diff options
| author | Le Ma <le.ma@amd.com> | 2024-08-07 17:33:00 +0800 |
|---|---|---|
| committer | Alex Deucher <alexander.deucher@amd.com> | 2024-12-10 10:26:51 -0500 |
| commit | 5a7c8c579dd1d35dc385724fd34ffe94f90d872f (patch) | |
| tree | dc337a41834ff96f046df05112ecd1ae6e3ba733 | |
| parent | 5690011a7006f8a2ce1dbf32d733c3b1454af6da (diff) | |
drm/amdkfd: update the cwsr area size for gfx950
Update cwsr area size for gfx950 to fit the new user queue buffer validation.
The size of LDS calculation is referred from gfx950 thunk implementation.
Signed-off-by: Le Ma <le.ma@amd.com>
Acked-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
| -rw-r--r-- | drivers/gpu/drm/amd/amdkfd/kfd_queue.c | 10 |
1 files changed, 6 insertions, 4 deletions
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_queue.c b/drivers/gpu/drm/amd/amdkfd/kfd_queue.c index ad29634f8b44..ecccd7adbab4 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_queue.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_queue.c @@ -394,7 +394,8 @@ static u32 kfd_get_vgpr_size_per_cu(u32 gfxv) if ((gfxv / 100 * 100) == 90400 || /* GFX_VERSION_AQUA_VANJARAM */ gfxv == 90010 || /* GFX_VERSION_ALDEBARAN */ - gfxv == 90008) /* GFX_VERSION_ARCTURUS */ + gfxv == 90008 || /* GFX_VERSION_ARCTURUS */ + gfxv == 90500) vgpr_size = 0x80000; else if (gfxv == 110000 || /* GFX_VERSION_PLUM_BONITO */ gfxv == 110001 || /* GFX_VERSION_WHEAT_NAS */ @@ -405,9 +406,10 @@ static u32 kfd_get_vgpr_size_per_cu(u32 gfxv) return vgpr_size; } -#define WG_CONTEXT_DATA_SIZE_PER_CU(gfxv) \ +#define WG_CONTEXT_DATA_SIZE_PER_CU(gfxv, props) \ (kfd_get_vgpr_size_per_cu(gfxv) + SGPR_SIZE_PER_CU +\ - LDS_SIZE_PER_CU + HWREG_SIZE_PER_CU) + (((gfxv) == 90500) ? (props->lds_size_in_kb << 10) : LDS_SIZE_PER_CU) +\ + HWREG_SIZE_PER_CU) #define CNTL_STACK_BYTES_PER_WAVE(gfxv) \ ((gfxv) >= 100100 ? 12 : 8) /* GFX_VERSION_NAVI10*/ @@ -431,7 +433,7 @@ void kfd_queue_ctx_save_restore_size(struct kfd_topology_device *dev) min(cu_num * 40, props->array_count / props->simd_arrays_per_engine * 512) : cu_num * 32; - wg_data_size = ALIGN(cu_num * WG_CONTEXT_DATA_SIZE_PER_CU(gfxv), PAGE_SIZE); + wg_data_size = ALIGN(cu_num * WG_CONTEXT_DATA_SIZE_PER_CU(gfxv, props), PAGE_SIZE); ctl_stack_size = wave_num * CNTL_STACK_BYTES_PER_WAVE(gfxv) + 8; ctl_stack_size = ALIGN(SIZEOF_HSA_USER_CONTEXT_SAVE_AREA_HEADER + ctl_stack_size, PAGE_SIZE); |