linux-stable.git/drivers/gpu/drm/amd/amdkfd, branch v4.16.2 Linux kernel stable tree drm/amdkfd: Deallocate SDMA queues correctly 2018-03-23T19:30:34+00:00 Felix Kuehling Felix.Kuehling@amd.com 2018-03-23T19:30:34+00:00 72a01d231dcb22a276209c7a924a8cd475fbaa9b Deallocate SDMA queues during abnormal process termination and when queue creation fails after the SDMA allocation. Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com> Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com> Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com> 
Deallocate SDMA queues during abnormal process termination and when
queue creation fails after the SDMA allocation.

Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
drm/amdkfd: Fix scratch memory with HWS enabled 2018-03-23T19:30:33+00:00 Felix Kuehling Felix.Kuehling@amd.com 2018-03-23T19:30:33+00:00 c70a36268799cf2f902b5a31e452571fcb96bfe9 Program sh_hidden_private_base_vmid correctly in the map-process PM4 packet. Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com> Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com> Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com> 
Program sh_hidden_private_base_vmid correctly in the map-process
PM4 packet.

Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
drm/amdkfd: Fix potential NULL pointer dereferences 2018-01-10T23:15:09+00:00 Gustavo A. R. Silva garsilva@embeddedor.com 2018-01-10T23:15:09+00:00 2e3dca53654ddc51bada4cee4ca32d0d266fc9dd In case kfd_get_process_device_data returns null, there are some null pointer dereferences in functions kfd_bind_processes_to_device and kfd_unbind_processes_from_device. Fix this by printing a WARN_ON for PDDs that aren't found and skip them with continue statements. Addresses-Coverity-ID: 1463794 ("Dereference null return value") Addresses-Coverity-ID: 1463772 ("Dereference null return value") Suggested-by: Felix Kuehling <felix.kuehling@amd.com> Signed-off-by: Gustavo A. R. Silva <garsilva@embeddedor.com> Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com> Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com> 
In case kfd_get_process_device_data returns null, there are some
null pointer dereferences in functions kfd_bind_processes_to_device
and kfd_unbind_processes_from_device.

Fix this by printing a WARN_ON for PDDs that aren't found and skip
them with continue statements.

Addresses-Coverity-ID: 1463794 ("Dereference null return value")
Addresses-Coverity-ID: 1463772 ("Dereference null return value")
Suggested-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Gustavo A. R. Silva <garsilva@embeddedor.com>
Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
drm/amdkfd: add ull suffix to 64bit defines 2018-01-10T10:55:17+00:00 Oded Gabbay oded.gabbay@gmail.com 2018-01-10T10:55:17+00:00 a1235e10ee9a9b4d5840c6edea74479fa87c9119 Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> 
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
drm/amdkfd: don't always call execute_queues_cpsch() 2018-01-02T18:10:50+00:00 Yong Zhao yong.zhao@amd.com 2018-01-02T18:10:50+00:00 40a526dc1edbe08000f49aeb6c7b01406d58a073 When destroying an inactive queue, we don't need to call execute_queues_cpsch. Signed-off-by: Yong Zhao <yong.zhao@amd.com> Reviewed-by: Oak Zeng <oak.zeng@amd.com> Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com> 
When destroying an inactive queue, we don't need to call
execute_queues_cpsch.

Signed-off-by: Yong Zhao <yong.zhao@amd.com>
Reviewed-by: Oak Zeng <oak.zeng@amd.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
drm/amdkfd: Fix return value 0 when execute_queues_cpsch fails 2018-01-02T18:10:49+00:00 Yong Zhao yong.zhao@amd.com 2018-01-02T18:10:49+00:00 9e8272240bee42ebafa9d4994c1d65916a761f5d Signed-off-by: Yong Zhao <yong.zhao@amd.com> Reviewed-by: Oak Zeng <oak.zeng@amd.com> Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com> 
Signed-off-by: Yong Zhao <yong.zhao@amd.com>
Reviewed-by: Oak Zeng <oak.zeng@amd.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
drm/amdkfd: Ignore ACPI CRAT for non-APU systems 2017-12-09T04:09:04+00:00 Harish Kasiviswanathan Harish.Kasiviswanathan@amd.com 2017-12-09T04:09:04+00:00 b441093e40e14d8305f0bee37cbaa3fa2e6ce8d0 Some AMD motherboards without an APU have a broken CRAT table which causes KFD initialization failures or incorrect information about NUMA nodes, CPU cores or system memory. Ignore CRAT tables without GPUs and rely on KFD's code to create a CRAT table for the CPU. Signed-off-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com> Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com> Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com> Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com> 
Some AMD motherboards without an APU have a broken CRAT table which
causes KFD initialization failures or incorrect information about
NUMA nodes, CPU cores or system memory. Ignore CRAT tables without
GPUs and rely on KFD's code to create a CRAT table for the CPU.

Signed-off-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com>
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
drm/amdkfd: Module option to disable CRAT table 2017-12-09T04:09:03+00:00 Felix Kuehling Felix.Kuehling@amd.com 2017-12-09T04:09:03+00:00 ebcfd1e276207e4436a6d15a81f7e151a103565c Some systems have broken CRAT tables. Add a module option to ignore a CRAT table. Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com> Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com> Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com> 
Some systems have broken CRAT tables. Add a module option to ignore
a CRAT table.

Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
drm/amdkfd: Add AQL Queue Memory flag on topology 2017-12-09T04:09:02+00:00 Ben Goz ben.goz@amd.com 2017-12-09T04:09:02+00:00 413e85d5d38f9b000561b6b9adf4182803b50af0 This is needed for enabling a user-mode workaround for an AQL queue wrapping HW bug on Tonga. Signed-off-by: Ben Goz <ben.goz@amd.com> Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com> Acked-by: Oded Gabbay <oded.gabbay@gmail.com> Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com> 
This is needed for enabling a user-mode workaround for an AQL queue
wrapping HW bug on Tonga.

Signed-off-by: Ben Goz <ben.goz@amd.com>
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Acked-by: Oded Gabbay <oded.gabbay@gmail.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
drm/amdkfd: Fixup incorrect info in the CZ CRAT table 2017-12-09T04:09:01+00:00 Philip Cox Philip.Cox@amd.com 2017-12-09T04:09:01+00:00 70f372bffcc84c9790acbae07d25d7dd0eca4d9f * Wrong value for max_waves_per_simd * Missing ATC capability bit Signed-off-by: Philip Cox <Philip.Cox@amd.com> Signed-off-by: Kent Russell <kent.russell@amd.com> Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com> Acked-by: Oded Gabbay <oded.gabbay@gmail.com> Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com> 
* Wrong value for max_waves_per_simd
* Missing ATC capability bit

Signed-off-by: Philip Cox <Philip.Cox@amd.com>
Signed-off-by: Kent Russell <kent.russell@amd.com>
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Acked-by: Oded Gabbay <oded.gabbay@gmail.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>