linux-stable.git/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c, branch v5.15.2 Linux kernel stable tree drm/amdgpu: revert "Add autodump debugfs node for gpu reset v8" 2021-11-06T13:13:31+00:00 Christian König christian.koenig@amd.com 2021-09-30T09:22:51+00:00 6ecad8906f05a44d3ce15ec3cba6871f2de93095 commit c8365dbda056578eebe164bf110816b1a39b4b7f upstream. This reverts commit 728e7e0cd61899208e924472b9e641dbeb0775c4. Further discussion reveals that this feature is severely broken and needs to be reverted ASAP. GPU reset can never be delayed by userspace even for debugging or otherwise we can run into in kernel deadlocks. Signed-off-by: Christian König <christian.koenig@amd.com> Acked-by: Alex Deucher <alexander.deucher@amd.com> Acked-by: Nirmoy Das <nirmoy.das@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> 
commit c8365dbda056578eebe164bf110816b1a39b4b7f upstream.

This reverts commit 728e7e0cd61899208e924472b9e641dbeb0775c4.

Further discussion reveals that this feature is severely broken
and needs to be reverted ASAP.

GPU reset can never be delayed by userspace even for debugging or
otherwise we can run into in kernel deadlocks.

Signed-off-by: Christian König <christian.koenig@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Acked-by: Nirmoy Das <nirmoy.das@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
drm/amdkfd: fix boot failure when iommu is disabled in Picasso. 2021-11-06T13:13:30+00:00 Yifan Zhang yifan1.zhang@amd.com 2021-10-11T12:37:01+00:00 f17dca0ab3f38b19c0f1b935f417f62d4a528723 commit afd18180c07026f94a80ff024acef5f4159084a4 upstream. When IOMMU disabled in sbios and kfd in iommuv2 path, iommuv2 init will fail. But this failure should not block amdgpu driver init. Reported-by: youling <youling257@gmail.com> Tested-by: youling <youling257@gmail.com> Signed-off-by: Yifan Zhang <yifan1.zhang@amd.com> Reviewed-by: James Zhu <James.Zhu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> 
commit afd18180c07026f94a80ff024acef5f4159084a4 upstream.

When IOMMU disabled in sbios and kfd in iommuv2 path, iommuv2
init will fail. But this failure should not block amdgpu driver init.

Reported-by: youling <youling257@gmail.com>
Tested-by: youling <youling257@gmail.com>
Signed-off-by: Yifan Zhang <yifan1.zhang@amd.com>
Reviewed-by: James Zhu <James.Zhu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
drm/amdgpu: handle the case of pci_channel_io_frozen only in amdgpu_pci_resume 2021-10-05T17:02:31+00:00 Guchun Chen guchun.chen@amd.com 2021-10-01T01:48:50+00:00 248b061689a40f4fed05252ee2c89f87cf26d7d8 In current code, when a PCI error state pci_channel_io_normal is detectd, it will report PCI_ERS_RESULT_CAN_RECOVER status to PCI driver, and PCI driver will continue the execution of PCI resume callback report_resume by pci_walk_bridge, and the callback will go into amdgpu_pci_resume finally, where write lock is releasd unconditionally without acquiring such lock first. In this case, a deadlock will happen when other threads start to acquire the read lock. To fix this, add a member in amdgpu_device strucutre to cache pci_channel_state, and only continue the execution in amdgpu_pci_resume when it's pci_channel_io_frozen. Fixes: c9a6b82f45e2 ("drm/amdgpu: Implement DPC recovery") Suggested-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com> Signed-off-by: Guchun Chen <guchun.chen@amd.com> Reviewed-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 
In current code, when a PCI error state pci_channel_io_normal is detectd,
it will report PCI_ERS_RESULT_CAN_RECOVER status to PCI driver, and PCI
driver will continue the execution of PCI resume callback report_resume by
pci_walk_bridge, and the callback will go into amdgpu_pci_resume
finally, where write lock is releasd unconditionally without acquiring
such lock first. In this case, a deadlock will happen when other threads
start to acquire the read lock.

To fix this, add a member in amdgpu_device strucutre to cache
pci_channel_state, and only continue the execution in amdgpu_pci_resume
when it's pci_channel_io_frozen.

Fixes: c9a6b82f45e2 ("drm/amdgpu: Implement DPC recovery")
Suggested-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Signed-off-by: Guchun Chen <guchun.chen@amd.com>
Reviewed-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amdgpu: init iommu after amdkfd device init 2021-10-05T17:02:20+00:00 Yifan Zhang yifan1.zhang@amd.com 2021-09-28T07:42:35+00:00 714d9e4574d54596973ee3b0624ee4a16264d700 This patch is to fix clinfo failure in Raven/Picasso: Number of platforms: 1 Platform Profile: FULL_PROFILE Platform Version: OpenCL 2.2 AMD-APP (3364.0) Platform Name: AMD Accelerated Parallel Processing Platform Vendor: Advanced Micro Devices, Inc. Platform Extensions: cl_khr_icd cl_amd_event_callback Platform Name: AMD Accelerated Parallel Processing Number of devices: 0 Signed-off-by: Yifan Zhang <yifan1.zhang@amd.com> Reviewed-by: James Zhu <James.Zhu@amd.com> Tested-by: James Zhu <James.Zhu@amd.com> Acked-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 
This patch is to fix clinfo failure in Raven/Picasso:

Number of platforms: 1
  Platform Profile: FULL_PROFILE
  Platform Version: OpenCL 2.2 AMD-APP (3364.0)
  Platform Name: AMD Accelerated Parallel Processing
  Platform Vendor: Advanced Micro Devices, Inc.
  Platform Extensions: cl_khr_icd cl_amd_event_callback

  Platform Name: AMD Accelerated Parallel Processing Number of devices: 0

Signed-off-by: Yifan Zhang <yifan1.zhang@amd.com>
Reviewed-by: James Zhu <James.Zhu@amd.com>
Tested-by: James Zhu <James.Zhu@amd.com>
Acked-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amdgpu: move iommu_resume before ip init/resume 2021-09-16T13:56:24+00:00 James Zhu James.Zhu@amd.com 2021-09-07T15:32:22+00:00 f02abeb0779700c308e661a412451b38962b8a0b Separate iommu_resume from kfd_resume, and move it before other amdgpu ip init/resume. Bug: https://bugzilla.kernel.org/show_bug.cgi?id=211277 Signed-off-by: James Zhu <James.Zhu@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Cc: stable@vger.kernel.org 
Separate iommu_resume from kfd_resume, and move it before
other amdgpu ip init/resume.

Bug: https://bugzilla.kernel.org/show_bug.cgi?id=211277
Signed-off-by: James Zhu <James.Zhu@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
drm/amdgpu: Cancel delayed work when GFXOFF is disabled 2021-08-20T16:09:44+00:00 Michel Dänzer mdaenzer@redhat.com 2021-08-17T08:23:25+00:00 90a9266269eb9f71af1f323c33e1dca53527bd22 schedule_delayed_work does not push back the work if it was already scheduled before, so amdgpu_device_delay_enable_gfx_off ran ~100 ms after the first time GFXOFF was disabled and re-enabled, even if GFXOFF was disabled and re-enabled again during those 100 ms. This resulted in frame drops / stutter with the upcoming mutter 41 release on Navi 14, due to constantly enabling GFXOFF in the HW and disabling it again (for getting the GPU clock counter). To fix this, call cancel_delayed_work_sync when the disable count transitions from 0 to 1, and only schedule the delayed work on the reverse transition, not if the disable count was already 0. This makes sure the delayed work doesn't run at unexpected times, and allows it to be lock-free. v2: * Use cancel_delayed_work_sync & mutex_trylock instead of mod_delayed_work. v3: * Make amdgpu_device_delay_enable_gfx_off lock-free (Christian König) v4: * Fix race condition between amdgpu_gfx_off_ctrl incrementing adev->gfx.gfx_off_req_count and amdgpu_device_delay_enable_gfx_off checking for it to be 0 (Evan Quan) Cc: stable@vger.kernel.org Reviewed-by: Evan Quan <evan.quan@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> # v3 Acked-by: Christian König <christian.koenig@amd.com> # v3 Signed-off-by: Michel Dänzer <mdaenzer@redhat.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 
schedule_delayed_work does not push back the work if it was already
scheduled before, so amdgpu_device_delay_enable_gfx_off ran ~100 ms
after the first time GFXOFF was disabled and re-enabled, even if GFXOFF
was disabled and re-enabled again during those 100 ms.

This resulted in frame drops / stutter with the upcoming mutter 41
release on Navi 14, due to constantly enabling GFXOFF in the HW and
disabling it again (for getting the GPU clock counter).

To fix this, call cancel_delayed_work_sync when the disable count
transitions from 0 to 1, and only schedule the delayed work on the
reverse transition, not if the disable count was already 0. This makes
sure the delayed work doesn't run at unexpected times, and allows it to
be lock-free.

v2:
* Use cancel_delayed_work_sync & mutex_trylock instead of
  mod_delayed_work.
v3:
* Make amdgpu_device_delay_enable_gfx_off lock-free (Christian König)
v4:
* Fix race condition between amdgpu_gfx_off_ctrl incrementing
  adev->gfx.gfx_off_req_count and amdgpu_device_delay_enable_gfx_off
  checking for it to be 0 (Evan Quan)

Cc: stable@vger.kernel.org
Reviewed-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> # v3
Acked-by: Christian König <christian.koenig@amd.com> # v3
Signed-off-by: Michel Dänzer <mdaenzer@redhat.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amd/amdgpu:flush ttm delayed work before cancel_sync 2021-08-18T22:23:00+00:00 YuBiao Wang YuBiao.Wang@amd.com 2021-08-17T09:36:33+00:00 691191a2f458e0176414cb5b3993b0c018cdc58c [Why] In some cases when we unload driver, warning call trace will show up in vram_mgr_fini which claims that LRU is not empty, caused by the ttm bo inside delay deleted queue. [How] We should flush delayed work to make sure the delay deleting is done. Signed-off-by: YuBiao Wang <YuBiao.Wang@amd.com> Reviewed-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 
[Why]
In some cases when we unload driver, warning call trace
will show up in vram_mgr_fini which claims that LRU is not empty, caused
by the ttm bo inside delay deleted queue.

[How]
We should flush delayed work to make sure the delay deleting is done.

Signed-off-by: YuBiao Wang <YuBiao.Wang@amd.com>
Reviewed-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amd/amdgpu embed hw_fence into amdgpu_job 2021-08-16T19:16:58+00:00 Jack Zhang Jack.Zhang1@amd.com 2021-05-12T07:06:35+00:00 c530b02f39850a639b72d01ebbf7e5d745c60831 Why: Previously hw fence is alloced separately with job. It caused historical lifetime issues and corner cases. The ideal situation is to take fence to manage both job and fence's lifetime, and simplify the design of gpu-scheduler. How: We propose to embed hw_fence into amdgpu_job. 1. We cover the normal job submission by this method. 2. For ib_test, and submit without a parent job keep the legacy way to create a hw fence separately. v2: use AMDGPU_FENCE_FLAG_EMBED_IN_JOB_BIT to show that the fence is embedded in a job. v3: remove redundant variable ring in amdgpu_job v4: add tdr sequence support for this feature. Add a job_run_counter to indicate whether this job is a resubmit job. v5 add missing handling in amdgpu_fence_enable_signaling Signed-off-by: Jingwen Chen <Jingwen.Chen2@amd.com> Signed-off-by: Jack Zhang <Jack.Zhang7@hotmail.com> Reviewed-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com> Reviewed by: Monk Liu <monk.liu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 
Why: Previously hw fence is alloced separately with job.
It caused historical lifetime issues and corner cases.
The ideal situation is to take fence to manage both job
and fence's lifetime, and simplify the design of gpu-scheduler.

How:
We propose to embed hw_fence into amdgpu_job.
1. We cover the normal job submission by this method.
2. For ib_test, and submit without a parent job keep the
legacy way to create a hw fence separately.
v2:
use AMDGPU_FENCE_FLAG_EMBED_IN_JOB_BIT to show that the fence is
embedded in a job.
v3:
remove redundant variable ring in amdgpu_job
v4:
add tdr sequence support for this feature. Add a job_run_counter to
indicate whether this job is a resubmit job.
v5
add missing handling in amdgpu_fence_enable_signaling

Signed-off-by: Jingwen Chen <Jingwen.Chen2@amd.com>
Signed-off-by: Jack Zhang <Jack.Zhang7@hotmail.com>
Reviewed-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Reviewed by: Monk Liu <monk.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amd/amdgpu: skip locking delayed work if not initialized. 2021-08-09T19:42:30+00:00 YuBiao Wang YuBiao.Wang@amd.com 2021-08-05T02:32:40+00:00 e78b3197dbf73fc0695dd019e388576d0a551830 When init failed in early init stage, amdgpu_object has not been initialized, so hasn't the ttm delayed queue functions. Signed-off-by: YuBiao Wang <YuBiao.Wang@amd.com> Reviewed-by: Emily.Deng <Emily.Deng@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 
When init failed in early init stage, amdgpu_object has
not been initialized, so hasn't the ttm delayed queue functions.

Signed-off-by: YuBiao Wang <YuBiao.Wang@amd.com>
Reviewed-by: Emily.Deng <Emily.Deng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amdgpu: avoid over-handle of fence driver fini in s3 test (v2) 2021-08-06T01:17:58+00:00 Guchun Chen guchun.chen@amd.com 2021-07-29T10:35:13+00:00 067f44c8b4590c3f24d21a037578a478590f2175 In amdgpu_fence_driver_hw_fini, no need to call drm_sched_fini to stop scheduler in s3 test, otherwise, fence related failure will arrive after resume. To fix this and for a better clean up, move drm_sched_fini from fence_hw_fini to fence_sw_fini, as it's part of driver shutdown, and should never be called in hw_fini. v2: rename amdgpu_fence_driver_init to amdgpu_fence_driver_sw_init, to keep sw_init and sw_fini paired. Bug: https://gitlab.freedesktop.org/drm/amd/-/issues/1668 Fixes: 8d35a2596164c1 ("drm/amdgpu: adjust fence driver enable sequence") Suggested-by: Christian König <christian.koenig@amd.com> Tested-by: Mike Lothian <mike@fireburn.co.uk> Signed-off-by: Guchun Chen <guchun.chen@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 
In amdgpu_fence_driver_hw_fini, no need to call drm_sched_fini to stop
scheduler in s3 test, otherwise, fence related failure will arrive
after resume. To fix this and for a better clean up, move drm_sched_fini
from fence_hw_fini to fence_sw_fini, as it's part of driver shutdown, and
should never be called in hw_fini.

v2: rename amdgpu_fence_driver_init to amdgpu_fence_driver_sw_init,
to keep sw_init and sw_fini paired.

Bug: https://gitlab.freedesktop.org/drm/amd/-/issues/1668
Fixes: 8d35a2596164c1 ("drm/amdgpu: adjust fence driver enable sequence")
Suggested-by: Christian König <christian.koenig@amd.com>
Tested-by: Mike Lothian <mike@fireburn.co.uk>
Signed-off-by: Guchun Chen <guchun.chen@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>