diff options
| author | Mukul Joshi <mukul.joshi@amd.com> | 2024-09-16 14:33:58 -0400 |
|---|---|---|
| committer | Alex Deucher <alexander.deucher@amd.com> | 2024-09-25 12:56:00 -0400 |
| commit | 6ae9e1aba97e4cdaa31a0bfdc07497ad0e915c84 (patch) | |
| tree | 821af9df5ef4c77d195ca1142f28ffdc750a2c88 /drivers/gpu/drm/amd/include | |
| parent | drm/amdgpu: skip coredump after job timeout in SRIOV (diff) | |
| download | linux-6ae9e1aba97e4cdaa31a0bfdc07497ad0e915c84.tar.gz linux-6ae9e1aba97e4cdaa31a0bfdc07497ad0e915c84.zip | |
drm/amdkfd: Update logic for CU occupancy calculations
Currently, the code uses the IH_VMID_X_LUT register to map
a queue's vmid to the corresponding PASID. This logic is racy
since CP can update the VMID-PASID mapping anytime especially
when there are more processes than number of vmids. Update the
logic to calculate CU occupancy by matching doorbell offset of
the queue with valid wave counts against the process's queues.
Signed-off-by: Mukul Joshi <mukul.joshi@amd.com>
Reviewed-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Diffstat (limited to 'drivers/gpu/drm/amd/include')
| -rw-r--r-- | drivers/gpu/drm/amd/include/kgd_kfd_interface.h | 10 |
1 files changed, 8 insertions, 2 deletions
diff --git a/drivers/gpu/drm/amd/include/kgd_kfd_interface.h b/drivers/gpu/drm/amd/include/kgd_kfd_interface.h index 7744ca3ef4b1..e3e635a31b8a 100644 --- a/drivers/gpu/drm/amd/include/kgd_kfd_interface.h +++ b/drivers/gpu/drm/amd/include/kgd_kfd_interface.h @@ -71,6 +71,11 @@ enum kgd_memory_pool { KGD_POOL_FRAMEBUFFER = 3, }; +struct kfd_cu_occupancy { + u32 wave_cnt; + u32 doorbell_off; +}; + /** * enum kfd_sched_policy * @@ -313,8 +318,9 @@ struct kfd2kgd_calls { uint32_t grace_period, uint32_t *reg_offset, uint32_t *reg_data); - void (*get_cu_occupancy)(struct amdgpu_device *adev, int pasid, - int *wave_cnt, int *max_waves_per_cu, uint32_t inst); + void (*get_cu_occupancy)(struct amdgpu_device *adev, + struct kfd_cu_occupancy *cu_occupancy, + int *max_waves_per_cu, uint32_t inst); void (*program_trap_handler_settings)(struct amdgpu_device *adev, uint32_t vmid, uint64_t tba_addr, uint64_t tma_addr, uint32_t inst); |