iv/linux
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
linux/drivers/gpu/drm/amd/amdkfd
History
Lang Yu 9c29282ecb drm/amdkfd: reserve the BO before validating it 
Fix a warning.

v2: Avoid unmapping attachment repeatedly when ERESTARTSYS.

v3: Lock the BO before accessing ttm->sg to avoid race conditions.(Felix)

[   41.708711] WARNING: CPU: 0 PID: 1463 at drivers/gpu/drm/ttm/ttm_bo.c:846 ttm_bo_validate+0x146/0x1b0 [ttm]
[   41.708989] Call Trace:
[   41.708992]  <TASK>
[   41.708996]  ? show_regs+0x6c/0x80
[   41.709000]  ? ttm_bo_validate+0x146/0x1b0 [ttm]
[   41.709008]  ? __warn+0x93/0x190
[   41.709014]  ? ttm_bo_validate+0x146/0x1b0 [ttm]
[   41.709024]  ? report_bug+0x1f9/0x210
[   41.709035]  ? handle_bug+0x46/0x80
[   41.709041]  ? exc_invalid_op+0x1d/0x80
[   41.709048]  ? asm_exc_invalid_op+0x1f/0x30
[   41.709057]  ? amdgpu_amdkfd_gpuvm_dmaunmap_mem+0x2c/0x80 [amdgpu]
[   41.709185]  ? ttm_bo_validate+0x146/0x1b0 [ttm]
[   41.709197]  ? amdgpu_amdkfd_gpuvm_dmaunmap_mem+0x2c/0x80 [amdgpu]
[   41.709337]  ? srso_alias_return_thunk+0x5/0x7f
[   41.709346]  kfd_mem_dmaunmap_attachment+0x9e/0x1e0 [amdgpu]
[   41.709467]  amdgpu_amdkfd_gpuvm_dmaunmap_mem+0x56/0x80 [amdgpu]
[   41.709586]  kfd_ioctl_unmap_memory_from_gpu+0x1b7/0x300 [amdgpu]
[   41.709710]  kfd_ioctl+0x1ec/0x650 [amdgpu]
[   41.709822]  ? __pfx_kfd_ioctl_unmap_memory_from_gpu+0x10/0x10 [amdgpu]
[   41.709945]  ? srso_alias_return_thunk+0x5/0x7f
[   41.709949]  ? tomoyo_file_ioctl+0x20/0x30
[   41.709959]  __x64_sys_ioctl+0x9c/0xd0
[   41.709967]  do_syscall_64+0x3f/0x90
[   41.709973]  entry_SYSCALL_64_after_hwframe+0x6e/0xd8

Fixes: 101b8104307e ("drm/amdkfd: Move dma unmapping after TLB flush")
Signed-off-by: Lang Yu <Lang.Yu@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-01-31 17:37:53 -05:00
..
cik_event_interrupt.c
drm/amdkfd: add send exception operation
2023-06-09 12:36:01 -04:00
cik_int.h
…
cik_regs.h
…
cwsr_trap_handler_gfx8.asm
…
cwsr_trap_handler_gfx9.asm
drm/amdkfd: Use S_ENDPGM_SAVED in trap handler
2024-01-31 17:22:16 -05:00
cwsr_trap_handler_gfx10.asm
drm/amdkfd: Use S_ENDPGM_SAVED in trap handler
2024-01-31 17:22:16 -05:00
cwsr_trap_handler.h
drm/amdkfd: Use S_ENDPGM_SAVED in trap handler
2024-01-31 17:22:16 -05:00
Kconfig
drm/amdkfd: drop IOMMUv2 support
2023-08-11 14:47:25 -04:00
kfd_chardev.c
drm/amdkfd: reserve the BO before validating it
2024-01-31 17:37:53 -05:00
kfd_crat.c
drm/amdkfd: Populate cache info for GFX 9.4.3
2023-11-03 11:59:51 -04:00
kfd_crat.h
drm/amdkfd: Update cache info reporting for GFX v9.4.3
2023-09-11 18:17:20 -04:00
kfd_debug.c
drm/amdkfd: fix and enable ttmp setup for gfx11
2023-07-27 14:59:29 -04:00
kfd_debug.h
drm/amdkfd: fix and enable ttmp setup for gfx11
2023-07-27 14:59:29 -04:00
kfd_debugfs.c
drm/amdkfd: Introduce kfd_node struct (v5)
2023-06-09 09:42:27 -04:00
kfd_device_queue_manager_cik.c
drm/amdkfd: drop IOMMUv2 support
2023-08-11 14:47:25 -04:00
kfd_device_queue_manager_v9.c
drm/amdkfd: drop IOMMUv2 support
2023-08-11 14:47:25 -04:00
kfd_device_queue_manager_v10.c
drm/amdkfd: rename device_queue_manager_init_v10_navi10()
2023-08-11 14:47:39 -04:00
kfd_device_queue_manager_v11.c
…
kfd_device_queue_manager_vi.c
drm/amdkfd: drop IOMMUv2 support
2023-08-11 14:47:25 -04:00
kfd_device_queue_manager.c
drm/amdkfd: get doorbell's absolute offset based on the db_size
2023-10-09 17:02:34 -04:00
kfd_device_queue_manager.h
drm/amdkfd: rename device_queue_manager_init_v10_navi10()
2023-08-11 14:47:39 -04:00
kfd_device.c
drm/amdgpu: Use function for IP version check
2023-09-20 12:23:28 -04:00
kfd_doorbell.c
drm/amdkfd: get doorbell's absolute offset based on the db_size
2023-10-09 17:02:34 -04:00
kfd_events.c
drm/amdkfd: Copy HW exception data to user event
2023-11-29 16:24:43 -05:00
kfd_events.h
drm/amdkfd: add event_age tracking when receiving interrupt
2023-06-15 11:37:55 -04:00
kfd_flat_memory.c
Revert "drm/amdkfd: Relocate TBA/TMA to opposite side of VM hole"
2024-01-05 16:10:44 -05:00
kfd_int_process_v9.c
drm/amdkfd: use mask to get v9 interrupt sq data bits correctly
2023-08-31 18:10:28 -04:00
kfd_int_process_v10.c
drm/amdkfd: ratelimited SQ interrupt messages
2023-08-30 15:51:16 -04:00
kfd_int_process_v11.c
drm/amdkfd: ratelimited SQ interrupt messages
2023-08-30 15:51:16 -04:00
kfd_interrupt.c
drm/amdkfd: Introduce kfd_node struct (v5)
2023-06-09 09:42:27 -04:00
kfd_kernel_queue.c
drm/amdkfd: Introduce kfd_node struct (v5)
2023-06-09 09:42:27 -04:00
kfd_kernel_queue.h
drm/amdkfd: Introduce kfd_node struct (v5)
2023-06-09 09:42:27 -04:00
kfd_migrate.c
drm/amdkfd: Correct partial migration virtual addr
2024-01-31 17:21:32 -05:00
kfd_migrate.h
drm/amdkfd: Use partial migrations/mapping for GPU/CPU page faults in SVM
2023-12-06 15:22:32 -05:00
kfd_module.c
drm/amdkfd: Fixed kfd_process cleanup on module exit.
2023-03-15 18:45:26 -04:00
kfd_mqd_manager_cik.c
drm/amdkfd: Update CU masking for GFX 9.4.3
2023-09-11 18:17:27 -04:00
kfd_mqd_manager_v9.c
drm/amdkfd: Update CU masking for GFX 9.4.3
2023-09-11 18:17:27 -04:00
kfd_mqd_manager_v10.c
drm/amdgpu/gfx10: set UNORD_DISPATCH in compute MQDs
2024-01-25 15:48:51 -05:00
kfd_mqd_manager_v11.c
drm/amdgpu/gfx11: set UNORD_DISPATCH in compute MQDs
2024-01-25 15:48:57 -05:00
kfd_mqd_manager_vi.c
drm/amdkfd: Update CU masking for GFX 9.4.3
2023-09-11 18:17:27 -04:00
kfd_mqd_manager.c
drm/amdkfd: drop struct kfd_cu_info
2023-10-03 15:41:13 -04:00
kfd_mqd_manager.h
drm/amdkfd: Update CU masking for GFX 9.4.3
2023-09-11 18:17:27 -04:00
kfd_packet_manager_v9.c
drm/amdgpu: Use function for IP version check
2023-09-20 12:23:28 -04:00
kfd_packet_manager_vi.c
drm/amdgpu: add configurable grace period for unmap queues
2023-06-09 12:35:31 -04:00
kfd_packet_manager.c
drm/amdgpu: add configurable grace period for unmap queues
2023-06-09 12:35:31 -04:00
kfd_pasid.c
…
kfd_pm4_headers_ai.h
drm/amdkfd: prepare map process for single process debug devices
2023-06-09 12:35:36 -04:00
kfd_pm4_headers_aldebaran.h
…
kfd_pm4_headers_vi.h
…
kfd_pm4_headers.h
…
kfd_pm4_opcodes.h
…
kfd_priv.h
drm/amdkfd: Use correct drm device for cgroup permission check
2024-01-31 17:22:48 -05:00
kfd_process_queue_manager.c
drm/amdkfd: only flush mes process context if mes support is there
2023-12-15 12:15:13 -05:00
kfd_process.c
drm/amdkfd: Fix sparse __rcu annotation warnings
2024-01-09 15:44:13 -05:00
kfd_queue.c
…
kfd_smi_events.c
drm/amdkfd: Update SMI events for GFX9.4.3
2023-06-09 09:46:31 -04:00
kfd_smi_events.h
drm/amdkfd: Update SMI events for GFX9.4.3
2023-06-09 09:46:31 -04:00
kfd_svm.c
drm/amdkfd: Fix 'node' NULL check in 'svm_range_get_range_boundaries()'
2024-01-15 18:33:00 -05:00
kfd_svm.h
drm/amdkfd: Use partial migrations/mapping for GPU/CPU page faults in SVM
2023-12-06 15:22:32 -05:00
kfd_topology.c
drm/amdkfd: Fix iterator used outside loop in 'kfd_add_peer_prop()'
2024-01-05 16:10:43 -05:00
kfd_topology.h
drm/amdkfd: Update cache info reporting for GFX v9.4.3
2023-09-11 18:17:20 -04:00
Makefile
drm/amdkfd: drop IOMMUv2 support
2023-08-11 14:47:25 -04:00
soc15_int.h
drm/amdkfd: Update interrupt handling for GFX 9.4.3
2023-06-30 13:11:35 -04:00