linux-stable.git/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c, branch linux-6.8.y Linux kernel stable tree drm/amdkfd: Reset GPU on queue preemption failure 2024-04-17T09:23:37+00:00 Harish Kasiviswanathan Harish.Kasiviswanathan@amd.com 2024-03-26T19:32:46+00:00 38042ce7675ed2330f2ae7fb0f0a71ffcf1d2bea commit 8bdfb4ea95ca738d33ef71376c21eba20130f2eb upstream. Currently, with F32 HWS GPU reset is only when unmap queue fails. However, if compute queue doesn't repond to preemption request in time unmap will return without any error. In this case, only preemption error is logged and Reset is not triggered. Call GPU reset in this case also. Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com> Reviewed-by: Mukul Joshi <mukul.joshi@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Cc: stable@vger.kernel.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> 
commit 8bdfb4ea95ca738d33ef71376c21eba20130f2eb upstream.

Currently, with F32 HWS GPU reset is only when unmap queue fails.

However, if compute queue doesn't repond to preemption request in time
unmap will return without any error. In this case, only preemption error
is logged and Reset is not triggered. Call GPU reset in this case also.

Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com>
Reviewed-by: Mukul Joshi <mukul.joshi@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
drm/amdkfd: get doorbell's absolute offset based on the db_size 2023-10-09T21:02:34+00:00 Arvind Yadav Arvind.Yadav@amd.com 2023-10-09T17:13:16+00:00 367a0af43373d4f791cc8b466a659ecf5aa52377 Here, Adding db_size in byte to find the doorbell's absolute offset for both 32-bit and 64-bit doorbell sizes. So that doorbell offset will be aligned based on the doorbell size. v2: - Addressed the review comment from Felix. v3: - Adding doorbell_size as parameter to get db absolute offset. v4: Squash the two patches into one. Cc: Christian Koenig <christian.koenig@amd.com> Cc: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Shashank Sharma <shashank.sharma@amd.com> Signed-off-by: Arvind Yadav <Arvind.Yadav@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 
Here, Adding db_size in byte to find the doorbell's
absolute offset for both 32-bit and 64-bit doorbell sizes.
So that doorbell offset will be aligned based on the doorbell
size.

v2:
- Addressed the review comment from Felix.
v3:
- Adding doorbell_size as parameter to get db absolute offset.
v4:
  Squash the two patches into one.

Cc: Christian Koenig <christian.koenig@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Shashank Sharma <shashank.sharma@amd.com>
Signed-off-by: Arvind Yadav <Arvind.Yadav@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amdkfd: Use gpu_offset for user queue's wptr 2023-09-20T20:24:09+00:00 YuBiao Wang YuBiao.Wang@amd.com 2023-09-15T02:47:50+00:00 b157df66d81a05a1cbfb5f6faf196afb72d20e88 Directly use tbo's start address will miss the domain start offset. Need to use gpu_offset instead. Signed-off-by: YuBiao Wang <YuBiao.Wang@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 
Directly use tbo's start address will miss the domain start offset. Need
to use gpu_offset instead.

Signed-off-by: YuBiao Wang <YuBiao.Wang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amdkfd: fix add queue process context clear without runtime enable 2023-09-20T20:23:57+00:00 Jonathan Kim jonathan.kim@amd.com 2023-09-11T16:18:43+00:00 d92e55565c53eff6d0c549090487d0a1ef704e1c There are cases where HSA runtime is not enabled through the AMDKFD_IOC_RUNTIME_ENABLE call when adding queues and the MES ADD_QUEUE API should clear the MES process context instead of SET_SHADER_DEBUGGER. Such examples are legacy HSA runtime builds that do not support the current exception handling and running KFD tests. The only time ADD_QUEUE.skip_process_ctx_clear is required is for debugger use cases where a debugged process is always runtime enabled when adding a queue. Tested-by: Shikai Guo <shikai.guo@amd.com> Signed-off-by: Jonathan Kim <jonathan.kim@amd.com> Acked-by: Felix Kuehling <felix.kuehling@amd.com> Reviewed-by: Eric Huang <jinhuieric.huang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 
There are cases where HSA runtime is not enabled through the
AMDKFD_IOC_RUNTIME_ENABLE call when adding queues and the MES ADD_QUEUE
API should clear the MES process context instead of SET_SHADER_DEBUGGER.
Such examples are legacy HSA runtime builds that do not support the
current exception handling and running KFD tests.

The only time ADD_QUEUE.skip_process_ctx_clear is required is for
debugger use cases where a debugged process is always runtime enabled
when adding a queue.

Tested-by: Shikai Guo <shikai.guo@amd.com>
Signed-off-by: Jonathan Kim <jonathan.kim@amd.com>
Acked-by: Felix Kuehling <felix.kuehling@amd.com>
Reviewed-by: Eric Huang <jinhuieric.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amdgpu: Rename KGD_MAX_QUEUES to AMDGPU_MAX_QUEUES 2023-09-11T21:15:41+00:00 Mukul Joshi mukul.joshi@amd.com 2023-09-06T15:01:55+00:00 68fa72a437ea3b8ad6dd1d03247fcd670b8717c1 Rename KGD_MAX_QUEUES to AMDGPU_MAX_QUEUES to conform with the naming convention followed in amdgpu_gfx.h. No functional change. Signed-off-by: Mukul Joshi <mukul.joshi@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 
Rename KGD_MAX_QUEUES to AMDGPU_MAX_QUEUES to conform with
the naming convention followed in amdgpu_gfx.h. No functional
change.

Signed-off-by: Mukul Joshi <mukul.joshi@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amdkfd: Fix reg offset for setting CWSR grace period 2023-09-11T21:07:33+00:00 Mukul Joshi mukul.joshi@amd.com 2023-08-28T18:18:23+00:00 56d6daa3c743ab48d2f747d9373bfc4db0c8fd5a This patch fixes the case where the code currently passes absolute register address and not the reg offset, which HWS expects, when sending the PM4 packet to set/update CWSR grace period. Additionally, cleanup the signature of build_grace_period_packet_info function as it no longer needs the inst parameter. Signed-off-by: Mukul Joshi <mukul.joshi@amd.com> Reviewed-by: Jonathan Kim <jonathan.kim@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 
This patch fixes the case where the code currently passes
absolute register address and not the reg offset, which HWS
expects, when sending the PM4 packet to set/update CWSR grace
period. Additionally, cleanup the signature of
build_grace_period_packet_info function as it no longer needs
the inst parameter.

Signed-off-by: Mukul Joshi <mukul.joshi@amd.com>
Reviewed-by: Jonathan Kim <jonathan.kim@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amdkfd: Replace pr_err with dev_err 2023-08-30T19:51:17+00:00 Asad Kamal asad.kamal@amd.com 2023-08-25T07:52:19+00:00 80c74918aa7dbea8ccd42e1c973e4f19580825c0 Replace pr_err with dev_err to show the bus-id of failing device with kfd queue errors Signed-off-by: Asad Kamal <asad.kamal@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 
Replace pr_err with dev_err to show the bus-id of
failing device with kfd queue errors

Signed-off-by: Asad Kamal <asad.kamal@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amdkfd: fix double assign skip process context clear 2023-08-15T22:07:42+00:00 Jonathan Kim jonathan.kim@amd.com 2023-08-10T18:55:11+00:00 8b4c350c4d0e2df6ed24e670697662db18520acf Remove redundant assignment when skipping process ctx clear. Signed-off-by: Jonathan Kim <jonathan.kim@amd.com> Reviewed-by: Felix Kuehling <felix.kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 
Remove redundant assignment when skipping process ctx clear.

Signed-off-by: Jonathan Kim <jonathan.kim@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amdkfd: Use memdup_user() rather than duplicating its implementation 2023-08-15T22:07:41+00:00 Atul Raut rauji.raut@gmail.com 2023-08-09T05:30:18+00:00 bd6040b0ea04aca3f90bc81ccd2aa816d20292d7 To prevent its redundant implementation and streamline code, use memdup_user. This fixes warnings reported by Coccinelle: ./drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c:2811:13-20: WARNING opportunity for memdup_user Signed-off-by: Atul Raut <rauji.raut@gmail.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 
To prevent its redundant implementation and streamline
code, use memdup_user.

This fixes warnings reported by Coccinelle:
./drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c:2811:13-20: WARNING opportunity for memdup_user

Signed-off-by: Atul Raut <rauji.raut@gmail.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amdkfd: rename device_queue_manager_init_v10_navi10() 2023-08-11T18:47:39+00:00 Alex Deucher alexander.deucher@amd.com 2023-08-07T21:57:51+00:00 80e28aaf93db246c7eb0fb6430d1ba2b140d63b5 Drop the navi10 in the name for consistency with other families. All gfx10 parts use the same implementation. Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Tested-by: Mike Lothian <mike@fireburn.co.uk> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 
Drop the navi10 in the name for consistency with other
families.  All gfx10 parts use the same implementation.

Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Tested-by: Mike Lothian <mike@fireburn.co.uk>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>