Pull printk updates from Petr Mladek:
- Allow creaing nbcon console drivers with an unsafe write_atomic()
callback that can only be called by the final nbcon_atomic_flush_unsafe().
Otherwise, the driver would rely on the kthread.
It is going to be used as the-best-effort approach for an
experimental nbcon netconsole driver, see
https://lore.kernel.org/r/20251121-nbcon-v1-2-503d17b2b4af@debian.org
Note that a safe .write_atomic() callback is supposed to work in NMI
context. But some networking drivers are not safe even in IRQ
context:
https://lore.kernel.org/r/oc46gdpmmlly5o44obvmoatfqo5bhpgv7pabpvb6sjuqioymcg@gjsma3ghoz35
In an ideal world, all networking drivers would be fixed first and
the atomic flush would be blocked only in NMI context. But it brings
the question how reliable networking drivers are when the system is
in a bad state. They might block flushing more reliable serial
consoles which are more suitable for serious debugging anyway.
- Allow to use the last 4 bytes of the printk ring buffer.
- Prevent queuing IRQ work and block printk kthreads when consoles are
suspended. Otherwise, they create non-necessary churn or even block
the suspend.
- Release console_lock() between each record in the kthread used for
legacy consoles on RT. It might significantly speed up the boot.
- Release nbcon context between each record in the atomic flush. It
prevents stalls of the related printk kthread after it has lost the
ownership in the middle of a record
- Add support for NBCON consoles into KDB
- Add %ptsP modifier for printing struct timespec64 and use it where
possible
- Misc code clean up
* tag 'printk-for-6.19' of git://git.kernel.org/pub/scm/linux/kernel/git/printk/linux: (48 commits)
printk: Use console_is_usable on console_unblank
arch: um: kmsg_dump: Use console_is_usable
drivers: serial: kgdboc: Drop checks for CON_ENABLED and CON_BOOT
lib/vsprintf: Unify FORMAT_STATE_NUM handlers
printk: Avoid irq_work for printk_deferred() on suspend
printk: Avoid scheduling irq_work on suspend
printk: Allow printk_trigger_flush() to flush all types
tracing: Switch to use %ptSp
scsi: snic: Switch to use %ptSp
scsi: fnic: Switch to use %ptSp
s390/dasd: Switch to use %ptSp
ptp: ocp: Switch to use %ptSp
pps: Switch to use %ptSp
PCI: epf-test: Switch to use %ptSp
net: dsa: sja1105: Switch to use %ptSp
mmc: mmc_test: Switch to use %ptSp
media: av7110: Switch to use %ptSp
ipmi: Switch to use %ptSp
igb: Switch to use %ptSp
e1000e: Switch to use %ptSp
...
The MMIO_REMAP BO is a special 4K IO page that does not have a ttm_tt
behind it. However, amdgpu_ttm_tt_pde_flags() was treating it like
normal TT/doorbell/preempt memory and unconditionally accessed
ttm->caching. For the MMIO_REMAP BO, ttm is NULL, so this leads to a
NULL pointer dereference when computing PDE flags.
Fix this by checking that ttm is non-NULL before reading ttm->caching.
This prevents the crash for MMIO_REMAP and also makes the code more
defensive if other BOs ever come through without a ttm_tt.
Fixes: fb5a52dbe9 ("drm/amdgpu: Implement TTM handling for MMIO_REMAP placement")
Suggested-by: Jesse Zhang <Jesse.Zhang@amd.com>
Suggested-by: Christian König <christian.koenig@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Jesse Zhang <Jesse.Zhang@amd.com>
Tested-by: Jesse Zhang <Jesse.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 0db94da5a0)
This fixes sparse mappings (aka. partially resident textures).
Check the correct flags.
Since a recent refactor, the code works with uAPI flags (for
mapping buffer objects), and not PTE (page table entry) flags.
Fixes: 6716a823d1 ("drm/amdgpu: rework how PTE flags are generated v3")
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 8feeab26c8)
[Why]
Accoreding to CP updated to RS64 on gfx11,
WRITE_DATA with PREEMPTION_META_MEMORY(dst_sel=8) is illegal for CP FW.
That packet is used for MCBP on F32 based system.
So it would lead to incorrect GRBM write and FW is not handling that
extra case correctly.
[How]
With gfx11 rs64 enabled, skip emit de meta data.
Signed-off-by: Yifan Zha <Yifan.Zha@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 8366cd442d)
Cc: stable@vger.kernel.org
During the suspend sequence VPE is already going to be power gated
as part of vpe_suspend(). It's unnecessary to call during calls to
amdgpu_device_set_pg_state().
It actually can expose a race condition with the firmware if s0i3
sequence starts as well. Drop these calls.
Cc: Peyton.Lee@amd.com
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 2a6c826cfe)
Cc: stable@vger.kernel.org
When the BO pointer provided to amdgpu_bo_create_kernel() points to
non-NULL, amdgpu_bo_create_kernel() takes it as a hint to pin that address
rather than allocate a new BO.
This functionality is never desired for allocating ISP buffers. A new BO
should always be created when isp_kernel_buffer_alloc() is called, per the
description for isp_kernel_buffer_alloc().
Ensure this by zeroing *bo right before the amdgpu_bo_create_kernel() call.
Fixes: 55d42f6169 ("drm/amd/amdgpu: Add helper functions for isp buffers")
Reviewed-by: Mario Limonciello (AMD) <superm1@kernel.org>
Reviewed-by: Pratap Nirujogi <pratap.nirujogi@amd.com>
Signed-off-by: Sultan Alsawaf <sultan@kerneltoast.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 73c8c29baa)
Fix a potential deadlock caused by inconsistent spinlock usage
between interrupt and process contexts in the userq fence driver.
The issue occurs when amdgpu_userq_fence_driver_process() is called
from both:
- Interrupt context: gfx_v11_0_eop_irq() -> amdgpu_userq_fence_driver_process()
- Process context: amdgpu_eviction_fence_suspend_worker() ->
amdgpu_userq_fence_driver_force_completion() -> amdgpu_userq_fence_driver_process()
In interrupt context, the spinlock was acquired without disabling
interrupts, leaving it in {IN-HARDIRQ-W} state. When the same lock
is acquired in process context, the kernel detects inconsistent
locking since the process context acquisition would enable interrupts
while holding a lock previously acquired in interrupt context.
Kernel log shows:
[ 4039.310790] inconsistent {IN-HARDIRQ-W} -> {HARDIRQ-ON-W} usage.
[ 4039.310804] kworker/7:2/409 [HC0[0]:SC0[0]:HE1:SE1] takes:
[ 4039.310818] ffff9284e1bed000 (&fence_drv->fence_list_lock){?...}-{3:3},
[ 4039.310993] {IN-HARDIRQ-W} state was registered at:
[ 4039.311004] lock_acquire+0xc6/0x300
[ 4039.311018] _raw_spin_lock+0x39/0x80
[ 4039.311031] amdgpu_userq_fence_driver_process.part.0+0x30/0x180 [amdgpu]
[ 4039.311146] amdgpu_userq_fence_driver_process+0x17/0x30 [amdgpu]
[ 4039.311257] gfx_v11_0_eop_irq+0x132/0x170 [amdgpu]
Fix by using spin_lock_irqsave()/spin_unlock_irqrestore() to properly
manage interrupt state regardless of calling context.
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Jesse Zhang <Jesse.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit ded3ad780c)
Cc: stable@vger.kernel.org
drm_sched_entity_init wasn't called yet, so the only thing to
do is to release allocated memory.
This doesn't fix any bug since entity is zero allocated and
drm_sched_entity_fini does nothing in this case.
Signed-off-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit ec49374ccb)
Certain multi-GPU configurations (especially GFX12) may hit
data corruption when a DCC-compressed VRAM surface is shared across GPUs
using peer-to-peer (P2P) DMA transfers.
Such surfaces rely on device-local metadata and cannot be safely accessed
through a remote GPU’s page tables. Attempting to import a DCC-enabled
surface through P2P leads to incorrect rendering or GPU faults.
This change disables P2P for DCC-enabled VRAM buffers that are contiguous
and allocated on GFX12+ hardware. In these cases, the importer falls back
to the standard system-memory path, avoiding invalid access to compressed
surfaces.
Future work could consider optional migration (VRAM→System→VRAM) if a
performance regression is observed when `attach->peer2peer = false`.
Tested on:
- Dual RX 9700 XT (Navi4x) setup
- GNOME and Wayland compositor scenarios
- Confirmed no corruption after disabling P2P under these conditions
v2: Remove check TTM_PL_VRAM & TTM_PL_FLAG_CONTIGUOUS.
v3: simplify for upsteam and fix ip version check (Alex)
Suggested-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Vitaly Prosyak <vitaly.prosyak@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 9dff2bb709)
Cc: stable@vger.kernel.org
commit c760bcda83 ("drm/amd: Check whether secure display TA loaded
successfully") attempted to fix extra messages, but failed to port the
cleanup that was in commit 5c6d52ff4b ("drm/amd: Don't try to enable
secure display TA multiple times") to prevent multiple tries.
Add that to the failure handling path even on a quick failure.
Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4679
Fixes: c760bcda83 ("drm/amd: Check whether secure display TA loaded successfully")
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 4104c0a454)
On PF passthrough environment, after hibernate and then resume, coralgemm
will cause gpu page fault.
Mode1 reset happens during hibernate, but partition mode is not restored
on resume, register mmCP_HYP_XCP_CTL and mmCP_PSP_XCP_CTL is not right
after resume. When CP access the MQD BO, wrong stride size is used,
this will cause out of bound access on the MQD BO, resulting page fault.
The fix is to ensure gfx_v9_4_3_switch_compute_partition() is called
when resume from a hibernation.
KFD resume is called separately during a reset recovery or resume from
suspend sequence. Hence it's not required to be called as part of
partition switch.
Signed-off-by: Samuel Zhang <guoqing.zhang@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 5d1b32cfe4)
For S3 on vangogh, PMFW needs to be notified before the
driver powers down RLC. This already happens in smu_disable_dpms()
so drop the superfluous call in amdgpu_device_suspend().
Co-developed-by: Mario Limonciello (AMD) <superm1@kernel.org>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Mario Limonciello (AMD) <superm1@kernel.org>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 960e30a61e)
These were not set so soft recovery was inadvertantly
disabled.
Fixes: 6ac55eab4f ("drm/amdgpu: move reset support type checks into the caller")
Reviewed-by: Jesse Zhang <Jesse.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 1972763505)
[Why]
Newer VPE microcode has functionality that will decrease DPM level
only when a workload has run for 2 or more seconds. If VPE is turned
off before this DPM decrease and the PMFW doesn't reset it when
power gating VPE, the SOC can get stuck with a higher DPM level.
This can happen from amdgpu's ring buffer test because it's a short
quick workload for VPE and VPE is turned off after 1s.
[How]
In idle handler besides checking fences are drained check PMFW version
to determine if it will reset DPM when power gating VPE. If PMFW will
not do this, then check VPE DPM level. If it is not DPM0 reschedule
delayed work again until it is.
v2: squash in return fix (Alex)
Cc: Peyton.Lee@amd.com
Reported-by: Sultan Alsawaf <sultan@kerneltoast.com>
Reviewed-by: Sultan Alsawaf <sultan@kerneltoast.com>
Tested-by: Sultan Alsawaf <sultan@kerneltoast.com>
Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4615
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 3ac635367e)
Cc: stable@vger.kernel.org
By design the MES will return an array result that is twice the number
of hung doorbells it can report.
i.e. if up k reported doorbells are supported, then the
second half of the array, also of length k, holds the HQD information
(type/queue/pipe) where queue 1 corresponds to index 0 and k,
queue 2 corresponds to index 1 and k + 1 etc ...
The driver will use the HDQ info to target queue/pipe reset for
hardware scheduled user compute queues.
Signed-off-by: Jonathan Kim <jonathan.kim@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Initialized doorbells should be set to invalid rather than 0 to prevent
driver from over counting hung doorbells since it checks against the
invalid value to begin with.
Signed-off-by: Jonathan Kim <jonathan.kim@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
GFX12 MES uses low 32 bits of status return for success (1 or 0)
and high bits for debug information if low bits are 0.
GFX11 MES doesn't do this so checking full 64-bit status return
for 1 or 0 is still valid.
Signed-off-by: Jonathan Kim <jonathan.kim@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
Previously, APU platforms (and other scenarios with uninitialized VRAM managers)
triggered a NULL pointer dereference in `ttm_resource_manager_usage()`. The root
cause is not that the `struct ttm_resource_manager *man` pointer itself is NULL,
but that `man->bdev` (the backing device pointer within the manager) remains
uninitialized (NULL) on APUs—since APUs lack dedicated VRAM and do not fully
set up VRAM manager structures. When `ttm_resource_manager_usage()` attempts to
acquire `man->bdev->lru_lock`, it dereferences the NULL `man->bdev`, leading to
a kernel OOPS.
1. **amdgpu_cs.c**: Extend the existing bandwidth control check in
`amdgpu_cs_get_threshold_for_moves()` to include a check for
`ttm_resource_manager_used()`. If the manager is not used (uninitialized
`bdev`), return 0 for migration thresholds immediately—skipping VRAM-specific
logic that would trigger the NULL dereference.
2. **amdgpu_kms.c**: Update the `AMDGPU_INFO_VRAM_USAGE` ioctl and memory info
reporting to use a conditional: if the manager is used, return the real VRAM
usage; otherwise, return 0. This avoids accessing `man->bdev` when it is
NULL.
3. **amdgpu_virt.c**: Modify the vf2pf (virtual function to physical function)
data write path. Use `ttm_resource_manager_used()` to check validity: if the
manager is usable, calculate `fb_usage` from VRAM usage; otherwise, set
`fb_usage` to 0 (APUs have no discrete framebuffer to report).
This approach is more robust than APU-specific checks because it:
- Works for all scenarios where the VRAM manager is uninitialized (not just APUs),
- Aligns with TTM's design by using its native helper function,
- Preserves correct behavior for discrete GPUs (which have fully initialized
`man->bdev` and pass the `ttm_resource_manager_used()` check).
v4: use ttm_resource_manager_used(&adev->mman.vram_mgr.manager) instead of checking the adev->gmc.is_app_apu flag (Christian)
Reviewed-by: Christian König <christian.koenig@amd.com>
Suggested-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Jesse Zhang <Jesse.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
The atomic variable vm_fault_info_updated is used to synchronize access to
adev->gmc.vm_fault_info between the interrupt handler and
get_vm_fault_info().
The default atomic functions like atomic_set() and atomic_read() do not
provide memory barriers. This allows for CPU instruction reordering,
meaning the memory accesses to vm_fault_info and the vm_fault_info_updated
flag are not guaranteed to occur in the intended order. This creates a
race condition that can lead to inconsistent or stale data being used.
The previous implementation, which used an explicit mb(), was incomplete
and inefficient. It failed to account for all potential CPU reorderings,
such as the access of vm_fault_info being reordered before the atomic_read
of the flag. This approach is also more verbose and less performant than
using the proper atomic functions with acquire/release semantics.
Fix this by switching to atomic_set_release() and atomic_read_acquire().
These functions provide the necessary acquire and release semantics,
which act as memory barriers to ensure the correct order of operations.
It is also more efficient and idiomatic than using explicit full memory
barriers.
Fixes: b97dfa27ef ("drm/amdgpu: save vm fault information for amdkfd")
Cc: stable@vger.kernel.org
Signed-off-by: Gui-Dong Han <hanguidong02@gmail.com>
Signed-off-by: Felix Kuehling <felix.kuehling@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
When we backup ring contents to reemit after a queue reset,
we don't backup ring contents from the bad context. When
we signal the fences, we should set an error on those
fences as well.
v2: misc cleanups
v3: add locking for fence error, fix comment (Christian)
v4: fix wrap around, locking (Christian)
Fixes: 77cc0da39c ("drm/amdgpu: track ring state associated with a fence")
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Chips which use the IP discovery firmware loaded by the driver
reported incorrect harvesting information in the ip discovery
table in sysfs because the driver only uses the ip discovery
firmware for populating sysfs and not for direct parsing for the
driver itself as such, the fields that are used to print the
harvesting info in sysfs report incorrect data for some IPs. Populate
the relevant fields for this case as well.
Fixes: 514678da56 ("drm/amdgpu/discovery: fix fw based ip discovery")
Acked-by: Tom St Denis <tom.stdenis@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
The Constant Engine found on gfx6-gfx10 HW has been a notorious source of
problems.
RADV never used it in the first place, radeonsi only used it for a few
releases around 2017 for gfx6-gfx9 before dropping support for it as
well.
While investigating another problem I just recently found that submitting
to the CE seems to be completely broken on gfx9 for quite a while.
Since nobody complained about that problem it most likely means that
nobody is using any of the affected radeonsi versions on current Linux
kernels any more.
So to potentially phase out the support for the CE and eliminate another
source of problems block submitting CE IBs unless it is enabled again
using a debug flag.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Acked-by: Timur Kristóf <timur.kristof@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Enabling ASPM causes randoms hangs on Tahiti and Oland on Zen4.
It's unclear if this is a platform-specific or GPU-specific issue.
Disable ASPM on SI for the time being.
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Pull more drm fixes from Dave Airlie:
"Just the follow up fixes for rc1 from the next branch, amdgpu and xe
mostly with a single v3d fix in there.
amdgpu:
- DC DCE6 fixes
- GPU reset fixes
- Secure diplay messaging cleanup
- MES fix
- GPUVM locking fixes
- PMFW messaging cleanup
- PCI US/DS switch handling fix
- VCN queue reset fix
- DC FPU handling fix
- DCN 3.5 fix
- DC mirroring fix
amdkfd:
- Fix kfd process ref leak
- mmap write lock handling fix
- Fix comments in IOCTL
xe:
- Fix build with clang 16
- Fix handling of invalid configfs syntax usage and spell out the
expected syntax in the documentation
- Do not try late bind firmware when running as VF since it shouldn't
handle firmware loading
- Fix idle assertion for local BOs
- Fix uninitialized variable for late binding
- Do not require perfmon_capable to expose free memory at page
granularity. Handle it like other drm drivers do
- Fix lock handling on suspend error path
- Fix I2C controller resume after S3
v3d:
- fix fence locking"
* tag 'drm-next-2025-10-11-1' of https://gitlab.freedesktop.org/drm/kernel: (34 commits)
drm/amd/display: Incorrect Mirror Cositing
drm/amd/display: Enable Dynamic DTBCLK Switch
drm/amdgpu: Report individual reset error
drm/amdgpu: partially revert "revert to old status lock handling v3"
drm/amd/display: Fix unsafe uses of kernel mode FPU
drm/amd/pm: Disable VCN queue reset on SMU v13.0.6 due to regression
drm/amdgpu: Fix general protection fault in amdgpu_vm_bo_reset_state_machine
drm/amdgpu: Check swus/ds for switch state save
drm/amdkfd: Fix two comments in kfd_ioctl.h
drm/amd/pm: Avoid interface mismatch messaging
drm/amdgpu: Merge amdgpu_vm_set_pasid into amdgpu_vm_init
drm/amd/amdgpu: Fix the mes version that support inv_tlbs
drm/amd: Check whether secure display TA loaded successfully
drm/amdkfd: Fix mmap write lock not release
drm/amdkfd: Fix kfd process ref leaking when userptr unmapping
drm/amdgpu: Fix for GPU reset being blocked by KIQ I/O.
drm/amd/display: Disable scaling on DCE6 for now
drm/amd/display: Properly disable scaling on DCE6
drm/amd/display: Properly clear SCL_*_FILTER_CONTROL on DCE6
drm/amd/display: Add missing DCE6 SCL_HORZ_FILTER_INIT* SRIs
...
If reinitialization of one of the GPUs fails after reset, it logs
failure on all subsequent GPUs eventhough they have resumed
successfully.
A sample log where only device at 0000:95:00.0 had a failure -
amdgpu 0000:15:00.0: amdgpu: GPU reset(19) succeeded!
amdgpu 0000:65:00.0: amdgpu: GPU reset(19) succeeded!
amdgpu 0000:75:00.0: amdgpu: GPU reset(19) succeeded!
amdgpu 0000:85:00.0: amdgpu: GPU reset(19) succeeded!
amdgpu 0000:95:00.0: amdgpu: GPU reset(19) failed
amdgpu 0000:e5:00.0: amdgpu: GPU reset(19) failed
amdgpu 0000:f5:00.0: amdgpu: GPU reset(19) failed
amdgpu 0000:05:00.0: amdgpu: GPU reset(19) failed
amdgpu 0000:15:00.0: amdgpu: GPU reset end with ret = -5
To avoid confusion, report the error for each device
separately and return the first error as the overall result.
Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Asad Kamal <asad.kamal@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
The CI systems are pointing out list corruptions, so we still need to
fix something here.
Keep the asserts, but revert the lock changes for now.
Fixes: 59e4405e9e ("drm/amdgpu: revert to old status lock handling v3")
Signed-off-by: Christian König <christian.koenig@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
As KFD no longer uses a separate PASID, the global amdgpu_vm_set_pasid()function is no longer necessary.
Merge its functionality directly intoamdgpu_vm_init() to simplify code flow and eliminate redundant locking.
v2: remove superflous check
adjust amdgpu_vm_fin and remove amdgpu_vm_set_pasid (Chritian)
v3: drop amdgpu_vm_assert_locked (Chritian)
Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4614
Fixes: 59e4405e9e ("drm/amdgpu: revert to old status lock handling v3")
Reviewed-by: Christian König <christian.koenig@amd.com>
Suggested-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Jesse Zhang <Jesse.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
MES version 0x83 is not stable to use the inv_tlbs API. Defer it to 0x84 vertsion.
Fixes: 85442bac84 ("drm/amd/amdgpu: Fix the mes version that support inv_tlbs")
Signed-off-by: Shaoyun Liu <shaoyun.liu@amd.com>
Reviewed-by: Michael Chen <michael.chen@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why]
Not all renoir hardware supports secure display. If the TA is present
but the feature isn't supported it will fail to load or send commands.
This shows ERR messages to the user that make it seems like there is
a problem.
[How]
Check the resp_status of the context to see if there was an error
before trying to send any secure display commands.
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/1415
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
kfd_lookup_process_by_pid hold the kfd process reference to ensure it
doesn't get destroyed while sending the segfault event to user space.
Calling kfd_lookup_process_by_pid as function parameter leaks the kfd
process refcount and miss the NULL pointer check if app process is
already destroyed.
Fixes: 2d274bf709 ("amd/amdkfd: Trigger segfault for early userptr unmmapping")
Signed-off-by: Philip Yang <Philip.Yang@amd.com>
Reviewed-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
There is some probability that reset workqueue is blocked by KIQ I/O for 10+ seconds after gpu hangs.
So we need to add a in_reset check during each KIQ register poll.
Signed-off-by: Heng Zhou <Heng.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Pull drm updates from Dave Airlie:
"cross-subsystem:
- i2c-hid: Make elan touch controllers power on after panel is
enabled
- dt bindings for STM32MP25 SoC
- pci vgaarb: use screen_info helpers
- rust pin-init updates
- add MEI driver for late binding firmware update/load
uapi:
- add ioctl for reassigning GEM handles
- provide boot_display attribute on boot-up devices
core:
- document DRM_MODE_PAGE_FLIP_EVENT
- add vendor specific recovery method to drm device wedged uevent
gem:
- Simplify gpuvm locking
ttm:
- add interface to populate buffers
sched:
- Fix race condition in trace code
atomic:
- Reallow no-op async page flips
display:
- dp: Fix command length
video:
- Improve pixel-format handling for struct screen_info
rust:
- drop Opaque<> from ioctl args
- Alloc:
- BorrowedPage type and AsPageIter traits
- Implement Vmalloc::to_page() and VmallocPageIter
- DMA/Scatterlist:
- Add dma::DataDirection and type alias for dma_addr_t
- Abstraction for struct scatterlist and sg_table
- DRM:
- simplify use of generics
- add DriverFile type alias
- drop Object::SIZE
- Rust:
- pin-init tree merge
- Various methods for AsBytes and FromBytes traits
gpuvm:
- Support madvice in Xe driver
gpusvm:
- fix hmm_pfn_to_map_order usage in gpusvm
bridge:
- Improve and fix ref counting on bridge management
- cdns-dsi: Various improvements to mode setting
- Support Solomon SSD2825 plus DT bindings
- Support Waveshare DSI2DPI plus DT bindings
- Support Content Protection property
- display-connector: Improve DP display detection
- Add support for Radxa Ra620 plus DT bindings
- adv7511: Provide SPD and HDMI infoframes
- it6505: Replace crypto_shash with sha()
- synopsys: Add support for DW DPTX Controller plus DT bindings
- adv7511: Write full Audio infoframe
- ite6263: Support vendor-specific infoframes
- simple: Add support for Realtek RTD2171 DP-to-HDMI plus DT bindings
panel:
- panel-edp: Support mt8189 Chromebooks; Support BOE NV140WUM-N64;
Support SHP LQ134Z1; Fixes
- panel-simple: Support Olimex LCD-OLinuXino-5CTS plus DT bindings
- Support Samsung AMS561RA01
- Support Hydis HV101HD1 plus DT bindings
- ilitek-ili9881c: Refactor mode setting; Add support for Bestar
BSD1218-A101KL68 LCD plus DT bindings
- lvds: Add support for Ampire AMP19201200B5TZQW-T03 to DT bindings
- edp: Add support for additonal mt8189 Chromebook panels
- lvds: Add DT bindings for EDT ETML0700Z8DHA
amdgpu:
- add CRIU support for gem objects
- RAS updates
- VCN SRAM load fixes
- EDID read fixes
- eDP ALPM support
- Documentation updates
- Rework PTE flag generation
- DCE6 fixes
- VCN devcoredump cleanup
- MMHUB client id fixes
- VCN 5.0.1 RAS support
- SMU 13.0.x updates
- Expanded PCIe DPC support
- Expanded VCN reset support
- VPE per queue reset support
- give kernel jobs unique id for tracing
- pre-populate exported buffers
- cyan skillfish updates
- make vbios build number available in sysfs
- userq updates
- HDCP updates
- support MMIO remap page as ttm pool
- JPEG parser updates
- DCE6 DC updates
- use devm for i2c buses
- GPUVM locking updates
- Drop non-DC DCE11 code
- improve fallback handling for pixel encoding
amdkfd:
- SVM/page migration fixes
- debugfs fixes
- add CRIO support for gem objects
- SVM updates
radeon:
- use dev_warn_once in CS parsers
xe:
- add madvise interface
- add DRM_IOCTL_XE_VM_QUERY_MEMORY_RANGE_ATTRS to query VMA count
and memory attributes
- drop L# bank mask reporting from media GT3 on Xe3+.
- add SLPC power_profile sysfs interface
- add configs attribs to add post/mid context-switch commands
- handle firmware reported hardware errors notifying userspace with
device wedged uevent
- use same dir structure across sysfs/debugfs
- cleanup and future proof vram region init
- add G-states and PCI link states to debugfs
- Add SRIOV support for CCS surfaces on Xe2+
- Enable SRIOV PF mode by default on supported platforms
- move flush to common code
- extended core workarounds for Xe2/3
- use DRM scheduler for delayed GT TLB invalidations
- configs improvements and allow VF device enablement
- prep work to expose mmio regions to userspace
- VF migration support added
- prepare GPU SVM for THP migration
- start fixing XE_PAGE_SIZE vs PAGE_SIZE
- add PSMI support for hw validation
- resize VF bars to max possible size according to number of VFs
- Ensure GT is in C0 during resume
- pre-populate exported buffers
- replace xe_hmm with gpusvm
- add more SVM GT stats to debugfs
- improve fake pci and WA kunnit handle for new platform testing
- Test GuC to GuC comms to add debugging
- use attribute groups to simplify sysfs registration
- add Late Binding firmware code to interact with MEI
i915:
- apply multiple JSL/EHL/Gen7/Gen6 workarounds properly
- protect against overflow in active_engine()
- Use try_cmpxchg64() in __active_lookup()
- include GuC registers in error state
- get rid of dev->struct_mutex
- iopoll: generalize read_poll_timout
- lots more display refactoring
- Reject HBR3 in any eDP Panel
- Prune modes for YUV420
- Display Wa fix, additions, and updates
- DP: Fix 2.7 Gbps link training on g4x
- DP: Adjust the idle pattern handling
- DP: Shuffle the link training code a bit
- Don't set/read the DSI C clock divider on GLK
- Enable_psr kernel parameter changes
- Type-C enabled/disconnected dp-alt sink
- Wildcat Lake enabling
- DP HDR updates
- DRAM detection
- wait PSR idle on dsb commit
- Remove FBC modulo 4 restriction for ADL-P+
- panic: refactor framebuffer allocation
habanalabs:
- debug/visibility improvements
- vmalloc-backed coherent mmap support
- HLDIO infrastructure
nova-core:
- various register!() macro improvements
- minor vbios/firmware fixes/refactoring
- advance firmware boot stages; process Booter and patch signatures
- process GSP and GSP bootloader
- Add r570.144 firmware bindings and update to it
- Move GSP boot code to own module
- Use new pin-init features to store driver's private data in a
single allocation
- Update ARef import from sync::aref
nova-drm:
- Update ARef import from sync::aref
tyr:
- initial driver skeleton for a rust driver for ARM Mali GPUs
- capable of powering up, query metadata and provide it to userspace.
msm:
- GPU and Core:
- in DT bindings describe clocks per GPU type
- GMU bandwidth voting for x1-85
- a623/a663 speedbins
- cleanup some remaining no-iommu leftovers after VM_BIND conversion
- fix GEM obj 32b size truncation
- add missing VM_BIND param validation
- IFPC for x1-85 and a750
- register xml and gen_header.py sync from mesa
- Display:
- add missing bindings for display on SC8180X
- added DisplayPort MST bindings
- conversion from round_rate() to determine_rate()
amdxdna:
- add IOCTL_AMDXDNA_GET_ARRAY
- support user space allocated buffers
- streamline PM interfaces
- Refactoring wrt. hardware contexts
- improve error reporting
nouveau:
- use GSP firmware by default
- improve error reporting
- Pre-populate exported buffers
ast:
- Clean up detection of DRAM config
exynos:
- add DSIM bridge driver support for Exynos7870
- Document Exynos7870 DSIM compatible in dt-binding
panthor:
- Print task/pid on errors
- Add support for Mali G710, G510, G310, Gx15, Gx20, Gx25
- Improve cache flushing
- Fail VM bind if BO has offset
renesas:
- convert to RUNTIME_PM_OPS
rcar-du:
- Make number of lanes configurable
- Use RUNTIME_PM_OPS
- Add support for DSI commands
rocket:
- Add driver for Rockchip NPU plus DT bindings
- Use kfree() and sizeof() correctly
- Test DMA status
rockchip:
- dsi2: Add support for RK3576 plus DT bindings
- Add support for RK3588 DPTX output
tidss:
- Use crtc_ fields for programming display mode
- Remove other drivers from aperture
pixpaper:
- Add support for Mayqueen Pixpaper plus DT bindings
v3d:
- Support querying nubmer of GPU resets for KHR_robustness
stm:
- Clean up logging
- ltdc: Add support support for STM32MP257F-EV1 plus DT bindings
sitronix:
- st7571-i2c: Add support for inverted displays and 2-bit grayscale
tidss:
- Convert to kernel's FIELD_ macros
vesadrm:
- Support 8-bit palette mode
imagination:
- Improve power management
- Add support for TH1520 GPU
- Support Risc-V architectures
v3d:
- Improve job management and locking
vkms:
- Support variants of ARGB8888, ARGB16161616, RGB565, RGB888 and P01x
- Spport YUV with 16-bit components"
* tag 'drm-next-2025-10-01' of https://gitlab.freedesktop.org/drm/kernel: (1455 commits)
drm/amd: Add name to modes from amdgpu_connector_add_common_modes()
drm/amd: Drop some common modes from amdgpu_connector_add_common_modes()
drm/amdgpu: update MODULE_PARM_DESC for freesync_video
drm/amd: Use dynamic array size declaration for amdgpu_connector_add_common_modes()
drm/amd/display: Share dce100_validate_global with DCE6-8
drm/amd/display: Share dce100_validate_bandwidth with DCE6-8
drm/amdgpu: Fix fence signaling race condition in userqueue
amd/amdkfd: enhance kfd process check in switch partition
amd/amdkfd: resolve a race in amdgpu_amdkfd_device_fini_sw
drm/amd/display: Reject modes with too high pixel clock on DCE6-10
drm/amd: Drop unnecessary check in amdgpu_connector_add_common_modes()
drm/amd/display: Only enable common modes for eDP and LVDS
drm/amdgpu: remove the redeclaration of variable i
drm/amdgpu/userq: assign an error code for invalid userq va
drm/amdgpu: revert "rework reserved VMID handling" v2
drm/amdgpu: remove leftover from enforcing isolation by VMID
drm/amdgpu: Add fallback to pipe reset if KCQ ring reset fails
accel/habanalabs: add Infineon version check
accel/habanalabs/gaudi2: read preboot status after recovering from dirty state
accel/habanalabs: add HL_GET_P_STATE passthrough type
...
Merge changes related to system sleep and runtime PM framework for
6.18-rc1:
- Annotate loops walking device links in the power management core
code as _srcu and add macros for walking device links to reduce the
likelihood of coding mistakes related to them (Rafael Wysocki)
- Document time units for *_time functions in the runtime PM API (Brian
Norris)
- Clear power.must_resume in noirq suspend error path to avoid resuming
a dependant device under a suspended parent or supplier (Rafael
Wysocki)
- Fix GFP mask handling during hybrid suspend and make the amdgpu
driver handle hybrid suspend correctly (Mario Limonciello, Rafael
Wysocki)
- Fix GFP mask handling after aborted hibernation in platform mode and
combine exit paths in power_down() to avoid code duplication (Rafael
Wysocki)
- Use vmalloc_array() and vcalloc() in the hibernation core to avoid
open-coded size computations (Qianfeng Rong)
- Fix typo in hibernation core code comment (Li Jun)
- Call pm_wakeup_clear() in the same place where other functions that do
bookkeeping prior to suspend_prepare() are called (Samuel Wu)
* pm-core:
PM: core: Add two macros for walking device links
PM: core: Annotate loops walking device links as _srcu
* pm-runtime:
PM: runtime: Documentation: ABI: Document time units for *_time
* pm-sleep:
PM: hibernate: Combine return paths in power_down()
PM: hibernate: Restrict GFP mask in power_down()
PM: hibernate: Fix pm_hibernation_mode_is_suspend() build breakage
drm/amd: Fix hybrid sleep
PM: hibernate: Add pm_hibernation_mode_is_suspend()
PM: hibernate: Fix hybrid-sleep
PM: sleep: core: Clear power.must_resume in noirq suspend error path
PM: sleep: Make pm_wakeup_clear() call more clear
PM: hibernate: Fix typo in memory bitmaps description comment
PM: hibernate: Use vmalloc_array() and vcalloc() to improve code
