summaryrefslogtreecommitdiff
path: root/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
diff options
context:
space:
mode:
Diffstat (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c')
-rw-r--r--drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c116
1 files changed, 42 insertions, 74 deletions
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
index cd89d2e46852..8b5452a8d330 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
@@ -253,53 +253,18 @@ void amdgpu_amdkfd_release_notify(struct amdgpu_bo *bo)
static int amdgpu_amdkfd_remove_eviction_fence(struct amdgpu_bo *bo,
struct amdgpu_amdkfd_fence *ef)
{
- struct dma_resv *resv = bo->tbo.base.resv;
- struct dma_resv_list *old, *new;
- unsigned int i, j, k;
+ struct dma_fence *replacement;
if (!ef)
return -EINVAL;
- old = dma_resv_shared_list(resv);
- if (!old)
- return 0;
-
- new = kmalloc(struct_size(new, shared, old->shared_max), GFP_KERNEL);
- if (!new)
- return -ENOMEM;
-
- /* Go through all the shared fences in the resevation object and sort
- * the interesting ones to the end of the list.
+ /* TODO: Instead of block before we should use the fence of the page
+ * table update and TLB flush here directly.
*/
- for (i = 0, j = old->shared_count, k = 0; i < old->shared_count; ++i) {
- struct dma_fence *f;
-
- f = rcu_dereference_protected(old->shared[i],
- dma_resv_held(resv));
-
- if (f->context == ef->base.context)
- RCU_INIT_POINTER(new->shared[--j], f);
- else
- RCU_INIT_POINTER(new->shared[k++], f);
- }
- new->shared_max = old->shared_max;
- new->shared_count = k;
-
- /* Install the new fence list, seqcount provides the barriers */
- write_seqcount_begin(&resv->seq);
- RCU_INIT_POINTER(resv->fence, new);
- write_seqcount_end(&resv->seq);
-
- /* Drop the references to the removed fences or move them to ef_list */
- for (i = j; i < old->shared_count; ++i) {
- struct dma_fence *f;
-
- f = rcu_dereference_protected(new->shared[i],
- dma_resv_held(resv));
- dma_fence_put(f);
- }
- kfree_rcu(old, rcu);
-
+ replacement = dma_fence_get_stub();
+ dma_resv_replace_fences(bo->tbo.base.resv, ef->base.context,
+ replacement, DMA_RESV_USAGE_READ);
+ dma_fence_put(replacement);
return 0;
}
@@ -1093,8 +1058,7 @@ static void unmap_bo_from_gpuvm(struct kgd_mem *mem,
static int update_gpuvm_pte(struct kgd_mem *mem,
struct kfd_mem_attachment *entry,
- struct amdgpu_sync *sync,
- bool *table_freed)
+ struct amdgpu_sync *sync)
{
struct amdgpu_bo_va *bo_va = entry->bo_va;
struct amdgpu_device *adev = entry->adev;
@@ -1105,7 +1069,7 @@ static int update_gpuvm_pte(struct kgd_mem *mem,
return ret;
/* Update the page tables */
- ret = amdgpu_vm_bo_update(adev, bo_va, false, table_freed);
+ ret = amdgpu_vm_bo_update(adev, bo_va, false);
if (ret) {
pr_err("amdgpu_vm_bo_update failed\n");
return ret;
@@ -1117,8 +1081,7 @@ static int update_gpuvm_pte(struct kgd_mem *mem,
static int map_bo_to_gpuvm(struct kgd_mem *mem,
struct kfd_mem_attachment *entry,
struct amdgpu_sync *sync,
- bool no_update_pte,
- bool *table_freed)
+ bool no_update_pte)
{
int ret;
@@ -1135,7 +1098,7 @@ static int map_bo_to_gpuvm(struct kgd_mem *mem,
if (no_update_pte)
return 0;
- ret = update_gpuvm_pte(mem, entry, sync, table_freed);
+ ret = update_gpuvm_pte(mem, entry, sync);
if (ret) {
pr_err("update_gpuvm_pte() failed\n");
goto update_gpuvm_pte_failed;
@@ -1268,7 +1231,7 @@ static int init_kfd_vm(struct amdgpu_vm *vm, void **process_info,
AMDGPU_FENCE_OWNER_KFD, false);
if (ret)
goto wait_pd_fail;
- ret = dma_resv_reserve_shared(vm->root.bo->tbo.base.resv, 1);
+ ret = dma_resv_reserve_fences(vm->root.bo->tbo.base.resv, 1);
if (ret)
goto reserve_shared_fail;
amdgpu_bo_fence(vm->root.bo,
@@ -1520,26 +1483,26 @@ int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu(
} else if (flags & KFD_IOC_ALLOC_MEM_FLAGS_GTT) {
domain = alloc_domain = AMDGPU_GEM_DOMAIN_GTT;
alloc_flags = 0;
- } else if (flags & KFD_IOC_ALLOC_MEM_FLAGS_USERPTR) {
+ } else {
domain = AMDGPU_GEM_DOMAIN_GTT;
alloc_domain = AMDGPU_GEM_DOMAIN_CPU;
alloc_flags = AMDGPU_GEM_CREATE_PREEMPTIBLE;
- if (!offset || !*offset)
- return -EINVAL;
- user_addr = untagged_addr(*offset);
- } else if (flags & (KFD_IOC_ALLOC_MEM_FLAGS_DOORBELL |
- KFD_IOC_ALLOC_MEM_FLAGS_MMIO_REMAP)) {
- domain = AMDGPU_GEM_DOMAIN_GTT;
- alloc_domain = AMDGPU_GEM_DOMAIN_CPU;
- bo_type = ttm_bo_type_sg;
- alloc_flags = 0;
- if (size > UINT_MAX)
+
+ if (flags & KFD_IOC_ALLOC_MEM_FLAGS_USERPTR) {
+ if (!offset || !*offset)
+ return -EINVAL;
+ user_addr = untagged_addr(*offset);
+ } else if (flags & (KFD_IOC_ALLOC_MEM_FLAGS_DOORBELL |
+ KFD_IOC_ALLOC_MEM_FLAGS_MMIO_REMAP)) {
+ bo_type = ttm_bo_type_sg;
+ if (size > UINT_MAX)
+ return -EINVAL;
+ sg = create_doorbell_sg(*offset, size);
+ if (!sg)
+ return -ENOMEM;
+ } else {
return -EINVAL;
- sg = create_doorbell_sg(*offset, size);
- if (!sg)
- return -ENOMEM;
- } else {
- return -EINVAL;
+ }
}
*mem = kzalloc(sizeof(struct kgd_mem), GFP_KERNEL);
@@ -1745,7 +1708,7 @@ int amdgpu_amdkfd_gpuvm_free_memory_of_gpu(
int amdgpu_amdkfd_gpuvm_map_memory_to_gpu(
struct amdgpu_device *adev, struct kgd_mem *mem,
- void *drm_priv, bool *table_freed)
+ void *drm_priv)
{
struct amdgpu_vm *avm = drm_priv_to_vm(drm_priv);
int ret;
@@ -1832,7 +1795,7 @@ int amdgpu_amdkfd_gpuvm_map_memory_to_gpu(
entry->va, entry->va + bo_size, entry);
ret = map_bo_to_gpuvm(mem, entry, ctx.sync,
- is_invalid_userptr, table_freed);
+ is_invalid_userptr);
if (ret) {
pr_err("Failed to map bo to gpuvm\n");
goto out_unreserve;
@@ -2300,7 +2263,7 @@ static int validate_invalid_user_pages(struct amdkfd_process_info *process_info)
continue;
kfd_mem_dmaunmap_attachment(mem, attachment);
- ret = update_gpuvm_pte(mem, attachment, &sync, NULL);
+ ret = update_gpuvm_pte(mem, attachment, &sync);
if (ret) {
pr_err("%s: update PTE failed\n", __func__);
/* make sure this gets validated again */
@@ -2482,6 +2445,8 @@ int amdgpu_amdkfd_gpuvm_restore_process_bos(void *info, struct dma_fence **ef)
struct amdgpu_bo *bo = mem->bo;
uint32_t domain = mem->domain;
struct kfd_mem_attachment *attachment;
+ struct dma_resv_iter cursor;
+ struct dma_fence *fence;
total_size += amdgpu_bo_size(bo);
@@ -2496,17 +2461,20 @@ int amdgpu_amdkfd_gpuvm_restore_process_bos(void *info, struct dma_fence **ef)
goto validate_map_fail;
}
}
- ret = amdgpu_sync_fence(&sync_obj, bo->tbo.moving);
- if (ret) {
- pr_debug("Memory eviction: Sync BO fence failed. Try again\n");
- goto validate_map_fail;
+ dma_resv_for_each_fence(&cursor, bo->tbo.base.resv,
+ DMA_RESV_USAGE_KERNEL, fence) {
+ ret = amdgpu_sync_fence(&sync_obj, fence);
+ if (ret) {
+ pr_debug("Memory eviction: Sync BO fence failed. Try again\n");
+ goto validate_map_fail;
+ }
}
list_for_each_entry(attachment, &mem->attachments, list) {
if (!attachment->is_mapped)
continue;
kfd_mem_dmaunmap_attachment(mem, attachment);
- ret = update_gpuvm_pte(mem, attachment, &sync_obj, NULL);
+ ret = update_gpuvm_pte(mem, attachment, &sync_obj);
if (ret) {
pr_debug("Memory eviction: update PTE failed. Try again\n");
goto validate_map_fail;
@@ -2606,7 +2574,7 @@ int amdgpu_amdkfd_add_gws_to_process(void *info, void *gws, struct kgd_mem **mem
* Add process eviction fence to bo so they can
* evict each other.
*/
- ret = dma_resv_reserve_shared(gws_bo->tbo.base.resv, 1);
+ ret = dma_resv_reserve_fences(gws_bo->tbo.base.resv, 1);
if (ret)
goto reserve_shared_fail;
amdgpu_bo_fence(gws_bo, &process_info->eviction_fence->base, true);
generated by cgit v1.2.3-70-g09d2 (git 2.48.1) at 2025-09-23 16:12:00 +0000