On Wed, Nov 6, 2024 at 1:49 AM Victor Zhao <Victor.Zhao@xxxxxxx> wrote:
>
> From: Gang Ba <gaba@xxxxxxx>
>
> amdgpu_amdkfd_alloc_gtt_mem currently allocates USWC memory.
> It uses write-combining for CPU access, which is slow for reading.
> Add a new parameter to amdgpu_amdkfd_alloc_gtt_mem to allocate
> normal GTT memory.
>
> Signed-off-by: Gang Ba <gaba@xxxxxxx>
> Reviewed-by: Felix Kuehling <Felix.Kuehling@xxxxxxx>
> Signed-off-by: Victor Zhao <Victor.Zhao@xxxxxxx>
Reviewed-by: Alex Deucher <alexander.deucher@xxxxxxx>
> ---
> drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c | 7 +++++--
> drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h | 2 +-
> drivers/gpu/drm/amd/amdkfd/kfd_device.c | 2 +-
> drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c | 2 +-
> drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c | 2 +-
> drivers/gpu/drm/amd/amdkfd/kfd_process.c | 2 +-
> drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c | 2 +-
> 7 files changed, 11 insertions(+), 8 deletions(-)
>
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
> index 713be49540d4..acb762855c24 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
> @@ -294,7 +294,7 @@ void amdgpu_amdkfd_gpu_reset(struct amdgpu_device *adev)
>
> int amdgpu_amdkfd_alloc_gtt_mem(struct amdgpu_device *adev, size_t size,
> void **mem_obj, uint64_t *gpu_addr,
> - void **cpu_ptr, bool cp_mqd_gfx9)
> + void **cpu_ptr, bool cp_mqd_gfx9, bool is_uswc_mode)
> {
> struct amdgpu_bo *bo = NULL;
> struct amdgpu_bo_param bp;
> @@ -305,7 +305,10 @@ int amdgpu_amdkfd_alloc_gtt_mem(struct amdgpu_device *adev, size_t size,
> bp.size = size;
> bp.byte_align = PAGE_SIZE;
> bp.domain = AMDGPU_GEM_DOMAIN_GTT;
> - bp.flags = AMDGPU_GEM_CREATE_CPU_GTT_USWC;
> + if (is_uswc_mode)
> + bp.flags = AMDGPU_GEM_CREATE_CPU_GTT_USWC;
> + else
> + bp.flags = 0;
> bp.type = ttm_bo_type_kernel;
> bp.resv = NULL;
> bp.bo_ptr_size = sizeof(struct amdgpu_bo);
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
> index 4b80ad860639..745121e0dd8e 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
> @@ -234,7 +234,7 @@ int amdgpu_amdkfd_bo_validate_and_fence(struct amdgpu_bo *bo,
> /* Shared API */
> int amdgpu_amdkfd_alloc_gtt_mem(struct amdgpu_device *adev, size_t size,
> void **mem_obj, uint64_t *gpu_addr,
> - void **cpu_ptr, bool mqd_gfx9);
> + void **cpu_ptr, bool mqd_gfx9, bool is_uswc_mode);
> void amdgpu_amdkfd_free_gtt_mem(struct amdgpu_device *adev, void **mem_obj);
> int amdgpu_amdkfd_alloc_gws(struct amdgpu_device *adev, size_t size,
> void **mem_obj);
> diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
> index 956198da7859..1f1d79ac5e6c 100644
> --- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c
> +++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
> @@ -779,7 +779,7 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
> if (amdgpu_amdkfd_alloc_gtt_mem(
> kfd->adev, size, &kfd->gtt_mem,
> &kfd->gtt_start_gpu_addr, &kfd->gtt_start_cpu_ptr,
> - false)) {
> + false, true)) {
> dev_err(kfd_device, "Could not allocate %d bytes\n", size);
> goto alloc_gtt_mem_failure;
> }
> diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
> index 38c19dc8311d..9672542578d4 100644
> --- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
> +++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
> @@ -2783,7 +2783,7 @@ static int allocate_hiq_sdma_mqd(struct device_queue_manager *dqm)
>
> retval = amdgpu_amdkfd_alloc_gtt_mem(dev->adev, size,
> &(mem_obj->gtt_mem), &(mem_obj->gpu_addr),
> - (void *)&(mem_obj->cpu_ptr), false);
> + (void *)&(mem_obj->cpu_ptr), false, true);
>
> return retval;
> }
> diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c
> index 84e8ea3a8a0c..c9882f1d4419 100644
> --- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c
> +++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c
> @@ -140,7 +140,7 @@ static struct kfd_mem_obj *allocate_mqd(struct kfd_node *node,
> NUM_XCC(node->xcc_mask),
> &(mqd_mem_obj->gtt_mem),
> &(mqd_mem_obj->gpu_addr),
> - (void *)&(mqd_mem_obj->cpu_ptr), true);
> + (void *)&(mqd_mem_obj->cpu_ptr), true, true);
>
> if (retval) {
> kfree(mqd_mem_obj);
> diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
> index 6bab6fc6a35d..e1d8fd11d2b1 100644
> --- a/drivers/gpu/drm/amd/amdkfd/kfd_process.c
> +++ b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
> @@ -1639,7 +1639,7 @@ struct kfd_process_device *kfd_create_process_device_data(struct kfd_node *dev,
> &pdd->proc_ctx_bo,
> &pdd->proc_ctx_gpu_addr,
> &pdd->proc_ctx_cpu_ptr,
> - false);
> + false, true);
> if (retval) {
> dev_err(dev->adev->dev,
> "failed to allocate process context bo\n");
> diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c
> index c76db22a1000..3a3be0e19fb9 100644
> --- a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c
> +++ b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c
> @@ -260,7 +260,7 @@ static int init_user_queue(struct process_queue_manager *pqm,
> &(*q)->gang_ctx_bo,
> &(*q)->gang_ctx_gpu_addr,
> &(*q)->gang_ctx_cpu_ptr,
> - false);
> + false, true);
> if (retval) {
> pr_err("failed to allocate gang context bo\n");
> goto cleanup;
> --
> 2.34.1
>
