aboutsummaryrefslogtreecommitdiff
path: root/drivers/gpu/drm/radeon/radeon_vm.c
diff options
context:
space:
mode:
Diffstat (limited to 'drivers/gpu/drm/radeon/radeon_vm.c')
-rw-r--r--drivers/gpu/drm/radeon/radeon_vm.c80
1 files changed, 50 insertions, 30 deletions
diff --git a/drivers/gpu/drm/radeon/radeon_vm.c b/drivers/gpu/drm/radeon/radeon_vm.c
index 2ab80a5331a4..0974fa17deb2 100644
--- a/drivers/gpu/drm/radeon/radeon_vm.c
+++ b/drivers/gpu/drm/radeon/radeon_vm.c
@@ -143,7 +143,7 @@ struct radeon_cs_reloc *radeon_vm_get_bos(struct radeon_device *rdev,
list[0].prefered_domains = RADEON_GEM_DOMAIN_VRAM;
list[0].allowed_domains = RADEON_GEM_DOMAIN_VRAM;
list[0].tv.bo = &vm->page_directory->tbo;
- list[0].tv.shared = false;
+ list[0].tv.shared = true;
list[0].tiling_flags = 0;
list[0].handle = 0;
list_add(&list[0].tv.head, head);
@@ -157,7 +157,7 @@ struct radeon_cs_reloc *radeon_vm_get_bos(struct radeon_device *rdev,
list[idx].prefered_domains = RADEON_GEM_DOMAIN_VRAM;
list[idx].allowed_domains = RADEON_GEM_DOMAIN_VRAM;
list[idx].tv.bo = &list[idx].robj->tbo;
- list[idx].tv.shared = false;
+ list[idx].tv.shared = true;
list[idx].tiling_flags = 0;
list[idx].handle = 0;
list_add(&list[idx++].tv.head, head);
@@ -387,35 +387,25 @@ static void radeon_vm_set_pages(struct radeon_device *rdev,
static int radeon_vm_clear_bo(struct radeon_device *rdev,
struct radeon_bo *bo)
{
- struct ttm_validate_buffer tv;
- struct ww_acquire_ctx ticket;
- struct list_head head;
struct radeon_ib ib;
unsigned entries;
uint64_t addr;
int r;
- memset(&tv, 0, sizeof(tv));
- tv.bo = &bo->tbo;
- tv.shared = false;
-
- INIT_LIST_HEAD(&head);
- list_add(&tv.head, &head);
-
- r = ttm_eu_reserve_buffers(&ticket, &head, true);
- if (r)
+ r = radeon_bo_reserve(bo, false);
+ if (r)
return r;
- r = ttm_bo_validate(&bo->tbo, &bo->placement, true, false);
- if (r)
- goto error;
+ r = ttm_bo_validate(&bo->tbo, &bo->placement, true, false);
+ if (r)
+ goto error_unreserve;
addr = radeon_bo_gpu_offset(bo);
entries = radeon_bo_size(bo) / 8;
r = radeon_ib_get(rdev, R600_RING_TYPE_DMA_INDEX, &ib, NULL, 256);
if (r)
- goto error;
+ goto error_unreserve;
ib.length_dw = 0;
@@ -425,15 +415,15 @@ static int radeon_vm_clear_bo(struct radeon_device *rdev,
r = radeon_ib_schedule(rdev, &ib, NULL, false);
if (r)
- goto error;
+ goto error_free;
- ttm_eu_fence_buffer_objects(&ticket, &head, &ib.fence->base);
- radeon_ib_free(rdev, &ib);
+ radeon_bo_fence(bo, ib.fence, false);
- return 0;
+error_free:
+ radeon_ib_free(rdev, &ib);
-error:
- ttm_eu_backoff_reservation(&ticket, &head);
+error_unreserve:
+ radeon_bo_unreserve(bo);
return r;
}
@@ -464,14 +454,14 @@ int radeon_vm_bo_set_addr(struct radeon_device *rdev,
if (soffset) {
/* make sure object fit at this offset */
- eoffset = soffset + size;
+ eoffset = soffset + size - 1;
if (soffset >= eoffset) {
return -EINVAL;
}
last_pfn = eoffset / RADEON_GPU_PAGE_SIZE;
- if (last_pfn > rdev->vm_manager.max_pfn) {
- dev_err(rdev->dev, "va above limit (0x%08X > 0x%08X)\n",
+ if (last_pfn >= rdev->vm_manager.max_pfn) {
+ dev_err(rdev->dev, "va above limit (0x%08X >= 0x%08X)\n",
last_pfn, rdev->vm_manager.max_pfn);
return -EINVAL;
}
@@ -485,7 +475,7 @@ int radeon_vm_bo_set_addr(struct radeon_device *rdev,
eoffset /= RADEON_GPU_PAGE_SIZE;
if (soffset || eoffset) {
struct interval_tree_node *it;
- it = interval_tree_iter_first(&vm->va, soffset, eoffset - 1);
+ it = interval_tree_iter_first(&vm->va, soffset, eoffset);
if (it && it != &bo_va->it) {
struct radeon_bo_va *tmp;
tmp = container_of(it, struct radeon_bo_va, it);
@@ -513,6 +503,8 @@ int radeon_vm_bo_set_addr(struct radeon_device *rdev,
tmp->addr = bo_va->addr;
tmp->bo = radeon_bo_ref(bo_va->bo);
list_add(&tmp->vm_status, &vm->freed);
+
+ bo_va->addr = 0;
}
interval_tree_remove(&bo_va->it, &vm->va);
@@ -522,7 +514,7 @@ int radeon_vm_bo_set_addr(struct radeon_device *rdev,
if (soffset || eoffset) {
bo_va->it.start = soffset;
- bo_va->it.last = eoffset - 1;
+ bo_va->it.last = eoffset;
interval_tree_insert(&bo_va->it, &vm->va);
}
@@ -710,6 +702,7 @@ int radeon_vm_update_page_directory(struct radeon_device *rdev,
radeon_ib_free(rdev, &ib);
return r;
}
+ radeon_bo_fence(pd, ib.fence, false);
radeon_fence_unref(&vm->fence);
vm->fence = radeon_fence_ref(ib.fence);
radeon_fence_unref(&vm->last_flush);
@@ -868,6 +861,31 @@ static void radeon_vm_update_ptes(struct radeon_device *rdev,
}
/**
+ * radeon_vm_fence_pts - fence page tables after an update
+ *
+ * @vm: requested vm
+ * @start: start of GPU address range
+ * @end: end of GPU address range
+ * @fence: fence to use
+ *
+ * Fence the page tables in the range @start - @end (cayman+).
+ *
+ * Global and local mutex must be locked!
+ */
+static void radeon_vm_fence_pts(struct radeon_vm *vm,
+ uint64_t start, uint64_t end,
+ struct radeon_fence *fence)
+{
+ unsigned i;
+
+ start >>= radeon_vm_block_size;
+ end = (end - 1) >> radeon_vm_block_size;
+
+ for (i = start; i <= end; ++i)
+ radeon_bo_fence(vm->page_tables[i].bo, fence, false);
+}
+
+/**
* radeon_vm_bo_update - map a bo into the vm page table
*
* @rdev: radeon_device pointer
@@ -979,6 +997,7 @@ int radeon_vm_bo_update(struct radeon_device *rdev,
radeon_ib_free(rdev, &ib);
return r;
}
+ radeon_vm_fence_pts(vm, bo_va->it.start, bo_va->it.last + 1, ib.fence);
radeon_fence_unref(&vm->fence);
vm->fence = radeon_fence_ref(ib.fence);
radeon_ib_free(rdev, &ib);
@@ -1058,7 +1077,8 @@ void radeon_vm_bo_rmv(struct radeon_device *rdev,
list_del(&bo_va->bo_list);
mutex_lock(&vm->mutex);
- interval_tree_remove(&bo_va->it, &vm->va);
+ if (bo_va->it.start || bo_va->it.last)
+ interval_tree_remove(&bo_va->it, &vm->va);
list_del(&bo_va->vm_status);
if (bo_va->addr) {
generated by cgit v1.2.3 (git 2.25.1) at 2024-06-13 00:24:32 +0000