MES FW uses address(mqd_addr + sizeof(struct mqd) + 3*sizeof(uint32_t)) as fence address and writes a 32 bit fence value to this address. Driver needs to allocate some extra memory(at least 4 DWs) in addition to sizeof(struct mqd) as mqd memory.
For gfx11/12, sizeof(struct mqd) < PAGE_SIZE, KGD allocates mqd memory with PAGE_SIZE aligned works. For gfx12.1, sizeof(struct mqd) == PAGE_SIZE, it doesn't work. KFD mqd manager hardcodes mqd size to PAGE_SIZE/MQD_SIZE across different IP versions to solve this issue. To avoid hardcoding in differnet places and across different IP versions. Let's use AMDGPU_MQD_SIZE_ALIGN instead. It is used in two places. 1. mqd memory alloction 2. mqd stride handling for multi xcc config v2: Use AMDGPU_GPU_PAGE_ALIGN. (Mukul) Signed-off-by: Lang Yu <[email protected]> Reviewed-by: David Belanger <[email protected]> (v1) --- drivers/gpu/drm/amd/amdgpu/amdgpu.h | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h index 9c11535c44c6..8c6edb0f58dd 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h @@ -1239,6 +1239,14 @@ struct amdgpu_device { struct amdgpu_kfd_dev kfd; }; +/* + * MES FW uses address(mqd_addr + sizeof(struct mqd) + 3*sizeof(uint32_t)) + * as fence address and writes a 32 bit fence value to this address. + * Driver needs to allocate at least 4 DWs extra memory in addition to + * sizeof(struct mqd). Add 8 DWs and align to AMDGPU_GPU_PAGE_SIZE for safety. + */ +#define AMDGPU_MQD_SIZE_ALIGN(mqd_size) AMDGPU_GPU_PAGE_ALIGN(((mqd_size) + 32)) + static inline uint32_t amdgpu_ip_version(const struct amdgpu_device *adev, uint8_t ip, uint8_t inst) { -- 2.34.1
