatory_flags = 0; @@ -1298,6 +1308,14 @@ static void bnxt_set_mandatory_attributes(struct bnxt_qplib_qp *qp, mandatory_flags |= CMDQ_MODIFY_QP_MODIFY_MASK_PKEY; } + if (_is_min_rnr_in_rtr_rts_mandatory(res->dattr->dev_cap_flags2) && + (qp->cur_qp_state == CMDQ_MODIFY_QP_NEW_STATE_RTR && + qp->state == CMDQ_MODIFY_QP_NEW_STATE_RTS)) { + if (qp->type == CMDQ_MODIFY_QP_QP_TYPE_RC) + mandatory_flags |= + CMDQ_MODIFY_QP_MODIFY_MASK_MIN_RNR_TIMER; + } + if (qp->type == CMDQ_MODIFY_QP_QP_TYPE_UD || qp->type == CMDQ_MODIFY_QP_QP_TYPE_GSI) mandatory_flags |= CMDQ_MODIFY_QP_MODIFY_MASK_QKEY; @@ -1338,7 +1356,7 @@ int bnxt_qplib_modify_qp(struct bnxt_qplib_res *res, struct bnxt_qplib_qp *qp) /* Set mandatory attributes for INIT -> RTR and RTR -> RTS transition */ if (_is_optimize_modify_qp_supported(res->dattr->dev_cap_flags2) && is_optimized_state_transition(qp)) - bnxt_set_mandatory_attributes(qp, &req); + bnxt_set_mandatory_attributes(res, qp, &req); } bmask = qp->modify_flags; req.modify_mask = cpu_to_le32(qp->modify_flags); @@ -1521,6 +1539,7 @@ int bnxt_qplib_query_qp(struct bnxt_qplib_res *res, struct bnxt_qplib_qp *qp) qp->dest_qpn = le32_to_cpu(sb->dest_qp_id); memcpy(qp->smac, sb->src_mac, 6); qp->vlan_id = le16_to_cpu(sb->vlan_pcp_vlan_dei_vlan_id); + qp->port_id = le16_to_cpu(sb->port_id); bail: dma_free_coherent(&rcfw->pdev->dev, sbuf.size, sbuf.sb, sbuf.dma_addr); @@ -2667,10 +2686,12 @@ static int bnxt_qplib_cq_process_req(struct bnxt_qplib_cq *cq, bnxt_qplib_add_flush_qp(qp); } else { /* Before we complete, do WA 9060 */ - if (do_wa9060(qp, cq, cq_cons, sq->swq_last, - cqe_sq_cons)) { - *lib_qp = qp; - goto out; + if (!bnxt_qplib_is_chip_gen_p5_p7(qp->cctx)) { + if (do_wa9060(qp, cq, cq_cons, sq->swq_last, + cqe_sq_cons)) { + *lib_qp = qp; + goto out; + } } if (swq->flags & SQ_SEND_FLAGS_SIGNAL_COMP) { cqe->status = CQ_REQ_STATUS_OK; diff --git a/drivers/infiniband/hw/bnxt_re/qplib_fp.h b/drivers/infiniband/hw/bnxt_re/qplib_fp.h index ef3424c81345..0660101b5310 100644 --- a/drivers/infiniband/hw/bnxt_re/qplib_fp.h +++ b/drivers/infiniband/hw/bnxt_re/qplib_fp.h @@ -114,7 +114,6 @@ struct bnxt_qplib_sge { u32 size; }; -#define BNXT_QPLIB_QP_MAX_SGL 6 struct bnxt_qplib_swq { u64 wr_id; int next_idx; @@ -154,7 +153,7 @@ struct bnxt_qplib_swqe { #define BNXT_QPLIB_SWQE_FLAGS_UC_FENCE BIT(2) #define BNXT_QPLIB_SWQE_FLAGS_SOLICIT_EVENT BIT(3) #define BNXT_QPLIB_SWQE_FLAGS_INLINE BIT(4) - struct bnxt_qplib_sge sg_list[BNXT_QPLIB_QP_MAX_SGL]; + struct bnxt_qplib_sge sg_list[BNXT_VAR_MAX_SGE]; int num_sge; /* Max inline data is 96 bytes */ u32 inline_len; @@ -299,6 +298,7 @@ struct bnxt_qplib_qp { u32 dest_qpn; u8 smac[6]; u16 vlan_id; + u16 port_id; u8 nw_type; struct bnxt_qplib_ah ah; diff --git a/drivers/infiniband/hw/bnxt_re/qplib_rcfw.c b/drivers/infiniband/hw/bnxt_re/qplib_rcfw.c index 5e90ea232de8..17e62f22683b 100644 --- a/drivers/infiniband/hw/bnxt_re/qplib_rcfw.c +++ b/drivers/infiniband/hw/bnxt_re/qplib_rcfw.c @@ -424,7 +424,8 @@ static int __send_message_basic_sanity(struct bnxt_qplib_rcfw *rcfw, /* Prevent posting if f/w is not in a state to process */ if (test_bit(ERR_DEVICE_DETACHED, &rcfw->cmdq.flags)) - return bnxt_qplib_map_rc(opcode); + return -ENXIO; + if (test_bit(FIRMWARE_STALL_DETECTED, &cmdq->flags)) return -ETIMEDOUT; @@ -493,7 +494,7 @@ static int __bnxt_qplib_rcfw_send_message(struct bnxt_qplib_rcfw *rcfw, rc = __send_message_basic_sanity(rcfw, msg, opcode); if (rc) - return rc; + return rc == -ENXIO ? bnxt_qplib_map_rc(opcode) : rc; rc = __send_message(rcfw, msg, opcode); if (rc) diff --git a/drivers/infiniband/hw/bnxt_re/qplib_res.h b/drivers/infiniband/hw/bnxt_re/qplib_res.h index 21fb148713a6..cbfc49a1a56d 100644 --- a/drivers/infiniband/hw/bnxt_re/qplib_res.h +++ b/drivers/infiniband/hw/bnxt_re/qplib_res.h @@ -584,6 +584,11 @@ static inline bool _is_optimize_modify_qp_supported(u16 dev_cap_ext_flags2) return dev_cap_ext_flags2 & CREQ_QUERY_FUNC_RESP_SB_OPTIMIZE_MODIFY_QP_SUPPORTED; } +static inline bool _is_min_rnr_in_rtr_rts_mandatory(u16 dev_cap_ext_flags2) +{ + return !!(dev_cap_ext_flags2 & CREQ_QUERY_FUNC_RESP_SB_MIN_RNR_RTR_RTS_OPT_SUPPORTED); +} + static inline bool _is_cq_coalescing_supported(u16 dev_cap_ext_flags2) { return dev_cap_ext_flags2 & CREQ_QUERY_FUNC_RESP_SB_CQ_COALESCING_SUPPORTED; diff --git a/drivers/infiniband/hw/bnxt_re/qplib_sp.c b/drivers/infiniband/hw/bnxt_re/qplib_sp.c index 7e20ae3d2c4f..9df3e3271577 100644 --- a/drivers/infiniband/hw/bnxt_re/qplib_sp.c +++ b/drivers/infiniband/hw/bnxt_re/qplib_sp.c @@ -129,12 +129,18 @@ int bnxt_qplib_get_dev_attr(struct bnxt_qplib_rcfw *rcfw, attr->max_qp_init_rd_atom = sb->max_qp_init_rd_atom > BNXT_QPLIB_MAX_OUT_RD_ATOM ? BNXT_QPLIB_MAX_OUT_RD_ATOM : sb->max_qp_init_rd_atom; - attr->max_qp_wqes = le16_to_cpu(sb->max_qp_wr); - /* - * 128 WQEs needs to be reserved for the HW (8916). Prevent - * reporting the max number - */ - attr->max_qp_wqes -= BNXT_QPLIB_RESERVED_QP_WRS + 1; + attr->max_qp_wqes = le16_to_cpu(sb->max_qp_wr) - 1; + if (!bnxt_qplib_is_chip_gen_p5_p7(rcfw->res->cctx)) { + /* + * 128 WQEs needs to be reserved for the HW (8916). Prevent + * reporting the max number on legacy devices + */ + attr->max_qp_wqes -= BNXT_QPLIB_RESERVED_QP_WRS + 1; + } + + /* Adjust for max_qp_wqes for variable wqe */ + if (cctx->modes.wqe_mode == BNXT_QPLIB_WQE_MODE_VARIABLE) + attr->max_qp_wqes = BNXT_VAR_MAX_WQE - 1; attr->max_qp_sges = cctx->modes.wqe_mode == BNXT_QPLIB_WQE_MODE_VARIABLE ? min_t(u32, sb->max_sge_var_wqe, BNXT_VAR_MAX_SGE) : 6; diff --git a/drivers/infiniband/hw/bnxt_re/roce_hsi.h b/drivers/infiniband/hw/bnxt_re/roce_hsi.h index a98fc9c2313e..0ee60fdc18b3 100644 --- a/drivers/infiniband/hw/bnxt_re/roce_hsi.h +++ b/drivers/infiniband/hw/bnxt_re/roce_hsi.h @@ -2215,6 +2215,7 @@ struct creq_query_func_resp_sb { #define CREQ_QUERY_FUNC_RESP_SB_REQ_RETRANSMISSION_SUPPORT_IQM_MSN_TABLE (0x2UL << 4) #define CREQ_QUERY_FUNC_RESP_SB_REQ_RETRANSMISSION_SUPPORT_LAST \ CREQ_QUERY_FUNC_RESP_SB_REQ_RETRANSMISSION_SUPPORT_IQM_MSN_TABLE + #define CREQ_QUERY_FUNC_RESP_SB_MIN_RNR_RTR_RTS_OPT_SUPPORTED 0x1000UL __le16 max_xp_qp_size; __le16 create_qp_batch_size; __le16 destroy_qp_batch_size; diff --git a/drivers/infiniband/hw/hns/hns_roce_hem.c b/drivers/infiniband/hw/hns/hns_roce_hem.c index f84521be3bea..605562122ecc 100644 --- a/drivers/infiniband/hw/hns/hns_roce_hem.c +++ b/drivers/infiniband/hw/hns/hns_roce_hem.c @@ -931,6 +931,7 @@ struct hns_roce_hem_item { size_t count; /* max ba numbers */ int start; /* start buf offset in this hem */ int end; /* end buf offset in this hem */ + bool exist_bt; }; /* All HEM items are linked in a tree structure */ @@ -959,6 +960,7 @@ hem_list_alloc_item(struct hns_roce_dev *hr_dev, int start, int end, int count, } } + hem->exist_bt = exist_bt; hem->count = count; hem->start = start; hem->end = end; @@ -969,22 +971,22 @@ hem_list_alloc_item(struct hns_roce_dev *hr_dev, int start, int end, int count, } static void hem_list_free_item(struct hns_roce_dev *hr_dev, - struct hns_roce_hem_item *hem, bool exist_bt) + struct hns_roce_hem_item *hem) { - if (exist_bt) + if (hem->exist_bt) dma_free_coherent(hr_dev->dev, hem->count * BA_BYTE_LEN, hem->addr, hem->dma_addr); kfree(hem); } static void hem_list_free_all(struct hns_roce_dev *hr_dev, - struct list_head *head, bool exist_bt) + struct list_head *head) { struct hns_roce_hem_item *hem, *temp_hem; list_for_each_entry_safe(hem, temp_hem, head, list) { list_del(&hem->list); - hem_list_free_item(hr_dev, hem, exist_bt); + hem_list_free_item(hr_dev, hem); } } @@ -1084,6 +1086,10 @@ int hns_roce_hem_list_calc_root_ba(const struct hns_roce_buf_region *regions, for (i = 0; i < region_cnt; i++) { r = (struct hns_roce_buf_region *)®ions[i]; + /* when r->hopnum = 0, the region should not occupy root_ba. */ + if (!r->hopnum) + continue; + if (r->hopnum > 1) { step = hem_list_calc_ba_range(r->hopnum, 1, unit); if (step > 0) @@ -1177,7 +1183,7 @@ static int hem_list_alloc_mid_bt(struct hns_roce_dev *hr_dev, err_exit: for (level = 1; level < hopnum; level++) - hem_list_free_all(hr_dev, &temp_list[level], true); + hem_list_free_all(hr_dev, &temp_list[level]); return ret; } @@ -1218,16 +1224,26 @@ static int alloc_fake_root_bt(struct hns_roce_dev *hr_dev, void *cpu_base, { struct hns_roce_hem_item *hem; + /* This is on the has_mtt branch, if r->hopnum + * is 0, there is no root_ba to reuse for the + * region's fake hem, so a dma_alloc request is + * necessary here. + */ hem = hem_list_alloc_item(hr_dev, r->offset, r->offset + r->count - 1, - r->count, false); + r->count, !r->hopnum); if (!hem) return -ENOMEM; - hem_list_assign_bt(hem, cpu_base, phy_base); + /* The root_ba can be reused only when r->hopnum > 0. */ + if (r->hopnum) + hem_list_assign_bt(hem, cpu_base, phy_base); list_add(&hem->list, branch_head); list_add(&hem->sibling, leaf_head); - return r->count; + /* If r->hopnum == 0, 0 is returned, + * so that the root_bt entry is not occupied. + */ + return r->hopnum ? r->count : 0; } static int setup_middle_bt(struct hns_roce_dev *hr_dev, void *cpu_base, @@ -1271,7 +1287,7 @@ setup_root_hem(struct hns_roce_dev *hr_dev, struct hns_roce_hem_list *hem_list, return -ENOMEM; total = 0; - for (i = 0; i < region_cnt && total < max_ba_num; i++) { + for (i = 0; i < region_cnt && total <= max_ba_num; i++) { r = ®ions[i]; if (!r->count) continue; @@ -1337,9 +1353,9 @@ static int hem_list_alloc_root_bt(struct hns_roce_dev *hr_dev, region_cnt); if (ret) { for (i = 0; i < region_cnt; i++) - hem_list_free_all(hr_dev, &head.branch[i], false); + hem_list_free_all(hr_dev, &head.branch[i]); - hem_list_free_all(hr_dev, &head.root, true); + hem_list_free_all(hr_dev, &head.root); } return ret; @@ -1402,10 +1418,9 @@ void hns_roce_hem_list_release(struct hns_roce_dev *hr_dev, for (i = 0; i < HNS_ROCE_MAX_BT_REGION; i++) for (j = 0; j < HNS_ROCE_MAX_BT_LEVEL; j++) - hem_list_free_all(hr_dev, &hem_list->mid_bt[i][j], - j != 0); + hem_list_free_all(hr_dev, &hem_list->mid_bt[i][j]); - hem_list_free_all(hr_dev, &hem_list->root_bt, true); + hem_list_free_all(hr_dev, &hem_list->root_bt); INIT_LIST_HEAD(&hem_list->btm_bt); hem_list->root_ba = 0; } diff --git a/drivers/infiniband/hw/hns/hns_roce_hw_v2.c b/drivers/infiniband/hw/hns/hns_roce_hw_v2.c index 697b17cca02e..0144e7210d05 100644 --- a/drivers/infiniband/hw/hns/hns_roce_hw_v2.c +++ b/drivers/infiniband/hw/hns/hns_roce_hw_v2.c @@ -468,7 +468,7 @@ static inline int set_ud_wqe(struct hns_roce_qp *qp, valid_num_sge = calc_wr_sge_num(wr, &msg_len); ret = set_ud_opcode(ud_sq_wqe, wr); - if (WARN_ON(ret)) + if (WARN_ON_ONCE(ret)) return ret; ud_sq_wqe->msg_len = cpu_to_le32(msg_len); @@ -572,7 +572,7 @@ static inline int set_rc_wqe(struct hns_roce_qp *qp, rc_sq_wqe->msg_len = cpu_to_le32(msg_len); ret = set_rc_opcode(hr_dev, rc_sq_wqe, wr); - if (WARN_ON(ret)) + if (WARN_ON_ONCE(ret)) return ret; hr_reg_write(rc_sq_wqe, RC_SEND_WQE_SO, @@ -670,6 +670,10 @@ static void write_dwqe(struct hns_roce_dev *hr_dev, struct hns_roce_qp *qp, #define HNS_ROCE_SL_SHIFT 2 struct hns_roce_v2_rc_send_wqe *rc_sq_wqe = wqe; + if (unlikely(qp->state == IB_QPS_ERR)) { + flush_cqe(hr_dev, qp); + return; + } /* All kinds of DirectWQE have the same header field layout */ hr_reg_enable(rc_sq_wqe, RC_SEND_WQE_FLAG); hr_reg_write(rc_sq_wqe, RC_SEND_WQE_DB_SL_L, qp->sl); @@ -5619,6 +5623,9 @@ static void put_dip_ctx_idx(struct hns_roce_dev *hr_dev, { struct hns_roce_dip *hr_dip = hr_qp->dip; + if (!hr_dip) + return; + xa_lock(&hr_dev->qp_table.dip_xa); hr_dip->qp_cnt--; diff --git a/drivers/infiniband/hw/hns/hns_roce_mr.c b/drivers/infiniband/hw/hns/hns_roce_mr.c index bf30b3a65a9b..55b9283bfc6f 100644 --- a/drivers/infiniband/hw/hns/hns_roce_mr.c +++ b/drivers/infiniband/hw/hns/hns_roce_mr.c @@ -814,11 +814,6 @@ int hns_roce_mtr_map(struct hns_roce_dev *hr_dev, struct hns_roce_mtr *mtr, for (i = 0, mapped_cnt = 0; i < mtr->hem_cfg.region_count && mapped_cnt < page_cnt; i++) { r = &mtr->hem_cfg.region[i]; - /* if hopnum is 0, no need to map pages in this region */ - if (!r->hopnum) { - mapped_cnt += r->count; - continue; - } if (r->offset + r->count > page_cnt) { ret = -EINVAL; diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c index bc7930d0c564..f5b59d02f4d3 100644 --- a/drivers/infiniband/hw/mlx5/main.c +++ b/drivers/infiniband/hw/mlx5/main.c @@ -2839,7 +2839,7 @@ static int mlx5_ib_get_plane_num(struct mlx5_core_dev *mdev, u8 *num_plane) int err; *num_plane = 0; - if (!MLX5_CAP_GEN(mdev, ib_virt)) + if (!MLX5_CAP_GEN(mdev, ib_virt) || !MLX5_CAP_GEN_2(mdev, multiplane)) return 0; err = mlx5_query_hca_vport_context(mdev, 0, 1, 0, &vport_ctx); @@ -3639,7 +3639,8 @@ static int mlx5_ib_init_multiport_master(struct mlx5_ib_dev *dev) list_for_each_entry(mpi, &mlx5_ib_unaffiliated_port_list, list) { if (dev->sys_image_guid == mpi->sys_image_guid && - (mlx5_core_native_port_num(mpi->mdev) - 1) == i) { + (mlx5_core_native_port_num(mpi->mdev) - 1) == i && + mlx5_core_same_coredev_type(dev->mdev, mpi->mdev)) { bound = mlx5_ib_bind_slave_port(dev, mpi); } @@ -4785,7 +4786,8 @@ static int mlx5r_mp_probe(struct auxiliary_device *adev, mutex_lock(&mlx5_ib_multiport_mutex); list_for_each_entry(dev, &mlx5_ib_dev_list, ib_dev_list) { - if (dev->sys_image_guid == mpi->sys_image_guid) + if (dev->sys_image_guid == mpi->sys_image_guid && + mlx5_core_same_coredev_type(dev->mdev, mpi->mdev)) bound = mlx5_ib_bind_slave_port(dev, mpi); if (bound) { diff --git a/drivers/infiniband/sw/rxe/rxe.c b/drivers/infiniband/sw/rxe/rxe.c index 255677bc12b2..1ba4a0c8726a 100644 --- a/drivers/infiniband/sw/rxe/rxe.c +++ b/drivers/infiniband/sw/rxe/rxe.c @@ -40,6 +40,8 @@ void rxe_dealloc(struct ib_device *ib_dev) /* initialize rxe device parameters */ static void rxe_init_device_param(struct rxe_dev *rxe) { + struct net_device *ndev; + rxe->max_inline_data = RXE_MAX_INLINE_DATA; rxe->attr.vendor_id = RXE_VENDOR_ID; @@ -71,8 +73,15 @@ static void rxe_init_device_param(struct rxe_dev *rxe) rxe->attr.max_fast_reg_page_list_len = RXE_MAX_FMR_PAGE_LIST_LEN; rxe->attr.max_pkeys = RXE_MAX_PKEYS; rxe->attr.local_ca_ack_delay = RXE_LOCAL_CA_ACK_DELAY; + + ndev = rxe_ib_device_get_netdev(&rxe->ib_dev); + if (!ndev) + return; + addrconf_addr_eui48((unsigned char *)&rxe->attr.sys_image_guid, - rxe->ndev->dev_addr); + ndev->dev_addr); + + dev_put(ndev); rxe->max_ucontext = RXE_MAX_UCONTEXT; } @@ -109,10 +118,15 @@ static void rxe_init_port_param(struct rxe_port *port) static void rxe_init_ports(struct rxe_dev *rxe) { struct rxe_port *port = &rxe->port; + struct net_device *ndev; rxe_init_port_param(port); + ndev = rxe_ib_device_get_netdev(&rxe->ib_dev); + if (!ndev) + return; addrconf_addr_eui48((unsigned char *)&port->port_guid, - rxe->ndev->dev_addr); + ndev->dev_addr); + dev_put(ndev); spin_lock_init(&port->port_lock); } @@ -167,12 +181,13 @@ void rxe_set_mtu(struct rxe_dev *rxe, unsigned int ndev_mtu) /* called by ifc layer to create new rxe device. * The caller should allocate memory for rxe by calling ib_alloc_device. */ -int rxe_add(struct rxe_dev *rxe, unsigned int mtu, const char *ibdev_name) +int rxe_add(struct rxe_dev *rxe, unsigned int mtu, const char *ibdev_name, + struct net_device *ndev) { rxe_init(rxe); rxe_set_mtu(rxe, mtu); - return rxe_register_device(rxe, ibdev_name); + return rxe_register_device(rxe, ibdev_name, ndev); } static int rxe_newlink(const char *ibdev_name, struct net_device *ndev) diff --git a/drivers/infiniband/sw/rxe/rxe.h b/drivers/infiniband/sw/rxe/rxe.h index d8fb2c7af30a..fe7f97066732 100644 --- a/drivers/infiniband/sw/rxe/rxe.h +++ b/drivers/infiniband/sw/rxe/rxe.h @@ -139,7 +139,8 @@ enum resp_states { void rxe_set_mtu(struct rxe_dev *rxe, unsigned int dev_mtu); -int rxe_add(struct rxe_dev *rxe, unsigned int mtu, const char *ibdev_name); +int rxe_add(struct rxe_dev *rxe, unsigned int mtu, const char *ibdev_name, + struct net_device *ndev); void rxe_rcv(struct sk_buff *skb); diff --git a/drivers/infiniband/sw/rxe/rxe_mcast.c b/drivers/infiniband/sw/rxe/rxe_mcast.c index 86cc2e18a7fd..07ff47bae31d 100644 --- a/drivers/infiniband/sw/rxe/rxe_mcast.c +++ b/drivers/infiniband/sw/rxe/rxe_mcast.c @@ -31,10 +31,19 @@ static int rxe_mcast_add(struct rxe_dev *rxe, union ib_gid *mgid) { unsigned char ll_addr[ETH_ALEN]; + struct net_device *ndev; + int ret; + + ndev = rxe_ib_device_get_netdev(&rxe->ib_dev); + if (!ndev) + return -ENODEV; ipv6_eth_mc_map((struct in6_addr *)mgid->raw, ll_addr); - return dev_mc_add(rxe->ndev, ll_addr); + ret = dev_mc_add(ndev, ll_addr); + dev_put(ndev); + + return ret; } /** @@ -47,10 +56,19 @@ static int rxe_mcast_add(struct rxe_dev *rxe, union ib_gid *mgid) static int rxe_mcast_del(struct rxe_dev *rxe, union ib_gid *mgid) { unsigned char ll_addr[ETH_ALEN]; + struct net_device *ndev; + int ret; + + ndev = rxe_ib_device_get_netdev(&rxe->ib_dev); + if (!ndev) + return -ENODEV; ipv6_eth_mc_map((struct in6_addr *)mgid->raw, ll_addr); - return dev_mc_del(rxe->ndev, ll_addr); + ret = dev_mc_del(ndev, ll_addr); + dev_put(ndev); + + return ret; } /** diff --git a/drivers/infiniband/sw/rxe/rxe_net.c b/drivers/infiniband/sw/rxe/rxe_net.c index 75d1407db52d..8cc64ceeb356 100644 --- a/drivers/infiniband/sw/rxe/rxe_net.c +++ b/drivers/infiniband/sw/rxe/rxe_net.c @@ -524,7 +524,16 @@ out: */ const char *rxe_parent_name(struct rxe_dev *rxe, unsigned int port_num) { - return rxe->ndev->name; + struct net_device *ndev; + char *ndev_name; + + ndev = rxe_ib_device_get_netdev(&rxe->ib_dev); + if (!ndev) + return NULL; + ndev_name = ndev->name; + dev_put(ndev); + + return ndev_name; } int rxe_net_add(const char *ibdev_name, struct net_device *ndev) @@ -536,10 +545,9 @@ int rxe_net_add(const char *ibdev_name, struct net_device *ndev) if (!rxe) return -ENOMEM; - rxe->ndev = ndev; ib_mark_name_assigned_by_user(&rxe->ib_dev); - err = rxe_add(rxe, ndev->mtu, ibdev_name); + err = rxe_add(rxe, ndev->mtu, ibdev_name, ndev); if (err) { ib_dealloc_device(&rxe->ib_dev); return err; @@ -587,10 +595,18 @@ void rxe_port_down(struct rxe_dev *rxe) void rxe_set_port_state(struct rxe_dev *rxe) { - if (netif_running(rxe->ndev) && netif_carrier_ok(rxe->ndev)) + struct net_device *ndev; + + ndev = rxe_ib_device_get_netdev(&rxe->ib_dev); + if (!ndev) + return; + + if (netif_running(ndev) && netif_carrier_ok(ndev)) rxe_port_up(rxe); else rxe_port_down(rxe); + + dev_put(ndev); } static int rxe_notify(struct notifier_block *not_blk, diff --git a/drivers/infiniband/sw/rxe/rxe_verbs.c b/drivers/infiniband/sw/rxe/rxe_verbs.c index 5c18f7e342f2..8a5fc20fd186 100644 --- a/drivers/infiniband/sw/rxe/rxe_verbs.c +++ b/drivers/infiniband/sw/rxe/rxe_verbs.c @@ -41,6 +41,7 @@ static int rxe_query_port(struct ib_device *ibdev, u32 port_num, struct ib_port_attr *attr) { struct rxe_dev *rxe = to_rdev(ibdev); + struct net_device *ndev; int err, ret; if (port_num != 1) { @@ -49,6 +50,12 @@ static int rxe_query_port(struct ib_device *ibdev, goto err_out; } + ndev = rxe_ib_device_get_netdev(ibdev); + if (!ndev) { + err = -ENODEV; + goto err_out; + } + memcpy(attr, &rxe->port.attr, sizeof(*attr)); mutex_lock(&rxe->usdev_lock); @@ -57,13 +64,14 @@ static int rxe_query_port(struct ib_device *ibdev, if (attr->state == IB_PORT_ACTIVE) attr->phys_state = IB_PORT_PHYS_STATE_LINK_UP; - else if (dev_get_flags(rxe->ndev) & IFF_UP) + else if (dev_get_flags(ndev) & IFF_UP) attr->phys_state = IB_PORT_PHYS_STATE_POLLING; else attr->phys_state = IB_PORT_PHYS_STATE_DISABLED; mutex_unlock(&rxe->usdev_lock); + dev_put(ndev); return ret; err_out: @@ -1425,9 +1433,16 @@ static const struct attribute_group rxe_attr_group = { static int rxe_enable_driver(struct ib_device *ib_dev) { struct rxe_dev *rxe = container_of(ib_dev, struct rxe_dev, ib_dev); + struct net_device *ndev; + + ndev = rxe_ib_device_get_netdev(ib_dev); + if (!ndev) + return -ENODEV; rxe_set_port_state(rxe); - dev_info(&rxe->ib_dev.dev, "added %s\n", netdev_name(rxe->ndev)); + dev_info(&rxe->ib_dev.dev, "added %s\n", netdev_name(ndev)); + + dev_put(ndev); return 0; } @@ -1495,7 +1510,8 @@ static const struct ib_device_ops rxe_dev_ops = { INIT_RDMA_OBJ_SIZE(ib_mw, rxe_mw, ibmw), }; -int rxe_register_device(struct rxe_dev *rxe, const char *ibdev_name) +int rxe_register_device(struct rxe_dev *rxe, const char *ibdev_name, + struct net_device *ndev) { int err; struct ib_device *dev = &rxe->ib_dev; @@ -1507,13 +1523,13 @@ int rxe_register_device(struct rxe_dev *rxe, const char *ibdev_name) dev->num_comp_vectors = num_possible_cpus(); dev->local_dma_lkey = 0; addrconf_addr_eui48((unsigned char *)&dev->node_guid, - rxe->ndev->dev_addr); + ndev->dev_addr); dev->uverbs_cmd_mask |= BIT_ULL(IB_USER_VERBS_CMD_POST_SEND) | BIT_ULL(IB_USER_VERBS_CMD_REQ_NOTIFY_CQ); ib_set_device_ops(dev, &rxe_dev_ops); - err = ib_device_set_netdev(&rxe->ib_dev, rxe->ndev, 1); + err = ib_device_set_netdev(&rxe->ib_dev, ndev, 1); if (err) return err; diff --git a/drivers/infiniband/sw/rxe/rxe_verbs.h b/drivers/infiniband/sw/rxe/rxe_verbs.h index 3c1354f82283..6573ceec0ef5 100644 --- a/drivers/infiniband/sw/rxe/rxe_verbs.h +++ b/drivers/infiniband/sw/rxe/rxe_verbs.h @@ -370,6 +370,7 @@ struct rxe_port { u32 qp_gsi_index; }; +#define RXE_PORT 1 struct rxe_dev { struct ib_device ib_dev; struct ib_device_attr attr; @@ -377,8 +378,6 @@ struct rxe_dev { int max_inline_data; struct mutex usdev_lock; - struct net_device *ndev; - struct rxe_pool uc_pool; struct rxe_pool pd_pool; struct rxe_pool ah_pool; @@ -406,6 +405,11 @@ struct rxe_dev { struct crypto_shash *tfm; }; +static inline struct net_device *rxe_ib_device_get_netdev(struct ib_device *dev) +{ + return ib_device_get_netdev(dev, RXE_PORT); +} + static inline void rxe_counter_inc(struct rxe_dev *rxe, enum rxe_counters index) { atomic64_inc(&rxe->stats_counters[index]); @@ -471,6 +475,7 @@ static inline struct rxe_pd *rxe_mw_pd(struct rxe_mw *mw) return to_rpd(mw->ibmw.pd); } -int rxe_register_device(struct rxe_dev *rxe, const char *ibdev_name); +int rxe_register_device(struct rxe_dev *rxe, const char *ibdev_name, + struct net_device *ndev); #endif /* RXE_VERBS_H */ diff --git a/drivers/infiniband/sw/siw/siw.h b/drivers/infiniband/sw/siw/siw.h index 86d4d6a2170e..ea5eee50dc39 100644 --- a/drivers/infiniband/sw/siw/siw.h +++ b/drivers/infiniband/sw/siw/siw.h @@ -46,6 +46,9 @@ */ #define SIW_IRQ_MAXBURST_SQ_ACTIVE 4 +/* There is always only a port 1 per siw device */ +#define SIW_PORT 1 + struct siw_dev_cap { int max_qp; int max_qp_wr; @@ -69,16 +72,12 @@ struct siw_pd { struct siw_device { struct ib_device base_dev; - struct net_device *netdev; struct siw_dev_cap attrs; u32 vendor_part_id; int numa_node; char raw_gid[ETH_ALEN]; - /* physical port state (only one port per device) */ - enum ib_port_state state; - spinlock_t lock; struct xarray qp_xa; diff --git a/drivers/infiniband/sw/siw/siw_cm.c b/drivers/infiniband/sw/siw/siw_cm.c index 86323918a570..708b13993fdf 100644 --- a/drivers/infiniband/sw/siw/siw_cm.c +++ b/drivers/infiniband/sw/siw/siw_cm.c @@ -1759,6 +1759,7 @@ int siw_create_listen(struct iw_cm_id *id, int backlog) { struct socket *s; struct siw_cep *cep = NULL; + struct net_device *ndev = NULL; struct siw_device *sdev = to_siw_dev(id->device); int addr_family = id->local_addr.ss_family; int rv = 0; @@ -1779,9 +1780,15 @@ int siw_create_listen(struct iw_cm_id *id, int backlog) struct sockaddr_in *laddr = &to_sockaddr_in(id->local_addr); /* For wildcard addr, limit binding to current device only */ - if (ipv4_is_zeronet(laddr->sin_addr.s_addr)) - s->sk->sk_bound_dev_if = sdev->netdev->ifindex; - + if (ipv4_is_zeronet(laddr->sin_addr.s_addr)) { + ndev = ib_device_get_netdev(id->device, SIW_PORT); + if (ndev) { + s->sk->sk_bound_dev_if = ndev->ifindex; + } else { + rv = -ENODEV; + goto error; + } + } rv = s->ops->bind(s, (struct sockaddr *)laddr, sizeof(struct sockaddr_in)); } else { @@ -1797,9 +1804,15 @@ int siw_create_listen(struct iw_cm_id *id, int backlog) } /* For wildcard addr, limit binding to current device only */ - if (ipv6_addr_any(&laddr->sin6_addr)) - s->sk->sk_bound_dev_if = sdev->netdev->ifindex; - + if (ipv6_addr_any(&laddr->sin6_addr)) { + ndev = ib_device_get_netdev(id->device, SIW_PORT); + if (ndev) { + s->sk->sk_bound_dev_if = ndev->ifindex; + } else { + rv = -ENODEV; + goto error; + } + } rv = s->ops->bind(s, (struct sockaddr *)laddr, sizeof(struct sockaddr_in6)); } @@ -1860,6 +1873,7 @@ int siw_create_listen(struct iw_cm_id *id, int backlog) } list_add_tail(&cep->listenq, (struct list_head *)id->provider_data); cep->state = SIW_EPSTATE_LISTENING; + dev_put(ndev); siw_dbg(id->device, "Listen at laddr %pISp\n", &id->local_addr); @@ -1879,6 +1893,7 @@ error: siw_cep_set_free_and_put(cep); } sock_release(s); + dev_put(ndev); return rv; } diff --git a/drivers/infiniband/sw/siw/siw_main.c b/drivers/infiniband/sw/siw/siw_main.c index 17abef48abcd..14d3103aee6f 100644 --- a/drivers/infiniband/sw/siw/siw_main.c +++ b/drivers/infiniband/sw/siw/siw_main.c @@ -287,7 +287,6 @@ static struct siw_device *siw_device_create(struct net_device *netdev) return NULL; base_dev = &sdev->base_dev; - sdev->netdev = netdev; if (netdev->addr_len) { memcpy(sdev->raw_gid, netdev->dev_addr, @@ -381,12 +380,10 @@ static int siw_netdev_event(struct notifier_block *nb, unsigned long event, switch (event) { case NETDEV_UP: - sdev->state = IB_PORT_ACTIVE; siw_port_event(sdev, 1, IB_EVENT_PORT_ACTIVE); break; case NETDEV_DOWN: - sdev->state = IB_PORT_DOWN; siw_port_event(sdev, 1, IB_EVENT_PORT_ERR); break; @@ -407,12 +404,8 @@ static int siw_netdev_event(struct notifier_block *nb, unsigned long event, siw_port_event(sdev, 1, IB_EVENT_LID_CHANGE); break; /* - * Todo: Below netdev events are currently not handled. + * All other events are not handled */ - case NETDEV_CHANGEMTU: - case NETDEV_CHANGE: - break; - default: break; } @@ -442,12 +435,6 @@ static int siw_newlink(const char *basedev_name, struct net_device *netdev) sdev = siw_device_create(netdev); if (sdev) { dev_dbg(&netdev->dev, "siw: new device\n"); - - if (netif_running(netdev) && netif_carrier_ok(netdev)) - sdev->state = IB_PORT_ACTIVE; - else - sdev->state = IB_PORT_DOWN; - ib_mark_name_assigned_by_user(&sdev->base_dev); rv = siw_device_register(sdev, basedev_name); if (rv) diff --git a/drivers/infiniband/sw/siw/siw_verbs.c b/drivers/infiniband/sw/siw/siw_verbs.c index 986666c19378..7ca0297d68a4 100644 --- a/drivers/infiniband/sw/siw/siw_verbs.c +++ b/drivers/infiniband/sw/siw/siw_verbs.c @@ -171,21 +171,29 @@ int siw_query_device(struct ib_device *base_dev, struct ib_device_attr *attr, int siw_query_port(struct ib_device *base_dev, u32 port, struct ib_port_attr *attr) { - struct siw_device *sdev = to_siw_dev(base_dev); + struct net_device *ndev; int rv; memset(attr, 0, sizeof(*attr)); rv = ib_get_eth_speed(base_dev, port, &attr->active_speed, &attr->active_width); + if (rv) + return rv; + + ndev = ib_device_get_netdev(base_dev, SIW_PORT); + if (!ndev) + return -ENODEV; + attr->gid_tbl_len = 1; attr->max_msg_sz = -1; - attr->max_mtu = ib_mtu_int_to_enum(sdev->netdev->mtu); - attr->active_mtu = ib_mtu_int_to_enum(sdev->netdev->mtu); - attr->phys_state = sdev->state == IB_PORT_ACTIVE ? + attr->max_mtu = ib_mtu_int_to_enum(ndev->max_mtu); + attr->active_mtu = ib_mtu_int_to_enum(READ_ONCE(ndev->mtu)); + attr->phys_state = (netif_running(ndev) && netif_carrier_ok(ndev)) ? IB_PORT_PHYS_STATE_LINK_UP : IB_PORT_PHYS_STATE_DISABLED; + attr->state = attr->phys_state == IB_PORT_PHYS_STATE_LINK_UP ? + IB_PORT_ACTIVE : IB_PORT_DOWN; attr->port_cap_flags = IB_PORT_CM_SUP | IB_PORT_DEVICE_MGMT_SUP; - attr->state = sdev->state; /* * All zero * @@ -199,6 +207,7 @@ int siw_query_port(struct ib_device *base_dev, u32 port, * attr->subnet_timeout = 0; * attr->init_type_repy = 0; */ + dev_put(ndev); return rv; } @@ -505,21 +514,24 @@ int siw_query_qp(struct ib_qp *base_qp, struct ib_qp_attr *qp_attr, int qp_attr_mask, struct ib_qp_init_attr *qp_init_attr) { struct siw_qp *qp; - struct siw_device *sdev; + struct net_device *ndev; - if (base_qp && qp_attr && qp_init_attr) { + if (base_qp && qp_attr && qp_init_attr) qp = to_siw_qp(base_qp); - sdev = to_siw_dev(base_qp->device); - } else { + else return -EINVAL; - } + + ndev = ib_device_get_netdev(base_qp->device, SIW_PORT); + if (!ndev) + return -ENODEV; + qp_attr->qp_state = siw_qp_state_to_ib_qp_state[qp->attrs.state]; qp_attr->cap.max_inline_data = SIW_MAX_INLINE; qp_attr->cap.max_send_wr = qp->attrs.sq_size; qp_attr->cap.max_send_sge = qp->attrs.sq_max_sges; qp_attr->cap.max_recv_wr = qp->attrs.rq_size; qp_attr->cap.max_recv_sge = qp->attrs.rq_max_sges; - qp_attr->path_mtu = ib_mtu_int_to_enum(sdev->netdev->mtu); + qp_attr->path_mtu = ib_mtu_int_to_enum(READ_ONCE(ndev->mtu)); qp_attr->max_rd_atomic = qp->attrs.irq_size; qp_attr->max_dest_rd_atomic = qp->attrs.orq_size; @@ -534,6 +546,7 @@ int siw_query_qp(struct ib_qp *base_qp, struct ib_qp_attr *qp_attr, qp_init_attr->cap = qp_attr->cap; + dev_put(ndev); return 0; } diff --git a/drivers/infiniband/ulp/rtrs/rtrs-srv.c b/drivers/infiniband/ulp/rtrs/rtrs-srv.c index e83d95647852..ef4abdea3c2d 100644 --- a/drivers/infiniband/ulp/rtrs/rtrs-srv.c +++ b/drivers/infiniband/ulp/rtrs/rtrs-srv.c @@ -349,6 +349,7 @@ static int send_io_resp_imm(struct rtrs_srv_con *con, struct rtrs_srv_op *id, struct rtrs_srv_mr *srv_mr; bool need_inval = false; enum ib_send_flags flags; + struct ib_sge list; u32 imm; int err; @@ -401,7 +402,6 @@ static int send_io_resp_imm(struct rtrs_srv_con *con, struct rtrs_srv_op *id, imm = rtrs_to_io_rsp_imm(id->msg_id, errno, need_inval); imm_wr.wr.next = NULL; if (always_invalidate) { - struct ib_sge list; struct rtrs_msg_rkey_rsp *msg; srv_mr = &srv_path->mrs[id->msg_id]; diff --git a/drivers/interconnect/icc-clk.c b/drivers/interconnect/icc-clk.c index b956e4050f38..88f311c11020 100644 --- a/drivers/interconnect/icc-clk.c +++ b/drivers/interconnect/icc-clk.c @@ -116,6 +116,11 @@ struct icc_provider *icc_clk_register(struct device *dev, } node->name = devm_kasprintf(dev, GFP_KERNEL, "%s_master", data[i].name); + if (!node->name) { + ret = -ENOMEM; + goto err; + } + node->data = &qp->clocks[i]; icc_node_add(node, provider); /* link to the next node, slave */ @@ -129,6 +134,11 @@ struct icc_provider *icc_clk_register(struct device *dev, } node->name = devm_kasprintf(dev, GFP_KERNEL, "%s_slave", data[i].name); + if (!node->name) { + ret = -ENOMEM; + goto err; + } + /* no data for slave node */ icc_node_add(node, provider); onecell->nodes[j++] = node; diff --git a/drivers/interconnect/qcom/icc-rpm.c b/drivers/interconnect/qcom/icc-rpm.c index a8ed435f696c..ea1042d38128 100644 --- a/drivers/interconnect/qcom/icc-rpm.c +++ b/drivers/interconnect/qcom/icc-rpm.c @@ -503,6 +503,7 @@ int qnoc_probe(struct platform_device *pdev) GFP_KERNEL); if (!data) return -ENOMEM; + data->num_nodes = num_nodes; qp->num_intf_clks = cd_num; for (i = 0; i < cd_num; i++) @@ -597,7 +598,6 @@ regmap_done: data->nodes[i] = node; } - data->num_nodes = num_nodes; clk_bulk_disable_unprepare(qp->num_intf_clks, qp->intf_clks); diff --git a/drivers/irqchip/irq-gic-v3-its.c b/drivers/irqchip/irq-gic-v3-its.c index 92244cfa0464..8c3ec5734f1e 100644 --- a/drivers/irqchip/irq-gic-v3-its.c +++ b/drivers/irqchip/irq-gic-v3-its.c @@ -2045,7 +2045,7 @@ static int its_irq_set_vcpu_affinity(struct irq_data *d, void *vcpu_info) if (!is_v4(its_dev->its)) return -EINVAL; - guard(raw_spinlock_irq)(&its_dev->event_map.vlpi_lock); + guard(raw_spinlock)(&its_dev->event_map.vlpi_lock); /* Unmap request? */ if (!info) diff --git a/drivers/irqchip/irq-gic-v3.c b/drivers/irqchip/irq-gic-v3.c index 79d8cc80693c..76dce0aac246 100644 --- a/drivers/irqchip/irq-gic-v3.c +++ b/drivers/irqchip/irq-gic-v3.c @@ -1522,7 +1522,7 @@ static int gic_retrigger(struct irq_data *data) static int gic_cpu_pm_notifier(struct notifier_block *self, unsigned long cmd, void *v) { - if (cmd == CPU_PM_EXIT) { + if (cmd == CPU_PM_EXIT || cmd == CPU_PM_ENTER_FAILED) { if (gic_dist_security_disabled()) gic_enable_redist(true); gic_cpu_sys_reg_enable(); diff --git a/drivers/irqchip/irq-sunxi-nmi.c b/drivers/irqchip/irq-sunxi-nmi.c index bb92fd85e975..0b4312152024 100644 --- a/drivers/irqchip/irq-sunxi-nmi.c +++ b/drivers/irqchip/irq-sunxi-nmi.c @@ -186,7 +186,8 @@ static int __init sunxi_sc_nmi_irq_init(struct device_node *node, gc->chip_types[0].chip.irq_unmask = irq_gc_mask_set_bit; gc->chip_types[0].chip.irq_eoi = irq_gc_ack_set_bit; gc->chip_types[0].chip.irq_set_type = sunxi_sc_nmi_set_type; - gc->chip_types[0].chip.flags = IRQCHIP_EOI_THREADED | IRQCHIP_EOI_IF_HANDLED; + gc->chip_types[0].chip.flags = IRQCHIP_EOI_THREADED | IRQCHIP_EOI_IF_HANDLED | + IRQCHIP_SKIP_SET_WAKE; gc->chip_types[0].regs.ack = reg_offs->pend; gc->chip_types[0].regs.mask = reg_offs->enable; gc->chip_types[0].regs.type = reg_offs->ctrl; diff --git a/drivers/irqchip/irqchip.c b/drivers/irqchip/irqchip.c index 1eeb0d0156ce..0ee7b6b71f5f 100644 --- a/drivers/irqchip/irqchip.c +++ b/drivers/irqchip/irqchip.c @@ -35,11 +35,10 @@ void __init irqchip_init(void) int platform_irqchip_probe(struct platform_device *pdev) { struct device_node *np = pdev->dev.of_node; - struct device_node *par_np = of_irq_find_parent(np); + struct device_node *par_np __free(device_node) = of_irq_find_parent(np); of_irq_init_cb_t irq_init_cb = of_device_get_match_data(&pdev->dev); if (!irq_init_cb) { - of_node_put(par_np); return -EINVAL; } @@ -55,7 +54,6 @@ int platform_irqchip_probe(struct platform_device *pdev) * interrupt controller can check for specific domains as necessary. */ if (par_np && !irq_find_matching_host(par_np, DOMAIN_BUS_ANY)) { - of_node_put(par_np); return -EPROBE_DEFER; } diff --git a/drivers/md/dm-ebs-target.c b/drivers/md/dm-ebs-target.c index ec5db1478b2f..18ae45dcbfb2 100644 --- a/drivers/md/dm-ebs-target.c +++ b/drivers/md/dm-ebs-target.c @@ -442,7 +442,7 @@ static int ebs_iterate_devices(struct dm_target *ti, static struct target_type ebs_target = { .name = "ebs", .version = {1, 0, 1}, - .features = DM_TARGET_PASSES_INTEGRITY, + .features = 0, .module = THIS_MODULE, .ctr = ebs_ctr, .dtr = ebs_dtr, diff --git a/drivers/md/dm-thin.c b/drivers/md/dm-thin.c index bf0f9dddd146..05cf4e3f2bbe 100644 --- a/drivers/md/dm-thin.c +++ b/drivers/md/dm-thin.c @@ -2332,10 +2332,9 @@ static struct thin_c *get_first_thin(struct pool *pool) struct thin_c *tc = NULL; rcu_read_lock(); - if (!list_empty(&pool->active_thins)) { - tc = list_entry_rcu(pool->active_thins.next, struct thin_c, list); + tc = list_first_or_null_rcu(&pool->active_thins, struct thin_c, list); + if (tc) thin_get(tc); - } rcu_read_unlock(); return tc; diff --git a/drivers/md/dm-verity-fec.c b/drivers/md/dm-verity-fec.c index b0ee199009fc..0c41949db784 100644 --- a/drivers/md/dm-verity-fec.c +++ b/drivers/md/dm-verity-fec.c @@ -40,35 +40,23 @@ static inline u64 fec_interleave(struct dm_verity *v, u64 offset) } /* - * Decode an RS block using Reed-Solomon. - */ -static int fec_decode_rs8(struct dm_verity *v, struct dm_verity_fec_io *fio, - u8 *data, u8 *fec, int neras) -{ - int i; - uint16_t par[DM_VERITY_FEC_RSM - DM_VERITY_FEC_MIN_RSN]; - - for (i = 0; i < v->fec->roots; i++) - par[i] = fec[i]; - - return decode_rs8(fio->rs, data, par, v->fec->rsn, NULL, neras, - fio->erasures, 0, NULL); -} - -/* * Read error-correcting codes for the requested RS block. Returns a pointer * to the data block. Caller is responsible for releasing buf. */ static u8 *fec_read_parity(struct dm_verity *v, u64 rsb, int index, - unsigned int *offset, struct dm_buffer **buf, - unsigned short ioprio) + unsigned int *offset, unsigned int par_buf_offset, + struct dm_buffer **buf, unsigned short ioprio) { u64 position, block, rem; u8 *res; + /* We have already part of parity bytes read, skip to the next block */ + if (par_buf_offset) + index++; + position = (index + rsb) * v->fec->roots; block = div64_u64_rem(position, v->fec->io_size, &rem); - *offset = (unsigned int)rem; + *offset = par_buf_offset ? 0 : (unsigned int)rem; res = dm_bufio_read_with_ioprio(v->fec->bufio, block, buf, ioprio); if (IS_ERR(res)) { @@ -128,11 +116,13 @@ static int fec_decode_bufs(struct dm_verity *v, struct dm_verity_io *io, { int r, corrected = 0, res; struct dm_buffer *buf; - unsigned int n, i, offset; + unsigned int n, i, j, offset, par_buf_offset = 0; + uint16_t par_buf[DM_VERITY_FEC_RSM - DM_VERITY_FEC_MIN_RSN]; u8 *par, *block; struct bio *bio = dm_bio_from_per_bio_data(io, v->ti->per_io_data_size); - par = fec_read_parity(v, rsb, block_offset, &offset, &buf, bio->bi_ioprio); + par = fec_read_parity(v, rsb, block_offset, &offset, + par_buf_offset, &buf, bio->bi_ioprio); if (IS_ERR(par)) return PTR_ERR(par); @@ -142,7 +132,11 @@ static int fec_decode_bufs(struct dm_verity *v, struct dm_verity_io *io, */ fec_for_each_buffer_rs_block(fio, n, i) { block = fec_buffer_rs_block(v, fio, n, i); - res = fec_decode_rs8(v, fio, block, &par[offset], neras); + for (j = 0; j < v->fec->roots - par_buf_offset; j++) + par_buf[par_buf_offset + j] = par[offset + j]; + /* Decode an RS block using Reed-Solomon */ + res = decode_rs8(fio->rs, block, par_buf, v->fec->rsn, + NULL, neras, fio->erasures, 0, NULL); if (res < 0) { r = res; goto error; @@ -155,12 +149,22 @@ static int fec_decode_bufs(struct dm_verity *v, struct dm_verity_io *io, if (block_offset >= 1 << v->data_dev_block_bits) goto done; - /* read the next block when we run out of parity bytes */ - offset += v->fec->roots; + /* Read the next block when we run out of parity bytes */ + offset += (v->fec->roots - par_buf_offset); + /* Check if parity bytes are split between blocks */ + if (offset < v->fec->io_size && (offset + v->fec->roots) > v->fec->io_size) { + par_buf_offset = v->fec->io_size - offset; + for (j = 0; j < par_buf_offset; j++) + par_buf[j] = par[offset + j]; + offset += par_buf_offset; + } else + par_buf_offset = 0; + if (offset >= v->fec->io_size) { dm_bufio_release(buf); - par = fec_read_parity(v, rsb, block_offset, &offset, &buf, bio->bi_ioprio); + par = fec_read_parity(v, rsb, block_offset, &offset, + par_buf_offset, &buf, bio->bi_ioprio); if (IS_ERR(par)) return PTR_ERR(par); } @@ -724,10 +728,7 @@ int verity_fec_ctr(struct dm_verity *v) return -E2BIG; } - if ((f->roots << SECTOR_SHIFT) & ((1 << v->data_dev_block_bits) - 1)) - f->io_size = 1 << v->data_dev_block_bits; - else - f->io_size = v->fec->roots << SECTOR_SHIFT; + f->io_size = 1 << v->data_dev_block_bits; f->bufio = dm_bufio_client_create(f->dev->bdev, f->io_size, diff --git a/drivers/md/persistent-data/dm-array.c b/drivers/md/persistent-data/dm-array.c index 157c9bd2fed7..8f8792e55806 100644 --- a/drivers/md/persistent-data/dm-array.c +++ b/drivers/md/persistent-data/dm-array.c @@ -917,23 +917,27 @@ static int load_ablock(struct dm_array_cursor *c) if (c->block) unlock_ablock(c->info, c->block); - c->block = NULL; - c->ab = NULL; c->index = 0; r = dm_btree_cursor_get_value(&c->cursor, &key, &value_le); if (r) { DMERR("dm_btree_cursor_get_value failed"); - dm_btree_cursor_end(&c->cursor); + goto out; } else { r = get_ablock(c->info, le64_to_cpu(value_le), &c->block, &c->ab); if (r) { DMERR("get_ablock failed"); - dm_btree_cursor_end(&c->cursor); + goto out; } } + return 0; + +out: + dm_btree_cursor_end(&c->cursor); + c->block = NULL; + c->ab = NULL; return r; } @@ -956,10 +960,10 @@ EXPORT_SYMBOL_GPL(dm_array_cursor_begin); void dm_array_cursor_end(struct dm_array_cursor *c) { - if (c->block) { + if (c->block) unlock_ablock(c->info, c->block); - dm_btree_cursor_end(&c->cursor); - } + + dm_btree_cursor_end(&c->cursor); } EXPORT_SYMBOL_GPL(dm_array_cursor_end); @@ -999,6 +1003,7 @@ int dm_array_cursor_skip(struct dm_array_cursor *c, uint32_t count) } count -= remaining; + c->index += (remaining - 1); r = dm_array_cursor_next(c); } while (!r); diff --git a/drivers/misc/mchp_pci1xxxx/mchp_pci1xxxx_gpio.c b/drivers/misc/mchp_pci1xxxx/mchp_pci1xxxx_gpio.c index e616e3ec2b42..3c1359d8d4e6 100644 --- a/drivers/misc/mchp_pci1xxxx/mchp_pci1xxxx_gpio.c +++ b/drivers/misc/mchp_pci1xxxx/mchp_pci1xxxx_gpio.c @@ -148,7 +148,7 @@ static int pci1xxxx_gpio_set_config(struct gpio_chip *gpio, unsigned int offset, pci1xxx_assign_bit(priv->reg_base, OPENDRAIN_OFFSET(offset), (offset % 32), true); break; default: - ret = -EOPNOTSUPP; + ret = -ENOTSUPP; break; } spin_unlock_irqrestore(&priv->lock, flags); @@ -277,7 +277,7 @@ static irqreturn_t pci1xxxx_gpio_irq_handler(int irq, void *dev_id) writel(BIT(bit), priv->reg_base + INTR_STATUS_OFFSET(gpiobank)); spin_unlock_irqrestore(&priv->lock, flags); irq = irq_find_mapping(gc->irq.domain, (bit + (gpiobank * 32))); - generic_handle_irq(irq); + handle_nested_irq(irq); } } spin_lock_irqsave(&priv->lock, flags); diff --git a/drivers/mmc/host/sdhci-msm.c b/drivers/mmc/host/sdhci-msm.c index e00208535bd1..319f0ebbe652 100644 --- a/drivers/mmc/host/sdhci-msm.c +++ b/drivers/mmc/host/sdhci-msm.c @@ -1867,20 +1867,20 @@ static int sdhci_msm_program_key(struct cqhci_host *cq_host, struct sdhci_msm_host *msm_host = sdhci_pltfm_priv(pltfm_host); union cqhci_crypto_cap_entry cap; + if (!(cfg->config_enable & CQHCI_CRYPTO_CONFIGURATION_ENABLE)) + return qcom_ice_evict_key(msm_host->ice, slot); + /* Only AES-256-XTS has been tested so far. */ cap = cq_host->crypto_cap_array[cfg->crypto_cap_idx]; if (cap.algorithm_id != CQHCI_CRYPTO_ALG_AES_XTS || cap.key_size != CQHCI_CRYPTO_KEY_SIZE_256) return -EINVAL; - if (cfg->config_enable & CQHCI_CRYPTO_CONFIGURATION_ENABLE) - return qcom_ice_program_key(msm_host->ice, - QCOM_ICE_CRYPTO_ALG_AES_XTS, - QCOM_ICE_CRYPTO_KEY_SIZE_256, - cfg->crypto_key, - cfg->data_unit_size, slot); - else - return qcom_ice_evict_key(msm_host->ice, slot); + return qcom_ice_program_key(msm_host->ice, + QCOM_ICE_CRYPTO_ALG_AES_XTS, + QCOM_ICE_CRYPTO_KEY_SIZE_256, + cfg->crypto_key, + cfg->data_unit_size, slot); } #else /* CONFIG_MMC_CRYPTO */ diff --git a/drivers/mtd/nand/raw/arasan-nand-controller.c b/drivers/mtd/nand/raw/arasan-nand-controller.c index db42aa0c7b6b..865754737f5f 100644 --- a/drivers/mtd/nand/raw/arasan-nand-controller.c +++ b/drivers/mtd/nand/raw/arasan-nand-controller.c @@ -1409,8 +1409,8 @@ static int anfc_parse_cs(struct arasan_nfc *nfc) * case, the "not" chosen CS is assigned to nfc->spare_cs and selected * whenever a GPIO CS must be asserted. */ - if (nfc->cs_array && nfc->ncs > 2) { - if (!nfc->cs_array[0] && !nfc->cs_array[1]) { + if (nfc->cs_array) { + if (nfc->ncs > 2 && !nfc->cs_array[0] && !nfc->cs_array[1]) { dev_err(nfc->dev, "Assign a single native CS when using GPIOs\n"); return -EINVAL; @@ -1478,8 +1478,15 @@ static int anfc_probe(struct platform_device *pdev) static void anfc_remove(struct platform_device *pdev) { + int i; struct arasan_nfc *nfc = platform_get_drvdata(pdev); + for (i = 0; i < nfc->ncs; i++) { + if (nfc->cs_array[i]) { + gpiod_put(nfc->cs_array[i]); + } + } + anfc_chips_cleanup(nfc); } diff --git a/drivers/mtd/nand/raw/atmel/pmecc.c b/drivers/mtd/nand/raw/atmel/pmecc.c index a22aab4ed4e8..3c7dee1be21d 100644 --- a/drivers/mtd/nand/raw/atmel/pmecc.c +++ b/drivers/mtd/nand/raw/atmel/pmecc.c @@ -380,10 +380,8 @@ atmel_pmecc_create_user(struct atmel_pmecc *pmecc, user->delta = user->dmu + req->ecc.strength + 1; gf_tables = atmel_pmecc_get_gf_tables(req); - if (IS_ERR(gf_tables)) { - kfree(user); + if (IS_ERR(gf_tables)) return ERR_CAST(gf_tables); - } user->gf_tables = gf_tables; diff --git a/drivers/mtd/nand/raw/diskonchip.c b/drivers/mtd/nand/raw/diskonchip.c index 8db7fc424571..70d6c2250f32 100644 --- a/drivers/mtd/nand/raw/diskonchip.c +++ b/drivers/mtd/nand/raw/diskonchip.c @@ -1098,7 +1098,7 @@ static inline int __init inftl_partscan(struct mtd_info *mtd, struct mtd_partiti (i == 0) && (ip->firstUnit > 0)) { parts[0].name = " DiskOnChip IPL / Media Header partition"; parts[0].offset = 0; - parts[0].size = mtd->erasesize * ip->firstUnit; + parts[0].size = (uint64_t)mtd->erasesize * ip->firstUnit; numparts = 1; } diff --git a/drivers/mtd/nand/raw/omap2.c b/drivers/mtd/nand/raw/omap2.c index d9141f3c0dd1..b8af3a3533fc 100644 --- a/drivers/mtd/nand/raw/omap2.c +++ b/drivers/mtd/nand/raw/omap2.c @@ -254,6 +254,10 @@ static int omap_prefetch_reset(int cs, struct omap_nand_info *info) /** * omap_nand_data_in_pref - NAND data in using prefetch engine + * @chip: NAND chip + * @buf: output buffer where NAND data is placed into + * @len: length of transfer + * @force_8bit: force 8-bit transfers */ static void omap_nand_data_in_pref(struct nand_chip *chip, void *buf, unsigned int len, bool force_8bit) @@ -297,6 +301,10 @@ static void omap_nand_data_in_pref(struct nand_chip *chip, void *buf, /** * omap_nand_data_out_pref - NAND data out using Write Posting engine + * @chip: NAND chip + * @buf: input buffer that is sent to NAND + * @len: length of transfer + * @force_8bit: force 8-bit transfers */ static void omap_nand_data_out_pref(struct nand_chip *chip, const void *buf, unsigned int len, @@ -440,6 +448,10 @@ out_copy: /** * omap_nand_data_in_dma_pref - NAND data in using DMA and Prefetch + * @chip: NAND chip + * @buf: output buffer where NAND data is placed into + * @len: length of transfer + * @force_8bit: force 8-bit transfers */ static void omap_nand_data_in_dma_pref(struct nand_chip *chip, void *buf, unsigned int len, bool force_8bit) @@ -460,6 +472,10 @@ static void omap_nand_data_in_dma_pref(struct nand_chip *chip, void *buf, /** * omap_nand_data_out_dma_pref - NAND data out using DMA and write posting + * @chip: NAND chip + * @buf: input buffer that is sent to NAND + * @len: length of transfer + * @force_8bit: force 8-bit transfers */ static void omap_nand_data_out_dma_pref(struct nand_chip *chip, const void *buf, unsigned int len, diff --git a/drivers/mtd/spi-nor/core.c b/drivers/mtd/spi-nor/core.c index 66949d9f0cc5..b6f374ded390 100644 --- a/drivers/mtd/spi-nor/core.c +++ b/drivers/mtd/spi-nor/core.c @@ -89,7 +89,7 @@ void spi_nor_spimem_setup_op(const struct spi_nor *nor, op->addr.buswidth = spi_nor_get_protocol_addr_nbits(proto); if (op->dummy.nbytes) - op->dummy.buswidth = spi_nor_get_protocol_data_nbits(proto); + op->dummy.buswidth = spi_nor_get_protocol_addr_nbits(proto); if (op->data.nbytes) op->data.buswidth = spi_nor_get_protocol_data_nbits(proto); diff --git a/drivers/net/dsa/microchip/ksz9477.c b/drivers/net/dsa/microchip/ksz9477.c index d16817e0476f..29fe79ea74cd 100644 --- a/drivers/net/dsa/microchip/ksz9477.c +++ b/drivers/net/dsa/microchip/ksz9477.c @@ -2,7 +2,7 @@ /* * Microchip KSZ9477 switch driver main logic * - * Copyright (C) 2017-2019 Microchip Technology Inc. + * Copyright (C) 2017-2024 Microchip Technology Inc. */ #include <linux/kernel.h> @@ -983,26 +983,51 @@ void ksz9477_get_caps(struct ksz_device *dev, int port, int ksz9477_set_ageing_time(struct ksz_device *dev, unsigned int msecs) { u32 secs = msecs / 1000; - u8 value; - u8 data; + u8 data, mult, value; + u32 max_val; int ret; - value = FIELD_GET(SW_AGE_PERIOD_7_0_M, secs); +#define MAX_TIMER_VAL ((1 << 8) - 1) - ret = ksz_write8(dev, REG_SW_LUE_CTRL_3, value); - if (ret < 0) - return ret; + /* The aging timer comprises a 3-bit multiplier and an 8-bit second + * value. Either of them cannot be zero. The maximum timer is then + * 7 * 255 = 1785 seconds. + */ + if (!secs) + secs = 1; - data = FIELD_GET(SW_AGE_PERIOD_10_8_M, secs); + /* Return error if too large. */ + else if (secs > 7 * MAX_TIMER_VAL) + return -EINVAL; ret = ksz_read8(dev, REG_SW_LUE_CTRL_0, &value); if (ret < 0) return ret; - value &= ~SW_AGE_CNT_M; - value |= FIELD_PREP(SW_AGE_CNT_M, data); + /* Check whether there is need to update the multiplier. */ + mult = FIELD_GET(SW_AGE_CNT_M, value); + max_val = MAX_TIMER_VAL; + if (mult > 0) { + /* Try to use the same multiplier already in the register as + * the hardware default uses multiplier 4 and 75 seconds for + * 300 seconds. + */ + max_val = DIV_ROUND_UP(secs, mult); + if (max_val > MAX_TIMER_VAL || max_val * mult != secs) + max_val = MAX_TIMER_VAL; + } + + data = DIV_ROUND_UP(secs, max_val); + if (mult != data) { + value &= ~SW_AGE_CNT_M; + value |= FIELD_PREP(SW_AGE_CNT_M, data); + ret = ksz_write8(dev, REG_SW_LUE_CTRL_0, value); + if (ret < 0) + return ret; + } - return ksz_write8(dev, REG_SW_LUE_CTRL_0, value); + value = DIV_ROUND_UP(secs, data); + return ksz_write8(dev, REG_SW_LUE_CTRL_3, value); } void ksz9477_port_queue_split(struct ksz_device *dev, int port) diff --git a/drivers/net/dsa/microchip/ksz9477_reg.h b/drivers/net/dsa/microchip/ksz9477_reg.h index 04235c22bf40..ff579920078e 100644 --- a/drivers/net/dsa/microchip/ksz9477_reg.h +++ b/drivers/net/dsa/microchip/ksz9477_reg.h @@ -2,7 +2,7 @@ /* * Microchip KSZ9477 register definitions * - * Copyright (C) 2017-2018 Microchip Technology Inc. + * Copyright (C) 2017-2024 Microchip Technology Inc. */ #ifndef __KSZ9477_REGS_H @@ -165,8 +165,6 @@ #define SW_VLAN_ENABLE BIT(7) #define SW_DROP_INVALID_VID BIT(6) #define SW_AGE_CNT_M GENMASK(5, 3) -#define SW_AGE_CNT_S 3 -#define SW_AGE_PERIOD_10_8_M GENMASK(10, 8) #define SW_RESV_MCAST_ENABLE BIT(2) #define SW_HASH_OPTION_M 0x03 #define SW_HASH_OPTION_CRC 1 diff --git a/drivers/net/dsa/microchip/lan937x_main.c b/drivers/net/dsa/microchip/lan937x_main.c index b7652efd632e..b1ae3b9de3d1 100644 --- a/drivers/net/dsa/microchip/lan937x_main.c +++ b/drivers/net/dsa/microchip/lan937x_main.c @@ -1,6 +1,6 @@ // SPDX-License-Identifier: GPL-2.0 /* Microchip LAN937X switch driver main logic - * Copyright (C) 2019-2022 Microchip Technology Inc. + * Copyright (C) 2019-2024 Microchip Technology Inc. */ #include <linux/kernel.h> #include <linux/module.h> @@ -461,10 +461,66 @@ int lan937x_change_mtu(struct ksz_device *dev, int port, int new_mtu) int lan937x_set_ageing_time(struct ksz_device *dev, unsigned int msecs) { - u32 secs = msecs / 1000; - u32 value; + u8 data, mult, value8; + bool in_msec = false; + u32 max_val, value; + u32 secs = msecs; int ret; +#define MAX_TIMER_VAL ((1 << 20) - 1) + + /* The aging timer comprises a 3-bit multiplier and a 20-bit second + * value. Either of them cannot be zero. The maximum timer is then + * 7 * 1048575 = 7340025 seconds. As this value is too large for + * practical use it can be interpreted as microseconds, making the + * maximum timer 7340 seconds with finer control. This allows for + * maximum 122 minutes compared to 29 minutes in KSZ9477 switch. + */ + if (msecs % 1000) + in_msec = true; + else + secs /= 1000; + if (!secs) + secs = 1; + + /* Return error if too large. */ + else if (secs > 7 * MAX_TIMER_VAL) + return -EINVAL; + + /* Configure how to interpret the number value. */ + ret = ksz_rmw8(dev, REG_SW_LUE_CTRL_2, SW_AGE_CNT_IN_MICROSEC, + in_msec ? SW_AGE_CNT_IN_MICROSEC : 0); + if (ret < 0) + return ret; + + ret = ksz_read8(dev, REG_SW_LUE_CTRL_0, &value8); + if (ret < 0) + return ret; + + /* Check whether there is need to update the multiplier. */ + mult = FIELD_GET(SW_AGE_CNT_M, value8); + max_val = MAX_TIMER_VAL; + if (mult > 0) { + /* Try to use the same multiplier already in the register as + * the hardware default uses multiplier 4 and 75 seconds for + * 300 seconds. + */ + max_val = DIV_ROUND_UP(secs, mult); + if (max_val > MAX_TIMER_VAL || max_val * mult != secs) + max_val = MAX_TIMER_VAL; + } + + data = DIV_ROUND_UP(secs, max_val); + if (mult != data) { + value8 &= ~SW_AGE_CNT_M; + value8 |= FIELD_PREP(SW_AGE_CNT_M, data); + ret = ksz_write8(dev, REG_SW_LUE_CTRL_0, value8); + if (ret < 0) + return ret; + } + + secs = DIV_ROUND_UP(secs, data); + value = FIELD_GET(SW_AGE_PERIOD_7_0_M, secs); ret = ksz_write8(dev, REG_SW_AGE_PERIOD__1, value); diff --git a/drivers/net/dsa/microchip/lan937x_reg.h b/drivers/net/dsa/microchip/lan937x_reg.h index 4ec93e421da4..72042fd64e5b 100644 --- a/drivers/net/dsa/microchip/lan937x_reg.h +++ b/drivers/net/dsa/microchip/lan937x_reg.h @@ -1,6 +1,6 @@ /* SPDX-License-Identifier: GPL-2.0 */ /* Microchip LAN937X switch register definitions - * Copyright (C) 2019-2021 Microchip Technology Inc. + * Copyright (C) 2019-2024 Microchip Technology Inc. */ #ifndef __LAN937X_REG_H #define __LAN937X_REG_H @@ -56,8 +56,7 @@ #define SW_VLAN_ENABLE BIT(7) #define SW_DROP_INVALID_VID BIT(6) -#define SW_AGE_CNT_M 0x7 -#define SW_AGE_CNT_S 3 +#define SW_AGE_CNT_M GENMASK(5, 3) #define SW_RESV_MCAST_ENABLE BIT(2) #define REG_SW_LUE_CTRL_1 0x0311 @@ -70,6 +69,10 @@ #define SW_FAST_AGING BIT(1) #define SW_LINK_AUTO_AGING BIT(0) +#define REG_SW_LUE_CTRL_2 0x0312 + +#define SW_AGE_CNT_IN_MICROSEC BIT(7) + #define REG_SW_AGE_PERIOD__1 0x0313 #define SW_AGE_PERIOD_7_0_M GENMASK(7, 0) diff --git a/drivers/net/ethernet/amd/pds_core/devlink.c b/drivers/net/ethernet/amd/pds_core/devlink.c index 2681889162a2..44971e71991f 100644 --- a/drivers/net/ethernet/amd/pds_core/devlink.c +++ b/drivers/net/ethernet/amd/pds_core/devlink.c @@ -118,7 +118,7 @@ int pdsc_dl_info_get(struct devlink *dl, struct devlink_info_req *req, if (err && err != -EIO) return err; - listlen = fw_list.num_fw_slots; + listlen = min(fw_list.num_fw_slots, ARRAY_SIZE(fw_list.fw_names)); for (i = 0; i < listlen; i++) { if (i < ARRAY_SIZE(fw_slotnames)) strscpy(buf, fw_slotnames[i], sizeof(buf)); diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-phy-v2.c b/drivers/net/ethernet/amd/xgbe/xgbe-phy-v2.c index 6a716337f48b..268399dfcf22 100644 --- a/drivers/net/ethernet/amd/xgbe/xgbe-phy-v2.c +++ b/drivers/net/ethernet/amd/xgbe/xgbe-phy-v2.c @@ -923,7 +923,6 @@ static void xgbe_phy_free_phy_device(struct xgbe_prv_data *pdata) static bool xgbe_phy_finisar_phy_quirks(struct xgbe_prv_data *pdata) { - __ETHTOOL_DECLARE_LINK_MODE_MASK(supported) = { 0, }; struct xgbe_phy_data *phy_data = pdata->phy_data; unsigned int phy_id = phy_data->phydev->phy_id; @@ -945,14 +944,7 @@ static bool xgbe_phy_finisar_phy_quirks(struct xgbe_prv_data *pdata) phy_write(phy_data->phydev, 0x04, 0x0d01); phy_write(phy_data->phydev, 0x00, 0x9140); - linkmode_set_bit_array(phy_10_100_features_array, - ARRAY_SIZE(phy_10_100_features_array), - supported); - linkmode_set_bit_array(phy_gbit_features_array, - ARRAY_SIZE(phy_gbit_features_array), - supported); - - linkmode_copy(phy_data->phydev->supported, supported); + linkmode_copy(phy_data->phydev->supported, PHY_GBIT_FEATURES); phy_support_asym_pause(phy_data->phydev); @@ -964,7 +956,6 @@ static bool xgbe_phy_finisar_phy_quirks(struct xgbe_prv_data *pdata) static bool xgbe_phy_belfuse_phy_quirks(struct xgbe_prv_data *pdata) { - __ETHTOOL_DECLARE_LINK_MODE_MASK(supported) = { 0, }; struct xgbe_phy_data *phy_data = pdata->phy_data; struct xgbe_sfp_eeprom *sfp_eeprom = &phy_data->sfp_eeprom; unsigned int phy_id = phy_data->phydev->phy_id; @@ -1028,13 +1019,7 @@ static bool xgbe_phy_belfuse_phy_quirks(struct xgbe_prv_data *pdata) reg = phy_read(phy_data->phydev, 0x00); phy_write(phy_data->phydev, 0x00, reg & ~0x00800); - linkmode_set_bit_array(phy_10_100_features_array, - ARRAY_SIZE(phy_10_100_features_array), - supported); - linkmode_set_bit_array(phy_gbit_features_array, - ARRAY_SIZE(phy_gbit_features_array), - supported); - linkmode_copy(phy_data->phydev->supported, supported); + linkmode_copy(phy_data->phydev->supported, PHY_GBIT_FEATURES); phy_support_asym_pause(phy_data->phydev); netif_dbg(pdata, drv, pdata->netdev, diff --git a/drivers/net/ethernet/broadcom/bcmsysport.c b/drivers/net/ethernet/broadcom/bcmsysport.c index 42672c63f108..bc4e1f3b3752 100644 --- a/drivers/net/ethernet/broadcom/bcmsysport.c +++ b/drivers/net/ethernet/broadcom/bcmsysport.c @@ -1933,7 +1933,11 @@ static int bcm_sysport_open(struct net_device *dev) unsigned int i; int ret; - clk_prepare_enable(priv->clk); + ret = clk_prepare_enable(priv->clk); + if (ret) { + netdev_err(dev, "could not enable priv clock\n"); + return ret; + } /* Reset UniMAC */ umac_reset(priv); @@ -2591,7 +2595,11 @@ static int bcm_sysport_probe(struct platform_device *pdev) goto err_deregister_notifier; } - clk_prepare_enable(priv->clk); + ret = clk_prepare_enable(priv->clk); + if (ret) { + dev_err(&pdev->dev, "could not enable priv clock\n"); + goto err_deregister_netdev; + } priv->rev = topctrl_readl(priv, REV_CNTL) & REV_MASK; dev_info(&pdev->dev, @@ -2605,6 +2613,8 @@ static int bcm_sysport_probe(struct platform_device *pdev) return 0; +err_deregister_netdev: + unregister_netdev(dev); err_deregister_notifier: unregister_netdevice_notifier(&priv->netdev_notifier); err_deregister_fixed_link: @@ -2774,7 +2784,12 @@ static int __maybe_unused bcm_sysport_resume(struct device *d) if (!netif_running(dev)) return 0; - clk_prepare_enable(priv->clk); + ret = clk_prepare_enable(priv->clk); + if (ret) { + netdev_err(dev, "could not enable priv clock\n"); + return ret; + } + if (priv->wolopts) clk_disable_unprepare(priv->wol_clk); diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index b86f980fa7ea..b6f844cac80e 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -2897,6 +2897,13 @@ static int bnxt_hwrm_handler(struct bnxt *bp, struct tx_cmp *txcmp) return 0; } +static bool bnxt_vnic_is_active(struct bnxt *bp) +{ + struct bnxt_vnic_info *vnic = &bp->vnic_info[0]; + + return vnic->fw_vnic_id != INVALID_HW_RING_ID && vnic->mru > 0; +} + static irqreturn_t bnxt_msix(int irq, void *dev_instance) { struct bnxt_napi *bnapi = dev_instance; @@ -3164,7 +3171,7 @@ static int bnxt_poll(struct napi_struct *napi, int budget) break; } } - if (bp->flags & BNXT_FLAG_DIM) { + if ((bp->flags & BNXT_FLAG_DIM) && bnxt_vnic_is_active(bp)) { struct dim_sample dim_sample = {}; dim_update_sample(cpr->event_ctr, @@ -3295,7 +3302,7 @@ static int bnxt_poll_p5(struct napi_struct *napi, int budget) poll_done: cpr_rx = &cpr->cp_ring_arr[0]; if (cpr_rx->cp_ring_type == BNXT_NQ_HDL_TYPE_RX && - (bp->flags & BNXT_FLAG_DIM)) { + (bp->flags & BNXT_FLAG_DIM) && bnxt_vnic_is_active(bp)) { struct dim_sample dim_sample = {}; dim_update_sample(cpr->event_ctr, @@ -4701,7 +4708,7 @@ void bnxt_set_ring_params(struct bnxt *bp) /* Changing allocation mode of RX rings. * TODO: Update when extending xdp_rxq_info to support allocation modes. */ -int bnxt_set_rx_skb_mode(struct bnxt *bp, bool page_mode) +static void __bnxt_set_rx_skb_mode(struct bnxt *bp, bool page_mode) { struct net_device *dev = bp->dev; @@ -4722,15 +4729,30 @@ int bnxt_set_rx_skb_mode(struct bnxt *bp, bool page_mode) bp->rx_skb_func = bnxt_rx_page_skb; } bp->rx_dir = DMA_BIDIRECTIONAL; - /* Disable LRO or GRO_HW */ - netdev_update_features(dev); } else { dev->max_mtu = bp->max_mtu; bp->flags &= ~BNXT_FLAG_RX_PAGE_MODE; bp->rx_dir = DMA_FROM_DEVICE; bp->rx_skb_func = bnxt_rx_skb; } - return 0; +} + +void bnxt_set_rx_skb_mode(struct bnxt *bp, bool page_mode) +{ + __bnxt_set_rx_skb_mode(bp, page_mode); + + if (!page_mode) { + int rx, tx; + + bnxt_get_max_rings(bp, &rx, &tx, true); + if (rx > 1) { + bp->flags &= ~BNXT_FLAG_NO_AGG_RINGS; + bp->dev->hw_features |= NETIF_F_LRO; + } + } + + /* Update LRO and GRO_HW availability */ + netdev_update_features(bp->dev); } static void bnxt_free_vnic_attributes(struct bnxt *bp) @@ -7266,6 +7288,26 @@ err_out: return rc; } +static void bnxt_cancel_dim(struct bnxt *bp) +{ + int i; + + /* DIM work is initialized in bnxt_enable_napi(). Proceed only + * if NAPI is enabled. + */ + if (!bp->bnapi || test_bit(BNXT_STATE_NAPI_DISABLED, &bp->state)) + return; + + /* Make sure NAPI sees that the VNIC is disabled */ + synchronize_net(); + for (i = 0; i < bp->rx_nr_rings; i++) { + struct bnxt_rx_ring_info *rxr = &bp->rx_ring[i]; + struct bnxt_napi *bnapi = rxr->bnapi; + + cancel_work_sync(&bnapi->cp_ring.dim.work); + } +} + static int hwrm_ring_free_send_msg(struct bnxt *bp, struct bnxt_ring_struct *ring, u32 ring_type, int cmpl_ring_id) @@ -7366,6 +7408,7 @@ static void bnxt_hwrm_ring_free(struct bnxt *bp, bool close_path) } } + bnxt_cancel_dim(bp); for (i = 0; i < bp->rx_nr_rings; i++) { bnxt_hwrm_rx_ring_free(bp, &bp->rx_ring[i], close_path); bnxt_hwrm_rx_agg_ring_free(bp, &bp->rx_ring[i], close_path); @@ -11309,8 +11352,6 @@ static void bnxt_disable_napi(struct bnxt *bp) if (bnapi->in_reset) cpr->sw_stats->rx.rx_resets++; napi_disable(&bnapi->napi); - if (bnapi->rx_ring) - cancel_work_sync(&cpr->dim.work); } } @@ -15572,8 +15613,10 @@ static int bnxt_queue_stop(struct net_device *dev, void *qmem, int idx) bnxt_hwrm_vnic_update(bp, vnic, VNIC_UPDATE_REQ_ENABLES_MRU_VALID); } - + /* Make sure NAPI sees that the VNIC is disabled */ + synchronize_net(); rxr = &bp->rx_ring[idx]; + cancel_work_sync(&rxr->bnapi->cp_ring.dim.work); bnxt_hwrm_rx_ring_free(bp, rxr, false); bnxt_hwrm_rx_agg_ring_free(bp, rxr, false); rxr->rx_next_cons = 0; @@ -16186,7 +16229,7 @@ static int bnxt_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) if (bp->max_fltr < BNXT_MAX_FLTR) bp->max_fltr = BNXT_MAX_FLTR; bnxt_init_l2_fltr_tbl(bp); - bnxt_set_rx_skb_mode(bp, false); + __bnxt_set_rx_skb_mode(bp, false); bnxt_set_tpa_flags(bp); bnxt_set_ring_params(bp); bnxt_rdma_aux_device_init(bp); diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h index 7df7a2233307..f11ed59203d9 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h @@ -2846,7 +2846,7 @@ u32 bnxt_fw_health_readl(struct bnxt *bp, int reg_idx); bool bnxt_bs_trace_avail(struct bnxt *bp, u16 type); void bnxt_set_tpa_flags(struct bnxt *bp); void bnxt_set_ring_params(struct bnxt *); -int bnxt_set_rx_skb_mode(struct bnxt *bp, bool page_mode); +void bnxt_set_rx_skb_mode(struct bnxt *bp, bool page_mode); void bnxt_insert_usr_fltr(struct bnxt *bp, struct bnxt_filter_base *fltr); void bnxt_del_one_usr_fltr(struct bnxt *bp, struct bnxt_filter_base *fltr); int bnxt_hwrm_func_drv_rgtr(struct bnxt *bp, unsigned long *bmap, diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c index b771c84cdd89..0ed26e3a28f4 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c @@ -208,7 +208,7 @@ int bnxt_send_msg(struct bnxt_en_dev *edev, rc = hwrm_req_replace(bp, req, fw_msg->msg, fw_msg->msg_len); if (rc) - return rc; + goto drop_req; hwrm_req_timeout(bp, req, fw_msg->timeout); resp = hwrm_req_hold(bp, req); @@ -220,6 +220,7 @@ int bnxt_send_msg(struct bnxt_en_dev *edev, memcpy(fw_msg->resp, resp, resp_len); } +drop_req: hwrm_req_drop(bp, req); return rc; } diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.c index f88b641533fc..dc51dce209d5 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.c @@ -422,15 +422,8 @@ static int bnxt_xdp_set(struct bnxt *bp, struct bpf_prog *prog) bnxt_set_rx_skb_mode(bp, true); xdp_features_set_redirect_target(dev, true); } else { - int rx, tx; - xdp_features_clear_redirect_target(dev); bnxt_set_rx_skb_mode(bp, false); - bnxt_get_max_rings(bp, &rx, &tx, true); - if (rx > 1) { - bp->flags &= ~BNXT_FLAG_NO_AGG_RINGS; - bp->dev->hw_features |= NETIF_F_LRO; - } } bp->tx_nr_rings_xdp = tx_xdp; bp->tx_nr_rings = bp->tx_nr_rings_per_tc * tc + tx_xdp; diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c index bc3af0054406..604dcfd49aa4 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c @@ -1799,7 +1799,10 @@ void cxgb4_remove_tid(struct tid_info *t, unsigned int chan, unsigned int tid, struct adapter *adap = container_of(t, struct adapter, tids); struct sk_buff *skb; - WARN_ON(tid_out_of_range(&adap->tids, tid)); + if (tid_out_of_range(&adap->tids, tid)) { + dev_err(adap->pdev_dev, "tid %d out of range\n", tid); + return; + } if (t->tid_tab[tid - adap->tids.tid_base]) { t->tid_tab[tid - adap->tids.tid_base] = NULL; diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c index 1b55047c0237..4566848e1d7c 100644 --- a/drivers/net/ethernet/freescale/fec_main.c +++ b/drivers/net/ethernet/freescale/fec_main.c @@ -1591,19 +1591,22 @@ static void fec_enet_tx(struct net_device *ndev, int budget) fec_enet_tx_queue(ndev, i, budget); } -static void fec_enet_update_cbd(struct fec_enet_priv_rx_q *rxq, +static int fec_enet_update_cbd(struct fec_enet_priv_rx_q *rxq, struct bufdesc *bdp, int index) { struct page *new_page; dma_addr_t phys_addr; new_page = page_pool_dev_alloc_pages(rxq->page_pool); - WARN_ON(!new_page); - rxq->rx_skb_info[index].page = new_page; + if (unlikely(!new_page)) + return -ENOMEM; + rxq->rx_skb_info[index].page = new_page; rxq->rx_skb_info[index].offset = FEC_ENET_XDP_HEADROOM; phys_addr = page_pool_get_dma_addr(new_page) + FEC_ENET_XDP_HEADROOM; bdp->cbd_bufaddr = cpu_to_fec32(phys_addr); + + return 0; } static u32 @@ -1698,6 +1701,7 @@ fec_enet_rx_queue(struct net_device *ndev, int budget, u16 queue_id) int cpu = smp_processor_id(); struct xdp_buff xdp; struct page *page; + __fec32 cbd_bufaddr; u32 sub_len = 4; #if !defined(CONFIG_M5272) @@ -1766,12 +1770,17 @@ fec_enet_rx_queue(struct net_device *ndev, int budget, u16 queue_id) index = fec_enet_get_bd_index(bdp, &rxq->bd); page = rxq->rx_skb_info[index].page; + cbd_bufaddr = bdp->cbd_bufaddr; + if (fec_enet_update_cbd(rxq, bdp, index)) { + ndev->stats.rx_dropped++; + goto rx_processing_done; + } + dma_sync_single_for_cpu(&fep->pdev->dev, - fec32_to_cpu(bdp->cbd_bufaddr), + fec32_to_cpu(cbd_bufaddr), pkt_len, DMA_FROM_DEVICE); prefetch(page_address(page)); - fec_enet_update_cbd(rxq, bdp, index); if (xdp_prog) { xdp_buff_clear_frags_flag(&xdp); diff --git a/drivers/net/ethernet/google/gve/gve.h b/drivers/net/ethernet/google/gve/gve.h index dd92949bb214..8167cc5fb0df 100644 --- a/drivers/net/ethernet/google/gve/gve.h +++ b/drivers/net/ethernet/google/gve/gve.h @@ -1140,6 +1140,7 @@ int gve_xdp_xmit_one(struct gve_priv *priv, struct gve_tx_ring *tx, void gve_xdp_tx_flush(struct gve_priv *priv, u32 xdp_qid); bool gve_tx_poll(struct gve_notify_block *block, int budget); bool gve_xdp_poll(struct gve_notify_block *block, int budget); +int gve_xsk_tx_poll(struct gve_notify_block *block, int budget); int gve_tx_alloc_rings_gqi(struct gve_priv *priv, struct gve_tx_alloc_rings_cfg *cfg); void gve_tx_free_rings_gqi(struct gve_priv *priv, diff --git a/drivers/net/ethernet/google/gve/gve_main.c b/drivers/net/ethernet/google/gve/gve_main.c index e171ca248f9a..533e659b15b3 100644 --- a/drivers/net/ethernet/google/gve/gve_main.c +++ b/drivers/net/ethernet/google/gve/gve_main.c @@ -333,6 +333,14 @@ int gve_napi_poll(struct napi_struct *napi, int budget) if (block->rx) { work_done = gve_rx_poll(block, budget); + + /* Poll XSK TX as part of RX NAPI. Setup re-poll based on max of + * TX and RX work done. + */ + if (priv->xdp_prog) + work_done = max_t(int, work_done, + gve_xsk_tx_poll(block, budget)); + reschedule |= work_done == budget; } @@ -922,11 +930,13 @@ static void gve_init_sync_stats(struct gve_priv *priv) static void gve_tx_get_curr_alloc_cfg(struct gve_priv *priv, struct gve_tx_alloc_rings_cfg *cfg) { + int num_xdp_queues = priv->xdp_prog ? priv->rx_cfg.num_queues : 0; + cfg->qcfg = &priv->tx_cfg; cfg->raw_addressing = !gve_is_qpl(priv); cfg->ring_size = priv->tx_desc_cnt; cfg->start_idx = 0; - cfg->num_rings = gve_num_tx_queues(priv); + cfg->num_rings = priv->tx_cfg.num_queues + num_xdp_queues; cfg->tx = priv->tx; } @@ -1623,8 +1633,8 @@ static int gve_xsk_pool_enable(struct net_device *dev, if (err) return err; - /* If XDP prog is not installed, return */ - if (!priv->xdp_prog) + /* If XDP prog is not installed or interface is down, return. */ + if (!priv->xdp_prog || !netif_running(dev)) return 0; rx = &priv->rx[qid]; @@ -1669,21 +1679,16 @@ static int gve_xsk_pool_disable(struct net_device *dev, if (qid >= priv->rx_cfg.num_queues) return -EINVAL; - /* If XDP prog is not installed, unmap DMA and return */ - if (!priv->xdp_prog) - goto done; - - tx_qid = gve_xdp_tx_queue_id(priv, qid); - if (!netif_running(dev)) { - priv->rx[qid].xsk_pool = NULL; - xdp_rxq_info_unreg(&priv->rx[qid].xsk_rxq); - priv->tx[tx_qid].xsk_pool = NULL; + /* If XDP prog is not installed or interface is down, unmap DMA and + * return. + */ + if (!priv->xdp_prog || !netif_running(dev)) goto done; - } napi_rx = &priv->ntfy_blocks[priv->rx[qid].ntfy_id].napi; napi_disable(napi_rx); /* make sure current rx poll is done */ + tx_qid = gve_xdp_tx_queue_id(priv, qid); napi_tx = &priv->ntfy_blocks[priv->tx[tx_qid].ntfy_id].napi; napi_disable(napi_tx); /* make sure current tx poll is done */ @@ -1709,24 +1714,20 @@ done: static int gve_xsk_wakeup(struct net_device *dev, u32 queue_id, u32 flags) { struct gve_priv *priv = netdev_priv(dev); - int tx_queue_id = gve_xdp_tx_queue_id(priv, queue_id); + struct napi_struct *napi; + + if (!gve_get_napi_enabled(priv)) + return -ENETDOWN; if (queue_id >= priv->rx_cfg.num_queues || !priv->xdp_prog) return -EINVAL; - if (flags & XDP_WAKEUP_TX) { - struct gve_tx_ring *tx = &priv->tx[tx_queue_id]; - struct napi_struct *napi = - &priv->ntfy_blocks[tx->ntfy_id].napi; - - if (!napi_if_scheduled_mark_missed(napi)) { - /* Call local_bh_enable to trigger SoftIRQ processing */ - local_bh_disable(); - napi_schedule(napi); - local_bh_enable(); - } - - tx->xdp_xsk_wakeup++; + napi = &priv->ntfy_blocks[gve_rx_idx_to_ntfy(priv, queue_id)].napi; + if (!napi_if_scheduled_mark_missed(napi)) { + /* Call local_bh_enable to trigger SoftIRQ processing */ + local_bh_disable(); + napi_schedule(napi); + local_bh_enable(); } return 0; @@ -1837,6 +1838,7 @@ int gve_adjust_queues(struct gve_priv *priv, { struct gve_tx_alloc_rings_cfg tx_alloc_cfg = {0}; struct gve_rx_alloc_rings_cfg rx_alloc_cfg = {0}; + int num_xdp_queues; int err; gve_get_curr_alloc_cfgs(priv, &tx_alloc_cfg, &rx_alloc_cfg); @@ -1847,6 +1849,10 @@ int gve_adjust_queues(struct gve_priv *priv, rx_alloc_cfg.qcfg = &new_rx_config; tx_alloc_cfg.num_rings = new_tx_config.num_queues; + /* Add dedicated XDP TX queues if enabled. */ + num_xdp_queues = priv->xdp_prog ? new_rx_config.num_queues : 0; + tx_alloc_cfg.num_rings += num_xdp_queues; + if (netif_running(priv->dev)) { err = gve_adjust_config(priv, &tx_alloc_cfg, &rx_alloc_cfg); return err; @@ -1899,6 +1905,9 @@ static void gve_turndown(struct gve_priv *priv) gve_clear_napi_enabled(priv); gve_clear_report_stats(priv); + + /* Make sure that all traffic is finished processing. */ + synchronize_net(); } static void gve_turnup(struct gve_priv *priv) @@ -2232,14 +2241,18 @@ static void gve_service_task(struct work_struct *work) static void gve_set_netdev_xdp_features(struct gve_priv *priv) { + xdp_features_t xdp_features; + if (priv->queue_format == GVE_GQI_QPL_FORMAT) { - priv->dev->xdp_features = NETDEV_XDP_ACT_BASIC; - priv->dev->xdp_features |= NETDEV_XDP_ACT_REDIRECT; - priv->dev->xdp_features |= NETDEV_XDP_ACT_NDO_XMIT; - priv->dev->xdp_features |= NETDEV_XDP_ACT_XSK_ZEROCOPY; + xdp_features = NETDEV_XDP_ACT_BASIC; + xdp_features |= NETDEV_XDP_ACT_REDIRECT; + xdp_features |= NETDEV_XDP_ACT_NDO_XMIT; + xdp_features |= NETDEV_XDP_ACT_XSK_ZEROCOPY; } else { - priv->dev->xdp_features = 0; + xdp_features = 0; } + + xdp_set_features_flag(priv->dev, xdp_features); } static int gve_init_priv(struct gve_priv *priv, bool skip_describe_device) diff --git a/drivers/net/ethernet/google/gve/gve_tx.c b/drivers/net/ethernet/google/gve/gve_tx.c index e7fb7d6d283d..4350ebd9c2bd 100644 --- a/drivers/net/ethernet/google/gve/gve_tx.c +++ b/drivers/net/ethernet/google/gve/gve_tx.c @@ -206,7 +206,10 @@ void gve_tx_stop_ring_gqi(struct gve_priv *priv, int idx) return; gve_remove_napi(priv, ntfy_idx); - gve_clean_tx_done(priv, tx, priv->tx_desc_cnt, false); + if (tx->q_num < priv->tx_cfg.num_queues) + gve_clean_tx_done(priv, tx, priv->tx_desc_cnt, false); + else + gve_clean_xdp_done(priv, tx, priv->tx_desc_cnt); netdev_tx_reset_queue(tx->netdev_txq); gve_tx_remove_from_block(priv, idx); } @@ -834,9 +837,12 @@ int gve_xdp_xmit(struct net_device *dev, int n, struct xdp_frame **frames, struct gve_tx_ring *tx; int i, err = 0, qid; - if (unlikely(flags & ~XDP_XMIT_FLAGS_MASK)) + if (unlikely(flags & ~XDP_XMIT_FLAGS_MASK) || !priv->xdp_prog) return -EINVAL; + if (!gve_get_napi_enabled(priv)) + return -ENETDOWN; + qid = gve_xdp_tx_queue_id(priv, smp_processor_id() % priv->num_xdp_queues); @@ -975,33 +981,41 @@ out: return sent; } +int gve_xsk_tx_poll(struct gve_notify_block *rx_block, int budget) +{ + struct gve_rx_ring *rx = rx_block->rx; + struct gve_priv *priv = rx->gve; + struct gve_tx_ring *tx; + int sent = 0; + + tx = &priv->tx[gve_xdp_tx_queue_id(priv, rx->q_num)]; + if (tx->xsk_pool) { + sent = gve_xsk_tx(priv, tx, budget); + + u64_stats_update_begin(&tx->statss); + tx->xdp_xsk_sent += sent; + u64_stats_update_end(&tx->statss); + if (xsk_uses_need_wakeup(tx->xsk_pool)) + xsk_set_tx_need_wakeup(tx->xsk_pool); + } + + return sent; +} + bool gve_xdp_poll(struct gve_notify_block *block, int budget) { struct gve_priv *priv = block->priv; struct gve_tx_ring *tx = block->tx; u32 nic_done; - bool repoll; u32 to_do; /* Find out how much work there is to be done */ nic_done = gve_tx_load_event_counter(priv, tx); to_do = min_t(u32, (nic_done - tx->done), budget); gve_clean_xdp_done(priv, tx, to_do); - repoll = nic_done != tx->done; - - if (tx->xsk_pool) { - int sent = gve_xsk_tx(priv, tx, budget); - - u64_stats_update_begin(&tx->statss); - tx->xdp_xsk_sent += sent; - u64_stats_update_end(&tx->statss); - repoll |= (sent == budget); - if (xsk_uses_need_wakeup(tx->xsk_pool)) - xsk_set_tx_need_wakeup(tx->xsk_pool); - } /* If we still have work we want to repoll */ - return repoll; + return nic_done != tx->done; } bool gve_tx_poll(struct gve_notify_block *block, int budget) diff --git a/drivers/net/ethernet/hisilicon/hns3/hnae3.h b/drivers/net/ethernet/hisilicon/hns3/hnae3.h index 710a8f9f2248..12ba380eb701 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hnae3.h +++ b/drivers/net/ethernet/hisilicon/hns3/hnae3.h @@ -916,9 +916,6 @@ struct hnae3_handle { u8 netdev_flags; struct dentry *hnae3_dbgfs; - /* protects concurrent contention between debugfs commands */ - struct mutex dbgfs_lock; - char **dbgfs_buf; /* Network interface message level enabled bits */ u32 msg_enable; diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.c b/drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.c index 807eb3bbb11c..9bbece25552b 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.c @@ -1260,69 +1260,55 @@ static int hns3_dbg_read_cmd(struct hns3_dbg_data *dbg_data, static ssize_t hns3_dbg_read(struct file *filp, char __user *buffer, size_t count, loff_t *ppos) { - struct hns3_dbg_data *dbg_data = filp->private_data; + char *buf = filp->private_data; + + return simple_read_from_buffer(buffer, count, ppos, buf, strlen(buf)); +} + +static int hns3_dbg_open(struct inode *inode, struct file *filp) +{ + struct hns3_dbg_data *dbg_data = inode->i_private; struct hnae3_handle *handle = dbg_data->handle; struct hns3_nic_priv *priv = handle->priv; - ssize_t size = 0; - char **save_buf; - char *read_buf; u32 index; + char *buf; int ret; + if (!test_bit(HNS3_NIC_STATE_INITED, &priv->state) || + test_bit(HNS3_NIC_STATE_RESETTING, &priv->state)) + return -EBUSY; + ret = hns3_dbg_get_cmd_index(dbg_data, &index); if (ret) return ret; - mutex_lock(&handle->dbgfs_lock); - save_buf = &handle->dbgfs_buf[index]; - - if (!test_bit(HNS3_NIC_STATE_INITED, &priv->state) || - test_bit(HNS3_NIC_STATE_RESETTING, &priv->state)) { - ret = -EBUSY; - goto out; - } - - if (*save_buf) { - read_buf = *save_buf; - } else { - read_buf = kvzalloc(hns3_dbg_cmd[index].buf_len, GFP_KERNEL); - if (!read_buf) { - ret = -ENOMEM; - goto out; - } - - /* save the buffer addr until the last read operation */ - *save_buf = read_buf; - - /* get data ready for the first time to read */ - ret = hns3_dbg_read_cmd(dbg_data, hns3_dbg_cmd[index].cmd, - read_buf, hns3_dbg_cmd[index].buf_len); - if (ret) - goto out; - } + buf = kvzalloc(hns3_dbg_cmd[index].buf_len, GFP_KERNEL); + if (!buf) + return -ENOMEM; - size = simple_read_from_buffer(buffer, count, ppos, read_buf, - strlen(read_buf)); - if (size > 0) { - mutex_unlock(&handle->dbgfs_lock); - return size; + ret = hns3_dbg_read_cmd(dbg_data, hns3_dbg_cmd[index].cmd, + buf, hns3_dbg_cmd[index].buf_len); + if (ret) { + kvfree(buf); + return ret; } -out: - /* free the buffer for the last read operation */ - if (*save_buf) { - kvfree(*save_buf); - *save_buf = NULL; - } + filp->private_data = buf; + return 0; +} - mutex_unlock(&handle->dbgfs_lock); - return ret; +static int hns3_dbg_release(struct inode *inode, struct file *filp) +{ + kvfree(filp->private_data); + filp->private_data = NULL; + return 0; } static const struct file_operations hns3_dbg_fops = { .owner = THIS_MODULE, - .open = simple_open, + .open = hns3_dbg_open, .read = hns3_dbg_read, + .release = hns3_dbg_release, }; static int hns3_dbg_bd_file_init(struct hnae3_handle *handle, u32 cmd) @@ -1379,13 +1365,6 @@ int hns3_dbg_init(struct hnae3_handle *handle) int ret; u32 i; - handle->dbgfs_buf = devm_kcalloc(&handle->pdev->dev, - ARRAY_SIZE(hns3_dbg_cmd), - sizeof(*handle->dbgfs_buf), - GFP_KERNEL); - if (!handle->dbgfs_buf) - return -ENOMEM; - hns3_dbg_dentry[HNS3_DBG_DENTRY_COMMON].dentry = debugfs_create_dir(name, hns3_dbgfs_root); handle->hnae3_dbgfs = hns3_dbg_dentry[HNS3_DBG_DENTRY_COMMON].dentry; @@ -1395,8 +1374,6 @@ int hns3_dbg_init(struct hnae3_handle *handle) debugfs_create_dir(hns3_dbg_dentry[i].name, handle->hnae3_dbgfs); - mutex_init(&handle->dbgfs_lock); - for (i = 0; i < ARRAY_SIZE(hns3_dbg_cmd); i++) { if ((hns3_dbg_cmd[i].cmd == HNAE3_DBG_CMD_TM_NODES && ae_dev->dev_version <= HNAE3_DEVICE_VERSION_V2) || @@ -1425,24 +1402,13 @@ int hns3_dbg_init(struct hnae3_handle *handle) out: debugfs_remove_recursive(handle->hnae3_dbgfs); handle->hnae3_dbgfs = NULL; - mutex_destroy(&handle->dbgfs_lock); return ret; } void hns3_dbg_uninit(struct hnae3_handle *handle) { - u32 i; - debugfs_remove_recursive(handle->hnae3_dbgfs); handle->hnae3_dbgfs = NULL; - - for (i = 0; i < ARRAY_SIZE(hns3_dbg_cmd); i++) - if (handle->dbgfs_buf[i]) { - kvfree(handle->dbgfs_buf[i]); - handle->dbgfs_buf[i] = NULL; - } - - mutex_destroy(&handle->dbgfs_lock); } void hns3_dbg_register_debugfs(const char *debugfs_dir_name) diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c index 43377a7b2426..a7e3b22f641c 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c @@ -2452,7 +2452,6 @@ static int hns3_nic_set_features(struct net_device *netdev, return ret; } - netdev->features = features; return 0; } diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c index 05942fa78b11..db7845009252 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c @@ -6,6 +6,7 @@ #include <linux/etherdevice.h> #include <linux/init.h> #include <linux/interrupt.h> +#include <linux/irq.h> #include <linux/kernel.h> #include <linux/module.h> #include <linux/netdevice.h> @@ -3574,6 +3575,17 @@ static int hclge_set_vf_link_state(struct hnae3_handle *handle, int vf, return ret; } +static void hclge_set_reset_pending(struct hclge_dev *hdev, + enum hnae3_reset_type reset_type) +{ + /* When an incorrect reset type is executed, the get_reset_level + * function generates the HNAE3_NONE_RESET flag. As a result, this + * type do not need to pending. + */ + if (reset_type != HNAE3_NONE_RESET) + set_bit(reset_type, &hdev->reset_pending); +} + static u32 hclge_check_event_cause(struct hclge_dev *hdev, u32 *clearval) { u32 cmdq_src_reg, msix_src_reg, hw_err_src_reg; @@ -3594,7 +3606,7 @@ static u32 hclge_check_event_cause(struct hclge_dev *hdev, u32 *clearval) */ if (BIT(HCLGE_VECTOR0_IMPRESET_INT_B) & msix_src_reg) { dev_info(&hdev->pdev->dev, "IMP reset interrupt\n"); - set_bit(HNAE3_IMP_RESET, &hdev->reset_pending); + hclge_set_reset_pending(hdev, HNAE3_IMP_RESET); set_bit(HCLGE_COMM_STATE_CMD_DISABLE, &hdev->hw.hw.comm_state); *clearval = BIT(HCLGE_VECTOR0_IMPRESET_INT_B); hdev->rst_stats.imp_rst_cnt++; @@ -3604,7 +3616,7 @@ static u32 hclge_check_event_cause(struct hclge_dev *hdev, u32 *clearval) if (BIT(HCLGE_VECTOR0_GLOBALRESET_INT_B) & msix_src_reg) { dev_info(&hdev->pdev->dev, "global reset interrupt\n"); set_bit(HCLGE_COMM_STATE_CMD_DISABLE, &hdev->hw.hw.comm_state); - set_bit(HNAE3_GLOBAL_RESET, &hdev->reset_pending); + hclge_set_reset_pending(hdev, HNAE3_GLOBAL_RESET); *clearval = BIT(HCLGE_VECTOR0_GLOBALRESET_INT_B); hdev->rst_stats.global_rst_cnt++; return HCLGE_VECTOR0_EVENT_RST; @@ -3759,7 +3771,7 @@ static int hclge_misc_irq_init(struct hclge_dev *hdev) snprintf(hdev->misc_vector.name, HNAE3_INT_NAME_LEN, "%s-misc-%s", HCLGE_NAME, pci_name(hdev->pdev)); ret = request_irq(hdev->misc_vector.vector_irq, hclge_misc_irq_handle, - 0, hdev->misc_vector.name, hdev); + IRQF_NO_AUTOEN, hdev->misc_vector.name, hdev); if (ret) { hclge_free_vector(hdev, 0); dev_err(&hdev->pdev->dev, "request misc irq(%d) fail\n", @@ -4052,7 +4064,7 @@ static void hclge_do_reset(struct hclge_dev *hdev) case HNAE3_FUNC_RESET: dev_info(&pdev->dev, "PF reset requested\n"); /* schedule again to check later */ - set_bit(HNAE3_FUNC_RESET, &hdev->reset_pending); + hclge_set_reset_pending(hdev, HNAE3_FUNC_RESET); hclge_reset_task_schedule(hdev); break; default: @@ -4086,6 +4098,8 @@ static enum hnae3_reset_type hclge_get_reset_level(struct hnae3_ae_dev *ae_dev, clear_bit(HNAE3_FLR_RESET, addr); } + clear_bit(HNAE3_NONE_RESET, addr); + if (hdev->reset_type != HNAE3_NONE_RESET && rst_level < hdev->reset_type) return HNAE3_NONE_RESET; @@ -4227,7 +4241,7 @@ static bool hclge_reset_err_handle(struct hclge_dev *hdev) return false; } else if (hdev->rst_stats.reset_fail_cnt < MAX_RESET_FAIL_CNT) { hdev->rst_stats.reset_fail_cnt++; - set_bit(hdev->reset_type, &hdev->reset_pending); + hclge_set_reset_pending(hdev, hdev->reset_type); dev_info(&hdev->pdev->dev, "re-schedule reset task(%u)\n", hdev->rst_stats.reset_fail_cnt); @@ -4470,8 +4484,20 @@ static void hclge_reset_event(struct pci_dev *pdev, struct hnae3_handle *handle) static void hclge_set_def_reset_request(struct hnae3_ae_dev *ae_dev, enum hnae3_reset_type rst_type) { +#define HCLGE_SUPPORT_RESET_TYPE \ + (BIT(HNAE3_FLR_RESET) | BIT(HNAE3_FUNC_RESET) | \ + BIT(HNAE3_GLOBAL_RESET) | BIT(HNAE3_IMP_RESET)) + struct hclge_dev *hdev = ae_dev->priv; + if (!(BIT(rst_type) & HCLGE_SUPPORT_RESET_TYPE)) { + /* To prevent reset triggered by hclge_reset_event */ + set_bit(HNAE3_NONE_RESET, &hdev->default_reset_request); + dev_warn(&hdev->pdev->dev, "unsupported reset type %d\n", + rst_type); + return; + } + set_bit(rst_type, &hdev->default_reset_request); } @@ -11881,9 +11907,6 @@ static int hclge_init_ae_dev(struct hnae3_ae_dev *ae_dev) hclge_init_rxd_adv_layout(hdev); - /* Enable MISC vector(vector0) */ - hclge_enable_vector(&hdev->misc_vector, true); - ret = hclge_init_wol(hdev); if (ret) dev_warn(&pdev->dev, @@ -11896,6 +11919,10 @@ static int hclge_init_ae_dev(struct hnae3_ae_dev *ae_dev) hclge_state_init(hdev); hdev->last_reset_time = jiffies; + /* Enable MISC vector(vector0) */ + enable_irq(hdev->misc_vector.vector_irq); + hclge_enable_vector(&hdev->misc_vector, true); + dev_info(&hdev->pdev->dev, "%s driver initialization finished.\n", HCLGE_DRIVER_NAME); @@ -12301,7 +12328,7 @@ static void hclge_uninit_ae_dev(struct hnae3_ae_dev *ae_dev) /* Disable MISC vector(vector0) */ hclge_enable_vector(&hdev->misc_vector, false); - synchronize_irq(hdev->misc_vector.vector_irq); + disable_irq(hdev->misc_vector.vector_irq); /* Disable all hw interrupts */ hclge_config_mac_tnl_int(hdev, false); diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_ptp.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_ptp.c index 5505caea88e9..bab16c2191b2 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_ptp.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_ptp.c @@ -58,6 +58,9 @@ bool hclge_ptp_set_tx_info(struct hnae3_handle *handle, struct sk_buff *skb) struct hclge_dev *hdev = vport->back; struct hclge_ptp *ptp = hdev->ptp; + if (!ptp) + return false; + if (!test_bit(HCLGE_PTP_FLAG_TX_EN, &ptp->flags) || test_and_set_bit(HCLGE_STATE_PTP_TX_HANDLING, &hdev->state)) { ptp->tx_skipped++; diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_regs.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_regs.c index 43c1c18fa81f..8c057192aae6 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_regs.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_regs.c @@ -510,9 +510,9 @@ out: static int hclge_fetch_pf_reg(struct hclge_dev *hdev, void *data, struct hnae3_knic_private_info *kinfo) { -#define HCLGE_RING_REG_OFFSET 0x200 #define HCLGE_RING_INT_REG_OFFSET 0x4 + struct hnae3_queue *tqp; int i, j, reg_num; int data_num_sum; u32 *reg = data; @@ -533,10 +533,11 @@ static int hclge_fetch_pf_reg(struct hclge_dev *hdev, void *data, reg_num = ARRAY_SIZE(ring_reg_addr_list); for (j = 0; j < kinfo->num_tqps; j++) { reg += hclge_reg_get_tlv(HCLGE_REG_TAG_RING, reg_num, reg); + tqp = kinfo->tqp[j]; for (i = 0; i < reg_num; i++) - *reg++ = hclge_read_dev(&hdev->hw, - ring_reg_addr_list[i] + - HCLGE_RING_REG_OFFSET * j); + *reg++ = readl_relaxed(tqp->io_base - + HCLGE_TQP_REG_OFFSET + + ring_reg_addr_list[i]); } data_num_sum += (reg_num + HCLGE_REG_TLV_SPACE) * kinfo->num_tqps; diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c index 2f6ffb88e700..163c6e59ea4c 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c @@ -1393,6 +1393,17 @@ static int hclgevf_notify_roce_client(struct hclgevf_dev *hdev, return ret; } +static void hclgevf_set_reset_pending(struct hclgevf_dev *hdev, + enum hnae3_reset_type reset_type) +{ + /* When an incorrect reset type is executed, the get_reset_level + * function generates the HNAE3_NONE_RESET flag. As a result, this + * type do not need to pending. + */ + if (reset_type != HNAE3_NONE_RESET) + set_bit(reset_type, &hdev->reset_pending); +} + static int hclgevf_reset_wait(struct hclgevf_dev *hdev) { #define HCLGEVF_RESET_WAIT_US 20000 @@ -1542,7 +1553,7 @@ static void hclgevf_reset_err_handle(struct hclgevf_dev *hdev) hdev->rst_stats.rst_fail_cnt); if (hdev->rst_stats.rst_fail_cnt < HCLGEVF_RESET_MAX_FAIL_CNT) - set_bit(hdev->reset_type, &hdev->reset_pending); + hclgevf_set_reset_pending(hdev, hdev->reset_type); if (hclgevf_is_reset_pending(hdev)) { set_bit(HCLGEVF_RESET_PENDING, &hdev->reset_state); @@ -1662,6 +1673,8 @@ static enum hnae3_reset_type hclgevf_get_reset_level(unsigned long *addr) clear_bit(HNAE3_FLR_RESET, addr); } + clear_bit(HNAE3_NONE_RESET, addr); + return rst_level; } @@ -1671,14 +1684,15 @@ static void hclgevf_reset_event(struct pci_dev *pdev, struct hnae3_ae_dev *ae_dev = pci_get_drvdata(pdev); struct hclgevf_dev *hdev = ae_dev->priv; - dev_info(&hdev->pdev->dev, "received reset request from VF enet\n"); - if (hdev->default_reset_request) hdev->reset_level = hclgevf_get_reset_level(&hdev->default_reset_request); else hdev->reset_level = HNAE3_VF_FUNC_RESET; + dev_info(&hdev->pdev->dev, "received reset request from VF enet, reset level is %d\n", + hdev->reset_level); + /* reset of this VF requested */ set_bit(HCLGEVF_RESET_REQUESTED, &hdev->reset_state); hclgevf_reset_task_schedule(hdev); @@ -1689,8 +1703,20 @@ static void hclgevf_reset_event(struct pci_dev *pdev, static void hclgevf_set_def_reset_request(struct hnae3_ae_dev *ae_dev, enum hnae3_reset_type rst_type) { +#define HCLGEVF_SUPPORT_RESET_TYPE \ + (BIT(HNAE3_VF_RESET) | BIT(HNAE3_VF_FUNC_RESET) | \ + BIT(HNAE3_VF_PF_FUNC_RESET) | BIT(HNAE3_VF_FULL_RESET) | \ + BIT(HNAE3_FLR_RESET) | BIT(HNAE3_VF_EXP_RESET)) + struct hclgevf_dev *hdev = ae_dev->priv; + if (!(BIT(rst_type) & HCLGEVF_SUPPORT_RESET_TYPE)) { + /* To prevent reset triggered by hclge_reset_event */ + set_bit(HNAE3_NONE_RESET, &hdev->default_reset_request); + dev_info(&hdev->pdev->dev, "unsupported reset type %d\n", + rst_type); + return; + } set_bit(rst_type, &hdev->default_reset_request); } @@ -1847,14 +1873,14 @@ static void hclgevf_reset_service_task(struct hclgevf_dev *hdev) */ if (hdev->reset_attempts > HCLGEVF_MAX_RESET_ATTEMPTS_CNT) { /* prepare for full reset of stack + pcie interface */ - set_bit(HNAE3_VF_FULL_RESET, &hdev->reset_pending); + hclgevf_set_reset_pending(hdev, HNAE3_VF_FULL_RESET); /* "defer" schedule the reset task again */ set_bit(HCLGEVF_RESET_PENDING, &hdev->reset_state); } else { hdev->reset_attempts++; - set_bit(hdev->reset_level, &hdev->reset_pending); + hclgevf_set_reset_pending(hdev, hdev->reset_level); set_bit(HCLGEVF_RESET_PENDING, &hdev->reset_state); } hclgevf_reset_task_schedule(hdev); @@ -1977,7 +2003,7 @@ static enum hclgevf_evt_cause hclgevf_check_evt_cause(struct hclgevf_dev *hdev, rst_ing_reg = hclgevf_read_dev(&hdev->hw, HCLGEVF_RST_ING); dev_info(&hdev->pdev->dev, "receive reset interrupt 0x%x!\n", rst_ing_reg); - set_bit(HNAE3_VF_RESET, &hdev->reset_pending); + hclgevf_set_reset_pending(hdev, HNAE3_VF_RESET); set_bit(HCLGEVF_RESET_PENDING, &hdev->reset_state); set_bit(HCLGE_COMM_STATE_CMD_DISABLE, &hdev->hw.hw.comm_state); *clearval = ~(1U << HCLGEVF_VECTOR0_RST_INT_B); @@ -2287,6 +2313,8 @@ static void hclgevf_state_init(struct hclgevf_dev *hdev) clear_bit(HCLGEVF_STATE_RST_FAIL, &hdev->state); INIT_DELAYED_WORK(&hdev->service_task, hclgevf_service_task); + /* timer needs to be initialized before misc irq */ + timer_setup(&hdev->reset_timer, hclgevf_reset_timer, 0); mutex_init(&hdev->mbx_resp.mbx_mutex); sema_init(&hdev->reset_sem, 1); @@ -2986,7 +3014,6 @@ static int hclgevf_init_hdev(struct hclgevf_dev *hdev) HCLGEVF_DRIVER_NAME); hclgevf_task_schedule(hdev, round_jiffies_relative(HZ)); - timer_setup(&hdev->reset_timer, hclgevf_reset_timer, 0); return 0; diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_regs.c b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_regs.c index 6db415d8b917..7d9d9dbc7560 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_regs.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_regs.c @@ -123,10 +123,10 @@ int hclgevf_get_regs_len(struct hnae3_handle *handle) void hclgevf_get_regs(struct hnae3_handle *handle, u32 *version, void *data) { -#define HCLGEVF_RING_REG_OFFSET 0x200 #define HCLGEVF_RING_INT_REG_OFFSET 0x4 struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle); + struct hnae3_queue *tqp; int i, j, reg_um; u32 *reg = data; @@ -147,10 +147,11 @@ void hclgevf_get_regs(struct hnae3_handle *handle, u32 *version, reg_um = ARRAY_SIZE(ring_reg_addr_list); for (j = 0; j < hdev->num_tqps; j++) { reg += hclgevf_reg_get_tlv(HCLGEVF_REG_TAG_RING, reg_um, reg); + tqp = &hdev->htqp[j].q; for (i = 0; i < reg_um; i++) - *reg++ = hclgevf_read_dev(&hdev->hw, - ring_reg_addr_list[i] + - HCLGEVF_RING_REG_OFFSET * j); + *reg++ = readl_relaxed(tqp->io_base - + HCLGEVF_TQP_REG_OFFSET + + ring_reg_addr_list[i]); } reg_um = ARRAY_SIZE(tqp_intr_reg_addr_list); diff --git a/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h b/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h index 1489a8ceec51..46f9726d9a8a 100644 --- a/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h +++ b/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h @@ -1665,6 +1665,7 @@ struct ice_aqc_get_port_options_elem { #define ICE_AQC_PORT_OPT_MAX_LANE_25G 5 #define ICE_AQC_PORT_OPT_MAX_LANE_50G 6 #define ICE_AQC_PORT_OPT_MAX_LANE_100G 7 +#define ICE_AQC_PORT_OPT_MAX_LANE_200G 8 u8 global_scid[2]; u8 phy_scid[2]; @@ -2264,6 +2265,8 @@ struct ice_aqc_get_pkg_info_resp { struct ice_aqc_get_pkg_info pkg_info[]; }; +#define ICE_AQC_GET_CGU_MAX_PHASE_ADJ GENMASK(30, 0) + /* Get CGU abilities command response data structure (indirect 0x0C61) */ struct ice_aqc_get_cgu_abilities { u8 num_inputs; diff --git a/drivers/net/ethernet/intel/ice/ice_common.c b/drivers/net/ethernet/intel/ice/ice_common.c index 496d86cbd13f..532024f34ce4 100644 --- a/drivers/net/ethernet/intel/ice/ice_common.c +++ b/drivers/net/ethernet/intel/ice/ice_common.c @@ -4096,6 +4096,57 @@ ice_aq_set_port_option(struct ice_hw *hw, u8 lport, u8 lport_valid, } /** + * ice_get_phy_lane_number - Get PHY lane number for current adapter + * @hw: pointer to the hw struct + * + * Return: PHY lane number on success, negative error code otherwise. + */ +int ice_get_phy_lane_number(struct ice_hw *hw) +{ + struct ice_aqc_get_port_options_elem *options; + unsigned int lport = 0; + unsigned int lane; + int err; + + options = kcalloc(ICE_AQC_PORT_OPT_MAX, sizeof(*options), GFP_KERNEL); + if (!options) + return -ENOMEM; + + for (lane = 0; lane < ICE_MAX_PORT_PER_PCI_DEV; lane++) { + u8 options_count = ICE_AQC_PORT_OPT_MAX; + u8 speed, active_idx, pending_idx; + bool active_valid, pending_valid; + + err = ice_aq_get_port_options(hw, options, &options_count, lane, + true, &active_idx, &active_valid, + &pending_idx, &pending_valid); + if (err) + goto err; + + if (!active_valid) + continue; + + speed = options[active_idx].max_lane_speed; + /* If we don't get speed for this lane, it's unoccupied */ + if (speed > ICE_AQC_PORT_OPT_MAX_LANE_200G) + continue; + + if (hw->pf_id == lport) { + kfree(options); + return lane; + } + + lport++; + } + + /* PHY lane not found */ + err = -ENXIO; +err: + kfree(options); + return err; +} + +/** * ice_aq_sff_eeprom * @hw: pointer to the HW struct * @lport: bits [7:0] = logical port, bit [8] = logical port valid diff --git a/drivers/net/ethernet/intel/ice/ice_common.h b/drivers/net/ethernet/intel/ice/ice_common.h index 27208a60cece..fe6f88cfd948 100644 --- a/drivers/net/ethernet/intel/ice/ice_common.h +++ b/drivers/net/ethernet/intel/ice/ice_common.h @@ -193,6 +193,7 @@ ice_aq_get_port_options(struct ice_hw *hw, int ice_aq_set_port_option(struct ice_hw *hw, u8 lport, u8 lport_valid, u8 new_option); +int ice_get_phy_lane_number(struct ice_hw *hw); int ice_aq_sff_eeprom(struct ice_hw *hw, u16 lport, u8 bus_addr, u16 mem_addr, u8 page, u8 set_page, u8 *data, u8 length, diff --git a/drivers/net/ethernet/intel/ice/ice_dpll.c b/drivers/net/ethernet/intel/ice/ice_dpll.c index d5ad6d84007c..38e151c7ea23 100644 --- a/drivers/net/ethernet/intel/ice/ice_dpll.c +++ b/drivers/net/ethernet/intel/ice/ice_dpll.c @@ -2065,6 +2065,18 @@ static int ice_dpll_init_worker(struct ice_pf *pf) } /** + * ice_dpll_phase_range_set - initialize phase adjust range helper + * @range: pointer to phase adjust range struct to be initialized + * @phase_adj: a value to be used as min(-)/max(+) boundary + */ +static void ice_dpll_phase_range_set(struct dpll_pin_phase_adjust_range *range, + u32 phase_adj) +{ + range->min = -phase_adj; + range->max = phase_adj; +} + +/** * ice_dpll_init_info_pins_generic - initializes generic pins info * @pf: board private structure * @input: if input pins initialized @@ -2105,8 +2117,8 @@ static int ice_dpll_init_info_pins_generic(struct ice_pf *pf, bool input) for (i = 0; i < pin_num; i++) { pins[i].idx = i; pins[i].prop.board_label = labels[i]; - pins[i].prop.phase_range.min = phase_adj_max; - pins[i].prop.phase_range.max = -phase_adj_max; + ice_dpll_phase_range_set(&pins[i].prop.phase_range, + phase_adj_max); pins[i].prop.capabilities = cap; pins[i].pf = pf; ret = ice_dpll_pin_state_update(pf, &pins[i], pin_type, NULL); @@ -2152,6 +2164,7 @@ ice_dpll_init_info_direct_pins(struct ice_pf *pf, struct ice_hw *hw = &pf->hw; struct ice_dpll_pin *pins; unsigned long caps; + u32 phase_adj_max; u8 freq_supp_num; bool input; @@ -2159,11 +2172,13 @@ ice_dpll_init_info_direct_pins(struct ice_pf *pf, case ICE_DPLL_PIN_TYPE_INPUT: pins = pf->dplls.inputs; num_pins = pf->dplls.num_inputs; + phase_adj_max = pf->dplls.input_phase_adj_max; input = true; break; case ICE_DPLL_PIN_TYPE_OUTPUT: pins = pf->dplls.outputs; num_pins = pf->dplls.num_outputs; + phase_adj_max = pf->dplls.output_phase_adj_max; input = false; break; default: @@ -2188,19 +2203,13 @@ ice_dpll_init_info_direct_pins(struct ice_pf *pf, return ret; caps |= (DPLL_PIN_CAPABILITIES_PRIORITY_CAN_CHANGE | DPLL_PIN_CAPABILITIES_STATE_CAN_CHANGE); - pins[i].prop.phase_range.min = - pf->dplls.input_phase_adj_max; - pins[i].prop.phase_range.max = - -pf->dplls.input_phase_adj_max; } else { - pins[i].prop.phase_range.min = - pf->dplls.output_phase_adj_max; - pins[i].prop.phase_range.max = - -pf->dplls.output_phase_adj_max; ret = ice_cgu_get_output_pin_state_caps(hw, i, &caps); if (ret) return ret; } + ice_dpll_phase_range_set(&pins[i].prop.phase_range, + phase_adj_max); pins[i].prop.capabilities = caps; ret = ice_dpll_pin_state_update(pf, &pins[i], pin_type, NULL); if (ret) @@ -2308,8 +2317,10 @@ static int ice_dpll_init_info(struct ice_pf *pf, bool cgu) dp->dpll_idx = abilities.pps_dpll_idx; d->num_inputs = abilities.num_inputs; d->num_outputs = abilities.num_outputs; - d->input_phase_adj_max = le32_to_cpu(abilities.max_in_phase_adj); - d->output_phase_adj_max = le32_to_cpu(abilities.max_out_phase_adj); + d->input_phase_adj_max = le32_to_cpu(abilities.max_in_phase_adj) & + ICE_AQC_GET_CGU_MAX_PHASE_ADJ; + d->output_phase_adj_max = le32_to_cpu(abilities.max_out_phase_adj) & + ICE_AQC_GET_CGU_MAX_PHASE_ADJ; alloc_size = sizeof(*d->inputs) * d->num_inputs; d->inputs = kzalloc(alloc_size, GFP_KERNEL); diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c index 0ab35607e5d5..89fa3d53d317 100644 --- a/drivers/net/ethernet/intel/ice/ice_main.c +++ b/drivers/net/ethernet/intel/ice/ice_main.c @@ -1144,7 +1144,7 @@ ice_link_event(struct ice_pf *pf, struct ice_port_info *pi, bool link_up, if (link_up == old_link && link_speed == old_link_speed) return 0; - ice_ptp_link_change(pf, pf->hw.pf_id, link_up); + ice_ptp_link_change(pf, link_up); if (ice_is_dcb_active(pf)) { if (test_bit(ICE_FLAG_DCB_ENA, pf->flags)) @@ -6790,7 +6790,7 @@ static int ice_up_complete(struct ice_vsi *vsi) ice_print_link_msg(vsi, true); netif_tx_start_all_queues(vsi->netdev); netif_carrier_on(vsi->netdev); - ice_ptp_link_change(pf, pf->hw.pf_id, true); + ice_ptp_link_change(pf, true); } /* Perform an initial read of the statistics registers now to @@ -7260,7 +7260,7 @@ int ice_down(struct ice_vsi *vsi) if (vsi->netdev) { vlan_err = ice_vsi_del_vlan_zero(vsi); - ice_ptp_link_change(vsi->back, vsi->back->hw.pf_id, false); + ice_ptp_link_change(vsi->back, false); netif_carrier_off(vsi->netdev); netif_tx_disable(vsi->netdev); } diff --git a/drivers/net/ethernet/intel/ice/ice_ptp.c b/drivers/net/ethernet/intel/ice/ice_ptp.c index a999fface272..efd770dfec44 100644 --- a/drivers/net/ethernet/intel/ice/ice_ptp.c +++ b/drivers/net/ethernet/intel/ice/ice_ptp.c @@ -1388,10 +1388,9 @@ ice_ptp_port_phy_restart(struct ice_ptp_port *ptp_port) /** * ice_ptp_link_change - Reconfigure PTP after link status change * @pf: Board private structure - * @port: Port for which the PHY start is set * @linkup: Link is up or down */ -void ice_ptp_link_change(struct ice_pf *pf, u8 port, bool linkup) +void ice_ptp_link_change(struct ice_pf *pf, bool linkup) { struct ice_ptp_port *ptp_port; struct ice_hw *hw = &pf->hw; @@ -1399,14 +1398,7 @@ void ice_ptp_link_change(struct ice_pf *pf, u8 port, bool linkup) if (pf->ptp.state != ICE_PTP_READY) return; - if (WARN_ON_ONCE(port >= hw->ptp.num_lports)) - return; - ptp_port = &pf->ptp.port; - if (ice_is_e825c(hw) && hw->ptp.is_2x50g_muxed_topo) - port *= 2; - if (WARN_ON_ONCE(ptp_port->port_num != port)) - return; /* Update cached link status for this port immediately */ ptp_port->link_up = linkup; @@ -3164,10 +3156,17 @@ void ice_ptp_init(struct ice_pf *pf) { struct ice_ptp *ptp = &pf->ptp; struct ice_hw *hw = &pf->hw; - int err; + int lane_num, err; ptp->state = ICE_PTP_INITIALIZING; + lane_num = ice_get_phy_lane_number(hw); + if (lane_num < 0) { + err = lane_num; + goto err_exit; + } + + ptp->port.port_num = (u8)lane_num; ice_ptp_init_hw(hw); ice_ptp_init_tx_interrupt_mode(pf); @@ -3188,10 +3187,6 @@ void ice_ptp_init(struct ice_pf *pf) if (err) goto err_exit; - ptp->port.port_num = hw->pf_id; - if (ice_is_e825c(hw) && hw->ptp.is_2x50g_muxed_topo) - ptp->port.port_num = hw->pf_id * 2; - err = ice_ptp_init_port(pf, &ptp->port); if (err) goto err_exit; diff --git a/drivers/net/ethernet/intel/ice/ice_ptp.h b/drivers/net/ethernet/intel/ice/ice_ptp.h index 824e73b677a4..c490d98fd9c6 100644 --- a/drivers/net/ethernet/intel/ice/ice_ptp.h +++ b/drivers/net/ethernet/intel/ice/ice_ptp.h @@ -310,7 +310,7 @@ void ice_ptp_prepare_for_reset(struct ice_pf *pf, enum ice_reset_req reset_type); void ice_ptp_init(struct ice_pf *pf); void ice_ptp_release(struct ice_pf *pf); -void ice_ptp_link_change(struct ice_pf *pf, u8 port, bool linkup); +void ice_ptp_link_change(struct ice_pf *pf, bool linkup); #else /* IS_ENABLED(CONFIG_PTP_1588_CLOCK) */ static inline int ice_ptp_set_ts_config(struct ice_pf *pf, struct ifreq *ifr) { @@ -358,7 +358,7 @@ static inline void ice_ptp_prepare_for_reset(struct ice_pf *pf, } static inline void ice_ptp_init(struct ice_pf *pf) { } static inline void ice_ptp_release(struct ice_pf *pf) { } -static inline void ice_ptp_link_change(struct ice_pf *pf, u8 port, bool linkup) +static inline void ice_ptp_link_change(struct ice_pf *pf, bool linkup) { } diff --git a/drivers/net/ethernet/intel/ice/ice_ptp_consts.h b/drivers/net/ethernet/intel/ice/ice_ptp_consts.h index 585ce200c60f..a8e57cf05a9c 100644 --- a/drivers/net/ethernet/intel/ice/ice_ptp_consts.h +++ b/drivers/net/ethernet/intel/ice/ice_ptp_consts.h @@ -131,7 +131,7 @@ struct ice_eth56g_mac_reg_cfg eth56g_mac_cfg[NUM_ICE_ETH56G_LNK_SPD] = { .rx_offset = { .serdes = 0xffffeb27, /* -10.42424 */ .no_fec = 0xffffcccd, /* -25.6 */ - .fc = 0xfffe0014, /* -255.96 */ + .fc = 0xfffc557b, /* -469.26 */ .sfd = 0x4a4, /* 2.32 */ .bs_ds = 0x32 /* 0.0969697 */ } @@ -761,9 +761,9 @@ const struct ice_vernier_info_e82x e822_vernier[NUM_ICE_PTP_LNK_SPD] = { /* rx_desk_rsgb_par */ 644531250, /* 644.53125 MHz Reed Solomon gearbox */ /* tx_desk_rsgb_pcs */ - 644531250, /* 644.53125 MHz Reed Solomon gearbox */ + 390625000, /* 390.625 MHz Reed Solomon gearbox */ /* rx_desk_rsgb_pcs */ - 644531250, /* 644.53125 MHz Reed Solomon gearbox */ + 390625000, /* 390.625 MHz Reed Solomon gearbox */ /* tx_fixed_delay */ 1620, /* pmd_adj_divisor */ diff --git a/drivers/net/ethernet/intel/ice/ice_ptp_hw.c b/drivers/net/ethernet/intel/ice/ice_ptp_hw.c index 518893f23372..02e84f5b1d45 100644 --- a/drivers/net/ethernet/intel/ice/ice_ptp_hw.c +++ b/drivers/net/ethernet/intel/ice/ice_ptp_hw.c @@ -901,30 +901,45 @@ static void ice_ptp_exec_tmr_cmd(struct ice_hw *hw) */ /** + * ice_ptp_get_dest_dev_e825 - get destination PHY for given port number + * @hw: pointer to the HW struct + * @port: destination port + * + * Return: destination sideband queue PHY device. + */ +static enum ice_sbq_msg_dev ice_ptp_get_dest_dev_e825(struct ice_hw *hw, + u8 port) +{ + /* On a single complex E825, PHY 0 is always destination device phy_0 + * and PHY 1 is phy_0_peer. + */ + if (port >= hw->ptp.ports_per_phy) + return eth56g_phy_1; + else + return eth56g_phy_0; +} + +/** * ice_write_phy_eth56g - Write a PHY port register * @hw: pointer to the HW struct - * @phy_idx: PHY index + * @port: destination port * @addr: PHY register address * @val: Value to write * * Return: 0 on success, other error codes when failed to write to PHY */ -static int ice_write_phy_eth56g(struct ice_hw *hw, u8 phy_idx, u32 addr, - u32 val) +static int ice_write_phy_eth56g(struct ice_hw *hw, u8 port, u32 addr, u32 val) { - struct ice_sbq_msg_input phy_msg; + struct ice_sbq_msg_input msg = { + .dest_dev = ice_ptp_get_dest_dev_e825(hw, port), + .opcode = ice_sbq_msg_wr, + .msg_addr_low = lower_16_bits(addr), + .msg_addr_high = upper_16_bits(addr), + .data = val + }; int err; - phy_msg.opcode = ice_sbq_msg_wr; - - phy_msg.msg_addr_low = lower_16_bits(addr); - phy_msg.msg_addr_high = upper_16_bits(addr); - - phy_msg.data = val; - phy_msg.dest_dev = hw->ptp.phy.eth56g.phy_addr[phy_idx]; - - err = ice_sbq_rw_reg(hw, &phy_msg, ICE_AQ_FLAG_RD); - + err = ice_sbq_rw_reg(hw, &msg, ICE_AQ_FLAG_RD); if (err) ice_debug(hw, ICE_DBG_PTP, "PTP failed to send msg to phy %d\n", err); @@ -935,41 +950,36 @@ static int ice_write_phy_eth56g(struct ice_hw *hw, u8 phy_idx, u32 addr, /** * ice_read_phy_eth56g - Read a PHY port register * @hw: pointer to the HW struct - * @phy_idx: PHY index + * @port: destination port * @addr: PHY register address * @val: Value to write * * Return: 0 on success, other error codes when failed to read from PHY */ -static int ice_read_phy_eth56g(struct ice_hw *hw, u8 phy_idx, u32 addr, - u32 *val) +static int ice_read_phy_eth56g(struct ice_hw *hw, u8 port, u32 addr, u32 *val) { - struct ice_sbq_msg_input phy_msg; + struct ice_sbq_msg_input msg = { + .dest_dev = ice_ptp_get_dest_dev_e825(hw, port), + .opcode = ice_sbq_msg_rd, + .msg_addr_low = lower_16_bits(addr), + .msg_addr_high = upper_16_bits(addr) + }; int err; - phy_msg.opcode = ice_sbq_msg_rd; - - phy_msg.msg_addr_low = lower_16_bits(addr); - phy_msg.msg_addr_high = upper_16_bits(addr); - - phy_msg.data = 0; - phy_msg.dest_dev = hw->ptp.phy.eth56g.phy_addr[phy_idx]; - - err = ice_sbq_rw_reg(hw, &phy_msg, ICE_AQ_FLAG_RD); - if (err) { + err = ice_sbq_rw_reg(hw, &msg, ICE_AQ_FLAG_RD); + if (err) ice_debug(hw, ICE_DBG_PTP, "PTP failed to send msg to phy %d\n", err); - return err; - } - - *val = phy_msg.data; + else + *val = msg.data; - return 0; + return err; } /** * ice_phy_res_address_eth56g - Calculate a PHY port register address - * @port: Port number to be written + * @hw: pointer to the HW struct + * @lane: Lane number to be written * @res_type: resource type (register/memory) * @offset: Offset from PHY port register base * @addr: The result address @@ -978,17 +988,19 @@ static int ice_read_phy_eth56g(struct ice_hw *hw, u8 phy_idx, u32 addr, * * %0 - success * * %EINVAL - invalid port number or resource type */ -static int ice_phy_res_address_eth56g(u8 port, enum eth56g_res_type res_type, - u32 offset, u32 *addr) +static int ice_phy_res_address_eth56g(struct ice_hw *hw, u8 lane, + enum eth56g_res_type res_type, + u32 offset, + u32 *addr) { - u8 lane = port % ICE_PORTS_PER_QUAD; - u8 phy = ICE_GET_QUAD_NUM(port); - if (res_type >= NUM_ETH56G_PHY_RES) return -EINVAL; - *addr = eth56g_phy_res[res_type].base[phy] + + /* Lanes 4..7 are in fact 0..3 on a second PHY */ + lane %= hw->ptp.ports_per_phy; + *addr = eth56g_phy_res[res_type].base[0] + lane * eth56g_phy_res[res_type].step + offset; + return 0; } @@ -1008,19 +1020,17 @@ static int ice_phy_res_address_eth56g(u8 port, enum eth56g_res_type res_type, static int ice_write_port_eth56g(struct ice_hw *hw, u8 port, u32 offset, u32 val, enum eth56g_res_type res_type) { - u8 phy_port = port % hw->ptp.ports_per_phy; - u8 phy_idx = port / hw->ptp.ports_per_phy; u32 addr; int err; if (port >= hw->ptp.num_lports) return -EINVAL; - err = ice_phy_res_address_eth56g(phy_port, res_type, offset, &addr); + err = ice_phy_res_address_eth56g(hw, port, res_type, offset, &addr); if (err) return err; - return ice_write_phy_eth56g(hw, phy_idx, addr, val); + return ice_write_phy_eth56g(hw, port, addr, val); } /** @@ -1039,19 +1049,17 @@ static int ice_write_port_eth56g(struct ice_hw *hw, u8 port, u32 offset, static int ice_read_port_eth56g(struct ice_hw *hw, u8 port, u32 offset, u32 *val, enum eth56g_res_type res_type) { - u8 phy_port = port % hw->ptp.ports_per_phy; - u8 phy_idx = port / hw->ptp.ports_per_phy; u32 addr; int err; if (port >= hw->ptp.num_lports) return -EINVAL; - err = ice_phy_res_address_eth56g(phy_port, res_type, offset, &addr); + err = ice_phy_res_address_eth56g(hw, port, res_type, offset, &addr); if (err) return err; - return ice_read_phy_eth56g(hw, phy_idx, addr, val); + return ice_read_phy_eth56g(hw, port, addr, val); } /** @@ -1201,6 +1209,56 @@ static int ice_write_port_mem_eth56g(struct ice_hw *hw, u8 port, u16 offset, } /** + * ice_write_quad_ptp_reg_eth56g - Write a PHY quad register + * @hw: pointer to the HW struct + * @offset: PHY register offset + * @port: Port number + * @val: Value to write + * + * Return: + * * %0 - success + * * %EIO - invalid port number or resource type + * * %other - failed to write to PHY + */ +static int ice_write_quad_ptp_reg_eth56g(struct ice_hw *hw, u8 port, + u32 offset, u32 val) +{ + u32 addr; + + if (port >= hw->ptp.num_lports) + return -EIO; + + addr = eth56g_phy_res[ETH56G_PHY_REG_PTP].base[0] + offset; + + return ice_write_phy_eth56g(hw, port, addr, val); +} + +/** + * ice_read_quad_ptp_reg_eth56g - Read a PHY quad register + * @hw: pointer to the HW struct + * @offset: PHY register offset + * @port: Port number + * @val: Value to read + * + * Return: + * * %0 - success + * * %EIO - invalid port number or resource type + * * %other - failed to read from PHY + */ +static int ice_read_quad_ptp_reg_eth56g(struct ice_hw *hw, u8 port, + u32 offset, u32 *val) +{ + u32 addr; + + if (port >= hw->ptp.num_lports) + return -EIO; + + addr = eth56g_phy_res[ETH56G_PHY_REG_PTP].base[0] + offset; + + return ice_read_phy_eth56g(hw, port, addr, val); +} + +/** * ice_is_64b_phy_reg_eth56g - Check if this is a 64bit PHY register * @low_addr: the low address to check * @high_addr: on return, contains the high address of the 64bit register @@ -1919,7 +1977,6 @@ ice_phy_get_speed_eth56g(struct ice_link_status *li) */ static int ice_phy_cfg_parpcs_eth56g(struct ice_hw *hw, u8 port) { - u8 port_blk = port & ~(ICE_PORTS_PER_QUAD - 1); u32 val; int err; @@ -1934,8 +1991,8 @@ static int ice_phy_cfg_parpcs_eth56g(struct ice_hw *hw, u8 port) switch (ice_phy_get_speed_eth56g(&hw->port_info->phy.link_info)) { case ICE_ETH56G_LNK_SPD_1G: case ICE_ETH56G_LNK_SPD_2_5G: - err = ice_read_ptp_reg_eth56g(hw, port_blk, - PHY_GPCS_CONFIG_REG0, &val); + err = ice_read_quad_ptp_reg_eth56g(hw, port, + PHY_GPCS_CONFIG_REG0, &val); if (err) { ice_debug(hw, ICE_DBG_PTP, "Failed to read PHY_GPCS_CONFIG_REG0, status: %d", err); @@ -1946,8 +2003,8 @@ static int ice_phy_cfg_parpcs_eth56g(struct ice_hw *hw, u8 port) val |= FIELD_PREP(PHY_GPCS_CONFIG_REG0_TX_THR_M, ICE_ETH56G_NOMINAL_TX_THRESH); - err = ice_write_ptp_reg_eth56g(hw, port_blk, - PHY_GPCS_CONFIG_REG0, val); + err = ice_write_quad_ptp_reg_eth56g(hw, port, + PHY_GPCS_CONFIG_REG0, val); if (err) { ice_debug(hw, ICE_DBG_PTP, "Failed to write PHY_GPCS_CONFIG_REG0, status: %d", err); @@ -1988,50 +2045,47 @@ static int ice_phy_cfg_parpcs_eth56g(struct ice_hw *hw, u8 port) */ int ice_phy_cfg_ptp_1step_eth56g(struct ice_hw *hw, u8 port) { - u8 port_blk = port & ~(ICE_PORTS_PER_QUAD - 1); - u8 blk_port = port & (ICE_PORTS_PER_QUAD - 1); + u8 quad_lane = port % ICE_PORTS_PER_QUAD; + u32 addr, val, peer_delay; bool enable, sfd_ena; - u32 val, peer_delay; int err; enable = hw->ptp.phy.eth56g.onestep_ena; peer_delay = hw->ptp.phy.eth56g.peer_delay; sfd_ena = hw->ptp.phy.eth56g.sfd_ena; - /* PHY_PTP_1STEP_CONFIG */ - err = ice_read_ptp_reg_eth56g(hw, port_blk, PHY_PTP_1STEP_CONFIG, &val); + addr = PHY_PTP_1STEP_CONFIG; + err = ice_read_quad_ptp_reg_eth56g(hw, port, addr, &val); if (err) return err; if (enable) - val |= blk_port; + val |= BIT(quad_lane); else - val &= ~blk_port; + val &= ~BIT(quad_lane); val &= ~(PHY_PTP_1STEP_T1S_UP64_M | PHY_PTP_1STEP_T1S_DELTA_M); - err = ice_write_ptp_reg_eth56g(hw, port_blk, PHY_PTP_1STEP_CONFIG, val); + err = ice_write_quad_ptp_reg_eth56g(hw, port, addr, val); if (err) return err; - /* PHY_PTP_1STEP_PEER_DELAY */ + addr = PHY_PTP_1STEP_PEER_DELAY(quad_lane); val = FIELD_PREP(PHY_PTP_1STEP_PD_DELAY_M, peer_delay); if (peer_delay) val |= PHY_PTP_1STEP_PD_ADD_PD_M; val |= PHY_PTP_1STEP_PD_DLY_V_M; - err = ice_write_ptp_reg_eth56g(hw, port_blk, - PHY_PTP_1STEP_PEER_DELAY(blk_port), val); + err = ice_write_quad_ptp_reg_eth56g(hw, port, addr, val); if (err) return err; val &= ~PHY_PTP_1STEP_PD_DLY_V_M; - err = ice_write_ptp_reg_eth56g(hw, port_blk, - PHY_PTP_1STEP_PEER_DELAY(blk_port), val); + err = ice_write_quad_ptp_reg_eth56g(hw, port, addr, val); if (err) return err; - /* PHY_MAC_XIF_MODE */ - err = ice_read_mac_reg_eth56g(hw, port, PHY_MAC_XIF_MODE, &val); + addr = PHY_MAC_XIF_MODE; + err = ice_read_mac_reg_eth56g(hw, port, addr, &val); if (err) return err; @@ -2051,7 +2105,7 @@ int ice_phy_cfg_ptp_1step_eth56g(struct ice_hw *hw, u8 port) FIELD_PREP(PHY_MAC_XIF_TS_BIN_MODE_M, enable) | FIELD_PREP(PHY_MAC_XIF_TS_SFD_ENA_M, sfd_ena); - return ice_write_mac_reg_eth56g(hw, port, PHY_MAC_XIF_MODE, val); + return ice_write_mac_reg_eth56g(hw, port, addr, val); } /** @@ -2093,21 +2147,22 @@ static u32 ice_ptp_calc_bitslip_eth56g(struct ice_hw *hw, u8 port, u32 bs, bool fc, bool rs, enum ice_eth56g_link_spd spd) { - u8 port_offset = port & (ICE_PORTS_PER_QUAD - 1); - u8 port_blk = port & ~(ICE_PORTS_PER_QUAD - 1); u32 bitslip; int err; if (!bs || rs) return 0; - if (spd == ICE_ETH56G_LNK_SPD_1G || spd == ICE_ETH56G_LNK_SPD_2_5G) + if (spd == ICE_ETH56G_LNK_SPD_1G || spd == ICE_ETH56G_LNK_SPD_2_5G) { err = ice_read_gpcs_reg_eth56g(hw, port, PHY_GPCS_BITSLIP, &bitslip); - else - err = ice_read_ptp_reg_eth56g(hw, port_blk, - PHY_REG_SD_BIT_SLIP(port_offset), - &bitslip); + } else { + u8 quad_lane = port % ICE_PORTS_PER_QUAD; + u32 addr; + + addr = PHY_REG_SD_BIT_SLIP(quad_lane); + err = ice_read_quad_ptp_reg_eth56g(hw, port, addr, &bitslip); + } if (err) return 0; @@ -2667,59 +2722,29 @@ static int ice_get_phy_tx_tstamp_ready_eth56g(struct ice_hw *hw, u8 port, } /** - * ice_is_muxed_topo - detect breakout 2x50G topology for E825C - * @hw: pointer to the HW struct - * - * Return: true if it's 2x50 breakout topology, false otherwise - */ -static bool ice_is_muxed_topo(struct ice_hw *hw) -{ - u8 link_topo; - bool mux; - u32 val; - - val = rd32(hw, GLGEN_SWITCH_MODE_CONFIG); - mux = FIELD_GET(GLGEN_SWITCH_MODE_CONFIG_25X4_QUAD_M, val); - val = rd32(hw, GLGEN_MAC_LINK_TOPO); - link_topo = FIELD_GET(GLGEN_MAC_LINK_TOPO_LINK_TOPO_M, val); - - return (mux && link_topo == ICE_LINK_TOPO_UP_TO_2_LINKS); -} - -/** - * ice_ptp_init_phy_e825c - initialize PHY parameters + * ice_ptp_init_phy_e825 - initialize PHY parameters * @hw: pointer to the HW struct */ -static void ice_ptp_init_phy_e825c(struct ice_hw *hw) +static void ice_ptp_init_phy_e825(struct ice_hw *hw) { struct ice_ptp_hw *ptp = &hw->ptp; struct ice_eth56g_params *params; - u8 phy; + u32 phy_rev; + int err; ptp->phy_model = ICE_PHY_ETH56G; params = &ptp->phy.eth56g; params->onestep_ena = false; params->peer_delay = 0; params->sfd_ena = false; - params->phy_addr[0] = eth56g_phy_0; - params->phy_addr[1] = eth56g_phy_1; params->num_phys = 2; ptp->ports_per_phy = 4; ptp->num_lports = params->num_phys * ptp->ports_per_phy; ice_sb_access_ena_eth56g(hw, true); - for (phy = 0; phy < params->num_phys; phy++) { - u32 phy_rev; - int err; - - err = ice_read_phy_eth56g(hw, phy, PHY_REG_REVISION, &phy_rev); - if (err || phy_rev != PHY_REVISION_ETH56G) { - ptp->phy_model = ICE_PHY_UNSUP; - return; - } - } - - ptp->is_2x50g_muxed_topo = ice_is_muxed_topo(hw); + err = ice_read_phy_eth56g(hw, hw->pf_id, PHY_REG_REVISION, &phy_rev); + if (err || phy_rev != PHY_REVISION_ETH56G) + ptp->phy_model = ICE_PHY_UNSUP; } /* E822 family functions @@ -2738,10 +2763,9 @@ static void ice_fill_phy_msg_e82x(struct ice_hw *hw, struct ice_sbq_msg_input *msg, u8 port, u16 offset) { - int phy_port, phy, quadtype; + int phy_port, quadtype; phy_port = port % hw->ptp.ports_per_phy; - phy = port / hw->ptp.ports_per_phy; quadtype = ICE_GET_QUAD_NUM(port) % ICE_GET_QUAD_NUM(hw->ptp.ports_per_phy); @@ -2753,12 +2777,7 @@ static void ice_fill_phy_msg_e82x(struct ice_hw *hw, msg->msg_addr_high = P_Q1_H(P_4_BASE + offset, phy_port); } - if (phy == 0) - msg->dest_dev = rmn_0; - else if (phy == 1) - msg->dest_dev = rmn_1; - else - msg->dest_dev = rmn_2; + msg->dest_dev = rmn_0; } /** @@ -5478,7 +5497,7 @@ void ice_ptp_init_hw(struct ice_hw *hw) else if (ice_is_e810(hw)) ice_ptp_init_phy_e810(ptp); else if (ice_is_e825c(hw)) - ice_ptp_init_phy_e825c(hw); + ice_ptp_init_phy_e825(hw); else ptp->phy_model = ICE_PHY_UNSUP; } diff --git a/drivers/net/ethernet/intel/ice/ice_type.h b/drivers/net/ethernet/intel/ice/ice_type.h index adb168860711..4a9ef722635f 100644 --- a/drivers/net/ethernet/intel/ice/ice_type.h +++ b/drivers/net/ethernet/intel/ice/ice_type.h @@ -850,7 +850,6 @@ struct ice_mbx_data { struct ice_eth56g_params { u8 num_phys; - u8 phy_addr[2]; bool onestep_ena; bool sfd_ena; u32 peer_delay; @@ -881,7 +880,6 @@ struct ice_ptp_hw { union ice_phy_params phy; u8 num_lports; u8 ports_per_phy; - bool is_2x50g_muxed_topo; }; /* Port hardware description */ diff --git a/drivers/net/ethernet/intel/igc/igc_base.c b/drivers/net/ethernet/intel/igc/igc_base.c index 9fae8bdec2a7..1613b562d17c 100644 --- a/drivers/net/ethernet/intel/igc/igc_base.c +++ b/drivers/net/ethernet/intel/igc/igc_base.c @@ -68,6 +68,10 @@ static s32 igc_init_nvm_params_base(struct igc_hw *hw) u32 eecd = rd32(IGC_EECD); u16 size; + /* failed to read reg and got all F's */ + if (!(~eecd)) + return -ENXIO; + size = FIELD_GET(IGC_EECD_SIZE_EX_MASK, eecd); /* Added to a constant, "size" becomes the left-shift value @@ -221,6 +225,8 @@ static s32 igc_get_invariants_base(struct igc_hw *hw) /* NVM initialization */ ret_val = igc_init_nvm_params_base(hw); + if (ret_val) + goto out; switch (hw->mac.type) { case igc_i225: ret_val = igc_init_nvm_params_i225(hw); diff --git a/drivers/net/ethernet/marvell/mv643xx_eth.c b/drivers/net/ethernet/marvell/mv643xx_eth.c index a06048719e84..67a6ff07c83d 100644 --- a/drivers/net/ethernet/marvell/mv643xx_eth.c +++ b/drivers/net/ethernet/marvell/mv643xx_eth.c @@ -2704,9 +2704,15 @@ static struct platform_device *port_platdev[3]; static void mv643xx_eth_shared_of_remove(void) { + struct mv643xx_eth_platform_data *pd; int n; for (n = 0; n < 3; n++) { + if (!port_platdev[n]) + continue; + pd = dev_get_platdata(&port_platdev[n]->dev); + if (pd) + of_node_put(pd->phy_node); platform_device_del(port_platdev[n]); port_platdev[n] = NULL; } @@ -2769,8 +2775,10 @@ static int mv643xx_eth_shared_of_add_port(struct platform_device *pdev, } ppdev = platform_device_alloc(MV643XX_ETH_NAME, dev_num); - if (!ppdev) - return -ENOMEM; + if (!ppdev) { + ret = -ENOMEM; + goto put_err; + } ppdev->dev.coherent_dma_mask = DMA_BIT_MASK(32); ppdev->dev.of_node = pnp; @@ -2792,6 +2800,8 @@ static int mv643xx_eth_shared_of_add_port(struct platform_device *pdev, port_err: platform_device_put(ppdev); +put_err: + of_node_put(ppd.phy_node); return ret; } diff --git a/drivers/net/ethernet/marvell/sky2.c b/drivers/net/ethernet/marvell/sky2.c index 3914cd9210d4..988fa28cfb5f 100644 --- a/drivers/net/ethernet/marvell/sky2.c +++ b/drivers/net/ethernet/marvell/sky2.c @@ -130,6 +130,7 @@ static const struct pci_device_id sky2_id_table[] = { { PCI_DEVICE(PCI_VENDOR_ID_MARVELL, 0x436C) }, /* 88E8072 */ { PCI_DEVICE(PCI_VENDOR_ID_MARVELL, 0x436D) }, /* 88E8055 */ { PCI_DEVICE(PCI_VENDOR_ID_MARVELL, 0x4370) }, /* 88E8075 */ + { PCI_DEVICE(PCI_VENDOR_ID_MARVELL, 0x4373) }, /* 88E8075 */ { PCI_DEVICE(PCI_VENDOR_ID_MARVELL, 0x4380) }, /* 88E8057 */ { PCI_DEVICE(PCI_VENDOR_ID_MARVELL, 0x4381) }, /* 88E8059 */ { PCI_DEVICE(PCI_VENDOR_ID_MARVELL, 0x4382) }, /* 88E8079 */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c index 6bd8a18e3af3..e733b81e18a2 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c @@ -1013,6 +1013,7 @@ static void cmd_work_handler(struct work_struct *work) complete(&ent->done); } up(&cmd->vars.sem); + complete(&ent->slotted); return; } } else { diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c index ca92e518be76..1baf8933a07c 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c @@ -724,6 +724,12 @@ static int mlx5e_xfrm_add_state(struct xfrm_state *x, /* check esn */ if (x->props.flags & XFRM_STATE_ESN) mlx5e_ipsec_update_esn_state(sa_entry); + else + /* According to RFC4303, section "3.3.3. Sequence Number Generation", + * the first packet sent using a given SA will contain a sequence + * number of 1. + */ + sa_entry->esn_state.esn = 1; mlx5e_ipsec_build_accel_xfrm_attrs(sa_entry, &sa_entry->attrs); @@ -768,9 +774,12 @@ static int mlx5e_xfrm_add_state(struct xfrm_state *x, MLX5_IPSEC_RESCHED); if (x->xso.type == XFRM_DEV_OFFLOAD_PACKET && - x->props.mode == XFRM_MODE_TUNNEL) - xa_set_mark(&ipsec->sadb, sa_entry->ipsec_obj_id, - MLX5E_IPSEC_TUNNEL_SA); + x->props.mode == XFRM_MODE_TUNNEL) { + xa_lock_bh(&ipsec->sadb); + __xa_set_mark(&ipsec->sadb, sa_entry->ipsec_obj_id, + MLX5E_IPSEC_TUNNEL_SA); + xa_unlock_bh(&ipsec->sadb); + } out: x->xso.offload_handle = (unsigned long)sa_entry; @@ -797,7 +806,6 @@ err_xfrm: static void mlx5e_xfrm_del_state(struct xfrm_state *x) { struct mlx5e_ipsec_sa_entry *sa_entry = to_ipsec_sa_entry(x); - struct mlx5_accel_esp_xfrm_attrs *attrs = &sa_entry->attrs; struct mlx5e_ipsec *ipsec = sa_entry->ipsec; struct mlx5e_ipsec_sa_entry *old; @@ -806,12 +814,6 @@ static void mlx5e_xfrm_del_state(struct xfrm_state *x) old = xa_erase_bh(&ipsec->sadb, sa_entry->ipsec_obj_id); WARN_ON(old != sa_entry); - - if (attrs->mode == XFRM_MODE_TUNNEL && - attrs->type == XFRM_DEV_OFFLOAD_PACKET) - /* Make sure that no ARP requests are running in parallel */ - flush_workqueue(ipsec->wq); - } static void mlx5e_xfrm_free_state(struct xfrm_state *x) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c index e51b03d4c717..57861d34d46f 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c @@ -1718,23 +1718,21 @@ static int tx_add_rule(struct mlx5e_ipsec_sa_entry *sa_entry) goto err_alloc; } - if (attrs->family == AF_INET) - setup_fte_addr4(spec, &attrs->saddr.a4, &attrs->daddr.a4); - else - setup_fte_addr6(spec, attrs->saddr.a6, attrs->daddr.a6); - setup_fte_no_frags(spec); setup_fte_upper_proto_match(spec, &attrs->upspec); switch (attrs->type) { case XFRM_DEV_OFFLOAD_CRYPTO: + if (attrs->family == AF_INET) + setup_fte_addr4(spec, &attrs->saddr.a4, &attrs->daddr.a4); + else + setup_fte_addr6(spec, attrs->saddr.a6, attrs->daddr.a6); setup_fte_spi(spec, attrs->spi, false); setup_fte_esp(spec); setup_fte_reg_a(spec); break; case XFRM_DEV_OFFLOAD_PACKET: - if (attrs->reqid) - setup_fte_reg_c4(spec, attrs->reqid); + setup_fte_reg_c4(spec, attrs->reqid); err = setup_pkt_reformat(ipsec, attrs, &flow_act); if (err) goto err_pkt_reformat; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c index 53cfa39188cb..820debf3fbbf 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c @@ -91,8 +91,9 @@ u32 mlx5_ipsec_device_caps(struct mlx5_core_dev *mdev) EXPORT_SYMBOL_GPL(mlx5_ipsec_device_caps); static void mlx5e_ipsec_packet_setup(void *obj, u32 pdn, - struct mlx5_accel_esp_xfrm_attrs *attrs) + struct mlx5e_ipsec_sa_entry *sa_entry) { + struct mlx5_accel_esp_xfrm_attrs *attrs = &sa_entry->attrs; void *aso_ctx; aso_ctx = MLX5_ADDR_OF(ipsec_obj, obj, ipsec_aso); @@ -120,8 +121,12 @@ static void mlx5e_ipsec_packet_setup(void *obj, u32 pdn, * active. */ MLX5_SET(ipsec_obj, obj, aso_return_reg, MLX5_IPSEC_ASO_REG_C_4_5); - if (attrs->dir == XFRM_DEV_OFFLOAD_OUT) + if (attrs->dir == XFRM_DEV_OFFLOAD_OUT) { MLX5_SET(ipsec_aso, aso_ctx, mode, MLX5_IPSEC_ASO_INC_SN); + if (!attrs->replay_esn.trigger) + MLX5_SET(ipsec_aso, aso_ctx, mode_parameter, + sa_entry->esn_state.esn); + } if (attrs->lft.hard_packet_limit != XFRM_INF) { MLX5_SET(ipsec_aso, aso_ctx, remove_flow_pkt_cnt, @@ -175,7 +180,7 @@ static int mlx5_create_ipsec_obj(struct mlx5e_ipsec_sa_entry *sa_entry) res = &mdev->mlx5e_res.hw_objs; if (attrs->type == XFRM_DEV_OFFLOAD_PACKET) - mlx5e_ipsec_packet_setup(obj, res->pdn, attrs); + mlx5e_ipsec_packet_setup(obj, res->pdn, sa_entry); err = mlx5_cmd_exec(mdev, in, sizeof(in), out, sizeof(out)); if (!err) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/macsec.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/macsec.c index cc9bcc420032..6ab02f3fc291 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/macsec.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/macsec.c @@ -339,9 +339,13 @@ static int mlx5e_macsec_init_sa_fs(struct macsec_context *ctx, { struct mlx5e_priv *priv = macsec_netdev_priv(ctx->netdev); struct mlx5_macsec_fs *macsec_fs = priv->mdev->macsec_fs; + const struct macsec_tx_sc *tx_sc = &ctx->secy->tx_sc; struct mlx5_macsec_rule_attrs rule_attrs; union mlx5_macsec_rule *macsec_rule; + if (is_tx && tx_sc->encoding_sa != sa->assoc_num) + return 0; + rule_attrs.macsec_obj_id = sa->macsec_obj_id; rule_attrs.sci = sa->sci; rule_attrs.assoc_num = sa->assoc_num; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index dd16d73000c3..0ec17c276bdd 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -6542,8 +6542,23 @@ static void _mlx5e_remove(struct auxiliary_device *adev) mlx5_core_uplink_netdev_set(mdev, NULL); mlx5e_dcbnl_delete_app(priv); - unregister_netdev(priv->netdev); - _mlx5e_suspend(adev, false); + /* When unload driver, the netdev is in registered state + * if it's from legacy mode. If from switchdev mode, it + * is already unregistered before changing to NIC profile. + */ + if (priv->netdev->reg_state == NETREG_REGISTERED) { + unregister_netdev(priv->netdev); + _mlx5e_suspend(adev, false); + } else { + struct mlx5_core_dev *pos; + int i; + + if (test_bit(MLX5E_STATE_DESTROYING, &priv->state)) + mlx5_sd_for_each_dev(i, mdev, pos) + mlx5e_destroy_mdev_resources(pos); + else + _mlx5e_suspend(adev, true); + } /* Avoid cleanup if profile rollback failed. */ if (priv->profile) priv->profile->cleanup(priv); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c index 554f9cb5b53f..fdff9fd8a89e 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c @@ -1509,6 +1509,21 @@ mlx5e_vport_uplink_rep_unload(struct mlx5e_rep_priv *rpriv) priv = netdev_priv(netdev); + /* This bit is set when using devlink to change eswitch mode from + * switchdev to legacy. As need to keep uplink netdev ifindex, we + * detach uplink representor profile and attach NIC profile only. + * The netdev will be unregistered later when unload NIC auxiliary + * driver for this case. + * We explicitly block devlink eswitch mode change if any IPSec rules + * offloaded, but can't block other cases, such as driver unload + * and devlink reload. We have to unregister netdev before profile + * change for those cases. This is to avoid resource leak because + * the offloaded rules don't have the chance to be unoffloaded before + * cleanup which is triggered by detach uplink representor profile. + */ + if (!(priv->mdev->priv.flags & MLX5_PRIV_FLAGS_SWITCH_LEGACY)) + unregister_netdev(netdev); + mlx5e_netdev_attach_nic_profile(priv); } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/ipsec_fs.c b/drivers/net/ethernet/mellanox/mlx5/core/esw/ipsec_fs.c index 5a0047bdcb51..ed977ae75fab 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/esw/ipsec_fs.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/ipsec_fs.c @@ -150,11 +150,11 @@ void mlx5_esw_ipsec_restore_dest_uplink(struct mlx5_core_dev *mdev) unsigned long i; int err; - xa_for_each(&esw->offloads.vport_reps, i, rep) { - rpriv = rep->rep_data[REP_ETH].priv; - if (!rpriv || !rpriv->netdev) + mlx5_esw_for_each_rep(esw, i, rep) { + if (atomic_read(&rep->rep_data[REP_ETH].state) != REP_LOADED) continue; + rpriv = rep->rep_data[REP_ETH].priv; rhashtable_walk_enter(&rpriv->tc_ht, &iter); rhashtable_walk_start(&iter); while ((flow = rhashtable_walk_next(&iter)) != NULL) { diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h index a83d41121db6..8573d36785f4 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h @@ -714,6 +714,9 @@ void mlx5e_tc_clean_fdb_peer_flows(struct mlx5_eswitch *esw); MLX5_CAP_GEN_2((esw->dev), ec_vf_vport_base) +\ (last) - 1) +#define mlx5_esw_for_each_rep(esw, i, rep) \ + xa_for_each(&((esw)->offloads.vport_reps), i, rep) + struct mlx5_eswitch *__must_check mlx5_devlink_eswitch_get(struct devlink *devlink); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c index d5b42b3a19fd..06076dd9ec64 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c @@ -53,9 +53,6 @@ #include "lag/lag.h" #include "en/tc/post_meter.h" -#define mlx5_esw_for_each_rep(esw, i, rep) \ - xa_for_each(&((esw)->offloads.vport_reps), i, rep) - /* There are two match-all miss flows, one for unicast dst mac and * one for multicast. */ @@ -3780,6 +3777,8 @@ int mlx5_devlink_eswitch_mode_set(struct devlink *devlink, u16 mode, esw->eswitch_operation_in_progress = true; up_write(&esw->mode_lock); + if (mode == DEVLINK_ESWITCH_MODE_LEGACY) + esw->dev->priv.flags |= MLX5_PRIV_FLAGS_SWITCH_LEGACY; mlx5_eswitch_disable_locked(esw); if (mode == DEVLINK_ESWITCH_MODE_SWITCHDEV) { if (mlx5_devlink_trap_get_num_active(esw->dev)) { diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c index 2eabfcc247c6..0ce999706d41 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c @@ -2709,6 +2709,7 @@ struct mlx5_flow_namespace *mlx5_get_flow_namespace(struct mlx5_core_dev *dev, break; case MLX5_FLOW_NAMESPACE_RDMA_TX: root_ns = steering->rdma_tx_root_ns; + prio = RDMA_TX_BYPASS_PRIO; break; case MLX5_FLOW_NAMESPACE_RDMA_RX_COUNTERS: root_ns = steering->rdma_rx_root_ns; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lag/port_sel.c b/drivers/net/ethernet/mellanox/mlx5/core/lag/port_sel.c index ab2717012b79..39e80704b1c4 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/lag/port_sel.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/lag/port_sel.c @@ -530,7 +530,7 @@ int mlx5_lag_port_sel_create(struct mlx5_lag *ldev, set_tt_map(port_sel, hash_type); err = mlx5_lag_create_definers(ldev, hash_type, ports); if (err) - return err; + goto clear_port_sel; if (port_sel->tunnel) { err = mlx5_lag_create_inner_ttc_table(ldev); @@ -549,6 +549,8 @@ destroy_inner: mlx5_destroy_ttc_table(port_sel->inner.ttc); destroy_definers: mlx5_lag_destroy_definers(ldev); +clear_port_sel: + memset(port_sel, 0, sizeof(*port_sel)); return err; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c b/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c index a96be98be032..b96909fbeb12 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c @@ -257,6 +257,7 @@ static int mlx5_sf_add(struct mlx5_core_dev *dev, struct mlx5_sf_table *table, return 0; esw_err: + mlx5_sf_function_id_erase(table, sf); mlx5_sf_free(table, sf); return err; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/sws/dr_send.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/sws/dr_send.c index 6fa06ba2d346..f57c84e5128b 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/sws/dr_send.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/sws/dr_send.c @@ -1067,7 +1067,6 @@ static struct mlx5dr_cq *dr_create_cq(struct mlx5_core_dev *mdev, int inlen, err, eqn; void *cqc, *in; __be64 *pas; - int vector; u32 i; cq = kzalloc(sizeof(*cq), GFP_KERNEL); @@ -1096,8 +1095,7 @@ static struct mlx5dr_cq *dr_create_cq(struct mlx5_core_dev *mdev, if (!in) goto err_cqwq; - vector = raw_smp_processor_id() % mlx5_comp_vectors_max(mdev); - err = mlx5_comp_eqn_get(mdev, vector, &eqn); + err = mlx5_comp_eqn_get(mdev, 0, &eqn); if (err) { kvfree(in); goto err_cqwq; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/wc.c b/drivers/net/ethernet/mellanox/mlx5/core/wc.c index 1bed75eca97d..740b719e7072 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/wc.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/wc.c @@ -382,6 +382,7 @@ err_alloc_bfreg: bool mlx5_wc_support_get(struct mlx5_core_dev *mdev) { + struct mutex *wc_state_lock = &mdev->wc_state_lock; struct mlx5_core_dev *parent = NULL; if (!MLX5_CAP_GEN(mdev, bf)) { @@ -400,32 +401,31 @@ bool mlx5_wc_support_get(struct mlx5_core_dev *mdev) */ goto out; - mutex_lock(&mdev->wc_state_lock); - - if (mdev->wc_state != MLX5_WC_STATE_UNINITIALIZED) - goto unlock; - #ifdef CONFIG_MLX5_SF - if (mlx5_core_is_sf(mdev)) + if (mlx5_core_is_sf(mdev)) { parent = mdev->priv.parent_mdev; + wc_state_lock = &parent->wc_state_lock; + } #endif - if (parent) { - mutex_lock(&parent->wc_state_lock); + mutex_lock(wc_state_lock); + if (mdev->wc_state != MLX5_WC_STATE_UNINITIALIZED) + goto unlock; + + if (parent) { mlx5_core_test_wc(parent); mlx5_core_dbg(mdev, "parent set wc_state=%d\n", parent->wc_state); mdev->wc_state = parent->wc_state; - mutex_unlock(&parent->wc_state_lock); + } else { + mlx5_core_test_wc(mdev); } - mlx5_core_test_wc(mdev); - unlock: - mutex_unlock(&mdev->wc_state_lock); + mutex_unlock(wc_state_lock); out: mlx5_core_dbg(mdev, "wc_state=%d\n", mdev->wc_state); diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.c index 4b5fd71c897d..32d2e61f2b82 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.c @@ -423,8 +423,7 @@ mlxsw_sp_span_gretap4_route(const struct net_device *to_dev, parms = mlxsw_sp_ipip_netdev_parms4(to_dev); ip_tunnel_init_flow(&fl4, parms.iph.protocol, *daddrp, *saddrp, - 0, 0, dev_net(to_dev), parms.link, tun->fwmark, 0, - 0); + 0, 0, tun->net, parms.link, tun->fwmark, 0, 0); rt = ip_route_output_key(tun->net, &fl4); if (IS_ERR(rt)) diff --git a/drivers/net/ethernet/meta/fbnic/Makefile b/drivers/net/ethernet/meta/fbnic/Makefile index 239b2258ec65..ea6214ca48e7 100644 --- a/drivers/net/ethernet/meta/fbnic/Makefile +++ b/drivers/net/ethernet/meta/fbnic/Makefile @@ -13,7 +13,6 @@ fbnic-y := fbnic_csr.o \ fbnic_ethtool.o \ fbnic_fw.o \ fbnic_hw_stats.o \ - fbnic_hwmon.o \ fbnic_irq.o \ fbnic_mac.o \ fbnic_netdev.o \ diff --git a/drivers/net/ethernet/meta/fbnic/fbnic.h b/drivers/net/ethernet/meta/fbnic/fbnic.h index 706ae6104c8e..744eb0d95449 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic.h +++ b/drivers/net/ethernet/meta/fbnic/fbnic.h @@ -20,7 +20,6 @@ struct fbnic_dev { struct device *dev; struct net_device *netdev; struct dentry *dbg_fbd; - struct device *hwmon; u32 __iomem *uc_addr0; u32 __iomem *uc_addr4; @@ -33,7 +32,6 @@ struct fbnic_dev { struct fbnic_fw_mbx mbx[FBNIC_IPC_MBX_INDICES]; struct fbnic_fw_cap fw_cap; - struct fbnic_fw_completion *cmpl_data; /* Lock protecting Tx Mailbox queue to prevent possible races */ spinlock_t fw_tx_lock; @@ -142,9 +140,6 @@ void fbnic_devlink_unregister(struct fbnic_dev *fbd); int fbnic_fw_enable_mbx(struct fbnic_dev *fbd); void fbnic_fw_disable_mbx(struct fbnic_dev *fbd); -void fbnic_hwmon_register(struct fbnic_dev *fbd); -void fbnic_hwmon_unregister(struct fbnic_dev *fbd); - int fbnic_pcs_irq_enable(struct fbnic_dev *fbd); void fbnic_pcs_irq_disable(struct fbnic_dev *fbd); diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_csr.c b/drivers/net/ethernet/meta/fbnic/fbnic_csr.c index 2118901b25e9..aeb9f333f4c7 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic_csr.c +++ b/drivers/net/ethernet/meta/fbnic/fbnic_csr.c @@ -64,7 +64,7 @@ static void fbnic_csr_get_regs_rpc_ram(struct fbnic_dev *fbd, u32 **data_p) u32 i, j; *(data++) = start; - *(data++) = end - 1; + *(data++) = end; /* FBNIC_RPC_TCAM_ACT */ for (i = 0; i < FBNIC_RPC_TCAM_ACT_NUM_ENTRIES; i++) { diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_fw.h b/drivers/net/ethernet/meta/fbnic/fbnic_fw.h index 7cd8841920e4..221faf8c6756 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic_fw.h +++ b/drivers/net/ethernet/meta/fbnic/fbnic_fw.h @@ -44,13 +44,6 @@ struct fbnic_fw_cap { u8 link_fec; }; -struct fbnic_fw_completion { - struct { - s32 millivolts; - s32 millidegrees; - } tsene; -}; - void fbnic_mbx_init(struct fbnic_dev *fbd); void fbnic_mbx_clean(struct fbnic_dev *fbd); void fbnic_mbx_poll(struct fbnic_dev *fbd); diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_hwmon.c b/drivers/net/ethernet/meta/fbnic/fbnic_hwmon.c deleted file mode 100644 index bcd1086e3768..000000000000 --- a/drivers/net/ethernet/meta/fbnic/fbnic_hwmon.c +++ /dev/null @@ -1,81 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0 -/* Copyright (c) Meta Platforms, Inc. and affiliates. */ - -#include <linux/hwmon.h> - -#include "fbnic.h" -#include "fbnic_mac.h" - -static int fbnic_hwmon_sensor_id(enum hwmon_sensor_types type) -{ - if (type == hwmon_temp) - return FBNIC_SENSOR_TEMP; - if (type == hwmon_in) - return FBNIC_SENSOR_VOLTAGE; - - return -EOPNOTSUPP; -} - -static umode_t fbnic_hwmon_is_visible(const void *drvdata, - enum hwmon_sensor_types type, - u32 attr, int channel) -{ - if (type == hwmon_temp && attr == hwmon_temp_input) - return 0444; - if (type == hwmon_in && attr == hwmon_in_input) - return 0444; - - return 0; -} - -static int fbnic_hwmon_read(struct device *dev, enum hwmon_sensor_types type, - u32 attr, int channel, long *val) -{ - struct fbnic_dev *fbd = dev_get_drvdata(dev); - const struct fbnic_mac *mac = fbd->mac; - int id; - - id = fbnic_hwmon_sensor_id(type); - return id < 0 ? id : mac->get_sensor(fbd, id, val); -} - -static const struct hwmon_ops fbnic_hwmon_ops = { - .is_visible = fbnic_hwmon_is_visible, - .read = fbnic_hwmon_read, -}; - -static const struct hwmon_channel_info *fbnic_hwmon_info[] = { - HWMON_CHANNEL_INFO(temp, HWMON_T_INPUT), - HWMON_CHANNEL_INFO(in, HWMON_I_INPUT), - NULL -}; - -static const struct hwmon_chip_info fbnic_chip_info = { - .ops = &fbnic_hwmon_ops, - .info = fbnic_hwmon_info, -}; - -void fbnic_hwmon_register(struct fbnic_dev *fbd) -{ - if (!IS_REACHABLE(CONFIG_HWMON)) - return; - - fbd->hwmon = hwmon_device_register_with_info(fbd->dev, "fbnic", - fbd, &fbnic_chip_info, - NULL); - if (IS_ERR(fbd->hwmon)) { - dev_notice(fbd->dev, - "Failed to register hwmon device %pe\n", - fbd->hwmon); - fbd->hwmon = NULL; - } -} - -void fbnic_hwmon_unregister(struct fbnic_dev *fbd) -{ - if (!IS_REACHABLE(CONFIG_HWMON) || !fbd->hwmon) - return; - - hwmon_device_unregister(fbd->hwmon); - fbd->hwmon = NULL; -} diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_mac.c b/drivers/net/ethernet/meta/fbnic/fbnic_mac.c index 80b82ff12c4d..7b654d0a6dac 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic_mac.c +++ b/drivers/net/ethernet/meta/fbnic/fbnic_mac.c @@ -686,27 +686,6 @@ fbnic_mac_get_eth_mac_stats(struct fbnic_dev *fbd, bool reset, MAC_STAT_TX_BROADCAST); } -static int fbnic_mac_get_sensor_asic(struct fbnic_dev *fbd, int id, long *val) -{ - struct fbnic_fw_completion fw_cmpl; - s32 *sensor; - - switch (id) { - case FBNIC_SENSOR_TEMP: - sensor = &fw_cmpl.tsene.millidegrees; - break; - case FBNIC_SENSOR_VOLTAGE: - sensor = &fw_cmpl.tsene.millivolts; - break; - default: - return -EINVAL; - } - - *val = *sensor; - - return 0; -} - static const struct fbnic_mac fbnic_mac_asic = { .init_regs = fbnic_mac_init_regs, .pcs_enable = fbnic_pcs_enable_asic, @@ -716,7 +695,6 @@ static const struct fbnic_mac fbnic_mac_asic = { .get_eth_mac_stats = fbnic_mac_get_eth_mac_stats, .link_down = fbnic_mac_link_down_asic, .link_up = fbnic_mac_link_up_asic, - .get_sensor = fbnic_mac_get_sensor_asic, }; /** diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_mac.h b/drivers/net/ethernet/meta/fbnic/fbnic_mac.h index 05a591653e09..476239a9d381 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic_mac.h +++ b/drivers/net/ethernet/meta/fbnic/fbnic_mac.h @@ -47,11 +47,6 @@ enum { #define FBNIC_LINK_MODE_PAM4 (FBNIC_LINK_50R1) #define FBNIC_LINK_MODE_MASK (FBNIC_LINK_AUTO - 1) -enum fbnic_sensor_id { - FBNIC_SENSOR_TEMP, /* Temp in millidegrees Centigrade */ - FBNIC_SENSOR_VOLTAGE, /* Voltage in millivolts */ -}; - /* This structure defines the interface hooks for the MAC. The MAC hooks * will be configured as a const struct provided with a set of function * pointers. @@ -88,8 +83,6 @@ struct fbnic_mac { void (*link_down)(struct fbnic_dev *fbd); void (*link_up)(struct fbnic_dev *fbd, bool tx_pause, bool rx_pause); - - int (*get_sensor)(struct fbnic_dev *fbd, int id, long *val); }; int fbnic_mac_init(struct fbnic_dev *fbd); diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_pci.c b/drivers/net/ethernet/meta/fbnic/fbnic_pci.c index 32702dc4a066..7ccf192f13d5 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic_pci.c +++ b/drivers/net/ethernet/meta/fbnic/fbnic_pci.c @@ -296,8 +296,6 @@ static int fbnic_probe(struct pci_dev *pdev, const struct pci_device_id *ent) /* Capture snapshot of hardware stats so netdev can calculate delta */ fbnic_reset_hw_stats(fbd); - fbnic_hwmon_register(fbd); - if (!fbd->dsn) { dev_warn(&pdev->dev, "Reading serial number failed\n"); goto init_failure_mode; @@ -360,7 +358,6 @@ static void fbnic_remove(struct pci_dev *pdev) fbnic_netdev_free(fbd); } - fbnic_hwmon_unregister(fbd); fbnic_dbg_fbd_exit(fbd); fbnic_devlink_unregister(fbd); fbnic_fw_disable_mbx(fbd); diff --git a/drivers/net/ethernet/microsoft/mana/gdma_main.c b/drivers/net/ethernet/microsoft/mana/gdma_main.c index 2dc0c6ad54be..be95336ce089 100644 --- a/drivers/net/ethernet/microsoft/mana/gdma_main.c +++ b/drivers/net/ethernet/microsoft/mana/gdma_main.c @@ -1656,9 +1656,9 @@ static int __init mana_driver_init(void) static void __exit mana_driver_exit(void) { - debugfs_remove(mana_debugfs_root); - pci_unregister_driver(&mana_driver); + + debugfs_remove(mana_debugfs_root); } module_init(mana_driver_init); diff --git a/drivers/net/ethernet/netronome/nfp/bpf/offload.c b/drivers/net/ethernet/netronome/nfp/bpf/offload.c index 9d97cd281f18..c03558adda91 100644 --- a/drivers/net/ethernet/netronome/nfp/bpf/offload.c +++ b/drivers/net/ethernet/netronome/nfp/bpf/offload.c @@ -458,7 +458,8 @@ int nfp_bpf_event_output(struct nfp_app_bpf *bpf, const void *data, map_id_full = be64_to_cpu(cbe->map_ptr); map_id = map_id_full; - if (len < sizeof(struct cmsg_bpf_event) + pkt_size + data_size) + if (size_add(pkt_size, data_size) > INT_MAX || + len < sizeof(struct cmsg_bpf_event) + pkt_size + data_size) return -EINVAL; if (cbe->hdr.ver != NFP_CCM_ABI_VERSION) return -EINVAL; diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c index 739707a7b40f..8a3959bb2360 100644 --- a/drivers/net/ethernet/realtek/r8169_main.c +++ b/drivers/net/ethernet/realtek/r8169_main.c @@ -16,7 +16,6 @@ #include <linux/clk.h> #include <linux/delay.h> #include <linux/ethtool.h> -#include <linux/hwmon.h> #include <linux/phy.h> #include <linux/if_vlan.h> #include <linux/in.h> @@ -5347,43 +5346,6 @@ static bool rtl_aspm_is_safe(struct rtl8169_private *tp) return false; } -static umode_t r8169_hwmon_is_visible(const void *drvdata, - enum hwmon_sensor_types type, - u32 attr, int channel) -{ - return 0444; -} - -static int r8169_hwmon_read(struct device *dev, enum hwmon_sensor_types type, - u32 attr, int channel, long *val) -{ - struct rtl8169_private *tp = dev_get_drvdata(dev); - int val_raw; - - val_raw = phy_read_paged(tp->phydev, 0xbd8, 0x12) & 0x3ff; - if (val_raw >= 512) - val_raw -= 1024; - - *val = 1000 * val_raw / 2; - - return 0; -} - -static const struct hwmon_ops r8169_hwmon_ops = { - .is_visible = r8169_hwmon_is_visible, - .read = r8169_hwmon_read, -}; - -static const struct hwmon_channel_info * const r8169_hwmon_info[] = { - HWMON_CHANNEL_INFO(temp, HWMON_T_INPUT), - NULL -}; - -static const struct hwmon_chip_info r8169_hwmon_chip_info = { - .ops = &r8169_hwmon_ops, - .info = r8169_hwmon_info, -}; - static int rtl_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) { struct rtl8169_private *tp; @@ -5563,12 +5525,6 @@ static int rtl_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) if (rc) return rc; - /* The temperature sensor is available from RTl8125B */ - if (IS_REACHABLE(CONFIG_HWMON) && tp->mac_version >= RTL_GIGA_MAC_VER_63) - /* ignore errors */ - devm_hwmon_device_register_with_info(&pdev->dev, "nic_temp", tp, - &r8169_hwmon_chip_info, - NULL); rc = register_netdev(dev); if (rc) return rc; diff --git a/drivers/net/ethernet/realtek/rtase/rtase_main.c b/drivers/net/ethernet/realtek/rtase/rtase_main.c index de7f11232593..c42c0516656b 100644 --- a/drivers/net/ethernet/realtek/rtase/rtase_main.c +++ b/drivers/net/ethernet/realtek/rtase/rtase_main.c @@ -1827,7 +1827,7 @@ static int rtase_alloc_msix(struct pci_dev *pdev, struct rtase_private *tp) for (i = 0; i < tp->int_nums; i++) { irq = pci_irq_vector(pdev, i); - if (!irq) { + if (irq < 0) { pci_disable_msix(pdev); return irq; } diff --git a/drivers/net/ethernet/renesas/ravb_main.c b/drivers/net/ethernet/renesas/ravb_main.c index ac0f093f647a..bc395294a32d 100644 --- a/drivers/net/ethernet/renesas/ravb_main.c +++ b/drivers/net/ethernet/renesas/ravb_main.c @@ -2763,6 +2763,7 @@ static const struct ravb_hw_info ravb_rzv2m_hw_info = { .net_features = NETIF_F_RXCSUM, .stats_len = ARRAY_SIZE(ravb_gstrings_stats), .tccr_mask = TCCR_TSRQ0 | TCCR_TSRQ1 | TCCR_TSRQ2 | TCCR_TSRQ3, + .tx_max_frame_size = SZ_2K, .rx_max_frame_size = SZ_2K, .rx_buffer_size = SZ_2K + SKB_DATA_ALIGN(sizeof(struct skb_shared_info)), diff --git a/drivers/net/ethernet/sfc/tc_conntrack.c b/drivers/net/ethernet/sfc/tc_conntrack.c index d90206f27161..c0603f54cec3 100644 --- a/drivers/net/ethernet/sfc/tc_conntrack.c +++ b/drivers/net/ethernet/sfc/tc_conntrack.c @@ -16,7 +16,7 @@ static int efx_tc_flow_block(enum tc_setup_type type, void *type_data, void *cb_priv); static const struct rhashtable_params efx_tc_ct_zone_ht_params = { - .key_len = offsetof(struct efx_tc_ct_zone, linkage), + .key_len = sizeof_field(struct efx_tc_ct_zone, zone), .key_offset = 0, .head_offset = offsetof(struct efx_tc_ct_zone, linkage), }; diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-tegra.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-tegra.c index 3827997d2132..dc903b846b1b 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-tegra.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-tegra.c @@ -1,4 +1,5 @@ // SPDX-License-Identifier: GPL-2.0-only +#include <linux/iommu.h> #include <linux/platform_device.h> #include <linux/of.h> #include <linux/module.h> @@ -19,6 +20,8 @@ struct tegra_mgbe { struct reset_control *rst_mac; struct reset_control *rst_pcs; + u32 iommu_sid; + void __iomem *hv; void __iomem *regs; void __iomem *xpcs; @@ -50,7 +53,6 @@ struct tegra_mgbe { #define MGBE_WRAP_COMMON_INTR_ENABLE 0x8704 #define MAC_SBD_INTR BIT(2) #define MGBE_WRAP_AXI_ASID0_CTRL 0x8400 -#define MGBE_SID 0x6 static int __maybe_unused tegra_mgbe_suspend(struct device *dev) { @@ -84,7 +86,7 @@ static int __maybe_unused tegra_mgbe_resume(struct device *dev) writel(MAC_SBD_INTR, mgbe->regs + MGBE_WRAP_COMMON_INTR_ENABLE); /* Program SID */ - writel(MGBE_SID, mgbe->hv + MGBE_WRAP_AXI_ASID0_CTRL); + writel(mgbe->iommu_sid, mgbe->hv + MGBE_WRAP_AXI_ASID0_CTRL); value = readl(mgbe->xpcs + XPCS_WRAP_UPHY_STATUS); if ((value & XPCS_WRAP_UPHY_STATUS_TX_P_UP) == 0) { @@ -241,6 +243,12 @@ static int tegra_mgbe_probe(struct platform_device *pdev) if (IS_ERR(mgbe->xpcs)) return PTR_ERR(mgbe->xpcs); + /* get controller's stream id from iommu property in device tree */ + if (!tegra_dev_iommu_get_stream_id(mgbe->dev, &mgbe->iommu_sid)) { + dev_err(mgbe->dev, "failed to get iommu stream id\n"); + return -EINVAL; + } + res.addr = mgbe->regs; res.irq = irq; @@ -346,7 +354,7 @@ static int tegra_mgbe_probe(struct platform_device *pdev) writel(MAC_SBD_INTR, mgbe->regs + MGBE_WRAP_COMMON_INTR_ENABLE); /* Program SID */ - writel(MGBE_SID, mgbe->hv + MGBE_WRAP_AXI_ASID0_CTRL); + writel(mgbe->iommu_sid, mgbe->hv + MGBE_WRAP_AXI_ASID0_CTRL); plat->flags |= STMMAC_FLAG_SERDES_UP_AFTER_PHY_LINKUP; diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c index 3ac32444e492..dc9884130b91 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c @@ -406,22 +406,6 @@ static int stmmac_of_get_mac_mode(struct device_node *np) } /** - * stmmac_remove_config_dt - undo the effects of stmmac_probe_config_dt() - * @pdev: platform_device structure - * @plat: driver data platform structure - * - * Release resources claimed by stmmac_probe_config_dt(). - */ -static void stmmac_remove_config_dt(struct platform_device *pdev, - struct plat_stmmacenet_data *plat) -{ - clk_disable_unprepare(plat->stmmac_clk); - clk_disable_unprepare(plat->pclk); - of_node_put(plat->phy_node); - of_node_put(plat->mdio_node); -} - -/** * stmmac_probe_config_dt - parse device-tree driver parameters * @pdev: platform_device structure * @mac: MAC address to use @@ -490,8 +474,10 @@ stmmac_probe_config_dt(struct platform_device *pdev, u8 *mac) dev_warn(&pdev->dev, "snps,phy-addr property is deprecated\n"); rc = stmmac_mdio_setup(plat, np, &pdev->dev); - if (rc) - return ERR_PTR(rc); + if (rc) { + ret = ERR_PTR(rc); + goto error_put_phy; + } of_property_read_u32(np, "tx-fifo-depth", &plat->tx_fifo_size); @@ -581,8 +567,8 @@ stmmac_probe_config_dt(struct platform_device *pdev, u8 *mac) dma_cfg = devm_kzalloc(&pdev->dev, sizeof(*dma_cfg), GFP_KERNEL); if (!dma_cfg) { - stmmac_remove_config_dt(pdev, plat); - return ERR_PTR(-ENOMEM); + ret = ERR_PTR(-ENOMEM); + goto error_put_mdio; } plat->dma_cfg = dma_cfg; @@ -610,8 +596,8 @@ stmmac_probe_config_dt(struct platform_device *pdev, u8 *mac) rc = stmmac_mtl_setup(pdev, plat); if (rc) { - stmmac_remove_config_dt(pdev, plat); - return ERR_PTR(rc); + ret = ERR_PTR(rc); + goto error_put_mdio; } /* clock setup */ @@ -663,6 +649,10 @@ error_hw_init: clk_disable_unprepare(plat->pclk); error_pclk_get: clk_disable_unprepare(plat->stmmac_clk); +error_put_mdio: + of_node_put(plat->mdio_node); +error_put_phy: + of_node_put(plat->phy_node); return ret; } @@ -671,16 +661,17 @@ static void devm_stmmac_remove_config_dt(void *data) { struct plat_stmmacenet_data *plat = data; - /* Platform data argument is unused */ - stmmac_remove_config_dt(NULL, plat); + clk_disable_unprepare(plat->stmmac_clk); + clk_disable_unprepare(plat->pclk); + of_node_put(plat->mdio_node); + of_node_put(plat->phy_node); } /** * devm_stmmac_probe_config_dt * @pdev: platform_device structure * @mac: MAC address to use - * Description: Devres variant of stmmac_probe_config_dt(). Does not require - * the user to call stmmac_remove_config_dt() at driver detach. + * Description: Devres variant of stmmac_probe_config_dt(). */ struct plat_stmmacenet_data * devm_stmmac_probe_config_dt(struct platform_device *pdev, u8 *mac) diff --git a/drivers/net/ethernet/ti/am65-cpsw-nuss.c b/drivers/net/ethernet/ti/am65-cpsw-nuss.c index 14e1df721f2e..5465bf872734 100644 --- a/drivers/net/ethernet/ti/am65-cpsw-nuss.c +++ b/drivers/net/ethernet/ti/am65-cpsw-nuss.c @@ -3551,7 +3551,7 @@ static int am65_cpsw_nuss_probe(struct platform_device *pdev) init_completion(&common->tdown_complete); common->tx_ch_num = AM65_CPSW_DEFAULT_TX_CHNS; common->rx_ch_num_flows = AM65_CPSW_DEFAULT_RX_CHN_FLOWS; - common->pf_p0_rx_ptype_rrobin = false; + common->pf_p0_rx_ptype_rrobin = true; common->default_vlan = 1; common->ports = devm_kcalloc(dev, common->port_num, diff --git a/drivers/net/ethernet/ti/cpsw_ale.c b/drivers/net/ethernet/ti/cpsw_ale.c index 52e4e350b734..5cc72a91f220 100644 --- a/drivers/net/ethernet/ti/cpsw_ale.c +++ b/drivers/net/ethernet/ti/cpsw_ale.c @@ -127,15 +127,15 @@ struct cpsw_ale_dev_id { static inline int cpsw_ale_get_field(u32 *ale_entry, u32 start, u32 bits) { - int idx, idx2; + int idx, idx2, index; u32 hi_val = 0; idx = start / 32; idx2 = (start + bits - 1) / 32; /* Check if bits to be fetched exceed a word */ if (idx != idx2) { - idx2 = 2 - idx2; /* flip */ - hi_val = ale_entry[idx2] << ((idx2 * 32) - start); + index = 2 - idx2; /* flip */ + hi_val = ale_entry[index] << ((idx2 * 32) - start); } start -= idx * 32; idx = 2 - idx; /* flip */ @@ -145,16 +145,16 @@ static inline int cpsw_ale_get_field(u32 *ale_entry, u32 start, u32 bits) static inline void cpsw_ale_set_field(u32 *ale_entry, u32 start, u32 bits, u32 value) { - int idx, idx2; + int idx, idx2, index; value &= BITMASK(bits); idx = start / 32; idx2 = (start + bits - 1) / 32; /* Check if bits to be set exceed a word */ if (idx != idx2) { - idx2 = 2 - idx2; /* flip */ - ale_entry[idx2] &= ~(BITMASK(bits + start - (idx2 * 32))); - ale_entry[idx2] |= (value >> ((idx2 * 32) - start)); + index = 2 - idx2; /* flip */ + ale_entry[index] &= ~(BITMASK(bits + start - (idx2 * 32))); + ale_entry[index] |= (value >> ((idx2 * 32) - start)); } start -= idx * 32; idx = 2 - idx; /* flip */ diff --git a/drivers/net/ethernet/ti/icssg/icss_iep.c b/drivers/net/ethernet/ti/icssg/icss_iep.c index 5d6d1cf78e93..768578c0d958 100644 --- a/drivers/net/ethernet/ti/icssg/icss_iep.c +++ b/drivers/net/ethernet/ti/icssg/icss_iep.c @@ -215,6 +215,9 @@ static void icss_iep_enable_shadow_mode(struct icss_iep *iep) for (cmp = IEP_MIN_CMP; cmp < IEP_MAX_CMP; cmp++) { regmap_update_bits(iep->map, ICSS_IEP_CMP_STAT_REG, IEP_CMP_STATUS(cmp), IEP_CMP_STATUS(cmp)); + + regmap_update_bits(iep->map, ICSS_IEP_CMP_CFG_REG, + IEP_CMP_CFG_CMP_EN(cmp), 0); } /* enable reset counter on CMP0 event */ @@ -780,6 +783,11 @@ int icss_iep_exit(struct icss_iep *iep) } icss_iep_disable(iep); + if (iep->pps_enabled) + icss_iep_pps_enable(iep, false); + else if (iep->perout_enabled) + icss_iep_perout_enable(iep, NULL, false); + return 0; } EXPORT_SYMBOL_GPL(icss_iep_exit); diff --git a/drivers/net/ethernet/ti/icssg/icssg_common.c b/drivers/net/ethernet/ti/icssg/icssg_common.c index fdebeb2f84e0..74f0f200a89d 100644 --- a/drivers/net/ethernet/ti/icssg/icssg_common.c +++ b/drivers/net/ethernet/ti/icssg/icssg_common.c @@ -855,31 +855,6 @@ irqreturn_t prueth_rx_irq(int irq, void *dev_id) } EXPORT_SYMBOL_GPL(prueth_rx_irq); -void prueth_emac_stop(struct prueth_emac *emac) -{ - struct prueth *prueth = emac->prueth; - int slice; - - switch (emac->port_id) { - case PRUETH_PORT_MII0: - slice = ICSS_SLICE0; - break; - case PRUETH_PORT_MII1: - slice = ICSS_SLICE1; - break; - default: - netdev_err(emac->ndev, "invalid port\n"); - return; - } - - emac->fw_running = 0; - if (!emac->is_sr1) - rproc_shutdown(prueth->txpru[slice]); - rproc_shutdown(prueth->rtu[slice]); - rproc_shutdown(prueth->pru[slice]); -} -EXPORT_SYMBOL_GPL(prueth_emac_stop); - void prueth_cleanup_tx_ts(struct prueth_emac *emac) { int i; diff --git a/drivers/net/ethernet/ti/icssg/icssg_config.c b/drivers/net/ethernet/ti/icssg/icssg_config.c index 5d2491c2943a..ddfd1c02a885 100644 --- a/drivers/net/ethernet/ti/icssg/icssg_config.c +++ b/drivers/net/ethernet/ti/icssg/icssg_config.c @@ -397,7 +397,7 @@ static int prueth_emac_buffer_setup(struct prueth_emac *emac) return 0; } -static void icssg_init_emac_mode(struct prueth *prueth) +void icssg_init_emac_mode(struct prueth *prueth) { /* When the device is configured as a bridge and it is being brought * back to the emac mode, the host mac address has to be set as 0. @@ -406,9 +406,6 @@ static void icssg_init_emac_mode(struct prueth *prueth) int i; u8 mac[ETH_ALEN] = { 0 }; - if (prueth->emacs_initialized) - return; - /* Set VLAN TABLE address base */ regmap_update_bits(prueth->miig_rt, FDB_GEN_CFG1, SMEM_VLAN_OFFSET_MASK, addr << SMEM_VLAN_OFFSET); @@ -423,15 +420,13 @@ static void icssg_init_emac_mode(struct prueth *prueth) /* Clear host MAC address */ icssg_class_set_host_mac_addr(prueth->miig_rt, mac); } +EXPORT_SYMBOL_GPL(icssg_init_emac_mode); -static void icssg_init_fw_offload_mode(struct prueth *prueth) +void icssg_init_fw_offload_mode(struct prueth *prueth) { u32 addr = prueth->shram.pa + EMAC_ICSSG_SWITCH_DEFAULT_VLAN_TABLE_OFFSET; int i; - if (prueth->emacs_initialized) - return; - /* Set VLAN TABLE address base */ regmap_update_bits(prueth->miig_rt, FDB_GEN_CFG1, SMEM_VLAN_OFFSET_MASK, addr << SMEM_VLAN_OFFSET); @@ -448,6 +443,7 @@ static void icssg_init_fw_offload_mode(struct prueth *prueth) icssg_class_set_host_mac_addr(prueth->miig_rt, prueth->hw_bridge_dev->dev_addr); icssg_set_pvid(prueth, prueth->default_vlan, PRUETH_PORT_HOST); } +EXPORT_SYMBOL_GPL(icssg_init_fw_offload_mode); int icssg_config(struct prueth *prueth, struct prueth_emac *emac, int slice) { @@ -455,11 +451,6 @@ int icssg_config(struct prueth *prueth, struct prueth_emac *emac, int slice) struct icssg_flow_cfg __iomem *flow_cfg; int ret; - if (prueth->is_switch_mode || prueth->is_hsr_offload_mode) - icssg_init_fw_offload_mode(prueth); - else - icssg_init_emac_mode(prueth); - memset_io(config, 0, TAS_GATE_MASK_LIST0); icssg_miig_queues_init(prueth, slice); @@ -786,3 +777,27 @@ void icssg_set_pvid(struct prueth *prueth, u8 vid, u8 port) writel(pvid, prueth->shram.va + EMAC_ICSSG_SWITCH_PORT0_DEFAULT_VLAN_OFFSET); } EXPORT_SYMBOL_GPL(icssg_set_pvid); + +int emac_fdb_flow_id_updated(struct prueth_emac *emac) +{ + struct mgmt_cmd_rsp fdb_cmd_rsp = { 0 }; + int slice = prueth_emac_slice(emac); + struct mgmt_cmd fdb_cmd = { 0 }; + int ret; + + fdb_cmd.header = ICSSG_FW_MGMT_CMD_HEADER; + fdb_cmd.type = ICSSG_FW_MGMT_FDB_CMD_TYPE_RX_FLOW; + fdb_cmd.seqnum = ++(emac->prueth->icssg_hwcmdseq); + fdb_cmd.param = 0; + + fdb_cmd.param |= (slice << 4); + fdb_cmd.cmd_args[0] = 0; + + ret = icssg_send_fdb_msg(emac, &fdb_cmd, &fdb_cmd_rsp); + if (ret) + return ret; + + WARN_ON(fdb_cmd.seqnum != fdb_cmd_rsp.seqnum); + return fdb_cmd_rsp.status == 1 ? 0 : -EINVAL; +} +EXPORT_SYMBOL_GPL(emac_fdb_flow_id_updated); diff --git a/drivers/net/ethernet/ti/icssg/icssg_config.h b/drivers/net/ethernet/ti/icssg/icssg_config.h index 92c2deaa3068..c884e9fa099e 100644 --- a/drivers/net/ethernet/ti/icssg/icssg_config.h +++ b/drivers/net/ethernet/ti/icssg/icssg_config.h @@ -55,6 +55,7 @@ struct icssg_rxq_ctx { #define ICSSG_FW_MGMT_FDB_CMD_TYPE 0x03 #define ICSSG_FW_MGMT_CMD_TYPE 0x04 #define ICSSG_FW_MGMT_PKT 0x80000000 +#define ICSSG_FW_MGMT_FDB_CMD_TYPE_RX_FLOW 0x05 struct icssg_r30_cmd { u32 cmd[4]; diff --git a/drivers/net/ethernet/ti/icssg/icssg_prueth.c b/drivers/net/ethernet/ti/icssg/icssg_prueth.c index c568c84a032b..d76fe6d05e10 100644 --- a/drivers/net/ethernet/ti/icssg/icssg_prueth.c +++ b/drivers/net/ethernet/ti/icssg/icssg_prueth.c @@ -164,11 +164,26 @@ static struct icssg_firmwares icssg_emac_firmwares[] = { } }; -static int prueth_emac_start(struct prueth *prueth, struct prueth_emac *emac) +static int prueth_start(struct rproc *rproc, const char *fw_name) +{ + int ret; + + ret = rproc_set_firmware(rproc, fw_name); + if (ret) + return ret; + return rproc_boot(rproc); +} + +static void prueth_shutdown(struct rproc *rproc) +{ + rproc_shutdown(rproc); +} + +static int prueth_emac_start(struct prueth *prueth) { struct icssg_firmwares *firmwares; struct device *dev = prueth->dev; - int slice, ret; + int ret, slice; if (prueth->is_switch_mode) firmwares = icssg_switch_firmwares; @@ -177,49 +192,126 @@ static int prueth_emac_start(struct prueth *prueth, struct prueth_emac *emac) else firmwares = icssg_emac_firmwares; - slice = prueth_emac_slice(emac); - if (slice < 0) { - netdev_err(emac->ndev, "invalid port\n"); - return -EINVAL; + for (slice = 0; slice < PRUETH_NUM_MACS; slice++) { + ret = prueth_start(prueth->pru[slice], firmwares[slice].pru); + if (ret) { + dev_err(dev, "failed to boot PRU%d: %d\n", slice, ret); + goto unwind_slices; + } + + ret = prueth_start(prueth->rtu[slice], firmwares[slice].rtu); + if (ret) { + dev_err(dev, "failed to boot RTU%d: %d\n", slice, ret); + rproc_shutdown(prueth->pru[slice]); + goto unwind_slices; + } + + ret = prueth_start(prueth->txpru[slice], firmwares[slice].txpru); + if (ret) { + dev_err(dev, "failed to boot TX_PRU%d: %d\n", slice, ret); + rproc_shutdown(prueth->rtu[slice]); + rproc_shutdown(prueth->pru[slice]); + goto unwind_slices; + } } - ret = icssg_config(prueth, emac, slice); - if (ret) - return ret; + return 0; - ret = rproc_set_firmware(prueth->pru[slice], firmwares[slice].pru); - ret = rproc_boot(prueth->pru[slice]); - if (ret) { - dev_err(dev, "failed to boot PRU%d: %d\n", slice, ret); - return -EINVAL; +unwind_slices: + while (--slice >= 0) { + prueth_shutdown(prueth->txpru[slice]); + prueth_shutdown(prueth->rtu[slice]); + prueth_shutdown(prueth->pru[slice]); } - ret = rproc_set_firmware(prueth->rtu[slice], firmwares[slice].rtu); - ret = rproc_boot(prueth->rtu[slice]); - if (ret) { - dev_err(dev, "failed to boot RTU%d: %d\n", slice, ret); - goto halt_pru; + return ret; +} + +static void prueth_emac_stop(struct prueth *prueth) +{ + int slice; + + for (slice = 0; slice < PRUETH_NUM_MACS; slice++) { + prueth_shutdown(prueth->txpru[slice]); + prueth_shutdown(prueth->rtu[slice]); + prueth_shutdown(prueth->pru[slice]); } +} + +static int prueth_emac_common_start(struct prueth *prueth) +{ + struct prueth_emac *emac; + int ret = 0; + int slice; + + if (!prueth->emac[ICSS_SLICE0] && !prueth->emac[ICSS_SLICE1]) + return -EINVAL; + + /* clear SMEM and MSMC settings for all slices */ + memset_io(prueth->msmcram.va, 0, prueth->msmcram.size); + memset_io(prueth->shram.va, 0, ICSSG_CONFIG_OFFSET_SLICE1 * PRUETH_NUM_MACS); + + icssg_class_default(prueth->miig_rt, ICSS_SLICE0, 0, false); + icssg_class_default(prueth->miig_rt, ICSS_SLICE1, 0, false); + + if (prueth->is_switch_mode || prueth->is_hsr_offload_mode) + icssg_init_fw_offload_mode(prueth); + else + icssg_init_emac_mode(prueth); + + for (slice = 0; slice < PRUETH_NUM_MACS; slice++) { + emac = prueth->emac[slice]; + if (!emac) + continue; + ret = icssg_config(prueth, emac, slice); + if (ret) + goto disable_class; + } + + ret = prueth_emac_start(prueth); + if (ret) + goto disable_class; - ret = rproc_set_firmware(prueth->txpru[slice], firmwares[slice].txpru); - ret = rproc_boot(prueth->txpru[slice]); + emac = prueth->emac[ICSS_SLICE0] ? prueth->emac[ICSS_SLICE0] : + prueth->emac[ICSS_SLICE1]; + ret = icss_iep_init(emac->iep, &prueth_iep_clockops, + emac, IEP_DEFAULT_CYCLE_TIME_NS); if (ret) { - dev_err(dev, "failed to boot TX_PRU%d: %d\n", slice, ret); - goto halt_rtu; + dev_err(prueth->dev, "Failed to initialize IEP module\n"); + goto stop_pruss; } - emac->fw_running = 1; return 0; -halt_rtu: - rproc_shutdown(prueth->rtu[slice]); +stop_pruss: + prueth_emac_stop(prueth); -halt_pru: - rproc_shutdown(prueth->pru[slice]); +disable_class: + icssg_class_disable(prueth->miig_rt, ICSS_SLICE0); + icssg_class_disable(prueth->miig_rt, ICSS_SLICE1); return ret; } +static int prueth_emac_common_stop(struct prueth *prueth) +{ + struct prueth_emac *emac; + + if (!prueth->emac[ICSS_SLICE0] && !prueth->emac[ICSS_SLICE1]) + return -EINVAL; + + icssg_class_disable(prueth->miig_rt, ICSS_SLICE0); + icssg_class_disable(prueth->miig_rt, ICSS_SLICE1); + + prueth_emac_stop(prueth); + + emac = prueth->emac[ICSS_SLICE0] ? prueth->emac[ICSS_SLICE0] : + prueth->emac[ICSS_SLICE1]; + icss_iep_exit(emac->iep); + + return 0; +} + /* called back by PHY layer if there is change in link state of hw port*/ static void emac_adjust_link(struct net_device *ndev) { @@ -374,9 +466,6 @@ static void prueth_iep_settime(void *clockops_data, u64 ns) u32 cycletime; int timeout; - if (!emac->fw_running) - return; - sc_descp = emac->prueth->shram.va + TIMESYNC_FW_WC_SETCLOCK_DESC_OFFSET; cycletime = IEP_DEFAULT_CYCLE_TIME_NS; @@ -543,23 +632,17 @@ static int emac_ndo_open(struct net_device *ndev) { struct prueth_emac *emac = netdev_priv(ndev); int ret, i, num_data_chn = emac->tx_ch_num; + struct icssg_flow_cfg __iomem *flow_cfg; struct prueth *prueth = emac->prueth; int slice = prueth_emac_slice(emac); struct device *dev = prueth->dev; int max_rx_flows; int rx_flow; - /* clear SMEM and MSMC settings for all slices */ - if (!prueth->emacs_initialized) { - memset_io(prueth->msmcram.va, 0, prueth->msmcram.size); - memset_io(prueth->shram.va, 0, ICSSG_CONFIG_OFFSET_SLICE1 * PRUETH_NUM_MACS); - } - /* set h/w MAC as user might have re-configured */ ether_addr_copy(emac->mac_addr, ndev->dev_addr); icssg_class_set_mac_addr(prueth->miig_rt, slice, emac->mac_addr); - icssg_class_default(prueth->miig_rt, slice, 0, false); icssg_ft1_set_mac_addr(prueth->miig_rt, slice, emac->mac_addr); /* Notify the stack of the actual queue counts. */ @@ -597,18 +680,23 @@ static int emac_ndo_open(struct net_device *ndev) goto cleanup_napi; } - /* reset and start PRU firmware */ - ret = prueth_emac_start(prueth, emac); - if (ret) - goto free_rx_irq; + if (!prueth->emacs_initialized) { + ret = prueth_emac_common_start(prueth); + if (ret) + goto free_rx_irq; + } - icssg_mii_update_mtu(prueth->mii_rt, slice, ndev->max_mtu); + flow_cfg = emac->dram.va + ICSSG_CONFIG_OFFSET + PSI_L_REGULAR_FLOW_ID_BASE_OFFSET; + writew(emac->rx_flow_id_base, &flow_cfg->rx_base_flow); + ret = emac_fdb_flow_id_updated(emac); - if (!prueth->emacs_initialized) { - ret = icss_iep_init(emac->iep, &prueth_iep_clockops, - emac, IEP_DEFAULT_CYCLE_TIME_NS); + if (ret) { + netdev_err(ndev, "Failed to update Rx Flow ID %d", ret); + goto stop; } + icssg_mii_update_mtu(prueth->mii_rt, slice, ndev->max_mtu); + ret = request_threaded_irq(emac->tx_ts_irq, NULL, prueth_tx_ts_irq, IRQF_ONESHOT, dev_name(dev), emac); if (ret) @@ -653,7 +741,8 @@ reset_rx_chn: free_tx_ts_irq: free_irq(emac->tx_ts_irq, emac); stop: - prueth_emac_stop(emac); + if (!prueth->emacs_initialized) + prueth_emac_common_stop(prueth); free_rx_irq: free_irq(emac->rx_chns.irq[rx_flow], emac); cleanup_napi: @@ -689,8 +778,6 @@ static int emac_ndo_stop(struct net_device *ndev) if (ndev->phydev) phy_stop(ndev->phydev); - icssg_class_disable(prueth->miig_rt, prueth_emac_slice(emac)); - if (emac->prueth->is_hsr_offload_mode) __dev_mc_unsync(ndev, icssg_prueth_hsr_del_mcast); else @@ -728,11 +815,9 @@ static int emac_ndo_stop(struct net_device *ndev) /* Destroying the queued work in ndo_stop() */ cancel_delayed_work_sync(&emac->stats_work); - if (prueth->emacs_initialized == 1) - icss_iep_exit(emac->iep); - /* stop PRUs */ - prueth_emac_stop(emac); + if (prueth->emacs_initialized == 1) + prueth_emac_common_stop(prueth); free_irq(emac->tx_ts_irq, emac); @@ -1053,10 +1138,11 @@ static void prueth_offload_fwd_mark_update(struct prueth *prueth) } } -static void prueth_emac_restart(struct prueth *prueth) +static int prueth_emac_restart(struct prueth *prueth) { struct prueth_emac *emac0 = prueth->emac[PRUETH_MAC0]; struct prueth_emac *emac1 = prueth->emac[PRUETH_MAC1]; + int ret; /* Detach the net_device for both PRUeth ports*/ if (netif_running(emac0->ndev)) @@ -1065,36 +1151,46 @@ static void prueth_emac_restart(struct prueth *prueth) netif_device_detach(emac1->ndev); /* Disable both PRUeth ports */ - icssg_set_port_state(emac0, ICSSG_EMAC_PORT_DISABLE); - icssg_set_port_state(emac1, ICSSG_EMAC_PORT_DISABLE); + ret = icssg_set_port_state(emac0, ICSSG_EMAC_PORT_DISABLE); + ret |= icssg_set_port_state(emac1, ICSSG_EMAC_PORT_DISABLE); + if (ret) + return ret; /* Stop both pru cores for both PRUeth ports*/ - prueth_emac_stop(emac0); - prueth->emacs_initialized--; - prueth_emac_stop(emac1); - prueth->emacs_initialized--; + ret = prueth_emac_common_stop(prueth); + if (ret) { + dev_err(prueth->dev, "Failed to stop the firmwares"); + return ret; + } /* Start both pru cores for both PRUeth ports */ - prueth_emac_start(prueth, emac0); - prueth->emacs_initialized++; - prueth_emac_start(prueth, emac1); - prueth->emacs_initialized++; + ret = prueth_emac_common_start(prueth); + if (ret) { + dev_err(prueth->dev, "Failed to start the firmwares"); + return ret; + } /* Enable forwarding for both PRUeth ports */ - icssg_set_port_state(emac0, ICSSG_EMAC_PORT_FORWARD); - icssg_set_port_state(emac1, ICSSG_EMAC_PORT_FORWARD); + ret = icssg_set_port_state(emac0, ICSSG_EMAC_PORT_FORWARD); + ret |= icssg_set_port_state(emac1, ICSSG_EMAC_PORT_FORWARD); /* Attache net_device for both PRUeth ports */ netif_device_attach(emac0->ndev); netif_device_attach(emac1->ndev); + + return ret; } static void icssg_change_mode(struct prueth *prueth) { struct prueth_emac *emac; - int mac; + int mac, ret; - prueth_emac_restart(prueth); + ret = prueth_emac_restart(prueth); + if (ret) { + dev_err(prueth->dev, "Failed to restart the firmwares, aborting the process"); + return; + } for (mac = PRUETH_MAC0; mac < PRUETH_NUM_MACS; mac++) { emac = prueth->emac[mac]; @@ -1173,13 +1269,18 @@ static void prueth_netdevice_port_unlink(struct net_device *ndev) { struct prueth_emac *emac = netdev_priv(ndev); struct prueth *prueth = emac->prueth; + int ret; prueth->br_members &= ~BIT(emac->port_id); if (prueth->is_switch_mode) { prueth->is_switch_mode = false; emac->port_vlan = 0; - prueth_emac_restart(prueth); + ret = prueth_emac_restart(prueth); + if (ret) { + dev_err(prueth->dev, "Failed to restart the firmwares, aborting the process"); + return; + } } prueth_offload_fwd_mark_update(prueth); @@ -1228,6 +1329,7 @@ static void prueth_hsr_port_unlink(struct net_device *ndev) struct prueth *prueth = emac->prueth; struct prueth_emac *emac0; struct prueth_emac *emac1; + int ret; emac0 = prueth->emac[PRUETH_MAC0]; emac1 = prueth->emac[PRUETH_MAC1]; @@ -1238,7 +1340,11 @@ static void prueth_hsr_port_unlink(struct net_device *ndev) emac0->port_vlan = 0; emac1->port_vlan = 0; prueth->hsr_dev = NULL; - prueth_emac_restart(prueth); + ret = prueth_emac_restart(prueth); + if (ret) { + dev_err(prueth->dev, "Failed to restart the firmwares, aborting the process"); + return; + } netdev_dbg(ndev, "Disabling HSR Offload mode\n"); } } @@ -1413,13 +1519,10 @@ static int prueth_probe(struct platform_device *pdev) prueth->pa_stats = NULL; } - if (eth0_node) { + if (eth0_node || eth1_node) { ret = prueth_get_cores(prueth, ICSS_SLICE0, false); if (ret) goto put_cores; - } - - if (eth1_node) { ret = prueth_get_cores(prueth, ICSS_SLICE1, false); if (ret) goto put_cores; @@ -1618,14 +1721,12 @@ put_pruss: pruss_put(prueth->pruss); put_cores: - if (eth1_node) { - prueth_put_cores(prueth, ICSS_SLICE1); - of_node_put(eth1_node); - } - - if (eth0_node) { + if (eth0_node || eth1_node) { prueth_put_cores(prueth, ICSS_SLICE0); of_node_put(eth0_node); + + prueth_put_cores(prueth, ICSS_SLICE1); + of_node_put(eth1_node); } return ret; diff --git a/drivers/net/ethernet/ti/icssg/icssg_prueth.h b/drivers/net/ethernet/ti/icssg/icssg_prueth.h index f5c1d473e9f9..5473315ea204 100644 --- a/drivers/net/ethernet/ti/icssg/icssg_prueth.h +++ b/drivers/net/ethernet/ti/icssg/icssg_prueth.h @@ -140,7 +140,6 @@ struct prueth_rx_chn { /* data for each emac port */ struct prueth_emac { bool is_sr1; - bool fw_running; struct prueth *prueth; struct net_device *ndev; u8 mac_addr[6]; @@ -361,6 +360,8 @@ int icssg_set_port_state(struct prueth_emac *emac, enum icssg_port_state_cmd state); void icssg_config_set_speed(struct prueth_emac *emac); void icssg_config_half_duplex(struct prueth_emac *emac); +void icssg_init_emac_mode(struct prueth *prueth); +void icssg_init_fw_offload_mode(struct prueth *prueth); /* Buffer queue helpers */ int icssg_queue_pop(struct prueth *prueth, u8 queue); @@ -377,6 +378,7 @@ void icssg_vtbl_modify(struct prueth_emac *emac, u8 vid, u8 port_mask, u8 untag_mask, bool add); u16 icssg_get_pvid(struct prueth_emac *emac); void icssg_set_pvid(struct prueth *prueth, u8 vid, u8 port); +int emac_fdb_flow_id_updated(struct prueth_emac *emac); #define prueth_napi_to_tx_chn(pnapi) \ container_of(pnapi, struct prueth_tx_chn, napi_tx) @@ -407,7 +409,6 @@ void emac_rx_timestamp(struct prueth_emac *emac, struct sk_buff *skb, u32 *psdata); enum netdev_tx icssg_ndo_start_xmit(struct sk_buff *skb, struct net_device *ndev); irqreturn_t prueth_rx_irq(int irq, void *dev_id); -void prueth_emac_stop(struct prueth_emac *emac); void prueth_cleanup_tx_ts(struct prueth_emac *emac); int icssg_napi_rx_poll(struct napi_struct *napi_rx, int budget); int prueth_prepare_rx_chan(struct prueth_emac *emac, diff --git a/drivers/net/ethernet/ti/icssg/icssg_prueth_sr1.c b/drivers/net/ethernet/ti/icssg/icssg_prueth_sr1.c index 5024f0647a0d..3dc86397c367 100644 --- a/drivers/net/ethernet/ti/icssg/icssg_prueth_sr1.c +++ b/drivers/net/ethernet/ti/icssg/icssg_prueth_sr1.c @@ -440,7 +440,6 @@ static int prueth_emac_start(struct prueth *prueth, struct prueth_emac *emac) goto halt_pru; } - emac->fw_running = 1; return 0; halt_pru: @@ -449,6 +448,29 @@ halt_pru: return ret; } +static void prueth_emac_stop(struct prueth_emac *emac) +{ + struct prueth *prueth = emac->prueth; + int slice; + + switch (emac->port_id) { + case PRUETH_PORT_MII0: + slice = ICSS_SLICE0; + break; + case PRUETH_PORT_MII1: + slice = ICSS_SLICE1; + break; + default: + netdev_err(emac->ndev, "invalid port\n"); + return; + } + + if (!emac->is_sr1) + rproc_shutdown(prueth->txpru[slice]); + rproc_shutdown(prueth->rtu[slice]); + rproc_shutdown(prueth->pru[slice]); +} + /** * emac_ndo_open - EMAC device open * @ndev: network adapter device diff --git a/drivers/net/ethernet/wangxun/libwx/wx_hw.c b/drivers/net/ethernet/wangxun/libwx/wx_hw.c index 1bf9c38e4125..deaf670c160e 100644 --- a/drivers/net/ethernet/wangxun/libwx/wx_hw.c +++ b/drivers/net/ethernet/wangxun/libwx/wx_hw.c @@ -334,27 +334,25 @@ int wx_host_interface_command(struct wx *wx, u32 *buffer, status = read_poll_timeout(rd32, hicr, hicr & WX_MNG_MBOX_CTL_FWRDY, 1000, timeout * 1000, false, wx, WX_MNG_MBOX_CTL); + buf[0] = rd32(wx, WX_MNG_MBOX); + if ((buf[0] & 0xff0000) >> 16 == 0x80) { + wx_err(wx, "Unknown FW command: 0x%x\n", buffer[0] & 0xff); + status = -EINVAL; + goto rel_out; + } + /* Check command completion */ if (status) { - wx_dbg(wx, "Command has failed with no status valid.\n"); - - buf[0] = rd32(wx, WX_MNG_MBOX); - if ((buffer[0] & 0xff) != (~buf[0] >> 24)) { - status = -EINVAL; - goto rel_out; - } - if ((buf[0] & 0xff0000) >> 16 == 0x80) { - wx_dbg(wx, "It's unknown cmd.\n"); - status = -EINVAL; - goto rel_out; - } - + wx_err(wx, "Command has failed with no status valid.\n"); wx_dbg(wx, "write value:\n"); for (i = 0; i < dword_len; i++) wx_dbg(wx, "%x ", buffer[i]); wx_dbg(wx, "read value:\n"); for (i = 0; i < dword_len; i++) wx_dbg(wx, "%x ", buf[i]); + wx_dbg(wx, "\ncheck: %x %x\n", buffer[0] & 0xff, ~buf[0] >> 24); + + goto rel_out; } if (!return_data) diff --git a/drivers/net/ethernet/xilinx/xilinx_axienet_main.c b/drivers/net/ethernet/xilinx/xilinx_axienet_main.c index 0f4b02fe6f85..ae743991117c 100644 --- a/drivers/net/ethernet/xilinx/xilinx_axienet_main.c +++ b/drivers/net/ethernet/xilinx/xilinx_axienet_main.c @@ -2056,6 +2056,12 @@ axienet_ethtools_set_coalesce(struct net_device *ndev, return -EBUSY; } + if (ecoalesce->rx_max_coalesced_frames > 255 || + ecoalesce->tx_max_coalesced_frames > 255) { + NL_SET_ERR_MSG(extack, "frames must be less than 256"); + return -EINVAL; + } + if (ecoalesce->rx_max_coalesced_frames) lp->coalesce_count_rx = ecoalesce->rx_max_coalesced_frames; if (ecoalesce->rx_coalesce_usecs) diff --git a/drivers/net/gtp.c b/drivers/net/gtp.c index 89a996ad8cd0..fbabada7d3ba 100644 --- a/drivers/net/gtp.c +++ b/drivers/net/gtp.c @@ -1524,8 +1524,8 @@ static int gtp_newlink(struct net *src_net, struct net_device *dev, goto out_encap; } - gn = net_generic(dev_net(dev), gtp_net_id); - list_add_rcu(>p->list, &gn->gtp_dev_list); + gn = net_generic(src_net, gtp_net_id); + list_add(>p->list, &gn->gtp_dev_list); dev->priv_destructor = gtp_destructor; netdev_dbg(dev, "registered new GTP interface\n"); @@ -1551,7 +1551,7 @@ static void gtp_dellink(struct net_device *dev, struct list_head *head) hlist_for_each_entry_safe(pctx, next, >p->tid_hash[i], hlist_tid) pdp_context_delete(pctx); - list_del_rcu(>p->list); + list_del(>p->list); unregister_netdevice_queue(dev, head); } @@ -2271,16 +2271,19 @@ static int gtp_genl_dump_pdp(struct sk_buff *skb, struct gtp_dev *last_gtp = (struct gtp_dev *)cb->args[2], *gtp; int i, j, bucket = cb->args[0], skip = cb->args[1]; struct net *net = sock_net(skb->sk); + struct net_device *dev; struct pdp_ctx *pctx; - struct gtp_net *gn; - - gn = net_generic(net, gtp_net_id); if (cb->args[4]) return 0; rcu_read_lock(); - list_for_each_entry_rcu(gtp, &gn->gtp_dev_list, list) { + for_each_netdev_rcu(net, dev) { + if (dev->rtnl_link_ops != >p_link_ops) + continue; + + gtp = netdev_priv(dev); + if (last_gtp && last_gtp != gtp) continue; else @@ -2475,9 +2478,14 @@ static void __net_exit gtp_net_exit_batch_rtnl(struct list_head *net_list, list_for_each_entry(net, net_list, exit_list) { struct gtp_net *gn = net_generic(net, gtp_net_id); - struct gtp_dev *gtp; + struct gtp_dev *gtp, *gtp_next; + struct net_device *dev; + + for_each_netdev(net, dev) + if (dev->rtnl_link_ops == >p_link_ops) + gtp_dellink(dev, dev_to_kill); - list_for_each_entry(gtp, &gn->gtp_dev_list, list) + list_for_each_entry_safe(gtp, gtp_next, &gn->gtp_dev_list, list) gtp_dellink(gtp->dev, dev_to_kill); } } diff --git a/drivers/net/ieee802154/ca8210.c b/drivers/net/ieee802154/ca8210.c index e685a7f946f0..753215ebc67c 100644 --- a/drivers/net/ieee802154/ca8210.c +++ b/drivers/net/ieee802154/ca8210.c @@ -3072,7 +3072,11 @@ static int ca8210_probe(struct spi_device *spi_device) spi_set_drvdata(priv->spi, priv); if (IS_ENABLED(CONFIG_IEEE802154_CA8210_DEBUGFS)) { cascoda_api_upstream = ca8210_test_int_driver_write; - ca8210_test_interface_init(priv); + ret = ca8210_test_interface_init(priv); + if (ret) { + dev_crit(&spi_device->dev, "ca8210_test_interface_init failed\n"); + goto error; + } } else { cascoda_api_upstream = NULL; } diff --git a/drivers/net/mctp/mctp-i3c.c b/drivers/net/mctp/mctp-i3c.c index 9adad59b8676..d247fe483c58 100644 --- a/drivers/net/mctp/mctp-i3c.c +++ b/drivers/net/mctp/mctp-i3c.c @@ -125,6 +125,8 @@ static int mctp_i3c_read(struct mctp_i3c_device *mi) xfer.data.in = skb_put(skb, mi->mrl); + /* Make sure netif_rx() is read in the same order as i3c. */ + mutex_lock(&mi->lock); rc = i3c_device_do_priv_xfers(mi->i3c, &xfer, 1); if (rc < 0) goto err; @@ -166,8 +168,10 @@ static int mctp_i3c_read(struct mctp_i3c_device *mi) stats->rx_dropped++; } + mutex_unlock(&mi->lock); return 0; err: + mutex_unlock(&mi->lock); kfree_skb(skb); return rc; } diff --git a/drivers/net/pcs/pcs-xpcs.c b/drivers/net/pcs/pcs-xpcs.c index 7246a910728d..3059435af596 100644 --- a/drivers/net/pcs/pcs-xpcs.c +++ b/drivers/net/pcs/pcs-xpcs.c @@ -684,7 +684,9 @@ static int xpcs_config_aneg_c37_sgmii(struct dw_xpcs *xpcs, if (ret < 0) return ret; - mask = DW_VR_MII_DIG_CTRL1_MAC_AUTO_SW; + val = 0; + mask = DW_VR_MII_DIG_CTRL1_2G5_EN | DW_VR_MII_DIG_CTRL1_MAC_AUTO_SW; + if (neg_mode == PHYLINK_PCS_NEG_INBAND_ENABLED) val = DW_VR_MII_DIG_CTRL1_MAC_AUTO_SW; diff --git a/drivers/net/pfcp.c b/drivers/net/pfcp.c index 69434fd13f96..68d0d9e92a22 100644 --- a/drivers/net/pfcp.c +++ b/drivers/net/pfcp.c @@ -206,8 +206,8 @@ static int pfcp_newlink(struct net *net, struct net_device *dev, goto exit_del_pfcp_sock; } - pn = net_generic(dev_net(dev), pfcp_net_id); - list_add_rcu(&pfcp->list, &pn->pfcp_dev_list); + pn = net_generic(net, pfcp_net_id); + list_add(&pfcp->list, &pn->pfcp_dev_list); netdev_dbg(dev, "registered new PFCP interface\n"); @@ -224,7 +224,7 @@ static void pfcp_dellink(struct net_device *dev, struct list_head *head) { struct pfcp_dev *pfcp = netdev_priv(dev); - list_del_rcu(&pfcp->list); + list_del(&pfcp->list); unregister_netdevice_queue(dev, head); } @@ -247,11 +247,16 @@ static int __net_init pfcp_net_init(struct net *net) static void __net_exit pfcp_net_exit(struct net *net) { struct pfcp_net *pn = net_generic(net, pfcp_net_id); - struct pfcp_dev *pfcp; + struct pfcp_dev *pfcp, *pfcp_next; + struct net_device *dev; LIST_HEAD(list); rtnl_lock(); - list_for_each_entry(pfcp, &pn->pfcp_dev_list, list) + for_each_netdev(net, dev) + if (dev->rtnl_link_ops == &pfcp_link_ops) + pfcp_dellink(dev, &list); + + list_for_each_entry_safe(pfcp, pfcp_next, &pn->pfcp_dev_list, list) pfcp_dellink(pfcp->dev, &list); unregister_netdevice_many(&list); diff --git a/drivers/net/phy/micrel.c b/drivers/net/phy/micrel.c index 3ef508840674..eeb33eb181ac 100644 --- a/drivers/net/phy/micrel.c +++ b/drivers/net/phy/micrel.c @@ -432,10 +432,12 @@ struct kszphy_ptp_priv { struct kszphy_priv { struct kszphy_ptp_priv ptp_priv; const struct kszphy_type *type; + struct clk *clk; int led_mode; u16 vct_ctrl1000; bool rmii_ref_clk_sel; bool rmii_ref_clk_sel_val; + bool clk_enable; u64 stats[ARRAY_SIZE(kszphy_hw_stats)]; }; @@ -2050,6 +2052,46 @@ static void kszphy_get_stats(struct phy_device *phydev, data[i] = kszphy_get_stat(phydev, i); } +static void kszphy_enable_clk(struct phy_device *phydev) +{ + struct kszphy_priv *priv = phydev->priv; + + if (!priv->clk_enable && priv->clk) { + clk_prepare_enable(priv->clk); + priv->clk_enable = true; + } +} + +static void kszphy_disable_clk(struct phy_device *phydev) +{ + struct kszphy_priv *priv = phydev->priv; + + if (priv->clk_enable && priv->clk) { + clk_disable_unprepare(priv->clk); + priv->clk_enable = false; + } +} + +static int kszphy_generic_resume(struct phy_device *phydev) +{ + kszphy_enable_clk(phydev); + + return genphy_resume(phydev); +} + +static int kszphy_generic_suspend(struct phy_device *phydev) +{ + int ret; + + ret = genphy_suspend(phydev); + if (ret) + return ret; + + kszphy_disable_clk(phydev); + + return 0; +} + static int kszphy_suspend(struct phy_device *phydev) { /* Disable PHY Interrupts */ @@ -2059,7 +2101,7 @@ static int kszphy_suspend(struct phy_device *phydev) phydev->drv->config_intr(phydev); } - return genphy_suspend(phydev); + return kszphy_generic_suspend(phydev); } static void kszphy_parse_led_mode(struct phy_device *phydev) @@ -2090,7 +2132,9 @@ static int kszphy_resume(struct phy_device *phydev) { int ret; - genphy_resume(phydev); + ret = kszphy_generic_resume(phydev); + if (ret) + return ret; /* After switching from power-down to normal mode, an internal global * reset is automatically generated. Wait a minimum of 1 ms before @@ -2112,6 +2156,24 @@ static int kszphy_resume(struct phy_device *phydev) return 0; } +/* Because of errata DS80000700A, receiver error following software + * power down. Suspend and resume callbacks only disable and enable + * external rmii reference clock. + */ +static int ksz8041_resume(struct phy_device *phydev) +{ + kszphy_enable_clk(phydev); + + return 0; +} + +static int ksz8041_suspend(struct phy_device *phydev) +{ + kszphy_disable_clk(phydev); + + return 0; +} + static int ksz9477_resume(struct phy_device *phydev) { int ret; @@ -2159,7 +2221,10 @@ static int ksz8061_resume(struct phy_device *phydev) if (!(ret & BMCR_PDOWN)) return 0; - genphy_resume(phydev); + ret = kszphy_generic_resume(phydev); + if (ret) + return ret; + usleep_range(1000, 2000); /* Re-program the value after chip is reset. */ @@ -2177,6 +2242,11 @@ static int ksz8061_resume(struct phy_device *phydev) return 0; } +static int ksz8061_suspend(struct phy_device *phydev) +{ + return kszphy_suspend(phydev); +} + static int kszphy_probe(struct phy_device *phydev) { const struct kszphy_type *type = phydev->drv->driver_data; @@ -2217,10 +2287,14 @@ static int kszphy_probe(struct phy_device *phydev) } else if (!clk) { /* unnamed clock from the generic ethernet-phy binding */ clk = devm_clk_get_optional_enabled(&phydev->mdio.dev, NULL); - if (IS_ERR(clk)) - return PTR_ERR(clk); } + if (IS_ERR(clk)) + return PTR_ERR(clk); + + clk_disable_unprepare(clk); + priv->clk = clk; + if (ksz8041_fiber_mode(phydev)) phydev->port = PORT_FIBRE; @@ -5290,6 +5364,21 @@ static int lan8841_probe(struct phy_device *phydev) return 0; } +static int lan8804_resume(struct phy_device *phydev) +{ + return kszphy_resume(phydev); +} + +static int lan8804_suspend(struct phy_device *phydev) +{ + return kszphy_generic_suspend(phydev); +} + +static int lan8841_resume(struct phy_device *phydev) +{ + return kszphy_generic_resume(phydev); +} + static int lan8841_suspend(struct phy_device *phydev) { struct kszphy_priv *priv = phydev->priv; @@ -5298,7 +5387,7 @@ static int lan8841_suspend(struct phy_device *phydev) if (ptp_priv->ptp_clock) ptp_cancel_worker_sync(ptp_priv->ptp_clock); - return genphy_suspend(phydev); + return kszphy_generic_suspend(phydev); } static struct phy_driver ksphy_driver[] = { @@ -5358,9 +5447,8 @@ static struct phy_driver ksphy_driver[] = { .get_sset_count = kszphy_get_sset_count, .get_strings = kszphy_get_strings, .get_stats = kszphy_get_stats, - /* No suspend/resume callbacks because of errata DS80000700A, - * receiver error following software power down. - */ + .suspend = ksz8041_suspend, + .resume = ksz8041_resume, }, { .phy_id = PHY_ID_KSZ8041RNLI, .phy_id_mask = MICREL_PHY_ID_MASK, @@ -5436,7 +5524,7 @@ static struct phy_driver ksphy_driver[] = { .soft_reset = genphy_soft_reset, .config_intr = kszphy_config_intr, .handle_interrupt = kszphy_handle_interrupt, - .suspend = kszphy_suspend, + .suspend = ksz8061_suspend, .resume = ksz8061_resume, }, { .phy_id = PHY_ID_KSZ9021, @@ -5507,8 +5595,8 @@ static struct phy_driver ksphy_driver[] = { .get_sset_count = kszphy_get_sset_count, .get_strings = kszphy_get_strings, .get_stats = kszphy_get_stats, - .suspend = genphy_suspend, - .resume = kszphy_resume, + .suspend = lan8804_suspend, + .resume = lan8804_resume, .config_intr = lan8804_config_intr, .handle_interrupt = lan8804_handle_interrupt, }, { @@ -5526,7 +5614,7 @@ static struct phy_driver ksphy_driver[] = { .get_strings = kszphy_get_strings, .get_stats = kszphy_get_stats, .suspend = lan8841_suspend, - .resume = genphy_resume, + .resume = lan8841_resume, .cable_test_start = lan8814_cable_test_start, .cable_test_get_status = ksz886x_cable_test_get_status, }, { diff --git a/drivers/net/pse-pd/tps23881.c b/drivers/net/pse-pd/tps23881.c index 5c4e88be46ee..8797ca1a8a21 100644 --- a/drivers/net/pse-pd/tps23881.c +++ b/drivers/net/pse-pd/tps23881.c @@ -64,15 +64,11 @@ static int tps23881_pi_enable(struct pse_controller_dev *pcdev, int id) if (id >= TPS23881_MAX_CHANS) return -ERANGE; - ret = i2c_smbus_read_word_data(client, TPS23881_REG_PW_STATUS); - if (ret < 0) - return ret; - chan = priv->port[id].chan[0]; if (chan < 4) - val = (u16)(ret | BIT(chan)); + val = BIT(chan); else - val = (u16)(ret | BIT(chan + 4)); + val = BIT(chan + 4); if (priv->port[id].is_4p) { chan = priv->port[id].chan[1]; @@ -100,15 +96,11 @@ static int tps23881_pi_disable(struct pse_controller_dev *pcdev, int id) if (id >= TPS23881_MAX_CHANS) return -ERANGE; - ret = i2c_smbus_read_word_data(client, TPS23881_REG_PW_STATUS); - if (ret < 0) - return ret; - chan = priv->port[id].chan[0]; if (chan < 4) - val = (u16)(ret | BIT(chan + 4)); + val = BIT(chan + 4); else - val = (u16)(ret | BIT(chan + 8)); + val = BIT(chan + 8); if (priv->port[id].is_4p) { chan = priv->port[id].chan[1]; diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/bz.c b/drivers/net/wireless/intel/iwlwifi/cfg/bz.c index cd1fe8490ae5..1c43f283ac4a 100644 --- a/drivers/net/wireless/intel/iwlwifi/cfg/bz.c +++ b/drivers/net/wireless/intel/iwlwifi/cfg/bz.c @@ -161,6 +161,7 @@ const struct iwl_cfg_trans_params iwl_gl_trans_cfg = { const char iwl_bz_name[] = "Intel(R) TBD Bz device"; const char iwl_fm_name[] = "Intel(R) Wi-Fi 7 BE201 320MHz"; +const char iwl_wh_name[] = "Intel(R) Wi-Fi 7 BE211 320MHz"; const char iwl_gl_name[] = "Intel(R) Wi-Fi 7 BE200 320MHz"; const char iwl_mtp_name[] = "Intel(R) Wi-Fi 7 BE202 160MHz"; diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-config.h b/drivers/net/wireless/intel/iwlwifi/iwl-config.h index 34c91deca57b..17721bb47e25 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-config.h +++ b/drivers/net/wireless/intel/iwlwifi/iwl-config.h @@ -545,6 +545,7 @@ extern const char iwl_ax231_name[]; extern const char iwl_ax411_name[]; extern const char iwl_bz_name[]; extern const char iwl_fm_name[]; +extern const char iwl_wh_name[]; extern const char iwl_gl_name[]; extern const char iwl_mtp_name[]; extern const char iwl_sc_name[]; diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/d3.c b/drivers/net/wireless/intel/iwlwifi/mvm/d3.c index f85c01e04ebf..7d973546c9fb 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/d3.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/d3.c @@ -2954,6 +2954,7 @@ static void iwl_mvm_query_set_freqs(struct iwl_mvm *mvm, int idx) { int i; + int n_channels = 0; if (fw_has_api(&mvm->fw->ucode_capa, IWL_UCODE_TLV_API_SCAN_OFFLOAD_CHANS)) { @@ -2962,7 +2963,7 @@ static void iwl_mvm_query_set_freqs(struct iwl_mvm *mvm, for (i = 0; i < SCAN_OFFLOAD_MATCHING_CHANNELS_LEN * 8; i++) if (matches[idx].matching_channels[i / 8] & (BIT(i % 8))) - match->channels[match->n_channels++] = + match->channels[n_channels++] = mvm->nd_channels[i]->center_freq; } else { struct iwl_scan_offload_profile_match_v1 *matches = @@ -2970,9 +2971,11 @@ static void iwl_mvm_query_set_freqs(struct iwl_mvm *mvm, for (i = 0; i < SCAN_OFFLOAD_MATCHING_CHANNELS_LEN_V1 * 8; i++) if (matches[idx].matching_channels[i / 8] & (BIT(i % 8))) - match->channels[match->n_channels++] = + match->channels[n_channels++] = mvm->nd_channels[i]->center_freq; } + /* We may have ended up with fewer channels than we allocated. */ + match->n_channels = n_channels; } /** @@ -3053,6 +3056,8 @@ static void iwl_mvm_query_netdetect_reasons(struct iwl_mvm *mvm, GFP_KERNEL); if (!net_detect || !n_matches) goto out_report_nd; + net_detect->n_matches = n_matches; + n_matches = 0; for_each_set_bit(i, &matched_profiles, mvm->n_nd_match_sets) { struct cfg80211_wowlan_nd_match *match; @@ -3066,8 +3071,9 @@ static void iwl_mvm_query_netdetect_reasons(struct iwl_mvm *mvm, GFP_KERNEL); if (!match) goto out_report_nd; + match->n_channels = n_channels; - net_detect->matches[net_detect->n_matches++] = match; + net_detect->matches[n_matches++] = match; /* We inverted the order of the SSIDs in the scan * request, so invert the index here. @@ -3082,6 +3088,8 @@ static void iwl_mvm_query_netdetect_reasons(struct iwl_mvm *mvm, iwl_mvm_query_set_freqs(mvm, d3_data->nd_results, match, i); } + /* We may have fewer matches than we allocated. */ + net_detect->n_matches = n_matches; out_report_nd: wakeup.net_detect = net_detect; diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c index 805fb249a0c6..8fb2aa282242 100644 --- a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c +++ b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c @@ -1106,19 +1106,54 @@ VISIBLE_IF_IWLWIFI_KUNIT const struct iwl_dev_info iwl_dev_info_table[] = { iwlax210_2ax_cfg_so_jf_b0, iwl9462_name), /* Bz */ -/* FIXME: need to change the naming according to the actual CRF */ _IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_MAC_TYPE_BZ, IWL_CFG_ANY, + IWL_CFG_RF_TYPE_HR2, IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_ANY, + iwl_cfg_bz, iwl_ax201_name), + + _IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY, + IWL_CFG_MAC_TYPE_BZ, IWL_CFG_ANY, + IWL_CFG_RF_TYPE_GF, IWL_CFG_ANY, IWL_CFG_ANY, + IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_ANY, + iwl_cfg_bz, iwl_ax211_name), + + _IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY, + IWL_CFG_MAC_TYPE_BZ, IWL_CFG_ANY, + IWL_CFG_RF_TYPE_FM, IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_ANY, iwl_cfg_bz, iwl_fm_name), _IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY, + IWL_CFG_MAC_TYPE_BZ, IWL_CFG_ANY, + IWL_CFG_RF_TYPE_WH, IWL_CFG_ANY, IWL_CFG_ANY, + IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_ANY, + iwl_cfg_bz, iwl_wh_name), + + _IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_MAC_TYPE_BZ_W, IWL_CFG_ANY, + IWL_CFG_RF_TYPE_HR2, IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_ANY, + iwl_cfg_bz, iwl_ax201_name), + + _IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY, + IWL_CFG_MAC_TYPE_BZ_W, IWL_CFG_ANY, + IWL_CFG_RF_TYPE_GF, IWL_CFG_ANY, IWL_CFG_ANY, + IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_ANY, + iwl_cfg_bz, iwl_ax211_name), + + _IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY, + IWL_CFG_MAC_TYPE_BZ_W, IWL_CFG_ANY, + IWL_CFG_RF_TYPE_FM, IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_ANY, iwl_cfg_bz, iwl_fm_name), + _IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY, + IWL_CFG_MAC_TYPE_BZ_W, IWL_CFG_ANY, + IWL_CFG_RF_TYPE_WH, IWL_CFG_ANY, IWL_CFG_ANY, + IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_ANY, + iwl_cfg_bz, iwl_wh_name), + /* Ga (Gl) */ _IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_MAC_TYPE_GL, IWL_CFG_ANY, diff --git a/drivers/net/wireless/st/cw1200/cw1200_spi.c b/drivers/net/wireless/st/cw1200/cw1200_spi.c index 862964a8cc87..52386dfb5f4a 100644 --- a/drivers/net/wireless/st/cw1200/cw1200_spi.c +++ b/drivers/net/wireless/st/cw1200/cw1200_spi.c @@ -442,8 +442,8 @@ static void cw1200_spi_disconnect(struct spi_device *func) cw1200_core_release(self->core); self->core = NULL; } + cw1200_spi_off(self, dev_get_platdata(&func->dev)); } - cw1200_spi_off(self, dev_get_platdata(&func->dev)); } static int __maybe_unused cw1200_spi_suspend(struct device *dev) diff --git a/drivers/net/wwan/iosm/iosm_ipc_mmio.c b/drivers/net/wwan/iosm/iosm_ipc_mmio.c index 63eb08c43c05..6764c13530b9 100644 --- a/drivers/net/wwan/iosm/iosm_ipc_mmio.c +++ b/drivers/net/wwan/iosm/iosm_ipc_mmio.c @@ -104,7 +104,7 @@ struct iosm_mmio *ipc_mmio_init(void __iomem *mmio, struct device *dev) break; msleep(20); - } while (retries-- > 0); + } while (--retries > 0); if (!retries) { dev_err(ipc_mmio->dev, "invalid exec stage %X", stage); diff --git a/drivers/net/wwan/t7xx/t7xx_state_monitor.c b/drivers/net/wwan/t7xx/t7xx_state_monitor.c index 3931c7a13f5a..cbdbb91e8381 100644 --- a/drivers/net/wwan/t7xx/t7xx_state_monitor.c +++ b/drivers/net/wwan/t7xx/t7xx_state_monitor.c @@ -104,14 +104,21 @@ void t7xx_fsm_broadcast_state(struct t7xx_fsm_ctl *ctl, enum md_state state) fsm_state_notify(ctl->md, state); } +static void fsm_release_command(struct kref *ref) +{ + struct t7xx_fsm_command *cmd = container_of(ref, typeof(*cmd), refcnt); + + kfree(cmd); +} + static void fsm_finish_command(struct t7xx_fsm_ctl *ctl, struct t7xx_fsm_command *cmd, int result) { if (cmd->flag & FSM_CMD_FLAG_WAIT_FOR_COMPLETION) { - *cmd->ret = result; - complete_all(cmd->done); + cmd->result = result; + complete_all(&cmd->done); } - kfree(cmd); + kref_put(&cmd->refcnt, fsm_release_command); } static void fsm_del_kf_event(struct t7xx_fsm_event *event) @@ -475,7 +482,6 @@ static int fsm_main_thread(void *data) int t7xx_fsm_append_cmd(struct t7xx_fsm_ctl *ctl, enum t7xx_fsm_cmd_state cmd_id, unsigned int flag) { - DECLARE_COMPLETION_ONSTACK(done); struct t7xx_fsm_command *cmd; unsigned long flags; int ret; @@ -487,11 +493,13 @@ int t7xx_fsm_append_cmd(struct t7xx_fsm_ctl *ctl, enum t7xx_fsm_cmd_state cmd_id INIT_LIST_HEAD(&cmd->entry); cmd->cmd_id = cmd_id; cmd->flag = flag; + kref_init(&cmd->refcnt); if (flag & FSM_CMD_FLAG_WAIT_FOR_COMPLETION) { - cmd->done = &done; - cmd->ret = &ret; + init_completion(&cmd->done); + kref_get(&cmd->refcnt); } + kref_get(&cmd->refcnt); spin_lock_irqsave(&ctl->command_lock, flags); list_add_tail(&cmd->entry, &ctl->command_queue); spin_unlock_irqrestore(&ctl->command_lock, flags); @@ -501,11 +509,11 @@ int t7xx_fsm_append_cmd(struct t7xx_fsm_ctl *ctl, enum t7xx_fsm_cmd_state cmd_id if (flag & FSM_CMD_FLAG_WAIT_FOR_COMPLETION) { unsigned long wait_ret; - wait_ret = wait_for_completion_timeout(&done, + wait_ret = wait_for_completion_timeout(&cmd->done, msecs_to_jiffies(FSM_CMD_TIMEOUT_MS)); - if (!wait_ret) - return -ETIMEDOUT; + ret = wait_ret ? cmd->result : -ETIMEDOUT; + kref_put(&cmd->refcnt, fsm_release_command); return ret; } diff --git a/drivers/net/wwan/t7xx/t7xx_state_monitor.h b/drivers/net/wwan/t7xx/t7xx_state_monitor.h index 7b0a9baf488c..6e0601bb752e 100644 --- a/drivers/net/wwan/t7xx/t7xx_state_monitor.h +++ b/drivers/net/wwan/t7xx/t7xx_state_monitor.h @@ -110,8 +110,9 @@ struct t7xx_fsm_command { struct list_head entry; enum t7xx_fsm_cmd_state cmd_id; unsigned int flag; - struct completion *done; - int *ret; + struct completion done; + int result; + struct kref refcnt; }; struct t7xx_fsm_notifier { diff --git a/drivers/nvme/host/nvme.h b/drivers/nvme/host/nvme.h index 2c76afd00390..7be92d07430e 100644 --- a/drivers/nvme/host/nvme.h +++ b/drivers/nvme/host/nvme.h @@ -173,6 +173,11 @@ enum nvme_quirks { * MSI (but not MSI-X) interrupts are broken and never fire. */ NVME_QUIRK_BROKEN_MSI = (1 << 21), + + /* + * Align dma pool segment size to 512 bytes + */ + NVME_QUIRK_DMAPOOL_ALIGN_512 = (1 << 22), }; /* diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c index fe0795e16e25..278bed4e35bb 100644 --- a/drivers/nvme/host/pci.c +++ b/drivers/nvme/host/pci.c @@ -2833,15 +2833,20 @@ static int nvme_disable_prepare_reset(struct nvme_dev *dev, bool shutdown) static int nvme_setup_prp_pools(struct nvme_dev *dev) { + size_t small_align = 256; + dev->prp_page_pool = dma_pool_create("prp list page", dev->dev, NVME_CTRL_PAGE_SIZE, NVME_CTRL_PAGE_SIZE, 0); if (!dev->prp_page_pool) return -ENOMEM; + if (dev->ctrl.quirks & NVME_QUIRK_DMAPOOL_ALIGN_512) + small_align = 512; + /* Optimisation for I/Os between 4k and 128k */ dev->prp_small_pool = dma_pool_create("prp list 256", dev->dev, - 256, 256, 0); + 256, small_align, 0); if (!dev->prp_small_pool) { dma_pool_destroy(dev->prp_page_pool); return -ENOMEM; @@ -3606,7 +3611,7 @@ static const struct pci_device_id nvme_id_table[] = { { PCI_VDEVICE(REDHAT, 0x0010), /* Qemu emulated controller */ .driver_data = NVME_QUIRK_BOGUS_NID, }, { PCI_DEVICE(0x1217, 0x8760), /* O2 Micro 64GB Steam Deck */ - .driver_data = NVME_QUIRK_QDEPTH_ONE }, + .driver_data = NVME_QUIRK_DMAPOOL_ALIGN_512, }, { PCI_DEVICE(0x126f, 0x2262), /* Silicon Motion generic */ .driver_data = NVME_QUIRK_NO_DEEPEST_PS | NVME_QUIRK_BOGUS_NID, }, diff --git a/drivers/nvme/host/tcp.c b/drivers/nvme/host/tcp.c index dc5bbca58c6d..841238f38fdd 100644 --- a/drivers/nvme/host/tcp.c +++ b/drivers/nvme/host/tcp.c @@ -2064,14 +2064,6 @@ static int nvme_tcp_alloc_io_queues(struct nvme_ctrl *ctrl) return __nvme_tcp_alloc_io_queues(ctrl); } -static void nvme_tcp_destroy_io_queues(struct nvme_ctrl *ctrl, bool remove) -{ - nvme_tcp_stop_io_queues(ctrl); - if (remove) - nvme_remove_io_tag_set(ctrl); - nvme_tcp_free_io_queues(ctrl); -} - static int nvme_tcp_configure_io_queues(struct nvme_ctrl *ctrl, bool new) { int ret, nr_queues; @@ -2216,9 +2208,11 @@ static void nvme_tcp_teardown_io_queues(struct nvme_ctrl *ctrl, nvme_sync_io_queues(ctrl); nvme_tcp_stop_io_queues(ctrl); nvme_cancel_tagset(ctrl); - if (remove) + if (remove) { nvme_unquiesce_io_queues(ctrl); - nvme_tcp_destroy_io_queues(ctrl, remove); + nvme_remove_io_tag_set(ctrl); + } + nvme_tcp_free_io_queues(ctrl); } static void nvme_tcp_reconnect_or_remove(struct nvme_ctrl *ctrl, @@ -2307,7 +2301,9 @@ destroy_io: nvme_sync_io_queues(ctrl); nvme_tcp_stop_io_queues(ctrl); nvme_cancel_tagset(ctrl); - nvme_tcp_destroy_io_queues(ctrl, new); + if (new) + nvme_remove_io_tag_set(ctrl); + nvme_tcp_free_io_queues(ctrl); } destroy_admin: nvme_stop_keep_alive(ctrl); diff --git a/drivers/nvme/target/admin-cmd.c b/drivers/nvme/target/admin-cmd.c index 3ddd8e44e148..e670dc185a96 100644 --- a/drivers/nvme/target/admin-cmd.c +++ b/drivers/nvme/target/admin-cmd.c @@ -275,7 +275,7 @@ static u16 nvmet_get_smart_log_all(struct nvmet_req *req, unsigned long idx; ctrl = req->sq->ctrl; - xa_for_each(&ctrl->subsys->namespaces, idx, ns) { + nvmet_for_each_enabled_ns(&ctrl->subsys->namespaces, idx, ns) { /* we don't have the right data for file backed ns */ if (!ns->bdev) continue; @@ -483,9 +483,10 @@ static u32 nvmet_format_ana_group(struct nvmet_req *req, u32 grpid, u32 count = 0; if (!(req->cmd->get_log_page.lsp & NVME_ANA_LOG_RGO)) { - xa_for_each(&ctrl->subsys->namespaces, idx, ns) + nvmet_for_each_enabled_ns(&ctrl->subsys->namespaces, idx, ns) { if (ns->anagrpid == grpid) desc->nsids[count++] = cpu_to_le32(ns->nsid); + } } desc->grpid = cpu_to_le32(grpid); @@ -925,7 +926,7 @@ static void nvmet_execute_identify_endgrp_list(struct nvmet_req *req) goto out; } - xa_for_each(&ctrl->subsys->namespaces, idx, ns) { + nvmet_for_each_enabled_ns(&ctrl->subsys->namespaces, idx, ns) { if (ns->nsid <= min_endgid) continue; @@ -968,7 +969,7 @@ static void nvmet_execute_identify_nslist(struct nvmet_req *req, bool match_css) goto out; } - xa_for_each(&ctrl->subsys->namespaces, idx, ns) { + nvmet_for_each_enabled_ns(&ctrl->subsys->namespaces, idx, ns) { if (ns->nsid <= min_nsid) continue; if (match_css && req->ns->csi != req->cmd->identify.csi) diff --git a/drivers/nvme/target/configfs.c b/drivers/nvme/target/configfs.c index 20cad722c060..e44ef69dffc2 100644 --- a/drivers/nvme/target/configfs.c +++ b/drivers/nvme/target/configfs.c @@ -812,18 +812,6 @@ static struct configfs_attribute *nvmet_ns_attrs[] = { NULL, }; -bool nvmet_subsys_nsid_exists(struct nvmet_subsys *subsys, u32 nsid) -{ - struct config_item *ns_item; - char name[12]; - - snprintf(name, sizeof(name), "%u", nsid); - mutex_lock(&subsys->namespaces_group.cg_subsys->su_mutex); - ns_item = config_group_find_item(&subsys->namespaces_group, name); - mutex_unlock(&subsys->namespaces_group.cg_subsys->su_mutex); - return ns_item != NULL; -} - static void nvmet_ns_release(struct config_item *item) { struct nvmet_ns *ns = to_nvmet_ns(item); @@ -2303,12 +2291,17 @@ static ssize_t nvmet_root_discovery_nqn_store(struct config_item *item, const char *page, size_t count) { struct list_head *entry; + char *old_nqn, *new_nqn; size_t len; len = strcspn(page, "\n"); if (!len || len > NVMF_NQN_FIELD_LEN - 1) return -EINVAL; + new_nqn = kstrndup(page, len, GFP_KERNEL); + if (!new_nqn) + return -ENOMEM; + down_write(&nvmet_config_sem); list_for_each(entry, &nvmet_subsystems_group.cg_children) { struct config_item *item = @@ -2317,13 +2310,15 @@ static ssize_t nvmet_root_discovery_nqn_store(struct config_item *item, if (!strncmp(config_item_name(item), page, len)) { pr_err("duplicate NQN %s\n", config_item_name(item)); up_write(&nvmet_config_sem); + kfree(new_nqn); return -EINVAL; } } - memset(nvmet_disc_subsys->subsysnqn, 0, NVMF_NQN_FIELD_LEN); - memcpy(nvmet_disc_subsys->subsysnqn, page, len); + old_nqn = nvmet_disc_subsys->subsysnqn; + nvmet_disc_subsys->subsysnqn = new_nqn; up_write(&nvmet_config_sem); + kfree(old_nqn); return len; } diff --git a/drivers/nvme/target/core.c b/drivers/nvme/target/core.c index 43c9888eea90..cdc4a09a6e8a 100644 --- a/drivers/nvme/target/core.c +++ b/drivers/nvme/target/core.c @@ -127,7 +127,7 @@ static u32 nvmet_max_nsid(struct nvmet_subsys *subsys) unsigned long idx; u32 nsid = 0; - xa_for_each(&subsys->namespaces, idx, cur) + nvmet_for_each_enabled_ns(&subsys->namespaces, idx, cur) nsid = cur->nsid; return nsid; @@ -441,11 +441,14 @@ u16 nvmet_req_find_ns(struct nvmet_req *req) struct nvmet_subsys *subsys = nvmet_req_subsys(req); req->ns = xa_load(&subsys->namespaces, nsid); - if (unlikely(!req->ns)) { + if (unlikely(!req->ns || !req->ns->enabled)) { req->error_loc = offsetof(struct nvme_common_command, nsid); - if (nvmet_subsys_nsid_exists(subsys, nsid)) - return NVME_SC_INTERNAL_PATH_ERROR; - return NVME_SC_INVALID_NS | NVME_STATUS_DNR; + if (!req->ns) /* ns doesn't exist! */ + return NVME_SC_INVALID_NS | NVME_STATUS_DNR; + + /* ns exists but it's disabled */ + req->ns = NULL; + return NVME_SC_INTERNAL_PATH_ERROR; } percpu_ref_get(&req->ns->ref); @@ -583,8 +586,6 @@ int nvmet_ns_enable(struct nvmet_ns *ns) goto out_unlock; ret = -EMFILE; - if (subsys->nr_namespaces == NVMET_MAX_NAMESPACES) - goto out_unlock; ret = nvmet_bdev_ns_enable(ns); if (ret == -ENOTBLK) @@ -599,38 +600,19 @@ int nvmet_ns_enable(struct nvmet_ns *ns) list_for_each_entry(ctrl, &subsys->ctrls, subsys_entry) nvmet_p2pmem_ns_add_p2p(ctrl, ns); - ret = percpu_ref_init(&ns->ref, nvmet_destroy_namespace, - 0, GFP_KERNEL); - if (ret) - goto out_dev_put; - - if (ns->nsid > subsys->max_nsid) - subsys->max_nsid = ns->nsid; - - ret = xa_insert(&subsys->namespaces, ns->nsid, ns, GFP_KERNEL); - if (ret) - goto out_restore_subsys_maxnsid; - if (ns->pr.enable) { ret = nvmet_pr_init_ns(ns); if (ret) - goto out_remove_from_subsys; + goto out_dev_put; } - subsys->nr_namespaces++; - nvmet_ns_changed(subsys, ns->nsid); ns->enabled = true; + xa_set_mark(&subsys->namespaces, ns->nsid, NVMET_NS_ENABLED); ret = 0; out_unlock: mutex_unlock(&subsys->lock); return ret; - -out_remove_from_subsys: - xa_erase(&subsys->namespaces, ns->nsid); -out_restore_subsys_maxnsid: - subsys->max_nsid = nvmet_max_nsid(subsys); - percpu_ref_exit(&ns->ref); out_dev_put: list_for_each_entry(ctrl, &subsys->ctrls, subsys_entry) pci_dev_put(radix_tree_delete(&ctrl->p2p_ns_map, ns->nsid)); @@ -649,15 +631,37 @@ void nvmet_ns_disable(struct nvmet_ns *ns) goto out_unlock; ns->enabled = false; - xa_erase(&ns->subsys->namespaces, ns->nsid); - if (ns->nsid == subsys->max_nsid) - subsys->max_nsid = nvmet_max_nsid(subsys); + xa_clear_mark(&subsys->namespaces, ns->nsid, NVMET_NS_ENABLED); list_for_each_entry(ctrl, &subsys->ctrls, subsys_entry) pci_dev_put(radix_tree_delete(&ctrl->p2p_ns_map, ns->nsid)); mutex_unlock(&subsys->lock); + if (ns->pr.enable) + nvmet_pr_exit_ns(ns); + + mutex_lock(&subsys->lock); + nvmet_ns_changed(subsys, ns->nsid); + nvmet_ns_dev_disable(ns); +out_unlock: + mutex_unlock(&subsys->lock); +} + +void nvmet_ns_free(struct nvmet_ns *ns) +{ + struct nvmet_subsys *subsys = ns->subsys; + + nvmet_ns_disable(ns); + + mutex_lock(&subsys->lock); + + xa_erase(&subsys->namespaces, ns->nsid); + if (ns->nsid == subsys->max_nsid) + subsys->max_nsid = nvmet_max_nsid(subsys); + + mutex_unlock(&subsys->lock); + /* * Now that we removed the namespaces from the lookup list, we * can kill the per_cpu ref and wait for any remaining references @@ -671,21 +675,9 @@ void nvmet_ns_disable(struct nvmet_ns *ns) wait_for_completion(&ns->disable_done); percpu_ref_exit(&ns->ref); - if (ns->pr.enable) - nvmet_pr_exit_ns(ns); - mutex_lock(&subsys->lock); - subsys->nr_namespaces--; - nvmet_ns_changed(subsys, ns->nsid); - nvmet_ns_dev_disable(ns); -out_unlock: mutex_unlock(&subsys->lock); -} - -void nvmet_ns_free(struct nvmet_ns *ns) -{ - nvmet_ns_disable(ns); down_write(&nvmet_ana_sem); nvmet_ana_group_enabled[ns->anagrpid]--; @@ -699,15 +691,33 @@ struct nvmet_ns *nvmet_ns_alloc(struct nvmet_subsys *subsys, u32 nsid) { struct nvmet_ns *ns; + mutex_lock(&subsys->lock); + + if (subsys->nr_namespaces == NVMET_MAX_NAMESPACES) + goto out_unlock; + ns = kzalloc(sizeof(*ns), GFP_KERNEL); if (!ns) - return NULL; + goto out_unlock; init_completion(&ns->disable_done); ns->nsid = nsid; ns->subsys = subsys; + if (percpu_ref_init(&ns->ref, nvmet_destroy_namespace, 0, GFP_KERNEL)) + goto out_free; + + if (ns->nsid > subsys->max_nsid) + subsys->max_nsid = nsid; + + if (xa_insert(&subsys->namespaces, ns->nsid, ns, GFP_KERNEL)) + goto out_exit; + + subsys->nr_namespaces++; + + mutex_unlock(&subsys->lock); + down_write(&nvmet_ana_sem); ns->anagrpid = NVMET_DEFAULT_ANA_GRPID; nvmet_ana_group_enabled[ns->anagrpid]++; @@ -718,6 +728,14 @@ struct nvmet_ns *nvmet_ns_alloc(struct nvmet_subsys *subsys, u32 nsid) ns->csi = NVME_CSI_NVM; return ns; +out_exit: + subsys->max_nsid = nvmet_max_nsid(subsys); + percpu_ref_exit(&ns->ref); +out_free: + kfree(ns); +out_unlock: + mutex_unlock(&subsys->lock); + return NULL; } static void nvmet_update_sq_head(struct nvmet_req *req) @@ -1495,7 +1513,7 @@ static void nvmet_setup_p2p_ns_map(struct nvmet_ctrl *ctrl, ctrl->p2p_client = get_device(p2p_client); - xa_for_each(&ctrl->subsys->namespaces, idx, ns) + nvmet_for_each_enabled_ns(&ctrl->subsys->namespaces, idx, ns) nvmet_p2pmem_ns_add_p2p(ctrl, ns); } diff --git a/drivers/nvme/target/io-cmd-bdev.c b/drivers/nvme/target/io-cmd-bdev.c index 6380b60fd490..c1f574fe3280 100644 --- a/drivers/nvme/target/io-cmd-bdev.c +++ b/drivers/nvme/target/io-cmd-bdev.c @@ -36,7 +36,7 @@ void nvmet_bdev_set_limits(struct block_device *bdev, struct nvme_id_ns *id) */ id->nsfeat |= 1 << 4; /* NPWG = Namespace Preferred Write Granularity. 0's based */ - id->npwg = lpp0b; + id->npwg = to0based(bdev_io_min(bdev) / bdev_logical_block_size(bdev)); /* NPWA = Namespace Preferred Write Alignment. 0's based */ id->npwa = id->npwg; /* NPDG = Namespace Preferred Deallocate Granularity. 0's based */ diff --git a/drivers/nvme/target/nvmet.h b/drivers/nvme/target/nvmet.h index f4df458df9db..b540216c0c9a 100644 --- a/drivers/nvme/target/nvmet.h +++ b/drivers/nvme/target/nvmet.h @@ -24,6 +24,7 @@ #define NVMET_DEFAULT_VS NVME_VS(2, 1, 0) +#define NVMET_NS_ENABLED XA_MARK_1 #define NVMET_ASYNC_EVENTS 4 #define NVMET_ERROR_LOG_SLOTS 128 #define NVMET_NO_ERROR_LOC ((u16)-1) @@ -33,6 +34,12 @@ #define NVMET_FR_MAX_SIZE 8 #define NVMET_PR_LOG_QUEUE_SIZE 64 +#define nvmet_for_each_ns(xa, index, entry) \ + xa_for_each(xa, index, entry) + +#define nvmet_for_each_enabled_ns(xa, index, entry) \ + xa_for_each_marked(xa, index, entry, NVMET_NS_ENABLED) + /* * Supported optional AENs: */ diff --git a/drivers/nvme/target/pr.c b/drivers/nvme/target/pr.c index 90e9f5bbe581..cd22d8333314 100644 --- a/drivers/nvme/target/pr.c +++ b/drivers/nvme/target/pr.c @@ -60,7 +60,7 @@ u16 nvmet_set_feat_resv_notif_mask(struct nvmet_req *req, u32 mask) goto success; } - xa_for_each(&ctrl->subsys->namespaces, idx, ns) { + nvmet_for_each_enabled_ns(&ctrl->subsys->namespaces, idx, ns) { if (ns->pr.enable) WRITE_ONCE(ns->pr.notify_mask, mask); } @@ -1056,7 +1056,7 @@ int nvmet_ctrl_init_pr(struct nvmet_ctrl *ctrl) * nvmet_pr_init_ns(), see more details in nvmet_ns_enable(). * So just check ns->pr.enable. */ - xa_for_each(&subsys->namespaces, idx, ns) { + nvmet_for_each_enabled_ns(&subsys->namespaces, idx, ns) { if (ns->pr.enable) { ret = nvmet_pr_alloc_and_insert_pc_ref(ns, ctrl->cntlid, &ctrl->hostid); @@ -1067,7 +1067,7 @@ int nvmet_ctrl_init_pr(struct nvmet_ctrl *ctrl) return 0; free_per_ctrl_refs: - xa_for_each(&subsys->namespaces, idx, ns) { + nvmet_for_each_enabled_ns(&subsys->namespaces, idx, ns) { if (ns->pr.enable) { pc_ref = xa_erase(&ns->pr_per_ctrl_refs, ctrl->cntlid); if (pc_ref) @@ -1087,7 +1087,7 @@ void nvmet_ctrl_destroy_pr(struct nvmet_ctrl *ctrl) kfifo_free(&ctrl->pr_log_mgr.log_queue); mutex_destroy(&ctrl->pr_log_mgr.lock); - xa_for_each(&ctrl->subsys->namespaces, idx, ns) { + nvmet_for_each_enabled_ns(&ctrl->subsys->namespaces, idx, ns) { if (ns->pr.enable) { pc_ref = xa_erase(&ns->pr_per_ctrl_refs, ctrl->cntlid); if (pc_ref) diff --git a/drivers/of/address.c b/drivers/of/address.c index c1f1c810e810..8770004d9b08 100644 --- a/drivers/of/address.c +++ b/drivers/of/address.c @@ -340,6 +340,15 @@ static int of_bus_default_flags_match(struct device_node *np) return of_property_present(np, "#address-cells") && (of_bus_n_addr_cells(np) == 3); } +static int of_bus_default_match(struct device_node *np) +{ + /* + * Check for presence first since of_bus_n_addr_cells() will warn when + * walking parent nodes. + */ + return of_property_present(np, "#address-cells"); +} + /* * Array of bus specific translators */ @@ -384,7 +393,7 @@ static const struct of_bus of_busses[] = { { .name = "default", .addresses = "reg", - .match = NULL, + .match = of_bus_default_match, .count_cells = of_bus_default_count_cells, .map = of_bus_default_map, .translate = of_bus_default_translate, @@ -399,7 +408,6 @@ static const struct of_bus *of_match_bus(struct device_node *np) for (i = 0; i < ARRAY_SIZE(of_busses); i++) if (!of_busses[i].match || of_busses[i].match(np)) return &of_busses[i]; - BUG(); return NULL; } @@ -521,6 +529,8 @@ static u64 __of_translate_address(struct device_node *node, if (parent == NULL) return OF_BAD_ADDR; bus = of_match_bus(parent); + if (!bus) + return OF_BAD_ADDR; /* Count address cells & copy address locally */ bus->count_cells(dev, &na, &ns); @@ -564,6 +574,8 @@ static u64 __of_translate_address(struct device_node *node, /* Get new parent bus and counts */ pbus = of_match_bus(parent); + if (!pbus) + return OF_BAD_ADDR; pbus->count_cells(dev, &pna, &pns); if (!OF_CHECK_COUNTS(pna, pns)) { pr_err("Bad cell count for %pOF\n", dev); @@ -703,7 +715,7 @@ const __be32 *__of_get_address(struct device_node *dev, int index, int bar_no, /* match the parent's bus type */ bus = of_match_bus(parent); - if (strcmp(bus->name, "pci") && (bar_no >= 0)) + if (!bus || (strcmp(bus->name, "pci") && (bar_no >= 0))) return NULL; /* Get "reg" or "assigned-addresses" property */ diff --git a/drivers/of/unittest-data/tests-platform.dtsi b/drivers/of/unittest-data/tests-platform.dtsi index fa39611071b3..cd310b26b50c 100644 --- a/drivers/of/unittest-data/tests-platform.dtsi +++ b/drivers/of/unittest-data/tests-platform.dtsi @@ -34,5 +34,18 @@ }; }; }; + + platform-tests-2 { + // No #address-cells or #size-cells + node { + #address-cells = <1>; + #size-cells = <1>; + + test-device@100 { + compatible = "test-sub-device"; + reg = <0x100 1>; + }; + }; + }; }; }; diff --git a/drivers/of/unittest.c b/drivers/of/unittest.c index 438fd70fa995..0fa0c0fd9a6a 100644 --- a/drivers/of/unittest.c +++ b/drivers/of/unittest.c @@ -1380,6 +1380,7 @@ static void __init of_unittest_bus_3cell_ranges(void) static void __init of_unittest_reg(void) { struct device_node *np; + struct resource res; int ret; u64 addr, size; @@ -1396,6 +1397,19 @@ static void __init of_unittest_reg(void) np, addr); of_node_put(np); + + np = of_find_node_by_path("/testcase-data/platform-tests-2/node/test-device@100"); + if (!np) { + pr_err("missing testcase data\n"); + return; + } + + ret = of_address_to_resource(np, 0, &res); + unittest(ret == -EINVAL, "of_address_to_resource(%pOF) expected error on untranslatable address\n", + np); + + of_node_put(np); + } struct of_unittest_expected_res { diff --git a/drivers/pci/msi/irqdomain.c b/drivers/pci/msi/irqdomain.c index 569125726b3e..d7ba8795d60f 100644 --- a/drivers/pci/msi/irqdomain.c +++ b/drivers/pci/msi/irqdomain.c @@ -350,8 +350,11 @@ bool pci_msi_domain_supports(struct pci_dev *pdev, unsigned int feature_mask, domain = dev_get_msi_domain(&pdev->dev); - if (!domain || !irq_domain_is_hierarchy(domain)) - return mode == ALLOW_LEGACY; + if (!domain || !irq_domain_is_hierarchy(domain)) { + if (IS_ENABLED(CONFIG_PCI_MSI_ARCH_FALLBACKS)) + return mode == ALLOW_LEGACY; + return false; + } if (!irq_domain_is_msi_parent(domain)) { /* diff --git a/drivers/pci/msi/msi.c b/drivers/pci/msi/msi.c index 3a45879d85db..2f647cac4cae 100644 --- a/drivers/pci/msi/msi.c +++ b/drivers/pci/msi/msi.c @@ -433,6 +433,10 @@ int __pci_enable_msi_range(struct pci_dev *dev, int minvec, int maxvec, if (WARN_ON_ONCE(dev->msi_enabled)) return -EINVAL; + /* Test for the availability of MSI support */ + if (!pci_msi_domain_supports(dev, 0, ALLOW_LEGACY)) + return -ENOTSUPP; + nvec = pci_msi_vec_count(dev); if (nvec < 0) return nvec; diff --git a/drivers/pci/pcie/bwctrl.c b/drivers/pci/pcie/bwctrl.c index b59cacc740fa..0a5e7efbce2c 100644 --- a/drivers/pci/pcie/bwctrl.c +++ b/drivers/pci/pcie/bwctrl.c @@ -303,14 +303,17 @@ static int pcie_bwnotif_probe(struct pcie_device *srv) if (ret) return ret; - ret = devm_request_irq(&srv->device, srv->irq, pcie_bwnotif_irq, - IRQF_SHARED, "PCIe bwctrl", srv); - if (ret) - return ret; - scoped_guard(rwsem_write, &pcie_bwctrl_setspeed_rwsem) { scoped_guard(rwsem_write, &pcie_bwctrl_lbms_rwsem) { - port->link_bwctrl = no_free_ptr(data); + port->link_bwctrl = data; + + ret = request_irq(srv->irq, pcie_bwnotif_irq, + IRQF_SHARED, "PCIe bwctrl", srv); + if (ret) { + port->link_bwctrl = NULL; + return ret; + } + pcie_bwnotif_enable(srv); } } @@ -331,11 +334,15 @@ static void pcie_bwnotif_remove(struct pcie_device *srv) pcie_cooling_device_unregister(data->cdev); - pcie_bwnotif_disable(srv->port); + scoped_guard(rwsem_write, &pcie_bwctrl_setspeed_rwsem) { + scoped_guard(rwsem_write, &pcie_bwctrl_lbms_rwsem) { + pcie_bwnotif_disable(srv->port); + + free_irq(srv->irq, srv); - scoped_guard(rwsem_write, &pcie_bwctrl_setspeed_rwsem) - scoped_guard(rwsem_write, &pcie_bwctrl_lbms_rwsem) srv->port->link_bwctrl = NULL; + } + } } static int pcie_bwnotif_suspend(struct pcie_device *srv) diff --git a/drivers/perf/riscv_pmu_sbi.c b/drivers/perf/riscv_pmu_sbi.c index 1aa303f76cc7..194c153e5d71 100644 --- a/drivers/perf/riscv_pmu_sbi.c +++ b/drivers/perf/riscv_pmu_sbi.c @@ -507,8 +507,7 @@ static int pmu_sbi_event_map(struct perf_event *event, u64 *econfig) { u32 type = event->attr.type; u64 config = event->attr.config; - u64 raw_config_val; - int ret; + int ret = -ENOENT; /* * Ensure we are finished checking standard hardware events for @@ -528,21 +527,23 @@ static int pmu_sbi_event_map(struct perf_event *event, u64 *econfig) case PERF_TYPE_RAW: /* * As per SBI specification, the upper 16 bits must be unused - * for a raw event. + * for a hardware raw event. * Bits 63:62 are used to distinguish between raw events * 00 - Hardware raw event * 10 - SBI firmware events * 11 - Risc-V platform specific firmware event */ - raw_config_val = config & RISCV_PMU_RAW_EVENT_MASK; + switch (config >> 62) { case 0: - ret = RISCV_PMU_RAW_EVENT_IDX; - *econfig = raw_config_val; + /* Return error any bits [48-63] is set as it is not allowed by the spec */ + if (!(config & ~RISCV_PMU_RAW_EVENT_MASK)) { + *econfig = config & RISCV_PMU_RAW_EVENT_MASK; + ret = RISCV_PMU_RAW_EVENT_IDX; + } break; case 2: - ret = (raw_config_val & 0xFFFF) | - (SBI_PMU_EVENT_TYPE_FW << 16); + ret = (config & 0xFFFF) | (SBI_PMU_EVENT_TYPE_FW << 16); break; case 3: /* @@ -551,12 +552,13 @@ static int pmu_sbi_event_map(struct perf_event *event, u64 *econfig) * Event data - raw event encoding */ ret = SBI_PMU_EVENT_TYPE_FW << 16 | RISCV_PLAT_FW_EVENT; - *econfig = raw_config_val; + *econfig = config & RISCV_PMU_PLAT_FW_EVENT_MASK; + break; + default: break; } break; default: - ret = -ENOENT; break; } diff --git a/drivers/phy/broadcom/phy-brcm-usb-init-synopsys.c b/drivers/phy/broadcom/phy-brcm-usb-init-synopsys.c index 950b7ae1d1a8..dc452610934a 100644 --- a/drivers/phy/broadcom/phy-brcm-usb-init-synopsys.c +++ b/drivers/phy/broadcom/phy-brcm-usb-init-synopsys.c @@ -325,6 +325,12 @@ static void usb_init_common_7216(struct brcm_usb_init_params *params) void __iomem *ctrl = params->regs[BRCM_REGS_CTRL]; USB_CTRL_UNSET(ctrl, USB_PM, XHC_S2_CLK_SWITCH_EN); + + /* + * The PHY might be in a bad state if it is already powered + * up. Toggle the power just in case. + */ + USB_CTRL_SET(ctrl, USB_PM, USB_PWRDN); USB_CTRL_UNSET(ctrl, USB_PM, USB_PWRDN); /* 1 millisecond - for USB clocks to settle down */ diff --git a/drivers/phy/freescale/phy-fsl-samsung-hdmi.c b/drivers/phy/freescale/phy-fsl-samsung-hdmi.c index 2c8038864357..d3ccf547ba1c 100644 --- a/drivers/phy/freescale/phy-fsl-samsung-hdmi.c +++ b/drivers/phy/freescale/phy-fsl-samsung-hdmi.c @@ -424,8 +424,7 @@ static unsigned long fsl_samsung_hdmi_phy_find_pms(unsigned long fout, u8 *p, u1 * Fvco = (M * f_ref) / P, * where f_ref is 24MHz. */ - tmp = (u64)_m * 24 * MHZ; - do_div(tmp, _p); + tmp = div64_ul((u64)_m * 24 * MHZ, _p); if (tmp < 750 * MHZ || tmp > 3000 * MHZ) continue; diff --git a/drivers/phy/mediatek/Kconfig b/drivers/phy/mediatek/Kconfig index 60e00057e8bc..ba6461350951 100644 --- a/drivers/phy/mediatek/Kconfig +++ b/drivers/phy/mediatek/Kconfig @@ -65,6 +65,7 @@ config PHY_MTK_HDMI depends on ARCH_MEDIATEK || COMPILE_TEST depends on COMMON_CLK depends on OF + depends on REGULATOR select GENERIC_PHY help Support HDMI PHY for Mediatek SoCs. diff --git a/drivers/phy/phy-core.c b/drivers/phy/phy-core.c index f053b525ccff..413f76e2d174 100644 --- a/drivers/phy/phy-core.c +++ b/drivers/phy/phy-core.c @@ -145,8 +145,10 @@ static struct phy_provider *of_phy_provider_lookup(struct device_node *node) return phy_provider; for_each_child_of_node(phy_provider->children, child) - if (child == node) + if (child == node) { + of_node_put(child); return phy_provider; + } } return ERR_PTR(-EPROBE_DEFER); @@ -629,8 +631,10 @@ static struct phy *_of_phy_get(struct device_node *np, int index) return ERR_PTR(-ENODEV); /* This phy type handled by the usb-phy subsystem for now */ - if (of_device_is_compatible(args.np, "usb-nop-xceiv")) - return ERR_PTR(-ENODEV); + if (of_device_is_compatible(args.np, "usb-nop-xceiv")) { + phy = ERR_PTR(-ENODEV); + goto out_put_node; + } mutex_lock(&phy_provider_mutex); phy_provider = of_phy_provider_lookup(args.np); @@ -652,6 +656,7 @@ out_put_module: out_unlock: mutex_unlock(&phy_provider_mutex); +out_put_node: of_node_put(args.np); return phy; @@ -737,7 +742,7 @@ void devm_phy_put(struct device *dev, struct phy *phy) if (!phy) return; - r = devres_destroy(dev, devm_phy_release, devm_phy_match, phy); + r = devres_release(dev, devm_phy_release, devm_phy_match, phy); dev_WARN_ONCE(dev, r, "couldn't find PHY resource\n"); } EXPORT_SYMBOL_GPL(devm_phy_put); @@ -1121,7 +1126,7 @@ void devm_phy_destroy(struct device *dev, struct phy *phy) { int r; - r = devres_destroy(dev, devm_phy_consume, devm_phy_match, phy); + r = devres_release(dev, devm_phy_consume, devm_phy_match, phy); dev_WARN_ONCE(dev, r, "couldn't find PHY resource\n"); } EXPORT_SYMBOL_GPL(devm_phy_destroy); @@ -1259,12 +1264,12 @@ EXPORT_SYMBOL_GPL(of_phy_provider_unregister); * of_phy_provider_unregister to unregister the phy provider. */ void devm_of_phy_provider_unregister(struct device *dev, - struct phy_provider *phy_provider) + struct phy_provider *phy_provider) { int r; - r = devres_destroy(dev, devm_phy_provider_release, devm_phy_match, - phy_provider); + r = devres_release(dev, devm_phy_provider_release, devm_phy_match, + phy_provider); dev_WARN_ONCE(dev, r, "couldn't find PHY provider device resource\n"); } EXPORT_SYMBOL_GPL(devm_of_phy_provider_unregister); diff --git a/drivers/phy/qualcomm/phy-qcom-qmp-usb.c b/drivers/phy/qualcomm/phy-qcom-qmp-usb.c index acd6075bf6d9..c9c337840715 100644 --- a/drivers/phy/qualcomm/phy-qcom-qmp-usb.c +++ b/drivers/phy/qualcomm/phy-qcom-qmp-usb.c @@ -1052,7 +1052,7 @@ static const struct qmp_phy_init_tbl sc8280xp_usb3_uniphy_rx_tbl[] = { QMP_PHY_INIT_CFG(QSERDES_V5_RX_UCDR_FASTLOCK_FO_GAIN, 0x2f), QMP_PHY_INIT_CFG(QSERDES_V5_RX_UCDR_FASTLOCK_COUNT_LOW, 0xff), QMP_PHY_INIT_CFG(QSERDES_V5_RX_UCDR_FASTLOCK_COUNT_HIGH, 0x0f), - QMP_PHY_INIT_CFG(QSERDES_V5_RX_UCDR_SO_GAIN, 0x0a), + QMP_PHY_INIT_CFG(QSERDES_V5_RX_UCDR_FO_GAIN, 0x0a), QMP_PHY_INIT_CFG(QSERDES_V5_RX_VGA_CAL_CNTRL1, 0x54), QMP_PHY_INIT_CFG(QSERDES_V5_RX_VGA_CAL_CNTRL2, 0x0f), QMP_PHY_INIT_CFG(QSERDES_V5_RX_RX_EQU_ADAPTOR_CNTRL2, 0x0f), diff --git a/drivers/phy/rockchip/phy-rockchip-naneng-combphy.c b/drivers/phy/rockchip/phy-rockchip-naneng-combphy.c index 0a9989e41237..2eb3329ca23f 100644 --- a/drivers/phy/rockchip/phy-rockchip-naneng-combphy.c +++ b/drivers/phy/rockchip/phy-rockchip-naneng-combphy.c @@ -309,7 +309,7 @@ static int rockchip_combphy_parse_dt(struct device *dev, struct rockchip_combphy priv->ext_refclk = device_property_present(dev, "rockchip,ext-refclk"); - priv->phy_rst = devm_reset_control_array_get_exclusive(dev); + priv->phy_rst = devm_reset_control_get(dev, "phy"); if (IS_ERR(priv->phy_rst)) return dev_err_probe(dev, PTR_ERR(priv->phy_rst), "failed to get phy reset\n"); diff --git a/drivers/phy/rockchip/phy-rockchip-samsung-hdptx.c b/drivers/phy/rockchip/phy-rockchip-samsung-hdptx.c index ceab9c71d3b5..0965b9d4f9cf 100644 --- a/drivers/phy/rockchip/phy-rockchip-samsung-hdptx.c +++ b/drivers/phy/rockchip/phy-rockchip-samsung-hdptx.c @@ -1101,6 +1101,8 @@ static int rk_hdptx_phy_probe(struct platform_device *pdev) return dev_err_probe(dev, PTR_ERR(hdptx->grf), "Could not get GRF syscon\n"); + platform_set_drvdata(pdev, hdptx); + ret = devm_pm_runtime_enable(dev); if (ret) return dev_err_probe(dev, ret, "Failed to enable runtime PM\n"); @@ -1110,7 +1112,6 @@ static int rk_hdptx_phy_probe(struct platform_device *pdev) return dev_err_probe(dev, PTR_ERR(hdptx->phy), "Failed to create HDMI PHY\n"); - platform_set_drvdata(pdev, hdptx); phy_set_drvdata(hdptx->phy, hdptx); phy_set_bus_width(hdptx->phy, 8); diff --git a/drivers/phy/st/phy-stm32-combophy.c b/drivers/phy/st/phy-stm32-combophy.c index 765bb34fe358..49e9fa90a681 100644 --- a/drivers/phy/st/phy-stm32-combophy.c +++ b/drivers/phy/st/phy-stm32-combophy.c @@ -122,6 +122,7 @@ static int stm32_impedance_tune(struct stm32_combophy *combophy) u32 max_vswing = imp_lookup[imp_size - 1].vswing[vswing_size - 1]; u32 min_vswing = imp_lookup[0].vswing[0]; u32 val; + u32 regval; if (!of_property_read_u32(combophy->dev->of_node, "st,output-micro-ohms", &val)) { if (val < min_imp || val > max_imp) { @@ -129,16 +130,20 @@ static int stm32_impedance_tune(struct stm32_combophy *combophy) return -EINVAL; } - for (imp_of = 0; imp_of < ARRAY_SIZE(imp_lookup); imp_of++) - if (imp_lookup[imp_of].microohm <= val) + regval = 0; + for (imp_of = 0; imp_of < ARRAY_SIZE(imp_lookup); imp_of++) { + if (imp_lookup[imp_of].microohm <= val) { + regval = FIELD_PREP(STM32MP25_PCIEPRG_IMPCTRL_OHM, imp_of); break; + } + } dev_dbg(combophy->dev, "Set %u micro-ohms output impedance\n", imp_lookup[imp_of].microohm); regmap_update_bits(combophy->regmap, SYSCFG_PCIEPRGCR, STM32MP25_PCIEPRG_IMPCTRL_OHM, - FIELD_PREP(STM32MP25_PCIEPRG_IMPCTRL_OHM, imp_of)); + regval); } else { regmap_read(combophy->regmap, SYSCFG_PCIEPRGCR, &val); imp_of = FIELD_GET(STM32MP25_PCIEPRG_IMPCTRL_OHM, val); @@ -150,16 +155,20 @@ static int stm32_impedance_tune(struct stm32_combophy *combophy) return -EINVAL; } - for (vswing_of = 0; vswing_of < ARRAY_SIZE(imp_lookup[imp_of].vswing); vswing_of++) - if (imp_lookup[imp_of].vswing[vswing_of] >= val) + regval = 0; + for (vswing_of = 0; vswing_of < ARRAY_SIZE(imp_lookup[imp_of].vswing); vswing_of++) { + if (imp_lookup[imp_of].vswing[vswing_of] >= val) { + regval = FIELD_PREP(STM32MP25_PCIEPRG_IMPCTRL_VSWING, vswing_of); break; + } + } dev_dbg(combophy->dev, "Set %u microvolt swing\n", imp_lookup[imp_of].vswing[vswing_of]); regmap_update_bits(combophy->regmap, SYSCFG_PCIEPRGCR, STM32MP25_PCIEPRG_IMPCTRL_VSWING, - FIELD_PREP(STM32MP25_PCIEPRG_IMPCTRL_VSWING, vswing_of)); + regval); } return 0; diff --git a/drivers/pinctrl/pinctrl-mcp23s08.c b/drivers/pinctrl/pinctrl-mcp23s08.c index d66c3a3e8429..b96e6368a956 100644 --- a/drivers/pinctrl/pinctrl-mcp23s08.c +++ b/drivers/pinctrl/pinctrl-mcp23s08.c @@ -86,6 +86,7 @@ const struct regmap_config mcp23x08_regmap = { .num_reg_defaults = ARRAY_SIZE(mcp23x08_defaults), .cache_type = REGCACHE_FLAT, .max_register = MCP_OLAT, + .disable_locking = true, /* mcp->lock protects the regmap */ }; EXPORT_SYMBOL_GPL(mcp23x08_regmap); @@ -132,6 +133,7 @@ const struct regmap_config mcp23x17_regmap = { .num_reg_defaults = ARRAY_SIZE(mcp23x17_defaults), .cache_type = REGCACHE_FLAT, .val_format_endian = REGMAP_ENDIAN_LITTLE, + .disable_locking = true, /* mcp->lock protects the regmap */ }; EXPORT_SYMBOL_GPL(mcp23x17_regmap); @@ -228,7 +230,9 @@ static int mcp_pinconf_get(struct pinctrl_dev *pctldev, unsigned int pin, switch (param) { case PIN_CONFIG_BIAS_PULL_UP: + mutex_lock(&mcp->lock); ret = mcp_read(mcp, MCP_GPPU, &data); + mutex_unlock(&mcp->lock); if (ret < 0) return ret; status = (data & BIT(pin)) ? 1 : 0; @@ -257,7 +261,9 @@ static int mcp_pinconf_set(struct pinctrl_dev *pctldev, unsigned int pin, switch (param) { case PIN_CONFIG_BIAS_PULL_UP: + mutex_lock(&mcp->lock); ret = mcp_set_bit(mcp, MCP_GPPU, pin, arg); + mutex_unlock(&mcp->lock); break; default: dev_dbg(mcp->dev, "Invalid config param %04x\n", param); diff --git a/drivers/platform/chrome/cros_ec_lpc.c b/drivers/platform/chrome/cros_ec_lpc.c index 924bf4d3cc77..8470b7f2b135 100644 --- a/drivers/platform/chrome/cros_ec_lpc.c +++ b/drivers/platform/chrome/cros_ec_lpc.c @@ -707,7 +707,7 @@ static const struct dmi_system_id cros_ec_lpc_dmi_table[] __initconst = { /* Framework Laptop (12th Gen Intel Core) */ .matches = { DMI_MATCH(DMI_SYS_VENDOR, "Framework"), - DMI_EXACT_MATCH(DMI_PRODUCT_NAME, "12th Gen Intel Core"), + DMI_EXACT_MATCH(DMI_PRODUCT_NAME, "Laptop (12th Gen Intel Core)"), }, .driver_data = (void *)&framework_laptop_mec_lpc_driver_data, }, @@ -715,7 +715,7 @@ static const struct dmi_system_id cros_ec_lpc_dmi_table[] __initconst = { /* Framework Laptop (13th Gen Intel Core) */ .matches = { DMI_MATCH(DMI_SYS_VENDOR, "Framework"), - DMI_EXACT_MATCH(DMI_PRODUCT_NAME, "13th Gen Intel Core"), + DMI_EXACT_MATCH(DMI_PRODUCT_NAME, "Laptop (13th Gen Intel Core)"), }, .driver_data = (void *)&framework_laptop_mec_lpc_driver_data, }, diff --git a/drivers/platform/x86/amd/pmc/pmc.c b/drivers/platform/x86/amd/pmc/pmc.c index 26b878ee5191..a254debb9256 100644 --- a/drivers/platform/x86/amd/pmc/pmc.c +++ b/drivers/platform/x86/amd/pmc/pmc.c @@ -947,6 +947,10 @@ static int amd_pmc_suspend_handler(struct device *dev) { struct amd_pmc_dev *pdev = dev_get_drvdata(dev); + /* + * Must be called only from the same set of dev_pm_ops handlers + * as i8042_pm_suspend() is called: currently just from .suspend. + */ if (pdev->disable_8042_wakeup && !disable_workarounds) { int rc = amd_pmc_wa_irq1(pdev); @@ -959,7 +963,9 @@ static int amd_pmc_suspend_handler(struct device *dev) return 0; } -static DEFINE_SIMPLE_DEV_PM_OPS(amd_pmc_pm, amd_pmc_suspend_handler, NULL); +static const struct dev_pm_ops amd_pmc_pm = { + .suspend = amd_pmc_suspend_handler, +}; static const struct pci_device_id pmc_pci_ids[] = { { PCI_DEVICE(PCI_VENDOR_ID_AMD, AMD_CPU_ID_PS) }, diff --git a/drivers/platform/x86/dell/dell-uart-backlight.c b/drivers/platform/x86/dell/dell-uart-backlight.c index 6e5dc7e3674f..bcc5c0f3bb4d 100644 --- a/drivers/platform/x86/dell/dell-uart-backlight.c +++ b/drivers/platform/x86/dell/dell-uart-backlight.c @@ -283,6 +283,9 @@ static int dell_uart_bl_serdev_probe(struct serdev_device *serdev) init_waitqueue_head(&dell_bl->wait_queue); dell_bl->dev = dev; + serdev_device_set_drvdata(serdev, dell_bl); + serdev_device_set_client_ops(serdev, &dell_uart_bl_serdev_ops); + ret = devm_serdev_device_open(dev, serdev); if (ret) return dev_err_probe(dev, ret, "opening UART device\n"); @@ -290,8 +293,6 @@ static int dell_uart_bl_serdev_probe(struct serdev_device *serdev) /* 9600 bps, no flow control, these are the default but set them to be sure */ serdev_device_set_baudrate(serdev, 9600); serdev_device_set_flow_control(serdev, false); - serdev_device_set_drvdata(serdev, dell_bl); - serdev_device_set_client_ops(serdev, &dell_uart_bl_serdev_ops); get_version[0] = DELL_SOF(GET_CMD_LEN); get_version[1] = CMD_GET_VERSION; diff --git a/drivers/platform/x86/hp/hp-wmi.c b/drivers/platform/x86/hp/hp-wmi.c index 81ccc96ffe40..20c55bab3b8c 100644 --- a/drivers/platform/x86/hp/hp-wmi.c +++ b/drivers/platform/x86/hp/hp-wmi.c @@ -64,7 +64,7 @@ static const char * const omen_thermal_profile_boards[] = { "874A", "8603", "8604", "8748", "886B", "886C", "878A", "878B", "878C", "88C8", "88CB", "8786", "8787", "8788", "88D1", "88D2", "88F4", "88FD", "88F5", "88F6", "88F7", "88FE", "88FF", "8900", "8901", "8902", "8912", - "8917", "8918", "8949", "894A", "89EB", "8BAD", "8A42" + "8917", "8918", "8949", "894A", "89EB", "8BAD", "8A42", "8A15" }; /* DMI Board names of Omen laptops that are specifically set to be thermal @@ -80,7 +80,7 @@ static const char * const omen_thermal_profile_force_v0_boards[] = { * "balanced" when reaching zero. */ static const char * const omen_timed_thermal_profile_boards[] = { - "8BAD", "8A42" + "8BAD", "8A42", "8A15" }; /* DMI Board names of Victus laptops */ diff --git a/drivers/platform/x86/intel/pmc/core_ssram.c b/drivers/platform/x86/intel/pmc/core_ssram.c index 50ebfd586d3f..739569803017 100644 --- a/drivers/platform/x86/intel/pmc/core_ssram.c +++ b/drivers/platform/x86/intel/pmc/core_ssram.c @@ -269,8 +269,12 @@ pmc_core_ssram_get_pmc(struct pmc_dev *pmcdev, int pmc_idx, u32 offset) /* * The secondary PMC BARS (which are behind hidden PCI devices) * are read from fixed offsets in MMIO of the primary PMC BAR. + * If a device is not present, the value will be 0. */ ssram_base = get_base(tmp_ssram, offset); + if (!ssram_base) + return 0; + ssram = ioremap(ssram_base, SSRAM_HDR_SIZE); if (!ssram) return -ENOMEM; diff --git a/drivers/platform/x86/intel/speed_select_if/isst_if_common.c b/drivers/platform/x86/intel/speed_select_if/isst_if_common.c index 1e46e30dae96..dbcd3087aaa4 100644 --- a/drivers/platform/x86/intel/speed_select_if/isst_if_common.c +++ b/drivers/platform/x86/intel/speed_select_if/isst_if_common.c @@ -804,6 +804,7 @@ EXPORT_SYMBOL_GPL(isst_if_cdev_unregister); static const struct x86_cpu_id isst_cpu_ids[] = { X86_MATCH_VFM(INTEL_ATOM_CRESTMONT, SST_HPM_SUPPORTED), X86_MATCH_VFM(INTEL_ATOM_CRESTMONT_X, SST_HPM_SUPPORTED), + X86_MATCH_VFM(INTEL_ATOM_DARKMONT_X, SST_HPM_SUPPORTED), X86_MATCH_VFM(INTEL_EMERALDRAPIDS_X, 0), X86_MATCH_VFM(INTEL_GRANITERAPIDS_D, SST_HPM_SUPPORTED), X86_MATCH_VFM(INTEL_GRANITERAPIDS_X, SST_HPM_SUPPORTED), diff --git a/drivers/platform/x86/intel/tpmi_power_domains.c b/drivers/platform/x86/intel/tpmi_power_domains.c index 603e7ded06a9..2f01cd22a6ee 100644 --- a/drivers/platform/x86/intel/tpmi_power_domains.c +++ b/drivers/platform/x86/intel/tpmi_power_domains.c @@ -81,6 +81,7 @@ static const struct x86_cpu_id tpmi_cpu_ids[] = { X86_MATCH_VFM(INTEL_GRANITERAPIDS_X, NULL), X86_MATCH_VFM(INTEL_ATOM_CRESTMONT_X, NULL), X86_MATCH_VFM(INTEL_ATOM_CRESTMONT, NULL), + X86_MATCH_VFM(INTEL_ATOM_DARKMONT_X, NULL), X86_MATCH_VFM(INTEL_GRANITERAPIDS_D, NULL), X86_MATCH_VFM(INTEL_PANTHERCOVE_X, NULL), {} diff --git a/drivers/platform/x86/lenovo-yoga-tab2-pro-1380-fastcharger.c b/drivers/platform/x86/lenovo-yoga-tab2-pro-1380-fastcharger.c index d2699ca24f34..a96b215cd2c5 100644 --- a/drivers/platform/x86/lenovo-yoga-tab2-pro-1380-fastcharger.c +++ b/drivers/platform/x86/lenovo-yoga-tab2-pro-1380-fastcharger.c @@ -199,14 +199,15 @@ static int yt2_1380_fc_serdev_probe(struct serdev_device *serdev) if (ret) return ret; + serdev_device_set_drvdata(serdev, fc); + serdev_device_set_client_ops(serdev, &yt2_1380_fc_serdev_ops); + ret = devm_serdev_device_open(dev, serdev); if (ret) return dev_err_probe(dev, ret, "opening UART device\n"); serdev_device_set_baudrate(serdev, 600); serdev_device_set_flow_control(serdev, false); - serdev_device_set_drvdata(serdev, fc); - serdev_device_set_client_ops(serdev, &yt2_1380_fc_serdev_ops); ret = devm_extcon_register_notifier_all(dev, fc->extcon, &fc->nb); if (ret) diff --git a/drivers/platform/x86/mlx-platform.c b/drivers/platform/x86/mlx-platform.c index 671021cd1f59..9c7f30a47f1f 100644 --- a/drivers/platform/x86/mlx-platform.c +++ b/drivers/platform/x86/mlx-platform.c @@ -6237,6 +6237,7 @@ fail_pci_set_dma_mask: fail_pci_request_regions: pci_disable_device(pci_dev); fail_pci_enable_device: + pci_dev_put(pci_dev); return err; } @@ -6247,6 +6248,7 @@ mlxplat_pci_fpga_device_exit(struct pci_dev *pci_bridge, iounmap(pci_bridge_addr); pci_release_regions(pci_bridge); pci_disable_device(pci_bridge); + pci_dev_put(pci_bridge); } static int diff --git a/drivers/platform/x86/thinkpad_acpi.c b/drivers/platform/x86/thinkpad_acpi.c index 6371a9f765c1..2cfb2ac3f465 100644 --- a/drivers/platform/x86/thinkpad_acpi.c +++ b/drivers/platform/x86/thinkpad_acpi.c @@ -184,7 +184,8 @@ enum tpacpi_hkey_event_t { */ TP_HKEY_EV_AMT_TOGGLE = 0x131a, /* Toggle AMT on/off */ TP_HKEY_EV_DOUBLETAP_TOGGLE = 0x131c, /* Toggle trackpoint doubletap on/off */ - TP_HKEY_EV_PROFILE_TOGGLE = 0x131f, /* Toggle platform profile */ + TP_HKEY_EV_PROFILE_TOGGLE = 0x131f, /* Toggle platform profile in 2024 systems */ + TP_HKEY_EV_PROFILE_TOGGLE2 = 0x1401, /* Toggle platform profile in 2025 + systems */ /* Reasons for waking up from S3/S4 */ TP_HKEY_EV_WKUP_S3_UNDOCK = 0x2304, /* undock requested, S3 */ @@ -11200,6 +11201,7 @@ static bool tpacpi_driver_event(const unsigned int hkey_event) tp_features.trackpoint_doubletap = !tp_features.trackpoint_doubletap; return true; case TP_HKEY_EV_PROFILE_TOGGLE: + case TP_HKEY_EV_PROFILE_TOGGLE2: platform_profile_cycle(); return true; } diff --git a/drivers/pmdomain/core.c b/drivers/pmdomain/core.c index bb11f467dc78..20a9efebbcb7 100644 --- a/drivers/pmdomain/core.c +++ b/drivers/pmdomain/core.c @@ -2142,6 +2142,11 @@ static int genpd_set_default_power_state(struct generic_pm_domain *genpd) return 0; } +static void genpd_provider_release(struct device *dev) +{ + /* nothing to be done here */ +} + static int genpd_alloc_data(struct generic_pm_domain *genpd) { struct genpd_governor_data *gd = NULL; @@ -2173,6 +2178,7 @@ static int genpd_alloc_data(struct generic_pm_domain *genpd) genpd->gd = gd; device_initialize(&genpd->dev); + genpd->dev.release = genpd_provider_release; if (!genpd_is_dev_name_fw(genpd)) { dev_set_name(&genpd->dev, "%s", genpd->name); diff --git a/drivers/pmdomain/imx/gpcv2.c b/drivers/pmdomain/imx/gpcv2.c index e67ecf99ef84..9bdb80fd7210 100644 --- a/drivers/pmdomain/imx/gpcv2.c +++ b/drivers/pmdomain/imx/gpcv2.c @@ -1458,12 +1458,12 @@ static int imx_gpcv2_probe(struct platform_device *pdev) .max_register = SZ_4K, }; struct device *dev = &pdev->dev; - struct device_node *pgc_np; + struct device_node *pgc_np __free(device_node) = + of_get_child_by_name(dev->of_node, "pgc"); struct regmap *regmap; void __iomem *base; int ret; - pgc_np = of_get_child_by_name(dev->of_node, "pgc"); if (!pgc_np) { dev_err(dev, "No power domains specified in DT\n"); return -EINVAL; diff --git a/drivers/pmdomain/imx/imx8mp-blk-ctrl.c b/drivers/pmdomain/imx/imx8mp-blk-ctrl.c index e3a0f64c144c..3668fe66b22c 100644 --- a/drivers/pmdomain/imx/imx8mp-blk-ctrl.c +++ b/drivers/pmdomain/imx/imx8mp-blk-ctrl.c @@ -770,7 +770,7 @@ static void imx8mp_blk_ctrl_remove(struct platform_device *pdev) of_genpd_del_provider(pdev->dev.of_node); - for (i = 0; bc->onecell_data.num_domains; i++) { + for (i = 0; i < bc->onecell_data.num_domains; i++) { struct imx8mp_blk_ctrl_domain *domain = &bc->domains[i]; pm_genpd_remove(&domain->genpd); diff --git a/drivers/power/supply/bq24190_charger.c b/drivers/power/supply/bq24190_charger.c index 2b393eb5c282..c47f32f152e6 100644 --- a/drivers/power/supply/bq24190_charger.c +++ b/drivers/power/supply/bq24190_charger.c @@ -567,6 +567,7 @@ static int bq24190_set_otg_vbus(struct bq24190_dev_info *bdi, bool enable) static int bq24296_set_otg_vbus(struct bq24190_dev_info *bdi, bool enable) { + union power_supply_propval val = { .intval = bdi->charge_type }; int ret; ret = pm_runtime_resume_and_get(bdi->dev); @@ -587,13 +588,18 @@ static int bq24296_set_otg_vbus(struct bq24190_dev_info *bdi, bool enable) ret = bq24190_write_mask(bdi, BQ24190_REG_POC, BQ24296_REG_POC_OTG_CONFIG_MASK, - BQ24296_REG_POC_CHG_CONFIG_SHIFT, + BQ24296_REG_POC_OTG_CONFIG_SHIFT, BQ24296_REG_POC_OTG_CONFIG_OTG); - } else + } else { ret = bq24190_write_mask(bdi, BQ24190_REG_POC, BQ24296_REG_POC_OTG_CONFIG_MASK, - BQ24296_REG_POC_CHG_CONFIG_SHIFT, + BQ24296_REG_POC_OTG_CONFIG_SHIFT, BQ24296_REG_POC_OTG_CONFIG_DISABLE); + if (ret < 0) + goto out; + + ret = bq24190_charger_set_charge_type(bdi, &val); + } out: pm_runtime_mark_last_busy(bdi->dev); diff --git a/drivers/power/supply/cros_charge-control.c b/drivers/power/supply/cros_charge-control.c index 17c53591ce19..9b0a7500296b 100644 --- a/drivers/power/supply/cros_charge-control.c +++ b/drivers/power/supply/cros_charge-control.c @@ -7,8 +7,10 @@ #include <acpi/battery.h> #include <linux/container_of.h> #include <linux/dmi.h> +#include <linux/lockdep.h> #include <linux/mod_devicetable.h> #include <linux/module.h> +#include <linux/mutex.h> #include <linux/platform_data/cros_ec_commands.h> #include <linux/platform_data/cros_ec_proto.h> #include <linux/platform_device.h> @@ -49,6 +51,7 @@ struct cros_chctl_priv { struct attribute *attributes[_CROS_CHCTL_ATTR_COUNT]; struct attribute_group group; + struct mutex lock; /* protects fields below and cros_ec */ enum power_supply_charge_behaviour current_behaviour; u8 current_start_threshold, current_end_threshold; }; @@ -85,6 +88,8 @@ static int cros_chctl_configure_ec(struct cros_chctl_priv *priv) { struct ec_params_charge_control req = {}; + lockdep_assert_held(&priv->lock); + req.cmd = EC_CHARGE_CONTROL_CMD_SET; switch (priv->current_behaviour) { @@ -134,11 +139,15 @@ static ssize_t cros_chctl_store_threshold(struct device *dev, struct cros_chctl_ return -EINVAL; if (is_end_threshold) { - if (val <= priv->current_start_threshold) + /* Start threshold is not exposed, use fixed value */ + if (priv->cmd_version == 2) + priv->current_start_threshold = val == 100 ? 0 : val; + + if (val < priv->current_start_threshold) return -EINVAL; priv->current_end_threshold = val; } else { - if (val >= priv->current_end_threshold) + if (val > priv->current_end_threshold) return -EINVAL; priv->current_start_threshold = val; } @@ -159,6 +168,7 @@ static ssize_t charge_control_start_threshold_show(struct device *dev, struct cros_chctl_priv *priv = cros_chctl_attr_to_priv(&attr->attr, CROS_CHCTL_ATTR_START_THRESHOLD); + guard(mutex)(&priv->lock); return sysfs_emit(buf, "%u\n", (unsigned int)priv->current_start_threshold); } @@ -169,6 +179,7 @@ static ssize_t charge_control_start_threshold_store(struct device *dev, struct cros_chctl_priv *priv = cros_chctl_attr_to_priv(&attr->attr, CROS_CHCTL_ATTR_START_THRESHOLD); + guard(mutex)(&priv->lock); return cros_chctl_store_threshold(dev, priv, 0, buf, count); } @@ -178,6 +189,7 @@ static ssize_t charge_control_end_threshold_show(struct device *dev, struct devi struct cros_chctl_priv *priv = cros_chctl_attr_to_priv(&attr->attr, CROS_CHCTL_ATTR_END_THRESHOLD); + guard(mutex)(&priv->lock); return sysfs_emit(buf, "%u\n", (unsigned int)priv->current_end_threshold); } @@ -187,6 +199,7 @@ static ssize_t charge_control_end_threshold_store(struct device *dev, struct dev struct cros_chctl_priv *priv = cros_chctl_attr_to_priv(&attr->attr, CROS_CHCTL_ATTR_END_THRESHOLD); + guard(mutex)(&priv->lock); return cros_chctl_store_threshold(dev, priv, 1, buf, count); } @@ -195,6 +208,7 @@ static ssize_t charge_behaviour_show(struct device *dev, struct device_attribute struct cros_chctl_priv *priv = cros_chctl_attr_to_priv(&attr->attr, CROS_CHCTL_ATTR_CHARGE_BEHAVIOUR); + guard(mutex)(&priv->lock); return power_supply_charge_behaviour_show(dev, EC_CHARGE_CONTROL_BEHAVIOURS, priv->current_behaviour, buf); } @@ -210,6 +224,7 @@ static ssize_t charge_behaviour_store(struct device *dev, struct device_attribut if (ret < 0) return ret; + guard(mutex)(&priv->lock); priv->current_behaviour = ret; ret = cros_chctl_configure_ec(priv); @@ -223,12 +238,10 @@ static umode_t cros_chtl_attr_is_visible(struct kobject *kobj, struct attribute { struct cros_chctl_priv *priv = cros_chctl_attr_to_priv(attr, n); - if (priv->cmd_version < 2) { - if (n == CROS_CHCTL_ATTR_START_THRESHOLD) - return 0; - if (n == CROS_CHCTL_ATTR_END_THRESHOLD) - return 0; - } + if (n == CROS_CHCTL_ATTR_START_THRESHOLD && priv->cmd_version < 3) + return 0; + else if (n == CROS_CHCTL_ATTR_END_THRESHOLD && priv->cmd_version < 2) + return 0; return attr->mode; } @@ -290,6 +303,10 @@ static int cros_chctl_probe(struct platform_device *pdev) if (!priv) return -ENOMEM; + ret = devm_mutex_init(dev, &priv->lock); + if (ret) + return ret; + ret = cros_ec_get_cmd_versions(cros_ec, EC_CMD_CHARGE_CONTROL); if (ret < 0) return ret; @@ -327,7 +344,8 @@ static int cros_chctl_probe(struct platform_device *pdev) priv->current_end_threshold = 100; /* Bring EC into well-known state */ - ret = cros_chctl_configure_ec(priv); + scoped_guard(mutex, &priv->lock) + ret = cros_chctl_configure_ec(priv); if (ret < 0) return ret; diff --git a/drivers/power/supply/gpio-charger.c b/drivers/power/supply/gpio-charger.c index 68212b39785b..6139f736ecbe 100644 --- a/drivers/power/supply/gpio-charger.c +++ b/drivers/power/supply/gpio-charger.c @@ -67,6 +67,14 @@ static int set_charge_current_limit(struct gpio_charger *gpio_charger, int val) if (gpio_charger->current_limit_map[i].limit_ua <= val) break; } + + /* + * If a valid charge current limit isn't found, default to smallest + * current limitation for safety reasons. + */ + if (i >= gpio_charger->current_limit_map_size) + i = gpio_charger->current_limit_map_size - 1; + mapping = gpio_charger->current_limit_map[i]; for (i = 0; i < ndescs; i++) { diff --git a/drivers/reset/reset-rzg2l-usbphy-ctrl.c b/drivers/reset/reset-rzg2l-usbphy-ctrl.c index 12d0535a874b..8a7f167e405e 100644 --- a/drivers/reset/reset-rzg2l-usbphy-ctrl.c +++ b/drivers/reset/reset-rzg2l-usbphy-ctrl.c @@ -176,6 +176,7 @@ static int rzg2l_usbphy_ctrl_probe(struct platform_device *pdev) vdev->dev.parent = dev; priv->vdev = vdev; + device_set_of_node_from_dev(&vdev->dev, dev); error = platform_device_add(vdev); if (error) goto err_device_put; diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c index 51c496ca9380..4411426a7894 100644 --- a/drivers/scsi/scsi_lib.c +++ b/drivers/scsi/scsi_lib.c @@ -210,6 +210,9 @@ static int scsi_check_passthrough(struct scsi_cmnd *scmd, struct scsi_sense_hdr sshdr; enum sam_status status; + if (!scmd->result) + return 0; + if (!failures) return 0; diff --git a/drivers/scsi/scsi_transport_iscsi.c b/drivers/scsi/scsi_transport_iscsi.c index fde7de3b1e55..9b47f91c5b97 100644 --- a/drivers/scsi/scsi_transport_iscsi.c +++ b/drivers/scsi/scsi_transport_iscsi.c @@ -4104,7 +4104,7 @@ iscsi_if_rx(struct sk_buff *skb) } do { /* - * special case for GET_STATS: + * special case for GET_STATS, GET_CHAP and GET_HOST_STATS: * on success - sending reply and stats from * inside of if_recv_msg(), * on error - fall through. @@ -4113,6 +4113,8 @@ iscsi_if_rx(struct sk_buff *skb) break; if (ev->type == ISCSI_UEVENT_GET_CHAP && !err) break; + if (ev->type == ISCSI_UEVENT_GET_HOST_STATS && !err) + break; err = iscsi_if_send_reply(portid, nlh->nlmsg_type, ev, sizeof(*ev)); if (err == -EAGAIN && --retries < 0) { diff --git a/drivers/staging/gpib/Kconfig b/drivers/staging/gpib/Kconfig index 259f3ff33646..81510db3072e 100644 --- a/drivers/staging/gpib/Kconfig +++ b/drivers/staging/gpib/Kconfig @@ -65,6 +65,8 @@ config GPIB_NI_PCI_ISA depends on ISA_BUS || PCI || PCMCIA depends on HAS_IOPORT depends on !X86_PAE + depends on PCMCIA || !PCMCIA + depends on HAS_IOPORT_MAP select GPIB_COMMON select GPIB_NEC7210 help @@ -89,6 +91,7 @@ config GPIB_CB7210 depends on HAS_IOPORT depends on ISA_BUS || PCI || PCMCIA depends on !X86_PAE + depends on PCMCIA || !PCMCIA select GPIB_COMMON select GPIB_NEC7210 help @@ -177,6 +180,7 @@ config GPIB_HP82341 config GPIB_INES tristate "INES" depends on PCI || ISA_BUS || PCMCIA + depends on PCMCIA || !PCMCIA depends on HAS_IOPORT depends on !X86_PAE select GPIB_COMMON @@ -199,8 +203,8 @@ config GPIB_INES called cb7210. config GPIB_PCMCIA - bool "PCMCIA/Cardbus support for NI MC and Ines boards" - depends on PCCARD && (GPIB_NI_PCI_ISA || GPIB_CB7210 || GPIB_INES) + def_bool y + depends on PCMCIA && (GPIB_NI_PCI_ISA || GPIB_CB7210 || GPIB_INES) help Enable PCMCIA/CArdbus support for National Instruments, measurement computing boards and Ines boards. diff --git a/drivers/staging/gpib/agilent_82350b/Makefile b/drivers/staging/gpib/agilent_82350b/Makefile index d9236c92e04b..f24e1e713a63 100644 --- a/drivers/staging/gpib/agilent_82350b/Makefile +++ b/drivers/staging/gpib/agilent_82350b/Makefile @@ -1,2 +1,2 @@ -obj-m += agilent_82350b.o +obj-$(CONFIG_GPIB_AGILENT_82350B) += agilent_82350b.o diff --git a/drivers/staging/gpib/agilent_82350b/agilent_82350b.c b/drivers/staging/gpib/agilent_82350b/agilent_82350b.c index 53006d0cc79c..8e2334fe5c9b 100644 --- a/drivers/staging/gpib/agilent_82350b/agilent_82350b.c +++ b/drivers/staging/gpib/agilent_82350b/agilent_82350b.c @@ -700,7 +700,7 @@ static int agilent_82350b_generic_attach(gpib_board_t *board, const gpib_board_c GPIB_82350A_REGION)); dev_dbg(board->gpib_dev, "%s: gpib base address remapped to 0x%p\n", driver_name, a_priv->gpib_base); - tms_priv->iobase = a_priv->gpib_base + TMS9914_BASE_REG; + tms_priv->mmiobase = a_priv->gpib_base + TMS9914_BASE_REG; a_priv->sram_base = ioremap(pci_resource_start(a_priv->pci_device, SRAM_82350A_REGION), pci_resource_len(a_priv->pci_device, @@ -724,7 +724,7 @@ static int agilent_82350b_generic_attach(gpib_board_t *board, const gpib_board_c pci_resource_len(a_priv->pci_device, GPIB_REGION)); dev_dbg(board->gpib_dev, "%s: gpib base address remapped to 0x%p\n", driver_name, a_priv->gpib_base); - tms_priv->iobase = a_priv->gpib_base + TMS9914_BASE_REG; + tms_priv->mmiobase = a_priv->gpib_base + TMS9914_BASE_REG; a_priv->sram_base = ioremap(pci_resource_start(a_priv->pci_device, SRAM_REGION), pci_resource_len(a_priv->pci_device, SRAM_REGION)); dev_dbg(board->gpib_dev, "%s: sram base address remapped to 0x%p\n", diff --git a/drivers/staging/gpib/agilent_82357a/Makefile b/drivers/staging/gpib/agilent_82357a/Makefile index 4a1d940fce2b..81a55c257a6e 100644 --- a/drivers/staging/gpib/agilent_82357a/Makefile +++ b/drivers/staging/gpib/agilent_82357a/Makefile @@ -1,4 +1,4 @@ -obj-m += agilent_82357a.o +obj-$(CONFIG_GPIB_AGILENT_82357A) += agilent_82357a.o diff --git a/drivers/staging/gpib/cb7210/Makefile b/drivers/staging/gpib/cb7210/Makefile index 22e0214fc17d..cda0725d6487 100644 --- a/drivers/staging/gpib/cb7210/Makefile +++ b/drivers/staging/gpib/cb7210/Makefile @@ -1,4 +1,4 @@ ccflags-$(CONFIG_GPIB_PCMCIA) := -DGPIB_PCMCIA -obj-m += cb7210.o +obj-$(CONFIG_GPIB_CB7210) += cb7210.o diff --git a/drivers/staging/gpib/cb7210/cb7210.c b/drivers/staging/gpib/cb7210/cb7210.c index 63df7f3eb3f3..59e41c97f518 100644 --- a/drivers/staging/gpib/cb7210/cb7210.c +++ b/drivers/staging/gpib/cb7210/cb7210.c @@ -971,12 +971,12 @@ int cb_pci_attach(gpib_board_t *board, const gpib_board_config_t *config) switch (cb_priv->pci_chip) { case PCI_CHIP_AMCC_S5933: cb_priv->amcc_iobase = pci_resource_start(cb_priv->pci_device, 0); - nec_priv->iobase = (void *)(pci_resource_start(cb_priv->pci_device, 1)); + nec_priv->iobase = pci_resource_start(cb_priv->pci_device, 1); cb_priv->fifo_iobase = pci_resource_start(cb_priv->pci_device, 2); break; case PCI_CHIP_QUANCOM: - nec_priv->iobase = (void *)(pci_resource_start(cb_priv->pci_device, 0)); - cb_priv->fifo_iobase = (unsigned long)nec_priv->iobase; + nec_priv->iobase = pci_resource_start(cb_priv->pci_device, 0); + cb_priv->fifo_iobase = nec_priv->iobase; break; default: pr_err("cb7210: bug! unhandled pci_chip=%i\n", cb_priv->pci_chip); @@ -1040,8 +1040,8 @@ int cb_isa_attach(gpib_board_t *board, const gpib_board_config_t *config) return retval; cb_priv = board->private_data; nec_priv = &cb_priv->nec7210_priv; - if (request_region((unsigned long)config->ibbase, cb7210_iosize, "cb7210") == 0) { - pr_err("gpib: ioports starting at 0x%p are already in use\n", config->ibbase); + if (request_region(config->ibbase, cb7210_iosize, "cb7210") == 0) { + pr_err("gpib: ioports starting at 0x%u are already in use\n", config->ibbase); return -EIO; } nec_priv->iobase = config->ibbase; @@ -1471,7 +1471,7 @@ int cb_pcmcia_attach(gpib_board_t *board, const gpib_board_config_t *config) (unsigned long)curr_dev->resource[0]->start); return -EIO; } - nec_priv->iobase = (void *)(unsigned long)curr_dev->resource[0]->start; + nec_priv->iobase = curr_dev->resource[0]->start; cb_priv->fifo_iobase = curr_dev->resource[0]->start; if (request_irq(curr_dev->irq, cb7210_interrupt, IRQF_SHARED, diff --git a/drivers/staging/gpib/cb7210/cb7210.h b/drivers/staging/gpib/cb7210/cb7210.h index 4ad976de2b68..c17cb22585f7 100644 --- a/drivers/staging/gpib/cb7210/cb7210.h +++ b/drivers/staging/gpib/cb7210/cb7210.h @@ -113,9 +113,9 @@ enum hs_regs { HS_STATUS = 0x8, /* HS_STATUS register */ }; -static inline unsigned long nec7210_iobase(const struct cb7210_priv *cb_priv) +static inline u32 nec7210_iobase(const struct cb7210_priv *cb_priv) { - return (unsigned long)(cb_priv->nec7210_priv.iobase); + return cb_priv->nec7210_priv.iobase; } static inline int cb7210_page_in_bits(unsigned int page) diff --git a/drivers/staging/gpib/cec/Makefile b/drivers/staging/gpib/cec/Makefile index f4638628ff29..b7141e23d4e0 100644 --- a/drivers/staging/gpib/cec/Makefile +++ b/drivers/staging/gpib/cec/Makefile @@ -1,3 +1,3 @@ -obj-m += cec_gpib.o +obj-$(CONFIG_GPIB_CEC_PCI) += cec_gpib.o diff --git a/drivers/staging/gpib/cec/cec_gpib.c b/drivers/staging/gpib/cec/cec_gpib.c index 3dc933deb401..9c00a874468c 100644 --- a/drivers/staging/gpib/cec/cec_gpib.c +++ b/drivers/staging/gpib/cec/cec_gpib.c @@ -297,8 +297,8 @@ int cec_pci_attach(gpib_board_t *board, const gpib_board_config_t *config) cec_priv->plx_iobase = pci_resource_start(cec_priv->pci_device, 1); pr_info(" plx9050 base address 0x%lx\n", cec_priv->plx_iobase); - nec_priv->iobase = (void *)(pci_resource_start(cec_priv->pci_device, 3)); - pr_info(" nec7210 base address 0x%p\n", nec_priv->iobase); + nec_priv->iobase = pci_resource_start(cec_priv->pci_device, 3); + pr_info(" nec7210 base address 0x%x\n", nec_priv->iobase); isr_flags |= IRQF_SHARED; if (request_irq(cec_priv->pci_device->irq, cec_interrupt, isr_flags, "pci-gpib", board)) { diff --git a/drivers/staging/gpib/common/gpib_os.c b/drivers/staging/gpib/common/gpib_os.c index 405237d8cb47..0962729d7dfe 100644 --- a/drivers/staging/gpib/common/gpib_os.c +++ b/drivers/staging/gpib/common/gpib_os.c @@ -116,56 +116,6 @@ int io_timed_out(gpib_board_t *board) return 0; } -void writeb_wrapper(unsigned int value, void *address) -{ - writeb(value, address); -}; -EXPORT_SYMBOL(writeb_wrapper); - -void writew_wrapper(unsigned int value, void *address) -{ - writew(value, address); -}; -EXPORT_SYMBOL(writew_wrapper); - -unsigned int readb_wrapper(void *address) -{ - return readb(address); -}; -EXPORT_SYMBOL(readb_wrapper); - -unsigned int readw_wrapper(void *address) -{ - return readw(address); -}; -EXPORT_SYMBOL(readw_wrapper); - -#ifdef CONFIG_HAS_IOPORT -void outb_wrapper(unsigned int value, void *address) -{ - outb(value, (unsigned long)(address)); -}; -EXPORT_SYMBOL(outb_wrapper); - -void outw_wrapper(unsigned int value, void *address) -{ - outw(value, (unsigned long)(address)); -}; -EXPORT_SYMBOL(outw_wrapper); - -unsigned int inb_wrapper(void *address) -{ - return inb((unsigned long)(address)); -}; -EXPORT_SYMBOL(inb_wrapper); - -unsigned int inw_wrapper(void *address) -{ - return inw((unsigned long)(address)); -}; -EXPORT_SYMBOL(inw_wrapper); -#endif - /* this is a function instead of a constant because of Suse * defining HZ to be a function call to get_hz() */ @@ -536,7 +486,7 @@ int dvrsp(gpib_board_t *board, unsigned int pad, int sad, return -1; } - if (pad > MAX_GPIB_PRIMARY_ADDRESS || sad > MAX_GPIB_SECONDARY_ADDRESS) { + if (pad > MAX_GPIB_PRIMARY_ADDRESS || sad > MAX_GPIB_SECONDARY_ADDRESS || sad < -1) { pr_err("gpib: bad address for serial poll"); return -1; } @@ -1623,7 +1573,7 @@ static int iobase_ioctl(gpib_board_config_t *config, unsigned long arg) if (WARN_ON_ONCE(sizeof(void *) > sizeof(base_addr))) return -EFAULT; - config->ibbase = (void *)(unsigned long)(base_addr); + config->ibbase = base_addr; return 0; } diff --git a/drivers/staging/gpib/eastwood/Makefile b/drivers/staging/gpib/eastwood/Makefile index c74056f959d0..384825195f77 100644 --- a/drivers/staging/gpib/eastwood/Makefile +++ b/drivers/staging/gpib/eastwood/Makefile @@ -1,3 +1,3 @@ -obj-m += fluke_gpib.o +obj-$(CONFIG_GPIB_FLUKE) += fluke_gpib.o diff --git a/drivers/staging/gpib/eastwood/fluke_gpib.c b/drivers/staging/gpib/eastwood/fluke_gpib.c index 3f938ab0c84d..5e59d38beb35 100644 --- a/drivers/staging/gpib/eastwood/fluke_gpib.c +++ b/drivers/staging/gpib/eastwood/fluke_gpib.c @@ -1011,12 +1011,12 @@ static int fluke_attach_impl(gpib_board_t *board, const gpib_board_config_t *con } e_priv->gpib_iomem_res = res; - nec_priv->iobase = ioremap(e_priv->gpib_iomem_res->start, + nec_priv->mmiobase = ioremap(e_priv->gpib_iomem_res->start, resource_size(e_priv->gpib_iomem_res)); - pr_info("gpib: iobase %lx remapped to %p, length=%d\n", - (unsigned long)e_priv->gpib_iomem_res->start, - nec_priv->iobase, (int)resource_size(e_priv->gpib_iomem_res)); - if (!nec_priv->iobase) { + pr_info("gpib: mmiobase %llx remapped to %p, length=%d\n", + (u64)e_priv->gpib_iomem_res->start, + nec_priv->mmiobase, (int)resource_size(e_priv->gpib_iomem_res)); + if (!nec_priv->mmiobase) { dev_err(&fluke_gpib_pdev->dev, "Could not map I/O memory\n"); return -ENOMEM; } @@ -1107,7 +1107,7 @@ void fluke_detach(gpib_board_t *board) gpib_free_pseudo_irq(board); nec_priv = &e_priv->nec7210_priv; - if (nec_priv->iobase) { + if (nec_priv->mmiobase) { fluke_paged_write_byte(e_priv, 0, ISR0_IMR0, ISR0_IMR0_PAGE); nec7210_board_reset(nec_priv, board); } diff --git a/drivers/staging/gpib/eastwood/fluke_gpib.h b/drivers/staging/gpib/eastwood/fluke_gpib.h index fcbd42f8f9af..4e2144d45270 100644 --- a/drivers/staging/gpib/eastwood/fluke_gpib.h +++ b/drivers/staging/gpib/eastwood/fluke_gpib.h @@ -72,7 +72,7 @@ static inline uint8_t fluke_read_byte_nolock(struct nec7210_priv *nec_priv, { u8 retval; - retval = readl(nec_priv->iobase + register_num * nec_priv->offset); + retval = readl(nec_priv->mmiobase + register_num * nec_priv->offset); return retval; } @@ -80,7 +80,7 @@ static inline uint8_t fluke_read_byte_nolock(struct nec7210_priv *nec_priv, static inline void fluke_write_byte_nolock(struct nec7210_priv *nec_priv, uint8_t data, int register_num) { - writel(data, nec_priv->iobase + register_num * nec_priv->offset); + writel(data, nec_priv->mmiobase + register_num * nec_priv->offset); } static inline uint8_t fluke_paged_read_byte(struct fluke_priv *e_priv, diff --git a/drivers/staging/gpib/fmh_gpib/fmh_gpib.c b/drivers/staging/gpib/fmh_gpib/fmh_gpib.c index 62791db1c34a..0662b20a45e7 100644 --- a/drivers/staging/gpib/fmh_gpib/fmh_gpib.c +++ b/drivers/staging/gpib/fmh_gpib/fmh_gpib.c @@ -24,6 +24,8 @@ #include <linux/slab.h> MODULE_LICENSE("GPL"); +MODULE_DESCRIPTION("GPIB Driver for fmh_gpib_core"); +MODULE_AUTHOR("Frank Mori Hess <fmh6jj@gmail.com>"); static irqreturn_t fmh_gpib_interrupt(int irq, void *arg); static int fmh_gpib_attach_holdoff_all(gpib_board_t *board, const gpib_board_config_t *config); @@ -1419,15 +1421,14 @@ static int fmh_gpib_attach_impl(gpib_board_t *board, const gpib_board_config_t * } e_priv->gpib_iomem_res = res; - nec_priv->iobase = ioremap(e_priv->gpib_iomem_res->start, + nec_priv->mmiobase = ioremap(e_priv->gpib_iomem_res->start, resource_size(e_priv->gpib_iomem_res)); - if (!nec_priv->iobase) { + if (!nec_priv->mmiobase) { dev_err(board->dev, "Could not map I/O memory for gpib\n"); return -ENOMEM; } - dev_info(board->dev, "iobase 0x%lx remapped to %p, length=%ld\n", - (unsigned long)e_priv->gpib_iomem_res->start, - nec_priv->iobase, (unsigned long)resource_size(e_priv->gpib_iomem_res)); + dev_info(board->dev, "iobase %pr remapped to %p\n", + e_priv->gpib_iomem_res, nec_priv->mmiobase); res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "dma_fifos"); if (!res) { @@ -1507,14 +1508,14 @@ void fmh_gpib_detach(gpib_board_t *board) free_irq(e_priv->irq, board); if (e_priv->fifo_base) fifos_write(e_priv, 0, FIFO_CONTROL_STATUS_REG); - if (nec_priv->iobase) { + if (nec_priv->mmiobase) { write_byte(nec_priv, 0, ISR0_IMR0_REG); nec7210_board_reset(nec_priv, board); } if (e_priv->fifo_base) iounmap(e_priv->fifo_base); - if (nec_priv->iobase) - iounmap(nec_priv->iobase); + if (nec_priv->mmiobase) + iounmap(nec_priv->mmiobase); if (e_priv->dma_port_res) { release_mem_region(e_priv->dma_port_res->start, resource_size(e_priv->dma_port_res)); @@ -1564,12 +1565,12 @@ static int fmh_gpib_pci_attach_impl(gpib_board_t *board, const gpib_board_config e_priv->gpib_iomem_res = &pci_device->resource[gpib_control_status_pci_resource_index]; e_priv->dma_port_res = &pci_device->resource[gpib_fifo_pci_resource_index]; - nec_priv->iobase = ioremap(pci_resource_start(pci_device, + nec_priv->mmiobase = ioremap(pci_resource_start(pci_device, gpib_control_status_pci_resource_index), pci_resource_len(pci_device, gpib_control_status_pci_resource_index)); dev_info(board->dev, "base address for gpib control/status registers remapped to 0x%p\n", - nec_priv->iobase); + nec_priv->mmiobase); if (e_priv->dma_port_res->flags & IORESOURCE_MEM) { e_priv->fifo_base = ioremap(pci_resource_start(pci_device, @@ -1632,14 +1633,14 @@ void fmh_gpib_pci_detach(gpib_board_t *board) free_irq(e_priv->irq, board); if (e_priv->fifo_base) fifos_write(e_priv, 0, FIFO_CONTROL_STATUS_REG); - if (nec_priv->iobase) { + if (nec_priv->mmiobase) { write_byte(nec_priv, 0, ISR0_IMR0_REG); nec7210_board_reset(nec_priv, board); } if (e_priv->fifo_base) iounmap(e_priv->fifo_base); - if (nec_priv->iobase) - iounmap(nec_priv->iobase); + if (nec_priv->mmiobase) + iounmap(nec_priv->mmiobase); if (e_priv->dma_port_res || e_priv->gpib_iomem_res) pci_release_regions(to_pci_dev(board->dev)); if (board->dev) diff --git a/drivers/staging/gpib/fmh_gpib/fmh_gpib.h b/drivers/staging/gpib/fmh_gpib/fmh_gpib.h index 43bfc89d2a6f..60b1bd6d3c15 100644 --- a/drivers/staging/gpib/fmh_gpib/fmh_gpib.h +++ b/drivers/staging/gpib/fmh_gpib/fmh_gpib.h @@ -127,13 +127,13 @@ static const unsigned int fifo_max_burst_length_mask = 0x00ff; static inline uint8_t gpib_cs_read_byte(struct nec7210_priv *nec_priv, unsigned int register_num) { - return readb(nec_priv->iobase + register_num * nec_priv->offset); + return readb(nec_priv->mmiobase + register_num * nec_priv->offset); } static inline void gpib_cs_write_byte(struct nec7210_priv *nec_priv, uint8_t data, unsigned int register_num) { - writeb(data, nec_priv->iobase + register_num * nec_priv->offset); + writeb(data, nec_priv->mmiobase + register_num * nec_priv->offset); } static inline uint16_t fifos_read(struct fmh_priv *fmh_priv, int register_num) diff --git a/drivers/staging/gpib/gpio/Makefile b/drivers/staging/gpib/gpio/Makefile index a31ded6e5924..00ea52abdda7 100644 --- a/drivers/staging/gpib/gpio/Makefile +++ b/drivers/staging/gpib/gpio/Makefile @@ -1,4 +1,4 @@ -obj-m += gpib_bitbang.o +obj-$(CONFIG_GPIB_GPIO) += gpib_bitbang.o diff --git a/drivers/staging/gpib/gpio/gpib_bitbang.c b/drivers/staging/gpib/gpio/gpib_bitbang.c index a2d562cbd65b..23550502e012 100644 --- a/drivers/staging/gpib/gpio/gpib_bitbang.c +++ b/drivers/staging/gpib/gpio/gpib_bitbang.c @@ -315,7 +315,7 @@ struct bb_priv { enum listener_function_state listener_state; }; -inline long usec_diff(struct timespec64 *a, struct timespec64 *b); +static inline long usec_diff(struct timespec64 *a, struct timespec64 *b); static void bb_buffer_print(unsigned char *buffer, size_t length, int cmd, int eoi); static void set_data_lines(u8 byte); static u8 get_data_lines(void); diff --git a/drivers/staging/gpib/hp_82335/Makefile b/drivers/staging/gpib/hp_82335/Makefile index 8b7a552e9355..305ce44ee48a 100644 --- a/drivers/staging/gpib/hp_82335/Makefile +++ b/drivers/staging/gpib/hp_82335/Makefile @@ -1,4 +1,4 @@ -obj-m += hp82335.o +obj-$(CONFIG_GPIB_HP82335) += hp82335.o diff --git a/drivers/staging/gpib/hp_82335/hp82335.c b/drivers/staging/gpib/hp_82335/hp82335.c index 40afe42aea47..ea78143c7ab6 100644 --- a/drivers/staging/gpib/hp_82335/hp82335.c +++ b/drivers/staging/gpib/hp_82335/hp82335.c @@ -9,6 +9,7 @@ */ #include "hp82335.h" +#include <linux/io.h> #include <linux/ioport.h> #include <linux/sched.h> #include <linux/module.h> @@ -233,7 +234,7 @@ static void hp82335_clear_interrupt(struct hp82335_priv *hp_priv) { struct tms9914_priv *tms_priv = &hp_priv->tms9914_priv; - writeb(0, tms_priv->iobase + HPREG_INTR_CLEAR); + writeb(0, tms_priv->mmiobase + HPREG_INTR_CLEAR); } int hp82335_attach(gpib_board_t *board, const gpib_board_config_t *config) @@ -241,7 +242,7 @@ int hp82335_attach(gpib_board_t *board, const gpib_board_config_t *config) struct hp82335_priv *hp_priv; struct tms9914_priv *tms_priv; int retval; - const unsigned long upper_iomem_base = (unsigned long)config->ibbase + hp82335_rom_size; + const unsigned long upper_iomem_base = config->ibbase + hp82335_rom_size; board->status = 0; @@ -253,7 +254,7 @@ int hp82335_attach(gpib_board_t *board, const gpib_board_config_t *config) tms_priv->write_byte = hp82335_write_byte; tms_priv->offset = 1; - switch ((unsigned long)(config->ibbase)) { + switch (config->ibbase) { case 0xc4000: case 0xc8000: case 0xcc000: @@ -271,7 +272,7 @@ int hp82335_attach(gpib_board_t *board, const gpib_board_config_t *config) case 0xfc000: break; default: - pr_err("hp82335: invalid base io address 0x%p\n", config->ibbase); + pr_err("hp82335: invalid base io address 0x%u\n", config->ibbase); return -EINVAL; } if (!request_mem_region(upper_iomem_base, hp82335_upper_iomem_size, "hp82335")) { @@ -280,9 +281,9 @@ int hp82335_attach(gpib_board_t *board, const gpib_board_config_t *config) return -EBUSY; } hp_priv->raw_iobase = upper_iomem_base; - tms_priv->iobase = ioremap(upper_iomem_base, hp82335_upper_iomem_size); + tms_priv->mmiobase = ioremap(upper_iomem_base, hp82335_upper_iomem_size); pr_info("hp82335: upper half of 82335 iomem region 0x%lx remapped to 0x%p\n", - hp_priv->raw_iobase, tms_priv->iobase); + hp_priv->raw_iobase, tms_priv->mmiobase); retval = request_irq(config->ibirq, hp82335_interrupt, 0, "hp82335", board); if (retval) { @@ -296,7 +297,7 @@ int hp82335_attach(gpib_board_t *board, const gpib_board_config_t *config) hp82335_clear_interrupt(hp_priv); - writeb(INTR_ENABLE, tms_priv->iobase + HPREG_CCR); + writeb(INTR_ENABLE, tms_priv->mmiobase + HPREG_CCR); tms9914_online(board, tms_priv); @@ -312,10 +313,10 @@ void hp82335_detach(gpib_board_t *board) tms_priv = &hp_priv->tms9914_priv; if (hp_priv->irq) free_irq(hp_priv->irq, board); - if (tms_priv->iobase) { - writeb(0, tms_priv->iobase + HPREG_CCR); + if (tms_priv->mmiobase) { + writeb(0, tms_priv->mmiobase + HPREG_CCR); tms9914_board_reset(tms_priv); - iounmap((void *)tms_priv->iobase); + iounmap(tms_priv->mmiobase); } if (hp_priv->raw_iobase) release_mem_region(hp_priv->raw_iobase, hp82335_upper_iomem_size); diff --git a/drivers/staging/gpib/hp_82341/Makefile b/drivers/staging/gpib/hp_82341/Makefile index 1fe7db4f8ca4..21367310a17e 100644 --- a/drivers/staging/gpib/hp_82341/Makefile +++ b/drivers/staging/gpib/hp_82341/Makefile @@ -1,2 +1,2 @@ -obj-m += hp_82341.o +obj-$(CONFIG_GPIB_HP82341) += hp_82341.o diff --git a/drivers/staging/gpib/hp_82341/hp_82341.c b/drivers/staging/gpib/hp_82341/hp_82341.c index 8ad1c885a9fb..71d481e88bd9 100644 --- a/drivers/staging/gpib/hp_82341/hp_82341.c +++ b/drivers/staging/gpib/hp_82341/hp_82341.c @@ -473,12 +473,12 @@ void hp_82341_free_private(gpib_board_t *board) static uint8_t hp_82341_read_byte(struct tms9914_priv *priv, unsigned int register_num) { - return inb((unsigned long)(priv->iobase) + register_num); + return inb(priv->iobase + register_num); } static void hp_82341_write_byte(struct tms9914_priv *priv, uint8_t data, unsigned int register_num) { - outb(data, (unsigned long)(priv->iobase) + register_num); + outb(data, priv->iobase + register_num); } static int hp_82341_find_isapnp_board(struct pnp_dev **dev) @@ -682,8 +682,8 @@ int hp_82341_attach(gpib_board_t *board, const gpib_board_config_t *config) { struct hp_82341_priv *hp_priv; struct tms9914_priv *tms_priv; - unsigned long start_addr; - void *iobase; + u32 start_addr; + u32 iobase; int irq; int i; int retval; @@ -704,7 +704,7 @@ int hp_82341_attach(gpib_board_t *board, const gpib_board_config_t *config) if (retval < 0) return retval; hp_priv->pnp_dev = dev; - iobase = (void *)(pnp_port_start(dev, 0)); + iobase = pnp_port_start(dev, 0); irq = pnp_irq(dev, 0); hp_priv->hw_version = HW_VERSION_82341D; hp_priv->io_region_offset = 0x8; @@ -714,9 +714,9 @@ int hp_82341_attach(gpib_board_t *board, const gpib_board_config_t *config) hp_priv->hw_version = HW_VERSION_82341C; hp_priv->io_region_offset = 0x400; } - pr_info("hp_82341: base io 0x%p\n", iobase); + pr_info("hp_82341: base io 0x%u\n", iobase); for (i = 0; i < hp_82341_num_io_regions; ++i) { - start_addr = (unsigned long)(iobase) + i * hp_priv->io_region_offset; + start_addr = iobase + i * hp_priv->io_region_offset; if (!request_region(start_addr, hp_82341_region_iosize, "hp_82341")) { pr_err("hp_82341: failed to allocate io ports 0x%lx-0x%lx\n", start_addr, @@ -725,7 +725,7 @@ int hp_82341_attach(gpib_board_t *board, const gpib_board_config_t *config) } hp_priv->iobase[i] = start_addr; } - tms_priv->iobase = (void *)(hp_priv->iobase[2]); + tms_priv->iobase = hp_priv->iobase[2]; if (hp_priv->hw_version == HW_VERSION_82341D) { retval = isapnp_cfg_begin(hp_priv->pnp_dev->card->number, hp_priv->pnp_dev->number); diff --git a/drivers/staging/gpib/include/gpibP.h b/drivers/staging/gpib/include/gpibP.h index 5fc42b645ab7..b97da577ba33 100644 --- a/drivers/staging/gpib/include/gpibP.h +++ b/drivers/staging/gpib/include/gpibP.h @@ -16,6 +16,7 @@ #include <linux/fs.h> #include <linux/interrupt.h> +#include <linux/io.h> void gpib_register_driver(gpib_interface_t *interface, struct module *mod); void gpib_unregister_driver(gpib_interface_t *interface); @@ -35,16 +36,5 @@ extern gpib_board_t board_array[GPIB_MAX_NUM_BOARDS]; extern struct list_head registered_drivers; -#include <linux/io.h> - -void writeb_wrapper(unsigned int value, void *address); -unsigned int readb_wrapper(void *address); -void outb_wrapper(unsigned int value, void *address); -unsigned int inb_wrapper(void *address); -void writew_wrapper(unsigned int value, void *address); -unsigned int readw_wrapper(void *address); -void outw_wrapper(unsigned int value, void *address); -unsigned int inw_wrapper(void *address); - #endif // _GPIB_P_H diff --git a/drivers/staging/gpib/include/gpib_types.h b/drivers/staging/gpib/include/gpib_types.h index ee2643da6d71..b41781a55a60 100644 --- a/drivers/staging/gpib/include/gpib_types.h +++ b/drivers/staging/gpib/include/gpib_types.h @@ -31,7 +31,8 @@ typedef struct { void *init_data; int init_data_length; /* IO base address to use for non-pnp cards (set by core, driver should make local copy) */ - void *ibbase; + u32 ibbase; + void __iomem *mmibbase; /* IRQ to use for non-pnp cards (set by core, driver should make local copy) */ unsigned int ibirq; /* dma channel to use for non-pnp cards (set by core, driver should make local copy) */ diff --git a/drivers/staging/gpib/include/nec7210.h b/drivers/staging/gpib/include/nec7210.h index c00aba4ce846..ca998c4a84bf 100644 --- a/drivers/staging/gpib/include/nec7210.h +++ b/drivers/staging/gpib/include/nec7210.h @@ -18,7 +18,10 @@ /* struct used to provide variables local to a nec7210 chip */ struct nec7210_priv { - void *iobase; +#ifdef CONFIG_HAS_IOPORT + u32 iobase; +#endif + void __iomem *mmiobase; unsigned int offset; // offset between successive nec7210 io addresses unsigned int dma_channel; u8 *dma_buffer; diff --git a/drivers/staging/gpib/include/tms9914.h b/drivers/staging/gpib/include/tms9914.h index 456b488212d2..d8c8d1c9b131 100644 --- a/drivers/staging/gpib/include/tms9914.h +++ b/drivers/staging/gpib/include/tms9914.h @@ -20,7 +20,10 @@ enum tms9914_holdoff_mode { /* struct used to provide variables local to a tms9914 chip */ struct tms9914_priv { - void *iobase; +#ifdef CONFIG_HAS_IOPORT + u32 iobase; +#endif + void __iomem *mmiobase; unsigned int offset; // offset between successive tms9914 io addresses unsigned int dma_channel; // software copy of bits written to interrupt mask registers diff --git a/drivers/staging/gpib/ines/Makefile b/drivers/staging/gpib/ines/Makefile index cdcaa59a4e39..6b6e480fd811 100644 --- a/drivers/staging/gpib/ines/Makefile +++ b/drivers/staging/gpib/ines/Makefile @@ -1,4 +1,4 @@ ccflags-$(CONFIG_GPIB_PCMCIA) := -DGPIB_PCMCIA -obj-m += ines_gpib.o +obj-$(CONFIG_GPIB_INES) += ines_gpib.o diff --git a/drivers/staging/gpib/ines/ines.h b/drivers/staging/gpib/ines/ines.h index 7e8302619998..eed038fd3f28 100644 --- a/drivers/staging/gpib/ines/ines.h +++ b/drivers/staging/gpib/ines/ines.h @@ -83,14 +83,14 @@ void ines_set_xfer_counter(struct ines_priv *priv, unsigned int count); /* inb/outb wrappers */ static inline unsigned int ines_inb(struct ines_priv *priv, unsigned int register_number) { - return inb((unsigned long)(priv->nec7210_priv.iobase) + + return inb(priv->nec7210_priv.iobase + register_number * priv->nec7210_priv.offset); } static inline void ines_outb(struct ines_priv *priv, unsigned int value, unsigned int register_number) { - outb(value, (unsigned long)(priv->nec7210_priv.iobase) + + outb(value, priv->nec7210_priv.iobase + register_number * priv->nec7210_priv.offset); } diff --git a/drivers/staging/gpib/ines/ines_gpib.c b/drivers/staging/gpib/ines/ines_gpib.c index 9d8387c3bf01..e18455ba842f 100644 --- a/drivers/staging/gpib/ines/ines_gpib.c +++ b/drivers/staging/gpib/ines/ines_gpib.c @@ -273,10 +273,10 @@ irqreturn_t ines_pci_interrupt(int irq, void *arg) struct nec7210_priv *nec_priv = &priv->nec7210_priv; if (priv->pci_chip_type == PCI_CHIP_QUANCOM) { - if ((inb((unsigned long)nec_priv->iobase + + if ((inb(nec_priv->iobase + QUANCOM_IRQ_CONTROL_STATUS_REG) & QUANCOM_IRQ_ASSERTED_BIT)) - outb(QUANCOM_IRQ_ENABLE_BIT, (unsigned long)(nec_priv->iobase) + + outb(QUANCOM_IRQ_ENABLE_BIT, nec_priv->iobase + QUANCOM_IRQ_CONTROL_STATUS_REG); } @@ -780,8 +780,8 @@ static int ines_common_pci_attach(gpib_board_t *board, const gpib_board_config_t if (pci_request_regions(ines_priv->pci_device, "ines-gpib")) return -1; - nec_priv->iobase = (void *)(pci_resource_start(ines_priv->pci_device, - found_id.gpib_region)); + nec_priv->iobase = pci_resource_start(ines_priv->pci_device, + found_id.gpib_region); ines_priv->pci_chip_type = found_id.pci_chip_type; nec_priv->offset = found_id.io_offset; @@ -840,7 +840,7 @@ static int ines_common_pci_attach(gpib_board_t *board, const gpib_board_config_t } break; case PCI_CHIP_QUANCOM: - outb(QUANCOM_IRQ_ENABLE_BIT, (unsigned long)(nec_priv->iobase) + + outb(QUANCOM_IRQ_ENABLE_BIT, nec_priv->iobase + QUANCOM_IRQ_CONTROL_STATUS_REG); break; case PCI_CHIP_QUICKLOGIC5030: @@ -899,8 +899,8 @@ int ines_isa_attach(gpib_board_t *board, const gpib_board_config_t *config) ines_priv = board->private_data; nec_priv = &ines_priv->nec7210_priv; - if (!request_region((unsigned long)config->ibbase, ines_isa_iosize, "ines_gpib")) { - pr_err("ines_gpib: ioports at 0x%p already in use\n", config->ibbase); + if (!request_region(config->ibbase, ines_isa_iosize, "ines_gpib")) { + pr_err("ines_gpib: ioports at 0x%x already in use\n", config->ibbase); return -1; } nec_priv->iobase = config->ibbase; @@ -931,7 +931,7 @@ void ines_pci_detach(gpib_board_t *board) break; case PCI_CHIP_QUANCOM: if (nec_priv->iobase) - outb(0, (unsigned long)(nec_priv->iobase) + + outb(0, nec_priv->iobase + QUANCOM_IRQ_CONTROL_STATUS_REG); break; default: @@ -960,7 +960,7 @@ void ines_isa_detach(gpib_board_t *board) free_irq(ines_priv->irq, board); if (nec_priv->iobase) { nec7210_board_reset(nec_priv, board); - release_region((unsigned long)(nec_priv->iobase), ines_isa_iosize); + release_region(nec_priv->iobase, ines_isa_iosize); } } ines_free_private(board); @@ -1355,7 +1355,7 @@ int ines_common_pcmcia_attach(gpib_board_t *board) return -1; } - nec_priv->iobase = (void *)(unsigned long)curr_dev->resource[0]->start; + nec_priv->iobase = curr_dev->resource[0]->start; nec7210_board_reset(nec_priv, board); @@ -1410,7 +1410,7 @@ void ines_pcmcia_detach(gpib_board_t *board) free_irq(ines_priv->irq, board); if (nec_priv->iobase) { nec7210_board_reset(nec_priv, board); - release_region((unsigned long)(nec_priv->iobase), ines_pcmcia_iosize); + release_region(nec_priv->iobase, ines_pcmcia_iosize); } } ines_free_private(board); diff --git a/drivers/staging/gpib/lpvo_usb_gpib/Makefile b/drivers/staging/gpib/lpvo_usb_gpib/Makefile index 137511acce63..360553488e6d 100644 --- a/drivers/staging/gpib/lpvo_usb_gpib/Makefile +++ b/drivers/staging/gpib/lpvo_usb_gpib/Makefile @@ -1,3 +1,3 @@ -obj-m += lpvo_usb_gpib.o +obj-$(CONFIG_GPIB_LPVO) += lpvo_usb_gpib.o diff --git a/drivers/staging/gpib/lpvo_usb_gpib/lpvo_usb_gpib.c b/drivers/staging/gpib/lpvo_usb_gpib/lpvo_usb_gpib.c index 267651a15fa0..93a05c792816 100644 --- a/drivers/staging/gpib/lpvo_usb_gpib/lpvo_usb_gpib.c +++ b/drivers/staging/gpib/lpvo_usb_gpib/lpvo_usb_gpib.c @@ -99,8 +99,8 @@ module_param(debug, int, 0644); #define USB_GPIB_DEBUG_ON "\nIBDE\xAA\n" #define USB_GPIB_SET_LISTEN "\nIBDT0\n" #define USB_GPIB_SET_TALK "\nIBDT1\n" -#define USB_GPIB_SET_LINES "\nIBDC\n" -#define USB_GPIB_SET_DATA "\nIBDM\n" +#define USB_GPIB_SET_LINES "\nIBDC.\n" +#define USB_GPIB_SET_DATA "\nIBDM.\n" #define USB_GPIB_READ_LINES "\nIBD?C\n" #define USB_GPIB_READ_DATA "\nIBD?M\n" #define USB_GPIB_READ_BUS "\nIBD??\n" @@ -210,7 +210,7 @@ static int skel_do_release(gpib_board_t *); * (unix time in sec and NANOsec) */ -inline int usec_diff(struct timespec64 *a, struct timespec64 *b) +static inline int usec_diff(struct timespec64 *a, struct timespec64 *b) { return ((a->tv_sec - b->tv_sec) * 1000000 + (a->tv_nsec - b->tv_nsec) / 1000); @@ -436,7 +436,7 @@ static void set_timeout(gpib_board_t *board) static int usb_gpib_attach(gpib_board_t *board, const gpib_board_config_t *config) { int retval, j; - int base = (long)config->ibbase; + u32 base = config->ibbase; char *device_path; int match; struct usb_device *udev; @@ -589,7 +589,7 @@ static int usb_gpib_command(gpib_board_t *board, size_t *bytes_written) { int i, retval; - char command[6] = "IBc\n"; + char command[6] = "IBc.\n"; DIA_LOG(1, "enter %p\n", board); @@ -608,7 +608,7 @@ static int usb_gpib_command(gpib_board_t *board, } /** - * disable_eos() - Disable END on eos byte (END on EOI only) + * usb_gpib_disable_eos() - Disable END on eos byte (END on EOI only) * * @board: the gpib_board data area for this gpib interface * @@ -624,7 +624,7 @@ static void usb_gpib_disable_eos(gpib_board_t *board) } /** - * enable_eos() - Enable END for reads when eos byte is received. + * usb_gpib_enable_eos() - Enable END for reads when eos byte is received. * * @board: the gpib_board data area for this gpib interface * @eos_byte: the 'eos' byte @@ -647,7 +647,7 @@ static int usb_gpib_enable_eos(gpib_board_t *board, } /** - * go_to_standby() - De-assert ATN + * usb_gpib_go_to_standby() - De-assert ATN * * @board: the gpib_board data area for this gpib interface */ @@ -664,7 +664,7 @@ static int usb_gpib_go_to_standby(gpib_board_t *board) } /** - * interface_clear() - Assert or de-assert IFC + * usb_gpib_interface_clear() - Assert or de-assert IFC * * @board: the gpib_board data area for this gpib interface * assert: 1: assert IFC; 0: de-assert IFC diff --git a/drivers/staging/gpib/nec7210/nec7210.c b/drivers/staging/gpib/nec7210/nec7210.c index 1d9951035497..c9a837fad96e 100644 --- a/drivers/staging/gpib/nec7210/nec7210.c +++ b/drivers/staging/gpib/nec7210/nec7210.c @@ -1035,7 +1035,7 @@ EXPORT_SYMBOL(nec7210_board_online); /* wrappers for io */ uint8_t nec7210_ioport_read_byte(struct nec7210_priv *priv, unsigned int register_num) { - return inb((unsigned long)(priv->iobase) + register_num * priv->offset); + return inb(priv->iobase + register_num * priv->offset); } EXPORT_SYMBOL(nec7210_ioport_read_byte); @@ -1047,7 +1047,7 @@ void nec7210_ioport_write_byte(struct nec7210_priv *priv, uint8_t data, unsigned */ nec7210_locking_ioport_write_byte(priv, data, register_num); else - outb(data, (unsigned long)(priv->iobase) + register_num * priv->offset); + outb(data, priv->iobase + register_num * priv->offset); } EXPORT_SYMBOL(nec7210_ioport_write_byte); @@ -1058,7 +1058,7 @@ uint8_t nec7210_locking_ioport_read_byte(struct nec7210_priv *priv, unsigned int unsigned long flags; spin_lock_irqsave(&priv->register_page_lock, flags); - retval = inb((unsigned long)(priv->iobase) + register_num * priv->offset); + retval = inb(priv->iobase + register_num * priv->offset); spin_unlock_irqrestore(&priv->register_page_lock, flags); return retval; } @@ -1072,7 +1072,7 @@ void nec7210_locking_ioport_write_byte(struct nec7210_priv *priv, uint8_t data, spin_lock_irqsave(&priv->register_page_lock, flags); if (register_num == AUXMR) udelay(1); - outb(data, (unsigned long)(priv->iobase) + register_num * priv->offset); + outb(data, priv->iobase + register_num * priv->offset); spin_unlock_irqrestore(&priv->register_page_lock, flags); } EXPORT_SYMBOL(nec7210_locking_ioport_write_byte); @@ -1080,7 +1080,7 @@ EXPORT_SYMBOL(nec7210_locking_ioport_write_byte); uint8_t nec7210_iomem_read_byte(struct nec7210_priv *priv, unsigned int register_num) { - return readb(priv->iobase + register_num * priv->offset); + return readb(priv->mmiobase + register_num * priv->offset); } EXPORT_SYMBOL(nec7210_iomem_read_byte); @@ -1092,7 +1092,7 @@ void nec7210_iomem_write_byte(struct nec7210_priv *priv, uint8_t data, unsigned */ nec7210_locking_iomem_write_byte(priv, data, register_num); else - writeb(data, priv->iobase + register_num * priv->offset); + writeb(data, priv->mmiobase + register_num * priv->offset); } EXPORT_SYMBOL(nec7210_iomem_write_byte); @@ -1102,7 +1102,7 @@ uint8_t nec7210_locking_iomem_read_byte(struct nec7210_priv *priv, unsigned int unsigned long flags; spin_lock_irqsave(&priv->register_page_lock, flags); - retval = readb(priv->iobase + register_num * priv->offset); + retval = readb(priv->mmiobase + register_num * priv->offset); spin_unlock_irqrestore(&priv->register_page_lock, flags); return retval; } @@ -1116,7 +1116,7 @@ void nec7210_locking_iomem_write_byte(struct nec7210_priv *priv, uint8_t data, spin_lock_irqsave(&priv->register_page_lock, flags); if (register_num == AUXMR) udelay(1); - writeb(data, priv->iobase + register_num * priv->offset); + writeb(data, priv->mmiobase + register_num * priv->offset); spin_unlock_irqrestore(&priv->register_page_lock, flags); } EXPORT_SYMBOL(nec7210_locking_iomem_write_byte); diff --git a/drivers/staging/gpib/ni_usb/Makefile b/drivers/staging/gpib/ni_usb/Makefile index e22b3b21a62c..469c5d16add3 100644 --- a/drivers/staging/gpib/ni_usb/Makefile +++ b/drivers/staging/gpib/ni_usb/Makefile @@ -1,4 +1,4 @@ -obj-m += ni_usb_gpib.o +obj-$(CONFIG_GPIB_NI_USB) += ni_usb_gpib.o diff --git a/drivers/staging/gpib/pc2/Makefile b/drivers/staging/gpib/pc2/Makefile index 8148425e0f87..481ee4296e1b 100644 --- a/drivers/staging/gpib/pc2/Makefile +++ b/drivers/staging/gpib/pc2/Makefile @@ -1,5 +1,5 @@ -obj-m += pc2_gpib.o +obj-$(CONFIG_GPIB_PC2) += pc2_gpib.o diff --git a/drivers/staging/gpib/pc2/pc2_gpib.c b/drivers/staging/gpib/pc2/pc2_gpib.c index 7b3b34f47341..d22af25125b1 100644 --- a/drivers/staging/gpib/pc2/pc2_gpib.c +++ b/drivers/staging/gpib/pc2/pc2_gpib.c @@ -426,7 +426,7 @@ int pc2_attach(gpib_board_t *board, const gpib_board_config_t *config) nec_priv = &pc2_priv->nec7210_priv; nec_priv->offset = pc2_reg_offset; - if (request_region((unsigned long)config->ibbase, pc2_iosize, "pc2") == 0) { + if (request_region(config->ibbase, pc2_iosize, "pc2") == 0) { pr_err("gpib: ioports are already in use\n"); return -1; } @@ -471,7 +471,7 @@ void pc2_detach(gpib_board_t *board) free_irq(pc2_priv->irq, board); if (nec_priv->iobase) { nec7210_board_reset(nec_priv, board); - release_region((unsigned long)(nec_priv->iobase), pc2_iosize); + release_region(nec_priv->iobase, pc2_iosize); } if (nec_priv->dma_buffer) { dma_free_coherent(board->dev, nec_priv->dma_buffer_length, @@ -498,14 +498,14 @@ static int pc2a_common_attach(gpib_board_t *board, const gpib_board_config_t *co nec_priv = &pc2_priv->nec7210_priv; nec_priv->offset = pc2a_reg_offset; - switch ((unsigned long)(config->ibbase)) { + switch (config->ibbase) { case 0x02e1: case 0x22e1: case 0x42e1: case 0x62e1: break; default: - pr_err("PCIIa base range invalid, must be one of 0x[0246]2e1, but is 0x%p\n", + pr_err("PCIIa base range invalid, must be one of 0x[0246]2e1, but is 0x%d\n", config->ibbase); return -1; } @@ -522,7 +522,7 @@ static int pc2a_common_attach(gpib_board_t *board, const gpib_board_config_t *co unsigned int err = 0; for (i = 0; i < num_registers; i++) { - if (check_region((unsigned long)config->ibbase + i * pc2a_reg_offset, 1)) + if (check_region(config->ibbase + i * pc2a_reg_offset, 1)) err++; } if (config->ibirq && check_region(pc2a_clear_intr_iobase + config->ibirq, 1)) @@ -533,11 +533,11 @@ static int pc2a_common_attach(gpib_board_t *board, const gpib_board_config_t *co } #endif for (i = 0; i < num_registers; i++) { - if (!request_region((unsigned long)config->ibbase + + if (!request_region(config->ibbase + i * pc2a_reg_offset, 1, "pc2a")) { pr_err("gpib: ioports are already in use"); for (j = 0; j < i; j++) - release_region((unsigned long)(config->ibbase) + + release_region(config->ibbase + j * pc2a_reg_offset, 1); return -1; } @@ -608,7 +608,7 @@ static void pc2a_common_detach(gpib_board_t *board, unsigned int num_registers) if (nec_priv->iobase) { nec7210_board_reset(nec_priv, board); for (i = 0; i < num_registers; i++) - release_region((unsigned long)nec_priv->iobase + + release_region(nec_priv->iobase + i * pc2a_reg_offset, 1); } if (pc2_priv->clear_intr_addr) diff --git a/drivers/staging/gpib/tms9914/Makefile b/drivers/staging/gpib/tms9914/Makefile index 81b7e3cf104c..4705ab07f413 100644 --- a/drivers/staging/gpib/tms9914/Makefile +++ b/drivers/staging/gpib/tms9914/Makefile @@ -1,5 +1,5 @@ -obj-m += tms9914.o +obj-$(CONFIG_GPIB_TMS9914) += tms9914.o diff --git a/drivers/staging/gpib/tms9914/tms9914.c b/drivers/staging/gpib/tms9914/tms9914.c index 152b243b845b..ec8e1d4d762f 100644 --- a/drivers/staging/gpib/tms9914/tms9914.c +++ b/drivers/staging/gpib/tms9914/tms9914.c @@ -866,14 +866,14 @@ EXPORT_SYMBOL_GPL(tms9914_online); // wrapper for inb uint8_t tms9914_ioport_read_byte(struct tms9914_priv *priv, unsigned int register_num) { - return inb((unsigned long)(priv->iobase) + register_num * priv->offset); + return inb(priv->iobase + register_num * priv->offset); } EXPORT_SYMBOL_GPL(tms9914_ioport_read_byte); // wrapper for outb void tms9914_ioport_write_byte(struct tms9914_priv *priv, uint8_t data, unsigned int register_num) { - outb(data, (unsigned long)(priv->iobase) + register_num * priv->offset); + outb(data, priv->iobase + register_num * priv->offset); if (register_num == AUXCR) udelay(1); } @@ -883,14 +883,14 @@ EXPORT_SYMBOL_GPL(tms9914_ioport_write_byte); // wrapper for readb uint8_t tms9914_iomem_read_byte(struct tms9914_priv *priv, unsigned int register_num) { - return readb(priv->iobase + register_num * priv->offset); + return readb(priv->mmiobase + register_num * priv->offset); } EXPORT_SYMBOL_GPL(tms9914_iomem_read_byte); // wrapper for writeb void tms9914_iomem_write_byte(struct tms9914_priv *priv, uint8_t data, unsigned int register_num) { - writeb(data, priv->iobase + register_num * priv->offset); + writeb(data, priv->mmiobase + register_num * priv->offset); if (register_num == AUXCR) udelay(1); } diff --git a/drivers/staging/gpib/tnt4882/Makefile b/drivers/staging/gpib/tnt4882/Makefile index f767c990db7a..a3c3fb96d5ed 100644 --- a/drivers/staging/gpib/tnt4882/Makefile +++ b/drivers/staging/gpib/tnt4882/Makefile @@ -1,5 +1,5 @@ ccflags-$(CONFIG_GPIB_PCMCIA) := -DGPIB_PCMCIA -obj-m += tnt4882.o +obj-$(CONFIG_GPIB_NI_PCI_ISA) += tnt4882.o tnt4882-objs := tnt4882_gpib.o mite.o diff --git a/drivers/staging/gpib/tnt4882/mite.c b/drivers/staging/gpib/tnt4882/mite.c index 0edf34d243e9..4bd352967616 100644 --- a/drivers/staging/gpib/tnt4882/mite.c +++ b/drivers/staging/gpib/tnt4882/mite.c @@ -148,72 +148,3 @@ void mite_list_devices(void) } pr_info("\n"); } - -int mite_bytes_transferred(struct mite_struct *mite, int chan) -{ - int dar, fcr; - - dar = readl(mite->mite_io_addr + MITE_DAR + CHAN_OFFSET(chan)); - fcr = readl(mite->mite_io_addr + MITE_FCR + CHAN_OFFSET(chan)) & 0x000000FF; - return dar - fcr; -} - -int mite_dma_tcr(struct mite_struct *mite) -{ - int tcr; - int lkar; - - lkar = readl(mite->mite_io_addr + CHAN_OFFSET(0) + MITE_LKAR); - tcr = readl(mite->mite_io_addr + CHAN_OFFSET(0) + MITE_TCR); - MDPRINTK("lkar=0x%08x tcr=%d\n", lkar, tcr); - - return tcr; -} - -void mite_dma_disarm(struct mite_struct *mite) -{ - int chor; - - /* disarm */ - chor = CHOR_ABORT; - writel(chor, mite->mite_io_addr + CHAN_OFFSET(0) + MITE_CHOR); -} - -void mite_dump_regs(struct mite_struct *mite) -{ - void *addr = 0; - unsigned long temp = 0; - - pr_info("mite address is =0x%p\n", mite->mite_io_addr); - - addr = mite->mite_io_addr + MITE_CHOR + CHAN_OFFSET(0); - pr_info("mite status[CHOR]at 0x%p =0x%08lx\n", addr, temp = readl(addr)); - //mite_decode(mite_CHOR_strings,temp); - addr = mite->mite_io_addr + MITE_CHCR + CHAN_OFFSET(0); - pr_info("mite status[CHCR]at 0x%p =0x%08lx\n", addr, temp = readl(addr)); - //mite_decode(mite_CHCR_strings,temp); - addr = mite->mite_io_addr + MITE_TCR + CHAN_OFFSET(0); - pr_info("mite status[TCR] at 0x%p =0x%08x\n", addr, readl(addr)); - addr = mite->mite_io_addr + MITE_MCR + CHAN_OFFSET(0); - pr_info("mite status[MCR] at 0x%p =0x%08lx\n", addr, temp = readl(addr)); - //mite_decode(mite_MCR_strings,temp); - addr = mite->mite_io_addr + MITE_MAR + CHAN_OFFSET(0); - pr_info("mite status[MAR] at 0x%p =0x%08x\n", addr, readl(addr)); - addr = mite->mite_io_addr + MITE_DCR + CHAN_OFFSET(0); - pr_info("mite status[DCR] at 0x%p =0x%08lx\n", addr, temp = readl(addr)); - //mite_decode(mite_CR_strings,temp); - addr = mite->mite_io_addr + MITE_DAR + CHAN_OFFSET(0); - pr_info("mite status[DAR] at 0x%p =0x%08x\n", addr, readl(addr)); - addr = mite->mite_io_addr + MITE_LKCR + CHAN_OFFSET(0); - pr_info("mite status[LKCR]at 0x%p =0x%08lx\n", addr, temp = readl(addr)); - //mite_decode(mite_CR_strings,temp); - addr = mite->mite_io_addr + MITE_LKAR + CHAN_OFFSET(0); - pr_info("mite status[LKAR]at 0x%p =0x%08x\n", addr, readl(addr)); - - addr = mite->mite_io_addr + MITE_CHSR + CHAN_OFFSET(0); - pr_info("mite status[CHSR]at 0x%p =0x%08lx\n", addr, temp = readl(addr)); - //mite_decode(mite_CHSR_strings,temp); - addr = mite->mite_io_addr + MITE_FCR + CHAN_OFFSET(0); - pr_info("mite status[FCR] at 0x%p =0x%08x\n\n", addr, readl(addr)); -} - diff --git a/drivers/staging/gpib/tnt4882/mite.h b/drivers/staging/gpib/tnt4882/mite.h index 6454d069b8cc..edb873435b51 100644 --- a/drivers/staging/gpib/tnt4882/mite.h +++ b/drivers/staging/gpib/tnt4882/mite.h @@ -34,9 +34,9 @@ struct mite_struct { struct pci_dev *pcidev; unsigned long mite_phys_addr; - void *mite_io_addr; + void __iomem *mite_io_addr; unsigned long daq_phys_addr; - void *daq_io_addr; + void __iomem *daq_io_addr; int DMA_CheckNearEnd; @@ -61,15 +61,6 @@ int mite_setup(struct mite_struct *mite); void mite_unsetup(struct mite_struct *mite); void mite_list_devices(void); -int mite_dma_tcr(struct mite_struct *mite); - -void mite_dma_arm(struct mite_struct *mite); -void mite_dma_disarm(struct mite_struct *mite); - -void mite_dump_regs(struct mite_struct *mite); -void mite_setregs(struct mite_struct *mite, unsigned long ll_start, int chan, int dir); -int mite_bytes_transferred(struct mite_struct *mite, int chan); - #define CHAN_OFFSET(x) (0x100 * (x)) /* DMA base for chan 0 is 0x500, chan 1 is 0x600 */ diff --git a/drivers/staging/gpib/tnt4882/tnt4882_gpib.c b/drivers/staging/gpib/tnt4882/tnt4882_gpib.c index e49a952fa0d8..ce91c3eb768c 100644 --- a/drivers/staging/gpib/tnt4882/tnt4882_gpib.c +++ b/drivers/staging/gpib/tnt4882/tnt4882_gpib.c @@ -45,10 +45,6 @@ struct tnt4882_priv { unsigned short imr0_bits; unsigned short imr3_bits; unsigned short auxg_bits; // bits written to auxiliary register G - void (*io_writeb)(unsigned int value, void *address); - void (*io_writew)(unsigned int value, void *address); - unsigned int (*io_readb)(void *address); - unsigned int (*io_readw)(void *address); }; // interface functions @@ -104,23 +100,23 @@ static const int atgpib_iosize = 32; /* paged io */ static inline unsigned int tnt_paged_readb(struct tnt4882_priv *priv, unsigned long offset) { - priv->io_writeb(AUX_PAGEIN, priv->nec7210_priv.iobase + AUXMR * priv->nec7210_priv.offset); + iowrite8(AUX_PAGEIN, priv->nec7210_priv.mmiobase + AUXMR * priv->nec7210_priv.offset); udelay(1); - return priv->io_readb(priv->nec7210_priv.iobase + offset); + return ioread8(priv->nec7210_priv.mmiobase + offset); } static inline void tnt_paged_writeb(struct tnt4882_priv *priv, unsigned int value, unsigned long offset) { - priv->io_writeb(AUX_PAGEIN, priv->nec7210_priv.iobase + AUXMR * priv->nec7210_priv.offset); + iowrite8(AUX_PAGEIN, priv->nec7210_priv.mmiobase + AUXMR * priv->nec7210_priv.offset); udelay(1); - priv->io_writeb(value, priv->nec7210_priv.iobase + offset); + iowrite8(value, priv->nec7210_priv.mmiobase + offset); } /* readb/writeb wrappers */ static inline unsigned short tnt_readb(struct tnt4882_priv *priv, unsigned long offset) { - void *address = priv->nec7210_priv.iobase + offset; + void *address = priv->nec7210_priv.mmiobase + offset; unsigned long flags; unsigned short retval; spinlock_t *register_lock = &priv->nec7210_priv.register_page_lock; @@ -134,7 +130,7 @@ static inline unsigned short tnt_readb(struct tnt4882_priv *priv, unsigned long switch (priv->nec7210_priv.type) { case TNT4882: case TNT5004: - retval = priv->io_readb(address); + retval = ioread8(address); break; case NAT4882: retval = tnt_paged_readb(priv, offset - tnt_pagein_offset); @@ -149,7 +145,7 @@ static inline unsigned short tnt_readb(struct tnt4882_priv *priv, unsigned long } break; default: - retval = priv->io_readb(address); + retval = ioread8(address); break; } spin_unlock_irqrestore(register_lock, flags); @@ -158,7 +154,7 @@ static inline unsigned short tnt_readb(struct tnt4882_priv *priv, unsigned long static inline void tnt_writeb(struct tnt4882_priv *priv, unsigned short value, unsigned long offset) { - void *address = priv->nec7210_priv.iobase + offset; + void *address = priv->nec7210_priv.mmiobase + offset; unsigned long flags; spinlock_t *register_lock = &priv->nec7210_priv.register_page_lock; @@ -170,7 +166,7 @@ static inline void tnt_writeb(struct tnt4882_priv *priv, unsigned short value, u switch (priv->nec7210_priv.type) { case TNT4882: case TNT5004: - priv->io_writeb(value, address); + iowrite8(value, address); break; case NAT4882: tnt_paged_writeb(priv, value, offset - tnt_pagein_offset); @@ -183,7 +179,7 @@ static inline void tnt_writeb(struct tnt4882_priv *priv, unsigned short value, u } break; default: - priv->io_writeb(value, address); + iowrite8(value, address); break; } spin_unlock_irqrestore(register_lock, flags); @@ -288,7 +284,7 @@ static int drain_fifo_words(struct tnt4882_priv *tnt_priv, uint8_t *buffer, int while (fifo_word_available(tnt_priv) && count + 2 <= num_bytes) { short word; - word = tnt_priv->io_readw(nec_priv->iobase + FIFOB); + word = ioread16(nec_priv->mmiobase + FIFOB); buffer[count++] = word & 0xff; buffer[count++] = (word >> 8) & 0xff; } @@ -573,7 +569,7 @@ static int generic_write(gpib_board_t *board, uint8_t *buffer, size_t length, word = buffer[count++] & 0xff; if (count < length) word |= (buffer[count++] << 8) & 0xff00; - tnt_priv->io_writew(word, nec_priv->iobase + FIFOB); + iowrite16(word, nec_priv->mmiobase + FIFOB); } // avoid unnecessary HR_NFF interrupts // tnt_priv->imr3_bits |= HR_NFF; @@ -1269,10 +1265,6 @@ int ni_pci_attach(gpib_board_t *board, const gpib_board_config_t *config) if (tnt4882_allocate_private(board)) return -ENOMEM; tnt_priv = board->private_data; - tnt_priv->io_writeb = writeb_wrapper; - tnt_priv->io_readb = readb_wrapper; - tnt_priv->io_writew = writew_wrapper; - tnt_priv->io_readw = readw_wrapper; nec_priv = &tnt_priv->nec7210_priv; nec_priv->type = TNT4882; nec_priv->read_byte = nec7210_locking_iomem_read_byte; @@ -1324,7 +1316,7 @@ int ni_pci_attach(gpib_board_t *board, const gpib_board_config_t *config) return retval; } - nec_priv->iobase = tnt_priv->mite->daq_io_addr; + nec_priv->mmiobase = tnt_priv->mite->daq_io_addr; // get irq if (request_irq(mite_irq(tnt_priv->mite), tnt4882_interrupt, isr_flags, @@ -1359,7 +1351,7 @@ void ni_pci_detach(gpib_board_t *board) if (tnt_priv) { nec_priv = &tnt_priv->nec7210_priv; - if (nec_priv->iobase) + if (nec_priv->mmiobase) tnt4882_board_reset(tnt_priv, board); if (tnt_priv->irq) free_irq(tnt_priv->irq, board); @@ -1400,7 +1392,7 @@ static int ni_isa_attach_common(gpib_board_t *board, const gpib_board_config_t * struct tnt4882_priv *tnt_priv; struct nec7210_priv *nec_priv; int isr_flags = 0; - void *iobase; + u32 iobase; int irq; board->status = 0; @@ -1408,10 +1400,6 @@ static int ni_isa_attach_common(gpib_board_t *board, const gpib_board_config_t * if (tnt4882_allocate_private(board)) return -ENOMEM; tnt_priv = board->private_data; - tnt_priv->io_writeb = outb_wrapper; - tnt_priv->io_readb = inb_wrapper; - tnt_priv->io_writew = outw_wrapper; - tnt_priv->io_readw = inw_wrapper; nec_priv = &tnt_priv->nec7210_priv; nec_priv->type = chipset; nec_priv->read_byte = nec7210_locking_ioport_read_byte; @@ -1427,18 +1415,20 @@ static int ni_isa_attach_common(gpib_board_t *board, const gpib_board_config_t * if (retval < 0) return retval; tnt_priv->pnp_dev = dev; - iobase = (void *)(pnp_port_start(dev, 0)); + iobase = pnp_port_start(dev, 0); irq = pnp_irq(dev, 0); } else { iobase = config->ibbase; irq = config->ibirq; } // allocate ioports - if (!request_region((unsigned long)(iobase), atgpib_iosize, "atgpib")) { + if (!request_region(iobase, atgpib_iosize, "atgpib")) { pr_err("tnt4882: failed to allocate ioports\n"); return -1; } - nec_priv->iobase = iobase; + nec_priv->mmiobase = ioport_map(iobase, atgpib_iosize); + if (!nec_priv->mmiobase) + return -1; // get irq if (request_irq(irq, tnt4882_interrupt, isr_flags, "atgpib", board)) { @@ -1478,8 +1468,10 @@ void ni_isa_detach(gpib_board_t *board) tnt4882_board_reset(tnt_priv, board); if (tnt_priv->irq) free_irq(tnt_priv->irq, board); + if (nec_priv->mmiobase) + ioport_unmap(nec_priv->mmiobase); if (nec_priv->iobase) - release_region((unsigned long)(nec_priv->iobase), atgpib_iosize); + release_region(nec_priv->iobase, atgpib_iosize); if (tnt_priv->pnp_dev) pnp_device_detach(tnt_priv->pnp_dev); } @@ -1817,10 +1809,6 @@ int ni_pcmcia_attach(gpib_board_t *board, const gpib_board_config_t *config) if (tnt4882_allocate_private(board)) return -ENOMEM; tnt_priv = board->private_data; - tnt_priv->io_writeb = outb_wrapper; - tnt_priv->io_readb = inb_wrapper; - tnt_priv->io_writew = outw_wrapper; - tnt_priv->io_readw = inw_wrapper; nec_priv = &tnt_priv->nec7210_priv; nec_priv->type = TNT4882; nec_priv->read_byte = nec7210_locking_ioport_read_byte; @@ -1835,7 +1823,10 @@ int ni_pcmcia_attach(gpib_board_t *board, const gpib_board_config_t *config) return -EIO; } - nec_priv->iobase = (void *)(unsigned long)curr_dev->resource[0]->start; + nec_priv->mmiobase = ioport_map(curr_dev->resource[0]->start, + resource_size(curr_dev->resource[0])); + if (!nec_priv->mmiobase) + return -1; // get irq if (request_irq(curr_dev->irq, tnt4882_interrupt, isr_flags, "tnt4882", board)) { @@ -1860,9 +1851,11 @@ void ni_pcmcia_detach(gpib_board_t *board) nec_priv = &tnt_priv->nec7210_priv; if (tnt_priv->irq) free_irq(tnt_priv->irq, board); + if (nec_priv->mmiobase) + ioport_unmap(nec_priv->mmiobase); if (nec_priv->iobase) { tnt4882_board_reset(tnt_priv, board); - release_region((unsigned long)nec_priv->iobase, pcmcia_gpib_iosize); + release_region(nec_priv->iobase, pcmcia_gpib_iosize); } } tnt4882_free_private(board); diff --git a/drivers/staging/iio/frequency/ad9832.c b/drivers/staging/iio/frequency/ad9832.c index 492612e8f8ba..140ee4f9c137 100644 --- a/drivers/staging/iio/frequency/ad9832.c +++ b/drivers/staging/iio/frequency/ad9832.c @@ -158,7 +158,7 @@ static int ad9832_write_frequency(struct ad9832_state *st, static int ad9832_write_phase(struct ad9832_state *st, unsigned long addr, unsigned long phase) { - if (phase > BIT(AD9832_PHASE_BITS)) + if (phase >= BIT(AD9832_PHASE_BITS)) return -EINVAL; st->phase_data[0] = cpu_to_be16((AD9832_CMD_PHA8BITSW << CMD_SHIFT) | diff --git a/drivers/staging/iio/frequency/ad9834.c b/drivers/staging/iio/frequency/ad9834.c index 47e7d7e6d920..6e99e008c5f4 100644 --- a/drivers/staging/iio/frequency/ad9834.c +++ b/drivers/staging/iio/frequency/ad9834.c @@ -131,7 +131,7 @@ static int ad9834_write_frequency(struct ad9834_state *st, static int ad9834_write_phase(struct ad9834_state *st, unsigned long addr, unsigned long phase) { - if (phase > BIT(AD9834_PHASE_BITS)) + if (phase >= BIT(AD9834_PHASE_BITS)) return -EINVAL; st->data = cpu_to_be16(addr | phase); diff --git a/drivers/thermal/thermal_of.c b/drivers/thermal/thermal_of.c index fab11b98ca49..5ab4ce4daaeb 100644 --- a/drivers/thermal/thermal_of.c +++ b/drivers/thermal/thermal_of.c @@ -160,6 +160,7 @@ static struct device_node *of_thermal_zone_find(struct device_node *sensor, int return ERR_PTR(ret); } + of_node_put(sensor_specs.np); if ((sensor == sensor_specs.np) && id == (sensor_specs.args_count ? sensor_specs.args[0] : 0)) { pr_debug("sensor %pOFn id=%d belongs to %pOFn\n", sensor, id, child); diff --git a/drivers/tty/serial/8250/8250_core.c b/drivers/tty/serial/8250/8250_core.c index 5f9f06911795..68baf75bdadc 100644 --- a/drivers/tty/serial/8250/8250_core.c +++ b/drivers/tty/serial/8250/8250_core.c @@ -812,6 +812,9 @@ int serial8250_register_8250_port(const struct uart_8250_port *up) uart->dl_write = up->dl_write; if (uart->port.type != PORT_8250_CIR) { + if (uart_console_registered(&uart->port)) + pm_runtime_get_sync(uart->port.dev); + if (serial8250_isa_config != NULL) serial8250_isa_config(0, &uart->port, &uart->capabilities); diff --git a/drivers/tty/serial/imx.c b/drivers/tty/serial/imx.c index 17f70e4bee43..9c59ec128bb4 100644 --- a/drivers/tty/serial/imx.c +++ b/drivers/tty/serial/imx.c @@ -2692,7 +2692,7 @@ static void imx_uart_enable_wakeup(struct imx_port *sport, bool on) { u32 ucr3; - uart_port_lock(&sport->port); + uart_port_lock_irq(&sport->port); ucr3 = imx_uart_readl(sport, UCR3); if (on) { @@ -2714,7 +2714,7 @@ static void imx_uart_enable_wakeup(struct imx_port *sport, bool on) imx_uart_writel(sport, ucr1, UCR1); } - uart_port_unlock(&sport->port); + uart_port_unlock_irq(&sport->port); } static int imx_uart_suspend_noirq(struct device *dev) diff --git a/drivers/tty/serial/stm32-usart.c b/drivers/tty/serial/stm32-usart.c index 7dc254546075..1ec5d8c3aef8 100644 --- a/drivers/tty/serial/stm32-usart.c +++ b/drivers/tty/serial/stm32-usart.c @@ -1051,14 +1051,14 @@ static void stm32_usart_break_ctl(struct uart_port *port, int break_state) const struct stm32_usart_offsets *ofs = &stm32_port->info->ofs; unsigned long flags; - spin_lock_irqsave(&port->lock, flags); + uart_port_lock_irqsave(port, &flags); if (break_state) stm32_usart_set_bits(port, ofs->rqr, USART_RQR_SBKRQ); else stm32_usart_clr_bits(port, ofs->rqr, USART_RQR_SBKRQ); - spin_unlock_irqrestore(&port->lock, flags); + uart_port_unlock_irqrestore(port, flags); } static int stm32_usart_startup(struct uart_port *port) diff --git a/drivers/ufs/core/ufshcd-priv.h b/drivers/ufs/core/ufshcd-priv.h index 9ffd94ddf8c7..786f20ef2238 100644 --- a/drivers/ufs/core/ufshcd-priv.h +++ b/drivers/ufs/core/ufshcd-priv.h @@ -237,12 +237,6 @@ static inline void ufshcd_vops_config_scaling_param(struct ufs_hba *hba, hba->vops->config_scaling_param(hba, p, data); } -static inline void ufshcd_vops_reinit_notify(struct ufs_hba *hba) -{ - if (hba->vops && hba->vops->reinit_notify) - hba->vops->reinit_notify(hba); -} - static inline int ufshcd_vops_mcq_config_resource(struct ufs_hba *hba) { if (hba->vops && hba->vops->mcq_config_resource) diff --git a/drivers/ufs/core/ufshcd.c b/drivers/ufs/core/ufshcd.c index fd53c9f402c3..3094f3c89e82 100644 --- a/drivers/ufs/core/ufshcd.c +++ b/drivers/ufs/core/ufshcd.c @@ -8858,7 +8858,6 @@ static int ufshcd_probe_hba(struct ufs_hba *hba, bool init_dev_params) ufshcd_device_reset(hba); ufs_put_device_desc(hba); ufshcd_hba_stop(hba); - ufshcd_vops_reinit_notify(hba); ret = ufshcd_hba_enable(hba); if (ret) { dev_err(hba->dev, "Host controller enable failed\n"); @@ -10590,14 +10589,17 @@ int ufshcd_init(struct ufs_hba *hba, void __iomem *mmio_base, unsigned int irq) } /* - * Set the default power management level for runtime and system PM. + * Set the default power management level for runtime and system PM if + * not set by the host controller drivers. * Default power saving mode is to keep UFS link in Hibern8 state * and UFS device in sleep state. */ - hba->rpm_lvl = ufs_get_desired_pm_lvl_for_dev_link_state( + if (!hba->rpm_lvl) + hba->rpm_lvl = ufs_get_desired_pm_lvl_for_dev_link_state( UFS_SLEEP_PWR_MODE, UIC_LINK_HIBERN8_STATE); - hba->spm_lvl = ufs_get_desired_pm_lvl_for_dev_link_state( + if (!hba->spm_lvl) + hba->spm_lvl = ufs_get_desired_pm_lvl_for_dev_link_state( UFS_SLEEP_PWR_MODE, UIC_LINK_HIBERN8_STATE); diff --git a/drivers/ufs/host/ufs-qcom.c b/drivers/ufs/host/ufs-qcom.c index 68040b2ab5f8..91e94fe990b4 100644 --- a/drivers/ufs/host/ufs-qcom.c +++ b/drivers/ufs/host/ufs-qcom.c @@ -368,6 +368,11 @@ static int ufs_qcom_power_up_sequence(struct ufs_hba *hba) if (ret) return ret; + if (phy->power_count) { + phy_power_off(phy); + phy_exit(phy); + } + /* phy initialization - calibrate the phy */ ret = phy_init(phy); if (ret) { @@ -866,6 +871,7 @@ static u32 ufs_qcom_get_ufs_hci_version(struct ufs_hba *hba) */ static void ufs_qcom_advertise_quirks(struct ufs_hba *hba) { + const struct ufs_qcom_drvdata *drvdata = of_device_get_match_data(hba->dev); struct ufs_qcom_host *host = ufshcd_get_variant(hba); if (host->hw_ver.major == 0x2) @@ -874,9 +880,8 @@ static void ufs_qcom_advertise_quirks(struct ufs_hba *hba) if (host->hw_ver.major > 0x3) hba->quirks |= UFSHCD_QUIRK_REINIT_AFTER_MAX_GEAR_SWITCH; - if (of_device_is_compatible(hba->dev->of_node, "qcom,sm8550-ufshc") || - of_device_is_compatible(hba->dev->of_node, "qcom,sm8650-ufshc")) - hba->quirks |= UFSHCD_QUIRK_BROKEN_LSDBS_CAP; + if (drvdata && drvdata->quirks) + hba->quirks |= drvdata->quirks; } static void ufs_qcom_set_phy_gear(struct ufs_qcom_host *host) @@ -1064,6 +1069,7 @@ static int ufs_qcom_init(struct ufs_hba *hba) struct device *dev = hba->dev; struct ufs_qcom_host *host; struct ufs_clk_info *clki; + const struct ufs_qcom_drvdata *drvdata = of_device_get_match_data(hba->dev); host = devm_kzalloc(dev, sizeof(*host), GFP_KERNEL); if (!host) @@ -1143,6 +1149,9 @@ static int ufs_qcom_init(struct ufs_hba *hba) dev_warn(dev, "%s: failed to configure the testbus %d\n", __func__, err); + if (drvdata && drvdata->no_phy_retention) + hba->spm_lvl = UFS_PM_LVL_5; + return 0; out_variant_clear: @@ -1579,13 +1588,6 @@ static void ufs_qcom_config_scaling_param(struct ufs_hba *hba, } #endif -static void ufs_qcom_reinit_notify(struct ufs_hba *hba) -{ - struct ufs_qcom_host *host = ufshcd_get_variant(hba); - - phy_power_off(host->generic_phy); -} - /* Resources */ static const struct ufshcd_res_info ufs_res_info[RES_MAX] = { {.name = "ufs_mem",}, @@ -1825,7 +1827,6 @@ static const struct ufs_hba_variant_ops ufs_hba_qcom_vops = { .device_reset = ufs_qcom_device_reset, .config_scaling_param = ufs_qcom_config_scaling_param, .program_key = ufs_qcom_ice_program_key, - .reinit_notify = ufs_qcom_reinit_notify, .mcq_config_resource = ufs_qcom_mcq_config_resource, .get_hba_mac = ufs_qcom_get_hba_mac, .op_runtime_config = ufs_qcom_op_runtime_config, @@ -1868,9 +1869,15 @@ static void ufs_qcom_remove(struct platform_device *pdev) platform_device_msi_free_irqs_all(hba->dev); } +static const struct ufs_qcom_drvdata ufs_qcom_sm8550_drvdata = { + .quirks = UFSHCD_QUIRK_BROKEN_LSDBS_CAP, + .no_phy_retention = true, +}; + static const struct of_device_id ufs_qcom_of_match[] __maybe_unused = { { .compatible = "qcom,ufshc" }, - { .compatible = "qcom,sm8550-ufshc" }, + { .compatible = "qcom,sm8550-ufshc", .data = &ufs_qcom_sm8550_drvdata }, + { .compatible = "qcom,sm8650-ufshc", .data = &ufs_qcom_sm8550_drvdata }, {}, }; MODULE_DEVICE_TABLE(of, ufs_qcom_of_match); diff --git a/drivers/ufs/host/ufs-qcom.h b/drivers/ufs/host/ufs-qcom.h index b9de170983c9..919f53682beb 100644 --- a/drivers/ufs/host/ufs-qcom.h +++ b/drivers/ufs/host/ufs-qcom.h @@ -217,6 +217,11 @@ struct ufs_qcom_host { bool esi_enabled; }; +struct ufs_qcom_drvdata { + enum ufshcd_quirks quirks; + bool no_phy_retention; +}; + static inline u32 ufs_qcom_get_debug_reg_offset(struct ufs_qcom_host *host, u32 reg) { diff --git a/drivers/usb/chipidea/ci_hdrc_imx.c b/drivers/usb/chipidea/ci_hdrc_imx.c index f2801700be8e..1a7fc638213e 100644 --- a/drivers/usb/chipidea/ci_hdrc_imx.c +++ b/drivers/usb/chipidea/ci_hdrc_imx.c @@ -370,25 +370,29 @@ static int ci_hdrc_imx_probe(struct platform_device *pdev) data->pinctrl = devm_pinctrl_get(dev); if (PTR_ERR(data->pinctrl) == -ENODEV) data->pinctrl = NULL; - else if (IS_ERR(data->pinctrl)) - return dev_err_probe(dev, PTR_ERR(data->pinctrl), + else if (IS_ERR(data->pinctrl)) { + ret = dev_err_probe(dev, PTR_ERR(data->pinctrl), "pinctrl get failed\n"); + goto err_put; + } data->hsic_pad_regulator = devm_regulator_get_optional(dev, "hsic"); if (PTR_ERR(data->hsic_pad_regulator) == -ENODEV) { /* no pad regulator is needed */ data->hsic_pad_regulator = NULL; - } else if (IS_ERR(data->hsic_pad_regulator)) - return dev_err_probe(dev, PTR_ERR(data->hsic_pad_regulator), + } else if (IS_ERR(data->hsic_pad_regulator)) { + ret = dev_err_probe(dev, PTR_ERR(data->hsic_pad_regulator), "Get HSIC pad regulator error\n"); + goto err_put; + } if (data->hsic_pad_regulator) { ret = regulator_enable(data->hsic_pad_regulator); if (ret) { dev_err(dev, "Failed to enable HSIC pad regulator\n"); - return ret; + goto err_put; } } } @@ -402,13 +406,14 @@ static int ci_hdrc_imx_probe(struct platform_device *pdev) dev_err(dev, "pinctrl_hsic_idle lookup failed, err=%ld\n", PTR_ERR(pinctrl_hsic_idle)); - return PTR_ERR(pinctrl_hsic_idle); + ret = PTR_ERR(pinctrl_hsic_idle); + goto err_put; } ret = pinctrl_select_state(data->pinctrl, pinctrl_hsic_idle); if (ret) { dev_err(dev, "hsic_idle select failed, err=%d\n", ret); - return ret; + goto err_put; } data->pinctrl_hsic_active = pinctrl_lookup_state(data->pinctrl, @@ -417,7 +422,8 @@ static int ci_hdrc_imx_probe(struct platform_device *pdev) dev_err(dev, "pinctrl_hsic_active lookup failed, err=%ld\n", PTR_ERR(data->pinctrl_hsic_active)); - return PTR_ERR(data->pinctrl_hsic_active); + ret = PTR_ERR(data->pinctrl_hsic_active); + goto err_put; } } @@ -527,6 +533,8 @@ disable_hsic_regulator: if (pdata.flags & CI_HDRC_PMQOS) cpu_latency_qos_remove_request(&data->pm_qos_req); data->ci_pdev = NULL; +err_put: + put_device(data->usbmisc_data->dev); return ret; } @@ -551,6 +559,7 @@ static void ci_hdrc_imx_remove(struct platform_device *pdev) if (data->hsic_pad_regulator) regulator_disable(data->hsic_pad_regulator); } + put_device(data->usbmisc_data->dev); } static void ci_hdrc_imx_shutdown(struct platform_device *pdev) diff --git a/drivers/usb/class/usblp.c b/drivers/usb/class/usblp.c index 5a2e43331064..ff1a941fd2ed 100644 --- a/drivers/usb/class/usblp.c +++ b/drivers/usb/class/usblp.c @@ -1337,11 +1337,12 @@ static int usblp_set_protocol(struct usblp *usblp, int protocol) if (protocol < USBLP_FIRST_PROTOCOL || protocol > USBLP_LAST_PROTOCOL) return -EINVAL; + alts = usblp->protocol[protocol].alt_setting; + if (alts < 0) + return -EINVAL; + /* Don't unnecessarily set the interface if there's a single alt. */ if (usblp->intf->num_altsetting > 1) { - alts = usblp->protocol[protocol].alt_setting; - if (alts < 0) - return -EINVAL; r = usb_set_interface(usblp->dev, usblp->ifnum, alts); if (r < 0) { printk(KERN_ERR "usblp: can't set desired altsetting %d on interface %d\n", diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c index 4b93c0bd1d4b..21ac9b464696 100644 --- a/drivers/usb/core/hub.c +++ b/drivers/usb/core/hub.c @@ -2663,13 +2663,13 @@ int usb_new_device(struct usb_device *udev) err = sysfs_create_link(&udev->dev.kobj, &port_dev->dev.kobj, "port"); if (err) - goto fail; + goto out_del_dev; err = sysfs_create_link(&port_dev->dev.kobj, &udev->dev.kobj, "device"); if (err) { sysfs_remove_link(&udev->dev.kobj, "port"); - goto fail; + goto out_del_dev; } if (!test_and_set_bit(port1, hub->child_usage_bits)) @@ -2683,6 +2683,8 @@ int usb_new_device(struct usb_device *udev) pm_runtime_put_sync_autosuspend(&udev->dev); return err; +out_del_dev: + device_del(&udev->dev); fail: usb_set_device_state(udev, USB_STATE_NOTATTACHED); pm_runtime_disable(&udev->dev); diff --git a/drivers/usb/core/port.c b/drivers/usb/core/port.c index 45d7af00f8d1..e857e532b35a 100644 --- a/drivers/usb/core/port.c +++ b/drivers/usb/core/port.c @@ -453,10 +453,11 @@ static int usb_port_runtime_suspend(struct device *dev) static void usb_port_shutdown(struct device *dev) { struct usb_port *port_dev = to_usb_port(dev); + struct usb_device *udev = port_dev->child; - if (port_dev->child) { - usb_disable_usb2_hardware_lpm(port_dev->child); - usb_unlocked_disable_lpm(port_dev->child); + if (udev && !udev->port_is_suspended) { + usb_disable_usb2_hardware_lpm(udev); + usb_unlocked_disable_lpm(udev); } } diff --git a/drivers/usb/dwc3/core.h b/drivers/usb/dwc3/core.h index ee73789326bc..f11570c8ffd0 100644 --- a/drivers/usb/dwc3/core.h +++ b/drivers/usb/dwc3/core.h @@ -464,6 +464,7 @@ #define DWC3_DCTL_TRGTULST_SS_INACT (DWC3_DCTL_TRGTULST(6)) /* These apply for core versions 1.94a and later */ +#define DWC3_DCTL_NYET_THRES_MASK (0xf << 20) #define DWC3_DCTL_NYET_THRES(n) (((n) & 0xf) << 20) #define DWC3_DCTL_KEEP_CONNECT BIT(19) diff --git a/drivers/usb/dwc3/dwc3-am62.c b/drivers/usb/dwc3/dwc3-am62.c index 5e3d1741701f..7d43da5f2897 100644 --- a/drivers/usb/dwc3/dwc3-am62.c +++ b/drivers/usb/dwc3/dwc3-am62.c @@ -309,6 +309,7 @@ static void dwc3_ti_remove(struct platform_device *pdev) pm_runtime_put_sync(dev); pm_runtime_disable(dev); + pm_runtime_dont_use_autosuspend(dev); pm_runtime_set_suspended(dev); } diff --git a/drivers/usb/dwc3/gadget.c b/drivers/usb/dwc3/gadget.c index 83dc7304d701..31a654c6f15b 100644 --- a/drivers/usb/dwc3/gadget.c +++ b/drivers/usb/dwc3/gadget.c @@ -4195,8 +4195,10 @@ static void dwc3_gadget_conndone_interrupt(struct dwc3 *dwc) WARN_ONCE(DWC3_VER_IS_PRIOR(DWC3, 240A) && dwc->has_lpm_erratum, "LPM Erratum not available on dwc3 revisions < 2.40a\n"); - if (dwc->has_lpm_erratum && !DWC3_VER_IS_PRIOR(DWC3, 240A)) + if (dwc->has_lpm_erratum && !DWC3_VER_IS_PRIOR(DWC3, 240A)) { + reg &= ~DWC3_DCTL_NYET_THRES_MASK; reg |= DWC3_DCTL_NYET_THRES(dwc->lpm_nyet_threshold); + } dwc3_gadget_dctl_write_safe(dwc, reg); } else { diff --git a/drivers/usb/gadget/Kconfig b/drivers/usb/gadget/Kconfig index 566ff0b1282a..76521555e3c1 100644 --- a/drivers/usb/gadget/Kconfig +++ b/drivers/usb/gadget/Kconfig @@ -211,6 +211,8 @@ config USB_F_MIDI config USB_F_MIDI2 tristate + select SND_UMP + select SND_UMP_LEGACY_RAWMIDI config USB_F_HID tristate @@ -445,8 +447,6 @@ config USB_CONFIGFS_F_MIDI2 depends on USB_CONFIGFS depends on SND select USB_LIBCOMPOSITE - select SND_UMP - select SND_UMP_LEGACY_RAWMIDI select USB_F_MIDI2 help The MIDI 2.0 function driver provides the generic emulated diff --git a/drivers/usb/gadget/configfs.c b/drivers/usb/gadget/configfs.c index 6499a88d346c..fba2a56dae97 100644 --- a/drivers/usb/gadget/configfs.c +++ b/drivers/usb/gadget/configfs.c @@ -827,11 +827,15 @@ static ssize_t gadget_string_s_store(struct config_item *item, const char *page, { struct gadget_string *string = to_gadget_string(item); int size = min(sizeof(string->string), len + 1); + ssize_t cpy_len; if (len > USB_MAX_STRING_LEN) return -EINVAL; - return strscpy(string->string, page, size); + cpy_len = strscpy(string->string, page, size); + if (cpy_len > 0 && string->string[cpy_len - 1] == '\n') + string->string[cpy_len - 1] = 0; + return len; } CONFIGFS_ATTR(gadget_string_, s); diff --git a/drivers/usb/gadget/function/f_fs.c b/drivers/usb/gadget/function/f_fs.c index ad79eb0f729b..2dea9e42a0f8 100644 --- a/drivers/usb/gadget/function/f_fs.c +++ b/drivers/usb/gadget/function/f_fs.c @@ -2285,7 +2285,7 @@ static int functionfs_bind(struct ffs_data *ffs, struct usb_composite_dev *cdev) struct usb_gadget_strings **lang; int first_id; - if (WARN_ON(ffs->state != FFS_ACTIVE + if ((ffs->state != FFS_ACTIVE || test_and_set_bit(FFS_FL_BOUND, &ffs->flags))) return -EBADFD; diff --git a/drivers/usb/gadget/function/f_uac2.c b/drivers/usb/gadget/function/f_uac2.c index ce5b77f89190..9b324821c93b 100644 --- a/drivers/usb/gadget/function/f_uac2.c +++ b/drivers/usb/gadget/function/f_uac2.c @@ -1185,6 +1185,7 @@ afunc_bind(struct usb_configuration *cfg, struct usb_function *fn) uac2->as_in_alt = 0; } + std_ac_if_desc.bNumEndpoints = 0; if (FUOUT_EN(uac2_opts) || FUIN_EN(uac2_opts)) { uac2->int_ep = usb_ep_autoconfig(gadget, &fs_ep_int_desc); if (!uac2->int_ep) { diff --git a/drivers/usb/gadget/function/u_serial.c b/drivers/usb/gadget/function/u_serial.c index 53d9fc41acc5..bc143a86c2dd 100644 --- a/drivers/usb/gadget/function/u_serial.c +++ b/drivers/usb/gadget/function/u_serial.c @@ -1420,6 +1420,10 @@ void gserial_disconnect(struct gserial *gser) /* REVISIT as above: how best to track this? */ port->port_line_coding = gser->port_line_coding; + /* disable endpoints, aborting down any active I/O */ + usb_ep_disable(gser->out); + usb_ep_disable(gser->in); + port->port_usb = NULL; gser->ioport = NULL; if (port->port.count > 0) { @@ -1431,10 +1435,6 @@ void gserial_disconnect(struct gserial *gser) spin_unlock(&port->port_lock); spin_unlock_irqrestore(&serial_port_lock, flags); - /* disable endpoints, aborting down any active I/O */ - usb_ep_disable(gser->out); - usb_ep_disable(gser->in); - /* finally, free any unused/unusable I/O buffers */ spin_lock_irqsave(&port->port_lock, flags); if (port->port.count == 0) diff --git a/drivers/usb/host/xhci-plat.c b/drivers/usb/host/xhci-plat.c index e6c9006bd568..db109b570c5c 100644 --- a/drivers/usb/host/xhci-plat.c +++ b/drivers/usb/host/xhci-plat.c @@ -290,7 +290,8 @@ int xhci_plat_probe(struct platform_device *pdev, struct device *sysdev, const s hcd->tpl_support = of_usb_host_tpl_support(sysdev->of_node); - if (priv && (priv->quirks & XHCI_SKIP_PHY_INIT)) + if ((priv && (priv->quirks & XHCI_SKIP_PHY_INIT)) || + (xhci->quirks & XHCI_SKIP_PHY_INIT)) hcd->skip_phy_initialization = 1; if (priv && (priv->quirks & XHCI_SG_TRB_CACHE_SIZE_QUIRK)) diff --git a/drivers/usb/serial/cp210x.c b/drivers/usb/serial/cp210x.c index c24101f0a07a..9960ac2b10b7 100644 --- a/drivers/usb/serial/cp210x.c +++ b/drivers/usb/serial/cp210x.c @@ -223,6 +223,7 @@ static const struct usb_device_id id_table[] = { { USB_DEVICE(0x19CF, 0x3000) }, /* Parrot NMEA GPS Flight Recorder */ { USB_DEVICE(0x1ADB, 0x0001) }, /* Schweitzer Engineering C662 Cable */ { USB_DEVICE(0x1B1C, 0x1C00) }, /* Corsair USB Dongle */ + { USB_DEVICE(0x1B93, 0x1013) }, /* Phoenix Contact UPS Device */ { USB_DEVICE(0x1BA4, 0x0002) }, /* Silicon Labs 358x factory default */ { USB_DEVICE(0x1BE3, 0x07A6) }, /* WAGO 750-923 USB Service Cable */ { USB_DEVICE(0x1D6F, 0x0010) }, /* Seluxit ApS RF Dongle */ diff --git a/drivers/usb/serial/option.c b/drivers/usb/serial/option.c index 64317b390d22..1e2ae0c6c41c 100644 --- a/drivers/usb/serial/option.c +++ b/drivers/usb/serial/option.c @@ -621,7 +621,7 @@ static void option_instat_callback(struct urb *urb); /* MeiG Smart Technology products */ #define MEIGSMART_VENDOR_ID 0x2dee -/* MeiG Smart SRM825L based on Qualcomm 315 */ +/* MeiG Smart SRM815/SRM825L based on Qualcomm 315 */ #define MEIGSMART_PRODUCT_SRM825L 0x4d22 /* MeiG Smart SLM320 based on UNISOC UIS8910 */ #define MEIGSMART_PRODUCT_SLM320 0x4d41 @@ -2405,6 +2405,7 @@ static const struct usb_device_id option_ids[] = { { USB_DEVICE_AND_INTERFACE_INFO(UNISOC_VENDOR_ID, LUAT_PRODUCT_AIR720U, 0xff, 0, 0) }, { USB_DEVICE_AND_INTERFACE_INFO(MEIGSMART_VENDOR_ID, MEIGSMART_PRODUCT_SLM320, 0xff, 0, 0) }, { USB_DEVICE_AND_INTERFACE_INFO(MEIGSMART_VENDOR_ID, MEIGSMART_PRODUCT_SLM770A, 0xff, 0, 0) }, + { USB_DEVICE_AND_INTERFACE_INFO(MEIGSMART_VENDOR_ID, MEIGSMART_PRODUCT_SRM825L, 0xff, 0, 0) }, { USB_DEVICE_AND_INTERFACE_INFO(MEIGSMART_VENDOR_ID, MEIGSMART_PRODUCT_SRM825L, 0xff, 0xff, 0x30) }, { USB_DEVICE_AND_INTERFACE_INFO(MEIGSMART_VENDOR_ID, MEIGSMART_PRODUCT_SRM825L, 0xff, 0xff, 0x40) }, { USB_DEVICE_AND_INTERFACE_INFO(MEIGSMART_VENDOR_ID, MEIGSMART_PRODUCT_SRM825L, 0xff, 0xff, 0x60) }, @@ -2412,6 +2413,7 @@ static const struct usb_device_id option_ids[] = { .driver_info = NCTRL(1) }, { USB_DEVICE_INTERFACE_CLASS(0x1bbb, 0x0640, 0xff), /* TCL IK512 ECM */ .driver_info = NCTRL(3) }, + { USB_DEVICE_INTERFACE_CLASS(0x2949, 0x8700, 0xff) }, /* Neoway N723-EA */ { } /* Terminating entry */ }; MODULE_DEVICE_TABLE(usb, option_ids); diff --git a/drivers/usb/storage/unusual_devs.h b/drivers/usb/storage/unusual_devs.h index e5ad23d86833..54f0b1c83317 100644 --- a/drivers/usb/storage/unusual_devs.h +++ b/drivers/usb/storage/unusual_devs.h @@ -255,6 +255,13 @@ UNUSUAL_DEV( 0x0421, 0x06aa, 0x1110, 0x1110, USB_SC_DEVICE, USB_PR_DEVICE, NULL, US_FL_MAX_SECTORS_64 ), +/* Added by Lubomir Rintel <lkundrak@v3.sk>, a very fine chap */ +UNUSUAL_DEV( 0x0421, 0x06c2, 0x0000, 0x0406, + "Nokia", + "Nokia 208", + USB_SC_DEVICE, USB_PR_DEVICE, NULL, + US_FL_MAX_SECTORS_64 ), + #ifdef NO_SDDR09 UNUSUAL_DEV( 0x0436, 0x0005, 0x0100, 0x0100, "Microtech", diff --git a/drivers/usb/typec/tcpm/maxim_contaminant.c b/drivers/usb/typec/tcpm/maxim_contaminant.c index 22163d8f9eb0..0cdda06592fd 100644 --- a/drivers/usb/typec/tcpm/maxim_contaminant.c +++ b/drivers/usb/typec/tcpm/maxim_contaminant.c @@ -135,7 +135,7 @@ static int max_contaminant_read_resistance_kohm(struct max_tcpci_chip *chip, mv = max_contaminant_read_adc_mv(chip, channel, sleep_msec, raw, true); if (mv < 0) - return ret; + return mv; /* OVP enable */ ret = regmap_update_bits(regmap, TCPC_VENDOR_CC_CTRL2, CCOVPDIS, 0); @@ -157,7 +157,7 @@ static int max_contaminant_read_resistance_kohm(struct max_tcpci_chip *chip, mv = max_contaminant_read_adc_mv(chip, channel, sleep_msec, raw, true); if (mv < 0) - return ret; + return mv; /* Disable current source */ ret = regmap_update_bits(regmap, TCPC_VENDOR_CC_CTRL2, SBURPCTRL, 0); if (ret < 0) diff --git a/drivers/usb/typec/tcpm/tcpci.c b/drivers/usb/typec/tcpm/tcpci.c index ed32583829be..24a6a4354df8 100644 --- a/drivers/usb/typec/tcpm/tcpci.c +++ b/drivers/usb/typec/tcpm/tcpci.c @@ -700,7 +700,7 @@ static int tcpci_init(struct tcpc_dev *tcpc) tcpci->alert_mask = reg; - return tcpci_write16(tcpci, TCPC_ALERT_MASK, reg); + return 0; } irqreturn_t tcpci_irq(struct tcpci *tcpci) @@ -923,22 +923,27 @@ static int tcpci_probe(struct i2c_client *client) chip->data.set_orientation = err; + chip->tcpci = tcpci_register_port(&client->dev, &chip->data); + if (IS_ERR(chip->tcpci)) + return PTR_ERR(chip->tcpci); + err = devm_request_threaded_irq(&client->dev, client->irq, NULL, _tcpci_irq, IRQF_SHARED | IRQF_ONESHOT, dev_name(&client->dev), chip); if (err < 0) - return err; + goto unregister_port; - /* - * Disable irq while registering port. If irq is configured as an edge - * irq this allow to keep track and process the irq as soon as it is enabled. - */ - disable_irq(client->irq); - chip->tcpci = tcpci_register_port(&client->dev, &chip->data); - enable_irq(client->irq); + /* Enable chip interrupts at last */ + err = tcpci_write16(chip->tcpci, TCPC_ALERT_MASK, chip->tcpci->alert_mask); + if (err < 0) + goto unregister_port; - return PTR_ERR_OR_ZERO(chip->tcpci); + return 0; + +unregister_port: + tcpci_unregister_port(chip->tcpci); + return err; } static void tcpci_remove(struct i2c_client *client) diff --git a/drivers/usb/typec/ucsi/ucsi_ccg.c b/drivers/usb/typec/ucsi/ucsi_ccg.c index fcb8e61136cf..740171f24ef9 100644 --- a/drivers/usb/typec/ucsi/ucsi_ccg.c +++ b/drivers/usb/typec/ucsi/ucsi_ccg.c @@ -646,7 +646,7 @@ static int ucsi_ccg_sync_control(struct ucsi *ucsi, u64 command) UCSI_CMD_CONNECTOR_MASK; if (con_index == 0) { ret = -EINVAL; - goto unlock; + goto err_put; } con = &uc->ucsi->connector[con_index - 1]; ucsi_ccg_update_set_new_cam_cmd(uc, con, &command); @@ -654,8 +654,8 @@ static int ucsi_ccg_sync_control(struct ucsi *ucsi, u64 command) ret = ucsi_sync_control_common(ucsi, command); +err_put: pm_runtime_put_sync(uc->dev); -unlock: mutex_unlock(&uc->lock); return ret; diff --git a/drivers/usb/typec/ucsi/ucsi_glink.c b/drivers/usb/typec/ucsi/ucsi_glink.c index 90948cd6d297..fed39d458090 100644 --- a/drivers/usb/typec/ucsi/ucsi_glink.c +++ b/drivers/usb/typec/ucsi/ucsi_glink.c @@ -185,6 +185,11 @@ static void pmic_glink_ucsi_connector_status(struct ucsi_connector *con) struct pmic_glink_ucsi *ucsi = ucsi_get_drvdata(con->ucsi); int orientation; + if (!UCSI_CONSTAT(con, CONNECTED)) { + typec_set_orientation(con->port, TYPEC_ORIENTATION_NONE); + return; + } + if (con->num > PMIC_GLINK_MAX_PORTS || !ucsi->port_orientation[con->num - 1]) return; diff --git a/drivers/vfio/pci/vfio_pci_core.c b/drivers/vfio/pci/vfio_pci_core.c index 1ab58da9f38a..1a4ed5a357d3 100644 --- a/drivers/vfio/pci/vfio_pci_core.c +++ b/drivers/vfio/pci/vfio_pci_core.c @@ -1661,14 +1661,15 @@ static vm_fault_t vfio_pci_mmap_huge_fault(struct vm_fault *vmf, unsigned long pfn, pgoff = vmf->pgoff - vma->vm_pgoff; vm_fault_t ret = VM_FAULT_SIGBUS; - if (order && (vmf->address & ((PAGE_SIZE << order) - 1) || + pfn = vma_to_pfn(vma) + pgoff; + + if (order && (pfn & ((1 << order) - 1) || + vmf->address & ((PAGE_SIZE << order) - 1) || vmf->address + (PAGE_SIZE << order) > vma->vm_end)) { ret = VM_FAULT_FALLBACK; goto out; } - pfn = vma_to_pfn(vma); - down_read(&vdev->memory_lock); if (vdev->pm_runtime_engaged || !__vfio_pci_memory_enabled(vdev)) @@ -1676,18 +1677,18 @@ static vm_fault_t vfio_pci_mmap_huge_fault(struct vm_fault *vmf, switch (order) { case 0: - ret = vmf_insert_pfn(vma, vmf->address, pfn + pgoff); + ret = vmf_insert_pfn(vma, vmf->address, pfn); break; #ifdef CONFIG_ARCH_SUPPORTS_PMD_PFNMAP case PMD_ORDER: - ret = vmf_insert_pfn_pmd(vmf, __pfn_to_pfn_t(pfn + pgoff, - PFN_DEV), false); + ret = vmf_insert_pfn_pmd(vmf, + __pfn_to_pfn_t(pfn, PFN_DEV), false); break; #endif #ifdef CONFIG_ARCH_SUPPORTS_PUD_PFNMAP case PUD_ORDER: - ret = vmf_insert_pfn_pud(vmf, __pfn_to_pfn_t(pfn + pgoff, - PFN_DEV), false); + ret = vmf_insert_pfn_pud(vmf, + __pfn_to_pfn_t(pfn, PFN_DEV), false); break; #endif default: diff --git a/drivers/virt/coco/tdx-guest/tdx-guest.c b/drivers/virt/coco/tdx-guest/tdx-guest.c index d7db6c824e13..224e7dde9cde 100644 --- a/drivers/virt/coco/tdx-guest/tdx-guest.c +++ b/drivers/virt/coco/tdx-guest/tdx-guest.c @@ -124,10 +124,8 @@ static void *alloc_quote_buf(void) if (!addr) return NULL; - if (set_memory_decrypted((unsigned long)addr, count)) { - free_pages_exact(addr, len); + if (set_memory_decrypted((unsigned long)addr, count)) return NULL; - } return addr; } diff --git a/drivers/virt/vboxguest/Kconfig b/drivers/virt/vboxguest/Kconfig index cc329887bfae..11b153e7454e 100644 --- a/drivers/virt/vboxguest/Kconfig +++ b/drivers/virt/vboxguest/Kconfig @@ -1,7 +1,7 @@ # SPDX-License-Identifier: GPL-2.0-only config VBOXGUEST tristate "Virtual Box Guest integration support" - depends on X86 && PCI && INPUT + depends on (ARM64 || X86) && PCI && INPUT help This is a driver for the Virtual Box Guest PCI device used in Virtual Box virtual machines. Enabling this driver will add diff --git a/drivers/watchdog/stm32_iwdg.c b/drivers/watchdog/stm32_iwdg.c index d700e0d49bb9..8ad06b54c5ad 100644 --- a/drivers/watchdog/stm32_iwdg.c +++ b/drivers/watchdog/stm32_iwdg.c @@ -286,7 +286,7 @@ static int stm32_iwdg_irq_init(struct platform_device *pdev, if (!wdt->data->has_early_wakeup) return 0; - irq = platform_get_irq(pdev, 0); + irq = platform_get_irq_optional(pdev, 0); if (irq <= 0) return 0; diff --git a/fs/9p/vfs_addr.c b/fs/9p/vfs_addr.c index 819c75233235..32619d146cbc 100644 --- a/fs/9p/vfs_addr.c +++ b/fs/9p/vfs_addr.c @@ -57,6 +57,8 @@ static void v9fs_issue_write(struct netfs_io_subrequest *subreq) int err, len; len = p9_client_write(fid, subreq->start, &subreq->io_iter, &err); + if (len > 0) + __set_bit(NETFS_SREQ_MADE_PROGRESS, &subreq->flags); netfs_write_subrequest_terminated(subreq, len ?: err, false); } @@ -79,11 +81,13 @@ static void v9fs_issue_read(struct netfs_io_subrequest *subreq) __set_bit(NETFS_SREQ_CLEAR_TAIL, &subreq->flags); if (pos + total >= i_size_read(rreq->inode)) __set_bit(NETFS_SREQ_HIT_EOF, &subreq->flags); - - if (!err) + if (!err && total) { subreq->transferred += total; + __set_bit(NETFS_SREQ_MADE_PROGRESS, &subreq->flags); + } - netfs_read_subreq_terminated(subreq, err, false); + subreq->error = err; + netfs_read_subreq_terminated(subreq); } /** diff --git a/fs/afs/Makefile b/fs/afs/Makefile index dcdc0f1bb76f..5efd7e13b304 100644 --- a/fs/afs/Makefile +++ b/fs/afs/Makefile @@ -11,6 +11,7 @@ kafs-y := \ cmservice.o \ dir.o \ dir_edit.o \ + dir_search.o \ dir_silly.o \ dynroot.o \ file.o \ diff --git a/fs/afs/addr_prefs.c b/fs/afs/addr_prefs.c index a189ff8a5034..c0384201b8fe 100644 --- a/fs/afs/addr_prefs.c +++ b/fs/afs/addr_prefs.c @@ -413,8 +413,10 @@ int afs_proc_addr_prefs_write(struct file *file, char *buf, size_t size) do { argc = afs_split_string(&buf, argv, ARRAY_SIZE(argv)); - if (argc < 0) - return argc; + if (argc < 0) { + ret = argc; + goto done; + } if (argc < 2) goto inval; diff --git a/fs/afs/afs.h b/fs/afs/afs.h index b488072aee87..ec3db00bd081 100644 --- a/fs/afs/afs.h +++ b/fs/afs/afs.h @@ -10,7 +10,7 @@ #include <linux/in.h> -#define AFS_MAXCELLNAME 256 /* Maximum length of a cell name */ +#define AFS_MAXCELLNAME 253 /* Maximum length of a cell name (DNS limited) */ #define AFS_MAXVOLNAME 64 /* Maximum length of a volume name */ #define AFS_MAXNSERVERS 8 /* Maximum servers in a basic volume record */ #define AFS_NMAXNSERVERS 13 /* Maximum servers in a N/U-class volume record */ diff --git a/fs/afs/afs_vl.h b/fs/afs/afs_vl.h index a06296c8827d..b835e25a2c02 100644 --- a/fs/afs/afs_vl.h +++ b/fs/afs/afs_vl.h @@ -13,6 +13,7 @@ #define AFS_VL_PORT 7003 /* volume location service port */ #define VL_SERVICE 52 /* RxRPC service ID for the Volume Location service */ #define YFS_VL_SERVICE 2503 /* Service ID for AuriStor upgraded VL service */ +#define YFS_VL_MAXCELLNAME 256 /* Maximum length of a cell name in YFS protocol */ enum AFSVL_Operations { VLGETENTRYBYID = 503, /* AFS Get VLDB entry by ID */ diff --git a/fs/afs/callback.c b/fs/afs/callback.c index 99b2c8172021..69e1dd55b160 100644 --- a/fs/afs/callback.c +++ b/fs/afs/callback.c @@ -41,7 +41,7 @@ static void afs_volume_init_callback(struct afs_volume *volume) list_for_each_entry(vnode, &volume->open_mmaps, cb_mmap_link) { if (vnode->cb_v_check != atomic_read(&volume->cb_v_break)) { - atomic64_set(&vnode->cb_expires_at, AFS_NO_CB_PROMISE); + afs_clear_cb_promise(vnode, afs_cb_promise_clear_vol_init_cb); queue_work(system_unbound_wq, &vnode->cb_work); } } @@ -79,7 +79,7 @@ void __afs_break_callback(struct afs_vnode *vnode, enum afs_cb_break_reason reas _enter(""); clear_bit(AFS_VNODE_NEW_CONTENT, &vnode->flags); - if (atomic64_xchg(&vnode->cb_expires_at, AFS_NO_CB_PROMISE) != AFS_NO_CB_PROMISE) { + if (afs_clear_cb_promise(vnode, afs_cb_promise_clear_cb_break)) { vnode->cb_break++; vnode->cb_v_check = atomic_read(&vnode->volume->cb_v_break); afs_clear_permits(vnode); diff --git a/fs/afs/cell.c b/fs/afs/cell.c index caa09875f520..cee42646736c 100644 --- a/fs/afs/cell.c +++ b/fs/afs/cell.c @@ -146,18 +146,20 @@ static struct afs_cell *afs_alloc_cell(struct afs_net *net, return ERR_PTR(-ENOMEM); } - cell->name = kmalloc(namelen + 1, GFP_KERNEL); + cell->name = kmalloc(1 + namelen + 1, GFP_KERNEL); if (!cell->name) { kfree(cell); return ERR_PTR(-ENOMEM); } - cell->net = net; + cell->name[0] = '.'; + cell->name++; cell->name_len = namelen; for (i = 0; i < namelen; i++) cell->name[i] = tolower(name[i]); cell->name[i] = 0; + cell->net = net; refcount_set(&cell->ref, 1); atomic_set(&cell->active, 0); INIT_WORK(&cell->manager, afs_manage_cell_work); @@ -211,7 +213,7 @@ parse_failed: if (ret == -EINVAL) printk(KERN_ERR "kAFS: bad VL server IP address\n"); error: - kfree(cell->name); + kfree(cell->name - 1); kfree(cell); _leave(" = %d", ret); return ERR_PTR(ret); @@ -365,6 +367,14 @@ int afs_cell_init(struct afs_net *net, const char *rootcell) len = cp - rootcell; } + if (len == 0 || !rootcell[0] || rootcell[0] == '.' || rootcell[len - 1] == '.') + return -EINVAL; + if (memchr(rootcell, '/', len)) + return -EINVAL; + cp = strstr(rootcell, ".."); + if (cp && cp < rootcell + len) + return -EINVAL; + /* allocate a cell record for the root cell */ new_root = afs_lookup_cell(net, rootcell, len, vllist, false); if (IS_ERR(new_root)) { @@ -502,7 +512,7 @@ static void afs_cell_destroy(struct rcu_head *rcu) afs_put_vlserverlist(net, rcu_access_pointer(cell->vl_servers)); afs_unuse_cell(net, cell->alias_of, afs_cell_trace_unuse_alias); key_put(cell->anonymous_key); - kfree(cell->name); + kfree(cell->name - 1); kfree(cell); afs_dec_cells_outstanding(net); @@ -710,7 +720,8 @@ static void afs_deactivate_cell(struct afs_net *net, struct afs_cell *cell) afs_proc_cell_remove(cell); mutex_lock(&net->proc_cells_lock); - hlist_del_rcu(&cell->proc_link); + if (!hlist_unhashed(&cell->proc_link)) + hlist_del_rcu(&cell->proc_link); afs_dynroot_rmdir(net, cell); mutex_unlock(&net->proc_cells_lock); diff --git a/fs/afs/dir.c b/fs/afs/dir.c index ada363af5aab..a843c36fc471 100644 --- a/fs/afs/dir.c +++ b/fs/afs/dir.c @@ -13,6 +13,7 @@ #include <linux/ctype.h> #include <linux/sched.h> #include <linux/iversion.h> +#include <linux/iov_iter.h> #include <linux/task_io_accounting_ops.h> #include "internal.h" #include "afs_fs.h" @@ -42,15 +43,6 @@ static int afs_symlink(struct mnt_idmap *idmap, struct inode *dir, static int afs_rename(struct mnt_idmap *idmap, struct inode *old_dir, struct dentry *old_dentry, struct inode *new_dir, struct dentry *new_dentry, unsigned int flags); -static bool afs_dir_release_folio(struct folio *folio, gfp_t gfp_flags); -static void afs_dir_invalidate_folio(struct folio *folio, size_t offset, - size_t length); - -static bool afs_dir_dirty_folio(struct address_space *mapping, - struct folio *folio) -{ - BUG(); /* This should never happen. */ -} const struct file_operations afs_dir_file_operations = { .open = afs_dir_open, @@ -75,10 +67,7 @@ const struct inode_operations afs_dir_inode_operations = { }; const struct address_space_operations afs_dir_aops = { - .dirty_folio = afs_dir_dirty_folio, - .release_folio = afs_dir_release_folio, - .invalidate_folio = afs_dir_invalidate_folio, - .migrate_folio = filemap_migrate_folio, + .writepages = afs_single_writepages, }; const struct dentry_operations afs_fs_dentry_operations = { @@ -99,152 +88,124 @@ struct afs_lookup_one_cookie { struct afs_lookup_cookie { struct dir_context ctx; struct qstr name; - bool found; - bool one_only; unsigned short nr_fids; struct afs_fid fids[50]; }; +static void afs_dir_unuse_cookie(struct afs_vnode *dvnode, int ret) +{ + if (ret == 0) { + struct afs_vnode_cache_aux aux; + loff_t i_size = i_size_read(&dvnode->netfs.inode); + + afs_set_cache_aux(dvnode, &aux); + fscache_unuse_cookie(afs_vnode_cache(dvnode), &aux, &i_size); + } else { + fscache_unuse_cookie(afs_vnode_cache(dvnode), NULL, NULL); + } +} + /* - * Drop the refs that we're holding on the folios we were reading into. We've - * got refs on the first nr_pages pages. + * Iterate through a kmapped directory segment, dumping a summary of + * the contents. */ -static void afs_dir_read_cleanup(struct afs_read *req) +static size_t afs_dir_dump_step(void *iter_base, size_t progress, size_t len, + void *priv, void *priv2) { - struct address_space *mapping = req->vnode->netfs.inode.i_mapping; - struct folio *folio; - pgoff_t last = req->nr_pages - 1; + do { + union afs_xdr_dir_block *block = iter_base; - XA_STATE(xas, &mapping->i_pages, 0); + pr_warn("[%05zx] %32phN\n", progress, block); + iter_base += AFS_DIR_BLOCK_SIZE; + progress += AFS_DIR_BLOCK_SIZE; + len -= AFS_DIR_BLOCK_SIZE; + } while (len > 0); - if (unlikely(!req->nr_pages)) - return; + return len; +} - rcu_read_lock(); - xas_for_each(&xas, folio, last) { - if (xas_retry(&xas, folio)) - continue; - BUG_ON(xa_is_value(folio)); - ASSERTCMP(folio->mapping, ==, mapping); +/* + * Dump the contents of a directory. + */ +static void afs_dir_dump(struct afs_vnode *dvnode) +{ + struct iov_iter iter; + unsigned long long i_size = i_size_read(&dvnode->netfs.inode); - folio_put(folio); - } + pr_warn("DIR %llx:%llx is=%llx\n", + dvnode->fid.vid, dvnode->fid.vnode, i_size); - rcu_read_unlock(); + iov_iter_folio_queue(&iter, ITER_SOURCE, dvnode->directory, 0, 0, i_size); + iterate_folioq(&iter, iov_iter_count(&iter), NULL, NULL, + afs_dir_dump_step); } /* * check that a directory folio is valid */ -static bool afs_dir_check_folio(struct afs_vnode *dvnode, struct folio *folio, - loff_t i_size) +static bool afs_dir_check_block(struct afs_vnode *dvnode, size_t progress, + union afs_xdr_dir_block *block) { - union afs_xdr_dir_block *block; - size_t offset, size; - loff_t pos; + if (block->hdr.magic != AFS_DIR_MAGIC) { + pr_warn("%s(%lx): [%zx] bad magic %04x\n", + __func__, dvnode->netfs.inode.i_ino, + progress, ntohs(block->hdr.magic)); + trace_afs_dir_check_failed(dvnode, progress); + trace_afs_file_error(dvnode, -EIO, afs_file_error_dir_bad_magic); + return false; + } - /* Determine how many magic numbers there should be in this folio, but - * we must take care because the directory may change size under us. + /* Make sure each block is NUL terminated so we can reasonably + * use string functions on it. The filenames in the folio + * *should* be NUL-terminated anyway. */ - pos = folio_pos(folio); - if (i_size <= pos) - goto checked; - - size = min_t(loff_t, folio_size(folio), i_size - pos); - for (offset = 0; offset < size; offset += sizeof(*block)) { - block = kmap_local_folio(folio, offset); - if (block->hdr.magic != AFS_DIR_MAGIC) { - printk("kAFS: %s(%lx): [%llx] bad magic %zx/%zx is %04hx\n", - __func__, dvnode->netfs.inode.i_ino, - pos, offset, size, ntohs(block->hdr.magic)); - trace_afs_dir_check_failed(dvnode, pos + offset, i_size); - kunmap_local(block); - trace_afs_file_error(dvnode, -EIO, afs_file_error_dir_bad_magic); - goto error; - } - - /* Make sure each block is NUL terminated so we can reasonably - * use string functions on it. The filenames in the folio - * *should* be NUL-terminated anyway. - */ - ((u8 *)block)[AFS_DIR_BLOCK_SIZE - 1] = 0; - - kunmap_local(block); - } -checked: + ((u8 *)block)[AFS_DIR_BLOCK_SIZE - 1] = 0; afs_stat_v(dvnode, n_read_dir); return true; - -error: - return false; } /* - * Dump the contents of a directory. + * Iterate through a kmapped directory segment, checking the content. */ -static void afs_dir_dump(struct afs_vnode *dvnode, struct afs_read *req) +static size_t afs_dir_check_step(void *iter_base, size_t progress, size_t len, + void *priv, void *priv2) { - union afs_xdr_dir_block *block; - struct address_space *mapping = dvnode->netfs.inode.i_mapping; - struct folio *folio; - pgoff_t last = req->nr_pages - 1; - size_t offset, size; - - XA_STATE(xas, &mapping->i_pages, 0); - - pr_warn("DIR %llx:%llx f=%llx l=%llx al=%llx\n", - dvnode->fid.vid, dvnode->fid.vnode, - req->file_size, req->len, req->actual_len); - pr_warn("DIR %llx %x %zx %zx\n", - req->pos, req->nr_pages, - req->iter->iov_offset, iov_iter_count(req->iter)); - - xas_for_each(&xas, folio, last) { - if (xas_retry(&xas, folio)) - continue; + struct afs_vnode *dvnode = priv; - BUG_ON(folio->mapping != mapping); + if (WARN_ON_ONCE(progress % AFS_DIR_BLOCK_SIZE || + len % AFS_DIR_BLOCK_SIZE)) + return len; - size = min_t(loff_t, folio_size(folio), req->actual_len - folio_pos(folio)); - for (offset = 0; offset < size; offset += sizeof(*block)) { - block = kmap_local_folio(folio, offset); - pr_warn("[%02lx] %32phN\n", folio->index + offset, block); - kunmap_local(block); - } - } + do { + if (!afs_dir_check_block(dvnode, progress, iter_base)) + break; + iter_base += AFS_DIR_BLOCK_SIZE; + len -= AFS_DIR_BLOCK_SIZE; + } while (len > 0); + + return len; } /* - * Check all the blocks in a directory. All the folios are held pinned. + * Check all the blocks in a directory. */ -static int afs_dir_check(struct afs_vnode *dvnode, struct afs_read *req) +static int afs_dir_check(struct afs_vnode *dvnode) { - struct address_space *mapping = dvnode->netfs.inode.i_mapping; - struct folio *folio; - pgoff_t last = req->nr_pages - 1; - int ret = 0; + struct iov_iter iter; + unsigned long long i_size = i_size_read(&dvnode->netfs.inode); + size_t checked = 0; - XA_STATE(xas, &mapping->i_pages, 0); - - if (unlikely(!req->nr_pages)) + if (unlikely(!i_size)) return 0; - rcu_read_lock(); - xas_for_each(&xas, folio, last) { - if (xas_retry(&xas, folio)) - continue; - - BUG_ON(folio->mapping != mapping); - - if (!afs_dir_check_folio(dvnode, folio, req->actual_len)) { - afs_dir_dump(dvnode, req); - ret = -EIO; - break; - } + iov_iter_folio_queue(&iter, ITER_SOURCE, dvnode->directory, 0, 0, i_size); + checked = iterate_folioq(&iter, iov_iter_count(&iter), dvnode, NULL, + afs_dir_check_step); + if (checked != i_size) { + afs_dir_dump(dvnode); + return -EIO; } - - rcu_read_unlock(); - return ret; + return 0; } /* @@ -264,134 +225,140 @@ static int afs_dir_open(struct inode *inode, struct file *file) } /* - * Read the directory into the pagecache in one go, scrubbing the previous - * contents. The list of folios is returned, pinning them so that they don't - * get reclaimed during the iteration. + * Read a file in a single download. */ -static struct afs_read *afs_read_dir(struct afs_vnode *dvnode, struct key *key) - __acquires(&dvnode->validate_lock) +static ssize_t afs_do_read_single(struct afs_vnode *dvnode, struct file *file) { - struct address_space *mapping = dvnode->netfs.inode.i_mapping; - struct afs_read *req; + struct iov_iter iter; + ssize_t ret; loff_t i_size; - int nr_pages, i; - int ret; - loff_t remote_size = 0; - - _enter(""); + bool is_dir = (S_ISDIR(dvnode->netfs.inode.i_mode) && + !test_bit(AFS_VNODE_MOUNTPOINT, &dvnode->flags)); - req = kzalloc(sizeof(*req), GFP_KERNEL); - if (!req) - return ERR_PTR(-ENOMEM); - - refcount_set(&req->usage, 1); - req->vnode = dvnode; - req->key = key_get(key); - req->cleanup = afs_dir_read_cleanup; - -expand: i_size = i_size_read(&dvnode->netfs.inode); - if (i_size < remote_size) - i_size = remote_size; - if (i_size < 2048) { - ret = afs_bad(dvnode, afs_file_error_dir_small); - goto error; - } - if (i_size > 2048 * 1024) { - trace_afs_file_error(dvnode, -EFBIG, afs_file_error_dir_big); - ret = -EFBIG; - goto error; + if (is_dir) { + if (i_size < AFS_DIR_BLOCK_SIZE) + return afs_bad(dvnode, afs_file_error_dir_small); + if (i_size > AFS_DIR_BLOCK_SIZE * 1024) { + trace_afs_file_error(dvnode, -EFBIG, afs_file_error_dir_big); + return -EFBIG; + } + } else { + if (i_size > AFSPATHMAX) { + trace_afs_file_error(dvnode, -EFBIG, afs_file_error_dir_big); + return -EFBIG; + } } - _enter("%llu", i_size); + /* Expand the storage. TODO: Shrink the storage too. */ + if (dvnode->directory_size < i_size) { + size_t cur_size = dvnode->directory_size; - nr_pages = (i_size + PAGE_SIZE - 1) / PAGE_SIZE; + ret = netfs_alloc_folioq_buffer(NULL, + &dvnode->directory, &cur_size, i_size, + mapping_gfp_mask(dvnode->netfs.inode.i_mapping)); + dvnode->directory_size = cur_size; + if (ret < 0) + return ret; + } - req->actual_len = i_size; /* May change */ - req->len = nr_pages * PAGE_SIZE; /* We can ask for more than there is */ - req->data_version = dvnode->status.data_version; /* May change */ - iov_iter_xarray(&req->def_iter, ITER_DEST, &dvnode->netfs.inode.i_mapping->i_pages, - 0, i_size); - req->iter = &req->def_iter; + iov_iter_folio_queue(&iter, ITER_DEST, dvnode->directory, 0, 0, dvnode->directory_size); - /* Fill in any gaps that we might find where the memory reclaimer has - * been at work and pin all the folios. If there are any gaps, we will - * need to reread the entire directory contents. + /* AFS requires us to perform the read of a directory synchronously as + * a single unit to avoid issues with the directory contents being + * changed between reads. */ - i = req->nr_pages; - while (i < nr_pages) { - struct folio *folio; - - folio = filemap_get_folio(mapping, i); - if (IS_ERR(folio)) { - if (test_and_clear_bit(AFS_VNODE_DIR_VALID, &dvnode->flags)) - afs_stat_v(dvnode, n_inval); - folio = __filemap_get_folio(mapping, - i, FGP_LOCK | FGP_CREAT, - mapping->gfp_mask); - if (IS_ERR(folio)) { - ret = PTR_ERR(folio); - goto error; - } - folio_attach_private(folio, (void *)1); - folio_unlock(folio); + ret = netfs_read_single(&dvnode->netfs.inode, file, &iter); + if (ret >= 0) { + i_size = i_size_read(&dvnode->netfs.inode); + if (i_size > ret) { + /* The content has grown, so we need to expand the + * buffer. + */ + ret = -ESTALE; + } else if (is_dir) { + int ret2 = afs_dir_check(dvnode); + + if (ret2 < 0) + ret = ret2; + } else if (i_size < folioq_folio_size(dvnode->directory, 0)) { + /* NUL-terminate a symlink. */ + char *symlink = kmap_local_folio(folioq_folio(dvnode->directory, 0), 0); + + symlink[i_size] = 0; + kunmap_local(symlink); } - - req->nr_pages += folio_nr_pages(folio); - i += folio_nr_pages(folio); } - /* If we're going to reload, we need to lock all the pages to prevent - * races. - */ + return ret; +} + +ssize_t afs_read_single(struct afs_vnode *dvnode, struct file *file) +{ + ssize_t ret; + + fscache_use_cookie(afs_vnode_cache(dvnode), false); + ret = afs_do_read_single(dvnode, file); + fscache_unuse_cookie(afs_vnode_cache(dvnode), NULL, NULL); + return ret; +} + +/* + * Read the directory into a folio_queue buffer in one go, scrubbing the + * previous contents. We return -ESTALE if the caller needs to call us again. + */ +ssize_t afs_read_dir(struct afs_vnode *dvnode, struct file *file) + __acquires(&dvnode->validate_lock) +{ + ssize_t ret; + loff_t i_size; + + i_size = i_size_read(&dvnode->netfs.inode); + ret = -ERESTARTSYS; if (down_read_killable(&dvnode->validate_lock) < 0) goto error; - if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags)) - goto success; + /* We only need to reread the data if it became invalid - or if we + * haven't read it yet. + */ + if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags) && + test_bit(AFS_VNODE_DIR_READ, &dvnode->flags)) { + ret = i_size; + goto valid; + } up_read(&dvnode->validate_lock); if (down_write_killable(&dvnode->validate_lock) < 0) goto error; - if (!test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags)) { - trace_afs_reload_dir(dvnode); - ret = afs_fetch_data(dvnode, req); - if (ret < 0) - goto error_unlock; - - task_io_account_read(PAGE_SIZE * req->nr_pages); - - if (req->len < req->file_size) { - /* The content has grown, so we need to expand the - * buffer. - */ - up_write(&dvnode->validate_lock); - remote_size = req->file_size; - goto expand; - } + if (!test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags)) + afs_invalidate_cache(dvnode, 0); - /* Validate the data we just read. */ - ret = afs_dir_check(dvnode, req); + if (!test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags) || + !test_bit(AFS_VNODE_DIR_READ, &dvnode->flags)) { + trace_afs_reload_dir(dvnode); + ret = afs_read_single(dvnode, file); if (ret < 0) goto error_unlock; // TODO: Trim excess pages set_bit(AFS_VNODE_DIR_VALID, &dvnode->flags); + set_bit(AFS_VNODE_DIR_READ, &dvnode->flags); + } else { + ret = i_size; } downgrade_write(&dvnode->validate_lock); -success: - return req; +valid: + return ret; error_unlock: up_write(&dvnode->validate_lock); error: - afs_put_read(req); - _leave(" = %d", ret); - return ERR_PTR(ret); + _leave(" = %zd", ret); + return ret; } /* @@ -399,79 +366,69 @@ error: */ static int afs_dir_iterate_block(struct afs_vnode *dvnode, struct dir_context *ctx, - union afs_xdr_dir_block *block, - unsigned blkoff) + union afs_xdr_dir_block *block) { union afs_xdr_dirent *dire; - unsigned offset, next, curr, nr_slots; + unsigned int blknum, base, hdr, pos, next, nr_slots; size_t nlen; int tmp; - _enter("%llx,%x", ctx->pos, blkoff); + blknum = ctx->pos / AFS_DIR_BLOCK_SIZE; + base = blknum * AFS_DIR_SLOTS_PER_BLOCK; + hdr = (blknum == 0 ? AFS_DIR_RESV_BLOCKS0 : AFS_DIR_RESV_BLOCKS); + pos = DIV_ROUND_UP(ctx->pos, AFS_DIR_DIRENT_SIZE) - base; - curr = (ctx->pos - blkoff) / sizeof(union afs_xdr_dirent); + _enter("%llx,%x", ctx->pos, blknum); /* walk through the block, an entry at a time */ - for (offset = (blkoff == 0 ? AFS_DIR_RESV_BLOCKS0 : AFS_DIR_RESV_BLOCKS); - offset < AFS_DIR_SLOTS_PER_BLOCK; - offset = next - ) { + for (unsigned int slot = hdr; slot < AFS_DIR_SLOTS_PER_BLOCK; slot = next) { /* skip entries marked unused in the bitmap */ - if (!(block->hdr.bitmap[offset / 8] & - (1 << (offset % 8)))) { - _debug("ENT[%zu.%u]: unused", - blkoff / sizeof(union afs_xdr_dir_block), offset); - next = offset + 1; - if (offset >= curr) - ctx->pos = blkoff + - next * sizeof(union afs_xdr_dirent); + if (!(block->hdr.bitmap[slot / 8] & + (1 << (slot % 8)))) { + _debug("ENT[%x]: Unused", base + slot); + next = slot + 1; + if (next >= pos) + ctx->pos = (base + next) * sizeof(union afs_xdr_dirent); continue; } /* got a valid entry */ - dire = &block->dirents[offset]; + dire = &block->dirents[slot]; nlen = strnlen(dire->u.name, - sizeof(*block) - - offset * sizeof(union afs_xdr_dirent)); + (unsigned long)(block + 1) - (unsigned long)dire->u.name - 1); if (nlen > AFSNAMEMAX - 1) { - _debug("ENT[%zu]: name too long (len %u/%zu)", - blkoff / sizeof(union afs_xdr_dir_block), - offset, nlen); + _debug("ENT[%x]: Name too long (len %zx)", + base + slot, nlen); return afs_bad(dvnode, afs_file_error_dir_name_too_long); } - _debug("ENT[%zu.%u]: %s %zu \"%s\"", - blkoff / sizeof(union afs_xdr_dir_block), offset, - (offset < curr ? "skip" : "fill"), + _debug("ENT[%x]: %s %zx \"%s\"", + base + slot, (slot < pos ? "skip" : "fill"), nlen, dire->u.name); nr_slots = afs_dir_calc_slots(nlen); - next = offset + nr_slots; + next = slot + nr_slots; if (next > AFS_DIR_SLOTS_PER_BLOCK) { - _debug("ENT[%zu.%u]:" - " %u extends beyond end dir block" - " (len %zu)", - blkoff / sizeof(union afs_xdr_dir_block), - offset, next, nlen); + _debug("ENT[%x]: extends beyond end dir block (len %zx)", + base + slot, nlen); return afs_bad(dvnode, afs_file_error_dir_over_end); } /* Check that the name-extension dirents are all allocated */ for (tmp = 1; tmp < nr_slots; tmp++) { - unsigned int ix = offset + tmp; - if (!(block->hdr.bitmap[ix / 8] & (1 << (ix % 8)))) { - _debug("ENT[%zu.u]:" - " %u unmarked extension (%u/%u)", - blkoff / sizeof(union afs_xdr_dir_block), - offset, tmp, nr_slots); + unsigned int xslot = slot + tmp; + + if (!(block->hdr.bitmap[xslot / 8] & (1 << (xslot % 8)))) { + _debug("ENT[%x]: Unmarked extension (%x/%x)", + base + slot, tmp, nr_slots); return afs_bad(dvnode, afs_file_error_dir_unmarked_ext); } } /* skip if starts before the current position */ - if (offset < curr) { - if (next > curr) - ctx->pos = blkoff + next * sizeof(union afs_xdr_dirent); + if (slot < pos) { + if (next > pos) + ctx->pos = (base + next) * sizeof(union afs_xdr_dirent); continue; } @@ -485,75 +442,110 @@ static int afs_dir_iterate_block(struct afs_vnode *dvnode, return 0; } - ctx->pos = blkoff + next * sizeof(union afs_xdr_dirent); + ctx->pos = (base + next) * sizeof(union afs_xdr_dirent); } _leave(" = 1 [more]"); return 1; } +struct afs_dir_iteration_ctx { + struct dir_context *dir_ctx; + int error; +}; + /* - * iterate through the data blob that lists the contents of an AFS directory + * Iterate through a kmapped directory segment. */ -static int afs_dir_iterate(struct inode *dir, struct dir_context *ctx, - struct key *key, afs_dataversion_t *_dir_version) +static size_t afs_dir_iterate_step(void *iter_base, size_t progress, size_t len, + void *priv, void *priv2) { - struct afs_vnode *dvnode = AFS_FS_I(dir); - union afs_xdr_dir_block *dblock; - struct afs_read *req; - struct folio *folio; - unsigned offset, size; + struct afs_dir_iteration_ctx *ctx = priv2; + struct afs_vnode *dvnode = priv; int ret; - _enter("{%lu},%u,,", dir->i_ino, (unsigned)ctx->pos); - - if (test_bit(AFS_VNODE_DELETED, &AFS_FS_I(dir)->flags)) { - _leave(" = -ESTALE"); - return -ESTALE; + if (WARN_ON_ONCE(progress % AFS_DIR_BLOCK_SIZE || + len % AFS_DIR_BLOCK_SIZE)) { + pr_err("Mis-iteration prog=%zx len=%zx\n", + progress % AFS_DIR_BLOCK_SIZE, + len % AFS_DIR_BLOCK_SIZE); + return len; } - req = afs_read_dir(dvnode, key); - if (IS_ERR(req)) - return PTR_ERR(req); - *_dir_version = req->data_version; + do { + ret = afs_dir_iterate_block(dvnode, ctx->dir_ctx, iter_base); + if (ret != 1) + break; - /* round the file position up to the next entry boundary */ - ctx->pos += sizeof(union afs_xdr_dirent) - 1; - ctx->pos &= ~(sizeof(union afs_xdr_dirent) - 1); + ctx->dir_ctx->pos = round_up(ctx->dir_ctx->pos, AFS_DIR_BLOCK_SIZE); + iter_base += AFS_DIR_BLOCK_SIZE; + len -= AFS_DIR_BLOCK_SIZE; + } while (len > 0); - /* walk through the blocks in sequence */ - ret = 0; - while (ctx->pos < req->actual_len) { - /* Fetch the appropriate folio from the directory and re-add it - * to the LRU. We have all the pages pinned with an extra ref. - */ - folio = __filemap_get_folio(dir->i_mapping, ctx->pos / PAGE_SIZE, - FGP_ACCESSED, 0); - if (IS_ERR(folio)) { - ret = afs_bad(dvnode, afs_file_error_dir_missing_page); - break; - } + return len; +} - offset = round_down(ctx->pos, sizeof(*dblock)) - folio_pos(folio); - size = min_t(loff_t, folio_size(folio), - req->actual_len - folio_pos(folio)); +/* + * Iterate through the directory folios. + */ +static int afs_dir_iterate_contents(struct inode *dir, struct dir_context *dir_ctx) +{ + struct afs_dir_iteration_ctx ctx = { .dir_ctx = dir_ctx }; + struct afs_vnode *dvnode = AFS_FS_I(dir); + struct iov_iter iter; + unsigned long long i_size = i_size_read(dir); - do { - dblock = kmap_local_folio(folio, offset); - ret = afs_dir_iterate_block(dvnode, ctx, dblock, - folio_pos(folio) + offset); - kunmap_local(dblock); - if (ret != 1) - goto out; + /* Round the file position up to the next entry boundary */ + dir_ctx->pos = round_up(dir_ctx->pos, sizeof(union afs_xdr_dirent)); - } while (offset += sizeof(*dblock), offset < size); + if (i_size <= 0 || dir_ctx->pos >= i_size) + return 0; - ret = 0; - } + iov_iter_folio_queue(&iter, ITER_SOURCE, dvnode->directory, 0, 0, i_size); + iov_iter_advance(&iter, round_down(dir_ctx->pos, AFS_DIR_BLOCK_SIZE)); + + iterate_folioq(&iter, iov_iter_count(&iter), dvnode, &ctx, + afs_dir_iterate_step); + + if (ctx.error == -ESTALE) + afs_invalidate_dir(dvnode, afs_dir_invalid_iter_stale); + return ctx.error; +} + +/* + * iterate through the data blob that lists the contents of an AFS directory + */ +static int afs_dir_iterate(struct inode *dir, struct dir_context *ctx, + struct file *file, afs_dataversion_t *_dir_version) +{ + struct afs_vnode *dvnode = AFS_FS_I(dir); + int retry_limit = 100; + int ret; + + _enter("{%lu},%llx,,", dir->i_ino, ctx->pos); + + do { + if (--retry_limit < 0) { + pr_warn("afs_read_dir(): Too many retries\n"); + ret = -ESTALE; + break; + } + ret = afs_read_dir(dvnode, file); + if (ret < 0) { + if (ret != -ESTALE) + break; + if (test_bit(AFS_VNODE_DELETED, &AFS_FS_I(dir)->flags)) { + ret = -ESTALE; + break; + } + continue; + } + *_dir_version = inode_peek_iversion_raw(dir); + + ret = afs_dir_iterate_contents(dir, ctx); + up_read(&dvnode->validate_lock); + } while (ret == -ESTALE); -out: - up_read(&dvnode->validate_lock); - afs_put_read(req); _leave(" = %d", ret); return ret; } @@ -565,8 +557,7 @@ static int afs_readdir(struct file *file, struct dir_context *ctx) { afs_dataversion_t dir_version; - return afs_dir_iterate(file_inode(file), ctx, afs_file_key(file), - &dir_version); + return afs_dir_iterate(file_inode(file), ctx, file, &dir_version); } /* @@ -607,7 +598,7 @@ static bool afs_lookup_one_filldir(struct dir_context *ctx, const char *name, * - just returns the FID the dentry name maps to if found */ static int afs_do_lookup_one(struct inode *dir, struct dentry *dentry, - struct afs_fid *fid, struct key *key, + struct afs_fid *fid, afs_dataversion_t *_dir_version) { struct afs_super_info *as = dir->i_sb->s_fs_info; @@ -621,7 +612,7 @@ static int afs_do_lookup_one(struct inode *dir, struct dentry *dentry, _enter("{%lu},%p{%pd},", dir->i_ino, dentry, dentry); /* search the directory */ - ret = afs_dir_iterate(dir, &cookie.ctx, key, _dir_version); + ret = afs_dir_iterate(dir, &cookie.ctx, NULL, _dir_version); if (ret < 0) { _leave(" = %d [iter]", ret); return ret; @@ -656,19 +647,10 @@ static bool afs_lookup_filldir(struct dir_context *ctx, const char *name, BUILD_BUG_ON(sizeof(union afs_xdr_dir_block) != 2048); BUILD_BUG_ON(sizeof(union afs_xdr_dirent) != 32); - if (cookie->found) { - if (cookie->nr_fids < 50) { - cookie->fids[cookie->nr_fids].vnode = ino; - cookie->fids[cookie->nr_fids].unique = dtype; - cookie->nr_fids++; - } - } else if (cookie->name.len == nlen && - memcmp(cookie->name.name, name, nlen) == 0) { - cookie->fids[1].vnode = ino; - cookie->fids[1].unique = dtype; - cookie->found = 1; - if (cookie->one_only) - return false; + if (cookie->nr_fids < 50) { + cookie->fids[cookie->nr_fids].vnode = ino; + cookie->fids[cookie->nr_fids].unique = dtype; + cookie->nr_fids++; } return cookie->nr_fids < 50; @@ -788,8 +770,7 @@ static bool afs_server_supports_ibulk(struct afs_vnode *dvnode) * files in one go and create inodes for them. The inode of the file we were * asked for is returned. */ -static struct inode *afs_do_lookup(struct inode *dir, struct dentry *dentry, - struct key *key) +static struct inode *afs_do_lookup(struct inode *dir, struct dentry *dentry) { struct afs_lookup_cookie *cookie; struct afs_vnode_param *vp; @@ -797,6 +778,7 @@ static struct inode *afs_do_lookup(struct inode *dir, struct dentry *dentry, struct afs_vnode *dvnode = AFS_FS_I(dir), *vnode; struct inode *inode = NULL, *ti; afs_dataversion_t data_version = READ_ONCE(dvnode->status.data_version); + bool supports_ibulk; long ret; int i; @@ -813,19 +795,19 @@ static struct inode *afs_do_lookup(struct inode *dir, struct dentry *dentry, cookie->nr_fids = 2; /* slot 1 is saved for the fid we actually want * and slot 0 for the directory */ - if (!afs_server_supports_ibulk(dvnode)) - cookie->one_only = true; - - /* search the directory */ - ret = afs_dir_iterate(dir, &cookie->ctx, key, &data_version); + /* Search the directory for the named entry using the hash table... */ + ret = afs_dir_search(dvnode, &dentry->d_name, &cookie->fids[1], &data_version); if (ret < 0) goto out; - dentry->d_fsdata = (void *)(unsigned long)data_version; + supports_ibulk = afs_server_supports_ibulk(dvnode); + if (supports_ibulk) { + /* ...then scan linearly from that point for entries to lookup-ahead. */ + cookie->ctx.pos = (ret + 1) * AFS_DIR_DIRENT_SIZE; + afs_dir_iterate(dir, &cookie->ctx, NULL, &data_version); + } - ret = -ENOENT; - if (!cookie->found) - goto out; + dentry->d_fsdata = (void *)(unsigned long)data_version; /* Check to see if we already have an inode for the primary fid. */ inode = ilookup5(dir->i_sb, cookie->fids[1].vnode, @@ -884,7 +866,7 @@ static struct inode *afs_do_lookup(struct inode *dir, struct dentry *dentry, * the whole operation. */ afs_op_set_error(op, -ENOTSUPP); - if (!cookie->one_only) { + if (supports_ibulk) { op->ops = &afs_inline_bulk_status_operation; afs_begin_vnode_operation(op); afs_wait_for_operation(op); @@ -926,8 +908,7 @@ out: /* * Look up an entry in a directory with @sys substitution. */ -static struct dentry *afs_lookup_atsys(struct inode *dir, struct dentry *dentry, - struct key *key) +static struct dentry *afs_lookup_atsys(struct inode *dir, struct dentry *dentry) { struct afs_sysnames *subs; struct afs_net *net = afs_i2net(dir); @@ -975,7 +956,6 @@ out_s: afs_put_sysnames(subs); kfree(buf); out_p: - key_put(key); return ret; } @@ -989,7 +969,6 @@ static struct dentry *afs_lookup(struct inode *dir, struct dentry *dentry, struct afs_fid fid = {}; struct inode *inode; struct dentry *d; - struct key *key; int ret; _enter("{%llx:%llu},%p{%pd},", @@ -1007,15 +986,9 @@ static struct dentry *afs_lookup(struct inode *dir, struct dentry *dentry, return ERR_PTR(-ESTALE); } - key = afs_request_key(dvnode->volume->cell); - if (IS_ERR(key)) { - _leave(" = %ld [key]", PTR_ERR(key)); - return ERR_CAST(key); - } - - ret = afs_validate(dvnode, key); + ret = afs_validate(dvnode, NULL); if (ret < 0) { - key_put(key); + afs_dir_unuse_cookie(dvnode, ret); _leave(" = %d [val]", ret); return ERR_PTR(ret); } @@ -1025,11 +998,10 @@ static struct dentry *afs_lookup(struct inode *dir, struct dentry *dentry, dentry->d_name.name[dentry->d_name.len - 3] == 's' && dentry->d_name.name[dentry->d_name.len - 2] == 'y' && dentry->d_name.name[dentry->d_name.len - 1] == 's') - return afs_lookup_atsys(dir, dentry, key); + return afs_lookup_atsys(dir, dentry); afs_stat_v(dvnode, n_lookup); - inode = afs_do_lookup(dir, dentry, key); - key_put(key); + inode = afs_do_lookup(dir, dentry); if (inode == ERR_PTR(-ENOENT)) inode = afs_try_auto_mntpt(dentry, dir); @@ -1155,7 +1127,7 @@ static int afs_d_revalidate(struct dentry *dentry, unsigned int flags) afs_stat_v(dir, n_reval); /* search the directory for this vnode */ - ret = afs_do_lookup_one(&dir->netfs.inode, dentry, &fid, key, &dir_version); + ret = afs_do_lookup_one(&dir->netfs.inode, dentry, &fid, &dir_version); switch (ret) { case 0: /* the filename maps to something */ @@ -1282,6 +1254,7 @@ void afs_check_for_remote_deletion(struct afs_operation *op) */ static void afs_vnode_new_inode(struct afs_operation *op) { + struct afs_vnode_param *dvp = &op->file[0]; struct afs_vnode_param *vp = &op->file[1]; struct afs_vnode *vnode; struct inode *inode; @@ -1301,6 +1274,10 @@ static void afs_vnode_new_inode(struct afs_operation *op) vnode = AFS_FS_I(inode); set_bit(AFS_VNODE_NEW_CONTENT, &vnode->flags); + if (S_ISDIR(inode->i_mode)) + afs_mkdir_init_dir(vnode, dvp->vnode); + else if (S_ISLNK(inode->i_mode)) + afs_init_new_symlink(vnode, op); if (!afs_op_error(op)) afs_cache_permit(vnode, op->key, vnode->cb_break, &vp->scb); d_instantiate(op->dentry, inode); @@ -1317,18 +1294,21 @@ static void afs_create_success(struct afs_operation *op) static void afs_create_edit_dir(struct afs_operation *op) { + struct netfs_cache_resources cres = {}; struct afs_vnode_param *dvp = &op->file[0]; struct afs_vnode_param *vp = &op->file[1]; struct afs_vnode *dvnode = dvp->vnode; _enter("op=%08x", op->debug_id); + fscache_begin_write_operation(&cres, afs_vnode_cache(dvnode)); down_write(&dvnode->validate_lock); if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags) && dvnode->status.data_version == dvp->dv_before + dvp->dv_delta) afs_edit_dir_add(dvnode, &op->dentry->d_name, &vp->fid, op->create.reason); up_write(&dvnode->validate_lock); + fscache_end_operation(&cres); } static void afs_create_put(struct afs_operation *op) @@ -1356,6 +1336,7 @@ static int afs_mkdir(struct mnt_idmap *idmap, struct inode *dir, { struct afs_operation *op; struct afs_vnode *dvnode = AFS_FS_I(dir); + int ret; _enter("{%llx:%llu},{%pd},%ho", dvnode->fid.vid, dvnode->fid.vnode, dentry, mode); @@ -1366,6 +1347,8 @@ static int afs_mkdir(struct mnt_idmap *idmap, struct inode *dir, return PTR_ERR(op); } + fscache_use_cookie(afs_vnode_cache(dvnode), true); + afs_op_set_vnode(op, 0, dvnode); op->file[0].dv_delta = 1; op->file[0].modification = true; @@ -1375,7 +1358,9 @@ static int afs_mkdir(struct mnt_idmap *idmap, struct inode *dir, op->create.reason = afs_edit_dir_for_mkdir; op->mtime = current_time(dir); op->ops = &afs_mkdir_operation; - return afs_do_sync_operation(op); + ret = afs_do_sync_operation(op); + afs_dir_unuse_cookie(dvnode, ret); + return ret; } /* @@ -1388,8 +1373,8 @@ static void afs_dir_remove_subdir(struct dentry *dentry) clear_nlink(&vnode->netfs.inode); set_bit(AFS_VNODE_DELETED, &vnode->flags); - atomic64_set(&vnode->cb_expires_at, AFS_NO_CB_PROMISE); - clear_bit(AFS_VNODE_DIR_VALID, &vnode->flags); + afs_clear_cb_promise(vnode, afs_cb_promise_clear_rmdir); + afs_invalidate_dir(vnode, afs_dir_invalid_subdir_removed); } } @@ -1403,18 +1388,21 @@ static void afs_rmdir_success(struct afs_operation *op) static void afs_rmdir_edit_dir(struct afs_operation *op) { + struct netfs_cache_resources cres = {}; struct afs_vnode_param *dvp = &op->file[0]; struct afs_vnode *dvnode = dvp->vnode; _enter("op=%08x", op->debug_id); afs_dir_remove_subdir(op->dentry); + fscache_begin_write_operation(&cres, afs_vnode_cache(dvnode)); down_write(&dvnode->validate_lock); if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags) && dvnode->status.data_version == dvp->dv_before + dvp->dv_delta) afs_edit_dir_remove(dvnode, &op->dentry->d_name, afs_edit_dir_for_rmdir); up_write(&dvnode->validate_lock); + fscache_end_operation(&cres); } static void afs_rmdir_put(struct afs_operation *op) @@ -1449,6 +1437,8 @@ static int afs_rmdir(struct inode *dir, struct dentry *dentry) if (IS_ERR(op)) return PTR_ERR(op); + fscache_use_cookie(afs_vnode_cache(dvnode), true); + afs_op_set_vnode(op, 0, dvnode); op->file[0].dv_delta = 1; op->file[0].modification = true; @@ -1472,10 +1462,18 @@ static int afs_rmdir(struct inode *dir, struct dentry *dentry) op->file[1].vnode = vnode; } - return afs_do_sync_operation(op); + ret = afs_do_sync_operation(op); + + /* Not all systems that can host afs servers have ENOTEMPTY. */ + if (ret == -EEXIST) + ret = -ENOTEMPTY; +out: + afs_dir_unuse_cookie(dvnode, ret); + return ret; error: - return afs_put_operation(op); + ret = afs_put_operation(op); + goto out; } /* @@ -1538,16 +1536,19 @@ static void afs_unlink_success(struct afs_operation *op) static void afs_unlink_edit_dir(struct afs_operation *op) { + struct netfs_cache_resources cres = {}; struct afs_vnode_param *dvp = &op->file[0]; struct afs_vnode *dvnode = dvp->vnode; _enter("op=%08x", op->debug_id); + fscache_begin_write_operation(&cres, afs_vnode_cache(dvnode)); down_write(&dvnode->validate_lock); if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags) && dvnode->status.data_version == dvp->dv_before + dvp->dv_delta) afs_edit_dir_remove(dvnode, &op->dentry->d_name, afs_edit_dir_for_unlink); up_write(&dvnode->validate_lock); + fscache_end_operation(&cres); } static void afs_unlink_put(struct afs_operation *op) @@ -1586,6 +1587,8 @@ static int afs_unlink(struct inode *dir, struct dentry *dentry) if (IS_ERR(op)) return PTR_ERR(op); + fscache_use_cookie(afs_vnode_cache(dvnode), true); + afs_op_set_vnode(op, 0, dvnode); op->file[0].dv_delta = 1; op->file[0].modification = true; @@ -1632,10 +1635,10 @@ static int afs_unlink(struct inode *dir, struct dentry *dentry) afs_wait_for_operation(op); } - return afs_put_operation(op); - error: - return afs_put_operation(op); + ret = afs_put_operation(op); + afs_dir_unuse_cookie(dvnode, ret); + return ret; } static const struct afs_operation_ops afs_create_operation = { @@ -1669,6 +1672,8 @@ static int afs_create(struct mnt_idmap *idmap, struct inode *dir, goto error; } + fscache_use_cookie(afs_vnode_cache(dvnode), true); + afs_op_set_vnode(op, 0, dvnode); op->file[0].dv_delta = 1; op->file[0].modification = true; @@ -1679,7 +1684,9 @@ static int afs_create(struct mnt_idmap *idmap, struct inode *dir, op->create.reason = afs_edit_dir_for_create; op->mtime = current_time(dir); op->ops = &afs_create_operation; - return afs_do_sync_operation(op); + ret = afs_do_sync_operation(op); + afs_dir_unuse_cookie(dvnode, ret); + return ret; error: d_drop(dentry); @@ -1744,6 +1751,8 @@ static int afs_link(struct dentry *from, struct inode *dir, goto error; } + fscache_use_cookie(afs_vnode_cache(dvnode), true); + ret = afs_validate(vnode, op->key); if (ret < 0) goto error_op; @@ -1759,10 +1768,13 @@ static int afs_link(struct dentry *from, struct inode *dir, op->dentry_2 = from; op->ops = &afs_link_operation; op->create.reason = afs_edit_dir_for_link; - return afs_do_sync_operation(op); + ret = afs_do_sync_operation(op); + afs_dir_unuse_cookie(dvnode, ret); + return ret; error_op: afs_put_operation(op); + afs_dir_unuse_cookie(dvnode, ret); error: d_drop(dentry); _leave(" = %d", ret); @@ -1806,6 +1818,8 @@ static int afs_symlink(struct mnt_idmap *idmap, struct inode *dir, goto error; } + fscache_use_cookie(afs_vnode_cache(dvnode), true); + afs_op_set_vnode(op, 0, dvnode); op->file[0].dv_delta = 1; @@ -1814,7 +1828,9 @@ static int afs_symlink(struct mnt_idmap *idmap, struct inode *dir, op->create.reason = afs_edit_dir_for_symlink; op->create.symlink = content; op->mtime = current_time(dir); - return afs_do_sync_operation(op); + ret = afs_do_sync_operation(op); + afs_dir_unuse_cookie(dvnode, ret); + return ret; error: d_drop(dentry); @@ -1846,6 +1862,7 @@ static void afs_rename_success(struct afs_operation *op) write_seqlock(&vnode->cb_lock); new_dv = vnode->status.data_version + 1; + trace_afs_set_dv(vnode, new_dv); vnode->status.data_version = new_dv; inode_set_iversion_raw(&vnode->netfs.inode, new_dv); @@ -1855,6 +1872,7 @@ static void afs_rename_success(struct afs_operation *op) static void afs_rename_edit_dir(struct afs_operation *op) { + struct netfs_cache_resources orig_cres = {}, new_cres = {}; struct afs_vnode_param *orig_dvp = &op->file[0]; struct afs_vnode_param *new_dvp = &op->file[1]; struct afs_vnode *orig_dvnode = orig_dvp->vnode; @@ -1871,6 +1889,10 @@ static void afs_rename_edit_dir(struct afs_operation *op) op->rename.rehash = NULL; } + fscache_begin_write_operation(&orig_cres, afs_vnode_cache(orig_dvnode)); + if (new_dvnode != orig_dvnode) + fscache_begin_write_operation(&new_cres, afs_vnode_cache(new_dvnode)); + down_write(&orig_dvnode->validate_lock); if (test_bit(AFS_VNODE_DIR_VALID, &orig_dvnode->flags) && orig_dvnode->status.data_version == orig_dvp->dv_before + orig_dvp->dv_delta) @@ -1920,6 +1942,9 @@ static void afs_rename_edit_dir(struct afs_operation *op) d_move(old_dentry, new_dentry); up_write(&new_dvnode->validate_lock); + fscache_end_operation(&orig_cres); + if (new_dvnode != orig_dvnode) + fscache_end_operation(&new_cres); } static void afs_rename_put(struct afs_operation *op) @@ -1972,6 +1997,10 @@ static int afs_rename(struct mnt_idmap *idmap, struct inode *old_dir, if (IS_ERR(op)) return PTR_ERR(op); + fscache_use_cookie(afs_vnode_cache(orig_dvnode), true); + if (new_dvnode != orig_dvnode) + fscache_use_cookie(afs_vnode_cache(new_dvnode), true); + ret = afs_validate(vnode, op->key); afs_op_set_error(op, ret); if (ret < 0) @@ -2039,47 +2068,43 @@ static int afs_rename(struct mnt_idmap *idmap, struct inode *old_dir, */ d_drop(old_dentry); - return afs_do_sync_operation(op); + ret = afs_do_sync_operation(op); +out: + afs_dir_unuse_cookie(orig_dvnode, ret); + if (new_dvnode != orig_dvnode) + afs_dir_unuse_cookie(new_dvnode, ret); + return ret; error: - return afs_put_operation(op); -} - -/* - * Release a directory folio and clean up its private state if it's not busy - * - return true if the folio can now be released, false if not - */ -static bool afs_dir_release_folio(struct folio *folio, gfp_t gfp_flags) -{ - struct afs_vnode *dvnode = AFS_FS_I(folio_inode(folio)); - - _enter("{{%llx:%llu}[%lu]}", dvnode->fid.vid, dvnode->fid.vnode, folio->index); - - folio_detach_private(folio); - - /* The directory will need reloading. */ - if (test_and_clear_bit(AFS_VNODE_DIR_VALID, &dvnode->flags)) - afs_stat_v(dvnode, n_relpg); - return true; + ret = afs_put_operation(op); + goto out; } /* - * Invalidate part or all of a folio. + * Write the file contents to the cache as a single blob. */ -static void afs_dir_invalidate_folio(struct folio *folio, size_t offset, - size_t length) +int afs_single_writepages(struct address_space *mapping, + struct writeback_control *wbc) { - struct afs_vnode *dvnode = AFS_FS_I(folio_inode(folio)); - - _enter("{%lu},%zu,%zu", folio->index, offset, length); - - BUG_ON(!folio_test_locked(folio)); + struct afs_vnode *dvnode = AFS_FS_I(mapping->host); + struct iov_iter iter; + bool is_dir = (S_ISDIR(dvnode->netfs.inode.i_mode) && + !test_bit(AFS_VNODE_MOUNTPOINT, &dvnode->flags)); + int ret = 0; - /* The directory will need reloading. */ - if (test_and_clear_bit(AFS_VNODE_DIR_VALID, &dvnode->flags)) - afs_stat_v(dvnode, n_inval); + /* Need to lock to prevent the folio queue and folios from being thrown + * away. + */ + down_read(&dvnode->validate_lock); + + if (is_dir ? + test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags) : + atomic64_read(&dvnode->cb_expires_at) != AFS_NO_CB_PROMISE) { + iov_iter_folio_queue(&iter, ITER_SOURCE, dvnode->directory, 0, 0, + i_size_read(&dvnode->netfs.inode)); + ret = netfs_writeback_single(mapping, wbc, &iter); + } - /* we clean up only if the entire folio is being invalidated */ - if (offset == 0 && length == folio_size(folio)) - folio_detach_private(folio); + up_read(&dvnode->validate_lock); + return ret; } diff --git a/fs/afs/dir_edit.c b/fs/afs/dir_edit.c index fe223fb78111..60a549f1d9c5 100644 --- a/fs/afs/dir_edit.c +++ b/fs/afs/dir_edit.c @@ -10,6 +10,7 @@ #include <linux/namei.h> #include <linux/pagemap.h> #include <linux/iversion.h> +#include <linux/folio_queue.h> #include "internal.h" #include "xdr_fs.h" @@ -105,23 +106,57 @@ static void afs_clear_contig_bits(union afs_xdr_dir_block *block, } /* - * Get a new directory folio. + * Get a specific block, extending the directory storage to cover it as needed. */ -static struct folio *afs_dir_get_folio(struct afs_vnode *vnode, pgoff_t index) +static union afs_xdr_dir_block *afs_dir_get_block(struct afs_dir_iter *iter, size_t block) { - struct address_space *mapping = vnode->netfs.inode.i_mapping; + struct folio_queue *fq; + struct afs_vnode *dvnode = iter->dvnode; struct folio *folio; + size_t blpos = block * AFS_DIR_BLOCK_SIZE; + size_t blend = (block + 1) * AFS_DIR_BLOCK_SIZE, fpos = iter->fpos; + int ret; + + if (dvnode->directory_size < blend) { + size_t cur_size = dvnode->directory_size; + + ret = netfs_alloc_folioq_buffer( + NULL, &dvnode->directory, &cur_size, blend, + mapping_gfp_mask(dvnode->netfs.inode.i_mapping)); + dvnode->directory_size = cur_size; + if (ret < 0) + goto fail; + } - folio = __filemap_get_folio(mapping, index, - FGP_LOCK | FGP_ACCESSED | FGP_CREAT, - mapping->gfp_mask); - if (IS_ERR(folio)) { - clear_bit(AFS_VNODE_DIR_VALID, &vnode->flags); - return NULL; + fq = iter->fq; + if (!fq) + fq = dvnode->directory; + + /* Search the folio queue for the folio containing the block... */ + for (; fq; fq = fq->next) { + for (int s = iter->fq_slot; s < folioq_count(fq); s++) { + size_t fsize = folioq_folio_size(fq, s); + + if (blend <= fpos + fsize) { + /* ... and then return the mapped block. */ + folio = folioq_folio(fq, s); + if (WARN_ON_ONCE(folio_pos(folio) != fpos)) + goto fail; + iter->fq = fq; + iter->fq_slot = s; + iter->fpos = fpos; + return kmap_local_folio(folio, blpos - fpos); + } + fpos += fsize; + } + iter->fq_slot = 0; } - if (!folio_test_private(folio)) - folio_attach_private(folio, (void *)1); - return folio; + +fail: + iter->fq = NULL; + iter->fq_slot = 0; + afs_invalidate_dir(dvnode, afs_dir_invalid_edit_get_block); + return NULL; } /* @@ -209,9 +244,8 @@ void afs_edit_dir_add(struct afs_vnode *vnode, { union afs_xdr_dir_block *meta, *block; union afs_xdr_dirent *de; - struct folio *folio0, *folio; - unsigned int need_slots, nr_blocks, b; - pgoff_t index; + struct afs_dir_iter iter = { .dvnode = vnode }; + unsigned int nr_blocks, b, entry; loff_t i_size; int slot; @@ -220,20 +254,17 @@ void afs_edit_dir_add(struct afs_vnode *vnode, i_size = i_size_read(&vnode->netfs.inode); if (i_size > AFS_DIR_BLOCK_SIZE * AFS_DIR_MAX_BLOCKS || (i_size & (AFS_DIR_BLOCK_SIZE - 1))) { - clear_bit(AFS_VNODE_DIR_VALID, &vnode->flags); + afs_invalidate_dir(vnode, afs_dir_invalid_edit_add_bad_size); return; } - folio0 = afs_dir_get_folio(vnode, 0); - if (!folio0) { - _leave(" [fgp]"); + meta = afs_dir_get_block(&iter, 0); + if (!meta) return; - } /* Work out how many slots we're going to need. */ - need_slots = afs_dir_calc_slots(name->len); + iter.nr_slots = afs_dir_calc_slots(name->len); - meta = kmap_local_folio(folio0, 0); if (i_size == 0) goto new_directory; nr_blocks = i_size / AFS_DIR_BLOCK_SIZE; @@ -245,22 +276,21 @@ void afs_edit_dir_add(struct afs_vnode *vnode, /* If the directory extended into a new folio, then we need to * tack a new folio on the end. */ - index = b / AFS_DIR_BLOCKS_PER_PAGE; if (nr_blocks >= AFS_DIR_MAX_BLOCKS) - goto error; - if (index >= folio_nr_pages(folio0)) { - folio = afs_dir_get_folio(vnode, index); - if (!folio) - goto error; - } else { - folio = folio0; - } + goto error_too_many_blocks; - block = kmap_local_folio(folio, b * AFS_DIR_BLOCK_SIZE - folio_pos(folio)); + /* Lower dir blocks have a counter in the header we can check. */ + if (b < AFS_DIR_BLOCKS_WITH_CTR && + meta->meta.alloc_ctrs[b] < iter.nr_slots) + continue; + + block = afs_dir_get_block(&iter, b); + if (!block) + goto error; /* Abandon the edit if we got a callback break. */ if (!test_bit(AFS_VNODE_DIR_VALID, &vnode->flags)) - goto invalidated; + goto already_invalidated; _debug("block %u: %2u %3u %u", b, @@ -275,31 +305,23 @@ void afs_edit_dir_add(struct afs_vnode *vnode, afs_set_i_size(vnode, (b + 1) * AFS_DIR_BLOCK_SIZE); } - /* Only lower dir blocks have a counter in the header. */ - if (b >= AFS_DIR_BLOCKS_WITH_CTR || - meta->meta.alloc_ctrs[b] >= need_slots) { - /* We need to try and find one or more consecutive - * slots to hold the entry. - */ - slot = afs_find_contig_bits(block, need_slots); - if (slot >= 0) { - _debug("slot %u", slot); - goto found_space; - } + /* We need to try and find one or more consecutive slots to + * hold the entry. + */ + slot = afs_find_contig_bits(block, iter.nr_slots); + if (slot >= 0) { + _debug("slot %u", slot); + goto found_space; } kunmap_local(block); - if (folio != folio0) { - folio_unlock(folio); - folio_put(folio); - } } /* There are no spare slots of sufficient size, yet the operation * succeeded. Download the directory again. */ trace_afs_edit_dir(vnode, why, afs_edit_dir_create_nospc, 0, 0, 0, 0, name->name); - clear_bit(AFS_VNODE_DIR_VALID, &vnode->flags); + afs_invalidate_dir(vnode, afs_dir_invalid_edit_add_no_slots); goto out_unmap; new_directory: @@ -307,8 +329,7 @@ new_directory: i_size = AFS_DIR_BLOCK_SIZE; afs_set_i_size(vnode, i_size); slot = AFS_DIR_RESV_BLOCKS0; - folio = folio0; - block = kmap_local_folio(folio, 0); + block = afs_dir_get_block(&iter, 0); nr_blocks = 1; b = 0; @@ -326,41 +347,39 @@ found_space: de->u.name[name->len] = 0; /* Adjust the bitmap. */ - afs_set_contig_bits(block, slot, need_slots); - kunmap_local(block); - if (folio != folio0) { - folio_unlock(folio); - folio_put(folio); - } + afs_set_contig_bits(block, slot, iter.nr_slots); /* Adjust the allocation counter. */ if (b < AFS_DIR_BLOCKS_WITH_CTR) - meta->meta.alloc_ctrs[b] -= need_slots; + meta->meta.alloc_ctrs[b] -= iter.nr_slots; + + /* Adjust the hash chain. */ + entry = b * AFS_DIR_SLOTS_PER_BLOCK + slot; + iter.bucket = afs_dir_hash_name(name); + de->u.hash_next = meta->meta.hashtable[iter.bucket]; + meta->meta.hashtable[iter.bucket] = htons(entry); + kunmap_local(block); inode_inc_iversion_raw(&vnode->netfs.inode); afs_stat_v(vnode, n_dir_cr); _debug("Insert %s in %u[%u]", name->name, b, slot); + netfs_single_mark_inode_dirty(&vnode->netfs.inode); + out_unmap: kunmap_local(meta); - folio_unlock(folio0); - folio_put(folio0); _leave(""); return; -invalidated: +already_invalidated: trace_afs_edit_dir(vnode, why, afs_edit_dir_create_inval, 0, 0, 0, 0, name->name); - clear_bit(AFS_VNODE_DIR_VALID, &vnode->flags); kunmap_local(block); - if (folio != folio0) { - folio_unlock(folio); - folio_put(folio); - } goto out_unmap; +error_too_many_blocks: + afs_invalidate_dir(vnode, afs_dir_invalid_edit_add_too_many_blocks); error: trace_afs_edit_dir(vnode, why, afs_edit_dir_create_error, 0, 0, 0, 0, name->name); - clear_bit(AFS_VNODE_DIR_VALID, &vnode->flags); goto out_unmap; } @@ -374,13 +393,14 @@ error: void afs_edit_dir_remove(struct afs_vnode *vnode, struct qstr *name, enum afs_edit_dir_reason why) { - union afs_xdr_dir_block *meta, *block; - union afs_xdr_dirent *de; - struct folio *folio0, *folio; - unsigned int need_slots, nr_blocks, b; - pgoff_t index; + union afs_xdr_dir_block *meta, *block, *pblock; + union afs_xdr_dirent *de, *pde; + struct afs_dir_iter iter = { .dvnode = vnode }; + struct afs_fid fid; + unsigned int b, slot, entry; loff_t i_size; - int slot; + __be16 next; + int found; _enter(",,{%d,%s},", name->len, name->name); @@ -388,81 +408,95 @@ void afs_edit_dir_remove(struct afs_vnode *vnode, if (i_size < AFS_DIR_BLOCK_SIZE || i_size > AFS_DIR_BLOCK_SIZE * AFS_DIR_MAX_BLOCKS || (i_size & (AFS_DIR_BLOCK_SIZE - 1))) { - clear_bit(AFS_VNODE_DIR_VALID, &vnode->flags); + afs_invalidate_dir(vnode, afs_dir_invalid_edit_rem_bad_size); return; } - nr_blocks = i_size / AFS_DIR_BLOCK_SIZE; - folio0 = afs_dir_get_folio(vnode, 0); - if (!folio0) { - _leave(" [fgp]"); + if (!afs_dir_init_iter(&iter, name)) return; - } - - /* Work out how many slots we're going to discard. */ - need_slots = afs_dir_calc_slots(name->len); - - meta = kmap_local_folio(folio0, 0); - - /* Find a block that has sufficient slots available. Each folio - * contains two or more directory blocks. - */ - for (b = 0; b < nr_blocks; b++) { - index = b / AFS_DIR_BLOCKS_PER_PAGE; - if (index >= folio_nr_pages(folio0)) { - folio = afs_dir_get_folio(vnode, index); - if (!folio) - goto error; - } else { - folio = folio0; - } - block = kmap_local_folio(folio, b * AFS_DIR_BLOCK_SIZE - folio_pos(folio)); - - /* Abandon the edit if we got a callback break. */ - if (!test_bit(AFS_VNODE_DIR_VALID, &vnode->flags)) - goto invalidated; - - if (b > AFS_DIR_BLOCKS_WITH_CTR || - meta->meta.alloc_ctrs[b] <= AFS_DIR_SLOTS_PER_BLOCK - 1 - need_slots) { - slot = afs_dir_scan_block(block, name, b); - if (slot >= 0) - goto found_dirent; - } + meta = afs_dir_find_block(&iter, 0); + if (!meta) + return; - kunmap_local(block); - if (folio != folio0) { - folio_unlock(folio); - folio_put(folio); - } + /* Find the entry in the blob. */ + found = afs_dir_search_bucket(&iter, name, &fid); + if (found < 0) { + /* Didn't find the dirent to clobber. Re-download. */ + trace_afs_edit_dir(vnode, why, afs_edit_dir_delete_noent, + 0, 0, 0, 0, name->name); + afs_invalidate_dir(vnode, afs_dir_invalid_edit_rem_wrong_name); + goto out_unmap; } - /* Didn't find the dirent to clobber. Download the directory again. */ - trace_afs_edit_dir(vnode, why, afs_edit_dir_delete_noent, - 0, 0, 0, 0, name->name); - clear_bit(AFS_VNODE_DIR_VALID, &vnode->flags); - goto out_unmap; + entry = found; + b = entry / AFS_DIR_SLOTS_PER_BLOCK; + slot = entry % AFS_DIR_SLOTS_PER_BLOCK; -found_dirent: + block = afs_dir_find_block(&iter, b); + if (!block) + goto error; + if (!test_bit(AFS_VNODE_DIR_VALID, &vnode->flags)) + goto already_invalidated; + + /* Check and clear the entry. */ de = &block->dirents[slot]; + if (de->u.valid != 1) + goto error_unmap; trace_afs_edit_dir(vnode, why, afs_edit_dir_delete, b, slot, ntohl(de->u.vnode), ntohl(de->u.unique), name->name); - memset(de, 0, sizeof(*de) * need_slots); - /* Adjust the bitmap. */ - afs_clear_contig_bits(block, slot, need_slots); - kunmap_local(block); - if (folio != folio0) { - folio_unlock(folio); - folio_put(folio); - } + afs_clear_contig_bits(block, slot, iter.nr_slots); /* Adjust the allocation counter. */ if (b < AFS_DIR_BLOCKS_WITH_CTR) - meta->meta.alloc_ctrs[b] += need_slots; + meta->meta.alloc_ctrs[b] += iter.nr_slots; + + /* Clear the constituent entries. */ + next = de->u.hash_next; + memset(de, 0, sizeof(*de) * iter.nr_slots); + kunmap_local(block); + + /* Adjust the hash chain: if iter->prev_entry is 0, the hashtable head + * index is previous; otherwise it's slot number of the previous entry. + */ + if (!iter.prev_entry) { + __be16 prev_next = meta->meta.hashtable[iter.bucket]; + + if (unlikely(prev_next != htons(entry))) { + pr_warn("%llx:%llx:%x: not head of chain b=%x p=%x,%x e=%x %*s", + vnode->fid.vid, vnode->fid.vnode, vnode->fid.unique, + iter.bucket, iter.prev_entry, prev_next, entry, + name->len, name->name); + goto error; + } + meta->meta.hashtable[iter.bucket] = next; + } else { + unsigned int pb = iter.prev_entry / AFS_DIR_SLOTS_PER_BLOCK; + unsigned int ps = iter.prev_entry % AFS_DIR_SLOTS_PER_BLOCK; + __be16 prev_next; + + pblock = afs_dir_find_block(&iter, pb); + if (!pblock) + goto error; + pde = &pblock->dirents[ps]; + prev_next = pde->u.hash_next; + if (prev_next != htons(entry)) { + kunmap_local(pblock); + pr_warn("%llx:%llx:%x: not prev in chain b=%x p=%x,%x e=%x %*s", + vnode->fid.vid, vnode->fid.vnode, vnode->fid.unique, + iter.bucket, iter.prev_entry, prev_next, entry, + name->len, name->name); + goto error; + } + pde->u.hash_next = next; + kunmap_local(pblock); + } + + netfs_single_mark_inode_dirty(&vnode->netfs.inode); inode_set_iversion_raw(&vnode->netfs.inode, vnode->status.data_version); afs_stat_v(vnode, n_dir_rm); @@ -470,26 +504,20 @@ found_dirent: out_unmap: kunmap_local(meta); - folio_unlock(folio0); - folio_put(folio0); _leave(""); return; -invalidated: +already_invalidated: + kunmap_local(block); trace_afs_edit_dir(vnode, why, afs_edit_dir_delete_inval, 0, 0, 0, 0, name->name); - clear_bit(AFS_VNODE_DIR_VALID, &vnode->flags); - kunmap_local(block); - if (folio != folio0) { - folio_unlock(folio); - folio_put(folio); - } goto out_unmap; +error_unmap: + kunmap_local(block); error: trace_afs_edit_dir(vnode, why, afs_edit_dir_delete_error, 0, 0, 0, 0, name->name); - clear_bit(AFS_VNODE_DIR_VALID, &vnode->flags); goto out_unmap; } @@ -502,9 +530,8 @@ void afs_edit_dir_update_dotdot(struct afs_vnode *vnode, struct afs_vnode *new_d { union afs_xdr_dir_block *block; union afs_xdr_dirent *de; - struct folio *folio; + struct afs_dir_iter iter = { .dvnode = vnode }; unsigned int nr_blocks, b; - pgoff_t index; loff_t i_size; int slot; @@ -512,39 +539,35 @@ void afs_edit_dir_update_dotdot(struct afs_vnode *vnode, struct afs_vnode *new_d i_size = i_size_read(&vnode->netfs.inode); if (i_size < AFS_DIR_BLOCK_SIZE) { - clear_bit(AFS_VNODE_DIR_VALID, &vnode->flags); + afs_invalidate_dir(vnode, afs_dir_invalid_edit_upd_bad_size); return; } + nr_blocks = i_size / AFS_DIR_BLOCK_SIZE; /* Find a block that has sufficient slots available. Each folio * contains two or more directory blocks. */ for (b = 0; b < nr_blocks; b++) { - index = b / AFS_DIR_BLOCKS_PER_PAGE; - folio = afs_dir_get_folio(vnode, index); - if (!folio) + block = afs_dir_get_block(&iter, b); + if (!block) goto error; - block = kmap_local_folio(folio, b * AFS_DIR_BLOCK_SIZE - folio_pos(folio)); - /* Abandon the edit if we got a callback break. */ if (!test_bit(AFS_VNODE_DIR_VALID, &vnode->flags)) - goto invalidated; + goto already_invalidated; slot = afs_dir_scan_block(block, &dotdot_name, b); if (slot >= 0) goto found_dirent; kunmap_local(block); - folio_unlock(folio); - folio_put(folio); } /* Didn't find the dirent to clobber. Download the directory again. */ trace_afs_edit_dir(vnode, why, afs_edit_dir_update_nodd, 0, 0, 0, 0, ".."); - clear_bit(AFS_VNODE_DIR_VALID, &vnode->flags); + afs_invalidate_dir(vnode, afs_dir_invalid_edit_upd_no_dd); goto out; found_dirent: @@ -556,26 +579,70 @@ found_dirent: ntohl(de->u.vnode), ntohl(de->u.unique), ".."); kunmap_local(block); - folio_unlock(folio); - folio_put(folio); + netfs_single_mark_inode_dirty(&vnode->netfs.inode); inode_set_iversion_raw(&vnode->netfs.inode, vnode->status.data_version); out: _leave(""); return; -invalidated: +already_invalidated: kunmap_local(block); - folio_unlock(folio); - folio_put(folio); trace_afs_edit_dir(vnode, why, afs_edit_dir_update_inval, 0, 0, 0, 0, ".."); - clear_bit(AFS_VNODE_DIR_VALID, &vnode->flags); goto out; error: trace_afs_edit_dir(vnode, why, afs_edit_dir_update_error, 0, 0, 0, 0, ".."); - clear_bit(AFS_VNODE_DIR_VALID, &vnode->flags); goto out; } + +/* + * Initialise a new directory. We need to fill in the "." and ".." entries. + */ +void afs_mkdir_init_dir(struct afs_vnode *dvnode, struct afs_vnode *parent_dvnode) +{ + union afs_xdr_dir_block *meta; + struct afs_dir_iter iter = { .dvnode = dvnode }; + union afs_xdr_dirent *de; + unsigned int slot = AFS_DIR_RESV_BLOCKS0; + loff_t i_size; + + i_size = i_size_read(&dvnode->netfs.inode); + if (i_size != AFS_DIR_BLOCK_SIZE) { + afs_invalidate_dir(dvnode, afs_dir_invalid_edit_add_bad_size); + return; + } + + meta = afs_dir_get_block(&iter, 0); + if (!meta) + return; + + afs_edit_init_block(meta, meta, 0); + + de = &meta->dirents[slot]; + de->u.valid = 1; + de->u.vnode = htonl(dvnode->fid.vnode); + de->u.unique = htonl(dvnode->fid.unique); + memcpy(de->u.name, ".", 2); + trace_afs_edit_dir(dvnode, afs_edit_dir_for_mkdir, afs_edit_dir_mkdir, 0, slot, + dvnode->fid.vnode, dvnode->fid.unique, "."); + slot++; + + de = &meta->dirents[slot]; + de->u.valid = 1; + de->u.vnode = htonl(parent_dvnode->fid.vnode); + de->u.unique = htonl(parent_dvnode->fid.unique); + memcpy(de->u.name, "..", 3); + trace_afs_edit_dir(dvnode, afs_edit_dir_for_mkdir, afs_edit_dir_mkdir, 0, slot, + parent_dvnode->fid.vnode, parent_dvnode->fid.unique, ".."); + + afs_set_contig_bits(meta, AFS_DIR_RESV_BLOCKS0, 2); + meta->meta.alloc_ctrs[0] -= 2; + kunmap_local(meta); + + netfs_single_mark_inode_dirty(&dvnode->netfs.inode); + set_bit(AFS_VNODE_DIR_VALID, &dvnode->flags); + set_bit(AFS_VNODE_DIR_READ, &dvnode->flags); +} diff --git a/fs/afs/dir_search.c b/fs/afs/dir_search.c new file mode 100644 index 000000000000..b25bd892db4d --- /dev/null +++ b/fs/afs/dir_search.c @@ -0,0 +1,227 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* Search a directory's hash table. + * + * Copyright (C) 2024 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + * + * https://tools.ietf.org/html/draft-keiser-afs3-directory-object-00 + */ + +#include <linux/kernel.h> +#include <linux/fs.h> +#include <linux/namei.h> +#include <linux/iversion.h> +#include "internal.h" +#include "afs_fs.h" +#include "xdr_fs.h" + +/* + * Calculate the name hash. + */ +unsigned int afs_dir_hash_name(const struct qstr *name) +{ + const unsigned char *p = name->name; + unsigned int hash = 0, i; + int bucket; + + for (i = 0; i < name->len; i++) + hash = (hash * 173) + p[i]; + bucket = hash & (AFS_DIR_HASHTBL_SIZE - 1); + if (hash > INT_MAX) { + bucket = AFS_DIR_HASHTBL_SIZE - bucket; + bucket &= (AFS_DIR_HASHTBL_SIZE - 1); + } + return bucket; +} + +/* + * Reset a directory iterator. + */ +static bool afs_dir_reset_iter(struct afs_dir_iter *iter) +{ + unsigned long long i_size = i_size_read(&iter->dvnode->netfs.inode); + unsigned int nblocks; + + /* Work out the maximum number of steps we can take. */ + nblocks = umin(i_size / AFS_DIR_BLOCK_SIZE, AFS_DIR_MAX_BLOCKS); + if (!nblocks) + return false; + iter->loop_check = nblocks * (AFS_DIR_SLOTS_PER_BLOCK - AFS_DIR_RESV_BLOCKS); + iter->prev_entry = 0; /* Hash head is previous */ + return true; +} + +/* + * Initialise a directory iterator for looking up a name. + */ +bool afs_dir_init_iter(struct afs_dir_iter *iter, const struct qstr *name) +{ + iter->nr_slots = afs_dir_calc_slots(name->len); + iter->bucket = afs_dir_hash_name(name); + return afs_dir_reset_iter(iter); +} + +/* + * Get a specific block. + */ +union afs_xdr_dir_block *afs_dir_find_block(struct afs_dir_iter *iter, size_t block) +{ + struct folio_queue *fq = iter->fq; + struct afs_vnode *dvnode = iter->dvnode; + struct folio *folio; + size_t blpos = block * AFS_DIR_BLOCK_SIZE; + size_t blend = (block + 1) * AFS_DIR_BLOCK_SIZE, fpos = iter->fpos; + int slot = iter->fq_slot; + + _enter("%zx,%d", block, slot); + + if (iter->block) { + kunmap_local(iter->block); + iter->block = NULL; + } + + if (dvnode->directory_size < blend) + goto fail; + + if (!fq || blpos < fpos) { + fq = dvnode->directory; + slot = 0; + fpos = 0; + } + + /* Search the folio queue for the folio containing the block... */ + for (; fq; fq = fq->next) { + for (; slot < folioq_count(fq); slot++) { + size_t fsize = folioq_folio_size(fq, slot); + + if (blend <= fpos + fsize) { + /* ... and then return the mapped block. */ + folio = folioq_folio(fq, slot); + if (WARN_ON_ONCE(folio_pos(folio) != fpos)) + goto fail; + iter->fq = fq; + iter->fq_slot = slot; + iter->fpos = fpos; + iter->block = kmap_local_folio(folio, blpos - fpos); + return iter->block; + } + fpos += fsize; + } + slot = 0; + } + +fail: + iter->fq = NULL; + iter->fq_slot = 0; + afs_invalidate_dir(dvnode, afs_dir_invalid_edit_get_block); + return NULL; +} + +/* + * Search through a directory bucket. + */ +int afs_dir_search_bucket(struct afs_dir_iter *iter, const struct qstr *name, + struct afs_fid *_fid) +{ + const union afs_xdr_dir_block *meta; + unsigned int entry; + int ret = -ESTALE; + + meta = afs_dir_find_block(iter, 0); + if (!meta) + return -ESTALE; + + entry = ntohs(meta->meta.hashtable[iter->bucket & (AFS_DIR_HASHTBL_SIZE - 1)]); + _enter("%x,%x", iter->bucket, entry); + + while (entry) { + const union afs_xdr_dir_block *block; + const union afs_xdr_dirent *dire; + unsigned int blnum = entry / AFS_DIR_SLOTS_PER_BLOCK; + unsigned int slot = entry % AFS_DIR_SLOTS_PER_BLOCK; + unsigned int resv = (blnum == 0 ? AFS_DIR_RESV_BLOCKS0 : AFS_DIR_RESV_BLOCKS); + + _debug("search %x", entry); + + if (slot < resv) { + kdebug("slot out of range h=%x rs=%2x sl=%2x-%2x", + iter->bucket, resv, slot, slot + iter->nr_slots - 1); + goto bad; + } + + block = afs_dir_find_block(iter, blnum); + if (!block) + goto bad; + dire = &block->dirents[slot]; + + if (slot + iter->nr_slots <= AFS_DIR_SLOTS_PER_BLOCK && + memcmp(dire->u.name, name->name, name->len) == 0 && + dire->u.name[name->len] == '\0') { + _fid->vnode = ntohl(dire->u.vnode); + _fid->unique = ntohl(dire->u.unique); + ret = entry; + goto found; + } + + iter->prev_entry = entry; + entry = ntohs(dire->u.hash_next); + if (!--iter->loop_check) { + kdebug("dir chain loop h=%x", iter->bucket); + goto bad; + } + } + + ret = -ENOENT; +found: + if (iter->block) { + kunmap_local(iter->block); + iter->block = NULL; + } + +bad: + if (ret == -ESTALE) + afs_invalidate_dir(iter->dvnode, afs_dir_invalid_iter_stale); + _leave(" = %d", ret); + return ret; +} + +/* + * Search the appropriate hash chain in the contents of an AFS directory. + */ +int afs_dir_search(struct afs_vnode *dvnode, struct qstr *name, + struct afs_fid *_fid, afs_dataversion_t *_dir_version) +{ + struct afs_dir_iter iter = { .dvnode = dvnode, }; + int ret, retry_limit = 3; + + _enter("{%lu},,,", dvnode->netfs.inode.i_ino); + + if (!afs_dir_init_iter(&iter, name)) + return -ENOENT; + do { + if (--retry_limit < 0) { + pr_warn("afs_read_dir(): Too many retries\n"); + ret = -ESTALE; + break; + } + ret = afs_read_dir(dvnode, NULL); + if (ret < 0) { + if (ret != -ESTALE) + break; + if (test_bit(AFS_VNODE_DELETED, &dvnode->flags)) { + ret = -ESTALE; + break; + } + continue; + } + *_dir_version = inode_peek_iversion_raw(&dvnode->netfs.inode); + + ret = afs_dir_search_bucket(&iter, name, _fid); + up_read(&dvnode->validate_lock); + if (ret == -ESTALE) + afs_dir_reset_iter(&iter); + } while (ret == -ESTALE); + + _leave(" = %d", ret); + return ret; +} diff --git a/fs/afs/dynroot.c b/fs/afs/dynroot.c index c4d2711e20ad..d8bf52f77d93 100644 --- a/fs/afs/dynroot.c +++ b/fs/afs/dynroot.c @@ -186,50 +186,6 @@ out: } /* - * Look up @cell in a dynroot directory. This is a substitution for the - * local cell name for the net namespace. - */ -static struct dentry *afs_lookup_atcell(struct dentry *dentry) -{ - struct afs_cell *cell; - struct afs_net *net = afs_d2net(dentry); - struct dentry *ret; - char *name; - int len; - - if (!net->ws_cell) - return ERR_PTR(-ENOENT); - - ret = ERR_PTR(-ENOMEM); - name = kmalloc(AFS_MAXCELLNAME + 1, GFP_KERNEL); - if (!name) - goto out_p; - - down_read(&net->cells_lock); - cell = net->ws_cell; - if (cell) { - len = cell->name_len; - memcpy(name, cell->name, len + 1); - } - up_read(&net->cells_lock); - - ret = ERR_PTR(-ENOENT); - if (!cell) - goto out_n; - - ret = lookup_one_len(name, dentry->d_parent, len); - - /* We don't want to d_add() the @cell dentry here as we don't want to - * the cached dentry to hide changes to the local cell name. - */ - -out_n: - kfree(name); -out_p: - return ret; -} - -/* * Look up an entry in a dynroot directory. */ static struct dentry *afs_dynroot_lookup(struct inode *dir, struct dentry *dentry, @@ -247,10 +203,6 @@ static struct dentry *afs_dynroot_lookup(struct inode *dir, struct dentry *dentr return ERR_PTR(-ENAMETOOLONG); } - if (dentry->d_name.len == 5 && - memcmp(dentry->d_name.name, "@cell", 5) == 0) - return afs_lookup_atcell(dentry); - return d_splice_alias(afs_try_auto_mntpt(dentry, dir), dentry); } @@ -271,7 +223,8 @@ const struct dentry_operations afs_dynroot_dentry_operations = { int afs_dynroot_mkdir(struct afs_net *net, struct afs_cell *cell) { struct super_block *sb = net->dynroot_sb; - struct dentry *root, *subdir; + struct dentry *root, *subdir, *dsubdir; + char *dotname = cell->name - 1; int ret; if (!sb || atomic_read(&sb->s_active) == 0) @@ -286,14 +239,43 @@ int afs_dynroot_mkdir(struct afs_net *net, struct afs_cell *cell) goto unlock; } - /* Note that we're retaining an extra ref on the dentry */ + dsubdir = lookup_one_len(dotname, root, cell->name_len + 1); + if (IS_ERR(dsubdir)) { + ret = PTR_ERR(dsubdir); + dput(subdir); + goto unlock; + } + + /* Note that we're retaining extra refs on the dentries. */ subdir->d_fsdata = (void *)1UL; + dsubdir->d_fsdata = (void *)1UL; ret = 0; unlock: inode_unlock(root->d_inode); return ret; } +static void afs_dynroot_rm_one_dir(struct dentry *root, const char *name, size_t name_len) +{ + struct dentry *subdir; + + /* Don't want to trigger a lookup call, which will re-add the cell */ + subdir = try_lookup_one_len(name, root, name_len); + if (IS_ERR_OR_NULL(subdir)) { + _debug("lookup %ld", PTR_ERR(subdir)); + return; + } + + _debug("rmdir %pd %u", subdir, d_count(subdir)); + + if (subdir->d_fsdata) { + _debug("unpin %u", d_count(subdir)); + subdir->d_fsdata = NULL; + dput(subdir); + } + dput(subdir); +} + /* * Remove a manually added cell mount directory. * - The caller must hold net->proc_cells_lock @@ -301,32 +283,141 @@ unlock: void afs_dynroot_rmdir(struct afs_net *net, struct afs_cell *cell) { struct super_block *sb = net->dynroot_sb; - struct dentry *root, *subdir; + char *dotname = cell->name - 1; if (!sb || atomic_read(&sb->s_active) == 0) return; - root = sb->s_root; - inode_lock(root->d_inode); + inode_lock(sb->s_root->d_inode); + afs_dynroot_rm_one_dir(sb->s_root, cell->name, cell->name_len); + afs_dynroot_rm_one_dir(sb->s_root, dotname, cell->name_len + 1); + inode_unlock(sb->s_root->d_inode); + _leave(""); +} - /* Don't want to trigger a lookup call, which will re-add the cell */ - subdir = try_lookup_one_len(cell->name, root, cell->name_len); - if (IS_ERR_OR_NULL(subdir)) { - _debug("lookup %ld", PTR_ERR(subdir)); - goto no_dentry; +static void afs_atcell_delayed_put_cell(void *arg) +{ + struct afs_cell *cell = arg; + + afs_put_cell(cell, afs_cell_trace_put_atcell); +} + +/* + * Read @cell or .@cell symlinks. + */ +static const char *afs_atcell_get_link(struct dentry *dentry, struct inode *inode, + struct delayed_call *done) +{ + struct afs_vnode *vnode = AFS_FS_I(inode); + struct afs_cell *cell; + struct afs_net *net = afs_i2net(inode); + const char *name; + bool dotted = vnode->fid.vnode == 3; + + if (!net->ws_cell) + return ERR_PTR(-ENOENT); + + down_read(&net->cells_lock); + + cell = net->ws_cell; + if (dotted) + name = cell->name - 1; + else + name = cell->name; + afs_get_cell(cell, afs_cell_trace_get_atcell); + set_delayed_call(done, afs_atcell_delayed_put_cell, cell); + + up_read(&net->cells_lock); + return name; +} + +static const struct inode_operations afs_atcell_inode_operations = { + .get_link = afs_atcell_get_link, +}; + +/* + * Look up @cell or .@cell in a dynroot directory. This is a substitution for + * the local cell name for the net namespace. + */ +static struct dentry *afs_dynroot_create_symlink(struct dentry *root, const char *name) +{ + struct afs_vnode *vnode; + struct afs_fid fid = { .vnode = 2, .unique = 1, }; + struct dentry *dentry; + struct inode *inode; + + if (name[0] == '.') + fid.vnode = 3; + + dentry = d_alloc_name(root, name); + if (!dentry) + return ERR_PTR(-ENOMEM); + + inode = iget5_locked(dentry->d_sb, fid.vnode, + afs_iget5_pseudo_test, afs_iget5_pseudo_set, &fid); + if (!inode) { + dput(dentry); + return ERR_PTR(-ENOMEM); } - _debug("rmdir %pd %u", subdir, d_count(subdir)); + vnode = AFS_FS_I(inode); - if (subdir->d_fsdata) { - _debug("unpin %u", d_count(subdir)); - subdir->d_fsdata = NULL; - dput(subdir); + /* there shouldn't be an existing inode */ + if (WARN_ON_ONCE(!(inode->i_state & I_NEW))) { + iput(inode); + dput(dentry); + return ERR_PTR(-EIO); } - dput(subdir); -no_dentry: + + netfs_inode_init(&vnode->netfs, NULL, false); + simple_inode_init_ts(inode); + set_nlink(inode, 1); + inode->i_size = 0; + inode->i_mode = S_IFLNK | 0555; + inode->i_op = &afs_atcell_inode_operations; + inode->i_uid = GLOBAL_ROOT_UID; + inode->i_gid = GLOBAL_ROOT_GID; + inode->i_blocks = 0; + inode->i_generation = 0; + inode->i_flags |= S_NOATIME; + + unlock_new_inode(inode); + d_splice_alias(inode, dentry); + return dentry; +} + +/* + * Create @cell and .@cell symlinks. + */ +static int afs_dynroot_symlink(struct afs_net *net) +{ + struct super_block *sb = net->dynroot_sb; + struct dentry *root, *symlink, *dsymlink; + int ret; + + /* Let the ->lookup op do the creation */ + root = sb->s_root; + inode_lock(root->d_inode); + symlink = afs_dynroot_create_symlink(root, "@cell"); + if (IS_ERR(symlink)) { + ret = PTR_ERR(symlink); + goto unlock; + } + + dsymlink = afs_dynroot_create_symlink(root, ".@cell"); + if (IS_ERR(dsymlink)) { + ret = PTR_ERR(dsymlink); + dput(symlink); + goto unlock; + } + + /* Note that we're retaining extra refs on the dentries. */ + symlink->d_fsdata = (void *)1UL; + dsymlink->d_fsdata = (void *)1UL; + ret = 0; +unlock: inode_unlock(root->d_inode); - _leave(""); + return ret; } /* @@ -341,6 +432,10 @@ int afs_dynroot_populate(struct super_block *sb) mutex_lock(&net->proc_cells_lock); net->dynroot_sb = sb; + ret = afs_dynroot_symlink(net); + if (ret < 0) + goto error; + hlist_for_each_entry(cell, &net->proc_cells, proc_link) { ret = afs_dynroot_mkdir(net, cell); if (ret < 0) diff --git a/fs/afs/file.c b/fs/afs/file.c index 6762eff97517..fc15497608c6 100644 --- a/fs/afs/file.c +++ b/fs/afs/file.c @@ -20,7 +20,6 @@ #include "internal.h" static int afs_file_mmap(struct file *file, struct vm_area_struct *vma); -static int afs_symlink_read_folio(struct file *file, struct folio *folio); static ssize_t afs_file_read_iter(struct kiocb *iocb, struct iov_iter *iter); static ssize_t afs_file_splice_read(struct file *in, loff_t *ppos, @@ -61,13 +60,6 @@ const struct address_space_operations afs_file_aops = { .writepages = afs_writepages, }; -const struct address_space_operations afs_symlink_aops = { - .read_folio = afs_symlink_read_folio, - .release_folio = netfs_release_folio, - .invalidate_folio = netfs_invalidate_folio, - .migrate_folio = filemap_migrate_folio, -}; - static const struct vm_operations_struct afs_vm_ops = { .open = afs_vm_open, .close = afs_vm_close, @@ -208,49 +200,12 @@ int afs_release(struct inode *inode, struct file *file) return ret; } -/* - * Allocate a new read record. - */ -struct afs_read *afs_alloc_read(gfp_t gfp) -{ - struct afs_read *req; - - req = kzalloc(sizeof(struct afs_read), gfp); - if (req) - refcount_set(&req->usage, 1); - - return req; -} - -/* - * Dispose of a ref to a read record. - */ -void afs_put_read(struct afs_read *req) -{ - if (refcount_dec_and_test(&req->usage)) { - if (req->cleanup) - req->cleanup(req); - key_put(req->key); - kfree(req); - } -} - static void afs_fetch_data_notify(struct afs_operation *op) { - struct afs_read *req = op->fetch.req; - struct netfs_io_subrequest *subreq = req->subreq; - int error = afs_op_error(op); - - req->error = error; - if (subreq) { - subreq->rreq->i_size = req->file_size; - if (req->pos + req->actual_len >= req->file_size) - __set_bit(NETFS_SREQ_HIT_EOF, &subreq->flags); - netfs_read_subreq_terminated(subreq, error, false); - req->subreq = NULL; - } else if (req->done) { - req->done(req); - } + struct netfs_io_subrequest *subreq = op->fetch.subreq; + + subreq->error = afs_op_error(op); + netfs_read_subreq_terminated(subreq); } static void afs_fetch_data_success(struct afs_operation *op) @@ -260,7 +215,7 @@ static void afs_fetch_data_success(struct afs_operation *op) _enter("op=%08x", op->debug_id); afs_vnode_commit_status(op, &op->file[0]); afs_stat_v(vnode, n_fetches); - atomic_long_add(op->fetch.req->actual_len, &op->net->n_fetch_bytes); + atomic_long_add(op->fetch.subreq->transferred, &op->net->n_fetch_bytes); afs_fetch_data_notify(op); } @@ -270,107 +225,188 @@ static void afs_fetch_data_aborted(struct afs_operation *op) afs_fetch_data_notify(op); } -static void afs_fetch_data_put(struct afs_operation *op) -{ - op->fetch.req->error = afs_op_error(op); - afs_put_read(op->fetch.req); -} - -static const struct afs_operation_ops afs_fetch_data_operation = { +const struct afs_operation_ops afs_fetch_data_operation = { .issue_afs_rpc = afs_fs_fetch_data, .issue_yfs_rpc = yfs_fs_fetch_data, .success = afs_fetch_data_success, .aborted = afs_fetch_data_aborted, .failed = afs_fetch_data_notify, - .put = afs_fetch_data_put, }; +static void afs_issue_read_call(struct afs_operation *op) +{ + op->call_responded = false; + op->call_error = 0; + op->call_abort_code = 0; + if (test_bit(AFS_SERVER_FL_IS_YFS, &op->server->flags)) + yfs_fs_fetch_data(op); + else + afs_fs_fetch_data(op); +} + +static void afs_end_read(struct afs_operation *op) +{ + if (op->call_responded && op->server) + set_bit(AFS_SERVER_FL_RESPONDING, &op->server->flags); + + if (!afs_op_error(op)) + afs_fetch_data_success(op); + else if (op->cumul_error.aborted) + afs_fetch_data_aborted(op); + else + afs_fetch_data_notify(op); + + afs_end_vnode_operation(op); + afs_put_operation(op); +} + +/* + * Perform I/O processing on an asynchronous call. The work item carries a ref + * to the call struct that we either need to release or to pass on. + */ +static void afs_read_receive(struct afs_call *call) +{ + struct afs_operation *op = call->op; + enum afs_call_state state; + + _enter(""); + + state = READ_ONCE(call->state); + if (state == AFS_CALL_COMPLETE) + return; + trace_afs_read_recv(op, call); + + while (state < AFS_CALL_COMPLETE && READ_ONCE(call->need_attention)) { + WRITE_ONCE(call->need_attention, false); + afs_deliver_to_call(call); + state = READ_ONCE(call->state); + } + + if (state < AFS_CALL_COMPLETE) { + netfs_read_subreq_progress(op->fetch.subreq); + if (rxrpc_kernel_check_life(call->net->socket, call->rxcall)) + return; + /* rxrpc terminated the call. */ + afs_set_call_complete(call, call->error, call->abort_code); + } + + op->call_abort_code = call->abort_code; + op->call_error = call->error; + op->call_responded = call->responded; + op->call = NULL; + call->op = NULL; + afs_put_call(call); + + /* If the call failed, then we need to crank the server rotation + * handle and try the next. + */ + if (afs_select_fileserver(op)) { + afs_issue_read_call(op); + return; + } + + afs_end_read(op); +} + +void afs_fetch_data_async_rx(struct work_struct *work) +{ + struct afs_call *call = container_of(work, struct afs_call, async_work); + + afs_read_receive(call); + afs_put_call(call); +} + +void afs_fetch_data_immediate_cancel(struct afs_call *call) +{ + if (call->async) { + afs_get_call(call, afs_call_trace_wake); + if (!queue_work(afs_async_calls, &call->async_work)) + afs_deferred_put_call(call); + flush_work(&call->async_work); + } +} + /* * Fetch file data from the volume. */ -int afs_fetch_data(struct afs_vnode *vnode, struct afs_read *req) +static void afs_issue_read(struct netfs_io_subrequest *subreq) { struct afs_operation *op; + struct afs_vnode *vnode = AFS_FS_I(subreq->rreq->inode); + struct key *key = subreq->rreq->netfs_priv; _enter("%s{%llx:%llu.%u},%x,,,", vnode->volume->name, vnode->fid.vid, vnode->fid.vnode, vnode->fid.unique, - key_serial(req->key)); + key_serial(key)); - op = afs_alloc_operation(req->key, vnode->volume); + op = afs_alloc_operation(key, vnode->volume); if (IS_ERR(op)) { - if (req->subreq) - netfs_read_subreq_terminated(req->subreq, PTR_ERR(op), false); - return PTR_ERR(op); + subreq->error = PTR_ERR(op); + netfs_read_subreq_terminated(subreq); + return; } afs_op_set_vnode(op, 0, vnode); - op->fetch.req = afs_get_read(req); + op->fetch.subreq = subreq; op->ops = &afs_fetch_data_operation; - return afs_do_sync_operation(op); -} - -static void afs_read_worker(struct work_struct *work) -{ - struct netfs_io_subrequest *subreq = container_of(work, struct netfs_io_subrequest, work); - struct afs_vnode *vnode = AFS_FS_I(subreq->rreq->inode); - struct afs_read *fsreq; - - fsreq = afs_alloc_read(GFP_NOFS); - if (!fsreq) - return netfs_read_subreq_terminated(subreq, -ENOMEM, false); - - fsreq->subreq = subreq; - fsreq->pos = subreq->start + subreq->transferred; - fsreq->len = subreq->len - subreq->transferred; - fsreq->key = key_get(subreq->rreq->netfs_priv); - fsreq->vnode = vnode; - fsreq->iter = &subreq->io_iter; trace_netfs_sreq(subreq, netfs_sreq_trace_submit); - afs_fetch_data(fsreq->vnode, fsreq); - afs_put_read(fsreq); -} - -static void afs_issue_read(struct netfs_io_subrequest *subreq) -{ - INIT_WORK(&subreq->work, afs_read_worker); - queue_work(system_long_wq, &subreq->work); -} -static int afs_symlink_read_folio(struct file *file, struct folio *folio) -{ - struct afs_vnode *vnode = AFS_FS_I(folio->mapping->host); - struct afs_read *fsreq; - int ret; + if (subreq->rreq->origin == NETFS_READAHEAD || + subreq->rreq->iocb) { + op->flags |= AFS_OPERATION_ASYNC; - fsreq = afs_alloc_read(GFP_NOFS); - if (!fsreq) - return -ENOMEM; + if (!afs_begin_vnode_operation(op)) { + subreq->error = afs_put_operation(op); + netfs_read_subreq_terminated(subreq); + return; + } - fsreq->pos = folio_pos(folio); - fsreq->len = folio_size(folio); - fsreq->vnode = vnode; - fsreq->iter = &fsreq->def_iter; - iov_iter_xarray(&fsreq->def_iter, ITER_DEST, &folio->mapping->i_pages, - fsreq->pos, fsreq->len); + if (!afs_select_fileserver(op)) { + afs_end_read(op); + return; + } - ret = afs_fetch_data(fsreq->vnode, fsreq); - if (ret == 0) - folio_mark_uptodate(folio); - folio_unlock(folio); - return ret; + afs_issue_read_call(op); + } else { + afs_do_sync_operation(op); + } } static int afs_init_request(struct netfs_io_request *rreq, struct file *file) { + struct afs_vnode *vnode = AFS_FS_I(rreq->inode); + if (file) rreq->netfs_priv = key_get(afs_file_key(file)); rreq->rsize = 256 * 1024; rreq->wsize = 256 * 1024 * 1024; + + switch (rreq->origin) { + case NETFS_READ_SINGLE: + if (!file) { + struct key *key = afs_request_key(vnode->volume->cell); + + if (IS_ERR(key)) + return PTR_ERR(key); + rreq->netfs_priv = key; + } + break; + case NETFS_WRITEBACK: + case NETFS_WRITETHROUGH: + case NETFS_UNBUFFERED_WRITE: + case NETFS_DIO_WRITE: + if (S_ISREG(rreq->inode->i_mode)) + rreq->io_streams[0].avail = true; + break; + case NETFS_WRITEBACK_SINGLE: + default: + break; + } return 0; } diff --git a/fs/afs/fs_operation.c b/fs/afs/fs_operation.c index 428721bbe4f6..8418813ee043 100644 --- a/fs/afs/fs_operation.c +++ b/fs/afs/fs_operation.c @@ -49,6 +49,105 @@ struct afs_operation *afs_alloc_operation(struct key *key, struct afs_volume *vo return op; } +struct afs_io_locker { + struct list_head link; + struct task_struct *task; + unsigned long have_lock; +}; + +/* + * Unlock the I/O lock on a vnode. + */ +static void afs_unlock_for_io(struct afs_vnode *vnode) +{ + struct afs_io_locker *locker; + + spin_lock(&vnode->lock); + locker = list_first_entry_or_null(&vnode->io_lock_waiters, + struct afs_io_locker, link); + if (locker) { + list_del(&locker->link); + smp_store_release(&locker->have_lock, 1); /* The unlock barrier. */ + smp_mb__after_atomic(); /* Store have_lock before task state */ + wake_up_process(locker->task); + } else { + clear_bit(AFS_VNODE_IO_LOCK, &vnode->flags); + } + spin_unlock(&vnode->lock); +} + +/* + * Lock the I/O lock on a vnode uninterruptibly. We can't use an ordinary + * mutex as lockdep will complain if we unlock it in the wrong thread. + */ +static void afs_lock_for_io(struct afs_vnode *vnode) +{ + struct afs_io_locker myself = { .task = current, }; + + spin_lock(&vnode->lock); + + if (!test_and_set_bit(AFS_VNODE_IO_LOCK, &vnode->flags)) { + spin_unlock(&vnode->lock); + return; + } + + list_add_tail(&myself.link, &vnode->io_lock_waiters); + spin_unlock(&vnode->lock); + + for (;;) { + set_current_state(TASK_UNINTERRUPTIBLE); + if (smp_load_acquire(&myself.have_lock)) /* The lock barrier */ + break; + schedule(); + } + __set_current_state(TASK_RUNNING); +} + +/* + * Lock the I/O lock on a vnode interruptibly. We can't use an ordinary mutex + * as lockdep will complain if we unlock it in the wrong thread. + */ +static int afs_lock_for_io_interruptible(struct afs_vnode *vnode) +{ + struct afs_io_locker myself = { .task = current, }; + int ret = 0; + + spin_lock(&vnode->lock); + + if (!test_and_set_bit(AFS_VNODE_IO_LOCK, &vnode->flags)) { + spin_unlock(&vnode->lock); + return 0; + } + + list_add_tail(&myself.link, &vnode->io_lock_waiters); + spin_unlock(&vnode->lock); + + for (;;) { + set_current_state(TASK_INTERRUPTIBLE); + if (smp_load_acquire(&myself.have_lock) || /* The lock barrier */ + signal_pending(current)) + break; + schedule(); + } + __set_current_state(TASK_RUNNING); + + /* If we got a signal, try to transfer the lock onto the next + * waiter. + */ + if (unlikely(signal_pending(current))) { + spin_lock(&vnode->lock); + if (myself.have_lock) { + spin_unlock(&vnode->lock); + afs_unlock_for_io(vnode); + } else { + list_del(&myself.link); + spin_unlock(&vnode->lock); + } + ret = -ERESTARTSYS; + } + return ret; +} + /* * Lock the vnode(s) being operated upon. */ @@ -60,7 +159,7 @@ static bool afs_get_io_locks(struct afs_operation *op) _enter(""); if (op->flags & AFS_OPERATION_UNINTR) { - mutex_lock(&vnode->io_lock); + afs_lock_for_io(vnode); op->flags |= AFS_OPERATION_LOCK_0; _leave(" = t [1]"); return true; @@ -72,7 +171,7 @@ static bool afs_get_io_locks(struct afs_operation *op) if (vnode2 > vnode) swap(vnode, vnode2); - if (mutex_lock_interruptible(&vnode->io_lock) < 0) { + if (afs_lock_for_io_interruptible(vnode) < 0) { afs_op_set_error(op, -ERESTARTSYS); op->flags |= AFS_OPERATION_STOP; _leave(" = f [I 0]"); @@ -81,10 +180,10 @@ static bool afs_get_io_locks(struct afs_operation *op) op->flags |= AFS_OPERATION_LOCK_0; if (vnode2) { - if (mutex_lock_interruptible_nested(&vnode2->io_lock, 1) < 0) { + if (afs_lock_for_io_interruptible(vnode2) < 0) { afs_op_set_error(op, -ERESTARTSYS); op->flags |= AFS_OPERATION_STOP; - mutex_unlock(&vnode->io_lock); + afs_unlock_for_io(vnode); op->flags &= ~AFS_OPERATION_LOCK_0; _leave(" = f [I 1]"); return false; @@ -104,9 +203,9 @@ static void afs_drop_io_locks(struct afs_operation *op) _enter(""); if (op->flags & AFS_OPERATION_LOCK_1) - mutex_unlock(&vnode2->io_lock); + afs_unlock_for_io(vnode2); if (op->flags & AFS_OPERATION_LOCK_0) - mutex_unlock(&vnode->io_lock); + afs_unlock_for_io(vnode); } static void afs_prepare_vnode(struct afs_operation *op, struct afs_vnode_param *vp, @@ -157,7 +256,7 @@ bool afs_begin_vnode_operation(struct afs_operation *op) /* * Tidy up a filesystem cursor and unlock the vnode. */ -static void afs_end_vnode_operation(struct afs_operation *op) +void afs_end_vnode_operation(struct afs_operation *op) { _enter(""); diff --git a/fs/afs/fsclient.c b/fs/afs/fsclient.c index 098fa034a1cc..1d9ecd5418d8 100644 --- a/fs/afs/fsclient.c +++ b/fs/afs/fsclient.c @@ -301,19 +301,19 @@ void afs_fs_fetch_status(struct afs_operation *op) static int afs_deliver_fs_fetch_data(struct afs_call *call) { struct afs_operation *op = call->op; + struct netfs_io_subrequest *subreq = op->fetch.subreq; struct afs_vnode_param *vp = &op->file[0]; - struct afs_read *req = op->fetch.req; const __be32 *bp; size_t count_before; int ret; _enter("{%u,%zu,%zu/%llu}", call->unmarshall, call->iov_len, iov_iter_count(call->iter), - req->actual_len); + call->remaining); switch (call->unmarshall) { case 0: - req->actual_len = 0; + call->remaining = 0; call->unmarshall++; if (call->operation_ID == FSFETCHDATA64) { afs_extract_to_tmp64(call); @@ -323,8 +323,8 @@ static int afs_deliver_fs_fetch_data(struct afs_call *call) } fallthrough; - /* Extract the returned data length into - * ->actual_len. This may indicate more or less data than was + /* Extract the returned data length into ->remaining. + * This may indicate more or less data than was * requested will be returned. */ case 1: @@ -333,42 +333,40 @@ static int afs_deliver_fs_fetch_data(struct afs_call *call) if (ret < 0) return ret; - req->actual_len = be64_to_cpu(call->tmp64); - _debug("DATA length: %llu", req->actual_len); + call->remaining = be64_to_cpu(call->tmp64); + _debug("DATA length: %llu", call->remaining); - if (req->actual_len == 0) + if (call->remaining == 0) goto no_more_data; - call->iter = req->iter; - call->iov_len = min(req->actual_len, req->len); + call->iter = &subreq->io_iter; + call->iov_len = umin(call->remaining, subreq->len - subreq->transferred); call->unmarshall++; fallthrough; /* extract the returned data */ case 2: count_before = call->iov_len; - _debug("extract data %zu/%llu", count_before, req->actual_len); + _debug("extract data %zu/%llu", count_before, call->remaining); ret = afs_extract_data(call, true); - if (req->subreq) { - req->subreq->transferred += count_before - call->iov_len; - netfs_read_subreq_progress(req->subreq, false); - } + subreq->transferred += count_before - call->iov_len; + call->remaining -= count_before - call->iov_len; if (ret < 0) return ret; call->iter = &call->def_iter; - if (req->actual_len <= req->len) + if (call->remaining) goto no_more_data; /* Discard any excess data the server gave us */ - afs_extract_discard(call, req->actual_len - req->len); + afs_extract_discard(call, call->remaining); call->unmarshall = 3; fallthrough; case 3: _debug("extract discard %zu/%llu", - iov_iter_count(call->iter), req->actual_len - req->len); + iov_iter_count(call->iter), call->remaining); ret = afs_extract_data(call, true); if (ret < 0) @@ -390,8 +388,8 @@ static int afs_deliver_fs_fetch_data(struct afs_call *call) xdr_decode_AFSCallBack(&bp, call, &vp->scb); xdr_decode_AFSVolSync(&bp, &op->volsync); - req->data_version = vp->scb.status.data_version; - req->file_size = vp->scb.status.size; + if (subreq->start + subreq->transferred >= vp->scb.status.size) + __set_bit(NETFS_SREQ_HIT_EOF, &subreq->flags); call->unmarshall++; fallthrough; @@ -410,14 +408,18 @@ static int afs_deliver_fs_fetch_data(struct afs_call *call) static const struct afs_call_type afs_RXFSFetchData = { .name = "FS.FetchData", .op = afs_FS_FetchData, + .async_rx = afs_fetch_data_async_rx, .deliver = afs_deliver_fs_fetch_data, + .immediate_cancel = afs_fetch_data_immediate_cancel, .destructor = afs_flat_call_destructor, }; static const struct afs_call_type afs_RXFSFetchData64 = { .name = "FS.FetchData64", .op = afs_FS_FetchData64, + .async_rx = afs_fetch_data_async_rx, .deliver = afs_deliver_fs_fetch_data, + .immediate_cancel = afs_fetch_data_immediate_cancel, .destructor = afs_flat_call_destructor, }; @@ -426,8 +428,8 @@ static const struct afs_call_type afs_RXFSFetchData64 = { */ static void afs_fs_fetch_data64(struct afs_operation *op) { + struct netfs_io_subrequest *subreq = op->fetch.subreq; struct afs_vnode_param *vp = &op->file[0]; - struct afs_read *req = op->fetch.req; struct afs_call *call; __be32 *bp; @@ -437,16 +439,19 @@ static void afs_fs_fetch_data64(struct afs_operation *op) if (!call) return afs_op_nomem(op); + if (op->flags & AFS_OPERATION_ASYNC) + call->async = true; + /* marshall the parameters */ bp = call->request; bp[0] = htonl(FSFETCHDATA64); bp[1] = htonl(vp->fid.vid); bp[2] = htonl(vp->fid.vnode); bp[3] = htonl(vp->fid.unique); - bp[4] = htonl(upper_32_bits(req->pos)); - bp[5] = htonl(lower_32_bits(req->pos)); + bp[4] = htonl(upper_32_bits(subreq->start + subreq->transferred)); + bp[5] = htonl(lower_32_bits(subreq->start + subreq->transferred)); bp[6] = 0; - bp[7] = htonl(lower_32_bits(req->len)); + bp[7] = htonl(lower_32_bits(subreq->len - subreq->transferred)); call->fid = vp->fid; trace_afs_make_fs_call(call, &vp->fid); @@ -458,9 +463,9 @@ static void afs_fs_fetch_data64(struct afs_operation *op) */ void afs_fs_fetch_data(struct afs_operation *op) { + struct netfs_io_subrequest *subreq = op->fetch.subreq; struct afs_vnode_param *vp = &op->file[0]; struct afs_call *call; - struct afs_read *req = op->fetch.req; __be32 *bp; if (test_bit(AFS_SERVER_FL_HAS_FS64, &op->server->flags)) @@ -472,16 +477,14 @@ void afs_fs_fetch_data(struct afs_operation *op) if (!call) return afs_op_nomem(op); - req->call_debug_id = call->debug_id; - /* marshall the parameters */ bp = call->request; bp[0] = htonl(FSFETCHDATA); bp[1] = htonl(vp->fid.vid); bp[2] = htonl(vp->fid.vnode); bp[3] = htonl(vp->fid.unique); - bp[4] = htonl(lower_32_bits(req->pos)); - bp[5] = htonl(lower_32_bits(req->len)); + bp[4] = htonl(lower_32_bits(subreq->start + subreq->transferred)); + bp[5] = htonl(lower_32_bits(subreq->len + subreq->transferred)); call->fid = vp->fid; trace_afs_make_fs_call(call, &vp->fid); @@ -1733,6 +1736,7 @@ static const struct afs_call_type afs_RXFSGetCapabilities = { .op = afs_FS_GetCapabilities, .deliver = afs_deliver_fs_get_capabilities, .done = afs_fileserver_probe_result, + .immediate_cancel = afs_fileserver_probe_result, .destructor = afs_fs_get_capabilities_destructor, }; diff --git a/fs/afs/inode.c b/fs/afs/inode.c index a95e77670b49..e9538e91f848 100644 --- a/fs/afs/inode.c +++ b/fs/afs/inode.c @@ -25,8 +25,94 @@ #include "internal.h" #include "afs_fs.h" +void afs_init_new_symlink(struct afs_vnode *vnode, struct afs_operation *op) +{ + size_t size = strlen(op->create.symlink) + 1; + size_t dsize = 0; + char *p; + + if (netfs_alloc_folioq_buffer(NULL, &vnode->directory, &dsize, size, + mapping_gfp_mask(vnode->netfs.inode.i_mapping)) < 0) + return; + + vnode->directory_size = dsize; + p = kmap_local_folio(folioq_folio(vnode->directory, 0), 0); + memcpy(p, op->create.symlink, size); + kunmap_local(p); + set_bit(AFS_VNODE_DIR_READ, &vnode->flags); + netfs_single_mark_inode_dirty(&vnode->netfs.inode); +} + +static void afs_put_link(void *arg) +{ + struct folio *folio = virt_to_folio(arg); + + kunmap_local(arg); + folio_put(folio); +} + +const char *afs_get_link(struct dentry *dentry, struct inode *inode, + struct delayed_call *callback) +{ + struct afs_vnode *vnode = AFS_FS_I(inode); + struct folio *folio; + char *content; + ssize_t ret; + + if (!dentry) { + /* RCU pathwalk. */ + if (!test_bit(AFS_VNODE_DIR_READ, &vnode->flags) || !afs_check_validity(vnode)) + return ERR_PTR(-ECHILD); + goto good; + } + + if (test_bit(AFS_VNODE_DIR_READ, &vnode->flags)) + goto fetch; + + ret = afs_validate(vnode, NULL); + if (ret < 0) + return ERR_PTR(ret); + + if (!test_and_clear_bit(AFS_VNODE_ZAP_DATA, &vnode->flags) && + test_bit(AFS_VNODE_DIR_READ, &vnode->flags)) + goto good; + +fetch: + ret = afs_read_single(vnode, NULL); + if (ret < 0) + return ERR_PTR(ret); + set_bit(AFS_VNODE_DIR_READ, &vnode->flags); + +good: + folio = folioq_folio(vnode->directory, 0); + folio_get(folio); + content = kmap_local_folio(folio, 0); + set_delayed_call(callback, afs_put_link, content); + return content; +} + +int afs_readlink(struct dentry *dentry, char __user *buffer, int buflen) +{ + DEFINE_DELAYED_CALL(done); + const char *content; + int len; + + content = afs_get_link(dentry, d_inode(dentry), &done); + if (IS_ERR(content)) { + do_delayed_call(&done); + return PTR_ERR(content); + } + + len = umin(strlen(content), buflen); + if (copy_to_user(buffer, content, len)) + len = -EFAULT; + do_delayed_call(&done); + return len; +} + static const struct inode_operations afs_symlink_inode_operations = { - .get_link = page_get_link, + .get_link = afs_get_link, + .readlink = afs_readlink, }; static noinline void dump_vnode(struct afs_vnode *vnode, struct afs_vnode *parent_vnode) @@ -110,7 +196,9 @@ static int afs_inode_init_from_status(struct afs_operation *op, inode->i_op = &afs_dir_inode_operations; inode->i_fop = &afs_dir_file_operations; inode->i_mapping->a_ops = &afs_dir_aops; - mapping_set_large_folios(inode->i_mapping); + __set_bit(NETFS_ICTX_SINGLE_NO_UPLOAD, &vnode->netfs.flags); + /* Assume locally cached directory data will be valid. */ + __set_bit(AFS_VNODE_DIR_VALID, &vnode->flags); break; case AFS_FTYPE_SYMLINK: /* Symlinks with a mode of 0644 are actually mountpoints. */ @@ -122,13 +210,13 @@ static int afs_inode_init_from_status(struct afs_operation *op, inode->i_mode = S_IFDIR | 0555; inode->i_op = &afs_mntpt_inode_operations; inode->i_fop = &afs_mntpt_file_operations; - inode->i_mapping->a_ops = &afs_symlink_aops; } else { inode->i_mode = S_IFLNK | status->mode; inode->i_op = &afs_symlink_inode_operations; - inode->i_mapping->a_ops = &afs_symlink_aops; } + inode->i_mapping->a_ops = &afs_dir_aops; inode_nohighmem(inode); + mapping_set_release_always(inode->i_mapping); break; default: dump_vnode(vnode, op->file[0].vnode != vnode ? op->file[0].vnode : NULL); @@ -140,15 +228,17 @@ static int afs_inode_init_from_status(struct afs_operation *op, afs_set_netfs_context(vnode); vnode->invalid_before = status->data_version; + trace_afs_set_dv(vnode, status->data_version); inode_set_iversion_raw(&vnode->netfs.inode, status->data_version); if (!vp->scb.have_cb) { /* it's a symlink we just created (the fileserver * didn't give us a callback) */ - atomic64_set(&vnode->cb_expires_at, AFS_NO_CB_PROMISE); + afs_clear_cb_promise(vnode, afs_cb_promise_set_new_symlink); } else { vnode->cb_server = op->server; - atomic64_set(&vnode->cb_expires_at, vp->scb.callback.expires_at); + afs_set_cb_promise(vnode, vp->scb.callback.expires_at, + afs_cb_promise_set_new_inode); } write_sequnlock(&vnode->cb_lock); @@ -207,12 +297,17 @@ static void afs_apply_status(struct afs_operation *op, if (vp->update_ctime) inode_set_ctime_to_ts(inode, op->ctime); - if (vnode->status.data_version != status->data_version) + if (vnode->status.data_version != status->data_version) { + trace_afs_set_dv(vnode, status->data_version); data_changed = true; + } vnode->status = *status; if (vp->dv_before + vp->dv_delta != status->data_version) { + trace_afs_dv_mismatch(vnode, vp->dv_before, vp->dv_delta, + status->data_version); + if (vnode->cb_ro_snapshot == atomic_read(&vnode->volume->cb_ro_snapshot) && atomic64_read(&vnode->cb_expires_at) != AFS_NO_CB_PROMISE) pr_warn("kAFS: vnode modified {%llx:%llu} %llx->%llx %s (op=%x)\n", @@ -223,12 +318,10 @@ static void afs_apply_status(struct afs_operation *op, op->debug_id); vnode->invalid_before = status->data_version; - if (vnode->status.type == AFS_FTYPE_DIR) { - if (test_and_clear_bit(AFS_VNODE_DIR_VALID, &vnode->flags)) - afs_stat_v(vnode, n_inval); - } else { + if (vnode->status.type == AFS_FTYPE_DIR) + afs_invalidate_dir(vnode, afs_dir_invalid_dv_mismatch); + else set_bit(AFS_VNODE_ZAP_DATA, &vnode->flags); - } change_size = true; data_changed = true; unexpected_jump = true; @@ -258,6 +351,8 @@ static void afs_apply_status(struct afs_operation *op, inode_set_ctime_to_ts(inode, t); inode_set_atime_to_ts(inode, t); } + if (op->ops == &afs_fetch_data_operation) + op->fetch.subreq->rreq->i_size = status->size; } } @@ -273,7 +368,7 @@ static void afs_apply_callback(struct afs_operation *op, if (!afs_cb_is_broken(vp->cb_break_before, vnode)) { if (op->volume->type == AFSVL_RWVOL) vnode->cb_server = op->server; - atomic64_set(&vnode->cb_expires_at, cb->expires_at); + afs_set_cb_promise(vnode, cb->expires_at, afs_cb_promise_set_apply_cb); } } @@ -435,7 +530,9 @@ static void afs_get_inode_cache(struct afs_vnode *vnode) } __packed key; struct afs_vnode_cache_aux aux; - if (vnode->status.type != AFS_FTYPE_FILE) { + if (vnode->status.type != AFS_FTYPE_FILE && + vnode->status.type != AFS_FTYPE_DIR && + vnode->status.type != AFS_FTYPE_SYMLINK) { vnode->netfs.cache = NULL; return; } @@ -637,6 +734,7 @@ int afs_drop_inode(struct inode *inode) void afs_evict_inode(struct inode *inode) { struct afs_vnode_cache_aux aux; + struct afs_super_info *sbi = AFS_FS_S(inode->i_sb); struct afs_vnode *vnode = AFS_FS_I(inode); _enter("{%llx:%llu.%d}", @@ -648,8 +746,22 @@ void afs_evict_inode(struct inode *inode) ASSERTCMP(inode->i_ino, ==, vnode->fid.vnode); + if ((S_ISDIR(inode->i_mode) || + S_ISLNK(inode->i_mode)) && + (inode->i_state & I_DIRTY) && + !sbi->dyn_root) { + struct writeback_control wbc = { + .sync_mode = WB_SYNC_ALL, + .for_sync = true, + .range_end = LLONG_MAX, + }; + + afs_single_writepages(inode->i_mapping, &wbc); + } + netfs_wait_for_outstanding_io(inode); truncate_inode_pages_final(&inode->i_data); + netfs_free_folioq_buffer(vnode->directory); afs_set_cache_aux(vnode, &aux); netfs_clear_inode_writeback(inode, &aux); diff --git a/fs/afs/internal.h b/fs/afs/internal.h index c9d620175e80..90f407774a9a 100644 --- a/fs/afs/internal.h +++ b/fs/afs/internal.h @@ -163,6 +163,7 @@ struct afs_call { spinlock_t state_lock; int error; /* error code */ u32 abort_code; /* Remote abort ID or 0 */ + unsigned long long remaining; /* How much is left to receive */ unsigned int max_lifespan; /* Maximum lifespan in secs to set if not 0 */ unsigned request_size; /* size of request data */ unsigned reply_max; /* maximum size of reply */ @@ -201,11 +202,17 @@ struct afs_call_type { /* clean up a call */ void (*destructor)(struct afs_call *call); + /* Async receive processing function */ + void (*async_rx)(struct work_struct *work); + /* Work function */ void (*work)(struct work_struct *work); /* Call done function (gets called immediately on success or failure) */ void (*done)(struct afs_call *call); + + /* Handle a call being immediately cancelled. */ + void (*immediate_cancel)(struct afs_call *call); }; /* @@ -233,28 +240,6 @@ static inline struct key *afs_file_key(struct file *file) } /* - * Record of an outstanding read operation on a vnode. - */ -struct afs_read { - loff_t pos; /* Where to start reading */ - loff_t len; /* How much we're asking for */ - loff_t actual_len; /* How much we're actually getting */ - loff_t file_size; /* File size returned by server */ - struct key *key; /* The key to use to reissue the read */ - struct afs_vnode *vnode; /* The file being read into. */ - struct netfs_io_subrequest *subreq; /* Fscache helper read request this belongs to */ - afs_dataversion_t data_version; /* Version number returned by server */ - refcount_t usage; - unsigned int call_debug_id; - unsigned int nr_pages; - int error; - void (*done)(struct afs_read *); - void (*cleanup)(struct afs_read *); - struct iov_iter *iter; /* Iterator representing the buffer */ - struct iov_iter def_iter; /* Default iterator */ -}; - -/* * AFS superblock private data * - there's one superblock per volume */ @@ -702,13 +687,14 @@ struct afs_vnode { struct afs_file_status status; /* AFS status info for this file */ afs_dataversion_t invalid_before; /* Child dentries are invalid before this */ struct afs_permits __rcu *permit_cache; /* cache of permits so far obtained */ - struct mutex io_lock; /* Lock for serialising I/O on this mutex */ + struct list_head io_lock_waiters; /* Threads waiting for the I/O lock */ struct rw_semaphore validate_lock; /* lock for validating this vnode */ struct rw_semaphore rmdir_lock; /* Lock for rmdir vs sillyrename */ struct key *silly_key; /* Silly rename key */ spinlock_t wb_lock; /* lock for wb_keys */ spinlock_t lock; /* waitqueue/flags lock */ unsigned long flags; +#define AFS_VNODE_IO_LOCK 0 /* Set if the I/O serialisation lock is held */ #define AFS_VNODE_UNSET 1 /* set if vnode attributes not yet set */ #define AFS_VNODE_DIR_VALID 2 /* Set if dir contents are valid */ #define AFS_VNODE_ZAP_DATA 3 /* set if vnode's data should be invalidated */ @@ -719,7 +705,9 @@ struct afs_vnode { #define AFS_VNODE_NEW_CONTENT 8 /* Set if file has new content (create/trunc-0) */ #define AFS_VNODE_SILLY_DELETED 9 /* Set if file has been silly-deleted */ #define AFS_VNODE_MODIFYING 10 /* Set if we're performing a modification op */ +#define AFS_VNODE_DIR_READ 11 /* Set if we've read a dir's contents */ + struct folio_queue *directory; /* Directory contents */ struct list_head wb_keys; /* List of keys available for writeback */ struct list_head pending_locks; /* locks waiting to be granted */ struct list_head granted_locks; /* locks granted on this file */ @@ -728,6 +716,7 @@ struct afs_vnode { ktime_t locked_at; /* Time at which lock obtained */ enum afs_lock_state lock_state : 8; afs_lock_type_t lock_type : 8; + unsigned int directory_size; /* Amount of space in ->directory */ /* outstanding callback notification on this file */ struct work_struct cb_work; /* Work for mmap'd files */ @@ -907,7 +896,7 @@ struct afs_operation { bool new_negative; } rename; struct { - struct afs_read *req; + struct netfs_io_subrequest *subreq; } fetch; struct { afs_lock_type_t type; @@ -959,6 +948,7 @@ struct afs_operation { #define AFS_OPERATION_TRIED_ALL 0x0400 /* Set if we've tried all the fileservers */ #define AFS_OPERATION_RETRY_SERVER 0x0800 /* Set if we should retry the current server */ #define AFS_OPERATION_DIR_CONFLICT 0x1000 /* Set if we detected a 3rd-party dir change */ +#define AFS_OPERATION_ASYNC 0x2000 /* Set if should run asynchronously */ }; /* @@ -983,6 +973,21 @@ static inline void afs_invalidate_cache(struct afs_vnode *vnode, unsigned int fl i_size_read(&vnode->netfs.inode), flags); } +/* + * Directory iteration management. + */ +struct afs_dir_iter { + struct afs_vnode *dvnode; + union afs_xdr_dir_block *block; + struct folio_queue *fq; + unsigned int fpos; + int fq_slot; + unsigned int loop_check; + u8 nr_slots; + u8 bucket; + unsigned int prev_entry; +}; + #include <trace/events/afs.h> /*****************************************************************************/ @@ -1064,8 +1069,13 @@ extern const struct inode_operations afs_dir_inode_operations; extern const struct address_space_operations afs_dir_aops; extern const struct dentry_operations afs_fs_dentry_operations; +ssize_t afs_read_single(struct afs_vnode *dvnode, struct file *file); +ssize_t afs_read_dir(struct afs_vnode *dvnode, struct file *file) + __acquires(&dvnode->validate_lock); extern void afs_d_release(struct dentry *); extern void afs_check_for_remote_deletion(struct afs_operation *); +int afs_single_writepages(struct address_space *mapping, + struct writeback_control *wbc); /* * dir_edit.c @@ -1075,6 +1085,18 @@ extern void afs_edit_dir_add(struct afs_vnode *, struct qstr *, struct afs_fid * extern void afs_edit_dir_remove(struct afs_vnode *, struct qstr *, enum afs_edit_dir_reason); void afs_edit_dir_update_dotdot(struct afs_vnode *vnode, struct afs_vnode *new_dvnode, enum afs_edit_dir_reason why); +void afs_mkdir_init_dir(struct afs_vnode *dvnode, struct afs_vnode *parent_vnode); + +/* + * dir_search.c + */ +unsigned int afs_dir_hash_name(const struct qstr *name); +bool afs_dir_init_iter(struct afs_dir_iter *iter, const struct qstr *name); +union afs_xdr_dir_block *afs_dir_find_block(struct afs_dir_iter *iter, size_t block); +int afs_dir_search_bucket(struct afs_dir_iter *iter, const struct qstr *name, + struct afs_fid *_fid); +int afs_dir_search(struct afs_vnode *dvnode, struct qstr *name, + struct afs_fid *_fid, afs_dataversion_t *_dir_version); /* * dir_silly.c @@ -1099,24 +1121,17 @@ extern void afs_dynroot_depopulate(struct super_block *); * file.c */ extern const struct address_space_operations afs_file_aops; -extern const struct address_space_operations afs_symlink_aops; extern const struct inode_operations afs_file_inode_operations; extern const struct file_operations afs_file_operations; +extern const struct afs_operation_ops afs_fetch_data_operation; extern const struct netfs_request_ops afs_req_ops; extern int afs_cache_wb_key(struct afs_vnode *, struct afs_file *); extern void afs_put_wb_key(struct afs_wb_key *); extern int afs_open(struct inode *, struct file *); extern int afs_release(struct inode *, struct file *); -extern int afs_fetch_data(struct afs_vnode *, struct afs_read *); -extern struct afs_read *afs_alloc_read(gfp_t); -extern void afs_put_read(struct afs_read *); - -static inline struct afs_read *afs_get_read(struct afs_read *req) -{ - refcount_inc(&req->usage); - return req; -} +void afs_fetch_data_async_rx(struct work_struct *work); +void afs_fetch_data_immediate_cancel(struct afs_call *call); /* * flock.c @@ -1168,6 +1183,7 @@ extern void afs_fs_store_acl(struct afs_operation *); extern struct afs_operation *afs_alloc_operation(struct key *, struct afs_volume *); extern int afs_put_operation(struct afs_operation *); extern bool afs_begin_vnode_operation(struct afs_operation *); +extern void afs_end_vnode_operation(struct afs_operation *op); extern void afs_wait_for_operation(struct afs_operation *); extern int afs_do_sync_operation(struct afs_operation *); @@ -1205,6 +1221,10 @@ extern void afs_fs_probe_cleanup(struct afs_net *); */ extern const struct afs_operation_ops afs_fetch_status_operation; +void afs_init_new_symlink(struct afs_vnode *vnode, struct afs_operation *op); +const char *afs_get_link(struct dentry *dentry, struct inode *inode, + struct delayed_call *callback); +int afs_readlink(struct dentry *dentry, char __user *buffer, int buflen); extern void afs_vnode_commit_status(struct afs_operation *, struct afs_vnode_param *); extern int afs_fetch_status(struct afs_vnode *, struct key *, bool, afs_access_t *); extern int afs_ilookup5_test_by_fid(struct inode *, void *); @@ -1336,6 +1356,7 @@ extern void afs_charge_preallocation(struct work_struct *); extern void afs_put_call(struct afs_call *); void afs_deferred_put_call(struct afs_call *call); void afs_make_call(struct afs_call *call, gfp_t gfp); +void afs_deliver_to_call(struct afs_call *call); void afs_wait_for_call_to_complete(struct afs_call *call); extern struct afs_call *afs_alloc_flat_call(struct afs_net *, const struct afs_call_type *, @@ -1346,6 +1367,28 @@ extern void afs_send_simple_reply(struct afs_call *, const void *, size_t); extern int afs_extract_data(struct afs_call *, bool); extern int afs_protocol_error(struct afs_call *, enum afs_eproto_cause); +static inline struct afs_call *afs_get_call(struct afs_call *call, + enum afs_call_trace why) +{ + int r; + + __refcount_inc(&call->ref, &r); + + trace_afs_call(call->debug_id, why, r + 1, + atomic_read(&call->net->nr_outstanding_calls), + __builtin_return_address(0)); + return call; +} + +static inline void afs_see_call(struct afs_call *call, enum afs_call_trace why) +{ + int r = refcount_read(&call->ref); + + trace_afs_call(call->debug_id, why, r, + atomic_read(&call->net->nr_outstanding_calls), + __builtin_return_address(0)); +} + static inline void afs_make_op_call(struct afs_operation *op, struct afs_call *call, gfp_t gfp) { @@ -1712,6 +1755,38 @@ static inline int afs_bad(struct afs_vnode *vnode, enum afs_file_error where) return -EIO; } +/* + * Set the callback promise on a vnode. + */ +static inline void afs_set_cb_promise(struct afs_vnode *vnode, time64_t expires_at, + enum afs_cb_promise_trace trace) +{ + atomic64_set(&vnode->cb_expires_at, expires_at); + trace_afs_cb_promise(vnode, trace); +} + +/* + * Clear the callback promise on a vnode, returning true if it was promised. + */ +static inline bool afs_clear_cb_promise(struct afs_vnode *vnode, + enum afs_cb_promise_trace trace) +{ + trace_afs_cb_promise(vnode, trace); + return atomic64_xchg(&vnode->cb_expires_at, AFS_NO_CB_PROMISE) != AFS_NO_CB_PROMISE; +} + +/* + * Mark a directory as being invalid. + */ +static inline void afs_invalidate_dir(struct afs_vnode *dvnode, + enum afs_dir_invalid_trace trace) +{ + if (test_and_clear_bit(AFS_VNODE_DIR_VALID, &dvnode->flags)) { + trace_afs_dir_invalid(dvnode, trace); + afs_stat_v(dvnode, n_inval); + } +} + /*****************************************************************************/ /* * debug tracing diff --git a/fs/afs/main.c b/fs/afs/main.c index a14f6013e316..1ae0067f772d 100644 --- a/fs/afs/main.c +++ b/fs/afs/main.c @@ -177,7 +177,7 @@ static int __init afs_init(void) afs_wq = alloc_workqueue("afs", 0, 0); if (!afs_wq) goto error_afs_wq; - afs_async_calls = alloc_workqueue("kafsd", WQ_MEM_RECLAIM, 0); + afs_async_calls = alloc_workqueue("kafsd", WQ_MEM_RECLAIM | WQ_UNBOUND, 0); if (!afs_async_calls) goto error_async; afs_lock_manager = alloc_workqueue("kafs_lockd", WQ_MEM_RECLAIM, 0); diff --git a/fs/afs/mntpt.c b/fs/afs/mntpt.c index 297487ee8323..507c25a5b2cb 100644 --- a/fs/afs/mntpt.c +++ b/fs/afs/mntpt.c @@ -30,7 +30,7 @@ const struct file_operations afs_mntpt_file_operations = { const struct inode_operations afs_mntpt_inode_operations = { .lookup = afs_mntpt_lookup, - .readlink = page_readlink, + .readlink = afs_readlink, .getattr = afs_getattr, }; @@ -118,9 +118,9 @@ static int afs_mntpt_set_params(struct fs_context *fc, struct dentry *mntpt) ctx->volnamesz = sizeof(afs_root_volume) - 1; } else { /* read the contents of the AFS special symlink */ - struct page *page; + DEFINE_DELAYED_CALL(cleanup); + const char *content; loff_t size = i_size_read(d_inode(mntpt)); - char *buf; if (src_as->cell) ctx->cell = afs_use_cell(src_as->cell, afs_cell_trace_use_mntpt); @@ -128,16 +128,16 @@ static int afs_mntpt_set_params(struct fs_context *fc, struct dentry *mntpt) if (size < 2 || size > PAGE_SIZE - 1) return -EINVAL; - page = read_mapping_page(d_inode(mntpt)->i_mapping, 0, NULL); - if (IS_ERR(page)) - return PTR_ERR(page); + content = afs_get_link(mntpt, d_inode(mntpt), &cleanup); + if (IS_ERR(content)) { + do_delayed_call(&cleanup); + return PTR_ERR(content); + } - buf = kmap(page); ret = -EINVAL; - if (buf[size - 1] == '.') - ret = vfs_parse_fs_string(fc, "source", buf, size - 1); - kunmap(page); - put_page(page); + if (content[size - 1] == '.') + ret = vfs_parse_fs_string(fc, "source", content, size - 1); + do_delayed_call(&cleanup); if (ret < 0) return ret; diff --git a/fs/afs/proc.c b/fs/afs/proc.c index 15eab053af6d..e7614f4f30c2 100644 --- a/fs/afs/proc.c +++ b/fs/afs/proc.c @@ -240,7 +240,13 @@ static int afs_proc_rootcell_write(struct file *file, char *buf, size_t size) /* determine command to perform */ _debug("rootcell=%s", buf); - ret = afs_cell_init(net, buf); + ret = -EEXIST; + inode_lock(file_inode(file)); + if (!net->ws_cell) + ret = afs_cell_init(net, buf); + else + printk("busy\n"); + inode_unlock(file_inode(file)); out: _leave(" = %d", ret); diff --git a/fs/afs/rotate.c b/fs/afs/rotate.c index d612983d6f38..a1c24f589d9e 100644 --- a/fs/afs/rotate.c +++ b/fs/afs/rotate.c @@ -99,7 +99,7 @@ static bool afs_start_fs_iteration(struct afs_operation *op, write_seqlock(&vnode->cb_lock); ASSERTCMP(cb_server, ==, vnode->cb_server); vnode->cb_server = NULL; - if (atomic64_xchg(&vnode->cb_expires_at, AFS_NO_CB_PROMISE) != AFS_NO_CB_PROMISE) + if (afs_clear_cb_promise(vnode, afs_cb_promise_clear_rotate_server)) vnode->cb_break++; write_sequnlock(&vnode->cb_lock); } @@ -583,7 +583,7 @@ selected_server: if (vnode->cb_server != server) { vnode->cb_server = server; vnode->cb_v_check = atomic_read(&vnode->volume->cb_v_break); - atomic64_set(&vnode->cb_expires_at, AFS_NO_CB_PROMISE); + afs_clear_cb_promise(vnode, afs_cb_promise_clear_server_change); } retry_server: diff --git a/fs/afs/rxrpc.c b/fs/afs/rxrpc.c index 9f2a3bb56ec6..886416ea1d96 100644 --- a/fs/afs/rxrpc.c +++ b/fs/afs/rxrpc.c @@ -149,7 +149,8 @@ static struct afs_call *afs_alloc_call(struct afs_net *net, call->net = net; call->debug_id = atomic_inc_return(&rxrpc_debug_id); refcount_set(&call->ref, 1); - INIT_WORK(&call->async_work, afs_process_async_call); + INIT_WORK(&call->async_work, type->async_rx ?: afs_process_async_call); + INIT_WORK(&call->work, call->type->work); INIT_WORK(&call->free_work, afs_deferred_free_worker); init_waitqueue_head(&call->waitq); spin_lock_init(&call->state_lock); @@ -235,27 +236,12 @@ void afs_deferred_put_call(struct afs_call *call) schedule_work(&call->free_work); } -static struct afs_call *afs_get_call(struct afs_call *call, - enum afs_call_trace why) -{ - int r; - - __refcount_inc(&call->ref, &r); - - trace_afs_call(call->debug_id, why, r + 1, - atomic_read(&call->net->nr_outstanding_calls), - __builtin_return_address(0)); - return call; -} - /* * Queue the call for actual work. */ static void afs_queue_call_work(struct afs_call *call) { if (call->type->work) { - INIT_WORK(&call->work, call->type->work); - afs_get_call(call, afs_call_trace_work); if (!queue_work(afs_wq, &call->work)) afs_put_call(call); @@ -430,11 +416,16 @@ void afs_make_call(struct afs_call *call, gfp_t gfp) return; error_do_abort: - if (ret != -ECONNABORTED) { + if (ret != -ECONNABORTED) rxrpc_kernel_abort_call(call->net->socket, rxcall, RX_USER_ABORT, ret, afs_abort_send_data_error); - } else { + if (call->async) { + afs_see_call(call, afs_call_trace_async_abort); + return; + } + + if (ret == -ECONNABORTED) { len = 0; iov_iter_kvec(&msg.msg_iter, ITER_DEST, NULL, 0, 0); rxrpc_kernel_recv_data(call->net->socket, rxcall, @@ -445,8 +436,10 @@ error_do_abort: call->error = ret; trace_afs_call_done(call); error_kill_call: - if (call->type->done) - call->type->done(call); + if (call->async) + afs_see_call(call, afs_call_trace_async_kill); + if (call->type->immediate_cancel) + call->type->immediate_cancel(call); /* We need to dispose of the extra ref we grabbed for an async call. * The call, however, might be queued on afs_async_calls and we need to @@ -501,7 +494,7 @@ static void afs_log_error(struct afs_call *call, s32 remote_abort) /* * deliver messages to a call */ -static void afs_deliver_to_call(struct afs_call *call) +void afs_deliver_to_call(struct afs_call *call) { enum afs_call_state state; size_t len; @@ -602,7 +595,6 @@ local_abort: abort_code = 0; call_complete: afs_set_call_complete(call, ret, remote_abort); - state = AFS_CALL_COMPLETE; goto done; } @@ -803,6 +795,7 @@ static int afs_deliver_cm_op_id(struct afs_call *call) return -ENOTSUPP; trace_afs_cb_call(call); + call->work.func = call->type->work; /* pass responsibility for the remainer of this message off to the * cache manager op */ diff --git a/fs/afs/super.c b/fs/afs/super.c index f3ba1c3e72f5..a9bee610674e 100644 --- a/fs/afs/super.c +++ b/fs/afs/super.c @@ -663,7 +663,7 @@ static void afs_i_init_once(void *_vnode) memset(vnode, 0, sizeof(*vnode)); inode_init_once(&vnode->netfs.inode); - mutex_init(&vnode->io_lock); + INIT_LIST_HEAD(&vnode->io_lock_waiters); init_rwsem(&vnode->validate_lock); spin_lock_init(&vnode->wb_lock); spin_lock_init(&vnode->lock); @@ -696,6 +696,8 @@ static struct inode *afs_alloc_inode(struct super_block *sb) vnode->volume = NULL; vnode->lock_key = NULL; vnode->permit_cache = NULL; + vnode->directory = NULL; + vnode->directory_size = 0; vnode->flags = 1 << AFS_VNODE_UNSET; vnode->lock_state = AFS_VNODE_LOCK_NONE; diff --git a/fs/afs/validation.c b/fs/afs/validation.c index bef8af12ebe2..0ba8336c9025 100644 --- a/fs/afs/validation.c +++ b/fs/afs/validation.c @@ -120,22 +120,31 @@ bool afs_check_validity(const struct afs_vnode *vnode) { const struct afs_volume *volume = vnode->volume; + enum afs_vnode_invalid_trace trace = afs_vnode_valid_trace; + time64_t cb_expires_at = atomic64_read(&vnode->cb_expires_at); time64_t deadline = ktime_get_real_seconds() + 10; if (test_bit(AFS_VNODE_DELETED, &vnode->flags)) return true; - if (atomic_read(&volume->cb_v_check) != atomic_read(&volume->cb_v_break) || - atomic64_read(&vnode->cb_expires_at) <= deadline || - volume->cb_expires_at <= deadline || - vnode->cb_ro_snapshot != atomic_read(&volume->cb_ro_snapshot) || - vnode->cb_scrub != atomic_read(&volume->cb_scrub) || - test_bit(AFS_VNODE_ZAP_DATA, &vnode->flags)) { - _debug("inval"); - return false; - } - - return true; + if (atomic_read(&volume->cb_v_check) != atomic_read(&volume->cb_v_break)) + trace = afs_vnode_invalid_trace_cb_v_break; + else if (cb_expires_at == AFS_NO_CB_PROMISE) + trace = afs_vnode_invalid_trace_no_cb_promise; + else if (cb_expires_at <= deadline) + trace = afs_vnode_invalid_trace_expired; + else if (volume->cb_expires_at <= deadline) + trace = afs_vnode_invalid_trace_vol_expired; + else if (vnode->cb_ro_snapshot != atomic_read(&volume->cb_ro_snapshot)) + trace = afs_vnode_invalid_trace_cb_ro_snapshot; + else if (vnode->cb_scrub != atomic_read(&volume->cb_scrub)) + trace = afs_vnode_invalid_trace_cb_scrub; + else if (test_bit(AFS_VNODE_ZAP_DATA, &vnode->flags)) + trace = afs_vnode_invalid_trace_zap_data; + else + return true; + trace_afs_vnode_invalid(vnode, trace); + return false; } /* diff --git a/fs/afs/vl_alias.c b/fs/afs/vl_alias.c index 9f36e14f1c2d..f9e76b604f31 100644 --- a/fs/afs/vl_alias.c +++ b/fs/afs/vl_alias.c @@ -253,6 +253,7 @@ static char *afs_vl_get_cell_name(struct afs_cell *cell, struct key *key) static int yfs_check_canonical_cell_name(struct afs_cell *cell, struct key *key) { struct afs_cell *master; + size_t name_len; char *cell_name; cell_name = afs_vl_get_cell_name(cell, key); @@ -264,8 +265,11 @@ static int yfs_check_canonical_cell_name(struct afs_cell *cell, struct key *key) return 0; } - master = afs_lookup_cell(cell->net, cell_name, strlen(cell_name), - NULL, false); + name_len = strlen(cell_name); + if (!name_len || name_len > AFS_MAXCELLNAME) + master = ERR_PTR(-EOPNOTSUPP); + else + master = afs_lookup_cell(cell->net, cell_name, name_len, NULL, false); kfree(cell_name); if (IS_ERR(master)) return PTR_ERR(master); diff --git a/fs/afs/vlclient.c b/fs/afs/vlclient.c index cac75f89b64a..3a23c0b08eb6 100644 --- a/fs/afs/vlclient.c +++ b/fs/afs/vlclient.c @@ -370,6 +370,7 @@ static const struct afs_call_type afs_RXVLGetCapabilities = { .name = "VL.GetCapabilities", .op = afs_VL_GetCapabilities, .deliver = afs_deliver_vl_get_capabilities, + .immediate_cancel = afs_vlserver_probe_result, .done = afs_vlserver_probe_result, .destructor = afs_destroy_vl_get_capabilities, }; @@ -697,7 +698,7 @@ static int afs_deliver_yfsvl_get_cell_name(struct afs_call *call) return ret; namesz = ntohl(call->tmp); - if (namesz > AFS_MAXCELLNAME) + if (namesz > YFS_VL_MAXCELLNAME) return afs_protocol_error(call, afs_eproto_cellname_len); paddedsz = (namesz + 3) & ~3; call->count = namesz; diff --git a/fs/afs/write.c b/fs/afs/write.c index 34107b55f834..18b0a9f1615e 100644 --- a/fs/afs/write.c +++ b/fs/afs/write.c @@ -122,7 +122,7 @@ static void afs_issue_write_worker(struct work_struct *work) if (subreq->debug_index == 3) return netfs_write_subrequest_terminated(subreq, -ENOANO, false); - if (!test_bit(NETFS_SREQ_RETRYING, &subreq->flags)) { + if (!subreq->retry_count) { set_bit(NETFS_SREQ_NEED_RETRY, &subreq->flags); return netfs_write_subrequest_terminated(subreq, -EAGAIN, false); } @@ -149,6 +149,9 @@ static void afs_issue_write_worker(struct work_struct *work) afs_wait_for_operation(op); ret = afs_put_operation(op); switch (ret) { + case 0: + __set_bit(NETFS_SREQ_MADE_PROGRESS, &subreq->flags); + break; case -EACCES: case -EPERM: case -ENOKEY: @@ -179,8 +182,8 @@ void afs_issue_write(struct netfs_io_subrequest *subreq) */ void afs_begin_writeback(struct netfs_io_request *wreq) { - afs_get_writeback_key(wreq); - wreq->io_streams[0].avail = true; + if (S_ISREG(wreq->inode->i_mode)) + afs_get_writeback_key(wreq); } /* @@ -193,6 +196,18 @@ void afs_retry_request(struct netfs_io_request *wreq, struct netfs_io_stream *st list_first_entry(&stream->subrequests, struct netfs_io_subrequest, rreq_link); + switch (wreq->origin) { + case NETFS_READAHEAD: + case NETFS_READPAGE: + case NETFS_READ_GAPS: + case NETFS_READ_SINGLE: + case NETFS_READ_FOR_WRITE: + case NETFS_DIO_READ: + return; + default: + break; + } + switch (subreq->error) { case -EACCES: case -EPERM: diff --git a/fs/afs/xdr_fs.h b/fs/afs/xdr_fs.h index 8ca868164507..cc5f143d21a3 100644 --- a/fs/afs/xdr_fs.h +++ b/fs/afs/xdr_fs.h @@ -88,7 +88,7 @@ union afs_xdr_dir_block { struct { struct afs_xdr_dir_hdr hdr; - u8 alloc_ctrs[AFS_DIR_MAX_BLOCKS]; + u8 alloc_ctrs[AFS_DIR_BLOCKS_WITH_CTR]; __be16 hashtable[AFS_DIR_HASHTBL_SIZE]; } meta; diff --git a/fs/afs/yfsclient.c b/fs/afs/yfsclient.c index 024227aba4cd..257af259c04a 100644 --- a/fs/afs/yfsclient.c +++ b/fs/afs/yfsclient.c @@ -352,19 +352,19 @@ static int yfs_deliver_status_and_volsync(struct afs_call *call) static int yfs_deliver_fs_fetch_data64(struct afs_call *call) { struct afs_operation *op = call->op; + struct netfs_io_subrequest *subreq = op->fetch.subreq; struct afs_vnode_param *vp = &op->file[0]; - struct afs_read *req = op->fetch.req; const __be32 *bp; size_t count_before; int ret; _enter("{%u,%zu, %zu/%llu}", call->unmarshall, call->iov_len, iov_iter_count(call->iter), - req->actual_len); + call->remaining); switch (call->unmarshall) { case 0: - req->actual_len = 0; + call->remaining = 0; afs_extract_to_tmp64(call); call->unmarshall++; fallthrough; @@ -379,42 +379,39 @@ static int yfs_deliver_fs_fetch_data64(struct afs_call *call) if (ret < 0) return ret; - req->actual_len = be64_to_cpu(call->tmp64); - _debug("DATA length: %llu", req->actual_len); + call->remaining = be64_to_cpu(call->tmp64); + _debug("DATA length: %llu", call->remaining); - if (req->actual_len == 0) + if (call->remaining == 0) goto no_more_data; - call->iter = req->iter; - call->iov_len = min(req->actual_len, req->len); + call->iter = &subreq->io_iter; + call->iov_len = min(call->remaining, subreq->len - subreq->transferred); call->unmarshall++; fallthrough; /* extract the returned data */ case 2: count_before = call->iov_len; - _debug("extract data %zu/%llu", count_before, req->actual_len); + _debug("extract data %zu/%llu", count_before, call->remaining); ret = afs_extract_data(call, true); - if (req->subreq) { - req->subreq->transferred += count_before - call->iov_len; - netfs_read_subreq_progress(req->subreq, false); - } + subreq->transferred += count_before - call->iov_len; if (ret < 0) return ret; call->iter = &call->def_iter; - if (req->actual_len <= req->len) + if (call->remaining) goto no_more_data; /* Discard any excess data the server gave us */ - afs_extract_discard(call, req->actual_len - req->len); + afs_extract_discard(call, call->remaining); call->unmarshall = 3; fallthrough; case 3: _debug("extract discard %zu/%llu", - iov_iter_count(call->iter), req->actual_len - req->len); + iov_iter_count(call->iter), call->remaining); ret = afs_extract_data(call, true); if (ret < 0) @@ -439,8 +436,8 @@ static int yfs_deliver_fs_fetch_data64(struct afs_call *call) xdr_decode_YFSCallBack(&bp, call, &vp->scb); xdr_decode_YFSVolSync(&bp, &op->volsync); - req->data_version = vp->scb.status.data_version; - req->file_size = vp->scb.status.size; + if (subreq->start + subreq->transferred >= vp->scb.status.size) + __set_bit(NETFS_SREQ_HIT_EOF, &subreq->flags); call->unmarshall++; fallthrough; @@ -459,7 +456,9 @@ static int yfs_deliver_fs_fetch_data64(struct afs_call *call) static const struct afs_call_type yfs_RXYFSFetchData64 = { .name = "YFS.FetchData64", .op = yfs_FS_FetchData64, + .async_rx = afs_fetch_data_async_rx, .deliver = yfs_deliver_fs_fetch_data64, + .immediate_cancel = afs_fetch_data_immediate_cancel, .destructor = afs_flat_call_destructor, }; @@ -468,14 +467,15 @@ static const struct afs_call_type yfs_RXYFSFetchData64 = { */ void yfs_fs_fetch_data(struct afs_operation *op) { + struct netfs_io_subrequest *subreq = op->fetch.subreq; struct afs_vnode_param *vp = &op->file[0]; - struct afs_read *req = op->fetch.req; struct afs_call *call; __be32 *bp; - _enter(",%x,{%llx:%llu},%llx,%llx", + _enter(",%x,{%llx:%llu},%llx,%zx", key_serial(op->key), vp->fid.vid, vp->fid.vnode, - req->pos, req->len); + subreq->start + subreq->transferred, + subreq->len - subreq->transferred); call = afs_alloc_flat_call(op->net, &yfs_RXYFSFetchData64, sizeof(__be32) * 2 + @@ -487,15 +487,16 @@ void yfs_fs_fetch_data(struct afs_operation *op) if (!call) return afs_op_nomem(op); - req->call_debug_id = call->debug_id; + if (op->flags & AFS_OPERATION_ASYNC) + call->async = true; /* marshall the parameters */ bp = call->request; bp = xdr_encode_u32(bp, YFSFETCHDATA64); bp = xdr_encode_u32(bp, 0); /* RPC flags */ bp = xdr_encode_YFSFid(bp, &vp->fid); - bp = xdr_encode_u64(bp, req->pos); - bp = xdr_encode_u64(bp, req->len); + bp = xdr_encode_u64(bp, subreq->start + subreq->transferred); + bp = xdr_encode_u64(bp, subreq->len - subreq->transferred); yfs_check_req(call, bp); call->fid = vp->fid; @@ -666,8 +667,9 @@ static int yfs_deliver_fs_remove_file2(struct afs_call *call) static void yfs_done_fs_remove_file2(struct afs_call *call) { if (call->error == -ECONNABORTED && - call->abort_code == RX_INVALID_OPERATION) { - set_bit(AFS_SERVER_FL_NO_RM2, &call->server->flags); + (call->abort_code == RX_INVALID_OPERATION || + call->abort_code == RXGEN_OPCODE)) { + set_bit(AFS_SERVER_FL_NO_RM2, &call->op->server->flags); call->op->flags |= AFS_OPERATION_DOWNGRADE; } } diff --git a/fs/backing-file.c b/fs/backing-file.c index cbdad8b68474..763fbe9b72b2 100644 --- a/fs/backing-file.c +++ b/fs/backing-file.c @@ -176,7 +176,7 @@ ssize_t backing_file_read_iter(struct file *file, struct iov_iter *iter, !(file->f_mode & FMODE_CAN_ODIRECT)) return -EINVAL; - old_cred = override_creds_light(ctx->cred); + old_cred = override_creds(ctx->cred); if (is_sync_kiocb(iocb)) { rwf_t rwf = iocb_to_rw_flags(flags); @@ -197,7 +197,7 @@ ssize_t backing_file_read_iter(struct file *file, struct iov_iter *iter, backing_aio_cleanup(aio, ret); } out: - revert_creds_light(old_cred); + revert_creds(old_cred); if (ctx->accessed) ctx->accessed(iocb->ki_filp); @@ -233,7 +233,7 @@ ssize_t backing_file_write_iter(struct file *file, struct iov_iter *iter, */ flags &= ~IOCB_DIO_CALLER_COMP; - old_cred = override_creds_light(ctx->cred); + old_cred = override_creds(ctx->cred); if (is_sync_kiocb(iocb)) { rwf_t rwf = iocb_to_rw_flags(flags); @@ -264,7 +264,7 @@ ssize_t backing_file_write_iter(struct file *file, struct iov_iter *iter, backing_aio_cleanup(aio, ret); } out: - revert_creds_light(old_cred); + revert_creds(old_cred); return ret; } @@ -281,9 +281,9 @@ ssize_t backing_file_splice_read(struct file *in, struct kiocb *iocb, if (WARN_ON_ONCE(!(in->f_mode & FMODE_BACKING))) return -EIO; - old_cred = override_creds_light(ctx->cred); + old_cred = override_creds(ctx->cred); ret = vfs_splice_read(in, &iocb->ki_pos, pipe, len, flags); - revert_creds_light(old_cred); + revert_creds(old_cred); if (ctx->accessed) ctx->accessed(iocb->ki_filp); @@ -310,11 +310,11 @@ ssize_t backing_file_splice_write(struct pipe_inode_info *pipe, if (ret) return ret; - old_cred = override_creds_light(ctx->cred); + old_cred = override_creds(ctx->cred); file_start_write(out); ret = out->f_op->splice_write(pipe, out, &iocb->ki_pos, len, flags); file_end_write(out); - revert_creds_light(old_cred); + revert_creds(old_cred); if (ctx->end_write) ctx->end_write(iocb, ret); @@ -338,9 +338,9 @@ int backing_file_mmap(struct file *file, struct vm_area_struct *vma, vma_set_file(vma, file); - old_cred = override_creds_light(ctx->cred); + old_cred = override_creds(ctx->cred); ret = call_mmap(vma->vm_file, vma); - revert_creds_light(old_cred); + revert_creds(old_cred); if (ctx->accessed) ctx->accessed(user_file); diff --git a/fs/bcachefs/Kconfig b/fs/bcachefs/Kconfig index ab6c95b895b3..464b927e4fff 100644 --- a/fs/bcachefs/Kconfig +++ b/fs/bcachefs/Kconfig @@ -90,7 +90,7 @@ config BCACHEFS_SIX_OPTIMISTIC_SPIN config BCACHEFS_PATH_TRACEPOINTS bool "Extra btree_path tracepoints" - depends on BCACHEFS_FS + depends on BCACHEFS_FS && TRACING help Enable extra tracepoints for debugging btree_path operations; we don't normally want these enabled because they happen at very high rates. diff --git a/fs/bcachefs/Makefile b/fs/bcachefs/Makefile index 56d20e219f59..d2689388d5e8 100644 --- a/fs/bcachefs/Makefile +++ b/fs/bcachefs/Makefile @@ -82,6 +82,7 @@ bcachefs-y := \ siphash.o \ six.o \ snapshot.o \ + str_hash.o \ subvolume.o \ super.o \ super-io.o \ diff --git a/fs/bcachefs/acl.c b/fs/bcachefs/acl.c index 87f1be9d4db4..99487727ae64 100644 --- a/fs/bcachefs/acl.c +++ b/fs/bcachefs/acl.c @@ -184,11 +184,6 @@ invalid: return ERR_PTR(-EINVAL); } -#define acl_for_each_entry(acl, acl_e) \ - for (acl_e = acl->a_entries; \ - acl_e < acl->a_entries + acl->a_count; \ - acl_e++) - /* * Convert from in-memory to filesystem representation. */ @@ -199,11 +194,11 @@ bch2_acl_to_xattr(struct btree_trans *trans, { struct bkey_i_xattr *xattr; bch_acl_header *acl_header; - const struct posix_acl_entry *acl_e; + const struct posix_acl_entry *acl_e, *pe; void *outptr; unsigned nr_short = 0, nr_long = 0, acl_len, u64s; - acl_for_each_entry(acl, acl_e) { + FOREACH_ACL_ENTRY(acl_e, acl, pe) { switch (acl_e->e_tag) { case ACL_USER: case ACL_GROUP: @@ -241,7 +236,7 @@ bch2_acl_to_xattr(struct btree_trans *trans, outptr = (void *) acl_header + sizeof(*acl_header); - acl_for_each_entry(acl, acl_e) { + FOREACH_ACL_ENTRY(acl_e, acl, pe) { bch_acl_entry *entry = outptr; entry->e_tag = cpu_to_le16(acl_e->e_tag); diff --git a/fs/bcachefs/alloc_background.c b/fs/bcachefs/alloc_background.c index c84a91572a1d..fc2ef33b67b3 100644 --- a/fs/bcachefs/alloc_background.c +++ b/fs/bcachefs/alloc_background.c @@ -198,7 +198,7 @@ static unsigned bch_alloc_v1_val_u64s(const struct bch_alloc *a) } int bch2_alloc_v1_validate(struct bch_fs *c, struct bkey_s_c k, - enum bch_validate_flags flags) + struct bkey_validate_context from) { struct bkey_s_c_alloc a = bkey_s_c_to_alloc(k); int ret = 0; @@ -213,7 +213,7 @@ fsck_err: } int bch2_alloc_v2_validate(struct bch_fs *c, struct bkey_s_c k, - enum bch_validate_flags flags) + struct bkey_validate_context from) { struct bkey_alloc_unpacked u; int ret = 0; @@ -226,7 +226,7 @@ fsck_err: } int bch2_alloc_v3_validate(struct bch_fs *c, struct bkey_s_c k, - enum bch_validate_flags flags) + struct bkey_validate_context from) { struct bkey_alloc_unpacked u; int ret = 0; @@ -239,7 +239,7 @@ fsck_err: } int bch2_alloc_v4_validate(struct bch_fs *c, struct bkey_s_c k, - enum bch_validate_flags flags) + struct bkey_validate_context from) { struct bch_alloc_v4 a; int ret = 0; @@ -322,9 +322,9 @@ fsck_err: void bch2_alloc_v4_swab(struct bkey_s k) { struct bch_alloc_v4 *a = bkey_s_to_alloc_v4(k).v; - struct bch_backpointer *bp, *bps; - a->journal_seq = swab64(a->journal_seq); + a->journal_seq_nonempty = swab64(a->journal_seq_nonempty); + a->journal_seq_empty = swab64(a->journal_seq_empty); a->flags = swab32(a->flags); a->dirty_sectors = swab32(a->dirty_sectors); a->cached_sectors = swab32(a->cached_sectors); @@ -333,13 +333,6 @@ void bch2_alloc_v4_swab(struct bkey_s k) a->stripe = swab32(a->stripe); a->nr_external_backpointers = swab32(a->nr_external_backpointers); a->stripe_sectors = swab32(a->stripe_sectors); - - bps = alloc_v4_backpointers(a); - for (bp = bps; bp < bps + BCH_ALLOC_V4_NR_BACKPOINTERS(a); bp++) { - bp->bucket_offset = swab40(bp->bucket_offset); - bp->bucket_len = swab32(bp->bucket_len); - bch2_bpos_swab(&bp->pos); - } } void bch2_alloc_to_text(struct printbuf *out, struct bch_fs *c, struct bkey_s_c k) @@ -354,16 +347,17 @@ void bch2_alloc_to_text(struct printbuf *out, struct bch_fs *c, struct bkey_s_c prt_printf(out, "gen %u oldest_gen %u data_type ", a->gen, a->oldest_gen); bch2_prt_data_type(out, a->data_type); prt_newline(out); - prt_printf(out, "journal_seq %llu\n", a->journal_seq); - prt_printf(out, "need_discard %llu\n", BCH_ALLOC_V4_NEED_DISCARD(a)); - prt_printf(out, "need_inc_gen %llu\n", BCH_ALLOC_V4_NEED_INC_GEN(a)); - prt_printf(out, "dirty_sectors %u\n", a->dirty_sectors); - prt_printf(out, "stripe_sectors %u\n", a->stripe_sectors); - prt_printf(out, "cached_sectors %u\n", a->cached_sectors); - prt_printf(out, "stripe %u\n", a->stripe); - prt_printf(out, "stripe_redundancy %u\n", a->stripe_redundancy); - prt_printf(out, "io_time[READ] %llu\n", a->io_time[READ]); - prt_printf(out, "io_time[WRITE] %llu\n", a->io_time[WRITE]); + prt_printf(out, "journal_seq_nonempty %llu\n", a->journal_seq_nonempty); + prt_printf(out, "journal_seq_empty %llu\n", a->journal_seq_empty); + prt_printf(out, "need_discard %llu\n", BCH_ALLOC_V4_NEED_DISCARD(a)); + prt_printf(out, "need_inc_gen %llu\n", BCH_ALLOC_V4_NEED_INC_GEN(a)); + prt_printf(out, "dirty_sectors %u\n", a->dirty_sectors); + prt_printf(out, "stripe_sectors %u\n", a->stripe_sectors); + prt_printf(out, "cached_sectors %u\n", a->cached_sectors); + prt_printf(out, "stripe %u\n", a->stripe); + prt_printf(out, "stripe_redundancy %u\n", a->stripe_redundancy); + prt_printf(out, "io_time[READ] %llu\n", a->io_time[READ]); + prt_printf(out, "io_time[WRITE] %llu\n", a->io_time[WRITE]); if (ca) prt_printf(out, "fragmentation %llu\n", alloc_lru_idx_fragmentation(*a, ca)); @@ -392,7 +386,7 @@ void __bch2_alloc_to_v4(struct bkey_s_c k, struct bch_alloc_v4 *out) struct bkey_alloc_unpacked u = bch2_alloc_unpack(k); *out = (struct bch_alloc_v4) { - .journal_seq = u.journal_seq, + .journal_seq_nonempty = u.journal_seq, .flags = u.need_discard, .gen = u.gen, .oldest_gen = u.oldest_gen, @@ -517,7 +511,7 @@ static unsigned alloc_gen(struct bkey_s_c k, unsigned offset) } int bch2_bucket_gens_validate(struct bch_fs *c, struct bkey_s_c k, - enum bch_validate_flags flags) + struct bkey_validate_context from) { int ret = 0; @@ -664,74 +658,80 @@ int bch2_alloc_read(struct bch_fs *c) /* Free space/discard btree: */ +static int __need_discard_or_freespace_err(struct btree_trans *trans, + struct bkey_s_c alloc_k, + bool set, bool discard, bool repair) +{ + struct bch_fs *c = trans->c; + enum bch_fsck_flags flags = FSCK_CAN_IGNORE|(repair ? FSCK_CAN_FIX : 0); + enum bch_sb_error_id err_id = discard + ? BCH_FSCK_ERR_need_discard_key_wrong + : BCH_FSCK_ERR_freespace_key_wrong; + enum btree_id btree = discard ? BTREE_ID_need_discard : BTREE_ID_freespace; + struct printbuf buf = PRINTBUF; + + bch2_bkey_val_to_text(&buf, c, alloc_k); + + int ret = __bch2_fsck_err(NULL, trans, flags, err_id, + "bucket incorrectly %sset in %s btree\n" + " %s", + set ? "" : "un", + bch2_btree_id_str(btree), + buf.buf); + if (ret == -BCH_ERR_fsck_ignore || + ret == -BCH_ERR_fsck_errors_not_fixed) + ret = 0; + + printbuf_exit(&buf); + return ret; +} + +#define need_discard_or_freespace_err(...) \ + fsck_err_wrap(__need_discard_or_freespace_err(__VA_ARGS__)) + +#define need_discard_or_freespace_err_on(cond, ...) \ + (unlikely(cond) ? need_discard_or_freespace_err(__VA_ARGS__) : false) + static int bch2_bucket_do_index(struct btree_trans *trans, struct bch_dev *ca, struct bkey_s_c alloc_k, const struct bch_alloc_v4 *a, bool set) { - struct bch_fs *c = trans->c; - struct btree_iter iter; - struct bkey_s_c old; - struct bkey_i *k; enum btree_id btree; - enum bch_bkey_type old_type = !set ? KEY_TYPE_set : KEY_TYPE_deleted; - enum bch_bkey_type new_type = set ? KEY_TYPE_set : KEY_TYPE_deleted; - struct printbuf buf = PRINTBUF; - int ret; + struct bpos pos; if (a->data_type != BCH_DATA_free && a->data_type != BCH_DATA_need_discard) return 0; - k = bch2_trans_kmalloc_nomemzero(trans, sizeof(*k)); - if (IS_ERR(k)) - return PTR_ERR(k); - - bkey_init(&k->k); - k->k.type = new_type; - switch (a->data_type) { case BCH_DATA_free: btree = BTREE_ID_freespace; - k->k.p = alloc_freespace_pos(alloc_k.k->p, *a); - bch2_key_resize(&k->k, 1); + pos = alloc_freespace_pos(alloc_k.k->p, *a); break; case BCH_DATA_need_discard: btree = BTREE_ID_need_discard; - k->k.p = alloc_k.k->p; + pos = alloc_k.k->p; break; default: return 0; } - old = bch2_bkey_get_iter(trans, &iter, btree, - bkey_start_pos(&k->k), - BTREE_ITER_intent); - ret = bkey_err(old); + struct btree_iter iter; + struct bkey_s_c old = bch2_bkey_get_iter(trans, &iter, btree, pos, BTREE_ITER_intent); + int ret = bkey_err(old); if (ret) return ret; - if (ca->mi.freespace_initialized && - c->curr_recovery_pass > BCH_RECOVERY_PASS_check_alloc_info && - bch2_trans_inconsistent_on(old.k->type != old_type, trans, - "incorrect key when %s %s:%llu:%llu:0 (got %s should be %s)\n" - " for %s", - set ? "setting" : "clearing", - bch2_btree_id_str(btree), - iter.pos.inode, - iter.pos.offset, - bch2_bkey_types[old.k->type], - bch2_bkey_types[old_type], - (bch2_bkey_val_to_text(&buf, c, alloc_k), buf.buf))) { - ret = -EIO; - goto err; - } + need_discard_or_freespace_err_on(ca->mi.freespace_initialized && + !old.k->type != set, + trans, alloc_k, set, + btree == BTREE_ID_need_discard, false); - ret = bch2_trans_update(trans, &iter, k, 0); -err: + ret = bch2_btree_bit_mod_iter(trans, &iter, set); +fsck_err: bch2_trans_iter_exit(trans, &iter); - printbuf_exit(&buf); return ret; } @@ -858,7 +858,10 @@ int bch2_trigger_alloc(struct btree_trans *trans, if (flags & BTREE_TRIGGER_transactional) { alloc_data_type_set(new_a, new_a->data_type); - if (bch2_bucket_sectors_total(*new_a) > bch2_bucket_sectors_total(*old_a)) { + int is_empty_delta = (int) data_type_is_empty(new_a->data_type) - + (int) data_type_is_empty(old_a->data_type); + + if (is_empty_delta < 0) { new_a->io_time[READ] = bch2_current_io_time(c, READ); new_a->io_time[WRITE]= bch2_current_io_time(c, WRITE); SET_BCH_ALLOC_V4_NEED_INC_GEN(new_a, true); @@ -928,37 +931,55 @@ int bch2_trigger_alloc(struct btree_trans *trans, } if ((flags & BTREE_TRIGGER_atomic) && (flags & BTREE_TRIGGER_insert)) { - u64 journal_seq = trans->journal_res.seq; - u64 bucket_journal_seq = new_a->journal_seq; + u64 transaction_seq = trans->journal_res.seq; + BUG_ON(!transaction_seq); - if ((flags & BTREE_TRIGGER_insert) && - data_type_is_empty(old_a->data_type) != - data_type_is_empty(new_a->data_type) && - new.k->type == KEY_TYPE_alloc_v4) { - struct bch_alloc_v4 *v = bkey_s_to_alloc_v4(new).v; + if (log_fsck_err_on(transaction_seq && new_a->journal_seq_nonempty > transaction_seq, + trans, alloc_key_journal_seq_in_future, + "bucket journal seq in future (currently at %llu)\n%s", + journal_cur_seq(&c->journal), + (bch2_bkey_val_to_text(&buf, c, new.s_c), buf.buf))) + new_a->journal_seq_nonempty = transaction_seq; - /* - * If the btree updates referring to a bucket weren't flushed - * before the bucket became empty again, then the we don't have - * to wait on a journal flush before we can reuse the bucket: - */ - v->journal_seq = bucket_journal_seq = - data_type_is_empty(new_a->data_type) && - (journal_seq == v->journal_seq || - bch2_journal_noflush_seq(&c->journal, v->journal_seq)) - ? 0 : journal_seq; + int is_empty_delta = (int) data_type_is_empty(new_a->data_type) - + (int) data_type_is_empty(old_a->data_type); + + /* + * Record journal sequence number of empty -> nonempty transition: + * Note that there may be multiple empty -> nonempty + * transitions, data in a bucket may be overwritten while we're + * still writing to it - so be careful to only record the first: + * */ + if (is_empty_delta < 0 && + new_a->journal_seq_empty <= c->journal.flushed_seq_ondisk) { + new_a->journal_seq_nonempty = transaction_seq; + new_a->journal_seq_empty = 0; } - if (!data_type_is_empty(old_a->data_type) && - data_type_is_empty(new_a->data_type) && - bucket_journal_seq) { - ret = bch2_set_bucket_needs_journal_commit(&c->buckets_waiting_for_journal, - c->journal.flushed_seq_ondisk, - new.k->p.inode, new.k->p.offset, - bucket_journal_seq); - if (bch2_fs_fatal_err_on(ret, c, - "setting bucket_needs_journal_commit: %s", bch2_err_str(ret))) - goto err; + /* + * Bucket becomes empty: mark it as waiting for a journal flush, + * unless updates since empty -> nonempty transition were never + * flushed - we may need to ask the journal not to flush + * intermediate sequence numbers: + */ + if (is_empty_delta > 0) { + if (new_a->journal_seq_nonempty == transaction_seq || + bch2_journal_noflush_seq(&c->journal, + new_a->journal_seq_nonempty, + transaction_seq)) { + new_a->journal_seq_nonempty = new_a->journal_seq_empty = 0; + } else { + new_a->journal_seq_empty = transaction_seq; + + ret = bch2_set_bucket_needs_journal_commit(&c->buckets_waiting_for_journal, + c->journal.flushed_seq_ondisk, + new.k->p.inode, new.k->p.offset, + transaction_seq); + if (bch2_fs_fatal_err_on(ret, c, + "setting bucket_needs_journal_commit: %s", + bch2_err_str(ret))) + goto err; + } } if (new_a->gen != old_a->gen) { @@ -974,7 +995,7 @@ int bch2_trigger_alloc(struct btree_trans *trans, #define eval_state(_a, expr) ({ const struct bch_alloc_v4 *a = _a; expr; }) #define statechange(expr) !eval_state(old_a, expr) && eval_state(new_a, expr) -#define bucket_flushed(a) (!a->journal_seq || a->journal_seq <= c->journal.flushed_seq_ondisk) +#define bucket_flushed(a) (a->journal_seq_empty <= c->journal.flushed_seq_ondisk) if (statechange(a->data_type == BCH_DATA_free) && bucket_flushed(new_a)) @@ -1006,6 +1027,7 @@ int bch2_trigger_alloc(struct btree_trans *trans, rcu_read_unlock(); } err: +fsck_err: printbuf_exit(&buf); bch2_dev_put(ca); return ret; @@ -1045,7 +1067,7 @@ static struct bkey_s_c bch2_get_key_or_hole(struct btree_iter *iter, struct bpos * btree node min/max is a closed interval, upto takes a half * open interval: */ - k = bch2_btree_iter_peek_upto(&iter2, end); + k = bch2_btree_iter_peek_max(&iter2, end); next = iter2.pos; bch2_trans_iter_exit(iter->trans, &iter2); @@ -1129,7 +1151,6 @@ int bch2_check_alloc_key(struct btree_trans *trans, struct bch_fs *c = trans->c; struct bch_alloc_v4 a_convert; const struct bch_alloc_v4 *a; - unsigned discard_key_type, freespace_key_type; unsigned gens_offset; struct bkey_s_c k; struct printbuf buf = PRINTBUF; @@ -1149,64 +1170,30 @@ int bch2_check_alloc_key(struct btree_trans *trans, a = bch2_alloc_to_v4(alloc_k, &a_convert); - discard_key_type = a->data_type == BCH_DATA_need_discard ? KEY_TYPE_set : 0; bch2_btree_iter_set_pos(discard_iter, alloc_k.k->p); k = bch2_btree_iter_peek_slot(discard_iter); ret = bkey_err(k); if (ret) goto err; - if (fsck_err_on(k.k->type != discard_key_type, - trans, need_discard_key_wrong, - "incorrect key in need_discard btree (got %s should be %s)\n" - " %s", - bch2_bkey_types[k.k->type], - bch2_bkey_types[discard_key_type], - (bch2_bkey_val_to_text(&buf, c, alloc_k), buf.buf))) { - struct bkey_i *update = - bch2_trans_kmalloc(trans, sizeof(*update)); - - ret = PTR_ERR_OR_ZERO(update); - if (ret) - goto err; - - bkey_init(&update->k); - update->k.type = discard_key_type; - update->k.p = discard_iter->pos; - - ret = bch2_trans_update(trans, discard_iter, update, 0); + bool is_discarded = a->data_type == BCH_DATA_need_discard; + if (need_discard_or_freespace_err_on(!!k.k->type != is_discarded, + trans, alloc_k, !is_discarded, true, true)) { + ret = bch2_btree_bit_mod_iter(trans, discard_iter, is_discarded); if (ret) goto err; } - freespace_key_type = a->data_type == BCH_DATA_free ? KEY_TYPE_set : 0; bch2_btree_iter_set_pos(freespace_iter, alloc_freespace_pos(alloc_k.k->p, *a)); k = bch2_btree_iter_peek_slot(freespace_iter); ret = bkey_err(k); if (ret) goto err; - if (fsck_err_on(k.k->type != freespace_key_type, - trans, freespace_key_wrong, - "incorrect key in freespace btree (got %s should be %s)\n" - " %s", - bch2_bkey_types[k.k->type], - bch2_bkey_types[freespace_key_type], - (printbuf_reset(&buf), - bch2_bkey_val_to_text(&buf, c, alloc_k), buf.buf))) { - struct bkey_i *update = - bch2_trans_kmalloc(trans, sizeof(*update)); - - ret = PTR_ERR_OR_ZERO(update); - if (ret) - goto err; - - bkey_init(&update->k); - update->k.type = freespace_key_type; - update->k.p = freespace_iter->pos; - bch2_key_resize(&update->k, 1); - - ret = bch2_trans_update(trans, freespace_iter, update, 0); + bool is_free = a->data_type == BCH_DATA_free; + if (need_discard_or_freespace_err_on(!!k.k->type != is_free, + trans, alloc_k, !is_free, false, true)) { + ret = bch2_btree_bit_mod_iter(trans, freespace_iter, is_free); if (ret) goto err; } @@ -1368,51 +1355,88 @@ fsck_err: return ret; } -static noinline_for_stack int bch2_check_discard_freespace_key(struct btree_trans *trans, - struct btree_iter *iter) +struct check_discard_freespace_key_async { + struct work_struct work; + struct bch_fs *c; + struct bbpos pos; +}; + +static int bch2_recheck_discard_freespace_key(struct btree_trans *trans, struct bbpos pos) +{ + struct btree_iter iter; + struct bkey_s_c k = bch2_bkey_get_iter(trans, &iter, pos.btree, pos.pos, 0); + int ret = bkey_err(k); + if (ret) + return ret; + + u8 gen; + ret = k.k->type != KEY_TYPE_set + ? bch2_check_discard_freespace_key(trans, &iter, &gen, false) + : 0; + bch2_trans_iter_exit(trans, &iter); + return ret; +} + +static void check_discard_freespace_key_work(struct work_struct *work) +{ + struct check_discard_freespace_key_async *w = + container_of(work, struct check_discard_freespace_key_async, work); + + bch2_trans_do(w->c, bch2_recheck_discard_freespace_key(trans, w->pos)); + bch2_write_ref_put(w->c, BCH_WRITE_REF_check_discard_freespace_key); + kfree(w); +} + +int bch2_check_discard_freespace_key(struct btree_trans *trans, struct btree_iter *iter, u8 *gen, + bool async_repair) { struct bch_fs *c = trans->c; - struct btree_iter alloc_iter; - struct bkey_s_c alloc_k; - struct bch_alloc_v4 a_convert; - const struct bch_alloc_v4 *a; - u64 genbits; - struct bpos pos; enum bch_data_type state = iter->btree_id == BTREE_ID_need_discard ? BCH_DATA_need_discard : BCH_DATA_free; struct printbuf buf = PRINTBUF; - int ret; - pos = iter->pos; - pos.offset &= ~(~0ULL << 56); - genbits = iter->pos.offset & (~0ULL << 56); + struct bpos bucket = iter->pos; + bucket.offset &= ~(~0ULL << 56); + u64 genbits = iter->pos.offset & (~0ULL << 56); - alloc_k = bch2_bkey_get_iter(trans, &alloc_iter, BTREE_ID_alloc, pos, 0); - ret = bkey_err(alloc_k); + struct btree_iter alloc_iter; + struct bkey_s_c alloc_k = bch2_bkey_get_iter(trans, &alloc_iter, + BTREE_ID_alloc, bucket, + async_repair ? BTREE_ITER_cached : 0); + int ret = bkey_err(alloc_k); if (ret) return ret; - if (fsck_err_on(!bch2_dev_bucket_exists(c, pos), - trans, need_discard_freespace_key_to_invalid_dev_bucket, - "entry in %s btree for nonexistant dev:bucket %llu:%llu", - bch2_btree_id_str(iter->btree_id), pos.inode, pos.offset)) - goto delete; + if (!bch2_dev_bucket_exists(c, bucket)) { + if (fsck_err(trans, need_discard_freespace_key_to_invalid_dev_bucket, + "entry in %s btree for nonexistant dev:bucket %llu:%llu", + bch2_btree_id_str(iter->btree_id), bucket.inode, bucket.offset)) + goto delete; + ret = 1; + goto out; + } - a = bch2_alloc_to_v4(alloc_k, &a_convert); + struct bch_alloc_v4 a_convert; + const struct bch_alloc_v4 *a = bch2_alloc_to_v4(alloc_k, &a_convert); + + if (a->data_type != state || + (state == BCH_DATA_free && + genbits != alloc_freespace_genbits(*a))) { + if (fsck_err(trans, need_discard_freespace_key_bad, + "%s\n incorrectly set at %s:%llu:%llu:0 (free %u, genbits %llu should be %llu)", + (bch2_bkey_val_to_text(&buf, c, alloc_k), buf.buf), + bch2_btree_id_str(iter->btree_id), + iter->pos.inode, + iter->pos.offset, + a->data_type == state, + genbits >> 56, alloc_freespace_genbits(*a) >> 56)) + goto delete; + ret = 1; + goto out; + } - if (fsck_err_on(a->data_type != state || - (state == BCH_DATA_free && - genbits != alloc_freespace_genbits(*a)), - trans, need_discard_freespace_key_bad, - "%s\n incorrectly set at %s:%llu:%llu:0 (free %u, genbits %llu should be %llu)", - (bch2_bkey_val_to_text(&buf, c, alloc_k), buf.buf), - bch2_btree_id_str(iter->btree_id), - iter->pos.inode, - iter->pos.offset, - a->data_type == state, - genbits >> 56, alloc_freespace_genbits(*a) >> 56)) - goto delete; + *gen = a->gen; out: fsck_err: bch2_set_btree_iter_dontneed(&alloc_iter); @@ -1420,11 +1444,40 @@ fsck_err: printbuf_exit(&buf); return ret; delete: - ret = bch2_btree_delete_extent_at(trans, iter, - iter->btree_id == BTREE_ID_freespace ? 1 : 0, 0) ?: - bch2_trans_commit(trans, NULL, NULL, - BCH_TRANS_COMMIT_no_enospc); - goto out; + if (!async_repair) { + ret = bch2_btree_bit_mod_iter(trans, iter, false) ?: + bch2_trans_commit(trans, NULL, NULL, + BCH_TRANS_COMMIT_no_enospc) ?: + -BCH_ERR_transaction_restart_commit; + goto out; + } else { + /* + * We can't repair here when called from the allocator path: the + * commit will recurse back into the allocator + */ + struct check_discard_freespace_key_async *w = + kzalloc(sizeof(*w), GFP_KERNEL); + if (!w) + goto out; + + if (!bch2_write_ref_tryget(c, BCH_WRITE_REF_check_discard_freespace_key)) { + kfree(w); + goto out; + } + + INIT_WORK(&w->work, check_discard_freespace_key_work); + w->c = c; + w->pos = BBPOS(iter->btree_id, iter->pos); + queue_work(c->write_ref_wq, &w->work); + goto out; + } +} + +static int bch2_check_discard_freespace_key_fsck(struct btree_trans *trans, struct btree_iter *iter) +{ + u8 gen; + int ret = bch2_check_discard_freespace_key(trans, iter, &gen, false); + return ret < 0 ? ret : 0; } /* @@ -1581,7 +1634,7 @@ bkey_err: ret = for_each_btree_key(trans, iter, BTREE_ID_need_discard, POS_MIN, BTREE_ITER_prefetch, k, - bch2_check_discard_freespace_key(trans, &iter)); + bch2_check_discard_freespace_key_fsck(trans, &iter)); if (ret) goto err; @@ -1594,7 +1647,7 @@ bkey_err: break; ret = bkey_err(k) ?: - bch2_check_discard_freespace_key(trans, &iter); + bch2_check_discard_freespace_key_fsck(trans, &iter); if (bch2_err_matches(ret, BCH_ERR_transaction_restart)) { ret = 0; continue; @@ -1757,7 +1810,8 @@ static int bch2_discard_one_bucket(struct btree_trans *trans, struct bch_dev *ca, struct btree_iter *need_discard_iter, struct bpos *discard_pos_done, - struct discard_buckets_state *s) + struct discard_buckets_state *s, + bool fastpath) { struct bch_fs *c = trans->c; struct bpos pos = need_discard_iter->pos; @@ -1793,44 +1847,23 @@ static int bch2_discard_one_bucket(struct btree_trans *trans, if (ret) goto out; - if (bch2_bucket_sectors_total(a->v)) { - if (bch2_trans_inconsistent_on(c->curr_recovery_pass > BCH_RECOVERY_PASS_check_alloc_info, - trans, "attempting to discard bucket with dirty data\n%s", - (bch2_bkey_val_to_text(&buf, c, k), buf.buf))) - ret = -EIO; - goto out; - } - if (a->v.data_type != BCH_DATA_need_discard) { - if (data_type_is_empty(a->v.data_type) && - BCH_ALLOC_V4_NEED_INC_GEN(&a->v)) { - a->v.gen++; - SET_BCH_ALLOC_V4_NEED_INC_GEN(&a->v, false); - goto write; + if (need_discard_or_freespace_err(trans, k, true, true, true)) { + ret = bch2_btree_bit_mod_iter(trans, need_discard_iter, false); + if (ret) + goto out; + goto commit; } - if (bch2_trans_inconsistent_on(c->curr_recovery_pass > BCH_RECOVERY_PASS_check_alloc_info, - trans, "bucket incorrectly set in need_discard btree\n" - "%s", - (bch2_bkey_val_to_text(&buf, c, k), buf.buf))) - ret = -EIO; goto out; } - if (a->v.journal_seq > c->journal.flushed_seq_ondisk) { - if (bch2_trans_inconsistent_on(c->curr_recovery_pass > BCH_RECOVERY_PASS_check_alloc_info, - trans, "clearing need_discard but journal_seq %llu > flushed_seq %llu\n%s", - a->v.journal_seq, - c->journal.flushed_seq_ondisk, - (bch2_bkey_val_to_text(&buf, c, k), buf.buf))) - ret = -EIO; - goto out; - } - - if (discard_in_flight_add(ca, iter.pos.offset, true)) - goto out; + if (!fastpath) { + if (discard_in_flight_add(ca, iter.pos.offset, true)) + goto out; - discard_locked = true; + discard_locked = true; + } if (!bkey_eq(*discard_pos_done, iter.pos) && ca->mi.discard && !c->opts.nochanges) { @@ -1844,6 +1877,7 @@ static int bch2_discard_one_bucket(struct btree_trans *trans, ca->mi.bucket_size, GFP_KERNEL); *discard_pos_done = iter.pos; + s->discarded++; ret = bch2_trans_relock_notrace(trans); if (ret) @@ -1851,22 +1885,25 @@ static int bch2_discard_one_bucket(struct btree_trans *trans, } SET_BCH_ALLOC_V4_NEED_DISCARD(&a->v, false); -write: alloc_data_type_set(&a->v, a->v.data_type); - ret = bch2_trans_update(trans, &iter, &a->k_i, 0) ?: - bch2_trans_commit(trans, NULL, NULL, - BCH_WATERMARK_btree| - BCH_TRANS_COMMIT_no_enospc); + ret = bch2_trans_update(trans, &iter, &a->k_i, 0); + if (ret) + goto out; +commit: + ret = bch2_trans_commit(trans, NULL, NULL, + BCH_WATERMARK_btree| + BCH_TRANS_COMMIT_no_enospc); if (ret) goto out; count_event(c, bucket_discard); - s->discarded++; out: +fsck_err: if (discard_locked) discard_in_flight_remove(ca, iter.pos.offset); - s->seen++; + if (!ret) + s->seen++; bch2_trans_iter_exit(trans, &iter); printbuf_exit(&buf); return ret; @@ -1886,11 +1923,11 @@ static void bch2_do_discards_work(struct work_struct *work) * successful commit: */ ret = bch2_trans_run(c, - for_each_btree_key_upto(trans, iter, + for_each_btree_key_max(trans, iter, BTREE_ID_need_discard, POS(ca->dev_idx, 0), POS(ca->dev_idx, U64_MAX), 0, k, - bch2_discard_one_bucket(trans, ca, &iter, &discard_pos_done, &s))); + bch2_discard_one_bucket(trans, ca, &iter, &discard_pos_done, &s, false))); trace_discard_buckets(c, s.seen, s.open, s.need_journal_commit, s.discarded, bch2_err_str(ret)); @@ -1923,27 +1960,29 @@ void bch2_do_discards(struct bch_fs *c) bch2_dev_do_discards(ca); } -static int bch2_clear_bucket_needs_discard(struct btree_trans *trans, struct bpos bucket) +static int bch2_do_discards_fast_one(struct btree_trans *trans, + struct bch_dev *ca, + u64 bucket, + struct bpos *discard_pos_done, + struct discard_buckets_state *s) { - struct btree_iter iter; - bch2_trans_iter_init(trans, &iter, BTREE_ID_alloc, bucket, BTREE_ITER_intent); - struct bkey_s_c k = bch2_btree_iter_peek_slot(&iter); - int ret = bkey_err(k); + struct btree_iter need_discard_iter; + struct bkey_s_c discard_k = bch2_bkey_get_iter(trans, &need_discard_iter, + BTREE_ID_need_discard, POS(ca->dev_idx, bucket), 0); + int ret = bkey_err(discard_k); if (ret) - goto err; - - struct bkey_i_alloc_v4 *a = bch2_alloc_to_v4_mut(trans, k); - ret = PTR_ERR_OR_ZERO(a); - if (ret) - goto err; + return ret; - BUG_ON(a->v.dirty_sectors); - SET_BCH_ALLOC_V4_NEED_DISCARD(&a->v, false); - alloc_data_type_set(&a->v, a->v.data_type); + if (log_fsck_err_on(discard_k.k->type != KEY_TYPE_set, + trans, discarding_bucket_not_in_need_discard_btree, + "attempting to discard bucket %u:%llu not in need_discard btree", + ca->dev_idx, bucket)) + goto out; - ret = bch2_trans_update(trans, &iter, &a->k_i, 0); -err: - bch2_trans_iter_exit(trans, &iter); + ret = bch2_discard_one_bucket(trans, ca, &need_discard_iter, discard_pos_done, s, true); +out: +fsck_err: + bch2_trans_iter_exit(trans, &need_discard_iter); return ret; } @@ -1951,6 +1990,10 @@ static void bch2_do_discards_fast_work(struct work_struct *work) { struct bch_dev *ca = container_of(work, struct bch_dev, discard_fast_work); struct bch_fs *c = ca->fs; + struct discard_buckets_state s = {}; + struct bpos discard_pos_done = POS_MAX; + struct btree_trans *trans = bch2_trans_get(c); + int ret = 0; while (1) { bool got_bucket = false; @@ -1971,16 +2014,8 @@ static void bch2_do_discards_fast_work(struct work_struct *work) if (!got_bucket) break; - if (ca->mi.discard && !c->opts.nochanges) - blkdev_issue_discard(ca->disk_sb.bdev, - bucket_to_sector(ca, bucket), - ca->mi.bucket_size, - GFP_KERNEL); - - int ret = bch2_trans_commit_do(c, NULL, NULL, - BCH_WATERMARK_btree| - BCH_TRANS_COMMIT_no_enospc, - bch2_clear_bucket_needs_discard(trans, POS(ca->dev_idx, bucket))); + ret = lockrestart_do(trans, + bch2_do_discards_fast_one(trans, ca, bucket, &discard_pos_done, &s)); bch_err_fn(c, ret); discard_in_flight_remove(ca, bucket); @@ -1989,6 +2024,9 @@ static void bch2_do_discards_fast_work(struct work_struct *work) break; } + trace_discard_buckets(c, s.seen, s.open, s.need_journal_commit, s.discarded, bch2_err_str(ret)); + + bch2_trans_put(trans); percpu_ref_put(&ca->io_ref); bch2_write_ref_put(c, BCH_WRITE_REF_discard_fast); } @@ -2030,8 +2068,11 @@ static int invalidate_one_bucket(struct btree_trans *trans, return 1; if (!bch2_dev_bucket_exists(c, bucket)) { - prt_str(&buf, "lru entry points to invalid bucket"); - goto err; + if (fsck_err(trans, lru_entry_to_invalid_bucket, + "lru key points to nonexistent device:bucket %llu:%llu", + bucket.inode, bucket.offset)) + return bch2_btree_bit_mod_buffered(trans, BTREE_ID_lru, lru_iter->pos, false); + goto out; } if (bch2_bucket_is_open_safe(c, bucket.inode, bucket.offset)) @@ -2072,28 +2113,9 @@ static int invalidate_one_bucket(struct btree_trans *trans, trace_and_count(c, bucket_invalidate, c, bucket.inode, bucket.offset, cached_sectors); --*nr_to_invalidate; out: +fsck_err: printbuf_exit(&buf); return ret; -err: - prt_str(&buf, "\n lru key: "); - bch2_bkey_val_to_text(&buf, c, lru_k); - - prt_str(&buf, "\n lru entry: "); - bch2_lru_pos_to_text(&buf, lru_iter->pos); - - prt_str(&buf, "\n alloc key: "); - if (!a) - bch2_bpos_to_text(&buf, bucket); - else - bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(&a->k_i)); - - bch_err(c, "%s", buf.buf); - if (c->curr_recovery_pass > BCH_RECOVERY_PASS_check_lrus) { - bch2_inconsistent_error(c); - ret = -EINVAL; - } - - goto out; } static struct bkey_s_c next_lru_key(struct btree_trans *trans, struct btree_iter *iter, @@ -2101,7 +2123,7 @@ static struct bkey_s_c next_lru_key(struct btree_trans *trans, struct btree_iter { struct bkey_s_c k; again: - k = bch2_btree_iter_peek_upto(iter, lru_pos(ca->dev_idx, U64_MAX, LRU_TIME_MAX)); + k = bch2_btree_iter_peek_max(iter, lru_pos(ca->dev_idx, U64_MAX, LRU_TIME_MAX)); if (!k.k && !*wrapped) { bch2_btree_iter_set_pos(iter, lru_pos(ca->dev_idx, 0, 0)); *wrapped = true; diff --git a/fs/bcachefs/alloc_background.h b/fs/bcachefs/alloc_background.h index 163a67b97a40..de25ba4ee94b 100644 --- a/fs/bcachefs/alloc_background.h +++ b/fs/bcachefs/alloc_background.h @@ -8,8 +8,6 @@ #include "debug.h" #include "super.h" -enum bch_validate_flags; - /* How out of date a pointer gen is allowed to be: */ #define BUCKET_GC_GEN_MAX 96U @@ -245,10 +243,14 @@ struct bkey_i_alloc_v4 *bch2_alloc_to_v4_mut(struct btree_trans *, struct bkey_s int bch2_bucket_io_time_reset(struct btree_trans *, unsigned, size_t, int); -int bch2_alloc_v1_validate(struct bch_fs *, struct bkey_s_c, enum bch_validate_flags); -int bch2_alloc_v2_validate(struct bch_fs *, struct bkey_s_c, enum bch_validate_flags); -int bch2_alloc_v3_validate(struct bch_fs *, struct bkey_s_c, enum bch_validate_flags); -int bch2_alloc_v4_validate(struct bch_fs *, struct bkey_s_c, enum bch_validate_flags); +int bch2_alloc_v1_validate(struct bch_fs *, struct bkey_s_c, + struct bkey_validate_context); +int bch2_alloc_v2_validate(struct bch_fs *, struct bkey_s_c, + struct bkey_validate_context); +int bch2_alloc_v3_validate(struct bch_fs *, struct bkey_s_c, + struct bkey_validate_context); +int bch2_alloc_v4_validate(struct bch_fs *, struct bkey_s_c, + struct bkey_validate_context); void bch2_alloc_v4_swab(struct bkey_s); void bch2_alloc_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c); @@ -282,7 +284,7 @@ void bch2_alloc_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c); }) int bch2_bucket_gens_validate(struct bch_fs *, struct bkey_s_c, - enum bch_validate_flags); + struct bkey_validate_context); void bch2_bucket_gens_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c); #define bch2_bkey_ops_bucket_gens ((struct bkey_ops) { \ @@ -307,6 +309,8 @@ int bch2_alloc_key_to_dev_counters(struct btree_trans *, struct bch_dev *, int bch2_trigger_alloc(struct btree_trans *, enum btree_id, unsigned, struct bkey_s_c, struct bkey_s, enum btree_iter_update_trigger_flags); + +int bch2_check_discard_freespace_key(struct btree_trans *, struct btree_iter *, u8 *, bool); int bch2_check_alloc_info(struct bch_fs *); int bch2_check_alloc_to_lru_refs(struct bch_fs *); void bch2_dev_do_discards(struct bch_dev *); diff --git a/fs/bcachefs/alloc_background_format.h b/fs/bcachefs/alloc_background_format.h index befdaa95c515..740238369a5a 100644 --- a/fs/bcachefs/alloc_background_format.h +++ b/fs/bcachefs/alloc_background_format.h @@ -58,7 +58,7 @@ LE32_BITMASK(BCH_ALLOC_V3_NEED_INC_GEN,struct bch_alloc_v3, flags, 1, 2) struct bch_alloc_v4 { struct bch_val v; - __u64 journal_seq; + __u64 journal_seq_nonempty; __u32 flags; __u8 gen; __u8 oldest_gen; @@ -70,7 +70,7 @@ struct bch_alloc_v4 { __u32 stripe; __u32 nr_external_backpointers; /* end of fields in original version of alloc_v4 */ - __u64 _fragmentation_lru; /* obsolete */ + __u64 journal_seq_empty; __u32 stripe_sectors; __u32 pad; } __packed __aligned(8); diff --git a/fs/bcachefs/alloc_foreground.c b/fs/bcachefs/alloc_foreground.c index 372178c8d416..6df41c331a52 100644 --- a/fs/bcachefs/alloc_foreground.c +++ b/fs/bcachefs/alloc_foreground.c @@ -107,14 +107,10 @@ void __bch2_open_bucket_put(struct bch_fs *c, struct open_bucket *ob) return; } - percpu_down_read(&c->mark_lock); spin_lock(&ob->lock); - ob->valid = false; ob->data_type = 0; - spin_unlock(&ob->lock); - percpu_up_read(&c->mark_lock); spin_lock(&c->freelist_lock); bch2_open_bucket_hash_remove(c, ob); @@ -156,6 +152,14 @@ static struct open_bucket *bch2_open_bucket_alloc(struct bch_fs *c) return ob; } +static inline bool is_superblock_bucket(struct bch_fs *c, struct bch_dev *ca, u64 b) +{ + if (c->curr_recovery_pass > BCH_RECOVERY_PASS_trans_mark_dev_sbs) + return false; + + return bch2_is_superblock_bucket(ca, b); +} + static void open_bucket_free_unused(struct bch_fs *c, struct open_bucket *ob) { BUG_ON(c->open_buckets_partial_nr >= @@ -175,20 +179,6 @@ static void open_bucket_free_unused(struct bch_fs *c, struct open_bucket *ob) closure_wake_up(&c->freelist_wait); } -/* _only_ for allocating the journal on a new device: */ -long bch2_bucket_alloc_new_fs(struct bch_dev *ca) -{ - while (ca->new_fs_bucket_idx < ca->mi.nbuckets) { - u64 b = ca->new_fs_bucket_idx++; - - if (!is_superblock_bucket(ca, b) && - (!ca->buckets_nouse || !test_bit(b, ca->buckets_nouse))) - return b; - } - - return -1; -} - static inline unsigned open_buckets_reserved(enum bch_watermark watermark) { switch (watermark) { @@ -206,33 +196,40 @@ static inline unsigned open_buckets_reserved(enum bch_watermark watermark) } } -static struct open_bucket *__try_alloc_bucket(struct bch_fs *c, struct bch_dev *ca, - u64 bucket, - enum bch_watermark watermark, - const struct bch_alloc_v4 *a, - struct bucket_alloc_state *s, - struct closure *cl) +static inline bool may_alloc_bucket(struct bch_fs *c, + struct bpos bucket, + struct bucket_alloc_state *s) { - struct open_bucket *ob; - - if (unlikely(ca->buckets_nouse && test_bit(bucket, ca->buckets_nouse))) { - s->skipped_nouse++; - return NULL; - } - - if (bch2_bucket_is_open(c, ca->dev_idx, bucket)) { + if (bch2_bucket_is_open(c, bucket.inode, bucket.offset)) { s->skipped_open++; - return NULL; + return false; } if (bch2_bucket_needs_journal_commit(&c->buckets_waiting_for_journal, - c->journal.flushed_seq_ondisk, ca->dev_idx, bucket)) { + c->journal.flushed_seq_ondisk, bucket.inode, bucket.offset)) { s->skipped_need_journal_commit++; - return NULL; + return false; } - if (bch2_bucket_nocow_is_locked(&c->nocow_locks, POS(ca->dev_idx, bucket))) { + if (bch2_bucket_nocow_is_locked(&c->nocow_locks, bucket)) { s->skipped_nocow++; + return false; + } + + return true; +} + +static struct open_bucket *__try_alloc_bucket(struct bch_fs *c, struct bch_dev *ca, + u64 bucket, u8 gen, + enum bch_watermark watermark, + struct bucket_alloc_state *s, + struct closure *cl) +{ + if (unlikely(is_superblock_bucket(c, ca, bucket))) + return NULL; + + if (unlikely(ca->buckets_nouse && test_bit(bucket, ca->buckets_nouse))) { + s->skipped_nouse++; return NULL; } @@ -254,14 +251,13 @@ static struct open_bucket *__try_alloc_bucket(struct bch_fs *c, struct bch_dev * return NULL; } - ob = bch2_open_bucket_alloc(c); + struct open_bucket *ob = bch2_open_bucket_alloc(c); spin_lock(&ob->lock); - ob->valid = true; ob->sectors_free = ca->mi.bucket_size; ob->dev = ca->dev_idx; - ob->gen = a->gen; + ob->gen = gen; ob->bucket = bucket; spin_unlock(&ob->lock); @@ -276,111 +272,29 @@ static struct open_bucket *__try_alloc_bucket(struct bch_fs *c, struct bch_dev * } static struct open_bucket *try_alloc_bucket(struct btree_trans *trans, struct bch_dev *ca, - enum bch_watermark watermark, u64 free_entry, + enum bch_watermark watermark, struct bucket_alloc_state *s, - struct bkey_s_c freespace_k, + struct btree_iter *freespace_iter, struct closure *cl) { struct bch_fs *c = trans->c; - struct btree_iter iter = { NULL }; - struct bkey_s_c k; - struct open_bucket *ob; - struct bch_alloc_v4 a_convert; - const struct bch_alloc_v4 *a; - u64 b = free_entry & ~(~0ULL << 56); - unsigned genbits = free_entry >> 56; - struct printbuf buf = PRINTBUF; - int ret; - - if (b < ca->mi.first_bucket || b >= ca->mi.nbuckets) { - prt_printf(&buf, "freespace btree has bucket outside allowed range %u-%llu\n" - " freespace key ", - ca->mi.first_bucket, ca->mi.nbuckets); - bch2_bkey_val_to_text(&buf, c, freespace_k); - bch2_trans_inconsistent(trans, "%s", buf.buf); - ob = ERR_PTR(-EIO); - goto err; - } + u64 b = freespace_iter->pos.offset & ~(~0ULL << 56); - k = bch2_bkey_get_iter(trans, &iter, - BTREE_ID_alloc, POS(ca->dev_idx, b), - BTREE_ITER_cached); - ret = bkey_err(k); - if (ret) { - ob = ERR_PTR(ret); - goto err; - } - - a = bch2_alloc_to_v4(k, &a_convert); - - if (a->data_type != BCH_DATA_free) { - if (c->curr_recovery_pass <= BCH_RECOVERY_PASS_check_alloc_info) { - ob = NULL; - goto err; - } - - prt_printf(&buf, "non free bucket in freespace btree\n" - " freespace key "); - bch2_bkey_val_to_text(&buf, c, freespace_k); - prt_printf(&buf, "\n "); - bch2_bkey_val_to_text(&buf, c, k); - bch2_trans_inconsistent(trans, "%s", buf.buf); - ob = ERR_PTR(-EIO); - goto err; - } - - if (genbits != (alloc_freespace_genbits(*a) >> 56) && - c->curr_recovery_pass > BCH_RECOVERY_PASS_check_alloc_info) { - prt_printf(&buf, "bucket in freespace btree with wrong genbits (got %u should be %llu)\n" - " freespace key ", - genbits, alloc_freespace_genbits(*a) >> 56); - bch2_bkey_val_to_text(&buf, c, freespace_k); - prt_printf(&buf, "\n "); - bch2_bkey_val_to_text(&buf, c, k); - bch2_trans_inconsistent(trans, "%s", buf.buf); - ob = ERR_PTR(-EIO); - goto err; - } - - if (c->curr_recovery_pass <= BCH_RECOVERY_PASS_check_extents_to_backpointers) { - struct bch_backpointer bp; - struct bpos bp_pos = POS_MIN; - - ret = bch2_get_next_backpointer(trans, ca, POS(ca->dev_idx, b), -1, - &bp_pos, &bp, - BTREE_ITER_nopreserve); - if (ret) { - ob = ERR_PTR(ret); - goto err; - } + if (!may_alloc_bucket(c, POS(ca->dev_idx, b), s)) + return NULL; - if (!bkey_eq(bp_pos, POS_MAX)) { - /* - * Bucket may have data in it - we don't call - * bc2h_trans_inconnsistent() because fsck hasn't - * finished yet - */ - ob = NULL; - goto err; - } - } + u8 gen; + int ret = bch2_check_discard_freespace_key(trans, freespace_iter, &gen, true); + if (ret < 0) + return ERR_PTR(ret); + if (ret) + return NULL; - ob = __try_alloc_bucket(c, ca, b, watermark, a, s, cl); - if (!ob) - bch2_set_btree_iter_dontneed(&iter); -err: - if (iter.path) - bch2_set_btree_iter_dontneed(&iter); - bch2_trans_iter_exit(trans, &iter); - printbuf_exit(&buf); - return ob; + return __try_alloc_bucket(c, ca, b, gen, watermark, s, cl); } /* * This path is for before the freespace btree is initialized: - * - * If ca->new_fs_bucket_idx is nonzero, we haven't yet marked superblock & - * journal buckets - journal buckets will be < ca->new_fs_bucket_idx */ static noinline struct open_bucket * bch2_bucket_alloc_early(struct btree_trans *trans, @@ -389,10 +303,11 @@ bch2_bucket_alloc_early(struct btree_trans *trans, struct bucket_alloc_state *s, struct closure *cl) { + struct bch_fs *c = trans->c; struct btree_iter iter, citer; struct bkey_s_c k, ck; struct open_bucket *ob = NULL; - u64 first_bucket = max_t(u64, ca->mi.first_bucket, ca->new_fs_bucket_idx); + u64 first_bucket = ca->mi.first_bucket; u64 *dev_alloc_cursor = &ca->alloc_cursor[s->btree_bitmap]; u64 alloc_start = max(first_bucket, *dev_alloc_cursor); u64 alloc_cursor = alloc_start; @@ -415,10 +330,6 @@ again: if (bkey_ge(k.k->p, POS(ca->dev_idx, ca->mi.nbuckets))) break; - if (ca->new_fs_bucket_idx && - is_superblock_bucket(ca, k.k->p.offset)) - continue; - if (s->btree_bitmap != BTREE_BITMAP_ANY && s->btree_bitmap != bch2_dev_btree_bitmap_marked_sectors(ca, bucket_to_sector(ca, bucket), ca->mi.bucket_size)) { @@ -452,7 +363,10 @@ again: s->buckets_seen++; - ob = __try_alloc_bucket(trans->c, ca, k.k->p.offset, watermark, a, s, cl); + ob = may_alloc_bucket(c, k.k->p, s) + ? __try_alloc_bucket(c, ca, k.k->p.offset, a->gen, + watermark, s, cl) + : NULL; next: bch2_set_btree_iter_dontneed(&citer); bch2_trans_iter_exit(trans, &citer); @@ -489,20 +403,21 @@ static struct open_bucket *bch2_bucket_alloc_freelist(struct btree_trans *trans, u64 alloc_start = max_t(u64, ca->mi.first_bucket, READ_ONCE(*dev_alloc_cursor)); u64 alloc_cursor = alloc_start; int ret; - - BUG_ON(ca->new_fs_bucket_idx); again: - for_each_btree_key_norestart(trans, iter, BTREE_ID_freespace, - POS(ca->dev_idx, alloc_cursor), 0, k, ret) { - if (k.k->p.inode != ca->dev_idx) - break; + for_each_btree_key_max_norestart(trans, iter, BTREE_ID_freespace, + POS(ca->dev_idx, alloc_cursor), + POS(ca->dev_idx, U64_MAX), + 0, k, ret) { + /* + * peek normally dosen't trim extents - they can span iter.pos, + * which is not what we want here: + */ + iter.k.size = iter.k.p.offset - iter.pos.offset; - for (alloc_cursor = max(alloc_cursor, bkey_start_offset(k.k)); - alloc_cursor < k.k->p.offset; - alloc_cursor++) { + while (iter.k.size) { s->buckets_seen++; - u64 bucket = alloc_cursor & ~(~0ULL << 56); + u64 bucket = iter.pos.offset & ~(~0ULL << 56); if (s->btree_bitmap != BTREE_BITMAP_ANY && s->btree_bitmap != bch2_dev_btree_bitmap_marked_sectors(ca, bucket_to_sector(ca, bucket), ca->mi.bucket_size)) { @@ -511,32 +426,36 @@ again: goto fail; bucket = sector_to_bucket(ca, - round_up(bucket_to_sector(ca, bucket) + 1, + round_up(bucket_to_sector(ca, bucket + 1), 1ULL << ca->mi.btree_bitmap_shift)); - u64 genbits = alloc_cursor >> 56; - alloc_cursor = bucket | (genbits << 56); + alloc_cursor = bucket|(iter.pos.offset & (~0ULL << 56)); - if (alloc_cursor > k.k->p.offset) - bch2_btree_iter_set_pos(&iter, POS(ca->dev_idx, alloc_cursor)); + bch2_btree_iter_set_pos(&iter, POS(ca->dev_idx, alloc_cursor)); s->skipped_mi_btree_bitmap++; - continue; + goto next; } - ob = try_alloc_bucket(trans, ca, watermark, - alloc_cursor, s, k, cl); + ob = try_alloc_bucket(trans, ca, watermark, s, &iter, cl); if (ob) { + if (!IS_ERR(ob)) + *dev_alloc_cursor = iter.pos.offset; bch2_set_btree_iter_dontneed(&iter); break; } - } + iter.k.size--; + iter.pos.offset++; + } +next: if (ob || ret) break; } fail: bch2_trans_iter_exit(trans, &iter); - if (!ob && ret) + BUG_ON(ob && ret); + + if (ret) ob = ERR_PTR(ret); if (!ob && alloc_start > ca->mi.first_bucket) { @@ -544,8 +463,6 @@ fail: goto again; } - *dev_alloc_cursor = alloc_cursor; - return ob; } @@ -595,6 +512,7 @@ static noinline void trace_bucket_alloc2(struct bch_fs *c, struct bch_dev *ca, * @watermark: how important is this allocation? * @data_type: BCH_DATA_journal, btree, user... * @cl: if not NULL, closure to be used to wait if buckets not available + * @nowait: if true, do not wait for buckets to become available * @usage: for secondarily also returning the current device usage * * Returns: an open_bucket on success, or an ERR_PTR() on failure. @@ -629,6 +547,10 @@ again: bch2_dev_do_invalidates(ca); if (!avail) { + if (watermark > BCH_WATERMARK_normal && + c->curr_recovery_pass <= BCH_RECOVERY_PASS_check_allocations) + goto alloc; + if (cl && !waiting) { closure_wait(&c->freelist_wait, cl); waiting = true; @@ -711,9 +633,9 @@ struct dev_alloc_list bch2_dev_alloc_list(struct bch_fs *c, unsigned i; for_each_set_bit(i, devs->d, BCH_SB_MEMBERS_MAX) - ret.devs[ret.nr++] = i; + ret.data[ret.nr++] = i; - bubble_sort(ret.devs, ret.nr, dev_stripe_cmp); + bubble_sort(ret.data, ret.nr, dev_stripe_cmp); return ret; } @@ -785,18 +707,13 @@ int bch2_bucket_alloc_set_trans(struct btree_trans *trans, struct closure *cl) { struct bch_fs *c = trans->c; - struct dev_alloc_list devs_sorted = - bch2_dev_alloc_list(c, stripe, devs_may_alloc); int ret = -BCH_ERR_insufficient_devices; BUG_ON(*nr_effective >= nr_replicas); - for (unsigned i = 0; i < devs_sorted.nr; i++) { - struct bch_dev_usage usage; - struct open_bucket *ob; - - unsigned dev = devs_sorted.devs[i]; - struct bch_dev *ca = bch2_dev_tryget_noerror(c, dev); + struct dev_alloc_list devs_sorted = bch2_dev_alloc_list(c, stripe, devs_may_alloc); + darray_for_each(devs_sorted, i) { + struct bch_dev *ca = bch2_dev_tryget_noerror(c, *i); if (!ca) continue; @@ -805,8 +722,9 @@ int bch2_bucket_alloc_set_trans(struct btree_trans *trans, continue; } - ob = bch2_bucket_alloc_trans(trans, ca, watermark, data_type, - cl, flags & BCH_WRITE_ALLOC_NOWAIT, &usage); + struct bch_dev_usage usage; + struct open_bucket *ob = bch2_bucket_alloc_trans(trans, ca, watermark, data_type, + cl, flags & BCH_WRITE_ALLOC_NOWAIT, &usage); if (!IS_ERR(ob)) bch2_dev_stripe_increment_inlined(ca, stripe, &usage); bch2_dev_put(ca); @@ -850,10 +768,6 @@ static int bucket_alloc_from_stripe(struct btree_trans *trans, struct closure *cl) { struct bch_fs *c = trans->c; - struct dev_alloc_list devs_sorted; - struct ec_stripe_head *h; - struct open_bucket *ob; - unsigned i, ec_idx; int ret = 0; if (nr_replicas < 2) @@ -862,34 +776,32 @@ static int bucket_alloc_from_stripe(struct btree_trans *trans, if (ec_open_bucket(c, ptrs)) return 0; - h = bch2_ec_stripe_head_get(trans, target, 0, nr_replicas - 1, watermark, cl); + struct ec_stripe_head *h = + bch2_ec_stripe_head_get(trans, target, 0, nr_replicas - 1, watermark, cl); if (IS_ERR(h)) return PTR_ERR(h); if (!h) return 0; - devs_sorted = bch2_dev_alloc_list(c, &wp->stripe, devs_may_alloc); - - for (i = 0; i < devs_sorted.nr; i++) - for (ec_idx = 0; ec_idx < h->s->nr_data; ec_idx++) { + struct dev_alloc_list devs_sorted = bch2_dev_alloc_list(c, &wp->stripe, devs_may_alloc); + darray_for_each(devs_sorted, i) + for (unsigned ec_idx = 0; ec_idx < h->s->nr_data; ec_idx++) { if (!h->s->blocks[ec_idx]) continue; - ob = c->open_buckets + h->s->blocks[ec_idx]; - if (ob->dev == devs_sorted.devs[i] && - !test_and_set_bit(ec_idx, h->s->blocks_allocated)) - goto got_bucket; + struct open_bucket *ob = c->open_buckets + h->s->blocks[ec_idx]; + if (ob->dev == *i && !test_and_set_bit(ec_idx, h->s->blocks_allocated)) { + ob->ec_idx = ec_idx; + ob->ec = h->s; + ec_stripe_new_get(h->s, STRIPE_REF_io); + + ret = add_new_bucket(c, ptrs, devs_may_alloc, + nr_replicas, nr_effective, + have_cache, ob); + goto out; + } } - goto out_put_head; -got_bucket: - ob->ec_idx = ec_idx; - ob->ec = h->s; - ec_stripe_new_get(h->s, STRIPE_REF_io); - - ret = add_new_bucket(c, ptrs, devs_may_alloc, - nr_replicas, nr_effective, - have_cache, ob); -out_put_head: +out: bch2_ec_stripe_head_put(c, h); return ret; } diff --git a/fs/bcachefs/alloc_foreground.h b/fs/bcachefs/alloc_foreground.h index 1a16fd5bd4f8..f25481a0d1a0 100644 --- a/fs/bcachefs/alloc_foreground.h +++ b/fs/bcachefs/alloc_foreground.h @@ -20,7 +20,7 @@ void bch2_reset_alloc_cursors(struct bch_fs *); struct dev_alloc_list { unsigned nr; - u8 devs[BCH_SB_MEMBERS_MAX]; + u8 data[BCH_SB_MEMBERS_MAX]; }; struct dev_alloc_list bch2_dev_alloc_list(struct bch_fs *, @@ -28,8 +28,6 @@ struct dev_alloc_list bch2_dev_alloc_list(struct bch_fs *, struct bch_devs_mask *); void bch2_dev_stripe_increment(struct bch_dev *, struct dev_stripe_state *); -long bch2_bucket_alloc_new_fs(struct bch_dev *); - static inline struct bch_dev *ob_dev(struct bch_fs *c, struct open_bucket *ob) { return bch2_dev_have_ref(c, ob->dev); diff --git a/fs/bcachefs/backpointers.c b/fs/bcachefs/backpointers.c index 654a58132a4d..ebeb6a5ff9d2 100644 --- a/fs/bcachefs/backpointers.c +++ b/fs/bcachefs/backpointers.c @@ -14,42 +14,8 @@ #include <linux/mm.h> -static bool extent_matches_bp(struct bch_fs *c, - enum btree_id btree_id, unsigned level, - struct bkey_s_c k, - struct bpos bucket, - struct bch_backpointer bp) -{ - struct bkey_ptrs_c ptrs = bch2_bkey_ptrs_c(k); - const union bch_extent_entry *entry; - struct extent_ptr_decoded p; - - rcu_read_lock(); - bkey_for_each_ptr_decode(k.k, ptrs, p, entry) { - struct bpos bucket2; - struct bch_backpointer bp2; - - if (p.ptr.cached) - continue; - - struct bch_dev *ca = bch2_dev_rcu(c, p.ptr.dev); - if (!ca) - continue; - - bch2_extent_ptr_to_bp(c, ca, btree_id, level, k, p, entry, &bucket2, &bp2); - if (bpos_eq(bucket, bucket2) && - !memcmp(&bp, &bp2, sizeof(bp))) { - rcu_read_unlock(); - return true; - } - } - rcu_read_unlock(); - - return false; -} - int bch2_backpointer_validate(struct bch_fs *c, struct bkey_s_c k, - enum bch_validate_flags flags) + struct bkey_validate_context from) { struct bkey_s_c_backpointer bp = bkey_s_c_to_backpointer(k); int ret = 0; @@ -59,67 +25,70 @@ int bch2_backpointer_validate(struct bch_fs *c, struct bkey_s_c k, "backpointer level bad: %u >= %u", bp.v->level, BTREE_MAX_DEPTH); - rcu_read_lock(); - struct bch_dev *ca = bch2_dev_rcu_noerror(c, bp.k->p.inode); - if (!ca) { - /* these will be caught by fsck */ - rcu_read_unlock(); - return 0; - } - - struct bpos bucket = bp_pos_to_bucket(ca, bp.k->p); - struct bpos bp_pos = bucket_pos_to_bp_noerror(ca, bucket, bp.v->bucket_offset); - rcu_read_unlock(); - - bkey_fsck_err_on((bp.v->bucket_offset >> MAX_EXTENT_COMPRESS_RATIO_SHIFT) >= ca->mi.bucket_size || - !bpos_eq(bp.k->p, bp_pos), - c, backpointer_bucket_offset_wrong, - "backpointer bucket_offset wrong"); + bkey_fsck_err_on(bp.k->p.inode == BCH_SB_MEMBER_INVALID, + c, backpointer_dev_bad, + "backpointer for BCH_SB_MEMBER_INVALID"); fsck_err: return ret; } -void bch2_backpointer_to_text(struct printbuf *out, const struct bch_backpointer *bp) +void bch2_backpointer_to_text(struct printbuf *out, struct bch_fs *c, struct bkey_s_c k) { - prt_printf(out, "btree=%s l=%u offset=%llu:%u len=%u pos=", - bch2_btree_id_str(bp->btree_id), - bp->level, - (u64) (bp->bucket_offset >> MAX_EXTENT_COMPRESS_RATIO_SHIFT), - (u32) bp->bucket_offset & ~(~0U << MAX_EXTENT_COMPRESS_RATIO_SHIFT), - bp->bucket_len); - bch2_bpos_to_text(out, bp->pos); -} + struct bkey_s_c_backpointer bp = bkey_s_c_to_backpointer(k); -void bch2_backpointer_k_to_text(struct printbuf *out, struct bch_fs *c, struct bkey_s_c k) -{ rcu_read_lock(); - struct bch_dev *ca = bch2_dev_rcu_noerror(c, k.k->p.inode); + struct bch_dev *ca = bch2_dev_rcu_noerror(c, bp.k->p.inode); if (ca) { - struct bpos bucket = bp_pos_to_bucket(ca, k.k->p); + u32 bucket_offset; + struct bpos bucket = bp_pos_to_bucket_and_offset(ca, bp.k->p, &bucket_offset); rcu_read_unlock(); - prt_str(out, "bucket="); - bch2_bpos_to_text(out, bucket); - prt_str(out, " "); + prt_printf(out, "bucket=%llu:%llu:%u ", bucket.inode, bucket.offset, bucket_offset); } else { rcu_read_unlock(); + prt_printf(out, "sector=%llu:%llu ", bp.k->p.inode, bp.k->p.offset >> MAX_EXTENT_COMPRESS_RATIO_SHIFT); } - bch2_backpointer_to_text(out, bkey_s_c_to_backpointer(k).v); + bch2_btree_id_level_to_text(out, bp.v->btree_id, bp.v->level); + prt_printf(out, " suboffset=%u len=%u gen=%u pos=", + (u32) bp.k->p.offset & ~(~0U << MAX_EXTENT_COMPRESS_RATIO_SHIFT), + bp.v->bucket_len, + bp.v->bucket_gen); + bch2_bpos_to_text(out, bp.v->pos); } void bch2_backpointer_swab(struct bkey_s k) { struct bkey_s_backpointer bp = bkey_s_to_backpointer(k); - bp.v->bucket_offset = swab40(bp.v->bucket_offset); bp.v->bucket_len = swab32(bp.v->bucket_len); bch2_bpos_swab(&bp.v->pos); } +static bool extent_matches_bp(struct bch_fs *c, + enum btree_id btree_id, unsigned level, + struct bkey_s_c k, + struct bkey_s_c_backpointer bp) +{ + struct bkey_ptrs_c ptrs = bch2_bkey_ptrs_c(k); + const union bch_extent_entry *entry; + struct extent_ptr_decoded p; + + bkey_for_each_ptr_decode(k.k, ptrs, p, entry) { + struct bkey_i_backpointer bp2; + bch2_extent_ptr_to_bp(c, btree_id, level, k, p, entry, &bp2); + + if (bpos_eq(bp.k->p, bp2.k.p) && + !memcmp(bp.v, &bp2.v, sizeof(bp2.v))) + return true; + } + + return false; +} + static noinline int backpointer_mod_err(struct btree_trans *trans, - struct bch_backpointer bp, - struct bkey_s_c bp_k, struct bkey_s_c orig_k, + struct bkey_i_backpointer *new_bp, + struct bkey_s_c found_bp, bool insert) { struct bch_fs *c = trans->c; @@ -127,12 +96,12 @@ static noinline int backpointer_mod_err(struct btree_trans *trans, if (insert) { prt_printf(&buf, "existing backpointer found when inserting "); - bch2_backpointer_to_text(&buf, &bp); + bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(&new_bp->k_i)); prt_newline(&buf); printbuf_indent_add(&buf, 2); prt_printf(&buf, "found "); - bch2_bkey_val_to_text(&buf, c, bp_k); + bch2_bkey_val_to_text(&buf, c, found_bp); prt_newline(&buf); prt_printf(&buf, "for "); @@ -144,11 +113,11 @@ static noinline int backpointer_mod_err(struct btree_trans *trans, printbuf_indent_add(&buf, 2); prt_printf(&buf, "searching for "); - bch2_backpointer_to_text(&buf, &bp); + bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(&new_bp->k_i)); prt_newline(&buf); prt_printf(&buf, "got "); - bch2_bkey_val_to_text(&buf, c, bp_k); + bch2_bkey_val_to_text(&buf, c, found_bp); prt_newline(&buf); prt_printf(&buf, "for "); @@ -167,161 +136,118 @@ static noinline int backpointer_mod_err(struct btree_trans *trans, } int bch2_bucket_backpointer_mod_nowritebuffer(struct btree_trans *trans, - struct bch_dev *ca, - struct bpos bucket, - struct bch_backpointer bp, struct bkey_s_c orig_k, + struct bkey_i_backpointer *bp, bool insert) { struct btree_iter bp_iter; - struct bkey_s_c k; - struct bkey_i_backpointer *bp_k; - int ret; - - bp_k = bch2_trans_kmalloc_nomemzero(trans, sizeof(struct bkey_i_backpointer)); - ret = PTR_ERR_OR_ZERO(bp_k); - if (ret) - return ret; - - bkey_backpointer_init(&bp_k->k_i); - bp_k->k.p = bucket_pos_to_bp(ca, bucket, bp.bucket_offset); - bp_k->v = bp; - - if (!insert) { - bp_k->k.type = KEY_TYPE_deleted; - set_bkey_val_u64s(&bp_k->k, 0); - } - - k = bch2_bkey_get_iter(trans, &bp_iter, BTREE_ID_backpointers, - bp_k->k.p, + struct bkey_s_c k = bch2_bkey_get_iter(trans, &bp_iter, BTREE_ID_backpointers, + bp->k.p, BTREE_ITER_intent| BTREE_ITER_slots| BTREE_ITER_with_updates); - ret = bkey_err(k); + int ret = bkey_err(k); if (ret) - goto err; + return ret; if (insert ? k.k->type : (k.k->type != KEY_TYPE_backpointer || - memcmp(bkey_s_c_to_backpointer(k).v, &bp, sizeof(bp)))) { - ret = backpointer_mod_err(trans, bp, k, orig_k, insert); + memcmp(bkey_s_c_to_backpointer(k).v, &bp->v, sizeof(bp->v)))) { + ret = backpointer_mod_err(trans, orig_k, bp, k, insert); if (ret) goto err; } - ret = bch2_trans_update(trans, &bp_iter, &bp_k->k_i, 0); + if (!insert) { + bp->k.type = KEY_TYPE_deleted; + set_bkey_val_u64s(&bp->k, 0); + } + + ret = bch2_trans_update(trans, &bp_iter, &bp->k_i, 0); err: bch2_trans_iter_exit(trans, &bp_iter); return ret; } -/* - * Find the next backpointer >= *bp_offset: - */ -int bch2_get_next_backpointer(struct btree_trans *trans, - struct bch_dev *ca, - struct bpos bucket, int gen, - struct bpos *bp_pos, - struct bch_backpointer *bp, - unsigned iter_flags) +static int bch2_backpointer_del(struct btree_trans *trans, struct bpos pos) { - struct bpos bp_end_pos = bucket_pos_to_bp(ca, bpos_nosnap_successor(bucket), 0); - struct btree_iter alloc_iter = { NULL }, bp_iter = { NULL }; - struct bkey_s_c k; - int ret = 0; - - if (bpos_ge(*bp_pos, bp_end_pos)) - goto done; - - if (gen >= 0) { - k = bch2_bkey_get_iter(trans, &alloc_iter, BTREE_ID_alloc, - bucket, BTREE_ITER_cached|iter_flags); - ret = bkey_err(k); - if (ret) - goto out; - - if (k.k->type != KEY_TYPE_alloc_v4 || - bkey_s_c_to_alloc_v4(k).v->gen != gen) - goto done; - } - - *bp_pos = bpos_max(*bp_pos, bucket_pos_to_bp(ca, bucket, 0)); - - for_each_btree_key_norestart(trans, bp_iter, BTREE_ID_backpointers, - *bp_pos, iter_flags, k, ret) { - if (bpos_ge(k.k->p, bp_end_pos)) - break; + return (likely(!bch2_backpointers_no_use_write_buffer) + ? bch2_btree_delete_at_buffered(trans, BTREE_ID_backpointers, pos) + : bch2_btree_delete(trans, BTREE_ID_backpointers, pos, 0)) ?: + bch2_trans_commit(trans, NULL, NULL, BCH_TRANS_COMMIT_no_enospc); +} - *bp_pos = k.k->p; - *bp = *bkey_s_c_to_backpointer(k).v; - goto out; - } -done: - *bp_pos = SPOS_MAX; -out: - bch2_trans_iter_exit(trans, &bp_iter); - bch2_trans_iter_exit(trans, &alloc_iter); - return ret; +static inline int bch2_backpointers_maybe_flush(struct btree_trans *trans, + struct bkey_s_c visiting_k, + struct bkey_buf *last_flushed) +{ + return likely(!bch2_backpointers_no_use_write_buffer) + ? bch2_btree_write_buffer_maybe_flush(trans, visiting_k, last_flushed) + : 0; } -static void backpointer_not_found(struct btree_trans *trans, - struct bpos bp_pos, - struct bch_backpointer bp, - struct bkey_s_c k) +static int backpointer_target_not_found(struct btree_trans *trans, + struct bkey_s_c_backpointer bp, + struct bkey_s_c target_k, + struct bkey_buf *last_flushed) { struct bch_fs *c = trans->c; struct printbuf buf = PRINTBUF; + int ret = 0; /* * If we're using the btree write buffer, the backpointer we were * looking at may have already been deleted - failure to find what it * pointed to is not an error: */ - if (likely(!bch2_backpointers_no_use_write_buffer)) - return; - - struct bpos bucket; - if (!bp_pos_to_bucket_nodev(c, bp_pos, &bucket)) - return; + ret = last_flushed + ? bch2_backpointers_maybe_flush(trans, bp.s_c, last_flushed) + : 0; + if (ret) + return ret; prt_printf(&buf, "backpointer doesn't match %s it points to:\n ", - bp.level ? "btree node" : "extent"); - prt_printf(&buf, "bucket: "); - bch2_bpos_to_text(&buf, bucket); - prt_printf(&buf, "\n "); + bp.v->level ? "btree node" : "extent"); + bch2_bkey_val_to_text(&buf, c, bp.s_c); - prt_printf(&buf, "backpointer pos: "); - bch2_bpos_to_text(&buf, bp_pos); prt_printf(&buf, "\n "); + bch2_bkey_val_to_text(&buf, c, target_k); - bch2_backpointer_to_text(&buf, &bp); - prt_printf(&buf, "\n "); - bch2_bkey_val_to_text(&buf, c, k); - if (c->curr_recovery_pass >= BCH_RECOVERY_PASS_check_extents_to_backpointers) - bch_err_ratelimited(c, "%s", buf.buf); - else - bch2_trans_inconsistent(trans, "%s", buf.buf); + struct bkey_ptrs_c ptrs = bch2_bkey_ptrs_c(target_k); + const union bch_extent_entry *entry; + struct extent_ptr_decoded p; + bkey_for_each_ptr_decode(target_k.k, ptrs, p, entry) + if (p.ptr.dev == bp.k->p.inode) { + prt_printf(&buf, "\n "); + struct bkey_i_backpointer bp2; + bch2_extent_ptr_to_bp(c, bp.v->btree_id, bp.v->level, target_k, p, entry, &bp2); + bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(&bp2.k_i)); + } + if (fsck_err(trans, backpointer_to_missing_ptr, + "%s", buf.buf)) + ret = bch2_backpointer_del(trans, bp.k->p); +fsck_err: printbuf_exit(&buf); + return ret; } struct bkey_s_c bch2_backpointer_get_key(struct btree_trans *trans, + struct bkey_s_c_backpointer bp, struct btree_iter *iter, - struct bpos bp_pos, - struct bch_backpointer bp, - unsigned iter_flags) + unsigned iter_flags, + struct bkey_buf *last_flushed) { - if (likely(!bp.level)) { - struct bch_fs *c = trans->c; + struct bch_fs *c = trans->c; - struct bpos bucket; - if (!bp_pos_to_bucket_nodev(c, bp_pos, &bucket)) - return bkey_s_c_err(-EIO); + if (unlikely(bp.v->btree_id >= btree_id_nr_alive(c))) + return bkey_s_c_null; + if (likely(!bp.v->level)) { bch2_trans_node_iter_init(trans, iter, - bp.btree_id, - bp.pos, + bp.v->btree_id, + bp.v->pos, 0, 0, iter_flags); struct bkey_s_c k = bch2_btree_iter_peek_slot(iter); @@ -330,67 +256,64 @@ struct bkey_s_c bch2_backpointer_get_key(struct btree_trans *trans, return k; } - if (k.k && extent_matches_bp(c, bp.btree_id, bp.level, k, bucket, bp)) + if (k.k && + extent_matches_bp(c, bp.v->btree_id, bp.v->level, k, bp)) return k; bch2_trans_iter_exit(trans, iter); - backpointer_not_found(trans, bp_pos, bp, k); - return bkey_s_c_null; + int ret = backpointer_target_not_found(trans, bp, k, last_flushed); + return ret ? bkey_s_c_err(ret) : bkey_s_c_null; } else { - struct btree *b = bch2_backpointer_get_node(trans, iter, bp_pos, bp); + struct btree *b = bch2_backpointer_get_node(trans, bp, iter, last_flushed); + if (IS_ERR_OR_NULL(b)) + return ((struct bkey_s_c) { .k = ERR_CAST(b) }); - if (IS_ERR_OR_NULL(b)) { - bch2_trans_iter_exit(trans, iter); - return IS_ERR(b) ? bkey_s_c_err(PTR_ERR(b)) : bkey_s_c_null; - } return bkey_i_to_s_c(&b->key); } } struct btree *bch2_backpointer_get_node(struct btree_trans *trans, + struct bkey_s_c_backpointer bp, struct btree_iter *iter, - struct bpos bp_pos, - struct bch_backpointer bp) + struct bkey_buf *last_flushed) { struct bch_fs *c = trans->c; - BUG_ON(!bp.level); - - struct bpos bucket; - if (!bp_pos_to_bucket_nodev(c, bp_pos, &bucket)) - return ERR_PTR(-EIO); + BUG_ON(!bp.v->level); bch2_trans_node_iter_init(trans, iter, - bp.btree_id, - bp.pos, + bp.v->btree_id, + bp.v->pos, 0, - bp.level - 1, + bp.v->level - 1, 0); struct btree *b = bch2_btree_iter_peek_node(iter); if (IS_ERR_OR_NULL(b)) goto err; - BUG_ON(b->c.level != bp.level - 1); + BUG_ON(b->c.level != bp.v->level - 1); - if (extent_matches_bp(c, bp.btree_id, bp.level, - bkey_i_to_s_c(&b->key), - bucket, bp)) + if (extent_matches_bp(c, bp.v->btree_id, bp.v->level, + bkey_i_to_s_c(&b->key), bp)) return b; if (btree_node_will_make_reachable(b)) { b = ERR_PTR(-BCH_ERR_backpointer_to_overwritten_btree_node); } else { - backpointer_not_found(trans, bp_pos, bp, bkey_i_to_s_c(&b->key)); - b = NULL; + int ret = backpointer_target_not_found(trans, bp, bkey_i_to_s_c(&b->key), last_flushed); + b = ret ? ERR_PTR(ret) : NULL; } err: bch2_trans_iter_exit(trans, iter); return b; } -static int bch2_check_btree_backpointer(struct btree_trans *trans, struct btree_iter *bp_iter, - struct bkey_s_c k) +static int bch2_check_backpointer_has_valid_bucket(struct btree_trans *trans, struct bkey_s_c k, + struct bkey_buf *last_flushed) { + if (k.k->type != KEY_TYPE_backpointer) + return 0; + struct bch_fs *c = trans->c; struct btree_iter alloc_iter = { NULL }; struct bkey_s_c alloc_k; @@ -399,10 +322,14 @@ static int bch2_check_btree_backpointer(struct btree_trans *trans, struct btree_ struct bpos bucket; if (!bp_pos_to_bucket_nodev_noerror(c, k.k->p, &bucket)) { + ret = bch2_backpointers_maybe_flush(trans, k, last_flushed); + if (ret) + goto out; + if (fsck_err(trans, backpointer_to_missing_device, "backpointer for missing device:\n%s", (bch2_bkey_val_to_text(&buf, c, k), buf.buf))) - ret = bch2_btree_delete_at(trans, bp_iter, 0); + ret = bch2_backpointer_del(trans, k.k->p); goto out; } @@ -411,13 +338,16 @@ static int bch2_check_btree_backpointer(struct btree_trans *trans, struct btree_ if (ret) goto out; - if (fsck_err_on(alloc_k.k->type != KEY_TYPE_alloc_v4, - trans, backpointer_to_missing_alloc, - "backpointer for nonexistent alloc key: %llu:%llu:0\n%s", - alloc_iter.pos.inode, alloc_iter.pos.offset, - (bch2_bkey_val_to_text(&buf, c, k), buf.buf))) { - ret = bch2_btree_delete_at(trans, bp_iter, 0); - goto out; + if (alloc_k.k->type != KEY_TYPE_alloc_v4) { + ret = bch2_backpointers_maybe_flush(trans, k, last_flushed); + if (ret) + goto out; + + if (fsck_err(trans, backpointer_to_missing_alloc, + "backpointer for nonexistent alloc key: %llu:%llu:0\n%s", + alloc_iter.pos.inode, alloc_iter.pos.offset, + (bch2_bkey_val_to_text(&buf, c, k), buf.buf))) + ret = bch2_backpointer_del(trans, k.k->p); } out: fsck_err: @@ -429,18 +359,24 @@ fsck_err: /* verify that every backpointer has a corresponding alloc key */ int bch2_check_btree_backpointers(struct bch_fs *c) { + struct bkey_buf last_flushed; + bch2_bkey_buf_init(&last_flushed); + bkey_init(&last_flushed.k->k); + int ret = bch2_trans_run(c, for_each_btree_key_commit(trans, iter, BTREE_ID_backpointers, POS_MIN, 0, k, NULL, NULL, BCH_TRANS_COMMIT_no_enospc, - bch2_check_btree_backpointer(trans, &iter, k))); + bch2_check_backpointer_has_valid_bucket(trans, k, &last_flushed))); + + bch2_bkey_buf_exit(&last_flushed, c); bch_err_fn(c, ret); return ret; } struct extents_to_bp_state { - struct bpos bucket_start; - struct bpos bucket_end; + struct bpos bp_start; + struct bpos bp_end; struct bkey_buf last_flushed; }; @@ -501,9 +437,13 @@ found: goto err; prt_str(&buf, "extents pointing to same space, but first extent checksum bad:"); - prt_printf(&buf, "\n %s ", bch2_btree_id_str(btree)); + prt_printf(&buf, "\n "); + bch2_btree_id_to_text(&buf, btree); + prt_str(&buf, " "); bch2_bkey_val_to_text(&buf, c, extent); - prt_printf(&buf, "\n %s ", bch2_btree_id_str(o_btree)); + prt_printf(&buf, "\n "); + bch2_btree_id_to_text(&buf, o_btree); + prt_str(&buf, " "); bch2_bkey_val_to_text(&buf, c, extent2); struct nonce nonce = extent_nonce(extent.k->bversion, p.crc); @@ -524,41 +464,25 @@ err: static int check_bp_exists(struct btree_trans *trans, struct extents_to_bp_state *s, - struct bpos bucket, - struct bch_backpointer bp, + struct bkey_i_backpointer *bp, struct bkey_s_c orig_k) { struct bch_fs *c = trans->c; - struct btree_iter bp_iter = {}; struct btree_iter other_extent_iter = {}; struct printbuf buf = PRINTBUF; - struct bkey_s_c bp_k; - int ret = 0; - struct bch_dev *ca = bch2_dev_bucket_tryget(c, bucket); - if (!ca) { - prt_str(&buf, "extent for nonexistent device:bucket "); - bch2_bpos_to_text(&buf, bucket); - prt_str(&buf, "\n "); - bch2_bkey_val_to_text(&buf, c, orig_k); - bch_err(c, "%s", buf.buf); - ret = -BCH_ERR_fsck_repair_unimplemented; - goto err; - } - - if (bpos_lt(bucket, s->bucket_start) || - bpos_gt(bucket, s->bucket_end)) - goto out; + if (bpos_lt(bp->k.p, s->bp_start) || + bpos_gt(bp->k.p, s->bp_end)) + return 0; - bp_k = bch2_bkey_get_iter(trans, &bp_iter, BTREE_ID_backpointers, - bucket_pos_to_bp(ca, bucket, bp.bucket_offset), - 0); - ret = bkey_err(bp_k); + struct btree_iter bp_iter; + struct bkey_s_c bp_k = bch2_bkey_get_iter(trans, &bp_iter, BTREE_ID_backpointers, bp->k.p, 0); + int ret = bkey_err(bp_k); if (ret) goto err; if (bp_k.k->type != KEY_TYPE_backpointer || - memcmp(bkey_s_c_to_backpointer(bp_k).v, &bp, sizeof(bp))) { + memcmp(bkey_s_c_to_backpointer(bp_k).v, &bp->v, sizeof(bp->v))) { ret = bch2_btree_write_buffer_maybe_flush(trans, orig_k, &s->last_flushed); if (ret) goto err; @@ -570,7 +494,6 @@ err: fsck_err: bch2_trans_iter_exit(trans, &other_extent_iter); bch2_trans_iter_exit(trans, &bp_iter); - bch2_dev_put(ca); printbuf_exit(&buf); return ret; check_existing_bp: @@ -578,10 +501,10 @@ check_existing_bp: if (bp_k.k->type != KEY_TYPE_backpointer) goto missing; - struct bch_backpointer other_bp = *bkey_s_c_to_backpointer(bp_k).v; + struct bkey_s_c_backpointer other_bp = bkey_s_c_to_backpointer(bp_k); struct bkey_s_c other_extent = - bch2_backpointer_get_key(trans, &other_extent_iter, bp_k.k->p, other_bp, 0); + bch2_backpointer_get_key(trans, other_bp, &other_extent_iter, 0, NULL); ret = bkey_err(other_extent); if (ret == -BCH_ERR_backpointer_to_overwritten_btree_node) ret = 0; @@ -600,19 +523,23 @@ check_existing_bp: bch_err(c, "%s", buf.buf); if (other_extent.k->size <= orig_k.k->size) { - ret = drop_dev_and_update(trans, other_bp.btree_id, other_extent, bucket.inode); + ret = drop_dev_and_update(trans, other_bp.v->btree_id, + other_extent, bp->k.p.inode); if (ret) goto err; goto out; } else { - ret = drop_dev_and_update(trans, bp.btree_id, orig_k, bucket.inode); + ret = drop_dev_and_update(trans, bp->v.btree_id, orig_k, bp->k.p.inode); if (ret) goto err; goto missing; } } - ret = check_extent_checksum(trans, other_bp.btree_id, other_extent, bp.btree_id, orig_k, bucket.inode); + ret = check_extent_checksum(trans, + other_bp.v->btree_id, other_extent, + bp->v.btree_id, orig_k, + bp->k.p.inode); if (ret < 0) goto err; if (ret) { @@ -620,7 +547,8 @@ check_existing_bp: goto missing; } - ret = check_extent_checksum(trans, bp.btree_id, orig_k, other_bp.btree_id, other_extent, bucket.inode); + ret = check_extent_checksum(trans, bp->v.btree_id, orig_k, + other_bp.v->btree_id, other_extent, bp->k.p.inode); if (ret < 0) goto err; if (ret) { @@ -629,7 +557,7 @@ check_existing_bp: } printbuf_reset(&buf); - prt_printf(&buf, "duplicate extents pointing to same space on dev %llu\n ", bucket.inode); + prt_printf(&buf, "duplicate extents pointing to same space on dev %llu\n ", bp->k.p.inode); bch2_bkey_val_to_text(&buf, c, orig_k); prt_str(&buf, "\n "); bch2_bkey_val_to_text(&buf, c, other_extent); @@ -638,21 +566,15 @@ check_existing_bp: goto err; missing: printbuf_reset(&buf); - prt_printf(&buf, "missing backpointer for btree=%s l=%u ", - bch2_btree_id_str(bp.btree_id), bp.level); + prt_str(&buf, "missing backpointer\n for: "); bch2_bkey_val_to_text(&buf, c, orig_k); - prt_printf(&buf, "\n got: "); + prt_printf(&buf, "\n want: "); + bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(&bp->k_i)); + prt_printf(&buf, "\n got: "); bch2_bkey_val_to_text(&buf, c, bp_k); - struct bkey_i_backpointer n_bp_k; - bkey_backpointer_init(&n_bp_k.k_i); - n_bp_k.k.p = bucket_pos_to_bp(ca, bucket, bp.bucket_offset); - n_bp_k.v = bp; - prt_printf(&buf, "\n want: "); - bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(&n_bp_k.k_i)); - if (fsck_err(trans, ptr_to_missing_backpointer, "%s", buf.buf)) - ret = bch2_bucket_backpointer_mod(trans, ca, bucket, bp, orig_k, true); + ret = bch2_bucket_backpointer_mod(trans, orig_k, bp, true); goto out; } @@ -663,31 +585,33 @@ static int check_extent_to_backpointers(struct btree_trans *trans, struct bkey_s_c k) { struct bch_fs *c = trans->c; - struct bkey_ptrs_c ptrs; + struct bkey_ptrs_c ptrs = bch2_bkey_ptrs_c(k); const union bch_extent_entry *entry; struct extent_ptr_decoded p; - int ret; - ptrs = bch2_bkey_ptrs_c(k); bkey_for_each_ptr_decode(k.k, ptrs, p, entry) { - struct bpos bucket_pos = POS_MIN; - struct bch_backpointer bp; - if (p.ptr.cached) continue; + if (p.ptr.dev == BCH_SB_MEMBER_INVALID) + continue; + rcu_read_lock(); struct bch_dev *ca = bch2_dev_rcu_noerror(c, p.ptr.dev); - if (ca) - bch2_extent_ptr_to_bp(c, ca, btree, level, k, p, entry, &bucket_pos, &bp); + bool check = ca && test_bit(PTR_BUCKET_NR(ca, &p.ptr), ca->bucket_backpointer_mismatches); + bool empty = ca && test_bit(PTR_BUCKET_NR(ca, &p.ptr), ca->bucket_backpointer_empty); rcu_read_unlock(); - if (!ca) - continue; + if (check || empty) { + struct bkey_i_backpointer bp; + bch2_extent_ptr_to_bp(c, btree, level, k, p, entry, &bp); - ret = check_bp_exists(trans, s, bucket_pos, bp, k); - if (ret) - return ret; + int ret = check + ? check_bp_exists(trans, s, &bp, k) + : bch2_bucket_backpointer_mod(trans, k, &bp, true); + if (ret) + return ret; + } } return 0; @@ -896,54 +820,330 @@ static int bch2_check_extents_to_backpointers_pass(struct btree_trans *trans, return 0; } +enum alloc_sector_counter { + ALLOC_dirty, + ALLOC_cached, + ALLOC_stripe, + ALLOC_SECTORS_NR +}; + +static enum alloc_sector_counter data_type_to_alloc_counter(enum bch_data_type t) +{ + switch (t) { + case BCH_DATA_btree: + case BCH_DATA_user: + return ALLOC_dirty; + case BCH_DATA_cached: + return ALLOC_cached; + case BCH_DATA_stripe: + return ALLOC_stripe; + default: + BUG(); + } +} + +static int check_bucket_backpointers_to_extents(struct btree_trans *, struct bch_dev *, struct bpos); + +static int check_bucket_backpointer_mismatch(struct btree_trans *trans, struct bkey_s_c alloc_k, + struct bkey_buf *last_flushed) +{ + struct bch_fs *c = trans->c; + struct bch_alloc_v4 a_convert; + const struct bch_alloc_v4 *a = bch2_alloc_to_v4(alloc_k, &a_convert); + bool need_commit = false; + + if (a->data_type == BCH_DATA_sb || + a->data_type == BCH_DATA_journal || + a->data_type == BCH_DATA_parity) + return 0; + + u32 sectors[ALLOC_SECTORS_NR]; + memset(sectors, 0, sizeof(sectors)); + + struct bch_dev *ca = bch2_dev_bucket_tryget_noerror(trans->c, alloc_k.k->p); + if (!ca) + return 0; + + struct btree_iter iter; + struct bkey_s_c bp_k; + int ret = 0; + for_each_btree_key_max_norestart(trans, iter, BTREE_ID_backpointers, + bucket_pos_to_bp_start(ca, alloc_k.k->p), + bucket_pos_to_bp_end(ca, alloc_k.k->p), 0, bp_k, ret) { + if (bp_k.k->type != KEY_TYPE_backpointer) + continue; + + struct bkey_s_c_backpointer bp = bkey_s_c_to_backpointer(bp_k); + + if (c->sb.version_upgrade_complete >= bcachefs_metadata_version_backpointer_bucket_gen && + (bp.v->bucket_gen != a->gen || + bp.v->pad)) { + ret = bch2_backpointer_del(trans, bp_k.k->p); + if (ret) + break; + + need_commit = true; + continue; + } + + if (bp.v->bucket_gen != a->gen) + continue; + + sectors[data_type_to_alloc_counter(bp.v->data_type)] += bp.v->bucket_len; + }; + bch2_trans_iter_exit(trans, &iter); + if (ret) + goto err; + + if (need_commit) { + ret = bch2_trans_commit(trans, NULL, NULL, BCH_TRANS_COMMIT_no_enospc); + if (ret) + goto err; + } + + /* Cached pointers don't have backpointers: */ + + if (sectors[ALLOC_dirty] != a->dirty_sectors || + sectors[ALLOC_stripe] != a->stripe_sectors) { + if (c->sb.version_upgrade_complete >= bcachefs_metadata_version_backpointer_bucket_gen) { + ret = bch2_backpointers_maybe_flush(trans, alloc_k, last_flushed); + if (ret) + goto err; + } + + if (sectors[ALLOC_dirty] > a->dirty_sectors || + sectors[ALLOC_stripe] > a->stripe_sectors) { + ret = check_bucket_backpointers_to_extents(trans, ca, alloc_k.k->p) ?: + -BCH_ERR_transaction_restart_nested; + goto err; + } + + if (!sectors[ALLOC_dirty] && + !sectors[ALLOC_stripe]) + __set_bit(alloc_k.k->p.offset, ca->bucket_backpointer_empty); + else + __set_bit(alloc_k.k->p.offset, ca->bucket_backpointer_mismatches); + } +err: + bch2_dev_put(ca); + return ret; +} + +static bool backpointer_node_has_missing(struct bch_fs *c, struct bkey_s_c k) +{ + switch (k.k->type) { + case KEY_TYPE_btree_ptr_v2: { + bool ret = false; + + rcu_read_lock(); + struct bpos pos = bkey_s_c_to_btree_ptr_v2(k).v->min_key; + while (pos.inode <= k.k->p.inode) { + if (pos.inode >= c->sb.nr_devices) + break; + + struct bch_dev *ca = bch2_dev_rcu_noerror(c, pos.inode); + if (!ca) + goto next; + + struct bpos bucket = bp_pos_to_bucket(ca, pos); + bucket.offset = find_next_bit(ca->bucket_backpointer_mismatches, + ca->mi.nbuckets, bucket.offset); + if (bucket.offset == ca->mi.nbuckets) + goto next; + + ret = bpos_le(bucket_pos_to_bp_end(ca, bucket), k.k->p); + if (ret) + break; +next: + pos = SPOS(pos.inode + 1, 0, 0); + } + rcu_read_unlock(); + + return ret; + } + case KEY_TYPE_btree_ptr: + return true; + default: + return false; + } +} + +static int btree_node_get_and_pin(struct btree_trans *trans, struct bkey_i *k, + enum btree_id btree, unsigned level) +{ + struct btree_iter iter; + bch2_trans_node_iter_init(trans, &iter, btree, k->k.p, 0, level, 0); + struct btree *b = bch2_btree_iter_peek_node(&iter); + int ret = PTR_ERR_OR_ZERO(b); + if (ret) + goto err; + + if (b) + bch2_node_pin(trans->c, b); +err: + bch2_trans_iter_exit(trans, &iter); + return ret; +} + +static int bch2_pin_backpointer_nodes_with_missing(struct btree_trans *trans, + struct bpos start, struct bpos *end) +{ + struct bch_fs *c = trans->c; + int ret = 0; + + struct bkey_buf tmp; + bch2_bkey_buf_init(&tmp); + + bch2_btree_cache_unpin(c); + + *end = SPOS_MAX; + + s64 mem_may_pin = mem_may_pin_bytes(c); + struct btree_iter iter; + bch2_trans_node_iter_init(trans, &iter, BTREE_ID_backpointers, start, + 0, 1, BTREE_ITER_prefetch); + ret = for_each_btree_key_continue(trans, iter, 0, k, ({ + if (!backpointer_node_has_missing(c, k)) + continue; + + mem_may_pin -= c->opts.btree_node_size; + if (mem_may_pin <= 0) + break; + + bch2_bkey_buf_reassemble(&tmp, c, k); + struct btree_path *path = btree_iter_path(trans, &iter); + + BUG_ON(path->level != 1); + + bch2_btree_node_prefetch(trans, path, tmp.k, path->btree_id, path->level - 1); + })); + if (ret) + return ret; + + struct bpos pinned = SPOS_MAX; + mem_may_pin = mem_may_pin_bytes(c); + bch2_trans_node_iter_init(trans, &iter, BTREE_ID_backpointers, start, + 0, 1, BTREE_ITER_prefetch); + ret = for_each_btree_key_continue(trans, iter, 0, k, ({ + if (!backpointer_node_has_missing(c, k)) + continue; + + mem_may_pin -= c->opts.btree_node_size; + if (mem_may_pin <= 0) { + *end = pinned; + break; + } + + bch2_bkey_buf_reassemble(&tmp, c, k); + struct btree_path *path = btree_iter_path(trans, &iter); + + BUG_ON(path->level != 1); + + int ret2 = btree_node_get_and_pin(trans, tmp.k, path->btree_id, path->level - 1); + + if (!ret2) + pinned = tmp.k->k.p; + + ret; + })); + if (ret) + return ret; + + return ret; +} + int bch2_check_extents_to_backpointers(struct bch_fs *c) { + int ret = 0; + + /* + * Can't allow devices to come/go/resize while we have bucket bitmaps + * allocated + */ + lockdep_assert_held(&c->state_lock); + + for_each_member_device(c, ca) { + BUG_ON(ca->bucket_backpointer_mismatches); + ca->bucket_backpointer_mismatches = kvcalloc(BITS_TO_LONGS(ca->mi.nbuckets), + sizeof(unsigned long), + GFP_KERNEL); + ca->bucket_backpointer_empty = kvcalloc(BITS_TO_LONGS(ca->mi.nbuckets), + sizeof(unsigned long), + GFP_KERNEL); + if (!ca->bucket_backpointer_mismatches || + !ca->bucket_backpointer_empty) { + bch2_dev_put(ca); + ret = -BCH_ERR_ENOMEM_backpointer_mismatches_bitmap; + goto err_free_bitmaps; + } + } + struct btree_trans *trans = bch2_trans_get(c); - struct extents_to_bp_state s = { .bucket_start = POS_MIN }; - int ret; + struct extents_to_bp_state s = { .bp_start = POS_MIN }; bch2_bkey_buf_init(&s.last_flushed); bkey_init(&s.last_flushed.k->k); + ret = for_each_btree_key(trans, iter, BTREE_ID_alloc, + POS_MIN, BTREE_ITER_prefetch, k, ({ + check_bucket_backpointer_mismatch(trans, k, &s.last_flushed); + })); + if (ret) + goto err; + + u64 nr_buckets = 0, nr_mismatches = 0, nr_empty = 0; + for_each_member_device(c, ca) { + nr_buckets += ca->mi.nbuckets; + nr_mismatches += bitmap_weight(ca->bucket_backpointer_mismatches, ca->mi.nbuckets); + nr_empty += bitmap_weight(ca->bucket_backpointer_empty, ca->mi.nbuckets); + } + + if (!nr_mismatches && !nr_empty) + goto err; + + bch_info(c, "scanning for missing backpointers in %llu/%llu buckets", + nr_mismatches + nr_empty, nr_buckets); + while (1) { - struct bbpos end; - ret = bch2_get_btree_in_memory_pos(trans, - BIT_ULL(BTREE_ID_backpointers), - BIT_ULL(BTREE_ID_backpointers), - BBPOS(BTREE_ID_backpointers, s.bucket_start), &end); + ret = bch2_pin_backpointer_nodes_with_missing(trans, s.bp_start, &s.bp_end); if (ret) break; - s.bucket_end = end.pos; - - if ( bpos_eq(s.bucket_start, POS_MIN) && - !bpos_eq(s.bucket_end, SPOS_MAX)) + if ( bpos_eq(s.bp_start, POS_MIN) && + !bpos_eq(s.bp_end, SPOS_MAX)) bch_verbose(c, "%s(): alloc info does not fit in ram, running in multiple passes with %zu nodes per pass", __func__, btree_nodes_fit_in_ram(c)); - if (!bpos_eq(s.bucket_start, POS_MIN) || - !bpos_eq(s.bucket_end, SPOS_MAX)) { + if (!bpos_eq(s.bp_start, POS_MIN) || + !bpos_eq(s.bp_end, SPOS_MAX)) { struct printbuf buf = PRINTBUF; prt_str(&buf, "check_extents_to_backpointers(): "); - bch2_bpos_to_text(&buf, s.bucket_start); + bch2_bpos_to_text(&buf, s.bp_start); prt_str(&buf, "-"); - bch2_bpos_to_text(&buf, s.bucket_end); + bch2_bpos_to_text(&buf, s.bp_end); bch_verbose(c, "%s", buf.buf); printbuf_exit(&buf); } ret = bch2_check_extents_to_backpointers_pass(trans, &s); - if (ret || bpos_eq(s.bucket_end, SPOS_MAX)) + if (ret || bpos_eq(s.bp_end, SPOS_MAX)) break; - s.bucket_start = bpos_successor(s.bucket_end); + s.bp_start = bpos_successor(s.bp_end); } +err: bch2_trans_put(trans); bch2_bkey_buf_exit(&s.last_flushed, c); - bch2_btree_cache_unpin(c); +err_free_bitmaps: + for_each_member_device(c, ca) { + kvfree(ca->bucket_backpointer_empty); + ca->bucket_backpointer_empty = NULL; + kvfree(ca->bucket_backpointer_mismatches); + ca->bucket_backpointer_mismatches = NULL; + } bch_err_fn(c, ret); return ret; @@ -959,44 +1159,43 @@ static int check_one_backpointer(struct btree_trans *trans, return 0; struct bkey_s_c_backpointer bp = bkey_s_c_to_backpointer(bp_k); - struct bch_fs *c = trans->c; - struct btree_iter iter; struct bbpos pos = bp_to_bbpos(*bp.v); - struct bkey_s_c k; - struct printbuf buf = PRINTBUF; - int ret; if (bbpos_cmp(pos, start) < 0 || bbpos_cmp(pos, end) > 0) return 0; - k = bch2_backpointer_get_key(trans, &iter, bp.k->p, *bp.v, 0); - ret = bkey_err(k); + struct btree_iter iter; + struct bkey_s_c k = bch2_backpointer_get_key(trans, bp, &iter, 0, last_flushed); + int ret = bkey_err(k); if (ret == -BCH_ERR_backpointer_to_overwritten_btree_node) return 0; if (ret) return ret; - if (!k.k) { - ret = bch2_btree_write_buffer_maybe_flush(trans, bp.s_c, last_flushed); - if (ret) - goto out; - - if (fsck_err(trans, backpointer_to_missing_ptr, - "backpointer for missing %s\n %s", - bp.v->level ? "btree node" : "extent", - (bch2_bkey_val_to_text(&buf, c, bp.s_c), buf.buf))) { - ret = bch2_btree_delete_at_buffered(trans, BTREE_ID_backpointers, bp.k->p); - goto out; - } - } -out: -fsck_err: bch2_trans_iter_exit(trans, &iter); - printbuf_exit(&buf); return ret; } +static int check_bucket_backpointers_to_extents(struct btree_trans *trans, + struct bch_dev *ca, struct bpos bucket) +{ + u32 restart_count = trans->restart_count; + struct bkey_buf last_flushed; + bch2_bkey_buf_init(&last_flushed); + bkey_init(&last_flushed.k->k); + + int ret = for_each_btree_key_max(trans, iter, BTREE_ID_backpointers, + bucket_pos_to_bp_start(ca, bucket), + bucket_pos_to_bp_end(ca, bucket), + 0, k, + check_one_backpointer(trans, BBPOS_MIN, BBPOS_MAX, k, &last_flushed) + ); + + bch2_bkey_buf_exit(&last_flushed, trans->c); + return ret ?: trans_was_restarted(trans, restart_count); +} + static int bch2_check_backpointers_to_extents_pass(struct btree_trans *trans, struct bbpos start, struct bbpos end) @@ -1009,9 +1208,8 @@ static int bch2_check_backpointers_to_extents_pass(struct btree_trans *trans, bkey_init(&last_flushed.k->k); progress_init(&progress, trans->c, BIT_ULL(BTREE_ID_backpointers)); - int ret = for_each_btree_key_commit(trans, iter, BTREE_ID_backpointers, - POS_MIN, BTREE_ITER_prefetch, k, - NULL, NULL, BCH_TRANS_COMMIT_no_enospc, ({ + int ret = for_each_btree_key(trans, iter, BTREE_ID_backpointers, + POS_MIN, BTREE_ITER_prefetch, k, ({ progress_update_iter(trans, &progress, &iter, "backpointers_to_extents"); check_one_backpointer(trans, start, end, k, &last_flushed); })); diff --git a/fs/bcachefs/backpointers.h b/fs/bcachefs/backpointers.h index 3b29fdf519dd..060dad1521ee 100644 --- a/fs/bcachefs/backpointers.h +++ b/fs/bcachefs/backpointers.h @@ -18,14 +18,14 @@ static inline u64 swab40(u64 x) ((x & 0xff00000000ULL) >> 32)); } -int bch2_backpointer_validate(struct bch_fs *, struct bkey_s_c k, enum bch_validate_flags); -void bch2_backpointer_to_text(struct printbuf *, const struct bch_backpointer *); -void bch2_backpointer_k_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c); +int bch2_backpointer_validate(struct bch_fs *, struct bkey_s_c k, + struct bkey_validate_context); +void bch2_backpointer_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c); void bch2_backpointer_swab(struct bkey_s); #define bch2_bkey_ops_backpointer ((struct bkey_ops) { \ .key_validate = bch2_backpointer_validate, \ - .val_to_text = bch2_backpointer_k_to_text, \ + .val_to_text = bch2_backpointer_to_text, \ .swab = bch2_backpointer_swab, \ .min_val_size = 32, \ }) @@ -43,22 +43,24 @@ static inline struct bpos bp_pos_to_bucket(const struct bch_dev *ca, struct bpos return POS(bp_pos.inode, sector_to_bucket(ca, bucket_sector)); } +static inline struct bpos bp_pos_to_bucket_and_offset(const struct bch_dev *ca, struct bpos bp_pos, + u32 *bucket_offset) +{ + u64 bucket_sector = bp_pos.offset >> MAX_EXTENT_COMPRESS_RATIO_SHIFT; + + return POS(bp_pos.inode, sector_to_bucket_and_offset(ca, bucket_sector, bucket_offset)); +} + static inline bool bp_pos_to_bucket_nodev_noerror(struct bch_fs *c, struct bpos bp_pos, struct bpos *bucket) { rcu_read_lock(); - struct bch_dev *ca = bch2_dev_rcu(c, bp_pos.inode); + struct bch_dev *ca = bch2_dev_rcu_noerror(c, bp_pos.inode); if (ca) *bucket = bp_pos_to_bucket(ca, bp_pos); rcu_read_unlock(); return ca != NULL; } -static inline bool bp_pos_to_bucket_nodev(struct bch_fs *c, struct bpos bp_pos, struct bpos *bucket) -{ - return !bch2_fs_inconsistent_on(!bp_pos_to_bucket_nodev_noerror(c, bp_pos, bucket), - c, "backpointer for missing device %llu", bp_pos.inode); -} - static inline struct bpos bucket_pos_to_bp_noerror(const struct bch_dev *ca, struct bpos bucket, u64 bucket_offset) @@ -80,31 +82,35 @@ static inline struct bpos bucket_pos_to_bp(const struct bch_dev *ca, return ret; } -int bch2_bucket_backpointer_mod_nowritebuffer(struct btree_trans *, struct bch_dev *, - struct bpos bucket, struct bch_backpointer, struct bkey_s_c, bool); +static inline struct bpos bucket_pos_to_bp_start(const struct bch_dev *ca, struct bpos bucket) +{ + return bucket_pos_to_bp(ca, bucket, 0); +} + +static inline struct bpos bucket_pos_to_bp_end(const struct bch_dev *ca, struct bpos bucket) +{ + return bpos_nosnap_predecessor(bucket_pos_to_bp(ca, bpos_nosnap_successor(bucket), 0)); +} + +int bch2_bucket_backpointer_mod_nowritebuffer(struct btree_trans *, + struct bkey_s_c, + struct bkey_i_backpointer *, + bool); static inline int bch2_bucket_backpointer_mod(struct btree_trans *trans, - struct bch_dev *ca, - struct bpos bucket, - struct bch_backpointer bp, struct bkey_s_c orig_k, + struct bkey_i_backpointer *bp, bool insert) { if (unlikely(bch2_backpointers_no_use_write_buffer)) - return bch2_bucket_backpointer_mod_nowritebuffer(trans, ca, bucket, bp, orig_k, insert); - - struct bkey_i_backpointer bp_k; - - bkey_backpointer_init(&bp_k.k_i); - bp_k.k.p = bucket_pos_to_bp(ca, bucket, bp.bucket_offset); - bp_k.v = bp; + return bch2_bucket_backpointer_mod_nowritebuffer(trans, orig_k, bp, insert); if (!insert) { - bp_k.k.type = KEY_TYPE_deleted; - set_bkey_val_u64s(&bp_k.k, 0); + bp->k.type = KEY_TYPE_deleted; + set_bkey_val_u64s(&bp->k, 0); } - return bch2_trans_update_buffered(trans, BTREE_ID_backpointers, &bp_k.k_i); + return bch2_trans_update_buffered(trans, BTREE_ID_backpointers, &bp->k_i); } static inline enum bch_data_type bch2_bkey_ptr_data_type(struct bkey_s_c k, @@ -134,44 +140,29 @@ static inline enum bch_data_type bch2_bkey_ptr_data_type(struct bkey_s_c k, } } -static inline void __bch2_extent_ptr_to_bp(struct bch_fs *c, struct bch_dev *ca, +static inline void bch2_extent_ptr_to_bp(struct bch_fs *c, enum btree_id btree_id, unsigned level, struct bkey_s_c k, struct extent_ptr_decoded p, const union bch_extent_entry *entry, - struct bpos *bucket_pos, struct bch_backpointer *bp, - u64 sectors) + struct bkey_i_backpointer *bp) { - u32 bucket_offset; - *bucket_pos = PTR_BUCKET_POS_OFFSET(ca, &p.ptr, &bucket_offset); - *bp = (struct bch_backpointer) { + bkey_backpointer_init(&bp->k_i); + bp->k.p = POS(p.ptr.dev, ((u64) p.ptr.offset << MAX_EXTENT_COMPRESS_RATIO_SHIFT) + p.crc.offset); + bp->v = (struct bch_backpointer) { .btree_id = btree_id, .level = level, .data_type = bch2_bkey_ptr_data_type(k, p, entry), - .bucket_offset = ((u64) bucket_offset << MAX_EXTENT_COMPRESS_RATIO_SHIFT) + - p.crc.offset, - .bucket_len = sectors, + .bucket_gen = p.ptr.gen, + .bucket_len = ptr_disk_sectors(level ? btree_sectors(c) : k.k->size, p), .pos = k.k->p, }; } -static inline void bch2_extent_ptr_to_bp(struct bch_fs *c, struct bch_dev *ca, - enum btree_id btree_id, unsigned level, - struct bkey_s_c k, struct extent_ptr_decoded p, - const union bch_extent_entry *entry, - struct bpos *bucket_pos, struct bch_backpointer *bp) -{ - u64 sectors = ptr_disk_sectors(level ? btree_sectors(c) : k.k->size, p); - - __bch2_extent_ptr_to_bp(c, ca, btree_id, level, k, p, entry, bucket_pos, bp, sectors); -} - -int bch2_get_next_backpointer(struct btree_trans *, struct bch_dev *ca, struct bpos, int, - struct bpos *, struct bch_backpointer *, unsigned); -struct bkey_s_c bch2_backpointer_get_key(struct btree_trans *, struct btree_iter *, - struct bpos, struct bch_backpointer, - unsigned); -struct btree *bch2_backpointer_get_node(struct btree_trans *, struct btree_iter *, - struct bpos, struct bch_backpointer); +struct bkey_buf; +struct bkey_s_c bch2_backpointer_get_key(struct btree_trans *, struct bkey_s_c_backpointer, + struct btree_iter *, unsigned, struct bkey_buf *); +struct btree *bch2_backpointer_get_node(struct btree_trans *, struct bkey_s_c_backpointer, + struct btree_iter *, struct bkey_buf *); int bch2_check_btree_backpointers(struct bch_fs *); int bch2_check_extents_to_backpointers(struct bch_fs *); diff --git a/fs/bcachefs/bbpos.h b/fs/bcachefs/bbpos.h index be2edced5213..63abe17f35ea 100644 --- a/fs/bcachefs/bbpos.h +++ b/fs/bcachefs/bbpos.h @@ -29,7 +29,7 @@ static inline struct bbpos bbpos_successor(struct bbpos pos) static inline void bch2_bbpos_to_text(struct printbuf *out, struct bbpos pos) { - prt_str(out, bch2_btree_id_str(pos.btree)); + bch2_btree_id_to_text(out, pos.btree); prt_char(out, ':'); bch2_bpos_to_text(out, pos.pos); } diff --git a/fs/bcachefs/bcachefs.h b/fs/bcachefs/bcachefs.h index e94a83b8113e..161cf2f05d2a 100644 --- a/fs/bcachefs/bcachefs.h +++ b/fs/bcachefs/bcachefs.h @@ -205,6 +205,7 @@ #include <linux/zstd.h> #include "bcachefs_format.h" +#include "btree_journal_iter_types.h" #include "disk_accounting_types.h" #include "errcode.h" #include "fifo.h" @@ -293,6 +294,8 @@ do { \ #define bch_info(c, fmt, ...) \ bch2_print(c, KERN_INFO bch2_fmt(c, fmt), ##__VA_ARGS__) +#define bch_info_ratelimited(c, fmt, ...) \ + bch2_print_ratelimited(c, KERN_INFO bch2_fmt(c, fmt), ##__VA_ARGS__) #define bch_notice(c, fmt, ...) \ bch2_print(c, KERN_NOTICE bch2_fmt(c, fmt), ##__VA_ARGS__) #define bch_warn(c, fmt, ...) \ @@ -352,6 +355,12 @@ do { \ bch_info(c, fmt, ##__VA_ARGS__); \ } while (0) +#define bch_verbose_ratelimited(c, fmt, ...) \ +do { \ + if ((c)->opts.verbose) \ + bch_info_ratelimited(c, fmt, ##__VA_ARGS__); \ +} while (0) + #define pr_verbose_init(opts, fmt, ...) \ do { \ if (opt_get(opts, verbose)) \ @@ -538,20 +547,20 @@ struct bch_dev { /* * Buckets: - * Per-bucket arrays are protected by c->mark_lock, bucket_lock and - * gc_gens_lock, for device resize - holding any is sufficient for - * access: Or rcu_read_lock(), but only for dev_ptr_stale(): + * Per-bucket arrays are protected by either rcu_read_lock or + * state_lock, for device resize. */ GENRADIX(struct bucket) buckets_gc; struct bucket_gens __rcu *bucket_gens; u8 *oldest_gen; unsigned long *buckets_nouse; - struct rw_semaphore bucket_lock; + + unsigned long *bucket_backpointer_mismatches; + unsigned long *bucket_backpointer_empty; struct bch_dev_usage __percpu *usage; /* Allocator: */ - u64 new_fs_bucket_idx; u64 alloc_cursor[3]; unsigned nr_open_buckets; @@ -606,6 +615,7 @@ struct bch_dev { x(going_ro) \ x(write_disable_complete) \ x(clean_shutdown) \ + x(recovery_running) \ x(fsck_running) \ x(initial_gc_unfixed) \ x(need_delete_dead_snapshots) \ @@ -650,28 +660,6 @@ struct journal_seq_blacklist_table { } entries[]; }; -struct journal_keys { - /* must match layout in darray_types.h */ - size_t nr, size; - struct journal_key { - u64 journal_seq; - u32 journal_offset; - enum btree_id btree_id:8; - unsigned level:8; - bool allocated; - bool overwritten; - struct bkey_i *k; - } *data; - /* - * Gap buffer: instead of all the empty space in the array being at the - * end of the buffer - from @nr to @size - the empty space is at @gap. - * This means that sequential insertions are O(n) instead of O(n^2). - */ - size_t gap; - atomic_t ref; - bool initial_ref_held; -}; - struct btree_trans_buf { struct btree_trans *trans; }; @@ -680,6 +668,7 @@ struct btree_trans_buf { ((subvol_inum) { BCACHEFS_ROOT_SUBVOL, BCACHEFS_ROOT_INO }) #define BCH_WRITE_REFS() \ + x(journal) \ x(trans) \ x(write) \ x(promote) \ @@ -692,6 +681,7 @@ struct btree_trans_buf { x(dio_write) \ x(discard) \ x(discard_fast) \ + x(check_discard_freespace_key) \ x(invalidate) \ x(delete_dead_snapshots) \ x(gc_gens) \ @@ -735,6 +725,12 @@ struct bch_fs { struct percpu_ref writes; #endif /* + * Certain operations are only allowed in single threaded mode, during + * recovery, and we want to assert that this is the case: + */ + struct task_struct *recovery_task; + + /* * Analagous to c->writes, for asynchronous ops that don't necessarily * need fs to be read-write */ @@ -764,6 +760,8 @@ struct bch_fs { __uuid_t user_uuid; u16 version; + u16 version_incompat; + u16 version_incompat_allowed; u16 version_min; u16 version_upgrade_complete; @@ -834,9 +832,10 @@ struct bch_fs { struct work_struct btree_interior_update_work; struct workqueue_struct *btree_node_rewrite_worker; - - struct list_head pending_node_rewrites; - struct mutex pending_node_rewrites_lock; + struct list_head btree_node_rewrites; + struct list_head btree_node_rewrites_pending; + spinlock_t btree_node_rewrites_lock; + struct closure_waitlist btree_node_rewrites_wait; /* btree_io.c: */ spinlock_t btree_write_error_lock; @@ -967,8 +966,7 @@ struct bch_fs { struct rhashtable promote_table; mempool_t compression_bounce[2]; - mempool_t compress_workspace[BCH_COMPRESSION_TYPE_NR]; - mempool_t decompress_workspace; + mempool_t compress_workspace[BCH_COMPRESSION_OPT_NR]; size_t zstd_workspace_size; struct crypto_shash *sha256; @@ -1027,6 +1025,7 @@ struct bch_fs { struct list_head vfs_inodes_list; struct mutex vfs_inodes_lock; struct rhashtable vfs_inodes_table; + struct rhltable vfs_inodes_by_inum_table; /* VFS IO PATH - fs-io.c */ struct bio_set writepage_bioset; @@ -1048,10 +1047,12 @@ struct bch_fs { * for signaling to the toplevel code which pass we want to run now. */ enum bch_recovery_pass curr_recovery_pass; + enum bch_recovery_pass next_recovery_pass; /* bitmask of recovery passes that we actually ran */ u64 recovery_passes_complete; /* never rewinds version of curr_recovery_pass */ enum bch_recovery_pass recovery_pass_done; + spinlock_t recovery_pass_lock; struct semaphore online_fsck_mutex; /* DEBUG JUNK */ @@ -1062,9 +1063,6 @@ struct bch_fs { struct btree_node *verify_ondisk; struct mutex verify_lock; - u64 *unused_inode_hints; - unsigned inode_shard_bits; - /* * A btree node on disk could have too many bsets for an iterator to fit * on the stack - have to dynamically allocate them @@ -1086,8 +1084,6 @@ struct bch_fs { u64 counters_on_mount[BCH_COUNTER_NR]; u64 __percpu *counters; - unsigned copy_gc_enabled:1; - struct bch2_time_stats times[BCH_TIME_STAT_NR]; struct btree_transaction_stats btree_transaction_stats[BCH_TRANSACTIONS_NR]; diff --git a/fs/bcachefs/bcachefs_format.h b/fs/bcachefs/bcachefs_format.h index 5004f6ba997c..f70f0108401f 100644 --- a/fs/bcachefs/bcachefs_format.h +++ b/fs/bcachefs/bcachefs_format.h @@ -418,7 +418,8 @@ static inline void bkey_init(struct bkey *k) x(snapshot_tree, 31) \ x(logged_op_truncate, 32) \ x(logged_op_finsert, 33) \ - x(accounting, 34) + x(accounting, 34) \ + x(inode_alloc_cursor, 35) enum bch_bkey_type { #define x(name, nr) KEY_TYPE_##name = nr, @@ -463,7 +464,8 @@ struct bch_backpointer { __u8 btree_id; __u8 level; __u8 data_type; - __u64 bucket_offset:40; + __u8 bucket_gen; + __u32 pad; __u32 bucket_len; struct bpos pos; } __packed __aligned(8); @@ -499,8 +501,6 @@ struct bch_sb_field { #include "disk_groups_format.h" #include "extents_format.h" #include "ec_format.h" -#include "dirent_format.h" -#include "disk_groups_format.h" #include "inode_format.h" #include "journal_seq_blacklist_format.h" #include "logged_ops_format.h" @@ -679,7 +679,14 @@ struct bch_sb_field_ext { x(disk_accounting_v3, BCH_VERSION(1, 10)) \ x(disk_accounting_inum, BCH_VERSION(1, 11)) \ x(rebalance_work_acct_fix, BCH_VERSION(1, 12)) \ - x(inode_has_child_snapshots, BCH_VERSION(1, 13)) + x(inode_has_child_snapshots, BCH_VERSION(1, 13)) \ + x(backpointer_bucket_gen, BCH_VERSION(1, 14)) \ + x(disk_accounting_big_endian, BCH_VERSION(1, 15)) \ + x(reflink_p_may_update_opts, BCH_VERSION(1, 16)) \ + x(inode_depth, BCH_VERSION(1, 17)) \ + x(persistent_inode_cursors, BCH_VERSION(1, 18)) \ + x(autofix_errors, BCH_VERSION(1, 19)) \ + x(directory_size, BCH_VERSION(1, 20)) enum bcachefs_metadata_version { bcachefs_metadata_version_min = 9, @@ -844,6 +851,10 @@ LE64_BITMASK(BCH_SB_VERSION_UPGRADE_COMPLETE, struct bch_sb, flags[5], 0, 16); LE64_BITMASK(BCH_SB_ALLOCATOR_STUCK_TIMEOUT, struct bch_sb, flags[5], 16, 32); +LE64_BITMASK(BCH_SB_VERSION_INCOMPAT, struct bch_sb, flags[5], 32, 48); +LE64_BITMASK(BCH_SB_VERSION_INCOMPAT_ALLOWED, + struct bch_sb, flags[5], 48, 64); +LE64_BITMASK(BCH_SB_SHARD_INUMS_NBITS, struct bch_sb, flags[6], 0, 4); static inline __u64 BCH_SB_COMPRESSION_TYPE(const struct bch_sb *sb) { @@ -896,21 +907,22 @@ static inline void SET_BCH_SB_BACKGROUND_COMPRESSION_TYPE(struct bch_sb *sb, __u x(new_varint, 15) \ x(journal_no_flush, 16) \ x(alloc_v2, 17) \ - x(extents_across_btree_nodes, 18) + x(extents_across_btree_nodes, 18) \ + x(incompat_version_field, 19) #define BCH_SB_FEATURES_ALWAYS \ - ((1ULL << BCH_FEATURE_new_extent_overwrite)| \ - (1ULL << BCH_FEATURE_extents_above_btree_updates)|\ - (1ULL << BCH_FEATURE_btree_updates_journalled)|\ - (1ULL << BCH_FEATURE_alloc_v2)|\ - (1ULL << BCH_FEATURE_extents_across_btree_nodes)) + (BIT_ULL(BCH_FEATURE_new_extent_overwrite)| \ + BIT_ULL(BCH_FEATURE_extents_above_btree_updates)|\ + BIT_ULL(BCH_FEATURE_btree_updates_journalled)|\ + BIT_ULL(BCH_FEATURE_alloc_v2)|\ + BIT_ULL(BCH_FEATURE_extents_across_btree_nodes)) #define BCH_SB_FEATURES_ALL \ (BCH_SB_FEATURES_ALWAYS| \ - (1ULL << BCH_FEATURE_new_siphash)| \ - (1ULL << BCH_FEATURE_btree_ptr_v2)| \ - (1ULL << BCH_FEATURE_new_varint)| \ - (1ULL << BCH_FEATURE_journal_no_flush)) + BIT_ULL(BCH_FEATURE_new_siphash)| \ + BIT_ULL(BCH_FEATURE_btree_ptr_v2)| \ + BIT_ULL(BCH_FEATURE_new_varint)| \ + BIT_ULL(BCH_FEATURE_journal_no_flush)) enum bch_sb_feature { #define x(f, n) BCH_FEATURE_##f, @@ -1032,7 +1044,7 @@ static inline _Bool bch2_csum_type_is_encryption(enum bch_csum_type type) x(crc64, 2) \ x(xxhash, 3) -enum bch_csum_opts { +enum bch_csum_opt { #define x(t, n) BCH_CSUM_OPT_##t = n, BCH_CSUM_OPTS() #undef x @@ -1221,6 +1233,15 @@ struct jset_entry_log { u8 d[]; } __packed __aligned(8); +static inline unsigned jset_entry_log_msg_bytes(struct jset_entry_log *l) +{ + unsigned b = vstruct_bytes(&l->entry) - offsetof(struct jset_entry_log, d); + + while (b && !l->d[b - 1]) + --b; + return b; +} + struct jset_entry_datetime { struct jset_entry entry; __le64 seconds; @@ -1268,14 +1289,18 @@ LE32_BITMASK(JSET_NO_FLUSH, struct jset, flags, 5, 6); /* Btree: */ enum btree_id_flags { - BTREE_ID_EXTENTS = BIT(0), - BTREE_ID_SNAPSHOTS = BIT(1), - BTREE_ID_SNAPSHOT_FIELD = BIT(2), - BTREE_ID_DATA = BIT(3), + BTREE_IS_extents = BIT(0), + BTREE_IS_snapshots = BIT(1), + BTREE_IS_snapshot_field = BIT(2), + BTREE_IS_data = BIT(3), + BTREE_IS_write_buffer = BIT(4), }; #define BCH_BTREE_IDS() \ - x(extents, 0, BTREE_ID_EXTENTS|BTREE_ID_SNAPSHOTS|BTREE_ID_DATA,\ + x(extents, 0, \ + BTREE_IS_extents| \ + BTREE_IS_snapshots| \ + BTREE_IS_data, \ BIT_ULL(KEY_TYPE_whiteout)| \ BIT_ULL(KEY_TYPE_error)| \ BIT_ULL(KEY_TYPE_cookie)| \ @@ -1283,17 +1308,20 @@ enum btree_id_flags { BIT_ULL(KEY_TYPE_reservation)| \ BIT_ULL(KEY_TYPE_reflink_p)| \ BIT_ULL(KEY_TYPE_inline_data)) \ - x(inodes, 1, BTREE_ID_SNAPSHOTS, \ + x(inodes, 1, \ + BTREE_IS_snapshots, \ BIT_ULL(KEY_TYPE_whiteout)| \ BIT_ULL(KEY_TYPE_inode)| \ BIT_ULL(KEY_TYPE_inode_v2)| \ BIT_ULL(KEY_TYPE_inode_v3)| \ BIT_ULL(KEY_TYPE_inode_generation)) \ - x(dirents, 2, BTREE_ID_SNAPSHOTS, \ + x(dirents, 2, \ + BTREE_IS_snapshots, \ BIT_ULL(KEY_TYPE_whiteout)| \ BIT_ULL(KEY_TYPE_hash_whiteout)| \ BIT_ULL(KEY_TYPE_dirent)) \ - x(xattrs, 3, BTREE_ID_SNAPSHOTS, \ + x(xattrs, 3, \ + BTREE_IS_snapshots, \ BIT_ULL(KEY_TYPE_whiteout)| \ BIT_ULL(KEY_TYPE_cookie)| \ BIT_ULL(KEY_TYPE_hash_whiteout)| \ @@ -1307,7 +1335,9 @@ enum btree_id_flags { BIT_ULL(KEY_TYPE_quota)) \ x(stripes, 6, 0, \ BIT_ULL(KEY_TYPE_stripe)) \ - x(reflink, 7, BTREE_ID_EXTENTS|BTREE_ID_DATA, \ + x(reflink, 7, \ + BTREE_IS_extents| \ + BTREE_IS_data, \ BIT_ULL(KEY_TYPE_reflink_v)| \ BIT_ULL(KEY_TYPE_indirect_inline_data)| \ BIT_ULL(KEY_TYPE_error)) \ @@ -1315,28 +1345,38 @@ enum btree_id_flags { BIT_ULL(KEY_TYPE_subvolume)) \ x(snapshots, 9, 0, \ BIT_ULL(KEY_TYPE_snapshot)) \ - x(lru, 10, 0, \ + x(lru, 10, \ + BTREE_IS_write_buffer, \ BIT_ULL(KEY_TYPE_set)) \ - x(freespace, 11, BTREE_ID_EXTENTS, \ + x(freespace, 11, \ + BTREE_IS_extents, \ BIT_ULL(KEY_TYPE_set)) \ x(need_discard, 12, 0, \ BIT_ULL(KEY_TYPE_set)) \ - x(backpointers, 13, 0, \ + x(backpointers, 13, \ + BTREE_IS_write_buffer, \ BIT_ULL(KEY_TYPE_backpointer)) \ x(bucket_gens, 14, 0, \ BIT_ULL(KEY_TYPE_bucket_gens)) \ x(snapshot_trees, 15, 0, \ BIT_ULL(KEY_TYPE_snapshot_tree)) \ - x(deleted_inodes, 16, BTREE_ID_SNAPSHOT_FIELD, \ + x(deleted_inodes, 16, \ + BTREE_IS_snapshot_field| \ + BTREE_IS_write_buffer, \ BIT_ULL(KEY_TYPE_set)) \ x(logged_ops, 17, 0, \ BIT_ULL(KEY_TYPE_logged_op_truncate)| \ - BIT_ULL(KEY_TYPE_logged_op_finsert)) \ - x(rebalance_work, 18, BTREE_ID_SNAPSHOT_FIELD, \ + BIT_ULL(KEY_TYPE_logged_op_finsert)| \ + BIT_ULL(KEY_TYPE_inode_alloc_cursor)) \ + x(rebalance_work, 18, \ + BTREE_IS_snapshot_field| \ + BTREE_IS_write_buffer, \ BIT_ULL(KEY_TYPE_set)|BIT_ULL(KEY_TYPE_cookie)) \ x(subvolume_children, 19, 0, \ BIT_ULL(KEY_TYPE_set)) \ - x(accounting, 20, BTREE_ID_SNAPSHOT_FIELD, \ + x(accounting, 20, \ + BTREE_IS_snapshot_field| \ + BTREE_IS_write_buffer, \ BIT_ULL(KEY_TYPE_accounting)) \ enum btree_id { @@ -1361,6 +1401,8 @@ static inline bool btree_id_is_alloc(enum btree_id id) case BTREE_ID_need_discard: case BTREE_ID_freespace: case BTREE_ID_bucket_gens: + case BTREE_ID_lru: + case BTREE_ID_accounting: return true; default: return false; diff --git a/fs/bcachefs/bkey.h b/fs/bcachefs/bkey.h index 41df24a53d97..054e2d5e8448 100644 --- a/fs/bcachefs/bkey.h +++ b/fs/bcachefs/bkey.h @@ -9,13 +9,6 @@ #include "util.h" #include "vstructs.h" -enum bch_validate_flags { - BCH_VALIDATE_write = BIT(0), - BCH_VALIDATE_commit = BIT(1), - BCH_VALIDATE_journal = BIT(2), - BCH_VALIDATE_silent = BIT(3), -}; - #if 0 /* diff --git a/fs/bcachefs/bkey_methods.c b/fs/bcachefs/bkey_methods.c index e7ac227ba7e8..15c93576b5c2 100644 --- a/fs/bcachefs/bkey_methods.c +++ b/fs/bcachefs/bkey_methods.c @@ -28,7 +28,7 @@ const char * const bch2_bkey_types[] = { }; static int deleted_key_validate(struct bch_fs *c, struct bkey_s_c k, - enum bch_validate_flags flags) + struct bkey_validate_context from) { return 0; } @@ -42,7 +42,7 @@ static int deleted_key_validate(struct bch_fs *c, struct bkey_s_c k, }) static int empty_val_key_validate(struct bch_fs *c, struct bkey_s_c k, - enum bch_validate_flags flags) + struct bkey_validate_context from) { int ret = 0; @@ -59,7 +59,7 @@ fsck_err: }) static int key_type_cookie_validate(struct bch_fs *c, struct bkey_s_c k, - enum bch_validate_flags flags) + struct bkey_validate_context from) { return 0; } @@ -83,7 +83,7 @@ static void key_type_cookie_to_text(struct printbuf *out, struct bch_fs *c, }) static int key_type_inline_data_validate(struct bch_fs *c, struct bkey_s_c k, - enum bch_validate_flags flags) + struct bkey_validate_context from) { return 0; } @@ -124,7 +124,7 @@ const struct bkey_ops bch2_bkey_null_ops = { }; int bch2_bkey_val_validate(struct bch_fs *c, struct bkey_s_c k, - enum bch_validate_flags flags) + struct bkey_validate_context from) { if (test_bit(BCH_FS_no_invalid_checks, &c->flags)) return 0; @@ -140,7 +140,7 @@ int bch2_bkey_val_validate(struct bch_fs *c, struct bkey_s_c k, if (!ops->key_validate) return 0; - ret = ops->key_validate(c, k, flags); + ret = ops->key_validate(c, k, from); fsck_err: return ret; } @@ -161,9 +161,10 @@ const char *bch2_btree_node_type_str(enum btree_node_type type) } int __bch2_bkey_validate(struct bch_fs *c, struct bkey_s_c k, - enum btree_node_type type, - enum bch_validate_flags flags) + struct bkey_validate_context from) { + enum btree_node_type type = __btree_node_type(from.level, from.btree); + if (test_bit(BCH_FS_no_invalid_checks, &c->flags)) return 0; @@ -177,7 +178,7 @@ int __bch2_bkey_validate(struct bch_fs *c, struct bkey_s_c k, return 0; bkey_fsck_err_on(k.k->type < KEY_TYPE_MAX && - (type == BKEY_TYPE_btree || (flags & BCH_VALIDATE_commit)) && + (type == BKEY_TYPE_btree || (from.flags & BCH_VALIDATE_commit)) && !(bch2_key_types_allowed[type] & BIT_ULL(k.k->type)), c, bkey_invalid_type_for_btree, "invalid key type for btree %s (%s)", @@ -228,15 +229,15 @@ fsck_err: } int bch2_bkey_validate(struct bch_fs *c, struct bkey_s_c k, - enum btree_node_type type, - enum bch_validate_flags flags) + struct bkey_validate_context from) { - return __bch2_bkey_validate(c, k, type, flags) ?: - bch2_bkey_val_validate(c, k, flags); + return __bch2_bkey_validate(c, k, from) ?: + bch2_bkey_val_validate(c, k, from); } int bch2_bkey_in_btree_node(struct bch_fs *c, struct btree *b, - struct bkey_s_c k, enum bch_validate_flags flags) + struct bkey_s_c k, + struct bkey_validate_context from) { int ret = 0; diff --git a/fs/bcachefs/bkey_methods.h b/fs/bcachefs/bkey_methods.h index 018fb72e32d3..bf34111cdf00 100644 --- a/fs/bcachefs/bkey_methods.h +++ b/fs/bcachefs/bkey_methods.h @@ -22,7 +22,7 @@ extern const struct bkey_ops bch2_bkey_null_ops; */ struct bkey_ops { int (*key_validate)(struct bch_fs *c, struct bkey_s_c k, - enum bch_validate_flags flags); + struct bkey_validate_context from); void (*val_to_text)(struct printbuf *, struct bch_fs *, struct bkey_s_c); void (*swab)(struct bkey_s); @@ -48,13 +48,14 @@ static inline const struct bkey_ops *bch2_bkey_type_ops(enum bch_bkey_type type) : &bch2_bkey_null_ops; } -int bch2_bkey_val_validate(struct bch_fs *, struct bkey_s_c, enum bch_validate_flags); -int __bch2_bkey_validate(struct bch_fs *, struct bkey_s_c, enum btree_node_type, - enum bch_validate_flags); -int bch2_bkey_validate(struct bch_fs *, struct bkey_s_c, enum btree_node_type, - enum bch_validate_flags); +int bch2_bkey_val_validate(struct bch_fs *, struct bkey_s_c, + struct bkey_validate_context); +int __bch2_bkey_validate(struct bch_fs *, struct bkey_s_c, + struct bkey_validate_context); +int bch2_bkey_validate(struct bch_fs *, struct bkey_s_c, + struct bkey_validate_context); int bch2_bkey_in_btree_node(struct bch_fs *, struct btree *, struct bkey_s_c, - enum bch_validate_flags); + struct bkey_validate_context from); void bch2_bpos_to_text(struct printbuf *, struct bpos); void bch2_bkey_to_text(struct printbuf *, const struct bkey *); diff --git a/fs/bcachefs/bkey_types.h b/fs/bcachefs/bkey_types.h index c9ae9e42b385..b4f328f9853c 100644 --- a/fs/bcachefs/bkey_types.h +++ b/fs/bcachefs/bkey_types.h @@ -210,4 +210,32 @@ static inline struct bkey_i_##name *bkey_##name##_init(struct bkey_i *_k)\ BCH_BKEY_TYPES(); #undef x +enum bch_validate_flags { + BCH_VALIDATE_write = BIT(0), + BCH_VALIDATE_commit = BIT(1), + BCH_VALIDATE_silent = BIT(2), +}; + +#define BKEY_VALIDATE_CONTEXTS() \ + x(unknown) \ + x(superblock) \ + x(journal) \ + x(btree_root) \ + x(btree_node) \ + x(commit) + +struct bkey_validate_context { + enum { +#define x(n) BKEY_VALIDATE_##n, + BKEY_VALIDATE_CONTEXTS() +#undef x + } from:8; + enum bch_validate_flags flags:8; + u8 level; + enum btree_id btree; + bool root:1; + unsigned journal_offset; + u64 journal_seq; +}; + #endif /* _BCACHEFS_BKEY_TYPES_H */ diff --git a/fs/bcachefs/btree_cache.c b/fs/bcachefs/btree_cache.c index 7123019ab3bc..672ca2c1d37d 100644 --- a/fs/bcachefs/btree_cache.c +++ b/fs/bcachefs/btree_cache.c @@ -222,7 +222,6 @@ void bch2_node_pin(struct bch_fs *c, struct btree *b) struct btree_cache *bc = &c->btree_cache; mutex_lock(&bc->lock); - BUG_ON(!__btree_node_pinned(bc, b)); if (b != btree_node_root(c, b) && !btree_node_pinned(b)) { set_btree_node_pinned(b); list_move(&b->list, &bc->live[1].list); @@ -326,7 +325,7 @@ void bch2_btree_node_update_key_early(struct btree_trans *trans, if (!IS_ERR_OR_NULL(b)) { mutex_lock(&c->btree_cache.lock); - bch2_btree_node_hash_remove(&c->btree_cache, b); + __bch2_btree_node_hash_remove(&c->btree_cache, b); bkey_copy(&b->key, new); ret = __bch2_btree_node_hash_insert(&c->btree_cache, b); @@ -1004,16 +1003,14 @@ static noinline void btree_bad_header(struct bch_fs *c, struct btree *b) return; prt_printf(&buf, - "btree node header doesn't match ptr\n" - "btree %s level %u\n" - "ptr: ", - bch2_btree_id_str(b->c.btree_id), b->c.level); + "btree node header doesn't match ptr: "); + bch2_btree_id_level_to_text(&buf, b->c.btree_id, b->c.level); + prt_str(&buf, "\nptr: "); bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(&b->key)); - prt_printf(&buf, "\nheader: btree %s level %llu\n" - "min ", - bch2_btree_id_str(BTREE_NODE_ID(b->data)), - BTREE_NODE_LEVEL(b->data)); + prt_str(&buf, "\nheader: "); + bch2_btree_id_level_to_text(&buf, BTREE_NODE_ID(b->data), BTREE_NODE_LEVEL(b->data)); + prt_str(&buf, "\nmin "); bch2_bpos_to_text(&buf, b->data->min_key); prt_printf(&buf, "\nmax "); @@ -1133,7 +1130,7 @@ retry: if (unlikely(btree_node_read_error(b))) { six_unlock_type(&b->c.lock, lock_type); - return ERR_PTR(-BCH_ERR_btree_node_read_error); + return ERR_PTR(-BCH_ERR_btree_node_read_err_cached); } EBUG_ON(b->c.btree_id != path->btree_id); @@ -1223,7 +1220,7 @@ struct btree *bch2_btree_node_get(struct btree_trans *trans, struct btree_path * if (unlikely(btree_node_read_error(b))) { six_unlock_type(&b->c.lock, lock_type); - return ERR_PTR(-BCH_ERR_btree_node_read_error); + return ERR_PTR(-BCH_ERR_btree_node_read_err_cached); } EBUG_ON(b->c.btree_id != path->btree_id); @@ -1305,7 +1302,7 @@ lock_node: if (unlikely(btree_node_read_error(b))) { six_unlock_read(&b->c.lock); - b = ERR_PTR(-BCH_ERR_btree_node_read_error); + b = ERR_PTR(-BCH_ERR_btree_node_read_err_cached); goto out; } @@ -1398,13 +1395,31 @@ void bch2_btree_id_to_text(struct printbuf *out, enum btree_id btree) prt_printf(out, "(unknown btree %u)", btree); } +void bch2_btree_id_level_to_text(struct printbuf *out, enum btree_id btree, unsigned level) +{ + prt_str(out, "btree="); + bch2_btree_id_to_text(out, btree); + prt_printf(out, " level=%u", level); +} + +void __bch2_btree_pos_to_text(struct printbuf *out, struct bch_fs *c, + enum btree_id btree, unsigned level, struct bkey_s_c k) +{ + bch2_btree_id_to_text(out, btree); + prt_printf(out, " level %u/", level); + struct btree_root *r = bch2_btree_id_root(c, btree); + if (r) + prt_printf(out, "%u", r->level); + else + prt_printf(out, "(unknown)"); + prt_printf(out, "\n "); + + bch2_bkey_val_to_text(out, c, k); +} + void bch2_btree_pos_to_text(struct printbuf *out, struct bch_fs *c, const struct btree *b) { - prt_printf(out, "%s level %u/%u\n ", - bch2_btree_id_str(b->c.btree_id), - b->c.level, - bch2_btree_id_root(c, b->c.btree_id)->level); - bch2_bkey_val_to_text(out, c, bkey_i_to_s_c(&b->key)); + __bch2_btree_pos_to_text(out, c, b->c.btree_id, b->c.level, bkey_i_to_s_c(&b->key)); } void bch2_btree_node_to_text(struct printbuf *out, struct bch_fs *c, const struct btree *b) @@ -1478,8 +1493,12 @@ void bch2_btree_cache_to_text(struct printbuf *out, const struct btree_cache *bc prt_printf(out, "cannibalize lock:\t%p\n", bc->alloc_lock); prt_newline(out); - for (unsigned i = 0; i < ARRAY_SIZE(bc->nr_by_btree); i++) - prt_btree_cache_line(out, c, bch2_btree_id_str(i), bc->nr_by_btree[i]); + for (unsigned i = 0; i < ARRAY_SIZE(bc->nr_by_btree); i++) { + bch2_btree_id_to_text(out, i); + prt_printf(out, "\t"); + prt_human_readable_u64(out, bc->nr_by_btree[i] * c->opts.btree_node_size); + prt_printf(out, " (%zu)\n", bc->nr_by_btree[i]); + } prt_newline(out); prt_printf(out, "freed:\t%zu\n", bc->nr_freed); diff --git a/fs/bcachefs/btree_cache.h b/fs/bcachefs/btree_cache.h index 66e86d1a178d..ca3c1b145330 100644 --- a/fs/bcachefs/btree_cache.h +++ b/fs/bcachefs/btree_cache.h @@ -128,19 +128,27 @@ static inline struct btree_root *bch2_btree_id_root(struct bch_fs *c, unsigned i } else { unsigned idx = id - BTREE_ID_NR; - EBUG_ON(idx >= c->btree_roots_extra.nr); + /* This can happen when we're called from btree_node_scan */ + if (idx >= c->btree_roots_extra.nr) + return NULL; + return &c->btree_roots_extra.data[idx]; } } static inline struct btree *btree_node_root(struct bch_fs *c, struct btree *b) { - return bch2_btree_id_root(c, b->c.btree_id)->b; + struct btree_root *r = bch2_btree_id_root(c, b->c.btree_id); + + return r ? r->b : NULL; } -const char *bch2_btree_id_str(enum btree_id); +const char *bch2_btree_id_str(enum btree_id); /* avoid */ void bch2_btree_id_to_text(struct printbuf *, enum btree_id); +void bch2_btree_id_level_to_text(struct printbuf *, enum btree_id, unsigned); +void __bch2_btree_pos_to_text(struct printbuf *, struct bch_fs *, + enum btree_id, unsigned, struct bkey_s_c); void bch2_btree_pos_to_text(struct printbuf *, struct bch_fs *, const struct btree *); void bch2_btree_node_to_text(struct printbuf *, struct bch_fs *, const struct btree *); void bch2_btree_cache_to_text(struct printbuf *, const struct btree_cache *); diff --git a/fs/bcachefs/btree_gc.c b/fs/bcachefs/btree_gc.c index 81dcf9e512c0..dd1d9b74076e 100644 --- a/fs/bcachefs/btree_gc.c +++ b/fs/bcachefs/btree_gc.c @@ -29,6 +29,7 @@ #include "move.h" #include "recovery_passes.h" #include "reflink.h" +#include "recovery.h" #include "replicas.h" #include "super-io.h" #include "trace.h" @@ -56,8 +57,8 @@ void bch2_gc_pos_to_text(struct printbuf *out, struct gc_pos *p) { prt_str(out, bch2_gc_phase_strs[p->phase]); prt_char(out, ' '); - bch2_btree_id_to_text(out, p->btree); - prt_printf(out, " l=%u ", p->level); + bch2_btree_id_level_to_text(out, p->btree, p->level); + prt_char(out, ' '); bch2_bpos_to_text(out, p->pos); } @@ -209,8 +210,9 @@ static int btree_check_node_boundaries(struct btree_trans *trans, struct btree * if (bpos_eq(expected_start, cur->data->min_key)) return 0; - prt_printf(&buf, " at btree %s level %u:\n parent: ", - bch2_btree_id_str(b->c.btree_id), b->c.level); + prt_printf(&buf, " at "); + bch2_btree_id_level_to_text(&buf, b->c.btree_id, b->c.level); + prt_printf(&buf, ":\n parent: "); bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(&b->key)); if (prev) { @@ -277,8 +279,9 @@ static int btree_repair_node_end(struct btree_trans *trans, struct btree *b, if (bpos_eq(child->key.k.p, b->key.k.p)) return 0; - prt_printf(&buf, "at btree %s level %u:\n parent: ", - bch2_btree_id_str(b->c.btree_id), b->c.level); + prt_printf(&buf, " at "); + bch2_btree_id_level_to_text(&buf, b->c.btree_id, b->c.level); + prt_printf(&buf, ":\n parent: "); bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(&b->key)); prt_str(&buf, "\n child: "); @@ -341,14 +344,14 @@ again: ret = PTR_ERR_OR_ZERO(cur); printbuf_reset(&buf); + bch2_btree_id_level_to_text(&buf, b->c.btree_id, b->c.level - 1); + prt_char(&buf, ' '); bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(cur_k.k)); if (mustfix_fsck_err_on(bch2_err_matches(ret, EIO), - trans, btree_node_unreadable, - "Topology repair: unreadable btree node at btree %s level %u:\n" + trans, btree_node_read_error, + "Topology repair: unreadable btree node at\n" " %s", - bch2_btree_id_str(b->c.btree_id), - b->c.level - 1, buf.buf)) { bch2_btree_node_evict(trans, cur_k.k); cur = NULL; @@ -357,11 +360,9 @@ again: if (ret) break; - if (!btree_id_is_alloc(b->c.btree_id)) { - ret = bch2_run_explicit_recovery_pass(c, BCH_RECOVERY_PASS_scan_for_btree_nodes); - if (ret) - break; - } + ret = bch2_btree_lost_data(c, b->c.btree_id); + if (ret) + break; continue; } @@ -370,7 +371,7 @@ again: break; if (bch2_btree_node_is_stale(c, cur)) { - bch_info(c, "btree node %s older than nodes found by scanning", buf.buf); + bch_info(c, "btree node older than nodes found by scanning\n %s", buf.buf); six_unlock_read(&cur->c.lock); bch2_btree_node_evict(trans, cur_k.k); ret = bch2_journal_key_delete(c, b->c.btree_id, @@ -478,14 +479,13 @@ again: } printbuf_reset(&buf); + bch2_btree_id_level_to_text(&buf, b->c.btree_id, b->c.level); + prt_newline(&buf); bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(&b->key)); if (mustfix_fsck_err_on(!have_child, trans, btree_node_topology_interior_node_empty, - "empty interior btree node at btree %s level %u\n" - " %s", - bch2_btree_id_str(b->c.btree_id), - b->c.level, buf.buf)) + "empty interior btree node at %s", buf.buf)) ret = DROP_THIS_NODE; err: fsck_err: @@ -511,6 +511,7 @@ int bch2_check_topology(struct bch_fs *c) { struct btree_trans *trans = bch2_trans_get(c); struct bpos pulled_from_scan = POS_MIN; + struct printbuf buf = PRINTBUF; int ret = 0; bch2_trans_srcu_unlock(trans); @@ -519,19 +520,22 @@ int bch2_check_topology(struct bch_fs *c) struct btree_root *r = bch2_btree_id_root(c, i); bool reconstructed_root = false; + printbuf_reset(&buf); + bch2_btree_id_to_text(&buf, i); + if (r->error) { - ret = bch2_run_explicit_recovery_pass(c, BCH_RECOVERY_PASS_scan_for_btree_nodes); + ret = bch2_btree_lost_data(c, i); if (ret) break; reconstruct_root: - bch_info(c, "btree root %s unreadable, must recover from scan", bch2_btree_id_str(i)); + bch_info(c, "btree root %s unreadable, must recover from scan", buf.buf); r->alive = false; r->error = 0; if (!bch2_btree_has_scanned_nodes(c, i)) { mustfix_fsck_err(trans, btree_root_unreadable_and_scan_found_nothing, - "no nodes found for btree %s, continue?", bch2_btree_id_str(i)); + "no nodes found for btree %s, continue?", buf.buf); bch2_btree_root_alloc_fake_trans(trans, i, 0); } else { bch2_btree_root_alloc_fake_trans(trans, i, 1); @@ -560,13 +564,14 @@ reconstruct_root: if (!reconstructed_root) goto reconstruct_root; - bch_err(c, "empty btree root %s", bch2_btree_id_str(i)); + bch_err(c, "empty btree root %s", buf.buf); bch2_btree_root_alloc_fake_trans(trans, i, 0); r->alive = false; ret = 0; } } fsck_err: + printbuf_exit(&buf); bch2_trans_put(trans); return ret; } @@ -713,6 +718,7 @@ static int bch2_gc_btrees(struct bch_fs *c) { struct btree_trans *trans = bch2_trans_get(c); enum btree_id ids[BTREE_ID_NR]; + struct printbuf buf = PRINTBUF; unsigned i; int ret = 0; @@ -727,14 +733,9 @@ static int bch2_gc_btrees(struct bch_fs *c) continue; ret = bch2_gc_btree(trans, btree, true); - - if (mustfix_fsck_err_on(bch2_err_matches(ret, EIO), - trans, btree_node_read_error, - "btree node read error for %s", - bch2_btree_id_str(btree))) - ret = bch2_run_explicit_recovery_pass(c, BCH_RECOVERY_PASS_check_topology); } -fsck_err: + + printbuf_exit(&buf); bch2_trans_put(trans); bch_err_fn(c, ret); return ret; @@ -802,7 +803,6 @@ static int bch2_alloc_write_key(struct btree_trans *trans, old = bch2_alloc_to_v4(k, &old_convert); gc = new = *old; - percpu_down_read(&c->mark_lock); __bucket_m_to_alloc(&gc, *gc_bucket(ca, iter->pos.offset)); old_gc = gc; @@ -813,7 +813,6 @@ static int bch2_alloc_write_key(struct btree_trans *trans, gc.data_type = old->data_type; gc.dirty_sectors = old->dirty_sectors; } - percpu_up_read(&c->mark_lock); /* * gc.data_type doesn't yet include need_discard & need_gc_gen states - @@ -831,11 +830,9 @@ static int bch2_alloc_write_key(struct btree_trans *trans, * safe w.r.t. transaction restarts, so fixup the gc_bucket so * we don't run it twice: */ - percpu_down_read(&c->mark_lock); struct bucket *gc_m = gc_bucket(ca, iter->pos.offset); gc_m->data_type = gc.data_type; gc_m->dirty_sectors = gc.dirty_sectors; - percpu_up_read(&c->mark_lock); } if (fsck_err_on(new.data_type != gc.data_type, @@ -895,11 +892,11 @@ static int bch2_gc_alloc_done(struct bch_fs *c) for_each_member_device(c, ca) { ret = bch2_trans_run(c, - for_each_btree_key_upto_commit(trans, iter, BTREE_ID_alloc, + for_each_btree_key_max_commit(trans, iter, BTREE_ID_alloc, POS(ca->dev_idx, ca->mi.first_bucket), POS(ca->dev_idx, ca->mi.nbuckets - 1), BTREE_ITER_slots|BTREE_ITER_prefetch, k, - NULL, NULL, BCH_TRANS_COMMIT_lazy_rw, + NULL, NULL, BCH_TRANS_COMMIT_no_enospc, bch2_alloc_write_key(trans, &iter, ca, k))); if (ret) { bch2_dev_put(ca); @@ -928,98 +925,6 @@ static int bch2_gc_alloc_start(struct bch_fs *c) return ret; } -static int bch2_gc_write_reflink_key(struct btree_trans *trans, - struct btree_iter *iter, - struct bkey_s_c k, - size_t *idx) -{ - struct bch_fs *c = trans->c; - const __le64 *refcount = bkey_refcount_c(k); - struct printbuf buf = PRINTBUF; - struct reflink_gc *r; - int ret = 0; - - if (!refcount) - return 0; - - while ((r = genradix_ptr(&c->reflink_gc_table, *idx)) && - r->offset < k.k->p.offset) - ++*idx; - - if (!r || - r->offset != k.k->p.offset || - r->size != k.k->size) { - bch_err(c, "unexpected inconsistency walking reflink table at gc finish"); - return -EINVAL; - } - - if (fsck_err_on(r->refcount != le64_to_cpu(*refcount), - trans, reflink_v_refcount_wrong, - "reflink key has wrong refcount:\n" - " %s\n" - " should be %u", - (bch2_bkey_val_to_text(&buf, c, k), buf.buf), - r->refcount)) { - struct bkey_i *new = bch2_bkey_make_mut_noupdate(trans, k); - ret = PTR_ERR_OR_ZERO(new); - if (ret) - goto out; - - if (!r->refcount) - new->k.type = KEY_TYPE_deleted; - else - *bkey_refcount(bkey_i_to_s(new)) = cpu_to_le64(r->refcount); - ret = bch2_trans_update(trans, iter, new, 0); - } -out: -fsck_err: - printbuf_exit(&buf); - return ret; -} - -static int bch2_gc_reflink_done(struct bch_fs *c) -{ - size_t idx = 0; - - int ret = bch2_trans_run(c, - for_each_btree_key_commit(trans, iter, - BTREE_ID_reflink, POS_MIN, - BTREE_ITER_prefetch, k, - NULL, NULL, BCH_TRANS_COMMIT_no_enospc, - bch2_gc_write_reflink_key(trans, &iter, k, &idx))); - c->reflink_gc_nr = 0; - return ret; -} - -static int bch2_gc_reflink_start(struct bch_fs *c) -{ - c->reflink_gc_nr = 0; - - int ret = bch2_trans_run(c, - for_each_btree_key(trans, iter, BTREE_ID_reflink, POS_MIN, - BTREE_ITER_prefetch, k, ({ - const __le64 *refcount = bkey_refcount_c(k); - - if (!refcount) - continue; - - struct reflink_gc *r = genradix_ptr_alloc(&c->reflink_gc_table, - c->reflink_gc_nr++, GFP_KERNEL); - if (!r) { - ret = -BCH_ERR_ENOMEM_gc_reflink_start; - break; - } - - r->offset = k.k->p.offset; - r->size = k.k->size; - r->refcount = 0; - 0; - }))); - - bch_err_fn(c, ret); - return ret; -} - static int bch2_gc_write_stripes_key(struct btree_trans *trans, struct btree_iter *iter, struct bkey_s_c k) @@ -1171,7 +1076,6 @@ static int gc_btree_gens_key(struct btree_trans *trans, if (unlikely(test_bit(BCH_FS_going_ro, &c->flags))) return -EROFS; - percpu_down_read(&c->mark_lock); rcu_read_lock(); bkey_for_each_ptr(ptrs, ptr) { struct bch_dev *ca = bch2_dev_rcu(c, ptr->dev); @@ -1180,7 +1084,6 @@ static int gc_btree_gens_key(struct btree_trans *trans, if (dev_ptr_stale(ca, ptr) > 16) { rcu_read_unlock(); - percpu_up_read(&c->mark_lock); goto update; } } @@ -1195,7 +1098,6 @@ static int gc_btree_gens_key(struct btree_trans *trans, *gen = ptr->gen; } rcu_read_unlock(); - percpu_up_read(&c->mark_lock); return 0; update: u = bch2_bkey_make_mut(trans, iter, &k, 0); @@ -1224,7 +1126,6 @@ static int bch2_alloc_write_oldest_gen(struct btree_trans *trans, struct bch_dev return ret; a_mut->v.oldest_gen = ca->oldest_gen[iter->pos.offset]; - alloc_data_type_set(&a_mut->v, a_mut->v.data_type); return bch2_trans_update(trans, iter, &a_mut->k_i, 0); } @@ -1337,9 +1238,16 @@ void bch2_gc_gens_async(struct bch_fs *c) bch2_write_ref_put(c, BCH_WRITE_REF_gc_gens); } -void bch2_fs_gc_init(struct bch_fs *c) +void bch2_fs_btree_gc_exit(struct bch_fs *c) { - seqcount_init(&c->gc_pos_lock); +} +int bch2_fs_btree_gc_init(struct bch_fs *c) +{ + seqcount_init(&c->gc_pos_lock); INIT_WORK(&c->gc_gens_work, bch2_gc_gens_work); + + init_rwsem(&c->gc_lock); + mutex_init(&c->gc_gens_lock); + return 0; } diff --git a/fs/bcachefs/btree_gc.h b/fs/bcachefs/btree_gc.h index 8a47e8bd0791..9693a90a48a2 100644 --- a/fs/bcachefs/btree_gc.h +++ b/fs/bcachefs/btree_gc.h @@ -82,6 +82,8 @@ void bch2_gc_pos_to_text(struct printbuf *, struct gc_pos *); int bch2_gc_gens(struct bch_fs *); void bch2_gc_gens_async(struct bch_fs *); -void bch2_fs_gc_init(struct bch_fs *); + +void bch2_fs_btree_gc_exit(struct bch_fs *); +int bch2_fs_btree_gc_init(struct bch_fs *); #endif /* _BCACHEFS_BTREE_GC_H */ diff --git a/fs/bcachefs/btree_io.c b/fs/bcachefs/btree_io.c index 839d68802e42..e371e60e3133 100644 --- a/fs/bcachefs/btree_io.c +++ b/fs/bcachefs/btree_io.c @@ -25,9 +25,8 @@ static void bch2_btree_node_header_to_text(struct printbuf *out, struct btree_node *bn) { - prt_printf(out, "btree=%s l=%u seq %llux\n", - bch2_btree_id_str(BTREE_NODE_ID(bn)), - (unsigned) BTREE_NODE_LEVEL(bn), bn->keys.seq); + bch2_btree_id_level_to_text(out, BTREE_NODE_ID(bn), BTREE_NODE_LEVEL(bn)); + prt_printf(out, " seq %llx %llu\n", bn->keys.seq, BTREE_NODE_SEQ(bn)); prt_str(out, "min: "); bch2_bpos_to_text(out, bn->min_key); prt_newline(out); @@ -490,8 +489,8 @@ void bch2_btree_init_next(struct btree_trans *trans, struct btree *b) if (b->nsets == MAX_BSETS && !btree_node_write_in_flight(b) && should_compact_all(c, b)) { - bch2_btree_node_write(c, b, SIX_LOCK_write, - BTREE_WRITE_init_next_bset); + bch2_btree_node_write_trans(trans, b, SIX_LOCK_write, + BTREE_WRITE_init_next_bset); reinit_iter = true; } @@ -832,13 +831,32 @@ fsck_err: return ret; } +static int btree_node_bkey_val_validate(struct bch_fs *c, struct btree *b, + struct bkey_s_c k, + enum bch_validate_flags flags) +{ + return bch2_bkey_val_validate(c, k, (struct bkey_validate_context) { + .from = BKEY_VALIDATE_btree_node, + .level = b->c.level, + .btree = b->c.btree_id, + .flags = flags + }); +} + static int bset_key_validate(struct bch_fs *c, struct btree *b, struct bkey_s_c k, - bool updated_range, int rw) + bool updated_range, + enum bch_validate_flags flags) { - return __bch2_bkey_validate(c, k, btree_node_type(b), 0) ?: - (!updated_range ? bch2_bkey_in_btree_node(c, b, k, 0) : 0) ?: - (rw == WRITE ? bch2_bkey_val_validate(c, k, 0) : 0); + struct bkey_validate_context from = (struct bkey_validate_context) { + .from = BKEY_VALIDATE_btree_node, + .level = b->c.level, + .btree = b->c.btree_id, + .flags = flags, + }; + return __bch2_bkey_validate(c, k, from) ?: + (!updated_range ? bch2_bkey_in_btree_node(c, b, k, from) : 0) ?: + (flags & BCH_VALIDATE_write ? btree_node_bkey_val_validate(c, b, k, flags) : 0); } static bool bkey_packed_valid(struct bch_fs *c, struct btree *b, @@ -855,7 +873,21 @@ static bool bkey_packed_valid(struct bch_fs *c, struct btree *b, struct bkey tmp; struct bkey_s u = __bkey_disassemble(b, k, &tmp); - return !__bch2_bkey_validate(c, u.s_c, btree_node_type(b), BCH_VALIDATE_silent); + return !__bch2_bkey_validate(c, u.s_c, + (struct bkey_validate_context) { + .from = BKEY_VALIDATE_btree_node, + .level = b->c.level, + .btree = b->c.btree_id, + .flags = BCH_VALIDATE_silent + }); +} + +static inline int btree_node_read_bkey_cmp(const struct btree *b, + const struct bkey_packed *l, + const struct bkey_packed *r) +{ + return bch2_bkey_cmp_packed(b, l, r) + ?: (int) bkey_deleted(r) - (int) bkey_deleted(l); } static int validate_bset_keys(struct bch_fs *c, struct btree *b, @@ -918,7 +950,7 @@ static int validate_bset_keys(struct bch_fs *c, struct btree *b, BSET_BIG_ENDIAN(i), write, &b->format, k); - if (prev && bkey_iter_cmp(b, prev, k) > 0) { + if (prev && btree_node_read_bkey_cmp(b, prev, k) >= 0) { struct bkey up = bkey_unpack_key(b, prev); printbuf_reset(&buf); @@ -965,6 +997,7 @@ drop_this_key: got_good_key: le16_add_cpu(&i->u64s, -next_good_key); memmove_u64s_down(k, bkey_p_next(k), (u64 *) vstruct_end(i) - (u64 *) k); + set_btree_node_need_rewrite(b); } fsck_err: printbuf_exit(&buf); @@ -1038,39 +1071,51 @@ int bch2_btree_node_read_done(struct bch_fs *c, struct bch_dev *ca, while (b->written < (ptr_written ?: btree_sectors(c))) { unsigned sectors; - struct nonce nonce; bool first = !b->written; - bool csum_bad; - if (!b->written) { + if (first) { + bne = NULL; i = &b->data->keys; + } else { + bne = write_block(b); + i = &bne->keys; - btree_err_on(!bch2_checksum_type_valid(c, BSET_CSUM_TYPE(i)), - -BCH_ERR_btree_node_read_err_want_retry, - c, ca, b, i, NULL, - bset_unknown_csum, - "unknown checksum type %llu", BSET_CSUM_TYPE(i)); - - nonce = btree_nonce(i, b->written << 9); + if (i->seq != b->data->keys.seq) + break; + } - struct bch_csum csum = csum_vstruct(c, BSET_CSUM_TYPE(i), nonce, b->data); - csum_bad = bch2_crc_cmp(b->data->csum, csum); - if (csum_bad) - bch2_io_error(ca, BCH_MEMBER_ERROR_checksum); + struct nonce nonce = btree_nonce(i, b->written << 9); + bool good_csum_type = bch2_checksum_type_valid(c, BSET_CSUM_TYPE(i)); - btree_err_on(csum_bad, - -BCH_ERR_btree_node_read_err_want_retry, - c, ca, b, i, NULL, - bset_bad_csum, - "%s", - (printbuf_reset(&buf), - bch2_csum_err_msg(&buf, BSET_CSUM_TYPE(i), b->data->csum, csum), - buf.buf)); - - ret = bset_encrypt(c, i, b->written << 9); - if (bch2_fs_fatal_err_on(ret, c, - "decrypting btree node: %s", bch2_err_str(ret))) - goto fsck_err; + btree_err_on(!good_csum_type, + bch2_csum_type_is_encryption(BSET_CSUM_TYPE(i)) + ? -BCH_ERR_btree_node_read_err_must_retry + : -BCH_ERR_btree_node_read_err_want_retry, + c, ca, b, i, NULL, + bset_unknown_csum, + "unknown checksum type %llu", BSET_CSUM_TYPE(i)); + + if (first) { + if (good_csum_type) { + struct bch_csum csum = csum_vstruct(c, BSET_CSUM_TYPE(i), nonce, b->data); + bool csum_bad = bch2_crc_cmp(b->data->csum, csum); + if (csum_bad) + bch2_io_error(ca, BCH_MEMBER_ERROR_checksum); + + btree_err_on(csum_bad, + -BCH_ERR_btree_node_read_err_want_retry, + c, ca, b, i, NULL, + bset_bad_csum, + "%s", + (printbuf_reset(&buf), + bch2_csum_err_msg(&buf, BSET_CSUM_TYPE(i), b->data->csum, csum), + buf.buf)); + + ret = bset_encrypt(c, i, b->written << 9); + if (bch2_fs_fatal_err_on(ret, c, + "decrypting btree node: %s", bch2_err_str(ret))) + goto fsck_err; + } btree_err_on(btree_node_type_is_extents(btree_node_type(b)) && !BTREE_NODE_NEW_EXTENT_OVERWRITE(b->data), @@ -1081,37 +1126,26 @@ int bch2_btree_node_read_done(struct bch_fs *c, struct bch_dev *ca, sectors = vstruct_sectors(b->data, c->block_bits); } else { - bne = write_block(b); - i = &bne->keys; - - if (i->seq != b->data->keys.seq) - break; - - btree_err_on(!bch2_checksum_type_valid(c, BSET_CSUM_TYPE(i)), - -BCH_ERR_btree_node_read_err_want_retry, - c, ca, b, i, NULL, - bset_unknown_csum, - "unknown checksum type %llu", BSET_CSUM_TYPE(i)); - - nonce = btree_nonce(i, b->written << 9); - struct bch_csum csum = csum_vstruct(c, BSET_CSUM_TYPE(i), nonce, bne); - csum_bad = bch2_crc_cmp(bne->csum, csum); - if (ca && csum_bad) - bch2_io_error(ca, BCH_MEMBER_ERROR_checksum); - - btree_err_on(csum_bad, - -BCH_ERR_btree_node_read_err_want_retry, - c, ca, b, i, NULL, - bset_bad_csum, - "%s", - (printbuf_reset(&buf), - bch2_csum_err_msg(&buf, BSET_CSUM_TYPE(i), bne->csum, csum), - buf.buf)); - - ret = bset_encrypt(c, i, b->written << 9); - if (bch2_fs_fatal_err_on(ret, c, - "decrypting btree node: %s", bch2_err_str(ret))) - goto fsck_err; + if (good_csum_type) { + struct bch_csum csum = csum_vstruct(c, BSET_CSUM_TYPE(i), nonce, bne); + bool csum_bad = bch2_crc_cmp(bne->csum, csum); + if (ca && csum_bad) + bch2_io_error(ca, BCH_MEMBER_ERROR_checksum); + + btree_err_on(csum_bad, + -BCH_ERR_btree_node_read_err_want_retry, + c, ca, b, i, NULL, + bset_bad_csum, + "%s", + (printbuf_reset(&buf), + bch2_csum_err_msg(&buf, BSET_CSUM_TYPE(i), bne->csum, csum), + buf.buf)); + + ret = bset_encrypt(c, i, b->written << 9); + if (bch2_fs_fatal_err_on(ret, c, + "decrypting btree node: %s", bch2_err_str(ret))) + goto fsck_err; + } sectors = vstruct_sectors(bne, c->block_bits); } @@ -1216,7 +1250,7 @@ int bch2_btree_node_read_done(struct bch_fs *c, struct bch_dev *ca, struct bkey tmp; struct bkey_s u = __bkey_disassemble(b, k, &tmp); - ret = bch2_bkey_val_validate(c, u.s_c, READ); + ret = btree_node_bkey_val_validate(c, b, u.s_c, READ); if (ret == -BCH_ERR_fsck_delete_bkey || (bch2_inject_invalid_keys && !bversion_cmp(u.k->bversion, MAX_VERSION))) { @@ -1226,6 +1260,7 @@ int bch2_btree_node_read_done(struct bch_fs *c, struct bch_dev *ca, memmove_u64s_down(k, bkey_p_next(k), (u64 *) vstruct_end(i) - (u64 *) k); set_btree_bset_end(b, b->set); + set_btree_node_need_rewrite(b); continue; } if (ret) @@ -1339,13 +1374,18 @@ start: rb->start_time); bio_put(&rb->bio); - if (saw_error && + if ((saw_error || + btree_node_need_rewrite(b)) && !btree_node_read_error(b) && c->curr_recovery_pass != BCH_RECOVERY_PASS_scan_for_btree_nodes) { - printbuf_reset(&buf); - bch2_bpos_to_text(&buf, b->key.k.p); - bch_err_ratelimited(c, "%s: rewriting btree node at btree=%s level=%u %s due to error", - __func__, bch2_btree_id_str(b->c.btree_id), b->c.level, buf.buf); + if (saw_error) { + printbuf_reset(&buf); + bch2_btree_id_level_to_text(&buf, b->c.btree_id, b->c.level); + prt_str(&buf, " "); + bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(&b->key)); + bch_err_ratelimited(c, "%s: rewriting btree node at due to error\n %s", + __func__, buf.buf); + } bch2_btree_node_rewrite_async(c, b); } @@ -1933,7 +1973,12 @@ static int validate_bset_for_write(struct bch_fs *c, struct btree *b, bool saw_error; int ret = bch2_bkey_validate(c, bkey_i_to_s_c(&b->key), - BKEY_TYPE_btree, WRITE); + (struct bkey_validate_context) { + .from = BKEY_VALIDATE_btree_node, + .level = b->c.level + 1, + .btree = b->c.btree_id, + .flags = BCH_VALIDATE_write, + }); if (ret) { bch2_fs_inconsistent(c, "invalid btree node key before write"); return ret; @@ -2300,6 +2345,34 @@ void bch2_btree_node_write(struct bch_fs *c, struct btree *b, } } +void bch2_btree_node_write_trans(struct btree_trans *trans, struct btree *b, + enum six_lock_type lock_type_held, + unsigned flags) +{ + struct bch_fs *c = trans->c; + + if (lock_type_held == SIX_LOCK_intent || + (lock_type_held == SIX_LOCK_read && + six_lock_tryupgrade(&b->c.lock))) { + __bch2_btree_node_write(c, b, flags); + + /* don't cycle lock unnecessarily: */ + if (btree_node_just_written(b) && + six_trylock_write(&b->c.lock)) { + bch2_btree_post_write_cleanup(c, b); + __bch2_btree_node_unlock_write(trans, b); + } + + if (lock_type_held == SIX_LOCK_read) + six_lock_downgrade(&b->c.lock); + } else { + __bch2_btree_node_write(c, b, flags); + if (lock_type_held == SIX_LOCK_write && + btree_node_just_written(b)) + bch2_btree_post_write_cleanup(c, b); + } +} + static bool __bch2_btree_flush_all(struct bch_fs *c, unsigned flag) { struct bucket_table *tbl; diff --git a/fs/bcachefs/btree_io.h b/fs/bcachefs/btree_io.h index 9b01ca3de907..6f9e4a6dacf7 100644 --- a/fs/bcachefs/btree_io.h +++ b/fs/bcachefs/btree_io.h @@ -144,11 +144,13 @@ enum btree_write_flags { void __bch2_btree_node_write(struct bch_fs *, struct btree *, unsigned); void bch2_btree_node_write(struct bch_fs *, struct btree *, enum six_lock_type, unsigned); +void bch2_btree_node_write_trans(struct btree_trans *, struct btree *, + enum six_lock_type, unsigned); -static inline void btree_node_write_if_need(struct bch_fs *c, struct btree *b, +static inline void btree_node_write_if_need(struct btree_trans *trans, struct btree *b, enum six_lock_type lock_held) { - bch2_btree_node_write(c, b, lock_held, BTREE_WRITE_ONLY_IF_NEED); + bch2_btree_node_write_trans(trans, b, lock_held, BTREE_WRITE_ONLY_IF_NEED); } bool bch2_btree_flush_all_reads(struct bch_fs *); diff --git a/fs/bcachefs/btree_iter.c b/fs/bcachefs/btree_iter.c index eef9b89c561d..367231ab1980 100644 --- a/fs/bcachefs/btree_iter.c +++ b/fs/bcachefs/btree_iter.c @@ -270,8 +270,10 @@ static void bch2_btree_iter_verify_entry_exit(struct btree_iter *iter) BUG_ON(!(iter->flags & BTREE_ITER_all_snapshots) && iter->pos.snapshot != iter->snapshot); - BUG_ON(bkey_lt(iter->pos, bkey_start_pos(&iter->k)) || - bkey_gt(iter->pos, iter->k.p)); + BUG_ON(iter->flags & BTREE_ITER_all_snapshots ? !bpos_eq(iter->pos, iter->k.p) : + !(iter->flags & BTREE_ITER_is_extents) ? !bkey_eq(iter->pos, iter->k.p) : + (bkey_lt(iter->pos, bkey_start_pos(&iter->k)) || + bkey_gt(iter->pos, iter->k.p))); } static int bch2_btree_iter_verify_ret(struct btree_iter *iter, struct bkey_s_c k) @@ -327,7 +329,7 @@ out: void bch2_assert_pos_locked(struct btree_trans *trans, enum btree_id id, struct bpos pos) { - bch2_trans_verify_not_unlocked(trans); + bch2_trans_verify_not_unlocked_or_in_restart(trans); struct btree_path *path; struct trans_for_each_path_inorder_iter iter; @@ -697,6 +699,19 @@ void bch2_trans_node_add(struct btree_trans *trans, bch2_trans_revalidate_updates_in_node(trans, b); } +void bch2_trans_node_drop(struct btree_trans *trans, + struct btree *b) +{ + struct btree_path *path; + unsigned i, level = b->c.level; + + trans_for_each_path(trans, path, i) + if (path->l[level].b == b) { + btree_node_unlock(trans, path, level); + path->l[level].b = ERR_PTR(-BCH_ERR_no_btree_node_init); + } +} + /* * A btree node has been modified in such a way as to invalidate iterators - fix * them: @@ -720,7 +735,7 @@ static inline int btree_path_lock_root(struct btree_trans *trans, unsigned long trace_ip) { struct bch_fs *c = trans->c; - struct btree *b, **rootp = &bch2_btree_id_root(c, path->btree_id)->b; + struct btree_root *r = bch2_btree_id_root(c, path->btree_id); enum six_lock_type lock_type; unsigned i; int ret; @@ -728,7 +743,12 @@ static inline int btree_path_lock_root(struct btree_trans *trans, EBUG_ON(path->nodes_locked); while (1) { - b = READ_ONCE(*rootp); + struct btree *b = READ_ONCE(r->b); + if (unlikely(!b)) { + BUG_ON(!r->error); + return r->error; + } + path->level = READ_ONCE(b->c.level); if (unlikely(path->level < depth_want)) { @@ -748,14 +768,12 @@ static inline int btree_path_lock_root(struct btree_trans *trans, ret = btree_node_lock(trans, path, &b->c, path->level, lock_type, trace_ip); if (unlikely(ret)) { - if (bch2_err_matches(ret, BCH_ERR_lock_fail_root_changed)) - continue; if (bch2_err_matches(ret, BCH_ERR_transaction_restart)) return ret; BUG(); } - if (likely(b == READ_ONCE(*rootp) && + if (likely(b == READ_ONCE(r->b) && b->c.level == path->level && !race_fault())) { for (i = 0; i < path->level; i++) @@ -825,6 +843,8 @@ static int btree_path_prefetch_j(struct btree_trans *trans, struct btree_path *p bch2_bkey_buf_init(&tmp); + jiter->fail_if_too_many_whiteouts = true; + while (nr-- && !ret) { if (!bch2_btree_node_relock(trans, path, path->level)) break; @@ -1000,7 +1020,7 @@ retry_all: bch2_trans_unlock(trans); cond_resched(); - trans_set_locked(trans); + trans_set_locked(trans, false); if (unlikely(trans->memory_allocation_failure)) { struct closure cl; @@ -1267,7 +1287,7 @@ __bch2_btree_path_set_pos(struct btree_trans *trans, { int cmp = bpos_cmp(new_pos, trans->paths[path_idx].pos); - bch2_trans_verify_not_in_restart(trans); + bch2_trans_verify_not_unlocked_or_in_restart(trans); EBUG_ON(!trans->paths[path_idx].ref); trace_btree_path_set_pos(trans, trans->paths + path_idx, &new_pos); @@ -1427,17 +1447,31 @@ void __noreturn bch2_trans_restart_error(struct btree_trans *trans, u32 restart_ (void *) trans->last_begin_ip); } -void __noreturn bch2_trans_in_restart_error(struct btree_trans *trans) +static void __noreturn bch2_trans_in_restart_error(struct btree_trans *trans) { +#ifdef CONFIG_BCACHEFS_DEBUG + struct printbuf buf = PRINTBUF; + bch2_prt_backtrace(&buf, &trans->last_restarted_trace); + panic("in transaction restart: %s, last restarted by\n%s", + bch2_err_str(trans->restarted), + buf.buf); +#else panic("in transaction restart: %s, last restarted by %pS\n", bch2_err_str(trans->restarted), (void *) trans->last_restarted_ip); +#endif } -void __noreturn bch2_trans_unlocked_error(struct btree_trans *trans) +void __noreturn bch2_trans_unlocked_or_in_restart_error(struct btree_trans *trans) { - panic("trans should be locked, unlocked by %pS\n", - (void *) trans->last_unlock_ip); + if (trans->restarted) + bch2_trans_in_restart_error(trans); + + if (!trans->locked) + panic("trans should be locked, unlocked by %pS\n", + (void *) trans->last_unlock_ip); + + BUG(); } noinline __cold @@ -1450,10 +1484,11 @@ void bch2_trans_updates_to_text(struct printbuf *buf, struct btree_trans *trans) trans_for_each_update(trans, i) { struct bkey_s_c old = { &i->old_k, i->old_v }; - prt_printf(buf, "update: btree=%s cached=%u %pS\n", - bch2_btree_id_str(i->btree_id), - i->cached, - (void *) i->ip_allocated); + prt_str(buf, "update: btree="); + bch2_btree_id_to_text(buf, i->btree_id); + prt_printf(buf, " cached=%u %pS\n", + i->cached, + (void *) i->ip_allocated); prt_printf(buf, " old "); bch2_bkey_val_to_text(buf, trans->c, old); @@ -1486,13 +1521,13 @@ static void bch2_btree_path_to_text_short(struct printbuf *out, struct btree_tra { struct btree_path *path = trans->paths + path_idx; - prt_printf(out, "path: idx %3u ref %u:%u %c %c %c btree=%s l=%u pos ", + prt_printf(out, "path: idx %3u ref %u:%u %c %c %c ", path_idx, path->ref, path->intent_ref, path->preserve ? 'P' : ' ', path->should_be_locked ? 'S' : ' ', - path->cached ? 'C' : 'B', - bch2_btree_id_str(path->btree_id), - path->level); + path->cached ? 'C' : 'B'); + bch2_btree_id_level_to_text(out, path->btree_id, path->level); + prt_str(out, " pos "); bch2_bpos_to_text(out, path->pos); if (!path->cached && btree_node_locked(path, path->level)) { @@ -1717,8 +1752,7 @@ btree_path_idx_t bch2_path_get(struct btree_trans *trans, struct trans_for_each_path_inorder_iter iter; btree_path_idx_t path_pos = 0, path_idx; - bch2_trans_verify_not_unlocked(trans); - bch2_trans_verify_not_in_restart(trans); + bch2_trans_verify_not_unlocked_or_in_restart(trans); bch2_trans_verify_locks(trans); btree_trans_sort_paths(trans); @@ -1833,7 +1867,7 @@ struct bkey_s_c bch2_btree_path_peek_slot(struct btree_path *path, struct bkey * !bkey_eq(path->pos, ck->key.pos)); *u = ck->k->k; - k = bkey_i_to_s_c(ck->k); + k = (struct bkey_s_c) { u, &ck->k->v }; } return k; @@ -1843,7 +1877,6 @@ hole: return (struct bkey_s_c) { u, NULL }; } - void bch2_set_btree_iter_dontneed(struct btree_iter *iter) { struct btree_trans *trans = iter->trans; @@ -1870,7 +1903,7 @@ bch2_btree_iter_traverse(struct btree_iter *iter) struct btree_trans *trans = iter->trans; int ret; - bch2_trans_verify_not_unlocked(trans); + bch2_trans_verify_not_unlocked_or_in_restart(trans); iter->path = bch2_btree_path_set_pos(trans, iter->path, btree_iter_search_key(iter), @@ -1945,7 +1978,7 @@ struct btree *bch2_btree_iter_next_node(struct btree_iter *iter) int ret; EBUG_ON(trans->paths[iter->path].cached); - bch2_trans_verify_not_in_restart(trans); + bch2_trans_verify_not_unlocked_or_in_restart(trans); bch2_btree_iter_verify(iter); ret = bch2_btree_path_traverse(trans, iter->path, iter->flags); @@ -2101,7 +2134,7 @@ static struct bkey_i *bch2_btree_journal_peek(struct btree_trans *trans, { struct btree_path *path = btree_iter_path(trans, iter); - return bch2_journal_keys_peek_upto(trans->c, iter->btree_id, + return bch2_journal_keys_peek_max(trans->c, iter->btree_id, path->level, path->pos, end_pos, @@ -2124,21 +2157,47 @@ struct bkey_s_c btree_trans_peek_slot_journal(struct btree_trans *trans, } static noinline -struct bkey_s_c btree_trans_peek_journal(struct btree_trans *trans, - struct btree_iter *iter, - struct bkey_s_c k) +void btree_trans_peek_journal(struct btree_trans *trans, + struct btree_iter *iter, + struct bkey_s_c *k) { struct btree_path *path = btree_iter_path(trans, iter); struct bkey_i *next_journal = bch2_btree_journal_peek(trans, iter, - k.k ? k.k->p : path_l(path)->b->key.k.p); - + k->k ? k->k->p : path_l(path)->b->key.k.p); if (next_journal) { iter->k = next_journal->k; - k = bkey_i_to_s_c(next_journal); + *k = bkey_i_to_s_c(next_journal); } +} - return k; +static struct bkey_i *bch2_btree_journal_peek_prev(struct btree_trans *trans, + struct btree_iter *iter, + struct bpos end_pos) +{ + struct btree_path *path = btree_iter_path(trans, iter); + + return bch2_journal_keys_peek_prev_min(trans->c, iter->btree_id, + path->level, + path->pos, + end_pos, + &iter->journal_idx); +} + +static noinline +void btree_trans_peek_prev_journal(struct btree_trans *trans, + struct btree_iter *iter, + struct bkey_s_c *k) +{ + struct btree_path *path = btree_iter_path(trans, iter); + struct bkey_i *next_journal = + bch2_btree_journal_peek_prev(trans, iter, + k->k ? k->k->p : path_l(path)->b->key.k.p); + + if (next_journal) { + iter->k = next_journal->k; + *k = bkey_i_to_s_c(next_journal); + } } /* @@ -2154,8 +2213,7 @@ struct bkey_s_c btree_trans_peek_key_cache(struct btree_iter *iter, struct bpos struct bkey_s_c k; int ret; - bch2_trans_verify_not_in_restart(trans); - bch2_trans_verify_not_unlocked(trans); + bch2_trans_verify_not_unlocked_or_in_restart(trans); if ((iter->flags & BTREE_ITER_key_cache_fill) && bpos_eq(iter->pos, pos)) @@ -2184,10 +2242,15 @@ struct bkey_s_c btree_trans_peek_key_cache(struct btree_iter *iter, struct bpos btree_path_set_should_be_locked(trans, trans->paths + iter->key_cache_path); k = bch2_btree_path_peek_slot(trans->paths + iter->key_cache_path, &u); - if (k.k && !bkey_err(k)) { - iter->k = u; - k.k = &iter->k; - } + if (!k.k) + return k; + + if ((iter->flags & BTREE_ITER_all_snapshots) && + !bpos_eq(pos, k.k->p)) + return bkey_s_c_null; + + iter->k = u; + k.k = &iter->k; return k; } @@ -2201,8 +2264,6 @@ static struct bkey_s_c __bch2_btree_iter_peek(struct btree_iter *iter, struct bp bch2_btree_iter_verify(iter); while (1) { - struct btree_path_level *l; - iter->path = bch2_btree_path_set_pos(trans, iter->path, search_key, iter->flags & BTREE_ITER_intent, btree_iter_ip_allocated(iter)); @@ -2212,17 +2273,17 @@ static struct bkey_s_c __bch2_btree_iter_peek(struct btree_iter *iter, struct bp /* ensure that iter->k is consistent with iter->pos: */ bch2_btree_iter_set_pos(iter, iter->pos); k = bkey_s_c_err(ret); - goto out; + break; } struct btree_path *path = btree_iter_path(trans, iter); - l = path_l(path); + struct btree_path_level *l = path_l(path); if (unlikely(!l->b)) { /* No btree nodes at requested level: */ bch2_btree_iter_set_pos(iter, SPOS_MAX); k = bkey_s_c_null; - goto out; + break; } btree_path_set_should_be_locked(trans, path); @@ -2233,15 +2294,14 @@ static struct bkey_s_c __bch2_btree_iter_peek(struct btree_iter *iter, struct bp k.k && (k2 = btree_trans_peek_key_cache(iter, k.k->p)).k) { k = k2; - ret = bkey_err(k); - if (ret) { + if (bkey_err(k)) { bch2_btree_iter_set_pos(iter, iter->pos); - goto out; + break; } } if (unlikely(iter->flags & BTREE_ITER_with_journal)) - k = btree_trans_peek_journal(trans, iter, k); + btree_trans_peek_journal(trans, iter, &k); if (unlikely((iter->flags & BTREE_ITER_with_updates) && trans->nr_updates)) @@ -2270,32 +2330,32 @@ static struct bkey_s_c __bch2_btree_iter_peek(struct btree_iter *iter, struct bp /* End of btree: */ bch2_btree_iter_set_pos(iter, SPOS_MAX); k = bkey_s_c_null; - goto out; + break; } } -out: - bch2_btree_iter_verify(iter); + bch2_btree_iter_verify(iter); return k; } /** - * bch2_btree_iter_peek_upto() - returns first key greater than or equal to + * bch2_btree_iter_peek_max() - returns first key greater than or equal to * iterator's current position * @iter: iterator to peek from * @end: search limit: returns keys less than or equal to @end * * Returns: key if found, or an error extractable with bkey_err(). */ -struct bkey_s_c bch2_btree_iter_peek_upto(struct btree_iter *iter, struct bpos end) +struct bkey_s_c bch2_btree_iter_peek_max(struct btree_iter *iter, struct bpos end) { struct btree_trans *trans = iter->trans; struct bpos search_key = btree_iter_search_key(iter); struct bkey_s_c k; - struct bpos iter_pos; + struct bpos iter_pos = iter->pos; int ret; - bch2_trans_verify_not_unlocked(trans); + bch2_trans_verify_not_unlocked_or_in_restart(trans); + bch2_btree_iter_verify_entry_exit(iter); EBUG_ON((iter->flags & BTREE_ITER_filter_snapshots) && bkey_eq(end, POS_MAX)); if (iter->update_path) { @@ -2304,8 +2364,6 @@ struct bkey_s_c bch2_btree_iter_peek_upto(struct btree_iter *iter, struct bpos e iter->update_path = 0; } - bch2_btree_iter_verify_entry_exit(iter); - while (1) { k = __bch2_btree_iter_peek(iter, search_key); if (unlikely(!k.k)) @@ -2313,75 +2371,75 @@ struct bkey_s_c bch2_btree_iter_peek_upto(struct btree_iter *iter, struct bpos e if (unlikely(bkey_err(k))) goto out_no_locked; - /* - * We need to check against @end before FILTER_SNAPSHOTS because - * if we get to a different inode that requested we might be - * seeing keys for a different snapshot tree that will all be - * filtered out. - * - * But we can't do the full check here, because bkey_start_pos() - * isn't monotonically increasing before FILTER_SNAPSHOTS, and - * that's what we check against in extents mode: - */ - if (unlikely(!(iter->flags & BTREE_ITER_is_extents) - ? bkey_gt(k.k->p, end) - : k.k->p.inode > end.inode)) - goto end; + if (iter->flags & BTREE_ITER_filter_snapshots) { + /* + * We need to check against @end before FILTER_SNAPSHOTS because + * if we get to a different inode that requested we might be + * seeing keys for a different snapshot tree that will all be + * filtered out. + * + * But we can't do the full check here, because bkey_start_pos() + * isn't monotonically increasing before FILTER_SNAPSHOTS, and + * that's what we check against in extents mode: + */ + if (unlikely(!(iter->flags & BTREE_ITER_is_extents) + ? bkey_gt(k.k->p, end) + : k.k->p.inode > end.inode)) + goto end; + + if (iter->update_path && + !bkey_eq(trans->paths[iter->update_path].pos, k.k->p)) { + bch2_path_put_nokeep(trans, iter->update_path, + iter->flags & BTREE_ITER_intent); + iter->update_path = 0; + } - if (iter->update_path && - !bkey_eq(trans->paths[iter->update_path].pos, k.k->p)) { - bch2_path_put_nokeep(trans, iter->update_path, - iter->flags & BTREE_ITER_intent); - iter->update_path = 0; - } + if ((iter->flags & BTREE_ITER_intent) && + !(iter->flags & BTREE_ITER_is_extents) && + !iter->update_path) { + struct bpos pos = k.k->p; - if ((iter->flags & BTREE_ITER_filter_snapshots) && - (iter->flags & BTREE_ITER_intent) && - !(iter->flags & BTREE_ITER_is_extents) && - !iter->update_path) { - struct bpos pos = k.k->p; + if (pos.snapshot < iter->snapshot) { + search_key = bpos_successor(k.k->p); + continue; + } - if (pos.snapshot < iter->snapshot) { - search_key = bpos_successor(k.k->p); - continue; - } + pos.snapshot = iter->snapshot; - pos.snapshot = iter->snapshot; + /* + * advance, same as on exit for iter->path, but only up + * to snapshot + */ + __btree_path_get(trans, trans->paths + iter->path, iter->flags & BTREE_ITER_intent); + iter->update_path = iter->path; + + iter->update_path = bch2_btree_path_set_pos(trans, + iter->update_path, pos, + iter->flags & BTREE_ITER_intent, + _THIS_IP_); + ret = bch2_btree_path_traverse(trans, iter->update_path, iter->flags); + if (unlikely(ret)) { + k = bkey_s_c_err(ret); + goto out_no_locked; + } + } /* - * advance, same as on exit for iter->path, but only up - * to snapshot + * We can never have a key in a leaf node at POS_MAX, so + * we don't have to check these successor() calls: */ - __btree_path_get(trans, trans->paths + iter->path, iter->flags & BTREE_ITER_intent); - iter->update_path = iter->path; - - iter->update_path = bch2_btree_path_set_pos(trans, - iter->update_path, pos, - iter->flags & BTREE_ITER_intent, - _THIS_IP_); - ret = bch2_btree_path_traverse(trans, iter->update_path, iter->flags); - if (unlikely(ret)) { - k = bkey_s_c_err(ret); - goto out_no_locked; + if (!bch2_snapshot_is_ancestor(trans->c, + iter->snapshot, + k.k->p.snapshot)) { + search_key = bpos_successor(k.k->p); + continue; } - } - - /* - * We can never have a key in a leaf node at POS_MAX, so - * we don't have to check these successor() calls: - */ - if ((iter->flags & BTREE_ITER_filter_snapshots) && - !bch2_snapshot_is_ancestor(trans->c, - iter->snapshot, - k.k->p.snapshot)) { - search_key = bpos_successor(k.k->p); - continue; - } - if (bkey_whiteout(k.k) && - !(iter->flags & BTREE_ITER_all_snapshots)) { - search_key = bkey_successor(iter, k.k->p); - continue; + if (bkey_whiteout(k.k) && + !(iter->flags & BTREE_ITER_key_cache_fill)) { + search_key = bkey_successor(iter, k.k->p); + continue; + } } /* @@ -2451,127 +2509,204 @@ struct bkey_s_c bch2_btree_iter_next(struct btree_iter *iter) return bch2_btree_iter_peek(iter); } -/** - * bch2_btree_iter_peek_prev() - returns first key less than or equal to - * iterator's current position - * @iter: iterator to peek from - * - * Returns: key if found, or an error extractable with bkey_err(). - */ -struct bkey_s_c bch2_btree_iter_peek_prev(struct btree_iter *iter) +static struct bkey_s_c __bch2_btree_iter_peek_prev(struct btree_iter *iter, struct bpos search_key) { struct btree_trans *trans = iter->trans; - struct bpos search_key = iter->pos; - struct bkey_s_c k; - struct bkey saved_k; - const struct bch_val *saved_v; - btree_path_idx_t saved_path = 0; - int ret; - - bch2_trans_verify_not_unlocked(trans); - EBUG_ON(btree_iter_path(trans, iter)->cached || - btree_iter_path(trans, iter)->level); - - if (iter->flags & BTREE_ITER_with_journal) - return bkey_s_c_err(-BCH_ERR_btree_iter_with_journal_not_supported); + struct bkey_s_c k, k2; bch2_btree_iter_verify(iter); - bch2_btree_iter_verify_entry_exit(iter); - - if (iter->flags & BTREE_ITER_filter_snapshots) - search_key.snapshot = U32_MAX; while (1) { iter->path = bch2_btree_path_set_pos(trans, iter->path, search_key, - iter->flags & BTREE_ITER_intent, - btree_iter_ip_allocated(iter)); + iter->flags & BTREE_ITER_intent, + btree_iter_ip_allocated(iter)); - ret = bch2_btree_path_traverse(trans, iter->path, iter->flags); + int ret = bch2_btree_path_traverse(trans, iter->path, iter->flags); if (unlikely(ret)) { /* ensure that iter->k is consistent with iter->pos: */ bch2_btree_iter_set_pos(iter, iter->pos); k = bkey_s_c_err(ret); - goto out_no_locked; + break; } struct btree_path *path = btree_iter_path(trans, iter); + struct btree_path_level *l = path_l(path); + + if (unlikely(!l->b)) { + /* No btree nodes at requested level: */ + bch2_btree_iter_set_pos(iter, SPOS_MAX); + k = bkey_s_c_null; + break; + } + + btree_path_set_should_be_locked(trans, path); - k = btree_path_level_peek(trans, path, &path->l[0], &iter->k); - if (!k.k || - ((iter->flags & BTREE_ITER_is_extents) - ? bpos_ge(bkey_start_pos(k.k), search_key) - : bpos_gt(k.k->p, search_key))) - k = btree_path_level_prev(trans, path, &path->l[0], &iter->k); + k = btree_path_level_peek_all(trans->c, l, &iter->k); + if (!k.k || bpos_gt(k.k->p, search_key)) { + k = btree_path_level_prev(trans, path, l, &iter->k); + + BUG_ON(k.k && bpos_gt(k.k->p, search_key)); + } + + if (unlikely(iter->flags & BTREE_ITER_with_key_cache) && + k.k && + (k2 = btree_trans_peek_key_cache(iter, k.k->p)).k) { + k = k2; + if (bkey_err(k2)) { + bch2_btree_iter_set_pos(iter, iter->pos); + break; + } + } + + if (unlikely(iter->flags & BTREE_ITER_with_journal)) + btree_trans_peek_prev_journal(trans, iter, &k); if (unlikely((iter->flags & BTREE_ITER_with_updates) && trans->nr_updates)) bch2_btree_trans_peek_prev_updates(trans, iter, &k); - if (likely(k.k)) { - if (iter->flags & BTREE_ITER_filter_snapshots) { - if (k.k->p.snapshot == iter->snapshot) - goto got_key; + if (likely(k.k && !bkey_deleted(k.k))) { + break; + } else if (k.k) { + search_key = bpos_predecessor(k.k->p); + } else if (likely(!bpos_eq(path->l[0].b->data->min_key, POS_MIN))) { + /* Advance to previous leaf node: */ + search_key = bpos_predecessor(path->l[0].b->data->min_key); + } else { + /* Start of btree: */ + bch2_btree_iter_set_pos(iter, POS_MIN); + k = bkey_s_c_null; + break; + } + } + + bch2_btree_iter_verify(iter); + return k; +} + +/** + * bch2_btree_iter_peek_prev_min() - returns first key less than or equal to + * iterator's current position + * @iter: iterator to peek from + * @end: search limit: returns keys greater than or equal to @end + * + * Returns: key if found, or an error extractable with bkey_err(). + */ +struct bkey_s_c bch2_btree_iter_peek_prev_min(struct btree_iter *iter, struct bpos end) +{ + if ((iter->flags & (BTREE_ITER_is_extents|BTREE_ITER_filter_snapshots)) && + !bkey_eq(iter->pos, POS_MAX)) { + /* + * bkey_start_pos(), for extents, is not monotonically + * increasing until after filtering for snapshots: + * + * Thus, for extents we need to search forward until we find a + * real visible extents - easiest to just use peek_slot() (which + * internally uses peek() for extents) + */ + struct bkey_s_c k = bch2_btree_iter_peek_slot(iter); + if (bkey_err(k)) + return k; + + if (!bkey_deleted(k.k) && + (!(iter->flags & BTREE_ITER_is_extents) || + bkey_lt(bkey_start_pos(k.k), iter->pos))) + return k; + } + + struct btree_trans *trans = iter->trans; + struct bpos search_key = iter->pos; + struct bkey_s_c k; + btree_path_idx_t saved_path = 0; + + bch2_trans_verify_not_unlocked_or_in_restart(trans); + bch2_btree_iter_verify_entry_exit(iter); + EBUG_ON((iter->flags & BTREE_ITER_filter_snapshots) && bpos_eq(end, POS_MIN)); + + while (1) { + k = __bch2_btree_iter_peek_prev(iter, search_key); + if (unlikely(!k.k)) + goto end; + if (unlikely(bkey_err(k))) + goto out_no_locked; + if (iter->flags & BTREE_ITER_filter_snapshots) { + struct btree_path *s = saved_path ? trans->paths + saved_path : NULL; + if (s && bpos_lt(k.k->p, SPOS(s->pos.inode, s->pos.offset, iter->snapshot))) { /* - * If we have a saved candidate, and we're no - * longer at the same _key_ (not pos), return - * that candidate + * If we have a saved candidate, and we're past + * the last possible snapshot overwrite, return + * it: */ - if (saved_path && !bkey_eq(k.k->p, saved_k.p)) { - bch2_path_put_nokeep(trans, iter->path, - iter->flags & BTREE_ITER_intent); - iter->path = saved_path; + bch2_path_put_nokeep(trans, iter->path, + iter->flags & BTREE_ITER_intent); + iter->path = saved_path; + saved_path = 0; + k = bch2_btree_path_peek_slot(btree_iter_path(trans, iter), &iter->k); + break; + } + + /* + * We need to check against @end before FILTER_SNAPSHOTS because + * if we get to a different inode that requested we might be + * seeing keys for a different snapshot tree that will all be + * filtered out. + */ + if (unlikely(bkey_lt(k.k->p, end))) + goto end; + + if (!bch2_snapshot_is_ancestor(trans->c, iter->snapshot, k.k->p.snapshot)) { + search_key = bpos_predecessor(k.k->p); + continue; + } + + if (k.k->p.snapshot != iter->snapshot) { + /* + * Have a key visible in iter->snapshot, but + * might have overwrites: - save it and keep + * searching. Unless it's a whiteout - then drop + * our previous saved candidate: + */ + if (saved_path) { + bch2_path_put_nokeep(trans, saved_path, + iter->flags & BTREE_ITER_intent); saved_path = 0; - iter->k = saved_k; - k.v = saved_v; - goto got_key; } - if (bch2_snapshot_is_ancestor(trans->c, - iter->snapshot, - k.k->p.snapshot)) { - if (saved_path) - bch2_path_put_nokeep(trans, saved_path, - iter->flags & BTREE_ITER_intent); + if (!bkey_whiteout(k.k)) { saved_path = btree_path_clone(trans, iter->path, iter->flags & BTREE_ITER_intent, _THIS_IP_); - path = btree_iter_path(trans, iter); - trace_btree_path_save_pos(trans, path, trans->paths + saved_path); - saved_k = *k.k; - saved_v = k.v; + trace_btree_path_save_pos(trans, + trans->paths + iter->path, + trans->paths + saved_path); } search_key = bpos_predecessor(k.k->p); continue; } -got_key: - if (bkey_whiteout(k.k) && - !(iter->flags & BTREE_ITER_all_snapshots)) { + + if (bkey_whiteout(k.k)) { search_key = bkey_predecessor(iter, k.k->p); - if (iter->flags & BTREE_ITER_filter_snapshots) - search_key.snapshot = U32_MAX; + search_key.snapshot = U32_MAX; continue; } - - btree_path_set_should_be_locked(trans, path); - break; - } else if (likely(!bpos_eq(path->l[0].b->data->min_key, POS_MIN))) { - /* Advance to previous leaf node: */ - search_key = bpos_predecessor(path->l[0].b->data->min_key); - } else { - /* Start of btree: */ - bch2_btree_iter_set_pos(iter, POS_MIN); - k = bkey_s_c_null; - goto out_no_locked; } - } - EBUG_ON(bkey_gt(bkey_start_pos(k.k), iter->pos)); + EBUG_ON(iter->flags & BTREE_ITER_all_snapshots ? bpos_gt(k.k->p, iter->pos) : + iter->flags & BTREE_ITER_is_extents ? bkey_ge(bkey_start_pos(k.k), iter->pos) : + bkey_gt(k.k->p, iter->pos)); + + if (unlikely(iter->flags & BTREE_ITER_all_snapshots ? bpos_lt(k.k->p, end) : + iter->flags & BTREE_ITER_is_extents ? bkey_le(k.k->p, end) : + bkey_lt(k.k->p, end))) + goto end; + + break; + } /* Extents can straddle iter->pos: */ - if (bkey_lt(k.k->p, iter->pos)) - iter->pos = k.k->p; + iter->pos = bpos_min(iter->pos, k.k->p);; if (iter->flags & BTREE_ITER_filter_snapshots) iter->pos.snapshot = iter->snapshot; @@ -2581,8 +2716,11 @@ out_no_locked: bch2_btree_iter_verify_entry_exit(iter); bch2_btree_iter_verify(iter); - return k; +end: + bch2_btree_iter_set_pos(iter, end); + k = bkey_s_c_null; + goto out_no_locked; } /** @@ -2607,7 +2745,7 @@ struct bkey_s_c bch2_btree_iter_peek_slot(struct btree_iter *iter) struct bkey_s_c k; int ret; - bch2_trans_verify_not_unlocked(trans); + bch2_trans_verify_not_unlocked_or_in_restart(trans); bch2_btree_iter_verify(iter); bch2_btree_iter_verify_entry_exit(iter); EBUG_ON(btree_iter_path(trans, iter)->level && (iter->flags & BTREE_ITER_with_key_cache)); @@ -2632,6 +2770,10 @@ struct bkey_s_c bch2_btree_iter_peek_slot(struct btree_iter *iter) goto out_no_locked; } + struct btree_path *path = btree_iter_path(trans, iter); + if (unlikely(!btree_path_node(path, path->level))) + return bkey_s_c_null; + if ((iter->flags & BTREE_ITER_cached) || !(iter->flags & (BTREE_ITER_is_extents|BTREE_ITER_filter_snapshots))) { k = bkey_s_c_null; @@ -2658,6 +2800,11 @@ struct bkey_s_c bch2_btree_iter_peek_slot(struct btree_iter *iter) k = bch2_btree_path_peek_slot(trans->paths + iter->path, &iter->k); if (unlikely(!k.k)) goto out_no_locked; + + if (unlikely(k.k->type == KEY_TYPE_whiteout && + (iter->flags & BTREE_ITER_filter_snapshots) && + !(iter->flags & BTREE_ITER_key_cache_fill))) + iter->k.type = KEY_TYPE_deleted; } else { struct bpos next; struct bpos end = iter->pos; @@ -2671,7 +2818,7 @@ struct bkey_s_c bch2_btree_iter_peek_slot(struct btree_iter *iter) struct btree_iter iter2; bch2_trans_copy_iter(&iter2, iter); - k = bch2_btree_iter_peek_upto(&iter2, end); + k = bch2_btree_iter_peek_max(&iter2, end); if (k.k && !bkey_err(k)) { swap(iter->key_cache_path, iter2.key_cache_path); @@ -2682,7 +2829,7 @@ struct bkey_s_c bch2_btree_iter_peek_slot(struct btree_iter *iter) } else { struct bpos pos = iter->pos; - k = bch2_btree_iter_peek_upto(iter, end); + k = bch2_btree_iter_peek_max(iter, end); if (unlikely(bkey_err(k))) bch2_btree_iter_set_pos(iter, pos); else @@ -2902,7 +3049,7 @@ void bch2_trans_iter_init_outlined(struct btree_trans *trans, unsigned flags) { bch2_trans_iter_init_common(trans, iter, btree_id, pos, 0, 0, - bch2_btree_iter_flags(trans, btree_id, flags), + bch2_btree_iter_flags(trans, btree_id, 0, flags), _RET_IP_); } @@ -2918,8 +3065,11 @@ void bch2_trans_node_iter_init(struct btree_trans *trans, flags |= BTREE_ITER_snapshot_field; flags |= BTREE_ITER_all_snapshots; + if (!depth && btree_id_cached(trans->c, btree_id)) + flags |= BTREE_ITER_with_key_cache; + bch2_trans_iter_init_common(trans, iter, btree_id, pos, locks_want, depth, - __bch2_btree_iter_flags(trans, btree_id, flags), + bch2_btree_iter_flags(trans, btree_id, depth, flags), _RET_IP_); iter->min_depth = depth; @@ -3122,14 +3272,14 @@ u32 bch2_trans_begin(struct btree_trans *trans) trans->last_begin_ip = _RET_IP_; - trans_set_locked(trans); + trans_set_locked(trans, false); if (trans->restarted) { bch2_btree_path_traverse_all(trans); trans->notrace_relock_fail = false; } - bch2_trans_verify_not_unlocked(trans); + bch2_trans_verify_not_unlocked_or_in_restart(trans); return trans->restart_count; } @@ -3228,7 +3378,7 @@ got_trans: trans->srcu_idx = srcu_read_lock(&c->btree_trans_barrier); trans->srcu_lock_time = jiffies; trans->srcu_held = true; - trans_set_locked(trans); + trans_set_locked(trans, false); closure_init_stack_release(&trans->ref); return trans; @@ -3262,6 +3412,9 @@ void bch2_trans_put(struct btree_trans *trans) { struct bch_fs *c = trans->c; + if (trans->restarted) + bch2_trans_in_restart_error(trans); + bch2_trans_unlock(trans); trans_for_each_update(trans, i) @@ -3285,6 +3438,10 @@ void bch2_trans_put(struct btree_trans *trans) closure_return_sync(&trans->ref); trans->locking_wait.task = NULL; +#ifdef CONFIG_BCACHEFS_DEBUG + darray_exit(&trans->last_restarted_trace); +#endif + unsigned long *paths_allocated = trans->paths_allocated; trans->paths_allocated = NULL; trans->paths = NULL; @@ -3338,8 +3495,9 @@ bch2_btree_bkey_cached_common_to_text(struct printbuf *out, pid = owner ? owner->pid : 0; rcu_read_unlock(); - prt_printf(out, "\t%px %c l=%u %s:", b, b->cached ? 'c' : 'b', - b->level, bch2_btree_id_str(b->btree_id)); + prt_printf(out, "\t%px %c ", b, b->cached ? 'c' : 'b'); + bch2_btree_id_to_text(out, b->btree_id); + prt_printf(out, " l=%u:", b->level); bch2_bpos_to_text(out, btree_node_pos(b)); prt_printf(out, "\t locks %u:%u:%u held by pid %u", @@ -3378,11 +3536,11 @@ void bch2_btree_trans_to_text(struct printbuf *out, struct btree_trans *trans) if (!path->nodes_locked) continue; - prt_printf(out, " path %u %c l=%u %s:", - idx, - path->cached ? 'c' : 'b', - path->level, - bch2_btree_id_str(path->btree_id)); + prt_printf(out, " path %u %c ", + idx, + path->cached ? 'c' : 'b'); + bch2_btree_id_to_text(out, path->btree_id); + prt_printf(out, " l=%u:", path->level); bch2_bpos_to_text(out, path->pos); prt_newline(out); @@ -3488,7 +3646,7 @@ int bch2_fs_btree_iter_init(struct bch_fs *c) #ifdef CONFIG_LOCKDEP fs_reclaim_acquire(GFP_KERNEL); struct btree_trans *trans = bch2_trans_get(c); - trans_set_locked(trans); + trans_set_locked(trans, false); bch2_trans_put(trans); fs_reclaim_release(GFP_KERNEL); #endif diff --git a/fs/bcachefs/btree_iter.h b/fs/bcachefs/btree_iter.h index 0bda054f80d7..b9538e6e6d65 100644 --- a/fs/bcachefs/btree_iter.h +++ b/fs/bcachefs/btree_iter.h @@ -23,6 +23,7 @@ static inline void __btree_path_get(struct btree_trans *trans, struct btree_path { unsigned idx = path - trans->paths; + EBUG_ON(idx >= trans->nr_paths); EBUG_ON(!test_bit(idx, trans->paths_allocated)); if (unlikely(path->ref == U8_MAX)) { bch2_dump_trans_paths_updates(trans); @@ -36,6 +37,7 @@ static inline void __btree_path_get(struct btree_trans *trans, struct btree_path static inline bool __btree_path_put(struct btree_trans *trans, struct btree_path *path, bool intent) { + EBUG_ON(path - trans->paths >= trans->nr_paths); EBUG_ON(!test_bit(path - trans->paths, trans->paths_allocated)); EBUG_ON(!path->ref); EBUG_ON(!path->intent_ref && intent); @@ -234,12 +236,12 @@ int __must_check bch2_btree_path_traverse_one(struct btree_trans *, btree_path_idx_t, unsigned, unsigned long); -static inline void bch2_trans_verify_not_unlocked(struct btree_trans *); +static inline void bch2_trans_verify_not_unlocked_or_in_restart(struct btree_trans *); static inline int __must_check bch2_btree_path_traverse(struct btree_trans *trans, btree_path_idx_t path, unsigned flags) { - bch2_trans_verify_not_unlocked(trans); + bch2_trans_verify_not_unlocked_or_in_restart(trans); if (trans->paths[path].uptodate < BTREE_ITER_NEED_RELOCK) return 0; @@ -324,38 +326,33 @@ static inline void bch2_trans_verify_not_restarted(struct btree_trans *trans, bch2_trans_restart_error(trans, restart_count); } -void __noreturn bch2_trans_in_restart_error(struct btree_trans *); +void __noreturn bch2_trans_unlocked_or_in_restart_error(struct btree_trans *); -static inline void bch2_trans_verify_not_in_restart(struct btree_trans *trans) +static inline void bch2_trans_verify_not_unlocked_or_in_restart(struct btree_trans *trans) { - if (trans->restarted) - bch2_trans_in_restart_error(trans); -} - -void __noreturn bch2_trans_unlocked_error(struct btree_trans *); - -static inline void bch2_trans_verify_not_unlocked(struct btree_trans *trans) -{ - if (!trans->locked) - bch2_trans_unlocked_error(trans); + if (trans->restarted || !trans->locked) + bch2_trans_unlocked_or_in_restart_error(trans); } __always_inline -static int btree_trans_restart_nounlock(struct btree_trans *trans, int err) +static int btree_trans_restart_ip(struct btree_trans *trans, int err, unsigned long ip) { BUG_ON(err <= 0); BUG_ON(!bch2_err_matches(-err, BCH_ERR_transaction_restart)); trans->restarted = err; - trans->last_restarted_ip = _THIS_IP_; + trans->last_restarted_ip = ip; +#ifdef CONFIG_BCACHEFS_DEBUG + darray_exit(&trans->last_restarted_trace); + bch2_save_backtrace(&trans->last_restarted_trace, current, 0, GFP_NOWAIT); +#endif return -err; } __always_inline static int btree_trans_restart(struct btree_trans *trans, int err) { - btree_trans_restart_nounlock(trans, err); - return -err; + return btree_trans_restart_ip(trans, err, _THIS_IP_); } bool bch2_btree_node_upgrade(struct btree_trans *, @@ -375,6 +372,7 @@ static inline void bch2_btree_path_downgrade(struct btree_trans *trans, void bch2_trans_downgrade(struct btree_trans *); void bch2_trans_node_add(struct btree_trans *trans, struct btree_path *, struct btree *); +void bch2_trans_node_drop(struct btree_trans *trans, struct btree *); void bch2_trans_node_reinit_iter(struct btree_trans *, struct btree *); int __must_check __bch2_btree_iter_traverse(struct btree_iter *iter); @@ -384,15 +382,21 @@ struct btree *bch2_btree_iter_peek_node(struct btree_iter *); struct btree *bch2_btree_iter_peek_node_and_restart(struct btree_iter *); struct btree *bch2_btree_iter_next_node(struct btree_iter *); -struct bkey_s_c bch2_btree_iter_peek_upto(struct btree_iter *, struct bpos); +struct bkey_s_c bch2_btree_iter_peek_max(struct btree_iter *, struct bpos); struct bkey_s_c bch2_btree_iter_next(struct btree_iter *); static inline struct bkey_s_c bch2_btree_iter_peek(struct btree_iter *iter) { - return bch2_btree_iter_peek_upto(iter, SPOS_MAX); + return bch2_btree_iter_peek_max(iter, SPOS_MAX); +} + +struct bkey_s_c bch2_btree_iter_peek_prev_min(struct btree_iter *, struct bpos); + +static inline struct bkey_s_c bch2_btree_iter_peek_prev(struct btree_iter *iter) +{ + return bch2_btree_iter_peek_prev_min(iter, POS_MIN); } -struct bkey_s_c bch2_btree_iter_peek_prev(struct btree_iter *); struct bkey_s_c bch2_btree_iter_prev(struct btree_iter *); struct bkey_s_c bch2_btree_iter_peek_slot(struct btree_iter *); @@ -443,10 +447,17 @@ static inline void bch2_btree_iter_set_snapshot(struct btree_iter *iter, u32 sna void bch2_trans_iter_exit(struct btree_trans *, struct btree_iter *); -static inline unsigned __bch2_btree_iter_flags(struct btree_trans *trans, - unsigned btree_id, - unsigned flags) +static inline unsigned bch2_btree_iter_flags(struct btree_trans *trans, + unsigned btree_id, + unsigned level, + unsigned flags) { + if (level || !btree_id_cached(trans->c, btree_id)) { + flags &= ~BTREE_ITER_cached; + flags &= ~BTREE_ITER_with_key_cache; + } else if (!(flags & BTREE_ITER_cached)) + flags |= BTREE_ITER_with_key_cache; + if (!(flags & (BTREE_ITER_all_snapshots|BTREE_ITER_not_extents)) && btree_id_is_extents(btree_id)) flags |= BTREE_ITER_is_extents; @@ -465,19 +476,6 @@ static inline unsigned __bch2_btree_iter_flags(struct btree_trans *trans, return flags; } -static inline unsigned bch2_btree_iter_flags(struct btree_trans *trans, - unsigned btree_id, - unsigned flags) -{ - if (!btree_id_cached(trans->c, btree_id)) { - flags &= ~BTREE_ITER_cached; - flags &= ~BTREE_ITER_with_key_cache; - } else if (!(flags & BTREE_ITER_cached)) - flags |= BTREE_ITER_with_key_cache; - - return __bch2_btree_iter_flags(trans, btree_id, flags); -} - static inline void bch2_trans_iter_init_common(struct btree_trans *trans, struct btree_iter *iter, unsigned btree_id, struct bpos pos, @@ -514,7 +512,7 @@ static inline void bch2_trans_iter_init(struct btree_trans *trans, if (__builtin_constant_p(btree_id) && __builtin_constant_p(flags)) bch2_trans_iter_init_common(trans, iter, btree_id, pos, 0, 0, - bch2_btree_iter_flags(trans, btree_id, flags), + bch2_btree_iter_flags(trans, btree_id, 0, flags), _THIS_IP_); else bch2_trans_iter_init_outlined(trans, iter, btree_id, pos, flags); @@ -593,13 +591,18 @@ static inline struct bkey_s_c bch2_bkey_get_iter(struct btree_trans *trans, bkey_s_c_to_##_type(__bch2_bkey_get_iter(_trans, _iter, \ _btree_id, _pos, _flags, KEY_TYPE_##_type)) +static inline void __bkey_val_copy(void *dst_v, unsigned dst_size, struct bkey_s_c src_k) +{ + unsigned b = min_t(unsigned, dst_size, bkey_val_bytes(src_k.k)); + memcpy(dst_v, src_k.v, b); + if (unlikely(b < dst_size)) + memset(dst_v + b, 0, dst_size - b); +} + #define bkey_val_copy(_dst_v, _src_k) \ do { \ - unsigned b = min_t(unsigned, sizeof(*_dst_v), \ - bkey_val_bytes(_src_k.k)); \ - memcpy(_dst_v, _src_k.v, b); \ - if (b < sizeof(*_dst_v)) \ - memset((void *) (_dst_v) + b, 0, sizeof(*_dst_v) - b); \ + BUILD_BUG_ON(!__typecheck(*_dst_v, *_src_k.v)); \ + __bkey_val_copy(_dst_v, sizeof(*_dst_v), _src_k.s_c); \ } while (0) static inline int __bch2_bkey_get_val_typed(struct btree_trans *trans, @@ -608,17 +611,10 @@ static inline int __bch2_bkey_get_val_typed(struct btree_trans *trans, unsigned val_size, void *val) { struct btree_iter iter; - struct bkey_s_c k; - int ret; - - k = __bch2_bkey_get_iter(trans, &iter, btree_id, pos, flags, type); - ret = bkey_err(k); + struct bkey_s_c k = __bch2_bkey_get_iter(trans, &iter, btree_id, pos, flags, type); + int ret = bkey_err(k); if (!ret) { - unsigned b = min_t(unsigned, bkey_val_bytes(k.k), val_size); - - memcpy(val, k.v, b); - if (unlikely(b < sizeof(*val))) - memset((void *) val + b, 0, sizeof(*val) - b); + __bkey_val_copy(val, val_size, k); bch2_trans_iter_exit(trans, &iter); } @@ -677,12 +673,12 @@ static inline struct bkey_s_c bch2_btree_iter_peek_type(struct btree_iter *iter, bch2_btree_iter_peek(iter); } -static inline struct bkey_s_c bch2_btree_iter_peek_upto_type(struct btree_iter *iter, +static inline struct bkey_s_c bch2_btree_iter_peek_max_type(struct btree_iter *iter, struct bpos end, unsigned flags) { if (!(flags & BTREE_ITER_slots)) - return bch2_btree_iter_peek_upto(iter, end); + return bch2_btree_iter_peek_max(iter, end); if (bkey_gt(iter->pos, end)) return bkey_s_c_null; @@ -746,7 +742,7 @@ transaction_restart: \ _ret2 ?: trans_was_restarted(_trans, _restart_count); \ }) -#define for_each_btree_key_upto_continue(_trans, _iter, \ +#define for_each_btree_key_max_continue(_trans, _iter, \ _end, _flags, _k, _do) \ ({ \ struct bkey_s_c _k; \ @@ -754,7 +750,7 @@ transaction_restart: \ \ do { \ _ret3 = lockrestart_do(_trans, ({ \ - (_k) = bch2_btree_iter_peek_upto_type(&(_iter), \ + (_k) = bch2_btree_iter_peek_max_type(&(_iter), \ _end, (_flags)); \ if (!(_k).k) \ break; \ @@ -768,9 +764,9 @@ transaction_restart: \ }) #define for_each_btree_key_continue(_trans, _iter, _flags, _k, _do) \ - for_each_btree_key_upto_continue(_trans, _iter, SPOS_MAX, _flags, _k, _do) + for_each_btree_key_max_continue(_trans, _iter, SPOS_MAX, _flags, _k, _do) -#define for_each_btree_key_upto(_trans, _iter, _btree_id, \ +#define for_each_btree_key_max(_trans, _iter, _btree_id, \ _start, _end, _flags, _k, _do) \ ({ \ bch2_trans_begin(trans); \ @@ -779,12 +775,12 @@ transaction_restart: \ bch2_trans_iter_init((_trans), &(_iter), (_btree_id), \ (_start), (_flags)); \ \ - for_each_btree_key_upto_continue(_trans, _iter, _end, _flags, _k, _do);\ + for_each_btree_key_max_continue(_trans, _iter, _end, _flags, _k, _do);\ }) #define for_each_btree_key(_trans, _iter, _btree_id, \ _start, _flags, _k, _do) \ - for_each_btree_key_upto(_trans, _iter, _btree_id, _start, \ + for_each_btree_key_max(_trans, _iter, _btree_id, _start, \ SPOS_MAX, _flags, _k, _do) #define for_each_btree_key_reverse(_trans, _iter, _btree_id, \ @@ -828,33 +824,33 @@ transaction_restart: \ (_do) ?: bch2_trans_commit(_trans, (_disk_res),\ (_journal_seq), (_commit_flags))) -#define for_each_btree_key_upto_commit(_trans, _iter, _btree_id, \ +#define for_each_btree_key_max_commit(_trans, _iter, _btree_id, \ _start, _end, _iter_flags, _k, \ _disk_res, _journal_seq, _commit_flags,\ _do) \ - for_each_btree_key_upto(_trans, _iter, _btree_id, _start, _end, _iter_flags, _k,\ + for_each_btree_key_max(_trans, _iter, _btree_id, _start, _end, _iter_flags, _k,\ (_do) ?: bch2_trans_commit(_trans, (_disk_res),\ (_journal_seq), (_commit_flags))) struct bkey_s_c bch2_btree_iter_peek_and_restart_outlined(struct btree_iter *); -#define for_each_btree_key_upto_norestart(_trans, _iter, _btree_id, \ +#define for_each_btree_key_max_norestart(_trans, _iter, _btree_id, \ _start, _end, _flags, _k, _ret) \ for (bch2_trans_iter_init((_trans), &(_iter), (_btree_id), \ (_start), (_flags)); \ - (_k) = bch2_btree_iter_peek_upto_type(&(_iter), _end, _flags),\ + (_k) = bch2_btree_iter_peek_max_type(&(_iter), _end, _flags),\ !((_ret) = bkey_err(_k)) && (_k).k; \ bch2_btree_iter_advance(&(_iter))) -#define for_each_btree_key_upto_continue_norestart(_iter, _end, _flags, _k, _ret)\ +#define for_each_btree_key_max_continue_norestart(_iter, _end, _flags, _k, _ret)\ for (; \ - (_k) = bch2_btree_iter_peek_upto_type(&(_iter), _end, _flags), \ + (_k) = bch2_btree_iter_peek_max_type(&(_iter), _end, _flags), \ !((_ret) = bkey_err(_k)) && (_k).k; \ bch2_btree_iter_advance(&(_iter))) #define for_each_btree_key_norestart(_trans, _iter, _btree_id, \ _start, _flags, _k, _ret) \ - for_each_btree_key_upto_norestart(_trans, _iter, _btree_id, _start,\ + for_each_btree_key_max_norestart(_trans, _iter, _btree_id, _start,\ SPOS_MAX, _flags, _k, _ret) #define for_each_btree_key_reverse_norestart(_trans, _iter, _btree_id, \ @@ -866,7 +862,7 @@ struct bkey_s_c bch2_btree_iter_peek_and_restart_outlined(struct btree_iter *); bch2_btree_iter_rewind(&(_iter))) #define for_each_btree_key_continue_norestart(_iter, _flags, _k, _ret) \ - for_each_btree_key_upto_continue_norestart(_iter, SPOS_MAX, _flags, _k, _ret) + for_each_btree_key_max_continue_norestart(_iter, SPOS_MAX, _flags, _k, _ret) /* * This should not be used in a fastpath, without first trying _do in diff --git a/fs/bcachefs/btree_journal_iter.c b/fs/bcachefs/btree_journal_iter.c index c1657182c275..6d25e3f85ce8 100644 --- a/fs/bcachefs/btree_journal_iter.c +++ b/fs/bcachefs/btree_journal_iter.c @@ -16,6 +16,17 @@ * operations for the regular btree iter code to use: */ +static inline size_t pos_to_idx(struct journal_keys *keys, size_t pos) +{ + size_t gap_size = keys->size - keys->nr; + + BUG_ON(pos >= keys->gap && pos < keys->gap + gap_size); + + if (pos >= keys->gap) + pos -= gap_size; + return pos; +} + static inline size_t idx_to_pos(struct journal_keys *keys, size_t idx) { size_t gap_size = keys->size - keys->nr; @@ -61,7 +72,7 @@ static size_t bch2_journal_key_search(struct journal_keys *keys, } /* Returns first non-overwritten key >= search key: */ -struct bkey_i *bch2_journal_keys_peek_upto(struct bch_fs *c, enum btree_id btree_id, +struct bkey_i *bch2_journal_keys_peek_max(struct bch_fs *c, enum btree_id btree_id, unsigned level, struct bpos pos, struct bpos end_pos, size_t *idx) { @@ -84,27 +95,92 @@ search: } } + struct bkey_i *ret = NULL; + rcu_read_lock(); /* for overwritten_ranges */ + while ((k = *idx < keys->nr ? idx_to_key(keys, *idx) : NULL)) { if (__journal_key_cmp(btree_id, level, end_pos, k) < 0) - return NULL; + break; if (k->overwritten) { - (*idx)++; + if (k->overwritten_range) + *idx = rcu_dereference(k->overwritten_range)->end; + else + *idx += 1; continue; } - if (__journal_key_cmp(btree_id, level, pos, k) <= 0) - return k->k; + if (__journal_key_cmp(btree_id, level, pos, k) <= 0) { + ret = k->k; + break; + } (*idx)++; iters++; if (iters == 10) { *idx = 0; + rcu_read_unlock(); goto search; } } - return NULL; + rcu_read_unlock(); + return ret; +} + +struct bkey_i *bch2_journal_keys_peek_prev_min(struct bch_fs *c, enum btree_id btree_id, + unsigned level, struct bpos pos, + struct bpos end_pos, size_t *idx) +{ + struct journal_keys *keys = &c->journal_keys; + unsigned iters = 0; + struct journal_key *k; + + BUG_ON(*idx > keys->nr); +search: + if (!*idx) + *idx = __bch2_journal_key_search(keys, btree_id, level, pos); + + while (*idx && + __journal_key_cmp(btree_id, level, end_pos, idx_to_key(keys, *idx - 1)) <= 0) { + (*idx)++; + iters++; + if (iters == 10) { + *idx = 0; + goto search; + } + } + + struct bkey_i *ret = NULL; + rcu_read_lock(); /* for overwritten_ranges */ + + while ((k = *idx < keys->nr ? idx_to_key(keys, *idx) : NULL)) { + if (__journal_key_cmp(btree_id, level, end_pos, k) > 0) + break; + + if (k->overwritten) { + if (k->overwritten_range) + *idx = rcu_dereference(k->overwritten_range)->start - 1; + else + *idx -= 1; + continue; + } + + if (__journal_key_cmp(btree_id, level, pos, k) >= 0) { + ret = k->k; + break; + } + + --(*idx); + iters++; + if (iters == 10) { + *idx = 0; + goto search; + } + } + + rcu_read_unlock(); + return ret; } struct bkey_i *bch2_journal_keys_peek_slot(struct bch_fs *c, enum btree_id btree_id, @@ -112,11 +188,12 @@ struct bkey_i *bch2_journal_keys_peek_slot(struct bch_fs *c, enum btree_id btree { size_t idx = 0; - return bch2_journal_keys_peek_upto(c, btree_id, level, pos, pos, &idx); + return bch2_journal_keys_peek_max(c, btree_id, level, pos, pos, &idx); } static void journal_iter_verify(struct journal_iter *iter) { +#ifdef CONFIG_BCACHEFS_DEBUG struct journal_keys *keys = iter->keys; size_t gap_size = keys->size - keys->nr; @@ -126,10 +203,10 @@ static void journal_iter_verify(struct journal_iter *iter) if (iter->idx < keys->size) { struct journal_key *k = keys->data + iter->idx; - int cmp = cmp_int(k->btree_id, iter->btree_id) ?: - cmp_int(k->level, iter->level); - BUG_ON(cmp < 0); + int cmp = __journal_key_btree_cmp(iter->btree_id, iter->level, k); + BUG_ON(cmp > 0); } +#endif } static void journal_iters_fix(struct bch_fs *c) @@ -182,7 +259,7 @@ int bch2_journal_key_insert_take(struct bch_fs *c, enum btree_id id, * Ensure these keys are done last by journal replay, to unblock * journal reclaim: */ - .journal_seq = U32_MAX, + .journal_seq = U64_MAX, }; struct journal_keys *keys = &c->journal_keys; size_t idx = bch2_journal_key_search(keys, id, level, k->k.p); @@ -290,6 +367,68 @@ bool bch2_key_deleted_in_journal(struct btree_trans *trans, enum btree_id btree, bkey_deleted(&keys->data[idx].k->k)); } +static void __bch2_journal_key_overwritten(struct journal_keys *keys, size_t pos) +{ + struct journal_key *k = keys->data + pos; + size_t idx = pos_to_idx(keys, pos); + + k->overwritten = true; + + struct journal_key *prev = idx > 0 ? keys->data + idx_to_pos(keys, idx - 1) : NULL; + struct journal_key *next = idx + 1 < keys->nr ? keys->data + idx_to_pos(keys, idx + 1) : NULL; + + bool prev_overwritten = prev && prev->overwritten; + bool next_overwritten = next && next->overwritten; + + struct journal_key_range_overwritten *prev_range = + prev_overwritten ? prev->overwritten_range : NULL; + struct journal_key_range_overwritten *next_range = + next_overwritten ? next->overwritten_range : NULL; + + BUG_ON(prev_range && prev_range->end != idx); + BUG_ON(next_range && next_range->start != idx + 1); + + if (prev_range && next_range) { + prev_range->end = next_range->end; + + keys->data[pos].overwritten_range = prev_range; + for (size_t i = next_range->start; i < next_range->end; i++) { + struct journal_key *ip = keys->data + idx_to_pos(keys, i); + BUG_ON(ip->overwritten_range != next_range); + ip->overwritten_range = prev_range; + } + + kfree_rcu_mightsleep(next_range); + } else if (prev_range) { + prev_range->end++; + k->overwritten_range = prev_range; + if (next_overwritten) { + prev_range->end++; + next->overwritten_range = prev_range; + } + } else if (next_range) { + next_range->start--; + k->overwritten_range = next_range; + if (prev_overwritten) { + next_range->start--; + prev->overwritten_range = next_range; + } + } else if (prev_overwritten || next_overwritten) { + struct journal_key_range_overwritten *r = kmalloc(sizeof(*r), GFP_KERNEL); + if (!r) + return; + + r->start = idx - (size_t) prev_overwritten; + r->end = idx + 1 + (size_t) next_overwritten; + + rcu_assign_pointer(k->overwritten_range, r); + if (prev_overwritten) + prev->overwritten_range = r; + if (next_overwritten) + next->overwritten_range = r; + } +} + void bch2_journal_key_overwritten(struct bch_fs *c, enum btree_id btree, unsigned level, struct bpos pos) { @@ -299,8 +438,12 @@ void bch2_journal_key_overwritten(struct bch_fs *c, enum btree_id btree, if (idx < keys->size && keys->data[idx].btree_id == btree && keys->data[idx].level == level && - bpos_eq(keys->data[idx].k->k.p, pos)) - keys->data[idx].overwritten = true; + bpos_eq(keys->data[idx].k->k.p, pos) && + !keys->data[idx].overwritten) { + mutex_lock(&keys->overwrite_lock); + __bch2_journal_key_overwritten(keys, idx); + mutex_unlock(&keys->overwrite_lock); + } } static void bch2_journal_iter_advance(struct journal_iter *iter) @@ -314,24 +457,32 @@ static void bch2_journal_iter_advance(struct journal_iter *iter) static struct bkey_s_c bch2_journal_iter_peek(struct journal_iter *iter) { + struct bkey_s_c ret = bkey_s_c_null; + journal_iter_verify(iter); + rcu_read_lock(); while (iter->idx < iter->keys->size) { struct journal_key *k = iter->keys->data + iter->idx; - int cmp = cmp_int(k->btree_id, iter->btree_id) ?: - cmp_int(k->level, iter->level); - if (cmp > 0) + int cmp = __journal_key_btree_cmp(iter->btree_id, iter->level, k); + if (cmp < 0) break; BUG_ON(cmp); - if (!k->overwritten) - return bkey_i_to_s_c(k->k); + if (!k->overwritten) { + ret = bkey_i_to_s_c(k->k); + break; + } - bch2_journal_iter_advance(iter); + if (k->overwritten_range) + iter->idx = idx_to_pos(iter->keys, rcu_dereference(k->overwritten_range)->end); + else + bch2_journal_iter_advance(iter); } + rcu_read_unlock(); - return bkey_s_c_null; + return ret; } static void bch2_journal_iter_exit(struct journal_iter *iter) @@ -382,6 +533,7 @@ static void btree_and_journal_iter_prefetch(struct btree_and_journal_iter *_iter : (level > 1 ? 1 : 16); iter.prefetch = false; + iter.fail_if_too_many_whiteouts = true; bch2_bkey_buf_init(&tmp); while (nr--) { @@ -400,6 +552,7 @@ static void btree_and_journal_iter_prefetch(struct btree_and_journal_iter *_iter struct bkey_s_c bch2_btree_and_journal_iter_peek(struct btree_and_journal_iter *iter) { struct bkey_s_c btree_k, journal_k = bkey_s_c_null, ret; + size_t iters = 0; if (iter->prefetch && iter->journal.level) btree_and_journal_iter_prefetch(iter); @@ -407,6 +560,11 @@ again: if (iter->at_end) return bkey_s_c_null; + iters++; + + if (iters > 20 && iter->fail_if_too_many_whiteouts) + return bkey_s_c_null; + while ((btree_k = bch2_journal_iter_peek_btree(iter)).k && bpos_lt(btree_k.k->p, iter->pos)) bch2_journal_iter_advance_btree(iter); @@ -481,16 +639,6 @@ void bch2_btree_and_journal_iter_init_node_iter(struct btree_trans *trans, /* sort and dedup all keys in the journal: */ -void bch2_journal_entries_free(struct bch_fs *c) -{ - struct journal_replay **i; - struct genradix_iter iter; - - genradix_for_each(&c->journal_entries, iter, i) - kvfree(*i); - genradix_free(&c->journal_entries); -} - /* * When keys compare equal, oldest compares first: */ @@ -515,15 +663,26 @@ void bch2_journal_keys_put(struct bch_fs *c) move_gap(keys, keys->nr); - darray_for_each(*keys, i) + darray_for_each(*keys, i) { + if (i->overwritten_range && + (i == &darray_last(*keys) || + i->overwritten_range != i[1].overwritten_range)) + kfree(i->overwritten_range); + if (i->allocated) kfree(i->k); + } kvfree(keys->data); keys->data = NULL; keys->nr = keys->gap = keys->size = 0; - bch2_journal_entries_free(c); + struct journal_replay **i; + struct genradix_iter iter; + + genradix_for_each(&c->journal_entries, iter, i) + kvfree(*i); + genradix_free(&c->journal_entries); } static void __journal_keys_sort(struct journal_keys *keys) @@ -628,8 +787,20 @@ void bch2_journal_keys_dump(struct bch_fs *c) darray_for_each(*keys, i) { printbuf_reset(&buf); + prt_printf(&buf, "btree="); + bch2_btree_id_to_text(&buf, i->btree_id); + prt_printf(&buf, " l=%u ", i->level); bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(i->k)); - pr_err("%s l=%u %s", bch2_btree_id_str(i->btree_id), i->level, buf.buf); + pr_err("%s", buf.buf); } printbuf_exit(&buf); } + +void bch2_fs_journal_keys_init(struct bch_fs *c) +{ + struct journal_keys *keys = &c->journal_keys; + + atomic_set(&keys->ref, 1); + keys->initial_ref_held = true; + mutex_init(&keys->overwrite_lock); +} diff --git a/fs/bcachefs/btree_journal_iter.h b/fs/bcachefs/btree_journal_iter.h index 1653de9d609b..2a3082919b8d 100644 --- a/fs/bcachefs/btree_journal_iter.h +++ b/fs/bcachefs/btree_journal_iter.h @@ -26,16 +26,24 @@ struct btree_and_journal_iter { struct bpos pos; bool at_end; bool prefetch; + bool fail_if_too_many_whiteouts; }; +static inline int __journal_key_btree_cmp(enum btree_id l_btree_id, + unsigned l_level, + const struct journal_key *r) +{ + return -cmp_int(l_level, r->level) ?: + cmp_int(l_btree_id, r->btree_id); +} + static inline int __journal_key_cmp(enum btree_id l_btree_id, unsigned l_level, struct bpos l_pos, const struct journal_key *r) { - return (cmp_int(l_btree_id, r->btree_id) ?: - cmp_int(l_level, r->level) ?: - bpos_cmp(l_pos, r->k->k.p)); + return __journal_key_btree_cmp(l_btree_id, l_level, r) ?: + bpos_cmp(l_pos, r->k->k.p); } static inline int journal_key_cmp(const struct journal_key *l, const struct journal_key *r) @@ -43,7 +51,9 @@ static inline int journal_key_cmp(const struct journal_key *l, const struct jour return __journal_key_cmp(l->btree_id, l->level, l->k->k.p, r); } -struct bkey_i *bch2_journal_keys_peek_upto(struct bch_fs *, enum btree_id, +struct bkey_i *bch2_journal_keys_peek_max(struct bch_fs *, enum btree_id, + unsigned, struct bpos, struct bpos, size_t *); +struct bkey_i *bch2_journal_keys_peek_prev_min(struct bch_fs *, enum btree_id, unsigned, struct bpos, struct bpos, size_t *); struct bkey_i *bch2_journal_keys_peek_slot(struct bch_fs *, enum btree_id, unsigned, struct bpos); @@ -79,8 +89,6 @@ static inline void bch2_journal_keys_put_initial(struct bch_fs *c) c->journal_keys.initial_ref_held = false; } -void bch2_journal_entries_free(struct bch_fs *); - int bch2_journal_keys_sort(struct bch_fs *); void bch2_shoot_down_journal_keys(struct bch_fs *, enum btree_id, @@ -89,4 +97,6 @@ void bch2_shoot_down_journal_keys(struct bch_fs *, enum btree_id, void bch2_journal_keys_dump(struct bch_fs *); +void bch2_fs_journal_keys_init(struct bch_fs *); + #endif /* _BCACHEFS_BTREE_JOURNAL_ITER_H */ diff --git a/fs/bcachefs/btree_journal_iter_types.h b/fs/bcachefs/btree_journal_iter_types.h new file mode 100644 index 000000000000..8b773823704f --- /dev/null +++ b/fs/bcachefs/btree_journal_iter_types.h @@ -0,0 +1,36 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef _BCACHEFS_BTREE_JOURNAL_ITER_TYPES_H +#define _BCACHEFS_BTREE_JOURNAL_ITER_TYPES_H + +struct journal_key_range_overwritten { + size_t start, end; +}; + +struct journal_key { + u64 journal_seq; + u32 journal_offset; + enum btree_id btree_id:8; + unsigned level:8; + bool allocated; + bool overwritten; + struct journal_key_range_overwritten __rcu * + overwritten_range; + struct bkey_i *k; +}; + +struct journal_keys { + /* must match layout in darray_types.h */ + size_t nr, size; + struct journal_key *data; + /* + * Gap buffer: instead of all the empty space in the array being at the + * end of the buffer - from @nr to @size - the empty space is at @gap. + * This means that sequential insertions are O(n) instead of O(n^2). + */ + size_t gap; + atomic_t ref; + bool initial_ref_held; + struct mutex overwrite_lock; +}; + +#endif /* _BCACHEFS_BTREE_JOURNAL_ITER_TYPES_H */ diff --git a/fs/bcachefs/btree_key_cache.c b/fs/bcachefs/btree_key_cache.c index 244610b1d0b5..3b62296c3100 100644 --- a/fs/bcachefs/btree_key_cache.c +++ b/fs/bcachefs/btree_key_cache.c @@ -197,7 +197,9 @@ out: return ck; } -static int btree_key_cache_create(struct btree_trans *trans, struct btree_path *path, +static int btree_key_cache_create(struct btree_trans *trans, + struct btree_path *path, + struct btree_path *ck_path, struct bkey_s_c k) { struct bch_fs *c = trans->c; @@ -217,7 +219,7 @@ static int btree_key_cache_create(struct btree_trans *trans, struct btree_path * key_u64s = min(256U, (key_u64s * 3) / 2); key_u64s = roundup_pow_of_two(key_u64s); - struct bkey_cached *ck = bkey_cached_alloc(trans, path, key_u64s); + struct bkey_cached *ck = bkey_cached_alloc(trans, ck_path, key_u64s); int ret = PTR_ERR_OR_ZERO(ck); if (ret) return ret; @@ -226,19 +228,19 @@ static int btree_key_cache_create(struct btree_trans *trans, struct btree_path * ck = bkey_cached_reuse(bc); if (unlikely(!ck)) { bch_err(c, "error allocating memory for key cache item, btree %s", - bch2_btree_id_str(path->btree_id)); + bch2_btree_id_str(ck_path->btree_id)); return -BCH_ERR_ENOMEM_btree_key_cache_create; } } ck->c.level = 0; - ck->c.btree_id = path->btree_id; - ck->key.btree_id = path->btree_id; - ck->key.pos = path->pos; + ck->c.btree_id = ck_path->btree_id; + ck->key.btree_id = ck_path->btree_id; + ck->key.pos = ck_path->pos; ck->flags = 1U << BKEY_CACHED_ACCESSED; if (unlikely(key_u64s > ck->u64s)) { - mark_btree_node_locked_noreset(path, 0, BTREE_NODE_UNLOCKED); + mark_btree_node_locked_noreset(ck_path, 0, BTREE_NODE_UNLOCKED); struct bkey_i *new_k = allocate_dropping_locks(trans, ret, kmalloc(key_u64s * sizeof(u64), _gfp)); @@ -258,22 +260,29 @@ static int btree_key_cache_create(struct btree_trans *trans, struct btree_path * bkey_reassemble(ck->k, k); + ret = bch2_btree_node_lock_write(trans, path, &path_l(path)->b->c); + if (unlikely(ret)) + goto err; + ret = rhashtable_lookup_insert_fast(&bc->table, &ck->hash, bch2_btree_key_cache_params); + + bch2_btree_node_unlock_write(trans, path, path_l(path)->b); + if (unlikely(ret)) /* raced with another fill? */ goto err; atomic_long_inc(&bc->nr_keys); six_unlock_write(&ck->c.lock); - enum six_lock_type lock_want = __btree_lock_want(path, 0); + enum six_lock_type lock_want = __btree_lock_want(ck_path, 0); if (lock_want == SIX_LOCK_read) six_lock_downgrade(&ck->c.lock); - btree_path_cached_set(trans, path, ck, (enum btree_node_locked_type) lock_want); - path->uptodate = BTREE_ITER_UPTODATE; + btree_path_cached_set(trans, ck_path, ck, (enum btree_node_locked_type) lock_want); + ck_path->uptodate = BTREE_ITER_UPTODATE; return 0; err: bkey_cached_free(bc, ck); - mark_btree_node_locked_noreset(path, 0, BTREE_NODE_UNLOCKED); + mark_btree_node_locked_noreset(ck_path, 0, BTREE_NODE_UNLOCKED); return ret; } @@ -282,10 +291,8 @@ static noinline int btree_key_cache_fill(struct btree_trans *trans, struct btree_path *ck_path, unsigned flags) { - if (flags & BTREE_ITER_cached_nofill) { - ck_path->uptodate = BTREE_ITER_UPTODATE; + if (flags & BTREE_ITER_cached_nofill) return 0; - } struct bch_fs *c = trans->c; struct btree_iter iter; @@ -293,6 +300,7 @@ static noinline int btree_key_cache_fill(struct btree_trans *trans, int ret; bch2_trans_iter_init(trans, &iter, ck_path->btree_id, ck_path->pos, + BTREE_ITER_intent| BTREE_ITER_key_cache_fill| BTREE_ITER_cached_nofill); iter.flags &= ~BTREE_ITER_with_journal; @@ -306,9 +314,19 @@ static noinline int btree_key_cache_fill(struct btree_trans *trans, if (unlikely(ret)) goto out; - ret = btree_key_cache_create(trans, ck_path, k); + ret = btree_key_cache_create(trans, btree_iter_path(trans, &iter), ck_path, k); if (ret) goto err; + + if (trace_key_cache_fill_enabled()) { + struct printbuf buf = PRINTBUF; + + bch2_bpos_to_text(&buf, ck_path->pos); + prt_char(&buf, ' '); + bch2_bkey_val_to_text(&buf, trans->c, k); + trace_key_cache_fill(trans, buf.buf); + printbuf_exit(&buf); + } out: /* We're not likely to need this iterator again: */ bch2_set_btree_iter_dontneed(&iter); @@ -424,8 +442,15 @@ static int btree_key_cache_flush_pos(struct btree_trans *trans, !test_bit(JOURNAL_space_low, &c->journal.flags)) commit_flags |= BCH_TRANS_COMMIT_no_journal_res; - ret = bch2_btree_iter_traverse(&b_iter) ?: - bch2_trans_update(trans, &b_iter, ck->k, + struct bkey_s_c btree_k = bch2_btree_iter_peek_slot(&b_iter); + ret = bkey_err(btree_k); + if (ret) + goto err; + + /* * Check that we're not violating cache coherency rules: */ + BUG_ON(bkey_deleted(btree_k.k)); + + ret = bch2_trans_update(trans, &b_iter, ck->k, BTREE_UPDATE_key_cache_reclaim| BTREE_UPDATE_internal_snapshot_node| BTREE_TRIGGER_norun) ?: @@ -433,7 +458,7 @@ static int btree_key_cache_flush_pos(struct btree_trans *trans, BCH_TRANS_COMMIT_no_check_rw| BCH_TRANS_COMMIT_no_enospc| commit_flags); - +err: bch2_fs_fatal_err_on(ret && !bch2_err_matches(ret, BCH_ERR_transaction_restart) && !bch2_err_matches(ret, BCH_ERR_journal_reclaim_would_deadlock) && @@ -586,8 +611,18 @@ void bch2_btree_key_cache_drop(struct btree_trans *trans, bkey_cached_free(bc, ck); mark_btree_node_locked(trans, path, 0, BTREE_NODE_UNLOCKED); - btree_path_set_dirty(path, BTREE_ITER_NEED_TRAVERSE); - path->should_be_locked = false; + + struct btree_path *path2; + unsigned i; + trans_for_each_path(trans, path2, i) + if (path2->l[0].b == (void *) ck) { + __bch2_btree_path_unlock(trans, path2); + path2->l[0].b = ERR_PTR(-BCH_ERR_no_btree_node_drop); + path2->should_be_locked = false; + btree_path_set_dirty(path2, BTREE_ITER_NEED_TRAVERSE); + } + + bch2_trans_verify_locks(trans); } static unsigned long bch2_btree_key_cache_scan(struct shrinker *shrink, diff --git a/fs/bcachefs/btree_locking.c b/fs/bcachefs/btree_locking.c index efe2a007b482..10b805a60f52 100644 --- a/fs/bcachefs/btree_locking.c +++ b/fs/bcachefs/btree_locking.c @@ -109,6 +109,12 @@ static noinline void lock_graph_pop_all(struct lock_graph *g) lock_graph_up(g); } +static noinline void lock_graph_pop_from(struct lock_graph *g, struct trans_waiting_for_lock *i) +{ + while (g->g + g->nr > i) + lock_graph_up(g); +} + static void __lock_graph_down(struct lock_graph *g, struct btree_trans *trans) { g->g[g->nr++] = (struct trans_waiting_for_lock) { @@ -124,15 +130,20 @@ static void lock_graph_down(struct lock_graph *g, struct btree_trans *trans) __lock_graph_down(g, trans); } -static bool lock_graph_remove_non_waiters(struct lock_graph *g) +static bool lock_graph_remove_non_waiters(struct lock_graph *g, + struct trans_waiting_for_lock *from) { struct trans_waiting_for_lock *i; - for (i = g->g + 1; i < g->g + g->nr; i++) + if (from->trans->locking != from->node_want) { + lock_graph_pop_from(g, from); + return true; + } + + for (i = from + 1; i < g->g + g->nr; i++) if (i->trans->locking != i->node_want || i->trans->locking_wait.start_time != i[-1].lock_start_time) { - while (g->g + g->nr > i) - lock_graph_up(g); + lock_graph_pop_from(g, i); return true; } @@ -179,13 +190,14 @@ static int btree_trans_abort_preference(struct btree_trans *trans) return 3; } -static noinline int break_cycle(struct lock_graph *g, struct printbuf *cycle) +static noinline int break_cycle(struct lock_graph *g, struct printbuf *cycle, + struct trans_waiting_for_lock *from) { struct trans_waiting_for_lock *i, *abort = NULL; unsigned best = 0, pref; int ret; - if (lock_graph_remove_non_waiters(g)) + if (lock_graph_remove_non_waiters(g, from)) return 0; /* Only checking, for debugfs: */ @@ -195,7 +207,7 @@ static noinline int break_cycle(struct lock_graph *g, struct printbuf *cycle) goto out; } - for (i = g->g; i < g->g + g->nr; i++) { + for (i = from; i < g->g + g->nr; i++) { pref = btree_trans_abort_preference(i->trans); if (pref > best) { abort = i; @@ -229,8 +241,9 @@ static noinline int break_cycle(struct lock_graph *g, struct printbuf *cycle) ret = abort_lock(g, abort); out: if (ret) - while (g->nr) - lock_graph_up(g); + lock_graph_pop_all(g); + else + lock_graph_pop_from(g, abort); return ret; } @@ -243,7 +256,7 @@ static int lock_graph_descend(struct lock_graph *g, struct btree_trans *trans, for (i = g->g; i < g->g + g->nr; i++) if (i->trans == trans) { closure_put(&trans->ref); - return break_cycle(g, cycle); + return break_cycle(g, cycle, i); } if (g->nr == ARRAY_SIZE(g->g)) { @@ -252,8 +265,7 @@ static int lock_graph_descend(struct lock_graph *g, struct btree_trans *trans, if (orig_trans->lock_may_not_fail) return 0; - while (g->nr) - lock_graph_up(g); + lock_graph_pop_all(g); if (cycle) return 0; @@ -281,7 +293,7 @@ int bch2_check_for_deadlock(struct btree_trans *trans, struct printbuf *cycle) g.nr = 0; - if (trans->lock_must_abort) { + if (trans->lock_must_abort && !trans->lock_may_not_fail) { if (cycle) return -1; @@ -336,7 +348,7 @@ next: * structures - which means it can't be blocked * waiting on a lock: */ - if (!lock_graph_remove_non_waiters(&g)) { + if (!lock_graph_remove_non_waiters(&g, g.g)) { /* * If lock_graph_remove_non_waiters() * didn't do anything, it must be @@ -512,7 +524,6 @@ bool bch2_btree_node_upgrade(struct btree_trans *trans, struct btree_path *path, unsigned level) { struct btree *b = path->l[level].b; - struct six_lock_count count = bch2_btree_node_lock_counts(trans, path, &b->c, level); if (!is_btree_node(path, level)) return false; @@ -536,24 +547,11 @@ bool bch2_btree_node_upgrade(struct btree_trans *trans, if (race_fault()) return false; - if (btree_node_locked(path, level)) { - bool ret; - - six_lock_readers_add(&b->c.lock, -count.n[SIX_LOCK_read]); - ret = six_lock_tryupgrade(&b->c.lock); - six_lock_readers_add(&b->c.lock, count.n[SIX_LOCK_read]); - - if (ret) - goto success; - } else { - if (six_relock_type(&b->c.lock, SIX_LOCK_intent, path->l[level].lock_seq)) - goto success; - } + if (btree_node_locked(path, level) + ? six_lock_tryupgrade(&b->c.lock) + : six_relock_type(&b->c.lock, SIX_LOCK_intent, path->l[level].lock_seq)) + goto success; - /* - * Do we already have an intent lock via another path? If so, just bump - * lock count: - */ if (btree_node_lock_seq_matches(path, b, level) && btree_node_lock_increment(trans, &b->c, level, BTREE_NODE_INTENT_LOCKED)) { btree_node_unlock(trans, path, level); @@ -782,7 +780,7 @@ static inline int __bch2_trans_relock(struct btree_trans *trans, bool trace) return bch2_trans_relock_fail(trans, path, &f, trace); } - trans_set_locked(trans); + trans_set_locked(trans, true); out: bch2_trans_verify_locks(trans); return 0; @@ -818,6 +816,17 @@ void bch2_trans_unlock_long(struct btree_trans *trans) bch2_trans_srcu_unlock(trans); } +void bch2_trans_unlock_write(struct btree_trans *trans) +{ + struct btree_path *path; + unsigned i; + + trans_for_each_path(trans, path, i) + for (unsigned l = 0; l < BTREE_MAX_DEPTH; l++) + if (btree_node_write_locked(path, l)) + bch2_btree_node_unlock_write(trans, path, path->l[l].b); +} + int __bch2_trans_mutex_lock(struct btree_trans *trans, struct mutex *lock) { @@ -856,6 +865,9 @@ void bch2_btree_path_verify_locks(struct btree_path *path) (want == BTREE_NODE_UNLOCKED || have != BTREE_NODE_WRITE_LOCKED) && want != have); + + BUG_ON(btree_node_locked(path, l) && + path->l[l].lock_seq != six_lock_seq(&path->l[l].b->c.lock)); } } diff --git a/fs/bcachefs/btree_locking.h b/fs/bcachefs/btree_locking.h index 7c07f9fa9add..b54ef48eb8cc 100644 --- a/fs/bcachefs/btree_locking.h +++ b/fs/bcachefs/btree_locking.h @@ -16,6 +16,7 @@ void bch2_btree_lock_init(struct btree_bkey_cached_common *, enum six_lock_init_flags); void bch2_trans_unlock_noassert(struct btree_trans *); +void bch2_trans_unlock_write(struct btree_trans *); static inline bool is_btree_node(struct btree_path *path, unsigned l) { @@ -75,13 +76,6 @@ static inline void mark_btree_node_locked_noreset(struct btree_path *path, path->nodes_locked |= (type + 1) << (level << 1); } -static inline void mark_btree_node_unlocked(struct btree_path *path, - unsigned level) -{ - EBUG_ON(btree_node_write_locked(path, level)); - mark_btree_node_locked_noreset(path, level, BTREE_NODE_UNLOCKED); -} - static inline void mark_btree_node_locked(struct btree_trans *trans, struct btree_path *path, unsigned level, @@ -124,19 +118,25 @@ static void btree_trans_lock_hold_time_update(struct btree_trans *trans, /* unlock: */ +void bch2_btree_node_unlock_write(struct btree_trans *, + struct btree_path *, struct btree *); + static inline void btree_node_unlock(struct btree_trans *trans, struct btree_path *path, unsigned level) { int lock_type = btree_node_locked_type(path, level); EBUG_ON(level >= BTREE_MAX_DEPTH); - EBUG_ON(lock_type == BTREE_NODE_WRITE_LOCKED); if (lock_type != BTREE_NODE_UNLOCKED) { + if (unlikely(lock_type == BTREE_NODE_WRITE_LOCKED)) { + bch2_btree_node_unlock_write(trans, path, path->l[level].b); + lock_type = BTREE_NODE_INTENT_LOCKED; + } six_unlock_type(&path->l[level].b->c.lock, lock_type); btree_trans_lock_hold_time_update(trans, path, level); + mark_btree_node_locked_noreset(path, level, BTREE_NODE_UNLOCKED); } - mark_btree_node_unlocked(path, level); } static inline int btree_path_lowest_level_locked(struct btree_path *path) @@ -163,35 +163,39 @@ static inline void __bch2_btree_path_unlock(struct btree_trans *trans, * succeed: */ static inline void +__bch2_btree_node_unlock_write(struct btree_trans *trans, struct btree *b) +{ + if (!b->c.lock.write_lock_recurse) { + struct btree_path *linked; + unsigned i; + + trans_for_each_path_with_node(trans, b, linked, i) + linked->l[b->c.level].lock_seq++; + } + + six_unlock_write(&b->c.lock); +} + +static inline void bch2_btree_node_unlock_write_inlined(struct btree_trans *trans, struct btree_path *path, struct btree *b) { - struct btree_path *linked; - unsigned i; - EBUG_ON(path->l[b->c.level].b != b); EBUG_ON(path->l[b->c.level].lock_seq != six_lock_seq(&b->c.lock)); EBUG_ON(btree_node_locked_type(path, b->c.level) != SIX_LOCK_write); mark_btree_node_locked_noreset(path, b->c.level, BTREE_NODE_INTENT_LOCKED); - - trans_for_each_path_with_node(trans, b, linked, i) - linked->l[b->c.level].lock_seq++; - - six_unlock_write(&b->c.lock); + __bch2_btree_node_unlock_write(trans, b); } -void bch2_btree_node_unlock_write(struct btree_trans *, - struct btree_path *, struct btree *); - int bch2_six_check_for_deadlock(struct six_lock *lock, void *p); /* lock: */ -static inline void trans_set_locked(struct btree_trans *trans) +static inline void trans_set_locked(struct btree_trans *trans, bool try) { if (!trans->locked) { - lock_acquire_exclusive(&trans->dep_map, 0, 0, NULL, _THIS_IP_); + lock_acquire_exclusive(&trans->dep_map, 0, try, NULL, _THIS_IP_); trans->locked = true; trans->last_unlock_ip = 0; @@ -282,7 +286,7 @@ static inline int btree_node_lock(struct btree_trans *trans, int ret = 0; EBUG_ON(level >= BTREE_MAX_DEPTH); - bch2_trans_verify_not_unlocked(trans); + bch2_trans_verify_not_unlocked_or_in_restart(trans); if (likely(six_trylock_type(&b->lock, type)) || btree_node_lock_increment(trans, b, level, (enum btree_node_locked_type) type) || diff --git a/fs/bcachefs/btree_node_scan.c b/fs/bcachefs/btree_node_scan.c index 30131c3bdd97..a7f06deee13c 100644 --- a/fs/bcachefs/btree_node_scan.c +++ b/fs/bcachefs/btree_node_scan.c @@ -12,6 +12,7 @@ #include "recovery_passes.h" #include <linux/kthread.h> +#include <linux/min_heap.h> #include <linux/sort.h> struct find_btree_nodes_worker { @@ -22,17 +23,15 @@ struct find_btree_nodes_worker { static void found_btree_node_to_text(struct printbuf *out, struct bch_fs *c, const struct found_btree_node *n) { - prt_printf(out, "%s l=%u seq=%u journal_seq=%llu cookie=%llx ", - bch2_btree_id_str(n->btree_id), n->level, n->seq, - n->journal_seq, n->cookie); + bch2_btree_id_level_to_text(out, n->btree_id, n->level); + prt_printf(out, " seq=%u journal_seq=%llu cookie=%llx ", + n->seq, n->journal_seq, n->cookie); bch2_bpos_to_text(out, n->min_key); prt_str(out, "-"); bch2_bpos_to_text(out, n->max_key); if (n->range_updated) prt_str(out, " range updated"); - if (n->overwritten) - prt_str(out, " overwritten"); for (unsigned i = 0; i < n->nr_ptrs; i++) { prt_char(out, ' '); @@ -140,6 +139,24 @@ static int found_btree_node_cmp_pos(const void *_l, const void *_r) -found_btree_node_cmp_time(l, r); } +static inline bool found_btree_node_cmp_pos_less(const void *l, const void *r, void *arg) +{ + return found_btree_node_cmp_pos(l, r) < 0; +} + +static inline void found_btree_node_swap(void *_l, void *_r, void *arg) +{ + struct found_btree_node *l = _l; + struct found_btree_node *r = _r; + + swap(*l, *r); +} + +static const struct min_heap_callbacks found_btree_node_heap_cbs = { + .less = found_btree_node_cmp_pos_less, + .swp = found_btree_node_swap, +}; + static void try_read_btree_node(struct find_btree_nodes *f, struct bch_dev *ca, struct bio *bio, struct btree_node *bn, u64 offset) { @@ -159,6 +176,9 @@ static void try_read_btree_node(struct find_btree_nodes *f, struct bch_dev *ca, return; if (bch2_csum_type_is_encryption(BSET_CSUM_TYPE(&bn->keys))) { + if (!c->chacha20) + return; + struct nonce nonce = btree_nonce(&bn->keys, 0); unsigned bytes = (void *) &bn->keys - (void *) &bn->flags; @@ -292,55 +312,48 @@ err: return f->ret ?: ret; } -static void bubble_up(struct found_btree_node *n, struct found_btree_node *end) +static bool nodes_overlap(const struct found_btree_node *l, + const struct found_btree_node *r) { - while (n + 1 < end && - found_btree_node_cmp_pos(n, n + 1) > 0) { - swap(n[0], n[1]); - n++; - } + return (l->btree_id == r->btree_id && + l->level == r->level && + bpos_gt(l->max_key, r->min_key)); } static int handle_overwrites(struct bch_fs *c, - struct found_btree_node *start, - struct found_btree_node *end) + struct found_btree_node *l, + found_btree_nodes *nodes_heap) { - struct found_btree_node *n; -again: - for (n = start + 1; - n < end && - n->btree_id == start->btree_id && - n->level == start->level && - bpos_lt(n->min_key, start->max_key); - n++) { - int cmp = found_btree_node_cmp_time(start, n); + struct found_btree_node *r; + + while ((r = min_heap_peek(nodes_heap)) && + nodes_overlap(l, r)) { + int cmp = found_btree_node_cmp_time(l, r); if (cmp > 0) { - if (bpos_cmp(start->max_key, n->max_key) >= 0) - n->overwritten = true; + if (bpos_cmp(l->max_key, r->max_key) >= 0) + min_heap_pop(nodes_heap, &found_btree_node_heap_cbs, NULL); else { - n->range_updated = true; - n->min_key = bpos_successor(start->max_key); - n->range_updated = true; - bubble_up(n, end); - goto again; + r->range_updated = true; + r->min_key = bpos_successor(l->max_key); + r->range_updated = true; + min_heap_sift_down(nodes_heap, 0, &found_btree_node_heap_cbs, NULL); } } else if (cmp < 0) { - BUG_ON(bpos_cmp(n->min_key, start->min_key) <= 0); + BUG_ON(bpos_eq(l->min_key, r->min_key)); - start->max_key = bpos_predecessor(n->min_key); - start->range_updated = true; - } else if (n->level) { - n->overwritten = true; + l->max_key = bpos_predecessor(r->min_key); + l->range_updated = true; + } else if (r->level) { + min_heap_pop(nodes_heap, &found_btree_node_heap_cbs, NULL); } else { - if (bpos_cmp(start->max_key, n->max_key) >= 0) - n->overwritten = true; + if (bpos_cmp(l->max_key, r->max_key) >= 0) + min_heap_pop(nodes_heap, &found_btree_node_heap_cbs, NULL); else { - n->range_updated = true; - n->min_key = bpos_successor(start->max_key); - n->range_updated = true; - bubble_up(n, end); - goto again; + r->range_updated = true; + r->min_key = bpos_successor(l->max_key); + r->range_updated = true; + min_heap_sift_down(nodes_heap, 0, &found_btree_node_heap_cbs, NULL); } } } @@ -352,6 +365,7 @@ int bch2_scan_for_btree_nodes(struct bch_fs *c) { struct find_btree_nodes *f = &c->found_btree_nodes; struct printbuf buf = PRINTBUF; + found_btree_nodes nodes_heap = {}; size_t dst; int ret = 0; @@ -406,29 +420,57 @@ int bch2_scan_for_btree_nodes(struct bch_fs *c) bch2_print_string_as_lines(KERN_INFO, buf.buf); } - dst = 0; - darray_for_each(f->nodes, i) { - if (i->overwritten) - continue; + swap(nodes_heap, f->nodes); + + { + /* darray must have same layout as a heap */ + min_heap_char real_heap; + BUILD_BUG_ON(sizeof(nodes_heap.nr) != sizeof(real_heap.nr)); + BUILD_BUG_ON(sizeof(nodes_heap.size) != sizeof(real_heap.size)); + BUILD_BUG_ON(offsetof(found_btree_nodes, nr) != offsetof(min_heap_char, nr)); + BUILD_BUG_ON(offsetof(found_btree_nodes, size) != offsetof(min_heap_char, size)); + } + + min_heapify_all(&nodes_heap, &found_btree_node_heap_cbs, NULL); - ret = handle_overwrites(c, i, &darray_top(f->nodes)); + if (nodes_heap.nr) { + ret = darray_push(&f->nodes, *min_heap_peek(&nodes_heap)); if (ret) goto err; - BUG_ON(i->overwritten); - f->nodes.data[dst++] = *i; + min_heap_pop(&nodes_heap, &found_btree_node_heap_cbs, NULL); } - f->nodes.nr = dst; - if (c->opts.verbose) { + while (true) { + ret = handle_overwrites(c, &darray_last(f->nodes), &nodes_heap); + if (ret) + goto err; + + if (!nodes_heap.nr) + break; + + ret = darray_push(&f->nodes, *min_heap_peek(&nodes_heap)); + if (ret) + goto err; + + min_heap_pop(&nodes_heap, &found_btree_node_heap_cbs, NULL); + } + + for (struct found_btree_node *n = f->nodes.data; n < &darray_last(f->nodes); n++) + BUG_ON(nodes_overlap(n, n + 1)); + + if (0 && c->opts.verbose) { printbuf_reset(&buf); prt_printf(&buf, "%s: nodes found after overwrites:\n", __func__); found_btree_nodes_to_text(&buf, c, f->nodes); bch2_print_string_as_lines(KERN_INFO, buf.buf); + } else { + bch_info(c, "btree node scan found %zu nodes after overwrites", f->nodes.nr); } eytzinger0_sort(f->nodes.data, f->nodes.nr, sizeof(f->nodes.data[0]), found_btree_node_cmp_pos, NULL); err: + darray_exit(&nodes_heap); printbuf_exit(&buf); return ret; } @@ -499,7 +541,9 @@ int bch2_get_scanned_nodes(struct bch_fs *c, enum btree_id btree, if (c->opts.verbose) { struct printbuf buf = PRINTBUF; - prt_printf(&buf, "recovering %s l=%u ", bch2_btree_id_str(btree), level); + prt_str(&buf, "recovery "); + bch2_btree_id_level_to_text(&buf, btree, level); + prt_str(&buf, " "); bch2_bpos_to_text(&buf, node_min); prt_str(&buf, " - "); bch2_bpos_to_text(&buf, node_max); @@ -533,7 +577,12 @@ int bch2_get_scanned_nodes(struct bch_fs *c, enum btree_id btree, bch_verbose(c, "%s(): recovering %s", __func__, buf.buf); printbuf_exit(&buf); - BUG_ON(bch2_bkey_validate(c, bkey_i_to_s_c(&tmp.k), BKEY_TYPE_btree, 0)); + BUG_ON(bch2_bkey_validate(c, bkey_i_to_s_c(&tmp.k), + (struct bkey_validate_context) { + .from = BKEY_VALIDATE_btree_node, + .level = level + 1, + .btree = btree, + })); ret = bch2_journal_key_insert(c, btree, level + 1, &tmp.k); if (ret) diff --git a/fs/bcachefs/btree_node_scan_types.h b/fs/bcachefs/btree_node_scan_types.h index b6c36c45d0be..2811b6857c97 100644 --- a/fs/bcachefs/btree_node_scan_types.h +++ b/fs/bcachefs/btree_node_scan_types.h @@ -6,7 +6,6 @@ struct found_btree_node { bool range_updated:1; - bool overwritten:1; u8 btree_id; u8 level; unsigned sectors_written; diff --git a/fs/bcachefs/btree_trans_commit.c b/fs/bcachefs/btree_trans_commit.c index 9bf471fa4361..6b79b672e0b1 100644 --- a/fs/bcachefs/btree_trans_commit.c +++ b/fs/bcachefs/btree_trans_commit.c @@ -133,7 +133,7 @@ static inline int bch2_trans_lock_write(struct btree_trans *trans) return 0; } -static inline void bch2_trans_unlock_write(struct btree_trans *trans) +static inline void bch2_trans_unlock_updates_write(struct btree_trans *trans) { if (likely(trans->write_locked)) { trans_for_each_update(trans, i) @@ -249,7 +249,7 @@ static int __btree_node_flush(struct journal *j, struct journal_entry_pin *pin, new |= 1 << BTREE_NODE_need_write; } while (!try_cmpxchg(&b->flags, &old, new)); - btree_node_write_if_need(c, b, SIX_LOCK_read); + btree_node_write_if_need(trans, b, SIX_LOCK_read); six_unlock_read(&b->c.lock); bch2_trans_put(trans); @@ -384,7 +384,7 @@ btree_key_can_insert_cached_slowpath(struct btree_trans *trans, unsigned flags, struct bkey_i *new_k; int ret; - bch2_trans_unlock_write(trans); + bch2_trans_unlock_updates_write(trans); bch2_trans_unlock(trans); new_k = kmalloc(new_u64s * sizeof(u64), GFP_KERNEL); @@ -479,8 +479,7 @@ static int run_one_mem_trigger(struct btree_trans *trans, old, flags); } -static int run_one_trans_trigger(struct btree_trans *trans, struct btree_insert_entry *i, - bool overwrite) +static int run_one_trans_trigger(struct btree_trans *trans, struct btree_insert_entry *i) { verify_update_old_key(trans, i); @@ -507,10 +506,10 @@ static int run_one_trans_trigger(struct btree_trans *trans, struct btree_insert_ return bch2_key_trigger(trans, i->btree_id, i->level, old, bkey_i_to_s(i->k), BTREE_TRIGGER_insert| BTREE_TRIGGER_overwrite|flags) ?: 1; - } else if (overwrite && !i->overwrite_trigger_run) { + } else if (!i->overwrite_trigger_run) { i->overwrite_trigger_run = true; return bch2_key_trigger_old(trans, i->btree_id, i->level, old, flags) ?: 1; - } else if (!overwrite && !i->insert_trigger_run) { + } else if (!i->insert_trigger_run) { i->insert_trigger_run = true; return bch2_key_trigger_new(trans, i->btree_id, i->level, bkey_i_to_s(i->k), flags) ?: 1; } else { @@ -519,39 +518,45 @@ static int run_one_trans_trigger(struct btree_trans *trans, struct btree_insert_ } static int run_btree_triggers(struct btree_trans *trans, enum btree_id btree_id, - unsigned btree_id_start) + unsigned *btree_id_updates_start) { - for (int overwrite = 1; overwrite >= 0; --overwrite) { - bool trans_trigger_run; + bool trans_trigger_run; - /* - * Running triggers will append more updates to the list of updates as - * we're walking it: - */ - do { - trans_trigger_run = false; - - for (unsigned i = btree_id_start; - i < trans->nr_updates && trans->updates[i].btree_id <= btree_id; - i++) { - if (trans->updates[i].btree_id != btree_id) - continue; + /* + * Running triggers will append more updates to the list of updates as + * we're walking it: + */ + do { + trans_trigger_run = false; - int ret = run_one_trans_trigger(trans, trans->updates + i, overwrite); - if (ret < 0) - return ret; - if (ret) - trans_trigger_run = true; + for (unsigned i = *btree_id_updates_start; + i < trans->nr_updates && trans->updates[i].btree_id <= btree_id; + i++) { + if (trans->updates[i].btree_id < btree_id) { + *btree_id_updates_start = i; + continue; } - } while (trans_trigger_run); - } + + int ret = run_one_trans_trigger(trans, trans->updates + i); + if (ret < 0) + return ret; + if (ret) + trans_trigger_run = true; + } + } while (trans_trigger_run); + + trans_for_each_update(trans, i) + BUG_ON(!(i->flags & BTREE_TRIGGER_norun) && + i->btree_id == btree_id && + btree_node_type_has_trans_triggers(i->bkey_type) && + (!i->insert_trigger_run || !i->overwrite_trigger_run)); return 0; } static int bch2_trans_commit_run_triggers(struct btree_trans *trans) { - unsigned btree_id = 0, btree_id_start = 0; + unsigned btree_id = 0, btree_id_updates_start = 0; int ret = 0; /* @@ -565,27 +570,15 @@ static int bch2_trans_commit_run_triggers(struct btree_trans *trans) if (btree_id == BTREE_ID_alloc) continue; - while (btree_id_start < trans->nr_updates && - trans->updates[btree_id_start].btree_id < btree_id) - btree_id_start++; - - ret = run_btree_triggers(trans, btree_id, btree_id_start); + ret = run_btree_triggers(trans, btree_id, &btree_id_updates_start); if (ret) return ret; } - for (unsigned idx = 0; idx < trans->nr_updates; idx++) { - struct btree_insert_entry *i = trans->updates + idx; - - if (i->btree_id > BTREE_ID_alloc) - break; - if (i->btree_id == BTREE_ID_alloc) { - ret = run_btree_triggers(trans, BTREE_ID_alloc, idx); - if (ret) - return ret; - break; - } - } + btree_id_updates_start = 0; + ret = run_btree_triggers(trans, BTREE_ID_alloc, &btree_id_updates_start); + if (ret) + return ret; #ifdef CONFIG_BCACHEFS_DEBUG trans_for_each_update(trans, i) @@ -609,14 +602,6 @@ static noinline int bch2_trans_commit_run_gc_triggers(struct btree_trans *trans) return 0; } -static struct bversion journal_pos_to_bversion(struct journal_res *res, unsigned offset) -{ - return (struct bversion) { - .hi = res->seq >> 32, - .lo = (res->seq << 32) | (res->offset + offset), - }; -} - static inline int bch2_trans_commit_write_locked(struct btree_trans *trans, unsigned flags, struct btree_insert_entry **stopped_at, @@ -627,12 +612,11 @@ bch2_trans_commit_write_locked(struct btree_trans *trans, unsigned flags, unsigned u64s = 0; int ret = 0; - bch2_trans_verify_not_unlocked(trans); - bch2_trans_verify_not_in_restart(trans); + bch2_trans_verify_not_unlocked_or_in_restart(trans); if (race_fault()) { trace_and_count(c, trans_restart_fault_inject, trans, trace_ip); - return btree_trans_restart_nounlock(trans, BCH_ERR_transaction_restart_fault_inject); + return btree_trans_restart(trans, BCH_ERR_transaction_restart_fault_inject); } /* @@ -701,25 +685,14 @@ bch2_trans_commit_write_locked(struct btree_trans *trans, unsigned flags, struct jset_entry *entry = trans->journal_entries; percpu_down_read(&c->mark_lock); - for (entry = trans->journal_entries; entry != (void *) ((u64 *) trans->journal_entries + trans->journal_entries_u64s); entry = vstruct_next(entry)) if (entry->type == BCH_JSET_ENTRY_write_buffer_keys && entry->start->k.type == KEY_TYPE_accounting) { - BUG_ON(!trans->journal_res.ref); - - struct bkey_i_accounting *a = bkey_i_to_accounting(entry->start); - - a->k.bversion = journal_pos_to_bversion(&trans->journal_res, - (u64 *) entry - (u64 *) trans->journal_entries); - BUG_ON(bversion_zero(a->k.bversion)); - - if (likely(!(flags & BCH_TRANS_COMMIT_skip_accounting_apply))) { - ret = bch2_accounting_mem_mod_locked(trans, accounting_i_to_s_c(a), BCH_ACCOUNTING_normal); - if (ret) - goto revert_fs_usage; - } + ret = bch2_accounting_trans_commit_hook(trans, bkey_i_to_accounting(entry->start), flags); + if (ret) + goto revert_fs_usage; } percpu_up_read(&c->mark_lock); @@ -739,33 +712,17 @@ bch2_trans_commit_write_locked(struct btree_trans *trans, unsigned flags, goto fatal_err; } - trans_for_each_update(trans, i) { - enum bch_validate_flags invalid_flags = 0; + struct bkey_validate_context validate_context = { .from = BKEY_VALIDATE_commit }; - if (!(flags & BCH_TRANS_COMMIT_no_journal_res)) - invalid_flags |= BCH_VALIDATE_write|BCH_VALIDATE_commit; - - ret = bch2_bkey_validate(c, bkey_i_to_s_c(i->k), - i->bkey_type, invalid_flags); - if (unlikely(ret)){ - bch2_trans_inconsistent(trans, "invalid bkey on insert from %s -> %ps\n", - trans->fn, (void *) i->ip_allocated); - goto fatal_err; - } - btree_insert_entry_checks(trans, i); - } + if (!(flags & BCH_TRANS_COMMIT_no_journal_res)) + validate_context.flags = BCH_VALIDATE_write|BCH_VALIDATE_commit; for (struct jset_entry *i = trans->journal_entries; i != (void *) ((u64 *) trans->journal_entries + trans->journal_entries_u64s); i = vstruct_next(i)) { - enum bch_validate_flags invalid_flags = 0; - - if (!(flags & BCH_TRANS_COMMIT_no_journal_res)) - invalid_flags |= BCH_VALIDATE_write|BCH_VALIDATE_commit; - ret = bch2_journal_entry_validate(c, NULL, i, bcachefs_metadata_version_current, - CPU_BIG_ENDIAN, invalid_flags); + CPU_BIG_ENDIAN, validate_context); if (unlikely(ret)) { bch2_trans_inconsistent(trans, "invalid journal entry on insert from %s\n", trans->fn); @@ -773,6 +730,19 @@ bch2_trans_commit_write_locked(struct btree_trans *trans, unsigned flags, } } + trans_for_each_update(trans, i) { + validate_context.level = i->level; + validate_context.btree = i->btree_id; + + ret = bch2_bkey_validate(c, bkey_i_to_s_c(i->k), validate_context); + if (unlikely(ret)){ + bch2_trans_inconsistent(trans, "invalid bkey on insert from %s -> %ps\n", + trans->fn, (void *) i->ip_allocated); + goto fatal_err; + } + btree_insert_entry_checks(trans, i); + } + if (likely(!(flags & BCH_TRANS_COMMIT_no_journal_res))) { struct journal *j = &c->journal; struct jset_entry *entry; @@ -833,13 +803,9 @@ revert_fs_usage: entry2 != entry; entry2 = vstruct_next(entry2)) if (entry2->type == BCH_JSET_ENTRY_write_buffer_keys && - entry2->start->k.type == KEY_TYPE_accounting) { - struct bkey_s_accounting a = bkey_i_to_s_accounting(entry2->start); - - bch2_accounting_neg(a); - bch2_accounting_mem_mod_locked(trans, a.c, BCH_ACCOUNTING_normal); - bch2_accounting_neg(a); - } + entry2->start->k.type == KEY_TYPE_accounting) + bch2_accounting_trans_commit_revert(trans, + bkey_i_to_accounting(entry2->start), flags); percpu_up_read(&c->mark_lock); return ret; } @@ -902,7 +868,7 @@ static inline int do_bch2_trans_commit(struct btree_trans *trans, unsigned flags if (!ret && unlikely(trans->journal_replay_not_finished)) bch2_drop_overwrites_from_journal(trans); - bch2_trans_unlock_write(trans); + bch2_trans_unlock_updates_write(trans); if (!ret && trans->journal_pin) bch2_journal_pin_add(&c->journal, trans->journal_res.seq, @@ -994,24 +960,6 @@ int bch2_trans_commit_error(struct btree_trans *trans, unsigned flags, return ret; } -static noinline int -bch2_trans_commit_get_rw_cold(struct btree_trans *trans, unsigned flags) -{ - struct bch_fs *c = trans->c; - int ret; - - if (likely(!(flags & BCH_TRANS_COMMIT_lazy_rw)) || - test_bit(BCH_FS_started, &c->flags)) - return -BCH_ERR_erofs_trans_commit; - - ret = drop_locks_do(trans, bch2_fs_read_write_early(c)); - if (ret) - return ret; - - bch2_write_ref_get(c, BCH_WRITE_REF_trans); - return 0; -} - /* * This is for updates done in the early part of fsck - btree_gc - before we've * gone RW. we only add the new key to the list of keys for journal replay to @@ -1022,6 +970,8 @@ do_bch2_trans_commit_to_journal_replay(struct btree_trans *trans) { struct bch_fs *c = trans->c; + BUG_ON(current != c->recovery_task); + trans_for_each_update(trans, i) { int ret = bch2_journal_key_insert(c, i->btree_id, i->level, i->k); if (ret) @@ -1047,8 +997,7 @@ int __bch2_trans_commit(struct btree_trans *trans, unsigned flags) struct bch_fs *c = trans->c; int ret = 0; - bch2_trans_verify_not_unlocked(trans); - bch2_trans_verify_not_in_restart(trans); + bch2_trans_verify_not_unlocked_or_in_restart(trans); if (!trans->nr_updates && !trans->journal_entries_u64s) @@ -1058,16 +1007,13 @@ int __bch2_trans_commit(struct btree_trans *trans, unsigned flags) if (ret) goto out_reset; - if (unlikely(!test_bit(BCH_FS_may_go_rw, &c->flags))) { - ret = do_bch2_trans_commit_to_journal_replay(trans); - goto out_reset; - } - if (!(flags & BCH_TRANS_COMMIT_no_check_rw) && unlikely(!bch2_write_ref_tryget(c, BCH_WRITE_REF_trans))) { - ret = bch2_trans_commit_get_rw_cold(trans, flags); - if (ret) - goto out_reset; + if (unlikely(!test_bit(BCH_FS_may_go_rw, &c->flags))) + ret = do_bch2_trans_commit_to_journal_replay(trans); + else + ret = -BCH_ERR_erofs_trans_commit; + goto out_reset; } EBUG_ON(test_bit(BCH_FS_clean_shutdown, &c->flags)); @@ -1112,8 +1058,7 @@ int __bch2_trans_commit(struct btree_trans *trans, unsigned flags) } retry: errored_at = NULL; - bch2_trans_verify_not_unlocked(trans); - bch2_trans_verify_not_in_restart(trans); + bch2_trans_verify_not_unlocked_or_in_restart(trans); if (likely(!(flags & BCH_TRANS_COMMIT_no_journal_res))) memset(&trans->journal_res, 0, sizeof(trans->journal_res)); memset(&trans->fs_usage_delta, 0, sizeof(trans->fs_usage_delta)); diff --git a/fs/bcachefs/btree_types.h b/fs/bcachefs/btree_types.h index 4568a41fefaf..a6f251eb4164 100644 --- a/fs/bcachefs/btree_types.h +++ b/fs/bcachefs/btree_types.h @@ -513,6 +513,9 @@ struct btree_trans { u64 last_begin_time; unsigned long last_begin_ip; unsigned long last_restarted_ip; +#ifdef CONFIG_BCACHEFS_DEBUG + bch_stacktrace last_restarted_trace; +#endif unsigned long last_unlock_ip; unsigned long srcu_lock_time; @@ -787,53 +790,64 @@ static inline bool btree_node_type_has_triggers(enum btree_node_type type) return BIT_ULL(type) & BTREE_NODE_TYPE_HAS_TRIGGERS; } -static inline bool btree_node_type_is_extents(enum btree_node_type type) +static inline bool btree_id_is_extents(enum btree_id btree) { const u64 mask = 0 -#define x(name, nr, flags, ...) |((!!((flags) & BTREE_ID_EXTENTS)) << (nr + 1)) +#define x(name, nr, flags, ...) |((!!((flags) & BTREE_IS_extents)) << nr) BCH_BTREE_IDS() #undef x ; - return BIT_ULL(type) & mask; + return BIT_ULL(btree) & mask; } -static inline bool btree_id_is_extents(enum btree_id btree) +static inline bool btree_node_type_is_extents(enum btree_node_type type) { - return btree_node_type_is_extents(__btree_node_type(0, btree)); + return type != BKEY_TYPE_btree && btree_id_is_extents(type - 1); +} + +static inline bool btree_type_has_snapshots(enum btree_id btree) +{ + const u64 mask = 0 +#define x(name, nr, flags, ...) |((!!((flags) & BTREE_IS_snapshots)) << nr) + BCH_BTREE_IDS() +#undef x + ; + + return BIT_ULL(btree) & mask; } -static inline bool btree_type_has_snapshots(enum btree_id id) +static inline bool btree_type_has_snapshot_field(enum btree_id btree) { const u64 mask = 0 -#define x(name, nr, flags, ...) |((!!((flags) & BTREE_ID_SNAPSHOTS)) << nr) +#define x(name, nr, flags, ...) |((!!((flags) & (BTREE_IS_snapshot_field|BTREE_IS_snapshots))) << nr) BCH_BTREE_IDS() #undef x ; - return BIT_ULL(id) & mask; + return BIT_ULL(btree) & mask; } -static inline bool btree_type_has_snapshot_field(enum btree_id id) +static inline bool btree_type_has_ptrs(enum btree_id btree) { const u64 mask = 0 -#define x(name, nr, flags, ...) |((!!((flags) & (BTREE_ID_SNAPSHOT_FIELD|BTREE_ID_SNAPSHOTS))) << nr) +#define x(name, nr, flags, ...) |((!!((flags) & BTREE_IS_data)) << nr) BCH_BTREE_IDS() #undef x ; - return BIT_ULL(id) & mask; + return BIT_ULL(btree) & mask; } -static inline bool btree_type_has_ptrs(enum btree_id id) +static inline bool btree_type_uses_write_buffer(enum btree_id btree) { const u64 mask = 0 -#define x(name, nr, flags, ...) |((!!((flags) & BTREE_ID_DATA)) << nr) +#define x(name, nr, flags, ...) |((!!((flags) & BTREE_IS_write_buffer)) << nr) BCH_BTREE_IDS() #undef x ; - return BIT_ULL(id) & mask; + return BIT_ULL(btree) & mask; } struct btree_root { diff --git a/fs/bcachefs/btree_update.c b/fs/bcachefs/btree_update.c index 5d809e8bd170..13d794f201a5 100644 --- a/fs/bcachefs/btree_update.c +++ b/fs/bcachefs/btree_update.c @@ -144,7 +144,7 @@ int __bch2_insert_snapshot_whiteouts(struct btree_trans *trans, !(ret = bkey_err(old_k)) && bkey_eq(old_pos, old_k.k->p)) { struct bpos whiteout_pos = - SPOS(new_pos.inode, new_pos.offset, old_k.k->p.snapshot);; + SPOS(new_pos.inode, new_pos.offset, old_k.k->p.snapshot); if (!bch2_snapshot_is_ancestor(c, old_k.k->p.snapshot, old_pos.snapshot) || snapshot_list_has_ancestor(c, &s, old_k.k->p.snapshot)) @@ -296,7 +296,7 @@ static int bch2_trans_update_extent(struct btree_trans *trans, BTREE_ITER_intent| BTREE_ITER_with_updates| BTREE_ITER_not_extents); - k = bch2_btree_iter_peek_upto(&iter, POS(insert->k.p.inode, U64_MAX)); + k = bch2_btree_iter_peek_max(&iter, POS(insert->k.p.inode, U64_MAX)); if ((ret = bkey_err(k))) goto err; if (!k.k) @@ -323,7 +323,7 @@ static int bch2_trans_update_extent(struct btree_trans *trans, goto out; next: bch2_btree_iter_advance(&iter); - k = bch2_btree_iter_peek_upto(&iter, POS(insert->k.p.inode, U64_MAX)); + k = bch2_btree_iter_peek_max(&iter, POS(insert->k.p.inode, U64_MAX)); if ((ret = bkey_err(k))) goto err; if (!k.k) @@ -588,12 +588,9 @@ struct jset_entry *__bch2_trans_jset_entry_alloc(struct btree_trans *trans, unsi int bch2_bkey_get_empty_slot(struct btree_trans *trans, struct btree_iter *iter, enum btree_id btree, struct bpos end) { - struct bkey_s_c k; - int ret = 0; - - bch2_trans_iter_init(trans, iter, btree, POS_MAX, BTREE_ITER_intent); - k = bch2_btree_iter_prev(iter); - ret = bkey_err(k); + bch2_trans_iter_init(trans, iter, btree, end, BTREE_ITER_intent); + struct bkey_s_c k = bch2_btree_iter_peek_prev(iter); + int ret = bkey_err(k); if (ret) goto err; @@ -672,27 +669,19 @@ int bch2_btree_insert(struct bch_fs *c, enum btree_id id, struct bkey_i *k, bch2_btree_insert_trans(trans, id, k, iter_flags)); } -int bch2_btree_delete_extent_at(struct btree_trans *trans, struct btree_iter *iter, - unsigned len, unsigned update_flags) +int bch2_btree_delete_at(struct btree_trans *trans, + struct btree_iter *iter, unsigned update_flags) { - struct bkey_i *k; - - k = bch2_trans_kmalloc(trans, sizeof(*k)); - if (IS_ERR(k)) - return PTR_ERR(k); + struct bkey_i *k = bch2_trans_kmalloc(trans, sizeof(*k)); + int ret = PTR_ERR_OR_ZERO(k); + if (ret) + return ret; bkey_init(&k->k); k->k.p = iter->pos; - bch2_key_resize(&k->k, len); return bch2_trans_update(trans, iter, k, update_flags); } -int bch2_btree_delete_at(struct btree_trans *trans, - struct btree_iter *iter, unsigned update_flags) -{ - return bch2_btree_delete_extent_at(trans, iter, 0, update_flags); -} - int bch2_btree_delete(struct btree_trans *trans, enum btree_id btree, struct bpos pos, unsigned update_flags) @@ -721,7 +710,7 @@ int bch2_btree_delete_range_trans(struct btree_trans *trans, enum btree_id id, int ret = 0; bch2_trans_iter_init(trans, &iter, id, start, BTREE_ITER_intent); - while ((k = bch2_btree_iter_peek_upto(&iter, end)).k) { + while ((k = bch2_btree_iter_peek_max(&iter, end)).k) { struct disk_reservation disk_res = bch2_disk_reservation_init(trans->c, 0); struct bkey_i delete; @@ -794,8 +783,7 @@ int bch2_btree_delete_range(struct bch_fs *c, enum btree_id id, return ret; } -int bch2_btree_bit_mod(struct btree_trans *trans, enum btree_id btree, - struct bpos pos, bool set) +int bch2_btree_bit_mod_iter(struct btree_trans *trans, struct btree_iter *iter, bool set) { struct bkey_i *k = bch2_trans_kmalloc(trans, sizeof(*k)); int ret = PTR_ERR_OR_ZERO(k); @@ -804,13 +792,21 @@ int bch2_btree_bit_mod(struct btree_trans *trans, enum btree_id btree, bkey_init(&k->k); k->k.type = set ? KEY_TYPE_set : KEY_TYPE_deleted; - k->k.p = pos; + k->k.p = iter->pos; + if (iter->flags & BTREE_ITER_is_extents) + bch2_key_resize(&k->k, 1); + return bch2_trans_update(trans, iter, k, 0); +} + +int bch2_btree_bit_mod(struct btree_trans *trans, enum btree_id btree, + struct bpos pos, bool set) +{ struct btree_iter iter; bch2_trans_iter_init(trans, &iter, btree, pos, BTREE_ITER_intent); - ret = bch2_btree_iter_traverse(&iter) ?: - bch2_trans_update(trans, &iter, k, 0); + int ret = bch2_btree_iter_traverse(&iter) ?: + bch2_btree_bit_mod_iter(trans, &iter, set); bch2_trans_iter_exit(trans, &iter); return ret; } @@ -827,10 +823,17 @@ int bch2_btree_bit_mod_buffered(struct btree_trans *trans, enum btree_id btree, return bch2_trans_update_buffered(trans, btree, &k); } -static int __bch2_trans_log_msg(struct btree_trans *trans, struct printbuf *buf, unsigned u64s) +int bch2_trans_log_msg(struct btree_trans *trans, struct printbuf *buf) { + unsigned u64s = DIV_ROUND_UP(buf->pos, sizeof(u64)); + prt_chars(buf, '\0', u64s * sizeof(u64) - buf->pos); + + int ret = buf->allocation_failure ? -BCH_ERR_ENOMEM_trans_log_msg : 0; + if (ret) + return ret; + struct jset_entry *e = bch2_trans_jset_entry_alloc(trans, jset_u64s(u64s)); - int ret = PTR_ERR_OR_ZERO(e); + ret = PTR_ERR_OR_ZERO(e); if (ret) return ret; @@ -865,9 +868,8 @@ __bch2_fs_log_msg(struct bch_fs *c, unsigned commit_flags, const char *fmt, memcpy(l->d, buf.buf, buf.pos); c->journal.early_journal_entries.nr += jset_u64s(u64s); } else { - ret = bch2_trans_commit_do(c, NULL, NULL, - BCH_TRANS_COMMIT_lazy_rw|commit_flags, - __bch2_trans_log_msg(trans, &buf, u64s)); + ret = bch2_trans_commit_do(c, NULL, NULL, commit_flags, + bch2_trans_log_msg(trans, &buf)); } err: printbuf_exit(&buf); diff --git a/fs/bcachefs/btree_update.h b/fs/bcachefs/btree_update.h index 70b3c989fac2..8f22ef9a7651 100644 --- a/fs/bcachefs/btree_update.h +++ b/fs/bcachefs/btree_update.h @@ -24,7 +24,6 @@ void bch2_btree_insert_key_leaf(struct btree_trans *, struct btree_path *, #define BCH_TRANS_COMMIT_FLAGS() \ x(no_enospc, "don't check for enospc") \ x(no_check_rw, "don't attempt to take a ref on c->writes") \ - x(lazy_rw, "go read-write if we haven't yet - only for use in recovery") \ x(no_journal_res, "don't take a journal reservation, instead " \ "pin journal entry referred to by trans->journal_res.seq") \ x(journal_reclaim, "operation required for journal reclaim; may return error" \ @@ -47,8 +46,6 @@ enum bch_trans_commit_flags { void bch2_trans_commit_flags_to_text(struct printbuf *, enum bch_trans_commit_flags); -int bch2_btree_delete_extent_at(struct btree_trans *, struct btree_iter *, - unsigned, unsigned); int bch2_btree_delete_at(struct btree_trans *, struct btree_iter *, unsigned); int bch2_btree_delete(struct btree_trans *, enum btree_id, struct bpos, unsigned); @@ -66,6 +63,7 @@ int bch2_btree_delete_range_trans(struct btree_trans *, enum btree_id, int bch2_btree_delete_range(struct bch_fs *, enum btree_id, struct bpos, struct bpos, unsigned, u64 *); +int bch2_btree_bit_mod_iter(struct btree_trans *, struct btree_iter *, bool); int bch2_btree_bit_mod(struct btree_trans *, enum btree_id, struct bpos, bool); int bch2_btree_bit_mod_buffered(struct btree_trans *, enum btree_id, struct bpos, bool); @@ -161,6 +159,7 @@ void bch2_trans_commit_hook(struct btree_trans *, struct btree_trans_commit_hook *); int __bch2_trans_commit(struct btree_trans *, unsigned); +int bch2_trans_log_msg(struct btree_trans *, struct printbuf *); __printf(2, 3) int bch2_fs_log_msg(struct bch_fs *, const char *, ...); __printf(2, 3) int bch2_journal_log_msg(struct bch_fs *, const char *, ...); @@ -244,7 +243,8 @@ static inline struct bkey_i *bch2_bkey_make_mut_noupdate(struct btree_trans *tra KEY_TYPE_##_type, sizeof(struct bkey_i_##_type))) static inline struct bkey_i *__bch2_bkey_make_mut(struct btree_trans *trans, struct btree_iter *iter, - struct bkey_s_c *k, unsigned flags, + struct bkey_s_c *k, + enum btree_iter_update_trigger_flags flags, unsigned type, unsigned min_bytes) { struct bkey_i *mut = __bch2_bkey_make_mut_noupdate(trans, *k, type, min_bytes); @@ -261,8 +261,9 @@ static inline struct bkey_i *__bch2_bkey_make_mut(struct btree_trans *trans, str return mut; } -static inline struct bkey_i *bch2_bkey_make_mut(struct btree_trans *trans, struct btree_iter *iter, - struct bkey_s_c *k, unsigned flags) +static inline struct bkey_i *bch2_bkey_make_mut(struct btree_trans *trans, + struct btree_iter *iter, struct bkey_s_c *k, + enum btree_iter_update_trigger_flags flags) { return __bch2_bkey_make_mut(trans, iter, k, flags, 0, 0); } @@ -274,7 +275,8 @@ static inline struct bkey_i *bch2_bkey_make_mut(struct btree_trans *trans, struc static inline struct bkey_i *__bch2_bkey_get_mut_noupdate(struct btree_trans *trans, struct btree_iter *iter, unsigned btree_id, struct bpos pos, - unsigned flags, unsigned type, unsigned min_bytes) + enum btree_iter_update_trigger_flags flags, + unsigned type, unsigned min_bytes) { struct bkey_s_c k = __bch2_bkey_get_iter(trans, iter, btree_id, pos, flags|BTREE_ITER_intent, type); @@ -289,7 +291,7 @@ static inline struct bkey_i *__bch2_bkey_get_mut_noupdate(struct btree_trans *tr static inline struct bkey_i *bch2_bkey_get_mut_noupdate(struct btree_trans *trans, struct btree_iter *iter, unsigned btree_id, struct bpos pos, - unsigned flags) + enum btree_iter_update_trigger_flags flags) { return __bch2_bkey_get_mut_noupdate(trans, iter, btree_id, pos, flags, 0, 0); } @@ -297,7 +299,8 @@ static inline struct bkey_i *bch2_bkey_get_mut_noupdate(struct btree_trans *tran static inline struct bkey_i *__bch2_bkey_get_mut(struct btree_trans *trans, struct btree_iter *iter, unsigned btree_id, struct bpos pos, - unsigned flags, unsigned type, unsigned min_bytes) + enum btree_iter_update_trigger_flags flags, + unsigned type, unsigned min_bytes) { struct bkey_i *mut = __bch2_bkey_get_mut_noupdate(trans, iter, btree_id, pos, flags|BTREE_ITER_intent, type, min_bytes); @@ -318,7 +321,8 @@ static inline struct bkey_i *__bch2_bkey_get_mut(struct btree_trans *trans, static inline struct bkey_i *bch2_bkey_get_mut_minsize(struct btree_trans *trans, struct btree_iter *iter, unsigned btree_id, struct bpos pos, - unsigned flags, unsigned min_bytes) + enum btree_iter_update_trigger_flags flags, + unsigned min_bytes) { return __bch2_bkey_get_mut(trans, iter, btree_id, pos, flags, 0, min_bytes); } @@ -326,7 +330,7 @@ static inline struct bkey_i *bch2_bkey_get_mut_minsize(struct btree_trans *trans static inline struct bkey_i *bch2_bkey_get_mut(struct btree_trans *trans, struct btree_iter *iter, unsigned btree_id, struct bpos pos, - unsigned flags) + enum btree_iter_update_trigger_flags flags) { return __bch2_bkey_get_mut(trans, iter, btree_id, pos, flags, 0, 0); } @@ -337,7 +341,8 @@ static inline struct bkey_i *bch2_bkey_get_mut(struct btree_trans *trans, KEY_TYPE_##_type, sizeof(struct bkey_i_##_type))) static inline struct bkey_i *__bch2_bkey_alloc(struct btree_trans *trans, struct btree_iter *iter, - unsigned flags, unsigned type, unsigned val_size) + enum btree_iter_update_trigger_flags flags, + unsigned type, unsigned val_size) { struct bkey_i *k = bch2_trans_kmalloc(trans, sizeof(*k) + val_size); int ret; diff --git a/fs/bcachefs/btree_update_interior.c b/fs/bcachefs/btree_update_interior.c index d596ef93239f..f4aeadbe53c1 100644 --- a/fs/bcachefs/btree_update_interior.c +++ b/fs/bcachefs/btree_update_interior.c @@ -58,11 +58,15 @@ int bch2_btree_node_check_topology(struct btree_trans *trans, struct btree *b) !bpos_eq(bkey_i_to_btree_ptr_v2(&b->key)->v.min_key, b->data->min_key)); + bch2_bkey_buf_init(&prev); + bkey_init(&prev.k->k); + bch2_btree_and_journal_iter_init_node_iter(trans, &iter, b); + if (b == btree_node_root(c, b)) { if (!bpos_eq(b->data->min_key, POS_MIN)) { printbuf_reset(&buf); bch2_bpos_to_text(&buf, b->data->min_key); - need_fsck_err(trans, btree_root_bad_min_key, + log_fsck_err(trans, btree_root_bad_min_key, "btree root with incorrect min_key: %s", buf.buf); goto topology_repair; } @@ -70,18 +74,14 @@ int bch2_btree_node_check_topology(struct btree_trans *trans, struct btree *b) if (!bpos_eq(b->data->max_key, SPOS_MAX)) { printbuf_reset(&buf); bch2_bpos_to_text(&buf, b->data->max_key); - need_fsck_err(trans, btree_root_bad_max_key, + log_fsck_err(trans, btree_root_bad_max_key, "btree root with incorrect max_key: %s", buf.buf); goto topology_repair; } } if (!b->c.level) - return 0; - - bch2_bkey_buf_init(&prev); - bkey_init(&prev.k->k); - bch2_btree_and_journal_iter_init_node_iter(trans, &iter, b); + goto out; while ((k = bch2_btree_and_journal_iter_peek(&iter)).k) { if (k.k->type != KEY_TYPE_btree_ptr_v2) @@ -97,16 +97,16 @@ int bch2_btree_node_check_topology(struct btree_trans *trans, struct btree *b) bch2_topology_error(c); printbuf_reset(&buf); - prt_str(&buf, "end of prev node doesn't match start of next node\n"), - prt_printf(&buf, " in btree %s level %u node ", - bch2_btree_id_str(b->c.btree_id), b->c.level); + prt_str(&buf, "end of prev node doesn't match start of next node\n in "); + bch2_btree_id_level_to_text(&buf, b->c.btree_id, b->c.level); + prt_str(&buf, " node "); bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(&b->key)); prt_str(&buf, "\n prev "); bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(prev.k)); prt_str(&buf, "\n next "); bch2_bkey_val_to_text(&buf, c, k); - need_fsck_err(trans, btree_node_topology_bad_min_key, "%s", buf.buf); + log_fsck_err(trans, btree_node_topology_bad_min_key, "%s", buf.buf); goto topology_repair; } @@ -118,25 +118,25 @@ int bch2_btree_node_check_topology(struct btree_trans *trans, struct btree *b) bch2_topology_error(c); printbuf_reset(&buf); - prt_str(&buf, "empty interior node\n"); - prt_printf(&buf, " in btree %s level %u node ", - bch2_btree_id_str(b->c.btree_id), b->c.level); + prt_str(&buf, "empty interior node\n in "); + bch2_btree_id_level_to_text(&buf, b->c.btree_id, b->c.level); + prt_str(&buf, " node "); bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(&b->key)); - need_fsck_err(trans, btree_node_topology_empty_interior_node, "%s", buf.buf); + log_fsck_err(trans, btree_node_topology_empty_interior_node, "%s", buf.buf); goto topology_repair; } else if (!bpos_eq(prev.k->k.p, b->key.k.p)) { bch2_topology_error(c); printbuf_reset(&buf); - prt_str(&buf, "last child node doesn't end at end of parent node\n"); - prt_printf(&buf, " in btree %s level %u node ", - bch2_btree_id_str(b->c.btree_id), b->c.level); + prt_str(&buf, "last child node doesn't end at end of parent node\n in "); + bch2_btree_id_level_to_text(&buf, b->c.btree_id, b->c.level); + prt_str(&buf, " node "); bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(&b->key)); prt_str(&buf, "\n last key "); bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(prev.k)); - need_fsck_err(trans, btree_node_topology_bad_max_key, "%s", buf.buf); + log_fsck_err(trans, btree_node_topology_bad_max_key, "%s", buf.buf); goto topology_repair; } out: @@ -146,13 +146,7 @@ fsck_err: printbuf_exit(&buf); return ret; topology_repair: - if ((c->opts.recovery_passes & BIT_ULL(BCH_RECOVERY_PASS_check_topology)) && - c->curr_recovery_pass > BCH_RECOVERY_PASS_check_topology) { - bch2_inconsistent_error(c); - ret = -BCH_ERR_btree_need_topology_repair; - } else { - ret = bch2_run_explicit_recovery_pass(c, BCH_RECOVERY_PASS_check_topology); - } + ret = bch2_topology_error(c); goto out; } @@ -244,7 +238,6 @@ static void bch2_btree_node_free_inmem(struct btree_trans *trans, struct btree *b) { struct bch_fs *c = trans->c; - unsigned i, level = b->c.level; bch2_btree_node_lock_write_nofail(trans, path, &b->c); @@ -255,13 +248,9 @@ static void bch2_btree_node_free_inmem(struct btree_trans *trans, mutex_unlock(&c->btree_cache.lock); six_unlock_write(&b->c.lock); - mark_btree_node_locked_noreset(path, level, BTREE_NODE_INTENT_LOCKED); + mark_btree_node_locked_noreset(path, b->c.level, BTREE_NODE_INTENT_LOCKED); - trans_for_each_path(trans, path, i) - if (path->l[level].b == b) { - btree_node_unlock(trans, path, level); - path->l[level].b = ERR_PTR(-BCH_ERR_no_btree_node_init); - } + bch2_trans_node_drop(trans, b); } static void bch2_btree_node_free_never_used(struct btree_update *as, @@ -270,8 +259,6 @@ static void bch2_btree_node_free_never_used(struct btree_update *as, { struct bch_fs *c = as->c; struct prealloc_nodes *p = &as->prealloc_nodes[b->c.lock.readers != NULL]; - struct btree_path *path; - unsigned i, level = b->c.level; BUG_ON(!list_empty(&b->write_blocked)); BUG_ON(b->will_make_reachable != (1UL|(unsigned long) as)); @@ -293,11 +280,7 @@ static void bch2_btree_node_free_never_used(struct btree_update *as, six_unlock_intent(&b->c.lock); - trans_for_each_path(trans, path, i) - if (path->l[level].b == b) { - btree_node_unlock(trans, path, level); - path->l[level].b = ERR_PTR(-BCH_ERR_no_btree_node_init); - } + bch2_trans_node_drop(trans, b); } static struct btree *__bch2_btree_node_alloc(struct btree_trans *trans, @@ -809,7 +792,7 @@ err: mark_btree_node_locked_noreset(path, b->c.level, BTREE_NODE_INTENT_LOCKED); six_unlock_write(&b->c.lock); - btree_node_write_if_need(c, b, SIX_LOCK_intent); + btree_node_write_if_need(trans, b, SIX_LOCK_intent); btree_node_unlock(trans, path, b->c.level); bch2_path_put(trans, path_idx, true); } @@ -830,7 +813,7 @@ err: b = as->new_nodes[i]; btree_node_lock_nopath_nofail(trans, &b->c, SIX_LOCK_read); - btree_node_write_if_need(c, b, SIX_LOCK_read); + btree_node_write_if_need(trans, b, SIX_LOCK_read); six_unlock_read(&b->c.lock); } @@ -1366,9 +1349,14 @@ static void bch2_insert_fixup_btree_ptr(struct btree_update *as, if (unlikely(!test_bit(JOURNAL_replay_done, &c->journal.flags))) bch2_journal_key_overwritten(c, b->c.btree_id, b->c.level, insert->k.p); - if (bch2_bkey_validate(c, bkey_i_to_s_c(insert), - btree_node_type(b), BCH_VALIDATE_write) ?: - bch2_bkey_in_btree_node(c, b, bkey_i_to_s_c(insert), BCH_VALIDATE_write)) { + struct bkey_validate_context from = (struct bkey_validate_context) { + .from = BKEY_VALIDATE_btree_node, + .level = b->c.level, + .btree = b->c.btree_id, + .flags = BCH_VALIDATE_commit, + }; + if (bch2_bkey_validate(c, bkey_i_to_s_c(insert), from) ?: + bch2_bkey_in_btree_node(c, b, bkey_i_to_s_c(insert), from)) { bch2_fs_inconsistent(c, "%s: inserting invalid bkey", __func__); dump_stack(); } @@ -1418,15 +1406,26 @@ bch2_btree_insert_keys_interior(struct btree_update *as, (bkey_cmp_left_packed(b, k, &insert->k.p) >= 0)) ; - while (!bch2_keylist_empty(keys)) { - insert = bch2_keylist_front(keys); + for (; + insert != keys->top && bpos_le(insert->k.p, b->key.k.p); + insert = bkey_next(insert)) + bch2_insert_fixup_btree_ptr(as, trans, path, b, &node_iter, insert); - if (bpos_gt(insert->k.p, b->key.k.p)) - break; + if (bch2_btree_node_check_topology(trans, b)) { + struct printbuf buf = PRINTBUF; - bch2_insert_fixup_btree_ptr(as, trans, path, b, &node_iter, insert); - bch2_keylist_pop_front(keys); + for (struct bkey_i *k = keys->keys; + k != insert; + k = bkey_next(k)) { + bch2_bkey_val_to_text(&buf, trans->c, bkey_i_to_s_c(k)); + prt_newline(&buf); + } + + panic("%s(): check_topology error: inserted keys\n%s", __func__, buf.buf); } + + memmove_u64s_down(keys->keys, insert, keys->top_p - insert->_data); + keys->top_p -= insert->_data - keys->keys_p; } static bool key_deleted_in_insert(struct keylist *insert_keys, struct bpos pos) @@ -1575,8 +1574,6 @@ static void btree_split_insert_keys(struct btree_update *as, bch2_btree_node_iter_init(&node_iter, b, &bch2_keylist_front(keys)->k.p); bch2_btree_insert_keys_interior(as, trans, path, b, node_iter, keys); - - BUG_ON(bch2_btree_node_check_topology(trans, b)); } } @@ -1599,8 +1596,6 @@ static int btree_split(struct btree_update *as, struct btree_trans *trans, if (ret) return ret; - bch2_btree_interior_update_will_free_node(as, b); - if (b->nr.live_u64s > BTREE_SPLIT_THRESHOLD(c)) { struct btree *n[2]; @@ -1699,16 +1694,18 @@ static int btree_split(struct btree_update *as, struct btree_trans *trans, if (ret) goto err; + bch2_btree_interior_update_will_free_node(as, b); + if (n3) { bch2_btree_update_get_open_buckets(as, n3); - bch2_btree_node_write(c, n3, SIX_LOCK_intent, 0); + bch2_btree_node_write_trans(trans, n3, SIX_LOCK_intent, 0); } if (n2) { bch2_btree_update_get_open_buckets(as, n2); - bch2_btree_node_write(c, n2, SIX_LOCK_intent, 0); + bch2_btree_node_write_trans(trans, n2, SIX_LOCK_intent, 0); } bch2_btree_update_get_open_buckets(as, n1); - bch2_btree_node_write(c, n1, SIX_LOCK_intent, 0); + bch2_btree_node_write_trans(trans, n1, SIX_LOCK_intent, 0); /* * The old node must be freed (in memory) _before_ unlocking the new @@ -1827,8 +1824,6 @@ static int bch2_btree_insert_node(struct btree_update *as, struct btree_trans *t btree_update_updated_node(as, b); bch2_btree_node_unlock_write(trans, path, b); - - BUG_ON(bch2_btree_node_check_topology(trans, b)); return 0; split: /* @@ -1905,7 +1900,7 @@ static void __btree_increase_depth(struct btree_update *as, struct btree_trans * BUG_ON(ret); bch2_btree_update_get_open_buckets(as, n); - bch2_btree_node_write(c, n, SIX_LOCK_intent, 0); + bch2_btree_node_write_trans(trans, n, SIX_LOCK_intent, 0); bch2_trans_node_add(trans, path, n); six_unlock_intent(&n->c.lock); @@ -1953,8 +1948,7 @@ int __bch2_foreground_maybe_merge(struct btree_trans *trans, u64 start_time = local_clock(); int ret = 0; - bch2_trans_verify_not_in_restart(trans); - bch2_trans_verify_not_unlocked(trans); + bch2_trans_verify_not_unlocked_or_in_restart(trans); BUG_ON(!trans->paths[path].should_be_locked); BUG_ON(!btree_node_locked(&trans->paths[path], level)); @@ -2058,9 +2052,6 @@ int __bch2_foreground_maybe_merge(struct btree_trans *trans, trace_and_count(c, btree_node_merge, trans, b); - bch2_btree_interior_update_will_free_node(as, b); - bch2_btree_interior_update_will_free_node(as, m); - n = bch2_btree_node_alloc(as, trans, b->c.level); SET_BTREE_NODE_SEQ(n->data, @@ -2096,10 +2087,13 @@ int __bch2_foreground_maybe_merge(struct btree_trans *trans, if (ret) goto err_free_update; + bch2_btree_interior_update_will_free_node(as, b); + bch2_btree_interior_update_will_free_node(as, m); + bch2_trans_verify_paths(trans); bch2_btree_update_get_open_buckets(as, n); - bch2_btree_node_write(c, n, SIX_LOCK_intent, 0); + bch2_btree_node_write_trans(trans, n, SIX_LOCK_intent, 0); bch2_btree_node_free_inmem(trans, trans->paths + path, b); bch2_btree_node_free_inmem(trans, trans->paths + sib_path, m); @@ -2150,8 +2144,6 @@ int bch2_btree_node_rewrite(struct btree_trans *trans, if (ret) goto out; - bch2_btree_interior_update_will_free_node(as, b); - n = bch2_btree_node_alloc_replacement(as, trans, b); bch2_btree_build_aux_trees(n); @@ -2175,8 +2167,10 @@ int bch2_btree_node_rewrite(struct btree_trans *trans, if (ret) goto err; + bch2_btree_interior_update_will_free_node(as, b); + bch2_btree_update_get_open_buckets(as, n); - bch2_btree_node_write(c, n, SIX_LOCK_intent, 0); + bch2_btree_node_write_trans(trans, n, SIX_LOCK_intent, 0); bch2_btree_node_free_inmem(trans, btree_iter_path(trans, iter), b); @@ -2201,42 +2195,50 @@ struct async_btree_rewrite { struct list_head list; enum btree_id btree_id; unsigned level; - struct bpos pos; - __le64 seq; + struct bkey_buf key; }; static int async_btree_node_rewrite_trans(struct btree_trans *trans, struct async_btree_rewrite *a) { - struct bch_fs *c = trans->c; struct btree_iter iter; - struct btree *b; - int ret; - - bch2_trans_node_iter_init(trans, &iter, a->btree_id, a->pos, + bch2_trans_node_iter_init(trans, &iter, + a->btree_id, a->key.k->k.p, BTREE_MAX_DEPTH, a->level, 0); - b = bch2_btree_iter_peek_node(&iter); - ret = PTR_ERR_OR_ZERO(b); + struct btree *b = bch2_btree_iter_peek_node(&iter); + int ret = PTR_ERR_OR_ZERO(b); if (ret) goto out; - if (!b || b->data->keys.seq != a->seq) { + bool found = b && btree_ptr_hash_val(&b->key) == btree_ptr_hash_val(a->key.k); + ret = found + ? bch2_btree_node_rewrite(trans, &iter, b, 0) + : -ENOENT; + +#if 0 + /* Tracepoint... */ + if (!ret || ret == -ENOENT) { + struct bch_fs *c = trans->c; struct printbuf buf = PRINTBUF; - if (b) - bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(&b->key)); - else - prt_str(&buf, "(null"); - bch_info(c, "%s: node to rewrite not found:, searching for seq %llu, got\n%s", - __func__, a->seq, buf.buf); + if (!ret) { + prt_printf(&buf, "rewrite node:\n "); + bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(a->key.k)); + } else { + prt_printf(&buf, "node to rewrite not found:\n want: "); + bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(a->key.k)); + prt_printf(&buf, "\n got: "); + if (b) + bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(&b->key)); + else + prt_str(&buf, "(null)"); + } + bch_info(c, "%s", buf.buf); printbuf_exit(&buf); - goto out; } - - ret = bch2_btree_node_rewrite(trans, &iter, b, 0); +#endif out: bch2_trans_iter_exit(trans, &iter); - return ret; } @@ -2247,81 +2249,97 @@ static void async_btree_node_rewrite_work(struct work_struct *work) struct bch_fs *c = a->c; int ret = bch2_trans_do(c, async_btree_node_rewrite_trans(trans, a)); - bch_err_fn_ratelimited(c, ret); + if (ret != -ENOENT) + bch_err_fn_ratelimited(c, ret); + + spin_lock(&c->btree_node_rewrites_lock); + list_del(&a->list); + spin_unlock(&c->btree_node_rewrites_lock); + + closure_wake_up(&c->btree_node_rewrites_wait); + + bch2_bkey_buf_exit(&a->key, c); bch2_write_ref_put(c, BCH_WRITE_REF_node_rewrite); kfree(a); } void bch2_btree_node_rewrite_async(struct bch_fs *c, struct btree *b) { - struct async_btree_rewrite *a; - int ret; - - a = kmalloc(sizeof(*a), GFP_NOFS); - if (!a) { - bch_err(c, "%s: error allocating memory", __func__); + struct async_btree_rewrite *a = kmalloc(sizeof(*a), GFP_NOFS); + if (!a) return; - } a->c = c; a->btree_id = b->c.btree_id; a->level = b->c.level; - a->pos = b->key.k.p; - a->seq = b->data->keys.seq; INIT_WORK(&a->work, async_btree_node_rewrite_work); - if (unlikely(!test_bit(BCH_FS_may_go_rw, &c->flags))) { - mutex_lock(&c->pending_node_rewrites_lock); - list_add(&a->list, &c->pending_node_rewrites); - mutex_unlock(&c->pending_node_rewrites_lock); - return; - } + bch2_bkey_buf_init(&a->key); + bch2_bkey_buf_copy(&a->key, c, &b->key); - if (!bch2_write_ref_tryget(c, BCH_WRITE_REF_node_rewrite)) { - if (test_bit(BCH_FS_started, &c->flags)) { - bch_err(c, "%s: error getting c->writes ref", __func__); - kfree(a); - return; - } + bool now = false, pending = false; - ret = bch2_fs_read_write_early(c); - bch_err_msg(c, ret, "going read-write"); - if (ret) { - kfree(a); - return; - } + spin_lock(&c->btree_node_rewrites_lock); + if (c->curr_recovery_pass > BCH_RECOVERY_PASS_journal_replay && + bch2_write_ref_tryget(c, BCH_WRITE_REF_node_rewrite)) { + list_add(&a->list, &c->btree_node_rewrites); + now = true; + } else if (!test_bit(BCH_FS_may_go_rw, &c->flags)) { + list_add(&a->list, &c->btree_node_rewrites_pending); + pending = true; + } + spin_unlock(&c->btree_node_rewrites_lock); - bch2_write_ref_get(c, BCH_WRITE_REF_node_rewrite); + if (now) { + queue_work(c->btree_node_rewrite_worker, &a->work); + } else if (pending) { + /* bch2_do_pending_node_rewrites will execute */ + } else { + bch2_bkey_buf_exit(&a->key, c); + kfree(a); } +} - queue_work(c->btree_node_rewrite_worker, &a->work); +void bch2_async_btree_node_rewrites_flush(struct bch_fs *c) +{ + closure_wait_event(&c->btree_node_rewrites_wait, + list_empty(&c->btree_node_rewrites)); } void bch2_do_pending_node_rewrites(struct bch_fs *c) { - struct async_btree_rewrite *a, *n; - - mutex_lock(&c->pending_node_rewrites_lock); - list_for_each_entry_safe(a, n, &c->pending_node_rewrites, list) { - list_del(&a->list); + while (1) { + spin_lock(&c->btree_node_rewrites_lock); + struct async_btree_rewrite *a = + list_pop_entry(&c->btree_node_rewrites_pending, + struct async_btree_rewrite, list); + if (a) + list_add(&a->list, &c->btree_node_rewrites); + spin_unlock(&c->btree_node_rewrites_lock); + + if (!a) + break; bch2_write_ref_get(c, BCH_WRITE_REF_node_rewrite); queue_work(c->btree_node_rewrite_worker, &a->work); } - mutex_unlock(&c->pending_node_rewrites_lock); } void bch2_free_pending_node_rewrites(struct bch_fs *c) { - struct async_btree_rewrite *a, *n; + while (1) { + spin_lock(&c->btree_node_rewrites_lock); + struct async_btree_rewrite *a = + list_pop_entry(&c->btree_node_rewrites_pending, + struct async_btree_rewrite, list); + spin_unlock(&c->btree_node_rewrites_lock); - mutex_lock(&c->pending_node_rewrites_lock); - list_for_each_entry_safe(a, n, &c->pending_node_rewrites, list) { - list_del(&a->list); + if (!a) + break; + bch2_bkey_buf_exit(&a->key, c); kfree(a); } - mutex_unlock(&c->pending_node_rewrites_lock); } static int __bch2_btree_node_update_key(struct btree_trans *trans, @@ -2575,8 +2593,9 @@ static void bch2_btree_update_to_text(struct printbuf *out, struct btree_update prt_printf(out, "%ps: ", (void *) as->ip_started); bch2_trans_commit_flags_to_text(out, as->flags); - prt_printf(out, " btree=%s l=%u-%u mode=%s nodes_written=%u cl.remaining=%u journal_seq=%llu\n", - bch2_btree_id_str(as->btree_id), + prt_str(out, " "); + bch2_btree_id_to_text(out, as->btree_id); + prt_printf(out, " l=%u-%u mode=%s nodes_written=%u cl.remaining=%u journal_seq=%llu\n", as->update_level_start, as->update_level_end, bch2_btree_update_modes[as->mode], @@ -2677,6 +2696,9 @@ void bch2_btree_reserve_cache_to_text(struct printbuf *out, struct bch_fs *c) void bch2_fs_btree_interior_update_exit(struct bch_fs *c) { + WARN_ON(!list_empty(&c->btree_node_rewrites)); + WARN_ON(!list_empty(&c->btree_node_rewrites_pending)); + if (c->btree_node_rewrite_worker) destroy_workqueue(c->btree_node_rewrite_worker); if (c->btree_interior_update_worker) @@ -2692,8 +2714,9 @@ void bch2_fs_btree_interior_update_init_early(struct bch_fs *c) mutex_init(&c->btree_interior_update_lock); INIT_WORK(&c->btree_interior_update_work, btree_interior_update_work); - INIT_LIST_HEAD(&c->pending_node_rewrites); - mutex_init(&c->pending_node_rewrites_lock); + INIT_LIST_HEAD(&c->btree_node_rewrites); + INIT_LIST_HEAD(&c->btree_node_rewrites_pending); + spin_lock_init(&c->btree_node_rewrites_lock); } int bch2_fs_btree_interior_update_init(struct bch_fs *c) diff --git a/fs/bcachefs/btree_update_interior.h b/fs/bcachefs/btree_update_interior.h index 10f400957f21..7930ffea3075 100644 --- a/fs/bcachefs/btree_update_interior.h +++ b/fs/bcachefs/btree_update_interior.h @@ -159,7 +159,7 @@ static inline int bch2_foreground_maybe_merge(struct btree_trans *trans, unsigned level, unsigned flags) { - bch2_trans_verify_not_unlocked(trans); + bch2_trans_verify_not_unlocked_or_in_restart(trans); return bch2_foreground_maybe_merge_sibling(trans, path, level, flags, btree_prev_sib) ?: @@ -334,6 +334,7 @@ void bch2_journal_entry_to_btree_root(struct bch_fs *, struct jset_entry *); struct jset_entry *bch2_btree_roots_to_journal_entries(struct bch_fs *, struct jset_entry *, unsigned long); +void bch2_async_btree_node_rewrites_flush(struct bch_fs *); void bch2_do_pending_node_rewrites(struct bch_fs *); void bch2_free_pending_node_rewrites(struct bch_fs *); diff --git a/fs/bcachefs/btree_write_buffer.c b/fs/bcachefs/btree_write_buffer.c index 1639c60dffa0..b56c4987b8c9 100644 --- a/fs/bcachefs/btree_write_buffer.c +++ b/fs/bcachefs/btree_write_buffer.c @@ -19,8 +19,6 @@ static int bch2_btree_write_buffer_journal_flush(struct journal *, struct journal_entry_pin *, u64); -static int bch2_journal_keys_to_write_buffer(struct bch_fs *, struct journal_buf *); - static inline bool __wb_key_ref_cmp(const struct wb_key_ref *l, const struct wb_key_ref *r) { return (cmp_int(l->hi, r->hi) ?: @@ -314,6 +312,8 @@ static int bch2_btree_write_buffer_flush_locked(struct btree_trans *trans) darray_for_each(wb->sorted, i) { struct btree_write_buffered_key *k = &wb->flushing.keys.data[i->idx]; + BUG_ON(!btree_type_uses_write_buffer(k->btree)); + for (struct wb_key_ref *n = i + 1; n < min(i + 4, &darray_top(wb->sorted)); n++) prefetch(&wb->flushing.keys.data[n->idx]); @@ -481,21 +481,55 @@ err: return ret; } -static int fetch_wb_keys_from_journal(struct bch_fs *c, u64 seq) +static int bch2_journal_keys_to_write_buffer(struct bch_fs *c, struct journal_buf *buf) +{ + struct journal_keys_to_wb dst; + int ret = 0; + + bch2_journal_keys_to_write_buffer_start(c, &dst, le64_to_cpu(buf->data->seq)); + + for_each_jset_entry_type(entry, buf->data, BCH_JSET_ENTRY_write_buffer_keys) { + jset_entry_for_each_key(entry, k) { + ret = bch2_journal_key_to_wb(c, &dst, entry->btree_id, k); + if (ret) + goto out; + } + + entry->type = BCH_JSET_ENTRY_btree_keys; + } +out: + ret = bch2_journal_keys_to_write_buffer_end(c, &dst) ?: ret; + return ret; +} + +static int fetch_wb_keys_from_journal(struct bch_fs *c, u64 max_seq) { struct journal *j = &c->journal; struct journal_buf *buf; + bool blocked; int ret = 0; - while (!ret && (buf = bch2_next_write_buffer_flush_journal_buf(j, seq))) { + while (!ret && (buf = bch2_next_write_buffer_flush_journal_buf(j, max_seq, &blocked))) { ret = bch2_journal_keys_to_write_buffer(c, buf); + + if (!blocked && !ret) { + spin_lock(&j->lock); + buf->need_flush_to_write_buffer = false; + spin_unlock(&j->lock); + } + mutex_unlock(&j->buf_lock); + + if (blocked) { + bch2_journal_unblock(j); + break; + } } return ret; } -static int btree_write_buffer_flush_seq(struct btree_trans *trans, u64 seq, +static int btree_write_buffer_flush_seq(struct btree_trans *trans, u64 max_seq, bool *did_work) { struct bch_fs *c = trans->c; @@ -505,7 +539,7 @@ static int btree_write_buffer_flush_seq(struct btree_trans *trans, u64 seq, do { bch2_trans_unlock(trans); - fetch_from_journal_err = fetch_wb_keys_from_journal(c, seq); + fetch_from_journal_err = fetch_wb_keys_from_journal(c, max_seq); *did_work |= wb->inc.keys.nr || wb->flushing.keys.nr; @@ -518,8 +552,8 @@ static int btree_write_buffer_flush_seq(struct btree_trans *trans, u64 seq, mutex_unlock(&wb->flushing.lock); } while (!ret && (fetch_from_journal_err || - (wb->inc.pin.seq && wb->inc.pin.seq <= seq) || - (wb->flushing.pin.seq && wb->flushing.pin.seq <= seq))); + (wb->inc.pin.seq && wb->inc.pin.seq <= max_seq) || + (wb->flushing.pin.seq && wb->flushing.pin.seq <= max_seq))); return ret; } @@ -600,6 +634,14 @@ int bch2_btree_write_buffer_maybe_flush(struct btree_trans *trans, bch2_bkey_buf_init(&tmp); if (!bkey_and_val_eq(referring_k, bkey_i_to_s_c(last_flushed->k))) { + if (trace_write_buffer_maybe_flush_enabled()) { + struct printbuf buf = PRINTBUF; + + bch2_bkey_val_to_text(&buf, c, referring_k); + trace_write_buffer_maybe_flush(trans, _RET_IP_, buf.buf); + printbuf_exit(&buf); + } + bch2_bkey_buf_reassemble(&tmp, c, referring_k); if (bkey_is_btree_ptr(referring_k.k)) { @@ -771,31 +813,6 @@ int bch2_journal_keys_to_write_buffer_end(struct bch_fs *c, struct journal_keys_ return ret; } -static int bch2_journal_keys_to_write_buffer(struct bch_fs *c, struct journal_buf *buf) -{ - struct journal_keys_to_wb dst; - int ret = 0; - - bch2_journal_keys_to_write_buffer_start(c, &dst, le64_to_cpu(buf->data->seq)); - - for_each_jset_entry_type(entry, buf->data, BCH_JSET_ENTRY_write_buffer_keys) { - jset_entry_for_each_key(entry, k) { - ret = bch2_journal_key_to_wb(c, &dst, entry->btree_id, k); - if (ret) - goto out; - } - - entry->type = BCH_JSET_ENTRY_btree_keys; - } - - spin_lock(&c->journal.lock); - buf->need_flush_to_write_buffer = false; - spin_unlock(&c->journal.lock); -out: - ret = bch2_journal_keys_to_write_buffer_end(c, &dst) ?: ret; - return ret; -} - static int wb_keys_resize(struct btree_write_buffer_keys *wb, size_t new_size) { if (wb->keys.size >= new_size) diff --git a/fs/bcachefs/buckets.c b/fs/bcachefs/buckets.c index ec7d9a59bea9..345b117a4a4a 100644 --- a/fs/bcachefs/buckets.c +++ b/fs/bcachefs/buckets.c @@ -18,7 +18,9 @@ #include "error.h" #include "inode.h" #include "movinggc.h" +#include "rebalance.h" #include "recovery.h" +#include "recovery_passes.h" #include "reflink.h" #include "replicas.h" #include "subvolume.h" @@ -260,8 +262,6 @@ int bch2_check_fix_ptrs(struct btree_trans *trans, struct printbuf buf = PRINTBUF; int ret = 0; - percpu_down_read(&c->mark_lock); - bkey_for_each_ptr_decode(k.k, ptrs_c, p, entry_c) { ret = bch2_check_fix_ptr(trans, k, p, entry_c, &do_update); if (ret) @@ -362,7 +362,6 @@ found: bch_info(c, "new key %s", buf.buf); } - percpu_up_read(&c->mark_lock); struct btree_iter iter; bch2_trans_node_iter_init(trans, &iter, btree, new->k.p, 0, level, BTREE_ITER_intent|BTREE_ITER_all_snapshots); @@ -371,8 +370,6 @@ found: BTREE_UPDATE_internal_snapshot_node| BTREE_TRIGGER_norun); bch2_trans_iter_exit(trans, &iter); - percpu_down_read(&c->mark_lock); - if (ret) goto err; @@ -380,7 +377,6 @@ found: bch2_btree_node_update_key_early(trans, btree, level - 1, k, new); } err: - percpu_up_read(&c->mark_lock); printbuf_exit(&buf); return ret; } @@ -401,8 +397,8 @@ int bch2_bucket_ref_update(struct btree_trans *trans, struct bch_dev *ca, BUG_ON(!sectors); if (gen_after(ptr->gen, b_gen)) { - bch2_fsck_err(trans, FSCK_CAN_IGNORE|FSCK_NEED_FSCK, - ptr_gen_newer_than_bucket_gen, + bch2_run_explicit_recovery_pass(c, BCH_RECOVERY_PASS_check_allocations); + log_fsck_err(trans, ptr_gen_newer_than_bucket_gen, "bucket %u:%zu gen %u data type %s: ptr gen %u newer than bucket gen\n" "while marking %s", ptr->dev, bucket_nr, b_gen, @@ -415,8 +411,8 @@ int bch2_bucket_ref_update(struct btree_trans *trans, struct bch_dev *ca, } if (gen_cmp(b_gen, ptr->gen) > BUCKET_GC_GEN_MAX) { - bch2_fsck_err(trans, FSCK_CAN_IGNORE|FSCK_NEED_FSCK, - ptr_too_stale, + bch2_run_explicit_recovery_pass(c, BCH_RECOVERY_PASS_check_allocations); + log_fsck_err(trans, ptr_too_stale, "bucket %u:%zu gen %u data type %s: ptr gen %u too stale\n" "while marking %s", ptr->dev, bucket_nr, b_gen, @@ -435,8 +431,8 @@ int bch2_bucket_ref_update(struct btree_trans *trans, struct bch_dev *ca, } if (b_gen != ptr->gen) { - bch2_fsck_err(trans, FSCK_CAN_IGNORE|FSCK_NEED_FSCK, - stale_dirty_ptr, + bch2_run_explicit_recovery_pass(c, BCH_RECOVERY_PASS_check_allocations); + log_fsck_err(trans, stale_dirty_ptr, "bucket %u:%zu gen %u (mem gen %u) data type %s: stale dirty ptr (gen %u)\n" "while marking %s", ptr->dev, bucket_nr, b_gen, @@ -451,8 +447,8 @@ int bch2_bucket_ref_update(struct btree_trans *trans, struct bch_dev *ca, } if (bucket_data_type_mismatch(bucket_data_type, ptr_data_type)) { - bch2_fsck_err(trans, FSCK_CAN_IGNORE|FSCK_NEED_FSCK, - ptr_bucket_data_type_mismatch, + bch2_run_explicit_recovery_pass(c, BCH_RECOVERY_PASS_check_allocations); + log_fsck_err(trans, ptr_bucket_data_type_mismatch, "bucket %u:%zu gen %u different types of data in same bucket: %s, %s\n" "while marking %s", ptr->dev, bucket_nr, b_gen, @@ -466,8 +462,8 @@ int bch2_bucket_ref_update(struct btree_trans *trans, struct bch_dev *ca, } if ((u64) *bucket_sectors + sectors > U32_MAX) { - bch2_fsck_err(trans, FSCK_CAN_IGNORE|FSCK_NEED_FSCK, - bucket_sector_count_overflow, + bch2_run_explicit_recovery_pass(c, BCH_RECOVERY_PASS_check_allocations); + log_fsck_err(trans, bucket_sector_count_overflow, "bucket %u:%zu gen %u data type %s sector count overflow: %u + %lli > U32_MAX\n" "while marking %s", ptr->dev, bucket_nr, b_gen, @@ -485,7 +481,9 @@ out: printbuf_exit(&buf); return ret; err: +fsck_err: bch2_dump_trans_updates(trans); + bch2_inconsistent_error(c); ret = -BCH_ERR_bucket_ref_update; goto out; } @@ -543,7 +541,8 @@ static int __mark_pointer(struct btree_trans *trans, struct bch_dev *ca, struct bkey_s_c k, const struct extent_ptr_decoded *p, s64 sectors, enum bch_data_type ptr_data_type, - struct bch_alloc_v4 *a) + struct bch_alloc_v4 *a, + bool insert) { u32 *dst_sectors = p->has_ec ? &a->stripe_sectors : !p->ptr.cached ? &a->dirty_sectors : @@ -553,8 +552,8 @@ static int __mark_pointer(struct btree_trans *trans, struct bch_dev *ca, if (ret) return ret; - - alloc_data_type_set(a, ptr_data_type); + if (insert) + alloc_data_type_set(a, ptr_data_type); return 0; } @@ -570,8 +569,10 @@ static int bch2_trigger_pointer(struct btree_trans *trans, struct printbuf buf = PRINTBUF; int ret = 0; - u64 abs_sectors = ptr_disk_sectors(level ? btree_sectors(c) : k.k->size, p); - *sectors = insert ? abs_sectors : -abs_sectors; + struct bkey_i_backpointer bp; + bch2_extent_ptr_to_bp(c, btree_id, level, k, p, entry, &bp); + + *sectors = insert ? bp.v.bucket_len : -(s64) bp.v.bucket_len; struct bch_dev *ca = bch2_dev_tryget(c, p.ptr.dev); if (unlikely(!ca)) { @@ -580,41 +581,36 @@ static int bch2_trigger_pointer(struct btree_trans *trans, goto err; } - struct bpos bucket; - struct bch_backpointer bp; - __bch2_extent_ptr_to_bp(trans->c, ca, btree_id, level, k, p, entry, &bucket, &bp, abs_sectors); + struct bpos bucket = PTR_BUCKET_POS(ca, &p.ptr); if (flags & BTREE_TRIGGER_transactional) { struct bkey_i_alloc_v4 *a = bch2_trans_start_alloc_update(trans, bucket, 0); ret = PTR_ERR_OR_ZERO(a) ?: - __mark_pointer(trans, ca, k, &p, *sectors, bp.data_type, &a->v); + __mark_pointer(trans, ca, k, &p, *sectors, bp.v.data_type, &a->v, insert); if (ret) goto err; if (!p.ptr.cached) { - ret = bch2_bucket_backpointer_mod(trans, ca, bucket, bp, k, insert); + ret = bch2_bucket_backpointer_mod(trans, k, &bp, insert); if (ret) goto err; } } if (flags & BTREE_TRIGGER_gc) { - percpu_down_read(&c->mark_lock); struct bucket *g = gc_bucket(ca, bucket.offset); if (bch2_fs_inconsistent_on(!g, c, "reference to invalid bucket on device %u\n %s", p.ptr.dev, (bch2_bkey_val_to_text(&buf, c, k), buf.buf))) { ret = -BCH_ERR_trigger_pointer; - goto err_unlock; + goto err; } bucket_lock(g); struct bch_alloc_v4 old = bucket_m_to_alloc(*g), new = old; - ret = __mark_pointer(trans, ca, k, &p, *sectors, bp.data_type, &new); + ret = __mark_pointer(trans, ca, k, &p, *sectors, bp.v.data_type, &new, insert); alloc_to_bucket(g, new); bucket_unlock(g); -err_unlock: - percpu_up_read(&c->mark_lock); if (!ret) ret = bch2_alloc_key_to_dev_counters(trans, ca, &old, &new, flags); @@ -951,6 +947,7 @@ static int __bch2_trans_mark_metadata_bucket(struct btree_trans *trans, enum bch_data_type type, unsigned sectors) { + struct bch_fs *c = trans->c; struct btree_iter iter; int ret = 0; @@ -960,8 +957,8 @@ static int __bch2_trans_mark_metadata_bucket(struct btree_trans *trans, return PTR_ERR(a); if (a->v.data_type && type && a->v.data_type != type) { - bch2_fsck_err(trans, FSCK_CAN_IGNORE|FSCK_NEED_FSCK, - bucket_metadata_type_mismatch, + bch2_run_explicit_recovery_pass(c, BCH_RECOVERY_PASS_check_allocations); + log_fsck_err(trans, bucket_metadata_type_mismatch, "bucket %llu:%llu gen %u different types of data in same bucket: %s, %s\n" "while marking %s", iter.pos.inode, iter.pos.offset, a->v.gen, @@ -979,6 +976,7 @@ static int __bch2_trans_mark_metadata_bucket(struct btree_trans *trans, ret = bch2_trans_update(trans, &iter, &a->k_i, 0); } err: +fsck_err: bch2_trans_iter_exit(trans, &iter); return ret; } @@ -990,11 +988,10 @@ static int bch2_mark_metadata_bucket(struct btree_trans *trans, struct bch_dev * struct bch_fs *c = trans->c; int ret = 0; - percpu_down_read(&c->mark_lock); struct bucket *g = gc_bucket(ca, b); if (bch2_fs_inconsistent_on(!g, c, "reference to invalid bucket on device %u when marking metadata type %s", ca->dev_idx, bch2_data_type_str(data_type))) - goto err_unlock; + goto err; bucket_lock(g); struct bch_alloc_v4 old = bucket_m_to_alloc(*g); @@ -1004,26 +1001,24 @@ static int bch2_mark_metadata_bucket(struct btree_trans *trans, struct bch_dev * "different types of data in same bucket: %s, %s", bch2_data_type_str(g->data_type), bch2_data_type_str(data_type))) - goto err; + goto err_unlock; if (bch2_fs_inconsistent_on((u64) g->dirty_sectors + sectors > ca->mi.bucket_size, c, "bucket %u:%llu gen %u data type %s sector count overflow: %u + %u > bucket size", ca->dev_idx, b, g->gen, bch2_data_type_str(g->data_type ?: data_type), g->dirty_sectors, sectors)) - goto err; + goto err_unlock; g->data_type = data_type; g->dirty_sectors += sectors; struct bch_alloc_v4 new = bucket_m_to_alloc(*g); bucket_unlock(g); - percpu_up_read(&c->mark_lock); ret = bch2_alloc_key_to_dev_counters(trans, ca, &old, &new, flags); return ret; -err: - bucket_unlock(g); err_unlock: - percpu_up_read(&c->mark_lock); + bucket_unlock(g); +err: return -BCH_ERR_metadata_bucket_inconsistency; } @@ -1155,6 +1150,31 @@ int bch2_trans_mark_dev_sbs(struct bch_fs *c) return bch2_trans_mark_dev_sbs_flags(c, BTREE_TRIGGER_transactional); } +bool bch2_is_superblock_bucket(struct bch_dev *ca, u64 b) +{ + struct bch_sb_layout *layout = &ca->disk_sb.sb->layout; + u64 b_offset = bucket_to_sector(ca, b); + u64 b_end = bucket_to_sector(ca, b + 1); + unsigned i; + + if (!b) + return true; + + for (i = 0; i < layout->nr_superblocks; i++) { + u64 offset = le64_to_cpu(layout->sb_offset[i]); + u64 end = offset + (1 << layout->sb_max_size_bits); + + if (!(offset >= b_end || end <= b_offset)) + return true; + } + + for (i = 0; i < ca->journal.nr; i++) + if (b == ca->journal.buckets[i]) + return true; + + return false; +} + /* Disk reservations: */ #define SECTORS_CACHE 1024 @@ -1238,7 +1258,7 @@ int bch2_buckets_nouse_alloc(struct bch_fs *c) for_each_member_device(c, ca) { BUG_ON(ca->buckets_nouse); - ca->buckets_nouse = kvmalloc(BITS_TO_LONGS(ca->mi.nbuckets) * + ca->buckets_nouse = bch2_kvmalloc(BITS_TO_LONGS(ca->mi.nbuckets) * sizeof(unsigned long), GFP_KERNEL|__GFP_ZERO); if (!ca->buckets_nouse) { @@ -1264,10 +1284,15 @@ int bch2_dev_buckets_resize(struct bch_fs *c, struct bch_dev *ca, u64 nbuckets) bool resize = ca->bucket_gens != NULL; int ret; - BUG_ON(resize && ca->buckets_nouse); + if (resize) + lockdep_assert_held(&c->state_lock); + + if (resize && ca->buckets_nouse) + return -BCH_ERR_no_resize_with_buckets_nouse; - if (!(bucket_gens = kvmalloc(sizeof(struct bucket_gens) + nbuckets, - GFP_KERNEL|__GFP_ZERO))) { + bucket_gens = bch2_kvmalloc(struct_size(bucket_gens, b, nbuckets), + GFP_KERNEL|__GFP_ZERO); + if (!bucket_gens) { ret = -BCH_ERR_ENOMEM_bucket_gens; goto err; } @@ -1277,19 +1302,16 @@ int bch2_dev_buckets_resize(struct bch_fs *c, struct bch_dev *ca, u64 nbuckets) bucket_gens->nbuckets_minus_first = bucket_gens->nbuckets - bucket_gens->first_bucket; - if (resize) { - down_write(&ca->bucket_lock); - percpu_down_write(&c->mark_lock); - } - old_bucket_gens = rcu_dereference_protected(ca->bucket_gens, 1); if (resize) { - size_t n = min(bucket_gens->nbuckets, old_bucket_gens->nbuckets); - + bucket_gens->nbuckets = min(bucket_gens->nbuckets, + old_bucket_gens->nbuckets); + bucket_gens->nbuckets_minus_first = + bucket_gens->nbuckets - bucket_gens->first_bucket; memcpy(bucket_gens->b, old_bucket_gens->b, - n); + bucket_gens->nbuckets); } rcu_assign_pointer(ca->bucket_gens, bucket_gens); @@ -1297,11 +1319,6 @@ int bch2_dev_buckets_resize(struct bch_fs *c, struct bch_dev *ca, u64 nbuckets) nbuckets = ca->mi.nbuckets; - if (resize) { - percpu_up_write(&c->mark_lock); - up_write(&ca->bucket_lock); - } - ret = 0; err: if (bucket_gens) diff --git a/fs/bcachefs/buckets.h b/fs/bcachefs/buckets.h index ccc78bfe2fd4..a9acdd6c0c86 100644 --- a/fs/bcachefs/buckets.h +++ b/fs/bcachefs/buckets.h @@ -82,16 +82,15 @@ static inline void bucket_lock(struct bucket *b) static inline struct bucket *gc_bucket(struct bch_dev *ca, size_t b) { - return genradix_ptr(&ca->buckets_gc, b); + return bucket_valid(ca, b) + ? genradix_ptr(&ca->buckets_gc, b) + : NULL; } static inline struct bucket_gens *bucket_gens(struct bch_dev *ca) { return rcu_dereference_check(ca->bucket_gens, - !ca->fs || - percpu_rwsem_is_held(&ca->fs->mark_lock) || - lockdep_is_held(&ca->fs->state_lock) || - lockdep_is_held(&ca->bucket_lock)); + lockdep_is_held(&ca->fs->state_lock)); } static inline u8 *bucket_gen(struct bch_dev *ca, size_t b) @@ -308,26 +307,7 @@ int bch2_trans_mark_dev_sbs_flags(struct bch_fs *, enum btree_iter_update_trigger_flags); int bch2_trans_mark_dev_sbs(struct bch_fs *); -static inline bool is_superblock_bucket(struct bch_dev *ca, u64 b) -{ - struct bch_sb_layout *layout = &ca->disk_sb.sb->layout; - u64 b_offset = bucket_to_sector(ca, b); - u64 b_end = bucket_to_sector(ca, b + 1); - unsigned i; - - if (!b) - return true; - - for (i = 0; i < layout->nr_superblocks; i++) { - u64 offset = le64_to_cpu(layout->sb_offset[i]); - u64 end = offset + (1 << layout->sb_max_size_bits); - - if (!(offset >= b_end || end <= b_offset)) - return true; - } - - return false; -} +bool bch2_is_superblock_bucket(struct bch_dev *, u64); static inline const char *bch2_data_type_str(enum bch_data_type type) { diff --git a/fs/bcachefs/buckets_types.h b/fs/bcachefs/buckets_types.h index 28bd09a253c8..7174047b8e92 100644 --- a/fs/bcachefs/buckets_types.h +++ b/fs/bcachefs/buckets_types.h @@ -24,7 +24,7 @@ struct bucket_gens { u16 first_bucket; size_t nbuckets; size_t nbuckets_minus_first; - u8 b[]; + u8 b[] __counted_by(nbuckets); }; struct bch_dev_usage { diff --git a/fs/bcachefs/chardev.c b/fs/bcachefs/chardev.c index 2182b555c112..46e9e32105a9 100644 --- a/fs/bcachefs/chardev.c +++ b/fs/bcachefs/chardev.c @@ -6,11 +6,11 @@ #include "buckets.h" #include "chardev.h" #include "disk_accounting.h" +#include "fsck.h" #include "journal.h" #include "move.h" #include "recovery_passes.h" #include "replicas.h" -#include "super.h" #include "super-io.h" #include "thread_with_file.h" @@ -127,130 +127,6 @@ static long bch2_ioctl_incremental(struct bch_ioctl_incremental __user *user_arg } #endif -struct fsck_thread { - struct thread_with_stdio thr; - struct bch_fs *c; - struct bch_opts opts; -}; - -static void bch2_fsck_thread_exit(struct thread_with_stdio *_thr) -{ - struct fsck_thread *thr = container_of(_thr, struct fsck_thread, thr); - kfree(thr); -} - -static int bch2_fsck_offline_thread_fn(struct thread_with_stdio *stdio) -{ - struct fsck_thread *thr = container_of(stdio, struct fsck_thread, thr); - struct bch_fs *c = thr->c; - - int ret = PTR_ERR_OR_ZERO(c); - if (ret) - return ret; - - ret = bch2_fs_start(thr->c); - if (ret) - goto err; - - if (test_bit(BCH_FS_errors_fixed, &c->flags)) { - bch2_stdio_redirect_printf(&stdio->stdio, false, "%s: errors fixed\n", c->name); - ret |= 1; - } - if (test_bit(BCH_FS_error, &c->flags)) { - bch2_stdio_redirect_printf(&stdio->stdio, false, "%s: still has errors\n", c->name); - ret |= 4; - } -err: - bch2_fs_stop(c); - return ret; -} - -static const struct thread_with_stdio_ops bch2_offline_fsck_ops = { - .exit = bch2_fsck_thread_exit, - .fn = bch2_fsck_offline_thread_fn, -}; - -static long bch2_ioctl_fsck_offline(struct bch_ioctl_fsck_offline __user *user_arg) -{ - struct bch_ioctl_fsck_offline arg; - struct fsck_thread *thr = NULL; - darray_str(devs) = {}; - long ret = 0; - - if (copy_from_user(&arg, user_arg, sizeof(arg))) - return -EFAULT; - - if (arg.flags) - return -EINVAL; - - if (!capable(CAP_SYS_ADMIN)) - return -EPERM; - - for (size_t i = 0; i < arg.nr_devs; i++) { - u64 dev_u64; - ret = copy_from_user_errcode(&dev_u64, &user_arg->devs[i], sizeof(u64)); - if (ret) - goto err; - - char *dev_str = strndup_user((char __user *)(unsigned long) dev_u64, PATH_MAX); - ret = PTR_ERR_OR_ZERO(dev_str); - if (ret) - goto err; - - ret = darray_push(&devs, dev_str); - if (ret) { - kfree(dev_str); - goto err; - } - } - - thr = kzalloc(sizeof(*thr), GFP_KERNEL); - if (!thr) { - ret = -ENOMEM; - goto err; - } - - thr->opts = bch2_opts_empty(); - - if (arg.opts) { - char *optstr = strndup_user((char __user *)(unsigned long) arg.opts, 1 << 16); - ret = PTR_ERR_OR_ZERO(optstr) ?: - bch2_parse_mount_opts(NULL, &thr->opts, NULL, optstr); - if (!IS_ERR(optstr)) - kfree(optstr); - - if (ret) - goto err; - } - - opt_set(thr->opts, stdio, (u64)(unsigned long)&thr->thr.stdio); - opt_set(thr->opts, read_only, 1); - opt_set(thr->opts, ratelimit_errors, 0); - - /* We need request_key() to be called before we punt to kthread: */ - opt_set(thr->opts, nostart, true); - - bch2_thread_with_stdio_init(&thr->thr, &bch2_offline_fsck_ops); - - thr->c = bch2_fs_open(devs.data, arg.nr_devs, thr->opts); - - if (!IS_ERR(thr->c) && - thr->c->opts.errors == BCH_ON_ERROR_panic) - thr->c->opts.errors = BCH_ON_ERROR_ro; - - ret = __bch2_run_thread_with_stdio(&thr->thr); -out: - darray_for_each(devs, i) - kfree(*i); - darray_exit(&devs); - return ret; -err: - if (thr) - bch2_fsck_thread_exit(&thr->thr); - pr_err("ret %s", bch2_err_str(ret)); - goto out; -} - static long bch2_global_ioctl(unsigned cmd, void __user *arg) { long ret; @@ -775,99 +651,6 @@ static long bch2_ioctl_disk_resize_journal(struct bch_fs *c, return ret; } -static int bch2_fsck_online_thread_fn(struct thread_with_stdio *stdio) -{ - struct fsck_thread *thr = container_of(stdio, struct fsck_thread, thr); - struct bch_fs *c = thr->c; - - c->stdio_filter = current; - c->stdio = &thr->thr.stdio; - - /* - * XXX: can we figure out a way to do this without mucking with c->opts? - */ - unsigned old_fix_errors = c->opts.fix_errors; - if (opt_defined(thr->opts, fix_errors)) - c->opts.fix_errors = thr->opts.fix_errors; - else - c->opts.fix_errors = FSCK_FIX_ask; - - c->opts.fsck = true; - set_bit(BCH_FS_fsck_running, &c->flags); - - c->curr_recovery_pass = BCH_RECOVERY_PASS_check_alloc_info; - int ret = bch2_run_online_recovery_passes(c); - - clear_bit(BCH_FS_fsck_running, &c->flags); - bch_err_fn(c, ret); - - c->stdio = NULL; - c->stdio_filter = NULL; - c->opts.fix_errors = old_fix_errors; - - up(&c->online_fsck_mutex); - bch2_ro_ref_put(c); - return ret; -} - -static const struct thread_with_stdio_ops bch2_online_fsck_ops = { - .exit = bch2_fsck_thread_exit, - .fn = bch2_fsck_online_thread_fn, -}; - -static long bch2_ioctl_fsck_online(struct bch_fs *c, - struct bch_ioctl_fsck_online arg) -{ - struct fsck_thread *thr = NULL; - long ret = 0; - - if (arg.flags) - return -EINVAL; - - if (!capable(CAP_SYS_ADMIN)) - return -EPERM; - - if (!bch2_ro_ref_tryget(c)) - return -EROFS; - - if (down_trylock(&c->online_fsck_mutex)) { - bch2_ro_ref_put(c); - return -EAGAIN; - } - - thr = kzalloc(sizeof(*thr), GFP_KERNEL); - if (!thr) { - ret = -ENOMEM; - goto err; - } - - thr->c = c; - thr->opts = bch2_opts_empty(); - - if (arg.opts) { - char *optstr = strndup_user((char __user *)(unsigned long) arg.opts, 1 << 16); - - ret = PTR_ERR_OR_ZERO(optstr) ?: - bch2_parse_mount_opts(c, &thr->opts, NULL, optstr); - if (!IS_ERR(optstr)) - kfree(optstr); - - if (ret) - goto err; - } - - ret = bch2_run_thread_with_stdio(&thr->thr, &bch2_online_fsck_ops); -err: - if (ret < 0) { - bch_err_fn(c, ret); - if (thr) - bch2_fsck_thread_exit(&thr->thr); - up(&c->online_fsck_mutex); - bch2_ro_ref_put(c); - } - return ret; -} - #define BCH_IOCTL(_name, _argtype) \ do { \ _argtype i; \ diff --git a/fs/bcachefs/checksum.c b/fs/bcachefs/checksum.c index ce8fc677bef9..23a383577d4c 100644 --- a/fs/bcachefs/checksum.c +++ b/fs/bcachefs/checksum.c @@ -2,6 +2,7 @@ #include "bcachefs.h" #include "checksum.h" #include "errcode.h" +#include "error.h" #include "super.h" #include "super-io.h" @@ -252,6 +253,10 @@ int bch2_encrypt(struct bch_fs *c, unsigned type, if (!bch2_csum_type_is_encryption(type)) return 0; + if (bch2_fs_inconsistent_on(!c->chacha20, + c, "attempting to encrypt without encryption key")) + return -BCH_ERR_no_encryption_key; + return do_encrypt(c->chacha20, nonce, data, len); } @@ -337,8 +342,9 @@ int __bch2_encrypt_bio(struct bch_fs *c, unsigned type, size_t sgl_len = 0; int ret = 0; - if (!bch2_csum_type_is_encryption(type)) - return 0; + if (bch2_fs_inconsistent_on(!c->chacha20, + c, "attempting to encrypt without encryption key")) + return -BCH_ERR_no_encryption_key; darray_init(&sgl); diff --git a/fs/bcachefs/checksum.h b/fs/bcachefs/checksum.h index e40499fde9a4..43b9d71f2f2b 100644 --- a/fs/bcachefs/checksum.h +++ b/fs/bcachefs/checksum.h @@ -109,7 +109,7 @@ int bch2_enable_encryption(struct bch_fs *, bool); void bch2_fs_encryption_exit(struct bch_fs *); int bch2_fs_encryption_init(struct bch_fs *); -static inline enum bch_csum_type bch2_csum_opt_to_type(enum bch_csum_opts type, +static inline enum bch_csum_type bch2_csum_opt_to_type(enum bch_csum_opt type, bool data) { switch (type) { diff --git a/fs/bcachefs/compress.c b/fs/bcachefs/compress.c index 1410365a8891..f99ff1819597 100644 --- a/fs/bcachefs/compress.c +++ b/fs/bcachefs/compress.c @@ -2,13 +2,33 @@ #include "bcachefs.h" #include "checksum.h" #include "compress.h" +#include "error.h" #include "extents.h" +#include "opts.h" #include "super-io.h" #include <linux/lz4.h> #include <linux/zlib.h> #include <linux/zstd.h> +static inline enum bch_compression_opts bch2_compression_type_to_opt(enum bch_compression_type type) +{ + switch (type) { + case BCH_COMPRESSION_TYPE_none: + case BCH_COMPRESSION_TYPE_incompressible: + return BCH_COMPRESSION_OPT_none; + case BCH_COMPRESSION_TYPE_lz4_old: + case BCH_COMPRESSION_TYPE_lz4: + return BCH_COMPRESSION_OPT_lz4; + case BCH_COMPRESSION_TYPE_gzip: + return BCH_COMPRESSION_OPT_gzip; + case BCH_COMPRESSION_TYPE_zstd: + return BCH_COMPRESSION_OPT_zstd; + default: + BUG(); + } +} + /* Bounce buffer: */ struct bbuf { void *b; @@ -158,6 +178,19 @@ static int __bio_uncompress(struct bch_fs *c, struct bio *src, void *workspace; int ret; + enum bch_compression_opts opt = bch2_compression_type_to_opt(crc.compression_type); + mempool_t *workspace_pool = &c->compress_workspace[opt]; + if (unlikely(!mempool_initialized(workspace_pool))) { + if (fsck_err(c, compression_type_not_marked_in_sb, + "compression type %s set but not marked in superblock", + __bch2_compression_types[crc.compression_type])) + ret = bch2_check_set_has_compressed_data(c, opt); + else + ret = -BCH_ERR_compression_workspace_not_initialized; + if (ret) + goto out; + } + src_data = bio_map_or_bounce(c, src, READ); switch (crc.compression_type) { @@ -176,13 +209,13 @@ static int __bio_uncompress(struct bch_fs *c, struct bio *src, .avail_out = dst_len, }; - workspace = mempool_alloc(&c->decompress_workspace, GFP_NOFS); + workspace = mempool_alloc(workspace_pool, GFP_NOFS); zlib_set_workspace(&strm, workspace); zlib_inflateInit2(&strm, -MAX_WBITS); ret = zlib_inflate(&strm, Z_FINISH); - mempool_free(workspace, &c->decompress_workspace); + mempool_free(workspace, workspace_pool); if (ret != Z_STREAM_END) goto err; @@ -195,14 +228,14 @@ static int __bio_uncompress(struct bch_fs *c, struct bio *src, if (real_src_len > src_len - 4) goto err; - workspace = mempool_alloc(&c->decompress_workspace, GFP_NOFS); + workspace = mempool_alloc(workspace_pool, GFP_NOFS); ctx = zstd_init_dctx(workspace, zstd_dctx_workspace_bound()); ret = zstd_decompress_dctx(ctx, dst_data, dst_len, src_data.b + 4, real_src_len); - mempool_free(workspace, &c->decompress_workspace); + mempool_free(workspace, workspace_pool); if (ret != dst_len) goto err; @@ -212,6 +245,7 @@ static int __bio_uncompress(struct bch_fs *c, struct bio *src, BUG(); } ret = 0; +fsck_err: out: bio_unmap_or_unbounce(c, src_data); return ret; @@ -394,8 +428,21 @@ static unsigned __bio_compress(struct bch_fs *c, unsigned pad; int ret = 0; - BUG_ON(compression_type >= BCH_COMPRESSION_TYPE_NR); - BUG_ON(!mempool_initialized(&c->compress_workspace[compression_type])); + /* bch2_compression_decode catches unknown compression types: */ + BUG_ON(compression.type >= BCH_COMPRESSION_OPT_NR); + + mempool_t *workspace_pool = &c->compress_workspace[compression.type]; + if (unlikely(!mempool_initialized(workspace_pool))) { + if (fsck_err(c, compression_opt_not_marked_in_sb, + "compression opt %s set but not marked in superblock", + bch2_compression_opts[compression.type])) { + ret = bch2_check_set_has_compressed_data(c, compression.type); + if (ret) /* memory allocation failure, don't compress */ + return 0; + } else { + return 0; + } + } /* If it's only one block, don't bother trying to compress: */ if (src->bi_iter.bi_size <= c->opts.block_size) @@ -404,7 +451,7 @@ static unsigned __bio_compress(struct bch_fs *c, dst_data = bio_map_or_bounce(c, dst, WRITE); src_data = bio_map_or_bounce(c, src, READ); - workspace = mempool_alloc(&c->compress_workspace[compression_type], GFP_NOFS); + workspace = mempool_alloc(workspace_pool, GFP_NOFS); *src_len = src->bi_iter.bi_size; *dst_len = dst->bi_iter.bi_size; @@ -447,7 +494,7 @@ static unsigned __bio_compress(struct bch_fs *c, *src_len = round_down(*src_len, block_bytes(c)); } - mempool_free(workspace, &c->compress_workspace[compression_type]); + mempool_free(workspace, workspace_pool); if (ret) goto err; @@ -477,6 +524,9 @@ out: err: ret = BCH_COMPRESSION_TYPE_incompressible; goto out; +fsck_err: + ret = 0; + goto out; } unsigned bch2_bio_compress(struct bch_fs *c, @@ -559,7 +609,6 @@ void bch2_fs_compress_exit(struct bch_fs *c) { unsigned i; - mempool_exit(&c->decompress_workspace); for (i = 0; i < ARRAY_SIZE(c->compress_workspace); i++) mempool_exit(&c->compress_workspace[i]); mempool_exit(&c->compression_bounce[WRITE]); @@ -568,7 +617,6 @@ void bch2_fs_compress_exit(struct bch_fs *c) static int __bch2_fs_compress_init(struct bch_fs *c, u64 features) { - size_t decompress_workspace_size = 0; ZSTD_parameters params = zstd_get_params(zstd_max_clevel(), c->opts.encoded_extent_max); @@ -576,19 +624,17 @@ static int __bch2_fs_compress_init(struct bch_fs *c, u64 features) struct { unsigned feature; - enum bch_compression_type type; + enum bch_compression_opts type; size_t compress_workspace; - size_t decompress_workspace; } compression_types[] = { - { BCH_FEATURE_lz4, BCH_COMPRESSION_TYPE_lz4, - max_t(size_t, LZ4_MEM_COMPRESS, LZ4HC_MEM_COMPRESS), - 0 }, - { BCH_FEATURE_gzip, BCH_COMPRESSION_TYPE_gzip, - zlib_deflate_workspacesize(MAX_WBITS, DEF_MEM_LEVEL), - zlib_inflate_workspacesize(), }, - { BCH_FEATURE_zstd, BCH_COMPRESSION_TYPE_zstd, - c->zstd_workspace_size, - zstd_dctx_workspace_bound() }, + { BCH_FEATURE_lz4, BCH_COMPRESSION_OPT_lz4, + max_t(size_t, LZ4_MEM_COMPRESS, LZ4HC_MEM_COMPRESS) }, + { BCH_FEATURE_gzip, BCH_COMPRESSION_OPT_gzip, + max(zlib_deflate_workspacesize(MAX_WBITS, DEF_MEM_LEVEL), + zlib_inflate_workspacesize()) }, + { BCH_FEATURE_zstd, BCH_COMPRESSION_OPT_zstd, + max(c->zstd_workspace_size, + zstd_dctx_workspace_bound()) }, }, *i; bool have_compressed = false; @@ -613,9 +659,6 @@ static int __bch2_fs_compress_init(struct bch_fs *c, u64 features) for (i = compression_types; i < compression_types + ARRAY_SIZE(compression_types); i++) { - decompress_workspace_size = - max(decompress_workspace_size, i->decompress_workspace); - if (!(features & (1 << i->feature))) continue; @@ -628,11 +671,6 @@ static int __bch2_fs_compress_init(struct bch_fs *c, u64 features) return -BCH_ERR_ENOMEM_compression_workspace_init; } - if (!mempool_initialized(&c->decompress_workspace) && - mempool_init_kvmalloc_pool(&c->decompress_workspace, - 1, decompress_workspace_size)) - return -BCH_ERR_ENOMEM_decompression_workspace_init; - return 0; } diff --git a/fs/bcachefs/darray.h b/fs/bcachefs/darray.h index 8f4c3f0665c4..c6151495985f 100644 --- a/fs/bcachefs/darray.h +++ b/fs/bcachefs/darray.h @@ -83,7 +83,7 @@ int __bch2_darray_resize_noprof(darray_char *, size_t, size_t, gfp_t); for (typeof(&(_d).data[0]) _i = (_d).data; _i < (_d).data + (_d).nr; _i++) #define darray_for_each_reverse(_d, _i) \ - for (typeof(&(_d).data[0]) _i = (_d).data + (_d).nr - 1; _i >= (_d).data; --_i) + for (typeof(&(_d).data[0]) _i = (_d).data + (_d).nr - 1; _i >= (_d).data && (_d).nr; --_i) #define darray_init(_d) \ do { \ diff --git a/fs/bcachefs/data_update.c b/fs/bcachefs/data_update.c index 8e75a852b358..585214931e05 100644 --- a/fs/bcachefs/data_update.c +++ b/fs/bcachefs/data_update.c @@ -110,11 +110,8 @@ static void trace_move_extent_fail2(struct data_update *m, { struct bch_fs *c = m->op.c; struct bkey_s_c old = bkey_i_to_s_c(m->k.k); - const union bch_extent_entry *entry; - struct bch_extent_ptr *ptr; - struct extent_ptr_decoded p; struct printbuf buf = PRINTBUF; - unsigned i, rewrites_found = 0; + unsigned rewrites_found = 0; if (!trace_move_extent_fail_enabled()) return; @@ -122,27 +119,25 @@ static void trace_move_extent_fail2(struct data_update *m, prt_str(&buf, msg); if (insert) { - i = 0; + const union bch_extent_entry *entry; + struct bch_extent_ptr *ptr; + struct extent_ptr_decoded p; + + unsigned ptr_bit = 1; bkey_for_each_ptr_decode(old.k, bch2_bkey_ptrs_c(old), p, entry) { - if (((1U << i) & m->data_opts.rewrite_ptrs) && + if ((ptr_bit & m->data_opts.rewrite_ptrs) && (ptr = bch2_extent_has_ptr(old, p, bkey_i_to_s(insert))) && !ptr->cached) - rewrites_found |= 1U << i; - i++; + rewrites_found |= ptr_bit; + ptr_bit <<= 1; } } - prt_printf(&buf, "\nrewrite ptrs: %u%u%u%u", - (m->data_opts.rewrite_ptrs & (1 << 0)) != 0, - (m->data_opts.rewrite_ptrs & (1 << 1)) != 0, - (m->data_opts.rewrite_ptrs & (1 << 2)) != 0, - (m->data_opts.rewrite_ptrs & (1 << 3)) != 0); + prt_str(&buf, "rewrites found:\t"); + bch2_prt_u64_base2(&buf, rewrites_found); + prt_newline(&buf); - prt_printf(&buf, "\nrewrites found: %u%u%u%u", - (rewrites_found & (1 << 0)) != 0, - (rewrites_found & (1 << 1)) != 0, - (rewrites_found & (1 << 2)) != 0, - (rewrites_found & (1 << 3)) != 0); + bch2_data_update_opts_to_text(&buf, c, &m->op.opts, &m->data_opts); prt_str(&buf, "\nold: "); bch2_bkey_val_to_text(&buf, c, old); @@ -194,7 +189,7 @@ static int __bch2_data_update_index_update(struct btree_trans *trans, struct bpos next_pos; bool should_check_enospc; s64 i_sectors_delta = 0, disk_sectors_delta = 0; - unsigned rewrites_found = 0, durability, i; + unsigned rewrites_found = 0, durability, ptr_bit; bch2_trans_begin(trans); @@ -231,16 +226,16 @@ static int __bch2_data_update_index_update(struct btree_trans *trans, * * Fist, drop rewrite_ptrs from @new: */ - i = 0; + ptr_bit = 1; bkey_for_each_ptr_decode(old.k, bch2_bkey_ptrs_c(old), p, entry_c) { - if (((1U << i) & m->data_opts.rewrite_ptrs) && + if ((ptr_bit & m->data_opts.rewrite_ptrs) && (ptr = bch2_extent_has_ptr(old, p, bkey_i_to_s(insert))) && !ptr->cached) { bch2_extent_ptr_set_cached(c, &m->op.opts, bkey_i_to_s(insert), ptr); - rewrites_found |= 1U << i; + rewrites_found |= ptr_bit; } - i++; + ptr_bit <<= 1; } if (m->data_opts.rewrite_ptrs && @@ -323,8 +318,11 @@ restart_drop_extra_replicas: * it's been hard to reproduce, so this should give us some more * information when it does occur: */ - int invalid = bch2_bkey_validate(c, bkey_i_to_s_c(insert), __btree_node_type(0, m->btree_id), - BCH_VALIDATE_commit); + int invalid = bch2_bkey_validate(c, bkey_i_to_s_c(insert), + (struct bkey_validate_context) { + .btree = m->btree_id, + .flags = BCH_VALIDATE_commit, + }); if (invalid) { struct printbuf buf = PRINTBUF; @@ -362,7 +360,7 @@ restart_drop_extra_replicas: k.k->p, bkey_start_pos(&insert->k)) ?: bch2_insert_snapshot_whiteouts(trans, m->btree_id, k.k->p, insert->k.p) ?: - bch2_bkey_set_needs_rebalance(c, insert, &op->opts) ?: + bch2_bkey_set_needs_rebalance(c, &op->opts, insert) ?: bch2_trans_update(trans, &iter, insert, BTREE_UPDATE_internal_snapshot_node) ?: bch2_trans_commit(trans, &op->res, @@ -540,7 +538,7 @@ void bch2_data_update_opts_to_text(struct printbuf *out, struct bch_fs *c, prt_newline(out); prt_str(out, "compression:\t"); - bch2_compression_opt_to_text(out, background_compression(*io_opts)); + bch2_compression_opt_to_text(out, io_opts->background_compression); prt_newline(out); prt_str(out, "opts.replicas:\t"); @@ -614,7 +612,7 @@ int bch2_data_update_init(struct btree_trans *trans, struct bkey_ptrs_c ptrs = bch2_bkey_ptrs_c(k); const union bch_extent_entry *entry; struct extent_ptr_decoded p; - unsigned i, reserve_sectors = k.k->size * data_opts.extra_replicas; + unsigned reserve_sectors = k.k->size * data_opts.extra_replicas; int ret = 0; /* @@ -622,7 +620,7 @@ int bch2_data_update_init(struct btree_trans *trans, * and we have to check for this because we go rw before repairing the * snapshots table - just skip it, we can move it later. */ - if (unlikely(k.k->p.snapshot && !bch2_snapshot_equiv(c, k.k->p.snapshot))) + if (unlikely(k.k->p.snapshot && !bch2_snapshot_exists(c, k.k->p.snapshot))) return -BCH_ERR_data_update_done; if (!bkey_get_dev_refs(c, k)) @@ -652,22 +650,22 @@ int bch2_data_update_init(struct btree_trans *trans, BCH_WRITE_DATA_ENCODED| BCH_WRITE_MOVE| m->data_opts.write_flags; - m->op.compression_opt = background_compression(io_opts); + m->op.compression_opt = io_opts.background_compression; m->op.watermark = m->data_opts.btree_insert_flags & BCH_WATERMARK_MASK; unsigned durability_have = 0, durability_removing = 0; - i = 0; + unsigned ptr_bit = 1; bkey_for_each_ptr_decode(k.k, ptrs, p, entry) { if (!p.ptr.cached) { rcu_read_lock(); - if (BIT(i) & m->data_opts.rewrite_ptrs) { + if (ptr_bit & m->data_opts.rewrite_ptrs) { if (crc_is_compressed(p.crc)) reserve_sectors += k.k->size; m->op.nr_replicas += bch2_extent_ptr_desired_durability(c, &p); durability_removing += bch2_extent_ptr_desired_durability(c, &p); - } else if (!(BIT(i) & m->data_opts.kill_ptrs)) { + } else if (!(ptr_bit & m->data_opts.kill_ptrs)) { bch2_dev_list_add_dev(&m->op.devs_have, p.ptr.dev); durability_have += bch2_extent_ptr_durability(c, &p); } @@ -687,7 +685,7 @@ int bch2_data_update_init(struct btree_trans *trans, if (p.crc.compression_type == BCH_COMPRESSION_TYPE_incompressible) m->op.incompressible = true; - i++; + ptr_bit <<= 1; } unsigned durability_required = max(0, (int) (io_opts.data_replicas - durability_have)); @@ -750,14 +748,14 @@ out: void bch2_data_update_opts_normalize(struct bkey_s_c k, struct data_update_opts *opts) { struct bkey_ptrs_c ptrs = bch2_bkey_ptrs_c(k); - unsigned i = 0; + unsigned ptr_bit = 1; bkey_for_each_ptr(ptrs, ptr) { - if ((opts->rewrite_ptrs & (1U << i)) && ptr->cached) { - opts->kill_ptrs |= 1U << i; - opts->rewrite_ptrs ^= 1U << i; + if ((opts->rewrite_ptrs & ptr_bit) && ptr->cached) { + opts->kill_ptrs |= ptr_bit; + opts->rewrite_ptrs ^= ptr_bit; } - i++; + ptr_bit <<= 1; } } diff --git a/fs/bcachefs/debug.c b/fs/bcachefs/debug.c index 45aec1afdb0e..b5de52a50d10 100644 --- a/fs/bcachefs/debug.c +++ b/fs/bcachefs/debug.c @@ -472,7 +472,9 @@ static void bch2_cached_btree_node_to_text(struct printbuf *out, struct bch_fs * if (!out->nr_tabstops) printbuf_tabstop_push(out, 32); - prt_printf(out, "%px btree=%s l=%u\n", b, bch2_btree_id_str(b->c.btree_id), b->c.level); + prt_printf(out, "%px ", b); + bch2_btree_id_level_to_text(out, b->c.btree_id, b->c.level); + prt_printf(out, "\n"); printbuf_indent_add(out, 2); diff --git a/fs/bcachefs/dirent.c b/fs/bcachefs/dirent.c index faffc98d5605..600eee936f13 100644 --- a/fs/bcachefs/dirent.c +++ b/fs/bcachefs/dirent.c @@ -101,7 +101,7 @@ const struct bch_hash_desc bch2_dirent_hash_desc = { }; int bch2_dirent_validate(struct bch_fs *c, struct bkey_s_c k, - enum bch_validate_flags flags) + struct bkey_validate_context from) { struct bkey_s_c_dirent d = bkey_s_c_to_dirent(k); struct qstr d_name = bch2_dirent_get_name(d); @@ -120,7 +120,7 @@ int bch2_dirent_validate(struct bch_fs *c, struct bkey_s_c k, * Check new keys don't exceed the max length * (older keys may be larger.) */ - bkey_fsck_err_on((flags & BCH_VALIDATE_commit) && d_name.len > BCH_NAME_MAX, + bkey_fsck_err_on((from.flags & BCH_VALIDATE_commit) && d_name.len > BCH_NAME_MAX, c, dirent_name_too_long, "dirent name too big (%u > %u)", d_name.len, BCH_NAME_MAX); @@ -266,7 +266,7 @@ int bch2_dirent_read_target(struct btree_trans *trans, subvol_inum dir, } else { target->subvol = le32_to_cpu(d.v->d_child_subvol); - ret = bch2_subvolume_get(trans, target->subvol, true, BTREE_ITER_cached, &s); + ret = bch2_subvolume_get(trans, target->subvol, true, &s); target->inum = le64_to_cpu(s.inode); } @@ -500,7 +500,7 @@ int bch2_empty_dir_snapshot(struct btree_trans *trans, u64 dir, u32 subvol, u32 struct bkey_s_c k; int ret; - for_each_btree_key_upto_norestart(trans, iter, BTREE_ID_dirents, + for_each_btree_key_max_norestart(trans, iter, BTREE_ID_dirents, SPOS(dir, 0, snapshot), POS(dir, U64_MAX), 0, k, ret) if (k.k->type == KEY_TYPE_dirent) { @@ -549,7 +549,7 @@ int bch2_readdir(struct bch_fs *c, subvol_inum inum, struct dir_context *ctx) bch2_bkey_buf_init(&sk); int ret = bch2_trans_run(c, - for_each_btree_key_in_subvolume_upto(trans, iter, BTREE_ID_dirents, + for_each_btree_key_in_subvolume_max(trans, iter, BTREE_ID_dirents, POS(inum.inum, ctx->pos), POS(inum.inum, U64_MAX), inum.subvol, 0, k, ({ diff --git a/fs/bcachefs/dirent.h b/fs/bcachefs/dirent.h index 53ad99666022..a633f83c1ac7 100644 --- a/fs/bcachefs/dirent.h +++ b/fs/bcachefs/dirent.h @@ -4,10 +4,10 @@ #include "str_hash.h" -enum bch_validate_flags; extern const struct bch_hash_desc bch2_dirent_hash_desc; -int bch2_dirent_validate(struct bch_fs *, struct bkey_s_c, enum bch_validate_flags); +int bch2_dirent_validate(struct bch_fs *, struct bkey_s_c, + struct bkey_validate_context); void bch2_dirent_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c); #define bch2_bkey_ops_dirent ((struct bkey_ops) { \ @@ -31,6 +31,11 @@ static inline unsigned dirent_val_u64s(unsigned len) sizeof(u64)); } +static inline unsigned int dirent_occupied_size(const struct qstr *name) +{ + return (BKEY_U64s + dirent_val_u64s(name->len)) * sizeof(u64); +} + int bch2_dirent_read_target(struct btree_trans *, subvol_inum, struct bkey_s_c_dirent, subvol_inum *); diff --git a/fs/bcachefs/disk_accounting.c b/fs/bcachefs/disk_accounting.c index 07eb8fa1b026..b32e91ba8be8 100644 --- a/fs/bcachefs/disk_accounting.c +++ b/fs/bcachefs/disk_accounting.c @@ -79,6 +79,8 @@ static inline void accounting_key_init(struct bkey_i *k, struct disk_accounting_ memcpy_u64s_small(acc->v.d, d, nr); } +static int bch2_accounting_update_sb_one(struct bch_fs *, struct bpos); + int bch2_disk_accounting_mod(struct btree_trans *trans, struct disk_accounting_pos *k, s64 *d, unsigned nr, bool gc) @@ -96,9 +98,16 @@ int bch2_disk_accounting_mod(struct btree_trans *trans, accounting_key_init(&k_i.k, k, d, nr); - return likely(!gc) - ? bch2_trans_update_buffered(trans, BTREE_ID_accounting, &k_i.k) - : bch2_accounting_mem_add(trans, bkey_i_to_s_c_accounting(&k_i.k), true); + if (unlikely(gc)) { + int ret = bch2_accounting_mem_add(trans, bkey_i_to_s_c_accounting(&k_i.k), true); + if (ret == -BCH_ERR_btree_insert_need_mark_replicas) + ret = drop_locks_do(trans, + bch2_accounting_update_sb_one(trans->c, disk_accounting_pos_to_bpos(k))) ?: + bch2_accounting_mem_add(trans, bkey_i_to_s_c_accounting(&k_i.k), true); + return ret; + } else { + return bch2_trans_update_buffered(trans, BTREE_ID_accounting, &k_i.k); + } } int bch2_mod_dev_cached_sectors(struct btree_trans *trans, @@ -127,14 +136,15 @@ static inline bool is_zero(char *start, char *end) #define field_end(p, member) (((void *) (&p.member)) + sizeof(p.member)) int bch2_accounting_validate(struct bch_fs *c, struct bkey_s_c k, - enum bch_validate_flags flags) + struct bkey_validate_context from) { struct disk_accounting_pos acc_k; bpos_to_disk_accounting_pos(&acc_k, k.k->p); void *end = &acc_k + 1; int ret = 0; - bkey_fsck_err_on(bversion_zero(k.k->bversion), + bkey_fsck_err_on((from.flags & BCH_VALIDATE_commit) && + bversion_zero(k.k->bversion), c, accounting_key_version_0, "accounting key with version=0"); @@ -217,7 +227,8 @@ void bch2_accounting_key_to_text(struct printbuf *out, struct disk_accounting_po prt_printf(out, "id=%u", k->snapshot.id); break; case BCH_DISK_ACCOUNTING_btree: - prt_printf(out, "btree=%s", bch2_btree_id_str(k->btree.id)); + prt_str(out, "btree="); + bch2_btree_id_to_text(out, k->btree.id); break; } } @@ -243,10 +254,10 @@ void bch2_accounting_swab(struct bkey_s k) } static inline void __accounting_to_replicas(struct bch_replicas_entry_v1 *r, - struct disk_accounting_pos acc) + struct disk_accounting_pos *acc) { - unsafe_memcpy(r, &acc.replicas, - replicas_entry_bytes(&acc.replicas), + unsafe_memcpy(r, &acc->replicas, + replicas_entry_bytes(&acc->replicas), "variable length struct"); } @@ -257,7 +268,7 @@ static inline bool accounting_to_replicas(struct bch_replicas_entry_v1 *r, struc switch (acc_k.type) { case BCH_DISK_ACCOUNTING_replicas: - __accounting_to_replicas(r, acc_k); + __accounting_to_replicas(r, &acc_k); return true; default: return false; @@ -322,6 +333,14 @@ static int __bch2_accounting_mem_insert(struct bch_fs *c, struct bkey_s_c_accoun eytzinger0_sort(acc->k.data, acc->k.nr, sizeof(acc->k.data[0]), accounting_pos_cmp, NULL); + + if (trace_accounting_mem_insert_enabled()) { + struct printbuf buf = PRINTBUF; + + bch2_accounting_to_text(&buf, c, a.s_c); + trace_accounting_mem_insert(c, buf.buf); + printbuf_exit(&buf); + } return 0; err: free_percpu(n.v[1]); @@ -461,32 +480,6 @@ int bch2_fs_accounting_read(struct bch_fs *c, darray_char *out_buf, unsigned acc return ret; } -void bch2_fs_accounting_to_text(struct printbuf *out, struct bch_fs *c) -{ - struct bch_accounting_mem *acc = &c->accounting; - - percpu_down_read(&c->mark_lock); - out->atomic++; - - eytzinger0_for_each(i, acc->k.nr) { - struct disk_accounting_pos acc_k; - bpos_to_disk_accounting_pos(&acc_k, acc->k.data[i].pos); - - bch2_accounting_key_to_text(out, &acc_k); - - u64 v[BCH_ACCOUNTING_MAX_COUNTERS]; - bch2_accounting_mem_read_counters(acc, i, v, ARRAY_SIZE(v), false); - - prt_str(out, ":"); - for (unsigned j = 0; j < acc->k.data[i].nr_counters; j++) - prt_printf(out, " %llu", v[j]); - prt_newline(out); - } - - --out->atomic; - percpu_up_read(&c->mark_lock); -} - static void bch2_accounting_free_counters(struct bch_accounting_mem *acc, bool gc) { darray_for_each(acc->k, e) { @@ -625,7 +618,7 @@ static int bch2_disk_accounting_validate_late(struct btree_trans *trans, switch (acc.type) { case BCH_DISK_ACCOUNTING_replicas: { struct bch_replicas_padded r; - __accounting_to_replicas(&r.e, acc); + __accounting_to_replicas(&r.e, &acc); for (unsigned i = 0; i < r.e.nr_devs; i++) if (r.e.devs[i] != BCH_SB_MEMBER_INVALID && @@ -699,11 +692,45 @@ int bch2_accounting_read(struct bch_fs *c) struct btree_trans *trans = bch2_trans_get(c); struct printbuf buf = PRINTBUF; - int ret = for_each_btree_key(trans, iter, - BTREE_ID_accounting, POS_MIN, + /* + * We might run more than once if we rewind to start topology repair or + * btree node scan - and those might cause us to get different results, + * so we can't just skip if we've already run. + * + * Instead, zero out any accounting we have: + */ + percpu_down_write(&c->mark_lock); + darray_for_each(acc->k, e) + percpu_memset(e->v[0], 0, sizeof(u64) * e->nr_counters); + for_each_member_device(c, ca) + percpu_memset(ca->usage, 0, sizeof(*ca->usage)); + percpu_memset(c->usage, 0, sizeof(*c->usage)); + percpu_up_write(&c->mark_lock); + + struct btree_iter iter; + bch2_trans_iter_init(trans, &iter, BTREE_ID_accounting, POS_MIN, + BTREE_ITER_prefetch|BTREE_ITER_all_snapshots); + iter.flags &= ~BTREE_ITER_with_journal; + int ret = for_each_btree_key_continue(trans, iter, BTREE_ITER_prefetch|BTREE_ITER_all_snapshots, k, ({ struct bkey u; struct bkey_s_c k = bch2_btree_path_peek_slot_exact(btree_iter_path(trans, &iter), &u); + + if (k.k->type != KEY_TYPE_accounting) + continue; + + struct disk_accounting_pos acc_k; + bpos_to_disk_accounting_pos(&acc_k, k.k->p); + + if (acc_k.type >= BCH_DISK_ACCOUNTING_TYPE_NR) + break; + + if (!bch2_accounting_is_mem(acc_k)) { + struct disk_accounting_pos next = { .type = acc_k.type + 1 }; + bch2_btree_iter_set_pos(&iter, disk_accounting_pos_to_bpos(&next)); + continue; + } + accounting_read_key(trans, k); })); if (ret) @@ -715,6 +742,12 @@ int bch2_accounting_read(struct bch_fs *c) darray_for_each(*keys, i) { if (i->k->k.type == KEY_TYPE_accounting) { + struct disk_accounting_pos acc_k; + bpos_to_disk_accounting_pos(&acc_k, i->k->k.p); + + if (!bch2_accounting_is_mem(acc_k)) + continue; + struct bkey_s_c k = bkey_i_to_s_c(i->k); unsigned idx = eytzinger0_find(acc->k.data, acc->k.nr, sizeof(acc->k.data[0]), @@ -748,15 +781,16 @@ int bch2_accounting_read(struct bch_fs *c) keys->gap = keys->nr = dst - keys->data; percpu_down_write(&c->mark_lock); - unsigned i = 0; - while (i < acc->k.nr) { - unsigned idx = inorder_to_eytzinger0(i, acc->k.nr); + darray_for_each_reverse(acc->k, i) { struct disk_accounting_pos acc_k; - bpos_to_disk_accounting_pos(&acc_k, acc->k.data[idx].pos); + bpos_to_disk_accounting_pos(&acc_k, i->pos); u64 v[BCH_ACCOUNTING_MAX_COUNTERS]; - bch2_accounting_mem_read_counters(acc, idx, v, ARRAY_SIZE(v), false); + memset(v, 0, sizeof(v)); + + for (unsigned j = 0; j < i->nr_counters; j++) + v[j] = percpu_u64_get(i->v[0] + j); /* * If the entry counters are zeroed, it should be treated as @@ -765,26 +799,25 @@ int bch2_accounting_read(struct bch_fs *c) * Remove it, so that if it's re-added it gets re-marked in the * superblock: */ - ret = bch2_is_zero(v, sizeof(v[0]) * acc->k.data[idx].nr_counters) + ret = bch2_is_zero(v, sizeof(v[0]) * i->nr_counters) ? -BCH_ERR_remove_disk_accounting_entry - : bch2_disk_accounting_validate_late(trans, acc_k, - v, acc->k.data[idx].nr_counters); + : bch2_disk_accounting_validate_late(trans, acc_k, v, i->nr_counters); if (ret == -BCH_ERR_remove_disk_accounting_entry) { - free_percpu(acc->k.data[idx].v[0]); - free_percpu(acc->k.data[idx].v[1]); - darray_remove_item(&acc->k, &acc->k.data[idx]); - eytzinger0_sort(acc->k.data, acc->k.nr, sizeof(acc->k.data[0]), - accounting_pos_cmp, NULL); + free_percpu(i->v[0]); + free_percpu(i->v[1]); + darray_remove_item(&acc->k, i); ret = 0; continue; } if (ret) goto fsck_err; - i++; } + eytzinger0_sort(acc->k.data, acc->k.nr, sizeof(acc->k.data[0]), + accounting_pos_cmp, NULL); + preempt_disable(); struct bch_fs_usage_base *usage = this_cpu_ptr(c->usage); @@ -804,7 +837,7 @@ int bch2_accounting_read(struct bch_fs *c) break; case BCH_DISK_ACCOUNTING_dev_data_type: rcu_read_lock(); - struct bch_dev *ca = bch2_dev_rcu(c, k.dev_data_type.dev); + struct bch_dev *ca = bch2_dev_rcu_noerror(c, k.dev_data_type.dev); if (ca) { struct bch_dev_usage_type __percpu *d = &ca->usage->d[k.dev_data_type.data_type]; percpu_u64_set(&d->buckets, v[0]); @@ -881,10 +914,13 @@ void bch2_verify_accounting_clean(struct bch_fs *c) bpos_to_disk_accounting_pos(&acc_k, k.k->p); if (acc_k.type >= BCH_DISK_ACCOUNTING_TYPE_NR) - continue; + break; - if (acc_k.type == BCH_DISK_ACCOUNTING_inum) + if (!bch2_accounting_is_mem(acc_k)) { + struct disk_accounting_pos next = { .type = acc_k.type + 1 }; + bch2_btree_iter_set_pos(&iter, disk_accounting_pos_to_bpos(&next)); continue; + } bch2_accounting_mem_read(c, k.k->p, v, nr); @@ -910,7 +946,7 @@ void bch2_verify_accounting_clean(struct bch_fs *c) break; case BCH_DISK_ACCOUNTING_dev_data_type: { rcu_read_lock(); - struct bch_dev *ca = bch2_dev_rcu(c, acc_k.dev_data_type.dev); + struct bch_dev *ca = bch2_dev_rcu_noerror(c, acc_k.dev_data_type.dev); if (!ca) { rcu_read_unlock(); continue; diff --git a/fs/bcachefs/disk_accounting.h b/fs/bcachefs/disk_accounting.h index 4ea6c8a092bc..5360cbb3ec29 100644 --- a/fs/bcachefs/disk_accounting.h +++ b/fs/bcachefs/disk_accounting.h @@ -2,6 +2,7 @@ #ifndef _BCACHEFS_DISK_ACCOUNTING_H #define _BCACHEFS_DISK_ACCOUNTING_H +#include "btree_update.h" #include "eytzinger.h" #include "sb-members.h" @@ -62,27 +63,32 @@ static inline void fs_usage_data_type_to_base(struct bch_fs_usage_base *fs_usage static inline void bpos_to_disk_accounting_pos(struct disk_accounting_pos *acc, struct bpos p) { - acc->_pad = p; + BUILD_BUG_ON(sizeof(*acc) != sizeof(p)); + #if __BYTE_ORDER__ == __ORDER_BIG_ENDIAN__ - bch2_bpos_swab(&acc->_pad); + acc->_pad = p; +#else + memcpy_swab(acc, &p, sizeof(p)); #endif } -static inline struct bpos disk_accounting_pos_to_bpos(struct disk_accounting_pos *k) +static inline struct bpos disk_accounting_pos_to_bpos(struct disk_accounting_pos *acc) { - struct bpos ret = k->_pad; - + struct bpos p; #if __BYTE_ORDER__ == __ORDER_BIG_ENDIAN__ - bch2_bpos_swab(&ret); + p = acc->_pad; +#else + memcpy_swab(&p, acc, sizeof(p)); #endif - return ret; + return p; } int bch2_disk_accounting_mod(struct btree_trans *, struct disk_accounting_pos *, s64 *, unsigned, bool); int bch2_mod_dev_cached_sectors(struct btree_trans *, unsigned, s64, bool); -int bch2_accounting_validate(struct bch_fs *, struct bkey_s_c, enum bch_validate_flags); +int bch2_accounting_validate(struct bch_fs *, struct bkey_s_c, + struct bkey_validate_context); void bch2_accounting_key_to_text(struct printbuf *, struct disk_accounting_pos *); void bch2_accounting_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c); void bch2_accounting_swab(struct bkey_s); @@ -112,6 +118,12 @@ enum bch_accounting_mode { int bch2_accounting_mem_insert(struct bch_fs *, struct bkey_s_c_accounting, enum bch_accounting_mode); void bch2_accounting_mem_gc(struct bch_fs *); +static inline bool bch2_accounting_is_mem(struct disk_accounting_pos acc) +{ + return acc.type < BCH_DISK_ACCOUNTING_TYPE_NR && + acc.type != BCH_DISK_ACCOUNTING_inum; +} + /* * Update in memory counters so they match the btree update we're doing; called * from transaction commit path @@ -126,9 +138,10 @@ static inline int bch2_accounting_mem_mod_locked(struct btree_trans *trans, bpos_to_disk_accounting_pos(&acc_k, a.k->p); bool gc = mode == BCH_ACCOUNTING_gc; - EBUG_ON(gc && !acc->gc_running); + if (gc && !acc->gc_running) + return 0; - if (acc_k.type == BCH_DISK_ACCOUNTING_inum) + if (!bch2_accounting_is_mem(acc_k)) return 0; if (mode == BCH_ACCOUNTING_normal) { @@ -141,7 +154,7 @@ static inline int bch2_accounting_mem_mod_locked(struct btree_trans *trans, break; case BCH_DISK_ACCOUNTING_dev_data_type: rcu_read_lock(); - struct bch_dev *ca = bch2_dev_rcu(c, acc_k.dev_data_type.dev); + struct bch_dev *ca = bch2_dev_rcu_noerror(c, acc_k.dev_data_type.dev); if (ca) { this_cpu_add(ca->usage->d[acc_k.dev_data_type.data_type].buckets, a.v->d[0]); this_cpu_add(ca->usage->d[acc_k.dev_data_type.data_type].sectors, a.v->d[1]); @@ -204,9 +217,45 @@ static inline void bch2_accounting_mem_read(struct bch_fs *c, struct bpos p, bch2_accounting_mem_read_counters(acc, idx, v, nr, false); } +static inline struct bversion journal_pos_to_bversion(struct journal_res *res, unsigned offset) +{ + EBUG_ON(!res->ref); + + return (struct bversion) { + .hi = res->seq >> 32, + .lo = (res->seq << 32) | (res->offset + offset), + }; +} + +static inline int bch2_accounting_trans_commit_hook(struct btree_trans *trans, + struct bkey_i_accounting *a, + unsigned commit_flags) +{ + a->k.bversion = journal_pos_to_bversion(&trans->journal_res, + (u64 *) a - (u64 *) trans->journal_entries); + + EBUG_ON(bversion_zero(a->k.bversion)); + + return likely(!(commit_flags & BCH_TRANS_COMMIT_skip_accounting_apply)) + ? bch2_accounting_mem_mod_locked(trans, accounting_i_to_s_c(a), BCH_ACCOUNTING_normal) + : 0; +} + +static inline void bch2_accounting_trans_commit_revert(struct btree_trans *trans, + struct bkey_i_accounting *a_i, + unsigned commit_flags) +{ + if (likely(!(commit_flags & BCH_TRANS_COMMIT_skip_accounting_apply))) { + struct bkey_s_accounting a = accounting_i_to_s(a_i); + + bch2_accounting_neg(a); + bch2_accounting_mem_mod_locked(trans, a.c, BCH_ACCOUNTING_normal); + bch2_accounting_neg(a); + } +} + int bch2_fs_replicas_usage_read(struct bch_fs *, darray_char *); int bch2_fs_accounting_read(struct bch_fs *, darray_char *, unsigned); -void bch2_fs_accounting_to_text(struct printbuf *, struct bch_fs *); int bch2_gc_accounting_start(struct bch_fs *); int bch2_gc_accounting_done(struct bch_fs *); diff --git a/fs/bcachefs/ec.c b/fs/bcachefs/ec.c index 6094afb0c6be..d2a5e76e6479 100644 --- a/fs/bcachefs/ec.c +++ b/fs/bcachefs/ec.c @@ -26,6 +26,7 @@ #include "util.h" #include <linux/sort.h> +#include <linux/string_choices.h> #ifdef __KERNEL__ @@ -109,7 +110,7 @@ struct ec_bio { /* Stripes btree keys: */ int bch2_stripe_validate(struct bch_fs *c, struct bkey_s_c k, - enum bch_validate_flags flags) + struct bkey_validate_context from) { const struct bch_stripe *s = bkey_s_c_to_stripe(k).v; int ret = 0; @@ -129,7 +130,7 @@ int bch2_stripe_validate(struct bch_fs *c, struct bkey_s_c k, "invalid csum granularity (%u >= 64)", s->csum_granularity_bits); - ret = bch2_bkey_ptrs_validate(c, k, flags); + ret = bch2_bkey_ptrs_validate(c, k, from); fsck_err: return ret; } @@ -304,13 +305,12 @@ static int mark_stripe_bucket(struct btree_trans *trans, } if (flags & BTREE_TRIGGER_gc) { - percpu_down_read(&c->mark_lock); struct bucket *g = gc_bucket(ca, bucket.offset); if (bch2_fs_inconsistent_on(!g, c, "reference to invalid bucket on device %u\n %s", ptr->dev, (bch2_bkey_val_to_text(&buf, c, s.s_c), buf.buf))) { ret = -BCH_ERR_mark_stripe; - goto err_unlock; + goto err; } bucket_lock(g); @@ -318,8 +318,7 @@ static int mark_stripe_bucket(struct btree_trans *trans, ret = __mark_stripe_bucket(trans, ca, s, ptr_idx, deleting, bucket, &new, flags); alloc_to_bucket(g, new); bucket_unlock(g); -err_unlock: - percpu_up_read(&c->mark_lock); + if (!ret) ret = bch2_alloc_key_to_dev_counters(trans, ca, &old, &new, flags); } @@ -732,7 +731,7 @@ static void ec_block_endio(struct bio *bio) ? BCH_MEMBER_ERROR_write : BCH_MEMBER_ERROR_read, "erasure coding %s error: %s", - bio_data_dir(bio) ? "write" : "read", + str_write_read(bio_data_dir(bio)), bch2_blk_status_to_str(bio->bi_status))) clear_bit(ec_bio->idx, ec_bio->buf->valid); @@ -909,7 +908,7 @@ err: bch2_bkey_val_to_text(&msgbuf, c, orig_k); bch_err_ratelimited(c, "error doing reconstruct read: %s\n %s", msg, msgbuf.buf); - printbuf_exit(&msgbuf);; + printbuf_exit(&msgbuf); ret = -BCH_ERR_stripe_reconstruct; goto out; } @@ -1266,11 +1265,11 @@ static int ec_stripe_update_extent(struct btree_trans *trans, struct bch_dev *ca, struct bpos bucket, u8 gen, struct ec_stripe_buf *s, - struct bpos *bp_pos) + struct bkey_s_c_backpointer bp, + struct bkey_buf *last_flushed) { struct bch_stripe *v = &bkey_i_to_stripe(&s->key)->v; struct bch_fs *c = trans->c; - struct bch_backpointer bp; struct btree_iter iter; struct bkey_s_c k; const struct bch_extent_ptr *ptr_c; @@ -1279,33 +1278,26 @@ static int ec_stripe_update_extent(struct btree_trans *trans, struct bkey_i *n; int ret, dev, block; - ret = bch2_get_next_backpointer(trans, ca, bucket, gen, - bp_pos, &bp, BTREE_ITER_cached); - if (ret) - return ret; - if (bpos_eq(*bp_pos, SPOS_MAX)) - return 0; - - if (bp.level) { + if (bp.v->level) { struct printbuf buf = PRINTBUF; struct btree_iter node_iter; struct btree *b; - b = bch2_backpointer_get_node(trans, &node_iter, *bp_pos, bp); + b = bch2_backpointer_get_node(trans, bp, &node_iter, last_flushed); bch2_trans_iter_exit(trans, &node_iter); if (!b) return 0; prt_printf(&buf, "found btree node in erasure coded bucket: b=%px\n", b); - bch2_backpointer_to_text(&buf, &bp); + bch2_bkey_val_to_text(&buf, c, bp.s_c); bch2_fs_inconsistent(c, "%s", buf.buf); printbuf_exit(&buf); return -EIO; } - k = bch2_backpointer_get_key(trans, &iter, *bp_pos, bp, BTREE_ITER_intent); + k = bch2_backpointer_get_key(trans, bp, &iter, BTREE_ITER_intent, last_flushed); ret = bkey_err(k); if (ret) return ret; @@ -1364,7 +1356,6 @@ static int ec_stripe_update_bucket(struct btree_trans *trans, struct ec_stripe_b struct bch_fs *c = trans->c; struct bch_stripe *v = &bkey_i_to_stripe(&s->key)->v; struct bch_extent_ptr ptr = v->ptrs[block]; - struct bpos bp_pos = POS_MIN; int ret = 0; struct bch_dev *ca = bch2_dev_tryget(c, ptr.dev); @@ -1373,19 +1364,27 @@ static int ec_stripe_update_bucket(struct btree_trans *trans, struct ec_stripe_b struct bpos bucket_pos = PTR_BUCKET_POS(ca, &ptr); - while (1) { - ret = commit_do(trans, NULL, NULL, - BCH_TRANS_COMMIT_no_check_rw| - BCH_TRANS_COMMIT_no_enospc, - ec_stripe_update_extent(trans, ca, bucket_pos, ptr.gen, s, &bp_pos)); - if (ret) - break; - if (bkey_eq(bp_pos, POS_MAX)) + struct bkey_buf last_flushed; + bch2_bkey_buf_init(&last_flushed); + bkey_init(&last_flushed.k->k); + + ret = for_each_btree_key_max_commit(trans, bp_iter, BTREE_ID_backpointers, + bucket_pos_to_bp_start(ca, bucket_pos), + bucket_pos_to_bp_end(ca, bucket_pos), 0, bp_k, + NULL, NULL, + BCH_TRANS_COMMIT_no_check_rw| + BCH_TRANS_COMMIT_no_enospc, ({ + if (bkey_ge(bp_k.k->p, bucket_pos_to_bp(ca, bpos_nosnap_successor(bucket_pos), 0))) break; - bp_pos = bpos_nosnap_successor(bp_pos); - } + if (bp_k.k->type != KEY_TYPE_backpointer) + continue; + ec_stripe_update_extent(trans, ca, bucket_pos, ptr.gen, s, + bkey_s_c_to_backpointer(bp_k), &last_flushed); + })); + + bch2_bkey_buf_exit(&last_flushed, c); bch2_dev_put(ca); return ret; } @@ -1707,7 +1706,7 @@ static void ec_stripe_key_init(struct bch_fs *c, set_bkey_val_u64s(&s->k, u64s); } -static int ec_new_stripe_alloc(struct bch_fs *c, struct ec_stripe_head *h) +static struct ec_stripe_new *ec_new_stripe_alloc(struct bch_fs *c, struct ec_stripe_head *h) { struct ec_stripe_new *s; @@ -1715,7 +1714,7 @@ static int ec_new_stripe_alloc(struct bch_fs *c, struct ec_stripe_head *h) s = kzalloc(sizeof(*s), GFP_KERNEL); if (!s) - return -BCH_ERR_ENOMEM_ec_new_stripe_alloc; + return NULL; mutex_init(&s->lock); closure_init(&s->iodone, NULL); @@ -1730,10 +1729,7 @@ static int ec_new_stripe_alloc(struct bch_fs *c, struct ec_stripe_head *h) ec_stripe_key_init(c, &s->new_stripe.key, s->nr_data, s->nr_parity, h->blocksize, h->disk_label); - - h->s = s; - h->nr_created++; - return 0; + return s; } static void ec_stripe_head_devs_update(struct bch_fs *c, struct ec_stripe_head *h) @@ -1878,25 +1874,26 @@ err: return h; } -static int new_stripe_alloc_buckets(struct btree_trans *trans, struct ec_stripe_head *h, +static int new_stripe_alloc_buckets(struct btree_trans *trans, + struct ec_stripe_head *h, struct ec_stripe_new *s, enum bch_watermark watermark, struct closure *cl) { struct bch_fs *c = trans->c; struct bch_devs_mask devs = h->devs; struct open_bucket *ob; struct open_buckets buckets; - struct bch_stripe *v = &bkey_i_to_stripe(&h->s->new_stripe.key)->v; + struct bch_stripe *v = &bkey_i_to_stripe(&s->new_stripe.key)->v; unsigned i, j, nr_have_parity = 0, nr_have_data = 0; bool have_cache = true; int ret = 0; - BUG_ON(v->nr_blocks != h->s->nr_data + h->s->nr_parity); - BUG_ON(v->nr_redundant != h->s->nr_parity); + BUG_ON(v->nr_blocks != s->nr_data + s->nr_parity); + BUG_ON(v->nr_redundant != s->nr_parity); /* * We bypass the sector allocator which normally does this: */ bitmap_and(devs.d, devs.d, c->rw_devs[BCH_DATA_user].d, BCH_SB_MEMBERS_MAX); - for_each_set_bit(i, h->s->blocks_gotten, v->nr_blocks) { + for_each_set_bit(i, s->blocks_gotten, v->nr_blocks) { /* * Note: we don't yet repair invalid blocks (failed/removed * devices) when reusing stripes - we still need a codepath to @@ -1906,21 +1903,21 @@ static int new_stripe_alloc_buckets(struct btree_trans *trans, struct ec_stripe_ if (v->ptrs[i].dev != BCH_SB_MEMBER_INVALID) __clear_bit(v->ptrs[i].dev, devs.d); - if (i < h->s->nr_data) + if (i < s->nr_data) nr_have_data++; else nr_have_parity++; } - BUG_ON(nr_have_data > h->s->nr_data); - BUG_ON(nr_have_parity > h->s->nr_parity); + BUG_ON(nr_have_data > s->nr_data); + BUG_ON(nr_have_parity > s->nr_parity); buckets.nr = 0; - if (nr_have_parity < h->s->nr_parity) { + if (nr_have_parity < s->nr_parity) { ret = bch2_bucket_alloc_set_trans(trans, &buckets, &h->parity_stripe, &devs, - h->s->nr_parity, + s->nr_parity, &nr_have_parity, &have_cache, 0, BCH_DATA_parity, @@ -1928,14 +1925,14 @@ static int new_stripe_alloc_buckets(struct btree_trans *trans, struct ec_stripe_ cl); open_bucket_for_each(c, &buckets, ob, i) { - j = find_next_zero_bit(h->s->blocks_gotten, - h->s->nr_data + h->s->nr_parity, - h->s->nr_data); - BUG_ON(j >= h->s->nr_data + h->s->nr_parity); + j = find_next_zero_bit(s->blocks_gotten, + s->nr_data + s->nr_parity, + s->nr_data); + BUG_ON(j >= s->nr_data + s->nr_parity); - h->s->blocks[j] = buckets.v[i]; + s->blocks[j] = buckets.v[i]; v->ptrs[j] = bch2_ob_ptr(c, ob); - __set_bit(j, h->s->blocks_gotten); + __set_bit(j, s->blocks_gotten); } if (ret) @@ -1943,11 +1940,11 @@ static int new_stripe_alloc_buckets(struct btree_trans *trans, struct ec_stripe_ } buckets.nr = 0; - if (nr_have_data < h->s->nr_data) { + if (nr_have_data < s->nr_data) { ret = bch2_bucket_alloc_set_trans(trans, &buckets, &h->block_stripe, &devs, - h->s->nr_data, + s->nr_data, &nr_have_data, &have_cache, 0, BCH_DATA_user, @@ -1955,13 +1952,13 @@ static int new_stripe_alloc_buckets(struct btree_trans *trans, struct ec_stripe_ cl); open_bucket_for_each(c, &buckets, ob, i) { - j = find_next_zero_bit(h->s->blocks_gotten, - h->s->nr_data, 0); - BUG_ON(j >= h->s->nr_data); + j = find_next_zero_bit(s->blocks_gotten, + s->nr_data, 0); + BUG_ON(j >= s->nr_data); - h->s->blocks[j] = buckets.v[i]; + s->blocks[j] = buckets.v[i]; v->ptrs[j] = bch2_ob_ptr(c, ob); - __set_bit(j, h->s->blocks_gotten); + __set_bit(j, s->blocks_gotten); } if (ret) @@ -2007,73 +2004,78 @@ static s64 get_existing_stripe(struct bch_fs *c, return ret; } -static int __bch2_ec_stripe_head_reuse(struct btree_trans *trans, struct ec_stripe_head *h) +static int init_new_stripe_from_existing(struct bch_fs *c, struct ec_stripe_new *s) { - struct bch_fs *c = trans->c; - struct bch_stripe *new_v = &bkey_i_to_stripe(&h->s->new_stripe.key)->v; - struct bch_stripe *existing_v; + struct bch_stripe *new_v = &bkey_i_to_stripe(&s->new_stripe.key)->v; + struct bch_stripe *existing_v = &bkey_i_to_stripe(&s->existing_stripe.key)->v; unsigned i; - s64 idx; - int ret; - - /* - * If we can't allocate a new stripe, and there's no stripes with empty - * blocks for us to reuse, that means we have to wait on copygc: - */ - idx = get_existing_stripe(c, h); - if (idx < 0) - return -BCH_ERR_stripe_alloc_blocked; - - ret = get_stripe_key_trans(trans, idx, &h->s->existing_stripe); - bch2_fs_fatal_err_on(ret && !bch2_err_matches(ret, BCH_ERR_transaction_restart), c, - "reading stripe key: %s", bch2_err_str(ret)); - if (ret) { - bch2_stripe_close(c, h->s); - return ret; - } - existing_v = &bkey_i_to_stripe(&h->s->existing_stripe.key)->v; - - BUG_ON(existing_v->nr_redundant != h->s->nr_parity); - h->s->nr_data = existing_v->nr_blocks - + BUG_ON(existing_v->nr_redundant != s->nr_parity); + s->nr_data = existing_v->nr_blocks - existing_v->nr_redundant; - ret = ec_stripe_buf_init(&h->s->existing_stripe, 0, h->blocksize); + int ret = ec_stripe_buf_init(&s->existing_stripe, 0, le16_to_cpu(existing_v->sectors)); if (ret) { - bch2_stripe_close(c, h->s); + bch2_stripe_close(c, s); return ret; } - BUG_ON(h->s->existing_stripe.size != h->blocksize); - BUG_ON(h->s->existing_stripe.size != le16_to_cpu(existing_v->sectors)); + BUG_ON(s->existing_stripe.size != le16_to_cpu(existing_v->sectors)); /* * Free buckets we initially allocated - they might conflict with * blocks from the stripe we're reusing: */ - for_each_set_bit(i, h->s->blocks_gotten, new_v->nr_blocks) { - bch2_open_bucket_put(c, c->open_buckets + h->s->blocks[i]); - h->s->blocks[i] = 0; + for_each_set_bit(i, s->blocks_gotten, new_v->nr_blocks) { + bch2_open_bucket_put(c, c->open_buckets + s->blocks[i]); + s->blocks[i] = 0; } - memset(h->s->blocks_gotten, 0, sizeof(h->s->blocks_gotten)); - memset(h->s->blocks_allocated, 0, sizeof(h->s->blocks_allocated)); + memset(s->blocks_gotten, 0, sizeof(s->blocks_gotten)); + memset(s->blocks_allocated, 0, sizeof(s->blocks_allocated)); - for (i = 0; i < existing_v->nr_blocks; i++) { + for (unsigned i = 0; i < existing_v->nr_blocks; i++) { if (stripe_blockcount_get(existing_v, i)) { - __set_bit(i, h->s->blocks_gotten); - __set_bit(i, h->s->blocks_allocated); + __set_bit(i, s->blocks_gotten); + __set_bit(i, s->blocks_allocated); } - ec_block_io(c, &h->s->existing_stripe, READ, i, &h->s->iodone); + ec_block_io(c, &s->existing_stripe, READ, i, &s->iodone); } - bkey_copy(&h->s->new_stripe.key, &h->s->existing_stripe.key); - h->s->have_existing_stripe = true; + bkey_copy(&s->new_stripe.key, &s->existing_stripe.key); + s->have_existing_stripe = true; return 0; } -static int __bch2_ec_stripe_head_reserve(struct btree_trans *trans, struct ec_stripe_head *h) +static int __bch2_ec_stripe_head_reuse(struct btree_trans *trans, struct ec_stripe_head *h, + struct ec_stripe_new *s) +{ + struct bch_fs *c = trans->c; + s64 idx; + int ret; + + /* + * If we can't allocate a new stripe, and there's no stripes with empty + * blocks for us to reuse, that means we have to wait on copygc: + */ + idx = get_existing_stripe(c, h); + if (idx < 0) + return -BCH_ERR_stripe_alloc_blocked; + + ret = get_stripe_key_trans(trans, idx, &s->existing_stripe); + bch2_fs_fatal_err_on(ret && !bch2_err_matches(ret, BCH_ERR_transaction_restart), c, + "reading stripe key: %s", bch2_err_str(ret)); + if (ret) { + bch2_stripe_close(c, s); + return ret; + } + + return init_new_stripe_from_existing(c, s); +} + +static int __bch2_ec_stripe_head_reserve(struct btree_trans *trans, struct ec_stripe_head *h, + struct ec_stripe_new *s) { struct bch_fs *c = trans->c; struct btree_iter iter; @@ -2082,15 +2084,19 @@ static int __bch2_ec_stripe_head_reserve(struct btree_trans *trans, struct ec_st struct bpos start_pos = bpos_max(min_pos, POS(0, c->ec_stripe_hint)); int ret; - if (!h->s->res.sectors) { - ret = bch2_disk_reservation_get(c, &h->s->res, + if (!s->res.sectors) { + ret = bch2_disk_reservation_get(c, &s->res, h->blocksize, - h->s->nr_parity, + s->nr_parity, BCH_DISK_RESERVATION_NOFAIL); if (ret) return ret; } + /* + * Allocate stripe slot + * XXX: we're going to need a bitrange btree of free stripes + */ for_each_btree_key_norestart(trans, iter, BTREE_ID_stripes, start_pos, BTREE_ITER_slots|BTREE_ITER_intent, k, ret) { if (bkey_gt(k.k->p, POS(0, U32_MAX))) { @@ -2105,7 +2111,7 @@ static int __bch2_ec_stripe_head_reserve(struct btree_trans *trans, struct ec_st } if (bkey_deleted(k.k) && - bch2_try_open_stripe(c, h->s, k.k->p.offset)) + bch2_try_open_stripe(c, s, k.k->p.offset)) break; } @@ -2116,16 +2122,16 @@ static int __bch2_ec_stripe_head_reserve(struct btree_trans *trans, struct ec_st ret = ec_stripe_mem_alloc(trans, &iter); if (ret) { - bch2_stripe_close(c, h->s); + bch2_stripe_close(c, s); goto err; } - h->s->new_stripe.key.k.p = iter.pos; + s->new_stripe.key.k.p = iter.pos; out: bch2_trans_iter_exit(trans, &iter); return ret; err: - bch2_disk_reservation_put(c, &h->s->res); + bch2_disk_reservation_put(c, &s->res); goto out; } @@ -2156,22 +2162,27 @@ struct ec_stripe_head *bch2_ec_stripe_head_get(struct btree_trans *trans, return h; if (!h->s) { - ret = ec_new_stripe_alloc(c, h); - if (ret) { + h->s = ec_new_stripe_alloc(c, h); + if (!h->s) { + ret = -BCH_ERR_ENOMEM_ec_new_stripe_alloc; bch_err(c, "failed to allocate new stripe"); goto err; } + + h->nr_created++; } - if (h->s->allocated) + struct ec_stripe_new *s = h->s; + + if (s->allocated) goto allocated; - if (h->s->have_existing_stripe) + if (s->have_existing_stripe) goto alloc_existing; /* First, try to allocate a full stripe: */ - ret = new_stripe_alloc_buckets(trans, h, BCH_WATERMARK_stripe, NULL) ?: - __bch2_ec_stripe_head_reserve(trans, h); + ret = new_stripe_alloc_buckets(trans, h, s, BCH_WATERMARK_stripe, NULL) ?: + __bch2_ec_stripe_head_reserve(trans, h, s); if (!ret) goto allocate_buf; if (bch2_err_matches(ret, BCH_ERR_transaction_restart) || @@ -2183,15 +2194,15 @@ struct ec_stripe_head *bch2_ec_stripe_head_get(struct btree_trans *trans, * existing stripe: */ while (1) { - ret = __bch2_ec_stripe_head_reuse(trans, h); + ret = __bch2_ec_stripe_head_reuse(trans, h, s); if (!ret) break; if (waiting || !cl || ret != -BCH_ERR_stripe_alloc_blocked) goto err; if (watermark == BCH_WATERMARK_copygc) { - ret = new_stripe_alloc_buckets(trans, h, watermark, NULL) ?: - __bch2_ec_stripe_head_reserve(trans, h); + ret = new_stripe_alloc_buckets(trans, h, s, watermark, NULL) ?: + __bch2_ec_stripe_head_reserve(trans, h, s); if (ret) goto err; goto allocate_buf; @@ -2209,19 +2220,19 @@ alloc_existing: * Retry allocating buckets, with the watermark for this * particular write: */ - ret = new_stripe_alloc_buckets(trans, h, watermark, cl); + ret = new_stripe_alloc_buckets(trans, h, s, watermark, cl); if (ret) goto err; allocate_buf: - ret = ec_stripe_buf_init(&h->s->new_stripe, 0, h->blocksize); + ret = ec_stripe_buf_init(&s->new_stripe, 0, h->blocksize); if (ret) goto err; - h->s->allocated = true; + s->allocated = true; allocated: - BUG_ON(!h->s->idx); - BUG_ON(!h->s->new_stripe.data[0]); + BUG_ON(!s->idx); + BUG_ON(!s->new_stripe.data[0]); BUG_ON(trans->restarted); return h; err: @@ -2286,7 +2297,7 @@ err: int bch2_dev_remove_stripes(struct bch_fs *c, unsigned dev_idx) { return bch2_trans_run(c, - for_each_btree_key_upto_commit(trans, iter, + for_each_btree_key_max_commit(trans, iter, BTREE_ID_alloc, POS(dev_idx, 0), POS(dev_idx, U64_MAX), BTREE_ITER_intent, k, NULL, NULL, 0, ({ @@ -2449,11 +2460,9 @@ void bch2_fs_ec_exit(struct bch_fs *c) while (1) { mutex_lock(&c->ec_stripe_head_lock); - h = list_first_entry_or_null(&c->ec_stripe_head_list, - struct ec_stripe_head, list); - if (h) - list_del(&h->list); + h = list_pop_entry(&c->ec_stripe_head_list, struct ec_stripe_head, list); mutex_unlock(&c->ec_stripe_head_lock); + if (!h) break; diff --git a/fs/bcachefs/ec.h b/fs/bcachefs/ec.h index 43326370b410..583ca6a226da 100644 --- a/fs/bcachefs/ec.h +++ b/fs/bcachefs/ec.h @@ -6,9 +6,8 @@ #include "buckets_types.h" #include "extents_types.h" -enum bch_validate_flags; - -int bch2_stripe_validate(struct bch_fs *, struct bkey_s_c, enum bch_validate_flags); +int bch2_stripe_validate(struct bch_fs *, struct bkey_s_c, + struct bkey_validate_context); void bch2_stripe_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c); int bch2_trigger_stripe(struct btree_trans *, enum btree_id, unsigned, diff --git a/fs/bcachefs/ec_format.h b/fs/bcachefs/ec_format.h index 64ef52e00078..b9770f24f213 100644 --- a/fs/bcachefs/ec_format.h +++ b/fs/bcachefs/ec_format.h @@ -20,6 +20,23 @@ struct bch_stripe { */ __u8 disk_label; + /* + * Variable length sections: + * - Pointers + * - Checksums + * 2D array of [stripe block/device][csum block], with checksum block + * size given by csum_granularity_bits + * - Block sector counts: per-block array of u16s + * + * XXX: + * Either checksums should have come last, or we should have included a + * checksum_size field (the size in bytes of the checksum itself, not + * the blocksize the checksum covers). + * + * Currently we aren't able to access the block sector counts if the + * checksum type is unknown. + */ + struct bch_extent_ptr ptrs[]; } __packed __aligned(8); diff --git a/fs/bcachefs/errcode.h b/fs/bcachefs/errcode.h index 9c4fe5cdbfb7..4590cd0c7c90 100644 --- a/fs/bcachefs/errcode.h +++ b/fs/bcachefs/errcode.h @@ -54,7 +54,8 @@ x(ENOMEM, ENOMEM_compression_bounce_read_init) \ x(ENOMEM, ENOMEM_compression_bounce_write_init) \ x(ENOMEM, ENOMEM_compression_workspace_init) \ - x(ENOMEM, ENOMEM_decompression_workspace_init) \ + x(ENOMEM, ENOMEM_backpointer_mismatches_bitmap) \ + x(EIO, compression_workspace_not_initialized) \ x(ENOMEM, ENOMEM_bucket_gens) \ x(ENOMEM, ENOMEM_buckets_nouse) \ x(ENOMEM, ENOMEM_usage_init) \ @@ -116,6 +117,8 @@ x(ENOENT, ENOENT_dirent_doesnt_match_inode) \ x(ENOENT, ENOENT_dev_not_found) \ x(ENOENT, ENOENT_dev_idx_not_found) \ + x(ENOENT, ENOENT_inode_no_backpointer) \ + x(ENOENT, ENOENT_no_snapshot_tree_subvol) \ x(ENOTEMPTY, ENOTEMPTY_dir_not_empty) \ x(ENOTEMPTY, ENOTEMPTY_subvol_not_empty) \ x(EEXIST, EEXIST_str_hash_set) \ @@ -148,6 +151,7 @@ x(BCH_ERR_transaction_restart, transaction_restart_split_race) \ x(BCH_ERR_transaction_restart, transaction_restart_write_buffer_flush) \ x(BCH_ERR_transaction_restart, transaction_restart_nested) \ + x(BCH_ERR_transaction_restart, transaction_restart_commit) \ x(0, no_btree_node) \ x(BCH_ERR_no_btree_node, no_btree_node_relock) \ x(BCH_ERR_no_btree_node, no_btree_node_upgrade) \ @@ -164,7 +168,6 @@ x(BCH_ERR_btree_insert_fail, btree_insert_need_journal_res) \ x(BCH_ERR_btree_insert_fail, btree_insert_need_journal_reclaim) \ x(0, backpointer_to_overwritten_btree_node) \ - x(0, lock_fail_root_changed) \ x(0, journal_reclaim_would_deadlock) \ x(EINVAL, fsck) \ x(BCH_ERR_fsck, fsck_fix) \ @@ -173,7 +176,9 @@ x(BCH_ERR_fsck, fsck_errors_not_fixed) \ x(BCH_ERR_fsck, fsck_repair_unimplemented) \ x(BCH_ERR_fsck, fsck_repair_impossible) \ - x(0, restart_recovery) \ + x(EINVAL, restart_recovery) \ + x(EINVAL, not_in_recovery) \ + x(EINVAL, cannot_rewind_recovery) \ x(0, data_update_done) \ x(EINVAL, device_state_not_allowed) \ x(EINVAL, member_info_missing) \ @@ -192,7 +197,9 @@ x(EINVAL, opt_parse_error) \ x(EINVAL, remove_with_metadata_missing_unimplemented)\ x(EINVAL, remove_would_lose_data) \ - x(EINVAL, btree_iter_with_journal_not_supported) \ + x(EINVAL, no_resize_with_buckets_nouse) \ + x(EINVAL, inode_unpack_error) \ + x(EINVAL, varint_decode_error) \ x(EROFS, erofs_trans_commit) \ x(EROFS, erofs_no_writes) \ x(EROFS, erofs_journal_err) \ @@ -241,7 +248,10 @@ x(BCH_ERR_invalid_sb, invalid_sb_downgrade) \ x(BCH_ERR_invalid, invalid_bkey) \ x(BCH_ERR_operation_blocked, nocow_lock_blocked) \ + x(EIO, journal_shutdown) \ + x(EIO, journal_flush_err) \ x(EIO, btree_node_read_err) \ + x(BCH_ERR_btree_node_read_err, btree_node_read_err_cached) \ x(EIO, sb_not_downgraded) \ x(EIO, btree_node_write_all_failed) \ x(EIO, btree_node_read_error) \ @@ -257,6 +267,8 @@ x(EIO, no_device_to_read_from) \ x(EIO, missing_indirect_extent) \ x(EIO, invalidate_stripe_to_dev) \ + x(EIO, no_encryption_key) \ + x(EIO, insufficient_journal_devices) \ x(BCH_ERR_btree_node_read_err, btree_node_read_err_fixable) \ x(BCH_ERR_btree_node_read_err, btree_node_read_err_want_retry) \ x(BCH_ERR_btree_node_read_err, btree_node_read_err_must_retry) \ @@ -305,6 +317,7 @@ static inline long bch2_err_class(long err) #define BLK_STS_REMOVED ((__force blk_status_t)128) +#include <linux/blk_types.h> const char *bch2_blk_status_to_str(blk_status_t); #endif /* _BCACHFES_ERRCODE_H */ diff --git a/fs/bcachefs/error.c b/fs/bcachefs/error.c index b679def8fb98..038da6a61f6b 100644 --- a/fs/bcachefs/error.c +++ b/fs/bcachefs/error.c @@ -1,7 +1,9 @@ // SPDX-License-Identifier: GPL-2.0 #include "bcachefs.h" +#include "btree_cache.h" #include "btree_iter.h" #include "error.h" +#include "fs-common.h" #include "journal.h" #include "recovery_passes.h" #include "super.h" @@ -33,7 +35,7 @@ bool bch2_inconsistent_error(struct bch_fs *c) int bch2_topology_error(struct bch_fs *c) { set_bit(BCH_FS_topology_error, &c->flags); - if (!test_bit(BCH_FS_fsck_running, &c->flags)) { + if (!test_bit(BCH_FS_recovery_running, &c->flags)) { bch2_inconsistent_error(c); return -BCH_ERR_btree_need_topology_repair; } else { @@ -218,6 +220,30 @@ static const u8 fsck_flags_extra[] = { #undef x }; +static int do_fsck_ask_yn(struct bch_fs *c, + struct btree_trans *trans, + struct printbuf *question, + const char *action) +{ + prt_str(question, ", "); + prt_str(question, action); + + if (bch2_fs_stdio_redirect(c)) + bch2_print(c, "%s", question->buf); + else + bch2_print_string_as_lines(KERN_ERR, question->buf); + + int ask = bch2_fsck_ask_yn(c, trans); + + if (trans) { + int ret = bch2_trans_relock(trans); + if (ret) + return ret; + } + + return ask; +} + int __bch2_fsck_err(struct bch_fs *c, struct btree_trans *trans, enum bch_fsck_flags flags, @@ -226,7 +252,7 @@ int __bch2_fsck_err(struct bch_fs *c, { struct fsck_err_state *s = NULL; va_list args; - bool print = true, suppressing = false, inconsistent = false; + bool print = true, suppressing = false, inconsistent = false, exiting = false; struct printbuf buf = PRINTBUF, *out = &buf; int ret = -BCH_ERR_fsck_ignore; const char *action_orig = "fix?", *action = action_orig; @@ -256,9 +282,10 @@ int __bch2_fsck_err(struct bch_fs *c, !trans && bch2_current_has_btree_trans(c)); - if ((flags & FSCK_CAN_FIX) && - test_bit(err, c->sb.errors_silent)) - return -BCH_ERR_fsck_fix; + if (test_bit(err, c->sb.errors_silent)) + return flags & FSCK_CAN_FIX + ? -BCH_ERR_fsck_fix + : -BCH_ERR_fsck_ignore; bch2_sb_error_count(c, err); @@ -289,16 +316,14 @@ int __bch2_fsck_err(struct bch_fs *c, */ if (s->last_msg && !strcmp(buf.buf, s->last_msg)) { ret = s->ret; - mutex_unlock(&c->fsck_error_msgs_lock); - goto err; + goto err_unlock; } kfree(s->last_msg); s->last_msg = kstrdup(buf.buf, GFP_KERNEL); if (!s->last_msg) { - mutex_unlock(&c->fsck_error_msgs_lock); ret = -ENOMEM; - goto err; + goto err_unlock; } if (c->opts.ratelimit_errors && @@ -318,13 +343,19 @@ int __bch2_fsck_err(struct bch_fs *c, prt_printf(out, bch2_log_msg(c, "")); #endif - if ((flags & FSCK_CAN_FIX) && - (flags & FSCK_AUTOFIX) && + if ((flags & FSCK_AUTOFIX) && (c->opts.errors == BCH_ON_ERROR_continue || c->opts.errors == BCH_ON_ERROR_fix_safe)) { prt_str(out, ", "); - prt_actioning(out, action); - ret = -BCH_ERR_fsck_fix; + if (flags & FSCK_CAN_FIX) { + prt_actioning(out, action); + ret = -BCH_ERR_fsck_fix; + } else { + prt_str(out, ", continuing"); + ret = -BCH_ERR_fsck_ignore; + } + + goto print; } else if (!test_bit(BCH_FS_fsck_running, &c->flags)) { if (c->opts.errors != BCH_ON_ERROR_continue || !(flags & (FSCK_CAN_FIX|FSCK_CAN_IGNORE))) { @@ -348,31 +379,18 @@ int __bch2_fsck_err(struct bch_fs *c, : c->opts.fix_errors; if (fix == FSCK_FIX_ask) { - prt_str(out, ", "); - prt_str(out, action); - - if (bch2_fs_stdio_redirect(c)) - bch2_print(c, "%s", out->buf); - else - bch2_print_string_as_lines(KERN_ERR, out->buf); print = false; - int ask = bch2_fsck_ask_yn(c, trans); - - if (trans) { - ret = bch2_trans_relock(trans); - if (ret) { - mutex_unlock(&c->fsck_error_msgs_lock); - goto err; - } - } + ret = do_fsck_ask_yn(c, trans, out, action); + if (ret < 0) + goto err_unlock; - if (ask >= YN_ALLNO && s) - s->fix = ask == YN_ALLNO + if (ret >= YN_ALLNO && s) + s->fix = ret == YN_ALLNO ? FSCK_FIX_no : FSCK_FIX_yes; - ret = ask & 1 + ret = ret & 1 ? -BCH_ERR_fsck_fix : -BCH_ERR_fsck_ignore; } else if (fix == FSCK_FIX_yes || @@ -385,9 +403,7 @@ int __bch2_fsck_err(struct bch_fs *c, prt_str(out, ", not "); prt_actioning(out, action); } - } else if (flags & FSCK_NEED_FSCK) { - prt_str(out, " (run fsck to correct)"); - } else { + } else if (!(flags & FSCK_CAN_IGNORE)) { prt_str(out, " (repair unimplemented)"); } @@ -396,14 +412,13 @@ int __bch2_fsck_err(struct bch_fs *c, !(flags & FSCK_CAN_IGNORE))) ret = -BCH_ERR_fsck_errors_not_fixed; - bool exiting = - test_bit(BCH_FS_fsck_running, &c->flags) && - (ret != -BCH_ERR_fsck_fix && - ret != -BCH_ERR_fsck_ignore); - - if (exiting) + if (test_bit(BCH_FS_fsck_running, &c->flags) && + (ret != -BCH_ERR_fsck_fix && + ret != -BCH_ERR_fsck_ignore)) { + exiting = true; print = true; - + } +print: if (print) { if (bch2_fs_stdio_redirect(c)) bch2_print(c, "%s\n", out->buf); @@ -419,17 +434,24 @@ int __bch2_fsck_err(struct bch_fs *c, if (s) s->ret = ret; - mutex_unlock(&c->fsck_error_msgs_lock); - if (inconsistent) bch2_inconsistent_error(c); - if (ret == -BCH_ERR_fsck_fix) { - set_bit(BCH_FS_errors_fixed, &c->flags); - } else { - set_bit(BCH_FS_errors_not_fixed, &c->flags); - set_bit(BCH_FS_error, &c->flags); + /* + * We don't yet track whether the filesystem currently has errors, for + * log_fsck_err()s: that would require us to track for every error type + * which recovery pass corrects it, to get the fsck exit status correct: + */ + if (flags & FSCK_CAN_FIX) { + if (ret == -BCH_ERR_fsck_fix) { + set_bit(BCH_FS_errors_fixed, &c->flags); + } else { + set_bit(BCH_FS_errors_not_fixed, &c->flags); + set_bit(BCH_FS_error, &c->flags); + } } +err_unlock: + mutex_unlock(&c->fsck_error_msgs_lock); err: if (action != action_orig) kfree(action); @@ -437,28 +459,52 @@ err: return ret; } +static const char * const bch2_bkey_validate_contexts[] = { +#define x(n) #n, + BKEY_VALIDATE_CONTEXTS() +#undef x + NULL +}; + int __bch2_bkey_fsck_err(struct bch_fs *c, struct bkey_s_c k, - enum bch_validate_flags validate_flags, + struct bkey_validate_context from, enum bch_sb_error_id err, const char *fmt, ...) { - if (validate_flags & BCH_VALIDATE_silent) + if (from.flags & BCH_VALIDATE_silent) return -BCH_ERR_fsck_delete_bkey; unsigned fsck_flags = 0; - if (!(validate_flags & (BCH_VALIDATE_write|BCH_VALIDATE_commit))) + if (!(from.flags & (BCH_VALIDATE_write|BCH_VALIDATE_commit))) { + if (test_bit(err, c->sb.errors_silent)) + return -BCH_ERR_fsck_delete_bkey; + fsck_flags |= FSCK_AUTOFIX|FSCK_CAN_FIX; + } + if (!WARN_ON(err >= ARRAY_SIZE(fsck_flags_extra))) + fsck_flags |= fsck_flags_extra[err]; struct printbuf buf = PRINTBUF; - va_list args; + prt_printf(&buf, "invalid bkey in %s", + bch2_bkey_validate_contexts[from.from]); + + if (from.from == BKEY_VALIDATE_journal) + prt_printf(&buf, " journal seq=%llu offset=%u", + from.journal_seq, from.journal_offset); + + prt_str(&buf, " btree="); + bch2_btree_id_to_text(&buf, from.btree); + prt_printf(&buf, " level=%u: ", from.level); - prt_str(&buf, "invalid bkey "); bch2_bkey_val_to_text(&buf, c, k); prt_str(&buf, "\n "); + + va_list args; va_start(args, fmt); prt_vprintf(&buf, fmt, args); va_end(args); + prt_str(&buf, ": delete?"); int ret = __bch2_fsck_err(c, NULL, fsck_flags, err, "%s", buf.buf); @@ -483,3 +529,36 @@ void bch2_flush_fsck_errs(struct bch_fs *c) mutex_unlock(&c->fsck_error_msgs_lock); } + +int bch2_inum_err_msg_trans(struct btree_trans *trans, struct printbuf *out, subvol_inum inum) +{ + u32 restart_count = trans->restart_count; + int ret = 0; + + /* XXX: we don't yet attempt to print paths when we don't know the subvol */ + if (inum.subvol) + ret = lockrestart_do(trans, bch2_inum_to_path(trans, inum, out)); + if (!inum.subvol || ret) + prt_printf(out, "inum %llu:%llu", inum.subvol, inum.inum); + + return trans_was_restarted(trans, restart_count); +} + +int bch2_inum_offset_err_msg_trans(struct btree_trans *trans, struct printbuf *out, + subvol_inum inum, u64 offset) +{ + int ret = bch2_inum_err_msg_trans(trans, out, inum); + prt_printf(out, " offset %llu: ", offset); + return ret; +} + +void bch2_inum_err_msg(struct bch_fs *c, struct printbuf *out, subvol_inum inum) +{ + bch2_trans_run(c, bch2_inum_err_msg_trans(trans, out, inum)); +} + +void bch2_inum_offset_err_msg(struct bch_fs *c, struct printbuf *out, + subvol_inum inum, u64 offset) +{ + bch2_trans_run(c, bch2_inum_offset_err_msg_trans(trans, out, inum, offset)); +} diff --git a/fs/bcachefs/error.h b/fs/bcachefs/error.h index 6551ada926b6..7acf2a27ca28 100644 --- a/fs/bcachefs/error.h +++ b/fs/bcachefs/error.h @@ -45,32 +45,11 @@ int bch2_topology_error(struct bch_fs *); bch2_inconsistent_error(c); \ }) -#define bch2_fs_inconsistent_on(cond, c, ...) \ +#define bch2_fs_inconsistent_on(cond, ...) \ ({ \ bool _ret = unlikely(!!(cond)); \ - \ - if (_ret) \ - bch2_fs_inconsistent(c, __VA_ARGS__); \ - _ret; \ -}) - -/* - * Later we might want to mark only the particular device inconsistent, not the - * entire filesystem: - */ - -#define bch2_dev_inconsistent(ca, ...) \ -do { \ - bch_err(ca, __VA_ARGS__); \ - bch2_inconsistent_error((ca)->fs); \ -} while (0) - -#define bch2_dev_inconsistent_on(cond, ca, ...) \ -({ \ - bool _ret = unlikely(!!(cond)); \ - \ if (_ret) \ - bch2_dev_inconsistent(ca, __VA_ARGS__); \ + bch2_fs_inconsistent(__VA_ARGS__); \ _ret; \ }) @@ -123,9 +102,9 @@ int __bch2_fsck_err(struct bch_fs *, struct btree_trans *, void bch2_flush_fsck_errs(struct bch_fs *); -#define __fsck_err(c, _flags, _err_type, ...) \ +#define fsck_err_wrap(_do) \ ({ \ - int _ret = bch2_fsck_err(c, _flags, _err_type, __VA_ARGS__); \ + int _ret = _do; \ if (_ret != -BCH_ERR_fsck_fix && \ _ret != -BCH_ERR_fsck_ignore) { \ ret = _ret; \ @@ -135,6 +114,8 @@ void bch2_flush_fsck_errs(struct bch_fs *); _ret == -BCH_ERR_fsck_fix; \ }) +#define __fsck_err(...) fsck_err_wrap(bch2_fsck_err(__VA_ARGS__)) + /* These macros return true if error should be fixed: */ /* XXX: mark in superblock that filesystem contains errors, if we ignore: */ @@ -149,12 +130,6 @@ void bch2_flush_fsck_errs(struct bch_fs *); (unlikely(cond) ? __fsck_err(c, _flags, _err_type, __VA_ARGS__) : false);\ }) -#define need_fsck_err_on(cond, c, _err_type, ...) \ - __fsck_err_on(cond, c, FSCK_CAN_IGNORE|FSCK_NEED_FSCK, _err_type, __VA_ARGS__) - -#define need_fsck_err(c, _err_type, ...) \ - __fsck_err(c, FSCK_CAN_IGNORE|FSCK_NEED_FSCK, _err_type, __VA_ARGS__) - #define mustfix_fsck_err(c, _err_type, ...) \ __fsck_err(c, FSCK_CAN_FIX, _err_type, __VA_ARGS__) @@ -167,11 +142,22 @@ void bch2_flush_fsck_errs(struct bch_fs *); #define fsck_err_on(cond, c, _err_type, ...) \ __fsck_err_on(cond, c, FSCK_CAN_FIX|FSCK_CAN_IGNORE, _err_type, __VA_ARGS__) +#define log_fsck_err(c, _err_type, ...) \ + __fsck_err(c, FSCK_CAN_IGNORE, _err_type, __VA_ARGS__) + +#define log_fsck_err_on(cond, ...) \ +({ \ + bool _ret = unlikely(!!(cond)); \ + if (_ret) \ + log_fsck_err(__VA_ARGS__); \ + _ret; \ +}) + enum bch_validate_flags; __printf(5, 6) int __bch2_bkey_fsck_err(struct bch_fs *, struct bkey_s_c, - enum bch_validate_flags, + struct bkey_validate_context from, enum bch_sb_error_id, const char *, ...); @@ -181,7 +167,7 @@ int __bch2_bkey_fsck_err(struct bch_fs *, */ #define bkey_fsck_err(c, _err_type, _err_msg, ...) \ do { \ - int _ret = __bch2_bkey_fsck_err(c, k, flags, \ + int _ret = __bch2_bkey_fsck_err(c, k, from, \ BCH_FSCK_ERR_##_err_type, \ _err_msg, ##__VA_ARGS__); \ if (_ret != -BCH_ERR_fsck_fix && \ @@ -252,4 +238,10 @@ void bch2_io_error(struct bch_dev *, enum bch_member_error_type); _ret; \ }) +int bch2_inum_err_msg_trans(struct btree_trans *, struct printbuf *, subvol_inum); +int bch2_inum_offset_err_msg_trans(struct btree_trans *, struct printbuf *, subvol_inum, u64); + +void bch2_inum_err_msg(struct bch_fs *, struct printbuf *, subvol_inum); +void bch2_inum_offset_err_msg(struct bch_fs *, struct printbuf *, subvol_inum, u64); + #endif /* _BCACHEFS_ERROR_H */ diff --git a/fs/bcachefs/extent_update.c b/fs/bcachefs/extent_update.c index 5f4fecb358da..6aac579a692a 100644 --- a/fs/bcachefs/extent_update.c +++ b/fs/bcachefs/extent_update.c @@ -64,7 +64,7 @@ static int count_iters_for_insert(struct btree_trans *trans, break; case KEY_TYPE_reflink_p: { struct bkey_s_c_reflink_p p = bkey_s_c_to_reflink_p(k); - u64 idx = le64_to_cpu(p.v->idx); + u64 idx = REFLINK_P_IDX(p.v); unsigned sectors = bpos_min(*end, p.k->p).offset - bkey_start_offset(p.k); struct btree_iter iter; @@ -128,7 +128,7 @@ int bch2_extent_atomic_end(struct btree_trans *trans, bch2_trans_copy_iter(©, iter); - for_each_btree_key_upto_continue_norestart(copy, insert->k.p, 0, k, ret) { + for_each_btree_key_max_continue_norestart(copy, insert->k.p, 0, k, ret) { unsigned offset = 0; if (bkey_gt(bkey_start_pos(&insert->k), bkey_start_pos(k.k))) diff --git a/fs/bcachefs/extents.c b/fs/bcachefs/extents.c index 37e3d69bec06..05d5f71a7ca9 100644 --- a/fs/bcachefs/extents.c +++ b/fs/bcachefs/extents.c @@ -21,6 +21,7 @@ #include "extents.h" #include "inode.h" #include "journal.h" +#include "rebalance.h" #include "replicas.h" #include "super.h" #include "super-io.h" @@ -88,6 +89,14 @@ static inline bool ptr_better(struct bch_fs *c, u64 l1 = dev_latency(c, p1.ptr.dev); u64 l2 = dev_latency(c, p2.ptr.dev); + /* + * Square the latencies, to bias more in favor of the faster + * device - we never want to stop issuing reads to the slower + * device altogether, so that we can update our latency numbers: + */ + l1 *= l1; + l2 *= l2; + /* Pick at random, biased in favor of the faster device: */ return bch2_rand_range(l1 + l2) > l1; @@ -169,7 +178,7 @@ int bch2_bkey_pick_read_device(struct bch_fs *c, struct bkey_s_c k, /* KEY_TYPE_btree_ptr: */ int bch2_btree_ptr_validate(struct bch_fs *c, struct bkey_s_c k, - enum bch_validate_flags flags) + struct bkey_validate_context from) { int ret = 0; @@ -177,7 +186,7 @@ int bch2_btree_ptr_validate(struct bch_fs *c, struct bkey_s_c k, c, btree_ptr_val_too_big, "value too big (%zu > %u)", bkey_val_u64s(k.k), BCH_REPLICAS_MAX); - ret = bch2_bkey_ptrs_validate(c, k, flags); + ret = bch2_bkey_ptrs_validate(c, k, from); fsck_err: return ret; } @@ -189,7 +198,7 @@ void bch2_btree_ptr_to_text(struct printbuf *out, struct bch_fs *c, } int bch2_btree_ptr_v2_validate(struct bch_fs *c, struct bkey_s_c k, - enum bch_validate_flags flags) + struct bkey_validate_context from) { struct bkey_s_c_btree_ptr_v2 bp = bkey_s_c_to_btree_ptr_v2(k); int ret = 0; @@ -203,12 +212,13 @@ int bch2_btree_ptr_v2_validate(struct bch_fs *c, struct bkey_s_c k, c, btree_ptr_v2_min_key_bad, "min_key > key"); - if (flags & BCH_VALIDATE_write) + if ((from.flags & BCH_VALIDATE_write) && + c->sb.version_min >= bcachefs_metadata_version_btree_ptr_sectors_written) bkey_fsck_err_on(!bp.v->sectors_written, c, btree_ptr_v2_written_0, "sectors_written == 0"); - ret = bch2_bkey_ptrs_validate(c, k, flags); + ret = bch2_bkey_ptrs_validate(c, k, from); fsck_err: return ret; } @@ -395,7 +405,7 @@ bool bch2_extent_merge(struct bch_fs *c, struct bkey_s l, struct bkey_s_c r) /* KEY_TYPE_reservation: */ int bch2_reservation_validate(struct bch_fs *c, struct bkey_s_c k, - enum bch_validate_flags flags) + struct bkey_validate_context from) { struct bkey_s_c_reservation r = bkey_s_c_to_reservation(k); int ret = 0; @@ -1120,6 +1130,57 @@ void bch2_extent_crc_unpacked_to_text(struct printbuf *out, struct bch_extent_cr bch2_prt_compression_type(out, crc->compression_type); } +static void bch2_extent_rebalance_to_text(struct printbuf *out, struct bch_fs *c, + const struct bch_extent_rebalance *r) +{ + prt_str(out, "rebalance:"); + + prt_printf(out, " replicas=%u", r->data_replicas); + if (r->data_replicas_from_inode) + prt_str(out, " (inode)"); + + prt_str(out, " checksum="); + bch2_prt_csum_opt(out, r->data_checksum); + if (r->data_checksum_from_inode) + prt_str(out, " (inode)"); + + if (r->background_compression || r->background_compression_from_inode) { + prt_str(out, " background_compression="); + bch2_compression_opt_to_text(out, r->background_compression); + + if (r->background_compression_from_inode) + prt_str(out, " (inode)"); + } + + if (r->background_target || r->background_target_from_inode) { + prt_str(out, " background_target="); + if (c) + bch2_target_to_text(out, c, r->background_target); + else + prt_printf(out, "%u", r->background_target); + + if (r->background_target_from_inode) + prt_str(out, " (inode)"); + } + + if (r->promote_target || r->promote_target_from_inode) { + prt_str(out, " promote_target="); + if (c) + bch2_target_to_text(out, c, r->promote_target); + else + prt_printf(out, "%u", r->promote_target); + + if (r->promote_target_from_inode) + prt_str(out, " (inode)"); + } + + if (r->erasure_code || r->erasure_code_from_inode) { + prt_printf(out, " ec=%u", r->erasure_code); + if (r->erasure_code_from_inode) + prt_str(out, " (inode)"); + } +} + void bch2_bkey_ptrs_to_text(struct printbuf *out, struct bch_fs *c, struct bkey_s_c k) { @@ -1155,18 +1216,10 @@ void bch2_bkey_ptrs_to_text(struct printbuf *out, struct bch_fs *c, (u64) ec->idx, ec->block); break; } - case BCH_EXTENT_ENTRY_rebalance: { - const struct bch_extent_rebalance *r = &entry->rebalance; - - prt_str(out, "rebalance: target "); - if (c) - bch2_target_to_text(out, c, r->target); - else - prt_printf(out, "%u", r->target); - prt_str(out, " compression "); - bch2_compression_opt_to_text(out, r->compression); + case BCH_EXTENT_ENTRY_rebalance: + bch2_extent_rebalance_to_text(out, c, &entry->rebalance); break; - } + default: prt_printf(out, "(invalid extent entry %.16llx)", *((u64 *) entry)); return; @@ -1178,13 +1231,19 @@ void bch2_bkey_ptrs_to_text(struct printbuf *out, struct bch_fs *c, static int extent_ptr_validate(struct bch_fs *c, struct bkey_s_c k, - enum bch_validate_flags flags, + struct bkey_validate_context from, const struct bch_extent_ptr *ptr, unsigned size_ondisk, bool metadata) { int ret = 0; + struct bkey_ptrs_c ptrs = bch2_bkey_ptrs_c(k); + bkey_for_each_ptr(ptrs, ptr2) + bkey_fsck_err_on(ptr != ptr2 && ptr->dev == ptr2->dev, + c, ptr_to_duplicate_device, + "multiple pointers to same device (%u)", ptr->dev); + /* bad pointers are repaired by check_fix_ptrs(): */ rcu_read_lock(); struct bch_dev *ca = bch2_dev_rcu_noerror(c, ptr->dev); @@ -1199,13 +1258,6 @@ static int extent_ptr_validate(struct bch_fs *c, unsigned bucket_size = ca->mi.bucket_size; rcu_read_unlock(); - struct bkey_ptrs_c ptrs = bch2_bkey_ptrs_c(k); - bkey_for_each_ptr(ptrs, ptr2) - bkey_fsck_err_on(ptr != ptr2 && ptr->dev == ptr2->dev, - c, ptr_to_duplicate_device, - "multiple pointers to same device (%u)", ptr->dev); - - bkey_fsck_err_on(bucket >= nbuckets, c, ptr_after_last_bucket, "pointer past last bucket (%llu > %llu)", bucket, nbuckets); @@ -1221,7 +1273,7 @@ fsck_err: } int bch2_bkey_ptrs_validate(struct bch_fs *c, struct bkey_s_c k, - enum bch_validate_flags flags) + struct bkey_validate_context from) { struct bkey_ptrs_c ptrs = bch2_bkey_ptrs_c(k); const union bch_extent_entry *entry; @@ -1248,7 +1300,7 @@ int bch2_bkey_ptrs_validate(struct bch_fs *c, struct bkey_s_c k, switch (extent_entry_type(entry)) { case BCH_EXTENT_ENTRY_ptr: - ret = extent_ptr_validate(c, k, flags, &entry->ptr, size_ondisk, false); + ret = extent_ptr_validate(c, k, from, &entry->ptr, size_ondisk, false); if (ret) return ret; @@ -1270,9 +1322,6 @@ int bch2_bkey_ptrs_validate(struct bch_fs *c, struct bkey_s_c k, case BCH_EXTENT_ENTRY_crc128: crc = bch2_extent_crc_unpack(k.k, entry_to_crc(entry)); - bkey_fsck_err_on(crc.offset + crc.live_size > crc.uncompressed_size, - c, ptr_crc_uncompressed_size_too_small, - "checksum offset + key size > uncompressed size"); bkey_fsck_err_on(!bch2_checksum_type_valid(c, crc.csum_type), c, ptr_crc_csum_type_unknown, "invalid checksum type"); @@ -1280,6 +1329,19 @@ int bch2_bkey_ptrs_validate(struct bch_fs *c, struct bkey_s_c k, c, ptr_crc_compression_type_unknown, "invalid compression type"); + bkey_fsck_err_on(crc.offset + crc.live_size > crc.uncompressed_size, + c, ptr_crc_uncompressed_size_too_small, + "checksum offset + key size > uncompressed size"); + bkey_fsck_err_on(crc_is_encoded(crc) && + (crc.uncompressed_size > c->opts.encoded_extent_max >> 9) && + (from.flags & (BCH_VALIDATE_write|BCH_VALIDATE_commit)), + c, ptr_crc_uncompressed_size_too_big, + "too large encoded extent"); + bkey_fsck_err_on(!crc_is_compressed(crc) && + crc.compressed_size != crc.uncompressed_size, + c, ptr_crc_uncompressed_size_mismatch, + "not compressed but compressed != uncompressed size"); + if (bch2_csum_type_is_encryption(crc.csum_type)) { if (nonce == UINT_MAX) nonce = crc.offset + crc.nonce; @@ -1293,12 +1355,6 @@ int bch2_bkey_ptrs_validate(struct bch_fs *c, struct bkey_s_c k, "redundant crc entry"); crc_since_last_ptr = true; - bkey_fsck_err_on(crc_is_encoded(crc) && - (crc.uncompressed_size > c->opts.encoded_extent_max >> 9) && - (flags & (BCH_VALIDATE_write|BCH_VALIDATE_commit)), - c, ptr_crc_uncompressed_size_too_big, - "too large encoded extent"); - size_ondisk = crc.compressed_size; break; case BCH_EXTENT_ENTRY_stripe_ptr: @@ -1391,166 +1447,6 @@ void bch2_ptr_swab(struct bkey_s k) } } -const struct bch_extent_rebalance *bch2_bkey_rebalance_opts(struct bkey_s_c k) -{ - struct bkey_ptrs_c ptrs = bch2_bkey_ptrs_c(k); - const union bch_extent_entry *entry; - - bkey_extent_entry_for_each(ptrs, entry) - if (__extent_entry_type(entry) == BCH_EXTENT_ENTRY_rebalance) - return &entry->rebalance; - - return NULL; -} - -unsigned bch2_bkey_ptrs_need_rebalance(struct bch_fs *c, struct bkey_s_c k, - unsigned target, unsigned compression) -{ - struct bkey_ptrs_c ptrs = bch2_bkey_ptrs_c(k); - unsigned rewrite_ptrs = 0; - - if (compression) { - unsigned compression_type = bch2_compression_opt_to_type(compression); - const union bch_extent_entry *entry; - struct extent_ptr_decoded p; - unsigned i = 0; - - bkey_for_each_ptr_decode(k.k, ptrs, p, entry) { - if (p.crc.compression_type == BCH_COMPRESSION_TYPE_incompressible || - p.ptr.unwritten) { - rewrite_ptrs = 0; - goto incompressible; - } - - if (!p.ptr.cached && p.crc.compression_type != compression_type) - rewrite_ptrs |= 1U << i; - i++; - } - } -incompressible: - if (target && bch2_target_accepts_data(c, BCH_DATA_user, target)) { - unsigned i = 0; - - bkey_for_each_ptr(ptrs, ptr) { - if (!ptr->cached && !bch2_dev_in_target(c, ptr->dev, target)) - rewrite_ptrs |= 1U << i; - i++; - } - } - - return rewrite_ptrs; -} - -bool bch2_bkey_needs_rebalance(struct bch_fs *c, struct bkey_s_c k) -{ - const struct bch_extent_rebalance *r = bch2_bkey_rebalance_opts(k); - - /* - * If it's an indirect extent, we don't delete the rebalance entry when - * done so that we know what options were applied - check if it still - * needs work done: - */ - if (r && - k.k->type == KEY_TYPE_reflink_v && - !bch2_bkey_ptrs_need_rebalance(c, k, r->target, r->compression)) - r = NULL; - - return r != NULL; -} - -static u64 __bch2_bkey_sectors_need_rebalance(struct bch_fs *c, struct bkey_s_c k, - unsigned target, unsigned compression) -{ - struct bkey_ptrs_c ptrs = bch2_bkey_ptrs_c(k); - const union bch_extent_entry *entry; - struct extent_ptr_decoded p; - u64 sectors = 0; - - if (compression) { - unsigned compression_type = bch2_compression_opt_to_type(compression); - - bkey_for_each_ptr_decode(k.k, ptrs, p, entry) { - if (p.crc.compression_type == BCH_COMPRESSION_TYPE_incompressible || - p.ptr.unwritten) { - sectors = 0; - goto incompressible; - } - - if (!p.ptr.cached && p.crc.compression_type != compression_type) - sectors += p.crc.compressed_size; - } - } -incompressible: - if (target && bch2_target_accepts_data(c, BCH_DATA_user, target)) { - bkey_for_each_ptr_decode(k.k, ptrs, p, entry) - if (!p.ptr.cached && !bch2_dev_in_target(c, p.ptr.dev, target)) - sectors += p.crc.compressed_size; - } - - return sectors; -} - -u64 bch2_bkey_sectors_need_rebalance(struct bch_fs *c, struct bkey_s_c k) -{ - const struct bch_extent_rebalance *r = bch2_bkey_rebalance_opts(k); - - return r ? __bch2_bkey_sectors_need_rebalance(c, k, r->target, r->compression) : 0; -} - -int bch2_bkey_set_needs_rebalance(struct bch_fs *c, struct bkey_i *_k, - struct bch_io_opts *opts) -{ - struct bkey_s k = bkey_i_to_s(_k); - struct bch_extent_rebalance *r; - unsigned target = opts->background_target; - unsigned compression = background_compression(*opts); - bool needs_rebalance; - - if (!bkey_extent_is_direct_data(k.k)) - return 0; - - /* get existing rebalance entry: */ - r = (struct bch_extent_rebalance *) bch2_bkey_rebalance_opts(k.s_c); - if (r) { - if (k.k->type == KEY_TYPE_reflink_v) { - /* - * indirect extents: existing options take precedence, - * so that we don't move extents back and forth if - * they're referenced by different inodes with different - * options: - */ - if (r->target) - target = r->target; - if (r->compression) - compression = r->compression; - } - - r->target = target; - r->compression = compression; - } - - needs_rebalance = bch2_bkey_ptrs_need_rebalance(c, k.s_c, target, compression); - - if (needs_rebalance && !r) { - union bch_extent_entry *new = bkey_val_end(k); - - new->rebalance.type = 1U << BCH_EXTENT_ENTRY_rebalance; - new->rebalance.compression = compression; - new->rebalance.target = target; - new->rebalance.unused = 0; - k.k->u64s += extent_entry_u64s(new); - } else if (!needs_rebalance && r && k.k->type != KEY_TYPE_reflink_v) { - /* - * For indirect extents, don't delete the rebalance entry when - * we're finished so that we know we specifically moved it or - * compressed it to its current location/compression type - */ - extent_entry_drop(k, (union bch_extent_entry *) r); - } - - return 0; -} - /* Generic extent code: */ int bch2_cut_front_s(struct bpos where, struct bkey_s k) @@ -1610,7 +1506,7 @@ int bch2_cut_front_s(struct bpos where, struct bkey_s k) case KEY_TYPE_reflink_p: { struct bkey_s_reflink_p p = bkey_s_to_reflink_p(k); - le64_add_cpu(&p.v->idx, sub); + SET_REFLINK_P_IDX(p.v, REFLINK_P_IDX(p.v) + sub); break; } case KEY_TYPE_inline_data: diff --git a/fs/bcachefs/extents.h b/fs/bcachefs/extents.h index bcffcf60aaaf..620b284aa34f 100644 --- a/fs/bcachefs/extents.h +++ b/fs/bcachefs/extents.h @@ -8,7 +8,6 @@ struct bch_fs; struct btree_trans; -enum bch_validate_flags; /* extent entries: */ @@ -410,12 +409,12 @@ int bch2_bkey_pick_read_device(struct bch_fs *, struct bkey_s_c, /* KEY_TYPE_btree_ptr: */ int bch2_btree_ptr_validate(struct bch_fs *, struct bkey_s_c, - enum bch_validate_flags); + struct bkey_validate_context); void bch2_btree_ptr_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c); int bch2_btree_ptr_v2_validate(struct bch_fs *, struct bkey_s_c, - enum bch_validate_flags); + struct bkey_validate_context); void bch2_btree_ptr_v2_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c); void bch2_btree_ptr_v2_compat(enum btree_id, unsigned, unsigned, int, struct bkey_s); @@ -452,7 +451,7 @@ bool bch2_extent_merge(struct bch_fs *, struct bkey_s, struct bkey_s_c); /* KEY_TYPE_reservation: */ int bch2_reservation_validate(struct bch_fs *, struct bkey_s_c, - enum bch_validate_flags); + struct bkey_validate_context); void bch2_reservation_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c); bool bch2_reservation_merge(struct bch_fs *, struct bkey_s, struct bkey_s_c); @@ -696,7 +695,7 @@ void bch2_extent_ptr_to_text(struct printbuf *out, struct bch_fs *, const struct void bch2_bkey_ptrs_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c); int bch2_bkey_ptrs_validate(struct bch_fs *, struct bkey_s_c, - enum bch_validate_flags); + struct bkey_validate_context); static inline bool bch2_extent_ptr_eq(struct bch_extent_ptr ptr1, struct bch_extent_ptr ptr2) @@ -710,15 +709,6 @@ static inline bool bch2_extent_ptr_eq(struct bch_extent_ptr ptr1, void bch2_ptr_swab(struct bkey_s); -const struct bch_extent_rebalance *bch2_bkey_rebalance_opts(struct bkey_s_c); -unsigned bch2_bkey_ptrs_need_rebalance(struct bch_fs *, struct bkey_s_c, - unsigned, unsigned); -bool bch2_bkey_needs_rebalance(struct bch_fs *, struct bkey_s_c); -u64 bch2_bkey_sectors_need_rebalance(struct bch_fs *, struct bkey_s_c); - -int bch2_bkey_set_needs_rebalance(struct bch_fs *, struct bkey_i *, - struct bch_io_opts *); - /* Generic extent code: */ enum bch_extent_overlap { diff --git a/fs/bcachefs/extents_format.h b/fs/bcachefs/extents_format.h index 3bd2fdbb0817..c198dfc376d6 100644 --- a/fs/bcachefs/extents_format.h +++ b/fs/bcachefs/extents_format.h @@ -201,19 +201,8 @@ struct bch_extent_stripe_ptr { #endif }; -struct bch_extent_rebalance { -#if defined(__LITTLE_ENDIAN_BITFIELD) - __u64 type:6, - unused:34, - compression:8, /* enum bch_compression_opt */ - target:16; -#elif defined (__BIG_ENDIAN_BITFIELD) - __u64 target:16, - compression:8, - unused:34, - type:6; -#endif -}; +/* bch_extent_rebalance: */ +#include "rebalance_format.h" union bch_extent_entry { #if __BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__ || __BITS_PER_LONG == 64 diff --git a/fs/bcachefs/fs-common.c b/fs/bcachefs/fs-common.c index 7e10a9ddcfd9..d70d9f634cea 100644 --- a/fs/bcachefs/fs-common.c +++ b/fs/bcachefs/fs-common.c @@ -69,9 +69,7 @@ int bch2_create_trans(struct btree_trans *trans, if (!snapshot_src.inum) { /* Inode wasn't specified, just snapshot: */ struct bch_subvolume s; - - ret = bch2_subvolume_get(trans, snapshot_src.subvol, true, - BTREE_ITER_cached, &s); + ret = bch2_subvolume_get(trans, snapshot_src.subvol, true, &s); if (ret) goto err; @@ -154,6 +152,7 @@ int bch2_create_trans(struct btree_trans *trans, if (is_subdir_for_nlink(new_inode)) dir_u->bi_nlink++; dir_u->bi_mtime = dir_u->bi_ctime = now; + dir_u->bi_size += dirent_occupied_size(name); ret = bch2_inode_write(trans, &dir_iter, dir_u); if (ret) @@ -172,6 +171,10 @@ int bch2_create_trans(struct btree_trans *trans, new_inode->bi_dir_offset = dir_offset; } + if (S_ISDIR(mode) && + !new_inode->bi_subvol) + new_inode->bi_depth = dir_u->bi_depth + 1; + inode_iter.flags &= ~BTREE_ITER_all_snapshots; bch2_btree_iter_set_snapshot(&inode_iter, snapshot); @@ -218,6 +221,7 @@ int bch2_link_trans(struct btree_trans *trans, } dir_u->bi_mtime = dir_u->bi_ctime = now; + dir_u->bi_size += dirent_occupied_size(name); dir_hash = bch2_hash_info_init(c, dir_u); @@ -320,6 +324,7 @@ int bch2_unlink_trans(struct btree_trans *trans, dir_u->bi_mtime = dir_u->bi_ctime = inode_u->bi_ctime = now; dir_u->bi_nlink -= is_subdir_for_nlink(inode_u); + dir_u->bi_size -= dirent_occupied_size(name); ret = bch2_hash_delete_at(trans, bch2_dirent_hash_desc, &dir_hash, &dirent_iter, @@ -458,6 +463,14 @@ int bch2_rename_trans(struct btree_trans *trans, goto err; } + if (mode == BCH_RENAME) { + src_dir_u->bi_size -= dirent_occupied_size(src_name); + dst_dir_u->bi_size += dirent_occupied_size(dst_name); + } + + if (mode == BCH_RENAME_OVERWRITE) + src_dir_u->bi_size -= dirent_occupied_size(src_name); + if (src_inode_u->bi_parent_subvol) src_inode_u->bi_parent_subvol = dst_dir.subvol; @@ -512,6 +525,15 @@ int bch2_rename_trans(struct btree_trans *trans, dst_dir_u->bi_nlink++; } + if (S_ISDIR(src_inode_u->bi_mode) && + !src_inode_u->bi_subvol) + src_inode_u->bi_depth = dst_dir_u->bi_depth + 1; + + if (mode == BCH_RENAME_EXCHANGE && + S_ISDIR(dst_inode_u->bi_mode) && + !dst_inode_u->bi_subvol) + dst_inode_u->bi_depth = src_dir_u->bi_depth + 1; + if (dst_inum.inum && is_subdir_for_nlink(dst_inode_u)) { dst_dir_u->bi_nlink--; src_dir_u->bi_nlink += mode == BCH_RENAME_EXCHANGE; @@ -548,3 +570,94 @@ err: bch2_trans_iter_exit(trans, &src_dir_iter); return ret; } + +static inline void prt_bytes_reversed(struct printbuf *out, const void *b, unsigned n) +{ + bch2_printbuf_make_room(out, n); + + unsigned can_print = min(n, printbuf_remaining(out)); + + b += n; + + for (unsigned i = 0; i < can_print; i++) + out->buf[out->pos++] = *((char *) --b); + + printbuf_nul_terminate(out); +} + +static inline void prt_str_reversed(struct printbuf *out, const char *s) +{ + prt_bytes_reversed(out, s, strlen(s)); +} + +static inline void reverse_bytes(void *b, size_t n) +{ + char *e = b + n, *s = b; + + while (s < e) { + --e; + swap(*s, *e); + s++; + } +} + +/* XXX: we don't yet attempt to print paths when we don't know the subvol */ +int bch2_inum_to_path(struct btree_trans *trans, subvol_inum inum, struct printbuf *path) +{ + unsigned orig_pos = path->pos; + int ret = 0; + + while (!(inum.subvol == BCACHEFS_ROOT_SUBVOL && + inum.inum == BCACHEFS_ROOT_INO)) { + struct bch_inode_unpacked inode; + ret = bch2_inode_find_by_inum_trans(trans, inum, &inode); + if (ret) + goto disconnected; + + if (!inode.bi_dir && !inode.bi_dir_offset) { + ret = -BCH_ERR_ENOENT_inode_no_backpointer; + goto disconnected; + } + + inum.subvol = inode.bi_parent_subvol ?: inum.subvol; + inum.inum = inode.bi_dir; + + u32 snapshot; + ret = bch2_subvolume_get_snapshot(trans, inum.subvol, &snapshot); + if (ret) + goto disconnected; + + struct btree_iter d_iter; + struct bkey_s_c_dirent d = bch2_bkey_get_iter_typed(trans, &d_iter, + BTREE_ID_dirents, SPOS(inode.bi_dir, inode.bi_dir_offset, snapshot), + 0, dirent); + ret = bkey_err(d.s_c); + if (ret) + goto disconnected; + + struct qstr dirent_name = bch2_dirent_get_name(d); + prt_bytes_reversed(path, dirent_name.name, dirent_name.len); + + prt_char(path, '/'); + + bch2_trans_iter_exit(trans, &d_iter); + } + + if (orig_pos == path->pos) + prt_char(path, '/'); +out: + ret = path->allocation_failure ? -ENOMEM : 0; + if (ret) + goto err; + + reverse_bytes(path->buf + orig_pos, path->pos - orig_pos); + return 0; +err: + return ret; +disconnected: + if (bch2_err_matches(ret, BCH_ERR_transaction_restart)) + goto err; + + prt_str_reversed(path, "(disconnected)"); + goto out; +} diff --git a/fs/bcachefs/fs-common.h b/fs/bcachefs/fs-common.h index c934e807b380..2b59210bb5e8 100644 --- a/fs/bcachefs/fs-common.h +++ b/fs/bcachefs/fs-common.h @@ -42,4 +42,6 @@ int bch2_rename_trans(struct btree_trans *, bool bch2_reinherit_attrs(struct bch_inode_unpacked *, struct bch_inode_unpacked *); +int bch2_inum_to_path(struct btree_trans *, subvol_inum, struct printbuf *); + #endif /* _BCACHEFS_FS_COMMON_H */ diff --git a/fs/bcachefs/fs-io-buffered.c b/fs/bcachefs/fs-io-buffered.c index 95972809e76d..ab1d5db2fa56 100644 --- a/fs/bcachefs/fs-io-buffered.c +++ b/fs/bcachefs/fs-io-buffered.c @@ -164,7 +164,8 @@ static void bchfs_read(struct btree_trans *trans, BTREE_ITER_slots); while (1) { struct bkey_s_c k; - unsigned bytes, sectors, offset_into_extent; + unsigned bytes, sectors; + s64 offset_into_extent; enum btree_id data_btree = BTREE_ID_extents; bch2_trans_begin(trans); @@ -197,7 +198,7 @@ static void bchfs_read(struct btree_trans *trans, k = bkey_i_to_s_c(sk.k); - sectors = min(sectors, k.k->size - offset_into_extent); + sectors = min_t(unsigned, sectors, k.k->size - offset_into_extent); if (readpages_iter) { ret = readpage_bio_extend(trans, readpages_iter, &rbio->bio, sectors, @@ -230,10 +231,12 @@ err: bch2_trans_iter_exit(trans, &iter); if (ret) { - bch_err_inum_offset_ratelimited(c, - iter.pos.inode, - iter.pos.offset << 9, - "read error %i from btree lookup", ret); + struct printbuf buf = PRINTBUF; + bch2_inum_offset_err_msg_trans(trans, &buf, inum, iter.pos.offset << 9); + prt_printf(&buf, "read error %i from btree lookup", ret); + bch_err_ratelimited(c, "%s", buf.buf); + printbuf_exit(&buf); + rbio->bio.bi_status = BLK_STS_IOERR; bio_endio(&rbio->bio); } @@ -248,6 +251,7 @@ void bch2_readahead(struct readahead_control *ractl) struct bch_io_opts opts; struct folio *folio; struct readpages_iter readpages_iter; + struct blk_plug plug; bch2_inode_opts_get(&opts, c, &inode->ei_inode); @@ -255,6 +259,16 @@ void bch2_readahead(struct readahead_control *ractl) if (ret) return; + /* + * Besides being a general performance optimization, plugging helps with + * avoiding btree transaction srcu warnings - submitting a bio can + * block, and we don't want todo that with the transaction locked. + * + * However, plugged bios are submitted when we schedule; we ideally + * would have our own scheduler hook to call unlock_long() before + * scheduling. + */ + blk_start_plug(&plug); bch2_pagecache_add_get(inode); struct btree_trans *trans = bch2_trans_get(c); @@ -281,7 +295,7 @@ void bch2_readahead(struct readahead_control *ractl) bch2_trans_put(trans); bch2_pagecache_add_put(inode); - + blk_finish_plug(&plug); darray_exit(&readpages_iter.folios); } @@ -296,9 +310,13 @@ int bch2_read_single_folio(struct folio *folio, struct address_space *mapping) struct bch_fs *c = inode->v.i_sb->s_fs_info; struct bch_read_bio *rbio; struct bch_io_opts opts; + struct blk_plug plug; int ret; DECLARE_COMPLETION_ONSTACK(done); + BUG_ON(folio_test_uptodate(folio)); + BUG_ON(folio_test_dirty(folio)); + if (!bch2_folio_create(folio, GFP_KERNEL)) return -ENOMEM; @@ -313,7 +331,9 @@ int bch2_read_single_folio(struct folio *folio, struct address_space *mapping) rbio->bio.bi_iter.bi_sector = folio_sector(folio); BUG_ON(!bio_add_folio(&rbio->bio, folio, folio_size(folio), 0)); + blk_start_plug(&plug); bch2_trans_run(c, (bchfs_read(trans, rbio, inode_inum(inode), NULL), 0)); + blk_finish_plug(&plug); wait_for_completion(&done); ret = blk_status_to_errno(rbio->bio.bi_status); @@ -605,15 +625,6 @@ do_io: BUG_ON(!bio_add_folio(&w->io->op.wbio.bio, folio, sectors << 9, offset << 9)); - /* Check for writing past i_size: */ - WARN_ONCE((bio_end_sector(&w->io->op.wbio.bio) << 9) > - round_up(i_size, block_bytes(c)) && - !test_bit(BCH_FS_emergency_ro, &c->flags), - "writing past i_size: %llu > %llu (unrounded %llu)\n", - bio_end_sector(&w->io->op.wbio.bio) << 9, - round_up(i_size, block_bytes(c)), - i_size); - w->io->op.res.sectors += reserved_sectors; w->io->op.i_sectors_delta -= dirty_sectors; w->io->op.new_i_size = i_size; @@ -669,7 +680,7 @@ int bch2_write_begin(struct file *file, struct address_space *mapping, folio = __filemap_get_folio(mapping, pos >> PAGE_SHIFT, FGP_WRITEBEGIN | fgf_set_order(len), mapping_gfp_mask(mapping)); - if (IS_ERR_OR_NULL(folio)) + if (IS_ERR(folio)) goto err_unlock; offset = pos - folio_pos(folio); diff --git a/fs/bcachefs/fs-io-direct.c b/fs/bcachefs/fs-io-direct.c index 6d3a05ae5da8..2089c36b5866 100644 --- a/fs/bcachefs/fs-io-direct.c +++ b/fs/bcachefs/fs-io-direct.c @@ -70,6 +70,7 @@ static int bch2_direct_IO_read(struct kiocb *req, struct iov_iter *iter) struct bch_io_opts opts; struct dio_read *dio; struct bio *bio; + struct blk_plug plug; loff_t offset = req->ki_pos; bool sync = is_sync_kiocb(req); size_t shorten; @@ -128,6 +129,8 @@ static int bch2_direct_IO_read(struct kiocb *req, struct iov_iter *iter) */ dio->should_dirty = iter_is_iovec(iter); + blk_start_plug(&plug); + goto start; while (iter->count) { bio = bio_alloc_bioset(NULL, @@ -160,6 +163,8 @@ start: bch2_read(c, rbio_init(bio, opts), inode_inum(inode)); } + blk_finish_plug(&plug); + iter->count += shorten; if (sync) { diff --git a/fs/bcachefs/fs-io-pagecache.c b/fs/bcachefs/fs-io-pagecache.c index 1d4910ea0f1d..e072900e6a5b 100644 --- a/fs/bcachefs/fs-io-pagecache.c +++ b/fs/bcachefs/fs-io-pagecache.c @@ -29,7 +29,7 @@ int bch2_filemap_get_contig_folios_d(struct address_space *mapping, break; f = __filemap_get_folio(mapping, pos >> PAGE_SHIFT, fgp_flags, gfp); - if (IS_ERR_OR_NULL(f)) + if (IS_ERR(f)) break; BUG_ON(fs->nr && folio_pos(f) != pos); @@ -199,7 +199,7 @@ int bch2_folio_set(struct bch_fs *c, subvol_inum inum, unsigned folio_idx = 0; return bch2_trans_run(c, - for_each_btree_key_in_subvolume_upto(trans, iter, BTREE_ID_extents, + for_each_btree_key_in_subvolume_max(trans, iter, BTREE_ID_extents, POS(inum.inum, offset), POS(inum.inum, U64_MAX), inum.subvol, BTREE_ITER_slots, k, ({ diff --git a/fs/bcachefs/fs-io.c b/fs/bcachefs/fs-io.c index 2456c41b215e..94bf34b9b65f 100644 --- a/fs/bcachefs/fs-io.c +++ b/fs/bcachefs/fs-io.c @@ -167,6 +167,34 @@ void __bch2_i_sectors_acct(struct bch_fs *c, struct bch_inode_info *inode, /* fsync: */ +static int bch2_get_inode_journal_seq_trans(struct btree_trans *trans, subvol_inum inum, + u64 *seq) +{ + struct printbuf buf = PRINTBUF; + struct bch_inode_unpacked u; + struct btree_iter iter; + int ret = bch2_inode_peek(trans, &iter, &u, inum, 0); + if (ret) + return ret; + + u64 cur_seq = journal_cur_seq(&trans->c->journal); + *seq = min(cur_seq, u.bi_journal_seq); + + if (fsck_err_on(u.bi_journal_seq > cur_seq, + trans, inode_journal_seq_in_future, + "inode journal seq in future (currently at %llu)\n%s", + cur_seq, + (bch2_inode_unpacked_to_text(&buf, &u), + buf.buf))) { + u.bi_journal_seq = cur_seq; + ret = bch2_inode_write(trans, &iter, &u); + } +fsck_err: + bch2_trans_iter_exit(trans, &iter); + printbuf_exit(&buf); + return ret; +} + /* * inode->ei_inode.bi_journal_seq won't be up to date since it's set in an * insert trigger: look up the btree inode instead @@ -180,9 +208,10 @@ static int bch2_flush_inode(struct bch_fs *c, if (!bch2_write_ref_tryget(c, BCH_WRITE_REF_fsync)) return -EROFS; - struct bch_inode_unpacked u; - int ret = bch2_inode_find_by_inum(c, inode_inum(inode), &u) ?: - bch2_journal_flush_seq(&c->journal, u.bi_journal_seq, TASK_INTERRUPTIBLE) ?: + u64 seq; + int ret = bch2_trans_commit_do(c, NULL, NULL, 0, + bch2_get_inode_journal_seq_trans(trans, inode_inum(inode), &seq)) ?: + bch2_journal_flush_seq(&c->journal, seq, TASK_INTERRUPTIBLE) ?: bch2_inode_flush_nocow_writes(c, inode); bch2_write_ref_put(c, BCH_WRITE_REF_fsync); return ret; @@ -222,7 +251,7 @@ static inline int range_has_data(struct bch_fs *c, u32 subvol, struct bpos end) { return bch2_trans_run(c, - for_each_btree_key_in_subvolume_upto(trans, iter, BTREE_ID_extents, start, end, + for_each_btree_key_in_subvolume_max(trans, iter, BTREE_ID_extents, start, end, subvol, 0, k, ({ bkey_extent_is_data(k.k) && !bkey_extent_is_unwritten(k); }))); @@ -256,7 +285,7 @@ static int __bch2_truncate_folio(struct bch_inode_info *inode, folio = __filemap_get_folio(mapping, index, FGP_LOCK|FGP_CREAT, GFP_KERNEL); - if (IS_ERR_OR_NULL(folio)) { + if (IS_ERR(folio)) { ret = -ENOMEM; goto out; } @@ -806,7 +835,7 @@ static int quota_reserve_range(struct bch_inode_info *inode, u64 sectors = end - start; int ret = bch2_trans_run(c, - for_each_btree_key_in_subvolume_upto(trans, iter, + for_each_btree_key_in_subvolume_max(trans, iter, BTREE_ID_extents, POS(inode->v.i_ino, start), POS(inode->v.i_ino, end - 1), @@ -877,11 +906,18 @@ loff_t bch2_remap_file_range(struct file *file_src, loff_t pos_src, bch2_mark_pagecache_unallocated(src, pos_src >> 9, (pos_src + aligned_len) >> 9); + /* + * XXX: we'd like to be telling bch2_remap_range() if we have + * permission to write to the source file, and thus if io path option + * changes should be propagated through the copy, but we need mnt_idmap + * from the pathwalk, awkward + */ ret = bch2_remap_range(c, inode_inum(dst), pos_dst >> 9, inode_inum(src), pos_src >> 9, aligned_len >> 9, - pos_dst + len, &i_sectors_delta); + pos_dst + len, &i_sectors_delta, + false); if (ret < 0) goto err; @@ -922,7 +958,7 @@ static loff_t bch2_seek_data(struct file *file, u64 offset) return -ENXIO; int ret = bch2_trans_run(c, - for_each_btree_key_in_subvolume_upto(trans, iter, BTREE_ID_extents, + for_each_btree_key_in_subvolume_max(trans, iter, BTREE_ID_extents, POS(inode->v.i_ino, offset >> 9), POS(inode->v.i_ino, U64_MAX), inum.subvol, 0, k, ({ @@ -958,7 +994,7 @@ static loff_t bch2_seek_hole(struct file *file, u64 offset) return -ENXIO; int ret = bch2_trans_run(c, - for_each_btree_key_in_subvolume_upto(trans, iter, BTREE_ID_extents, + for_each_btree_key_in_subvolume_max(trans, iter, BTREE_ID_extents, POS(inode->v.i_ino, offset >> 9), POS(inode->v.i_ino, U64_MAX), inum.subvol, BTREE_ITER_slots, k, ({ diff --git a/fs/bcachefs/fs-ioctl.c b/fs/bcachefs/fs-ioctl.c index 405cf08bda34..15725b4ce393 100644 --- a/fs/bcachefs/fs-ioctl.c +++ b/fs/bcachefs/fs-ioctl.c @@ -406,7 +406,7 @@ static long bch2_ioctl_subvolume_create(struct bch_fs *c, struct file *filp, sync_inodes_sb(c->vfs_sb); up_read(&c->vfs_sb->s_umount); } -retry: + if (arg.src_ptr) { error = user_path_at(arg.dirfd, (const char __user *)(unsigned long)arg.src_ptr, @@ -486,11 +486,6 @@ err3: err2: if (arg.src_ptr) path_put(&src_path); - - if (retry_estale(error, lookup_flags)) { - lookup_flags |= LOOKUP_REVAL; - goto retry; - } err1: return error; } diff --git a/fs/bcachefs/fs.c b/fs/bcachefs/fs.c index a41d0d8a2f7b..90ade8f648d9 100644 --- a/fs/bcachefs/fs.c +++ b/fs/bcachefs/fs.c @@ -23,6 +23,7 @@ #include "journal.h" #include "keylist.h" #include "quota.h" +#include "rebalance.h" #include "snapshot.h" #include "super.h" #include "xattr.h" @@ -38,6 +39,7 @@ #include <linux/posix_acl.h> #include <linux/random.h> #include <linux/seq_file.h> +#include <linux/siphash.h> #include <linux/statfs.h> #include <linux/string.h> #include <linux/xattr.h> @@ -65,6 +67,9 @@ void bch2_inode_update_after_write(struct btree_trans *trans, i_gid_write(&inode->v, bi->bi_gid); inode->v.i_mode = bi->bi_mode; + if (fields & ATTR_SIZE) + i_size_write(&inode->v, bi->bi_size); + if (fields & ATTR_ATIME) inode_set_atime_to_ts(&inode->v, bch2_time_to_timespec(c, bi->bi_atime)); if (fields & ATTR_MTIME) @@ -89,10 +94,25 @@ int __must_check bch2_write_inode(struct bch_fs *c, retry: bch2_trans_begin(trans); - ret = bch2_inode_peek(trans, &iter, &inode_u, inode_inum(inode), - BTREE_ITER_intent) ?: - (set ? set(trans, inode, &inode_u, p) : 0) ?: - bch2_inode_write(trans, &iter, &inode_u) ?: + ret = bch2_inode_peek(trans, &iter, &inode_u, inode_inum(inode), BTREE_ITER_intent); + if (ret) + goto err; + + struct bch_extent_rebalance old_r = bch2_inode_rebalance_opts_get(c, &inode_u); + + ret = (set ? set(trans, inode, &inode_u, p) : 0); + if (ret) + goto err; + + struct bch_extent_rebalance new_r = bch2_inode_rebalance_opts_get(c, &inode_u); + + if (memcmp(&old_r, &new_r, sizeof(new_r))) { + ret = bch2_set_rebalance_needs_scan_trans(trans, inode_u.bi_inum); + if (ret) + goto err; + } + + ret = bch2_inode_write(trans, &iter, &inode_u) ?: bch2_trans_commit(trans, NULL, NULL, BCH_TRANS_COMMIT_no_enospc); /* @@ -101,7 +121,7 @@ retry: */ if (!ret) bch2_inode_update_after_write(trans, inode, &inode_u, fields); - +err: bch2_trans_iter_exit(trans, &iter); if (bch2_err_matches(ret, BCH_ERR_transaction_restart)) @@ -160,8 +180,9 @@ static bool subvol_inum_eq(subvol_inum a, subvol_inum b) static u32 bch2_vfs_inode_hash_fn(const void *data, u32 len, u32 seed) { const subvol_inum *inum = data; + siphash_key_t k = { .key[0] = seed }; - return jhash(&inum->inum, sizeof(inum->inum), seed); + return siphash_2u64(inum->subvol, inum->inum, &k); } static u32 bch2_vfs_inode_obj_hash_fn(const void *data, u32 len, u32 seed) @@ -190,11 +211,18 @@ static const struct rhashtable_params bch2_vfs_inodes_params = { .automatic_shrinking = true, }; +static const struct rhashtable_params bch2_vfs_inodes_by_inum_params = { + .head_offset = offsetof(struct bch_inode_info, by_inum_hash), + .key_offset = offsetof(struct bch_inode_info, ei_inum.inum), + .key_len = sizeof(u64), + .automatic_shrinking = true, +}; + int bch2_inode_or_descendents_is_open(struct btree_trans *trans, struct bpos p) { struct bch_fs *c = trans->c; - struct rhashtable *ht = &c->vfs_inodes_table; - subvol_inum inum = (subvol_inum) { .inum = p.offset }; + struct rhltable *ht = &c->vfs_inodes_by_inum_table; + u64 inum = p.offset; DARRAY(u32) subvols; int ret = 0; @@ -219,15 +247,15 @@ restart_from_top: struct rhash_lock_head __rcu *const *bkt; struct rhash_head *he; unsigned int hash; - struct bucket_table *tbl = rht_dereference_rcu(ht->tbl, ht); + struct bucket_table *tbl = rht_dereference_rcu(ht->ht.tbl, &ht->ht); restart: - hash = rht_key_hashfn(ht, tbl, &inum, bch2_vfs_inodes_params); + hash = rht_key_hashfn(&ht->ht, tbl, &inum, bch2_vfs_inodes_by_inum_params); bkt = rht_bucket(tbl, hash); do { struct bch_inode_info *inode; rht_for_each_entry_rcu_from(inode, he, rht_ptr_rcu(bkt), tbl, hash, hash) { - if (inode->ei_inum.inum == inum.inum) { + if (inode->ei_inum.inum == inum) { ret = darray_push_gfp(&subvols, inode->ei_inum.subvol, GFP_NOWAIT|__GFP_NOWARN); if (ret) { @@ -248,7 +276,7 @@ restart: /* Ensure we see any new tables. */ smp_rmb(); - tbl = rht_dereference_rcu(tbl->future_tbl, ht); + tbl = rht_dereference_rcu(tbl->future_tbl, &ht->ht); if (unlikely(tbl)) goto restart; rcu_read_unlock(); @@ -327,7 +355,11 @@ static void bch2_inode_hash_remove(struct bch_fs *c, struct bch_inode_info *inod spin_unlock(&inode->v.i_lock); if (remove) { - int ret = rhashtable_remove_fast(&c->vfs_inodes_table, + int ret = rhltable_remove(&c->vfs_inodes_by_inum_table, + &inode->by_inum_hash, bch2_vfs_inodes_by_inum_params); + BUG_ON(ret); + + ret = rhashtable_remove_fast(&c->vfs_inodes_table, &inode->hash, bch2_vfs_inodes_params); BUG_ON(ret); inode->v.i_hash.pprev = NULL; @@ -372,6 +404,11 @@ retry: discard_new_inode(&inode->v); return old; } else { + int ret = rhltable_insert(&c->vfs_inodes_by_inum_table, + &inode->by_inum_hash, + bch2_vfs_inodes_by_inum_params); + BUG_ON(ret); + inode_fake_hash(&inode->v); inode_sb_list_add(&inode->v); @@ -465,7 +502,7 @@ struct inode *bch2_vfs_inode_get(struct bch_fs *c, subvol_inum inum) struct bch_inode_unpacked inode_u; struct bch_subvolume subvol; int ret = lockrestart_do(trans, - bch2_subvolume_get(trans, inum.subvol, true, 0, &subvol) ?: + bch2_subvolume_get(trans, inum.subvol, true, &subvol) ?: bch2_inode_find_by_inum_trans(trans, inum, &inode_u)) ?: PTR_ERR_OR_ZERO(inode = bch2_inode_hash_init_insert(trans, inum, &inode_u, &subvol)); bch2_trans_put(trans); @@ -535,8 +572,7 @@ retry: inum.subvol = inode_u.bi_subvol ?: dir->ei_inum.subvol; inum.inum = inode_u.bi_inum; - ret = bch2_subvolume_get(trans, inum.subvol, true, - BTREE_ITER_with_updates, &subvol) ?: + ret = bch2_subvolume_get(trans, inum.subvol, true, &subvol) ?: bch2_trans_commit(trans, NULL, &journal_seq, 0); if (unlikely(ret)) { bch2_quota_acct(c, bch_qid(&inode_u), Q_INO, -1, @@ -549,7 +585,7 @@ err_before_quota: if (!(flags & BCH_CREATE_TMPFILE)) { bch2_inode_update_after_write(trans, dir, &dir_u, - ATTR_MTIME|ATTR_CTIME); + ATTR_MTIME|ATTR_CTIME|ATTR_SIZE); mutex_unlock(&dir->ei_update_lock); } @@ -617,7 +653,7 @@ static struct bch_inode_info *bch2_lookup_trans(struct btree_trans *trans, struct bch_subvolume subvol; struct bch_inode_unpacked inode_u; - ret = bch2_subvolume_get(trans, inum.subvol, true, 0, &subvol) ?: + ret = bch2_subvolume_get(trans, inum.subvol, true, &subvol) ?: bch2_inode_find_by_inum_nowarn_trans(trans, inum, &inode_u) ?: PTR_ERR_OR_ZERO(inode = bch2_inode_hash_init_insert(trans, inum, &inode_u, &subvol)); @@ -628,7 +664,7 @@ static struct bch_inode_info *bch2_lookup_trans(struct btree_trans *trans, goto err; /* regular files may have hardlinks: */ - if (bch2_fs_inconsistent_on(bch2_inode_should_have_bp(&inode_u) && + if (bch2_fs_inconsistent_on(bch2_inode_should_have_single_bp(&inode_u) && !bkey_eq(k.k->p, POS(inode_u.bi_dir, inode_u.bi_dir_offset)), c, "dirent points to inode that does not point back:\n %s", @@ -706,7 +742,7 @@ static int __bch2_link(struct bch_fs *c, if (likely(!ret)) { bch2_inode_update_after_write(trans, dir, &dir_u, - ATTR_MTIME|ATTR_CTIME); + ATTR_MTIME|ATTR_CTIME|ATTR_SIZE); bch2_inode_update_after_write(trans, inode, &inode_u, ATTR_CTIME); } @@ -759,7 +795,7 @@ int __bch2_unlink(struct inode *vdir, struct dentry *dentry, goto err; bch2_inode_update_after_write(trans, dir, &dir_u, - ATTR_MTIME|ATTR_CTIME); + ATTR_MTIME|ATTR_CTIME|ATTR_SIZE); bch2_inode_update_after_write(trans, inode, &inode_u, ATTR_MTIME); @@ -937,11 +973,11 @@ err_tx_restart: dst_inode->v.i_ino != dst_inode_u.bi_inum); bch2_inode_update_after_write(trans, src_dir, &src_dir_u, - ATTR_MTIME|ATTR_CTIME); + ATTR_MTIME|ATTR_CTIME|ATTR_SIZE); if (src_dir != dst_dir) bch2_inode_update_after_write(trans, dst_dir, &dst_dir_u, - ATTR_MTIME|ATTR_CTIME); + ATTR_MTIME|ATTR_CTIME|ATTR_SIZE); bch2_inode_update_after_write(trans, src_inode, &src_inode_u, ATTR_CTIME); @@ -1245,7 +1281,6 @@ static int bch2_fiemap(struct inode *vinode, struct fiemap_extent_info *info, struct btree_iter iter; struct bkey_s_c k; struct bkey_buf cur, prev; - unsigned offset_into_extent, sectors; bool have_extent = false; int ret = 0; @@ -1278,7 +1313,7 @@ static int bch2_fiemap(struct inode *vinode, struct fiemap_extent_info *info, bch2_btree_iter_set_snapshot(&iter, snapshot); - k = bch2_btree_iter_peek_upto(&iter, end); + k = bch2_btree_iter_peek_max(&iter, end); ret = bkey_err(k); if (ret) continue; @@ -1292,9 +1327,8 @@ static int bch2_fiemap(struct inode *vinode, struct fiemap_extent_info *info, continue; } - offset_into_extent = iter.pos.offset - - bkey_start_offset(k.k); - sectors = k.k->size - offset_into_extent; + s64 offset_into_extent = iter.pos.offset - bkey_start_offset(k.k); + unsigned sectors = k.k->size - offset_into_extent; bch2_bkey_buf_reassemble(&cur, c, k); @@ -1306,7 +1340,7 @@ static int bch2_fiemap(struct inode *vinode, struct fiemap_extent_info *info, k = bkey_i_to_s_c(cur.k); bch2_bkey_buf_realloc(&prev, c, k.k->u64s); - sectors = min(sectors, k.k->size - offset_into_extent); + sectors = min_t(unsigned, sectors, k.k->size - offset_into_extent); bch2_cut_front(POS(k.k->p.inode, bkey_start_offset(k.k) + @@ -1736,7 +1770,6 @@ static void bch2_vfs_inode_init(struct btree_trans *trans, bch2_inode_update_after_write(trans, inode, bi, ~0); inode->v.i_blocks = bi->bi_sectors; - inode->v.i_ino = bi->bi_inum; inode->v.i_rdev = bi->bi_dev; inode->v.i_generation = bi->bi_generation; inode->v.i_size = bi->bi_size; @@ -2200,7 +2233,8 @@ got_sb: sb->s_time_gran = c->sb.nsec_per_time_unit; sb->s_time_min = div_s64(S64_MIN, c->sb.time_units_per_sec) + 1; sb->s_time_max = div_s64(S64_MAX, c->sb.time_units_per_sec); - sb->s_uuid = c->sb.user_uuid; + super_set_uuid(sb, c->sb.user_uuid.b, sizeof(c->sb.user_uuid)); + super_set_sysfs_name_uuid(sb); sb->s_shrink->seeks = 0; c->vfs_sb = sb; strscpy(sb->s_id, c->name, sizeof(sb->s_id)); @@ -2345,13 +2379,16 @@ static int bch2_init_fs_context(struct fs_context *fc) void bch2_fs_vfs_exit(struct bch_fs *c) { + if (c->vfs_inodes_by_inum_table.ht.tbl) + rhltable_destroy(&c->vfs_inodes_by_inum_table); if (c->vfs_inodes_table.tbl) rhashtable_destroy(&c->vfs_inodes_table); } int bch2_fs_vfs_init(struct bch_fs *c) { - return rhashtable_init(&c->vfs_inodes_table, &bch2_vfs_inodes_params); + return rhashtable_init(&c->vfs_inodes_table, &bch2_vfs_inodes_params) ?: + rhltable_init(&c->vfs_inodes_by_inum_table, &bch2_vfs_inodes_by_inum_params); } static struct file_system_type bcache_fs_type = { diff --git a/fs/bcachefs/fs.h b/fs/bcachefs/fs.h index 59f9f7ae728d..dd2198541455 100644 --- a/fs/bcachefs/fs.h +++ b/fs/bcachefs/fs.h @@ -14,6 +14,7 @@ struct bch_inode_info { struct inode v; struct rhash_head hash; + struct rhlist_head by_inum_hash; subvol_inum ei_inum; struct list_head ei_vfs_inode_list; diff --git a/fs/bcachefs/fsck.c b/fs/bcachefs/fsck.c index 75c8a97a6954..8fcf7c8e5ede 100644 --- a/fs/bcachefs/fsck.c +++ b/fs/bcachefs/fsck.c @@ -1,6 +1,7 @@ // SPDX-License-Identifier: GPL-2.0 #include "bcachefs.h" +#include "bcachefs_ioctl.h" #include "bkey_buf.h" #include "btree_cache.h" #include "btree_update.h" @@ -16,6 +17,7 @@ #include "recovery_passes.h" #include "snapshot.h" #include "super.h" +#include "thread_with_file.h" #include "xattr.h" #include <linux/bsearch.h> @@ -73,7 +75,7 @@ static s64 bch2_count_inode_sectors(struct btree_trans *trans, u64 inum, { u64 sectors = 0; - int ret = for_each_btree_key_upto(trans, iter, BTREE_ID_extents, + int ret = for_each_btree_key_max(trans, iter, BTREE_ID_extents, SPOS(inum, 0, snapshot), POS(inum, U64_MAX), 0, k, ({ @@ -90,7 +92,7 @@ static s64 bch2_count_subdirs(struct btree_trans *trans, u64 inum, { u64 subdirs = 0; - int ret = for_each_btree_key_upto(trans, iter, BTREE_ID_dirents, + int ret = for_each_btree_key_max(trans, iter, BTREE_ID_dirents, SPOS(inum, 0, snapshot), POS(inum, U64_MAX), 0, k, ({ @@ -107,7 +109,7 @@ static int subvol_lookup(struct btree_trans *trans, u32 subvol, u32 *snapshot, u64 *inum) { struct bch_subvolume s; - int ret = bch2_subvolume_get(trans, subvol, false, 0, &s); + int ret = bch2_subvolume_get(trans, subvol, false, &s); *snapshot = le32_to_cpu(s.snapshot); *inum = le64_to_cpu(s.inode); @@ -170,7 +172,7 @@ static int lookup_dirent_in_snapshot(struct btree_trans *trans, if (ret) return ret; - struct bkey_s_c_dirent d = bkey_s_c_to_dirent(bch2_btree_iter_peek_slot(&iter)); + struct bkey_s_c_dirent d = bkey_s_c_to_dirent(k); *target = le64_to_cpu(d.v->d_inum); *type = d.v->d_type; bch2_trans_iter_exit(trans, &iter); @@ -203,6 +205,36 @@ err: return ret; } +/* + * Find any subvolume associated with a tree of snapshots + * We can't rely on master_subvol - it might have been deleted. + */ +static int find_snapshot_tree_subvol(struct btree_trans *trans, + u32 tree_id, u32 *subvol) +{ + struct btree_iter iter; + struct bkey_s_c k; + int ret; + + for_each_btree_key_norestart(trans, iter, BTREE_ID_snapshots, POS_MIN, 0, k, ret) { + if (k.k->type != KEY_TYPE_snapshot) + continue; + + struct bkey_s_c_snapshot s = bkey_s_c_to_snapshot(k); + if (le32_to_cpu(s.v->tree) != tree_id) + continue; + + if (s.v->subvol) { + *subvol = le32_to_cpu(s.v->subvol); + goto found; + } + } + ret = -BCH_ERR_ENOENT_no_snapshot_tree_subvol; +found: + bch2_trans_iter_exit(trans, &iter); + return ret; +} + /* Get lost+found, create if it doesn't exist: */ static int lookup_lostfound(struct btree_trans *trans, u32 snapshot, struct bch_inode_unpacked *lostfound, @@ -210,6 +242,7 @@ static int lookup_lostfound(struct btree_trans *trans, u32 snapshot, { struct bch_fs *c = trans->c; struct qstr lostfound_str = QSTR("lost+found"); + struct btree_iter lostfound_iter = { NULL }; u64 inum = 0; unsigned d_type = 0; int ret; @@ -220,20 +253,24 @@ static int lookup_lostfound(struct btree_trans *trans, u32 snapshot, if (ret) return ret; - subvol_inum root_inum = { .subvol = le32_to_cpu(st.master_subvol) }; + u32 subvolid; + ret = find_snapshot_tree_subvol(trans, + bch2_snapshot_tree(c, snapshot), &subvolid); + bch_err_msg(c, ret, "finding subvol associated with snapshot tree %u", + bch2_snapshot_tree(c, snapshot)); + if (ret) + return ret; struct bch_subvolume subvol; - ret = bch2_subvolume_get(trans, le32_to_cpu(st.master_subvol), - false, 0, &subvol); - bch_err_msg(c, ret, "looking up root subvol %u for snapshot %u", - le32_to_cpu(st.master_subvol), snapshot); + ret = bch2_subvolume_get(trans, subvolid, false, &subvol); + bch_err_msg(c, ret, "looking up subvol %u for snapshot %u", subvolid, snapshot); if (ret) return ret; if (!subvol.inode) { struct btree_iter iter; struct bkey_i_subvolume *subvol = bch2_bkey_get_mut_typed(trans, &iter, - BTREE_ID_subvolumes, POS(0, le32_to_cpu(st.master_subvol)), + BTREE_ID_subvolumes, POS(0, subvolid), 0, subvolume); ret = PTR_ERR_OR_ZERO(subvol); if (ret) @@ -243,13 +280,16 @@ static int lookup_lostfound(struct btree_trans *trans, u32 snapshot, bch2_trans_iter_exit(trans, &iter); } - root_inum.inum = le64_to_cpu(subvol.inode); + subvol_inum root_inum = { + .subvol = subvolid, + .inum = le64_to_cpu(subvol.inode) + }; struct bch_inode_unpacked root_inode; struct bch_hash_info root_hash_info; ret = lookup_inode(trans, root_inum.inum, snapshot, &root_inode); bch_err_msg(c, ret, "looking up root inode %llu for subvol %u", - root_inum.inum, le32_to_cpu(st.master_subvol)); + root_inum.inum, subvolid); if (ret) return ret; @@ -288,11 +328,16 @@ create_lostfound: * XXX: we could have a nicer log message here if we had a nice way to * walk backpointers to print a path */ - bch_notice(c, "creating lost+found in subvol %llu snapshot %u", - root_inum.subvol, le32_to_cpu(st.root_snapshot)); + struct printbuf path = PRINTBUF; + ret = bch2_inum_to_path(trans, root_inum, &path); + if (ret) + goto err; + + bch_notice(c, "creating %s/lost+found in subvol %llu snapshot %u", + path.buf, root_inum.subvol, snapshot); + printbuf_exit(&path); u64 now = bch2_current_time(c); - struct btree_iter lostfound_iter = { NULL }; u64 cpu = raw_smp_processor_id(); bch2_inode_init_early(c, lostfound); @@ -451,7 +496,9 @@ static int reattach_inode(struct btree_trans *trans, struct bch_inode_unpacked * continue; struct bch_inode_unpacked child_inode; - bch2_inode_unpack(k, &child_inode); + ret = bch2_inode_unpack(k, &child_inode); + if (ret) + break; if (!inode_should_reattach(&child_inode)) { ret = maybe_delete_dirent(trans, @@ -482,6 +529,13 @@ static int reattach_inode(struct btree_trans *trans, struct bch_inode_unpacked * return ret; } +static struct bkey_s_c_dirent dirent_get_by_pos(struct btree_trans *trans, + struct btree_iter *iter, + struct bpos pos) +{ + return bch2_bkey_get_iter_typed(trans, iter, BTREE_ID_dirents, pos, 0, dirent); +} + static int remove_backpointer(struct btree_trans *trans, struct bch_inode_unpacked *inode) { @@ -490,13 +544,11 @@ static int remove_backpointer(struct btree_trans *trans, struct bch_fs *c = trans->c; struct btree_iter iter; - struct bkey_s_c_dirent d = - bch2_bkey_get_iter_typed(trans, &iter, BTREE_ID_dirents, - SPOS(inode->bi_dir, inode->bi_dir_offset, inode->bi_snapshot), 0, - dirent); - int ret = bkey_err(d) ?: - dirent_points_to_inode(c, d, inode) ?: - __remove_dirent(trans, d.k->p); + struct bkey_s_c_dirent d = dirent_get_by_pos(trans, &iter, + SPOS(inode->bi_dir, inode->bi_dir_offset, inode->bi_snapshot)); + int ret = bkey_err(d) ?: + dirent_points_to_inode(c, d, inode) ?: + __remove_dirent(trans, d.k->p); bch2_trans_iter_exit(trans, &iter); return ret; } @@ -613,7 +665,7 @@ static int reconstruct_inode(struct btree_trans *trans, enum btree_id btree, u32 struct btree_iter iter = {}; bch2_trans_iter_init(trans, &iter, BTREE_ID_extents, SPOS(inum, U64_MAX, snapshot), 0); - struct bkey_s_c k = bch2_btree_iter_peek_prev(&iter); + struct bkey_s_c k = bch2_btree_iter_peek_prev_min(&iter, POS(inum, 0)); bch2_trans_iter_exit(trans, &iter); int ret = bkey_err(k); if (ret) @@ -780,11 +832,13 @@ struct inode_walker { struct bpos last_pos; DARRAY(struct inode_walker_entry) inodes; + snapshot_id_list deletes; }; static void inode_walker_exit(struct inode_walker *w) { darray_exit(&w->inodes); + darray_exit(&w->deletes); } static struct inode_walker inode_walker_init(void) @@ -797,9 +851,8 @@ static int add_inode(struct bch_fs *c, struct inode_walker *w, { struct bch_inode_unpacked u; - BUG_ON(bch2_inode_unpack(inode, &u)); - - return darray_push(&w->inodes, ((struct inode_walker_entry) { + return bch2_inode_unpack(inode, &u) ?: + darray_push(&w->inodes, ((struct inode_walker_entry) { .inode = u, .snapshot = inode.k->p.snapshot, })); @@ -909,8 +962,9 @@ static int get_visible_inodes(struct btree_trans *trans, int ret; w->inodes.nr = 0; + w->deletes.nr = 0; - for_each_btree_key_norestart(trans, iter, BTREE_ID_inodes, POS(0, inum), + for_each_btree_key_reverse_norestart(trans, iter, BTREE_ID_inodes, SPOS(0, inum, s->pos.snapshot), BTREE_ITER_all_snapshots, k, ret) { if (k.k->p.offset != inum) break; @@ -918,10 +972,13 @@ static int get_visible_inodes(struct btree_trans *trans, if (!ref_visible(c, s, s->pos.snapshot, k.k->p.snapshot)) continue; - if (bkey_is_inode(k.k)) - add_inode(c, w, k); + if (snapshot_list_has_ancestor(c, &w->deletes, k.k->p.snapshot)) + continue; - if (k.k->p.snapshot >= s->pos.snapshot) + ret = bkey_is_inode(k.k) + ? add_inode(c, w, k) + : snapshot_list_add(c, &w->deletes, k.k->p.snapshot); + if (ret) break; } bch2_trans_iter_exit(trans, &iter); @@ -929,69 +986,16 @@ static int get_visible_inodes(struct btree_trans *trans, return ret; } -static int dirent_has_target(struct btree_trans *trans, struct bkey_s_c_dirent d) -{ - if (d.v->d_type == DT_SUBVOL) { - u32 snap; - u64 inum; - int ret = subvol_lookup(trans, le32_to_cpu(d.v->d_child_subvol), &snap, &inum); - if (ret && !bch2_err_matches(ret, ENOENT)) - return ret; - return !ret; - } else { - struct btree_iter iter; - struct bkey_s_c k = bch2_bkey_get_iter(trans, &iter, BTREE_ID_inodes, - SPOS(0, le64_to_cpu(d.v->d_inum), d.k->p.snapshot), 0); - int ret = bkey_err(k); - if (ret) - return ret; - - ret = bkey_is_inode(k.k); - bch2_trans_iter_exit(trans, &iter); - return ret; - } -} - /* * Prefer to delete the first one, since that will be the one at the wrong * offset: * return value: 0 -> delete k1, 1 -> delete k2 */ -static int hash_pick_winner(struct btree_trans *trans, - const struct bch_hash_desc desc, - struct bch_hash_info *hash_info, - struct bkey_s_c k1, - struct bkey_s_c k2) -{ - if (bkey_val_bytes(k1.k) == bkey_val_bytes(k2.k) && - !memcmp(k1.v, k2.v, bkey_val_bytes(k1.k))) - return 0; - - switch (desc.btree_id) { - case BTREE_ID_dirents: { - int ret = dirent_has_target(trans, bkey_s_c_to_dirent(k1)); - if (ret < 0) - return ret; - if (!ret) - return 0; - - ret = dirent_has_target(trans, bkey_s_c_to_dirent(k2)); - if (ret < 0) - return ret; - if (!ret) - return 1; - return 2; - } - default: - return 0; - } -} - -static int fsck_update_backpointers(struct btree_trans *trans, - struct snapshots_seen *s, - const struct bch_hash_desc desc, - struct bch_hash_info *hash_info, - struct bkey_i *new) +int bch2_fsck_update_backpointers(struct btree_trans *trans, + struct snapshots_seen *s, + const struct bch_hash_desc desc, + struct bch_hash_info *hash_info, + struct bkey_i *new) { if (new->k.type != KEY_TYPE_dirent) return 0; @@ -1019,160 +1023,6 @@ err: return ret; } -static int fsck_rename_dirent(struct btree_trans *trans, - struct snapshots_seen *s, - const struct bch_hash_desc desc, - struct bch_hash_info *hash_info, - struct bkey_s_c_dirent old) -{ - struct qstr old_name = bch2_dirent_get_name(old); - struct bkey_i_dirent *new = bch2_trans_kmalloc(trans, bkey_bytes(old.k) + 32); - int ret = PTR_ERR_OR_ZERO(new); - if (ret) - return ret; - - bkey_dirent_init(&new->k_i); - dirent_copy_target(new, old); - new->k.p = old.k->p; - - for (unsigned i = 0; i < 1000; i++) { - unsigned len = sprintf(new->v.d_name, "%.*s.fsck_renamed-%u", - old_name.len, old_name.name, i); - unsigned u64s = BKEY_U64s + dirent_val_u64s(len); - - if (u64s > U8_MAX) - return -EINVAL; - - new->k.u64s = u64s; - - ret = bch2_hash_set_in_snapshot(trans, bch2_dirent_hash_desc, hash_info, - (subvol_inum) { 0, old.k->p.inode }, - old.k->p.snapshot, &new->k_i, - BTREE_UPDATE_internal_snapshot_node); - if (!bch2_err_matches(ret, EEXIST)) - break; - } - - if (ret) - return ret; - - return fsck_update_backpointers(trans, s, desc, hash_info, &new->k_i); -} - -static int hash_check_key(struct btree_trans *trans, - struct snapshots_seen *s, - const struct bch_hash_desc desc, - struct bch_hash_info *hash_info, - struct btree_iter *k_iter, struct bkey_s_c hash_k) -{ - struct bch_fs *c = trans->c; - struct btree_iter iter = { NULL }; - struct printbuf buf = PRINTBUF; - struct bkey_s_c k; - u64 hash; - int ret = 0; - - if (hash_k.k->type != desc.key_type) - return 0; - - hash = desc.hash_bkey(hash_info, hash_k); - - if (likely(hash == hash_k.k->p.offset)) - return 0; - - if (hash_k.k->p.offset < hash) - goto bad_hash; - - for_each_btree_key_norestart(trans, iter, desc.btree_id, - SPOS(hash_k.k->p.inode, hash, hash_k.k->p.snapshot), - BTREE_ITER_slots, k, ret) { - if (bkey_eq(k.k->p, hash_k.k->p)) - break; - - if (k.k->type == desc.key_type && - !desc.cmp_bkey(k, hash_k)) - goto duplicate_entries; - - if (bkey_deleted(k.k)) { - bch2_trans_iter_exit(trans, &iter); - goto bad_hash; - } - } -out: - bch2_trans_iter_exit(trans, &iter); - printbuf_exit(&buf); - return ret; -bad_hash: - if (fsck_err(trans, hash_table_key_wrong_offset, - "hash table key at wrong offset: btree %s inode %llu offset %llu, hashed to %llu\n %s", - bch2_btree_id_str(desc.btree_id), hash_k.k->p.inode, hash_k.k->p.offset, hash, - (printbuf_reset(&buf), - bch2_bkey_val_to_text(&buf, c, hash_k), buf.buf))) { - struct bkey_i *new = bch2_bkey_make_mut_noupdate(trans, hash_k); - if (IS_ERR(new)) - return PTR_ERR(new); - - k = bch2_hash_set_or_get_in_snapshot(trans, &iter, desc, hash_info, - (subvol_inum) { 0, hash_k.k->p.inode }, - hash_k.k->p.snapshot, new, - STR_HASH_must_create| - BTREE_ITER_with_updates| - BTREE_UPDATE_internal_snapshot_node); - ret = bkey_err(k); - if (ret) - goto out; - if (k.k) - goto duplicate_entries; - - ret = bch2_hash_delete_at(trans, desc, hash_info, k_iter, - BTREE_UPDATE_internal_snapshot_node) ?: - fsck_update_backpointers(trans, s, desc, hash_info, new) ?: - bch2_trans_commit(trans, NULL, NULL, BCH_TRANS_COMMIT_no_enospc) ?: - -BCH_ERR_transaction_restart_nested; - goto out; - } -fsck_err: - goto out; -duplicate_entries: - ret = hash_pick_winner(trans, desc, hash_info, hash_k, k); - if (ret < 0) - goto out; - - if (!fsck_err(trans, hash_table_key_duplicate, - "duplicate hash table keys%s:\n%s", - ret != 2 ? "" : ", both point to valid inodes", - (printbuf_reset(&buf), - bch2_bkey_val_to_text(&buf, c, hash_k), - prt_newline(&buf), - bch2_bkey_val_to_text(&buf, c, k), - buf.buf))) - goto out; - - switch (ret) { - case 0: - ret = bch2_hash_delete_at(trans, desc, hash_info, k_iter, 0); - break; - case 1: - ret = bch2_hash_delete_at(trans, desc, hash_info, &iter, 0); - break; - case 2: - ret = fsck_rename_dirent(trans, s, desc, hash_info, bkey_s_c_to_dirent(hash_k)) ?: - bch2_hash_delete_at(trans, desc, hash_info, k_iter, 0); - goto out; - } - - ret = bch2_trans_commit(trans, NULL, NULL, 0) ?: - -BCH_ERR_transaction_restart_nested; - goto out; -} - -static struct bkey_s_c_dirent dirent_get_by_pos(struct btree_trans *trans, - struct btree_iter *iter, - struct bpos pos) -{ - return bch2_bkey_get_iter_typed(trans, iter, BTREE_ID_dirents, pos, 0, dirent); -} - static struct bkey_s_c_dirent inode_get_dirent(struct btree_trans *trans, struct btree_iter *iter, struct bch_inode_unpacked *inode, @@ -1260,12 +1110,43 @@ static int get_snapshot_root_inode(struct btree_trans *trans, goto err; BUG(); found_root: - BUG_ON(bch2_inode_unpack(k, root)); + ret = bch2_inode_unpack(k, root); err: bch2_trans_iter_exit(trans, &iter); return ret; } +static int check_directory_size(struct btree_trans *trans, + struct bch_inode_unpacked *inode_u, + struct bkey_s_c inode_k, bool *write_inode) +{ + struct btree_iter iter; + struct bkey_s_c k; + u64 new_size = 0; + int ret; + + for_each_btree_key_max_norestart(trans, iter, BTREE_ID_dirents, + SPOS(inode_k.k->p.offset, 0, inode_k.k->p.snapshot), + POS(inode_k.k->p.offset, U64_MAX), + 0, k, ret) { + if (k.k->type != KEY_TYPE_dirent) + continue; + + struct bkey_s_c_dirent dirent = bkey_s_c_to_dirent(k); + struct qstr name = bch2_dirent_get_name(dirent); + + new_size += dirent_occupied_size(&name); + } + bch2_trans_iter_exit(trans, &iter); + + if (!ret && inode_u->bi_size != new_size) { + inode_u->bi_size = new_size; + *write_inode = true; + } + + return ret; +} + static int check_inode(struct btree_trans *trans, struct btree_iter *iter, struct bkey_s_c k, @@ -1291,7 +1172,9 @@ static int check_inode(struct btree_trans *trans, if (!bkey_is_inode(k.k)) return 0; - BUG_ON(bch2_inode_unpack(k, &u)); + ret = bch2_inode_unpack(k, &u); + if (ret) + goto err; if (snapshot_root->bi_inum != u.bi_inum) { ret = get_snapshot_root_inode(trans, snapshot_root, u.bi_inum); @@ -1302,7 +1185,7 @@ static int check_inode(struct btree_trans *trans, if (fsck_err_on(u.bi_hash_seed != snapshot_root->bi_hash_seed || INODE_STR_HASH(&u) != INODE_STR_HASH(snapshot_root), trans, inode_snapshot_mismatch, - "inodes in different snapshots don't match")) { + "inode hash info in different snapshots don't match")) { u.bi_hash_seed = snapshot_root->bi_hash_seed; SET_INODE_STR_HASH(&u, INODE_STR_HASH(snapshot_root)); do_update = true; @@ -1392,7 +1275,7 @@ static int check_inode(struct btree_trans *trans, if (fsck_err_on(!ret, trans, inode_unlinked_and_not_open, - "inode %llu%u unlinked and not open", + "inode %llu:%u unlinked and not open", u.bi_inum, u.bi_snapshot)) { ret = bch2_inode_rm_snapshot(trans, u.bi_inum, iter->pos.snapshot); bch_err_msg(c, ret, "in fsck deleting inode"); @@ -1415,7 +1298,7 @@ static int check_inode(struct btree_trans *trans, if (u.bi_subvol) { struct bch_subvolume s; - ret = bch2_subvolume_get(trans, u.bi_subvol, false, 0, &s); + ret = bch2_subvolume_get(trans, u.bi_subvol, false, &s); if (ret && !bch2_err_matches(ret, ENOENT)) goto err; @@ -1441,6 +1324,27 @@ static int check_inode(struct btree_trans *trans, do_update = true; } } + + if (fsck_err_on(u.bi_journal_seq > journal_cur_seq(&c->journal), + trans, inode_journal_seq_in_future, + "inode journal seq in future (currently at %llu)\n%s", + journal_cur_seq(&c->journal), + (printbuf_reset(&buf), + bch2_inode_unpacked_to_text(&buf, &u), + buf.buf))) { + u.bi_journal_seq = journal_cur_seq(&c->journal); + do_update = true; + } + + if (S_ISDIR(u.bi_mode)) { + ret = check_directory_size(trans, &u, k, &do_update); + + fsck_err_on(ret, + trans, directory_size_mismatch, + "directory inode %llu:%u with the mismatch directory size", + u.bi_inum, k.k->p.snapshot); + ret = 0; + } do_update: if (do_update) { ret = __bch2_fsck_write_inode(trans, &u); @@ -1502,7 +1406,9 @@ static int find_oldest_inode_needs_reattach(struct btree_trans *trans, break; struct bch_inode_unpacked parent_inode; - bch2_inode_unpack(k, &parent_inode); + ret = bch2_inode_unpack(k, &parent_inode); + if (ret) + break; if (!inode_should_reattach(&parent_inode)) break; @@ -1525,7 +1431,9 @@ static int check_unreachable_inode(struct btree_trans *trans, return 0; struct bch_inode_unpacked inode; - BUG_ON(bch2_inode_unpack(k, &inode)); + ret = bch2_inode_unpack(k, &inode); + if (ret) + return ret; if (!inode_should_reattach(&inode)) return 0; @@ -1649,7 +1557,7 @@ static int check_i_sectors_notnested(struct btree_trans *trans, struct inode_wal if (i->count != count2) { bch_err_ratelimited(c, "fsck counted i_sectors wrong for inode %llu:%u: got %llu should be %llu", w->last_pos.inode, i->snapshot, i->count, count2); - return -BCH_ERR_internal_fsck_err; + i->count = count2; } if (fsck_err_on(!(i->inode.bi_flags & BCH_INODE_i_sectors_dirty), @@ -1753,7 +1661,7 @@ static int overlapping_extents_found(struct btree_trans *trans, bch2_trans_iter_init(trans, &iter1, btree, pos1, BTREE_ITER_all_snapshots| BTREE_ITER_not_extents); - k1 = bch2_btree_iter_peek_upto(&iter1, POS(pos1.inode, U64_MAX)); + k1 = bch2_btree_iter_peek_max(&iter1, POS(pos1.inode, U64_MAX)); ret = bkey_err(k1); if (ret) goto err; @@ -1778,7 +1686,7 @@ static int overlapping_extents_found(struct btree_trans *trans, while (1) { bch2_btree_iter_advance(&iter2); - k2 = bch2_btree_iter_peek_upto(&iter2, POS(pos1.inode, U64_MAX)); + k2 = bch2_btree_iter_peek_max(&iter2, POS(pos1.inode, U64_MAX)); ret = bkey_err(k2); if (ret) goto err; @@ -2156,7 +2064,7 @@ static int check_dirent_inode_dirent(struct btree_trans *trans, return __bch2_fsck_write_inode(trans, target); } - if (bch2_inode_should_have_bp(target) && + if (bch2_inode_should_have_single_bp(target) && !fsck_err(trans, inode_wrong_backpointer, "dirent points to inode that does not point back:\n %s", (bch2_bkey_val_to_text(&buf, c, d.s_c), @@ -2480,7 +2388,7 @@ static int check_dirent(struct btree_trans *trans, struct btree_iter *iter, *hash_info = bch2_hash_info_init(c, &i->inode); dir->first_this_inode = false; - ret = hash_check_key(trans, s, bch2_dirent_hash_desc, hash_info, iter, k); + ret = bch2_str_hash_check_key(trans, s, &bch2_dirent_hash_desc, hash_info, iter, k); if (ret < 0) goto err; if (ret) { @@ -2519,6 +2427,30 @@ static int check_dirent(struct btree_trans *trans, struct btree_iter *iter, if (ret) goto err; } + + darray_for_each(target->deletes, i) + if (fsck_err_on(!snapshot_list_has_id(&s->ids, *i), + trans, dirent_to_overwritten_inode, + "dirent points to inode overwritten in snapshot %u:\n%s", + *i, + (printbuf_reset(&buf), + bch2_bkey_val_to_text(&buf, c, k), + buf.buf))) { + struct btree_iter delete_iter; + bch2_trans_iter_init(trans, &delete_iter, + BTREE_ID_dirents, + SPOS(k.k->p.inode, k.k->p.offset, *i), + BTREE_ITER_intent); + ret = bch2_btree_iter_traverse(&delete_iter) ?: + bch2_hash_delete_at(trans, bch2_dirent_hash_desc, + hash_info, + &delete_iter, + BTREE_UPDATE_internal_snapshot_node); + bch2_trans_iter_exit(trans, &delete_iter); + if (ret) + goto err; + + } } ret = bch2_trans_commit(trans, NULL, NULL, BCH_TRANS_COMMIT_no_enospc); @@ -2594,7 +2526,7 @@ static int check_xattr(struct btree_trans *trans, struct btree_iter *iter, *hash_info = bch2_hash_info_init(c, &i->inode); inode->first_this_inode = false; - ret = hash_check_key(trans, NULL, bch2_xattr_hash_desc, hash_info, iter, k); + ret = bch2_str_hash_check_key(trans, NULL, &bch2_xattr_hash_desc, hash_info, iter, k); bch_err_fn(c, ret); return ret; } @@ -2774,6 +2706,48 @@ struct pathbuf_entry { typedef DARRAY(struct pathbuf_entry) pathbuf; +static int bch2_bi_depth_renumber_one(struct btree_trans *trans, struct pathbuf_entry *p, + u32 new_depth) +{ + struct btree_iter iter; + struct bkey_s_c k = bch2_bkey_get_iter(trans, &iter, BTREE_ID_inodes, + SPOS(0, p->inum, p->snapshot), 0); + + struct bch_inode_unpacked inode; + int ret = bkey_err(k) ?: + !bkey_is_inode(k.k) ? -BCH_ERR_ENOENT_inode + : bch2_inode_unpack(k, &inode); + if (ret) + goto err; + + if (inode.bi_depth != new_depth) { + inode.bi_depth = new_depth; + ret = __bch2_fsck_write_inode(trans, &inode) ?: + bch2_trans_commit(trans, NULL, NULL, 0); + } +err: + bch2_trans_iter_exit(trans, &iter); + return ret; +} + +static int bch2_bi_depth_renumber(struct btree_trans *trans, pathbuf *path, u32 new_bi_depth) +{ + u32 restart_count = trans->restart_count; + int ret = 0; + + darray_for_each_reverse(*path, i) { + ret = nested_lockrestart_do(trans, + bch2_bi_depth_renumber_one(trans, i, new_bi_depth)); + bch_err_fn(trans->c, ret); + if (ret) + break; + + new_bi_depth++; + } + + return ret ?: trans_was_restarted(trans, restart_count); +} + static bool path_is_dup(pathbuf *p, u64 inum, u32 snapshot) { darray_for_each(*p, i) @@ -2783,21 +2757,21 @@ static bool path_is_dup(pathbuf *p, u64 inum, u32 snapshot) return false; } -static int check_path(struct btree_trans *trans, pathbuf *p, struct bkey_s_c inode_k) +static int check_path_loop(struct btree_trans *trans, struct bkey_s_c inode_k) { struct bch_fs *c = trans->c; struct btree_iter inode_iter = {}; - struct bch_inode_unpacked inode; + pathbuf path = {}; struct printbuf buf = PRINTBUF; u32 snapshot = inode_k.k->p.snapshot; + bool redo_bi_depth = false; + u32 min_bi_depth = U32_MAX; int ret = 0; - p->nr = 0; - - BUG_ON(bch2_inode_unpack(inode_k, &inode)); - - if (!S_ISDIR(inode.bi_mode)) - return 0; + struct bch_inode_unpacked inode; + ret = bch2_inode_unpack(inode_k, &inode); + if (ret) + return ret; while (!inode.bi_subvol) { struct btree_iter dirent_iter; @@ -2807,7 +2781,7 @@ static int check_path(struct btree_trans *trans, pathbuf *p, struct bkey_s_c ino d = inode_get_dirent(trans, &dirent_iter, &inode, &parent_snapshot); ret = bkey_err(d.s_c); if (ret && !bch2_err_matches(ret, ENOENT)) - break; + goto out; if (!ret && (ret = dirent_points_to_inode(c, d, &inode))) bch2_trans_iter_exit(trans, &dirent_iter); @@ -2822,7 +2796,7 @@ static int check_path(struct btree_trans *trans, pathbuf *p, struct bkey_s_c ino bch2_trans_iter_exit(trans, &dirent_iter); - ret = darray_push(p, ((struct pathbuf_entry) { + ret = darray_push(&path, ((struct pathbuf_entry) { .inum = inode.bi_inum, .snapshot = snapshot, })); @@ -2834,22 +2808,32 @@ static int check_path(struct btree_trans *trans, pathbuf *p, struct bkey_s_c ino bch2_trans_iter_exit(trans, &inode_iter); inode_k = bch2_bkey_get_iter(trans, &inode_iter, BTREE_ID_inodes, SPOS(0, inode.bi_dir, snapshot), 0); + + struct bch_inode_unpacked parent_inode; ret = bkey_err(inode_k) ?: !bkey_is_inode(inode_k.k) ? -BCH_ERR_ENOENT_inode - : bch2_inode_unpack(inode_k, &inode); + : bch2_inode_unpack(inode_k, &parent_inode); if (ret) { /* Should have been caught in dirents pass */ bch_err_msg(c, ret, "error looking up parent directory"); - break; + goto out; } + min_bi_depth = parent_inode.bi_depth; + + if (parent_inode.bi_depth < inode.bi_depth && + min_bi_depth < U16_MAX) + break; + + inode = parent_inode; snapshot = inode_k.k->p.snapshot; + redo_bi_depth = true; - if (path_is_dup(p, inode.bi_inum, snapshot)) { + if (path_is_dup(&path, inode.bi_inum, snapshot)) { /* XXX print path */ bch_err(c, "directory structure loop"); - darray_for_each(*p, i) + darray_for_each(path, i) pr_err("%llu:%u", i->inum, i->snapshot); pr_err("%llu:%u", inode.bi_inum, snapshot); @@ -2862,12 +2846,20 @@ static int check_path(struct btree_trans *trans, pathbuf *p, struct bkey_s_c ino ret = reattach_inode(trans, &inode); bch_err_msg(c, ret, "reattaching inode %llu", inode.bi_inum); } - break; + + goto out; } } + + if (inode.bi_subvol) + min_bi_depth = 0; + + if (redo_bi_depth) + ret = bch2_bi_depth_renumber(trans, &path, min_bi_depth); out: fsck_err: bch2_trans_iter_exit(trans, &inode_iter); + darray_exit(&path); printbuf_exit(&buf); bch_err_fn(c, ret); return ret; @@ -2879,24 +2871,20 @@ fsck_err: */ int bch2_check_directory_structure(struct bch_fs *c) { - pathbuf path = { 0, }; - int ret; - - ret = bch2_trans_run(c, + int ret = bch2_trans_run(c, for_each_btree_key_commit(trans, iter, BTREE_ID_inodes, POS_MIN, BTREE_ITER_intent| BTREE_ITER_prefetch| BTREE_ITER_all_snapshots, k, NULL, NULL, BCH_TRANS_COMMIT_no_enospc, ({ - if (!bkey_is_inode(k.k)) + if (!S_ISDIR(bkey_inode_mode(k))) continue; if (bch2_inode_flags(k) & BCH_INODE_unlinked) continue; - check_path(trans, &path, k); + check_path_loop(trans, k); }))); - darray_exit(&path); bch_err_fn(c, ret); return ret; @@ -2994,7 +2982,9 @@ static int check_nlinks_find_hardlinks(struct bch_fs *c, /* Should never fail, checked by bch2_inode_invalid: */ struct bch_inode_unpacked u; - BUG_ON(bch2_inode_unpack(k, &u)); + _ret3 = bch2_inode_unpack(k, &u); + if (_ret3) + break; /* * Backpointer and directory structure checks are sufficient for @@ -3072,7 +3062,9 @@ static int check_nlinks_update_inode(struct btree_trans *trans, struct btree_ite if (!bkey_is_inode(k.k)) return 0; - BUG_ON(bch2_inode_unpack(k, &u)); + ret = bch2_inode_unpack(k, &u); + if (ret) + return ret; if (S_ISDIR(u.bi_mode)) return 0; @@ -3194,3 +3186,223 @@ int bch2_fix_reflink_p(struct bch_fs *c) bch_err_fn(c, ret); return ret; } + +#ifndef NO_BCACHEFS_CHARDEV + +struct fsck_thread { + struct thread_with_stdio thr; + struct bch_fs *c; + struct bch_opts opts; +}; + +static void bch2_fsck_thread_exit(struct thread_with_stdio *_thr) +{ + struct fsck_thread *thr = container_of(_thr, struct fsck_thread, thr); + kfree(thr); +} + +static int bch2_fsck_offline_thread_fn(struct thread_with_stdio *stdio) +{ + struct fsck_thread *thr = container_of(stdio, struct fsck_thread, thr); + struct bch_fs *c = thr->c; + + int ret = PTR_ERR_OR_ZERO(c); + if (ret) + return ret; + + ret = bch2_fs_start(thr->c); + if (ret) + goto err; + + if (test_bit(BCH_FS_errors_fixed, &c->flags)) { + bch2_stdio_redirect_printf(&stdio->stdio, false, "%s: errors fixed\n", c->name); + ret |= 1; + } + if (test_bit(BCH_FS_error, &c->flags)) { + bch2_stdio_redirect_printf(&stdio->stdio, false, "%s: still has errors\n", c->name); + ret |= 4; + } +err: + bch2_fs_stop(c); + return ret; +} + +static const struct thread_with_stdio_ops bch2_offline_fsck_ops = { + .exit = bch2_fsck_thread_exit, + .fn = bch2_fsck_offline_thread_fn, +}; + +long bch2_ioctl_fsck_offline(struct bch_ioctl_fsck_offline __user *user_arg) +{ + struct bch_ioctl_fsck_offline arg; + struct fsck_thread *thr = NULL; + darray_str(devs) = {}; + long ret = 0; + + if (copy_from_user(&arg, user_arg, sizeof(arg))) + return -EFAULT; + + if (arg.flags) + return -EINVAL; + + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + + for (size_t i = 0; i < arg.nr_devs; i++) { + u64 dev_u64; + ret = copy_from_user_errcode(&dev_u64, &user_arg->devs[i], sizeof(u64)); + if (ret) + goto err; + + char *dev_str = strndup_user((char __user *)(unsigned long) dev_u64, PATH_MAX); + ret = PTR_ERR_OR_ZERO(dev_str); + if (ret) + goto err; + + ret = darray_push(&devs, dev_str); + if (ret) { + kfree(dev_str); + goto err; + } + } + + thr = kzalloc(sizeof(*thr), GFP_KERNEL); + if (!thr) { + ret = -ENOMEM; + goto err; + } + + thr->opts = bch2_opts_empty(); + + if (arg.opts) { + char *optstr = strndup_user((char __user *)(unsigned long) arg.opts, 1 << 16); + ret = PTR_ERR_OR_ZERO(optstr) ?: + bch2_parse_mount_opts(NULL, &thr->opts, NULL, optstr); + if (!IS_ERR(optstr)) + kfree(optstr); + + if (ret) + goto err; + } + + opt_set(thr->opts, stdio, (u64)(unsigned long)&thr->thr.stdio); + opt_set(thr->opts, read_only, 1); + opt_set(thr->opts, ratelimit_errors, 0); + + /* We need request_key() to be called before we punt to kthread: */ + opt_set(thr->opts, nostart, true); + + bch2_thread_with_stdio_init(&thr->thr, &bch2_offline_fsck_ops); + + thr->c = bch2_fs_open(devs.data, arg.nr_devs, thr->opts); + + if (!IS_ERR(thr->c) && + thr->c->opts.errors == BCH_ON_ERROR_panic) + thr->c->opts.errors = BCH_ON_ERROR_ro; + + ret = __bch2_run_thread_with_stdio(&thr->thr); +out: + darray_for_each(devs, i) + kfree(*i); + darray_exit(&devs); + return ret; +err: + if (thr) + bch2_fsck_thread_exit(&thr->thr); + pr_err("ret %s", bch2_err_str(ret)); + goto out; +} + +static int bch2_fsck_online_thread_fn(struct thread_with_stdio *stdio) +{ + struct fsck_thread *thr = container_of(stdio, struct fsck_thread, thr); + struct bch_fs *c = thr->c; + + c->stdio_filter = current; + c->stdio = &thr->thr.stdio; + + /* + * XXX: can we figure out a way to do this without mucking with c->opts? + */ + unsigned old_fix_errors = c->opts.fix_errors; + if (opt_defined(thr->opts, fix_errors)) + c->opts.fix_errors = thr->opts.fix_errors; + else + c->opts.fix_errors = FSCK_FIX_ask; + + c->opts.fsck = true; + set_bit(BCH_FS_fsck_running, &c->flags); + + c->curr_recovery_pass = BCH_RECOVERY_PASS_check_alloc_info; + int ret = bch2_run_online_recovery_passes(c); + + clear_bit(BCH_FS_fsck_running, &c->flags); + bch_err_fn(c, ret); + + c->stdio = NULL; + c->stdio_filter = NULL; + c->opts.fix_errors = old_fix_errors; + + up(&c->online_fsck_mutex); + bch2_ro_ref_put(c); + return ret; +} + +static const struct thread_with_stdio_ops bch2_online_fsck_ops = { + .exit = bch2_fsck_thread_exit, + .fn = bch2_fsck_online_thread_fn, +}; + +long bch2_ioctl_fsck_online(struct bch_fs *c, struct bch_ioctl_fsck_online arg) +{ + struct fsck_thread *thr = NULL; + long ret = 0; + + if (arg.flags) + return -EINVAL; + + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + + if (!bch2_ro_ref_tryget(c)) + return -EROFS; + + if (down_trylock(&c->online_fsck_mutex)) { + bch2_ro_ref_put(c); + return -EAGAIN; + } + + thr = kzalloc(sizeof(*thr), GFP_KERNEL); + if (!thr) { + ret = -ENOMEM; + goto err; + } + + thr->c = c; + thr->opts = bch2_opts_empty(); + + if (arg.opts) { + char *optstr = strndup_user((char __user *)(unsigned long) arg.opts, 1 << 16); + + ret = PTR_ERR_OR_ZERO(optstr) ?: + bch2_parse_mount_opts(c, &thr->opts, NULL, optstr); + if (!IS_ERR(optstr)) + kfree(optstr); + + if (ret) + goto err; + } + + ret = bch2_run_thread_with_stdio(&thr->thr, &bch2_online_fsck_ops); +err: + if (ret < 0) { + bch_err_fn(c, ret); + if (thr) + bch2_fsck_thread_exit(&thr->thr); + up(&c->online_fsck_mutex); + bch2_ro_ref_put(c); + } + return ret; +} + +#endif /* NO_BCACHEFS_CHARDEV */ diff --git a/fs/bcachefs/fsck.h b/fs/bcachefs/fsck.h index 1cca31011530..574948278cd4 100644 --- a/fs/bcachefs/fsck.h +++ b/fs/bcachefs/fsck.h @@ -2,6 +2,14 @@ #ifndef _BCACHEFS_FSCK_H #define _BCACHEFS_FSCK_H +#include "str_hash.h" + +int bch2_fsck_update_backpointers(struct btree_trans *, + struct snapshots_seen *, + const struct bch_hash_desc, + struct bch_hash_info *, + struct bkey_i *); + int bch2_check_inodes(struct bch_fs *); int bch2_check_extents(struct bch_fs *); int bch2_check_indirect_extents(struct bch_fs *); @@ -14,4 +22,7 @@ int bch2_check_directory_structure(struct bch_fs *); int bch2_check_nlinks(struct bch_fs *); int bch2_fix_reflink_p(struct bch_fs *); +long bch2_ioctl_fsck_offline(struct bch_ioctl_fsck_offline __user *); +long bch2_ioctl_fsck_online(struct bch_fs *, struct bch_ioctl_fsck_online); + #endif /* _BCACHEFS_FSCK_H */ diff --git a/fs/bcachefs/inode.c b/fs/bcachefs/inode.c index 039cb7a22244..04ec05206f8c 100644 --- a/fs/bcachefs/inode.c +++ b/fs/bcachefs/inode.c @@ -14,6 +14,7 @@ #include "extent_update.h" #include "fs.h" #include "inode.h" +#include "opts.h" #include "str_hash.h" #include "snapshot.h" #include "subvolume.h" @@ -47,10 +48,10 @@ static int inode_decode_field(const u8 *in, const u8 *end, u8 *p; if (in >= end) - return -1; + return -BCH_ERR_inode_unpack_error; if (!*in) - return -1; + return -BCH_ERR_inode_unpack_error; /* * position of highest set bit indicates number of bytes: @@ -60,7 +61,7 @@ static int inode_decode_field(const u8 *in, const u8 *end, bytes = byte_table[shift - 1]; if (in + bytes > end) - return -1; + return -BCH_ERR_inode_unpack_error; p = (u8 *) be + 16 - bytes; memcpy(p, in, bytes); @@ -176,7 +177,7 @@ static noinline int bch2_inode_unpack_v1(struct bkey_s_c_inode inode, return ret; \ \ if (field_bits > sizeof(unpacked->_name) * 8) \ - return -1; \ + return -BCH_ERR_inode_unpack_error; \ \ unpacked->_name = field[1]; \ in += ret; @@ -217,7 +218,7 @@ static int bch2_inode_unpack_v2(struct bch_inode_unpacked *unpacked, \ unpacked->_name = v[0]; \ if (v[1] || v[0] != unpacked->_name) \ - return -1; \ + return -BCH_ERR_inode_unpack_error; \ fieldnr++; BCH_INODE_FIELDS_v2() @@ -268,7 +269,7 @@ static int bch2_inode_unpack_v3(struct bkey_s_c k, \ unpacked->_name = v[0]; \ if (v[1] || v[0] != unpacked->_name) \ - return -1; \ + return -BCH_ERR_inode_unpack_error; \ fieldnr++; BCH_INODE_FIELDS_v3() @@ -428,7 +429,7 @@ struct bkey_i *bch2_inode_to_v3(struct btree_trans *trans, struct bkey_i *k) } static int __bch2_inode_validate(struct bch_fs *c, struct bkey_s_c k, - enum bch_validate_flags flags) + struct bkey_validate_context from) { struct bch_inode_unpacked unpacked; int ret = 0; @@ -468,7 +469,7 @@ fsck_err: } int bch2_inode_validate(struct bch_fs *c, struct bkey_s_c k, - enum bch_validate_flags flags) + struct bkey_validate_context from) { struct bkey_s_c_inode inode = bkey_s_c_to_inode(k); int ret = 0; @@ -478,13 +479,13 @@ int bch2_inode_validate(struct bch_fs *c, struct bkey_s_c k, "invalid str hash type (%llu >= %u)", INODEv1_STR_HASH(inode.v), BCH_STR_HASH_NR); - ret = __bch2_inode_validate(c, k, flags); + ret = __bch2_inode_validate(c, k, from); fsck_err: return ret; } int bch2_inode_v2_validate(struct bch_fs *c, struct bkey_s_c k, - enum bch_validate_flags flags) + struct bkey_validate_context from) { struct bkey_s_c_inode_v2 inode = bkey_s_c_to_inode_v2(k); int ret = 0; @@ -494,13 +495,13 @@ int bch2_inode_v2_validate(struct bch_fs *c, struct bkey_s_c k, "invalid str hash type (%llu >= %u)", INODEv2_STR_HASH(inode.v), BCH_STR_HASH_NR); - ret = __bch2_inode_validate(c, k, flags); + ret = __bch2_inode_validate(c, k, from); fsck_err: return ret; } int bch2_inode_v3_validate(struct bch_fs *c, struct bkey_s_c k, - enum bch_validate_flags flags) + struct bkey_validate_context from) { struct bkey_s_c_inode_v3 inode = bkey_s_c_to_inode_v3(k); int ret = 0; @@ -518,7 +519,7 @@ int bch2_inode_v3_validate(struct bch_fs *c, struct bkey_s_c k, "invalid str hash type (%llu >= %u)", INODEv3_STR_HASH(inode.v), BCH_STR_HASH_NR); - ret = __bch2_inode_validate(c, k, flags); + ret = __bch2_inode_validate(c, k, from); fsck_err: return ret; } @@ -617,7 +618,7 @@ bch2_bkey_get_iter_snapshot_parent(struct btree_trans *trans, struct btree_iter struct bkey_s_c k; int ret = 0; - for_each_btree_key_upto_norestart(trans, *iter, btree, + for_each_btree_key_max_norestart(trans, *iter, btree, bpos_successor(pos), SPOS(pos.inode, pos.offset, U32_MAX), flags|BTREE_ITER_all_snapshots, k, ret) @@ -652,7 +653,7 @@ int __bch2_inode_has_child_snapshots(struct btree_trans *trans, struct bpos pos) struct bkey_s_c k; int ret = 0; - for_each_btree_key_upto_norestart(trans, iter, + for_each_btree_key_max_norestart(trans, iter, BTREE_ID_inodes, POS(0, pos.offset), bpos_predecessor(pos), BTREE_ITER_all_snapshots| BTREE_ITER_with_updates, k, ret) @@ -779,7 +780,7 @@ int bch2_trigger_inode(struct btree_trans *trans, } int bch2_inode_generation_validate(struct bch_fs *c, struct bkey_s_c k, - enum bch_validate_flags flags) + struct bkey_validate_context from) { int ret = 0; @@ -798,6 +799,28 @@ void bch2_inode_generation_to_text(struct printbuf *out, struct bch_fs *c, prt_printf(out, "generation: %u", le32_to_cpu(gen.v->bi_generation)); } +int bch2_inode_alloc_cursor_validate(struct bch_fs *c, struct bkey_s_c k, + struct bkey_validate_context from) +{ + int ret = 0; + + bkey_fsck_err_on(k.k->p.inode != LOGGED_OPS_INUM_inode_cursors, + c, inode_alloc_cursor_inode_bad, + "k.p.inode bad"); +fsck_err: + return ret; +} + +void bch2_inode_alloc_cursor_to_text(struct printbuf *out, struct bch_fs *c, + struct bkey_s_c k) +{ + struct bkey_s_c_inode_alloc_cursor i = bkey_s_c_to_inode_alloc_cursor(k); + + prt_printf(out, "idx %llu generation %llu", + le64_to_cpu(i.v->idx), + le64_to_cpu(i.v->gen)); +} + void bch2_inode_init_early(struct bch_fs *c, struct bch_inode_unpacked *inode_u) { @@ -858,43 +881,78 @@ static inline u32 bkey_generation(struct bkey_s_c k) } } -/* - * This just finds an empty slot: - */ -int bch2_inode_create(struct btree_trans *trans, - struct btree_iter *iter, - struct bch_inode_unpacked *inode_u, - u32 snapshot, u64 cpu) +static struct bkey_i_inode_alloc_cursor * +bch2_inode_alloc_cursor_get(struct btree_trans *trans, u64 cpu, u64 *min, u64 *max) { struct bch_fs *c = trans->c; - struct bkey_s_c k; - u64 min, max, start, pos, *hint; - int ret = 0; - unsigned bits = (c->opts.inodes_32bit ? 31 : 63); - if (c->opts.shard_inode_numbers) { - bits -= c->inode_shard_bits; + u64 cursor_idx = c->opts.inodes_32bit ? 0 : cpu + 1; - min = (cpu << bits); - max = (cpu << bits) | ~(ULLONG_MAX << bits); + cursor_idx &= ~(~0ULL << c->opts.shard_inode_numbers_bits); - min = max_t(u64, min, BLOCKDEV_INODE_MAX); - hint = c->unused_inode_hints + cpu; + struct btree_iter iter; + struct bkey_s_c k = bch2_bkey_get_iter(trans, &iter, + BTREE_ID_logged_ops, + POS(LOGGED_OPS_INUM_inode_cursors, cursor_idx), + BTREE_ITER_cached); + int ret = bkey_err(k); + if (ret) + return ERR_PTR(ret); + + struct bkey_i_inode_alloc_cursor *cursor = + k.k->type == KEY_TYPE_inode_alloc_cursor + ? bch2_bkey_make_mut_typed(trans, &iter, &k, 0, inode_alloc_cursor) + : bch2_bkey_alloc(trans, &iter, 0, inode_alloc_cursor); + ret = PTR_ERR_OR_ZERO(cursor); + if (ret) + goto err; + + if (c->opts.inodes_32bit) { + *min = BLOCKDEV_INODE_MAX; + *max = INT_MAX; } else { - min = BLOCKDEV_INODE_MAX; - max = ~(ULLONG_MAX << bits); - hint = c->unused_inode_hints; + cursor->v.bits = c->opts.shard_inode_numbers_bits; + + unsigned bits = 63 - c->opts.shard_inode_numbers_bits; + + *min = max(cpu << bits, (u64) INT_MAX + 1); + *max = (cpu << bits) | ~(ULLONG_MAX << bits); } - start = READ_ONCE(*hint); + if (le64_to_cpu(cursor->v.idx) < *min) + cursor->v.idx = cpu_to_le64(*min); - if (start >= max || start < min) - start = min; + if (le64_to_cpu(cursor->v.idx) >= *max) { + cursor->v.idx = cpu_to_le64(*min); + le32_add_cpu(&cursor->v.gen, 1); + } +err: + bch2_trans_iter_exit(trans, &iter); + return ret ? ERR_PTR(ret) : cursor; +} + +/* + * This just finds an empty slot: + */ +int bch2_inode_create(struct btree_trans *trans, + struct btree_iter *iter, + struct bch_inode_unpacked *inode_u, + u32 snapshot, u64 cpu) +{ + u64 min, max; + struct bkey_i_inode_alloc_cursor *cursor = + bch2_inode_alloc_cursor_get(trans, cpu, &min, &max); + int ret = PTR_ERR_OR_ZERO(cursor); + if (ret) + return ret; + + u64 start = le64_to_cpu(cursor->v.idx); + u64 pos = start; - pos = start; bch2_trans_iter_init(trans, iter, BTREE_ID_inodes, POS(0, pos), BTREE_ITER_all_snapshots| BTREE_ITER_intent); + struct bkey_s_c k; again: while ((k = bch2_btree_iter_peek(iter)).k && !(ret = bkey_err(k)) && @@ -924,6 +982,7 @@ again: /* Retry from start */ pos = start = min; bch2_btree_iter_set_pos(iter, POS(0, pos)); + le32_add_cpu(&cursor->v.gen, 1); goto again; found_slot: bch2_btree_iter_set_pos(iter, SPOS(0, pos, snapshot)); @@ -934,9 +993,9 @@ found_slot: return ret; } - *hint = k.k->p.offset; inode_u->bi_inum = k.k->p.offset; - inode_u->bi_generation = bkey_generation(k); + inode_u->bi_generation = le64_to_cpu(cursor->v.gen); + cursor->v.idx = cpu_to_le64(k.k->p.offset + 1); return 0; } @@ -966,7 +1025,7 @@ static int bch2_inode_delete_keys(struct btree_trans *trans, bch2_btree_iter_set_snapshot(&iter, snapshot); - k = bch2_btree_iter_peek_upto(&iter, end); + k = bch2_btree_iter_peek_max(&iter, end); ret = bkey_err(k); if (ret) goto err; @@ -998,8 +1057,6 @@ int bch2_inode_rm(struct bch_fs *c, subvol_inum inum) { struct btree_trans *trans = bch2_trans_get(c); struct btree_iter iter = { NULL }; - struct bkey_i_inode_generation delete; - struct bch_inode_unpacked inode_u; struct bkey_s_c k; u32 snapshot; int ret; @@ -1039,13 +1096,7 @@ retry: goto err; } - bch2_inode_unpack(k, &inode_u); - - bkey_inode_generation_init(&delete.k_i); - delete.k.p = iter.pos; - delete.v.bi_generation = cpu_to_le32(inode_u.bi_generation + 1); - - ret = bch2_trans_update(trans, &iter, &delete.k_i, 0) ?: + ret = bch2_btree_delete_at(trans, &iter, 0) ?: bch2_trans_commit(trans, NULL, NULL, BCH_TRANS_COMMIT_no_enospc); err: @@ -1141,12 +1192,17 @@ struct bch_opts bch2_inode_opts_to_opts(struct bch_inode_unpacked *inode) void bch2_inode_opts_get(struct bch_io_opts *opts, struct bch_fs *c, struct bch_inode_unpacked *inode) { -#define x(_name, _bits) opts->_name = inode_opt_get(c, inode, _name); +#define x(_name, _bits) \ + if ((inode)->bi_##_name) { \ + opts->_name = inode->bi_##_name - 1; \ + opts->_name##_from_inode = true; \ + } else { \ + opts->_name = c->opts._name; \ + } BCH_INODE_OPTS() #undef x - if (opts->nocow) - opts->compression = opts->background_compression = opts->data_checksum = opts->erasure_code = 0; + bch2_io_opts_fixups(opts); } int bch2_inum_opts_get(struct btree_trans *trans, subvol_inum inum, struct bch_io_opts *opts) @@ -1380,7 +1436,8 @@ again: NULL, NULL, BCH_TRANS_COMMIT_no_enospc, ({ ret = may_delete_deleted_inode(trans, &iter, k.k->p, &need_another_pass); if (ret > 0) { - bch_verbose(c, "deleting unlinked inode %llu:%u", k.k->p.offset, k.k->p.snapshot); + bch_verbose_ratelimited(c, "deleting unlinked inode %llu:%u", + k.k->p.offset, k.k->p.snapshot); ret = bch2_inode_rm_snapshot(trans, k.k->p.offset, k.k->p.snapshot); /* diff --git a/fs/bcachefs/inode.h b/fs/bcachefs/inode.h index eab82b5eb897..d2e134528f0e 100644 --- a/fs/bcachefs/inode.h +++ b/fs/bcachefs/inode.h @@ -7,15 +7,14 @@ #include "opts.h" #include "snapshot.h" -enum bch_validate_flags; extern const char * const bch2_inode_opts[]; int bch2_inode_validate(struct bch_fs *, struct bkey_s_c, - enum bch_validate_flags); + struct bkey_validate_context); int bch2_inode_v2_validate(struct bch_fs *, struct bkey_s_c, - enum bch_validate_flags); + struct bkey_validate_context); int bch2_inode_v3_validate(struct bch_fs *, struct bkey_s_c, - enum bch_validate_flags); + struct bkey_validate_context); void bch2_inode_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c); int __bch2_inode_has_child_snapshots(struct btree_trans *, struct bpos); @@ -60,7 +59,7 @@ static inline bool bkey_is_inode(const struct bkey *k) } int bch2_inode_generation_validate(struct bch_fs *, struct bkey_s_c, - enum bch_validate_flags); + struct bkey_validate_context); void bch2_inode_generation_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c); #define bch2_bkey_ops_inode_generation ((struct bkey_ops) { \ @@ -69,6 +68,16 @@ void bch2_inode_generation_to_text(struct printbuf *, struct bch_fs *, struct bk .min_val_size = 8, \ }) +int bch2_inode_alloc_cursor_validate(struct bch_fs *, struct bkey_s_c, + struct bkey_validate_context); +void bch2_inode_alloc_cursor_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c); + +#define bch2_bkey_ops_inode_alloc_cursor ((struct bkey_ops) { \ + .key_validate = bch2_inode_alloc_cursor_validate, \ + .val_to_text = bch2_inode_alloc_cursor_to_text, \ + .min_val_size = 16, \ +}) + #if 0 typedef struct { u64 lo; @@ -220,6 +229,20 @@ static inline u32 bch2_inode_flags(struct bkey_s_c k) } } +static inline unsigned bkey_inode_mode(struct bkey_s_c k) +{ + switch (k.k->type) { + case KEY_TYPE_inode: + return le16_to_cpu(bkey_s_c_to_inode(k).v->bi_mode); + case KEY_TYPE_inode_v2: + return le16_to_cpu(bkey_s_c_to_inode_v2(k).v->bi_mode); + case KEY_TYPE_inode_v3: + return INODEv3_MODE(bkey_s_c_to_inode_v3(k).v); + default: + return 0; + } +} + /* i_nlink: */ static inline unsigned nlink_bias(umode_t mode) @@ -249,7 +272,7 @@ static inline void bch2_inode_nlink_set(struct bch_inode_unpacked *bi, int bch2_inode_nlink_inc(struct bch_inode_unpacked *); void bch2_inode_nlink_dec(struct btree_trans *, struct bch_inode_unpacked *); -static inline bool bch2_inode_should_have_bp(struct bch_inode_unpacked *inode) +static inline bool bch2_inode_should_have_single_bp(struct bch_inode_unpacked *inode) { bool inode_has_bp = inode->bi_dir || inode->bi_dir_offset; @@ -262,6 +285,14 @@ void bch2_inode_opts_get(struct bch_io_opts *, struct bch_fs *, struct bch_inode_unpacked *); int bch2_inum_opts_get(struct btree_trans*, subvol_inum, struct bch_io_opts *); +static inline struct bch_extent_rebalance +bch2_inode_rebalance_opts_get(struct bch_fs *c, struct bch_inode_unpacked *inode) +{ + struct bch_io_opts io_opts; + bch2_inode_opts_get(&io_opts, c, inode); + return io_opts_to_rebalance_opts(&io_opts); +} + int bch2_inode_rm_snapshot(struct btree_trans *, u64, u32); int bch2_delete_dead_inodes(struct bch_fs *); diff --git a/fs/bcachefs/inode_format.h b/fs/bcachefs/inode_format.h index 7928d0c6954f..b99a5bf1a75e 100644 --- a/fs/bcachefs/inode_format.h +++ b/fs/bcachefs/inode_format.h @@ -101,7 +101,9 @@ struct bch_inode_generation { x(bi_dir_offset, 64) \ x(bi_subvol, 32) \ x(bi_parent_subvol, 32) \ - x(bi_nocow, 8) + x(bi_nocow, 8) \ + x(bi_depth, 32) \ + x(bi_inodes_32bit, 8) /* subset of BCH_INODE_FIELDS */ #define BCH_INODE_OPTS() \ @@ -114,7 +116,8 @@ struct bch_inode_generation { x(foreground_target, 16) \ x(background_target, 16) \ x(erasure_code, 16) \ - x(nocow, 8) + x(nocow, 8) \ + x(inodes_32bit, 8) enum inode_opt_id { #define x(name, ...) \ @@ -164,4 +167,12 @@ LE64_BITMASK(INODEv3_FIELDS_START, struct bch_inode_v3, bi_flags, 31, 36); LE64_BITMASK(INODEv3_MODE, struct bch_inode_v3, bi_flags, 36, 52); +struct bch_inode_alloc_cursor { + struct bch_val v; + __u8 bits; + __u8 pad; + __le32 gen; + __le64 idx; +}; + #endif /* _BCACHEFS_INODE_FORMAT_H */ diff --git a/fs/bcachefs/io_misc.c b/fs/bcachefs/io_misc.c index f283051758d6..5353979117b0 100644 --- a/fs/bcachefs/io_misc.c +++ b/fs/bcachefs/io_misc.c @@ -113,11 +113,13 @@ int bch2_extent_fallocate(struct btree_trans *trans, err: if (!ret && sectors_allocated) bch2_increment_clock(c, sectors_allocated, WRITE); - if (should_print_err(ret)) - bch_err_inum_offset_ratelimited(c, - inum.inum, - iter->pos.offset << 9, - "%s(): error: %s", __func__, bch2_err_str(ret)); + if (should_print_err(ret)) { + struct printbuf buf = PRINTBUF; + bch2_inum_offset_err_msg_trans(trans, &buf, inum, iter->pos.offset << 9); + prt_printf(&buf, "fallocate error: %s", bch2_err_str(ret)); + bch_err_ratelimited(c, "%s", buf.buf); + printbuf_exit(&buf); + } err_noprint: bch2_open_buckets_put(c, &open_buckets); bch2_disk_reservation_put(c, &disk_res); @@ -164,9 +166,9 @@ int bch2_fpunch_at(struct btree_trans *trans, struct btree_iter *iter, bch2_btree_iter_set_snapshot(iter, snapshot); /* - * peek_upto() doesn't have ideal semantics for extents: + * peek_max() doesn't have ideal semantics for extents: */ - k = bch2_btree_iter_peek_upto(iter, end_pos); + k = bch2_btree_iter_peek_max(iter, end_pos); if (!k.k) break; @@ -426,8 +428,8 @@ case LOGGED_OP_FINSERT_shift_extents: bch2_btree_iter_set_pos(&iter, SPOS(inum.inum, pos, snapshot)); k = insert - ? bch2_btree_iter_peek_prev(&iter) - : bch2_btree_iter_peek_upto(&iter, POS(inum.inum, U64_MAX)); + ? bch2_btree_iter_peek_prev_min(&iter, POS(inum.inum, 0)) + : bch2_btree_iter_peek_max(&iter, POS(inum.inum, U64_MAX)); if ((ret = bkey_err(k))) goto btree_err; @@ -461,7 +463,7 @@ case LOGGED_OP_FINSERT_shift_extents: op->v.pos = cpu_to_le64(insert ? bkey_start_offset(&delete.k) : delete.k.p.offset); - ret = bch2_bkey_set_needs_rebalance(c, copy, &opts) ?: + ret = bch2_bkey_set_needs_rebalance(c, &opts, copy) ?: bch2_btree_insert_trans(trans, BTREE_ID_extents, &delete, 0) ?: bch2_btree_insert_trans(trans, BTREE_ID_extents, copy, 0) ?: bch2_logged_op_update(trans, &op->k_i) ?: diff --git a/fs/bcachefs/io_read.c b/fs/bcachefs/io_read.c index b3b934a87c6d..8c7b2d3d779d 100644 --- a/fs/bcachefs/io_read.c +++ b/fs/bcachefs/io_read.c @@ -21,6 +21,7 @@ #include "io_read.h" #include "io_misc.h" #include "io_write.h" +#include "reflink.h" #include "subvolume.h" #include "trace.h" @@ -90,13 +91,18 @@ static const struct rhashtable_params bch_promote_params = { .automatic_shrinking = true, }; +static inline bool have_io_error(struct bch_io_failures *failed) +{ + return failed && failed->nr; +} + static inline int should_promote(struct bch_fs *c, struct bkey_s_c k, struct bpos pos, struct bch_io_opts opts, unsigned flags, struct bch_io_failures *failed) { - if (!failed) { + if (!have_io_error(failed)) { BUG_ON(!opts.promote_target); if (!(flags & BCH_READ_MAY_PROMOTE)) @@ -223,7 +229,7 @@ static struct promote_op *__promote_alloc(struct btree_trans *trans, struct data_update_opts update_opts = {}; - if (!failed) { + if (!have_io_error(failed)) { update_opts.target = opts.promote_target; update_opts.extra_replicas = 1; update_opts.write_flags = BCH_WRITE_ALLOC_NOWAIT|BCH_WRITE_CACHED; @@ -231,11 +237,11 @@ static struct promote_op *__promote_alloc(struct btree_trans *trans, update_opts.target = opts.foreground_target; struct bkey_ptrs_c ptrs = bch2_bkey_ptrs_c(k); - unsigned i = 0; + unsigned ptr_bit = 1; bkey_for_each_ptr(ptrs, ptr) { if (bch2_dev_io_failures(failed, ptr->dev)) - update_opts.rewrite_ptrs |= BIT(i); - i++; + update_opts.rewrite_ptrs |= ptr_bit; + ptr_bit <<= 1; } } @@ -285,7 +291,7 @@ static struct promote_op *promote_alloc(struct btree_trans *trans, * if failed != NULL we're not actually doing a promote, we're * recovering from an io/checksum error */ - bool promote_full = (failed || + bool promote_full = (have_io_error(failed) || *read_full || READ_ONCE(c->opts.promote_whole_extents)); /* data might have to be decompressed in the write path: */ @@ -321,6 +327,20 @@ nopromote: /* Read */ +static int bch2_read_err_msg_trans(struct btree_trans *trans, struct printbuf *out, + struct bch_read_bio *rbio, struct bpos read_pos) +{ + return bch2_inum_offset_err_msg_trans(trans, out, + (subvol_inum) { rbio->subvol, read_pos.inode }, + read_pos.offset << 9); +} + +static void bch2_read_err_msg(struct bch_fs *c, struct printbuf *out, + struct bch_read_bio *rbio, struct bpos read_pos) +{ + bch2_trans_run(c, bch2_read_err_msg_trans(trans, out, rbio, read_pos)); +} + #define READ_RETRY_AVOID 1 #define READ_RETRY 2 #define READ_ERR 3 @@ -499,6 +519,29 @@ static void bch2_rbio_error(struct bch_read_bio *rbio, int retry, } } +static void bch2_read_io_err(struct work_struct *work) +{ + struct bch_read_bio *rbio = + container_of(work, struct bch_read_bio, work); + struct bio *bio = &rbio->bio; + struct bch_fs *c = rbio->c; + struct bch_dev *ca = rbio->have_ioref ? bch2_dev_have_ref(c, rbio->pick.ptr.dev) : NULL; + struct printbuf buf = PRINTBUF; + + bch2_read_err_msg(c, &buf, rbio, rbio->read_pos); + prt_printf(&buf, "data read error: %s", bch2_blk_status_to_str(bio->bi_status)); + + if (ca) { + bch2_io_error(ca, BCH_MEMBER_ERROR_read); + bch_err_ratelimited(ca, "%s", buf.buf); + } else { + bch_err_ratelimited(c, "%s", buf.buf); + } + + printbuf_exit(&buf); + bch2_rbio_error(rbio, READ_RETRY_AVOID, bio->bi_status); +} + static int __bch2_rbio_narrow_crcs(struct btree_trans *trans, struct bch_read_bio *rbio) { @@ -562,6 +605,73 @@ static noinline void bch2_rbio_narrow_crcs(struct bch_read_bio *rbio) __bch2_rbio_narrow_crcs(trans, rbio)); } +static void bch2_read_csum_err(struct work_struct *work) +{ + struct bch_read_bio *rbio = + container_of(work, struct bch_read_bio, work); + struct bch_fs *c = rbio->c; + struct bio *src = &rbio->bio; + struct bch_extent_crc_unpacked crc = rbio->pick.crc; + struct nonce nonce = extent_nonce(rbio->version, crc); + struct bch_csum csum = bch2_checksum_bio(c, crc.csum_type, nonce, src); + struct printbuf buf = PRINTBUF; + + bch2_read_err_msg(c, &buf, rbio, rbio->read_pos); + prt_str(&buf, "data "); + bch2_csum_err_msg(&buf, crc.csum_type, rbio->pick.crc.csum, csum); + + struct bch_dev *ca = rbio->have_ioref ? bch2_dev_have_ref(c, rbio->pick.ptr.dev) : NULL; + if (ca) { + bch2_io_error(ca, BCH_MEMBER_ERROR_checksum); + bch_err_ratelimited(ca, "%s", buf.buf); + } else { + bch_err_ratelimited(c, "%s", buf.buf); + } + + bch2_rbio_error(rbio, READ_RETRY_AVOID, BLK_STS_IOERR); + printbuf_exit(&buf); +} + +static void bch2_read_decompress_err(struct work_struct *work) +{ + struct bch_read_bio *rbio = + container_of(work, struct bch_read_bio, work); + struct bch_fs *c = rbio->c; + struct printbuf buf = PRINTBUF; + + bch2_read_err_msg(c, &buf, rbio, rbio->read_pos); + prt_str(&buf, "decompression error"); + + struct bch_dev *ca = rbio->have_ioref ? bch2_dev_have_ref(c, rbio->pick.ptr.dev) : NULL; + if (ca) + bch_err_ratelimited(ca, "%s", buf.buf); + else + bch_err_ratelimited(c, "%s", buf.buf); + + bch2_rbio_error(rbio, READ_ERR, BLK_STS_IOERR); + printbuf_exit(&buf); +} + +static void bch2_read_decrypt_err(struct work_struct *work) +{ + struct bch_read_bio *rbio = + container_of(work, struct bch_read_bio, work); + struct bch_fs *c = rbio->c; + struct printbuf buf = PRINTBUF; + + bch2_read_err_msg(c, &buf, rbio, rbio->read_pos); + prt_str(&buf, "decrypt error"); + + struct bch_dev *ca = rbio->have_ioref ? bch2_dev_have_ref(c, rbio->pick.ptr.dev) : NULL; + if (ca) + bch_err_ratelimited(ca, "%s", buf.buf); + else + bch_err_ratelimited(c, "%s", buf.buf); + + bch2_rbio_error(rbio, READ_ERR, BLK_STS_IOERR); + printbuf_exit(&buf); +} + /* Inner part that may run in process context */ static void __bch2_read_endio(struct work_struct *work) { @@ -668,33 +778,13 @@ csum_err: goto out; } - struct printbuf buf = PRINTBUF; - buf.atomic++; - prt_str(&buf, "data "); - bch2_csum_err_msg(&buf, crc.csum_type, rbio->pick.crc.csum, csum); - - struct bch_dev *ca = rbio->have_ioref ? bch2_dev_have_ref(c, rbio->pick.ptr.dev) : NULL; - if (ca) { - bch_err_inum_offset_ratelimited(ca, - rbio->read_pos.inode, - rbio->read_pos.offset << 9, - "data %s", buf.buf); - bch2_io_error(ca, BCH_MEMBER_ERROR_checksum); - } - printbuf_exit(&buf); - bch2_rbio_error(rbio, READ_RETRY_AVOID, BLK_STS_IOERR); + bch2_rbio_punt(rbio, bch2_read_csum_err, RBIO_CONTEXT_UNBOUND, system_unbound_wq); goto out; decompression_err: - bch_err_inum_offset_ratelimited(c, rbio->read_pos.inode, - rbio->read_pos.offset << 9, - "decompression error"); - bch2_rbio_error(rbio, READ_ERR, BLK_STS_IOERR); + bch2_rbio_punt(rbio, bch2_read_decompress_err, RBIO_CONTEXT_UNBOUND, system_unbound_wq); goto out; decrypt_err: - bch_err_inum_offset_ratelimited(c, rbio->read_pos.inode, - rbio->read_pos.offset << 9, - "decrypt error"); - bch2_rbio_error(rbio, READ_ERR, BLK_STS_IOERR); + bch2_rbio_punt(rbio, bch2_read_decrypt_err, RBIO_CONTEXT_UNBOUND, system_unbound_wq); goto out; } @@ -715,16 +805,8 @@ static void bch2_read_endio(struct bio *bio) if (!rbio->split) rbio->bio.bi_end_io = rbio->end_io; - if (bio->bi_status) { - if (ca) { - bch_err_inum_offset_ratelimited(ca, - rbio->read_pos.inode, - rbio->read_pos.offset, - "data read error: %s", - bch2_blk_status_to_str(bio->bi_status)); - bch2_io_error(ca, BCH_MEMBER_ERROR_read); - } - bch2_rbio_error(rbio, READ_RETRY_AVOID, bio->bi_status); + if (unlikely(bio->bi_status)) { + bch2_rbio_punt(rbio, bch2_read_io_err, RBIO_CONTEXT_UNBOUND, system_unbound_wq); return; } @@ -750,45 +832,6 @@ static void bch2_read_endio(struct bio *bio) bch2_rbio_punt(rbio, __bch2_read_endio, context, wq); } -int __bch2_read_indirect_extent(struct btree_trans *trans, - unsigned *offset_into_extent, - struct bkey_buf *orig_k) -{ - struct btree_iter iter; - struct bkey_s_c k; - u64 reflink_offset; - int ret; - - reflink_offset = le64_to_cpu(bkey_i_to_reflink_p(orig_k->k)->v.idx) + - *offset_into_extent; - - k = bch2_bkey_get_iter(trans, &iter, BTREE_ID_reflink, - POS(0, reflink_offset), 0); - ret = bkey_err(k); - if (ret) - goto err; - - if (k.k->type != KEY_TYPE_reflink_v && - k.k->type != KEY_TYPE_indirect_inline_data) { - bch_err_inum_offset_ratelimited(trans->c, - orig_k->k->k.p.inode, - orig_k->k->k.p.offset << 9, - "%llu len %u points to nonexistent indirect extent %llu", - orig_k->k->k.p.offset, - orig_k->k->k.size, - reflink_offset); - bch2_inconsistent_error(trans->c); - ret = -BCH_ERR_missing_indirect_extent; - goto err; - } - - *offset_into_extent = iter.pos.offset - bkey_start_offset(k.k); - bch2_bkey_buf_reassemble(orig_k, trans->c, k); -err: - bch2_trans_iter_exit(trans, &iter); - return ret; -} - static noinline void read_from_stale_dirty_pointer(struct btree_trans *trans, struct bch_dev *ca, struct bkey_s_c k, @@ -868,15 +911,24 @@ retry_pick: if (!pick_ret) goto hole; - if (pick_ret < 0) { + if (unlikely(pick_ret < 0)) { + struct printbuf buf = PRINTBUF; + bch2_read_err_msg_trans(trans, &buf, orig, read_pos); + prt_printf(&buf, "no device to read from: %s\n ", bch2_err_str(pick_ret)); + bch2_bkey_val_to_text(&buf, c, k); + + bch_err_ratelimited(c, "%s", buf.buf); + printbuf_exit(&buf); + goto err; + } + + if (unlikely(bch2_csum_type_is_encryption(pick.crc.csum_type)) && !c->chacha20) { struct printbuf buf = PRINTBUF; + bch2_read_err_msg_trans(trans, &buf, orig, read_pos); + prt_printf(&buf, "attempting to read encrypted data without encryption key\n "); bch2_bkey_val_to_text(&buf, c, k); - bch_err_inum_offset_ratelimited(c, - read_pos.inode, read_pos.offset << 9, - "no device to read from: %s\n %s", - bch2_err_str(pick_ret), - buf.buf); + bch_err_ratelimited(c, "%s", buf.buf); printbuf_exit(&buf); goto err; } @@ -942,7 +994,7 @@ retry_pick: bounce = true; } - if (orig->opts.promote_target)// || failed) + if (orig->opts.promote_target || have_io_error(failed)) promote = promote_alloc(trans, iter, k, &pick, orig->opts, flags, &rbio, &bounce, &read_full, failed); @@ -1062,11 +1114,15 @@ get_bio: } if (!rbio->pick.idx) { - if (!rbio->have_ioref) { - bch_err_inum_offset_ratelimited(c, - read_pos.inode, - read_pos.offset << 9, - "no device to read from"); + if (unlikely(!rbio->have_ioref)) { + struct printbuf buf = PRINTBUF; + bch2_read_err_msg_trans(trans, &buf, rbio, read_pos); + prt_printf(&buf, "no device to read from:\n "); + bch2_bkey_val_to_text(&buf, c, k); + + bch_err_ratelimited(c, "%s", buf.buf); + printbuf_exit(&buf); + bch2_rbio_error(rbio, READ_RETRY_AVOID, BLK_STS_IOERR); goto out; } @@ -1164,7 +1220,6 @@ void __bch2_read(struct bch_fs *c, struct bch_read_bio *rbio, BTREE_ITER_slots); while (1) { - unsigned bytes, sectors, offset_into_extent; enum btree_id data_btree = BTREE_ID_extents; bch2_trans_begin(trans); @@ -1184,9 +1239,9 @@ void __bch2_read(struct bch_fs *c, struct bch_read_bio *rbio, if (ret) goto err; - offset_into_extent = iter.pos.offset - + s64 offset_into_extent = iter.pos.offset - bkey_start_offset(k.k); - sectors = k.k->size - offset_into_extent; + unsigned sectors = k.k->size - offset_into_extent; bch2_bkey_buf_reassemble(&sk, c, k); @@ -1201,9 +1256,9 @@ void __bch2_read(struct bch_fs *c, struct bch_read_bio *rbio, * With indirect extents, the amount of data to read is the min * of the original extent and the indirect extent: */ - sectors = min(sectors, k.k->size - offset_into_extent); + sectors = min_t(unsigned, sectors, k.k->size - offset_into_extent); - bytes = min(sectors, bvec_iter_sectors(bvec_iter)) << 9; + unsigned bytes = min(sectors, bvec_iter_sectors(bvec_iter)) << 9; swap(bvec_iter.bi_size, bytes); if (bvec_iter.bi_size == bytes) @@ -1229,16 +1284,20 @@ err: } bch2_trans_iter_exit(trans, &iter); - bch2_trans_put(trans); - bch2_bkey_buf_exit(&sk, c); if (ret) { - bch_err_inum_offset_ratelimited(c, inum.inum, - bvec_iter.bi_sector << 9, - "read error %i from btree lookup", ret); + struct printbuf buf = PRINTBUF; + bch2_inum_offset_err_msg_trans(trans, &buf, inum, bvec_iter.bi_sector << 9); + prt_printf(&buf, "read error %i from btree lookup", ret); + bch_err_ratelimited(c, "%s", buf.buf); + printbuf_exit(&buf); + rbio->bio.bi_status = BLK_STS_IOERR; bch2_rbio_done(rbio); } + + bch2_trans_put(trans); + bch2_bkey_buf_exit(&sk, c); } void bch2_fs_io_read_exit(struct bch_fs *c) diff --git a/fs/bcachefs/io_read.h b/fs/bcachefs/io_read.h index d9c18bb7d403..a82e8a94ccb6 100644 --- a/fs/bcachefs/io_read.h +++ b/fs/bcachefs/io_read.h @@ -3,6 +3,7 @@ #define _BCACHEFS_IO_READ_H #include "bkey_buf.h" +#include "reflink.h" struct bch_read_bio { struct bch_fs *c; @@ -79,19 +80,32 @@ struct bch_devs_mask; struct cache_promote_op; struct extent_ptr_decoded; -int __bch2_read_indirect_extent(struct btree_trans *, unsigned *, - struct bkey_buf *); - static inline int bch2_read_indirect_extent(struct btree_trans *trans, enum btree_id *data_btree, - unsigned *offset_into_extent, - struct bkey_buf *k) + s64 *offset_into_extent, + struct bkey_buf *extent) { - if (k->k->k.type != KEY_TYPE_reflink_p) + if (extent->k->k.type != KEY_TYPE_reflink_p) return 0; *data_btree = BTREE_ID_reflink; - return __bch2_read_indirect_extent(trans, offset_into_extent, k); + struct btree_iter iter; + struct bkey_s_c k = bch2_lookup_indirect_extent(trans, &iter, + offset_into_extent, + bkey_i_to_s_c_reflink_p(extent->k), + true, 0); + int ret = bkey_err(k); + if (ret) + return ret; + + if (bkey_deleted(k.k)) { + bch2_trans_iter_exit(trans, &iter); + return -BCH_ERR_missing_indirect_extent; + } + + bch2_bkey_buf_reassemble(extent, trans->c, k); + bch2_trans_iter_exit(trans, &iter); + return 0; } enum bch_read_flags { diff --git a/fs/bcachefs/io_write.c b/fs/bcachefs/io_write.c index 96720adcfee0..3e71860f66b9 100644 --- a/fs/bcachefs/io_write.c +++ b/fs/bcachefs/io_write.c @@ -164,7 +164,7 @@ int bch2_sum_sector_overwrites(struct btree_trans *trans, bch2_trans_copy_iter(&iter, extent_iter); - for_each_btree_key_upto_continue_norestart(iter, + for_each_btree_key_max_continue_norestart(iter, new->k.p, BTREE_ITER_slots, old, ret) { s64 sectors = min(new->k.p.offset, old.k->p.offset) - max(bkey_start_offset(&new->k), @@ -216,6 +216,7 @@ static inline int bch2_extent_update_i_size_sectors(struct btree_trans *trans, SPOS(0, extent_iter->pos.inode, extent_iter->snapshot), + BTREE_ITER_intent| BTREE_ITER_cached); int ret = bkey_err(k); if (unlikely(ret)) @@ -369,7 +370,7 @@ static int bch2_write_index_default(struct bch_write_op *op) bkey_start_pos(&sk.k->k), BTREE_ITER_slots|BTREE_ITER_intent); - ret = bch2_bkey_set_needs_rebalance(c, sk.k, &op->opts) ?: + ret = bch2_bkey_set_needs_rebalance(c, &op->opts, sk.k) ?: bch2_extent_update(trans, inum, &iter, sk.k, &op->res, op->new_i_size, &op->i_sectors_delta, @@ -395,6 +396,21 @@ static int bch2_write_index_default(struct bch_write_op *op) /* Writes */ +static void __bch2_write_op_error(struct printbuf *out, struct bch_write_op *op, + u64 offset) +{ + bch2_inum_offset_err_msg(op->c, out, + (subvol_inum) { op->subvol, op->pos.inode, }, + offset << 9); + prt_printf(out, "write error%s: ", + op->flags & BCH_WRITE_MOVE ? "(internal move)" : ""); +} + +static void bch2_write_op_error(struct printbuf *out, struct bch_write_op *op) +{ + __bch2_write_op_error(out, op, op->pos.offset); +} + void bch2_submit_wbio_replicas(struct bch_write_bio *wbio, struct bch_fs *c, enum bch_data_type type, const struct bkey_i *k, @@ -531,14 +547,14 @@ static void __bch2_write_index(struct bch_write_op *op) op->written += sectors_start - keylist_sectors(keys); - if (ret && !bch2_err_matches(ret, EROFS)) { + if (unlikely(ret && !bch2_err_matches(ret, EROFS))) { struct bkey_i *insert = bch2_keylist_front(&op->insert_keys); - bch_err_inum_offset_ratelimited(c, - insert->k.p.inode, insert->k.p.offset << 9, - "%s write error while doing btree update: %s", - op->flags & BCH_WRITE_MOVE ? "move" : "user", - bch2_err_str(ret)); + struct printbuf buf = PRINTBUF; + __bch2_write_op_error(&buf, op, bkey_start_offset(&insert->k)); + prt_printf(&buf, "btree update error: %s", bch2_err_str(ret)); + bch_err_ratelimited(c, "%s", buf.buf); + printbuf_exit(&buf); } if (ret) @@ -621,9 +637,7 @@ void bch2_write_point_do_index_updates(struct work_struct *work) while (1) { spin_lock_irq(&wp->writes_lock); - op = list_first_entry_or_null(&wp->writes, struct bch_write_op, wp_list); - if (op) - list_del(&op->wp_list); + op = list_pop_entry(&wp->writes, struct bch_write_op, wp_list); wp_update_state(wp, op != NULL); spin_unlock_irq(&wp->writes_lock); @@ -1080,11 +1094,14 @@ do_write: *_dst = dst; return more; csum_err: - bch_err_inum_offset_ratelimited(c, - op->pos.inode, - op->pos.offset << 9, - "%s write error: error verifying existing checksum while rewriting existing data (memory corruption?)", - op->flags & BCH_WRITE_MOVE ? "move" : "user"); + { + struct printbuf buf = PRINTBUF; + bch2_write_op_error(&buf, op); + prt_printf(&buf, "error verifying existing checksum while rewriting existing data (memory corruption?)"); + bch_err_ratelimited(c, "%s", buf.buf); + printbuf_exit(&buf); + } + ret = -EIO; err: if (to_wbio(dst)->bounce) @@ -1165,7 +1182,7 @@ static void bch2_nocow_write_convert_unwritten(struct bch_write_op *op) struct btree_trans *trans = bch2_trans_get(c); for_each_keylist_key(&op->insert_keys, orig) { - int ret = for_each_btree_key_upto_commit(trans, iter, BTREE_ID_extents, + int ret = for_each_btree_key_max_commit(trans, iter, BTREE_ID_extents, bkey_start_pos(&orig->k), orig->k.p, BTREE_ITER_intent, k, NULL, NULL, BCH_TRANS_COMMIT_no_enospc, ({ @@ -1175,11 +1192,11 @@ static void bch2_nocow_write_convert_unwritten(struct bch_write_op *op) if (ret && !bch2_err_matches(ret, EROFS)) { struct bkey_i *insert = bch2_keylist_front(&op->insert_keys); - bch_err_inum_offset_ratelimited(c, - insert->k.p.inode, insert->k.p.offset << 9, - "%s write error while doing btree update: %s", - op->flags & BCH_WRITE_MOVE ? "move" : "user", - bch2_err_str(ret)); + struct printbuf buf = PRINTBUF; + __bch2_write_op_error(&buf, op, bkey_start_offset(&insert->k)); + prt_printf(&buf, "btree update error: %s", bch2_err_str(ret)); + bch_err_ratelimited(c, "%s", buf.buf); + printbuf_exit(&buf); } if (ret) { @@ -1339,17 +1356,19 @@ err: if (bch2_err_matches(ret, BCH_ERR_transaction_restart)) goto retry; + bch2_trans_put(trans); + darray_exit(&buckets); + if (ret) { - bch_err_inum_offset_ratelimited(c, - op->pos.inode, op->pos.offset << 9, - "%s: btree lookup error %s", __func__, bch2_err_str(ret)); + struct printbuf buf = PRINTBUF; + bch2_write_op_error(&buf, op); + prt_printf(&buf, "%s(): btree lookup error: %s", __func__, bch2_err_str(ret)); + bch_err_ratelimited(c, "%s", buf.buf); + printbuf_exit(&buf); op->error = ret; op->flags |= BCH_WRITE_SUBMITTED; } - bch2_trans_put(trans); - darray_exit(&buckets); - /* fallback to cow write path? */ if (!(op->flags & BCH_WRITE_SUBMITTED)) { closure_sync(&op->cl); @@ -1462,14 +1481,14 @@ err: if (ret <= 0) { op->flags |= BCH_WRITE_SUBMITTED; - if (ret < 0) { - if (!(op->flags & BCH_WRITE_ALLOC_NOWAIT)) - bch_err_inum_offset_ratelimited(c, - op->pos.inode, - op->pos.offset << 9, - "%s(): %s error: %s", __func__, - op->flags & BCH_WRITE_MOVE ? "move" : "user", - bch2_err_str(ret)); + if (unlikely(ret < 0)) { + if (!(op->flags & BCH_WRITE_ALLOC_NOWAIT)) { + struct printbuf buf = PRINTBUF; + bch2_write_op_error(&buf, op); + prt_printf(&buf, "%s(): %s", __func__, bch2_err_str(ret)); + bch_err_ratelimited(c, "%s", buf.buf); + printbuf_exit(&buf); + } op->error = ret; break; } @@ -1595,12 +1614,11 @@ CLOSURE_CALLBACK(bch2_write) bch2_keylist_init(&op->insert_keys, op->inline_keys); wbio_init(bio)->put_bio = false; - if (bio->bi_iter.bi_size & (c->opts.block_size - 1)) { - bch_err_inum_offset_ratelimited(c, - op->pos.inode, - op->pos.offset << 9, - "%s write error: misaligned write", - op->flags & BCH_WRITE_MOVE ? "move" : "user"); + if (unlikely(bio->bi_iter.bi_size & (c->opts.block_size - 1))) { + struct printbuf buf = PRINTBUF; + bch2_write_op_error(&buf, op); + prt_printf(&buf, "misaligned write"); + printbuf_exit(&buf); op->error = -EIO; goto err; } diff --git a/fs/bcachefs/journal.c b/fs/bcachefs/journal.c index 2dc0d60c1745..2cd20114b74b 100644 --- a/fs/bcachefs/journal.c +++ b/fs/bcachefs/journal.c @@ -217,6 +217,12 @@ void bch2_journal_buf_put_final(struct journal *j, u64 seq) if (__bch2_journal_pin_put(j, seq)) bch2_journal_reclaim_fast(j); bch2_journal_do_writes(j); + + /* + * for __bch2_next_write_buffer_flush_journal_buf(), when quiescing an + * open journal entry + */ + wake_up(&j->wait); } /* @@ -251,6 +257,9 @@ static void __journal_entry_close(struct journal *j, unsigned closed_val, bool t if (!__journal_entry_is_open(old)) return; + if (old.cur_entry_offset == JOURNAL_ENTRY_BLOCKED_VAL) + old.cur_entry_offset = j->cur_entry_offset_if_blocked; + /* Close out old buffer: */ buf->data->u64s = cpu_to_le32(old.cur_entry_offset); @@ -373,6 +382,10 @@ static int journal_entry_open(struct journal *j) if (nr_unwritten_journal_entries(j) == ARRAY_SIZE(j->buf)) return JOURNAL_ERR_max_in_flight; + if (bch2_fs_fatal_err_on(journal_cur_seq(j) >= JOURNAL_SEQ_MAX, + c, "cannot start: journal seq overflow")) + return JOURNAL_ERR_insufficient_devices; /* -EROFS */ + BUG_ON(!j->cur_entry_sectors); buf->expires = @@ -664,7 +677,7 @@ out: * @seq: seq to flush * @parent: closure object to wait with * Returns: 1 if @seq has already been flushed, 0 if @seq is being flushed, - * -EIO if @seq will never be flushed + * -BCH_ERR_journal_flush_err if @seq will never be flushed * * Like bch2_journal_wait_on_seq, except that it triggers a write immediately if * necessary @@ -687,7 +700,7 @@ int bch2_journal_flush_seq_async(struct journal *j, u64 seq, /* Recheck under lock: */ if (j->err_seq && seq >= j->err_seq) { - ret = -EIO; + ret = -BCH_ERR_journal_flush_err; goto out; } @@ -794,10 +807,11 @@ int bch2_journal_flush(struct journal *j) } /* - * bch2_journal_noflush_seq - tell the journal not to issue any flushes before + * bch2_journal_noflush_seq - ask the journal not to issue any flushes in the + * range [start, end) * @seq */ -bool bch2_journal_noflush_seq(struct journal *j, u64 seq) +bool bch2_journal_noflush_seq(struct journal *j, u64 start, u64 end) { struct bch_fs *c = container_of(j, struct bch_fs, journal); u64 unwritten_seq; @@ -806,15 +820,15 @@ bool bch2_journal_noflush_seq(struct journal *j, u64 seq) if (!(c->sb.features & (1ULL << BCH_FEATURE_journal_no_flush))) return false; - if (seq <= c->journal.flushed_seq_ondisk) + if (c->journal.flushed_seq_ondisk >= start) return false; spin_lock(&j->lock); - if (seq <= c->journal.flushed_seq_ondisk) + if (c->journal.flushed_seq_ondisk >= start) goto out; for (unwritten_seq = journal_last_unwritten_seq(j); - unwritten_seq < seq; + unwritten_seq < end; unwritten_seq++) { struct journal_buf *buf = journal_seq_to_buf(j, unwritten_seq); @@ -831,19 +845,14 @@ out: return ret; } -int bch2_journal_meta(struct journal *j) +static int __bch2_journal_meta(struct journal *j) { - struct journal_buf *buf; - struct journal_res res; - int ret; - - memset(&res, 0, sizeof(res)); - - ret = bch2_journal_res_get(j, &res, jset_u64s(0), 0); + struct journal_res res = {}; + int ret = bch2_journal_res_get(j, &res, jset_u64s(0), 0); if (ret) return ret; - buf = j->buf + (res.seq & JOURNAL_BUF_MASK); + struct journal_buf *buf = j->buf + (res.seq & JOURNAL_BUF_MASK); buf->must_flush = true; if (!buf->flush_time) { @@ -856,27 +865,70 @@ int bch2_journal_meta(struct journal *j) return bch2_journal_flush_seq(j, res.seq, TASK_UNINTERRUPTIBLE); } +int bch2_journal_meta(struct journal *j) +{ + struct bch_fs *c = container_of(j, struct bch_fs, journal); + + if (!bch2_write_ref_tryget(c, BCH_WRITE_REF_journal)) + return -EROFS; + + int ret = __bch2_journal_meta(j); + bch2_write_ref_put(c, BCH_WRITE_REF_journal); + return ret; +} + /* block/unlock the journal: */ void bch2_journal_unblock(struct journal *j) { spin_lock(&j->lock); - j->blocked--; + if (!--j->blocked && + j->cur_entry_offset_if_blocked < JOURNAL_ENTRY_CLOSED_VAL && + j->reservations.cur_entry_offset == JOURNAL_ENTRY_BLOCKED_VAL) { + union journal_res_state old, new; + + old.v = atomic64_read(&j->reservations.counter); + do { + new.v = old.v; + new.cur_entry_offset = j->cur_entry_offset_if_blocked; + } while (!atomic64_try_cmpxchg(&j->reservations.counter, &old.v, new.v)); + } spin_unlock(&j->lock); journal_wake(j); } +static void __bch2_journal_block(struct journal *j) +{ + if (!j->blocked++) { + union journal_res_state old, new; + + old.v = atomic64_read(&j->reservations.counter); + do { + j->cur_entry_offset_if_blocked = old.cur_entry_offset; + + if (j->cur_entry_offset_if_blocked >= JOURNAL_ENTRY_CLOSED_VAL) + break; + + new.v = old.v; + new.cur_entry_offset = JOURNAL_ENTRY_BLOCKED_VAL; + } while (!atomic64_try_cmpxchg(&j->reservations.counter, &old.v, new.v)); + + journal_cur_buf(j)->data->u64s = cpu_to_le32(old.cur_entry_offset); + } +} + void bch2_journal_block(struct journal *j) { spin_lock(&j->lock); - j->blocked++; + __bch2_journal_block(j); spin_unlock(&j->lock); journal_quiesce(j); } -static struct journal_buf *__bch2_next_write_buffer_flush_journal_buf(struct journal *j, u64 max_seq) +static struct journal_buf *__bch2_next_write_buffer_flush_journal_buf(struct journal *j, + u64 max_seq, bool *blocked) { struct journal_buf *ret = NULL; @@ -893,13 +945,17 @@ static struct journal_buf *__bch2_next_write_buffer_flush_journal_buf(struct jou struct journal_buf *buf = j->buf + idx; if (buf->need_flush_to_write_buffer) { - if (seq == journal_cur_seq(j)) - __journal_entry_close(j, JOURNAL_ENTRY_CLOSED_VAL, true); - union journal_res_state s; s.v = atomic64_read_acquire(&j->reservations.counter); - ret = journal_state_count(s, idx) + unsigned open = seq == journal_cur_seq(j) && __journal_entry_is_open(s); + + if (open && !*blocked) { + __bch2_journal_block(j); + *blocked = true; + } + + ret = journal_state_count(s, idx) > open ? ERR_PTR(-EAGAIN) : buf; break; @@ -912,11 +968,17 @@ static struct journal_buf *__bch2_next_write_buffer_flush_journal_buf(struct jou return ret; } -struct journal_buf *bch2_next_write_buffer_flush_journal_buf(struct journal *j, u64 max_seq) +struct journal_buf *bch2_next_write_buffer_flush_journal_buf(struct journal *j, + u64 max_seq, bool *blocked) { struct journal_buf *ret; + *blocked = false; + + wait_event(j->wait, (ret = __bch2_next_write_buffer_flush_journal_buf(j, + max_seq, blocked)) != ERR_PTR(-EAGAIN)); + if (IS_ERR_OR_NULL(ret) && *blocked) + bch2_journal_unblock(j); - wait_event(j->wait, (ret = __bch2_next_write_buffer_flush_journal_buf(j, max_seq)) != ERR_PTR(-EAGAIN)); return ret; } @@ -945,19 +1007,17 @@ static int __bch2_set_nr_journal_buckets(struct bch_dev *ca, unsigned nr, } for (nr_got = 0; nr_got < nr_want; nr_got++) { - if (new_fs) { - bu[nr_got] = bch2_bucket_alloc_new_fs(ca); - if (bu[nr_got] < 0) { - ret = -BCH_ERR_ENOSPC_bucket_alloc; - break; - } - } else { - ob[nr_got] = bch2_bucket_alloc(c, ca, BCH_WATERMARK_normal, - BCH_DATA_journal, cl); - ret = PTR_ERR_OR_ZERO(ob[nr_got]); - if (ret) - break; + enum bch_watermark watermark = new_fs + ? BCH_WATERMARK_btree + : BCH_WATERMARK_normal; + ob[nr_got] = bch2_bucket_alloc(c, ca, watermark, + BCH_DATA_journal, cl); + ret = PTR_ERR_OR_ZERO(ob[nr_got]); + if (ret) + break; + + if (!new_fs) { ret = bch2_trans_run(c, bch2_trans_mark_metadata_bucket(trans, ca, ob[nr_got]->bucket, BCH_DATA_journal, @@ -967,9 +1027,9 @@ static int __bch2_set_nr_journal_buckets(struct bch_dev *ca, unsigned nr, bch_err_msg(c, ret, "marking new journal buckets"); break; } - - bu[nr_got] = ob[nr_got]->bucket; } + + bu[nr_got] = ob[nr_got]->bucket; } if (!nr_got) @@ -1009,8 +1069,7 @@ static int __bch2_set_nr_journal_buckets(struct bch_dev *ca, unsigned nr, if (ret) goto err_unblock; - if (!new_fs) - bch2_write_super(c); + bch2_write_super(c); /* Commit: */ if (c) @@ -1044,9 +1103,8 @@ err_unblock: bu[i], BCH_DATA_free, 0, BTREE_TRIGGER_transactional)); err_free: - if (!new_fs) - for (i = 0; i < nr_got; i++) - bch2_open_bucket_put(c, ob[i]); + for (i = 0; i < nr_got; i++) + bch2_open_bucket_put(c, ob[i]); kfree(new_bucket_seq); kfree(new_buckets); @@ -1193,7 +1251,7 @@ void bch2_fs_journal_stop(struct journal *j) * Always write a new journal entry, to make sure the clock hands are up * to date (and match the superblock) */ - bch2_journal_meta(j); + __bch2_journal_meta(j); journal_quiesce(j); cancel_delayed_work_sync(&j->write_work); @@ -1217,6 +1275,11 @@ int bch2_fs_journal_start(struct journal *j, u64 cur_seq) bool had_entries = false; u64 last_seq = cur_seq, nr, seq; + if (cur_seq >= JOURNAL_SEQ_MAX) { + bch_err(c, "cannot start: journal seq overflow"); + return -EINVAL; + } + genradix_for_each_reverse(&c->journal_entries, iter, _i) { i = *_i; @@ -1474,6 +1537,9 @@ void __bch2_journal_debug_to_text(struct printbuf *out, struct journal *j) case JOURNAL_ENTRY_CLOSED_VAL: prt_printf(out, "closed\n"); break; + case JOURNAL_ENTRY_BLOCKED_VAL: + prt_printf(out, "blocked\n"); + break; default: prt_printf(out, "%u/%u\n", s.cur_entry_offset, j->cur_entry_u64s); break; @@ -1499,6 +1565,9 @@ void __bch2_journal_debug_to_text(struct printbuf *out, struct journal *j) printbuf_indent_sub(out, 2); for_each_member_device_rcu(c, ca, &c->rw_devs[BCH_DATA_journal]) { + if (!ca->mi.durability) + continue; + struct journal_device *ja = &ca->journal; if (!test_bit(ca->dev_idx, c->rw_devs[BCH_DATA_journal].d)) @@ -1508,6 +1577,7 @@ void __bch2_journal_debug_to_text(struct printbuf *out, struct journal *j) continue; prt_printf(out, "dev %u:\n", ca->dev_idx); + prt_printf(out, "durability %u:\n", ca->mi.durability); printbuf_indent_add(out, 2); prt_printf(out, "nr\t%u\n", ja->nr); prt_printf(out, "bucket size\t%u\n", ca->mi.bucket_size); @@ -1519,6 +1589,8 @@ void __bch2_journal_debug_to_text(struct printbuf *out, struct journal *j) printbuf_indent_sub(out, 2); } + prt_printf(out, "replicas want %u need %u\n", c->opts.metadata_replicas, c->opts.metadata_replicas_required); + rcu_read_unlock(); --out->atomic; diff --git a/fs/bcachefs/journal.h b/fs/bcachefs/journal.h index 2762be6f9814..cb0df0663946 100644 --- a/fs/bcachefs/journal.h +++ b/fs/bcachefs/journal.h @@ -285,7 +285,8 @@ static inline void bch2_journal_buf_put(struct journal *j, unsigned idx, u64 seq spin_lock(&j->lock); bch2_journal_buf_put_final(j, seq); spin_unlock(&j->lock); - } + } else if (unlikely(s.cur_entry_offset == JOURNAL_ENTRY_BLOCKED_VAL)) + wake_up(&j->wait); } /* @@ -403,7 +404,7 @@ void bch2_journal_flush_async(struct journal *, struct closure *); int bch2_journal_flush_seq(struct journal *, u64, unsigned); int bch2_journal_flush(struct journal *); -bool bch2_journal_noflush_seq(struct journal *, u64); +bool bch2_journal_noflush_seq(struct journal *, u64, u64); int bch2_journal_meta(struct journal *); void bch2_journal_halt(struct journal *); @@ -411,7 +412,7 @@ void bch2_journal_halt(struct journal *); static inline int bch2_journal_error(struct journal *j) { return j->reservations.cur_entry_offset == JOURNAL_ENTRY_ERROR_VAL - ? -EIO : 0; + ? -BCH_ERR_journal_shutdown : 0; } struct bch_dev; @@ -424,7 +425,7 @@ static inline void bch2_journal_set_replay_done(struct journal *j) void bch2_journal_unblock(struct journal *); void bch2_journal_block(struct journal *); -struct journal_buf *bch2_next_write_buffer_flush_journal_buf(struct journal *j, u64 max_seq); +struct journal_buf *bch2_next_write_buffer_flush_journal_buf(struct journal *, u64, bool *); void __bch2_journal_debug_to_text(struct printbuf *, struct journal *); void bch2_journal_debug_to_text(struct printbuf *, struct journal *); diff --git a/fs/bcachefs/journal_io.c b/fs/bcachefs/journal_io.c index fb35dd336331..7f2efe85a805 100644 --- a/fs/bcachefs/journal_io.c +++ b/fs/bcachefs/journal_io.c @@ -17,6 +17,8 @@ #include "sb-clean.h" #include "trace.h" +#include <linux/string_choices.h> + void bch2_journal_pos_from_member_info_set(struct bch_fs *c) { lockdep_assert_held(&c->sb_lock); @@ -299,7 +301,7 @@ static void journal_entry_err_msg(struct printbuf *out, journal_entry_err_msg(&_buf, version, jset, entry); \ prt_printf(&_buf, msg, ##__VA_ARGS__); \ \ - switch (flags & BCH_VALIDATE_write) { \ + switch (from.flags & BCH_VALIDATE_write) { \ case READ: \ mustfix_fsck_err(c, _err, "%s", _buf.buf); \ break; \ @@ -325,11 +327,11 @@ static void journal_entry_err_msg(struct printbuf *out, static int journal_validate_key(struct bch_fs *c, struct jset *jset, struct jset_entry *entry, - unsigned level, enum btree_id btree_id, struct bkey_i *k, - unsigned version, int big_endian, - enum bch_validate_flags flags) + struct bkey_validate_context from, + unsigned version, int big_endian) { + enum bch_validate_flags flags = from.flags; int write = flags & BCH_VALIDATE_write; void *next = vstruct_next(entry); int ret = 0; @@ -364,11 +366,10 @@ static int journal_validate_key(struct bch_fs *c, } if (!write) - bch2_bkey_compat(level, btree_id, version, big_endian, + bch2_bkey_compat(from.level, from.btree, version, big_endian, write, NULL, bkey_to_packed(k)); - ret = bch2_bkey_validate(c, bkey_i_to_s_c(k), - __btree_node_type(level, btree_id), write); + ret = bch2_bkey_validate(c, bkey_i_to_s_c(k), from); if (ret == -BCH_ERR_fsck_delete_bkey) { le16_add_cpu(&entry->u64s, -((u16) k->k.u64s)); memmove(k, bkey_next(k), next - (void *) bkey_next(k)); @@ -379,7 +380,7 @@ static int journal_validate_key(struct bch_fs *c, goto fsck_err; if (write) - bch2_bkey_compat(level, btree_id, version, big_endian, + bch2_bkey_compat(from.level, from.btree, version, big_endian, write, NULL, bkey_to_packed(k)); fsck_err: return ret; @@ -389,16 +390,15 @@ static int journal_entry_btree_keys_validate(struct bch_fs *c, struct jset *jset, struct jset_entry *entry, unsigned version, int big_endian, - enum bch_validate_flags flags) + struct bkey_validate_context from) { struct bkey_i *k = entry->start; + from.level = entry->level; + from.btree = entry->btree_id; + while (k != vstruct_last(entry)) { - int ret = journal_validate_key(c, jset, entry, - entry->level, - entry->btree_id, - k, version, big_endian, - flags|BCH_VALIDATE_journal); + int ret = journal_validate_key(c, jset, entry, k, from, version, big_endian); if (ret == FSCK_DELETED_KEY) continue; else if (ret) @@ -421,7 +421,8 @@ static void journal_entry_btree_keys_to_text(struct printbuf *out, struct bch_fs bch2_prt_jset_entry_type(out, entry->type); prt_str(out, ": "); } - prt_printf(out, "btree=%s l=%u ", bch2_btree_id_str(entry->btree_id), entry->level); + bch2_btree_id_level_to_text(out, entry->btree_id, entry->level); + prt_char(out, ' '); bch2_bkey_val_to_text(out, c, bkey_i_to_s_c(k)); first = false; } @@ -431,11 +432,15 @@ static int journal_entry_btree_root_validate(struct bch_fs *c, struct jset *jset, struct jset_entry *entry, unsigned version, int big_endian, - enum bch_validate_flags flags) + struct bkey_validate_context from) { struct bkey_i *k = entry->start; int ret = 0; + from.root = true; + from.level = entry->level + 1; + from.btree = entry->btree_id; + if (journal_entry_err_on(!entry->u64s || le16_to_cpu(entry->u64s) != k->k.u64s, c, version, jset, entry, @@ -452,8 +457,7 @@ static int journal_entry_btree_root_validate(struct bch_fs *c, return 0; } - ret = journal_validate_key(c, jset, entry, 1, entry->btree_id, k, - version, big_endian, flags); + ret = journal_validate_key(c, jset, entry, k, from, version, big_endian); if (ret == FSCK_DELETED_KEY) ret = 0; fsck_err: @@ -470,7 +474,7 @@ static int journal_entry_prio_ptrs_validate(struct bch_fs *c, struct jset *jset, struct jset_entry *entry, unsigned version, int big_endian, - enum bch_validate_flags flags) + struct bkey_validate_context from) { /* obsolete, don't care: */ return 0; @@ -485,7 +489,7 @@ static int journal_entry_blacklist_validate(struct bch_fs *c, struct jset *jset, struct jset_entry *entry, unsigned version, int big_endian, - enum bch_validate_flags flags) + struct bkey_validate_context from) { int ret = 0; @@ -512,7 +516,7 @@ static int journal_entry_blacklist_v2_validate(struct bch_fs *c, struct jset *jset, struct jset_entry *entry, unsigned version, int big_endian, - enum bch_validate_flags flags) + struct bkey_validate_context from) { struct jset_entry_blacklist_v2 *bl_entry; int ret = 0; @@ -554,7 +558,7 @@ static int journal_entry_usage_validate(struct bch_fs *c, struct jset *jset, struct jset_entry *entry, unsigned version, int big_endian, - enum bch_validate_flags flags) + struct bkey_validate_context from) { struct jset_entry_usage *u = container_of(entry, struct jset_entry_usage, entry); @@ -588,7 +592,7 @@ static int journal_entry_data_usage_validate(struct bch_fs *c, struct jset *jset, struct jset_entry *entry, unsigned version, int big_endian, - enum bch_validate_flags flags) + struct bkey_validate_context from) { struct jset_entry_data_usage *u = container_of(entry, struct jset_entry_data_usage, entry); @@ -632,7 +636,7 @@ static int journal_entry_clock_validate(struct bch_fs *c, struct jset *jset, struct jset_entry *entry, unsigned version, int big_endian, - enum bch_validate_flags flags) + struct bkey_validate_context from) { struct jset_entry_clock *clock = container_of(entry, struct jset_entry_clock, entry); @@ -665,14 +669,14 @@ static void journal_entry_clock_to_text(struct printbuf *out, struct bch_fs *c, struct jset_entry_clock *clock = container_of(entry, struct jset_entry_clock, entry); - prt_printf(out, "%s=%llu", clock->rw ? "write" : "read", le64_to_cpu(clock->time)); + prt_printf(out, "%s=%llu", str_write_read(clock->rw), le64_to_cpu(clock->time)); } static int journal_entry_dev_usage_validate(struct bch_fs *c, struct jset *jset, struct jset_entry *entry, unsigned version, int big_endian, - enum bch_validate_flags flags) + struct bkey_validate_context from) { struct jset_entry_dev_usage *u = container_of(entry, struct jset_entry_dev_usage, entry); @@ -729,7 +733,7 @@ static int journal_entry_log_validate(struct bch_fs *c, struct jset *jset, struct jset_entry *entry, unsigned version, int big_endian, - enum bch_validate_flags flags) + struct bkey_validate_context from) { return 0; } @@ -738,19 +742,19 @@ static void journal_entry_log_to_text(struct printbuf *out, struct bch_fs *c, struct jset_entry *entry) { struct jset_entry_log *l = container_of(entry, struct jset_entry_log, entry); - unsigned bytes = vstruct_bytes(entry) - offsetof(struct jset_entry_log, d); - prt_printf(out, "%.*s", bytes, l->d); + prt_printf(out, "%.*s", jset_entry_log_msg_bytes(l), l->d); } static int journal_entry_overwrite_validate(struct bch_fs *c, struct jset *jset, struct jset_entry *entry, unsigned version, int big_endian, - enum bch_validate_flags flags) + struct bkey_validate_context from) { + from.flags = 0; return journal_entry_btree_keys_validate(c, jset, entry, - version, big_endian, READ); + version, big_endian, from); } static void journal_entry_overwrite_to_text(struct printbuf *out, struct bch_fs *c, @@ -763,10 +767,10 @@ static int journal_entry_write_buffer_keys_validate(struct bch_fs *c, struct jset *jset, struct jset_entry *entry, unsigned version, int big_endian, - enum bch_validate_flags flags) + struct bkey_validate_context from) { return journal_entry_btree_keys_validate(c, jset, entry, - version, big_endian, READ); + version, big_endian, from); } static void journal_entry_write_buffer_keys_to_text(struct printbuf *out, struct bch_fs *c, @@ -779,7 +783,7 @@ static int journal_entry_datetime_validate(struct bch_fs *c, struct jset *jset, struct jset_entry *entry, unsigned version, int big_endian, - enum bch_validate_flags flags) + struct bkey_validate_context from) { unsigned bytes = vstruct_bytes(entry); unsigned expected = 16; @@ -809,7 +813,7 @@ static void journal_entry_datetime_to_text(struct printbuf *out, struct bch_fs * struct jset_entry_ops { int (*validate)(struct bch_fs *, struct jset *, struct jset_entry *, unsigned, int, - enum bch_validate_flags); + struct bkey_validate_context); void (*to_text)(struct printbuf *, struct bch_fs *, struct jset_entry *); }; @@ -827,11 +831,11 @@ int bch2_journal_entry_validate(struct bch_fs *c, struct jset *jset, struct jset_entry *entry, unsigned version, int big_endian, - enum bch_validate_flags flags) + struct bkey_validate_context from) { return entry->type < BCH_JSET_ENTRY_NR ? bch2_jset_entry_ops[entry->type].validate(c, jset, entry, - version, big_endian, flags) + version, big_endian, from) : 0; } @@ -849,10 +853,18 @@ void bch2_journal_entry_to_text(struct printbuf *out, struct bch_fs *c, static int jset_validate_entries(struct bch_fs *c, struct jset *jset, enum bch_validate_flags flags) { + struct bkey_validate_context from = { + .flags = flags, + .from = BKEY_VALIDATE_journal, + .journal_seq = le64_to_cpu(jset->seq), + }; + unsigned version = le32_to_cpu(jset->version); int ret = 0; vstruct_for_each(jset, entry) { + from.journal_offset = (u64 *) entry - jset->_data; + if (journal_entry_err_on(vstruct_next(entry) > vstruct_last(jset), c, version, jset, entry, journal_entry_past_jset_end, @@ -861,8 +873,8 @@ static int jset_validate_entries(struct bch_fs *c, struct jset *jset, break; } - ret = bch2_journal_entry_validate(c, jset, entry, - version, JSET_BIG_ENDIAN(jset), flags); + ret = bch2_journal_entry_validate(c, jset, entry, version, + JSET_BIG_ENDIAN(jset), from); if (ret) break; } @@ -875,13 +887,17 @@ static int jset_validate(struct bch_fs *c, struct jset *jset, u64 sector, enum bch_validate_flags flags) { - unsigned version; + struct bkey_validate_context from = { + .flags = flags, + .from = BKEY_VALIDATE_journal, + .journal_seq = le64_to_cpu(jset->seq), + }; int ret = 0; if (le64_to_cpu(jset->magic) != jset_magic(c)) return JOURNAL_ENTRY_NONE; - version = le32_to_cpu(jset->version); + unsigned version = le32_to_cpu(jset->version); if (journal_entry_err_on(!bch2_version_compatible(version), c, version, jset, NULL, jset_unsupported_version, @@ -926,15 +942,16 @@ static int jset_validate_early(struct bch_fs *c, unsigned bucket_sectors_left, unsigned sectors_read) { - size_t bytes = vstruct_bytes(jset); - unsigned version; - enum bch_validate_flags flags = BCH_VALIDATE_journal; + struct bkey_validate_context from = { + .from = BKEY_VALIDATE_journal, + .journal_seq = le64_to_cpu(jset->seq), + }; int ret = 0; if (le64_to_cpu(jset->magic) != jset_magic(c)) return JOURNAL_ENTRY_NONE; - version = le32_to_cpu(jset->version); + unsigned version = le32_to_cpu(jset->version); if (journal_entry_err_on(!bch2_version_compatible(version), c, version, jset, NULL, jset_unsupported_version, @@ -947,6 +964,7 @@ static int jset_validate_early(struct bch_fs *c, return -EINVAL; } + size_t bytes = vstruct_bytes(jset); if (bytes > (sectors_read << 9) && sectors_read < bucket_sectors_left) return JOURNAL_ENTRY_REREAD; @@ -1231,8 +1249,6 @@ int bch2_journal_read(struct bch_fs *c, * those entries will be blacklisted: */ genradix_for_each_reverse(&c->journal_entries, radix_iter, _i) { - enum bch_validate_flags flags = BCH_VALIDATE_journal; - i = *_i; if (journal_replay_ignore(i)) @@ -1252,6 +1268,10 @@ int bch2_journal_read(struct bch_fs *c, continue; } + struct bkey_validate_context from = { + .from = BKEY_VALIDATE_journal, + .journal_seq = le64_to_cpu(i->j.seq), + }; if (journal_entry_err_on(le64_to_cpu(i->j.last_seq) > le64_to_cpu(i->j.seq), c, le32_to_cpu(i->j.version), &i->j, NULL, jset_last_seq_newer_than_seq, @@ -1411,27 +1431,50 @@ fsck_err: /* journal write: */ +static void journal_advance_devs_to_next_bucket(struct journal *j, + struct dev_alloc_list *devs, + unsigned sectors, u64 seq) +{ + struct bch_fs *c = container_of(j, struct bch_fs, journal); + + darray_for_each(*devs, i) { + struct bch_dev *ca = rcu_dereference(c->devs[*i]); + if (!ca) + continue; + + struct journal_device *ja = &ca->journal; + + if (sectors > ja->sectors_free && + sectors <= ca->mi.bucket_size && + bch2_journal_dev_buckets_available(j, ja, + journal_space_discarded)) { + ja->cur_idx = (ja->cur_idx + 1) % ja->nr; + ja->sectors_free = ca->mi.bucket_size; + + /* + * ja->bucket_seq[ja->cur_idx] must always have + * something sensible: + */ + ja->bucket_seq[ja->cur_idx] = le64_to_cpu(seq); + } + } +} + static void __journal_write_alloc(struct journal *j, struct journal_buf *w, - struct dev_alloc_list *devs_sorted, + struct dev_alloc_list *devs, unsigned sectors, unsigned *replicas, unsigned replicas_want) { struct bch_fs *c = container_of(j, struct bch_fs, journal); - struct journal_device *ja; - struct bch_dev *ca; - unsigned i; - if (*replicas >= replicas_want) - return; - - for (i = 0; i < devs_sorted->nr; i++) { - ca = rcu_dereference(c->devs[devs_sorted->devs[i]]); + darray_for_each(*devs, i) { + struct bch_dev *ca = rcu_dereference(c->devs[*i]); if (!ca) continue; - ja = &ca->journal; + struct journal_device *ja = &ca->journal; /* * Check that we can use this device, and aren't already using @@ -1477,65 +1520,53 @@ static int journal_write_alloc(struct journal *j, struct journal_buf *w) { struct bch_fs *c = container_of(j, struct bch_fs, journal); struct bch_devs_mask devs; - struct journal_device *ja; - struct bch_dev *ca; struct dev_alloc_list devs_sorted; unsigned sectors = vstruct_sectors(w->data, c->block_bits); unsigned target = c->opts.metadata_target ?: c->opts.foreground_target; - unsigned i, replicas = 0, replicas_want = + unsigned replicas = 0, replicas_want = READ_ONCE(c->opts.metadata_replicas); unsigned replicas_need = min_t(unsigned, replicas_want, READ_ONCE(c->opts.metadata_replicas_required)); + bool advance_done = false; rcu_read_lock(); -retry: - devs = target_rw_devs(c, BCH_DATA_journal, target); - devs_sorted = bch2_dev_alloc_list(c, &j->wp.stripe, &devs); + /* We might run more than once if we have to stop and do discards: */ + struct bkey_ptrs_c ptrs = bch2_bkey_ptrs_c(bkey_i_to_s_c(&w->key)); + bkey_for_each_ptr(ptrs, p) { + struct bch_dev *ca = bch2_dev_rcu_noerror(c, p->dev); + if (ca) + replicas += ca->mi.durability; + } - __journal_write_alloc(j, w, &devs_sorted, - sectors, &replicas, replicas_want); +retry_target: + devs = target_rw_devs(c, BCH_DATA_journal, target); + devs_sorted = bch2_dev_alloc_list(c, &j->wp.stripe, &devs); +retry_alloc: + __journal_write_alloc(j, w, &devs_sorted, sectors, &replicas, replicas_want); - if (replicas >= replicas_want) + if (likely(replicas >= replicas_want)) goto done; - for (i = 0; i < devs_sorted.nr; i++) { - ca = rcu_dereference(c->devs[devs_sorted.devs[i]]); - if (!ca) - continue; - - ja = &ca->journal; - - if (sectors > ja->sectors_free && - sectors <= ca->mi.bucket_size && - bch2_journal_dev_buckets_available(j, ja, - journal_space_discarded)) { - ja->cur_idx = (ja->cur_idx + 1) % ja->nr; - ja->sectors_free = ca->mi.bucket_size; - - /* - * ja->bucket_seq[ja->cur_idx] must always have - * something sensible: - */ - ja->bucket_seq[ja->cur_idx] = le64_to_cpu(w->data->seq); - } + if (!advance_done) { + journal_advance_devs_to_next_bucket(j, &devs_sorted, sectors, w->data->seq); + advance_done = true; + goto retry_alloc; } - __journal_write_alloc(j, w, &devs_sorted, - sectors, &replicas, replicas_want); - if (replicas < replicas_want && target) { /* Retry from all devices: */ target = 0; - goto retry; + advance_done = false; + goto retry_target; } done: rcu_read_unlock(); BUG_ON(bkey_val_u64s(&w->key.k) > BCH_REPLICAS_MAX); - return replicas >= replicas_need ? 0 : -EROFS; + return replicas >= replicas_need ? 0 : -BCH_ERR_insufficient_journal_devices; } static void journal_buf_realloc(struct journal *j, struct journal_buf *buf) @@ -2023,19 +2054,21 @@ CLOSURE_CALLBACK(bch2_journal_write) bch2_journal_do_discards(j); } - if (ret) { + if (ret && !bch2_journal_error(j)) { struct printbuf buf = PRINTBUF; buf.atomic++; - prt_printf(&buf, bch2_fmt(c, "Unable to allocate journal write at seq %llu: %s"), + prt_printf(&buf, bch2_fmt(c, "Unable to allocate journal write at seq %llu for %zu sectors: %s"), le64_to_cpu(w->data->seq), + vstruct_sectors(w->data, c->block_bits), bch2_err_str(ret)); __bch2_journal_debug_to_text(&buf, j); spin_unlock(&j->lock); bch2_print_string_as_lines(KERN_ERR, buf.buf); printbuf_exit(&buf); - goto err; } + if (ret) + goto err; /* * write is allocated, no longer need to account for it in diff --git a/fs/bcachefs/journal_io.h b/fs/bcachefs/journal_io.h index 2ca9cde30ea8..12b39fcb4424 100644 --- a/fs/bcachefs/journal_io.h +++ b/fs/bcachefs/journal_io.h @@ -63,7 +63,7 @@ static inline struct jset_entry *__jset_entry_type_next(struct jset *jset, int bch2_journal_entry_validate(struct bch_fs *, struct jset *, struct jset_entry *, unsigned, int, - enum bch_validate_flags); + struct bkey_validate_context); void bch2_journal_entry_to_text(struct printbuf *, struct bch_fs *, struct jset_entry *); diff --git a/fs/bcachefs/journal_reclaim.c b/fs/bcachefs/journal_reclaim.c index ace291f175dd..3c8242606da7 100644 --- a/fs/bcachefs/journal_reclaim.c +++ b/fs/bcachefs/journal_reclaim.c @@ -38,6 +38,9 @@ unsigned bch2_journal_dev_buckets_available(struct journal *j, struct journal_device *ja, enum journal_space_from from) { + if (!ja->nr) + return 0; + unsigned available = (journal_space_from(ja, from) - ja->cur_idx - 1 + ja->nr) % ja->nr; @@ -137,14 +140,18 @@ static struct journal_space __journal_space_available(struct journal *j, unsigne struct bch_fs *c = container_of(j, struct bch_fs, journal); unsigned pos, nr_devs = 0; struct journal_space space, dev_space[BCH_SB_MEMBERS_MAX]; + unsigned min_bucket_size = U32_MAX; BUG_ON(nr_devs_want > ARRAY_SIZE(dev_space)); rcu_read_lock(); for_each_member_device_rcu(c, ca, &c->rw_devs[BCH_DATA_journal]) { - if (!ca->journal.nr) + if (!ca->journal.nr || + !ca->mi.durability) continue; + min_bucket_size = min(min_bucket_size, ca->mi.bucket_size); + space = journal_dev_space_available(j, ca, from); if (!space.next_entry) continue; @@ -164,7 +171,9 @@ static struct journal_space __journal_space_available(struct journal *j, unsigne * We sorted largest to smallest, and we want the smallest out of the * @nr_devs_want largest devices: */ - return dev_space[nr_devs_want - 1]; + space = dev_space[nr_devs_want - 1]; + space.next_entry = min(space.next_entry, min_bucket_size); + return space; } void bch2_journal_space_available(struct journal *j) @@ -758,10 +767,12 @@ static int bch2_journal_reclaim_thread(void *arg) journal_empty = fifo_empty(&j->pin); spin_unlock(&j->lock); + long timeout = j->next_reclaim - jiffies; + if (journal_empty) schedule(); - else if (time_after(j->next_reclaim, jiffies)) - schedule_timeout(j->next_reclaim - jiffies); + else if (timeout > 0) + schedule_timeout(timeout); else break; } diff --git a/fs/bcachefs/journal_types.h b/fs/bcachefs/journal_types.h index 19183fcf7ad7..e9bd716fbb71 100644 --- a/fs/bcachefs/journal_types.h +++ b/fs/bcachefs/journal_types.h @@ -9,6 +9,9 @@ #include "super_types.h" #include "fifo.h" +/* btree write buffer steals 8 bits for its own purposes: */ +#define JOURNAL_SEQ_MAX ((1ULL << 56) - 1) + #define JOURNAL_BUF_BITS 2 #define JOURNAL_BUF_NR (1U << JOURNAL_BUF_BITS) #define JOURNAL_BUF_MASK (JOURNAL_BUF_NR - 1) @@ -112,6 +115,7 @@ union journal_res_state { */ #define JOURNAL_ENTRY_OFFSET_MAX ((1U << 20) - 1) +#define JOURNAL_ENTRY_BLOCKED_VAL (JOURNAL_ENTRY_OFFSET_MAX - 2) #define JOURNAL_ENTRY_CLOSED_VAL (JOURNAL_ENTRY_OFFSET_MAX - 1) #define JOURNAL_ENTRY_ERROR_VAL (JOURNAL_ENTRY_OFFSET_MAX) @@ -193,6 +197,7 @@ struct journal { * insufficient devices: */ enum journal_errors cur_entry_error; + unsigned cur_entry_offset_if_blocked; unsigned buf_size_want; /* diff --git a/fs/bcachefs/logged_ops.c b/fs/bcachefs/logged_ops.c index 60e00702d1a4..75f27ec26f85 100644 --- a/fs/bcachefs/logged_ops.c +++ b/fs/bcachefs/logged_ops.c @@ -63,8 +63,10 @@ fsck_err: int bch2_resume_logged_ops(struct bch_fs *c) { int ret = bch2_trans_run(c, - for_each_btree_key(trans, iter, - BTREE_ID_logged_ops, POS_MIN, + for_each_btree_key_max(trans, iter, + BTREE_ID_logged_ops, + POS(LOGGED_OPS_INUM_logged_ops, 0), + POS(LOGGED_OPS_INUM_logged_ops, U64_MAX), BTREE_ITER_prefetch, k, resume_logged_op(trans, &iter, k))); bch_err_fn(c, ret); @@ -74,9 +76,8 @@ int bch2_resume_logged_ops(struct bch_fs *c) static int __bch2_logged_op_start(struct btree_trans *trans, struct bkey_i *k) { struct btree_iter iter; - int ret; - - ret = bch2_bkey_get_empty_slot(trans, &iter, BTREE_ID_logged_ops, POS_MAX); + int ret = bch2_bkey_get_empty_slot(trans, &iter, + BTREE_ID_logged_ops, POS(LOGGED_OPS_INUM_logged_ops, U64_MAX)); if (ret) return ret; diff --git a/fs/bcachefs/logged_ops_format.h b/fs/bcachefs/logged_ops_format.h index 6a4bf7129dba..cfb67c95d4c8 100644 --- a/fs/bcachefs/logged_ops_format.h +++ b/fs/bcachefs/logged_ops_format.h @@ -2,6 +2,11 @@ #ifndef _BCACHEFS_LOGGED_OPS_FORMAT_H #define _BCACHEFS_LOGGED_OPS_FORMAT_H +enum logged_ops_inums { + LOGGED_OPS_INUM_logged_ops, + LOGGED_OPS_INUM_inode_cursors, +}; + struct bch_logged_op_truncate { struct bch_val v; __le32 subvol; diff --git a/fs/bcachefs/lru.c b/fs/bcachefs/lru.c index 10857eccdeaf..ce794d55818f 100644 --- a/fs/bcachefs/lru.c +++ b/fs/bcachefs/lru.c @@ -12,7 +12,7 @@ /* KEY_TYPE_lru is obsolete: */ int bch2_lru_validate(struct bch_fs *c, struct bkey_s_c k, - enum bch_validate_flags flags) + struct bkey_validate_context from) { int ret = 0; @@ -192,7 +192,7 @@ int bch2_check_lrus(struct bch_fs *c) int ret = bch2_trans_run(c, for_each_btree_key_commit(trans, iter, BTREE_ID_lru, POS_MIN, BTREE_ITER_prefetch, k, - NULL, NULL, BCH_TRANS_COMMIT_no_enospc|BCH_TRANS_COMMIT_lazy_rw, + NULL, NULL, BCH_TRANS_COMMIT_no_enospc, bch2_check_lru_key(trans, &iter, k, &last_flushed))); bch2_bkey_buf_exit(&last_flushed, c); diff --git a/fs/bcachefs/lru.h b/fs/bcachefs/lru.h index e6a7d8241bb8..f31a6cf1514c 100644 --- a/fs/bcachefs/lru.h +++ b/fs/bcachefs/lru.h @@ -33,7 +33,7 @@ static inline enum bch_lru_type lru_type(struct bkey_s_c l) return BCH_LRU_read; } -int bch2_lru_validate(struct bch_fs *, struct bkey_s_c, enum bch_validate_flags); +int bch2_lru_validate(struct bch_fs *, struct bkey_s_c, struct bkey_validate_context); void bch2_lru_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c); void bch2_lru_pos_to_text(struct printbuf *, struct bpos); diff --git a/fs/bcachefs/move.c b/fs/bcachefs/move.c index 67fb651f4af4..160b4374160a 100644 --- a/fs/bcachefs/move.c +++ b/fs/bcachefs/move.c @@ -21,6 +21,8 @@ #include "journal_reclaim.h" #include "keylist.h" #include "move.h" +#include "rebalance.h" +#include "reflink.h" #include "replicas.h" #include "snapshot.h" #include "super-io.h" @@ -196,6 +198,13 @@ void bch2_moving_ctxt_exit(struct moving_context *ctxt) list_del(&ctxt->list); mutex_unlock(&c->moving_context_lock); + /* + * Generally, releasing a transaction within a transaction restart means + * an unhandled transaction restart: but this can happen legitimately + * within the move code, e.g. when bch2_move_ratelimit() tells us to + * exit before we've retried + */ + bch2_trans_begin(ctxt->trans); bch2_trans_put(ctxt->trans); memset(ctxt, 0, sizeof(*ctxt)); } @@ -379,34 +388,42 @@ err: return ret; } -struct bch_io_opts *bch2_move_get_io_opts(struct btree_trans *trans, +static struct bch_io_opts *bch2_move_get_io_opts(struct btree_trans *trans, struct per_snapshot_io_opts *io_opts, + struct bpos extent_pos, /* extent_iter, extent_k may be in reflink btree */ + struct btree_iter *extent_iter, struct bkey_s_c extent_k) { struct bch_fs *c = trans->c; u32 restart_count = trans->restart_count; + struct bch_io_opts *opts_ret = &io_opts->fs_io_opts; int ret = 0; - if (io_opts->cur_inum != extent_k.k->p.inode) { + if (extent_k.k->type == KEY_TYPE_reflink_v) + goto out; + + if (io_opts->cur_inum != extent_pos.inode) { io_opts->d.nr = 0; - ret = for_each_btree_key(trans, iter, BTREE_ID_inodes, POS(0, extent_k.k->p.inode), + ret = for_each_btree_key(trans, iter, BTREE_ID_inodes, POS(0, extent_pos.inode), BTREE_ITER_all_snapshots, k, ({ - if (k.k->p.offset != extent_k.k->p.inode) + if (k.k->p.offset != extent_pos.inode) break; if (!bkey_is_inode(k.k)) continue; struct bch_inode_unpacked inode; - BUG_ON(bch2_inode_unpack(k, &inode)); + _ret3 = bch2_inode_unpack(k, &inode); + if (_ret3) + break; struct snapshot_io_opts_entry e = { .snapshot = k.k->p.snapshot }; bch2_inode_opts_get(&e.io_opts, trans->c, &inode); darray_push(&io_opts->d, e); })); - io_opts->cur_inum = extent_k.k->p.inode; + io_opts->cur_inum = extent_pos.inode; } ret = ret ?: trans_was_restarted(trans, restart_count); @@ -415,43 +432,46 @@ struct bch_io_opts *bch2_move_get_io_opts(struct btree_trans *trans, if (extent_k.k->p.snapshot) darray_for_each(io_opts->d, i) - if (bch2_snapshot_is_ancestor(c, extent_k.k->p.snapshot, i->snapshot)) - return &i->io_opts; - - return &io_opts->fs_io_opts; + if (bch2_snapshot_is_ancestor(c, extent_k.k->p.snapshot, i->snapshot)) { + opts_ret = &i->io_opts; + break; + } +out: + ret = bch2_get_update_rebalance_opts(trans, opts_ret, extent_iter, extent_k); + if (ret) + return ERR_PTR(ret); + return opts_ret; } int bch2_move_get_io_opts_one(struct btree_trans *trans, struct bch_io_opts *io_opts, + struct btree_iter *extent_iter, struct bkey_s_c extent_k) { - struct btree_iter iter; - struct bkey_s_c k; - int ret; + struct bch_fs *c = trans->c; + + *io_opts = bch2_opts_to_inode_opts(c->opts); /* reflink btree? */ - if (!extent_k.k->p.inode) { - *io_opts = bch2_opts_to_inode_opts(trans->c->opts); - return 0; - } + if (!extent_k.k->p.inode) + goto out; - k = bch2_bkey_get_iter(trans, &iter, BTREE_ID_inodes, + struct btree_iter inode_iter; + struct bkey_s_c inode_k = bch2_bkey_get_iter(trans, &inode_iter, BTREE_ID_inodes, SPOS(0, extent_k.k->p.inode, extent_k.k->p.snapshot), BTREE_ITER_cached); - ret = bkey_err(k); + int ret = bkey_err(inode_k); if (bch2_err_matches(ret, BCH_ERR_transaction_restart)) return ret; - if (!ret && bkey_is_inode(k.k)) { + if (!ret && bkey_is_inode(inode_k.k)) { struct bch_inode_unpacked inode; - bch2_inode_unpack(k, &inode); - bch2_inode_opts_get(io_opts, trans->c, &inode); - } else { - *io_opts = bch2_opts_to_inode_opts(trans->c->opts); + bch2_inode_unpack(inode_k, &inode); + bch2_inode_opts_get(io_opts, c, &inode); } - - bch2_trans_iter_exit(trans, &iter); - return 0; + bch2_trans_iter_exit(trans, &inode_iter); +out: + return bch2_get_update_rebalance_opts(trans, io_opts, extent_iter, extent_k); } int bch2_move_ratelimit(struct moving_context *ctxt) @@ -509,9 +529,15 @@ static int bch2_move_data_btree(struct moving_context *ctxt, struct per_snapshot_io_opts snapshot_io_opts; struct bch_io_opts *io_opts; struct bkey_buf sk; - struct btree_iter iter; + struct btree_iter iter, reflink_iter = {}; struct bkey_s_c k; struct data_update_opts data_opts; + /* + * If we're moving a single file, also process reflinked data it points + * to (this includes propagating changed io_opts from the inode to the + * extent): + */ + bool walk_indirect = start.inode == end.inode; int ret = 0, ret2; per_snapshot_io_opts_init(&snapshot_io_opts, c); @@ -531,6 +557,8 @@ static int bch2_move_data_btree(struct moving_context *ctxt, bch2_ratelimit_reset(ctxt->rate); while (!bch2_move_ratelimit(ctxt)) { + struct btree_iter *extent_iter = &iter; + bch2_trans_begin(trans); k = bch2_btree_iter_peek(&iter); @@ -549,10 +577,36 @@ static int bch2_move_data_btree(struct moving_context *ctxt, if (ctxt->stats) ctxt->stats->pos = BBPOS(iter.btree_id, iter.pos); + if (walk_indirect && + k.k->type == KEY_TYPE_reflink_p && + REFLINK_P_MAY_UPDATE_OPTIONS(bkey_s_c_to_reflink_p(k).v)) { + struct bkey_s_c_reflink_p p = bkey_s_c_to_reflink_p(k); + s64 offset_into_extent = iter.pos.offset - bkey_start_offset(k.k); + + bch2_trans_iter_exit(trans, &reflink_iter); + k = bch2_lookup_indirect_extent(trans, &reflink_iter, &offset_into_extent, p, true, 0); + ret = bkey_err(k); + if (bch2_err_matches(ret, BCH_ERR_transaction_restart)) + continue; + if (ret) + break; + + if (bkey_deleted(k.k)) + goto next_nondata; + + /* + * XXX: reflink pointers may point to multiple indirect + * extents, so don't advance past the entire reflink + * pointer - need to fixup iter->k + */ + extent_iter = &reflink_iter; + } + if (!bkey_extent_is_direct_data(k.k)) goto next_nondata; - io_opts = bch2_move_get_io_opts(trans, &snapshot_io_opts, k); + io_opts = bch2_move_get_io_opts(trans, &snapshot_io_opts, + iter.pos, extent_iter, k); ret = PTR_ERR_OR_ZERO(io_opts); if (ret) continue; @@ -568,7 +622,7 @@ static int bch2_move_data_btree(struct moving_context *ctxt, bch2_bkey_buf_reassemble(&sk, c, k); k = bkey_i_to_s_c(sk.k); - ret2 = bch2_move_extent(ctxt, NULL, &iter, k, *io_opts, data_opts); + ret2 = bch2_move_extent(ctxt, NULL, extent_iter, k, *io_opts, data_opts); if (ret2) { if (bch2_err_matches(ret2, BCH_ERR_transaction_restart)) continue; @@ -589,6 +643,7 @@ next_nondata: bch2_btree_iter_advance(&iter); } + bch2_trans_iter_exit(trans, &reflink_iter); bch2_trans_iter_exit(trans, &iter); bch2_bkey_buf_exit(&sk, c); per_snapshot_io_opts_exit(&snapshot_io_opts); @@ -654,16 +709,12 @@ int bch2_evacuate_bucket(struct moving_context *ctxt, struct bch_fs *c = trans->c; bool is_kthread = current->flags & PF_KTHREAD; struct bch_io_opts io_opts = bch2_opts_to_inode_opts(c->opts); - struct btree_iter iter; + struct btree_iter iter = {}, bp_iter = {}; struct bkey_buf sk; - struct bch_backpointer bp; - struct bch_alloc_v4 a_convert; - const struct bch_alloc_v4 *a; struct bkey_s_c k; struct data_update_opts data_opts; - unsigned dirty_sectors, bucket_size; - u64 fragmentation; - struct bpos bp_pos = POS_MIN; + unsigned sectors_moved = 0; + struct bkey_buf last_flushed; int ret = 0; struct bch_dev *ca = bch2_dev_tryget(c, bucket.inode); @@ -672,6 +723,8 @@ int bch2_evacuate_bucket(struct moving_context *ctxt, trace_bucket_evacuate(c, &bucket); + bch2_bkey_buf_init(&last_flushed); + bkey_init(&last_flushed.k->k); bch2_bkey_buf_init(&sk); /* @@ -679,21 +732,13 @@ int bch2_evacuate_bucket(struct moving_context *ctxt, */ bch2_trans_begin(trans); - bch2_trans_iter_init(trans, &iter, BTREE_ID_alloc, - bucket, BTREE_ITER_cached); - ret = lockrestart_do(trans, - bkey_err(k = bch2_btree_iter_peek_slot(&iter))); - bch2_trans_iter_exit(trans, &iter); + bch2_trans_iter_init(trans, &bp_iter, BTREE_ID_backpointers, + bucket_pos_to_bp_start(ca, bucket), 0); bch_err_msg(c, ret, "looking up alloc key"); if (ret) goto err; - a = bch2_alloc_to_v4(k, &a_convert); - dirty_sectors = bch2_bucket_sectors_dirty(*a); - bucket_size = ca->mi.bucket_size; - fragmentation = alloc_lru_idx_fragmentation(*a, ca); - ret = bch2_btree_write_buffer_tryflush(trans); bch_err_msg(c, ret, "flushing btree write buffer"); if (ret) @@ -705,18 +750,23 @@ int bch2_evacuate_bucket(struct moving_context *ctxt, bch2_trans_begin(trans); - ret = bch2_get_next_backpointer(trans, ca, bucket, gen, - &bp_pos, &bp, - BTREE_ITER_cached); + k = bch2_btree_iter_peek(&bp_iter); + ret = bkey_err(k); if (bch2_err_matches(ret, BCH_ERR_transaction_restart)) continue; if (ret) goto err; - if (bkey_eq(bp_pos, POS_MAX)) + + if (!k.k || bkey_gt(k.k->p, bucket_pos_to_bp_end(ca, bucket))) break; - if (!bp.level) { - k = bch2_backpointer_get_key(trans, &iter, bp_pos, bp, 0); + if (k.k->type != KEY_TYPE_backpointer) + goto next; + + struct bkey_s_c_backpointer bp = bkey_s_c_to_backpointer(k); + + if (!bp.v->level) { + k = bch2_backpointer_get_key(trans, bp, &iter, 0, &last_flushed); ret = bkey_err(k); if (bch2_err_matches(ret, BCH_ERR_transaction_restart)) continue; @@ -728,7 +778,7 @@ int bch2_evacuate_bucket(struct moving_context *ctxt, bch2_bkey_buf_reassemble(&sk, c, k); k = bkey_i_to_s_c(sk.k); - ret = bch2_move_get_io_opts_one(trans, &io_opts, k); + ret = bch2_move_get_io_opts_one(trans, &io_opts, &iter, k); if (ret) { bch2_trans_iter_exit(trans, &iter); continue; @@ -738,14 +788,18 @@ int bch2_evacuate_bucket(struct moving_context *ctxt, data_opts.target = io_opts.background_target; data_opts.rewrite_ptrs = 0; + unsigned sectors = bp.v->bucket_len; /* move_extent will drop locks */ unsigned i = 0; - bkey_for_each_ptr(bch2_bkey_ptrs_c(k), ptr) { - if (ptr->dev == bucket.inode) { - data_opts.rewrite_ptrs |= 1U << i; - if (ptr->cached) { + const union bch_extent_entry *entry; + struct extent_ptr_decoded p; + bkey_for_each_ptr_decode(k.k, bch2_bkey_ptrs_c(k), p, entry) { + if (p.ptr.dev == bucket.inode) { + if (p.ptr.cached) { bch2_trans_iter_exit(trans, &iter); goto next; } + data_opts.rewrite_ptrs |= 1U << i; + break; } i++; } @@ -765,14 +819,15 @@ int bch2_evacuate_bucket(struct moving_context *ctxt, goto err; if (ctxt->stats) - atomic64_add(k.k->size, &ctxt->stats->sectors_seen); + atomic64_add(sectors, &ctxt->stats->sectors_seen); + sectors_moved += sectors; } else { struct btree *b; - b = bch2_backpointer_get_node(trans, &iter, bp_pos, bp); + b = bch2_backpointer_get_node(trans, bp, &iter, &last_flushed); ret = PTR_ERR_OR_ZERO(b); if (ret == -BCH_ERR_backpointer_to_overwritten_btree_node) - continue; + goto next; if (bch2_err_matches(ret, BCH_ERR_transaction_restart)) continue; if (ret) @@ -796,15 +851,18 @@ int bch2_evacuate_bucket(struct moving_context *ctxt, atomic64_add(sectors, &ctxt->stats->sectors_seen); atomic64_add(sectors, &ctxt->stats->sectors_moved); } + sectors_moved += btree_sectors(c); } next: - bp_pos = bpos_nosnap_successor(bp_pos); + bch2_btree_iter_advance(&bp_iter); } - trace_evacuate_bucket(c, &bucket, dirty_sectors, bucket_size, fragmentation, ret); + trace_evacuate_bucket(c, &bucket, sectors_moved, ca->mi.bucket_size, ret); err: + bch2_trans_iter_exit(trans, &bp_iter); bch2_dev_put(ca); bch2_bkey_buf_exit(&sk, c); + bch2_bkey_buf_exit(&last_flushed, c); return ret; } diff --git a/fs/bcachefs/move.h b/fs/bcachefs/move.h index 9baf3093a678..51e0505a8156 100644 --- a/fs/bcachefs/move.h +++ b/fs/bcachefs/move.h @@ -110,9 +110,8 @@ static inline void per_snapshot_io_opts_exit(struct per_snapshot_io_opts *io_opt darray_exit(&io_opts->d); } -struct bch_io_opts *bch2_move_get_io_opts(struct btree_trans *, - struct per_snapshot_io_opts *, struct bkey_s_c); -int bch2_move_get_io_opts_one(struct btree_trans *, struct bch_io_opts *, struct bkey_s_c); +int bch2_move_get_io_opts_one(struct btree_trans *, struct bch_io_opts *, + struct btree_iter *, struct bkey_s_c); int bch2_scan_old_btree_nodes(struct bch_fs *, struct bch_move_stats *); diff --git a/fs/bcachefs/movinggc.c b/fs/bcachefs/movinggc.c index d658be90f737..85c361e78ba5 100644 --- a/fs/bcachefs/movinggc.c +++ b/fs/bcachefs/movinggc.c @@ -167,7 +167,7 @@ static int bch2_copygc_get_buckets(struct moving_context *ctxt, bch2_trans_begin(trans); - ret = for_each_btree_key_upto(trans, iter, BTREE_ID_lru, + ret = for_each_btree_key_max(trans, iter, BTREE_ID_lru, lru_pos(BCH_LRU_FRAGMENTATION_START, 0, 0), lru_pos(BCH_LRU_FRAGMENTATION_START, U64_MAX, LRU_TIME_MAX), 0, k, ({ @@ -350,9 +350,9 @@ static int bch2_copygc_thread(void *arg) bch2_trans_unlock_long(ctxt.trans); cond_resched(); - if (!c->copy_gc_enabled) { + if (!c->opts.copygc_enabled) { move_buckets_wait(&ctxt, buckets, true); - kthread_wait_freezable(c->copy_gc_enabled || + kthread_wait_freezable(c->opts.copygc_enabled || kthread_should_stop()); } diff --git a/fs/bcachefs/opts.c b/fs/bcachefs/opts.c index 0e2ee262fbd4..6772faf385a5 100644 --- a/fs/bcachefs/opts.c +++ b/fs/bcachefs/opts.c @@ -1,6 +1,7 @@ // SPDX-License-Identifier: GPL-2.0 #include <linux/kernel.h> +#include <linux/fs_parser.h> #include "bcachefs.h" #include "compress.h" @@ -48,12 +49,12 @@ static const char * const __bch2_csum_types[] = { NULL }; -const char * const bch2_csum_opts[] = { +const char * const __bch2_csum_opts[] = { BCH_CSUM_OPTS() NULL }; -static const char * const __bch2_compression_types[] = { +const char * const __bch2_compression_types[] = { BCH_COMPRESSION_TYPES() NULL }; @@ -113,6 +114,7 @@ void bch2_prt_##name(struct printbuf *out, type t) \ PRT_STR_OPT_BOUNDSCHECKED(jset_entry_type, enum bch_jset_entry_type); PRT_STR_OPT_BOUNDSCHECKED(fs_usage_type, enum bch_fs_usage_type); PRT_STR_OPT_BOUNDSCHECKED(data_type, enum bch_data_type); +PRT_STR_OPT_BOUNDSCHECKED(csum_opt, enum bch_csum_opt); PRT_STR_OPT_BOUNDSCHECKED(csum_type, enum bch_csum_type); PRT_STR_OPT_BOUNDSCHECKED(compression_type, enum bch_compression_type); PRT_STR_OPT_BOUNDSCHECKED(str_hash_type, enum bch_str_hash_type); @@ -333,17 +335,18 @@ int bch2_opt_parse(struct bch_fs *c, switch (opt->type) { case BCH_OPT_BOOL: if (val) { - ret = kstrtou64(val, 10, res); + ret = lookup_constant(bool_names, val, -BCH_ERR_option_not_bool); + if (ret != -BCH_ERR_option_not_bool) { + *res = ret; + } else { + if (err) + prt_printf(err, "%s: must be bool", opt->attr.name); + return ret; + } } else { - ret = 0; *res = 1; } - if (ret < 0 || (*res != 0 && *res != 1)) { - if (err) - prt_printf(err, "%s: must be bool", opt->attr.name); - return ret < 0 ? ret : -BCH_ERR_option_not_bool; - } break; case BCH_OPT_UINT: if (!val) { @@ -710,11 +713,14 @@ void bch2_opt_set_sb(struct bch_fs *c, struct bch_dev *ca, struct bch_io_opts bch2_opts_to_inode_opts(struct bch_opts src) { - return (struct bch_io_opts) { + struct bch_io_opts opts = { #define x(_name, _bits) ._name = src._name, BCH_INODE_OPTS() #undef x }; + + bch2_io_opts_fixups(&opts); + return opts; } bool bch2_opt_is_inode_opt(enum bch_opt_id id) diff --git a/fs/bcachefs/opts.h b/fs/bcachefs/opts.h index 23dda014e331..e763d52e0f38 100644 --- a/fs/bcachefs/opts.h +++ b/fs/bcachefs/opts.h @@ -16,7 +16,8 @@ extern const char * const bch2_version_upgrade_opts[]; extern const char * const bch2_sb_features[]; extern const char * const bch2_sb_compat[]; extern const char * const __bch2_btree_ids[]; -extern const char * const bch2_csum_opts[]; +extern const char * const __bch2_csum_opts[]; +extern const char * const __bch2_compression_types[]; extern const char * const bch2_compression_opts[]; extern const char * const __bch2_str_hash_types[]; extern const char * const bch2_str_hash_opts[]; @@ -27,6 +28,7 @@ extern const char * const bch2_d_types[]; void bch2_prt_jset_entry_type(struct printbuf *, enum bch_jset_entry_type); void bch2_prt_fs_usage_type(struct printbuf *, enum bch_fs_usage_type); void bch2_prt_data_type(struct printbuf *, enum bch_data_type); +void bch2_prt_csum_opt(struct printbuf *, enum bch_csum_opt); void bch2_prt_csum_type(struct printbuf *, enum bch_csum_type); void bch2_prt_compression_type(struct printbuf *, enum bch_compression_type); void bch2_prt_str_hash_type(struct printbuf *, enum bch_str_hash_type); @@ -171,12 +173,12 @@ enum fsck_err_opts { "size", "Maximum size of checksummed/compressed extents")\ x(metadata_checksum, u8, \ OPT_FS|OPT_FORMAT|OPT_MOUNT|OPT_RUNTIME, \ - OPT_STR(bch2_csum_opts), \ + OPT_STR(__bch2_csum_opts), \ BCH_SB_META_CSUM_TYPE, BCH_CSUM_OPT_crc32c, \ NULL, NULL) \ x(data_checksum, u8, \ OPT_FS|OPT_INODE|OPT_FORMAT|OPT_MOUNT|OPT_RUNTIME, \ - OPT_STR(bch2_csum_opts), \ + OPT_STR(__bch2_csum_opts), \ BCH_SB_DATA_CSUM_TYPE, BCH_CSUM_OPT_crc32c, \ NULL, NULL) \ x(compression, u8, \ @@ -220,14 +222,14 @@ enum fsck_err_opts { BCH_SB_ERASURE_CODE, false, \ NULL, "Enable erasure coding (DO NOT USE YET)") \ x(inodes_32bit, u8, \ - OPT_FS|OPT_FORMAT|OPT_MOUNT|OPT_RUNTIME, \ + OPT_FS|OPT_INODE|OPT_FORMAT|OPT_MOUNT|OPT_RUNTIME, \ OPT_BOOL(), \ BCH_SB_INODE_32BIT, true, \ NULL, "Constrain inode numbers to 32 bits") \ - x(shard_inode_numbers, u8, \ - OPT_FS|OPT_FORMAT|OPT_MOUNT|OPT_RUNTIME, \ - OPT_BOOL(), \ - BCH_SB_SHARD_INUMS, true, \ + x(shard_inode_numbers_bits, u8, \ + OPT_FS|OPT_FORMAT, \ + OPT_UINT(0, 8), \ + BCH_SB_SHARD_INUMS_NBITS, 0, \ NULL, "Shard new inode numbers by CPU id") \ x(inodes_use_key_cache, u8, \ OPT_FS|OPT_FORMAT|OPT_MOUNT, \ @@ -473,6 +475,18 @@ enum fsck_err_opts { BCH2_NO_SB_OPT, true, \ NULL, "Enable nocow mode: enables runtime locking in\n"\ "data move path needed if nocow will ever be in use\n")\ + x(copygc_enabled, u8, \ + OPT_FS|OPT_MOUNT, \ + OPT_BOOL(), \ + BCH2_NO_SB_OPT, true, \ + NULL, "Enable copygc: disable for debugging, or to\n"\ + "quiet the system when doing performance testing\n")\ + x(rebalance_enabled, u8, \ + OPT_FS|OPT_MOUNT, \ + OPT_BOOL(), \ + BCH2_NO_SB_OPT, true, \ + NULL, "Enable rebalance: disable for debugging, or to\n"\ + "quiet the system when doing performance testing\n")\ x(no_data_io, u8, \ OPT_MOUNT, \ OPT_BOOL(), \ @@ -488,7 +502,7 @@ enum fsck_err_opts { OPT_DEVICE, \ OPT_UINT(0, S64_MAX), \ BCH2_NO_SB_OPT, 0, \ - "size", "Size of filesystem on device") \ + "size", "Specifies the bucket size; must be greater than the btree node size")\ x(durability, u8, \ OPT_DEVICE|OPT_SB_FIELD_ONE_BIAS, \ OPT_UINT(0, BCH_REPLICAS_MAX), \ @@ -624,14 +638,39 @@ struct bch_io_opts { #define x(_name, _bits) u##_bits _name; BCH_INODE_OPTS() #undef x +#define x(_name, _bits) u64 _name##_from_inode:1; + BCH_INODE_OPTS() +#undef x }; -static inline unsigned background_compression(struct bch_io_opts opts) +static inline void bch2_io_opts_fixups(struct bch_io_opts *opts) { - return opts.background_compression ?: opts.compression; + if (!opts->background_target) + opts->background_target = opts->foreground_target; + if (!opts->background_compression) + opts->background_compression = opts->compression; + if (opts->nocow) { + opts->compression = opts->background_compression = 0; + opts->data_checksum = 0; + opts->erasure_code = 0; + } } struct bch_io_opts bch2_opts_to_inode_opts(struct bch_opts); bool bch2_opt_is_inode_opt(enum bch_opt_id); +/* rebalance opts: */ + +static inline struct bch_extent_rebalance io_opts_to_rebalance_opts(struct bch_io_opts *opts) +{ + return (struct bch_extent_rebalance) { + .type = BIT(BCH_EXTENT_ENTRY_rebalance), +#define x(_name) \ + ._name = opts->_name, \ + ._name##_from_inode = opts->_name##_from_inode, + BCH_REBALANCE_OPTS() +#undef x + }; +}; + #endif /* _BCACHEFS_OPTS_H */ diff --git a/fs/bcachefs/printbuf.h b/fs/bcachefs/printbuf.h index 1d570387b77f..d0dd398baa2b 100644 --- a/fs/bcachefs/printbuf.h +++ b/fs/bcachefs/printbuf.h @@ -251,16 +251,23 @@ static inline void prt_hex_byte_upper(struct printbuf *out, u8 byte) printbuf_nul_terminate_reserved(out); } +static inline void printbuf_reset_keep_tabstops(struct printbuf *buf) +{ + buf->pos = 0; + buf->allocation_failure = 0; + buf->last_newline = 0; + buf->last_field = 0; + buf->indent = 0; + buf->cur_tabstop = 0; +} + /** * printbuf_reset - re-use a printbuf without freeing and re-initializing it: */ static inline void printbuf_reset(struct printbuf *buf) { - buf->pos = 0; - buf->allocation_failure = 0; - buf->indent = 0; + printbuf_reset_keep_tabstops(buf); buf->nr_tabstops = 0; - buf->cur_tabstop = 0; } /** diff --git a/fs/bcachefs/quota.c b/fs/bcachefs/quota.c index 74f45a8162ad..8b857fc33244 100644 --- a/fs/bcachefs/quota.c +++ b/fs/bcachefs/quota.c @@ -60,7 +60,7 @@ const struct bch_sb_field_ops bch_sb_field_ops_quota = { }; int bch2_quota_validate(struct bch_fs *c, struct bkey_s_c k, - enum bch_validate_flags flags) + struct bkey_validate_context from) { int ret = 0; diff --git a/fs/bcachefs/quota.h b/fs/bcachefs/quota.h index a62abcc5332a..1551800ff44c 100644 --- a/fs/bcachefs/quota.h +++ b/fs/bcachefs/quota.h @@ -5,10 +5,10 @@ #include "inode.h" #include "quota_types.h" -enum bch_validate_flags; extern const struct bch_sb_field_ops bch_sb_field_ops_quota; -int bch2_quota_validate(struct bch_fs *, struct bkey_s_c, enum bch_validate_flags); +int bch2_quota_validate(struct bch_fs *, struct bkey_s_c, + struct bkey_validate_context); void bch2_quota_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c); #define bch2_bkey_ops_quota ((struct bkey_ops) { \ diff --git a/fs/bcachefs/rcu_pending.c b/fs/bcachefs/rcu_pending.c index 40a20192eee8..bef2aa1b8bcd 100644 --- a/fs/bcachefs/rcu_pending.c +++ b/fs/bcachefs/rcu_pending.c @@ -25,21 +25,37 @@ enum rcu_pending_special { #define RCU_PENDING_KVFREE_FN ((rcu_pending_process_fn) (ulong) RCU_PENDING_KVFREE) #define RCU_PENDING_CALL_RCU_FN ((rcu_pending_process_fn) (ulong) RCU_PENDING_CALL_RCU) -static inline unsigned long __get_state_synchronize_rcu(struct srcu_struct *ssp) +#ifdef __KERNEL__ +typedef unsigned long rcu_gp_poll_state_t; + +static inline bool rcu_gp_poll_cookie_eq(rcu_gp_poll_state_t l, rcu_gp_poll_state_t r) +{ + return l == r; +} +#else +typedef struct urcu_gp_poll_state rcu_gp_poll_state_t; + +static inline bool rcu_gp_poll_cookie_eq(rcu_gp_poll_state_t l, rcu_gp_poll_state_t r) +{ + return l.grace_period_id == r.grace_period_id; +} +#endif + +static inline rcu_gp_poll_state_t __get_state_synchronize_rcu(struct srcu_struct *ssp) { return ssp ? get_state_synchronize_srcu(ssp) : get_state_synchronize_rcu(); } -static inline unsigned long __start_poll_synchronize_rcu(struct srcu_struct *ssp) +static inline rcu_gp_poll_state_t __start_poll_synchronize_rcu(struct srcu_struct *ssp) { return ssp ? start_poll_synchronize_srcu(ssp) : start_poll_synchronize_rcu(); } -static inline bool __poll_state_synchronize_rcu(struct srcu_struct *ssp, unsigned long cookie) +static inline bool __poll_state_synchronize_rcu(struct srcu_struct *ssp, rcu_gp_poll_state_t cookie) { return ssp ? poll_state_synchronize_srcu(ssp, cookie) @@ -71,13 +87,13 @@ struct rcu_pending_seq { GENRADIX(struct rcu_head *) objs; size_t nr; struct rcu_head **cursor; - unsigned long seq; + rcu_gp_poll_state_t seq; }; struct rcu_pending_list { struct rcu_head *head; struct rcu_head *tail; - unsigned long seq; + rcu_gp_poll_state_t seq; }; struct rcu_pending_pcpu { @@ -316,10 +332,10 @@ static void rcu_pending_rcu_cb(struct rcu_head *rcu) } static __always_inline struct rcu_pending_seq * -get_object_radix(struct rcu_pending_pcpu *p, unsigned long seq) +get_object_radix(struct rcu_pending_pcpu *p, rcu_gp_poll_state_t seq) { darray_for_each_reverse(p->objs, objs) - if (objs->seq == seq) + if (rcu_gp_poll_cookie_eq(objs->seq, seq)) return objs; if (darray_push_gfp(&p->objs, ((struct rcu_pending_seq) { .seq = seq }), GFP_ATOMIC)) @@ -329,7 +345,7 @@ get_object_radix(struct rcu_pending_pcpu *p, unsigned long seq) } static noinline bool -rcu_pending_enqueue_list(struct rcu_pending_pcpu *p, unsigned long seq, +rcu_pending_enqueue_list(struct rcu_pending_pcpu *p, rcu_gp_poll_state_t seq, struct rcu_head *head, void *ptr, unsigned long *flags) { @@ -364,7 +380,7 @@ rcu_pending_enqueue_list(struct rcu_pending_pcpu *p, unsigned long seq, again: for (struct rcu_pending_list *i = p->lists; i < p->lists + NUM_ACTIVE_RCU_POLL_OLDSTATE; i++) { - if (i->seq == seq) { + if (rcu_gp_poll_cookie_eq(i->seq, seq)) { rcu_pending_list_add(i, head); return false; } @@ -408,7 +424,7 @@ __rcu_pending_enqueue(struct rcu_pending *pending, struct rcu_head *head, struct rcu_pending_pcpu *p; struct rcu_pending_seq *objs; struct genradix_node *new_node = NULL; - unsigned long seq, flags; + unsigned long flags; bool start_gp = false; BUG_ON((ptr != NULL) != (pending->process == RCU_PENDING_KVFREE_FN)); @@ -416,7 +432,7 @@ __rcu_pending_enqueue(struct rcu_pending *pending, struct rcu_head *head, local_irq_save(flags); p = this_cpu_ptr(pending->p); spin_lock(&p->lock); - seq = __get_state_synchronize_rcu(pending->srcu); + rcu_gp_poll_state_t seq = __get_state_synchronize_rcu(pending->srcu); restart: if (may_sleep && unlikely(process_finished_items(pending, p, flags))) diff --git a/fs/bcachefs/rebalance.c b/fs/bcachefs/rebalance.c index cd6647374353..4adc74cd3f70 100644 --- a/fs/bcachefs/rebalance.c +++ b/fs/bcachefs/rebalance.c @@ -24,6 +24,192 @@ #include <linux/kthread.h> #include <linux/sched/cputime.h> +/* bch_extent_rebalance: */ + +static const struct bch_extent_rebalance *bch2_bkey_rebalance_opts(struct bkey_s_c k) +{ + struct bkey_ptrs_c ptrs = bch2_bkey_ptrs_c(k); + const union bch_extent_entry *entry; + + bkey_extent_entry_for_each(ptrs, entry) + if (__extent_entry_type(entry) == BCH_EXTENT_ENTRY_rebalance) + return &entry->rebalance; + + return NULL; +} + +static inline unsigned bch2_bkey_ptrs_need_compress(struct bch_fs *c, + struct bch_io_opts *opts, + struct bkey_s_c k, + struct bkey_ptrs_c ptrs) +{ + if (!opts->background_compression) + return 0; + + unsigned compression_type = bch2_compression_opt_to_type(opts->background_compression); + const union bch_extent_entry *entry; + struct extent_ptr_decoded p; + unsigned ptr_bit = 1; + unsigned rewrite_ptrs = 0; + + bkey_for_each_ptr_decode(k.k, ptrs, p, entry) { + if (p.crc.compression_type == BCH_COMPRESSION_TYPE_incompressible || + p.ptr.unwritten) + return 0; + + if (!p.ptr.cached && p.crc.compression_type != compression_type) + rewrite_ptrs |= ptr_bit; + ptr_bit <<= 1; + } + + return rewrite_ptrs; +} + +static inline unsigned bch2_bkey_ptrs_need_move(struct bch_fs *c, + struct bch_io_opts *opts, + struct bkey_ptrs_c ptrs) +{ + if (!opts->background_target || + !bch2_target_accepts_data(c, BCH_DATA_user, opts->background_target)) + return 0; + + unsigned ptr_bit = 1; + unsigned rewrite_ptrs = 0; + + bkey_for_each_ptr(ptrs, ptr) { + if (!ptr->cached && !bch2_dev_in_target(c, ptr->dev, opts->background_target)) + rewrite_ptrs |= ptr_bit; + ptr_bit <<= 1; + } + + return rewrite_ptrs; +} + +static unsigned bch2_bkey_ptrs_need_rebalance(struct bch_fs *c, + struct bch_io_opts *opts, + struct bkey_s_c k) +{ + struct bkey_ptrs_c ptrs = bch2_bkey_ptrs_c(k); + + return bch2_bkey_ptrs_need_compress(c, opts, k, ptrs) | + bch2_bkey_ptrs_need_move(c, opts, ptrs); +} + +u64 bch2_bkey_sectors_need_rebalance(struct bch_fs *c, struct bkey_s_c k) +{ + const struct bch_extent_rebalance *opts = bch2_bkey_rebalance_opts(k); + if (!opts) + return 0; + + struct bkey_ptrs_c ptrs = bch2_bkey_ptrs_c(k); + const union bch_extent_entry *entry; + struct extent_ptr_decoded p; + u64 sectors = 0; + + if (opts->background_compression) { + unsigned compression_type = bch2_compression_opt_to_type(opts->background_compression); + + bkey_for_each_ptr_decode(k.k, ptrs, p, entry) { + if (p.crc.compression_type == BCH_COMPRESSION_TYPE_incompressible || + p.ptr.unwritten) { + sectors = 0; + goto incompressible; + } + + if (!p.ptr.cached && p.crc.compression_type != compression_type) + sectors += p.crc.compressed_size; + } + } +incompressible: + if (opts->background_target && + bch2_target_accepts_data(c, BCH_DATA_user, opts->background_target)) { + bkey_for_each_ptr_decode(k.k, ptrs, p, entry) + if (!p.ptr.cached && !bch2_dev_in_target(c, p.ptr.dev, opts->background_target)) + sectors += p.crc.compressed_size; + } + + return sectors; +} + +static bool bch2_bkey_rebalance_needs_update(struct bch_fs *c, struct bch_io_opts *opts, + struct bkey_s_c k) +{ + if (!bkey_extent_is_direct_data(k.k)) + return 0; + + const struct bch_extent_rebalance *old = bch2_bkey_rebalance_opts(k); + + if (k.k->type == KEY_TYPE_reflink_v || bch2_bkey_ptrs_need_rebalance(c, opts, k)) { + struct bch_extent_rebalance new = io_opts_to_rebalance_opts(opts); + return old == NULL || memcmp(old, &new, sizeof(new)); + } else { + return old != NULL; + } +} + +int bch2_bkey_set_needs_rebalance(struct bch_fs *c, struct bch_io_opts *opts, + struct bkey_i *_k) +{ + if (!bkey_extent_is_direct_data(&_k->k)) + return 0; + + struct bkey_s k = bkey_i_to_s(_k); + struct bch_extent_rebalance *old = + (struct bch_extent_rebalance *) bch2_bkey_rebalance_opts(k.s_c); + + if (k.k->type == KEY_TYPE_reflink_v || bch2_bkey_ptrs_need_rebalance(c, opts, k.s_c)) { + if (!old) { + old = bkey_val_end(k); + k.k->u64s += sizeof(*old) / sizeof(u64); + } + + *old = io_opts_to_rebalance_opts(opts); + } else { + if (old) + extent_entry_drop(k, (union bch_extent_entry *) old); + } + + return 0; +} + +int bch2_get_update_rebalance_opts(struct btree_trans *trans, + struct bch_io_opts *io_opts, + struct btree_iter *iter, + struct bkey_s_c k) +{ + BUG_ON(iter->flags & BTREE_ITER_is_extents); + BUG_ON(iter->flags & BTREE_ITER_filter_snapshots); + + const struct bch_extent_rebalance *r = k.k->type == KEY_TYPE_reflink_v + ? bch2_bkey_rebalance_opts(k) : NULL; + if (r) { +#define x(_name) \ + if (r->_name##_from_inode) { \ + io_opts->_name = r->_name; \ + io_opts->_name##_from_inode = true; \ + } + BCH_REBALANCE_OPTS() +#undef x + } + + if (!bch2_bkey_rebalance_needs_update(trans->c, io_opts, k)) + return 0; + + struct bkey_i *n = bch2_trans_kmalloc(trans, bkey_bytes(k.k) + 8); + int ret = PTR_ERR_OR_ZERO(n); + if (ret) + return ret; + + bkey_reassemble(n, k); + + /* On successfull transaction commit, @k was invalidated: */ + + return bch2_bkey_set_needs_rebalance(trans->c, io_opts, n) ?: + bch2_trans_update(trans, iter, n, BTREE_UPDATE_internal_snapshot_node) ?: + bch2_trans_commit(trans, NULL, NULL, 0) ?: + -BCH_ERR_transaction_restart_nested; +} + #define REBALANCE_WORK_SCAN_OFFSET (U64_MAX - 1) static const char * const bch2_rebalance_state_strs[] = { @@ -33,7 +219,7 @@ static const char * const bch2_rebalance_state_strs[] = { #undef x }; -static int __bch2_set_rebalance_needs_scan(struct btree_trans *trans, u64 inum) +int bch2_set_rebalance_needs_scan_trans(struct btree_trans *trans, u64 inum) { struct btree_iter iter; struct bkey_s_c k; @@ -71,9 +257,8 @@ err: int bch2_set_rebalance_needs_scan(struct bch_fs *c, u64 inum) { int ret = bch2_trans_commit_do(c, NULL, NULL, - BCH_TRANS_COMMIT_no_enospc| - BCH_TRANS_COMMIT_lazy_rw, - __bch2_set_rebalance_needs_scan(trans, inum)); + BCH_TRANS_COMMIT_no_enospc, + bch2_set_rebalance_needs_scan_trans(trans, inum)); rebalance_wakeup(c); return ret; } @@ -121,6 +306,9 @@ static int bch2_bkey_clear_needs_rebalance(struct btree_trans *trans, struct btree_iter *iter, struct bkey_s_c k) { + if (!bch2_bkey_rebalance_opts(k)) + return 0; + struct bkey_i *n = bch2_bkey_make_mut(trans, iter, &k, 0); int ret = PTR_ERR_OR_ZERO(n); if (ret) @@ -134,31 +322,27 @@ static int bch2_bkey_clear_needs_rebalance(struct btree_trans *trans, static struct bkey_s_c next_rebalance_extent(struct btree_trans *trans, struct bpos work_pos, struct btree_iter *extent_iter, + struct bch_io_opts *io_opts, struct data_update_opts *data_opts) { struct bch_fs *c = trans->c; - struct bkey_s_c k; bch2_trans_iter_exit(trans, extent_iter); bch2_trans_iter_init(trans, extent_iter, work_pos.inode ? BTREE_ID_extents : BTREE_ID_reflink, work_pos, BTREE_ITER_all_snapshots); - k = bch2_btree_iter_peek_slot(extent_iter); + struct bkey_s_c k = bch2_btree_iter_peek_slot(extent_iter); if (bkey_err(k)) return k; - const struct bch_extent_rebalance *r = k.k ? bch2_bkey_rebalance_opts(k) : NULL; - if (!r) { - /* raced due to btree write buffer, nothing to do */ - return bkey_s_c_null; - } + int ret = bch2_move_get_io_opts_one(trans, io_opts, extent_iter, k); + if (ret) + return bkey_s_c_err(ret); memset(data_opts, 0, sizeof(*data_opts)); - - data_opts->rewrite_ptrs = - bch2_bkey_ptrs_need_rebalance(c, k, r->target, r->compression); - data_opts->target = r->target; + data_opts->rewrite_ptrs = bch2_bkey_ptrs_need_rebalance(c, io_opts, k); + data_opts->target = io_opts->background_target; data_opts->write_flags |= BCH_WRITE_ONLY_SPECIFIED_DEVS; if (!data_opts->rewrite_ptrs) { @@ -178,12 +362,28 @@ static struct bkey_s_c next_rebalance_extent(struct btree_trans *trans, if (trace_rebalance_extent_enabled()) { struct printbuf buf = PRINTBUF; - prt_str(&buf, "target="); - bch2_target_to_text(&buf, c, r->target); - prt_str(&buf, " compression="); - bch2_compression_opt_to_text(&buf, r->compression); - prt_str(&buf, " "); bch2_bkey_val_to_text(&buf, c, k); + prt_newline(&buf); + + struct bkey_ptrs_c ptrs = bch2_bkey_ptrs_c(k); + + unsigned p = bch2_bkey_ptrs_need_compress(c, io_opts, k, ptrs); + if (p) { + prt_str(&buf, "compression="); + bch2_compression_opt_to_text(&buf, io_opts->background_compression); + prt_str(&buf, " "); + bch2_prt_u64_base2(&buf, p); + prt_newline(&buf); + } + + p = bch2_bkey_ptrs_need_move(c, io_opts, ptrs); + if (p) { + prt_str(&buf, "move="); + bch2_target_to_text(&buf, c, io_opts->background_target); + prt_str(&buf, " "); + bch2_prt_u64_base2(&buf, p); + prt_newline(&buf); + } trace_rebalance_extent(c, buf.buf); printbuf_exit(&buf); @@ -212,14 +412,10 @@ static int do_rebalance_extent(struct moving_context *ctxt, bch2_bkey_buf_init(&sk); ret = bkey_err(k = next_rebalance_extent(trans, work_pos, - extent_iter, &data_opts)); + extent_iter, &io_opts, &data_opts)); if (ret || !k.k) goto out; - ret = bch2_move_get_io_opts_one(trans, &io_opts, k); - if (ret) - goto out; - atomic64_add(k.k->size, &ctxt->stats->sectors_seen); /* @@ -253,20 +449,8 @@ static bool rebalance_pred(struct bch_fs *c, void *arg, struct bch_io_opts *io_opts, struct data_update_opts *data_opts) { - unsigned target, compression; - - if (k.k->p.inode) { - target = io_opts->background_target; - compression = background_compression(*io_opts); - } else { - const struct bch_extent_rebalance *r = bch2_bkey_rebalance_opts(k); - - target = r ? r->target : io_opts->background_target; - compression = r ? r->compression : background_compression(*io_opts); - } - - data_opts->rewrite_ptrs = bch2_bkey_ptrs_need_rebalance(c, k, target, compression); - data_opts->target = target; + data_opts->rewrite_ptrs = bch2_bkey_ptrs_need_rebalance(c, io_opts, k); + data_opts->target = io_opts->background_target; data_opts->write_flags |= BCH_WRITE_ONLY_SPECIFIED_DEVS; return data_opts->rewrite_ptrs != 0; } @@ -338,9 +522,9 @@ static int do_rebalance(struct moving_context *ctxt) BTREE_ITER_all_snapshots); while (!bch2_move_ratelimit(ctxt)) { - if (!r->enabled) { + if (!c->opts.rebalance_enabled) { bch2_moving_ctxt_flush_all(ctxt); - kthread_wait_freezable(r->enabled || + kthread_wait_freezable(c->opts.rebalance_enabled || kthread_should_stop()); } diff --git a/fs/bcachefs/rebalance.h b/fs/bcachefs/rebalance.h index 28a52638f16c..0a0821ab895d 100644 --- a/fs/bcachefs/rebalance.h +++ b/fs/bcachefs/rebalance.h @@ -2,8 +2,18 @@ #ifndef _BCACHEFS_REBALANCE_H #define _BCACHEFS_REBALANCE_H +#include "compress.h" +#include "disk_groups.h" #include "rebalance_types.h" +u64 bch2_bkey_sectors_need_rebalance(struct bch_fs *, struct bkey_s_c); +int bch2_bkey_set_needs_rebalance(struct bch_fs *, struct bch_io_opts *, struct bkey_i *); +int bch2_get_update_rebalance_opts(struct btree_trans *, + struct bch_io_opts *, + struct btree_iter *, + struct bkey_s_c); + +int bch2_set_rebalance_needs_scan_trans(struct btree_trans *, u64); int bch2_set_rebalance_needs_scan(struct bch_fs *, u64 inum); int bch2_set_fs_needs_rebalance(struct bch_fs *); diff --git a/fs/bcachefs/rebalance_format.h b/fs/bcachefs/rebalance_format.h new file mode 100644 index 000000000000..ff9a1342a22b --- /dev/null +++ b/fs/bcachefs/rebalance_format.h @@ -0,0 +1,53 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef _BCACHEFS_REBALANCE_FORMAT_H +#define _BCACHEFS_REBALANCE_FORMAT_H + +struct bch_extent_rebalance { +#if defined(__LITTLE_ENDIAN_BITFIELD) + __u64 type:6, + unused:3, + + promote_target_from_inode:1, + erasure_code_from_inode:1, + data_checksum_from_inode:1, + background_compression_from_inode:1, + data_replicas_from_inode:1, + background_target_from_inode:1, + + promote_target:16, + erasure_code:1, + data_checksum:4, + data_replicas:4, + background_compression:8, /* enum bch_compression_opt */ + background_target:16; +#elif defined (__BIG_ENDIAN_BITFIELD) + __u64 background_target:16, + background_compression:8, + data_replicas:4, + data_checksum:4, + erasure_code:1, + promote_target:16, + + background_target_from_inode:1, + data_replicas_from_inode:1, + background_compression_from_inode:1, + data_checksum_from_inode:1, + erasure_code_from_inode:1, + promote_target_from_inode:1, + + unused:3, + type:6; +#endif +}; + +/* subset of BCH_INODE_OPTS */ +#define BCH_REBALANCE_OPTS() \ + x(data_checksum) \ + x(background_compression) \ + x(data_replicas) \ + x(promote_target) \ + x(background_target) \ + x(erasure_code) + +#endif /* _BCACHEFS_REBALANCE_FORMAT_H */ + diff --git a/fs/bcachefs/rebalance_types.h b/fs/bcachefs/rebalance_types.h index 0fffb536c1d0..fe5098c17dfc 100644 --- a/fs/bcachefs/rebalance_types.h +++ b/fs/bcachefs/rebalance_types.h @@ -30,8 +30,6 @@ struct bch_fs_rebalance { struct bbpos scan_start; struct bbpos scan_end; struct bch_move_stats scan_stats; - - unsigned enabled:1; }; #endif /* _BCACHEFS_REBALANCE_TYPES_H */ diff --git a/fs/bcachefs/recovery.c b/fs/bcachefs/recovery.c index 3c7f941dde39..98825437381c 100644 --- a/fs/bcachefs/recovery.c +++ b/fs/bcachefs/recovery.c @@ -34,21 +34,83 @@ #define QSTR(n) { { { .len = strlen(n) } }, .name = n } -void bch2_btree_lost_data(struct bch_fs *c, enum btree_id btree) +int bch2_btree_lost_data(struct bch_fs *c, enum btree_id btree) { - if (btree >= BTREE_ID_NR_MAX) - return; - u64 b = BIT_ULL(btree); + int ret = 0; + + mutex_lock(&c->sb_lock); + struct bch_sb_field_ext *ext = bch2_sb_field_get(c->disk_sb.sb, ext); if (!(c->sb.btrees_lost_data & b)) { - bch_err(c, "flagging btree %s lost data", bch2_btree_id_str(btree)); + struct printbuf buf = PRINTBUF; + bch2_btree_id_to_text(&buf, btree); + bch_err(c, "flagging btree %s lost data", buf.buf); + printbuf_exit(&buf); + ext->btrees_lost_data |= cpu_to_le64(b); + } - mutex_lock(&c->sb_lock); - bch2_sb_field_get(c->disk_sb.sb, ext)->btrees_lost_data |= cpu_to_le64(b); - bch2_write_super(c); - mutex_unlock(&c->sb_lock); + /* Once we have runtime self healing for topology errors we won't need this: */ + ret = bch2_run_explicit_recovery_pass_persistent_locked(c, BCH_RECOVERY_PASS_check_topology) ?: ret; + + /* Btree node accounting will be off: */ + __set_bit_le64(BCH_FSCK_ERR_accounting_mismatch, ext->errors_silent); + ret = bch2_run_explicit_recovery_pass_persistent_locked(c, BCH_RECOVERY_PASS_check_allocations) ?: ret; + +#ifdef CONFIG_BCACHEFS_DEBUG + /* + * These are much more minor, and don't need to be corrected right away, + * but in debug mode we want the next fsck run to be clean: + */ + ret = bch2_run_explicit_recovery_pass_persistent_locked(c, BCH_RECOVERY_PASS_check_lrus) ?: ret; + ret = bch2_run_explicit_recovery_pass_persistent_locked(c, BCH_RECOVERY_PASS_check_backpointers_to_extents) ?: ret; +#endif + + switch (btree) { + case BTREE_ID_alloc: + ret = bch2_run_explicit_recovery_pass_persistent_locked(c, BCH_RECOVERY_PASS_check_alloc_info) ?: ret; + + __set_bit_le64(BCH_FSCK_ERR_alloc_key_data_type_wrong, ext->errors_silent); + __set_bit_le64(BCH_FSCK_ERR_alloc_key_gen_wrong, ext->errors_silent); + __set_bit_le64(BCH_FSCK_ERR_alloc_key_dirty_sectors_wrong, ext->errors_silent); + __set_bit_le64(BCH_FSCK_ERR_alloc_key_cached_sectors_wrong, ext->errors_silent); + __set_bit_le64(BCH_FSCK_ERR_alloc_key_stripe_wrong, ext->errors_silent); + __set_bit_le64(BCH_FSCK_ERR_alloc_key_stripe_redundancy_wrong, ext->errors_silent); + goto out; + case BTREE_ID_backpointers: + ret = bch2_run_explicit_recovery_pass_persistent_locked(c, BCH_RECOVERY_PASS_check_btree_backpointers) ?: ret; + ret = bch2_run_explicit_recovery_pass_persistent_locked(c, BCH_RECOVERY_PASS_check_extents_to_backpointers) ?: ret; + goto out; + case BTREE_ID_need_discard: + ret = bch2_run_explicit_recovery_pass_persistent_locked(c, BCH_RECOVERY_PASS_check_alloc_info) ?: ret; + goto out; + case BTREE_ID_freespace: + ret = bch2_run_explicit_recovery_pass_persistent_locked(c, BCH_RECOVERY_PASS_check_alloc_info) ?: ret; + goto out; + case BTREE_ID_bucket_gens: + ret = bch2_run_explicit_recovery_pass_persistent_locked(c, BCH_RECOVERY_PASS_check_alloc_info) ?: ret; + goto out; + case BTREE_ID_lru: + ret = bch2_run_explicit_recovery_pass_persistent_locked(c, BCH_RECOVERY_PASS_check_alloc_info) ?: ret; + goto out; + case BTREE_ID_accounting: + ret = bch2_run_explicit_recovery_pass_persistent_locked(c, BCH_RECOVERY_PASS_check_allocations) ?: ret; + goto out; + default: + ret = bch2_run_explicit_recovery_pass_persistent_locked(c, BCH_RECOVERY_PASS_scan_for_btree_nodes) ?: ret; + goto out; } +out: + bch2_write_super(c); + mutex_unlock(&c->sb_lock); + + return ret; +} + +static void kill_btree(struct bch_fs *c, enum btree_id btree) +{ + bch2_btree_id_root(c, btree)->alive = false; + bch2_shoot_down_journal_keys(c, btree, 0, BTREE_MAX_DEPTH, POS_MIN, SPOS_MAX); } /* for -o reconstruct_alloc: */ @@ -79,6 +141,8 @@ static void bch2_reconstruct_alloc(struct bch_fs *c) __set_bit_le64(BCH_FSCK_ERR_fs_usage_persistent_reserved_wrong, ext->errors_silent); __set_bit_le64(BCH_FSCK_ERR_fs_usage_replicas_wrong, ext->errors_silent); + __set_bit_le64(BCH_FSCK_ERR_alloc_key_to_missing_lru_entry, ext->errors_silent); + __set_bit_le64(BCH_FSCK_ERR_alloc_key_data_type_wrong, ext->errors_silent); __set_bit_le64(BCH_FSCK_ERR_alloc_key_gen_wrong, ext->errors_silent); __set_bit_le64(BCH_FSCK_ERR_alloc_key_dirty_sectors_wrong, ext->errors_silent); @@ -99,16 +163,9 @@ static void bch2_reconstruct_alloc(struct bch_fs *c) bch2_write_super(c); mutex_unlock(&c->sb_lock); - bch2_shoot_down_journal_keys(c, BTREE_ID_alloc, - 0, BTREE_MAX_DEPTH, POS_MIN, SPOS_MAX); - bch2_shoot_down_journal_keys(c, BTREE_ID_backpointers, - 0, BTREE_MAX_DEPTH, POS_MIN, SPOS_MAX); - bch2_shoot_down_journal_keys(c, BTREE_ID_need_discard, - 0, BTREE_MAX_DEPTH, POS_MIN, SPOS_MAX); - bch2_shoot_down_journal_keys(c, BTREE_ID_freespace, - 0, BTREE_MAX_DEPTH, POS_MIN, SPOS_MAX); - bch2_shoot_down_journal_keys(c, BTREE_ID_bucket_gens, - 0, BTREE_MAX_DEPTH, POS_MIN, SPOS_MAX); + for (unsigned i = 0; i < btree_id_nr_alive(c); i++) + if (btree_id_is_alloc(i)) + kill_btree(c, i); } /* @@ -354,10 +411,13 @@ int bch2_journal_replay(struct bch_fs *c) ? BCH_TRANS_COMMIT_no_journal_res|BCH_WATERMARK_reclaim : 0), bch2_journal_replay_key(trans, k)); - bch_err_msg(c, ret, "while replaying key at btree %s level %u:", - bch2_btree_id_str(k->btree_id), k->level); - if (ret) + if (ret) { + struct printbuf buf = PRINTBUF; + bch2_btree_id_level_to_text(&buf, k->btree_id, k->level); + bch_err_msg(c, ret, "while replaying key at %s:", buf.buf); + printbuf_exit(&buf); goto err; + } BUG_ON(k->btree_id != BTREE_ID_accounting && !k->overwritten); } @@ -403,7 +463,9 @@ static int journal_replay_entry_early(struct bch_fs *c, switch (entry->type) { case BCH_JSET_ENTRY_btree_root: { - struct btree_root *r; + + if (unlikely(!entry->u64s)) + return 0; if (fsck_err_on(entry->btree_id >= BTREE_ID_NR_MAX, c, invalid_btree_id, @@ -417,15 +479,11 @@ static int journal_replay_entry_early(struct bch_fs *c, return ret; } - r = bch2_btree_id_root(c, entry->btree_id); + struct btree_root *r = bch2_btree_id_root(c, entry->btree_id); - if (entry->u64s) { - r->level = entry->level; - bkey_copy(&r->key, (struct bkey_i *) entry->start); - r->error = 0; - } else { - r->error = -BCH_ERR_btree_node_read_error; - } + r->level = entry->level; + bkey_copy(&r->key, (struct bkey_i *) entry->start); + r->error = 0; r->alive = true; break; } @@ -505,6 +563,7 @@ static int journal_replay_early(struct bch_fs *c, static int read_btree_roots(struct bch_fs *c) { + struct printbuf buf = PRINTBUF; int ret = 0; for (unsigned i = 0; i < btree_id_nr_alive(c); i++) { @@ -513,33 +572,22 @@ static int read_btree_roots(struct bch_fs *c) if (!r->alive) continue; - if (btree_id_is_alloc(i) && c->opts.reconstruct_alloc) - continue; + printbuf_reset(&buf); + bch2_btree_id_level_to_text(&buf, i, r->level); if (mustfix_fsck_err_on((ret = r->error), c, btree_root_bkey_invalid, "invalid btree root %s", - bch2_btree_id_str(i)) || + buf.buf) || mustfix_fsck_err_on((ret = r->error = bch2_btree_root_read(c, i, &r->key, r->level)), c, btree_root_read_error, - "error reading btree root %s l=%u: %s", - bch2_btree_id_str(i), r->level, bch2_err_str(ret))) { - if (btree_id_is_alloc(i)) { - c->opts.recovery_passes |= BIT_ULL(BCH_RECOVERY_PASS_check_allocations); - c->opts.recovery_passes |= BIT_ULL(BCH_RECOVERY_PASS_check_alloc_info); - c->opts.recovery_passes |= BIT_ULL(BCH_RECOVERY_PASS_check_lrus); - c->opts.recovery_passes |= BIT_ULL(BCH_RECOVERY_PASS_check_extents_to_backpointers); - c->opts.recovery_passes |= BIT_ULL(BCH_RECOVERY_PASS_check_alloc_to_lru_refs); - c->sb.compat &= ~(1ULL << BCH_COMPAT_alloc_info); + "error reading btree root %s: %s", + buf.buf, bch2_err_str(ret))) { + if (btree_id_is_alloc(i)) r->error = 0; - } else if (!(c->opts.recovery_passes & BIT_ULL(BCH_RECOVERY_PASS_scan_for_btree_nodes))) { - bch_info(c, "will run btree node scan"); - c->opts.recovery_passes |= BIT_ULL(BCH_RECOVERY_PASS_scan_for_btree_nodes); - c->opts.recovery_passes |= BIT_ULL(BCH_RECOVERY_PASS_check_topology); - } - ret = 0; - bch2_btree_lost_data(c, i); + ret = bch2_btree_lost_data(c, i); + BUG_ON(ret); } } @@ -553,6 +601,7 @@ static int read_btree_roots(struct bch_fs *c) } } fsck_err: + printbuf_exit(&buf); return ret; } @@ -563,6 +612,7 @@ static bool check_version_upgrade(struct bch_fs *c) bch2_latest_compatible_version(c->sb.version)); unsigned old_version = c->sb.version_upgrade_complete ?: c->sb.version; unsigned new_version = 0; + bool ret = false; if (old_version < bcachefs_metadata_required_upgrade_below) { if (c->opts.version_upgrade == BCH_VERSION_UPGRADE_incompatible || @@ -618,14 +668,32 @@ static bool check_version_upgrade(struct bch_fs *c) } bch_info(c, "%s", buf.buf); + printbuf_exit(&buf); - bch2_sb_upgrade(c, new_version); + ret = true; + } + if (new_version > c->sb.version_incompat && + c->opts.version_upgrade == BCH_VERSION_UPGRADE_incompatible) { + struct printbuf buf = PRINTBUF; + + prt_str(&buf, "Now allowing incompatible features up to "); + bch2_version_to_text(&buf, new_version); + prt_str(&buf, ", previously allowed up to "); + bch2_version_to_text(&buf, c->sb.version_incompat_allowed); + prt_newline(&buf); + + bch_info(c, "%s", buf.buf); printbuf_exit(&buf); - return true; + + ret = true; } - return false; + if (ret) + bch2_sb_upgrade(c, new_version, + c->opts.version_upgrade == BCH_VERSION_UPGRADE_incompatible); + + return ret; } int bch2_fs_recovery(struct bch_fs *c) @@ -660,8 +728,13 @@ int bch2_fs_recovery(struct bch_fs *c) goto err; } - if (c->opts.norecovery) - c->opts.recovery_pass_last = BCH_RECOVERY_PASS_journal_replay - 1; + if (c->opts.norecovery) { + c->opts.recovery_pass_last = c->opts.recovery_pass_last + ? min(c->opts.recovery_pass_last, BCH_RECOVERY_PASS_snapshots_read) + : BCH_RECOVERY_PASS_snapshots_read; + c->opts.nochanges = true; + c->opts.read_only = true; + } mutex_lock(&c->sb_lock); struct bch_sb_field_ext *ext = bch2_sb_field_get(c->disk_sb.sb, ext); @@ -708,17 +781,20 @@ int bch2_fs_recovery(struct bch_fs *c) c->opts.recovery_passes |= bch2_recovery_passes_from_stable(le64_to_cpu(ext->recovery_passes_required[0])); + if (c->sb.version_upgrade_complete < bcachefs_metadata_version_autofix_errors) { + SET_BCH_SB_ERROR_ACTION(c->disk_sb.sb, BCH_ON_ERROR_fix_safe); + write_sb = true; + } + if (write_sb) bch2_write_super(c); mutex_unlock(&c->sb_lock); - if (c->opts.fsck && IS_ENABLED(CONFIG_BCACHEFS_DEBUG)) - c->opts.recovery_passes |= BIT_ULL(BCH_RECOVERY_PASS_check_topology); - if (c->opts.fsck) set_bit(BCH_FS_fsck_running, &c->flags); if (c->sb.clean) set_bit(BCH_FS_clean_recovery, &c->flags); + set_bit(BCH_FS_recovery_running, &c->flags); ret = bch2_blacklist_table_initialize(c); if (ret) { @@ -807,15 +883,15 @@ use_clean: c->journal_replay_seq_start = last_seq; c->journal_replay_seq_end = blacklist_seq - 1; - if (c->opts.reconstruct_alloc) - bch2_reconstruct_alloc(c); - zero_out_btree_mem_ptr(&c->journal_keys); ret = journal_replay_early(c, clean); if (ret) goto err; + if (c->opts.reconstruct_alloc) + bch2_reconstruct_alloc(c); + /* * After an unclean shutdown, skip then next few journal sequence * numbers as they may have been referenced by btree writes that @@ -870,16 +946,17 @@ use_clean: */ set_bit(BCH_FS_may_go_rw, &c->flags); clear_bit(BCH_FS_fsck_running, &c->flags); + clear_bit(BCH_FS_recovery_running, &c->flags); /* in case we don't run journal replay, i.e. norecovery mode */ set_bit(BCH_FS_accounting_replay_done, &c->flags); + bch2_async_btree_node_rewrites_flush(c); + /* fsync if we fixed errors */ - if (test_bit(BCH_FS_errors_fixed, &c->flags) && - bch2_write_ref_tryget(c, BCH_WRITE_REF_fsync)) { + if (test_bit(BCH_FS_errors_fixed, &c->flags)) { bch2_journal_flush_all_pins(&c->journal); bch2_journal_meta(&c->journal); - bch2_write_ref_put(c, BCH_WRITE_REF_fsync); } /* If we fixed errors, verify that fs is actually clean now: */ @@ -1021,7 +1098,7 @@ int bch2_fs_initialize(struct bch_fs *c) bch2_check_version_downgrade(c); if (c->opts.version_upgrade != BCH_VERSION_UPGRADE_none) { - bch2_sb_upgrade(c, bcachefs_metadata_version_current); + bch2_sb_upgrade(c, bcachefs_metadata_version_current, false); SET_BCH_SB_VERSION_UPGRADE_COMPLETE(c->disk_sb.sb, bcachefs_metadata_version_current); bch2_write_super(c); } @@ -1035,7 +1112,6 @@ int bch2_fs_initialize(struct bch_fs *c) bch2_write_super(c); mutex_unlock(&c->sb_lock); - c->curr_recovery_pass = BCH_RECOVERY_PASS_NR; set_bit(BCH_FS_btree_running, &c->flags); set_bit(BCH_FS_may_go_rw, &c->flags); @@ -1076,9 +1152,6 @@ int bch2_fs_initialize(struct bch_fs *c) if (ret) goto err; - for_each_online_member(c, ca) - ca->new_fs_bucket_idx = 0; - ret = bch2_fs_freespace_init(c); if (ret) goto err; @@ -1137,6 +1210,7 @@ int bch2_fs_initialize(struct bch_fs *c) bch2_write_super(c); mutex_unlock(&c->sb_lock); + c->curr_recovery_pass = BCH_RECOVERY_PASS_NR; return 0; err: bch_err_fn(c, ret); diff --git a/fs/bcachefs/recovery.h b/fs/bcachefs/recovery.h index 4bf818de1f2f..b0d55754b21b 100644 --- a/fs/bcachefs/recovery.h +++ b/fs/bcachefs/recovery.h @@ -2,7 +2,7 @@ #ifndef _BCACHEFS_RECOVERY_H #define _BCACHEFS_RECOVERY_H -void bch2_btree_lost_data(struct bch_fs *, enum btree_id); +int bch2_btree_lost_data(struct bch_fs *, enum btree_id); int bch2_journal_replay(struct bch_fs *); diff --git a/fs/bcachefs/recovery_passes.c b/fs/bcachefs/recovery_passes.c index dff589ddc984..0b3c951c32da 100644 --- a/fs/bcachefs/recovery_passes.c +++ b/fs/bcachefs/recovery_passes.c @@ -46,7 +46,7 @@ static int bch2_set_may_go_rw(struct bch_fs *c) set_bit(BCH_FS_may_go_rw, &c->flags); - if (keys->nr || c->opts.fsck || !c->sb.clean || c->opts.recovery_passes) + if (keys->nr || !c->opts.read_only || c->opts.fsck || !c->sb.clean || c->opts.recovery_passes) return bch2_fs_read_write_early(c); return 0; } @@ -100,20 +100,34 @@ u64 bch2_recovery_passes_from_stable(u64 v) /* * For when we need to rewind recovery passes and run a pass we skipped: */ -int bch2_run_explicit_recovery_pass(struct bch_fs *c, - enum bch_recovery_pass pass) +static int __bch2_run_explicit_recovery_pass(struct bch_fs *c, + enum bch_recovery_pass pass) { - if (c->opts.recovery_passes & BIT_ULL(pass)) + if (c->curr_recovery_pass == ARRAY_SIZE(recovery_pass_fns)) + return -BCH_ERR_not_in_recovery; + + if (c->recovery_passes_complete & BIT_ULL(pass)) return 0; - bch_info(c, "running explicit recovery pass %s (%u), currently at %s (%u)", - bch2_recovery_passes[pass], pass, - bch2_recovery_passes[c->curr_recovery_pass], c->curr_recovery_pass); + bool print = !(c->opts.recovery_passes & BIT_ULL(pass)); + + if (pass < BCH_RECOVERY_PASS_set_may_go_rw && + c->curr_recovery_pass >= BCH_RECOVERY_PASS_set_may_go_rw) { + if (print) + bch_info(c, "need recovery pass %s (%u), but already rw", + bch2_recovery_passes[pass], pass); + return -BCH_ERR_cannot_rewind_recovery; + } + + if (print) + bch_info(c, "running explicit recovery pass %s (%u), currently at %s (%u)", + bch2_recovery_passes[pass], pass, + bch2_recovery_passes[c->curr_recovery_pass], c->curr_recovery_pass); c->opts.recovery_passes |= BIT_ULL(pass); - if (c->curr_recovery_pass >= pass) { - c->curr_recovery_pass = pass; + if (c->curr_recovery_pass > pass) { + c->next_recovery_pass = pass; c->recovery_passes_complete &= (1ULL << pass) >> 1; return -BCH_ERR_restart_recovery; } else { @@ -121,6 +135,27 @@ int bch2_run_explicit_recovery_pass(struct bch_fs *c, } } +int bch2_run_explicit_recovery_pass(struct bch_fs *c, + enum bch_recovery_pass pass) +{ + unsigned long flags; + spin_lock_irqsave(&c->recovery_pass_lock, flags); + int ret = __bch2_run_explicit_recovery_pass(c, pass); + spin_unlock_irqrestore(&c->recovery_pass_lock, flags); + return ret; +} + +int bch2_run_explicit_recovery_pass_persistent_locked(struct bch_fs *c, + enum bch_recovery_pass pass) +{ + lockdep_assert_held(&c->sb_lock); + + struct bch_sb_field_ext *ext = bch2_sb_field_get(c->disk_sb.sb, ext); + __set_bit_le64(bch2_recovery_pass_to_stable(pass), ext->recovery_passes_required); + + return bch2_run_explicit_recovery_pass(c, pass); +} + int bch2_run_explicit_recovery_pass_persistent(struct bch_fs *c, enum bch_recovery_pass pass) { @@ -233,31 +268,48 @@ int bch2_run_recovery_passes(struct bch_fs *c) */ c->opts.recovery_passes_exclude &= ~BCH_RECOVERY_PASS_set_may_go_rw; - while (c->curr_recovery_pass < ARRAY_SIZE(recovery_pass_fns)) { - if (c->opts.recovery_pass_last && - c->curr_recovery_pass > c->opts.recovery_pass_last) - break; - - if (should_run_recovery_pass(c, c->curr_recovery_pass)) { - unsigned pass = c->curr_recovery_pass; + while (c->curr_recovery_pass < ARRAY_SIZE(recovery_pass_fns) && !ret) { + c->next_recovery_pass = c->curr_recovery_pass + 1; - ret = bch2_run_recovery_pass(c, c->curr_recovery_pass) ?: - bch2_journal_flush(&c->journal); - if (bch2_err_matches(ret, BCH_ERR_restart_recovery) || - (ret && c->curr_recovery_pass < pass)) - continue; - if (ret) - break; + spin_lock_irq(&c->recovery_pass_lock); + unsigned pass = c->curr_recovery_pass; - c->recovery_passes_complete |= BIT_ULL(c->curr_recovery_pass); + if (c->opts.recovery_pass_last && + c->curr_recovery_pass > c->opts.recovery_pass_last) { + spin_unlock_irq(&c->recovery_pass_lock); + break; } - c->recovery_pass_done = max(c->recovery_pass_done, c->curr_recovery_pass); - - if (!test_bit(BCH_FS_error, &c->flags)) - bch2_clear_recovery_pass_required(c, c->curr_recovery_pass); - - c->curr_recovery_pass++; + if (!should_run_recovery_pass(c, pass)) { + c->curr_recovery_pass++; + c->recovery_pass_done = max(c->recovery_pass_done, pass); + spin_unlock_irq(&c->recovery_pass_lock); + continue; + } + spin_unlock_irq(&c->recovery_pass_lock); + + ret = bch2_run_recovery_pass(c, pass) ?: + bch2_journal_flush(&c->journal); + + if (!ret && !test_bit(BCH_FS_error, &c->flags)) + bch2_clear_recovery_pass_required(c, pass); + + spin_lock_irq(&c->recovery_pass_lock); + if (c->next_recovery_pass < c->curr_recovery_pass) { + /* + * bch2_run_explicit_recovery_pass() was called: we + * can't always catch -BCH_ERR_restart_recovery because + * it may have been called from another thread (btree + * node read completion) + */ + ret = 0; + c->recovery_passes_complete &= ~(~0ULL << c->curr_recovery_pass); + } else { + c->recovery_passes_complete |= BIT_ULL(pass); + c->recovery_pass_done = max(c->recovery_pass_done, pass); + } + c->curr_recovery_pass = c->next_recovery_pass; + spin_unlock_irq(&c->recovery_pass_lock); } return ret; diff --git a/fs/bcachefs/recovery_passes.h b/fs/bcachefs/recovery_passes.h index 99b464e127b8..7d7339c8fa29 100644 --- a/fs/bcachefs/recovery_passes.h +++ b/fs/bcachefs/recovery_passes.h @@ -9,6 +9,7 @@ u64 bch2_recovery_passes_from_stable(u64 v); u64 bch2_fsck_recovery_passes(void); int bch2_run_explicit_recovery_pass(struct bch_fs *, enum bch_recovery_pass); +int bch2_run_explicit_recovery_pass_persistent_locked(struct bch_fs *, enum bch_recovery_pass); int bch2_run_explicit_recovery_pass_persistent(struct bch_fs *, enum bch_recovery_pass); int bch2_run_online_recovery_passes(struct bch_fs *); diff --git a/fs/bcachefs/recovery_passes_types.h b/fs/bcachefs/recovery_passes_types.h index 94dc20ca2065..418557960ed6 100644 --- a/fs/bcachefs/recovery_passes_types.h +++ b/fs/bcachefs/recovery_passes_types.h @@ -8,53 +8,59 @@ #define PASS_ALWAYS BIT(3) #define PASS_ONLINE BIT(4) +#ifdef CONFIG_BCACHEFS_DEBUG +#define PASS_FSCK_DEBUG BIT(1) +#else +#define PASS_FSCK_DEBUG 0 +#endif + /* * Passes may be reordered, but the second field is a persistent identifier and * must never change: */ -#define BCH_RECOVERY_PASSES() \ - x(recovery_pass_empty, 41, PASS_SILENT) \ - x(scan_for_btree_nodes, 37, 0) \ - x(check_topology, 4, 0) \ - x(accounting_read, 39, PASS_ALWAYS) \ - x(alloc_read, 0, PASS_ALWAYS) \ - x(stripes_read, 1, PASS_ALWAYS) \ - x(initialize_subvolumes, 2, 0) \ - x(snapshots_read, 3, PASS_ALWAYS) \ - x(check_allocations, 5, PASS_FSCK) \ - x(trans_mark_dev_sbs, 6, PASS_ALWAYS|PASS_SILENT) \ - x(fs_journal_alloc, 7, PASS_ALWAYS|PASS_SILENT) \ - x(set_may_go_rw, 8, PASS_ALWAYS|PASS_SILENT) \ - x(journal_replay, 9, PASS_ALWAYS) \ - x(check_alloc_info, 10, PASS_ONLINE|PASS_FSCK) \ - x(check_lrus, 11, PASS_ONLINE|PASS_FSCK) \ - x(check_btree_backpointers, 12, PASS_ONLINE|PASS_FSCK) \ - x(check_backpointers_to_extents, 13, PASS_ONLINE|PASS_FSCK) \ - x(check_extents_to_backpointers, 14, PASS_ONLINE|PASS_FSCK) \ - x(check_alloc_to_lru_refs, 15, PASS_ONLINE|PASS_FSCK) \ - x(fs_freespace_init, 16, PASS_ALWAYS|PASS_SILENT) \ - x(bucket_gens_init, 17, 0) \ - x(reconstruct_snapshots, 38, 0) \ - x(check_snapshot_trees, 18, PASS_ONLINE|PASS_FSCK) \ - x(check_snapshots, 19, PASS_ONLINE|PASS_FSCK) \ - x(check_subvols, 20, PASS_ONLINE|PASS_FSCK) \ - x(check_subvol_children, 35, PASS_ONLINE|PASS_FSCK) \ - x(delete_dead_snapshots, 21, PASS_ONLINE|PASS_FSCK) \ - x(fs_upgrade_for_subvolumes, 22, 0) \ - x(check_inodes, 24, PASS_FSCK) \ - x(check_extents, 25, PASS_FSCK) \ - x(check_indirect_extents, 26, PASS_FSCK) \ - x(check_dirents, 27, PASS_FSCK) \ - x(check_xattrs, 28, PASS_FSCK) \ - x(check_root, 29, PASS_ONLINE|PASS_FSCK) \ - x(check_unreachable_inodes, 40, PASS_ONLINE|PASS_FSCK) \ - x(check_subvolume_structure, 36, PASS_ONLINE|PASS_FSCK) \ - x(check_directory_structure, 30, PASS_ONLINE|PASS_FSCK) \ - x(check_nlinks, 31, PASS_FSCK) \ - x(resume_logged_ops, 23, PASS_ALWAYS) \ - x(delete_dead_inodes, 32, PASS_ALWAYS) \ - x(fix_reflink_p, 33, 0) \ - x(set_fs_needs_rebalance, 34, 0) \ +#define BCH_RECOVERY_PASSES() \ + x(recovery_pass_empty, 41, PASS_SILENT) \ + x(scan_for_btree_nodes, 37, 0) \ + x(check_topology, 4, 0) \ + x(accounting_read, 39, PASS_ALWAYS) \ + x(alloc_read, 0, PASS_ALWAYS) \ + x(stripes_read, 1, PASS_ALWAYS) \ + x(initialize_subvolumes, 2, 0) \ + x(snapshots_read, 3, PASS_ALWAYS) \ + x(check_allocations, 5, PASS_FSCK) \ + x(trans_mark_dev_sbs, 6, PASS_ALWAYS|PASS_SILENT) \ + x(fs_journal_alloc, 7, PASS_ALWAYS|PASS_SILENT) \ + x(set_may_go_rw, 8, PASS_ALWAYS|PASS_SILENT) \ + x(journal_replay, 9, PASS_ALWAYS) \ + x(check_alloc_info, 10, PASS_ONLINE|PASS_FSCK) \ + x(check_lrus, 11, PASS_ONLINE|PASS_FSCK) \ + x(check_btree_backpointers, 12, PASS_ONLINE|PASS_FSCK) \ + x(check_backpointers_to_extents, 13, PASS_ONLINE|PASS_FSCK_DEBUG) \ + x(check_extents_to_backpointers, 14, PASS_ONLINE|PASS_FSCK) \ + x(check_alloc_to_lru_refs, 15, PASS_ONLINE|PASS_FSCK) \ + x(fs_freespace_init, 16, PASS_ALWAYS|PASS_SILENT) \ + x(bucket_gens_init, 17, 0) \ + x(reconstruct_snapshots, 38, 0) \ + x(check_snapshot_trees, 18, PASS_ONLINE|PASS_FSCK) \ + x(check_snapshots, 19, PASS_ONLINE|PASS_FSCK) \ + x(check_subvols, 20, PASS_ONLINE|PASS_FSCK) \ + x(check_subvol_children, 35, PASS_ONLINE|PASS_FSCK) \ + x(delete_dead_snapshots, 21, PASS_ONLINE|PASS_FSCK) \ + x(fs_upgrade_for_subvolumes, 22, 0) \ + x(check_inodes, 24, PASS_FSCK) \ + x(check_extents, 25, PASS_FSCK) \ + x(check_indirect_extents, 26, PASS_ONLINE|PASS_FSCK) \ + x(check_dirents, 27, PASS_FSCK) \ + x(check_xattrs, 28, PASS_FSCK) \ + x(check_root, 29, PASS_ONLINE|PASS_FSCK) \ + x(check_unreachable_inodes, 40, PASS_FSCK) \ + x(check_subvolume_structure, 36, PASS_ONLINE|PASS_FSCK) \ + x(check_directory_structure, 30, PASS_ONLINE|PASS_FSCK) \ + x(check_nlinks, 31, PASS_FSCK) \ + x(resume_logged_ops, 23, PASS_ALWAYS) \ + x(delete_dead_inodes, 32, PASS_ALWAYS) \ + x(fix_reflink_p, 33, 0) \ + x(set_fs_needs_rebalance, 34, 0) /* We normally enumerate recovery passes in the order we run them: */ enum bch_recovery_pass { diff --git a/fs/bcachefs/reflink.c b/fs/bcachefs/reflink.c index f457925fa362..93ba4f4e47ca 100644 --- a/fs/bcachefs/reflink.c +++ b/fs/bcachefs/reflink.c @@ -15,6 +15,17 @@ #include <linux/sched/signal.h> +static inline bool bkey_extent_is_reflink_data(const struct bkey *k) +{ + switch (k->type) { + case KEY_TYPE_reflink_v: + case KEY_TYPE_indirect_inline_data: + return true; + default: + return false; + } +} + static inline unsigned bkey_type_to_indirect(const struct bkey *k) { switch (k->type) { @@ -30,15 +41,15 @@ static inline unsigned bkey_type_to_indirect(const struct bkey *k) /* reflink pointers */ int bch2_reflink_p_validate(struct bch_fs *c, struct bkey_s_c k, - enum bch_validate_flags flags) + struct bkey_validate_context from) { struct bkey_s_c_reflink_p p = bkey_s_c_to_reflink_p(k); int ret = 0; - bkey_fsck_err_on(le64_to_cpu(p.v->idx) < le32_to_cpu(p.v->front_pad), + bkey_fsck_err_on(REFLINK_P_IDX(p.v) < le32_to_cpu(p.v->front_pad), c, reflink_p_front_pad_bad, "idx < front_pad (%llu < %u)", - le64_to_cpu(p.v->idx), le32_to_cpu(p.v->front_pad)); + REFLINK_P_IDX(p.v), le32_to_cpu(p.v->front_pad)); fsck_err: return ret; } @@ -49,7 +60,7 @@ void bch2_reflink_p_to_text(struct printbuf *out, struct bch_fs *c, struct bkey_s_c_reflink_p p = bkey_s_c_to_reflink_p(k); prt_printf(out, "idx %llu front_pad %u back_pad %u", - le64_to_cpu(p.v->idx), + REFLINK_P_IDX(p.v), le32_to_cpu(p.v->front_pad), le32_to_cpu(p.v->back_pad)); } @@ -65,49 +76,250 @@ bool bch2_reflink_p_merge(struct bch_fs *c, struct bkey_s _l, struct bkey_s_c _r */ return false; - if (le64_to_cpu(l.v->idx) + l.k->size != le64_to_cpu(r.v->idx)) + if (REFLINK_P_IDX(l.v) + l.k->size != REFLINK_P_IDX(r.v)) + return false; + + if (REFLINK_P_ERROR(l.v) != REFLINK_P_ERROR(r.v)) return false; bch2_key_resize(l.k, l.k->size + r.k->size); return true; } +/* indirect extents */ + +int bch2_reflink_v_validate(struct bch_fs *c, struct bkey_s_c k, + struct bkey_validate_context from) +{ + int ret = 0; + + bkey_fsck_err_on(bkey_gt(k.k->p, POS(0, REFLINK_P_IDX_MAX)), + c, reflink_v_pos_bad, + "indirect extent above maximum position 0:%llu", + REFLINK_P_IDX_MAX); + + ret = bch2_bkey_ptrs_validate(c, k, from); +fsck_err: + return ret; +} + +void bch2_reflink_v_to_text(struct printbuf *out, struct bch_fs *c, + struct bkey_s_c k) +{ + struct bkey_s_c_reflink_v r = bkey_s_c_to_reflink_v(k); + + prt_printf(out, "refcount: %llu ", le64_to_cpu(r.v->refcount)); + + bch2_bkey_ptrs_to_text(out, c, k); +} + +#if 0 +Currently disabled, needs to be debugged: + +bool bch2_reflink_v_merge(struct bch_fs *c, struct bkey_s _l, struct bkey_s_c _r) +{ + struct bkey_s_reflink_v l = bkey_s_to_reflink_v(_l); + struct bkey_s_c_reflink_v r = bkey_s_c_to_reflink_v(_r); + + return l.v->refcount == r.v->refcount && bch2_extent_merge(c, _l, _r); +} +#endif + +/* indirect inline data */ + +int bch2_indirect_inline_data_validate(struct bch_fs *c, struct bkey_s_c k, + struct bkey_validate_context from) +{ + return 0; +} + +void bch2_indirect_inline_data_to_text(struct printbuf *out, + struct bch_fs *c, struct bkey_s_c k) +{ + struct bkey_s_c_indirect_inline_data d = bkey_s_c_to_indirect_inline_data(k); + unsigned datalen = bkey_inline_data_bytes(k.k); + + prt_printf(out, "refcount %llu datalen %u: %*phN", + le64_to_cpu(d.v->refcount), datalen, + min(datalen, 32U), d.v->data); +} + +/* lookup */ + +static int bch2_indirect_extent_not_missing(struct btree_trans *trans, struct bkey_s_c_reflink_p p, + bool should_commit) +{ + struct bkey_i_reflink_p *new = bch2_bkey_make_mut_noupdate_typed(trans, p.s_c, reflink_p); + int ret = PTR_ERR_OR_ZERO(new); + if (ret) + return ret; + + SET_REFLINK_P_ERROR(&new->v, false); + ret = bch2_btree_insert_trans(trans, BTREE_ID_extents, &new->k_i, BTREE_TRIGGER_norun); + if (ret) + return ret; + + if (!should_commit) + return 0; + + return bch2_trans_commit(trans, NULL, NULL, BCH_TRANS_COMMIT_no_enospc) ?: + -BCH_ERR_transaction_restart_nested; +} + +static int bch2_indirect_extent_missing_error(struct btree_trans *trans, + struct bkey_s_c_reflink_p p, + u64 missing_start, u64 missing_end, + bool should_commit) +{ + if (REFLINK_P_ERROR(p.v)) + return -BCH_ERR_missing_indirect_extent; + + struct bch_fs *c = trans->c; + u64 live_start = REFLINK_P_IDX(p.v); + u64 live_end = REFLINK_P_IDX(p.v) + p.k->size; + u64 refd_start = live_start - le32_to_cpu(p.v->front_pad); + u64 refd_end = live_end + le32_to_cpu(p.v->back_pad); + struct printbuf buf = PRINTBUF; + int ret = 0; + + BUG_ON(missing_start < refd_start); + BUG_ON(missing_end > refd_end); + + if (fsck_err(trans, reflink_p_to_missing_reflink_v, + "pointer to missing indirect extent\n" + " %s\n" + " missing range %llu-%llu", + (bch2_bkey_val_to_text(&buf, c, p.s_c), buf.buf), + missing_start, missing_end)) { + struct bkey_i_reflink_p *new = bch2_bkey_make_mut_noupdate_typed(trans, p.s_c, reflink_p); + ret = PTR_ERR_OR_ZERO(new); + if (ret) + goto err; + + /* + * Is the missing range not actually needed? + * + * p.v->idx refers to the data that we actually want, but if the + * indirect extent we point to was bigger, front_pad and back_pad + * indicate the range we took a reference on. + */ + + if (missing_end <= live_start) { + new->v.front_pad = cpu_to_le32(live_start - missing_end); + } else if (missing_start >= live_end) { + new->v.back_pad = cpu_to_le32(missing_start - live_end); + } else { + struct bpos new_start = bkey_start_pos(&new->k); + struct bpos new_end = new->k.p; + + if (missing_start > live_start) + new_start.offset += missing_start - live_start; + if (missing_end < live_end) + new_end.offset -= live_end - missing_end; + + bch2_cut_front(new_start, &new->k_i); + bch2_cut_back(new_end, &new->k_i); + + SET_REFLINK_P_ERROR(&new->v, true); + } + + ret = bch2_btree_insert_trans(trans, BTREE_ID_extents, &new->k_i, BTREE_TRIGGER_norun); + if (ret) + goto err; + + if (should_commit) + ret = bch2_trans_commit(trans, NULL, NULL, BCH_TRANS_COMMIT_no_enospc) ?: + -BCH_ERR_transaction_restart_nested; + } +err: +fsck_err: + printbuf_exit(&buf); + return ret; +} + +/* + * This is used from the read path, which doesn't expect to have to do a + * transaction commit, and from triggers, which should not be doing a commit: + */ +struct bkey_s_c bch2_lookup_indirect_extent(struct btree_trans *trans, + struct btree_iter *iter, + s64 *offset_into_extent, + struct bkey_s_c_reflink_p p, + bool should_commit, + unsigned iter_flags) +{ + BUG_ON(*offset_into_extent < -((s64) le32_to_cpu(p.v->front_pad))); + BUG_ON(*offset_into_extent >= p.k->size + le32_to_cpu(p.v->back_pad)); + + u64 reflink_offset = REFLINK_P_IDX(p.v) + *offset_into_extent; + + struct bkey_s_c k = bch2_bkey_get_iter(trans, iter, BTREE_ID_reflink, + POS(0, reflink_offset), iter_flags); + if (bkey_err(k)) + return k; + + if (unlikely(!bkey_extent_is_reflink_data(k.k))) { + bch2_trans_iter_exit(trans, iter); + + unsigned size = min((u64) k.k->size, + REFLINK_P_IDX(p.v) + p.k->size + le32_to_cpu(p.v->back_pad) - + reflink_offset); + bch2_key_resize(&iter->k, size); + + int ret = bch2_indirect_extent_missing_error(trans, p, reflink_offset, + k.k->p.offset, should_commit); + if (ret) + return bkey_s_c_err(ret); + } else if (unlikely(REFLINK_P_ERROR(p.v))) { + bch2_trans_iter_exit(trans, iter); + + int ret = bch2_indirect_extent_not_missing(trans, p, should_commit); + if (ret) + return bkey_s_c_err(ret); + } + + *offset_into_extent = reflink_offset - bkey_start_offset(k.k); + return k; +} + +/* reflink pointer trigger */ + static int trans_trigger_reflink_p_segment(struct btree_trans *trans, struct bkey_s_c_reflink_p p, u64 *idx, enum btree_iter_update_trigger_flags flags) { struct bch_fs *c = trans->c; - struct btree_iter iter; - struct bkey_i *k; - __le64 *refcount; - int add = !(flags & BTREE_TRIGGER_overwrite) ? 1 : -1; struct printbuf buf = PRINTBUF; - int ret; - k = bch2_bkey_get_mut_noupdate(trans, &iter, - BTREE_ID_reflink, POS(0, *idx), - BTREE_ITER_with_updates); - ret = PTR_ERR_OR_ZERO(k); + s64 offset_into_extent = *idx - REFLINK_P_IDX(p.v); + struct btree_iter iter; + struct bkey_s_c k = bch2_lookup_indirect_extent(trans, &iter, &offset_into_extent, p, false, + BTREE_ITER_intent| + BTREE_ITER_with_updates); + int ret = bkey_err(k); if (ret) - goto err; + return ret; - refcount = bkey_refcount(bkey_i_to_s(k)); - if (!refcount) { - bch2_bkey_val_to_text(&buf, c, p.s_c); - bch2_trans_inconsistent(trans, - "nonexistent indirect extent at %llu while marking\n %s", - *idx, buf.buf); - ret = -EIO; - goto err; + if (bkey_deleted(k.k)) { + if (!(flags & BTREE_TRIGGER_overwrite)) + ret = -BCH_ERR_missing_indirect_extent; + goto next; } + struct bkey_i *new = bch2_bkey_make_mut_noupdate(trans, k); + ret = PTR_ERR_OR_ZERO(new); + if (ret) + goto err; + + __le64 *refcount = bkey_refcount(bkey_i_to_s(new)); if (!*refcount && (flags & BTREE_TRIGGER_overwrite)) { bch2_bkey_val_to_text(&buf, c, p.s_c); - bch2_trans_inconsistent(trans, - "indirect extent refcount underflow at %llu while marking\n %s", - *idx, buf.buf); - ret = -EIO; - goto err; + prt_printf(&buf, "\n "); + bch2_bkey_val_to_text(&buf, c, k); + log_fsck_err(trans, reflink_refcount_underflow, + "indirect extent refcount underflow while marking\n %s", + buf.buf); + goto next; } if (flags & BTREE_TRIGGER_insert) { @@ -115,25 +327,26 @@ static int trans_trigger_reflink_p_segment(struct btree_trans *trans, u64 pad; pad = max_t(s64, le32_to_cpu(v->front_pad), - le64_to_cpu(v->idx) - bkey_start_offset(&k->k)); + REFLINK_P_IDX(v) - bkey_start_offset(&new->k)); BUG_ON(pad > U32_MAX); v->front_pad = cpu_to_le32(pad); pad = max_t(s64, le32_to_cpu(v->back_pad), - k->k.p.offset - p.k->size - le64_to_cpu(v->idx)); + new->k.p.offset - p.k->size - REFLINK_P_IDX(v)); BUG_ON(pad > U32_MAX); v->back_pad = cpu_to_le32(pad); } - le64_add_cpu(refcount, add); + le64_add_cpu(refcount, !(flags & BTREE_TRIGGER_overwrite) ? 1 : -1); bch2_btree_iter_set_pos_to_extent_start(&iter); - ret = bch2_trans_update(trans, &iter, k, 0); + ret = bch2_trans_update(trans, &iter, new, 0); if (ret) goto err; - - *idx = k->k.p.offset; +next: + *idx = k.k->p.offset; err: +fsck_err: bch2_trans_iter_exit(trans, &iter); printbuf_exit(&buf); return ret; @@ -147,9 +360,7 @@ static s64 gc_trigger_reflink_p_segment(struct btree_trans *trans, struct bch_fs *c = trans->c; struct reflink_gc *r; int add = !(flags & BTREE_TRIGGER_overwrite) ? 1 : -1; - u64 start = le64_to_cpu(p.v->idx); - u64 end = le64_to_cpu(p.v->idx) + p.k->size; - u64 next_idx = end + le32_to_cpu(p.v->back_pad); + u64 next_idx = REFLINK_P_IDX(p.v) + p.k->size + le32_to_cpu(p.v->back_pad); s64 ret = 0; struct printbuf buf = PRINTBUF; @@ -168,36 +379,14 @@ static s64 gc_trigger_reflink_p_segment(struct btree_trans *trans, *idx = r->offset; return 0; not_found: - BUG_ON(!(flags & BTREE_TRIGGER_check_repair)); - - if (fsck_err(trans, reflink_p_to_missing_reflink_v, - "pointer to missing indirect extent\n" - " %s\n" - " missing range %llu-%llu", - (bch2_bkey_val_to_text(&buf, c, p.s_c), buf.buf), - *idx, next_idx)) { - struct bkey_i *update = bch2_bkey_make_mut_noupdate(trans, p.s_c); - ret = PTR_ERR_OR_ZERO(update); + if (flags & BTREE_TRIGGER_check_repair) { + ret = bch2_indirect_extent_missing_error(trans, p, *idx, next_idx, false); if (ret) goto err; - - if (next_idx <= start) { - bkey_i_to_reflink_p(update)->v.front_pad = cpu_to_le32(start - next_idx); - } else if (*idx >= end) { - bkey_i_to_reflink_p(update)->v.back_pad = cpu_to_le32(*idx - end); - } else { - bkey_error_init(update); - update->k.p = p.k->p; - update->k.size = p.k->size; - set_bkey_val_u64s(&update->k, 0); - } - - ret = bch2_btree_insert_trans(trans, BTREE_ID_extents, update, BTREE_TRIGGER_norun); } *idx = next_idx; err: -fsck_err: printbuf_exit(&buf); return ret; } @@ -210,8 +399,8 @@ static int __trigger_reflink_p(struct btree_trans *trans, struct bkey_s_c_reflink_p p = bkey_s_c_to_reflink_p(k); int ret = 0; - u64 idx = le64_to_cpu(p.v->idx) - le32_to_cpu(p.v->front_pad); - u64 end = le64_to_cpu(p.v->idx) + p.k->size + le32_to_cpu(p.v->back_pad); + u64 idx = REFLINK_P_IDX(p.v) - le32_to_cpu(p.v->front_pad); + u64 end = REFLINK_P_IDX(p.v) + p.k->size + le32_to_cpu(p.v->back_pad); if (flags & BTREE_TRIGGER_transactional) { while (idx < end && !ret) @@ -253,35 +442,7 @@ int bch2_trigger_reflink_p(struct btree_trans *trans, return trigger_run_overwrite_then_insert(__trigger_reflink_p, trans, btree_id, level, old, new, flags); } -/* indirect extents */ - -int bch2_reflink_v_validate(struct bch_fs *c, struct bkey_s_c k, - enum bch_validate_flags flags) -{ - return bch2_bkey_ptrs_validate(c, k, flags); -} - -void bch2_reflink_v_to_text(struct printbuf *out, struct bch_fs *c, - struct bkey_s_c k) -{ - struct bkey_s_c_reflink_v r = bkey_s_c_to_reflink_v(k); - - prt_printf(out, "refcount: %llu ", le64_to_cpu(r.v->refcount)); - - bch2_bkey_ptrs_to_text(out, c, k); -} - -#if 0 -Currently disabled, needs to be debugged: - -bool bch2_reflink_v_merge(struct bch_fs *c, struct bkey_s _l, struct bkey_s_c _r) -{ - struct bkey_s_reflink_v l = bkey_s_to_reflink_v(_l); - struct bkey_s_c_reflink_v r = bkey_s_c_to_reflink_v(_r); - - return l.v->refcount == r.v->refcount && bch2_extent_merge(c, _l, _r); -} -#endif +/* indirect extent trigger */ static inline void check_indirect_extent_deleting(struct bkey_s new, @@ -307,25 +468,6 @@ int bch2_trigger_reflink_v(struct btree_trans *trans, return bch2_trigger_extent(trans, btree_id, level, old, new, flags); } -/* indirect inline data */ - -int bch2_indirect_inline_data_validate(struct bch_fs *c, struct bkey_s_c k, - enum bch_validate_flags flags) -{ - return 0; -} - -void bch2_indirect_inline_data_to_text(struct printbuf *out, - struct bch_fs *c, struct bkey_s_c k) -{ - struct bkey_s_c_indirect_inline_data d = bkey_s_c_to_indirect_inline_data(k); - unsigned datalen = bkey_inline_data_bytes(k.k); - - prt_printf(out, "refcount %llu datalen %u: %*phN", - le64_to_cpu(d.v->refcount), datalen, - min(datalen, 32U), d.v->data); -} - int bch2_trigger_indirect_inline_data(struct btree_trans *trans, enum btree_id btree_id, unsigned level, struct bkey_s_c old, struct bkey_s new, @@ -336,9 +478,12 @@ int bch2_trigger_indirect_inline_data(struct btree_trans *trans, return 0; } +/* create */ + static int bch2_make_extent_indirect(struct btree_trans *trans, struct btree_iter *extent_iter, - struct bkey_i *orig) + struct bkey_i *orig, + bool reflink_p_may_update_opts_field) { struct bch_fs *c = trans->c; struct btree_iter reflink_iter = { NULL }; @@ -358,6 +503,14 @@ static int bch2_make_extent_indirect(struct btree_trans *trans, if (ret) goto err; + /* + * XXX: we're assuming that 56 bits will be enough for the life of the + * filesystem: we need to implement wraparound, with a cursor in the + * logged ops btree: + */ + if (bkey_ge(reflink_iter.pos, POS(0, REFLINK_P_IDX_MAX - orig->k.size))) + return -ENOSPC; + r_v = bch2_trans_kmalloc(trans, sizeof(__le64) + bkey_bytes(&orig->k)); ret = PTR_ERR_OR_ZERO(r_v); if (ret) @@ -394,7 +547,10 @@ static int bch2_make_extent_indirect(struct btree_trans *trans, memset(&r_p->v, 0, sizeof(r_p->v)); #endif - r_p->v.idx = cpu_to_le64(bkey_start_offset(&r_v->k)); + SET_REFLINK_P_IDX(&r_p->v, bkey_start_offset(&r_v->k)); + + if (reflink_p_may_update_opts_field) + SET_REFLINK_P_MAY_UPDATE_OPTIONS(&r_p->v, true); ret = bch2_trans_update(trans, extent_iter, &r_p->k_i, BTREE_UPDATE_internal_snapshot_node); @@ -409,7 +565,7 @@ static struct bkey_s_c get_next_src(struct btree_iter *iter, struct bpos end) struct bkey_s_c k; int ret; - for_each_btree_key_upto_continue_norestart(*iter, end, 0, k, ret) { + for_each_btree_key_max_continue_norestart(*iter, end, 0, k, ret) { if (bkey_extent_is_unwritten(k)) continue; @@ -426,7 +582,8 @@ s64 bch2_remap_range(struct bch_fs *c, subvol_inum dst_inum, u64 dst_offset, subvol_inum src_inum, u64 src_offset, u64 remap_sectors, - u64 new_i_size, s64 *i_sectors_delta) + u64 new_i_size, s64 *i_sectors_delta, + bool may_change_src_io_path_opts) { struct btree_trans *trans; struct btree_iter dst_iter, src_iter; @@ -439,6 +596,8 @@ s64 bch2_remap_range(struct bch_fs *c, struct bpos src_want; u64 dst_done = 0; u32 dst_snapshot, src_snapshot; + bool reflink_p_may_update_opts_field = + bch2_request_incompat_feature(c, bcachefs_metadata_version_reflink_p_may_update_opts); int ret = 0, ret2 = 0; if (!bch2_write_ref_tryget(c, BCH_WRITE_REF_reflink)) @@ -520,7 +679,8 @@ s64 bch2_remap_range(struct bch_fs *c, src_k = bkey_i_to_s_c(new_src.k); ret = bch2_make_extent_indirect(trans, &src_iter, - new_src.k); + new_src.k, + reflink_p_may_update_opts_field); if (ret) continue; @@ -533,11 +693,15 @@ s64 bch2_remap_range(struct bch_fs *c, struct bkey_i_reflink_p *dst_p = bkey_reflink_p_init(new_dst.k); - u64 offset = le64_to_cpu(src_p.v->idx) + + u64 offset = REFLINK_P_IDX(src_p.v) + (src_want.offset - bkey_start_offset(src_k.k)); - dst_p->v.idx = cpu_to_le64(offset); + SET_REFLINK_P_IDX(&dst_p->v, offset); + + if (reflink_p_may_update_opts_field && + may_change_src_io_path_opts) + SET_REFLINK_P_MAY_UPDATE_OPTIONS(&dst_p->v, true); } else { BUG(); } @@ -547,7 +711,7 @@ s64 bch2_remap_range(struct bch_fs *c, min(src_k.k->p.offset - src_want.offset, dst_end.offset - dst_iter.pos.offset)); - ret = bch2_bkey_set_needs_rebalance(c, new_dst.k, &opts) ?: + ret = bch2_bkey_set_needs_rebalance(c, &opts, new_dst.k) ?: bch2_extent_update(trans, dst_inum, &dst_iter, new_dst.k, &disk_res, new_i_size, i_sectors_delta, @@ -591,3 +755,97 @@ err: return dst_done ?: ret ?: ret2; } + +/* fsck */ + +static int bch2_gc_write_reflink_key(struct btree_trans *trans, + struct btree_iter *iter, + struct bkey_s_c k, + size_t *idx) +{ + struct bch_fs *c = trans->c; + const __le64 *refcount = bkey_refcount_c(k); + struct printbuf buf = PRINTBUF; + struct reflink_gc *r; + int ret = 0; + + if (!refcount) + return 0; + + while ((r = genradix_ptr(&c->reflink_gc_table, *idx)) && + r->offset < k.k->p.offset) + ++*idx; + + if (!r || + r->offset != k.k->p.offset || + r->size != k.k->size) { + bch_err(c, "unexpected inconsistency walking reflink table at gc finish"); + return -EINVAL; + } + + if (fsck_err_on(r->refcount != le64_to_cpu(*refcount), + trans, reflink_v_refcount_wrong, + "reflink key has wrong refcount:\n" + " %s\n" + " should be %u", + (bch2_bkey_val_to_text(&buf, c, k), buf.buf), + r->refcount)) { + struct bkey_i *new = bch2_bkey_make_mut_noupdate(trans, k); + ret = PTR_ERR_OR_ZERO(new); + if (ret) + goto out; + + if (!r->refcount) + new->k.type = KEY_TYPE_deleted; + else + *bkey_refcount(bkey_i_to_s(new)) = cpu_to_le64(r->refcount); + ret = bch2_trans_update(trans, iter, new, 0); + } +out: +fsck_err: + printbuf_exit(&buf); + return ret; +} + +int bch2_gc_reflink_done(struct bch_fs *c) +{ + size_t idx = 0; + + int ret = bch2_trans_run(c, + for_each_btree_key_commit(trans, iter, + BTREE_ID_reflink, POS_MIN, + BTREE_ITER_prefetch, k, + NULL, NULL, BCH_TRANS_COMMIT_no_enospc, + bch2_gc_write_reflink_key(trans, &iter, k, &idx))); + c->reflink_gc_nr = 0; + return ret; +} + +int bch2_gc_reflink_start(struct bch_fs *c) +{ + c->reflink_gc_nr = 0; + + int ret = bch2_trans_run(c, + for_each_btree_key(trans, iter, BTREE_ID_reflink, POS_MIN, + BTREE_ITER_prefetch, k, ({ + const __le64 *refcount = bkey_refcount_c(k); + + if (!refcount) + continue; + + struct reflink_gc *r = genradix_ptr_alloc(&c->reflink_gc_table, + c->reflink_gc_nr++, GFP_KERNEL); + if (!r) { + ret = -BCH_ERR_ENOMEM_gc_reflink_start; + break; + } + + r->offset = k.k->p.offset; + r->size = k.k->size; + r->refcount = 0; + 0; + }))); + + bch_err_fn(c, ret); + return ret; +} diff --git a/fs/bcachefs/reflink.h b/fs/bcachefs/reflink.h index 51afe11d8ed6..1632780bdf18 100644 --- a/fs/bcachefs/reflink.h +++ b/fs/bcachefs/reflink.h @@ -2,9 +2,8 @@ #ifndef _BCACHEFS_REFLINK_H #define _BCACHEFS_REFLINK_H -enum bch_validate_flags; - -int bch2_reflink_p_validate(struct bch_fs *, struct bkey_s_c, enum bch_validate_flags); +int bch2_reflink_p_validate(struct bch_fs *, struct bkey_s_c, + struct bkey_validate_context); void bch2_reflink_p_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c); bool bch2_reflink_p_merge(struct bch_fs *, struct bkey_s, struct bkey_s_c); int bch2_trigger_reflink_p(struct btree_trans *, enum btree_id, unsigned, @@ -19,7 +18,8 @@ int bch2_trigger_reflink_p(struct btree_trans *, enum btree_id, unsigned, .min_val_size = 16, \ }) -int bch2_reflink_v_validate(struct bch_fs *, struct bkey_s_c, enum bch_validate_flags); +int bch2_reflink_v_validate(struct bch_fs *, struct bkey_s_c, + struct bkey_validate_context); void bch2_reflink_v_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c); int bch2_trigger_reflink_v(struct btree_trans *, enum btree_id, unsigned, struct bkey_s_c, struct bkey_s, @@ -34,7 +34,7 @@ int bch2_trigger_reflink_v(struct btree_trans *, enum btree_id, unsigned, }) int bch2_indirect_inline_data_validate(struct bch_fs *, struct bkey_s_c, - enum bch_validate_flags); + struct bkey_validate_context); void bch2_indirect_inline_data_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c); int bch2_trigger_indirect_inline_data(struct btree_trans *, @@ -73,7 +73,15 @@ static inline __le64 *bkey_refcount(struct bkey_s k) } } +struct bkey_s_c bch2_lookup_indirect_extent(struct btree_trans *, struct btree_iter *, + s64 *, struct bkey_s_c_reflink_p, + bool, unsigned); + s64 bch2_remap_range(struct bch_fs *, subvol_inum, u64, - subvol_inum, u64, u64, u64, s64 *); + subvol_inum, u64, u64, u64, s64 *, + bool); + +int bch2_gc_reflink_done(struct bch_fs *); +int bch2_gc_reflink_start(struct bch_fs *); #endif /* _BCACHEFS_REFLINK_H */ diff --git a/fs/bcachefs/reflink_format.h b/fs/bcachefs/reflink_format.h index 6772eebb1fc6..92995e4f898e 100644 --- a/fs/bcachefs/reflink_format.h +++ b/fs/bcachefs/reflink_format.h @@ -4,7 +4,7 @@ struct bch_reflink_p { struct bch_val v; - __le64 idx; + __le64 idx_flags; /* * A reflink pointer might point to an indirect extent which is then * later split (by copygc or rebalance). If we only pointed to part of @@ -17,6 +17,11 @@ struct bch_reflink_p { __le32 back_pad; } __packed __aligned(8); +LE64_BITMASK(REFLINK_P_IDX, struct bch_reflink_p, idx_flags, 0, 56); +LE64_BITMASK(REFLINK_P_ERROR, struct bch_reflink_p, idx_flags, 56, 57); +LE64_BITMASK(REFLINK_P_MAY_UPDATE_OPTIONS, + struct bch_reflink_p, idx_flags, 57, 58); + struct bch_reflink_v { struct bch_val v; __le64 refcount; diff --git a/fs/bcachefs/sb-clean.c b/fs/bcachefs/sb-clean.c index 005275281804..59c8770e4a0e 100644 --- a/fs/bcachefs/sb-clean.c +++ b/fs/bcachefs/sb-clean.c @@ -23,6 +23,10 @@ int bch2_sb_clean_validate_late(struct bch_fs *c, struct bch_sb_field_clean *clean, int write) { + struct bkey_validate_context from = { + .flags = write, + .from = BKEY_VALIDATE_superblock, + }; struct jset_entry *entry; int ret; @@ -40,7 +44,7 @@ int bch2_sb_clean_validate_late(struct bch_fs *c, struct bch_sb_field_clean *cle ret = bch2_journal_entry_validate(c, NULL, entry, le16_to_cpu(c->disk_sb.sb->version), BCH_SB_BIG_ENDIAN(c->disk_sb.sb), - write); + from); if (ret) return ret; } diff --git a/fs/bcachefs/sb-counters_format.h b/fs/bcachefs/sb-counters_format.h index 62ea478215d0..fdcf598f08b1 100644 --- a/fs/bcachefs/sb-counters_format.h +++ b/fs/bcachefs/sb-counters_format.h @@ -2,86 +2,91 @@ #ifndef _BCACHEFS_SB_COUNTERS_FORMAT_H #define _BCACHEFS_SB_COUNTERS_FORMAT_H -#define BCH_PERSISTENT_COUNTERS() \ - x(io_read, 0) \ - x(io_write, 1) \ - x(io_move, 2) \ - x(bucket_invalidate, 3) \ - x(bucket_discard, 4) \ - x(bucket_alloc, 5) \ - x(bucket_alloc_fail, 6) \ - x(btree_cache_scan, 7) \ - x(btree_cache_reap, 8) \ - x(btree_cache_cannibalize, 9) \ - x(btree_cache_cannibalize_lock, 10) \ - x(btree_cache_cannibalize_lock_fail, 11) \ - x(btree_cache_cannibalize_unlock, 12) \ - x(btree_node_write, 13) \ - x(btree_node_read, 14) \ - x(btree_node_compact, 15) \ - x(btree_node_merge, 16) \ - x(btree_node_split, 17) \ - x(btree_node_rewrite, 18) \ - x(btree_node_alloc, 19) \ - x(btree_node_free, 20) \ - x(btree_node_set_root, 21) \ - x(btree_path_relock_fail, 22) \ - x(btree_path_upgrade_fail, 23) \ - x(btree_reserve_get_fail, 24) \ - x(journal_entry_full, 25) \ - x(journal_full, 26) \ - x(journal_reclaim_finish, 27) \ - x(journal_reclaim_start, 28) \ - x(journal_write, 29) \ - x(read_promote, 30) \ - x(read_bounce, 31) \ - x(read_split, 33) \ - x(read_retry, 32) \ - x(read_reuse_race, 34) \ - x(move_extent_read, 35) \ - x(move_extent_write, 36) \ - x(move_extent_finish, 37) \ - x(move_extent_fail, 38) \ - x(move_extent_start_fail, 39) \ - x(copygc, 40) \ - x(copygc_wait, 41) \ - x(gc_gens_end, 42) \ - x(gc_gens_start, 43) \ - x(trans_blocked_journal_reclaim, 44) \ - x(trans_restart_btree_node_reused, 45) \ - x(trans_restart_btree_node_split, 46) \ - x(trans_restart_fault_inject, 47) \ - x(trans_restart_iter_upgrade, 48) \ - x(trans_restart_journal_preres_get, 49) \ - x(trans_restart_journal_reclaim, 50) \ - x(trans_restart_journal_res_get, 51) \ - x(trans_restart_key_cache_key_realloced, 52) \ - x(trans_restart_key_cache_raced, 53) \ - x(trans_restart_mark_replicas, 54) \ - x(trans_restart_mem_realloced, 55) \ - x(trans_restart_memory_allocation_failure, 56) \ - x(trans_restart_relock, 57) \ - x(trans_restart_relock_after_fill, 58) \ - x(trans_restart_relock_key_cache_fill, 59) \ - x(trans_restart_relock_next_node, 60) \ - x(trans_restart_relock_parent_for_fill, 61) \ - x(trans_restart_relock_path, 62) \ - x(trans_restart_relock_path_intent, 63) \ - x(trans_restart_too_many_iters, 64) \ - x(trans_restart_traverse, 65) \ - x(trans_restart_upgrade, 66) \ - x(trans_restart_would_deadlock, 67) \ - x(trans_restart_would_deadlock_write, 68) \ - x(trans_restart_injected, 69) \ - x(trans_restart_key_cache_upgrade, 70) \ - x(trans_traverse_all, 71) \ - x(transaction_commit, 72) \ - x(write_super, 73) \ - x(trans_restart_would_deadlock_recursion_limit, 74) \ - x(trans_restart_write_buffer_flush, 75) \ - x(trans_restart_split_race, 76) \ - x(write_buffer_flush_slowpath, 77) \ - x(write_buffer_flush_sync, 78) +enum counters_flags { + TYPE_COUNTER = BIT(0), /* event counters */ + TYPE_SECTORS = BIT(1), /* amount counters, the unit is sectors */ +}; + +#define BCH_PERSISTENT_COUNTERS() \ + x(io_read, 0, TYPE_SECTORS) \ + x(io_write, 1, TYPE_SECTORS) \ + x(io_move, 2, TYPE_SECTORS) \ + x(bucket_invalidate, 3, TYPE_COUNTER) \ + x(bucket_discard, 4, TYPE_COUNTER) \ + x(bucket_alloc, 5, TYPE_COUNTER) \ + x(bucket_alloc_fail, 6, TYPE_COUNTER) \ + x(btree_cache_scan, 7, TYPE_COUNTER) \ + x(btree_cache_reap, 8, TYPE_COUNTER) \ + x(btree_cache_cannibalize, 9, TYPE_COUNTER) \ + x(btree_cache_cannibalize_lock, 10, TYPE_COUNTER) \ + x(btree_cache_cannibalize_lock_fail, 11, TYPE_COUNTER) \ + x(btree_cache_cannibalize_unlock, 12, TYPE_COUNTER) \ + x(btree_node_write, 13, TYPE_COUNTER) \ + x(btree_node_read, 14, TYPE_COUNTER) \ + x(btree_node_compact, 15, TYPE_COUNTER) \ + x(btree_node_merge, 16, TYPE_COUNTER) \ + x(btree_node_split, 17, TYPE_COUNTER) \ + x(btree_node_rewrite, 18, TYPE_COUNTER) \ + x(btree_node_alloc, 19, TYPE_COUNTER) \ + x(btree_node_free, 20, TYPE_COUNTER) \ + x(btree_node_set_root, 21, TYPE_COUNTER) \ + x(btree_path_relock_fail, 22, TYPE_COUNTER) \ + x(btree_path_upgrade_fail, 23, TYPE_COUNTER) \ + x(btree_reserve_get_fail, 24, TYPE_COUNTER) \ + x(journal_entry_full, 25, TYPE_COUNTER) \ + x(journal_full, 26, TYPE_COUNTER) \ + x(journal_reclaim_finish, 27, TYPE_COUNTER) \ + x(journal_reclaim_start, 28, TYPE_COUNTER) \ + x(journal_write, 29, TYPE_COUNTER) \ + x(read_promote, 30, TYPE_COUNTER) \ + x(read_bounce, 31, TYPE_COUNTER) \ + x(read_split, 33, TYPE_COUNTER) \ + x(read_retry, 32, TYPE_COUNTER) \ + x(read_reuse_race, 34, TYPE_COUNTER) \ + x(move_extent_read, 35, TYPE_SECTORS) \ + x(move_extent_write, 36, TYPE_SECTORS) \ + x(move_extent_finish, 37, TYPE_SECTORS) \ + x(move_extent_fail, 38, TYPE_COUNTER) \ + x(move_extent_start_fail, 39, TYPE_COUNTER) \ + x(copygc, 40, TYPE_COUNTER) \ + x(copygc_wait, 41, TYPE_COUNTER) \ + x(gc_gens_end, 42, TYPE_COUNTER) \ + x(gc_gens_start, 43, TYPE_COUNTER) \ + x(trans_blocked_journal_reclaim, 44, TYPE_COUNTER) \ + x(trans_restart_btree_node_reused, 45, TYPE_COUNTER) \ + x(trans_restart_btree_node_split, 46, TYPE_COUNTER) \ + x(trans_restart_fault_inject, 47, TYPE_COUNTER) \ + x(trans_restart_iter_upgrade, 48, TYPE_COUNTER) \ + x(trans_restart_journal_preres_get, 49, TYPE_COUNTER) \ + x(trans_restart_journal_reclaim, 50, TYPE_COUNTER) \ + x(trans_restart_journal_res_get, 51, TYPE_COUNTER) \ + x(trans_restart_key_cache_key_realloced, 52, TYPE_COUNTER) \ + x(trans_restart_key_cache_raced, 53, TYPE_COUNTER) \ + x(trans_restart_mark_replicas, 54, TYPE_COUNTER) \ + x(trans_restart_mem_realloced, 55, TYPE_COUNTER) \ + x(trans_restart_memory_allocation_failure, 56, TYPE_COUNTER) \ + x(trans_restart_relock, 57, TYPE_COUNTER) \ + x(trans_restart_relock_after_fill, 58, TYPE_COUNTER) \ + x(trans_restart_relock_key_cache_fill, 59, TYPE_COUNTER) \ + x(trans_restart_relock_next_node, 60, TYPE_COUNTER) \ + x(trans_restart_relock_parent_for_fill, 61, TYPE_COUNTER) \ + x(trans_restart_relock_path, 62, TYPE_COUNTER) \ + x(trans_restart_relock_path_intent, 63, TYPE_COUNTER) \ + x(trans_restart_too_many_iters, 64, TYPE_COUNTER) \ + x(trans_restart_traverse, 65, TYPE_COUNTER) \ + x(trans_restart_upgrade, 66, TYPE_COUNTER) \ + x(trans_restart_would_deadlock, 67, TYPE_COUNTER) \ + x(trans_restart_would_deadlock_write, 68, TYPE_COUNTER) \ + x(trans_restart_injected, 69, TYPE_COUNTER) \ + x(trans_restart_key_cache_upgrade, 70, TYPE_COUNTER) \ + x(trans_traverse_all, 71, TYPE_COUNTER) \ + x(transaction_commit, 72, TYPE_COUNTER) \ + x(write_super, 73, TYPE_COUNTER) \ + x(trans_restart_would_deadlock_recursion_limit, 74, TYPE_COUNTER) \ + x(trans_restart_write_buffer_flush, 75, TYPE_COUNTER) \ + x(trans_restart_split_race, 76, TYPE_COUNTER) \ + x(write_buffer_flush_slowpath, 77, TYPE_COUNTER) \ + x(write_buffer_flush_sync, 78, TYPE_COUNTER) enum bch_persistent_counters { #define x(t, n, ...) BCH_COUNTER_##t, diff --git a/fs/bcachefs/sb-downgrade.c b/fs/bcachefs/sb-downgrade.c index 8767c33c2b51..14f6b6a5fb38 100644 --- a/fs/bcachefs/sb-downgrade.c +++ b/fs/bcachefs/sb-downgrade.c @@ -81,7 +81,19 @@ BCH_FSCK_ERR_accounting_mismatch) \ x(inode_has_child_snapshots, \ BIT_ULL(BCH_RECOVERY_PASS_check_inodes), \ - BCH_FSCK_ERR_inode_has_child_snapshots_wrong) + BCH_FSCK_ERR_inode_has_child_snapshots_wrong) \ + x(backpointer_bucket_gen, \ + BIT_ULL(BCH_RECOVERY_PASS_check_extents_to_backpointers),\ + BCH_FSCK_ERR_backpointer_to_missing_ptr, \ + BCH_FSCK_ERR_ptr_to_missing_backpointer) \ + x(disk_accounting_big_endian, \ + BIT_ULL(BCH_RECOVERY_PASS_check_allocations), \ + BCH_FSCK_ERR_accounting_mismatch, \ + BCH_FSCK_ERR_accounting_key_replicas_nr_devs_0, \ + BCH_FSCK_ERR_accounting_key_junk_at_end) \ + x(directory_size, \ + BIT_ULL(BCH_RECOVERY_PASS_check_inodes), \ + BCH_FSCK_ERR_directory_size_mismatch) \ #define DOWNGRADE_TABLE() \ x(bucket_stripe_sectors, \ @@ -117,7 +129,19 @@ BCH_FSCK_ERR_bkey_version_in_future) \ x(rebalance_work_acct_fix, \ BIT_ULL(BCH_RECOVERY_PASS_check_allocations), \ - BCH_FSCK_ERR_accounting_mismatch) + BCH_FSCK_ERR_accounting_mismatch, \ + BCH_FSCK_ERR_accounting_key_replicas_nr_devs_0, \ + BCH_FSCK_ERR_accounting_key_junk_at_end) \ + x(backpointer_bucket_gen, \ + BIT_ULL(BCH_RECOVERY_PASS_check_extents_to_backpointers),\ + BCH_FSCK_ERR_backpointer_bucket_offset_wrong, \ + BCH_FSCK_ERR_backpointer_to_missing_ptr, \ + BCH_FSCK_ERR_ptr_to_missing_backpointer) \ + x(disk_accounting_big_endian, \ + BIT_ULL(BCH_RECOVERY_PASS_check_allocations), \ + BCH_FSCK_ERR_accounting_mismatch, \ + BCH_FSCK_ERR_accounting_key_replicas_nr_devs_0, \ + BCH_FSCK_ERR_accounting_key_junk_at_end) struct upgrade_downgrade_entry { u64 recovery_passes; diff --git a/fs/bcachefs/sb-errors_format.h b/fs/bcachefs/sb-errors_format.h index 9feb6739f77a..0b4fe899209b 100644 --- a/fs/bcachefs/sb-errors_format.h +++ b/fs/bcachefs/sb-errors_format.h @@ -5,9 +5,8 @@ enum bch_fsck_flags { FSCK_CAN_FIX = 1 << 0, FSCK_CAN_IGNORE = 1 << 1, - FSCK_NEED_FSCK = 1 << 2, - FSCK_NO_RATELIMIT = 1 << 3, - FSCK_AUTOFIX = 1 << 4, + FSCK_NO_RATELIMIT = 1 << 2, + FSCK_AUTOFIX = 1 << 3, }; #define BCH_SB_ERRS() \ @@ -59,7 +58,7 @@ enum bch_fsck_flags { x(bset_empty, 45, 0) \ x(bset_bad_seq, 46, 0) \ x(bset_blacklisted_journal_seq, 47, 0) \ - x(first_bset_blacklisted_journal_seq, 48, 0) \ + x(first_bset_blacklisted_journal_seq, 48, FSCK_AUTOFIX) \ x(btree_node_bad_btree, 49, 0) \ x(btree_node_bad_level, 50, 0) \ x(btree_node_bad_min_key, 51, 0) \ @@ -68,17 +67,17 @@ enum bch_fsck_flags { x(btree_node_bkey_past_bset_end, 54, 0) \ x(btree_node_bkey_bad_format, 55, 0) \ x(btree_node_bad_bkey, 56, 0) \ - x(btree_node_bkey_out_of_order, 57, 0) \ - x(btree_root_bkey_invalid, 58, 0) \ - x(btree_root_read_error, 59, 0) \ + x(btree_node_bkey_out_of_order, 57, FSCK_AUTOFIX) \ + x(btree_root_bkey_invalid, 58, FSCK_AUTOFIX) \ + x(btree_root_read_error, 59, FSCK_AUTOFIX) \ x(btree_root_bad_min_key, 60, 0) \ x(btree_root_bad_max_key, 61, 0) \ - x(btree_node_read_error, 62, 0) \ - x(btree_node_topology_bad_min_key, 63, 0) \ - x(btree_node_topology_bad_max_key, 64, 0) \ - x(btree_node_topology_overwritten_by_prev_node, 65, 0) \ - x(btree_node_topology_overwritten_by_next_node, 66, 0) \ - x(btree_node_topology_interior_node_empty, 67, 0) \ + x(btree_node_read_error, 62, FSCK_AUTOFIX) \ + x(btree_node_topology_bad_min_key, 63, FSCK_AUTOFIX) \ + x(btree_node_topology_bad_max_key, 64, FSCK_AUTOFIX) \ + x(btree_node_topology_overwritten_by_prev_node, 65, FSCK_AUTOFIX) \ + x(btree_node_topology_overwritten_by_next_node, 66, FSCK_AUTOFIX) \ + x(btree_node_topology_interior_node_empty, 67, FSCK_AUTOFIX) \ x(fs_usage_hidden_wrong, 68, FSCK_AUTOFIX) \ x(fs_usage_btree_wrong, 69, FSCK_AUTOFIX) \ x(fs_usage_data_wrong, 70, FSCK_AUTOFIX) \ @@ -123,11 +122,12 @@ enum bch_fsck_flags { x(alloc_key_cached_sectors_wrong, 109, FSCK_AUTOFIX) \ x(alloc_key_stripe_wrong, 110, FSCK_AUTOFIX) \ x(alloc_key_stripe_redundancy_wrong, 111, FSCK_AUTOFIX) \ + x(alloc_key_journal_seq_in_future, 298, FSCK_AUTOFIX) \ x(bucket_sector_count_overflow, 112, 0) \ x(bucket_metadata_type_mismatch, 113, 0) \ - x(need_discard_key_wrong, 114, 0) \ - x(freespace_key_wrong, 115, 0) \ - x(freespace_hole_missing, 116, 0) \ + x(need_discard_key_wrong, 114, FSCK_AUTOFIX) \ + x(freespace_key_wrong, 115, FSCK_AUTOFIX) \ + x(freespace_hole_missing, 116, FSCK_AUTOFIX) \ x(bucket_gens_val_size_bad, 117, 0) \ x(bucket_gens_key_wrong, 118, FSCK_AUTOFIX) \ x(bucket_gens_hole_wrong, 119, FSCK_AUTOFIX) \ @@ -139,9 +139,10 @@ enum bch_fsck_flags { x(discarding_bucket_not_in_need_discard_btree, 291, 0) \ x(backpointer_bucket_offset_wrong, 125, 0) \ x(backpointer_level_bad, 294, 0) \ - x(backpointer_to_missing_device, 126, 0) \ - x(backpointer_to_missing_alloc, 127, 0) \ - x(backpointer_to_missing_ptr, 128, 0) \ + x(backpointer_dev_bad, 297, 0) \ + x(backpointer_to_missing_device, 126, FSCK_AUTOFIX) \ + x(backpointer_to_missing_alloc, 127, FSCK_AUTOFIX) \ + x(backpointer_to_missing_ptr, 128, FSCK_AUTOFIX) \ x(lru_entry_at_time_0, 129, FSCK_AUTOFIX) \ x(lru_entry_to_invalid_bucket, 130, FSCK_AUTOFIX) \ x(lru_entry_bad, 131, FSCK_AUTOFIX) \ @@ -167,14 +168,15 @@ enum bch_fsck_flags { x(ptr_to_incorrect_stripe, 151, 0) \ x(ptr_gen_newer_than_bucket_gen, 152, 0) \ x(ptr_too_stale, 153, 0) \ - x(stale_dirty_ptr, 154, 0) \ + x(stale_dirty_ptr, 154, FSCK_AUTOFIX) \ x(ptr_bucket_data_type_mismatch, 155, 0) \ x(ptr_cached_and_erasure_coded, 156, 0) \ x(ptr_crc_uncompressed_size_too_small, 157, 0) \ + x(ptr_crc_uncompressed_size_too_big, 161, 0) \ + x(ptr_crc_uncompressed_size_mismatch, 300, 0) \ x(ptr_crc_csum_type_unknown, 158, 0) \ x(ptr_crc_compression_type_unknown, 159, 0) \ x(ptr_crc_redundant, 160, 0) \ - x(ptr_crc_uncompressed_size_too_big, 161, 0) \ x(ptr_crc_nonce_mismatch, 162, 0) \ x(ptr_stripe_redundant, 163, 0) \ x(reservation_key_nr_replicas_invalid, 164, 0) \ @@ -209,6 +211,7 @@ enum bch_fsck_flags { x(bkey_in_missing_snapshot, 190, 0) \ x(inode_pos_inode_nonzero, 191, 0) \ x(inode_pos_blockdev_range, 192, 0) \ + x(inode_alloc_cursor_inode_bad, 301, 0) \ x(inode_unpack_error, 193, 0) \ x(inode_str_hash_invalid, 194, 0) \ x(inode_v3_fields_start_bad, 195, 0) \ @@ -232,6 +235,7 @@ enum bch_fsck_flags { x(inode_wrong_nlink, 209, FSCK_AUTOFIX) \ x(inode_has_child_snapshots_wrong, 287, 0) \ x(inode_unreachable, 210, FSCK_AUTOFIX) \ + x(inode_journal_seq_in_future, 299, FSCK_AUTOFIX) \ x(deleted_inode_but_clean, 211, FSCK_AUTOFIX) \ x(deleted_inode_missing, 212, FSCK_AUTOFIX) \ x(deleted_inode_is_dir, 213, FSCK_AUTOFIX) \ @@ -252,6 +256,7 @@ enum bch_fsck_flags { x(dirent_in_missing_dir_inode, 227, 0) \ x(dirent_in_non_dir_inode, 228, 0) \ x(dirent_to_missing_inode, 229, 0) \ + x(dirent_to_overwritten_inode, 302, 0) \ x(dirent_to_missing_subvol, 230, 0) \ x(dirent_to_itself, 231, 0) \ x(quota_type_invalid, 232, 0) \ @@ -288,7 +293,7 @@ enum bch_fsck_flags { x(btree_root_unreadable_and_scan_found_nothing, 263, 0) \ x(snapshot_node_missing, 264, 0) \ x(dup_backpointer_to_bad_csum_extent, 265, 0) \ - x(btree_bitmap_not_marked, 266, 0) \ + x(btree_bitmap_not_marked, 266, FSCK_AUTOFIX) \ x(sb_clean_entry_overrun, 267, 0) \ x(btree_ptr_v2_written_0, 268, 0) \ x(subvol_snapshot_bad, 269, 0) \ @@ -306,7 +311,10 @@ enum bch_fsck_flags { x(accounting_key_replicas_devs_unsorted, 280, FSCK_AUTOFIX) \ x(accounting_key_version_0, 282, FSCK_AUTOFIX) \ x(logged_op_but_clean, 283, FSCK_AUTOFIX) \ - x(MAX, 295, 0) + x(compression_opt_not_marked_in_sb, 295, FSCK_AUTOFIX) \ + x(compression_type_not_marked_in_sb, 296, FSCK_AUTOFIX) \ + x(directory_size_mismatch, 303, FSCK_AUTOFIX) \ + x(MAX, 304, 0) enum bch_sb_error_id { #define x(t, n, ...) BCH_FSCK_ERR_##t = n, diff --git a/fs/bcachefs/six.c b/fs/bcachefs/six.c index 617d07e53b20..7e7c66a1e1a6 100644 --- a/fs/bcachefs/six.c +++ b/fs/bcachefs/six.c @@ -491,8 +491,12 @@ static int six_lock_slowpath(struct six_lock *lock, enum six_lock_type type, list_del(&wait->list); raw_spin_unlock(&lock->wait_lock); - if (unlikely(acquired)) + if (unlikely(acquired)) { do_six_unlock_type(lock, type); + } else if (type == SIX_LOCK_write) { + six_clear_bitmask(lock, SIX_LOCK_HELD_write); + six_lock_wakeup(lock, atomic_read(&lock->state), SIX_LOCK_read); + } break; } @@ -501,10 +505,6 @@ static int six_lock_slowpath(struct six_lock *lock, enum six_lock_type type, __set_current_state(TASK_RUNNING); out: - if (ret && type == SIX_LOCK_write) { - six_clear_bitmask(lock, SIX_LOCK_HELD_write); - six_lock_wakeup(lock, atomic_read(&lock->state), SIX_LOCK_read); - } trace_contention_end(lock, 0); return ret; @@ -616,8 +616,6 @@ void six_unlock_ip(struct six_lock *lock, enum six_lock_type type, unsigned long if (type != SIX_LOCK_write) six_release(&lock->dep_map, ip); - else - lock->seq++; if (type == SIX_LOCK_intent && lock->intent_lock_recurse) { @@ -625,6 +623,15 @@ void six_unlock_ip(struct six_lock *lock, enum six_lock_type type, unsigned long return; } + if (type == SIX_LOCK_write && + lock->write_lock_recurse) { + --lock->write_lock_recurse; + return; + } + + if (type == SIX_LOCK_write) + lock->seq++; + do_six_unlock_type(lock, type); } EXPORT_SYMBOL_GPL(six_unlock_ip); @@ -735,13 +742,13 @@ void six_lock_increment(struct six_lock *lock, enum six_lock_type type) atomic_add(l[type].lock_val, &lock->state); } break; + case SIX_LOCK_write: + lock->write_lock_recurse++; + fallthrough; case SIX_LOCK_intent: EBUG_ON(!(atomic_read(&lock->state) & SIX_LOCK_HELD_intent)); lock->intent_lock_recurse++; break; - case SIX_LOCK_write: - BUG(); - break; } } EXPORT_SYMBOL_GPL(six_lock_increment); diff --git a/fs/bcachefs/six.h b/fs/bcachefs/six.h index 68d46fd7f391..c142e06b7a3a 100644 --- a/fs/bcachefs/six.h +++ b/fs/bcachefs/six.h @@ -137,6 +137,7 @@ struct six_lock { atomic_t state; u32 seq; unsigned intent_lock_recurse; + unsigned write_lock_recurse; struct task_struct *owner; unsigned __percpu *readers; raw_spinlock_t wait_lock; diff --git a/fs/bcachefs/snapshot.c b/fs/bcachefs/snapshot.c index ae57638506c3..c54091a28909 100644 --- a/fs/bcachefs/snapshot.c +++ b/fs/bcachefs/snapshot.c @@ -2,6 +2,7 @@ #include "bcachefs.h" #include "bkey_buf.h" +#include "btree_cache.h" #include "btree_key_cache.h" #include "btree_update.h" #include "buckets.h" @@ -32,7 +33,7 @@ void bch2_snapshot_tree_to_text(struct printbuf *out, struct bch_fs *c, } int bch2_snapshot_tree_validate(struct bch_fs *c, struct bkey_s_c k, - enum bch_validate_flags flags) + struct bkey_validate_context from) { int ret = 0; @@ -225,7 +226,7 @@ void bch2_snapshot_to_text(struct printbuf *out, struct bch_fs *c, } int bch2_snapshot_validate(struct bch_fs *c, struct bkey_s_c k, - enum bch_validate_flags flags) + struct bkey_validate_context from) { struct bkey_s_c_snapshot s; u32 i, id; @@ -279,23 +280,6 @@ fsck_err: return ret; } -static void __set_is_ancestor_bitmap(struct bch_fs *c, u32 id) -{ - struct snapshot_t *t = snapshot_t_mut(c, id); - u32 parent = id; - - while ((parent = bch2_snapshot_parent_early(c, parent)) && - parent - id - 1 < IS_ANCESTOR_BITMAP) - __set_bit(parent - id - 1, t->is_ancestor); -} - -static void set_is_ancestor_bitmap(struct bch_fs *c, u32 id) -{ - mutex_lock(&c->snapshot_table_lock); - __set_is_ancestor_bitmap(c, id); - mutex_unlock(&c->snapshot_table_lock); -} - static int __bch2_mark_snapshot(struct btree_trans *trans, enum btree_id btree, unsigned level, struct bkey_s_c old, struct bkey_s_c new, @@ -317,6 +301,7 @@ static int __bch2_mark_snapshot(struct btree_trans *trans, if (new.k->type == KEY_TYPE_snapshot) { struct bkey_s_c_snapshot s = bkey_s_c_to_snapshot(new); + t->live = true; t->parent = le32_to_cpu(s.v->parent); t->children[0] = le32_to_cpu(s.v->children[0]); t->children[1] = le32_to_cpu(s.v->children[1]); @@ -335,7 +320,11 @@ static int __bch2_mark_snapshot(struct btree_trans *trans, t->skip[2] = 0; } - __set_is_ancestor_bitmap(c, id); + u32 parent = id; + + while ((parent = bch2_snapshot_parent_early(c, parent)) && + parent - id - 1 < IS_ANCESTOR_BITMAP) + __set_bit(parent - id - 1, t->is_ancestor); if (BCH_SNAPSHOT_DELETED(s.v)) { set_bit(BCH_FS_need_delete_dead_snapshots, &c->flags); @@ -365,70 +354,6 @@ int bch2_snapshot_lookup(struct btree_trans *trans, u32 id, BTREE_ITER_with_updates, snapshot, s); } -static int bch2_snapshot_live(struct btree_trans *trans, u32 id) -{ - struct bch_snapshot v; - int ret; - - if (!id) - return 0; - - ret = bch2_snapshot_lookup(trans, id, &v); - if (bch2_err_matches(ret, ENOENT)) - bch_err(trans->c, "snapshot node %u not found", id); - if (ret) - return ret; - - return !BCH_SNAPSHOT_DELETED(&v); -} - -/* - * If @k is a snapshot with just one live child, it's part of a linear chain, - * which we consider to be an equivalence class: and then after snapshot - * deletion cleanup, there should only be a single key at a given position in - * this equivalence class. - * - * This sets the equivalence class of @k to be the child's equivalence class, if - * it's part of such a linear chain: this correctly sets equivalence classes on - * startup if we run leaf to root (i.e. in natural key order). - */ -static int bch2_snapshot_set_equiv(struct btree_trans *trans, struct bkey_s_c k) -{ - struct bch_fs *c = trans->c; - unsigned i, nr_live = 0, live_idx = 0; - struct bkey_s_c_snapshot snap; - u32 id = k.k->p.offset, child[2]; - - if (k.k->type != KEY_TYPE_snapshot) - return 0; - - snap = bkey_s_c_to_snapshot(k); - - child[0] = le32_to_cpu(snap.v->children[0]); - child[1] = le32_to_cpu(snap.v->children[1]); - - for (i = 0; i < 2; i++) { - int ret = bch2_snapshot_live(trans, child[i]); - - if (ret < 0) - return ret; - - if (ret) - live_idx = i; - nr_live += ret; - } - - mutex_lock(&c->snapshot_table_lock); - - snapshot_t_mut(c, id)->equiv = nr_live == 1 - ? snapshot_t_mut(c, child[live_idx])->equiv - : id; - - mutex_unlock(&c->snapshot_table_lock); - - return 0; -} - /* fsck: */ static u32 bch2_snapshot_child(struct bch_fs *c, u32 id, unsigned child) @@ -506,7 +431,6 @@ static int bch2_snapshot_tree_master_subvol(struct btree_trans *trans, break; } } - bch2_trans_iter_exit(trans, &iter); if (!ret && !found) { @@ -536,6 +460,7 @@ static int check_snapshot_tree(struct btree_trans *trans, struct bch_snapshot s; struct bch_subvolume subvol; struct printbuf buf = PRINTBUF; + struct btree_iter snapshot_iter = {}; u32 root_id; int ret; @@ -545,22 +470,35 @@ static int check_snapshot_tree(struct btree_trans *trans, st = bkey_s_c_to_snapshot_tree(k); root_id = le32_to_cpu(st.v->root_snapshot); - ret = bch2_snapshot_lookup(trans, root_id, &s); + struct bkey_s_c_snapshot snapshot_k = + bch2_bkey_get_iter_typed(trans, &snapshot_iter, BTREE_ID_snapshots, + POS(0, root_id), 0, snapshot); + ret = bkey_err(snapshot_k); if (ret && !bch2_err_matches(ret, ENOENT)) goto err; + if (!ret) + bkey_val_copy(&s, snapshot_k); + if (fsck_err_on(ret || root_id != bch2_snapshot_root(c, root_id) || st.k->p.offset != le32_to_cpu(s.tree), trans, snapshot_tree_to_missing_snapshot, "snapshot tree points to missing/incorrect snapshot:\n %s", - (bch2_bkey_val_to_text(&buf, c, st.s_c), buf.buf))) { + (bch2_bkey_val_to_text(&buf, c, st.s_c), + prt_newline(&buf), + ret + ? prt_printf(&buf, "(%s)", bch2_err_str(ret)) + : bch2_bkey_val_to_text(&buf, c, snapshot_k.s_c), + buf.buf))) { ret = bch2_btree_delete_at(trans, iter, 0); goto err; } - ret = bch2_subvolume_get(trans, le32_to_cpu(st.v->master_subvol), - false, 0, &subvol); + if (!st.v->master_subvol) + goto out; + + ret = bch2_subvolume_get(trans, le32_to_cpu(st.v->master_subvol), false, &subvol); if (ret && !bch2_err_matches(ret, ENOENT)) goto err; @@ -603,8 +541,10 @@ static int check_snapshot_tree(struct btree_trans *trans, u->v.master_subvol = cpu_to_le32(subvol_id); st = snapshot_tree_i_to_s_c(u); } +out: err: fsck_err: + bch2_trans_iter_exit(trans, &snapshot_iter); printbuf_exit(&buf); return ret; } @@ -799,7 +739,7 @@ static int check_snapshot(struct btree_trans *trans, if (should_have_subvol) { id = le32_to_cpu(s.subvol); - ret = bch2_subvolume_get(trans, id, 0, false, &subvol); + ret = bch2_subvolume_get(trans, id, false, &subvol); if (bch2_err_matches(ret, ENOENT)) bch_err(c, "snapshot points to nonexistent subvolume:\n %s", (bch2_bkey_val_to_text(&buf, c, k), buf.buf)); @@ -902,7 +842,7 @@ static int check_snapshot_exists(struct btree_trans *trans, u32 id) { struct bch_fs *c = trans->c; - if (bch2_snapshot_equiv(c, id)) + if (bch2_snapshot_exists(c, id)) return 0; /* Do we need to reconstruct the snapshot_tree entry as well? */ @@ -951,8 +891,7 @@ static int check_snapshot_exists(struct btree_trans *trans, u32 id) return bch2_btree_insert_trans(trans, BTREE_ID_snapshots, &snapshot->k_i, 0) ?: bch2_mark_snapshot(trans, BTREE_ID_snapshots, 0, - bkey_s_c_null, bkey_i_to_s(&snapshot->k_i), 0) ?: - bch2_snapshot_set_equiv(trans, bkey_i_to_s_c(&snapshot->k_i)); + bkey_s_c_null, bkey_i_to_s(&snapshot->k_i), 0); } /* Figure out which snapshot nodes belong in the same tree: */ @@ -1050,7 +989,7 @@ int bch2_reconstruct_snapshots(struct bch_fs *c) snapshot_id_list_to_text(&buf, t); darray_for_each(*t, id) { - if (fsck_err_on(!bch2_snapshot_equiv(c, *id), + if (fsck_err_on(!bch2_snapshot_exists(c, *id), trans, snapshot_node_missing, "snapshot node %u from tree %s missing, recreate?", *id, buf.buf)) { if (t->nr > 1) { @@ -1083,10 +1022,12 @@ int bch2_check_key_has_snapshot(struct btree_trans *trans, struct printbuf buf = PRINTBUF; int ret = 0; - if (fsck_err_on(!bch2_snapshot_equiv(c, k.k->p.snapshot), + if (fsck_err_on(!bch2_snapshot_exists(c, k.k->p.snapshot), trans, bkey_in_missing_snapshot, "key in missing snapshot %s, delete?", - (bch2_bkey_val_to_text(&buf, c, k), buf.buf))) + (bch2_btree_id_to_text(&buf, iter->btree_id), + prt_char(&buf, ' '), + bch2_bkey_val_to_text(&buf, c, k), buf.buf))) ret = bch2_btree_delete_at(trans, iter, BTREE_UPDATE_internal_snapshot_node) ?: 1; fsck_err: @@ -1100,13 +1041,11 @@ fsck_err: int bch2_snapshot_node_set_deleted(struct btree_trans *trans, u32 id) { struct btree_iter iter; - struct bkey_i_snapshot *s; - int ret = 0; - - s = bch2_bkey_get_mut_typed(trans, &iter, + struct bkey_i_snapshot *s = + bch2_bkey_get_mut_typed(trans, &iter, BTREE_ID_snapshots, POS(0, id), 0, snapshot); - ret = PTR_ERR_OR_ZERO(s); + int ret = PTR_ERR_OR_ZERO(s); if (unlikely(ret)) { bch2_fs_inconsistent_on(bch2_err_matches(ret, ENOENT), trans->c, "missing snapshot %u", id); @@ -1294,10 +1233,6 @@ static int create_snapids(struct btree_trans *trans, u32 parent, u32 tree, goto err; new_snapids[i] = iter.pos.offset; - - mutex_lock(&c->snapshot_table_lock); - snapshot_t_mut(c, new_snapids[i])->equiv = new_snapids[i]; - mutex_unlock(&c->snapshot_table_lock); } err: bch2_trans_iter_exit(trans, &iter); @@ -1403,129 +1338,153 @@ int bch2_snapshot_node_create(struct btree_trans *trans, u32 parent, * that key to snapshot leaf nodes, where we can mutate it */ -static int delete_dead_snapshots_process_key(struct btree_trans *trans, - struct btree_iter *iter, - struct bkey_s_c k, - snapshot_id_list *deleted, - snapshot_id_list *equiv_seen, - struct bpos *last_pos) +struct snapshot_interior_delete { + u32 id; + u32 live_child; +}; +typedef DARRAY(struct snapshot_interior_delete) interior_delete_list; + +static inline u32 interior_delete_has_id(interior_delete_list *l, u32 id) { - int ret = bch2_check_key_has_snapshot(trans, iter, k); - if (ret) - return ret < 0 ? ret : 0; + darray_for_each(*l, i) + if (i->id == id) + return i->live_child; + return 0; +} - struct bch_fs *c = trans->c; - u32 equiv = bch2_snapshot_equiv(c, k.k->p.snapshot); - if (!equiv) /* key for invalid snapshot node, but we chose not to delete */ +static unsigned __live_child(struct snapshot_table *t, u32 id, + snapshot_id_list *delete_leaves, + interior_delete_list *delete_interior) +{ + struct snapshot_t *s = __snapshot_t(t, id); + if (!s) return 0; - if (!bkey_eq(k.k->p, *last_pos)) - equiv_seen->nr = 0; + for (unsigned i = 0; i < ARRAY_SIZE(s->children); i++) + if (s->children[i] && + !snapshot_list_has_id(delete_leaves, s->children[i]) && + !interior_delete_has_id(delete_interior, s->children[i])) + return s->children[i]; - if (snapshot_list_has_id(deleted, k.k->p.snapshot)) - return bch2_btree_delete_at(trans, iter, - BTREE_UPDATE_internal_snapshot_node); + for (unsigned i = 0; i < ARRAY_SIZE(s->children); i++) { + u32 live_child = s->children[i] + ? __live_child(t, s->children[i], delete_leaves, delete_interior) + : 0; + if (live_child) + return live_child; + } - if (!bpos_eq(*last_pos, k.k->p) && - snapshot_list_has_id(equiv_seen, equiv)) - return bch2_btree_delete_at(trans, iter, - BTREE_UPDATE_internal_snapshot_node); + return 0; +} - *last_pos = k.k->p; +static unsigned live_child(struct bch_fs *c, u32 id, + snapshot_id_list *delete_leaves, + interior_delete_list *delete_interior) +{ + rcu_read_lock(); + u32 ret = __live_child(rcu_dereference(c->snapshots), id, + delete_leaves, delete_interior); + rcu_read_unlock(); + return ret; +} - ret = snapshot_list_add_nodup(c, equiv_seen, equiv); - if (ret) - return ret; +static int delete_dead_snapshots_process_key(struct btree_trans *trans, + struct btree_iter *iter, + struct bkey_s_c k, + snapshot_id_list *delete_leaves, + interior_delete_list *delete_interior) +{ + if (snapshot_list_has_id(delete_leaves, k.k->p.snapshot)) + return bch2_btree_delete_at(trans, iter, + BTREE_UPDATE_internal_snapshot_node); - /* - * When we have a linear chain of snapshot nodes, we consider - * those to form an equivalence class: we're going to collapse - * them all down to a single node, and keep the leaf-most node - - * which has the same id as the equivalence class id. - * - * If there are multiple keys in different snapshots at the same - * position, we're only going to keep the one in the newest - * snapshot (we delete the others above) - the rest have been - * overwritten and are redundant, and for the key we're going to keep we - * need to move it to the equivalance class ID if it's not there - * already. - */ - if (equiv != k.k->p.snapshot) { + u32 live_child = interior_delete_has_id(delete_interior, k.k->p.snapshot); + if (live_child) { struct bkey_i *new = bch2_bkey_make_mut_noupdate(trans, k); int ret = PTR_ERR_OR_ZERO(new); if (ret) return ret; - new->k.p.snapshot = equiv; - - struct btree_iter new_iter; - bch2_trans_iter_init(trans, &new_iter, iter->btree_id, new->k.p, - BTREE_ITER_all_snapshots| - BTREE_ITER_cached| - BTREE_ITER_intent); + new->k.p.snapshot = live_child; - ret = bch2_btree_iter_traverse(&new_iter) ?: - bch2_trans_update(trans, &new_iter, new, - BTREE_UPDATE_internal_snapshot_node) ?: - bch2_btree_delete_at(trans, iter, - BTREE_UPDATE_internal_snapshot_node); - bch2_trans_iter_exit(trans, &new_iter); + struct btree_iter dst_iter; + struct bkey_s_c dst_k = bch2_bkey_get_iter(trans, &dst_iter, + iter->btree_id, new->k.p, + BTREE_ITER_all_snapshots| + BTREE_ITER_intent); + ret = bkey_err(dst_k); if (ret) return ret; + + ret = (bkey_deleted(dst_k.k) + ? bch2_trans_update(trans, &dst_iter, new, + BTREE_UPDATE_internal_snapshot_node) + : 0) ?: + bch2_btree_delete_at(trans, iter, + BTREE_UPDATE_internal_snapshot_node); + bch2_trans_iter_exit(trans, &dst_iter); + return ret; } return 0; } -static int bch2_snapshot_needs_delete(struct btree_trans *trans, struct bkey_s_c k) +/* + * For a given snapshot, if it doesn't have a subvolume that points to it, and + * it doesn't have child snapshot nodes - it's now redundant and we can mark it + * as deleted. + */ +static int check_should_delete_snapshot(struct btree_trans *trans, struct bkey_s_c k, + snapshot_id_list *delete_leaves, + interior_delete_list *delete_interior) { - struct bkey_s_c_snapshot snap; - u32 children[2]; - int ret; - if (k.k->type != KEY_TYPE_snapshot) return 0; - snap = bkey_s_c_to_snapshot(k); - if (BCH_SNAPSHOT_DELETED(snap.v) || - BCH_SNAPSHOT_SUBVOL(snap.v)) + struct bch_fs *c = trans->c; + struct bkey_s_c_snapshot s = bkey_s_c_to_snapshot(k); + unsigned live_children = 0; + + if (BCH_SNAPSHOT_SUBVOL(s.v)) return 0; - children[0] = le32_to_cpu(snap.v->children[0]); - children[1] = le32_to_cpu(snap.v->children[1]); + for (unsigned i = 0; i < 2; i++) { + u32 child = le32_to_cpu(s.v->children[i]); - ret = bch2_snapshot_live(trans, children[0]) ?: - bch2_snapshot_live(trans, children[1]); - if (ret < 0) - return ret; - return !ret; -} + live_children += child && + !snapshot_list_has_id(delete_leaves, child); + } -/* - * For a given snapshot, if it doesn't have a subvolume that points to it, and - * it doesn't have child snapshot nodes - it's now redundant and we can mark it - * as deleted. - */ -static int bch2_delete_redundant_snapshot(struct btree_trans *trans, struct bkey_s_c k) -{ - int ret = bch2_snapshot_needs_delete(trans, k); + if (live_children == 0) { + return snapshot_list_add(c, delete_leaves, s.k->p.offset); + } else if (live_children == 1) { + struct snapshot_interior_delete d = { + .id = s.k->p.offset, + .live_child = live_child(c, s.k->p.offset, delete_leaves, delete_interior), + }; + + if (!d.live_child) { + bch_err(c, "error finding live child of snapshot %u", d.id); + return -EINVAL; + } - return ret <= 0 - ? ret - : bch2_snapshot_node_set_deleted(trans, k.k->p.offset); + return darray_push(delete_interior, d); + } else { + return 0; + } } static inline u32 bch2_snapshot_nth_parent_skip(struct bch_fs *c, u32 id, u32 n, - snapshot_id_list *skip) + interior_delete_list *skip) { rcu_read_lock(); - while (snapshot_list_has_id(skip, id)) + while (interior_delete_has_id(skip, id)) id = __bch2_snapshot_parent(c, id); while (n--) { do { id = __bch2_snapshot_parent(c, id); - } while (snapshot_list_has_id(skip, id)); + } while (interior_delete_has_id(skip, id)); } rcu_read_unlock(); @@ -1534,7 +1493,7 @@ static inline u32 bch2_snapshot_nth_parent_skip(struct bch_fs *c, u32 id, u32 n, static int bch2_fix_child_of_deleted_snapshot(struct btree_trans *trans, struct btree_iter *iter, struct bkey_s_c k, - snapshot_id_list *deleted) + interior_delete_list *deleted) { struct bch_fs *c = trans->c; u32 nr_deleted_ancestors = 0; @@ -1544,7 +1503,7 @@ static int bch2_fix_child_of_deleted_snapshot(struct btree_trans *trans, if (k.k->type != KEY_TYPE_snapshot) return 0; - if (snapshot_list_has_id(deleted, k.k->p.offset)) + if (interior_delete_has_id(deleted, k.k->p.offset)) return 0; s = bch2_bkey_make_mut_noupdate_typed(trans, k, snapshot); @@ -1553,7 +1512,7 @@ static int bch2_fix_child_of_deleted_snapshot(struct btree_trans *trans, return ret; darray_for_each(*deleted, i) - nr_deleted_ancestors += bch2_snapshot_is_ancestor(c, s->k.p.offset, *i); + nr_deleted_ancestors += bch2_snapshot_is_ancestor(c, s->k.p.offset, i->id); if (!nr_deleted_ancestors) return 0; @@ -1571,7 +1530,7 @@ static int bch2_fix_child_of_deleted_snapshot(struct btree_trans *trans, for (unsigned j = 0; j < ARRAY_SIZE(s->v.skip); j++) { u32 id = le32_to_cpu(s->v.skip[j]); - if (snapshot_list_has_id(deleted, id)) { + if (interior_delete_has_id(deleted, id)) { id = bch2_snapshot_nth_parent_skip(c, parent, depth > 1 @@ -1590,51 +1549,45 @@ static int bch2_fix_child_of_deleted_snapshot(struct btree_trans *trans, int bch2_delete_dead_snapshots(struct bch_fs *c) { - struct btree_trans *trans; - snapshot_id_list deleted = { 0 }; - snapshot_id_list deleted_interior = { 0 }; - int ret = 0; - if (!test_and_clear_bit(BCH_FS_need_delete_dead_snapshots, &c->flags)) return 0; - trans = bch2_trans_get(c); + struct btree_trans *trans = bch2_trans_get(c); + snapshot_id_list delete_leaves = {}; + interior_delete_list delete_interior = {}; + int ret = 0; /* * For every snapshot node: If we have no live children and it's not * pointed to by a subvolume, delete it: */ - ret = for_each_btree_key_commit(trans, iter, BTREE_ID_snapshots, - POS_MIN, 0, k, - NULL, NULL, 0, - bch2_delete_redundant_snapshot(trans, k)); - bch_err_msg(c, ret, "deleting redundant snapshots"); + ret = for_each_btree_key(trans, iter, BTREE_ID_snapshots, POS_MIN, 0, k, + check_should_delete_snapshot(trans, k, &delete_leaves, &delete_interior)); + if (!bch2_err_matches(ret, EROFS)) + bch_err_msg(c, ret, "walking snapshots"); if (ret) goto err; - ret = for_each_btree_key(trans, iter, BTREE_ID_snapshots, - POS_MIN, 0, k, - bch2_snapshot_set_equiv(trans, k)); - bch_err_msg(c, ret, "in bch2_snapshots_set_equiv"); - if (ret) + if (!delete_leaves.nr && !delete_interior.nr) goto err; - ret = for_each_btree_key(trans, iter, BTREE_ID_snapshots, - POS_MIN, 0, k, ({ - if (k.k->type != KEY_TYPE_snapshot) - continue; + { + struct printbuf buf = PRINTBUF; + prt_printf(&buf, "deleting leaves"); + darray_for_each(delete_leaves, i) + prt_printf(&buf, " %u", *i); - BCH_SNAPSHOT_DELETED(bkey_s_c_to_snapshot(k).v) - ? snapshot_list_add(c, &deleted, k.k->p.offset) - : 0; - })); - bch_err_msg(c, ret, "walking snapshots"); - if (ret) - goto err; + prt_printf(&buf, " interior"); + darray_for_each(delete_interior, i) + prt_printf(&buf, " %u->%u", i->id, i->live_child); + + ret = commit_do(trans, NULL, NULL, 0, bch2_trans_log_msg(trans, &buf)); + printbuf_exit(&buf); + if (ret) + goto err; + } for (unsigned btree = 0; btree < BTREE_ID_NR; btree++) { - struct bpos last_pos = POS_MIN; - snapshot_id_list equiv_seen = { 0 }; struct disk_reservation res = { 0 }; if (!btree_type_has_snapshots(btree)) @@ -1644,33 +1597,26 @@ int bch2_delete_dead_snapshots(struct bch_fs *c) btree, POS_MIN, BTREE_ITER_prefetch|BTREE_ITER_all_snapshots, k, &res, NULL, BCH_TRANS_COMMIT_no_enospc, - delete_dead_snapshots_process_key(trans, &iter, k, &deleted, - &equiv_seen, &last_pos)); + delete_dead_snapshots_process_key(trans, &iter, k, + &delete_leaves, + &delete_interior)); bch2_disk_reservation_put(c, &res); - darray_exit(&equiv_seen); - bch_err_msg(c, ret, "deleting keys from dying snapshots"); + if (!bch2_err_matches(ret, EROFS)) + bch_err_msg(c, ret, "deleting keys from dying snapshots"); if (ret) goto err; } - bch2_trans_unlock(trans); - down_write(&c->snapshot_create_lock); - - ret = for_each_btree_key(trans, iter, BTREE_ID_snapshots, - POS_MIN, 0, k, ({ - u32 snapshot = k.k->p.offset; - u32 equiv = bch2_snapshot_equiv(c, snapshot); - - equiv != snapshot - ? snapshot_list_add(c, &deleted_interior, snapshot) - : 0; - })); - - bch_err_msg(c, ret, "walking snapshots"); - if (ret) - goto err_create_lock; + darray_for_each(delete_leaves, i) { + ret = commit_do(trans, NULL, NULL, 0, + bch2_snapshot_node_delete(trans, *i)); + if (!bch2_err_matches(ret, EROFS)) + bch_err_msg(c, ret, "deleting snapshot %u", *i); + if (ret) + goto err; + } /* * Fixing children of deleted snapshots can't be done completely @@ -1680,32 +1626,24 @@ int bch2_delete_dead_snapshots(struct bch_fs *c) ret = for_each_btree_key_commit(trans, iter, BTREE_ID_snapshots, POS_MIN, BTREE_ITER_intent, k, NULL, NULL, BCH_TRANS_COMMIT_no_enospc, - bch2_fix_child_of_deleted_snapshot(trans, &iter, k, &deleted_interior)); + bch2_fix_child_of_deleted_snapshot(trans, &iter, k, &delete_interior)); if (ret) - goto err_create_lock; - - darray_for_each(deleted, i) { - ret = commit_do(trans, NULL, NULL, 0, - bch2_snapshot_node_delete(trans, *i)); - bch_err_msg(c, ret, "deleting snapshot %u", *i); - if (ret) - goto err_create_lock; - } + goto err; - darray_for_each(deleted_interior, i) { + darray_for_each(delete_interior, i) { ret = commit_do(trans, NULL, NULL, 0, - bch2_snapshot_node_delete(trans, *i)); - bch_err_msg(c, ret, "deleting snapshot %u", *i); + bch2_snapshot_node_delete(trans, i->id)); + if (!bch2_err_matches(ret, EROFS)) + bch_err_msg(c, ret, "deleting snapshot %u", i->id); if (ret) - goto err_create_lock; + goto err; } -err_create_lock: - up_write(&c->snapshot_create_lock); err: - darray_exit(&deleted_interior); - darray_exit(&deleted); + darray_exit(&delete_interior); + darray_exit(&delete_leaves); bch2_trans_put(trans); - bch_err_fn(c, ret); + if (!bch2_err_matches(ret, EROFS)) + bch_err_fn(c, ret); return ret; } @@ -1721,8 +1659,12 @@ void bch2_delete_dead_snapshots_work(struct work_struct *work) void bch2_delete_dead_snapshots_async(struct bch_fs *c) { - if (bch2_write_ref_tryget(c, BCH_WRITE_REF_delete_dead_snapshots) && - !queue_work(c->write_ref_wq, &c->snapshot_delete_work)) + if (!bch2_write_ref_tryget(c, BCH_WRITE_REF_delete_dead_snapshots)) + return; + + BUG_ON(!test_bit(BCH_FS_may_go_rw, &c->flags)); + + if (!queue_work(c->write_ref_wq, &c->snapshot_delete_work)) bch2_write_ref_put(c, BCH_WRITE_REF_delete_dead_snapshots); } @@ -1735,18 +1677,10 @@ int __bch2_key_has_snapshot_overwrites(struct btree_trans *trans, struct bkey_s_c k; int ret; - bch2_trans_iter_init(trans, &iter, id, pos, - BTREE_ITER_not_extents| - BTREE_ITER_all_snapshots); - while (1) { - k = bch2_btree_iter_prev(&iter); - ret = bkey_err(k); - if (ret) - break; - - if (!k.k) - break; - + for_each_btree_key_reverse_norestart(trans, iter, id, bpos_predecessor(pos), + BTREE_ITER_not_extents| + BTREE_ITER_all_snapshots, + k, ret) { if (!bkey_eq(pos, k.k->p)) break; @@ -1760,37 +1694,36 @@ int __bch2_key_has_snapshot_overwrites(struct btree_trans *trans, return ret; } -static int bch2_check_snapshot_needs_deletion(struct btree_trans *trans, struct bkey_s_c k) +static bool interior_snapshot_needs_delete(struct bkey_s_c_snapshot snap) { - struct bch_fs *c = trans->c; - struct bkey_s_c_snapshot snap; - int ret = 0; + /* If there's one child, it's redundant and keys will be moved to the child */ + return !!snap.v->children[0] + !!snap.v->children[1] == 1; +} +static int bch2_check_snapshot_needs_deletion(struct btree_trans *trans, struct bkey_s_c k) +{ if (k.k->type != KEY_TYPE_snapshot) return 0; - snap = bkey_s_c_to_snapshot(k); + struct bkey_s_c_snapshot snap = bkey_s_c_to_snapshot(k); if (BCH_SNAPSHOT_DELETED(snap.v) || - bch2_snapshot_equiv(c, k.k->p.offset) != k.k->p.offset || - (ret = bch2_snapshot_needs_delete(trans, k)) > 0) { - set_bit(BCH_FS_need_delete_dead_snapshots, &c->flags); - return 0; - } + interior_snapshot_needs_delete(snap)) + set_bit(BCH_FS_need_delete_dead_snapshots, &trans->c->flags); - return ret; + return 0; } int bch2_snapshots_read(struct bch_fs *c) { + /* + * Initializing the is_ancestor bitmaps requires ancestors to already be + * initialized - so mark in reverse: + */ int ret = bch2_trans_run(c, - for_each_btree_key(trans, iter, BTREE_ID_snapshots, - POS_MIN, 0, k, + for_each_btree_key_reverse(trans, iter, BTREE_ID_snapshots, + POS_MAX, 0, k, __bch2_mark_snapshot(trans, BTREE_ID_snapshots, 0, bkey_s_c_null, k, 0) ?: - bch2_snapshot_set_equiv(trans, k) ?: - bch2_check_snapshot_needs_deletion(trans, k)) ?: - for_each_btree_key(trans, iter, BTREE_ID_snapshots, - POS_MIN, 0, k, - (set_is_ancestor_bitmap(c, k.k->p.offset), 0))); + bch2_check_snapshot_needs_deletion(trans, k))); bch_err_fn(c, ret); /* diff --git a/fs/bcachefs/snapshot.h b/fs/bcachefs/snapshot.h index 29c94716293e..00373cf32e7b 100644 --- a/fs/bcachefs/snapshot.h +++ b/fs/bcachefs/snapshot.h @@ -2,11 +2,9 @@ #ifndef _BCACHEFS_SNAPSHOT_H #define _BCACHEFS_SNAPSHOT_H -enum bch_validate_flags; - void bch2_snapshot_tree_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c); int bch2_snapshot_tree_validate(struct bch_fs *, struct bkey_s_c, - enum bch_validate_flags); + struct bkey_validate_context); #define bch2_bkey_ops_snapshot_tree ((struct bkey_ops) { \ .key_validate = bch2_snapshot_tree_validate, \ @@ -19,7 +17,8 @@ struct bkey_i_snapshot_tree *__bch2_snapshot_tree_create(struct btree_trans *); int bch2_snapshot_tree_lookup(struct btree_trans *, u32, struct bch_snapshot_tree *); void bch2_snapshot_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c); -int bch2_snapshot_validate(struct bch_fs *, struct bkey_s_c, enum bch_validate_flags); +int bch2_snapshot_validate(struct bch_fs *, struct bkey_s_c, + struct bkey_validate_context); int bch2_mark_snapshot(struct btree_trans *, enum btree_id, unsigned, struct bkey_s_c, struct bkey_s, enum btree_iter_update_trigger_flags); @@ -120,19 +119,19 @@ static inline u32 bch2_snapshot_root(struct bch_fs *c, u32 id) return id; } -static inline u32 __bch2_snapshot_equiv(struct bch_fs *c, u32 id) +static inline bool __bch2_snapshot_exists(struct bch_fs *c, u32 id) { const struct snapshot_t *s = snapshot_t(c, id); - return s ? s->equiv : 0; + return s ? s->live : 0; } -static inline u32 bch2_snapshot_equiv(struct bch_fs *c, u32 id) +static inline bool bch2_snapshot_exists(struct bch_fs *c, u32 id) { rcu_read_lock(); - id = __bch2_snapshot_equiv(c, id); + bool ret = __bch2_snapshot_exists(c, id); rcu_read_unlock(); - return id; + return ret; } static inline int bch2_snapshot_is_internal_node(struct bch_fs *c, u32 id) diff --git a/fs/bcachefs/str_hash.c b/fs/bcachefs/str_hash.c new file mode 100644 index 000000000000..8c2c5539de2e --- /dev/null +++ b/fs/bcachefs/str_hash.c @@ -0,0 +1,295 @@ +// SPDX-License-Identifier: GPL-2.0 + +#include "bcachefs.h" +#include "btree_cache.h" +#include "btree_update.h" +#include "dirent.h" +#include "fsck.h" +#include "str_hash.h" +#include "subvolume.h" + +static int bch2_dirent_has_target(struct btree_trans *trans, struct bkey_s_c_dirent d) +{ + if (d.v->d_type == DT_SUBVOL) { + struct bch_subvolume subvol; + int ret = bch2_subvolume_get(trans, le32_to_cpu(d.v->d_child_subvol), + false, &subvol); + if (ret && !bch2_err_matches(ret, ENOENT)) + return ret; + return !ret; + } else { + struct btree_iter iter; + struct bkey_s_c k = bch2_bkey_get_iter(trans, &iter, BTREE_ID_inodes, + SPOS(0, le64_to_cpu(d.v->d_inum), d.k->p.snapshot), 0); + int ret = bkey_err(k); + if (ret) + return ret; + + ret = bkey_is_inode(k.k); + bch2_trans_iter_exit(trans, &iter); + return ret; + } +} + +static int fsck_rename_dirent(struct btree_trans *trans, + struct snapshots_seen *s, + const struct bch_hash_desc desc, + struct bch_hash_info *hash_info, + struct bkey_s_c_dirent old) +{ + struct qstr old_name = bch2_dirent_get_name(old); + struct bkey_i_dirent *new = bch2_trans_kmalloc(trans, bkey_bytes(old.k) + 32); + int ret = PTR_ERR_OR_ZERO(new); + if (ret) + return ret; + + bkey_dirent_init(&new->k_i); + dirent_copy_target(new, old); + new->k.p = old.k->p; + + for (unsigned i = 0; i < 1000; i++) { + unsigned len = sprintf(new->v.d_name, "%.*s.fsck_renamed-%u", + old_name.len, old_name.name, i); + unsigned u64s = BKEY_U64s + dirent_val_u64s(len); + + if (u64s > U8_MAX) + return -EINVAL; + + new->k.u64s = u64s; + + ret = bch2_hash_set_in_snapshot(trans, bch2_dirent_hash_desc, hash_info, + (subvol_inum) { 0, old.k->p.inode }, + old.k->p.snapshot, &new->k_i, + BTREE_UPDATE_internal_snapshot_node); + if (!bch2_err_matches(ret, EEXIST)) + break; + } + + if (ret) + return ret; + + return bch2_fsck_update_backpointers(trans, s, desc, hash_info, &new->k_i); +} + +static int hash_pick_winner(struct btree_trans *trans, + const struct bch_hash_desc desc, + struct bch_hash_info *hash_info, + struct bkey_s_c k1, + struct bkey_s_c k2) +{ + if (bkey_val_bytes(k1.k) == bkey_val_bytes(k2.k) && + !memcmp(k1.v, k2.v, bkey_val_bytes(k1.k))) + return 0; + + switch (desc.btree_id) { + case BTREE_ID_dirents: { + int ret = bch2_dirent_has_target(trans, bkey_s_c_to_dirent(k1)); + if (ret < 0) + return ret; + if (!ret) + return 0; + + ret = bch2_dirent_has_target(trans, bkey_s_c_to_dirent(k2)); + if (ret < 0) + return ret; + if (!ret) + return 1; + return 2; + } + default: + return 0; + } +} + +static int repair_inode_hash_info(struct btree_trans *trans, + struct bch_inode_unpacked *snapshot_root) +{ + struct btree_iter iter; + struct bkey_s_c k; + int ret = 0; + + for_each_btree_key_reverse_norestart(trans, iter, BTREE_ID_inodes, + SPOS(0, snapshot_root->bi_inum, snapshot_root->bi_snapshot - 1), + BTREE_ITER_all_snapshots, k, ret) { + if (k.k->p.offset != snapshot_root->bi_inum) + break; + if (!bkey_is_inode(k.k)) + continue; + + struct bch_inode_unpacked inode; + ret = bch2_inode_unpack(k, &inode); + if (ret) + break; + + if (fsck_err_on(inode.bi_hash_seed != snapshot_root->bi_hash_seed || + INODE_STR_HASH(&inode) != INODE_STR_HASH(snapshot_root), + trans, inode_snapshot_mismatch, + "inode hash info in different snapshots don't match")) { + inode.bi_hash_seed = snapshot_root->bi_hash_seed; + SET_INODE_STR_HASH(&inode, INODE_STR_HASH(snapshot_root)); + ret = __bch2_fsck_write_inode(trans, &inode) ?: + bch2_trans_commit(trans, NULL, NULL, BCH_TRANS_COMMIT_no_enospc) ?: + -BCH_ERR_transaction_restart_nested; + break; + } + } +fsck_err: + bch2_trans_iter_exit(trans, &iter); + return ret; +} + +/* + * All versions of the same inode in different snapshots must have the same hash + * seed/type: verify that the hash info we're using matches the root + */ +static int check_inode_hash_info_matches_root(struct btree_trans *trans, u64 inum, + struct bch_hash_info *hash_info) +{ + struct bch_fs *c = trans->c; + struct btree_iter iter; + struct bkey_s_c k; + int ret = 0; + + for_each_btree_key_reverse_norestart(trans, iter, BTREE_ID_inodes, SPOS(0, inum, U32_MAX), + BTREE_ITER_all_snapshots, k, ret) { + if (k.k->p.offset != inum) + break; + if (bkey_is_inode(k.k)) + goto found; + } + bch_err(c, "%s(): inum %llu not found", __func__, inum); + ret = -BCH_ERR_fsck_repair_unimplemented; + goto err; +found:; + struct bch_inode_unpacked inode; + ret = bch2_inode_unpack(k, &inode); + if (ret) + goto err; + + struct bch_hash_info hash2 = bch2_hash_info_init(c, &inode); + if (hash_info->type != hash2.type || + memcmp(&hash_info->siphash_key, &hash2.siphash_key, sizeof(hash2.siphash_key))) { + ret = repair_inode_hash_info(trans, &inode); + if (!ret) { + bch_err(c, "inode hash info mismatch with root, but mismatch not found\n" + "%u %llx %llx\n" + "%u %llx %llx", + hash_info->type, + hash_info->siphash_key.k0, + hash_info->siphash_key.k1, + hash2.type, + hash2.siphash_key.k0, + hash2.siphash_key.k1); + ret = -BCH_ERR_fsck_repair_unimplemented; + } + } +err: + bch2_trans_iter_exit(trans, &iter); + return ret; +} + +int __bch2_str_hash_check_key(struct btree_trans *trans, + struct snapshots_seen *s, + const struct bch_hash_desc *desc, + struct bch_hash_info *hash_info, + struct btree_iter *k_iter, struct bkey_s_c hash_k) +{ + struct bch_fs *c = trans->c; + struct btree_iter iter = { NULL }; + struct printbuf buf = PRINTBUF; + struct bkey_s_c k; + int ret = 0; + + u64 hash = desc->hash_bkey(hash_info, hash_k); + if (hash_k.k->p.offset < hash) + goto bad_hash; + + for_each_btree_key_norestart(trans, iter, desc->btree_id, + SPOS(hash_k.k->p.inode, hash, hash_k.k->p.snapshot), + BTREE_ITER_slots, k, ret) { + if (bkey_eq(k.k->p, hash_k.k->p)) + break; + + if (k.k->type == desc->key_type && + !desc->cmp_bkey(k, hash_k)) + goto duplicate_entries; + + if (bkey_deleted(k.k)) { + bch2_trans_iter_exit(trans, &iter); + goto bad_hash; + } + } +out: + bch2_trans_iter_exit(trans, &iter); + printbuf_exit(&buf); + return ret; +bad_hash: + /* + * Before doing any repair, check hash_info itself: + */ + ret = check_inode_hash_info_matches_root(trans, hash_k.k->p.inode, hash_info); + if (ret) + goto out; + + if (fsck_err(trans, hash_table_key_wrong_offset, + "hash table key at wrong offset: btree %s inode %llu offset %llu, hashed to %llu\n %s", + bch2_btree_id_str(desc->btree_id), hash_k.k->p.inode, hash_k.k->p.offset, hash, + (printbuf_reset(&buf), + bch2_bkey_val_to_text(&buf, c, hash_k), buf.buf))) { + struct bkey_i *new = bch2_bkey_make_mut_noupdate(trans, hash_k); + if (IS_ERR(new)) + return PTR_ERR(new); + + k = bch2_hash_set_or_get_in_snapshot(trans, &iter, *desc, hash_info, + (subvol_inum) { 0, hash_k.k->p.inode }, + hash_k.k->p.snapshot, new, + STR_HASH_must_create| + BTREE_ITER_with_updates| + BTREE_UPDATE_internal_snapshot_node); + ret = bkey_err(k); + if (ret) + goto out; + if (k.k) + goto duplicate_entries; + + ret = bch2_hash_delete_at(trans, *desc, hash_info, k_iter, + BTREE_UPDATE_internal_snapshot_node) ?: + bch2_fsck_update_backpointers(trans, s, *desc, hash_info, new) ?: + bch2_trans_commit(trans, NULL, NULL, BCH_TRANS_COMMIT_no_enospc) ?: + -BCH_ERR_transaction_restart_nested; + goto out; + } +fsck_err: + goto out; +duplicate_entries: + ret = hash_pick_winner(trans, *desc, hash_info, hash_k, k); + if (ret < 0) + goto out; + + if (!fsck_err(trans, hash_table_key_duplicate, + "duplicate hash table keys%s:\n%s", + ret != 2 ? "" : ", both point to valid inodes", + (printbuf_reset(&buf), + bch2_bkey_val_to_text(&buf, c, hash_k), + prt_newline(&buf), + bch2_bkey_val_to_text(&buf, c, k), + buf.buf))) + goto out; + + switch (ret) { + case 0: + ret = bch2_hash_delete_at(trans, *desc, hash_info, k_iter, 0); + break; + case 1: + ret = bch2_hash_delete_at(trans, *desc, hash_info, &iter, 0); + break; + case 2: + ret = fsck_rename_dirent(trans, s, *desc, hash_info, bkey_s_c_to_dirent(hash_k)) ?: + bch2_hash_delete_at(trans, *desc, hash_info, k_iter, 0); + goto out; + } + + ret = bch2_trans_commit(trans, NULL, NULL, 0) ?: + -BCH_ERR_transaction_restart_nested; + goto out; +} diff --git a/fs/bcachefs/str_hash.h b/fs/bcachefs/str_hash.h index ec2b1feea520..55a4ac7bf220 100644 --- a/fs/bcachefs/str_hash.h +++ b/fs/bcachefs/str_hash.h @@ -160,7 +160,7 @@ bch2_hash_lookup_in_snapshot(struct btree_trans *trans, struct bkey_s_c k; int ret; - for_each_btree_key_upto_norestart(trans, *iter, desc.btree_id, + for_each_btree_key_max_norestart(trans, *iter, desc.btree_id, SPOS(inum.inum, desc.hash_key(info, key), snapshot), POS(inum.inum, U64_MAX), BTREE_ITER_slots|flags, k, ret) { @@ -210,7 +210,7 @@ bch2_hash_hole(struct btree_trans *trans, if (ret) return ret; - for_each_btree_key_upto_norestart(trans, *iter, desc.btree_id, + for_each_btree_key_max_norestart(trans, *iter, desc.btree_id, SPOS(inum.inum, desc.hash_key(info, key), snapshot), POS(inum.inum, U64_MAX), BTREE_ITER_slots|BTREE_ITER_intent, k, ret) @@ -265,7 +265,7 @@ struct bkey_s_c bch2_hash_set_or_get_in_snapshot(struct btree_trans *trans, bool found = false; int ret; - for_each_btree_key_upto_norestart(trans, *iter, desc.btree_id, + for_each_btree_key_max_norestart(trans, *iter, desc.btree_id, SPOS(insert->k.p.inode, desc.hash_bkey(info, bkey_i_to_s_c(insert)), snapshot), @@ -393,4 +393,26 @@ int bch2_hash_delete(struct btree_trans *trans, return ret; } +struct snapshots_seen; +int __bch2_str_hash_check_key(struct btree_trans *, + struct snapshots_seen *, + const struct bch_hash_desc *, + struct bch_hash_info *, + struct btree_iter *, struct bkey_s_c); + +static inline int bch2_str_hash_check_key(struct btree_trans *trans, + struct snapshots_seen *s, + const struct bch_hash_desc *desc, + struct bch_hash_info *hash_info, + struct btree_iter *k_iter, struct bkey_s_c hash_k) +{ + if (hash_k.k->type != desc->key_type) + return 0; + + if (likely(desc->hash_bkey(hash_info, hash_k) == hash_k.k->p.offset)) + return 0; + + return __bch2_str_hash_check_key(trans, s, desc, hash_info, k_iter, hash_k); +} + #endif /* _BCACHEFS_STR_HASH_H */ diff --git a/fs/bcachefs/subvolume.c b/fs/bcachefs/subvolume.c index 80e5efaff524..e3d0475232e5 100644 --- a/fs/bcachefs/subvolume.c +++ b/fs/bcachefs/subvolume.c @@ -207,7 +207,7 @@ int bch2_check_subvol_children(struct bch_fs *c) /* Subvolumes: */ int bch2_subvolume_validate(struct bch_fs *c, struct bkey_s_c k, - enum bch_validate_flags flags) + struct bkey_validate_context from) { struct bkey_s_c_subvolume subvol = bkey_s_c_to_subvolume(k); int ret = 0; @@ -286,11 +286,11 @@ int bch2_subvol_has_children(struct btree_trans *trans, u32 subvol) static __always_inline int bch2_subvolume_get_inlined(struct btree_trans *trans, unsigned subvol, bool inconsistent_if_not_found, - int iter_flags, struct bch_subvolume *s) { int ret = bch2_bkey_get_val_typed(trans, BTREE_ID_subvolumes, POS(0, subvol), - iter_flags, subvolume, s); + BTREE_ITER_cached| + BTREE_ITER_with_updates, subvolume, s); bch2_fs_inconsistent_on(bch2_err_matches(ret, ENOENT) && inconsistent_if_not_found, trans->c, "missing subvolume %u", subvol); @@ -299,16 +299,15 @@ bch2_subvolume_get_inlined(struct btree_trans *trans, unsigned subvol, int bch2_subvolume_get(struct btree_trans *trans, unsigned subvol, bool inconsistent_if_not_found, - int iter_flags, struct bch_subvolume *s) { - return bch2_subvolume_get_inlined(trans, subvol, inconsistent_if_not_found, iter_flags, s); + return bch2_subvolume_get_inlined(trans, subvol, inconsistent_if_not_found, s); } int bch2_subvol_is_ro_trans(struct btree_trans *trans, u32 subvol) { struct bch_subvolume s; - int ret = bch2_subvolume_get_inlined(trans, subvol, true, 0, &s); + int ret = bch2_subvolume_get_inlined(trans, subvol, true, &s); if (ret) return ret; @@ -328,7 +327,7 @@ int bch2_snapshot_get_subvol(struct btree_trans *trans, u32 snapshot, struct bch_snapshot snap; return bch2_snapshot_lookup(trans, snapshot, &snap) ?: - bch2_subvolume_get(trans, le32_to_cpu(snap.subvol), true, 0, subvol); + bch2_subvolume_get(trans, le32_to_cpu(snap.subvol), true, subvol); } int __bch2_subvolume_get_snapshot(struct btree_trans *trans, u32 subvolid, @@ -396,8 +395,7 @@ static int bch2_subvolumes_reparent(struct btree_trans *trans, u32 subvolid_to_d struct bch_subvolume s; return lockrestart_do(trans, - bch2_subvolume_get(trans, subvolid_to_delete, true, - BTREE_ITER_cached, &s)) ?: + bch2_subvolume_get(trans, subvolid_to_delete, true, &s)) ?: for_each_btree_key_commit(trans, iter, BTREE_ID_subvolumes, POS_MIN, BTREE_ITER_prefetch, k, NULL, NULL, BCH_TRANS_COMMIT_no_enospc, @@ -411,26 +409,56 @@ static int bch2_subvolumes_reparent(struct btree_trans *trans, u32 subvolid_to_d */ static int __bch2_subvolume_delete(struct btree_trans *trans, u32 subvolid) { - struct btree_iter iter; - struct bkey_s_c_subvolume subvol; - u32 snapid; - int ret = 0; + struct btree_iter subvol_iter = {}, snapshot_iter = {}, snapshot_tree_iter = {}; - subvol = bch2_bkey_get_iter_typed(trans, &iter, + struct bkey_s_c_subvolume subvol = + bch2_bkey_get_iter_typed(trans, &subvol_iter, BTREE_ID_subvolumes, POS(0, subvolid), BTREE_ITER_cached|BTREE_ITER_intent, subvolume); - ret = bkey_err(subvol); + int ret = bkey_err(subvol); bch2_fs_inconsistent_on(bch2_err_matches(ret, ENOENT), trans->c, "missing subvolume %u", subvolid); if (ret) - return ret; + goto err; - snapid = le32_to_cpu(subvol.v->snapshot); + u32 snapid = le32_to_cpu(subvol.v->snapshot); + + struct bkey_s_c_snapshot snapshot = + bch2_bkey_get_iter_typed(trans, &snapshot_iter, + BTREE_ID_snapshots, POS(0, snapid), + 0, snapshot); + ret = bkey_err(subvol); + bch2_fs_inconsistent_on(bch2_err_matches(ret, ENOENT), trans->c, + "missing snapshot %u", snapid); + if (ret) + goto err; + + u32 treeid = le32_to_cpu(snapshot.v->tree); - ret = bch2_btree_delete_at(trans, &iter, 0) ?: + struct bkey_s_c_snapshot_tree snapshot_tree = + bch2_bkey_get_iter_typed(trans, &snapshot_tree_iter, + BTREE_ID_snapshot_trees, POS(0, treeid), + 0, snapshot_tree); + + if (le32_to_cpu(snapshot_tree.v->master_subvol) == subvolid) { + struct bkey_i_snapshot_tree *snapshot_tree_mut = + bch2_bkey_make_mut_typed(trans, &snapshot_tree_iter, + &snapshot_tree.s_c, + 0, snapshot_tree); + ret = PTR_ERR_OR_ZERO(snapshot_tree_mut); + if (ret) + goto err; + + snapshot_tree_mut->v.master_subvol = 0; + } + + ret = bch2_btree_delete_at(trans, &subvol_iter, 0) ?: bch2_snapshot_node_set_deleted(trans, snapid); - bch2_trans_iter_exit(trans, &iter); +err: + bch2_trans_iter_exit(trans, &snapshot_tree_iter); + bch2_trans_iter_exit(trans, &snapshot_iter); + bch2_trans_iter_exit(trans, &subvol_iter); return ret; } @@ -675,7 +703,7 @@ err: /* set bi_subvol on root inode */ int bch2_fs_upgrade_for_subvolumes(struct bch_fs *c) { - int ret = bch2_trans_commit_do(c, NULL, NULL, BCH_TRANS_COMMIT_lazy_rw, + int ret = bch2_trans_commit_do(c, NULL, NULL, BCH_TRANS_COMMIT_no_enospc, __bch2_fs_upgrade_for_subvolumes(trans)); bch_err_fn(c, ret); return ret; diff --git a/fs/bcachefs/subvolume.h b/fs/bcachefs/subvolume.h index f897d106e142..910f6196700e 100644 --- a/fs/bcachefs/subvolume.h +++ b/fs/bcachefs/subvolume.h @@ -5,12 +5,11 @@ #include "darray.h" #include "subvolume_types.h" -enum bch_validate_flags; - int bch2_check_subvols(struct bch_fs *); int bch2_check_subvol_children(struct bch_fs *); -int bch2_subvolume_validate(struct bch_fs *, struct bkey_s_c, enum bch_validate_flags); +int bch2_subvolume_validate(struct bch_fs *, struct bkey_s_c, + struct bkey_validate_context); void bch2_subvolume_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c); int bch2_subvolume_trigger(struct btree_trans *, enum btree_id, unsigned, struct bkey_s_c, struct bkey_s, @@ -25,7 +24,7 @@ int bch2_subvolume_trigger(struct btree_trans *, enum btree_id, unsigned, int bch2_subvol_has_children(struct btree_trans *, u32); int bch2_subvolume_get(struct btree_trans *, unsigned, - bool, int, struct bch_subvolume *); + bool, struct bch_subvolume *); int __bch2_subvolume_get_snapshot(struct btree_trans *, u32, u32 *, bool); int bch2_subvolume_get_snapshot(struct btree_trans *, u32, u32 *); @@ -34,7 +33,7 @@ int bch2_subvol_is_ro_trans(struct btree_trans *, u32); int bch2_subvol_is_ro(struct bch_fs *, u32); static inline struct bkey_s_c -bch2_btree_iter_peek_in_subvolume_upto_type(struct btree_iter *iter, struct bpos end, +bch2_btree_iter_peek_in_subvolume_max_type(struct btree_iter *iter, struct bpos end, u32 subvolid, unsigned flags) { u32 snapshot; @@ -43,10 +42,10 @@ bch2_btree_iter_peek_in_subvolume_upto_type(struct btree_iter *iter, struct bpos return bkey_s_c_err(ret); bch2_btree_iter_set_snapshot(iter, snapshot); - return bch2_btree_iter_peek_upto_type(iter, end, flags); + return bch2_btree_iter_peek_max_type(iter, end, flags); } -#define for_each_btree_key_in_subvolume_upto_continue(_trans, _iter, \ +#define for_each_btree_key_in_subvolume_max_continue(_trans, _iter, \ _end, _subvolid, _flags, _k, _do) \ ({ \ struct bkey_s_c _k; \ @@ -54,7 +53,7 @@ bch2_btree_iter_peek_in_subvolume_upto_type(struct btree_iter *iter, struct bpos \ do { \ _ret3 = lockrestart_do(_trans, ({ \ - (_k) = bch2_btree_iter_peek_in_subvolume_upto_type(&(_iter), \ + (_k) = bch2_btree_iter_peek_in_subvolume_max_type(&(_iter), \ _end, _subvolid, (_flags)); \ if (!(_k).k) \ break; \ @@ -67,14 +66,14 @@ bch2_btree_iter_peek_in_subvolume_upto_type(struct btree_iter *iter, struct bpos _ret3; \ }) -#define for_each_btree_key_in_subvolume_upto(_trans, _iter, _btree_id, \ +#define for_each_btree_key_in_subvolume_max(_trans, _iter, _btree_id, \ _start, _end, _subvolid, _flags, _k, _do) \ ({ \ struct btree_iter _iter; \ bch2_trans_iter_init((_trans), &(_iter), (_btree_id), \ (_start), (_flags)); \ \ - for_each_btree_key_in_subvolume_upto_continue(_trans, _iter, \ + for_each_btree_key_in_subvolume_max_continue(_trans, _iter, \ _end, _subvolid, _flags, _k, _do); \ }) diff --git a/fs/bcachefs/subvolume_types.h b/fs/bcachefs/subvolume_types.h index f2ec4277c2a5..1549d6daf7af 100644 --- a/fs/bcachefs/subvolume_types.h +++ b/fs/bcachefs/subvolume_types.h @@ -9,13 +9,13 @@ typedef DARRAY(u32) snapshot_id_list; #define IS_ANCESTOR_BITMAP 128 struct snapshot_t { + bool live; u32 parent; u32 skip[3]; u32 depth; u32 children[2]; u32 subvol; /* Nonzero only if a subvolume points to this node: */ u32 tree; - u32 equiv; unsigned long is_ancestor[BITS_TO_LONGS(IS_ANCESTOR_BITMAP)]; }; diff --git a/fs/bcachefs/super-io.c b/fs/bcachefs/super-io.c index 7c71594f6a8b..8037ccbacf6a 100644 --- a/fs/bcachefs/super-io.c +++ b/fs/bcachefs/super-io.c @@ -23,6 +23,7 @@ #include <linux/backing-dev.h> #include <linux/sort.h> +#include <linux/string_choices.h> static const struct blk_holder_ops bch2_sb_handle_bdev_ops = { }; @@ -41,7 +42,7 @@ static const struct bch2_metadata_version bch2_metadata_versions[] = { #undef x }; -void bch2_version_to_text(struct printbuf *out, unsigned v) +void bch2_version_to_text(struct printbuf *out, enum bcachefs_metadata_version v) { const char *str = "(unknown version)"; @@ -54,7 +55,7 @@ void bch2_version_to_text(struct printbuf *out, unsigned v) prt_printf(out, "%u.%u: %s", BCH_VERSION_MAJOR(v), BCH_VERSION_MINOR(v), str); } -unsigned bch2_latest_compatible_version(unsigned v) +enum bcachefs_metadata_version bch2_latest_compatible_version(enum bcachefs_metadata_version v) { if (!BCH_VERSION_MAJOR(v)) return v; @@ -68,6 +69,16 @@ unsigned bch2_latest_compatible_version(unsigned v) return v; } +void bch2_set_version_incompat(struct bch_fs *c, enum bcachefs_metadata_version version) +{ + mutex_lock(&c->sb_lock); + SET_BCH_SB_VERSION_INCOMPAT(c->disk_sb.sb, + max(BCH_SB_VERSION_INCOMPAT(c->disk_sb.sb), version)); + c->disk_sb.sb->features[0] |= cpu_to_le64(BCH_FEATURE_incompat_version_field); + bch2_write_super(c); + mutex_unlock(&c->sb_lock); +} + const char * const bch2_sb_fields[] = { #define x(name, nr) #name, BCH_SB_FIELDS() @@ -368,6 +379,12 @@ static int bch2_sb_validate(struct bch_sb_handle *disk_sb, return -BCH_ERR_invalid_sb_features; } + if (BCH_VERSION_MAJOR(le16_to_cpu(sb->version)) > BCH_VERSION_MAJOR(bcachefs_metadata_version_current) || + BCH_SB_VERSION_INCOMPAT(sb) > bcachefs_metadata_version_current) { + prt_printf(out, "Filesystem has incompatible version"); + return -BCH_ERR_invalid_sb_features; + } + block_size = le16_to_cpu(sb->block_size); if (block_size > PAGE_SECTORS) { @@ -406,6 +423,21 @@ static int bch2_sb_validate(struct bch_sb_handle *disk_sb, return -BCH_ERR_invalid_sb_time_precision; } + /* old versions didn't know to downgrade this field */ + if (BCH_SB_VERSION_INCOMPAT_ALLOWED(sb) > le16_to_cpu(sb->version)) + SET_BCH_SB_VERSION_INCOMPAT_ALLOWED(sb, le16_to_cpu(sb->version)); + + if (BCH_SB_VERSION_INCOMPAT(sb) > BCH_SB_VERSION_INCOMPAT_ALLOWED(sb)) { + prt_printf(out, "Invalid version_incompat "); + bch2_version_to_text(out, BCH_SB_VERSION_INCOMPAT(sb)); + prt_str(out, " > incompat_allowed "); + bch2_version_to_text(out, BCH_SB_VERSION_INCOMPAT_ALLOWED(sb)); + if (flags & BCH_VALIDATE_write) + return -BCH_ERR_invalid_sb_version; + else + SET_BCH_SB_VERSION_INCOMPAT_ALLOWED(sb, BCH_SB_VERSION_INCOMPAT(sb)); + } + if (!flags) { /* * Been seeing a bug where these are getting inexplicably @@ -428,6 +460,11 @@ static int bch2_sb_validate(struct bch_sb_handle *disk_sb, SET_BCH_SB_PROMOTE_WHOLE_EXTENTS(sb, true); } +#ifdef __KERNEL__ + if (!BCH_SB_SHARD_INUMS_NBITS(sb)) + SET_BCH_SB_SHARD_INUMS_NBITS(sb, ilog2(roundup_pow_of_two(num_online_cpus()))); +#endif + for (opt_id = 0; opt_id < bch2_opts_nr; opt_id++) { const struct bch_option *opt = bch2_opt_table + opt_id; @@ -519,6 +556,9 @@ static void bch2_sb_update(struct bch_fs *c) c->sb.uuid = src->uuid; c->sb.user_uuid = src->user_uuid; c->sb.version = le16_to_cpu(src->version); + c->sb.version_incompat = BCH_SB_VERSION_INCOMPAT(src); + c->sb.version_incompat_allowed + = BCH_SB_VERSION_INCOMPAT_ALLOWED(src); c->sb.version_min = le16_to_cpu(src->version_min); c->sb.version_upgrade_complete = BCH_SB_VERSION_UPGRADE_COMPLETE(src); c->sb.nr_devices = src->nr_devices; @@ -676,7 +716,8 @@ reread: } enum bch_csum_type csum_type = BCH_SB_CSUM_TYPE(sb->sb); - if (csum_type >= BCH_CSUM_NR) { + if (csum_type >= BCH_CSUM_NR || + bch2_csum_type_is_encryption(csum_type)) { prt_printf(err, "unknown checksum type %llu", BCH_SB_CSUM_TYPE(sb->sb)); return -BCH_ERR_invalid_sb_csum_type; } @@ -878,7 +919,7 @@ static void write_super_endio(struct bio *bio) ? BCH_MEMBER_ERROR_write : BCH_MEMBER_ERROR_read, "superblock %s error: %s", - bio_data_dir(bio) ? "write" : "read", + str_write_read(bio_data_dir(bio)), bch2_blk_status_to_str(bio->bi_status))) ca->sb_write_error = 1; @@ -891,14 +932,15 @@ static void read_back_super(struct bch_fs *c, struct bch_dev *ca) struct bch_sb *sb = ca->disk_sb.sb; struct bio *bio = ca->disk_sb.bio; + memset(ca->sb_read_scratch, 0, BCH_SB_READ_SCRATCH_BUF_SIZE); + bio_reset(bio, ca->disk_sb.bdev, REQ_OP_READ|REQ_SYNC|REQ_META); bio->bi_iter.bi_sector = le64_to_cpu(sb->layout.sb_offset[0]); bio->bi_end_io = write_super_endio; bio->bi_private = ca; - bch2_bio_map(bio, ca->sb_read_scratch, PAGE_SIZE); + bch2_bio_map(bio, ca->sb_read_scratch, BCH_SB_READ_SCRATCH_BUF_SIZE); - this_cpu_add(ca->io_done->sectors[READ][BCH_DATA_sb], - bio_sectors(bio)); + this_cpu_add(ca->io_done->sectors[READ][BCH_DATA_sb], bio_sectors(bio)); percpu_ref_get(&ca->io_ref); closure_bio_submit(bio, &c->sb_write); @@ -1042,9 +1084,16 @@ int bch2_write_super(struct bch_fs *c) ": Superblock write was silently dropped! (seq %llu expected %llu)", le64_to_cpu(ca->sb_read_scratch->seq), ca->disk_sb.seq); - bch2_fs_fatal_error(c, "%s", buf.buf); + + if (c->opts.errors != BCH_ON_ERROR_continue && + c->opts.errors != BCH_ON_ERROR_fix_safe) { + ret = -BCH_ERR_erofs_sb_err; + bch2_fs_fatal_error(c, "%s", buf.buf); + } else { + bch_err(c, "%s", buf.buf); + } + printbuf_exit(&buf); - ret = -BCH_ERR_erofs_sb_err; } if (le64_to_cpu(ca->sb_read_scratch->seq) > ca->disk_sb.seq) { @@ -1149,6 +1198,8 @@ bool bch2_check_version_downgrade(struct bch_fs *c) */ if (BCH_SB_VERSION_UPGRADE_COMPLETE(c->disk_sb.sb) > bcachefs_metadata_version_current) SET_BCH_SB_VERSION_UPGRADE_COMPLETE(c->disk_sb.sb, bcachefs_metadata_version_current); + if (BCH_SB_VERSION_INCOMPAT_ALLOWED(c->disk_sb.sb) > bcachefs_metadata_version_current) + SET_BCH_SB_VERSION_INCOMPAT_ALLOWED(c->disk_sb.sb, bcachefs_metadata_version_current); if (c->sb.version > bcachefs_metadata_version_current) c->disk_sb.sb->version = cpu_to_le16(bcachefs_metadata_version_current); if (c->sb.version_min > bcachefs_metadata_version_current) @@ -1157,7 +1208,7 @@ bool bch2_check_version_downgrade(struct bch_fs *c) return ret; } -void bch2_sb_upgrade(struct bch_fs *c, unsigned new_version) +void bch2_sb_upgrade(struct bch_fs *c, unsigned new_version, bool incompat) { lockdep_assert_held(&c->sb_lock); @@ -1167,6 +1218,10 @@ void bch2_sb_upgrade(struct bch_fs *c, unsigned new_version) c->disk_sb.sb->version = cpu_to_le16(new_version); c->disk_sb.sb->features[0] |= cpu_to_le64(BCH_SB_FEATURES_ALL); + + if (incompat) + SET_BCH_SB_VERSION_INCOMPAT_ALLOWED(c->disk_sb.sb, + max(BCH_SB_VERSION_INCOMPAT_ALLOWED(c->disk_sb.sb), new_version)); } static int bch2_sb_ext_validate(struct bch_sb *sb, struct bch_sb_field *f, @@ -1331,6 +1386,14 @@ void bch2_sb_to_text(struct printbuf *out, struct bch_sb *sb, bch2_version_to_text(out, le16_to_cpu(sb->version)); prt_newline(out); + prt_printf(out, "Incompatible features allowed:\t"); + bch2_version_to_text(out, BCH_SB_VERSION_INCOMPAT_ALLOWED(sb)); + prt_newline(out); + + prt_printf(out, "Incompatible features in use:\t"); + bch2_version_to_text(out, BCH_SB_VERSION_INCOMPAT(sb)); + prt_newline(out); + prt_printf(out, "Version upgrade complete:\t"); bch2_version_to_text(out, BCH_SB_VERSION_UPGRADE_COMPLETE(sb)); prt_newline(out); diff --git a/fs/bcachefs/super-io.h b/fs/bcachefs/super-io.h index fadd364e2802..f1ab4f943720 100644 --- a/fs/bcachefs/super-io.h +++ b/fs/bcachefs/super-io.h @@ -10,14 +10,29 @@ #include <asm/byteorder.h> +#define BCH_SB_READ_SCRATCH_BUF_SIZE 4096 + static inline bool bch2_version_compatible(u16 version) { return BCH_VERSION_MAJOR(version) <= BCH_VERSION_MAJOR(bcachefs_metadata_version_current) && version >= bcachefs_metadata_version_min; } -void bch2_version_to_text(struct printbuf *, unsigned); -unsigned bch2_latest_compatible_version(unsigned); +void bch2_version_to_text(struct printbuf *, enum bcachefs_metadata_version); +enum bcachefs_metadata_version bch2_latest_compatible_version(enum bcachefs_metadata_version); + +void bch2_set_version_incompat(struct bch_fs *, enum bcachefs_metadata_version); + +static inline bool bch2_request_incompat_feature(struct bch_fs *c, + enum bcachefs_metadata_version version) +{ + if (unlikely(version > c->sb.version_incompat)) { + if (version > c->sb.version_incompat_allowed) + return false; + bch2_set_version_incompat(c, version); + } + return true; +} static inline size_t bch2_sb_field_bytes(struct bch_sb_field *f) { @@ -92,7 +107,7 @@ static inline void bch2_check_set_feature(struct bch_fs *c, unsigned feat) } bool bch2_check_version_downgrade(struct bch_fs *); -void bch2_sb_upgrade(struct bch_fs *, unsigned); +void bch2_sb_upgrade(struct bch_fs *, unsigned, bool); void __bch2_sb_field_to_text(struct printbuf *, struct bch_sb *, struct bch_sb_field *); diff --git a/fs/bcachefs/super.c b/fs/bcachefs/super.c index a6ed9a0bf1c7..d97ea7bd1171 100644 --- a/fs/bcachefs/super.c +++ b/fs/bcachefs/super.c @@ -290,7 +290,7 @@ static void __bch2_fs_read_only(struct bch_fs *c) bch2_fs_journal_stop(&c->journal); - bch_info(c, "%sshutdown complete, journal seq %llu", + bch_info(c, "%sclean shutdown complete, journal seq %llu", test_bit(BCH_FS_clean_shutdown, &c->flags) ? "" : "un", c->journal.seq_ondisk); @@ -441,6 +441,8 @@ static int __bch2_fs_read_write(struct bch_fs *c, bool early) { int ret; + BUG_ON(!test_bit(BCH_FS_may_go_rw, &c->flags)); + if (test_bit(BCH_FS_initial_gc_unfixed, &c->flags)) { bch_err(c, "cannot go rw, unfixed btree errors"); return -BCH_ERR_erofs_unfixed_errors; @@ -561,6 +563,7 @@ static void __bch2_fs_free(struct bch_fs *c) bch2_io_clock_exit(&c->io_clock[WRITE]); bch2_io_clock_exit(&c->io_clock[READ]); bch2_fs_compress_exit(c); + bch2_fs_btree_gc_exit(c); bch2_journal_keys_put_initial(c); bch2_find_btree_nodes_exit(&c->found_btree_nodes); BUG_ON(atomic_read(&c->journal_keys.ref)); @@ -584,7 +587,6 @@ static void __bch2_fs_free(struct bch_fs *c) #endif kfree(rcu_dereference_protected(c->disk_groups, 1)); kfree(c->journal_seq_blacklist_table); - kfree(c->unused_inode_hints); if (c->write_ref_wq) destroy_workqueue(c->write_ref_wq); @@ -766,21 +768,17 @@ static struct bch_fs *bch2_fs_alloc(struct bch_sb *sb, struct bch_opts opts) refcount_set(&c->ro_ref, 1); init_waitqueue_head(&c->ro_ref_wait); + spin_lock_init(&c->recovery_pass_lock); sema_init(&c->online_fsck_mutex, 1); - init_rwsem(&c->gc_lock); - mutex_init(&c->gc_gens_lock); - atomic_set(&c->journal_keys.ref, 1); - c->journal_keys.initial_ref_held = true; - for (i = 0; i < BCH_TIME_STAT_NR; i++) bch2_time_stats_init(&c->times[i]); - bch2_fs_gc_init(c); bch2_fs_copygc_init(c); bch2_fs_btree_key_cache_init_early(&c->btree_key_cache); bch2_fs_btree_iter_init_early(c); bch2_fs_btree_interior_update_init_early(c); + bch2_fs_journal_keys_init(c); bch2_fs_allocator_background_init(c); bch2_fs_allocator_foreground_init(c); bch2_fs_rebalance_init(c); @@ -809,9 +807,6 @@ static struct bch_fs *bch2_fs_alloc(struct bch_sb *sb, struct bch_opts opts) INIT_LIST_HEAD(&c->vfs_inodes_list); mutex_init(&c->vfs_inodes_lock); - c->copy_gc_enabled = 1; - c->rebalance.enabled = 1; - c->journal.flush_write_time = &c->times[BCH_TIME_journal_flush_write]; c->journal.noflush_write_time = &c->times[BCH_TIME_journal_noflush_write]; c->journal.flush_seq_time = &c->times[BCH_TIME_journal_flush_seq]; @@ -873,8 +868,6 @@ static struct bch_fs *bch2_fs_alloc(struct bch_sb *sb, struct bch_opts opts) (btree_blocks(c) + 1) * 2 * sizeof(struct sort_iter_set); - c->inode_shard_bits = ilog2(roundup_pow_of_two(num_possible_cpus())); - if (!(c->btree_update_wq = alloc_workqueue("bcachefs", WQ_HIGHPRI|WQ_FREEZABLE|WQ_MEM_RECLAIM|WQ_UNBOUND, 512)) || !(c->btree_io_complete_wq = alloc_workqueue("bcachefs_btree_io", @@ -901,9 +894,7 @@ static struct bch_fs *bch2_fs_alloc(struct bch_sb *sb, struct bch_opts opts) !(c->online_reserved = alloc_percpu(u64)) || mempool_init_kvmalloc_pool(&c->btree_bounce_pool, 1, c->opts.btree_node_size) || - mempool_init_kmalloc_pool(&c->large_bkey_pool, 1, 2048) || - !(c->unused_inode_hints = kcalloc(1U << c->inode_shard_bits, - sizeof(u64), GFP_KERNEL))) { + mempool_init_kmalloc_pool(&c->large_bkey_pool, 1, 2048)) { ret = -BCH_ERR_ENOMEM_fs_other_alloc; goto err; } @@ -917,6 +908,7 @@ static struct bch_fs *bch2_fs_alloc(struct bch_sb *sb, struct bch_opts opts) bch2_fs_btree_cache_init(c) ?: bch2_fs_btree_key_cache_init(&c->btree_key_cache) ?: bch2_fs_btree_interior_update_init(c) ?: + bch2_fs_btree_gc_init(c) ?: bch2_fs_buckets_waiting_for_journal_init(c) ?: bch2_fs_btree_write_buffer_init(c) ?: bch2_fs_subvolumes_init(c) ?: @@ -1033,9 +1025,12 @@ int bch2_fs_start(struct bch_fs *c) bch2_dev_allocator_add(c, ca); bch2_recalc_capacity(c); + c->recovery_task = current; ret = BCH_SB_INITIALIZED(c->disk_sb.sb) ? bch2_fs_recovery(c) : bch2_fs_initialize(c); + c->recovery_task = NULL; + if (ret) goto err; @@ -1120,12 +1115,12 @@ static int bch2_dev_in_fs(struct bch_sb_handle *fs, prt_bdevname(&buf, fs->bdev); prt_char(&buf, ' '); - bch2_prt_datetime(&buf, le64_to_cpu(fs->sb->write_time));; + bch2_prt_datetime(&buf, le64_to_cpu(fs->sb->write_time)); prt_newline(&buf); prt_bdevname(&buf, sb->bdev); prt_char(&buf, ' '); - bch2_prt_datetime(&buf, le64_to_cpu(sb->sb->write_time));; + bch2_prt_datetime(&buf, le64_to_cpu(sb->sb->write_time)); prt_newline(&buf); if (!opts->no_splitbrain_check) @@ -1198,7 +1193,7 @@ static void bch2_dev_free(struct bch_dev *ca) free_percpu(ca->io_done); bch2_dev_buckets_free(ca); - free_page((unsigned long) ca->sb_read_scratch); + kfree(ca->sb_read_scratch); bch2_time_stats_quantiles_exit(&ca->io_latency[WRITE]); bch2_time_stats_quantiles_exit(&ca->io_latency[READ]); @@ -1309,8 +1304,6 @@ static struct bch_dev *__bch2_dev_alloc(struct bch_fs *c, init_completion(&ca->ref_completion); init_completion(&ca->io_ref_completion); - init_rwsem(&ca->bucket_lock); - INIT_WORK(&ca->io_error_work, bch2_io_error_work); bch2_time_stats_quantiles_init(&ca->io_latency[READ]); @@ -1337,7 +1330,7 @@ static struct bch_dev *__bch2_dev_alloc(struct bch_fs *c, if (percpu_ref_init(&ca->io_ref, bch2_dev_io_ref_complete, PERCPU_REF_INIT_DEAD, GFP_KERNEL) || - !(ca->sb_read_scratch = (void *) __get_free_page(GFP_KERNEL)) || + !(ca->sb_read_scratch = kmalloc(BCH_SB_READ_SCRATCH_BUF_SIZE, GFP_KERNEL)) || bch2_dev_buckets_alloc(c, ca) || !(ca->io_done = alloc_percpu(*ca->io_done))) goto err; @@ -1366,7 +1359,6 @@ static int bch2_dev_alloc(struct bch_fs *c, unsigned dev_idx) { struct bch_member member = bch2_sb_member_get(c->disk_sb.sb, dev_idx); struct bch_dev *ca = NULL; - int ret = 0; if (bch2_fs_init_fault("dev_alloc")) goto err; @@ -1378,10 +1370,8 @@ static int bch2_dev_alloc(struct bch_fs *c, unsigned dev_idx) ca->fs = c; bch2_dev_attach(c, ca, dev_idx); - return ret; + return 0; err: - if (ca) - bch2_dev_free(ca); return -BCH_ERR_ENOMEM_dev_alloc; } @@ -1751,11 +1741,6 @@ int bch2_dev_add(struct bch_fs *c, const char *path) if (ret) goto err; - ret = bch2_dev_journal_alloc(ca, true); - bch_err_msg(c, ret, "allocating journal"); - if (ret) - goto err; - down_write(&c->state_lock); mutex_lock(&c->sb_lock); @@ -1806,11 +1791,14 @@ int bch2_dev_add(struct bch_fs *c, const char *path) if (ret) goto err_late; - ca->new_fs_bucket_idx = 0; - if (ca->mi.state == BCH_MEMBER_STATE_rw) __bch2_dev_read_write(c, ca); + ret = bch2_dev_journal_alloc(ca, false); + bch_err_msg(c, ret, "allocating journal"); + if (ret) + goto err_late; + up_write(&c->state_lock); return 0; diff --git a/fs/bcachefs/super.h b/fs/bcachefs/super.h index dada09331d2e..fa6d52216510 100644 --- a/fs/bcachefs/super.h +++ b/fs/bcachefs/super.h @@ -34,16 +34,6 @@ void bch2_fs_read_only(struct bch_fs *); int bch2_fs_read_write(struct bch_fs *); int bch2_fs_read_write_early(struct bch_fs *); -/* - * Only for use in the recovery/fsck path: - */ -static inline void bch2_fs_lazy_rw(struct bch_fs *c) -{ - if (!test_bit(BCH_FS_rw, &c->flags) && - !test_bit(BCH_FS_was_rw, &c->flags)) - bch2_fs_read_write_early(c); -} - void __bch2_fs_stop(struct bch_fs *); void bch2_fs_free(struct bch_fs *); void bch2_fs_stop(struct bch_fs *); diff --git a/fs/bcachefs/sysfs.c b/fs/bcachefs/sysfs.c index 03e59f86f360..a7eb1f511484 100644 --- a/fs/bcachefs/sysfs.c +++ b/fs/bcachefs/sysfs.c @@ -146,7 +146,7 @@ write_attribute(trigger_journal_writes); write_attribute(trigger_btree_cache_shrink); write_attribute(trigger_btree_key_cache_shrink); write_attribute(trigger_freelist_wakeup); -rw_attribute(gc_gens_pos); +read_attribute(gc_gens_pos); read_attribute(uuid); read_attribute(minor); @@ -203,7 +203,6 @@ read_attribute(disk_groups); read_attribute(has_data); read_attribute(alloc_debug); -read_attribute(accounting); read_attribute(usage_base); #define x(t, n, ...) read_attribute(t); @@ -211,12 +210,11 @@ BCH_PERSISTENT_COUNTERS() #undef x rw_attribute(discard); +read_attribute(state); rw_attribute(label); -rw_attribute(copy_gc_enabled); read_attribute(copy_gc_wait); -rw_attribute(rebalance_enabled); sysfs_pd_controller_attribute(rebalance); read_attribute(rebalance_status); @@ -237,11 +235,6 @@ write_attribute(perf_test); BCH_TIME_STATS() #undef x -static struct attribute sysfs_state_rw = { - .name = "state", - .mode = 0444, -}; - static size_t bch2_btree_cache_size(struct bch_fs *c) { struct btree_cache *bc = &c->btree_cache; @@ -302,7 +295,8 @@ static int bch2_compression_stats_to_text(struct printbuf *out, struct bch_fs *c static void bch2_gc_gens_pos_to_text(struct printbuf *out, struct bch_fs *c) { - prt_printf(out, "%s: ", bch2_btree_id_str(c->gc_gens_btree)); + bch2_btree_id_to_text(out, c->gc_gens_btree); + prt_printf(out, ": "); bch2_bpos_to_text(out, c->gc_gens_pos); prt_printf(out, "\n"); } @@ -339,9 +333,6 @@ SHOW(bch2_fs) if (attr == &sysfs_gc_gens_pos) bch2_gc_gens_pos_to_text(out, c); - sysfs_printf(copy_gc_enabled, "%i", c->copy_gc_enabled); - - sysfs_printf(rebalance_enabled, "%i", c->rebalance.enabled); sysfs_pd_controller_show(rebalance, &c->rebalance.pd); /* XXX */ if (attr == &sysfs_copy_gc_wait) @@ -405,9 +396,6 @@ SHOW(bch2_fs) if (attr == &sysfs_alloc_debug) bch2_fs_alloc_debug_to_text(out, c); - if (attr == &sysfs_accounting) - bch2_fs_accounting_to_text(out, c); - if (attr == &sysfs_usage_base) bch2_fs_usage_base_to_text(out, c); @@ -418,23 +406,6 @@ STORE(bch2_fs) { struct bch_fs *c = container_of(kobj, struct bch_fs, kobj); - if (attr == &sysfs_copy_gc_enabled) { - ssize_t ret = strtoul_safe(buf, c->copy_gc_enabled) - ?: (ssize_t) size; - - if (c->copygc_thread) - wake_up_process(c->copygc_thread); - return ret; - } - - if (attr == &sysfs_rebalance_enabled) { - ssize_t ret = strtoul_safe(buf, c->rebalance.enabled) - ?: (ssize_t) size; - - rebalance_wakeup(c); - return ret; - } - sysfs_pd_controller_store(rebalance, &c->rebalance.pd); /* Debugging: */ @@ -534,15 +505,22 @@ SHOW(bch2_fs_counters) printbuf_tabstop_push(out, 32); - #define x(t, ...) \ + #define x(t, n, f, ...) \ if (attr == &sysfs_##t) { \ counter = percpu_u64_get(&c->counters[BCH_COUNTER_##t]);\ counter_since_mount = counter - c->counters_on_mount[BCH_COUNTER_##t];\ + if (f & TYPE_SECTORS) { \ + counter <<= 9; \ + counter_since_mount <<= 9; \ + } \ + \ prt_printf(out, "since mount:\t"); \ + (f & TYPE_COUNTER) ? prt_u64(out, counter_since_mount) :\ prt_human_readable_u64(out, counter_since_mount); \ prt_newline(out); \ \ prt_printf(out, "since filesystem creation:\t"); \ + (f & TYPE_COUNTER) ? prt_u64(out, counter) : \ prt_human_readable_u64(out, counter); \ prt_newline(out); \ } @@ -610,10 +588,8 @@ struct attribute *bch2_fs_internal_files[] = { &sysfs_gc_gens_pos, - &sysfs_copy_gc_enabled, &sysfs_copy_gc_wait, - &sysfs_rebalance_enabled, sysfs_pd_controller_files(rebalance), &sysfs_moving_ctxts, @@ -622,7 +598,6 @@ struct attribute *bch2_fs_internal_files[] = { &sysfs_disk_groups, &sysfs_alloc_debug, - &sysfs_accounting, &sysfs_usage_base, NULL }; @@ -682,6 +657,13 @@ STORE(bch2_fs_opts_dir) (id == Opt_compression && !c->opts.background_compression))) bch2_set_rebalance_needs_scan(c, 0); + if (v && id == Opt_rebalance_enabled) + rebalance_wakeup(c); + + if (v && id == Opt_copygc_enabled && + c->copygc_thread) + wake_up_process(c->copygc_thread); + ret = size; err: bch2_write_ref_put(c, BCH_WRITE_REF_sysfs); @@ -790,7 +772,7 @@ SHOW(bch2_dev) prt_char(out, '\n'); } - if (attr == &sysfs_state_rw) { + if (attr == &sysfs_state) { prt_string_option(out, bch2_member_states, ca->mi.state); prt_char(out, '\n'); } @@ -870,7 +852,7 @@ struct attribute *bch2_dev_files[] = { /* settings: */ &sysfs_discard, - &sysfs_state_rw, + &sysfs_state, &sysfs_label, &sysfs_has_data, diff --git a/fs/bcachefs/tests.c b/fs/bcachefs/tests.c index fb5c1543e52f..6c6469814637 100644 --- a/fs/bcachefs/tests.c +++ b/fs/bcachefs/tests.c @@ -131,7 +131,7 @@ static int test_iterate(struct bch_fs *c, u64 nr) i = 0; ret = bch2_trans_run(c, - for_each_btree_key_upto(trans, iter, BTREE_ID_xattrs, + for_each_btree_key_max(trans, iter, BTREE_ID_xattrs, SPOS(0, 0, U32_MAX), POS(0, U64_MAX), 0, k, ({ BUG_ON(k.k->p.offset != i++); @@ -186,7 +186,7 @@ static int test_iterate_extents(struct bch_fs *c, u64 nr) i = 0; ret = bch2_trans_run(c, - for_each_btree_key_upto(trans, iter, BTREE_ID_extents, + for_each_btree_key_max(trans, iter, BTREE_ID_extents, SPOS(0, 0, U32_MAX), POS(0, U64_MAX), 0, k, ({ BUG_ON(bkey_start_offset(k.k) != i); @@ -242,7 +242,7 @@ static int test_iterate_slots(struct bch_fs *c, u64 nr) i = 0; ret = bch2_trans_run(c, - for_each_btree_key_upto(trans, iter, BTREE_ID_xattrs, + for_each_btree_key_max(trans, iter, BTREE_ID_xattrs, SPOS(0, 0, U32_MAX), POS(0, U64_MAX), 0, k, ({ BUG_ON(k.k->p.offset != i); @@ -259,7 +259,7 @@ static int test_iterate_slots(struct bch_fs *c, u64 nr) i = 0; ret = bch2_trans_run(c, - for_each_btree_key_upto(trans, iter, BTREE_ID_xattrs, + for_each_btree_key_max(trans, iter, BTREE_ID_xattrs, SPOS(0, 0, U32_MAX), POS(0, U64_MAX), BTREE_ITER_slots, k, ({ if (i >= nr * 2) @@ -302,7 +302,7 @@ static int test_iterate_slots_extents(struct bch_fs *c, u64 nr) i = 0; ret = bch2_trans_run(c, - for_each_btree_key_upto(trans, iter, BTREE_ID_extents, + for_each_btree_key_max(trans, iter, BTREE_ID_extents, SPOS(0, 0, U32_MAX), POS(0, U64_MAX), 0, k, ({ BUG_ON(bkey_start_offset(k.k) != i + 8); @@ -320,7 +320,7 @@ static int test_iterate_slots_extents(struct bch_fs *c, u64 nr) i = 0; ret = bch2_trans_run(c, - for_each_btree_key_upto(trans, iter, BTREE_ID_extents, + for_each_btree_key_max(trans, iter, BTREE_ID_extents, SPOS(0, 0, U32_MAX), POS(0, U64_MAX), BTREE_ITER_slots, k, ({ if (i == nr) @@ -349,10 +349,10 @@ static int test_peek_end(struct bch_fs *c, u64 nr) bch2_trans_iter_init(trans, &iter, BTREE_ID_xattrs, SPOS(0, 0, U32_MAX), 0); - lockrestart_do(trans, bkey_err(k = bch2_btree_iter_peek_upto(&iter, POS(0, U64_MAX)))); + lockrestart_do(trans, bkey_err(k = bch2_btree_iter_peek_max(&iter, POS(0, U64_MAX)))); BUG_ON(k.k); - lockrestart_do(trans, bkey_err(k = bch2_btree_iter_peek_upto(&iter, POS(0, U64_MAX)))); + lockrestart_do(trans, bkey_err(k = bch2_btree_iter_peek_max(&iter, POS(0, U64_MAX)))); BUG_ON(k.k); bch2_trans_iter_exit(trans, &iter); @@ -369,10 +369,10 @@ static int test_peek_end_extents(struct bch_fs *c, u64 nr) bch2_trans_iter_init(trans, &iter, BTREE_ID_extents, SPOS(0, 0, U32_MAX), 0); - lockrestart_do(trans, bkey_err(k = bch2_btree_iter_peek_upto(&iter, POS(0, U64_MAX)))); + lockrestart_do(trans, bkey_err(k = bch2_btree_iter_peek_max(&iter, POS(0, U64_MAX)))); BUG_ON(k.k); - lockrestart_do(trans, bkey_err(k = bch2_btree_iter_peek_upto(&iter, POS(0, U64_MAX)))); + lockrestart_do(trans, bkey_err(k = bch2_btree_iter_peek_max(&iter, POS(0, U64_MAX)))); BUG_ON(k.k); bch2_trans_iter_exit(trans, &iter); @@ -488,7 +488,7 @@ static int test_snapshot_filter(struct bch_fs *c, u32 snapid_lo, u32 snapid_hi) trans = bch2_trans_get(c); bch2_trans_iter_init(trans, &iter, BTREE_ID_xattrs, SPOS(0, 0, snapid_lo), 0); - lockrestart_do(trans, bkey_err(k = bch2_btree_iter_peek_upto(&iter, POS(0, U64_MAX)))); + lockrestart_do(trans, bkey_err(k = bch2_btree_iter_peek_max(&iter, POS(0, U64_MAX)))); BUG_ON(k.k->p.snapshot != U32_MAX); @@ -672,7 +672,7 @@ static int __do_delete(struct btree_trans *trans, struct bpos pos) bch2_trans_iter_init(trans, &iter, BTREE_ID_xattrs, pos, BTREE_ITER_intent); - k = bch2_btree_iter_peek_upto(&iter, POS(0, U64_MAX)); + k = bch2_btree_iter_peek_max(&iter, POS(0, U64_MAX)); ret = bkey_err(k); if (ret) goto err; @@ -726,7 +726,7 @@ static int seq_insert(struct bch_fs *c, u64 nr) static int seq_lookup(struct bch_fs *c, u64 nr) { return bch2_trans_run(c, - for_each_btree_key_upto(trans, iter, BTREE_ID_xattrs, + for_each_btree_key_max(trans, iter, BTREE_ID_xattrs, SPOS(0, 0, U32_MAX), POS(0, U64_MAX), 0, k, 0)); diff --git a/fs/bcachefs/trace.h b/fs/bcachefs/trace.h index 5597b9d6297f..9d40b7d4ea29 100644 --- a/fs/bcachefs/trace.h +++ b/fs/bcachefs/trace.h @@ -199,6 +199,30 @@ DECLARE_EVENT_CLASS(bio, (unsigned long long)__entry->sector, __entry->nr_sector) ); +/* disk_accounting.c */ + +TRACE_EVENT(accounting_mem_insert, + TP_PROTO(struct bch_fs *c, const char *acc), + TP_ARGS(c, acc), + + TP_STRUCT__entry( + __field(dev_t, dev ) + __field(unsigned, new_nr ) + __string(acc, acc ) + ), + + TP_fast_assign( + __entry->dev = c->dev; + __entry->new_nr = c->accounting.k.nr; + __assign_str(acc); + ), + + TP_printk("%d,%d entries %u added %s", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->new_nr, + __get_str(acc)) +); + /* fs.c: */ TRACE_EVENT(bch2_sync_fs, TP_PROTO(struct super_block *sb, int wait), @@ -848,8 +872,8 @@ TRACE_EVENT(move_data, TRACE_EVENT(evacuate_bucket, TP_PROTO(struct bch_fs *c, struct bpos *bucket, unsigned sectors, unsigned bucket_size, - u64 fragmentation, int ret), - TP_ARGS(c, bucket, sectors, bucket_size, fragmentation, ret), + int ret), + TP_ARGS(c, bucket, sectors, bucket_size, ret), TP_STRUCT__entry( __field(dev_t, dev ) @@ -857,7 +881,6 @@ TRACE_EVENT(evacuate_bucket, __field(u64, bucket ) __field(u32, sectors ) __field(u32, bucket_size ) - __field(u64, fragmentation ) __field(int, ret ) ), @@ -867,15 +890,14 @@ TRACE_EVENT(evacuate_bucket, __entry->bucket = bucket->offset; __entry->sectors = sectors; __entry->bucket_size = bucket_size; - __entry->fragmentation = fragmentation; __entry->ret = ret; ), - TP_printk("%d,%d %llu:%llu sectors %u/%u fragmentation %llu ret %i", + TP_printk("%d,%d %llu:%llu sectors %u/%u ret %i", MAJOR(__entry->dev), MINOR(__entry->dev), __entry->member, __entry->bucket, __entry->sectors, __entry->bucket_size, - __entry->fragmentation, __entry->ret) + __entry->ret) ); TRACE_EVENT(copygc, @@ -1316,6 +1338,12 @@ TRACE_EVENT(trans_restart_key_cache_key_realloced, __entry->new_u64s) ); +DEFINE_EVENT(transaction_event, trans_restart_write_buffer_flush, + TP_PROTO(struct btree_trans *trans, + unsigned long caller_ip), + TP_ARGS(trans, caller_ip) +); + TRACE_EVENT(path_downgrade, TP_PROTO(struct btree_trans *trans, unsigned long caller_ip, @@ -1352,10 +1380,21 @@ TRACE_EVENT(path_downgrade, __entry->pos_snapshot) ); -DEFINE_EVENT(transaction_event, trans_restart_write_buffer_flush, - TP_PROTO(struct btree_trans *trans, - unsigned long caller_ip), - TP_ARGS(trans, caller_ip) +TRACE_EVENT(key_cache_fill, + TP_PROTO(struct btree_trans *trans, const char *key), + TP_ARGS(trans, key), + + TP_STRUCT__entry( + __array(char, trans_fn, 32 ) + __string(key, key ) + ), + + TP_fast_assign( + strscpy(__entry->trans_fn, trans->fn, sizeof(__entry->trans_fn)); + __assign_str(key); + ), + + TP_printk("%s %s", __entry->trans_fn, __get_str(key)) ); TRACE_EVENT(write_buffer_flush, @@ -1414,6 +1453,24 @@ TRACE_EVENT(write_buffer_flush_slowpath, TP_printk("%zu/%zu", __entry->slowpath, __entry->total) ); +TRACE_EVENT(write_buffer_maybe_flush, + TP_PROTO(struct btree_trans *trans, unsigned long caller_ip, const char *key), + TP_ARGS(trans, caller_ip, key), + + TP_STRUCT__entry( + __array(char, trans_fn, 32 ) + __field(unsigned long, caller_ip ) + __string(key, key ) + ), + + TP_fast_assign( + strscpy(__entry->trans_fn, trans->fn, sizeof(__entry->trans_fn)); + __assign_str(key); + ), + + TP_printk("%s %pS %s", __entry->trans_fn, (void *) __entry->caller_ip, __get_str(key)) +); + DEFINE_EVENT(fs_str, rebalance_extent, TP_PROTO(struct bch_fs *c, const char *str), TP_ARGS(c, str) diff --git a/fs/bcachefs/util.h b/fs/bcachefs/util.h index fb02c1c36004..1a1720116071 100644 --- a/fs/bcachefs/util.h +++ b/fs/bcachefs/util.h @@ -55,6 +55,16 @@ static inline size_t buf_pages(void *p, size_t len) PAGE_SIZE); } +static inline void *bch2_kvmalloc(size_t n, gfp_t flags) +{ + void *p = unlikely(n >= INT_MAX) + ? vmalloc(n) + : kvmalloc(n, flags & ~__GFP_ZERO); + if (p && (flags & __GFP_ZERO)) + memset(p, 0, n); + return p; +} + #define init_heap(heap, _size, gfp) \ ({ \ (heap)->nr = 0; \ @@ -317,6 +327,19 @@ do { \ _ptr ? container_of(_ptr, type, member) : NULL; \ }) +static inline struct list_head *list_pop(struct list_head *head) +{ + if (list_empty(head)) + return NULL; + + struct list_head *ret = head->next; + list_del_init(ret); + return ret; +} + +#define list_pop_entry(head, type, member) \ + container_of_or_null(list_pop(head), type, member) + /* Does linear interpolation between powers of two */ static inline unsigned fract_exp_two(unsigned x, unsigned fract_bits) { @@ -696,4 +719,13 @@ static inline bool test_bit_le64(size_t bit, __le64 *addr) return (addr[bit / 64] & cpu_to_le64(BIT_ULL(bit % 64))) != 0; } +static inline void memcpy_swab(void *_dst, void *_src, size_t len) +{ + u8 *dst = _dst + len; + u8 *src = _src; + + while (len--) + *--dst = *src++; +} + #endif /* _BCACHEFS_UTIL_H */ diff --git a/fs/bcachefs/varint.c b/fs/bcachefs/varint.c index 6a78553d9b0c..6620ecae26af 100644 --- a/fs/bcachefs/varint.c +++ b/fs/bcachefs/varint.c @@ -9,6 +9,7 @@ #include <valgrind/memcheck.h> #endif +#include "errcode.h" #include "varint.h" /** @@ -53,7 +54,7 @@ int bch2_varint_decode(const u8 *in, const u8 *end, u64 *out) u64 v; if (unlikely(in + bytes > end)) - return -1; + return -BCH_ERR_varint_decode_error; if (likely(bytes < 9)) { __le64 v_le = 0; @@ -115,7 +116,7 @@ int bch2_varint_decode_fast(const u8 *in, const u8 *end, u64 *out) unsigned bytes = ffz(*in) + 1; if (unlikely(in + bytes > end)) - return -1; + return -BCH_ERR_varint_decode_error; if (likely(bytes < 9)) { v >>= bytes; diff --git a/fs/bcachefs/xattr.c b/fs/bcachefs/xattr.c index 952aca400faf..aed7c6984173 100644 --- a/fs/bcachefs/xattr.c +++ b/fs/bcachefs/xattr.c @@ -71,7 +71,7 @@ const struct bch_hash_desc bch2_xattr_hash_desc = { }; int bch2_xattr_validate(struct bch_fs *c, struct bkey_s_c k, - enum bch_validate_flags flags) + struct bkey_validate_context from) { struct bkey_s_c_xattr xattr = bkey_s_c_to_xattr(k); unsigned val_u64s = xattr_val_u64s(xattr.v->x_name_len, @@ -309,7 +309,7 @@ ssize_t bch2_xattr_list(struct dentry *dentry, char *buffer, size_t buffer_size) u64 offset = 0, inum = inode->ei_inode.bi_inum; int ret = bch2_trans_run(c, - for_each_btree_key_in_subvolume_upto(trans, iter, BTREE_ID_xattrs, + for_each_btree_key_in_subvolume_max(trans, iter, BTREE_ID_xattrs, POS(inum, offset), POS(inum, U64_MAX), inode->ei_inum.subvol, 0, k, ({ @@ -565,13 +565,6 @@ static int bch2_xattr_bcachefs_set(const struct xattr_handler *handler, ret = bch2_write_inode(c, inode, inode_opt_set_fn, &s, 0); err: mutex_unlock(&inode->ei_update_lock); - - if (value && - (opt_id == Opt_background_target || - opt_id == Opt_background_compression || - (opt_id == Opt_compression && !inode_opt_get(c, &inode->ei_inode, background_compression)))) - bch2_set_rebalance_needs_scan(c, inode->ei_inode.bi_inum); - err_class_exit: return bch2_err_class(ret); } @@ -609,7 +602,7 @@ static const struct xattr_handler bch_xattr_bcachefs_effective_handler = { #endif /* NO_BCACHEFS_FS */ -const struct xattr_handler *bch2_xattr_handlers[] = { +const struct xattr_handler * const bch2_xattr_handlers[] = { &bch_xattr_user_handler, &bch_xattr_trusted_handler, &bch_xattr_security_handler, diff --git a/fs/bcachefs/xattr.h b/fs/bcachefs/xattr.h index c188a5ad64ce..132fbbd15a66 100644 --- a/fs/bcachefs/xattr.h +++ b/fs/bcachefs/xattr.h @@ -6,7 +6,8 @@ extern const struct bch_hash_desc bch2_xattr_hash_desc; -int bch2_xattr_validate(struct bch_fs *, struct bkey_s_c, enum bch_validate_flags); +int bch2_xattr_validate(struct bch_fs *, struct bkey_s_c, + struct bkey_validate_context); void bch2_xattr_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c); #define bch2_bkey_ops_xattr ((struct bkey_ops) { \ @@ -44,6 +45,6 @@ int bch2_xattr_set(struct btree_trans *, subvol_inum, ssize_t bch2_xattr_list(struct dentry *, char *, size_t); -extern const struct xattr_handler *bch2_xattr_handlers[]; +extern const struct xattr_handler * const bch2_xattr_handlers[]; #endif /* _BCACHEFS_XATTR_H */ diff --git a/fs/binfmt_flat.c b/fs/binfmt_flat.c index 390808ce935d..b5b5ca1a44f7 100644 --- a/fs/binfmt_flat.c +++ b/fs/binfmt_flat.c @@ -478,7 +478,7 @@ static int load_flat_file(struct linux_binprm *bprm, * 28 bits (256 MB) is way more than reasonable in this case. * If some top bits are set we have probable binary corruption. */ - if ((text_len | data_len | bss_len | stack_len | full_data) >> 28) { + if ((text_len | data_len | bss_len | stack_len | relocs | full_data) >> 28) { pr_err("bad header\n"); ret = -ENOEXEC; goto err; diff --git a/fs/binfmt_misc.c b/fs/binfmt_misc.c index 6a3a16f91051..5a7ebd160724 100644 --- a/fs/binfmt_misc.c +++ b/fs/binfmt_misc.c @@ -1001,7 +1001,7 @@ static int bm_fill_super(struct super_block *sb, struct fs_context *fc) /* * If it turns out that most user namespaces actually want to * register their own binary type handler and therefore all - * create their own separate binfm_misc mounts we should + * create their own separate binfmt_misc mounts we should * consider turning this into a kmem cache. */ misc = kzalloc(sizeof(struct binfmt_misc), GFP_KERNEL); diff --git a/fs/btrfs/Makefile b/fs/btrfs/Makefile index 3cfc440c636c..2d5f0482678b 100644 --- a/fs/btrfs/Makefile +++ b/fs/btrfs/Makefile @@ -44,4 +44,4 @@ btrfs-$(CONFIG_BTRFS_FS_RUN_SANITY_TESTS) += tests/free-space-tests.o \ tests/extent-buffer-tests.o tests/btrfs-tests.o \ tests/extent-io-tests.o tests/inode-tests.o tests/qgroup-tests.o \ tests/free-space-tree-tests.o tests/extent-map-tests.o \ - tests/raid-stripe-tree-tests.o + tests/raid-stripe-tree-tests.o tests/delayed-refs-tests.o diff --git a/fs/btrfs/async-thread.c b/fs/btrfs/async-thread.c index 361a866c1995..a4c51600a408 100644 --- a/fs/btrfs/async-thread.c +++ b/fs/btrfs/async-thread.c @@ -18,7 +18,7 @@ enum { }; #define NO_THRESHOLD (-1) -#define DFT_THRESHOLD (32) +#define DEFAULT_THRESHOLD (32) struct btrfs_workqueue { struct workqueue_struct *normal_wq; @@ -94,9 +94,9 @@ struct btrfs_workqueue *btrfs_alloc_workqueue(struct btrfs_fs_info *fs_info, ret->limit_active = limit_active; if (thresh == 0) - thresh = DFT_THRESHOLD; + thresh = DEFAULT_THRESHOLD; /* For low threshold, disabling threshold is a better choice */ - if (thresh < DFT_THRESHOLD) { + if (thresh < DEFAULT_THRESHOLD) { ret->current_active = limit_active; ret->thresh = NO_THRESHOLD; } else { diff --git a/fs/btrfs/backref.c b/fs/btrfs/backref.c index 04f53ca548e1..3d3923cfc357 100644 --- a/fs/btrfs/backref.c +++ b/fs/btrfs/backref.c @@ -250,6 +250,21 @@ static int prelim_ref_compare(const struct prelim_ref *ref1, return 0; } +static int prelim_ref_rb_add_cmp(const struct rb_node *new, + const struct rb_node *exist) +{ + const struct prelim_ref *ref_new = + rb_entry(new, struct prelim_ref, rbnode); + const struct prelim_ref *ref_exist = + rb_entry(exist, struct prelim_ref, rbnode); + + /* + * prelim_ref_compare() expects the first parameter as the existing one, + * different from the rb_find_add_cached() order. + */ + return prelim_ref_compare(ref_exist, ref_new); +} + static void update_share_count(struct share_check *sc, int oldcount, int newcount, const struct prelim_ref *newref) { @@ -278,55 +293,39 @@ static void prelim_ref_insert(const struct btrfs_fs_info *fs_info, struct share_check *sc) { struct rb_root_cached *root; - struct rb_node **p; - struct rb_node *parent = NULL; - struct prelim_ref *ref; - int result; - bool leftmost = true; + struct rb_node *exist; root = &preftree->root; - p = &root->rb_root.rb_node; + exist = rb_find_add_cached(&newref->rbnode, root, prelim_ref_rb_add_cmp); + if (exist) { + struct prelim_ref *ref = rb_entry(exist, struct prelim_ref, rbnode); + /* Identical refs, merge them and free @newref */ + struct extent_inode_elem *eie = ref->inode_list; - while (*p) { - parent = *p; - ref = rb_entry(parent, struct prelim_ref, rbnode); - result = prelim_ref_compare(ref, newref); - if (result < 0) { - p = &(*p)->rb_left; - } else if (result > 0) { - p = &(*p)->rb_right; - leftmost = false; - } else { - /* Identical refs, merge them and free @newref */ - struct extent_inode_elem *eie = ref->inode_list; - - while (eie && eie->next) - eie = eie->next; + while (eie && eie->next) + eie = eie->next; - if (!eie) - ref->inode_list = newref->inode_list; - else - eie->next = newref->inode_list; - trace_btrfs_prelim_ref_merge(fs_info, ref, newref, - preftree->count); - /* - * A delayed ref can have newref->count < 0. - * The ref->count is updated to follow any - * BTRFS_[ADD|DROP]_DELAYED_REF actions. - */ - update_share_count(sc, ref->count, - ref->count + newref->count, newref); - ref->count += newref->count; - free_pref(newref); - return; - } + if (!eie) + ref->inode_list = newref->inode_list; + else + eie->next = newref->inode_list; + trace_btrfs_prelim_ref_merge(fs_info, ref, newref, + preftree->count); + /* + * A delayed ref can have newref->count < 0. + * The ref->count is updated to follow any + * BTRFS_[ADD|DROP]_DELAYED_REF actions. + */ + update_share_count(sc, ref->count, + ref->count + newref->count, newref); + ref->count += newref->count; + free_pref(newref); + return; } update_share_count(sc, 0, newref->count, newref); preftree->count++; trace_btrfs_prelim_ref_insert(fs_info, newref, NULL, preftree->count); - rb_link_node(&newref->rbnode, parent, p); - rb_insert_color_cached(&newref->rbnode, root, leftmost); } /* @@ -3022,9 +3021,6 @@ void btrfs_backref_init_cache(struct btrfs_fs_info *fs_info, cache->rb_root = RB_ROOT; for (i = 0; i < BTRFS_MAX_LEVEL; i++) INIT_LIST_HEAD(&cache->pending[i]); - INIT_LIST_HEAD(&cache->changed); - INIT_LIST_HEAD(&cache->detached); - INIT_LIST_HEAD(&cache->leaves); INIT_LIST_HEAD(&cache->pending_edge); INIT_LIST_HEAD(&cache->useless_node); cache->fs_info = fs_info; @@ -3132,29 +3128,17 @@ void btrfs_backref_drop_node(struct btrfs_backref_cache *tree, void btrfs_backref_cleanup_node(struct btrfs_backref_cache *cache, struct btrfs_backref_node *node) { - struct btrfs_backref_node *upper; struct btrfs_backref_edge *edge; if (!node) return; - BUG_ON(!node->lowest && !node->detached); while (!list_empty(&node->upper)) { edge = list_entry(node->upper.next, struct btrfs_backref_edge, list[LOWER]); - upper = edge->node[UPPER]; list_del(&edge->list[LOWER]); list_del(&edge->list[UPPER]); btrfs_backref_free_edge(cache, edge); - - /* - * Add the node to leaf node list if no other child block - * cached. - */ - if (list_empty(&upper->lower)) { - list_add_tail(&upper->lower, &cache->leaves); - upper->lowest = 1; - } } btrfs_backref_drop_node(cache, node); @@ -3166,33 +3150,13 @@ void btrfs_backref_cleanup_node(struct btrfs_backref_cache *cache, void btrfs_backref_release_cache(struct btrfs_backref_cache *cache) { struct btrfs_backref_node *node; - int i; - - while (!list_empty(&cache->detached)) { - node = list_entry(cache->detached.next, - struct btrfs_backref_node, list); - btrfs_backref_cleanup_node(cache, node); - } - while (!list_empty(&cache->leaves)) { - node = list_entry(cache->leaves.next, - struct btrfs_backref_node, lower); + while ((node = rb_entry_safe(rb_first(&cache->rb_root), + struct btrfs_backref_node, rb_node))) btrfs_backref_cleanup_node(cache, node); - } - for (i = 0; i < BTRFS_MAX_LEVEL; i++) { - while (!list_empty(&cache->pending[i])) { - node = list_first_entry(&cache->pending[i], - struct btrfs_backref_node, - list); - btrfs_backref_cleanup_node(cache, node); - } - } ASSERT(list_empty(&cache->pending_edge)); ASSERT(list_empty(&cache->useless_node)); - ASSERT(list_empty(&cache->changed)); - ASSERT(list_empty(&cache->detached)); - ASSERT(RB_EMPTY_ROOT(&cache->rb_root)); ASSERT(!cache->nr_nodes); ASSERT(!cache->nr_edges); } @@ -3316,8 +3280,12 @@ static int handle_indirect_tree_backref(struct btrfs_trans_handle *trans, root = btrfs_get_fs_root(fs_info, ref_key->offset, false); if (IS_ERR(root)) return PTR_ERR(root); - if (!test_bit(BTRFS_ROOT_SHAREABLE, &root->state)) - cur->cowonly = 1; + + /* We shouldn't be using backref cache for non-shareable roots. */ + if (unlikely(!test_bit(BTRFS_ROOT_SHAREABLE, &root->state))) { + btrfs_put_root(root); + return -EUCLEAN; + } if (btrfs_root_level(&root->root_item) == cur->level) { /* Tree root */ @@ -3403,8 +3371,15 @@ static int handle_indirect_tree_backref(struct btrfs_trans_handle *trans, goto out; } upper->owner = btrfs_header_owner(eb); - if (!test_bit(BTRFS_ROOT_SHAREABLE, &root->state)) - upper->cowonly = 1; + + /* We shouldn't be using backref cache for non shareable roots. */ + if (unlikely(!test_bit(BTRFS_ROOT_SHAREABLE, &root->state))) { + btrfs_put_root(root); + btrfs_backref_free_edge(cache, edge); + btrfs_backref_free_node(cache, upper); + ret = -EUCLEAN; + goto out; + } /* * If we know the block isn't shared we can avoid @@ -3595,15 +3570,9 @@ int btrfs_backref_finish_upper_links(struct btrfs_backref_cache *cache, ASSERT(start->checked); - /* Insert this node to cache if it's not COW-only */ - if (!start->cowonly) { - rb_node = rb_simple_insert(&cache->rb_root, start->bytenr, - &start->rb_node); - if (rb_node) - btrfs_backref_panic(cache->fs_info, start->bytenr, - -EEXIST); - list_add_tail(&start->lower, &cache->leaves); - } + rb_node = rb_simple_insert(&cache->rb_root, start->bytenr, &start->rb_node); + if (rb_node) + btrfs_backref_panic(cache->fs_info, start->bytenr, -EEXIST); /* * Use breadth first search to iterate all related edges. @@ -3642,11 +3611,6 @@ int btrfs_backref_finish_upper_links(struct btrfs_backref_cache *cache, * parents have already been linked. */ if (!RB_EMPTY_NODE(&upper->rb_node)) { - if (upper->lowest) { - list_del_init(&upper->lower); - upper->lowest = 0; - } - list_add_tail(&edge->list[UPPER], &upper->lower); continue; } @@ -3657,23 +3621,13 @@ int btrfs_backref_finish_upper_links(struct btrfs_backref_cache *cache, return -EUCLEAN; } - /* Sanity check, COW-only node has non-COW-only parent */ - if (start->cowonly != upper->cowonly) { - ASSERT(0); + rb_node = rb_simple_insert(&cache->rb_root, upper->bytenr, + &upper->rb_node); + if (unlikely(rb_node)) { + btrfs_backref_panic(cache->fs_info, upper->bytenr, -EEXIST); return -EUCLEAN; } - /* Only cache non-COW-only (subvolume trees) tree blocks */ - if (!upper->cowonly) { - rb_node = rb_simple_insert(&cache->rb_root, upper->bytenr, - &upper->rb_node); - if (rb_node) { - btrfs_backref_panic(cache->fs_info, - upper->bytenr, -EEXIST); - return -EUCLEAN; - } - } - list_add_tail(&edge->list[UPPER], &upper->lower); /* diff --git a/fs/btrfs/backref.h b/fs/btrfs/backref.h index e8c22cccb5c1..74e614031274 100644 --- a/fs/btrfs/backref.h +++ b/fs/btrfs/backref.h @@ -318,6 +318,12 @@ struct btrfs_backref_node { u64 bytenr; }; /* Use rb_simple_node for search/insert */ + /* + * This is a sanity check, whenever we COW a block we will update + * new_bytenr with it's current location, and we will check this in + * various places to validate that the cache makes sense, it shouldn't + * be used for anything else. + */ u64 new_bytenr; /* Objectid of tree block owner, can be not uptodate */ u64 owner; @@ -335,10 +341,6 @@ struct btrfs_backref_node { struct extent_buffer *eb; /* Level of the tree block */ unsigned int level:8; - /* Is the block in a non-shareable tree */ - unsigned int cowonly:1; - /* 1 if no child node is in the cache */ - unsigned int lowest:1; /* Is the extent buffer locked */ unsigned int locked:1; /* Has the block been processed */ @@ -391,12 +393,6 @@ struct btrfs_backref_cache { * level blocks may not reflect the new location */ struct list_head pending[BTRFS_MAX_LEVEL]; - /* List of backref nodes with no child node */ - struct list_head leaves; - /* List of blocks that have been COWed in current transaction */ - struct list_head changed; - /* List of detached backref node. */ - struct list_head detached; u64 last_trans; diff --git a/fs/btrfs/bio.c b/fs/btrfs/bio.c index 7ea6f0b43b95..bc2555c44a12 100644 --- a/fs/btrfs/bio.c +++ b/fs/btrfs/bio.c @@ -453,6 +453,14 @@ static void btrfs_submit_dev_bio(struct btrfs_device *dev, struct bio *bio) (unsigned long)dev->bdev->bd_dev, btrfs_dev_name(dev), dev->devid, bio->bi_iter.bi_size); + /* + * Track reads if tracking is enabled; ignore I/O operations before the + * filesystem is fully initialized. + */ + if (dev->fs_devices->collect_fs_stats && bio_op(bio) == REQ_OP_READ && dev->fs_info) + percpu_counter_add(&dev->fs_info->stats_read_blocks, + bio->bi_iter.bi_size >> dev->fs_info->sectorsize_bits); + if (bio->bi_opf & REQ_BTRFS_CGROUP_PUNT) blkcg_punt_bio_submit(bio); else @@ -725,8 +733,7 @@ static bool btrfs_submit_chunk(struct btrfs_bio *bbio, int mirror_num) bio->bi_opf |= REQ_OP_ZONE_APPEND; } - if (is_data_bbio(bbio) && bioc && - btrfs_need_stripe_tree_update(bioc->fs_info, bioc->map_type)) { + if (is_data_bbio(bbio) && bioc && bioc->use_rst) { /* * No locking for the list update, as we only add to * the list in the I/O submission path, and list diff --git a/fs/btrfs/block-group.c b/fs/btrfs/block-group.c index 4427c1b835e8..c0a8f7d92acc 100644 --- a/fs/btrfs/block-group.c +++ b/fs/btrfs/block-group.c @@ -173,43 +173,41 @@ void btrfs_put_block_group(struct btrfs_block_group *cache) } } +static int btrfs_bg_start_cmp(const struct rb_node *new, + const struct rb_node *exist) +{ + const struct btrfs_block_group *new_bg = + rb_entry(new, struct btrfs_block_group, cache_node); + const struct btrfs_block_group *exist_bg = + rb_entry(exist, struct btrfs_block_group, cache_node); + + if (new_bg->start < exist_bg->start) + return -1; + if (new_bg->start > exist_bg->start) + return 1; + return 0; +} + /* * This adds the block group to the fs_info rb tree for the block group cache */ static int btrfs_add_block_group_cache(struct btrfs_fs_info *info, struct btrfs_block_group *block_group) { - struct rb_node **p; - struct rb_node *parent = NULL; - struct btrfs_block_group *cache; - bool leftmost = true; + struct rb_node *exist; + int ret = 0; ASSERT(block_group->length != 0); write_lock(&info->block_group_cache_lock); - p = &info->block_group_cache_tree.rb_root.rb_node; - - while (*p) { - parent = *p; - cache = rb_entry(parent, struct btrfs_block_group, cache_node); - if (block_group->start < cache->start) { - p = &(*p)->rb_left; - } else if (block_group->start > cache->start) { - p = &(*p)->rb_right; - leftmost = false; - } else { - write_unlock(&info->block_group_cache_lock); - return -EEXIST; - } - } - - rb_link_node(&block_group->cache_node, parent, p); - rb_insert_color_cached(&block_group->cache_node, - &info->block_group_cache_tree, leftmost); + exist = rb_find_add_cached(&block_group->cache_node, + &info->block_group_cache_tree, btrfs_bg_start_cmp); + if (exist) + ret = -EEXIST; write_unlock(&info->block_group_cache_lock); - return 0; + return ret; } /* @@ -1223,7 +1221,7 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans, block_group->space_info->total_bytes -= block_group->length; block_group->space_info->bytes_readonly -= (block_group->length - block_group->zone_unusable); - btrfs_space_info_update_bytes_zone_unusable(fs_info, block_group->space_info, + btrfs_space_info_update_bytes_zone_unusable(block_group->space_info, -block_group->zone_unusable); block_group->space_info->disk_total -= block_group->length * factor; @@ -1396,8 +1394,7 @@ static int inc_block_group_ro(struct btrfs_block_group *cache, int force) if (btrfs_is_zoned(cache->fs_info)) { /* Migrate zone_unusable bytes to readonly */ sinfo->bytes_readonly += cache->zone_unusable; - btrfs_space_info_update_bytes_zone_unusable(cache->fs_info, sinfo, - -cache->zone_unusable); + btrfs_space_info_update_bytes_zone_unusable(sinfo, -cache->zone_unusable); cache->zone_unusable = 0; } cache->ro++; @@ -1645,8 +1642,7 @@ void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info) spin_lock(&space_info->lock); spin_lock(&block_group->lock); - btrfs_space_info_update_bytes_pinned(fs_info, space_info, - -block_group->pinned); + btrfs_space_info_update_bytes_pinned(space_info, -block_group->pinned); space_info->bytes_readonly += block_group->pinned; block_group->pinned = 0; @@ -2672,7 +2668,6 @@ static int insert_dev_extent(struct btrfs_trans_handle *trans, btrfs_set_dev_extent_chunk_offset(leaf, extent, chunk_offset); btrfs_set_dev_extent_length(leaf, extent, num_bytes); - btrfs_mark_buffer_dirty(trans, leaf); out: btrfs_free_path(path); return ret; @@ -3060,8 +3055,7 @@ void btrfs_dec_block_group_ro(struct btrfs_block_group *cache) (cache->alloc_offset - cache->used - cache->pinned - cache->reserved) + (cache->length - cache->zone_capacity); - btrfs_space_info_update_bytes_zone_unusable(cache->fs_info, sinfo, - cache->zone_unusable); + btrfs_space_info_update_bytes_zone_unusable(sinfo, cache->zone_unusable); sinfo->bytes_readonly -= cache->zone_unusable; } num_bytes = cache->length - cache->reserved - @@ -3123,7 +3117,6 @@ static int update_block_group_item(struct btrfs_trans_handle *trans, cache->global_root_id); btrfs_set_stack_block_group_flags(&bgi, cache->flags); write_extent_buffer(leaf, &bgi, bi, sizeof(bgi)); - btrfs_mark_buffer_dirty(trans, leaf); fail: btrfs_release_path(path); /* @@ -3699,7 +3692,7 @@ int btrfs_update_block_group(struct btrfs_trans_handle *trans, old_val -= num_bytes; cache->used = old_val; cache->pinned += num_bytes; - btrfs_space_info_update_bytes_pinned(info, space_info, num_bytes); + btrfs_space_info_update_bytes_pinned(space_info, num_bytes); space_info->bytes_used -= num_bytes; space_info->disk_used -= num_bytes * factor; if (READ_ONCE(space_info->periodic_reclaim)) @@ -3781,8 +3774,7 @@ int btrfs_add_reserved_bytes(struct btrfs_block_group *cache, space_info->bytes_reserved += num_bytes; trace_btrfs_space_reservation(cache->fs_info, "space_info", space_info->flags, num_bytes, 1); - btrfs_space_info_update_bytes_may_use(cache->fs_info, - space_info, -ram_bytes); + btrfs_space_info_update_bytes_may_use(space_info, -ram_bytes); if (delalloc) cache->delalloc_bytes += num_bytes; diff --git a/fs/btrfs/block-rsv.c b/fs/btrfs/block-rsv.c index a07b9594dc70..3f3608299c0b 100644 --- a/fs/btrfs/block-rsv.c +++ b/fs/btrfs/block-rsv.c @@ -150,9 +150,7 @@ static u64 block_rsv_release_bytes(struct btrfs_fs_info *fs_info, spin_unlock(&dest->lock); } if (num_bytes) - btrfs_space_info_free_bytes_may_use(fs_info, - space_info, - num_bytes); + btrfs_space_info_free_bytes_may_use(space_info, num_bytes); } if (qgroup_to_release_ret) *qgroup_to_release_ret = qgroup_to_release; @@ -383,13 +381,11 @@ void btrfs_update_global_block_rsv(struct btrfs_fs_info *fs_info) if (block_rsv->reserved < block_rsv->size) { num_bytes = block_rsv->size - block_rsv->reserved; - btrfs_space_info_update_bytes_may_use(fs_info, sinfo, - num_bytes); + btrfs_space_info_update_bytes_may_use(sinfo, num_bytes); block_rsv->reserved = block_rsv->size; } else if (block_rsv->reserved > block_rsv->size) { num_bytes = block_rsv->reserved - block_rsv->size; - btrfs_space_info_update_bytes_may_use(fs_info, sinfo, - -num_bytes); + btrfs_space_info_update_bytes_may_use(sinfo, -num_bytes); block_rsv->reserved = block_rsv->size; btrfs_try_granting_tickets(fs_info, sinfo); } diff --git a/fs/btrfs/btrfs_inode.h b/fs/btrfs/btrfs_inode.h index aa1f55cd81b7..b2fa33911c28 100644 --- a/fs/btrfs/btrfs_inode.h +++ b/fs/btrfs/btrfs_inode.h @@ -526,7 +526,7 @@ bool btrfs_data_csum_ok(struct btrfs_bio *bbio, struct btrfs_device *dev, u32 bio_offset, struct bio_vec *bv); noinline int can_nocow_extent(struct inode *inode, u64 offset, u64 *len, struct btrfs_file_extent *file_extent, - bool nowait, bool strict); + bool nowait); void btrfs_del_delalloc_inode(struct btrfs_inode *inode); struct inode *btrfs_lookup_dentry(struct inode *dir, struct dentry *dentry); diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c index 693dc27ffb89..92071ca0655f 100644 --- a/fs/btrfs/ctree.c +++ b/fs/btrfs/ctree.c @@ -37,19 +37,6 @@ static int push_node_left(struct btrfs_trans_handle *trans, static int balance_node_right(struct btrfs_trans_handle *trans, struct extent_buffer *dst_buf, struct extent_buffer *src_buf); - -static const struct btrfs_csums { - u16 size; - const char name[10]; - const char driver[12]; -} btrfs_csums[] = { - [BTRFS_CSUM_TYPE_CRC32] = { .size = 4, .name = "crc32c" }, - [BTRFS_CSUM_TYPE_XXHASH] = { .size = 8, .name = "xxhash64" }, - [BTRFS_CSUM_TYPE_SHA256] = { .size = 32, .name = "sha256" }, - [BTRFS_CSUM_TYPE_BLAKE2] = { .size = 32, .name = "blake2b", - .driver = "blake2b-256" }, -}; - /* * The leaf data grows from end-to-front in the node. this returns the address * of the start of the last item, which is the stop of the leaf data stack. @@ -148,44 +135,6 @@ static inline void copy_leaf_items(const struct extent_buffer *dst, nr_items * sizeof(struct btrfs_item)); } -/* This exists for btrfs-progs usages. */ -u16 btrfs_csum_type_size(u16 type) -{ - return btrfs_csums[type].size; -} - -int btrfs_super_csum_size(const struct btrfs_super_block *s) -{ - u16 t = btrfs_super_csum_type(s); - /* - * csum type is validated at mount time - */ - return btrfs_csum_type_size(t); -} - -const char *btrfs_super_csum_name(u16 csum_type) -{ - /* csum type is validated at mount time */ - return btrfs_csums[csum_type].name; -} - -/* - * Return driver name if defined, otherwise the name that's also a valid driver - * name - */ -const char *btrfs_super_csum_driver(u16 csum_type) -{ - /* csum type is validated at mount time */ - return btrfs_csums[csum_type].driver[0] ? - btrfs_csums[csum_type].driver : - btrfs_csums[csum_type].name; -} - -size_t __attribute_const__ btrfs_get_num_csums(void) -{ - return ARRAY_SIZE(btrfs_csums); -} - struct btrfs_path *btrfs_alloc_path(void) { might_sleep(); @@ -226,22 +175,6 @@ noinline void btrfs_release_path(struct btrfs_path *p) } /* - * We want the transaction abort to print stack trace only for errors where the - * cause could be a bug, eg. due to ENOSPC, and not for common errors that are - * caused by external factors. - */ -bool __cold abort_should_print_stack(int error) -{ - switch (error) { - case -EIO: - case -EROFS: - case -ENOMEM: - return false; - } - return true; -} - -/* * safely gets a reference on the root node of a tree. A lock * is not taken, so a concurrent writer may put a different node * at the root of the tree. See btrfs_lock_root_node for the @@ -654,6 +587,8 @@ int btrfs_force_cow_block(struct btrfs_trans_handle *trans, goto error_unlock_cow; } } + + trace_btrfs_cow_block(root, buf, cow); if (unlock_orig) btrfs_tree_unlock(buf); free_extent_buffer_stale(buf); @@ -710,7 +645,6 @@ int btrfs_cow_block(struct btrfs_trans_handle *trans, { struct btrfs_fs_info *fs_info = root->fs_info; u64 search_start; - int ret; if (unlikely(test_bit(BTRFS_ROOT_DELETING, &root->state))) { btrfs_abort_transaction(trans, -EUCLEAN); @@ -751,12 +685,8 @@ int btrfs_cow_block(struct btrfs_trans_handle *trans, * Also We don't care about the error, as it's handled internally. */ btrfs_qgroup_trace_subtree_after_cow(trans, root, buf); - ret = btrfs_force_cow_block(trans, root, buf, parent, parent_slot, - cow_ret, search_start, 0, nest); - - trace_btrfs_cow_block(root, buf, *cow_ret); - - return ret; + return btrfs_force_cow_block(trans, root, buf, parent, parent_slot, + cow_ret, search_start, 0, nest); } ALLOW_ERROR_INJECTION(btrfs_cow_block, ERRNO); @@ -3903,6 +3833,7 @@ static noinline int setup_leaf_for_split(struct btrfs_trans_handle *trans, btrfs_item_key_to_cpu(leaf, &key, path->slots[0]); BUG_ON(key.type != BTRFS_EXTENT_DATA_KEY && + key.type != BTRFS_RAID_STRIPE_KEY && key.type != BTRFS_EXTENT_CSUM_KEY); if (btrfs_leaf_free_space(leaf) >= ins_len) diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h index 2c341956a01c..1096a80a64e7 100644 --- a/fs/btrfs/ctree.h +++ b/fs/btrfs/ctree.h @@ -7,7 +7,6 @@ #define BTRFS_CTREE_H #include "linux/cleanup.h" -#include <linux/pagemap.h> #include <linux/spinlock.h> #include <linux/rbtree.h> #include <linux/mutex.h> @@ -506,20 +505,6 @@ static inline u32 BTRFS_MAX_XATTR_SIZE(const struct btrfs_fs_info *info) return BTRFS_MAX_ITEM_SIZE(info) - sizeof(struct btrfs_dir_item); } -#define BTRFS_BYTES_TO_BLKS(fs_info, bytes) \ - ((bytes) >> (fs_info)->sectorsize_bits) - -static inline gfp_t btrfs_alloc_write_mask(struct address_space *mapping) -{ - return mapping_gfp_constraint(mapping, ~__GFP_FS); -} - -void btrfs_error_unpin_extent_range(struct btrfs_fs_info *fs_info, u64 start, u64 end); -int btrfs_discard_extent(struct btrfs_fs_info *fs_info, u64 bytenr, - u64 num_bytes, u64 *actual_bytes); -int btrfs_trim_fs(struct btrfs_fs_info *fs_info, struct fstrim_range *range); - -/* ctree.c */ int __init btrfs_ctree_init(void); void __cold btrfs_ctree_exit(void); @@ -756,18 +741,4 @@ static inline bool btrfs_is_data_reloc_root(const struct btrfs_root *root) return root->root_key.objectid == BTRFS_DATA_RELOC_TREE_OBJECTID; } -u16 btrfs_csum_type_size(u16 type); -int btrfs_super_csum_size(const struct btrfs_super_block *s); -const char *btrfs_super_csum_name(u16 csum_type); -const char *btrfs_super_csum_driver(u16 csum_type); -size_t __attribute_const__ btrfs_get_num_csums(void); - -/* - * We use folio flag owner_2 to indicate there is an ordered extent with - * unfinished IO. - */ -#define folio_test_ordered(folio) folio_test_owner_2(folio) -#define folio_set_ordered(folio) folio_set_owner_2(folio) -#define folio_clear_ordered(folio) folio_clear_owner_2(folio) - #endif diff --git a/fs/btrfs/delalloc-space.c b/fs/btrfs/delalloc-space.c index 7aa8a395d838..88e900e5a43d 100644 --- a/fs/btrfs/delalloc-space.c +++ b/fs/btrfs/delalloc-space.c @@ -176,7 +176,7 @@ void btrfs_free_reserved_data_space_noquota(struct btrfs_fs_info *fs_info, ASSERT(IS_ALIGNED(len, fs_info->sectorsize)); data_sinfo = fs_info->data_sinfo; - btrfs_space_info_free_bytes_may_use(fs_info, data_sinfo, len); + btrfs_space_info_free_bytes_may_use(data_sinfo, len); } /* diff --git a/fs/btrfs/delayed-inode.c b/fs/btrfs/delayed-inode.c index 508bdbae29a0..0b4933c6a889 100644 --- a/fs/btrfs/delayed-inode.c +++ b/fs/btrfs/delayed-inode.c @@ -366,40 +366,35 @@ static struct btrfs_delayed_item *__btrfs_lookup_delayed_item( return NULL; } +static int btrfs_delayed_item_cmp(const struct rb_node *new, + const struct rb_node *exist) +{ + const struct btrfs_delayed_item *new_item = + rb_entry(new, struct btrfs_delayed_item, rb_node); + const struct btrfs_delayed_item *exist_item = + rb_entry(exist, struct btrfs_delayed_item, rb_node); + + if (new_item->index < exist_item->index) + return -1; + if (new_item->index > exist_item->index) + return 1; + return 0; +} + static int __btrfs_add_delayed_item(struct btrfs_delayed_node *delayed_node, struct btrfs_delayed_item *ins) { - struct rb_node **p, *node; - struct rb_node *parent_node = NULL; struct rb_root_cached *root; - struct btrfs_delayed_item *item; - bool leftmost = true; + struct rb_node *exist; if (ins->type == BTRFS_DELAYED_INSERTION_ITEM) root = &delayed_node->ins_root; else root = &delayed_node->del_root; - p = &root->rb_root.rb_node; - node = &ins->rb_node; - - while (*p) { - parent_node = *p; - item = rb_entry(parent_node, struct btrfs_delayed_item, - rb_node); - - if (item->index < ins->index) { - p = &(*p)->rb_right; - leftmost = false; - } else if (item->index > ins->index) { - p = &(*p)->rb_left; - } else { - return -EEXIST; - } - } - - rb_link_node(node, parent_node, p); - rb_insert_color_cached(node, root, leftmost); + exist = rb_find_add_cached(&ins->rb_node, root, btrfs_delayed_item_cmp); + if (exist) + return -EEXIST; if (ins->type == BTRFS_DELAYED_INSERTION_ITEM && ins->index >= delayed_node->index_cnt) @@ -1038,7 +1033,6 @@ static int __btrfs_update_delayed_inode(struct btrfs_trans_handle *trans, struct btrfs_inode_item); write_extent_buffer(leaf, &node->inode_item, (unsigned long)inode_item, sizeof(struct btrfs_inode_item)); - btrfs_mark_buffer_dirty(trans, leaf); if (!test_bit(BTRFS_DELAYED_NODE_DEL_IREF, &node->flags)) goto out; @@ -1561,8 +1555,7 @@ release_node: return ret; } -static int btrfs_delete_delayed_insertion_item(struct btrfs_fs_info *fs_info, - struct btrfs_delayed_node *node, +static int btrfs_delete_delayed_insertion_item(struct btrfs_delayed_node *node, u64 index) { struct btrfs_delayed_item *item; @@ -1620,7 +1613,7 @@ int btrfs_delete_delayed_dir_index(struct btrfs_trans_handle *trans, if (IS_ERR(node)) return PTR_ERR(node); - ret = btrfs_delete_delayed_insertion_item(trans->fs_info, node, index); + ret = btrfs_delete_delayed_insertion_item(node, index); if (!ret) goto end; diff --git a/fs/btrfs/delayed-ref.c b/fs/btrfs/delayed-ref.c index 0d878dbbabba..98c5b61dabe8 100644 --- a/fs/btrfs/delayed-ref.c +++ b/fs/btrfs/delayed-ref.c @@ -93,6 +93,9 @@ void btrfs_update_delayed_refs_rsv(struct btrfs_trans_handle *trans) u64 num_bytes; u64 reserved_bytes; + if (btrfs_is_testing(fs_info)) + return; + num_bytes = btrfs_calc_delayed_ref_bytes(fs_info, trans->delayed_ref_updates); num_bytes += btrfs_calc_delayed_ref_csum_bytes(fs_info, trans->delayed_ref_csum_deletions); @@ -254,7 +257,7 @@ int btrfs_delayed_refs_rsv_refill(struct btrfs_fs_info *fs_info, spin_unlock(&block_rsv->lock); if (to_free > 0) - btrfs_space_info_free_bytes_may_use(fs_info, space_info, to_free); + btrfs_space_info_free_bytes_may_use(space_info, to_free); if (refilled_bytes > 0) trace_btrfs_space_reservation(fs_info, "delayed_refs_rsv", 0, @@ -265,8 +268,8 @@ int btrfs_delayed_refs_rsv_refill(struct btrfs_fs_info *fs_info, /* * compare two delayed data backrefs with same bytenr and type */ -static int comp_data_refs(struct btrfs_delayed_ref_node *ref1, - struct btrfs_delayed_ref_node *ref2) +static int comp_data_refs(const struct btrfs_delayed_ref_node *ref1, + const struct btrfs_delayed_ref_node *ref2) { if (ref1->data_ref.objectid < ref2->data_ref.objectid) return -1; @@ -279,8 +282,8 @@ static int comp_data_refs(struct btrfs_delayed_ref_node *ref1, return 0; } -static int comp_refs(struct btrfs_delayed_ref_node *ref1, - struct btrfs_delayed_ref_node *ref2, +static int comp_refs(const struct btrfs_delayed_ref_node *ref1, + const struct btrfs_delayed_ref_node *ref2, bool check_seq) { int ret = 0; @@ -314,34 +317,25 @@ static int comp_refs(struct btrfs_delayed_ref_node *ref1, return 0; } +static int cmp_refs_node(const struct rb_node *new, const struct rb_node *exist) +{ + const struct btrfs_delayed_ref_node *new_node = + rb_entry(new, struct btrfs_delayed_ref_node, ref_node); + const struct btrfs_delayed_ref_node *exist_node = + rb_entry(exist, struct btrfs_delayed_ref_node, ref_node); + + return comp_refs(new_node, exist_node, true); +} + static struct btrfs_delayed_ref_node* tree_insert(struct rb_root_cached *root, struct btrfs_delayed_ref_node *ins) { - struct rb_node **p = &root->rb_root.rb_node; struct rb_node *node = &ins->ref_node; - struct rb_node *parent_node = NULL; - struct btrfs_delayed_ref_node *entry; - bool leftmost = true; - - while (*p) { - int comp; - - parent_node = *p; - entry = rb_entry(parent_node, struct btrfs_delayed_ref_node, - ref_node); - comp = comp_refs(ins, entry, true); - if (comp < 0) { - p = &(*p)->rb_left; - } else if (comp > 0) { - p = &(*p)->rb_right; - leftmost = false; - } else { - return entry; - } - } + struct rb_node *exist; - rb_link_node(node, parent_node, p); - rb_insert_color_cached(node, root, leftmost); + exist = rb_find_add_cached(node, root, cmp_refs_node); + if (exist) + return rb_entry(exist, struct btrfs_delayed_ref_node, ref_node); return NULL; } @@ -555,6 +549,32 @@ void btrfs_delete_ref_head(const struct btrfs_fs_info *fs_info, delayed_refs->num_heads_ready--; } +struct btrfs_delayed_ref_node *btrfs_select_delayed_ref(struct btrfs_delayed_ref_head *head) +{ + struct btrfs_delayed_ref_node *ref; + + lockdep_assert_held(&head->mutex); + lockdep_assert_held(&head->lock); + + if (RB_EMPTY_ROOT(&head->ref_tree.rb_root)) + return NULL; + + /* + * Select a delayed ref of type BTRFS_ADD_DELAYED_REF first. + * This is to prevent a ref count from going down to zero, which deletes + * the extent item from the extent tree, when there still are references + * to add, which would fail because they would not find the extent item. + */ + if (!list_empty(&head->ref_add_list)) + return list_first_entry(&head->ref_add_list, + struct btrfs_delayed_ref_node, add_list); + + ref = rb_entry(rb_first_cached(&head->ref_tree), + struct btrfs_delayed_ref_node, ref_node); + ASSERT(list_empty(&ref->add_list)); + return ref; +} + /* * Helper to insert the ref_node to the tail or merge with tail. * @@ -1234,6 +1254,7 @@ void btrfs_destroy_delayed_refs(struct btrfs_transaction *trans) { struct btrfs_delayed_ref_root *delayed_refs = &trans->delayed_refs; struct btrfs_fs_info *fs_info = trans->fs_info; + bool testing = btrfs_is_testing(fs_info); spin_lock(&delayed_refs->lock); while (true) { @@ -1263,7 +1284,7 @@ void btrfs_destroy_delayed_refs(struct btrfs_transaction *trans) spin_unlock(&delayed_refs->lock); mutex_unlock(&head->mutex); - if (pin_bytes) { + if (!testing && pin_bytes) { struct btrfs_block_group *bg; bg = btrfs_lookup_block_group(fs_info, head->bytenr); @@ -1281,8 +1302,7 @@ void btrfs_destroy_delayed_refs(struct btrfs_transaction *trans) spin_lock(&bg->space_info->lock); spin_lock(&bg->lock); bg->pinned += head->num_bytes; - btrfs_space_info_update_bytes_pinned(fs_info, - bg->space_info, + btrfs_space_info_update_bytes_pinned(bg->space_info, head->num_bytes); bg->reserved -= head->num_bytes; bg->space_info->bytes_reserved -= head->num_bytes; @@ -1295,12 +1315,15 @@ void btrfs_destroy_delayed_refs(struct btrfs_transaction *trans) btrfs_error_unpin_extent_range(fs_info, head->bytenr, head->bytenr + head->num_bytes - 1); } - btrfs_cleanup_ref_head_accounting(fs_info, delayed_refs, head); + if (!testing) + btrfs_cleanup_ref_head_accounting(fs_info, delayed_refs, head); btrfs_put_delayed_ref_head(head); cond_resched(); spin_lock(&delayed_refs->lock); } - btrfs_qgroup_destroy_extent_records(trans); + + if (!testing) + btrfs_qgroup_destroy_extent_records(trans); spin_unlock(&delayed_refs->lock); } diff --git a/fs/btrfs/delayed-ref.h b/fs/btrfs/delayed-ref.h index 611fb3388f82..a35067cebb97 100644 --- a/fs/btrfs/delayed-ref.h +++ b/fs/btrfs/delayed-ref.h @@ -402,6 +402,7 @@ struct btrfs_delayed_ref_head *btrfs_select_ref_head( struct btrfs_delayed_ref_root *delayed_refs); void btrfs_unselect_ref_head(struct btrfs_delayed_ref_root *delayed_refs, struct btrfs_delayed_ref_head *head); +struct btrfs_delayed_ref_node *btrfs_select_delayed_ref(struct btrfs_delayed_ref_head *head); int btrfs_check_delayed_seq(struct btrfs_fs_info *fs_info, u64 seq); diff --git a/fs/btrfs/dev-replace.c b/fs/btrfs/dev-replace.c index ac8e97ed13f7..f86fbea0b3de 100644 --- a/fs/btrfs/dev-replace.c +++ b/fs/btrfs/dev-replace.c @@ -440,9 +440,6 @@ int btrfs_run_dev_replace(struct btrfs_trans_handle *trans) dev_replace->cursor_right); dev_replace->item_needs_writeback = 0; up_write(&dev_replace->rwsem); - - btrfs_mark_buffer_dirty(trans, eb); - out: btrfs_free_path(path); diff --git a/fs/btrfs/dir-item.c b/fs/btrfs/dir-item.c index 1ea5d8fcfbf7..ccf91de29f80 100644 --- a/fs/btrfs/dir-item.c +++ b/fs/btrfs/dir-item.c @@ -92,7 +92,6 @@ int btrfs_insert_xattr_item(struct btrfs_trans_handle *trans, write_extent_buffer(leaf, name, name_ptr, name_len); write_extent_buffer(leaf, data, data_ptr, data_len); - btrfs_mark_buffer_dirty(trans, path->nodes[0]); return ret; } @@ -152,7 +151,6 @@ int btrfs_insert_dir_item(struct btrfs_trans_handle *trans, name_ptr = (unsigned long)(dir_item + 1); write_extent_buffer(leaf, name->name, name_ptr, name->len); - btrfs_mark_buffer_dirty(trans, leaf); second_insert: /* FIXME, use some real flag for selecting the extra index */ diff --git a/fs/btrfs/direct-io.c b/fs/btrfs/direct-io.c index a7c3e221378d..8567af46e16f 100644 --- a/fs/btrfs/direct-io.c +++ b/fs/btrfs/direct-io.c @@ -248,8 +248,7 @@ static int btrfs_get_blocks_direct_write(struct extent_map **map, len = min(len, em->len - (start - em->start)); block_start = extent_map_block_start(em) + (start - em->start); - if (can_nocow_extent(inode, start, &len, - &file_extent, false, false) == 1) { + if (can_nocow_extent(inode, start, &len, &file_extent, false) == 1) { bg = btrfs_inc_nocow_writers(fs_info, block_start); if (bg) can_nocow = true; diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index eff0dd1ae62f..f09db62e61a1 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -226,7 +226,7 @@ int btrfs_read_extent_buffer(struct extent_buffer *eb, while (1) { clear_bit(EXTENT_BUFFER_CORRUPT, &eb->bflags); - ret = read_extent_buffer_pages(eb, WAIT_COMPLETE, mirror_num, check); + ret = read_extent_buffer_pages(eb, mirror_num, check); if (!ret) break; @@ -1258,6 +1258,7 @@ void btrfs_free_fs_info(struct btrfs_fs_info *fs_info) { struct percpu_counter *em_counter = &fs_info->evictable_extent_maps; + percpu_counter_destroy(&fs_info->stats_read_blocks); percpu_counter_destroy(&fs_info->dirty_metadata_bytes); percpu_counter_destroy(&fs_info->delalloc_bytes); percpu_counter_destroy(&fs_info->ordered_bytes); @@ -2327,6 +2328,71 @@ out: return ret; } +static int validate_sys_chunk_array(const struct btrfs_fs_info *fs_info, + const struct btrfs_super_block *sb) +{ + unsigned int cur = 0; /* Offset inside the sys chunk array */ + /* + * At sb read time, fs_info is not fully initialized. Thus we have + * to use super block sectorsize, which should have been validated. + */ + const u32 sectorsize = btrfs_super_sectorsize(sb); + u32 sys_array_size = btrfs_super_sys_array_size(sb); + + if (sys_array_size > BTRFS_SYSTEM_CHUNK_ARRAY_SIZE) { + btrfs_err(fs_info, "system chunk array too big %u > %u", + sys_array_size, BTRFS_SYSTEM_CHUNK_ARRAY_SIZE); + return -EUCLEAN; + } + + while (cur < sys_array_size) { + struct btrfs_disk_key *disk_key; + struct btrfs_chunk *chunk; + struct btrfs_key key; + u64 type; + u16 num_stripes; + u32 len; + int ret; + + disk_key = (struct btrfs_disk_key *)(sb->sys_chunk_array + cur); + len = sizeof(*disk_key); + + if (cur + len > sys_array_size) + goto short_read; + cur += len; + + btrfs_disk_key_to_cpu(&key, disk_key); + if (key.type != BTRFS_CHUNK_ITEM_KEY) { + btrfs_err(fs_info, + "unexpected item type %u in sys_array at offset %u", + key.type, cur); + return -EUCLEAN; + } + chunk = (struct btrfs_chunk *)(sb->sys_chunk_array + cur); + num_stripes = btrfs_stack_chunk_num_stripes(chunk); + if (cur + btrfs_chunk_item_size(num_stripes) > sys_array_size) + goto short_read; + type = btrfs_stack_chunk_type(chunk); + if (!(type & BTRFS_BLOCK_GROUP_SYSTEM)) { + btrfs_err(fs_info, + "invalid chunk type %llu in sys_array at offset %u", + type, cur); + return -EUCLEAN; + } + ret = btrfs_check_chunk_valid(fs_info, NULL, chunk, key.offset, + sectorsize); + if (ret < 0) + return ret; + cur += btrfs_chunk_item_size(num_stripes); + } + return 0; +short_read: + btrfs_err(fs_info, + "super block sys chunk array short read, cur=%u sys_array_size=%u", + cur, sys_array_size); + return -EUCLEAN; +} + /* * Real super block validation * NOTE: super csum type and incompat features will not be checked here. @@ -2495,6 +2561,8 @@ int btrfs_validate_super(const struct btrfs_fs_info *fs_info, ret = -EINVAL; } + ret = validate_sys_chunk_array(fs_info, sb); + /* * Obvious sys_chunk_array corruptions, it must hold at least one key * and one chunk @@ -2856,6 +2924,10 @@ static int init_mount_fs_info(struct btrfs_fs_info *fs_info, struct super_block if (ret) return ret; + ret = percpu_counter_init(&fs_info->stats_read_blocks, 0, GFP_KERNEL); + if (ret) + return ret; + fs_info->dirty_metadata_batch = PAGE_SIZE * (1 + ilog2(nr_cpu_ids)); @@ -3321,6 +3393,7 @@ int __cold open_ctree(struct super_block *sb, struct btrfs_fs_devices *fs_device fs_info->sectors_per_page = (PAGE_SIZE >> fs_info->sectorsize_bits); fs_info->csums_per_leaf = BTRFS_MAX_ITEM_SIZE(fs_info) / fs_info->csum_size; fs_info->stripesize = stripesize; + fs_info->fs_devices->fs_info = fs_info; /* * Handle the space caching options appropriately now that we have the diff --git a/fs/btrfs/disk-io.h b/fs/btrfs/disk-io.h index a7051e2570c1..587842991b24 100644 --- a/fs/btrfs/disk-io.h +++ b/fs/btrfs/disk-io.h @@ -96,9 +96,6 @@ struct btrfs_root *btrfs_alloc_dummy_root(struct btrfs_fs_info *fs_info); /* * This function is used to grab the root, and avoid it is freed when we * access it. But it doesn't ensure that the tree is not dropped. - * - * If you want to ensure the whole tree is safe, you should use - * fs_info->subvol_srcu */ static inline struct btrfs_root *btrfs_grab_root(struct btrfs_root *root) { diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 3c6f7fecbb9a..3014a1a23efd 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -570,7 +570,6 @@ static noinline int insert_extent_data_ref(struct btrfs_trans_handle *trans, btrfs_set_extent_data_ref_count(leaf, ref, num_refs); } } - btrfs_mark_buffer_dirty(trans, leaf); ret = 0; fail: btrfs_release_path(path); @@ -618,7 +617,6 @@ static noinline int remove_extent_data_ref(struct btrfs_trans_handle *trans, btrfs_set_extent_data_ref_count(leaf, ref1, num_refs); else if (key.type == BTRFS_SHARED_DATA_REF_KEY) btrfs_set_shared_data_ref_count(leaf, ref2, num_refs); - btrfs_mark_buffer_dirty(trans, leaf); } return ret; } @@ -1050,7 +1048,6 @@ void setup_inline_extent_backref(struct btrfs_trans_handle *trans, } else { btrfs_set_extent_inline_ref_offset(leaf, iref, root_objectid); } - btrfs_mark_buffer_dirty(trans, leaf); } static int lookup_extent_backref(struct btrfs_trans_handle *trans, @@ -1195,7 +1192,6 @@ static noinline_for_stack int update_inline_extent_backref( item_size -= size; btrfs_truncate_item(trans, path, item_size, 1); } - btrfs_mark_buffer_dirty(trans, leaf); return 0; } @@ -1260,12 +1256,12 @@ static int btrfs_issue_discard(struct block_device *bdev, u64 start, u64 len, { int j, ret = 0; u64 bytes_left, end; - u64 aligned_start = ALIGN(start, 1 << SECTOR_SHIFT); + u64 aligned_start = ALIGN(start, SECTOR_SIZE); /* Adjust the range to be aligned to 512B sectors if necessary. */ if (start != aligned_start) { len -= aligned_start - start; - len = round_down(len, 1 << SECTOR_SHIFT); + len = round_down(len, SECTOR_SIZE); start = aligned_start; } @@ -1527,7 +1523,6 @@ static int __btrfs_inc_extent_ref(struct btrfs_trans_handle *trans, if (extent_op) __run_delayed_extent_op(extent_op, leaf, item); - btrfs_mark_buffer_dirty(trans, leaf); btrfs_release_path(path); /* now insert the actual backref */ @@ -1711,8 +1706,6 @@ again: ei = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_extent_item); __run_delayed_extent_op(extent_op, leaf, ei); - - btrfs_mark_buffer_dirty(trans, leaf); out: btrfs_free_path(path); return ret; @@ -1803,30 +1796,6 @@ static int run_one_delayed_ref(struct btrfs_trans_handle *trans, return ret; } -static inline struct btrfs_delayed_ref_node * -select_delayed_ref(struct btrfs_delayed_ref_head *head) -{ - struct btrfs_delayed_ref_node *ref; - - if (RB_EMPTY_ROOT(&head->ref_tree.rb_root)) - return NULL; - - /* - * Select a delayed ref of type BTRFS_ADD_DELAYED_REF first. - * This is to prevent a ref count from going down to zero, which deletes - * the extent item from the extent tree, when there still are references - * to add, which would fail because they would not find the extent item. - */ - if (!list_empty(&head->ref_add_list)) - return list_first_entry(&head->ref_add_list, - struct btrfs_delayed_ref_node, add_list); - - ref = rb_entry(rb_first_cached(&head->ref_tree), - struct btrfs_delayed_ref_node, ref_node); - ASSERT(list_empty(&ref->add_list)); - return ref; -} - static struct btrfs_delayed_extent_op *cleanup_extent_op( struct btrfs_delayed_ref_head *head) { @@ -1959,7 +1928,7 @@ static int btrfs_run_delayed_refs_for_head(struct btrfs_trans_handle *trans, lockdep_assert_held(&locked_ref->mutex); lockdep_assert_held(&locked_ref->lock); - while ((ref = select_delayed_ref(locked_ref))) { + while ((ref = btrfs_select_delayed_ref(locked_ref))) { if (ref->seq && btrfs_check_delayed_seq(fs_info, ref->seq)) { spin_unlock(&locked_ref->lock); @@ -2230,10 +2199,11 @@ int btrfs_set_disk_extent_flags(struct btrfs_trans_handle *trans, return ret; } -static noinline int check_delayed_ref(struct btrfs_root *root, +static noinline int check_delayed_ref(struct btrfs_inode *inode, struct btrfs_path *path, - u64 objectid, u64 offset, u64 bytenr) + u64 offset, u64 bytenr) { + struct btrfs_root *root = inode->root; struct btrfs_delayed_ref_head *head; struct btrfs_delayed_ref_node *ref; struct btrfs_delayed_ref_root *delayed_refs; @@ -2307,7 +2277,7 @@ static noinline int check_delayed_ref(struct btrfs_root *root, * then we have a cross reference. */ if (ref->ref_root != btrfs_root_id(root) || - ref_owner != objectid || ref_offset != offset) { + ref_owner != btrfs_ino(inode) || ref_offset != offset) { ret = 1; break; } @@ -2318,11 +2288,53 @@ static noinline int check_delayed_ref(struct btrfs_root *root, return ret; } -static noinline int check_committed_ref(struct btrfs_root *root, +/* + * Check if there are references for a data extent other than the one belonging + * to the given inode and offset. + * + * @inode: The only inode we expect to find associated with the data extent. + * @path: A path to use for searching the extent tree. + * @offset: The only offset we expect to find associated with the data extent. + * @bytenr: The logical address of the data extent. + * + * When the extent does not have any other references other than the one we + * expect to find, we always return a value of 0 with the path having a locked + * leaf that contains the extent's extent item - this is necessary to ensure + * we don't race with a task running delayed references, and our caller must + * have such a path when calling check_delayed_ref() - it must lock a delayed + * ref head while holding the leaf locked. In case the extent item is not found + * in the extent tree, we return -ENOENT with the path having the leaf (locked) + * where the extent item should be, in order to prevent races with another task + * running delayed references, so that we don't miss any reference when calling + * check_delayed_ref(). + * + * Note: this may return false positives, and this is because we want to be + * quick here as we're called in write paths (when flushing delalloc and + * in the direct IO write path). For example we can have an extent with + * a single reference but that reference is not inlined, or we may have + * many references in the extent tree but we also have delayed references + * that cancel all the reference except the one for our inode and offset, + * but it would be expensive to do such checks and complex due to all + * locking to avoid races between the checks and flushing delayed refs, + * plus non-inline references may be located on leaves other than the one + * that contains the extent item in the extent tree. The important thing + * here is to not return false negatives and that the false positives are + * not very common. + * + * Returns: 0 if there are no cross references and with the path having a locked + * leaf from the extent tree that contains the extent's extent item. + * + * 1 if there are cross references (false positives can happen). + * + * < 0 in case of an error. In case of -ENOENT the leaf in the extent + * tree where the extent item should be located at is read locked and + * accessible in the given path. + */ +static noinline int check_committed_ref(struct btrfs_inode *inode, struct btrfs_path *path, - u64 objectid, u64 offset, u64 bytenr, - bool strict) + u64 offset, u64 bytenr) { + struct btrfs_root *root = inode->root; struct btrfs_fs_info *fs_info = root->fs_info; struct btrfs_root *extent_root = btrfs_extent_root(fs_info, bytenr); struct extent_buffer *leaf; @@ -2341,35 +2353,32 @@ static noinline int check_committed_ref(struct btrfs_root *root, ret = btrfs_search_slot(NULL, extent_root, &key, path, 0, 0); if (ret < 0) - goto out; + return ret; if (ret == 0) { /* * Key with offset -1 found, there would have to exist an extent * item with such offset, but this is out of the valid range. */ - ret = -EUCLEAN; - goto out; + return -EUCLEAN; } - ret = -ENOENT; if (path->slots[0] == 0) - goto out; + return -ENOENT; path->slots[0]--; leaf = path->nodes[0]; btrfs_item_key_to_cpu(leaf, &key, path->slots[0]); if (key.objectid != bytenr || key.type != BTRFS_EXTENT_ITEM_KEY) - goto out; + return -ENOENT; - ret = 1; item_size = btrfs_item_size(leaf, path->slots[0]); ei = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_extent_item); expected_size = sizeof(*ei) + btrfs_extent_inline_ref_size(BTRFS_EXTENT_DATA_REF_KEY); /* No inline refs; we need to bail before checking for owner ref. */ if (item_size == sizeof(*ei)) - goto out; + return 1; /* Check for an owner ref; skip over it to the real inline refs. */ iref = (struct btrfs_extent_inline_ref *)(ei + 1); @@ -2377,56 +2386,69 @@ static noinline int check_committed_ref(struct btrfs_root *root, if (btrfs_fs_incompat(fs_info, SIMPLE_QUOTA) && type == BTRFS_EXTENT_OWNER_REF_KEY) { expected_size += btrfs_extent_inline_ref_size(BTRFS_EXTENT_OWNER_REF_KEY); iref = (struct btrfs_extent_inline_ref *)(iref + 1); + type = btrfs_get_extent_inline_ref_type(leaf, iref, BTRFS_REF_TYPE_DATA); } /* If extent item has more than 1 inline ref then it's shared */ if (item_size != expected_size) - goto out; - - /* - * If extent created before last snapshot => it's shared unless the - * snapshot has been deleted. Use the heuristic if strict is false. - */ - if (!strict && - (btrfs_extent_generation(leaf, ei) <= - btrfs_root_last_snapshot(&root->root_item))) - goto out; + return 1; /* If this extent has SHARED_DATA_REF then it's shared */ - type = btrfs_get_extent_inline_ref_type(leaf, iref, BTRFS_REF_TYPE_DATA); if (type != BTRFS_EXTENT_DATA_REF_KEY) - goto out; + return 1; ref = (struct btrfs_extent_data_ref *)(&iref->offset); if (btrfs_extent_refs(leaf, ei) != btrfs_extent_data_ref_count(leaf, ref) || btrfs_extent_data_ref_root(leaf, ref) != btrfs_root_id(root) || - btrfs_extent_data_ref_objectid(leaf, ref) != objectid || + btrfs_extent_data_ref_objectid(leaf, ref) != btrfs_ino(inode) || btrfs_extent_data_ref_offset(leaf, ref) != offset) - goto out; + return 1; - ret = 0; -out: - return ret; + return 0; } -int btrfs_cross_ref_exist(struct btrfs_root *root, u64 objectid, u64 offset, - u64 bytenr, bool strict, struct btrfs_path *path) +int btrfs_cross_ref_exist(struct btrfs_inode *inode, u64 offset, + u64 bytenr, struct btrfs_path *path) { int ret; do { - ret = check_committed_ref(root, path, objectid, - offset, bytenr, strict); + ret = check_committed_ref(inode, path, offset, bytenr); if (ret && ret != -ENOENT) goto out; - ret = check_delayed_ref(root, path, objectid, offset, bytenr); + /* + * The path must have a locked leaf from the extent tree where + * the extent item for our extent is located, in case it exists, + * or where it should be located in case it doesn't exist yet + * because it's new and its delayed ref was not yet flushed. + * We need to lock the delayed ref head at check_delayed_ref(), + * if one exists, while holding the leaf locked in order to not + * race with delayed ref flushing, missing references and + * incorrectly reporting that the extent is not shared. + */ + if (IS_ENABLED(CONFIG_BTRFS_ASSERT)) { + struct extent_buffer *leaf = path->nodes[0]; + + ASSERT(leaf != NULL); + btrfs_assert_tree_read_locked(leaf); + + if (ret != -ENOENT) { + struct btrfs_key key; + + btrfs_item_key_to_cpu(leaf, &key, path->slots[0]); + ASSERT(key.objectid == bytenr); + ASSERT(key.type == BTRFS_EXTENT_ITEM_KEY); + } + } + + ret = check_delayed_ref(inode, path, offset, bytenr); } while (ret == -EAGAIN && !path->nowait); out: btrfs_release_path(path); - if (btrfs_is_data_reloc_root(root)) + if (btrfs_is_data_reloc_root(inode->root)) WARN_ON(ret > 0); return ret; } @@ -2571,13 +2593,10 @@ static int pin_down_extent(struct btrfs_trans_handle *trans, struct btrfs_block_group *cache, u64 bytenr, u64 num_bytes, int reserved) { - struct btrfs_fs_info *fs_info = cache->fs_info; - spin_lock(&cache->space_info->lock); spin_lock(&cache->lock); cache->pinned += num_bytes; - btrfs_space_info_update_bytes_pinned(fs_info, cache->space_info, - num_bytes); + btrfs_space_info_update_bytes_pinned(cache->space_info, num_bytes); if (reserved) { cache->reserved -= num_bytes; cache->space_info->bytes_reserved -= num_bytes; @@ -2724,15 +2743,15 @@ static int unpin_extent_range(struct btrfs_fs_info *fs_info, { struct btrfs_block_group *cache = NULL; struct btrfs_space_info *space_info; - struct btrfs_block_rsv *global_rsv = &fs_info->global_block_rsv; struct btrfs_free_cluster *cluster = NULL; - u64 len; u64 total_unpinned = 0; u64 empty_cluster = 0; bool readonly; int ret = 0; while (start <= end) { + u64 len; + readonly = false; if (!cache || start >= cache->start + cache->length) { @@ -2778,37 +2797,19 @@ static int unpin_extent_range(struct btrfs_fs_info *fs_info, spin_lock(&space_info->lock); spin_lock(&cache->lock); cache->pinned -= len; - btrfs_space_info_update_bytes_pinned(fs_info, space_info, -len); + btrfs_space_info_update_bytes_pinned(space_info, -len); space_info->max_extent_size = 0; if (cache->ro) { space_info->bytes_readonly += len; readonly = true; } else if (btrfs_is_zoned(fs_info)) { /* Need reset before reusing in a zoned block group */ - btrfs_space_info_update_bytes_zone_unusable(fs_info, space_info, - len); + btrfs_space_info_update_bytes_zone_unusable(space_info, len); readonly = true; } spin_unlock(&cache->lock); - if (!readonly && return_free_space && - global_rsv->space_info == space_info) { - spin_lock(&global_rsv->lock); - if (!global_rsv->full) { - u64 to_add = min(len, global_rsv->size - - global_rsv->reserved); - - global_rsv->reserved += to_add; - btrfs_space_info_update_bytes_may_use(fs_info, - space_info, to_add); - if (global_rsv->reserved >= global_rsv->size) - global_rsv->full = 1; - len -= to_add; - } - spin_unlock(&global_rsv->lock); - } - /* Add to any tickets we may have */ - if (!readonly && return_free_space && len) - btrfs_try_granting_tickets(fs_info, space_info); + if (!readonly && return_free_space) + btrfs_return_free_space(space_info, len); spin_unlock(&space_info->lock); } @@ -3259,7 +3260,6 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans, } } else { btrfs_set_extent_refs(leaf, ei, refs); - btrfs_mark_buffer_dirty(trans, leaf); } if (found_extent) { ret = remove_extent_backref(trans, extent_root, path, @@ -4827,7 +4827,6 @@ static int alloc_reserved_file_extent(struct btrfs_trans_handle *trans, btrfs_set_extent_data_ref_count(leaf, ref, ref_mod); } - btrfs_mark_buffer_dirty(trans, path->nodes[0]); btrfs_free_path(path); return alloc_reserved_extent(trans, ins->objectid, ins->offset); @@ -4902,7 +4901,6 @@ static int alloc_reserved_tree_block(struct btrfs_trans_handle *trans, btrfs_set_extent_inline_ref_offset(leaf, iref, node->ref_root); } - btrfs_mark_buffer_dirty(trans, leaf); btrfs_free_path(path); return alloc_reserved_extent(trans, node->bytenr, fs_info->nodesize); diff --git a/fs/btrfs/extent-tree.h b/fs/btrfs/extent-tree.h index 2ad51130c037..cfa52264f678 100644 --- a/fs/btrfs/extent-tree.h +++ b/fs/btrfs/extent-tree.h @@ -116,8 +116,7 @@ int btrfs_pin_extent(struct btrfs_trans_handle *trans, u64 bytenr, u64 num, int btrfs_pin_extent_for_log_replay(struct btrfs_trans_handle *trans, const struct extent_buffer *eb); int btrfs_exclude_logged_extents(struct extent_buffer *eb); -int btrfs_cross_ref_exist(struct btrfs_root *root, - u64 objectid, u64 offset, u64 bytenr, bool strict, +int btrfs_cross_ref_exist(struct btrfs_inode *inode, u64 offset, u64 bytenr, struct btrfs_path *path); struct extent_buffer *btrfs_alloc_tree_block(struct btrfs_trans_handle *trans, struct btrfs_root *root, @@ -163,5 +162,9 @@ int btrfs_drop_subtree(struct btrfs_trans_handle *trans, struct btrfs_root *root, struct extent_buffer *node, struct extent_buffer *parent); +void btrfs_error_unpin_extent_range(struct btrfs_fs_info *fs_info, u64 start, u64 end); +int btrfs_discard_extent(struct btrfs_fs_info *fs_info, u64 bytenr, + u64 num_bytes, u64 *actual_bytes); +int btrfs_trim_fs(struct btrfs_fs_info *fs_info, struct fstrim_range *range); #endif diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index b923d0cec61c..d9f856358704 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -198,9 +198,8 @@ static void __process_folios_contig(struct address_space *mapping, u64 end, unsigned long page_ops) { struct btrfs_fs_info *fs_info = inode_to_fs_info(mapping->host); - pgoff_t start_index = start >> PAGE_SHIFT; + pgoff_t index = start >> PAGE_SHIFT; pgoff_t end_index = end >> PAGE_SHIFT; - pgoff_t index = start_index; struct folio_batch fbatch; int i; @@ -221,7 +220,7 @@ static void __process_folios_contig(struct address_space *mapping, } } -static noinline void __unlock_for_delalloc(const struct inode *inode, +static noinline void unlock_delalloc_folio(const struct inode *inode, const struct folio *locked_folio, u64 start, u64 end) { @@ -242,9 +241,8 @@ static noinline int lock_delalloc_folios(struct inode *inode, { struct btrfs_fs_info *fs_info = inode_to_fs_info(inode); struct address_space *mapping = inode->i_mapping; - pgoff_t start_index = start >> PAGE_SHIFT; + pgoff_t index = start >> PAGE_SHIFT; pgoff_t end_index = end >> PAGE_SHIFT; - pgoff_t index = start_index; u64 processed_end = start; struct folio_batch fbatch; @@ -288,8 +286,7 @@ static noinline int lock_delalloc_folios(struct inode *inode, out: folio_batch_release(&fbatch); if (processed_end > start) - __unlock_for_delalloc(inode, locked_folio, start, - processed_end); + unlock_delalloc_folio(inode, locked_folio, start, processed_end); return -EAGAIN; } @@ -390,7 +387,7 @@ again: unlock_extent(tree, delalloc_start, delalloc_end, &cached_state); if (!ret) { - __unlock_for_delalloc(inode, locked_folio, delalloc_start, + unlock_delalloc_folio(inode, locked_folio, delalloc_start, delalloc_end); cond_resched(); goto again; @@ -710,6 +707,7 @@ static void alloc_new_bio(struct btrfs_inode *inode, bbio = btrfs_bio_alloc(BIO_MAX_VECS, bio_ctrl->opf, fs_info, bio_ctrl->end_io_func, NULL); bbio->bio.bi_iter.bi_sector = disk_bytenr >> SECTOR_SHIFT; + bbio->bio.bi_write_hint = inode->vfs_inode.i_write_hint; bbio->inode = inode; bbio->file_offset = file_offset; bio_ctrl->bbio = bbio; @@ -862,11 +860,6 @@ static int attach_extent_buffer_folio(struct extent_buffer *eb, return ret; } -int set_page_extent_mapped(struct page *page) -{ - return set_folio_extent_mapped(page_folio(page)); -} - int set_folio_extent_mapped(struct folio *folio) { struct btrfs_fs_info *fs_info; @@ -901,9 +894,9 @@ void clear_folio_extent_mapped(struct folio *folio) folio_detach_private(folio); } -static struct extent_map *__get_extent_map(struct inode *inode, - struct folio *folio, u64 start, - u64 len, struct extent_map **em_cached) +static struct extent_map *get_extent_map(struct btrfs_inode *inode, + struct folio *folio, u64 start, + u64 len, struct extent_map **em_cached) { struct extent_map *em; struct extent_state *cached_state = NULL; @@ -922,14 +915,14 @@ static struct extent_map *__get_extent_map(struct inode *inode, *em_cached = NULL; } - btrfs_lock_and_flush_ordered_range(BTRFS_I(inode), start, start + len - 1, &cached_state); - em = btrfs_get_extent(BTRFS_I(inode), folio, start, len); + btrfs_lock_and_flush_ordered_range(inode, start, start + len - 1, &cached_state); + em = btrfs_get_extent(inode, folio, start, len); if (!IS_ERR(em)) { BUG_ON(*em_cached); refcount_inc(&em->refs); *em_cached = em; } - unlock_extent(&BTRFS_I(inode)->io_tree, start, start + len - 1, &cached_state); + unlock_extent(&inode->io_tree, start, start + len - 1, &cached_state); return em; } @@ -985,8 +978,7 @@ static int btrfs_do_readpage(struct folio *folio, struct extent_map **em_cached, end_folio_read(folio, true, cur, iosize); break; } - em = __get_extent_map(inode, folio, cur, end - cur + 1, - em_cached); + em = get_extent_map(BTRFS_I(inode), folio, cur, end - cur + 1, em_cached); if (IS_ERR(em)) { end_folio_read(folio, false, cur, end + 1 - cur); return PTR_ERR(em); @@ -1142,14 +1134,19 @@ static bool find_next_delalloc_bitmap(struct folio *folio, } /* - * helper for extent_writepage(), doing all of the delayed allocation setup. + * Do all of the delayed allocation setup. * - * This returns 1 if btrfs_run_delalloc_range function did all the work required - * to write the page (copy into inline extent). In this case the IO has - * been started and the page is already unlocked. + * Return >0 if all the dirty blocks are submitted async (compression) or inlined. + * The @folio should no longer be touched (treat it as already unlocked). * - * This returns 0 if all went well (page still locked) - * This returns < 0 if there were errors (page still locked) + * Return 0 if there is still dirty block that needs to be submitted through + * extent_writepage_io(). + * bio_ctrl->submit_bitmap will indicate which blocks of the folio should be + * submitted, and @folio is still kept locked. + * + * Return <0 if there is any error hit. + * Any allocated ordered extent range covering this folio will be marked + * finished (IOERR), and @folio is still kept locked. */ static noinline_for_stack int writepage_delalloc(struct btrfs_inode *inode, struct folio *folio, @@ -1167,6 +1164,16 @@ static noinline_for_stack int writepage_delalloc(struct btrfs_inode *inode, * last delalloc end. */ u64 last_delalloc_end = 0; + /* + * The range end (exclusive) of the last successfully finished delalloc + * range. + * Any range covered by ordered extent must either be manually marked + * finished (error handling), or has IO submitted (and finish the + * ordered extent normally). + * + * This records the end of ordered extent cleanup if we hit an error. + */ + u64 last_finished_delalloc_end = page_start; u64 delalloc_start = page_start; u64 delalloc_end = page_end; u64 delalloc_to_write = 0; @@ -1235,11 +1242,28 @@ static noinline_for_stack int writepage_delalloc(struct btrfs_inode *inode, found_len = last_delalloc_end + 1 - found_start; if (ret >= 0) { + /* + * Some delalloc range may be created by previous folios. + * Thus we still need to clean up this range during error + * handling. + */ + last_finished_delalloc_end = found_start; /* No errors hit so far, run the current delalloc range. */ ret = btrfs_run_delalloc_range(inode, folio, found_start, found_start + found_len - 1, wbc); + if (ret >= 0) + last_finished_delalloc_end = found_start + found_len; + if (unlikely(ret < 0)) + btrfs_err_rl(fs_info, +"failed to run delalloc range, root=%lld ino=%llu folio=%llu submit_bitmap=%*pbl start=%llu len=%u: %d", + btrfs_root_id(inode->root), + btrfs_ino(inode), + folio_pos(folio), + fs_info->sectors_per_page, + &bio_ctrl->submit_bitmap, + found_start, found_len, ret); } else { /* * We've hit an error during previous delalloc range, @@ -1247,7 +1271,7 @@ static noinline_for_stack int writepage_delalloc(struct btrfs_inode *inode, */ unlock_extent(&inode->io_tree, found_start, found_start + found_len - 1, NULL); - __unlock_for_delalloc(&inode->vfs_inode, folio, + unlock_delalloc_folio(&inode->vfs_inode, folio, found_start, found_start + found_len - 1); } @@ -1274,8 +1298,22 @@ static noinline_for_stack int writepage_delalloc(struct btrfs_inode *inode, delalloc_start = found_start + found_len; } - if (ret < 0) + /* + * It's possible we had some ordered extents created before we hit + * an error, cleanup non-async successfully created delalloc ranges. + */ + if (unlikely(ret < 0)) { + unsigned int bitmap_size = min( + (last_finished_delalloc_end - page_start) >> + fs_info->sectorsize_bits, + fs_info->sectors_per_page); + + for_each_set_bit(bit, &bio_ctrl->submit_bitmap, bitmap_size) + btrfs_mark_ordered_io_finished(inode, folio, + page_start + (bit << fs_info->sectorsize_bits), + fs_info->sectorsize, false); return ret; + } out: if (last_delalloc_end) delalloc_end = last_delalloc_end; @@ -1335,7 +1373,7 @@ static int submit_one_sector(struct btrfs_inode *inode, em = btrfs_get_extent(inode, NULL, filepos, sectorsize); if (IS_ERR(em)) - return PTR_ERR_OR_ZERO(em); + return PTR_ERR(em); extent_offset = filepos - em->start; em_end = extent_map_end(em); @@ -1391,6 +1429,7 @@ static noinline_for_stack int extent_writepage_io(struct btrfs_inode *inode, struct btrfs_fs_info *fs_info = inode->root->fs_info; unsigned long range_bitmap = 0; bool submitted_io = false; + bool error = false; const u64 folio_start = folio_pos(folio); u64 cur; int bit; @@ -1433,11 +1472,26 @@ static noinline_for_stack int extent_writepage_io(struct btrfs_inode *inode, break; } ret = submit_one_sector(inode, folio, cur, bio_ctrl, i_size); - if (ret < 0) - goto out; + if (unlikely(ret < 0)) { + /* + * bio_ctrl may contain a bio crossing several folios. + * Submit it immediately so that the bio has a chance + * to finish normally, other than marked as error. + */ + submit_one_bio(bio_ctrl); + /* + * Failed to grab the extent map which should be very rare. + * Since there is no bio submitted to finish the ordered + * extent, we have to manually finish this sector. + */ + btrfs_mark_ordered_io_finished(inode, folio, cur, + fs_info->sectorsize, false); + error = true; + continue; + } submitted_io = true; } -out: + /* * If we didn't submitted any sector (>= i_size), folio dirty get * cleared but PAGECACHE_TAG_DIRTY is not cleared (only cleared @@ -1445,8 +1499,11 @@ out: * * Here we set writeback and clear for the range. If the full folio * is no longer dirty then we clear the PAGECACHE_TAG_DIRTY tag. + * + * If we hit any error, the corresponding sector will still be dirty + * thus no need to clear PAGECACHE_TAG_DIRTY. */ - if (!submitted_io) { + if (!submitted_io && !error) { btrfs_folio_set_writeback(fs_info, folio, start, len); btrfs_folio_clear_writeback(fs_info, folio, start, len); } @@ -1464,15 +1521,14 @@ out: */ static int extent_writepage(struct folio *folio, struct btrfs_bio_ctrl *bio_ctrl) { - struct inode *inode = folio->mapping->host; - struct btrfs_fs_info *fs_info = inode_to_fs_info(inode); - const u64 page_start = folio_pos(folio); + struct btrfs_inode *inode = BTRFS_I(folio->mapping->host); + struct btrfs_fs_info *fs_info = inode->root->fs_info; int ret; size_t pg_offset; - loff_t i_size = i_size_read(inode); + loff_t i_size = i_size_read(&inode->vfs_inode); unsigned long end_index = i_size >> PAGE_SHIFT; - trace_extent_writepage(folio, inode, bio_ctrl->wbc); + trace_extent_writepage(folio, &inode->vfs_inode, bio_ctrl->wbc); WARN_ON(!folio_test_locked(folio)); @@ -1496,26 +1552,28 @@ static int extent_writepage(struct folio *folio, struct btrfs_bio_ctrl *bio_ctrl if (ret < 0) goto done; - ret = writepage_delalloc(BTRFS_I(inode), folio, bio_ctrl); + ret = writepage_delalloc(inode, folio, bio_ctrl); if (ret == 1) return 0; if (ret) goto done; - ret = extent_writepage_io(BTRFS_I(inode), folio, folio_pos(folio), + ret = extent_writepage_io(inode, folio, folio_pos(folio), PAGE_SIZE, bio_ctrl, i_size); if (ret == 1) return 0; + if (ret < 0) + btrfs_err_rl(fs_info, +"failed to submit blocks, root=%lld inode=%llu folio=%llu submit_bitmap=%*pbl: %d", + btrfs_root_id(inode->root), btrfs_ino(inode), + folio_pos(folio), fs_info->sectors_per_page, + &bio_ctrl->submit_bitmap, ret); bio_ctrl->wbc->nr_to_write--; done: - if (ret) { - btrfs_mark_ordered_io_finished(BTRFS_I(inode), folio, - page_start, PAGE_SIZE, !ret); + if (ret < 0) mapping_set_error(folio->mapping, ret); - } - /* * Only unlock ranges that are submitted. As there can be some async * submitted ranges inside the folio. @@ -1525,12 +1583,6 @@ done: return ret; } -void wait_on_extent_buffer_writeback(struct extent_buffer *eb) -{ - wait_on_bit_io(&eb->bflags, EXTENT_BUFFER_WRITEBACK, - TASK_UNINTERRUPTIBLE); -} - /* * Lock extent buffer status and pages for writeback. * @@ -1671,11 +1723,10 @@ static void end_bbio_meta_write(struct btrfs_bio *bbio) { struct extent_buffer *eb = bbio->private; struct btrfs_fs_info *fs_info = eb->fs_info; - bool uptodate = !bbio->bio.bi_status; struct folio_iter fi; u32 bio_offset = 0; - if (!uptodate) + if (bbio->bio.bi_status != BLK_STS_OK) set_btree_ioerr(eb); bio_for_each_folio_all(fi, &bbio->bio) { @@ -2292,11 +2343,8 @@ void extent_write_locked_range(struct inode *inode, const struct folio *locked_f if (ret == 1) goto next_page; - if (ret) { - btrfs_mark_ordered_io_finished(BTRFS_I(inode), folio, - cur, cur_len, !ret); + if (ret) mapping_set_error(mapping, ret); - } btrfs_folio_end_lock(fs_info, folio, cur, cur_len); if (ret < 0) found_error = true; @@ -2495,11 +2543,6 @@ next: return try_release_extent_state(io_tree, folio); } -static void __free_extent_buffer(struct extent_buffer *eb) -{ - kmem_cache_free(extent_buffer_cache, eb); -} - static int extent_buffer_under_io(const struct extent_buffer *eb) { return (test_bit(EXTENT_BUFFER_WRITEBACK, &eb->bflags) || @@ -2580,8 +2623,8 @@ static void detach_extent_buffer_folio(const struct extent_buffer *eb, struct fo spin_unlock(&folio->mapping->i_private_lock); } -/* Release all pages attached to the extent buffer */ -static void btrfs_release_extent_buffer_pages(const struct extent_buffer *eb) +/* Release all folios attached to the extent buffer */ +static void btrfs_release_extent_buffer_folios(const struct extent_buffer *eb) { ASSERT(!extent_buffer_under_io(eb)); @@ -2603,9 +2646,9 @@ static void btrfs_release_extent_buffer_pages(const struct extent_buffer *eb) */ static inline void btrfs_release_extent_buffer(struct extent_buffer *eb) { - btrfs_release_extent_buffer_pages(eb); + btrfs_release_extent_buffer_folios(eb); btrfs_leak_debug_del_eb(eb); - __free_extent_buffer(eb); + kmem_cache_free(extent_buffer_cache, eb); } static struct extent_buffer * @@ -2703,7 +2746,7 @@ err: folio_put(eb->folios[i]); } } - __free_extent_buffer(eb); + kmem_cache_free(extent_buffer_cache, eb); return NULL; } @@ -2830,13 +2873,12 @@ free_eb: } #endif -static struct extent_buffer *grab_extent_buffer( - struct btrfs_fs_info *fs_info, struct page *page) +static struct extent_buffer *grab_extent_buffer(struct btrfs_fs_info *fs_info, + struct folio *folio) { - struct folio *folio = page_folio(page); struct extent_buffer *exists; - lockdep_assert_held(&page->mapping->i_private_lock); + lockdep_assert_held(&folio->mapping->i_private_lock); /* * For subpage case, we completely rely on radix tree to ensure we @@ -2851,7 +2893,7 @@ static struct extent_buffer *grab_extent_buffer( return NULL; /* - * We could have already allocated an eb for this page and attached one + * We could have already allocated an eb for this folio and attached one * so lets see if we can get a ref on the existing eb, and if we can we * know it's good and we can just return that one, else we know we can * just overwrite folio private. @@ -2860,16 +2902,19 @@ static struct extent_buffer *grab_extent_buffer( if (atomic_inc_not_zero(&exists->refs)) return exists; - WARN_ON(PageDirty(page)); + WARN_ON(folio_test_dirty(folio)); folio_detach_private(folio); return NULL; } -static int check_eb_alignment(struct btrfs_fs_info *fs_info, u64 start) +/* + * Validate alignment constraints of eb at logical address @start. + */ +static bool check_eb_alignment(struct btrfs_fs_info *fs_info, u64 start) { if (!IS_ALIGNED(start, fs_info->sectorsize)) { btrfs_err(fs_info, "bad tree block start %llu", start); - return -EINVAL; + return true; } if (fs_info->nodesize < PAGE_SIZE && @@ -2877,14 +2922,14 @@ static int check_eb_alignment(struct btrfs_fs_info *fs_info, u64 start) btrfs_err(fs_info, "tree block crosses page boundary, start %llu nodesize %u", start, fs_info->nodesize); - return -EINVAL; + return true; } if (fs_info->nodesize >= PAGE_SIZE && !PAGE_ALIGNED(start)) { btrfs_err(fs_info, "tree block is not page aligned, start %llu nodesize %u", start, fs_info->nodesize); - return -EINVAL; + return true; } if (!IS_ALIGNED(start, fs_info->nodesize) && !test_and_set_bit(BTRFS_FS_UNALIGNED_TREE_BLOCK, &fs_info->flags)) { @@ -2892,10 +2937,9 @@ static int check_eb_alignment(struct btrfs_fs_info *fs_info, u64 start) "tree block not nodesize aligned, start %llu nodesize %u, can be resolved by a full metadata balance", start, fs_info->nodesize); } - return 0; + return false; } - /* * Return 0 if eb->folios[i] is attached to btree inode successfully. * Return >0 if there is already another extent buffer for the range, @@ -2951,8 +2995,7 @@ finish: } else if (existing_folio) { struct extent_buffer *existing_eb; - existing_eb = grab_extent_buffer(fs_info, - folio_page(existing_folio, 0)); + existing_eb = grab_extent_buffer(fs_info, existing_folio); if (existing_eb) { /* The extent buffer still exists, we can use it directly. */ *found_eb_ret = existing_eb; @@ -3149,7 +3192,7 @@ again: * live buffer and won't free them prematurely. */ for (int i = 0; i < num_folios; i++) - unlock_page(folio_page(eb->folios[i], 0)); + folio_unlock(eb->folios[i]); return eb; out: @@ -3173,7 +3216,7 @@ out: for (int i = 0; i < attached; i++) { ASSERT(eb->folios[i]); detach_extent_buffer_folio(eb, eb->folios[i]); - unlock_page(folio_page(eb->folios[i], 0)); + folio_unlock(eb->folios[i]); folio_put(eb->folios[i]); eb->folios[i] = NULL; } @@ -3195,7 +3238,7 @@ static inline void btrfs_release_extent_buffer_rcu(struct rcu_head *head) struct extent_buffer *eb = container_of(head, struct extent_buffer, rcu_head); - __free_extent_buffer(eb); + kmem_cache_free(extent_buffer_cache, eb); } static int release_extent_buffer(struct extent_buffer *eb) @@ -3219,11 +3262,11 @@ static int release_extent_buffer(struct extent_buffer *eb) } btrfs_leak_debug_del_eb(eb); - /* Should be safe to release our pages at this point */ - btrfs_release_extent_buffer_pages(eb); + /* Should be safe to release folios at this point. */ + btrfs_release_extent_buffer_folios(eb); #ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS if (unlikely(test_bit(EXTENT_BUFFER_UNMAPPED, &eb->bflags))) { - __free_extent_buffer(eb); + kmem_cache_free(extent_buffer_cache, eb); return 1; } #endif @@ -3382,12 +3425,12 @@ void set_extent_buffer_dirty(struct extent_buffer *eb) * the above race. */ if (subpage) - lock_page(folio_page(eb->folios[0], 0)); + folio_lock(eb->folios[0]); for (int i = 0; i < num_folios; i++) btrfs_folio_set_dirty(eb->fs_info, eb->folios[i], eb->start, eb->len); if (subpage) - unlock_page(folio_page(eb->folios[0], 0)); + folio_unlock(eb->folios[0]); percpu_counter_add_batch(&eb->fs_info->dirty_metadata_bytes, eb->len, eb->fs_info->dirty_metadata_batch); @@ -3497,8 +3540,8 @@ static void end_bbio_meta_read(struct btrfs_bio *bbio) bio_put(&bbio->bio); } -int read_extent_buffer_pages(struct extent_buffer *eb, int wait, int mirror_num, - const struct btrfs_tree_parent_check *check) +int read_extent_buffer_pages_nowait(struct extent_buffer *eb, int mirror_num, + const struct btrfs_tree_parent_check *check) { struct btrfs_bio *bbio; bool ret; @@ -3516,7 +3559,7 @@ int read_extent_buffer_pages(struct extent_buffer *eb, int wait, int mirror_num, /* Someone else is already reading the buffer, just wait for it. */ if (test_and_set_bit(EXTENT_BUFFER_READING, &eb->bflags)) - goto done; + return 0; /* * Between the initial test_bit(EXTENT_BUFFER_UPTODATE) and the above @@ -3556,14 +3599,21 @@ int read_extent_buffer_pages(struct extent_buffer *eb, int wait, int mirror_num, } } btrfs_submit_bbio(bbio, mirror_num); + return 0; +} -done: - if (wait == WAIT_COMPLETE) { - wait_on_bit_io(&eb->bflags, EXTENT_BUFFER_READING, TASK_UNINTERRUPTIBLE); - if (!test_bit(EXTENT_BUFFER_UPTODATE, &eb->bflags)) - return -EIO; - } +int read_extent_buffer_pages(struct extent_buffer *eb, int mirror_num, + const struct btrfs_tree_parent_check *check) +{ + int ret; + ret = read_extent_buffer_pages_nowait(eb, mirror_num, check); + if (ret < 0) + return ret; + + wait_on_bit_io(&eb->bflags, EXTENT_BUFFER_READING, TASK_UNINTERRUPTIBLE); + if (!test_bit(EXTENT_BUFFER_UPTODATE, &eb->bflags)) + return -EIO; return 0; } @@ -4294,7 +4344,7 @@ void btrfs_readahead_tree_block(struct btrfs_fs_info *fs_info, return; } - ret = read_extent_buffer_pages(eb, WAIT_NONE, 0, &check); + ret = read_extent_buffer_pages_nowait(eb, 0, &check); if (ret < 0) free_extent_buffer_stale(eb); else diff --git a/fs/btrfs/extent_io.h b/fs/btrfs/extent_io.h index 8a36117ed453..6c5328bfabc2 100644 --- a/fs/btrfs/extent_io.h +++ b/fs/btrfs/extent_io.h @@ -248,7 +248,6 @@ int btree_write_cache_pages(struct address_space *mapping, struct writeback_control *wbc); void btrfs_readahead(struct readahead_control *rac); int set_folio_extent_mapped(struct folio *folio); -int set_page_extent_mapped(struct page *page); void clear_folio_extent_mapped(struct folio *folio); struct extent_buffer *alloc_extent_buffer(struct btrfs_fs_info *fs_info, @@ -262,12 +261,17 @@ struct extent_buffer *find_extent_buffer(struct btrfs_fs_info *fs_info, u64 start); void free_extent_buffer(struct extent_buffer *eb); void free_extent_buffer_stale(struct extent_buffer *eb); -#define WAIT_NONE 0 -#define WAIT_COMPLETE 1 -#define WAIT_PAGE_LOCK 2 -int read_extent_buffer_pages(struct extent_buffer *eb, int wait, int mirror_num, +int read_extent_buffer_pages(struct extent_buffer *eb, int mirror_num, const struct btrfs_tree_parent_check *parent_check); -void wait_on_extent_buffer_writeback(struct extent_buffer *eb); +int read_extent_buffer_pages_nowait(struct extent_buffer *eb, int mirror_num, + const struct btrfs_tree_parent_check *parent_check); + +static inline void wait_on_extent_buffer_writeback(struct extent_buffer *eb) +{ + wait_on_bit_io(&eb->bflags, EXTENT_BUFFER_WRITEBACK, + TASK_UNINTERRUPTIBLE); +} + void btrfs_readahead_tree_block(struct btrfs_fs_info *fs_info, u64 bytenr, u64 owner_root, u64 gen, int level); void btrfs_readahead_node_child(struct extent_buffer *node, int slot); diff --git a/fs/btrfs/file-item.c b/fs/btrfs/file-item.c index 886749b39672..d04a3b47b1fb 100644 --- a/fs/btrfs/file-item.c +++ b/fs/btrfs/file-item.c @@ -190,8 +190,6 @@ int btrfs_insert_hole_extent(struct btrfs_trans_handle *trans, btrfs_set_file_extent_compression(leaf, item, 0); btrfs_set_file_extent_encryption(leaf, item, 0); btrfs_set_file_extent_other_encoding(leaf, item, 0); - - btrfs_mark_buffer_dirty(trans, leaf); out: btrfs_free_path(path); return ret; @@ -1259,7 +1257,6 @@ found: ins_size /= csum_size; total_bytes += ins_size * fs_info->sectorsize; - btrfs_mark_buffer_dirty(trans, path->nodes[0]); if (total_bytes < sums->len) { btrfs_release_path(path); cond_resched(); diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index 14e27473c5bc..36f51c311bb1 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -36,52 +36,7 @@ #include "ioctl.h" #include "file.h" #include "super.h" - -/* - * Helper to fault in page and copy. This should go away and be replaced with - * calls into generic code. - */ -static noinline int btrfs_copy_from_user(loff_t pos, size_t write_bytes, - struct folio *folio, struct iov_iter *i) -{ - size_t copied = 0; - size_t total_copied = 0; - int offset = offset_in_page(pos); - - while (write_bytes > 0) { - size_t count = min_t(size_t, PAGE_SIZE - offset, write_bytes); - /* - * Copy data from userspace to the current page - */ - copied = copy_folio_from_iter_atomic(folio, offset, count, i); - - /* Flush processor's dcache for this page */ - flush_dcache_folio(folio); - - /* - * if we get a partial write, we can end up with - * partially up to date page. These add - * a lot of complexity, so make sure they don't - * happen by forcing this copy to be retried. - * - * The rest of the btrfs_file_write code will fall - * back to page at a time copies after we return 0. - */ - if (unlikely(copied < count)) { - if (!folio_test_uptodate(folio)) { - iov_iter_revert(i, copied); - copied = 0; - } - if (!copied) - break; - } - - write_bytes -= copied; - total_copied += copied; - offset += copied; - } - return total_copied; -} +#include "print-tree.h" /* * Unlock folio after btrfs_file_write() is done with it. @@ -106,7 +61,7 @@ static void btrfs_drop_folio(struct btrfs_fs_info *fs_info, struct folio *folio, } /* - * After btrfs_copy_from_user(), update the following things for delalloc: + * After copy_folio_from_iter_atomic(), update the following things for delalloc: * - Mark newly dirtied folio as DELALLOC in the io tree. * Used to advise which range is to be written back. * - Mark modified folio as Uptodate/Dirty and not needing COW fixup @@ -224,7 +179,7 @@ int btrfs_drop_extents(struct btrfs_trans_handle *trans, if (args->drop_cache) btrfs_drop_extent_map_range(inode, args->start, args->end - 1, false); - if (args->start >= inode->disk_i_size && !args->replace_extent) + if (data_race(args->start >= inode->disk_i_size) && !args->replace_extent) modify_tree = 0; update_refs = (btrfs_root_id(root) != BTRFS_TREE_LOG_OBJECTID); @@ -245,7 +200,11 @@ int btrfs_drop_extents(struct btrfs_trans_handle *trans, next_slot: leaf = path->nodes[0]; if (path->slots[0] >= btrfs_header_nritems(leaf)) { - BUG_ON(del_nr > 0); + if (WARN_ON(del_nr > 0)) { + btrfs_print_leaf(leaf); + ret = -EINVAL; + break; + } ret = btrfs_next_leaf(root, path); if (ret < 0) break; @@ -321,7 +280,11 @@ next_slot: * | -------- extent -------- | */ if (args->start > key.offset && args->end < extent_end) { - BUG_ON(del_nr > 0); + if (WARN_ON(del_nr > 0)) { + btrfs_print_leaf(leaf); + ret = -EINVAL; + break; + } if (extent_type == BTRFS_FILE_EXTENT_INLINE) { ret = -EOPNOTSUPP; break; @@ -351,7 +314,6 @@ next_slot: btrfs_set_file_extent_offset(leaf, fi, extent_offset); btrfs_set_file_extent_num_bytes(leaf, fi, extent_end - args->start); - btrfs_mark_buffer_dirty(trans, leaf); if (update_refs && disk_bytenr > 0) { struct btrfs_ref ref = { @@ -397,7 +359,6 @@ next_slot: btrfs_set_file_extent_offset(leaf, fi, extent_offset); btrfs_set_file_extent_num_bytes(leaf, fi, extent_end - args->end); - btrfs_mark_buffer_dirty(trans, leaf); if (update_refs && disk_bytenr > 0) args->bytes_found += args->end - key.offset; break; @@ -409,7 +370,11 @@ next_slot: * | -------- extent -------- | */ if (args->start > key.offset && args->end >= extent_end) { - BUG_ON(del_nr > 0); + if (WARN_ON(del_nr > 0)) { + btrfs_print_leaf(leaf); + ret = -EINVAL; + break; + } if (extent_type == BTRFS_FILE_EXTENT_INLINE) { ret = -EOPNOTSUPP; break; @@ -417,7 +382,6 @@ next_slot: btrfs_set_file_extent_num_bytes(leaf, fi, args->start - key.offset); - btrfs_mark_buffer_dirty(trans, leaf); if (update_refs && disk_bytenr > 0) args->bytes_found += extent_end - args->start; if (args->end == extent_end) @@ -437,7 +401,11 @@ delete_extent_item: del_slot = path->slots[0]; del_nr = 1; } else { - BUG_ON(del_slot + del_nr != path->slots[0]); + if (WARN_ON(del_slot + del_nr != path->slots[0])) { + btrfs_print_leaf(leaf); + ret = -EINVAL; + break; + } del_nr++; } @@ -668,7 +636,6 @@ again: trans->transid); btrfs_set_file_extent_num_bytes(leaf, fi, end - other_start); - btrfs_mark_buffer_dirty(trans, leaf); goto out; } } @@ -697,7 +664,6 @@ again: other_end - start); btrfs_set_file_extent_offset(leaf, fi, start - orig_offset); - btrfs_mark_buffer_dirty(trans, leaf); goto out; } } @@ -731,7 +697,6 @@ again: btrfs_set_file_extent_offset(leaf, fi, split - orig_offset); btrfs_set_file_extent_num_bytes(leaf, fi, extent_end - split); - btrfs_mark_buffer_dirty(trans, leaf); ref.action = BTRFS_ADD_DELAYED_REF; ref.bytenr = bytenr; @@ -810,7 +775,6 @@ again: btrfs_set_file_extent_type(leaf, fi, BTRFS_FILE_EXTENT_REG); btrfs_set_file_extent_generation(leaf, fi, trans->transid); - btrfs_mark_buffer_dirty(trans, leaf); } else { fi = btrfs_item_ptr(leaf, del_slot - 1, struct btrfs_file_extent_item); @@ -819,7 +783,6 @@ again: btrfs_set_file_extent_generation(leaf, fi, trans->transid); btrfs_set_file_extent_num_bytes(leaf, fi, extent_end - key.offset); - btrfs_mark_buffer_dirty(trans, leaf); ret = btrfs_del_items(trans, root, path, del_slot, del_nr); if (ret < 0) { @@ -1052,7 +1015,7 @@ int btrfs_check_nocow_lock(struct btrfs_inode *inode, loff_t pos, &cached_state); } ret = can_nocow_extent(&inode->vfs_inode, lockstart, &num_bytes, - NULL, nowait, false); + NULL, nowait); if (ret <= 0) btrfs_drew_write_unlock(&root->snapshot_lock); else @@ -1252,7 +1215,23 @@ again: break; } - copied = btrfs_copy_from_user(pos, write_bytes, folio, i); + copied = copy_folio_from_iter_atomic(folio, + offset_in_folio(folio, pos), write_bytes, i); + flush_dcache_folio(folio); + + /* + * If we get a partial write, we can end up with partially + * uptodate page. Although if sector size < page size we can + * handle it, but if it's not sector aligned it can cause + * a lot of complexity, so make sure they don't happen by + * forcing retry this copy. + */ + if (unlikely(copied < write_bytes)) { + if (!folio_test_uptodate(folio)) { + iov_iter_revert(i, copied); + copied = 0; + } + } num_sectors = BTRFS_BYTES_TO_BLKS(fs_info, reserve_bytes); dirty_sectors = round_up(copied + sector_offset, @@ -2029,7 +2008,6 @@ static int fill_holes(struct btrfs_trans_handle *trans, btrfs_set_file_extent_ram_bytes(leaf, fi, num_bytes); btrfs_set_file_extent_offset(leaf, fi, 0); btrfs_set_file_extent_generation(leaf, fi, trans->transid); - btrfs_mark_buffer_dirty(trans, leaf); goto out; } @@ -2046,7 +2024,6 @@ static int fill_holes(struct btrfs_trans_handle *trans, btrfs_set_file_extent_ram_bytes(leaf, fi, num_bytes); btrfs_set_file_extent_offset(leaf, fi, 0); btrfs_set_file_extent_generation(leaf, fi, trans->transid); - btrfs_mark_buffer_dirty(trans, leaf); goto out; } btrfs_release_path(path); @@ -2194,7 +2171,6 @@ static int btrfs_insert_replace_extent(struct btrfs_trans_handle *trans, btrfs_set_file_extent_num_bytes(leaf, extent, replace_len); if (extent_info->is_new_extent) btrfs_set_file_extent_generation(leaf, extent, trans->transid); - btrfs_mark_buffer_dirty(trans, leaf); btrfs_release_path(path); ret = btrfs_inode_set_file_extent_range(inode, extent_info->file_offset, diff --git a/fs/btrfs/free-space-cache.c b/fs/btrfs/free-space-cache.c index cfa52ef40b06..d42b6f882f57 100644 --- a/fs/btrfs/free-space-cache.c +++ b/fs/btrfs/free-space-cache.c @@ -12,7 +12,7 @@ #include <linux/error-injection.h> #include <linux/sched/mm.h> #include <linux/string_choices.h> -#include "ctree.h" +#include "extent-tree.h" #include "fs.h" #include "messages.h" #include "misc.h" @@ -198,7 +198,6 @@ static int __create_free_space_inode(struct btrfs_root *root, btrfs_set_inode_nlink(leaf, inode_item, 1); btrfs_set_inode_transid(leaf, inode_item, trans->transid); btrfs_set_inode_block_group(leaf, inode_item, offset); - btrfs_mark_buffer_dirty(trans, leaf); btrfs_release_path(path); key.objectid = BTRFS_FREE_SPACE_OBJECTID; @@ -216,7 +215,6 @@ static int __create_free_space_inode(struct btrfs_root *root, struct btrfs_free_space_header); memzero_extent_buffer(leaf, (unsigned long)header, sizeof(*header)); btrfs_set_free_space_key(leaf, header, &disk_key); - btrfs_mark_buffer_dirty(trans, leaf); btrfs_release_path(path); return 0; @@ -463,7 +461,7 @@ static int io_ctl_prepare_pages(struct btrfs_io_ctl *io_ctl, bool uptodate) return -ENOMEM; } - ret = set_page_extent_mapped(page); + ret = set_folio_extent_mapped(page_folio(page)); if (ret < 0) { unlock_page(page); put_page(page); @@ -1189,7 +1187,6 @@ update_cache_item(struct btrfs_trans_handle *trans, btrfs_set_free_space_entries(leaf, header, entries); btrfs_set_free_space_bitmaps(leaf, header, bitmaps); btrfs_set_free_space_generation(leaf, header, trans->transid); - btrfs_mark_buffer_dirty(trans, leaf); btrfs_release_path(path); return 0; diff --git a/fs/btrfs/free-space-tree.c b/fs/btrfs/free-space-tree.c index 7ba50e133921..cae540ec15ed 100644 --- a/fs/btrfs/free-space-tree.c +++ b/fs/btrfs/free-space-tree.c @@ -89,7 +89,6 @@ static int add_new_free_space_info(struct btrfs_trans_handle *trans, struct btrfs_free_space_info); btrfs_set_free_space_extent_count(leaf, info, 0); btrfs_set_free_space_flags(leaf, info, 0); - btrfs_mark_buffer_dirty(trans, leaf); ret = 0; out: @@ -287,7 +286,6 @@ int convert_free_space_to_bitmaps(struct btrfs_trans_handle *trans, flags |= BTRFS_FREE_SPACE_USING_BITMAPS; btrfs_set_free_space_flags(leaf, info, flags); expected_extent_count = btrfs_free_space_extent_count(leaf, info); - btrfs_mark_buffer_dirty(trans, leaf); btrfs_release_path(path); if (extent_count != expected_extent_count) { @@ -324,7 +322,6 @@ int convert_free_space_to_bitmaps(struct btrfs_trans_handle *trans, ptr = btrfs_item_ptr_offset(leaf, path->slots[0]); write_extent_buffer(leaf, bitmap_cursor, ptr, data_size); - btrfs_mark_buffer_dirty(trans, leaf); btrfs_release_path(path); i += extent_size; @@ -430,7 +427,6 @@ int convert_free_space_to_extents(struct btrfs_trans_handle *trans, flags &= ~BTRFS_FREE_SPACE_USING_BITMAPS; btrfs_set_free_space_flags(leaf, info, flags); expected_extent_count = btrfs_free_space_extent_count(leaf, info); - btrfs_mark_buffer_dirty(trans, leaf); btrfs_release_path(path); nrbits = block_group->length >> block_group->fs_info->sectorsize_bits; @@ -495,7 +491,6 @@ static int update_free_space_extent_count(struct btrfs_trans_handle *trans, extent_count += new_extents; btrfs_set_free_space_extent_count(path->nodes[0], info, extent_count); - btrfs_mark_buffer_dirty(trans, path->nodes[0]); btrfs_release_path(path); if (!(flags & BTRFS_FREE_SPACE_USING_BITMAPS) && @@ -1350,6 +1345,12 @@ int btrfs_rebuild_free_space_tree(struct btrfs_fs_info *fs_info) btrfs_end_transaction(trans); return ret; } + if (btrfs_should_end_transaction(trans)) { + btrfs_end_transaction(trans); + trans = btrfs_start_transaction(free_space_root, 1); + if (IS_ERR(trans)) + return PTR_ERR(trans); + } node = rb_next(node); } diff --git a/fs/btrfs/fs.c b/fs/btrfs/fs.c index 31c1648bc0b4..09cfb43580cb 100644 --- a/fs/btrfs/fs.c +++ b/fs/btrfs/fs.c @@ -4,6 +4,136 @@ #include "ctree.h" #include "fs.h" #include "accessors.h" +#include "volumes.h" + +static const struct btrfs_csums { + u16 size; + const char name[10]; + const char driver[12]; +} btrfs_csums[] = { + [BTRFS_CSUM_TYPE_CRC32] = { .size = 4, .name = "crc32c" }, + [BTRFS_CSUM_TYPE_XXHASH] = { .size = 8, .name = "xxhash64" }, + [BTRFS_CSUM_TYPE_SHA256] = { .size = 32, .name = "sha256" }, + [BTRFS_CSUM_TYPE_BLAKE2] = { .size = 32, .name = "blake2b", + .driver = "blake2b-256" }, +}; + +/* This exists for btrfs-progs usages. */ +u16 btrfs_csum_type_size(u16 type) +{ + return btrfs_csums[type].size; +} + +int btrfs_super_csum_size(const struct btrfs_super_block *s) +{ + u16 t = btrfs_super_csum_type(s); + + /* csum type is validated at mount time. */ + return btrfs_csum_type_size(t); +} + +const char *btrfs_super_csum_name(u16 csum_type) +{ + /* csum type is validated at mount time. */ + return btrfs_csums[csum_type].name; +} + +/* + * Return driver name if defined, otherwise the name that's also a valid driver + * name. + */ +const char *btrfs_super_csum_driver(u16 csum_type) +{ + /* csum type is validated at mount time */ + return btrfs_csums[csum_type].driver[0] ? + btrfs_csums[csum_type].driver : + btrfs_csums[csum_type].name; +} + +size_t __attribute_const__ btrfs_get_num_csums(void) +{ + return ARRAY_SIZE(btrfs_csums); +} + +/* + * Start exclusive operation @type, return true on success. + */ +bool btrfs_exclop_start(struct btrfs_fs_info *fs_info, + enum btrfs_exclusive_operation type) +{ + bool ret = false; + + spin_lock(&fs_info->super_lock); + if (fs_info->exclusive_operation == BTRFS_EXCLOP_NONE) { + fs_info->exclusive_operation = type; + ret = true; + } + spin_unlock(&fs_info->super_lock); + + return ret; +} + +/* + * Conditionally allow to enter the exclusive operation in case it's compatible + * with the running one. This must be paired with btrfs_exclop_start_unlock() + * and btrfs_exclop_finish(). + * + * Compatibility: + * - the same type is already running + * - when trying to add a device and balance has been paused + * - not BTRFS_EXCLOP_NONE - this is intentionally incompatible and the caller + * must check the condition first that would allow none -> @type + */ +bool btrfs_exclop_start_try_lock(struct btrfs_fs_info *fs_info, + enum btrfs_exclusive_operation type) +{ + spin_lock(&fs_info->super_lock); + if (fs_info->exclusive_operation == type || + (fs_info->exclusive_operation == BTRFS_EXCLOP_BALANCE_PAUSED && + type == BTRFS_EXCLOP_DEV_ADD)) + return true; + + spin_unlock(&fs_info->super_lock); + return false; +} + +void btrfs_exclop_start_unlock(struct btrfs_fs_info *fs_info) +{ + spin_unlock(&fs_info->super_lock); +} + +void btrfs_exclop_finish(struct btrfs_fs_info *fs_info) +{ + spin_lock(&fs_info->super_lock); + WRITE_ONCE(fs_info->exclusive_operation, BTRFS_EXCLOP_NONE); + spin_unlock(&fs_info->super_lock); + sysfs_notify(&fs_info->fs_devices->fsid_kobj, NULL, "exclusive_operation"); +} + +void btrfs_exclop_balance(struct btrfs_fs_info *fs_info, + enum btrfs_exclusive_operation op) +{ + switch (op) { + case BTRFS_EXCLOP_BALANCE_PAUSED: + spin_lock(&fs_info->super_lock); + ASSERT(fs_info->exclusive_operation == BTRFS_EXCLOP_BALANCE || + fs_info->exclusive_operation == BTRFS_EXCLOP_DEV_ADD || + fs_info->exclusive_operation == BTRFS_EXCLOP_NONE || + fs_info->exclusive_operation == BTRFS_EXCLOP_BALANCE_PAUSED); + fs_info->exclusive_operation = BTRFS_EXCLOP_BALANCE_PAUSED; + spin_unlock(&fs_info->super_lock); + break; + case BTRFS_EXCLOP_BALANCE: + spin_lock(&fs_info->super_lock); + ASSERT(fs_info->exclusive_operation == BTRFS_EXCLOP_BALANCE_PAUSED); + fs_info->exclusive_operation = BTRFS_EXCLOP_BALANCE; + spin_unlock(&fs_info->super_lock); + break; + default: + btrfs_warn(fs_info, + "invalid exclop balance operation %d requested", op); + } +} void __btrfs_set_fs_incompat(struct btrfs_fs_info *fs_info, u64 flag, const char *name) diff --git a/fs/btrfs/fs.h b/fs/btrfs/fs.h index 79a1a3d6f04d..b572d6b9730b 100644 --- a/fs/btrfs/fs.h +++ b/fs/btrfs/fs.h @@ -14,10 +14,10 @@ #include <linux/lockdep.h> #include <linux/spinlock.h> #include <linux/mutex.h> -#include <linux/rwlock_types.h> #include <linux/rwsem.h> #include <linux/semaphore.h> #include <linux/list.h> +#include <linux/pagemap.h> #include <linux/radix-tree.h> #include <linux/workqueue.h> #include <linux/wait.h> @@ -627,6 +627,9 @@ struct btrfs_fs_info { struct kobject *qgroups_kobj; struct kobject *discard_kobj; + /* Track the number of blocks (sectors) read by the filesystem. */ + struct percpu_counter stats_read_blocks; + /* Used to keep from writing metadata until there is a nice batch */ struct percpu_counter dirty_metadata_bytes; struct percpu_counter delalloc_bytes; @@ -887,6 +890,11 @@ struct btrfs_fs_info { #define inode_to_fs_info(_inode) (BTRFS_I(_Generic((_inode), \ struct inode *: (_inode)))->root->fs_info) +static inline gfp_t btrfs_alloc_write_mask(struct address_space *mapping) +{ + return mapping_gfp_constraint(mapping, ~__GFP_FS); +} + static inline u64 btrfs_get_fs_generation(const struct btrfs_fs_info *fs_info) { return READ_ONCE(fs_info->generation); @@ -953,6 +961,8 @@ static inline u64 btrfs_calc_metadata_size(const struct btrfs_fs_info *fs_info, #define BTRFS_MAX_EXTENT_ITEM_SIZE(r) ((BTRFS_LEAF_DATA_SIZE(r->fs_info) >> 4) - \ sizeof(struct btrfs_item)) +#define BTRFS_BYTES_TO_BLKS(fs_info, bytes) ((bytes) >> (fs_info)->sectorsize_bits) + static inline bool btrfs_is_zoned(const struct btrfs_fs_info *fs_info) { return IS_ENABLED(CONFIG_BLK_DEV_ZONED) && fs_info->zone_size > 0; @@ -982,6 +992,17 @@ void btrfs_exclop_balance(struct btrfs_fs_info *fs_info, int btrfs_check_ioctl_vol_args_path(const struct btrfs_ioctl_vol_args *vol_args); +u16 btrfs_csum_type_size(u16 type); +int btrfs_super_csum_size(const struct btrfs_super_block *s); +const char *btrfs_super_csum_name(u16 csum_type); +const char *btrfs_super_csum_driver(u16 csum_type); +size_t __attribute_const__ btrfs_get_num_csums(void); + +static inline bool btrfs_is_empty_uuid(const u8 *uuid) +{ + return uuid_is_null((const uuid_t *)uuid); +} + /* Compatibility and incompatibility defines */ void __btrfs_set_fs_incompat(struct btrfs_fs_info *fs_info, u64 flag, const char *name); @@ -1058,6 +1079,14 @@ static inline void btrfs_wake_unfinished_drop(struct btrfs_fs_info *fs_info) (unlikely(test_bit(BTRFS_FS_STATE_LOG_CLEANUP_ERROR, \ &(fs_info)->fs_state))) +/* + * We use folio flag owner_2 to indicate there is an ordered extent with + * unfinished IO. + */ +#define folio_test_ordered(folio) folio_test_owner_2(folio) +#define folio_set_ordered(folio) folio_set_owner_2(folio) +#define folio_clear_ordered(folio) folio_clear_owner_2(folio) + #ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS #define EXPORT_FOR_TESTS diff --git a/fs/btrfs/inode-item.c b/fs/btrfs/inode-item.c index 29572dfaf878..448aa1a682d6 100644 --- a/fs/btrfs/inode-item.c +++ b/fs/btrfs/inode-item.c @@ -298,8 +298,6 @@ static int btrfs_insert_inode_extref(struct btrfs_trans_handle *trans, ptr = (unsigned long)&extref->name; write_extent_buffer(path->nodes[0], name->name, ptr, name->len); - btrfs_mark_buffer_dirty(trans, path->nodes[0]); - out: btrfs_free_path(path); return ret; @@ -363,8 +361,6 @@ int btrfs_insert_inode_ref(struct btrfs_trans_handle *trans, ptr = (unsigned long)(ref + 1); } write_extent_buffer(path->nodes[0], name->name, ptr, name->len); - btrfs_mark_buffer_dirty(trans, path->nodes[0]); - out: btrfs_free_path(path); @@ -590,7 +586,6 @@ search_again: num_dec = (orig_num_bytes - extent_num_bytes); if (extent_start != 0) control->sub_bytes += num_dec; - btrfs_mark_buffer_dirty(trans, leaf); } else { extent_num_bytes = btrfs_file_extent_disk_num_bytes(leaf, fi); diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 488edca8333a..fe2c810335ff 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -393,34 +393,13 @@ void btrfs_inode_unlock(struct btrfs_inode *inode, unsigned int ilock_flags) * extent (btrfs_finish_ordered_io()). */ static inline void btrfs_cleanup_ordered_extents(struct btrfs_inode *inode, - struct folio *locked_folio, u64 offset, u64 bytes) { unsigned long index = offset >> PAGE_SHIFT; unsigned long end_index = (offset + bytes - 1) >> PAGE_SHIFT; - u64 page_start = 0, page_end = 0; struct folio *folio; - if (locked_folio) { - page_start = folio_pos(locked_folio); - page_end = page_start + folio_size(locked_folio) - 1; - } - while (index <= end_index) { - /* - * For locked page, we will call btrfs_mark_ordered_io_finished - * through btrfs_mark_ordered_io_finished() on it - * in run_delalloc_range() for the error handling, which will - * clear page Ordered and run the ordered extent accounting. - * - * Here we can't just clear the Ordered bit, or - * btrfs_mark_ordered_io_finished() would skip the accounting - * for the page range, and the ordered extent will never finish. - */ - if (locked_folio && index == (page_start >> PAGE_SHIFT)) { - index++; - continue; - } folio = filemap_get_folio(inode->vfs_inode.i_mapping, index); index++; if (IS_ERR(folio)) @@ -436,23 +415,6 @@ static inline void btrfs_cleanup_ordered_extents(struct btrfs_inode *inode, folio_put(folio); } - if (locked_folio) { - /* The locked page covers the full range, nothing needs to be done */ - if (bytes + offset <= page_start + folio_size(locked_folio)) - return; - /* - * In case this page belongs to the delalloc range being - * instantiated then skip it, since the first page of a range is - * going to be properly cleaned up by the caller of - * run_delalloc_range - */ - if (page_start >= offset && page_end <= (offset + bytes - 1)) { - bytes = offset + bytes - folio_pos(locked_folio) - - folio_size(locked_folio); - offset = folio_pos(locked_folio) + folio_size(locked_folio); - } - } - return btrfs_mark_ordered_io_finished(inode, NULL, offset, bytes, false); } @@ -564,7 +526,6 @@ static int insert_inline_extent(struct btrfs_trans_handle *trans, kunmap_local(kaddr); folio_put(folio); } - btrfs_mark_buffer_dirty(trans, leaf); btrfs_release_path(path); /* @@ -1129,19 +1090,14 @@ static void submit_uncompressed_range(struct btrfs_inode *inode, &wbc, false); wbc_detach_inode(&wbc); if (ret < 0) { - btrfs_cleanup_ordered_extents(inode, locked_folio, - start, end - start + 1); - if (locked_folio) { - const u64 page_start = folio_pos(locked_folio); - - folio_start_writeback(locked_folio); - folio_end_writeback(locked_folio); - btrfs_mark_ordered_io_finished(inode, locked_folio, - page_start, PAGE_SIZE, - !ret); - mapping_set_error(locked_folio->mapping, ret); - folio_unlock(locked_folio); - } + btrfs_cleanup_ordered_extents(inode, start, end - start + 1); + if (locked_folio) + btrfs_folio_end_lock(inode->root->fs_info, locked_folio, + start, async_extent->ram_size); + btrfs_err_rl(inode->root->fs_info, + "%s failed, root=%llu inode=%llu start=%llu len=%llu: %d", + __func__, btrfs_root_id(inode->root), + btrfs_ino(inode), start, async_extent->ram_size, ret); } } @@ -1373,6 +1329,17 @@ static noinline int cow_file_range(struct btrfs_inode *inode, alloc_hint = btrfs_get_extent_allocation_hint(inode, start, num_bytes); /* + * We're not doing compressed IO, don't unlock the first page (which + * the caller expects to stay locked), don't clear any dirty bits and + * don't set any writeback bits. + * + * Do set the Ordered (Private2) bit so we know this page was properly + * setup for writepage. + */ + page_ops = (keep_locked ? 0 : PAGE_UNLOCK); + page_ops |= PAGE_SET_ORDERED; + + /* * Relocation relies on the relocated extents to have exactly the same * size as the original extents. Normally writeback for relocation data * extents follows a NOCOW path because relocation preallocates the @@ -1431,6 +1398,10 @@ static noinline int cow_file_range(struct btrfs_inode *inode, file_extent.offset = 0; file_extent.compression = BTRFS_COMPRESS_NONE; + /* + * Locked range will be released either during error clean up or + * after the whole range is finished. + */ lock_extent(&inode->io_tree, start, start + cur_alloc_size - 1, &cached); @@ -1476,21 +1447,6 @@ static noinline int cow_file_range(struct btrfs_inode *inode, btrfs_dec_block_group_reservations(fs_info, ins.objectid); - /* - * We're not doing compressed IO, don't unlock the first page - * (which the caller expects to stay locked), don't clear any - * dirty bits and don't set any writeback bits - * - * Do set the Ordered flag so we know this page was - * properly setup for writepage. - */ - page_ops = (keep_locked ? 0 : PAGE_UNLOCK); - page_ops |= PAGE_SET_ORDERED; - - extent_clear_unlock_delalloc(inode, start, start + cur_alloc_size - 1, - locked_folio, &cached, - EXTENT_LOCKED | EXTENT_DELALLOC, - page_ops); if (num_bytes < cur_alloc_size) num_bytes = 0; else @@ -1507,6 +1463,8 @@ static noinline int cow_file_range(struct btrfs_inode *inode, if (ret) goto out_unlock; } + extent_clear_unlock_delalloc(inode, orig_start, end, locked_folio, &cached, + EXTENT_LOCKED | EXTENT_DELALLOC, page_ops); done: if (done_offset) *done_offset = end; @@ -1527,35 +1485,30 @@ out_unlock: * We process each region below. */ - clear_bits = EXTENT_LOCKED | EXTENT_DELALLOC | EXTENT_DELALLOC_NEW | - EXTENT_DEFRAG | EXTENT_CLEAR_META_RESV; - page_ops = PAGE_UNLOCK | PAGE_START_WRITEBACK | PAGE_END_WRITEBACK; - /* * For the range (1). We have already instantiated the ordered extents * for this region. They are cleaned up by * btrfs_cleanup_ordered_extents() in e.g, - * btrfs_run_delalloc_range(). EXTENT_LOCKED | EXTENT_DELALLOC are - * already cleared in the above loop. And, EXTENT_DELALLOC_NEW | - * EXTENT_DEFRAG | EXTENT_CLEAR_META_RESV are handled by the cleanup - * function. + * btrfs_run_delalloc_range(). + * EXTENT_DELALLOC_NEW | EXTENT_DEFRAG | EXTENT_CLEAR_META_RESV + * are also handled by the cleanup function. * - * However, in case of @keep_locked, we still need to unlock the pages - * (except @locked_folio) to ensure all the pages are unlocked. + * So here we only clear EXTENT_LOCKED and EXTENT_DELALLOC flag, and + * finish the writeback of the involved folios, which will be never submitted. */ - if (keep_locked && orig_start < start) { + if (orig_start < start) { + clear_bits = EXTENT_LOCKED | EXTENT_DELALLOC; + page_ops = PAGE_UNLOCK | PAGE_START_WRITEBACK | PAGE_END_WRITEBACK; + if (!locked_folio) mapping_set_error(inode->vfs_inode.i_mapping, ret); extent_clear_unlock_delalloc(inode, orig_start, start - 1, - locked_folio, NULL, 0, page_ops); + locked_folio, NULL, clear_bits, page_ops); } - /* - * At this point we're unlocked, we want to make sure we're only - * clearing these flags under the extent lock, so lock the rest of the - * range and clear everything up. - */ - lock_extent(&inode->io_tree, start, end, NULL); + clear_bits = EXTENT_LOCKED | EXTENT_DELALLOC | EXTENT_DELALLOC_NEW | + EXTENT_DEFRAG | EXTENT_CLEAR_META_RESV; + page_ops = PAGE_UNLOCK | PAGE_START_WRITEBACK | PAGE_END_WRITEBACK; /* * For the range (2). If we reserved an extent for our delalloc range @@ -1589,6 +1542,10 @@ out_unlock: btrfs_qgroup_free_data(inode, NULL, start + cur_alloc_size, end - start - cur_alloc_size + 1, NULL); } + btrfs_err_rl(fs_info, + "%s failed, root=%llu inode=%llu start=%llu len=%llu: %d", + __func__, btrfs_root_id(inode->root), + btrfs_ino(inode), orig_start, end + 1 - orig_start, ret); return ret; } @@ -1809,7 +1766,7 @@ static int fallback_to_cow(struct btrfs_inode *inode, bytes = range_bytes; spin_lock(&sinfo->lock); - btrfs_space_info_update_bytes_may_use(fs_info, sinfo, bytes); + btrfs_space_info_update_bytes_may_use(sinfo, bytes); spin_unlock(&sinfo->lock); if (count > 0) @@ -1837,7 +1794,6 @@ struct can_nocow_file_extent_args { /* End file offset (inclusive) of the range we want to NOCOW. */ u64 end; bool writeback_path; - bool strict; /* * Free the path passed to can_nocow_file_extent() once it's not needed * anymore. @@ -1892,8 +1848,7 @@ static int can_nocow_file_extent(struct btrfs_path *path, * for its subvolume was created, then this implies the extent is shared, * hence we must COW. */ - if (!args->strict && - btrfs_file_extent_generation(leaf, fi) <= + if (btrfs_file_extent_generation(leaf, fi) <= btrfs_root_last_snapshot(&root->root_item)) goto out; @@ -1922,9 +1877,8 @@ static int can_nocow_file_extent(struct btrfs_path *path, */ btrfs_release_path(path); - ret = btrfs_cross_ref_exist(root, btrfs_ino(inode), - key->offset - args->file_extent.offset, - args->file_extent.disk_bytenr, args->strict, path); + ret = btrfs_cross_ref_exist(inode, key->offset - args->file_extent.offset, + args->file_extent.disk_bytenr, path); WARN_ON_ONCE(ret > 0 && is_freespace_inode); if (ret != 0) goto out; @@ -1971,6 +1925,53 @@ static int can_nocow_file_extent(struct btrfs_path *path, } /* + * Cleanup the dirty folios which will never be submitted due to error. + * + * When running a delalloc range, we may need to split the ranges (due to + * fragmentation or NOCOW). If we hit an error in the later part, we will error + * out and previously successfully executed range will never be submitted, thus + * we have to cleanup those folios by clearing their dirty flag, starting and + * finishing the writeback. + */ +static void cleanup_dirty_folios(struct btrfs_inode *inode, + struct folio *locked_folio, + u64 start, u64 end, int error) +{ + struct btrfs_fs_info *fs_info = inode->root->fs_info; + struct address_space *mapping = inode->vfs_inode.i_mapping; + pgoff_t start_index = start >> PAGE_SHIFT; + pgoff_t end_index = end >> PAGE_SHIFT; + u32 len; + + ASSERT(end + 1 - start < U32_MAX); + ASSERT(IS_ALIGNED(start, fs_info->sectorsize) && + IS_ALIGNED(end + 1, fs_info->sectorsize)); + len = end + 1 - start; + + /* + * Handle the locked folio first. + * The btrfs_folio_clamp_*() helpers can handle range out of the folio case. + */ + btrfs_folio_clamp_finish_io(fs_info, locked_folio, start, len); + + for (pgoff_t index = start_index; index <= end_index; index++) { + struct folio *folio; + + /* Already handled at the beginning. */ + if (index == locked_folio->index) + continue; + folio = __filemap_get_folio(mapping, index, FGP_LOCK, GFP_NOFS); + /* Cache already dropped, no need to do any cleanup. */ + if (IS_ERR(folio)) + continue; + btrfs_folio_clamp_finish_io(fs_info, locked_folio, start, len); + folio_unlock(folio); + folio_put(folio); + } + mapping_set_error(mapping, error); +} + +/* * when nowcow writeback call back. This checks for snapshots or COW copies * of the extents that exist in the file, and COWs the file as required. * @@ -1985,6 +1986,11 @@ static noinline int run_delalloc_nocow(struct btrfs_inode *inode, struct btrfs_root *root = inode->root; struct btrfs_path *path; u64 cow_start = (u64)-1; + /* + * If not 0, represents the inclusive end of the last fallback_to_cow() + * range. Only for error handling. + */ + u64 cow_end = 0; u64 cur_offset = start; int ret; bool check_prev = true; @@ -2145,6 +2151,7 @@ must_cow: found_key.offset - 1); cow_start = (u64)-1; if (ret) { + cow_end = found_key.offset - 1; btrfs_dec_nocow_writers(nocow_bg); goto error; } @@ -2218,11 +2225,12 @@ must_cow: cow_start = cur_offset; if (cow_start != (u64)-1) { - cur_offset = end; ret = fallback_to_cow(inode, locked_folio, cow_start, end); cow_start = (u64)-1; - if (ret) + if (ret) { + cow_end = end; goto error; + } } btrfs_free_path(path); @@ -2230,12 +2238,41 @@ must_cow: error: /* + * There are several error cases: + * + * 1) Failed without falling back to COW + * start cur_offset end + * |/////////////| | + * + * For range [start, cur_offset) the folios are already unlocked (except + * @locked_folio), EXTENT_DELALLOC already removed. + * Only need to clear the dirty flag as they will never be submitted. + * Ordered extent and extent maps are handled by + * btrfs_mark_ordered_io_finished() inside run_delalloc_range(). + * + * 2) Failed with error from fallback_to_cow() + * start cur_offset cow_end end + * |/////////////|-----------| | + * + * For range [start, cur_offset) it's the same as case 1). + * But for range [cur_offset, cow_end), the folios have dirty flag + * cleared and unlocked, EXTENT_DEALLLOC cleared by cow_file_range(). + * + * Thus we should not call extent_clear_unlock_delalloc() on range + * [cur_offset, cow_end), as the folios are already unlocked. + * + * So clear the folio dirty flags for [start, cur_offset) first. + */ + if (cur_offset > start) + cleanup_dirty_folios(inode, locked_folio, start, cur_offset - 1, ret); + + /* * If an error happened while a COW region is outstanding, cur_offset - * needs to be reset to cow_start to ensure the COW region is unlocked - * as well. + * needs to be reset to @cow_end + 1 to skip the COW range, as + * cow_file_range() will do the proper cleanup at error. */ - if (cow_start != (u64)-1) - cur_offset = cow_start; + if (cow_end) + cur_offset = cow_end + 1; /* * We need to lock the extent here because we're clearing DELALLOC and @@ -2255,6 +2292,10 @@ error: btrfs_qgroup_free_data(inode, NULL, cur_offset, end - cur_offset + 1, NULL); } btrfs_free_path(path); + btrfs_err_rl(fs_info, + "%s failed, root=%llu inode=%llu start=%llu len=%llu: %d", + __func__, btrfs_root_id(inode->root), + btrfs_ino(inode), start, end + 1 - start, ret); return ret; } @@ -2305,8 +2346,7 @@ int btrfs_run_delalloc_range(struct btrfs_inode *inode, struct folio *locked_fol out: if (ret < 0) - btrfs_cleanup_ordered_extents(inode, locked_folio, start, - end - start + 1); + btrfs_cleanup_ordered_extents(inode, start, end - start + 1); return ret; } @@ -2921,7 +2961,6 @@ static int insert_reserved_file_extent(struct btrfs_trans_handle *trans, btrfs_item_ptr_offset(leaf, path->slots[0]), sizeof(struct btrfs_file_extent_item)); - btrfs_mark_buffer_dirty(trans, leaf); btrfs_release_path(path); /* @@ -4085,7 +4124,6 @@ static noinline int btrfs_update_inode_item(struct btrfs_trans_handle *trans, struct btrfs_inode_item); fill_inode_item(trans, leaf, inode_item, &inode->vfs_inode); - btrfs_mark_buffer_dirty(trans, leaf); btrfs_set_inode_last_trans(trans, inode); ret = 0; failed: @@ -6380,7 +6418,6 @@ int btrfs_create_new_inode(struct btrfs_trans_handle *trans, } } - btrfs_mark_buffer_dirty(trans, path->nodes[0]); /* * We don't need the path anymore, plus inheriting properties, adding * ACLs, security xattrs, orphan item or adding the link, will result in @@ -7011,8 +7048,6 @@ static bool btrfs_extent_readonly(struct btrfs_fs_info *fs_info, u64 bytenr) * @orig_start: (optional) Return the original file offset of the file extent * @orig_len: (optional) Return the original on-disk length of the file extent * @ram_bytes: (optional) Return the ram_bytes of the file extent - * @strict: if true, omit optimizations that might force us into unnecessary - * cow. e.g., don't trust generation number. * * Return: * >0 and update @len if we can do nocow write @@ -7024,7 +7059,7 @@ static bool btrfs_extent_readonly(struct btrfs_fs_info *fs_info, u64 bytenr) */ noinline int can_nocow_extent(struct inode *inode, u64 offset, u64 *len, struct btrfs_file_extent *file_extent, - bool nowait, bool strict) + bool nowait) { struct btrfs_fs_info *fs_info = inode_to_fs_info(inode); struct can_nocow_file_extent_args nocow_args = { 0 }; @@ -7077,7 +7112,6 @@ noinline int can_nocow_extent(struct inode *inode, u64 offset, u64 *len, nocow_args.start = offset; nocow_args.end = offset + *len - 1; - nocow_args.strict = strict; nocow_args.free_path = true; ret = can_nocow_file_extent(path, &key, BTRFS_I(inode), &nocow_args); @@ -8027,31 +8061,45 @@ static int btrfs_rename_exchange(struct inode *old_dir, /* src is a subvolume */ if (old_ino == BTRFS_FIRST_FREE_OBJECTID) { ret = btrfs_unlink_subvol(trans, BTRFS_I(old_dir), old_dentry); + if (ret) { + btrfs_abort_transaction(trans, ret); + goto out_fail; + } } else { /* src is an inode */ ret = __btrfs_unlink_inode(trans, BTRFS_I(old_dir), BTRFS_I(old_dentry->d_inode), old_name, &old_rename_ctx); - if (!ret) - ret = btrfs_update_inode(trans, BTRFS_I(old_inode)); - } - if (ret) { - btrfs_abort_transaction(trans, ret); - goto out_fail; + if (ret) { + btrfs_abort_transaction(trans, ret); + goto out_fail; + } + ret = btrfs_update_inode(trans, BTRFS_I(old_inode)); + if (ret) { + btrfs_abort_transaction(trans, ret); + goto out_fail; + } } /* dest is a subvolume */ if (new_ino == BTRFS_FIRST_FREE_OBJECTID) { ret = btrfs_unlink_subvol(trans, BTRFS_I(new_dir), new_dentry); + if (ret) { + btrfs_abort_transaction(trans, ret); + goto out_fail; + } } else { /* dest is an inode */ ret = __btrfs_unlink_inode(trans, BTRFS_I(new_dir), BTRFS_I(new_dentry->d_inode), new_name, &new_rename_ctx); - if (!ret) - ret = btrfs_update_inode(trans, BTRFS_I(new_inode)); - } - if (ret) { - btrfs_abort_transaction(trans, ret); - goto out_fail; + if (ret) { + btrfs_abort_transaction(trans, ret); + goto out_fail; + } + ret = btrfs_update_inode(trans, BTRFS_I(new_inode)); + if (ret) { + btrfs_abort_transaction(trans, ret); + goto out_fail; + } } ret = btrfs_add_link(trans, BTRFS_I(new_dir), BTRFS_I(old_inode), @@ -8287,16 +8335,23 @@ static int btrfs_rename(struct mnt_idmap *idmap, if (unlikely(old_ino == BTRFS_FIRST_FREE_OBJECTID)) { ret = btrfs_unlink_subvol(trans, BTRFS_I(old_dir), old_dentry); + if (ret) { + btrfs_abort_transaction(trans, ret); + goto out_fail; + } } else { ret = __btrfs_unlink_inode(trans, BTRFS_I(old_dir), BTRFS_I(d_inode(old_dentry)), &old_fname.disk_name, &rename_ctx); - if (!ret) - ret = btrfs_update_inode(trans, BTRFS_I(old_inode)); - } - if (ret) { - btrfs_abort_transaction(trans, ret); - goto out_fail; + if (ret) { + btrfs_abort_transaction(trans, ret); + goto out_fail; + } + ret = btrfs_update_inode(trans, BTRFS_I(old_inode)); + if (ret) { + btrfs_abort_transaction(trans, ret); + goto out_fail; + } } if (new_inode) { @@ -8304,18 +8359,27 @@ static int btrfs_rename(struct mnt_idmap *idmap, if (unlikely(btrfs_ino(BTRFS_I(new_inode)) == BTRFS_EMPTY_SUBVOL_DIR_OBJECTID)) { ret = btrfs_unlink_subvol(trans, BTRFS_I(new_dir), new_dentry); + if (ret) { + btrfs_abort_transaction(trans, ret); + goto out_fail; + } BUG_ON(new_inode->i_nlink == 0); } else { ret = btrfs_unlink_inode(trans, BTRFS_I(new_dir), BTRFS_I(d_inode(new_dentry)), &new_fname.disk_name); + if (ret) { + btrfs_abort_transaction(trans, ret); + goto out_fail; + } } - if (!ret && new_inode->i_nlink == 0) + if (new_inode->i_nlink == 0) { ret = btrfs_orphan_add(trans, BTRFS_I(d_inode(new_dentry))); - if (ret) { - btrfs_abort_transaction(trans, ret); - goto out_fail; + if (ret) { + btrfs_abort_transaction(trans, ret); + goto out_fail; + } } } @@ -8655,7 +8719,6 @@ static int btrfs_symlink(struct mnt_idmap *idmap, struct inode *dir, ptr = btrfs_file_extent_inline_start(ei); write_extent_buffer(leaf, symname, ptr, name_len); - btrfs_mark_buffer_dirty(trans, leaf); btrfs_free_path(path); d_instantiate_new(dentry, inode); @@ -9078,9 +9141,9 @@ out: } struct btrfs_encoded_read_private { - wait_queue_head_t wait; + struct completion done; void *uring_ctx; - atomic_t pending; + refcount_t pending_refs; blk_status_t status; }; @@ -9099,14 +9162,14 @@ static void btrfs_encoded_read_endio(struct btrfs_bio *bbio) */ WRITE_ONCE(priv->status, bbio->bio.bi_status); } - if (atomic_dec_and_test(&priv->pending)) { + if (refcount_dec_and_test(&priv->pending_refs)) { int err = blk_status_to_errno(READ_ONCE(priv->status)); if (priv->uring_ctx) { btrfs_uring_read_extent_endio(priv->uring_ctx, err); kfree(priv); } else { - wake_up(&priv->wait); + complete(&priv->done); } } bio_put(&bbio->bio); @@ -9126,8 +9189,8 @@ int btrfs_encoded_read_regular_fill_pages(struct btrfs_inode *inode, if (!priv) return -ENOMEM; - init_waitqueue_head(&priv->wait); - atomic_set(&priv->pending, 1); + init_completion(&priv->done); + refcount_set(&priv->pending_refs, 1); priv->status = 0; priv->uring_ctx = uring_ctx; @@ -9140,7 +9203,7 @@ int btrfs_encoded_read_regular_fill_pages(struct btrfs_inode *inode, size_t bytes = min_t(u64, disk_io_size, PAGE_SIZE); if (bio_add_page(&bbio->bio, pages[i], bytes, 0) < bytes) { - atomic_inc(&priv->pending); + refcount_inc(&priv->pending_refs); btrfs_submit_bbio(bbio, 0); bbio = btrfs_bio_alloc(BIO_MAX_VECS, REQ_OP_READ, fs_info, @@ -9155,11 +9218,11 @@ int btrfs_encoded_read_regular_fill_pages(struct btrfs_inode *inode, disk_io_size -= bytes; } while (disk_io_size); - atomic_inc(&priv->pending); + refcount_inc(&priv->pending_refs); btrfs_submit_bbio(bbio, 0); if (uring_ctx) { - if (atomic_dec_return(&priv->pending) == 0) { + if (refcount_dec_and_test(&priv->pending_refs)) { ret = blk_status_to_errno(READ_ONCE(priv->status)); btrfs_uring_read_extent_endio(uring_ctx, ret); kfree(priv); @@ -9168,8 +9231,8 @@ int btrfs_encoded_read_regular_fill_pages(struct btrfs_inode *inode, return -EIOCBQUEUED; } else { - if (atomic_dec_return(&priv->pending) != 0) - io_wait_event(priv->wait, !atomic_read(&priv->pending)); + if (!refcount_dec_and_test(&priv->pending_refs)) + wait_for_completion_io(&priv->done); /* See btrfs_encoded_read_endio() for ordering. */ ret = blk_status_to_errno(READ_ONCE(priv->status)); kfree(priv); @@ -9799,15 +9862,25 @@ static int btrfs_swap_activate(struct swap_info_struct *sis, struct file *file, struct btrfs_fs_info *fs_info = root->fs_info; struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree; struct extent_state *cached_state = NULL; - struct extent_map *em = NULL; struct btrfs_chunk_map *map = NULL; struct btrfs_device *device = NULL; struct btrfs_swap_info bsi = { .lowest_ppage = (sector_t)-1ULL, }; + struct btrfs_backref_share_check_ctx *backref_ctx = NULL; + struct btrfs_path *path = NULL; int ret = 0; u64 isize; - u64 start; + u64 prev_extent_end = 0; + + /* + * Acquire the inode's mmap lock to prevent races with memory mapped + * writes, as they could happen after we flush delalloc below and before + * we lock the extent range further below. The inode was already locked + * up in the call chain. + */ + btrfs_assert_inode_locked(BTRFS_I(inode)); + down_write(&BTRFS_I(inode)->i_mmap_lock); /* * If the swap file was just created, make sure delalloc is done. If the @@ -9816,22 +9889,32 @@ static int btrfs_swap_activate(struct swap_info_struct *sis, struct file *file, */ ret = btrfs_wait_ordered_range(BTRFS_I(inode), 0, (u64)-1); if (ret) - return ret; + goto out_unlock_mmap; /* * The inode is locked, so these flags won't change after we check them. */ if (BTRFS_I(inode)->flags & BTRFS_INODE_COMPRESS) { btrfs_warn(fs_info, "swapfile must not be compressed"); - return -EINVAL; + ret = -EINVAL; + goto out_unlock_mmap; } if (!(BTRFS_I(inode)->flags & BTRFS_INODE_NODATACOW)) { btrfs_warn(fs_info, "swapfile must not be copy-on-write"); - return -EINVAL; + ret = -EINVAL; + goto out_unlock_mmap; } if (!(BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM)) { btrfs_warn(fs_info, "swapfile must not be checksummed"); - return -EINVAL; + ret = -EINVAL; + goto out_unlock_mmap; + } + + path = btrfs_alloc_path(); + backref_ctx = btrfs_alloc_backref_share_check_ctx(); + if (!path || !backref_ctx) { + ret = -ENOMEM; + goto out_unlock_mmap; } /* @@ -9846,7 +9929,8 @@ static int btrfs_swap_activate(struct swap_info_struct *sis, struct file *file, if (!btrfs_exclop_start(fs_info, BTRFS_EXCLOP_SWAP_ACTIVATE)) { btrfs_warn(fs_info, "cannot activate swapfile while exclusive operation is running"); - return -EBUSY; + ret = -EBUSY; + goto out_unlock_mmap; } /* @@ -9860,7 +9944,8 @@ static int btrfs_swap_activate(struct swap_info_struct *sis, struct file *file, btrfs_exclop_finish(fs_info); btrfs_warn(fs_info, "cannot activate swapfile because snapshot creation is in progress"); - return -EINVAL; + ret = -EINVAL; + goto out_unlock_mmap; } /* * Snapshots can create extents which require COW even if NODATACOW is @@ -9881,7 +9966,8 @@ static int btrfs_swap_activate(struct swap_info_struct *sis, struct file *file, btrfs_warn(fs_info, "cannot activate swapfile because subvolume %llu is being deleted", btrfs_root_id(root)); - return -EPERM; + ret = -EPERM; + goto out_unlock_mmap; } atomic_inc(&root->nr_swapfiles); spin_unlock(&root->root_item_lock); @@ -9889,24 +9975,39 @@ static int btrfs_swap_activate(struct swap_info_struct *sis, struct file *file, isize = ALIGN_DOWN(inode->i_size, fs_info->sectorsize); lock_extent(io_tree, 0, isize - 1, &cached_state); - start = 0; - while (start < isize) { - u64 logical_block_start, physical_block_start; + while (prev_extent_end < isize) { + struct btrfs_key key; + struct extent_buffer *leaf; + struct btrfs_file_extent_item *ei; struct btrfs_block_group *bg; - u64 len = isize - start; + u64 logical_block_start; + u64 physical_block_start; + u64 extent_gen; + u64 disk_bytenr; + u64 len; - em = btrfs_get_extent(BTRFS_I(inode), NULL, start, len); - if (IS_ERR(em)) { - ret = PTR_ERR(em); + key.objectid = btrfs_ino(BTRFS_I(inode)); + key.type = BTRFS_EXTENT_DATA_KEY; + key.offset = prev_extent_end; + + ret = btrfs_search_slot(NULL, root, &key, path, 0, 0); + if (ret < 0) goto out; - } - if (em->disk_bytenr == EXTENT_MAP_HOLE) { + /* + * If key not found it means we have an implicit hole (NO_HOLES + * is enabled). + */ + if (ret > 0) { btrfs_warn(fs_info, "swapfile must not have holes"); ret = -EINVAL; goto out; } - if (em->disk_bytenr == EXTENT_MAP_INLINE) { + + leaf = path->nodes[0]; + ei = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_file_extent_item); + + if (btrfs_file_extent_type(leaf, ei) == BTRFS_FILE_EXTENT_INLINE) { /* * It's unlikely we'll ever actually find ourselves * here, as a file small enough to fit inline won't be @@ -9918,23 +10019,45 @@ static int btrfs_swap_activate(struct swap_info_struct *sis, struct file *file, ret = -EINVAL; goto out; } - if (extent_map_is_compressed(em)) { + + if (btrfs_file_extent_compression(leaf, ei) != BTRFS_COMPRESS_NONE) { btrfs_warn(fs_info, "swapfile must not be compressed"); ret = -EINVAL; goto out; } - logical_block_start = extent_map_block_start(em) + (start - em->start); - len = min(len, em->len - (start - em->start)); - free_extent_map(em); - em = NULL; + disk_bytenr = btrfs_file_extent_disk_bytenr(leaf, ei); + if (disk_bytenr == 0) { + btrfs_warn(fs_info, "swapfile must not have holes"); + ret = -EINVAL; + goto out; + } + + logical_block_start = disk_bytenr + btrfs_file_extent_offset(leaf, ei); + extent_gen = btrfs_file_extent_generation(leaf, ei); + prev_extent_end = btrfs_file_extent_end(path); + + if (prev_extent_end > isize) + len = isize - key.offset; + else + len = btrfs_file_extent_num_bytes(leaf, ei); + + backref_ctx->curr_leaf_bytenr = leaf->start; - ret = can_nocow_extent(inode, start, &len, NULL, false, true); + /* + * Don't need the path anymore, release to avoid deadlocks when + * calling btrfs_is_data_extent_shared() because when joining a + * transaction it can block waiting for the current one's commit + * which in turn may be trying to lock the same leaf to flush + * delayed items for example. + */ + btrfs_release_path(path); + + ret = btrfs_is_data_extent_shared(BTRFS_I(inode), disk_bytenr, + extent_gen, backref_ctx); if (ret < 0) { goto out; - } else if (ret) { - ret = 0; - } else { + } else if (ret > 0) { btrfs_warn(fs_info, "swapfile must not be copy-on-write"); ret = -EINVAL; @@ -9969,7 +10092,6 @@ static int btrfs_swap_activate(struct swap_info_struct *sis, struct file *file, physical_block_start = (map->stripes[0].physical + (logical_block_start - map->start)); - len = min(len, map->chunk_len - (logical_block_start - map->start)); btrfs_free_chunk_map(map); map = NULL; @@ -10010,20 +10132,23 @@ static int btrfs_swap_activate(struct swap_info_struct *sis, struct file *file, if (ret) goto out; } - bsi.start = start; + bsi.start = key.offset; bsi.block_start = physical_block_start; bsi.block_len = len; } - start += len; + if (fatal_signal_pending(current)) { + ret = -EINTR; + goto out; + } + + cond_resched(); } if (bsi.block_len) ret = btrfs_add_swap_extent(sis, &bsi); out: - if (!IS_ERR_OR_NULL(em)) - free_extent_map(em); if (!IS_ERR_OR_NULL(map)) btrfs_free_chunk_map(map); @@ -10036,6 +10161,10 @@ out: btrfs_exclop_finish(fs_info); +out_unlock_mmap: + up_write(&BTRFS_I(inode)->i_mmap_lock); + btrfs_free_backref_share_ctx(backref_ctx); + btrfs_free_path(path); if (ret) return ret; diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c index 3af8bb0c8d75..ae98269a5e3a 100644 --- a/fs/btrfs/ioctl.c +++ b/fs/btrfs/ioctl.c @@ -403,86 +403,6 @@ update_flags: return ret; } -/* - * Start exclusive operation @type, return true on success - */ -bool btrfs_exclop_start(struct btrfs_fs_info *fs_info, - enum btrfs_exclusive_operation type) -{ - bool ret = false; - - spin_lock(&fs_info->super_lock); - if (fs_info->exclusive_operation == BTRFS_EXCLOP_NONE) { - fs_info->exclusive_operation = type; - ret = true; - } - spin_unlock(&fs_info->super_lock); - - return ret; -} - -/* - * Conditionally allow to enter the exclusive operation in case it's compatible - * with the running one. This must be paired with btrfs_exclop_start_unlock and - * btrfs_exclop_finish. - * - * Compatibility: - * - the same type is already running - * - when trying to add a device and balance has been paused - * - not BTRFS_EXCLOP_NONE - this is intentionally incompatible and the caller - * must check the condition first that would allow none -> @type - */ -bool btrfs_exclop_start_try_lock(struct btrfs_fs_info *fs_info, - enum btrfs_exclusive_operation type) -{ - spin_lock(&fs_info->super_lock); - if (fs_info->exclusive_operation == type || - (fs_info->exclusive_operation == BTRFS_EXCLOP_BALANCE_PAUSED && - type == BTRFS_EXCLOP_DEV_ADD)) - return true; - - spin_unlock(&fs_info->super_lock); - return false; -} - -void btrfs_exclop_start_unlock(struct btrfs_fs_info *fs_info) -{ - spin_unlock(&fs_info->super_lock); -} - -void btrfs_exclop_finish(struct btrfs_fs_info *fs_info) -{ - spin_lock(&fs_info->super_lock); - WRITE_ONCE(fs_info->exclusive_operation, BTRFS_EXCLOP_NONE); - spin_unlock(&fs_info->super_lock); - sysfs_notify(&fs_info->fs_devices->fsid_kobj, NULL, "exclusive_operation"); -} - -void btrfs_exclop_balance(struct btrfs_fs_info *fs_info, - enum btrfs_exclusive_operation op) -{ - switch (op) { - case BTRFS_EXCLOP_BALANCE_PAUSED: - spin_lock(&fs_info->super_lock); - ASSERT(fs_info->exclusive_operation == BTRFS_EXCLOP_BALANCE || - fs_info->exclusive_operation == BTRFS_EXCLOP_DEV_ADD || - fs_info->exclusive_operation == BTRFS_EXCLOP_NONE || - fs_info->exclusive_operation == BTRFS_EXCLOP_BALANCE_PAUSED); - fs_info->exclusive_operation = BTRFS_EXCLOP_BALANCE_PAUSED; - spin_unlock(&fs_info->super_lock); - break; - case BTRFS_EXCLOP_BALANCE: - spin_lock(&fs_info->super_lock); - ASSERT(fs_info->exclusive_operation == BTRFS_EXCLOP_BALANCE_PAUSED); - fs_info->exclusive_operation = BTRFS_EXCLOP_BALANCE; - spin_unlock(&fs_info->super_lock); - break; - default: - btrfs_warn(fs_info, - "invalid exclop balance operation %d requested", op); - } -} - static int btrfs_ioctl_getversion(struct inode *inode, int __user *arg) { return put_user(inode->i_generation, arg); @@ -551,17 +471,6 @@ static noinline int btrfs_ioctl_fitrim(struct btrfs_fs_info *fs_info, return ret; } -int __pure btrfs_is_empty_uuid(const u8 *uuid) -{ - int i; - - for (i = 0; i < BTRFS_UUID_SIZE; i++) { - if (uuid[i]) - return 0; - } - return 1; -} - /* * Calculate the number of transaction items to reserve for creating a subvolume * or snapshot, not including the inode, directory entries, or parent directory. @@ -3007,7 +2916,6 @@ static long btrfs_ioctl_default_subvol(struct file *file, void __user *argp) btrfs_cpu_key_to_disk(&disk_key, &new_root->root_key); btrfs_set_dir_item_key(path->nodes[0], di, &disk_key); - btrfs_mark_buffer_dirty(trans, path->nodes[0]); btrfs_release_path(path); btrfs_set_fs_incompat(fs_info, DEFAULT_SUBVOL); @@ -4878,25 +4786,29 @@ out_fail: return ret; } +struct btrfs_uring_encoded_data { + struct btrfs_ioctl_encoded_io_args args; + struct iovec iovstack[UIO_FASTIOV]; + struct iovec *iov; + struct iov_iter iter; +}; + static int btrfs_uring_encoded_read(struct io_uring_cmd *cmd, unsigned int issue_flags) { size_t copy_end_kernel = offsetofend(struct btrfs_ioctl_encoded_io_args, flags); size_t copy_end; - struct btrfs_ioctl_encoded_io_args args = { 0 }; int ret; u64 disk_bytenr, disk_io_size; struct file *file; struct btrfs_inode *inode; struct btrfs_fs_info *fs_info; struct extent_io_tree *io_tree; - struct iovec iovstack[UIO_FASTIOV]; - struct iovec *iov = iovstack; - struct iov_iter iter; loff_t pos; struct kiocb kiocb; struct extent_state *cached_state = NULL; u64 start, lockend; void __user *sqe_addr; + struct btrfs_uring_encoded_data *data = io_uring_cmd_get_async_data(cmd)->op_data; if (!capable(CAP_SYS_ADMIN)) { ret = -EPERM; @@ -4910,43 +4822,64 @@ static int btrfs_uring_encoded_read(struct io_uring_cmd *cmd, unsigned int issue if (issue_flags & IO_URING_F_COMPAT) { #if defined(CONFIG_64BIT) && defined(CONFIG_COMPAT) - struct btrfs_ioctl_encoded_io_args_32 args32; - copy_end = offsetofend(struct btrfs_ioctl_encoded_io_args_32, flags); - if (copy_from_user(&args32, sqe_addr, copy_end)) { - ret = -EFAULT; - goto out_acct; - } - args.iov = compat_ptr(args32.iov); - args.iovcnt = args32.iovcnt; - args.offset = args32.offset; - args.flags = args32.flags; #else return -ENOTTY; #endif } else { copy_end = copy_end_kernel; - if (copy_from_user(&args, sqe_addr, copy_end)) { - ret = -EFAULT; + } + + if (!data) { + data = kzalloc(sizeof(*data), GFP_NOFS); + if (!data) { + ret = -ENOMEM; goto out_acct; } - } - if (args.flags != 0) - return -EINVAL; + io_uring_cmd_get_async_data(cmd)->op_data = data; - ret = import_iovec(ITER_DEST, args.iov, args.iovcnt, ARRAY_SIZE(iovstack), - &iov, &iter); - if (ret < 0) - goto out_acct; + if (issue_flags & IO_URING_F_COMPAT) { +#if defined(CONFIG_64BIT) && defined(CONFIG_COMPAT) + struct btrfs_ioctl_encoded_io_args_32 args32; - if (iov_iter_count(&iter) == 0) { - ret = 0; - goto out_free; + if (copy_from_user(&args32, sqe_addr, copy_end)) { + ret = -EFAULT; + goto out_acct; + } + + data->args.iov = compat_ptr(args32.iov); + data->args.iovcnt = args32.iovcnt; + data->args.offset = args32.offset; + data->args.flags = args32.flags; +#endif + } else { + if (copy_from_user(&data->args, sqe_addr, copy_end)) { + ret = -EFAULT; + goto out_acct; + } + } + + if (data->args.flags != 0) { + ret = -EINVAL; + goto out_acct; + } + + data->iov = data->iovstack; + ret = import_iovec(ITER_DEST, data->args.iov, data->args.iovcnt, + ARRAY_SIZE(data->iovstack), &data->iov, + &data->iter); + if (ret < 0) + goto out_acct; + + if (iov_iter_count(&data->iter) == 0) { + ret = 0; + goto out_free; + } } - pos = args.offset; - ret = rw_verify_area(READ, file, &pos, args.len); + pos = data->args.offset; + ret = rw_verify_area(READ, file, &pos, data->args.len); if (ret < 0) goto out_free; @@ -4959,15 +4892,16 @@ static int btrfs_uring_encoded_read(struct io_uring_cmd *cmd, unsigned int issue start = ALIGN_DOWN(pos, fs_info->sectorsize); lockend = start + BTRFS_MAX_UNCOMPRESSED - 1; - ret = btrfs_encoded_read(&kiocb, &iter, &args, &cached_state, + ret = btrfs_encoded_read(&kiocb, &data->iter, &data->args, &cached_state, &disk_bytenr, &disk_io_size); if (ret < 0 && ret != -EIOCBQUEUED) goto out_free; file_accessed(file); - if (copy_to_user(sqe_addr + copy_end, (const char *)&args + copy_end_kernel, - sizeof(args) - copy_end_kernel)) { + if (copy_to_user(sqe_addr + copy_end, + (const char *)&data->args + copy_end_kernel, + sizeof(data->args) - copy_end_kernel)) { if (ret == -EIOCBQUEUED) { unlock_extent(io_tree, start, lockend, &cached_state); btrfs_inode_unlock(inode, BTRFS_ILOCK_SHARED); @@ -4977,40 +4911,22 @@ static int btrfs_uring_encoded_read(struct io_uring_cmd *cmd, unsigned int issue } if (ret == -EIOCBQUEUED) { - u64 count; - - /* - * If we've optimized things by storing the iovecs on the stack, - * undo this. - */ - if (!iov) { - iov = kmalloc(sizeof(struct iovec) * args.iovcnt, GFP_NOFS); - if (!iov) { - unlock_extent(io_tree, start, lockend, &cached_state); - btrfs_inode_unlock(inode, BTRFS_ILOCK_SHARED); - ret = -ENOMEM; - goto out_acct; - } - - memcpy(iov, iovstack, sizeof(struct iovec) * args.iovcnt); - } - - count = min_t(u64, iov_iter_count(&iter), disk_io_size); + u64 count = min_t(u64, iov_iter_count(&data->iter), disk_io_size); /* Match ioctl by not returning past EOF if uncompressed. */ - if (!args.compression) - count = min_t(u64, count, args.len); + if (!data->args.compression) + count = min_t(u64, count, data->args.len); - ret = btrfs_uring_read_extent(&kiocb, &iter, start, lockend, - cached_state, disk_bytenr, - disk_io_size, count, - args.compression, iov, cmd); + ret = btrfs_uring_read_extent(&kiocb, &data->iter, start, lockend, + cached_state, disk_bytenr, disk_io_size, + count, data->args.compression, + data->iov, cmd); goto out_acct; } out_free: - kfree(iov); + kfree(data->iov); out_acct: if (ret > 0) @@ -5020,6 +4936,128 @@ out_acct: return ret; } +static int btrfs_uring_encoded_write(struct io_uring_cmd *cmd, unsigned int issue_flags) +{ + loff_t pos; + struct kiocb kiocb; + struct file *file; + ssize_t ret; + void __user *sqe_addr; + struct btrfs_uring_encoded_data *data = io_uring_cmd_get_async_data(cmd)->op_data; + + if (!capable(CAP_SYS_ADMIN)) { + ret = -EPERM; + goto out_acct; + } + + file = cmd->file; + sqe_addr = u64_to_user_ptr(READ_ONCE(cmd->sqe->addr)); + + if (!(file->f_mode & FMODE_WRITE)) { + ret = -EBADF; + goto out_acct; + } + + if (!data) { + data = kzalloc(sizeof(*data), GFP_NOFS); + if (!data) { + ret = -ENOMEM; + goto out_acct; + } + + io_uring_cmd_get_async_data(cmd)->op_data = data; + + if (issue_flags & IO_URING_F_COMPAT) { +#if defined(CONFIG_64BIT) && defined(CONFIG_COMPAT) + struct btrfs_ioctl_encoded_io_args_32 args32; + + if (copy_from_user(&args32, sqe_addr, sizeof(args32))) { + ret = -EFAULT; + goto out_acct; + } + data->args.iov = compat_ptr(args32.iov); + data->args.iovcnt = args32.iovcnt; + data->args.offset = args32.offset; + data->args.flags = args32.flags; + data->args.len = args32.len; + data->args.unencoded_len = args32.unencoded_len; + data->args.unencoded_offset = args32.unencoded_offset; + data->args.compression = args32.compression; + data->args.encryption = args32.encryption; + memcpy(data->args.reserved, args32.reserved, + sizeof(data->args.reserved)); +#else + ret = -ENOTTY; + goto out_acct; +#endif + } else { + if (copy_from_user(&data->args, sqe_addr, sizeof(data->args))) { + ret = -EFAULT; + goto out_acct; + } + } + + ret = -EINVAL; + if (data->args.flags != 0) + goto out_acct; + if (memchr_inv(data->args.reserved, 0, sizeof(data->args.reserved))) + goto out_acct; + if (data->args.compression == BTRFS_ENCODED_IO_COMPRESSION_NONE && + data->args.encryption == BTRFS_ENCODED_IO_ENCRYPTION_NONE) + goto out_acct; + if (data->args.compression >= BTRFS_ENCODED_IO_COMPRESSION_TYPES || + data->args.encryption >= BTRFS_ENCODED_IO_ENCRYPTION_TYPES) + goto out_acct; + if (data->args.unencoded_offset > data->args.unencoded_len) + goto out_acct; + if (data->args.len > data->args.unencoded_len - data->args.unencoded_offset) + goto out_acct; + + data->iov = data->iovstack; + ret = import_iovec(ITER_SOURCE, data->args.iov, data->args.iovcnt, + ARRAY_SIZE(data->iovstack), &data->iov, + &data->iter); + if (ret < 0) + goto out_acct; + + if (iov_iter_count(&data->iter) == 0) { + ret = 0; + goto out_iov; + } + } + + if (issue_flags & IO_URING_F_NONBLOCK) { + ret = -EAGAIN; + goto out_acct; + } + + pos = data->args.offset; + ret = rw_verify_area(WRITE, file, &pos, data->args.len); + if (ret < 0) + goto out_iov; + + init_sync_kiocb(&kiocb, file); + ret = kiocb_set_rw_flags(&kiocb, 0, WRITE); + if (ret) + goto out_iov; + kiocb.ki_pos = pos; + + file_start_write(file); + + ret = btrfs_do_write_iter(&kiocb, &data->iter, &data->args); + if (ret > 0) + fsnotify_modify(file); + + file_end_write(file); +out_iov: + kfree(data->iov); +out_acct: + if (ret > 0) + add_wchar(current, ret); + inc_syscw(current); + return ret; +} + int btrfs_uring_cmd(struct io_uring_cmd *cmd, unsigned int issue_flags) { switch (cmd->cmd_op) { @@ -5028,6 +5066,12 @@ int btrfs_uring_cmd(struct io_uring_cmd *cmd, unsigned int issue_flags) case BTRFS_IOC_ENCODED_READ_32: #endif return btrfs_uring_encoded_read(cmd, issue_flags); + + case BTRFS_IOC_ENCODED_WRITE: +#if defined(CONFIG_64BIT) && defined(CONFIG_COMPAT) + case BTRFS_IOC_ENCODED_WRITE_32: +#endif + return btrfs_uring_encoded_write(cmd, issue_flags); } return -EINVAL; @@ -5300,6 +5344,8 @@ long btrfs_ioctl(struct file *file, unsigned int return fsverity_ioctl_enable(file, (const void __user *)argp); case FS_IOC_MEASURE_VERITY: return fsverity_ioctl_measure(file, argp); + case FS_IOC_READ_VERITY_METADATA: + return fsverity_ioctl_read_metadata(file, argp); case BTRFS_IOC_ENCODED_READ: return btrfs_ioctl_encoded_read(file, argp, false); case BTRFS_IOC_ENCODED_WRITE: diff --git a/fs/btrfs/ioctl.h b/fs/btrfs/ioctl.h index 2b760c8778f8..ce915fcda43b 100644 --- a/fs/btrfs/ioctl.h +++ b/fs/btrfs/ioctl.h @@ -19,7 +19,6 @@ int btrfs_fileattr_set(struct mnt_idmap *idmap, struct dentry *dentry, struct fileattr *fa); int btrfs_ioctl_get_supported_features(void __user *arg); void btrfs_sync_inode_flags_to_i_flags(struct inode *inode); -int __pure btrfs_is_empty_uuid(const u8 *uuid); void btrfs_update_ioctl_balance_args(struct btrfs_fs_info *fs_info, struct btrfs_ioctl_balance_args *bargs); int btrfs_uring_cmd(struct io_uring_cmd *cmd, unsigned int issue_flags); diff --git a/fs/btrfs/locking.h b/fs/btrfs/locking.h index 35036b151bf5..c69e57ff804b 100644 --- a/fs/btrfs/locking.h +++ b/fs/btrfs/locking.h @@ -199,8 +199,13 @@ static inline void btrfs_assert_tree_write_locked(struct extent_buffer *eb) { lockdep_assert_held_write(&eb->lock); } +static inline void btrfs_assert_tree_read_locked(struct extent_buffer *eb) +{ + lockdep_assert_held_read(&eb->lock); +} #else static inline void btrfs_assert_tree_write_locked(struct extent_buffer *eb) { } +static inline void btrfs_assert_tree_read_locked(struct extent_buffer *eb) { } #endif void btrfs_unlock_up_safe(struct btrfs_path *path, int level); diff --git a/fs/btrfs/qgroup.c b/fs/btrfs/qgroup.c index a6f92836c9b1..b90fabe302e6 100644 --- a/fs/btrfs/qgroup.c +++ b/fs/btrfs/qgroup.c @@ -673,9 +673,6 @@ static int add_qgroup_relation_item(struct btrfs_trans_handle *trans, u64 src, key.offset = dst; ret = btrfs_insert_empty_item(trans, quota_root, path, &key, 0); - - btrfs_mark_buffer_dirty(trans, path->nodes[0]); - btrfs_free_path(path); return ret; } @@ -752,8 +749,6 @@ static int add_qgroup_item(struct btrfs_trans_handle *trans, btrfs_set_qgroup_info_excl(leaf, qgroup_info, 0); btrfs_set_qgroup_info_excl_cmpr(leaf, qgroup_info, 0); - btrfs_mark_buffer_dirty(trans, leaf); - btrfs_release_path(path); key.type = BTRFS_QGROUP_LIMIT_KEY; @@ -771,8 +766,6 @@ static int add_qgroup_item(struct btrfs_trans_handle *trans, btrfs_set_qgroup_limit_rsv_rfer(leaf, qgroup_limit, 0); btrfs_set_qgroup_limit_rsv_excl(leaf, qgroup_limit, 0); - btrfs_mark_buffer_dirty(trans, leaf); - ret = 0; out: btrfs_free_path(path); @@ -859,9 +852,6 @@ static int update_qgroup_limit_item(struct btrfs_trans_handle *trans, btrfs_set_qgroup_limit_max_excl(l, qgroup_limit, qgroup->max_excl); btrfs_set_qgroup_limit_rsv_rfer(l, qgroup_limit, qgroup->rsv_rfer); btrfs_set_qgroup_limit_rsv_excl(l, qgroup_limit, qgroup->rsv_excl); - - btrfs_mark_buffer_dirty(trans, l); - out: btrfs_free_path(path); return ret; @@ -905,9 +895,6 @@ static int update_qgroup_info_item(struct btrfs_trans_handle *trans, btrfs_set_qgroup_info_rfer_cmpr(l, qgroup_info, qgroup->rfer_cmpr); btrfs_set_qgroup_info_excl(l, qgroup_info, qgroup->excl); btrfs_set_qgroup_info_excl_cmpr(l, qgroup_info, qgroup->excl_cmpr); - - btrfs_mark_buffer_dirty(trans, l); - out: btrfs_free_path(path); return ret; @@ -947,9 +934,6 @@ static int update_qgroup_status_item(struct btrfs_trans_handle *trans) btrfs_set_qgroup_status_generation(l, ptr, trans->transid); btrfs_set_qgroup_status_rescan(l, ptr, fs_info->qgroup_rescan_progress.objectid); - - btrfs_mark_buffer_dirty(trans, l); - out: btrfs_free_path(path); return ret; @@ -1121,6 +1105,7 @@ int btrfs_quota_enable(struct btrfs_fs_info *fs_info, fs_info->qgroup_flags = BTRFS_QGROUP_STATUS_FLAG_ON; if (simple) { fs_info->qgroup_flags |= BTRFS_QGROUP_STATUS_FLAG_SIMPLE_MODE; + btrfs_set_fs_incompat(fs_info, SIMPLE_QUOTA); btrfs_set_qgroup_status_enable_gen(leaf, ptr, trans->transid); } else { fs_info->qgroup_flags |= BTRFS_QGROUP_STATUS_FLAG_INCONSISTENT; @@ -1129,8 +1114,6 @@ int btrfs_quota_enable(struct btrfs_fs_info *fs_info, BTRFS_QGROUP_STATUS_FLAGS_MASK); btrfs_set_qgroup_status_rescan(leaf, ptr, 0); - btrfs_mark_buffer_dirty(trans, leaf); - key.objectid = 0; key.type = BTRFS_ROOT_REF_KEY; key.offset = 0; @@ -1254,8 +1237,6 @@ out_add_root: spin_lock(&fs_info->qgroup_lock); fs_info->quota_root = quota_root; set_bit(BTRFS_FS_QUOTA_ENABLED, &fs_info->flags); - if (simple) - btrfs_set_fs_incompat(fs_info, SIMPLE_QUOTA); spin_unlock(&fs_info->qgroup_lock); /* Skip rescan for simple qgroups. */ @@ -1839,9 +1820,19 @@ int btrfs_remove_qgroup(struct btrfs_trans_handle *trans, u64 qgroupid) * Thus its reserved space should all be zero, no matter if qgroup * is consistent or the mode. */ - WARN_ON(qgroup->rsv.values[BTRFS_QGROUP_RSV_DATA] || - qgroup->rsv.values[BTRFS_QGROUP_RSV_META_PREALLOC] || - qgroup->rsv.values[BTRFS_QGROUP_RSV_META_PERTRANS]); + if (qgroup->rsv.values[BTRFS_QGROUP_RSV_DATA] || + qgroup->rsv.values[BTRFS_QGROUP_RSV_META_PREALLOC] || + qgroup->rsv.values[BTRFS_QGROUP_RSV_META_PERTRANS]) { + WARN_ON(IS_ENABLED(CONFIG_BTRFS_DEBUG)); + btrfs_warn_rl(fs_info, +"to be deleted qgroup %u/%llu has non-zero numbers, data %llu meta prealloc %llu meta pertrans %llu", + btrfs_qgroup_level(qgroup->qgroupid), + btrfs_qgroup_subvolid(qgroup->qgroupid), + qgroup->rsv.values[BTRFS_QGROUP_RSV_DATA], + qgroup->rsv.values[BTRFS_QGROUP_RSV_META_PREALLOC], + qgroup->rsv.values[BTRFS_QGROUP_RSV_META_PERTRANS]); + + } /* * The same for rfer/excl numbers, but that's only if our qgroup is * consistent and if it's in regular qgroup mode. @@ -1850,8 +1841,9 @@ int btrfs_remove_qgroup(struct btrfs_trans_handle *trans, u64 qgroupid) */ if (btrfs_qgroup_mode(fs_info) == BTRFS_QGROUP_MODE_FULL && !(fs_info->qgroup_flags & BTRFS_QGROUP_STATUS_FLAG_INCONSISTENT)) { - if (WARN_ON(qgroup->rfer || qgroup->excl || - qgroup->rfer_cmpr || qgroup->excl_cmpr)) { + if (qgroup->rfer || qgroup->excl || + qgroup->rfer_cmpr || qgroup->excl_cmpr) { + WARN_ON(IS_ENABLED(CONFIG_BTRFS_DEBUG)); btrfs_warn_rl(fs_info, "to be deleted qgroup %u/%llu has non-zero numbers, rfer %llu rfer_cmpr %llu excl %llu excl_cmpr %llu", btrfs_qgroup_level(qgroup->qgroupid), diff --git a/fs/btrfs/raid-stripe-tree.c b/fs/btrfs/raid-stripe-tree.c index 9ffc79f250fb..1834011ccc49 100644 --- a/fs/btrfs/raid-stripe-tree.c +++ b/fs/btrfs/raid-stripe-tree.c @@ -13,12 +13,13 @@ #include "volumes.h" #include "print-tree.h" -static void btrfs_partially_delete_raid_extent(struct btrfs_trans_handle *trans, +static int btrfs_partially_delete_raid_extent(struct btrfs_trans_handle *trans, struct btrfs_path *path, const struct btrfs_key *oldkey, u64 newlen, u64 frontpad) { - struct btrfs_stripe_extent *extent; + struct btrfs_root *stripe_root = trans->fs_info->stripe_root; + struct btrfs_stripe_extent *extent, *newitem; struct extent_buffer *leaf; int slot; size_t item_size; @@ -27,23 +28,39 @@ static void btrfs_partially_delete_raid_extent(struct btrfs_trans_handle *trans, .type = BTRFS_RAID_STRIPE_KEY, .offset = newlen, }; + int ret; + ASSERT(newlen > 0); ASSERT(oldkey->type == BTRFS_RAID_STRIPE_KEY); leaf = path->nodes[0]; slot = path->slots[0]; item_size = btrfs_item_size(leaf, slot); + + newitem = kzalloc(item_size, GFP_NOFS); + if (!newitem) + return -ENOMEM; + extent = btrfs_item_ptr(leaf, slot, struct btrfs_stripe_extent); for (int i = 0; i < btrfs_num_raid_stripes(item_size); i++) { struct btrfs_raid_stride *stride = &extent->strides[i]; u64 phys; - phys = btrfs_raid_stride_physical(leaf, stride); - btrfs_set_raid_stride_physical(leaf, stride, phys + frontpad); + phys = btrfs_raid_stride_physical(leaf, stride) + frontpad; + btrfs_set_stack_raid_stride_physical(&newitem->strides[i], phys); } - btrfs_set_item_key_safe(trans, path, &newkey); + ret = btrfs_del_item(trans, stripe_root, path); + if (ret) + goto out; + + btrfs_release_path(path); + ret = btrfs_insert_item(trans, stripe_root, &newkey, newitem, item_size); + +out: + kfree(newitem); + return ret; } int btrfs_delete_raid_extent(struct btrfs_trans_handle *trans, u64 start, u64 length) @@ -59,9 +76,22 @@ int btrfs_delete_raid_extent(struct btrfs_trans_handle *trans, u64 start, u64 le int slot; int ret; - if (!stripe_root) + if (!btrfs_fs_incompat(fs_info, RAID_STRIPE_TREE) || !stripe_root) return 0; + if (!btrfs_is_testing(fs_info)) { + struct btrfs_chunk_map *map; + bool use_rst; + + map = btrfs_find_chunk_map(fs_info, start, length); + if (!map) + return -EINVAL; + use_rst = btrfs_need_stripe_tree_update(fs_info, map->type); + btrfs_free_chunk_map(map); + if (!use_rst) + return 0; + } + path = btrfs_alloc_path(); if (!path) return -ENOMEM; @@ -85,6 +115,37 @@ int btrfs_delete_raid_extent(struct btrfs_trans_handle *trans, u64 start, u64 le found_end = found_start + key.offset; ret = 0; + /* + * The stripe extent starts before the range we want to delete, + * but the range spans more than one stripe extent: + * + * |--- RAID Stripe Extent ---||--- RAID Stripe Extent ---| + * |--- keep ---|--- drop ---| + * + * This means we have to get the previous item, truncate its + * length and then restart the search. + */ + if (found_start > start) { + if (slot == 0) { + ret = btrfs_previous_item(stripe_root, path, start, + BTRFS_RAID_STRIPE_KEY); + if (ret) { + if (ret > 0) + ret = -ENOENT; + break; + } + } else { + path->slots[0]--; + } + + leaf = path->nodes[0]; + slot = path->slots[0]; + btrfs_item_key_to_cpu(leaf, &key, slot); + found_start = key.objectid; + found_end = found_start + key.offset; + ASSERT(found_start <= start); + } + if (key.type != BTRFS_RAID_STRIPE_KEY) break; @@ -96,6 +157,54 @@ int btrfs_delete_raid_extent(struct btrfs_trans_handle *trans, u64 start, u64 le found_start, found_end); /* + * The stripe extent starts before the range we want to delete + * and ends after the range we want to delete, i.e. we're + * punching a hole in the stripe extent: + * + * |--- RAID Stripe Extent ---| + * | keep |--- drop ---| keep | + * + * This means we need to a) truncate the existing item and b) + * create a second item for the remaining range. + */ + if (found_start < start && found_end > end) { + size_t item_size; + u64 diff_start = start - found_start; + u64 diff_end = found_end - end; + struct btrfs_stripe_extent *extent; + struct btrfs_key newkey = { + .objectid = end, + .type = BTRFS_RAID_STRIPE_KEY, + .offset = diff_end, + }; + + /* The "right" item. */ + ret = btrfs_duplicate_item(trans, stripe_root, path, &newkey); + if (ret) + break; + + item_size = btrfs_item_size(leaf, path->slots[0]); + extent = btrfs_item_ptr(leaf, path->slots[0], + struct btrfs_stripe_extent); + + for (int i = 0; i < btrfs_num_raid_stripes(item_size); i++) { + struct btrfs_raid_stride *stride = &extent->strides[i]; + u64 phys; + + phys = btrfs_raid_stride_physical(leaf, stride); + phys += diff_start + length; + btrfs_set_raid_stride_physical(leaf, stride, phys); + } + + /* The "left" item. */ + path->slots[0]--; + btrfs_item_key_to_cpu(leaf, &key, path->slots[0]); + btrfs_partially_delete_raid_extent(trans, path, &key, + diff_start, 0); + break; + } + + /* * The stripe extent starts before the range we want to delete: * * |--- RAID Stripe Extent ---| @@ -105,11 +214,18 @@ int btrfs_delete_raid_extent(struct btrfs_trans_handle *trans, u64 start, u64 le * length to the new size and then re-insert the item. */ if (found_start < start) { - u64 diff = start - found_start; + u64 diff_start = start - found_start; btrfs_partially_delete_raid_extent(trans, path, &key, - diff, 0); - break; + diff_start, 0); + + start += (key.offset - diff_start); + length -= (key.offset - diff_start); + if (length == 0) + break; + + btrfs_release_path(path); + continue; } /* @@ -122,13 +238,16 @@ int btrfs_delete_raid_extent(struct btrfs_trans_handle *trans, u64 start, u64 le * length to the new size and then re-insert the item. */ if (found_end > end) { - u64 diff = found_end - end; + u64 diff_end = found_end - end; btrfs_partially_delete_raid_extent(trans, path, &key, - diff, diff); + key.offset - length, + length); + ASSERT(key.offset - diff_end == length); break; } + /* Finally we can delete the whole item, no more special cases. */ ret = btrfs_del_item(trans, stripe_root, path); if (ret) break; @@ -169,7 +288,6 @@ static int update_raid_extent_item(struct btrfs_trans_handle *trans, write_extent_buffer(leaf, stripe_extent, btrfs_item_ptr_offset(leaf, slot), item_size); - btrfs_mark_buffer_dirty(trans, leaf); btrfs_free_path(path); return ret; @@ -199,12 +317,8 @@ int btrfs_insert_one_raid_extent(struct btrfs_trans_handle *trans, for (int i = 0; i < num_stripes; i++) { u64 devid = bioc->stripes[i].dev->devid; u64 physical = bioc->stripes[i].physical; - u64 length = bioc->stripes[i].length; struct btrfs_raid_stride *raid_stride = &stripe_extent->strides[i]; - if (length == 0) - length = bioc->size; - btrfs_set_stack_raid_stride_devid(raid_stride, devid); btrfs_set_stack_raid_stride_physical(raid_stride, physical); } diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c index bf267bdfa8f8..af0969b70b53 100644 --- a/fs/btrfs/relocation.c +++ b/fs/btrfs/relocation.c @@ -342,12 +342,6 @@ static bool handle_useless_nodes(struct reloc_control *rc, if (cur == node) ret = true; - /* The node is the lowest node */ - if (cur->lowest) { - list_del_init(&cur->lower); - cur->lowest = 0; - } - /* Cleanup the lower edges */ while (!list_empty(&cur->lower)) { struct btrfs_backref_edge *edge; @@ -373,7 +367,6 @@ static bool handle_useless_nodes(struct reloc_control *rc, * cache to avoid unnecessary backref lookup. */ if (cur->level > 0) { - list_add(&cur->list, &cache->detached); cur->detached = 1; } else { rb_erase(&cur->rb_node, &cache->rb_root); @@ -426,7 +419,6 @@ static noinline_for_stack struct btrfs_backref_node *build_backref_tree( goto out; } - node->lowest = 1; cur = node; /* Breadth-first search to build backref cache */ @@ -470,92 +462,6 @@ out: } /* - * helper to add backref node for the newly created snapshot. - * the backref node is created by cloning backref node that - * corresponds to root of source tree - */ -static int clone_backref_node(struct btrfs_trans_handle *trans, - struct reloc_control *rc, - const struct btrfs_root *src, - struct btrfs_root *dest) -{ - struct btrfs_root *reloc_root = src->reloc_root; - struct btrfs_backref_cache *cache = &rc->backref_cache; - struct btrfs_backref_node *node = NULL; - struct btrfs_backref_node *new_node; - struct btrfs_backref_edge *edge; - struct btrfs_backref_edge *new_edge; - struct rb_node *rb_node; - - rb_node = rb_simple_search(&cache->rb_root, src->commit_root->start); - if (rb_node) { - node = rb_entry(rb_node, struct btrfs_backref_node, rb_node); - if (node->detached) - node = NULL; - else - BUG_ON(node->new_bytenr != reloc_root->node->start); - } - - if (!node) { - rb_node = rb_simple_search(&cache->rb_root, - reloc_root->commit_root->start); - if (rb_node) { - node = rb_entry(rb_node, struct btrfs_backref_node, - rb_node); - BUG_ON(node->detached); - } - } - - if (!node) - return 0; - - new_node = btrfs_backref_alloc_node(cache, dest->node->start, - node->level); - if (!new_node) - return -ENOMEM; - - new_node->lowest = node->lowest; - new_node->checked = 1; - new_node->root = btrfs_grab_root(dest); - ASSERT(new_node->root); - - if (!node->lowest) { - list_for_each_entry(edge, &node->lower, list[UPPER]) { - new_edge = btrfs_backref_alloc_edge(cache); - if (!new_edge) - goto fail; - - btrfs_backref_link_edge(new_edge, edge->node[LOWER], - new_node, LINK_UPPER); - } - } else { - list_add_tail(&new_node->lower, &cache->leaves); - } - - rb_node = rb_simple_insert(&cache->rb_root, new_node->bytenr, - &new_node->rb_node); - if (rb_node) - btrfs_backref_panic(trans->fs_info, new_node->bytenr, -EEXIST); - - if (!new_node->lowest) { - list_for_each_entry(new_edge, &new_node->lower, list[UPPER]) { - list_add_tail(&new_edge->list[LOWER], - &new_edge->node[LOWER]->upper); - } - } - return 0; -fail: - while (!list_empty(&new_node->lower)) { - new_edge = list_entry(new_node->lower.next, - struct btrfs_backref_edge, list[UPPER]); - list_del(&new_edge->list[UPPER]); - btrfs_backref_free_edge(cache, new_edge); - } - btrfs_backref_free_node(cache, new_node); - return -ENOMEM; -} - -/* * helper to add 'address of tree root -> reloc tree' mapping */ static int __add_reloc_root(struct btrfs_root *root) @@ -950,7 +856,6 @@ int replace_file_extents(struct btrfs_trans_handle *trans, u32 i; int ret = 0; int first = 1; - int dirty = 0; if (rc->stage != UPDATE_DATA_PTRS) return 0; @@ -1030,7 +935,6 @@ int replace_file_extents(struct btrfs_trans_handle *trans, } btrfs_set_file_extent_disk_bytenr(leaf, fi, new_bytenr); - dirty = 1; key.offset -= btrfs_file_extent_offset(leaf, fi); ref.action = BTRFS_ADD_DELAYED_REF; @@ -1061,8 +965,6 @@ int replace_file_extents(struct btrfs_trans_handle *trans, break; } } - if (dirty) - btrfs_mark_buffer_dirty(trans, leaf); if (inode) btrfs_add_delayed_iput(inode); return ret; @@ -1255,13 +1157,11 @@ again: */ btrfs_set_node_blockptr(parent, slot, new_bytenr); btrfs_set_node_ptr_generation(parent, slot, new_ptr_gen); - btrfs_mark_buffer_dirty(trans, parent); btrfs_set_node_blockptr(path->nodes[level], path->slots[level], old_bytenr); btrfs_set_node_ptr_generation(path->nodes[level], path->slots[level], old_ptr_gen); - btrfs_mark_buffer_dirty(trans, path->nodes[level]); ref.action = BTRFS_ADD_DELAYED_REF; ref.bytenr = old_bytenr; @@ -2058,100 +1958,72 @@ struct btrfs_root *select_reloc_root(struct btrfs_trans_handle *trans, int index = 0; int ret; - next = node; - while (1) { - cond_resched(); - next = walk_up_backref(next, edges, &index); - root = next->root; + next = walk_up_backref(node, edges, &index); + root = next->root; - /* - * If there is no root, then our references for this block are - * incomplete, as we should be able to walk all the way up to a - * block that is owned by a root. - * - * This path is only for SHAREABLE roots, so if we come upon a - * non-SHAREABLE root then we have backrefs that resolve - * improperly. - * - * Both of these cases indicate file system corruption, or a bug - * in the backref walking code. - */ - if (!root) { - ASSERT(0); - btrfs_err(trans->fs_info, - "bytenr %llu doesn't have a backref path ending in a root", - node->bytenr); - return ERR_PTR(-EUCLEAN); - } - if (!test_bit(BTRFS_ROOT_SHAREABLE, &root->state)) { - ASSERT(0); - btrfs_err(trans->fs_info, - "bytenr %llu has multiple refs with one ending in a non-shareable root", - node->bytenr); - return ERR_PTR(-EUCLEAN); - } - - if (btrfs_root_id(root) == BTRFS_TREE_RELOC_OBJECTID) { - ret = record_reloc_root_in_trans(trans, root); - if (ret) - return ERR_PTR(ret); - break; - } + /* + * If there is no root, then our references for this block are + * incomplete, as we should be able to walk all the way up to a block + * that is owned by a root. + * + * This path is only for SHAREABLE roots, so if we come upon a + * non-SHAREABLE root then we have backrefs that resolve improperly. + * + * Both of these cases indicate file system corruption, or a bug in the + * backref walking code. + */ + if (unlikely(!root)) { + btrfs_err(trans->fs_info, + "bytenr %llu doesn't have a backref path ending in a root", + node->bytenr); + return ERR_PTR(-EUCLEAN); + } + if (unlikely(!test_bit(BTRFS_ROOT_SHAREABLE, &root->state))) { + btrfs_err(trans->fs_info, + "bytenr %llu has multiple refs with one ending in a non-shareable root", + node->bytenr); + return ERR_PTR(-EUCLEAN); + } - ret = btrfs_record_root_in_trans(trans, root); + if (btrfs_root_id(root) == BTRFS_TREE_RELOC_OBJECTID) { + ret = record_reloc_root_in_trans(trans, root); if (ret) return ERR_PTR(ret); - root = root->reloc_root; - - /* - * We could have raced with another thread which failed, so - * root->reloc_root may not be set, return ENOENT in this case. - */ - if (!root) - return ERR_PTR(-ENOENT); + goto found; + } - if (next->new_bytenr != root->node->start) { - /* - * We just created the reloc root, so we shouldn't have - * ->new_bytenr set and this shouldn't be in the changed - * list. If it is then we have multiple roots pointing - * at the same bytenr which indicates corruption, or - * we've made a mistake in the backref walking code. - */ - ASSERT(next->new_bytenr == 0); - ASSERT(list_empty(&next->list)); - if (next->new_bytenr || !list_empty(&next->list)) { - btrfs_err(trans->fs_info, - "bytenr %llu possibly has multiple roots pointing at the same bytenr %llu", - node->bytenr, next->bytenr); - return ERR_PTR(-EUCLEAN); - } + ret = btrfs_record_root_in_trans(trans, root); + if (ret) + return ERR_PTR(ret); + root = root->reloc_root; - next->new_bytenr = root->node->start; - btrfs_put_root(next->root); - next->root = btrfs_grab_root(root); - ASSERT(next->root); - list_add_tail(&next->list, - &rc->backref_cache.changed); - mark_block_processed(rc, next); - break; - } + /* + * We could have raced with another thread which failed, so + * root->reloc_root may not be set, return ENOENT in this case. + */ + if (!root) + return ERR_PTR(-ENOENT); - WARN_ON(1); - root = NULL; - next = walk_down_backref(edges, &index); - if (!next || next->level <= node->level) - break; - } - if (!root) { + if (next->new_bytenr) { /* - * This can happen if there's fs corruption or if there's a bug - * in the backref lookup code. + * We just created the reloc root, so we shouldn't have + * ->new_bytenr set yet. If it is then we have multiple roots + * pointing at the same bytenr which indicates corruption, or + * we've made a mistake in the backref walking code. */ - ASSERT(0); - return ERR_PTR(-ENOENT); + ASSERT(next->new_bytenr == 0); + btrfs_err(trans->fs_info, + "bytenr %llu possibly has multiple roots pointing at the same bytenr %llu", + node->bytenr, next->bytenr); + return ERR_PTR(-EUCLEAN); } + next->new_bytenr = root->node->start; + btrfs_put_root(next->root); + next->root = btrfs_grab_root(root); + ASSERT(next->root); + mark_block_processed(rc, next); +found: next = node; /* setup backref node path for btrfs_reloc_cow_block */ while (1) { @@ -2247,17 +2119,11 @@ static noinline_for_stack u64 calcu_metadata_size(struct reloc_control *rc, return num_bytes; } -static int reserve_metadata_space(struct btrfs_trans_handle *trans, - struct reloc_control *rc, - struct btrfs_backref_node *node) +static int refill_metadata_space(struct btrfs_trans_handle *trans, + struct reloc_control *rc, u64 num_bytes) { - struct btrfs_root *root = rc->extent_root; - struct btrfs_fs_info *fs_info = root->fs_info; - u64 num_bytes; + struct btrfs_fs_info *fs_info = trans->fs_info; int ret; - u64 tmp; - - num_bytes = calcu_metadata_size(rc, node) * 2; trans->block_rsv = rc->block_rsv; rc->reserved_bytes += num_bytes; @@ -2270,7 +2136,8 @@ static int reserve_metadata_space(struct btrfs_trans_handle *trans, ret = btrfs_block_rsv_refill(fs_info, rc->block_rsv, num_bytes, BTRFS_RESERVE_FLUSH_LIMIT); if (ret) { - tmp = fs_info->nodesize * RELOCATION_RESERVED_NODES; + u64 tmp = fs_info->nodesize * RELOCATION_RESERVED_NODES; + while (tmp <= rc->reserved_bytes) tmp <<= 1; /* @@ -2288,6 +2155,16 @@ static int reserve_metadata_space(struct btrfs_trans_handle *trans, return 0; } +static int reserve_metadata_space(struct btrfs_trans_handle *trans, + struct reloc_control *rc, + struct btrfs_backref_node *node) +{ + u64 num_bytes; + + num_bytes = calcu_metadata_size(rc, node) * 2; + return refill_metadata_space(trans, rc, num_bytes); +} + /* * relocate a block tree, and then update pointers in upper level * blocks that reference the block to point to the new location. @@ -2442,7 +2319,7 @@ next: if (!ret && node->pending) { btrfs_backref_drop_node_buffer(node); - list_move_tail(&node->list, &rc->backref_cache.changed); + list_del_init(&node->list); node->pending = 0; } @@ -2605,8 +2482,7 @@ static int relocate_tree_block(struct btrfs_trans_handle *trans, /* * This block was the root block of a root, and this is * the first time we're processing the block and thus it - * should not have had the ->new_bytenr modified and - * should have not been included on the changed list. + * should not have had the ->new_bytenr modified. * * However in the case of corruption we could have * multiple refs pointing to the same block improperly, @@ -2616,8 +2492,7 @@ static int relocate_tree_block(struct btrfs_trans_handle *trans, * normal user in the case of corruption. */ ASSERT(node->new_bytenr == 0); - ASSERT(list_empty(&node->list)); - if (node->new_bytenr || !list_empty(&node->list)) { + if (node->new_bytenr) { btrfs_err(root->fs_info, "bytenr %llu has improper references to it", node->bytenr); @@ -2640,17 +2515,12 @@ static int relocate_tree_block(struct btrfs_trans_handle *trans, btrfs_put_root(node->root); node->root = btrfs_grab_root(root); ASSERT(node->root); - list_add_tail(&node->list, &rc->backref_cache.changed); } else { - path->lowest_level = node->level; - if (root == root->fs_info->chunk_root) - btrfs_reserve_chunk_metadata(trans, false); - ret = btrfs_search_slot(trans, root, key, path, 0, 1); - btrfs_release_path(path); - if (root == root->fs_info->chunk_root) - btrfs_trans_release_chunk_metadata(trans); - if (ret > 0) - ret = 0; + btrfs_err(root->fs_info, + "bytenr %llu resolved to a non-shareable root", + node->bytenr); + ret = -EUCLEAN; + goto out; } if (!ret) update_processed_blocks(rc, node); @@ -2658,11 +2528,50 @@ static int relocate_tree_block(struct btrfs_trans_handle *trans, ret = do_relocation(trans, rc, node, key, path, 1); } out: - if (ret || node->level == 0 || node->cowonly) + if (ret || node->level == 0) btrfs_backref_cleanup_node(&rc->backref_cache, node); return ret; } +static int relocate_cowonly_block(struct btrfs_trans_handle *trans, + struct reloc_control *rc, struct tree_block *block, + struct btrfs_path *path) +{ + struct btrfs_fs_info *fs_info = trans->fs_info; + struct btrfs_root *root; + u64 num_bytes; + int nr_levels; + int ret; + + root = btrfs_get_fs_root(fs_info, block->owner, true); + if (IS_ERR(root)) + return PTR_ERR(root); + + nr_levels = max(btrfs_header_level(root->node) - block->level, 0) + 1; + + num_bytes = fs_info->nodesize * nr_levels; + ret = refill_metadata_space(trans, rc, num_bytes); + if (ret) { + btrfs_put_root(root); + return ret; + } + path->lowest_level = block->level; + if (root == root->fs_info->chunk_root) + btrfs_reserve_chunk_metadata(trans, false); + + ret = btrfs_search_slot(trans, root, &block->key, path, 0, 1); + path->lowest_level = 0; + btrfs_release_path(path); + + if (root == root->fs_info->chunk_root) + btrfs_trans_release_chunk_metadata(trans); + if (ret > 0) + ret = 0; + btrfs_put_root(root); + + return ret; +} + /* * relocate a list of blocks */ @@ -2702,6 +2611,20 @@ int relocate_tree_blocks(struct btrfs_trans_handle *trans, /* Do tree relocation */ rbtree_postorder_for_each_entry_safe(block, next, blocks, rb_node) { + /* + * For COWonly blocks, or the data reloc tree, we only need to + * COW down to the block, there's no need to generate a backref + * tree. + */ + if (block->owner && + (!is_fstree(block->owner) || + block->owner == BTRFS_DATA_RELOC_TREE_OBJECTID)) { + ret = relocate_cowonly_block(trans, rc, block, path); + if (ret) + break; + continue; + } + node = build_backref_tree(trans, rc, &block->key, block->level, block->bytenr); if (IS_ERR(node)) { @@ -2902,6 +2825,7 @@ static int relocate_one_folio(struct reloc_control *rc, const bool use_rst = btrfs_need_stripe_tree_update(fs_info, rc->block_group->flags); ASSERT(index <= last_index); +again: folio = filemap_lock_folio(inode->i_mapping, index); if (IS_ERR(folio)) { @@ -2937,11 +2861,16 @@ static int relocate_one_folio(struct reloc_control *rc, ret = -EIO; goto release_folio; } + if (folio->mapping != inode->i_mapping) { + folio_unlock(folio); + folio_put(folio); + goto again; + } } /* * We could have lost folio private when we dropped the lock to read the - * folio above, make sure we set_page_extent_mapped here so we have any + * folio above, make sure we set_folio_extent_mapped() here so we have any * of the subpage blocksize stuff we need in place. */ ret = set_folio_extent_mapped(folio); @@ -3793,7 +3722,6 @@ static int __insert_orphan_inode(struct btrfs_trans_handle *trans, btrfs_set_inode_mode(leaf, item, S_IFREG | 0600); btrfs_set_inode_flags(leaf, item, BTRFS_INODE_NOCOMPRESS | BTRFS_INODE_PREALLOC); - btrfs_mark_buffer_dirty(trans, leaf); out: btrfs_free_path(path); return ret; @@ -4399,8 +4327,18 @@ int btrfs_reloc_cow_block(struct btrfs_trans_handle *trans, WARN_ON(!first_cow && level == 0); node = rc->backref_cache.path[level]; - BUG_ON(node->bytenr != buf->start && - node->new_bytenr != buf->start); + + /* + * If node->bytenr != buf->start and node->new_bytenr != + * buf->start then we've got the wrong backref node for what we + * expected to see here and the cache is incorrect. + */ + if (unlikely(node->bytenr != buf->start && node->new_bytenr != buf->start)) { + btrfs_err(fs_info, +"bytenr %llu was found but our backref cache was expecting %llu or %llu", + buf->start, node->bytenr, node->new_bytenr); + return -EUCLEAN; + } btrfs_backref_drop_node_buffer(node); atomic_inc(&cow->refs); @@ -4500,10 +4438,7 @@ int btrfs_reloc_post_snapshot(struct btrfs_trans_handle *trans, return ret; } new_root->reloc_root = btrfs_grab_root(reloc_root); - - if (rc->create_reloc_tree) - ret = clone_backref_node(trans, rc, root, reloc_root); - return ret; + return 0; } /* diff --git a/fs/btrfs/root-tree.c b/fs/btrfs/root-tree.c index 33962671a96c..e22e6b06927a 100644 --- a/fs/btrfs/root-tree.c +++ b/fs/btrfs/root-tree.c @@ -197,7 +197,6 @@ int btrfs_update_root(struct btrfs_trans_handle *trans, struct btrfs_root btrfs_set_root_generation_v2(item, btrfs_root_generation(item)); write_extent_buffer(l, item, ptr, sizeof(*item)); - btrfs_mark_buffer_dirty(trans, path->nodes[0]); out: btrfs_free_path(path); return ret; @@ -447,7 +446,6 @@ again: btrfs_set_root_ref_name_len(leaf, ref, name->len); ptr = (unsigned long)(ref + 1); write_extent_buffer(leaf, name->name, ptr, name->len); - btrfs_mark_buffer_dirty(trans, leaf); if (key.type == BTRFS_ROOT_BACKREF_KEY) { btrfs_release_path(path); diff --git a/fs/btrfs/scrub.c b/fs/btrfs/scrub.c index 204c928beaf9..531312efee8d 100644 --- a/fs/btrfs/scrub.c +++ b/fs/btrfs/scrub.c @@ -1541,6 +1541,10 @@ static int scrub_find_fill_first_stripe(struct btrfs_block_group *bg, u64 extent_gen; int ret; + if (unlikely(!extent_root)) { + btrfs_err(fs_info, "no valid extent root for scrub"); + return -EUCLEAN; + } memset(stripe->sectors, 0, sizeof(struct scrub_sector_verification) * stripe->nr_sectors); scrub_stripe_reset_bitmaps(stripe); diff --git a/fs/btrfs/send.c b/fs/btrfs/send.c index 7254279c3cc9..f437138fefbc 100644 --- a/fs/btrfs/send.c +++ b/fs/btrfs/send.c @@ -5280,6 +5280,7 @@ static int put_file_data(struct send_ctx *sctx, u64 offset, u32 len) unsigned cur_len = min_t(unsigned, len, PAGE_SIZE - pg_offset); +again: folio = filemap_lock_folio(mapping, index); if (IS_ERR(folio)) { page_cache_sync_readahead(mapping, @@ -5312,6 +5313,11 @@ static int put_file_data(struct send_ctx *sctx, u64 offset, u32 len) ret = -EIO; break; } + if (folio->mapping != mapping) { + folio_unlock(folio); + folio_put(folio); + goto again; + } } memcpy_from_folio(sctx->send_buf + sctx->send_size, folio, @@ -7253,7 +7259,7 @@ static int changed_cb(struct btrfs_path *left_path, enum btrfs_compare_tree_result result, struct send_ctx *sctx) { - int ret = 0; + int ret; /* * We can not hold the commit root semaphore here. This is because in @@ -7313,7 +7319,6 @@ static int changed_cb(struct btrfs_path *left_path, return 0; } result = BTRFS_COMPARE_TREE_CHANGED; - ret = 0; } sctx->left_path = left_path; diff --git a/fs/btrfs/space-info.c b/fs/btrfs/space-info.c index 255e85f78313..a341d087567a 100644 --- a/fs/btrfs/space-info.c +++ b/fs/btrfs/space-info.c @@ -14,6 +14,7 @@ #include "fs.h" #include "accessors.h" #include "extent-tree.h" +#include "zoned.h" /* * HOW DOES SPACE RESERVATION WORK @@ -127,6 +128,14 @@ * churn a lot and we can avoid making some extent tree modifications if we * are able to delay for as long as possible. * + * RESET_ZONES + * This state works only for the zoned mode. On the zoned mode, we cannot + * reuse once allocated then freed region until we reset the zone, due to + * the sequential write zone requirement. The RESET_ZONES state resets the + * zones of an unused block group and let us reuse the space. The reusing + * is faster than removing the block group and allocating another block + * group on the zones. + * * ALLOC_CHUNK * We will skip this the first time through space reservation, because of * overcommit and we don't want to have a lot of useless metadata space when @@ -316,7 +325,7 @@ void btrfs_add_bg_to_space_info(struct btrfs_fs_info *info, found->bytes_used += block_group->used; found->disk_used += block_group->used * factor; found->bytes_readonly += block_group->bytes_super; - btrfs_space_info_update_bytes_zone_unusable(info, found, block_group->zone_unusable); + btrfs_space_info_update_bytes_zone_unusable(found, block_group->zone_unusable); if (block_group->length > 0) found->full = 0; btrfs_try_granting_tickets(info, found); @@ -489,9 +498,7 @@ again: if ((used + ticket->bytes <= space_info->total_bytes) || btrfs_can_overcommit(fs_info, space_info, ticket->bytes, flush)) { - btrfs_space_info_update_bytes_may_use(fs_info, - space_info, - ticket->bytes); + btrfs_space_info_update_bytes_may_use(space_info, ticket->bytes); remove_ticket(space_info, ticket); ticket->bytes = 0; space_info->tickets_id++; @@ -834,6 +841,9 @@ static void flush_space(struct btrfs_fs_info *fs_info, */ ret = btrfs_commit_current_transaction(root); break; + case RESET_ZONES: + ret = btrfs_reset_unused_block_groups(space_info, num_bytes); + break; default: ret = -ENOSPC; break; @@ -1086,9 +1096,14 @@ static void btrfs_async_reclaim_metadata_space(struct work_struct *work) enum btrfs_flush_state flush_state; int commit_cycles = 0; u64 last_tickets_id; + enum btrfs_flush_state final_state; fs_info = container_of(work, struct btrfs_fs_info, async_reclaim_work); space_info = btrfs_find_space_info(fs_info, BTRFS_BLOCK_GROUP_METADATA); + if (btrfs_is_zoned(fs_info)) + final_state = RESET_ZONES; + else + final_state = COMMIT_TRANS; spin_lock(&space_info->lock); to_reclaim = btrfs_calc_reclaim_metadata_size(fs_info, space_info); @@ -1141,7 +1156,7 @@ static void btrfs_async_reclaim_metadata_space(struct work_struct *work) if (flush_state == ALLOC_CHUNK_FORCE && !commit_cycles) flush_state++; - if (flush_state > COMMIT_TRANS) { + if (flush_state > final_state) { commit_cycles++; if (commit_cycles > 2) { if (maybe_fail_all_tickets(fs_info, space_info)) { @@ -1155,7 +1170,7 @@ static void btrfs_async_reclaim_metadata_space(struct work_struct *work) } } spin_unlock(&space_info->lock); - } while (flush_state <= COMMIT_TRANS); + } while (flush_state <= final_state); } /* @@ -1286,6 +1301,10 @@ static void btrfs_preempt_reclaim_metadata_space(struct work_struct *work) * This is where we reclaim all of the pinned space generated by running the * iputs * + * RESET_ZONES + * This state works only for the zoned mode. We scan the unused block group + * list and reset the zones and reuse the block group. + * * ALLOC_CHUNK_FORCE * For data we start with alloc chunk force, however we could have been full * before, and then the transaction commit could have freed new block groups, @@ -1295,6 +1314,7 @@ static const enum btrfs_flush_state data_flush_states[] = { FLUSH_DELALLOC_FULL, RUN_DELAYED_IPUTS, COMMIT_TRANS, + RESET_ZONES, ALLOC_CHUNK_FORCE, }; @@ -1386,6 +1406,7 @@ void btrfs_init_async_reclaim_work(struct btrfs_fs_info *fs_info) static const enum btrfs_flush_state priority_flush_states[] = { FLUSH_DELAYED_ITEMS_NR, FLUSH_DELAYED_ITEMS, + RESET_ZONES, ALLOC_CHUNK, }; @@ -1399,6 +1420,7 @@ static const enum btrfs_flush_state evict_flush_states[] = { FLUSH_DELALLOC_FULL, ALLOC_CHUNK, COMMIT_TRANS, + RESET_ZONES, }; static void priority_reclaim_metadata_space(struct btrfs_fs_info *fs_info, @@ -1690,8 +1712,7 @@ static int __reserve_bytes(struct btrfs_fs_info *fs_info, if (!pending_tickets && ((used + orig_bytes <= space_info->total_bytes) || btrfs_can_overcommit(fs_info, space_info, orig_bytes, flush))) { - btrfs_space_info_update_bytes_may_use(fs_info, space_info, - orig_bytes); + btrfs_space_info_update_bytes_may_use(space_info, orig_bytes); ret = 0; } @@ -1703,8 +1724,7 @@ static int __reserve_bytes(struct btrfs_fs_info *fs_info, if (ret && unlikely(flush == BTRFS_RESERVE_FLUSH_EMERGENCY)) { used = btrfs_space_info_used(space_info, false); if (used + orig_bytes <= space_info->total_bytes) { - btrfs_space_info_update_bytes_may_use(fs_info, space_info, - orig_bytes); + btrfs_space_info_update_bytes_may_use(space_info, orig_bytes); ret = 0; } } @@ -2082,3 +2102,32 @@ void btrfs_reclaim_sweep(const struct btrfs_fs_info *fs_info) do_reclaim_sweep(space_info, raid); } } + +void btrfs_return_free_space(struct btrfs_space_info *space_info, u64 len) +{ + struct btrfs_fs_info *fs_info = space_info->fs_info; + struct btrfs_block_rsv *global_rsv = &fs_info->global_block_rsv; + + lockdep_assert_held(&space_info->lock); + + /* Prioritize the global reservation to receive the freed space. */ + if (global_rsv->space_info != space_info) + goto grant; + + spin_lock(&global_rsv->lock); + if (!global_rsv->full) { + u64 to_add = min(len, global_rsv->size - global_rsv->reserved); + + global_rsv->reserved += to_add; + btrfs_space_info_update_bytes_may_use(space_info, to_add); + if (global_rsv->reserved >= global_rsv->size) + global_rsv->full = 1; + len -= to_add; + } + spin_unlock(&global_rsv->lock); + +grant: + /* Add to any tickets we may have. */ + if (len) + btrfs_try_granting_tickets(fs_info, space_info); +} diff --git a/fs/btrfs/space-info.h b/fs/btrfs/space-info.h index efbecc0c5258..a96efdb5e681 100644 --- a/fs/btrfs/space-info.h +++ b/fs/btrfs/space-info.h @@ -79,6 +79,10 @@ enum btrfs_reserve_flush_enum { BTRFS_RESERVE_FLUSH_EMERGENCY, }; +/* + * Please be aware that the order of enum values will be the order of the reclaim + * process in btrfs_async_reclaim_metadata_space(). + */ enum btrfs_flush_state { FLUSH_DELAYED_ITEMS_NR = 1, FLUSH_DELAYED_ITEMS = 2, @@ -91,6 +95,7 @@ enum btrfs_flush_state { ALLOC_CHUNK_FORCE = 9, RUN_DELAYED_IPUTS = 10, COMMIT_TRANS = 11, + RESET_ZONES = 12, }; struct btrfs_space_info { @@ -229,10 +234,10 @@ static inline bool btrfs_mixed_space_info(const struct btrfs_space_info *space_i */ #define DECLARE_SPACE_INFO_UPDATE(name, trace_name) \ static inline void \ -btrfs_space_info_update_##name(struct btrfs_fs_info *fs_info, \ - struct btrfs_space_info *sinfo, \ +btrfs_space_info_update_##name(struct btrfs_space_info *sinfo, \ s64 bytes) \ { \ + struct btrfs_fs_info *fs_info = sinfo->fs_info; \ const u64 abs_bytes = (bytes < 0) ? -bytes : bytes; \ lockdep_assert_held(&sinfo->lock); \ trace_update_##name(fs_info, sinfo, sinfo->name, bytes); \ @@ -275,13 +280,12 @@ int btrfs_can_overcommit(struct btrfs_fs_info *fs_info, enum btrfs_reserve_flush_enum flush); static inline void btrfs_space_info_free_bytes_may_use( - struct btrfs_fs_info *fs_info, struct btrfs_space_info *space_info, u64 num_bytes) { spin_lock(&space_info->lock); - btrfs_space_info_update_bytes_may_use(fs_info, space_info, -num_bytes); - btrfs_try_granting_tickets(fs_info, space_info); + btrfs_space_info_update_bytes_may_use(space_info, -num_bytes); + btrfs_try_granting_tickets(space_info->fs_info, space_info); spin_unlock(&space_info->lock); } int btrfs_reserve_data_bytes(struct btrfs_fs_info *fs_info, u64 bytes, @@ -295,5 +299,6 @@ void btrfs_set_periodic_reclaim_ready(struct btrfs_space_info *space_info, bool bool btrfs_should_periodic_reclaim(struct btrfs_space_info *space_info); int btrfs_calc_reclaim_threshold(const struct btrfs_space_info *space_info); void btrfs_reclaim_sweep(const struct btrfs_fs_info *fs_info); +void btrfs_return_free_space(struct btrfs_space_info *space_info, u64 len); #endif /* BTRFS_SPACE_INFO_H */ diff --git a/fs/btrfs/subpage.c b/fs/btrfs/subpage.c index 8c68059ac1b0..722acf768396 100644 --- a/fs/btrfs/subpage.c +++ b/fs/btrfs/subpage.c @@ -635,6 +635,28 @@ IMPLEMENT_BTRFS_PAGE_OPS(ordered, folio_set_ordered, folio_clear_ordered, IMPLEMENT_BTRFS_PAGE_OPS(checked, folio_set_checked, folio_clear_checked, folio_test_checked); +#define GET_SUBPAGE_BITMAP(subpage, fs_info, name, dst) \ +{ \ + const int sectors_per_page = fs_info->sectors_per_page; \ + \ + ASSERT(sectors_per_page < BITS_PER_LONG); \ + *dst = bitmap_read(subpage->bitmaps, \ + sectors_per_page * btrfs_bitmap_nr_##name, \ + sectors_per_page); \ +} + +#define SUBPAGE_DUMP_BITMAP(fs_info, folio, name, start, len) \ +{ \ + const struct btrfs_subpage *subpage = folio_get_private(folio); \ + unsigned long bitmap; \ + \ + GET_SUBPAGE_BITMAP(subpage, fs_info, name, &bitmap); \ + btrfs_warn(fs_info, \ + "dumpping bitmap start=%llu len=%u folio=%llu " #name "_bitmap=%*pbl", \ + start, len, folio_pos(folio), \ + fs_info->sectors_per_page, &bitmap); \ +} + /* * Make sure not only the page dirty bit is cleared, but also subpage dirty bit * is cleared. @@ -660,6 +682,10 @@ void btrfs_folio_assert_not_dirty(const struct btrfs_fs_info *fs_info, subpage = folio_get_private(folio); ASSERT(subpage); spin_lock_irqsave(&subpage->lock, flags); + if (unlikely(!bitmap_test_range_all_zero(subpage->bitmaps, start_bit, nbits))) { + SUBPAGE_DUMP_BITMAP(fs_info, folio, dirty, start, len); + ASSERT(bitmap_test_range_all_zero(subpage->bitmaps, start_bit, nbits)); + } ASSERT(bitmap_test_range_all_zero(subpage->bitmaps, start_bit, nbits)); spin_unlock_irqrestore(&subpage->lock, flags); } @@ -689,23 +715,16 @@ void btrfs_folio_set_lock(const struct btrfs_fs_info *fs_info, nbits = len >> fs_info->sectorsize_bits; spin_lock_irqsave(&subpage->lock, flags); /* Target range should not yet be locked. */ - ASSERT(bitmap_test_range_all_zero(subpage->bitmaps, start_bit, nbits)); + if (unlikely(!bitmap_test_range_all_zero(subpage->bitmaps, start_bit, nbits))) { + SUBPAGE_DUMP_BITMAP(fs_info, folio, locked, start, len); + ASSERT(bitmap_test_range_all_zero(subpage->bitmaps, start_bit, nbits)); + } bitmap_set(subpage->bitmaps, start_bit, nbits); ret = atomic_add_return(nbits, &subpage->nr_locked); ASSERT(ret <= fs_info->sectors_per_page); spin_unlock_irqrestore(&subpage->lock, flags); } -#define GET_SUBPAGE_BITMAP(subpage, fs_info, name, dst) \ -{ \ - const int sectors_per_page = fs_info->sectors_per_page; \ - \ - ASSERT(sectors_per_page < BITS_PER_LONG); \ - *dst = bitmap_read(subpage->bitmaps, \ - sectors_per_page * btrfs_bitmap_nr_##name, \ - sectors_per_page); \ -} - void __cold btrfs_subpage_dump_bitmap(const struct btrfs_fs_info *fs_info, struct folio *folio, u64 start, u32 len) { @@ -716,6 +735,7 @@ void __cold btrfs_subpage_dump_bitmap(const struct btrfs_fs_info *fs_info, unsigned long writeback_bitmap; unsigned long ordered_bitmap; unsigned long checked_bitmap; + unsigned long locked_bitmap; unsigned long flags; ASSERT(folio_test_private(folio) && folio_get_private(folio)); @@ -728,15 +748,16 @@ void __cold btrfs_subpage_dump_bitmap(const struct btrfs_fs_info *fs_info, GET_SUBPAGE_BITMAP(subpage, fs_info, writeback, &writeback_bitmap); GET_SUBPAGE_BITMAP(subpage, fs_info, ordered, &ordered_bitmap); GET_SUBPAGE_BITMAP(subpage, fs_info, checked, &checked_bitmap); - GET_SUBPAGE_BITMAP(subpage, fs_info, locked, &checked_bitmap); + GET_SUBPAGE_BITMAP(subpage, fs_info, locked, &locked_bitmap); spin_unlock_irqrestore(&subpage->lock, flags); dump_page(folio_page(folio, 0), "btrfs subpage dump"); btrfs_warn(fs_info, -"start=%llu len=%u page=%llu, bitmaps uptodate=%*pbl dirty=%*pbl writeback=%*pbl ordered=%*pbl checked=%*pbl", +"start=%llu len=%u page=%llu, bitmaps uptodate=%*pbl dirty=%*pbl locked=%*pbl writeback=%*pbl ordered=%*pbl checked=%*pbl", start, len, folio_pos(folio), sectors_per_page, &uptodate_bitmap, sectors_per_page, &dirty_bitmap, + sectors_per_page, &locked_bitmap, sectors_per_page, &writeback_bitmap, sectors_per_page, &ordered_bitmap, sectors_per_page, &checked_bitmap); diff --git a/fs/btrfs/subpage.h b/fs/btrfs/subpage.h index 428fa9389fd4..44fff1f4eac4 100644 --- a/fs/btrfs/subpage.h +++ b/fs/btrfs/subpage.h @@ -137,6 +137,19 @@ DECLARE_BTRFS_SUBPAGE_OPS(writeback); DECLARE_BTRFS_SUBPAGE_OPS(ordered); DECLARE_BTRFS_SUBPAGE_OPS(checked); +/* + * Helper for error cleanup, where a folio will have its dirty flag cleared, + * with writeback started and finished. + */ +static inline void btrfs_folio_clamp_finish_io(struct btrfs_fs_info *fs_info, + struct folio *locked_folio, + u64 start, u32 len) +{ + btrfs_folio_clamp_clear_dirty(fs_info, locked_folio, start, len); + btrfs_folio_clamp_set_writeback(fs_info, locked_folio, start, len); + btrfs_folio_clamp_clear_writeback(fs_info, locked_folio, start, len); +} + bool btrfs_subpage_clear_and_test_dirty(const struct btrfs_fs_info *fs_info, struct folio *folio, u64 start, u32 len); diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c index 7dfe5005129a..f809c3200c21 100644 --- a/fs/btrfs/super.c +++ b/fs/btrfs/super.c @@ -971,7 +971,7 @@ static int btrfs_fill_super(struct super_block *sb, err = open_ctree(sb, fs_devices); if (err) { - btrfs_err(fs_info, "open_ctree failed"); + btrfs_err(fs_info, "open_ctree failed: %d", err); return err; } @@ -2446,6 +2446,9 @@ static __cold void btrfs_interface_exit(void) static int __init btrfs_print_mod_info(void) { static const char options[] = "" +#ifdef CONFIG_BTRFS_EXPERIMENTAL + ", experimental=on" +#endif #ifdef CONFIG_BTRFS_DEBUG ", debug=on" #endif @@ -2466,7 +2469,17 @@ static int __init btrfs_print_mod_info(void) ", fsverity=no" #endif ; + +#ifdef CONFIG_BTRFS_EXPERIMENTAL + if (btrfs_get_mod_read_policy() == NULL) + pr_info("Btrfs loaded%s\n", options); + else + pr_info("Btrfs loaded%s, read_policy=%s\n", + options, btrfs_get_mod_read_policy()); +#else pr_info("Btrfs loaded%s\n", options); +#endif + return 0; } @@ -2524,6 +2537,11 @@ static const struct init_sequence mod_init_seq[] = { }, { .init_func = extent_map_init, .exit_func = extent_map_exit, +#ifdef CONFIG_BTRFS_EXPERIMENTAL + }, { + .init_func = btrfs_read_policy_init, + .exit_func = NULL, +#endif }, { .init_func = ordered_data_init, .exit_func = ordered_data_exit, diff --git a/fs/btrfs/sysfs.c b/fs/btrfs/sysfs.c index fdcbf650ac31..53b846d99ece 100644 --- a/fs/btrfs/sysfs.c +++ b/fs/btrfs/sysfs.c @@ -1118,7 +1118,7 @@ static ssize_t btrfs_nodesize_show(struct kobject *kobj, { struct btrfs_fs_info *fs_info = to_fs_info(kobj); - return sysfs_emit(buf, "%u\n", fs_info->super_copy->nodesize); + return sysfs_emit(buf, "%u\n", fs_info->nodesize); } BTRFS_ATTR(, nodesize, btrfs_nodesize_show); @@ -1128,7 +1128,7 @@ static ssize_t btrfs_sectorsize_show(struct kobject *kobj, { struct btrfs_fs_info *fs_info = to_fs_info(kobj); - return sysfs_emit(buf, "%u\n", fs_info->super_copy->sectorsize); + return sysfs_emit(buf, "%u\n", fs_info->sectorsize); } BTRFS_ATTR(, sectorsize, btrfs_sectorsize_show); @@ -1180,7 +1180,7 @@ static ssize_t btrfs_clone_alignment_show(struct kobject *kobj, { struct btrfs_fs_info *fs_info = to_fs_info(kobj); - return sysfs_emit(buf, "%u\n", fs_info->super_copy->sectorsize); + return sysfs_emit(buf, "%u\n", fs_info->sectorsize); } BTRFS_ATTR(, clone_alignment, btrfs_clone_alignment_show); @@ -1305,7 +1305,73 @@ static ssize_t btrfs_temp_fsid_show(struct kobject *kobj, } BTRFS_ATTR(, temp_fsid, btrfs_temp_fsid_show); -static const char * const btrfs_read_policy_name[] = { "pid" }; +static const char *btrfs_read_policy_name[] = { + "pid", +#ifdef CONFIG_BTRFS_EXPERIMENTAL + "round-robin", + "devid", +#endif +}; + +#ifdef CONFIG_BTRFS_EXPERIMENTAL + +/* Global module configuration parameters. */ +static char *read_policy; +char *btrfs_get_mod_read_policy(void) +{ + return read_policy; +} + +/* Set perms to 0, disable /sys/module/btrfs/parameter/read_policy interface. */ +module_param(read_policy, charp, 0); +MODULE_PARM_DESC(read_policy, +"Global read policy: pid (default), round-robin[:<min_contig_read>], devid[:<devid>]"); +#endif + +int btrfs_read_policy_to_enum(const char *str, s64 *value_ret) +{ + char param[32] = { 0 }; + char __maybe_unused *value_str; + + if (!str || strlen(str) == 0) + return 0; + + strncpy(param, str, sizeof(param) - 1); + +#ifdef CONFIG_BTRFS_EXPERIMENTAL + /* Separate value from input in policy:value format. */ + value_str = strchr(param, ':'); + if (value_str) { + int ret; + + *value_str = 0; + value_str++; + if (!value_ret) + return -EINVAL; + ret = kstrtos64(value_str, 10, value_ret); + if (ret) + return -EINVAL; + if (*value_ret < 0) + return -ERANGE; + } +#endif + + return sysfs_match_string(btrfs_read_policy_name, param); +} + +#ifdef CONFIG_BTRFS_EXPERIMENTAL +int __init btrfs_read_policy_init(void) +{ + s64 value; + + if (btrfs_read_policy_to_enum(read_policy, &value) == -EINVAL) { + btrfs_err(NULL, "invalid read policy or value %s", read_policy); + return -EINVAL; + } + + return 0; +} +#endif static ssize_t btrfs_read_policy_show(struct kobject *kobj, struct kobj_attribute *a, char *buf) @@ -1316,14 +1382,25 @@ static ssize_t btrfs_read_policy_show(struct kobject *kobj, int i; for (i = 0; i < BTRFS_NR_READ_POLICY; i++) { - if (policy == i) - ret += sysfs_emit_at(buf, ret, "%s[%s]", - (ret == 0 ? "" : " "), - btrfs_read_policy_name[i]); - else - ret += sysfs_emit_at(buf, ret, "%s%s", - (ret == 0 ? "" : " "), - btrfs_read_policy_name[i]); + if (ret != 0) + ret += sysfs_emit_at(buf, ret, " "); + + if (i == policy) + ret += sysfs_emit_at(buf, ret, "["); + + ret += sysfs_emit_at(buf, ret, "%s", btrfs_read_policy_name[i]); + +#ifdef CONFIG_BTRFS_EXPERIMENTAL + if (i == BTRFS_READ_POLICY_RR) + ret += sysfs_emit_at(buf, ret, ":%u", + READ_ONCE(fs_devices->rr_min_contig_read)); + + if (i == BTRFS_READ_POLICY_DEVID) + ret += sysfs_emit_at(buf, ret, ":%llu", + READ_ONCE(fs_devices->read_devid)); +#endif + if (i == policy) + ret += sysfs_emit_at(buf, ret, "]"); } ret += sysfs_emit_at(buf, ret, "\n"); @@ -1336,21 +1413,80 @@ static ssize_t btrfs_read_policy_store(struct kobject *kobj, const char *buf, size_t len) { struct btrfs_fs_devices *fs_devices = to_fs_devs(kobj); - int i; + int index; + s64 value = -1; - for (i = 0; i < BTRFS_NR_READ_POLICY; i++) { - if (sysfs_streq(buf, btrfs_read_policy_name[i])) { - if (i != READ_ONCE(fs_devices->read_policy)) { - WRITE_ONCE(fs_devices->read_policy, i); - btrfs_info(fs_devices->fs_info, - "read policy set to '%s'", - btrfs_read_policy_name[i]); + index = btrfs_read_policy_to_enum(buf, &value); + if (index < 0) + return -EINVAL; + +#ifdef CONFIG_BTRFS_EXPERIMENTAL + /* If moving from RR then disable collecting fs stats. */ + if (fs_devices->read_policy == BTRFS_READ_POLICY_RR && index != BTRFS_READ_POLICY_RR) + fs_devices->collect_fs_stats = false; + + if (index == BTRFS_READ_POLICY_RR) { + if (value != -1) { + const u32 sectorsize = fs_devices->fs_info->sectorsize; + + if (!IS_ALIGNED(value, sectorsize)) { + u64 temp_value = round_up(value, sectorsize); + + btrfs_debug(fs_devices->fs_info, +"read_policy: min contig read %lld should be multiple of sectorsize %u, rounded to %llu", + value, sectorsize, temp_value); + value = temp_value; } - return len; + } else { + value = BTRFS_DEFAULT_RR_MIN_CONTIG_READ; + } + + if (index != READ_ONCE(fs_devices->read_policy) || + value != READ_ONCE(fs_devices->rr_min_contig_read)) { + WRITE_ONCE(fs_devices->read_policy, index); + WRITE_ONCE(fs_devices->rr_min_contig_read, value); + + btrfs_info(fs_devices->fs_info, "read policy set to '%s:%lld'", + btrfs_read_policy_name[index], value); + } + + fs_devices->collect_fs_stats = true; + + return len; + } + + if (index == BTRFS_READ_POLICY_DEVID) { + if (value != -1) { + BTRFS_DEV_LOOKUP_ARGS(args); + + /* Validate input devid. */ + args.devid = value; + if (btrfs_find_device(fs_devices, &args) == NULL) + return -EINVAL; + } else { + /* Set default devid to the devid of the latest device. */ + value = fs_devices->latest_dev->devid; } + + if (index != READ_ONCE(fs_devices->read_policy) || + value != READ_ONCE(fs_devices->read_devid)) { + WRITE_ONCE(fs_devices->read_policy, index); + WRITE_ONCE(fs_devices->read_devid, value); + + btrfs_info(fs_devices->fs_info, "read policy set to '%s:%llu'", + btrfs_read_policy_name[index], value); + } + + return len; + } +#endif + if (index != READ_ONCE(fs_devices->read_policy)) { + WRITE_ONCE(fs_devices->read_policy, index); + btrfs_info(fs_devices->fs_info, "read policy set to '%s'", + btrfs_read_policy_name[index]); } - return -EINVAL; + return len; } BTRFS_ATTR_RW(, read_policy, btrfs_read_policy_show, btrfs_read_policy_store); diff --git a/fs/btrfs/sysfs.h b/fs/btrfs/sysfs.h index e6a284c59809..3fc5c6f90dc4 100644 --- a/fs/btrfs/sysfs.h +++ b/fs/btrfs/sysfs.h @@ -47,5 +47,11 @@ void btrfs_sysfs_del_qgroups(struct btrfs_fs_info *fs_info); int btrfs_sysfs_add_qgroups(struct btrfs_fs_info *fs_info); void btrfs_sysfs_del_one_qgroup(struct btrfs_fs_info *fs_info, struct btrfs_qgroup *qgroup); +int btrfs_read_policy_to_enum(const char *str, s64 *value); + +#ifdef CONFIG_BTRFS_EXPERIMENTAL +int __init btrfs_read_policy_init(void); +char *btrfs_get_mod_read_policy(void); +#endif #endif diff --git a/fs/btrfs/tests/btrfs-tests.c b/fs/btrfs/tests/btrfs-tests.c index e607b5d52fb1..5eff8d7d2360 100644 --- a/fs/btrfs/tests/btrfs-tests.c +++ b/fs/btrfs/tests/btrfs-tests.c @@ -30,6 +30,7 @@ const char *test_error[] = { [TEST_ALLOC_EXTENT_MAP] = "cannot allocate extent map", [TEST_ALLOC_CHUNK_MAP] = "cannot allocate chunk map", [TEST_ALLOC_IO_CONTEXT] = "cannot allocate io context", + [TEST_ALLOC_TRANSACTION] = "cannot allocate transaction", }; static const struct super_operations btrfs_test_super_ops = { @@ -142,6 +143,11 @@ struct btrfs_fs_info *btrfs_alloc_dummy_fs_info(u32 nodesize, u32 sectorsize) fs_info->nodesize = nodesize; fs_info->sectorsize = sectorsize; fs_info->sectorsize_bits = ilog2(sectorsize); + + /* CRC32C csum size. */ + fs_info->csum_size = 4; + fs_info->csums_per_leaf = BTRFS_MAX_ITEM_SIZE(fs_info) / + fs_info->csum_size; set_bit(BTRFS_FS_STATE_DUMMY_FS_INFO, &fs_info->fs_state); test_mnt->mnt_sb->s_fs_info = fs_info; @@ -247,6 +253,15 @@ void btrfs_free_dummy_block_group(struct btrfs_block_group *cache) kfree(cache); } +void btrfs_init_dummy_transaction(struct btrfs_transaction *trans, struct btrfs_fs_info *fs_info) +{ + memset(trans, 0, sizeof(*trans)); + trans->fs_info = fs_info; + xa_init(&trans->delayed_refs.head_refs); + xa_init(&trans->delayed_refs.dirty_extents); + spin_lock_init(&trans->delayed_refs.lock); +} + void btrfs_init_dummy_trans(struct btrfs_trans_handle *trans, struct btrfs_fs_info *fs_info) { @@ -295,6 +310,9 @@ int btrfs_run_sanity_tests(void) ret = btrfs_test_raid_stripe_tree(sectorsize, nodesize); if (ret) goto out; + ret = btrfs_test_delayed_refs(sectorsize, nodesize); + if (ret) + goto out; } } ret = btrfs_test_extent_map(); diff --git a/fs/btrfs/tests/btrfs-tests.h b/fs/btrfs/tests/btrfs-tests.h index b524ecf2f452..4307bdaa6749 100644 --- a/fs/btrfs/tests/btrfs-tests.h +++ b/fs/btrfs/tests/btrfs-tests.h @@ -6,6 +6,8 @@ #ifndef BTRFS_TESTS_H #define BTRFS_TESTS_H +#include <linux/types.h> + #ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS int btrfs_run_sanity_tests(void); @@ -25,12 +27,14 @@ enum { TEST_ALLOC_EXTENT_MAP, TEST_ALLOC_CHUNK_MAP, TEST_ALLOC_IO_CONTEXT, + TEST_ALLOC_TRANSACTION, }; extern const char *test_error[]; struct btrfs_root; struct btrfs_trans_handle; +struct btrfs_transaction; int btrfs_test_extent_buffer_operations(u32 sectorsize, u32 nodesize); int btrfs_test_free_space_cache(u32 sectorsize, u32 nodesize); @@ -40,6 +44,7 @@ int btrfs_test_qgroups(u32 sectorsize, u32 nodesize); int btrfs_test_free_space_tree(u32 sectorsize, u32 nodesize); int btrfs_test_raid_stripe_tree(u32 sectorsize, u32 nodesize); int btrfs_test_extent_map(void); +int btrfs_test_delayed_refs(u32 sectorsize, u32 nodesize); struct inode *btrfs_new_test_inode(void); struct btrfs_fs_info *btrfs_alloc_dummy_fs_info(u32 nodesize, u32 sectorsize); void btrfs_free_dummy_fs_info(struct btrfs_fs_info *fs_info); @@ -49,6 +54,7 @@ btrfs_alloc_dummy_block_group(struct btrfs_fs_info *fs_info, unsigned long lengt void btrfs_free_dummy_block_group(struct btrfs_block_group *cache); void btrfs_init_dummy_trans(struct btrfs_trans_handle *trans, struct btrfs_fs_info *fs_info); +void btrfs_init_dummy_transaction(struct btrfs_transaction *trans, struct btrfs_fs_info *fs_info); struct btrfs_device *btrfs_alloc_dummy_device(struct btrfs_fs_info *fs_info); #else static inline int btrfs_run_sanity_tests(void) diff --git a/fs/btrfs/tests/delayed-refs-tests.c b/fs/btrfs/tests/delayed-refs-tests.c new file mode 100644 index 000000000000..6558508c2ddf --- /dev/null +++ b/fs/btrfs/tests/delayed-refs-tests.c @@ -0,0 +1,1015 @@ +// SPDX-License-Identifier: GPL-2.0 +#include <linux/sizes.h> +#include "btrfs-tests.h" +#include "../transaction.h" +#include "../delayed-ref.h" +#include "../extent-tree.h" + +#define FAKE_ROOT_OBJECTID 256 +#define FAKE_BYTENR 0 +#define FAKE_LEVEL 1 +#define FAKE_INO 256 +#define FAKE_FILE_OFFSET 0 +#define FAKE_PARENT SZ_1M + +struct ref_head_check { + u64 bytenr; + u64 num_bytes; + int ref_mod; + int total_ref_mod; + int must_insert; +}; + +struct ref_node_check { + u64 bytenr; + u64 num_bytes; + int ref_mod; + enum btrfs_delayed_ref_action action; + u8 type; + u64 parent; + u64 root; + u64 owner; + u64 offset; +}; + +static enum btrfs_ref_type ref_type_from_disk_ref_type(u8 type) +{ + if ((type == BTRFS_TREE_BLOCK_REF_KEY) || + (type == BTRFS_SHARED_BLOCK_REF_KEY)) + return BTRFS_REF_METADATA; + return BTRFS_REF_DATA; +} + +static void delete_delayed_ref_head(struct btrfs_trans_handle *trans, + struct btrfs_delayed_ref_head *head) +{ + struct btrfs_fs_info *fs_info = trans->fs_info; + struct btrfs_delayed_ref_root *delayed_refs = + &trans->transaction->delayed_refs; + + spin_lock(&delayed_refs->lock); + spin_lock(&head->lock); + btrfs_delete_ref_head(fs_info, delayed_refs, head); + spin_unlock(&head->lock); + spin_unlock(&delayed_refs->lock); + + btrfs_delayed_ref_unlock(head); + btrfs_put_delayed_ref_head(head); +} + +static void delete_delayed_ref_node(struct btrfs_delayed_ref_head *head, + struct btrfs_delayed_ref_node *node) +{ + rb_erase_cached(&node->ref_node, &head->ref_tree); + RB_CLEAR_NODE(&node->ref_node); + if (!list_empty(&node->add_list)) + list_del_init(&node->add_list); + btrfs_put_delayed_ref(node); +} + +static int validate_ref_head(struct btrfs_delayed_ref_head *head, + struct ref_head_check *check) +{ + if (head->bytenr != check->bytenr) { + test_err("invalid bytenr have: %llu want: %llu", head->bytenr, + check->bytenr); + return -EINVAL; + } + + if (head->num_bytes != check->num_bytes) { + test_err("invalid num_bytes have: %llu want: %llu", + head->num_bytes, check->num_bytes); + return -EINVAL; + } + + if (head->ref_mod != check->ref_mod) { + test_err("invalid ref_mod have: %d want: %d", head->ref_mod, + check->ref_mod); + return -EINVAL; + } + + if (head->total_ref_mod != check->total_ref_mod) { + test_err("invalid total_ref_mod have: %d want: %d", + head->total_ref_mod, check->total_ref_mod); + return -EINVAL; + } + + if (head->must_insert_reserved != check->must_insert) { + test_err("invalid must_insert have: %d want: %d", + head->must_insert_reserved, check->must_insert); + return -EINVAL; + } + + return 0; +} + +static int validate_ref_node(struct btrfs_delayed_ref_node *node, + struct ref_node_check *check) +{ + if (node->bytenr != check->bytenr) { + test_err("invalid bytenr have: %llu want: %llu", node->bytenr, + check->bytenr); + return -EINVAL; + } + + if (node->num_bytes != check->num_bytes) { + test_err("invalid num_bytes have: %llu want: %llu", + node->num_bytes, check->num_bytes); + return -EINVAL; + } + + if (node->ref_mod != check->ref_mod) { + test_err("invalid ref_mod have: %d want: %d", node->ref_mod, + check->ref_mod); + return -EINVAL; + } + + if (node->action != check->action) { + test_err("invalid action have: %d want: %d", node->action, + check->action); + return -EINVAL; + } + + if (node->parent != check->parent) { + test_err("invalid parent have: %llu want: %llu", node->parent, + check->parent); + return -EINVAL; + } + + if (node->ref_root != check->root) { + test_err("invalid root have: %llu want: %llu", node->ref_root, + check->root); + return -EINVAL; + } + + if (node->type != check->type) { + test_err("invalid type have: %d want: %d", node->type, + check->type); + return -EINVAL; + } + + if (btrfs_delayed_ref_owner(node) != check->owner) { + test_err("invalid owner have: %llu want: %llu", + btrfs_delayed_ref_owner(node), check->owner); + return -EINVAL; + } + + if (btrfs_delayed_ref_offset(node) != check->offset) { + test_err("invalid offset have: %llu want: %llu", + btrfs_delayed_ref_offset(node), check->offset); + return -EINVAL; + } + + return 0; +} + +static int simple_test(struct btrfs_trans_handle *trans, + struct ref_head_check *head_check, + struct ref_node_check *node_check) +{ + struct btrfs_delayed_ref_root *delayed_refs = + &trans->transaction->delayed_refs; + struct btrfs_fs_info *fs_info = trans->fs_info; + struct btrfs_delayed_ref_head *head; + struct btrfs_delayed_ref_node *node; + struct btrfs_ref ref = { + .type = ref_type_from_disk_ref_type(node_check->type), + .action = node_check->action, + .parent = node_check->parent, + .ref_root = node_check->root, + .bytenr = node_check->bytenr, + .num_bytes = fs_info->nodesize, + }; + int ret; + + if (ref.type == BTRFS_REF_METADATA) + btrfs_init_tree_ref(&ref, node_check->owner, node_check->root, + false); + else + btrfs_init_data_ref(&ref, node_check->owner, node_check->offset, + node_check->root, true); + + if (ref.type == BTRFS_REF_METADATA) + ret = btrfs_add_delayed_tree_ref(trans, &ref, NULL); + else + ret = btrfs_add_delayed_data_ref(trans, &ref, 0); + if (ret) { + test_err("failed ref action %d", ret); + return ret; + } + + head = btrfs_select_ref_head(fs_info, delayed_refs); + if (IS_ERR_OR_NULL(head)) { + if (IS_ERR(head)) + test_err("failed to select delayed ref head: %ld", + PTR_ERR(head)); + else + test_err("failed to find delayed ref head"); + return -EINVAL; + } + + ret = -EINVAL; + if (validate_ref_head(head, head_check)) + goto out; + + spin_lock(&head->lock); + node = btrfs_select_delayed_ref(head); + spin_unlock(&head->lock); + if (!node) { + test_err("failed to select delayed ref"); + goto out; + } + + if (validate_ref_node(node, node_check)) + goto out; + ret = 0; +out: + btrfs_unselect_ref_head(delayed_refs, head); + btrfs_destroy_delayed_refs(trans->transaction); + return ret; +} + +/* + * These are simple tests, make sure that our btrfs_ref's get turned into the + * appropriate btrfs_delayed_ref_node based on their settings and action. + */ +static int simple_tests(struct btrfs_trans_handle *trans) +{ + struct btrfs_fs_info *fs_info = trans->fs_info; + struct ref_head_check head_check = { + .bytenr = FAKE_BYTENR, + .num_bytes = fs_info->nodesize, + .ref_mod = 1, + .total_ref_mod = 1, + }; + struct ref_node_check node_check = { + .bytenr = FAKE_BYTENR, + .num_bytes = fs_info->nodesize, + .ref_mod = 1, + .action = BTRFS_ADD_DELAYED_REF, + .type = BTRFS_TREE_BLOCK_REF_KEY, + .parent = 0, + .root = FAKE_ROOT_OBJECTID, + .owner = FAKE_LEVEL, + .offset = 0, + }; + + if (simple_test(trans, &head_check, &node_check)) { + test_err("single add tree block failed"); + return -EINVAL; + } + + node_check.type = BTRFS_EXTENT_DATA_REF_KEY; + node_check.owner = FAKE_INO; + node_check.offset = FAKE_FILE_OFFSET; + + if (simple_test(trans, &head_check, &node_check)) { + test_err("single add extent data failed"); + return -EINVAL; + } + + node_check.parent = FAKE_PARENT; + node_check.type = BTRFS_SHARED_BLOCK_REF_KEY; + node_check.owner = FAKE_LEVEL; + node_check.offset = 0; + + if (simple_test(trans, &head_check, &node_check)) { + test_err("single add shared block failed"); + return -EINVAL; + } + + node_check.type = BTRFS_SHARED_DATA_REF_KEY; + node_check.owner = FAKE_INO; + node_check.offset = FAKE_FILE_OFFSET; + + if (simple_test(trans, &head_check, &node_check)) { + test_err("single add shared data failed"); + return -EINVAL; + } + + head_check.ref_mod = -1; + head_check.total_ref_mod = -1; + node_check.action = BTRFS_DROP_DELAYED_REF; + node_check.type = BTRFS_TREE_BLOCK_REF_KEY; + node_check.owner = FAKE_LEVEL; + node_check.offset = 0; + node_check.parent = 0; + + if (simple_test(trans, &head_check, &node_check)) { + test_err("single drop tree block failed"); + return -EINVAL; + } + + node_check.type = BTRFS_EXTENT_DATA_REF_KEY; + node_check.owner = FAKE_INO; + node_check.offset = FAKE_FILE_OFFSET; + + if (simple_test(trans, &head_check, &node_check)) { + test_err("single drop extent data failed"); + return -EINVAL; + } + + node_check.parent = FAKE_PARENT; + node_check.type = BTRFS_SHARED_BLOCK_REF_KEY; + node_check.owner = FAKE_LEVEL; + node_check.offset = 0; + if (simple_test(trans, &head_check, &node_check)) { + test_err("single drop shared block failed"); + return -EINVAL; + } + + node_check.type = BTRFS_SHARED_DATA_REF_KEY; + node_check.owner = FAKE_INO; + node_check.offset = FAKE_FILE_OFFSET; + if (simple_test(trans, &head_check, &node_check)) { + test_err("single drop shared data failed"); + return -EINVAL; + } + + return 0; +} + +/* + * Merge tests, validate that we do delayed ref merging properly, the ref counts + * all end up properly, and delayed refs are deleted once they're no longer + * needed. + */ +static int merge_tests(struct btrfs_trans_handle *trans, + enum btrfs_ref_type type) +{ + struct btrfs_fs_info *fs_info = trans->fs_info; + struct btrfs_delayed_ref_head *head = NULL; + struct btrfs_delayed_ref_node *node; + struct btrfs_ref ref = { + .type = type, + .action = BTRFS_ADD_DELAYED_REF, + .parent = 0, + .ref_root = FAKE_ROOT_OBJECTID, + .bytenr = FAKE_BYTENR, + .num_bytes = fs_info->nodesize, + }; + struct ref_head_check head_check = { + .bytenr = FAKE_BYTENR, + .num_bytes = fs_info->nodesize, + .ref_mod = 0, + .total_ref_mod = 0, + }; + struct ref_node_check node_check = { + .bytenr = FAKE_BYTENR, + .num_bytes = fs_info->nodesize, + .ref_mod = 2, + .action = BTRFS_ADD_DELAYED_REF, + .parent = 0, + .root = FAKE_ROOT_OBJECTID, + }; + int ret; + + /* + * First add a ref and then drop it, make sure we get a head ref with a + * 0 total ref mod and no nodes. + */ + if (type == BTRFS_REF_METADATA) { + node_check.type = BTRFS_TREE_BLOCK_REF_KEY; + node_check.owner = FAKE_LEVEL; + btrfs_init_tree_ref(&ref, FAKE_LEVEL, FAKE_ROOT_OBJECTID, false); + } else { + node_check.type = BTRFS_EXTENT_DATA_REF_KEY; + node_check.owner = FAKE_INO; + node_check.offset = FAKE_FILE_OFFSET; + btrfs_init_data_ref(&ref, FAKE_INO, FAKE_FILE_OFFSET, + FAKE_ROOT_OBJECTID, true); + } + + if (type == BTRFS_REF_METADATA) + ret = btrfs_add_delayed_tree_ref(trans, &ref, NULL); + else + ret = btrfs_add_delayed_data_ref(trans, &ref, 0); + if (ret) { + test_err("failed ref action %d", ret); + return ret; + } + + ref.action = BTRFS_DROP_DELAYED_REF; + if (type == BTRFS_REF_METADATA) + ret = btrfs_add_delayed_tree_ref(trans, &ref, NULL); + else + ret = btrfs_add_delayed_data_ref(trans, &ref, 0); + if (ret) { + test_err("failed ref action %d", ret); + goto out; + } + + head = btrfs_select_ref_head(fs_info, &trans->transaction->delayed_refs); + if (IS_ERR_OR_NULL(head)) { + if (IS_ERR(head)) + test_err("failed to select delayed ref head: %ld", + PTR_ERR(head)); + else + test_err("failed to find delayed ref head"); + goto out; + } + + ret = -EINVAL; + if (validate_ref_head(head, &head_check)) { + test_err("single add and drop failed"); + goto out; + } + + spin_lock(&head->lock); + node = btrfs_select_delayed_ref(head); + spin_unlock(&head->lock); + if (node) { + test_err("found node when none should exist"); + goto out; + } + + delete_delayed_ref_head(trans, head); + head = NULL; + + /* + * Add a ref, then add another ref, make sure we get a head ref with a + * 2 total ref mod and 1 node. + */ + ref.action = BTRFS_ADD_DELAYED_REF; + if (type == BTRFS_REF_METADATA) + ret = btrfs_add_delayed_tree_ref(trans, &ref, NULL); + else + ret = btrfs_add_delayed_data_ref(trans, &ref, 0); + if (ret) { + test_err("failed ref action %d", ret); + goto out; + } + + if (type == BTRFS_REF_METADATA) + ret = btrfs_add_delayed_tree_ref(trans, &ref, NULL); + else + ret = btrfs_add_delayed_data_ref(trans, &ref, 0); + if (ret) { + test_err("failed ref action %d", ret); + goto out; + } + + head = btrfs_select_ref_head(fs_info, &trans->transaction->delayed_refs); + if (IS_ERR_OR_NULL(head)) { + if (IS_ERR(head)) + test_err("failed to select delayed ref head: %ld", + PTR_ERR(head)); + else + test_err("failed to find delayed ref head"); + goto out; + } + + head_check.ref_mod = 2; + head_check.total_ref_mod = 2; + ret = -EINVAL; + if (validate_ref_head(head, &head_check)) { + test_err("double add failed"); + goto out; + } + + spin_lock(&head->lock); + node = btrfs_select_delayed_ref(head); + spin_unlock(&head->lock); + if (!node) { + test_err("failed to select delayed ref"); + goto out; + } + + if (validate_ref_node(node, &node_check)) { + test_err("node check failed"); + goto out; + } + + delete_delayed_ref_node(head, node); + + spin_lock(&head->lock); + node = btrfs_select_delayed_ref(head); + spin_unlock(&head->lock); + if (node) { + test_err("found node when none should exist"); + goto out; + } + delete_delayed_ref_head(trans, head); + head = NULL; + + /* Add two drop refs, make sure they are merged properly. */ + ref.action = BTRFS_DROP_DELAYED_REF; + if (type == BTRFS_REF_METADATA) + ret = btrfs_add_delayed_tree_ref(trans, &ref, NULL); + else + ret = btrfs_add_delayed_data_ref(trans, &ref, 0); + if (ret) { + test_err("failed ref action %d", ret); + goto out; + } + + if (type == BTRFS_REF_METADATA) + ret = btrfs_add_delayed_tree_ref(trans, &ref, NULL); + else + ret = btrfs_add_delayed_data_ref(trans, &ref, 0); + if (ret) { + test_err("failed ref action %d", ret); + goto out; + } + + head = btrfs_select_ref_head(fs_info, &trans->transaction->delayed_refs); + if (IS_ERR_OR_NULL(head)) { + if (IS_ERR(head)) + test_err("failed to select delayed ref head: %ld", + PTR_ERR(head)); + else + test_err("failed to find delayed ref head"); + goto out; + } + + head_check.ref_mod = -2; + head_check.total_ref_mod = -2; + ret = -EINVAL; + if (validate_ref_head(head, &head_check)) { + test_err("double drop failed"); + goto out; + } + + node_check.action = BTRFS_DROP_DELAYED_REF; + spin_lock(&head->lock); + node = btrfs_select_delayed_ref(head); + spin_unlock(&head->lock); + if (!node) { + test_err("failed to select delayed ref"); + goto out; + } + + if (validate_ref_node(node, &node_check)) { + test_err("node check failed"); + goto out; + } + + delete_delayed_ref_node(head, node); + + spin_lock(&head->lock); + node = btrfs_select_delayed_ref(head); + spin_unlock(&head->lock); + if (node) { + test_err("found node when none should exist"); + goto out; + } + delete_delayed_ref_head(trans, head); + head = NULL; + + /* Add multiple refs, then drop until we go negative again. */ + ref.action = BTRFS_ADD_DELAYED_REF; + for (int i = 0; i < 10; i++) { + if (type == BTRFS_REF_METADATA) + ret = btrfs_add_delayed_tree_ref(trans, &ref, NULL); + else + ret = btrfs_add_delayed_data_ref(trans, &ref, 0); + if (ret) { + test_err("failed ref action %d", ret); + goto out; + } + } + + ref.action = BTRFS_DROP_DELAYED_REF; + for (int i = 0; i < 12; i++) { + if (type == BTRFS_REF_METADATA) + ret = btrfs_add_delayed_tree_ref(trans, &ref, NULL); + else + ret = btrfs_add_delayed_data_ref(trans, &ref, 0); + if (ret) { + test_err("failed ref action %d", ret); + goto out; + } + } + + head = btrfs_select_ref_head(fs_info, &trans->transaction->delayed_refs); + if (IS_ERR_OR_NULL(head)) { + if (IS_ERR(head)) + test_err("failed to select delayed ref head: %ld", + PTR_ERR(head)); + else + test_err("failed to find delayed ref head"); + ret = -EINVAL; + goto out; + } + + head_check.ref_mod = -2; + head_check.total_ref_mod = -2; + ret = -EINVAL; + if (validate_ref_head(head, &head_check)) { + test_err("double drop failed"); + goto out; + } + + spin_lock(&head->lock); + node = btrfs_select_delayed_ref(head); + spin_unlock(&head->lock); + if (!node) { + test_err("failed to select delayed ref"); + goto out; + } + + if (validate_ref_node(node, &node_check)) { + test_err("node check failed"); + goto out; + } + + delete_delayed_ref_node(head, node); + + spin_lock(&head->lock); + node = btrfs_select_delayed_ref(head); + spin_unlock(&head->lock); + if (node) { + test_err("found node when none should exist"); + goto out; + } + + delete_delayed_ref_head(trans, head); + head = NULL; + + /* Drop multiple refs, then add until we go positive again. */ + ref.action = BTRFS_DROP_DELAYED_REF; + for (int i = 0; i < 10; i++) { + if (type == BTRFS_REF_METADATA) + ret = btrfs_add_delayed_tree_ref(trans, &ref, NULL); + else + ret = btrfs_add_delayed_data_ref(trans, &ref, 0); + if (ret) { + test_err("failed ref action %d", ret); + goto out; + } + } + + ref.action = BTRFS_ADD_DELAYED_REF; + for (int i = 0; i < 12; i++) { + if (type == BTRFS_REF_METADATA) + ret = btrfs_add_delayed_tree_ref(trans, &ref, NULL); + else + ret = btrfs_add_delayed_data_ref(trans, &ref, 0); + if (ret) { + test_err("failed ref action %d", ret); + goto out; + } + } + + head = btrfs_select_ref_head(fs_info, &trans->transaction->delayed_refs); + if (IS_ERR_OR_NULL(head)) { + if (IS_ERR(head)) + test_err("failed to select delayed ref head: %ld", + PTR_ERR(head)); + else + test_err("failed to find delayed ref head"); + ret = -EINVAL; + goto out; + } + + head_check.ref_mod = 2; + head_check.total_ref_mod = 2; + ret = -EINVAL; + if (validate_ref_head(head, &head_check)) { + test_err("add and drop to positive failed"); + goto out; + } + + node_check.action = BTRFS_ADD_DELAYED_REF; + spin_lock(&head->lock); + node = btrfs_select_delayed_ref(head); + spin_unlock(&head->lock); + if (!node) { + test_err("failed to select delayed ref"); + goto out; + } + + if (validate_ref_node(node, &node_check)) { + test_err("node check failed"); + goto out; + } + + delete_delayed_ref_node(head, node); + + spin_lock(&head->lock); + node = btrfs_select_delayed_ref(head); + spin_unlock(&head->lock); + if (node) { + test_err("found node when none should exist"); + goto out; + } + delete_delayed_ref_head(trans, head); + head = NULL; + + /* + * Add a bunch of refs with different roots and parents, then drop them + * all, make sure everything is properly merged. + */ + ref.action = BTRFS_ADD_DELAYED_REF; + for (int i = 0; i < 50; i++) { + if (!(i % 2)) { + ref.parent = 0; + ref.ref_root = FAKE_ROOT_OBJECTID + i; + } else { + ref.parent = FAKE_PARENT + (i * fs_info->nodesize); + } + if (type == BTRFS_REF_METADATA) + ret = btrfs_add_delayed_tree_ref(trans, &ref, NULL); + else + ret = btrfs_add_delayed_data_ref(trans, &ref, 0); + if (ret) { + test_err("failed ref action %d", ret); + goto out; + } + } + + ref.action = BTRFS_DROP_DELAYED_REF; + for (int i = 0; i < 50; i++) { + if (!(i % 2)) { + ref.parent = 0; + ref.ref_root = FAKE_ROOT_OBJECTID + i; + } else { + ref.parent = FAKE_PARENT + (i * fs_info->nodesize); + } + if (type == BTRFS_REF_METADATA) + ret = btrfs_add_delayed_tree_ref(trans, &ref, NULL); + else + ret = btrfs_add_delayed_data_ref(trans, &ref, 0); + if (ret) { + test_err("failed ref action %d", ret); + goto out; + } + } + + head = btrfs_select_ref_head(fs_info, &trans->transaction->delayed_refs); + if (IS_ERR_OR_NULL(head)) { + if (IS_ERR(head)) + test_err("failed to select delayed ref head: %ld", + PTR_ERR(head)); + else + test_err("failed to find delayed ref head"); + ret = -EINVAL; + goto out; + } + + head_check.ref_mod = 0; + head_check.total_ref_mod = 0; + ret = -EINVAL; + if (validate_ref_head(head, &head_check)) { + test_err("add and drop multiple failed"); + goto out; + } + + spin_lock(&head->lock); + node = btrfs_select_delayed_ref(head); + spin_unlock(&head->lock); + if (node) { + test_err("found node when none should exist"); + goto out; + } + ret = 0; +out: + if (!IS_ERR_OR_NULL(head)) + btrfs_unselect_ref_head(&trans->transaction->delayed_refs, head); + btrfs_destroy_delayed_refs(trans->transaction); + return ret; +} + +/* + * Basic test to validate we always get the add operations first followed by any + * delete operations. + */ +static int select_delayed_refs_test(struct btrfs_trans_handle *trans) +{ + struct btrfs_delayed_ref_root *delayed_refs = + &trans->transaction->delayed_refs; + struct btrfs_fs_info *fs_info = trans->fs_info; + struct btrfs_delayed_ref_head *head = NULL; + struct btrfs_delayed_ref_node *node; + struct btrfs_ref ref = { + .type = BTRFS_REF_METADATA, + .action = BTRFS_DROP_DELAYED_REF, + .parent = 0, + .ref_root = FAKE_ROOT_OBJECTID, + .bytenr = FAKE_BYTENR, + .num_bytes = fs_info->nodesize, + }; + struct ref_head_check head_check = { + .bytenr = FAKE_BYTENR, + .num_bytes = fs_info->nodesize, + .ref_mod = 0, + .total_ref_mod = 0, + }; + struct ref_node_check node_check = { + .bytenr = FAKE_BYTENR, + .num_bytes = fs_info->nodesize, + .ref_mod = 1, + .action = BTRFS_ADD_DELAYED_REF, + .type = BTRFS_TREE_BLOCK_REF_KEY, + .parent = 0, + .owner = FAKE_LEVEL, + .offset = 0, + }; + int ret; + + /* Add the drop first. */ + btrfs_init_tree_ref(&ref, FAKE_LEVEL, FAKE_ROOT_OBJECTID, false); + ret = btrfs_add_delayed_tree_ref(trans, &ref, NULL); + if (ret) { + test_err("failed ref action %d", ret); + return ret; + } + + /* + * Now add the add, and make it a different root so it's logically later + * in the rb tree. + */ + ref.action = BTRFS_ADD_DELAYED_REF; + ref.ref_root = FAKE_ROOT_OBJECTID + 1; + ret = btrfs_add_delayed_tree_ref(trans, &ref, NULL); + if (ret) { + test_err("failed ref action %d", ret); + goto out; + } + + head = btrfs_select_ref_head(fs_info, delayed_refs); + if (IS_ERR_OR_NULL(head)) { + if (IS_ERR(head)) + test_err("failed to select delayed ref head: %ld", + PTR_ERR(head)); + else + test_err("failed to find delayed ref head"); + ret = -EINVAL; + head = NULL; + goto out; + } + + ret = -EINVAL; + if (validate_ref_head(head, &head_check)) { + test_err("head check failed"); + goto out; + } + + spin_lock(&head->lock); + node = btrfs_select_delayed_ref(head); + spin_unlock(&head->lock); + if (!node) { + test_err("failed to select delayed ref"); + goto out; + } + + node_check.root = FAKE_ROOT_OBJECTID + 1; + if (validate_ref_node(node, &node_check)) { + test_err("node check failed"); + goto out; + } + delete_delayed_ref_node(head, node); + + spin_lock(&head->lock); + node = btrfs_select_delayed_ref(head); + spin_unlock(&head->lock); + if (!node) { + test_err("failed to select delayed ref"); + goto out; + } + + node_check.action = BTRFS_DROP_DELAYED_REF; + node_check.root = FAKE_ROOT_OBJECTID; + if (validate_ref_node(node, &node_check)) { + test_err("node check failed"); + goto out; + } + delete_delayed_ref_node(head, node); + delete_delayed_ref_head(trans, head); + head = NULL; + + /* + * Now we're going to do the same thing, but we're going to have an add + * that gets deleted because of a merge, and make sure we still have + * another add in place. + */ + ref.action = BTRFS_DROP_DELAYED_REF; + ref.ref_root = FAKE_ROOT_OBJECTID; + ret = btrfs_add_delayed_tree_ref(trans, &ref, NULL); + if (ret) { + test_err("failed ref action %d", ret); + goto out; + } + + ref.action = BTRFS_ADD_DELAYED_REF; + ref.ref_root = FAKE_ROOT_OBJECTID + 1; + ret = btrfs_add_delayed_tree_ref(trans, &ref, NULL); + if (ret) { + test_err("failed ref action %d", ret); + goto out; + } + + ref.action = BTRFS_DROP_DELAYED_REF; + ret = btrfs_add_delayed_tree_ref(trans, &ref, NULL); + if (ret) { + test_err("failed ref action %d", ret); + goto out; + } + + ref.action = BTRFS_ADD_DELAYED_REF; + ref.ref_root = FAKE_ROOT_OBJECTID + 2; + ret = btrfs_add_delayed_tree_ref(trans, &ref, NULL); + if (ret) { + test_err("failed ref action %d", ret); + goto out; + } + + head = btrfs_select_ref_head(fs_info, delayed_refs); + if (IS_ERR_OR_NULL(head)) { + if (IS_ERR(head)) + test_err("failed to select delayed ref head: %ld", + PTR_ERR(head)); + else + test_err("failed to find delayed ref head"); + ret = -EINVAL; + head = NULL; + goto out; + } + + ret = -EINVAL; + if (validate_ref_head(head, &head_check)) { + test_err("head check failed"); + goto out; + } + + spin_lock(&head->lock); + node = btrfs_select_delayed_ref(head); + spin_unlock(&head->lock); + if (!node) { + test_err("failed to select delayed ref"); + goto out; + } + + node_check.action = BTRFS_ADD_DELAYED_REF; + node_check.root = FAKE_ROOT_OBJECTID + 2; + if (validate_ref_node(node, &node_check)) { + test_err("node check failed"); + goto out; + } + delete_delayed_ref_node(head, node); + + spin_lock(&head->lock); + node = btrfs_select_delayed_ref(head); + spin_unlock(&head->lock); + if (!node) { + test_err("failed to select delayed ref"); + goto out; + } + + node_check.action = BTRFS_DROP_DELAYED_REF; + node_check.root = FAKE_ROOT_OBJECTID; + if (validate_ref_node(node, &node_check)) { + test_err("node check failed"); + goto out; + } + delete_delayed_ref_node(head, node); + ret = 0; +out: + if (head) + btrfs_unselect_ref_head(delayed_refs, head); + btrfs_destroy_delayed_refs(trans->transaction); + return ret; +} + +int btrfs_test_delayed_refs(u32 sectorsize, u32 nodesize) +{ + struct btrfs_transaction *transaction; + struct btrfs_trans_handle trans; + struct btrfs_fs_info *fs_info; + int ret; + + test_msg("running delayed refs tests"); + + fs_info = btrfs_alloc_dummy_fs_info(nodesize, sectorsize); + if (!fs_info) { + test_std_err(TEST_ALLOC_FS_INFO); + return -ENOMEM; + } + transaction = kmalloc(sizeof(*transaction), GFP_KERNEL); + if (!transaction) { + test_std_err(TEST_ALLOC_TRANSACTION); + ret = -ENOMEM; + goto out_free_fs_info; + } + btrfs_init_dummy_trans(&trans, fs_info); + btrfs_init_dummy_transaction(transaction, fs_info); + trans.transaction = transaction; + + ret = simple_tests(&trans); + if (!ret) { + test_msg("running delayed refs merg tests on metadata refs"); + ret = merge_tests(&trans, BTRFS_REF_METADATA); + } + + if (!ret) { + test_msg("running delayed refs merg tests on data refs"); + ret = merge_tests(&trans, BTRFS_REF_DATA); + } + + if (!ret) + ret = select_delayed_refs_test(&trans); + +out_free_fs_info: + btrfs_free_dummy_fs_info(fs_info); + return ret; +} diff --git a/fs/btrfs/tests/raid-stripe-tree-tests.c b/fs/btrfs/tests/raid-stripe-tree-tests.c index 30f17eb7b6a8..a7bc58a5c1e2 100644 --- a/fs/btrfs/tests/raid-stripe-tree-tests.c +++ b/fs/btrfs/tests/raid-stripe-tree-tests.c @@ -14,6 +14,8 @@ #define RST_TEST_NUM_DEVICES (2) #define RST_TEST_RAID1_TYPE (BTRFS_BLOCK_GROUP_DATA | BTRFS_BLOCK_GROUP_RAID1) +#define SZ_48K (SZ_32K + SZ_16K) + typedef int (*test_func_t)(struct btrfs_trans_handle *trans); static struct btrfs_device *btrfs_device_by_devid(struct btrfs_fs_devices *fs_devices, @@ -30,6 +32,613 @@ static struct btrfs_device *btrfs_device_by_devid(struct btrfs_fs_devices *fs_de } /* + * Test creating a range of three extents and then punch a hole in the middle, + * deleting all of the middle extents and partially deleting the "book ends". + */ +static int test_punch_hole_3extents(struct btrfs_trans_handle *trans) +{ + struct btrfs_fs_info *fs_info = trans->fs_info; + struct btrfs_io_context *bioc; + struct btrfs_io_stripe io_stripe = { 0 }; + u64 map_type = RST_TEST_RAID1_TYPE; + u64 logical1 = SZ_1M; + u64 len1 = SZ_1M; + u64 logical2 = logical1 + len1; + u64 len2 = SZ_1M; + u64 logical3 = logical2 + len2; + u64 len3 = SZ_1M; + u64 hole_start = logical1 + SZ_256K; + u64 hole_len = SZ_2M; + int ret; + + bioc = alloc_btrfs_io_context(fs_info, logical1, RST_TEST_NUM_DEVICES); + if (!bioc) { + test_std_err(TEST_ALLOC_IO_CONTEXT); + ret = -ENOMEM; + goto out; + } + + io_stripe.dev = btrfs_device_by_devid(fs_info->fs_devices, 0); + + /* Prepare for the test, 1st create 3 x 1M extents. */ + bioc->map_type = map_type; + bioc->size = len1; + + for (int i = 0; i < RST_TEST_NUM_DEVICES; i++) { + struct btrfs_io_stripe *stripe = &bioc->stripes[i]; + + stripe->dev = btrfs_device_by_devid(fs_info->fs_devices, i); + if (!stripe->dev) { + test_err("cannot find device with devid %d", i); + ret = -EINVAL; + goto out; + } + + stripe->physical = logical1 + i * SZ_1G; + } + + ret = btrfs_insert_one_raid_extent(trans, bioc); + if (ret) { + test_err("inserting RAID extent failed: %d", ret); + goto out; + } + + bioc->logical = logical2; + bioc->size = len2; + for (int i = 0; i < RST_TEST_NUM_DEVICES; i++) { + struct btrfs_io_stripe *stripe = &bioc->stripes[i]; + + stripe->dev = btrfs_device_by_devid(fs_info->fs_devices, i); + if (!stripe->dev) { + test_err("cannot find device with devid %d", i); + ret = -EINVAL; + goto out; + } + + stripe->physical = logical2 + i * SZ_1G; + } + + ret = btrfs_insert_one_raid_extent(trans, bioc); + if (ret) { + test_err("inserting RAID extent failed: %d", ret); + goto out; + } + + bioc->logical = logical3; + bioc->size = len3; + for (int i = 0; i < RST_TEST_NUM_DEVICES; i++) { + struct btrfs_io_stripe *stripe = &bioc->stripes[i]; + + stripe->dev = btrfs_device_by_devid(fs_info->fs_devices, i); + if (!stripe->dev) { + test_err("cannot find device with devid %d", i); + ret = -EINVAL; + goto out; + } + + stripe->physical = logical3 + i * SZ_1G; + } + + ret = btrfs_insert_one_raid_extent(trans, bioc); + if (ret) { + test_err("inserting RAID extent failed: %d", ret); + goto out; + } + + /* + * Delete a range starting at logical1 + 256K and 2M in length. Extent + * 1 is truncated to 256k length, extent 2 is completely dropped and + * extent 3 is moved 256K to the right. + */ + ret = btrfs_delete_raid_extent(trans, hole_start, hole_len); + if (ret) { + test_err("deleting RAID extent [%llu, %llu] failed", + hole_start, hole_start + hole_len); + goto out; + } + + /* Get the first extent and check its size. */ + ret = btrfs_get_raid_extent_offset(fs_info, logical1, &len1, map_type, + 0, &io_stripe); + if (ret) { + test_err("lookup of RAID extent [%llu, %llu] failed", + logical1, logical1 + len1); + goto out; + } + + if (io_stripe.physical != logical1) { + test_err("invalid physical address, expected %llu, got %llu", + logical1, io_stripe.physical); + ret = -EINVAL; + goto out; + } + + if (len1 != SZ_256K) { + test_err("invalid stripe length, expected %llu, got %llu", + (u64)SZ_256K, len1); + ret = -EINVAL; + goto out; + } + + /* Get the second extent and check it's absent. */ + ret = btrfs_get_raid_extent_offset(fs_info, logical2, &len2, map_type, + 0, &io_stripe); + if (ret != -ENODATA) { + test_err("lookup of RAID extent [%llu, %llu] succeeded should fail", + logical2, logical2 + len2); + ret = -EINVAL; + goto out; + } + + /* Get the third extent and check its size. */ + logical3 += SZ_256K; + ret = btrfs_get_raid_extent_offset(fs_info, logical3, &len3, map_type, + 0, &io_stripe); + if (ret) { + test_err("lookup of RAID extent [%llu, %llu] failed", + logical3, logical3 + len3); + goto out; + } + + if (io_stripe.physical != logical3) { + test_err("invalid physical address, expected %llu, got %llu", + logical3 + SZ_256K, io_stripe.physical); + ret = -EINVAL; + goto out; + } + + if (len3 != SZ_1M - SZ_256K) { + test_err("invalid stripe length, expected %llu, got %llu", + (u64)SZ_1M - SZ_256K, len3); + ret = -EINVAL; + goto out; + } + + ret = btrfs_delete_raid_extent(trans, logical1, len1); + if (ret) { + test_err("deleting RAID extent [%llu, %llu] failed", + logical1, logical1 + len1); + goto out; + } + + ret = btrfs_delete_raid_extent(trans, logical3, len3); + if (ret) { + test_err("deleting RAID extent [%llu, %llu] failed", + logical1, logical1 + len1); + goto out; + } + +out: + btrfs_put_bioc(bioc); + return ret; +} + +static int test_delete_two_extents(struct btrfs_trans_handle *trans) +{ + struct btrfs_fs_info *fs_info = trans->fs_info; + struct btrfs_io_context *bioc; + struct btrfs_io_stripe io_stripe = { 0 }; + u64 map_type = RST_TEST_RAID1_TYPE; + u64 logical1 = SZ_1M; + u64 len1 = SZ_1M; + u64 logical2 = logical1 + len1; + u64 len2 = SZ_1M; + u64 logical3 = logical2 + len2; + u64 len3 = SZ_1M; + int ret; + + bioc = alloc_btrfs_io_context(fs_info, logical1, RST_TEST_NUM_DEVICES); + if (!bioc) { + test_std_err(TEST_ALLOC_IO_CONTEXT); + ret = -ENOMEM; + goto out; + } + + io_stripe.dev = btrfs_device_by_devid(fs_info->fs_devices, 0); + + /* Prepare for the test, 1st create 3 x 1M extents. */ + bioc->map_type = map_type; + bioc->size = len1; + + for (int i = 0; i < RST_TEST_NUM_DEVICES; i++) { + struct btrfs_io_stripe *stripe = &bioc->stripes[i]; + + stripe->dev = btrfs_device_by_devid(fs_info->fs_devices, i); + if (!stripe->dev) { + test_err("cannot find device with devid %d", i); + ret = -EINVAL; + goto out; + } + + stripe->physical = logical1 + i * SZ_1G; + } + + ret = btrfs_insert_one_raid_extent(trans, bioc); + if (ret) { + test_err("inserting RAID extent failed: %d", ret); + goto out; + } + + bioc->logical = logical2; + bioc->size = len2; + for (int i = 0; i < RST_TEST_NUM_DEVICES; i++) { + struct btrfs_io_stripe *stripe = &bioc->stripes[i]; + + stripe->dev = btrfs_device_by_devid(fs_info->fs_devices, i); + if (!stripe->dev) { + test_err("cannot find device with devid %d", i); + ret = -EINVAL; + goto out; + } + + stripe->physical = logical2 + i * SZ_1G; + } + + ret = btrfs_insert_one_raid_extent(trans, bioc); + if (ret) { + test_err("inserting RAID extent failed: %d", ret); + goto out; + } + + bioc->logical = logical3; + bioc->size = len3; + for (int i = 0; i < RST_TEST_NUM_DEVICES; i++) { + struct btrfs_io_stripe *stripe = &bioc->stripes[i]; + + stripe->dev = btrfs_device_by_devid(fs_info->fs_devices, i); + if (!stripe->dev) { + test_err("cannot find device with devid %d", i); + ret = -EINVAL; + goto out; + } + + stripe->physical = logical3 + i * SZ_1G; + } + + ret = btrfs_insert_one_raid_extent(trans, bioc); + if (ret) { + test_err("inserting RAID extent failed: %d", ret); + goto out; + } + + /* + * Delete a range starting at logical1 and 2M in length. Extents 1 + * and 2 are dropped and extent 3 is kept as is. + */ + ret = btrfs_delete_raid_extent(trans, logical1, len1 + len2); + if (ret) { + test_err("deleting RAID extent [%llu, %llu] failed", + logical1, logical1 + len1 + len2); + goto out; + } + + ret = btrfs_get_raid_extent_offset(fs_info, logical1, &len1, map_type, + 0, &io_stripe); + if (ret != -ENODATA) { + test_err("lookup of RAID extent [%llu, %llu] succeeded, should fail", + logical1, len1); + goto out; + } + + ret = btrfs_get_raid_extent_offset(fs_info, logical2, &len2, map_type, + 0, &io_stripe); + if (ret != -ENODATA) { + test_err("lookup of RAID extent [%llu, %llu] succeeded, should fail", + logical2, len2); + goto out; + } + + ret = btrfs_get_raid_extent_offset(fs_info, logical3, &len3, map_type, + 0, &io_stripe); + if (ret) { + test_err("lookup of RAID extent [%llu, %llu] failed", + logical3, len3); + goto out; + } + + if (io_stripe.physical != logical3) { + test_err("invalid physical address, expected %llu, got %llu", + logical3, io_stripe.physical); + ret = -EINVAL; + goto out; + } + + if (len3 != SZ_1M) { + test_err("invalid stripe length, expected %llu, got %llu", + (u64)SZ_1M, len3); + ret = -EINVAL; + goto out; + } + + ret = btrfs_delete_raid_extent(trans, logical3, len3); +out: + btrfs_put_bioc(bioc); + return ret; +} + +/* Test punching a hole into a single RAID stripe-extent. */ +static int test_punch_hole(struct btrfs_trans_handle *trans) +{ + struct btrfs_fs_info *fs_info = trans->fs_info; + struct btrfs_io_context *bioc; + struct btrfs_io_stripe io_stripe = { 0 }; + u64 map_type = RST_TEST_RAID1_TYPE; + u64 logical1 = SZ_1M; + u64 hole_start = logical1 + SZ_32K; + u64 hole_len = SZ_64K; + u64 logical2 = hole_start + hole_len; + u64 len = SZ_1M; + u64 len1 = SZ_32K; + u64 len2 = len - len1 - hole_len; + int ret; + + bioc = alloc_btrfs_io_context(fs_info, logical1, RST_TEST_NUM_DEVICES); + if (!bioc) { + test_std_err(TEST_ALLOC_IO_CONTEXT); + ret = -ENOMEM; + goto out; + } + + io_stripe.dev = btrfs_device_by_devid(fs_info->fs_devices, 0); + bioc->map_type = map_type; + bioc->size = len; + + for (int i = 0; i < RST_TEST_NUM_DEVICES; i++) { + struct btrfs_io_stripe *stripe = &bioc->stripes[i]; + + stripe->dev = btrfs_device_by_devid(fs_info->fs_devices, i); + if (!stripe->dev) { + test_err("cannot find device with devid %d", i); + ret = -EINVAL; + goto out; + } + + stripe->physical = logical1 + i * SZ_1G; + } + + ret = btrfs_insert_one_raid_extent(trans, bioc); + if (ret) { + test_err("inserting RAID extent failed: %d", ret); + goto out; + } + + ret = btrfs_get_raid_extent_offset(fs_info, logical1, &len, map_type, 0, + &io_stripe); + if (ret) { + test_err("lookup of RAID extent [%llu, %llu] failed", logical1, + logical1 + len); + goto out; + } + + if (io_stripe.physical != logical1) { + test_err("invalid physical address, expected %llu got %llu", + logical1, io_stripe.physical); + ret = -EINVAL; + goto out; + } + + if (len != SZ_1M) { + test_err("invalid stripe length, expected %llu got %llu", + (u64)SZ_1M, len); + ret = -EINVAL; + goto out; + } + + ret = btrfs_delete_raid_extent(trans, hole_start, hole_len); + if (ret) { + test_err("deleting RAID extent [%llu, %llu] failed", + hole_start, hole_start + hole_len); + goto out; + } + + ret = btrfs_get_raid_extent_offset(fs_info, logical1, &len1, map_type, + 0, &io_stripe); + if (ret) { + test_err("lookup of RAID extent [%llu, %llu] failed", + logical1, logical1 + len1); + goto out; + } + + if (io_stripe.physical != logical1) { + test_err("invalid physical address, expected %llu, got %llu", + logical1, io_stripe.physical); + ret = -EINVAL; + goto out; + } + + if (len1 != SZ_32K) { + test_err("invalid stripe length, expected %llu, got %llu", + (u64)SZ_32K, len1); + ret = -EINVAL; + goto out; + } + + ret = btrfs_get_raid_extent_offset(fs_info, logical2, &len2, map_type, + 0, &io_stripe); + if (ret) { + test_err("lookup of RAID extent [%llu, %llu] failed", logical2, + logical2 + len2); + goto out; + } + + if (io_stripe.physical != logical2) { + test_err("invalid physical address, expected %llu, got %llu", + logical2, io_stripe.physical); + ret = -EINVAL; + goto out; + } + + if (len2 != len - len1 - hole_len) { + test_err("invalid length, expected %llu, got %llu", + len - len1 - hole_len, len2); + ret = -EINVAL; + goto out; + } + + /* Check for the absence of the hole. */ + ret = btrfs_get_raid_extent_offset(fs_info, hole_start, &hole_len, + map_type, 0, &io_stripe); + if (ret != -ENODATA) { + ret = -EINVAL; + test_err("lookup of RAID extent [%llu, %llu] succeeded, should fail", + hole_start, hole_start + SZ_64K); + goto out; + } + + ret = btrfs_delete_raid_extent(trans, logical1, len1); + if (ret) + goto out; + + ret = btrfs_delete_raid_extent(trans, logical2, len2); +out: + btrfs_put_bioc(bioc); + return ret; +} + +/* + * Test a 1M RST write that spans two adjacent RST items on disk and then + * delete a portion starting in the first item and spanning into the second + * item. This is similar to test_front_delete(), but spanning multiple items. + */ +static int test_front_delete_prev_item(struct btrfs_trans_handle *trans) +{ + struct btrfs_fs_info *fs_info = trans->fs_info; + struct btrfs_io_context *bioc; + struct btrfs_io_stripe io_stripe = { 0 }; + u64 map_type = RST_TEST_RAID1_TYPE; + u64 logical1 = SZ_1M; + u64 logical2 = SZ_2M; + u64 len = SZ_1M; + int ret; + + bioc = alloc_btrfs_io_context(fs_info, logical1, RST_TEST_NUM_DEVICES); + if (!bioc) { + test_std_err(TEST_ALLOC_IO_CONTEXT); + ret = -ENOMEM; + goto out; + } + + io_stripe.dev = btrfs_device_by_devid(fs_info->fs_devices, 0); + bioc->map_type = map_type; + bioc->size = len; + + /* Insert RAID extent 1. */ + for (int i = 0; i < RST_TEST_NUM_DEVICES; i++) { + struct btrfs_io_stripe *stripe = &bioc->stripes[i]; + + stripe->dev = btrfs_device_by_devid(fs_info->fs_devices, i); + if (!stripe->dev) { + test_err("cannot find device with devid %d", i); + ret = -EINVAL; + goto out; + } + + stripe->physical = logical1 + i * SZ_1G; + } + + ret = btrfs_insert_one_raid_extent(trans, bioc); + if (ret) { + test_err("inserting RAID extent failed: %d", ret); + goto out; + } + + bioc->logical = logical2; + /* Insert RAID extent 2, directly adjacent to it. */ + for (int i = 0; i < RST_TEST_NUM_DEVICES; i++) { + struct btrfs_io_stripe *stripe = &bioc->stripes[i]; + + stripe->dev = btrfs_device_by_devid(fs_info->fs_devices, i); + if (!stripe->dev) { + test_err("cannot find device with devid %d", i); + ret = -EINVAL; + goto out; + } + + stripe->physical = logical2 + i * SZ_1G; + } + + ret = btrfs_insert_one_raid_extent(trans, bioc); + if (ret) { + test_err("inserting RAID extent failed: %d", ret); + goto out; + } + + ret = btrfs_delete_raid_extent(trans, logical1 + SZ_512K, SZ_1M); + if (ret) { + test_err("deleting RAID extent [%llu, %llu] failed", + logical1 + SZ_512K, (u64)SZ_1M); + goto out; + } + + /* Verify item 1 is truncated to 512K. */ + ret = btrfs_get_raid_extent_offset(fs_info, logical1, &len, map_type, 0, + &io_stripe); + if (ret) { + test_err("lookup of RAID extent [%llu, %llu] failed", logical1, + logical1 + len); + goto out; + } + + if (io_stripe.physical != logical1) { + test_err("invalid physical address, expected %llu got %llu", + logical1, io_stripe.physical); + ret = -EINVAL; + goto out; + } + + if (len != SZ_512K) { + test_err("invalid stripe length, expected %llu got %llu", + (u64)SZ_512K, len); + ret = -EINVAL; + goto out; + } + + /* Verify item 2's start is moved by 512K. */ + ret = btrfs_get_raid_extent_offset(fs_info, logical2 + SZ_512K, &len, + map_type, 0, &io_stripe); + if (ret) { + test_err("lookup of RAID extent [%llu, %llu] failed", + logical2 + SZ_512K, logical2 + len); + goto out; + } + + if (io_stripe.physical != logical2 + SZ_512K) { + test_err("invalid physical address, expected %llu got %llu", + logical2 + SZ_512K, io_stripe.physical); + ret = -EINVAL; + goto out; + } + + if (len != SZ_512K) { + test_err("invalid stripe length, expected %llu got %llu", + (u64)SZ_512K, len); + ret = -EINVAL; + goto out; + } + + /* Verify there's a hole at [1M+512K, 2M+512K] . */ + len = SZ_1M; + ret = btrfs_get_raid_extent_offset(fs_info, logical1 + SZ_512K, &len, + map_type, 0, &io_stripe); + if (ret != -ENODATA) { + test_err("lookup of RAID [%llu, %llu] succeeded, should fail", + logical1 + SZ_512K, logical1 + SZ_512K + len); + goto out; + } + + /* Clean up after us. */ + ret = btrfs_delete_raid_extent(trans, logical1, SZ_512K); + if (ret) + goto out; + + ret = btrfs_delete_raid_extent(trans, logical2 + SZ_512K, SZ_512K); + +out: + btrfs_put_bioc(bioc); + return ret; +} + +/* * Test a 64K RST write on a 2 disk RAID1 at a logical address of 1M and then * delete the 1st 32K, making the new start address 1M+32K. */ @@ -94,45 +703,45 @@ static int test_front_delete(struct btrfs_trans_handle *trans) goto out; } - ret = btrfs_delete_raid_extent(trans, logical, SZ_32K); + ret = btrfs_delete_raid_extent(trans, logical, SZ_16K); if (ret) { test_err("deleting RAID extent [%llu, %llu] failed", logical, - logical + SZ_32K); + logical + SZ_16K); goto out; } - len = SZ_32K; - ret = btrfs_get_raid_extent_offset(fs_info, logical + SZ_32K, &len, + len -= SZ_16K; + ret = btrfs_get_raid_extent_offset(fs_info, logical + SZ_16K, &len, map_type, 0, &io_stripe); if (ret) { test_err("lookup of RAID extent [%llu, %llu] failed", - logical + SZ_32K, logical + SZ_32K + len); + logical + SZ_16K, logical + SZ_64K); goto out; } - if (io_stripe.physical != logical + SZ_32K) { + if (io_stripe.physical != logical + SZ_16K) { test_err("invalid physical address, expected %llu, got %llu", - logical + SZ_32K, io_stripe.physical); + logical + SZ_16K, io_stripe.physical); ret = -EINVAL; goto out; } - if (len != SZ_32K) { + if (len != SZ_48K) { test_err("invalid stripe length, expected %llu, got %llu", - (u64)SZ_32K, len); + (u64)SZ_48K, len); ret = -EINVAL; goto out; } ret = btrfs_get_raid_extent_offset(fs_info, logical, &len, map_type, 0, &io_stripe); - if (!ret) { + if (ret != -ENODATA) { ret = -EINVAL; test_err("lookup of RAID extent [%llu, %llu] succeeded, should fail", - logical, logical + SZ_32K); + logical, logical + SZ_16K); goto out; } - ret = btrfs_delete_raid_extent(trans, logical + SZ_32K, SZ_32K); + ret = btrfs_delete_raid_extent(trans, logical + SZ_16K, SZ_48K); out: btrfs_put_bioc(bioc); return ret; @@ -209,14 +818,14 @@ static int test_tail_delete(struct btrfs_trans_handle *trans) goto out; } - ret = btrfs_delete_raid_extent(trans, logical + SZ_32K, SZ_32K); + ret = btrfs_delete_raid_extent(trans, logical + SZ_48K, SZ_16K); if (ret) { test_err("deleting RAID extent [%llu, %llu] failed", - logical + SZ_32K, logical + SZ_64K); + logical + SZ_48K, logical + SZ_64K); goto out; } - len = SZ_32K; + len = SZ_48K; ret = btrfs_get_raid_extent_offset(fs_info, logical, &len, map_type, 0, &io_stripe); if (ret) { test_err("lookup of RAID extent [%llu, %llu] failed", logical, @@ -231,9 +840,19 @@ static int test_tail_delete(struct btrfs_trans_handle *trans) goto out; } - if (len != SZ_32K) { + if (len != SZ_48K) { test_err("invalid stripe length, expected %llu, got %llu", - (u64)SZ_32K, len); + (u64)SZ_48K, len); + ret = -EINVAL; + goto out; + } + + len = SZ_16K; + ret = btrfs_get_raid_extent_offset(fs_info, logical + SZ_48K, &len, + map_type, 0, &io_stripe); + if (ret != -ENODATA) { + test_err("lookup of RAID extent [%llu, %llu] succeeded should fail", + logical + SZ_48K, logical + SZ_64K); ret = -EINVAL; goto out; } @@ -456,6 +1075,10 @@ static const test_func_t tests[] = { test_create_update_delete, test_tail_delete, test_front_delete, + test_front_delete_prev_item, + test_punch_hole, + test_punch_hole_3extents, + test_delete_two_extents, }; static int run_test(test_func_t test, u32 sectorsize, u32 nodesize) @@ -478,8 +1101,8 @@ static int run_test(test_func_t test, u32 sectorsize, u32 nodesize) ret = PTR_ERR(root); goto out; } - btrfs_set_super_compat_ro_flags(root->fs_info->super_copy, - BTRFS_FEATURE_INCOMPAT_RAID_STRIPE_TREE); + btrfs_set_super_incompat_flags(root->fs_info->super_copy, + BTRFS_FEATURE_INCOMPAT_RAID_STRIPE_TREE); root->root_key.objectid = BTRFS_RAID_STRIPE_TREE_OBJECTID; root->root_key.type = BTRFS_ROOT_ITEM_KEY; root->root_key.offset = 0; diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c index dc0b837efd5d..15312013f2a3 100644 --- a/fs/btrfs/transaction.c +++ b/fs/btrfs/transaction.c @@ -795,8 +795,7 @@ alloc_fail: if (num_bytes) btrfs_block_rsv_release(fs_info, trans_rsv, num_bytes, NULL); if (delayed_refs_bytes) - btrfs_space_info_free_bytes_may_use(fs_info, trans_rsv->space_info, - delayed_refs_bytes); + btrfs_space_info_free_bytes_may_use(trans_rsv->space_info, delayed_refs_bytes); reserve_fail: btrfs_qgroup_free_meta_prealloc(root, qgroup_reserved); return ERR_PTR(ret); diff --git a/fs/btrfs/transaction.h b/fs/btrfs/transaction.h index 184fa5c0062a..9f7c777af635 100644 --- a/fs/btrfs/transaction.h +++ b/fs/btrfs/transaction.h @@ -227,7 +227,21 @@ static inline void btrfs_clear_skip_qgroup(struct btrfs_trans_handle *trans) delayed_refs->qgroup_to_skip = 0; } -bool __cold abort_should_print_stack(int error); +/* + * We want the transaction abort to print stack trace only for errors where the + * cause could be a bug, eg. due to ENOSPC, and not for common errors that are + * caused by external factors. + */ +static inline bool btrfs_abort_should_print_stack(int error) +{ + switch (error) { + case -EIO: + case -EROFS: + case -ENOMEM: + return false; + } + return true; +} /* * Call btrfs_abort_transaction as early as possible when an error condition is @@ -240,7 +254,7 @@ do { \ if (!test_and_set_bit(BTRFS_FS_STATE_TRANS_ABORTED, \ &((trans)->fs_info->fs_state))) { \ __first = true; \ - if (WARN(abort_should_print_stack(error), \ + if (WARN(btrfs_abort_should_print_stack(error), \ KERN_ERR \ "BTRFS: Transaction aborted (error %d)\n", \ (error))) { \ diff --git a/fs/btrfs/tree-checker.c b/fs/btrfs/tree-checker.c index dfeee033f31f..43979891f7c8 100644 --- a/fs/btrfs/tree-checker.c +++ b/fs/btrfs/tree-checker.c @@ -764,22 +764,19 @@ static int check_block_group_item(struct extent_buffer *leaf, return 0; } -__printf(4, 5) +__printf(5, 6) __cold -static void chunk_err(const struct extent_buffer *leaf, +static void chunk_err(const struct btrfs_fs_info *fs_info, + const struct extent_buffer *leaf, const struct btrfs_chunk *chunk, u64 logical, const char *fmt, ...) { - const struct btrfs_fs_info *fs_info = leaf->fs_info; - bool is_sb; + bool is_sb = !leaf; struct va_format vaf; va_list args; int i; int slot = -1; - /* Only superblock eb is able to have such small offset */ - is_sb = (leaf->start == BTRFS_SUPER_INFO_OFFSET); - if (!is_sb) { /* * Get the slot number by iterating through all slots, this @@ -812,13 +809,17 @@ static void chunk_err(const struct extent_buffer *leaf, /* * The common chunk check which could also work on super block sys chunk array. * + * If @leaf is NULL, then @chunk must be an on-stack chunk item. + * (For superblock sys_chunk array, and fs_info->sectorsize is unreliable) + * * Return -EUCLEAN if anything is corrupted. * Return 0 if everything is OK. */ -int btrfs_check_chunk_valid(struct extent_buffer *leaf, - struct btrfs_chunk *chunk, u64 logical) +int btrfs_check_chunk_valid(const struct btrfs_fs_info *fs_info, + const struct extent_buffer *leaf, + const struct btrfs_chunk *chunk, u64 logical, + u32 sectorsize) { - struct btrfs_fs_info *fs_info = leaf->fs_info; u64 length; u64 chunk_end; u64 stripe_len; @@ -826,63 +827,73 @@ int btrfs_check_chunk_valid(struct extent_buffer *leaf, u16 sub_stripes; u64 type; u64 features; + u32 chunk_sector_size; bool mixed = false; int raid_index; int nparity; int ncopies; - length = btrfs_chunk_length(leaf, chunk); - stripe_len = btrfs_chunk_stripe_len(leaf, chunk); - num_stripes = btrfs_chunk_num_stripes(leaf, chunk); - sub_stripes = btrfs_chunk_sub_stripes(leaf, chunk); - type = btrfs_chunk_type(leaf, chunk); + if (leaf) { + length = btrfs_chunk_length(leaf, chunk); + stripe_len = btrfs_chunk_stripe_len(leaf, chunk); + num_stripes = btrfs_chunk_num_stripes(leaf, chunk); + sub_stripes = btrfs_chunk_sub_stripes(leaf, chunk); + type = btrfs_chunk_type(leaf, chunk); + chunk_sector_size = btrfs_chunk_sector_size(leaf, chunk); + } else { + length = btrfs_stack_chunk_length(chunk); + stripe_len = btrfs_stack_chunk_stripe_len(chunk); + num_stripes = btrfs_stack_chunk_num_stripes(chunk); + sub_stripes = btrfs_stack_chunk_sub_stripes(chunk); + type = btrfs_stack_chunk_type(chunk); + chunk_sector_size = btrfs_stack_chunk_sector_size(chunk); + } raid_index = btrfs_bg_flags_to_raid_index(type); ncopies = btrfs_raid_array[raid_index].ncopies; nparity = btrfs_raid_array[raid_index].nparity; if (unlikely(!num_stripes)) { - chunk_err(leaf, chunk, logical, + chunk_err(fs_info, leaf, chunk, logical, "invalid chunk num_stripes, have %u", num_stripes); return -EUCLEAN; } if (unlikely(num_stripes < ncopies)) { - chunk_err(leaf, chunk, logical, + chunk_err(fs_info, leaf, chunk, logical, "invalid chunk num_stripes < ncopies, have %u < %d", num_stripes, ncopies); return -EUCLEAN; } if (unlikely(nparity && num_stripes == nparity)) { - chunk_err(leaf, chunk, logical, + chunk_err(fs_info, leaf, chunk, logical, "invalid chunk num_stripes == nparity, have %u == %d", num_stripes, nparity); return -EUCLEAN; } - if (unlikely(!IS_ALIGNED(logical, fs_info->sectorsize))) { - chunk_err(leaf, chunk, logical, + if (unlikely(!IS_ALIGNED(logical, sectorsize))) { + chunk_err(fs_info, leaf, chunk, logical, "invalid chunk logical, have %llu should aligned to %u", - logical, fs_info->sectorsize); + logical, sectorsize); return -EUCLEAN; } - if (unlikely(btrfs_chunk_sector_size(leaf, chunk) != fs_info->sectorsize)) { - chunk_err(leaf, chunk, logical, + if (unlikely(chunk_sector_size != sectorsize)) { + chunk_err(fs_info, leaf, chunk, logical, "invalid chunk sectorsize, have %u expect %u", - btrfs_chunk_sector_size(leaf, chunk), - fs_info->sectorsize); + chunk_sector_size, sectorsize); return -EUCLEAN; } - if (unlikely(!length || !IS_ALIGNED(length, fs_info->sectorsize))) { - chunk_err(leaf, chunk, logical, + if (unlikely(!length || !IS_ALIGNED(length, sectorsize))) { + chunk_err(fs_info, leaf, chunk, logical, "invalid chunk length, have %llu", length); return -EUCLEAN; } if (unlikely(check_add_overflow(logical, length, &chunk_end))) { - chunk_err(leaf, chunk, logical, + chunk_err(fs_info, leaf, chunk, logical, "invalid chunk logical start and length, have logical start %llu length %llu", logical, length); return -EUCLEAN; } if (unlikely(!is_power_of_2(stripe_len) || stripe_len != BTRFS_STRIPE_LEN)) { - chunk_err(leaf, chunk, logical, + chunk_err(fs_info, leaf, chunk, logical, "invalid chunk stripe length: %llu", stripe_len); return -EUCLEAN; @@ -896,30 +907,29 @@ int btrfs_check_chunk_valid(struct extent_buffer *leaf, * Thus it should be a good way to catch obvious bitflips. */ if (unlikely(length >= btrfs_stripe_nr_to_offset(U32_MAX))) { - chunk_err(leaf, chunk, logical, + chunk_err(fs_info, leaf, chunk, logical, "chunk length too large: have %llu limit %llu", length, btrfs_stripe_nr_to_offset(U32_MAX)); return -EUCLEAN; } if (unlikely(type & ~(BTRFS_BLOCK_GROUP_TYPE_MASK | BTRFS_BLOCK_GROUP_PROFILE_MASK))) { - chunk_err(leaf, chunk, logical, + chunk_err(fs_info, leaf, chunk, logical, "unrecognized chunk type: 0x%llx", ~(BTRFS_BLOCK_GROUP_TYPE_MASK | - BTRFS_BLOCK_GROUP_PROFILE_MASK) & - btrfs_chunk_type(leaf, chunk)); + BTRFS_BLOCK_GROUP_PROFILE_MASK) & type); return -EUCLEAN; } if (unlikely(!has_single_bit_set(type & BTRFS_BLOCK_GROUP_PROFILE_MASK) && (type & BTRFS_BLOCK_GROUP_PROFILE_MASK) != 0)) { - chunk_err(leaf, chunk, logical, + chunk_err(fs_info, leaf, chunk, logical, "invalid chunk profile flag: 0x%llx, expect 0 or 1 bit set", type & BTRFS_BLOCK_GROUP_PROFILE_MASK); return -EUCLEAN; } if (unlikely((type & BTRFS_BLOCK_GROUP_TYPE_MASK) == 0)) { - chunk_err(leaf, chunk, logical, + chunk_err(fs_info, leaf, chunk, logical, "missing chunk type flag, have 0x%llx one bit must be set in 0x%llx", type, BTRFS_BLOCK_GROUP_TYPE_MASK); return -EUCLEAN; @@ -928,7 +938,7 @@ int btrfs_check_chunk_valid(struct extent_buffer *leaf, if (unlikely((type & BTRFS_BLOCK_GROUP_SYSTEM) && (type & (BTRFS_BLOCK_GROUP_METADATA | BTRFS_BLOCK_GROUP_DATA)))) { - chunk_err(leaf, chunk, logical, + chunk_err(fs_info, leaf, chunk, logical, "system chunk with data or metadata type: 0x%llx", type); return -EUCLEAN; @@ -941,7 +951,7 @@ int btrfs_check_chunk_valid(struct extent_buffer *leaf, if (!mixed) { if (unlikely((type & BTRFS_BLOCK_GROUP_METADATA) && (type & BTRFS_BLOCK_GROUP_DATA))) { - chunk_err(leaf, chunk, logical, + chunk_err(fs_info, leaf, chunk, logical, "mixed chunk type in non-mixed mode: 0x%llx", type); return -EUCLEAN; } @@ -963,7 +973,7 @@ int btrfs_check_chunk_valid(struct extent_buffer *leaf, num_stripes != btrfs_raid_array[BTRFS_RAID_DUP].dev_stripes) || ((type & BTRFS_BLOCK_GROUP_PROFILE_MASK) == 0 && num_stripes != btrfs_raid_array[BTRFS_RAID_SINGLE].dev_stripes))) { - chunk_err(leaf, chunk, logical, + chunk_err(fs_info, leaf, chunk, logical, "invalid num_stripes:sub_stripes %u:%u for profile %llu", num_stripes, sub_stripes, type & BTRFS_BLOCK_GROUP_PROFILE_MASK); @@ -983,14 +993,15 @@ static int check_leaf_chunk_item(struct extent_buffer *leaf, struct btrfs_chunk *chunk, struct btrfs_key *key, int slot) { + struct btrfs_fs_info *fs_info = leaf->fs_info; int num_stripes; if (unlikely(btrfs_item_size(leaf, slot) < sizeof(struct btrfs_chunk))) { - chunk_err(leaf, chunk, key->offset, + chunk_err(fs_info, leaf, chunk, key->offset, "invalid chunk item size: have %u expect [%zu, %u)", btrfs_item_size(leaf, slot), sizeof(struct btrfs_chunk), - BTRFS_LEAF_DATA_SIZE(leaf->fs_info)); + BTRFS_LEAF_DATA_SIZE(fs_info)); return -EUCLEAN; } @@ -1001,14 +1012,15 @@ static int check_leaf_chunk_item(struct extent_buffer *leaf, if (unlikely(btrfs_chunk_item_size(num_stripes) != btrfs_item_size(leaf, slot))) { - chunk_err(leaf, chunk, key->offset, + chunk_err(fs_info, leaf, chunk, key->offset, "invalid chunk item size: have %u expect %lu", btrfs_item_size(leaf, slot), btrfs_chunk_item_size(num_stripes)); return -EUCLEAN; } out: - return btrfs_check_chunk_valid(leaf, chunk, key->offset); + return btrfs_check_chunk_valid(fs_info, leaf, chunk, key->offset, + fs_info->sectorsize); } __printf(3, 4) diff --git a/fs/btrfs/tree-checker.h b/fs/btrfs/tree-checker.h index db67f96cbe4b..eb201f4ec3c7 100644 --- a/fs/btrfs/tree-checker.h +++ b/fs/btrfs/tree-checker.h @@ -10,6 +10,7 @@ #include <uapi/linux/btrfs_tree.h> struct extent_buffer; +struct btrfs_fs_info; struct btrfs_chunk; struct btrfs_key; @@ -66,8 +67,10 @@ enum btrfs_tree_block_status __btrfs_check_node(struct extent_buffer *node); int btrfs_check_leaf(struct extent_buffer *leaf); int btrfs_check_node(struct extent_buffer *node); -int btrfs_check_chunk_valid(struct extent_buffer *leaf, - struct btrfs_chunk *chunk, u64 logical); +int btrfs_check_chunk_valid(const struct btrfs_fs_info *fs_info, + const struct extent_buffer *leaf, + const struct btrfs_chunk *chunk, u64 logical, + u32 sectorsize); int btrfs_check_eb_owner(const struct extent_buffer *eb, u64 root_owner); int btrfs_verify_level_key(struct extent_buffer *eb, const struct btrfs_tree_parent_check *check); diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c index c8d6587688b3..955d1677e865 100644 --- a/fs/btrfs/tree-log.c +++ b/fs/btrfs/tree-log.c @@ -590,7 +590,6 @@ insert: } } no_copy: - btrfs_mark_buffer_dirty(trans, path->nodes[0]); btrfs_release_path(path); return 0; } @@ -3588,7 +3587,6 @@ static noinline int insert_dir_log_key(struct btrfs_trans_handle *trans, last_offset = max(last_offset, curr_end); } btrfs_set_dir_log_end(path->nodes[0], item, last_offset); - btrfs_mark_buffer_dirty(trans, path->nodes[0]); btrfs_release_path(path); return 0; } @@ -4566,7 +4564,6 @@ copy_item: dst_index++; } - btrfs_mark_buffer_dirty(trans, dst_path->nodes[0]); btrfs_release_path(dst_path); out: kfree(ins_data); @@ -4776,7 +4773,6 @@ static int log_one_extent(struct btrfs_trans_handle *trans, write_extent_buffer(leaf, &fi, btrfs_item_ptr_offset(leaf, path->slots[0]), sizeof(fi)); - btrfs_mark_buffer_dirty(trans, leaf); btrfs_release_path(path); diff --git a/fs/btrfs/uuid-tree.c b/fs/btrfs/uuid-tree.c index aca2861f2187..17b5e81123a1 100644 --- a/fs/btrfs/uuid-tree.c +++ b/fs/btrfs/uuid-tree.c @@ -140,8 +140,6 @@ int btrfs_uuid_tree_add(struct btrfs_trans_handle *trans, const u8 *uuid, u8 typ ret = 0; subid_le = cpu_to_le64(subid_cpu); write_extent_buffer(eb, &subid_le, offset, sizeof(subid_le)); - btrfs_mark_buffer_dirty(trans, eb); - out: btrfs_free_path(path); return ret; diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c index 1cccaf9c2b0d..0a0776489055 100644 --- a/fs/btrfs/volumes.c +++ b/fs/btrfs/volumes.c @@ -13,8 +13,8 @@ #include <linux/list_sort.h> #include <linux/namei.h> #include "misc.h" -#include "ctree.h" #include "disk-io.h" +#include "extent-tree.h" #include "transaction.h" #include "volumes.h" #include "raid56.h" @@ -48,6 +48,7 @@ struct btrfs_io_geometry { u64 raid56_full_stripe_start; int max_errors; enum btrfs_map_op op; + bool use_rst; }; const struct btrfs_raid_attr btrfs_raid_array[BTRFS_NR_RAID_TYPES] = { @@ -797,6 +798,10 @@ static int get_canonical_dev_path(const char *dev_path, char *canonical) if (ret) goto out; resolved_path = d_path(&path, path_buf, PATH_MAX); + if (IS_ERR(resolved_path)) { + ret = PTR_ERR(resolved_path); + goto out; + } ret = strscpy(canonical, resolved_path, PATH_MAX); out: kfree(path_buf); @@ -1298,6 +1303,7 @@ static int open_fs_devices(struct btrfs_fs_devices *fs_devices, struct btrfs_device *device; struct btrfs_device *latest_dev = NULL; struct btrfs_device *tmp_device; + s64 __maybe_unused value = 0; int ret = 0; list_for_each_entry_safe(device, tmp_device, &fs_devices->devices, @@ -1327,7 +1333,23 @@ static int open_fs_devices(struct btrfs_fs_devices *fs_devices, fs_devices->latest_dev = latest_dev; fs_devices->total_rw_bytes = 0; fs_devices->chunk_alloc_policy = BTRFS_CHUNK_ALLOC_REGULAR; +#ifdef CONFIG_BTRFS_EXPERIMENTAL + fs_devices->rr_min_contig_read = BTRFS_DEFAULT_RR_MIN_CONTIG_READ; + fs_devices->read_devid = latest_dev->devid; + fs_devices->read_policy = btrfs_read_policy_to_enum(btrfs_get_mod_read_policy(), + &value); + if (fs_devices->read_policy == BTRFS_READ_POLICY_RR) + fs_devices->collect_fs_stats = true; + + if (value) { + if (fs_devices->read_policy == BTRFS_READ_POLICY_RR) + fs_devices->rr_min_contig_read = value; + if (fs_devices->read_policy == BTRFS_READ_POLICY_DEVID) + fs_devices->read_devid = value; + } +#else fs_devices->read_policy = BTRFS_READ_POLICY_PID; +#endif return 0; } @@ -2045,7 +2067,6 @@ static int btrfs_add_dev_item(struct btrfs_trans_handle *trans, ptr = btrfs_device_fsid(dev_item); write_extent_buffer(leaf, trans->fs_info->fs_devices->metadata_uuid, ptr, BTRFS_FSID_SIZE); - btrfs_mark_buffer_dirty(trans, leaf); ret = 0; out: @@ -2741,11 +2762,9 @@ next_slot: device = btrfs_find_device(fs_info->fs_devices, &args); BUG_ON(!device); /* Logic error */ - if (device->fs_devices->seeding) { + if (device->fs_devices->seeding) btrfs_set_device_generation(leaf, dev_item, device->generation); - btrfs_mark_buffer_dirty(trans, leaf); - } path->slots[0]++; goto next_slot; @@ -3038,8 +3057,6 @@ static noinline int btrfs_update_device(struct btrfs_trans_handle *trans, btrfs_device_get_disk_total_bytes(device)); btrfs_set_device_bytes_used(leaf, dev_item, btrfs_device_get_bytes_used(device)); - btrfs_mark_buffer_dirty(trans, leaf); - out: btrfs_free_path(path); return ret; @@ -3748,10 +3765,7 @@ static int insert_balance_item(struct btrfs_fs_info *fs_info, btrfs_set_balance_meta(leaf, item, &disk_bargs); btrfs_cpu_balance_args_to_disk(&disk_bargs, &bctl->sys); btrfs_set_balance_sys(leaf, item, &disk_bargs); - btrfs_set_balance_flags(leaf, item, bctl->flags); - - btrfs_mark_buffer_dirty(trans, leaf); out: btrfs_free_path(path); err = btrfs_commit_transaction(trans); @@ -5513,33 +5527,34 @@ void btrfs_remove_chunk_map(struct btrfs_fs_info *fs_info, struct btrfs_chunk_ma btrfs_free_chunk_map(map); } +static int btrfs_chunk_map_cmp(const struct rb_node *new, + const struct rb_node *exist) +{ + const struct btrfs_chunk_map *new_map = + rb_entry(new, struct btrfs_chunk_map, rb_node); + const struct btrfs_chunk_map *exist_map = + rb_entry(exist, struct btrfs_chunk_map, rb_node); + + if (new_map->start == exist_map->start) + return 0; + if (new_map->start < exist_map->start) + return -1; + return 1; +} + EXPORT_FOR_TESTS int btrfs_add_chunk_map(struct btrfs_fs_info *fs_info, struct btrfs_chunk_map *map) { - struct rb_node **p; - struct rb_node *parent = NULL; - bool leftmost = true; + struct rb_node *exist; write_lock(&fs_info->mapping_tree_lock); - p = &fs_info->mapping_tree.rb_root.rb_node; - while (*p) { - struct btrfs_chunk_map *entry; - - parent = *p; - entry = rb_entry(parent, struct btrfs_chunk_map, rb_node); - - if (map->start < entry->start) { - p = &(*p)->rb_left; - } else if (map->start > entry->start) { - p = &(*p)->rb_right; - leftmost = false; - } else { - write_unlock(&fs_info->mapping_tree_lock); - return -EEXIST; - } + exist = rb_find_add_cached(&map->rb_node, &fs_info->mapping_tree, + btrfs_chunk_map_cmp); + + if (exist) { + write_unlock(&fs_info->mapping_tree_lock); + return -EEXIST; } - rb_link_node(&map->rb_node, parent, p); - rb_insert_color_cached(&map->rb_node, &fs_info->mapping_tree, leftmost); chunk_map_device_set_bits(map, CHUNK_ALLOCATED); chunk_map_device_clear_bits(map, CHUNK_TRIMMED); write_unlock(&fs_info->mapping_tree_lock); @@ -5959,6 +5974,76 @@ unsigned long btrfs_full_stripe_len(struct btrfs_fs_info *fs_info, return len; } +#ifdef CONFIG_BTRFS_EXPERIMENTAL +static int btrfs_read_preferred(struct btrfs_chunk_map *map, int first, int num_stripes) +{ + for (int index = first; index < first + num_stripes; index++) { + const struct btrfs_device *device = map->stripes[index].dev; + + if (device->devid == READ_ONCE(device->fs_devices->read_devid)) + return index; + } + + /* If no read-preferred device is set use the first stripe. */ + return first; +} + +struct stripe_mirror { + u64 devid; + int num; +}; + +static int btrfs_cmp_devid(const void *a, const void *b) +{ + const struct stripe_mirror *s1 = (const struct stripe_mirror *)a; + const struct stripe_mirror *s2 = (const struct stripe_mirror *)b; + + if (s1->devid < s2->devid) + return -1; + if (s1->devid > s2->devid) + return 1; + return 0; +} + +/* + * Select a stripe for reading using the round-robin algorithm. + * + * 1. Compute the read cycle as the total sectors read divided by the minimum + * sectors per device. + * 2. Determine the stripe number for the current read by taking the modulus + * of the read cycle with the total number of stripes: + * + * stripe index = (total sectors / min sectors per dev) % num stripes + * + * The calculated stripe index is then used to select the corresponding device + * from the list of devices, which is ordered by devid. + */ +static int btrfs_read_rr(const struct btrfs_chunk_map *map, int first, int num_stripes) +{ + struct stripe_mirror stripes[BTRFS_RAID1_MAX_MIRRORS] = { 0 }; + struct btrfs_device *device = map->stripes[first].dev; + struct btrfs_fs_info *fs_info = device->fs_devices->fs_info; + unsigned int read_cycle; + unsigned int total_reads; + unsigned int min_reads_per_dev; + + total_reads = percpu_counter_sum(&fs_info->stats_read_blocks); + min_reads_per_dev = READ_ONCE(fs_info->fs_devices->rr_min_contig_read) >> + fs_info->sectorsize_bits; + + for (int index = 0, i = first; i < first + num_stripes; i++) { + stripes[index].devid = map->stripes[i].dev->devid; + stripes[index].num = i; + index++; + } + sort(stripes, num_stripes, sizeof(struct stripe_mirror), + btrfs_cmp_devid, NULL); + + read_cycle = total_reads / min_reads_per_dev; + return stripes[read_cycle % num_stripes].num; +} +#endif + static int find_live_mirror(struct btrfs_fs_info *fs_info, struct btrfs_chunk_map *map, int first, int dev_replace_is_ongoing) @@ -5988,6 +6073,14 @@ static int find_live_mirror(struct btrfs_fs_info *fs_info, case BTRFS_READ_POLICY_PID: preferred_mirror = first + (current->pid % num_stripes); break; +#ifdef CONFIG_BTRFS_EXPERIMENTAL + case BTRFS_READ_POLICY_RR: + preferred_mirror = btrfs_read_rr(map, first, num_stripes); + break; + case BTRFS_READ_POLICY_DEVID: + preferred_mirror = btrfs_read_preferred(map, first, num_stripes); + break; +#endif } if (dev_replace_is_ongoing && @@ -6346,8 +6439,7 @@ static int set_io_stripe(struct btrfs_fs_info *fs_info, u64 logical, { dst->dev = map->stripes[io_geom->stripe_index].dev; - if (io_geom->op == BTRFS_MAP_READ && - btrfs_need_stripe_tree_update(fs_info, map->type)) + if (io_geom->op == BTRFS_MAP_READ && io_geom->use_rst) return btrfs_get_raid_extent_offset(fs_info, logical, length, map->type, io_geom->stripe_index, dst); @@ -6362,7 +6454,7 @@ static bool is_single_device_io(struct btrfs_fs_info *fs_info, const struct btrfs_io_stripe *smap, const struct btrfs_chunk_map *map, int num_alloc_stripes, - enum btrfs_map_op op, int mirror_num) + struct btrfs_io_geometry *io_geom) { if (!smap) return false; @@ -6370,10 +6462,10 @@ static bool is_single_device_io(struct btrfs_fs_info *fs_info, if (num_alloc_stripes != 1) return false; - if (btrfs_need_stripe_tree_update(fs_info, map->type) && op != BTRFS_MAP_READ) + if (io_geom->use_rst && io_geom->op != BTRFS_MAP_READ) return false; - if ((map->type & BTRFS_BLOCK_GROUP_RAID56_MASK) && mirror_num > 1) + if ((map->type & BTRFS_BLOCK_GROUP_RAID56_MASK) && io_geom->mirror_num > 1) return false; return true; @@ -6579,6 +6671,7 @@ int btrfs_map_block(struct btrfs_fs_info *fs_info, enum btrfs_map_op op, io_geom.raid56_full_stripe_start = (u64)-1; max_len = btrfs_max_io_len(map, map_offset, &io_geom); *length = min_t(u64, map->chunk_len - map_offset, max_len); + io_geom.use_rst = btrfs_need_stripe_tree_update(fs_info, map->type); if (dev_replace->replace_task != current) down_read(&dev_replace->rwsem); @@ -6647,8 +6740,7 @@ int btrfs_map_block(struct btrfs_fs_info *fs_info, enum btrfs_map_op op, * physical block information on the stack instead of allocating an * I/O context structure. */ - if (is_single_device_io(fs_info, smap, map, num_alloc_stripes, op, - io_geom.mirror_num)) { + if (is_single_device_io(fs_info, smap, map, num_alloc_stripes, &io_geom)) { ret = set_io_stripe(fs_info, logical, length, smap, map, &io_geom); if (mirror_num_ret) *mirror_num_ret = io_geom.mirror_num; @@ -6662,6 +6754,7 @@ int btrfs_map_block(struct btrfs_fs_info *fs_info, enum btrfs_map_op op, goto out; } bioc->map_type = map->type; + bioc->use_rst = io_geom.use_rst; /* * For RAID56 full map, we need to make sure the stripes[] follows the @@ -7002,16 +7095,6 @@ static int read_one_chunk(struct btrfs_key *key, struct extent_buffer *leaf, warn_32bit_meta_chunk(fs_info, logical, length, type); #endif - /* - * Only need to verify chunk item if we're reading from sys chunk array, - * as chunk item in tree block is already verified by tree-checker. - */ - if (leaf->start == BTRFS_SUPER_INFO_OFFSET) { - ret = btrfs_check_chunk_valid(leaf, chunk, logical); - if (ret) - return ret; - } - map = btrfs_find_chunk_map(fs_info, logical, 1); /* already mapped? */ @@ -7269,16 +7352,11 @@ int btrfs_read_sys_array(struct btrfs_fs_info *fs_info) { struct btrfs_super_block *super_copy = fs_info->super_copy; struct extent_buffer *sb; - struct btrfs_disk_key *disk_key; - struct btrfs_chunk *chunk; u8 *array_ptr; unsigned long sb_array_offset; int ret = 0; - u32 num_stripes; u32 array_size; - u32 len = 0; u32 cur_offset; - u64 type; struct btrfs_key key; ASSERT(BTRFS_SUPER_INFO_SIZE <= fs_info->nodesize); @@ -7301,10 +7379,15 @@ int btrfs_read_sys_array(struct btrfs_fs_info *fs_info) cur_offset = 0; while (cur_offset < array_size) { - disk_key = (struct btrfs_disk_key *)array_ptr; - len = sizeof(*disk_key); - if (cur_offset + len > array_size) - goto out_short_read; + struct btrfs_chunk *chunk; + struct btrfs_disk_key *disk_key = (struct btrfs_disk_key *)array_ptr; + u32 len = sizeof(*disk_key); + + /* + * The sys_chunk_array has been already verified at super block + * read time. Only do ASSERT()s for basic checks. + */ + ASSERT(cur_offset + len <= array_size); btrfs_disk_key_to_cpu(&key, disk_key); @@ -7312,44 +7395,14 @@ int btrfs_read_sys_array(struct btrfs_fs_info *fs_info) sb_array_offset += len; cur_offset += len; - if (key.type != BTRFS_CHUNK_ITEM_KEY) { - btrfs_err(fs_info, - "unexpected item type %u in sys_array at offset %u", - (u32)key.type, cur_offset); - ret = -EIO; - break; - } + ASSERT(key.type == BTRFS_CHUNK_ITEM_KEY); chunk = (struct btrfs_chunk *)sb_array_offset; - /* - * At least one btrfs_chunk with one stripe must be present, - * exact stripe count check comes afterwards - */ - len = btrfs_chunk_item_size(1); - if (cur_offset + len > array_size) - goto out_short_read; + ASSERT(btrfs_chunk_type(sb, chunk) & BTRFS_BLOCK_GROUP_SYSTEM); - num_stripes = btrfs_chunk_num_stripes(sb, chunk); - if (!num_stripes) { - btrfs_err(fs_info, - "invalid number of stripes %u in sys_array at offset %u", - num_stripes, cur_offset); - ret = -EIO; - break; - } + len = btrfs_chunk_item_size(btrfs_chunk_num_stripes(sb, chunk)); - type = btrfs_chunk_type(sb, chunk); - if ((type & BTRFS_BLOCK_GROUP_SYSTEM) == 0) { - btrfs_err(fs_info, - "invalid chunk type %llu in sys_array at offset %u", - type, cur_offset); - ret = -EIO; - break; - } - - len = btrfs_chunk_item_size(num_stripes); - if (cur_offset + len > array_size) - goto out_short_read; + ASSERT(cur_offset + len <= array_size); ret = read_one_chunk(&key, sb, chunk); if (ret) @@ -7362,13 +7415,6 @@ int btrfs_read_sys_array(struct btrfs_fs_info *fs_info) clear_extent_buffer_uptodate(sb); free_extent_buffer_stale(sb); return ret; - -out_short_read: - btrfs_err(fs_info, "sys_array too short to read %u bytes at offset %u", - len, cur_offset); - clear_extent_buffer_uptodate(sb); - free_extent_buffer_stale(sb); - return -EIO; } /* @@ -7568,8 +7614,6 @@ int btrfs_init_devices_late(struct btrfs_fs_info *fs_info) struct btrfs_device *device; int ret = 0; - fs_devices->fs_info = fs_info; - mutex_lock(&fs_devices->device_list_mutex); list_for_each_entry(device, &fs_devices->devices, dev_list) device->fs_info = fs_info; @@ -7745,8 +7789,6 @@ static int update_dev_stat_item(struct btrfs_trans_handle *trans, for (i = 0; i < BTRFS_DEV_STAT_VALUES_MAX; i++) btrfs_set_dev_stats_value(eb, ptr, i, btrfs_dev_stat_read(device, i)); - btrfs_mark_buffer_dirty(trans, eb); - out: btrfs_free_path(path); return ret; diff --git a/fs/btrfs/volumes.h b/fs/btrfs/volumes.h index 3a416b1bc24c..120f65e21eeb 100644 --- a/fs/btrfs/volumes.h +++ b/fs/btrfs/volumes.h @@ -296,6 +296,9 @@ enum btrfs_chunk_allocation_policy { BTRFS_CHUNK_ALLOC_ZONED, }; +#define BTRFS_DEFAULT_RR_MIN_CONTIG_READ (SZ_256K) +/* Keep in sync with raid_attr table, current maximum is RAID1C4. */ +#define BTRFS_RAID1_MAX_MIRRORS (4) /* * Read policies for mirrored block group profiles, read picks the stripe based * on these policies. @@ -303,6 +306,12 @@ enum btrfs_chunk_allocation_policy { enum btrfs_read_policy { /* Use process PID to choose the stripe */ BTRFS_READ_POLICY_PID, +#ifdef CONFIG_BTRFS_EXPERIMENTAL + /* Balancing RAID1 reads across all striped devices (round-robin). */ + BTRFS_READ_POLICY_RR, + /* Read from a specific device. */ + BTRFS_READ_POLICY_DEVID, +#endif BTRFS_NR_READ_POLICY, }; @@ -417,6 +426,8 @@ struct btrfs_fs_devices { bool seeding; /* The mount needs to use a randomly generated fsid. */ bool temp_fsid; + /* Enable/disable the filesystem stats tracking. */ + bool collect_fs_stats; struct btrfs_fs_info *fs_info; /* sysfs kobjects */ @@ -431,6 +442,15 @@ struct btrfs_fs_devices { enum btrfs_read_policy read_policy; #ifdef CONFIG_BTRFS_EXPERIMENTAL + /* + * Minimum contiguous reads before switching to next device, the unit + * is one block/sectorsize. + */ + u32 rr_min_contig_read; + + /* Device to be used for reading in case of RAID1. */ + u64 read_devid; + /* Checksum mode - offload it or do it synchronously. */ enum btrfs_offload_csum_mode offload_csum_mode; #endif @@ -485,6 +505,7 @@ struct btrfs_io_context { struct bio *orig_bio; atomic_t error; u16 max_errors; + bool use_rst; u64 logical; u64 size; diff --git a/fs/btrfs/xattr.c b/fs/btrfs/xattr.c index bc18710d1dcf..3e0edbcf73e1 100644 --- a/fs/btrfs/xattr.c +++ b/fs/btrfs/xattr.c @@ -204,7 +204,6 @@ int btrfs_setxattr(struct btrfs_trans_handle *trans, struct inode *inode, btrfs_set_dir_data_len(leaf, di, size); data_ptr = ((unsigned long)(di + 1)) + name_len; write_extent_buffer(leaf, value, data_ptr, size); - btrfs_mark_buffer_dirty(trans, leaf); } else { /* * Insert, and we had space for the xattr, so path->slots[0] is diff --git a/fs/btrfs/zlib.c b/fs/btrfs/zlib.c index ddf0d5a448a7..c9e92c6941ec 100644 --- a/fs/btrfs/zlib.c +++ b/fs/btrfs/zlib.c @@ -174,10 +174,10 @@ int zlib_compress_folios(struct list_head *ws, struct address_space *mapping, copy_page(workspace->buf + i * PAGE_SIZE, data_in); start += PAGE_SIZE; - workspace->strm.avail_in = - (in_buf_folios << PAGE_SHIFT); } workspace->strm.next_in = workspace->buf; + workspace->strm.avail_in = min(bytes_left, + in_buf_folios << PAGE_SHIFT); } else { unsigned int pg_off; unsigned int cur_len; diff --git a/fs/btrfs/zoned.c b/fs/btrfs/zoned.c index 11ed523e528e..73e0aa9fc08a 100644 --- a/fs/btrfs/zoned.c +++ b/fs/btrfs/zoned.c @@ -748,8 +748,9 @@ int btrfs_check_zoned_mode(struct btrfs_fs_info *fs_info) (u64)lim->max_segments << PAGE_SHIFT), fs_info->sectorsize); fs_info->fs_devices->chunk_alloc_policy = BTRFS_CHUNK_ALLOC_ZONED; - if (fs_info->max_zone_append_size < fs_info->max_extent_size) - fs_info->max_extent_size = fs_info->max_zone_append_size; + + fs_info->max_extent_size = min_not_zero(fs_info->max_extent_size, + fs_info->max_zone_append_size); /* * Check mount options here, because we might change fs_info->zoned @@ -2651,3 +2652,127 @@ void btrfs_check_active_zone_reservation(struct btrfs_fs_info *fs_info) } spin_unlock(&fs_info->zone_active_bgs_lock); } + +/* + * Reset the zones of unused block groups from @space_info->bytes_zone_unusable. + * + * @space_info: the space to work on + * @num_bytes: targeting reclaim bytes + * + * This one resets the zones of a block group, so we can reuse the region + * without removing the block group. On the other hand, btrfs_delete_unused_bgs() + * just removes a block group and frees up the underlying zones. So, we still + * need to allocate a new block group to reuse the zones. + * + * Resetting is faster than deleting/recreating a block group. It is similar + * to freeing the logical space on the regular mode. However, we cannot change + * the block group's profile with this operation. + */ +int btrfs_reset_unused_block_groups(struct btrfs_space_info *space_info, u64 num_bytes) +{ + struct btrfs_fs_info *fs_info = space_info->fs_info; + const sector_t zone_size_sectors = fs_info->zone_size >> SECTOR_SHIFT; + + if (!btrfs_is_zoned(fs_info)) + return 0; + + while (num_bytes > 0) { + struct btrfs_chunk_map *map; + struct btrfs_block_group *bg = NULL; + bool found = false; + u64 reclaimed = 0; + + /* + * Here, we choose a fully zone_unusable block group. It's + * technically possible to reset a partly zone_unusable block + * group, which still has some free space left. However, + * handling that needs to cope with the allocation side, which + * makes the logic more complex. So, let's handle the easy case + * for now. + */ + spin_lock(&fs_info->unused_bgs_lock); + list_for_each_entry(bg, &fs_info->unused_bgs, bg_list) { + if ((bg->flags & BTRFS_BLOCK_GROUP_TYPE_MASK) != space_info->flags) + continue; + + /* + * Use trylock to avoid locking order violation. In + * btrfs_reclaim_bgs_work(), the lock order is + * &bg->lock -> &fs_info->unused_bgs_lock. We skip a + * block group if we cannot take its lock. + */ + if (!spin_trylock(&bg->lock)) + continue; + if (btrfs_is_block_group_used(bg) || bg->zone_unusable < bg->length) { + spin_unlock(&bg->lock); + continue; + } + spin_unlock(&bg->lock); + found = true; + break; + } + if (!found) { + spin_unlock(&fs_info->unused_bgs_lock); + return 0; + } + + list_del_init(&bg->bg_list); + btrfs_put_block_group(bg); + spin_unlock(&fs_info->unused_bgs_lock); + + /* + * Since the block group is fully zone_unusable and we cannot + * allocate from this block group anymore, we don't need to set + * this block group read-only. + */ + + down_read(&fs_info->dev_replace.rwsem); + map = bg->physical_map; + for (int i = 0; i < map->num_stripes; i++) { + struct btrfs_io_stripe *stripe = &map->stripes[i]; + unsigned int nofs_flags; + int ret; + + nofs_flags = memalloc_nofs_save(); + ret = blkdev_zone_mgmt(stripe->dev->bdev, REQ_OP_ZONE_RESET, + stripe->physical >> SECTOR_SHIFT, + zone_size_sectors); + memalloc_nofs_restore(nofs_flags); + + if (ret) { + up_read(&fs_info->dev_replace.rwsem); + return ret; + } + } + up_read(&fs_info->dev_replace.rwsem); + + spin_lock(&space_info->lock); + spin_lock(&bg->lock); + ASSERT(!btrfs_is_block_group_used(bg)); + if (bg->ro) { + spin_unlock(&bg->lock); + spin_unlock(&space_info->lock); + continue; + } + + reclaimed = bg->alloc_offset; + bg->zone_unusable = bg->length - bg->zone_capacity; + bg->alloc_offset = 0; + /* + * This holds because we currently reset fully used then freed + * block group. + */ + ASSERT(reclaimed == bg->zone_capacity); + bg->free_space_ctl->free_space += reclaimed; + space_info->bytes_zone_unusable -= reclaimed; + spin_unlock(&bg->lock); + btrfs_return_free_space(space_info, reclaimed); + spin_unlock(&space_info->lock); + + if (num_bytes <= reclaimed) + break; + num_bytes -= reclaimed; + } + + return 0; +} diff --git a/fs/btrfs/zoned.h b/fs/btrfs/zoned.h index 7612e6572605..9672bf4c3335 100644 --- a/fs/btrfs/zoned.h +++ b/fs/btrfs/zoned.h @@ -96,6 +96,7 @@ int btrfs_zone_finish_one_bg(struct btrfs_fs_info *fs_info); int btrfs_zoned_activate_one_bg(struct btrfs_fs_info *fs_info, struct btrfs_space_info *space_info, bool do_finish); void btrfs_check_active_zone_reservation(struct btrfs_fs_info *fs_info); +int btrfs_reset_unused_block_groups(struct btrfs_space_info *space_info, u64 num_bytes); #else /* CONFIG_BLK_DEV_ZONED */ static inline int btrfs_get_dev_zone_info_all_devices(struct btrfs_fs_info *fs_info) @@ -265,6 +266,12 @@ static inline int btrfs_zoned_activate_one_bg(struct btrfs_fs_info *fs_info, static inline void btrfs_check_active_zone_reservation(struct btrfs_fs_info *fs_info) { } +static inline int btrfs_reset_unused_block_groups(struct btrfs_space_info *space_info, + u64 num_bytes) +{ + return 0; +} + #endif static inline bool btrfs_dev_is_sequential(struct btrfs_device *device, u64 pos) diff --git a/fs/cachefiles/daemon.c b/fs/cachefiles/daemon.c index 89b11336a836..1806bff8e59b 100644 --- a/fs/cachefiles/daemon.c +++ b/fs/cachefiles/daemon.c @@ -15,6 +15,7 @@ #include <linux/namei.h> #include <linux/poll.h> #include <linux/mount.h> +#include <linux/security.h> #include <linux/statfs.h> #include <linux/ctype.h> #include <linux/string.h> @@ -576,7 +577,7 @@ static int cachefiles_daemon_dir(struct cachefiles_cache *cache, char *args) */ static int cachefiles_daemon_secctx(struct cachefiles_cache *cache, char *args) { - char *secctx; + int err; _enter(",%s", args); @@ -585,16 +586,16 @@ static int cachefiles_daemon_secctx(struct cachefiles_cache *cache, char *args) return -EINVAL; } - if (cache->secctx) { + if (cache->have_secid) { pr_err("Second security context specified\n"); return -EINVAL; } - secctx = kstrdup(args, GFP_KERNEL); - if (!secctx) - return -ENOMEM; + err = security_secctx_to_secid(args, strlen(args), &cache->secid); + if (err) + return err; - cache->secctx = secctx; + cache->have_secid = true; return 0; } @@ -820,7 +821,6 @@ static void cachefiles_daemon_unbind(struct cachefiles_cache *cache) put_cred(cache->cache_cred); kfree(cache->rootdirname); - kfree(cache->secctx); kfree(cache->tag); _leave(""); diff --git a/fs/cachefiles/internal.h b/fs/cachefiles/internal.h index 7b99bd98de75..38c236e38cef 100644 --- a/fs/cachefiles/internal.h +++ b/fs/cachefiles/internal.h @@ -122,7 +122,6 @@ struct cachefiles_cache { #define CACHEFILES_STATE_CHANGED 3 /* T if state changed (poll trigger) */ #define CACHEFILES_ONDEMAND_MODE 4 /* T if in on-demand read mode */ char *rootdirname; /* name of cache root directory */ - char *secctx; /* LSM security context */ char *tag; /* cache binding tag */ refcount_t unbind_pincount;/* refcount to do daemon unbind */ struct xarray reqs; /* xarray of pending on-demand requests */ @@ -130,6 +129,8 @@ struct cachefiles_cache { struct xarray ondemand_ids; /* xarray for ondemand_id allocation */ u32 ondemand_id_next; u32 msg_id_next; + u32 secid; /* LSM security id */ + bool have_secid; /* whether "secid" was set */ }; static inline bool cachefiles_in_ondemand_mode(struct cachefiles_cache *cache) diff --git a/fs/cachefiles/io.c b/fs/cachefiles/io.c index 6a821a959b59..92058ae43488 100644 --- a/fs/cachefiles/io.c +++ b/fs/cachefiles/io.c @@ -13,6 +13,7 @@ #include <linux/falloc.h> #include <linux/sched/mm.h> #include <trace/events/fscache.h> +#include <trace/events/netfs.h> #include "internal.h" struct cachefiles_kiocb { @@ -366,6 +367,7 @@ static int cachefiles_write(struct netfs_cache_resources *cres, if (!fscache_wait_for_operation(cres, FSCACHE_WANT_WRITE)) { if (term_func) term_func(term_func_priv, -ENOBUFS, false); + trace_netfs_sreq(term_func_priv, netfs_sreq_trace_cache_nowrite); return -ENOBUFS; } @@ -695,6 +697,7 @@ static void cachefiles_issue_write(struct netfs_io_subrequest *subreq) iov_iter_truncate(&subreq->io_iter, len); } + trace_netfs_sreq(subreq, netfs_sreq_trace_cache_prepare); cachefiles_begin_secure(cache, &saved_cred); ret = __cachefiles_prepare_write(object, cachefiles_cres_file(cres), &start, &len, len, true); @@ -704,6 +707,7 @@ static void cachefiles_issue_write(struct netfs_io_subrequest *subreq) return; } + trace_netfs_sreq(subreq, netfs_sreq_trace_cache_write); cachefiles_write(&subreq->rreq->cache_resources, subreq->start, &subreq->io_iter, netfs_write_subrequest_terminated, subreq); diff --git a/fs/cachefiles/security.c b/fs/cachefiles/security.c index fe777164f1d8..fc6611886b3b 100644 --- a/fs/cachefiles/security.c +++ b/fs/cachefiles/security.c @@ -18,7 +18,7 @@ int cachefiles_get_security_ID(struct cachefiles_cache *cache) struct cred *new; int ret; - _enter("{%s}", cache->secctx); + _enter("{%u}", cache->have_secid ? cache->secid : 0); new = prepare_kernel_cred(current); if (!new) { @@ -26,8 +26,8 @@ int cachefiles_get_security_ID(struct cachefiles_cache *cache) goto error; } - if (cache->secctx) { - ret = set_security_override_from_ctx(new, cache->secctx); + if (cache->have_secid) { + ret = set_security_override(new, cache->secid); if (ret < 0) { put_cred(new); pr_err("Security denies permission to nominate security context: error %d\n", diff --git a/fs/cachefiles/xattr.c b/fs/cachefiles/xattr.c index 7c6f260a3be5..52383b1d0ba6 100644 --- a/fs/cachefiles/xattr.c +++ b/fs/cachefiles/xattr.c @@ -77,6 +77,7 @@ int cachefiles_set_object_xattr(struct cachefiles_object *object) trace_cachefiles_vfs_error(object, file_inode(file), ret, cachefiles_trace_setxattr_error); trace_cachefiles_coherency(object, file_inode(file)->i_ino, + be64_to_cpup((__be64 *)buf->data), buf->content, cachefiles_coherency_set_fail); if (ret != -ENOMEM) @@ -85,6 +86,7 @@ int cachefiles_set_object_xattr(struct cachefiles_object *object) "Failed to set xattr with error %d", ret); } else { trace_cachefiles_coherency(object, file_inode(file)->i_ino, + be64_to_cpup((__be64 *)buf->data), buf->content, cachefiles_coherency_set_ok); } @@ -126,7 +128,10 @@ int cachefiles_check_auxdata(struct cachefiles_object *object, struct file *file object, "Failed to read aux with error %zd", xlen); why = cachefiles_coherency_check_xattr; - } else if (buf->type != CACHEFILES_COOKIE_TYPE_DATA) { + goto out; + } + + if (buf->type != CACHEFILES_COOKIE_TYPE_DATA) { why = cachefiles_coherency_check_type; } else if (memcmp(buf->data, p, len) != 0) { why = cachefiles_coherency_check_aux; @@ -141,7 +146,9 @@ int cachefiles_check_auxdata(struct cachefiles_object *object, struct file *file ret = 0; } +out: trace_cachefiles_coherency(object, file_inode(file)->i_ino, + be64_to_cpup((__be64 *)buf->data), buf->content, why); kfree(buf); return ret; diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c index 85936f6d2bf7..f5224a566b69 100644 --- a/fs/ceph/addr.c +++ b/fs/ceph/addr.c @@ -223,10 +223,13 @@ static void finish_netfs_read(struct ceph_osd_request *req) subreq->len, i_size_read(req->r_inode)); /* no object means success but no data */ - if (err == -ENOENT) + if (err == -ENOENT) { + __set_bit(NETFS_SREQ_CLEAR_TAIL, &subreq->flags); + __set_bit(NETFS_SREQ_MADE_PROGRESS, &subreq->flags); err = 0; - else if (err == -EBLOCKLISTED) + } else if (err == -EBLOCKLISTED) { fsc->blocklisted = true; + } if (err >= 0) { if (sparse && err > 0) @@ -242,6 +245,8 @@ static void finish_netfs_read(struct ceph_osd_request *req) if (err > subreq->len) err = subreq->len; } + if (err > 0) + __set_bit(NETFS_SREQ_CLEAR_TAIL, &subreq->flags); } if (osd_data->type == CEPH_OSD_DATA_TYPE_PAGES) { @@ -253,8 +258,9 @@ static void finish_netfs_read(struct ceph_osd_request *req) subreq->transferred = err; err = 0; } + subreq->error = err; trace_netfs_sreq(subreq, netfs_sreq_trace_io_progress); - netfs_read_subreq_terminated(subreq, err, false); + netfs_read_subreq_terminated(subreq); iput(req->r_inode); ceph_dec_osd_stopping_blocker(fsc->mdsc); } @@ -314,7 +320,9 @@ static bool ceph_netfs_issue_op_inline(struct netfs_io_subrequest *subreq) ceph_mdsc_put_request(req); out: - netfs_read_subreq_terminated(subreq, err, false); + subreq->error = err; + trace_netfs_sreq(subreq, netfs_sreq_trace_io_progress); + netfs_read_subreq_terminated(subreq); return true; } @@ -426,8 +434,10 @@ static void ceph_netfs_issue_read(struct netfs_io_subrequest *subreq) ceph_osdc_start_request(req->r_osdc, req); out: ceph_osdc_put_request(req); - if (err) - netfs_read_subreq_terminated(subreq, err, false); + if (err) { + subreq->error = err; + netfs_read_subreq_terminated(subreq); + } doutc(cl, "%llx.%llx result %d\n", ceph_vinop(inode), err); } diff --git a/fs/dcache.c b/fs/dcache.c index b4d5e9e1e43d..1a01d7a6a7a9 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -1681,9 +1681,8 @@ static struct dentry *__d_alloc(struct super_block *sb, const struct qstr *name) /* Make sure we always see the terminating NUL character */ smp_store_release(&dentry->d_name.name, dname); /* ^^^ */ - dentry->d_lockref.count = 1; dentry->d_flags = 0; - spin_lock_init(&dentry->d_lock); + lockref_init(&dentry->d_lockref, 1); seqcount_spinlock_init(&dentry->d_seq, &dentry->d_lock); dentry->d_inode = NULL; dentry->d_parent = dentry; diff --git a/fs/debugfs/file.c b/fs/debugfs/file.c index 47dc96dfe386..16e198a26339 100644 --- a/fs/debugfs/file.c +++ b/fs/debugfs/file.c @@ -64,22 +64,13 @@ const struct file_operations *debugfs_real_fops(const struct file *filp) } EXPORT_SYMBOL_GPL(debugfs_real_fops); -/** - * debugfs_file_get - mark the beginning of file data access - * @dentry: the dentry object whose data is being accessed. - * - * Up to a matching call to debugfs_file_put(), any successive call - * into the file removing functions debugfs_remove() and - * debugfs_remove_recursive() will block. Since associated private - * file data may only get freed after a successful return of any of - * the removal functions, you may safely access it after a successful - * call to debugfs_file_get() without worrying about lifetime issues. - * - * If -%EIO is returned, the file has already been removed and thus, - * it is not safe to access any of its data. If, on the other hand, - * it is allowed to access the file data, zero is returned. - */ -int debugfs_file_get(struct dentry *dentry) +enum dbgfs_get_mode { + DBGFS_GET_ALREADY, + DBGFS_GET_REGULAR, + DBGFS_GET_SHORT, +}; + +static int __debugfs_file_get(struct dentry *dentry, enum dbgfs_get_mode mode) { struct debugfs_fsdata *fsd; void *d_fsd; @@ -96,15 +87,17 @@ int debugfs_file_get(struct dentry *dentry) if (!((unsigned long)d_fsd & DEBUGFS_FSDATA_IS_REAL_FOPS_BIT)) { fsd = d_fsd; } else { + if (WARN_ON(mode == DBGFS_GET_ALREADY)) + return -EINVAL; + fsd = kmalloc(sizeof(*fsd), GFP_KERNEL); if (!fsd) return -ENOMEM; - if ((unsigned long)d_fsd & DEBUGFS_FSDATA_IS_SHORT_FOPS_BIT) { + if (mode == DBGFS_GET_SHORT) { fsd->real_fops = NULL; fsd->short_fops = (void *)((unsigned long)d_fsd & - ~(DEBUGFS_FSDATA_IS_REAL_FOPS_BIT | - DEBUGFS_FSDATA_IS_SHORT_FOPS_BIT)); + ~DEBUGFS_FSDATA_IS_REAL_FOPS_BIT); } else { fsd->real_fops = (void *)((unsigned long)d_fsd & ~DEBUGFS_FSDATA_IS_REAL_FOPS_BIT); @@ -138,6 +131,26 @@ int debugfs_file_get(struct dentry *dentry) return 0; } + +/** + * debugfs_file_get - mark the beginning of file data access + * @dentry: the dentry object whose data is being accessed. + * + * Up to a matching call to debugfs_file_put(), any successive call + * into the file removing functions debugfs_remove() and + * debugfs_remove_recursive() will block. Since associated private + * file data may only get freed after a successful return of any of + * the removal functions, you may safely access it after a successful + * call to debugfs_file_get() without worrying about lifetime issues. + * + * If -%EIO is returned, the file has already been removed and thus, + * it is not safe to access any of its data. If, on the other hand, + * it is allowed to access the file data, zero is returned. + */ +int debugfs_file_get(struct dentry *dentry) +{ + return __debugfs_file_get(dentry, DBGFS_GET_ALREADY); +} EXPORT_SYMBOL_GPL(debugfs_file_get); /** @@ -267,7 +280,7 @@ static int open_proxy_open(struct inode *inode, struct file *filp) const struct file_operations *real_fops = NULL; int r; - r = debugfs_file_get(dentry); + r = __debugfs_file_get(dentry, DBGFS_GET_REGULAR); if (r) return r == -EIO ? -ENOENT : r; @@ -424,7 +437,8 @@ static void __full_proxy_fops_init(struct file_operations *proxy_fops, proxy_fops->unlocked_ioctl = full_proxy_unlocked_ioctl; } -static int full_proxy_open(struct inode *inode, struct file *filp) +static int full_proxy_open(struct inode *inode, struct file *filp, + enum dbgfs_get_mode mode) { struct dentry *dentry = F_DENTRY(filp); const struct file_operations *real_fops; @@ -432,7 +446,7 @@ static int full_proxy_open(struct inode *inode, struct file *filp) struct debugfs_fsdata *fsd; int r; - r = debugfs_file_get(dentry); + r = __debugfs_file_get(dentry, mode); if (r) return r == -EIO ? -ENOENT : r; @@ -491,8 +505,22 @@ out: return r; } +static int full_proxy_open_regular(struct inode *inode, struct file *filp) +{ + return full_proxy_open(inode, filp, DBGFS_GET_REGULAR); +} + const struct file_operations debugfs_full_proxy_file_operations = { - .open = full_proxy_open, + .open = full_proxy_open_regular, +}; + +static int full_proxy_open_short(struct inode *inode, struct file *filp) +{ + return full_proxy_open(inode, filp, DBGFS_GET_SHORT); +} + +const struct file_operations debugfs_full_short_proxy_file_operations = { + .open = full_proxy_open_short, }; ssize_t debugfs_attr_read(struct file *file, char __user *buf, diff --git a/fs/debugfs/inode.c b/fs/debugfs/inode.c index 38a9c7eb97e6..e752009de929 100644 --- a/fs/debugfs/inode.c +++ b/fs/debugfs/inode.c @@ -229,7 +229,7 @@ static void debugfs_release_dentry(struct dentry *dentry) return; /* check it wasn't a dir (no fsdata) or automount (no real_fops) */ - if (fsd && fsd->real_fops) { + if (fsd && (fsd->real_fops || fsd->short_fops)) { WARN_ON(!list_empty(&fsd->cancellations)); mutex_destroy(&fsd->cancellations_mtx); } @@ -455,8 +455,7 @@ struct dentry *debugfs_create_file_full(const char *name, umode_t mode, const struct file_operations *fops) { if (WARN_ON((unsigned long)fops & - (DEBUGFS_FSDATA_IS_SHORT_FOPS_BIT | - DEBUGFS_FSDATA_IS_REAL_FOPS_BIT))) + DEBUGFS_FSDATA_IS_REAL_FOPS_BIT)) return ERR_PTR(-EINVAL); return __debugfs_create_file(name, mode, parent, data, @@ -471,15 +470,13 @@ struct dentry *debugfs_create_file_short(const char *name, umode_t mode, const struct debugfs_short_fops *fops) { if (WARN_ON((unsigned long)fops & - (DEBUGFS_FSDATA_IS_SHORT_FOPS_BIT | - DEBUGFS_FSDATA_IS_REAL_FOPS_BIT))) + DEBUGFS_FSDATA_IS_REAL_FOPS_BIT)) return ERR_PTR(-EINVAL); return __debugfs_create_file(name, mode, parent, data, - fops ? &debugfs_full_proxy_file_operations : + fops ? &debugfs_full_short_proxy_file_operations : &debugfs_noop_file_operations, - (const void *)((unsigned long)fops | - DEBUGFS_FSDATA_IS_SHORT_FOPS_BIT)); + fops); } EXPORT_SYMBOL_GPL(debugfs_create_file_short); diff --git a/fs/debugfs/internal.h b/fs/debugfs/internal.h index a3edfa4f0d8e..bbae4a228ef4 100644 --- a/fs/debugfs/internal.h +++ b/fs/debugfs/internal.h @@ -15,6 +15,7 @@ struct file_operations; extern const struct file_operations debugfs_noop_file_operations; extern const struct file_operations debugfs_open_proxy_file_operations; extern const struct file_operations debugfs_full_proxy_file_operations; +extern const struct file_operations debugfs_full_short_proxy_file_operations; struct debugfs_fsdata { const struct file_operations *real_fops; @@ -40,11 +41,6 @@ struct debugfs_fsdata { * pointer gets its lowest bit set. */ #define DEBUGFS_FSDATA_IS_REAL_FOPS_BIT BIT(0) -/* - * A dentry's ->d_fsdata, when pointing to real fops, is with - * short fops instead of full fops. - */ -#define DEBUGFS_FSDATA_IS_SHORT_FOPS_BIT BIT(1) /* Access BITS */ #define DEBUGFS_ALLOW_API BIT(0) diff --git a/fs/dlm/config.c b/fs/dlm/config.c index b2f21aa00719..cf9ba6fd7a28 100644 --- a/fs/dlm/config.c +++ b/fs/dlm/config.c @@ -935,7 +935,7 @@ int dlm_comm_seq(int nodeid, uint32_t *seq, bool locked) mutex_unlock(&clusters_root.subsys.su_mutex); } if (!cm) - return -EEXIST; + return -ENOENT; *seq = cm->seq; put_comm(cm); diff --git a/fs/dlm/lock.c b/fs/dlm/lock.c index fc1d710166e9..c8ff88f1cdcf 100644 --- a/fs/dlm/lock.c +++ b/fs/dlm/lock.c @@ -824,9 +824,12 @@ static int find_rsb_dir(struct dlm_ls *ls, const void *name, int len, r->res_first_lkid = 0; } - /* A dir record will not be on the scan list. */ - if (r->res_dir_nodeid != our_nodeid) - del_scan(ls, r); + /* we always deactivate scan timer for the rsb, when + * we move it out of the inactive state as rsb state + * can be changed and scan timers are only for inactive + * rsbs. + */ + del_scan(ls, r); list_move(&r->res_slow_list, &ls->ls_slow_active); rsb_clear_flag(r, RSB_INACTIVE); kref_init(&r->res_ref); /* ref is now used in active state */ @@ -989,10 +992,10 @@ static int find_rsb_nodir(struct dlm_ls *ls, const void *name, int len, r->res_nodeid = 0; } + del_scan(ls, r); list_move(&r->res_slow_list, &ls->ls_slow_active); rsb_clear_flag(r, RSB_INACTIVE); kref_init(&r->res_ref); - del_scan(ls, r); write_unlock_bh(&ls->ls_rsbtbl_lock); goto out; @@ -1337,9 +1340,13 @@ static int _dlm_master_lookup(struct dlm_ls *ls, int from_nodeid, const char *na __dlm_master_lookup(ls, r, our_nodeid, from_nodeid, true, flags, r_nodeid, result); - /* A dir record rsb should never be on scan list. */ - /* Try to fix this with del_scan? */ - WARN_ON(!list_empty(&r->res_scan_list)); + /* A dir record rsb should never be on scan list. + * Except when we are the dir and master node. + * This function should only be called by the dir + * node. + */ + WARN_ON(!list_empty(&r->res_scan_list) && + r->res_master_nodeid != our_nodeid); write_unlock_bh(&ls->ls_rsbtbl_lock); @@ -1430,16 +1437,23 @@ static void deactivate_rsb(struct kref *kref) list_move(&r->res_slow_list, &ls->ls_slow_inactive); /* - * When the rsb becomes unused: - * - If it's not a dir record for a remote master rsb, - * then it is put on the scan list to be freed. - * - If it's a dir record for a remote master rsb, - * then it is kept in the inactive state until - * receive_remove() from the master node. + * When the rsb becomes unused, there are two possibilities: + * 1. Leave the inactive rsb in place (don't remove it). + * 2. Add it to the scan list to be removed. + * + * 1 is done when the rsb is acting as the dir record + * for a remotely mastered rsb. The rsb must be left + * in place as an inactive rsb to act as the dir record. + * + * 2 is done when a) the rsb is not the master and not the + * dir record, b) when the rsb is both the master and the + * dir record, c) when the rsb is master but not dir record. + * + * (If no directory is used, the rsb can always be removed.) */ - if (!dlm_no_directory(ls) && - (r->res_master_nodeid != our_nodeid) && - (dlm_dir_nodeid(r) != our_nodeid)) + if (dlm_no_directory(ls) || + (r->res_master_nodeid == our_nodeid || + dlm_dir_nodeid(r) != our_nodeid)) add_scan(ls, r); if (r->res_lvbptr) { diff --git a/fs/dlm/lowcomms.c b/fs/dlm/lowcomms.c index df40c3fd1070..d28141829c05 100644 --- a/fs/dlm/lowcomms.c +++ b/fs/dlm/lowcomms.c @@ -462,7 +462,8 @@ static bool dlm_lowcomms_con_has_addr(const struct connection *con, int dlm_lowcomms_addr(int nodeid, struct sockaddr_storage *addr) { struct connection *con; - bool ret, idx; + bool ret; + int idx; idx = srcu_read_lock(&connections_srcu); con = nodeid2con(nodeid, GFP_NOFS); diff --git a/fs/erofs/zdata.c b/fs/erofs/zdata.c index 19ef4ff2a134..254f6ad2c336 100644 --- a/fs/erofs/zdata.c +++ b/fs/erofs/zdata.c @@ -747,8 +747,7 @@ static int z_erofs_register_pcluster(struct z_erofs_decompress_frontend *fe) if (IS_ERR(pcl)) return PTR_ERR(pcl); - spin_lock_init(&pcl->lockref.lock); - pcl->lockref.count = 1; /* one ref for this request */ + lockref_init(&pcl->lockref, 1); /* one ref for this request */ pcl->algorithmformat = map->m_algorithmformat; pcl->length = 0; pcl->partial = true; diff --git a/fs/exec.c b/fs/exec.c index 98cb7ba9983c..2f0acef8908e 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -1194,16 +1194,16 @@ static int unshare_sighand(struct task_struct *me) } /* - * These functions flushes out all traces of the currently running executable - * so that a new one can be started + * This is unlocked -- the string will always be NUL-terminated, but + * may show overlapping contents if racing concurrent reads. */ - void __set_task_comm(struct task_struct *tsk, const char *buf, bool exec) { - task_lock(tsk); + size_t len = min(strlen(buf), sizeof(tsk->comm) - 1); + trace_task_rename(tsk, buf); - strscpy_pad(tsk->comm, buf, sizeof(tsk->comm)); - task_unlock(tsk); + memcpy(tsk->comm, buf, len); + memset(&tsk->comm[len], 0, sizeof(tsk->comm) - len); perf_event_comm(tsk, exec); } @@ -1341,7 +1341,28 @@ int begin_new_exec(struct linux_binprm * bprm) set_dumpable(current->mm, SUID_DUMP_USER); perf_event_exec(); - __set_task_comm(me, kbasename(bprm->filename), true); + + /* + * If the original filename was empty, alloc_bprm() made up a path + * that will probably not be useful to admins running ps or similar. + * Let's fix it up to be something reasonable. + */ + if (bprm->comm_from_dentry) { + /* + * Hold RCU lock to keep the name from being freed behind our back. + * Use acquire semantics to make sure the terminating NUL from + * __d_alloc() is seen. + * + * Note, we're deliberately sloppy here. We don't need to care about + * detecting a concurrent rename and just want a terminated name. + */ + rcu_read_lock(); + __set_task_comm(me, smp_load_acquire(&bprm->file->f_path.dentry->d_name.name), + true); + rcu_read_unlock(); + } else { + __set_task_comm(me, kbasename(bprm->filename), true); + } /* An exec changes our domain. We are no longer part of the thread group */ @@ -1517,11 +1538,13 @@ static struct linux_binprm *alloc_bprm(int fd, struct filename *filename, int fl if (fd == AT_FDCWD || filename->name[0] == '/') { bprm->filename = filename->name; } else { - if (filename->name[0] == '\0') + if (filename->name[0] == '\0') { bprm->fdpath = kasprintf(GFP_KERNEL, "/dev/fd/%d", fd); - else + bprm->comm_from_dentry = 1; + } else { bprm->fdpath = kasprintf(GFP_KERNEL, "/dev/fd/%d/%s", fd, filename->name); + } if (!bprm->fdpath) goto out_free; @@ -1719,13 +1742,11 @@ int remove_arg_zero(struct linux_binprm *bprm) } EXPORT_SYMBOL(remove_arg_zero); -#define printable(c) (((c)=='\t') || ((c)=='\n') || (0x20<=(c) && (c)<=0x7e)) /* * cycle the list of binary formats handler, until one recognizes the image */ static int search_binary_handler(struct linux_binprm *bprm) { - bool need_retry = IS_ENABLED(CONFIG_MODULES); struct linux_binfmt *fmt; int retval; @@ -1737,8 +1758,6 @@ static int search_binary_handler(struct linux_binprm *bprm) if (retval) return retval; - retval = -ENOENT; - retry: read_lock(&binfmt_lock); list_for_each_entry(fmt, &formats, lh) { if (!try_module_get(fmt->module)) @@ -1756,17 +1775,7 @@ static int search_binary_handler(struct linux_binprm *bprm) } read_unlock(&binfmt_lock); - if (need_retry) { - if (printable(bprm->buf[0]) && printable(bprm->buf[1]) && - printable(bprm->buf[2]) && printable(bprm->buf[3])) - return retval; - if (request_module("binfmt-%04x", *(ushort *)(bprm->buf + 2)) < 0) - return retval; - need_retry = false; - goto retry; - } - - return retval; + return -ENOEXEC; } /* binfmt handlers will call back into begin_new_exec() on success. */ @@ -1904,9 +1913,6 @@ static int do_execveat_common(int fd, struct filename *filename, } retval = count(argv, MAX_ARG_STRINGS); - if (retval == 0) - pr_warn_once("process '%s' launched '%s' with NULL argv: empty string added\n", - current->comm, bprm->filename); if (retval < 0) goto out_free; bprm->argc = retval; @@ -1944,6 +1950,9 @@ static int do_execveat_common(int fd, struct filename *filename, if (retval < 0) goto out_free; bprm->argc = 1; + + pr_warn_once("process '%s' launched '%s' with NULL argv: empty string added\n", + current->comm, bprm->filename); } retval = bprm_execve(bprm); diff --git a/fs/exfat/dir.c b/fs/exfat/dir.c index fe0a9b8a0cd0..3103b932b674 100644 --- a/fs/exfat/dir.c +++ b/fs/exfat/dir.c @@ -122,7 +122,7 @@ static int exfat_readdir(struct inode *inode, loff_t *cpos, struct exfat_dir_ent type = exfat_get_entry_type(ep); if (type == TYPE_UNUSED) { brelse(bh); - break; + goto out; } if (type != TYPE_FILE && type != TYPE_DIR) { @@ -170,6 +170,7 @@ static int exfat_readdir(struct inode *inode, loff_t *cpos, struct exfat_dir_ent } } +out: dir_entry->namebuf.lfn[0] = '\0'; *cpos = EXFAT_DEN_TO_B(dentry); return 0; diff --git a/fs/exfat/fatent.c b/fs/exfat/fatent.c index 773c320d68f3..9e5492ac409b 100644 --- a/fs/exfat/fatent.c +++ b/fs/exfat/fatent.c @@ -216,6 +216,16 @@ static int __exfat_free_cluster(struct inode *inode, struct exfat_chain *p_chain if (err) goto dec_used_clus; + + if (num_clusters >= sbi->num_clusters - EXFAT_FIRST_CLUSTER) { + /* + * The cluster chain includes a loop, scan the + * bitmap to get the number of used clusters. + */ + exfat_count_used_clusters(sb, &sbi->used_clusters); + + return 0; + } } while (clu != EXFAT_EOF_CLUSTER); } diff --git a/fs/exfat/file.c b/fs/exfat/file.c index fb38769c3e39..05b51e721783 100644 --- a/fs/exfat/file.c +++ b/fs/exfat/file.c @@ -545,6 +545,7 @@ static int exfat_extend_valid_size(struct file *file, loff_t new_valid_size) while (pos < new_valid_size) { u32 len; struct folio *folio; + unsigned long off; len = PAGE_SIZE - (pos & (PAGE_SIZE - 1)); if (pos + len > new_valid_size) @@ -554,6 +555,9 @@ static int exfat_extend_valid_size(struct file *file, loff_t new_valid_size) if (err) goto out; + off = offset_in_folio(folio, pos); + folio_zero_new_buffers(folio, off, off + len); + err = ops->write_end(file, mapping, pos, len, len, folio, NULL); if (err < 0) goto out; @@ -563,6 +567,8 @@ static int exfat_extend_valid_size(struct file *file, loff_t new_valid_size) cond_resched(); } + return 0; + out: return err; } diff --git a/fs/exfat/namei.c b/fs/exfat/namei.c index 97d2774760fe..099f80645072 100644 --- a/fs/exfat/namei.c +++ b/fs/exfat/namei.c @@ -330,8 +330,8 @@ static int exfat_find_empty_entry(struct inode *inode, while ((dentry = exfat_search_empty_slot(sb, &hint_femp, p_dir, num_entries, es)) < 0) { - if (dentry == -EIO) - break; + if (dentry != -ENOSPC) + return dentry; if (exfat_check_max_dentries(inode)) return -ENOSPC; diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 89aade6f45f6..7c54ae5fcbd4 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -5006,10 +5006,11 @@ struct inode *__ext4_iget(struct super_block *sb, unsigned long ino, if (IS_ENCRYPTED(inode)) { inode->i_op = &ext4_encrypted_symlink_inode_operations; } else if (ext4_inode_is_fast_symlink(inode)) { - inode->i_link = (char *)ei->i_data; inode->i_op = &ext4_fast_symlink_inode_operations; nd_terminate_link(ei->i_data, inode->i_size, sizeof(ei->i_data) - 1); + inode_set_cached_link(inode, (char *)ei->i_data, + inode->i_size); } else { inode->i_op = &ext4_symlink_inode_operations; } diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c index bcf2737078b8..536d56d15072 100644 --- a/fs/ext4/namei.c +++ b/fs/ext4/namei.c @@ -3418,7 +3418,6 @@ retry: inode->i_op = &ext4_symlink_inode_operations; } else { inode->i_op = &ext4_fast_symlink_inode_operations; - inode->i_link = (char *)&EXT4_I(inode)->i_data; } } @@ -3434,6 +3433,9 @@ retry: disk_link.len); inode->i_size = disk_link.len - 1; EXT4_I(inode)->i_disksize = inode->i_size; + if (!IS_ENCRYPTED(inode)) + inode_set_cached_link(inode, (char *)&EXT4_I(inode)->i_data, + inode->i_size); } err = ext4_add_nondir(handle, dentry, &inode); if (handle) diff --git a/fs/fhandle.c b/fs/fhandle.c index ec9145047dfc..3e092ae6d142 100644 --- a/fs/fhandle.c +++ b/fs/fhandle.c @@ -187,17 +187,6 @@ static int get_path_from_fd(int fd, struct path *root) return 0; } -enum handle_to_path_flags { - HANDLE_CHECK_PERMS = (1 << 0), - HANDLE_CHECK_SUBTREE = (1 << 1), -}; - -struct handle_to_path_ctx { - struct path root; - enum handle_to_path_flags flags; - unsigned int fh_flags; -}; - static int vfs_dentry_acceptable(void *context, struct dentry *dentry) { struct handle_to_path_ctx *ctx = context; @@ -261,50 +250,55 @@ static int do_handle_to_path(struct file_handle *handle, struct path *path, { int handle_dwords; struct vfsmount *mnt = ctx->root.mnt; + struct dentry *dentry; /* change the handle size to multiple of sizeof(u32) */ handle_dwords = handle->handle_bytes >> 2; - path->dentry = exportfs_decode_fh_raw(mnt, - (struct fid *)handle->f_handle, - handle_dwords, handle->handle_type, - ctx->fh_flags, - vfs_dentry_acceptable, ctx); - if (IS_ERR_OR_NULL(path->dentry)) { - if (path->dentry == ERR_PTR(-ENOMEM)) + dentry = exportfs_decode_fh_raw(mnt, (struct fid *)handle->f_handle, + handle_dwords, handle->handle_type, + ctx->fh_flags, vfs_dentry_acceptable, + ctx); + if (IS_ERR_OR_NULL(dentry)) { + if (dentry == ERR_PTR(-ENOMEM)) return -ENOMEM; return -ESTALE; } + path->dentry = dentry; path->mnt = mntget(mnt); return 0; } -/* - * Allow relaxed permissions of file handles if the caller has the - * ability to mount the filesystem or create a bind-mount of the - * provided @mountdirfd. - * - * In both cases the caller may be able to get an unobstructed way to - * the encoded file handle. If the caller is only able to create a - * bind-mount we need to verify that there are no locked mounts on top - * of it that could prevent us from getting to the encoded file. - * - * In principle, locked mounts can prevent the caller from mounting the - * filesystem but that only applies to procfs and sysfs neither of which - * support decoding file handles. - */ -static inline bool may_decode_fh(struct handle_to_path_ctx *ctx, - unsigned int o_flags) +static inline int may_decode_fh(struct handle_to_path_ctx *ctx, + unsigned int o_flags) { struct path *root = &ctx->root; + if (capable(CAP_DAC_READ_SEARCH)) + return 0; + /* - * Restrict to O_DIRECTORY to provide a deterministic API that avoids a - * confusing api in the face of disconnected non-dir dentries. + * Allow relaxed permissions of file handles if the caller has + * the ability to mount the filesystem or create a bind-mount of + * the provided @mountdirfd. + * + * In both cases the caller may be able to get an unobstructed + * way to the encoded file handle. If the caller is only able to + * create a bind-mount we need to verify that there are no + * locked mounts on top of it that could prevent us from getting + * to the encoded file. + * + * In principle, locked mounts can prevent the caller from + * mounting the filesystem but that only applies to procfs and + * sysfs neither of which support decoding file handles. + * + * Restrict to O_DIRECTORY to provide a deterministic API that + * avoids a confusing api in the face of disconnected non-dir + * dentries. * * There's only one dentry for each directory inode (VFS rule)... */ if (!(o_flags & O_DIRECTORY)) - return false; + return -EPERM; if (ns_capable(root->mnt->mnt_sb->s_user_ns, CAP_SYS_ADMIN)) ctx->flags = HANDLE_CHECK_PERMS; @@ -314,14 +308,14 @@ static inline bool may_decode_fh(struct handle_to_path_ctx *ctx, !has_locked_children(real_mount(root->mnt), root->dentry)) ctx->flags = HANDLE_CHECK_PERMS | HANDLE_CHECK_SUBTREE; else - return false; + return -EPERM; /* Are we able to override DAC permissions? */ if (!ns_capable(current_user_ns(), CAP_DAC_READ_SEARCH)) - return false; + return -EPERM; ctx->fh_flags = EXPORT_FH_DIR_ONLY; - return true; + return 0; } static int handle_to_path(int mountdirfd, struct file_handle __user *ufh, @@ -331,15 +325,19 @@ static int handle_to_path(int mountdirfd, struct file_handle __user *ufh, struct file_handle f_handle; struct file_handle *handle = NULL; struct handle_to_path_ctx ctx = {}; + const struct export_operations *eops; retval = get_path_from_fd(mountdirfd, &ctx.root); if (retval) goto out_err; - if (!capable(CAP_DAC_READ_SEARCH) && !may_decode_fh(&ctx, o_flags)) { - retval = -EPERM; + eops = ctx.root.mnt->mnt_sb->s_export_op; + if (eops && eops->permission) + retval = eops->permission(&ctx, o_flags); + else + retval = may_decode_fh(&ctx, o_flags); + if (retval) goto out_path; - } if (copy_from_user(&f_handle, ufh, sizeof(struct file_handle))) { retval = -EFAULT; @@ -398,29 +396,28 @@ static long do_handle_open(int mountdirfd, struct file_handle __user *ufh, int open_flag) { long retval = 0; - struct path path; + struct path path __free(path_put) = {}; struct file *file; - int fd; + const struct export_operations *eops; retval = handle_to_path(mountdirfd, ufh, &path, open_flag); if (retval) return retval; - fd = get_unused_fd_flags(open_flag); - if (fd < 0) { - path_put(&path); + CLASS(get_unused_fd, fd)(O_CLOEXEC); + if (fd < 0) return fd; - } - file = file_open_root(&path, "", open_flag, 0); - if (IS_ERR(file)) { - put_unused_fd(fd); - retval = PTR_ERR(file); - } else { - retval = fd; - fd_install(fd, file); - } - path_put(&path); - return retval; + + eops = path.mnt->mnt_sb->s_export_op; + if (eops->open) + file = eops->open(&path, open_flag); + else + file = file_open_root(&path, "", open_flag, 0); + if (IS_ERR(file)) + return PTR_ERR(file); + + fd_install(fd, file); + return take_fd(fd); } /** diff --git a/fs/file.c b/fs/file.c index fb1011cf6b4a..d868cdb95d1e 100644 --- a/fs/file.c +++ b/fs/file.c @@ -22,6 +22,7 @@ #include <linux/close_range.h> #include <linux/file_ref.h> #include <net/sock.h> +#include <linux/init_task.h> #include "internal.h" @@ -278,10 +279,6 @@ repeat: if (nr < fdt->max_fds) return 0; - /* Can we expand? */ - if (nr >= sysctl_nr_open) - return -EMFILE; - if (unlikely(files->resize_in_progress)) { spin_unlock(&files->file_lock); wait_event(files->resize_wait, !files->resize_in_progress); @@ -289,6 +286,10 @@ repeat: goto repeat; } + /* Can we expand? */ + if (unlikely(nr >= sysctl_nr_open)) + return -EMFILE; + /* All good, so we try */ files->resize_in_progress = true; error = expand_fdtable(files, nr); @@ -1230,17 +1231,9 @@ __releases(&files->file_lock) /* * We need to detect attempts to do dup2() over allocated but still - * not finished descriptor. NB: OpenBSD avoids that at the price of - * extra work in their equivalent of fget() - they insert struct - * file immediately after grabbing descriptor, mark it larval if - * more work (e.g. actual opening) is needed and make sure that - * fget() treats larval files as absent. Potentially interesting, - * but while extra work in fget() is trivial, locking implications - * and amount of surgery on open()-related paths in VFS are not. - * FreeBSD fails with -EBADF in the same situation, NetBSD "solution" - * deadlocks in rather amusing ways, AFAICS. All of that is out of - * scope of POSIX or SUS, since neither considers shared descriptor - * tables and this condition does not arise without those. + * not finished descriptor. + * + * POSIX is silent on the issue, we return -EBUSY. */ fdt = files_fdtable(files); fd = array_index_nospec(fd, fdt->max_fds); diff --git a/fs/file_table.c b/fs/file_table.c index 976736be47cb..a32171d2b83f 100644 --- a/fs/file_table.c +++ b/fs/file_table.c @@ -128,7 +128,7 @@ static struct ctl_table fs_stat_sysctls[] = { .data = &sysctl_nr_open, .maxlen = sizeof(unsigned int), .mode = 0644, - .proc_handler = proc_dointvec_minmax, + .proc_handler = proc_douintvec_minmax, .extra1 = &sysctl_nr_open_min, .extra2 = &sysctl_nr_open_max, }, @@ -478,6 +478,8 @@ static void ____fput(struct callback_head *work) __fput(container_of(work, struct file, f_task_work)); } +static DECLARE_DELAYED_WORK(delayed_fput_work, delayed_fput); + /* * If kernel thread really needs to have the final fput() it has done * to complete, call this. The only user right now is the boot - we @@ -491,11 +493,10 @@ static void ____fput(struct callback_head *work) void flush_delayed_fput(void) { delayed_fput(NULL); + flush_delayed_work(&delayed_fput_work); } EXPORT_SYMBOL_GPL(flush_delayed_fput); -static DECLARE_DELAYED_WORK(delayed_fput_work, delayed_fput); - void fput(struct file *file) { if (file_ref_put(&file->f_ref)) { diff --git a/fs/fs_context.c b/fs/fs_context.c index 98589aae5208..582d33e81117 100644 --- a/fs/fs_context.c +++ b/fs/fs_context.c @@ -493,7 +493,7 @@ static void put_fc_log(struct fs_context *fc) if (log) { if (refcount_dec_and_test(&log->usage)) { fc->log.log = NULL; - for (i = 0; i <= 7; i++) + for (i = 0; i < ARRAY_SIZE(log->buffer) ; i++) if (log->need_free & (1 << i)) kfree(log->buffer[i]); kfree(log); diff --git a/fs/fs_parser.c b/fs/fs_parser.c index 16fa61ef56bf..e635a81e17d9 100644 --- a/fs/fs_parser.c +++ b/fs/fs_parser.c @@ -13,7 +13,7 @@ #include <linux/namei.h> #include "internal.h" -static const struct constant_table bool_names[] = { +const struct constant_table bool_names[] = { { "0", false }, { "1", true }, { "false", false }, @@ -22,6 +22,7 @@ static const struct constant_table bool_names[] = { { "yes", true }, { }, }; +EXPORT_SYMBOL(bool_names); static const struct constant_table * __lookup_constant(const struct constant_table *tbl, const char *name) diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c index 494ac372ace0..e540d05549ff 100644 --- a/fs/fuse/dir.c +++ b/fs/fuse/dir.c @@ -1681,6 +1681,8 @@ static int fuse_dir_open(struct inode *inode, struct file *file) */ if (ff->open_flags & (FOPEN_STREAM | FOPEN_NONSEEKABLE)) nonseekable_open(inode, file); + if (!(ff->open_flags & FOPEN_KEEP_CACHE)) + invalidate_inode_pages2(inode->i_mapping); } return err; diff --git a/fs/fuse/file.c b/fs/fuse/file.c index 88d0946b5bc9..7d92a5479998 100644 --- a/fs/fuse/file.c +++ b/fs/fuse/file.c @@ -1541,8 +1541,10 @@ static int fuse_get_user_pages(struct fuse_args_pages *ap, struct iov_iter *ii, */ struct page **pages = kzalloc(max_pages * sizeof(struct page *), GFP_KERNEL); - if (!pages) - return -ENOMEM; + if (!pages) { + ret = -ENOMEM; + goto out; + } while (nbytes < *nbytesp && nr_pages < max_pages) { unsigned nfolios, i; @@ -1557,18 +1559,22 @@ static int fuse_get_user_pages(struct fuse_args_pages *ap, struct iov_iter *ii, nbytes += ret; - ret += start; - /* Currently, all folios in FUSE are one page */ - nfolios = DIV_ROUND_UP(ret, PAGE_SIZE); + nfolios = DIV_ROUND_UP(ret + start, PAGE_SIZE); + + for (i = 0; i < nfolios; i++) { + struct folio *folio = page_folio(pages[i]); + unsigned int offset = start + + (folio_page_idx(folio, pages[i]) << PAGE_SHIFT); + unsigned int len = min_t(unsigned int, ret, PAGE_SIZE - start); - ap->descs[ap->num_folios].offset = start; - fuse_folio_descs_length_init(ap->descs, ap->num_folios, nfolios); - for (i = 0; i < nfolios; i++) - ap->folios[i + ap->num_folios] = page_folio(pages[i]); + ap->descs[ap->num_folios].offset = offset; + ap->descs[ap->num_folios].length = len; + ap->folios[ap->num_folios] = folio; + start = 0; + ret -= len; + ap->num_folios++; + } - ap->num_folios += nfolios; - ap->descs[ap->num_folios - 1].length -= - (PAGE_SIZE - ret) & (PAGE_SIZE - 1); nr_pages += nfolios; } kfree(pages); @@ -1584,6 +1590,7 @@ static int fuse_get_user_pages(struct fuse_args_pages *ap, struct iov_iter *ii, else ap->args.out_pages = true; +out: *nbytesp = nbytes; return ret < 0 ? ret : 0; diff --git a/fs/gfs2/file.c b/fs/gfs2/file.c index 1e73cf87ff88..c9bb3be21d2b 100644 --- a/fs/gfs2/file.c +++ b/fs/gfs2/file.c @@ -251,6 +251,7 @@ static int do_gfs2_set_flags(struct inode *inode, u32 reqflags, u32 mask) error = filemap_fdatawait(inode->i_mapping); if (error) goto out; + truncate_inode_pages(inode->i_mapping, 0); if (new_flags & GFS2_DIF_JDATA) gfs2_ordered_del_inode(ip); } diff --git a/fs/gfs2/quota.c b/fs/gfs2/quota.c index 72b48f6f5561..58bc5013ca49 100644 --- a/fs/gfs2/quota.c +++ b/fs/gfs2/quota.c @@ -236,8 +236,7 @@ static struct gfs2_quota_data *qd_alloc(unsigned hash, struct gfs2_sbd *sdp, str return NULL; qd->qd_sbd = sdp; - qd->qd_lockref.count = 0; - spin_lock_init(&qd->qd_lockref.lock); + lockref_init(&qd->qd_lockref, 0); qd->qd_id = qid; qd->qd_slot = -1; INIT_LIST_HEAD(&qd->qd_lru); diff --git a/fs/gfs2/quota.h b/fs/gfs2/quota.h index f462d9cb3087..988f38dc5b2c 100644 --- a/fs/gfs2/quota.h +++ b/fs/gfs2/quota.h @@ -44,8 +44,8 @@ static inline int gfs2_quota_lock_check(struct gfs2_inode *ip, int ret; ap->allowed = UINT_MAX; /* Assume we are permitted a whole lot */ - if (capable(CAP_SYS_RESOURCE) || - sdp->sd_args.ar_quota == GFS2_QUOTA_OFF) + if (sdp->sd_args.ar_quota == GFS2_QUOTA_OFF || + capable(CAP_SYS_RESOURCE)) return 0; ret = gfs2_quota_lock(ip, NO_UID_QUOTA_CHANGE, NO_GID_QUOTA_CHANGE); if (ret) diff --git a/fs/hfs/super.c b/fs/hfs/super.c index 3bee9b5dba5e..fe09c2093a93 100644 --- a/fs/hfs/super.c +++ b/fs/hfs/super.c @@ -349,11 +349,13 @@ static int hfs_fill_super(struct super_block *sb, struct fs_context *fc) goto bail_no_root; res = hfs_cat_find_brec(sb, HFS_ROOT_CNID, &fd); if (!res) { - if (fd.entrylength > sizeof(rec) || fd.entrylength < 0) { + if (fd.entrylength != sizeof(rec.dir)) { res = -EIO; goto bail_hfs_find; } hfs_bnode_read(fd.bnode, &rec, fd.entryoffset, fd.entrylength); + if (rec.type != HFS_CDR_DIR) + res = -EIO; } if (res) goto bail_hfs_find; diff --git a/fs/iomap/buffered-io.c b/fs/iomap/buffered-io.c index 955f19e27e47..d303e6c8900c 100644 --- a/fs/iomap/buffered-io.c +++ b/fs/iomap/buffered-io.c @@ -1138,7 +1138,7 @@ static void iomap_write_delalloc_scan(struct inode *inode, start_byte, end_byte, iomap, punch); /* move offset to start of next folio in range */ - start_byte = folio_next_index(folio) << PAGE_SHIFT; + start_byte = folio_pos(folio) + folio_size(folio); folio_unlock(folio); folio_put(folio); } @@ -1774,7 +1774,8 @@ static bool iomap_can_add_to_ioend(struct iomap_writepage_ctx *wpc, loff_t pos) */ static int iomap_add_to_ioend(struct iomap_writepage_ctx *wpc, struct writeback_control *wbc, struct folio *folio, - struct inode *inode, loff_t pos, unsigned len) + struct inode *inode, loff_t pos, loff_t end_pos, + unsigned len) { struct iomap_folio_state *ifs = folio->private; size_t poff = offset_in_folio(folio, pos); @@ -1793,15 +1794,60 @@ new_ioend: if (ifs) atomic_add(len, &ifs->write_bytes_pending); + + /* + * Clamp io_offset and io_size to the incore EOF so that ondisk + * file size updates in the ioend completion are byte-accurate. + * This avoids recovering files with zeroed tail regions when + * writeback races with appending writes: + * + * Thread 1: Thread 2: + * ------------ ----------- + * write [A, A+B] + * update inode size to A+B + * submit I/O [A, A+BS] + * write [A+B, A+B+C] + * update inode size to A+B+C + * <I/O completes, updates disk size to min(A+B+C, A+BS)> + * <power failure> + * + * After reboot: + * 1) with A+B+C < A+BS, the file has zero padding in range + * [A+B, A+B+C] + * + * |< Block Size (BS) >| + * |DDDDDDDDDDDD0000000000000| + * ^ ^ ^ + * A A+B A+B+C + * (EOF) + * + * 2) with A+B+C > A+BS, the file has zero padding in range + * [A+B, A+BS] + * + * |< Block Size (BS) >|< Block Size (BS) >| + * |DDDDDDDDDDDD0000000000000|00000000000000000000000000| + * ^ ^ ^ ^ + * A A+B A+BS A+B+C + * (EOF) + * + * D = Valid Data + * 0 = Zero Padding + * + * Note that this defeats the ability to chain the ioends of + * appending writes. + */ wpc->ioend->io_size += len; + if (wpc->ioend->io_offset + wpc->ioend->io_size > end_pos) + wpc->ioend->io_size = end_pos - wpc->ioend->io_offset; + wbc_account_cgroup_owner(wbc, folio, len); return 0; } static int iomap_writepage_map_blocks(struct iomap_writepage_ctx *wpc, struct writeback_control *wbc, struct folio *folio, - struct inode *inode, u64 pos, unsigned dirty_len, - unsigned *count) + struct inode *inode, u64 pos, u64 end_pos, + unsigned dirty_len, unsigned *count) { int error; @@ -1826,7 +1872,7 @@ static int iomap_writepage_map_blocks(struct iomap_writepage_ctx *wpc, break; default: error = iomap_add_to_ioend(wpc, wbc, folio, inode, pos, - map_len); + end_pos, map_len); if (!error) (*count)++; break; @@ -1897,11 +1943,11 @@ static bool iomap_writepage_handle_eof(struct folio *folio, struct inode *inode, * remaining memory is zeroed when mapped, and writes to that * region are not written out to the file. * - * Also adjust the writeback range to skip all blocks entirely - * beyond i_size. + * Also adjust the end_pos to the end of file and skip writeback + * for all blocks entirely beyond i_size. */ folio_zero_segment(folio, poff, folio_size(folio)); - *end_pos = round_up(isize, i_blocksize(inode)); + *end_pos = isize; } return true; @@ -1914,6 +1960,7 @@ static int iomap_writepage_map(struct iomap_writepage_ctx *wpc, struct inode *inode = folio->mapping->host; u64 pos = folio_pos(folio); u64 end_pos = pos + folio_size(folio); + u64 end_aligned = 0; unsigned count = 0; int error = 0; u32 rlen; @@ -1955,9 +2002,10 @@ static int iomap_writepage_map(struct iomap_writepage_ctx *wpc, /* * Walk through the folio to find dirty areas to write back. */ - while ((rlen = iomap_find_dirty_range(folio, &pos, end_pos))) { + end_aligned = round_up(end_pos, i_blocksize(inode)); + while ((rlen = iomap_find_dirty_range(folio, &pos, end_aligned))) { error = iomap_writepage_map_blocks(wpc, wbc, folio, inode, - pos, rlen, &count); + pos, end_pos, rlen, &count); if (error) break; pos += rlen; diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c index 9153ff3a08e7..e8e80761ac73 100644 --- a/fs/jbd2/commit.c +++ b/fs/jbd2/commit.c @@ -772,9 +772,9 @@ start_journal_io: /* * If the journal is not located on the file system device, * then we must flush the file system device before we issue - * the commit record + * the commit record and update the journal tail sequence. */ - if (commit_transaction->t_need_data_flush && + if ((commit_transaction->t_need_data_flush || update_tail) && (journal->j_fs_dev != journal->j_dev) && (journal->j_flags & JBD2_BARRIER)) blkdev_issue_flush(journal->j_fs_dev); diff --git a/fs/jbd2/revoke.c b/fs/jbd2/revoke.c index 4556e4689024..ce63d5fde9c3 100644 --- a/fs/jbd2/revoke.c +++ b/fs/jbd2/revoke.c @@ -654,7 +654,7 @@ static void flush_descriptor(journal_t *journal, set_buffer_jwrite(descriptor); BUFFER_TRACE(descriptor, "write"); set_buffer_dirty(descriptor); - write_dirty_buffer(descriptor, REQ_SYNC); + write_dirty_buffer(descriptor, JBD2_JOURNAL_REQ_FLAGS); } #endif diff --git a/fs/libfs.c b/fs/libfs.c index 748ac5923154..5b6120b19e99 100644 --- a/fs/libfs.c +++ b/fs/libfs.c @@ -245,9 +245,16 @@ const struct inode_operations simple_dir_inode_operations = { }; EXPORT_SYMBOL(simple_dir_inode_operations); -/* 0 is '.', 1 is '..', so always start with offset 2 or more */ +/* simple_offset_add() never assigns these to a dentry */ enum { - DIR_OFFSET_MIN = 2, + DIR_OFFSET_FIRST = 2, /* Find first real entry */ + DIR_OFFSET_EOD = S32_MAX, +}; + +/* simple_offset_add() allocation range */ +enum { + DIR_OFFSET_MIN = DIR_OFFSET_FIRST + 1, + DIR_OFFSET_MAX = DIR_OFFSET_EOD - 1, }; static void offset_set(struct dentry *dentry, long offset) @@ -291,9 +298,10 @@ int simple_offset_add(struct offset_ctx *octx, struct dentry *dentry) return -EBUSY; ret = mtree_alloc_cyclic(&octx->mt, &offset, dentry, DIR_OFFSET_MIN, - LONG_MAX, &octx->next_offset, GFP_KERNEL); - if (ret < 0) - return ret; + DIR_OFFSET_MAX, &octx->next_offset, + GFP_KERNEL); + if (unlikely(ret < 0)) + return ret == -EBUSY ? -ENOSPC : ret; offset_set(dentry, offset); return 0; @@ -330,38 +338,6 @@ void simple_offset_remove(struct offset_ctx *octx, struct dentry *dentry) } /** - * simple_offset_empty - Check if a dentry can be unlinked - * @dentry: dentry to be tested - * - * Returns 0 if @dentry is a non-empty directory; otherwise returns 1. - */ -int simple_offset_empty(struct dentry *dentry) -{ - struct inode *inode = d_inode(dentry); - struct offset_ctx *octx; - struct dentry *child; - unsigned long index; - int ret = 1; - - if (!inode || !S_ISDIR(inode->i_mode)) - return ret; - - index = DIR_OFFSET_MIN; - octx = inode->i_op->get_offset_ctx(inode); - mt_for_each(&octx->mt, child, index, LONG_MAX) { - spin_lock(&child->d_lock); - if (simple_positive(child)) { - spin_unlock(&child->d_lock); - ret = 0; - break; - } - spin_unlock(&child->d_lock); - } - - return ret; -} - -/** * simple_offset_rename - handle directory offsets for rename * @old_dir: parent directory of source entry * @old_dentry: dentry of source entry @@ -454,14 +430,6 @@ void simple_offset_destroy(struct offset_ctx *octx) mtree_destroy(&octx->mt); } -static int offset_dir_open(struct inode *inode, struct file *file) -{ - struct offset_ctx *ctx = inode->i_op->get_offset_ctx(inode); - - file->private_data = (void *)ctx->next_offset; - return 0; -} - /** * offset_dir_llseek - Advance the read position of a directory descriptor * @file: an open directory whose position is to be updated @@ -475,9 +443,6 @@ static int offset_dir_open(struct inode *inode, struct file *file) */ static loff_t offset_dir_llseek(struct file *file, loff_t offset, int whence) { - struct inode *inode = file->f_inode; - struct offset_ctx *ctx = inode->i_op->get_offset_ctx(inode); - switch (whence) { case SEEK_CUR: offset += file->f_pos; @@ -490,62 +455,89 @@ static loff_t offset_dir_llseek(struct file *file, loff_t offset, int whence) return -EINVAL; } - /* In this case, ->private_data is protected by f_pos_lock */ - if (!offset) - file->private_data = (void *)ctx->next_offset; return vfs_setpos(file, offset, LONG_MAX); } -static struct dentry *offset_find_next(struct offset_ctx *octx, loff_t offset) +static struct dentry *find_positive_dentry(struct dentry *parent, + struct dentry *dentry, + bool next) { - MA_STATE(mas, &octx->mt, offset, offset); + struct dentry *found = NULL; + + spin_lock(&parent->d_lock); + if (next) + dentry = d_next_sibling(dentry); + else if (!dentry) + dentry = d_first_child(parent); + hlist_for_each_entry_from(dentry, d_sib) { + if (!simple_positive(dentry)) + continue; + spin_lock_nested(&dentry->d_lock, DENTRY_D_LOCK_NESTED); + if (simple_positive(dentry)) + found = dget_dlock(dentry); + spin_unlock(&dentry->d_lock); + if (likely(found)) + break; + } + spin_unlock(&parent->d_lock); + return found; +} + +static noinline_for_stack struct dentry * +offset_dir_lookup(struct dentry *parent, loff_t offset) +{ + struct inode *inode = d_inode(parent); + struct offset_ctx *octx = inode->i_op->get_offset_ctx(inode); struct dentry *child, *found = NULL; - rcu_read_lock(); - child = mas_find(&mas, LONG_MAX); - if (!child) - goto out; - spin_lock(&child->d_lock); - if (simple_positive(child)) - found = dget_dlock(child); - spin_unlock(&child->d_lock); -out: - rcu_read_unlock(); + MA_STATE(mas, &octx->mt, offset, offset); + + if (offset == DIR_OFFSET_FIRST) + found = find_positive_dentry(parent, NULL, false); + else { + rcu_read_lock(); + child = mas_find(&mas, DIR_OFFSET_MAX); + found = find_positive_dentry(parent, child, false); + rcu_read_unlock(); + } return found; } static bool offset_dir_emit(struct dir_context *ctx, struct dentry *dentry) { struct inode *inode = d_inode(dentry); - long offset = dentry2offset(dentry); - return ctx->actor(ctx, dentry->d_name.name, dentry->d_name.len, offset, - inode->i_ino, fs_umode_to_dtype(inode->i_mode)); + return dir_emit(ctx, dentry->d_name.name, dentry->d_name.len, + inode->i_ino, fs_umode_to_dtype(inode->i_mode)); } -static void offset_iterate_dir(struct inode *inode, struct dir_context *ctx, long last_index) +static void offset_iterate_dir(struct file *file, struct dir_context *ctx) { - struct offset_ctx *octx = inode->i_op->get_offset_ctx(inode); + struct dentry *dir = file->f_path.dentry; struct dentry *dentry; + dentry = offset_dir_lookup(dir, ctx->pos); + if (!dentry) + goto out_eod; while (true) { - dentry = offset_find_next(octx, ctx->pos); - if (!dentry) - return; - - if (dentry2offset(dentry) >= last_index) { - dput(dentry); - return; - } + struct dentry *next; - if (!offset_dir_emit(ctx, dentry)) { - dput(dentry); - return; - } + ctx->pos = dentry2offset(dentry); + if (!offset_dir_emit(ctx, dentry)) + break; - ctx->pos = dentry2offset(dentry) + 1; + next = find_positive_dentry(dir, dentry, true); dput(dentry); + + if (!next) + goto out_eod; + dentry = next; } + dput(dentry); + return; + +out_eod: + ctx->pos = DIR_OFFSET_EOD; } /** @@ -565,6 +557,8 @@ static void offset_iterate_dir(struct inode *inode, struct dir_context *ctx, lon * * On return, @ctx->pos contains an offset that will read the next entry * in this directory when offset_readdir() is called again with @ctx. + * Caller places this value in the d_off field of the last entry in the + * user's buffer. * * Return values: * %0 - Complete @@ -572,19 +566,17 @@ static void offset_iterate_dir(struct inode *inode, struct dir_context *ctx, lon static int offset_readdir(struct file *file, struct dir_context *ctx) { struct dentry *dir = file->f_path.dentry; - long last_index = (long)file->private_data; lockdep_assert_held(&d_inode(dir)->i_rwsem); if (!dir_emit_dots(file, ctx)) return 0; - - offset_iterate_dir(d_inode(dir), ctx, last_index); + if (ctx->pos != DIR_OFFSET_EOD) + offset_iterate_dir(file, ctx); return 0; } const struct file_operations simple_offset_dir_operations = { - .open = offset_dir_open, .llseek = offset_dir_llseek, .iterate_shared = offset_readdir, .read = generic_read_dir, @@ -673,6 +665,7 @@ static int pseudo_fs_fill_super(struct super_block *s, struct fs_context *fc) s->s_blocksize_bits = PAGE_SHIFT; s->s_magic = ctx->magic; s->s_op = ctx->ops ?: &simple_super_operations; + s->s_export_op = ctx->eops; s->s_xattr = ctx->xattr; s->s_time_gran = 1; root = new_inode(s); diff --git a/fs/mount.h b/fs/mount.h index 185fc56afc13..ffb613cdfeee 100644 --- a/fs/mount.h +++ b/fs/mount.h @@ -8,15 +8,23 @@ struct mnt_namespace { struct ns_common ns; struct mount * root; - struct rb_root mounts; /* Protected by namespace_sem */ + struct { + struct rb_root mounts; /* Protected by namespace_sem */ + struct rb_node *mnt_last_node; /* last (rightmost) mount in the rbtree */ + struct rb_node *mnt_first_node; /* first (leftmost) mount in the rbtree */ + }; struct user_namespace *user_ns; struct ucounts *ucounts; u64 seq; /* Sequence number to prevent loops */ - wait_queue_head_t poll; + union { + wait_queue_head_t poll; + struct rcu_head mnt_ns_rcu; + }; u64 event; unsigned int nr_mounts; /* # of mounts in the namespace */ unsigned int pending_mounts; struct rb_node mnt_ns_tree_node; /* node in the mnt_ns_tree */ + struct list_head mnt_ns_list; /* entry in the sequential list of mounts namespace */ refcount_t passive; /* number references not pinning @mounts */ } __randomize_layout; @@ -38,6 +46,7 @@ struct mount { struct dentry *mnt_mountpoint; struct vfsmount mnt; union { + struct rb_node mnt_node; /* node in the ns->mounts rbtree */ struct rcu_head mnt_rcu; struct llist_node mnt_llist; }; @@ -51,10 +60,7 @@ struct mount { struct list_head mnt_child; /* and going through their mnt_child */ struct list_head mnt_instance; /* mount instance on sb->s_mounts */ const char *mnt_devname; /* Name of device e.g. /dev/dsk/hda1 */ - union { - struct rb_node mnt_node; /* Under ns->mounts */ - struct list_head mnt_list; - }; + struct list_head mnt_list; struct list_head mnt_expire; /* link in fs-specific expiry list */ struct list_head mnt_share; /* circular list of shared mounts */ struct list_head mnt_slave_list;/* list of slave mounts */ @@ -145,24 +151,28 @@ static inline bool is_anon_ns(struct mnt_namespace *ns) return ns->seq == 0; } +static inline bool mnt_ns_attached(const struct mount *mnt) +{ + return !RB_EMPTY_NODE(&mnt->mnt_node); +} + static inline void move_from_ns(struct mount *mnt, struct list_head *dt_list) { - WARN_ON(!(mnt->mnt.mnt_flags & MNT_ONRB)); - mnt->mnt.mnt_flags &= ~MNT_ONRB; - rb_erase(&mnt->mnt_node, &mnt->mnt_ns->mounts); + struct mnt_namespace *ns = mnt->mnt_ns; + WARN_ON(!mnt_ns_attached(mnt)); + if (ns->mnt_last_node == &mnt->mnt_node) + ns->mnt_last_node = rb_prev(&mnt->mnt_node); + if (ns->mnt_first_node == &mnt->mnt_node) + ns->mnt_first_node = rb_next(&mnt->mnt_node); + rb_erase(&mnt->mnt_node, &ns->mounts); + RB_CLEAR_NODE(&mnt->mnt_node); list_add_tail(&mnt->mnt_list, dt_list); } bool has_locked_children(struct mount *mnt, struct dentry *dentry); -struct mnt_namespace *__lookup_next_mnt_ns(struct mnt_namespace *mnt_ns, bool previous); -static inline struct mnt_namespace *lookup_next_mnt_ns(struct mnt_namespace *mntns) -{ - return __lookup_next_mnt_ns(mntns, false); -} -static inline struct mnt_namespace *lookup_prev_mnt_ns(struct mnt_namespace *mntns) -{ - return __lookup_next_mnt_ns(mntns, true); -} +struct mnt_namespace *get_sequential_mnt_ns(struct mnt_namespace *mnt_ns, + bool previous); + static inline struct mnt_namespace *to_mnt_ns(struct ns_common *ns) { return container_of(ns, struct mnt_namespace, ns); diff --git a/fs/namei.c b/fs/namei.c index 9d30c7aa9aa6..e56c29a22d26 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -5272,19 +5272,16 @@ SYSCALL_DEFINE2(rename, const char __user *, oldname, const char __user *, newna getname(newname), 0); } -int readlink_copy(char __user *buffer, int buflen, const char *link) +int readlink_copy(char __user *buffer, int buflen, const char *link, int linklen) { - int len = PTR_ERR(link); - if (IS_ERR(link)) - goto out; + int copylen; - len = strlen(link); - if (len > (unsigned) buflen) - len = buflen; - if (copy_to_user(buffer, link, len)) - len = -EFAULT; -out: - return len; + copylen = linklen; + if (unlikely(copylen > (unsigned) buflen)) + copylen = buflen; + if (copy_to_user(buffer, link, copylen)) + copylen = -EFAULT; + return copylen; } /** @@ -5304,6 +5301,9 @@ int vfs_readlink(struct dentry *dentry, char __user *buffer, int buflen) const char *link; int res; + if (inode->i_opflags & IOP_CACHED_LINK) + return readlink_copy(buffer, buflen, inode->i_link, inode->i_linklen); + if (unlikely(!(inode->i_opflags & IOP_DEFAULT_READLINK))) { if (unlikely(inode->i_op->readlink)) return inode->i_op->readlink(dentry, buffer, buflen); @@ -5322,7 +5322,7 @@ int vfs_readlink(struct dentry *dentry, char __user *buffer, int buflen) if (IS_ERR(link)) return PTR_ERR(link); } - res = readlink_copy(buffer, buflen, link); + res = readlink_copy(buffer, buflen, link, strlen(link)); do_delayed_call(&done); return res; } @@ -5391,10 +5391,14 @@ EXPORT_SYMBOL(page_put_link); int page_readlink(struct dentry *dentry, char __user *buffer, int buflen) { + const char *link; + int res; + DEFINE_DELAYED_CALL(done); - int res = readlink_copy(buffer, buflen, - page_get_link(dentry, d_inode(dentry), - &done)); + link = page_get_link(dentry, d_inode(dentry), &done); + res = PTR_ERR(link); + if (!IS_ERR(link)) + res = readlink_copy(buffer, buflen, link, strlen(link)); do_delayed_call(&done); return res; } diff --git a/fs/namespace.c b/fs/namespace.c index 23e81c2a1e3f..4013fbac354a 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -32,7 +32,7 @@ #include <linux/fs_context.h> #include <linux/shmem_fs.h> #include <linux/mnt_idmapping.h> -#include <linux/nospec.h> +#include <linux/pidfs.h> #include "pnode.h" #include "internal.h" @@ -66,12 +66,12 @@ static int __init set_mphash_entries(char *str) __setup("mphash_entries=", set_mphash_entries); static u64 event; -static DEFINE_IDA(mnt_id_ida); +static DEFINE_XARRAY_FLAGS(mnt_id_xa, XA_FLAGS_ALLOC); static DEFINE_IDA(mnt_group_ida); /* Don't allow confusion with old 32bit mount ID */ #define MNT_UNIQUE_ID_OFFSET (1ULL << 31) -static atomic64_t mnt_id_ctr = ATOMIC64_INIT(MNT_UNIQUE_ID_OFFSET); +static u64 mnt_id_ctr = MNT_UNIQUE_ID_OFFSET; static struct hlist_head *mount_hashtable __ro_after_init; static struct hlist_head *mountpoint_hashtable __ro_after_init; @@ -79,8 +79,10 @@ static struct kmem_cache *mnt_cache __ro_after_init; static DECLARE_RWSEM(namespace_sem); static HLIST_HEAD(unmounted); /* protected by namespace_sem */ static LIST_HEAD(ex_mountpoints); /* protected by namespace_sem */ -static DEFINE_RWLOCK(mnt_ns_tree_lock); +static DEFINE_SEQLOCK(mnt_ns_tree_lock); + static struct rb_root mnt_ns_tree = RB_ROOT; /* protected by mnt_ns_tree_lock */ +static LIST_HEAD(mnt_ns_list); /* protected by mnt_ns_tree_lock */ struct mount_kattr { unsigned int attr_set; @@ -106,17 +108,6 @@ EXPORT_SYMBOL_GPL(fs_kobj); */ __cacheline_aligned_in_smp DEFINE_SEQLOCK(mount_lock); -static int mnt_ns_cmp(u64 seq, const struct mnt_namespace *ns) -{ - u64 seq_b = ns->seq; - - if (seq < seq_b) - return -1; - if (seq > seq_b) - return 1; - return 0; -} - static inline struct mnt_namespace *node_to_mnt_ns(const struct rb_node *node) { if (!node) @@ -124,25 +115,52 @@ static inline struct mnt_namespace *node_to_mnt_ns(const struct rb_node *node) return rb_entry(node, struct mnt_namespace, mnt_ns_tree_node); } -static bool mnt_ns_less(struct rb_node *a, const struct rb_node *b) +static int mnt_ns_cmp(struct rb_node *a, const struct rb_node *b) { struct mnt_namespace *ns_a = node_to_mnt_ns(a); struct mnt_namespace *ns_b = node_to_mnt_ns(b); u64 seq_a = ns_a->seq; + u64 seq_b = ns_b->seq; - return mnt_ns_cmp(seq_a, ns_b) < 0; + if (seq_a < seq_b) + return -1; + if (seq_a > seq_b) + return 1; + return 0; +} + +static inline void mnt_ns_tree_write_lock(void) +{ + write_seqlock(&mnt_ns_tree_lock); +} + +static inline void mnt_ns_tree_write_unlock(void) +{ + write_sequnlock(&mnt_ns_tree_lock); } static void mnt_ns_tree_add(struct mnt_namespace *ns) { - guard(write_lock)(&mnt_ns_tree_lock); - rb_add(&ns->mnt_ns_tree_node, &mnt_ns_tree, mnt_ns_less); + struct rb_node *node, *prev; + + mnt_ns_tree_write_lock(); + node = rb_find_add_rcu(&ns->mnt_ns_tree_node, &mnt_ns_tree, mnt_ns_cmp); + /* + * If there's no previous entry simply add it after the + * head and if there is add it after the previous entry. + */ + prev = rb_prev(&ns->mnt_ns_tree_node); + if (!prev) + list_add_rcu(&ns->mnt_ns_list, &mnt_ns_list); + else + list_add_rcu(&ns->mnt_ns_list, &node_to_mnt_ns(prev)->mnt_ns_list); + mnt_ns_tree_write_unlock(); + + WARN_ON_ONCE(node); } static void mnt_ns_release(struct mnt_namespace *ns) { - lockdep_assert_not_held(&mnt_ns_tree_lock); - /* keep alive for {list,stat}mount() */ if (refcount_dec_and_test(&ns->passive)) { put_user_ns(ns->user_ns); @@ -151,41 +169,34 @@ static void mnt_ns_release(struct mnt_namespace *ns) } DEFINE_FREE(mnt_ns_release, struct mnt_namespace *, if (_T) mnt_ns_release(_T)) +static void mnt_ns_release_rcu(struct rcu_head *rcu) +{ + mnt_ns_release(container_of(rcu, struct mnt_namespace, mnt_ns_rcu)); +} + static void mnt_ns_tree_remove(struct mnt_namespace *ns) { /* remove from global mount namespace list */ if (!is_anon_ns(ns)) { - guard(write_lock)(&mnt_ns_tree_lock); + mnt_ns_tree_write_lock(); rb_erase(&ns->mnt_ns_tree_node, &mnt_ns_tree); + list_bidir_del_rcu(&ns->mnt_ns_list); + mnt_ns_tree_write_unlock(); } - mnt_ns_release(ns); + call_rcu(&ns->mnt_ns_rcu, mnt_ns_release_rcu); } -/* - * Returns the mount namespace which either has the specified id, or has the - * next smallest id afer the specified one. - */ -static struct mnt_namespace *mnt_ns_find_id_at(u64 mnt_ns_id) +static int mnt_ns_find(const void *key, const struct rb_node *node) { - struct rb_node *node = mnt_ns_tree.rb_node; - struct mnt_namespace *ret = NULL; - - lockdep_assert_held(&mnt_ns_tree_lock); + const u64 mnt_ns_id = *(u64 *)key; + const struct mnt_namespace *ns = node_to_mnt_ns(node); - while (node) { - struct mnt_namespace *n = node_to_mnt_ns(node); - - if (mnt_ns_id <= n->seq) { - ret = node_to_mnt_ns(node); - if (mnt_ns_id == n->seq) - break; - node = node->rb_left; - } else { - node = node->rb_right; - } - } - return ret; + if (mnt_ns_id < ns->seq) + return -1; + if (mnt_ns_id > ns->seq) + return 1; + return 0; } /* @@ -195,18 +206,37 @@ static struct mnt_namespace *mnt_ns_find_id_at(u64 mnt_ns_id) * namespace the @namespace_sem must first be acquired. If the namespace has * already shut down before acquiring @namespace_sem, {list,stat}mount() will * see that the mount rbtree of the namespace is empty. + * + * Note the lookup is lockless protected by a sequence counter. We only + * need to guard against false negatives as false positives aren't + * possible. So if we didn't find a mount namespace and the sequence + * counter has changed we need to retry. If the sequence counter is + * still the same we know the search actually failed. */ static struct mnt_namespace *lookup_mnt_ns(u64 mnt_ns_id) { - struct mnt_namespace *ns; + struct mnt_namespace *ns; + struct rb_node *node; + unsigned int seq; - guard(read_lock)(&mnt_ns_tree_lock); - ns = mnt_ns_find_id_at(mnt_ns_id); - if (!ns || ns->seq != mnt_ns_id) - return NULL; + guard(rcu)(); + do { + seq = read_seqbegin(&mnt_ns_tree_lock); + node = rb_find_rcu(&mnt_ns_id, &mnt_ns_tree, mnt_ns_find); + if (node) + break; + } while (read_seqretry(&mnt_ns_tree_lock, seq)); - refcount_inc(&ns->passive); - return ns; + if (!node) + return NULL; + + /* + * The last reference count is put with RCU delay so we can + * unconditonally acquire a reference here. + */ + ns = node_to_mnt_ns(node); + refcount_inc(&ns->passive); + return ns; } static inline void lock_mount_hash(void) @@ -236,18 +266,19 @@ static inline struct hlist_head *mp_hash(struct dentry *dentry) static int mnt_alloc_id(struct mount *mnt) { - int res = ida_alloc(&mnt_id_ida, GFP_KERNEL); + int res; - if (res < 0) - return res; - mnt->mnt_id = res; - mnt->mnt_id_unique = atomic64_inc_return(&mnt_id_ctr); - return 0; + xa_lock(&mnt_id_xa); + res = __xa_alloc(&mnt_id_xa, &mnt->mnt_id, mnt, XA_LIMIT(1, INT_MAX), GFP_KERNEL); + if (!res) + mnt->mnt_id_unique = ++mnt_id_ctr; + xa_unlock(&mnt_id_xa); + return res; } static void mnt_free_id(struct mount *mnt) { - ida_free(&mnt_id_ida, mnt->mnt_id); + xa_erase(&mnt_id_xa, mnt->mnt_id); } /* @@ -344,6 +375,7 @@ static struct mount *alloc_vfsmnt(const char *name) INIT_HLIST_NODE(&mnt->mnt_mp_list); INIT_LIST_HEAD(&mnt->mnt_umounting); INIT_HLIST_HEAD(&mnt->mnt_stuck_children); + RB_CLEAR_NODE(&mnt->mnt_node); mnt->mnt.mnt_idmap = &nop_mnt_idmap; } return mnt; @@ -1123,19 +1155,27 @@ static void mnt_add_to_ns(struct mnt_namespace *ns, struct mount *mnt) { struct rb_node **link = &ns->mounts.rb_node; struct rb_node *parent = NULL; + bool mnt_first_node = true, mnt_last_node = true; - WARN_ON(mnt->mnt.mnt_flags & MNT_ONRB); + WARN_ON(mnt_ns_attached(mnt)); mnt->mnt_ns = ns; while (*link) { parent = *link; - if (mnt->mnt_id_unique < node_to_mount(parent)->mnt_id_unique) + if (mnt->mnt_id_unique < node_to_mount(parent)->mnt_id_unique) { link = &parent->rb_left; - else + mnt_last_node = false; + } else { link = &parent->rb_right; + mnt_first_node = false; + } } + + if (mnt_last_node) + ns->mnt_last_node = &mnt->mnt_node; + if (mnt_first_node) + ns->mnt_first_node = &mnt->mnt_node; rb_link_node(&mnt->mnt_node, parent, link); rb_insert_color(&mnt->mnt_node, &ns->mounts); - mnt->mnt.mnt_flags |= MNT_ONRB; } /* @@ -1305,7 +1345,7 @@ static struct mount *clone_mnt(struct mount *old, struct dentry *root, } mnt->mnt.mnt_flags = old->mnt.mnt_flags; - mnt->mnt.mnt_flags &= ~(MNT_WRITE_HOLD|MNT_MARKED|MNT_INTERNAL|MNT_ONRB); + mnt->mnt.mnt_flags &= ~(MNT_WRITE_HOLD|MNT_MARKED|MNT_INTERNAL); atomic_inc(&sb->s_active); mnt->mnt.mnt_idmap = mnt_idmap_get(mnt_idmap(&old->mnt)); @@ -1763,7 +1803,7 @@ static void umount_tree(struct mount *mnt, enum umount_tree_flags how) /* Gather the mounts to umount */ for (p = mnt; p; p = next_mnt(p, mnt)) { p->mnt.mnt_flags |= MNT_UMOUNT; - if (p->mnt.mnt_flags & MNT_ONRB) + if (mnt_ns_attached(p)) move_from_ns(p, &tmp_list); else list_move(&p->mnt_list, &tmp_list); @@ -1912,16 +1952,14 @@ static int do_umount(struct mount *mnt, int flags) event++; if (flags & MNT_DETACH) { - if (mnt->mnt.mnt_flags & MNT_ONRB || - !list_empty(&mnt->mnt_list)) + if (mnt_ns_attached(mnt) || !list_empty(&mnt->mnt_list)) umount_tree(mnt, UMOUNT_PROPAGATE); retval = 0; } else { shrink_submounts(mnt); retval = -EBUSY; if (!propagate_mount_busy(mnt, 2)) { - if (mnt->mnt.mnt_flags & MNT_ONRB || - !list_empty(&mnt->mnt_list)) + if (mnt_ns_attached(mnt) || !list_empty(&mnt->mnt_list)) umount_tree(mnt, UMOUNT_PROPAGATE|UMOUNT_SYNC); retval = 0; } @@ -2055,9 +2093,15 @@ SYSCALL_DEFINE1(oldumount, char __user *, name) static bool is_mnt_ns_file(struct dentry *dentry) { + struct ns_common *ns; + /* Is this a proxy for a mount namespace? */ - return dentry->d_op == &ns_dentry_operations && - dentry->d_fsdata == &mntns_operations; + if (dentry->d_op != &ns_dentry_operations) + return false; + + ns = d_inode(dentry)->i_private; + + return ns->ops == &mntns_operations; } struct ns_common *from_mnt_ns(struct mnt_namespace *mnt) @@ -2065,30 +2109,34 @@ struct ns_common *from_mnt_ns(struct mnt_namespace *mnt) return &mnt->ns; } -struct mnt_namespace *__lookup_next_mnt_ns(struct mnt_namespace *mntns, bool previous) +struct mnt_namespace *get_sequential_mnt_ns(struct mnt_namespace *mntns, bool previous) { - guard(read_lock)(&mnt_ns_tree_lock); + guard(rcu)(); + for (;;) { - struct rb_node *node; + struct list_head *list; if (previous) - node = rb_prev(&mntns->mnt_ns_tree_node); + list = rcu_dereference(list_bidir_prev_rcu(&mntns->mnt_ns_list)); else - node = rb_next(&mntns->mnt_ns_tree_node); - if (!node) + list = rcu_dereference(list_next_rcu(&mntns->mnt_ns_list)); + if (list_is_head(list, &mnt_ns_list)) return ERR_PTR(-ENOENT); - mntns = node_to_mnt_ns(node); - node = &mntns->mnt_ns_tree_node; + mntns = list_entry_rcu(list, struct mnt_namespace, mnt_ns_list); + /* + * The last passive reference count is put with RCU + * delay so accessing the mount namespace is not just + * safe but all relevant members are still valid. + */ if (!ns_capable_noaudit(mntns->user_ns, CAP_SYS_ADMIN)) continue; /* - * Holding mnt_ns_tree_lock prevents the mount namespace from - * being freed but it may well be on it's deathbed. We want an - * active reference, not just a passive one here as we're - * persisting the mount namespace. + * We need an active reference count as we're persisting + * the mount namespace and it might already be on its + * deathbed. */ if (!refcount_inc_not_zero(&mntns->ns.count)) continue; @@ -2732,8 +2780,13 @@ static struct mount *__do_loopback(struct path *old_path, int recurse) if (IS_MNT_UNBINDABLE(old)) return mnt; - if (!check_mnt(old) && old_path->dentry->d_op != &ns_dentry_operations) - return mnt; + if (!check_mnt(old)) { + const struct dentry_operations *d_op = old_path->dentry->d_op; + + if (d_op != &ns_dentry_operations && + d_op != &pidfs_dentry_operations) + return mnt; + } if (!recurse && has_locked_children(old, old_path->dentry)) return mnt; @@ -3835,7 +3888,7 @@ int path_mount(const char *dev_name, struct path *path, data_page); } -long do_mount(const char *dev_name, const char __user *dir_name, +int do_mount(const char *dev_name, const char __user *dir_name, const char *type_page, unsigned long flags, void *data_page) { struct path path; @@ -3905,6 +3958,7 @@ static struct mnt_namespace *alloc_mnt_ns(struct user_namespace *user_ns, bool a refcount_set(&new_ns->ns.count, 1); refcount_set(&new_ns->passive, 1); new_ns->mounts = RB_ROOT; + INIT_LIST_HEAD(&new_ns->mnt_ns_list); RB_CLEAR_NODE(&new_ns->mnt_ns_tree_node); init_waitqueue_head(&new_ns->poll); new_ns->user_ns = get_user_ns(user_ns); @@ -3984,7 +4038,6 @@ struct mnt_namespace *copy_mnt_ns(unsigned long flags, struct mnt_namespace *ns, while (p->mnt.mnt_root != q->mnt.mnt_root) p = next_mnt(skip_mnt_tree(p), old); } - mnt_ns_tree_add(new_ns); namespace_unlock(); if (rootmnt) @@ -3992,6 +4045,7 @@ struct mnt_namespace *copy_mnt_ns(unsigned long flags, struct mnt_namespace *ns, if (pwdmnt) mntput(pwdmnt); + mnt_ns_tree_add(new_ns); return new_ns; } @@ -5038,6 +5092,10 @@ static int statmount_mnt_opts(struct kstatmount *s, struct seq_file *seq) if (sb->s_op->show_options) { size_t start = seq->count; + err = security_sb_show_options(seq, sb); + if (err) + return err; + err = sb->s_op->show_options(seq, mnt->mnt_root); if (err) return err; @@ -5525,9 +5583,9 @@ static ssize_t do_listmount(struct mnt_namespace *ns, u64 mnt_parent_id, if (!last_mnt_id) { if (reverse) - first = node_to_mount(rb_last(&ns->mounts)); + first = node_to_mount(ns->mnt_last_node); else - first = node_to_mount(rb_first(&ns->mounts)); + first = node_to_mount(ns->mnt_first_node); } else { if (reverse) first = mnt_find_id_at_reverse(ns, last_mnt_id - 1); diff --git a/fs/netfs/Makefile b/fs/netfs/Makefile index d08b0bfb6756..b43188d64bd8 100644 --- a/fs/netfs/Makefile +++ b/fs/netfs/Makefile @@ -13,8 +13,11 @@ netfs-y := \ read_collect.o \ read_pgpriv2.o \ read_retry.o \ + read_single.o \ + rolling_buffer.o \ write_collect.o \ - write_issue.o + write_issue.o \ + write_retry.o netfs-$(CONFIG_NETFS_STATS) += stats.o diff --git a/fs/netfs/buffered_read.c b/fs/netfs/buffered_read.c index 7ac34550c403..f761d44b3436 100644 --- a/fs/netfs/buffered_read.c +++ b/fs/netfs/buffered_read.c @@ -64,37 +64,6 @@ static int netfs_begin_cache_read(struct netfs_io_request *rreq, struct netfs_in } /* - * Decant the list of folios to read into a rolling buffer. - */ -static size_t netfs_load_buffer_from_ra(struct netfs_io_request *rreq, - struct folio_queue *folioq, - struct folio_batch *put_batch) -{ - unsigned int order, nr; - size_t size = 0; - - nr = __readahead_batch(rreq->ractl, (struct page **)folioq->vec.folios, - ARRAY_SIZE(folioq->vec.folios)); - folioq->vec.nr = nr; - for (int i = 0; i < nr; i++) { - struct folio *folio = folioq_folio(folioq, i); - - trace_netfs_folio(folio, netfs_folio_trace_read); - order = folio_order(folio); - folioq->orders[i] = order; - size += PAGE_SIZE << order; - - if (!folio_batch_add(put_batch, folio)) - folio_batch_release(put_batch); - } - - for (int i = nr; i < folioq_nr_slots(folioq); i++) - folioq_clear(folioq, i); - - return size; -} - -/* * netfs_prepare_read_iterator - Prepare the subreq iterator for I/O * @subreq: The subrequest to be set up * @@ -128,19 +97,12 @@ static ssize_t netfs_prepare_read_iterator(struct netfs_io_subrequest *subreq) folio_batch_init(&put_batch); while (rreq->submitted < subreq->start + rsize) { - struct folio_queue *tail = rreq->buffer_tail, *new; - size_t added; - - new = kmalloc(sizeof(*new), GFP_NOFS); - if (!new) - return -ENOMEM; - netfs_stat(&netfs_n_folioq); - folioq_init(new); - new->prev = tail; - tail->next = new; - rreq->buffer_tail = new; - added = netfs_load_buffer_from_ra(rreq, new, &put_batch); - rreq->iter.count += added; + ssize_t added; + + added = rolling_buffer_load_from_ra(&rreq->buffer, rreq->ractl, + &put_batch); + if (added < 0) + return added; rreq->submitted += added; } folio_batch_release(&put_batch); @@ -148,7 +110,7 @@ static ssize_t netfs_prepare_read_iterator(struct netfs_io_subrequest *subreq) subreq->len = rsize; if (unlikely(rreq->io_streams[0].sreq_max_segs)) { - size_t limit = netfs_limit_iter(&rreq->iter, 0, rsize, + size_t limit = netfs_limit_iter(&rreq->buffer.iter, 0, rsize, rreq->io_streams[0].sreq_max_segs); if (limit < rsize) { @@ -157,20 +119,10 @@ static ssize_t netfs_prepare_read_iterator(struct netfs_io_subrequest *subreq) } } - subreq->io_iter = rreq->iter; - - if (iov_iter_is_folioq(&subreq->io_iter)) { - if (subreq->io_iter.folioq_slot >= folioq_nr_slots(subreq->io_iter.folioq)) { - subreq->io_iter.folioq = subreq->io_iter.folioq->next; - subreq->io_iter.folioq_slot = 0; - } - subreq->curr_folioq = (struct folio_queue *)subreq->io_iter.folioq; - subreq->curr_folioq_slot = subreq->io_iter.folioq_slot; - subreq->curr_folio_order = subreq->curr_folioq->orders[subreq->curr_folioq_slot]; - } + subreq->io_iter = rreq->buffer.iter; iov_iter_truncate(&subreq->io_iter, subreq->len); - iov_iter_advance(&rreq->iter, subreq->len); + rolling_buffer_advance(&rreq->buffer, subreq->len); return subreq->len; } @@ -179,25 +131,14 @@ static enum netfs_io_source netfs_cache_prepare_read(struct netfs_io_request *rr loff_t i_size) { struct netfs_cache_resources *cres = &rreq->cache_resources; + enum netfs_io_source source; if (!cres->ops) return NETFS_DOWNLOAD_FROM_SERVER; - return cres->ops->prepare_read(subreq, i_size); -} - -static void netfs_cache_read_terminated(void *priv, ssize_t transferred_or_error, - bool was_async) -{ - struct netfs_io_subrequest *subreq = priv; - - if (transferred_or_error < 0) { - netfs_read_subreq_terminated(subreq, transferred_or_error, was_async); - return; - } + source = cres->ops->prepare_read(subreq, i_size); + trace_netfs_sreq(subreq, netfs_sreq_trace_prepare); + return source; - if (transferred_or_error > 0) - subreq->transferred += transferred_or_error; - netfs_read_subreq_terminated(subreq, 0, was_async); } /* @@ -214,6 +155,47 @@ static void netfs_read_cache_to_pagecache(struct netfs_io_request *rreq, netfs_cache_read_terminated, subreq); } +static void netfs_issue_read(struct netfs_io_request *rreq, + struct netfs_io_subrequest *subreq) +{ + struct netfs_io_stream *stream = &rreq->io_streams[0]; + + __set_bit(NETFS_SREQ_IN_PROGRESS, &subreq->flags); + + /* We add to the end of the list whilst the collector may be walking + * the list. The collector only goes nextwards and uses the lock to + * remove entries off of the front. + */ + spin_lock(&rreq->lock); + list_add_tail(&subreq->rreq_link, &stream->subrequests); + if (list_is_first(&subreq->rreq_link, &stream->subrequests)) { + stream->front = subreq; + if (!stream->active) { + stream->collected_to = stream->front->start; + /* Store list pointers before active flag */ + smp_store_release(&stream->active, true); + } + } + + spin_unlock(&rreq->lock); + + switch (subreq->source) { + case NETFS_DOWNLOAD_FROM_SERVER: + rreq->netfs_ops->issue_read(subreq); + break; + case NETFS_READ_FROM_CACHE: + netfs_read_cache_to_pagecache(rreq, subreq); + break; + default: + __set_bit(NETFS_SREQ_CLEAR_TAIL, &subreq->flags); + subreq->error = 0; + iov_iter_zero(subreq->len, &subreq->io_iter); + subreq->transferred = subreq->len; + netfs_read_subreq_terminated(subreq); + break; + } +} + /* * Perform a read to the pagecache from a series of sources of different types, * slicing up the region to be read according to available cache blocks and @@ -226,11 +208,9 @@ static void netfs_read_to_pagecache(struct netfs_io_request *rreq) ssize_t size = rreq->len; int ret = 0; - atomic_inc(&rreq->nr_outstanding); - do { struct netfs_io_subrequest *subreq; - enum netfs_io_source source = NETFS_DOWNLOAD_FROM_SERVER; + enum netfs_io_source source = NETFS_SOURCE_UNKNOWN; ssize_t slice; subreq = netfs_alloc_subrequest(rreq); @@ -242,20 +222,14 @@ static void netfs_read_to_pagecache(struct netfs_io_request *rreq) subreq->start = start; subreq->len = size; - atomic_inc(&rreq->nr_outstanding); - spin_lock_bh(&rreq->lock); - list_add_tail(&subreq->rreq_link, &rreq->subrequests); - subreq->prev_donated = rreq->prev_donated; - rreq->prev_donated = 0; - trace_netfs_sreq(subreq, netfs_sreq_trace_added); - spin_unlock_bh(&rreq->lock); - source = netfs_cache_prepare_read(rreq, subreq, rreq->i_size); subreq->source = source; if (source == NETFS_DOWNLOAD_FROM_SERVER) { unsigned long long zp = umin(ictx->zero_point, rreq->i_size); size_t len = subreq->len; + if (unlikely(rreq->origin == NETFS_READ_SINGLE)) + zp = rreq->i_size; if (subreq->start >= zp) { subreq->source = source = NETFS_FILL_WITH_ZEROES; goto fill_with_zeroes; @@ -276,24 +250,17 @@ static void netfs_read_to_pagecache(struct netfs_io_request *rreq) if (rreq->netfs_ops->prepare_read) { ret = rreq->netfs_ops->prepare_read(subreq); if (ret < 0) { - atomic_dec(&rreq->nr_outstanding); + subreq->error = ret; + /* Not queued - release both refs. */ + netfs_put_subrequest(subreq, false, + netfs_sreq_trace_put_cancel); netfs_put_subrequest(subreq, false, netfs_sreq_trace_put_cancel); break; } trace_netfs_sreq(subreq, netfs_sreq_trace_prepare); } - - slice = netfs_prepare_read_iterator(subreq); - if (slice < 0) { - atomic_dec(&rreq->nr_outstanding); - netfs_put_subrequest(subreq, false, netfs_sreq_trace_put_cancel); - ret = slice; - break; - } - - rreq->netfs_ops->issue_read(subreq); - goto done; + goto issue; } fill_with_zeroes: @@ -301,82 +268,50 @@ static void netfs_read_to_pagecache(struct netfs_io_request *rreq) subreq->source = NETFS_FILL_WITH_ZEROES; trace_netfs_sreq(subreq, netfs_sreq_trace_submit); netfs_stat(&netfs_n_rh_zero); - slice = netfs_prepare_read_iterator(subreq); - __set_bit(NETFS_SREQ_CLEAR_TAIL, &subreq->flags); - netfs_read_subreq_terminated(subreq, 0, false); - goto done; + goto issue; } if (source == NETFS_READ_FROM_CACHE) { trace_netfs_sreq(subreq, netfs_sreq_trace_submit); - slice = netfs_prepare_read_iterator(subreq); - netfs_read_cache_to_pagecache(rreq, subreq); - goto done; + goto issue; } pr_err("Unexpected read source %u\n", source); WARN_ON_ONCE(1); break; - done: + issue: + slice = netfs_prepare_read_iterator(subreq); + if (slice < 0) { + ret = slice; + subreq->error = ret; + trace_netfs_sreq(subreq, netfs_sreq_trace_cancel); + /* Not queued - release both refs. */ + netfs_put_subrequest(subreq, false, netfs_sreq_trace_put_cancel); + netfs_put_subrequest(subreq, false, netfs_sreq_trace_put_cancel); + break; + } size -= slice; start += slice; + if (size <= 0) { + smp_wmb(); /* Write lists before ALL_QUEUED. */ + set_bit(NETFS_RREQ_ALL_QUEUED, &rreq->flags); + } + + netfs_issue_read(rreq, subreq); cond_resched(); } while (size > 0); - if (atomic_dec_and_test(&rreq->nr_outstanding)) - netfs_rreq_terminated(rreq, false); + if (unlikely(size > 0)) { + smp_wmb(); /* Write lists before ALL_QUEUED. */ + set_bit(NETFS_RREQ_ALL_QUEUED, &rreq->flags); + netfs_wake_read_collector(rreq); + } /* Defer error return as we may need to wait for outstanding I/O. */ cmpxchg(&rreq->error, 0, ret); } -/* - * Wait for the read operation to complete, successfully or otherwise. - */ -static int netfs_wait_for_read(struct netfs_io_request *rreq) -{ - int ret; - - trace_netfs_rreq(rreq, netfs_rreq_trace_wait_ip); - wait_on_bit(&rreq->flags, NETFS_RREQ_IN_PROGRESS, TASK_UNINTERRUPTIBLE); - ret = rreq->error; - if (ret == 0 && rreq->submitted < rreq->len) { - trace_netfs_failure(rreq, NULL, ret, netfs_fail_short_read); - ret = -EIO; - } - - return ret; -} - -/* - * Set up the initial folioq of buffer folios in the rolling buffer and set the - * iterator to refer to it. - */ -static int netfs_prime_buffer(struct netfs_io_request *rreq) -{ - struct folio_queue *folioq; - struct folio_batch put_batch; - size_t added; - - folioq = kmalloc(sizeof(*folioq), GFP_KERNEL); - if (!folioq) - return -ENOMEM; - netfs_stat(&netfs_n_folioq); - folioq_init(folioq); - rreq->buffer = folioq; - rreq->buffer_tail = folioq; - rreq->submitted = rreq->start; - iov_iter_folio_queue(&rreq->iter, ITER_DEST, folioq, 0, 0, 0); - - folio_batch_init(&put_batch); - added = netfs_load_buffer_from_ra(rreq, folioq, &put_batch); - folio_batch_release(&put_batch); - rreq->iter.count += added; - rreq->submitted += added; - return 0; -} - /** * netfs_readahead - Helper to manage a read request * @ractl: The description of the readahead request @@ -405,6 +340,8 @@ void netfs_readahead(struct readahead_control *ractl) if (IS_ERR(rreq)) return; + __set_bit(NETFS_RREQ_OFFLOAD_COLLECTION, &rreq->flags); + ret = netfs_begin_cache_read(rreq, ictx); if (ret == -ENOMEM || ret == -EINTR || ret == -ERESTARTSYS) goto cleanup_free; @@ -416,7 +353,8 @@ void netfs_readahead(struct readahead_control *ractl) netfs_rreq_expand(rreq, ractl); rreq->ractl = ractl; - if (netfs_prime_buffer(rreq) < 0) + rreq->submitted = rreq->start; + if (rolling_buffer_init(&rreq->buffer, rreq->debug_id, ITER_DEST) < 0) goto cleanup_free; netfs_read_to_pagecache(rreq); @@ -432,23 +370,18 @@ EXPORT_SYMBOL(netfs_readahead); /* * Create a rolling buffer with a single occupying folio. */ -static int netfs_create_singular_buffer(struct netfs_io_request *rreq, struct folio *folio) +static int netfs_create_singular_buffer(struct netfs_io_request *rreq, struct folio *folio, + unsigned int rollbuf_flags) { - struct folio_queue *folioq; + ssize_t added; - folioq = kmalloc(sizeof(*folioq), GFP_KERNEL); - if (!folioq) + if (rolling_buffer_init(&rreq->buffer, rreq->debug_id, ITER_DEST) < 0) return -ENOMEM; - netfs_stat(&netfs_n_folioq); - folioq_init(folioq); - folioq_append(folioq, folio); - BUG_ON(folioq_folio(folioq, 0) != folio); - BUG_ON(folioq_folio_order(folioq, 0) != folio_order(folio)); - rreq->buffer = folioq; - rreq->buffer_tail = folioq; - rreq->submitted = rreq->start + rreq->len; - iov_iter_folio_queue(&rreq->iter, ITER_DEST, folioq, 0, 0, rreq->len); + added = rolling_buffer_append(&rreq->buffer, folio, rollbuf_flags); + if (added < 0) + return added; + rreq->submitted = rreq->start + added; rreq->ractl = (struct readahead_control *)1UL; return 0; } @@ -516,7 +449,7 @@ static int netfs_read_gaps(struct file *file, struct folio *folio) } if (to < flen) bvec_set_folio(&bvec[i++], folio, flen - to, to); - iov_iter_bvec(&rreq->iter, ITER_DEST, bvec, i, rreq->len); + iov_iter_bvec(&rreq->buffer.iter, ITER_DEST, bvec, i, rreq->len); rreq->submitted = rreq->start + flen; netfs_read_to_pagecache(rreq); @@ -525,7 +458,7 @@ static int netfs_read_gaps(struct file *file, struct folio *folio) folio_put(sink); ret = netfs_wait_for_read(rreq); - if (ret == 0) { + if (ret >= 0) { flush_dcache_folio(folio); folio_mark_uptodate(folio); } @@ -584,7 +517,7 @@ int netfs_read_folio(struct file *file, struct folio *folio) trace_netfs_read(rreq, rreq->start, rreq->len, netfs_read_trace_readpage); /* Set up the output buffer */ - ret = netfs_create_singular_buffer(rreq, folio); + ret = netfs_create_singular_buffer(rreq, folio, 0); if (ret < 0) goto discard; @@ -741,7 +674,7 @@ retry: trace_netfs_read(rreq, pos, len, netfs_read_trace_write_begin); /* Set up the output buffer */ - ret = netfs_create_singular_buffer(rreq, folio); + ret = netfs_create_singular_buffer(rreq, folio, 0); if (ret < 0) goto error_put; @@ -806,15 +739,14 @@ int netfs_prefetch_for_write(struct file *file, struct folio *folio, trace_netfs_read(rreq, start, flen, netfs_read_trace_prefetch_for_write); /* Set up the output buffer */ - ret = netfs_create_singular_buffer(rreq, folio); + ret = netfs_create_singular_buffer(rreq, folio, NETFS_ROLLBUF_PAGECACHE_MARK); if (ret < 0) goto error_put; - folioq_mark2(rreq->buffer, 0); netfs_read_to_pagecache(rreq); ret = netfs_wait_for_read(rreq); netfs_put_request(rreq, false, netfs_rreq_trace_put_return); - return ret; + return ret < 0 ? ret : 0; error_put: netfs_put_request(rreq, false, netfs_rreq_trace_put_discard); diff --git a/fs/netfs/direct_read.c b/fs/netfs/direct_read.c index b1a66a6e6bc2..0bf3c2f5a710 100644 --- a/fs/netfs/direct_read.c +++ b/fs/netfs/direct_read.c @@ -25,7 +25,7 @@ static void netfs_prepare_dio_read_iterator(struct netfs_io_subrequest *subreq) subreq->len = rsize; if (unlikely(rreq->io_streams[0].sreq_max_segs)) { - size_t limit = netfs_limit_iter(&rreq->iter, 0, rsize, + size_t limit = netfs_limit_iter(&rreq->buffer.iter, 0, rsize, rreq->io_streams[0].sreq_max_segs); if (limit < rsize) { @@ -36,9 +36,9 @@ static void netfs_prepare_dio_read_iterator(struct netfs_io_subrequest *subreq) trace_netfs_sreq(subreq, netfs_sreq_trace_prepare); - subreq->io_iter = rreq->iter; + subreq->io_iter = rreq->buffer.iter; iov_iter_truncate(&subreq->io_iter, subreq->len); - iov_iter_advance(&rreq->iter, subreq->len); + iov_iter_advance(&rreq->buffer.iter, subreq->len); } /* @@ -47,12 +47,11 @@ static void netfs_prepare_dio_read_iterator(struct netfs_io_subrequest *subreq) */ static int netfs_dispatch_unbuffered_reads(struct netfs_io_request *rreq) { + struct netfs_io_stream *stream = &rreq->io_streams[0]; unsigned long long start = rreq->start; ssize_t size = rreq->len; int ret = 0; - atomic_set(&rreq->nr_outstanding, 1); - do { struct netfs_io_subrequest *subreq; ssize_t slice; @@ -67,19 +66,25 @@ static int netfs_dispatch_unbuffered_reads(struct netfs_io_request *rreq) subreq->start = start; subreq->len = size; - atomic_inc(&rreq->nr_outstanding); - spin_lock_bh(&rreq->lock); - list_add_tail(&subreq->rreq_link, &rreq->subrequests); - subreq->prev_donated = rreq->prev_donated; - rreq->prev_donated = 0; + __set_bit(NETFS_SREQ_IN_PROGRESS, &subreq->flags); + + spin_lock(&rreq->lock); + list_add_tail(&subreq->rreq_link, &stream->subrequests); + if (list_is_first(&subreq->rreq_link, &stream->subrequests)) { + stream->front = subreq; + if (!stream->active) { + stream->collected_to = stream->front->start; + /* Store list pointers before active flag */ + smp_store_release(&stream->active, true); + } + } trace_netfs_sreq(subreq, netfs_sreq_trace_added); - spin_unlock_bh(&rreq->lock); + spin_unlock(&rreq->lock); netfs_stat(&netfs_n_rh_download); if (rreq->netfs_ops->prepare_read) { ret = rreq->netfs_ops->prepare_read(subreq); if (ret < 0) { - atomic_dec(&rreq->nr_outstanding); netfs_put_subrequest(subreq, false, netfs_sreq_trace_put_cancel); break; } @@ -87,20 +92,32 @@ static int netfs_dispatch_unbuffered_reads(struct netfs_io_request *rreq) netfs_prepare_dio_read_iterator(subreq); slice = subreq->len; - rreq->netfs_ops->issue_read(subreq); - size -= slice; start += slice; rreq->submitted += slice; + if (size <= 0) { + smp_wmb(); /* Write lists before ALL_QUEUED. */ + set_bit(NETFS_RREQ_ALL_QUEUED, &rreq->flags); + } + + rreq->netfs_ops->issue_read(subreq); + if (test_bit(NETFS_RREQ_PAUSE, &rreq->flags)) + netfs_wait_for_pause(rreq); + if (test_bit(NETFS_RREQ_FAILED, &rreq->flags)) + break; if (test_bit(NETFS_RREQ_BLOCKED, &rreq->flags) && test_bit(NETFS_RREQ_NONBLOCK, &rreq->flags)) break; cond_resched(); } while (size > 0); - if (atomic_dec_and_test(&rreq->nr_outstanding)) - netfs_rreq_terminated(rreq, false); + if (unlikely(size > 0)) { + smp_wmb(); /* Write lists before ALL_QUEUED. */ + set_bit(NETFS_RREQ_ALL_QUEUED, &rreq->flags); + netfs_wake_read_collector(rreq); + } + return ret; } @@ -133,21 +150,10 @@ static int netfs_unbuffered_read(struct netfs_io_request *rreq, bool sync) goto out; } - if (sync) { - trace_netfs_rreq(rreq, netfs_rreq_trace_wait_ip); - wait_on_bit(&rreq->flags, NETFS_RREQ_IN_PROGRESS, - TASK_UNINTERRUPTIBLE); - - ret = rreq->error; - if (ret == 0 && rreq->submitted < rreq->len && - rreq->origin != NETFS_DIO_READ) { - trace_netfs_failure(rreq, NULL, ret, netfs_fail_short_read); - ret = -EIO; - } - } else { + if (sync) + ret = netfs_wait_for_read(rreq); + else ret = -EIOCBQUEUED; - } - out: _leave(" = %d", ret); return ret; @@ -199,15 +205,15 @@ ssize_t netfs_unbuffered_read_iter_locked(struct kiocb *iocb, struct iov_iter *i * the request. */ if (user_backed_iter(iter)) { - ret = netfs_extract_user_iter(iter, rreq->len, &rreq->iter, 0); + ret = netfs_extract_user_iter(iter, rreq->len, &rreq->buffer.iter, 0); if (ret < 0) goto out; - rreq->direct_bv = (struct bio_vec *)rreq->iter.bvec; + rreq->direct_bv = (struct bio_vec *)rreq->buffer.iter.bvec; rreq->direct_bv_count = ret; rreq->direct_bv_unpin = iov_iter_extract_will_pin(iter); - rreq->len = iov_iter_count(&rreq->iter); + rreq->len = iov_iter_count(&rreq->buffer.iter); } else { - rreq->iter = *iter; + rreq->buffer.iter = *iter; rreq->len = orig_count; rreq->direct_bv_unpin = false; iov_iter_advance(iter, orig_count); @@ -215,8 +221,10 @@ ssize_t netfs_unbuffered_read_iter_locked(struct kiocb *iocb, struct iov_iter *i // TODO: Set up bounce buffer if needed - if (!sync) + if (!sync) { rreq->iocb = iocb; + __set_bit(NETFS_RREQ_OFFLOAD_COLLECTION, &rreq->flags); + } ret = netfs_unbuffered_read(rreq, sync); if (ret < 0) diff --git a/fs/netfs/direct_write.c b/fs/netfs/direct_write.c index 88f2adfab75e..42ce53cc216e 100644 --- a/fs/netfs/direct_write.c +++ b/fs/netfs/direct_write.c @@ -67,20 +67,23 @@ ssize_t netfs_unbuffered_write_iter_locked(struct kiocb *iocb, struct iov_iter * * allocate a sufficiently large bvec array and may shorten the * request. */ - if (async || user_backed_iter(iter)) { - n = netfs_extract_user_iter(iter, len, &wreq->iter, 0); + if (user_backed_iter(iter)) { + n = netfs_extract_user_iter(iter, len, &wreq->buffer.iter, 0); if (n < 0) { ret = n; goto out; } - wreq->direct_bv = (struct bio_vec *)wreq->iter.bvec; + wreq->direct_bv = (struct bio_vec *)wreq->buffer.iter.bvec; wreq->direct_bv_count = n; wreq->direct_bv_unpin = iov_iter_extract_will_pin(iter); } else { - wreq->iter = *iter; + /* If this is a kernel-generated async DIO request, + * assume that any resources the iterator points to + * (eg. a bio_vec array) will persist till the end of + * the op. + */ + wreq->buffer.iter = *iter; } - - wreq->io_iter = wreq->iter; } __set_bit(NETFS_RREQ_USE_IO_ITER, &wreq->flags); @@ -92,7 +95,7 @@ ssize_t netfs_unbuffered_write_iter_locked(struct kiocb *iocb, struct iov_iter * __set_bit(NETFS_RREQ_UPLOAD_TO_SERVER, &wreq->flags); if (async) wreq->iocb = iocb; - wreq->len = iov_iter_count(&wreq->io_iter); + wreq->len = iov_iter_count(&wreq->buffer.iter); wreq->cleanup = netfs_cleanup_dio_write; ret = netfs_unbuffered_write(wreq, is_sync_kiocb(iocb), wreq->len); if (ret < 0) { @@ -104,7 +107,6 @@ ssize_t netfs_unbuffered_write_iter_locked(struct kiocb *iocb, struct iov_iter * trace_netfs_rreq(wreq, netfs_rreq_trace_wait_ip); wait_on_bit(&wreq->flags, NETFS_RREQ_IN_PROGRESS, TASK_UNINTERRUPTIBLE); - smp_rmb(); /* Read error/transferred after RIP flag */ ret = wreq->error; if (ret == 0) { ret = wreq->transferred; diff --git a/fs/netfs/internal.h b/fs/netfs/internal.h index c562aec3b483..eb76f98c894b 100644 --- a/fs/netfs/internal.h +++ b/fs/netfs/internal.h @@ -23,6 +23,7 @@ /* * buffered_read.c */ +void netfs_cache_read_terminated(void *priv, ssize_t transferred_or_error, bool was_async); int netfs_prefetch_for_write(struct file *file, struct folio *folio, size_t offset, size_t len); @@ -58,11 +59,8 @@ static inline void netfs_proc_del_rreq(struct netfs_io_request *rreq) {} /* * misc.c */ -struct folio_queue *netfs_buffer_make_space(struct netfs_io_request *rreq); -int netfs_buffer_append_folio(struct netfs_io_request *rreq, struct folio *folio, - bool needs_put); -struct folio_queue *netfs_delete_buffer_head(struct netfs_io_request *wreq); -void netfs_clear_buffer(struct netfs_io_request *rreq); +struct folio_queue *netfs_buffer_make_space(struct netfs_io_request *rreq, + enum netfs_folioq_trace trace); void netfs_reset_iter(struct netfs_io_subrequest *subreq); /* @@ -84,20 +82,27 @@ static inline void netfs_see_request(struct netfs_io_request *rreq, trace_netfs_rreq_ref(rreq->debug_id, refcount_read(&rreq->ref), what); } +static inline void netfs_see_subrequest(struct netfs_io_subrequest *subreq, + enum netfs_sreq_ref_trace what) +{ + trace_netfs_sreq_ref(subreq->rreq->debug_id, subreq->debug_index, + refcount_read(&subreq->ref), what); +} + /* * read_collect.c */ -void netfs_read_termination_worker(struct work_struct *work); -void netfs_rreq_terminated(struct netfs_io_request *rreq, bool was_async); +void netfs_read_collection_worker(struct work_struct *work); +void netfs_wake_read_collector(struct netfs_io_request *rreq); +void netfs_cache_read_terminated(void *priv, ssize_t transferred_or_error, bool was_async); +ssize_t netfs_wait_for_read(struct netfs_io_request *rreq); +void netfs_wait_for_pause(struct netfs_io_request *rreq); /* * read_pgpriv2.c */ -void netfs_pgpriv2_mark_copy_to_cache(struct netfs_io_subrequest *subreq, - struct netfs_io_request *rreq, - struct folio_queue *folioq, - int slot); -void netfs_pgpriv2_write_to_the_cache(struct netfs_io_request *rreq); +void netfs_pgpriv2_copy_to_cache(struct netfs_io_request *rreq, struct folio *folio); +void netfs_pgpriv2_end_copy_to_cache(struct netfs_io_request *rreq); bool netfs_pgpriv2_unlock_copied_folios(struct netfs_io_request *wreq); /* @@ -113,6 +118,7 @@ void netfs_unlock_abandoned_read_pages(struct netfs_io_request *rreq); extern atomic_t netfs_n_rh_dio_read; extern atomic_t netfs_n_rh_readahead; extern atomic_t netfs_n_rh_read_folio; +extern atomic_t netfs_n_rh_read_single; extern atomic_t netfs_n_rh_rreq; extern atomic_t netfs_n_rh_sreq; extern atomic_t netfs_n_rh_download; @@ -181,9 +187,9 @@ void netfs_reissue_write(struct netfs_io_stream *stream, struct iov_iter *source); void netfs_issue_write(struct netfs_io_request *wreq, struct netfs_io_stream *stream); -int netfs_advance_write(struct netfs_io_request *wreq, - struct netfs_io_stream *stream, - loff_t start, size_t len, bool to_eof); +size_t netfs_advance_write(struct netfs_io_request *wreq, + struct netfs_io_stream *stream, + loff_t start, size_t len, bool to_eof); struct netfs_io_request *netfs_begin_writethrough(struct kiocb *iocb, size_t len); int netfs_advance_writethrough(struct netfs_io_request *wreq, struct writeback_control *wbc, struct folio *folio, size_t copied, bool to_page_end, @@ -193,6 +199,11 @@ int netfs_end_writethrough(struct netfs_io_request *wreq, struct writeback_contr int netfs_unbuffered_write(struct netfs_io_request *wreq, bool may_wait, size_t len); /* + * write_retry.c + */ +void netfs_retry_writes(struct netfs_io_request *wreq); + +/* * Miscellaneous functions. */ static inline bool netfs_is_cache_enabled(struct netfs_inode *ctx) diff --git a/fs/netfs/main.c b/fs/netfs/main.c index 6c7be1377ee0..4e3e62040831 100644 --- a/fs/netfs/main.c +++ b/fs/netfs/main.c @@ -37,9 +37,11 @@ static const char *netfs_origins[nr__netfs_io_origin] = { [NETFS_READAHEAD] = "RA", [NETFS_READPAGE] = "RP", [NETFS_READ_GAPS] = "RG", + [NETFS_READ_SINGLE] = "R1", [NETFS_READ_FOR_WRITE] = "RW", [NETFS_DIO_READ] = "DR", [NETFS_WRITEBACK] = "WB", + [NETFS_WRITEBACK_SINGLE] = "W1", [NETFS_WRITETHROUGH] = "WT", [NETFS_UNBUFFERED_WRITE] = "UW", [NETFS_DIO_WRITE] = "DW", @@ -69,7 +71,7 @@ static int netfs_requests_seq_show(struct seq_file *m, void *v) refcount_read(&rreq->ref), rreq->flags, rreq->error, - atomic_read(&rreq->nr_outstanding), + 0, rreq->start, rreq->submitted, rreq->len); seq_putc(m, '\n'); return 0; @@ -116,7 +118,7 @@ static int __init netfs_init(void) goto error_reqpool; netfs_subrequest_slab = kmem_cache_create("netfs_subrequest", - sizeof(struct netfs_io_subrequest), 0, + sizeof(struct netfs_io_subrequest) + 16, 0, SLAB_HWCACHE_ALIGN | SLAB_ACCOUNT, NULL); if (!netfs_subrequest_slab) diff --git a/fs/netfs/misc.c b/fs/netfs/misc.c index 78fe5796b2b2..7099aa07737a 100644 --- a/fs/netfs/misc.c +++ b/fs/netfs/misc.c @@ -8,113 +8,101 @@ #include <linux/swap.h> #include "internal.h" -/* - * Make sure there's space in the rolling queue. +/** + * netfs_alloc_folioq_buffer - Allocate buffer space into a folio queue + * @mapping: Address space to set on the folio (or NULL). + * @_buffer: Pointer to the folio queue to add to (may point to a NULL; updated). + * @_cur_size: Current size of the buffer (updated). + * @size: Target size of the buffer. + * @gfp: The allocation constraints. */ -struct folio_queue *netfs_buffer_make_space(struct netfs_io_request *rreq) +int netfs_alloc_folioq_buffer(struct address_space *mapping, + struct folio_queue **_buffer, + size_t *_cur_size, ssize_t size, gfp_t gfp) { - struct folio_queue *tail = rreq->buffer_tail, *prev; - unsigned int prev_nr_slots = 0; - - if (WARN_ON_ONCE(!rreq->buffer && tail) || - WARN_ON_ONCE(rreq->buffer && !tail)) - return ERR_PTR(-EIO); - - prev = tail; - if (prev) { - if (!folioq_full(tail)) - return tail; - prev_nr_slots = folioq_nr_slots(tail); - } - - tail = kmalloc(sizeof(*tail), GFP_NOFS); - if (!tail) - return ERR_PTR(-ENOMEM); - netfs_stat(&netfs_n_folioq); - folioq_init(tail); - tail->prev = prev; - if (prev) - /* [!] NOTE: After we set prev->next, the consumer is entirely - * at liberty to delete prev. - */ - WRITE_ONCE(prev->next, tail); - - rreq->buffer_tail = tail; - if (!rreq->buffer) { - rreq->buffer = tail; - iov_iter_folio_queue(&rreq->io_iter, ITER_SOURCE, tail, 0, 0, 0); - } else { - /* Make sure we don't leave the master iterator pointing to a - * block that might get immediately consumed. - */ - if (rreq->io_iter.folioq == prev && - rreq->io_iter.folioq_slot == prev_nr_slots) { - rreq->io_iter.folioq = tail; - rreq->io_iter.folioq_slot = 0; + struct folio_queue *tail = *_buffer, *p; + + size = round_up(size, PAGE_SIZE); + if (*_cur_size >= size) + return 0; + + if (tail) + while (tail->next) + tail = tail->next; + + do { + struct folio *folio; + int order = 0, slot; + + if (!tail || folioq_full(tail)) { + p = netfs_folioq_alloc(0, GFP_NOFS, netfs_trace_folioq_alloc_buffer); + if (!p) + return -ENOMEM; + if (tail) { + tail->next = p; + p->prev = tail; + } else { + *_buffer = p; + } + tail = p; } - } - rreq->buffer_tail_slot = 0; - return tail; -} -/* - * Append a folio to the rolling queue. - */ -int netfs_buffer_append_folio(struct netfs_io_request *rreq, struct folio *folio, - bool needs_put) -{ - struct folio_queue *tail; - unsigned int slot, order = folio_order(folio); + if (size - *_cur_size > PAGE_SIZE) + order = umin(ilog2(size - *_cur_size) - PAGE_SHIFT, + MAX_PAGECACHE_ORDER); - tail = netfs_buffer_make_space(rreq); - if (IS_ERR(tail)) - return PTR_ERR(tail); + folio = folio_alloc(gfp, order); + if (!folio && order > 0) + folio = folio_alloc(gfp, 0); + if (!folio) + return -ENOMEM; - rreq->io_iter.count += PAGE_SIZE << order; + folio->mapping = mapping; + folio->index = *_cur_size / PAGE_SIZE; + trace_netfs_folio(folio, netfs_folio_trace_alloc_buffer); + slot = folioq_append_mark(tail, folio); + *_cur_size += folioq_folio_size(tail, slot); + } while (*_cur_size < size); - slot = folioq_append(tail, folio); - /* Store the counter after setting the slot. */ - smp_store_release(&rreq->buffer_tail_slot, slot); return 0; } +EXPORT_SYMBOL(netfs_alloc_folioq_buffer); -/* - * Delete the head of a rolling queue. +/** + * netfs_free_folioq_buffer - Free a folio queue. + * @fq: The start of the folio queue to free + * + * Free up a chain of folio_queues and, if marked, the marked folios they point + * to. */ -struct folio_queue *netfs_delete_buffer_head(struct netfs_io_request *wreq) +void netfs_free_folioq_buffer(struct folio_queue *fq) { - struct folio_queue *head = wreq->buffer, *next = head->next; - - if (next) - next->prev = NULL; - netfs_stat_d(&netfs_n_folioq); - kfree(head); - wreq->buffer = next; - return next; -} + struct folio_queue *next; + struct folio_batch fbatch; -/* - * Clear out a rolling queue. - */ -void netfs_clear_buffer(struct netfs_io_request *rreq) -{ - struct folio_queue *p; + folio_batch_init(&fbatch); + + for (; fq; fq = next) { + for (int slot = 0; slot < folioq_count(fq); slot++) { + struct folio *folio = folioq_folio(fq, slot); - while ((p = rreq->buffer)) { - rreq->buffer = p->next; - for (int slot = 0; slot < folioq_count(p); slot++) { - struct folio *folio = folioq_folio(p, slot); - if (!folio) + if (!folio || + !folioq_is_marked(fq, slot)) continue; - if (folioq_is_marked(p, slot)) { - trace_netfs_folio(folio, netfs_folio_trace_put); - folio_put(folio); - } + + trace_netfs_folio(folio, netfs_folio_trace_put); + if (folio_batch_add(&fbatch, folio)) + folio_batch_release(&fbatch); } + netfs_stat_d(&netfs_n_folioq); - kfree(p); + next = fq->next; + kfree(fq); } + + folio_batch_release(&fbatch); } +EXPORT_SYMBOL(netfs_free_folioq_buffer); /* * Reset the subrequest iterator to refer just to the region remaining to be diff --git a/fs/netfs/objects.c b/fs/netfs/objects.c index 31e388ec6e48..dc6b41ef18b0 100644 --- a/fs/netfs/objects.c +++ b/fs/netfs/objects.c @@ -48,17 +48,20 @@ struct netfs_io_request *netfs_alloc_request(struct address_space *mapping, spin_lock_init(&rreq->lock); INIT_LIST_HEAD(&rreq->io_streams[0].subrequests); INIT_LIST_HEAD(&rreq->io_streams[1].subrequests); - INIT_LIST_HEAD(&rreq->subrequests); + init_waitqueue_head(&rreq->waitq); refcount_set(&rreq->ref, 1); if (origin == NETFS_READAHEAD || origin == NETFS_READPAGE || origin == NETFS_READ_GAPS || + origin == NETFS_READ_SINGLE || origin == NETFS_READ_FOR_WRITE || - origin == NETFS_DIO_READ) - INIT_WORK(&rreq->work, netfs_read_termination_worker); - else + origin == NETFS_DIO_READ) { + INIT_WORK(&rreq->work, netfs_read_collection_worker); + rreq->io_streams[0].avail = true; + } else { INIT_WORK(&rreq->work, netfs_write_collection_worker); + } __set_bit(NETFS_RREQ_IN_PROGRESS, &rreq->flags); if (file && file->f_flags & O_NONBLOCK) @@ -92,14 +95,6 @@ void netfs_clear_subrequests(struct netfs_io_request *rreq, bool was_async) struct netfs_io_stream *stream; int s; - while (!list_empty(&rreq->subrequests)) { - subreq = list_first_entry(&rreq->subrequests, - struct netfs_io_subrequest, rreq_link); - list_del(&subreq->rreq_link); - netfs_put_subrequest(subreq, was_async, - netfs_sreq_trace_put_clear); - } - for (s = 0; s < ARRAY_SIZE(rreq->io_streams); s++) { stream = &rreq->io_streams[s]; while (!list_empty(&stream->subrequests)) { @@ -143,7 +138,7 @@ static void netfs_free_request(struct work_struct *work) } kvfree(rreq->direct_bv); } - netfs_clear_buffer(rreq); + rolling_buffer_clear(&rreq->buffer); if (atomic_dec_and_test(&ictx->io_count)) wake_up_var(&ictx->io_count); diff --git a/fs/netfs/read_collect.c b/fs/netfs/read_collect.c index 3cbb289535a8..f65affa5a9e4 100644 --- a/fs/netfs/read_collect.c +++ b/fs/netfs/read_collect.c @@ -14,6 +14,14 @@ #include <linux/task_io_accounting_ops.h> #include "internal.h" +/* Notes made in the collector */ +#define HIT_PENDING 0x01 /* A front op was still pending */ +#define MADE_PROGRESS 0x04 /* Made progress cleaning up a stream or the folio set */ +#define BUFFERED 0x08 /* The pagecache needs cleaning up */ +#define NEED_RETRY 0x10 /* A front op requests retrying */ +#define COPY_TO_CACHE 0x40 /* Need to copy subrequest to cache */ +#define ABANDON_SREQ 0x80 /* Need to abandon untransferred part of subrequest */ + /* * Clear the unread part of an I/O request. */ @@ -31,14 +39,18 @@ static void netfs_clear_unread(struct netfs_io_subrequest *subreq) * cache the folio, we set the group to NETFS_FOLIO_COPY_TO_CACHE, mark it * dirty and let writeback handle it. */ -static void netfs_unlock_read_folio(struct netfs_io_subrequest *subreq, - struct netfs_io_request *rreq, +static void netfs_unlock_read_folio(struct netfs_io_request *rreq, struct folio_queue *folioq, int slot) { struct netfs_folio *finfo; struct folio *folio = folioq_folio(folioq, slot); + if (unlikely(folio_pos(folio) < rreq->abandon_to)) { + trace_netfs_folio(folio, netfs_folio_trace_abandon); + goto just_unlock; + } + flush_dcache_folio(folio); folio_mark_uptodate(folio); @@ -53,7 +65,7 @@ static void netfs_unlock_read_folio(struct netfs_io_subrequest *subreq, kfree(finfo); } - if (test_bit(NETFS_SREQ_COPY_TO_CACHE, &subreq->flags)) { + if (test_bit(NETFS_RREQ_FOLIO_COPY_TO_CACHE, &rreq->flags)) { if (!WARN_ON_ONCE(folio_get_private(folio) != NULL)) { trace_netfs_folio(folio, netfs_folio_trace_copy_to_cache); folio_attach_private(folio, NETFS_FOLIO_COPY_TO_CACHE); @@ -62,12 +74,15 @@ static void netfs_unlock_read_folio(struct netfs_io_subrequest *subreq, } else { trace_netfs_folio(folio, netfs_folio_trace_read_done); } + + folioq_clear(folioq, slot); } else { // TODO: Use of PG_private_2 is deprecated. - if (test_bit(NETFS_SREQ_COPY_TO_CACHE, &subreq->flags)) - netfs_pgpriv2_mark_copy_to_cache(subreq, rreq, folioq, slot); + if (test_bit(NETFS_RREQ_FOLIO_COPY_TO_CACHE, &rreq->flags)) + netfs_pgpriv2_copy_to_cache(rreq, folio); } +just_unlock: if (!test_bit(NETFS_RREQ_DONT_UNLOCK_FOLIOS, &rreq->flags)) { if (folio->index == rreq->no_unlock_folio && test_bit(NETFS_RREQ_NO_UNLOCK_FOLIO, &rreq->flags)) { @@ -82,234 +97,244 @@ static void netfs_unlock_read_folio(struct netfs_io_subrequest *subreq, } /* - * Unlock any folios that are now completely read. Returns true if the - * subrequest is removed from the list. + * Unlock any folios we've finished with. */ -static bool netfs_consume_read_data(struct netfs_io_subrequest *subreq, bool was_async) +static void netfs_read_unlock_folios(struct netfs_io_request *rreq, + unsigned int *notes) { - struct netfs_io_subrequest *prev, *next; - struct netfs_io_request *rreq = subreq->rreq; - struct folio_queue *folioq = subreq->curr_folioq; - size_t avail, prev_donated, next_donated, fsize, part, excess; - loff_t fpos, start; - loff_t fend; - int slot = subreq->curr_folioq_slot; - - if (WARN(subreq->transferred > subreq->len, - "Subreq overread: R%x[%x] %zu > %zu", - rreq->debug_id, subreq->debug_index, - subreq->transferred, subreq->len)) - subreq->transferred = subreq->len; - -next_folio: - fsize = PAGE_SIZE << subreq->curr_folio_order; - fpos = round_down(subreq->start + subreq->consumed, fsize); - fend = fpos + fsize; - - if (WARN_ON_ONCE(!folioq) || - WARN_ON_ONCE(!folioq_folio(folioq, slot)) || - WARN_ON_ONCE(folioq_folio(folioq, slot)->index != fpos / PAGE_SIZE)) { - pr_err("R=%08x[%x] s=%llx-%llx ctl=%zx/%zx/%zx sl=%u\n", - rreq->debug_id, subreq->debug_index, - subreq->start, subreq->start + subreq->transferred - 1, - subreq->consumed, subreq->transferred, subreq->len, - slot); - if (folioq) { - struct folio *folio = folioq_folio(folioq, slot); - - pr_err("folioq: orders=%02x%02x%02x%02x\n", - folioq->orders[0], folioq->orders[1], - folioq->orders[2], folioq->orders[3]); - if (folio) - pr_err("folio: %llx-%llx ix=%llx o=%u qo=%u\n", - fpos, fend - 1, folio_pos(folio), folio_order(folio), - folioq_folio_order(folioq, slot)); - } - } + struct folio_queue *folioq = rreq->buffer.tail; + unsigned long long collected_to = rreq->collected_to; + unsigned int slot = rreq->buffer.first_tail_slot; -donation_changed: - /* Try to consume the current folio if we've hit or passed the end of - * it. There's a possibility that this subreq doesn't start at the - * beginning of the folio, in which case we need to donate to/from the - * preceding subreq. - * - * We also need to include any potential donation back from the - * following subreq. - */ - prev_donated = READ_ONCE(subreq->prev_donated); - next_donated = READ_ONCE(subreq->next_donated); - if (prev_donated || next_donated) { - spin_lock_bh(&rreq->lock); - prev_donated = subreq->prev_donated; - next_donated = subreq->next_donated; - subreq->start -= prev_donated; - subreq->len += prev_donated; - subreq->transferred += prev_donated; - prev_donated = subreq->prev_donated = 0; - if (subreq->transferred == subreq->len) { - subreq->len += next_donated; - subreq->transferred += next_donated; - next_donated = subreq->next_donated = 0; + if (rreq->cleaned_to >= rreq->collected_to) + return; + + // TODO: Begin decryption + + if (slot >= folioq_nr_slots(folioq)) { + folioq = rolling_buffer_delete_spent(&rreq->buffer); + if (!folioq) { + rreq->front_folio_order = 0; + return; } - trace_netfs_sreq(subreq, netfs_sreq_trace_add_donations); - spin_unlock_bh(&rreq->lock); + slot = 0; } - avail = subreq->transferred; - if (avail == subreq->len) - avail += next_donated; - start = subreq->start; - if (subreq->consumed == 0) { - start -= prev_donated; - avail += prev_donated; - } else { - start += subreq->consumed; - avail -= subreq->consumed; - } - part = umin(avail, fsize); - - trace_netfs_progress(subreq, start, avail, part); - - if (start + avail >= fend) { - if (fpos == start) { - /* Flush, unlock and mark for caching any folio we've just read. */ - subreq->consumed = fend - subreq->start; - netfs_unlock_read_folio(subreq, rreq, folioq, slot); - folioq_mark2(folioq, slot); - if (subreq->consumed >= subreq->len) - goto remove_subreq; - } else if (fpos < start) { - excess = fend - subreq->start; - - spin_lock_bh(&rreq->lock); - /* If we complete first on a folio split with the - * preceding subreq, donate to that subreq - otherwise - * we get the responsibility. - */ - if (subreq->prev_donated != prev_donated) { - spin_unlock_bh(&rreq->lock); - goto donation_changed; - } + for (;;) { + struct folio *folio; + unsigned long long fpos, fend; + unsigned int order; + size_t fsize; - if (list_is_first(&subreq->rreq_link, &rreq->subrequests)) { - spin_unlock_bh(&rreq->lock); - pr_err("Can't donate prior to front\n"); - goto bad; - } + if (*notes & COPY_TO_CACHE) + set_bit(NETFS_RREQ_FOLIO_COPY_TO_CACHE, &rreq->flags); - prev = list_prev_entry(subreq, rreq_link); - WRITE_ONCE(prev->next_donated, prev->next_donated + excess); - subreq->start += excess; - subreq->len -= excess; - subreq->transferred -= excess; - trace_netfs_donate(rreq, subreq, prev, excess, - netfs_trace_donate_tail_to_prev); - trace_netfs_sreq(subreq, netfs_sreq_trace_donate_to_prev); - - if (subreq->consumed >= subreq->len) - goto remove_subreq_locked; - spin_unlock_bh(&rreq->lock); - } else { - pr_err("fpos > start\n"); - goto bad; - } + folio = folioq_folio(folioq, slot); + if (WARN_ONCE(!folio_test_locked(folio), + "R=%08x: folio %lx is not locked\n", + rreq->debug_id, folio->index)) + trace_netfs_folio(folio, netfs_folio_trace_not_locked); + + order = folioq_folio_order(folioq, slot); + rreq->front_folio_order = order; + fsize = PAGE_SIZE << order; + fpos = folio_pos(folio); + fend = umin(fpos + fsize, rreq->i_size); + + trace_netfs_collect_folio(rreq, folio, fend, collected_to); - /* Advance the rolling buffer to the next folio. */ + /* Unlock any folio we've transferred all of. */ + if (collected_to < fend) + break; + + netfs_unlock_read_folio(rreq, folioq, slot); + WRITE_ONCE(rreq->cleaned_to, fpos + fsize); + *notes |= MADE_PROGRESS; + + clear_bit(NETFS_RREQ_FOLIO_COPY_TO_CACHE, &rreq->flags); + + /* Clean up the head folioq. If we clear an entire folioq, then + * we can get rid of it provided it's not also the tail folioq + * being filled by the issuer. + */ + folioq_clear(folioq, slot); slot++; if (slot >= folioq_nr_slots(folioq)) { + folioq = rolling_buffer_delete_spent(&rreq->buffer); + if (!folioq) + goto done; slot = 0; - folioq = folioq->next; - subreq->curr_folioq = folioq; + trace_netfs_folioq(folioq, netfs_trace_folioq_read_progress); } - subreq->curr_folioq_slot = slot; - if (folioq && folioq_folio(folioq, slot)) - subreq->curr_folio_order = folioq->orders[slot]; - if (!was_async) - cond_resched(); - goto next_folio; + + if (fpos + fsize >= collected_to) + break; } - /* Deal with partial progress. */ - if (subreq->transferred < subreq->len) - return false; + rreq->buffer.tail = folioq; +done: + rreq->buffer.first_tail_slot = slot; +} - /* Donate the remaining downloaded data to one of the neighbouring - * subrequests. Note that we may race with them doing the same thing. +/* + * Collect and assess the results of various read subrequests. We may need to + * retry some of the results. + * + * Note that we have a sequence of subrequests, which may be drawing on + * different sources and may or may not be the same size or starting position + * and may not even correspond in boundary alignment. + */ +static void netfs_collect_read_results(struct netfs_io_request *rreq) +{ + struct netfs_io_subrequest *front, *remove; + struct netfs_io_stream *stream = &rreq->io_streams[0]; + unsigned int notes; + + _enter("%llx-%llx", rreq->start, rreq->start + rreq->len); + trace_netfs_rreq(rreq, netfs_rreq_trace_collect); + trace_netfs_collect(rreq); + +reassess: + if (rreq->origin == NETFS_READAHEAD || + rreq->origin == NETFS_READPAGE || + rreq->origin == NETFS_READ_FOR_WRITE) + notes = BUFFERED; + else + notes = 0; + + /* Remove completed subrequests from the front of the stream and + * advance the completion point. We stop when we hit something that's + * in progress. The issuer thread may be adding stuff to the tail + * whilst we're doing this. */ - spin_lock_bh(&rreq->lock); + front = READ_ONCE(stream->front); + while (front) { + size_t transferred; - if (subreq->prev_donated != prev_donated || - subreq->next_donated != next_donated) { - spin_unlock_bh(&rreq->lock); - cond_resched(); - goto donation_changed; - } + trace_netfs_collect_sreq(rreq, front); + _debug("sreq [%x] %llx %zx/%zx", + front->debug_index, front->start, front->transferred, front->len); - /* Deal with the trickiest case: that this subreq is in the middle of a - * folio, not touching either edge, but finishes first. In such a - * case, we donate to the previous subreq, if there is one, so that the - * donation is only handled when that completes - and remove this - * subreq from the list. - * - * If the previous subreq finished first, we will have acquired their - * donation and should be able to unlock folios and/or donate nextwards. - */ - if (!subreq->consumed && - !prev_donated && - !list_is_first(&subreq->rreq_link, &rreq->subrequests)) { - prev = list_prev_entry(subreq, rreq_link); - WRITE_ONCE(prev->next_donated, prev->next_donated + subreq->len); - subreq->start += subreq->len; - subreq->len = 0; - subreq->transferred = 0; - trace_netfs_donate(rreq, subreq, prev, subreq->len, - netfs_trace_donate_to_prev); - trace_netfs_sreq(subreq, netfs_sreq_trace_donate_to_prev); - goto remove_subreq_locked; + if (stream->collected_to < front->start) { + trace_netfs_collect_gap(rreq, stream, front->start, 'F'); + stream->collected_to = front->start; + } + + if (test_bit(NETFS_SREQ_IN_PROGRESS, &front->flags)) + notes |= HIT_PENDING; + smp_rmb(); /* Read counters after IN_PROGRESS flag. */ + transferred = READ_ONCE(front->transferred); + + /* If we can now collect the next folio, do so. We don't want + * to defer this as we have to decide whether we need to copy + * to the cache or not, and that may differ between adjacent + * subreqs. + */ + if (notes & BUFFERED) { + size_t fsize = PAGE_SIZE << rreq->front_folio_order; + + /* Clear the tail of a short read. */ + if (!(notes & HIT_PENDING) && + front->error == 0 && + transferred < front->len && + (test_bit(NETFS_SREQ_HIT_EOF, &front->flags) || + test_bit(NETFS_SREQ_CLEAR_TAIL, &front->flags))) { + netfs_clear_unread(front); + transferred = front->transferred = front->len; + trace_netfs_sreq(front, netfs_sreq_trace_clear); + } + + stream->collected_to = front->start + transferred; + rreq->collected_to = stream->collected_to; + + if (test_bit(NETFS_SREQ_COPY_TO_CACHE, &front->flags)) + notes |= COPY_TO_CACHE; + + if (test_bit(NETFS_SREQ_FAILED, &front->flags)) { + rreq->abandon_to = front->start + front->len; + front->transferred = front->len; + transferred = front->len; + trace_netfs_rreq(rreq, netfs_rreq_trace_set_abandon); + } + if (front->start + transferred >= rreq->cleaned_to + fsize || + test_bit(NETFS_SREQ_HIT_EOF, &front->flags)) + netfs_read_unlock_folios(rreq, ¬es); + } else { + stream->collected_to = front->start + transferred; + rreq->collected_to = stream->collected_to; + } + + /* Stall if the front is still undergoing I/O. */ + if (notes & HIT_PENDING) + break; + + if (test_bit(NETFS_SREQ_FAILED, &front->flags)) { + if (!stream->failed) { + stream->error = front->error; + rreq->error = front->error; + set_bit(NETFS_RREQ_FAILED, &rreq->flags); + stream->failed = true; + } + notes |= MADE_PROGRESS | ABANDON_SREQ; + } else if (test_bit(NETFS_SREQ_NEED_RETRY, &front->flags)) { + stream->need_retry = true; + notes |= NEED_RETRY | MADE_PROGRESS; + break; + } else { + if (!stream->failed) + stream->transferred = stream->collected_to - rreq->start; + notes |= MADE_PROGRESS; + } + + /* Remove if completely consumed. */ + stream->source = front->source; + spin_lock(&rreq->lock); + + remove = front; + trace_netfs_sreq(front, netfs_sreq_trace_discard); + list_del_init(&front->rreq_link); + front = list_first_entry_or_null(&stream->subrequests, + struct netfs_io_subrequest, rreq_link); + stream->front = front; + spin_unlock(&rreq->lock); + netfs_put_subrequest(remove, false, + notes & ABANDON_SREQ ? + netfs_sreq_trace_put_abandon : + netfs_sreq_trace_put_done); } - /* If we can't donate down the chain, donate up the chain instead. */ - excess = subreq->len - subreq->consumed + next_donated; + trace_netfs_collect_stream(rreq, stream); + trace_netfs_collect_state(rreq, rreq->collected_to, notes); - if (!subreq->consumed) - excess += prev_donated; + if (!(notes & BUFFERED)) + rreq->cleaned_to = rreq->collected_to; - if (list_is_last(&subreq->rreq_link, &rreq->subrequests)) { - rreq->prev_donated = excess; - trace_netfs_donate(rreq, subreq, NULL, excess, - netfs_trace_donate_to_deferred_next); - } else { - next = list_next_entry(subreq, rreq_link); - WRITE_ONCE(next->prev_donated, excess); - trace_netfs_donate(rreq, subreq, next, excess, - netfs_trace_donate_to_next); + if (notes & NEED_RETRY) + goto need_retry; + if ((notes & MADE_PROGRESS) && test_bit(NETFS_RREQ_PAUSE, &rreq->flags)) { + trace_netfs_rreq(rreq, netfs_rreq_trace_unpause); + clear_bit_unlock(NETFS_RREQ_PAUSE, &rreq->flags); + smp_mb__after_atomic(); /* Set PAUSE before task state */ + wake_up(&rreq->waitq); + } + + if (notes & MADE_PROGRESS) { + //cond_resched(); + goto reassess; } - trace_netfs_sreq(subreq, netfs_sreq_trace_donate_to_next); - subreq->len = subreq->consumed; - subreq->transferred = subreq->consumed; - goto remove_subreq_locked; - -remove_subreq: - spin_lock_bh(&rreq->lock); -remove_subreq_locked: - subreq->consumed = subreq->len; - list_del(&subreq->rreq_link); - spin_unlock_bh(&rreq->lock); - netfs_put_subrequest(subreq, false, netfs_sreq_trace_put_consumed); - return true; - -bad: - /* Errr... prev and next both donated to us, but insufficient to finish - * the folio. + +out: + _leave(" = %x", notes); + return; + +need_retry: + /* Okay... We're going to have to retry parts of the stream. Note + * that any partially completed op will have had any wholly transferred + * folios removed from it. */ - printk("R=%08x[%x] s=%llx-%llx %zx/%zx/%zx\n", - rreq->debug_id, subreq->debug_index, - subreq->start, subreq->start + subreq->transferred - 1, - subreq->consumed, subreq->transferred, subreq->len); - printk("folio: %llx-%llx\n", fpos, fend - 1); - printk("donated: prev=%zx next=%zx\n", prev_donated, next_donated); - printk("s=%llx av=%zx part=%zx\n", start, avail, part); - BUG(); + _debug("retry"); + netfs_retry_reads(rreq); + goto out; } /* @@ -318,12 +343,13 @@ bad: static void netfs_rreq_assess_dio(struct netfs_io_request *rreq) { struct netfs_io_subrequest *subreq; + struct netfs_io_stream *stream = &rreq->io_streams[0]; unsigned int i; /* Collect unbuffered reads and direct reads, adding up the transfer * sizes until we find the first short or failed subrequest. */ - list_for_each_entry(subreq, &rreq->subrequests, rreq_link) { + list_for_each_entry(subreq, &stream->subrequests, rreq_link) { rreq->transferred += subreq->transferred; if (subreq->transferred < subreq->len || @@ -356,89 +382,135 @@ static void netfs_rreq_assess_dio(struct netfs_io_request *rreq) } /* - * Assess the state of a read request and decide what to do next. + * Do processing after reading a monolithic single object. + */ +static void netfs_rreq_assess_single(struct netfs_io_request *rreq) +{ + struct netfs_io_stream *stream = &rreq->io_streams[0]; + + if (!rreq->error && stream->source == NETFS_DOWNLOAD_FROM_SERVER && + fscache_resources_valid(&rreq->cache_resources)) { + trace_netfs_rreq(rreq, netfs_rreq_trace_dirty); + netfs_single_mark_inode_dirty(rreq->inode); + } + + if (rreq->iocb) { + rreq->iocb->ki_pos += rreq->transferred; + if (rreq->iocb->ki_complete) + rreq->iocb->ki_complete( + rreq->iocb, rreq->error ? rreq->error : rreq->transferred); + } + if (rreq->netfs_ops->done) + rreq->netfs_ops->done(rreq); +} + +/* + * Perform the collection of subrequests and folios. * * Note that we're in normal kernel thread context at this point, possibly * running on a workqueue. */ -static void netfs_rreq_assess(struct netfs_io_request *rreq) +static void netfs_read_collection(struct netfs_io_request *rreq) { - trace_netfs_rreq(rreq, netfs_rreq_trace_assess); + struct netfs_io_stream *stream = &rreq->io_streams[0]; - //netfs_rreq_is_still_valid(rreq); + netfs_collect_read_results(rreq); - if (test_and_clear_bit(NETFS_RREQ_NEED_RETRY, &rreq->flags)) { - netfs_retry_reads(rreq); + /* We're done when the app thread has finished posting subreqs and the + * queue is empty. + */ + if (!test_bit(NETFS_RREQ_ALL_QUEUED, &rreq->flags)) + return; + smp_rmb(); /* Read ALL_QUEUED before subreq lists. */ + + if (!list_empty(&stream->subrequests)) return; - } - if (rreq->origin == NETFS_DIO_READ || - rreq->origin == NETFS_READ_GAPS) + /* Okay, declare that all I/O is complete. */ + rreq->transferred = stream->transferred; + trace_netfs_rreq(rreq, netfs_rreq_trace_complete); + + //netfs_rreq_is_still_valid(rreq); + + switch (rreq->origin) { + case NETFS_DIO_READ: + case NETFS_READ_GAPS: netfs_rreq_assess_dio(rreq); + break; + case NETFS_READ_SINGLE: + netfs_rreq_assess_single(rreq); + break; + default: + break; + } task_io_account_read(rreq->transferred); trace_netfs_rreq(rreq, netfs_rreq_trace_wake_ip); - clear_bit_unlock(NETFS_RREQ_IN_PROGRESS, &rreq->flags); - wake_up_bit(&rreq->flags, NETFS_RREQ_IN_PROGRESS); + clear_and_wake_up_bit(NETFS_RREQ_IN_PROGRESS, &rreq->flags); trace_netfs_rreq(rreq, netfs_rreq_trace_done); netfs_clear_subrequests(rreq, false); netfs_unlock_abandoned_read_pages(rreq); - if (unlikely(test_bit(NETFS_RREQ_USE_PGPRIV2, &rreq->flags))) - netfs_pgpriv2_write_to_the_cache(rreq); + if (unlikely(rreq->copy_to_cache)) + netfs_pgpriv2_end_copy_to_cache(rreq); } -void netfs_read_termination_worker(struct work_struct *work) +void netfs_read_collection_worker(struct work_struct *work) { - struct netfs_io_request *rreq = - container_of(work, struct netfs_io_request, work); + struct netfs_io_request *rreq = container_of(work, struct netfs_io_request, work); + netfs_see_request(rreq, netfs_rreq_trace_see_work); - netfs_rreq_assess(rreq); - netfs_put_request(rreq, false, netfs_rreq_trace_put_work_complete); + if (test_bit(NETFS_RREQ_IN_PROGRESS, &rreq->flags)) + netfs_read_collection(rreq); + netfs_put_request(rreq, false, netfs_rreq_trace_put_work); } /* - * Handle the completion of all outstanding I/O operations on a read request. - * We inherit a ref from the caller. + * Wake the collection work item. */ -void netfs_rreq_terminated(struct netfs_io_request *rreq, bool was_async) +void netfs_wake_read_collector(struct netfs_io_request *rreq) { - if (!was_async) - return netfs_rreq_assess(rreq); - if (!work_pending(&rreq->work)) { - netfs_get_request(rreq, netfs_rreq_trace_get_work); - if (!queue_work(system_unbound_wq, &rreq->work)) - netfs_put_request(rreq, was_async, netfs_rreq_trace_put_work_nq); + if (test_bit(NETFS_RREQ_OFFLOAD_COLLECTION, &rreq->flags)) { + if (!work_pending(&rreq->work)) { + netfs_get_request(rreq, netfs_rreq_trace_get_work); + if (!queue_work(system_unbound_wq, &rreq->work)) + netfs_put_request(rreq, true, netfs_rreq_trace_put_work_nq); + } + } else { + trace_netfs_rreq(rreq, netfs_rreq_trace_wake_queue); + wake_up(&rreq->waitq); } } /** * netfs_read_subreq_progress - Note progress of a read operation. * @subreq: The read request that has terminated. - * @was_async: True if we're in an asynchronous context. * * This tells the read side of netfs lib that a contributory I/O operation has * made some progress and that it may be possible to unlock some folios. * * Before calling, the filesystem should update subreq->transferred to track * the amount of data copied into the output buffer. - * - * If @was_async is true, the caller might be running in softirq or interrupt - * context and we can't sleep. */ -void netfs_read_subreq_progress(struct netfs_io_subrequest *subreq, - bool was_async) +void netfs_read_subreq_progress(struct netfs_io_subrequest *subreq) { struct netfs_io_request *rreq = subreq->rreq; + struct netfs_io_stream *stream = &rreq->io_streams[0]; + size_t fsize = PAGE_SIZE << rreq->front_folio_order; trace_netfs_sreq(subreq, netfs_sreq_trace_progress); - if (subreq->transferred > subreq->consumed && + /* If we are at the head of the queue, wake up the collector, + * getting a ref to it if we were the ones to do so. + */ + if (subreq->start + subreq->transferred > rreq->cleaned_to + fsize && (rreq->origin == NETFS_READAHEAD || rreq->origin == NETFS_READPAGE || - rreq->origin == NETFS_READ_FOR_WRITE)) { - netfs_consume_read_data(subreq, was_async); - __clear_bit(NETFS_SREQ_NO_PROGRESS, &subreq->flags); + rreq->origin == NETFS_READ_FOR_WRITE) && + list_is_first(&subreq->rreq_link, &stream->subrequests) + ) { + __set_bit(NETFS_SREQ_MADE_PROGRESS, &subreq->flags); + netfs_wake_read_collector(rreq); } } EXPORT_SYMBOL(netfs_read_subreq_progress); @@ -446,27 +518,23 @@ EXPORT_SYMBOL(netfs_read_subreq_progress); /** * netfs_read_subreq_terminated - Note the termination of an I/O operation. * @subreq: The I/O request that has terminated. - * @error: Error code indicating type of completion. - * @was_async: The termination was asynchronous * * This tells the read helper that a contributory I/O operation has terminated, * one way or another, and that it should integrate the results. * - * The caller indicates the outcome of the operation through @error, supplying - * 0 to indicate a successful or retryable transfer (if NETFS_SREQ_NEED_RETRY - * is set) or a negative error code. The helper will look after reissuing I/O - * operations as appropriate and writing downloaded data to the cache. + * The caller indicates the outcome of the operation through @subreq->error, + * supplying 0 to indicate a successful or retryable transfer (if + * NETFS_SREQ_NEED_RETRY is set) or a negative error code. The helper will + * look after reissuing I/O operations as appropriate and writing downloaded + * data to the cache. * * Before calling, the filesystem should update subreq->transferred to track * the amount of data copied into the output buffer. - * - * If @was_async is true, the caller might be running in softirq or interrupt - * context and we can't sleep. */ -void netfs_read_subreq_terminated(struct netfs_io_subrequest *subreq, - int error, bool was_async) +void netfs_read_subreq_terminated(struct netfs_io_subrequest *subreq) { struct netfs_io_request *rreq = subreq->rreq; + struct netfs_io_stream *stream = &rreq->io_streams[0]; switch (subreq->source) { case NETFS_READ_FROM_CACHE: @@ -479,68 +547,156 @@ void netfs_read_subreq_terminated(struct netfs_io_subrequest *subreq, break; } - if (rreq->origin != NETFS_DIO_READ) { - /* Collect buffered reads. - * - * If the read completed validly short, then we can clear the - * tail before going on to unlock the folios. - */ - if (error == 0 && subreq->transferred < subreq->len && - (test_bit(NETFS_SREQ_HIT_EOF, &subreq->flags) || - test_bit(NETFS_SREQ_CLEAR_TAIL, &subreq->flags))) { - netfs_clear_unread(subreq); - subreq->transferred = subreq->len; - trace_netfs_sreq(subreq, netfs_sreq_trace_clear); - } - if (subreq->transferred > subreq->consumed && - (rreq->origin == NETFS_READAHEAD || - rreq->origin == NETFS_READPAGE || - rreq->origin == NETFS_READ_FOR_WRITE)) { - netfs_consume_read_data(subreq, was_async); - __clear_bit(NETFS_SREQ_NO_PROGRESS, &subreq->flags); - } - rreq->transferred += subreq->transferred; - } - /* Deal with retry requests, short reads and errors. If we retry * but don't make progress, we abandon the attempt. */ - if (!error && subreq->transferred < subreq->len) { + if (!subreq->error && subreq->transferred < subreq->len) { if (test_bit(NETFS_SREQ_HIT_EOF, &subreq->flags)) { trace_netfs_sreq(subreq, netfs_sreq_trace_hit_eof); + } else if (test_bit(NETFS_SREQ_CLEAR_TAIL, &subreq->flags)) { + trace_netfs_sreq(subreq, netfs_sreq_trace_need_clear); + } else if (test_bit(NETFS_SREQ_NEED_RETRY, &subreq->flags)) { + trace_netfs_sreq(subreq, netfs_sreq_trace_need_retry); + } else if (test_bit(NETFS_SREQ_MADE_PROGRESS, &subreq->flags)) { + __set_bit(NETFS_SREQ_NEED_RETRY, &subreq->flags); + trace_netfs_sreq(subreq, netfs_sreq_trace_partial_read); } else { + __set_bit(NETFS_SREQ_FAILED, &subreq->flags); + subreq->error = -ENODATA; trace_netfs_sreq(subreq, netfs_sreq_trace_short); - if (subreq->transferred > subreq->consumed) { - __set_bit(NETFS_SREQ_NEED_RETRY, &subreq->flags); - __clear_bit(NETFS_SREQ_NO_PROGRESS, &subreq->flags); - set_bit(NETFS_RREQ_NEED_RETRY, &rreq->flags); - } else if (!__test_and_set_bit(NETFS_SREQ_NO_PROGRESS, &subreq->flags)) { - __set_bit(NETFS_SREQ_NEED_RETRY, &subreq->flags); - set_bit(NETFS_RREQ_NEED_RETRY, &rreq->flags); - } else { - __set_bit(NETFS_SREQ_FAILED, &subreq->flags); - error = -ENODATA; - } } } - subreq->error = error; - trace_netfs_sreq(subreq, netfs_sreq_trace_terminated); - - if (unlikely(error < 0)) { - trace_netfs_failure(rreq, subreq, error, netfs_fail_read); + if (unlikely(subreq->error < 0)) { + trace_netfs_failure(rreq, subreq, subreq->error, netfs_fail_read); if (subreq->source == NETFS_READ_FROM_CACHE) { netfs_stat(&netfs_n_rh_read_failed); + __set_bit(NETFS_SREQ_NEED_RETRY, &subreq->flags); } else { netfs_stat(&netfs_n_rh_download_failed); - set_bit(NETFS_RREQ_FAILED, &rreq->flags); - rreq->error = subreq->error; + __set_bit(NETFS_SREQ_FAILED, &subreq->flags); } + trace_netfs_rreq(rreq, netfs_rreq_trace_set_pause); + set_bit(NETFS_RREQ_PAUSE, &rreq->flags); } - if (atomic_dec_and_test(&rreq->nr_outstanding)) - netfs_rreq_terminated(rreq, was_async); + trace_netfs_sreq(subreq, netfs_sreq_trace_terminated); + + clear_bit_unlock(NETFS_SREQ_IN_PROGRESS, &subreq->flags); + smp_mb__after_atomic(); /* Clear IN_PROGRESS before task state */ + + /* If we are at the head of the queue, wake up the collector. */ + if (list_is_first(&subreq->rreq_link, &stream->subrequests)) + netfs_wake_read_collector(rreq); - netfs_put_subrequest(subreq, was_async, netfs_sreq_trace_put_terminated); + netfs_put_subrequest(subreq, true, netfs_sreq_trace_put_terminated); } EXPORT_SYMBOL(netfs_read_subreq_terminated); + +/* + * Handle termination of a read from the cache. + */ +void netfs_cache_read_terminated(void *priv, ssize_t transferred_or_error, bool was_async) +{ + struct netfs_io_subrequest *subreq = priv; + + if (transferred_or_error > 0) { + subreq->error = 0; + if (transferred_or_error > 0) { + subreq->transferred += transferred_or_error; + __set_bit(NETFS_SREQ_MADE_PROGRESS, &subreq->flags); + } + } else { + subreq->error = transferred_or_error; + } + netfs_read_subreq_terminated(subreq); +} + +/* + * Wait for the read operation to complete, successfully or otherwise. + */ +ssize_t netfs_wait_for_read(struct netfs_io_request *rreq) +{ + struct netfs_io_subrequest *subreq; + struct netfs_io_stream *stream = &rreq->io_streams[0]; + DEFINE_WAIT(myself); + ssize_t ret; + + for (;;) { + trace_netfs_rreq(rreq, netfs_rreq_trace_wait_queue); + prepare_to_wait(&rreq->waitq, &myself, TASK_UNINTERRUPTIBLE); + + subreq = list_first_entry_or_null(&stream->subrequests, + struct netfs_io_subrequest, rreq_link); + if (subreq && + (!test_bit(NETFS_SREQ_IN_PROGRESS, &subreq->flags) || + test_bit(NETFS_SREQ_MADE_PROGRESS, &subreq->flags))) { + __set_current_state(TASK_RUNNING); + netfs_read_collection(rreq); + continue; + } + + if (!test_bit(NETFS_RREQ_IN_PROGRESS, &rreq->flags)) + break; + + schedule(); + trace_netfs_rreq(rreq, netfs_rreq_trace_woke_queue); + } + + finish_wait(&rreq->waitq, &myself); + + ret = rreq->error; + if (ret == 0) { + ret = rreq->transferred; + switch (rreq->origin) { + case NETFS_DIO_READ: + case NETFS_READ_SINGLE: + ret = rreq->transferred; + break; + default: + if (rreq->submitted < rreq->len) { + trace_netfs_failure(rreq, NULL, ret, netfs_fail_short_read); + ret = -EIO; + } + break; + } + } + + return ret; +} + +/* + * Wait for a paused read operation to unpause or complete in some manner. + */ +void netfs_wait_for_pause(struct netfs_io_request *rreq) +{ + struct netfs_io_subrequest *subreq; + struct netfs_io_stream *stream = &rreq->io_streams[0]; + DEFINE_WAIT(myself); + + trace_netfs_rreq(rreq, netfs_rreq_trace_wait_pause); + + for (;;) { + trace_netfs_rreq(rreq, netfs_rreq_trace_wait_queue); + prepare_to_wait(&rreq->waitq, &myself, TASK_UNINTERRUPTIBLE); + + subreq = list_first_entry_or_null(&stream->subrequests, + struct netfs_io_subrequest, rreq_link); + if (subreq && + (!test_bit(NETFS_SREQ_IN_PROGRESS, &subreq->flags) || + test_bit(NETFS_SREQ_MADE_PROGRESS, &subreq->flags))) { + __set_current_state(TASK_RUNNING); + netfs_read_collection(rreq); + continue; + } + + if (!test_bit(NETFS_RREQ_IN_PROGRESS, &rreq->flags) || + !test_bit(NETFS_RREQ_PAUSE, &rreq->flags)) + break; + + schedule(); + trace_netfs_rreq(rreq, netfs_rreq_trace_woke_queue); + } + + finish_wait(&rreq->waitq, &myself); +} diff --git a/fs/netfs/read_pgpriv2.c b/fs/netfs/read_pgpriv2.c index ba5af89d37fa..cf7727060215 100644 --- a/fs/netfs/read_pgpriv2.c +++ b/fs/netfs/read_pgpriv2.c @@ -14,52 +14,11 @@ #include "internal.h" /* - * [DEPRECATED] Mark page as requiring copy-to-cache using PG_private_2. The - * third mark in the folio queue is used to indicate that this folio needs - * writing. - */ -void netfs_pgpriv2_mark_copy_to_cache(struct netfs_io_subrequest *subreq, - struct netfs_io_request *rreq, - struct folio_queue *folioq, - int slot) -{ - struct folio *folio = folioq_folio(folioq, slot); - - trace_netfs_folio(folio, netfs_folio_trace_copy_to_cache); - folio_start_private_2(folio); - folioq_mark3(folioq, slot); -} - -/* - * [DEPRECATED] Cancel PG_private_2 on all marked folios in the event of an - * unrecoverable error. - */ -static void netfs_pgpriv2_cancel(struct folio_queue *folioq) -{ - struct folio *folio; - int slot; - - while (folioq) { - if (!folioq->marks3) { - folioq = folioq->next; - continue; - } - - slot = __ffs(folioq->marks3); - folio = folioq_folio(folioq, slot); - - trace_netfs_folio(folio, netfs_folio_trace_cancel_copy); - folio_end_private_2(folio); - folioq_unmark3(folioq, slot); - } -} - -/* * [DEPRECATED] Copy a folio to the cache with PG_private_2 set. */ -static int netfs_pgpriv2_copy_folio(struct netfs_io_request *wreq, struct folio *folio) +static void netfs_pgpriv2_copy_folio(struct netfs_io_request *creq, struct folio *folio) { - struct netfs_io_stream *cache = &wreq->io_streams[1]; + struct netfs_io_stream *cache = &creq->io_streams[1]; size_t fsize = folio_size(folio), flen = fsize; loff_t fpos = folio_pos(folio), i_size; bool to_eof = false; @@ -70,17 +29,17 @@ static int netfs_pgpriv2_copy_folio(struct netfs_io_request *wreq, struct folio * of the page to beyond it, but cannot move i_size into or through the * page since we have it locked. */ - i_size = i_size_read(wreq->inode); + i_size = i_size_read(creq->inode); if (fpos >= i_size) { /* mmap beyond eof. */ _debug("beyond eof"); folio_end_private_2(folio); - return 0; + return; } - if (fpos + fsize > wreq->i_size) - wreq->i_size = i_size; + if (fpos + fsize > creq->i_size) + creq->i_size = i_size; if (flen > i_size - fpos) { flen = i_size - fpos; @@ -94,8 +53,10 @@ static int netfs_pgpriv2_copy_folio(struct netfs_io_request *wreq, struct folio trace_netfs_folio(folio, netfs_folio_trace_store_copy); /* Attach the folio to the rolling buffer. */ - if (netfs_buffer_append_folio(wreq, folio, false) < 0) - return -ENOMEM; + if (rolling_buffer_append(&creq->buffer, folio, 0) < 0) { + clear_bit(NETFS_RREQ_FOLIO_COPY_TO_CACHE, &creq->flags); + return; + } cache->submit_extendable_to = fsize; cache->submit_off = 0; @@ -109,11 +70,11 @@ static int netfs_pgpriv2_copy_folio(struct netfs_io_request *wreq, struct folio do { ssize_t part; - wreq->io_iter.iov_offset = cache->submit_off; + creq->buffer.iter.iov_offset = cache->submit_off; - atomic64_set(&wreq->issued_to, fpos + cache->submit_off); + atomic64_set(&creq->issued_to, fpos + cache->submit_off); cache->submit_extendable_to = fsize - cache->submit_off; - part = netfs_advance_write(wreq, cache, fpos + cache->submit_off, + part = netfs_advance_write(creq, cache, fpos + cache->submit_off, cache->submit_len, to_eof); cache->submit_off += part; if (part > cache->submit_len) @@ -122,94 +83,95 @@ static int netfs_pgpriv2_copy_folio(struct netfs_io_request *wreq, struct folio cache->submit_len -= part; } while (cache->submit_len > 0); - wreq->io_iter.iov_offset = 0; - iov_iter_advance(&wreq->io_iter, fsize); - atomic64_set(&wreq->issued_to, fpos + fsize); + creq->buffer.iter.iov_offset = 0; + rolling_buffer_advance(&creq->buffer, fsize); + atomic64_set(&creq->issued_to, fpos + fsize); if (flen < fsize) - netfs_issue_write(wreq, cache); - - _leave(" = 0"); - return 0; + netfs_issue_write(creq, cache); } /* - * [DEPRECATED] Go through the buffer and write any folios that are marked with - * the third mark to the cache. + * [DEPRECATED] Set up copying to the cache. */ -void netfs_pgpriv2_write_to_the_cache(struct netfs_io_request *rreq) +static struct netfs_io_request *netfs_pgpriv2_begin_copy_to_cache( + struct netfs_io_request *rreq, struct folio *folio) { - struct netfs_io_request *wreq; - struct folio_queue *folioq; - struct folio *folio; - int error = 0; - int slot = 0; - - _enter(""); + struct netfs_io_request *creq; if (!fscache_resources_valid(&rreq->cache_resources)) - goto couldnt_start; - - /* Need the first folio to be able to set up the op. */ - for (folioq = rreq->buffer; folioq; folioq = folioq->next) { - if (folioq->marks3) { - slot = __ffs(folioq->marks3); - break; - } - } - if (!folioq) - return; - folio = folioq_folio(folioq, slot); + goto cancel; - wreq = netfs_create_write_req(rreq->mapping, NULL, folio_pos(folio), + creq = netfs_create_write_req(rreq->mapping, NULL, folio_pos(folio), NETFS_PGPRIV2_COPY_TO_CACHE); - if (IS_ERR(wreq)) { - kleave(" [create %ld]", PTR_ERR(wreq)); - goto couldnt_start; - } + if (IS_ERR(creq)) + goto cancel; + + if (!creq->io_streams[1].avail) + goto cancel_put; - trace_netfs_write(wreq, netfs_write_trace_copy_to_cache); + trace_netfs_write(creq, netfs_write_trace_copy_to_cache); netfs_stat(&netfs_n_wh_copy_to_cache); + rreq->copy_to_cache = creq; + return creq; + +cancel_put: + netfs_put_request(creq, false, netfs_rreq_trace_put_return); +cancel: + rreq->copy_to_cache = ERR_PTR(-ENOBUFS); + clear_bit(NETFS_RREQ_FOLIO_COPY_TO_CACHE, &rreq->flags); + return ERR_PTR(-ENOBUFS); +} - for (;;) { - error = netfs_pgpriv2_copy_folio(wreq, folio); - if (error < 0) - break; +/* + * [DEPRECATED] Mark page as requiring copy-to-cache using PG_private_2 and add + * it to the copy write request. + */ +void netfs_pgpriv2_copy_to_cache(struct netfs_io_request *rreq, struct folio *folio) +{ + struct netfs_io_request *creq = rreq->copy_to_cache; - folioq_unmark3(folioq, slot); - if (!folioq->marks3) { - folioq = folioq->next; - if (!folioq) - break; - } + if (!creq) + creq = netfs_pgpriv2_begin_copy_to_cache(rreq, folio); + if (IS_ERR(creq)) + return; - slot = __ffs(folioq->marks3); - folio = folioq_folio(folioq, slot); - } + trace_netfs_folio(folio, netfs_folio_trace_copy_to_cache); + folio_start_private_2(folio); + netfs_pgpriv2_copy_folio(creq, folio); +} - netfs_issue_write(wreq, &wreq->io_streams[1]); +/* + * [DEPRECATED] End writing to the cache, flushing out any outstanding writes. + */ +void netfs_pgpriv2_end_copy_to_cache(struct netfs_io_request *rreq) +{ + struct netfs_io_request *creq = rreq->copy_to_cache; + + if (IS_ERR_OR_NULL(creq)) + return; + + netfs_issue_write(creq, &creq->io_streams[1]); smp_wmb(); /* Write lists before ALL_QUEUED. */ - set_bit(NETFS_RREQ_ALL_QUEUED, &wreq->flags); + set_bit(NETFS_RREQ_ALL_QUEUED, &creq->flags); - netfs_put_request(wreq, false, netfs_rreq_trace_put_return); - _leave(" = %d", error); -couldnt_start: - netfs_pgpriv2_cancel(rreq->buffer); + netfs_put_request(creq, false, netfs_rreq_trace_put_return); + creq->copy_to_cache = NULL; } /* * [DEPRECATED] Remove the PG_private_2 mark from any folios we've finished * copying. */ -bool netfs_pgpriv2_unlock_copied_folios(struct netfs_io_request *wreq) +bool netfs_pgpriv2_unlock_copied_folios(struct netfs_io_request *creq) { - struct folio_queue *folioq = wreq->buffer; - unsigned long long collected_to = wreq->collected_to; - unsigned int slot = wreq->buffer_head_slot; + struct folio_queue *folioq = creq->buffer.tail; + unsigned long long collected_to = creq->collected_to; + unsigned int slot = creq->buffer.first_tail_slot; bool made_progress = false; if (slot >= folioq_nr_slots(folioq)) { - folioq = netfs_delete_buffer_head(wreq); + folioq = rolling_buffer_delete_spent(&creq->buffer); slot = 0; } @@ -221,16 +183,16 @@ bool netfs_pgpriv2_unlock_copied_folios(struct netfs_io_request *wreq) folio = folioq_folio(folioq, slot); if (WARN_ONCE(!folio_test_private_2(folio), "R=%08x: folio %lx is not marked private_2\n", - wreq->debug_id, folio->index)) + creq->debug_id, folio->index)) trace_netfs_folio(folio, netfs_folio_trace_not_under_wback); fpos = folio_pos(folio); fsize = folio_size(folio); flen = fsize; - fend = min_t(unsigned long long, fpos + flen, wreq->i_size); + fend = min_t(unsigned long long, fpos + flen, creq->i_size); - trace_netfs_collect_folio(wreq, folio, fend, collected_to); + trace_netfs_collect_folio(creq, folio, fend, collected_to); /* Unlock any folio we've transferred all of. */ if (collected_to < fend) @@ -238,7 +200,7 @@ bool netfs_pgpriv2_unlock_copied_folios(struct netfs_io_request *wreq) trace_netfs_folio(folio, netfs_folio_trace_end_copy); folio_end_private_2(folio); - wreq->cleaned_to = fpos + fsize; + creq->cleaned_to = fpos + fsize; made_progress = true; /* Clean up the head folioq. If we clear an entire folioq, then @@ -248,9 +210,9 @@ bool netfs_pgpriv2_unlock_copied_folios(struct netfs_io_request *wreq) folioq_clear(folioq, slot); slot++; if (slot >= folioq_nr_slots(folioq)) { - if (READ_ONCE(wreq->buffer_tail) == folioq) - break; - folioq = netfs_delete_buffer_head(wreq); + folioq = rolling_buffer_delete_spent(&creq->buffer); + if (!folioq) + goto done; slot = 0; } @@ -258,7 +220,8 @@ bool netfs_pgpriv2_unlock_copied_folios(struct netfs_io_request *wreq) break; } - wreq->buffer = folioq; - wreq->buffer_head_slot = slot; + creq->buffer.tail = folioq; +done: + creq->buffer.first_tail_slot = slot; return made_progress; } diff --git a/fs/netfs/read_retry.c b/fs/netfs/read_retry.c index 0350592ea804..2290af0d51ac 100644 --- a/fs/netfs/read_retry.c +++ b/fs/netfs/read_retry.c @@ -12,15 +12,7 @@ static void netfs_reissue_read(struct netfs_io_request *rreq, struct netfs_io_subrequest *subreq) { - struct iov_iter *io_iter = &subreq->io_iter; - - if (iov_iter_is_folioq(io_iter)) { - subreq->curr_folioq = (struct folio_queue *)io_iter->folioq; - subreq->curr_folioq_slot = io_iter->folioq_slot; - subreq->curr_folio_order = subreq->curr_folioq->orders[subreq->curr_folioq_slot]; - } - - atomic_inc(&rreq->nr_outstanding); + __clear_bit(NETFS_SREQ_MADE_PROGRESS, &subreq->flags); __set_bit(NETFS_SREQ_IN_PROGRESS, &subreq->flags); netfs_get_subrequest(subreq, netfs_sreq_trace_get_resubmit); subreq->rreq->netfs_ops->issue_read(subreq); @@ -33,13 +25,12 @@ static void netfs_reissue_read(struct netfs_io_request *rreq, static void netfs_retry_read_subrequests(struct netfs_io_request *rreq) { struct netfs_io_subrequest *subreq; - struct netfs_io_stream *stream0 = &rreq->io_streams[0]; - LIST_HEAD(sublist); - LIST_HEAD(queue); + struct netfs_io_stream *stream = &rreq->io_streams[0]; + struct list_head *next; _enter("R=%x", rreq->debug_id); - if (list_empty(&rreq->subrequests)) + if (list_empty(&stream->subrequests)) return; if (rreq->netfs_ops->retry_request) @@ -49,13 +40,13 @@ static void netfs_retry_read_subrequests(struct netfs_io_request *rreq) * up to the first permanently failed one. */ if (!rreq->netfs_ops->prepare_read && - !test_bit(NETFS_RREQ_COPY_TO_CACHE, &rreq->flags)) { - struct netfs_io_subrequest *subreq; - - list_for_each_entry(subreq, &rreq->subrequests, rreq_link) { + !rreq->cache_resources.ops) { + list_for_each_entry(subreq, &stream->subrequests, rreq_link) { if (test_bit(NETFS_SREQ_FAILED, &subreq->flags)) break; if (__test_and_clear_bit(NETFS_SREQ_NEED_RETRY, &subreq->flags)) { + __clear_bit(NETFS_SREQ_MADE_PROGRESS, &subreq->flags); + subreq->retry_count++; netfs_reset_iter(subreq); netfs_reissue_read(rreq, subreq); } @@ -73,48 +64,44 @@ static void netfs_retry_read_subrequests(struct netfs_io_request *rreq) * populating with smaller subrequests. In the event that the subreq * we just launched finishes before we insert the next subreq, it'll * fill in rreq->prev_donated instead. - + * * Note: Alternatively, we could split the tail subrequest right before * we reissue it and fix up the donations under lock. */ - list_splice_init(&rreq->subrequests, &queue); + next = stream->subrequests.next; do { - struct netfs_io_subrequest *from; + struct netfs_io_subrequest *from, *to, *tmp; struct iov_iter source; unsigned long long start, len; - size_t part, deferred_next_donated = 0; + size_t part; bool boundary = false; /* Go through the subreqs and find the next span of contiguous * buffer that we then rejig (cifs, for example, needs the * rsize renegotiating) and reissue. */ - from = list_first_entry(&queue, struct netfs_io_subrequest, rreq_link); - list_move_tail(&from->rreq_link, &sublist); + from = list_entry(next, struct netfs_io_subrequest, rreq_link); + to = from; start = from->start + from->transferred; len = from->len - from->transferred; - _debug("from R=%08x[%x] s=%llx ctl=%zx/%zx/%zx", + _debug("from R=%08x[%x] s=%llx ctl=%zx/%zx", rreq->debug_id, from->debug_index, - from->start, from->consumed, from->transferred, from->len); + from->start, from->transferred, from->len); if (test_bit(NETFS_SREQ_FAILED, &from->flags) || !test_bit(NETFS_SREQ_NEED_RETRY, &from->flags)) goto abandon; - deferred_next_donated = from->next_donated; - while ((subreq = list_first_entry_or_null( - &queue, struct netfs_io_subrequest, rreq_link))) { - if (subreq->start != start + len || - subreq->transferred > 0 || + list_for_each_continue(next, &stream->subrequests) { + subreq = list_entry(next, struct netfs_io_subrequest, rreq_link); + if (subreq->start + subreq->transferred != start + len || + test_bit(NETFS_SREQ_BOUNDARY, &subreq->flags) || !test_bit(NETFS_SREQ_NEED_RETRY, &subreq->flags)) break; - list_move_tail(&subreq->rreq_link, &sublist); - len += subreq->len; - deferred_next_donated = subreq->next_donated; - if (test_bit(NETFS_SREQ_BOUNDARY, &subreq->flags)) - break; + to = subreq; + len += to->len; } _debug(" - range: %llx-%llx %llx", start, start + len - 1, len); @@ -127,36 +114,31 @@ static void netfs_retry_read_subrequests(struct netfs_io_request *rreq) source.count = len; /* Work through the sublist. */ - while ((subreq = list_first_entry_or_null( - &sublist, struct netfs_io_subrequest, rreq_link))) { - list_del(&subreq->rreq_link); - + subreq = from; + list_for_each_entry_from(subreq, &stream->subrequests, rreq_link) { + if (!len) + break; subreq->source = NETFS_DOWNLOAD_FROM_SERVER; subreq->start = start - subreq->transferred; subreq->len = len + subreq->transferred; - stream0->sreq_max_len = subreq->len; - __clear_bit(NETFS_SREQ_NEED_RETRY, &subreq->flags); - __set_bit(NETFS_SREQ_RETRYING, &subreq->flags); + __clear_bit(NETFS_SREQ_MADE_PROGRESS, &subreq->flags); + subreq->retry_count++; - spin_lock_bh(&rreq->lock); - list_add_tail(&subreq->rreq_link, &rreq->subrequests); - subreq->prev_donated += rreq->prev_donated; - rreq->prev_donated = 0; trace_netfs_sreq(subreq, netfs_sreq_trace_retry); - spin_unlock_bh(&rreq->lock); - - BUG_ON(!len); /* Renegotiate max_len (rsize) */ - if (rreq->netfs_ops->prepare_read(subreq) < 0) { + stream->sreq_max_len = subreq->len; + if (rreq->netfs_ops->prepare_read && + rreq->netfs_ops->prepare_read(subreq) < 0) { trace_netfs_sreq(subreq, netfs_sreq_trace_reprep_failed); __set_bit(NETFS_SREQ_FAILED, &subreq->flags); + goto abandon; } - part = umin(len, stream0->sreq_max_len); - if (unlikely(rreq->io_streams[0].sreq_max_segs)) - part = netfs_limit_iter(&source, 0, part, stream0->sreq_max_segs); + part = umin(len, stream->sreq_max_len); + if (unlikely(stream->sreq_max_segs)) + part = netfs_limit_iter(&source, 0, part, stream->sreq_max_segs); subreq->len = subreq->transferred + part; subreq->io_iter = source; iov_iter_truncate(&subreq->io_iter, part); @@ -166,58 +148,105 @@ static void netfs_retry_read_subrequests(struct netfs_io_request *rreq) if (!len) { if (boundary) __set_bit(NETFS_SREQ_BOUNDARY, &subreq->flags); - subreq->next_donated = deferred_next_donated; } else { __clear_bit(NETFS_SREQ_BOUNDARY, &subreq->flags); - subreq->next_donated = 0; } + netfs_get_subrequest(subreq, netfs_sreq_trace_get_resubmit); netfs_reissue_read(rreq, subreq); - if (!len) + if (subreq == to) break; - - /* If we ran out of subrequests, allocate another. */ - if (list_empty(&sublist)) { - subreq = netfs_alloc_subrequest(rreq); - if (!subreq) - goto abandon; - subreq->source = NETFS_DOWNLOAD_FROM_SERVER; - subreq->start = start; - - /* We get two refs, but need just one. */ - netfs_put_subrequest(subreq, false, netfs_sreq_trace_new); - trace_netfs_sreq(subreq, netfs_sreq_trace_split); - list_add_tail(&subreq->rreq_link, &sublist); - } } /* If we managed to use fewer subreqs, we can discard the - * excess. + * excess; if we used the same number, then we're done. */ - while ((subreq = list_first_entry_or_null( - &sublist, struct netfs_io_subrequest, rreq_link))) { - trace_netfs_sreq(subreq, netfs_sreq_trace_discard); - list_del(&subreq->rreq_link); - netfs_put_subrequest(subreq, false, netfs_sreq_trace_put_done); + if (!len) { + if (subreq == to) + continue; + list_for_each_entry_safe_from(subreq, tmp, + &stream->subrequests, rreq_link) { + trace_netfs_sreq(subreq, netfs_sreq_trace_discard); + list_del(&subreq->rreq_link); + netfs_put_subrequest(subreq, false, netfs_sreq_trace_put_done); + if (subreq == to) + break; + } + continue; } - } while (!list_empty(&queue)); + /* We ran out of subrequests, so we need to allocate some more + * and insert them after. + */ + do { + subreq = netfs_alloc_subrequest(rreq); + if (!subreq) { + subreq = to; + goto abandon_after; + } + subreq->source = NETFS_DOWNLOAD_FROM_SERVER; + subreq->start = start; + subreq->len = len; + subreq->debug_index = atomic_inc_return(&rreq->subreq_counter); + subreq->stream_nr = stream->stream_nr; + subreq->retry_count = 1; + + trace_netfs_sreq_ref(rreq->debug_id, subreq->debug_index, + refcount_read(&subreq->ref), + netfs_sreq_trace_new); + netfs_get_subrequest(subreq, netfs_sreq_trace_get_resubmit); + + list_add(&subreq->rreq_link, &to->rreq_link); + to = list_next_entry(to, rreq_link); + trace_netfs_sreq(subreq, netfs_sreq_trace_retry); + + stream->sreq_max_len = umin(len, rreq->rsize); + stream->sreq_max_segs = 0; + if (unlikely(stream->sreq_max_segs)) + part = netfs_limit_iter(&source, 0, part, stream->sreq_max_segs); + + netfs_stat(&netfs_n_rh_download); + if (rreq->netfs_ops->prepare_read(subreq) < 0) { + trace_netfs_sreq(subreq, netfs_sreq_trace_reprep_failed); + __set_bit(NETFS_SREQ_FAILED, &subreq->flags); + goto abandon; + } + + part = umin(len, stream->sreq_max_len); + subreq->len = subreq->transferred + part; + subreq->io_iter = source; + iov_iter_truncate(&subreq->io_iter, part); + iov_iter_advance(&source, part); + + len -= part; + start += part; + if (!len && boundary) { + __set_bit(NETFS_SREQ_BOUNDARY, &to->flags); + boundary = false; + } + + netfs_reissue_read(rreq, subreq); + } while (len); + + } while (!list_is_head(next, &stream->subrequests)); return; - /* If we hit ENOMEM, fail all remaining subrequests */ + /* If we hit an error, fail all remaining incomplete subrequests */ +abandon_after: + if (list_is_last(&subreq->rreq_link, &stream->subrequests)) + return; + subreq = list_next_entry(subreq, rreq_link); abandon: - list_splice_init(&sublist, &queue); - list_for_each_entry(subreq, &queue, rreq_link) { - if (!subreq->error) - subreq->error = -ENOMEM; - __clear_bit(NETFS_SREQ_FAILED, &subreq->flags); + list_for_each_entry_from(subreq, &stream->subrequests, rreq_link) { + if (!subreq->error && + !test_bit(NETFS_SREQ_FAILED, &subreq->flags) && + !test_bit(NETFS_SREQ_NEED_RETRY, &subreq->flags)) + continue; + subreq->error = -ENOMEM; + __set_bit(NETFS_SREQ_FAILED, &subreq->flags); __clear_bit(NETFS_SREQ_NEED_RETRY, &subreq->flags); - __clear_bit(NETFS_SREQ_RETRYING, &subreq->flags); } - spin_lock_bh(&rreq->lock); - list_splice_tail_init(&queue, &rreq->subrequests); - spin_unlock_bh(&rreq->lock); } /* @@ -225,14 +254,19 @@ abandon: */ void netfs_retry_reads(struct netfs_io_request *rreq) { - trace_netfs_rreq(rreq, netfs_rreq_trace_resubmit); + struct netfs_io_subrequest *subreq; + struct netfs_io_stream *stream = &rreq->io_streams[0]; - atomic_inc(&rreq->nr_outstanding); + /* Wait for all outstanding I/O to quiesce before performing retries as + * we may need to renegotiate the I/O sizes. + */ + list_for_each_entry(subreq, &stream->subrequests, rreq_link) { + wait_on_bit(&subreq->flags, NETFS_SREQ_IN_PROGRESS, + TASK_UNINTERRUPTIBLE); + } + trace_netfs_rreq(rreq, netfs_rreq_trace_resubmit); netfs_retry_read_subrequests(rreq); - - if (atomic_dec_and_test(&rreq->nr_outstanding)) - netfs_rreq_terminated(rreq, false); } /* @@ -243,7 +277,7 @@ void netfs_unlock_abandoned_read_pages(struct netfs_io_request *rreq) { struct folio_queue *p; - for (p = rreq->buffer; p; p = p->next) { + for (p = rreq->buffer.tail; p; p = p->next) { for (int slot = 0; slot < folioq_count(p); slot++) { struct folio *folio = folioq_folio(p, slot); diff --git a/fs/netfs/read_single.c b/fs/netfs/read_single.c new file mode 100644 index 000000000000..fea0ecdecc53 --- /dev/null +++ b/fs/netfs/read_single.c @@ -0,0 +1,195 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* Single, monolithic object support (e.g. AFS directory). + * + * Copyright (C) 2024 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + */ + +#include <linux/export.h> +#include <linux/fs.h> +#include <linux/mm.h> +#include <linux/pagemap.h> +#include <linux/slab.h> +#include <linux/uio.h> +#include <linux/sched/mm.h> +#include <linux/task_io_accounting_ops.h> +#include <linux/netfs.h> +#include "internal.h" + +/** + * netfs_single_mark_inode_dirty - Mark a single, monolithic object inode dirty + * @inode: The inode to mark + * + * Mark an inode that contains a single, monolithic object as dirty so that its + * writepages op will get called. If set, the SINGLE_NO_UPLOAD flag indicates + * that the object will only be written to the cache and not uploaded (e.g. AFS + * directory contents). + */ +void netfs_single_mark_inode_dirty(struct inode *inode) +{ + struct netfs_inode *ictx = netfs_inode(inode); + bool cache_only = test_bit(NETFS_ICTX_SINGLE_NO_UPLOAD, &ictx->flags); + bool caching = fscache_cookie_enabled(netfs_i_cookie(netfs_inode(inode))); + + if (cache_only && !caching) + return; + + mark_inode_dirty(inode); + + if (caching && !(inode->i_state & I_PINNING_NETFS_WB)) { + bool need_use = false; + + spin_lock(&inode->i_lock); + if (!(inode->i_state & I_PINNING_NETFS_WB)) { + inode->i_state |= I_PINNING_NETFS_WB; + need_use = true; + } + spin_unlock(&inode->i_lock); + + if (need_use) + fscache_use_cookie(netfs_i_cookie(ictx), true); + } + +} +EXPORT_SYMBOL(netfs_single_mark_inode_dirty); + +static int netfs_single_begin_cache_read(struct netfs_io_request *rreq, struct netfs_inode *ctx) +{ + return fscache_begin_read_operation(&rreq->cache_resources, netfs_i_cookie(ctx)); +} + +static void netfs_single_cache_prepare_read(struct netfs_io_request *rreq, + struct netfs_io_subrequest *subreq) +{ + struct netfs_cache_resources *cres = &rreq->cache_resources; + + if (!cres->ops) { + subreq->source = NETFS_DOWNLOAD_FROM_SERVER; + return; + } + subreq->source = cres->ops->prepare_read(subreq, rreq->i_size); + trace_netfs_sreq(subreq, netfs_sreq_trace_prepare); + +} + +static void netfs_single_read_cache(struct netfs_io_request *rreq, + struct netfs_io_subrequest *subreq) +{ + struct netfs_cache_resources *cres = &rreq->cache_resources; + + _enter("R=%08x[%x]", rreq->debug_id, subreq->debug_index); + netfs_stat(&netfs_n_rh_read); + cres->ops->read(cres, subreq->start, &subreq->io_iter, NETFS_READ_HOLE_FAIL, + netfs_cache_read_terminated, subreq); +} + +/* + * Perform a read to a buffer from the cache or the server. Only a single + * subreq is permitted as the object must be fetched in a single transaction. + */ +static int netfs_single_dispatch_read(struct netfs_io_request *rreq) +{ + struct netfs_io_stream *stream = &rreq->io_streams[0]; + struct netfs_io_subrequest *subreq; + int ret = 0; + + subreq = netfs_alloc_subrequest(rreq); + if (!subreq) + return -ENOMEM; + + subreq->source = NETFS_SOURCE_UNKNOWN; + subreq->start = 0; + subreq->len = rreq->len; + subreq->io_iter = rreq->buffer.iter; + + __set_bit(NETFS_SREQ_IN_PROGRESS, &subreq->flags); + + spin_lock(&rreq->lock); + list_add_tail(&subreq->rreq_link, &stream->subrequests); + trace_netfs_sreq(subreq, netfs_sreq_trace_added); + stream->front = subreq; + /* Store list pointers before active flag */ + smp_store_release(&stream->active, true); + spin_unlock(&rreq->lock); + + netfs_single_cache_prepare_read(rreq, subreq); + switch (subreq->source) { + case NETFS_DOWNLOAD_FROM_SERVER: + netfs_stat(&netfs_n_rh_download); + if (rreq->netfs_ops->prepare_read) { + ret = rreq->netfs_ops->prepare_read(subreq); + if (ret < 0) + goto cancel; + } + + rreq->netfs_ops->issue_read(subreq); + rreq->submitted += subreq->len; + break; + case NETFS_READ_FROM_CACHE: + trace_netfs_sreq(subreq, netfs_sreq_trace_submit); + netfs_single_read_cache(rreq, subreq); + rreq->submitted += subreq->len; + ret = 0; + break; + default: + pr_warn("Unexpected single-read source %u\n", subreq->source); + WARN_ON_ONCE(true); + ret = -EIO; + break; + } + + smp_wmb(); /* Write lists before ALL_QUEUED. */ + set_bit(NETFS_RREQ_ALL_QUEUED, &rreq->flags); + return ret; +cancel: + netfs_put_subrequest(subreq, false, netfs_sreq_trace_put_cancel); + return ret; +} + +/** + * netfs_read_single - Synchronously read a single blob of pages. + * @inode: The inode to read from. + * @file: The file we're using to read or NULL. + * @iter: The buffer we're reading into. + * + * Fulfil a read request for a single monolithic object by drawing data from + * the cache if possible, or the netfs if not. The buffer may be larger than + * the file content; unused beyond the EOF will be zero-filled. The content + * will be read with a single I/O request (though this may be retried). + * + * The calling netfs must initialise a netfs context contiguous to the vfs + * inode before calling this. + * + * This is usable whether or not caching is enabled. If caching is enabled, + * the data will be stored as a single object into the cache. + */ +ssize_t netfs_read_single(struct inode *inode, struct file *file, struct iov_iter *iter) +{ + struct netfs_io_request *rreq; + struct netfs_inode *ictx = netfs_inode(inode); + ssize_t ret; + + rreq = netfs_alloc_request(inode->i_mapping, file, 0, iov_iter_count(iter), + NETFS_READ_SINGLE); + if (IS_ERR(rreq)) + return PTR_ERR(rreq); + + ret = netfs_single_begin_cache_read(rreq, ictx); + if (ret == -ENOMEM || ret == -EINTR || ret == -ERESTARTSYS) + goto cleanup_free; + + netfs_stat(&netfs_n_rh_read_single); + trace_netfs_read(rreq, 0, rreq->len, netfs_read_trace_read_single); + + rreq->buffer.iter = *iter; + netfs_single_dispatch_read(rreq); + + ret = netfs_wait_for_read(rreq); + netfs_put_request(rreq, true, netfs_rreq_trace_put_return); + return ret; + +cleanup_free: + netfs_put_request(rreq, false, netfs_rreq_trace_put_failed); + return ret; +} +EXPORT_SYMBOL(netfs_read_single); diff --git a/fs/netfs/rolling_buffer.c b/fs/netfs/rolling_buffer.c new file mode 100644 index 000000000000..75d97af14b4a --- /dev/null +++ b/fs/netfs/rolling_buffer.c @@ -0,0 +1,226 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* Rolling buffer helpers + * + * Copyright (C) 2024 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + */ + +#include <linux/bitops.h> +#include <linux/pagemap.h> +#include <linux/rolling_buffer.h> +#include <linux/slab.h> +#include "internal.h" + +static atomic_t debug_ids; + +/** + * netfs_folioq_alloc - Allocate a folio_queue struct + * @rreq_id: Associated debugging ID for tracing purposes + * @gfp: Allocation constraints + * @trace: Trace tag to indicate the purpose of the allocation + * + * Allocate, initialise and account the folio_queue struct and log a trace line + * to mark the allocation. + */ +struct folio_queue *netfs_folioq_alloc(unsigned int rreq_id, gfp_t gfp, + unsigned int /*enum netfs_folioq_trace*/ trace) +{ + struct folio_queue *fq; + + fq = kmalloc(sizeof(*fq), gfp); + if (fq) { + netfs_stat(&netfs_n_folioq); + folioq_init(fq, rreq_id); + fq->debug_id = atomic_inc_return(&debug_ids); + trace_netfs_folioq(fq, trace); + } + return fq; +} +EXPORT_SYMBOL(netfs_folioq_alloc); + +/** + * netfs_folioq_free - Free a folio_queue struct + * @folioq: The object to free + * @trace: Trace tag to indicate which free + * + * Free and unaccount the folio_queue struct. + */ +void netfs_folioq_free(struct folio_queue *folioq, + unsigned int /*enum netfs_trace_folioq*/ trace) +{ + trace_netfs_folioq(folioq, trace); + netfs_stat_d(&netfs_n_folioq); + kfree(folioq); +} +EXPORT_SYMBOL(netfs_folioq_free); + +/* + * Initialise a rolling buffer. We allocate an empty folio queue struct to so + * that the pointers can be independently driven by the producer and the + * consumer. + */ +int rolling_buffer_init(struct rolling_buffer *roll, unsigned int rreq_id, + unsigned int direction) +{ + struct folio_queue *fq; + + fq = netfs_folioq_alloc(rreq_id, GFP_NOFS, netfs_trace_folioq_rollbuf_init); + if (!fq) + return -ENOMEM; + + roll->head = fq; + roll->tail = fq; + iov_iter_folio_queue(&roll->iter, direction, fq, 0, 0, 0); + return 0; +} + +/* + * Add another folio_queue to a rolling buffer if there's no space left. + */ +int rolling_buffer_make_space(struct rolling_buffer *roll) +{ + struct folio_queue *fq, *head = roll->head; + + if (!folioq_full(head)) + return 0; + + fq = netfs_folioq_alloc(head->rreq_id, GFP_NOFS, netfs_trace_folioq_make_space); + if (!fq) + return -ENOMEM; + fq->prev = head; + + roll->head = fq; + if (folioq_full(head)) { + /* Make sure we don't leave the master iterator pointing to a + * block that might get immediately consumed. + */ + if (roll->iter.folioq == head && + roll->iter.folioq_slot == folioq_nr_slots(head)) { + roll->iter.folioq = fq; + roll->iter.folioq_slot = 0; + } + } + + /* Make sure the initialisation is stored before the next pointer. + * + * [!] NOTE: After we set head->next, the consumer is at liberty to + * immediately delete the old head. + */ + smp_store_release(&head->next, fq); + return 0; +} + +/* + * Decant the list of folios to read into a rolling buffer. + */ +ssize_t rolling_buffer_load_from_ra(struct rolling_buffer *roll, + struct readahead_control *ractl, + struct folio_batch *put_batch) +{ + struct folio_queue *fq; + struct page **vec; + int nr, ix, to; + ssize_t size = 0; + + if (rolling_buffer_make_space(roll) < 0) + return -ENOMEM; + + fq = roll->head; + vec = (struct page **)fq->vec.folios; + nr = __readahead_batch(ractl, vec + folio_batch_count(&fq->vec), + folio_batch_space(&fq->vec)); + ix = fq->vec.nr; + to = ix + nr; + fq->vec.nr = to; + for (; ix < to; ix++) { + struct folio *folio = folioq_folio(fq, ix); + unsigned int order = folio_order(folio); + + fq->orders[ix] = order; + size += PAGE_SIZE << order; + trace_netfs_folio(folio, netfs_folio_trace_read); + if (!folio_batch_add(put_batch, folio)) + folio_batch_release(put_batch); + } + WRITE_ONCE(roll->iter.count, roll->iter.count + size); + + /* Store the counter after setting the slot. */ + smp_store_release(&roll->next_head_slot, to); + + for (; ix < folioq_nr_slots(fq); ix++) + folioq_clear(fq, ix); + + return size; +} + +/* + * Append a folio to the rolling buffer. + */ +ssize_t rolling_buffer_append(struct rolling_buffer *roll, struct folio *folio, + unsigned int flags) +{ + ssize_t size = folio_size(folio); + int slot; + + if (rolling_buffer_make_space(roll) < 0) + return -ENOMEM; + + slot = folioq_append(roll->head, folio); + if (flags & ROLLBUF_MARK_1) + folioq_mark(roll->head, slot); + if (flags & ROLLBUF_MARK_2) + folioq_mark2(roll->head, slot); + + WRITE_ONCE(roll->iter.count, roll->iter.count + size); + + /* Store the counter after setting the slot. */ + smp_store_release(&roll->next_head_slot, slot); + return size; +} + +/* + * Delete a spent buffer from a rolling queue and return the next in line. We + * don't return the last buffer to keep the pointers independent, but return + * NULL instead. + */ +struct folio_queue *rolling_buffer_delete_spent(struct rolling_buffer *roll) +{ + struct folio_queue *spent = roll->tail, *next = READ_ONCE(spent->next); + + if (!next) + return NULL; + next->prev = NULL; + netfs_folioq_free(spent, netfs_trace_folioq_delete); + roll->tail = next; + return next; +} + +/* + * Clear out a rolling queue. Folios that have mark 1 set are put. + */ +void rolling_buffer_clear(struct rolling_buffer *roll) +{ + struct folio_batch fbatch; + struct folio_queue *p; + + folio_batch_init(&fbatch); + + while ((p = roll->tail)) { + roll->tail = p->next; + for (int slot = 0; slot < folioq_count(p); slot++) { + struct folio *folio = folioq_folio(p, slot); + + if (!folio) + continue; + if (folioq_is_marked(p, slot)) { + trace_netfs_folio(folio, netfs_folio_trace_put); + if (!folio_batch_add(&fbatch, folio)) + folio_batch_release(&fbatch); + } + } + + netfs_folioq_free(p, netfs_trace_folioq_clear); + } + + folio_batch_release(&fbatch); +} diff --git a/fs/netfs/stats.c b/fs/netfs/stats.c index 8e63516b40f6..f1af344266cc 100644 --- a/fs/netfs/stats.c +++ b/fs/netfs/stats.c @@ -12,6 +12,7 @@ atomic_t netfs_n_rh_dio_read; atomic_t netfs_n_rh_readahead; atomic_t netfs_n_rh_read_folio; +atomic_t netfs_n_rh_read_single; atomic_t netfs_n_rh_rreq; atomic_t netfs_n_rh_sreq; atomic_t netfs_n_rh_download; @@ -46,10 +47,11 @@ atomic_t netfs_n_folioq; int netfs_stats_show(struct seq_file *m, void *v) { - seq_printf(m, "Reads : DR=%u RA=%u RF=%u WB=%u WBZ=%u\n", + seq_printf(m, "Reads : DR=%u RA=%u RF=%u RS=%u WB=%u WBZ=%u\n", atomic_read(&netfs_n_rh_dio_read), atomic_read(&netfs_n_rh_readahead), atomic_read(&netfs_n_rh_read_folio), + atomic_read(&netfs_n_rh_read_single), atomic_read(&netfs_n_rh_write_begin), atomic_read(&netfs_n_rh_write_zskip)); seq_printf(m, "Writes : BW=%u WT=%u DW=%u WP=%u 2C=%u\n", diff --git a/fs/netfs/write_collect.c b/fs/netfs/write_collect.c index 1d438be2e1b4..294f67795f79 100644 --- a/fs/netfs/write_collect.c +++ b/fs/netfs/write_collect.c @@ -17,10 +17,38 @@ #define HIT_PENDING 0x01 /* A front op was still pending */ #define NEED_REASSESS 0x02 /* Need to loop round and reassess */ #define MADE_PROGRESS 0x04 /* Made progress cleaning up a stream or the folio set */ -#define BUFFERED 0x08 /* The pagecache needs cleaning up */ +#define NEED_UNLOCK 0x08 /* The pagecache needs unlocking */ #define NEED_RETRY 0x10 /* A front op requests retrying */ #define SAW_FAILURE 0x20 /* One stream or hit a permanent failure */ +static void netfs_dump_request(const struct netfs_io_request *rreq) +{ + pr_err("Request R=%08x r=%d fl=%lx or=%x e=%ld\n", + rreq->debug_id, refcount_read(&rreq->ref), rreq->flags, + rreq->origin, rreq->error); + pr_err(" st=%llx tsl=%zx/%llx/%llx\n", + rreq->start, rreq->transferred, rreq->submitted, rreq->len); + pr_err(" cci=%llx/%llx/%llx\n", + rreq->cleaned_to, rreq->collected_to, atomic64_read(&rreq->issued_to)); + pr_err(" iw=%pSR\n", rreq->netfs_ops->issue_write); + for (int i = 0; i < NR_IO_STREAMS; i++) { + const struct netfs_io_subrequest *sreq; + const struct netfs_io_stream *s = &rreq->io_streams[i]; + + pr_err(" str[%x] s=%x e=%d acnf=%u,%u,%u,%u\n", + s->stream_nr, s->source, s->error, + s->avail, s->active, s->need_retry, s->failed); + pr_err(" str[%x] ct=%llx t=%zx\n", + s->stream_nr, s->collected_to, s->transferred); + list_for_each_entry(sreq, &s->subrequests, rreq_link) { + pr_err(" sreq[%x:%x] sc=%u s=%llx t=%zx/%zx r=%d f=%lx\n", + sreq->stream_nr, sreq->debug_index, sreq->source, + sreq->start, sreq->transferred, sreq->len, + refcount_read(&sreq->ref), sreq->flags); + } + } +} + /* * Successful completion of write of a folio to the server and/or cache. Note * that we are not allowed to lock the folio here on pain of deadlocking with @@ -83,9 +111,15 @@ end_wb: static void netfs_writeback_unlock_folios(struct netfs_io_request *wreq, unsigned int *notes) { - struct folio_queue *folioq = wreq->buffer; + struct folio_queue *folioq = wreq->buffer.tail; unsigned long long collected_to = wreq->collected_to; - unsigned int slot = wreq->buffer_head_slot; + unsigned int slot = wreq->buffer.first_tail_slot; + + if (WARN_ON_ONCE(!folioq)) { + pr_err("[!] Writeback unlock found empty rolling buffer!\n"); + netfs_dump_request(wreq); + return; + } if (wreq->origin == NETFS_PGPRIV2_COPY_TO_CACHE) { if (netfs_pgpriv2_unlock_copied_folios(wreq)) @@ -94,7 +128,9 @@ static void netfs_writeback_unlock_folios(struct netfs_io_request *wreq, } if (slot >= folioq_nr_slots(folioq)) { - folioq = netfs_delete_buffer_head(wreq); + folioq = rolling_buffer_delete_spent(&wreq->buffer); + if (!folioq) + return; slot = 0; } @@ -134,9 +170,9 @@ static void netfs_writeback_unlock_folios(struct netfs_io_request *wreq, folioq_clear(folioq, slot); slot++; if (slot >= folioq_nr_slots(folioq)) { - if (READ_ONCE(wreq->buffer_tail) == folioq) - break; - folioq = netfs_delete_buffer_head(wreq); + folioq = rolling_buffer_delete_spent(&wreq->buffer); + if (!folioq) + goto done; slot = 0; } @@ -144,223 +180,9 @@ static void netfs_writeback_unlock_folios(struct netfs_io_request *wreq, break; } - wreq->buffer = folioq; - wreq->buffer_head_slot = slot; -} - -/* - * Perform retries on the streams that need it. - */ -static void netfs_retry_write_stream(struct netfs_io_request *wreq, - struct netfs_io_stream *stream) -{ - struct list_head *next; - - _enter("R=%x[%x:]", wreq->debug_id, stream->stream_nr); - - if (list_empty(&stream->subrequests)) - return; - - if (stream->source == NETFS_UPLOAD_TO_SERVER && - wreq->netfs_ops->retry_request) - wreq->netfs_ops->retry_request(wreq, stream); - - if (unlikely(stream->failed)) - return; - - /* If there's no renegotiation to do, just resend each failed subreq. */ - if (!stream->prepare_write) { - struct netfs_io_subrequest *subreq; - - list_for_each_entry(subreq, &stream->subrequests, rreq_link) { - if (test_bit(NETFS_SREQ_FAILED, &subreq->flags)) - break; - if (__test_and_clear_bit(NETFS_SREQ_NEED_RETRY, &subreq->flags)) { - struct iov_iter source = subreq->io_iter; - - iov_iter_revert(&source, subreq->len - source.count); - __set_bit(NETFS_SREQ_RETRYING, &subreq->flags); - netfs_get_subrequest(subreq, netfs_sreq_trace_get_resubmit); - netfs_reissue_write(stream, subreq, &source); - } - } - return; - } - - next = stream->subrequests.next; - - do { - struct netfs_io_subrequest *subreq = NULL, *from, *to, *tmp; - struct iov_iter source; - unsigned long long start, len; - size_t part; - bool boundary = false; - - /* Go through the stream and find the next span of contiguous - * data that we then rejig (cifs, for example, needs the wsize - * renegotiating) and reissue. - */ - from = list_entry(next, struct netfs_io_subrequest, rreq_link); - to = from; - start = from->start + from->transferred; - len = from->len - from->transferred; - - if (test_bit(NETFS_SREQ_FAILED, &from->flags) || - !test_bit(NETFS_SREQ_NEED_RETRY, &from->flags)) - return; - - list_for_each_continue(next, &stream->subrequests) { - subreq = list_entry(next, struct netfs_io_subrequest, rreq_link); - if (subreq->start + subreq->transferred != start + len || - test_bit(NETFS_SREQ_BOUNDARY, &subreq->flags) || - !test_bit(NETFS_SREQ_NEED_RETRY, &subreq->flags)) - break; - to = subreq; - len += to->len; - } - - /* Determine the set of buffers we're going to use. Each - * subreq gets a subset of a single overall contiguous buffer. - */ - netfs_reset_iter(from); - source = from->io_iter; - source.count = len; - - /* Work through the sublist. */ - subreq = from; - list_for_each_entry_from(subreq, &stream->subrequests, rreq_link) { - if (!len) - break; - /* Renegotiate max_len (wsize) */ - trace_netfs_sreq(subreq, netfs_sreq_trace_retry); - __clear_bit(NETFS_SREQ_NEED_RETRY, &subreq->flags); - __set_bit(NETFS_SREQ_RETRYING, &subreq->flags); - stream->prepare_write(subreq); - - part = min(len, stream->sreq_max_len); - subreq->len = part; - subreq->start = start; - subreq->transferred = 0; - len -= part; - start += part; - if (len && subreq == to && - __test_and_clear_bit(NETFS_SREQ_BOUNDARY, &to->flags)) - boundary = true; - - netfs_get_subrequest(subreq, netfs_sreq_trace_get_resubmit); - netfs_reissue_write(stream, subreq, &source); - if (subreq == to) - break; - } - - /* If we managed to use fewer subreqs, we can discard the - * excess; if we used the same number, then we're done. - */ - if (!len) { - if (subreq == to) - continue; - list_for_each_entry_safe_from(subreq, tmp, - &stream->subrequests, rreq_link) { - trace_netfs_sreq(subreq, netfs_sreq_trace_discard); - list_del(&subreq->rreq_link); - netfs_put_subrequest(subreq, false, netfs_sreq_trace_put_done); - if (subreq == to) - break; - } - continue; - } - - /* We ran out of subrequests, so we need to allocate some more - * and insert them after. - */ - do { - subreq = netfs_alloc_subrequest(wreq); - subreq->source = to->source; - subreq->start = start; - subreq->debug_index = atomic_inc_return(&wreq->subreq_counter); - subreq->stream_nr = to->stream_nr; - __set_bit(NETFS_SREQ_RETRYING, &subreq->flags); - - trace_netfs_sreq_ref(wreq->debug_id, subreq->debug_index, - refcount_read(&subreq->ref), - netfs_sreq_trace_new); - netfs_get_subrequest(subreq, netfs_sreq_trace_get_resubmit); - - list_add(&subreq->rreq_link, &to->rreq_link); - to = list_next_entry(to, rreq_link); - trace_netfs_sreq(subreq, netfs_sreq_trace_retry); - - stream->sreq_max_len = len; - stream->sreq_max_segs = INT_MAX; - switch (stream->source) { - case NETFS_UPLOAD_TO_SERVER: - netfs_stat(&netfs_n_wh_upload); - stream->sreq_max_len = umin(len, wreq->wsize); - break; - case NETFS_WRITE_TO_CACHE: - netfs_stat(&netfs_n_wh_write); - break; - default: - WARN_ON_ONCE(1); - } - - stream->prepare_write(subreq); - - part = umin(len, stream->sreq_max_len); - subreq->len = subreq->transferred + part; - len -= part; - start += part; - if (!len && boundary) { - __set_bit(NETFS_SREQ_BOUNDARY, &to->flags); - boundary = false; - } - - netfs_reissue_write(stream, subreq, &source); - if (!len) - break; - - } while (len); - - } while (!list_is_head(next, &stream->subrequests)); -} - -/* - * Perform retries on the streams that need it. If we're doing content - * encryption and the server copy changed due to a third-party write, we may - * need to do an RMW cycle and also rewrite the data to the cache. - */ -static void netfs_retry_writes(struct netfs_io_request *wreq) -{ - struct netfs_io_subrequest *subreq; - struct netfs_io_stream *stream; - int s; - - /* Wait for all outstanding I/O to quiesce before performing retries as - * we may need to renegotiate the I/O sizes. - */ - for (s = 0; s < NR_IO_STREAMS; s++) { - stream = &wreq->io_streams[s]; - if (!stream->active) - continue; - - list_for_each_entry(subreq, &stream->subrequests, rreq_link) { - wait_on_bit(&subreq->flags, NETFS_SREQ_IN_PROGRESS, - TASK_UNINTERRUPTIBLE); - } - } - - // TODO: Enc: Fetch changed partial pages - // TODO: Enc: Reencrypt content if needed. - // TODO: Enc: Wind back transferred point. - // TODO: Enc: Mark cache pages for retry. - - for (s = 0; s < NR_IO_STREAMS; s++) { - stream = &wreq->io_streams[s]; - if (stream->need_retry) { - stream->need_retry = false; - netfs_retry_write_stream(wreq, stream); - } - } + wreq->buffer.tail = folioq; +done: + wreq->buffer.first_tail_slot = slot; } /* @@ -391,7 +213,7 @@ reassess_streams: if (wreq->origin == NETFS_WRITEBACK || wreq->origin == NETFS_WRITETHROUGH || wreq->origin == NETFS_PGPRIV2_COPY_TO_CACHE) - notes = BUFFERED; + notes = NEED_UNLOCK; else notes = 0; @@ -450,14 +272,14 @@ reassess_streams: cancel: /* Remove if completely consumed. */ - spin_lock_bh(&wreq->lock); + spin_lock(&wreq->lock); remove = front; list_del_init(&front->rreq_link); front = list_first_entry_or_null(&stream->subrequests, struct netfs_io_subrequest, rreq_link); stream->front = front; - spin_unlock_bh(&wreq->lock); + spin_unlock(&wreq->lock); netfs_put_subrequest(remove, false, notes & SAW_FAILURE ? netfs_sreq_trace_put_cancel : @@ -488,7 +310,7 @@ reassess_streams: trace_netfs_collect_state(wreq, wreq->collected_to, notes); /* Unlock any folios that we have now finished with. */ - if (notes & BUFFERED) { + if (notes & NEED_UNLOCK) { if (wreq->cleaned_to < wreq->collected_to) netfs_writeback_unlock_folios(wreq, ¬es); } else { @@ -502,7 +324,8 @@ reassess_streams: if ((notes & MADE_PROGRESS) && test_bit(NETFS_RREQ_PAUSE, &wreq->flags)) { trace_netfs_rreq(wreq, netfs_rreq_trace_unpause); clear_bit_unlock(NETFS_RREQ_PAUSE, &wreq->flags); - wake_up_bit(&wreq->flags, NETFS_RREQ_PAUSE); + smp_mb__after_atomic(); /* Set PAUSE before task state */ + wake_up(&wreq->waitq); } if (notes & NEED_REASSESS) { @@ -605,8 +428,7 @@ void netfs_write_collection_worker(struct work_struct *work) _debug("finished"); trace_netfs_rreq(wreq, netfs_rreq_trace_wake_ip); - clear_bit_unlock(NETFS_RREQ_IN_PROGRESS, &wreq->flags); - wake_up_bit(&wreq->flags, NETFS_RREQ_IN_PROGRESS); + clear_and_wake_up_bit(NETFS_RREQ_IN_PROGRESS, &wreq->flags); if (wreq->iocb) { size_t written = min(wreq->transferred, wreq->len); @@ -714,8 +536,7 @@ void netfs_write_subrequest_terminated(void *_op, ssize_t transferred_or_error, trace_netfs_sreq(subreq, netfs_sreq_trace_terminated); - clear_bit_unlock(NETFS_SREQ_IN_PROGRESS, &subreq->flags); - wake_up_bit(&subreq->flags, NETFS_SREQ_IN_PROGRESS); + clear_and_wake_up_bit(NETFS_SREQ_IN_PROGRESS, &subreq->flags); /* If we are at the head of the queue, wake up the collector, * transferring a ref to it if we were the ones to do so. diff --git a/fs/netfs/write_issue.c b/fs/netfs/write_issue.c index bf6d507578e5..69727411683e 100644 --- a/fs/netfs/write_issue.c +++ b/fs/netfs/write_issue.c @@ -94,9 +94,10 @@ struct netfs_io_request *netfs_create_write_req(struct address_space *mapping, { struct netfs_io_request *wreq; struct netfs_inode *ictx; - bool is_buffered = (origin == NETFS_WRITEBACK || - origin == NETFS_WRITETHROUGH || - origin == NETFS_PGPRIV2_COPY_TO_CACHE); + bool is_cacheable = (origin == NETFS_WRITEBACK || + origin == NETFS_WRITEBACK_SINGLE || + origin == NETFS_WRITETHROUGH || + origin == NETFS_PGPRIV2_COPY_TO_CACHE); wreq = netfs_alloc_request(mapping, file, start, 0, origin); if (IS_ERR(wreq)) @@ -105,8 +106,10 @@ struct netfs_io_request *netfs_create_write_req(struct address_space *mapping, _enter("R=%x", wreq->debug_id); ictx = netfs_inode(wreq->inode); - if (is_buffered && netfs_is_cache_enabled(ictx)) + if (is_cacheable && netfs_is_cache_enabled(ictx)) fscache_begin_write_operation(&wreq->cache_resources, netfs_i_cookie(ictx)); + if (rolling_buffer_init(&wreq->buffer, wreq->debug_id, ITER_SOURCE) < 0) + goto nomem; wreq->cleaned_to = wreq->start; @@ -129,6 +132,10 @@ struct netfs_io_request *netfs_create_write_req(struct address_space *mapping, } return wreq; +nomem: + wreq->error = -ENOMEM; + netfs_put_request(wreq, false, netfs_rreq_trace_put_failed); + return ERR_PTR(-ENOMEM); } /** @@ -153,16 +160,15 @@ static void netfs_prepare_write(struct netfs_io_request *wreq, loff_t start) { struct netfs_io_subrequest *subreq; - struct iov_iter *wreq_iter = &wreq->io_iter; + struct iov_iter *wreq_iter = &wreq->buffer.iter; /* Make sure we don't point the iterator at a used-up folio_queue * struct being used as a placeholder to prevent the queue from * collapsing. In such a case, extend the queue. */ if (iov_iter_is_folioq(wreq_iter) && - wreq_iter->folioq_slot >= folioq_nr_slots(wreq_iter->folioq)) { - netfs_buffer_make_space(wreq); - } + wreq_iter->folioq_slot >= folioq_nr_slots(wreq_iter->folioq)) + rolling_buffer_make_space(&wreq->buffer); subreq = netfs_alloc_subrequest(wreq); subreq->source = stream->source; @@ -198,7 +204,7 @@ static void netfs_prepare_write(struct netfs_io_request *wreq, * the list. The collector only goes nextwards and uses the lock to * remove entries off of the front. */ - spin_lock_bh(&wreq->lock); + spin_lock(&wreq->lock); list_add_tail(&subreq->rreq_link, &stream->subrequests); if (list_is_first(&subreq->rreq_link, &stream->subrequests)) { stream->front = subreq; @@ -209,7 +215,7 @@ static void netfs_prepare_write(struct netfs_io_request *wreq, } } - spin_unlock_bh(&wreq->lock); + spin_unlock(&wreq->lock); stream->construct = subreq; } @@ -244,6 +250,8 @@ void netfs_reissue_write(struct netfs_io_stream *stream, iov_iter_advance(source, size); iov_iter_truncate(&subreq->io_iter, size); + subreq->retry_count++; + __clear_bit(NETFS_SREQ_MADE_PROGRESS, &subreq->flags); __set_bit(NETFS_SREQ_IN_PROGRESS, &subreq->flags); netfs_do_issue_write(stream, subreq); } @@ -266,9 +274,9 @@ void netfs_issue_write(struct netfs_io_request *wreq, * we can avoid overrunning the credits obtained (cifs) and try to parallelise * content-crypto preparation with network writes. */ -int netfs_advance_write(struct netfs_io_request *wreq, - struct netfs_io_stream *stream, - loff_t start, size_t len, bool to_eof) +size_t netfs_advance_write(struct netfs_io_request *wreq, + struct netfs_io_stream *stream, + loff_t start, size_t len, bool to_eof) { struct netfs_io_subrequest *subreq = stream->construct; size_t part; @@ -325,6 +333,9 @@ static int netfs_write_folio(struct netfs_io_request *wreq, _enter(""); + if (rolling_buffer_make_space(&wreq->buffer) < 0) + return -ENOMEM; + /* netfs_perform_write() may shift i_size around the page or from out * of the page to beyond it, but cannot move i_size into or through the * page since we have it locked. @@ -429,7 +440,7 @@ static int netfs_write_folio(struct netfs_io_request *wreq, } /* Attach the folio to the rolling buffer. */ - netfs_buffer_append_folio(wreq, folio, false); + rolling_buffer_append(&wreq->buffer, folio, 0); /* Move the submission point forward to allow for write-streaming data * not starting at the front of the page. We don't do write-streaming @@ -442,7 +453,8 @@ static int netfs_write_folio(struct netfs_io_request *wreq, stream = &wreq->io_streams[s]; stream->submit_off = foff; stream->submit_len = flen; - if ((stream->source == NETFS_WRITE_TO_CACHE && streamw) || + if (!stream->avail || + (stream->source == NETFS_WRITE_TO_CACHE && streamw) || (stream->source == NETFS_UPLOAD_TO_SERVER && fgroup == NETFS_FOLIO_COPY_TO_CACHE)) { stream->submit_off = UINT_MAX; @@ -476,7 +488,7 @@ static int netfs_write_folio(struct netfs_io_request *wreq, /* Advance the iterator(s). */ if (stream->submit_off > iter_off) { - iov_iter_advance(&wreq->io_iter, stream->submit_off - iter_off); + rolling_buffer_advance(&wreq->buffer, stream->submit_off - iter_off); iter_off = stream->submit_off; } @@ -494,7 +506,7 @@ static int netfs_write_folio(struct netfs_io_request *wreq, } if (fsize > iter_off) - iov_iter_advance(&wreq->io_iter, fsize - iter_off); + rolling_buffer_advance(&wreq->buffer, fsize - iter_off); atomic64_set(&wreq->issued_to, fpos + fsize); if (!debug) @@ -633,7 +645,7 @@ int netfs_advance_writethrough(struct netfs_io_request *wreq, struct writeback_c struct folio **writethrough_cache) { _enter("R=%x ic=%zu ws=%u cp=%zu tp=%u", - wreq->debug_id, wreq->iter.count, wreq->wsize, copied, to_page_end); + wreq->debug_id, wreq->buffer.iter.count, wreq->wsize, copied, to_page_end); if (!*writethrough_cache) { if (folio_test_dirty(folio)) @@ -708,10 +720,10 @@ int netfs_unbuffered_write(struct netfs_io_request *wreq, bool may_wait, size_t part = netfs_advance_write(wreq, upload, start, len, false); start += part; len -= part; - iov_iter_advance(&wreq->io_iter, part); + rolling_buffer_advance(&wreq->buffer, part); if (test_bit(NETFS_RREQ_PAUSE, &wreq->flags)) { trace_netfs_rreq(wreq, netfs_rreq_trace_wait_pause); - wait_on_bit(&wreq->flags, NETFS_RREQ_PAUSE, TASK_UNINTERRUPTIBLE); + wait_event(wreq->waitq, !test_bit(NETFS_RREQ_PAUSE, &wreq->flags)); } if (test_bit(NETFS_RREQ_FAILED, &wreq->flags)) break; @@ -721,3 +733,194 @@ int netfs_unbuffered_write(struct netfs_io_request *wreq, bool may_wait, size_t _leave(" = %d", error); return error; } + +/* + * Write some of a pending folio data back to the server and/or the cache. + */ +static int netfs_write_folio_single(struct netfs_io_request *wreq, + struct folio *folio) +{ + struct netfs_io_stream *upload = &wreq->io_streams[0]; + struct netfs_io_stream *cache = &wreq->io_streams[1]; + struct netfs_io_stream *stream; + size_t iter_off = 0; + size_t fsize = folio_size(folio), flen; + loff_t fpos = folio_pos(folio); + bool to_eof = false; + bool no_debug = false; + + _enter(""); + + flen = folio_size(folio); + if (flen > wreq->i_size - fpos) { + flen = wreq->i_size - fpos; + folio_zero_segment(folio, flen, fsize); + to_eof = true; + } else if (flen == wreq->i_size - fpos) { + to_eof = true; + } + + _debug("folio %zx/%zx", flen, fsize); + + if (!upload->avail && !cache->avail) { + trace_netfs_folio(folio, netfs_folio_trace_cancel_store); + return 0; + } + + if (!upload->construct) + trace_netfs_folio(folio, netfs_folio_trace_store); + else + trace_netfs_folio(folio, netfs_folio_trace_store_plus); + + /* Attach the folio to the rolling buffer. */ + folio_get(folio); + rolling_buffer_append(&wreq->buffer, folio, NETFS_ROLLBUF_PUT_MARK); + + /* Move the submission point forward to allow for write-streaming data + * not starting at the front of the page. We don't do write-streaming + * with the cache as the cache requires DIO alignment. + * + * Also skip uploading for data that's been read and just needs copying + * to the cache. + */ + for (int s = 0; s < NR_IO_STREAMS; s++) { + stream = &wreq->io_streams[s]; + stream->submit_off = 0; + stream->submit_len = flen; + if (!stream->avail) { + stream->submit_off = UINT_MAX; + stream->submit_len = 0; + } + } + + /* Attach the folio to one or more subrequests. For a big folio, we + * could end up with thousands of subrequests if the wsize is small - + * but we might need to wait during the creation of subrequests for + * network resources (eg. SMB credits). + */ + for (;;) { + ssize_t part; + size_t lowest_off = ULONG_MAX; + int choose_s = -1; + + /* Always add to the lowest-submitted stream first. */ + for (int s = 0; s < NR_IO_STREAMS; s++) { + stream = &wreq->io_streams[s]; + if (stream->submit_len > 0 && + stream->submit_off < lowest_off) { + lowest_off = stream->submit_off; + choose_s = s; + } + } + + if (choose_s < 0) + break; + stream = &wreq->io_streams[choose_s]; + + /* Advance the iterator(s). */ + if (stream->submit_off > iter_off) { + rolling_buffer_advance(&wreq->buffer, stream->submit_off - iter_off); + iter_off = stream->submit_off; + } + + atomic64_set(&wreq->issued_to, fpos + stream->submit_off); + stream->submit_extendable_to = fsize - stream->submit_off; + part = netfs_advance_write(wreq, stream, fpos + stream->submit_off, + stream->submit_len, to_eof); + stream->submit_off += part; + if (part > stream->submit_len) + stream->submit_len = 0; + else + stream->submit_len -= part; + if (part > 0) + no_debug = true; + } + + wreq->buffer.iter.iov_offset = 0; + if (fsize > iter_off) + rolling_buffer_advance(&wreq->buffer, fsize - iter_off); + atomic64_set(&wreq->issued_to, fpos + fsize); + + if (!no_debug) + kdebug("R=%x: No submit", wreq->debug_id); + _leave(" = 0"); + return 0; +} + +/** + * netfs_writeback_single - Write back a monolithic payload + * @mapping: The mapping to write from + * @wbc: Hints from the VM + * @iter: Data to write, must be ITER_FOLIOQ. + * + * Write a monolithic, non-pagecache object back to the server and/or + * the cache. + */ +int netfs_writeback_single(struct address_space *mapping, + struct writeback_control *wbc, + struct iov_iter *iter) +{ + struct netfs_io_request *wreq; + struct netfs_inode *ictx = netfs_inode(mapping->host); + struct folio_queue *fq; + size_t size = iov_iter_count(iter); + int ret; + + if (WARN_ON_ONCE(!iov_iter_is_folioq(iter))) + return -EIO; + + if (!mutex_trylock(&ictx->wb_lock)) { + if (wbc->sync_mode == WB_SYNC_NONE) { + netfs_stat(&netfs_n_wb_lock_skip); + return 0; + } + netfs_stat(&netfs_n_wb_lock_wait); + mutex_lock(&ictx->wb_lock); + } + + wreq = netfs_create_write_req(mapping, NULL, 0, NETFS_WRITEBACK_SINGLE); + if (IS_ERR(wreq)) { + ret = PTR_ERR(wreq); + goto couldnt_start; + } + + trace_netfs_write(wreq, netfs_write_trace_writeback); + netfs_stat(&netfs_n_wh_writepages); + + if (__test_and_set_bit(NETFS_RREQ_UPLOAD_TO_SERVER, &wreq->flags)) + wreq->netfs_ops->begin_writeback(wreq); + + for (fq = (struct folio_queue *)iter->folioq; fq; fq = fq->next) { + for (int slot = 0; slot < folioq_count(fq); slot++) { + struct folio *folio = folioq_folio(fq, slot); + size_t part = umin(folioq_folio_size(fq, slot), size); + + _debug("wbiter %lx %llx", folio->index, atomic64_read(&wreq->issued_to)); + + ret = netfs_write_folio_single(wreq, folio); + if (ret < 0) + goto stop; + size -= part; + if (size <= 0) + goto stop; + } + } + +stop: + for (int s = 0; s < NR_IO_STREAMS; s++) + netfs_issue_write(wreq, &wreq->io_streams[s]); + smp_wmb(); /* Write lists before ALL_QUEUED. */ + set_bit(NETFS_RREQ_ALL_QUEUED, &wreq->flags); + + mutex_unlock(&ictx->wb_lock); + + netfs_put_request(wreq, false, netfs_rreq_trace_put_return); + _leave(" = %d", ret); + return ret; + +couldnt_start: + mutex_unlock(&ictx->wb_lock); + _leave(" = %d", ret); + return ret; +} +EXPORT_SYMBOL(netfs_writeback_single); diff --git a/fs/netfs/write_retry.c b/fs/netfs/write_retry.c new file mode 100644 index 000000000000..c841a851dd73 --- /dev/null +++ b/fs/netfs/write_retry.c @@ -0,0 +1,232 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* Network filesystem write retrying. + * + * Copyright (C) 2024 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + */ + +#include <linux/fs.h> +#include <linux/mm.h> +#include <linux/pagemap.h> +#include <linux/slab.h> +#include "internal.h" + +/* + * Perform retries on the streams that need it. + */ +static void netfs_retry_write_stream(struct netfs_io_request *wreq, + struct netfs_io_stream *stream) +{ + struct list_head *next; + + _enter("R=%x[%x:]", wreq->debug_id, stream->stream_nr); + + if (list_empty(&stream->subrequests)) + return; + + if (stream->source == NETFS_UPLOAD_TO_SERVER && + wreq->netfs_ops->retry_request) + wreq->netfs_ops->retry_request(wreq, stream); + + if (unlikely(stream->failed)) + return; + + /* If there's no renegotiation to do, just resend each failed subreq. */ + if (!stream->prepare_write) { + struct netfs_io_subrequest *subreq; + + list_for_each_entry(subreq, &stream->subrequests, rreq_link) { + if (test_bit(NETFS_SREQ_FAILED, &subreq->flags)) + break; + if (__test_and_clear_bit(NETFS_SREQ_NEED_RETRY, &subreq->flags)) { + struct iov_iter source = subreq->io_iter; + + iov_iter_revert(&source, subreq->len - source.count); + netfs_get_subrequest(subreq, netfs_sreq_trace_get_resubmit); + netfs_reissue_write(stream, subreq, &source); + } + } + return; + } + + next = stream->subrequests.next; + + do { + struct netfs_io_subrequest *subreq = NULL, *from, *to, *tmp; + struct iov_iter source; + unsigned long long start, len; + size_t part; + bool boundary = false; + + /* Go through the stream and find the next span of contiguous + * data that we then rejig (cifs, for example, needs the wsize + * renegotiating) and reissue. + */ + from = list_entry(next, struct netfs_io_subrequest, rreq_link); + to = from; + start = from->start + from->transferred; + len = from->len - from->transferred; + + if (test_bit(NETFS_SREQ_FAILED, &from->flags) || + !test_bit(NETFS_SREQ_NEED_RETRY, &from->flags)) + return; + + list_for_each_continue(next, &stream->subrequests) { + subreq = list_entry(next, struct netfs_io_subrequest, rreq_link); + if (subreq->start + subreq->transferred != start + len || + test_bit(NETFS_SREQ_BOUNDARY, &subreq->flags) || + !test_bit(NETFS_SREQ_NEED_RETRY, &subreq->flags)) + break; + to = subreq; + len += to->len; + } + + /* Determine the set of buffers we're going to use. Each + * subreq gets a subset of a single overall contiguous buffer. + */ + netfs_reset_iter(from); + source = from->io_iter; + source.count = len; + + /* Work through the sublist. */ + subreq = from; + list_for_each_entry_from(subreq, &stream->subrequests, rreq_link) { + if (!len) + break; + + subreq->start = start; + subreq->len = len; + __clear_bit(NETFS_SREQ_NEED_RETRY, &subreq->flags); + subreq->retry_count++; + trace_netfs_sreq(subreq, netfs_sreq_trace_retry); + + /* Renegotiate max_len (wsize) */ + stream->sreq_max_len = len; + stream->prepare_write(subreq); + + part = umin(len, stream->sreq_max_len); + if (unlikely(stream->sreq_max_segs)) + part = netfs_limit_iter(&source, 0, part, stream->sreq_max_segs); + subreq->len = part; + subreq->transferred = 0; + len -= part; + start += part; + if (len && subreq == to && + __test_and_clear_bit(NETFS_SREQ_BOUNDARY, &to->flags)) + boundary = true; + + netfs_get_subrequest(subreq, netfs_sreq_trace_get_resubmit); + netfs_reissue_write(stream, subreq, &source); + if (subreq == to) + break; + } + + /* If we managed to use fewer subreqs, we can discard the + * excess; if we used the same number, then we're done. + */ + if (!len) { + if (subreq == to) + continue; + list_for_each_entry_safe_from(subreq, tmp, + &stream->subrequests, rreq_link) { + trace_netfs_sreq(subreq, netfs_sreq_trace_discard); + list_del(&subreq->rreq_link); + netfs_put_subrequest(subreq, false, netfs_sreq_trace_put_done); + if (subreq == to) + break; + } + continue; + } + + /* We ran out of subrequests, so we need to allocate some more + * and insert them after. + */ + do { + subreq = netfs_alloc_subrequest(wreq); + subreq->source = to->source; + subreq->start = start; + subreq->debug_index = atomic_inc_return(&wreq->subreq_counter); + subreq->stream_nr = to->stream_nr; + subreq->retry_count = 1; + + trace_netfs_sreq_ref(wreq->debug_id, subreq->debug_index, + refcount_read(&subreq->ref), + netfs_sreq_trace_new); + netfs_get_subrequest(subreq, netfs_sreq_trace_get_resubmit); + + list_add(&subreq->rreq_link, &to->rreq_link); + to = list_next_entry(to, rreq_link); + trace_netfs_sreq(subreq, netfs_sreq_trace_retry); + + stream->sreq_max_len = len; + stream->sreq_max_segs = INT_MAX; + switch (stream->source) { + case NETFS_UPLOAD_TO_SERVER: + netfs_stat(&netfs_n_wh_upload); + stream->sreq_max_len = umin(len, wreq->wsize); + break; + case NETFS_WRITE_TO_CACHE: + netfs_stat(&netfs_n_wh_write); + break; + default: + WARN_ON_ONCE(1); + } + + stream->prepare_write(subreq); + + part = umin(len, stream->sreq_max_len); + subreq->len = subreq->transferred + part; + len -= part; + start += part; + if (!len && boundary) { + __set_bit(NETFS_SREQ_BOUNDARY, &to->flags); + boundary = false; + } + + netfs_reissue_write(stream, subreq, &source); + if (!len) + break; + + } while (len); + + } while (!list_is_head(next, &stream->subrequests)); +} + +/* + * Perform retries on the streams that need it. If we're doing content + * encryption and the server copy changed due to a third-party write, we may + * need to do an RMW cycle and also rewrite the data to the cache. + */ +void netfs_retry_writes(struct netfs_io_request *wreq) +{ + struct netfs_io_subrequest *subreq; + struct netfs_io_stream *stream; + int s; + + /* Wait for all outstanding I/O to quiesce before performing retries as + * we may need to renegotiate the I/O sizes. + */ + for (s = 0; s < NR_IO_STREAMS; s++) { + stream = &wreq->io_streams[s]; + if (!stream->active) + continue; + + list_for_each_entry(subreq, &stream->subrequests, rreq_link) { + wait_on_bit(&subreq->flags, NETFS_SREQ_IN_PROGRESS, + TASK_UNINTERRUPTIBLE); + } + } + + // TODO: Enc: Fetch changed partial pages + // TODO: Enc: Reencrypt content if needed. + // TODO: Enc: Wind back transferred point. + // TODO: Enc: Mark cache pages for retry. + + for (s = 0; s < NR_IO_STREAMS; s++) { + stream = &wreq->io_streams[s]; + if (stream->need_retry) { + stream->need_retry = false; + netfs_retry_write_stream(wreq, stream); + } + } +} diff --git a/fs/nfs/fscache.c b/fs/nfs/fscache.c index 810269ee0a50..e278a1ad1ca3 100644 --- a/fs/nfs/fscache.c +++ b/fs/nfs/fscache.c @@ -263,6 +263,12 @@ int nfs_netfs_readahead(struct readahead_control *ractl) static atomic_t nfs_netfs_debug_id; static int nfs_netfs_init_request(struct netfs_io_request *rreq, struct file *file) { + if (!file) { + if (WARN_ON_ONCE(rreq->origin != NETFS_PGPRIV2_COPY_TO_CACHE)) + return -EIO; + return 0; + } + rreq->netfs_priv = get_nfs_open_context(nfs_file_open_context(file)); rreq->debug_id = atomic_inc_return(&nfs_netfs_debug_id); /* [DEPRECATED] Use PG_private_2 to mark folio being written to the cache. */ @@ -274,7 +280,8 @@ static int nfs_netfs_init_request(struct netfs_io_request *rreq, struct file *fi static void nfs_netfs_free_request(struct netfs_io_request *rreq) { - put_nfs_open_context(rreq->netfs_priv); + if (rreq->netfs_priv) + put_nfs_open_context(rreq->netfs_priv); } static struct nfs_netfs_io_data *nfs_netfs_alloc(struct netfs_io_subrequest *sreq) @@ -307,8 +314,10 @@ static void nfs_netfs_issue_read(struct netfs_io_subrequest *sreq) &nfs_async_read_completion_ops); netfs = nfs_netfs_alloc(sreq); - if (!netfs) - return netfs_read_subreq_terminated(sreq, -ENOMEM, false); + if (!netfs) { + sreq->error = -ENOMEM; + return netfs_read_subreq_terminated(sreq); + } pgio.pg_netfs = netfs; /* used in completion */ diff --git a/fs/nfs/fscache.h b/fs/nfs/fscache.h index 772d485e96d3..9d86868f4998 100644 --- a/fs/nfs/fscache.h +++ b/fs/nfs/fscache.h @@ -74,7 +74,8 @@ static inline void nfs_netfs_put(struct nfs_netfs_io_data *netfs) */ netfs->sreq->transferred = min_t(s64, netfs->sreq->len, atomic64_read(&netfs->transferred)); - netfs_read_subreq_terminated(netfs->sreq, netfs->error, false); + netfs->sreq->error = netfs->error; + netfs_read_subreq_terminated(netfs->sreq); kfree(netfs); } static inline void nfs_netfs_inode_init(struct nfs_inode *nfsi) diff --git a/fs/nfsd/auth.c b/fs/nfsd/auth.c index 93e33d1ee891..4dc327e02456 100644 --- a/fs/nfsd/auth.c +++ b/fs/nfsd/auth.c @@ -27,7 +27,7 @@ int nfsd_setuser(struct svc_cred *cred, struct svc_export *exp) int flags = nfsexp_flags(cred, exp); /* discard any old override before preparing the new set */ - revert_creds(get_cred(current_real_cred())); + put_cred(revert_creds(get_cred(current_real_cred()))); new = prepare_creds(); if (!new) return -ENOMEM; @@ -80,7 +80,6 @@ int nfsd_setuser(struct svc_cred *cred, struct svc_export *exp) new->cap_effective = cap_raise_nfsd_set(new->cap_effective, new->cap_permitted); put_cred(override_creds(new)); - put_cred(new); return 0; oom: diff --git a/fs/nfsd/export.c b/fs/nfsd/export.c index eacafe46e3b6..aa4712362b3b 100644 --- a/fs/nfsd/export.c +++ b/fs/nfsd/export.c @@ -40,24 +40,15 @@ #define EXPKEY_HASHMAX (1 << EXPKEY_HASHBITS) #define EXPKEY_HASHMASK (EXPKEY_HASHMAX -1) -static void expkey_put_work(struct work_struct *work) +static void expkey_put(struct kref *ref) { - struct svc_expkey *key = - container_of(to_rcu_work(work), struct svc_expkey, ek_rcu_work); + struct svc_expkey *key = container_of(ref, struct svc_expkey, h.ref); if (test_bit(CACHE_VALID, &key->h.flags) && !test_bit(CACHE_NEGATIVE, &key->h.flags)) path_put(&key->ek_path); auth_domain_put(key->ek_client); - kfree(key); -} - -static void expkey_put(struct kref *ref) -{ - struct svc_expkey *key = container_of(ref, struct svc_expkey, h.ref); - - INIT_RCU_WORK(&key->ek_rcu_work, expkey_put_work); - queue_rcu_work(system_wq, &key->ek_rcu_work); + kfree_rcu(key, ek_rcu); } static int expkey_upcall(struct cache_detail *cd, struct cache_head *h) @@ -364,26 +355,16 @@ static void export_stats_destroy(struct export_stats *stats) EXP_STATS_COUNTERS_NUM); } -static void svc_export_put_work(struct work_struct *work) +static void svc_export_put(struct kref *ref) { - struct svc_export *exp = - container_of(to_rcu_work(work), struct svc_export, ex_rcu_work); - + struct svc_export *exp = container_of(ref, struct svc_export, h.ref); path_put(&exp->ex_path); auth_domain_put(exp->ex_client); nfsd4_fslocs_free(&exp->ex_fslocs); export_stats_destroy(exp->ex_stats); kfree(exp->ex_stats); kfree(exp->ex_uuid); - kfree(exp); -} - -static void svc_export_put(struct kref *ref) -{ - struct svc_export *exp = container_of(ref, struct svc_export, h.ref); - - INIT_RCU_WORK(&exp->ex_rcu_work, svc_export_put_work); - queue_rcu_work(system_wq, &exp->ex_rcu_work); + kfree_rcu(exp, ex_rcu); } static int svc_export_upcall(struct cache_detail *cd, struct cache_head *h) diff --git a/fs/nfsd/export.h b/fs/nfsd/export.h index 6f2fbaae01fa..4d92b99c1ffd 100644 --- a/fs/nfsd/export.h +++ b/fs/nfsd/export.h @@ -75,7 +75,7 @@ struct svc_export { u32 ex_layout_types; struct nfsd4_deviceid_map *ex_devid_map; struct cache_detail *cd; - struct rcu_work ex_rcu_work; + struct rcu_head ex_rcu; unsigned long ex_xprtsec_modes; struct export_stats *ex_stats; }; @@ -92,7 +92,7 @@ struct svc_expkey { u32 ek_fsid[6]; struct path ek_path; - struct rcu_work ek_rcu_work; + struct rcu_head ek_rcu; }; #define EX_ISSYNC(exp) (!((exp)->ex_flags & NFSEXP_ASYNC)) diff --git a/fs/nfsd/filecache.c b/fs/nfsd/filecache.c index a1cdba42c4fa..dc5c9d8e8202 100644 --- a/fs/nfsd/filecache.c +++ b/fs/nfsd/filecache.c @@ -1248,7 +1248,7 @@ nfsd_file_acquire_local(struct net *net, struct svc_cred *cred, beres = nfsd_file_do_acquire(NULL, net, cred, client, fhp, may_flags, NULL, pnf, true); - revert_creds(save_cred); + put_cred(revert_creds(save_cred)); return beres; } diff --git a/fs/nfsd/nfs4callback.c b/fs/nfsd/nfs4callback.c index 3877b53e429f..c083e539e898 100644 --- a/fs/nfsd/nfs4callback.c +++ b/fs/nfsd/nfs4callback.c @@ -1100,7 +1100,7 @@ static int setup_callback_client(struct nfs4_client *clp, struct nfs4_cb_conn *c args.authflavor = clp->cl_cred.cr_flavor; clp->cl_cb_ident = conn->cb_ident; } else { - if (!conn->cb_xprt) + if (!conn->cb_xprt || !ses) return -EINVAL; clp->cl_cb_session = ses; args.bc_xprt = conn->cb_xprt; @@ -1522,8 +1522,6 @@ static void nfsd4_process_cb_update(struct nfsd4_callback *cb) ses = c->cn_session; } spin_unlock(&clp->cl_lock); - if (!c) - return; err = setup_callback_client(clp, &conn, ses); if (err) { diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c index f8a10f90bc7a..ad44ad49274f 100644 --- a/fs/nfsd/nfs4proc.c +++ b/fs/nfsd/nfs4proc.c @@ -1347,7 +1347,6 @@ static void nfs4_put_copy(struct nfsd4_copy *copy) { if (!refcount_dec_and_test(©->refcount)) return; - atomic_dec(©->cp_nn->pending_async_copies); kfree(copy->cp_src); kfree(copy); } @@ -1870,6 +1869,7 @@ do_callback: set_bit(NFSD4_COPY_F_COMPLETED, ©->cp_flags); trace_nfsd_copy_async_done(copy); nfsd4_send_cb_offload(copy); + atomic_dec(©->cp_nn->pending_async_copies); return 0; } @@ -1927,19 +1927,19 @@ nfsd4_copy(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, /* Arbitrary cap on number of pending async copy operations */ if (atomic_inc_return(&nn->pending_async_copies) > (int)rqstp->rq_pool->sp_nrthreads) - goto out_err; + goto out_dec_async_copy_err; async_copy->cp_src = kmalloc(sizeof(*async_copy->cp_src), GFP_KERNEL); if (!async_copy->cp_src) - goto out_err; + goto out_dec_async_copy_err; if (!nfs4_init_copy_state(nn, copy)) - goto out_err; + goto out_dec_async_copy_err; memcpy(&result->cb_stateid, ©->cp_stateid.cs_stid, sizeof(result->cb_stateid)); dup_copy_fields(copy, async_copy); async_copy->copy_task = kthread_create(nfsd4_do_async_copy, async_copy, "%s", "copy thread"); if (IS_ERR(async_copy->copy_task)) - goto out_err; + goto out_dec_async_copy_err; spin_lock(&async_copy->cp_clp->async_lock); list_add(&async_copy->copies, &async_copy->cp_clp->async_copies); @@ -1954,6 +1954,9 @@ out: trace_nfsd_copy_done(copy, status); release_copy_files(copy); return status; +out_dec_async_copy_err: + if (async_copy) + atomic_dec(&nn->pending_async_copies); out_err: if (nfsd4_ssc_is_inter(copy)) { /* diff --git a/fs/nfsd/nfs4recover.c b/fs/nfsd/nfs4recover.c index 4a765555bf84..7f2ceeb118a4 100644 --- a/fs/nfsd/nfs4recover.c +++ b/fs/nfsd/nfs4recover.c @@ -82,14 +82,13 @@ nfs4_save_creds(const struct cred **original_creds) new->fsuid = GLOBAL_ROOT_UID; new->fsgid = GLOBAL_ROOT_GID; *original_creds = override_creds(new); - put_cred(new); return 0; } static void nfs4_reset_creds(const struct cred *original) { - revert_creds(original); + put_cred(revert_creds(original)); } static void diff --git a/fs/nfsd/nfsfh.c b/fs/nfsd/nfsfh.c index 6a831cb242df..98d6459724a7 100644 --- a/fs/nfsd/nfsfh.c +++ b/fs/nfsd/nfsfh.c @@ -222,7 +222,6 @@ static __be32 nfsd_set_fh_dentry(struct svc_rqst *rqstp, struct net *net, cap_raise_nfsd_set(new->cap_effective, new->cap_permitted); put_cred(override_creds(new)); - put_cred(new); } else { error = nfsd_setuser_and_check_port(rqstp, cred, exp); if (error) diff --git a/fs/notify/fdinfo.c b/fs/notify/fdinfo.c index dec553034027..e933f9c65d90 100644 --- a/fs/notify/fdinfo.c +++ b/fs/notify/fdinfo.c @@ -47,10 +47,8 @@ static void show_mark_fhandle(struct seq_file *m, struct inode *inode) size = f->handle_bytes >> 2; ret = exportfs_encode_fid(inode, (struct fid *)f->f_handle, &size); - if ((ret == FILEID_INVALID) || (ret < 0)) { - WARN_ONCE(1, "Can't encode file handler for inotify: %d\n", ret); + if ((ret == FILEID_INVALID) || (ret < 0)) return; - } f->handle_type = ret; f->handle_bytes = size * sizeof(u32); diff --git a/fs/nsfs.c b/fs/nsfs.c index c675fc40ce2d..663f8656158d 100644 --- a/fs/nsfs.c +++ b/fs/nsfs.c @@ -274,10 +274,7 @@ static long ns_ioctl(struct file *filp, unsigned int ioctl, if (usize < MNT_NS_INFO_SIZE_VER0) return -EINVAL; - if (previous) - mnt_ns = lookup_prev_mnt_ns(to_mnt_ns(ns)); - else - mnt_ns = lookup_next_mnt_ns(to_mnt_ns(ns)); + mnt_ns = get_sequential_mnt_ns(to_mnt_ns(ns), previous); if (IS_ERR(mnt_ns)) return PTR_ERR(mnt_ns); diff --git a/fs/ocfs2/dir.c b/fs/ocfs2/dir.c index 213206ebdd58..7799f4d16ce9 100644 --- a/fs/ocfs2/dir.c +++ b/fs/ocfs2/dir.c @@ -1065,26 +1065,39 @@ int ocfs2_find_entry(const char *name, int namelen, { struct buffer_head *bh; struct ocfs2_dir_entry *res_dir = NULL; + int ret = 0; if (ocfs2_dir_indexed(dir)) return ocfs2_find_entry_dx(name, namelen, dir, lookup); + if (unlikely(i_size_read(dir) <= 0)) { + ret = -EFSCORRUPTED; + mlog_errno(ret); + goto out; + } /* * The unindexed dir code only uses part of the lookup * structure, so there's no reason to push it down further * than this. */ - if (OCFS2_I(dir)->ip_dyn_features & OCFS2_INLINE_DATA_FL) + if (OCFS2_I(dir)->ip_dyn_features & OCFS2_INLINE_DATA_FL) { + if (unlikely(i_size_read(dir) > dir->i_sb->s_blocksize)) { + ret = -EFSCORRUPTED; + mlog_errno(ret); + goto out; + } bh = ocfs2_find_entry_id(name, namelen, dir, &res_dir); - else + } else { bh = ocfs2_find_entry_el(name, namelen, dir, &res_dir); + } if (bh == NULL) return -ENOENT; lookup->dl_leaf_bh = bh; lookup->dl_entry = res_dir; - return 0; +out: + return ret; } /* @@ -2010,6 +2023,7 @@ int ocfs2_lookup_ino_from_name(struct inode *dir, const char *name, * * Return 0 if the name does not exist * Return -EEXIST if the directory contains the name + * Return -EFSCORRUPTED if found corruption * * Callers should have i_rwsem + a cluster lock on dir */ @@ -2023,9 +2037,12 @@ int ocfs2_check_dir_for_entry(struct inode *dir, trace_ocfs2_check_dir_for_entry( (unsigned long long)OCFS2_I(dir)->ip_blkno, namelen, name); - if (ocfs2_find_entry(name, namelen, dir, &lookup) == 0) { + ret = ocfs2_find_entry(name, namelen, dir, &lookup); + if (ret == 0) { ret = -EEXIST; mlog_errno(ret); + } else if (ret == -ENOENT) { + ret = 0; } ocfs2_free_dir_lookup_result(&lookup); diff --git a/fs/ocfs2/quota_global.c b/fs/ocfs2/quota_global.c index 2b0daced98eb..3404e7a30c33 100644 --- a/fs/ocfs2/quota_global.c +++ b/fs/ocfs2/quota_global.c @@ -893,7 +893,7 @@ static int ocfs2_get_next_id(struct super_block *sb, struct kqid *qid) int status = 0; trace_ocfs2_get_next_id(from_kqid(&init_user_ns, *qid), type); - if (!sb_has_quota_loaded(sb, type)) { + if (!sb_has_quota_active(sb, type)) { status = -ESRCH; goto out; } diff --git a/fs/ocfs2/quota_local.c b/fs/ocfs2/quota_local.c index 73d3367c533b..2956d888c131 100644 --- a/fs/ocfs2/quota_local.c +++ b/fs/ocfs2/quota_local.c @@ -867,6 +867,7 @@ out: brelse(oinfo->dqi_libh); brelse(oinfo->dqi_lqi_bh); kfree(oinfo); + info->dqi_priv = NULL; return status; } diff --git a/fs/open.c b/fs/open.c index e6911101fe71..ffcfef67ac86 100644 --- a/fs/open.c +++ b/fs/open.c @@ -402,7 +402,6 @@ static bool access_need_override_creds(int flags) static const struct cred *access_override_creds(void) { - const struct cred *old_cred; struct cred *override_cred; override_cred = prepare_creds(); @@ -447,13 +446,7 @@ static const struct cred *access_override_creds(void) * freeing. */ override_cred->non_rcu = 1; - - old_cred = override_creds(override_cred); - - /* override_cred() gets its own ref */ - put_cred(override_cred); - - return old_cred; + return override_creds(override_cred); } static long do_faccessat(int dfd, const char __user *filename, int mode, int flags) @@ -523,7 +516,7 @@ out_path_release: } out: if (old_cred) - revert_creds(old_cred); + put_cred(revert_creds(old_cred)); return res; } diff --git a/fs/overlayfs/copy_up.c b/fs/overlayfs/copy_up.c index 3601ddfeddc2..0c28e5fa3407 100644 --- a/fs/overlayfs/copy_up.c +++ b/fs/overlayfs/copy_up.c @@ -415,13 +415,13 @@ int ovl_set_attr(struct ovl_fs *ofs, struct dentry *upperdentry, return err; } -struct ovl_fh *ovl_encode_real_fh(struct ovl_fs *ofs, struct dentry *real, +struct ovl_fh *ovl_encode_real_fh(struct ovl_fs *ofs, struct inode *realinode, bool is_upper) { struct ovl_fh *fh; int fh_type, dwords; int buflen = MAX_HANDLE_SZ; - uuid_t *uuid = &real->d_sb->s_uuid; + uuid_t *uuid = &realinode->i_sb->s_uuid; int err; /* Make sure the real fid stays 32bit aligned */ @@ -438,13 +438,13 @@ struct ovl_fh *ovl_encode_real_fh(struct ovl_fs *ofs, struct dentry *real, * the price or reconnecting the dentry. */ dwords = buflen >> 2; - fh_type = exportfs_encode_fh(real, (void *)fh->fb.fid, &dwords, 0); + fh_type = exportfs_encode_inode_fh(realinode, (void *)fh->fb.fid, + &dwords, NULL, 0); buflen = (dwords << 2); err = -EIO; - if (WARN_ON(fh_type < 0) || - WARN_ON(buflen > MAX_HANDLE_SZ) || - WARN_ON(fh_type == FILEID_INVALID)) + if (fh_type < 0 || fh_type == FILEID_INVALID || + WARN_ON(buflen > MAX_HANDLE_SZ)) goto out_err; fh->fb.version = OVL_FH_VERSION; @@ -480,7 +480,7 @@ struct ovl_fh *ovl_get_origin_fh(struct ovl_fs *ofs, struct dentry *origin) if (!ovl_can_decode_fh(origin->d_sb)) return NULL; - return ovl_encode_real_fh(ofs, origin, false); + return ovl_encode_real_fh(ofs, d_inode(origin), false); } int ovl_set_origin_fh(struct ovl_fs *ofs, const struct ovl_fh *fh, @@ -505,7 +505,7 @@ static int ovl_set_upper_fh(struct ovl_fs *ofs, struct dentry *upper, const struct ovl_fh *fh; int err; - fh = ovl_encode_real_fh(ofs, upper, true); + fh = ovl_encode_real_fh(ofs, d_inode(upper), true); if (IS_ERR(fh)) return PTR_ERR(fh); diff --git a/fs/overlayfs/dir.c b/fs/overlayfs/dir.c index 08e683917d12..c9993ff66fc2 100644 --- a/fs/overlayfs/dir.c +++ b/fs/overlayfs/dir.c @@ -575,12 +575,12 @@ static const struct cred *ovl_setup_cred_for_create(struct dentry *dentry, } /* - * Caller is going to match this with revert_creds_light() and drop + * Caller is going to match this with revert_creds() and drop * referenec on the returned creds. * We must be called with creator creds already, otherwise we risk * leaking creds. */ - old_cred = override_creds_light(override_cred); + old_cred = override_creds(override_cred); WARN_ON_ONCE(old_cred != ovl_creds(dentry->d_sb)); return override_cred; diff --git a/fs/overlayfs/export.c b/fs/overlayfs/export.c index 5868cb222955..444aeeccb6da 100644 --- a/fs/overlayfs/export.c +++ b/fs/overlayfs/export.c @@ -176,35 +176,37 @@ static int ovl_connect_layer(struct dentry *dentry) * * Return 0 for upper file handle, > 0 for lower file handle or < 0 on error. */ -static int ovl_check_encode_origin(struct dentry *dentry) +static int ovl_check_encode_origin(struct inode *inode) { - struct ovl_fs *ofs = OVL_FS(dentry->d_sb); + struct ovl_fs *ofs = OVL_FS(inode->i_sb); bool decodable = ofs->config.nfs_export; + struct dentry *dentry; + int err; /* No upper layer? */ if (!ovl_upper_mnt(ofs)) return 1; /* Lower file handle for non-upper non-decodable */ - if (!ovl_dentry_upper(dentry) && !decodable) + if (!ovl_inode_upper(inode) && !decodable) return 1; /* Upper file handle for pure upper */ - if (!ovl_dentry_lower(dentry)) + if (!ovl_inode_lower(inode)) return 0; /* * Root is never indexed, so if there's an upper layer, encode upper for * root. */ - if (dentry == dentry->d_sb->s_root) + if (inode == d_inode(inode->i_sb->s_root)) return 0; /* * Upper decodable file handle for non-indexed upper. */ - if (ovl_dentry_upper(dentry) && decodable && - !ovl_test_flag(OVL_INDEX, d_inode(dentry))) + if (ovl_inode_upper(inode) && decodable && + !ovl_test_flag(OVL_INDEX, inode)) return 0; /* @@ -213,14 +215,23 @@ static int ovl_check_encode_origin(struct dentry *dentry) * ovl_connect_layer() will try to make origin's layer "connected" by * copying up a "connectable" ancestor. */ - if (d_is_dir(dentry) && decodable) - return ovl_connect_layer(dentry); + if (!decodable || !S_ISDIR(inode->i_mode)) + return 1; + + dentry = d_find_any_alias(inode); + if (!dentry) + return -ENOENT; + + err = ovl_connect_layer(dentry); + dput(dentry); + if (err < 0) + return err; /* Lower file handle for indexed and non-upper dir/non-dir */ return 1; } -static int ovl_dentry_to_fid(struct ovl_fs *ofs, struct dentry *dentry, +static int ovl_dentry_to_fid(struct ovl_fs *ofs, struct inode *inode, u32 *fid, int buflen) { struct ovl_fh *fh = NULL; @@ -231,13 +242,13 @@ static int ovl_dentry_to_fid(struct ovl_fs *ofs, struct dentry *dentry, * Check if we should encode a lower or upper file handle and maybe * copy up an ancestor to make lower file handle connectable. */ - err = enc_lower = ovl_check_encode_origin(dentry); + err = enc_lower = ovl_check_encode_origin(inode); if (enc_lower < 0) goto fail; /* Encode an upper or lower file handle */ - fh = ovl_encode_real_fh(ofs, enc_lower ? ovl_dentry_lower(dentry) : - ovl_dentry_upper(dentry), !enc_lower); + fh = ovl_encode_real_fh(ofs, enc_lower ? ovl_inode_lower(inode) : + ovl_inode_upper(inode), !enc_lower); if (IS_ERR(fh)) return PTR_ERR(fh); @@ -251,8 +262,8 @@ out: return err; fail: - pr_warn_ratelimited("failed to encode file handle (%pd2, err=%i)\n", - dentry, err); + pr_warn_ratelimited("failed to encode file handle (ino=%lu, err=%i)\n", + inode->i_ino, err); goto out; } @@ -260,19 +271,13 @@ static int ovl_encode_fh(struct inode *inode, u32 *fid, int *max_len, struct inode *parent) { struct ovl_fs *ofs = OVL_FS(inode->i_sb); - struct dentry *dentry; int bytes, buflen = *max_len << 2; /* TODO: encode connectable file handles */ if (parent) return FILEID_INVALID; - dentry = d_find_any_alias(inode); - if (!dentry) - return FILEID_INVALID; - - bytes = ovl_dentry_to_fid(ofs, dentry, fid, buflen); - dput(dentry); + bytes = ovl_dentry_to_fid(ofs, inode, fid, buflen); if (bytes <= 0) return FILEID_INVALID; diff --git a/fs/overlayfs/namei.c b/fs/overlayfs/namei.c index 7e27b7d4adee..cea820cb3b55 100644 --- a/fs/overlayfs/namei.c +++ b/fs/overlayfs/namei.c @@ -542,7 +542,7 @@ int ovl_verify_origin_xattr(struct ovl_fs *ofs, struct dentry *dentry, struct ovl_fh *fh; int err; - fh = ovl_encode_real_fh(ofs, real, is_upper); + fh = ovl_encode_real_fh(ofs, d_inode(real), is_upper); err = PTR_ERR(fh); if (IS_ERR(fh)) { fh = NULL; @@ -738,7 +738,7 @@ int ovl_get_index_name(struct ovl_fs *ofs, struct dentry *origin, struct ovl_fh *fh; int err; - fh = ovl_encode_real_fh(ofs, origin, false); + fh = ovl_encode_real_fh(ofs, d_inode(origin), false); if (IS_ERR(fh)) return PTR_ERR(fh); diff --git a/fs/overlayfs/overlayfs.h b/fs/overlayfs/overlayfs.h index b361f35762be..0021e2025020 100644 --- a/fs/overlayfs/overlayfs.h +++ b/fs/overlayfs/overlayfs.h @@ -865,7 +865,7 @@ int ovl_copy_up_with_data(struct dentry *dentry); int ovl_maybe_copy_up(struct dentry *dentry, int flags); int ovl_copy_xattr(struct super_block *sb, const struct path *path, struct dentry *new); int ovl_set_attr(struct ovl_fs *ofs, struct dentry *upper, struct kstat *stat); -struct ovl_fh *ovl_encode_real_fh(struct ovl_fs *ofs, struct dentry *real, +struct ovl_fh *ovl_encode_real_fh(struct ovl_fs *ofs, struct inode *realinode, bool is_upper); struct ovl_fh *ovl_get_origin_fh(struct ovl_fs *ofs, struct dentry *origin); int ovl_set_origin_fh(struct ovl_fs *ofs, const struct ovl_fh *fh, diff --git a/fs/overlayfs/util.c b/fs/overlayfs/util.c index 9aa7493b1e10..0819c739cc2f 100644 --- a/fs/overlayfs/util.c +++ b/fs/overlayfs/util.c @@ -65,12 +65,12 @@ const struct cred *ovl_override_creds(struct super_block *sb) { struct ovl_fs *ofs = OVL_FS(sb); - return override_creds_light(ofs->creator_cred); + return override_creds(ofs->creator_cred); } void ovl_revert_creds(const struct cred *old_cred) { - revert_creds_light(old_cred); + revert_creds(old_cred); } /* diff --git a/fs/pidfs.c b/fs/pidfs.c index 618abb1fa1b8..049352f973de 100644 --- a/fs/pidfs.c +++ b/fs/pidfs.c @@ -1,5 +1,6 @@ // SPDX-License-Identifier: GPL-2.0 #include <linux/anon_inodes.h> +#include <linux/exportfs.h> #include <linux/file.h> #include <linux/fs.h> #include <linux/cgroup.h> @@ -23,6 +24,97 @@ #include "internal.h" #include "mount.h" +static struct rb_root pidfs_ino_tree = RB_ROOT; + +#if BITS_PER_LONG == 32 +static inline unsigned long pidfs_ino(u64 ino) +{ + return lower_32_bits(ino); +} + +/* On 32 bit the generation number are the upper 32 bits. */ +static inline u32 pidfs_gen(u64 ino) +{ + return upper_32_bits(ino); +} + +#else + +/* On 64 bit simply return ino. */ +static inline unsigned long pidfs_ino(u64 ino) +{ + return ino; +} + +/* On 64 bit the generation number is 0. */ +static inline u32 pidfs_gen(u64 ino) +{ + return 0; +} +#endif + +static int pidfs_ino_cmp(struct rb_node *a, const struct rb_node *b) +{ + struct pid *pid_a = rb_entry(a, struct pid, pidfs_node); + struct pid *pid_b = rb_entry(b, struct pid, pidfs_node); + u64 pid_ino_a = pid_a->ino; + u64 pid_ino_b = pid_b->ino; + + if (pid_ino_a < pid_ino_b) + return -1; + if (pid_ino_a > pid_ino_b) + return 1; + return 0; +} + +void pidfs_add_pid(struct pid *pid) +{ + static u64 pidfs_ino_nr = 2; + + /* + * On 64 bit nothing special happens. The 64bit number assigned + * to struct pid is the inode number. + * + * On 32 bit the 64 bit number assigned to struct pid is split + * into two 32 bit numbers. The lower 32 bits are used as the + * inode number and the upper 32 bits are used as the inode + * generation number. + * + * On 32 bit pidfs_ino() will return the lower 32 bit. When + * pidfs_ino() returns zero a wrap around happened. When a + * wraparound happens the 64 bit number will be incremented by 2 + * so inode numbering starts at 2 again. + * + * On 64 bit comparing two pidfds is as simple as comparing + * inode numbers. + * + * When a wraparound happens on 32 bit multiple pidfds with the + * same inode number are likely to exist (This isn't a problem + * since before pidfs pidfds used the anonymous inode meaning + * all pidfds had the same inode number.). Userspace can + * reconstruct the 64 bit identifier by retrieving both the + * inode number and the inode generation number to compare or + * use file handles. + */ + if (pidfs_ino(pidfs_ino_nr) == 0) + pidfs_ino_nr += 2; + + pid->ino = pidfs_ino_nr; + pid->stashed = NULL; + pidfs_ino_nr++; + + write_seqcount_begin(&pidmap_lock_seq); + rb_find_add_rcu(&pid->pidfs_node, &pidfs_ino_tree, pidfs_ino_cmp); + write_seqcount_end(&pidmap_lock_seq); +} + +void pidfs_remove_pid(struct pid *pid) +{ + write_seqcount_begin(&pidmap_lock_seq); + rb_erase(&pid->pidfs_node, &pidfs_ino_tree); + write_seqcount_end(&pidmap_lock_seq); +} + #ifdef CONFIG_PROC_FS /** * pidfd_show_fdinfo - print information about a pidfd @@ -190,6 +282,27 @@ static long pidfd_info(struct task_struct *task, unsigned int cmd, unsigned long return 0; } +static bool pidfs_ioctl_valid(unsigned int cmd) +{ + switch (cmd) { + case FS_IOC_GETVERSION: + case PIDFD_GET_CGROUP_NAMESPACE: + case PIDFD_GET_INFO: + case PIDFD_GET_IPC_NAMESPACE: + case PIDFD_GET_MNT_NAMESPACE: + case PIDFD_GET_NET_NAMESPACE: + case PIDFD_GET_PID_FOR_CHILDREN_NAMESPACE: + case PIDFD_GET_TIME_NAMESPACE: + case PIDFD_GET_TIME_FOR_CHILDREN_NAMESPACE: + case PIDFD_GET_UTS_NAMESPACE: + case PIDFD_GET_USER_NAMESPACE: + case PIDFD_GET_PID_NAMESPACE: + return true; + } + + return false; +} + static long pidfd_ioctl(struct file *file, unsigned int cmd, unsigned long arg) { struct task_struct *task __free(put_task) = NULL; @@ -198,6 +311,17 @@ static long pidfd_ioctl(struct file *file, unsigned int cmd, unsigned long arg) struct ns_common *ns_common = NULL; struct pid_namespace *pid_ns; + if (!pidfs_ioctl_valid(cmd)) + return -ENOIOCTLCMD; + + if (cmd == FS_IOC_GETVERSION) { + if (!arg) + return -EINVAL; + + __u32 __user *argp = (__u32 __user *)arg; + return put_user(file_inode(file)->i_generation, argp); + } + task = get_pid_task(pid, PIDTYPE_PID); if (!task) return -ESRCH; @@ -318,40 +442,6 @@ struct pid *pidfd_pid(const struct file *file) static struct vfsmount *pidfs_mnt __ro_after_init; -#if BITS_PER_LONG == 32 -/* - * Provide a fallback mechanism for 32-bit systems so processes remain - * reliably comparable by inode number even on those systems. - */ -static DEFINE_IDA(pidfd_inum_ida); - -static int pidfs_inum(struct pid *pid, unsigned long *ino) -{ - int ret; - - ret = ida_alloc_range(&pidfd_inum_ida, RESERVED_PIDS + 1, - UINT_MAX, GFP_ATOMIC); - if (ret < 0) - return -ENOSPC; - - *ino = ret; - return 0; -} - -static inline void pidfs_free_inum(unsigned long ino) -{ - if (ino > 0) - ida_free(&pidfd_inum_ida, ino); -} -#else -static inline int pidfs_inum(struct pid *pid, unsigned long *ino) -{ - *ino = pid->ino; - return 0; -} -#define pidfs_free_inum(ino) ((void)(ino)) -#endif - /* * The vfs falls back to simple_setattr() if i_op->setattr() isn't * implemented. Let's reject it completely until we have a clean @@ -403,7 +493,6 @@ static void pidfs_evict_inode(struct inode *inode) clear_inode(inode); put_pid(pid); - pidfs_free_inum(inode->i_ino); } static const struct super_operations pidfs_sops = { @@ -421,25 +510,149 @@ static char *pidfs_dname(struct dentry *dentry, char *buffer, int buflen) return dynamic_dname(buffer, buflen, "anon_inode:[pidfd]"); } -static const struct dentry_operations pidfs_dentry_operations = { +const struct dentry_operations pidfs_dentry_operations = { .d_delete = always_delete_dentry, .d_dname = pidfs_dname, .d_prune = stashed_dentry_prune, }; +static int pidfs_encode_fh(struct inode *inode, u32 *fh, int *max_len, + struct inode *parent) +{ + const struct pid *pid = inode->i_private; + + if (*max_len < 2) { + *max_len = 2; + return FILEID_INVALID; + } + + *max_len = 2; + *(u64 *)fh = pid->ino; + return FILEID_KERNFS; +} + +static int pidfs_ino_find(const void *key, const struct rb_node *node) +{ + const u64 pid_ino = *(u64 *)key; + const struct pid *pid = rb_entry(node, struct pid, pidfs_node); + + if (pid_ino < pid->ino) + return -1; + if (pid_ino > pid->ino) + return 1; + return 0; +} + +/* Find a struct pid based on the inode number. */ +static struct pid *pidfs_ino_get_pid(u64 ino) +{ + struct pid *pid; + struct rb_node *node; + unsigned int seq; + + guard(rcu)(); + do { + seq = read_seqcount_begin(&pidmap_lock_seq); + node = rb_find_rcu(&ino, &pidfs_ino_tree, pidfs_ino_find); + if (node) + break; + } while (read_seqcount_retry(&pidmap_lock_seq, seq)); + + if (!node) + return NULL; + + pid = rb_entry(node, struct pid, pidfs_node); + + /* Within our pid namespace hierarchy? */ + if (pid_vnr(pid) == 0) + return NULL; + + return get_pid(pid); +} + +static struct dentry *pidfs_fh_to_dentry(struct super_block *sb, + struct fid *fid, int fh_len, + int fh_type) +{ + int ret; + u64 pid_ino; + struct path path; + struct pid *pid; + + if (fh_len < 2) + return NULL; + + switch (fh_type) { + case FILEID_KERNFS: + pid_ino = *(u64 *)fid; + break; + default: + return NULL; + } + + pid = pidfs_ino_get_pid(pid_ino); + if (!pid) + return NULL; + + ret = path_from_stashed(&pid->stashed, pidfs_mnt, pid, &path); + if (ret < 0) + return ERR_PTR(ret); + + mntput(path.mnt); + return path.dentry; +} + +/* + * Make sure that we reject any nonsensical flags that users pass via + * open_by_handle_at(). Note that PIDFD_THREAD is defined as O_EXCL, and + * PIDFD_NONBLOCK as O_NONBLOCK. + */ +#define VALID_FILE_HANDLE_OPEN_FLAGS \ + (O_RDONLY | O_WRONLY | O_RDWR | O_NONBLOCK | O_CLOEXEC | O_EXCL) + +static int pidfs_export_permission(struct handle_to_path_ctx *ctx, + unsigned int oflags) +{ + if (oflags & ~(VALID_FILE_HANDLE_OPEN_FLAGS | O_LARGEFILE)) + return -EINVAL; + + /* + * pidfd_ino_get_pid() will verify that the struct pid is part + * of the caller's pid namespace hierarchy. No further + * permission checks are needed. + */ + return 0; +} + +static struct file *pidfs_export_open(struct path *path, unsigned int oflags) +{ + /* + * Clear O_LARGEFILE as open_by_handle_at() forces it and raise + * O_RDWR as pidfds always are. + */ + oflags &= ~O_LARGEFILE; + return dentry_open(path, oflags | O_RDWR, current_cred()); +} + +static const struct export_operations pidfs_export_operations = { + .encode_fh = pidfs_encode_fh, + .fh_to_dentry = pidfs_fh_to_dentry, + .open = pidfs_export_open, + .permission = pidfs_export_permission, +}; + static int pidfs_init_inode(struct inode *inode, void *data) { + const struct pid *pid = data; + inode->i_private = data; inode->i_flags |= S_PRIVATE; inode->i_mode |= S_IRWXU; inode->i_op = &pidfs_inode_operations; inode->i_fop = &pidfs_file_operations; - /* - * Inode numbering for pidfs start at RESERVED_PIDS + 1. This - * avoids collisions with the root inode which is 1 for pseudo - * filesystems. - */ - return pidfs_inum(data, &inode->i_ino); + inode->i_ino = pidfs_ino(pid->ino); + inode->i_generation = pidfs_gen(pid->ino); + return 0; } static void pidfs_put_data(void *data) @@ -462,6 +675,7 @@ static int pidfs_init_fs_context(struct fs_context *fc) return -ENOMEM; ctx->ops = &pidfs_sops; + ctx->eops = &pidfs_export_operations; ctx->dops = &pidfs_dentry_operations; fc->s_fs_info = (void *)&pidfs_stashed_ops; return 0; diff --git a/fs/pipe.c b/fs/pipe.c index 12b22c2723b7..82fede0f2111 100644 --- a/fs/pipe.c +++ b/fs/pipe.c @@ -253,7 +253,7 @@ pipe_read(struct kiocb *iocb, struct iov_iter *to) size_t total_len = iov_iter_count(to); struct file *filp = iocb->ki_filp; struct pipe_inode_info *pipe = filp->private_data; - bool was_full, wake_next_reader = false; + bool wake_writer = false, wake_next_reader = false; ssize_t ret; /* Null read succeeds. */ @@ -264,14 +264,13 @@ pipe_read(struct kiocb *iocb, struct iov_iter *to) mutex_lock(&pipe->mutex); /* - * We only wake up writers if the pipe was full when we started - * reading in order to avoid unnecessary wakeups. + * We only wake up writers if the pipe was full when we started reading + * and it is no longer full after reading to avoid unnecessary wakeups. * * But when we do wake up writers, we do so using a sync wakeup * (WF_SYNC), because we want them to get going and generate more * data for us. */ - was_full = pipe_full(pipe->head, pipe->tail, pipe->max_usage); for (;;) { /* Read ->head with a barrier vs post_one_notification() */ unsigned int head = smp_load_acquire(&pipe->head); @@ -340,8 +339,10 @@ pipe_read(struct kiocb *iocb, struct iov_iter *to) buf->len = 0; } - if (!buf->len) + if (!buf->len) { + wake_writer |= pipe_full(head, tail, pipe->max_usage); tail = pipe_update_tail(pipe, buf, tail); + } total_len -= chars; if (!total_len) break; /* common path: read succeeded */ @@ -377,7 +378,7 @@ pipe_read(struct kiocb *iocb, struct iov_iter *to) * _very_ unlikely case that the pipe was full, but we got * no data. */ - if (unlikely(was_full)) + if (unlikely(wake_writer)) wake_up_interruptible_sync_poll(&pipe->wr_wait, EPOLLOUT | EPOLLWRNORM); kill_fasync(&pipe->fasync_writers, SIGIO, POLL_OUT); @@ -390,15 +391,15 @@ pipe_read(struct kiocb *iocb, struct iov_iter *to) if (wait_event_interruptible_exclusive(pipe->rd_wait, pipe_readable(pipe)) < 0) return -ERESTARTSYS; - mutex_lock(&pipe->mutex); - was_full = pipe_full(pipe->head, pipe->tail, pipe->max_usage); + wake_writer = false; wake_next_reader = true; + mutex_lock(&pipe->mutex); } if (pipe_empty(pipe->head, pipe->tail)) wake_next_reader = false; mutex_unlock(&pipe->mutex); - if (was_full) + if (wake_writer) wake_up_interruptible_sync_poll(&pipe->wr_wait, EPOLLOUT | EPOLLWRNORM); if (wake_next_reader) wake_up_interruptible_sync_poll(&pipe->rd_wait, EPOLLIN | EPOLLRDNORM); diff --git a/fs/pnode.c b/fs/pnode.c index a799e0315cc9..ef048f008bdd 100644 --- a/fs/pnode.c +++ b/fs/pnode.c @@ -611,10 +611,10 @@ int propagate_umount(struct list_head *list) continue; } else if (child->mnt.mnt_flags & MNT_UMOUNT) { /* - * We have come accross an partially unmounted - * mount in list that has not been visited yet. - * Remember it has been visited and continue - * about our merry way. + * We have come across a partially unmounted + * mount in a list that has not been visited + * yet. Remember it has been visited and + * continue about our merry way. */ list_add_tail(&child->mnt_umounting, &visited); continue; diff --git a/fs/proc/array.c b/fs/proc/array.c index 55ed3510d2bb..d6a0369caa93 100644 --- a/fs/proc/array.c +++ b/fs/proc/array.c @@ -500,7 +500,7 @@ static int do_task_stat(struct seq_file *m, struct pid_namespace *ns, * a program is not able to use ptrace(2) in that case. It is * safe because the task has stopped executing permanently. */ - if (permitted && (task->flags & (PF_EXITING|PF_DUMPCORE))) { + if (permitted && (task->flags & (PF_EXITING|PF_DUMPCORE|PF_POSTCOREDUMP))) { if (try_get_task_stack(task)) { eip = KSTK_EIP(task); esp = KSTK_ESP(task); diff --git a/fs/proc/kcore.c b/fs/proc/kcore.c index e376f48c4b8b..1cb33771bf9f 100644 --- a/fs/proc/kcore.c +++ b/fs/proc/kcore.c @@ -65,7 +65,11 @@ static inline void kc_unxlate_dev_mem_ptr(phys_addr_t phys, void *virt) #endif static LIST_HEAD(kclist_head); -static DECLARE_RWSEM(kclist_lock); +static int kcore_nphdr; +static size_t kcore_phdrs_len; +static size_t kcore_notes_len; +static size_t kcore_data_offset; +DEFINE_STATIC_PERCPU_RWSEM(kclist_lock); static int kcore_need_update = 1; /* @@ -101,33 +105,32 @@ void __init kclist_add(struct kcore_list *new, void *addr, size_t size, list_add_tail(&new->list, &kclist_head); } -static size_t get_kcore_size(int *nphdr, size_t *phdrs_len, size_t *notes_len, - size_t *data_offset) +static void update_kcore_size(void) { size_t try, size; struct kcore_list *m; - *nphdr = 1; /* PT_NOTE */ + kcore_nphdr = 1; /* PT_NOTE */ size = 0; list_for_each_entry(m, &kclist_head, list) { try = kc_vaddr_to_offset((size_t)m->addr + m->size); if (try > size) size = try; - *nphdr = *nphdr + 1; + kcore_nphdr++; } - *phdrs_len = *nphdr * sizeof(struct elf_phdr); - *notes_len = (4 * sizeof(struct elf_note) + - 3 * ALIGN(sizeof(CORE_STR), 4) + - VMCOREINFO_NOTE_NAME_BYTES + - ALIGN(sizeof(struct elf_prstatus), 4) + - ALIGN(sizeof(struct elf_prpsinfo), 4) + - ALIGN(arch_task_struct_size, 4) + - ALIGN(vmcoreinfo_size, 4)); - *data_offset = PAGE_ALIGN(sizeof(struct elfhdr) + *phdrs_len + - *notes_len); - return *data_offset + size; + kcore_phdrs_len = kcore_nphdr * sizeof(struct elf_phdr); + kcore_notes_len = (4 * sizeof(struct elf_note) + + 3 * ALIGN(sizeof(CORE_STR), 4) + + VMCOREINFO_NOTE_NAME_BYTES + + ALIGN(sizeof(struct elf_prstatus), 4) + + ALIGN(sizeof(struct elf_prpsinfo), 4) + + ALIGN(arch_task_struct_size, 4) + + ALIGN(vmcoreinfo_size, 4)); + kcore_data_offset = PAGE_ALIGN(sizeof(struct elfhdr) + kcore_phdrs_len + + kcore_notes_len); + proc_root_kcore->size = kcore_data_offset + size; } #ifdef CONFIG_HIGHMEM @@ -270,12 +273,10 @@ static int kcore_update_ram(void) { LIST_HEAD(list); LIST_HEAD(garbage); - int nphdr; - size_t phdrs_len, notes_len, data_offset; struct kcore_list *tmp, *pos; int ret = 0; - down_write(&kclist_lock); + percpu_down_write(&kclist_lock); if (!xchg(&kcore_need_update, 0)) goto out; @@ -293,11 +294,10 @@ static int kcore_update_ram(void) } list_splice_tail(&list, &kclist_head); - proc_root_kcore->size = get_kcore_size(&nphdr, &phdrs_len, ¬es_len, - &data_offset); + update_kcore_size(); out: - up_write(&kclist_lock); + percpu_up_write(&kclist_lock); list_for_each_entry_safe(pos, tmp, &garbage, list) { list_del(&pos->list); kfree(pos); @@ -326,27 +326,24 @@ static ssize_t read_kcore_iter(struct kiocb *iocb, struct iov_iter *iter) struct file *file = iocb->ki_filp; char *buf = file->private_data; loff_t *fpos = &iocb->ki_pos; - size_t phdrs_offset, notes_offset, data_offset; + size_t phdrs_offset, notes_offset; size_t page_offline_frozen = 1; - size_t phdrs_len, notes_len; struct kcore_list *m; size_t tsz; - int nphdr; unsigned long start; size_t buflen = iov_iter_count(iter); size_t orig_buflen = buflen; int ret = 0; - down_read(&kclist_lock); + percpu_down_read(&kclist_lock); /* * Don't race against drivers that set PageOffline() and expect no * further page access. */ page_offline_freeze(); - get_kcore_size(&nphdr, &phdrs_len, ¬es_len, &data_offset); phdrs_offset = sizeof(struct elfhdr); - notes_offset = phdrs_offset + phdrs_len; + notes_offset = phdrs_offset + kcore_phdrs_len; /* ELF file header. */ if (buflen && *fpos < sizeof(struct elfhdr)) { @@ -368,7 +365,7 @@ static ssize_t read_kcore_iter(struct kiocb *iocb, struct iov_iter *iter) .e_flags = ELF_CORE_EFLAGS, .e_ehsize = sizeof(struct elfhdr), .e_phentsize = sizeof(struct elf_phdr), - .e_phnum = nphdr, + .e_phnum = kcore_nphdr, }; tsz = min_t(size_t, buflen, sizeof(struct elfhdr) - *fpos); @@ -382,10 +379,10 @@ static ssize_t read_kcore_iter(struct kiocb *iocb, struct iov_iter *iter) } /* ELF program headers. */ - if (buflen && *fpos < phdrs_offset + phdrs_len) { + if (buflen && *fpos < phdrs_offset + kcore_phdrs_len) { struct elf_phdr *phdrs, *phdr; - phdrs = kzalloc(phdrs_len, GFP_KERNEL); + phdrs = kzalloc(kcore_phdrs_len, GFP_KERNEL); if (!phdrs) { ret = -ENOMEM; goto out; @@ -393,13 +390,14 @@ static ssize_t read_kcore_iter(struct kiocb *iocb, struct iov_iter *iter) phdrs[0].p_type = PT_NOTE; phdrs[0].p_offset = notes_offset; - phdrs[0].p_filesz = notes_len; + phdrs[0].p_filesz = kcore_notes_len; phdr = &phdrs[1]; list_for_each_entry(m, &kclist_head, list) { phdr->p_type = PT_LOAD; phdr->p_flags = PF_R | PF_W | PF_X; - phdr->p_offset = kc_vaddr_to_offset(m->addr) + data_offset; + phdr->p_offset = kc_vaddr_to_offset(m->addr) + + kcore_data_offset; phdr->p_vaddr = (size_t)m->addr; if (m->type == KCORE_RAM) phdr->p_paddr = __pa(m->addr); @@ -412,7 +410,8 @@ static ssize_t read_kcore_iter(struct kiocb *iocb, struct iov_iter *iter) phdr++; } - tsz = min_t(size_t, buflen, phdrs_offset + phdrs_len - *fpos); + tsz = min_t(size_t, buflen, + phdrs_offset + kcore_phdrs_len - *fpos); if (copy_to_iter((char *)phdrs + *fpos - phdrs_offset, tsz, iter) != tsz) { kfree(phdrs); @@ -426,7 +425,7 @@ static ssize_t read_kcore_iter(struct kiocb *iocb, struct iov_iter *iter) } /* ELF note segment. */ - if (buflen && *fpos < notes_offset + notes_len) { + if (buflen && *fpos < notes_offset + kcore_notes_len) { struct elf_prstatus prstatus = {}; struct elf_prpsinfo prpsinfo = { .pr_sname = 'R', @@ -438,7 +437,7 @@ static ssize_t read_kcore_iter(struct kiocb *iocb, struct iov_iter *iter) strscpy(prpsinfo.pr_psargs, saved_command_line, sizeof(prpsinfo.pr_psargs)); - notes = kzalloc(notes_len, GFP_KERNEL); + notes = kzalloc(kcore_notes_len, GFP_KERNEL); if (!notes) { ret = -ENOMEM; goto out; @@ -459,9 +458,10 @@ static ssize_t read_kcore_iter(struct kiocb *iocb, struct iov_iter *iter) */ append_kcore_note(notes, &i, VMCOREINFO_NOTE_NAME, 0, vmcoreinfo_data, - min(vmcoreinfo_size, notes_len - i)); + min(vmcoreinfo_size, kcore_notes_len - i)); - tsz = min_t(size_t, buflen, notes_offset + notes_len - *fpos); + tsz = min_t(size_t, buflen, + notes_offset + kcore_notes_len - *fpos); if (copy_to_iter(notes + *fpos - notes_offset, tsz, iter) != tsz) { kfree(notes); ret = -EFAULT; @@ -477,7 +477,7 @@ static ssize_t read_kcore_iter(struct kiocb *iocb, struct iov_iter *iter) * Check to see if our file offset matches with any of * the addresses in the elf_phdr on our list. */ - start = kc_offset_to_vaddr(*fpos - data_offset); + start = kc_offset_to_vaddr(*fpos - kcore_data_offset); if ((tsz = (PAGE_SIZE - (start & ~PAGE_MASK))) > buflen) tsz = buflen; @@ -626,7 +626,7 @@ skip: out: page_offline_thaw(); - up_read(&kclist_lock); + percpu_up_read(&kclist_lock); if (ret) return ret; return orig_buflen - buflen; @@ -663,6 +663,7 @@ static int release_kcore(struct inode *inode, struct file *file) } static const struct proc_ops kcore_proc_ops = { + .proc_flags = PROC_ENTRY_PERMANENT, .proc_read_iter = read_kcore_iter, .proc_open = open_kcore, .proc_release = release_kcore, diff --git a/fs/proc/namespaces.c b/fs/proc/namespaces.c index 8e159fc78c0a..c610224faf10 100644 --- a/fs/proc/namespaces.c +++ b/fs/proc/namespaces.c @@ -83,7 +83,7 @@ static int proc_ns_readlink(struct dentry *dentry, char __user *buffer, int bufl if (ptrace_may_access(task, PTRACE_MODE_READ_FSCREDS)) { res = ns_get_name(name, sizeof(name), task, ns_ops); if (res >= 0) - res = readlink_copy(buffer, buflen, name); + res = readlink_copy(buffer, buflen, name, strlen(name)); } put_task_struct(task); return res; diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c index 38a5a3e9cba2..f02cd362309a 100644 --- a/fs/proc/task_mmu.c +++ b/fs/proc/task_mmu.c @@ -1810,7 +1810,7 @@ static int pagemap_pmd_range(pmd_t *pmdp, unsigned long addr, unsigned long end, } for (; addr != end; addr += PAGE_SIZE, idx++) { - unsigned long cur_flags = flags; + u64 cur_flags = flags; pagemap_entry_t pme; if (folio && (flags & PM_PRESENT) && diff --git a/fs/proc/vmcore.c b/fs/proc/vmcore.c index 3d8a82cee63e..658bf199d424 100644 --- a/fs/proc/vmcore.c +++ b/fs/proc/vmcore.c @@ -404,6 +404,8 @@ static ssize_t __read_vmcore(struct iov_iter *iter, loff_t *fpos) if (!iov_iter_count(iter)) return acc; } + + cond_resched(); } return acc; diff --git a/fs/pstore/blk.c b/fs/pstore/blk.c index 65b2473e22ff..fa6b8cb788a1 100644 --- a/fs/pstore/blk.c +++ b/fs/pstore/blk.c @@ -89,7 +89,7 @@ static struct pstore_device_info *pstore_device_info; _##name_ = check_size(name, alignsize); \ else \ _##name_ = 0; \ - /* Synchronize module parameters with resuls. */ \ + /* Synchronize module parameters with results. */ \ name = _##name_ / 1024; \ dev->zone.name = _##name_; \ } @@ -121,7 +121,7 @@ static int __register_pstore_device(struct pstore_device_info *dev) if (pstore_device_info) return -EBUSY; - /* zero means not limit on which backends to attempt to store. */ + /* zero means no limit on which backends attempt to store. */ if (!dev->flags) dev->flags = UINT_MAX; diff --git a/fs/pstore/zone.c b/fs/pstore/zone.c index 694db616663f..ceb5639a0629 100644 --- a/fs/pstore/zone.c +++ b/fs/pstore/zone.c @@ -1212,6 +1212,11 @@ static struct pstore_zone **psz_init_zones(enum pstore_type_id type, } c = total_size / record_size; + if (unlikely(!c)) { + pr_err("zone %s total_size too small\n", name); + return ERR_PTR(-EINVAL); + } + zones = kcalloc(c, sizeof(*zones), GFP_KERNEL); if (!zones) { pr_err("allocate for zones %s failed\n", name); diff --git a/fs/qnx6/inode.c b/fs/qnx6/inode.c index 85925ec0051a..3310d1ad4d0e 100644 --- a/fs/qnx6/inode.c +++ b/fs/qnx6/inode.c @@ -179,8 +179,7 @@ static int qnx6_statfs(struct dentry *dentry, struct kstatfs *buf) */ static const char *qnx6_checkroot(struct super_block *s) { - static char match_root[2][3] = {".\0\0", "..\0"}; - int i, error = 0; + int error = 0; struct qnx6_dir_entry *dir_entry; struct inode *root = d_inode(s->s_root); struct address_space *mapping = root->i_mapping; @@ -189,11 +188,9 @@ static const char *qnx6_checkroot(struct super_block *s) if (IS_ERR(folio)) return "error reading root directory"; dir_entry = kmap_local_folio(folio, 0); - for (i = 0; i < 2; i++) { - /* maximum 3 bytes - due to match_root limitation */ - if (strncmp(dir_entry[i].de_fname, match_root[i], 3)) - error = 1; - } + if (memcmp(dir_entry[0].de_fname, ".", 2) || + memcmp(dir_entry[1].de_fname, "..", 3)) + error = 1; folio_release_kmap(folio, dir_entry); if (error) return "error reading root directory."; diff --git a/fs/select.c b/fs/select.c index e223d1fe9d55..7da531b1cf6b 100644 --- a/fs/select.c +++ b/fs/select.c @@ -786,7 +786,7 @@ static inline int get_sigset_argpack(struct sigset_argpack *to, } return 0; Efault: - user_access_end(); + user_read_access_end(); return -EFAULT; } @@ -1355,7 +1355,7 @@ static inline int get_compat_sigset_argpack(struct compat_sigset_argpack *to, } return 0; Efault: - user_access_end(); + user_read_access_end(); return -EFAULT; } diff --git a/fs/smb/client/cifsproto.h b/fs/smb/client/cifsproto.h index 754417cb3294..d26f9bbb5382 100644 --- a/fs/smb/client/cifsproto.h +++ b/fs/smb/client/cifsproto.h @@ -614,8 +614,6 @@ int cifs_alloc_hash(const char *name, struct shash_desc **sdesc); void cifs_free_hash(struct shash_desc **sdesc); int cifs_try_adding_channels(struct cifs_ses *ses); -bool is_server_using_iface(struct TCP_Server_Info *server, - struct cifs_server_iface *iface); bool is_ses_using_iface(struct cifs_ses *ses, struct cifs_server_iface *iface); void cifs_ses_mark_for_reconnect(struct cifs_ses *ses); diff --git a/fs/smb/client/cifssmb.c b/fs/smb/client/cifssmb.c index bd42a419458e..7f1cacc89dbb 100644 --- a/fs/smb/client/cifssmb.c +++ b/fs/smb/client/cifssmb.c @@ -152,8 +152,17 @@ again: spin_unlock(&ses->ses_lock); rc = cifs_negotiate_protocol(0, ses, server); - if (!rc) + if (!rc) { rc = cifs_setup_session(0, ses, server, ses->local_nls); + if ((rc == -EACCES) || (rc == -EHOSTDOWN) || (rc == -EKEYREVOKED)) { + /* + * Try alternate password for next reconnect if an alternate + * password is available. + */ + if (ses->password2) + swap(ses->password2, ses->password); + } + } /* do we need to reconnect tcon? */ if (rc || !tcon->need_reconnect) { @@ -1258,14 +1267,6 @@ openRetry: return rc; } -static void cifs_readv_worker(struct work_struct *work) -{ - struct cifs_io_subrequest *rdata = - container_of(work, struct cifs_io_subrequest, subreq.work); - - netfs_read_subreq_terminated(&rdata->subreq, rdata->result, false); -} - static void cifs_readv_callback(struct mid_q_entry *mid) { @@ -1319,20 +1320,24 @@ cifs_readv_callback(struct mid_q_entry *mid) } if (rdata->result == -ENODATA) { - __set_bit(NETFS_SREQ_HIT_EOF, &rdata->subreq.flags); rdata->result = 0; + __set_bit(NETFS_SREQ_HIT_EOF, &rdata->subreq.flags); } else { size_t trans = rdata->subreq.transferred + rdata->got_bytes; if (trans < rdata->subreq.len && rdata->subreq.start + trans == ictx->remote_i_size) { - __set_bit(NETFS_SREQ_HIT_EOF, &rdata->subreq.flags); rdata->result = 0; + __set_bit(NETFS_SREQ_HIT_EOF, &rdata->subreq.flags); + } else if (rdata->got_bytes > 0) { + __set_bit(NETFS_SREQ_MADE_PROGRESS, &rdata->subreq.flags); } + if (rdata->got_bytes) + __set_bit(NETFS_SREQ_MADE_PROGRESS, &rdata->subreq.flags); } rdata->credits.value = 0; + rdata->subreq.error = rdata->result; rdata->subreq.transferred += rdata->got_bytes; - INIT_WORK(&rdata->subreq.work, cifs_readv_worker); queue_work(cifsiod_wq, &rdata->subreq.work); release_mid(mid); add_credits(server, &credits, 0); @@ -1670,10 +1675,13 @@ cifs_writev_callback(struct mid_q_entry *mid) if (written > wdata->subreq.len) written &= 0xFFFF; - if (written < wdata->subreq.len) + if (written < wdata->subreq.len) { result = -ENOSPC; - else + } else { result = written; + if (written > 0) + __set_bit(NETFS_SREQ_MADE_PROGRESS, &wdata->subreq.flags); + } break; case MID_REQUEST_SUBMITTED: case MID_RETRY_NEEDED: diff --git a/fs/smb/client/connect.c b/fs/smb/client/connect.c index ddcc9e514a0e..eaa6be4456d0 100644 --- a/fs/smb/client/connect.c +++ b/fs/smb/client/connect.c @@ -1044,6 +1044,7 @@ clean_demultiplex_info(struct TCP_Server_Info *server) /* Release netns reference for this server. */ put_net(cifs_net_ns(server)); kfree(server->leaf_fullpath); + kfree(server->hostname); kfree(server); length = atomic_dec_return(&tcpSesAllocCount); @@ -1670,8 +1671,6 @@ cifs_put_tcp_session(struct TCP_Server_Info *server, int from_reconnect) kfree_sensitive(server->session_key.response); server->session_key.response = NULL; server->session_key.len = 0; - kfree(server->hostname); - server->hostname = NULL; task = xchg(&server->tsk, NULL); if (task) diff --git a/fs/smb/client/file.c b/fs/smb/client/file.c index a58a3333ecc3..79de2f2f9c41 100644 --- a/fs/smb/client/file.c +++ b/fs/smb/client/file.c @@ -227,7 +227,8 @@ static void cifs_issue_read(struct netfs_io_subrequest *subreq) return; failed: - netfs_read_subreq_terminated(subreq, rc, false); + subreq->error = rc; + netfs_read_subreq_terminated(subreq); } /* @@ -990,7 +991,11 @@ int cifs_open(struct inode *inode, struct file *file) } /* Get the cached handle as SMB2 close is deferred */ - rc = cifs_get_readable_path(tcon, full_path, &cfile); + if (OPEN_FMODE(file->f_flags) & FMODE_WRITE) { + rc = cifs_get_writable_path(tcon, full_path, FIND_WR_FSUID_ONLY, &cfile); + } else { + rc = cifs_get_readable_path(tcon, full_path, &cfile); + } if (rc == 0) { if (file->f_flags == cfile->f_flags) { file->private_data = cfile; diff --git a/fs/smb/client/namespace.c b/fs/smb/client/namespace.c index 0f788031b740..e3f9213131c4 100644 --- a/fs/smb/client/namespace.c +++ b/fs/smb/client/namespace.c @@ -196,11 +196,28 @@ static struct vfsmount *cifs_do_automount(struct path *path) struct smb3_fs_context tmp; char *full_path; struct vfsmount *mnt; + struct cifs_sb_info *mntpt_sb; + struct cifs_ses *ses; if (IS_ROOT(mntpt)) return ERR_PTR(-ESTALE); - cur_ctx = CIFS_SB(mntpt->d_sb)->ctx; + mntpt_sb = CIFS_SB(mntpt->d_sb); + ses = cifs_sb_master_tcon(mntpt_sb)->ses; + cur_ctx = mntpt_sb->ctx; + + /* + * At this point, the root session should be in the mntpt sb. We should + * bring the sb context passwords in sync with the root session's + * passwords. This would help prevent unnecessary retries and password + * swaps for automounts. + */ + mutex_lock(&ses->session_mutex); + rc = smb3_sync_session_ctx_passwords(mntpt_sb, ses); + mutex_unlock(&ses->session_mutex); + + if (rc) + return ERR_PTR(rc); fc = fs_context_for_submount(path->mnt->mnt_sb->s_type, mntpt); if (IS_ERR(fc)) diff --git a/fs/smb/client/sess.c b/fs/smb/client/sess.c index 3306fb655136..91d4d409cb1d 100644 --- a/fs/smb/client/sess.c +++ b/fs/smb/client/sess.c @@ -27,31 +27,6 @@ static int cifs_ses_add_channel(struct cifs_ses *ses, struct cifs_server_iface *iface); -bool -is_server_using_iface(struct TCP_Server_Info *server, - struct cifs_server_iface *iface) -{ - struct sockaddr_in *i4 = (struct sockaddr_in *)&iface->sockaddr; - struct sockaddr_in6 *i6 = (struct sockaddr_in6 *)&iface->sockaddr; - struct sockaddr_in *s4 = (struct sockaddr_in *)&server->dstaddr; - struct sockaddr_in6 *s6 = (struct sockaddr_in6 *)&server->dstaddr; - - if (server->dstaddr.ss_family != iface->sockaddr.ss_family) - return false; - if (server->dstaddr.ss_family == AF_INET) { - if (s4->sin_addr.s_addr != i4->sin_addr.s_addr) - return false; - } else if (server->dstaddr.ss_family == AF_INET6) { - if (memcmp(&s6->sin6_addr, &i6->sin6_addr, - sizeof(i6->sin6_addr)) != 0) - return false; - } else { - /* unknown family.. */ - return false; - } - return true; -} - bool is_ses_using_iface(struct cifs_ses *ses, struct cifs_server_iface *iface) { int i; diff --git a/fs/smb/client/smb2ops.c b/fs/smb/client/smb2ops.c index 87cb1872db28..7121d9e0f404 100644 --- a/fs/smb/client/smb2ops.c +++ b/fs/smb/client/smb2ops.c @@ -4388,7 +4388,7 @@ static struct folio_queue *cifs_alloc_folioq_buffer(ssize_t size) p = kmalloc(sizeof(*p), GFP_NOFS); if (!p) goto nomem; - folioq_init(p); + folioq_init(p, 0); if (tail) { tail->next = p; p->prev = tail; diff --git a/fs/smb/client/smb2pdu.c b/fs/smb/client/smb2pdu.c index 959359301250..9f54596a6866 100644 --- a/fs/smb/client/smb2pdu.c +++ b/fs/smb/client/smb2pdu.c @@ -4500,14 +4500,6 @@ smb2_new_read_req(void **buf, unsigned int *total_len, return rc; } -static void smb2_readv_worker(struct work_struct *work) -{ - struct cifs_io_subrequest *rdata = - container_of(work, struct cifs_io_subrequest, subreq.work); - - netfs_read_subreq_terminated(&rdata->subreq, rdata->result, false); -} - static void smb2_readv_callback(struct mid_q_entry *mid) { @@ -4615,15 +4607,17 @@ smb2_readv_callback(struct mid_q_entry *mid) __set_bit(NETFS_SREQ_HIT_EOF, &rdata->subreq.flags); rdata->result = 0; } + if (rdata->got_bytes) + __set_bit(NETFS_SREQ_MADE_PROGRESS, &rdata->subreq.flags); } trace_smb3_rw_credits(rreq_debug_id, subreq_debug_index, rdata->credits.value, server->credits, server->in_flight, 0, cifs_trace_rw_credits_read_response_clear); rdata->credits.value = 0; + rdata->subreq.error = rdata->result; rdata->subreq.transferred += rdata->got_bytes; trace_netfs_sreq(&rdata->subreq, netfs_sreq_trace_io_progress); - INIT_WORK(&rdata->subreq.work, smb2_readv_worker); - queue_work(cifsiod_wq, &rdata->subreq.work); + netfs_read_subreq_terminated(&rdata->subreq); release_mid(mid); trace_smb3_rw_credits(rreq_debug_id, subreq_debug_index, 0, server->credits, server->in_flight, @@ -4842,10 +4836,12 @@ smb2_writev_callback(struct mid_q_entry *mid) cifs_stats_bytes_written(tcon, written); - if (written < wdata->subreq.len) + if (written < wdata->subreq.len) { wdata->result = -ENOSPC; - else + } else if (written > 0) { wdata->subreq.len = written; + __set_bit(NETFS_SREQ_MADE_PROGRESS, &wdata->subreq.flags); + } break; case MID_REQUEST_SUBMITTED: case MID_RETRY_NEEDED: @@ -5014,7 +5010,7 @@ smb2_async_writev(struct cifs_io_subrequest *wdata) } #endif - if (test_bit(NETFS_SREQ_RETRYING, &wdata->subreq.flags)) + if (wdata->subreq.retry_count > 0) smb2_set_replay(server, &rqst); cifs_dbg(FYI, "async write at %llu %u bytes iter=%zx\n", diff --git a/fs/smb/server/smb2pdu.c b/fs/smb/server/smb2pdu.c index 23e21845f928..772deec5b90f 100644 --- a/fs/smb/server/smb2pdu.c +++ b/fs/smb/server/smb2pdu.c @@ -695,6 +695,9 @@ void smb2_send_interim_resp(struct ksmbd_work *work, __le32 status) struct smb2_hdr *rsp_hdr; struct ksmbd_work *in_work = ksmbd_alloc_work_struct(); + if (!in_work) + return; + if (allocate_interim_rsp_buf(in_work)) { pr_err("smb_allocate_rsp_buf failed!\n"); ksmbd_free_work_struct(in_work); @@ -3991,6 +3994,26 @@ static int smb2_populate_readdir_entry(struct ksmbd_conn *conn, int info_level, posix_info->DeviceId = cpu_to_le32(ksmbd_kstat->kstat->rdev); posix_info->HardLinks = cpu_to_le32(ksmbd_kstat->kstat->nlink); posix_info->Mode = cpu_to_le32(ksmbd_kstat->kstat->mode & 0777); + switch (ksmbd_kstat->kstat->mode & S_IFMT) { + case S_IFDIR: + posix_info->Mode |= cpu_to_le32(POSIX_TYPE_DIR << POSIX_FILETYPE_SHIFT); + break; + case S_IFLNK: + posix_info->Mode |= cpu_to_le32(POSIX_TYPE_SYMLINK << POSIX_FILETYPE_SHIFT); + break; + case S_IFCHR: + posix_info->Mode |= cpu_to_le32(POSIX_TYPE_CHARDEV << POSIX_FILETYPE_SHIFT); + break; + case S_IFBLK: + posix_info->Mode |= cpu_to_le32(POSIX_TYPE_BLKDEV << POSIX_FILETYPE_SHIFT); + break; + case S_IFIFO: + posix_info->Mode |= cpu_to_le32(POSIX_TYPE_FIFO << POSIX_FILETYPE_SHIFT); + break; + case S_IFSOCK: + posix_info->Mode |= cpu_to_le32(POSIX_TYPE_SOCKET << POSIX_FILETYPE_SHIFT); + } + posix_info->Inode = cpu_to_le64(ksmbd_kstat->kstat->ino); posix_info->DosAttributes = S_ISDIR(ksmbd_kstat->kstat->mode) ? @@ -5181,6 +5204,26 @@ static int find_file_posix_info(struct smb2_query_info_rsp *rsp, file_info->AllocationSize = cpu_to_le64(stat.blocks << 9); file_info->HardLinks = cpu_to_le32(stat.nlink); file_info->Mode = cpu_to_le32(stat.mode & 0777); + switch (stat.mode & S_IFMT) { + case S_IFDIR: + file_info->Mode |= cpu_to_le32(POSIX_TYPE_DIR << POSIX_FILETYPE_SHIFT); + break; + case S_IFLNK: + file_info->Mode |= cpu_to_le32(POSIX_TYPE_SYMLINK << POSIX_FILETYPE_SHIFT); + break; + case S_IFCHR: + file_info->Mode |= cpu_to_le32(POSIX_TYPE_CHARDEV << POSIX_FILETYPE_SHIFT); + break; + case S_IFBLK: + file_info->Mode |= cpu_to_le32(POSIX_TYPE_BLKDEV << POSIX_FILETYPE_SHIFT); + break; + case S_IFIFO: + file_info->Mode |= cpu_to_le32(POSIX_TYPE_FIFO << POSIX_FILETYPE_SHIFT); + break; + case S_IFSOCK: + file_info->Mode |= cpu_to_le32(POSIX_TYPE_SOCKET << POSIX_FILETYPE_SHIFT); + } + file_info->DeviceId = cpu_to_le32(stat.rdev); /* diff --git a/fs/smb/server/smb2pdu.h b/fs/smb/server/smb2pdu.h index 649dacf7e8c4..17a0b18a8406 100644 --- a/fs/smb/server/smb2pdu.h +++ b/fs/smb/server/smb2pdu.h @@ -502,4 +502,14 @@ static inline void *smb2_get_msg(void *buf) return buf + 4; } +#define POSIX_TYPE_FILE 0 +#define POSIX_TYPE_DIR 1 +#define POSIX_TYPE_SYMLINK 2 +#define POSIX_TYPE_CHARDEV 3 +#define POSIX_TYPE_BLKDEV 4 +#define POSIX_TYPE_FIFO 5 +#define POSIX_TYPE_SOCKET 6 + +#define POSIX_FILETYPE_SHIFT 12 + #endif /* _SMB2PDU_H */ diff --git a/fs/smb/server/smb_common.c b/fs/smb/server/smb_common.c index f51cd0816b1a..425c756bcfb8 100644 --- a/fs/smb/server/smb_common.c +++ b/fs/smb/server/smb_common.c @@ -781,10 +781,6 @@ int __ksmbd_override_fsids(struct ksmbd_work *work, WARN_ON(work->saved_cred); work->saved_cred = override_creds(cred); - if (!work->saved_cred) { - abort_creds(cred); - return -EINVAL; - } return 0; } @@ -796,13 +792,11 @@ int ksmbd_override_fsids(struct ksmbd_work *work) void ksmbd_revert_fsids(struct ksmbd_work *work) { const struct cred *cred; - WARN_ON(!work->saved_cred); - cred = current_cred(); - revert_creds(work->saved_cred); - put_cred(cred); + cred = revert_creds(work->saved_cred); work->saved_cred = NULL; + put_cred(cred); } __le32 smb_map_generic_desired_access(__le32 daccess) diff --git a/fs/smb/server/transport_rdma.c b/fs/smb/server/transport_rdma.c index 0ef3c9f0bfeb..c3785a5434f9 100644 --- a/fs/smb/server/transport_rdma.c +++ b/fs/smb/server/transport_rdma.c @@ -2283,8 +2283,7 @@ out: ibdev = ib_device_get_by_netdev(netdev, RDMA_DRIVER_UNKNOWN); if (ibdev) { - if (rdma_frwr_is_supported(&ibdev->attrs)) - rdma_capable = true; + rdma_capable = rdma_frwr_is_supported(&ibdev->attrs); ib_device_put(ibdev); } } diff --git a/fs/smb/server/vfs.c b/fs/smb/server/vfs.c index 88d167a5f8b7..40f08eac519c 100644 --- a/fs/smb/server/vfs.c +++ b/fs/smb/server/vfs.c @@ -1264,6 +1264,8 @@ int ksmbd_vfs_kern_path_locked(struct ksmbd_work *work, char *name, filepath, flags, path); + if (!is_last) + next[0] = '/'; if (err) goto out2; else if (is_last) @@ -1271,7 +1273,6 @@ int ksmbd_vfs_kern_path_locked(struct ksmbd_work *work, char *name, path_put(parent_path); *parent_path = *path; - next[0] = '/'; remain_len -= filename_len + 1; } diff --git a/fs/stat.c b/fs/stat.c index 0870e969a8a0..2c0e111a098a 100644 --- a/fs/stat.c +++ b/fs/stat.c @@ -725,6 +725,7 @@ cp_statx(const struct kstat *stat, struct statx __user *buffer) tmp.stx_mnt_id = stat->mnt_id; tmp.stx_dio_mem_align = stat->dio_mem_align; tmp.stx_dio_offset_align = stat->dio_offset_align; + tmp.stx_dio_read_offset_align = stat->dio_read_offset_align; tmp.stx_subvol = stat->subvol; tmp.stx_atomic_write_unit_min = stat->atomic_write_unit_min; tmp.stx_atomic_write_unit_max = stat->atomic_write_unit_max; diff --git a/fs/vboxsf/Kconfig b/fs/vboxsf/Kconfig index b84586ae08b3..d4694026db8b 100644 --- a/fs/vboxsf/Kconfig +++ b/fs/vboxsf/Kconfig @@ -1,6 +1,6 @@ config VBOXSF_FS tristate "VirtualBox guest shared folder (vboxsf) support" - depends on X86 && VBOXGUEST + depends on (ARM64 || X86) && VBOXGUEST select NLS help VirtualBox hosts can share folders with guests, this driver diff --git a/fs/xfs/libxfs/xfs_rtgroup.h b/fs/xfs/libxfs/xfs_rtgroup.h index 7e7e491ff06f..2d7822644eff 100644 --- a/fs/xfs/libxfs/xfs_rtgroup.h +++ b/fs/xfs/libxfs/xfs_rtgroup.h @@ -272,7 +272,7 @@ static inline int xfs_initialize_rtgroups(struct xfs_mount *mp, } # define xfs_rtgroup_extents(mp, rgno) (0) -# define xfs_update_last_rtgroup_size(mp, rgno) (-EOPNOTSUPP) +# define xfs_update_last_rtgroup_size(mp, rgno) (0) # define xfs_rtgroup_lock(rtg, gf) ((void)0) # define xfs_rtgroup_unlock(rtg, gf) ((void)0) # define xfs_rtgroup_trans_join(tp, rtg, gf) ((void)0) diff --git a/fs/xfs/xfs_dquot.c b/fs/xfs/xfs_dquot.c index f11d475898f2..201c26322ede 100644 --- a/fs/xfs/xfs_dquot.c +++ b/fs/xfs/xfs_dquot.c @@ -87,8 +87,9 @@ xfs_dquot_detach_buf( } spin_unlock(&qlip->qli_lock); if (bp) { + xfs_buf_lock(bp); list_del_init(&qlip->qli_item.li_bio_list); - xfs_buf_rele(bp); + xfs_buf_relse(bp); } } diff --git a/fs/xfs/xfs_ioctl.c b/fs/xfs/xfs_ioctl.c index 0789c18aaa18..f95103325318 100644 --- a/fs/xfs/xfs_ioctl.c +++ b/fs/xfs/xfs_ioctl.c @@ -1204,7 +1204,16 @@ xfs_file_ioctl( struct xfs_buftarg *target = xfs_inode_buftarg(ip); struct dioattr da; - da.d_mem = da.d_miniosz = target->bt_logical_sectorsize; + da.d_mem = target->bt_logical_sectorsize; + + /* + * See xfs_report_dioalign() for an explanation about why this + * reports a value larger than the sector size for COW inodes. + */ + if (xfs_is_cow_inode(ip)) + da.d_miniosz = xfs_inode_alloc_unitsize(ip); + else + da.d_miniosz = target->bt_logical_sectorsize; da.d_maxiosz = INT_MAX & ~(da.d_miniosz - 1); if (copy_to_user(arg, &da, sizeof(da))) diff --git a/fs/xfs/xfs_iops.c b/fs/xfs/xfs_iops.c index 207e0dadffc3..40289fe6f5b2 100644 --- a/fs/xfs/xfs_iops.c +++ b/fs/xfs/xfs_iops.c @@ -573,17 +573,43 @@ xfs_stat_blksize( } static void -xfs_get_atomic_write_attr( +xfs_report_dioalign( struct xfs_inode *ip, - unsigned int *unit_min, - unsigned int *unit_max) + struct kstat *stat) { - if (!xfs_inode_can_atomicwrite(ip)) { - *unit_min = *unit_max = 0; - return; - } + struct xfs_buftarg *target = xfs_inode_buftarg(ip); + struct block_device *bdev = target->bt_bdev; + + stat->result_mask |= STATX_DIOALIGN | STATX_DIO_READ_ALIGN; + stat->dio_mem_align = bdev_dma_alignment(bdev) + 1; - *unit_min = *unit_max = ip->i_mount->m_sb.sb_blocksize; + /* + * For COW inodes, we can only perform out of place writes of entire + * allocation units (blocks or RT extents). + * For writes smaller than the allocation unit, we must fall back to + * buffered I/O to perform read-modify-write cycles. At best this is + * highly inefficient; at worst it leads to page cache invalidation + * races. Tell applications to avoid this by reporting the larger write + * alignment in dio_offset_align, and the smaller read alignment in + * dio_read_offset_align. + */ + stat->dio_read_offset_align = bdev_logical_block_size(bdev); + if (xfs_is_cow_inode(ip)) + stat->dio_offset_align = xfs_inode_alloc_unitsize(ip); + else + stat->dio_offset_align = stat->dio_read_offset_align; +} + +static void +xfs_report_atomic_write( + struct xfs_inode *ip, + struct kstat *stat) +{ + unsigned int unit_min = 0, unit_max = 0; + + if (xfs_inode_can_atomicwrite(ip)) + unit_min = unit_max = ip->i_mount->m_sb.sb_blocksize; + generic_fill_statx_atomic_writes(stat, unit_min, unit_max); } STATIC int @@ -647,22 +673,10 @@ xfs_vn_getattr( stat->rdev = inode->i_rdev; break; case S_IFREG: - if (request_mask & STATX_DIOALIGN) { - struct xfs_buftarg *target = xfs_inode_buftarg(ip); - struct block_device *bdev = target->bt_bdev; - - stat->result_mask |= STATX_DIOALIGN; - stat->dio_mem_align = bdev_dma_alignment(bdev) + 1; - stat->dio_offset_align = bdev_logical_block_size(bdev); - } - if (request_mask & STATX_WRITE_ATOMIC) { - unsigned int unit_min, unit_max; - - xfs_get_atomic_write_attr(ip, &unit_min, - &unit_max); - generic_fill_statx_atomic_writes(stat, - unit_min, unit_max); - } + if (request_mask & (STATX_DIOALIGN | STATX_DIO_READ_ALIGN)) + xfs_report_dioalign(ip, stat); + if (request_mask & STATX_WRITE_ATOMIC) + xfs_report_atomic_write(ip, stat); fallthrough; default: stat->blksize = xfs_stat_blksize(ip); diff --git a/include/kvm/arm_pmu.h b/include/kvm/arm_pmu.h index e61dd7dd2286..147bd3ee4f7b 100644 --- a/include/kvm/arm_pmu.h +++ b/include/kvm/arm_pmu.h @@ -53,8 +53,7 @@ u64 kvm_pmu_get_pmceid(struct kvm_vcpu *vcpu, bool pmceid1); void kvm_pmu_vcpu_init(struct kvm_vcpu *vcpu); void kvm_pmu_vcpu_reset(struct kvm_vcpu *vcpu); void kvm_pmu_vcpu_destroy(struct kvm_vcpu *vcpu); -void kvm_pmu_disable_counter_mask(struct kvm_vcpu *vcpu, u64 val); -void kvm_pmu_enable_counter_mask(struct kvm_vcpu *vcpu, u64 val); +void kvm_pmu_reprogram_counter_mask(struct kvm_vcpu *vcpu, u64 val); void kvm_pmu_flush_hwstate(struct kvm_vcpu *vcpu); void kvm_pmu_sync_hwstate(struct kvm_vcpu *vcpu); bool kvm_pmu_should_notify_user(struct kvm_vcpu *vcpu); @@ -127,8 +126,7 @@ static inline u64 kvm_pmu_accessible_counter_mask(struct kvm_vcpu *vcpu) static inline void kvm_pmu_vcpu_init(struct kvm_vcpu *vcpu) {} static inline void kvm_pmu_vcpu_reset(struct kvm_vcpu *vcpu) {} static inline void kvm_pmu_vcpu_destroy(struct kvm_vcpu *vcpu) {} -static inline void kvm_pmu_disable_counter_mask(struct kvm_vcpu *vcpu, u64 val) {} -static inline void kvm_pmu_enable_counter_mask(struct kvm_vcpu *vcpu, u64 val) {} +static inline void kvm_pmu_reprogram_counter_mask(struct kvm_vcpu *vcpu, u64 val) {} static inline void kvm_pmu_flush_hwstate(struct kvm_vcpu *vcpu) {} static inline void kvm_pmu_sync_hwstate(struct kvm_vcpu *vcpu) {} static inline bool kvm_pmu_should_notify_user(struct kvm_vcpu *vcpu) diff --git a/include/linux/binfmts.h b/include/linux/binfmts.h index e6c00e860951..3305c849abd6 100644 --- a/include/linux/binfmts.h +++ b/include/linux/binfmts.h @@ -42,7 +42,9 @@ struct linux_binprm { * Set when errors can no longer be returned to the * original userspace. */ - point_of_no_return:1; + point_of_no_return:1, + /* Set when "comm" must come from the dentry. */ + comm_from_dentry:1; struct file *executable; /* Executable to pass to the interpreter */ struct file *interpreter; struct file *file; diff --git a/include/linux/bus/stm32_firewall_device.h b/include/linux/bus/stm32_firewall_device.h index 18e0a2fc3816..5178b72bc920 100644 --- a/include/linux/bus/stm32_firewall_device.h +++ b/include/linux/bus/stm32_firewall_device.h @@ -115,7 +115,7 @@ void stm32_firewall_release_access_by_id(struct stm32_firewall *firewall, u32 su #else /* CONFIG_STM32_FIREWALL */ int stm32_firewall_get_firewall(struct device_node *np, struct stm32_firewall *firewall, - unsigned int nb_firewall); + unsigned int nb_firewall) { return -ENODEV; } diff --git a/include/linux/coredump.h b/include/linux/coredump.h index 45e598fe3476..77e6e195d1d6 100644 --- a/include/linux/coredump.h +++ b/include/linux/coredump.h @@ -52,8 +52,8 @@ extern void do_coredump(const kernel_siginfo_t *siginfo); #define __COREDUMP_PRINTK(Level, Format, ...) \ do { \ char comm[TASK_COMM_LEN]; \ - \ - get_task_comm(comm, current); \ + /* This will always be NUL terminated. */ \ + memcpy(comm, current->comm, sizeof(comm)); \ printk_ratelimited(Level "coredump: %d(%*pE): " Format "\n", \ task_tgid_vnr(current), (int)strlen(comm), comm, ##__VA_ARGS__); \ } while (0) \ diff --git a/include/linux/cred.h b/include/linux/cred.h index e4a3155fe409..0c3c4b16b469 100644 --- a/include/linux/cred.h +++ b/include/linux/cred.h @@ -155,8 +155,6 @@ extern struct cred *prepare_creds(void); extern struct cred *prepare_exec_creds(void); extern int commit_creds(struct cred *); extern void abort_creds(struct cred *); -extern const struct cred *override_creds(const struct cred *); -extern void revert_creds(const struct cred *); extern struct cred *prepare_kernel_cred(struct task_struct *); extern int set_security_override(struct cred *, u32); extern int set_security_override_from_ctx(struct cred *, const char *); @@ -172,12 +170,7 @@ static inline bool cap_ambient_invariant_ok(const struct cred *cred) cred->cap_inheritable)); } -/* - * Override creds without bumping reference count. Caller must ensure - * reference remains valid or has taken reference. Almost always not the - * interface you want. Use override_creds()/revert_creds() instead. - */ -static inline const struct cred *override_creds_light(const struct cred *override_cred) +static inline const struct cred *override_creds(const struct cred *override_cred) { const struct cred *old = current->cred; @@ -185,35 +178,12 @@ static inline const struct cred *override_creds_light(const struct cred *overrid return old; } -static inline void revert_creds_light(const struct cred *revert_cred) -{ - rcu_assign_pointer(current->cred, revert_cred); -} - -/** - * get_new_cred_many - Get references on a new set of credentials - * @cred: The new credentials to reference - * @nr: Number of references to acquire - * - * Get references on the specified set of new credentials. The caller must - * release all acquired references. - */ -static inline struct cred *get_new_cred_many(struct cred *cred, int nr) +static inline const struct cred *revert_creds(const struct cred *revert_cred) { - atomic_long_add(nr, &cred->usage); - return cred; -} + const struct cred *override_cred = current->cred; -/** - * get_new_cred - Get a reference on a new set of credentials - * @cred: The new credentials to reference - * - * Get a reference on the specified set of new credentials. The caller must - * release the reference. - */ -static inline struct cred *get_new_cred(struct cred *cred) -{ - return get_new_cred_many(cred, 1); + rcu_assign_pointer(current->cred, revert_cred); + return override_cred; } /** @@ -236,7 +206,8 @@ static inline const struct cred *get_cred_many(const struct cred *cred, int nr) if (!cred) return cred; nonconst_cred->non_rcu = 0; - return get_new_cred_many(nonconst_cred, nr); + atomic_long_add(nr, &nonconst_cred->usage); + return cred; } /* diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h index b137fdb56093..346251bf1026 100644 --- a/include/linux/dmaengine.h +++ b/include/linux/dmaengine.h @@ -84,7 +84,7 @@ enum dma_transfer_direction { DMA_TRANS_NONE, }; -/** +/* * Interleaved Transfer Request * ---------------------------- * A chunk is collection of contiguous bytes to be transferred. @@ -223,7 +223,7 @@ enum sum_check_bits { }; /** - * enum pq_check_flags - result of async_{xor,pq}_zero_sum operations + * enum sum_check_flags - result of async_{xor,pq}_zero_sum operations * @SUM_CHECK_P_RESULT - 1 if xor zero sum error, 0 otherwise * @SUM_CHECK_Q_RESULT - 1 if reed-solomon zero sum error, 0 otherwise */ @@ -286,7 +286,7 @@ typedef struct { DECLARE_BITMAP(bits, DMA_TX_TYPE_END); } dma_cap_mask_t; * pointer to the engine's metadata area * 4. Read out the metadata from the pointer * - * Note: the two mode is not compatible and clients must use one mode for a + * Warning: the two modes are not compatible and clients must use one mode for a * descriptor. */ enum dma_desc_metadata_mode { @@ -594,9 +594,13 @@ struct dma_descriptor_metadata_ops { * @phys: physical address of the descriptor * @chan: target channel for this operation * @tx_submit: accept the descriptor, assign ordered cookie and mark the + * @desc_free: driver's callback function to free a resusable descriptor + * after completion * descriptor pending. To be pushed on .issue_pending() call * @callback: routine to call after this operation is complete + * @callback_result: error result from a DMA transaction * @callback_param: general parameter to pass to the callback routine + * @unmap: hook for generic DMA unmap data * @desc_metadata_mode: core managed metadata mode to protect mixed use of * DESC_METADATA_CLIENT or DESC_METADATA_ENGINE. Otherwise * DESC_METADATA_NONE @@ -827,6 +831,9 @@ struct dma_filter { * @device_prep_dma_memset: prepares a memset operation * @device_prep_dma_memset_sg: prepares a memset operation over a scatter list * @device_prep_dma_interrupt: prepares an end of chain interrupt operation + * @device_prep_peripheral_dma_vec: prepares a scatter-gather DMA transfer, + * where the address and size of each segment is located in one entry of + * the dma_vec array. * @device_prep_slave_sg: prepares a slave dma operation * @device_prep_dma_cyclic: prepare a cyclic dma operation suitable for audio. * The function takes a buffer of size buf_len. The callback function will diff --git a/include/linux/exportfs.h b/include/linux/exportfs.h index 4cc8801e50e3..a087606ace19 100644 --- a/include/linux/exportfs.h +++ b/include/linux/exportfs.h @@ -3,6 +3,7 @@ #define LINUX_EXPORTFS_H 1 #include <linux/types.h> +#include <linux/path.h> struct dentry; struct iattr; @@ -156,6 +157,17 @@ struct fid { }; }; +enum handle_to_path_flags { + HANDLE_CHECK_PERMS = (1 << 0), + HANDLE_CHECK_SUBTREE = (1 << 1), +}; + +struct handle_to_path_ctx { + struct path root; + enum handle_to_path_flags flags; + unsigned int fh_flags; +}; + #define EXPORT_FH_CONNECTABLE 0x1 /* Encode file handle with parent */ #define EXPORT_FH_FID 0x2 /* File handle may be non-decodeable */ #define EXPORT_FH_DIR_ONLY 0x4 /* Only decode file handle for a directory */ @@ -225,6 +237,12 @@ struct fid { * is also a directory. In the event that it cannot be found, or storage * space cannot be allocated, a %ERR_PTR should be returned. * + * permission: + * Allow filesystems to specify a custom permission function. + * + * open: + * Allow filesystems to specify a custom open function. + * * commit_metadata: * @commit_metadata should commit metadata changes to stable storage. * @@ -251,6 +269,8 @@ struct export_operations { bool write, u32 *device_generation); int (*commit_blocks)(struct inode *inode, struct iomap *iomaps, int nr_iomaps, struct iattr *iattr); + int (*permission)(struct handle_to_path_ctx *ctx, unsigned int oflags); + struct file * (*open)(struct path *path, unsigned int oflags); #define EXPORT_OP_NOWCC (0x1) /* don't collect v3 wcc data */ #define EXPORT_OP_NOSUBTREECHK (0x2) /* no subtree checking */ #define EXPORT_OP_CLOSE_BEFORE_UNLINK (0x4) /* close files before unlink */ diff --git a/include/linux/fiemap.h b/include/linux/fiemap.h index c50882f19235..966092ffa89a 100644 --- a/include/linux/fiemap.h +++ b/include/linux/fiemap.h @@ -5,12 +5,18 @@ #include <uapi/linux/fiemap.h> #include <linux/fs.h> +/** + * struct fiemap_extent_info - fiemap request to a filesystem + * @fi_flags: Flags as passed from user + * @fi_extents_mapped: Number of mapped extents + * @fi_extents_max: Size of fiemap_extent array + * @fi_extents_start: Start of fiemap_extent array + */ struct fiemap_extent_info { - unsigned int fi_flags; /* Flags as passed from user */ - unsigned int fi_extents_mapped; /* Number of mapped extents */ - unsigned int fi_extents_max; /* Size of fiemap_extent array */ - struct fiemap_extent __user *fi_extents_start; /* Start of - fiemap_extent array */ + unsigned int fi_flags; + unsigned int fi_extents_mapped; + unsigned int fi_extents_max; + struct fiemap_extent __user *fi_extents_start; }; int fiemap_prep(struct inode *inode, struct fiemap_extent_info *fieinfo, diff --git a/include/linux/folio_queue.h b/include/linux/folio_queue.h index 3abe614ef5f0..4d3f8074c137 100644 --- a/include/linux/folio_queue.h +++ b/include/linux/folio_queue.h @@ -37,16 +37,20 @@ struct folio_queue { #if PAGEVEC_SIZE > BITS_PER_LONG #error marks is not big enough #endif + unsigned int rreq_id; + unsigned int debug_id; }; /** * folioq_init - Initialise a folio queue segment * @folioq: The segment to initialise + * @rreq_id: The request identifier to use in tracelines. * - * Initialise a folio queue segment. Note that the folio pointers are - * left uninitialised. + * Initialise a folio queue segment and set an identifier to be used in traces. + * + * Note that the folio pointers are left uninitialised. */ -static inline void folioq_init(struct folio_queue *folioq) +static inline void folioq_init(struct folio_queue *folioq, unsigned int rreq_id) { folio_batch_init(&folioq->vec); folioq->next = NULL; @@ -54,6 +58,8 @@ static inline void folioq_init(struct folio_queue *folioq) folioq->marks = 0; folioq->marks2 = 0; folioq->marks3 = 0; + folioq->rreq_id = rreq_id; + folioq->debug_id = 0; } /** diff --git a/include/linux/fs.h b/include/linux/fs.h index 7e29433c5ecc..2b8447f53200 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -322,6 +322,7 @@ struct readahead_control; #define IOCB_NOWAIT (__force int) RWF_NOWAIT #define IOCB_APPEND (__force int) RWF_APPEND #define IOCB_ATOMIC (__force int) RWF_ATOMIC +#define IOCB_DONTCACHE (__force int) RWF_DONTCACHE /* non-RWF related bits - start at 16 */ #define IOCB_EVENTFD (1 << 16) @@ -356,7 +357,8 @@ struct readahead_control; { IOCB_SYNC, "SYNC" }, \ { IOCB_NOWAIT, "NOWAIT" }, \ { IOCB_APPEND, "APPEND" }, \ - { IOCB_ATOMIC, "ATOMIC"}, \ + { IOCB_ATOMIC, "ATOMIC" }, \ + { IOCB_DONTCACHE, "DONTCACHE" }, \ { IOCB_EVENTFD, "EVENTFD"}, \ { IOCB_DIRECT, "DIRECT" }, \ { IOCB_WRITE, "WRITE" }, \ @@ -626,6 +628,7 @@ is_uncached_acl(struct posix_acl *acl) #define IOP_XATTR 0x0008 #define IOP_DEFAULT_READLINK 0x0010 #define IOP_MGTIME 0x0020 +#define IOP_CACHED_LINK 0x0040 /* * Keep mostly read-only and often accessed (especially for @@ -723,7 +726,10 @@ struct inode { }; struct file_lock_context *i_flctx; struct address_space i_data; - struct list_head i_devices; + union { + struct list_head i_devices; + int i_linklen; + }; union { struct pipe_inode_info *i_pipe; struct cdev *i_cdev; @@ -749,6 +755,13 @@ struct inode { void *i_private; /* fs or device private pointer */ } __randomize_layout; +static inline void inode_set_cached_link(struct inode *inode, char *link, int linklen) +{ + inode->i_link = link; + inode->i_linklen = linklen; + inode->i_opflags |= IOP_CACHED_LINK; +} + /* * Get bit address from inode->i_state to use with wait_var_event() * infrastructre. @@ -2127,6 +2140,8 @@ struct file_operations { #define FOP_UNSIGNED_OFFSET ((__force fop_flags_t)(1 << 5)) /* Supports asynchronous lock callbacks */ #define FOP_ASYNC_LOCK ((__force fop_flags_t)(1 << 6)) +/* File system supports uncached read/write buffered IO */ +#define FOP_DONTCACHE ((__force fop_flags_t)(1 << 7)) /* Wrap a directory iterator that needs exclusive inode access */ int wrap_directory_iterator(struct file *, struct dir_context *, @@ -3351,7 +3366,7 @@ extern const struct file_operations generic_ro_fops; #define special_file(m) (S_ISCHR(m)||S_ISBLK(m)||S_ISFIFO(m)||S_ISSOCK(m)) -extern int readlink_copy(char __user *, int, const char *); +extern int readlink_copy(char __user *, int, const char *, int); extern int page_readlink(struct dentry *, char __user *, int); extern const char *page_get_link(struct dentry *, struct inode *, struct delayed_call *); @@ -3468,7 +3483,6 @@ struct offset_ctx { void simple_offset_init(struct offset_ctx *octx); int simple_offset_add(struct offset_ctx *octx, struct dentry *dentry); void simple_offset_remove(struct offset_ctx *octx, struct dentry *dentry); -int simple_offset_empty(struct dentry *dentry); int simple_offset_rename(struct inode *old_dir, struct dentry *old_dentry, struct inode *new_dir, struct dentry *new_dentry); int simple_offset_rename_exchange(struct inode *old_dir, @@ -3614,6 +3628,14 @@ static inline int kiocb_set_rw_flags(struct kiocb *ki, rwf_t flags, if (!(ki->ki_filp->f_mode & FMODE_CAN_ATOMIC_WRITE)) return -EOPNOTSUPP; } + if (flags & RWF_DONTCACHE) { + /* file system must support it */ + if (!(ki->ki_filp->f_op->fop_flags & FOP_DONTCACHE)) + return -EOPNOTSUPP; + /* DAX mappings not supported */ + if (IS_DAX(ki->ki_filp->f_mapping->host)) + return -EOPNOTSUPP; + } kiocb_flags |= (__force int) (flags & RWF_SUPPORTED); if (flags & RWF_SYNC) kiocb_flags |= IOCB_DSYNC; diff --git a/include/linux/fs_parser.h b/include/linux/fs_parser.h index 3cef566088fc..53e566efd5fd 100644 --- a/include/linux/fs_parser.h +++ b/include/linux/fs_parser.h @@ -84,6 +84,8 @@ extern int fs_lookup_param(struct fs_context *fc, extern int lookup_constant(const struct constant_table tbl[], const char *name, int not_found); +extern const struct constant_table bool_names[]; + #ifdef CONFIG_VALIDATE_FS_PARSER extern bool validate_constant_table(const struct constant_table *tbl, size_t tbl_size, int low, int high, int special); diff --git a/include/linux/hrtimer.h b/include/linux/hrtimer.h index 7ef5f7ef31a9..f7bfdcf0dda3 100644 --- a/include/linux/hrtimer.h +++ b/include/linux/hrtimer.h @@ -386,6 +386,7 @@ extern void __init hrtimers_init(void); extern void sysrq_timer_list_show(void); int hrtimers_prepare_cpu(unsigned int cpu); +int hrtimers_cpu_starting(unsigned int cpu); #ifdef CONFIG_HOTPLUG_CPU int hrtimers_cpu_dying(unsigned int cpu); #else diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h index c1645c86eed9..d65b5d71b93b 100644 --- a/include/linux/if_vlan.h +++ b/include/linux/if_vlan.h @@ -585,13 +585,16 @@ static inline int vlan_get_tag(const struct sk_buff *skb, u16 *vlan_tci) * vlan_get_protocol - get protocol EtherType. * @skb: skbuff to query * @type: first vlan protocol + * @mac_offset: MAC offset * @depth: buffer to store length of eth and vlan tags in bytes * * Returns the EtherType of the packet, regardless of whether it is * vlan encapsulated (normal or hardware accelerated) or not. */ -static inline __be16 __vlan_get_protocol(const struct sk_buff *skb, __be16 type, - int *depth) +static inline __be16 __vlan_get_protocol_offset(const struct sk_buff *skb, + __be16 type, + int mac_offset, + int *depth) { unsigned int vlan_depth = skb->mac_len, parse_depth = VLAN_MAX_DEPTH; @@ -610,7 +613,8 @@ static inline __be16 __vlan_get_protocol(const struct sk_buff *skb, __be16 type, do { struct vlan_hdr vhdr, *vh; - vh = skb_header_pointer(skb, vlan_depth, sizeof(vhdr), &vhdr); + vh = skb_header_pointer(skb, mac_offset + vlan_depth, + sizeof(vhdr), &vhdr); if (unlikely(!vh || !--parse_depth)) return 0; @@ -625,6 +629,12 @@ static inline __be16 __vlan_get_protocol(const struct sk_buff *skb, __be16 type, return type; } +static inline __be16 __vlan_get_protocol(const struct sk_buff *skb, __be16 type, + int *depth) +{ + return __vlan_get_protocol_offset(skb, type, 0, depth); +} + /** * vlan_get_protocol - get protocol EtherType. * @skb: skbuff to query diff --git a/include/linux/io_uring/cmd.h b/include/linux/io_uring/cmd.h index 0d5448c0b86c..a3ce553413de 100644 --- a/include/linux/io_uring/cmd.h +++ b/include/linux/io_uring/cmd.h @@ -18,6 +18,11 @@ struct io_uring_cmd { u8 pdu[32]; /* available inline for free use */ }; +struct io_uring_cmd_data { + struct io_uring_sqe sqes[2]; + void *op_data; +}; + static inline const void *io_uring_sqe_cmd(const struct io_uring_sqe *sqe) { return sqe->cmd; @@ -113,4 +118,9 @@ static inline struct task_struct *io_uring_cmd_get_task(struct io_uring_cmd *cmd return cmd_to_io_kiocb(cmd)->tctx->task; } +static inline struct io_uring_cmd_data *io_uring_cmd_get_async_data(struct io_uring_cmd *cmd) +{ + return cmd_to_io_kiocb(cmd)->async_data; +} + #endif /* _LINUX_IO_URING_CMD_H */ diff --git a/include/linux/iomap.h b/include/linux/iomap.h index 5675af6b740c..75bf54e76f3b 100644 --- a/include/linux/iomap.h +++ b/include/linux/iomap.h @@ -335,7 +335,7 @@ struct iomap_ioend { u16 io_type; u16 io_flags; /* IOMAP_F_* */ struct inode *io_inode; /* file being written to */ - size_t io_size; /* size of the extent */ + size_t io_size; /* size of data within eof */ loff_t io_offset; /* offset in the file */ sector_t io_sector; /* start sector of ioend */ struct bio io_bio; /* MUST BE LAST! */ diff --git a/include/linux/lockref.h b/include/linux/lockref.h index c3a1f78bc884..c39f119659ba 100644 --- a/include/linux/lockref.h +++ b/include/linux/lockref.h @@ -34,14 +34,24 @@ struct lockref { }; }; -extern void lockref_get(struct lockref *); -extern int lockref_put_return(struct lockref *); -extern int lockref_get_not_zero(struct lockref *); -extern int lockref_put_not_zero(struct lockref *); -extern int lockref_put_or_lock(struct lockref *); - -extern void lockref_mark_dead(struct lockref *); -extern int lockref_get_not_dead(struct lockref *); +/** + * lockref_init - Initialize a lockref + * @lockref: pointer to lockref structure + * @count: initial count + */ +static inline void lockref_init(struct lockref *lockref, unsigned int count) +{ + spin_lock_init(&lockref->lock); + lockref->count = count; +} + +void lockref_get(struct lockref *lockref); +int lockref_put_return(struct lockref *lockref); +bool lockref_get_not_zero(struct lockref *lockref); +bool lockref_put_or_lock(struct lockref *lockref); + +void lockref_mark_dead(struct lockref *lockref); +bool lockref_get_not_dead(struct lockref *lockref); /* Must be called under spinlock for reliable results */ static inline bool __lockref_is_dead(const struct lockref *l) diff --git a/include/linux/memfd.h b/include/linux/memfd.h index 3f2cf339ceaf..d437e3070850 100644 --- a/include/linux/memfd.h +++ b/include/linux/memfd.h @@ -7,6 +7,7 @@ #ifdef CONFIG_MEMFD_CREATE extern long memfd_fcntl(struct file *file, unsigned int cmd, unsigned int arg); struct folio *memfd_alloc_folio(struct file *memfd, pgoff_t idx); +unsigned int *memfd_file_seals_ptr(struct file *file); #else static inline long memfd_fcntl(struct file *f, unsigned int c, unsigned int a) { @@ -16,6 +17,19 @@ static inline struct folio *memfd_alloc_folio(struct file *memfd, pgoff_t idx) { return ERR_PTR(-EINVAL); } + +static inline unsigned int *memfd_file_seals_ptr(struct file *file) +{ + return NULL; +} #endif +/* Retrieve memfd seals associated with the file, if any. */ +static inline unsigned int memfd_file_seals(struct file *file) +{ + unsigned int *sealsp = memfd_file_seals_ptr(file); + + return sealsp ? *sealsp : 0; +} + #endif /* __LINUX_MEMFD_H */ diff --git a/include/linux/min_heap.h b/include/linux/min_heap.h index e781727c8916..6325f6ffb895 100644 --- a/include/linux/min_heap.h +++ b/include/linux/min_heap.h @@ -15,8 +15,8 @@ */ #define MIN_HEAP_PREALLOCATED(_type, _name, _nr) \ struct _name { \ - int nr; \ - int size; \ + size_t nr; \ + size_t size; \ _type *data; \ _type preallocated[_nr]; \ } diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h index fc7e6153b73d..ea48eb879a0f 100644 --- a/include/linux/mlx5/driver.h +++ b/include/linux/mlx5/driver.h @@ -524,6 +524,7 @@ enum { * creation/deletion on drivers rescan. Unset during device attach. */ MLX5_PRIV_FLAGS_DETACH = 1 << 2, + MLX5_PRIV_FLAGS_SWITCH_LEGACY = 1 << 3, }; struct mlx5_adev { @@ -1202,6 +1203,12 @@ static inline bool mlx5_core_is_vf(const struct mlx5_core_dev *dev) return dev->coredev_type == MLX5_COREDEV_VF; } +static inline bool mlx5_core_same_coredev_type(const struct mlx5_core_dev *dev1, + const struct mlx5_core_dev *dev2) +{ + return dev1->coredev_type == dev2->coredev_type; +} + static inline bool mlx5_core_is_ecpf(const struct mlx5_core_dev *dev) { return dev->caps.embedded_cpu; diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h index 4fbbcf35498b..48d47181c7cd 100644 --- a/include/linux/mlx5/mlx5_ifc.h +++ b/include/linux/mlx5/mlx5_ifc.h @@ -2119,7 +2119,9 @@ struct mlx5_ifc_cmd_hca_cap_2_bits { u8 migration_in_chunks[0x1]; u8 reserved_at_d1[0x1]; u8 sf_eq_usage[0x1]; - u8 reserved_at_d3[0xd]; + u8 reserved_at_d3[0x5]; + u8 multiplane[0x1]; + u8 reserved_at_d9[0x7]; u8 cross_vhca_object_to_object_supported[0x20]; diff --git a/include/linux/mm.h b/include/linux/mm.h index 338a76ce9083..b1c3db9cf355 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -3125,6 +3125,7 @@ static inline bool pagetable_pmd_ctor(struct ptdesc *ptdesc) if (!pmd_ptlock_init(ptdesc)) return false; __folio_set_pgtable(folio); + ptdesc_pmd_pts_init(ptdesc); lruvec_stat_add_folio(folio, NR_PAGETABLE); return true; } @@ -4101,6 +4102,37 @@ void mem_dump_obj(void *object); static inline void mem_dump_obj(void *object) {} #endif +static inline bool is_write_sealed(int seals) +{ + return seals & (F_SEAL_WRITE | F_SEAL_FUTURE_WRITE); +} + +/** + * is_readonly_sealed - Checks whether write-sealed but mapped read-only, + * in which case writes should be disallowing moving + * forwards. + * @seals: the seals to check + * @vm_flags: the VMA flags to check + * + * Returns whether readonly sealed, in which case writess should be disallowed + * going forward. + */ +static inline bool is_readonly_sealed(int seals, vm_flags_t vm_flags) +{ + /* + * Since an F_SEAL_[FUTURE_]WRITE sealed memfd can be mapped as + * MAP_SHARED and read-only, take care to not allow mprotect to + * revert protections on such mappings. Do this only for shared + * mappings. For private mappings, don't need to mask + * VM_MAYWRITE as we still want them to be COW-writable. + */ + if (is_write_sealed(seals) && + ((vm_flags & (VM_SHARED | VM_WRITE)) == VM_SHARED)) + return true; + + return false; +} + /** * seal_check_write - Check for F_SEAL_WRITE or F_SEAL_FUTURE_WRITE flags and * handle them. @@ -4112,24 +4144,15 @@ static inline void mem_dump_obj(void *object) {} */ static inline int seal_check_write(int seals, struct vm_area_struct *vma) { - if (seals & (F_SEAL_WRITE | F_SEAL_FUTURE_WRITE)) { - /* - * New PROT_WRITE and MAP_SHARED mmaps are not allowed when - * write seals are active. - */ - if ((vma->vm_flags & VM_SHARED) && (vma->vm_flags & VM_WRITE)) - return -EPERM; - - /* - * Since an F_SEAL_[FUTURE_]WRITE sealed memfd can be mapped as - * MAP_SHARED and read-only, take care to not allow mprotect to - * revert protections on such mappings. Do this only for shared - * mappings. For private mappings, don't need to mask - * VM_MAYWRITE as we still want them to be COW-writable. - */ - if (vma->vm_flags & VM_SHARED) - vm_flags_clear(vma, VM_MAYWRITE); - } + if (!is_write_sealed(seals)) + return 0; + + /* + * New PROT_WRITE and MAP_SHARED mmaps are not allowed when + * write seals are active. + */ + if ((vma->vm_flags & VM_SHARED) && (vma->vm_flags & VM_WRITE)) + return -EPERM; return 0; } diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h index 7361a8f3ab68..332cee285662 100644 --- a/include/linux/mm_types.h +++ b/include/linux/mm_types.h @@ -445,6 +445,7 @@ FOLIO_MATCH(compound_head, _head_2a); * @pt_index: Used for s390 gmap. * @pt_mm: Used for x86 pgds. * @pt_frag_refcount: For fragmented page table tracking. Powerpc only. + * @pt_share_count: Used for HugeTLB PMD page table share count. * @_pt_pad_2: Padding to ensure proper alignment. * @ptl: Lock for the page table. * @__page_type: Same as page->page_type. Unused for page tables. @@ -471,6 +472,9 @@ struct ptdesc { pgoff_t pt_index; struct mm_struct *pt_mm; atomic_t pt_frag_refcount; +#ifdef CONFIG_HUGETLB_PMD_PAGE_TABLE_SHARING + atomic_t pt_share_count; +#endif }; union { @@ -516,6 +520,32 @@ static_assert(sizeof(struct ptdesc) <= sizeof(struct page)); const struct page *: (const struct ptdesc *)(p), \ struct page *: (struct ptdesc *)(p))) +#ifdef CONFIG_HUGETLB_PMD_PAGE_TABLE_SHARING +static inline void ptdesc_pmd_pts_init(struct ptdesc *ptdesc) +{ + atomic_set(&ptdesc->pt_share_count, 0); +} + +static inline void ptdesc_pmd_pts_inc(struct ptdesc *ptdesc) +{ + atomic_inc(&ptdesc->pt_share_count); +} + +static inline void ptdesc_pmd_pts_dec(struct ptdesc *ptdesc) +{ + atomic_dec(&ptdesc->pt_share_count); +} + +static inline int ptdesc_pmd_pts_count(struct ptdesc *ptdesc) +{ + return atomic_read(&ptdesc->pt_share_count); +} +#else +static inline void ptdesc_pmd_pts_init(struct ptdesc *ptdesc) +{ +} +#endif + /* * Used for sizing the vmemmap region on some architectures */ diff --git a/include/linux/module.h b/include/linux/module.h index 94acbacdcdf1..b3a643435357 100644 --- a/include/linux/module.h +++ b/include/linux/module.h @@ -773,7 +773,8 @@ void *__module_writable_address(struct module *mod, void *loc); static inline void *module_writable_address(struct module *mod, void *loc) { - if (!IS_ENABLED(CONFIG_ARCH_HAS_EXECMEM_ROX) || !mod) + if (!IS_ENABLED(CONFIG_ARCH_HAS_EXECMEM_ROX) || !mod || + mod->state != MODULE_STATE_UNFORMED) return loc; return __module_writable_address(mod, loc); } diff --git a/include/linux/mount.h b/include/linux/mount.h index c34c18b4e8f3..dcc17ce8a959 100644 --- a/include/linux/mount.h +++ b/include/linux/mount.h @@ -50,7 +50,7 @@ struct path; #define MNT_ATIME_MASK (MNT_NOATIME | MNT_NODIRATIME | MNT_RELATIME ) #define MNT_INTERNAL_FLAGS (MNT_SHARED | MNT_WRITE_HOLD | MNT_INTERNAL | \ - MNT_DOOMED | MNT_SYNC_UMOUNT | MNT_MARKED | MNT_ONRB) + MNT_DOOMED | MNT_SYNC_UMOUNT | MNT_MARKED) #define MNT_INTERNAL 0x4000 @@ -64,7 +64,6 @@ struct path; #define MNT_SYNC_UMOUNT 0x2000000 #define MNT_MARKED 0x4000000 #define MNT_UMOUNT 0x8000000 -#define MNT_ONRB 0x10000000 struct vfsmount { struct dentry *mnt_root; /* root of the mounted tree */ @@ -76,7 +75,7 @@ struct vfsmount { static inline struct mnt_idmap *mnt_idmap(const struct vfsmount *mnt) { /* Pairs with smp_store_release() in do_idmap_mount(). */ - return smp_load_acquire(&mnt->mnt_idmap); + return READ_ONCE(mnt->mnt_idmap); } extern int mnt_want_write(struct vfsmount *mnt); @@ -114,7 +113,7 @@ extern struct vfsmount *kern_mount(struct file_system_type *); extern void kern_unmount(struct vfsmount *mnt); extern int may_umount_tree(struct vfsmount *); extern int may_umount(struct vfsmount *); -extern long do_mount(const char *, const char __user *, +int do_mount(const char *, const char __user *, const char *, unsigned long, void *); extern struct vfsmount *collect_mounts(const struct path *); extern void drop_collected_mounts(struct vfsmount *); diff --git a/include/linux/netfs.h b/include/linux/netfs.h index 5eaceef41e6c..071d05d81d38 100644 --- a/include/linux/netfs.h +++ b/include/linux/netfs.h @@ -18,9 +18,11 @@ #include <linux/fs.h> #include <linux/pagemap.h> #include <linux/uio.h> +#include <linux/rolling_buffer.h> enum netfs_sreq_ref_trace; typedef struct mempool_s mempool_t; +struct folio_queue; /** * folio_start_private_2 - Start an fscache write on a folio. [DEPRECATED] @@ -71,6 +73,7 @@ struct netfs_inode { #define NETFS_ICTX_UNBUFFERED 1 /* I/O should not use the pagecache */ #define NETFS_ICTX_WRITETHROUGH 2 /* Write-through caching */ #define NETFS_ICTX_MODIFIED_ATTR 3 /* Indicate change in mtime/ctime */ +#define NETFS_ICTX_SINGLE_NO_UPLOAD 4 /* Monolithic payload, cache but no upload */ }; /* @@ -178,39 +181,35 @@ struct netfs_io_subrequest { unsigned long long start; /* Where to start the I/O */ size_t len; /* Size of the I/O */ size_t transferred; /* Amount of data transferred */ - size_t consumed; /* Amount of read data consumed */ - size_t prev_donated; /* Amount of data donated from previous subreq */ - size_t next_donated; /* Amount of data donated from next subreq */ refcount_t ref; short error; /* 0 or error that occurred */ unsigned short debug_index; /* Index in list (for debugging output) */ unsigned int nr_segs; /* Number of segs in io_iter */ + u8 retry_count; /* The number of retries (0 on initial pass) */ enum netfs_io_source source; /* Where to read from/write to */ unsigned char stream_nr; /* I/O stream this belongs to */ - unsigned char curr_folioq_slot; /* Folio currently being read */ - unsigned char curr_folio_order; /* Order of folio */ - struct folio_queue *curr_folioq; /* Queue segment in which current folio resides */ unsigned long flags; #define NETFS_SREQ_COPY_TO_CACHE 0 /* Set if should copy the data to the cache */ #define NETFS_SREQ_CLEAR_TAIL 1 /* Set if the rest of the read should be cleared */ #define NETFS_SREQ_SEEK_DATA_READ 3 /* Set if ->read() should SEEK_DATA first */ -#define NETFS_SREQ_NO_PROGRESS 4 /* Set if we didn't manage to read any data */ +#define NETFS_SREQ_MADE_PROGRESS 4 /* Set if we transferred at least some data */ #define NETFS_SREQ_ONDEMAND 5 /* Set if it's from on-demand read mode */ #define NETFS_SREQ_BOUNDARY 6 /* Set if ends on hard boundary (eg. ceph object) */ #define NETFS_SREQ_HIT_EOF 7 /* Set if short due to EOF */ #define NETFS_SREQ_IN_PROGRESS 8 /* Unlocked when the subrequest completes */ #define NETFS_SREQ_NEED_RETRY 9 /* Set if the filesystem requests a retry */ -#define NETFS_SREQ_RETRYING 10 /* Set if we're retrying */ -#define NETFS_SREQ_FAILED 11 /* Set if the subreq failed unretryably */ +#define NETFS_SREQ_FAILED 10 /* Set if the subreq failed unretryably */ }; enum netfs_io_origin { NETFS_READAHEAD, /* This read was triggered by readahead */ NETFS_READPAGE, /* This read is a synchronous read */ NETFS_READ_GAPS, /* This read is a synchronous read to fill gaps */ + NETFS_READ_SINGLE, /* This read should be treated as a single object */ NETFS_READ_FOR_WRITE, /* This read is to prepare a write */ NETFS_DIO_READ, /* This is a direct I/O read */ NETFS_WRITEBACK, /* This write was triggered by writepages */ + NETFS_WRITEBACK_SINGLE, /* This monolithic write was triggered by writepages */ NETFS_WRITETHROUGH, /* This write was made by netfs_perform_write() */ NETFS_UNBUFFERED_WRITE, /* This is an unbuffered write */ NETFS_DIO_WRITE, /* This is a direct I/O write */ @@ -231,16 +230,16 @@ struct netfs_io_request { struct address_space *mapping; /* The mapping being accessed */ struct kiocb *iocb; /* AIO completion vector */ struct netfs_cache_resources cache_resources; + struct netfs_io_request *copy_to_cache; /* Request to write just-read data to the cache */ struct readahead_control *ractl; /* Readahead descriptor */ struct list_head proc_link; /* Link in netfs_iorequests */ - struct list_head subrequests; /* Contributory I/O operations */ struct netfs_io_stream io_streams[2]; /* Streams of parallel I/O operations */ #define NR_IO_STREAMS 2 //wreq->nr_io_streams struct netfs_group *group; /* Writeback group being written back */ - struct folio_queue *buffer; /* Head of I/O buffer */ - struct folio_queue *buffer_tail; /* Tail of I/O buffer */ - struct iov_iter iter; /* Unencrypted-side iterator */ - struct iov_iter io_iter; /* I/O (Encrypted-side) iterator */ + struct rolling_buffer buffer; /* Unencrypted buffer */ +#define NETFS_ROLLBUF_PUT_MARK ROLLBUF_MARK_1 +#define NETFS_ROLLBUF_PAGECACHE_MARK ROLLBUF_MARK_2 + wait_queue_head_t waitq; /* Processor waiter */ void *netfs_priv; /* Private data for the netfs */ void *netfs_priv2; /* Private data for the netfs */ struct bio_vec *direct_bv; /* DIO buffer list (when handling iovec-iter) */ @@ -251,29 +250,28 @@ struct netfs_io_request { atomic_t subreq_counter; /* Next subreq->debug_index */ unsigned int nr_group_rel; /* Number of refs to release on ->group */ spinlock_t lock; /* Lock for queuing subreqs */ - atomic_t nr_outstanding; /* Number of ops in progress */ unsigned long long submitted; /* Amount submitted for I/O so far */ unsigned long long len; /* Length of the request */ size_t transferred; /* Amount to be indicated as transferred */ long error; /* 0 or error that occurred */ enum netfs_io_origin origin; /* Origin of the request */ bool direct_bv_unpin; /* T if direct_bv[] must be unpinned */ - u8 buffer_head_slot; /* First slot in ->buffer */ - u8 buffer_tail_slot; /* Next slot in ->buffer_tail */ unsigned long long i_size; /* Size of the file */ unsigned long long start; /* Start position */ atomic64_t issued_to; /* Write issuer folio cursor */ unsigned long long collected_to; /* Point we've collected to */ unsigned long long cleaned_to; /* Position we've cleaned folios to */ + unsigned long long abandon_to; /* Position to abandon folios to */ pgoff_t no_unlock_folio; /* Don't unlock this folio after read */ - size_t prev_donated; /* Fallback for subreq->prev_donated */ + unsigned char front_folio_order; /* Order (size) of front folio */ refcount_t ref; unsigned long flags; -#define NETFS_RREQ_COPY_TO_CACHE 1 /* Need to write to the cache */ +#define NETFS_RREQ_OFFLOAD_COLLECTION 0 /* Offload collection to workqueue */ #define NETFS_RREQ_NO_UNLOCK_FOLIO 2 /* Don't unlock no_unlock_folio on completion */ #define NETFS_RREQ_DONT_UNLOCK_FOLIOS 3 /* Don't unlock the folios on completion */ #define NETFS_RREQ_FAILED 4 /* The request failed */ #define NETFS_RREQ_IN_PROGRESS 5 /* Unlocked when the request completes */ +#define NETFS_RREQ_FOLIO_COPY_TO_CACHE 6 /* Copy current folio to cache from read */ #define NETFS_RREQ_UPLOAD_TO_SERVER 8 /* Need to write to the server */ #define NETFS_RREQ_NONBLOCK 9 /* Don't block if possible (O_NONBLOCK) */ #define NETFS_RREQ_BLOCKED 10 /* We blocked */ @@ -410,6 +408,13 @@ ssize_t netfs_unbuffered_write_iter_locked(struct kiocb *iocb, struct iov_iter * struct netfs_group *netfs_group); ssize_t netfs_file_write_iter(struct kiocb *iocb, struct iov_iter *from); +/* Single, monolithic object read/write API. */ +void netfs_single_mark_inode_dirty(struct inode *inode); +ssize_t netfs_read_single(struct inode *inode, struct file *file, struct iov_iter *iter); +int netfs_writeback_single(struct address_space *mapping, + struct writeback_control *wbc, + struct iov_iter *iter); + /* Address operations API */ struct readahead_control; void netfs_readahead(struct readahead_control *); @@ -429,10 +434,8 @@ bool netfs_release_folio(struct folio *folio, gfp_t gfp); vm_fault_t netfs_page_mkwrite(struct vm_fault *vmf, struct netfs_group *netfs_group); /* (Sub)request management API. */ -void netfs_read_subreq_progress(struct netfs_io_subrequest *subreq, - bool was_async); -void netfs_read_subreq_terminated(struct netfs_io_subrequest *subreq, - int error, bool was_async); +void netfs_read_subreq_progress(struct netfs_io_subrequest *subreq); +void netfs_read_subreq_terminated(struct netfs_io_subrequest *subreq); void netfs_get_subrequest(struct netfs_io_subrequest *subreq, enum netfs_sreq_ref_trace what); void netfs_put_subrequest(struct netfs_io_subrequest *subreq, @@ -454,6 +457,18 @@ void netfs_end_io_write(struct inode *inode); int netfs_start_io_direct(struct inode *inode); void netfs_end_io_direct(struct inode *inode); +/* Miscellaneous APIs. */ +struct folio_queue *netfs_folioq_alloc(unsigned int rreq_id, gfp_t gfp, + unsigned int trace /*enum netfs_folioq_trace*/); +void netfs_folioq_free(struct folio_queue *folioq, + unsigned int trace /*enum netfs_trace_folioq*/); + +/* Buffer wrangling helpers API. */ +int netfs_alloc_folioq_buffer(struct address_space *mapping, + struct folio_queue **_buffer, + size_t *_cur_size, ssize_t size, gfp_t gfp); +void netfs_free_folioq_buffer(struct folio_queue *fq); + /** * netfs_inode - Get the netfs inode context from the inode * @inode: The inode to query diff --git a/include/linux/percpu-defs.h b/include/linux/percpu-defs.h index 35842d1e3879..5b520fe86b60 100644 --- a/include/linux/percpu-defs.h +++ b/include/linux/percpu-defs.h @@ -221,10 +221,7 @@ do { \ } while (0) #define PERCPU_PTR(__p) \ -({ \ - unsigned long __pcpu_ptr = (__force unsigned long)(__p); \ - (typeof(*(__p)) __force __kernel *)(__pcpu_ptr); \ -}) + (typeof(*(__p)) __force __kernel *)((__force unsigned long)(__p)) #ifdef CONFIG_SMP diff --git a/include/linux/pid.h b/include/linux/pid.h index a3aad9b4074c..98837a1ff0f3 100644 --- a/include/linux/pid.h +++ b/include/linux/pid.h @@ -59,6 +59,7 @@ struct pid spinlock_t lock; struct dentry *stashed; u64 ino; + struct rb_node pidfs_node; /* lists of tasks that use this pid */ struct hlist_head tasks[PIDTYPE_MAX]; struct hlist_head inodes; @@ -68,6 +69,7 @@ struct pid struct upid numbers[]; }; +extern seqcount_spinlock_t pidmap_lock_seq; extern struct pid init_struct_pid; struct file; @@ -106,9 +108,6 @@ extern void exchange_tids(struct task_struct *task, struct task_struct *old); extern void transfer_pid(struct task_struct *old, struct task_struct *new, enum pid_type); -extern int pid_max; -extern int pid_max_min, pid_max_max; - /* * look up a PID in the hash table. Must be called with the tasklist_lock * or rcu_read_lock() held. diff --git a/include/linux/pid_namespace.h b/include/linux/pid_namespace.h index f9f9931e02d6..7c67a5811199 100644 --- a/include/linux/pid_namespace.h +++ b/include/linux/pid_namespace.h @@ -30,6 +30,7 @@ struct pid_namespace { struct task_struct *child_reaper; struct kmem_cache *pid_cachep; unsigned int level; + int pid_max; struct pid_namespace *parent; #ifdef CONFIG_BSD_PROCESS_ACCT struct fs_pin *bacct; @@ -38,9 +39,14 @@ struct pid_namespace { struct ucounts *ucounts; int reboot; /* group exit code if this pidns was rebooted */ struct ns_common ns; -#if defined(CONFIG_SYSCTL) && defined(CONFIG_MEMFD_CREATE) + struct work_struct work; +#ifdef CONFIG_SYSCTL + struct ctl_table_set set; + struct ctl_table_header *sysctls; +#if defined(CONFIG_MEMFD_CREATE) int memfd_noexec_scope; #endif +#endif } __randomize_layout; extern struct pid_namespace init_pid_ns; @@ -117,6 +123,8 @@ static inline int reboot_pid_ns(struct pid_namespace *pid_ns, int cmd) extern struct pid_namespace *task_active_pid_ns(struct task_struct *tsk); void pidhash_init(void); void pid_idr_init(void); +int register_pidns_sysctls(struct pid_namespace *pidns); +void unregister_pidns_sysctls(struct pid_namespace *pidns); static inline bool task_is_in_init_pid_ns(struct task_struct *tsk) { diff --git a/include/linux/pidfs.h b/include/linux/pidfs.h index 75bdf9807802..7c830d0dec9a 100644 --- a/include/linux/pidfs.h +++ b/include/linux/pidfs.h @@ -4,5 +4,8 @@ struct file *pidfs_alloc_file(struct pid *pid, unsigned int flags); void __init pidfs_init(void); +void pidfs_add_pid(struct pid *pid); +void pidfs_remove_pid(struct pid *pid); +extern const struct dentry_operations pidfs_dentry_operations; #endif /* _LINUX_PID_FS_H */ diff --git a/include/linux/platform_data/amd_qdma.h b/include/linux/platform_data/amd_qdma.h index 576d952f97ed..967a6ef31cf9 100644 --- a/include/linux/platform_data/amd_qdma.h +++ b/include/linux/platform_data/amd_qdma.h @@ -26,11 +26,13 @@ struct dma_slave_map; * @max_mm_channels: Maximum number of MM DMA channels in each direction * @device_map: DMA slave map * @irq_index: The index of first IRQ + * @dma_dev: The device pointer for dma operations */ struct qdma_platdata { u32 max_mm_channels; u32 irq_index; struct dma_slave_map *device_map; + struct device *dma_dev; }; #endif /* _PLATDATA_AMD_QDMA_H */ diff --git a/include/linux/poll.h b/include/linux/poll.h index d1ea4f3714a8..12bb18e8b978 100644 --- a/include/linux/poll.h +++ b/include/linux/poll.h @@ -25,14 +25,14 @@ struct poll_table_struct; -/* +/* * structures and helpers for f_op->poll implementations */ typedef void (*poll_queue_proc)(struct file *, wait_queue_head_t *, struct poll_table_struct *); /* - * Do not touch the structure directly, use the access functions - * poll_does_not_wait() and poll_requested_events() instead. + * Do not touch the structure directly, use the access function + * poll_requested_events() instead. */ typedef struct poll_table_struct { poll_queue_proc _qproc; @@ -41,18 +41,16 @@ typedef struct poll_table_struct { static inline void poll_wait(struct file * filp, wait_queue_head_t * wait_address, poll_table *p) { - if (p && p->_qproc && wait_address) + if (p && p->_qproc) { p->_qproc(filp, wait_address, p); -} - -/* - * Return true if it is guaranteed that poll will not wait. This is the case - * if the poll() of another file descriptor in the set got an event, so there - * is no need for waiting. - */ -static inline bool poll_does_not_wait(const poll_table *p) -{ - return p == NULL || p->_qproc == NULL; + /* + * This memory barrier is paired in the wq_has_sleeper(). + * See the comment above prepare_to_wait(), we need to + * ensure that subsequent tests in this thread can't be + * reordered with __add_wait_queue() in _qproc() paths. + */ + smp_mb(); + } } /* diff --git a/include/linux/pruss_driver.h b/include/linux/pruss_driver.h index c9a31c567e85..2e18fef1a2e1 100644 --- a/include/linux/pruss_driver.h +++ b/include/linux/pruss_driver.h @@ -144,32 +144,32 @@ static inline int pruss_release_mem_region(struct pruss *pruss, static inline int pruss_cfg_get_gpmux(struct pruss *pruss, enum pruss_pru_id pru_id, u8 *mux) { - return ERR_PTR(-EOPNOTSUPP); + return -EOPNOTSUPP; } static inline int pruss_cfg_set_gpmux(struct pruss *pruss, enum pruss_pru_id pru_id, u8 mux) { - return ERR_PTR(-EOPNOTSUPP); + return -EOPNOTSUPP; } static inline int pruss_cfg_gpimode(struct pruss *pruss, enum pruss_pru_id pru_id, enum pruss_gpi_mode mode) { - return ERR_PTR(-EOPNOTSUPP); + return -EOPNOTSUPP; } static inline int pruss_cfg_miirt_enable(struct pruss *pruss, bool enable) { - return ERR_PTR(-EOPNOTSUPP); + return -EOPNOTSUPP; } static inline int pruss_cfg_xfr_enable(struct pruss *pruss, enum pru_type pru_type, - bool enable); + bool enable) { - return ERR_PTR(-EOPNOTSUPP); + return -EOPNOTSUPP; } #endif /* CONFIG_TI_PRUSS */ diff --git a/include/linux/pseudo_fs.h b/include/linux/pseudo_fs.h index 730f77381d55..2503f7625d65 100644 --- a/include/linux/pseudo_fs.h +++ b/include/linux/pseudo_fs.h @@ -5,6 +5,7 @@ struct pseudo_fs_context { const struct super_operations *ops; + const struct export_operations *eops; const struct xattr_handler * const *xattr; const struct dentry_operations *dops; unsigned long magic; diff --git a/include/linux/rbtree.h b/include/linux/rbtree.h index 7c173aa64e1e..8d2ba3749866 100644 --- a/include/linux/rbtree.h +++ b/include/linux/rbtree.h @@ -211,6 +211,43 @@ rb_add(struct rb_node *node, struct rb_root *tree, } /** + * rb_find_add_cached() - find equivalent @node in @tree, or add @node + * @node: node to look-for / insert + * @tree: tree to search / modify + * @cmp: operator defining the node order + * + * Returns the rb_node matching @node, or NULL when no match is found and @node + * is inserted. + */ +static __always_inline struct rb_node * +rb_find_add_cached(struct rb_node *node, struct rb_root_cached *tree, + int (*cmp)(const struct rb_node *new, const struct rb_node *exist)) +{ + bool leftmost = true; + struct rb_node **link = &tree->rb_root.rb_node; + struct rb_node *parent = NULL; + int c; + + while (*link) { + parent = *link; + c = cmp(node, parent); + + if (c < 0) { + link = &parent->rb_left; + } else if (c > 0) { + link = &parent->rb_right; + leftmost = false; + } else { + return parent; + } + } + + rb_link_node(node, parent, link); + rb_insert_color_cached(node, tree, leftmost); + return NULL; +} + +/** * rb_find_add() - find equivalent @node in @tree, or add @node * @node: node to look-for / insert * @tree: tree to search / modify diff --git a/include/linux/rculist.h b/include/linux/rculist.h index 14dfa6008467..1b11926ddd47 100644 --- a/include/linux/rculist.h +++ b/include/linux/rculist.h @@ -30,6 +30,17 @@ static inline void INIT_LIST_HEAD_RCU(struct list_head *list) * way, we must not access it directly */ #define list_next_rcu(list) (*((struct list_head __rcu **)(&(list)->next))) +/* + * Return the ->prev pointer of a list_head in an rcu safe way. Don't + * access it directly. + * + * Any list traversed with list_bidir_prev_rcu() must never use + * list_del_rcu(). Doing so will poison the ->prev pointer that + * list_bidir_prev_rcu() relies on, which will result in segfaults. + * To prevent these segfaults, use list_bidir_del_rcu() instead + * of list_del_rcu(). + */ +#define list_bidir_prev_rcu(list) (*((struct list_head __rcu **)(&(list)->prev))) /** * list_tail_rcu - returns the prev pointer of the head of the list @@ -159,6 +170,39 @@ static inline void list_del_rcu(struct list_head *entry) } /** + * list_bidir_del_rcu - deletes entry from list without re-initialization + * @entry: the element to delete from the list. + * + * In contrast to list_del_rcu() doesn't poison the prev pointer thus + * allowing backwards traversal via list_bidir_prev_rcu(). + * + * Note: list_empty() on entry does not return true after this because + * the entry is in a special undefined state that permits RCU-based + * lockfree reverse traversal. In particular this means that we can not + * poison the forward and backwards pointers that may still be used for + * walking the list. + * + * The caller must take whatever precautions are necessary (such as + * holding appropriate locks) to avoid racing with another list-mutation + * primitive, such as list_bidir_del_rcu() or list_add_rcu(), running on + * this same list. However, it is perfectly legal to run concurrently + * with the _rcu list-traversal primitives, such as + * list_for_each_entry_rcu(). + * + * Note that list_del_rcu() and list_bidir_del_rcu() must not be used on + * the same list. + * + * Note that the caller is not permitted to immediately free + * the newly deleted entry. Instead, either synchronize_rcu() + * or call_rcu() must be used to defer freeing until an RCU + * grace period has elapsed. + */ +static inline void list_bidir_del_rcu(struct list_head *entry) +{ + __list_del_entry(entry); +} + +/** * hlist_del_init_rcu - deletes entry from hash list with re-initialization * @n: the element to delete from the hash list. * diff --git a/include/linux/regulator/consumer.h b/include/linux/regulator/consumer.h index 8c3c372ad735..bcba3935c6f9 100644 --- a/include/linux/regulator/consumer.h +++ b/include/linux/regulator/consumer.h @@ -168,29 +168,6 @@ int devm_regulator_get_enable_read_voltage(struct device *dev, const char *id); void regulator_put(struct regulator *regulator); void devm_regulator_put(struct regulator *regulator); -#if IS_ENABLED(CONFIG_OF) -struct regulator *__must_check of_regulator_get_optional(struct device *dev, - struct device_node *node, - const char *id); -struct regulator *__must_check devm_of_regulator_get_optional(struct device *dev, - struct device_node *node, - const char *id); -#else -static inline struct regulator *__must_check of_regulator_get_optional(struct device *dev, - struct device_node *node, - const char *id) -{ - return ERR_PTR(-ENODEV); -} - -static inline struct regulator *__must_check devm_of_regulator_get_optional(struct device *dev, - struct device_node *node, - const char *id) -{ - return ERR_PTR(-ENODEV); -} -#endif - int regulator_register_supply_alias(struct device *dev, const char *id, struct device *alias_dev, const char *alias_id); @@ -223,8 +200,6 @@ int regulator_disable_deferred(struct regulator *regulator, int ms); int __must_check regulator_bulk_get(struct device *dev, int num_consumers, struct regulator_bulk_data *consumers); -int __must_check of_regulator_bulk_get_all(struct device *dev, struct device_node *np, - struct regulator_bulk_data **consumers); int __must_check devm_regulator_bulk_get(struct device *dev, int num_consumers, struct regulator_bulk_data *consumers); void devm_regulator_bulk_put(struct regulator_bulk_data *consumers); @@ -373,20 +348,6 @@ devm_regulator_get_optional(struct device *dev, const char *id) return ERR_PTR(-ENODEV); } -static inline struct regulator *__must_check of_regulator_get_optional(struct device *dev, - struct device_node *node, - const char *id) -{ - return ERR_PTR(-ENODEV); -} - -static inline struct regulator *__must_check devm_of_regulator_get_optional(struct device *dev, - struct device_node *node, - const char *id) -{ - return ERR_PTR(-ENODEV); -} - static inline void regulator_put(struct regulator *regulator) { } @@ -483,12 +444,6 @@ static inline int devm_regulator_bulk_get(struct device *dev, int num_consumers, return 0; } -static inline int of_regulator_bulk_get_all(struct device *dev, struct device_node *np, - struct regulator_bulk_data **consumers) -{ - return 0; -} - static inline int devm_regulator_bulk_get_const( struct device *dev, int num_consumers, const struct regulator_bulk_data *in_consumers, @@ -700,6 +655,38 @@ regulator_is_equal(struct regulator *reg1, struct regulator *reg2) } #endif +#if IS_ENABLED(CONFIG_OF) && IS_ENABLED(CONFIG_REGULATOR) +struct regulator *__must_check of_regulator_get_optional(struct device *dev, + struct device_node *node, + const char *id); +struct regulator *__must_check devm_of_regulator_get_optional(struct device *dev, + struct device_node *node, + const char *id); +int __must_check of_regulator_bulk_get_all(struct device *dev, struct device_node *np, + struct regulator_bulk_data **consumers); +#else +static inline struct regulator *__must_check of_regulator_get_optional(struct device *dev, + struct device_node *node, + const char *id) +{ + return ERR_PTR(-ENODEV); +} + +static inline struct regulator *__must_check devm_of_regulator_get_optional(struct device *dev, + struct device_node *node, + const char *id) +{ + return ERR_PTR(-ENODEV); +} + +static inline int of_regulator_bulk_get_all(struct device *dev, struct device_node *np, + struct regulator_bulk_data **consumers) +{ + return 0; +} + +#endif + static inline int regulator_set_voltage_triplet(struct regulator *regulator, int min_uV, int target_uV, int max_uV) diff --git a/include/linux/rolling_buffer.h b/include/linux/rolling_buffer.h new file mode 100644 index 000000000000..ac15b1ffdd83 --- /dev/null +++ b/include/linux/rolling_buffer.h @@ -0,0 +1,61 @@ +/* SPDX-License-Identifier: GPL-2.0-or-later */ +/* Rolling buffer of folios + * + * Copyright (C) 2024 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + */ + +#ifndef _ROLLING_BUFFER_H +#define _ROLLING_BUFFER_H + +#include <linux/folio_queue.h> +#include <linux/uio.h> + +/* + * Rolling buffer. Whilst the buffer is live and in use, folios and folio + * queue segments can be added to one end by one thread and removed from the + * other end by another thread. The buffer isn't allowed to be empty; it must + * always have at least one folio_queue in it so that neither side has to + * modify both queue pointers. + * + * The iterator in the buffer is extended as buffers are inserted. It can be + * snapshotted to use a segment of the buffer. + */ +struct rolling_buffer { + struct folio_queue *head; /* Producer's insertion point */ + struct folio_queue *tail; /* Consumer's removal point */ + struct iov_iter iter; /* Iterator tracking what's left in the buffer */ + u8 next_head_slot; /* Next slot in ->head */ + u8 first_tail_slot; /* First slot in ->tail */ +}; + +/* + * Snapshot of a rolling buffer. + */ +struct rolling_buffer_snapshot { + struct folio_queue *curr_folioq; /* Queue segment in which current folio resides */ + unsigned char curr_slot; /* Folio currently being read */ + unsigned char curr_order; /* Order of folio */ +}; + +/* Marks to store per-folio in the internal folio_queue structs. */ +#define ROLLBUF_MARK_1 BIT(0) +#define ROLLBUF_MARK_2 BIT(1) + +int rolling_buffer_init(struct rolling_buffer *roll, unsigned int rreq_id, + unsigned int direction); +int rolling_buffer_make_space(struct rolling_buffer *roll); +ssize_t rolling_buffer_load_from_ra(struct rolling_buffer *roll, + struct readahead_control *ractl, + struct folio_batch *put_batch); +ssize_t rolling_buffer_append(struct rolling_buffer *roll, struct folio *folio, + unsigned int flags); +struct folio_queue *rolling_buffer_delete_spent(struct rolling_buffer *roll); +void rolling_buffer_clear(struct rolling_buffer *roll); + +static inline void rolling_buffer_advance(struct rolling_buffer *roll, size_t amount) +{ + iov_iter_advance(&roll->iter, amount); +} + +#endif /* _ROLLING_BUFFER_H */ diff --git a/include/linux/sched.h b/include/linux/sched.h index 66b311fbd5d6..3c7eb16ab1d5 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1637,8 +1637,9 @@ static inline unsigned int __task_state_index(unsigned int tsk_state, * We're lying here, but rather than expose a completely new task state * to userspace, we can make this appear as if the task has gone through * a regular rt_mutex_lock() call. + * Report frozen tasks as uninterruptible. */ - if (tsk_state & TASK_RTLOCK_WAIT) + if ((tsk_state & TASK_RTLOCK_WAIT) || (tsk_state & TASK_FROZEN)) state = TASK_UNINTERRUPTIBLE; return fls(state); @@ -1943,11 +1944,10 @@ static inline void kick_process(struct task_struct *tsk) { } #endif extern void __set_task_comm(struct task_struct *tsk, const char *from, bool exec); - -static inline void set_task_comm(struct task_struct *tsk, const char *from) -{ - __set_task_comm(tsk, from, false); -} +#define set_task_comm(tsk, from) ({ \ + BUILD_BUG_ON(sizeof(from) != TASK_COMM_LEN); \ + __set_task_comm(tsk, from, false); \ +}) /* * - Why not use task_lock()? diff --git a/include/linux/seccomp.h b/include/linux/seccomp.h index 341980599c71..e45531455d3b 100644 --- a/include/linux/seccomp.h +++ b/include/linux/seccomp.h @@ -55,10 +55,10 @@ struct seccomp_data; #ifdef CONFIG_HAVE_ARCH_SECCOMP_FILTER static inline int secure_computing(void) { return 0; } -static inline int __secure_computing(const struct seccomp_data *sd) { return 0; } #else static inline void secure_computing_strict(int this_syscall) { return; } #endif +static inline int __secure_computing(const struct seccomp_data *sd) { return 0; } static inline long prctl_get_seccomp(void) { diff --git a/include/linux/seqlock.h b/include/linux/seqlock.h index 5298765d6ca4..eb20dcaa51b5 100644 --- a/include/linux/seqlock.h +++ b/include/linux/seqlock.h @@ -272,7 +272,7 @@ SEQCOUNT_LOCKNAME(mutex, struct mutex, true, mutex) ({ \ unsigned __seq; \ \ - while ((__seq = seqprop_sequence(s)) & 1) \ + while (unlikely((__seq = seqprop_sequence(s)) & 1)) \ cpu_relax(); \ \ kcsan_atomic_next(KCSAN_SEQLOCK_REGION_MAX); \ diff --git a/include/linux/stat.h b/include/linux/stat.h index 3d900c86981c..9d8382e23a9c 100644 --- a/include/linux/stat.h +++ b/include/linux/stat.h @@ -52,6 +52,7 @@ struct kstat { u64 mnt_id; u32 dio_mem_align; u32 dio_offset_align; + u32 dio_read_offset_align; u64 change_cookie; u64 subvol; u32 atomic_write_unit_min; diff --git a/include/linux/trace_events.h b/include/linux/trace_events.h index 91b8ffbdfa8c..58ad4ead33fc 100644 --- a/include/linux/trace_events.h +++ b/include/linux/trace_events.h @@ -364,7 +364,7 @@ struct trace_event_call { struct list_head list; struct trace_event_class *class; union { - char *name; + const char *name; /* Set TRACE_EVENT_FL_TRACEPOINT flag when using "tp" */ struct tracepoint *tp; }; diff --git a/include/linux/userfaultfd_k.h b/include/linux/userfaultfd_k.h index cb40f1a1d081..75342022d144 100644 --- a/include/linux/userfaultfd_k.h +++ b/include/linux/userfaultfd_k.h @@ -247,6 +247,13 @@ static inline bool vma_can_userfault(struct vm_area_struct *vma, vma_is_shmem(vma); } +static inline bool vma_has_uffd_without_event_remap(struct vm_area_struct *vma) +{ + struct userfaultfd_ctx *uffd_ctx = vma->vm_userfaultfd_ctx.ctx; + + return uffd_ctx && (uffd_ctx->features & UFFD_FEATURE_EVENT_REMAP) == 0; +} + extern int dup_userfaultfd(struct vm_area_struct *, struct list_head *); extern void dup_userfaultfd_complete(struct list_head *); void dup_userfaultfd_fail(struct list_head *); @@ -402,6 +409,11 @@ static inline bool userfaultfd_wp_async(struct vm_area_struct *vma) return false; } +static inline bool vma_has_uffd_without_event_remap(struct vm_area_struct *vma) +{ + return false; +} + #endif /* CONFIG_USERFAULTFD */ static inline bool userfaultfd_wp_use_markers(struct vm_area_struct *vma) diff --git a/include/linux/vermagic.h b/include/linux/vermagic.h index a54046bf37e5..939ceabcaf06 100644 --- a/include/linux/vermagic.h +++ b/include/linux/vermagic.h @@ -15,10 +15,10 @@ #else #define MODULE_VERMAGIC_SMP "" #endif -#ifdef CONFIG_PREEMPT_BUILD -#define MODULE_VERMAGIC_PREEMPT "preempt " -#elif defined(CONFIG_PREEMPT_RT) +#ifdef CONFIG_PREEMPT_RT #define MODULE_VERMAGIC_PREEMPT "preempt_rt " +#elif defined(CONFIG_PREEMPT_BUILD) +#define MODULE_VERMAGIC_PREEMPT "preempt " #else #define MODULE_VERMAGIC_PREEMPT "" #endif diff --git a/include/net/busy_poll.h b/include/net/busy_poll.h index c858270141bc..c39a426ebf52 100644 --- a/include/net/busy_poll.h +++ b/include/net/busy_poll.h @@ -174,12 +174,4 @@ static inline void sk_mark_napi_id_once(struct sock *sk, #endif } -static inline void sk_mark_napi_id_once_xdp(struct sock *sk, - const struct xdp_buff *xdp) -{ -#ifdef CONFIG_NET_RX_BUSY_POLL - __sk_mark_napi_id_once(sk, xdp->rxq->napi_id); -#endif -} - #endif /* _LINUX_NET_BUSY_POLL_H */ diff --git a/include/net/inet_connection_sock.h b/include/net/inet_connection_sock.h index 3c82fad904d4..c7f42844c79a 100644 --- a/include/net/inet_connection_sock.h +++ b/include/net/inet_connection_sock.h @@ -282,7 +282,7 @@ static inline int inet_csk_reqsk_queue_len(const struct sock *sk) static inline int inet_csk_reqsk_queue_is_full(const struct sock *sk) { - return inet_csk_reqsk_queue_len(sk) >= READ_ONCE(sk->sk_max_ack_backlog); + return inet_csk_reqsk_queue_len(sk) > READ_ONCE(sk->sk_max_ack_backlog); } bool inet_csk_reqsk_queue_drop(struct sock *sk, struct request_sock *req); diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h index 4afa64c81304..0027beca5cd5 100644 --- a/include/net/netfilter/nf_tables.h +++ b/include/net/netfilter/nf_tables.h @@ -733,15 +733,18 @@ struct nft_set_ext_tmpl { /** * struct nft_set_ext - set extensions * - * @genmask: generation mask + * @genmask: generation mask, but also flags (see NFT_SET_ELEM_DEAD_BIT) * @offset: offsets of individual extension types * @data: beginning of extension data + * + * This structure must be aligned to word size, otherwise atomic bitops + * on genmask field can cause alignment failure on some archs. */ struct nft_set_ext { u8 genmask; u8 offset[NFT_SET_EXT_NUM]; char data[]; -}; +} __aligned(BITS_PER_LONG / 8); static inline void nft_set_ext_prepare(struct nft_set_ext_tmpl *tmpl) { diff --git a/include/net/page_pool/helpers.h b/include/net/page_pool/helpers.h index 793e6fd78bc5..60a5347922be 100644 --- a/include/net/page_pool/helpers.h +++ b/include/net/page_pool/helpers.h @@ -294,7 +294,7 @@ static inline long page_pool_unref_page(struct page *page, long nr) static inline void page_pool_ref_netmem(netmem_ref netmem) { - atomic_long_inc(&netmem_to_page(netmem)->pp_ref_count); + atomic_long_inc(netmem_get_pp_ref_count_ref(netmem)); } static inline void page_pool_ref_page(struct page *page) diff --git a/include/net/sock.h b/include/net/sock.h index c383126f691d..691ca7695d1d 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -2297,7 +2297,7 @@ static inline bool skwq_has_sleeper(struct socket_wq *wq) } /** - * sock_poll_wait - place memory barrier behind the poll_wait call. + * sock_poll_wait - wrapper for the poll_wait call. * @filp: file * @sock: socket to wait on * @p: poll_table @@ -2307,15 +2307,12 @@ static inline bool skwq_has_sleeper(struct socket_wq *wq) static inline void sock_poll_wait(struct file *filp, struct socket *sock, poll_table *p) { - if (!poll_does_not_wait(p)) { - poll_wait(filp, &sock->wq.wait, p); - /* We need to be sure we are in sync with the - * socket flags modification. - * - * This memory barrier is paired in the wq_has_sleeper. - */ - smp_mb(); - } + /* Provides a barrier we need to be sure we are in sync + * with the socket flags modification. + * + * This memory barrier is paired in the wq_has_sleeper. + */ + poll_wait(filp, &sock->wq.wait, p); } static inline void skb_set_hash_from_sk(struct sk_buff *skb, struct sock *sk) diff --git a/include/net/xdp.h b/include/net/xdp.h index e6770dd40c91..b5b10f2b88e5 100644 --- a/include/net/xdp.h +++ b/include/net/xdp.h @@ -62,7 +62,6 @@ struct xdp_rxq_info { u32 queue_index; u32 reg_state; struct xdp_mem_info mem; - unsigned int napi_id; u32 frag_size; } ____cacheline_aligned; /* perf critical, avoid false-sharing */ diff --git a/include/net/xdp_sock_drv.h b/include/net/xdp_sock_drv.h index 40085afd9160..7a7316d9c0da 100644 --- a/include/net/xdp_sock_drv.h +++ b/include/net/xdp_sock_drv.h @@ -59,15 +59,6 @@ static inline void xsk_pool_fill_cb(struct xsk_buff_pool *pool, xp_fill_cb(pool, desc); } -static inline unsigned int xsk_pool_get_napi_id(struct xsk_buff_pool *pool) -{ -#ifdef CONFIG_NET_RX_BUSY_POLL - return pool->heads[0].xdp.rxq->napi_id; -#else - return 0; -#endif -} - static inline void xsk_pool_dma_unmap(struct xsk_buff_pool *pool, unsigned long attrs) { @@ -306,11 +297,6 @@ static inline void xsk_pool_fill_cb(struct xsk_buff_pool *pool, { } -static inline unsigned int xsk_pool_get_napi_id(struct xsk_buff_pool *pool) -{ - return 0; -} - static inline void xsk_pool_dma_unmap(struct xsk_buff_pool *pool, unsigned long attrs) { diff --git a/include/trace/events/afs.h b/include/trace/events/afs.h index a0aed1a428a1..b0db89058c91 100644 --- a/include/trace/events/afs.h +++ b/include/trace/events/afs.h @@ -118,6 +118,8 @@ enum yfs_cm_operation { */ #define afs_call_traces \ EM(afs_call_trace_alloc, "ALLOC") \ + EM(afs_call_trace_async_abort, "ASYAB") \ + EM(afs_call_trace_async_kill, "ASYKL") \ EM(afs_call_trace_free, "FREE ") \ EM(afs_call_trace_get, "GET ") \ EM(afs_call_trace_put, "PUT ") \ @@ -168,12 +170,14 @@ enum yfs_cm_operation { #define afs_cell_traces \ EM(afs_cell_trace_alloc, "ALLOC ") \ EM(afs_cell_trace_free, "FREE ") \ + EM(afs_cell_trace_get_atcell, "GET atcell") \ EM(afs_cell_trace_get_queue_dns, "GET q-dns ") \ EM(afs_cell_trace_get_queue_manage, "GET q-mng ") \ EM(afs_cell_trace_get_queue_new, "GET q-new ") \ EM(afs_cell_trace_get_vol, "GET vol ") \ EM(afs_cell_trace_insert, "INSERT ") \ EM(afs_cell_trace_manage, "MANAGE ") \ + EM(afs_cell_trace_put_atcell, "PUT atcell") \ EM(afs_cell_trace_put_candidate, "PUT candid") \ EM(afs_cell_trace_put_destroy, "PUT destry") \ EM(afs_cell_trace_put_queue_work, "PUT q-work") \ @@ -323,6 +327,44 @@ enum yfs_cm_operation { EM(yfs_CB_TellMeAboutYourself, "YFSCB.TellMeAboutYourself") \ E_(yfs_CB_CallBack, "YFSCB.CallBack") +#define afs_cb_promise_traces \ + EM(afs_cb_promise_clear_cb_break, "CLEAR cb-break") \ + EM(afs_cb_promise_clear_rmdir, "CLEAR rmdir") \ + EM(afs_cb_promise_clear_rotate_server, "CLEAR rot-srv") \ + EM(afs_cb_promise_clear_server_change, "CLEAR srv-chg") \ + EM(afs_cb_promise_clear_vol_init_cb, "CLEAR vol-init-cb") \ + EM(afs_cb_promise_set_apply_cb, "SET apply-cb") \ + EM(afs_cb_promise_set_new_inode, "SET new-inode") \ + E_(afs_cb_promise_set_new_symlink, "SET new-symlink") + +#define afs_vnode_invalid_traces \ + EM(afs_vnode_invalid_trace_cb_ro_snapshot, "cb-ro-snapshot") \ + EM(afs_vnode_invalid_trace_cb_scrub, "cb-scrub") \ + EM(afs_vnode_invalid_trace_cb_v_break, "cb-v-break") \ + EM(afs_vnode_invalid_trace_expired, "expired") \ + EM(afs_vnode_invalid_trace_no_cb_promise, "no-cb-promise") \ + EM(afs_vnode_invalid_trace_vol_expired, "vol-expired") \ + EM(afs_vnode_invalid_trace_zap_data, "zap-data") \ + E_(afs_vnode_valid_trace, "valid") + +#define afs_dir_invalid_traces \ + EM(afs_dir_invalid_edit_add_bad_size, "edit-add-bad-size") \ + EM(afs_dir_invalid_edit_add_no_slots, "edit-add-no-slots") \ + EM(afs_dir_invalid_edit_add_too_many_blocks, "edit-add-too-many-blocks") \ + EM(afs_dir_invalid_edit_get_block, "edit-get-block") \ + EM(afs_dir_invalid_edit_mkdir, "edit-mkdir") \ + EM(afs_dir_invalid_edit_rem_bad_size, "edit-rem-bad-size") \ + EM(afs_dir_invalid_edit_rem_wrong_name, "edit-rem-wrong_name") \ + EM(afs_dir_invalid_edit_upd_bad_size, "edit-upd-bad-size") \ + EM(afs_dir_invalid_edit_upd_no_dd, "edit-upd-no-dotdot") \ + EM(afs_dir_invalid_dv_mismatch, "dv-mismatch") \ + EM(afs_dir_invalid_inval_folio, "inv-folio") \ + EM(afs_dir_invalid_iter_stale, "iter-stale") \ + EM(afs_dir_invalid_reclaimed_folio, "reclaimed-folio") \ + EM(afs_dir_invalid_release_folio, "rel-folio") \ + EM(afs_dir_invalid_remote, "remote") \ + E_(afs_dir_invalid_subdir_removed, "subdir-removed") + #define afs_edit_dir_ops \ EM(afs_edit_dir_create, "create") \ EM(afs_edit_dir_create_error, "c_fail") \ @@ -332,6 +374,7 @@ enum yfs_cm_operation { EM(afs_edit_dir_delete_error, "d_err ") \ EM(afs_edit_dir_delete_inval, "d_invl") \ EM(afs_edit_dir_delete_noent, "d_nent") \ + EM(afs_edit_dir_mkdir, "mk_ent") \ EM(afs_edit_dir_update_dd, "u_ddot") \ EM(afs_edit_dir_update_error, "u_fail") \ EM(afs_edit_dir_update_inval, "u_invl") \ @@ -385,6 +428,7 @@ enum yfs_cm_operation { EM(afs_file_error_dir_over_end, "DIR_ENT_OVER_END") \ EM(afs_file_error_dir_small, "DIR_SMALL") \ EM(afs_file_error_dir_unmarked_ext, "DIR_UNMARKED_EXT") \ + EM(afs_file_error_symlink_big, "SYM_BIG") \ EM(afs_file_error_mntpt, "MNTPT_READ_FAILED") \ E_(afs_file_error_writeback_fail, "WRITEBACK_FAILED") @@ -487,7 +531,9 @@ enum yfs_cm_operation { enum afs_alist_trace { afs_alist_traces } __mode(byte); enum afs_call_trace { afs_call_traces } __mode(byte); enum afs_cb_break_reason { afs_cb_break_reasons } __mode(byte); +enum afs_cb_promise_trace { afs_cb_promise_traces } __mode(byte); enum afs_cell_trace { afs_cell_traces } __mode(byte); +enum afs_dir_invalid_trace { afs_dir_invalid_traces} __mode(byte); enum afs_edit_dir_op { afs_edit_dir_ops } __mode(byte); enum afs_edit_dir_reason { afs_edit_dir_reasons } __mode(byte); enum afs_eproto_cause { afs_eproto_causes } __mode(byte); @@ -498,6 +544,7 @@ enum afs_flock_operation { afs_flock_operations } __mode(byte); enum afs_io_error { afs_io_errors } __mode(byte); enum afs_rotate_trace { afs_rotate_traces } __mode(byte); enum afs_server_trace { afs_server_traces } __mode(byte); +enum afs_vnode_invalid_trace { afs_vnode_invalid_traces} __mode(byte); enum afs_volume_trace { afs_volume_traces } __mode(byte); #endif /* end __AFS_GENERATE_TRACE_ENUMS_ONCE_ONLY */ @@ -513,8 +560,10 @@ enum afs_volume_trace { afs_volume_traces } __mode(byte); afs_alist_traces; afs_call_traces; afs_cb_break_reasons; +afs_cb_promise_traces; afs_cell_traces; afs_cm_operations; +afs_dir_invalid_traces; afs_edit_dir_ops; afs_edit_dir_reasons; afs_eproto_causes; @@ -526,6 +575,7 @@ afs_fs_operations; afs_io_errors; afs_rotate_traces; afs_server_traces; +afs_vnode_invalid_traces; afs_vl_operations; yfs_cm_operations; @@ -670,7 +720,7 @@ TRACE_EVENT(afs_make_fs_call, } ), - TP_printk("c=%08x %06llx:%06llx:%06x %s", + TP_printk("c=%08x V=%llx i=%llx:%x %s", __entry->call, __entry->fid.vid, __entry->fid.vnode, @@ -704,7 +754,7 @@ TRACE_EVENT(afs_make_fs_calli, } ), - TP_printk("c=%08x %06llx:%06llx:%06x %s i=%u", + TP_printk("c=%08x V=%llx i=%llx:%x %s i=%u", __entry->call, __entry->fid.vid, __entry->fid.vnode, @@ -741,7 +791,7 @@ TRACE_EVENT(afs_make_fs_call1, __entry->name[__len] = 0; ), - TP_printk("c=%08x %06llx:%06llx:%06x %s \"%s\"", + TP_printk("c=%08x V=%llx i=%llx:%x %s \"%s\"", __entry->call, __entry->fid.vid, __entry->fid.vnode, @@ -782,7 +832,7 @@ TRACE_EVENT(afs_make_fs_call2, __entry->name2[__len2] = 0; ), - TP_printk("c=%08x %06llx:%06llx:%06x %s \"%s\" \"%s\"", + TP_printk("c=%08x V=%llx i=%llx:%x %s \"%s\" \"%s\"", __entry->call, __entry->fid.vid, __entry->fid.vnode, @@ -887,9 +937,9 @@ TRACE_EVENT(afs_sent_data, ); TRACE_EVENT(afs_dir_check_failed, - TP_PROTO(struct afs_vnode *vnode, loff_t off, loff_t i_size), + TP_PROTO(struct afs_vnode *vnode, loff_t off), - TP_ARGS(vnode, off, i_size), + TP_ARGS(vnode, off), TP_STRUCT__entry( __field(struct afs_vnode *, vnode) @@ -900,7 +950,7 @@ TRACE_EVENT(afs_dir_check_failed, TP_fast_assign( __entry->vnode = vnode; __entry->off = off; - __entry->i_size = i_size; + __entry->i_size = i_size_read(&vnode->netfs.inode); ), TP_printk("vn=%p %llx/%llx", @@ -1002,7 +1052,7 @@ TRACE_EVENT(afs_edit_dir, __entry->name[__len] = 0; ), - TP_printk("d=%x:%x %s %s %u[%u] f=%x:%x \"%s\"", + TP_printk("di=%x:%x %s %s %u[%u] fi=%x:%x \"%s\"", __entry->vnode, __entry->unique, __print_symbolic(__entry->why, afs_edit_dir_reasons), __print_symbolic(__entry->op, afs_edit_dir_ops), @@ -1011,6 +1061,122 @@ TRACE_EVENT(afs_edit_dir, __entry->name) ); +TRACE_EVENT(afs_dir_invalid, + TP_PROTO(const struct afs_vnode *dvnode, enum afs_dir_invalid_trace trace), + + TP_ARGS(dvnode, trace), + + TP_STRUCT__entry( + __field(unsigned int, vnode) + __field(unsigned int, unique) + __field(enum afs_dir_invalid_trace, trace) + ), + + TP_fast_assign( + __entry->vnode = dvnode->fid.vnode; + __entry->unique = dvnode->fid.unique; + __entry->trace = trace; + ), + + TP_printk("di=%x:%x %s", + __entry->vnode, __entry->unique, + __print_symbolic(__entry->trace, afs_dir_invalid_traces)) + ); + +TRACE_EVENT(afs_cb_promise, + TP_PROTO(const struct afs_vnode *vnode, enum afs_cb_promise_trace trace), + + TP_ARGS(vnode, trace), + + TP_STRUCT__entry( + __field(unsigned int, vnode) + __field(unsigned int, unique) + __field(enum afs_cb_promise_trace, trace) + ), + + TP_fast_assign( + __entry->vnode = vnode->fid.vnode; + __entry->unique = vnode->fid.unique; + __entry->trace = trace; + ), + + TP_printk("di=%x:%x %s", + __entry->vnode, __entry->unique, + __print_symbolic(__entry->trace, afs_cb_promise_traces)) + ); + +TRACE_EVENT(afs_vnode_invalid, + TP_PROTO(const struct afs_vnode *vnode, enum afs_vnode_invalid_trace trace), + + TP_ARGS(vnode, trace), + + TP_STRUCT__entry( + __field(unsigned int, vnode) + __field(unsigned int, unique) + __field(enum afs_vnode_invalid_trace, trace) + ), + + TP_fast_assign( + __entry->vnode = vnode->fid.vnode; + __entry->unique = vnode->fid.unique; + __entry->trace = trace; + ), + + TP_printk("di=%x:%x %s", + __entry->vnode, __entry->unique, + __print_symbolic(__entry->trace, afs_vnode_invalid_traces)) + ); + +TRACE_EVENT(afs_set_dv, + TP_PROTO(const struct afs_vnode *dvnode, u64 new_dv), + + TP_ARGS(dvnode, new_dv), + + TP_STRUCT__entry( + __field(unsigned int, vnode) + __field(unsigned int, unique) + __field(u64, old_dv) + __field(u64, new_dv) + ), + + TP_fast_assign( + __entry->vnode = dvnode->fid.vnode; + __entry->unique = dvnode->fid.unique; + __entry->old_dv = dvnode->status.data_version; + __entry->new_dv = new_dv; + ), + + TP_printk("di=%x:%x dv=%llx -> dv=%llx", + __entry->vnode, __entry->unique, + __entry->old_dv, __entry->new_dv) + ); + +TRACE_EVENT(afs_dv_mismatch, + TP_PROTO(const struct afs_vnode *dvnode, u64 before_dv, int delta, u64 new_dv), + + TP_ARGS(dvnode, before_dv, delta, new_dv), + + TP_STRUCT__entry( + __field(unsigned int, vnode) + __field(unsigned int, unique) + __field(int, delta) + __field(u64, before_dv) + __field(u64, new_dv) + ), + + TP_fast_assign( + __entry->vnode = dvnode->fid.vnode; + __entry->unique = dvnode->fid.unique; + __entry->delta = delta; + __entry->before_dv = before_dv; + __entry->new_dv = new_dv; + ), + + TP_printk("di=%x:%x xdv=%llx+%d dv=%llx", + __entry->vnode, __entry->unique, + __entry->before_dv, __entry->delta, __entry->new_dv) + ); + TRACE_EVENT(afs_protocol_error, TP_PROTO(struct afs_call *call, enum afs_eproto_cause cause), @@ -1611,6 +1777,36 @@ TRACE_EVENT(afs_make_call, __entry->fid.unique) ); +TRACE_EVENT(afs_read_recv, + TP_PROTO(const struct afs_operation *op, const struct afs_call *call), + + TP_ARGS(op, call), + + TP_STRUCT__entry( + __field(unsigned int, rreq) + __field(unsigned int, sreq) + __field(unsigned int, op) + __field(unsigned int, op_flags) + __field(unsigned int, call) + __field(enum afs_call_state, call_state) + ), + + TP_fast_assign( + __entry->op = op->debug_id; + __entry->sreq = op->fetch.subreq->debug_index; + __entry->rreq = op->fetch.subreq->rreq->debug_id; + __entry->op_flags = op->flags; + __entry->call = call->debug_id; + __entry->call_state = call->state; + ), + + TP_printk("R=%08x[%x] OP=%08x c=%08x cs=%x of=%x", + __entry->rreq, __entry->sreq, + __entry->op, + __entry->call, __entry->call_state, + __entry->op_flags) + ); + #endif /* _TRACE_AFS_H */ /* This part must be outside protection */ diff --git a/include/trace/events/btrfs.h b/include/trace/events/btrfs.h index 4df93ca9b7a8..549ab3b41961 100644 --- a/include/trace/events/btrfs.h +++ b/include/trace/events/btrfs.h @@ -100,7 +100,8 @@ struct find_free_extent_ctl; EM( ALLOC_CHUNK, "ALLOC_CHUNK") \ EM( ALLOC_CHUNK_FORCE, "ALLOC_CHUNK_FORCE") \ EM( RUN_DELAYED_IPUTS, "RUN_DELAYED_IPUTS") \ - EMe(COMMIT_TRANS, "COMMIT_TRANS") + EM( COMMIT_TRANS, "COMMIT_TRANS") \ + EMe(RESET_ZONES, "RESET_ZONES") /* * First define the enums in the above macros to be exported to userspace via diff --git a/include/trace/events/cachefiles.h b/include/trace/events/cachefiles.h index 7d931db02b93..a743b2a35ea7 100644 --- a/include/trace/events/cachefiles.h +++ b/include/trace/events/cachefiles.h @@ -223,10 +223,10 @@ TRACE_EVENT(cachefiles_ref, /* Note that obj may be NULL */ TP_STRUCT__entry( - __field(unsigned int, obj ) - __field(unsigned int, cookie ) - __field(enum cachefiles_obj_ref_trace, why ) - __field(int, usage ) + __field(unsigned int, obj) + __field(unsigned int, cookie) + __field(enum cachefiles_obj_ref_trace, why) + __field(int, usage) ), TP_fast_assign( @@ -249,10 +249,10 @@ TRACE_EVENT(cachefiles_lookup, TP_ARGS(obj, dir, de), TP_STRUCT__entry( - __field(unsigned int, obj ) - __field(short, error ) - __field(unsigned long, dino ) - __field(unsigned long, ino ) + __field(unsigned int, obj) + __field(short, error) + __field(unsigned long, dino) + __field(unsigned long, ino) ), TP_fast_assign( @@ -273,8 +273,8 @@ TRACE_EVENT(cachefiles_mkdir, TP_ARGS(dir, subdir), TP_STRUCT__entry( - __field(unsigned int, dir ) - __field(unsigned int, subdir ) + __field(unsigned int, dir) + __field(unsigned int, subdir) ), TP_fast_assign( @@ -293,8 +293,8 @@ TRACE_EVENT(cachefiles_tmpfile, TP_ARGS(obj, backer), TP_STRUCT__entry( - __field(unsigned int, obj ) - __field(unsigned int, backer ) + __field(unsigned int, obj) + __field(unsigned int, backer) ), TP_fast_assign( @@ -313,8 +313,8 @@ TRACE_EVENT(cachefiles_link, TP_ARGS(obj, backer), TP_STRUCT__entry( - __field(unsigned int, obj ) - __field(unsigned int, backer ) + __field(unsigned int, obj) + __field(unsigned int, backer) ), TP_fast_assign( @@ -336,9 +336,9 @@ TRACE_EVENT(cachefiles_unlink, /* Note that obj may be NULL */ TP_STRUCT__entry( - __field(unsigned int, obj ) - __field(unsigned int, ino ) - __field(enum fscache_why_object_killed, why ) + __field(unsigned int, obj) + __field(unsigned int, ino) + __field(enum fscache_why_object_killed, why) ), TP_fast_assign( @@ -361,9 +361,9 @@ TRACE_EVENT(cachefiles_rename, /* Note that obj may be NULL */ TP_STRUCT__entry( - __field(unsigned int, obj ) - __field(unsigned int, ino ) - __field(enum fscache_why_object_killed, why ) + __field(unsigned int, obj) + __field(unsigned int, ino) + __field(enum fscache_why_object_killed, why) ), TP_fast_assign( @@ -380,17 +380,20 @@ TRACE_EVENT(cachefiles_rename, TRACE_EVENT(cachefiles_coherency, TP_PROTO(struct cachefiles_object *obj, ino_t ino, + u64 disk_aux, enum cachefiles_content content, enum cachefiles_coherency_trace why), - TP_ARGS(obj, ino, content, why), + TP_ARGS(obj, ino, disk_aux, content, why), /* Note that obj may be NULL */ TP_STRUCT__entry( - __field(unsigned int, obj ) - __field(enum cachefiles_coherency_trace, why ) - __field(enum cachefiles_content, content ) - __field(u64, ino ) + __field(unsigned int, obj) + __field(enum cachefiles_coherency_trace, why) + __field(enum cachefiles_content, content) + __field(u64, ino) + __field(u64, aux) + __field(u64, disk_aux) ), TP_fast_assign( @@ -398,13 +401,17 @@ TRACE_EVENT(cachefiles_coherency, __entry->why = why; __entry->content = content; __entry->ino = ino; + __entry->aux = be64_to_cpup((__be64 *)obj->cookie->inline_aux); + __entry->disk_aux = disk_aux; ), - TP_printk("o=%08x %s B=%llx c=%u", + TP_printk("o=%08x %s B=%llx c=%u aux=%llx dsk=%llx", __entry->obj, __print_symbolic(__entry->why, cachefiles_coherency_traces), __entry->ino, - __entry->content) + __entry->content, + __entry->aux, + __entry->disk_aux) ); TRACE_EVENT(cachefiles_vol_coherency, @@ -416,9 +423,9 @@ TRACE_EVENT(cachefiles_vol_coherency, /* Note that obj may be NULL */ TP_STRUCT__entry( - __field(unsigned int, vol ) - __field(enum cachefiles_coherency_trace, why ) - __field(u64, ino ) + __field(unsigned int, vol) + __field(enum cachefiles_coherency_trace, why) + __field(u64, ino) ), TP_fast_assign( @@ -445,14 +452,14 @@ TRACE_EVENT(cachefiles_prep_read, TP_ARGS(obj, start, len, flags, source, why, cache_inode, netfs_inode), TP_STRUCT__entry( - __field(unsigned int, obj ) - __field(unsigned short, flags ) - __field(enum netfs_io_source, source ) - __field(enum cachefiles_prepare_read_trace, why ) - __field(size_t, len ) - __field(loff_t, start ) - __field(unsigned int, netfs_inode ) - __field(unsigned int, cache_inode ) + __field(unsigned int, obj) + __field(unsigned short, flags) + __field(enum netfs_io_source, source) + __field(enum cachefiles_prepare_read_trace, why) + __field(size_t, len) + __field(loff_t, start) + __field(unsigned int, netfs_inode) + __field(unsigned int, cache_inode) ), TP_fast_assign( @@ -484,10 +491,10 @@ TRACE_EVENT(cachefiles_read, TP_ARGS(obj, backer, start, len), TP_STRUCT__entry( - __field(unsigned int, obj ) - __field(unsigned int, backer ) - __field(size_t, len ) - __field(loff_t, start ) + __field(unsigned int, obj) + __field(unsigned int, backer) + __field(size_t, len) + __field(loff_t, start) ), TP_fast_assign( @@ -513,10 +520,10 @@ TRACE_EVENT(cachefiles_write, TP_ARGS(obj, backer, start, len), TP_STRUCT__entry( - __field(unsigned int, obj ) - __field(unsigned int, backer ) - __field(size_t, len ) - __field(loff_t, start ) + __field(unsigned int, obj) + __field(unsigned int, backer) + __field(size_t, len) + __field(loff_t, start) ), TP_fast_assign( @@ -540,11 +547,11 @@ TRACE_EVENT(cachefiles_trunc, TP_ARGS(obj, backer, from, to, why), TP_STRUCT__entry( - __field(unsigned int, obj ) - __field(unsigned int, backer ) - __field(enum cachefiles_trunc_trace, why ) - __field(loff_t, from ) - __field(loff_t, to ) + __field(unsigned int, obj) + __field(unsigned int, backer) + __field(enum cachefiles_trunc_trace, why) + __field(loff_t, from) + __field(loff_t, to) ), TP_fast_assign( @@ -571,8 +578,8 @@ TRACE_EVENT(cachefiles_mark_active, /* Note that obj may be NULL */ TP_STRUCT__entry( - __field(unsigned int, obj ) - __field(ino_t, inode ) + __field(unsigned int, obj) + __field(ino_t, inode) ), TP_fast_assign( @@ -592,8 +599,8 @@ TRACE_EVENT(cachefiles_mark_failed, /* Note that obj may be NULL */ TP_STRUCT__entry( - __field(unsigned int, obj ) - __field(ino_t, inode ) + __field(unsigned int, obj) + __field(ino_t, inode) ), TP_fast_assign( @@ -613,8 +620,8 @@ TRACE_EVENT(cachefiles_mark_inactive, /* Note that obj may be NULL */ TP_STRUCT__entry( - __field(unsigned int, obj ) - __field(ino_t, inode ) + __field(unsigned int, obj) + __field(ino_t, inode) ), TP_fast_assign( @@ -633,10 +640,10 @@ TRACE_EVENT(cachefiles_vfs_error, TP_ARGS(obj, backer, error, where), TP_STRUCT__entry( - __field(unsigned int, obj ) - __field(unsigned int, backer ) - __field(enum cachefiles_error_trace, where ) - __field(short, error ) + __field(unsigned int, obj) + __field(unsigned int, backer) + __field(enum cachefiles_error_trace, where) + __field(short, error) ), TP_fast_assign( @@ -660,10 +667,10 @@ TRACE_EVENT(cachefiles_io_error, TP_ARGS(obj, backer, error, where), TP_STRUCT__entry( - __field(unsigned int, obj ) - __field(unsigned int, backer ) - __field(enum cachefiles_error_trace, where ) - __field(short, error ) + __field(unsigned int, obj) + __field(unsigned int, backer) + __field(enum cachefiles_error_trace, where) + __field(short, error) ), TP_fast_assign( @@ -687,11 +694,11 @@ TRACE_EVENT(cachefiles_ondemand_open, TP_ARGS(obj, msg, load), TP_STRUCT__entry( - __field(unsigned int, obj ) - __field(unsigned int, msg_id ) - __field(unsigned int, object_id ) - __field(unsigned int, fd ) - __field(unsigned int, flags ) + __field(unsigned int, obj) + __field(unsigned int, msg_id) + __field(unsigned int, object_id) + __field(unsigned int, fd) + __field(unsigned int, flags) ), TP_fast_assign( @@ -717,9 +724,9 @@ TRACE_EVENT(cachefiles_ondemand_copen, TP_ARGS(obj, msg_id, len), TP_STRUCT__entry( - __field(unsigned int, obj ) - __field(unsigned int, msg_id ) - __field(long, len ) + __field(unsigned int, obj) + __field(unsigned int, msg_id) + __field(long, len) ), TP_fast_assign( @@ -740,9 +747,9 @@ TRACE_EVENT(cachefiles_ondemand_close, TP_ARGS(obj, msg), TP_STRUCT__entry( - __field(unsigned int, obj ) - __field(unsigned int, msg_id ) - __field(unsigned int, object_id ) + __field(unsigned int, obj) + __field(unsigned int, msg_id) + __field(unsigned int, object_id) ), TP_fast_assign( @@ -764,11 +771,11 @@ TRACE_EVENT(cachefiles_ondemand_read, TP_ARGS(obj, msg, load), TP_STRUCT__entry( - __field(unsigned int, obj ) - __field(unsigned int, msg_id ) - __field(unsigned int, object_id ) - __field(loff_t, start ) - __field(size_t, len ) + __field(unsigned int, obj) + __field(unsigned int, msg_id) + __field(unsigned int, object_id) + __field(loff_t, start) + __field(size_t, len) ), TP_fast_assign( @@ -793,8 +800,8 @@ TRACE_EVENT(cachefiles_ondemand_cread, TP_ARGS(obj, msg_id), TP_STRUCT__entry( - __field(unsigned int, obj ) - __field(unsigned int, msg_id ) + __field(unsigned int, obj) + __field(unsigned int, msg_id) ), TP_fast_assign( @@ -814,10 +821,10 @@ TRACE_EVENT(cachefiles_ondemand_fd_write, TP_ARGS(obj, backer, start, len), TP_STRUCT__entry( - __field(unsigned int, obj ) - __field(unsigned int, backer ) - __field(loff_t, start ) - __field(size_t, len ) + __field(unsigned int, obj) + __field(unsigned int, backer) + __field(loff_t, start) + __field(size_t, len) ), TP_fast_assign( @@ -840,8 +847,8 @@ TRACE_EVENT(cachefiles_ondemand_fd_release, TP_ARGS(obj, object_id), TP_STRUCT__entry( - __field(unsigned int, obj ) - __field(unsigned int, object_id ) + __field(unsigned int, obj) + __field(unsigned int, object_id) ), TP_fast_assign( diff --git a/include/trace/events/hugetlbfs.h b/include/trace/events/hugetlbfs.h index 8331c904a9ba..59605dfaeeb4 100644 --- a/include/trace/events/hugetlbfs.h +++ b/include/trace/events/hugetlbfs.h @@ -23,7 +23,7 @@ TRACE_EVENT(hugetlbfs_alloc_inode, TP_fast_assign( __entry->dev = inode->i_sb->s_dev; __entry->ino = inode->i_ino; - __entry->dir = dir->i_ino; + __entry->dir = dir ? dir->i_ino : 0; __entry->mode = mode; ), diff --git a/include/trace/events/mmflags.h b/include/trace/events/mmflags.h index bb8a59c6caa2..d36c857dd249 100644 --- a/include/trace/events/mmflags.h +++ b/include/trace/events/mmflags.h @@ -13,6 +13,69 @@ * Thus most bits set go first. */ +/* These define the values that are enums (the bits) */ +#define TRACE_GFP_FLAGS_GENERAL \ + TRACE_GFP_EM(DMA) \ + TRACE_GFP_EM(HIGHMEM) \ + TRACE_GFP_EM(DMA32) \ + TRACE_GFP_EM(MOVABLE) \ + TRACE_GFP_EM(RECLAIMABLE) \ + TRACE_GFP_EM(HIGH) \ + TRACE_GFP_EM(IO) \ + TRACE_GFP_EM(FS) \ + TRACE_GFP_EM(ZERO) \ + TRACE_GFP_EM(DIRECT_RECLAIM) \ + TRACE_GFP_EM(KSWAPD_RECLAIM) \ + TRACE_GFP_EM(WRITE) \ + TRACE_GFP_EM(NOWARN) \ + TRACE_GFP_EM(RETRY_MAYFAIL) \ + TRACE_GFP_EM(NOFAIL) \ + TRACE_GFP_EM(NORETRY) \ + TRACE_GFP_EM(MEMALLOC) \ + TRACE_GFP_EM(COMP) \ + TRACE_GFP_EM(NOMEMALLOC) \ + TRACE_GFP_EM(HARDWALL) \ + TRACE_GFP_EM(THISNODE) \ + TRACE_GFP_EM(ACCOUNT) \ + TRACE_GFP_EM(ZEROTAGS) + +#ifdef CONFIG_KASAN_HW_TAGS +# define TRACE_GFP_FLAGS_KASAN \ + TRACE_GFP_EM(SKIP_ZERO) \ + TRACE_GFP_EM(SKIP_KASAN) +#else +# define TRACE_GFP_FLAGS_KASAN +#endif + +#ifdef CONFIG_LOCKDEP +# define TRACE_GFP_FLAGS_LOCKDEP \ + TRACE_GFP_EM(NOLOCKDEP) +#else +# define TRACE_GFP_FLAGS_LOCKDEP +#endif + +#ifdef CONFIG_SLAB_OBJ_EXT +# define TRACE_GFP_FLAGS_SLAB \ + TRACE_GFP_EM(NO_OBJ_EXT) +#else +# define TRACE_GFP_FLAGS_SLAB +#endif + +#define TRACE_GFP_FLAGS \ + TRACE_GFP_FLAGS_GENERAL \ + TRACE_GFP_FLAGS_KASAN \ + TRACE_GFP_FLAGS_LOCKDEP \ + TRACE_GFP_FLAGS_SLAB + +#undef TRACE_GFP_EM +#define TRACE_GFP_EM(a) TRACE_DEFINE_ENUM(___GFP_##a##_BIT); + +TRACE_GFP_FLAGS + +/* Just in case these are ever used */ +TRACE_DEFINE_ENUM(___GFP_UNUSED_BIT); +TRACE_DEFINE_ENUM(___GFP_LAST_BIT); + #define gfpflag_string(flag) {(__force unsigned long)flag, #flag} #define __def_gfpflag_names \ diff --git a/include/trace/events/netfs.h b/include/trace/events/netfs.h index bf511bca896e..6e699cadcb29 100644 --- a/include/trace/events/netfs.h +++ b/include/trace/events/netfs.h @@ -21,6 +21,7 @@ EM(netfs_read_trace_readahead, "READAHEAD") \ EM(netfs_read_trace_readpage, "READPAGE ") \ EM(netfs_read_trace_read_gaps, "READ-GAPS") \ + EM(netfs_read_trace_read_single, "READ-SNGL") \ EM(netfs_read_trace_prefetch_for_write, "PREFETCHW") \ E_(netfs_read_trace_write_begin, "WRITEBEGN") @@ -35,9 +36,11 @@ EM(NETFS_READAHEAD, "RA") \ EM(NETFS_READPAGE, "RP") \ EM(NETFS_READ_GAPS, "RG") \ + EM(NETFS_READ_SINGLE, "R1") \ EM(NETFS_READ_FOR_WRITE, "RW") \ EM(NETFS_DIO_READ, "DR") \ EM(NETFS_WRITEBACK, "WB") \ + EM(NETFS_WRITEBACK_SINGLE, "W1") \ EM(NETFS_WRITETHROUGH, "WT") \ EM(NETFS_UNBUFFERED_WRITE, "UW") \ EM(NETFS_DIO_WRITE, "DW") \ @@ -47,17 +50,23 @@ EM(netfs_rreq_trace_assess, "ASSESS ") \ EM(netfs_rreq_trace_copy, "COPY ") \ EM(netfs_rreq_trace_collect, "COLLECT") \ + EM(netfs_rreq_trace_complete, "COMPLET") \ + EM(netfs_rreq_trace_dirty, "DIRTY ") \ EM(netfs_rreq_trace_done, "DONE ") \ EM(netfs_rreq_trace_free, "FREE ") \ EM(netfs_rreq_trace_redirty, "REDIRTY") \ EM(netfs_rreq_trace_resubmit, "RESUBMT") \ + EM(netfs_rreq_trace_set_abandon, "S-ABNDN") \ EM(netfs_rreq_trace_set_pause, "PAUSE ") \ EM(netfs_rreq_trace_unlock, "UNLOCK ") \ EM(netfs_rreq_trace_unlock_pgpriv2, "UNLCK-2") \ EM(netfs_rreq_trace_unmark, "UNMARK ") \ EM(netfs_rreq_trace_wait_ip, "WAIT-IP") \ EM(netfs_rreq_trace_wait_pause, "WT-PAUS") \ + EM(netfs_rreq_trace_wait_queue, "WAIT-Q ") \ EM(netfs_rreq_trace_wake_ip, "WAKE-IP") \ + EM(netfs_rreq_trace_wake_queue, "WAKE-Q ") \ + EM(netfs_rreq_trace_woke_queue, "WOKE-Q ") \ EM(netfs_rreq_trace_unpause, "UNPAUSE") \ E_(netfs_rreq_trace_write_done, "WR-DONE") @@ -74,6 +83,10 @@ #define netfs_sreq_traces \ EM(netfs_sreq_trace_add_donations, "+DON ") \ EM(netfs_sreq_trace_added, "ADD ") \ + EM(netfs_sreq_trace_cache_nowrite, "CA-NW") \ + EM(netfs_sreq_trace_cache_prepare, "CA-PR") \ + EM(netfs_sreq_trace_cache_write, "CA-WR") \ + EM(netfs_sreq_trace_cancel, "CANCL") \ EM(netfs_sreq_trace_clear, "CLEAR") \ EM(netfs_sreq_trace_discard, "DSCRD") \ EM(netfs_sreq_trace_donate_to_prev, "DON-P") \ @@ -84,6 +97,9 @@ EM(netfs_sreq_trace_hit_eof, "EOF ") \ EM(netfs_sreq_trace_io_progress, "IO ") \ EM(netfs_sreq_trace_limited, "LIMIT") \ + EM(netfs_sreq_trace_need_clear, "N-CLR") \ + EM(netfs_sreq_trace_partial_read, "PARTR") \ + EM(netfs_sreq_trace_need_retry, "NRTRY") \ EM(netfs_sreq_trace_prepare, "PREP ") \ EM(netfs_sreq_trace_prep_failed, "PRPFL") \ EM(netfs_sreq_trace_progress, "PRGRS") \ @@ -129,6 +145,7 @@ EM(netfs_sreq_trace_get_submit, "GET SUBMIT") \ EM(netfs_sreq_trace_get_short_read, "GET SHORTRD") \ EM(netfs_sreq_trace_new, "NEW ") \ + EM(netfs_sreq_trace_put_abandon, "PUT ABANDON") \ EM(netfs_sreq_trace_put_cancel, "PUT CANCEL ") \ EM(netfs_sreq_trace_put_clear, "PUT CLEAR ") \ EM(netfs_sreq_trace_put_consumed, "PUT CONSUME") \ @@ -152,6 +169,7 @@ EM(netfs_streaming_filled_page, "mod-streamw-f") \ EM(netfs_streaming_cont_filled_page, "mod-streamw-f+") \ EM(netfs_folio_trace_abandon, "abandon") \ + EM(netfs_folio_trace_alloc_buffer, "alloc-buf") \ EM(netfs_folio_trace_cancel_copy, "cancel-copy") \ EM(netfs_folio_trace_cancel_store, "cancel-store") \ EM(netfs_folio_trace_clear, "clear") \ @@ -168,6 +186,7 @@ EM(netfs_folio_trace_mkwrite, "mkwrite") \ EM(netfs_folio_trace_mkwrite_plus, "mkwrite+") \ EM(netfs_folio_trace_not_under_wback, "!wback") \ + EM(netfs_folio_trace_not_locked, "!locked") \ EM(netfs_folio_trace_put, "put") \ EM(netfs_folio_trace_read, "read") \ EM(netfs_folio_trace_read_done, "read-done") \ @@ -191,6 +210,14 @@ EM(netfs_trace_donate_to_next, "to-next") \ E_(netfs_trace_donate_to_deferred_next, "defer-next") +#define netfs_folioq_traces \ + EM(netfs_trace_folioq_alloc_buffer, "alloc-buf") \ + EM(netfs_trace_folioq_clear, "clear") \ + EM(netfs_trace_folioq_delete, "delete") \ + EM(netfs_trace_folioq_make_space, "make-space") \ + EM(netfs_trace_folioq_rollbuf_init, "roll-init") \ + E_(netfs_trace_folioq_read_progress, "r-progress") + #ifndef __NETFS_DECLARE_TRACE_ENUMS_ONCE_ONLY #define __NETFS_DECLARE_TRACE_ENUMS_ONCE_ONLY @@ -209,6 +236,7 @@ enum netfs_sreq_ref_trace { netfs_sreq_ref_traces } __mode(byte); enum netfs_folio_trace { netfs_folio_traces } __mode(byte); enum netfs_collect_contig_trace { netfs_collect_contig_traces } __mode(byte); enum netfs_donate_trace { netfs_donate_traces } __mode(byte); +enum netfs_folioq_trace { netfs_folioq_traces } __mode(byte); #endif @@ -232,6 +260,7 @@ netfs_sreq_ref_traces; netfs_folio_traces; netfs_collect_contig_traces; netfs_donate_traces; +netfs_folioq_traces; /* * Now redefine the EM() and E_() macros to map the enums to the strings that @@ -250,13 +279,13 @@ TRACE_EVENT(netfs_read, TP_ARGS(rreq, start, len, what), TP_STRUCT__entry( - __field(unsigned int, rreq ) - __field(unsigned int, cookie ) - __field(loff_t, i_size ) - __field(loff_t, start ) - __field(size_t, len ) - __field(enum netfs_read_trace, what ) - __field(unsigned int, netfs_inode ) + __field(unsigned int, rreq) + __field(unsigned int, cookie) + __field(loff_t, i_size) + __field(loff_t, start) + __field(size_t, len) + __field(enum netfs_read_trace, what) + __field(unsigned int, netfs_inode) ), TP_fast_assign( @@ -284,10 +313,10 @@ TRACE_EVENT(netfs_rreq, TP_ARGS(rreq, what), TP_STRUCT__entry( - __field(unsigned int, rreq ) - __field(unsigned int, flags ) - __field(enum netfs_io_origin, origin ) - __field(enum netfs_rreq_trace, what ) + __field(unsigned int, rreq) + __field(unsigned int, flags) + __field(enum netfs_io_origin, origin) + __field(enum netfs_rreq_trace, what) ), TP_fast_assign( @@ -311,15 +340,16 @@ TRACE_EVENT(netfs_sreq, TP_ARGS(sreq, what), TP_STRUCT__entry( - __field(unsigned int, rreq ) - __field(unsigned short, index ) - __field(short, error ) - __field(unsigned short, flags ) - __field(enum netfs_io_source, source ) - __field(enum netfs_sreq_trace, what ) - __field(size_t, len ) - __field(size_t, transferred ) - __field(loff_t, start ) + __field(unsigned int, rreq) + __field(unsigned short, index) + __field(short, error) + __field(unsigned short, flags) + __field(enum netfs_io_source, source) + __field(enum netfs_sreq_trace, what) + __field(u8, slot) + __field(size_t, len) + __field(size_t, transferred) + __field(loff_t, start) ), TP_fast_assign( @@ -332,15 +362,16 @@ TRACE_EVENT(netfs_sreq, __entry->len = sreq->len; __entry->transferred = sreq->transferred; __entry->start = sreq->start; + __entry->slot = sreq->io_iter.folioq_slot; ), - TP_printk("R=%08x[%x] %s %s f=%02x s=%llx %zx/%zx e=%d", + TP_printk("R=%08x[%x] %s %s f=%02x s=%llx %zx/%zx s=%u e=%d", __entry->rreq, __entry->index, __print_symbolic(__entry->source, netfs_sreq_sources), __print_symbolic(__entry->what, netfs_sreq_traces), __entry->flags, __entry->start, __entry->transferred, __entry->len, - __entry->error) + __entry->slot, __entry->error) ); TRACE_EVENT(netfs_failure, @@ -351,15 +382,15 @@ TRACE_EVENT(netfs_failure, TP_ARGS(rreq, sreq, error, what), TP_STRUCT__entry( - __field(unsigned int, rreq ) - __field(short, index ) - __field(short, error ) - __field(unsigned short, flags ) - __field(enum netfs_io_source, source ) - __field(enum netfs_failure, what ) - __field(size_t, len ) - __field(size_t, transferred ) - __field(loff_t, start ) + __field(unsigned int, rreq) + __field(short, index) + __field(short, error) + __field(unsigned short, flags) + __field(enum netfs_io_source, source) + __field(enum netfs_failure, what) + __field(size_t, len) + __field(size_t, transferred) + __field(loff_t, start) ), TP_fast_assign( @@ -390,9 +421,9 @@ TRACE_EVENT(netfs_rreq_ref, TP_ARGS(rreq_debug_id, ref, what), TP_STRUCT__entry( - __field(unsigned int, rreq ) - __field(int, ref ) - __field(enum netfs_rreq_ref_trace, what ) + __field(unsigned int, rreq) + __field(int, ref) + __field(enum netfs_rreq_ref_trace, what) ), TP_fast_assign( @@ -414,10 +445,10 @@ TRACE_EVENT(netfs_sreq_ref, TP_ARGS(rreq_debug_id, subreq_debug_index, ref, what), TP_STRUCT__entry( - __field(unsigned int, rreq ) - __field(unsigned int, subreq ) - __field(int, ref ) - __field(enum netfs_sreq_ref_trace, what ) + __field(unsigned int, rreq) + __field(unsigned int, subreq) + __field(int, ref) + __field(enum netfs_sreq_ref_trace, what) ), TP_fast_assign( @@ -465,10 +496,10 @@ TRACE_EVENT(netfs_write_iter, TP_ARGS(iocb, from), TP_STRUCT__entry( - __field(unsigned long long, start ) - __field(size_t, len ) - __field(unsigned int, flags ) - __field(unsigned int, ino ) + __field(unsigned long long, start) + __field(size_t, len) + __field(unsigned int, flags) + __field(unsigned int, ino) ), TP_fast_assign( @@ -489,12 +520,12 @@ TRACE_EVENT(netfs_write, TP_ARGS(wreq, what), TP_STRUCT__entry( - __field(unsigned int, wreq ) - __field(unsigned int, cookie ) - __field(unsigned int, ino ) - __field(enum netfs_write_trace, what ) - __field(unsigned long long, start ) - __field(unsigned long long, len ) + __field(unsigned int, wreq) + __field(unsigned int, cookie) + __field(unsigned int, ino) + __field(enum netfs_write_trace, what) + __field(unsigned long long, start) + __field(unsigned long long, len) ), TP_fast_assign( @@ -522,10 +553,10 @@ TRACE_EVENT(netfs_collect, TP_ARGS(wreq), TP_STRUCT__entry( - __field(unsigned int, wreq ) - __field(unsigned int, len ) - __field(unsigned long long, transferred ) - __field(unsigned long long, start ) + __field(unsigned int, wreq) + __field(unsigned int, len) + __field(unsigned long long, transferred) + __field(unsigned long long, start) ), TP_fast_assign( @@ -548,12 +579,12 @@ TRACE_EVENT(netfs_collect_sreq, TP_ARGS(wreq, subreq), TP_STRUCT__entry( - __field(unsigned int, wreq ) - __field(unsigned int, subreq ) - __field(unsigned int, stream ) - __field(unsigned int, len ) - __field(unsigned int, transferred ) - __field(unsigned long long, start ) + __field(unsigned int, wreq) + __field(unsigned int, subreq) + __field(unsigned int, stream) + __field(unsigned int, len) + __field(unsigned int, transferred) + __field(unsigned long long, start) ), TP_fast_assign( @@ -579,11 +610,11 @@ TRACE_EVENT(netfs_collect_folio, TP_ARGS(wreq, folio, fend, collected_to), TP_STRUCT__entry( - __field(unsigned int, wreq ) - __field(unsigned long, index ) - __field(unsigned long long, fend ) - __field(unsigned long long, cleaned_to ) - __field(unsigned long long, collected_to ) + __field(unsigned int, wreq) + __field(unsigned long, index) + __field(unsigned long long, fend) + __field(unsigned long long, cleaned_to) + __field(unsigned long long, collected_to) ), TP_fast_assign( @@ -608,10 +639,10 @@ TRACE_EVENT(netfs_collect_state, TP_ARGS(wreq, collected_to, notes), TP_STRUCT__entry( - __field(unsigned int, wreq ) - __field(unsigned int, notes ) - __field(unsigned long long, collected_to ) - __field(unsigned long long, cleaned_to ) + __field(unsigned int, wreq) + __field(unsigned int, notes) + __field(unsigned long long, collected_to) + __field(unsigned long long, cleaned_to) ), TP_fast_assign( @@ -680,69 +711,27 @@ TRACE_EVENT(netfs_collect_stream, __entry->collected_to, __entry->front) ); -TRACE_EVENT(netfs_progress, - TP_PROTO(const struct netfs_io_subrequest *subreq, - unsigned long long start, size_t avail, size_t part), +TRACE_EVENT(netfs_folioq, + TP_PROTO(const struct folio_queue *fq, + enum netfs_folioq_trace trace), - TP_ARGS(subreq, start, avail, part), + TP_ARGS(fq, trace), TP_STRUCT__entry( __field(unsigned int, rreq) - __field(unsigned int, subreq) - __field(unsigned int, consumed) - __field(unsigned int, transferred) - __field(unsigned long long, f_start) - __field(unsigned int, f_avail) - __field(unsigned int, f_part) - __field(unsigned char, slot) + __field(unsigned int, id) + __field(enum netfs_folioq_trace, trace) ), TP_fast_assign( - __entry->rreq = subreq->rreq->debug_id; - __entry->subreq = subreq->debug_index; - __entry->consumed = subreq->consumed; - __entry->transferred = subreq->transferred; - __entry->f_start = start; - __entry->f_avail = avail; - __entry->f_part = part; - __entry->slot = subreq->curr_folioq_slot; - ), - - TP_printk("R=%08x[%02x] s=%llx ct=%x/%x pa=%x/%x sl=%x", - __entry->rreq, __entry->subreq, __entry->f_start, - __entry->consumed, __entry->transferred, - __entry->f_part, __entry->f_avail, __entry->slot) - ); - -TRACE_EVENT(netfs_donate, - TP_PROTO(const struct netfs_io_request *rreq, - const struct netfs_io_subrequest *from, - const struct netfs_io_subrequest *to, - size_t amount, - enum netfs_donate_trace trace), - - TP_ARGS(rreq, from, to, amount, trace), - - TP_STRUCT__entry( - __field(unsigned int, rreq) - __field(unsigned int, from) - __field(unsigned int, to) - __field(unsigned int, amount) - __field(enum netfs_donate_trace, trace) - ), - - TP_fast_assign( - __entry->rreq = rreq->debug_id; - __entry->from = from->debug_index; - __entry->to = to ? to->debug_index : -1; - __entry->amount = amount; + __entry->rreq = fq ? fq->rreq_id : 0; + __entry->id = fq ? fq->debug_id : 0; __entry->trace = trace; ), - TP_printk("R=%08x[%02x] -> [%02x] %s am=%x", - __entry->rreq, __entry->from, __entry->to, - __print_symbolic(__entry->trace, netfs_donate_traces), - __entry->amount) + TP_printk("R=%08x fq=%x %s", + __entry->rreq, __entry->id, + __print_symbolic(__entry->trace, netfs_folioq_traces)) ); #undef EM diff --git a/include/uapi/linux/fiemap.h b/include/uapi/linux/fiemap.h index 24ca0c00cae3..9d9e8ae32b41 100644 --- a/include/uapi/linux/fiemap.h +++ b/include/uapi/linux/fiemap.h @@ -14,37 +14,56 @@ #include <linux/types.h> +/** + * struct fiemap_extent - description of one fiemap extent + * @fe_logical: byte offset of the extent in the file + * @fe_physical: byte offset of extent on disk + * @fe_length: length in bytes for this extent + * @fe_flags: FIEMAP_EXTENT_* flags for this extent + */ struct fiemap_extent { - __u64 fe_logical; /* logical offset in bytes for the start of - * the extent from the beginning of the file */ - __u64 fe_physical; /* physical offset in bytes for the start - * of the extent from the beginning of the disk */ - __u64 fe_length; /* length in bytes for this extent */ + __u64 fe_logical; + __u64 fe_physical; + __u64 fe_length; + /* private: */ __u64 fe_reserved64[2]; - __u32 fe_flags; /* FIEMAP_EXTENT_* flags for this extent */ + /* public: */ + __u32 fe_flags; + /* private: */ __u32 fe_reserved[3]; }; +/** + * struct fiemap - file extent mappings + * @fm_start: byte offset (inclusive) at which to start mapping (in) + * @fm_length: logical length of mapping which userspace wants (in) + * @fm_flags: FIEMAP_FLAG_* flags for request (in/out) + * @fm_mapped_extents: number of extents that were mapped (out) + * @fm_extent_count: size of fm_extents array (in) + * @fm_extents: array of mapped extents (out) + */ struct fiemap { - __u64 fm_start; /* logical offset (inclusive) at - * which to start mapping (in) */ - __u64 fm_length; /* logical length of mapping which - * userspace wants (in) */ - __u32 fm_flags; /* FIEMAP_FLAG_* flags for request (in/out) */ - __u32 fm_mapped_extents;/* number of extents that were mapped (out) */ - __u32 fm_extent_count; /* size of fm_extents array (in) */ + __u64 fm_start; + __u64 fm_length; + __u32 fm_flags; + __u32 fm_mapped_extents; + __u32 fm_extent_count; + /* private: */ __u32 fm_reserved; - struct fiemap_extent fm_extents[]; /* array of mapped extents (out) */ + /* public: */ + struct fiemap_extent fm_extents[]; }; #define FIEMAP_MAX_OFFSET (~0ULL) +/* flags used in fm_flags: */ #define FIEMAP_FLAG_SYNC 0x00000001 /* sync file data before map */ #define FIEMAP_FLAG_XATTR 0x00000002 /* map extended attribute tree */ #define FIEMAP_FLAG_CACHE 0x00000004 /* request caching of the extents */ #define FIEMAP_FLAGS_COMPAT (FIEMAP_FLAG_SYNC | FIEMAP_FLAG_XATTR) +/* flags used in fe_flags: */ #define FIEMAP_EXTENT_LAST 0x00000001 /* Last extent in file. */ #define FIEMAP_EXTENT_UNKNOWN 0x00000002 /* Data location unknown. */ #define FIEMAP_EXTENT_DELALLOC 0x00000004 /* Location still pending. diff --git a/include/uapi/linux/fs.h b/include/uapi/linux/fs.h index 753971770733..56a4f93a08f4 100644 --- a/include/uapi/linux/fs.h +++ b/include/uapi/linux/fs.h @@ -332,9 +332,13 @@ typedef int __bitwise __kernel_rwf_t; /* Atomic Write */ #define RWF_ATOMIC ((__force __kernel_rwf_t)0x00000040) +/* buffered IO that drops the cache after reading or writing data */ +#define RWF_DONTCACHE ((__force __kernel_rwf_t)0x00000080) + /* mask of flags supported by the kernel */ #define RWF_SUPPORTED (RWF_HIPRI | RWF_DSYNC | RWF_SYNC | RWF_NOWAIT |\ - RWF_APPEND | RWF_NOAPPEND | RWF_ATOMIC) + RWF_APPEND | RWF_NOAPPEND | RWF_ATOMIC |\ + RWF_DONTCACHE) #define PROCFS_IOCTL_MAGIC 'f' diff --git a/include/uapi/linux/mptcp_pm.h b/include/uapi/linux/mptcp_pm.h index 50589e5dd6a3..84fa8a21dfd0 100644 --- a/include/uapi/linux/mptcp_pm.h +++ b/include/uapi/linux/mptcp_pm.h @@ -12,31 +12,33 @@ /** * enum mptcp_event_type * @MPTCP_EVENT_UNSPEC: unused event - * @MPTCP_EVENT_CREATED: token, family, saddr4 | saddr6, daddr4 | daddr6, - * sport, dport A new MPTCP connection has been created. It is the good time - * to allocate memory and send ADD_ADDR if needed. Depending on the + * @MPTCP_EVENT_CREATED: A new MPTCP connection has been created. It is the + * good time to allocate memory and send ADD_ADDR if needed. Depending on the * traffic-patterns it can take a long time until the MPTCP_EVENT_ESTABLISHED - * is sent. - * @MPTCP_EVENT_ESTABLISHED: token, family, saddr4 | saddr6, daddr4 | daddr6, - * sport, dport A MPTCP connection is established (can start new subflows). - * @MPTCP_EVENT_CLOSED: token A MPTCP connection has stopped. - * @MPTCP_EVENT_ANNOUNCED: token, rem_id, family, daddr4 | daddr6 [, dport] A - * new address has been announced by the peer. - * @MPTCP_EVENT_REMOVED: token, rem_id An address has been lost by the peer. - * @MPTCP_EVENT_SUB_ESTABLISHED: token, family, loc_id, rem_id, saddr4 | - * saddr6, daddr4 | daddr6, sport, dport, backup, if_idx [, error] A new - * subflow has been established. 'error' should not be set. - * @MPTCP_EVENT_SUB_CLOSED: token, family, loc_id, rem_id, saddr4 | saddr6, - * daddr4 | daddr6, sport, dport, backup, if_idx [, error] A subflow has been - * closed. An error (copy of sk_err) could be set if an error has been - * detected for this subflow. - * @MPTCP_EVENT_SUB_PRIORITY: token, family, loc_id, rem_id, saddr4 | saddr6, - * daddr4 | daddr6, sport, dport, backup, if_idx [, error] The priority of a - * subflow has changed. 'error' should not be set. - * @MPTCP_EVENT_LISTENER_CREATED: family, sport, saddr4 | saddr6 A new PM - * listener is created. - * @MPTCP_EVENT_LISTENER_CLOSED: family, sport, saddr4 | saddr6 A PM listener - * is closed. + * is sent. Attributes: token, family, saddr4 | saddr6, daddr4 | daddr6, + * sport, dport, server-side. + * @MPTCP_EVENT_ESTABLISHED: A MPTCP connection is established (can start new + * subflows). Attributes: token, family, saddr4 | saddr6, daddr4 | daddr6, + * sport, dport, server-side. + * @MPTCP_EVENT_CLOSED: A MPTCP connection has stopped. Attribute: token. + * @MPTCP_EVENT_ANNOUNCED: A new address has been announced by the peer. + * Attributes: token, rem_id, family, daddr4 | daddr6 [, dport]. + * @MPTCP_EVENT_REMOVED: An address has been lost by the peer. Attributes: + * token, rem_id. + * @MPTCP_EVENT_SUB_ESTABLISHED: A new subflow has been established. 'error' + * should not be set. Attributes: token, family, loc_id, rem_id, saddr4 | + * saddr6, daddr4 | daddr6, sport, dport, backup, if_idx [, error]. + * @MPTCP_EVENT_SUB_CLOSED: A subflow has been closed. An error (copy of + * sk_err) could be set if an error has been detected for this subflow. + * Attributes: token, family, loc_id, rem_id, saddr4 | saddr6, daddr4 | + * daddr6, sport, dport, backup, if_idx [, error]. + * @MPTCP_EVENT_SUB_PRIORITY: The priority of a subflow has changed. 'error' + * should not be set. Attributes: token, family, loc_id, rem_id, saddr4 | + * saddr6, daddr4 | daddr6, sport, dport, backup, if_idx [, error]. + * @MPTCP_EVENT_LISTENER_CREATED: A new PM listener is created. Attributes: + * family, sport, saddr4 | saddr6. + * @MPTCP_EVENT_LISTENER_CLOSED: A PM listener is closed. Attributes: family, + * sport, saddr4 | saddr6. */ enum mptcp_event_type { MPTCP_EVENT_UNSPEC, diff --git a/include/uapi/linux/stat.h b/include/uapi/linux/stat.h index 887a25286441..f78ee3670dd5 100644 --- a/include/uapi/linux/stat.h +++ b/include/uapi/linux/stat.h @@ -98,43 +98,93 @@ struct statx_timestamp { */ struct statx { /* 0x00 */ - __u32 stx_mask; /* What results were written [uncond] */ - __u32 stx_blksize; /* Preferred general I/O size [uncond] */ - __u64 stx_attributes; /* Flags conveying information about the file [uncond] */ + /* What results were written [uncond] */ + __u32 stx_mask; + + /* Preferred general I/O size [uncond] */ + __u32 stx_blksize; + + /* Flags conveying information about the file [uncond] */ + __u64 stx_attributes; + /* 0x10 */ - __u32 stx_nlink; /* Number of hard links */ - __u32 stx_uid; /* User ID of owner */ - __u32 stx_gid; /* Group ID of owner */ - __u16 stx_mode; /* File mode */ + /* Number of hard links */ + __u32 stx_nlink; + + /* User ID of owner */ + __u32 stx_uid; + + /* Group ID of owner */ + __u32 stx_gid; + + /* File mode */ + __u16 stx_mode; __u16 __spare0[1]; + /* 0x20 */ - __u64 stx_ino; /* Inode number */ - __u64 stx_size; /* File size */ - __u64 stx_blocks; /* Number of 512-byte blocks allocated */ - __u64 stx_attributes_mask; /* Mask to show what's supported in stx_attributes */ + /* Inode number */ + __u64 stx_ino; + + /* File size */ + __u64 stx_size; + + /* Number of 512-byte blocks allocated */ + __u64 stx_blocks; + + /* Mask to show what's supported in stx_attributes */ + __u64 stx_attributes_mask; + /* 0x40 */ - struct statx_timestamp stx_atime; /* Last access time */ - struct statx_timestamp stx_btime; /* File creation time */ - struct statx_timestamp stx_ctime; /* Last attribute change time */ - struct statx_timestamp stx_mtime; /* Last data modification time */ + /* Last access time */ + struct statx_timestamp stx_atime; + + /* File creation time */ + struct statx_timestamp stx_btime; + + /* Last attribute change time */ + struct statx_timestamp stx_ctime; + + /* Last data modification time */ + struct statx_timestamp stx_mtime; + /* 0x80 */ - __u32 stx_rdev_major; /* Device ID of special file [if bdev/cdev] */ + /* Device ID of special file [if bdev/cdev] */ + __u32 stx_rdev_major; __u32 stx_rdev_minor; - __u32 stx_dev_major; /* ID of device containing file [uncond] */ + + /* ID of device containing file [uncond] */ + __u32 stx_dev_major; __u32 stx_dev_minor; + /* 0x90 */ __u64 stx_mnt_id; - __u32 stx_dio_mem_align; /* Memory buffer alignment for direct I/O */ - __u32 stx_dio_offset_align; /* File offset alignment for direct I/O */ + + /* Memory buffer alignment for direct I/O */ + __u32 stx_dio_mem_align; + + /* File offset alignment for direct I/O */ + __u32 stx_dio_offset_align; + /* 0xa0 */ - __u64 stx_subvol; /* Subvolume identifier */ - __u32 stx_atomic_write_unit_min; /* Min atomic write unit in bytes */ - __u32 stx_atomic_write_unit_max; /* Max atomic write unit in bytes */ + /* Subvolume identifier */ + __u64 stx_subvol; + + /* Min atomic write unit in bytes */ + __u32 stx_atomic_write_unit_min; + + /* Max atomic write unit in bytes */ + __u32 stx_atomic_write_unit_max; + /* 0xb0 */ - __u32 stx_atomic_write_segments_max; /* Max atomic write segment count */ - __u32 __spare1[1]; + /* Max atomic write segment count */ + __u32 stx_atomic_write_segments_max; + + /* File offset alignment for direct I/O reads */ + __u32 stx_dio_read_offset_align; + /* 0xb8 */ __u64 __spare3[9]; /* Spare space for future expansion */ + /* 0x100 */ }; @@ -164,6 +214,7 @@ struct statx { #define STATX_MNT_ID_UNIQUE 0x00004000U /* Want/got extended stx_mount_id */ #define STATX_SUBVOL 0x00008000U /* Want/got stx_subvol */ #define STATX_WRITE_ATOMIC 0x00010000U /* Want/got atomic_write_* fields */ +#define STATX_DIO_READ_ALIGN 0x00020000U /* Want/got dio read alignment info */ #define STATX__RESERVED 0x80000000U /* Reserved for future struct statx expansion */ diff --git a/include/uapi/linux/stddef.h b/include/uapi/linux/stddef.h index 58154117d9b0..a6fce46aeb37 100644 --- a/include/uapi/linux/stddef.h +++ b/include/uapi/linux/stddef.h @@ -8,6 +8,13 @@ #define __always_inline inline #endif +/* Not all C++ standards support type declarations inside an anonymous union */ +#ifndef __cplusplus +#define __struct_group_tag(TAG) TAG +#else +#define __struct_group_tag(TAG) +#endif + /** * __struct_group() - Create a mirrored named and anonyomous struct * @@ -20,13 +27,13 @@ * and size: one anonymous and one named. The former's members can be used * normally without sub-struct naming, and the latter can be used to * reason about the start, end, and size of the group of struct members. - * The named struct can also be explicitly tagged for layer reuse, as well - * as both having struct attributes appended. + * The named struct can also be explicitly tagged for layer reuse (C only), + * as well as both having struct attributes appended. */ #define __struct_group(TAG, NAME, ATTRS, MEMBERS...) \ union { \ struct { MEMBERS } ATTRS; \ - struct TAG { MEMBERS } ATTRS NAME; \ + struct __struct_group_tag(TAG) { MEMBERS } ATTRS NAME; \ } ATTRS #ifdef __cplusplus diff --git a/include/ufs/ufshcd.h b/include/ufs/ufshcd.h index d650ae6b58d3..74e5b9960c54 100644 --- a/include/ufs/ufshcd.h +++ b/include/ufs/ufshcd.h @@ -329,7 +329,6 @@ struct ufs_pwr_mode_info { * @program_key: program or evict an inline encryption key * @fill_crypto_prdt: initialize crypto-related fields in the PRDT * @event_notify: called to notify important events - * @reinit_notify: called to notify reinit of UFSHCD during max gear switch * @mcq_config_resource: called to configure MCQ platform resources * @get_hba_mac: reports maximum number of outstanding commands supported by * the controller. Should be implemented for UFSHCI 4.0 or later @@ -381,7 +380,6 @@ struct ufs_hba_variant_ops { void *prdt, unsigned int num_segments); void (*event_notify)(struct ufs_hba *hba, enum ufs_event_type evt, void *data); - void (*reinit_notify)(struct ufs_hba *); int (*mcq_config_resource)(struct ufs_hba *hba); int (*get_hba_mac)(struct ufs_hba *hba); int (*op_runtime_config)(struct ufs_hba *hba); diff --git a/io_uring/eventfd.c b/io_uring/eventfd.c index fab936d31ba8..100d5da94cb9 100644 --- a/io_uring/eventfd.c +++ b/io_uring/eventfd.c @@ -33,20 +33,18 @@ static void io_eventfd_free(struct rcu_head *rcu) kfree(ev_fd); } -static void io_eventfd_do_signal(struct rcu_head *rcu) +static void io_eventfd_put(struct io_ev_fd *ev_fd) { - struct io_ev_fd *ev_fd = container_of(rcu, struct io_ev_fd, rcu); - - eventfd_signal_mask(ev_fd->cq_ev_fd, EPOLL_URING_WAKE); - if (refcount_dec_and_test(&ev_fd->refs)) - io_eventfd_free(rcu); + call_rcu(&ev_fd->rcu, io_eventfd_free); } -static void io_eventfd_put(struct io_ev_fd *ev_fd) +static void io_eventfd_do_signal(struct rcu_head *rcu) { - if (refcount_dec_and_test(&ev_fd->refs)) - call_rcu(&ev_fd->rcu, io_eventfd_free); + struct io_ev_fd *ev_fd = container_of(rcu, struct io_ev_fd, rcu); + + eventfd_signal_mask(ev_fd->cq_ev_fd, EPOLL_URING_WAKE); + io_eventfd_put(ev_fd); } static void io_eventfd_release(struct io_ev_fd *ev_fd, bool put_ref) diff --git a/io_uring/io-wq.c b/io_uring/io-wq.c index a38f36b68060..5d0928f37471 100644 --- a/io_uring/io-wq.c +++ b/io_uring/io-wq.c @@ -634,7 +634,7 @@ static int io_wq_worker(void *data) struct io_wq_acct *acct = io_wq_get_acct(worker); struct io_wq *wq = worker->wq; bool exit_mask = false, last_timeout = false; - char buf[TASK_COMM_LEN]; + char buf[TASK_COMM_LEN] = {}; set_mask_bits(&worker->flags, 0, BIT(IO_WORKER_F_UP) | BIT(IO_WORKER_F_RUNNING)); diff --git a/io_uring/io_uring.c b/io_uring/io_uring.c index d3403c8216db..4758f1ba902b 100644 --- a/io_uring/io_uring.c +++ b/io_uring/io_uring.c @@ -320,7 +320,7 @@ static __cold struct io_ring_ctx *io_ring_ctx_alloc(struct io_uring_params *p) ret |= io_alloc_cache_init(&ctx->rw_cache, IO_ALLOC_CACHE_MAX, sizeof(struct io_async_rw)); ret |= io_alloc_cache_init(&ctx->uring_cache, IO_ALLOC_CACHE_MAX, - sizeof(struct uring_cache)); + sizeof(struct io_uring_cmd_data)); spin_lock_init(&ctx->msg_lock); ret |= io_alloc_cache_init(&ctx->msg_cache, IO_ALLOC_CACHE_MAX, sizeof(struct io_kiocb)); @@ -1226,10 +1226,7 @@ static void io_req_normal_work_add(struct io_kiocb *req) /* SQPOLL doesn't need the task_work added, it'll run it itself */ if (ctx->flags & IORING_SETUP_SQPOLL) { - struct io_sq_data *sqd = ctx->sq_data; - - if (sqd->thread) - __set_notify_signal(sqd->thread); + __set_notify_signal(tctx->task); return; } @@ -2813,13 +2810,12 @@ static __poll_t io_uring_poll(struct file *file, poll_table *wait) if (unlikely(!ctx->poll_activated)) io_activate_pollwq(ctx); - - poll_wait(file, &ctx->poll_wq, wait); /* - * synchronizes with barrier from wq_has_sleeper call in - * io_commit_cqring + * provides mb() which pairs with barrier from wq_has_sleeper + * call in io_commit_cqring */ - smp_rmb(); + poll_wait(file, &ctx->poll_wq, wait); + if (!io_sqring_full(ctx)) mask |= EPOLLOUT | EPOLLWRNORM; diff --git a/io_uring/io_uring.h b/io_uring/io_uring.h index 12abee607e4a..492cbbf2c23b 100644 --- a/io_uring/io_uring.h +++ b/io_uring/io_uring.h @@ -125,6 +125,9 @@ static inline void io_lockdep_assert_cq_locked(struct io_ring_ctx *ctx) #if defined(CONFIG_PROVE_LOCKING) lockdep_assert(in_task()); + if (ctx->flags & IORING_SETUP_DEFER_TASKRUN) + lockdep_assert_held(&ctx->uring_lock); + if (ctx->flags & IORING_SETUP_IOPOLL) { lockdep_assert_held(&ctx->uring_lock); } else if (!ctx->task_complete) { @@ -136,9 +139,7 @@ static inline void io_lockdep_assert_cq_locked(struct io_ring_ctx *ctx) * Not from an SQE, as those cannot be submitted, but via * updating tagged resources. */ - if (percpu_ref_is_dying(&ctx->refs)) - lockdep_assert(current_work()); - else + if (!percpu_ref_is_dying(&ctx->refs)) lockdep_assert(current == ctx->submitter_task); } #endif diff --git a/io_uring/kbuf.c b/io_uring/kbuf.c index d407576ddfb7..eec5eb7de843 100644 --- a/io_uring/kbuf.c +++ b/io_uring/kbuf.c @@ -139,6 +139,7 @@ static void __user *io_ring_buffer_select(struct io_kiocb *req, size_t *len, struct io_uring_buf_ring *br = bl->buf_ring; __u16 tail, head = bl->head; struct io_uring_buf *buf; + void __user *ret; tail = smp_load_acquire(&br->tail); if (unlikely(tail == head)) @@ -153,6 +154,7 @@ static void __user *io_ring_buffer_select(struct io_kiocb *req, size_t *len, req->flags |= REQ_F_BUFFER_RING | REQ_F_BUFFERS_COMMIT; req->buf_list = bl; req->buf_index = buf->bid; + ret = u64_to_user_ptr(buf->addr); if (issue_flags & IO_URING_F_UNLOCKED || !io_file_can_poll(req)) { /* @@ -168,7 +170,7 @@ static void __user *io_ring_buffer_select(struct io_kiocb *req, size_t *len, io_kbuf_commit(req, bl, *len, 1); req->buf_list = NULL; } - return u64_to_user_ptr(buf->addr); + return ret; } void __user *io_buffer_select(struct io_kiocb *req, size_t *len, diff --git a/io_uring/net.c b/io_uring/net.c index df1f7dc6f1c8..c6cd38cc5dc4 100644 --- a/io_uring/net.c +++ b/io_uring/net.c @@ -754,6 +754,7 @@ static int io_recvmsg_prep_setup(struct io_kiocb *req) if (req->opcode == IORING_OP_RECV) { kmsg->msg.msg_name = NULL; kmsg->msg.msg_namelen = 0; + kmsg->msg.msg_inq = 0; kmsg->msg.msg_control = NULL; kmsg->msg.msg_get_inq = 1; kmsg->msg.msg_controllen = 0; diff --git a/io_uring/opdef.c b/io_uring/opdef.c index 3de75eca1c92..e8baef4e5146 100644 --- a/io_uring/opdef.c +++ b/io_uring/opdef.c @@ -7,6 +7,7 @@ #include <linux/fs.h> #include <linux/file.h> #include <linux/io_uring.h> +#include <linux/io_uring/cmd.h> #include "io_uring.h" #include "opdef.h" @@ -414,7 +415,7 @@ const struct io_issue_def io_issue_defs[] = { .plug = 1, .iopoll = 1, .iopoll_queue = 1, - .async_size = 2 * sizeof(struct io_uring_sqe), + .async_size = sizeof(struct io_uring_cmd_data), .prep = io_uring_cmd_prep, .issue = io_uring_cmd, }, diff --git a/io_uring/register.c b/io_uring/register.c index fdd44914c39c..371aec87e078 100644 --- a/io_uring/register.c +++ b/io_uring/register.c @@ -405,8 +405,8 @@ static int io_register_resize_rings(struct io_ring_ctx *ctx, void __user *arg) { struct io_ring_ctx_rings o = { }, n = { }, *to_free = NULL; size_t size, sq_array_offset; + unsigned i, tail, old_head; struct io_uring_params p; - unsigned i, tail; void *ptr; int ret; @@ -449,10 +449,18 @@ static int io_register_resize_rings(struct io_ring_ctx *ctx, void __user *arg) if (IS_ERR(n.rings)) return PTR_ERR(n.rings); - n.rings->sq_ring_mask = p.sq_entries - 1; - n.rings->cq_ring_mask = p.cq_entries - 1; - n.rings->sq_ring_entries = p.sq_entries; - n.rings->cq_ring_entries = p.cq_entries; + /* + * At this point n.rings is shared with userspace, just like o.rings + * is as well. While we don't expect userspace to modify it while + * a resize is in progress, and it's most likely that userspace will + * shoot itself in the foot if it does, we can't always assume good + * intent... Use read/write once helpers from here on to indicate the + * shared nature of it. + */ + WRITE_ONCE(n.rings->sq_ring_mask, p.sq_entries - 1); + WRITE_ONCE(n.rings->cq_ring_mask, p.cq_entries - 1); + WRITE_ONCE(n.rings->sq_ring_entries, p.sq_entries); + WRITE_ONCE(n.rings->cq_ring_entries, p.cq_entries); if (copy_to_user(arg, &p, sizeof(p))) { io_register_free_rings(&p, &n); @@ -509,20 +517,22 @@ static int io_register_resize_rings(struct io_ring_ctx *ctx, void __user *arg) * rings can't hold what is already there, then fail the operation. */ n.sq_sqes = ptr; - tail = o.rings->sq.tail; - if (tail - o.rings->sq.head > p.sq_entries) + tail = READ_ONCE(o.rings->sq.tail); + old_head = READ_ONCE(o.rings->sq.head); + if (tail - old_head > p.sq_entries) goto overflow; - for (i = o.rings->sq.head; i < tail; i++) { + for (i = old_head; i < tail; i++) { unsigned src_head = i & (ctx->sq_entries - 1); - unsigned dst_head = i & n.rings->sq_ring_mask; + unsigned dst_head = i & (p.sq_entries - 1); n.sq_sqes[dst_head] = o.sq_sqes[src_head]; } - n.rings->sq.head = o.rings->sq.head; - n.rings->sq.tail = o.rings->sq.tail; + WRITE_ONCE(n.rings->sq.head, READ_ONCE(o.rings->sq.head)); + WRITE_ONCE(n.rings->sq.tail, READ_ONCE(o.rings->sq.tail)); - tail = o.rings->cq.tail; - if (tail - o.rings->cq.head > p.cq_entries) { + tail = READ_ONCE(o.rings->cq.tail); + old_head = READ_ONCE(o.rings->cq.head); + if (tail - old_head > p.cq_entries) { overflow: /* restore old rings, and return -EOVERFLOW via cleanup path */ ctx->rings = o.rings; @@ -531,21 +541,21 @@ overflow: ret = -EOVERFLOW; goto out; } - for (i = o.rings->cq.head; i < tail; i++) { + for (i = old_head; i < tail; i++) { unsigned src_head = i & (ctx->cq_entries - 1); - unsigned dst_head = i & n.rings->cq_ring_mask; + unsigned dst_head = i & (p.cq_entries - 1); n.rings->cqes[dst_head] = o.rings->cqes[src_head]; } - n.rings->cq.head = o.rings->cq.head; - n.rings->cq.tail = o.rings->cq.tail; + WRITE_ONCE(n.rings->cq.head, READ_ONCE(o.rings->cq.head)); + WRITE_ONCE(n.rings->cq.tail, READ_ONCE(o.rings->cq.tail)); /* invalidate cached cqe refill */ ctx->cqe_cached = ctx->cqe_sentinel = NULL; - n.rings->sq_dropped = o.rings->sq_dropped; - n.rings->sq_flags = o.rings->sq_flags; - n.rings->cq_flags = o.rings->cq_flags; - n.rings->cq_overflow = o.rings->cq_overflow; + WRITE_ONCE(n.rings->sq_dropped, READ_ONCE(o.rings->sq_dropped)); + WRITE_ONCE(n.rings->sq_flags, READ_ONCE(o.rings->sq_flags)); + WRITE_ONCE(n.rings->cq_flags, READ_ONCE(o.rings->cq_flags)); + WRITE_ONCE(n.rings->cq_overflow, READ_ONCE(o.rings->cq_overflow)); /* all done, store old pointers and assign new ones */ if (!(ctx->flags & IORING_SETUP_NO_SQARRAY)) diff --git a/io_uring/rsrc.c b/io_uring/rsrc.c index 077f84684c18..69937d0c94f9 100644 --- a/io_uring/rsrc.c +++ b/io_uring/rsrc.c @@ -997,7 +997,7 @@ static int io_clone_buffers(struct io_ring_ctx *ctx, struct io_ring_ctx *src_ctx dst_node = io_rsrc_node_alloc(ctx, IORING_RSRC_BUFFER); if (!dst_node) { ret = -ENOMEM; - goto out_put_free; + goto out_unlock; } refcount_inc(&src_node->buf->refs); @@ -1033,14 +1033,6 @@ static int io_clone_buffers(struct io_ring_ctx *ctx, struct io_ring_ctx *src_ctx mutex_lock(&src_ctx->uring_lock); /* someone raced setting up buffers, dump ours */ ret = -EBUSY; -out_put_free: - i = data.nr; - while (i--) { - if (data.nodes[i]) { - io_buffer_unmap(src_ctx, data.nodes[i]); - kfree(data.nodes[i]); - } - } out_unlock: io_rsrc_data_free(ctx, &data); mutex_unlock(&src_ctx->uring_lock); diff --git a/io_uring/rw.c b/io_uring/rw.c index 0bcb83e4ce3c..29bb3010f9c0 100644 --- a/io_uring/rw.c +++ b/io_uring/rw.c @@ -983,6 +983,8 @@ int io_read_mshot(struct io_kiocb *req, unsigned int issue_flags) io_kbuf_recycle(req, issue_flags); if (ret < 0) req_set_fail(req); + } else if (!(req->flags & REQ_F_APOLL_MULTISHOT)) { + cflags = io_put_kbuf(req, ret, issue_flags); } else { /* * Any successful return value will keep the multishot read diff --git a/io_uring/sqpoll.c b/io_uring/sqpoll.c index 6df5e649c413..d037cc68e9d3 100644 --- a/io_uring/sqpoll.c +++ b/io_uring/sqpoll.c @@ -264,12 +264,16 @@ static int io_sq_thread(void *data) struct io_ring_ctx *ctx; struct rusage start; unsigned long timeout = 0; - char buf[TASK_COMM_LEN]; + char buf[TASK_COMM_LEN] = {}; DEFINE_WAIT(wait); /* offload context creation failed, just exit */ - if (!current->io_uring) + if (!current->io_uring) { + mutex_lock(&sqd->lock); + sqd->thread = NULL; + mutex_unlock(&sqd->lock); goto err_out; + } snprintf(buf, sizeof(buf), "iou-sqp-%d", sqd->task_pid); set_task_comm(current, buf); @@ -405,6 +409,7 @@ void io_sqpoll_wait_sq(struct io_ring_ctx *ctx) __cold int io_sq_offload_create(struct io_ring_ctx *ctx, struct io_uring_params *p) { + struct task_struct *task_to_put = NULL; int ret; /* Retain compatibility with failing for an invalid attach attempt */ @@ -480,6 +485,7 @@ __cold int io_sq_offload_create(struct io_ring_ctx *ctx, } sqd->thread = tsk; + task_to_put = get_task_struct(tsk); ret = io_uring_alloc_task_context(tsk, ctx); wake_up_new_task(tsk); if (ret) @@ -490,11 +496,15 @@ __cold int io_sq_offload_create(struct io_ring_ctx *ctx, goto err; } + if (task_to_put) + put_task_struct(task_to_put); return 0; err_sqpoll: complete(&ctx->sq_data->exited); err: io_sq_thread_finish(ctx); + if (task_to_put) + put_task_struct(task_to_put); return ret; } diff --git a/io_uring/timeout.c b/io_uring/timeout.c index bbe58638eca7..e9cec9e4dc2f 100644 --- a/io_uring/timeout.c +++ b/io_uring/timeout.c @@ -85,7 +85,27 @@ static void io_timeout_complete(struct io_kiocb *req, struct io_tw_state *ts) io_req_task_complete(req, ts); } -static bool io_kill_timeout(struct io_kiocb *req, int status) +static __cold bool io_flush_killed_timeouts(struct list_head *list, int err) +{ + if (list_empty(list)) + return false; + + while (!list_empty(list)) { + struct io_timeout *timeout; + struct io_kiocb *req; + + timeout = list_first_entry(list, struct io_timeout, list); + list_del_init(&timeout->list); + req = cmd_to_io_kiocb(timeout); + if (err) + req_set_fail(req); + io_req_queue_tw_complete(req, err); + } + + return true; +} + +static void io_kill_timeout(struct io_kiocb *req, struct list_head *list) __must_hold(&req->ctx->timeout_lock) { struct io_timeout_data *io = req->async_data; @@ -93,21 +113,17 @@ static bool io_kill_timeout(struct io_kiocb *req, int status) if (hrtimer_try_to_cancel(&io->timer) != -1) { struct io_timeout *timeout = io_kiocb_to_cmd(req, struct io_timeout); - if (status) - req_set_fail(req); atomic_set(&req->ctx->cq_timeouts, atomic_read(&req->ctx->cq_timeouts) + 1); - list_del_init(&timeout->list); - io_req_queue_tw_complete(req, status); - return true; + list_move_tail(&timeout->list, list); } - return false; } __cold void io_flush_timeouts(struct io_ring_ctx *ctx) { - u32 seq; struct io_timeout *timeout, *tmp; + LIST_HEAD(list); + u32 seq; raw_spin_lock_irq(&ctx->timeout_lock); seq = ctx->cached_cq_tail - atomic_read(&ctx->cq_timeouts); @@ -131,10 +147,11 @@ __cold void io_flush_timeouts(struct io_ring_ctx *ctx) if (events_got < events_needed) break; - io_kill_timeout(req, 0); + io_kill_timeout(req, &list); } ctx->cq_last_tm_flush = seq; raw_spin_unlock_irq(&ctx->timeout_lock); + io_flush_killed_timeouts(&list, 0); } static void io_req_tw_fail_links(struct io_kiocb *link, struct io_tw_state *ts) @@ -410,10 +427,12 @@ static int io_timeout_update(struct io_ring_ctx *ctx, __u64 user_data, timeout->off = 0; /* noseq */ data = req->async_data; + data->ts = *ts; + list_add_tail(&timeout->list, &ctx->timeout_list); hrtimer_init(&data->timer, io_timeout_get_clock(data), mode); data->timer.function = io_timeout_fn; - hrtimer_start(&data->timer, timespec64_to_ktime(*ts), mode); + hrtimer_start(&data->timer, timespec64_to_ktime(data->ts), mode); return 0; } @@ -661,7 +680,7 @@ __cold bool io_kill_timeouts(struct io_ring_ctx *ctx, struct io_uring_task *tctx bool cancel_all) { struct io_timeout *timeout, *tmp; - int canceled = 0; + LIST_HEAD(list); /* * completion_lock is needed for io_match_task(). Take it before @@ -672,11 +691,11 @@ __cold bool io_kill_timeouts(struct io_ring_ctx *ctx, struct io_uring_task *tctx list_for_each_entry_safe(timeout, tmp, &ctx->timeout_list, list) { struct io_kiocb *req = cmd_to_io_kiocb(timeout); - if (io_match_task(req, tctx, cancel_all) && - io_kill_timeout(req, -ECANCELED)) - canceled++; + if (io_match_task(req, tctx, cancel_all)) + io_kill_timeout(req, &list); } raw_spin_unlock_irq(&ctx->timeout_lock); spin_unlock(&ctx->completion_lock); - return canceled != 0; + + return io_flush_killed_timeouts(&list, -ECANCELED); } diff --git a/io_uring/uring_cmd.c b/io_uring/uring_cmd.c index af842e9b4eb9..ce7726a04883 100644 --- a/io_uring/uring_cmd.c +++ b/io_uring/uring_cmd.c @@ -16,26 +16,35 @@ #include "rsrc.h" #include "uring_cmd.h" -static struct uring_cache *io_uring_async_get(struct io_kiocb *req) +static struct io_uring_cmd_data *io_uring_async_get(struct io_kiocb *req) { struct io_ring_ctx *ctx = req->ctx; - struct uring_cache *cache; + struct io_uring_cmd_data *cache; cache = io_alloc_cache_get(&ctx->uring_cache); if (cache) { + cache->op_data = NULL; req->flags |= REQ_F_ASYNC_DATA; req->async_data = cache; return cache; } - if (!io_alloc_async_data(req)) - return req->async_data; + if (!io_alloc_async_data(req)) { + cache = req->async_data; + cache->op_data = NULL; + return cache; + } return NULL; } static void io_req_uring_cleanup(struct io_kiocb *req, unsigned int issue_flags) { struct io_uring_cmd *ioucmd = io_kiocb_to_cmd(req, struct io_uring_cmd); - struct uring_cache *cache = req->async_data; + struct io_uring_cmd_data *cache = req->async_data; + + if (cache->op_data) { + kfree(cache->op_data); + cache->op_data = NULL; + } if (issue_flags & IO_URING_F_UNLOCKED) return; @@ -183,7 +192,7 @@ static int io_uring_cmd_prep_setup(struct io_kiocb *req, const struct io_uring_sqe *sqe) { struct io_uring_cmd *ioucmd = io_kiocb_to_cmd(req, struct io_uring_cmd); - struct uring_cache *cache; + struct io_uring_cmd_data *cache; cache = io_uring_async_get(req); if (unlikely(!cache)) @@ -260,7 +269,7 @@ int io_uring_cmd(struct io_kiocb *req, unsigned int issue_flags) ret = file->f_op->uring_cmd(ioucmd, issue_flags); if (ret == -EAGAIN) { - struct uring_cache *cache = req->async_data; + struct io_uring_cmd_data *cache = req->async_data; if (ioucmd->sqe != (void *) cache) memcpy(cache, ioucmd->sqe, uring_sqe_size(req->ctx)); diff --git a/io_uring/uring_cmd.h b/io_uring/uring_cmd.h index 7dba0f1efc58..f6837ee0955b 100644 --- a/io_uring/uring_cmd.h +++ b/io_uring/uring_cmd.h @@ -1,9 +1,5 @@ // SPDX-License-Identifier: GPL-2.0 -struct uring_cache { - struct io_uring_sqe sqes[2]; -}; - int io_uring_cmd(struct io_kiocb *req, unsigned int issue_flags); int io_uring_cmd_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe); diff --git a/kernel/cgroup/cpuset.c b/kernel/cgroup/cpuset.c index f321ed515f3a..0f910c828973 100644 --- a/kernel/cgroup/cpuset.c +++ b/kernel/cgroup/cpuset.c @@ -197,10 +197,8 @@ static struct cpuset top_cpuset = { /* * There are two global locks guarding cpuset structures - cpuset_mutex and - * callback_lock. We also require taking task_lock() when dereferencing a - * task's cpuset pointer. See "The task_lock() exception", at the end of this - * comment. The cpuset code uses only cpuset_mutex. Other kernel subsystems - * can use cpuset_lock()/cpuset_unlock() to prevent change to cpuset + * callback_lock. The cpuset code uses only cpuset_mutex. Other kernel + * subsystems can use cpuset_lock()/cpuset_unlock() to prevent change to cpuset * structures. Note that cpuset_mutex needs to be a mutex as it is used in * paths that rely on priority inheritance (e.g. scheduler - on RT) for * correctness. @@ -229,9 +227,6 @@ static struct cpuset top_cpuset = { * The cpuset_common_seq_show() handlers only hold callback_lock across * small pieces of code, such as when reading out possibly multi-word * cpumasks and nodemasks. - * - * Accessing a task's cpuset should be done in accordance with the - * guidelines for accessing subsystem state in kernel/cgroup.c */ static DEFINE_MUTEX(cpuset_mutex); @@ -890,7 +885,15 @@ v2: */ if (cgrpv2) { for (i = 0; i < ndoms; i++) { - cpumask_copy(doms[i], csa[i]->effective_cpus); + /* + * The top cpuset may contain some boot time isolated + * CPUs that need to be excluded from the sched domain. + */ + if (csa[i] == &top_cpuset) + cpumask_and(doms[i], csa[i]->effective_cpus, + housekeeping_cpumask(HK_TYPE_DOMAIN)); + else + cpumask_copy(doms[i], csa[i]->effective_cpus); if (dattr) dattr[i] = SD_ATTR_INIT; } @@ -3121,29 +3124,6 @@ ssize_t cpuset_write_resmask(struct kernfs_open_file *of, int retval = -ENODEV; buf = strstrip(buf); - - /* - * CPU or memory hotunplug may leave @cs w/o any execution - * resources, in which case the hotplug code asynchronously updates - * configuration and transfers all tasks to the nearest ancestor - * which can execute. - * - * As writes to "cpus" or "mems" may restore @cs's execution - * resources, wait for the previously scheduled operations before - * proceeding, so that we don't end up keep removing tasks added - * after execution capability is restored. - * - * cpuset_handle_hotplug may call back into cgroup core asynchronously - * via cgroup_transfer_tasks() and waiting for it from a cgroupfs - * operation like this one can lead to a deadlock through kernfs - * active_ref protection. Let's break the protection. Losing the - * protection is okay as we check whether @cs is online after - * grabbing cpuset_mutex anyway. This only happens on the legacy - * hierarchies. - */ - css_get(&cs->css); - kernfs_break_active_protection(of->kn); - cpus_read_lock(); mutex_lock(&cpuset_mutex); if (!is_cpuset_online(cs)) @@ -3176,8 +3156,6 @@ ssize_t cpuset_write_resmask(struct kernfs_open_file *of, out_unlock: mutex_unlock(&cpuset_mutex); cpus_read_unlock(); - kernfs_unbreak_active_protection(of->kn); - css_put(&cs->css); flush_workqueue(cpuset_migrate_mm_wq); return retval ?: nbytes; } diff --git a/kernel/cpu.c b/kernel/cpu.c index b605334f8ee6..0509a9733745 100644 --- a/kernel/cpu.c +++ b/kernel/cpu.c @@ -2179,7 +2179,7 @@ static struct cpuhp_step cpuhp_hp_states[] = { }, [CPUHP_AP_HRTIMERS_DYING] = { .name = "hrtimers:dying", - .startup.single = NULL, + .startup.single = hrtimers_cpu_starting, .teardown.single = hrtimers_cpu_dying, }, [CPUHP_AP_TICK_DYING] = { diff --git a/kernel/cred.c b/kernel/cred.c index da7da250f7c8..9676965c0981 100644 --- a/kernel/cred.c +++ b/kernel/cred.c @@ -477,56 +477,6 @@ void abort_creds(struct cred *new) EXPORT_SYMBOL(abort_creds); /** - * override_creds - Override the current process's subjective credentials - * @new: The credentials to be assigned - * - * Install a set of temporary override subjective credentials on the current - * process, returning the old set for later reversion. - */ -const struct cred *override_creds(const struct cred *new) -{ - const struct cred *old; - - kdebug("override_creds(%p{%ld})", new, - atomic_long_read(&new->usage)); - - /* - * NOTE! This uses 'get_new_cred()' rather than 'get_cred()'. - * - * That means that we do not clear the 'non_rcu' flag, since - * we are only installing the cred into the thread-synchronous - * '->cred' pointer, not the '->real_cred' pointer that is - * visible to other threads under RCU. - */ - get_new_cred((struct cred *)new); - old = override_creds_light(new); - - kdebug("override_creds() = %p{%ld}", old, - atomic_long_read(&old->usage)); - return old; -} -EXPORT_SYMBOL(override_creds); - -/** - * revert_creds - Revert a temporary subjective credentials override - * @old: The credentials to be restored - * - * Revert a temporary set of override subjective credentials to an old set, - * discarding the override set. - */ -void revert_creds(const struct cred *old) -{ - const struct cred *override = current->cred; - - kdebug("revert_creds(%p{%ld})", old, - atomic_long_read(&old->usage)); - - revert_creds_light(old); - put_cred(override); -} -EXPORT_SYMBOL(revert_creds); - -/** * cred_fscmp - Compare two credentials with respect to filesystem access. * @a: The first credential * @b: The second credential diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c index fa04b14a7d72..5d71ef85420c 100644 --- a/kernel/events/uprobes.c +++ b/kernel/events/uprobes.c @@ -1915,6 +1915,7 @@ void uprobe_free_utask(struct task_struct *t) if (!utask) return; + t->utask = NULL; WARN_ON_ONCE(utask->active_uprobe || utask->xol_vaddr); timer_delete_sync(&utask->ri_timer); @@ -1924,7 +1925,6 @@ void uprobe_free_utask(struct task_struct *t) ri = free_ret_instance(ri, true /* cleanup_hprobe */); kfree(utask); - t->utask = NULL; } #define RI_TIMER_PERIOD (HZ / 10) /* 100 ms */ diff --git a/kernel/gen_kheaders.sh b/kernel/gen_kheaders.sh index 383fd43ac612..7e1340da5aca 100755 --- a/kernel/gen_kheaders.sh +++ b/kernel/gen_kheaders.sh @@ -89,6 +89,7 @@ find $cpio_dir -type f -print0 | # Create archive and try to normalize metadata for reproducibility. tar "${KBUILD_BUILD_TIMESTAMP:+--mtime=$KBUILD_BUILD_TIMESTAMP}" \ + --exclude=".__afs*" --exclude=".nfs*" \ --owner=0 --group=0 --sort=name --numeric-owner --mode=u=rw,go=r,a+X \ -I $XZ -cf $tarfile -C $cpio_dir/ . > /dev/null diff --git a/kernel/kcov.c b/kernel/kcov.c index 28a6be6e64fd..187ba1b80bda 100644 --- a/kernel/kcov.c +++ b/kernel/kcov.c @@ -166,7 +166,7 @@ static void kcov_remote_area_put(struct kcov_remote_area *area, * Unlike in_serving_softirq(), this function returns false when called during * a hardirq or an NMI that happened in the softirq context. */ -static inline bool in_softirq_really(void) +static __always_inline bool in_softirq_really(void) { return in_serving_softirq() && !in_hardirq() && !in_nmi(); } diff --git a/kernel/kthread.c b/kernel/kthread.c index a5ac612b1609..1eb6f62a9165 100644 --- a/kernel/kthread.c +++ b/kernel/kthread.c @@ -738,10 +738,11 @@ EXPORT_SYMBOL(kthread_stop_put); int kthreadd(void *unused) { + static const char comm[TASK_COMM_LEN] = "kthreadd"; struct task_struct *tsk = current; /* Setup a clean context for our children to inherit. */ - set_task_comm(tsk, "kthreadd"); + set_task_comm(tsk, comm); ignore_signals(tsk); set_cpus_allowed_ptr(tsk, housekeeping_cpumask(HK_TYPE_KTHREAD)); set_mems_allowed(node_states[N_MEMORY]); diff --git a/kernel/locking/rtmutex.c b/kernel/locking/rtmutex.c index e858de203eb6..697a56d3d949 100644 --- a/kernel/locking/rtmutex.c +++ b/kernel/locking/rtmutex.c @@ -1292,7 +1292,13 @@ static int __sched task_blocks_on_rt_mutex(struct rt_mutex_base *lock, */ get_task_struct(owner); + preempt_disable(); raw_spin_unlock_irq(&lock->wait_lock); + /* wake up any tasks on the wake_q before calling rt_mutex_adjust_prio_chain */ + wake_up_q(wake_q); + wake_q_init(wake_q); + preempt_enable(); + res = rt_mutex_adjust_prio_chain(owner, chwalk, lock, next_lock, waiter, task); @@ -1596,6 +1602,7 @@ static void __sched remove_waiter(struct rt_mutex_base *lock, * or TASK_UNINTERRUPTIBLE) * @timeout: the pre-initialized and started timer, or NULL for none * @waiter: the pre-initialized rt_mutex_waiter + * @wake_q: wake_q of tasks to wake when we drop the lock->wait_lock * * Must be called with lock->wait_lock held and interrupts disabled */ @@ -1603,7 +1610,8 @@ static int __sched rt_mutex_slowlock_block(struct rt_mutex_base *lock, struct ww_acquire_ctx *ww_ctx, unsigned int state, struct hrtimer_sleeper *timeout, - struct rt_mutex_waiter *waiter) + struct rt_mutex_waiter *waiter, + struct wake_q_head *wake_q) __releases(&lock->wait_lock) __acquires(&lock->wait_lock) { struct rt_mutex *rtm = container_of(lock, struct rt_mutex, rtmutex); @@ -1634,7 +1642,13 @@ static int __sched rt_mutex_slowlock_block(struct rt_mutex_base *lock, owner = rt_mutex_owner(lock); else owner = NULL; + preempt_disable(); raw_spin_unlock_irq(&lock->wait_lock); + if (wake_q) { + wake_up_q(wake_q); + wake_q_init(wake_q); + } + preempt_enable(); if (!owner || !rtmutex_spin_on_owner(lock, waiter, owner)) rt_mutex_schedule(); @@ -1708,7 +1722,7 @@ static int __sched __rt_mutex_slowlock(struct rt_mutex_base *lock, ret = task_blocks_on_rt_mutex(lock, waiter, current, ww_ctx, chwalk, wake_q); if (likely(!ret)) - ret = rt_mutex_slowlock_block(lock, ww_ctx, state, NULL, waiter); + ret = rt_mutex_slowlock_block(lock, ww_ctx, state, NULL, waiter, wake_q); if (likely(!ret)) { /* acquired the lock */ diff --git a/kernel/locking/rtmutex_api.c b/kernel/locking/rtmutex_api.c index 33ea31d6a7b3..191e4720e546 100644 --- a/kernel/locking/rtmutex_api.c +++ b/kernel/locking/rtmutex_api.c @@ -383,7 +383,7 @@ int __sched rt_mutex_wait_proxy_lock(struct rt_mutex_base *lock, raw_spin_lock_irq(&lock->wait_lock); /* sleep on the mutex */ set_current_state(TASK_INTERRUPTIBLE); - ret = rt_mutex_slowlock_block(lock, NULL, TASK_INTERRUPTIBLE, to, waiter); + ret = rt_mutex_slowlock_block(lock, NULL, TASK_INTERRUPTIBLE, to, waiter, NULL); /* * try_to_take_rt_mutex() sets the waiter bit unconditionally. We might * have to fix that up. diff --git a/kernel/pid.c b/kernel/pid.c index 115448e89c3e..3a10a7b6fcf8 100644 --- a/kernel/pid.c +++ b/kernel/pid.c @@ -43,6 +43,7 @@ #include <linux/sched/task.h> #include <linux/idr.h> #include <linux/pidfs.h> +#include <linux/seqlock.h> #include <net/sock.h> #include <uapi/linux/pidfd.h> @@ -60,15 +61,8 @@ struct pid init_struct_pid = { }, } }; -int pid_max = PID_MAX_DEFAULT; - -int pid_max_min = RESERVED_PIDS + 1; -int pid_max_max = PID_MAX_LIMIT; -/* - * Pseudo filesystems start inode numbering after one. We use Reserved - * PIDs as a natural offset. - */ -static u64 pidfs_ino = RESERVED_PIDS; +static int pid_max_min = RESERVED_PIDS + 1; +static int pid_max_max = PID_MAX_LIMIT; /* * PID-map pages start out as NULL, they get allocated upon @@ -87,6 +81,7 @@ struct pid_namespace init_pid_ns = { #ifdef CONFIG_PID_NS .ns.ops = &pidns_operations, #endif + .pid_max = PID_MAX_DEFAULT, #if defined(CONFIG_SYSCTL) && defined(CONFIG_MEMFD_CREATE) .memfd_noexec_scope = MEMFD_NOEXEC_SCOPE_EXEC, #endif @@ -108,6 +103,7 @@ EXPORT_SYMBOL_GPL(init_pid_ns); */ static __cacheline_aligned_in_smp DEFINE_SPINLOCK(pidmap_lock); +seqcount_spinlock_t pidmap_lock_seq = SEQCNT_SPINLOCK_ZERO(pidmap_lock_seq, &pidmap_lock); void put_pid(struct pid *pid) { @@ -158,6 +154,7 @@ void free_pid(struct pid *pid) idr_remove(&ns->idr, upid->nr); } + pidfs_remove_pid(pid); spin_unlock_irqrestore(&pidmap_lock, flags); call_rcu(&pid->rcu, delayed_put_pid); @@ -193,6 +190,7 @@ struct pid *alloc_pid(struct pid_namespace *ns, pid_t *set_tid, for (i = ns->level; i >= 0; i--) { int tid = 0; + int pid_max = READ_ONCE(tmp->pid_max); if (set_tid_size) { tid = set_tid[ns->level - i]; @@ -273,22 +271,24 @@ struct pid *alloc_pid(struct pid_namespace *ns, pid_t *set_tid, INIT_HLIST_HEAD(&pid->inodes); upid = pid->numbers + ns->level; + idr_preload(GFP_KERNEL); spin_lock_irq(&pidmap_lock); if (!(ns->pid_allocated & PIDNS_ADDING)) goto out_unlock; - pid->stashed = NULL; - pid->ino = ++pidfs_ino; + pidfs_add_pid(pid); for ( ; upid >= pid->numbers; --upid) { /* Make the PID visible to find_pid_ns. */ idr_replace(&upid->ns->idr, pid, upid->nr); upid->ns->pid_allocated++; } spin_unlock_irq(&pidmap_lock); + idr_preload_end(); return pid; out_unlock: spin_unlock_irq(&pidmap_lock); + idr_preload_end(); put_pid_ns(ns); out_free: @@ -644,17 +644,118 @@ SYSCALL_DEFINE2(pidfd_open, pid_t, pid, unsigned int, flags) return fd; } +#ifdef CONFIG_SYSCTL +static struct ctl_table_set *pid_table_root_lookup(struct ctl_table_root *root) +{ + return &task_active_pid_ns(current)->set; +} + +static int set_is_seen(struct ctl_table_set *set) +{ + return &task_active_pid_ns(current)->set == set; +} + +static int pid_table_root_permissions(struct ctl_table_header *head, + const struct ctl_table *table) +{ + struct pid_namespace *pidns = + container_of(head->set, struct pid_namespace, set); + int mode = table->mode; + + if (ns_capable(pidns->user_ns, CAP_SYS_ADMIN) || + uid_eq(current_euid(), make_kuid(pidns->user_ns, 0))) + mode = (mode & S_IRWXU) >> 6; + else if (in_egroup_p(make_kgid(pidns->user_ns, 0))) + mode = (mode & S_IRWXG) >> 3; + else + mode = mode & S_IROTH; + return (mode << 6) | (mode << 3) | mode; +} + +static void pid_table_root_set_ownership(struct ctl_table_header *head, + kuid_t *uid, kgid_t *gid) +{ + struct pid_namespace *pidns = + container_of(head->set, struct pid_namespace, set); + kuid_t ns_root_uid; + kgid_t ns_root_gid; + + ns_root_uid = make_kuid(pidns->user_ns, 0); + if (uid_valid(ns_root_uid)) + *uid = ns_root_uid; + + ns_root_gid = make_kgid(pidns->user_ns, 0); + if (gid_valid(ns_root_gid)) + *gid = ns_root_gid; +} + +static struct ctl_table_root pid_table_root = { + .lookup = pid_table_root_lookup, + .permissions = pid_table_root_permissions, + .set_ownership = pid_table_root_set_ownership, +}; + +static struct ctl_table pid_table[] = { + { + .procname = "pid_max", + .data = &init_pid_ns.pid_max, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = &pid_max_min, + .extra2 = &pid_max_max, + }, +}; +#endif + +int register_pidns_sysctls(struct pid_namespace *pidns) +{ +#ifdef CONFIG_SYSCTL + struct ctl_table *tbl; + + setup_sysctl_set(&pidns->set, &pid_table_root, set_is_seen); + + tbl = kmemdup(pid_table, sizeof(pid_table), GFP_KERNEL); + if (!tbl) + return -ENOMEM; + tbl->data = &pidns->pid_max; + pidns->pid_max = min(pid_max_max, max_t(int, pidns->pid_max, + PIDS_PER_CPU_DEFAULT * num_possible_cpus())); + + pidns->sysctls = __register_sysctl_table(&pidns->set, "kernel", tbl, + ARRAY_SIZE(pid_table)); + if (!pidns->sysctls) { + kfree(tbl); + retire_sysctl_set(&pidns->set); + return -ENOMEM; + } +#endif + return 0; +} + +void unregister_pidns_sysctls(struct pid_namespace *pidns) +{ +#ifdef CONFIG_SYSCTL + const struct ctl_table *tbl; + + tbl = pidns->sysctls->ctl_table_arg; + unregister_sysctl_table(pidns->sysctls); + retire_sysctl_set(&pidns->set); + kfree(tbl); +#endif +} + void __init pid_idr_init(void) { /* Verify no one has done anything silly: */ BUILD_BUG_ON(PID_MAX_LIMIT >= PIDNS_ADDING); /* bump default and minimum pid_max based on number of cpus */ - pid_max = min(pid_max_max, max_t(int, pid_max, - PIDS_PER_CPU_DEFAULT * num_possible_cpus())); + init_pid_ns.pid_max = min(pid_max_max, max_t(int, init_pid_ns.pid_max, + PIDS_PER_CPU_DEFAULT * num_possible_cpus())); pid_max_min = max_t(int, pid_max_min, PIDS_PER_CPU_MIN * num_possible_cpus()); - pr_info("pid_max: default: %u minimum: %u\n", pid_max, pid_max_min); + pr_info("pid_max: default: %u minimum: %u\n", init_pid_ns.pid_max, pid_max_min); idr_init(&init_pid_ns.idr); @@ -665,6 +766,16 @@ void __init pid_idr_init(void) NULL); } +static __init int pid_namespace_sysctl_init(void) +{ +#ifdef CONFIG_SYSCTL + /* "kernel" directory will have already been initialized. */ + BUG_ON(register_pidns_sysctls(&init_pid_ns)); +#endif + return 0; +} +subsys_initcall(pid_namespace_sysctl_init); + static struct file *__pidfd_fget(struct task_struct *task, int fd) { struct file *file; diff --git a/kernel/pid_namespace.c b/kernel/pid_namespace.c index d70ab49d5b4a..f1ffa032fc32 100644 --- a/kernel/pid_namespace.c +++ b/kernel/pid_namespace.c @@ -70,6 +70,8 @@ static void dec_pid_namespaces(struct ucounts *ucounts) dec_ucount(ucounts, UCOUNT_PID_NAMESPACES); } +static void destroy_pid_namespace_work(struct work_struct *work); + static struct pid_namespace *create_pid_namespace(struct user_namespace *user_ns, struct pid_namespace *parent_pid_ns) { @@ -105,17 +107,27 @@ static struct pid_namespace *create_pid_namespace(struct user_namespace *user_ns goto out_free_idr; ns->ns.ops = &pidns_operations; + ns->pid_max = parent_pid_ns->pid_max; + err = register_pidns_sysctls(ns); + if (err) + goto out_free_inum; + refcount_set(&ns->ns.count, 1); ns->level = level; ns->parent = get_pid_ns(parent_pid_ns); ns->user_ns = get_user_ns(user_ns); ns->ucounts = ucounts; ns->pid_allocated = PIDNS_ADDING; + INIT_WORK(&ns->work, destroy_pid_namespace_work); + #if defined(CONFIG_SYSCTL) && defined(CONFIG_MEMFD_CREATE) ns->memfd_noexec_scope = pidns_memfd_noexec_scope(parent_pid_ns); #endif + return ns; +out_free_inum: + ns_free_inum(&ns->ns); out_free_idr: idr_destroy(&ns->idr); kmem_cache_free(pid_ns_cachep, ns); @@ -137,12 +149,28 @@ static void delayed_free_pidns(struct rcu_head *p) static void destroy_pid_namespace(struct pid_namespace *ns) { + unregister_pidns_sysctls(ns); + ns_free_inum(&ns->ns); idr_destroy(&ns->idr); call_rcu(&ns->rcu, delayed_free_pidns); } +static void destroy_pid_namespace_work(struct work_struct *work) +{ + struct pid_namespace *ns = + container_of(work, struct pid_namespace, work); + + do { + struct pid_namespace *parent; + + parent = ns->parent; + destroy_pid_namespace(ns); + ns = parent; + } while (ns != &init_pid_ns && refcount_dec_and_test(&ns->ns.count)); +} + struct pid_namespace *copy_pid_ns(unsigned long flags, struct user_namespace *user_ns, struct pid_namespace *old_ns) { @@ -155,15 +183,8 @@ struct pid_namespace *copy_pid_ns(unsigned long flags, void put_pid_ns(struct pid_namespace *ns) { - struct pid_namespace *parent; - - while (ns != &init_pid_ns) { - parent = ns->parent; - if (!refcount_dec_and_test(&ns->ns.count)) - break; - destroy_pid_namespace(ns); - ns = parent; - } + if (ns && ns != &init_pid_ns && refcount_dec_and_test(&ns->ns.count)) + schedule_work(&ns->work); } EXPORT_SYMBOL_GPL(put_pid_ns); @@ -274,6 +295,7 @@ static int pid_ns_ctl_handler(const struct ctl_table *table, int write, next = idr_get_cursor(&pid_ns->idr) - 1; tmp.data = &next; + tmp.extra2 = &pid_ns->pid_max; ret = proc_dointvec_minmax(&tmp, write, buffer, lenp, ppos); if (!ret && write) idr_set_cursor(&pid_ns->idr, next + 1); @@ -281,7 +303,6 @@ static int pid_ns_ctl_handler(const struct ctl_table *table, int write, return ret; } -extern int pid_max; static struct ctl_table pid_ns_ctl_table[] = { { .procname = "ns_last_pid", @@ -289,7 +310,7 @@ static struct ctl_table pid_ns_ctl_table[] = { .mode = 0666, /* permissions are checked in the handler */ .proc_handler = pid_ns_ctl_handler, .extra1 = SYSCTL_ZERO, - .extra2 = &pid_max, + .extra2 = &init_pid_ns.pid_max, }, }; #endif /* CONFIG_CHECKPOINT_RESTORE */ diff --git a/kernel/sched/ext.c b/kernel/sched/ext.c index 7fff1d045477..19813b387ef9 100644 --- a/kernel/sched/ext.c +++ b/kernel/sched/ext.c @@ -2747,6 +2747,7 @@ static int balance_one(struct rq *rq, struct task_struct *prev) { struct scx_dsp_ctx *dspc = this_cpu_ptr(scx_dsp_ctx); bool prev_on_scx = prev->sched_class == &ext_sched_class; + bool prev_on_rq = prev->scx.flags & SCX_TASK_QUEUED; int nr_loops = SCX_DSP_MAX_LOOPS; lockdep_assert_rq_held(rq); @@ -2779,8 +2780,7 @@ static int balance_one(struct rq *rq, struct task_struct *prev) * See scx_ops_disable_workfn() for the explanation on the * bypassing test. */ - if ((prev->scx.flags & SCX_TASK_QUEUED) && - prev->scx.slice && !scx_rq_bypassing(rq)) { + if (prev_on_rq && prev->scx.slice && !scx_rq_bypassing(rq)) { rq->scx.flags |= SCX_RQ_BAL_KEEP; goto has_tasks; } @@ -2813,6 +2813,10 @@ static int balance_one(struct rq *rq, struct task_struct *prev) flush_dispatch_buf(rq); + if (prev_on_rq && prev->scx.slice) { + rq->scx.flags |= SCX_RQ_BAL_KEEP; + goto has_tasks; + } if (rq->scx.local_dsq.nr) goto has_tasks; if (consume_global_dsq(rq)) @@ -2838,8 +2842,7 @@ no_tasks: * Didn't find another task to run. Keep running @prev unless * %SCX_OPS_ENQ_LAST is in effect. */ - if ((prev->scx.flags & SCX_TASK_QUEUED) && - (!static_branch_unlikely(&scx_ops_enq_last) || + if (prev_on_rq && (!static_branch_unlikely(&scx_ops_enq_last) || scx_rq_bypassing(rq))) { rq->scx.flags |= SCX_RQ_BAL_KEEP; goto has_tasks; @@ -3034,7 +3037,7 @@ static void put_prev_task_scx(struct rq *rq, struct task_struct *p, */ if (p->scx.slice && !scx_rq_bypassing(rq)) { dispatch_enqueue(&rq->scx.local_dsq, p, SCX_ENQ_HEAD); - return; + goto switch_class; } /* @@ -3051,6 +3054,7 @@ static void put_prev_task_scx(struct rq *rq, struct task_struct *p, } } +switch_class: if (next && next->sched_class != &ext_sched_class) switch_class(rq, next); } @@ -3586,16 +3590,8 @@ static void reset_idle_masks(void) cpumask_copy(idle_masks.smt, cpu_online_mask); } -void __scx_update_idle(struct rq *rq, bool idle) +static void update_builtin_idle(int cpu, bool idle) { - int cpu = cpu_of(rq); - - if (SCX_HAS_OP(update_idle) && !scx_rq_bypassing(rq)) { - SCX_CALL_OP(SCX_KF_REST, update_idle, cpu_of(rq), idle); - if (!static_branch_unlikely(&scx_builtin_idle_enabled)) - return; - } - if (idle) cpumask_set_cpu(cpu, idle_masks.cpu); else @@ -3622,6 +3618,57 @@ void __scx_update_idle(struct rq *rq, bool idle) #endif } +/* + * Update the idle state of a CPU to @idle. + * + * If @do_notify is true, ops.update_idle() is invoked to notify the scx + * scheduler of an actual idle state transition (idle to busy or vice + * versa). If @do_notify is false, only the idle state in the idle masks is + * refreshed without invoking ops.update_idle(). + * + * This distinction is necessary, because an idle CPU can be "reserved" and + * awakened via scx_bpf_pick_idle_cpu() + scx_bpf_kick_cpu(), marking it as + * busy even if no tasks are dispatched. In this case, the CPU may return + * to idle without a true state transition. Refreshing the idle masks + * without invoking ops.update_idle() ensures accurate idle state tracking + * while avoiding unnecessary updates and maintaining balanced state + * transitions. + */ +void __scx_update_idle(struct rq *rq, bool idle, bool do_notify) +{ + int cpu = cpu_of(rq); + + lockdep_assert_rq_held(rq); + + /* + * Trigger ops.update_idle() only when transitioning from a task to + * the idle thread and vice versa. + * + * Idle transitions are indicated by do_notify being set to true, + * managed by put_prev_task_idle()/set_next_task_idle(). + */ + if (SCX_HAS_OP(update_idle) && do_notify && !scx_rq_bypassing(rq)) + SCX_CALL_OP(SCX_KF_REST, update_idle, cpu_of(rq), idle); + + /* + * Update the idle masks: + * - for real idle transitions (do_notify == true) + * - for idle-to-idle transitions (indicated by the previous task + * being the idle thread, managed by pick_task_idle()) + * + * Skip updating idle masks if the previous task is not the idle + * thread, since set_next_task_idle() has already handled it when + * transitioning from a task to the idle thread (calling this + * function with do_notify == true). + * + * In this way we can avoid updating the idle masks twice, + * unnecessarily. + */ + if (static_branch_likely(&scx_builtin_idle_enabled)) + if (do_notify || is_idle_task(rq->curr)) + update_builtin_idle(cpu, idle); +} + static void handle_hotplug(struct rq *rq, bool online) { int cpu = cpu_of(rq); @@ -4744,10 +4791,9 @@ static void scx_ops_bypass(bool bypass) */ for_each_possible_cpu(cpu) { struct rq *rq = cpu_rq(cpu); - struct rq_flags rf; struct task_struct *p, *n; - rq_lock(rq, &rf); + raw_spin_rq_lock(rq); if (bypass) { WARN_ON_ONCE(rq->scx.flags & SCX_RQ_BYPASSING); @@ -4763,7 +4809,7 @@ static void scx_ops_bypass(bool bypass) * sees scx_rq_bypassing() before moving tasks to SCX. */ if (!scx_enabled()) { - rq_unlock_irqrestore(rq, &rf); + raw_spin_rq_unlock(rq); continue; } @@ -4783,10 +4829,11 @@ static void scx_ops_bypass(bool bypass) sched_enq_and_set_task(&ctx); } - rq_unlock(rq, &rf); - /* resched to restore ticks and idle state */ - resched_cpu(cpu); + if (cpu_online(cpu) || cpu == smp_processor_id()) + resched_curr(rq); + + raw_spin_rq_unlock(rq); } atomic_dec(&scx_ops_breather_depth); @@ -7013,7 +7060,7 @@ __bpf_kfunc int bpf_iter_scx_dsq_new(struct bpf_iter_scx_dsq *it, u64 dsq_id, return -ENOENT; INIT_LIST_HEAD(&kit->cursor.node); - kit->cursor.flags |= SCX_DSQ_LNODE_ITER_CURSOR | flags; + kit->cursor.flags = SCX_DSQ_LNODE_ITER_CURSOR | flags; kit->cursor.priv = READ_ONCE(kit->dsq->seq); return 0; diff --git a/kernel/sched/ext.h b/kernel/sched/ext.h index b1675bb59fc4..4d022d17ac7d 100644 --- a/kernel/sched/ext.h +++ b/kernel/sched/ext.h @@ -57,15 +57,15 @@ static inline void init_sched_ext_class(void) {} #endif /* CONFIG_SCHED_CLASS_EXT */ #if defined(CONFIG_SCHED_CLASS_EXT) && defined(CONFIG_SMP) -void __scx_update_idle(struct rq *rq, bool idle); +void __scx_update_idle(struct rq *rq, bool idle, bool do_notify); -static inline void scx_update_idle(struct rq *rq, bool idle) +static inline void scx_update_idle(struct rq *rq, bool idle, bool do_notify) { if (scx_enabled()) - __scx_update_idle(rq, idle); + __scx_update_idle(rq, idle, do_notify); } #else -static inline void scx_update_idle(struct rq *rq, bool idle) {} +static inline void scx_update_idle(struct rq *rq, bool idle, bool do_notify) {} #endif #ifdef CONFIG_CGROUP_SCHED diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 3e9ca38512de..26958431deb7 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -689,21 +689,16 @@ u64 avg_vruntime(struct cfs_rq *cfs_rq) * * XXX could add max_slice to the augmented data to track this. */ -static s64 entity_lag(u64 avruntime, struct sched_entity *se) +static void update_entity_lag(struct cfs_rq *cfs_rq, struct sched_entity *se) { s64 vlag, limit; - vlag = avruntime - se->vruntime; - limit = calc_delta_fair(max_t(u64, 2*se->slice, TICK_NSEC), se); - - return clamp(vlag, -limit, limit); -} - -static void update_entity_lag(struct cfs_rq *cfs_rq, struct sched_entity *se) -{ SCHED_WARN_ON(!se->on_rq); - se->vlag = entity_lag(avg_vruntime(cfs_rq), se); + vlag = avg_vruntime(cfs_rq) - se->vruntime; + limit = calc_delta_fair(max_t(u64, 2*se->slice, TICK_NSEC), se); + + se->vlag = clamp(vlag, -limit, limit); } /* @@ -3774,137 +3769,32 @@ static inline void dequeue_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *se) { } #endif -static void reweight_eevdf(struct sched_entity *se, u64 avruntime, - unsigned long weight) -{ - unsigned long old_weight = se->load.weight; - s64 vlag, vslice; - - /* - * VRUNTIME - * -------- - * - * COROLLARY #1: The virtual runtime of the entity needs to be - * adjusted if re-weight at !0-lag point. - * - * Proof: For contradiction assume this is not true, so we can - * re-weight without changing vruntime at !0-lag point. - * - * Weight VRuntime Avg-VRuntime - * before w v V - * after w' v' V' - * - * Since lag needs to be preserved through re-weight: - * - * lag = (V - v)*w = (V'- v')*w', where v = v' - * ==> V' = (V - v)*w/w' + v (1) - * - * Let W be the total weight of the entities before reweight, - * since V' is the new weighted average of entities: - * - * V' = (WV + w'v - wv) / (W + w' - w) (2) - * - * by using (1) & (2) we obtain: - * - * (WV + w'v - wv) / (W + w' - w) = (V - v)*w/w' + v - * ==> (WV-Wv+Wv+w'v-wv)/(W+w'-w) = (V - v)*w/w' + v - * ==> (WV - Wv)/(W + w' - w) + v = (V - v)*w/w' + v - * ==> (V - v)*W/(W + w' - w) = (V - v)*w/w' (3) - * - * Since we are doing at !0-lag point which means V != v, we - * can simplify (3): - * - * ==> W / (W + w' - w) = w / w' - * ==> Ww' = Ww + ww' - ww - * ==> W * (w' - w) = w * (w' - w) - * ==> W = w (re-weight indicates w' != w) - * - * So the cfs_rq contains only one entity, hence vruntime of - * the entity @v should always equal to the cfs_rq's weighted - * average vruntime @V, which means we will always re-weight - * at 0-lag point, thus breach assumption. Proof completed. - * - * - * COROLLARY #2: Re-weight does NOT affect weighted average - * vruntime of all the entities. - * - * Proof: According to corollary #1, Eq. (1) should be: - * - * (V - v)*w = (V' - v')*w' - * ==> v' = V' - (V - v)*w/w' (4) - * - * According to the weighted average formula, we have: - * - * V' = (WV - wv + w'v') / (W - w + w') - * = (WV - wv + w'(V' - (V - v)w/w')) / (W - w + w') - * = (WV - wv + w'V' - Vw + wv) / (W - w + w') - * = (WV + w'V' - Vw) / (W - w + w') - * - * ==> V'*(W - w + w') = WV + w'V' - Vw - * ==> V' * (W - w) = (W - w) * V (5) - * - * If the entity is the only one in the cfs_rq, then reweight - * always occurs at 0-lag point, so V won't change. Or else - * there are other entities, hence W != w, then Eq. (5) turns - * into V' = V. So V won't change in either case, proof done. - * - * - * So according to corollary #1 & #2, the effect of re-weight - * on vruntime should be: - * - * v' = V' - (V - v) * w / w' (4) - * = V - (V - v) * w / w' - * = V - vl * w / w' - * = V - vl' - */ - if (avruntime != se->vruntime) { - vlag = entity_lag(avruntime, se); - vlag = div_s64(vlag * old_weight, weight); - se->vruntime = avruntime - vlag; - } - - /* - * DEADLINE - * -------- - * - * When the weight changes, the virtual time slope changes and - * we should adjust the relative virtual deadline accordingly. - * - * d' = v' + (d - v)*w/w' - * = V' - (V - v)*w/w' + (d - v)*w/w' - * = V - (V - v)*w/w' + (d - v)*w/w' - * = V + (d - V)*w/w' - */ - vslice = (s64)(se->deadline - avruntime); - vslice = div_s64(vslice * old_weight, weight); - se->deadline = avruntime + vslice; -} +static void place_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags); static void reweight_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, unsigned long weight) { bool curr = cfs_rq->curr == se; - u64 avruntime; if (se->on_rq) { /* commit outstanding execution time */ update_curr(cfs_rq); - avruntime = avg_vruntime(cfs_rq); + update_entity_lag(cfs_rq, se); + se->deadline -= se->vruntime; + se->rel_deadline = 1; if (!curr) __dequeue_entity(cfs_rq, se); update_load_sub(&cfs_rq->load, se->load.weight); } dequeue_load_avg(cfs_rq, se); - if (se->on_rq) { - reweight_eevdf(se, avruntime, weight); - } else { - /* - * Because we keep se->vlag = V - v_i, while: lag_i = w_i*(V - v_i), - * we need to scale se->vlag when w_i changes. - */ - se->vlag = div_s64(se->vlag * se->load.weight, weight); - } + /* + * Because we keep se->vlag = V - v_i, while: lag_i = w_i*(V - v_i), + * we need to scale se->vlag when w_i changes. + */ + se->vlag = div_s64(se->vlag * se->load.weight, weight); + if (se->rel_deadline) + se->deadline = div_s64(se->deadline * se->load.weight, weight); update_load_set(&se->load, weight); @@ -3919,6 +3809,7 @@ static void reweight_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, enqueue_load_avg(cfs_rq, se); if (se->on_rq) { update_load_add(&cfs_rq->load, se->load.weight); + place_entity(cfs_rq, se, 0); if (!curr) __enqueue_entity(cfs_rq, se); @@ -4065,7 +3956,11 @@ static void update_cfs_group(struct sched_entity *se) struct cfs_rq *gcfs_rq = group_cfs_rq(se); long shares; - if (!gcfs_rq) + /* + * When a group becomes empty, preserve its weight. This matters for + * DELAY_DEQUEUE. + */ + if (!gcfs_rq || !gcfs_rq->load.weight) return; if (throttled_hierarchy(gcfs_rq)) @@ -5359,7 +5254,7 @@ place_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) se->vruntime = vruntime - lag; - if (sched_feat(PLACE_REL_DEADLINE) && se->rel_deadline) { + if (se->rel_deadline) { se->deadline += se->vruntime; se->rel_deadline = 0; return; diff --git a/kernel/sched/idle.c b/kernel/sched/idle.c index 621696269584..2c85c86b455f 100644 --- a/kernel/sched/idle.c +++ b/kernel/sched/idle.c @@ -452,19 +452,20 @@ static void wakeup_preempt_idle(struct rq *rq, struct task_struct *p, int flags) static void put_prev_task_idle(struct rq *rq, struct task_struct *prev, struct task_struct *next) { dl_server_update_idle_time(rq, prev); - scx_update_idle(rq, false); + scx_update_idle(rq, false, true); } static void set_next_task_idle(struct rq *rq, struct task_struct *next, bool first) { update_idle_core(rq); - scx_update_idle(rq, true); + scx_update_idle(rq, true, true); schedstat_inc(rq->sched_goidle); next->se.exec_start = rq_clock_task(rq); } struct task_struct *pick_task_idle(struct rq *rq) { + scx_update_idle(rq, true, false); return rq->idle; } diff --git a/kernel/signal.c b/kernel/signal.c index 989b1cc9116a..a2afd54303f0 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -2007,11 +2007,22 @@ void posixtimer_send_sigqueue(struct k_itimer *tmr) if (!list_empty(&q->list)) { /* - * If task group is exiting with the signal already pending, - * wait for __exit_signal() to do its job. Otherwise if - * ignored, it's not supposed to be queued. Try to survive. + * The signal was ignored and blocked. The timer + * expiry queued it because blocked signals are + * queued independent of the ignored state. + * + * The unblocking set SIGPENDING, but the signal + * was not yet dequeued from the pending list. + * So prepare_signal() sees unblocked and ignored, + * which ends up here. Leave it queued like a + * regular signal. + * + * The same happens when the task group is exiting + * and the signal is already queued. + * prepare_signal() treats SIGNAL_GROUP_EXIT as + * ignored independent of its queued state. This + * gets cleaned up in __exit_signal(). */ - WARN_ON_ONCE(!(t->signal->flags & SIGNAL_GROUP_EXIT)); goto out; } @@ -2046,17 +2057,25 @@ void posixtimer_send_sigqueue(struct k_itimer *tmr) goto out; } - /* This should never happen and leaks a reference count */ - if (WARN_ON_ONCE(!hlist_unhashed(&tmr->ignored_list))) - hlist_del_init(&tmr->ignored_list); - if (unlikely(!list_empty(&q->list))) { /* This holds a reference count already */ result = TRACE_SIGNAL_ALREADY_PENDING; goto out; } - posixtimer_sigqueue_getref(q); + /* + * If the signal is on the ignore list, it got blocked after it was + * ignored earlier. But nothing lifted the ignore. Move it back to + * the pending list to be consistent with the regular signal + * handling. This already holds a reference count. + * + * If it's not on the ignore list acquire a reference count. + */ + if (likely(hlist_unhashed(&tmr->ignored_list))) + posixtimer_sigqueue_getref(q); + else + hlist_del_init(&tmr->ignored_list); + posixtimer_queue_sigqueue(q, t, tmr->it_pid_type); result = TRACE_SIGNAL_DELIVERED; out: diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 5c9202cb8f59..7ae7a4136855 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -1804,15 +1804,6 @@ static struct ctl_table kern_table[] = { }, #endif { - .procname = "pid_max", - .data = &pid_max, - .maxlen = sizeof (int), - .mode = 0644, - .proc_handler = proc_dointvec_minmax, - .extra1 = &pid_max_min, - .extra2 = &pid_max_max, - }, - { .procname = "panic_on_oops", .data = &panic_on_oops, .maxlen = sizeof(int), diff --git a/kernel/time/hrtimer.c b/kernel/time/hrtimer.c index 80fe3749d2db..030426c8c944 100644 --- a/kernel/time/hrtimer.c +++ b/kernel/time/hrtimer.c @@ -2202,6 +2202,15 @@ int hrtimers_prepare_cpu(unsigned int cpu) } cpu_base->cpu = cpu; + hrtimer_cpu_base_init_expiry_lock(cpu_base); + return 0; +} + +int hrtimers_cpu_starting(unsigned int cpu) +{ + struct hrtimer_cpu_base *cpu_base = this_cpu_ptr(&hrtimer_bases); + + /* Clear out any left over state from a CPU down operation */ cpu_base->active_bases = 0; cpu_base->hres_active = 0; cpu_base->hang_detected = 0; @@ -2210,7 +2219,6 @@ int hrtimers_prepare_cpu(unsigned int cpu) cpu_base->expires_next = KTIME_MAX; cpu_base->softirq_expires_next = KTIME_MAX; cpu_base->online = 1; - hrtimer_cpu_base_init_expiry_lock(cpu_base); return 0; } @@ -2286,5 +2294,6 @@ int hrtimers_cpu_dying(unsigned int dying_cpu) void __init hrtimers_init(void) { hrtimers_prepare_cpu(smp_processor_id()); + hrtimers_cpu_starting(smp_processor_id()); open_softirq(HRTIMER_SOFTIRQ, hrtimer_run_softirq); } diff --git a/kernel/time/timer_migration.c b/kernel/time/timer_migration.c index 8d57f7686bb0..066c9ddca4ec 100644 --- a/kernel/time/timer_migration.c +++ b/kernel/time/timer_migration.c @@ -534,8 +534,13 @@ static void __walk_groups(up_f up, struct tmigr_walk *data, break; child = group; - group = group->parent; + /* + * Pairs with the store release on group connection + * to make sure group initialization is visible. + */ + group = READ_ONCE(group->parent); data->childmask = child->groupmask; + WARN_ON_ONCE(!data->childmask); } while (group); } @@ -564,7 +569,7 @@ static struct tmigr_event *tmigr_next_groupevt(struct tmigr_group *group) while ((node = timerqueue_getnext(&group->events))) { evt = container_of(node, struct tmigr_event, nextevt); - if (!evt->ignore) { + if (!READ_ONCE(evt->ignore)) { WRITE_ONCE(group->next_expiry, evt->nextevt.expires); return evt; } @@ -660,7 +665,7 @@ static bool tmigr_active_up(struct tmigr_group *group, * lock is held while updating the ignore flag in idle path. So this * state change will not be lost. */ - group->groupevt.ignore = true; + WRITE_ONCE(group->groupevt.ignore, true); return walk_done; } @@ -721,6 +726,7 @@ bool tmigr_update_events(struct tmigr_group *group, struct tmigr_group *child, union tmigr_state childstate, groupstate; bool remote = data->remote; bool walk_done = false; + bool ignore; u64 nextexp; if (child) { @@ -739,11 +745,19 @@ bool tmigr_update_events(struct tmigr_group *group, struct tmigr_group *child, nextexp = child->next_expiry; evt = &child->groupevt; - evt->ignore = (nextexp == KTIME_MAX) ? true : false; + /* + * This can race with concurrent idle exit (activate). + * If the current writer wins, a useless remote expiration may + * be scheduled. If the activate wins, the event is properly + * ignored. + */ + ignore = (nextexp == KTIME_MAX) ? true : false; + WRITE_ONCE(evt->ignore, ignore); } else { nextexp = data->nextexp; first_childevt = evt = data->evt; + ignore = evt->ignore; /* * Walking the hierarchy is required in any case when a @@ -769,7 +783,7 @@ bool tmigr_update_events(struct tmigr_group *group, struct tmigr_group *child, * first event information of the group is updated properly and * also handled properly, so skip this fast return path. */ - if (evt->ignore && !remote && group->parent) + if (ignore && !remote && group->parent) return true; raw_spin_lock(&group->lock); @@ -783,7 +797,7 @@ bool tmigr_update_events(struct tmigr_group *group, struct tmigr_group *child, * queue when the expiry time changed only or when it could be ignored. */ if (timerqueue_node_queued(&evt->nextevt)) { - if ((evt->nextevt.expires == nextexp) && !evt->ignore) { + if ((evt->nextevt.expires == nextexp) && !ignore) { /* Make sure not to miss a new CPU event with the same expiry */ evt->cpu = first_childevt->cpu; goto check_toplvl; @@ -793,7 +807,7 @@ bool tmigr_update_events(struct tmigr_group *group, struct tmigr_group *child, WRITE_ONCE(group->next_expiry, KTIME_MAX); } - if (evt->ignore) { + if (ignore) { /* * When the next child event could be ignored (nextexp is * KTIME_MAX) and there was no remote timer handling before or @@ -1487,6 +1501,21 @@ static void tmigr_init_group(struct tmigr_group *group, unsigned int lvl, s.seq = 0; atomic_set(&group->migr_state, s.state); + /* + * If this is a new top-level, prepare its groupmask in advance. + * This avoids accidents where yet another new top-level is + * created in the future and made visible before the current groupmask. + */ + if (list_empty(&tmigr_level_list[lvl])) { + group->groupmask = BIT(0); + /* + * The previous top level has prepared its groupmask already, + * simply account it as the first child. + */ + if (lvl > 0) + group->num_children = 1; + } + timerqueue_init_head(&group->events); timerqueue_init(&group->groupevt.nextevt); group->groupevt.nextevt.expires = KTIME_MAX; @@ -1550,8 +1579,25 @@ static void tmigr_connect_child_parent(struct tmigr_group *child, raw_spin_lock_irq(&child->lock); raw_spin_lock_nested(&parent->lock, SINGLE_DEPTH_NESTING); - child->parent = parent; - child->groupmask = BIT(parent->num_children++); + if (activate) { + /* + * @child is the old top and @parent the new one. In this + * case groupmask is pre-initialized and @child already + * accounted, along with its new sibling corresponding to the + * CPU going up. + */ + WARN_ON_ONCE(child->groupmask != BIT(0) || parent->num_children != 2); + } else { + /* Adding @child for the CPU going up to @parent. */ + child->groupmask = BIT(parent->num_children++); + } + + /* + * Make sure parent initialization is visible before publishing it to a + * racing CPU entering/exiting idle. This RELEASE barrier enforces an + * address dependency that pairs with the READ_ONCE() in __walk_groups(). + */ + smp_store_release(&child->parent, parent); raw_spin_unlock(&parent->lock); raw_spin_unlock_irq(&child->lock); diff --git a/kernel/trace/fgraph.c b/kernel/trace/fgraph.c index ddedcb50917f..30e3ddc8a8a8 100644 --- a/kernel/trace/fgraph.c +++ b/kernel/trace/fgraph.c @@ -833,7 +833,7 @@ static unsigned long __ftrace_return_to_handler(struct fgraph_ret_regs *ret_regs #endif { for_each_set_bit(i, &bitmap, sizeof(bitmap) * BITS_PER_BYTE) { - struct fgraph_ops *gops = fgraph_array[i]; + struct fgraph_ops *gops = READ_ONCE(fgraph_array[i]); if (gops == &fgraph_stub) continue; diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 9b17efb1a87d..2e113f8b13a2 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -902,16 +902,13 @@ static void profile_graph_return(struct ftrace_graph_ret *trace, } static struct fgraph_ops fprofiler_ops = { - .ops = { - .flags = FTRACE_OPS_FL_INITIALIZED, - INIT_OPS_HASH(fprofiler_ops.ops) - }, .entryfunc = &profile_graph_entry, .retfunc = &profile_graph_return, }; static int register_ftrace_profiler(void) { + ftrace_ops_set_global_filter(&fprofiler_ops.ops); return register_ftrace_graph(&fprofiler_ops); } @@ -922,12 +919,11 @@ static void unregister_ftrace_profiler(void) #else static struct ftrace_ops ftrace_profile_ops __read_mostly = { .func = function_profile_call, - .flags = FTRACE_OPS_FL_INITIALIZED, - INIT_OPS_HASH(ftrace_profile_ops) }; static int register_ftrace_profiler(void) { + ftrace_ops_set_global_filter(&ftrace_profile_ops); return register_ftrace_function(&ftrace_profile_ops); } diff --git a/kernel/trace/pid_list.c b/kernel/trace/pid_list.c index 4966e6bbdf6f..c62b9b3cfb3d 100644 --- a/kernel/trace/pid_list.c +++ b/kernel/trace/pid_list.c @@ -414,7 +414,7 @@ struct trace_pid_list *trace_pid_list_alloc(void) int i; /* According to linux/thread.h, pids can be no bigger that 30 bits */ - WARN_ON_ONCE(pid_max > (1 << 30)); + WARN_ON_ONCE(init_pid_ns.pid_max > (1 << 30)); pid_list = kzalloc(sizeof(*pid_list), GFP_KERNEL); if (!pid_list) diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 957f941a08e7..b6e40e8791fa 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -4122,6 +4122,7 @@ print_trace_header(struct seq_file *m, struct trace_iterator *iter) preempt_model_none() ? "server" : preempt_model_voluntary() ? "desktop" : preempt_model_full() ? "preempt" : + preempt_model_lazy() ? "lazy" : preempt_model_rt() ? "preempt_rt" : "unknown", /* These are reserved for later use */ @@ -5087,6 +5088,9 @@ tracing_cpumask_write(struct file *filp, const char __user *ubuf, cpumask_var_t tracing_cpumask_new; int err; + if (count == 0 || count > KMALLOC_MAX_SIZE) + return -EINVAL; + if (!zalloc_cpumask_var(&tracing_cpumask_new, GFP_KERNEL)) return -ENOMEM; diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h index 9691b47b5f3d..179676db622e 100644 --- a/kernel/trace/trace.h +++ b/kernel/trace/trace.h @@ -717,8 +717,6 @@ extern unsigned long tracing_thresh; /* PID filtering */ -extern int pid_max; - bool trace_find_filtered_pid(struct trace_pid_list *filtered_pids, pid_t search_pid); bool trace_ignore_this_task(struct trace_pid_list *filtered_pids, diff --git a/kernel/trace/trace_events.c b/kernel/trace/trace_events.c index 1545cc8b49d0..770e7ed91716 100644 --- a/kernel/trace/trace_events.c +++ b/kernel/trace/trace_events.c @@ -365,6 +365,18 @@ static bool process_string(const char *fmt, int len, struct trace_event_call *ca } while (s < e); /* + * Check for arrays. If the argument has: foo[REC->val] + * then it is very likely that foo is an array of strings + * that are safe to use. + */ + r = strstr(s, "["); + if (r && r < e) { + r = strstr(r, "REC->"); + if (r && r < e) + return true; + } + + /* * If there's any strings in the argument consider this arg OK as it * could be: REC->field ? "foo" : "bar" and we don't want to get into * verifying that logic here. diff --git a/kernel/trace/trace_irqsoff.c b/kernel/trace/trace_irqsoff.c index fce064e20570..a4e799c1e767 100644 --- a/kernel/trace/trace_irqsoff.c +++ b/kernel/trace/trace_irqsoff.c @@ -182,6 +182,7 @@ static int irqsoff_graph_entry(struct ftrace_graph_ent *trace, struct trace_array_cpu *data; unsigned long flags; unsigned int trace_ctx; + u64 *calltime; int ret; if (ftrace_graph_ignore_func(gops, trace)) @@ -199,6 +200,12 @@ static int irqsoff_graph_entry(struct ftrace_graph_ent *trace, if (!func_prolog_dec(tr, &data, &flags)) return 0; + calltime = fgraph_reserve_data(gops->idx, sizeof(*calltime)); + if (!calltime) + return 0; + + *calltime = trace_clock_local(); + trace_ctx = tracing_gen_ctx_flags(flags); ret = __trace_graph_entry(tr, trace, trace_ctx); atomic_dec(&data->disabled); @@ -213,12 +220,19 @@ static void irqsoff_graph_return(struct ftrace_graph_ret *trace, struct trace_array_cpu *data; unsigned long flags; unsigned int trace_ctx; + u64 *calltime; + int size; ftrace_graph_addr_finish(gops, trace); if (!func_prolog_dec(tr, &data, &flags)) return; + calltime = fgraph_retrieve_data(gops->idx, &size); + if (!calltime) + return; + trace->calltime = *calltime; + trace_ctx = tracing_gen_ctx_flags(flags); __trace_graph_return(tr, trace, trace_ctx); atomic_dec(&data->disabled); diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c index 263fac44d3ca..0642ea174849 100644 --- a/kernel/trace/trace_kprobe.c +++ b/kernel/trace/trace_kprobe.c @@ -725,7 +725,7 @@ static int trace_kprobe_module_callback(struct notifier_block *nb, static struct notifier_block trace_kprobe_module_nb = { .notifier_call = trace_kprobe_module_callback, - .priority = 1 /* Invoked after kprobe module callback */ + .priority = 2 /* Invoked after kprobe and jump_label module callback */ }; static int trace_kprobe_register_module_notifier(void) { @@ -940,8 +940,10 @@ static int __trace_kprobe_create(int argc, const char *argv[]) } /* a symbol specified */ symbol = kstrdup(argv[1], GFP_KERNEL); - if (!symbol) - return -ENOMEM; + if (!symbol) { + ret = -ENOMEM; + goto error; + } tmp = strchr(symbol, '%'); if (tmp) { diff --git a/kernel/trace/trace_sched_switch.c b/kernel/trace/trace_sched_switch.c index 573b5d8e8a28..cb49f7279dc8 100644 --- a/kernel/trace/trace_sched_switch.c +++ b/kernel/trace/trace_sched_switch.c @@ -442,7 +442,7 @@ int trace_alloc_tgid_map(void) if (tgid_map) return 0; - tgid_map_max = pid_max; + tgid_map_max = init_pid_ns.pid_max; map = kvcalloc(tgid_map_max + 1, sizeof(*tgid_map), GFP_KERNEL); if (!map) diff --git a/kernel/trace/trace_sched_wakeup.c b/kernel/trace/trace_sched_wakeup.c index d6c7f18daa15..c58292e424d5 100644 --- a/kernel/trace/trace_sched_wakeup.c +++ b/kernel/trace/trace_sched_wakeup.c @@ -118,6 +118,7 @@ static int wakeup_graph_entry(struct ftrace_graph_ent *trace, struct trace_array *tr = wakeup_trace; struct trace_array_cpu *data; unsigned int trace_ctx; + u64 *calltime; int ret = 0; if (ftrace_graph_ignore_func(gops, trace)) @@ -135,6 +136,12 @@ static int wakeup_graph_entry(struct ftrace_graph_ent *trace, if (!func_prolog_preempt_disable(tr, &data, &trace_ctx)) return 0; + calltime = fgraph_reserve_data(gops->idx, sizeof(*calltime)); + if (!calltime) + return 0; + + *calltime = trace_clock_local(); + ret = __trace_graph_entry(tr, trace, trace_ctx); atomic_dec(&data->disabled); preempt_enable_notrace(); @@ -148,12 +155,19 @@ static void wakeup_graph_return(struct ftrace_graph_ret *trace, struct trace_array *tr = wakeup_trace; struct trace_array_cpu *data; unsigned int trace_ctx; + u64 *calltime; + int size; ftrace_graph_addr_finish(gops, trace); if (!func_prolog_preempt_disable(tr, &data, &trace_ctx)) return; + calltime = fgraph_retrieve_data(gops->idx, &size); + if (!calltime) + return; + trace->calltime = *calltime; + __trace_graph_return(tr, trace, trace_ctx); atomic_dec(&data->disabled); diff --git a/kernel/watch_queue.c b/kernel/watch_queue.c index 1895fbc32bcb..5267adeaa403 100644 --- a/kernel/watch_queue.c +++ b/kernel/watch_queue.c @@ -71,7 +71,7 @@ static void watch_queue_pipe_buf_release(struct pipe_inode_info *pipe, bit /= WATCH_QUEUE_NOTE_SIZE; page = buf->page; - bit += page->index; + bit += page->private; set_bit(bit, wqueue->notes_bitmap); generic_pipe_buf_release(pipe, buf); @@ -278,7 +278,7 @@ long watch_queue_set_size(struct pipe_inode_info *pipe, unsigned int nr_notes) pages[i] = alloc_page(GFP_KERNEL); if (!pages[i]) goto error_p; - pages[i]->index = i * WATCH_QUEUE_NOTES_PER_PAGE; + pages[i]->private = i * WATCH_QUEUE_NOTES_PER_PAGE; } bitmap = bitmap_alloc(nr_notes, GFP_KERNEL); diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 8b07576814a5..9362484a653c 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -2508,6 +2508,7 @@ static void __queue_delayed_work(int cpu, struct workqueue_struct *wq, return; } + WARN_ON_ONCE(cpu != WORK_CPU_UNBOUND && !cpu_online(cpu)); dwork->wq = wq; dwork->cpu = cpu; timer->expires = jiffies + delay; @@ -2533,6 +2534,12 @@ static void __queue_delayed_work(int cpu, struct workqueue_struct *wq, * @dwork: work to queue * @delay: number of jiffies to wait before queueing * + * We queue the delayed_work to a specific CPU, for non-zero delays the + * caller must ensure it is online and can't go away. Callers that fail + * to ensure this, may get @dwork->timer queued to an offlined CPU and + * this will prevent queueing of @dwork->work unless the offlined CPU + * becomes online again. + * * Return: %false if @work was already on a queue, %true otherwise. If * @delay is zero and @dwork is idle, it will be scheduled for immediate * execution. @@ -3680,23 +3687,27 @@ void workqueue_softirq_dead(unsigned int cpu) * check_flush_dependency - check for flush dependency sanity * @target_wq: workqueue being flushed * @target_work: work item being flushed (NULL for workqueue flushes) + * @from_cancel: are we called from the work cancel path * * %current is trying to flush the whole @target_wq or @target_work on it. - * If @target_wq doesn't have %WQ_MEM_RECLAIM, verify that %current is not - * reclaiming memory or running on a workqueue which doesn't have - * %WQ_MEM_RECLAIM as that can break forward-progress guarantee leading to - * a deadlock. + * If this is not the cancel path (which implies work being flushed is either + * already running, or will not be at all), check if @target_wq doesn't have + * %WQ_MEM_RECLAIM and verify that %current is not reclaiming memory or running + * on a workqueue which doesn't have %WQ_MEM_RECLAIM as that can break forward- + * progress guarantee leading to a deadlock. */ static void check_flush_dependency(struct workqueue_struct *target_wq, - struct work_struct *target_work) + struct work_struct *target_work, + bool from_cancel) { - work_func_t target_func = target_work ? target_work->func : NULL; + work_func_t target_func; struct worker *worker; - if (target_wq->flags & WQ_MEM_RECLAIM) + if (from_cancel || target_wq->flags & WQ_MEM_RECLAIM) return; worker = current_wq_worker(); + target_func = target_work ? target_work->func : NULL; WARN_ONCE(current->flags & PF_MEMALLOC, "workqueue: PF_MEMALLOC task %d(%s) is flushing !WQ_MEM_RECLAIM %s:%ps", @@ -3980,7 +3991,7 @@ void __flush_workqueue(struct workqueue_struct *wq) list_add_tail(&this_flusher.list, &wq->flusher_overflow); } - check_flush_dependency(wq, NULL); + check_flush_dependency(wq, NULL, false); mutex_unlock(&wq->mutex); @@ -4155,7 +4166,7 @@ static bool start_flush_work(struct work_struct *work, struct wq_barrier *barr, } wq = pwq->wq; - check_flush_dependency(wq, work); + check_flush_dependency(wq, work, from_cancel); insert_wq_barrier(pwq, barr, work, worker); raw_spin_unlock_irq(&pool->lock); @@ -5641,6 +5652,7 @@ static void wq_adjust_max_active(struct workqueue_struct *wq) } while (activated); } +__printf(1, 0) static struct workqueue_struct *__alloc_workqueue(const char *fmt, unsigned int flags, int max_active, va_list args) diff --git a/lib/alloc_tag.c b/lib/alloc_tag.c index 7dcebf118a3e..65e706e1bc19 100644 --- a/lib/alloc_tag.c +++ b/lib/alloc_tag.c @@ -195,6 +195,9 @@ void pgalloc_tag_swap(struct folio *new, struct folio *old) union codetag_ref ref_old, ref_new; struct alloc_tag *tag_old, *tag_new; + if (!mem_alloc_profiling_enabled()) + return; + tag_old = pgalloc_tag_get(&old->page); if (!tag_old) return; diff --git a/lib/kunit_iov_iter.c b/lib/kunit_iov_iter.c index 13e15687675a..10a560feb66e 100644 --- a/lib/kunit_iov_iter.c +++ b/lib/kunit_iov_iter.c @@ -392,7 +392,7 @@ static void __init iov_kunit_load_folioq(struct kunit *test, if (folioq_full(p)) { p->next = kzalloc(sizeof(struct folio_queue), GFP_KERNEL); KUNIT_ASSERT_NOT_ERR_OR_NULL(test, p->next); - folioq_init(p->next); + folioq_init(p->next, 0); p->next->prev = p; p = p->next; } @@ -409,7 +409,7 @@ static struct folio_queue *iov_kunit_create_folioq(struct kunit *test) folioq = kzalloc(sizeof(struct folio_queue), GFP_KERNEL); KUNIT_ASSERT_NOT_ERR_OR_NULL(test, folioq); kunit_add_action_or_reset(test, iov_kunit_destroy_folioq, folioq); - folioq_init(folioq); + folioq_init(folioq, 0); return folioq; } diff --git a/lib/lockref.c b/lib/lockref.c index 2afe4c5d8919..5d8e3ef3860e 100644 --- a/lib/lockref.c +++ b/lib/lockref.c @@ -58,23 +58,22 @@ EXPORT_SYMBOL(lockref_get); * @lockref: pointer to lockref structure * Return: 1 if count updated successfully or 0 if count was zero */ -int lockref_get_not_zero(struct lockref *lockref) +bool lockref_get_not_zero(struct lockref *lockref) { - int retval; + bool retval = false; CMPXCHG_LOOP( new.count++; if (old.count <= 0) - return 0; + return false; , - return 1; + return true; ); spin_lock(&lockref->lock); - retval = 0; if (lockref->count > 0) { lockref->count++; - retval = 1; + retval = true; } spin_unlock(&lockref->lock); return retval; @@ -82,39 +81,11 @@ int lockref_get_not_zero(struct lockref *lockref) EXPORT_SYMBOL(lockref_get_not_zero); /** - * lockref_put_not_zero - Decrements count unless count <= 1 before decrement - * @lockref: pointer to lockref structure - * Return: 1 if count updated successfully or 0 if count would become zero - */ -int lockref_put_not_zero(struct lockref *lockref) -{ - int retval; - - CMPXCHG_LOOP( - new.count--; - if (old.count <= 1) - return 0; - , - return 1; - ); - - spin_lock(&lockref->lock); - retval = 0; - if (lockref->count > 1) { - lockref->count--; - retval = 1; - } - spin_unlock(&lockref->lock); - return retval; -} -EXPORT_SYMBOL(lockref_put_not_zero); - -/** * lockref_put_return - Decrement reference count if possible * @lockref: pointer to lockref structure * * Decrement the reference count and return the new value. - * If the lockref was dead or locked, return an error. + * If the lockref was dead or locked, return -1. */ int lockref_put_return(struct lockref *lockref) { @@ -134,22 +105,22 @@ EXPORT_SYMBOL(lockref_put_return); * @lockref: pointer to lockref structure * Return: 1 if count updated successfully or 0 if count <= 1 and lock taken */ -int lockref_put_or_lock(struct lockref *lockref) +bool lockref_put_or_lock(struct lockref *lockref) { CMPXCHG_LOOP( new.count--; if (old.count <= 1) break; , - return 1; + return true; ); spin_lock(&lockref->lock); if (lockref->count <= 1) - return 0; + return false; lockref->count--; spin_unlock(&lockref->lock); - return 1; + return true; } EXPORT_SYMBOL(lockref_put_or_lock); @@ -169,23 +140,22 @@ EXPORT_SYMBOL(lockref_mark_dead); * @lockref: pointer to lockref structure * Return: 1 if count updated successfully or 0 if lockref was dead */ -int lockref_get_not_dead(struct lockref *lockref) +bool lockref_get_not_dead(struct lockref *lockref) { - int retval; + bool retval = false; CMPXCHG_LOOP( new.count++; if (old.count < 0) - return 0; + return false; , - return 1; + return true; ); spin_lock(&lockref->lock); - retval = 0; if (lockref->count >= 0) { lockref->count++; - retval = 1; + retval = true; } spin_unlock(&lockref->lock); return retval; diff --git a/lib/maple_tree.c b/lib/maple_tree.c index d0ae808f3a14..047397136f15 100644 --- a/lib/maple_tree.c +++ b/lib/maple_tree.c @@ -4354,6 +4354,7 @@ int mas_alloc_cyclic(struct ma_state *mas, unsigned long *startp, ret = 1; } if (ret < 0 && range_lo > min) { + mas_reset(mas); ret = mas_empty_area(mas, min, range_hi, 1); if (ret == 0) ret = 1; diff --git a/lib/vsprintf.c b/lib/vsprintf.c index 9d3dac38a3f4..56fe96319292 100644 --- a/lib/vsprintf.c +++ b/lib/vsprintf.c @@ -160,8 +160,7 @@ long long simple_strtoll(const char *cp, char **endp, unsigned int base) } EXPORT_SYMBOL(simple_strtoll); -static noinline_for_stack -int skip_atoi(const char **s) +static inline int skip_atoi(const char **s) { int i = 0; @@ -407,7 +406,7 @@ int num_to_str(char *buf, int size, unsigned long long num, unsigned int width) return len + width; } -#define SIGN 1 /* unsigned/signed, must be 1 */ +#define SIGN 1 /* unsigned/signed */ #define LEFT 2 /* left justified */ #define PLUS 4 /* show plus */ #define SPACE 8 /* space if plus */ @@ -415,38 +414,26 @@ int num_to_str(char *buf, int size, unsigned long long num, unsigned int width) #define SMALL 32 /* use lowercase in hex (must be 32 == 0x20) */ #define SPECIAL 64 /* prefix hex with "0x", octal with "0" */ -static_assert(SIGN == 1); static_assert(ZEROPAD == ('0' - ' ')); static_assert(SMALL == ('a' ^ 'A')); -enum format_type { - FORMAT_TYPE_NONE, /* Just a string part */ - FORMAT_TYPE_WIDTH, - FORMAT_TYPE_PRECISION, - FORMAT_TYPE_CHAR, - FORMAT_TYPE_STR, - FORMAT_TYPE_PTR, - FORMAT_TYPE_PERCENT_CHAR, - FORMAT_TYPE_INVALID, - FORMAT_TYPE_LONG_LONG, - FORMAT_TYPE_ULONG, - FORMAT_TYPE_LONG, - FORMAT_TYPE_UBYTE, - FORMAT_TYPE_BYTE, - FORMAT_TYPE_USHORT, - FORMAT_TYPE_SHORT, - FORMAT_TYPE_UINT, - FORMAT_TYPE_INT, - FORMAT_TYPE_SIZE_T, - FORMAT_TYPE_PTRDIFF +enum format_state { + FORMAT_STATE_NONE, /* Just a string part */ + FORMAT_STATE_NUM, + FORMAT_STATE_WIDTH, + FORMAT_STATE_PRECISION, + FORMAT_STATE_CHAR, + FORMAT_STATE_STR, + FORMAT_STATE_PTR, + FORMAT_STATE_PERCENT_CHAR, + FORMAT_STATE_INVALID, }; struct printf_spec { - unsigned int type:8; /* format_type enum */ - signed int field_width:24; /* width of output field */ - unsigned int flags:8; /* flags to number() */ - unsigned int base:8; /* number base, 8, 10 or 16 only */ - signed int precision:16; /* # of digits/chars */ + unsigned char flags; /* flags to number() */ + unsigned char base; /* number base, 8, 10 or 16 only */ + short precision; /* # of digits/chars */ + int field_width; /* width of output field */ } __packed; static_assert(sizeof(struct printf_spec) == 8); @@ -579,7 +566,6 @@ char *special_hex_number(char *buf, char *end, unsigned long long num, int size) { struct printf_spec spec; - spec.type = FORMAT_TYPE_PTR; spec.field_width = 2 + 2 * size; /* 0x + hex */ spec.flags = SPECIAL | SMALL | ZEROPAD; spec.base = 16; @@ -2530,6 +2516,26 @@ char *pointer(const char *fmt, char *buf, char *end, void *ptr, } } +struct fmt { + const char *str; + unsigned char state; // enum format_state + unsigned char size; // size of numbers +}; + +#define SPEC_CHAR(x, flag) [(x)-32] = flag +static unsigned char spec_flag(unsigned char c) +{ + static const unsigned char spec_flag_array[] = { + SPEC_CHAR(' ', SPACE), + SPEC_CHAR('#', SPECIAL), + SPEC_CHAR('+', PLUS), + SPEC_CHAR('-', LEFT), + SPEC_CHAR('0', ZEROPAD), + }; + c -= 32; + return (c < sizeof(spec_flag_array)) ? spec_flag_array[c] : 0; +} + /* * Helper function to decode printf style format. * Each call decode a token from the format and return the @@ -2552,181 +2558,141 @@ char *pointer(const char *fmt, char *buf, char *end, void *ptr, * @qualifier: qualifier of a number (long, size_t, ...) */ static noinline_for_stack -int format_decode(const char *fmt, struct printf_spec *spec) +struct fmt format_decode(struct fmt fmt, struct printf_spec *spec) { - const char *start = fmt; - char qualifier; + const char *start = fmt.str; + char flag; /* we finished early by reading the field width */ - if (spec->type == FORMAT_TYPE_WIDTH) { + if (unlikely(fmt.state == FORMAT_STATE_WIDTH)) { if (spec->field_width < 0) { spec->field_width = -spec->field_width; spec->flags |= LEFT; } - spec->type = FORMAT_TYPE_NONE; + fmt.state = FORMAT_STATE_NONE; goto precision; } /* we finished early by reading the precision */ - if (spec->type == FORMAT_TYPE_PRECISION) { + if (unlikely(fmt.state == FORMAT_STATE_PRECISION)) { if (spec->precision < 0) spec->precision = 0; - spec->type = FORMAT_TYPE_NONE; + fmt.state = FORMAT_STATE_NONE; goto qualifier; } /* By default */ - spec->type = FORMAT_TYPE_NONE; + fmt.state = FORMAT_STATE_NONE; - for (; *fmt ; ++fmt) { - if (*fmt == '%') + for (; *fmt.str ; fmt.str++) { + if (*fmt.str == '%') break; } /* Return the current non-format string */ - if (fmt != start || !*fmt) - return fmt - start; + if (fmt.str != start || !*fmt.str) + return fmt; - /* Process flags */ + /* Process flags. This also skips the first '%' */ spec->flags = 0; - - while (1) { /* this also skips first '%' */ - bool found = true; - - ++fmt; - - switch (*fmt) { - case '-': spec->flags |= LEFT; break; - case '+': spec->flags |= PLUS; break; - case ' ': spec->flags |= SPACE; break; - case '#': spec->flags |= SPECIAL; break; - case '0': spec->flags |= ZEROPAD; break; - default: found = false; - } - - if (!found) - break; - } + do { + /* this also skips first '%' */ + flag = spec_flag(*++fmt.str); + spec->flags |= flag; + } while (flag); /* get field width */ spec->field_width = -1; - if (isdigit(*fmt)) - spec->field_width = skip_atoi(&fmt); - else if (*fmt == '*') { + if (isdigit(*fmt.str)) + spec->field_width = skip_atoi(&fmt.str); + else if (unlikely(*fmt.str == '*')) { /* it's the next argument */ - spec->type = FORMAT_TYPE_WIDTH; - return ++fmt - start; + fmt.state = FORMAT_STATE_WIDTH; + fmt.str++; + return fmt; } precision: /* get the precision */ spec->precision = -1; - if (*fmt == '.') { - ++fmt; - if (isdigit(*fmt)) { - spec->precision = skip_atoi(&fmt); + if (unlikely(*fmt.str == '.')) { + fmt.str++; + if (isdigit(*fmt.str)) { + spec->precision = skip_atoi(&fmt.str); if (spec->precision < 0) spec->precision = 0; - } else if (*fmt == '*') { + } else if (*fmt.str == '*') { /* it's the next argument */ - spec->type = FORMAT_TYPE_PRECISION; - return ++fmt - start; + fmt.state = FORMAT_STATE_PRECISION; + fmt.str++; + return fmt; } } qualifier: - /* get the conversion qualifier */ - qualifier = 0; - if (*fmt == 'h' || _tolower(*fmt) == 'l' || - *fmt == 'z' || *fmt == 't') { - qualifier = *fmt++; - if (unlikely(qualifier == *fmt)) { - if (qualifier == 'l') { - qualifier = 'L'; - ++fmt; - } else if (qualifier == 'h') { - qualifier = 'H'; - ++fmt; - } - } - } - - /* default base */ + /* Set up default numeric format */ spec->base = 10; - switch (*fmt) { - case 'c': - spec->type = FORMAT_TYPE_CHAR; - return ++fmt - start; - - case 's': - spec->type = FORMAT_TYPE_STR; - return ++fmt - start; - - case 'p': - spec->type = FORMAT_TYPE_PTR; - return ++fmt - start; - - case '%': - spec->type = FORMAT_TYPE_PERCENT_CHAR; - return ++fmt - start; - - /* integer number formats - set up the flags and "break" */ - case 'o': - spec->base = 8; - break; - - case 'x': - spec->flags |= SMALL; - fallthrough; - - case 'X': - spec->base = 16; - break; + fmt.state = FORMAT_STATE_NUM; + fmt.size = sizeof(int); + static const struct format_state { + unsigned char state; + unsigned char size; + unsigned char flags_or_double_size; + unsigned char base; + } lookup_state[256] = { + // Length + ['l'] = { 0, sizeof(long), sizeof(long long) }, + ['L'] = { 0, sizeof(long long) }, + ['h'] = { 0, sizeof(short), sizeof(char) }, + ['H'] = { 0, sizeof(char) }, // Questionable historical + ['z'] = { 0, sizeof(size_t) }, + ['t'] = { 0, sizeof(ptrdiff_t) }, + + // Non-numeric formats + ['c'] = { FORMAT_STATE_CHAR }, + ['s'] = { FORMAT_STATE_STR }, + ['p'] = { FORMAT_STATE_PTR }, + ['%'] = { FORMAT_STATE_PERCENT_CHAR }, + + // Numerics + ['o'] = { FORMAT_STATE_NUM, 0, 0, 8 }, + ['x'] = { FORMAT_STATE_NUM, 0, SMALL, 16 }, + ['X'] = { FORMAT_STATE_NUM, 0, 0, 16 }, + ['d'] = { FORMAT_STATE_NUM, 0, SIGN, 10 }, + ['i'] = { FORMAT_STATE_NUM, 0, SIGN, 10 }, + ['u'] = { FORMAT_STATE_NUM, 0, 0, 10, }, - case 'd': - case 'i': - spec->flags |= SIGN; - break; - case 'u': - break; - - case 'n': /* * Since %n poses a greater security risk than * utility, treat it as any other invalid or * unsupported format specifier. */ - fallthrough; + }; - default: - WARN_ONCE(1, "Please remove unsupported %%%c in format string\n", *fmt); - spec->type = FORMAT_TYPE_INVALID; - return fmt - start; - } - - if (qualifier == 'L') - spec->type = FORMAT_TYPE_LONG_LONG; - else if (qualifier == 'l') { - BUILD_BUG_ON(FORMAT_TYPE_ULONG + SIGN != FORMAT_TYPE_LONG); - spec->type = FORMAT_TYPE_ULONG + (spec->flags & SIGN); - } else if (qualifier == 'z') { - spec->type = FORMAT_TYPE_SIZE_T; - } else if (qualifier == 't') { - spec->type = FORMAT_TYPE_PTRDIFF; - } else if (qualifier == 'H') { - BUILD_BUG_ON(FORMAT_TYPE_UBYTE + SIGN != FORMAT_TYPE_BYTE); - spec->type = FORMAT_TYPE_UBYTE + (spec->flags & SIGN); - } else if (qualifier == 'h') { - BUILD_BUG_ON(FORMAT_TYPE_USHORT + SIGN != FORMAT_TYPE_SHORT); - spec->type = FORMAT_TYPE_USHORT + (spec->flags & SIGN); - } else { - BUILD_BUG_ON(FORMAT_TYPE_UINT + SIGN != FORMAT_TYPE_INT); - spec->type = FORMAT_TYPE_UINT + (spec->flags & SIGN); + const struct format_state *p = lookup_state + (u8)*fmt.str; + if (p->size) { + fmt.size = p->size; + if (p->flags_or_double_size && fmt.str[0] == fmt.str[1]) { + fmt.size = p->flags_or_double_size; + fmt.str++; + } + fmt.str++; + p = lookup_state + *fmt.str; + } + if (p->state) { + if (p->base) + spec->base = p->base; + spec->flags |= p->flags_or_double_size; + fmt.state = p->state; + fmt.str++; + return fmt; } - return ++fmt - start; + WARN_ONCE(1, "Please remove unsupported %%%c in format string\n", *fmt.str); + fmt.state = FORMAT_STATE_INVALID; + return fmt; } static void @@ -2747,11 +2713,27 @@ set_precision(struct printf_spec *spec, int prec) } } +/* + * Turn a 1/2/4-byte value into a 64-bit one for printing: truncate + * as necessary and deal with signedness. + * + * 'size' is the size of the value in bytes. + */ +static unsigned long long convert_num_spec(unsigned int val, int size, struct printf_spec spec) +{ + unsigned int shift = 32 - size*8; + + val <<= shift; + if (!(spec.flags & SIGN)) + return val >> shift; + return (int)val >> shift; +} + /** * vsnprintf - Format a string and place it in a buffer * @buf: The buffer to place the result into * @size: The size of the buffer, including the trailing null space - * @fmt: The format string to use + * @fmt_str: The format string to use * @args: Arguments for the format string * * This function generally follows C99 vsnprintf, but has some @@ -2775,11 +2757,14 @@ set_precision(struct printf_spec *spec, int prec) * * If you're not already dealing with a va_list consider using snprintf(). */ -int vsnprintf(char *buf, size_t size, const char *fmt, va_list args) +int vsnprintf(char *buf, size_t size, const char *fmt_str, va_list args) { - unsigned long long num; char *str, *end; struct printf_spec spec = {0}; + struct fmt fmt = { + .str = fmt_str, + .state = FORMAT_STATE_NONE, + }; /* Reject out-of-range values early. Large positive sizes are used for unknown buffer sizes. */ @@ -2795,33 +2780,43 @@ int vsnprintf(char *buf, size_t size, const char *fmt, va_list args) size = end - buf; } - while (*fmt) { - const char *old_fmt = fmt; - int read = format_decode(fmt, &spec); + while (*fmt.str) { + const char *old_fmt = fmt.str; - fmt += read; + fmt = format_decode(fmt, &spec); - switch (spec.type) { - case FORMAT_TYPE_NONE: { - int copy = read; + switch (fmt.state) { + case FORMAT_STATE_NONE: { + int read = fmt.str - old_fmt; if (str < end) { + int copy = read; if (copy > end - str) copy = end - str; memcpy(str, old_fmt, copy); } str += read; - break; + continue; } - case FORMAT_TYPE_WIDTH: + case FORMAT_STATE_NUM: { + unsigned long long num; + if (fmt.size <= sizeof(int)) + num = convert_num_spec(va_arg(args, int), fmt.size, spec); + else + num = va_arg(args, long long); + str = number(str, end, num, spec); + continue; + } + + case FORMAT_STATE_WIDTH: set_field_width(&spec, va_arg(args, int)); - break; + continue; - case FORMAT_TYPE_PRECISION: + case FORMAT_STATE_PRECISION: set_precision(&spec, va_arg(args, int)); - break; + continue; - case FORMAT_TYPE_CHAR: { + case FORMAT_STATE_CHAR: { char c; if (!(spec.flags & LEFT)) { @@ -2841,27 +2836,27 @@ int vsnprintf(char *buf, size_t size, const char *fmt, va_list args) *str = ' '; ++str; } - break; + continue; } - case FORMAT_TYPE_STR: + case FORMAT_STATE_STR: str = string(str, end, va_arg(args, char *), spec); - break; + continue; - case FORMAT_TYPE_PTR: - str = pointer(fmt, str, end, va_arg(args, void *), + case FORMAT_STATE_PTR: + str = pointer(fmt.str, str, end, va_arg(args, void *), spec); - while (isalnum(*fmt)) - fmt++; - break; + while (isalnum(*fmt.str)) + fmt.str++; + continue; - case FORMAT_TYPE_PERCENT_CHAR: + case FORMAT_STATE_PERCENT_CHAR: if (str < end) *str = '%'; ++str; - break; + continue; - case FORMAT_TYPE_INVALID: + default: /* * Presumably the arguments passed gcc's type * checking, but there is no safe or sane way @@ -2871,47 +2866,6 @@ int vsnprintf(char *buf, size_t size, const char *fmt, va_list args) * sync. */ goto out; - - default: - switch (spec.type) { - case FORMAT_TYPE_LONG_LONG: - num = va_arg(args, long long); - break; - case FORMAT_TYPE_ULONG: - num = va_arg(args, unsigned long); - break; - case FORMAT_TYPE_LONG: - num = va_arg(args, long); - break; - case FORMAT_TYPE_SIZE_T: - if (spec.flags & SIGN) - num = va_arg(args, ssize_t); - else - num = va_arg(args, size_t); - break; - case FORMAT_TYPE_PTRDIFF: - num = va_arg(args, ptrdiff_t); - break; - case FORMAT_TYPE_UBYTE: - num = (unsigned char) va_arg(args, int); - break; - case FORMAT_TYPE_BYTE: - num = (signed char) va_arg(args, int); - break; - case FORMAT_TYPE_USHORT: - num = (unsigned short) va_arg(args, int); - break; - case FORMAT_TYPE_SHORT: - num = (short) va_arg(args, int); - break; - case FORMAT_TYPE_INT: - num = (int) va_arg(args, int); - break; - default: - num = va_arg(args, unsigned int); - } - - str = number(str, end, num, spec); } } @@ -3067,7 +3021,7 @@ EXPORT_SYMBOL(sprintf); * vbin_printf - Parse a format string and place args' binary value in a buffer * @bin_buf: The buffer to place args' binary value * @size: The size of the buffer(by words(32bits), not characters) - * @fmt: The format string to use + * @fmt_str: The format string to use * @args: Arguments for the format string * * The format follows C99 vsnprintf, except %n is ignored, and its argument @@ -3080,8 +3034,12 @@ EXPORT_SYMBOL(sprintf); * If the return value is greater than @size, the resulting bin_buf is NOT * valid for bstr_printf(). */ -int vbin_printf(u32 *bin_buf, size_t size, const char *fmt, va_list args) +int vbin_printf(u32 *bin_buf, size_t size, const char *fmt_str, va_list args) { + struct fmt fmt = { + .str = fmt_str, + .state = FORMAT_STATE_NONE, + }; struct printf_spec spec = {0}; char *str, *end; int width; @@ -3113,31 +3071,29 @@ int vbin_printf(u32 *bin_buf, size_t size, const char *fmt, va_list args) value; \ }) - while (*fmt) { - int read = format_decode(fmt, &spec); + while (*fmt.str) { + fmt = format_decode(fmt, &spec); - fmt += read; - - switch (spec.type) { - case FORMAT_TYPE_NONE: - case FORMAT_TYPE_PERCENT_CHAR: + switch (fmt.state) { + case FORMAT_STATE_NONE: + case FORMAT_STATE_PERCENT_CHAR: break; - case FORMAT_TYPE_INVALID: + case FORMAT_STATE_INVALID: goto out; - case FORMAT_TYPE_WIDTH: - case FORMAT_TYPE_PRECISION: + case FORMAT_STATE_WIDTH: + case FORMAT_STATE_PRECISION: width = (int)save_arg(int); /* Pointers may require the width */ - if (*fmt == 'p') + if (*fmt.str == 'p') set_field_width(&spec, width); break; - case FORMAT_TYPE_CHAR: + case FORMAT_STATE_CHAR: save_arg(char); break; - case FORMAT_TYPE_STR: { + case FORMAT_STATE_STR: { const char *save_str = va_arg(args, char *); const char *err_msg; size_t len; @@ -3153,9 +3109,9 @@ int vbin_printf(u32 *bin_buf, size_t size, const char *fmt, va_list args) break; } - case FORMAT_TYPE_PTR: + case FORMAT_STATE_PTR: /* Dereferenced pointers must be done now */ - switch (*fmt) { + switch (*fmt.str) { /* Dereference of functions is still OK */ case 'S': case 's': @@ -3165,11 +3121,11 @@ int vbin_printf(u32 *bin_buf, size_t size, const char *fmt, va_list args) save_arg(void *); break; default: - if (!isalnum(*fmt)) { + if (!isalnum(*fmt.str)) { save_arg(void *); break; } - str = pointer(fmt, str, end, va_arg(args, void *), + str = pointer(fmt.str, str, end, va_arg(args, void *), spec); if (str + 1 < end) *str++ = '\0'; @@ -3177,35 +3133,14 @@ int vbin_printf(u32 *bin_buf, size_t size, const char *fmt, va_list args) end[-1] = '\0'; /* Must be nul terminated */ } /* skip all alphanumeric pointer suffixes */ - while (isalnum(*fmt)) - fmt++; + while (isalnum(*fmt.str)) + fmt.str++; break; - default: - switch (spec.type) { - - case FORMAT_TYPE_LONG_LONG: + case FORMAT_STATE_NUM: + if (fmt.size > sizeof(int)) { save_arg(long long); - break; - case FORMAT_TYPE_ULONG: - case FORMAT_TYPE_LONG: - save_arg(unsigned long); - break; - case FORMAT_TYPE_SIZE_T: - save_arg(size_t); - break; - case FORMAT_TYPE_PTRDIFF: - save_arg(ptrdiff_t); - break; - case FORMAT_TYPE_UBYTE: - case FORMAT_TYPE_BYTE: - save_arg(char); - break; - case FORMAT_TYPE_USHORT: - case FORMAT_TYPE_SHORT: - save_arg(short); - break; - default: + } else { save_arg(int); } } @@ -3221,7 +3156,7 @@ EXPORT_SYMBOL_GPL(vbin_printf); * bstr_printf - Format a string from binary arguments and place it in a buffer * @buf: The buffer to place the result into * @size: The size of the buffer, including the trailing null space - * @fmt: The format string to use + * @fmt_str: The format string to use * @bin_buf: Binary arguments for the format string * * This function like C99 vsnprintf, but the difference is that vsnprintf gets @@ -3239,8 +3174,12 @@ EXPORT_SYMBOL_GPL(vbin_printf); * return is greater than or equal to @size, the resulting * string is truncated. */ -int bstr_printf(char *buf, size_t size, const char *fmt, const u32 *bin_buf) +int bstr_printf(char *buf, size_t size, const char *fmt_str, const u32 *bin_buf) { + struct fmt fmt = { + .str = fmt_str, + .state = FORMAT_STATE_NONE, + }; struct printf_spec spec = {0}; char *str, *end; const char *args = (const char *)bin_buf; @@ -3272,33 +3211,33 @@ int bstr_printf(char *buf, size_t size, const char *fmt, const u32 *bin_buf) size = end - buf; } - while (*fmt) { - const char *old_fmt = fmt; - int read = format_decode(fmt, &spec); - - fmt += read; + while (*fmt.str) { + const char *old_fmt = fmt.str; + unsigned long long num; - switch (spec.type) { - case FORMAT_TYPE_NONE: { - int copy = read; + fmt = format_decode(fmt, &spec); + switch (fmt.state) { + case FORMAT_STATE_NONE: { + int read = fmt.str - old_fmt; if (str < end) { + int copy = read; if (copy > end - str) copy = end - str; memcpy(str, old_fmt, copy); } str += read; - break; + continue; } - case FORMAT_TYPE_WIDTH: + case FORMAT_STATE_WIDTH: set_field_width(&spec, get_arg(int)); - break; + continue; - case FORMAT_TYPE_PRECISION: + case FORMAT_STATE_PRECISION: set_precision(&spec, get_arg(int)); - break; + continue; - case FORMAT_TYPE_CHAR: { + case FORMAT_STATE_CHAR: { char c; if (!(spec.flags & LEFT)) { @@ -3317,21 +3256,21 @@ int bstr_printf(char *buf, size_t size, const char *fmt, const u32 *bin_buf) *str = ' '; ++str; } - break; + continue; } - case FORMAT_TYPE_STR: { + case FORMAT_STATE_STR: { const char *str_arg = args; args += strlen(str_arg) + 1; str = string(str, end, (char *)str_arg, spec); - break; + continue; } - case FORMAT_TYPE_PTR: { + case FORMAT_STATE_PTR: { bool process = false; int copy, len; /* Non function dereferences were already done */ - switch (*fmt) { + switch (*fmt.str) { case 'S': case 's': case 'x': @@ -3340,7 +3279,7 @@ int bstr_printf(char *buf, size_t size, const char *fmt, const u32 *bin_buf) process = true; break; default: - if (!isalnum(*fmt)) { + if (!isalnum(*fmt.str)) { process = true; break; } @@ -3355,63 +3294,32 @@ int bstr_printf(char *buf, size_t size, const char *fmt, const u32 *bin_buf) } } if (process) - str = pointer(fmt, str, end, get_arg(void *), spec); + str = pointer(fmt.str, str, end, get_arg(void *), spec); - while (isalnum(*fmt)) - fmt++; - break; + while (isalnum(*fmt.str)) + fmt.str++; + continue; } - case FORMAT_TYPE_PERCENT_CHAR: + case FORMAT_STATE_PERCENT_CHAR: if (str < end) *str = '%'; ++str; - break; + continue; - case FORMAT_TYPE_INVALID: + case FORMAT_STATE_INVALID: goto out; - default: { - unsigned long long num; - - switch (spec.type) { - - case FORMAT_TYPE_LONG_LONG: + case FORMAT_STATE_NUM: + if (fmt.size > sizeof(int)) { num = get_arg(long long); - break; - case FORMAT_TYPE_ULONG: - case FORMAT_TYPE_LONG: - num = get_arg(unsigned long); - break; - case FORMAT_TYPE_SIZE_T: - num = get_arg(size_t); - break; - case FORMAT_TYPE_PTRDIFF: - num = get_arg(ptrdiff_t); - break; - case FORMAT_TYPE_UBYTE: - num = get_arg(unsigned char); - break; - case FORMAT_TYPE_BYTE: - num = get_arg(signed char); - break; - case FORMAT_TYPE_USHORT: - num = get_arg(unsigned short); - break; - case FORMAT_TYPE_SHORT: - num = get_arg(short); - break; - case FORMAT_TYPE_UINT: - num = get_arg(unsigned int); - break; - default: - num = get_arg(int); + } else { + num = convert_num_spec(get_arg(int), fmt.size, spec); } - str = number(str, end, num, spec); - } /* default: */ - } /* switch(spec.type) */ - } /* while(*fmt) */ + continue; + } + } /* while(*fmt.str) */ out: if (size > 0) { diff --git a/mm/damon/core.c b/mm/damon/core.c index 8b8e2933dcd4..0776452a1abb 100644 --- a/mm/damon/core.c +++ b/mm/damon/core.c @@ -868,6 +868,11 @@ static int damon_commit_schemes(struct damon_ctx *dst, struct damon_ctx *src) NUMA_NO_NODE); if (!new_scheme) return -ENOMEM; + err = damos_commit(new_scheme, src_scheme); + if (err) { + damon_destroy_scheme(new_scheme); + return err; + } damon_add_scheme(dst, new_scheme); } return 0; @@ -961,8 +966,11 @@ static int damon_commit_targets( return -ENOMEM; err = damon_commit_target(new_target, false, src_target, damon_target_has_pid(src)); - if (err) + if (err) { + damon_destroy_target(new_target); return err; + } + damon_add_target(dst, new_target); } return 0; } diff --git a/mm/filemap.c b/mm/filemap.c index f61cf51c2238..4f476411a9a2 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -124,15 +124,6 @@ * ->private_lock (zap_pte_range->block_dirty_folio) */ -static void mapping_set_update(struct xa_state *xas, - struct address_space *mapping) -{ - if (dax_mapping(mapping) || shmem_mapping(mapping)) - return; - xas_set_update(xas, workingset_update_node); - xas_set_lru(xas, &shadow_nodes); -} - static void page_cache_delete(struct address_space *mapping, struct folio *folio, void *shadow) { @@ -1532,7 +1523,7 @@ void folio_end_read(struct folio *folio, bool success) /* Must be in bottom byte for x86 to work */ BUILD_BUG_ON(PG_uptodate > 7); VM_BUG_ON_FOLIO(!folio_test_locked(folio), folio); - VM_BUG_ON_FOLIO(folio_test_uptodate(folio), folio); + VM_BUG_ON_FOLIO(success && folio_test_uptodate(folio), folio); if (likely(success)) mask |= 1 << PG_uptodate; @@ -3005,7 +2996,7 @@ static inline loff_t folio_seek_hole_data(struct xa_state *xas, if (ops->is_partially_uptodate(folio, offset, bsz) == seek_data) break; - start = (start + bsz) & ~(bsz - 1); + start = (start + bsz) & ~((u64)bsz - 1); offset += bsz; } while (offset < folio_size(folio)); unlock: diff --git a/mm/huge_memory.c b/mm/huge_memory.c index e53d83b3e5cf..db64116a4f84 100644 --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -2206,6 +2206,16 @@ static pmd_t move_soft_dirty_pmd(pmd_t pmd) return pmd; } +static pmd_t clear_uffd_wp_pmd(pmd_t pmd) +{ + if (pmd_present(pmd)) + pmd = pmd_clear_uffd_wp(pmd); + else if (is_swap_pmd(pmd)) + pmd = pmd_swp_clear_uffd_wp(pmd); + + return pmd; +} + bool move_huge_pmd(struct vm_area_struct *vma, unsigned long old_addr, unsigned long new_addr, pmd_t *old_pmd, pmd_t *new_pmd) { @@ -2244,6 +2254,8 @@ bool move_huge_pmd(struct vm_area_struct *vma, unsigned long old_addr, pgtable_trans_huge_deposit(mm, new_pmd, pgtable); } pmd = move_soft_dirty_pmd(pmd); + if (vma_has_uffd_without_event_remap(vma)) + pmd = clear_uffd_wp_pmd(pmd); set_pmd_at(mm, new_addr, new_pmd, pmd); if (force_flush) flush_pmd_tlb_range(vma, old_addr, old_addr + PMD_SIZE); diff --git a/mm/hugetlb.c b/mm/hugetlb.c index cec4b121193f..eaaec19caa7c 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -5402,6 +5402,7 @@ static void move_huge_pte(struct vm_area_struct *vma, unsigned long old_addr, unsigned long new_addr, pte_t *src_pte, pte_t *dst_pte, unsigned long sz) { + bool need_clear_uffd_wp = vma_has_uffd_without_event_remap(vma); struct hstate *h = hstate_vma(vma); struct mm_struct *mm = vma->vm_mm; spinlock_t *src_ptl, *dst_ptl; @@ -5418,7 +5419,18 @@ static void move_huge_pte(struct vm_area_struct *vma, unsigned long old_addr, spin_lock_nested(src_ptl, SINGLE_DEPTH_NESTING); pte = huge_ptep_get_and_clear(mm, old_addr, src_pte); - set_huge_pte_at(mm, new_addr, dst_pte, pte, sz); + + if (need_clear_uffd_wp && pte_marker_uffd_wp(pte)) + huge_pte_clear(mm, new_addr, dst_pte, sz); + else { + if (need_clear_uffd_wp) { + if (pte_present(pte)) + pte = huge_pte_clear_uffd_wp(pte); + else if (is_swap_pte(pte)) + pte = pte_swp_clear_uffd_wp(pte); + } + set_huge_pte_at(mm, new_addr, dst_pte, pte, sz); + } if (src_ptl != dst_ptl) spin_unlock(src_ptl); @@ -7211,7 +7223,7 @@ pte_t *huge_pmd_share(struct mm_struct *mm, struct vm_area_struct *vma, spte = hugetlb_walk(svma, saddr, vma_mmu_pagesize(svma)); if (spte) { - get_page(virt_to_page(spte)); + ptdesc_pmd_pts_inc(virt_to_ptdesc(spte)); break; } } @@ -7226,7 +7238,7 @@ pte_t *huge_pmd_share(struct mm_struct *mm, struct vm_area_struct *vma, (pmd_t *)((unsigned long)spte & PAGE_MASK)); mm_inc_nr_pmds(mm); } else { - put_page(virt_to_page(spte)); + ptdesc_pmd_pts_dec(virt_to_ptdesc(spte)); } spin_unlock(&mm->page_table_lock); out: @@ -7238,10 +7250,6 @@ out: /* * unmap huge page backed by shared pte. * - * Hugetlb pte page is ref counted at the time of mapping. If pte is shared - * indicated by page_count > 1, unmap is achieved by clearing pud and - * decrementing the ref count. If count == 1, the pte page is not shared. - * * Called with page table lock held. * * returns: 1 successfully unmapped a shared pte page @@ -7250,18 +7258,20 @@ out: int huge_pmd_unshare(struct mm_struct *mm, struct vm_area_struct *vma, unsigned long addr, pte_t *ptep) { + unsigned long sz = huge_page_size(hstate_vma(vma)); pgd_t *pgd = pgd_offset(mm, addr); p4d_t *p4d = p4d_offset(pgd, addr); pud_t *pud = pud_offset(p4d, addr); i_mmap_assert_write_locked(vma->vm_file->f_mapping); hugetlb_vma_assert_locked(vma); - BUG_ON(page_count(virt_to_page(ptep)) == 0); - if (page_count(virt_to_page(ptep)) == 1) + if (sz != PMD_SIZE) + return 0; + if (!ptdesc_pmd_pts_count(virt_to_ptdesc(ptep))) return 0; pud_clear(pud); - put_page(virt_to_page(ptep)); + ptdesc_pmd_pts_dec(virt_to_ptdesc(ptep)); mm_dec_nr_pmds(mm); return 1; } diff --git a/mm/internal.h b/mm/internal.h index 3bd08bafad04..9826f7dce607 100644 --- a/mm/internal.h +++ b/mm/internal.h @@ -1504,6 +1504,12 @@ static inline void shrinker_debugfs_remove(struct dentry *debugfs_entry, /* Only track the nodes of mappings with shadow entries */ void workingset_update_node(struct xa_node *node); extern struct list_lru shadow_nodes; +#define mapping_set_update(xas, mapping) do { \ + if (!dax_mapping(mapping) && !shmem_mapping(mapping)) { \ + xas_set_update(xas, workingset_update_node); \ + xas_set_lru(xas, &shadow_nodes); \ + } \ +} while (0) /* mremap.c */ unsigned long move_page_tables(struct vm_area_struct *vma, diff --git a/mm/khugepaged.c b/mm/khugepaged.c index 6f8d46d107b4..bad1e130eda8 100644 --- a/mm/khugepaged.c +++ b/mm/khugepaged.c @@ -19,6 +19,7 @@ #include <linux/rcupdate_wait.h> #include <linux/swapops.h> #include <linux/shmem_fs.h> +#include <linux/dax.h> #include <linux/ksm.h> #include <asm/tlb.h> @@ -1837,6 +1838,8 @@ static int collapse_file(struct mm_struct *mm, unsigned long addr, if (result != SCAN_SUCCEED) goto out; + mapping_set_update(&xas, mapping); + __folio_set_locked(new_folio); if (is_shmem) __folio_set_swapbacked(new_folio); @@ -2419,7 +2422,7 @@ skip: VM_BUG_ON(khugepaged_scan.address < hstart || khugepaged_scan.address + HPAGE_PMD_SIZE > hend); - if (IS_ENABLED(CONFIG_SHMEM) && vma->vm_file) { + if (IS_ENABLED(CONFIG_SHMEM) && !vma_is_anonymous(vma)) { struct file *file = get_file(vma->vm_file); pgoff_t pgoff = linear_page_index(vma, khugepaged_scan.address); @@ -2765,7 +2768,7 @@ int madvise_collapse(struct vm_area_struct *vma, struct vm_area_struct **prev, mmap_assert_locked(mm); memset(cc->node_load, 0, sizeof(cc->node_load)); nodes_clear(cc->alloc_nmask); - if (IS_ENABLED(CONFIG_SHMEM) && vma->vm_file) { + if (IS_ENABLED(CONFIG_SHMEM) && !vma_is_anonymous(vma)) { struct file *file = get_file(vma->vm_file); pgoff_t pgoff = linear_page_index(vma, addr); diff --git a/mm/kmemleak.c b/mm/kmemleak.c index 2a945c07ae99..820ba3b5cbfc 100644 --- a/mm/kmemleak.c +++ b/mm/kmemleak.c @@ -373,7 +373,7 @@ static void print_unreferenced(struct seq_file *seq, for (i = 0; i < nr_entries; i++) { void *ptr = (void *)entries[i]; - warn_or_seq_printf(seq, " [<%pK>] %pS\n", ptr, ptr); + warn_or_seq_printf(seq, " %pS\n", ptr); } } @@ -1093,7 +1093,7 @@ void __ref kmemleak_alloc_percpu(const void __percpu *ptr, size_t size, pr_debug("%s(0x%px, %zu)\n", __func__, ptr, size); if (kmemleak_enabled && ptr && !IS_ERR_PCPU(ptr)) - create_object_percpu((__force unsigned long)ptr, size, 0, gfp); + create_object_percpu((__force unsigned long)ptr, size, 1, gfp); } EXPORT_SYMBOL_GPL(kmemleak_alloc_percpu); diff --git a/mm/list_lru.c b/mm/list_lru.c index f93ada6a207b..7d69434c70e0 100644 --- a/mm/list_lru.c +++ b/mm/list_lru.c @@ -77,7 +77,6 @@ again: spin_lock(&l->lock); nr_items = READ_ONCE(l->nr_items); if (likely(nr_items != LONG_MIN)) { - WARN_ON(nr_items < 0); rcu_read_unlock(); return l; } @@ -450,6 +449,7 @@ static void memcg_reparent_list_lru_one(struct list_lru *lru, int nid, list_splice_init(&src->list, &dst->list); if (src->nr_items) { + WARN_ON(src->nr_items < 0); dst->nr_items += src->nr_items; set_shrinker_bit(dst_memcg, nid, lru_shrinker_id(lru)); } diff --git a/mm/memfd.c b/mm/memfd.c index c17c3ea701a1..35a370d75c9a 100644 --- a/mm/memfd.c +++ b/mm/memfd.c @@ -170,7 +170,7 @@ static int memfd_wait_for_pins(struct address_space *mapping) return error; } -static unsigned int *memfd_file_seals_ptr(struct file *file) +unsigned int *memfd_file_seals_ptr(struct file *file) { if (shmem_file(file)) return &SHMEM_I(file_inode(file))->seals; diff --git a/mm/mempolicy.c b/mm/mempolicy.c index 04f35659717a..162407fbf2bc 100644 --- a/mm/mempolicy.c +++ b/mm/mempolicy.c @@ -2268,7 +2268,8 @@ struct page *alloc_pages_mpol_noprof(gfp_t gfp, unsigned int order, page = __alloc_pages_noprof(gfp, order, nid, nodemask); - if (unlikely(pol->mode == MPOL_INTERLEAVE) && page) { + if (unlikely(pol->mode == MPOL_INTERLEAVE || + pol->mode == MPOL_WEIGHTED_INTERLEAVE) && page) { /* skip NUMA_INTERLEAVE_HIT update if numa stats is disabled */ if (static_branch_likely(&vm_numa_stat_key) && page_to_nid(page) == nid) { diff --git a/mm/mmap.c b/mm/mmap.c index d32b7e701058..aec208f90337 100644 --- a/mm/mmap.c +++ b/mm/mmap.c @@ -47,6 +47,7 @@ #include <linux/oom.h> #include <linux/sched/mm.h> #include <linux/ksm.h> +#include <linux/memfd.h> #include <linux/uaccess.h> #include <asm/cacheflush.h> @@ -368,6 +369,7 @@ unsigned long do_mmap(struct file *file, unsigned long addr, if (file) { struct inode *inode = file_inode(file); + unsigned int seals = memfd_file_seals(file); unsigned long flags_mask; if (!file_mmap_ok(file, inode, pgoff, len)) @@ -408,6 +410,8 @@ unsigned long do_mmap(struct file *file, unsigned long addr, vm_flags |= VM_SHARED | VM_MAYSHARE; if (!(file->f_mode & FMODE_WRITE)) vm_flags &= ~(VM_MAYWRITE | VM_SHARED); + else if (is_readonly_sealed(seals, vm_flags)) + vm_flags &= ~VM_MAYWRITE; fallthrough; case MAP_PRIVATE: if (!(file->f_mode & FMODE_READ)) @@ -888,7 +892,7 @@ __get_unmapped_area(struct file *file, unsigned long addr, unsigned long len, if (get_area) { addr = get_area(file, addr, len, pgoff, flags); - } else if (IS_ENABLED(CONFIG_TRANSPARENT_HUGEPAGE) + } else if (IS_ENABLED(CONFIG_TRANSPARENT_HUGEPAGE) && !file && !addr /* no hint */ && IS_ALIGNED(len, PMD_SIZE)) { /* Ensures that larger anonymous mappings are THP aligned. */ diff --git a/mm/mremap.c b/mm/mremap.c index 60473413836b..cff7f552f909 100644 --- a/mm/mremap.c +++ b/mm/mremap.c @@ -138,6 +138,7 @@ static int move_ptes(struct vm_area_struct *vma, pmd_t *old_pmd, struct vm_area_struct *new_vma, pmd_t *new_pmd, unsigned long new_addr, bool need_rmap_locks) { + bool need_clear_uffd_wp = vma_has_uffd_without_event_remap(vma); struct mm_struct *mm = vma->vm_mm; pte_t *old_pte, *new_pte, pte; pmd_t dummy_pmdval; @@ -216,7 +217,18 @@ static int move_ptes(struct vm_area_struct *vma, pmd_t *old_pmd, force_flush = true; pte = move_pte(pte, old_addr, new_addr); pte = move_soft_dirty_pte(pte); - set_pte_at(mm, new_addr, new_pte, pte); + + if (need_clear_uffd_wp && pte_marker_uffd_wp(pte)) + pte_clear(mm, new_addr, new_pte); + else { + if (need_clear_uffd_wp) { + if (pte_present(pte)) + pte = pte_clear_uffd_wp(pte); + else if (is_swap_pte(pte)) + pte = pte_swp_clear_uffd_wp(pte); + } + set_pte_at(mm, new_addr, new_pte, pte); + } } arch_leave_lazy_mmu_mode(); @@ -278,6 +290,15 @@ static bool move_normal_pmd(struct vm_area_struct *vma, unsigned long old_addr, if (WARN_ON_ONCE(!pmd_none(*new_pmd))) return false; + /* If this pmd belongs to a uffd vma with remap events disabled, we need + * to ensure that the uffd-wp state is cleared from all pgtables. This + * means recursing into lower page tables in move_page_tables(), and we + * can reuse the existing code if we simply treat the entry as "not + * moved". + */ + if (vma_has_uffd_without_event_remap(vma)) + return false; + /* * We don't have to worry about the ordering of src and dst * ptlocks because exclusive mmap_lock prevents deadlock. @@ -333,6 +354,15 @@ static bool move_normal_pud(struct vm_area_struct *vma, unsigned long old_addr, if (WARN_ON_ONCE(!pud_none(*new_pud))) return false; + /* If this pud belongs to a uffd vma with remap events disabled, we need + * to ensure that the uffd-wp state is cleared from all pgtables. This + * means recursing into lower page tables in move_page_tables(), and we + * can reuse the existing code if we simply treat the entry as "not + * moved". + */ + if (vma_has_uffd_without_event_remap(vma)) + return false; + /* * We don't have to worry about the ordering of src and dst * ptlocks because exclusive mmap_lock prevents deadlock. diff --git a/mm/page-writeback.c b/mm/page-writeback.c index d213ead95675..d9861e42b2bd 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -692,6 +692,8 @@ static unsigned long bdi_ratio_from_pages(unsigned long pages) unsigned long ratio; global_dirty_limits(&background_thresh, &dirty_thresh); + if (!dirty_thresh) + return -EINVAL; ratio = div64_u64(pages * 100ULL * BDI_RATIO_SCALE, dirty_thresh); return ratio; @@ -790,13 +792,15 @@ int bdi_set_min_bytes(struct backing_dev_info *bdi, u64 min_bytes) { int ret; unsigned long pages = min_bytes >> PAGE_SHIFT; - unsigned long min_ratio; + long min_ratio; ret = bdi_check_pages_limit(pages); if (ret) return ret; min_ratio = bdi_ratio_from_pages(pages); + if (min_ratio < 0) + return min_ratio; return __bdi_set_min_ratio(bdi, min_ratio); } @@ -809,13 +813,15 @@ int bdi_set_max_bytes(struct backing_dev_info *bdi, u64 max_bytes) { int ret; unsigned long pages = max_bytes >> PAGE_SHIFT; - unsigned long max_ratio; + long max_ratio; ret = bdi_check_pages_limit(pages); if (ret) return ret; max_ratio = bdi_ratio_from_pages(pages); + if (max_ratio < 0) + return max_ratio; return __bdi_set_max_ratio(bdi, max_ratio); } diff --git a/mm/page_alloc.c b/mm/page_alloc.c index cae7b93864c2..01eab25edf89 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -5692,10 +5692,13 @@ __meminit void zone_pcp_init(struct zone *zone) zone->present_pages, zone_batchsize(zone)); } +static void setup_per_zone_lowmem_reserve(void); + void adjust_managed_page_count(struct page *page, long count) { atomic_long_add(count, &page_zone(page)->managed_pages); totalram_pages_add(count); + setup_per_zone_lowmem_reserve(); } EXPORT_SYMBOL(adjust_managed_page_count); diff --git a/mm/readahead.c b/mm/readahead.c index ea650b8b02fb..e151f4b13ca4 100644 --- a/mm/readahead.c +++ b/mm/readahead.c @@ -646,7 +646,11 @@ void page_cache_async_ra(struct readahead_control *ractl, 1UL << order); if (index == expected) { ra->start += ra->size; - ra->size = get_next_ra_size(ra, max_pages); + /* + * In the case of MADV_HUGEPAGE, the actual size might exceed + * the readahead window. + */ + ra->size = max(ra->size, get_next_ra_size(ra, max_pages)); ra->async_size = ra->size; goto readit; } diff --git a/mm/shmem.c b/mm/shmem.c index f6fb053ac50d..532afd8e049c 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -1535,7 +1535,7 @@ try_split: !shmem_falloc->waitq && index >= shmem_falloc->start && index < shmem_falloc->next) - shmem_falloc->nr_unswapped++; + shmem_falloc->nr_unswapped += nr_pages; else shmem_falloc = NULL; spin_unlock(&inode->i_lock); @@ -1689,6 +1689,7 @@ unsigned long shmem_allowable_huge_orders(struct inode *inode, unsigned long mask = READ_ONCE(huge_shmem_orders_always); unsigned long within_size_orders = READ_ONCE(huge_shmem_orders_within_size); unsigned long vm_flags = vma ? vma->vm_flags : 0; + pgoff_t aligned_index; bool global_huge; loff_t i_size; int order; @@ -1723,9 +1724,9 @@ unsigned long shmem_allowable_huge_orders(struct inode *inode, /* Allow mTHP that will be fully within i_size. */ order = highest_order(within_size_orders); while (within_size_orders) { - index = round_up(index + 1, order); + aligned_index = round_up(index + 1, 1 << order); i_size = round_up(i_size_read(inode), PAGE_SIZE); - if (i_size >> PAGE_SHIFT >= index) { + if (i_size >> PAGE_SHIFT >= aligned_index) { mask |= within_size_orders; break; } @@ -3820,7 +3821,7 @@ static int shmem_unlink(struct inode *dir, struct dentry *dentry) static int shmem_rmdir(struct inode *dir, struct dentry *dentry) { - if (!simple_offset_empty(dentry)) + if (!simple_empty(dentry)) return -ENOTEMPTY; drop_nlink(d_inode(dentry)); @@ -3877,7 +3878,7 @@ static int shmem_rename2(struct mnt_idmap *idmap, return simple_offset_rename_exchange(old_dir, old_dentry, new_dir, new_dentry); - if (!simple_offset_empty(new_dentry)) + if (!simple_empty(new_dentry)) return -ENOTEMPTY; if (flags & RENAME_WHITEOUT) { @@ -3916,6 +3917,7 @@ static int shmem_symlink(struct mnt_idmap *idmap, struct inode *dir, int len; struct inode *inode; struct folio *folio; + char *link; len = strlen(symname) + 1; if (len > PAGE_SIZE) @@ -3937,12 +3939,13 @@ static int shmem_symlink(struct mnt_idmap *idmap, struct inode *dir, inode->i_size = len-1; if (len <= SHORT_SYMLINK_LEN) { - inode->i_link = kmemdup(symname, len, GFP_KERNEL); - if (!inode->i_link) { + link = kmemdup(symname, len, GFP_KERNEL); + if (!link) { error = -ENOMEM; goto out_remove_offset; } inode->i_op = &shmem_short_symlink_operations; + inode_set_cached_link(inode, link, len - 1); } else { inode_nohighmem(inode); inode->i_mapping->a_ops = &shmem_aops; @@ -4367,7 +4370,7 @@ static int shmem_parse_opt_casefold(struct fs_context *fc, struct fs_parameter * bool latest_version) { struct shmem_options *ctx = fc->fs_private; - unsigned int version = UTF8_LATEST; + int version = UTF8_LATEST; struct unicode_map *encoding; char *version_str = param->string + 5; diff --git a/mm/util.c b/mm/util.c index c1c3b06ab4f9..60aa40f612b8 100644 --- a/mm/util.c +++ b/mm/util.c @@ -297,12 +297,7 @@ void *memdup_user_nul(const void __user *src, size_t len) { char *p; - /* - * Always use GFP_KERNEL, since copy_from_user() can sleep and - * cause pagefault, which makes it pointless to use GFP_NOFS - * or GFP_ATOMIC. - */ - p = kmalloc_track_caller(len + 1, GFP_KERNEL); + p = kmem_buckets_alloc_track_caller(user_buckets, len + 1, GFP_USER | __GFP_NOWARN); if (!p) return ERR_PTR(-ENOMEM); diff --git a/mm/vmscan.c b/mm/vmscan.c index 76378bc257e3..b1ec5ece067e 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -374,7 +374,14 @@ unsigned long zone_reclaimable_pages(struct zone *zone) if (can_reclaim_anon_pages(NULL, zone_to_nid(zone), NULL)) nr += zone_page_state_snapshot(zone, NR_ZONE_INACTIVE_ANON) + zone_page_state_snapshot(zone, NR_ZONE_ACTIVE_ANON); - + /* + * If there are no reclaimable file-backed or anonymous pages, + * ensure zones with sufficient free pages are not skipped. + * This prevents zones like DMA32 from being ignored in reclaim + * scenarios where they can still help alleviate memory pressure. + */ + if (nr == 0) + nr = zone_page_state_snapshot(zone, NR_FREE_PAGES); return nr; } @@ -4635,6 +4642,9 @@ retry: reset_batch_size(walk); } + __mod_lruvec_state(lruvec, PGDEMOTE_KSWAPD + reclaimer_offset(), + stat.nr_demoted); + item = PGSTEAL_KSWAPD + reclaimer_offset(); if (!cgroup_reclaim(sc)) __count_vm_events(item, reclaimed); diff --git a/mm/vmstat.c b/mm/vmstat.c index 4d016314a56c..16bfe1c694dd 100644 --- a/mm/vmstat.c +++ b/mm/vmstat.c @@ -2122,10 +2122,20 @@ static void __init start_shepherd_timer(void) { int cpu; - for_each_possible_cpu(cpu) + for_each_possible_cpu(cpu) { INIT_DEFERRABLE_WORK(per_cpu_ptr(&vmstat_work, cpu), vmstat_update); + /* + * For secondary CPUs during CPU hotplug scenarios, + * vmstat_cpu_online() will enable the work. + * mm/vmstat:online enables and disables vmstat_work + * symmetrically during CPU hotplug events. + */ + if (!cpu_online(cpu)) + disable_delayed_work_sync(&per_cpu(vmstat_work, cpu)); + } + schedule_delayed_work(&shepherd, round_jiffies_relative(sysctl_stat_interval)); } @@ -2148,13 +2158,14 @@ static int vmstat_cpu_online(unsigned int cpu) if (!node_state(cpu_to_node(cpu), N_CPU)) { node_set_state(cpu_to_node(cpu), N_CPU); } + enable_delayed_work(&per_cpu(vmstat_work, cpu)); return 0; } static int vmstat_cpu_down_prep(unsigned int cpu) { - cancel_delayed_work_sync(&per_cpu(vmstat_work, cpu)); + disable_delayed_work_sync(&per_cpu(vmstat_work, cpu)); return 0; } diff --git a/mm/zswap.c b/mm/zswap.c index f6316b66fb23..b84c20d889b1 100644 --- a/mm/zswap.c +++ b/mm/zswap.c @@ -251,7 +251,7 @@ static struct zswap_pool *zswap_pool_create(char *type, char *compressor) struct zswap_pool *pool; char name[38]; /* 'zswap' + 32 char (max) num + \0 */ gfp_t gfp = __GFP_NORETRY | __GFP_NOWARN | __GFP_KSWAPD_RECLAIM; - int ret; + int ret, cpu; if (!zswap_has_pool) { /* if either are unset, pool initialization failed, and we @@ -285,6 +285,9 @@ static struct zswap_pool *zswap_pool_create(char *type, char *compressor) goto error; } + for_each_possible_cpu(cpu) + mutex_init(&per_cpu_ptr(pool->acomp_ctx, cpu)->mutex); + ret = cpuhp_state_add_instance(CPUHP_MM_ZSWP_POOL_PREPARE, &pool->node); if (ret) @@ -817,36 +820,41 @@ static int zswap_cpu_comp_prepare(unsigned int cpu, struct hlist_node *node) { struct zswap_pool *pool = hlist_entry(node, struct zswap_pool, node); struct crypto_acomp_ctx *acomp_ctx = per_cpu_ptr(pool->acomp_ctx, cpu); - struct crypto_acomp *acomp; - struct acomp_req *req; + struct crypto_acomp *acomp = NULL; + struct acomp_req *req = NULL; + u8 *buffer = NULL; int ret; - mutex_init(&acomp_ctx->mutex); - - acomp_ctx->buffer = kmalloc_node(PAGE_SIZE * 2, GFP_KERNEL, cpu_to_node(cpu)); - if (!acomp_ctx->buffer) - return -ENOMEM; + buffer = kmalloc_node(PAGE_SIZE * 2, GFP_KERNEL, cpu_to_node(cpu)); + if (!buffer) { + ret = -ENOMEM; + goto fail; + } acomp = crypto_alloc_acomp_node(pool->tfm_name, 0, 0, cpu_to_node(cpu)); if (IS_ERR(acomp)) { pr_err("could not alloc crypto acomp %s : %ld\n", pool->tfm_name, PTR_ERR(acomp)); ret = PTR_ERR(acomp); - goto acomp_fail; + goto fail; } - acomp_ctx->acomp = acomp; - acomp_ctx->is_sleepable = acomp_is_async(acomp); - req = acomp_request_alloc(acomp_ctx->acomp); + req = acomp_request_alloc(acomp); if (!req) { pr_err("could not alloc crypto acomp_request %s\n", pool->tfm_name); ret = -ENOMEM; - goto req_fail; + goto fail; } - acomp_ctx->req = req; + /* + * Only hold the mutex after completing allocations, otherwise we may + * recurse into zswap through reclaim and attempt to hold the mutex + * again resulting in a deadlock. + */ + mutex_lock(&acomp_ctx->mutex); crypto_init_wait(&acomp_ctx->wait); + /* * if the backend of acomp is async zip, crypto_req_done() will wakeup * crypto_wait_req(); if the backend of acomp is scomp, the callback @@ -855,12 +863,17 @@ static int zswap_cpu_comp_prepare(unsigned int cpu, struct hlist_node *node) acomp_request_set_callback(req, CRYPTO_TFM_REQ_MAY_BACKLOG, crypto_req_done, &acomp_ctx->wait); + acomp_ctx->buffer = buffer; + acomp_ctx->acomp = acomp; + acomp_ctx->is_sleepable = acomp_is_async(acomp); + acomp_ctx->req = req; + mutex_unlock(&acomp_ctx->mutex); return 0; -req_fail: - crypto_free_acomp(acomp_ctx->acomp); -acomp_fail: - kfree(acomp_ctx->buffer); +fail: + if (acomp) + crypto_free_acomp(acomp); + kfree(buffer); return ret; } @@ -869,17 +882,45 @@ static int zswap_cpu_comp_dead(unsigned int cpu, struct hlist_node *node) struct zswap_pool *pool = hlist_entry(node, struct zswap_pool, node); struct crypto_acomp_ctx *acomp_ctx = per_cpu_ptr(pool->acomp_ctx, cpu); + mutex_lock(&acomp_ctx->mutex); if (!IS_ERR_OR_NULL(acomp_ctx)) { if (!IS_ERR_OR_NULL(acomp_ctx->req)) acomp_request_free(acomp_ctx->req); + acomp_ctx->req = NULL; if (!IS_ERR_OR_NULL(acomp_ctx->acomp)) crypto_free_acomp(acomp_ctx->acomp); kfree(acomp_ctx->buffer); } + mutex_unlock(&acomp_ctx->mutex); return 0; } +static struct crypto_acomp_ctx *acomp_ctx_get_cpu_lock(struct zswap_pool *pool) +{ + struct crypto_acomp_ctx *acomp_ctx; + + for (;;) { + acomp_ctx = raw_cpu_ptr(pool->acomp_ctx); + mutex_lock(&acomp_ctx->mutex); + if (likely(acomp_ctx->req)) + return acomp_ctx; + /* + * It is possible that we were migrated to a different CPU after + * getting the per-CPU ctx but before the mutex was acquired. If + * the old CPU got offlined, zswap_cpu_comp_dead() could have + * already freed ctx->req (among other things) and set it to + * NULL. Just try again on the new CPU that we ended up on. + */ + mutex_unlock(&acomp_ctx->mutex); + } +} + +static void acomp_ctx_put_unlock(struct crypto_acomp_ctx *acomp_ctx) +{ + mutex_unlock(&acomp_ctx->mutex); +} + static bool zswap_compress(struct page *page, struct zswap_entry *entry, struct zswap_pool *pool) { @@ -893,10 +934,7 @@ static bool zswap_compress(struct page *page, struct zswap_entry *entry, gfp_t gfp; u8 *dst; - acomp_ctx = raw_cpu_ptr(pool->acomp_ctx); - - mutex_lock(&acomp_ctx->mutex); - + acomp_ctx = acomp_ctx_get_cpu_lock(pool); dst = acomp_ctx->buffer; sg_init_table(&input, 1); sg_set_page(&input, page, PAGE_SIZE, 0); @@ -949,7 +987,7 @@ unlock: else if (alloc_ret) zswap_reject_alloc_fail++; - mutex_unlock(&acomp_ctx->mutex); + acomp_ctx_put_unlock(acomp_ctx); return comp_ret == 0 && alloc_ret == 0; } @@ -960,9 +998,7 @@ static void zswap_decompress(struct zswap_entry *entry, struct folio *folio) struct crypto_acomp_ctx *acomp_ctx; u8 *src; - acomp_ctx = raw_cpu_ptr(entry->pool->acomp_ctx); - mutex_lock(&acomp_ctx->mutex); - + acomp_ctx = acomp_ctx_get_cpu_lock(entry->pool); src = zpool_map_handle(zpool, entry->handle, ZPOOL_MM_RO); /* * If zpool_map_handle is atomic, we cannot reliably utilize its mapped buffer @@ -986,10 +1022,10 @@ static void zswap_decompress(struct zswap_entry *entry, struct folio *folio) acomp_request_set_params(acomp_ctx->req, &input, &output, entry->length, PAGE_SIZE); BUG_ON(crypto_wait_req(crypto_acomp_decompress(acomp_ctx->req), &acomp_ctx->wait)); BUG_ON(acomp_ctx->req->dlen != PAGE_SIZE); - mutex_unlock(&acomp_ctx->mutex); if (src != acomp_ctx->buffer) zpool_unmap_handle(zpool, entry->handle); + acomp_ctx_put_unlock(acomp_ctx); } /********************************* diff --git a/net/802/psnap.c b/net/802/psnap.c index fca9d454905f..389df460c8c4 100644 --- a/net/802/psnap.c +++ b/net/802/psnap.c @@ -55,11 +55,11 @@ static int snap_rcv(struct sk_buff *skb, struct net_device *dev, goto drop; rcu_read_lock(); - proto = find_snap_client(skb_transport_header(skb)); + proto = find_snap_client(skb->data); if (proto) { /* Pass the frame on. */ - skb->transport_header += 5; skb_pull_rcsum(skb, 5); + skb_reset_transport_header(skb); rc = proto->rcvfunc(skb, dev, &snap_packet_type, orig_dev); } rcu_read_unlock(); diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c index c86f4e42e69c..7b2b04d6b856 100644 --- a/net/bluetooth/hci_sync.c +++ b/net/bluetooth/hci_sync.c @@ -1031,9 +1031,9 @@ static bool adv_use_rpa(struct hci_dev *hdev, uint32_t flags) static int hci_set_random_addr_sync(struct hci_dev *hdev, bdaddr_t *rpa) { - /* If we're advertising or initiating an LE connection we can't - * go ahead and change the random address at this time. This is - * because the eventual initiator address used for the + /* If a random_addr has been set we're advertising or initiating an LE + * connection we can't go ahead and change the random address at this + * time. This is because the eventual initiator address used for the * subsequently created connection will be undefined (some * controllers use the new address and others the one we had * when the operation started). @@ -1041,8 +1041,9 @@ static int hci_set_random_addr_sync(struct hci_dev *hdev, bdaddr_t *rpa) * In this kind of scenario skip the update and let the random * address be updated at the next cycle. */ - if (hci_dev_test_flag(hdev, HCI_LE_ADV) || - hci_lookup_le_connect(hdev)) { + if (bacmp(&hdev->random_addr, BDADDR_ANY) && + (hci_dev_test_flag(hdev, HCI_LE_ADV) || + hci_lookup_le_connect(hdev))) { bt_dev_dbg(hdev, "Deferring random address update"); hci_dev_set_flag(hdev, HCI_RPA_EXPIRED); return 0; diff --git a/net/bluetooth/mgmt.c b/net/bluetooth/mgmt.c index b31192d473d0..de47ad999d7b 100644 --- a/net/bluetooth/mgmt.c +++ b/net/bluetooth/mgmt.c @@ -7655,6 +7655,24 @@ static void device_added(struct sock *sk, struct hci_dev *hdev, mgmt_event(MGMT_EV_DEVICE_ADDED, hdev, &ev, sizeof(ev), sk); } +static void add_device_complete(struct hci_dev *hdev, void *data, int err) +{ + struct mgmt_pending_cmd *cmd = data; + struct mgmt_cp_add_device *cp = cmd->param; + + if (!err) { + device_added(cmd->sk, hdev, &cp->addr.bdaddr, cp->addr.type, + cp->action); + device_flags_changed(NULL, hdev, &cp->addr.bdaddr, + cp->addr.type, hdev->conn_flags, + PTR_UINT(cmd->user_data)); + } + + mgmt_cmd_complete(cmd->sk, hdev->id, MGMT_OP_ADD_DEVICE, + mgmt_status(err), &cp->addr, sizeof(cp->addr)); + mgmt_pending_free(cmd); +} + static int add_device_sync(struct hci_dev *hdev, void *data) { return hci_update_passive_scan_sync(hdev); @@ -7663,6 +7681,7 @@ static int add_device_sync(struct hci_dev *hdev, void *data) static int add_device(struct sock *sk, struct hci_dev *hdev, void *data, u16 len) { + struct mgmt_pending_cmd *cmd; struct mgmt_cp_add_device *cp = data; u8 auto_conn, addr_type; struct hci_conn_params *params; @@ -7743,9 +7762,24 @@ static int add_device(struct sock *sk, struct hci_dev *hdev, current_flags = params->flags; } - err = hci_cmd_sync_queue(hdev, add_device_sync, NULL, NULL); - if (err < 0) + cmd = mgmt_pending_new(sk, MGMT_OP_ADD_DEVICE, hdev, data, len); + if (!cmd) { + err = -ENOMEM; goto unlock; + } + + cmd->user_data = UINT_PTR(current_flags); + + err = hci_cmd_sync_queue(hdev, add_device_sync, cmd, + add_device_complete); + if (err < 0) { + err = mgmt_cmd_complete(sk, hdev->id, MGMT_OP_ADD_DEVICE, + MGMT_STATUS_FAILED, &cp->addr, + sizeof(cp->addr)); + mgmt_pending_free(cmd); + } + + goto unlock; added: device_added(sk, hdev, &cp->addr.bdaddr, cp->addr.type, cp->action); diff --git a/net/bluetooth/rfcomm/tty.c b/net/bluetooth/rfcomm/tty.c index af80d599c337..21a5b5535ebc 100644 --- a/net/bluetooth/rfcomm/tty.c +++ b/net/bluetooth/rfcomm/tty.c @@ -201,14 +201,14 @@ static ssize_t address_show(struct device *tty_dev, struct device_attribute *attr, char *buf) { struct rfcomm_dev *dev = dev_get_drvdata(tty_dev); - return sprintf(buf, "%pMR\n", &dev->dst); + return sysfs_emit(buf, "%pMR\n", &dev->dst); } static ssize_t channel_show(struct device *tty_dev, struct device_attribute *attr, char *buf) { struct rfcomm_dev *dev = dev_get_drvdata(tty_dev); - return sprintf(buf, "%d\n", dev->channel); + return sysfs_emit(buf, "%d\n", dev->channel); } static DEVICE_ATTR_RO(address); diff --git a/net/core/dev.c b/net/core/dev.c index 45a8c3dd4a64..a9f62f5aeb84 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -753,6 +753,36 @@ int dev_fill_forward_path(const struct net_device *dev, const u8 *daddr, } EXPORT_SYMBOL_GPL(dev_fill_forward_path); +/* must be called under rcu_read_lock(), as we dont take a reference */ +static struct napi_struct *napi_by_id(unsigned int napi_id) +{ + unsigned int hash = napi_id % HASH_SIZE(napi_hash); + struct napi_struct *napi; + + hlist_for_each_entry_rcu(napi, &napi_hash[hash], napi_hash_node) + if (napi->napi_id == napi_id) + return napi; + + return NULL; +} + +/* must be called under rcu_read_lock(), as we dont take a reference */ +struct napi_struct *netdev_napi_by_id(struct net *net, unsigned int napi_id) +{ + struct napi_struct *napi; + + napi = napi_by_id(napi_id); + if (!napi) + return NULL; + + if (WARN_ON_ONCE(!napi->dev)) + return NULL; + if (!net_eq(net, dev_net(napi->dev))) + return NULL; + + return napi; +} + /** * __dev_get_by_name - find a device by its name * @net: the applicable net namespace @@ -3642,8 +3672,10 @@ int skb_csum_hwoffload_help(struct sk_buff *skb, if (features & (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM)) { if (vlan_get_protocol(skb) == htons(ETH_P_IPV6) && - skb_network_header_len(skb) != sizeof(struct ipv6hdr)) + skb_network_header_len(skb) != sizeof(struct ipv6hdr) && + !ipv6_has_hopopt_jumbo(skb)) goto sw_checksum; + switch (skb->csum_offset) { case offsetof(struct tcphdr, check): case offsetof(struct udphdr, check): @@ -6291,19 +6323,6 @@ bool napi_complete_done(struct napi_struct *n, int work_done) } EXPORT_SYMBOL(napi_complete_done); -/* must be called under rcu_read_lock(), as we dont take a reference */ -struct napi_struct *napi_by_id(unsigned int napi_id) -{ - unsigned int hash = napi_id % HASH_SIZE(napi_hash); - struct napi_struct *napi; - - hlist_for_each_entry_rcu(napi, &napi_hash[hash], napi_hash_node) - if (napi->napi_id == napi_id) - return napi; - - return NULL; -} - static void skb_defer_free_flush(struct softnet_data *sd) { struct sk_buff *skb, *next; diff --git a/net/core/dev.h b/net/core/dev.h index d043dee25a68..deb5eae5749f 100644 --- a/net/core/dev.h +++ b/net/core/dev.h @@ -22,6 +22,8 @@ struct sd_flow_limit { extern int netdev_flow_limit_table_len; +struct napi_struct *netdev_napi_by_id(struct net *net, unsigned int napi_id); + #ifdef CONFIG_PROC_FS int __init dev_proc_init(void); #else @@ -269,7 +271,6 @@ void xdp_do_check_flushed(struct napi_struct *napi); static inline void xdp_do_check_flushed(struct napi_struct *napi) { } #endif -struct napi_struct *napi_by_id(unsigned int napi_id); void kick_defer_list_purge(struct softnet_data *sd, unsigned int cpu); #define XMIT_RECURSION_LIMIT 8 diff --git a/net/core/filter.c b/net/core/filter.c index 834614071727..2fb45a86f3dd 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -11251,6 +11251,7 @@ BPF_CALL_4(sk_select_reuseport, struct sk_reuseport_kern *, reuse_kern, bool is_sockarray = map->map_type == BPF_MAP_TYPE_REUSEPORT_SOCKARRAY; struct sock_reuseport *reuse; struct sock *selected_sk; + int err; selected_sk = map->ops->map_lookup_elem(map, key); if (!selected_sk) @@ -11258,10 +11259,6 @@ BPF_CALL_4(sk_select_reuseport, struct sk_reuseport_kern *, reuse_kern, reuse = rcu_dereference(selected_sk->sk_reuseport_cb); if (!reuse) { - /* Lookup in sock_map can return TCP ESTABLISHED sockets. */ - if (sk_is_refcounted(selected_sk)) - sock_put(selected_sk); - /* reuseport_array has only sk with non NULL sk_reuseport_cb. * The only (!reuse) case here is - the sk has already been * unhashed (e.g. by close()), so treat it as -ENOENT. @@ -11269,24 +11266,33 @@ BPF_CALL_4(sk_select_reuseport, struct sk_reuseport_kern *, reuse_kern, * Other maps (e.g. sock_map) do not provide this guarantee and * the sk may never be in the reuseport group to begin with. */ - return is_sockarray ? -ENOENT : -EINVAL; + err = is_sockarray ? -ENOENT : -EINVAL; + goto error; } if (unlikely(reuse->reuseport_id != reuse_kern->reuseport_id)) { struct sock *sk = reuse_kern->sk; - if (sk->sk_protocol != selected_sk->sk_protocol) - return -EPROTOTYPE; - else if (sk->sk_family != selected_sk->sk_family) - return -EAFNOSUPPORT; - - /* Catch all. Likely bound to a different sockaddr. */ - return -EBADFD; + if (sk->sk_protocol != selected_sk->sk_protocol) { + err = -EPROTOTYPE; + } else if (sk->sk_family != selected_sk->sk_family) { + err = -EAFNOSUPPORT; + } else { + /* Catch all. Likely bound to a different sockaddr. */ + err = -EBADFD; + } + goto error; } reuse_kern->selected_sk = selected_sk; return 0; +error: + /* Lookup in sock_map can return TCP ESTABLISHED sockets. */ + if (sk_is_refcounted(selected_sk)) + sock_put(selected_sk); + + return err; } static const struct bpf_func_proto sk_select_reuseport_proto = { diff --git a/net/core/link_watch.c b/net/core/link_watch.c index 1b4d39e38084..cb04ef2b9807 100644 --- a/net/core/link_watch.c +++ b/net/core/link_watch.c @@ -42,14 +42,18 @@ static unsigned int default_operstate(const struct net_device *dev) * first check whether lower is indeed the source of its down state. */ if (!netif_carrier_ok(dev)) { - int iflink = dev_get_iflink(dev); struct net_device *peer; + int iflink; /* If called from netdev_run_todo()/linkwatch_sync_dev(), * dev_net(dev) can be already freed, and RTNL is not held. */ - if (dev->reg_state == NETREG_UNREGISTERED || - iflink == dev->ifindex) + if (dev->reg_state <= NETREG_REGISTERED) + iflink = dev_get_iflink(dev); + else + iflink = dev->ifindex; + + if (iflink == dev->ifindex) return IF_OPER_DOWN; ASSERT_RTNL(); diff --git a/net/core/netdev-genl-gen.c b/net/core/netdev-genl-gen.c index a89cbd8d87c3..996ac6a449eb 100644 --- a/net/core/netdev-genl-gen.c +++ b/net/core/netdev-genl-gen.c @@ -197,6 +197,16 @@ static const struct genl_multicast_group netdev_nl_mcgrps[] = { [NETDEV_NLGRP_PAGE_POOL] = { "page-pool", }, }; +static void __netdev_nl_sock_priv_init(void *priv) +{ + netdev_nl_sock_priv_init(priv); +} + +static void __netdev_nl_sock_priv_destroy(void *priv) +{ + netdev_nl_sock_priv_destroy(priv); +} + struct genl_family netdev_nl_family __ro_after_init = { .name = NETDEV_FAMILY_NAME, .version = NETDEV_FAMILY_VERSION, @@ -208,6 +218,6 @@ struct genl_family netdev_nl_family __ro_after_init = { .mcgrps = netdev_nl_mcgrps, .n_mcgrps = ARRAY_SIZE(netdev_nl_mcgrps), .sock_priv_size = sizeof(struct list_head), - .sock_priv_init = (void *)netdev_nl_sock_priv_init, - .sock_priv_destroy = (void *)netdev_nl_sock_priv_destroy, + .sock_priv_init = __netdev_nl_sock_priv_init, + .sock_priv_destroy = __netdev_nl_sock_priv_destroy, }; diff --git a/net/core/netdev-genl.c b/net/core/netdev-genl.c index 2d3ae0cd3ad2..a3bdaf075b6b 100644 --- a/net/core/netdev-genl.c +++ b/net/core/netdev-genl.c @@ -167,8 +167,6 @@ netdev_nl_napi_fill_one(struct sk_buff *rsp, struct napi_struct *napi, void *hdr; pid_t pid; - if (WARN_ON_ONCE(!napi->dev)) - return -EINVAL; if (!(napi->dev->flags & IFF_UP)) return 0; @@ -176,8 +174,7 @@ netdev_nl_napi_fill_one(struct sk_buff *rsp, struct napi_struct *napi, if (!hdr) return -EMSGSIZE; - if (napi->napi_id >= MIN_NAPI_ID && - nla_put_u32(rsp, NETDEV_A_NAPI_ID, napi->napi_id)) + if (nla_put_u32(rsp, NETDEV_A_NAPI_ID, napi->napi_id)) goto nla_put_failure; if (nla_put_u32(rsp, NETDEV_A_NAPI_IFINDEX, napi->dev->ifindex)) @@ -235,7 +232,7 @@ int netdev_nl_napi_get_doit(struct sk_buff *skb, struct genl_info *info) rtnl_lock(); rcu_read_lock(); - napi = napi_by_id(napi_id); + napi = netdev_napi_by_id(genl_info_net(info), napi_id); if (napi) { err = netdev_nl_napi_fill_one(rsp, napi, info); } else { @@ -246,8 +243,12 @@ int netdev_nl_napi_get_doit(struct sk_buff *skb, struct genl_info *info) rcu_read_unlock(); rtnl_unlock(); - if (err) + if (err) { + goto err_free_msg; + } else if (!rsp->len) { + err = -ENOENT; goto err_free_msg; + } return genlmsg_reply(rsp, info); @@ -268,6 +269,8 @@ netdev_nl_napi_dump_one(struct net_device *netdev, struct sk_buff *rsp, return err; list_for_each_entry(napi, &netdev->napi_list, dev_list) { + if (napi->napi_id < MIN_NAPI_ID) + continue; if (ctx->napi_id && napi->napi_id >= ctx->napi_id) continue; @@ -350,7 +353,7 @@ int netdev_nl_napi_set_doit(struct sk_buff *skb, struct genl_info *info) rtnl_lock(); rcu_read_lock(); - napi = napi_by_id(napi_id); + napi = netdev_napi_by_id(genl_info_net(info), napi_id); if (napi) { err = netdev_nl_napi_set_config(napi, info); } else { diff --git a/net/core/netpoll.c b/net/core/netpoll.c index 2e459b9d88eb..96a6ed37d4cc 100644 --- a/net/core/netpoll.c +++ b/net/core/netpoll.c @@ -627,6 +627,8 @@ int __netpoll_setup(struct netpoll *np, struct net_device *ndev) const struct net_device_ops *ops; int err; + skb_queue_head_init(&np->skb_pool); + if (ndev->priv_flags & IFF_DISABLE_NETPOLL) { np_err(np, "%s doesn't support polling, aborting\n", ndev->name); @@ -662,6 +664,9 @@ int __netpoll_setup(struct netpoll *np, struct net_device *ndev) strscpy(np->dev_name, ndev->name, IFNAMSIZ); npinfo->netpoll = np; + /* fill up the skb queue */ + refill_skbs(np); + /* last thing to do is link it to the net device structure */ rcu_assign_pointer(ndev->npinfo, npinfo); @@ -681,8 +686,6 @@ int netpoll_setup(struct netpoll *np) struct in_device *in_dev; int err; - skb_queue_head_init(&np->skb_pool); - rtnl_lock(); if (np->dev_name[0]) { struct net *net = current->nsproxy->net_ns; @@ -782,9 +785,6 @@ put_noaddr: } } - /* fill up the skb queue */ - refill_skbs(np); - err = __netpoll_setup(np, ndev); if (err) goto flush; diff --git a/net/core/pktgen.c b/net/core/pktgen.c index 7e23cacbe66e..4cb547fae91f 100644 --- a/net/core/pktgen.c +++ b/net/core/pktgen.c @@ -851,6 +851,9 @@ static ssize_t get_imix_entries(const char __user *buffer, unsigned long weight; unsigned long size; + if (pkt_dev->n_imix_entries >= MAX_IMIX_ENTRIES) + return -E2BIG; + len = num_arg(&buffer[i], max_digits, &size); if (len < 0) return len; @@ -880,9 +883,6 @@ static ssize_t get_imix_entries(const char __user *buffer, i++; pkt_dev->n_imix_entries++; - - if (pkt_dev->n_imix_entries > MAX_IMIX_ENTRIES) - return -E2BIG; } while (c == ' '); return i; diff --git a/net/core/sock.c b/net/core/sock.c index 74729d20cd00..be84885f9290 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -1295,7 +1295,10 @@ int sk_setsockopt(struct sock *sk, int level, int optname, sk->sk_reuse = (valbool ? SK_CAN_REUSE : SK_NO_REUSE); break; case SO_REUSEPORT: - sk->sk_reuseport = valbool; + if (valbool && !sk_is_inet(sk)) + ret = -EOPNOTSUPP; + else + sk->sk_reuseport = valbool; break; case SO_DONTROUTE: sock_valbool_flag(sk, SOCK_LOCALROUTE, valbool); diff --git a/net/core/xdp.c b/net/core/xdp.c index bcc5551c6424..2315feed94ef 100644 --- a/net/core/xdp.c +++ b/net/core/xdp.c @@ -186,7 +186,6 @@ int __xdp_rxq_info_reg(struct xdp_rxq_info *xdp_rxq, xdp_rxq_info_init(xdp_rxq); xdp_rxq->dev = dev; xdp_rxq->queue_index = queue_index; - xdp_rxq->napi_id = napi_id; xdp_rxq->frag_size = frag_size; xdp_rxq->reg_state = REG_STATE_REGISTERED; diff --git a/net/ipv4/ip_tunnel.c b/net/ipv4/ip_tunnel.c index 25505f9b724c..09b73acf037a 100644 --- a/net/ipv4/ip_tunnel.c +++ b/net/ipv4/ip_tunnel.c @@ -294,7 +294,7 @@ static int ip_tunnel_bind_dev(struct net_device *dev) ip_tunnel_init_flow(&fl4, iph->protocol, iph->daddr, iph->saddr, tunnel->parms.o_key, - iph->tos & INET_DSCP_MASK, dev_net(dev), + iph->tos & INET_DSCP_MASK, tunnel->net, tunnel->parms.link, tunnel->fwmark, 0, 0); rt = ip_route_output_key(tunnel->net, &fl4); @@ -611,7 +611,7 @@ void ip_md_tunnel_xmit(struct sk_buff *skb, struct net_device *dev, } ip_tunnel_init_flow(&fl4, proto, key->u.ipv4.dst, key->u.ipv4.src, tunnel_id_to_key32(key->tun_id), - tos & INET_DSCP_MASK, dev_net(dev), 0, skb->mark, + tos & INET_DSCP_MASK, tunnel->net, 0, skb->mark, skb_get_hash(skb), key->flow_flags); if (!tunnel_hlen) @@ -774,7 +774,7 @@ void ip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev, ip_tunnel_init_flow(&fl4, protocol, dst, tnl_params->saddr, tunnel->parms.o_key, tos & INET_DSCP_MASK, - dev_net(dev), READ_ONCE(tunnel->parms.link), + tunnel->net, READ_ONCE(tunnel->parms.link), tunnel->fwmark, skb_get_hash(skb), 0); if (ip_tunnel_encap(skb, &tunnel->encap, &protocol, &fl4) < 0) diff --git a/net/ipv4/route.c b/net/ipv4/route.c index 0fbec3509618..e1564b95fab0 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -2445,6 +2445,7 @@ martian_destination: net_warn_ratelimited("martian destination %pI4 from %pI4, dev %s\n", &daddr, &saddr, dev->name); #endif + goto out; e_nobufs: reason = SKB_DROP_REASON_NOMEM; diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 5bdf13ac26ef..4811727b8a02 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -7328,6 +7328,7 @@ int tcp_conn_request(struct request_sock_ops *rsk_ops, if (unlikely(!inet_csk_reqsk_queue_hash_add(sk, req, req->timeout))) { reqsk_free(req); + dst_release(dst); return 0; } diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index a38c8b1f44db..c26f6c4b7bb4 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -896,7 +896,7 @@ static void tcp_v4_send_reset(const struct sock *sk, struct sk_buff *skb, sock_net_set(ctl_sk, net); if (sk) { ctl_sk->sk_mark = (sk->sk_state == TCP_TIME_WAIT) ? - inet_twsk(sk)->tw_mark : sk->sk_mark; + inet_twsk(sk)->tw_mark : READ_ONCE(sk->sk_mark); ctl_sk->sk_priority = (sk->sk_state == TCP_TIME_WAIT) ? inet_twsk(sk)->tw_priority : READ_ONCE(sk->sk_priority); transmit_time = tcp_transmit_time(sk); diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index e8953e88efef..86d282618515 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -533,7 +533,7 @@ begin: return NULL; } -/* In hash4, rehash can happen in connect(), where hash4_cnt keeps unchanged. */ +/* udp_rehash4() only checks hslot4, and hash4_cnt is not processed. */ static void udp_rehash4(struct udp_table *udptable, struct sock *sk, u16 newhash4) { @@ -582,15 +582,13 @@ void udp_lib_hash4(struct sock *sk, u16 hash) struct net *net = sock_net(sk); struct udp_table *udptable; - /* Connected udp socket can re-connect to another remote address, - * so rehash4 is needed. + /* Connected udp socket can re-connect to another remote address, which + * will be handled by rehash. Thus no need to redo hash4 here. */ - udptable = net->ipv4.udp_table; - if (udp_hashed4(sk)) { - udp_rehash4(udptable, sk, hash); + if (udp_hashed4(sk)) return; - } + udptable = net->ipv4.udp_table; hslot = udp_hashslot(udptable, net, udp_sk(sk)->udp_port_hash); hslot2 = udp_hashslot2(udptable, udp_sk(sk)->udp_portaddr_hash); hslot4 = udp_hashslot4(udptable, hash); @@ -2173,14 +2171,14 @@ void udp_lib_rehash(struct sock *sk, u16 newhash, u16 newhash4) struct udp_table *udptable = udp_get_table_prot(sk); struct udp_hslot *hslot, *hslot2, *nhslot2; + hslot = udp_hashslot(udptable, sock_net(sk), + udp_sk(sk)->udp_port_hash); hslot2 = udp_hashslot2(udptable, udp_sk(sk)->udp_portaddr_hash); nhslot2 = udp_hashslot2(udptable, newhash); udp_sk(sk)->udp_portaddr_hash = newhash; if (hslot2 != nhslot2 || rcu_access_pointer(sk->sk_reuseport_cb)) { - hslot = udp_hashslot(udptable, sock_net(sk), - udp_sk(sk)->udp_port_hash); /* we must lock primary chain too */ spin_lock_bh(&hslot->lock); if (rcu_access_pointer(sk->sk_reuseport_cb)) @@ -2199,19 +2197,29 @@ void udp_lib_rehash(struct sock *sk, u16 newhash, u16 newhash4) spin_unlock(&nhslot2->lock); } - if (udp_hashed4(sk)) { - udp_rehash4(udptable, sk, newhash4); + spin_unlock_bh(&hslot->lock); + } + + /* Now process hash4 if necessary: + * (1) update hslot4; + * (2) update hslot2->hash4_cnt. + * Note that hslot2/hslot4 should be checked separately, as + * either of them may change with the other unchanged. + */ + if (udp_hashed4(sk)) { + spin_lock_bh(&hslot->lock); - if (hslot2 != nhslot2) { - spin_lock(&hslot2->lock); - udp_hash4_dec(hslot2); - spin_unlock(&hslot2->lock); + udp_rehash4(udptable, sk, newhash4); + if (hslot2 != nhslot2) { + spin_lock(&hslot2->lock); + udp_hash4_dec(hslot2); + spin_unlock(&hslot2->lock); - spin_lock(&nhslot2->lock); - udp_hash4_inc(nhslot2); - spin_unlock(&nhslot2->lock); - } + spin_lock(&nhslot2->lock); + udp_hash4_inc(nhslot2); + spin_unlock(&nhslot2->lock); } + spin_unlock_bh(&hslot->lock); } } diff --git a/net/ipv6/ila/ila_xlat.c b/net/ipv6/ila/ila_xlat.c index 7646e401c630..1d41b2ab4884 100644 --- a/net/ipv6/ila/ila_xlat.c +++ b/net/ipv6/ila/ila_xlat.c @@ -195,6 +195,8 @@ static const struct nf_hook_ops ila_nf_hook_ops[] = { }, }; +static DEFINE_MUTEX(ila_mutex); + static int ila_add_mapping(struct net *net, struct ila_xlat_params *xp) { struct ila_net *ilan = net_generic(net, ila_net_id); @@ -202,16 +204,20 @@ static int ila_add_mapping(struct net *net, struct ila_xlat_params *xp) spinlock_t *lock = ila_get_lock(ilan, xp->ip.locator_match); int err = 0, order; - if (!ilan->xlat.hooks_registered) { + if (!READ_ONCE(ilan->xlat.hooks_registered)) { /* We defer registering net hooks in the namespace until the * first mapping is added. */ - err = nf_register_net_hooks(net, ila_nf_hook_ops, - ARRAY_SIZE(ila_nf_hook_ops)); + mutex_lock(&ila_mutex); + if (!ilan->xlat.hooks_registered) { + err = nf_register_net_hooks(net, ila_nf_hook_ops, + ARRAY_SIZE(ila_nf_hook_ops)); + if (!err) + WRITE_ONCE(ilan->xlat.hooks_registered, true); + } + mutex_unlock(&ila_mutex); if (err) return err; - - ilan->xlat.hooks_registered = true; } ila = kzalloc(sizeof(*ila), GFP_KERNEL); diff --git a/net/llc/llc_input.c b/net/llc/llc_input.c index 51bccfb00a9c..61b0159b2fbe 100644 --- a/net/llc/llc_input.c +++ b/net/llc/llc_input.c @@ -124,8 +124,8 @@ static inline int llc_fixup_skb(struct sk_buff *skb) if (unlikely(!pskb_may_pull(skb, llc_len))) return 0; - skb->transport_header += llc_len; skb_pull(skb, llc_len); + skb_reset_transport_header(skb); if (skb->protocol == htons(ETH_P_802_2)) { __be16 pdulen; s32 data_size; diff --git a/net/mac802154/iface.c b/net/mac802154/iface.c index c0e2da5072be..9e4631fade90 100644 --- a/net/mac802154/iface.c +++ b/net/mac802154/iface.c @@ -684,6 +684,10 @@ void ieee802154_if_remove(struct ieee802154_sub_if_data *sdata) ASSERT_RTNL(); mutex_lock(&sdata->local->iflist_mtx); + if (list_empty(&sdata->local->interfaces)) { + mutex_unlock(&sdata->local->iflist_mtx); + return; + } list_del_rcu(&sdata->list); mutex_unlock(&sdata->local->iflist_mtx); diff --git a/net/mptcp/ctrl.c b/net/mptcp/ctrl.c index 38d8121331d4..b0dd008e2114 100644 --- a/net/mptcp/ctrl.c +++ b/net/mptcp/ctrl.c @@ -102,16 +102,15 @@ static void mptcp_pernet_set_defaults(struct mptcp_pernet *pernet) } #ifdef CONFIG_SYSCTL -static int mptcp_set_scheduler(const struct net *net, const char *name) +static int mptcp_set_scheduler(char *scheduler, const char *name) { - struct mptcp_pernet *pernet = mptcp_get_pernet(net); struct mptcp_sched_ops *sched; int ret = 0; rcu_read_lock(); sched = mptcp_sched_find(name); if (sched) - strscpy(pernet->scheduler, name, MPTCP_SCHED_NAME_MAX); + strscpy(scheduler, name, MPTCP_SCHED_NAME_MAX); else ret = -ENOENT; rcu_read_unlock(); @@ -122,7 +121,7 @@ static int mptcp_set_scheduler(const struct net *net, const char *name) static int proc_scheduler(const struct ctl_table *ctl, int write, void *buffer, size_t *lenp, loff_t *ppos) { - const struct net *net = current->nsproxy->net_ns; + char (*scheduler)[MPTCP_SCHED_NAME_MAX] = ctl->data; char val[MPTCP_SCHED_NAME_MAX]; struct ctl_table tbl = { .data = val, @@ -130,11 +129,11 @@ static int proc_scheduler(const struct ctl_table *ctl, int write, }; int ret; - strscpy(val, mptcp_get_scheduler(net), MPTCP_SCHED_NAME_MAX); + strscpy(val, *scheduler, MPTCP_SCHED_NAME_MAX); ret = proc_dostring(&tbl, write, buffer, lenp, ppos); if (write && ret == 0) - ret = mptcp_set_scheduler(net, val); + ret = mptcp_set_scheduler(*scheduler, val); return ret; } @@ -161,7 +160,9 @@ static int proc_blackhole_detect_timeout(const struct ctl_table *table, int write, void *buffer, size_t *lenp, loff_t *ppos) { - struct mptcp_pernet *pernet = mptcp_get_pernet(current->nsproxy->net_ns); + struct mptcp_pernet *pernet = container_of(table->data, + struct mptcp_pernet, + blackhole_timeout); int ret; ret = proc_dointvec_minmax(table, write, buffer, lenp, ppos); @@ -228,7 +229,7 @@ static struct ctl_table mptcp_sysctl_table[] = { { .procname = "available_schedulers", .maxlen = MPTCP_SCHED_BUF_MAX, - .mode = 0644, + .mode = 0444, .proc_handler = proc_available_schedulers, }, { diff --git a/net/mptcp/options.c b/net/mptcp/options.c index 1603b3702e22..123f3f297284 100644 --- a/net/mptcp/options.c +++ b/net/mptcp/options.c @@ -607,7 +607,6 @@ static bool mptcp_established_options_dss(struct sock *sk, struct sk_buff *skb, } opts->ext_copy.use_ack = 1; opts->suboptions = OPTION_MPTCP_DSS; - WRITE_ONCE(msk->old_wspace, __mptcp_space((struct sock *)msk)); /* Add kind/length/subtype/flag overhead if mapping is not populated */ if (dss_size == 0) @@ -667,8 +666,15 @@ static bool mptcp_established_options_add_addr(struct sock *sk, struct sk_buff * &echo, &drop_other_suboptions)) return false; + /* + * Later on, mptcp_write_options() will enforce mutually exclusion with + * DSS, bail out if such option is set and we can't drop it. + */ if (drop_other_suboptions) remaining += opt_size; + else if (opts->suboptions & OPTION_MPTCP_DSS) + return false; + len = mptcp_add_addr_len(opts->addr.family, echo, !!opts->addr.port); if (remaining < len) return false; @@ -1281,7 +1287,7 @@ static void mptcp_set_rwin(struct tcp_sock *tp, struct tcphdr *th) } MPTCP_INC_STATS(sock_net(ssk), MPTCP_MIB_RCVWNDCONFLICT); } - return; + goto update_wspace; } if (rcv_wnd_new != rcv_wnd_old) { @@ -1306,6 +1312,9 @@ raise_win: th->window = htons(new_win); MPTCP_INC_STATS(sock_net(ssk), MPTCP_MIB_RCVWNDSHARED); } + +update_wspace: + WRITE_ONCE(msk->old_wspace, tp->rcv_wnd); } __sum16 __mptcp_make_csum(u64 data_seq, u32 subflow_seq, u16 data_len, __wsum sum) diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index 08a72242428c..1b2e7cbb577f 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -136,6 +136,7 @@ static bool mptcp_try_coalesce(struct sock *sk, struct sk_buff *to, int delta; if (MPTCP_SKB_CB(from)->offset || + ((to->len + from->len) > (sk->sk_rcvbuf >> 3)) || !skb_try_coalesce(to, from, &fragstolen, &delta)) return false; @@ -528,13 +529,13 @@ static void mptcp_send_ack(struct mptcp_sock *msk) mptcp_subflow_send_ack(mptcp_subflow_tcp_sock(subflow)); } -static void mptcp_subflow_cleanup_rbuf(struct sock *ssk) +static void mptcp_subflow_cleanup_rbuf(struct sock *ssk, int copied) { bool slow; slow = lock_sock_fast(ssk); if (tcp_can_send_ack(ssk)) - tcp_cleanup_rbuf(ssk, 1); + tcp_cleanup_rbuf(ssk, copied); unlock_sock_fast(ssk, slow); } @@ -551,7 +552,7 @@ static bool mptcp_subflow_could_cleanup(const struct sock *ssk, bool rx_empty) (ICSK_ACK_PUSHED2 | ICSK_ACK_PUSHED))); } -static void mptcp_cleanup_rbuf(struct mptcp_sock *msk) +static void mptcp_cleanup_rbuf(struct mptcp_sock *msk, int copied) { int old_space = READ_ONCE(msk->old_wspace); struct mptcp_subflow_context *subflow; @@ -559,14 +560,14 @@ static void mptcp_cleanup_rbuf(struct mptcp_sock *msk) int space = __mptcp_space(sk); bool cleanup, rx_empty; - cleanup = (space > 0) && (space >= (old_space << 1)); - rx_empty = !__mptcp_rmem(sk); + cleanup = (space > 0) && (space >= (old_space << 1)) && copied; + rx_empty = !__mptcp_rmem(sk) && copied; mptcp_for_each_subflow(msk, subflow) { struct sock *ssk = mptcp_subflow_tcp_sock(subflow); if (cleanup || mptcp_subflow_could_cleanup(ssk, rx_empty)) - mptcp_subflow_cleanup_rbuf(ssk); + mptcp_subflow_cleanup_rbuf(ssk, copied); } } @@ -1939,6 +1940,8 @@ do_error: goto out; } +static void mptcp_rcv_space_adjust(struct mptcp_sock *msk, int copied); + static int __mptcp_recvmsg_mskq(struct mptcp_sock *msk, struct msghdr *msg, size_t len, int flags, @@ -1992,6 +1995,7 @@ static int __mptcp_recvmsg_mskq(struct mptcp_sock *msk, break; } + mptcp_rcv_space_adjust(msk, copied); return copied; } @@ -2217,9 +2221,6 @@ static int mptcp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, copied += bytes_read; - /* be sure to advertise window change */ - mptcp_cleanup_rbuf(msk); - if (skb_queue_empty(&msk->receive_queue) && __mptcp_move_skbs(msk)) continue; @@ -2268,7 +2269,7 @@ static int mptcp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, } pr_debug("block timeout %ld\n", timeo); - mptcp_rcv_space_adjust(msk, copied); + mptcp_cleanup_rbuf(msk, copied); err = sk_wait_data(sk, &timeo, NULL); if (err < 0) { err = copied ? : err; @@ -2276,7 +2277,7 @@ static int mptcp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, } } - mptcp_rcv_space_adjust(msk, copied); + mptcp_cleanup_rbuf(msk, copied); out_err: if (cmsg_flags && copied >= 0) { diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h index a93e661ef5c4..73526f1d768f 100644 --- a/net/mptcp/protocol.h +++ b/net/mptcp/protocol.h @@ -760,10 +760,15 @@ static inline u64 mptcp_data_avail(const struct mptcp_sock *msk) static inline bool mptcp_epollin_ready(const struct sock *sk) { + u64 data_avail = mptcp_data_avail(mptcp_sk(sk)); + + if (!data_avail) + return false; + /* mptcp doesn't have to deal with small skbs in the receive queue, - * at it can always coalesce them + * as it can always coalesce them */ - return (mptcp_data_avail(mptcp_sk(sk)) >= sk->sk_rcvlowat) || + return (data_avail >= sk->sk_rcvlowat) || (mem_cgroup_sockets_enabled && sk->sk_memcg && mem_cgroup_under_socket_pressure(sk->sk_memcg)) || READ_ONCE(tcp_memory_pressure); diff --git a/net/ncsi/internal.h b/net/ncsi/internal.h index ef0f8f73826f..4e0842df5234 100644 --- a/net/ncsi/internal.h +++ b/net/ncsi/internal.h @@ -289,6 +289,7 @@ enum { ncsi_dev_state_config_sp = 0x0301, ncsi_dev_state_config_cis, ncsi_dev_state_config_oem_gma, + ncsi_dev_state_config_apply_mac, ncsi_dev_state_config_clear_vids, ncsi_dev_state_config_svf, ncsi_dev_state_config_ev, @@ -322,6 +323,7 @@ struct ncsi_dev_priv { #define NCSI_DEV_RESHUFFLE 4 #define NCSI_DEV_RESET 8 /* Reset state of NC */ unsigned int gma_flag; /* OEM GMA flag */ + struct sockaddr pending_mac; /* MAC address received from GMA */ spinlock_t lock; /* Protect the NCSI device */ unsigned int package_probe_id;/* Current ID during probe */ unsigned int package_num; /* Number of packages */ diff --git a/net/ncsi/ncsi-manage.c b/net/ncsi/ncsi-manage.c index 5cf55bde366d..bf276eaf9330 100644 --- a/net/ncsi/ncsi-manage.c +++ b/net/ncsi/ncsi-manage.c @@ -1038,7 +1038,7 @@ static void ncsi_configure_channel(struct ncsi_dev_priv *ndp) : ncsi_dev_state_config_clear_vids; break; case ncsi_dev_state_config_oem_gma: - nd->state = ncsi_dev_state_config_clear_vids; + nd->state = ncsi_dev_state_config_apply_mac; nca.package = np->id; nca.channel = nc->id; @@ -1050,10 +1050,22 @@ static void ncsi_configure_channel(struct ncsi_dev_priv *ndp) nca.type = NCSI_PKT_CMD_OEM; ret = ncsi_gma_handler(&nca, nc->version.mf_id); } - if (ret < 0) + if (ret < 0) { + nd->state = ncsi_dev_state_config_clear_vids; schedule_work(&ndp->work); + } break; + case ncsi_dev_state_config_apply_mac: + rtnl_lock(); + ret = dev_set_mac_address(dev, &ndp->pending_mac, NULL); + rtnl_unlock(); + if (ret < 0) + netdev_warn(dev, "NCSI: 'Writing MAC address to device failed\n"); + + nd->state = ncsi_dev_state_config_clear_vids; + + fallthrough; case ncsi_dev_state_config_clear_vids: case ncsi_dev_state_config_svf: case ncsi_dev_state_config_ev: diff --git a/net/ncsi/ncsi-rsp.c b/net/ncsi/ncsi-rsp.c index e28be33bdf2c..14bd66909ca4 100644 --- a/net/ncsi/ncsi-rsp.c +++ b/net/ncsi/ncsi-rsp.c @@ -628,16 +628,14 @@ static int ncsi_rsp_handler_snfc(struct ncsi_request *nr) static int ncsi_rsp_handler_oem_gma(struct ncsi_request *nr, int mfr_id) { struct ncsi_dev_priv *ndp = nr->ndp; + struct sockaddr *saddr = &ndp->pending_mac; struct net_device *ndev = ndp->ndev.dev; struct ncsi_rsp_oem_pkt *rsp; - struct sockaddr saddr; u32 mac_addr_off = 0; - int ret = 0; /* Get the response header */ rsp = (struct ncsi_rsp_oem_pkt *)skb_network_header(nr->rsp); - saddr.sa_family = ndev->type; ndev->priv_flags |= IFF_LIVE_ADDR_CHANGE; if (mfr_id == NCSI_OEM_MFR_BCM_ID) mac_addr_off = BCM_MAC_ADDR_OFFSET; @@ -646,22 +644,17 @@ static int ncsi_rsp_handler_oem_gma(struct ncsi_request *nr, int mfr_id) else if (mfr_id == NCSI_OEM_MFR_INTEL_ID) mac_addr_off = INTEL_MAC_ADDR_OFFSET; - memcpy(saddr.sa_data, &rsp->data[mac_addr_off], ETH_ALEN); + saddr->sa_family = ndev->type; + memcpy(saddr->sa_data, &rsp->data[mac_addr_off], ETH_ALEN); if (mfr_id == NCSI_OEM_MFR_BCM_ID || mfr_id == NCSI_OEM_MFR_INTEL_ID) - eth_addr_inc((u8 *)saddr.sa_data); - if (!is_valid_ether_addr((const u8 *)saddr.sa_data)) + eth_addr_inc((u8 *)saddr->sa_data); + if (!is_valid_ether_addr((const u8 *)saddr->sa_data)) return -ENXIO; /* Set the flag for GMA command which should only be called once */ ndp->gma_flag = 1; - rtnl_lock(); - ret = dev_set_mac_address(ndev, &saddr, NULL); - rtnl_unlock(); - if (ret < 0) - netdev_warn(ndev, "NCSI: 'Writing mac address to device failed\n"); - - return ret; + return 0; } /* Response handler for Mellanox card */ diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index 9db3e2b0b1c3..456446d7af20 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -2517,12 +2517,15 @@ void *nf_ct_alloc_hashtable(unsigned int *sizep, int nulls) struct hlist_nulls_head *hash; unsigned int nr_slots, i; - if (*sizep > (UINT_MAX / sizeof(struct hlist_nulls_head))) + if (*sizep > (INT_MAX / sizeof(struct hlist_nulls_head))) return NULL; BUILD_BUG_ON(sizeof(struct hlist_nulls_head) != sizeof(struct hlist_head)); nr_slots = *sizep = roundup(*sizep, PAGE_SIZE / sizeof(struct hlist_nulls_head)); + if (nr_slots > (INT_MAX / sizeof(struct hlist_nulls_head))) + return NULL; + hash = kvcalloc(nr_slots, sizeof(struct hlist_nulls_head), GFP_KERNEL); if (hash && nulls) diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index 0b9f1e8dfe49..c4af283356e7 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -8822,6 +8822,7 @@ static void nft_unregister_flowtable_hook(struct net *net, } static void __nft_unregister_flowtable_net_hooks(struct net *net, + struct nft_flowtable *flowtable, struct list_head *hook_list, bool release_netdev) { @@ -8829,6 +8830,8 @@ static void __nft_unregister_flowtable_net_hooks(struct net *net, list_for_each_entry_safe(hook, next, hook_list, list) { nf_unregister_net_hook(net, &hook->ops); + flowtable->data.type->setup(&flowtable->data, hook->ops.dev, + FLOW_BLOCK_UNBIND); if (release_netdev) { list_del(&hook->list); kfree_rcu(hook, rcu); @@ -8837,9 +8840,10 @@ static void __nft_unregister_flowtable_net_hooks(struct net *net, } static void nft_unregister_flowtable_net_hooks(struct net *net, + struct nft_flowtable *flowtable, struct list_head *hook_list) { - __nft_unregister_flowtable_net_hooks(net, hook_list, false); + __nft_unregister_flowtable_net_hooks(net, flowtable, hook_list, false); } static int nft_register_flowtable_net_hooks(struct net *net, @@ -9481,8 +9485,6 @@ static void nf_tables_flowtable_destroy(struct nft_flowtable *flowtable) flowtable->data.type->free(&flowtable->data); list_for_each_entry_safe(hook, next, &flowtable->hook_list, list) { - flowtable->data.type->setup(&flowtable->data, hook->ops.dev, - FLOW_BLOCK_UNBIND); list_del_rcu(&hook->list); kfree_rcu(hook, rcu); } @@ -10870,6 +10872,7 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb) &nft_trans_flowtable_hooks(trans), trans->msg_type); nft_unregister_flowtable_net_hooks(net, + nft_trans_flowtable(trans), &nft_trans_flowtable_hooks(trans)); } else { list_del_rcu(&nft_trans_flowtable(trans)->list); @@ -10878,6 +10881,7 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb) NULL, trans->msg_type); nft_unregister_flowtable_net_hooks(net, + nft_trans_flowtable(trans), &nft_trans_flowtable(trans)->hook_list); } break; @@ -11140,11 +11144,13 @@ static int __nf_tables_abort(struct net *net, enum nfnl_abort_action action) case NFT_MSG_NEWFLOWTABLE: if (nft_trans_flowtable_update(trans)) { nft_unregister_flowtable_net_hooks(net, + nft_trans_flowtable(trans), &nft_trans_flowtable_hooks(trans)); } else { nft_use_dec_restore(&table->use); list_del_rcu(&nft_trans_flowtable(trans)->list); nft_unregister_flowtable_net_hooks(net, + nft_trans_flowtable(trans), &nft_trans_flowtable(trans)->hook_list); } break; @@ -11737,7 +11743,8 @@ static void __nft_release_hook(struct net *net, struct nft_table *table) list_for_each_entry(chain, &table->chains, list) __nf_tables_unregister_hook(net, table, chain, true); list_for_each_entry(flowtable, &table->flowtables, list) - __nft_unregister_flowtable_net_hooks(net, &flowtable->hook_list, + __nft_unregister_flowtable_net_hooks(net, flowtable, + &flowtable->hook_list, true); } diff --git a/net/netrom/nr_route.c b/net/netrom/nr_route.c index 2b5e246b8d9a..b94cb2ffbaf8 100644 --- a/net/netrom/nr_route.c +++ b/net/netrom/nr_route.c @@ -754,6 +754,12 @@ int nr_route_frame(struct sk_buff *skb, ax25_cb *ax25) int ret; struct sk_buff *skbn; + /* + * Reject malformed packets early. Check that it contains at least 2 + * addresses and 1 byte more for Time-To-Live + */ + if (skb->len < 2 * sizeof(ax25_address) + 1) + return 0; nr_src = (ax25_address *)(skb->data + 0); nr_dest = (ax25_address *)(skb->data + 7); diff --git a/net/openvswitch/actions.c b/net/openvswitch/actions.c index 16e260014684..704c858cf209 100644 --- a/net/openvswitch/actions.c +++ b/net/openvswitch/actions.c @@ -934,7 +934,9 @@ static void do_output(struct datapath *dp, struct sk_buff *skb, int out_port, { struct vport *vport = ovs_vport_rcu(dp, out_port); - if (likely(vport && netif_carrier_ok(vport->dev))) { + if (likely(vport && + netif_running(vport->dev) && + netif_carrier_ok(vport->dev))) { u16 mru = OVS_CB(skb)->mru; u32 cutlen = OVS_CB(skb)->cutlen; diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c index 886c0dd47b66..2d73769d67f4 100644 --- a/net/packet/af_packet.c +++ b/net/packet/af_packet.c @@ -538,10 +538,8 @@ static void *packet_current_frame(struct packet_sock *po, return packet_lookup_frame(po, rb, rb->head, status); } -static u16 vlan_get_tci(struct sk_buff *skb, struct net_device *dev) +static u16 vlan_get_tci(const struct sk_buff *skb, struct net_device *dev) { - u8 *skb_orig_data = skb->data; - int skb_orig_len = skb->len; struct vlan_hdr vhdr, *vh; unsigned int header_len; @@ -562,33 +560,21 @@ static u16 vlan_get_tci(struct sk_buff *skb, struct net_device *dev) else return 0; - skb_push(skb, skb->data - skb_mac_header(skb)); - vh = skb_header_pointer(skb, header_len, sizeof(vhdr), &vhdr); - if (skb_orig_data != skb->data) { - skb->data = skb_orig_data; - skb->len = skb_orig_len; - } + vh = skb_header_pointer(skb, skb_mac_offset(skb) + header_len, + sizeof(vhdr), &vhdr); if (unlikely(!vh)) return 0; return ntohs(vh->h_vlan_TCI); } -static __be16 vlan_get_protocol_dgram(struct sk_buff *skb) +static __be16 vlan_get_protocol_dgram(const struct sk_buff *skb) { __be16 proto = skb->protocol; - if (unlikely(eth_type_vlan(proto))) { - u8 *skb_orig_data = skb->data; - int skb_orig_len = skb->len; - - skb_push(skb, skb->data - skb_mac_header(skb)); - proto = __vlan_get_protocol(skb, proto, NULL); - if (skb_orig_data != skb->data) { - skb->data = skb_orig_data; - skb->len = skb_orig_len; - } - } + if (unlikely(eth_type_vlan(proto))) + proto = __vlan_get_protocol_offset(skb, proto, + skb_mac_offset(skb), NULL); return proto; } diff --git a/net/rds/tcp.c b/net/rds/tcp.c index 351ac1747224..0581c53e6517 100644 --- a/net/rds/tcp.c +++ b/net/rds/tcp.c @@ -61,8 +61,10 @@ static atomic_t rds_tcp_unloading = ATOMIC_INIT(0); static struct kmem_cache *rds_tcp_conn_slab; -static int rds_tcp_skbuf_handler(const struct ctl_table *ctl, int write, - void *buffer, size_t *lenp, loff_t *fpos); +static int rds_tcp_sndbuf_handler(const struct ctl_table *ctl, int write, + void *buffer, size_t *lenp, loff_t *fpos); +static int rds_tcp_rcvbuf_handler(const struct ctl_table *ctl, int write, + void *buffer, size_t *lenp, loff_t *fpos); static int rds_tcp_min_sndbuf = SOCK_MIN_SNDBUF; static int rds_tcp_min_rcvbuf = SOCK_MIN_RCVBUF; @@ -74,7 +76,7 @@ static struct ctl_table rds_tcp_sysctl_table[] = { /* data is per-net pointer */ .maxlen = sizeof(int), .mode = 0644, - .proc_handler = rds_tcp_skbuf_handler, + .proc_handler = rds_tcp_sndbuf_handler, .extra1 = &rds_tcp_min_sndbuf, }, #define RDS_TCP_RCVBUF 1 @@ -83,7 +85,7 @@ static struct ctl_table rds_tcp_sysctl_table[] = { /* data is per-net pointer */ .maxlen = sizeof(int), .mode = 0644, - .proc_handler = rds_tcp_skbuf_handler, + .proc_handler = rds_tcp_rcvbuf_handler, .extra1 = &rds_tcp_min_rcvbuf, }, }; @@ -682,10 +684,10 @@ static void rds_tcp_sysctl_reset(struct net *net) spin_unlock_irq(&rds_tcp_conn_lock); } -static int rds_tcp_skbuf_handler(const struct ctl_table *ctl, int write, +static int rds_tcp_skbuf_handler(struct rds_tcp_net *rtn, + const struct ctl_table *ctl, int write, void *buffer, size_t *lenp, loff_t *fpos) { - struct net *net = current->nsproxy->net_ns; int err; err = proc_dointvec_minmax(ctl, write, buffer, lenp, fpos); @@ -694,11 +696,34 @@ static int rds_tcp_skbuf_handler(const struct ctl_table *ctl, int write, *(int *)(ctl->extra1)); return err; } - if (write) + + if (write && rtn->rds_tcp_listen_sock && rtn->rds_tcp_listen_sock->sk) { + struct net *net = sock_net(rtn->rds_tcp_listen_sock->sk); + rds_tcp_sysctl_reset(net); + } + return 0; } +static int rds_tcp_sndbuf_handler(const struct ctl_table *ctl, int write, + void *buffer, size_t *lenp, loff_t *fpos) +{ + struct rds_tcp_net *rtn = container_of(ctl->data, struct rds_tcp_net, + sndbuf_size); + + return rds_tcp_skbuf_handler(rtn, ctl, write, buffer, lenp, fpos); +} + +static int rds_tcp_rcvbuf_handler(const struct ctl_table *ctl, int write, + void *buffer, size_t *lenp, loff_t *fpos) +{ + struct rds_tcp_net *rtn = container_of(ctl->data, struct rds_tcp_net, + rcvbuf_size); + + return rds_tcp_skbuf_handler(rtn, ctl, write, buffer, lenp, fpos); +} + static void rds_tcp_exit(void) { rds_tcp_set_unloading(); diff --git a/net/sched/cls_flow.c b/net/sched/cls_flow.c index 5502998aace7..5c2580a07530 100644 --- a/net/sched/cls_flow.c +++ b/net/sched/cls_flow.c @@ -356,7 +356,8 @@ static const struct nla_policy flow_policy[TCA_FLOW_MAX + 1] = { [TCA_FLOW_KEYS] = { .type = NLA_U32 }, [TCA_FLOW_MODE] = { .type = NLA_U32 }, [TCA_FLOW_BASECLASS] = { .type = NLA_U32 }, - [TCA_FLOW_RSHIFT] = { .type = NLA_U32 }, + [TCA_FLOW_RSHIFT] = NLA_POLICY_MAX(NLA_U32, + 31 /* BITS_PER_U32 - 1 */), [TCA_FLOW_ADDEND] = { .type = NLA_U32 }, [TCA_FLOW_MASK] = { .type = NLA_U32 }, [TCA_FLOW_XOR] = { .type = NLA_U32 }, diff --git a/net/sched/sch_cake.c b/net/sched/sch_cake.c index 8d8b2db4653c..2c2e2a67f3b2 100644 --- a/net/sched/sch_cake.c +++ b/net/sched/sch_cake.c @@ -627,6 +627,63 @@ static bool cake_ddst(int flow_mode) return (flow_mode & CAKE_FLOW_DUAL_DST) == CAKE_FLOW_DUAL_DST; } +static void cake_dec_srchost_bulk_flow_count(struct cake_tin_data *q, + struct cake_flow *flow, + int flow_mode) +{ + if (likely(cake_dsrc(flow_mode) && + q->hosts[flow->srchost].srchost_bulk_flow_count)) + q->hosts[flow->srchost].srchost_bulk_flow_count--; +} + +static void cake_inc_srchost_bulk_flow_count(struct cake_tin_data *q, + struct cake_flow *flow, + int flow_mode) +{ + if (likely(cake_dsrc(flow_mode) && + q->hosts[flow->srchost].srchost_bulk_flow_count < CAKE_QUEUES)) + q->hosts[flow->srchost].srchost_bulk_flow_count++; +} + +static void cake_dec_dsthost_bulk_flow_count(struct cake_tin_data *q, + struct cake_flow *flow, + int flow_mode) +{ + if (likely(cake_ddst(flow_mode) && + q->hosts[flow->dsthost].dsthost_bulk_flow_count)) + q->hosts[flow->dsthost].dsthost_bulk_flow_count--; +} + +static void cake_inc_dsthost_bulk_flow_count(struct cake_tin_data *q, + struct cake_flow *flow, + int flow_mode) +{ + if (likely(cake_ddst(flow_mode) && + q->hosts[flow->dsthost].dsthost_bulk_flow_count < CAKE_QUEUES)) + q->hosts[flow->dsthost].dsthost_bulk_flow_count++; +} + +static u16 cake_get_flow_quantum(struct cake_tin_data *q, + struct cake_flow *flow, + int flow_mode) +{ + u16 host_load = 1; + + if (cake_dsrc(flow_mode)) + host_load = max(host_load, + q->hosts[flow->srchost].srchost_bulk_flow_count); + + if (cake_ddst(flow_mode)) + host_load = max(host_load, + q->hosts[flow->dsthost].dsthost_bulk_flow_count); + + /* The get_random_u16() is a way to apply dithering to avoid + * accumulating roundoff errors + */ + return (q->flow_quantum * quantum_div[host_load] + + get_random_u16()) >> 16; +} + static u32 cake_hash(struct cake_tin_data *q, const struct sk_buff *skb, int flow_mode, u16 flow_override, u16 host_override) { @@ -773,10 +830,8 @@ skip_hash: allocate_dst = cake_ddst(flow_mode); if (q->flows[outer_hash + k].set == CAKE_SET_BULK) { - if (allocate_src) - q->hosts[q->flows[reduced_hash].srchost].srchost_bulk_flow_count--; - if (allocate_dst) - q->hosts[q->flows[reduced_hash].dsthost].dsthost_bulk_flow_count--; + cake_dec_srchost_bulk_flow_count(q, &q->flows[outer_hash + k], flow_mode); + cake_dec_dsthost_bulk_flow_count(q, &q->flows[outer_hash + k], flow_mode); } found: /* reserve queue for future packets in same flow */ @@ -801,9 +856,10 @@ found: q->hosts[outer_hash + k].srchost_tag = srchost_hash; found_src: srchost_idx = outer_hash + k; - if (q->flows[reduced_hash].set == CAKE_SET_BULK) - q->hosts[srchost_idx].srchost_bulk_flow_count++; q->flows[reduced_hash].srchost = srchost_idx; + + if (q->flows[reduced_hash].set == CAKE_SET_BULK) + cake_inc_srchost_bulk_flow_count(q, &q->flows[reduced_hash], flow_mode); } if (allocate_dst) { @@ -824,9 +880,10 @@ found_src: q->hosts[outer_hash + k].dsthost_tag = dsthost_hash; found_dst: dsthost_idx = outer_hash + k; - if (q->flows[reduced_hash].set == CAKE_SET_BULK) - q->hosts[dsthost_idx].dsthost_bulk_flow_count++; q->flows[reduced_hash].dsthost = dsthost_idx; + + if (q->flows[reduced_hash].set == CAKE_SET_BULK) + cake_inc_dsthost_bulk_flow_count(q, &q->flows[reduced_hash], flow_mode); } } @@ -1839,10 +1896,6 @@ static s32 cake_enqueue(struct sk_buff *skb, struct Qdisc *sch, /* flowchain */ if (!flow->set || flow->set == CAKE_SET_DECAYING) { - struct cake_host *srchost = &b->hosts[flow->srchost]; - struct cake_host *dsthost = &b->hosts[flow->dsthost]; - u16 host_load = 1; - if (!flow->set) { list_add_tail(&flow->flowchain, &b->new_flows); } else { @@ -1852,18 +1905,8 @@ static s32 cake_enqueue(struct sk_buff *skb, struct Qdisc *sch, flow->set = CAKE_SET_SPARSE; b->sparse_flow_count++; - if (cake_dsrc(q->flow_mode)) - host_load = max(host_load, srchost->srchost_bulk_flow_count); - - if (cake_ddst(q->flow_mode)) - host_load = max(host_load, dsthost->dsthost_bulk_flow_count); - - flow->deficit = (b->flow_quantum * - quantum_div[host_load]) >> 16; + flow->deficit = cake_get_flow_quantum(b, flow, q->flow_mode); } else if (flow->set == CAKE_SET_SPARSE_WAIT) { - struct cake_host *srchost = &b->hosts[flow->srchost]; - struct cake_host *dsthost = &b->hosts[flow->dsthost]; - /* this flow was empty, accounted as a sparse flow, but actually * in the bulk rotation. */ @@ -1871,12 +1914,8 @@ static s32 cake_enqueue(struct sk_buff *skb, struct Qdisc *sch, b->sparse_flow_count--; b->bulk_flow_count++; - if (cake_dsrc(q->flow_mode)) - srchost->srchost_bulk_flow_count++; - - if (cake_ddst(q->flow_mode)) - dsthost->dsthost_bulk_flow_count++; - + cake_inc_srchost_bulk_flow_count(b, flow, q->flow_mode); + cake_inc_dsthost_bulk_flow_count(b, flow, q->flow_mode); } if (q->buffer_used > q->buffer_max_used) @@ -1933,13 +1972,11 @@ static struct sk_buff *cake_dequeue(struct Qdisc *sch) { struct cake_sched_data *q = qdisc_priv(sch); struct cake_tin_data *b = &q->tins[q->cur_tin]; - struct cake_host *srchost, *dsthost; ktime_t now = ktime_get(); struct cake_flow *flow; struct list_head *head; bool first_flow = true; struct sk_buff *skb; - u16 host_load; u64 delay; u32 len; @@ -2039,11 +2076,6 @@ retry: q->cur_flow = flow - b->flows; first_flow = false; - /* triple isolation (modified DRR++) */ - srchost = &b->hosts[flow->srchost]; - dsthost = &b->hosts[flow->dsthost]; - host_load = 1; - /* flow isolation (DRR++) */ if (flow->deficit <= 0) { /* Keep all flows with deficits out of the sparse and decaying @@ -2055,11 +2087,8 @@ retry: b->sparse_flow_count--; b->bulk_flow_count++; - if (cake_dsrc(q->flow_mode)) - srchost->srchost_bulk_flow_count++; - - if (cake_ddst(q->flow_mode)) - dsthost->dsthost_bulk_flow_count++; + cake_inc_srchost_bulk_flow_count(b, flow, q->flow_mode); + cake_inc_dsthost_bulk_flow_count(b, flow, q->flow_mode); flow->set = CAKE_SET_BULK; } else { @@ -2071,19 +2100,7 @@ retry: } } - if (cake_dsrc(q->flow_mode)) - host_load = max(host_load, srchost->srchost_bulk_flow_count); - - if (cake_ddst(q->flow_mode)) - host_load = max(host_load, dsthost->dsthost_bulk_flow_count); - - WARN_ON(host_load > CAKE_QUEUES); - - /* The get_random_u16() is a way to apply dithering to avoid - * accumulating roundoff errors - */ - flow->deficit += (b->flow_quantum * quantum_div[host_load] + - get_random_u16()) >> 16; + flow->deficit += cake_get_flow_quantum(b, flow, q->flow_mode); list_move_tail(&flow->flowchain, &b->old_flows); goto retry; @@ -2107,11 +2124,8 @@ retry: if (flow->set == CAKE_SET_BULK) { b->bulk_flow_count--; - if (cake_dsrc(q->flow_mode)) - srchost->srchost_bulk_flow_count--; - - if (cake_ddst(q->flow_mode)) - dsthost->dsthost_bulk_flow_count--; + cake_dec_srchost_bulk_flow_count(b, flow, q->flow_mode); + cake_dec_dsthost_bulk_flow_count(b, flow, q->flow_mode); b->decaying_flow_count++; } else if (flow->set == CAKE_SET_SPARSE || @@ -2129,12 +2143,8 @@ retry: else if (flow->set == CAKE_SET_BULK) { b->bulk_flow_count--; - if (cake_dsrc(q->flow_mode)) - srchost->srchost_bulk_flow_count--; - - if (cake_ddst(q->flow_mode)) - dsthost->dsthost_bulk_flow_count--; - + cake_dec_srchost_bulk_flow_count(b, flow, q->flow_mode); + cake_dec_dsthost_bulk_flow_count(b, flow, q->flow_mode); } else b->decaying_flow_count--; diff --git a/net/sctp/associola.c b/net/sctp/associola.c index c45c192b7878..0b0794f164cf 100644 --- a/net/sctp/associola.c +++ b/net/sctp/associola.c @@ -137,7 +137,8 @@ static struct sctp_association *sctp_association_init( = 5 * asoc->rto_max; asoc->timeouts[SCTP_EVENT_TIMEOUT_SACK] = asoc->sackdelay; - asoc->timeouts[SCTP_EVENT_TIMEOUT_AUTOCLOSE] = sp->autoclose * HZ; + asoc->timeouts[SCTP_EVENT_TIMEOUT_AUTOCLOSE] = + (unsigned long)sp->autoclose * HZ; /* Initializes the timers */ for (i = SCTP_EVENT_TIMEOUT_NONE; i < SCTP_NUM_TIMEOUT_TYPES; ++i) diff --git a/net/sctp/sysctl.c b/net/sctp/sysctl.c index e5a5af343c4c..8e1e97be4df7 100644 --- a/net/sctp/sysctl.c +++ b/net/sctp/sysctl.c @@ -387,7 +387,8 @@ static struct ctl_table sctp_net_table[] = { static int proc_sctp_do_hmac_alg(const struct ctl_table *ctl, int write, void *buffer, size_t *lenp, loff_t *ppos) { - struct net *net = current->nsproxy->net_ns; + struct net *net = container_of(ctl->data, struct net, + sctp.sctp_hmac_alg); struct ctl_table tbl; bool changed = false; char *none = "none"; @@ -432,7 +433,7 @@ static int proc_sctp_do_hmac_alg(const struct ctl_table *ctl, int write, static int proc_sctp_do_rto_min(const struct ctl_table *ctl, int write, void *buffer, size_t *lenp, loff_t *ppos) { - struct net *net = current->nsproxy->net_ns; + struct net *net = container_of(ctl->data, struct net, sctp.rto_min); unsigned int min = *(unsigned int *) ctl->extra1; unsigned int max = *(unsigned int *) ctl->extra2; struct ctl_table tbl; @@ -460,7 +461,7 @@ static int proc_sctp_do_rto_min(const struct ctl_table *ctl, int write, static int proc_sctp_do_rto_max(const struct ctl_table *ctl, int write, void *buffer, size_t *lenp, loff_t *ppos) { - struct net *net = current->nsproxy->net_ns; + struct net *net = container_of(ctl->data, struct net, sctp.rto_max); unsigned int min = *(unsigned int *) ctl->extra1; unsigned int max = *(unsigned int *) ctl->extra2; struct ctl_table tbl; @@ -498,7 +499,7 @@ static int proc_sctp_do_alpha_beta(const struct ctl_table *ctl, int write, static int proc_sctp_do_auth(const struct ctl_table *ctl, int write, void *buffer, size_t *lenp, loff_t *ppos) { - struct net *net = current->nsproxy->net_ns; + struct net *net = container_of(ctl->data, struct net, sctp.auth_enable); struct ctl_table tbl; int new_value, ret; @@ -527,7 +528,7 @@ static int proc_sctp_do_auth(const struct ctl_table *ctl, int write, static int proc_sctp_do_udp_port(const struct ctl_table *ctl, int write, void *buffer, size_t *lenp, loff_t *ppos) { - struct net *net = current->nsproxy->net_ns; + struct net *net = container_of(ctl->data, struct net, sctp.udp_port); unsigned int min = *(unsigned int *)ctl->extra1; unsigned int max = *(unsigned int *)ctl->extra2; struct ctl_table tbl; @@ -568,7 +569,8 @@ static int proc_sctp_do_udp_port(const struct ctl_table *ctl, int write, static int proc_sctp_do_probe_interval(const struct ctl_table *ctl, int write, void *buffer, size_t *lenp, loff_t *ppos) { - struct net *net = current->nsproxy->net_ns; + struct net *net = container_of(ctl->data, struct net, + sctp.probe_interval); struct ctl_table tbl; int ret, new_value; diff --git a/net/tls/tls_sw.c b/net/tls/tls_sw.c index bbf26cc4f6ee..7bcc9b4408a2 100644 --- a/net/tls/tls_sw.c +++ b/net/tls/tls_sw.c @@ -458,7 +458,7 @@ int tls_tx_records(struct sock *sk, int flags) tx_err: if (rc < 0 && rc != -EAGAIN) - tls_err_abort(sk, -EBADMSG); + tls_err_abort(sk, rc); return rc; } diff --git a/net/vmw_vsock/af_vsock.c b/net/vmw_vsock/af_vsock.c index 5cf8109f672a..fa9d1b49599b 100644 --- a/net/vmw_vsock/af_vsock.c +++ b/net/vmw_vsock/af_vsock.c @@ -491,6 +491,15 @@ int vsock_assign_transport(struct vsock_sock *vsk, struct vsock_sock *psk) */ vsk->transport->release(vsk); vsock_deassign_transport(vsk); + + /* transport's release() and destruct() can touch some socket + * state, since we are reassigning the socket to a new transport + * during vsock_connect(), let's reset these fields to have a + * clean state. + */ + sock_reset_flag(sk, SOCK_DONE); + sk->sk_state = TCP_CLOSE; + vsk->peer_shutdown = 0; } /* We increase the module refcnt to prevent the transport unloading @@ -870,6 +879,9 @@ EXPORT_SYMBOL_GPL(vsock_create_connected); s64 vsock_stream_has_data(struct vsock_sock *vsk) { + if (WARN_ON(!vsk->transport)) + return 0; + return vsk->transport->stream_has_data(vsk); } EXPORT_SYMBOL_GPL(vsock_stream_has_data); @@ -878,6 +890,9 @@ s64 vsock_connectible_has_data(struct vsock_sock *vsk) { struct sock *sk = sk_vsock(vsk); + if (WARN_ON(!vsk->transport)) + return 0; + if (sk->sk_type == SOCK_SEQPACKET) return vsk->transport->seqpacket_has_data(vsk); else @@ -887,6 +902,9 @@ EXPORT_SYMBOL_GPL(vsock_connectible_has_data); s64 vsock_stream_has_space(struct vsock_sock *vsk) { + if (WARN_ON(!vsk->transport)) + return 0; + return vsk->transport->stream_has_space(vsk); } EXPORT_SYMBOL_GPL(vsock_stream_has_space); diff --git a/net/vmw_vsock/virtio_transport_common.c b/net/vmw_vsock/virtio_transport_common.c index 9acc13ab3f82..7f7de6d88096 100644 --- a/net/vmw_vsock/virtio_transport_common.c +++ b/net/vmw_vsock/virtio_transport_common.c @@ -26,6 +26,9 @@ /* Threshold for detecting small packets to copy */ #define GOOD_COPY_LEN 128 +static void virtio_transport_cancel_close_work(struct vsock_sock *vsk, + bool cancel_timeout); + static const struct virtio_transport * virtio_transport_get_ops(struct vsock_sock *vsk) { @@ -1109,6 +1112,8 @@ void virtio_transport_destruct(struct vsock_sock *vsk) { struct virtio_vsock_sock *vvs = vsk->trans; + virtio_transport_cancel_close_work(vsk, true); + kfree(vvs); vsk->trans = NULL; } @@ -1204,17 +1209,11 @@ static void virtio_transport_wait_close(struct sock *sk, long timeout) } } -static void virtio_transport_do_close(struct vsock_sock *vsk, - bool cancel_timeout) +static void virtio_transport_cancel_close_work(struct vsock_sock *vsk, + bool cancel_timeout) { struct sock *sk = sk_vsock(vsk); - sock_set_flag(sk, SOCK_DONE); - vsk->peer_shutdown = SHUTDOWN_MASK; - if (vsock_stream_has_data(vsk) <= 0) - sk->sk_state = TCP_CLOSING; - sk->sk_state_change(sk); - if (vsk->close_work_scheduled && (!cancel_timeout || cancel_delayed_work(&vsk->close_work))) { vsk->close_work_scheduled = false; @@ -1226,6 +1225,20 @@ static void virtio_transport_do_close(struct vsock_sock *vsk, } } +static void virtio_transport_do_close(struct vsock_sock *vsk, + bool cancel_timeout) +{ + struct sock *sk = sk_vsock(vsk); + + sock_set_flag(sk, SOCK_DONE); + vsk->peer_shutdown = SHUTDOWN_MASK; + if (vsock_stream_has_data(vsk) <= 0) + sk->sk_state = TCP_CLOSING; + sk->sk_state_change(sk); + + virtio_transport_cancel_close_work(vsk, cancel_timeout); +} + static void virtio_transport_close_timeout(struct work_struct *work) { struct vsock_sock *vsk = @@ -1628,8 +1641,11 @@ void virtio_transport_recv_pkt(struct virtio_transport *t, lock_sock(sk); - /* Check if sk has been closed before lock_sock */ - if (sock_flag(sk, SOCK_DONE)) { + /* Check if sk has been closed or assigned to another transport before + * lock_sock (note: listener sockets are not assigned to any transport) + */ + if (sock_flag(sk, SOCK_DONE) || + (sk->sk_state != TCP_LISTEN && vsk->transport != &t->transport)) { (void)virtio_transport_reset_no_sock(t, skb); release_sock(sk); sock_put(sk); diff --git a/net/vmw_vsock/vsock_bpf.c b/net/vmw_vsock/vsock_bpf.c index 4aa6e74ec295..f201d9eca1df 100644 --- a/net/vmw_vsock/vsock_bpf.c +++ b/net/vmw_vsock/vsock_bpf.c @@ -77,6 +77,7 @@ static int vsock_bpf_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, int flags, int *addr_len) { struct sk_psock *psock; + struct vsock_sock *vsk; int copied; psock = sk_psock_get(sk); @@ -84,6 +85,13 @@ static int vsock_bpf_recvmsg(struct sock *sk, struct msghdr *msg, return __vsock_recvmsg(sk, msg, len, flags); lock_sock(sk); + vsk = vsock_sk(sk); + + if (!vsk->transport) { + copied = -ENODEV; + goto out; + } + if (vsock_has_data(sk, psock) && sk_psock_queue_empty(psock)) { release_sock(sk); sk_psock_put(sk, psock); @@ -108,6 +116,7 @@ static int vsock_bpf_recvmsg(struct sock *sk, struct msghdr *msg, copied = sk_msg_recvmsg(sk, psock, msg, len, flags); } +out: release_sock(sk); sk_psock_put(sk, psock); diff --git a/net/xdp/xsk.c b/net/xdp/xsk.c index 3fa70286c846..89d2bef96469 100644 --- a/net/xdp/xsk.c +++ b/net/xdp/xsk.c @@ -322,7 +322,6 @@ static int xsk_rcv_check(struct xdp_sock *xs, struct xdp_buff *xdp, u32 len) return -ENOSPC; } - sk_mark_napi_id_once_xdp(&xs->sk, xdp); return 0; } @@ -908,11 +907,8 @@ static int __xsk_sendmsg(struct socket *sock, struct msghdr *m, size_t total_len if (unlikely(!xs->tx)) return -ENOBUFS; - if (sk_can_busy_loop(sk)) { - if (xs->zc) - __sk_mark_napi_id_once(sk, xsk_pool_get_napi_id(xs->pool)); + if (sk_can_busy_loop(sk)) sk_busy_loop(sk, 1); /* only support non-blocking sockets */ - } if (xs->zc && xsk_no_wakeup(sk)) return 0; @@ -1298,6 +1294,14 @@ static int xsk_bind(struct socket *sock, struct sockaddr *addr, int addr_len) xs->queue_id = qid; xp_add_xsk(xs->pool, xs); + if (xs->zc && qid < dev->real_num_rx_queues) { + struct netdev_rx_queue *rxq; + + rxq = __netif_get_rx_queue(dev, qid); + if (rxq->napi) + __sk_mark_napi_id_once(sk, rxq->napi->napi_id); + } + out_unlock: if (err) { dev_put(dev); diff --git a/rust/kernel/workqueue.rs b/rust/kernel/workqueue.rs index 4d1d2062f6eb..fd3e97192ed8 100644 --- a/rust/kernel/workqueue.rs +++ b/rust/kernel/workqueue.rs @@ -519,7 +519,15 @@ impl_has_work! { impl{T} HasWork<Self> for ClosureWork<T> { self.work } } -// SAFETY: TODO. +// SAFETY: The `__enqueue` implementation in RawWorkItem uses a `work_struct` initialized with the +// `run` method of this trait as the function pointer because: +// - `__enqueue` gets the `work_struct` from the `Work` field, using `T::raw_get_work`. +// - The only safe way to create a `Work` object is through `Work::new`. +// - `Work::new` makes sure that `T::Pointer::run` is passed to `init_work_with_key`. +// - Finally `Work` and `RawWorkItem` guarantee that the correct `Work` field +// will be used because of the ID const generic bound. This makes sure that `T::raw_get_work` +// uses the correct offset for the `Work` field, and `Work::new` picks the correct +// implementation of `WorkItemPointer` for `Arc<T>`. unsafe impl<T, const ID: u64> WorkItemPointer<ID> for Arc<T> where T: WorkItem<ID, Pointer = Self>, @@ -537,7 +545,13 @@ where } } -// SAFETY: TODO. +// SAFETY: The `work_struct` raw pointer is guaranteed to be valid for the duration of the call to +// the closure because we get it from an `Arc`, which means that the ref count will be at least 1, +// and we don't drop the `Arc` ourselves. If `queue_work_on` returns true, it is further guaranteed +// to be valid until a call to the function pointer in `work_struct` because we leak the memory it +// points to, and only reclaim it if the closure returns false, or in `WorkItemPointer::run`, which +// is what the function pointer in the `work_struct` must be pointing to, according to the safety +// requirements of `WorkItemPointer`. unsafe impl<T, const ID: u64> RawWorkItem<ID> for Arc<T> where T: WorkItem<ID, Pointer = Self>, diff --git a/samples/vfs/.gitignore b/samples/vfs/.gitignore index 79212d91285b..8708341bc082 100644 --- a/samples/vfs/.gitignore +++ b/samples/vfs/.gitignore @@ -1,3 +1,5 @@ # SPDX-License-Identifier: GPL-2.0-only /test-fsmount +/test-list-all-mounts /test-statx +/mountinfo diff --git a/samples/vfs/Makefile b/samples/vfs/Makefile index 6377a678134a..6554b73a75c8 100644 --- a/samples/vfs/Makefile +++ b/samples/vfs/Makefile @@ -1,4 +1,4 @@ # SPDX-License-Identifier: GPL-2.0-only -userprogs-always-y += test-fsmount test-statx +userprogs-always-y += test-fsmount test-statx mountinfo test-list-all-mounts userccflags += -I usr/include diff --git a/samples/vfs/mountinfo.c b/samples/vfs/mountinfo.c new file mode 100644 index 000000000000..bc78275cac69 --- /dev/null +++ b/samples/vfs/mountinfo.c @@ -0,0 +1,274 @@ +// SPDX-License-Identifier: GPL-2.0-or-later + +/* + * Use pidfds, nsfds, listmount() and statmount() mimic the + * contents of /proc/self/mountinfo. + */ +#define _GNU_SOURCE +#define __SANE_USERSPACE_TYPES__ +#include <stdio.h> +#include <stdint.h> +#include <unistd.h> +#include <alloca.h> +#include <getopt.h> +#include <stdlib.h> +#include <stdbool.h> +#include <errno.h> + +#include "samples-vfs.h" + +/* max mounts per listmount call */ +#define MAXMOUNTS 1024 + +/* size of struct statmount (including trailing string buffer) */ +#define STATMOUNT_BUFSIZE 4096 + +static bool ext_format; + +#ifndef __NR_pidfd_open +#define __NR_pidfd_open -1 +#endif + +/* + * There are no bindings in glibc for listmount() and statmount() (yet), + * make our own here. + */ +static int statmount(__u64 mnt_id, __u64 mnt_ns_id, __u64 mask, + struct statmount *buf, size_t bufsize, + unsigned int flags) +{ + struct mnt_id_req req = { + .size = MNT_ID_REQ_SIZE_VER0, + .mnt_id = mnt_id, + .param = mask, + }; + + if (mnt_ns_id) { + req.size = MNT_ID_REQ_SIZE_VER1; + req.mnt_ns_id = mnt_ns_id; + } + + return syscall(__NR_statmount, &req, buf, bufsize, flags); +} + +static ssize_t listmount(__u64 mnt_id, __u64 mnt_ns_id, __u64 last_mnt_id, + __u64 list[], size_t num, unsigned int flags) +{ + struct mnt_id_req req = { + .size = MNT_ID_REQ_SIZE_VER0, + .mnt_id = mnt_id, + .param = last_mnt_id, + }; + + if (mnt_ns_id) { + req.size = MNT_ID_REQ_SIZE_VER1; + req.mnt_ns_id = mnt_ns_id; + } + + return syscall(__NR_listmount, &req, list, num, flags); +} + +static void show_mnt_attrs(__u64 flags) +{ + printf("%s", flags & MOUNT_ATTR_RDONLY ? "ro" : "rw"); + + if (flags & MOUNT_ATTR_NOSUID) + printf(",nosuid"); + if (flags & MOUNT_ATTR_NODEV) + printf(",nodev"); + if (flags & MOUNT_ATTR_NOEXEC) + printf(",noexec"); + + switch (flags & MOUNT_ATTR__ATIME) { + case MOUNT_ATTR_RELATIME: + printf(",relatime"); + break; + case MOUNT_ATTR_NOATIME: + printf(",noatime"); + break; + case MOUNT_ATTR_STRICTATIME: + /* print nothing */ + break; + } + + if (flags & MOUNT_ATTR_NODIRATIME) + printf(",nodiratime"); + if (flags & MOUNT_ATTR_NOSYMFOLLOW) + printf(",nosymfollow"); + if (flags & MOUNT_ATTR_IDMAP) + printf(",idmapped"); +} + +static void show_propagation(struct statmount *sm) +{ + if (sm->mnt_propagation & MS_SHARED) + printf(" shared:%llu", sm->mnt_peer_group); + if (sm->mnt_propagation & MS_SLAVE) { + printf(" master:%llu", sm->mnt_master); + if (sm->propagate_from && sm->propagate_from != sm->mnt_master) + printf(" propagate_from:%llu", sm->propagate_from); + } + if (sm->mnt_propagation & MS_UNBINDABLE) + printf(" unbindable"); +} + +static void show_sb_flags(__u64 flags) +{ + printf("%s", flags & MS_RDONLY ? "ro" : "rw"); + if (flags & MS_SYNCHRONOUS) + printf(",sync"); + if (flags & MS_DIRSYNC) + printf(",dirsync"); + if (flags & MS_MANDLOCK) + printf(",mand"); + if (flags & MS_LAZYTIME) + printf(",lazytime"); +} + +static int dump_mountinfo(__u64 mnt_id, __u64 mnt_ns_id) +{ + int ret; + struct statmount *buf = alloca(STATMOUNT_BUFSIZE); + const __u64 mask = STATMOUNT_SB_BASIC | STATMOUNT_MNT_BASIC | + STATMOUNT_PROPAGATE_FROM | STATMOUNT_FS_TYPE | + STATMOUNT_MNT_ROOT | STATMOUNT_MNT_POINT | + STATMOUNT_MNT_OPTS | STATMOUNT_FS_SUBTYPE | + STATMOUNT_SB_SOURCE; + + ret = statmount(mnt_id, mnt_ns_id, mask, buf, STATMOUNT_BUFSIZE, 0); + if (ret < 0) { + perror("statmount"); + return 1; + } + + if (ext_format) + printf("0x%llx 0x%llx 0x%llx ", mnt_ns_id, mnt_id, buf->mnt_parent_id); + + printf("%u %u %u:%u %s %s ", buf->mnt_id_old, buf->mnt_parent_id_old, + buf->sb_dev_major, buf->sb_dev_minor, + &buf->str[buf->mnt_root], + &buf->str[buf->mnt_point]); + show_mnt_attrs(buf->mnt_attr); + show_propagation(buf); + + printf(" - %s", &buf->str[buf->fs_type]); + if (buf->mask & STATMOUNT_FS_SUBTYPE) + printf(".%s", &buf->str[buf->fs_subtype]); + if (buf->mask & STATMOUNT_SB_SOURCE) + printf(" %s ", &buf->str[buf->sb_source]); + else + printf(" :none "); + + show_sb_flags(buf->sb_flags); + if (buf->mask & STATMOUNT_MNT_OPTS) + printf(",%s", &buf->str[buf->mnt_opts]); + printf("\n"); + return 0; +} + +static int dump_mounts(__u64 mnt_ns_id) +{ + __u64 mntid[MAXMOUNTS]; + __u64 last_mnt_id = 0; + ssize_t count; + int i; + + /* + * Get a list of all mntids in mnt_ns_id. If it returns MAXMOUNTS + * mounts, then go again until we get everything. + */ + do { + count = listmount(LSMT_ROOT, mnt_ns_id, last_mnt_id, mntid, MAXMOUNTS, 0); + if (count < 0 || count > MAXMOUNTS) { + errno = count < 0 ? errno : count; + perror("listmount"); + return 1; + } + + /* Walk the returned mntids and print info about each */ + for (i = 0; i < count; ++i) { + int ret = dump_mountinfo(mntid[i], mnt_ns_id); + + if (ret != 0) + return ret; + } + /* Set up last_mnt_id to pick up where we left off */ + last_mnt_id = mntid[count - 1]; + } while (count == MAXMOUNTS); + return 0; +} + +static void usage(const char * const prog) +{ + printf("Usage:\n"); + printf("%s [-e] [-p pid] [-r] [-h]\n", prog); + printf(" -e: extended format\n"); + printf(" -h: print usage message\n"); + printf(" -p: get mount namespace from given pid\n"); + printf(" -r: recursively print all mounts in all child namespaces\n"); +} + +int main(int argc, char * const *argv) +{ + struct mnt_ns_info mni = { .size = MNT_NS_INFO_SIZE_VER0 }; + int pidfd, mntns, ret, opt; + pid_t pid = getpid(); + bool recursive = false; + + while ((opt = getopt(argc, argv, "ehp:r")) != -1) { + switch (opt) { + case 'e': + ext_format = true; + break; + case 'h': + usage(argv[0]); + return 0; + case 'p': + pid = atoi(optarg); + break; + case 'r': + recursive = true; + break; + } + } + + /* Get a pidfd for pid */ + pidfd = syscall(__NR_pidfd_open, pid, 0); + if (pidfd < 0) { + perror("pidfd_open"); + return 1; + } + + /* Get the mnt namespace for pidfd */ + mntns = ioctl(pidfd, PIDFD_GET_MNT_NAMESPACE, NULL); + if (mntns < 0) { + perror("PIDFD_GET_MNT_NAMESPACE"); + return 1; + } + close(pidfd); + + /* get info about mntns. In particular, the mnt_ns_id */ + ret = ioctl(mntns, NS_MNT_GET_INFO, &mni); + if (ret < 0) { + perror("NS_MNT_GET_INFO"); + return 1; + } + + do { + int ret; + + ret = dump_mounts(mni.mnt_ns_id); + if (ret) + return ret; + + if (!recursive) + break; + + /* get the next mntns (and overwrite the old mount ns info) */ + ret = ioctl(mntns, NS_MNT_GET_NEXT, &mni); + close(mntns); + mntns = ret; + } while (mntns >= 0); + + return 0; +} diff --git a/samples/vfs/samples-vfs.h b/samples/vfs/samples-vfs.h new file mode 100644 index 000000000000..103e1e7c4cec --- /dev/null +++ b/samples/vfs/samples-vfs.h @@ -0,0 +1,241 @@ +/* SPDX-License-Identifier: GPL-2.0 */ + +#ifndef __SAMPLES_VFS_H +#define __SAMPLES_VFS_H + +#include <errno.h> +#include <linux/types.h> +#include <sys/ioctl.h> +#include <sys/syscall.h> + +#define die_errno(format, ...) \ + do { \ + fprintf(stderr, "%m | %s: %d: %s: " format "\n", __FILE__, \ + __LINE__, __func__, ##__VA_ARGS__); \ + exit(EXIT_FAILURE); \ + } while (0) + +struct statmount { + __u32 size; /* Total size, including strings */ + __u32 mnt_opts; /* [str] Options (comma separated, escaped) */ + __u64 mask; /* What results were written */ + __u32 sb_dev_major; /* Device ID */ + __u32 sb_dev_minor; + __u64 sb_magic; /* ..._SUPER_MAGIC */ + __u32 sb_flags; /* SB_{RDONLY,SYNCHRONOUS,DIRSYNC,LAZYTIME} */ + __u32 fs_type; /* [str] Filesystem type */ + __u64 mnt_id; /* Unique ID of mount */ + __u64 mnt_parent_id; /* Unique ID of parent (for root == mnt_id) */ + __u32 mnt_id_old; /* Reused IDs used in proc/.../mountinfo */ + __u32 mnt_parent_id_old; + __u64 mnt_attr; /* MOUNT_ATTR_... */ + __u64 mnt_propagation; /* MS_{SHARED,SLAVE,PRIVATE,UNBINDABLE} */ + __u64 mnt_peer_group; /* ID of shared peer group */ + __u64 mnt_master; /* Mount receives propagation from this ID */ + __u64 propagate_from; /* Propagation from in current namespace */ + __u32 mnt_root; /* [str] Root of mount relative to root of fs */ + __u32 mnt_point; /* [str] Mountpoint relative to current root */ + __u64 mnt_ns_id; /* ID of the mount namespace */ + __u32 fs_subtype; /* [str] Subtype of fs_type (if any) */ + __u32 sb_source; /* [str] Source string of the mount */ + __u32 opt_num; /* Number of fs options */ + __u32 opt_array; /* [str] Array of nul terminated fs options */ + __u32 opt_sec_num; /* Number of security options */ + __u32 opt_sec_array; /* [str] Array of nul terminated security options */ + __u64 __spare2[46]; + char str[]; /* Variable size part containing strings */ +}; + +struct mnt_id_req { + __u32 size; + __u32 spare; + __u64 mnt_id; + __u64 param; + __u64 mnt_ns_id; +}; + +#ifndef MNT_ID_REQ_SIZE_VER0 +#define MNT_ID_REQ_SIZE_VER0 24 /* sizeof first published struct */ +#endif + +#ifndef MNT_ID_REQ_SIZE_VER1 +#define MNT_ID_REQ_SIZE_VER1 32 /* sizeof second published struct */ +#endif + +/* Get the id for a mount namespace */ +#ifndef NS_GET_MNTNS_ID +#define NS_GET_MNTNS_ID _IO(0xb7, 0x5) +#endif + +struct mnt_ns_info { + __u32 size; + __u32 nr_mounts; + __u64 mnt_ns_id; +}; + +#ifndef MNT_NS_INFO_SIZE_VER0 +#define MNT_NS_INFO_SIZE_VER0 16 /* size of first published struct */ +#endif + +#ifndef NS_MNT_GET_INFO +#define NS_MNT_GET_INFO _IOR(0xb7, 10, struct mnt_ns_info) +#endif + +#ifndef NS_MNT_GET_NEXT +#define NS_MNT_GET_NEXT _IOR(0xb7, 11, struct mnt_ns_info) +#endif + +#ifndef NS_MNT_GET_PREV +#define NS_MNT_GET_PREV _IOR(0xb7, 12, struct mnt_ns_info) +#endif + +#ifndef PIDFD_GET_MNT_NAMESPACE +#define PIDFD_GET_MNT_NAMESPACE _IO(0xFF, 3) +#endif + +#ifndef __NR_listmount +#define __NR_listmount 458 +#endif + +#ifndef __NR_statmount +#define __NR_statmount 457 +#endif + +#ifndef LSMT_ROOT +#define LSMT_ROOT 0xffffffffffffffff /* root mount */ +#endif + +/* @mask bits for statmount(2) */ +#ifndef STATMOUNT_SB_BASIC +#define STATMOUNT_SB_BASIC 0x00000001U /* Want/got sb_... */ +#endif + +#ifndef STATMOUNT_MNT_BASIC +#define STATMOUNT_MNT_BASIC 0x00000002U /* Want/got mnt_... */ +#endif + +#ifndef STATMOUNT_PROPAGATE_FROM +#define STATMOUNT_PROPAGATE_FROM 0x00000004U /* Want/got propagate_from */ +#endif + +#ifndef STATMOUNT_MNT_ROOT +#define STATMOUNT_MNT_ROOT 0x00000008U /* Want/got mnt_root */ +#endif + +#ifndef STATMOUNT_MNT_POINT +#define STATMOUNT_MNT_POINT 0x00000010U /* Want/got mnt_point */ +#endif + +#ifndef STATMOUNT_FS_TYPE +#define STATMOUNT_FS_TYPE 0x00000020U /* Want/got fs_type */ +#endif + +#ifndef STATMOUNT_MNT_NS_ID +#define STATMOUNT_MNT_NS_ID 0x00000040U /* Want/got mnt_ns_id */ +#endif + +#ifndef STATMOUNT_MNT_OPTS +#define STATMOUNT_MNT_OPTS 0x00000080U /* Want/got mnt_opts */ +#endif + +#ifndef STATMOUNT_FS_SUBTYPE +#define STATMOUNT_FS_SUBTYPE 0x00000100U /* Want/got fs_subtype */ +#endif + +#ifndef STATMOUNT_SB_SOURCE +#define STATMOUNT_SB_SOURCE 0x00000200U /* Want/got sb_source */ +#endif + +#ifndef STATMOUNT_OPT_ARRAY +#define STATMOUNT_OPT_ARRAY 0x00000400U /* Want/got opt_... */ +#endif + +#ifndef STATMOUNT_OPT_SEC_ARRAY +#define STATMOUNT_OPT_SEC_ARRAY 0x00000800U /* Want/got opt_sec... */ +#endif + +#ifndef STATX_MNT_ID_UNIQUE +#define STATX_MNT_ID_UNIQUE 0x00004000U /* Want/got extended stx_mount_id */ +#endif + +#ifndef MOUNT_ATTR_RDONLY +#define MOUNT_ATTR_RDONLY 0x00000001 /* Mount read-only */ +#endif + +#ifndef MOUNT_ATTR_NOSUID +#define MOUNT_ATTR_NOSUID 0x00000002 /* Ignore suid and sgid bits */ +#endif + +#ifndef MOUNT_ATTR_NODEV +#define MOUNT_ATTR_NODEV 0x00000004 /* Disallow access to device special files */ +#endif + +#ifndef MOUNT_ATTR_NOEXEC +#define MOUNT_ATTR_NOEXEC 0x00000008 /* Disallow program execution */ +#endif + +#ifndef MOUNT_ATTR__ATIME +#define MOUNT_ATTR__ATIME 0x00000070 /* Setting on how atime should be updated */ +#endif + +#ifndef MOUNT_ATTR_RELATIME +#define MOUNT_ATTR_RELATIME 0x00000000 /* - Update atime relative to mtime/ctime. */ +#endif + +#ifndef MOUNT_ATTR_NOATIME +#define MOUNT_ATTR_NOATIME 0x00000010 /* - Do not update access times. */ +#endif + +#ifndef MOUNT_ATTR_STRICTATIME +#define MOUNT_ATTR_STRICTATIME 0x00000020 /* - Always perform atime updates */ +#endif + +#ifndef MOUNT_ATTR_NODIRATIME +#define MOUNT_ATTR_NODIRATIME 0x00000080 /* Do not update directory access times */ +#endif + +#ifndef MOUNT_ATTR_IDMAP +#define MOUNT_ATTR_IDMAP 0x00100000 /* Idmap mount to @userns_fd in struct mount_attr. */ +#endif + +#ifndef MOUNT_ATTR_NOSYMFOLLOW +#define MOUNT_ATTR_NOSYMFOLLOW 0x00200000 /* Do not follow symlinks */ +#endif + +#ifndef MS_RDONLY +#define MS_RDONLY 1 /* Mount read-only */ +#endif + +#ifndef MS_SYNCHRONOUS +#define MS_SYNCHRONOUS 16 /* Writes are synced at once */ +#endif + +#ifndef MS_MANDLOCK +#define MS_MANDLOCK 64 /* Allow mandatory locks on an FS */ +#endif + +#ifndef MS_DIRSYNC +#define MS_DIRSYNC 128 /* Directory modifications are synchronous */ +#endif + +#ifndef MS_UNBINDABLE +#define MS_UNBINDABLE (1<<17) /* change to unbindable */ +#endif + +#ifndef MS_PRIVATE +#define MS_PRIVATE (1<<18) /* change to private */ +#endif + +#ifndef MS_SLAVE +#define MS_SLAVE (1<<19) /* change to slave */ +#endif + +#ifndef MS_SHARED +#define MS_SHARED (1<<20) /* change to shared */ +#endif + +#ifndef MS_LAZYTIME +#define MS_LAZYTIME (1<<25) /* Update the on-disk [acm]times lazily */ +#endif + +#endif /* __SAMPLES_VFS_H */ diff --git a/samples/vfs/test-list-all-mounts.c b/samples/vfs/test-list-all-mounts.c new file mode 100644 index 000000000000..1a02ea4593e3 --- /dev/null +++ b/samples/vfs/test-list-all-mounts.c @@ -0,0 +1,150 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +// Copyright (c) 2024 Christian Brauner <brauner@kernel.org> + +#define _GNU_SOURCE +#include <errno.h> +#include <limits.h> +#include <linux/types.h> +#include <inttypes.h> +#include <stdio.h> + +#include "../../tools/testing/selftests/pidfd/pidfd.h" +#include "samples-vfs.h" + +static int __statmount(__u64 mnt_id, __u64 mnt_ns_id, __u64 mask, + struct statmount *stmnt, size_t bufsize, + unsigned int flags) +{ + struct mnt_id_req req = { + .size = MNT_ID_REQ_SIZE_VER1, + .mnt_id = mnt_id, + .param = mask, + .mnt_ns_id = mnt_ns_id, + }; + + return syscall(__NR_statmount, &req, stmnt, bufsize, flags); +} + +static struct statmount *sys_statmount(__u64 mnt_id, __u64 mnt_ns_id, + __u64 mask, unsigned int flags) +{ + size_t bufsize = 1 << 15; + struct statmount *stmnt = NULL, *tmp = NULL; + int ret; + + for (;;) { + tmp = realloc(stmnt, bufsize); + if (!tmp) + goto out; + + stmnt = tmp; + ret = __statmount(mnt_id, mnt_ns_id, mask, stmnt, bufsize, flags); + if (!ret) + return stmnt; + + if (errno != EOVERFLOW) + goto out; + + bufsize <<= 1; + if (bufsize >= UINT_MAX / 2) + goto out; + } + +out: + free(stmnt); + return NULL; +} + +static ssize_t sys_listmount(__u64 mnt_id, __u64 last_mnt_id, __u64 mnt_ns_id, + __u64 list[], size_t num, unsigned int flags) +{ + struct mnt_id_req req = { + .size = MNT_ID_REQ_SIZE_VER1, + .mnt_id = mnt_id, + .param = last_mnt_id, + .mnt_ns_id = mnt_ns_id, + }; + + return syscall(__NR_listmount, &req, list, num, flags); +} + +int main(int argc, char *argv[]) +{ +#define LISTMNT_BUFFER 10 + __u64 list[LISTMNT_BUFFER], last_mnt_id = 0; + int ret, pidfd, fd_mntns; + struct mnt_ns_info info = {}; + + pidfd = sys_pidfd_open(getpid(), 0); + if (pidfd < 0) + die_errno("pidfd_open failed"); + + fd_mntns = ioctl(pidfd, PIDFD_GET_MNT_NAMESPACE, 0); + if (fd_mntns < 0) + die_errno("ioctl(PIDFD_GET_MNT_NAMESPACE) failed"); + + ret = ioctl(fd_mntns, NS_MNT_GET_INFO, &info); + if (ret < 0) + die_errno("ioctl(NS_GET_MNTNS_ID) failed"); + + printf("Listing %u mounts for mount namespace %" PRIu64 "\n", + info.nr_mounts, (uint64_t)info.mnt_ns_id); + for (;;) { + ssize_t nr_mounts; +next: + nr_mounts = sys_listmount(LSMT_ROOT, last_mnt_id, + info.mnt_ns_id, list, LISTMNT_BUFFER, + 0); + if (nr_mounts <= 0) { + int fd_mntns_next; + + printf("Finished listing %u mounts for mount namespace %" PRIu64 "\n\n", + info.nr_mounts, (uint64_t)info.mnt_ns_id); + fd_mntns_next = ioctl(fd_mntns, NS_MNT_GET_NEXT, &info); + if (fd_mntns_next < 0) { + if (errno == ENOENT) { + printf("Finished listing all mount namespaces\n"); + exit(0); + } + die_errno("ioctl(NS_MNT_GET_NEXT) failed"); + } + close(fd_mntns); + fd_mntns = fd_mntns_next; + last_mnt_id = 0; + printf("Listing %u mounts for mount namespace %" PRIu64 "\n", + info.nr_mounts, (uint64_t)info.mnt_ns_id); + goto next; + } + + for (size_t cur = 0; cur < nr_mounts; cur++) { + struct statmount *stmnt; + + last_mnt_id = list[cur]; + + stmnt = sys_statmount(last_mnt_id, info.mnt_ns_id, + STATMOUNT_SB_BASIC | + STATMOUNT_MNT_BASIC | + STATMOUNT_MNT_ROOT | + STATMOUNT_MNT_POINT | + STATMOUNT_MNT_NS_ID | + STATMOUNT_MNT_OPTS | + STATMOUNT_FS_TYPE, 0); + if (!stmnt) { + printf("Failed to statmount(%" PRIu64 ") in mount namespace(%" PRIu64 ")\n", + (uint64_t)last_mnt_id, (uint64_t)info.mnt_ns_id); + continue; + } + + printf("mnt_id:\t\t%" PRIu64 "\nmnt_parent_id:\t%" PRIu64 "\nfs_type:\t%s\nmnt_root:\t%s\nmnt_point:\t%s\nmnt_opts:\t%s\n\n", + (uint64_t)stmnt->mnt_id, + (uint64_t)stmnt->mnt_parent_id, + stmnt->str + stmnt->fs_type, + stmnt->str + stmnt->mnt_root, + stmnt->str + stmnt->mnt_point, + stmnt->str + stmnt->mnt_opts); + free(stmnt); + } + } + + exit(0); +} diff --git a/scripts/decode_stacktrace.sh b/scripts/decode_stacktrace.sh index 46fa18b80fc1..17abc4e7a985 100755 --- a/scripts/decode_stacktrace.sh +++ b/scripts/decode_stacktrace.sh @@ -286,6 +286,18 @@ handle_line() { last=$(( $last - 1 )) fi + # Extract info after the symbol if present. E.g.: + # func_name+0x54/0x80 (P) + # ^^^ + # The regex assumes only uppercase letters will be used. To be + # extended if needed. + local info_str="" + if [[ ${words[$last]} =~ \([A-Z]*\) ]]; then + info_str=${words[$last]} + unset words[$last] + last=$(( $last - 1 )) + fi + if [[ ${words[$last]} =~ \[([^]]+)\] ]]; then module=${words[$last]} # some traces format is "(%pS)", which like "(foo+0x0/0x1 [bar])" @@ -313,9 +325,9 @@ handle_line() { # Add up the line number to the symbol if [[ -z ${module} ]] then - echo "${words[@]}" "$symbol" + echo "${words[@]}" "$symbol ${info_str}" else - echo "${words[@]}" "$symbol $module" + echo "${words[@]}" "$symbol $module ${info_str}" fi } diff --git a/scripts/mksysmap b/scripts/mksysmap index c12723a04655..3accbdb269ac 100755 --- a/scripts/mksysmap +++ b/scripts/mksysmap @@ -26,7 +26,7 @@ # (do not forget a space before each pattern) # local symbols for ARM, MIPS, etc. -/ \\$/d +/ \$/d # local labels, .LBB, .Ltmpxxx, .L__unnamed_xx, .LASANPC, etc. / \.L/d @@ -39,7 +39,7 @@ / __pi_\.L/d # arm64 local symbols in non-VHE KVM namespace -/ __kvm_nvhe_\\$/d +/ __kvm_nvhe_\$/d / __kvm_nvhe_\.L/d # lld arm/aarch64/mips thunks diff --git a/scripts/mod/file2alias.c b/scripts/mod/file2alias.c index 5b5745f00eb3..19ec72a69e90 100644 --- a/scripts/mod/file2alias.c +++ b/scripts/mod/file2alias.c @@ -132,7 +132,8 @@ struct devtable { * based at address m. */ #define DEF_FIELD(m, devid, f) \ - typeof(((struct devid *)0)->f) f = TO_NATIVE(*(typeof(f) *)((m) + OFF_##devid##_##f)) + typeof(((struct devid *)0)->f) f = \ + get_unaligned_native((typeof(f) *)((m) + OFF_##devid##_##f)) /* Define a variable f that holds the address of field f of struct devid * based at address m. Due to the way typeof works, for a field of type @@ -600,7 +601,7 @@ static void do_pnp_card_entry(struct module *mod, void *symval) static void do_pcmcia_entry(struct module *mod, void *symval) { char alias[256] = {}; - unsigned int i; + DEF_FIELD(symval, pcmcia_device_id, match_flags); DEF_FIELD(symval, pcmcia_device_id, manf_id); DEF_FIELD(symval, pcmcia_device_id, card_id); @@ -609,10 +610,6 @@ static void do_pcmcia_entry(struct module *mod, void *symval) DEF_FIELD(symval, pcmcia_device_id, device_no); DEF_FIELD_ADDR(symval, pcmcia_device_id, prod_id_hash); - for (i=0; i<4; i++) { - (*prod_id_hash)[i] = TO_NATIVE((*prod_id_hash)[i]); - } - ADD(alias, "m", match_flags & PCMCIA_DEV_ID_MATCH_MANF_ID, manf_id); ADD(alias, "c", match_flags & PCMCIA_DEV_ID_MATCH_CARD_ID, @@ -623,10 +620,14 @@ static void do_pcmcia_entry(struct module *mod, void *symval) function); ADD(alias, "pfn", match_flags & PCMCIA_DEV_ID_MATCH_DEVICE_NO, device_no); - ADD(alias, "pa", match_flags & PCMCIA_DEV_ID_MATCH_PROD_ID1, (*prod_id_hash)[0]); - ADD(alias, "pb", match_flags & PCMCIA_DEV_ID_MATCH_PROD_ID2, (*prod_id_hash)[1]); - ADD(alias, "pc", match_flags & PCMCIA_DEV_ID_MATCH_PROD_ID3, (*prod_id_hash)[2]); - ADD(alias, "pd", match_flags & PCMCIA_DEV_ID_MATCH_PROD_ID4, (*prod_id_hash)[3]); + ADD(alias, "pa", match_flags & PCMCIA_DEV_ID_MATCH_PROD_ID1, + get_unaligned_native(*prod_id_hash + 0)); + ADD(alias, "pb", match_flags & PCMCIA_DEV_ID_MATCH_PROD_ID2, + get_unaligned_native(*prod_id_hash + 1)); + ADD(alias, "pc", match_flags & PCMCIA_DEV_ID_MATCH_PROD_ID3, + get_unaligned_native(*prod_id_hash + 2)); + ADD(alias, "pd", match_flags & PCMCIA_DEV_ID_MATCH_PROD_ID4, + get_unaligned_native(*prod_id_hash + 3)); module_alias_printf(mod, true, "pcmcia:%s", alias); } @@ -654,10 +655,9 @@ static void do_input(char *alias, { unsigned int i; - for (i = min / BITS_PER_LONG; i < max / BITS_PER_LONG + 1; i++) - arr[i] = TO_NATIVE(arr[i]); - for (i = min; i < max; i++) - if (arr[i / BITS_PER_LONG] & (1ULL << (i%BITS_PER_LONG))) + for (i = min; i <= max; i++) + if (get_unaligned_native(arr + i / BITS_PER_LONG) & + (1ULL << (i % BITS_PER_LONG))) sprintf(alias + strlen(alias), "%X,*", i); } @@ -812,15 +812,13 @@ static void do_virtio_entry(struct module *mod, void *symval) * Each byte of the guid will be represented by two hex characters * in the name. */ - static void do_vmbus_entry(struct module *mod, void *symval) { - int i; DEF_FIELD_ADDR(symval, hv_vmbus_device_id, guid); - char guid_name[(sizeof(*guid) + 1) * 2]; + char guid_name[sizeof(*guid) * 2 + 1]; - for (i = 0; i < (sizeof(*guid) * 2); i += 2) - sprintf(&guid_name[i], "%02x", TO_NATIVE((guid->b)[i/2])); + for (int i = 0; i < sizeof(*guid); i++) + sprintf(&guid_name[i * 2], "%02x", guid->b[i]); module_alias_printf(mod, false, "vmbus:%s", guid_name); } diff --git a/scripts/mod/modpost.c b/scripts/mod/modpost.c index 94ee49207a45..7ea59dc4926b 100644 --- a/scripts/mod/modpost.c +++ b/scripts/mod/modpost.c @@ -1138,9 +1138,9 @@ static Elf_Addr addend_386_rel(uint32_t *location, unsigned int r_type) { switch (r_type) { case R_386_32: - return TO_NATIVE(*location); + return get_unaligned_native(location); case R_386_PC32: - return TO_NATIVE(*location) + 4; + return get_unaligned_native(location) + 4; } return (Elf_Addr)(-1); @@ -1161,24 +1161,24 @@ static Elf_Addr addend_arm_rel(void *loc, Elf_Sym *sym, unsigned int r_type) switch (r_type) { case R_ARM_ABS32: case R_ARM_REL32: - inst = TO_NATIVE(*(uint32_t *)loc); + inst = get_unaligned_native((uint32_t *)loc); return inst + sym->st_value; case R_ARM_MOVW_ABS_NC: case R_ARM_MOVT_ABS: - inst = TO_NATIVE(*(uint32_t *)loc); + inst = get_unaligned_native((uint32_t *)loc); offset = sign_extend32(((inst & 0xf0000) >> 4) | (inst & 0xfff), 15); return offset + sym->st_value; case R_ARM_PC24: case R_ARM_CALL: case R_ARM_JUMP24: - inst = TO_NATIVE(*(uint32_t *)loc); + inst = get_unaligned_native((uint32_t *)loc); offset = sign_extend32((inst & 0x00ffffff) << 2, 25); return offset + sym->st_value + 8; case R_ARM_THM_MOVW_ABS_NC: case R_ARM_THM_MOVT_ABS: - upper = TO_NATIVE(*(uint16_t *)loc); - lower = TO_NATIVE(*((uint16_t *)loc + 1)); + upper = get_unaligned_native((uint16_t *)loc); + lower = get_unaligned_native((uint16_t *)loc + 1); offset = sign_extend32(((upper & 0x000f) << 12) | ((upper & 0x0400) << 1) | ((lower & 0x7000) >> 4) | @@ -1195,8 +1195,8 @@ static Elf_Addr addend_arm_rel(void *loc, Elf_Sym *sym, unsigned int r_type) * imm11 = lower[10:0] * imm32 = SignExtend(S:J2:J1:imm6:imm11:'0') */ - upper = TO_NATIVE(*(uint16_t *)loc); - lower = TO_NATIVE(*((uint16_t *)loc + 1)); + upper = get_unaligned_native((uint16_t *)loc); + lower = get_unaligned_native((uint16_t *)loc + 1); sign = (upper >> 10) & 1; j1 = (lower >> 13) & 1; @@ -1219,8 +1219,8 @@ static Elf_Addr addend_arm_rel(void *loc, Elf_Sym *sym, unsigned int r_type) * I2 = NOT(J2 XOR S) * imm32 = SignExtend(S:I1:I2:imm10:imm11:'0') */ - upper = TO_NATIVE(*(uint16_t *)loc); - lower = TO_NATIVE(*((uint16_t *)loc + 1)); + upper = get_unaligned_native((uint16_t *)loc); + lower = get_unaligned_native((uint16_t *)loc + 1); sign = (upper >> 10) & 1; j1 = (lower >> 13) & 1; @@ -1241,7 +1241,7 @@ static Elf_Addr addend_mips_rel(uint32_t *location, unsigned int r_type) { uint32_t inst; - inst = TO_NATIVE(*location); + inst = get_unaligned_native(location); switch (r_type) { case R_MIPS_LO16: return inst & 0xffff; diff --git a/scripts/mod/modpost.h b/scripts/mod/modpost.h index 8b72c227ebf4..ffd0a52a606e 100644 --- a/scripts/mod/modpost.h +++ b/scripts/mod/modpost.h @@ -65,6 +65,20 @@ #define TO_NATIVE(x) \ (target_is_big_endian == host_is_big_endian ? x : bswap(x)) +#define __get_unaligned_t(type, ptr) ({ \ + const struct { type x; } __attribute__((__packed__)) *__pptr = \ + (typeof(__pptr))(ptr); \ + __pptr->x; \ +}) + +#define get_unaligned(ptr) __get_unaligned_t(typeof(*(ptr)), (ptr)) + +#define get_unaligned_native(ptr) \ +({ \ + typeof(*(ptr)) _val = get_unaligned(ptr); \ + TO_NATIVE(_val); \ +}) + #define ARRAY_SIZE(arr) (sizeof(arr) / sizeof((arr)[0])) #define strstarts(str, prefix) (strncmp(str, prefix, strlen(prefix)) == 0) diff --git a/scripts/package/PKGBUILD b/scripts/package/PKGBUILD index f83493838cf9..dca706617adc 100644 --- a/scripts/package/PKGBUILD +++ b/scripts/package/PKGBUILD @@ -103,7 +103,7 @@ _package-headers() { _package-api-headers() { pkgdesc="Kernel headers sanitized for use in userspace" - provides=(linux-api-headers) + provides=(linux-api-headers="${pkgver}") conflicts=(linux-api-headers) _prologue diff --git a/scripts/sorttable.h b/scripts/sorttable.h index 7bd0184380d3..a7c5445baf00 100644 --- a/scripts/sorttable.h +++ b/scripts/sorttable.h @@ -110,7 +110,7 @@ static inline unsigned long orc_ip(const int *ip) static int orc_sort_cmp(const void *_a, const void *_b) { - struct orc_entry *orc_a; + struct orc_entry *orc_a, *orc_b; const int *a = g_orc_ip_table + *(int *)_a; const int *b = g_orc_ip_table + *(int *)_b; unsigned long a_val = orc_ip(a); @@ -128,6 +128,9 @@ static int orc_sort_cmp(const void *_a, const void *_b) * whitelisted .o files which didn't get objtool generation. */ orc_a = g_orc_table + (a - g_orc_ip_table); + orc_b = g_orc_table + (b - g_orc_ip_table); + if (orc_a->type == ORC_TYPE_UNDEFINED && orc_b->type == ORC_TYPE_UNDEFINED) + return 0; return orc_a->type == ORC_TYPE_UNDEFINED ? -1 : 1; } diff --git a/security/apparmor/apparmorfs.c b/security/apparmor/apparmorfs.c index 2c0185ebc900..c07d150685d7 100644 --- a/security/apparmor/apparmorfs.c +++ b/security/apparmor/apparmorfs.c @@ -2612,7 +2612,7 @@ static int policy_readlink(struct dentry *dentry, char __user *buffer, res = snprintf(name, sizeof(name), "%s:[%lu]", AAFS_NAME, d_inode(dentry)->i_ino); if (res > 0 && res < sizeof(name)) - res = readlink_copy(buffer, buflen, name); + res = readlink_copy(buffer, buflen, name, strlen(name)); else res = -ENOENT; diff --git a/security/selinux/avc.c b/security/selinux/avc.c index cc0b0af20296..1f2680bcc43a 100644 --- a/security/selinux/avc.c +++ b/security/selinux/avc.c @@ -174,13 +174,15 @@ int avc_get_hash_stats(char *page) * using a linked list for extended_perms_decision lookup because the list is * always small. i.e. less than 5, typically 1 */ -static struct extended_perms_decision *avc_xperms_decision_lookup(u8 driver, - struct avc_xperms_node *xp_node) +static struct extended_perms_decision * +avc_xperms_decision_lookup(u8 driver, u8 base_perm, + struct avc_xperms_node *xp_node) { struct avc_xperms_decision_node *xpd_node; list_for_each_entry(xpd_node, &xp_node->xpd_head, xpd_list) { - if (xpd_node->xpd.driver == driver) + if (xpd_node->xpd.driver == driver && + xpd_node->xpd.base_perm == base_perm) return &xpd_node->xpd; } return NULL; @@ -205,11 +207,12 @@ avc_xperms_has_perm(struct extended_perms_decision *xpd, } static void avc_xperms_allow_perm(struct avc_xperms_node *xp_node, - u8 driver, u8 perm) + u8 driver, u8 base_perm, u8 perm) { struct extended_perms_decision *xpd; security_xperm_set(xp_node->xp.drivers.p, driver); - xpd = avc_xperms_decision_lookup(driver, xp_node); + xp_node->xp.base_perms |= base_perm; + xpd = avc_xperms_decision_lookup(driver, base_perm, xp_node); if (xpd && xpd->allowed) security_xperm_set(xpd->allowed->p, perm); } @@ -245,6 +248,7 @@ static void avc_xperms_free(struct avc_xperms_node *xp_node) static void avc_copy_xperms_decision(struct extended_perms_decision *dest, struct extended_perms_decision *src) { + dest->base_perm = src->base_perm; dest->driver = src->driver; dest->used = src->used; if (dest->used & XPERMS_ALLOWED) @@ -272,6 +276,7 @@ static inline void avc_quick_copy_xperms_decision(u8 perm, */ u8 i = perm >> 5; + dest->base_perm = src->base_perm; dest->used = src->used; if (dest->used & XPERMS_ALLOWED) dest->allowed->p[i] = src->allowed->p[i]; @@ -357,6 +362,7 @@ static int avc_xperms_populate(struct avc_node *node, memcpy(dest->xp.drivers.p, src->xp.drivers.p, sizeof(dest->xp.drivers.p)); dest->xp.len = src->xp.len; + dest->xp.base_perms = src->xp.base_perms; /* for each source xpd allocate a destination xpd and copy */ list_for_each_entry(src_xpd, &src->xpd_head, xpd_list) { @@ -807,6 +813,7 @@ out: * @event : Updating event * @perms : Permission mask bits * @driver: xperm driver information + * @base_perm: the base permission associated with the extended permission * @xperm: xperm permissions * @ssid: AVC entry source sid * @tsid: AVC entry target sid @@ -820,10 +827,9 @@ out: * otherwise, this function updates the AVC entry. The original AVC-entry object * will release later by RCU. */ -static int avc_update_node(u32 event, u32 perms, u8 driver, u8 xperm, u32 ssid, - u32 tsid, u16 tclass, u32 seqno, - struct extended_perms_decision *xpd, - u32 flags) +static int avc_update_node(u32 event, u32 perms, u8 driver, u8 base_perm, + u8 xperm, u32 ssid, u32 tsid, u16 tclass, u32 seqno, + struct extended_perms_decision *xpd, u32 flags) { u32 hvalue; int rc = 0; @@ -880,7 +886,7 @@ static int avc_update_node(u32 event, u32 perms, u8 driver, u8 xperm, u32 ssid, case AVC_CALLBACK_GRANT: node->ae.avd.allowed |= perms; if (node->ae.xp_node && (flags & AVC_EXTENDED_PERMS)) - avc_xperms_allow_perm(node->ae.xp_node, driver, xperm); + avc_xperms_allow_perm(node->ae.xp_node, driver, base_perm, xperm); break; case AVC_CALLBACK_TRY_REVOKE: case AVC_CALLBACK_REVOKE: @@ -987,10 +993,9 @@ static noinline void avc_compute_av(u32 ssid, u32 tsid, u16 tclass, avc_insert(ssid, tsid, tclass, avd, xp_node); } -static noinline int avc_denied(u32 ssid, u32 tsid, - u16 tclass, u32 requested, - u8 driver, u8 xperm, unsigned int flags, - struct av_decision *avd) +static noinline int avc_denied(u32 ssid, u32 tsid, u16 tclass, u32 requested, + u8 driver, u8 base_perm, u8 xperm, + unsigned int flags, struct av_decision *avd) { if (flags & AVC_STRICT) return -EACCES; @@ -999,7 +1004,7 @@ static noinline int avc_denied(u32 ssid, u32 tsid, !(avd->flags & AVD_FLAGS_PERMISSIVE)) return -EACCES; - avc_update_node(AVC_CALLBACK_GRANT, requested, driver, + avc_update_node(AVC_CALLBACK_GRANT, requested, driver, base_perm, xperm, ssid, tsid, tclass, avd->seqno, NULL, flags); return 0; } @@ -1012,7 +1017,8 @@ static noinline int avc_denied(u32 ssid, u32 tsid, * driver field is used to specify which set contains the permission. */ int avc_has_extended_perms(u32 ssid, u32 tsid, u16 tclass, u32 requested, - u8 driver, u8 xperm, struct common_audit_data *ad) + u8 driver, u8 base_perm, u8 xperm, + struct common_audit_data *ad) { struct avc_node *node; struct av_decision avd; @@ -1047,22 +1053,23 @@ int avc_has_extended_perms(u32 ssid, u32 tsid, u16 tclass, u32 requested, local_xpd.auditallow = &auditallow; local_xpd.dontaudit = &dontaudit; - xpd = avc_xperms_decision_lookup(driver, xp_node); + xpd = avc_xperms_decision_lookup(driver, base_perm, xp_node); if (unlikely(!xpd)) { /* * Compute the extended_perms_decision only if the driver - * is flagged + * is flagged and the base permission is known. */ - if (!security_xperm_test(xp_node->xp.drivers.p, driver)) { + if (!security_xperm_test(xp_node->xp.drivers.p, driver) || + !(xp_node->xp.base_perms & base_perm)) { avd.allowed &= ~requested; goto decision; } rcu_read_unlock(); - security_compute_xperms_decision(ssid, tsid, tclass, - driver, &local_xpd); + security_compute_xperms_decision(ssid, tsid, tclass, driver, + base_perm, &local_xpd); rcu_read_lock(); - avc_update_node(AVC_CALLBACK_ADD_XPERMS, requested, - driver, xperm, ssid, tsid, tclass, avd.seqno, + avc_update_node(AVC_CALLBACK_ADD_XPERMS, requested, driver, + base_perm, xperm, ssid, tsid, tclass, avd.seqno, &local_xpd, 0); } else { avc_quick_copy_xperms_decision(xperm, &local_xpd, xpd); @@ -1075,8 +1082,8 @@ int avc_has_extended_perms(u32 ssid, u32 tsid, u16 tclass, u32 requested, decision: denied = requested & ~(avd.allowed); if (unlikely(denied)) - rc = avc_denied(ssid, tsid, tclass, requested, - driver, xperm, AVC_EXTENDED_PERMS, &avd); + rc = avc_denied(ssid, tsid, tclass, requested, driver, + base_perm, xperm, AVC_EXTENDED_PERMS, &avd); rcu_read_unlock(); @@ -1110,7 +1117,7 @@ static noinline int avc_perm_nonode(u32 ssid, u32 tsid, u16 tclass, avc_compute_av(ssid, tsid, tclass, avd, &xp_node); denied = requested & ~(avd->allowed); if (unlikely(denied)) - return avc_denied(ssid, tsid, tclass, requested, 0, 0, + return avc_denied(ssid, tsid, tclass, requested, 0, 0, 0, flags, avd); return 0; } @@ -1158,7 +1165,7 @@ inline int avc_has_perm_noaudit(u32 ssid, u32 tsid, rcu_read_unlock(); if (unlikely(denied)) - return avc_denied(ssid, tsid, tclass, requested, 0, 0, + return avc_denied(ssid, tsid, tclass, requested, 0, 0, 0, flags, avd); return 0; } diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c index 366c87a40bd1..171dd7fceac5 100644 --- a/security/selinux/hooks.c +++ b/security/selinux/hooks.c @@ -3688,8 +3688,8 @@ static int ioctl_has_perm(const struct cred *cred, struct file *file, return 0; isec = inode_security(inode); - rc = avc_has_extended_perms(ssid, isec->sid, isec->sclass, - requested, driver, xperm, &ad); + rc = avc_has_extended_perms(ssid, isec->sid, isec->sclass, requested, + driver, AVC_EXT_IOCTL, xperm, &ad); out: return rc; } @@ -5952,7 +5952,7 @@ static int nlmsg_sock_has_extended_perms(struct sock *sk, u32 perms, u16 nlmsg_t xperm = nlmsg_type & 0xff; return avc_has_extended_perms(current_sid(), sksec->sid, sksec->sclass, - perms, driver, xperm, &ad); + perms, driver, AVC_EXT_NLMSG, xperm, &ad); } static int selinux_netlink_send(struct sock *sk, struct sk_buff *skb) diff --git a/security/selinux/include/avc.h b/security/selinux/include/avc.h index 96a614d47df8..281f40103663 100644 --- a/security/selinux/include/avc.h +++ b/security/selinux/include/avc.h @@ -136,8 +136,11 @@ int avc_has_perm_noaudit(u32 ssid, u32 tsid, u16 tclass, u32 requested, int avc_has_perm(u32 ssid, u32 tsid, u16 tclass, u32 requested, struct common_audit_data *auditdata); +#define AVC_EXT_IOCTL (1 << 0) /* Cache entry for an ioctl extended permission */ +#define AVC_EXT_NLMSG (1 << 1) /* Cache entry for an nlmsg extended permission */ int avc_has_extended_perms(u32 ssid, u32 tsid, u16 tclass, u32 requested, - u8 driver, u8 perm, struct common_audit_data *ad); + u8 driver, u8 base_perm, u8 perm, + struct common_audit_data *ad); u32 avc_policy_seqno(void); diff --git a/security/selinux/include/security.h b/security/selinux/include/security.h index c7f2731abd03..700bd6c8bb38 100644 --- a/security/selinux/include/security.h +++ b/security/selinux/include/security.h @@ -239,6 +239,7 @@ struct extended_perms_data { struct extended_perms_decision { u8 used; u8 driver; + u8 base_perm; struct extended_perms_data *allowed; struct extended_perms_data *auditallow; struct extended_perms_data *dontaudit; @@ -246,6 +247,7 @@ struct extended_perms_decision { struct extended_perms { u16 len; /* length associated decision chain */ + u8 base_perms; /* which base permissions are covered */ struct extended_perms_data drivers; /* flag drivers that are used */ }; @@ -257,6 +259,7 @@ void security_compute_av(u32 ssid, u32 tsid, u16 tclass, struct extended_perms *xperms); void security_compute_xperms_decision(u32 ssid, u32 tsid, u16 tclass, u8 driver, + u8 base_perm, struct extended_perms_decision *xpermd); void security_compute_av_user(u32 ssid, u32 tsid, u16 tclass, diff --git a/security/selinux/ss/services.c b/security/selinux/ss/services.c index 3d5c563cfc4c..d9f58b5d0f49 100644 --- a/security/selinux/ss/services.c +++ b/security/selinux/ss/services.c @@ -582,7 +582,7 @@ static void type_attribute_bounds_av(struct policydb *policydb, } /* - * Flag which drivers have permissions. + * Flag which drivers have permissions and which base permissions are covered. */ void services_compute_xperms_drivers( struct extended_perms *xperms, @@ -592,12 +592,19 @@ void services_compute_xperms_drivers( switch (node->datum.u.xperms->specified) { case AVTAB_XPERMS_IOCTLDRIVER: + xperms->base_perms |= AVC_EXT_IOCTL; /* if one or more driver has all permissions allowed */ for (i = 0; i < ARRAY_SIZE(xperms->drivers.p); i++) xperms->drivers.p[i] |= node->datum.u.xperms->perms.p[i]; break; case AVTAB_XPERMS_IOCTLFUNCTION: + xperms->base_perms |= AVC_EXT_IOCTL; + /* if allowing permissions within a driver */ + security_xperm_set(xperms->drivers.p, + node->datum.u.xperms->driver); + break; case AVTAB_XPERMS_NLMSG: + xperms->base_perms |= AVC_EXT_NLMSG; /* if allowing permissions within a driver */ security_xperm_set(xperms->drivers.p, node->datum.u.xperms->driver); @@ -631,8 +638,7 @@ static void context_struct_compute_av(struct policydb *policydb, avd->auditallow = 0; avd->auditdeny = 0xffffffff; if (xperms) { - memset(&xperms->drivers, 0, sizeof(xperms->drivers)); - xperms->len = 0; + memset(xperms, 0, sizeof(*xperms)); } if (unlikely(!tclass || tclass > policydb->p_classes.nprim)) { @@ -969,13 +975,19 @@ void services_compute_xperms_decision(struct extended_perms_decision *xpermd, { switch (node->datum.u.xperms->specified) { case AVTAB_XPERMS_IOCTLFUNCTION: - case AVTAB_XPERMS_NLMSG: - if (xpermd->driver != node->datum.u.xperms->driver) + if (xpermd->base_perm != AVC_EXT_IOCTL || + xpermd->driver != node->datum.u.xperms->driver) return; break; case AVTAB_XPERMS_IOCTLDRIVER: - if (!security_xperm_test(node->datum.u.xperms->perms.p, - xpermd->driver)) + if (xpermd->base_perm != AVC_EXT_IOCTL || + !security_xperm_test(node->datum.u.xperms->perms.p, + xpermd->driver)) + return; + break; + case AVTAB_XPERMS_NLMSG: + if (xpermd->base_perm != AVC_EXT_NLMSG || + xpermd->driver != node->datum.u.xperms->driver) return; break; default: @@ -1010,6 +1022,7 @@ void security_compute_xperms_decision(u32 ssid, u32 tsid, u16 orig_tclass, u8 driver, + u8 base_perm, struct extended_perms_decision *xpermd) { struct selinux_policy *policy; @@ -1023,6 +1036,7 @@ void security_compute_xperms_decision(u32 ssid, struct ebitmap_node *snode, *tnode; unsigned int i, j; + xpermd->base_perm = base_perm; xpermd->driver = driver; xpermd->used = 0; memset(xpermd->allowed->p, 0, sizeof(xpermd->allowed->p)); diff --git a/sound/core/compress_offload.c b/sound/core/compress_offload.c index 86ed2fbee0c8..840bb9cfe789 100644 --- a/sound/core/compress_offload.c +++ b/sound/core/compress_offload.c @@ -1025,7 +1025,7 @@ static u64 snd_compr_seqno_next(struct snd_compr_stream *stream) static int snd_compr_task_new(struct snd_compr_stream *stream, struct snd_compr_task *utask) { struct snd_compr_task_runtime *task; - int retval; + int retval, fd_i, fd_o; if (stream->runtime->total_tasks >= stream->runtime->fragments) return -EBUSY; @@ -1039,19 +1039,27 @@ static int snd_compr_task_new(struct snd_compr_stream *stream, struct snd_compr_ retval = stream->ops->task_create(stream, task); if (retval < 0) goto cleanup; - utask->input_fd = dma_buf_fd(task->input, O_WRONLY|O_CLOEXEC); - if (utask->input_fd < 0) { - retval = utask->input_fd; + /* similar functionality as in dma_buf_fd(), but ensure that both + file descriptors are allocated before fd_install() */ + if (!task->input || !task->input->file || !task->output || !task->output->file) { + retval = -EINVAL; goto cleanup; } - utask->output_fd = dma_buf_fd(task->output, O_RDONLY|O_CLOEXEC); - if (utask->output_fd < 0) { - retval = utask->output_fd; + fd_i = get_unused_fd_flags(O_WRONLY|O_CLOEXEC); + if (fd_i < 0) + goto cleanup; + fd_o = get_unused_fd_flags(O_RDONLY|O_CLOEXEC); + if (fd_o < 0) { + put_unused_fd(fd_i); goto cleanup; } /* keep dmabuf reference until freed with task free ioctl */ - dma_buf_get(utask->input_fd); - dma_buf_get(utask->output_fd); + get_dma_buf(task->input); + get_dma_buf(task->output); + fd_install(fd_i, task->input->file); + fd_install(fd_o, task->output->file); + utask->input_fd = fd_i; + utask->output_fd = fd_o; list_add_tail(&task->list, &stream->runtime->tasks); stream->runtime->total_tasks++; return 0; @@ -1069,7 +1077,7 @@ static int snd_compr_task_create(struct snd_compr_stream *stream, unsigned long return -EPERM; task = memdup_user((void __user *)arg, sizeof(*task)); if (IS_ERR(task)) - return PTR_ERR(no_free_ptr(task)); + return PTR_ERR(task); retval = snd_compr_task_new(stream, task); if (retval >= 0) if (copy_to_user((void __user *)arg, task, sizeof(*task))) @@ -1130,7 +1138,7 @@ static int snd_compr_task_start_ioctl(struct snd_compr_stream *stream, unsigned return -EPERM; task = memdup_user((void __user *)arg, sizeof(*task)); if (IS_ERR(task)) - return PTR_ERR(no_free_ptr(task)); + return PTR_ERR(task); retval = snd_compr_task_start(stream, task); if (retval >= 0) if (copy_to_user((void __user *)arg, task, sizeof(*task))) @@ -1174,18 +1182,18 @@ typedef void (*snd_compr_seq_func_t)(struct snd_compr_stream *stream, static int snd_compr_task_seq(struct snd_compr_stream *stream, unsigned long arg, snd_compr_seq_func_t fcn) { - struct snd_compr_task_runtime *task; + struct snd_compr_task_runtime *task, *temp; __u64 seqno; int retval; if (stream->runtime->state != SNDRV_PCM_STATE_SETUP) return -EPERM; - retval = get_user(seqno, (__u64 __user *)arg); - if (retval < 0) - return retval; + retval = copy_from_user(&seqno, (__u64 __user *)arg, sizeof(seqno)); + if (retval) + return -EFAULT; retval = 0; if (seqno == 0) { - list_for_each_entry_reverse(task, &stream->runtime->tasks, list) + list_for_each_entry_safe_reverse(task, temp, &stream->runtime->tasks, list) fcn(stream, task); } else { task = snd_compr_find_task(stream, seqno); @@ -1221,7 +1229,7 @@ static int snd_compr_task_status_ioctl(struct snd_compr_stream *stream, unsigned return -EPERM; status = memdup_user((void __user *)arg, sizeof(*status)); if (IS_ERR(status)) - return PTR_ERR(no_free_ptr(status)); + return PTR_ERR(status); retval = snd_compr_task_status(stream, status); if (retval >= 0) if (copy_to_user((void __user *)arg, status, sizeof(*status))) @@ -1247,6 +1255,7 @@ void snd_compr_task_finished(struct snd_compr_stream *stream, } EXPORT_SYMBOL_GPL(snd_compr_task_finished); +MODULE_IMPORT_NS("DMA_BUF"); #endif /* CONFIG_SND_COMPRESS_ACCEL */ static long snd_compr_ioctl(struct file *f, unsigned int cmd, unsigned long arg) diff --git a/sound/core/memalloc.c b/sound/core/memalloc.c index 13b71069ae18..b3853583d2ae 100644 --- a/sound/core/memalloc.c +++ b/sound/core/memalloc.c @@ -505,7 +505,7 @@ static void *snd_dma_wc_alloc(struct snd_dma_buffer *dmab, size_t size) if (!p) return NULL; dmab->addr = dma_map_single(dmab->dev.dev, p, size, DMA_BIDIRECTIONAL); - if (dmab->addr == DMA_MAPPING_ERROR) { + if (dma_mapping_error(dmab->dev.dev, dmab->addr)) { do_free_pages(dmab->area, size, true); return NULL; } diff --git a/sound/core/seq/oss/seq_oss_synth.c b/sound/core/seq/oss/seq_oss_synth.c index e3394919daa0..51ee4c00a843 100644 --- a/sound/core/seq/oss/seq_oss_synth.c +++ b/sound/core/seq/oss/seq_oss_synth.c @@ -66,6 +66,7 @@ static struct seq_oss_synth midi_synth_dev = { }; static DEFINE_SPINLOCK(register_lock); +static DEFINE_MUTEX(sysex_mutex); /* * prototypes @@ -497,6 +498,7 @@ snd_seq_oss_synth_sysex(struct seq_oss_devinfo *dp, int dev, unsigned char *buf, if (!info) return -ENXIO; + guard(mutex)(&sysex_mutex); sysex = info->sysex; if (sysex == NULL) { sysex = kzalloc(sizeof(*sysex), GFP_KERNEL); diff --git a/sound/core/seq/seq_clientmgr.c b/sound/core/seq/seq_clientmgr.c index 3930e2f9082f..77b6ac9b5c11 100644 --- a/sound/core/seq/seq_clientmgr.c +++ b/sound/core/seq/seq_clientmgr.c @@ -1275,10 +1275,16 @@ static int snd_seq_ioctl_set_client_info(struct snd_seq_client *client, if (client->type != client_info->type) return -EINVAL; - /* check validity of midi_version field */ - if (client->user_pversion >= SNDRV_PROTOCOL_VERSION(1, 0, 3) && - client_info->midi_version > SNDRV_SEQ_CLIENT_UMP_MIDI_2_0) - return -EINVAL; + if (client->user_pversion >= SNDRV_PROTOCOL_VERSION(1, 0, 3)) { + /* check validity of midi_version field */ + if (client_info->midi_version > SNDRV_SEQ_CLIENT_UMP_MIDI_2_0) + return -EINVAL; + + /* check if UMP is supported in kernel */ + if (!IS_ENABLED(CONFIG_SND_SEQ_UMP) && + client_info->midi_version > 0) + return -EINVAL; + } /* fill the info fields */ if (client_info->name[0]) diff --git a/sound/core/ump.c b/sound/core/ump.c index fe4d39ae1159..9198bff4768c 100644 --- a/sound/core/ump.c +++ b/sound/core/ump.c @@ -1244,7 +1244,7 @@ static int fill_legacy_mapping(struct snd_ump_endpoint *ump) num = 0; for (i = 0; i < SNDRV_UMP_MAX_GROUPS; i++) - if ((group_maps & (1U << i)) && ump->groups[i].valid) + if (group_maps & (1U << i)) ump->legacy_mapping[num++] = i; return num; diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c index 61ba5dc35b8b..ad66378d7321 100644 --- a/sound/pci/hda/patch_realtek.c +++ b/sound/pci/hda/patch_realtek.c @@ -10641,6 +10641,8 @@ static const struct hda_quirk alc269_fixup_tbl[] = { SND_PCI_QUIRK(0x1043, 0x1e1f, "ASUS Vivobook 15 X1504VAP", ALC2XX_FIXUP_HEADSET_MIC), SND_PCI_QUIRK(0x1043, 0x1e51, "ASUS Zephyrus M15", ALC294_FIXUP_ASUS_GU502_PINS), SND_PCI_QUIRK(0x1043, 0x1e5e, "ASUS ROG Strix G513", ALC294_FIXUP_ASUS_G513_PINS), + SND_PCI_QUIRK(0x1043, 0x1e63, "ASUS H7606W", ALC285_FIXUP_ASUS_GU605_SPI_SPEAKER2_TO_DAC1), + SND_PCI_QUIRK(0x1043, 0x1e83, "ASUS GA605W", ALC285_FIXUP_ASUS_GU605_SPI_SPEAKER2_TO_DAC1), SND_PCI_QUIRK(0x1043, 0x1e8e, "ASUS Zephyrus G15", ALC289_FIXUP_ASUS_GA401), SND_PCI_QUIRK(0x1043, 0x1eb3, "ASUS Ally RCLA72", ALC287_FIXUP_TAS2781_I2C), SND_PCI_QUIRK(0x1043, 0x1ed3, "ASUS HN7306W", ALC287_FIXUP_CS35L41_I2C_2), @@ -10930,8 +10932,8 @@ static const struct hda_quirk alc269_fixup_tbl[] = { SND_PCI_QUIRK(0x17aa, 0x38e0, "Yoga Y990 Intel VECO Dual", ALC287_FIXUP_TAS2781_I2C), SND_PCI_QUIRK(0x17aa, 0x38f8, "Yoga Book 9i", ALC287_FIXUP_TAS2781_I2C), SND_PCI_QUIRK(0x17aa, 0x38df, "Y990 YG DUAL", ALC287_FIXUP_TAS2781_I2C), - SND_PCI_QUIRK(0x17aa, 0x38f9, "Thinkbook 16P Gen5", ALC287_FIXUP_CS35L41_I2C_2), - SND_PCI_QUIRK(0x17aa, 0x38fa, "Thinkbook 16P Gen5", ALC287_FIXUP_CS35L41_I2C_2), + SND_PCI_QUIRK(0x17aa, 0x38f9, "Thinkbook 16P Gen5", ALC287_FIXUP_MG_RTKC_CSAMP_CS35L41_I2C_THINKPAD), + SND_PCI_QUIRK(0x17aa, 0x38fa, "Thinkbook 16P Gen5", ALC287_FIXUP_MG_RTKC_CSAMP_CS35L41_I2C_THINKPAD), SND_PCI_QUIRK(0x17aa, 0x38fd, "ThinkBook plus Gen5 Hybrid", ALC287_FIXUP_TAS2781_I2C), SND_PCI_QUIRK(0x17aa, 0x3902, "Lenovo E50-80", ALC269_FIXUP_DMIC_THINKPAD_ACPI), SND_PCI_QUIRK(0x17aa, 0x3913, "Lenovo 145", ALC236_FIXUP_LENOVO_INV_DMIC), @@ -10995,6 +10997,7 @@ static const struct hda_quirk alc269_fixup_tbl[] = { SND_PCI_QUIRK(0x1d72, 0x1901, "RedmiBook 14", ALC256_FIXUP_ASUS_HEADSET_MIC), SND_PCI_QUIRK(0x1d72, 0x1945, "Redmi G", ALC256_FIXUP_ASUS_HEADSET_MIC), SND_PCI_QUIRK(0x1d72, 0x1947, "RedmiBook Air", ALC255_FIXUP_XIAOMI_HEADSET_MIC), + SND_PCI_QUIRK(0x1f66, 0x0105, "Ayaneo Portable Game Player", ALC287_FIXUP_CS35L41_I2C_2), SND_PCI_QUIRK(0x2782, 0x0214, "VAIO VJFE-CL", ALC269_FIXUP_LIMIT_INT_MIC_BOOST), SND_PCI_QUIRK(0x2782, 0x0228, "Infinix ZERO BOOK 13", ALC269VB_FIXUP_INFINIX_ZERO_BOOK_13), SND_PCI_QUIRK(0x2782, 0x0232, "CHUWI CoreBook XPro", ALC269VB_FIXUP_CHUWI_COREBOOK_XPRO), @@ -11009,6 +11012,7 @@ static const struct hda_quirk alc269_fixup_tbl[] = { SND_PCI_QUIRK(0xf111, 0x0001, "Framework Laptop", ALC295_FIXUP_FRAMEWORK_LAPTOP_MIC_NO_PRESENCE), SND_PCI_QUIRK(0xf111, 0x0006, "Framework Laptop", ALC295_FIXUP_FRAMEWORK_LAPTOP_MIC_NO_PRESENCE), SND_PCI_QUIRK(0xf111, 0x0009, "Framework Laptop", ALC295_FIXUP_FRAMEWORK_LAPTOP_MIC_NO_PRESENCE), + SND_PCI_QUIRK(0xf111, 0x000c, "Framework Laptop", ALC295_FIXUP_FRAMEWORK_LAPTOP_MIC_NO_PRESENCE), #if 0 /* Below is a quirk table taken from the old code. diff --git a/sound/pci/hda/tas2781_hda_i2c.c b/sound/pci/hda/tas2781_hda_i2c.c index 0af015806aba..0e42b87dadb8 100644 --- a/sound/pci/hda/tas2781_hda_i2c.c +++ b/sound/pci/hda/tas2781_hda_i2c.c @@ -142,6 +142,9 @@ static int tas2781_read_acpi(struct tasdevice_priv *p, const char *hid) } sub = acpi_get_subsystem_id(ACPI_HANDLE(physdev)); if (IS_ERR(sub)) { + /* No subsys id in older tas2563 projects. */ + if (!strncmp(hid, "INT8866", sizeof("INT8866"))) + goto end_2563; dev_err(p->dev, "Failed to get SUBSYS ID.\n"); ret = PTR_ERR(sub); goto err; @@ -164,6 +167,7 @@ static int tas2781_read_acpi(struct tasdevice_priv *p, const char *hid) p->speaker_id = NULL; } +end_2563: acpi_dev_free_resource_list(&resources); strscpy(p->dev_name, hid, sizeof(p->dev_name)); put_device(physdev); diff --git a/sound/sh/sh_dac_audio.c b/sound/sh/sh_dac_audio.c index a4d07438ad64..3f5422145c5e 100644 --- a/sound/sh/sh_dac_audio.c +++ b/sound/sh/sh_dac_audio.c @@ -163,7 +163,7 @@ static int snd_sh_dac_pcm_copy(struct snd_pcm_substream *substream, /* channel is not used (interleaved data) */ struct snd_sh_dac *chip = snd_pcm_substream_chip(substream); - if (copy_from_iter(chip->data_buffer + pos, src, count) != count) + if (copy_from_iter(chip->data_buffer + pos, count, src) != count) return -EFAULT; chip->buffer_end = chip->data_buffer + pos + count; diff --git a/sound/soc/amd/ps/pci-ps.c b/sound/soc/amd/ps/pci-ps.c index 823a69bf778b..4575326d0635 100644 --- a/sound/soc/amd/ps/pci-ps.c +++ b/sound/soc/amd/ps/pci-ps.c @@ -375,11 +375,18 @@ static int get_acp63_device_config(struct pci_dev *pci, struct acp63_dev_data *a { struct acpi_device *pdm_dev; const union acpi_object *obj; + acpi_handle handle; + acpi_integer dmic_status; u32 config; bool is_dmic_dev = false; bool is_sdw_dev = false; + bool wov_en, dmic_en; int ret; + /* IF WOV entry not found, enable dmic based on acp-audio-device-type entry*/ + wov_en = true; + dmic_en = false; + config = readl(acp_data->acp63_base + ACP_PIN_CONFIG); switch (config) { case ACP_CONFIG_4: @@ -412,10 +419,18 @@ static int get_acp63_device_config(struct pci_dev *pci, struct acp63_dev_data *a if (!acpi_dev_get_property(pdm_dev, "acp-audio-device-type", ACPI_TYPE_INTEGER, &obj) && obj->integer.value == ACP_DMIC_DEV) - is_dmic_dev = true; + dmic_en = true; } + + handle = ACPI_HANDLE(&pci->dev); + ret = acpi_evaluate_integer(handle, "_WOV", NULL, &dmic_status); + if (!ACPI_FAILURE(ret)) + wov_en = dmic_status; } + if (dmic_en && wov_en) + is_dmic_dev = true; + if (acp_data->is_sdw_config) { ret = acp_scan_sdw_devices(&pci->dev, ACP63_SDW_ADDR); if (!ret && acp_data->info.link_mask) diff --git a/sound/soc/codecs/Kconfig b/sound/soc/codecs/Kconfig index 0f2df7c91e18..0b9e87dc2b6c 100644 --- a/sound/soc/codecs/Kconfig +++ b/sound/soc/codecs/Kconfig @@ -2451,6 +2451,7 @@ config SND_SOC_WM8993 config SND_SOC_WM8994 tristate + depends on MFD_WM8994 config SND_SOC_WM8995 tristate diff --git a/sound/soc/codecs/cs42l43.c b/sound/soc/codecs/cs42l43.c index 4236f78beec0..83c21c17fb80 100644 --- a/sound/soc/codecs/cs42l43.c +++ b/sound/soc/codecs/cs42l43.c @@ -2404,6 +2404,7 @@ static int cs42l43_codec_runtime_resume(struct device *dev) static const struct dev_pm_ops cs42l43_codec_pm_ops = { RUNTIME_PM_OPS(NULL, cs42l43_codec_runtime_resume, NULL) + SET_SYSTEM_SLEEP_PM_OPS(pm_runtime_force_suspend, pm_runtime_force_resume) }; static const struct platform_device_id cs42l43_codec_id_table[] = { diff --git a/sound/soc/codecs/es8316.c b/sound/soc/codecs/es8316.c index 61729e5b50a8..f508df01145b 100644 --- a/sound/soc/codecs/es8316.c +++ b/sound/soc/codecs/es8316.c @@ -39,7 +39,9 @@ struct es8316_priv { struct snd_soc_jack *jack; int irq; unsigned int sysclk; - unsigned int allowed_rates[ARRAY_SIZE(supported_mclk_lrck_ratios)]; + /* ES83xx supports halving the MCLK so it supports twice as many rates + */ + unsigned int allowed_rates[ARRAY_SIZE(supported_mclk_lrck_ratios) * 2]; struct snd_pcm_hw_constraint_list sysclk_constraints; bool jd_inverted; }; @@ -386,6 +388,12 @@ static int es8316_set_dai_sysclk(struct snd_soc_dai *codec_dai, if (freq % ratio == 0) es8316->allowed_rates[count++] = freq / ratio; + + /* We also check if the halved MCLK produces a valid rate + * since the codec supports halving the MCLK. + */ + if ((freq / ratio) % 2 == 0) + es8316->allowed_rates[count++] = freq / ratio / 2; } if (count) { diff --git a/sound/soc/codecs/es8326.c b/sound/soc/codecs/es8326.c index a5603b617688..b06eead7e0f6 100644 --- a/sound/soc/codecs/es8326.c +++ b/sound/soc/codecs/es8326.c @@ -616,7 +616,7 @@ static int es8326_mute(struct snd_soc_dai *dai, int mute, int direction) 0x0F, 0x0F); if (es8326->version > ES8326_VERSION_B) { regmap_update_bits(es8326->regmap, ES8326_VMIDSEL, 0x40, 0x40); - regmap_update_bits(es8326->regmap, ES8326_ANA_MICBIAS, 0x70, 0x10); + regmap_update_bits(es8326->regmap, ES8326_ANA_MICBIAS, 0x70, 0x30); } } } else { @@ -631,6 +631,8 @@ static int es8326_mute(struct snd_soc_dai *dai, int mute, int direction) regmap_write(es8326->regmap, ES8326_HPR_OFFSET_INI, offset_r); es8326->calibrated = true; } + regmap_update_bits(es8326->regmap, ES8326_CLK_INV, 0xc0, 0x00); + regmap_update_bits(es8326->regmap, ES8326_CLK_MUX, 0x80, 0x00); if (direction == SNDRV_PCM_STREAM_PLAYBACK) { regmap_update_bits(es8326->regmap, ES8326_DAC_DSM, 0x01, 0x01); usleep_range(1000, 5000); @@ -645,7 +647,7 @@ static int es8326_mute(struct snd_soc_dai *dai, int mute, int direction) } else { msleep(300); if (es8326->version > ES8326_VERSION_B) { - regmap_update_bits(es8326->regmap, ES8326_ANA_MICBIAS, 0x70, 0x50); + regmap_update_bits(es8326->regmap, ES8326_ANA_MICBIAS, 0x70, 0x70); regmap_update_bits(es8326->regmap, ES8326_VMIDSEL, 0x40, 0x00); } regmap_update_bits(es8326->regmap, ES8326_ADC_MUTE, @@ -676,6 +678,10 @@ static int es8326_set_bias_level(struct snd_soc_component *codec, regmap_write(es8326->regmap, ES8326_ANA_PDN, 0x00); regmap_update_bits(es8326->regmap, ES8326_CLK_CTL, 0x20, 0x20); regmap_update_bits(es8326->regmap, ES8326_RESET, 0x02, 0x00); + if (es8326->version > ES8326_VERSION_B) { + regmap_update_bits(es8326->regmap, ES8326_VMIDSEL, 0x40, 0x40); + regmap_update_bits(es8326->regmap, ES8326_ANA_MICBIAS, 0x70, 0x30); + } break; case SND_SOC_BIAS_PREPARE: break; @@ -683,6 +689,12 @@ static int es8326_set_bias_level(struct snd_soc_component *codec, regmap_write(es8326->regmap, ES8326_ANA_PDN, 0x3b); regmap_update_bits(es8326->regmap, ES8326_CLK_CTL, 0x20, 0x00); regmap_write(es8326->regmap, ES8326_SDINOUT1_IO, ES8326_IO_INPUT); + if (es8326->version > ES8326_VERSION_B) { + regmap_update_bits(es8326->regmap, ES8326_VMIDSEL, 0x40, 0x40); + regmap_update_bits(es8326->regmap, ES8326_ANA_MICBIAS, 0x70, 0x10); + } + regmap_update_bits(es8326->regmap, ES8326_CLK_INV, 0xc0, 0xc0); + regmap_update_bits(es8326->regmap, ES8326_CLK_MUX, 0x80, 0x80); break; case SND_SOC_BIAS_OFF: clk_disable_unprepare(es8326->mclk); @@ -773,7 +785,10 @@ static void es8326_jack_button_handler(struct work_struct *work) case 0x6f: case 0x4b: /* button volume up */ - cur_button = SND_JACK_BTN_1; + if ((iface == 0x6f) && (es8326->version > ES8326_VERSION_B)) + cur_button = SND_JACK_BTN_0; + else + cur_button = SND_JACK_BTN_1; break; case 0x27: /* button volume down */ @@ -1082,7 +1097,7 @@ static void es8326_init(struct snd_soc_component *component) regmap_write(es8326->regmap, ES8326_ADC2_SRC, 0x66); es8326_disable_micbias(es8326->component); if (es8326->version > ES8326_VERSION_B) { - regmap_update_bits(es8326->regmap, ES8326_ANA_MICBIAS, 0x73, 0x13); + regmap_update_bits(es8326->regmap, ES8326_ANA_MICBIAS, 0x73, 0x10); regmap_update_bits(es8326->regmap, ES8326_VMIDSEL, 0x40, 0x40); } diff --git a/sound/soc/codecs/rt722-sdca.c b/sound/soc/codecs/rt722-sdca.c index 908846e994df..e17a142d03b9 100644 --- a/sound/soc/codecs/rt722-sdca.c +++ b/sound/soc/codecs/rt722-sdca.c @@ -1468,13 +1468,18 @@ static void rt722_sdca_jack_preset(struct rt722_sdca_priv *rt722) 0x008d); /* check HP calibration FSM status */ for (loop_check = 0; loop_check < chk_cnt; loop_check++) { + usleep_range(10000, 11000); ret = rt722_sdca_index_read(rt722, RT722_VENDOR_CALI, RT722_DAC_DC_CALI_CTL3, &calib_status); - if (ret < 0 || loop_check == chk_cnt) + if (ret < 0) dev_dbg(&rt722->slave->dev, "calibration failed!, ret=%d\n", ret); if ((calib_status & 0x0040) == 0x0) break; } + + if (loop_check == chk_cnt) + dev_dbg(&rt722->slave->dev, "%s, calibration time-out!\n", __func__); + /* Set ADC09 power entity floating control */ rt722_sdca_index_write(rt722, RT722_VENDOR_HDA_CTL, RT722_ADC0A_08_PDE_FLOAT_CTL, 0x2a12); diff --git a/sound/soc/codecs/tas2781-i2c.c b/sound/soc/codecs/tas2781-i2c.c index fb8cd2284fe8..728bf78ae71f 100644 --- a/sound/soc/codecs/tas2781-i2c.c +++ b/sound/soc/codecs/tas2781-i2c.c @@ -78,7 +78,7 @@ static const struct bulk_reg_val tas2781_cali_start_reg[] = { X2781_CL_STT_VAL(TAS2781_PRM_INT_MASK_REG, 0xfe, false), X2781_CL_STT_VAL(TAS2781_PRM_CLK_CFG_REG, 0xdd, false), X2781_CL_STT_VAL(TAS2781_PRM_RSVD_REG, 0x20, false), - X2781_CL_STT_VAL(TAS2781_PRM_TEST_57_REG, 0x14, false), + X2781_CL_STT_VAL(TAS2781_PRM_TEST_57_REG, 0x14, true), X2781_CL_STT_VAL(TAS2781_PRM_TEST_62_REG, 0x45, true), X2781_CL_STT_VAL(TAS2781_PRM_PVDD_UVLO_REG, 0x03, false), X2781_CL_STT_VAL(TAS2781_PRM_CHNL_0_REG, 0xa8, false), diff --git a/sound/soc/intel/boards/sof_sdw.c b/sound/soc/intel/boards/sof_sdw.c index f3369e569abc..c9f9c9b0de9b 100644 --- a/sound/soc/intel/boards/sof_sdw.c +++ b/sound/soc/intel/boards/sof_sdw.c @@ -632,7 +632,7 @@ static const struct dmi_system_id sof_sdw_quirk_table[] = { .callback = sof_sdw_quirk_cb, .matches = { DMI_MATCH(DMI_SYS_VENDOR, "LENOVO"), - DMI_EXACT_MATCH(DMI_PRODUCT_SKU, "233C") + DMI_MATCH(DMI_PRODUCT_NAME, "21QB") }, /* Note this quirk excludes the CODEC mic */ .driver_data = (void *)(SOC_SDW_CODEC_MIC), @@ -641,9 +641,26 @@ static const struct dmi_system_id sof_sdw_quirk_table[] = { .callback = sof_sdw_quirk_cb, .matches = { DMI_MATCH(DMI_SYS_VENDOR, "LENOVO"), - DMI_EXACT_MATCH(DMI_PRODUCT_SKU, "233B") + DMI_MATCH(DMI_PRODUCT_NAME, "21QA") }, - .driver_data = (void *)(SOC_SDW_SIDECAR_AMPS), + /* Note this quirk excludes the CODEC mic */ + .driver_data = (void *)(SOC_SDW_CODEC_MIC), + }, + { + .callback = sof_sdw_quirk_cb, + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "LENOVO"), + DMI_MATCH(DMI_PRODUCT_NAME, "21Q6") + }, + .driver_data = (void *)(SOC_SDW_SIDECAR_AMPS | SOC_SDW_CODEC_MIC), + }, + { + .callback = sof_sdw_quirk_cb, + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "LENOVO"), + DMI_MATCH(DMI_PRODUCT_NAME, "21Q7") + }, + .driver_data = (void *)(SOC_SDW_SIDECAR_AMPS | SOC_SDW_CODEC_MIC), }, /* ArrowLake devices */ diff --git a/sound/soc/mediatek/common/mtk-afe-platform-driver.c b/sound/soc/mediatek/common/mtk-afe-platform-driver.c index 9b72b2a7ae91..6b6330583941 100644 --- a/sound/soc/mediatek/common/mtk-afe-platform-driver.c +++ b/sound/soc/mediatek/common/mtk-afe-platform-driver.c @@ -120,8 +120,8 @@ int mtk_afe_pcm_new(struct snd_soc_component *component, struct mtk_base_afe *afe = snd_soc_component_get_drvdata(component); size = afe->mtk_afe_hardware->buffer_bytes_max; - snd_pcm_set_managed_buffer_all(pcm, SNDRV_DMA_TYPE_DEV, - afe->dev, size, size); + snd_pcm_set_managed_buffer_all(pcm, SNDRV_DMA_TYPE_DEV, afe->dev, 0, size); + return 0; } EXPORT_SYMBOL_GPL(mtk_afe_pcm_new); diff --git a/sound/soc/renesas/rcar/adg.c b/sound/soc/renesas/rcar/adg.c index 0f190abf00e7..191f212d338c 100644 --- a/sound/soc/renesas/rcar/adg.c +++ b/sound/soc/renesas/rcar/adg.c @@ -374,12 +374,12 @@ int rsnd_adg_ssi_clk_try_start(struct rsnd_mod *ssi_mod, unsigned int rate) return 0; } -void rsnd_adg_clk_control(struct rsnd_priv *priv, int enable) +int rsnd_adg_clk_control(struct rsnd_priv *priv, int enable) { struct rsnd_adg *adg = rsnd_priv_to_adg(priv); struct rsnd_mod *adg_mod = rsnd_mod_get(adg); struct clk *clk; - int i; + int ret = 0, i; if (enable) { rsnd_mod_bset(adg_mod, BRGCKR, 0x80770000, adg->ckr); @@ -389,18 +389,33 @@ void rsnd_adg_clk_control(struct rsnd_priv *priv, int enable) for_each_rsnd_clkin(clk, adg, i) { if (enable) { - clk_prepare_enable(clk); + ret = clk_prepare_enable(clk); /* * We shouldn't use clk_get_rate() under * atomic context. Let's keep it when * rsnd_adg_clk_enable() was called */ + if (ret < 0) + break; + adg->clkin_rate[i] = clk_get_rate(clk); } else { - clk_disable_unprepare(clk); + if (adg->clkin_rate[i]) + clk_disable_unprepare(clk); + + adg->clkin_rate[i] = 0; } } + + /* + * rsnd_adg_clk_enable() might return error (_disable() will not). + * We need to rollback in such case + */ + if (ret < 0) + rsnd_adg_clk_disable(priv); + + return ret; } static struct clk *rsnd_adg_create_null_clk(struct rsnd_priv *priv, @@ -753,7 +768,10 @@ int rsnd_adg_probe(struct rsnd_priv *priv) if (ret) return ret; - rsnd_adg_clk_enable(priv); + ret = rsnd_adg_clk_enable(priv); + if (ret) + return ret; + rsnd_adg_clk_dbg_info(priv, NULL); return 0; diff --git a/sound/soc/renesas/rcar/core.c b/sound/soc/renesas/rcar/core.c index e2234928c9e8..d3709fd0409e 100644 --- a/sound/soc/renesas/rcar/core.c +++ b/sound/soc/renesas/rcar/core.c @@ -2086,9 +2086,7 @@ static int __maybe_unused rsnd_resume(struct device *dev) { struct rsnd_priv *priv = dev_get_drvdata(dev); - rsnd_adg_clk_enable(priv); - - return 0; + return rsnd_adg_clk_enable(priv); } static const struct dev_pm_ops rsnd_pm_ops = { diff --git a/sound/soc/renesas/rcar/rsnd.h b/sound/soc/renesas/rcar/rsnd.h index 3c164d8e3b16..a5f54b65313c 100644 --- a/sound/soc/renesas/rcar/rsnd.h +++ b/sound/soc/renesas/rcar/rsnd.h @@ -608,7 +608,7 @@ int rsnd_adg_set_cmd_timsel_gen2(struct rsnd_mod *cmd_mod, struct rsnd_dai_stream *io); #define rsnd_adg_clk_enable(priv) rsnd_adg_clk_control(priv, 1) #define rsnd_adg_clk_disable(priv) rsnd_adg_clk_control(priv, 0) -void rsnd_adg_clk_control(struct rsnd_priv *priv, int enable); +int rsnd_adg_clk_control(struct rsnd_priv *priv, int enable); void rsnd_adg_clk_dbg_info(struct rsnd_priv *priv, struct seq_file *m); /* diff --git a/sound/soc/samsung/Kconfig b/sound/soc/samsung/Kconfig index 4b1ea7b2c796..60b4b7b75215 100644 --- a/sound/soc/samsung/Kconfig +++ b/sound/soc/samsung/Kconfig @@ -127,8 +127,9 @@ config SND_SOC_SAMSUNG_TM2_WM5110 config SND_SOC_SAMSUNG_ARIES_WM8994 tristate "SoC I2S Audio support for WM8994 on Aries" - depends on SND_SOC_SAMSUNG && MFD_WM8994 && IIO && EXTCON + depends on SND_SOC_SAMSUNG && I2C && IIO && EXTCON select SND_SOC_BT_SCO + select MFD_WM8994 select SND_SOC_WM8994 select SND_SAMSUNG_I2S help @@ -140,8 +141,9 @@ config SND_SOC_SAMSUNG_ARIES_WM8994 config SND_SOC_SAMSUNG_MIDAS_WM1811 tristate "SoC I2S Audio support for Midas boards" - depends on SND_SOC_SAMSUNG && IIO + depends on SND_SOC_SAMSUNG && I2C && IIO select SND_SAMSUNG_I2S + select MFD_WM8994 select SND_SOC_WM8994 help Say Y if you want to add support for SoC audio on the Midas boards. diff --git a/sound/soc/sof/intel/hda-dai.c b/sound/soc/sof/intel/hda-dai.c index c13f89b7065e..0db2a3e554fb 100644 --- a/sound/soc/sof/intel/hda-dai.c +++ b/sound/soc/sof/intel/hda-dai.c @@ -103,8 +103,10 @@ hda_dai_get_ops(struct snd_pcm_substream *substream, struct snd_soc_dai *cpu_dai return sdai->platform_private; } -int hda_link_dma_cleanup(struct snd_pcm_substream *substream, struct hdac_ext_stream *hext_stream, - struct snd_soc_dai *cpu_dai) +static int +hda_link_dma_cleanup(struct snd_pcm_substream *substream, + struct hdac_ext_stream *hext_stream, + struct snd_soc_dai *cpu_dai, bool release) { const struct hda_dai_widget_dma_ops *ops = hda_dai_get_ops(substream, cpu_dai); struct sof_intel_hda_stream *hda_stream; @@ -128,6 +130,17 @@ int hda_link_dma_cleanup(struct snd_pcm_substream *substream, struct hdac_ext_st snd_hdac_ext_bus_link_clear_stream_id(hlink, stream_tag); } + if (!release) { + /* + * Force stream reconfiguration without releasing the channel on + * subsequent stream restart (without free), including LinkDMA + * reset. + * The stream is released via hda_dai_hw_free() + */ + hext_stream->link_prepared = 0; + return 0; + } + if (ops->release_hext_stream) ops->release_hext_stream(sdev, cpu_dai, substream); @@ -211,7 +224,7 @@ static int __maybe_unused hda_dai_hw_free(struct snd_pcm_substream *substream, if (!hext_stream) return 0; - return hda_link_dma_cleanup(substream, hext_stream, cpu_dai); + return hda_link_dma_cleanup(substream, hext_stream, cpu_dai, true); } static int __maybe_unused hda_dai_hw_params_data(struct snd_pcm_substream *substream, @@ -304,7 +317,8 @@ static int __maybe_unused hda_dai_trigger(struct snd_pcm_substream *substream, i switch (cmd) { case SNDRV_PCM_TRIGGER_STOP: case SNDRV_PCM_TRIGGER_SUSPEND: - ret = hda_link_dma_cleanup(substream, hext_stream, dai); + ret = hda_link_dma_cleanup(substream, hext_stream, dai, + cmd == SNDRV_PCM_TRIGGER_STOP ? false : true); if (ret < 0) { dev_err(sdev->dev, "%s: failed to clean up link DMA\n", __func__); return ret; @@ -660,8 +674,7 @@ static int hda_dai_suspend(struct hdac_bus *bus) } ret = hda_link_dma_cleanup(hext_stream->link_substream, - hext_stream, - cpu_dai); + hext_stream, cpu_dai, true); if (ret < 0) return ret; } diff --git a/sound/soc/sof/intel/hda.h b/sound/soc/sof/intel/hda.h index 22bd9c3c8216..ee4ccc1a5490 100644 --- a/sound/soc/sof/intel/hda.h +++ b/sound/soc/sof/intel/hda.h @@ -1038,8 +1038,6 @@ const struct hda_dai_widget_dma_ops * hda_select_dai_widget_ops(struct snd_sof_dev *sdev, struct snd_sof_widget *swidget); int hda_dai_config(struct snd_soc_dapm_widget *w, unsigned int flags, struct snd_sof_dai_config_data *data); -int hda_link_dma_cleanup(struct snd_pcm_substream *substream, struct hdac_ext_stream *hext_stream, - struct snd_soc_dai *cpu_dai); static inline struct snd_sof_dev *widget_to_sdev(struct snd_soc_dapm_widget *w) { diff --git a/sound/usb/mixer_us16x08.c b/sound/usb/mixer_us16x08.c index 6eb7d93b358d..20ac32635f1f 100644 --- a/sound/usb/mixer_us16x08.c +++ b/sound/usb/mixer_us16x08.c @@ -687,7 +687,7 @@ static int snd_us16x08_meter_get(struct snd_kcontrol *kcontrol, struct usb_mixer_elem_info *elem = kcontrol->private_data; struct snd_usb_audio *chip = elem->head.mixer->chip; struct snd_us16x08_meter_store *store = elem->private_data; - u8 meter_urb[64]; + u8 meter_urb[64] = {0}; switch (kcontrol->private_value) { case 0: { diff --git a/tools/include/uapi/linux/stddef.h b/tools/include/uapi/linux/stddef.h index bb6ea517efb5..c53cde425406 100644 --- a/tools/include/uapi/linux/stddef.h +++ b/tools/include/uapi/linux/stddef.h @@ -8,6 +8,13 @@ #define __always_inline __inline__ #endif +/* Not all C++ standards support type declarations inside an anonymous union */ +#ifndef __cplusplus +#define __struct_group_tag(TAG) TAG +#else +#define __struct_group_tag(TAG) +#endif + /** * __struct_group() - Create a mirrored named and anonyomous struct * @@ -20,14 +27,14 @@ * and size: one anonymous and one named. The former's members can be used * normally without sub-struct naming, and the latter can be used to * reason about the start, end, and size of the group of struct members. - * The named struct can also be explicitly tagged for layer reuse, as well - * as both having struct attributes appended. + * The named struct can also be explicitly tagged for layer reuse (C only), + * as well as both having struct attributes appended. */ #define __struct_group(TAG, NAME, ATTRS, MEMBERS...) \ union { \ struct { MEMBERS } ATTRS; \ - struct TAG { MEMBERS } ATTRS NAME; \ - } + struct __struct_group_tag(TAG) { MEMBERS } ATTRS NAME; \ + } ATTRS /** * __DECLARE_FLEX_ARRAY() - Declare a flexible array usable in a union diff --git a/tools/net/ynl/ynl-gen-c.py b/tools/net/ynl/ynl-gen-c.py index d8201c4b1520..6750fdb42564 100755 --- a/tools/net/ynl/ynl-gen-c.py +++ b/tools/net/ynl/ynl-gen-c.py @@ -2384,6 +2384,17 @@ def print_kernel_family_struct_src(family, cw): if not kernel_can_gen_family_struct(family): return + if 'sock-priv' in family.kernel_family: + # Generate "trampolines" to make CFI happy + cw.write_func("static void", f"__{family.c_name}_nl_sock_priv_init", + [f"{family.c_name}_nl_sock_priv_init(priv);"], + ["void *priv"]) + cw.nl() + cw.write_func("static void", f"__{family.c_name}_nl_sock_priv_destroy", + [f"{family.c_name}_nl_sock_priv_destroy(priv);"], + ["void *priv"]) + cw.nl() + cw.block_start(f"struct genl_family {family.ident_name}_nl_family __ro_after_init =") cw.p('.name\t\t= ' + family.fam_key + ',') cw.p('.version\t= ' + family.ver_key + ',') @@ -2401,9 +2412,8 @@ def print_kernel_family_struct_src(family, cw): cw.p(f'.n_mcgrps\t= ARRAY_SIZE({family.c_name}_nl_mcgrps),') if 'sock-priv' in family.kernel_family: cw.p(f'.sock_priv_size\t= sizeof({family.kernel_family["sock-priv"]}),') - # Force cast here, actual helpers take pointer to the real type. - cw.p(f'.sock_priv_init\t= (void *){family.c_name}_nl_sock_priv_init,') - cw.p(f'.sock_priv_destroy = (void *){family.c_name}_nl_sock_priv_destroy,') + cw.p(f'.sock_priv_init\t= __{family.c_name}_nl_sock_priv_init,') + cw.p(f'.sock_priv_destroy = __{family.c_name}_nl_sock_priv_destroy,') cw.block_end(';') diff --git a/tools/objtool/noreturns.h b/tools/objtool/noreturns.h index f37614cc2c1b..b2174894f9f7 100644 --- a/tools/objtool/noreturns.h +++ b/tools/objtool/noreturns.h @@ -19,6 +19,7 @@ NORETURN(__x64_sys_exit_group) NORETURN(arch_cpu_idle_dead) NORETURN(bch2_trans_in_restart_error) NORETURN(bch2_trans_restart_error) +NORETURN(bch2_trans_unlocked_error) NORETURN(cpu_bringup_and_idle) NORETURN(cpu_startup_entry) NORETURN(do_exit) diff --git a/tools/sched_ext/include/scx/common.bpf.h b/tools/sched_ext/include/scx/common.bpf.h index 2f36b7b6418d..625f5b046776 100644 --- a/tools/sched_ext/include/scx/common.bpf.h +++ b/tools/sched_ext/include/scx/common.bpf.h @@ -40,9 +40,9 @@ void scx_bpf_dsq_insert(struct task_struct *p, u64 dsq_id, u64 slice, u64 enq_fl void scx_bpf_dsq_insert_vtime(struct task_struct *p, u64 dsq_id, u64 slice, u64 vtime, u64 enq_flags) __ksym __weak; u32 scx_bpf_dispatch_nr_slots(void) __ksym; void scx_bpf_dispatch_cancel(void) __ksym; -bool scx_bpf_dsq_move_to_local(u64 dsq_id) __ksym; -void scx_bpf_dsq_move_set_slice(struct bpf_iter_scx_dsq *it__iter, u64 slice) __ksym; -void scx_bpf_dsq_move_set_vtime(struct bpf_iter_scx_dsq *it__iter, u64 vtime) __ksym; +bool scx_bpf_dsq_move_to_local(u64 dsq_id) __ksym __weak; +void scx_bpf_dsq_move_set_slice(struct bpf_iter_scx_dsq *it__iter, u64 slice) __ksym __weak; +void scx_bpf_dsq_move_set_vtime(struct bpf_iter_scx_dsq *it__iter, u64 vtime) __ksym __weak; bool scx_bpf_dsq_move(struct bpf_iter_scx_dsq *it__iter, struct task_struct *p, u64 dsq_id, u64 enq_flags) __ksym __weak; bool scx_bpf_dsq_move_vtime(struct bpf_iter_scx_dsq *it__iter, struct task_struct *p, u64 dsq_id, u64 enq_flags) __ksym __weak; u32 scx_bpf_reenqueue_local(void) __ksym; diff --git a/tools/sched_ext/scx_central.c b/tools/sched_ext/scx_central.c index 21deea320bd7..e938156ed0a0 100644 --- a/tools/sched_ext/scx_central.c +++ b/tools/sched_ext/scx_central.c @@ -97,7 +97,7 @@ restart: SCX_BUG_ON(!cpuset, "Failed to allocate cpuset"); CPU_ZERO(cpuset); CPU_SET(skel->rodata->central_cpu, cpuset); - SCX_BUG_ON(sched_setaffinity(0, sizeof(cpuset), cpuset), + SCX_BUG_ON(sched_setaffinity(0, sizeof(*cpuset), cpuset), "Failed to affinitize to central CPU %d (max %d)", skel->rodata->central_cpu, skel->rodata->nr_cpu_ids - 1); CPU_FREE(cpuset); diff --git a/tools/testing/selftests/alsa/Makefile b/tools/testing/selftests/alsa/Makefile index 944279160fed..8dab90ad22bb 100644 --- a/tools/testing/selftests/alsa/Makefile +++ b/tools/testing/selftests/alsa/Makefile @@ -27,5 +27,5 @@ include ../lib.mk $(OUTPUT)/libatest.so: conf.c alsa-local.h $(CC) $(CFLAGS) -shared -fPIC $< $(LDLIBS) -o $@ -$(OUTPUT)/%: %.c $(TEST_GEN_PROGS_EXTENDED) alsa-local.h +$(OUTPUT)/%: %.c $(OUTPUT)/libatest.so alsa-local.h $(CC) $(CFLAGS) $< $(LDLIBS) -latest -o $@ diff --git a/tools/testing/selftests/cgroup/test_cpuset_prs.sh b/tools/testing/selftests/cgroup/test_cpuset_prs.sh index 03c1bdaed2c3..400a696a0d21 100755 --- a/tools/testing/selftests/cgroup/test_cpuset_prs.sh +++ b/tools/testing/selftests/cgroup/test_cpuset_prs.sh @@ -86,15 +86,15 @@ echo "" > test/cpuset.cpus # # If isolated CPUs have been reserved at boot time (as shown in -# cpuset.cpus.isolated), these isolated CPUs should be outside of CPUs 0-7 +# cpuset.cpus.isolated), these isolated CPUs should be outside of CPUs 0-8 # that will be used by this script for testing purpose. If not, some of -# the tests may fail incorrectly. These isolated CPUs will also be removed -# before being compared with the expected results. +# the tests may fail incorrectly. These pre-isolated CPUs should stay in +# an isolated state throughout the testing process for now. # BOOT_ISOLCPUS=$(cat $CGROUP2/cpuset.cpus.isolated) if [[ -n "$BOOT_ISOLCPUS" ]] then - [[ $(echo $BOOT_ISOLCPUS | sed -e "s/[,-].*//") -le 7 ]] && + [[ $(echo $BOOT_ISOLCPUS | sed -e "s/[,-].*//") -le 8 ]] && skip_test "Pre-isolated CPUs ($BOOT_ISOLCPUS) overlap CPUs to be tested" echo "Pre-isolated CPUs: $BOOT_ISOLCPUS" fi @@ -684,14 +684,18 @@ check_isolcpus() fi # + # Appending pre-isolated CPUs + # Even though CPU #8 isn't used for testing, it can't be pre-isolated + # to make appending those CPUs easier. + # + [[ -n "$BOOT_ISOLCPUS" ]] && { + EXPECT_VAL=${EXPECT_VAL:+${EXPECT_VAL},}${BOOT_ISOLCPUS} + EXPECT_VAL2=${EXPECT_VAL2:+${EXPECT_VAL2},}${BOOT_ISOLCPUS} + } + + # # Check cpuset.cpus.isolated cpumask # - if [[ -z "$BOOT_ISOLCPUS" ]] - then - ISOLCPUS=$(cat $ISCPUS) - else - ISOLCPUS=$(cat $ISCPUS | sed -e "s/,*$BOOT_ISOLCPUS//") - fi [[ "$EXPECT_VAL2" != "$ISOLCPUS" ]] && { # Take a 50ms pause and try again pause 0.05 @@ -731,8 +735,6 @@ check_isolcpus() fi done [[ "$ISOLCPUS" = *- ]] && ISOLCPUS=${ISOLCPUS}$LASTISOLCPU - [[ -n "BOOT_ISOLCPUS" ]] && - ISOLCPUS=$(echo $ISOLCPUS | sed -e "s/,*$BOOT_ISOLCPUS//") [[ "$EXPECT_VAL" = "$ISOLCPUS" ]] } @@ -836,8 +838,11 @@ run_state_test() # if available [[ -n "$ICPUS" ]] && { check_isolcpus $ICPUS - [[ $? -ne 0 ]] && test_fail $I "isolated CPU" \ - "Expect $ICPUS, get $ISOLCPUS instead" + [[ $? -ne 0 ]] && { + [[ -n "$BOOT_ISOLCPUS" ]] && ICPUS=${ICPUS},${BOOT_ISOLCPUS} + test_fail $I "isolated CPU" \ + "Expect $ICPUS, get $ISOLCPUS instead" + } } reset_cgroup_states # diff --git a/tools/testing/selftests/coredump/Makefile b/tools/testing/selftests/coredump/Makefile new file mode 100644 index 000000000000..ed210037b29d --- /dev/null +++ b/tools/testing/selftests/coredump/Makefile @@ -0,0 +1,7 @@ +# SPDX-License-Identifier: GPL-2.0-only +CFLAGS = $(KHDR_INCLUDES) + +TEST_GEN_PROGS := stackdump_test +TEST_FILES := stackdump + +include ../lib.mk diff --git a/tools/testing/selftests/coredump/README.rst b/tools/testing/selftests/coredump/README.rst new file mode 100644 index 000000000000..164a7aa181c8 --- /dev/null +++ b/tools/testing/selftests/coredump/README.rst @@ -0,0 +1,50 @@ +coredump selftest +================= + +Background context +------------------ + +`coredump` is a feature which dumps a process's memory space when the process terminates +unexpectedly (e.g. due to segmentation fault), which can be useful for debugging. By default, +`coredump` dumps the memory to the file named `core`, but this behavior can be changed by writing a +different file name to `/proc/sys/kernel/core_pattern`. Furthermore, `coredump` can be piped to a +user-space program by writing the pipe symbol (`|`) followed by the command to be executed to +`/proc/sys/kernel/core_pattern`. For the full description, see `man 5 core`. + +The piped user program may be interested in reading the stack pointers of the crashed process. The +crashed process's stack pointers can be read from `procfs`: it is the `kstkesp` field in +`/proc/$PID/stat`. See `man 5 proc` for all the details. + +The problem +----------- +While a thread is active, the stack pointer is unsafe to read and therefore the `kstkesp` field +reads zero. But when the thread is dead (e.g. during a coredump), this field should have valid +value. + +However, this was broken in the past and `kstkesp` was zero even during coredump: + +* commit 0a1eb2d474ed ("fs/proc: Stop reporting eip and esp in /proc/PID/stat") changed kstkesp to + always be zero + +* commit fd7d56270b52 ("fs/proc: Report eip/esp in /prod/PID/stat for coredumping") fixed it for the + coredumping thread. However, other threads in a coredumping process still had the problem. + +* commit cb8f381f1613 ("fs/proc/array.c: allow reporting eip/esp for all coredumping threads") fixed + for all threads in a coredumping process. + +* commit 92307383082d ("coredump: Don't perform any cleanups before dumping core") broke it again + for the other threads in a coredumping process. + +The problem has been fixed now, but considering the history, it may appear again in the future. + +The goal of this test +--------------------- +This test detects problem with reading `kstkesp` during coredump by doing the following: + +#. Tell the kernel to execute the "stackdump" script when a coredump happens. This script + reads the stack pointers of all threads of crashed processes. + +#. Spawn a child process who creates some threads and then crashes. + +#. Read the output from the "stackdump" script, and make sure all stack pointer values are + non-zero. diff --git a/tools/testing/selftests/coredump/stackdump b/tools/testing/selftests/coredump/stackdump new file mode 100755 index 000000000000..96714ce42d12 --- /dev/null +++ b/tools/testing/selftests/coredump/stackdump @@ -0,0 +1,14 @@ +#!/bin/sh +# SPDX-License-Identifier: GPL-2.0 + +CRASH_PROGRAM_ID=$1 +STACKDUMP_FILE=$2 + +TMP=$(mktemp) + +for t in /proc/$CRASH_PROGRAM_ID/task/*; do + tid=$(basename $t) + cat /proc/$tid/stat | awk '{print $29}' >> $TMP +done + +mv $TMP $STACKDUMP_FILE diff --git a/tools/testing/selftests/coredump/stackdump_test.c b/tools/testing/selftests/coredump/stackdump_test.c new file mode 100644 index 000000000000..137b2364a082 --- /dev/null +++ b/tools/testing/selftests/coredump/stackdump_test.c @@ -0,0 +1,151 @@ +// SPDX-License-Identifier: GPL-2.0 + +#include <fcntl.h> +#include <libgen.h> +#include <linux/limits.h> +#include <pthread.h> +#include <string.h> +#include <sys/resource.h> +#include <unistd.h> + +#include "../kselftest_harness.h" + +#define STACKDUMP_FILE "stack_values" +#define STACKDUMP_SCRIPT "stackdump" +#define NUM_THREAD_SPAWN 128 + +static void *do_nothing(void *) +{ + while (1) + pause(); +} + +static void crashing_child(void) +{ + pthread_t thread; + int i; + + for (i = 0; i < NUM_THREAD_SPAWN; ++i) + pthread_create(&thread, NULL, do_nothing, NULL); + + /* crash on purpose */ + i = *(int *)NULL; +} + +FIXTURE(coredump) +{ + char original_core_pattern[256]; +}; + +FIXTURE_SETUP(coredump) +{ + char buf[PATH_MAX]; + FILE *file; + char *dir; + int ret; + + file = fopen("/proc/sys/kernel/core_pattern", "r"); + ASSERT_NE(NULL, file); + + ret = fread(self->original_core_pattern, 1, sizeof(self->original_core_pattern), file); + ASSERT_TRUE(ret || feof(file)); + ASSERT_LT(ret, sizeof(self->original_core_pattern)); + + self->original_core_pattern[ret] = '\0'; + + ret = fclose(file); + ASSERT_EQ(0, ret); +} + +FIXTURE_TEARDOWN(coredump) +{ + const char *reason; + FILE *file; + int ret; + + unlink(STACKDUMP_FILE); + + file = fopen("/proc/sys/kernel/core_pattern", "w"); + if (!file) { + reason = "Unable to open core_pattern"; + goto fail; + } + + ret = fprintf(file, "%s", self->original_core_pattern); + if (ret < 0) { + reason = "Unable to write to core_pattern"; + goto fail; + } + + ret = fclose(file); + if (ret) { + reason = "Unable to close core_pattern"; + goto fail; + } + + return; +fail: + /* This should never happen */ + fprintf(stderr, "Failed to cleanup stackdump test: %s\n", reason); +} + +TEST_F(coredump, stackdump) +{ + struct sigaction action = {}; + unsigned long long stack; + char *test_dir, *line; + size_t line_length; + char buf[PATH_MAX]; + int ret, i; + FILE *file; + pid_t pid; + + /* + * Step 1: Setup core_pattern so that the stackdump script is executed when the child + * process crashes + */ + ret = readlink("/proc/self/exe", buf, sizeof(buf)); + ASSERT_NE(-1, ret); + ASSERT_LT(ret, sizeof(buf)); + buf[ret] = '\0'; + + test_dir = dirname(buf); + + file = fopen("/proc/sys/kernel/core_pattern", "w"); + ASSERT_NE(NULL, file); + + ret = fprintf(file, "|%1$s/%2$s %%P %1$s/%3$s", test_dir, STACKDUMP_SCRIPT, STACKDUMP_FILE); + ASSERT_LT(0, ret); + + ret = fclose(file); + ASSERT_EQ(0, ret); + + /* Step 2: Create a process who spawns some threads then crashes */ + pid = fork(); + ASSERT_TRUE(pid >= 0); + if (pid == 0) + crashing_child(); + + /* + * Step 3: Wait for the stackdump script to write the stack pointers to the stackdump file + */ + for (i = 0; i < 10; ++i) { + file = fopen(STACKDUMP_FILE, "r"); + if (file) + break; + sleep(1); + } + ASSERT_NE(file, NULL); + + /* Step 4: Make sure all stack pointer values are non-zero */ + for (i = 0; -1 != getline(&line, &line_length, file); ++i) { + stack = strtoull(line, NULL, 10); + ASSERT_NE(stack, 0); + } + + ASSERT_EQ(i, 1 + NUM_THREAD_SPAWN); + + fclose(file); +} + +TEST_HARNESS_MAIN diff --git a/tools/testing/selftests/drivers/net/netdevsim/tc-mq-visibility.sh b/tools/testing/selftests/drivers/net/netdevsim/tc-mq-visibility.sh index fd13c8cfb7a8..b411fe66510f 100755 --- a/tools/testing/selftests/drivers/net/netdevsim/tc-mq-visibility.sh +++ b/tools/testing/selftests/drivers/net/netdevsim/tc-mq-visibility.sh @@ -58,9 +58,12 @@ for root in mq mqprio; do ethtool -L $NDEV combined 4 n_child_assert 4 "One real queue, rest default" - # Graft some - tcq replace parent 100:1 handle 204: - n_child_assert 3 "Grafted" + # Remove real one + tcq del parent 100:4 handle 204: + + # Replace default with pfifo + tcq replace parent 100:1 handle 205: pfifo limit 1000 + n_child_assert 3 "Deleting real one, replacing default one with pfifo" ethtool -L $NDEV combined 1 n_child_assert 1 "Grafted, one" diff --git a/tools/testing/selftests/drivers/net/queues.py b/tools/testing/selftests/drivers/net/queues.py index 9c5473abbd78..38303da957ee 100755 --- a/tools/testing/selftests/drivers/net/queues.py +++ b/tools/testing/selftests/drivers/net/queues.py @@ -1,10 +1,12 @@ #!/usr/bin/env python3 # SPDX-License-Identifier: GPL-2.0 -from lib.py import ksft_run, ksft_exit, ksft_eq, KsftSkipEx -from lib.py import EthtoolFamily, NetdevFamily +from lib.py import ksft_disruptive, ksft_exit, ksft_run +from lib.py import ksft_eq, ksft_raises, KsftSkipEx +from lib.py import EthtoolFamily, NetdevFamily, NlError from lib.py import NetDrvEnv -from lib.py import cmd +from lib.py import cmd, defer, ip +import errno import glob @@ -59,9 +61,27 @@ def addremove_queues(cfg, nl) -> None: ksft_eq(queues, expected) +@ksft_disruptive +def check_down(cfg, nl) -> None: + # Check the NAPI IDs before interface goes down and hides them + napis = nl.napi_get({'ifindex': cfg.ifindex}, dump=True) + + ip(f"link set dev {cfg.dev['ifname']} down") + defer(ip, f"link set dev {cfg.dev['ifname']} up") + + with ksft_raises(NlError) as cm: + nl.queue_get({'ifindex': cfg.ifindex, 'id': 0, 'type': 'rx'}) + ksft_eq(cm.exception.nl_msg.error, -errno.ENOENT) + + if napis: + with ksft_raises(NlError) as cm: + nl.napi_get({'id': napis[0]['id']}) + ksft_eq(cm.exception.nl_msg.error, -errno.ENOENT) + + def main() -> None: with NetDrvEnv(__file__, queue_count=100) as cfg: - ksft_run([get_queues, addremove_queues], args=(cfg, NetdevFamily())) + ksft_run([get_queues, addremove_queues, check_down], args=(cfg, NetdevFamily())) ksft_exit() diff --git a/tools/testing/selftests/exec/execveat.c b/tools/testing/selftests/exec/execveat.c index 071e03532cba..8fb7395fd35b 100644 --- a/tools/testing/selftests/exec/execveat.c +++ b/tools/testing/selftests/exec/execveat.c @@ -23,9 +23,11 @@ #include "../kselftest.h" -#define TESTS_EXPECTED 51 +#define TESTS_EXPECTED 54 #define TEST_NAME_LEN (PATH_MAX * 4) +#define CHECK_COMM "CHECK_COMM" + static char longpath[2 * PATH_MAX] = ""; static char *envp[] = { "IN_TEST=yes", NULL, NULL }; static char *argv[] = { "execveat", "99", NULL }; @@ -237,6 +239,29 @@ static int check_execveat_pathmax(int root_dfd, const char *src, int is_script) return fail; } +static int check_execveat_comm(int fd, char *argv0, char *expected) +{ + char buf[128], *old_env, *old_argv0; + int ret; + + snprintf(buf, sizeof(buf), CHECK_COMM "=%s", expected); + + old_env = envp[1]; + envp[1] = buf; + + old_argv0 = argv[0]; + argv[0] = argv0; + + ksft_print_msg("Check execveat(AT_EMPTY_PATH)'s comm is %s\n", + expected); + ret = check_execveat_invoked_rc(fd, "", AT_EMPTY_PATH, 0, 0); + + envp[1] = old_env; + argv[0] = old_argv0; + + return ret; +} + static int run_tests(void) { int fail = 0; @@ -389,6 +414,14 @@ static int run_tests(void) fail += check_execveat_pathmax(root_dfd, "execveat", 0); fail += check_execveat_pathmax(root_dfd, "script", 1); + + /* /proc/pid/comm gives filename by default */ + fail += check_execveat_comm(fd, "sentinel", "execveat"); + /* /proc/pid/comm gives argv[0] when invoked via link */ + fail += check_execveat_comm(fd_symlink, "sentinel", "execveat"); + /* /proc/pid/comm gives filename if NULL is passed */ + fail += check_execveat_comm(fd, NULL, "execveat"); + return fail; } @@ -415,9 +448,13 @@ int main(int argc, char **argv) int ii; int rc; const char *verbose = getenv("VERBOSE"); + const char *check_comm = getenv(CHECK_COMM); - if (argc >= 2) { - /* If we are invoked with an argument, don't run tests. */ + if (argc >= 2 || check_comm) { + /* + * If we are invoked with an argument, or no arguments but a + * command to check, don't run tests. + */ const char *in_test = getenv("IN_TEST"); if (verbose) { @@ -426,6 +463,38 @@ int main(int argc, char **argv) ksft_print_msg("\t[%d]='%s\n'", ii, argv[ii]); } + /* If the tests wanted us to check the command, do so. */ + if (check_comm) { + /* TASK_COMM_LEN == 16 */ + char buf[32]; + int fd, ret; + + fd = open("/proc/self/comm", O_RDONLY); + if (fd < 0) { + ksft_perror("open() comm failed"); + exit(1); + } + + ret = read(fd, buf, sizeof(buf)); + if (ret < 0) { + ksft_perror("read() comm failed"); + close(fd); + exit(1); + } + close(fd); + + // trim off the \n + buf[ret-1] = 0; + + if (strcmp(buf, check_comm)) { + ksft_print_msg("bad comm, got: %s expected: %s\n", + buf, check_comm); + exit(1); + } + + exit(0); + } + /* Check expected environment transferred. */ if (!in_test || strcmp(in_test, "yes") != 0) { ksft_print_msg("no IN_TEST=yes in env\n"); diff --git a/tools/testing/selftests/nsfs/.gitignore b/tools/testing/selftests/filesystems/nsfs/.gitignore index ed79ebdf286e..92a8249006d1 100644 --- a/tools/testing/selftests/nsfs/.gitignore +++ b/tools/testing/selftests/filesystems/nsfs/.gitignore @@ -1,3 +1,4 @@ # SPDX-License-Identifier: GPL-2.0-only owner pidns +iterate_mntns diff --git a/tools/testing/selftests/nsfs/Makefile b/tools/testing/selftests/filesystems/nsfs/Makefile index dd9bd50b7b93..231aaa7dfd95 100644 --- a/tools/testing/selftests/nsfs/Makefile +++ b/tools/testing/selftests/filesystems/nsfs/Makefile @@ -1,6 +1,6 @@ # SPDX-License-Identifier: GPL-2.0-only -TEST_GEN_PROGS := owner pidns +TEST_GEN_PROGS := owner pidns iterate_mntns CFLAGS := -Wall -Werror -include ../lib.mk +include ../../lib.mk diff --git a/tools/testing/selftests/nsfs/config b/tools/testing/selftests/filesystems/nsfs/config index 598d0a225fc9..598d0a225fc9 100644 --- a/tools/testing/selftests/nsfs/config +++ b/tools/testing/selftests/filesystems/nsfs/config diff --git a/tools/testing/selftests/filesystems/nsfs/iterate_mntns.c b/tools/testing/selftests/filesystems/nsfs/iterate_mntns.c new file mode 100644 index 000000000000..457cf76f3c5f --- /dev/null +++ b/tools/testing/selftests/filesystems/nsfs/iterate_mntns.c @@ -0,0 +1,149 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +// Copyright (c) 2024 Christian Brauner <brauner@kernel.org> + +#define _GNU_SOURCE +#include <fcntl.h> +#include <sched.h> +#include <stdio.h> +#include <string.h> +#include <sys/stat.h> +#include <sys/mount.h> +#include <unistd.h> + +#include "../../kselftest_harness.h" + +#define MNT_NS_COUNT 11 +#define MNT_NS_LAST_INDEX 10 + +struct mnt_ns_info { + __u32 size; + __u32 nr_mounts; + __u64 mnt_ns_id; +}; + +#define MNT_NS_INFO_SIZE_VER0 16 /* size of first published struct */ + +/* Get information about namespace. */ +#define NS_MNT_GET_INFO _IOR(0xb7, 10, struct mnt_ns_info) +/* Get next namespace. */ +#define NS_MNT_GET_NEXT _IOR(0xb7, 11, struct mnt_ns_info) +/* Get previous namespace. */ +#define NS_MNT_GET_PREV _IOR(0xb7, 12, struct mnt_ns_info) + +FIXTURE(iterate_mount_namespaces) { + int fd_mnt_ns[MNT_NS_COUNT]; + __u64 mnt_ns_id[MNT_NS_COUNT]; +}; + +FIXTURE_SETUP(iterate_mount_namespaces) +{ + for (int i = 0; i < MNT_NS_COUNT; i++) + self->fd_mnt_ns[i] = -EBADF; + + /* + * Creating a new user namespace let's us guarantee that we only see + * mount namespaces that we did actually create. + */ + ASSERT_EQ(unshare(CLONE_NEWUSER), 0); + + for (int i = 0; i < MNT_NS_COUNT; i++) { + struct mnt_ns_info info = {}; + + ASSERT_EQ(unshare(CLONE_NEWNS), 0); + self->fd_mnt_ns[i] = open("/proc/self/ns/mnt", O_RDONLY | O_CLOEXEC); + ASSERT_GE(self->fd_mnt_ns[i], 0); + ASSERT_EQ(ioctl(self->fd_mnt_ns[i], NS_MNT_GET_INFO, &info), 0); + self->mnt_ns_id[i] = info.mnt_ns_id; + } +} + +FIXTURE_TEARDOWN(iterate_mount_namespaces) +{ + for (int i = 0; i < MNT_NS_COUNT; i++) { + if (self->fd_mnt_ns[i] < 0) + continue; + ASSERT_EQ(close(self->fd_mnt_ns[i]), 0); + } +} + +TEST_F(iterate_mount_namespaces, iterate_all_forward) +{ + int fd_mnt_ns_cur, count = 0; + + fd_mnt_ns_cur = fcntl(self->fd_mnt_ns[0], F_DUPFD_CLOEXEC); + ASSERT_GE(fd_mnt_ns_cur, 0); + + for (;; count++) { + struct mnt_ns_info info = {}; + int fd_mnt_ns_next; + + fd_mnt_ns_next = ioctl(fd_mnt_ns_cur, NS_MNT_GET_NEXT, &info); + if (fd_mnt_ns_next < 0 && errno == ENOENT) + break; + ASSERT_GE(fd_mnt_ns_next, 0); + ASSERT_EQ(close(fd_mnt_ns_cur), 0); + fd_mnt_ns_cur = fd_mnt_ns_next; + } + ASSERT_EQ(count, MNT_NS_LAST_INDEX); +} + +TEST_F(iterate_mount_namespaces, iterate_all_backwards) +{ + int fd_mnt_ns_cur, count = 0; + + fd_mnt_ns_cur = fcntl(self->fd_mnt_ns[MNT_NS_LAST_INDEX], F_DUPFD_CLOEXEC); + ASSERT_GE(fd_mnt_ns_cur, 0); + + for (;; count++) { + struct mnt_ns_info info = {}; + int fd_mnt_ns_prev; + + fd_mnt_ns_prev = ioctl(fd_mnt_ns_cur, NS_MNT_GET_PREV, &info); + if (fd_mnt_ns_prev < 0 && errno == ENOENT) + break; + ASSERT_GE(fd_mnt_ns_prev, 0); + ASSERT_EQ(close(fd_mnt_ns_cur), 0); + fd_mnt_ns_cur = fd_mnt_ns_prev; + } + ASSERT_EQ(count, MNT_NS_LAST_INDEX); +} + +TEST_F(iterate_mount_namespaces, iterate_forward) +{ + int fd_mnt_ns_cur; + + ASSERT_EQ(setns(self->fd_mnt_ns[0], CLONE_NEWNS), 0); + + fd_mnt_ns_cur = self->fd_mnt_ns[0]; + for (int i = 1; i < MNT_NS_COUNT; i++) { + struct mnt_ns_info info = {}; + int fd_mnt_ns_next; + + fd_mnt_ns_next = ioctl(fd_mnt_ns_cur, NS_MNT_GET_NEXT, &info); + ASSERT_GE(fd_mnt_ns_next, 0); + ASSERT_EQ(close(fd_mnt_ns_cur), 0); + fd_mnt_ns_cur = fd_mnt_ns_next; + ASSERT_EQ(info.mnt_ns_id, self->mnt_ns_id[i]); + } +} + +TEST_F(iterate_mount_namespaces, iterate_backward) +{ + int fd_mnt_ns_cur; + + ASSERT_EQ(setns(self->fd_mnt_ns[MNT_NS_LAST_INDEX], CLONE_NEWNS), 0); + + fd_mnt_ns_cur = self->fd_mnt_ns[MNT_NS_LAST_INDEX]; + for (int i = MNT_NS_LAST_INDEX - 1; i >= 0; i--) { + struct mnt_ns_info info = {}; + int fd_mnt_ns_prev; + + fd_mnt_ns_prev = ioctl(fd_mnt_ns_cur, NS_MNT_GET_PREV, &info); + ASSERT_GE(fd_mnt_ns_prev, 0); + ASSERT_EQ(close(fd_mnt_ns_cur), 0); + fd_mnt_ns_cur = fd_mnt_ns_prev; + ASSERT_EQ(info.mnt_ns_id, self->mnt_ns_id[i]); + } +} + +TEST_HARNESS_MAIN diff --git a/tools/testing/selftests/nsfs/owner.c b/tools/testing/selftests/filesystems/nsfs/owner.c index 96a976c74550..96a976c74550 100644 --- a/tools/testing/selftests/nsfs/owner.c +++ b/tools/testing/selftests/filesystems/nsfs/owner.c diff --git a/tools/testing/selftests/nsfs/pidns.c b/tools/testing/selftests/filesystems/nsfs/pidns.c index e3c772c6a7c7..e3c772c6a7c7 100644 --- a/tools/testing/selftests/nsfs/pidns.c +++ b/tools/testing/selftests/filesystems/nsfs/pidns.c diff --git a/tools/testing/selftests/filesystems/statmount/Makefile b/tools/testing/selftests/filesystems/statmount/Makefile index 3af3136e35a4..14ee91a41650 100644 --- a/tools/testing/selftests/filesystems/statmount/Makefile +++ b/tools/testing/selftests/filesystems/statmount/Makefile @@ -1,6 +1,6 @@ # SPDX-License-Identifier: GPL-2.0-or-later CFLAGS += -Wall -O2 -g $(KHDR_INCLUDES) -TEST_GEN_PROGS := statmount_test statmount_test_ns +TEST_GEN_PROGS := statmount_test statmount_test_ns listmount_test include ../../lib.mk diff --git a/tools/testing/selftests/filesystems/statmount/listmount_test.c b/tools/testing/selftests/filesystems/statmount/listmount_test.c new file mode 100644 index 000000000000..15f0834f7557 --- /dev/null +++ b/tools/testing/selftests/filesystems/statmount/listmount_test.c @@ -0,0 +1,66 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +// Copyright (c) 2024 Christian Brauner <brauner@kernel.org> + +#define _GNU_SOURCE +#include <fcntl.h> +#include <sched.h> +#include <stdio.h> +#include <string.h> +#include <sys/stat.h> +#include <sys/mount.h> +#include <unistd.h> + +#include "statmount.h" +#include "../../kselftest_harness.h" + +#ifndef LISTMOUNT_REVERSE +#define LISTMOUNT_REVERSE (1 << 0) /* List later mounts first */ +#endif + +#define LISTMNT_BUFFER 10 + +/* Check that all mount ids are in increasing order. */ +TEST(listmount_forward) +{ + uint64_t list[LISTMNT_BUFFER], last_mnt_id = 0; + + for (;;) { + ssize_t nr_mounts; + + nr_mounts = listmount(LSMT_ROOT, 0, last_mnt_id, + list, LISTMNT_BUFFER, 0); + ASSERT_GE(nr_mounts, 0); + if (nr_mounts == 0) + break; + + for (size_t cur = 0; cur < nr_mounts; cur++) { + if (cur < nr_mounts - 1) + ASSERT_LT(list[cur], list[cur + 1]); + last_mnt_id = list[cur]; + } + } +} + +/* Check that all mount ids are in decreasing order. */ +TEST(listmount_backward) +{ + uint64_t list[LISTMNT_BUFFER], last_mnt_id = 0; + + for (;;) { + ssize_t nr_mounts; + + nr_mounts = listmount(LSMT_ROOT, 0, last_mnt_id, + list, LISTMNT_BUFFER, LISTMOUNT_REVERSE); + ASSERT_GE(nr_mounts, 0); + if (nr_mounts == 0) + break; + + for (size_t cur = 0; cur < nr_mounts; cur++) { + if (cur < nr_mounts - 1) + ASSERT_GT(list[cur], list[cur + 1]); + last_mnt_id = list[cur]; + } + } +} + +TEST_HARNESS_MAIN diff --git a/tools/testing/selftests/kvm/aarch64/set_id_regs.c b/tools/testing/selftests/kvm/aarch64/set_id_regs.c index a79b7f18452d..3a97c160b5fe 100644 --- a/tools/testing/selftests/kvm/aarch64/set_id_regs.c +++ b/tools/testing/selftests/kvm/aarch64/set_id_regs.c @@ -152,7 +152,6 @@ static const struct reg_ftr_bits ftr_id_aa64mmfr0_el1[] = { REG_FTR_BITS(FTR_LOWER_SAFE, ID_AA64MMFR0_EL1, BIGENDEL0, 0), REG_FTR_BITS(FTR_LOWER_SAFE, ID_AA64MMFR0_EL1, SNSMEM, 0), REG_FTR_BITS(FTR_LOWER_SAFE, ID_AA64MMFR0_EL1, BIGEND, 0), - REG_FTR_BITS(FTR_LOWER_SAFE, ID_AA64MMFR0_EL1, ASIDBITS, 0), REG_FTR_BITS(FTR_LOWER_SAFE, ID_AA64MMFR0_EL1, PARANGE, 0), REG_FTR_END, }; diff --git a/tools/testing/selftests/kvm/s390x/ucontrol_test.c b/tools/testing/selftests/kvm/s390x/ucontrol_test.c index 0c112319dab1..135ee22856cf 100644 --- a/tools/testing/selftests/kvm/s390x/ucontrol_test.c +++ b/tools/testing/selftests/kvm/s390x/ucontrol_test.c @@ -210,10 +210,13 @@ TEST_F(uc_kvm, uc_attr_mem_limit) struct kvm_device_attr attr = { .group = KVM_S390_VM_MEM_CTRL, .attr = KVM_S390_VM_MEM_LIMIT_SIZE, - .addr = (unsigned long)&limit, + .addr = (u64)&limit, }; int rc; + rc = ioctl(self->vm_fd, KVM_HAS_DEVICE_ATTR, &attr); + EXPECT_EQ(0, rc); + rc = ioctl(self->vm_fd, KVM_GET_DEVICE_ATTR, &attr); EXPECT_EQ(0, rc); EXPECT_EQ(~0UL, limit); @@ -635,4 +638,171 @@ TEST_F(uc_kvm, uc_skey) uc_assert_diag44(self); } +static char uc_flic_b[PAGE_SIZE]; +static struct kvm_s390_io_adapter uc_flic_ioa = { .id = 0 }; +static struct kvm_s390_io_adapter_req uc_flic_ioam = { .id = 0 }; +static struct kvm_s390_ais_req uc_flic_asim = { .isc = 0 }; +static struct kvm_s390_ais_all uc_flic_asima = { .simm = 0 }; +static struct uc_flic_attr_test { + char *name; + struct kvm_device_attr a; + int hasrc; + int geterrno; + int seterrno; +} uc_flic_attr_tests[] = { + { + .name = "KVM_DEV_FLIC_GET_ALL_IRQS", + .seterrno = EINVAL, + .a = { + .group = KVM_DEV_FLIC_GET_ALL_IRQS, + .addr = (u64)&uc_flic_b, + .attr = PAGE_SIZE, + }, + }, + { + .name = "KVM_DEV_FLIC_ENQUEUE", + .geterrno = EINVAL, + .a = { .group = KVM_DEV_FLIC_ENQUEUE, }, + }, + { + .name = "KVM_DEV_FLIC_CLEAR_IRQS", + .geterrno = EINVAL, + .a = { .group = KVM_DEV_FLIC_CLEAR_IRQS, }, + }, + { + .name = "KVM_DEV_FLIC_ADAPTER_REGISTER", + .geterrno = EINVAL, + .a = { + .group = KVM_DEV_FLIC_ADAPTER_REGISTER, + .addr = (u64)&uc_flic_ioa, + }, + }, + { + .name = "KVM_DEV_FLIC_ADAPTER_MODIFY", + .geterrno = EINVAL, + .seterrno = EINVAL, + .a = { + .group = KVM_DEV_FLIC_ADAPTER_MODIFY, + .addr = (u64)&uc_flic_ioam, + .attr = sizeof(uc_flic_ioam), + }, + }, + { + .name = "KVM_DEV_FLIC_CLEAR_IO_IRQ", + .geterrno = EINVAL, + .seterrno = EINVAL, + .a = { + .group = KVM_DEV_FLIC_CLEAR_IO_IRQ, + .attr = 32, + }, + }, + { + .name = "KVM_DEV_FLIC_AISM", + .geterrno = EINVAL, + .seterrno = ENOTSUP, + .a = { + .group = KVM_DEV_FLIC_AISM, + .addr = (u64)&uc_flic_asim, + }, + }, + { + .name = "KVM_DEV_FLIC_AIRQ_INJECT", + .geterrno = EINVAL, + .a = { .group = KVM_DEV_FLIC_AIRQ_INJECT, }, + }, + { + .name = "KVM_DEV_FLIC_AISM_ALL", + .geterrno = ENOTSUP, + .seterrno = ENOTSUP, + .a = { + .group = KVM_DEV_FLIC_AISM_ALL, + .addr = (u64)&uc_flic_asima, + .attr = sizeof(uc_flic_asima), + }, + }, + { + .name = "KVM_DEV_FLIC_APF_ENABLE", + .geterrno = EINVAL, + .seterrno = EINVAL, + .a = { .group = KVM_DEV_FLIC_APF_ENABLE, }, + }, + { + .name = "KVM_DEV_FLIC_APF_DISABLE_WAIT", + .geterrno = EINVAL, + .seterrno = EINVAL, + .a = { .group = KVM_DEV_FLIC_APF_DISABLE_WAIT, }, + }, +}; + +TEST_F(uc_kvm, uc_flic_attrs) +{ + struct kvm_create_device cd = { .type = KVM_DEV_TYPE_FLIC }; + struct kvm_device_attr attr; + u64 value; + int rc, i; + + rc = ioctl(self->vm_fd, KVM_CREATE_DEVICE, &cd); + ASSERT_EQ(0, rc) TH_LOG("create device failed with err %s (%i)", + strerror(errno), errno); + + for (i = 0; i < ARRAY_SIZE(uc_flic_attr_tests); i++) { + TH_LOG("test %s", uc_flic_attr_tests[i].name); + attr = (struct kvm_device_attr) { + .group = uc_flic_attr_tests[i].a.group, + .attr = uc_flic_attr_tests[i].a.attr, + .addr = uc_flic_attr_tests[i].a.addr, + }; + if (attr.addr == 0) + attr.addr = (u64)&value; + + rc = ioctl(cd.fd, KVM_HAS_DEVICE_ATTR, &attr); + EXPECT_EQ(uc_flic_attr_tests[i].hasrc, !!rc) + TH_LOG("expected dev attr missing %s", + uc_flic_attr_tests[i].name); + + rc = ioctl(cd.fd, KVM_GET_DEVICE_ATTR, &attr); + EXPECT_EQ(!!uc_flic_attr_tests[i].geterrno, !!rc) + TH_LOG("get dev attr rc not expected on %s %s (%i)", + uc_flic_attr_tests[i].name, + strerror(errno), errno); + if (uc_flic_attr_tests[i].geterrno) + EXPECT_EQ(uc_flic_attr_tests[i].geterrno, errno) + TH_LOG("get dev attr errno not expected on %s %s (%i)", + uc_flic_attr_tests[i].name, + strerror(errno), errno); + + rc = ioctl(cd.fd, KVM_SET_DEVICE_ATTR, &attr); + EXPECT_EQ(!!uc_flic_attr_tests[i].seterrno, !!rc) + TH_LOG("set sev attr rc not expected on %s %s (%i)", + uc_flic_attr_tests[i].name, + strerror(errno), errno); + if (uc_flic_attr_tests[i].seterrno) + EXPECT_EQ(uc_flic_attr_tests[i].seterrno, errno) + TH_LOG("set dev attr errno not expected on %s %s (%i)", + uc_flic_attr_tests[i].name, + strerror(errno), errno); + } + + close(cd.fd); +} + +TEST_F(uc_kvm, uc_set_gsi_routing) +{ + struct kvm_irq_routing *routing = kvm_gsi_routing_create(); + struct kvm_irq_routing_entry ue = { + .type = KVM_IRQ_ROUTING_S390_ADAPTER, + .gsi = 1, + .u.adapter = (struct kvm_irq_routing_s390_adapter) { + .ind_addr = 0, + }, + }; + int rc; + + routing->entries[0] = ue; + routing->nr = 1; + rc = ioctl(self->vm_fd, KVM_SET_GSI_ROUTING, routing); + ASSERT_EQ(-1, rc) TH_LOG("err %s (%i)", strerror(errno), errno); + ASSERT_EQ(EINVAL, errno) TH_LOG("err %s (%i)", strerror(errno), errno); +} + TEST_HARNESS_MAIN diff --git a/tools/testing/selftests/memfd/memfd_test.c b/tools/testing/selftests/memfd/memfd_test.c index 0a0b55516028..c0c53451a16d 100644 --- a/tools/testing/selftests/memfd/memfd_test.c +++ b/tools/testing/selftests/memfd/memfd_test.c @@ -282,6 +282,24 @@ static void *mfd_assert_mmap_shared(int fd) return p; } +static void *mfd_assert_mmap_read_shared(int fd) +{ + void *p; + + p = mmap(NULL, + mfd_def_size, + PROT_READ, + MAP_SHARED, + fd, + 0); + if (p == MAP_FAILED) { + printf("mmap() failed: %m\n"); + abort(); + } + + return p; +} + static void *mfd_assert_mmap_private(int fd) { void *p; @@ -980,6 +998,30 @@ static void test_seal_future_write(void) close(fd); } +static void test_seal_write_map_read_shared(void) +{ + int fd; + void *p; + + printf("%s SEAL-WRITE-MAP-READ\n", memfd_str); + + fd = mfd_assert_new("kern_memfd_seal_write_map_read", + mfd_def_size, + MFD_CLOEXEC | MFD_ALLOW_SEALING); + + mfd_assert_add_seals(fd, F_SEAL_WRITE); + mfd_assert_has_seals(fd, F_SEAL_WRITE); + + p = mfd_assert_mmap_read_shared(fd); + + mfd_assert_read(fd); + mfd_assert_read_shared(fd); + mfd_fail_write(fd); + + munmap(p, mfd_def_size); + close(fd); +} + /* * Test SEAL_SHRINK * Test whether SEAL_SHRINK actually prevents shrinking @@ -1593,6 +1635,7 @@ int main(int argc, char **argv) test_seal_write(); test_seal_future_write(); + test_seal_write_map_read_shared(); test_seal_shrink(); test_seal_grow(); test_seal_resize(); diff --git a/tools/testing/selftests/mm/cow.c b/tools/testing/selftests/mm/cow.c index 32c6ccc2a6be..1238e1c5aae1 100644 --- a/tools/testing/selftests/mm/cow.c +++ b/tools/testing/selftests/mm/cow.c @@ -758,7 +758,7 @@ static void do_run_with_base_page(test_fn fn, bool swapout) } /* Populate a base page. */ - memset(mem, 0, pagesize); + memset(mem, 1, pagesize); if (swapout) { madvise(mem, pagesize, MADV_PAGEOUT); @@ -824,12 +824,12 @@ static void do_run_with_thp(test_fn fn, enum thp_run thp_run, size_t thpsize) * Try to populate a THP. Touch the first sub-page and test if * we get the last sub-page populated automatically. */ - mem[0] = 0; + mem[0] = 1; if (!pagemap_is_populated(pagemap_fd, mem + thpsize - pagesize)) { ksft_test_result_skip("Did not get a THP populated\n"); goto munmap; } - memset(mem, 0, thpsize); + memset(mem, 1, thpsize); size = thpsize; switch (thp_run) { @@ -1012,7 +1012,7 @@ static void run_with_hugetlb(test_fn fn, const char *desc, size_t hugetlbsize) } /* Populate an huge page. */ - memset(mem, 0, hugetlbsize); + memset(mem, 1, hugetlbsize); /* * We need a total of two hugetlb pages to handle COW/unsharing diff --git a/tools/testing/selftests/net/forwarding/local_termination.sh b/tools/testing/selftests/net/forwarding/local_termination.sh index c35548767756..ecd34f364125 100755 --- a/tools/testing/selftests/net/forwarding/local_termination.sh +++ b/tools/testing/selftests/net/forwarding/local_termination.sh @@ -7,7 +7,6 @@ ALL_TESTS="standalone vlan_unaware_bridge vlan_aware_bridge test_vlan \ NUM_NETIFS=2 PING_COUNT=1 REQUIRE_MTOOLS=yes -REQUIRE_MZ=no source lib.sh diff --git a/tools/testing/selftests/net/mptcp/mptcp_connect.c b/tools/testing/selftests/net/mptcp/mptcp_connect.c index 4209b9569039..414addef9a45 100644 --- a/tools/testing/selftests/net/mptcp/mptcp_connect.c +++ b/tools/testing/selftests/net/mptcp/mptcp_connect.c @@ -25,6 +25,8 @@ #include <sys/types.h> #include <sys/mman.h> +#include <arpa/inet.h> + #include <netdb.h> #include <netinet/in.h> @@ -1211,23 +1213,42 @@ static void parse_setsock_options(const char *name) exit(1); } -void xdisconnect(int fd, int addrlen) +void xdisconnect(int fd) { - struct sockaddr_storage empty; + socklen_t addrlen = sizeof(struct sockaddr_storage); + struct sockaddr_storage addr, empty; int msec_sleep = 10; - int queued = 1; - int i; + void *raw_addr; + int i, cmdlen; + char cmd[128]; + + /* get the local address and convert it to string */ + if (getsockname(fd, (struct sockaddr *)&addr, &addrlen) < 0) + xerror("getsockname"); + + if (addr.ss_family == AF_INET) + raw_addr = &(((struct sockaddr_in *)&addr)->sin_addr); + else if (addr.ss_family == AF_INET6) + raw_addr = &(((struct sockaddr_in6 *)&addr)->sin6_addr); + else + xerror("bad family"); + + strcpy(cmd, "ss -M | grep -q "); + cmdlen = strlen(cmd); + if (!inet_ntop(addr.ss_family, raw_addr, &cmd[cmdlen], + sizeof(cmd) - cmdlen)) + xerror("inet_ntop"); shutdown(fd, SHUT_WR); - /* while until the pending data is completely flushed, the later + /* + * wait until the pending data is completely flushed and all + * the MPTCP sockets reached the closed status. * disconnect will bypass/ignore/drop any pending data. */ for (i = 0; ; i += msec_sleep) { - if (ioctl(fd, SIOCOUTQ, &queued) < 0) - xerror("can't query out socket queue: %d", errno); - - if (!queued) + /* closed socket are not listed by 'ss' */ + if (system(cmd) != 0) break; if (i > poll_timeout) @@ -1281,9 +1302,9 @@ again: return ret; if (cfg_truncate > 0) { - xdisconnect(fd, peer->ai_addrlen); + xdisconnect(fd); } else if (--cfg_repeat > 0) { - xdisconnect(fd, peer->ai_addrlen); + xdisconnect(fd); /* the socket could be unblocking at this point, we need the * connect to be blocking diff --git a/tools/testing/selftests/pid_namespace/.gitignore b/tools/testing/selftests/pid_namespace/.gitignore index 93ab9d7e5b7e..5118f0f3edf4 100644 --- a/tools/testing/selftests/pid_namespace/.gitignore +++ b/tools/testing/selftests/pid_namespace/.gitignore @@ -1 +1,2 @@ +pid_max regression_enomem diff --git a/tools/testing/selftests/pid_namespace/Makefile b/tools/testing/selftests/pid_namespace/Makefile index 9286a1d22cd3..b972f55d07ae 100644 --- a/tools/testing/selftests/pid_namespace/Makefile +++ b/tools/testing/selftests/pid_namespace/Makefile @@ -1,7 +1,7 @@ # SPDX-License-Identifier: GPL-2.0 CFLAGS += -g $(KHDR_INCLUDES) -TEST_GEN_PROGS = regression_enomem +TEST_GEN_PROGS = regression_enomem pid_max LOCAL_HDRS += $(selfdir)/pidfd/pidfd.h diff --git a/tools/testing/selftests/pid_namespace/pid_max.c b/tools/testing/selftests/pid_namespace/pid_max.c new file mode 100644 index 000000000000..51c414faabb0 --- /dev/null +++ b/tools/testing/selftests/pid_namespace/pid_max.c @@ -0,0 +1,358 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#define _GNU_SOURCE +#include <assert.h> +#include <errno.h> +#include <fcntl.h> +#include <linux/types.h> +#include <sched.h> +#include <signal.h> +#include <stdio.h> +#include <stdlib.h> +#include <string.h> +#include <syscall.h> +#include <sys/wait.h> + +#include "../kselftest_harness.h" +#include "../pidfd/pidfd.h" + +#define __STACK_SIZE (8 * 1024 * 1024) +static pid_t do_clone(int (*fn)(void *), void *arg, int flags) +{ + char *stack; + pid_t ret; + + stack = malloc(__STACK_SIZE); + if (!stack) + return -ENOMEM; + +#ifdef __ia64__ + ret = __clone2(fn, stack, __STACK_SIZE, flags | SIGCHLD, arg); +#else + ret = clone(fn, stack + __STACK_SIZE, flags | SIGCHLD, arg); +#endif + free(stack); + return ret; +} + +static int pid_max_cb(void *data) +{ + int fd, ret; + pid_t pid; + + ret = mount("", "/", NULL, MS_PRIVATE | MS_REC, 0); + if (ret) { + fprintf(stderr, "%m - Failed to make rootfs private mount\n"); + return -1; + } + + umount2("/proc", MNT_DETACH); + + ret = mount("proc", "/proc", "proc", 0, NULL); + if (ret) { + fprintf(stderr, "%m - Failed to mount proc\n"); + return -1; + } + + fd = open("/proc/sys/kernel/pid_max", O_RDWR | O_CLOEXEC | O_NOCTTY); + if (fd < 0) { + fprintf(stderr, "%m - Failed to open pid_max\n"); + return -1; + } + + ret = write(fd, "500", sizeof("500") - 1); + if (ret < 0) { + fprintf(stderr, "%m - Failed to write pid_max\n"); + return -1; + } + + for (int i = 0; i < 501; i++) { + pid = fork(); + if (pid == 0) + exit(EXIT_SUCCESS); + wait_for_pid(pid); + if (pid > 500) { + fprintf(stderr, "Managed to create pid number beyond limit\n"); + return -1; + } + } + + return 0; +} + +static int pid_max_nested_inner(void *data) +{ + int fret = -1; + pid_t pids[2]; + int fd, i, ret; + + ret = mount("", "/", NULL, MS_PRIVATE | MS_REC, 0); + if (ret) { + fprintf(stderr, "%m - Failed to make rootfs private mount\n"); + return fret; + } + + umount2("/proc", MNT_DETACH); + + ret = mount("proc", "/proc", "proc", 0, NULL); + if (ret) { + fprintf(stderr, "%m - Failed to mount proc\n"); + return fret; + } + + fd = open("/proc/sys/kernel/pid_max", O_RDWR | O_CLOEXEC | O_NOCTTY); + if (fd < 0) { + fprintf(stderr, "%m - Failed to open pid_max\n"); + return fret; + } + + ret = write(fd, "500", sizeof("500") - 1); + close(fd); + if (ret < 0) { + fprintf(stderr, "%m - Failed to write pid_max\n"); + return fret; + } + + pids[0] = fork(); + if (pids[0] < 0) { + fprintf(stderr, "Failed to create first new process\n"); + return fret; + } + + if (pids[0] == 0) + exit(EXIT_SUCCESS); + + pids[1] = fork(); + wait_for_pid(pids[0]); + if (pids[1] >= 0) { + if (pids[1] == 0) + exit(EXIT_SUCCESS); + wait_for_pid(pids[1]); + + fprintf(stderr, "Managed to create process even though ancestor pid namespace had a limit\n"); + return fret; + } + + /* Now make sure that we wrap pids at 400. */ + for (i = 0; i < 510; i++) { + pid_t pid; + + pid = fork(); + if (pid < 0) + return fret; + + if (pid == 0) + exit(EXIT_SUCCESS); + + wait_for_pid(pid); + if (pid >= 500) { + fprintf(stderr, "Managed to create process with pid %d beyond configured limit\n", pid); + return fret; + } + } + + return 0; +} + +static int pid_max_nested_outer(void *data) +{ + int fret = -1, nr_procs = 400; + pid_t pids[1000]; + int fd, i, ret; + pid_t pid; + + ret = mount("", "/", NULL, MS_PRIVATE | MS_REC, 0); + if (ret) { + fprintf(stderr, "%m - Failed to make rootfs private mount\n"); + return fret; + } + + umount2("/proc", MNT_DETACH); + + ret = mount("proc", "/proc", "proc", 0, NULL); + if (ret) { + fprintf(stderr, "%m - Failed to mount proc\n"); + return fret; + } + + fd = open("/proc/sys/kernel/pid_max", O_RDWR | O_CLOEXEC | O_NOCTTY); + if (fd < 0) { + fprintf(stderr, "%m - Failed to open pid_max\n"); + return fret; + } + + ret = write(fd, "400", sizeof("400") - 1); + close(fd); + if (ret < 0) { + fprintf(stderr, "%m - Failed to write pid_max\n"); + return fret; + } + + /* + * Create 397 processes. This leaves room for do_clone() (398) and + * one more 399. So creating another process needs to fail. + */ + for (nr_procs = 0; nr_procs < 396; nr_procs++) { + pid = fork(); + if (pid < 0) + goto reap; + + if (pid == 0) + exit(EXIT_SUCCESS); + + pids[nr_procs] = pid; + } + + pid = do_clone(pid_max_nested_inner, NULL, CLONE_NEWPID | CLONE_NEWNS); + if (pid < 0) { + fprintf(stderr, "%m - Failed to clone nested pidns\n"); + goto reap; + } + + if (wait_for_pid(pid)) { + fprintf(stderr, "%m - Nested pid_max failed\n"); + goto reap; + } + + fret = 0; + +reap: + for (int i = 0; i < nr_procs; i++) + wait_for_pid(pids[i]); + + return fret; +} + +static int pid_max_nested_limit_inner(void *data) +{ + int fret = -1, nr_procs = 400; + int fd, ret; + pid_t pid; + pid_t pids[1000]; + + ret = mount("", "/", NULL, MS_PRIVATE | MS_REC, 0); + if (ret) { + fprintf(stderr, "%m - Failed to make rootfs private mount\n"); + return fret; + } + + umount2("/proc", MNT_DETACH); + + ret = mount("proc", "/proc", "proc", 0, NULL); + if (ret) { + fprintf(stderr, "%m - Failed to mount proc\n"); + return fret; + } + + fd = open("/proc/sys/kernel/pid_max", O_RDWR | O_CLOEXEC | O_NOCTTY); + if (fd < 0) { + fprintf(stderr, "%m - Failed to open pid_max\n"); + return fret; + } + + ret = write(fd, "500", sizeof("500") - 1); + close(fd); + if (ret < 0) { + fprintf(stderr, "%m - Failed to write pid_max\n"); + return fret; + } + + for (nr_procs = 0; nr_procs < 500; nr_procs++) { + pid = fork(); + if (pid < 0) + break; + + if (pid == 0) + exit(EXIT_SUCCESS); + + pids[nr_procs] = pid; + } + + if (nr_procs >= 400) { + fprintf(stderr, "Managed to create processes beyond the configured outer limit\n"); + goto reap; + } + + fret = 0; + +reap: + for (int i = 0; i < nr_procs; i++) + wait_for_pid(pids[i]); + + return fret; +} + +static int pid_max_nested_limit_outer(void *data) +{ + int fd, ret; + pid_t pid; + + ret = mount("", "/", NULL, MS_PRIVATE | MS_REC, 0); + if (ret) { + fprintf(stderr, "%m - Failed to make rootfs private mount\n"); + return -1; + } + + umount2("/proc", MNT_DETACH); + + ret = mount("proc", "/proc", "proc", 0, NULL); + if (ret) { + fprintf(stderr, "%m - Failed to mount proc\n"); + return -1; + } + + fd = open("/proc/sys/kernel/pid_max", O_RDWR | O_CLOEXEC | O_NOCTTY); + if (fd < 0) { + fprintf(stderr, "%m - Failed to open pid_max\n"); + return -1; + } + + ret = write(fd, "400", sizeof("400") - 1); + close(fd); + if (ret < 0) { + fprintf(stderr, "%m - Failed to write pid_max\n"); + return -1; + } + + pid = do_clone(pid_max_nested_limit_inner, NULL, CLONE_NEWPID | CLONE_NEWNS); + if (pid < 0) { + fprintf(stderr, "%m - Failed to clone nested pidns\n"); + return -1; + } + + if (wait_for_pid(pid)) { + fprintf(stderr, "%m - Nested pid_max failed\n"); + return -1; + } + + return 0; +} + +TEST(pid_max_simple) +{ + pid_t pid; + + + pid = do_clone(pid_max_cb, NULL, CLONE_NEWPID | CLONE_NEWNS); + ASSERT_GT(pid, 0); + ASSERT_EQ(0, wait_for_pid(pid)); +} + +TEST(pid_max_nested_limit) +{ + pid_t pid; + + pid = do_clone(pid_max_nested_limit_outer, NULL, CLONE_NEWPID | CLONE_NEWNS); + ASSERT_GT(pid, 0); + ASSERT_EQ(0, wait_for_pid(pid)); +} + +TEST(pid_max_nested) +{ + pid_t pid; + + pid = do_clone(pid_max_nested_outer, NULL, CLONE_NEWPID | CLONE_NEWNS); + ASSERT_GT(pid, 0); + ASSERT_EQ(0, wait_for_pid(pid)); +} + +TEST_HARNESS_MAIN diff --git a/tools/testing/selftests/pidfd/.gitignore b/tools/testing/selftests/pidfd/.gitignore index 973198a3ec3d..bf92481f925c 100644 --- a/tools/testing/selftests/pidfd/.gitignore +++ b/tools/testing/selftests/pidfd/.gitignore @@ -6,3 +6,5 @@ pidfd_wait pidfd_fdinfo_test pidfd_getfd_test pidfd_setns_test +pidfd_file_handle_test +pidfd_bind_mount diff --git a/tools/testing/selftests/pidfd/Makefile b/tools/testing/selftests/pidfd/Makefile index d731e3e76d5b..301343a11b62 100644 --- a/tools/testing/selftests/pidfd/Makefile +++ b/tools/testing/selftests/pidfd/Makefile @@ -2,7 +2,8 @@ CFLAGS += -g $(KHDR_INCLUDES) -pthread -Wall TEST_GEN_PROGS := pidfd_test pidfd_fdinfo_test pidfd_open_test \ - pidfd_poll_test pidfd_wait pidfd_getfd_test pidfd_setns_test + pidfd_poll_test pidfd_wait pidfd_getfd_test pidfd_setns_test \ + pidfd_file_handle_test pidfd_bind_mount include ../lib.mk diff --git a/tools/testing/selftests/pidfd/pidfd.h b/tools/testing/selftests/pidfd/pidfd.h index 88d6830ee004..0b96ac4b8ce5 100644 --- a/tools/testing/selftests/pidfd/pidfd.h +++ b/tools/testing/selftests/pidfd/pidfd.h @@ -12,11 +12,11 @@ #include <stdlib.h> #include <string.h> #include <syscall.h> -#include <sys/mount.h> #include <sys/types.h> #include <sys/wait.h> #include "../kselftest.h" +#include "../clone3/clone3_selftests.h" #ifndef P_PIDFD #define P_PIDFD 3 @@ -68,6 +68,11 @@ #define PIDFD_SKIP 3 #define PIDFD_XFAIL 4 +static inline int sys_waitid(int which, pid_t pid, siginfo_t *info, int options) +{ + return syscall(__NR_waitid, which, pid, info, options, NULL); +} + static inline int wait_for_pid(pid_t pid) { int status, ret; @@ -114,4 +119,37 @@ static inline int sys_memfd_create(const char *name, unsigned int flags) return syscall(__NR_memfd_create, name, flags); } +static inline pid_t create_child(int *pidfd, unsigned flags) +{ + struct __clone_args args = { + .flags = CLONE_PIDFD | flags, + .exit_signal = SIGCHLD, + .pidfd = ptr_to_u64(pidfd), + }; + + return sys_clone3(&args, sizeof(struct __clone_args)); +} + +static inline ssize_t read_nointr(int fd, void *buf, size_t count) +{ + ssize_t ret; + + do { + ret = read(fd, buf, count); + } while (ret < 0 && errno == EINTR); + + return ret; +} + +static inline ssize_t write_nointr(int fd, const void *buf, size_t count) +{ + ssize_t ret; + + do { + ret = write(fd, buf, count); + } while (ret < 0 && errno == EINTR); + + return ret; +} + #endif /* __PIDFD_H */ diff --git a/tools/testing/selftests/pidfd/pidfd_bind_mount.c b/tools/testing/selftests/pidfd/pidfd_bind_mount.c new file mode 100644 index 000000000000..7822dd080258 --- /dev/null +++ b/tools/testing/selftests/pidfd/pidfd_bind_mount.c @@ -0,0 +1,188 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +// Copyright (c) 2024 Christian Brauner <brauner@kernel.org> + +#define _GNU_SOURCE +#include <fcntl.h> +#include <limits.h> +#include <sched.h> +#include <stdio.h> +#include <string.h> +#include <linux/fs.h> +#include <sys/ioctl.h> +#include <sys/stat.h> +#include <sys/mount.h> +#include <unistd.h> + +#include "pidfd.h" +#include "../kselftest_harness.h" + +#ifndef __NR_open_tree + #if defined __alpha__ + #define __NR_open_tree 538 + #elif defined _MIPS_SIM + #if _MIPS_SIM == _MIPS_SIM_ABI32 /* o32 */ + #define __NR_open_tree 4428 + #endif + #if _MIPS_SIM == _MIPS_SIM_NABI32 /* n32 */ + #define __NR_open_tree 6428 + #endif + #if _MIPS_SIM == _MIPS_SIM_ABI64 /* n64 */ + #define __NR_open_tree 5428 + #endif + #elif defined __ia64__ + #define __NR_open_tree (428 + 1024) + #else + #define __NR_open_tree 428 + #endif +#endif + +#ifndef __NR_move_mount + #if defined __alpha__ + #define __NR_move_mount 539 + #elif defined _MIPS_SIM + #if _MIPS_SIM == _MIPS_SIM_ABI32 /* o32 */ + #define __NR_move_mount 4429 + #endif + #if _MIPS_SIM == _MIPS_SIM_NABI32 /* n32 */ + #define __NR_move_mount 6429 + #endif + #if _MIPS_SIM == _MIPS_SIM_ABI64 /* n64 */ + #define __NR_move_mount 5429 + #endif + #elif defined __ia64__ + #define __NR_move_mount (428 + 1024) + #else + #define __NR_move_mount 429 + #endif +#endif + +#ifndef MOVE_MOUNT_F_EMPTY_PATH +#define MOVE_MOUNT_F_EMPTY_PATH 0x00000004 /* Empty from path permitted */ +#endif + +#ifndef MOVE_MOUNT_F_EMPTY_PATH +#define MOVE_MOUNT_T_EMPTY_PATH 0x00000040 /* Empty to path permitted */ +#endif + +static inline int sys_move_mount(int from_dfd, const char *from_pathname, + int to_dfd, const char *to_pathname, + unsigned int flags) +{ + return syscall(__NR_move_mount, from_dfd, from_pathname, to_dfd, + to_pathname, flags); +} + +#ifndef OPEN_TREE_CLONE +#define OPEN_TREE_CLONE 1 +#endif + +#ifndef OPEN_TREE_CLOEXEC +#define OPEN_TREE_CLOEXEC O_CLOEXEC +#endif + +#ifndef AT_RECURSIVE +#define AT_RECURSIVE 0x8000 /* Apply to the entire subtree */ +#endif + +static inline int sys_open_tree(int dfd, const char *filename, unsigned int flags) +{ + return syscall(__NR_open_tree, dfd, filename, flags); +} + +FIXTURE(pidfd_bind_mount) { + char template[PATH_MAX]; + int fd_tmp; + int pidfd; + struct stat st1; + struct stat st2; + __u32 gen1; + __u32 gen2; + bool must_unmount; +}; + +FIXTURE_SETUP(pidfd_bind_mount) +{ + self->fd_tmp = -EBADF; + self->must_unmount = false; + ASSERT_EQ(unshare(CLONE_NEWNS), 0); + ASSERT_LE(snprintf(self->template, PATH_MAX, "%s", P_tmpdir "/pidfd_bind_mount_XXXXXX"), PATH_MAX); + self->fd_tmp = mkstemp(self->template); + ASSERT_GE(self->fd_tmp, 0); + self->pidfd = sys_pidfd_open(getpid(), 0); + ASSERT_GE(self->pidfd, 0); + ASSERT_GE(fstat(self->pidfd, &self->st1), 0); + ASSERT_EQ(ioctl(self->pidfd, FS_IOC_GETVERSION, &self->gen1), 0); +} + +FIXTURE_TEARDOWN(pidfd_bind_mount) +{ + ASSERT_EQ(close(self->fd_tmp), 0); + if (self->must_unmount) + ASSERT_EQ(umount2(self->template, 0), 0); + ASSERT_EQ(unlink(self->template), 0); +} + +/* + * Test that a detached mount can be created for a pidfd and then + * attached to the filesystem hierarchy. + */ +TEST_F(pidfd_bind_mount, bind_mount) +{ + int fd_tree; + + fd_tree = sys_open_tree(self->pidfd, "", OPEN_TREE_CLONE | OPEN_TREE_CLOEXEC | AT_EMPTY_PATH); + ASSERT_GE(fd_tree, 0); + + ASSERT_EQ(move_mount(fd_tree, "", self->fd_tmp, "", MOVE_MOUNT_F_EMPTY_PATH | MOVE_MOUNT_T_EMPTY_PATH), 0); + self->must_unmount = true; + + ASSERT_EQ(close(fd_tree), 0); +} + +/* Test that a pidfd can be reopened through procfs. */ +TEST_F(pidfd_bind_mount, reopen) +{ + int pidfd; + char proc_path[PATH_MAX]; + + sprintf(proc_path, "/proc/self/fd/%d", self->pidfd); + pidfd = open(proc_path, O_RDONLY | O_NOCTTY | O_CLOEXEC); + ASSERT_GE(pidfd, 0); + + ASSERT_GE(fstat(self->pidfd, &self->st2), 0); + ASSERT_EQ(ioctl(self->pidfd, FS_IOC_GETVERSION, &self->gen2), 0); + + ASSERT_TRUE(self->st1.st_dev == self->st2.st_dev && self->st1.st_ino == self->st2.st_ino); + ASSERT_TRUE(self->gen1 == self->gen2); + + ASSERT_EQ(close(pidfd), 0); +} + +/* + * Test that a detached mount can be created for a pidfd and then + * attached to the filesystem hierarchy and reopened. + */ +TEST_F(pidfd_bind_mount, bind_mount_reopen) +{ + int fd_tree, fd_pidfd_mnt; + + fd_tree = sys_open_tree(self->pidfd, "", OPEN_TREE_CLONE | OPEN_TREE_CLOEXEC | AT_EMPTY_PATH); + ASSERT_GE(fd_tree, 0); + + ASSERT_EQ(move_mount(fd_tree, "", self->fd_tmp, "", MOVE_MOUNT_F_EMPTY_PATH | MOVE_MOUNT_T_EMPTY_PATH), 0); + self->must_unmount = true; + + fd_pidfd_mnt = openat(-EBADF, self->template, O_RDONLY | O_NOCTTY | O_CLOEXEC); + ASSERT_GE(fd_pidfd_mnt, 0); + + ASSERT_GE(fstat(fd_tree, &self->st2), 0); + ASSERT_EQ(ioctl(fd_pidfd_mnt, FS_IOC_GETVERSION, &self->gen2), 0); + + ASSERT_TRUE(self->st1.st_dev == self->st2.st_dev && self->st1.st_ino == self->st2.st_ino); + ASSERT_TRUE(self->gen1 == self->gen2); + + ASSERT_EQ(close(fd_tree), 0); + ASSERT_EQ(close(fd_pidfd_mnt), 0); +} + +TEST_HARNESS_MAIN diff --git a/tools/testing/selftests/pidfd/pidfd_file_handle_test.c b/tools/testing/selftests/pidfd/pidfd_file_handle_test.c new file mode 100644 index 000000000000..439b9c6c0457 --- /dev/null +++ b/tools/testing/selftests/pidfd/pidfd_file_handle_test.c @@ -0,0 +1,503 @@ +// SPDX-License-Identifier: GPL-2.0 + +#define _GNU_SOURCE +#include <errno.h> +#include <fcntl.h> +#include <limits.h> +#include <linux/types.h> +#include <poll.h> +#include <sched.h> +#include <signal.h> +#include <stdio.h> +#include <stdlib.h> +#include <string.h> +#include <syscall.h> +#include <sys/prctl.h> +#include <sys/wait.h> +#include <unistd.h> +#include <sys/socket.h> +#include <linux/kcmp.h> +#include <sys/stat.h> + +#include "pidfd.h" +#include "../kselftest_harness.h" + +FIXTURE(file_handle) +{ + pid_t pid; + int pidfd; + + pid_t child_pid1; + int child_pidfd1; + + pid_t child_pid2; + int child_pidfd2; + + pid_t child_pid3; + int child_pidfd3; +}; + +FIXTURE_SETUP(file_handle) +{ + int ret; + int ipc_sockets[2]; + char c; + + self->pid = getpid(); + self->pidfd = sys_pidfd_open(self->pid, 0); + ASSERT_GE(self->pidfd, 0); + + ret = socketpair(AF_LOCAL, SOCK_STREAM | SOCK_CLOEXEC, 0, ipc_sockets); + EXPECT_EQ(ret, 0); + + self->child_pid1 = create_child(&self->child_pidfd1, CLONE_NEWUSER); + EXPECT_GE(self->child_pid1, 0); + + if (self->child_pid1 == 0) { + close(ipc_sockets[0]); + + if (write_nointr(ipc_sockets[1], "1", 1) < 0) + _exit(EXIT_FAILURE); + + close(ipc_sockets[1]); + + pause(); + _exit(EXIT_SUCCESS); + } + + close(ipc_sockets[1]); + ASSERT_EQ(read_nointr(ipc_sockets[0], &c, 1), 1); + close(ipc_sockets[0]); + + ret = socketpair(AF_LOCAL, SOCK_STREAM | SOCK_CLOEXEC, 0, ipc_sockets); + EXPECT_EQ(ret, 0); + + self->child_pid2 = create_child(&self->child_pidfd2, CLONE_NEWUSER | CLONE_NEWPID); + EXPECT_GE(self->child_pid2, 0); + + if (self->child_pid2 == 0) { + close(ipc_sockets[0]); + + if (write_nointr(ipc_sockets[1], "1", 1) < 0) + _exit(EXIT_FAILURE); + + close(ipc_sockets[1]); + + pause(); + _exit(EXIT_SUCCESS); + } + + close(ipc_sockets[1]); + ASSERT_EQ(read_nointr(ipc_sockets[0], &c, 1), 1); + close(ipc_sockets[0]); + + ret = socketpair(AF_LOCAL, SOCK_STREAM | SOCK_CLOEXEC, 0, ipc_sockets); + EXPECT_EQ(ret, 0); + + self->child_pid3 = create_child(&self->child_pidfd3, CLONE_NEWUSER | CLONE_NEWPID); + EXPECT_GE(self->child_pid3, 0); + + if (self->child_pid3 == 0) { + close(ipc_sockets[0]); + + if (write_nointr(ipc_sockets[1], "1", 1) < 0) + _exit(EXIT_FAILURE); + + close(ipc_sockets[1]); + + pause(); + _exit(EXIT_SUCCESS); + } + + close(ipc_sockets[1]); + ASSERT_EQ(read_nointr(ipc_sockets[0], &c, 1), 1); + close(ipc_sockets[0]); +} + +FIXTURE_TEARDOWN(file_handle) +{ + EXPECT_EQ(close(self->pidfd), 0); + + EXPECT_EQ(sys_pidfd_send_signal(self->child_pidfd1, SIGKILL, NULL, 0), 0); + if (self->child_pidfd1 >= 0) + EXPECT_EQ(0, close(self->child_pidfd1)); + + EXPECT_EQ(sys_waitid(P_PID, self->child_pid1, NULL, WEXITED), 0); + + EXPECT_EQ(sys_pidfd_send_signal(self->child_pidfd2, SIGKILL, NULL, 0), 0); + if (self->child_pidfd2 >= 0) + EXPECT_EQ(0, close(self->child_pidfd2)); + + EXPECT_EQ(sys_waitid(P_PID, self->child_pid2, NULL, WEXITED), 0); + + if (self->child_pidfd3 >= 0) { + EXPECT_EQ(sys_pidfd_send_signal(self->child_pidfd3, SIGKILL, NULL, 0), 0); + EXPECT_EQ(0, close(self->child_pidfd3)); + EXPECT_EQ(sys_waitid(P_PID, self->child_pid3, NULL, WEXITED), 0); + } +} + +/* + * Test that we can decode a pidfs file handle in the same pid + * namespace. + */ +TEST_F(file_handle, file_handle_same_pidns) +{ + int mnt_id; + struct file_handle *fh; + int pidfd = -EBADF; + struct stat st1, st2; + + fh = malloc(sizeof(struct file_handle) + MAX_HANDLE_SZ); + ASSERT_NE(fh, NULL); + memset(fh, 0, sizeof(struct file_handle) + MAX_HANDLE_SZ); + fh->handle_bytes = MAX_HANDLE_SZ; + + ASSERT_EQ(name_to_handle_at(self->child_pidfd1, "", fh, &mnt_id, AT_EMPTY_PATH), 0); + + ASSERT_EQ(fstat(self->child_pidfd1, &st1), 0); + + pidfd = open_by_handle_at(self->pidfd, fh, 0); + ASSERT_GE(pidfd, 0); + + ASSERT_EQ(fstat(pidfd, &st2), 0); + ASSERT_TRUE(st1.st_dev == st2.st_dev && st1.st_ino == st2.st_ino); + + ASSERT_EQ(close(pidfd), 0); + + pidfd = open_by_handle_at(self->pidfd, fh, O_CLOEXEC); + ASSERT_GE(pidfd, 0); + + ASSERT_EQ(fstat(pidfd, &st2), 0); + ASSERT_TRUE(st1.st_dev == st2.st_dev && st1.st_ino == st2.st_ino); + + ASSERT_EQ(close(pidfd), 0); + + pidfd = open_by_handle_at(self->pidfd, fh, O_NONBLOCK); + ASSERT_GE(pidfd, 0); + + ASSERT_EQ(fstat(pidfd, &st2), 0); + ASSERT_TRUE(st1.st_dev == st2.st_dev && st1.st_ino == st2.st_ino); + + ASSERT_EQ(close(pidfd), 0); + + free(fh); +} + +/* + * Test that we can decode a pidfs file handle from a child pid + * namespace. + */ +TEST_F(file_handle, file_handle_child_pidns) +{ + int mnt_id; + struct file_handle *fh; + int pidfd = -EBADF; + struct stat st1, st2; + + fh = malloc(sizeof(struct file_handle) + MAX_HANDLE_SZ); + ASSERT_NE(fh, NULL); + memset(fh, 0, sizeof(struct file_handle) + MAX_HANDLE_SZ); + fh->handle_bytes = MAX_HANDLE_SZ; + + ASSERT_EQ(name_to_handle_at(self->child_pidfd2, "", fh, &mnt_id, AT_EMPTY_PATH), 0); + + ASSERT_EQ(fstat(self->child_pidfd2, &st1), 0); + + pidfd = open_by_handle_at(self->pidfd, fh, 0); + ASSERT_GE(pidfd, 0); + + ASSERT_EQ(fstat(pidfd, &st2), 0); + ASSERT_TRUE(st1.st_dev == st2.st_dev && st1.st_ino == st2.st_ino); + + ASSERT_EQ(close(pidfd), 0); + + pidfd = open_by_handle_at(self->pidfd, fh, O_CLOEXEC); + ASSERT_GE(pidfd, 0); + + ASSERT_EQ(fstat(pidfd, &st2), 0); + ASSERT_TRUE(st1.st_dev == st2.st_dev && st1.st_ino == st2.st_ino); + + ASSERT_EQ(close(pidfd), 0); + + pidfd = open_by_handle_at(self->pidfd, fh, O_NONBLOCK); + ASSERT_GE(pidfd, 0); + + ASSERT_EQ(fstat(pidfd, &st2), 0); + ASSERT_TRUE(st1.st_dev == st2.st_dev && st1.st_ino == st2.st_ino); + + ASSERT_EQ(close(pidfd), 0); + + free(fh); +} + +/* + * Test that we fail to decode a pidfs file handle from an ancestor + * child pid namespace. + */ +TEST_F(file_handle, file_handle_foreign_pidns) +{ + int mnt_id; + struct file_handle *fh; + pid_t pid; + + fh = malloc(sizeof(struct file_handle) + MAX_HANDLE_SZ); + ASSERT_NE(fh, NULL); + memset(fh, 0, sizeof(struct file_handle) + MAX_HANDLE_SZ); + fh->handle_bytes = MAX_HANDLE_SZ; + + ASSERT_EQ(name_to_handle_at(self->pidfd, "", fh, &mnt_id, AT_EMPTY_PATH), 0); + + ASSERT_EQ(setns(self->child_pidfd2, CLONE_NEWUSER | CLONE_NEWPID), 0); + + pid = fork(); + ASSERT_GE(pid, 0); + + if (pid == 0) { + int pidfd = open_by_handle_at(self->pidfd, fh, 0); + if (pidfd >= 0) { + TH_LOG("Managed to open pidfd outside of the caller's pid namespace hierarchy"); + _exit(1); + } + _exit(0); + } + + ASSERT_EQ(wait_for_pid(pid), 0); + + free(fh); +} + +/* + * Test that we can decode a pidfs file handle of a process that has + * exited but not been reaped. + */ +TEST_F(file_handle, pid_has_exited) +{ + int mnt_id, pidfd, child_pidfd3; + struct file_handle *fh; + struct stat st1, st2; + + fh = malloc(sizeof(struct file_handle) + MAX_HANDLE_SZ); + ASSERT_NE(fh, NULL); + memset(fh, 0, sizeof(struct file_handle) + MAX_HANDLE_SZ); + fh->handle_bytes = MAX_HANDLE_SZ; + + ASSERT_EQ(name_to_handle_at(self->child_pidfd3, "", fh, &mnt_id, AT_EMPTY_PATH), 0); + + ASSERT_EQ(fstat(self->child_pidfd3, &st1), 0); + + pidfd = open_by_handle_at(self->pidfd, fh, 0); + ASSERT_GE(pidfd, 0); + + ASSERT_EQ(fstat(pidfd, &st2), 0); + ASSERT_TRUE(st1.st_dev == st2.st_dev && st1.st_ino == st2.st_ino); + + ASSERT_EQ(close(pidfd), 0); + + child_pidfd3 = self->child_pidfd3; + self->child_pidfd3 = -EBADF; + EXPECT_EQ(sys_pidfd_send_signal(child_pidfd3, SIGKILL, NULL, 0), 0); + EXPECT_EQ(close(child_pidfd3), 0); + EXPECT_EQ(sys_waitid(P_PID, self->child_pid3, NULL, WEXITED | WNOWAIT), 0); + + pidfd = open_by_handle_at(self->pidfd, fh, 0); + ASSERT_GE(pidfd, 0); + + EXPECT_EQ(sys_waitid(P_PID, self->child_pid3, NULL, WEXITED), 0); +} + +/* + * Test that we fail to decode a pidfs file handle of a process that has + * already been reaped. + */ +TEST_F(file_handle, pid_has_been_reaped) +{ + int mnt_id, pidfd, child_pidfd3; + struct file_handle *fh; + struct stat st1, st2; + + fh = malloc(sizeof(struct file_handle) + MAX_HANDLE_SZ); + ASSERT_NE(fh, NULL); + memset(fh, 0, sizeof(struct file_handle) + MAX_HANDLE_SZ); + fh->handle_bytes = MAX_HANDLE_SZ; + + ASSERT_EQ(name_to_handle_at(self->child_pidfd3, "", fh, &mnt_id, AT_EMPTY_PATH), 0); + + ASSERT_EQ(fstat(self->child_pidfd3, &st1), 0); + + pidfd = open_by_handle_at(self->pidfd, fh, 0); + ASSERT_GE(pidfd, 0); + + ASSERT_EQ(fstat(pidfd, &st2), 0); + ASSERT_TRUE(st1.st_dev == st2.st_dev && st1.st_ino == st2.st_ino); + + ASSERT_EQ(close(pidfd), 0); + + child_pidfd3 = self->child_pidfd3; + self->child_pidfd3 = -EBADF; + EXPECT_EQ(sys_pidfd_send_signal(child_pidfd3, SIGKILL, NULL, 0), 0); + EXPECT_EQ(close(child_pidfd3), 0); + EXPECT_EQ(sys_waitid(P_PID, self->child_pid3, NULL, WEXITED), 0); + + pidfd = open_by_handle_at(self->pidfd, fh, 0); + ASSERT_LT(pidfd, 0); +} + +/* + * Test valid flags to open a pidfd file handle. Note, that + * PIDFD_NONBLOCK is defined as O_NONBLOCK and O_NONBLOCK is an alias to + * O_NDELAY. Also note that PIDFD_THREAD is an alias for O_EXCL. + */ +TEST_F(file_handle, open_by_handle_at_valid_flags) +{ + int mnt_id; + struct file_handle *fh; + int pidfd = -EBADF; + struct stat st1, st2; + + fh = malloc(sizeof(struct file_handle) + MAX_HANDLE_SZ); + ASSERT_NE(fh, NULL); + memset(fh, 0, sizeof(struct file_handle) + MAX_HANDLE_SZ); + fh->handle_bytes = MAX_HANDLE_SZ; + + ASSERT_EQ(name_to_handle_at(self->child_pidfd2, "", fh, &mnt_id, AT_EMPTY_PATH), 0); + + ASSERT_EQ(fstat(self->child_pidfd2, &st1), 0); + + pidfd = open_by_handle_at(self->pidfd, fh, + O_RDONLY | + O_WRONLY | + O_RDWR | + O_NONBLOCK | + O_NDELAY | + O_CLOEXEC | + O_EXCL); + ASSERT_GE(pidfd, 0); + + ASSERT_EQ(fstat(pidfd, &st2), 0); + ASSERT_TRUE(st1.st_dev == st2.st_dev && st1.st_ino == st2.st_ino); + + ASSERT_EQ(close(pidfd), 0); +} + +/* + * Test that invalid flags passed to open a pidfd file handle are + * rejected. + */ +TEST_F(file_handle, open_by_handle_at_invalid_flags) +{ + int mnt_id; + struct file_handle *fh; + int pidfd = -EBADF; + static const struct invalid_pidfs_file_handle_flags { + int oflag; + const char *oflag_name; + } invalid_pidfs_file_handle_flags[] = { + { FASYNC, "FASYNC" }, + { O_CREAT, "O_CREAT" }, + { O_NOCTTY, "O_NOCTTY" }, + { O_CREAT, "O_CREAT" }, + { O_TRUNC, "O_TRUNC" }, + { O_APPEND, "O_APPEND" }, + { O_SYNC, "O_SYNC" }, + { O_DSYNC, "O_DSYNC" }, + { O_DIRECT, "O_DIRECT" }, + { O_DIRECTORY, "O_DIRECTORY" }, + { O_NOFOLLOW, "O_NOFOLLOW" }, + { O_NOATIME, "O_NOATIME" }, + { O_PATH, "O_PATH" }, + { O_TMPFILE, "O_TMPFILE" }, + /* + * O_LARGEFILE is added implicitly by + * open_by_handle_at() so pidfs simply masks it off. + */ + }; + + fh = malloc(sizeof(struct file_handle) + MAX_HANDLE_SZ); + ASSERT_NE(fh, NULL); + memset(fh, 0, sizeof(struct file_handle) + MAX_HANDLE_SZ); + fh->handle_bytes = MAX_HANDLE_SZ; + + ASSERT_EQ(name_to_handle_at(self->child_pidfd2, "", fh, &mnt_id, AT_EMPTY_PATH), 0); + + for (int i = 0; i < ARRAY_SIZE(invalid_pidfs_file_handle_flags); i++) { + pidfd = open_by_handle_at(self->pidfd, fh, invalid_pidfs_file_handle_flags[i].oflag); + ASSERT_LT(pidfd, 0) { + TH_LOG("open_by_handle_at() succeeded with invalid flags: %s", invalid_pidfs_file_handle_flags[i].oflag_name); + } + } +} + +/* Test that lookup fails. */ +TEST_F(file_handle, lookup_must_fail) +{ + int mnt_id; + struct file_handle *fh; + + fh = malloc(sizeof(struct file_handle) + MAX_HANDLE_SZ); + ASSERT_NE(fh, NULL); + memset(fh, 0, sizeof(struct file_handle) + MAX_HANDLE_SZ); + fh->handle_bytes = MAX_HANDLE_SZ; + + ASSERT_NE(name_to_handle_at(self->child_pidfd2, "lookup-is-not-possible-with-pidfs", fh, &mnt_id, AT_EMPTY_PATH), 0); + ASSERT_EQ(errno, ENOTDIR); + ASSERT_NE(name_to_handle_at(self->child_pidfd2, "lookup-is-not-possible-with-pidfs", fh, &mnt_id, 0), 0); + ASSERT_EQ(errno, ENOTDIR); +} + +#ifndef AT_HANDLE_CONNECTABLE +#define AT_HANDLE_CONNECTABLE 0x002 +#endif + +/* + * Test that AT_HANDLE_CONNECTABLE is rejected. Connectable file handles + * don't make sense for pidfs. Note that currently AT_HANDLE_CONNECTABLE + * is rejected because it is incompatible with AT_EMPTY_PATH which is + * required with pidfds as we don't support lookup. + */ +TEST_F(file_handle, invalid_name_to_handle_at_flags) +{ + int mnt_id; + struct file_handle *fh; + + fh = malloc(sizeof(struct file_handle) + MAX_HANDLE_SZ); + ASSERT_NE(fh, NULL); + memset(fh, 0, sizeof(struct file_handle) + MAX_HANDLE_SZ); + fh->handle_bytes = MAX_HANDLE_SZ; + + ASSERT_NE(name_to_handle_at(self->child_pidfd2, "", fh, &mnt_id, AT_EMPTY_PATH | AT_HANDLE_CONNECTABLE), 0); +} + +#ifndef AT_HANDLE_FID +#define AT_HANDLE_FID 0x200 +#endif + +/* + * Test that a request with AT_HANDLE_FID always leads to decodable file + * handle as pidfs always provides export operations. + */ +TEST_F(file_handle, valid_name_to_handle_at_flags) +{ + int mnt_id, pidfd; + struct file_handle *fh; + struct stat st1, st2; + + fh = malloc(sizeof(struct file_handle) + MAX_HANDLE_SZ); + ASSERT_NE(fh, NULL); + memset(fh, 0, sizeof(struct file_handle) + MAX_HANDLE_SZ); + fh->handle_bytes = MAX_HANDLE_SZ; + + ASSERT_EQ(name_to_handle_at(self->child_pidfd2, "", fh, &mnt_id, AT_EMPTY_PATH | AT_HANDLE_FID), 0); + + ASSERT_EQ(fstat(self->child_pidfd2, &st1), 0); + + pidfd = open_by_handle_at(self->pidfd, fh, 0); + ASSERT_GE(pidfd, 0); + + ASSERT_EQ(fstat(pidfd, &st2), 0); + ASSERT_TRUE(st1.st_dev == st2.st_dev && st1.st_ino == st2.st_ino); + + ASSERT_EQ(close(pidfd), 0); +} + +TEST_HARNESS_MAIN diff --git a/tools/testing/selftests/pidfd/pidfd_setns_test.c b/tools/testing/selftests/pidfd/pidfd_setns_test.c index 7c2a4349170a..222f8131283b 100644 --- a/tools/testing/selftests/pidfd/pidfd_setns_test.c +++ b/tools/testing/selftests/pidfd/pidfd_setns_test.c @@ -19,7 +19,6 @@ #include <linux/ioctl.h> #include "pidfd.h" -#include "../clone3/clone3_selftests.h" #include "../kselftest_harness.h" #ifndef PIDFS_IOCTL_MAGIC @@ -118,22 +117,6 @@ FIXTURE(current_nsset) int child_pidfd_derived_nsfds2[PIDFD_NS_MAX]; }; -static int sys_waitid(int which, pid_t pid, int options) -{ - return syscall(__NR_waitid, which, pid, NULL, options, NULL); -} - -pid_t create_child(int *pidfd, unsigned flags) -{ - struct __clone_args args = { - .flags = CLONE_PIDFD | flags, - .exit_signal = SIGCHLD, - .pidfd = ptr_to_u64(pidfd), - }; - - return sys_clone3(&args, sizeof(struct clone_args)); -} - static bool switch_timens(void) { int fd, ret; @@ -150,28 +133,6 @@ static bool switch_timens(void) return ret == 0; } -static ssize_t read_nointr(int fd, void *buf, size_t count) -{ - ssize_t ret; - - do { - ret = read(fd, buf, count); - } while (ret < 0 && errno == EINTR); - - return ret; -} - -static ssize_t write_nointr(int fd, const void *buf, size_t count) -{ - ssize_t ret; - - do { - ret = write(fd, buf, count); - } while (ret < 0 && errno == EINTR); - - return ret; -} - FIXTURE_SETUP(current_nsset) { int i, proc_fd, ret; @@ -229,7 +190,7 @@ FIXTURE_SETUP(current_nsset) _exit(EXIT_SUCCESS); } - ASSERT_EQ(sys_waitid(P_PID, self->child_pid_exited, WEXITED | WNOWAIT), 0); + ASSERT_EQ(sys_waitid(P_PID, self->child_pid_exited, NULL, WEXITED | WNOWAIT), 0); self->pidfd = sys_pidfd_open(self->pid, 0); EXPECT_GE(self->pidfd, 0) { @@ -432,9 +393,9 @@ FIXTURE_TEARDOWN(current_nsset) EXPECT_EQ(0, close(self->child_pidfd1)); if (self->child_pidfd2 >= 0) EXPECT_EQ(0, close(self->child_pidfd2)); - ASSERT_EQ(sys_waitid(P_PID, self->child_pid_exited, WEXITED), 0); - ASSERT_EQ(sys_waitid(P_PID, self->child_pid1, WEXITED), 0); - ASSERT_EQ(sys_waitid(P_PID, self->child_pid2, WEXITED), 0); + ASSERT_EQ(sys_waitid(P_PID, self->child_pid_exited, NULL, WEXITED), 0); + ASSERT_EQ(sys_waitid(P_PID, self->child_pid1, NULL, WEXITED), 0); + ASSERT_EQ(sys_waitid(P_PID, self->child_pid2, NULL, WEXITED), 0); } static int preserve_ns(const int pid, const char *ns) diff --git a/tools/testing/selftests/pidfd/pidfd_wait.c b/tools/testing/selftests/pidfd/pidfd_wait.c index 0dcb8365ddc3..1e2d49751cde 100644 --- a/tools/testing/selftests/pidfd/pidfd_wait.c +++ b/tools/testing/selftests/pidfd/pidfd_wait.c @@ -26,22 +26,11 @@ #define SKIP(s, ...) XFAIL(s, ##__VA_ARGS__) #endif -static pid_t sys_clone3(struct clone_args *args) -{ - return syscall(__NR_clone3, args, sizeof(struct clone_args)); -} - -static int sys_waitid(int which, pid_t pid, siginfo_t *info, int options, - struct rusage *ru) -{ - return syscall(__NR_waitid, which, pid, info, options, ru); -} - TEST(wait_simple) { int pidfd = -1; pid_t parent_tid = -1; - struct clone_args args = { + struct __clone_args args = { .parent_tid = ptr_to_u64(&parent_tid), .pidfd = ptr_to_u64(&pidfd), .flags = CLONE_PIDFD | CLONE_PARENT_SETTID, @@ -55,7 +44,7 @@ TEST(wait_simple) pidfd = open("/proc/self", O_DIRECTORY | O_RDONLY | O_CLOEXEC); ASSERT_GE(pidfd, 0); - pid = sys_waitid(P_PIDFD, pidfd, &info, WEXITED, NULL); + pid = sys_waitid(P_PIDFD, pidfd, &info, WEXITED); ASSERT_NE(pid, 0); EXPECT_EQ(close(pidfd), 0); pidfd = -1; @@ -63,18 +52,18 @@ TEST(wait_simple) pidfd = open("/dev/null", O_RDONLY | O_CLOEXEC); ASSERT_GE(pidfd, 0); - pid = sys_waitid(P_PIDFD, pidfd, &info, WEXITED, NULL); + pid = sys_waitid(P_PIDFD, pidfd, &info, WEXITED); ASSERT_NE(pid, 0); EXPECT_EQ(close(pidfd), 0); pidfd = -1; - pid = sys_clone3(&args); + pid = sys_clone3(&args, sizeof(args)); ASSERT_GE(pid, 0); if (pid == 0) exit(EXIT_SUCCESS); - pid = sys_waitid(P_PIDFD, pidfd, &info, WEXITED, NULL); + pid = sys_waitid(P_PIDFD, pidfd, &info, WEXITED); ASSERT_GE(pid, 0); ASSERT_EQ(WIFEXITED(info.si_status), true); ASSERT_EQ(WEXITSTATUS(info.si_status), 0); @@ -89,7 +78,7 @@ TEST(wait_states) { int pidfd = -1; pid_t parent_tid = -1; - struct clone_args args = { + struct __clone_args args = { .parent_tid = ptr_to_u64(&parent_tid), .pidfd = ptr_to_u64(&pidfd), .flags = CLONE_PIDFD | CLONE_PARENT_SETTID, @@ -102,7 +91,7 @@ TEST(wait_states) }; ASSERT_EQ(pipe(pfd), 0); - pid = sys_clone3(&args); + pid = sys_clone3(&args, sizeof(args)); ASSERT_GE(pid, 0); if (pid == 0) { @@ -117,28 +106,28 @@ TEST(wait_states) } close(pfd[0]); - ASSERT_EQ(sys_waitid(P_PIDFD, pidfd, &info, WSTOPPED, NULL), 0); + ASSERT_EQ(sys_waitid(P_PIDFD, pidfd, &info, WSTOPPED), 0); ASSERT_EQ(info.si_signo, SIGCHLD); ASSERT_EQ(info.si_code, CLD_STOPPED); ASSERT_EQ(info.si_pid, parent_tid); ASSERT_EQ(sys_pidfd_send_signal(pidfd, SIGCONT, NULL, 0), 0); - ASSERT_EQ(sys_waitid(P_PIDFD, pidfd, &info, WCONTINUED, NULL), 0); + ASSERT_EQ(sys_waitid(P_PIDFD, pidfd, &info, WCONTINUED), 0); ASSERT_EQ(write(pfd[1], "C", 1), 1); close(pfd[1]); ASSERT_EQ(info.si_signo, SIGCHLD); ASSERT_EQ(info.si_code, CLD_CONTINUED); ASSERT_EQ(info.si_pid, parent_tid); - ASSERT_EQ(sys_waitid(P_PIDFD, pidfd, &info, WUNTRACED, NULL), 0); + ASSERT_EQ(sys_waitid(P_PIDFD, pidfd, &info, WUNTRACED), 0); ASSERT_EQ(info.si_signo, SIGCHLD); ASSERT_EQ(info.si_code, CLD_STOPPED); ASSERT_EQ(info.si_pid, parent_tid); ASSERT_EQ(sys_pidfd_send_signal(pidfd, SIGKILL, NULL, 0), 0); - ASSERT_EQ(sys_waitid(P_PIDFD, pidfd, &info, WEXITED, NULL), 0); + ASSERT_EQ(sys_waitid(P_PIDFD, pidfd, &info, WEXITED), 0); ASSERT_EQ(info.si_signo, SIGCHLD); ASSERT_EQ(info.si_code, CLD_KILLED); ASSERT_EQ(info.si_pid, parent_tid); @@ -151,7 +140,7 @@ TEST(wait_nonblock) int pidfd; unsigned int flags = 0; pid_t parent_tid = -1; - struct clone_args args = { + struct __clone_args args = { .parent_tid = ptr_to_u64(&parent_tid), .flags = CLONE_PARENT_SETTID, .exit_signal = SIGCHLD, @@ -173,12 +162,12 @@ TEST(wait_nonblock) SKIP(return, "Skipping PIDFD_NONBLOCK test"); } - ret = sys_waitid(P_PIDFD, pidfd, &info, WEXITED, NULL); + ret = sys_waitid(P_PIDFD, pidfd, &info, WEXITED); ASSERT_LT(ret, 0); ASSERT_EQ(errno, ECHILD); EXPECT_EQ(close(pidfd), 0); - pid = sys_clone3(&args); + pid = sys_clone3(&args, sizeof(args)); ASSERT_GE(pid, 0); if (pid == 0) { @@ -201,7 +190,7 @@ TEST(wait_nonblock) * Callers need to see EAGAIN/EWOULDBLOCK with non-blocking pidfd when * child processes exist but none have exited. */ - ret = sys_waitid(P_PIDFD, pidfd, &info, WEXITED, NULL); + ret = sys_waitid(P_PIDFD, pidfd, &info, WEXITED); ASSERT_LT(ret, 0); ASSERT_EQ(errno, EAGAIN); @@ -210,19 +199,19 @@ TEST(wait_nonblock) * WNOHANG raised explicitly when child processes exist but none have * exited. */ - ret = sys_waitid(P_PIDFD, pidfd, &info, WEXITED | WNOHANG, NULL); + ret = sys_waitid(P_PIDFD, pidfd, &info, WEXITED | WNOHANG); ASSERT_EQ(ret, 0); ASSERT_EQ(fcntl(pidfd, F_SETFL, (flags & ~O_NONBLOCK)), 0); - ASSERT_EQ(sys_waitid(P_PIDFD, pidfd, &info, WSTOPPED, NULL), 0); + ASSERT_EQ(sys_waitid(P_PIDFD, pidfd, &info, WSTOPPED), 0); ASSERT_EQ(info.si_signo, SIGCHLD); ASSERT_EQ(info.si_code, CLD_STOPPED); ASSERT_EQ(info.si_pid, parent_tid); ASSERT_EQ(sys_pidfd_send_signal(pidfd, SIGCONT, NULL, 0), 0); - ASSERT_EQ(sys_waitid(P_PIDFD, pidfd, &info, WEXITED, NULL), 0); + ASSERT_EQ(sys_waitid(P_PIDFD, pidfd, &info, WEXITED), 0); ASSERT_EQ(info.si_signo, SIGCHLD); ASSERT_EQ(info.si_code, CLD_EXITED); ASSERT_EQ(info.si_pid, parent_tid); diff --git a/tools/testing/selftests/riscv/abi/pointer_masking.c b/tools/testing/selftests/riscv/abi/pointer_masking.c index dee41b7ee3e3..059d2e87eb1f 100644 --- a/tools/testing/selftests/riscv/abi/pointer_masking.c +++ b/tools/testing/selftests/riscv/abi/pointer_masking.c @@ -185,8 +185,20 @@ static void test_fork_exec(void) } } +static bool pwrite_wrapper(int fd, void *buf, size_t count, const char *msg) +{ + int ret = pwrite(fd, buf, count, 0); + + if (ret != count) { + ksft_perror(msg); + return false; + } + return true; +} + static void test_tagged_addr_abi_sysctl(void) { + char *err_pwrite_msg = "failed to write to /proc/sys/abi/tagged_addr_disabled\n"; char value; int fd; @@ -200,14 +212,18 @@ static void test_tagged_addr_abi_sysctl(void) } value = '1'; - pwrite(fd, &value, 1, 0); - ksft_test_result(set_tagged_addr_ctrl(min_pmlen, true) == -EINVAL, - "sysctl disabled\n"); + if (!pwrite_wrapper(fd, &value, 1, "write '1'")) + ksft_test_result_fail(err_pwrite_msg); + else + ksft_test_result(set_tagged_addr_ctrl(min_pmlen, true) == -EINVAL, + "sysctl disabled\n"); value = '0'; - pwrite(fd, &value, 1, 0); - ksft_test_result(set_tagged_addr_ctrl(min_pmlen, true) == 0, - "sysctl enabled\n"); + if (!pwrite_wrapper(fd, &value, 1, "write '0'")) + ksft_test_result_fail(err_pwrite_msg); + else + ksft_test_result(set_tagged_addr_ctrl(min_pmlen, true) == 0, + "sysctl enabled\n"); set_tagged_addr_ctrl(0, false); diff --git a/tools/testing/selftests/riscv/vector/v_initval_nolibc.c b/tools/testing/selftests/riscv/vector/v_initval_nolibc.c index 1dd94197da30..6174ffe016dc 100644 --- a/tools/testing/selftests/riscv/vector/v_initval_nolibc.c +++ b/tools/testing/selftests/riscv/vector/v_initval_nolibc.c @@ -25,6 +25,8 @@ int main(void) unsigned long vl; char *datap, *tmp; + ksft_set_plan(1); + datap = malloc(MAX_VSIZE); if (!datap) { ksft_test_result_fail("fail to allocate memory for size = %d\n", MAX_VSIZE); @@ -63,6 +65,8 @@ int main(void) } free(datap); + + ksft_test_result_pass("tests for v_initval_nolibc pass\n"); ksft_exit_pass(); return 0; } diff --git a/tools/testing/selftests/riscv/vector/vstate_prctl.c b/tools/testing/selftests/riscv/vector/vstate_prctl.c index 895177f6bf4c..40b3bffcbb40 100644 --- a/tools/testing/selftests/riscv/vector/vstate_prctl.c +++ b/tools/testing/selftests/riscv/vector/vstate_prctl.c @@ -76,6 +76,8 @@ int main(void) long flag, expected; long rc; + ksft_set_plan(1); + pair.key = RISCV_HWPROBE_KEY_IMA_EXT_0; rc = riscv_hwprobe(&pair, 1, 0, NULL, 0); if (rc < 0) { diff --git a/tools/testing/selftests/sched_ext/ddsp_bogus_dsq_fail.bpf.c b/tools/testing/selftests/sched_ext/ddsp_bogus_dsq_fail.bpf.c index 37d9bf6fb745..6f4c3f5a1c5d 100644 --- a/tools/testing/selftests/sched_ext/ddsp_bogus_dsq_fail.bpf.c +++ b/tools/testing/selftests/sched_ext/ddsp_bogus_dsq_fail.bpf.c @@ -20,7 +20,7 @@ s32 BPF_STRUCT_OPS(ddsp_bogus_dsq_fail_select_cpu, struct task_struct *p, * If we dispatch to a bogus DSQ that will fall back to the * builtin global DSQ, we fail gracefully. */ - scx_bpf_dispatch_vtime(p, 0xcafef00d, SCX_SLICE_DFL, + scx_bpf_dsq_insert_vtime(p, 0xcafef00d, SCX_SLICE_DFL, p->scx.dsq_vtime, 0); return cpu; } diff --git a/tools/testing/selftests/sched_ext/ddsp_vtimelocal_fail.bpf.c b/tools/testing/selftests/sched_ext/ddsp_vtimelocal_fail.bpf.c index dffc97d9cdf1..e4a55027778f 100644 --- a/tools/testing/selftests/sched_ext/ddsp_vtimelocal_fail.bpf.c +++ b/tools/testing/selftests/sched_ext/ddsp_vtimelocal_fail.bpf.c @@ -17,8 +17,8 @@ s32 BPF_STRUCT_OPS(ddsp_vtimelocal_fail_select_cpu, struct task_struct *p, if (cpu >= 0) { /* Shouldn't be allowed to vtime dispatch to a builtin DSQ. */ - scx_bpf_dispatch_vtime(p, SCX_DSQ_LOCAL, SCX_SLICE_DFL, - p->scx.dsq_vtime, 0); + scx_bpf_dsq_insert_vtime(p, SCX_DSQ_LOCAL, SCX_SLICE_DFL, + p->scx.dsq_vtime, 0); return cpu; } diff --git a/tools/testing/selftests/sched_ext/dsp_local_on.bpf.c b/tools/testing/selftests/sched_ext/dsp_local_on.bpf.c index 6a7db1502c29..fbda6bf54671 100644 --- a/tools/testing/selftests/sched_ext/dsp_local_on.bpf.c +++ b/tools/testing/selftests/sched_ext/dsp_local_on.bpf.c @@ -43,9 +43,12 @@ void BPF_STRUCT_OPS(dsp_local_on_dispatch, s32 cpu, struct task_struct *prev) if (!p) return; - target = bpf_get_prandom_u32() % nr_cpus; + if (p->nr_cpus_allowed == nr_cpus) + target = bpf_get_prandom_u32() % nr_cpus; + else + target = scx_bpf_task_cpu(p); - scx_bpf_dispatch(p, SCX_DSQ_LOCAL_ON | target, SCX_SLICE_DFL, 0); + scx_bpf_dsq_insert(p, SCX_DSQ_LOCAL_ON | target, SCX_SLICE_DFL, 0); bpf_task_release(p); } diff --git a/tools/testing/selftests/sched_ext/dsp_local_on.c b/tools/testing/selftests/sched_ext/dsp_local_on.c index 472851b56854..0ff27e57fe43 100644 --- a/tools/testing/selftests/sched_ext/dsp_local_on.c +++ b/tools/testing/selftests/sched_ext/dsp_local_on.c @@ -34,9 +34,10 @@ static enum scx_test_status run(void *ctx) /* Just sleeping is fine, plenty of scheduling events happening */ sleep(1); - SCX_EQ(skel->data->uei.kind, EXIT_KIND(SCX_EXIT_ERROR)); bpf_link__destroy(link); + SCX_EQ(skel->data->uei.kind, EXIT_KIND(SCX_EXIT_UNREG)); + return SCX_TEST_PASS; } @@ -50,7 +51,7 @@ static void cleanup(void *ctx) struct scx_test dsp_local_on = { .name = "dsp_local_on", .description = "Verify we can directly dispatch tasks to a local DSQs " - "from osp.dispatch()", + "from ops.dispatch()", .setup = setup, .run = run, .cleanup = cleanup, diff --git a/tools/testing/selftests/sched_ext/enq_select_cpu_fails.bpf.c b/tools/testing/selftests/sched_ext/enq_select_cpu_fails.bpf.c index 1efb50d61040..a7cf868d5e31 100644 --- a/tools/testing/selftests/sched_ext/enq_select_cpu_fails.bpf.c +++ b/tools/testing/selftests/sched_ext/enq_select_cpu_fails.bpf.c @@ -31,7 +31,7 @@ void BPF_STRUCT_OPS(enq_select_cpu_fails_enqueue, struct task_struct *p, /* Can only call from ops.select_cpu() */ scx_bpf_select_cpu_dfl(p, 0, 0, &found); - scx_bpf_dispatch(p, SCX_DSQ_GLOBAL, SCX_SLICE_DFL, enq_flags); + scx_bpf_dsq_insert(p, SCX_DSQ_GLOBAL, SCX_SLICE_DFL, enq_flags); } SEC(".struct_ops.link") diff --git a/tools/testing/selftests/sched_ext/exit.bpf.c b/tools/testing/selftests/sched_ext/exit.bpf.c index d75d4faf07f6..4bc36182d3ff 100644 --- a/tools/testing/selftests/sched_ext/exit.bpf.c +++ b/tools/testing/selftests/sched_ext/exit.bpf.c @@ -33,7 +33,7 @@ void BPF_STRUCT_OPS(exit_enqueue, struct task_struct *p, u64 enq_flags) if (exit_point == EXIT_ENQUEUE) EXIT_CLEANLY(); - scx_bpf_dispatch(p, DSQ_ID, SCX_SLICE_DFL, enq_flags); + scx_bpf_dsq_insert(p, DSQ_ID, SCX_SLICE_DFL, enq_flags); } void BPF_STRUCT_OPS(exit_dispatch, s32 cpu, struct task_struct *p) @@ -41,7 +41,7 @@ void BPF_STRUCT_OPS(exit_dispatch, s32 cpu, struct task_struct *p) if (exit_point == EXIT_DISPATCH) EXIT_CLEANLY(); - scx_bpf_consume(DSQ_ID); + scx_bpf_dsq_move_to_local(DSQ_ID); } void BPF_STRUCT_OPS(exit_enable, struct task_struct *p) diff --git a/tools/testing/selftests/sched_ext/maximal.bpf.c b/tools/testing/selftests/sched_ext/maximal.bpf.c index 4d4cd8d966db..430f5e13bf55 100644 --- a/tools/testing/selftests/sched_ext/maximal.bpf.c +++ b/tools/testing/selftests/sched_ext/maximal.bpf.c @@ -12,6 +12,8 @@ char _license[] SEC("license") = "GPL"; +#define DSQ_ID 0 + s32 BPF_STRUCT_OPS(maximal_select_cpu, struct task_struct *p, s32 prev_cpu, u64 wake_flags) { @@ -20,7 +22,7 @@ s32 BPF_STRUCT_OPS(maximal_select_cpu, struct task_struct *p, s32 prev_cpu, void BPF_STRUCT_OPS(maximal_enqueue, struct task_struct *p, u64 enq_flags) { - scx_bpf_dispatch(p, SCX_DSQ_GLOBAL, SCX_SLICE_DFL, enq_flags); + scx_bpf_dsq_insert(p, DSQ_ID, SCX_SLICE_DFL, enq_flags); } void BPF_STRUCT_OPS(maximal_dequeue, struct task_struct *p, u64 deq_flags) @@ -28,7 +30,7 @@ void BPF_STRUCT_OPS(maximal_dequeue, struct task_struct *p, u64 deq_flags) void BPF_STRUCT_OPS(maximal_dispatch, s32 cpu, struct task_struct *prev) { - scx_bpf_consume(SCX_DSQ_GLOBAL); + scx_bpf_dsq_move_to_local(DSQ_ID); } void BPF_STRUCT_OPS(maximal_runnable, struct task_struct *p, u64 enq_flags) @@ -123,7 +125,7 @@ void BPF_STRUCT_OPS(maximal_cgroup_set_weight, struct cgroup *cgrp, u32 weight) s32 BPF_STRUCT_OPS_SLEEPABLE(maximal_init) { - return 0; + return scx_bpf_create_dsq(DSQ_ID, -1); } void BPF_STRUCT_OPS(maximal_exit, struct scx_exit_info *info) diff --git a/tools/testing/selftests/sched_ext/select_cpu_dfl.bpf.c b/tools/testing/selftests/sched_ext/select_cpu_dfl.bpf.c index f171ac470970..13d0f5be788d 100644 --- a/tools/testing/selftests/sched_ext/select_cpu_dfl.bpf.c +++ b/tools/testing/selftests/sched_ext/select_cpu_dfl.bpf.c @@ -30,7 +30,7 @@ void BPF_STRUCT_OPS(select_cpu_dfl_enqueue, struct task_struct *p, } scx_bpf_put_idle_cpumask(idle_mask); - scx_bpf_dispatch(p, SCX_DSQ_GLOBAL, SCX_SLICE_DFL, enq_flags); + scx_bpf_dsq_insert(p, SCX_DSQ_GLOBAL, SCX_SLICE_DFL, enq_flags); } SEC(".struct_ops.link") diff --git a/tools/testing/selftests/sched_ext/select_cpu_dfl_nodispatch.bpf.c b/tools/testing/selftests/sched_ext/select_cpu_dfl_nodispatch.bpf.c index 9efdbb7da928..815f1d5d61ac 100644 --- a/tools/testing/selftests/sched_ext/select_cpu_dfl_nodispatch.bpf.c +++ b/tools/testing/selftests/sched_ext/select_cpu_dfl_nodispatch.bpf.c @@ -67,7 +67,7 @@ void BPF_STRUCT_OPS(select_cpu_dfl_nodispatch_enqueue, struct task_struct *p, saw_local = true; } - scx_bpf_dispatch(p, dsq_id, SCX_SLICE_DFL, enq_flags); + scx_bpf_dsq_insert(p, dsq_id, SCX_SLICE_DFL, enq_flags); } s32 BPF_STRUCT_OPS(select_cpu_dfl_nodispatch_init_task, diff --git a/tools/testing/selftests/sched_ext/select_cpu_dispatch.bpf.c b/tools/testing/selftests/sched_ext/select_cpu_dispatch.bpf.c index 59bfc4f36167..4bb99699e920 100644 --- a/tools/testing/selftests/sched_ext/select_cpu_dispatch.bpf.c +++ b/tools/testing/selftests/sched_ext/select_cpu_dispatch.bpf.c @@ -29,7 +29,7 @@ s32 BPF_STRUCT_OPS(select_cpu_dispatch_select_cpu, struct task_struct *p, cpu = prev_cpu; dispatch: - scx_bpf_dispatch(p, dsq_id, SCX_SLICE_DFL, 0); + scx_bpf_dsq_insert(p, dsq_id, SCX_SLICE_DFL, 0); return cpu; } diff --git a/tools/testing/selftests/sched_ext/select_cpu_dispatch_bad_dsq.bpf.c b/tools/testing/selftests/sched_ext/select_cpu_dispatch_bad_dsq.bpf.c index 3bbd5fcdfb18..2a75de11b2cf 100644 --- a/tools/testing/selftests/sched_ext/select_cpu_dispatch_bad_dsq.bpf.c +++ b/tools/testing/selftests/sched_ext/select_cpu_dispatch_bad_dsq.bpf.c @@ -18,7 +18,7 @@ s32 BPF_STRUCT_OPS(select_cpu_dispatch_bad_dsq_select_cpu, struct task_struct *p s32 prev_cpu, u64 wake_flags) { /* Dispatching to a random DSQ should fail. */ - scx_bpf_dispatch(p, 0xcafef00d, SCX_SLICE_DFL, 0); + scx_bpf_dsq_insert(p, 0xcafef00d, SCX_SLICE_DFL, 0); return prev_cpu; } diff --git a/tools/testing/selftests/sched_ext/select_cpu_dispatch_dbl_dsp.bpf.c b/tools/testing/selftests/sched_ext/select_cpu_dispatch_dbl_dsp.bpf.c index 0fda57fe0ecf..99d075695c97 100644 --- a/tools/testing/selftests/sched_ext/select_cpu_dispatch_dbl_dsp.bpf.c +++ b/tools/testing/selftests/sched_ext/select_cpu_dispatch_dbl_dsp.bpf.c @@ -18,8 +18,8 @@ s32 BPF_STRUCT_OPS(select_cpu_dispatch_dbl_dsp_select_cpu, struct task_struct *p s32 prev_cpu, u64 wake_flags) { /* Dispatching twice in a row is disallowed. */ - scx_bpf_dispatch(p, SCX_DSQ_GLOBAL, SCX_SLICE_DFL, 0); - scx_bpf_dispatch(p, SCX_DSQ_GLOBAL, SCX_SLICE_DFL, 0); + scx_bpf_dsq_insert(p, SCX_DSQ_GLOBAL, SCX_SLICE_DFL, 0); + scx_bpf_dsq_insert(p, SCX_DSQ_GLOBAL, SCX_SLICE_DFL, 0); return prev_cpu; } diff --git a/tools/testing/selftests/sched_ext/select_cpu_vtime.bpf.c b/tools/testing/selftests/sched_ext/select_cpu_vtime.bpf.c index e6c67bcf5e6e..bfcb96cd4954 100644 --- a/tools/testing/selftests/sched_ext/select_cpu_vtime.bpf.c +++ b/tools/testing/selftests/sched_ext/select_cpu_vtime.bpf.c @@ -2,8 +2,8 @@ /* * A scheduler that validates that enqueue flags are properly stored and * applied at dispatch time when a task is directly dispatched from - * ops.select_cpu(). We validate this by using scx_bpf_dispatch_vtime(), and - * making the test a very basic vtime scheduler. + * ops.select_cpu(). We validate this by using scx_bpf_dsq_insert_vtime(), + * and making the test a very basic vtime scheduler. * * Copyright (c) 2024 Meta Platforms, Inc. and affiliates. * Copyright (c) 2024 David Vernet <dvernet@meta.com> @@ -47,13 +47,13 @@ s32 BPF_STRUCT_OPS(select_cpu_vtime_select_cpu, struct task_struct *p, cpu = prev_cpu; scx_bpf_test_and_clear_cpu_idle(cpu); ddsp: - scx_bpf_dispatch_vtime(p, VTIME_DSQ, SCX_SLICE_DFL, task_vtime(p), 0); + scx_bpf_dsq_insert_vtime(p, VTIME_DSQ, SCX_SLICE_DFL, task_vtime(p), 0); return cpu; } void BPF_STRUCT_OPS(select_cpu_vtime_dispatch, s32 cpu, struct task_struct *p) { - if (scx_bpf_consume(VTIME_DSQ)) + if (scx_bpf_dsq_move_to_local(VTIME_DSQ)) consumed = true; } diff --git a/tools/testing/selftests/tc-testing/tc-tests/filters/flow.json b/tools/testing/selftests/tc-testing/tc-tests/filters/flow.json index 996448afe31b..91d120548bf5 100644 --- a/tools/testing/selftests/tc-testing/tc-tests/filters/flow.json +++ b/tools/testing/selftests/tc-testing/tc-tests/filters/flow.json @@ -78,10 +78,10 @@ "setup": [ "$TC qdisc add dev $DEV1 ingress" ], - "cmdUnderTest": "$TC filter add dev $DEV1 parent ffff: handle 1 prio 1 protocol ip flow map key dst rshift 0xff", + "cmdUnderTest": "$TC filter add dev $DEV1 parent ffff: handle 1 prio 1 protocol ip flow map key dst rshift 0x1f", "expExitCode": "0", "verifyCmd": "$TC filter get dev $DEV1 parent ffff: handle 1 protocol ip prio 1 flow", - "matchPattern": "filter parent ffff: protocol ip pref 1 flow chain [0-9]+ handle 0x1 map keys dst rshift 255 baseclass", + "matchPattern": "filter parent ffff: protocol ip pref 1 flow chain [0-9]+ handle 0x1 map keys dst rshift 31 baseclass", "matchCount": "1", "teardown": [ "$TC qdisc del dev $DEV1 ingress" diff --git a/tools/testing/shared/linux/maple_tree.h b/tools/testing/shared/linux/maple_tree.h index 06c89bdcc515..f67d47d32857 100644 --- a/tools/testing/shared/linux/maple_tree.h +++ b/tools/testing/shared/linux/maple_tree.h @@ -2,6 +2,6 @@ #define atomic_t int32_t #define atomic_inc(x) uatomic_inc(x) #define atomic_read(x) uatomic_read(x) -#define atomic_set(x, y) do {} while (0) +#define atomic_set(x, y) uatomic_set(x, y) #define U8_MAX UCHAR_MAX #include "../../../../include/linux/maple_tree.h" diff --git a/tools/testing/vma/linux/atomic.h b/tools/testing/vma/linux/atomic.h index e01f66f98982..3e1b6adc027b 100644 --- a/tools/testing/vma/linux/atomic.h +++ b/tools/testing/vma/linux/atomic.h @@ -6,7 +6,7 @@ #define atomic_t int32_t #define atomic_inc(x) uatomic_inc(x) #define atomic_read(x) uatomic_read(x) -#define atomic_set(x, y) do {} while (0) +#define atomic_set(x, y) uatomic_set(x, y) #define U8_MAX UCHAR_MAX #endif /* _LINUX_ATOMIC_H */ diff --git a/tools/tracing/rtla/src/timerlat_hist.c b/tools/tracing/rtla/src/timerlat_hist.c index 8b66387e5f35..4403cc4eba30 100644 --- a/tools/tracing/rtla/src/timerlat_hist.c +++ b/tools/tracing/rtla/src/timerlat_hist.c @@ -282,6 +282,21 @@ static void timerlat_hist_header(struct osnoise_tool *tool) } /* + * format_summary_value - format a line of summary value (min, max or avg) + * of hist data + */ +static void format_summary_value(struct trace_seq *seq, + int count, + unsigned long long val, + bool avg) +{ + if (count) + trace_seq_printf(seq, "%9llu ", avg ? val / count : val); + else + trace_seq_printf(seq, "%9c ", '-'); +} + +/* * timerlat_print_summary - print the summary of the hist data to the output */ static void @@ -328,29 +343,23 @@ timerlat_print_summary(struct timerlat_hist_params *params, if (!data->hist[cpu].irq_count && !data->hist[cpu].thread_count) continue; - if (!params->no_irq) { - if (data->hist[cpu].irq_count) - trace_seq_printf(trace->seq, "%9llu ", - data->hist[cpu].min_irq); - else - trace_seq_printf(trace->seq, " - "); - } + if (!params->no_irq) + format_summary_value(trace->seq, + data->hist[cpu].irq_count, + data->hist[cpu].min_irq, + false); - if (!params->no_thread) { - if (data->hist[cpu].thread_count) - trace_seq_printf(trace->seq, "%9llu ", - data->hist[cpu].min_thread); - else - trace_seq_printf(trace->seq, " - "); - } + if (!params->no_thread) + format_summary_value(trace->seq, + data->hist[cpu].thread_count, + data->hist[cpu].min_thread, + false); - if (params->user_hist) { - if (data->hist[cpu].user_count) - trace_seq_printf(trace->seq, "%9llu ", - data->hist[cpu].min_user); - else - trace_seq_printf(trace->seq, " - "); - } + if (params->user_hist) + format_summary_value(trace->seq, + data->hist[cpu].user_count, + data->hist[cpu].min_user, + false); } trace_seq_printf(trace->seq, "\n"); @@ -364,29 +373,23 @@ timerlat_print_summary(struct timerlat_hist_params *params, if (!data->hist[cpu].irq_count && !data->hist[cpu].thread_count) continue; - if (!params->no_irq) { - if (data->hist[cpu].irq_count) - trace_seq_printf(trace->seq, "%9llu ", - data->hist[cpu].sum_irq / data->hist[cpu].irq_count); - else - trace_seq_printf(trace->seq, " - "); - } + if (!params->no_irq) + format_summary_value(trace->seq, + data->hist[cpu].irq_count, + data->hist[cpu].sum_irq, + true); - if (!params->no_thread) { - if (data->hist[cpu].thread_count) - trace_seq_printf(trace->seq, "%9llu ", - data->hist[cpu].sum_thread / data->hist[cpu].thread_count); - else - trace_seq_printf(trace->seq, " - "); - } + if (!params->no_thread) + format_summary_value(trace->seq, + data->hist[cpu].thread_count, + data->hist[cpu].sum_thread, + true); - if (params->user_hist) { - if (data->hist[cpu].user_count) - trace_seq_printf(trace->seq, "%9llu ", - data->hist[cpu].sum_user / data->hist[cpu].user_count); - else - trace_seq_printf(trace->seq, " - "); - } + if (params->user_hist) + format_summary_value(trace->seq, + data->hist[cpu].user_count, + data->hist[cpu].sum_user, + true); } trace_seq_printf(trace->seq, "\n"); @@ -400,29 +403,23 @@ timerlat_print_summary(struct timerlat_hist_params *params, if (!data->hist[cpu].irq_count && !data->hist[cpu].thread_count) continue; - if (!params->no_irq) { - if (data->hist[cpu].irq_count) - trace_seq_printf(trace->seq, "%9llu ", - data->hist[cpu].max_irq); - else - trace_seq_printf(trace->seq, " - "); - } + if (!params->no_irq) + format_summary_value(trace->seq, + data->hist[cpu].irq_count, + data->hist[cpu].max_irq, + false); - if (!params->no_thread) { - if (data->hist[cpu].thread_count) - trace_seq_printf(trace->seq, "%9llu ", - data->hist[cpu].max_thread); - else - trace_seq_printf(trace->seq, " - "); - } + if (!params->no_thread) + format_summary_value(trace->seq, + data->hist[cpu].thread_count, + data->hist[cpu].max_thread, + false); - if (params->user_hist) { - if (data->hist[cpu].user_count) - trace_seq_printf(trace->seq, "%9llu ", - data->hist[cpu].max_user); - else - trace_seq_printf(trace->seq, " - "); - } + if (params->user_hist) + format_summary_value(trace->seq, + data->hist[cpu].user_count, + data->hist[cpu].max_user, + false); } trace_seq_printf(trace->seq, "\n"); trace_seq_do_printf(trace->seq); @@ -506,16 +503,22 @@ timerlat_print_stats_all(struct timerlat_hist_params *params, trace_seq_printf(trace->seq, "min: "); if (!params->no_irq) - trace_seq_printf(trace->seq, "%9llu ", - sum.min_irq); + format_summary_value(trace->seq, + sum.irq_count, + sum.min_irq, + false); if (!params->no_thread) - trace_seq_printf(trace->seq, "%9llu ", - sum.min_thread); + format_summary_value(trace->seq, + sum.thread_count, + sum.min_thread, + false); if (params->user_hist) - trace_seq_printf(trace->seq, "%9llu ", - sum.min_user); + format_summary_value(trace->seq, + sum.user_count, + sum.min_user, + false); trace_seq_printf(trace->seq, "\n"); @@ -523,16 +526,22 @@ timerlat_print_stats_all(struct timerlat_hist_params *params, trace_seq_printf(trace->seq, "avg: "); if (!params->no_irq) - trace_seq_printf(trace->seq, "%9llu ", - sum.sum_irq / sum.irq_count); + format_summary_value(trace->seq, + sum.irq_count, + sum.sum_irq, + true); if (!params->no_thread) - trace_seq_printf(trace->seq, "%9llu ", - sum.sum_thread / sum.thread_count); + format_summary_value(trace->seq, + sum.thread_count, + sum.sum_thread, + true); if (params->user_hist) - trace_seq_printf(trace->seq, "%9llu ", - sum.sum_user / sum.user_count); + format_summary_value(trace->seq, + sum.user_count, + sum.sum_user, + true); trace_seq_printf(trace->seq, "\n"); @@ -540,16 +549,22 @@ timerlat_print_stats_all(struct timerlat_hist_params *params, trace_seq_printf(trace->seq, "max: "); if (!params->no_irq) - trace_seq_printf(trace->seq, "%9llu ", - sum.max_irq); + format_summary_value(trace->seq, + sum.irq_count, + sum.max_irq, + false); if (!params->no_thread) - trace_seq_printf(trace->seq, "%9llu ", - sum.max_thread); + format_summary_value(trace->seq, + sum.thread_count, + sum.max_thread, + false); if (params->user_hist) - trace_seq_printf(trace->seq, "%9llu ", - sum.max_user); + format_summary_value(trace->seq, + sum.user_count, + sum.max_user, + false); trace_seq_printf(trace->seq, "\n"); trace_seq_do_printf(trace->seq); |