diff options
| author | Dave Airlie <airlied@redhat.com> | 2024-06-27 17:18:49 +1000 |
|---|---|---|
| committer | Dave Airlie <airlied@redhat.com> | 2024-06-27 17:18:49 +1000 |
| commit | 365aa9f573995b46ca14a24165d85e31160e47b9 (patch) | |
| tree | 392d9656fc902e8198ee89efe278c78376478f8a /drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c | |
| parent | 541b1b0a8fc235bca355921eb7f3f59a8efa3e9a (diff) | |
| parent | 1ecef5589320fd56af599b624d59c355d162ac7b (diff) | |
Merge tag 'amd-drm-next-6.11-2024-06-22' of https://gitlab.freedesktop.org/agd5f/linux into drm-next
amd-drm-next-6.11-2024-06-22:
amdgpu:
- HPD fixes
- PSR fixes
- DCC updates
- DCN 4.0.1 fixes
- FAMS fixes
- Misc code cleanups
- SR-IOV fixes
- GPUVM TLB flush cleanups
- Make VCN less verbose
- ACPI backlight fixes
- MES fixes
- Firmware loading cleanups
- Replay fixes
- LTTPR fixes
- Trap handler fixes
- Cursor and overlay fixes
- Primary plane zpos fixes
- DML 2.1 fixes
- RAS updates
- USB4 fixes
- MALL fixes
- Reserved VMID fix
- Silence UBSAN warnings
amdkfd:
- Misc code cleanups
From: Alex Deucher <alexander.deucher@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240622152523.2267072-1-alexander.deucher@amd.com
Signed-off-by: Dave Airlie <airlied@redhat.com>
Diffstat (limited to 'drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c')
| -rw-r--r-- | drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c | 39 |
1 files changed, 16 insertions, 23 deletions
diff --git a/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c b/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c index f4c47492e0cd..6b71ee85ee65 100644 --- a/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c +++ b/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c @@ -249,38 +249,30 @@ static int xgpu_ai_set_mailbox_ack_irq(struct amdgpu_device *adev, return 0; } -static void xgpu_ai_mailbox_flr_work(struct work_struct *work) +static void xgpu_ai_ready_to_reset(struct amdgpu_device *adev) { - struct amdgpu_virt *virt = container_of(work, struct amdgpu_virt, flr_work); - struct amdgpu_device *adev = container_of(virt, struct amdgpu_device, virt); - int timeout = AI_MAILBOX_POLL_FLR_TIMEDOUT; - - /* block amdgpu_gpu_recover till msg FLR COMPLETE received, - * otherwise the mailbox msg will be ruined/reseted by - * the VF FLR. - */ - if (atomic_cmpxchg(&adev->reset_domain->in_gpu_reset, 0, 1) != 0) - return; - - down_write(&adev->reset_domain->sem); - - amdgpu_virt_fini_data_exchange(adev); - xgpu_ai_mailbox_trans_msg(adev, IDH_READY_TO_RESET, 0, 0, 0); +} +static int xgpu_ai_wait_reset(struct amdgpu_device *adev) +{ + int timeout = AI_MAILBOX_POLL_FLR_TIMEDOUT; do { if (xgpu_ai_mailbox_peek_msg(adev) == IDH_FLR_NOTIFICATION_CMPL) - goto flr_done; - + return 0; msleep(10); timeout -= 10; } while (timeout > 1); - dev_warn(adev->dev, "waiting IDH_FLR_NOTIFICATION_CMPL timeout\n"); + return -ETIME; +} -flr_done: - atomic_set(&adev->reset_domain->in_gpu_reset, 0); - up_write(&adev->reset_domain->sem); +static void xgpu_ai_mailbox_flr_work(struct work_struct *work) +{ + struct amdgpu_virt *virt = container_of(work, struct amdgpu_virt, flr_work); + struct amdgpu_device *adev = container_of(virt, struct amdgpu_device, virt); + + amdgpu_virt_fini_data_exchange(adev); /* Trigger recovery for world switch failure if no TDR */ if (amdgpu_device_should_recover_gpu(adev) @@ -417,7 +409,8 @@ const struct amdgpu_virt_ops xgpu_ai_virt_ops = { .req_full_gpu = xgpu_ai_request_full_gpu_access, .rel_full_gpu = xgpu_ai_release_full_gpu_access, .reset_gpu = xgpu_ai_request_reset, - .wait_reset = NULL, + .ready_to_reset = xgpu_ai_ready_to_reset, + .wait_reset = xgpu_ai_wait_reset, .trans_msg = xgpu_ai_mailbox_trans_msg, .req_init_data = xgpu_ai_request_init_data, .ras_poison_handler = xgpu_ai_ras_poison_handler, |