6.10/chimeraos #19
Closed
6.10/chimeraos #19
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Our default behavior continues to match the vanilla kernel.
The Nvidia proprietary driver has some bugs that leads to issues if used with the simpledrm driver. The most noticeable is that does not register an emulated fbdev device. It just relies on a fbdev to be registered by another driver, that could be that could be attached to the framebuffer console. On UEFI machines, this is the efifb driver. This means that disabling the efifb driver will cause virtual consoles to not be present in the system when using the Nvidia driver. Legacy BIOS is not affected just because fbcon is not used there, but instead vgacon. Unless a VGA mode is specified using the vga= kernel command line option, in that case the vesafb driver is used instead and its fbdev attached to the fbcon. This is a problem because with CONFIG_SYSFB_SIMPLEFB=y, the sysfb platform code attempts to register a "simple-framebuffer" platform device (that is matched against simpledrm) and only registers either an "efi-framebuffer" or "vesa-framebuffer" if this fails to be registered due the video modes not being compatible. The Nvidia driver relying on another driver to register the fbdev is quite fragile, since it can't really assume those will stick around. For example there are patches posted to remove the EFI and VESA platform devices once a real DRM or fbdev driver probes. But in any case, moving to a simpledrm + emulated fbdev only breaks this assumption and causes users to not have VT if the Nvidia driver is used. So to prevent this, let's add a workaround and make the sysfb to skip the "simple-framebuffer" registration when nvidia-drm.modeset=1 option is set. This is quite horrible, but honestly I can't think of any other approach. For this to work, the CONFIG_FB_EFI and CONFIG_FB_VESA config options must be enabled besides CONFIG_DRM_SIMPLEDRM. Signed-off-by: Javier Martinez Canillas <javierm@redhat.com> Cherry-picked-for: https://bugs.archlinux.org/task/73720
To mitigate https://zolutal.github.io/aslrnt/; do this with a patch to avoid having to enable `CONFIG_EXPERT`.
Add quirk orientation for AYA NEO 2. The name appears without spaces in dmi strings. That made it difficult to reuse the 2021 match and the display is greater in resolution. Tested by the JELOS team that has been patching their own kernel for a while now and confirmed by users in the AYA NEO and ChimeraOS discord servers. Signed-off-by: Joaquín Ignacio Aramendía <samsagax@gmail.com>
Add quirk orientation for AYA NEO Founder. The name appears with spaces in dmi strings as other devices of the brand. The panel is the same as the NEXT and 2021 models. Those could not be reused as the former has VENDOR name as "AYANEO" without spaces and the latter has "AYADEVICE". Tested by the JELOS team that has been patching their own kernel for a while now and confirmed by users in the AYA NEO and ChimeraOS discord servers. Signed-off-by: Joaquín Ignacio Aramendía <samsagax@gmail.com>
Add quirk orientation for AYA NEO GEEK. One of the more recent devices by the brand. The name appears without spaces in dmi strings. The board name is completely different to the previous models making it difficult to reuse their quirks despite being the same resolution and mounting. Tested by the JELOS team that has been patching their own kernel for a while now and confirmed by users in the AYA NEO and ChimeraOS discord servers. Signed-off-by: Joaquín Ignacio Aramendía <samsagax@gmail.com>
Add quirk orientation for the Ayn Loki Zero. This also has been tested/used by the JELOS team. Signed-off-by: Bouke Sybren Haarsma <boukehaarsma23@gmail.com> Reviewed-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Hans de Goede <hdegoede@redhat.com> Link: https://patchwork.freedesktop.org/patch/msgid/20240728124731.168452-2-boukehaarsma23@gmail.com
Add quirk orientation for Ayn Loki Max model. This has been tested by JELOS team that uses their own patched kernel for a while now and confirmed by users in the ChimeraOS discord servers. Signed-off-by: Bouke Sybren Haarsma <boukehaarsma23@gmail.com> Reviewed-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Hans de Goede <hdegoede@redhat.com> Link: https://patchwork.freedesktop.org/patch/msgid/20240728124731.168452-3-boukehaarsma23@gmail.com
We are enabling a large set of color calibration features to enhance KMS color mgmt but these properties are specific of AMD display HW, and cannot be provided by other vendors. Therefore, set a config option to enable AMD driver-private properties used on Steam Deck color mgmt pipeline. Replace the agreed name `AMD_PRIVATE_COLOR` with our downstream version `CONFIG_DRM_AMD_COLOR_STEAMDECK`. Signed-off-by: Melissa Wen <mwen@igalia.com>
This patch adds a new sysfs event, which will indicate the userland about a GPU reset, and can also provide some information like: - process ID of the process involved with the GPU reset - process name of the involved process - the GPU status info (using flags) This patch also introduces the first flag of the flags bitmap, which can be appended as and when required. V2: Addressed review comments from Christian and Amar - move the reset information structure to DRM layer - drop _ctx from struct name - make pid 32 bit(than 64) - set flag when VRAM invalid (than valid) - add process name as well (Amar) Cc: Alexandar Deucher <alexander.deucher@amd.com> Cc: Christian Koenig <christian.koenig@amd.com> Cc: Amaranath Somalapuram <amaranath.somalapuram@amd.com> Signed-off-by: Shashank Sharma <shashank.sharma@amd.com> (cherry picked from commit 90230bd9d9c7d979038547460c9a2cbbeff8d6b9) [Forward port to 6.0] Signed-off-by: Cristian Ciocaltea <cristian.ciocaltea@collabora.com>
This patch adds a work function, which sends a GPU reset uevent and some contextual infomration, like the PID and some status flags. This work should be scheduled during a GPU reset. The userspace can do some recovery and post-processing work based on this event and information. V2: Addressed review comments from Christian - Changed the name of the work to gpu_reset_event_work - Added a structure to accommodate some additional information (like a PID and some flags) - Do not add new structure in amdgpu.h Cc: Alexander Deucher <alexander.deucher@amd.com> Cc: Christian Koenig <christian.koenig@amd.com> Cc: Amaranath Somalapuram <amaranath.somalapuram@amd.com> Signed-off-by: Shashank Sharma <shashank.sharma@amd.com> (cherry picked from commit f63b09e78126f7da67b69409e2cce1d3ab2d7f46) [Forward port to 6.0] Signed-off-by: Cristian Ciocaltea <cristian.ciocaltea@collabora.com>
Schedule work function with valid PID, process name, and vram lost status during a GPU reset/ recovery. Signed-off-by: Somalapuram Amaranath <Amaranath.Somalapuram@amd.com> (cherry picked from commit 293c019a84c6402b08db9579819b555b01cd613b) [Forward ported to 6.0] Signed-off-by: Cristian Ciocaltea <cristian.ciocaltea@collabora.com> [Forward ported to 6.9] Signed-off-by: Bouke Sybren Haarsma <boukehaarsma23@gmail.com>
Add support for the OrangePi NEO-01. It uses different registers for PWM manual mode, set PWM, and read fan speed than previous devices. Valid PWM input and duty cycle is 1-244, we scale this from 1-255 to maintain compatibility with the existing interface. Add OneXPlayer 2 series, OneXFly, and X1 series models. The 2/X1 series use new registers for turbo button takeover and read fan speed. X1 has an Intel variant so change the CPU detection at init to only check for the affected devices. While at it, adjust formatting of some constants and reorder all cases alphabetically for consistency. Rename OXP_OLD constants to OXP_MINI for disambiguation. Update code comments for clarity. Add support for AYANEO models 2S, AIR 1S, Flip series, GEEK 1S, and KUN. Signed-off-by: Derek J. Clark <derekjohn.clark@gmail.com> Tested-by: Kevin Greenberg <kdgreenberg234@protonmail.com> Tested-by: Joshua Tam <csinaction@pm.me> Tested-by: Parth Menon <parthasarathymenon@gmail.com> Tested-by: Philip Müller <philm@manjaro.org> Reported-by: kernel test robot <lkp@intel.com> Closes: https://lore.kernel.org/oe-kbuild-all/202408160329.TLNbIwRC-lkp@intel.com/
Bumps the sensitivity of AMD sfh gyro and accelerometers by removing the division operation and rebasing the units in the hid descriptor. This helps with the gyro deadzone of the Legion Go. Should not affect existing devices.
The ultra-low power BMI260 is an IMU consisting of a 16-bit tri-axial gyroscope and a 16-bit tri-axial accelerometer combining precise acceleration, angular rate measurement and intelligent on-chip motion-triggered interrupt features. The driver supports the BMI260 over I2C. It is based on the BMI160 driver, and like that driver supports accelerometer and gyroscope reading, as well as data ready interrupts.
Contribution by CVMagic (https://github.com/CVMagic) aw87xxx: Use strscpy instead of strlcpy awinic: i2c_driver cleanup and fixes
… also fixed Reset Pin GPIO initialization issue with Ayn Loki Mini
…f specified in ACPI
bmi150-accel and bmi323-imu are declared in an almost identical way in the ACPI and in some devices such as the Asus RC71L the "ROTM" property can be found: parse and use the ACPI-defined mount-matrix. Co-developed-by: Luke D. Jones <luke@ljones.dev> Co-developed-by: Jonathan LoBue <jlobue10@gmail.com> Signed-off-by: Denis Benato <benato.denis96@gmail.com>
The bmi323 is mounted on some devices that are powered by an internal battery: help in reducing system overall power drain while the system is in s2idle or the imu driver is not loaded by resetting it in its lowest power draining state. Signed-off-by: Denis Benato <benato.denis96@gmail.com>
The new ASUS ROG Ally X functions almost exactly the same as the previous model, so we can use the same quirks. Signed-off-by: Luke D. Jones <luke@ljones.dev> Signed-off-by: Jiri Kosina <jkosina@suse.com>
Patch to test fixing TAS2781 amp getting bound properly to i2c for ASUS ROG ALLY X.
Add support for vivobook fan profiles wmi call on the ASUS VIVOBOOK to adjust power limits. These fan profiles have a different device id than the ROG series and different order. This reorders the existing modes. As part of keeping the patch clean the throttle_thermal_policy_available boolean stored in the driver struct is removed and throttle_thermal_policy_dev is used in place (as on init it is zeroed). Co-developed-by: Luke D. Jones <luke@ljones.dev> Signed-off-by: Luke D. Jones <luke@ljones.dev> Signed-off-by: Mohamed Ghanmi <mohamed.ghanmi@supcom.tn> Reviewed-by: Luke D. Jones <luke@ljones.dev> Link: https://lore.kernel.org/r/20240609144849.2532-2-mohamed.ghanmi@supcom.tn Reviewed-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Hans de Goede <hdegoede@redhat.com>
On almost all ASUS ROG series laptops the MCU used for the USB keyboard also has a HID packet used for setting the brightness. This is usually the same as the WMI method. But in some laptops the WMI method either is missing or doesn't work, so we should default to the HID control. Signed-off-by: Luke D. Jones <luke@ljones.dev> Acked-by: Benjamin Tissoires <bentiss@kernel.org> Link: https://lore.kernel.org/r/20240713074733.77334-2-luke@ljones.dev Reviewed-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Add the VID/PID for ASUS ROG RAIKIRI PRO to xpad_device and the VID to xpad_table. Signed-off-by: Luke D. Jones <luke@ljones.dev>
Add more verbose debug print in the WMI method calls. This helps a lot with debugging various issues working with regular users as the WMI methods can be traced now. Signed-off-by: Luke D. Jones <luke@ljones.dev>
…ered On some newer laptops it appears that an AMD driver can register a platform_profile handler. If this happens then the asus_wmi driver would error with -EEXIST when trying to register its own handler leaving the user with a possibly unusable system - this is especially true for laptops with an MCU that emit a stream of HID packets, some of which can be misinterpreted as shutdown signals. We can safely continue loading the driver instead of bombing out. Signed-off-by: Luke D. Jones <luke@ljones.dev>
The new ROG Ally X functions the same as the previus model so we can use the same method to ensure the MCU USB devices wake and reconnect correctly. Given that two devices marks the start of a trend, this patch also adds a quirk table to make future additions easier if the MCU is the same. Signed-off-by: Luke D. Jones <luke@ljones.dev>
The fw_attributes_class provides a much cleaner interface to all of the attributes introduced to asus-wmi. This patch moves all of these extra attributes over to fw_attributes_class, and shifts the bulk of these definitions to a new kernel module to reduce the clutter of asus-wmi with the intention of deprecating the asus-wmi attributes in future. The work applies only to WMI methods which don't have a clearly defined place within the sysfs and as a result ended up lumped together in /sys/devices/platform/asus-nb-wmi/ with no standard API. Where possible the fw attrs now implement defaults, min, max, scalar, choices, etc. As en example dgpu_disable becomes: /sys/class/firmware-attributes/asus-armoury/attributes/dgpu_disable/ ├── current_value ├── display_name ├── possible_values └── type as do other attributes. Signed-off-by: Luke D. Jones <luke@ljones.dev>
Implement the dgpu TGP control under the asus-armoury module using the fw_attributes class. Signed-off-by: Luke D. Jones <luke@ljones.dev>
Implement the APU memory size control under the asus-armoury module using the fw_attributes class. This allows the APU allocated memory size to be adjusted depending on the users priority. A reboot is required after change. Signed-off-by: Luke D. Jones <luke@ljones.dev>
Implement Intel core enablement under the asus-armoury module using the fw_attributes class. This allows users to enable or disable preformance or efficiency cores depending on their requirements. After change a reboot is required. Signed-off-by: Luke D. Jones <luke@ljones.dev>
With the existence of the asus-bioscfg module the attributes no-longer need to live under the /sys/devices/platform/asus-nb-wmi/ path. Deprecate all those that were implemented in asus-bioscfg with the goal of removing them fully in the next LTS cycle. Signed-off-by: Luke D. Jones <luke@ljones.dev>
All the _DSM methods for the LPS0 method are called back to back currently. The intended use of the screen off and screen on calls is supposed to be matching the screen being turned on or off though. Add support for other parts of the kernel to call such a callback. Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
This allows notifying the BIOS with the LPS0 _DSM for "Screen off" and "Screen on". Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
This lets the DRM core notify on screen events instead of calling them back to back. Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Normally systems don't wake up from s2idle when the AC adapter has changed, but users or other drivers may have a desire to configure this behavior. Add the ability for the LPS0 device to wake the system when AC adapter changes. Also add a new member to `struct acpi_s2idle_dev_ops` to let drivers indicate the system should wakeup by default from AC adapter changes. Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
This driver adds full support of the ASUS ROG Ally gamepad:
- dinput is translated to XBox controller (Ally-X only)
- default mode has the QAM buttons mapped (Ally-X only)
* left is XBox button
* right is an XBox + A combo for steam QAM
- force feedback is supported (Ally-X only)
- LED brightness control (0-2)
- LED multicolor class support for all 4 LED, individually addressable
- Support all configuration
Where "Ally-X only" is mentioned, this is because the Ally 1 uses the
existing xpad driver.
The configuration options available are:
- Gamepad mode (game, wasd, mouse)
- Remapping each button, plus macro map (hold a macro button and press other)
- Joystrick and trigger deadzones
- Gamepad vibration intensity
- Leds (using multicolor class)
- Button turbo abilities (per button)
- Joystick repsonse curves
- Joystick anti-deadzones
The attribute path tree looks like this:
- `./sys/../<USB HID>/`
- `joystick_left/
- `deadzone`
- `mapping` (mouse, wasd, custom)
- `anti_deadzone`
- `response_curve`
- `calibration`
- `calibration_reset`
- `trigger_left/
- `deadzone`
- `response_curve`
- `calibration`
- `calibration_reset`
- `gamepad_mode`
- `button_mapping`
- `A`
- `B`
- `dpad_left`
- etc
No settings are applied until `apply_all` is written to. The exception is
for calibrations.
While there is calibration ability, it can be difficult to get correct
and is heavily device dependent, as such it is set when written and not
when `apply_all` is written to. On driver load the set calibrations are
retrieved - this may be what you've set in Linux, Windows, or factory
defaults.
As a note: the proper way to set calibrations is to read the joystick
voltage from the EC whcih currently is not possible.
Signed-off-by: Luke D. Jones <luke@ljones.dev>
Adjust the CSEE quirk call ordering and time delay. Attempts to fix issues where the MCU isn't provided enough time to complete USB device disconnection or reconnection, particularly in powersave mode. Signed-off-by: Luke D. Jones <luke@ljones.dev>
BoukeHaarsma23
force-pushed
the
6.10/chimeraos
branch
from
f283500 to
2b5cb76
Compare
flukejones
pushed a commit
to flukejones/linux
that referenced
this pull request
Nov 27, 2024
The action force umount(umount -f) will attempt to kill all rpc_task even
umount operation may ultimately fail if some files remain open.
Consequently, if an action attempts to open a file, it can potentially
send two rpc_task to nfs server.
NFS CLIENT
thread1 thread2
open("file")
...
nfs4_do_open
_nfs4_do_open
_nfs4_open_and_get_state
_nfs4_proc_open
nfs4_run_open_task
/* rpc_task1 */
rpc_run_task
rpc_wait_for_completion_task
umount -f
nfs_umount_begin
rpc_killall_tasks
rpc_signal_task
rpc_task1 been wakeup
and return -512
_nfs4_do_open // while loop
...
nfs4_run_open_task
/* rpc_task2 */
rpc_run_task
rpc_wait_for_completion_task
While processing an open request, nfsd will first attempt to find or
allocate an nfs4_openowner. If it finds an nfs4_openowner that is not
marked as NFS4_OO_CONFIRMED, this nfs4_openowner will released. Since
two rpc_task can attempt to open the same file simultaneously from the
client to server, and because two instances of nfsd can run
concurrently, this situation can lead to lots of memory leak.
Additionally, when we echo 0 to /proc/fs/nfsd/threads, warning will be
triggered.
NFS SERVER
nfsd1 nfsd2 echo 0 > /proc/fs/nfsd/threads
nfsd4_open
nfsd4_process_open1
find_or_alloc_open_stateowner
// alloc oo1, stateid1
nfsd4_open
nfsd4_process_open1
find_or_alloc_open_stateowner
// find oo1, without NFS4_OO_CONFIRMED
release_openowner
unhash_openowner_locked
list_del_init(&oo->oo_perclient)
// cannot find this oo
// from client, LEAK!!!
alloc_stateowner // alloc oo2
nfsd4_process_open2
init_open_stateid
// associate oo1
// with stateid1, stateid1 LEAK!!!
nfs4_get_vfs_file
// alloc nfsd_file1 and nfsd_file_mark1
// all LEAK!!!
nfsd4_process_open2
...
write_threads
...
nfsd_destroy_serv
nfsd_shutdown_net
nfs4_state_shutdown_net
nfs4_state_destroy_net
destroy_client
__destroy_client
// won't find oo1!!!
nfsd_shutdown_generic
nfsd_file_cache_shutdown
kmem_cache_destroy
for nfsd_file_slab
and nfsd_file_mark_slab
// bark since nfsd_file1
// and nfsd_file_mark1
// still alive
=======================================================================
BUG nfsd_file (Not tainted): Objects remaining in nfsd_file on
__kmem_cache_shutdown()
-----------------------------------------------------------------------
Slab 0xffd4000004438a80 objects=34 used=1 fp=0xff11000110e2ad28
flags=0x17ffffc0000240(workingset|head|node=0|zone=2|lastcpupid=0x1fffff)
CPU: 4 UID: 0 PID: 757 Comm: sh Not tainted 6.12.0-rc6+ ChimeraOS#19
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS
1.16.1-2.fc37 04/01/2014
Call Trace:
<TASK>
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+0x1ae/0x6d0
ksys_write+0xc1/0x160
do_syscall_64+0x5f/0x170
entry_SYSCALL_64_after_hwframe+0x76/0x7e
Disabling lock debugging due to kernel taint
Object 0xff11000110e2ac38 @offset=3128
Allocated in nfsd_file_do_acquire+0x20f/0xa30 [nfsd] age=1635 cpu=3
pid=800
nfsd_file_do_acquire+0x20f/0xa30 [nfsd]
nfsd_file_acquire_opened+0x5f/0x90 [nfsd]
nfs4_get_vfs_file+0x4c9/0x570 [nfsd]
nfsd4_process_open2+0x713/0x1070 [nfsd]
nfsd4_open+0x74b/0x8b0 [nfsd]
nfsd4_proc_compound+0x70b/0xc20 [nfsd]
nfsd_dispatch+0x1b4/0x3a0 [nfsd]
svc_process_common+0x5b8/0xc50 [sunrpc]
svc_process+0x2ab/0x3b0 [sunrpc]
svc_handle_xprt+0x681/0xa20 [sunrpc]
nfsd+0x183/0x220 [nfsd]
kthread+0x199/0x1e0
ret_from_fork+0x31/0x60
ret_from_fork_asm+0x1a/0x30
Add nfs4_openowner_unhashed to help found unhashed nfs4_openowner, and
break nfsd4_open process to fix this problem.
Cc: stable@vger.kernel.org # v5.4+
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Signed-off-by: Yang Erkun <yangerkun@huawei.com>
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
NeroReflex
pushed a commit
that referenced
this pull request
Dec 12, 2024
commit 98100e8 upstream. The action force umount(umount -f) will attempt to kill all rpc_task even umount operation may ultimately fail if some files remain open. Consequently, if an action attempts to open a file, it can potentially send two rpc_task to nfs server. NFS CLIENT thread1 thread2 open("file") ... nfs4_do_open _nfs4_do_open _nfs4_open_and_get_state _nfs4_proc_open nfs4_run_open_task /* rpc_task1 */ rpc_run_task rpc_wait_for_completion_task umount -f nfs_umount_begin rpc_killall_tasks rpc_signal_task rpc_task1 been wakeup and return -512 _nfs4_do_open // while loop ... nfs4_run_open_task /* rpc_task2 */ rpc_run_task rpc_wait_for_completion_task While processing an open request, nfsd will first attempt to find or allocate an nfs4_openowner. If it finds an nfs4_openowner that is not marked as NFS4_OO_CONFIRMED, this nfs4_openowner will released. Since two rpc_task can attempt to open the same file simultaneously from the client to server, and because two instances of nfsd can run concurrently, this situation can lead to lots of memory leak. Additionally, when we echo 0 to /proc/fs/nfsd/threads, warning will be triggered. NFS SERVER nfsd1 nfsd2 echo 0 > /proc/fs/nfsd/threads nfsd4_open nfsd4_process_open1 find_or_alloc_open_stateowner // alloc oo1, stateid1 nfsd4_open nfsd4_process_open1 find_or_alloc_open_stateowner // find oo1, without NFS4_OO_CONFIRMED release_openowner unhash_openowner_locked list_del_init(&oo->oo_perclient) // cannot find this oo // from client, LEAK!!! alloc_stateowner // alloc oo2 nfsd4_process_open2 init_open_stateid // associate oo1 // with stateid1, stateid1 LEAK!!! nfs4_get_vfs_file // alloc nfsd_file1 and nfsd_file_mark1 // all LEAK!!! nfsd4_process_open2 ... write_threads ... nfsd_destroy_serv nfsd_shutdown_net nfs4_state_shutdown_net nfs4_state_destroy_net destroy_client __destroy_client // won't find oo1!!! nfsd_shutdown_generic nfsd_file_cache_shutdown kmem_cache_destroy for nfsd_file_slab and nfsd_file_mark_slab // bark since nfsd_file1 // and nfsd_file_mark1 // still alive ======================================================================= BUG nfsd_file (Not tainted): Objects remaining in nfsd_file on __kmem_cache_shutdown() ----------------------------------------------------------------------- Slab 0xffd4000004438a80 objects=34 used=1 fp=0xff11000110e2ad28 flags=0x17ffffc0000240(workingset|head|node=0|zone=2|lastcpupid=0x1fffff) CPU: 4 UID: 0 PID: 757 Comm: sh Not tainted 6.12.0-rc6+ #19 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.16.1-2.fc37 04/01/2014 Call Trace: <TASK> 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+0x1ae/0x6d0 ksys_write+0xc1/0x160 do_syscall_64+0x5f/0x170 entry_SYSCALL_64_after_hwframe+0x76/0x7e Disabling lock debugging due to kernel taint Object 0xff11000110e2ac38 @offset=3128 Allocated in nfsd_file_do_acquire+0x20f/0xa30 [nfsd] age=1635 cpu=3 pid=800 nfsd_file_do_acquire+0x20f/0xa30 [nfsd] nfsd_file_acquire_opened+0x5f/0x90 [nfsd] nfs4_get_vfs_file+0x4c9/0x570 [nfsd] nfsd4_process_open2+0x713/0x1070 [nfsd] nfsd4_open+0x74b/0x8b0 [nfsd] nfsd4_proc_compound+0x70b/0xc20 [nfsd] nfsd_dispatch+0x1b4/0x3a0 [nfsd] svc_process_common+0x5b8/0xc50 [sunrpc] svc_process+0x2ab/0x3b0 [sunrpc] svc_handle_xprt+0x681/0xa20 [sunrpc] nfsd+0x183/0x220 [nfsd] kthread+0x199/0x1e0 ret_from_fork+0x31/0x60 ret_from_fork_asm+0x1a/0x30 Add nfs4_openowner_unhashed to help found unhashed nfs4_openowner, and break nfsd4_open process to fix this problem. Cc: stable@vger.kernel.org # v5.4+ Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Yang Erkun <yangerkun@huawei.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Same story as with 6.9. I had to drop and cherry-pick. Best way to resolve is to
reset --hardon to my branch.