summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2017-06-29iwlwifi: pcie: improve debug in iwl_pcie_rx_handle_rb()Johannes Berg
Print the queue for the existing debug message and add a new debug message indicating where the RB ended. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
2017-06-29iwlwifi: mvm: support aggs of 64 frames in A000 familyLiad Kaufman
A SCD bug was fixed in the A000 family, allowing to support aggregations of 64 frames (rather than 63). Signed-off-by: Liad Kaufman <liad.kaufman@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
2017-06-29iwlwifi: pcie: improve "invalid queue" warningJohannes Berg
Print out both queue IDs to be able to see what went wrong. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
2017-06-29iwlwifi: mvm: use proper CDB check in PHY context modifyJohannes Berg
When the firmware supports CDB, PHY contexts cannot be modified to change their band, but need to be added/remove instead. Instead of relying on iwl_mvm_has_new_tx_api(), check the right FW capa flag IWL_UCODE_TLV_CAPA_BINDING_CDB_SUPPORT and remove the comment. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
2017-06-29iwlwifi: add twelve new 9560 series PCI IDsTzipi Peres
Add twelve new PCI IDs for the 9560 series. Signed-off-by: Tzipi Peres <tzipi.peres@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
2017-06-29iwlwifi: add the new a000_2ax seriesTzipi Peres
Add a new config struct for the new a000 2ax series and add the five PCI ID for it. Signed-off-by: Tzipi Peres <tzipi.peres@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
2017-06-29iwlwifi: mvm: simplify CHECK_MLME_TRIGGER macroLuca Coelho
There's no reason to pass mvm and trig as parameters to the macro, since it will be expanded inside the function itself. Also remove the bogus buf parameter which doesn't exist and is not used. Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
2017-06-29iwlwifi: mvm: change sta_id to u8Sharon Dvir
The sta_id variable is used as an index in an array, should be unsigned. Found by Klocwork. Fixes: 9f9af3d7d303 ("iwlwifi: mvm: re-aggregate shared queue after unsharing") Signed-off-by: Sharon Dvir <sharon.dvir@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
2017-06-29iwlwifi: pcie: fix 9000-series RF-kill interrupt propagationJohannes Berg
A hardware issue on 9000 series devices sometimes causes RF-kill interrupts to not be propagated to the host properly if ASPM is enabled. Work around this by setting the right hardware bit to allow it to interrupt the host for this reason (rfkill). Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
2017-06-29iwlwifi: pcie: only apply retention workaround on 9000-series A-stepJohannes Berg
Due to a hardware issue, certain power saving had to be disabled. However, this issue was fixed in B-step, so the workaround only needs to apply to A-step. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
2017-06-29iwlwifi: dvm: use macros for format stringsJohannes Berg
Some static checkers (e.g. smatch) complain if a non-constant format string is used, even if that's a static const variable. Since there's no impact on code generation, just change those format strings to be macros. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
2017-06-29iwlwifi: mvm: fix a bunch of kernel-doc warningsJohannes Berg
Fix the kernel-doc, and remove some fields even the firmware doesn't use in ToF, RX, scan, station and generic FW APIS. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
2017-06-29iwlwifi: mvm: add documentation for all command IDsJohannes Berg
Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
2017-06-29iwlwifi: mvm: use __le16 even for reserved fieldsJohannes Berg
Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
2017-06-29iwlwifi: mvm: remove various unused command IDs/structsJohannes Berg
Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
2017-06-29iwlwifi: mvm: fix various "Excess ... description" kernel-doc warningsJohannes Berg
Fix various "Excess struct/union/enum/typedef member '...' description in '...'" warnings from kernel-doc, mostly caused by typos. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
2017-06-29iwlwifi: mvm: remove some CamelCase from firmware APIJohannes Berg
Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
2017-06-29iwlwifi: mvm: Add debugfs entry to retrieve SAR geographic profileHaim Dreyfuss
Add a debugfs entry to get a verbose description of the power settings used in each band with the currently selected SAR geographic profile. Signed-off-by: Haim Dreyfuss <haim.dreyfuss@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
2017-06-29iwlwifi: mvm: refactor geo initHaim Dreyfuss
We are going to add debugfs entry to retrieve the current geographic profile being used in the FW. Currently the driver reads those tables from the BIOS and passes them to the FW. To prepare for this retrieving we want to store those tables in the driver. Signed-off-by: Haim Dreyfuss <haim.dreyfuss@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
2017-06-29arm64: fix endianness annotation for reloc_insn_movw() & reloc_insn_imm()Luc Van Oostenryck
Here the functions reloc_insn_movw() & reloc_insn_imm() are used to read, modify and write back ARM instructions, which are always stored in memory in little-endian order. These values are thus correctly converted to/from native order but the pointers used to hold their addresses are declared as for native order values. Fix this by declaring the pointers as __le32* and remove the casts that are now unneeded. Signed-off-by: Luc Van Oostenryck <luc.vanoostenryck@gmail.com> Signed-off-by: Will Deacon <will.deacon@arm.com>
2017-06-29arm64: fix endianness annotation for aarch64_insn_write()Luc Van Oostenryck
aarch64_insn_write() is used to write an instruction. As on ARM64 in-memory instructions are always stored in little-endian order, this function, taking the instruction opcode in native order, correctly convert it to little-endian before sending it to an helper function __aarch64_insn_write() which will do the effective write. This is all good, but the variable and argument holding the converted value are not annotated for a little-endian value but left for native values. Fix this by adjusting the prototype of the helper and directly using the result of cpu_to_le32() without passing by an intermediate variable (which was not a distinct one but the same as the one holding the native value). Signed-off-by: Luc Van Oostenryck <luc.vanoostenryck@gmail.com> Signed-off-by: Will Deacon <will.deacon@arm.com>
2017-06-29arm64: fix endianness annotation in aarch64_insn_read()Luc Van Oostenryck
The function arch64_insn_read() is used to read an instruction. On AM64 instructions are always stored in little-endian order and thus the function correctly do a little-to-native endian conversion to the value just read. However, the variable used to hold the value before the conversion is not declared for a little-endian value but for a native one. Fix this by using the correct type for the declaration: __le32 Note: This only works because the function reading the value, probe_kernel_read((), takes a void pointer and void pointers are endian-agnostic. Otherwise probe_kernel_read() should also be properly annotated (or worse, need to be specialized). Signed-off-by: Luc Van Oostenryck <luc.vanoostenryck@gmail.com> Signed-off-by: Will Deacon <will.deacon@arm.com>
2017-06-29arm64: fix endianness annotation in call_undef_hook()Luc Van Oostenryck
Here we're reading thumb or ARM instructions, which are always stored in memory in little-endian order. These values are thus correctly converted to native order but the intermediate value should be annotated as for little-endian values. Fix this by declaring the intermediate var as __le32 or __le16. Signed-off-by: Luc Van Oostenryck <luc.vanoostenryck@gmail.com> Signed-off-by: Will Deacon <will.deacon@arm.com>
2017-06-29arm64: fix endianness annotation for debug-monitors.cLuc Van Oostenryck
Here we're reading thumb or ARM instructions, which are always stored in memory in little-endian order. These values are thus correctly converted to native order but the intermediate value should be annotated as for little-endian values. Fix this by declaring the intermediate var as __le32 or __le16. Signed-off-by: Luc Van Oostenryck <luc.vanoostenryck@gmail.com> Signed-off-by: Will Deacon <will.deacon@arm.com>
2017-06-29ras: mark stub functions as 'inline'Arnd Bergmann
With CONFIG_RAS disabled, we get two harmless warnings about unused functions: include/linux/ras.h:37:13: error: 'log_arm_hw_error' defined but not used [-Werror=unused-function] static void log_arm_hw_error(struct cper_sec_proc_arm *err) { return; } include/linux/ras.h:33:13: error: 'log_non_standard_event' defined but not used [-Werror=unused-function] static void log_non_standard_event(const guid_t *sec_type, Clearly these are meant to be 'inline', like the other stubs in the same header. Fixes: 297b64c74385 ("ras: acpi / apei: generate trace event for unrecognized CPER section") Fixes: e9279e83ad1f ("trace, ras: add ARM processor error trace event") Acked-by: Borislav Petkov <bp@suse.de> Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Will Deacon <will.deacon@arm.com>
2017-06-29gpiolib: fix filtering out unwanted eventsBartosz Golaszewski
GPIOEVENT_REQUEST_BOTH_EDGES is not a single flag, but a binary OR of GPIOEVENT_REQUEST_RISING_EDGE and GPIOEVENT_REQUEST_FALLING_EDGE. The expression 'le->eflags & GPIOEVENT_REQUEST_BOTH_EDGES' we'll get evaluated to true even if only one event type was requested. Fix it by checking both RISING & FALLING flags explicitly. Cc: stable@vger.kernel.org Fixes: 61f922db7221 ("gpio: userspace ABI for reading GPIO line events") Signed-off-by: Bartosz Golaszewski <brgl@bgdev.pl> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2017-06-29EDAC, pnd2: Fix Apollo Lake DIMM detectionTony Luck
Non-existent or empty DIMM slots result in error return from RD_REGP(). But we shouldn't give up on failure. So long as we find at least one DIMM we can continue. Signed-off-by: Tony Luck <tony.luck@intel.com> Cc: Qiuxu Zhuo <qiuxu.zhuo@intel.com> Cc: linux-edac <linux-edac@vger.kernel.org> Link: http://lkml.kernel.org/r/20170628234407.21521-1-tony.luck@intel.com Signed-off-by: Borislav Petkov <bp@suse.de>
2017-06-29EDAC, i5000, i5400: Fix definition of NRECMEMB registerJérémy Lefaure
In the i5000 and i5400 drivers, the NRECMEMB register is defined as a 16-bit value, which results in wrong shifts in the code, as reported by sparse. In the datasheets ([1], section 3.9.22.20 and [2], section 3.9.22.21), this register is a 32-bit register. A u32 value for the register fixes the wrong shifts warnings and matches the datasheet. Also fix the mask to access to the CAS bits [27:16] in the i5000 driver. [1]: https://www.intel.com/content/dam/doc/datasheet/5000p-5000v-5000z-chipset-memory-controller-hub-datasheet.pdf [2]: https://www.intel.se/content/dam/doc/datasheet/5400-chipset-memory-controller-hub-datasheet.pdf Signed-off-by: Jérémy Lefaure <jeremy.lefaure@lse.epita.fr> Cc: linux-edac <linux-edac@vger.kernel.org> Link: http://lkml.kernel.org/r/20170629005729.8478-1-jeremy.lefaure@lse.epita.fr Signed-off-by: Borislav Petkov <bp@suse.de>
2017-06-29sched/numa: Hide numa_wake_affine() from UP buildThomas Gleixner
Stephen reported the following build warning in UP: kernel/sched/fair.c:2657:9: warning: 'struct sched_domain' declared inside parameter list ^ /home/sfr/next/next/kernel/sched/fair.c:2657:9: warning: its scope is only this definition or declaration, which is probably not what you want Hide the numa_wake_affine() inline stub on UP builds to get rid of it. Fixes: 3fed382b46ba ("sched/numa: Implement NUMA node level wake_affine()") Reported-by: Stephen Rothwell <sfr@canb.auug.org.au> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Cc: Rik van Riel <riel@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org>
2017-06-29PM / Domains: Fix missing default_power_down_ok commentKrzysztof Kozlowski
Commit fc5cbf0c94b6 (PM / Domains: Support for multiple states) split out some code out of default_power_down_ok() function so the documentation has to be moved to appropriate place. Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org> Acked-by: Ulf Hansson <ulf.hansson@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2017-06-29PM / Domains: Fix unsafe iteration over modified list of domainsKrzysztof Kozlowski
of_genpd_remove_last() iterates over list of domains and removes matching element thus it has to use safe version of list iteration. Fixes: 17926551c98a (PM / Domains: Add support for removing nested PM domains by provider) Cc: 4.9+ <stable@vger.kernel.org> # 4.9+ Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org> Acked-by: Ulf Hansson <ulf.hansson@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2017-06-29PM / Domains: Fix unsafe iteration over modified list of domain providersKrzysztof Kozlowski
of_genpd_del_provider() iterates over list of domain provides and removes matching element thus it has to use safe version of list iteration. Fixes: aa42240ab254 (PM / Domains: Add generic OF-based PM domain look-up) Cc: 3.19+ <stable@vger.kernel.org> # 3.19+ Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org> Acked-by: Ulf Hansson <ulf.hansson@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2017-06-29PM / Domains: Fix unsafe iteration over modified list of device linksKrzysztof Kozlowski
pm_genpd_remove_subdomain() iterates over domain's master_links list and removes matching element thus it has to use safe version of list iteration. Fixes: f721889ff65a ("PM / Domains: Support for generic I/O PM domains (v8)") Cc: 3.1+ <stable@vger.kernel.org> # 3.1+ Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org> Acked-by: Ulf Hansson <ulf.hansson@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2017-06-29PM / Domains: Handle safely genpd_syscore_switch() call on non-genpd deviceKrzysztof Kozlowski
genpd_syscore_switch() had two problems: 1. It silently assumed that device, it is being called for, belongs to generic power domain and used container_of() on its power domain pointer. Such assumption might not be true always. 2. It iterated over list of generic power domains without holding gpd_list_lock mutex thus list could have been modified at the same time. Usage of genpd_lookup_dev() solves both problems as it is safe a call for non-generic power domains and uses mutex when iterating. Reported-by: Ulf Hansson <ulf.hansson@linaro.org> Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org> Acked-by: Ulf Hansson <ulf.hansson@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2017-06-29PM / Domains: Call driver's noirq callbacksMikko Perttunen
Currently genpd installs its own noirq callbacks, but never calls down to the driver's corresponding callbacks. Add these calls. Signed-off-by: Mikko Perttunen <mperttunen@nvidia.com> Acked-by: Ulf Hansson <ulf.hansson@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2017-06-28arch: remove unused macro/function thread_saved_pc()Tobias Klauser
The only user of thread_saved_pc() in non-arch-specific code was removed in commit 8243d5597793 ("sched/core: Remove pointless printout in sched_show_task()"). Remove the implementations as well. Some architectures use thread_saved_pc() in their arch-specific code. Leave their thread_saved_pc() intact. Signed-off-by: Tobias Klauser <tklauser@distanz.ch> Acked-by: Geert Uytterhoeven <geert@linux-m68k.org> Cc: Ingo Molnar <mingo@kernel.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-06-29timers: Make the cpu base lock rawSebastian Andrzej Siewior
The timers cpu base lock could not be converted to a raw spinlock becaue the lock held time was non-deterministic due to cascading and long lasting timer wheel traversals. The rework of the timer wheel to the new non-cascading model removed also the wheel traversals and the lock held times are deterministic now. This allows to make the lock raw and thereby unbreaks NOHz* on preempt-RT. Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Link: http://lkml.kernel.org/r/20170627161538.30257-1-bigeasy@linutronix.de Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
2017-06-29ACPI / EC: Add quirk for GL720VMKCarlo Caione
ASUS GL720VMK is also affected by the EC GPE preference issue. Signed-off-by: Carlo Caione <carlo@caione.org> Signed-off-by: Lv Zheng <lv.zheng@intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2017-06-29ACPI / EC: Fix media keys not working problem on some Asus laptopsChris Chiu
Some Asus laptops (verified on X550VXK/FX502VD/FX502VE) get no interrupts when pressing media keys thus the corresponding functions are not invoked. It's due to the _GPE defines in DSDT for EC returns differnt value compared to the GPE Number in ECDT. Confirmed with Asus that the vale in ECDT is the correct one. This commit uses DMI quirks to prevent calling _GPE when doing ec_parse_device() and keep the ECDT GPE number setting for the EC device. With previous commit, it is ensured that if there is an ECDT, it can always be kept as boot_ec, this patch thus can implement a quirk on top of the determined ECDT boot_ec. Link: https://phabricator.endlessm.com/T16033 Link: https://phabricator.endlessm.com/T16722 Link: https://bugzilla.kernel.org/show_bug.cgi?id=195651 Tested-by: Daniel Drake <drake@endlessm.com> Signed-off-by: Chris Chiu <chiu@endlessm.com> Signed-off-by: Carlo Caione <carlo@caione.org> Signed-off-by: Lv Zheng <lv.zheng@intel.com> Signed-off-by: Lv Zheng <lv.zheng@intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2017-06-29ACPI / EC: Add support to skip boot stage DSDT probeLv Zheng
We prepared _INI/_STA methods for \_SB, \_SB.PCI0, \_SB.LID0 and \_SB.EC, _HID(PNP0C09)/_CRS/_GPE for \_SB.EC to poke Windows behavior with qemu, we got the following execution sequence: \_SB._INI \_SB.PCI0._STA \_SB.LID0._STA \_SB.EC._STA \_SB.PCI0._INI \_SB.LID0._INI \_SB.EC._INI There is no extra DSDT EC device enumeration process occurring before the main ACPI device enumeration process. That means acpi_ec_dsdt_probe() is not Windows-compatible. Tracking back, it was added by the following commit: Commit: c5279dee26c0e8d7c4200993bfc4b540d2469598 Subject: ACPI: EC: Add some basic check for ECDT data but that commit was misguided. Why we shouldn't enumerate DSDT EC before the main ACPI device enumeration? The only way to know if the DSDT EC is valid would be to evaluate its _STA control method, but it's not safe to evaluate this control method that early and out of the ACPI enumeration process, because _STA may refer to entities (such as resources or ACPI device objects) that may not have been initialized before OSPM starts to enumerate them via the main ACPI device enumeration. But after we had reverted back to the expected behavior, a regression was reported. On that platform, there is no ECDT, but the platform control methods access EC operation region earlier than Linux expects causing some ACPI method execution errors. For this reason, we just go back to old behavior to still probe DSDT EC as the boot EC. However, that turns out to lead to yet another functional breakage and in order to work around all of the problems, we skip boot stage DSDT probe when the ECDT exists so that a later quirk can always use correct ECDT GPE setting. Link: http://bugzilla.kernel.org/show_bug.cgi?id=11880 Link: http://bugzilla.kernel.org/show_bug.cgi?id=119261 Link: http://bugzilla.kernel.org/show_bug.cgi?id=195651 Tested-by: Daniel Drake <drake@endlessm.com> Signed-off-by: Lv Zheng <lv.zheng@intel.com> [ rjw: Changelog & comments massage ] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2017-06-29ACPI / EC: Enhance boot EC sanity checkLv Zheng
It's reported that some buggy BIOS tables can contain 2 DSDT ECs, one of them is invalid but acpi_ec_dsdt_probe() fails to pick the valid one. This patch simply enhances sanity checks in ec_parse_device() as a workaround to skip probing wrong namespace ECs. Link: https://bugzilla.kernel.org/show_bug.cgi?id=195651 Tested-by: Daniel Drake <drake@endlessm.com> Signed-off-by: Lv Zheng <lv.zheng@intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2017-06-28ACPI / video: Add quirks for the Dell Precision 7510Shih-Yuan Lee (FourDollars)
The Dell Precision 7510 has a broken acpi-video implementation. The backlight control bits work, but when the brightness is changed via the acpi-video interface the backlight flickers annoyingly before settling at the new brightness, switching to using the native interface fixes the flickering so add a quirk for this (the vendor interface can not change the brightness at all). Signed-off-by: Shih-Yuan Lee (FourDollars) <sylee@canonical.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2017-06-28block: provide bio_uninit() free freeing integrity/task associationsJens Axboe
Wen reports significant memory leaks with DIF and O_DIRECT: "With nvme devive + T10 enabled, On a system it has 256GB and started logging /proc/meminfo & /proc/slabinfo for every minute and in an hour it increased by 15968128 kB or ~15+GB.. Approximately 256 MB / minute leaking. /proc/meminfo | grep SUnreclaim... SUnreclaim: 6752128 kB SUnreclaim: 6874880 kB SUnreclaim: 7238080 kB .... SUnreclaim: 22307264 kB SUnreclaim: 22485888 kB SUnreclaim: 22720256 kB When testcases with T10 enabled call into __blkdev_direct_IO_simple, code doesn't free memory allocated by bio_integrity_alloc. The patch fixes the issue. HTX has been run with +60 hours without failure." Since __blkdev_direct_IO_simple() allocates the bio on the stack, it doesn't go through the regular bio free. This means that any ancillary data allocated with the bio through the stack is not freed. Hence, we can leak the integrity data associated with the bio, if the device is using DIF/DIX. Fix this by providing a bio_uninit() and export it, so that we can use it to free this data. Note that this is a minimal fix for this issue. Any current user of bio's that are allocated outside of bio_alloc_bioset() suffers from this issue, most notably some drivers. We will fix those in a more comprehensive patch for 4.13. This also means that the commit marked as being fixed by this isn't the real culprit, it's just the most obvious one out there. Fixes: 542ff7bf18c6 ("block: new direct I/O implementation") Reported-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2017-06-28nvme: Allocate queues for all possible CPUsChristoph Hellwig
Unlike most drіvers that simply pass the maximum possible vectors to pci_alloc_irq_vectors NVMe needs to configure the device before allocting the vectors, so it needs a manual update for the new scheme of using all present CPUs. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Jens Axboe <axboe@kernel.dk> Cc: Keith Busch <keith.busch@intel.com> Cc: linux-block@vger.kernel.org Cc: linux-nvme@lists.infradead.org Link: http://lkml.kernel.org/r/20170626102058.10200-4-hch@lst.de Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
2017-06-28blk-mq: Create hctx for each present CPUChristoph Hellwig
Currently we only create hctx for online CPUs, which can lead to a lot of churn due to frequent soft offline / online operations. Instead allocate one for each present CPU to avoid this and dramatically simplify the code. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Jens Axboe <axboe@kernel.dk> Cc: Keith Busch <keith.busch@intel.com> Cc: linux-block@vger.kernel.org Cc: linux-nvme@lists.infradead.org Link: http://lkml.kernel.org/r/20170626102058.10200-3-hch@lst.de Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
2017-06-28blk-mq: Include all present CPUs in the default queue mappingChristoph Hellwig
This way we get a nice distribution independent of the current cpu online / offline state. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Jens Axboe <axboe@kernel.dk> Cc: Keith Busch <keith.busch@intel.com> Cc: linux-block@vger.kernel.org Cc: linux-nvme@lists.infradead.org Link: http://lkml.kernel.org/r/20170626102058.10200-2-hch@lst.de Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
2017-06-28x86/PCI: Select CONFIG_PCI_LOCKLESS_CONFIGThomas Gleixner
All x86 PCI configuration space accessors have either their own serialization or can operate completely lockless (ECAM). Disable the global lock in the generic PCI configuration space accessors. Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Acked-by: Bjorn Helgaas <helgaas@kernel.org> Cc: Andi Kleen <ak@linux.intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Stephane Eranian <eranian@google.com> Cc: Borislav Petkov <bp@alien8.de> Cc: linux-pci@vger.kernel.org Link: http://lkml.kernel.org/r/20170316215057.295079391@linutronix.de Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
2017-06-28PCI: Provide Kconfig option for lockless config space accessorsThomas Gleixner
The generic PCI configuration space accessors are globally serialized via pci_lock. On larger systems this causes massive lock contention when the configuration space has to be accessed frequently. One such access pattern is the Intel Uncore performance counter unit. Provide a kernel config option which can be selected by an architecture when the low level PCI configuration space accessors in the architecture use their own serialization or can operate completely lockless. Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Acked-by: Bjorn Helgaas <helgaas@kernel.org> Cc: Andi Kleen <ak@linux.intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Stephane Eranian <eranian@google.com> Cc: Borislav Petkov <bp@alien8.de> Cc: linux-pci@vger.kernel.org Link: http://lkml.kernel.org/r/20170316215057.205961140@linutronix.de Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
2017-06-28x86/PCI/ce4100: Properly lock accessor functionsThomas Gleixner
x86 wants to get rid of the global pci_lock protecting the config space accessors so ECAM mode can operate completely lockless, but the CE4100 PCI code relies on that to protect the simulation registers. Restructure the code so it uses the x86 specific pci_config_lock to serialize the inner workings of the CE4100 PCI magic. That allows to remove the global locking via pci_lock later. Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Acked-by: Bjorn Helgaas <helgaas@kernel.org> Cc: Andi Kleen <ak@linux.intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Stephane Eranian <eranian@google.com> Cc: Borislav Petkov <bp@alien8.de> Cc: linux-pci@vger.kernel.org Link: http://lkml.kernel.org/r/20170316215057.126873574@linutronix.de Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
2017-06-28x86/PCI: Abort if legacy init failsThomas Gleixner
If the legacy PCI init fails, then there are no PCI config space accesors available, but the code continues and tries to scan the busses, which fails due to the lack of config space accessors. Return right away, if the last init fallback fails. Switch the few printks to pr_info while at it. Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Acked-by: Bjorn Helgaas <helgaas@kernel.org> Cc: Andi Kleen <ak@linux.intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Stephane Eranian <eranian@google.com> Cc: Borislav Petkov <bp@alien8.de> Cc: linux-pci@vger.kernel.org Link: http://lkml.kernel.org/r/20170316215057.047576516@linutronix.de Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
generated by cgit (git 2.34.1) at 2025-09-13 00:32:43 +0000