summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2024-10-219p: fix slab cache name creation for realLinus Torvalds
This was attempted by using the dev_name in the slab cache name, but as Omar Sandoval pointed out, that can be an arbitrary string, eg something like "/dev/root". Which in turn trips verify_dirent_name(), which fails if a filename contains a slash. So just make it use a sequence counter, and make it an atomic_t to avoid any possible races or locking issues. Reported-and-tested-by: Omar Sandoval <osandov@fb.com> Link: https://lore.kernel.org/all/ZxafcO8KWMlXaeWE@telecaster.dhcp.thefacebook.com/ Fixes: 79efebae4afc ("9p: Avoid creating multiple slab caches with the same name") Acked-by: Vlastimil Babka <vbabka@suse.cz> Cc: Dominique Martinet <asmadeus@codewreck.org> Cc: Thorsten Leemhuis <regressions@leemhuis.info> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2024-10-21x86/lam: Disable ADDRESS_MASKING in most casesPawan Gupta
Linear Address Masking (LAM) has a weakness related to transient execution as described in the SLAM paper[1]. Unless Linear Address Space Separation (LASS) is enabled this weakness may be exploitable. Until kernel adds support for LASS[2], only allow LAM for COMPILE_TEST, or when speculation mitigations have been disabled at compile time, otherwise keep LAM disabled. There are no processors in market that support LAM yet, so currently nobody is affected by this issue. [1] SLAM: https://download.vusec.net/papers/slam_sp24.pdf [2] LASS: https://lore.kernel.org/lkml/20230609183632.48706-1-alexander.shishkin@linux.intel.com/ [ dhansen: update SPECULATION_MITIGATIONS -> CPU_MITIGATIONS ] Signed-off-by: Pawan Gupta <pawan.kumar.gupta@linux.intel.com> Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com> Reviewed-by: Sohil Mehta <sohil.mehta@intel.com> Acked-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Cc:stable@vger.kernel.org Link: https://lore.kernel.org/all/5373262886f2783f054256babdf5a98545dc986b.1706068222.git.pawan.kumar.gupta%40linux.intel.com
2024-10-21Merge tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvmLinus Torvalds
Pull kvm fixes from Paolo Bonzini: "ARM64: - Fix the guest view of the ID registers, making the relevant fields writable from userspace (affecting ID_AA64DFR0_EL1 and ID_AA64PFR1_EL1) - Correcly expose S1PIE to guests, fixing a regression introduced in 6.12-rc1 with the S1POE support - Fix the recycling of stage-2 shadow MMUs by tracking the context (are we allowed to block or not) as well as the recycling state - Address a couple of issues with the vgic when userspace misconfigures the emulation, resulting in various splats. Headaches courtesy of our Syzkaller friends - Stop wasting space in the HYP idmap, as we are dangerously close to the 4kB limit, and this has already exploded in -next - Fix another race in vgic_init() - Fix a UBSAN error when faking the cache topology with MTE enabled RISCV: - RISCV: KVM: use raw_spinlock for critical section in imsic x86: - A bandaid for lack of XCR0 setup in selftests, which causes trouble if the compiler is configured to have x86-64-v3 (with AVX) as the default ISA. Proper XCR0 setup will come in the next merge window. - Fix an issue where KVM would not ignore low bits of the nested CR3 and potentially leak up to 31 bytes out of the guest memory's bounds - Fix case in which an out-of-date cached value for the segments could by returned by KVM_GET_SREGS. - More cleanups for KVM_X86_QUIRK_SLOT_ZAP_ALL - Override MTRR state for KVM confidential guests, making it WB by default as is already the case for Hyper-V guests. Generic: - Remove a couple of unused functions" * tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm: (27 commits) RISCV: KVM: use raw_spinlock for critical section in imsic KVM: selftests: Fix out-of-bounds reads in CPUID test's array lookups KVM: selftests: x86: Avoid using SSE/AVX instructions KVM: nSVM: Ignore nCR3[4:0] when loading PDPTEs from memory KVM: VMX: reset the segment cache after segment init in vmx_vcpu_reset() KVM: x86: Clean up documentation for KVM_X86_QUIRK_SLOT_ZAP_ALL KVM: x86/mmu: Add lockdep assert to enforce safe usage of kvm_unmap_gfn_range() KVM: x86/mmu: Zap only SPs that shadow gPTEs when deleting memslot x86/kvm: Override default caching mode for SEV-SNP and TDX KVM: Remove unused kvm_vcpu_gfn_to_pfn_atomic KVM: Remove unused kvm_vcpu_gfn_to_pfn KVM: arm64: Ensure vgic_ready() is ordered against MMIO registration KVM: arm64: vgic: Don't check for vgic_ready() when setting NR_IRQS KVM: arm64: Fix shift-out-of-bounds bug KVM: arm64: Shave a few bytes from the EL2 idmap code KVM: arm64: Don't eagerly teardown the vgic on init error KVM: arm64: Expose S1PIE to guests KVM: arm64: nv: Clarify safety of allowing TLBI unmaps to reschedule KVM: arm64: nv: Punt stage-2 recycling to a vCPU request KVM: arm64: nv: Do not block when unmapping stage-2 if disallowed ...
2024-10-21dt-bindings: iio: adc: add ad7779 docRamona Alexandra Nechita
Add dt bindings for AD7779 8-channel, simultaneous sampling ADC family with eight full Σ-Δ ADCs on chip and ultra-low input current to allow direct sensor connection. Signed-off-by: Ramona Alexandra Nechita <ramona.nechita@analog.com> Reviewed-by: Conor Dooley <conor.dooley@microchip.com> Link: https://patch.msgid.link/20241014143204.30195-2-ramona.nechita@analog.com Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2024-10-21iio: adc: ad7606: Disable PWM usage for non backend versionGuillaume Stols
Since the pwm was introduced before backend, there was a mock use, with a GPIO emulation. Now that iio backend is introduced, the mock use can be removed. Signed-off-by: Guillaume Stols <gstols@baylibre.com> Link: https://patch.msgid.link/20241015-ad7606_add_iio_backend_support-v5-8-654faf1ae08c@baylibre.com Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2024-10-21iio: adc: ad7606: Add iio-backend supportGuillaume Stols
- Basic support for iio backend. - Supports IIO_CHAN_INFO_SAMP_FREQ R/W. - Only hardware mode is available, and that IIO_CHAN_INFO_RAW is not supported if iio-backend mode is selected. Signed-off-by: Guillaume Stols <gstols@baylibre.com> Link: https://patch.msgid.link/20241015-ad7606_add_iio_backend_support-v5-7-654faf1ae08c@baylibre.com Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2024-10-21iio: adc: ad7606: Introduce num_adc_channelsGuillaume Stols
This variable determines how many hardware channels has the chip, oppositely to the num_channels that can contain more channels, e.g a timestamp channel in our case. Introducing this variable avoids decreasing the former num_channels variable when reading the ADC's channels, and clarifies a bit the code. Reviewed-by: Nuno Sa <nuno.sa@analog.com> Signed-off-by: Guillaume Stols <gstols@baylibre.com> Link: https://patch.msgid.link/20241015-ad7606_add_iio_backend_support-v5-6-654faf1ae08c@baylibre.com Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2024-10-21iio: adc: ad7606: Add compatibility to fw_nodesGuillaume Stols
On the parallel version, the current implementation is only compatible with id tables and won't work with fw_nodes, this commit intends to fix it. Doing so required to declare ad7606_chip_info structures in the .h file so to make them accessible to all the driver files that can set a pointer to the corresponding chip as the driver data. Signed-off-by: Guillaume Stols <gstols@baylibre.com> Link: https://patch.msgid.link/20241015-ad7606_add_iio_backend_support-v5-5-654faf1ae08c@baylibre.com Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2024-10-21iio: adc: ad7606: Add PWM support for conversion triggerGuillaume Stols
Until now, the conversion were triggered by setting high the GPIO connected to the convst pin. This commit gives the possibility to connect the convst pin to a PWM. Connecting a PWM allows to have a better control on the samplerate, but it must be handled with care, as it is completely decorrelated of the driver's busy pin handling. Hence it is not recommended to be used "as is" but must be exploited in conjunction with IIO backend, and for now only a mock functionality is enabled, i.e PWM never swings, but is used as a GPIO, i.e duty_cycle == period equals high state, duty_cycle == 0 equals low state. This mock functionality will be disabled after the IIO backend usecase is introduced. Signed-off-by: Guillaume Stols <gstols@baylibre.com> Link: https://patch.msgid.link/20241015-ad7606_add_iio_backend_support-v5-4-654faf1ae08c@baylibre.com Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2024-10-21Documentation: iio: Document ad7606 driverGuillaume Stols
The Analog Devices Inc. AD7606 (and similar chips) are complex ADCs that will benefit from a detailed driver documentation. This documents the current features supported by the driver. Signed-off-by: Guillaume Stols <gstols@baylibre.com> Link: https://patch.msgid.link/20241015-ad7606_add_iio_backend_support-v5-3-654faf1ae08c@baylibre.com Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2024-10-21dt-bindings: iio: adc: ad7606: Add iio backend bindingsGuillaume Stols
Add the required properties for iio-backend support, as well as an example and the conditions to mutually exclude interruption and conversion trigger with iio-backend. The iio-backend's function is to controls the communication, and thus the interruption pin won't be available anymore. As a consequence, the conversion pin must be controlled externally since we will miss information about when every single conversion cycle (i.e conversion + data transfer) ends, hence a PWM is introduced to trigger the conversions. Reviewed-by: Rob Herring (Arm) <robh@kernel.org> Signed-off-by: Guillaume Stols <gstols@baylibre.com> Link: https://patch.msgid.link/20241015-ad7606_add_iio_backend_support-v5-2-654faf1ae08c@baylibre.com Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2024-10-21dt-bindings: iio: adc: ad7606: Remove spi-cpha from requiredGuillaume Stols
The documentation is erroneously stating that spi-cpha is required, and the example is erroneously setting both spi-cpol and spi-cpha. According to the datasheet, only cpol should be set. On zedboard for instance, setting the devicetree as in the example will simply not work. Fixes: 416f882c3b40 ("dt-bindings: iio: adc: Migrate AD7606 documentation to yaml") Fixes: 6e33a125df66 ("dt-bindings: iio: adc: Add docs for AD7606 ADC") Reviewed-by: Rob Herring (Arm) <robh@kernel.org> Signed-off-by: Guillaume Stols <gstols@baylibre.com> Link: https://patch.msgid.link/20241015-ad7606_add_iio_backend_support-v5-1-654faf1ae08c@baylibre.com Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2024-10-21iio: addac: ad74413r: simplify with cleanup.hNuno Sa
Make use of mutex guard() and IIO iio_device_claim_direct_scoped() to simplify code and error handling. While at it, use devm_mutex_init() to initialize the mutex. Signed-off-by: Nuno Sa <nuno.sa@analog.com> Link: https://patch.msgid.link/20241016-dev-ad74413r-minor-improv-v1-3-13c9c769237d@analog.com Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2024-10-21iio: addac: ad74413r: use devm_regulator_get_enable_read_voltage()Nuno Sa
It's highly unlikely for the converter ref voltage to change at runtime. Hence, let's read the voltage and save it (instead of the regulator struct). While at it, simplify the code by using devm_regulator_get_enable_read_voltage(). Signed-off-by: Nuno Sa <nuno.sa@analog.com> Link: https://patch.msgid.link/20241016-dev-ad74413r-minor-improv-v1-2-13c9c769237d@analog.com Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2024-10-21iio: addac: ad74413r: drop reset_gpio from struct ad74413r_stateNuno Sa
We just need the reset gpio during probe so there's no need to keep it in our state struct. Hence, move devm_gpiod_get_optional() into ad74413r_reset() and use a local struct gpio_desc. While at it, request the gpio in the asserted state (GPIOD_OUT_HIGH) so that we already perform the reset while requesting the gpio saving us a call to gpiod_set_value_cansleep(). Also, explicitly include <gpio/consumer.h> for devm_gpiod_get_optional(). Signed-off-by: Nuno Sa <nuno.sa@analog.com> Link: https://patch.msgid.link/20241016-dev-ad74413r-minor-improv-v1-1-13c9c769237d@analog.com Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2024-10-21iio: light: veml6070: use field to set integration timeJavier Carrasco
Define the integration time within the configuration register as a field to easy its handling as an index, preparing the driver to support configurable integration times. Signed-off-by: Javier Carrasco <javier.carrasco.cruz@gmail.com> Link: https://patch.msgid.link/20241017-veml6070-integration-time-v1-2-3507d17d562a@gmail.com Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2024-10-21iio: light: veml6070: use unsigned int instead of unsignedJavier Carrasco
Trivial modification to use the recommended keyword 'int' after 'unsigned' for unsigned integers. Signed-off-by: Javier Carrasco <javier.carrasco.cruz@gmail.com> Link: https://patch.msgid.link/20241017-veml6070-integration-time-v1-1-3507d17d562a@gmail.com Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2024-10-21iio: adc: ad7192: properly check spi_get_device_match_data()Nuno Sa
spi_get_device_match_data() can return a NULL pointer. Hence, let's check for it. Signed-off-by: Nuno Sa <nuno.sa@analog.com> Link: https://patch.msgid.link/20241014-fix-error-check-v1-1-089e1003d12f@analog.com Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2024-10-21iio: pressure: bmp280: Move bmp085 interrupt to new configurationVasileios Amoiridis
This commit intends to add the old BMP085 sensor to the new IRQ interface of the driver for consistence. No functional changes intended. The BMP085 sensor is equivalent with the BMP180 with the only difference of BMP085 having an extra interrupt pin to inform about an End of Conversion. Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Signed-off-by: Vasileios Amoiridis <vassilisamir@gmail.com> Link: https://patch.msgid.link/20241017233022.238250-5-vassilisamir@gmail.com Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2024-10-21iio: pressure: bmp280: Add data ready trigger supportVasileios Amoiridis
The BMP3xx and BMP5xx sensors have an interrupt pin which can be used as a trigger for when there are data ready in the sensor for pick up. This use case is used along with NORMAL_MODE in the sensor, which allows the sensor to do consecutive measurements depending on the ODR rate value. The trigger pin can be configured to be open-drain or push-pull and either rising or falling edge. No support is added yet for interrupts for FIFO, WATERMARK and out of range values. Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Signed-off-by: Vasileios Amoiridis <vassilisamir@gmail.com> Link: https://patch.msgid.link/20241017233022.238250-4-vassilisamir@gmail.com Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2024-10-21dt-bindings: iio: pressure: bmp085: Add interrupts for BMP3xx and BMP5xx devicesVasileios Amoiridis
Add interrupt options for BMP3xx and BMP5xx devices as well. Acked-by: Conor Dooley <conor.dooley@microchip.com> Signed-off-by: Vasileios Amoiridis <vassilisamir@gmail.com> Link: https://patch.msgid.link/20241017233022.238250-3-vassilisamir@gmail.com Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2024-10-21iio: pressure: bmp280: Use sleep and forced mode for oneshot capturesVasileios Amoiridis
Add forced mode support in sensors BMP28x, BME28x, BMP3xx and BMP58x. Sensors BMP18x and BMP085 are old and do not support this feature so their operation is not affected at all. Essentially, up to now, the rest of the sensors were used in normal mode all the time. This means that they are continuously doing measurements even though these measurements are not used. Even though the sensor does provide PM support, to cover all the possible use cases, the sensor needs to go into sleep mode and wake up whenever necessary. The idea is that the sensor is by default in sleep mode, wakes up in forced mode when a oneshot capture is requested, or in normal mode when the buffer is enabled. The difference lays in the fact that in forced mode, the sensor does only one conversion and goes back to sleep while in normal mode, the sensor does continuous measurements with the frequency that was set in the ODR registers. The bmpX_chip_config() functions which are responsible for applying the requested configuration to the sensor, are modified accordingly in order to set the sensor by default in sleep mode. DEEP STANDBY, Low Power NORMAL and CONTINUOUS modes, supported only by the BMP58x version, are not added. Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Signed-off-by: Vasileios Amoiridis <vassilisamir@gmail.com> Link: https://patch.msgid.link/20241017233022.238250-2-vassilisamir@gmail.com Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2024-10-21iio: accel: adxl355: Fix typo "accelaration"WangYuli
There is a spelling mistake of 'accelaration' in comments which should be 'acceleration'. Signed-off-by: WangYuli <wangyuli@uniontech.com> Link: https://patch.msgid.link/9F137828F9F185FD+20241017092221.361511-1-wangyuli@uniontech.com Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2024-10-21iio: dac: ad8460: add SPI device match tableDavid Lechner
Add SPI device match table for ADI AD8460 DAC. As described in [1], this is required for the module to automatically load, even when using DT. [1]: https://lore.kernel.org/all/20210921192149.50740-1-broonie@kernel.org/ Signed-off-by: David Lechner <dlechner@baylibre.com> Link: https://patch.msgid.link/20241018-iio-dac-ad8460-add-spi-match-table-v1-1-84a5f903bf50@baylibre.com Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2024-10-21iio: gyro: list adis16137 in Kconfig descriptionYo-Jung (Leo) Lin
The adis16136 driver also supports the adis16137 model, but it is not mentioned in the Kconfig help. Add it to the description in Kconfig. Signed-off-by: Yo-Jung Lin (Leo) <0xff07@gmail.com> Link: https://patch.msgid.link/20241019034213.429464-1-0xff07@gmail.com Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2024-10-21iio: dac: ad8460: fix DT compatibleDavid Lechner
Fix the DT compatible string in the of_device_id table to match the binding documentation. There should not be a space after the comma. Fixes: a976ef24c625 ("iio: dac: support the ad8460 Waveform DAC") Signed-off-by: David Lechner <dlechner@baylibre.com> Link: https://patch.msgid.link/20241018-iio-adc-ad8460-fix-dt-compatible-v1-1-058231638527@baylibre.com Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2024-10-21iio: magnetometer: bmc150_magn: Drop most likely fake ACPI IDsAndy Shevchenko
The commits in question do not proove that ACPI IDs exist. Quite likely it was a cargo cult addition while doing that for DT-based enumeration. Drop most likely fake ACPI IDs. The to be removed IDs has been checked against the following resources: 1) DuckDuckGo 2) Google 3) MS catalog: https://www.catalog.update.microsoft.com/Search.aspx This gives no useful results in regard to DSDT, moreover, the official vendor IDs in the registry for Bosh are BSG and BOSC. Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Reviewed-by: Hans de Goede <hdegoede@redhat.com> Link: https://patch.msgid.link/20241018145805.2181682-1-andriy.shevchenko@linux.intel.com Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2024-10-21iio: gyro: bmg160: Drop most likely fake ACPI IDsAndy Shevchenko
The commits in question do not proove that ACPI IDs exist. Quite likely it was a cargo cult addition while doing that for DT-based enumeration. Drop most likely fake ACPI IDs. The to be removed IDs has been checked against the following resources: 1) DuckDuckGo 2) Google 3) MS catalog: https://www.catalog.update.microsoft.com/Search.aspx This gives no useful results in regard to DSDT, moreover, the official vendor IDs in the registry for Bosh are BSG and BOSC. Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Reviewed-by: Hans de Goede <hdegoede@redhat.com> Link: https://patch.msgid.link/20241018145732.2181309-1-andriy.shevchenko@linux.intel.com Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2024-10-21iio: imu: smi240: add driverShen Jianping
add the iio driver for bosch imu smi240. The smi240 is a combined three axis angular rate and three axis acceleration sensor module with a measurement range of +/-300°/s and up to 16g. A synchronous acc and gyro sampling can be triggered by setting the capture bit in spi read command. Implemented features: * raw data access for each axis through sysfs * tiggered buffer for continuous sampling * synchronous acc and gyro data from tiggered buffer Signed-off-by: Shen Jianping <Jianping.Shen@de.bosch.com> Link: https://patch.msgid.link/20241018135234.5446-3-Jianping.Shen@de.bosch.com Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2024-10-21dt-bindings: iio: imu: smi240: add Bosch smi240Shen Jianping
add devicetree binding for Bosch imu smi240. The smi240 is a combined three axis angular rate and three axis acceleration sensor module. * The smi240 requires VDD and VDDIO * Provides only spi interface. Reviewed-by: Rob Herring (Arm) <robh@kernel.org> Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org> Reviewed-by: Conor Dooley <conor.dooley@microchip.com> Signed-off-by: Shen Jianping <Jianping.Shen@de.bosch.com> Link: https://patch.msgid.link/20241018135234.5446-2-Jianping.Shen@de.bosch.com Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2024-10-21Merge tag 'probes-fixes-v6.12-rc4' of ↵Linus Torvalds
git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace Pull uprobe fix from Masami Hiramatsu: - uprobe: avoid out-of-bounds memory access of fetching args Uprobe trace events can cause out-of-bounds memory access when fetching user-space data which is bigger than one page, because it does not check the local CPU buffer size when reading the data. This checks the read data size and cut it down to the local CPU buffer size. * tag 'probes-fixes-v6.12-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace: uprobe: avoid out-of-bounds memory access of fetching args
2024-10-21phy: tegra: xusb: Add error pointer check in xusb.cDipendra Khadka
Add error pointer check after tegra_xusb_find_lane(). Fixes: e8f7d2f409a1 ("phy: tegra: xusb: Add usb-phy support") Signed-off-by: Dipendra Khadka <kdipendra88@gmail.com> Acked-by: Thierry Reding <treding@nvidia.com> Link: https://lore.kernel.org/r/20240930191101.13184-1-kdipendra88@gmail.com Signed-off-by: Vinod Koul <vkoul@kernel.org>
2024-10-21dt-bindings: phy: qcom,sc8280xp-qmp-pcie-phy: Fix X1E80100 resets entriesAbel Vesa
The PCIe 6a PHY is actually Gen4 4-lanes capable. So the gen4x4 compatible describes it. But according to the schema, currently the gen4x4 compatible doesn't require both PHY and PHY-nocsr resets, while the HW does. So fix that by adding the gen4x4 compatible alongside the gen4x2 one for the resets description. Fixes: 0c5f4d23f776 ("dt-bindings: phy: qcom,sc8280xp-qmp-pcie-phy: Document the X1E80100 QMP PCIe PHY Gen4 x4") Reported-by: kernel test robot <lkp@intel.com> Closes: https://lore.kernel.org/oe-kbuild-all/202410182029.n2zPkuGx-lkp@intel.com/ Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org> Reviewed-by: Johan Hovold <johan+linaro@kernel.org> Signed-off-by: Abel Vesa <abel.vesa@linaro.org> Link: https://lore.kernel.org/r/20241021-phy-qcom-qmp-pcie-fix-x1e80100-gen4x4-resets-v3-1-1918c46fc37c@linaro.org Signed-off-by: Vinod Koul <vkoul@kernel.org>
2024-10-21phy: freescale: imx8m-pcie: Do CMN_RST just before PHY PLL lock checkRichard Zhu
When enable initcall_debug together with higher debug level below. CONFIG_CONSOLE_LOGLEVEL_DEFAULT=9 CONFIG_CONSOLE_LOGLEVEL_QUIET=9 CONFIG_MESSAGE_LOGLEVEL_DEFAULT=7 The initialization of i.MX8MP PCIe PHY might be timeout failed randomly. To fix this issue, adjust the sequence of the resets refer to the power up sequence listed below. i.MX8MP PCIe PHY power up sequence: /--------------------------------------------- 1.8v supply ---------/ /--------------------------------------------------- 0.8v supply ---/ ---\ /-------------------------------------------------- X REFCLK Valid Reference Clock ---/ \-------------------------------------------------- ------------------------------------------- | i_init_restn -------------- ------------------------------------ | i_cmn_rstn --------------------- ------------------------------- | o_pll_lock_done -------------------------- Logs: imx6q-pcie 33800000.pcie: host bridge /soc@0/pcie@33800000 ranges: imx6q-pcie 33800000.pcie: IO 0x001ff80000..0x001ff8ffff -> 0x0000000000 imx6q-pcie 33800000.pcie: MEM 0x0018000000..0x001fefffff -> 0x0018000000 probe of clk_imx8mp_audiomix.reset.0 returned 0 after 1052 usecs probe of 30e20000.clock-controller returned 0 after 32971 usecs phy phy-32f00000.pcie-phy.4: phy poweron failed --> -110 probe of 30e10000.dma-controller returned 0 after 10235 usecs imx6q-pcie 33800000.pcie: waiting for PHY ready timeout! dwhdmi-imx 32fd8000.hdmi: Detected HDMI TX controller v2.13a with HDCP (samsung_dw_hdmi_phy2) imx6q-pcie 33800000.pcie: probe with driver imx6q-pcie failed with error -110 Fixes: dce9edff16ee ("phy: freescale: imx8m-pcie: Add i.MX8MP PCIe PHY support") Cc: stable@vger.kernel.org Signed-off-by: Richard Zhu <hongxing.zhu@nxp.com> Signed-off-by: Frank Li <Frank.Li@nxp.com> v2 changes: - Rebase to latest fixes branch of linux-phy git repo. - Richard's environment have problem and can't sent out patch. So I help post this fix patch. Link: https://lore.kernel.org/r/20241021155241.943665-1-Frank.Li@nxp.com Signed-off-by: Vinod Koul <vkoul@kernel.org>
2024-10-21Merge tag 'vfs-6.12-rc5.fixes' of ↵Linus Torvalds
git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs Pull vfs fixes from Christian Brauner: "afs: - Fix a lock recursion in afs_wake_up_async_call() on ->notify_lock netfs: - Drop the references to a folio immediately after the folio has been extracted to prevent races with future I/O collection - Fix a documenation build error - Downgrade the i_rwsem for buffered writes to fix a cifs reported performance regression when switching to netfslib vfs: - Explicitly return -E2BIG from openat2() if the specified size is unexpectedly large. This aligns openat2() with other extensible struct based system calls - When copying a mount namespace ensure that we only try to remove the new copy from the mount namespace rbtree if it has already been added to it nilfs: - Clear the buffer delay flag when clearing the buffer state clags when a buffer head is discarded to prevent a kernel OOPs ocfs2: - Fix an unitialized value warning in ocfs2_setattr() proc: - Fix a kernel doc warning" * tag 'vfs-6.12-rc5.fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs: proc: Fix W=1 build kernel-doc warning afs: Fix lock recursion fs: Fix uninitialized value issue in from_kuid and from_kgid fs: don't try and remove empty rbtree node netfs: Downgrade i_rwsem for a buffered write nilfs2: fix kernel bug due to missing clearing of buffer delay flag openat2: explicitly return -E2BIG for (usize > PAGE_SIZE) netfs: fix documentation build error netfs: In readahead, put the folio refs as soon extracted
2024-10-21Merge tag 'v6.12-p4' of ↵Linus Torvalds
git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6 Pull crypto fix from Herbert Xu: "Fix a regression in mpi that broke RSA" * tag 'v6.12-p4' of git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6: crypto: lib/mpi - Fix an "Uninitialized scalar variable" issue
2024-10-21RDMA/bnxt_re: synchronize the qp-handle table arraySelvin Xavier
There is a race between the CREQ tasklet and destroy qp when accessing the qp-handle table. There is a chance of reading a valid qp-handle in the CREQ tasklet handler while the QP is already moving ahead with the destruction. Fixing this race by implementing a table-lock to synchronize the access. Fixes: f218d67ef004 ("RDMA/bnxt_re: Allow posting when QPs are in error") Fixes: 84cf229f4001 ("RDMA/bnxt_re: Fix the qp table indexing") Link: https://patch.msgid.link/r/1728912975-19346-3-git-send-email-selvin.xavier@broadcom.com Signed-off-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com> Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2024-10-21RDMA/bnxt_re: Fix the usage of control path spin locksSelvin Xavier
Control path completion processing always runs in tasklet context. To synchronize with the posting thread, there is no need to use the irq variant of spin lock. Use spin_lock_bh instead. Fixes: 1ac5a4047975 ("RDMA/bnxt_re: Add bnxt_re RoCE driver") Link: https://patch.msgid.link/r/1728912975-19346-2-git-send-email-selvin.xavier@broadcom.com Signed-off-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com> Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2024-10-21RDMA/mlx5: Round max_rd_atomic/max_dest_rd_atomic up instead of downPatrisious Haddad
After the cited commit below max_dest_rd_atomic and max_rd_atomic values are being rounded down to the next power of 2. As opposed to the old behavior and mlx4 driver where they used to be rounded up instead. In order to stay consistent with older code and other drivers, revert to using fls round function which rounds up to the next power of 2. Fixes: f18e26af6aba ("RDMA/mlx5: Convert modify QP to use MLX5_SET macros") Link: https://patch.msgid.link/r/d85515d6ef21a2fa8ef4c8293dce9b58df8a6297.1728550179.git.leon@kernel.org Signed-off-by: Patrisious Haddad <phaddad@nvidia.com> Reviewed-by: Maher Sanalla <msanalla@nvidia.com> Signed-off-by: Leon Romanovsky <leonro@nvidia.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2024-10-21RDMA/cxgb4: Dump vendor specific QP detailsLeon Romanovsky
Restore the missing functionality to dump vendor specific QP details, which was mistakenly removed in the commit mentioned in Fixes line. Fixes: 5cc34116ccec ("RDMA: Add dedicated QP resource tracker function") Link: https://patch.msgid.link/r/ed9844829135cfdcac7d64285688195a5cd43f82.1728323026.git.leonro@nvidia.com Reported-by: Dr. David Alan Gilbert <linux@treblig.org> Closes: https://lore.kernel.org/all/Zv_4qAxuC0dLmgXP@gallifrey Signed-off-by: Leon Romanovsky <leonro@nvidia.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2024-10-21iomap: turn iomap_want_unshare_iter into an inline functionChristoph Hellwig
iomap_want_unshare_iter currently sits in fs/iomap/buffered-io.c, which depends on CONFIG_BLOCK. It is also in used in fs/dax.c whіch has no such dependency. Given that it is a trivial check turn it into an inline in include/linux/iomap.h to fix the DAX && !BLOCK build. Fixes: 6ef6a0e821d3 ("iomap: share iomap_unshare_iter predicate code with fsdax") Reported-by: kernel test robot <lkp@intel.com> Signed-off-by: Christoph Hellwig <hch@lst.de> Link: https://lore.kernel.org/r/20241015041350.118403-1-hch@lst.de Reviewed-by: Brian Foster <bfoster@redhat.com> Signed-off-by: Christian Brauner <brauner@kernel.org>
2024-10-21spi: mtk-snfi: fix kerneldoc for mtk_snand_is_page_ops()Bartosz Golaszewski
The op argument is missing the colon and is not picked up by the kerneldoc generator. Fix it to address the following build warning: drivers/spi/spi-mtk-snfi.c:1201: warning: Function parameter or struct member 'op' not described in 'mtk_snand_is_page_ops' Fixes: 764f1b748164 ("spi: add driver for MTK SPI NAND Flash Interface") Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org> Link: https://patch.msgid.link/20241021142113.71081-1-brgl@bgdev.pl Signed-off-by: Mark Brown <broonie@kernel.org>
2024-10-21platform/x86: dell-wmi: Ignore suspend notificationsArmin Wolf
Some machines like the Dell G15 5155 emit WMI events when suspending/resuming. Ignore those WMI events. Tested-by: siddharth.manthan@gmail.com Signed-off-by: Armin Wolf <W_Armin@gmx.de> Acked-by: Pali Rohár <pali@kernel.org> Link: https://lore.kernel.org/r/20241014220529.397390-1-W_Armin@gmx.de Reviewed-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Hans de Goede <hdegoede@redhat.com>
2024-10-21platform/x86/intel/pmc: Fix pmc_core_iounmap to call iounmap for valid addressesVamsi Krishna Brahmajosyula
Commit 50c6dbdfd16e ("x86/ioremap: Improve iounmap() address range checks") introduces a WARN when adrress ranges of iounmap are invalid. On Thinkpad P1 Gen 7 (Meteor Lake-P) this caused the following warning to appear: WARNING: CPU: 7 PID: 713 at arch/x86/mm/ioremap.c:461 iounmap+0x58/0x1f0 Modules linked in: rfkill(+) snd_timer(+) fjes(+) snd soundcore intel_pmc_core(+) int3403_thermal(+) int340x_thermal_zone intel_vsec pmt_telemetry acpi_pad pmt_class acpi_tad int3400_thermal acpi_thermal_rel joydev loop nfnetlink zram xe drm_suballoc_helper nouveau i915 mxm_wmi drm_ttm_helper gpu_sched drm_gpuvm drm_exec drm_buddy i2c_algo_bit crct10dif_pclmul crc32_pclmul ttm crc32c_intel polyval_clmulni rtsx_pci_sdmmc ucsi_acpi polyval_generic mmc_core hid_multitouch drm_display_helper ghash_clmulni_intel typec_ucsi nvme sha512_ssse3 video sha256_ssse3 nvme_core intel_vpu sha1_ssse3 rtsx_pci cec typec nvme_auth i2c_hid_acpi i2c_hid wmi pinctrl_meteorlake serio_raw ip6_tables ip_tables fuse CPU: 7 UID: 0 PID: 713 Comm: (udev-worker) Not tainted 6.12.0-rc2iounmap+ #42 Hardware name: LENOVO 21KWCTO1WW/21KWCTO1WW, BIOS N48ET19W (1.06 ) 07/18/2024 RIP: 0010:iounmap+0x58/0x1f0 Code: 85 6a 01 00 00 48 8b 05 e6 e2 28 04 48 39 c5 72 19 eb 26 cc cc cc 48 ba 00 00 00 00 00 00 32 00 48 8d 44 02 ff 48 39 c5 72 23 <0f> 0b 48 83 c4 08 5b 5d 41 5c c3 cc cc cc cc 48 ba 00 00 00 00 00 RSP: 0018:ffff888131eff038 EFLAGS: 00010207 RAX: ffffc90000000000 RBX: 0000000000000000 RCX: ffff888e33b80000 RDX: dffffc0000000000 RSI: ffff888e33bc29c0 RDI: 0000000000000000 RBP: 0000000000000000 R08: ffff8881598a8000 R09: ffff888e2ccedc10 R10: 0000000000000003 R11: ffffffffb3367634 R12: 00000000fe000000 R13: ffff888101d0da28 R14: ffffffffc2e437e0 R15: ffff888110b03b28 FS: 00007f3c1d4b3980(0000) GS:ffff888e33b80000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00005651cfc93578 CR3: 0000000124e4c002 CR4: 0000000000f70ef0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000ffff07f0 DR7: 0000000000000400 PKRU: 55555554 Call Trace: <TASK> ? __warn.cold+0xb6/0x176 ? iounmap+0x58/0x1f0 ? report_bug+0x1f4/0x2b0 ? handle_bug+0x58/0x90 ? exc_invalid_op+0x17/0x40 ? asm_exc_invalid_op+0x1a/0x20 ? iounmap+0x58/0x1f0 pmc_core_ssram_get_pmc+0x477/0x6c0 [intel_pmc_core] ? __pfx_pmc_core_ssram_get_pmc+0x10/0x10 [intel_pmc_core] ? __pfx_do_pci_enable_device+0x10/0x10 ? pci_wait_for_pending+0x60/0x110 ? pci_enable_device_flags+0x1e3/0x2e0 ? __pfx_mtl_core_init+0x10/0x10 [intel_pmc_core] pmc_core_ssram_init+0x7f/0x110 [intel_pmc_core] mtl_core_init+0xda/0x130 [intel_pmc_core] ? __mutex_init+0xb9/0x130 pmc_core_probe+0x27e/0x10b0 [intel_pmc_core] ? _raw_spin_lock_irqsave+0x96/0xf0 ? __pfx_pmc_core_probe+0x10/0x10 [intel_pmc_core] ? __pfx_mutex_unlock+0x10/0x10 ? __pfx_mutex_lock+0x10/0x10 ? device_pm_check_callbacks+0x82/0x370 ? acpi_dev_pm_attach+0x234/0x2b0 platform_probe+0x9f/0x150 really_probe+0x1e0/0x8a0 __driver_probe_device+0x18c/0x370 ? __pfx___driver_attach+0x10/0x10 driver_probe_device+0x4a/0x120 __driver_attach+0x190/0x4a0 ? __pfx___driver_attach+0x10/0x10 bus_for_each_dev+0x103/0x180 ? __pfx_bus_for_each_dev+0x10/0x10 ? klist_add_tail+0x136/0x270 bus_add_driver+0x2fc/0x540 driver_register+0x1a5/0x360 ? __pfx_pmc_core_driver_init+0x10/0x10 [intel_pmc_core] do_one_initcall+0xa4/0x380 ? __pfx_do_one_initcall+0x10/0x10 ? kasan_unpoison+0x44/0x70 do_init_module+0x296/0x800 load_module+0x5090/0x6ce0 ? __pfx_load_module+0x10/0x10 ? ima_post_read_file+0x193/0x200 ? __pfx_ima_post_read_file+0x10/0x10 ? rw_verify_area+0x152/0x4c0 ? kernel_read_file+0x257/0x750 ? __pfx_kernel_read_file+0x10/0x10 ? __pfx_filemap_get_read_batch+0x10/0x10 ? init_module_from_file+0xd1/0x130 init_module_from_file+0xd1/0x130 ? __pfx_init_module_from_file+0x10/0x10 ? __pfx__raw_spin_lock+0x10/0x10 ? __pfx_cred_has_capability.isra.0+0x10/0x10 idempotent_init_module+0x236/0x770 ? __pfx_idempotent_init_module+0x10/0x10 ? fdget+0x58/0x3f0 ? security_capable+0x7d/0x110 __x64_sys_finit_module+0xbe/0x130 do_syscall_64+0x82/0x160 ? __pfx_filemap_read+0x10/0x10 ? __pfx___fsnotify_parent+0x10/0x10 ? vfs_read+0x3a6/0xa30 ? vfs_read+0x3a6/0xa30 ? __seccomp_filter+0x175/0xc60 ? __pfx___seccomp_filter+0x10/0x10 ? fdget_pos+0x1ce/0x500 ? syscall_exit_to_user_mode_prepare+0x149/0x170 ? syscall_exit_to_user_mode+0x10/0x210 ? do_syscall_64+0x8e/0x160 ? switch_fpu_return+0xe3/0x1f0 ? syscall_exit_to_user_mode+0x1d5/0x210 ? do_syscall_64+0x8e/0x160 ? exc_page_fault+0x76/0xf0 entry_SYSCALL_64_after_hwframe+0x76/0x7e RIP: 0033:0x7f3c1d6d155d Code: ff c3 66 2e 0f 1f 84 00 00 00 00 00 90 f3 0f 1e fa 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d 83 58 0f 00 f7 d8 64 89 01 48 RSP: 002b:00007ffe6309db38 EFLAGS: 00000246 ORIG_RAX: 0000000000000139 RAX: ffffffffffffffda RBX: 0000557c212550a0 RCX: 00007f3c1d6d155d RDX: 0000000000000000 RSI: 00007f3c1cd943bd RDI: 0000000000000025 RBP: 00007ffe6309dbf0 R08: 00007f3c1d7c7b20 R09: 00007ffe6309db80 R10: 0000557c21255270 R11: 0000000000000246 R12: 00007f3c1cd943bd R13: 0000000000020000 R14: 0000557c21255c80 R15: 0000557c21255240 </TASK> no_free_ptr(tmp_ssram) sets tmp_ssram NULL while assigning ssram. pmc_core_iounmap calls iounmap unconditionally causing the above warning to appear during boot. Fix it by checking for a valid address before calling iounmap. Also in the function pmc_core_ssram_get_pmc return -ENOMEM when ioremap fails similar to other instances in the file. Fixes: a01486dc4bb1 ("platform/x86/intel/pmc: Cleanup SSRAM discovery") Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com> Reviewed-by: David E. Box <david.e.box@linux.intel.com> Signed-off-by: Vamsi Krishna Brahmajosyula <vamsikrishna.brahmajosyula@gmail.com> Link: https://lore.kernel.org/r/20241018104958.14195-1-vamsikrishna.brahmajosyula@gmail.com Reviewed-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Hans de Goede <hdegoede@redhat.com>
2024-10-21nfsd: cancel nfsd_shrinker_work using sync mode in nfs4_state_shutdown_netYang Erkun
In the normal case, when we excute `echo 0 > /proc/fs/nfsd/threads`, the function `nfs4_state_destroy_net` in `nfs4_state_shutdown_net` will release all resources related to the hashed `nfs4_client`. If the `nfsd_client_shrinker` is running concurrently, the `expire_client` function will first unhash this client and then destroy it. This can lead to the following warning. Additionally, numerous use-after-free errors may occur as well. nfsd_client_shrinker echo 0 > /proc/fs/nfsd/threads expire_client nfsd_shutdown_net unhash_client ... nfs4_state_shutdown_net /* won't wait shrinker exit */ /* cancel_work(&nn->nfsd_shrinker_work) * nfsd_file for this /* won't destroy unhashed client1 */ * client1 still alive nfs4_state_destroy_net */ nfsd_file_cache_shutdown /* trigger warning */ kmem_cache_destroy(nfsd_file_slab) kmem_cache_destroy(nfsd_file_mark_slab) /* release nfsd_file and mark */ __destroy_client ==================================================================== BUG nfsd_file (Not tainted): Objects remaining in nfsd_file on __kmem_cache_shutdown() -------------------------------------------------------------------- CPU: 4 UID: 0 PID: 764 Comm: sh Not tainted 6.12.0-rc3+ #1 dump_stack_lvl+0x53/0x70 slab_err+0xb0/0xf0 __kmem_cache_shutdown+0x15c/0x310 kmem_cache_destroy+0x66/0x160 nfsd_file_cache_shutdown+0xac/0x210 [nfsd] nfsd_destroy_serv+0x251/0x2a0 [nfsd] nfsd_svc+0x125/0x1e0 [nfsd] write_threads+0x16a/0x2a0 [nfsd] nfsctl_transaction_write+0x74/0xa0 [nfsd] vfs_write+0x1a5/0x6d0 ksys_write+0xc1/0x160 do_syscall_64+0x5f/0x170 entry_SYSCALL_64_after_hwframe+0x76/0x7e ==================================================================== BUG nfsd_file_mark (Tainted: G B W ): Objects remaining nfsd_file_mark on __kmem_cache_shutdown() -------------------------------------------------------------------- dump_stack_lvl+0x53/0x70 slab_err+0xb0/0xf0 __kmem_cache_shutdown+0x15c/0x310 kmem_cache_destroy+0x66/0x160 nfsd_file_cache_shutdown+0xc8/0x210 [nfsd] nfsd_destroy_serv+0x251/0x2a0 [nfsd] nfsd_svc+0x125/0x1e0 [nfsd] write_threads+0x16a/0x2a0 [nfsd] nfsctl_transaction_write+0x74/0xa0 [nfsd] vfs_write+0x1a5/0x6d0 ksys_write+0xc1/0x160 do_syscall_64+0x5f/0x170 entry_SYSCALL_64_after_hwframe+0x76/0x7e To resolve this issue, cancel `nfsd_shrinker_work` using synchronous mode in nfs4_state_shutdown_net. Fixes: 7c24fa225081 ("NFSD: replace delayed_work with work_struct for nfsd_client_shrinker") Signed-off-by: Yang Erkun <yangerkun@huaweicloud.com> Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
2024-10-21LoongArch: Set initial pte entry with PAGE_GLOBAL for kernel spaceBibo Mao
There are two pages in one TLB entry on LoongArch system. For kernel space, it requires both two pte entries (buddies) with PAGE_GLOBAL bit set, otherwise HW treats it as non-global tlb, there will be potential problems if tlb entry for kernel space is not global. Such as fail to flush kernel tlb with the function local_flush_tlb_kernel_range() which supposed only flush tlb with global bit. Kernel address space areas include percpu, vmalloc, vmemmap, fixmap and kasan areas. For these areas both two consecutive page table entries should be enabled with PAGE_GLOBAL bit. So with function set_pte() and pte_clear(), pte buddy entry is checked and set besides its own pte entry. However it is not atomic operation to set both two pte entries, there is problem with test_vmalloc test case. So function kernel_pte_init() is added to init a pte table when it is created for kernel address space, and the default initial pte value is PAGE_GLOBAL rather than zero at beginning. Then only its own pte entry need update with function set_pte() and pte_clear(), nothing to do with the pte buddy entry. Signed-off-by: Bibo Mao <maobibo@loongson.cn> Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>
2024-10-21LoongArch: Don't crash in stack_top() for tasks without vDSOThomas Weißschuh
Not all tasks have a vDSO mapped, for example kthreads never do. If such a task ever ends up calling stack_top(), it will derefence the NULL vdso pointer and crash. This can for example happen when using kunit: [<9000000000203874>] stack_top+0x58/0xa8 [<90000000002956cc>] arch_pick_mmap_layout+0x164/0x220 [<90000000003c284c>] kunit_vm_mmap_init+0x108/0x12c [<90000000003c1fbc>] __kunit_add_resource+0x38/0x8c [<90000000003c2704>] kunit_vm_mmap+0x88/0xc8 [<9000000000410b14>] usercopy_test_init+0xbc/0x25c [<90000000003c1db4>] kunit_try_run_case+0x5c/0x184 [<90000000003c3d54>] kunit_generic_run_threadfn_adapter+0x24/0x48 [<900000000022e4bc>] kthread+0xc8/0xd4 [<9000000000200ce8>] ret_from_kernel_thread+0xc/0xa4 Fixes: 803b0fc5c3f2 ("LoongArch: Add process management") Signed-off-by: Thomas Weißschuh <thomas.weissschuh@linutronix.de> Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>
2024-10-21LoongArch: Set correct size for vDSO code mappingHuacai Chen
The current size of vDSO code mapping is hardcoded to PAGE_SIZE. This cannot work for 4KB page size after commit 18efd0b10e0fd77 ("LoongArch: vDSO: Wire up getrandom() vDSO implementation") because the code size increases to 8KB. Thus set the code mapping size to its real size, i.e. PAGE_ALIGN(vdso_end - vdso_start). Fixes: 18efd0b10e0fd77 ("LoongArch: vDSO: Wire up getrandom() vDSO implementation") Reviewed-by: Xi Ruoyao <xry111@xry111.site> Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>
2024-10-21LoongArch: Enable IRQ if do_ale() triggered in irq-enabled contextHuacai Chen
Unaligned access exception can be triggered in irq-enabled context such as user mode, in this case do_ale() may call get_user() which may cause sleep. Then we will get: BUG: sleeping function called from invalid context at arch/loongarch/kernel/access-helper.h:7 in_atomic(): 0, irqs_disabled(): 1, non_block: 0, pid: 129, name: modprobe preempt_count: 0, expected: 0 RCU nest depth: 0, expected: 0 CPU: 0 UID: 0 PID: 129 Comm: modprobe Tainted: G W 6.12.0-rc1+ #1723 Tainted: [W]=WARN Stack : 9000000105e0bd48 0000000000000000 9000000003803944 9000000105e08000 9000000105e0bc70 9000000105e0bc78 0000000000000000 0000000000000000 9000000105e0bc78 0000000000000001 9000000185e0ba07 9000000105e0b890 ffffffffffffffff 9000000105e0bc78 73924b81763be05b 9000000100194500 000000000000020c 000000000000000a 0000000000000000 0000000000000003 00000000000023f0 00000000000e1401 00000000072f8000 0000007ffbb0e260 0000000000000000 0000000000000000 9000000005437650 90000000055d5000 0000000000000000 0000000000000003 0000007ffbb0e1f0 0000000000000000 0000005567b00490 0000000000000000 9000000003803964 0000007ffbb0dfec 00000000000000b0 0000000000000007 0000000000000003 0000000000071c1d ... Call Trace: [<9000000003803964>] show_stack+0x64/0x1a0 [<9000000004c57464>] dump_stack_lvl+0x74/0xb0 [<9000000003861ab4>] __might_resched+0x154/0x1a0 [<900000000380c96c>] emulate_load_store_insn+0x6c/0xf60 [<9000000004c58118>] do_ale+0x78/0x180 [<9000000003801bc8>] handle_ale+0x128/0x1e0 So enable IRQ if unaligned access exception is triggered in irq-enabled context to fix it. Cc: stable@vger.kernel.org Reported-by: Binbin Zhou <zhoubinbin@loongson.cn> Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>
2024-10-21LoongArch: Get correct cores_per_package for SMT systemsHuacai Chen
In loongson_sysconf, The "core" of cores_per_node and cores_per_package stands for a logical core, which means in a SMT system it stands for a thread indeed. This information is gotten from SMBIOS Type4 Structure, so in order to get a correct cores_per_package for both SMT and non-SMT systems in parse_cpu_table() we should use SMBIOS_THREAD_PACKAGE_OFFSET instead of SMBIOS_CORE_PACKAGE_OFFSET. Cc: stable@vger.kernel.org Reported-by: Chao Li <lichao@loongson.cn> Tested-by: Chao Li <lichao@loongson.cn> Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>
generated by cgit (git 2.34.1) at 2025-09-15 14:03:48 +0000