diff options
Diffstat (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c')
-rw-r--r-- | drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c | 83 |
1 files changed, 65 insertions, 18 deletions
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c index f82a2dd83874..eacd810fc09b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c @@ -32,6 +32,9 @@ #include "amdgpu.h" #include "amdgpu_trace.h" +#define AMDGPU_BO_LIST_MAX_PRIORITY 32u +#define AMDGPU_BO_LIST_NUM_BUCKETS (AMDGPU_BO_LIST_MAX_PRIORITY + 1) + static int amdgpu_bo_list_create(struct amdgpu_fpriv *fpriv, struct amdgpu_bo_list **result, int *id) @@ -88,8 +91,9 @@ static int amdgpu_bo_list_set(struct amdgpu_device *adev, struct amdgpu_bo *gws_obj = adev->gds.gws_gfx_bo; struct amdgpu_bo *oa_obj = adev->gds.oa_gfx_bo; - bool has_userptr = false; + unsigned last_entry = 0, first_userptr = num_entries; unsigned i; + int r; array = drm_malloc_ab(num_entries, sizeof(struct amdgpu_bo_list_entry)); if (!array) @@ -97,33 +101,43 @@ static int amdgpu_bo_list_set(struct amdgpu_device *adev, memset(array, 0, num_entries * sizeof(struct amdgpu_bo_list_entry)); for (i = 0; i < num_entries; ++i) { - struct amdgpu_bo_list_entry *entry = &array[i]; + struct amdgpu_bo_list_entry *entry; struct drm_gem_object *gobj; + struct amdgpu_bo *bo; + struct mm_struct *usermm; gobj = drm_gem_object_lookup(adev->ddev, filp, info[i].bo_handle); - if (!gobj) + if (!gobj) { + r = -ENOENT; goto error_free; + } - entry->robj = amdgpu_bo_ref(gem_to_amdgpu_bo(gobj)); + bo = amdgpu_bo_ref(gem_to_amdgpu_bo(gobj)); drm_gem_object_unreference_unlocked(gobj); - entry->priority = info[i].bo_priority; - entry->prefered_domains = entry->robj->initial_domain; - entry->allowed_domains = entry->prefered_domains; - if (entry->allowed_domains == AMDGPU_GEM_DOMAIN_VRAM) - entry->allowed_domains |= AMDGPU_GEM_DOMAIN_GTT; - if (amdgpu_ttm_tt_has_userptr(entry->robj->tbo.ttm)) { - has_userptr = true; - entry->prefered_domains = AMDGPU_GEM_DOMAIN_GTT; - entry->allowed_domains = AMDGPU_GEM_DOMAIN_GTT; + + usermm = amdgpu_ttm_tt_get_usermm(bo->tbo.ttm); + if (usermm) { + if (usermm != current->mm) { + amdgpu_bo_unref(&bo); + r = -EPERM; + goto error_free; + } + entry = &array[--first_userptr]; + } else { + entry = &array[last_entry++]; } + + entry->robj = bo; + entry->priority = min(info[i].bo_priority, + AMDGPU_BO_LIST_MAX_PRIORITY); entry->tv.bo = &entry->robj->tbo; entry->tv.shared = true; - if (entry->prefered_domains == AMDGPU_GEM_DOMAIN_GDS) + if (entry->robj->prefered_domains == AMDGPU_GEM_DOMAIN_GDS) gds_obj = entry->robj; - if (entry->prefered_domains == AMDGPU_GEM_DOMAIN_GWS) + if (entry->robj->prefered_domains == AMDGPU_GEM_DOMAIN_GWS) gws_obj = entry->robj; - if (entry->prefered_domains == AMDGPU_GEM_DOMAIN_OA) + if (entry->robj->prefered_domains == AMDGPU_GEM_DOMAIN_OA) oa_obj = entry->robj; trace_amdgpu_bo_list_set(list, entry->robj); @@ -137,15 +151,17 @@ static int amdgpu_bo_list_set(struct amdgpu_device *adev, list->gds_obj = gds_obj; list->gws_obj = gws_obj; list->oa_obj = oa_obj; - list->has_userptr = has_userptr; + list->first_userptr = first_userptr; list->array = array; list->num_entries = num_entries; return 0; error_free: + while (i--) + amdgpu_bo_unref(&array[i].robj); drm_free_large(array); - return -ENOENT; + return r; } struct amdgpu_bo_list * @@ -161,6 +177,37 @@ amdgpu_bo_list_get(struct amdgpu_fpriv *fpriv, int id) return result; } +void amdgpu_bo_list_get_list(struct amdgpu_bo_list *list, + struct list_head *validated) +{ + /* This is based on the bucket sort with O(n) time complexity. + * An item with priority "i" is added to bucket[i]. The lists are then + * concatenated in descending order. + */ + struct list_head bucket[AMDGPU_BO_LIST_NUM_BUCKETS]; + unsigned i; + + for (i = 0; i < AMDGPU_BO_LIST_NUM_BUCKETS; i++) + INIT_LIST_HEAD(&bucket[i]); + + /* Since buffers which appear sooner in the relocation list are + * likely to be used more often than buffers which appear later + * in the list, the sort mustn't change the ordering of buffers + * with the same priority, i.e. it must be stable. + */ + for (i = 0; i < list->num_entries; i++) { + unsigned priority = list->array[i].priority; + + list_add_tail(&list->array[i].tv.head, + &bucket[priority]); + list->array[i].user_pages = NULL; + } + + /* Connect the sorted buckets in the output list. */ + for (i = 0; i < AMDGPU_BO_LIST_NUM_BUCKETS; i++) + list_splice(&bucket[i], validated); +} + void amdgpu_bo_list_put(struct amdgpu_bo_list *list) { mutex_unlock(&list->lock); |