diff options
| author | Dave Airlie <airlied@redhat.com> | 2024-04-30 14:42:54 +1000 |
|---|---|---|
| committer | Dave Airlie <airlied@redhat.com> | 2024-04-30 14:43:00 +1000 |
| commit | 4a56c0ed5aa0bcbe1f5f7d755fb1fe1ebf48ae9c (patch) | |
| tree | ec7d1f08d654ef4b6e596c891eadb504df2f611e /drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h | |
| parent | 68b89e23c2282877b0d411e07a3ef90490d6fe30 (diff) | |
| parent | b77bef36015c501f1e0f51db72c55e6dcd8bdd48 (diff) | |
Merge tag 'amd-drm-next-6.10-2024-04-26' of https://gitlab.freedesktop.org/agd5f/linux into drm-next
amd-drm-next-6.10-2024-04-26:
amdgpu:
- Misc code cleanups and refactors
- Support setting reset method at runtime
- Report OD status
- SMU 14.0.1 fixes
- SDMA 4.4.2 fixes
- VPE fixes
- MES fixes
- Update BO eviction priorities
- UMSCH fixes
- Reset fixes
- Freesync fixes
- GFXIP 9.4.3 fixes
- SDMA 5.2 fixes
- MES UAF fix
- RAS updates
- Devcoredump updates for dumping IP state
- DSC fixes
- JPEG fix
- Fix VRAM memory accounting
- VCN 5.0 fixes
- MES fixes
- UMC 12.0 updates
- Modify contiguous flags handling
- Initial support for mapping kernel queues via MES
amdkfd:
- Fix rescheduling of restore worker
- VRAM accounting for SVM migrations
- mGPU fix
- Enable SQ watchpoint for gfx10
Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Alex Deucher <alexander.deucher@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240426221245.1613332-1-alexander.deucher@amd.com
Diffstat (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h')
| -rw-r--r-- | drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h | 19 |
1 files changed, 18 insertions, 1 deletions
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h index 563b0249247e..5f50c69c3cec 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h @@ -52,6 +52,8 @@ #define LOOP_UMC_EACH_NODE_INST_AND_CH(node_inst, umc_inst, ch_inst) \ LOOP_UMC_NODE_INST((node_inst)) LOOP_UMC_INST_AND_CH((umc_inst), (ch_inst)) +/* Page retirement tag */ +#define UMC_ECC_NEW_DETECTED_TAG 0x1 typedef int (*umc_func)(struct amdgpu_device *adev, uint32_t node_inst, uint32_t umc_inst, uint32_t ch_inst, void *data); @@ -66,6 +68,8 @@ struct amdgpu_umc_ras { void *ras_error_status); bool (*check_ecc_err_status)(struct amdgpu_device *adev, enum amdgpu_mca_error_type type, void *ras_error_status); + int (*update_ecc_status)(struct amdgpu_device *adev, + uint64_t status, uint64_t ipid, uint64_t addr); }; struct amdgpu_umc_funcs { @@ -102,10 +106,13 @@ int amdgpu_umc_ras_sw_init(struct amdgpu_device *adev); int amdgpu_umc_ras_late_init(struct amdgpu_device *adev, struct ras_common_if *ras_block); int amdgpu_umc_poison_handler(struct amdgpu_device *adev, enum amdgpu_ras_block block, uint32_t reset); +int amdgpu_umc_pasid_poison_handler(struct amdgpu_device *adev, + enum amdgpu_ras_block block, uint16_t pasid, + pasid_notify pasid_fn, void *data, uint32_t reset); int amdgpu_umc_process_ecc_irq(struct amdgpu_device *adev, struct amdgpu_irq_src *source, struct amdgpu_iv_entry *entry); -void amdgpu_umc_fill_error_record(struct ras_err_data *err_data, +int amdgpu_umc_fill_error_record(struct ras_err_data *err_data, uint64_t err_addr, uint64_t retired_page, uint32_t channel_index, @@ -122,4 +129,14 @@ int amdgpu_umc_loop_channels(struct amdgpu_device *adev, int amdgpu_umc_bad_page_polling_timeout(struct amdgpu_device *adev, uint32_t reset, uint32_t timeout_ms); + +int amdgpu_umc_update_ecc_status(struct amdgpu_device *adev, + uint64_t status, uint64_t ipid, uint64_t addr); +int amdgpu_umc_build_pages_hash(struct amdgpu_device *adev, + uint64_t *pfns, int len, uint64_t *val); +int amdgpu_umc_logs_ecc_err(struct amdgpu_device *adev, + struct radix_tree_root *ecc_tree, struct ras_ecc_err *ecc_err); + +void amdgpu_umc_handle_bad_pages(struct amdgpu_device *adev, + void *ras_error_status); #endif |