* [PATCH v2 1/2] gpu/buddy: replace dual-tree/force_merge with decoupled clear tracker
@ 2026-05-04 11:10 Arunpravin Paneer Selvam
2026-05-04 11:10 ` [PATCH v2 2/2] gpu/tests/buddy: add clear-tracker allocation latency benchmarks Arunpravin Paneer Selvam
` (5 more replies)
0 siblings, 6 replies; 8+ messages in thread
From: Arunpravin Paneer Selvam @ 2026-05-04 11:10 UTC (permalink / raw)
To: matthew.auld, christian.koenig, dri-devel, intel-gfx, intel-xe,
amd-gfx
Cc: alexander.deucher, Arunpravin Paneer Selvam
The current buddy allocator maintains separate clear_tree[] and
dirty_tree[] rbtrees per order, preventing coalescing between cleared
and dirty buddies. Under mixed workloads, this creates a merge barrier:
adjacent buddies frequently end up split across trees, forcing reliance
on __force_merge() during allocation.
__force_merge() performs an O(N x max_order) scan under the VRAM manager
lock, leading to allocation stalls and failures for large contiguous
requests even when sufficient total free memory is available.
Solution
Replace the dual-tree design with:
- A single free_tree[order] rbtree for dirty and mixed free blocks
(fully cleared free blocks float outside this tree)
- A lightweight out-of-band clear tracker (gpu_clear_tracker)
Fully cleared free blocks are tracked outside the buddy trees using an
augmented interval rbtree, enabling O(log E) lookup of the largest
cleared extents.
Buddy coalescing is now unconditional in __gpu_buddy_free(), regardless
of clear/dirty state. This removes the merge barrier and eliminates the
need for __force_merge().
Benefits
- Correct high-order allocations after mixed clear/dirty workloads
- Elimination of O(N x max_order) merge cost from the allocation path
- O(log E) cleared-extent lookup replacing O(N) scans
- Predictable allocation latency under fragmentation
- Reduced complexity with a single tree per order
Test:
dEQP-VK.memory.allocation.basic.size_8KiB.reverse.count_4000
Below data is from /sys/kernel/debug/dri/1/amdgpu_vram_mm:
Base (dual-tree), before VKCTS test:
order- 6 free: 6 MiB, blocks: 26
order- 5 free: 1 MiB, blocks: 15
order- 4 free: 960 KiB, blocks: 15
order- 3 free: 5 MiB, blocks: 171
order- 2 free: 2 MiB, blocks: 176
order- 1 free: 1 MiB, blocks: 165
order- 0 free: 16 KiB, blocks: 4
Base (dual-tree), after VKCTS test:
order- 6 free: 768 KiB, blocks: 3
order- 5 free: 499 MiB, blocks: 3999
order- 4 free: 250 MiB, blocks: 4001
order- 3 free: 129 MiB, blocks: 4157
order- 2 free: 65 MiB, blocks: 4161
order- 1 free: 63 MiB, blocks: 8138
order- 0 free: 20 KiB, blocks: 5
Clear tracker, before VKCTS test:
order- 6 free: 4 MiB, blocks: 19
order- 5 free: 2 MiB, blocks: 18
order- 4 free: 704 KiB, blocks: 11
order- 3 free: 5 MiB, blocks: 168
order- 2 free: 2 MiB, blocks: 174
order- 1 free: 1 MiB, blocks: 167
order- 0 free: 32 KiB, blocks: 8
Clear tracker, after VKCTS test:
order- 6 free: 4 MiB, blocks: 19
order- 5 free: 2 MiB, blocks: 18
order- 4 free: 704 KiB, blocks: 11
order- 3 free: 5 MiB, blocks: 168
order- 2 free: 2 MiB, blocks: 174
order- 1 free: 1 MiB, blocks: 167
order- 0 free: 28 KiB, blocks: 7
v2:
- Code-style cleanup and minor refactoring
- Renamed locals for clarity
Cc: Matthew Auld <matthew.auld@intel.com>
Cc: Christian König <christian.koenig@amd.com>
Signed-off-by: Arunpravin Paneer Selvam <Arunpravin.PaneerSelvam@amd.com>
---
drivers/gpu/buddy.c | 997 ++++++++++++++++++++++--------------
drivers/gpu/drm/drm_buddy.c | 44 +-
include/linux/gpu_buddy.h | 61 ++-
3 files changed, 692 insertions(+), 410 deletions(-)
diff --git a/drivers/gpu/buddy.c b/drivers/gpu/buddy.c
index 52686672e99f..8c6bd8b95cdb 100644
--- a/drivers/gpu/buddy.c
+++ b/drivers/gpu/buddy.c
@@ -34,6 +34,334 @@
#endif
static struct kmem_cache *slab_blocks;
+static struct kmem_cache *slab_extents;
+
+/*
+ * Clear tracker
+ * -------------
+ *
+ * The clear tracker maintains an augmented interval rbtree of contiguous
+ * cleared (zeroed) address ranges, decoupled from the buddy free trees.
+ * Each node covers a maximal coalesced run; adjacent extents are merged
+ * on insertion so the tree always holds the smallest possible number of
+ * extents. The augmentation field @subtree_max_size lets the allocator
+ * locate the largest cleared extent in O(log E).
+ */
+
+static u64 extent_size(struct gpu_clear_extent *clear_extent)
+{
+ return clear_extent->end - clear_extent->start;
+}
+
+RB_DECLARE_CALLBACKS_MAX(static, gpu_clear_augment_cb,
+ struct gpu_clear_extent, rb,
+ u64, subtree_max_size,
+ extent_size)
+
+static struct gpu_clear_extent *extent_alloc(void)
+{
+ return kmem_cache_zalloc(slab_extents, GFP_KERNEL);
+}
+
+static void extent_free(struct gpu_clear_extent *clear_extent)
+{
+ kmem_cache_free(slab_extents, clear_extent);
+}
+
+/* Return the rightmost extent whose start is strictly below @offset. */
+static struct gpu_clear_extent *
+prev_extent(struct gpu_clear_tracker *clear_tracker, u64 offset)
+{
+ struct rb_node *rb = clear_tracker->root.rb_node;
+ struct gpu_clear_extent *clear_extent = NULL;
+
+ while (rb) {
+ struct gpu_clear_extent *tmp_extent =
+ rb_entry(rb, struct gpu_clear_extent, rb);
+
+ if (tmp_extent->start < offset) {
+ clear_extent = tmp_extent;
+ rb = rb->rb_right;
+ } else {
+ rb = rb->rb_left;
+ }
+ }
+
+ return clear_extent;
+}
+
+/* Return the leftmost extent whose start is at or above @offset. */
+static struct gpu_clear_extent *
+next_extent(struct gpu_clear_tracker *clear_tracker, u64 offset)
+{
+ struct rb_node *rb = clear_tracker->root.rb_node;
+ struct gpu_clear_extent *clear_extent = NULL;
+
+ while (rb) {
+ struct gpu_clear_extent *tmp_extent =
+ rb_entry(rb, struct gpu_clear_extent, rb);
+
+ if (tmp_extent->start >= offset) {
+ clear_extent = tmp_extent;
+ rb = rb->rb_left;
+ } else {
+ rb = rb->rb_right;
+ }
+ }
+
+ return clear_extent;
+}
+
+static void insert_extent(struct gpu_clear_tracker *clear_tracker,
+ struct gpu_clear_extent *clear_extent)
+{
+ struct rb_node **link = &clear_tracker->root.rb_node;
+ struct rb_node *parent = NULL;
+
+ while (*link) {
+ struct gpu_clear_extent *tmp_extent;
+
+ parent = *link;
+ tmp_extent = rb_entry(parent, struct gpu_clear_extent, rb);
+
+ if (clear_extent->start < tmp_extent->start)
+ link = &parent->rb_left;
+ else
+ link = &parent->rb_right;
+ }
+
+ clear_extent->subtree_max_size = extent_size(clear_extent);
+ rb_link_node(&clear_extent->rb, parent, link);
+ rb_insert_augmented(&clear_extent->rb, &clear_tracker->root, &gpu_clear_augment_cb);
+}
+
+static void remove_extent(struct gpu_clear_tracker *clear_tracker,
+ struct gpu_clear_extent *clear_extent)
+{
+ rb_erase_augmented(&clear_extent->rb, &clear_tracker->root, &gpu_clear_augment_cb);
+ RB_CLEAR_NODE(&clear_extent->rb);
+}
+
+static void gpu_clear_tracker_init(struct gpu_clear_tracker *clear_tracker)
+{
+ clear_tracker->root = RB_ROOT;
+ clear_tracker->total_clear = 0;
+}
+
+static void gpu_clear_tracker_fini(struct gpu_clear_tracker *clear_tracker)
+{
+ struct rb_node *rb;
+
+ while ((rb = rb_first(&clear_tracker->root))) {
+ struct gpu_clear_extent *clear_extent =
+ rb_entry(rb, struct gpu_clear_extent, rb);
+
+ remove_extent(clear_tracker, clear_extent);
+ extent_free(clear_extent);
+ }
+
+ clear_tracker->total_clear = 0;
+}
+
+/*
+ * Mark the range [start, start + size] as cleared. Merge with the neighbour on
+ * each side if they are contiguous, so the tree never holds two adjacent ranges.
+ */
+static void gpu_clear_tracker_mark_clear(struct gpu_clear_tracker *clear_tracker,
+ u64 start, u64 size)
+{
+ struct gpu_clear_extent *left, *right, *clear_extent;
+ u64 end = start + size;
+
+ if (!size)
+ return;
+
+ /* Find contiguous neighbours, if any. */
+ left = prev_extent(clear_tracker, start);
+ if (left && left->end != start)
+ left = NULL;
+
+ right = next_extent(clear_tracker, end);
+ if (right && right->start != end)
+ right = NULL;
+
+ if (left && right) {
+ /* Merge left + new + right into a single extent. */
+ remove_extent(clear_tracker, left);
+ remove_extent(clear_tracker, right);
+ left->end = right->end;
+ extent_free(right);
+ insert_extent(clear_tracker, left);
+ } else if (left) {
+ /* Extend left neighbour rightwards. */
+ remove_extent(clear_tracker, left);
+ left->end = end;
+ insert_extent(clear_tracker, left);
+ } else if (right) {
+ /* Extend right neighbour leftwards. */
+ remove_extent(clear_tracker, right);
+ right->start = start;
+ insert_extent(clear_tracker, right);
+ } else {
+ /* Standalone extent. */
+ clear_extent = extent_alloc();
+ if (!clear_extent)
+ return;
+
+ clear_extent->start = start;
+ clear_extent->end = end;
+ insert_extent(clear_tracker, clear_extent);
+ }
+
+ clear_tracker->total_clear += size;
+}
+
+/*
+ * Mark the range [start, start + size] as dirty. Remove the range from every
+ * overlapping clear extent, splitting one extent in two if the dirty range
+ * falls strictly inside it.
+ */
+static void gpu_clear_tracker_mark_dirty(struct gpu_clear_tracker *clear_tracker,
+ u64 start, u64 size)
+{
+ struct gpu_clear_extent *clear_extent, *next;
+ u64 end = start + size;
+
+ if (!size)
+ return;
+
+ clear_extent = prev_extent(clear_tracker, start + 1);
+ if (!clear_extent)
+ clear_extent = next_extent(clear_tracker, start);
+
+ while (clear_extent && clear_extent->start < end) {
+ struct rb_node *next_node = rb_next(&clear_extent->rb);
+ u64 extent_start = clear_extent->start;
+ u64 extent_end = clear_extent->end;
+
+ if (next_node)
+ next = rb_entry(next_node, struct gpu_clear_extent, rb);
+ else
+ next = NULL;
+
+ /* Skip a non-overlapping neighbour returned by prev_extent(). */
+ if (extent_end <= start) {
+ clear_extent = next;
+ continue;
+ }
+
+ if (extent_start < start && extent_end > end) {
+ /* Dirty range falls strictly inside: split into left + right. */
+ struct gpu_clear_extent *right = extent_alloc();
+ if (!right)
+ return;
+
+ remove_extent(clear_tracker, clear_extent);
+
+ clear_extent->end = start;
+ right->start = end;
+ right->end = extent_end;
+
+ insert_extent(clear_tracker, clear_extent);
+ insert_extent(clear_tracker, right);
+
+ clear_tracker->total_clear -= size;
+ } else if (extent_start >= start && extent_end <= end) {
+ /* Extent fully covered: drop it. */
+ remove_extent(clear_tracker, clear_extent);
+ extent_free(clear_extent);
+
+ clear_tracker->total_clear -= (extent_end - extent_start);
+ } else if (extent_start < start) {
+ /* Extent overlaps from the left: trim its right end. */
+ remove_extent(clear_tracker, clear_extent);
+ clear_extent->end = start;
+ insert_extent(clear_tracker, clear_extent);
+
+ clear_tracker->total_clear -= (extent_end - start);
+ } else {
+ /* Extent overlaps from the right: trim its left end. */
+ remove_extent(clear_tracker, clear_extent);
+ clear_extent->start = end;
+ insert_extent(clear_tracker, clear_extent);
+
+ clear_tracker->total_clear -= (end - extent_start);
+ }
+
+ clear_extent = next;
+ }
+}
+
+/*
+ * Returns true if the range [start, start + size] lies entirely within
+ * a single clear extent in the tracker, i.e. the whole range is known
+ * to be cleared.
+ */
+static bool gpu_clear_tracker_is_clear(struct gpu_clear_tracker *clear_tracker,
+ u64 start, u64 size)
+{
+ struct gpu_clear_extent *clear_extent;
+ u64 end = start + size;
+
+ clear_extent = prev_extent(clear_tracker, start + 1);
+ if (!clear_extent)
+ return false;
+
+ return clear_extent->start <= start && clear_extent->end >= end;
+}
+
+/*
+ * Search the tracker for a clear extent of at least @min_size bytes
+ * and return it. The walk prefers higher-address candidates: at each
+ * node it descends into the right subtree first, then checks the node
+ * itself, and only falls back to the left subtree if neither matches.
+ *
+ * The augmented @subtree_max_size lets us skip whole subtrees that
+ * cannot contain a large-enough extent, so the search runs in
+ * O(log E) where E is the number of clear extents.
+ *
+ * Returns NULL if no extent satisfies @min_size.
+ */
+static struct gpu_clear_extent *
+gpu_clear_tracker_find(struct gpu_clear_tracker *clear_tracker, u64 min_size)
+{
+ struct rb_node *rb = clear_tracker->root.rb_node;
+
+ while (rb) {
+ struct gpu_clear_extent *clear_extent =
+ rb_entry(rb, struct gpu_clear_extent, rb);
+ struct rb_node *right = rb->rb_right;
+ struct rb_node *left = rb->rb_left;
+
+ /* Prefer the right (higher-address) subtree. */
+ if (right) {
+ struct gpu_clear_extent *r =
+ rb_entry(right, struct gpu_clear_extent, rb);
+
+ if (r->subtree_max_size >= min_size) {
+ rb = right;
+ continue;
+ }
+ }
+
+ if (extent_size(clear_extent) >= min_size)
+ return clear_extent;
+
+ if (left) {
+ struct gpu_clear_extent *l =
+ rb_entry(left, struct gpu_clear_extent, rb);
+
+ if (l->subtree_max_size >= min_size) {
+ rb = left;
+ continue;
+ }
+ }
+
+ break;
+ }
+
+ return NULL;
+}
static unsigned int
gpu_buddy_block_state(struct gpu_buddy_block *block)
@@ -101,13 +429,6 @@ static void gpu_block_free(struct gpu_buddy *mm,
kmem_cache_free(slab_blocks, block);
}
-static enum gpu_buddy_free_tree
-get_block_tree(struct gpu_buddy_block *block)
-{
- return gpu_buddy_block_is_clear(block) ?
- GPU_BUDDY_CLEAR_TREE : GPU_BUDDY_DIRTY_TREE;
-}
-
static struct gpu_buddy_block *
rbtree_get_free_block(const struct rb_node *node)
{
@@ -120,14 +441,8 @@ rbtree_last_free_block(struct rb_root *root)
return rbtree_get_free_block(rb_last(root));
}
-static bool rbtree_is_empty(struct rb_root *root)
-{
- return RB_EMPTY_ROOT(root);
-}
-
static void rbtree_insert(struct gpu_buddy *mm,
- struct gpu_buddy_block *block,
- enum gpu_buddy_free_tree tree)
+ struct gpu_buddy_block *block)
{
struct rb_node **link, *parent = NULL;
unsigned int block_alignment, order;
@@ -137,7 +452,7 @@ static void rbtree_insert(struct gpu_buddy *mm,
order = gpu_buddy_block_order(block);
block_alignment = gpu_buddy_block_offset_alignment(block);
- root = &mm->free_trees[tree][order];
+ root = &mm->free_tree[order];
link = &root->rb_node;
while (*link) {
@@ -167,26 +482,14 @@ static void rbtree_remove(struct gpu_buddy *mm,
struct gpu_buddy_block *block)
{
unsigned int order = gpu_buddy_block_order(block);
- enum gpu_buddy_free_tree tree;
- struct rb_root *root;
- tree = get_block_tree(block);
- root = &mm->free_trees[tree][order];
+ if (gpu_buddy_block_is_clear(block))
+ return;
- rb_erase_augmented(&block->rb, root, &gpu_buddy_augment_cb);
+ rb_erase_augmented(&block->rb, &mm->free_tree[order], &gpu_buddy_augment_cb);
RB_CLEAR_NODE(&block->rb);
}
-static void clear_reset(struct gpu_buddy_block *block)
-{
- block->header &= ~GPU_BUDDY_HEADER_CLEAR;
-}
-
-static void mark_cleared(struct gpu_buddy_block *block)
-{
- block->header |= GPU_BUDDY_HEADER_CLEAR;
-}
-
static void mark_allocated(struct gpu_buddy *mm,
struct gpu_buddy_block *block)
{
@@ -199,13 +502,18 @@ static void mark_allocated(struct gpu_buddy *mm,
static void mark_free(struct gpu_buddy *mm,
struct gpu_buddy_block *block)
{
- enum gpu_buddy_free_tree tree;
-
block->header &= ~GPU_BUDDY_HEADER_STATE;
block->header |= GPU_BUDDY_FREE;
- tree = get_block_tree(block);
- rbtree_insert(mm, block, tree);
+ if (gpu_clear_tracker_is_clear(&mm->clear,
+ gpu_buddy_block_offset(block),
+ gpu_buddy_block_size(mm, block))) {
+ block->header |= GPU_BUDDY_HEADER_CLEAR;
+ RB_CLEAR_NODE(&block->rb);
+ } else {
+ block->header &= ~GPU_BUDDY_HEADER_CLEAR;
+ rbtree_insert(mm, block);
+ }
}
static void mark_split(struct gpu_buddy *mm,
@@ -243,36 +551,18 @@ __get_buddy(struct gpu_buddy_block *block)
}
static unsigned int __gpu_buddy_free(struct gpu_buddy *mm,
- struct gpu_buddy_block *block,
- bool force_merge)
+ struct gpu_buddy_block *block)
{
struct gpu_buddy_block *parent;
unsigned int order;
while ((parent = block->parent)) {
- struct gpu_buddy_block *buddy;
-
- buddy = __get_buddy(block);
+ struct gpu_buddy_block *buddy = __get_buddy(block);
if (!gpu_buddy_block_is_free(buddy))
break;
- if (!force_merge) {
- /*
- * Check the block and its buddy clear state and exit
- * the loop if they both have the dissimilar state.
- */
- if (gpu_buddy_block_is_clear(block) !=
- gpu_buddy_block_is_clear(buddy))
- break;
-
- if (gpu_buddy_block_is_clear(block))
- mark_cleared(parent);
- }
-
rbtree_remove(mm, buddy);
- if (force_merge && gpu_buddy_block_is_clear(buddy))
- mm->clear_avail -= gpu_buddy_block_size(mm, buddy);
gpu_block_free(mm, block);
gpu_block_free(mm, buddy);
@@ -286,66 +576,15 @@ static unsigned int __gpu_buddy_free(struct gpu_buddy *mm,
return order;
}
-static int __force_merge(struct gpu_buddy *mm,
- u64 start,
- u64 end,
- unsigned int min_order)
+static void undo_partial_split(struct gpu_buddy *mm,
+ struct gpu_buddy_block *block)
{
- unsigned int tree, order;
- int i;
-
- if (!min_order)
- return -ENOMEM;
-
- if (min_order > mm->max_order)
- return -EINVAL;
-
- for_each_free_tree(tree) {
- for (i = min_order - 1; i >= 0; i--) {
- struct rb_node *iter = rb_last(&mm->free_trees[tree][i]);
-
- while (iter) {
- struct gpu_buddy_block *block, *buddy;
- u64 block_start, block_end;
-
- block = rbtree_get_free_block(iter);
- iter = rb_prev(iter);
-
- if (!block || !block->parent)
- continue;
-
- block_start = gpu_buddy_block_offset(block);
- block_end = block_start + gpu_buddy_block_size(mm, block) - 1;
-
- if (!contains(start, end, block_start, block_end))
- continue;
+ struct gpu_buddy_block *buddy = __get_buddy(block);
- buddy = __get_buddy(block);
- if (!gpu_buddy_block_is_free(buddy))
- continue;
-
- gpu_buddy_assert(gpu_buddy_block_is_clear(block) !=
- gpu_buddy_block_is_clear(buddy));
-
- /*
- * Advance to the next node when the current node is the buddy,
- * as freeing the block will also remove its buddy from the tree.
- */
- if (iter == &buddy->rb)
- iter = rb_prev(iter);
-
- rbtree_remove(mm, block);
- if (gpu_buddy_block_is_clear(block))
- mm->clear_avail -= gpu_buddy_block_size(mm, block);
-
- order = __gpu_buddy_free(mm, block, true);
- if (order >= min_order)
- return 0;
- }
- }
- }
-
- return -ENOMEM;
+ if (buddy &&
+ gpu_buddy_block_is_free(block) &&
+ gpu_buddy_block_is_free(buddy))
+ __gpu_buddy_free(mm, block);
}
/**
@@ -362,7 +601,7 @@ static int __force_merge(struct gpu_buddy *mm,
*/
int gpu_buddy_init(struct gpu_buddy *mm, u64 size, u64 chunk_size)
{
- unsigned int i, j, root_count = 0;
+ unsigned int root_count = 0;
u64 offset = 0;
if (size < chunk_size)
@@ -384,22 +623,13 @@ int gpu_buddy_init(struct gpu_buddy *mm, u64 size, u64 chunk_size)
BUG_ON(mm->max_order > GPU_BUDDY_MAX_ORDER);
- mm->free_trees = kmalloc_array(GPU_BUDDY_MAX_FREE_TREES,
- sizeof(*mm->free_trees),
- GFP_KERNEL);
- if (!mm->free_trees)
+ mm->free_tree = kcalloc(mm->max_order + 1,
+ sizeof(struct rb_root),
+ GFP_KERNEL);
+ if (!mm->free_tree)
return -ENOMEM;
- for_each_free_tree(i) {
- mm->free_trees[i] = kmalloc_array(mm->max_order + 1,
- sizeof(struct rb_root),
- GFP_KERNEL);
- if (!mm->free_trees[i])
- goto out_free_tree;
-
- for (j = 0; j <= mm->max_order; ++j)
- mm->free_trees[i][j] = RB_ROOT;
- }
+ gpu_clear_tracker_init(&mm->clear);
mm->n_roots = hweight64(size);
@@ -444,9 +674,8 @@ int gpu_buddy_init(struct gpu_buddy *mm, u64 size, u64 chunk_size)
gpu_block_free(mm, mm->roots[root_count]);
kfree(mm->roots);
out_free_tree:
- while (i--)
- kfree(mm->free_trees[i]);
- kfree(mm->free_trees);
+ gpu_clear_tracker_fini(&mm->clear);
+ kfree(mm->free_tree);
return -ENOMEM;
}
EXPORT_SYMBOL(gpu_buddy_init);
@@ -460,7 +689,7 @@ EXPORT_SYMBOL(gpu_buddy_init);
*/
void gpu_buddy_fini(struct gpu_buddy *mm)
{
- u64 root_size, size, start;
+ u64 root_size, size;
unsigned int order;
int i;
@@ -468,22 +697,17 @@ void gpu_buddy_fini(struct gpu_buddy *mm)
for (i = 0; i < mm->n_roots; ++i) {
order = ilog2(size) - ilog2(mm->chunk_size);
- start = gpu_buddy_block_offset(mm->roots[i]);
- __force_merge(mm, start, start + size, order);
+ root_size = mm->chunk_size << order;
gpu_buddy_assert(gpu_buddy_block_is_free(mm->roots[i]));
-
gpu_block_free(mm, mm->roots[i]);
-
- root_size = mm->chunk_size << order;
size -= root_size;
}
gpu_buddy_assert(mm->avail == mm->size);
- for_each_free_tree(i)
- kfree(mm->free_trees[i]);
- kfree(mm->free_trees);
+ gpu_clear_tracker_fini(&mm->clear);
+ kfree(mm->free_tree);
kfree(mm->roots);
}
EXPORT_SYMBOL(gpu_buddy_fini);
@@ -509,13 +733,6 @@ static int split_block(struct gpu_buddy *mm,
}
mark_split(mm, block);
-
- if (gpu_buddy_block_is_clear(block)) {
- mark_cleared(block->left);
- mark_cleared(block->right);
- clear_reset(block);
- }
-
mark_free(mm, block->left);
mark_free(mm, block->right);
@@ -533,41 +750,55 @@ static int split_block(struct gpu_buddy *mm,
*/
void gpu_buddy_reset_clear(struct gpu_buddy *mm, bool is_clear)
{
- enum gpu_buddy_free_tree src_tree, dst_tree;
- u64 root_size, size, start;
- unsigned int order;
- int i;
+ unsigned int i;
- size = mm->size;
- for (i = 0; i < mm->n_roots; ++i) {
- order = ilog2(size) - ilog2(mm->chunk_size);
- start = gpu_buddy_block_offset(mm->roots[i]);
- __force_merge(mm, start, start + size, order);
+ gpu_clear_tracker_fini(&mm->clear);
+ gpu_clear_tracker_init(&mm->clear);
+
+ if (is_clear) {
+ for (i = 0; i <= mm->max_order; ++i) {
+ struct rb_node *node;
+
+ node = rb_first(&mm->free_tree[i]);
+ while (node) {
+ struct gpu_buddy_block *block =
+ rb_entry(node, struct gpu_buddy_block, rb);
+
+ node = rb_next(node);
+ rb_erase_augmented(&block->rb, &mm->free_tree[i], &gpu_buddy_augment_cb);
+ RB_CLEAR_NODE(&block->rb);
+ block->header |= GPU_BUDDY_HEADER_CLEAR;
+ gpu_clear_tracker_mark_clear(&mm->clear,
+ gpu_buddy_block_offset(block),
+ gpu_buddy_block_size(mm, block));
+ }
+ }
+ } else {
+ LIST_HEAD(dfs);
- root_size = mm->chunk_size << order;
- size -= root_size;
- }
+ for (i = 0; i < mm->n_roots; ++i)
+ list_add(&mm->roots[i]->tmp_link, &dfs);
- src_tree = is_clear ? GPU_BUDDY_DIRTY_TREE : GPU_BUDDY_CLEAR_TREE;
- dst_tree = is_clear ? GPU_BUDDY_CLEAR_TREE : GPU_BUDDY_DIRTY_TREE;
+ while (!list_empty(&dfs)) {
+ struct gpu_buddy_block *block =
+ list_first_entry(&dfs, struct gpu_buddy_block, tmp_link);
- for (i = 0; i <= mm->max_order; ++i) {
- struct rb_root *root = &mm->free_trees[src_tree][i];
- struct gpu_buddy_block *block, *tmp;
+ list_del(&block->tmp_link);
- rbtree_postorder_for_each_entry_safe(block, tmp, root, rb) {
- rbtree_remove(mm, block);
- if (is_clear) {
- mark_cleared(block);
- mm->clear_avail += gpu_buddy_block_size(mm, block);
- } else {
- clear_reset(block);
- mm->clear_avail -= gpu_buddy_block_size(mm, block);
+ if (gpu_buddy_block_is_split(block)) {
+ list_add(&block->right->tmp_link, &dfs);
+ list_add(&block->left->tmp_link, &dfs);
+ continue;
}
- rbtree_insert(mm, block, dst_tree);
+ if (gpu_buddy_block_is_free(block) && gpu_buddy_block_is_clear(block)) {
+ block->header &= ~GPU_BUDDY_HEADER_CLEAR;
+ rbtree_insert(mm, block);
+ }
}
}
+
+ mm->clear_avail = mm->clear.total_clear;
}
EXPORT_SYMBOL(gpu_buddy_reset_clear);
@@ -580,12 +811,21 @@ EXPORT_SYMBOL(gpu_buddy_reset_clear);
void gpu_buddy_free_block(struct gpu_buddy *mm,
struct gpu_buddy_block *block)
{
+ bool was_clear = gpu_buddy_block_is_clear(block);
+ u64 size = gpu_buddy_block_size(mm, block);
+ u64 offset = gpu_buddy_block_offset(block);
+
BUG_ON(!gpu_buddy_block_is_allocated(block));
- mm->avail += gpu_buddy_block_size(mm, block);
- if (gpu_buddy_block_is_clear(block))
- mm->clear_avail += gpu_buddy_block_size(mm, block);
- __gpu_buddy_free(mm, block, false);
+ block->header &= ~GPU_BUDDY_HEADER_CLEAR;
+ mm->avail += size;
+
+ if (was_clear) {
+ gpu_clear_tracker_mark_clear(&mm->clear, offset, size);
+ mm->clear_avail = mm->clear.total_clear;
+ }
+
+ __gpu_buddy_free(mm, block);
}
EXPORT_SYMBOL(gpu_buddy_free_block);
@@ -599,10 +839,15 @@ static void __gpu_buddy_free_list(struct gpu_buddy *mm,
gpu_buddy_assert(!(mark_dirty && mark_clear));
list_for_each_entry_safe(block, on, objects, link) {
+ /*
+ * Propagate the caller's clear/dirty intent onto the block header
+ * before handing it to gpu_buddy_free_block(), which will then
+ * update the clear tracker accordingly.
+ */
if (mark_clear)
- mark_cleared(block);
+ block->header |= GPU_BUDDY_HEADER_CLEAR;
else if (mark_dirty)
- clear_reset(block);
+ block->header &= ~GPU_BUDDY_HEADER_CLEAR;
gpu_buddy_free_block(mm, block);
cond_resched();
}
@@ -637,23 +882,14 @@ void gpu_buddy_free_list(struct gpu_buddy *mm,
}
EXPORT_SYMBOL(gpu_buddy_free_list);
-static bool block_incompatible(struct gpu_buddy_block *block, unsigned int flags)
-{
- bool needs_clear = flags & GPU_BUDDY_CLEAR_ALLOCATION;
-
- return needs_clear != gpu_buddy_block_is_clear(block);
-}
-
static struct gpu_buddy_block *
__alloc_range_bias(struct gpu_buddy *mm,
u64 start, u64 end,
unsigned int order,
- unsigned long flags,
- bool fallback)
+ unsigned long flags)
{
u64 req_size = mm->chunk_size << order;
struct gpu_buddy_block *block;
- struct gpu_buddy_block *buddy;
LIST_HEAD(dfs);
int err;
int i;
@@ -696,9 +932,6 @@ __alloc_range_bias(struct gpu_buddy *mm,
continue;
}
- if (!fallback && block_incompatible(block, flags))
- continue;
-
if (contains(start, end, block_start, block_end) &&
order == gpu_buddy_block_order(block)) {
/*
@@ -716,68 +949,32 @@ __alloc_range_bias(struct gpu_buddy *mm,
goto err_undo;
}
- list_add(&block->right->tmp_link, &dfs);
list_add(&block->left->tmp_link, &dfs);
+ list_add(&block->right->tmp_link, &dfs);
} while (1);
return ERR_PTR(-ENOSPC);
err_undo:
- /*
- * We really don't want to leave around a bunch of split blocks, since
- * bigger is better, so make sure we merge everything back before we
- * free the allocated blocks.
- */
- buddy = __get_buddy(block);
- if (buddy &&
- (gpu_buddy_block_is_free(block) &&
- gpu_buddy_block_is_free(buddy)))
- __gpu_buddy_free(mm, block, false);
+ undo_partial_split(mm, block);
return ERR_PTR(err);
}
-static struct gpu_buddy_block *
-__gpu_buddy_alloc_range_bias(struct gpu_buddy *mm,
- u64 start, u64 end,
- unsigned int order,
- unsigned long flags)
-{
- struct gpu_buddy_block *block;
- bool fallback = false;
-
- block = __alloc_range_bias(mm, start, end, order,
- flags, fallback);
- if (IS_ERR(block))
- return __alloc_range_bias(mm, start, end, order,
- flags, !fallback);
-
- return block;
-}
-
static struct gpu_buddy_block *
get_maxblock(struct gpu_buddy *mm,
unsigned int order,
- enum gpu_buddy_free_tree tree)
+ unsigned long flags)
{
- struct gpu_buddy_block *max_block = NULL, *block = NULL;
- struct rb_root *root;
+ struct gpu_buddy_block *max_block = NULL, *block;
unsigned int i;
for (i = order; i <= mm->max_order; ++i) {
- root = &mm->free_trees[tree][i];
- block = rbtree_last_free_block(root);
+ block = rbtree_last_free_block(&mm->free_tree[i]);
if (!block)
continue;
-
- if (!max_block) {
+ if (!max_block ||
+ gpu_buddy_block_offset(block) > gpu_buddy_block_offset(max_block))
max_block = block;
- continue;
- }
-
- if (gpu_buddy_block_offset(block) >
- gpu_buddy_block_offset(max_block)) {
- max_block = block;
- }
}
return max_block;
@@ -789,44 +986,23 @@ alloc_from_freetree(struct gpu_buddy *mm,
unsigned long flags)
{
struct gpu_buddy_block *block = NULL;
- struct rb_root *root;
- enum gpu_buddy_free_tree tree;
unsigned int tmp;
int err;
- tree = (flags & GPU_BUDDY_CLEAR_ALLOCATION) ?
- GPU_BUDDY_CLEAR_TREE : GPU_BUDDY_DIRTY_TREE;
-
if (flags & GPU_BUDDY_TOPDOWN_ALLOCATION) {
- block = get_maxblock(mm, order, tree);
+ block = get_maxblock(mm, order, flags);
if (block)
- /* Store the obtained block order */
tmp = gpu_buddy_block_order(block);
} else {
for (tmp = order; tmp <= mm->max_order; ++tmp) {
- /* Get RB tree root for this order and tree */
- root = &mm->free_trees[tree][tmp];
- block = rbtree_last_free_block(root);
+ block = rbtree_last_free_block(&mm->free_tree[tmp]);
if (block)
break;
}
}
- if (!block) {
- /* Try allocating from the other tree */
- tree = (tree == GPU_BUDDY_CLEAR_TREE) ?
- GPU_BUDDY_DIRTY_TREE : GPU_BUDDY_CLEAR_TREE;
-
- for (tmp = order; tmp <= mm->max_order; ++tmp) {
- root = &mm->free_trees[tree][tmp];
- block = rbtree_last_free_block(root);
- if (block)
- break;
- }
-
- if (!block)
- return ERR_PTR(-ENOSPC);
- }
+ if (!block)
+ return ERR_PTR(-ENOSPC);
BUG_ON(!gpu_buddy_block_is_free(block));
@@ -835,14 +1011,18 @@ alloc_from_freetree(struct gpu_buddy *mm,
if (unlikely(err))
goto err_undo;
- block = block->right;
+ if (!(flags & GPU_BUDDY_CLEAR_ALLOCATION) &&
+ gpu_buddy_block_is_clear(block->right))
+ block = block->left;
+ else
+ block = block->right;
tmp--;
}
return block;
err_undo:
if (tmp != order)
- __gpu_buddy_free(mm, block, false);
+ __gpu_buddy_free(mm, block);
return ERR_PTR(err);
}
@@ -863,12 +1043,11 @@ static bool gpu_buddy_subtree_can_satisfy(struct rb_node *node,
static struct gpu_buddy_block *
gpu_buddy_find_block_aligned(struct gpu_buddy *mm,
- enum gpu_buddy_free_tree tree,
unsigned int order,
unsigned int alignment,
unsigned long flags)
{
- struct rb_root *root = &mm->free_trees[tree][order];
+ struct rb_root *root = &mm->free_tree[order];
struct rb_node *rb = root->rb_node;
while (rb) {
@@ -906,8 +1085,6 @@ gpu_buddy_offset_aligned_allocation(struct gpu_buddy *mm,
{
struct gpu_buddy_block *block = NULL;
unsigned int order, tmp, alignment;
- struct gpu_buddy_block *buddy;
- enum gpu_buddy_free_tree tree;
unsigned long pages;
int err;
@@ -915,19 +1092,8 @@ gpu_buddy_offset_aligned_allocation(struct gpu_buddy *mm,
pages = size >> ilog2(mm->chunk_size);
order = fls(pages) - 1;
- tree = (flags & GPU_BUDDY_CLEAR_ALLOCATION) ?
- GPU_BUDDY_CLEAR_TREE : GPU_BUDDY_DIRTY_TREE;
-
for (tmp = order; tmp <= mm->max_order; ++tmp) {
- block = gpu_buddy_find_block_aligned(mm, tree, tmp,
- alignment, flags);
- if (!block) {
- tree = (tree == GPU_BUDDY_CLEAR_TREE) ?
- GPU_BUDDY_DIRTY_TREE : GPU_BUDDY_CLEAR_TREE;
- block = gpu_buddy_find_block_aligned(mm, tree, tmp,
- alignment, flags);
- }
-
+ block = gpu_buddy_find_block_aligned(mm, tmp, alignment, flags);
if (block)
break;
}
@@ -954,27 +1120,18 @@ gpu_buddy_offset_aligned_allocation(struct gpu_buddy *mm,
return block;
err_undo:
- /*
- * We really don't want to leave around a bunch of split blocks, since
- * bigger is better, so make sure we merge everything back before we
- * free the allocated blocks.
- */
- buddy = __get_buddy(block);
- if (buddy &&
- (gpu_buddy_block_is_free(block) &&
- gpu_buddy_block_is_free(buddy)))
- __gpu_buddy_free(mm, block, false);
+ undo_partial_split(mm, block);
return ERR_PTR(err);
}
static int __alloc_range(struct gpu_buddy *mm,
struct list_head *dfs,
u64 start, u64 size,
+ unsigned long flags,
struct list_head *blocks,
u64 *total_allocated_on_err)
{
struct gpu_buddy_block *block;
- struct gpu_buddy_block *buddy;
u64 total_allocated = 0;
LIST_HEAD(allocated);
u64 end;
@@ -1007,16 +1164,24 @@ static int __alloc_range(struct gpu_buddy *mm,
if (contains(start, end, block_start, block_end)) {
if (gpu_buddy_block_is_free(block)) {
+ u64 bsize = gpu_buddy_block_size(mm, block);
+ u64 boff = gpu_buddy_block_offset(block);
+
mark_allocated(mm, block);
- total_allocated += gpu_buddy_block_size(mm, block);
- mm->avail -= gpu_buddy_block_size(mm, block);
- if (gpu_buddy_block_is_clear(block))
- mm->clear_avail -= gpu_buddy_block_size(mm, block);
+ total_allocated += bsize;
+ mm->avail -= bsize;
+
+ if (gpu_clear_tracker_is_clear(&mm->clear,
+ boff, bsize)) {
+ if (flags & GPU_BUDDY_CLEAR_ALLOCATION)
+ block->header |= GPU_BUDDY_HEADER_CLEAR;
+ }
+ gpu_clear_tracker_mark_dirty(&mm->clear,
+ boff, bsize);
+ mm->clear_avail = mm->clear.total_clear;
+
list_add_tail(&block->link, &allocated);
continue;
- } else if (!mm->clear_avail) {
- err = -ENOSPC;
- goto err_free;
}
}
@@ -1040,16 +1205,7 @@ static int __alloc_range(struct gpu_buddy *mm,
return 0;
err_undo:
- /*
- * We really don't want to leave around a bunch of split blocks, since
- * bigger is better, so make sure we merge everything back before we
- * free the allocated blocks.
- */
- buddy = __get_buddy(block);
- if (buddy &&
- (gpu_buddy_block_is_free(block) &&
- gpu_buddy_block_is_free(buddy)))
- __gpu_buddy_free(mm, block, false);
+ undo_partial_split(mm, block);
err_free:
if (err == -ENOSPC && total_allocated_on_err) {
@@ -1065,6 +1221,7 @@ static int __alloc_range(struct gpu_buddy *mm,
static int __gpu_buddy_alloc_range(struct gpu_buddy *mm,
u64 start,
u64 size,
+ unsigned long flags,
u64 *total_allocated_on_err,
struct list_head *blocks)
{
@@ -1074,20 +1231,23 @@ static int __gpu_buddy_alloc_range(struct gpu_buddy *mm,
for (i = 0; i < mm->n_roots; ++i)
list_add_tail(&mm->roots[i]->tmp_link, &dfs);
- return __alloc_range(mm, &dfs, start, size,
+ return __alloc_range(mm, &dfs, start, size, flags,
blocks, total_allocated_on_err);
}
static int __alloc_contig_try_harder(struct gpu_buddy *mm,
u64 size,
u64 min_block_size,
+ unsigned long flags,
struct list_head *blocks)
{
u64 rhs_offset, lhs_offset, lhs_size, filled;
struct gpu_buddy_block *block;
- unsigned int tree, order;
LIST_HEAD(blocks_lhs);
+ struct rb_root *root;
+ struct rb_node *iter;
unsigned long pages;
+ unsigned int order;
u64 modify_size;
int err;
@@ -1097,45 +1257,40 @@ static int __alloc_contig_try_harder(struct gpu_buddy *mm,
if (order == 0)
return -ENOSPC;
- for_each_free_tree(tree) {
- struct rb_root *root;
- struct rb_node *iter;
-
- root = &mm->free_trees[tree][order];
- if (rbtree_is_empty(root))
- continue;
+ root = &mm->free_tree[order];
+ if (RB_EMPTY_ROOT(root))
+ return -ENOSPC;
- iter = rb_last(root);
- while (iter) {
- block = rbtree_get_free_block(iter);
-
- /* Allocate blocks traversing RHS */
- rhs_offset = gpu_buddy_block_offset(block);
- err = __gpu_buddy_alloc_range(mm, rhs_offset, size,
- &filled, blocks);
- if (!err || err != -ENOSPC)
- return err;
-
- lhs_size = max((size - filled), min_block_size);
- if (!IS_ALIGNED(lhs_size, min_block_size))
- lhs_size = round_up(lhs_size, min_block_size);
-
- /* Allocate blocks traversing LHS */
- lhs_offset = gpu_buddy_block_offset(block) - lhs_size;
- err = __gpu_buddy_alloc_range(mm, lhs_offset, lhs_size,
- NULL, &blocks_lhs);
- if (!err) {
- list_splice(&blocks_lhs, blocks);
- return 0;
- } else if (err != -ENOSPC) {
- gpu_buddy_free_list_internal(mm, blocks);
- return err;
- }
- /* Free blocks for the next iteration */
+ iter = rb_last(root);
+ while (iter) {
+ block = rbtree_get_free_block(iter);
+
+ /* Allocate blocks traversing RHS */
+ rhs_offset = gpu_buddy_block_offset(block);
+ err = __gpu_buddy_alloc_range(mm, rhs_offset, size,
+ flags, &filled, blocks);
+ if (!err || err != -ENOSPC)
+ return err;
+
+ lhs_size = max((size - filled), min_block_size);
+ if (!IS_ALIGNED(lhs_size, min_block_size))
+ lhs_size = round_up(lhs_size, min_block_size);
+
+ /* Allocate blocks traversing LHS */
+ lhs_offset = gpu_buddy_block_offset(block) - lhs_size;
+ err = __gpu_buddy_alloc_range(mm, lhs_offset, lhs_size,
+ flags, NULL, &blocks_lhs);
+ if (!err) {
+ list_splice(&blocks_lhs, blocks);
+ return 0;
+ } else if (err != -ENOSPC) {
gpu_buddy_free_list_internal(mm, blocks);
-
- iter = rb_prev(iter);
+ return err;
}
+ /* Free blocks for the next iteration */
+ gpu_buddy_free_list_internal(mm, blocks);
+
+ iter = rb_prev(iter);
}
return -ENOSPC;
@@ -1169,6 +1324,7 @@ int gpu_buddy_block_trim(struct gpu_buddy *mm,
struct gpu_buddy_block *block;
u64 block_start, block_end;
LIST_HEAD(dfs);
+ bool was_clear;
u64 new_start;
int err;
@@ -1209,22 +1365,38 @@ int gpu_buddy_block_trim(struct gpu_buddy *mm,
}
list_del(&block->link);
+
+ was_clear = gpu_buddy_block_is_clear(block);
+ block->header &= ~GPU_BUDDY_HEADER_CLEAR;
+
+ if (was_clear) {
+ gpu_clear_tracker_mark_clear(&mm->clear,
+ gpu_buddy_block_offset(block),
+ gpu_buddy_block_size(mm, block));
+ mm->clear_avail = mm->clear.total_clear;
+ }
+
mark_free(mm, block);
mm->avail += gpu_buddy_block_size(mm, block);
- if (gpu_buddy_block_is_clear(block))
- mm->clear_avail += gpu_buddy_block_size(mm, block);
/* Prevent recursively freeing this node */
parent = block->parent;
block->parent = NULL;
list_add(&block->tmp_link, &dfs);
- err = __alloc_range(mm, &dfs, new_start, new_size, blocks, NULL);
+ err = __alloc_range(mm, &dfs, new_start, new_size,
+ was_clear ? GPU_BUDDY_CLEAR_ALLOCATION : 0,
+ blocks, NULL);
if (err) {
mark_allocated(mm, block);
mm->avail -= gpu_buddy_block_size(mm, block);
- if (gpu_buddy_block_is_clear(block))
- mm->clear_avail -= gpu_buddy_block_size(mm, block);
+ if (was_clear) {
+ gpu_clear_tracker_mark_dirty(&mm->clear,
+ gpu_buddy_block_offset(block),
+ gpu_buddy_block_size(mm, block));
+ mm->clear_avail = mm->clear.total_clear;
+ block->header |= GPU_BUDDY_HEADER_CLEAR;
+ }
list_add(&block->link, blocks);
}
@@ -1233,6 +1405,21 @@ int gpu_buddy_block_trim(struct gpu_buddy *mm,
}
EXPORT_SYMBOL(gpu_buddy_block_trim);
+static bool clear_steer_window(struct gpu_buddy *mm, u64 min_sz,
+ u64 *start, u64 *end, unsigned long *flags)
+{
+ struct gpu_clear_extent *ext =
+ gpu_clear_tracker_find(&mm->clear, min_sz);
+
+ if (!ext)
+ return false;
+
+ *start = ext->start;
+ *end = ext->end;
+ *flags |= GPU_BUDDY_RANGE_ALLOCATION;
+ return true;
+}
+
static struct gpu_buddy_block *
__gpu_buddy_alloc_blocks(struct gpu_buddy *mm,
u64 start, u64 end,
@@ -1240,18 +1427,24 @@ __gpu_buddy_alloc_blocks(struct gpu_buddy *mm,
unsigned int order,
unsigned long flags)
{
+ /* Steer cleared allocations to a cleared extent that fits the order */
+ if (!(flags & GPU_BUDDY_RANGE_ALLOCATION) &&
+ (flags & GPU_BUDDY_CLEAR_ALLOCATION) && mm->clear_avail)
+ clear_steer_window(mm, mm->chunk_size << order,
+ &start, &end, &flags);
+
if (flags & GPU_BUDDY_RANGE_ALLOCATION)
/* Allocate traversing within the range */
- return __gpu_buddy_alloc_range_bias(mm, start, end,
- order, flags);
- else if (size < min_block_size)
+ return __alloc_range_bias(mm, start, end, order, flags);
+
+ if (size < min_block_size)
/* Allocate from an offset-aligned region without size rounding */
return gpu_buddy_offset_aligned_allocation(mm, size,
min_block_size,
flags);
- else
- /* Allocate from freetree */
- return alloc_from_freetree(mm, order, flags);
+
+ /* Allocate from freetree */
+ return alloc_from_freetree(mm, order, flags);
}
/**
@@ -1310,7 +1503,7 @@ int gpu_buddy_alloc_blocks(struct gpu_buddy *mm,
if (!IS_ALIGNED(start | end, min_block_size))
return -EINVAL;
- return __gpu_buddy_alloc_range(mm, start, size, NULL, blocks);
+ return __gpu_buddy_alloc_range(mm, start, size, flags, NULL, blocks);
}
original_size = size;
@@ -1336,7 +1529,7 @@ int gpu_buddy_alloc_blocks(struct gpu_buddy *mm,
if ((flags & GPU_BUDDY_CONTIGUOUS_ALLOCATION) &&
!(flags & GPU_BUDDY_RANGE_ALLOCATION))
return __alloc_contig_try_harder(mm, original_size,
- original_min_size, blocks);
+ original_min_size, flags, blocks);
return -EINVAL;
}
@@ -1351,8 +1544,6 @@ int gpu_buddy_alloc_blocks(struct gpu_buddy *mm,
BUG_ON(size >= min_block_size && order < min_order);
do {
- unsigned int fallback_order;
-
block = __gpu_buddy_alloc_blocks(mm, start,
end,
size,
@@ -1362,30 +1553,11 @@ int gpu_buddy_alloc_blocks(struct gpu_buddy *mm,
if (!IS_ERR(block))
break;
- if (size < min_block_size) {
- fallback_order = order;
- } else if (order == min_order) {
- fallback_order = min_order;
- } else {
+ if (size >= min_block_size && order > min_order) {
order--;
continue;
}
- /* Try allocation through force merge method */
- if (mm->clear_avail &&
- !__force_merge(mm, start, end, fallback_order)) {
- block = __gpu_buddy_alloc_blocks(mm, start,
- end,
- size,
- min_block_size,
- fallback_order,
- flags);
- if (!IS_ERR(block)) {
- order = fallback_order;
- break;
- }
- }
-
/*
* Try contiguous block allocation through
* try harder method.
@@ -1395,6 +1567,7 @@ int gpu_buddy_alloc_blocks(struct gpu_buddy *mm,
return __alloc_contig_try_harder(mm,
original_size,
original_min_size,
+ flags,
blocks);
err = -ENOSPC;
goto err_free;
@@ -1402,8 +1575,23 @@ int gpu_buddy_alloc_blocks(struct gpu_buddy *mm,
mark_allocated(mm, block);
mm->avail -= gpu_buddy_block_size(mm, block);
- if (gpu_buddy_block_is_clear(block))
- mm->clear_avail -= gpu_buddy_block_size(mm, block);
+
+ /*
+ * Mark the block CLEAR only if the caller requested cleared
+ * memory and the entire block lies within a cleared extent.
+ * Either way, drop the block's range from the clear tracker
+ * since it is now allocated.
+ */
+ if (flags & GPU_BUDDY_CLEAR_ALLOCATION &&
+ gpu_clear_tracker_is_clear(&mm->clear,
+ gpu_buddy_block_offset(block),
+ gpu_buddy_block_size(mm, block)))
+ block->header |= GPU_BUDDY_HEADER_CLEAR;
+
+ gpu_clear_tracker_mark_dirty(&mm->clear,
+ gpu_buddy_block_offset(block),
+ gpu_buddy_block_size(mm, block));
+ mm->clear_avail = mm->clear.total_clear;
kmemleak_update_trace(block);
list_add_tail(&block->link, &allocated);
@@ -1473,26 +1661,52 @@ EXPORT_SYMBOL(gpu_buddy_block_print);
*/
void gpu_buddy_print(struct gpu_buddy *mm)
{
+ u64 *clear_count;
+ unsigned int i;
int order;
pr_info("chunk_size: %lluKiB, total: %lluMiB, free: %lluMiB, clear_free: %lluMiB\n",
mm->chunk_size >> 10, mm->size >> 20, mm->avail >> 20, mm->clear_avail >> 20);
+ clear_count = kcalloc(mm->max_order + 1, sizeof(*clear_count), GFP_KERNEL);
+
+ if (clear_count) {
+ LIST_HEAD(dfs);
+
+ for (i = 0; i < mm->n_roots; i++)
+ list_add_tail(&mm->roots[i]->tmp_link, &dfs);
+
+ while (!list_empty(&dfs)) {
+ struct gpu_buddy_block *block =
+ list_first_entry(&dfs, struct gpu_buddy_block, tmp_link);
+
+ list_del(&block->tmp_link);
+
+ if (gpu_buddy_block_is_split(block)) {
+ list_add(&block->right->tmp_link, &dfs);
+ list_add(&block->left->tmp_link, &dfs);
+ } else if (gpu_buddy_block_is_free(block) &&
+ gpu_buddy_block_is_clear(block)) {
+ clear_count[gpu_buddy_block_order(block)]++;
+ }
+ }
+ }
+
for (order = mm->max_order; order >= 0; order--) {
struct gpu_buddy_block *block, *tmp;
struct rb_root *root;
u64 count = 0, free;
- unsigned int tree;
- for_each_free_tree(tree) {
- root = &mm->free_trees[tree][order];
+ root = &mm->free_tree[order];
- rbtree_postorder_for_each_entry_safe(block, tmp, root, rb) {
- BUG_ON(!gpu_buddy_block_is_free(block));
- count++;
- }
+ rbtree_postorder_for_each_entry_safe(block, tmp, root, rb) {
+ BUG_ON(!gpu_buddy_block_is_free(block));
+ count++;
}
+ if (clear_count)
+ count += clear_count[order];
+
free = count * (mm->chunk_size << order);
if (free < SZ_1M)
pr_info("order-%2d free: %8llu KiB, blocks: %llu\n",
@@ -1501,11 +1715,14 @@ void gpu_buddy_print(struct gpu_buddy *mm)
pr_info("order-%2d free: %8llu MiB, blocks: %llu\n",
order, free >> 20, count);
}
+
+ kfree(clear_count);
}
EXPORT_SYMBOL(gpu_buddy_print);
static void gpu_buddy_module_exit(void)
{
+ kmem_cache_destroy(slab_extents);
kmem_cache_destroy(slab_blocks);
}
@@ -1515,6 +1732,12 @@ static int __init gpu_buddy_module_init(void)
if (!slab_blocks)
return -ENOMEM;
+ slab_extents = KMEM_CACHE(gpu_clear_extent, 0);
+ if (!slab_extents) {
+ kmem_cache_destroy(slab_blocks);
+ return -ENOMEM;
+ }
+
return 0;
}
diff --git a/drivers/gpu/drm/drm_buddy.c b/drivers/gpu/drm/drm_buddy.c
index 841f3de5f307..a0ba2e61ec23 100644
--- a/drivers/gpu/drm/drm_buddy.c
+++ b/drivers/gpu/drm/drm_buddy.c
@@ -9,6 +9,7 @@
#include <linux/kmemleak.h>
#include <linux/module.h>
#include <linux/sizes.h>
+#include <linux/slab.h>
#include <linux/gpu_buddy.h>
#include <drm/drm_buddy.h>
@@ -40,26 +41,51 @@ EXPORT_SYMBOL(drm_buddy_block_print);
*/
void drm_buddy_print(struct gpu_buddy *mm, struct drm_printer *p)
{
+ u64 *clear_count;
int order;
+ unsigned int i;
drm_printf(p, "chunk_size: %lluKiB, total: %lluMiB, free: %lluMiB, clear_free: %lluMiB\n",
mm->chunk_size >> 10, mm->size >> 20, mm->avail >> 20, mm->clear_avail >> 20);
+ clear_count = kcalloc(mm->max_order + 1, sizeof(*clear_count), GFP_KERNEL);
+
+ if (clear_count) {
+ LIST_HEAD(dfs);
+
+ for (i = 0; i < mm->n_roots; i++)
+ list_add_tail(&mm->roots[i]->tmp_link, &dfs);
+
+ while (!list_empty(&dfs)) {
+ struct gpu_buddy_block *block =
+ list_first_entry(&dfs, struct gpu_buddy_block, tmp_link);
+
+ list_del(&block->tmp_link);
+
+ if ((block->header & GPU_BUDDY_HEADER_STATE) == GPU_BUDDY_SPLIT) {
+ list_add(&block->right->tmp_link, &dfs);
+ list_add(&block->left->tmp_link, &dfs);
+ } else if (gpu_buddy_block_is_free(block) &&
+ gpu_buddy_block_is_clear(block)) {
+ clear_count[gpu_buddy_block_order(block)]++;
+ }
+ }
+ }
+
for (order = mm->max_order; order >= 0; order--) {
struct gpu_buddy_block *block, *tmp;
struct rb_root *root;
u64 count = 0, free;
- unsigned int tree;
-
- for_each_free_tree(tree) {
- root = &mm->free_trees[tree][order];
- rbtree_postorder_for_each_entry_safe(block, tmp, root, rb) {
- BUG_ON(!gpu_buddy_block_is_free(block));
- count++;
- }
+ root = &mm->free_tree[order];
+ rbtree_postorder_for_each_entry_safe(block, tmp, root, rb) {
+ BUG_ON(!gpu_buddy_block_is_free(block));
+ count++;
}
+ if (clear_count)
+ count += clear_count[order];
+
drm_printf(p, "order-%2d ", order);
free = count * (mm->chunk_size << order);
@@ -70,6 +96,8 @@ void drm_buddy_print(struct gpu_buddy *mm, struct drm_printer *p)
drm_printf(p, ", blocks: %llu\n", count);
}
+
+ kfree(clear_count);
}
EXPORT_SYMBOL(drm_buddy_print);
diff --git a/include/linux/gpu_buddy.h b/include/linux/gpu_buddy.h
index 5fa917ba5450..435c873f3469 100644
--- a/include/linux/gpu_buddy.h
+++ b/include/linux/gpu_buddy.h
@@ -67,15 +67,6 @@
*/
#define GPU_BUDDY_TRIM_DISABLE BIT(5)
-enum gpu_buddy_free_tree {
- GPU_BUDDY_CLEAR_TREE = 0,
- GPU_BUDDY_DIRTY_TREE,
- GPU_BUDDY_MAX_FREE_TREES,
-};
-
-#define for_each_free_tree(tree) \
- for ((tree) = 0; (tree) < GPU_BUDDY_MAX_FREE_TREES; (tree)++)
-
/**
* struct gpu_buddy_block - Block within a buddy allocator
*
@@ -103,6 +94,14 @@ struct gpu_buddy_block {
#define GPU_BUDDY_ALLOCATED (1 << 10)
#define GPU_BUDDY_FREE (2 << 10)
#define GPU_BUDDY_SPLIT (3 << 10)
+/*
+ * GPU_BUDDY_HEADER_CLEAR has two roles:
+ * - FREE state: set when the block's full range is cleared (tracker
+ * confirmed). Cleared free blocks float in the buddy
+ * tree and are NOT inserted into free_tree[].
+ * - ALLOCATED state: set when the block was served from cleared memory,
+ * informing the caller that no GPU clear pass is needed.
+ */
#define GPU_BUDDY_HEADER_CLEAR GENMASK_ULL(9, 9)
/* Free to be used, if needed in the future */
#define GPU_BUDDY_HEADER_UNUSED GENMASK_ULL(8, 6)
@@ -135,6 +134,38 @@ struct gpu_buddy_block {
/* Order-zero must be at least SZ_4K */
#define GPU_BUDDY_MAX_ORDER (63 - 12)
+/**
+ * struct gpu_clear_extent - a contiguous cleared (zeroed) address range
+ *
+ * Tracks a single contiguous address range whose memory content is known
+ * to be zeroed. Extents are non-overlapping and stored in an augmented
+ * red-black tree sorted by @start. The augmented value @subtree_max_size
+ * allows O(log N) search for an extent of at least a given size.
+ */
+struct gpu_clear_extent {
+/* private: */
+ struct rb_node rb;
+ u64 start;
+ u64 end;
+ u64 subtree_max_size;
+};
+
+/**
+ * struct gpu_clear_tracker - tracks cleared (zeroed) address intervals
+ *
+ * Maintains a set of non-overlapping cleared extents as an augmented
+ * red-black tree. The tracker is embedded inside struct gpu_buddy and
+ * replaces the former dual (clear/dirty) free-tree scheme.
+ *
+ * @total_clear: Total bytes of cleared memory currently tracked.
+ */
+struct gpu_clear_tracker {
+/* private: */
+ struct rb_root root;
+/* public: */
+ u64 total_clear;
+};
+
/**
* struct gpu_buddy - GPU binary buddy allocator
*
@@ -158,13 +189,12 @@ struct gpu_buddy_block {
struct gpu_buddy {
/* private: */
/*
- * Array of red-black trees for free block management.
- * Indexed as free_trees[clear/dirty][order] where:
- * - Index 0 (GPU_BUDDY_CLEAR_TREE): blocks with zeroed content
- * - Index 1 (GPU_BUDDY_DIRTY_TREE): blocks with unknown content
- * Each tree holds free blocks of the corresponding order.
+ * One RB-tree per order containing only dirty/mixed free blocks.
+ * Cleared free blocks are NOT inserted here; they float in the buddy
+ * tree and are located exclusively via the @clear tracker.
*/
- struct rb_root **free_trees;
+ struct rb_root *free_tree;
+
/*
* Array of root blocks representing the top-level blocks of the
* binary tree(s). Multiple roots exist when the total size is not
@@ -172,6 +202,7 @@ struct gpu_buddy {
* that fits in the remaining space.
*/
struct gpu_buddy_block **roots;
+ struct gpu_clear_tracker clear;
/* public: */
unsigned int n_roots;
unsigned int max_order;
base-commit: d37690b5e02418a2365548300628ef3895a24ed2
--
2.34.1
^ permalink raw reply related [flat|nested] 8+ messages in thread
* [PATCH v2 2/2] gpu/tests/buddy: add clear-tracker allocation latency benchmarks
2026-05-04 11:10 [PATCH v2 1/2] gpu/buddy: replace dual-tree/force_merge with decoupled clear tracker Arunpravin Paneer Selvam
@ 2026-05-04 11:10 ` Arunpravin Paneer Selvam
2026-05-04 14:48 ` ✗ CI.checkpatch: warning for series starting with [v2,1/2] gpu/buddy: replace dual-tree/force_merge with decoupled clear tracker Patchwork
` (4 subsequent siblings)
5 siblings, 0 replies; 8+ messages in thread
From: Arunpravin Paneer Selvam @ 2026-05-04 11:10 UTC (permalink / raw)
To: matthew.auld, christian.koenig, dri-devel, intel-gfx, intel-xe,
amd-gfx
Cc: alexander.deucher, Arunpravin Paneer Selvam
Add gpu_test_buddy_clear_tracker_performance test case that measures
allocation latency before and after replacing the dual-tree /
force_merge design with a decoupled clear tracker.
Two scenarios are covered.
1. Single contiguous allocation after fragmentation. A 4 GiB pool is
filled with 4 KiB blocks and freed in alternating clear/dirty order,
so every buddy pair ends up split across the two trees and cannot
coalesce at free() time. A single contiguous 4 GiB allocation then
takes ~61 ms on the dual-tree design (the alloc path has to invoke
__force_merge() to climb back up to max_order) and ~25 ms with the
clear tracker (the pool is already coalesced at free() time).
2. Repeated allocations from a fragmented pool. Same 4 GiB pool, freed
with even-indexed blocks cleared and odd-indexed dirty so every
adjacent buddy pair sits on opposite sides of the merge barrier.
16384 x 256 KiB allocations then take ~80 ms on the dual-tree
design (each alloc pays the __force_merge() cost) and ~39 ms with
the clear tracker (free-time merging makes each alloc an O(log N)
split).
v2:
- Removed unwanted sub tests
Cc: Matthew Auld <matthew.auld@intel.com>
Cc: Christian König <christian.koenig@amd.com>
Signed-off-by: Arunpravin Paneer Selvam <Arunpravin.PaneerSelvam@amd.com>
---
drivers/gpu/tests/gpu_buddy_test.c | 140 +++++++++++++++++++++++++----
1 file changed, 122 insertions(+), 18 deletions(-)
diff --git a/drivers/gpu/tests/gpu_buddy_test.c b/drivers/gpu/tests/gpu_buddy_test.c
index 7df5c2ae83bb..5040216ead2b 100644
--- a/drivers/gpu/tests/gpu_buddy_test.c
+++ b/drivers/gpu/tests/gpu_buddy_test.c
@@ -38,7 +38,7 @@ static void gpu_test_buddy_subtree_offset_alignment_stress(struct kunit *test)
};
struct list_head allocated[ARRAY_SIZE(alignments)];
unsigned int i, max_subtree_align = 0;
- int ret, tree, order;
+ int ret, order;
struct gpu_buddy mm;
KUNIT_ASSERT_FALSE_MSG(test, gpu_buddy_init(&mm, mm_size, SZ_4K),
@@ -78,14 +78,11 @@ static void gpu_test_buddy_subtree_offset_alignment_stress(struct kunit *test)
}
for (order = mm.max_order; order >= 0 && !root; order--) {
- for (tree = 0; tree < 2; tree++) {
- node = mm.free_trees[tree][order].rb_node;
- if (node) {
- root = container_of(node,
- struct gpu_buddy_block,
- rb);
- break;
- }
+ node = mm.free_tree[order].rb_node;
+ if (node) {
+ root = container_of(node,
+ struct gpu_buddy_block,
+ rb);
}
}
@@ -97,15 +94,13 @@ static void gpu_test_buddy_subtree_offset_alignment_stress(struct kunit *test)
gpu_buddy_free_list(&mm, &allocated[i], 0);
for (order = 0; order <= mm.max_order; order++) {
- for (tree = 0; tree < 2; tree++) {
- node = mm.free_trees[tree][order].rb_node;
- if (!node)
- continue;
-
- block = container_of(node, struct gpu_buddy_block, rb);
- max_subtree_align = max(max_subtree_align,
- block->subtree_max_alignment);
- }
+ node = mm.free_tree[order].rb_node;
+ if (!node)
+ continue;
+
+ block = container_of(node, struct gpu_buddy_block, rb);
+ max_subtree_align = max(max_subtree_align,
+ block->subtree_max_alignment);
}
KUNIT_EXPECT_GE(test, max_subtree_align, ilog2(alignments[i]));
@@ -289,6 +284,114 @@ static void gpu_test_buddy_fragmentation_performance(struct kunit *test)
gpu_buddy_fini(&mm);
}
+static void gpu_test_buddy_clear_tracker_performance(struct kunit *test)
+{
+ struct gpu_buddy_block *block, *tmp;
+ unsigned long elapsed_ms;
+ LIST_HEAD(clear_blocks);
+ LIST_HEAD(dirty_blocks);
+ LIST_HEAD(allocated);
+ struct gpu_buddy mm;
+ LIST_HEAD(results);
+ ktime_t start, end;
+ int i, count;
+
+ /*
+ * Contiguous alloc latency after alternating clear/dirty fragmentation
+ *
+ * Fill a 4 GiB pool with 4 KiB allocations, partition them into
+ * alternating cleared and dirty sets, then free both. In the old
+ * dual-tree design every adjacent buddy pair has one cleared half and
+ * one dirty half, so the pair sits on opposite sides of the clear/dirty
+ * merge barrier and cannot be coalesced at free() time. The pool
+ * stays fully fragmented and the subsequent contiguous 4 GiB allocation
+ * has to invoke __force_merge() to climb back up to max_order before
+ * it can succeed. With the clear-tracker design buddy pairs coalesce
+ * unconditionally during free(), so the pool is already at max_order
+ * before the timed alloc begins and __force_merge() is not needed.
+ */
+ KUNIT_ASSERT_FALSE_MSG(test, gpu_buddy_init(&mm, SZ_4G, SZ_4K),
+ "buddy_init failed\n");
+
+ for (i = 0; i < SZ_4G / SZ_4K; i++)
+ KUNIT_ASSERT_FALSE_MSG(test,
+ gpu_buddy_alloc_blocks(&mm, 0, SZ_4G, SZ_4K, SZ_4K,
+ &allocated, 0),
+ "buddy_alloc hit an error size=%u\n", SZ_4K);
+
+ count = 0;
+ list_for_each_entry_safe(block, tmp, &allocated, link) {
+ if (count++ % 2 == 0)
+ list_move_tail(&block->link, &clear_blocks);
+ else
+ list_move_tail(&block->link, &dirty_blocks);
+ }
+
+ gpu_buddy_free_list(&mm, &clear_blocks, GPU_BUDDY_CLEARED);
+ gpu_buddy_free_list(&mm, &dirty_blocks, 0);
+
+ start = ktime_get();
+ KUNIT_ASSERT_FALSE_MSG(test,
+ gpu_buddy_alloc_blocks(&mm, 0, SZ_4G, SZ_4G, SZ_4K,
+ &results, 0),
+ "contiguous alloc failed\n");
+ end = ktime_get();
+ elapsed_ms = ktime_to_ms(ktime_sub(end, start));
+
+ kunit_info(test, "Contiguous alloc after fragmentation: %lu ms\n",
+ elapsed_ms);
+
+ gpu_buddy_free_list(&mm, &results, 0);
+ gpu_buddy_fini(&mm);
+
+ /*
+ * Repeated alloc throughput from a maximally fragmented pool
+ *
+ * Fill a 4 GiB pool with 4 KiB allocations, free even-indexed blocks
+ * as cleared and odd-indexed blocks as dirty. The alternating pattern
+ * ensures every adjacent buddy pair has one cleared half and one dirty
+ * half, so each pair lands on opposite sides of the old merge barrier.
+ * Each of the 16 384 x 256 KiB allocations in the timed loop has to
+ * pay the __force_merge() cost on the alloc path under the old design.
+ * With the clear-tracker design the pool collapses to one max_order
+ * block during free(), so each alloc is a simple O(log N) split.
+ */
+ KUNIT_ASSERT_FALSE_MSG(test, gpu_buddy_init(&mm, SZ_4G, SZ_4K),
+ "buddy_init failed\n");
+
+ for (i = 0; i < SZ_4G / SZ_4K; i++)
+ KUNIT_ASSERT_FALSE_MSG(test,
+ gpu_buddy_alloc_blocks(&mm, 0, SZ_4G, SZ_4K, SZ_4K,
+ &allocated, 0),
+ "buddy_alloc hit an error size=%u\n", SZ_4K);
+
+ count = 0;
+ list_for_each_entry_safe(block, tmp, &allocated, link) {
+ if (count++ % 2 == 0)
+ list_move_tail(&block->link, &clear_blocks);
+ else
+ list_move_tail(&block->link, &dirty_blocks);
+ }
+
+ gpu_buddy_free_list(&mm, &clear_blocks, GPU_BUDDY_CLEARED);
+ gpu_buddy_free_list(&mm, &dirty_blocks, 0);
+
+ start = ktime_get();
+ for (i = 0; i < SZ_4G / SZ_256K; i++)
+ KUNIT_ASSERT_FALSE_MSG(test,
+ gpu_buddy_alloc_blocks(&mm, 0, SZ_4G, SZ_256K, SZ_4K,
+ &results, 0),
+ "buddy_alloc hit an error size=%u\n", SZ_256K);
+ end = ktime_get();
+ elapsed_ms = ktime_to_ms(ktime_sub(end, start));
+
+ kunit_info(test, "Repeated 256 KiB allocs from fragmented pool: %lu ms\n",
+ elapsed_ms);
+
+ gpu_buddy_free_list(&mm, &results, 0);
+ gpu_buddy_fini(&mm);
+}
+
static void gpu_test_buddy_alloc_range_bias(struct kunit *test)
{
u32 mm_size, size, ps, bias_size, bias_start, bias_end, bias_rem;
@@ -1402,6 +1505,7 @@ static struct kunit_case gpu_buddy_tests[] = {
KUNIT_CASE(gpu_test_buddy_alloc_range),
KUNIT_CASE(gpu_test_buddy_alloc_range_bias),
KUNIT_CASE_SLOW(gpu_test_buddy_fragmentation_performance),
+ KUNIT_CASE_SLOW(gpu_test_buddy_clear_tracker_performance),
KUNIT_CASE(gpu_test_buddy_alloc_exceeds_max_order),
KUNIT_CASE(gpu_test_buddy_offset_aligned_allocation),
KUNIT_CASE(gpu_test_buddy_subtree_offset_alignment_stress),
--
2.34.1
^ permalink raw reply related [flat|nested] 8+ messages in thread
* ✗ CI.checkpatch: warning for series starting with [v2,1/2] gpu/buddy: replace dual-tree/force_merge with decoupled clear tracker
2026-05-04 11:10 [PATCH v2 1/2] gpu/buddy: replace dual-tree/force_merge with decoupled clear tracker Arunpravin Paneer Selvam
2026-05-04 11:10 ` [PATCH v2 2/2] gpu/tests/buddy: add clear-tracker allocation latency benchmarks Arunpravin Paneer Selvam
@ 2026-05-04 14:48 ` Patchwork
2026-05-04 14:49 ` ✓ CI.KUnit: success " Patchwork
` (3 subsequent siblings)
5 siblings, 0 replies; 8+ messages in thread
From: Patchwork @ 2026-05-04 14:48 UTC (permalink / raw)
To: Arunpravin Paneer Selvam; +Cc: intel-xe
== Series Details ==
Series: series starting with [v2,1/2] gpu/buddy: replace dual-tree/force_merge with decoupled clear tracker
URL : https://patchwork.freedesktop.org/series/165893/
State : warning
== Summary ==
+ KERNEL=/kernel
+ git clone https://gitlab.freedesktop.org/drm/maintainer-tools mt
Cloning into 'mt'...
warning: redirecting to https://gitlab.freedesktop.org/drm/maintainer-tools.git/
+ git -C mt rev-list -n1 origin/master
c8c12e558adaef7a4d125d83b6e1f8824bc13b82
+ cd /kernel
+ git config --global --add safe.directory /kernel
+ git log -n1
commit 75bd739a55c8e43ce7d9bfb7cb6acb3b014dd040
Author: Arunpravin Paneer Selvam <Arunpravin.PaneerSelvam@amd.com>
Date: Mon May 4 16:40:55 2026 +0530
gpu/tests/buddy: add clear-tracker allocation latency benchmarks
Add gpu_test_buddy_clear_tracker_performance test case that measures
allocation latency before and after replacing the dual-tree /
force_merge design with a decoupled clear tracker.
Two scenarios are covered.
1. Single contiguous allocation after fragmentation. A 4 GiB pool is
filled with 4 KiB blocks and freed in alternating clear/dirty order,
so every buddy pair ends up split across the two trees and cannot
coalesce at free() time. A single contiguous 4 GiB allocation then
takes ~61 ms on the dual-tree design (the alloc path has to invoke
__force_merge() to climb back up to max_order) and ~25 ms with the
clear tracker (the pool is already coalesced at free() time).
2. Repeated allocations from a fragmented pool. Same 4 GiB pool, freed
with even-indexed blocks cleared and odd-indexed dirty so every
adjacent buddy pair sits on opposite sides of the merge barrier.
16384 x 256 KiB allocations then take ~80 ms on the dual-tree
design (each alloc pays the __force_merge() cost) and ~39 ms with
the clear tracker (free-time merging makes each alloc an O(log N)
split).
v2:
- Removed unwanted sub tests
Cc: Matthew Auld <matthew.auld@intel.com>
Cc: Christian König <christian.koenig@amd.com>
Signed-off-by: Arunpravin Paneer Selvam <Arunpravin.PaneerSelvam@amd.com>
+ /mt/dim checkpatch 190806bad83a94cee94343c1e937c8447046976b drm-intel
456fe204242d gpu/buddy: replace dual-tree/force_merge with decoupled clear tracker
-:319: WARNING:LINE_SPACING: Missing a blank line after declarations
#319: FILE: drivers/gpu/buddy.c:256:
+ struct gpu_clear_extent *right = extent_alloc();
+ if (!right)
-:654: CHECK:PARENTHESIS_ALIGNMENT: Alignment should match open parenthesis
#654: FILE: drivers/gpu/buddy.c:627:
+ mm->free_tree = kcalloc(mm->max_order + 1,
+ sizeof(struct rb_root),
-:762: WARNING:LONG_LINE: line length of 105 exceeds 100 columns
#762: FILE: drivers/gpu/buddy.c:768:
+ rb_erase_augmented(&block->rb, &mm->free_tree[i], &gpu_buddy_augment_cb);
-:1524: WARNING:AVOID_BUG: Do not crash the kernel unless it is absolutely unavoidable--use WARN_ON_ONCE() plus recovery code (if feasible) instead of BUG() or variants
#1524: FILE: drivers/gpu/buddy.c:1703:
+ BUG_ON(!gpu_buddy_block_is_free(block));
-:1624: WARNING:AVOID_BUG: Do not crash the kernel unless it is absolutely unavoidable--use WARN_ON_ONCE() plus recovery code (if feasible) instead of BUG() or variants
#1624: FILE: drivers/gpu/drm/drm_buddy.c:82:
+ BUG_ON(!gpu_buddy_block_is_free(block));
total: 0 errors, 4 warnings, 1 checks, 1588 lines checked
75bd739a55c8 gpu/tests/buddy: add clear-tracker allocation latency benchmarks
^ permalink raw reply [flat|nested] 8+ messages in thread
* ✓ CI.KUnit: success for series starting with [v2,1/2] gpu/buddy: replace dual-tree/force_merge with decoupled clear tracker
2026-05-04 11:10 [PATCH v2 1/2] gpu/buddy: replace dual-tree/force_merge with decoupled clear tracker Arunpravin Paneer Selvam
2026-05-04 11:10 ` [PATCH v2 2/2] gpu/tests/buddy: add clear-tracker allocation latency benchmarks Arunpravin Paneer Selvam
2026-05-04 14:48 ` ✗ CI.checkpatch: warning for series starting with [v2,1/2] gpu/buddy: replace dual-tree/force_merge with decoupled clear tracker Patchwork
@ 2026-05-04 14:49 ` Patchwork
2026-05-04 15:43 ` ✓ Xe.CI.BAT: " Patchwork
` (2 subsequent siblings)
5 siblings, 0 replies; 8+ messages in thread
From: Patchwork @ 2026-05-04 14:49 UTC (permalink / raw)
To: Arunpravin Paneer Selvam; +Cc: intel-xe
== Series Details ==
Series: series starting with [v2,1/2] gpu/buddy: replace dual-tree/force_merge with decoupled clear tracker
URL : https://patchwork.freedesktop.org/series/165893/
State : success
== Summary ==
+ trap cleanup EXIT
+ /kernel/tools/testing/kunit/kunit.py run --kunitconfig /kernel/drivers/gpu/drm/xe/.kunitconfig
[14:48:15] Configuring KUnit Kernel ...
Generating .config ...
Populating config with:
$ make ARCH=um O=.kunit olddefconfig
[14:48:20] Building KUnit Kernel ...
Populating config with:
$ make ARCH=um O=.kunit olddefconfig
Building with:
$ make all compile_commands.json scripts_gdb ARCH=um O=.kunit --jobs=48
[14:48:51] Starting KUnit Kernel (1/1)...
[14:48:51] ============================================================
Running tests with:
$ .kunit/linux kunit.enable=1 mem=1G console=tty kunit_shutdown=halt
[14:48:51] ================== guc_buf (11 subtests) ===================
[14:48:51] [PASSED] test_smallest
[14:48:51] [PASSED] test_largest
[14:48:51] [PASSED] test_granular
[14:48:51] [PASSED] test_unique
[14:48:51] [PASSED] test_overlap
[14:48:51] [PASSED] test_reusable
[14:48:51] [PASSED] test_too_big
[14:48:51] [PASSED] test_flush
[14:48:51] [PASSED] test_lookup
[14:48:51] [PASSED] test_data
[14:48:51] [PASSED] test_class
[14:48:51] ===================== [PASSED] guc_buf =====================
[14:48:51] =================== guc_dbm (7 subtests) ===================
[14:48:51] [PASSED] test_empty
[14:48:51] [PASSED] test_default
[14:48:51] ======================== test_size ========================
[14:48:51] [PASSED] 4
[14:48:51] [PASSED] 8
[14:48:51] [PASSED] 32
[14:48:51] [PASSED] 256
[14:48:51] ==================== [PASSED] test_size ====================
[14:48:51] ======================= test_reuse ========================
[14:48:51] [PASSED] 4
[14:48:51] [PASSED] 8
[14:48:51] [PASSED] 32
[14:48:51] [PASSED] 256
[14:48:51] =================== [PASSED] test_reuse ====================
[14:48:51] =================== test_range_overlap ====================
[14:48:51] [PASSED] 4
[14:48:51] [PASSED] 8
[14:48:51] [PASSED] 32
[14:48:51] [PASSED] 256
[14:48:51] =============== [PASSED] test_range_overlap ================
[14:48:51] =================== test_range_compact ====================
[14:48:51] [PASSED] 4
[14:48:51] [PASSED] 8
[14:48:51] [PASSED] 32
[14:48:51] [PASSED] 256
[14:48:51] =============== [PASSED] test_range_compact ================
[14:48:51] ==================== test_range_spare =====================
[14:48:51] [PASSED] 4
[14:48:51] [PASSED] 8
[14:48:51] [PASSED] 32
[14:48:51] [PASSED] 256
[14:48:51] ================ [PASSED] test_range_spare =================
[14:48:51] ===================== [PASSED] guc_dbm =====================
[14:48:51] =================== guc_idm (6 subtests) ===================
[14:48:51] [PASSED] bad_init
[14:48:51] [PASSED] no_init
[14:48:51] [PASSED] init_fini
[14:48:51] [PASSED] check_used
[14:48:51] [PASSED] check_quota
[14:48:51] [PASSED] check_all
[14:48:51] ===================== [PASSED] guc_idm =====================
[14:48:51] ================== no_relay (3 subtests) ===================
[14:48:51] [PASSED] xe_drops_guc2pf_if_not_ready
[14:48:51] [PASSED] xe_drops_guc2vf_if_not_ready
[14:48:51] [PASSED] xe_rejects_send_if_not_ready
[14:48:51] ==================== [PASSED] no_relay =====================
[14:48:51] ================== pf_relay (14 subtests) ==================
[14:48:51] [PASSED] pf_rejects_guc2pf_too_short
[14:48:51] [PASSED] pf_rejects_guc2pf_too_long
[14:48:51] [PASSED] pf_rejects_guc2pf_no_payload
[14:48:51] [PASSED] pf_fails_no_payload
[14:48:51] [PASSED] pf_fails_bad_origin
[14:48:51] [PASSED] pf_fails_bad_type
[14:48:51] [PASSED] pf_txn_reports_error
[14:48:51] [PASSED] pf_txn_sends_pf2guc
[14:48:51] [PASSED] pf_sends_pf2guc
[14:48:51] [SKIPPED] pf_loopback_nop
[14:48:51] [SKIPPED] pf_loopback_echo
[14:48:51] [SKIPPED] pf_loopback_fail
[14:48:51] [SKIPPED] pf_loopback_busy
[14:48:51] [SKIPPED] pf_loopback_retry
[14:48:51] ==================== [PASSED] pf_relay =====================
[14:48:51] ================== vf_relay (3 subtests) ===================
[14:48:51] [PASSED] vf_rejects_guc2vf_too_short
[14:48:51] [PASSED] vf_rejects_guc2vf_too_long
[14:48:51] [PASSED] vf_rejects_guc2vf_no_payload
[14:48:51] ==================== [PASSED] vf_relay =====================
[14:48:51] ================ pf_gt_config (9 subtests) =================
[14:48:51] [PASSED] fair_contexts_1vf
[14:48:51] [PASSED] fair_doorbells_1vf
[14:48:51] [PASSED] fair_ggtt_1vf
[14:48:51] ====================== fair_vram_1vf ======================
[14:48:51] [PASSED] 3.50 GiB
[14:48:51] [PASSED] 11.5 GiB
[14:48:51] [PASSED] 15.5 GiB
[14:48:51] [PASSED] 31.5 GiB
[14:48:51] [PASSED] 63.5 GiB
[14:48:51] [PASSED] 1.91 GiB
[14:48:51] ================== [PASSED] fair_vram_1vf ==================
[14:48:51] ================ fair_vram_1vf_admin_only =================
[14:48:51] [PASSED] 3.50 GiB
[14:48:51] [PASSED] 11.5 GiB
[14:48:51] [PASSED] 15.5 GiB
[14:48:51] [PASSED] 31.5 GiB
[14:48:51] [PASSED] 63.5 GiB
[14:48:51] [PASSED] 1.91 GiB
[14:48:51] ============ [PASSED] fair_vram_1vf_admin_only =============
[14:48:51] ====================== fair_contexts ======================
[14:48:51] [PASSED] 1 VF
[14:48:51] [PASSED] 2 VFs
[14:48:51] [PASSED] 3 VFs
[14:48:51] [PASSED] 4 VFs
[14:48:51] [PASSED] 5 VFs
[14:48:51] [PASSED] 6 VFs
[14:48:51] [PASSED] 7 VFs
[14:48:51] [PASSED] 8 VFs
[14:48:51] [PASSED] 9 VFs
[14:48:51] [PASSED] 10 VFs
[14:48:51] [PASSED] 11 VFs
[14:48:51] [PASSED] 12 VFs
[14:48:51] [PASSED] 13 VFs
[14:48:51] [PASSED] 14 VFs
[14:48:51] [PASSED] 15 VFs
[14:48:51] [PASSED] 16 VFs
[14:48:51] [PASSED] 17 VFs
[14:48:51] [PASSED] 18 VFs
[14:48:51] [PASSED] 19 VFs
[14:48:51] [PASSED] 20 VFs
[14:48:51] [PASSED] 21 VFs
[14:48:51] [PASSED] 22 VFs
[14:48:51] [PASSED] 23 VFs
[14:48:51] [PASSED] 24 VFs
[14:48:51] [PASSED] 25 VFs
[14:48:51] [PASSED] 26 VFs
[14:48:51] [PASSED] 27 VFs
[14:48:51] [PASSED] 28 VFs
[14:48:51] [PASSED] 29 VFs
[14:48:51] [PASSED] 30 VFs
[14:48:51] [PASSED] 31 VFs
[14:48:51] [PASSED] 32 VFs
[14:48:51] [PASSED] 33 VFs
[14:48:51] [PASSED] 34 VFs
[14:48:51] [PASSED] 35 VFs
[14:48:51] [PASSED] 36 VFs
[14:48:51] [PASSED] 37 VFs
[14:48:51] [PASSED] 38 VFs
[14:48:51] [PASSED] 39 VFs
[14:48:51] [PASSED] 40 VFs
[14:48:51] [PASSED] 41 VFs
[14:48:51] [PASSED] 42 VFs
[14:48:51] [PASSED] 43 VFs
[14:48:51] [PASSED] 44 VFs
[14:48:51] [PASSED] 45 VFs
[14:48:51] [PASSED] 46 VFs
[14:48:51] [PASSED] 47 VFs
[14:48:51] [PASSED] 48 VFs
[14:48:51] [PASSED] 49 VFs
[14:48:51] [PASSED] 50 VFs
[14:48:51] [PASSED] 51 VFs
[14:48:51] [PASSED] 52 VFs
[14:48:51] [PASSED] 53 VFs
[14:48:51] [PASSED] 54 VFs
[14:48:51] [PASSED] 55 VFs
[14:48:51] [PASSED] 56 VFs
[14:48:51] [PASSED] 57 VFs
[14:48:51] [PASSED] 58 VFs
[14:48:51] [PASSED] 59 VFs
[14:48:51] [PASSED] 60 VFs
[14:48:51] [PASSED] 61 VFs
[14:48:51] [PASSED] 62 VFs
[14:48:51] [PASSED] 63 VFs
[14:48:51] ================== [PASSED] fair_contexts ==================
[14:48:51] ===================== fair_doorbells ======================
[14:48:51] [PASSED] 1 VF
[14:48:51] [PASSED] 2 VFs
[14:48:51] [PASSED] 3 VFs
[14:48:51] [PASSED] 4 VFs
[14:48:51] [PASSED] 5 VFs
[14:48:51] [PASSED] 6 VFs
[14:48:51] [PASSED] 7 VFs
[14:48:51] [PASSED] 8 VFs
[14:48:51] [PASSED] 9 VFs
[14:48:51] [PASSED] 10 VFs
[14:48:51] [PASSED] 11 VFs
[14:48:51] [PASSED] 12 VFs
[14:48:51] [PASSED] 13 VFs
[14:48:51] [PASSED] 14 VFs
[14:48:51] [PASSED] 15 VFs
[14:48:51] [PASSED] 16 VFs
[14:48:51] [PASSED] 17 VFs
[14:48:51] [PASSED] 18 VFs
[14:48:51] [PASSED] 19 VFs
[14:48:51] [PASSED] 20 VFs
[14:48:51] [PASSED] 21 VFs
[14:48:51] [PASSED] 22 VFs
[14:48:51] [PASSED] 23 VFs
[14:48:51] [PASSED] 24 VFs
[14:48:51] [PASSED] 25 VFs
[14:48:51] [PASSED] 26 VFs
[14:48:51] [PASSED] 27 VFs
[14:48:51] [PASSED] 28 VFs
[14:48:51] [PASSED] 29 VFs
[14:48:51] [PASSED] 30 VFs
[14:48:51] [PASSED] 31 VFs
[14:48:51] [PASSED] 32 VFs
[14:48:51] [PASSED] 33 VFs
[14:48:51] [PASSED] 34 VFs
[14:48:51] [PASSED] 35 VFs
[14:48:51] [PASSED] 36 VFs
[14:48:51] [PASSED] 37 VFs
[14:48:51] [PASSED] 38 VFs
[14:48:51] [PASSED] 39 VFs
[14:48:51] [PASSED] 40 VFs
[14:48:51] [PASSED] 41 VFs
[14:48:51] [PASSED] 42 VFs
[14:48:51] [PASSED] 43 VFs
[14:48:51] [PASSED] 44 VFs
[14:48:51] [PASSED] 45 VFs
[14:48:51] [PASSED] 46 VFs
[14:48:51] [PASSED] 47 VFs
[14:48:51] [PASSED] 48 VFs
[14:48:51] [PASSED] 49 VFs
[14:48:51] [PASSED] 50 VFs
[14:48:51] [PASSED] 51 VFs
[14:48:51] [PASSED] 52 VFs
[14:48:51] [PASSED] 53 VFs
[14:48:51] [PASSED] 54 VFs
[14:48:51] [PASSED] 55 VFs
[14:48:51] [PASSED] 56 VFs
[14:48:51] [PASSED] 57 VFs
[14:48:51] [PASSED] 58 VFs
[14:48:51] [PASSED] 59 VFs
[14:48:51] [PASSED] 60 VFs
[14:48:51] [PASSED] 61 VFs
[14:48:51] [PASSED] 62 VFs
[14:48:51] [PASSED] 63 VFs
[14:48:51] ================= [PASSED] fair_doorbells ==================
[14:48:51] ======================== fair_ggtt ========================
[14:48:51] [PASSED] 1 VF
[14:48:51] [PASSED] 2 VFs
[14:48:51] [PASSED] 3 VFs
[14:48:51] [PASSED] 4 VFs
[14:48:51] [PASSED] 5 VFs
[14:48:51] [PASSED] 6 VFs
[14:48:51] [PASSED] 7 VFs
[14:48:51] [PASSED] 8 VFs
[14:48:51] [PASSED] 9 VFs
[14:48:51] [PASSED] 10 VFs
[14:48:51] [PASSED] 11 VFs
[14:48:51] [PASSED] 12 VFs
[14:48:51] [PASSED] 13 VFs
[14:48:51] [PASSED] 14 VFs
[14:48:51] [PASSED] 15 VFs
[14:48:51] [PASSED] 16 VFs
[14:48:51] [PASSED] 17 VFs
[14:48:51] [PASSED] 18 VFs
[14:48:51] [PASSED] 19 VFs
[14:48:51] [PASSED] 20 VFs
[14:48:51] [PASSED] 21 VFs
[14:48:51] [PASSED] 22 VFs
[14:48:51] [PASSED] 23 VFs
[14:48:51] [PASSED] 24 VFs
[14:48:51] [PASSED] 25 VFs
[14:48:51] [PASSED] 26 VFs
[14:48:51] [PASSED] 27 VFs
[14:48:51] [PASSED] 28 VFs
[14:48:51] [PASSED] 29 VFs
[14:48:51] [PASSED] 30 VFs
[14:48:51] [PASSED] 31 VFs
[14:48:51] [PASSED] 32 VFs
[14:48:51] [PASSED] 33 VFs
[14:48:51] [PASSED] 34 VFs
[14:48:51] [PASSED] 35 VFs
[14:48:51] [PASSED] 36 VFs
[14:48:51] [PASSED] 37 VFs
[14:48:51] [PASSED] 38 VFs
[14:48:51] [PASSED] 39 VFs
[14:48:51] [PASSED] 40 VFs
[14:48:51] [PASSED] 41 VFs
[14:48:51] [PASSED] 42 VFs
[14:48:51] [PASSED] 43 VFs
[14:48:51] [PASSED] 44 VFs
[14:48:51] [PASSED] 45 VFs
[14:48:51] [PASSED] 46 VFs
[14:48:51] [PASSED] 47 VFs
[14:48:51] [PASSED] 48 VFs
[14:48:51] [PASSED] 49 VFs
[14:48:51] [PASSED] 50 VFs
[14:48:51] [PASSED] 51 VFs
[14:48:51] [PASSED] 52 VFs
[14:48:51] [PASSED] 53 VFs
[14:48:51] [PASSED] 54 VFs
[14:48:51] [PASSED] 55 VFs
[14:48:51] [PASSED] 56 VFs
[14:48:51] [PASSED] 57 VFs
[14:48:51] [PASSED] 58 VFs
[14:48:51] [PASSED] 59 VFs
[14:48:51] [PASSED] 60 VFs
[14:48:51] [PASSED] 61 VFs
[14:48:51] [PASSED] 62 VFs
[14:48:51] [PASSED] 63 VFs
[14:48:51] ==================== [PASSED] fair_ggtt ====================
[14:48:51] ======================== fair_vram ========================
[14:48:51] [PASSED] 1 VF
[14:48:51] [PASSED] 2 VFs
[14:48:51] [PASSED] 3 VFs
[14:48:51] [PASSED] 4 VFs
[14:48:51] [PASSED] 5 VFs
[14:48:51] [PASSED] 6 VFs
[14:48:51] [PASSED] 7 VFs
[14:48:51] [PASSED] 8 VFs
[14:48:51] [PASSED] 9 VFs
[14:48:51] [PASSED] 10 VFs
[14:48:51] [PASSED] 11 VFs
[14:48:51] [PASSED] 12 VFs
[14:48:51] [PASSED] 13 VFs
[14:48:51] [PASSED] 14 VFs
[14:48:51] [PASSED] 15 VFs
[14:48:51] [PASSED] 16 VFs
[14:48:51] [PASSED] 17 VFs
[14:48:51] [PASSED] 18 VFs
[14:48:51] [PASSED] 19 VFs
[14:48:51] [PASSED] 20 VFs
[14:48:51] [PASSED] 21 VFs
[14:48:51] [PASSED] 22 VFs
[14:48:51] [PASSED] 23 VFs
[14:48:51] [PASSED] 24 VFs
[14:48:51] [PASSED] 25 VFs
[14:48:51] [PASSED] 26 VFs
[14:48:51] [PASSED] 27 VFs
[14:48:51] [PASSED] 28 VFs
[14:48:51] [PASSED] 29 VFs
[14:48:51] [PASSED] 30 VFs
[14:48:51] [PASSED] 31 VFs
[14:48:51] [PASSED] 32 VFs
[14:48:51] [PASSED] 33 VFs
[14:48:51] [PASSED] 34 VFs
[14:48:51] [PASSED] 35 VFs
[14:48:51] [PASSED] 36 VFs
[14:48:51] [PASSED] 37 VFs
[14:48:51] [PASSED] 38 VFs
[14:48:51] [PASSED] 39 VFs
[14:48:51] [PASSED] 40 VFs
[14:48:51] [PASSED] 41 VFs
[14:48:51] [PASSED] 42 VFs
[14:48:51] [PASSED] 43 VFs
[14:48:51] [PASSED] 44 VFs
[14:48:51] [PASSED] 45 VFs
[14:48:51] [PASSED] 46 VFs
[14:48:51] [PASSED] 47 VFs
[14:48:51] [PASSED] 48 VFs
[14:48:51] [PASSED] 49 VFs
[14:48:51] [PASSED] 50 VFs
[14:48:51] [PASSED] 51 VFs
[14:48:51] [PASSED] 52 VFs
[14:48:51] [PASSED] 53 VFs
[14:48:51] [PASSED] 54 VFs
[14:48:51] [PASSED] 55 VFs
[14:48:51] [PASSED] 56 VFs
[14:48:51] [PASSED] 57 VFs
[14:48:51] [PASSED] 58 VFs
[14:48:51] [PASSED] 59 VFs
[14:48:51] [PASSED] 60 VFs
[14:48:51] [PASSED] 61 VFs
[14:48:51] [PASSED] 62 VFs
[14:48:51] [PASSED] 63 VFs
[14:48:51] ==================== [PASSED] fair_vram ====================
[14:48:51] ================== [PASSED] pf_gt_config ===================
[14:48:51] ===================== lmtt (1 subtest) =====================
[14:48:51] ======================== test_ops =========================
[14:48:51] [PASSED] 2-level
[14:48:51] [PASSED] multi-level
[14:48:51] ==================== [PASSED] test_ops =====================
[14:48:51] ====================== [PASSED] lmtt =======================
[14:48:51] ================= pf_service (11 subtests) =================
[14:48:51] [PASSED] pf_negotiate_any
[14:48:51] [PASSED] pf_negotiate_base_match
[14:48:51] [PASSED] pf_negotiate_base_newer
[14:48:51] [PASSED] pf_negotiate_base_next
[14:48:51] [SKIPPED] pf_negotiate_base_older
[14:48:51] [PASSED] pf_negotiate_base_prev
[14:48:51] [PASSED] pf_negotiate_latest_match
[14:48:51] [PASSED] pf_negotiate_latest_newer
[14:48:51] [PASSED] pf_negotiate_latest_next
[14:48:51] [SKIPPED] pf_negotiate_latest_older
[14:48:51] [SKIPPED] pf_negotiate_latest_prev
[14:48:51] =================== [PASSED] pf_service ====================
[14:48:51] ================= xe_guc_g2g (2 subtests) ==================
[14:48:51] ============== xe_live_guc_g2g_kunit_default ==============
[14:48:51] ========= [SKIPPED] xe_live_guc_g2g_kunit_default ==========
[14:48:51] ============== xe_live_guc_g2g_kunit_allmem ===============
[14:48:51] ========== [SKIPPED] xe_live_guc_g2g_kunit_allmem ==========
[14:48:51] =================== [SKIPPED] xe_guc_g2g ===================
[14:48:51] =================== xe_mocs (2 subtests) ===================
[14:48:51] ================ xe_live_mocs_kernel_kunit ================
[14:48:51] =========== [SKIPPED] xe_live_mocs_kernel_kunit ============
[14:48:51] ================ xe_live_mocs_reset_kunit =================
[14:48:51] ============ [SKIPPED] xe_live_mocs_reset_kunit ============
[14:48:51] ==================== [SKIPPED] xe_mocs =====================
[14:48:51] ================= xe_migrate (2 subtests) ==================
[14:48:51] ================= xe_migrate_sanity_kunit =================
[14:48:51] ============ [SKIPPED] xe_migrate_sanity_kunit =============
[14:48:51] ================== xe_validate_ccs_kunit ==================
[14:48:51] ============= [SKIPPED] xe_validate_ccs_kunit ==============
[14:48:51] =================== [SKIPPED] xe_migrate ===================
[14:48:51] ================== xe_dma_buf (1 subtest) ==================
[14:48:51] ==================== xe_dma_buf_kunit =====================
[14:48:51] ================ [SKIPPED] xe_dma_buf_kunit ================
[14:48:51] =================== [SKIPPED] xe_dma_buf ===================
[14:48:51] ================= xe_bo_shrink (1 subtest) =================
[14:48:51] =================== xe_bo_shrink_kunit ====================
[14:48:51] =============== [SKIPPED] xe_bo_shrink_kunit ===============
[14:48:51] ================== [SKIPPED] xe_bo_shrink ==================
[14:48:51] ==================== xe_bo (2 subtests) ====================
[14:48:51] ================== xe_ccs_migrate_kunit ===================
[14:48:51] ============== [SKIPPED] xe_ccs_migrate_kunit ==============
[14:48:51] ==================== xe_bo_evict_kunit ====================
[14:48:51] =============== [SKIPPED] xe_bo_evict_kunit ================
[14:48:51] ===================== [SKIPPED] xe_bo ======================
[14:48:51] ==================== args (13 subtests) ====================
[14:48:51] [PASSED] count_args_test
[14:48:51] [PASSED] call_args_example
[14:48:51] [PASSED] call_args_test
[14:48:51] [PASSED] drop_first_arg_example
[14:48:51] [PASSED] drop_first_arg_test
[14:48:51] [PASSED] first_arg_example
[14:48:51] [PASSED] first_arg_test
[14:48:51] [PASSED] last_arg_example
[14:48:51] [PASSED] last_arg_test
[14:48:51] [PASSED] pick_arg_example
[14:48:51] [PASSED] if_args_example
[14:48:51] [PASSED] if_args_test
[14:48:51] [PASSED] sep_comma_example
[14:48:51] ====================== [PASSED] args =======================
[14:48:51] =================== xe_pci (3 subtests) ====================
[14:48:51] ==================== check_graphics_ip ====================
[14:48:51] [PASSED] 12.00 Xe_LP
[14:48:51] [PASSED] 12.10 Xe_LP+
[14:48:51] [PASSED] 12.55 Xe_HPG
[14:48:51] [PASSED] 12.60 Xe_HPC
[14:48:51] [PASSED] 12.70 Xe_LPG
[14:48:51] [PASSED] 12.71 Xe_LPG
[14:48:51] [PASSED] 12.74 Xe_LPG+
[14:48:51] [PASSED] 20.01 Xe2_HPG
[14:48:51] [PASSED] 20.02 Xe2_HPG
[14:48:51] [PASSED] 20.04 Xe2_LPG
[14:48:51] [PASSED] 30.00 Xe3_LPG
[14:48:51] [PASSED] 30.01 Xe3_LPG
[14:48:51] [PASSED] 30.03 Xe3_LPG
[14:48:51] [PASSED] 30.04 Xe3_LPG
[14:48:51] [PASSED] 30.05 Xe3_LPG
[14:48:51] [PASSED] 35.10 Xe3p_LPG
[14:48:51] [PASSED] 35.11 Xe3p_XPC
[14:48:51] ================ [PASSED] check_graphics_ip ================
[14:48:51] ===================== check_media_ip ======================
[14:48:51] [PASSED] 12.00 Xe_M
[14:48:51] [PASSED] 12.55 Xe_HPM
[14:48:51] [PASSED] 13.00 Xe_LPM+
[14:48:51] [PASSED] 13.01 Xe2_HPM
[14:48:51] [PASSED] 20.00 Xe2_LPM
[14:48:51] [PASSED] 30.00 Xe3_LPM
[14:48:51] [PASSED] 30.02 Xe3_LPM
[14:48:51] [PASSED] 35.00 Xe3p_LPM
[14:48:51] [PASSED] 35.03 Xe3p_HPM
[14:48:51] ================= [PASSED] check_media_ip ==================
[14:48:51] =================== check_platform_desc ===================
[14:48:51] [PASSED] 0x9A60 (TIGERLAKE)
[14:48:51] [PASSED] 0x9A68 (TIGERLAKE)
[14:48:51] [PASSED] 0x9A70 (TIGERLAKE)
[14:48:51] [PASSED] 0x9A40 (TIGERLAKE)
[14:48:51] [PASSED] 0x9A49 (TIGERLAKE)
[14:48:51] [PASSED] 0x9A59 (TIGERLAKE)
[14:48:51] [PASSED] 0x9A78 (TIGERLAKE)
[14:48:51] [PASSED] 0x9AC0 (TIGERLAKE)
[14:48:51] [PASSED] 0x9AC9 (TIGERLAKE)
[14:48:51] [PASSED] 0x9AD9 (TIGERLAKE)
[14:48:51] [PASSED] 0x9AF8 (TIGERLAKE)
[14:48:51] [PASSED] 0x4C80 (ROCKETLAKE)
[14:48:51] [PASSED] 0x4C8A (ROCKETLAKE)
[14:48:51] [PASSED] 0x4C8B (ROCKETLAKE)
[14:48:51] [PASSED] 0x4C8C (ROCKETLAKE)
[14:48:51] [PASSED] 0x4C90 (ROCKETLAKE)
[14:48:51] [PASSED] 0x4C9A (ROCKETLAKE)
[14:48:51] [PASSED] 0x4680 (ALDERLAKE_S)
[14:48:51] [PASSED] 0x4682 (ALDERLAKE_S)
[14:48:51] [PASSED] 0x4688 (ALDERLAKE_S)
[14:48:51] [PASSED] 0x468A (ALDERLAKE_S)
[14:48:51] [PASSED] 0x468B (ALDERLAKE_S)
[14:48:51] [PASSED] 0x4690 (ALDERLAKE_S)
[14:48:51] [PASSED] 0x4692 (ALDERLAKE_S)
[14:48:51] [PASSED] 0x4693 (ALDERLAKE_S)
[14:48:51] [PASSED] 0x46A0 (ALDERLAKE_P)
[14:48:51] [PASSED] 0x46A1 (ALDERLAKE_P)
[14:48:51] [PASSED] 0x46A2 (ALDERLAKE_P)
[14:48:51] [PASSED] 0x46A3 (ALDERLAKE_P)
[14:48:51] [PASSED] 0x46A6 (ALDERLAKE_P)
[14:48:51] [PASSED] 0x46A8 (ALDERLAKE_P)
[14:48:51] [PASSED] 0x46AA (ALDERLAKE_P)
[14:48:51] [PASSED] 0x462A (ALDERLAKE_P)
[14:48:51] [PASSED] 0x4626 (ALDERLAKE_P)
[14:48:51] [PASSED] 0x4628 (ALDERLAKE_P)
[14:48:51] [PASSED] 0x46B0 (ALDERLAKE_P)
[14:48:51] [PASSED] 0x46B1 (ALDERLAKE_P)
[14:48:51] [PASSED] 0x46B2 (ALDERLAKE_P)
[14:48:51] [PASSED] 0x46B3 (ALDERLAKE_P)
[14:48:51] [PASSED] 0x46C0 (ALDERLAKE_P)
[14:48:51] [PASSED] 0x46C1 (ALDERLAKE_P)
[14:48:51] [PASSED] 0x46C2 (ALDERLAKE_P)
[14:48:51] [PASSED] 0x46C3 (ALDERLAKE_P)
[14:48:51] [PASSED] 0x46D0 (ALDERLAKE_N)
[14:48:51] [PASSED] 0x46D1 (ALDERLAKE_N)
[14:48:51] [PASSED] 0x46D2 (ALDERLAKE_N)
[14:48:51] [PASSED] 0x46D3 (ALDERLAKE_N)
[14:48:51] [PASSED] 0x46D4 (ALDERLAKE_N)
[14:48:51] [PASSED] 0xA721 (ALDERLAKE_P)
[14:48:51] [PASSED] 0xA7A1 (ALDERLAKE_P)
[14:48:51] [PASSED] 0xA7A9 (ALDERLAKE_P)
[14:48:51] [PASSED] 0xA7AC (ALDERLAKE_P)
[14:48:51] [PASSED] 0xA7AD (ALDERLAKE_P)
[14:48:51] [PASSED] 0xA720 (ALDERLAKE_P)
[14:48:51] [PASSED] 0xA7A0 (ALDERLAKE_P)
[14:48:51] [PASSED] 0xA7A8 (ALDERLAKE_P)
[14:48:51] [PASSED] 0xA7AA (ALDERLAKE_P)
[14:48:51] [PASSED] 0xA7AB (ALDERLAKE_P)
[14:48:51] [PASSED] 0xA780 (ALDERLAKE_S)
[14:48:51] [PASSED] 0xA781 (ALDERLAKE_S)
[14:48:51] [PASSED] 0xA782 (ALDERLAKE_S)
[14:48:51] [PASSED] 0xA783 (ALDERLAKE_S)
[14:48:51] [PASSED] 0xA788 (ALDERLAKE_S)
[14:48:51] [PASSED] 0xA789 (ALDERLAKE_S)
[14:48:51] [PASSED] 0xA78A (ALDERLAKE_S)
[14:48:51] [PASSED] 0xA78B (ALDERLAKE_S)
[14:48:51] [PASSED] 0x4905 (DG1)
[14:48:51] [PASSED] 0x4906 (DG1)
[14:48:51] [PASSED] 0x4907 (DG1)
[14:48:51] [PASSED] 0x4908 (DG1)
[14:48:51] [PASSED] 0x4909 (DG1)
[14:48:51] [PASSED] 0x56C0 (DG2)
[14:48:51] [PASSED] 0x56C2 (DG2)
[14:48:51] [PASSED] 0x56C1 (DG2)
[14:48:51] [PASSED] 0x7D51 (METEORLAKE)
[14:48:51] [PASSED] 0x7DD1 (METEORLAKE)
[14:48:51] [PASSED] 0x7D41 (METEORLAKE)
[14:48:51] [PASSED] 0x7D67 (METEORLAKE)
[14:48:51] [PASSED] 0xB640 (METEORLAKE)
[14:48:51] [PASSED] 0x56A0 (DG2)
[14:48:51] [PASSED] 0x56A1 (DG2)
[14:48:51] [PASSED] 0x56A2 (DG2)
[14:48:51] [PASSED] 0x56BE (DG2)
[14:48:51] [PASSED] 0x56BF (DG2)
[14:48:51] [PASSED] 0x5690 (DG2)
[14:48:51] [PASSED] 0x5691 (DG2)
[14:48:51] [PASSED] 0x5692 (DG2)
[14:48:51] [PASSED] 0x56A5 (DG2)
[14:48:51] [PASSED] 0x56A6 (DG2)
[14:48:51] [PASSED] 0x56B0 (DG2)
[14:48:51] [PASSED] 0x56B1 (DG2)
[14:48:51] [PASSED] 0x56BA (DG2)
[14:48:51] [PASSED] 0x56BB (DG2)
[14:48:51] [PASSED] 0x56BC (DG2)
[14:48:51] [PASSED] 0x56BD (DG2)
[14:48:51] [PASSED] 0x5693 (DG2)
[14:48:51] [PASSED] 0x5694 (DG2)
[14:48:51] [PASSED] 0x5695 (DG2)
[14:48:51] [PASSED] 0x56A3 (DG2)
[14:48:51] [PASSED] 0x56A4 (DG2)
[14:48:51] [PASSED] 0x56B2 (DG2)
[14:48:51] [PASSED] 0x56B3 (DG2)
[14:48:51] [PASSED] 0x5696 (DG2)
[14:48:51] [PASSED] 0x5697 (DG2)
[14:48:51] [PASSED] 0xB69 (PVC)
[14:48:51] [PASSED] 0xB6E (PVC)
[14:48:51] [PASSED] 0xBD4 (PVC)
[14:48:51] [PASSED] 0xBD5 (PVC)
[14:48:51] [PASSED] 0xBD6 (PVC)
[14:48:51] [PASSED] 0xBD7 (PVC)
[14:48:51] [PASSED] 0xBD8 (PVC)
[14:48:51] [PASSED] 0xBD9 (PVC)
[14:48:51] [PASSED] 0xBDA (PVC)
[14:48:51] [PASSED] 0xBDB (PVC)
[14:48:51] [PASSED] 0xBE0 (PVC)
[14:48:51] [PASSED] 0xBE1 (PVC)
[14:48:51] [PASSED] 0xBE5 (PVC)
[14:48:51] [PASSED] 0x7D40 (METEORLAKE)
[14:48:51] [PASSED] 0x7D45 (METEORLAKE)
[14:48:51] [PASSED] 0x7D55 (METEORLAKE)
[14:48:51] [PASSED] 0x7D60 (METEORLAKE)
[14:48:51] [PASSED] 0x7DD5 (METEORLAKE)
[14:48:51] [PASSED] 0x6420 (LUNARLAKE)
[14:48:51] [PASSED] 0x64A0 (LUNARLAKE)
[14:48:51] [PASSED] 0x64B0 (LUNARLAKE)
[14:48:51] [PASSED] 0xE202 (BATTLEMAGE)
[14:48:51] [PASSED] 0xE209 (BATTLEMAGE)
[14:48:51] [PASSED] 0xE20B (BATTLEMAGE)
[14:48:51] [PASSED] 0xE20C (BATTLEMAGE)
[14:48:51] [PASSED] 0xE20D (BATTLEMAGE)
[14:48:51] [PASSED] 0xE210 (BATTLEMAGE)
[14:48:51] [PASSED] 0xE211 (BATTLEMAGE)
[14:48:51] [PASSED] 0xE212 (BATTLEMAGE)
[14:48:51] [PASSED] 0xE216 (BATTLEMAGE)
[14:48:51] [PASSED] 0xE220 (BATTLEMAGE)
[14:48:51] [PASSED] 0xE221 (BATTLEMAGE)
[14:48:51] [PASSED] 0xE222 (BATTLEMAGE)
[14:48:51] [PASSED] 0xE223 (BATTLEMAGE)
[14:48:51] [PASSED] 0xB080 (PANTHERLAKE)
[14:48:51] [PASSED] 0xB081 (PANTHERLAKE)
[14:48:51] [PASSED] 0xB082 (PANTHERLAKE)
[14:48:51] [PASSED] 0xB083 (PANTHERLAKE)
[14:48:51] [PASSED] 0xB084 (PANTHERLAKE)
[14:48:51] [PASSED] 0xB085 (PANTHERLAKE)
[14:48:51] [PASSED] 0xB086 (PANTHERLAKE)
[14:48:51] [PASSED] 0xB087 (PANTHERLAKE)
[14:48:51] [PASSED] 0xB08F (PANTHERLAKE)
[14:48:51] [PASSED] 0xB090 (PANTHERLAKE)
[14:48:51] [PASSED] 0xB0A0 (PANTHERLAKE)
[14:48:51] [PASSED] 0xB0B0 (PANTHERLAKE)
[14:48:51] [PASSED] 0xFD80 (PANTHERLAKE)
[14:48:51] [PASSED] 0xFD81 (PANTHERLAKE)
[14:48:51] [PASSED] 0xD740 (NOVALAKE_S)
[14:48:51] [PASSED] 0xD741 (NOVALAKE_S)
[14:48:51] [PASSED] 0xD742 (NOVALAKE_S)
[14:48:51] [PASSED] 0xD743 (NOVALAKE_S)
[14:48:51] [PASSED] 0xD744 (NOVALAKE_S)
[14:48:51] [PASSED] 0xD745 (NOVALAKE_S)
[14:48:51] [PASSED] 0x674C (CRESCENTISLAND)
[14:48:51] [PASSED] 0xD750 (NOVALAKE_P)
[14:48:51] [PASSED] 0xD751 (NOVALAKE_P)
[14:48:51] [PASSED] 0xD752 (NOVALAKE_P)
[14:48:51] [PASSED] 0xD753 (NOVALAKE_P)
[14:48:51] [PASSED] 0xD754 (NOVALAKE_P)
[14:48:51] [PASSED] 0xD755 (NOVALAKE_P)
[14:48:51] [PASSED] 0xD756 (NOVALAKE_P)
[14:48:51] [PASSED] 0xD757 (NOVALAKE_P)
[14:48:51] [PASSED] 0xD75F (NOVALAKE_P)
[14:48:51] =============== [PASSED] check_platform_desc ===============
[14:48:51] ===================== [PASSED] xe_pci ======================
[14:48:51] =================== xe_rtp (2 subtests) ====================
[14:48:51] =============== xe_rtp_process_to_sr_tests ================
[14:48:51] [PASSED] coalesce-same-reg
[14:48:51] [PASSED] no-match-no-add
[14:48:51] [PASSED] match-or
[14:48:51] [PASSED] match-or-xfail
[14:48:51] [PASSED] no-match-no-add-multiple-rules
[14:48:51] [PASSED] two-regs-two-entries
[14:48:51] [PASSED] clr-one-set-other
[14:48:51] [PASSED] set-field
[14:48:51] [PASSED] conflict-duplicate
[14:48:51] [PASSED] conflict-not-disjoint
[14:48:51] [PASSED] conflict-reg-type
[14:48:51] =========== [PASSED] xe_rtp_process_to_sr_tests ============
[14:48:51] ================== xe_rtp_process_tests ===================
[14:48:51] [PASSED] active1
[14:48:51] [PASSED] active2
[14:48:51] [PASSED] active-inactive
[14:48:51] [PASSED] inactive-active
[14:48:51] [PASSED] inactive-1st_or_active-inactive
[14:48:51] [PASSED] inactive-2nd_or_active-inactive
[14:48:51] [PASSED] inactive-last_or_active-inactive
[14:48:51] [PASSED] inactive-no_or_active-inactive
[14:48:51] ============== [PASSED] xe_rtp_process_tests ===============
[14:48:51] ===================== [PASSED] xe_rtp ======================
[14:48:51] ==================== xe_wa (1 subtest) =====================
[14:48:51] ======================== xe_wa_gt =========================
[14:48:51] [PASSED] TIGERLAKE B0
[14:48:51] [PASSED] DG1 A0
[14:48:51] [PASSED] DG1 B0
[14:48:51] [PASSED] ALDERLAKE_S A0
[14:48:51] [PASSED] ALDERLAKE_S B0
[14:48:51] [PASSED] ALDERLAKE_S C0
[14:48:51] [PASSED] ALDERLAKE_S D0
[14:48:51] [PASSED] ALDERLAKE_P A0
[14:48:51] [PASSED] ALDERLAKE_P B0
[14:48:51] [PASSED] ALDERLAKE_P C0
[14:48:51] [PASSED] ALDERLAKE_S RPLS D0
[14:48:51] [PASSED] ALDERLAKE_P RPLU E0
[14:48:51] [PASSED] DG2 G10 C0
[14:48:51] [PASSED] DG2 G11 B1
[14:48:51] [PASSED] DG2 G12 A1
[14:48:51] [PASSED] METEORLAKE 12.70(Xe_LPG) A0 13.00(Xe_LPM+) A0
[14:48:51] [PASSED] METEORLAKE 12.71(Xe_LPG) A0 13.00(Xe_LPM+) A0
[14:48:51] [PASSED] METEORLAKE 12.74(Xe_LPG+) A0 13.00(Xe_LPM+) A0
[14:48:51] [PASSED] LUNARLAKE 20.04(Xe2_LPG) A0 20.00(Xe2_LPM) A0
[14:48:51] [PASSED] LUNARLAKE 20.04(Xe2_LPG) B0 20.00(Xe2_LPM) A0
[14:48:51] [PASSED] BATTLEMAGE 20.01(Xe2_HPG) A0 13.01(Xe2_HPM) A1
[14:48:51] [PASSED] PANTHERLAKE 30.00(Xe3_LPG) A0 30.00(Xe3_LPM) A0
[14:48:51] ==================== [PASSED] xe_wa_gt =====================
[14:48:51] ====================== [PASSED] xe_wa ======================
[14:48:51] ============================================================
[14:48:51] Testing complete. Ran 597 tests: passed: 579, skipped: 18
[14:48:51] Elapsed time: 36.217s total, 4.333s configuring, 31.218s building, 0.616s running
+ /kernel/tools/testing/kunit/kunit.py run --kunitconfig /kernel/drivers/gpu/drm/tests/.kunitconfig
[14:48:52] Configuring KUnit Kernel ...
Regenerating .config ...
Populating config with:
$ make ARCH=um O=.kunit olddefconfig
[14:48:53] Building KUnit Kernel ...
Populating config with:
$ make ARCH=um O=.kunit olddefconfig
Building with:
$ make all compile_commands.json scripts_gdb ARCH=um O=.kunit --jobs=48
[14:49:17] Starting KUnit Kernel (1/1)...
[14:49:17] ============================================================
Running tests with:
$ .kunit/linux kunit.enable=1 mem=1G console=tty kunit_shutdown=halt
[14:49:17] ============ drm_test_pick_cmdline (2 subtests) ============
[14:49:17] [PASSED] drm_test_pick_cmdline_res_1920_1080_60
[14:49:17] =============== drm_test_pick_cmdline_named ===============
[14:49:17] [PASSED] NTSC
[14:49:17] [PASSED] NTSC-J
[14:49:17] [PASSED] PAL
[14:49:17] [PASSED] PAL-M
[14:49:17] =========== [PASSED] drm_test_pick_cmdline_named ===========
[14:49:17] ============== [PASSED] drm_test_pick_cmdline ==============
[14:49:17] == drm_test_atomic_get_connector_for_encoder (1 subtest) ===
[14:49:17] [PASSED] drm_test_drm_atomic_get_connector_for_encoder
[14:49:17] ==== [PASSED] drm_test_atomic_get_connector_for_encoder ====
[14:49:17] =========== drm_validate_clone_mode (2 subtests) ===========
[14:49:17] ============== drm_test_check_in_clone_mode ===============
[14:49:17] [PASSED] in_clone_mode
[14:49:17] [PASSED] not_in_clone_mode
[14:49:17] ========== [PASSED] drm_test_check_in_clone_mode ===========
[14:49:17] =============== drm_test_check_valid_clones ===============
[14:49:17] [PASSED] not_in_clone_mode
[14:49:17] [PASSED] valid_clone
[14:49:17] [PASSED] invalid_clone
[14:49:17] =========== [PASSED] drm_test_check_valid_clones ===========
[14:49:17] ============= [PASSED] drm_validate_clone_mode =============
[14:49:17] ============= drm_validate_modeset (1 subtest) =============
[14:49:17] [PASSED] drm_test_check_connector_changed_modeset
[14:49:17] ============== [PASSED] drm_validate_modeset ===============
[14:49:17] ====== drm_test_bridge_get_current_state (2 subtests) ======
[14:49:17] [PASSED] drm_test_drm_bridge_get_current_state_atomic
[14:49:17] [PASSED] drm_test_drm_bridge_get_current_state_legacy
[14:49:17] ======== [PASSED] drm_test_bridge_get_current_state ========
[14:49:17] ====== drm_test_bridge_helper_reset_crtc (3 subtests) ======
[14:49:17] [PASSED] drm_test_drm_bridge_helper_reset_crtc_atomic
[14:49:17] [PASSED] drm_test_drm_bridge_helper_reset_crtc_atomic_disabled
[14:49:17] [PASSED] drm_test_drm_bridge_helper_reset_crtc_legacy
[14:49:17] ======== [PASSED] drm_test_bridge_helper_reset_crtc ========
[14:49:17] ============== drm_bridge_alloc (2 subtests) ===============
[14:49:17] [PASSED] drm_test_drm_bridge_alloc_basic
[14:49:17] [PASSED] drm_test_drm_bridge_alloc_get_put
[14:49:17] ================ [PASSED] drm_bridge_alloc =================
[14:49:17] ============= drm_cmdline_parser (40 subtests) =============
[14:49:17] [PASSED] drm_test_cmdline_force_d_only
[14:49:17] [PASSED] drm_test_cmdline_force_D_only_dvi
[14:49:17] [PASSED] drm_test_cmdline_force_D_only_hdmi
[14:49:17] [PASSED] drm_test_cmdline_force_D_only_not_digital
[14:49:17] [PASSED] drm_test_cmdline_force_e_only
[14:49:17] [PASSED] drm_test_cmdline_res
[14:49:17] [PASSED] drm_test_cmdline_res_vesa
[14:49:17] [PASSED] drm_test_cmdline_res_vesa_rblank
[14:49:17] [PASSED] drm_test_cmdline_res_rblank
[14:49:17] [PASSED] drm_test_cmdline_res_bpp
[14:49:17] [PASSED] drm_test_cmdline_res_refresh
[14:49:17] [PASSED] drm_test_cmdline_res_bpp_refresh
[14:49:17] [PASSED] drm_test_cmdline_res_bpp_refresh_interlaced
[14:49:17] [PASSED] drm_test_cmdline_res_bpp_refresh_margins
[14:49:17] [PASSED] drm_test_cmdline_res_bpp_refresh_force_off
[14:49:17] [PASSED] drm_test_cmdline_res_bpp_refresh_force_on
[14:49:17] [PASSED] drm_test_cmdline_res_bpp_refresh_force_on_analog
[14:49:17] [PASSED] drm_test_cmdline_res_bpp_refresh_force_on_digital
[14:49:17] [PASSED] drm_test_cmdline_res_bpp_refresh_interlaced_margins_force_on
[14:49:17] [PASSED] drm_test_cmdline_res_margins_force_on
[14:49:17] [PASSED] drm_test_cmdline_res_vesa_margins
[14:49:17] [PASSED] drm_test_cmdline_name
[14:49:17] [PASSED] drm_test_cmdline_name_bpp
[14:49:17] [PASSED] drm_test_cmdline_name_option
[14:49:17] [PASSED] drm_test_cmdline_name_bpp_option
[14:49:17] [PASSED] drm_test_cmdline_rotate_0
[14:49:17] [PASSED] drm_test_cmdline_rotate_90
[14:49:17] [PASSED] drm_test_cmdline_rotate_180
[14:49:17] [PASSED] drm_test_cmdline_rotate_270
[14:49:17] [PASSED] drm_test_cmdline_hmirror
[14:49:17] [PASSED] drm_test_cmdline_vmirror
[14:49:17] [PASSED] drm_test_cmdline_margin_options
[14:49:17] [PASSED] drm_test_cmdline_multiple_options
[14:49:17] [PASSED] drm_test_cmdline_bpp_extra_and_option
[14:49:17] [PASSED] drm_test_cmdline_extra_and_option
[14:49:17] [PASSED] drm_test_cmdline_freestanding_options
[14:49:17] [PASSED] drm_test_cmdline_freestanding_force_e_and_options
[14:49:17] [PASSED] drm_test_cmdline_panel_orientation
[14:49:17] ================ drm_test_cmdline_invalid =================
[14:49:17] [PASSED] margin_only
[14:49:17] [PASSED] interlace_only
[14:49:17] [PASSED] res_missing_x
[14:49:17] [PASSED] res_missing_y
[14:49:17] [PASSED] res_bad_y
[14:49:17] [PASSED] res_missing_y_bpp
[14:49:17] [PASSED] res_bad_bpp
[14:49:17] [PASSED] res_bad_refresh
[14:49:17] [PASSED] res_bpp_refresh_force_on_off
[14:49:17] [PASSED] res_invalid_mode
[14:49:17] [PASSED] res_bpp_wrong_place_mode
[14:49:17] [PASSED] name_bpp_refresh
[14:49:17] [PASSED] name_refresh
[14:49:17] [PASSED] name_refresh_wrong_mode
[14:49:17] [PASSED] name_refresh_invalid_mode
[14:49:17] [PASSED] rotate_multiple
[14:49:17] [PASSED] rotate_invalid_val
[14:49:17] [PASSED] rotate_truncated
[14:49:17] [PASSED] invalid_option
[14:49:17] [PASSED] invalid_tv_option
[14:49:17] [PASSED] truncated_tv_option
[14:49:17] ============ [PASSED] drm_test_cmdline_invalid =============
[14:49:17] =============== drm_test_cmdline_tv_options ===============
[14:49:17] [PASSED] NTSC
[14:49:17] [PASSED] NTSC_443
[14:49:17] [PASSED] NTSC_J
[14:49:17] [PASSED] PAL
[14:49:17] [PASSED] PAL_M
[14:49:17] [PASSED] PAL_N
[14:49:17] [PASSED] SECAM
[14:49:17] [PASSED] MONO_525
[14:49:17] [PASSED] MONO_625
[14:49:17] =========== [PASSED] drm_test_cmdline_tv_options ===========
[14:49:17] =============== [PASSED] drm_cmdline_parser ================
[14:49:17] ========== drmm_connector_hdmi_init (20 subtests) ==========
[14:49:17] [PASSED] drm_test_connector_hdmi_init_valid
[14:49:17] [PASSED] drm_test_connector_hdmi_init_bpc_8
[14:49:17] [PASSED] drm_test_connector_hdmi_init_bpc_10
[14:49:17] [PASSED] drm_test_connector_hdmi_init_bpc_12
[14:49:17] [PASSED] drm_test_connector_hdmi_init_bpc_invalid
[14:49:17] [PASSED] drm_test_connector_hdmi_init_bpc_null
[14:49:17] [PASSED] drm_test_connector_hdmi_init_formats_empty
[14:49:17] [PASSED] drm_test_connector_hdmi_init_formats_no_rgb
[14:49:17] === drm_test_connector_hdmi_init_formats_yuv420_allowed ===
[14:49:17] [PASSED] supported_formats=0x9 yuv420_allowed=1
[14:49:17] [PASSED] supported_formats=0x9 yuv420_allowed=0
[14:49:17] [PASSED] supported_formats=0x5 yuv420_allowed=1
[14:49:17] [PASSED] supported_formats=0x5 yuv420_allowed=0
[14:49:17] === [PASSED] drm_test_connector_hdmi_init_formats_yuv420_allowed ===
[14:49:17] [PASSED] drm_test_connector_hdmi_init_null_ddc
[14:49:17] [PASSED] drm_test_connector_hdmi_init_null_product
[14:49:17] [PASSED] drm_test_connector_hdmi_init_null_vendor
[14:49:17] [PASSED] drm_test_connector_hdmi_init_product_length_exact
[14:49:17] [PASSED] drm_test_connector_hdmi_init_product_length_too_long
[14:49:17] [PASSED] drm_test_connector_hdmi_init_product_valid
[14:49:17] [PASSED] drm_test_connector_hdmi_init_vendor_length_exact
[14:49:17] [PASSED] drm_test_connector_hdmi_init_vendor_length_too_long
[14:49:17] [PASSED] drm_test_connector_hdmi_init_vendor_valid
[14:49:17] ========= drm_test_connector_hdmi_init_type_valid =========
[14:49:17] [PASSED] HDMI-A
[14:49:17] [PASSED] HDMI-B
[14:49:17] ===== [PASSED] drm_test_connector_hdmi_init_type_valid =====
[14:49:17] ======== drm_test_connector_hdmi_init_type_invalid ========
[14:49:17] [PASSED] Unknown
[14:49:17] [PASSED] VGA
[14:49:17] [PASSED] DVI-I
[14:49:17] [PASSED] DVI-D
[14:49:17] [PASSED] DVI-A
[14:49:17] [PASSED] Composite
[14:49:17] [PASSED] SVIDEO
[14:49:17] [PASSED] LVDS
[14:49:17] [PASSED] Component
[14:49:17] [PASSED] DIN
[14:49:17] [PASSED] DP
[14:49:17] [PASSED] TV
[14:49:17] [PASSED] eDP
[14:49:17] [PASSED] Virtual
[14:49:17] [PASSED] DSI
[14:49:17] [PASSED] DPI
[14:49:17] [PASSED] Writeback
[14:49:17] [PASSED] SPI
[14:49:17] [PASSED] USB
[14:49:17] ==== [PASSED] drm_test_connector_hdmi_init_type_invalid ====
[14:49:17] ============ [PASSED] drmm_connector_hdmi_init =============
[14:49:17] ============= drmm_connector_init (3 subtests) =============
[14:49:17] [PASSED] drm_test_drmm_connector_init
[14:49:17] [PASSED] drm_test_drmm_connector_init_null_ddc
[14:49:17] ========= drm_test_drmm_connector_init_type_valid =========
[14:49:17] [PASSED] Unknown
[14:49:17] [PASSED] VGA
[14:49:17] [PASSED] DVI-I
[14:49:17] [PASSED] DVI-D
[14:49:17] [PASSED] DVI-A
[14:49:17] [PASSED] Composite
[14:49:17] [PASSED] SVIDEO
[14:49:17] [PASSED] LVDS
[14:49:17] [PASSED] Component
[14:49:17] [PASSED] DIN
[14:49:17] [PASSED] DP
[14:49:17] [PASSED] HDMI-A
[14:49:17] [PASSED] HDMI-B
[14:49:17] [PASSED] TV
[14:49:17] [PASSED] eDP
[14:49:17] [PASSED] Virtual
[14:49:17] [PASSED] DSI
[14:49:17] [PASSED] DPI
[14:49:17] [PASSED] Writeback
[14:49:17] [PASSED] SPI
[14:49:17] [PASSED] USB
[14:49:17] ===== [PASSED] drm_test_drmm_connector_init_type_valid =====
[14:49:17] =============== [PASSED] drmm_connector_init ===============
[14:49:17] ========= drm_connector_dynamic_init (6 subtests) ==========
[14:49:17] [PASSED] drm_test_drm_connector_dynamic_init
[14:49:17] [PASSED] drm_test_drm_connector_dynamic_init_null_ddc
[14:49:17] [PASSED] drm_test_drm_connector_dynamic_init_not_added
[14:49:17] [PASSED] drm_test_drm_connector_dynamic_init_properties
[14:49:17] ===== drm_test_drm_connector_dynamic_init_type_valid ======
[14:49:17] [PASSED] Unknown
[14:49:17] [PASSED] VGA
[14:49:17] [PASSED] DVI-I
[14:49:17] [PASSED] DVI-D
[14:49:17] [PASSED] DVI-A
[14:49:17] [PASSED] Composite
[14:49:17] [PASSED] SVIDEO
[14:49:17] [PASSED] LVDS
[14:49:17] [PASSED] Component
[14:49:17] [PASSED] DIN
[14:49:17] [PASSED] DP
[14:49:17] [PASSED] HDMI-A
[14:49:17] [PASSED] HDMI-B
[14:49:17] [PASSED] TV
[14:49:17] [PASSED] eDP
[14:49:17] [PASSED] Virtual
[14:49:17] [PASSED] DSI
[14:49:17] [PASSED] DPI
[14:49:17] [PASSED] Writeback
[14:49:17] [PASSED] SPI
[14:49:17] [PASSED] USB
[14:49:17] = [PASSED] drm_test_drm_connector_dynamic_init_type_valid ==
[14:49:17] ======== drm_test_drm_connector_dynamic_init_name =========
[14:49:17] [PASSED] Unknown
[14:49:17] [PASSED] VGA
[14:49:17] [PASSED] DVI-I
[14:49:17] [PASSED] DVI-D
[14:49:17] [PASSED] DVI-A
[14:49:17] [PASSED] Composite
[14:49:17] [PASSED] SVIDEO
[14:49:17] [PASSED] LVDS
[14:49:17] [PASSED] Component
[14:49:17] [PASSED] DIN
[14:49:17] [PASSED] DP
[14:49:17] [PASSED] HDMI-A
[14:49:17] [PASSED] HDMI-B
[14:49:17] [PASSED] TV
[14:49:17] [PASSED] eDP
[14:49:17] [PASSED] Virtual
[14:49:17] [PASSED] DSI
[14:49:17] [PASSED] DPI
[14:49:17] [PASSED] Writeback
[14:49:17] [PASSED] SPI
[14:49:17] [PASSED] USB
[14:49:17] ==== [PASSED] drm_test_drm_connector_dynamic_init_name =====
[14:49:17] =========== [PASSED] drm_connector_dynamic_init ============
[14:49:17] ==== drm_connector_dynamic_register_early (4 subtests) =====
[14:49:17] [PASSED] drm_test_drm_connector_dynamic_register_early_on_list
[14:49:17] [PASSED] drm_test_drm_connector_dynamic_register_early_defer
[14:49:17] [PASSED] drm_test_drm_connector_dynamic_register_early_no_init
[14:49:17] [PASSED] drm_test_drm_connector_dynamic_register_early_no_mode_object
[14:49:17] ====== [PASSED] drm_connector_dynamic_register_early =======
[14:49:17] ======= drm_connector_dynamic_register (7 subtests) ========
[14:49:17] [PASSED] drm_test_drm_connector_dynamic_register_on_list
[14:49:17] [PASSED] drm_test_drm_connector_dynamic_register_no_defer
[14:49:17] [PASSED] drm_test_drm_connector_dynamic_register_no_init
[14:49:17] [PASSED] drm_test_drm_connector_dynamic_register_mode_object
[14:49:17] [PASSED] drm_test_drm_connector_dynamic_register_sysfs
[14:49:17] [PASSED] drm_test_drm_connector_dynamic_register_sysfs_name
[14:49:17] [PASSED] drm_test_drm_connector_dynamic_register_debugfs
[14:49:17] ========= [PASSED] drm_connector_dynamic_register ==========
[14:49:17] = drm_connector_attach_broadcast_rgb_property (2 subtests) =
[14:49:17] [PASSED] drm_test_drm_connector_attach_broadcast_rgb_property
[14:49:17] [PASSED] drm_test_drm_connector_attach_broadcast_rgb_property_hdmi_connector
[14:49:17] === [PASSED] drm_connector_attach_broadcast_rgb_property ===
[14:49:17] ========== drm_get_tv_mode_from_name (2 subtests) ==========
[14:49:17] ========== drm_test_get_tv_mode_from_name_valid ===========
[14:49:17] [PASSED] NTSC
[14:49:17] [PASSED] NTSC-443
[14:49:17] [PASSED] NTSC-J
[14:49:17] [PASSED] PAL
[14:49:17] [PASSED] PAL-M
[14:49:17] [PASSED] PAL-N
[14:49:17] [PASSED] SECAM
[14:49:17] [PASSED] Mono
[14:49:17] ====== [PASSED] drm_test_get_tv_mode_from_name_valid =======
[14:49:17] [PASSED] drm_test_get_tv_mode_from_name_truncated
[14:49:17] ============ [PASSED] drm_get_tv_mode_from_name ============
[14:49:17] = drm_test_connector_hdmi_compute_mode_clock (12 subtests) =
[14:49:17] [PASSED] drm_test_drm_hdmi_compute_mode_clock_rgb
[14:49:17] [PASSED] drm_test_drm_hdmi_compute_mode_clock_rgb_10bpc
[14:49:17] [PASSED] drm_test_drm_hdmi_compute_mode_clock_rgb_10bpc_vic_1
[14:49:17] [PASSED] drm_test_drm_hdmi_compute_mode_clock_rgb_12bpc
[14:49:17] [PASSED] drm_test_drm_hdmi_compute_mode_clock_rgb_12bpc_vic_1
[14:49:17] [PASSED] drm_test_drm_hdmi_compute_mode_clock_rgb_double
[14:49:17] = drm_test_connector_hdmi_compute_mode_clock_yuv420_valid =
[14:49:17] [PASSED] VIC 96
[14:49:17] [PASSED] VIC 97
[14:49:17] [PASSED] VIC 101
[14:49:17] [PASSED] VIC 102
[14:49:17] [PASSED] VIC 106
[14:49:17] [PASSED] VIC 107
[14:49:17] === [PASSED] drm_test_connector_hdmi_compute_mode_clock_yuv420_valid ===
[14:49:17] [PASSED] drm_test_connector_hdmi_compute_mode_clock_yuv420_10_bpc
[14:49:17] [PASSED] drm_test_connector_hdmi_compute_mode_clock_yuv420_12_bpc
[14:49:17] [PASSED] drm_test_connector_hdmi_compute_mode_clock_yuv422_8_bpc
[14:49:17] [PASSED] drm_test_connector_hdmi_compute_mode_clock_yuv422_10_bpc
[14:49:17] [PASSED] drm_test_connector_hdmi_compute_mode_clock_yuv422_12_bpc
[14:49:17] === [PASSED] drm_test_connector_hdmi_compute_mode_clock ====
[14:49:17] == drm_hdmi_connector_get_broadcast_rgb_name (2 subtests) ==
[14:49:17] === drm_test_drm_hdmi_connector_get_broadcast_rgb_name ====
[14:49:17] [PASSED] Automatic
[14:49:17] [PASSED] Full
[14:49:17] [PASSED] Limited 16:235
[14:49:17] === [PASSED] drm_test_drm_hdmi_connector_get_broadcast_rgb_name ===
[14:49:17] [PASSED] drm_test_drm_hdmi_connector_get_broadcast_rgb_name_invalid
[14:49:17] ==== [PASSED] drm_hdmi_connector_get_broadcast_rgb_name ====
[14:49:17] == drm_hdmi_connector_get_output_format_name (2 subtests) ==
[14:49:17] === drm_test_drm_hdmi_connector_get_output_format_name ====
[14:49:17] [PASSED] RGB
[14:49:17] [PASSED] YUV 4:2:0
[14:49:17] [PASSED] YUV 4:2:2
[14:49:17] [PASSED] YUV 4:4:4
[14:49:17] === [PASSED] drm_test_drm_hdmi_connector_get_output_format_name ===
[14:49:17] [PASSED] drm_test_drm_hdmi_connector_get_output_format_name_invalid
[14:49:17] ==== [PASSED] drm_hdmi_connector_get_output_format_name ====
[14:49:17] ============= drm_damage_helper (21 subtests) ==============
[14:49:17] [PASSED] drm_test_damage_iter_no_damage
[14:49:17] [PASSED] drm_test_damage_iter_no_damage_fractional_src
[14:49:17] [PASSED] drm_test_damage_iter_no_damage_src_moved
[14:49:17] [PASSED] drm_test_damage_iter_no_damage_fractional_src_moved
[14:49:17] [PASSED] drm_test_damage_iter_no_damage_not_visible
[14:49:17] [PASSED] drm_test_damage_iter_no_damage_no_crtc
[14:49:17] [PASSED] drm_test_damage_iter_no_damage_no_fb
[14:49:17] [PASSED] drm_test_damage_iter_simple_damage
[14:49:17] [PASSED] drm_test_damage_iter_single_damage
[14:49:17] [PASSED] drm_test_damage_iter_single_damage_intersect_src
[14:49:17] [PASSED] drm_test_damage_iter_single_damage_outside_src
[14:49:17] [PASSED] drm_test_damage_iter_single_damage_fractional_src
[14:49:17] [PASSED] drm_test_damage_iter_single_damage_intersect_fractional_src
[14:49:17] [PASSED] drm_test_damage_iter_single_damage_outside_fractional_src
[14:49:17] [PASSED] drm_test_damage_iter_single_damage_src_moved
[14:49:17] [PASSED] drm_test_damage_iter_single_damage_fractional_src_moved
[14:49:17] [PASSED] drm_test_damage_iter_damage
[14:49:17] [PASSED] drm_test_damage_iter_damage_one_intersect
[14:49:17] [PASSED] drm_test_damage_iter_damage_one_outside
[14:49:17] [PASSED] drm_test_damage_iter_damage_src_moved
[14:49:17] [PASSED] drm_test_damage_iter_damage_not_visible
[14:49:17] ================ [PASSED] drm_damage_helper ================
[14:49:17] ============== drm_dp_mst_helper (3 subtests) ==============
[14:49:17] ============== drm_test_dp_mst_calc_pbn_mode ==============
[14:49:17] [PASSED] Clock 154000 BPP 30 DSC disabled
[14:49:17] [PASSED] Clock 234000 BPP 30 DSC disabled
[14:49:17] [PASSED] Clock 297000 BPP 24 DSC disabled
[14:49:17] [PASSED] Clock 332880 BPP 24 DSC enabled
[14:49:17] [PASSED] Clock 324540 BPP 24 DSC enabled
[14:49:17] ========== [PASSED] drm_test_dp_mst_calc_pbn_mode ==========
[14:49:17] ============== drm_test_dp_mst_calc_pbn_div ===============
[14:49:17] [PASSED] Link rate 2000000 lane count 4
[14:49:17] [PASSED] Link rate 2000000 lane count 2
[14:49:17] [PASSED] Link rate 2000000 lane count 1
[14:49:17] [PASSED] Link rate 1350000 lane count 4
[14:49:17] [PASSED] Link rate 1350000 lane count 2
[14:49:17] [PASSED] Link rate 1350000 lane count 1
[14:49:17] [PASSED] Link rate 1000000 lane count 4
[14:49:17] [PASSED] Link rate 1000000 lane count 2
[14:49:17] [PASSED] Link rate 1000000 lane count 1
[14:49:17] [PASSED] Link rate 810000 lane count 4
[14:49:17] [PASSED] Link rate 810000 lane count 2
[14:49:17] [PASSED] Link rate 810000 lane count 1
[14:49:17] [PASSED] Link rate 540000 lane count 4
[14:49:17] [PASSED] Link rate 540000 lane count 2
[14:49:17] [PASSED] Link rate 540000 lane count 1
[14:49:17] [PASSED] Link rate 270000 lane count 4
[14:49:17] [PASSED] Link rate 270000 lane count 2
[14:49:17] [PASSED] Link rate 270000 lane count 1
[14:49:17] [PASSED] Link rate 162000 lane count 4
[14:49:17] [PASSED] Link rate 162000 lane count 2
[14:49:17] [PASSED] Link rate 162000 lane count 1
[14:49:17] ========== [PASSED] drm_test_dp_mst_calc_pbn_div ===========
[14:49:17] ========= drm_test_dp_mst_sideband_msg_req_decode =========
[14:49:17] [PASSED] DP_ENUM_PATH_RESOURCES with port number
[14:49:17] [PASSED] DP_POWER_UP_PHY with port number
[14:49:17] [PASSED] DP_POWER_DOWN_PHY with port number
[14:49:17] [PASSED] DP_ALLOCATE_PAYLOAD with SDP stream sinks
[14:49:17] [PASSED] DP_ALLOCATE_PAYLOAD with port number
[14:49:17] [PASSED] DP_ALLOCATE_PAYLOAD with VCPI
[14:49:17] [PASSED] DP_ALLOCATE_PAYLOAD with PBN
[14:49:17] [PASSED] DP_QUERY_PAYLOAD with port number
[14:49:17] [PASSED] DP_QUERY_PAYLOAD with VCPI
[14:49:17] [PASSED] DP_REMOTE_DPCD_READ with port number
[14:49:17] [PASSED] DP_REMOTE_DPCD_READ with DPCD address
[14:49:17] [PASSED] DP_REMOTE_DPCD_READ with max number of bytes
[14:49:17] [PASSED] DP_REMOTE_DPCD_WRITE with port number
[14:49:17] [PASSED] DP_REMOTE_DPCD_WRITE with DPCD address
[14:49:17] [PASSED] DP_REMOTE_DPCD_WRITE with data array
[14:49:17] [PASSED] DP_REMOTE_I2C_READ with port number
[14:49:17] [PASSED] DP_REMOTE_I2C_READ with I2C device ID
[14:49:17] [PASSED] DP_REMOTE_I2C_READ with transactions array
[14:49:17] [PASSED] DP_REMOTE_I2C_WRITE with port number
[14:49:17] [PASSED] DP_REMOTE_I2C_WRITE with I2C device ID
[14:49:17] [PASSED] DP_REMOTE_I2C_WRITE with data array
[14:49:17] [PASSED] DP_QUERY_STREAM_ENC_STATUS with stream ID
[14:49:17] [PASSED] DP_QUERY_STREAM_ENC_STATUS with client ID
[14:49:17] [PASSED] DP_QUERY_STREAM_ENC_STATUS with stream event
[14:49:17] [PASSED] DP_QUERY_STREAM_ENC_STATUS with valid stream event
[14:49:17] [PASSED] DP_QUERY_STREAM_ENC_STATUS with stream behavior
[14:49:17] [PASSED] DP_QUERY_STREAM_ENC_STATUS with a valid stream behavior
[14:49:17] ===== [PASSED] drm_test_dp_mst_sideband_msg_req_decode =====
[14:49:17] ================ [PASSED] drm_dp_mst_helper ================
[14:49:17] ================== drm_exec (7 subtests) ===================
[14:49:17] [PASSED] sanitycheck
[14:49:17] [PASSED] test_lock
[14:49:17] [PASSED] test_lock_unlock
[14:49:17] [PASSED] test_duplicates
[14:49:17] [PASSED] test_prepare
[14:49:17] [PASSED] test_prepare_array
[14:49:17] [PASSED] test_multiple_loops
[14:49:17] ==================== [PASSED] drm_exec =====================
[14:49:17] =========== drm_format_helper_test (17 subtests) ===========
[14:49:17] ============== drm_test_fb_xrgb8888_to_gray8 ==============
[14:49:17] [PASSED] single_pixel_source_buffer
[14:49:17] [PASSED] single_pixel_clip_rectangle
[14:49:17] [PASSED] well_known_colors
[14:49:17] [PASSED] destination_pitch
[14:49:17] ========== [PASSED] drm_test_fb_xrgb8888_to_gray8 ==========
[14:49:17] ============= drm_test_fb_xrgb8888_to_rgb332 ==============
[14:49:17] [PASSED] single_pixel_source_buffer
[14:49:17] [PASSED] single_pixel_clip_rectangle
[14:49:17] [PASSED] well_known_colors
[14:49:17] [PASSED] destination_pitch
[14:49:17] ========= [PASSED] drm_test_fb_xrgb8888_to_rgb332 ==========
[14:49:17] ============= drm_test_fb_xrgb8888_to_rgb565 ==============
[14:49:17] [PASSED] single_pixel_source_buffer
[14:49:17] [PASSED] single_pixel_clip_rectangle
[14:49:17] [PASSED] well_known_colors
[14:49:17] [PASSED] destination_pitch
[14:49:17] ========= [PASSED] drm_test_fb_xrgb8888_to_rgb565 ==========
[14:49:17] ============ drm_test_fb_xrgb8888_to_xrgb1555 =============
[14:49:17] [PASSED] single_pixel_source_buffer
[14:49:17] [PASSED] single_pixel_clip_rectangle
[14:49:17] [PASSED] well_known_colors
[14:49:17] [PASSED] destination_pitch
[14:49:17] ======== [PASSED] drm_test_fb_xrgb8888_to_xrgb1555 =========
[14:49:17] ============ drm_test_fb_xrgb8888_to_argb1555 =============
[14:49:17] [PASSED] single_pixel_source_buffer
[14:49:17] [PASSED] single_pixel_clip_rectangle
[14:49:17] [PASSED] well_known_colors
[14:49:17] [PASSED] destination_pitch
[14:49:17] ======== [PASSED] drm_test_fb_xrgb8888_to_argb1555 =========
[14:49:17] ============ drm_test_fb_xrgb8888_to_rgba5551 =============
[14:49:17] [PASSED] single_pixel_source_buffer
[14:49:17] [PASSED] single_pixel_clip_rectangle
[14:49:17] [PASSED] well_known_colors
[14:49:17] [PASSED] destination_pitch
[14:49:17] ======== [PASSED] drm_test_fb_xrgb8888_to_rgba5551 =========
[14:49:17] ============= drm_test_fb_xrgb8888_to_rgb888 ==============
[14:49:17] [PASSED] single_pixel_source_buffer
[14:49:17] [PASSED] single_pixel_clip_rectangle
[14:49:17] [PASSED] well_known_colors
[14:49:17] [PASSED] destination_pitch
[14:49:17] ========= [PASSED] drm_test_fb_xrgb8888_to_rgb888 ==========
[14:49:17] ============= drm_test_fb_xrgb8888_to_bgr888 ==============
[14:49:17] [PASSED] single_pixel_source_buffer
[14:49:17] [PASSED] single_pixel_clip_rectangle
[14:49:17] [PASSED] well_known_colors
[14:49:17] [PASSED] destination_pitch
[14:49:17] ========= [PASSED] drm_test_fb_xrgb8888_to_bgr888 ==========
[14:49:17] ============ drm_test_fb_xrgb8888_to_argb8888 =============
[14:49:17] [PASSED] single_pixel_source_buffer
[14:49:17] [PASSED] single_pixel_clip_rectangle
[14:49:17] [PASSED] well_known_colors
[14:49:17] [PASSED] destination_pitch
[14:49:17] ======== [PASSED] drm_test_fb_xrgb8888_to_argb8888 =========
[14:49:17] =========== drm_test_fb_xrgb8888_to_xrgb2101010 ===========
[14:49:17] [PASSED] single_pixel_source_buffer
[14:49:17] [PASSED] single_pixel_clip_rectangle
[14:49:17] [PASSED] well_known_colors
[14:49:17] [PASSED] destination_pitch
[14:49:17] ======= [PASSED] drm_test_fb_xrgb8888_to_xrgb2101010 =======
[14:49:17] =========== drm_test_fb_xrgb8888_to_argb2101010 ===========
[14:49:17] [PASSED] single_pixel_source_buffer
[14:49:17] [PASSED] single_pixel_clip_rectangle
[14:49:17] [PASSED] well_known_colors
[14:49:17] [PASSED] destination_pitch
[14:49:17] ======= [PASSED] drm_test_fb_xrgb8888_to_argb2101010 =======
[14:49:17] ============== drm_test_fb_xrgb8888_to_mono ===============
[14:49:17] [PASSED] single_pixel_source_buffer
[14:49:17] [PASSED] single_pixel_clip_rectangle
[14:49:17] [PASSED] well_known_colors
[14:49:17] [PASSED] destination_pitch
[14:49:17] ========== [PASSED] drm_test_fb_xrgb8888_to_mono ===========
[14:49:17] ==================== drm_test_fb_swab =====================
[14:49:17] [PASSED] single_pixel_source_buffer
[14:49:17] [PASSED] single_pixel_clip_rectangle
[14:49:17] [PASSED] well_known_colors
[14:49:17] [PASSED] destination_pitch
[14:49:17] ================ [PASSED] drm_test_fb_swab =================
[14:49:17] ============ drm_test_fb_xrgb8888_to_xbgr8888 =============
[14:49:17] [PASSED] single_pixel_source_buffer
[14:49:17] [PASSED] single_pixel_clip_rectangle
[14:49:17] [PASSED] well_known_colors
[14:49:17] [PASSED] destination_pitch
[14:49:17] ======== [PASSED] drm_test_fb_xrgb8888_to_xbgr8888 =========
[14:49:17] ============ drm_test_fb_xrgb8888_to_abgr8888 =============
[14:49:17] [PASSED] single_pixel_source_buffer
[14:49:17] [PASSED] single_pixel_clip_rectangle
[14:49:17] [PASSED] well_known_colors
[14:49:17] [PASSED] destination_pitch
[14:49:17] ======== [PASSED] drm_test_fb_xrgb8888_to_abgr8888 =========
[14:49:17] ================= drm_test_fb_clip_offset =================
[14:49:17] [PASSED] pass through
[14:49:17] [PASSED] horizontal offset
[14:49:17] [PASSED] vertical offset
[14:49:17] [PASSED] horizontal and vertical offset
[14:49:17] [PASSED] horizontal offset (custom pitch)
[14:49:17] [PASSED] vertical offset (custom pitch)
[14:49:17] [PASSED] horizontal and vertical offset (custom pitch)
[14:49:17] ============= [PASSED] drm_test_fb_clip_offset =============
[14:49:17] =================== drm_test_fb_memcpy ====================
[14:49:17] [PASSED] single_pixel_source_buffer: XR24 little-endian (0x34325258)
[14:49:17] [PASSED] single_pixel_source_buffer: XRA8 little-endian (0x38415258)
[14:49:17] [PASSED] single_pixel_source_buffer: YU24 little-endian (0x34325559)
[14:49:17] [PASSED] single_pixel_clip_rectangle: XB24 little-endian (0x34324258)
[14:49:17] [PASSED] single_pixel_clip_rectangle: XRA8 little-endian (0x38415258)
[14:49:17] [PASSED] single_pixel_clip_rectangle: YU24 little-endian (0x34325559)
[14:49:17] [PASSED] well_known_colors: XB24 little-endian (0x34324258)
[14:49:17] [PASSED] well_known_colors: XRA8 little-endian (0x38415258)
[14:49:17] [PASSED] well_known_colors: YU24 little-endian (0x34325559)
[14:49:17] [PASSED] destination_pitch: XB24 little-endian (0x34324258)
[14:49:17] [PASSED] destination_pitch: XRA8 little-endian (0x38415258)
[14:49:17] [PASSED] destination_pitch: YU24 little-endian (0x34325559)
[14:49:17] =============== [PASSED] drm_test_fb_memcpy ================
[14:49:17] ============= [PASSED] drm_format_helper_test ==============
[14:49:17] ================= drm_format (18 subtests) =================
[14:49:17] [PASSED] drm_test_format_block_width_invalid
[14:49:17] [PASSED] drm_test_format_block_width_one_plane
[14:49:17] [PASSED] drm_test_format_block_width_two_plane
[14:49:17] [PASSED] drm_test_format_block_width_three_plane
[14:49:17] [PASSED] drm_test_format_block_width_tiled
[14:49:17] [PASSED] drm_test_format_block_height_invalid
[14:49:17] [PASSED] drm_test_format_block_height_one_plane
[14:49:17] [PASSED] drm_test_format_block_height_two_plane
[14:49:17] [PASSED] drm_test_format_block_height_three_plane
[14:49:17] [PASSED] drm_test_format_block_height_tiled
[14:49:17] [PASSED] drm_test_format_min_pitch_invalid
[14:49:17] [PASSED] drm_test_format_min_pitch_one_plane_8bpp
[14:49:17] [PASSED] drm_test_format_min_pitch_one_plane_16bpp
[14:49:17] [PASSED] drm_test_format_min_pitch_one_plane_24bpp
[14:49:17] [PASSED] drm_test_format_min_pitch_one_plane_32bpp
[14:49:17] [PASSED] drm_test_format_min_pitch_two_plane
[14:49:17] [PASSED] drm_test_format_min_pitch_three_plane_8bpp
[14:49:17] [PASSED] drm_test_format_min_pitch_tiled
[14:49:17] =================== [PASSED] drm_format ====================
[14:49:17] ============== drm_framebuffer (10 subtests) ===============
[14:49:17] ========== drm_test_framebuffer_check_src_coords ==========
[14:49:17] [PASSED] Success: source fits into fb
[14:49:17] [PASSED] Fail: overflowing fb with x-axis coordinate
[14:49:17] [PASSED] Fail: overflowing fb with y-axis coordinate
[14:49:17] [PASSED] Fail: overflowing fb with source width
[14:49:17] [PASSED] Fail: overflowing fb with source height
[14:49:17] ====== [PASSED] drm_test_framebuffer_check_src_coords ======
[14:49:17] [PASSED] drm_test_framebuffer_cleanup
[14:49:17] =============== drm_test_framebuffer_create ===============
[14:49:17] [PASSED] ABGR8888 normal sizes
[14:49:17] [PASSED] ABGR8888 max sizes
[14:49:17] [PASSED] ABGR8888 pitch greater than min required
[14:49:17] [PASSED] ABGR8888 pitch less than min required
[14:49:17] [PASSED] ABGR8888 Invalid width
[14:49:17] [PASSED] ABGR8888 Invalid buffer handle
[14:49:17] [PASSED] No pixel format
[14:49:17] [PASSED] ABGR8888 Width 0
[14:49:17] [PASSED] ABGR8888 Height 0
[14:49:17] [PASSED] ABGR8888 Out of bound height * pitch combination
[14:49:17] [PASSED] ABGR8888 Large buffer offset
[14:49:17] [PASSED] ABGR8888 Buffer offset for inexistent plane
[14:49:17] [PASSED] ABGR8888 Invalid flag
[14:49:17] [PASSED] ABGR8888 Set DRM_MODE_FB_MODIFIERS without modifiers
[14:49:17] [PASSED] ABGR8888 Valid buffer modifier
[14:49:17] [PASSED] ABGR8888 Invalid buffer modifier(DRM_FORMAT_MOD_SAMSUNG_64_32_TILE)
[14:49:17] [PASSED] ABGR8888 Extra pitches without DRM_MODE_FB_MODIFIERS
[14:49:17] [PASSED] ABGR8888 Extra pitches with DRM_MODE_FB_MODIFIERS
[14:49:17] [PASSED] NV12 Normal sizes
[14:49:17] [PASSED] NV12 Max sizes
[14:49:17] [PASSED] NV12 Invalid pitch
[14:49:17] [PASSED] NV12 Invalid modifier/missing DRM_MODE_FB_MODIFIERS flag
[14:49:17] [PASSED] NV12 different modifier per-plane
[14:49:17] [PASSED] NV12 with DRM_FORMAT_MOD_SAMSUNG_64_32_TILE
[14:49:17] [PASSED] NV12 Valid modifiers without DRM_MODE_FB_MODIFIERS
[14:49:17] [PASSED] NV12 Modifier for inexistent plane
[14:49:17] [PASSED] NV12 Handle for inexistent plane
[14:49:17] [PASSED] NV12 Handle for inexistent plane without DRM_MODE_FB_MODIFIERS
[14:49:17] [PASSED] YVU420 DRM_MODE_FB_MODIFIERS set without modifier
[14:49:17] [PASSED] YVU420 Normal sizes
[14:49:17] [PASSED] YVU420 Max sizes
[14:49:17] [PASSED] YVU420 Invalid pitch
[14:49:17] [PASSED] YVU420 Different pitches
[14:49:17] [PASSED] YVU420 Different buffer offsets/pitches
[14:49:17] [PASSED] YVU420 Modifier set just for plane 0, without DRM_MODE_FB_MODIFIERS
[14:49:17] [PASSED] YVU420 Modifier set just for planes 0, 1, without DRM_MODE_FB_MODIFIERS
[14:49:17] [PASSED] YVU420 Modifier set just for plane 0, 1, with DRM_MODE_FB_MODIFIERS
[14:49:17] [PASSED] YVU420 Valid modifier
[14:49:17] [PASSED] YVU420 Different modifiers per plane
[14:49:17] [PASSED] YVU420 Modifier for inexistent plane
[14:49:17] [PASSED] YUV420_10BIT Invalid modifier(DRM_FORMAT_MOD_LINEAR)
[14:49:17] [PASSED] X0L2 Normal sizes
[14:49:17] [PASSED] X0L2 Max sizes
[14:49:17] [PASSED] X0L2 Invalid pitch
[14:49:17] [PASSED] X0L2 Pitch greater than minimum required
[14:49:17] [PASSED] X0L2 Handle for inexistent plane
[14:49:17] [PASSED] X0L2 Offset for inexistent plane, without DRM_MODE_FB_MODIFIERS set
[14:49:17] [PASSED] X0L2 Modifier without DRM_MODE_FB_MODIFIERS set
[14:49:17] [PASSED] X0L2 Valid modifier
[14:49:17] [PASSED] X0L2 Modifier for inexistent plane
[14:49:17] =========== [PASSED] drm_test_framebuffer_create ===========
[14:49:17] [PASSED] drm_test_framebuffer_free
[14:49:17] [PASSED] drm_test_framebuffer_init
[14:49:17] [PASSED] drm_test_framebuffer_init_bad_format
[14:49:17] [PASSED] drm_test_framebuffer_init_dev_mismatch
[14:49:17] [PASSED] drm_test_framebuffer_lookup
[14:49:17] [PASSED] drm_test_framebuffer_lookup_inexistent
[14:49:17] [PASSED] drm_test_framebuffer_modifiers_not_supported
[14:49:17] ================= [PASSED] drm_framebuffer =================
[14:49:17] ================ drm_gem_shmem (8 subtests) ================
[14:49:17] [PASSED] drm_gem_shmem_test_obj_create
[14:49:17] [PASSED] drm_gem_shmem_test_obj_create_private
[14:49:17] [PASSED] drm_gem_shmem_test_pin_pages
[14:49:17] [PASSED] drm_gem_shmem_test_vmap
[14:49:17] [PASSED] drm_gem_shmem_test_get_sg_table
[14:49:17] [PASSED] drm_gem_shmem_test_get_pages_sgt
[14:49:17] [PASSED] drm_gem_shmem_test_madvise
[14:49:17] [PASSED] drm_gem_shmem_test_purge
[14:49:17] ================== [PASSED] drm_gem_shmem ==================
[14:49:17] === drm_atomic_helper_connector_hdmi_check (27 subtests) ===
[14:49:17] [PASSED] drm_test_check_broadcast_rgb_auto_cea_mode
[14:49:17] [PASSED] drm_test_check_broadcast_rgb_auto_cea_mode_vic_1
[14:49:17] [PASSED] drm_test_check_broadcast_rgb_full_cea_mode
[14:49:17] [PASSED] drm_test_check_broadcast_rgb_full_cea_mode_vic_1
[14:49:17] [PASSED] drm_test_check_broadcast_rgb_limited_cea_mode
[14:49:17] [PASSED] drm_test_check_broadcast_rgb_limited_cea_mode_vic_1
[14:49:17] ====== drm_test_check_broadcast_rgb_cea_mode_yuv420 =======
[14:49:17] [PASSED] Automatic
[14:49:17] [PASSED] Full
[14:49:17] [PASSED] Limited 16:235
[14:49:17] == [PASSED] drm_test_check_broadcast_rgb_cea_mode_yuv420 ===
[14:49:17] [PASSED] drm_test_check_broadcast_rgb_crtc_mode_changed
[14:49:17] [PASSED] drm_test_check_broadcast_rgb_crtc_mode_not_changed
[14:49:17] [PASSED] drm_test_check_disable_connector
[14:49:17] [PASSED] drm_test_check_hdmi_funcs_reject_rate
[14:49:17] [PASSED] drm_test_check_max_tmds_rate_bpc_fallback_rgb
[14:49:17] [PASSED] drm_test_check_max_tmds_rate_bpc_fallback_yuv420
[14:49:17] [PASSED] drm_test_check_max_tmds_rate_bpc_fallback_ignore_yuv422
[14:49:17] [PASSED] drm_test_check_max_tmds_rate_bpc_fallback_ignore_yuv420
[14:49:17] [PASSED] drm_test_check_driver_unsupported_fallback_yuv420
[14:49:17] [PASSED] drm_test_check_output_bpc_crtc_mode_changed
[14:49:17] [PASSED] drm_test_check_output_bpc_crtc_mode_not_changed
[14:49:17] [PASSED] drm_test_check_output_bpc_dvi
[14:49:17] [PASSED] drm_test_check_output_bpc_format_vic_1
[14:49:17] [PASSED] drm_test_check_output_bpc_format_display_8bpc_only
[14:49:17] [PASSED] drm_test_check_output_bpc_format_display_rgb_only
[14:49:17] [PASSED] drm_test_check_output_bpc_format_driver_8bpc_only
[14:49:17] [PASSED] drm_test_check_output_bpc_format_driver_rgb_only
[14:49:17] [PASSED] drm_test_check_tmds_char_rate_rgb_8bpc
[14:49:17] [PASSED] drm_test_check_tmds_char_rate_rgb_10bpc
[14:49:17] [PASSED] drm_test_check_tmds_char_rate_rgb_12bpc
[14:49:17] ===== [PASSED] drm_atomic_helper_connector_hdmi_check ======
[14:49:17] === drm_atomic_helper_connector_hdmi_reset (6 subtests) ====
[14:49:17] [PASSED] drm_test_check_broadcast_rgb_value
[14:49:17] [PASSED] drm_test_check_bpc_8_value
[14:49:17] [PASSED] drm_test_check_bpc_10_value
[14:49:17] [PASSED] drm_test_check_bpc_12_value
[14:49:17] [PASSED] drm_test_check_format_value
[14:49:17] [PASSED] drm_test_check_tmds_char_value
[14:49:17] ===== [PASSED] drm_atomic_helper_connector_hdmi_reset ======
[14:49:17] = drm_atomic_helper_connector_hdmi_mode_valid (4 subtests) =
[14:49:17] [PASSED] drm_test_check_mode_valid
[14:49:17] [PASSED] drm_test_check_mode_valid_reject
[14:49:17] [PASSED] drm_test_check_mode_valid_reject_rate
[14:49:17] [PASSED] drm_test_check_mode_valid_reject_max_clock
[14:49:17] === [PASSED] drm_atomic_helper_connector_hdmi_mode_valid ===
[14:49:17] = drm_atomic_helper_connector_hdmi_infoframes (5 subtests) =
[14:49:17] [PASSED] drm_test_check_infoframes
[14:49:17] [PASSED] drm_test_check_reject_avi_infoframe
[14:49:18] [PASSED] drm_test_check_reject_hdr_infoframe_bpc_8
[14:49:18] [PASSED] drm_test_check_reject_hdr_infoframe_bpc_10
[14:49:18] [PASSED] drm_test_check_reject_audio_infoframe
[14:49:18] === [PASSED] drm_atomic_helper_connector_hdmi_infoframes ===
[14:49:18] ================= drm_managed (2 subtests) =================
[14:49:18] [PASSED] drm_test_managed_release_action
[14:49:18] [PASSED] drm_test_managed_run_action
[14:49:18] =================== [PASSED] drm_managed ===================
[14:49:18] =================== drm_mm (6 subtests) ====================
[14:49:18] [PASSED] drm_test_mm_init
[14:49:18] [PASSED] drm_test_mm_debug
[14:49:18] [PASSED] drm_test_mm_align32
[14:49:18] [PASSED] drm_test_mm_align64
[14:49:18] [PASSED] drm_test_mm_lowest
[14:49:18] [PASSED] drm_test_mm_highest
[14:49:18] ===================== [PASSED] drm_mm ======================
[14:49:18] ============= drm_modes_analog_tv (5 subtests) =============
[14:49:18] [PASSED] drm_test_modes_analog_tv_mono_576i
[14:49:18] [PASSED] drm_test_modes_analog_tv_ntsc_480i
[14:49:18] [PASSED] drm_test_modes_analog_tv_ntsc_480i_inlined
[14:49:18] [PASSED] drm_test_modes_analog_tv_pal_576i
[14:49:18] [PASSED] drm_test_modes_analog_tv_pal_576i_inlined
[14:49:18] =============== [PASSED] drm_modes_analog_tv ===============
[14:49:18] ============== drm_plane_helper (2 subtests) ===============
[14:49:18] =============== drm_test_check_plane_state ================
[14:49:18] [PASSED] clipping_simple
[14:49:18] [PASSED] clipping_rotate_reflect
[14:49:18] [PASSED] positioning_simple
[14:49:18] [PASSED] upscaling
[14:49:18] [PASSED] downscaling
[14:49:18] [PASSED] rounding1
[14:49:18] [PASSED] rounding2
[14:49:18] [PASSED] rounding3
[14:49:18] [PASSED] rounding4
[14:49:18] =========== [PASSED] drm_test_check_plane_state ============
[14:49:18] =========== drm_test_check_invalid_plane_state ============
[14:49:18] [PASSED] positioning_invalid
[14:49:18] [PASSED] upscaling_invalid
[14:49:18] [PASSED] downscaling_invalid
[14:49:18] ======= [PASSED] drm_test_check_invalid_plane_state ========
[14:49:18] ================ [PASSED] drm_plane_helper =================
[14:49:18] ====== drm_connector_helper_tv_get_modes (1 subtest) =======
[14:49:18] ====== drm_test_connector_helper_tv_get_modes_check =======
[14:49:18] [PASSED] None
[14:49:18] [PASSED] PAL
[14:49:18] [PASSED] NTSC
[14:49:18] [PASSED] Both, NTSC Default
[14:49:18] [PASSED] Both, PAL Default
[14:49:18] [PASSED] Both, NTSC Default, with PAL on command-line
[14:49:18] [PASSED] Both, PAL Default, with NTSC on command-line
[14:49:18] == [PASSED] drm_test_connector_helper_tv_get_modes_check ===
[14:49:18] ======== [PASSED] drm_connector_helper_tv_get_modes ========
[14:49:18] ================== drm_rect (9 subtests) ===================
[14:49:18] [PASSED] drm_test_rect_clip_scaled_div_by_zero
[14:49:18] [PASSED] drm_test_rect_clip_scaled_not_clipped
[14:49:18] [PASSED] drm_test_rect_clip_scaled_clipped
[14:49:18] [PASSED] drm_test_rect_clip_scaled_signed_vs_unsigned
[14:49:18] ================= drm_test_rect_intersect =================
[14:49:18] [PASSED] top-left x bottom-right: 2x2+1+1 x 2x2+0+0
[14:49:18] [PASSED] top-right x bottom-left: 2x2+0+0 x 2x2+1-1
[14:49:18] [PASSED] bottom-left x top-right: 2x2+1-1 x 2x2+0+0
[14:49:18] [PASSED] bottom-right x top-left: 2x2+0+0 x 2x2+1+1
[14:49:18] [PASSED] right x left: 2x1+0+0 x 3x1+1+0
[14:49:18] [PASSED] left x right: 3x1+1+0 x 2x1+0+0
[14:49:18] [PASSED] up x bottom: 1x2+0+0 x 1x3+0-1
[14:49:18] [PASSED] bottom x up: 1x3+0-1 x 1x2+0+0
[14:49:18] [PASSED] touching corner: 1x1+0+0 x 2x2+1+1
[14:49:18] [PASSED] touching side: 1x1+0+0 x 1x1+1+0
[14:49:18] [PASSED] equal rects: 2x2+0+0 x 2x2+0+0
[14:49:18] [PASSED] inside another: 2x2+0+0 x 1x1+1+1
[14:49:18] [PASSED] far away: 1x1+0+0 x 1x1+3+6
[14:49:18] [PASSED] points intersecting: 0x0+5+10 x 0x0+5+10
[14:49:18] [PASSED] points not intersecting: 0x0+0+0 x 0x0+5+10
[14:49:18] ============= [PASSED] drm_test_rect_intersect =============
[14:49:18] ================ drm_test_rect_calc_hscale ================
[14:49:18] [PASSED] normal use
[14:49:18] [PASSED] out of max range
[14:49:18] [PASSED] out of min range
[14:49:18] [PASSED] zero dst
[14:49:18] [PASSED] negative src
[14:49:18] [PASSED] negative dst
[14:49:18] ============ [PASSED] drm_test_rect_calc_hscale ============
[14:49:18] ================ drm_test_rect_calc_vscale ================
[14:49:18] [PASSED] normal use
[14:49:18] [PASSED] out of max range
[14:49:18] [PASSED] out of min range
[14:49:18] [PASSED] zero dst
[14:49:18] [PASSED] negative src
[14:49:18] [PASSED] negative dst
[14:49:18] ============ [PASSED] drm_test_rect_calc_vscale ============
[14:49:18] ================== drm_test_rect_rotate ===================
[14:49:18] [PASSED] reflect-x
[14:49:18] [PASSED] reflect-y
[14:49:18] [PASSED] rotate-0
[14:49:18] [PASSED] rotate-90
[14:49:18] [PASSED] rotate-180
[14:49:18] [PASSED] rotate-270
[14:49:18] ============== [PASSED] drm_test_rect_rotate ===============
[14:49:18] ================ drm_test_rect_rotate_inv =================
[14:49:18] [PASSED] reflect-x
[14:49:18] [PASSED] reflect-y
[14:49:18] [PASSED] rotate-0
[14:49:18] [PASSED] rotate-90
[14:49:18] [PASSED] rotate-180
[14:49:18] [PASSED] rotate-270
[14:49:18] ============ [PASSED] drm_test_rect_rotate_inv =============
[14:49:18] ==================== [PASSED] drm_rect =====================
[14:49:18] ============ drm_sysfb_modeset_test (1 subtest) ============
[14:49:18] ============ drm_test_sysfb_build_fourcc_list =============
[14:49:18] [PASSED] no native formats
[14:49:18] [PASSED] XRGB8888 as native format
[14:49:18] [PASSED] remove duplicates
[14:49:18] [PASSED] convert alpha formats
[14:49:18] [PASSED] random formats
[14:49:18] ======== [PASSED] drm_test_sysfb_build_fourcc_list =========
[14:49:18] ============= [PASSED] drm_sysfb_modeset_test ==============
[14:49:18] ================== drm_fixp (2 subtests) ===================
[14:49:18] [PASSED] drm_test_int2fixp
[14:49:18] [PASSED] drm_test_sm2fixp
[14:49:18] ==================== [PASSED] drm_fixp =====================
[14:49:18] ============================================================
[14:49:18] Testing complete. Ran 621 tests: passed: 621
[14:49:18] Elapsed time: 25.980s total, 1.726s configuring, 24.129s building, 0.125s running
+ /kernel/tools/testing/kunit/kunit.py run --kunitconfig /kernel/drivers/gpu/drm/ttm/tests/.kunitconfig
[14:49:18] Configuring KUnit Kernel ...
Regenerating .config ...
Populating config with:
$ make ARCH=um O=.kunit olddefconfig
[14:49:19] Building KUnit Kernel ...
Populating config with:
$ make ARCH=um O=.kunit olddefconfig
Building with:
$ make all compile_commands.json scripts_gdb ARCH=um O=.kunit --jobs=48
[14:49:29] Starting KUnit Kernel (1/1)...
[14:49:29] ============================================================
Running tests with:
$ .kunit/linux kunit.enable=1 mem=1G console=tty kunit_shutdown=halt
[14:49:29] ================= ttm_device (5 subtests) ==================
[14:49:29] [PASSED] ttm_device_init_basic
[14:49:29] [PASSED] ttm_device_init_multiple
[14:49:29] [PASSED] ttm_device_fini_basic
[14:49:29] [PASSED] ttm_device_init_no_vma_man
[14:49:29] ================== ttm_device_init_pools ==================
[14:49:29] [PASSED] No DMA allocations, no DMA32 required
[14:49:29] [PASSED] DMA allocations, DMA32 required
[14:49:29] [PASSED] No DMA allocations, DMA32 required
[14:49:29] [PASSED] DMA allocations, no DMA32 required
[14:49:29] ============== [PASSED] ttm_device_init_pools ==============
[14:49:29] =================== [PASSED] ttm_device ====================
[14:49:29] ================== ttm_pool (8 subtests) ===================
[14:49:29] ================== ttm_pool_alloc_basic ===================
[14:49:29] [PASSED] One page
[14:49:29] [PASSED] More than one page
[14:49:29] [PASSED] Above the allocation limit
[14:49:29] [PASSED] One page, with coherent DMA mappings enabled
[14:49:29] [PASSED] Above the allocation limit, with coherent DMA mappings enabled
[14:49:29] ============== [PASSED] ttm_pool_alloc_basic ===============
[14:49:29] ============== ttm_pool_alloc_basic_dma_addr ==============
[14:49:29] [PASSED] One page
[14:49:29] [PASSED] More than one page
[14:49:29] [PASSED] Above the allocation limit
[14:49:29] [PASSED] One page, with coherent DMA mappings enabled
[14:49:29] [PASSED] Above the allocation limit, with coherent DMA mappings enabled
[14:49:29] ========== [PASSED] ttm_pool_alloc_basic_dma_addr ==========
[14:49:29] [PASSED] ttm_pool_alloc_order_caching_match
[14:49:29] [PASSED] ttm_pool_alloc_caching_mismatch
[14:49:29] [PASSED] ttm_pool_alloc_order_mismatch
[14:49:29] [PASSED] ttm_pool_free_dma_alloc
[14:49:29] [PASSED] ttm_pool_free_no_dma_alloc
[14:49:29] [PASSED] ttm_pool_fini_basic
[14:49:29] ==================== [PASSED] ttm_pool =====================
[14:49:29] ================ ttm_resource (8 subtests) =================
[14:49:29] ================= ttm_resource_init_basic =================
[14:49:29] [PASSED] Init resource in TTM_PL_SYSTEM
[14:49:29] [PASSED] Init resource in TTM_PL_VRAM
[14:49:29] [PASSED] Init resource in a private placement
[14:49:29] [PASSED] Init resource in TTM_PL_SYSTEM, set placement flags
[14:49:29] ============= [PASSED] ttm_resource_init_basic =============
[14:49:29] [PASSED] ttm_resource_init_pinned
[14:49:29] [PASSED] ttm_resource_fini_basic
[14:49:29] [PASSED] ttm_resource_manager_init_basic
[14:49:29] [PASSED] ttm_resource_manager_usage_basic
[14:49:29] [PASSED] ttm_resource_manager_set_used_basic
[14:49:29] [PASSED] ttm_sys_man_alloc_basic
[14:49:29] [PASSED] ttm_sys_man_free_basic
[14:49:29] ================== [PASSED] ttm_resource ===================
[14:49:29] =================== ttm_tt (15 subtests) ===================
[14:49:29] ==================== ttm_tt_init_basic ====================
[14:49:29] [PASSED] Page-aligned size
[14:49:29] [PASSED] Extra pages requested
[14:49:29] ================ [PASSED] ttm_tt_init_basic ================
[14:49:29] [PASSED] ttm_tt_init_misaligned
[14:49:29] [PASSED] ttm_tt_fini_basic
[14:49:29] [PASSED] ttm_tt_fini_sg
[14:49:29] [PASSED] ttm_tt_fini_shmem
[14:49:29] [PASSED] ttm_tt_create_basic
[14:49:29] [PASSED] ttm_tt_create_invalid_bo_type
[14:49:29] [PASSED] ttm_tt_create_ttm_exists
[14:49:29] [PASSED] ttm_tt_create_failed
[14:49:29] [PASSED] ttm_tt_destroy_basic
[14:49:29] [PASSED] ttm_tt_populate_null_ttm
[14:49:29] [PASSED] ttm_tt_populate_populated_ttm
[14:49:29] [PASSED] ttm_tt_unpopulate_basic
[14:49:29] [PASSED] ttm_tt_unpopulate_empty_ttm
[14:49:29] [PASSED] ttm_tt_swapin_basic
[14:49:29] ===================== [PASSED] ttm_tt ======================
[14:49:29] =================== ttm_bo (14 subtests) ===================
[14:49:29] =========== ttm_bo_reserve_optimistic_no_ticket ===========
[14:49:29] [PASSED] Cannot be interrupted and sleeps
[14:49:29] [PASSED] Cannot be interrupted, locks straight away
[14:49:29] [PASSED] Can be interrupted, sleeps
[14:49:29] ======= [PASSED] ttm_bo_reserve_optimistic_no_ticket =======
[14:49:29] [PASSED] ttm_bo_reserve_locked_no_sleep
[14:49:29] [PASSED] ttm_bo_reserve_no_wait_ticket
[14:49:29] [PASSED] ttm_bo_reserve_double_resv
[14:49:29] [PASSED] ttm_bo_reserve_interrupted
[14:49:29] [PASSED] ttm_bo_reserve_deadlock
[14:49:29] [PASSED] ttm_bo_unreserve_basic
[14:49:29] [PASSED] ttm_bo_unreserve_pinned
[14:49:29] [PASSED] ttm_bo_unreserve_bulk
[14:49:29] [PASSED] ttm_bo_fini_basic
[14:49:29] [PASSED] ttm_bo_fini_shared_resv
[14:49:29] [PASSED] ttm_bo_pin_basic
[14:49:29] [PASSED] ttm_bo_pin_unpin_resource
[14:49:29] [PASSED] ttm_bo_multiple_pin_one_unpin
[14:49:29] ===================== [PASSED] ttm_bo ======================
[14:49:29] ============== ttm_bo_validate (22 subtests) ===============
[14:49:29] ============== ttm_bo_init_reserved_sys_man ===============
[14:49:29] [PASSED] Buffer object for userspace
[14:49:29] [PASSED] Kernel buffer object
[14:49:29] [PASSED] Shared buffer object
[14:49:29] ========== [PASSED] ttm_bo_init_reserved_sys_man ===========
[14:49:29] ============== ttm_bo_init_reserved_mock_man ==============
[14:49:29] [PASSED] Buffer object for userspace
[14:49:29] [PASSED] Kernel buffer object
[14:49:29] [PASSED] Shared buffer object
[14:49:29] ========== [PASSED] ttm_bo_init_reserved_mock_man ==========
[14:49:29] [PASSED] ttm_bo_init_reserved_resv
[14:49:29] ================== ttm_bo_validate_basic ==================
[14:49:29] [PASSED] Buffer object for userspace
[14:49:29] [PASSED] Kernel buffer object
[14:49:29] [PASSED] Shared buffer object
[14:49:29] ============== [PASSED] ttm_bo_validate_basic ==============
[14:49:29] [PASSED] ttm_bo_validate_invalid_placement
[14:49:29] ============= ttm_bo_validate_same_placement ==============
[14:49:29] [PASSED] System manager
[14:49:29] [PASSED] VRAM manager
[14:49:29] ========= [PASSED] ttm_bo_validate_same_placement ==========
[14:49:29] [PASSED] ttm_bo_validate_failed_alloc
[14:49:29] [PASSED] ttm_bo_validate_pinned
[14:49:29] [PASSED] ttm_bo_validate_busy_placement
[14:49:29] ================ ttm_bo_validate_multihop =================
[14:49:29] [PASSED] Buffer object for userspace
[14:49:29] [PASSED] Kernel buffer object
[14:49:29] [PASSED] Shared buffer object
[14:49:29] ============ [PASSED] ttm_bo_validate_multihop =============
[14:49:29] ========== ttm_bo_validate_no_placement_signaled ==========
[14:49:29] [PASSED] Buffer object in system domain, no page vector
[14:49:29] [PASSED] Buffer object in system domain with an existing page vector
[14:49:29] ====== [PASSED] ttm_bo_validate_no_placement_signaled ======
[14:49:29] ======== ttm_bo_validate_no_placement_not_signaled ========
[14:49:29] [PASSED] Buffer object for userspace
[14:49:29] [PASSED] Kernel buffer object
[14:49:29] [PASSED] Shared buffer object
[14:49:29] ==== [PASSED] ttm_bo_validate_no_placement_not_signaled ====
[14:49:29] [PASSED] ttm_bo_validate_move_fence_signaled
[14:49:29] ========= ttm_bo_validate_move_fence_not_signaled =========
[14:49:29] [PASSED] Waits for GPU
[14:49:29] [PASSED] Tries to lock straight away
[14:49:29] ===== [PASSED] ttm_bo_validate_move_fence_not_signaled =====
[14:49:29] [PASSED] ttm_bo_validate_swapout
[14:49:29] [PASSED] ttm_bo_validate_happy_evict
[14:49:29] [PASSED] ttm_bo_validate_all_pinned_evict
[14:49:29] [PASSED] ttm_bo_validate_allowed_only_evict
[14:49:29] [PASSED] ttm_bo_validate_deleted_evict
[14:49:29] [PASSED] ttm_bo_validate_busy_domain_evict
[14:49:29] [PASSED] ttm_bo_validate_evict_gutting
[14:49:29] [PASSED] ttm_bo_validate_recrusive_evict
[14:49:29] ================= [PASSED] ttm_bo_validate =================
[14:49:29] ============================================================
[14:49:29] Testing complete. Ran 102 tests: passed: 102
[14:49:29] Elapsed time: 11.481s total, 1.692s configuring, 9.575s building, 0.177s running
+ cleanup
++ stat -c %u:%g /kernel
+ chown -R 1003:1003 /kernel
^ permalink raw reply [flat|nested] 8+ messages in thread
* ✓ Xe.CI.BAT: success for series starting with [v2,1/2] gpu/buddy: replace dual-tree/force_merge with decoupled clear tracker
2026-05-04 11:10 [PATCH v2 1/2] gpu/buddy: replace dual-tree/force_merge with decoupled clear tracker Arunpravin Paneer Selvam
` (2 preceding siblings ...)
2026-05-04 14:49 ` ✓ CI.KUnit: success " Patchwork
@ 2026-05-04 15:43 ` Patchwork
2026-05-04 19:24 ` ✓ Xe.CI.FULL: " Patchwork
2026-05-08 15:19 ` [PATCH v2 1/2] " Matthew Auld
5 siblings, 0 replies; 8+ messages in thread
From: Patchwork @ 2026-05-04 15:43 UTC (permalink / raw)
To: Arunpravin Paneer Selvam; +Cc: intel-xe
[-- Attachment #1: Type: text/plain, Size: 1013 bytes --]
== Series Details ==
Series: series starting with [v2,1/2] gpu/buddy: replace dual-tree/force_merge with decoupled clear tracker
URL : https://patchwork.freedesktop.org/series/165893/
State : success
== Summary ==
CI Bug Log - changes from xe-4976-5e3e393d3eab6774e7aa49acd82f4bba2d8ef6a5_BAT -> xe-pw-165893v1_BAT
====================================================
Summary
-------
**SUCCESS**
No regressions found.
Participating hosts (13 -> 13)
------------------------------
No changes in participating hosts
Changes
-------
No changes found
Build changes
-------------
* Linux: xe-4976-5e3e393d3eab6774e7aa49acd82f4bba2d8ef6a5 -> xe-pw-165893v1
IGT_8882: 3ef61dd87d23e6c8d7956b5a72ba01a6597a3fd1 @ https://gitlab.freedesktop.org/drm/igt-gpu-tools.git
xe-4976-5e3e393d3eab6774e7aa49acd82f4bba2d8ef6a5: 5e3e393d3eab6774e7aa49acd82f4bba2d8ef6a5
xe-pw-165893v1: 165893v1
== Logs ==
For more details see: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/index.html
[-- Attachment #2: Type: text/html, Size: 1561 bytes --]
^ permalink raw reply [flat|nested] 8+ messages in thread
* ✓ Xe.CI.FULL: success for series starting with [v2,1/2] gpu/buddy: replace dual-tree/force_merge with decoupled clear tracker
2026-05-04 11:10 [PATCH v2 1/2] gpu/buddy: replace dual-tree/force_merge with decoupled clear tracker Arunpravin Paneer Selvam
` (3 preceding siblings ...)
2026-05-04 15:43 ` ✓ Xe.CI.BAT: " Patchwork
@ 2026-05-04 19:24 ` Patchwork
2026-05-08 15:19 ` [PATCH v2 1/2] " Matthew Auld
5 siblings, 0 replies; 8+ messages in thread
From: Patchwork @ 2026-05-04 19:24 UTC (permalink / raw)
To: Arunpravin Paneer Selvam; +Cc: intel-xe
[-- Attachment #1: Type: text/plain, Size: 37679 bytes --]
== Series Details ==
Series: series starting with [v2,1/2] gpu/buddy: replace dual-tree/force_merge with decoupled clear tracker
URL : https://patchwork.freedesktop.org/series/165893/
State : success
== Summary ==
CI Bug Log - changes from xe-4976-5e3e393d3eab6774e7aa49acd82f4bba2d8ef6a5_FULL -> xe-pw-165893v1_FULL
====================================================
Summary
-------
**SUCCESS**
No regressions found.
Participating hosts (2 -> 2)
------------------------------
No changes in participating hosts
Known issues
------------
Here are the changes found in xe-pw-165893v1_FULL that come from known issues:
### IGT changes ###
#### Issues hit ####
* igt@kms_big_fb@4-tiled-32bpp-rotate-270:
- shard-lnl: NOTRUN -> [SKIP][1] ([Intel XE#1407]) +1 other test skip
[1]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-lnl-3/igt@kms_big_fb@4-tiled-32bpp-rotate-270.html
* igt@kms_big_fb@4-tiled-8bpp-rotate-0:
- shard-lnl: [PASS][2] -> [ABORT][3] ([Intel XE#4760])
[2]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-4976-5e3e393d3eab6774e7aa49acd82f4bba2d8ef6a5/shard-lnl-8/igt@kms_big_fb@4-tiled-8bpp-rotate-0.html
[3]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-lnl-8/igt@kms_big_fb@4-tiled-8bpp-rotate-0.html
* igt@kms_big_fb@x-tiled-16bpp-rotate-270:
- shard-bmg: NOTRUN -> [SKIP][4] ([Intel XE#2327])
[4]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-bmg-9/igt@kms_big_fb@x-tiled-16bpp-rotate-270.html
* igt@kms_big_fb@x-tiled-8bpp-rotate-180:
- shard-bmg: [PASS][5] -> [SKIP][6] ([Intel XE#2231] / [Intel XE#6703])
[5]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-4976-5e3e393d3eab6774e7aa49acd82f4bba2d8ef6a5/shard-bmg-5/igt@kms_big_fb@x-tiled-8bpp-rotate-180.html
[6]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-bmg-2/igt@kms_big_fb@x-tiled-8bpp-rotate-180.html
* igt@kms_big_fb@y-tiled-addfb-size-offset-overflow:
- shard-lnl: NOTRUN -> [SKIP][7] ([Intel XE#1477] / [Intel XE#7361])
[7]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-lnl-3/igt@kms_big_fb@y-tiled-addfb-size-offset-overflow.html
* igt@kms_big_fb@y-tiled-max-hw-stride-32bpp-rotate-0-hflip-async-flip:
- shard-lnl: NOTRUN -> [SKIP][8] ([Intel XE#1124]) +4 other tests skip
[8]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-lnl-5/igt@kms_big_fb@y-tiled-max-hw-stride-32bpp-rotate-0-hflip-async-flip.html
* igt@kms_big_fb@y-tiled-max-hw-stride-32bpp-rotate-180-async-flip:
- shard-bmg: NOTRUN -> [SKIP][9] ([Intel XE#1124]) +1 other test skip
[9]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-bmg-9/igt@kms_big_fb@y-tiled-max-hw-stride-32bpp-rotate-180-async-flip.html
* igt@kms_big_fb@yf-tiled-addfb-size-offset-overflow:
- shard-bmg: NOTRUN -> [SKIP][10] ([Intel XE#607] / [Intel XE#7361])
[10]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-bmg-9/igt@kms_big_fb@yf-tiled-addfb-size-offset-overflow.html
* igt@kms_bw@linear-tiling-3-displays-target-2160x1440p:
- shard-lnl: NOTRUN -> [SKIP][11] ([Intel XE#367])
[11]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-lnl-5/igt@kms_bw@linear-tiling-3-displays-target-2160x1440p.html
* igt@kms_ccs@bad-rotation-90-4-tiled-lnl-ccs@pipe-c-dp-2:
- shard-bmg: NOTRUN -> [SKIP][12] ([Intel XE#2652]) +8 other tests skip
[12]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-bmg-9/igt@kms_ccs@bad-rotation-90-4-tiled-lnl-ccs@pipe-c-dp-2.html
* igt@kms_ccs@crc-primary-basic-4-tiled-mtl-rc-ccs:
- shard-lnl: NOTRUN -> [SKIP][13] ([Intel XE#2887]) +6 other tests skip
[13]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-lnl-5/igt@kms_ccs@crc-primary-basic-4-tiled-mtl-rc-ccs.html
* igt@kms_ccs@crc-primary-suspend-4-tiled-bmg-ccs:
- shard-bmg: [PASS][14] -> [INCOMPLETE][15] ([Intel XE#7084]) +1 other test incomplete
[14]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-4976-5e3e393d3eab6774e7aa49acd82f4bba2d8ef6a5/shard-bmg-5/igt@kms_ccs@crc-primary-suspend-4-tiled-bmg-ccs.html
[15]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-bmg-6/igt@kms_ccs@crc-primary-suspend-4-tiled-bmg-ccs.html
* igt@kms_ccs@crc-sprite-planes-basic-4-tiled-dg2-rc-ccs:
- shard-bmg: NOTRUN -> [SKIP][16] ([Intel XE#2887]) +2 other tests skip
[16]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-bmg-9/igt@kms_ccs@crc-sprite-planes-basic-4-tiled-dg2-rc-ccs.html
* igt@kms_chamelium_audio@dp-audio-edid:
- shard-lnl: NOTRUN -> [SKIP][17] ([Intel XE#373]) +2 other tests skip
[17]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-lnl-3/igt@kms_chamelium_audio@dp-audio-edid.html
* igt@kms_chamelium_color@ctm-0-25:
- shard-lnl: NOTRUN -> [SKIP][18] ([Intel XE#306] / [Intel XE#7358])
[18]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-lnl-5/igt@kms_chamelium_color@ctm-0-25.html
* igt@kms_chamelium_color@ctm-limited-range:
- shard-bmg: NOTRUN -> [SKIP][19] ([Intel XE#2325] / [Intel XE#7358])
[19]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-bmg-9/igt@kms_chamelium_color@ctm-limited-range.html
* igt@kms_chamelium_hpd@dp-hpd-storm-disable:
- shard-bmg: NOTRUN -> [SKIP][20] ([Intel XE#2252])
[20]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-bmg-9/igt@kms_chamelium_hpd@dp-hpd-storm-disable.html
* igt@kms_content_protection@atomic-dpms:
- shard-lnl: NOTRUN -> [SKIP][21] ([Intel XE#7642])
[21]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-lnl-3/igt@kms_content_protection@atomic-dpms.html
* igt@kms_cursor_crc@cursor-offscreen-max-size:
- shard-bmg: NOTRUN -> [SKIP][22] ([Intel XE#2320])
[22]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-bmg-9/igt@kms_cursor_crc@cursor-offscreen-max-size.html
* igt@kms_cursor_crc@cursor-random-128x42:
- shard-lnl: NOTRUN -> [SKIP][23] ([Intel XE#1424])
[23]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-lnl-5/igt@kms_cursor_crc@cursor-random-128x42.html
* igt@kms_cursor_legacy@cursorb-vs-flipa-legacy:
- shard-lnl: NOTRUN -> [SKIP][24] ([Intel XE#309] / [Intel XE#7343]) +2 other tests skip
[24]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-lnl-5/igt@kms_cursor_legacy@cursorb-vs-flipa-legacy.html
* igt@kms_dp_link_training@uhbr-sst:
- shard-lnl: NOTRUN -> [SKIP][25] ([Intel XE#4354] / [Intel XE#5870])
[25]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-lnl-3/igt@kms_dp_link_training@uhbr-sst.html
* igt@kms_dp_linktrain_fallback@dsc-fallback:
- shard-lnl: NOTRUN -> [SKIP][26] ([Intel XE#4331] / [Intel XE#7227])
[26]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-lnl-5/igt@kms_dp_linktrain_fallback@dsc-fallback.html
* igt@kms_dsc@dsc-basic:
- shard-bmg: NOTRUN -> [SKIP][27] ([Intel XE#2244])
[27]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-bmg-9/igt@kms_dsc@dsc-basic.html
* igt@kms_flip@2x-flip-vs-suspend-interruptible:
- shard-lnl: NOTRUN -> [SKIP][28] ([Intel XE#1421]) +1 other test skip
[28]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-lnl-5/igt@kms_flip@2x-flip-vs-suspend-interruptible.html
* igt@kms_flip@flip-vs-expired-vblank@a-edp1:
- shard-lnl: NOTRUN -> [FAIL][29] ([Intel XE#301]) +1 other test fail
[29]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-lnl-3/igt@kms_flip@flip-vs-expired-vblank@a-edp1.html
* igt@kms_flip_scaled_crc@flip-64bpp-4tile-to-32bpp-4tile-downscaling@pipe-a-default-mode:
- shard-lnl: NOTRUN -> [SKIP][30] ([Intel XE#1397] / [Intel XE#7385]) +1 other test skip
[30]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-lnl-5/igt@kms_flip_scaled_crc@flip-64bpp-4tile-to-32bpp-4tile-downscaling@pipe-a-default-mode.html
* igt@kms_flip_scaled_crc@flip-64bpp-linear-to-16bpp-linear-downscaling:
- shard-lnl: NOTRUN -> [SKIP][31] ([Intel XE#1397] / [Intel XE#1745] / [Intel XE#7385]) +1 other test skip
[31]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-lnl-3/igt@kms_flip_scaled_crc@flip-64bpp-linear-to-16bpp-linear-downscaling.html
* igt@kms_flip_scaled_crc@flip-64bpp-yftile-to-16bpp-yftile-downscaling:
- shard-lnl: NOTRUN -> [SKIP][32] ([Intel XE#7178] / [Intel XE#7351])
[32]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-lnl-3/igt@kms_flip_scaled_crc@flip-64bpp-yftile-to-16bpp-yftile-downscaling.html
* igt@kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytile-downscaling:
- shard-bmg: NOTRUN -> [SKIP][33] ([Intel XE#7178] / [Intel XE#7351]) +1 other test skip
[33]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-bmg-9/igt@kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytile-downscaling.html
* igt@kms_force_connector_basic@force-edid:
- shard-lnl: NOTRUN -> [SKIP][34] ([Intel XE#352])
[34]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-lnl-5/igt@kms_force_connector_basic@force-edid.html
* igt@kms_frontbuffer_tracking@drrs-2p-primscrn-pri-indfb-draw-mmap-wc:
- shard-bmg: NOTRUN -> [SKIP][35] ([Intel XE#2311]) +4 other tests skip
[35]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-bmg-9/igt@kms_frontbuffer_tracking@drrs-2p-primscrn-pri-indfb-draw-mmap-wc.html
* igt@kms_frontbuffer_tracking@drrs-argb161616f-draw-mmap-wc:
- shard-bmg: NOTRUN -> [SKIP][36] ([Intel XE#7061] / [Intel XE#7356]) +1 other test skip
[36]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-bmg-9/igt@kms_frontbuffer_tracking@drrs-argb161616f-draw-mmap-wc.html
* igt@kms_frontbuffer_tracking@drrs-rgb101010-draw-mmap-wc:
- shard-lnl: NOTRUN -> [SKIP][37] ([Intel XE#6312] / [Intel XE#651]) +3 other tests skip
[37]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-lnl-5/igt@kms_frontbuffer_tracking@drrs-rgb101010-draw-mmap-wc.html
* igt@kms_frontbuffer_tracking@fbc-1p-offscreen-pri-shrfb-draw-blt:
- shard-bmg: NOTRUN -> [SKIP][38] ([Intel XE#4141]) +3 other tests skip
[38]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-bmg-9/igt@kms_frontbuffer_tracking@fbc-1p-offscreen-pri-shrfb-draw-blt.html
* igt@kms_frontbuffer_tracking@fbc-argb161616f-draw-mmap-wc:
- shard-lnl: NOTRUN -> [SKIP][39] ([Intel XE#7061] / [Intel XE#7356])
[39]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-lnl-3/igt@kms_frontbuffer_tracking@fbc-argb161616f-draw-mmap-wc.html
* igt@kms_frontbuffer_tracking@fbcdrrs-1p-offscreen-pri-shrfb-draw-mmap-wc:
- shard-lnl: NOTRUN -> [SKIP][40] ([Intel XE#6312])
[40]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-lnl-5/igt@kms_frontbuffer_tracking@fbcdrrs-1p-offscreen-pri-shrfb-draw-mmap-wc.html
* igt@kms_frontbuffer_tracking@psr-1p-primscrn-cur-indfb-draw-render:
- shard-bmg: NOTRUN -> [SKIP][41] ([Intel XE#2313]) +3 other tests skip
[41]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-bmg-9/igt@kms_frontbuffer_tracking@psr-1p-primscrn-cur-indfb-draw-render.html
* igt@kms_frontbuffer_tracking@psr-2p-scndscrn-pri-indfb-draw-render:
- shard-lnl: NOTRUN -> [SKIP][42] ([Intel XE#5812] / [Intel XE#656]) +19 other tests skip
[42]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-lnl-5/igt@kms_frontbuffer_tracking@psr-2p-scndscrn-pri-indfb-draw-render.html
* igt@kms_hdr@invalid-metadata-sizes:
- shard-lnl: NOTRUN -> [SKIP][43] ([Intel XE#1503])
[43]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-lnl-3/igt@kms_hdr@invalid-metadata-sizes.html
* igt@kms_joiner@basic-force-ultra-joiner:
- shard-lnl: NOTRUN -> [SKIP][44] ([Intel XE#6900] / [Intel XE#7362])
[44]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-lnl-5/igt@kms_joiner@basic-force-ultra-joiner.html
* igt@kms_pipe_stress@stress-xrgb8888-yftiled:
- shard-lnl: NOTRUN -> [SKIP][45] ([Intel XE#6912] / [Intel XE#7375])
[45]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-lnl-3/igt@kms_pipe_stress@stress-xrgb8888-yftiled.html
* igt@kms_plane@pixel-format-4-tiled-dg2-mc-ccs-modifier:
- shard-lnl: NOTRUN -> [SKIP][46] ([Intel XE#7283]) +1 other test skip
[46]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-lnl-5/igt@kms_plane@pixel-format-4-tiled-dg2-mc-ccs-modifier.html
* igt@kms_plane@pixel-format-4-tiled-lnl-ccs-modifier@pipe-a-plane-5:
- shard-lnl: NOTRUN -> [SKIP][47] ([Intel XE#7130]) +1 other test skip
[47]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-lnl-3/igt@kms_plane@pixel-format-4-tiled-lnl-ccs-modifier@pipe-a-plane-5.html
* igt@kms_plane@pixel-format-4-tiled-mtl-mc-ccs-modifier-source-clamping:
- shard-bmg: NOTRUN -> [SKIP][48] ([Intel XE#7283]) +1 other test skip
[48]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-bmg-9/igt@kms_plane@pixel-format-4-tiled-mtl-mc-ccs-modifier-source-clamping.html
* igt@kms_psr2_sf@pr-cursor-plane-move-continuous-exceed-fully-sf:
- shard-bmg: NOTRUN -> [SKIP][49] ([Intel XE#1489]) +2 other tests skip
[49]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-bmg-9/igt@kms_psr2_sf@pr-cursor-plane-move-continuous-exceed-fully-sf.html
* igt@kms_psr2_sf@pr-overlay-primary-update-sf-dmg-area:
- shard-lnl: NOTRUN -> [SKIP][50] ([Intel XE#2893] / [Intel XE#7304]) +1 other test skip
[50]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-lnl-3/igt@kms_psr2_sf@pr-overlay-primary-update-sf-dmg-area.html
* igt@kms_psr@fbc-pr-dpms:
- shard-lnl: NOTRUN -> [SKIP][51] ([Intel XE#1406]) +1 other test skip
[51]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-lnl-3/igt@kms_psr@fbc-pr-dpms.html
* igt@kms_psr@fbc-psr2-primary-page-flip:
- shard-lnl: NOTRUN -> [SKIP][52] ([Intel XE#1406] / [Intel XE#7345])
[52]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-lnl-3/igt@kms_psr@fbc-psr2-primary-page-flip.html
* igt@kms_psr@fbc-psr2-primary-page-flip@edp-1:
- shard-lnl: NOTRUN -> [SKIP][53] ([Intel XE#1406] / [Intel XE#4609])
[53]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-lnl-3/igt@kms_psr@fbc-psr2-primary-page-flip@edp-1.html
* igt@kms_psr@fbc-psr2-sprite-render:
- shard-bmg: NOTRUN -> [SKIP][54] ([Intel XE#2234] / [Intel XE#2850]) +3 other tests skip
[54]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-bmg-9/igt@kms_psr@fbc-psr2-sprite-render.html
* igt@kms_rotation_crc@bad-tiling:
- shard-lnl: NOTRUN -> [SKIP][55] ([Intel XE#3414] / [Intel XE#3904] / [Intel XE#7342]) +1 other test skip
[55]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-lnl-5/igt@kms_rotation_crc@bad-tiling.html
* igt@kms_rotation_crc@primary-yf-tiled-reflect-x-270:
- shard-bmg: NOTRUN -> [SKIP][56] ([Intel XE#3904] / [Intel XE#7342])
[56]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-bmg-9/igt@kms_rotation_crc@primary-yf-tiled-reflect-x-270.html
* igt@kms_setmode@invalid-clone-exclusive-crtc:
- shard-lnl: NOTRUN -> [SKIP][57] ([Intel XE#1435])
[57]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-lnl-3/igt@kms_setmode@invalid-clone-exclusive-crtc.html
* igt@kms_sharpness_filter@filter-formats:
- shard-bmg: NOTRUN -> [SKIP][58] ([Intel XE#6503])
[58]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-bmg-9/igt@kms_sharpness_filter@filter-formats.html
* igt@kms_vrr@seamless-rr-switch-drrs:
- shard-lnl: NOTRUN -> [SKIP][59] ([Intel XE#1499])
[59]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-lnl-5/igt@kms_vrr@seamless-rr-switch-drrs.html
* igt@xe_eudebug@basic-connect:
- shard-lnl: NOTRUN -> [SKIP][60] ([Intel XE#7636]) +4 other tests skip
[60]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-lnl-3/igt@xe_eudebug@basic-connect.html
* igt@xe_eudebug@vm-bind-clear:
- shard-bmg: NOTRUN -> [SKIP][61] ([Intel XE#7636]) +2 other tests skip
[61]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-bmg-9/igt@xe_eudebug@vm-bind-clear.html
* igt@xe_evict@evict-small-multi-queue-priority-cm:
- shard-bmg: NOTRUN -> [SKIP][62] ([Intel XE#7140])
[62]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-bmg-9/igt@xe_evict@evict-small-multi-queue-priority-cm.html
* igt@xe_evict_ccs@evict-overcommit-simple:
- shard-lnl: NOTRUN -> [SKIP][63] ([Intel XE#6540] / [Intel XE#688]) +4 other tests skip
[63]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-lnl-5/igt@xe_evict_ccs@evict-overcommit-simple.html
* igt@xe_exec_balancer@once-parallel-rebind:
- shard-lnl: NOTRUN -> [SKIP][64] ([Intel XE#7482]) +8 other tests skip
[64]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-lnl-3/igt@xe_exec_balancer@once-parallel-rebind.html
* igt@xe_exec_basic@multigpu-no-exec-basic-defer-bind:
- shard-lnl: NOTRUN -> [SKIP][65] ([Intel XE#1392]) +1 other test skip
[65]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-lnl-3/igt@xe_exec_basic@multigpu-no-exec-basic-defer-bind.html
* igt@xe_exec_basic@multigpu-no-exec-bindexecqueue:
- shard-bmg: NOTRUN -> [SKIP][66] ([Intel XE#2322] / [Intel XE#7372])
[66]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-bmg-9/igt@xe_exec_basic@multigpu-no-exec-bindexecqueue.html
* igt@xe_exec_basic@once-bindexecqueue-userptr-invalidate-race:
- shard-bmg: [PASS][67] -> [SKIP][68] ([Intel XE#6703]) +32 other tests skip
[67]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-4976-5e3e393d3eab6774e7aa49acd82f4bba2d8ef6a5/shard-bmg-5/igt@xe_exec_basic@once-bindexecqueue-userptr-invalidate-race.html
[68]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-bmg-2/igt@xe_exec_basic@once-bindexecqueue-userptr-invalidate-race.html
* igt@xe_exec_fault_mode@many-execqueues-multi-queue-invalid-fault:
- shard-lnl: NOTRUN -> [SKIP][69] ([Intel XE#7136]) +5 other tests skip
[69]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-lnl-5/igt@xe_exec_fault_mode@many-execqueues-multi-queue-invalid-fault.html
* igt@xe_exec_fault_mode@twice-multi-queue-prefetch:
- shard-bmg: NOTRUN -> [SKIP][70] ([Intel XE#7136]) +1 other test skip
[70]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-bmg-9/igt@xe_exec_fault_mode@twice-multi-queue-prefetch.html
* igt@xe_exec_multi_queue@many-execs-priority:
- shard-lnl: NOTRUN -> [SKIP][71] ([Intel XE#6874]) +13 other tests skip
[71]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-lnl-3/igt@xe_exec_multi_queue@many-execs-priority.html
* igt@xe_exec_multi_queue@two-queues-dyn-priority:
- shard-bmg: NOTRUN -> [SKIP][72] ([Intel XE#6874]) +5 other tests skip
[72]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-bmg-9/igt@xe_exec_multi_queue@two-queues-dyn-priority.html
* igt@xe_exec_reset@cm-multi-queue-close-fd:
- shard-lnl: NOTRUN -> [SKIP][73] ([Intel XE#7866])
[73]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-lnl-3/igt@xe_exec_reset@cm-multi-queue-close-fd.html
* igt@xe_exec_threads@threads-multi-queue-cm-shared-vm-basic:
- shard-bmg: NOTRUN -> [SKIP][74] ([Intel XE#7138]) +1 other test skip
[74]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-bmg-9/igt@xe_exec_threads@threads-multi-queue-cm-shared-vm-basic.html
* igt@xe_exec_threads@threads-multi-queue-shared-vm-basic:
- shard-lnl: NOTRUN -> [SKIP][75] ([Intel XE#7138]) +4 other tests skip
[75]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-lnl-5/igt@xe_exec_threads@threads-multi-queue-shared-vm-basic.html
* igt@xe_live_ktest@xe_migrate@xe_validate_ccs_kunit:
- shard-lnl: NOTRUN -> [SKIP][76] ([Intel XE#2229])
[76]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-lnl-5/igt@xe_live_ktest@xe_migrate@xe_validate_ccs_kunit.html
* igt@xe_mmap@pci-membarrier:
- shard-lnl: NOTRUN -> [SKIP][77] ([Intel XE#5100] / [Intel XE#7322] / [Intel XE#7408])
[77]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-lnl-3/igt@xe_mmap@pci-membarrier.html
* igt@xe_multigpu_svm@mgpu-atomic-op-conflict:
- shard-lnl: NOTRUN -> [SKIP][78] ([Intel XE#6964]) +2 other tests skip
[78]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-lnl-3/igt@xe_multigpu_svm@mgpu-atomic-op-conflict.html
* igt@xe_page_reclaim@binds-full-pd:
- shard-lnl: NOTRUN -> [SKIP][79] ([Intel XE#7793])
[79]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-lnl-3/igt@xe_page_reclaim@binds-full-pd.html
* igt@xe_pmu@fn-engine-activity-load:
- shard-lnl: NOTRUN -> [SKIP][80] ([Intel XE#4650] / [Intel XE#7347])
[80]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-lnl-3/igt@xe_pmu@fn-engine-activity-load.html
* igt@xe_pxp@pxp-optout:
- shard-bmg: NOTRUN -> [SKIP][81] ([Intel XE#4733] / [Intel XE#7417])
[81]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-bmg-9/igt@xe_pxp@pxp-optout.html
* igt@xe_query@multigpu-query-topology-l3-bank-mask:
- shard-lnl: NOTRUN -> [SKIP][82] ([Intel XE#944])
[82]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-lnl-3/igt@xe_query@multigpu-query-topology-l3-bank-mask.html
* igt@xe_sriov_admin@default-sched-attributes-vfs-disabled:
- shard-lnl: NOTRUN -> [SKIP][83] ([Intel XE#7174])
[83]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-lnl-3/igt@xe_sriov_admin@default-sched-attributes-vfs-disabled.html
* igt@xe_sriov_vram@vf-access-after-resize-down:
- shard-lnl: NOTRUN -> [SKIP][84] ([Intel XE#6376] / [Intel XE#7330] / [Intel XE#7422])
[84]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-lnl-5/igt@xe_sriov_vram@vf-access-after-resize-down.html
* igt@xe_vm@overcommit-nonfault-vram-no-lr:
- shard-lnl: NOTRUN -> [SKIP][85] ([Intel XE#7892])
[85]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-lnl-3/igt@xe_vm@overcommit-nonfault-vram-no-lr.html
#### Possible fixes ####
* igt@kms_addfb_basic@invalid-get-prop:
- shard-bmg: [DMESG-WARN][86] -> [PASS][87]
[86]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-4976-5e3e393d3eab6774e7aa49acd82f4bba2d8ef6a5/shard-bmg-2/igt@kms_addfb_basic@invalid-get-prop.html
[87]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-bmg-9/igt@kms_addfb_basic@invalid-get-prop.html
* igt@xe_evict@evict-mixed-many-threads-small:
- shard-bmg: [INCOMPLETE][88] ([Intel XE#6321]) -> [PASS][89]
[88]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-4976-5e3e393d3eab6774e7aa49acd82f4bba2d8ef6a5/shard-bmg-5/igt@xe_evict@evict-mixed-many-threads-small.html
[89]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-bmg-2/igt@xe_evict@evict-mixed-many-threads-small.html
#### Warnings ####
* igt@kms_ccs@crc-primary-suspend-y-tiled-gen12-rc-ccs-cc:
- shard-bmg: [SKIP][90] ([Intel XE#3432]) -> [SKIP][91] ([Intel XE#2231] / [Intel XE#6703])
[90]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-4976-5e3e393d3eab6774e7aa49acd82f4bba2d8ef6a5/shard-bmg-5/igt@kms_ccs@crc-primary-suspend-y-tiled-gen12-rc-ccs-cc.html
[91]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-bmg-2/igt@kms_ccs@crc-primary-suspend-y-tiled-gen12-rc-ccs-cc.html
* igt@kms_chamelium_hpd@hdmi-hpd-after-suspend:
- shard-bmg: [SKIP][92] ([Intel XE#2252]) -> [SKIP][93] ([Intel XE#6703])
[92]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-4976-5e3e393d3eab6774e7aa49acd82f4bba2d8ef6a5/shard-bmg-5/igt@kms_chamelium_hpd@hdmi-hpd-after-suspend.html
[93]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-bmg-2/igt@kms_chamelium_hpd@hdmi-hpd-after-suspend.html
* igt@kms_feature_discovery@display-3x:
- shard-bmg: [SKIP][94] ([Intel XE#2373] / [Intel XE#7448]) -> [SKIP][95] ([Intel XE#6703])
[94]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-4976-5e3e393d3eab6774e7aa49acd82f4bba2d8ef6a5/shard-bmg-5/igt@kms_feature_discovery@display-3x.html
[95]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-bmg-2/igt@kms_feature_discovery@display-3x.html
* igt@kms_frontbuffer_tracking@drrs-2p-primscrn-pri-indfb-draw-blt:
- shard-bmg: [SKIP][96] ([Intel XE#2311]) -> [SKIP][97] ([Intel XE#2231] / [Intel XE#6703]) +2 other tests skip
[96]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-4976-5e3e393d3eab6774e7aa49acd82f4bba2d8ef6a5/shard-bmg-5/igt@kms_frontbuffer_tracking@drrs-2p-primscrn-pri-indfb-draw-blt.html
[97]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-bmg-2/igt@kms_frontbuffer_tracking@drrs-2p-primscrn-pri-indfb-draw-blt.html
* igt@kms_frontbuffer_tracking@fbc-2p-scndscrn-spr-indfb-onoff:
- shard-bmg: [SKIP][98] ([Intel XE#4141]) -> [SKIP][99] ([Intel XE#2231] / [Intel XE#6703]) +1 other test skip
[98]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-4976-5e3e393d3eab6774e7aa49acd82f4bba2d8ef6a5/shard-bmg-5/igt@kms_frontbuffer_tracking@fbc-2p-scndscrn-spr-indfb-onoff.html
[99]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-bmg-2/igt@kms_frontbuffer_tracking@fbc-2p-scndscrn-spr-indfb-onoff.html
* igt@kms_frontbuffer_tracking@fbcpsr-1p-primscrn-indfb-plflip-blt:
- shard-bmg: [SKIP][100] ([Intel XE#2313]) -> [SKIP][101] ([Intel XE#2231] / [Intel XE#6703]) +1 other test skip
[100]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-4976-5e3e393d3eab6774e7aa49acd82f4bba2d8ef6a5/shard-bmg-5/igt@kms_frontbuffer_tracking@fbcpsr-1p-primscrn-indfb-plflip-blt.html
[101]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-bmg-2/igt@kms_frontbuffer_tracking@fbcpsr-1p-primscrn-indfb-plflip-blt.html
* igt@kms_psr@fbc-pr-no-drrs:
- shard-bmg: [SKIP][102] ([Intel XE#2234] / [Intel XE#2850]) -> [SKIP][103] ([Intel XE#2231] / [Intel XE#6703])
[102]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-4976-5e3e393d3eab6774e7aa49acd82f4bba2d8ef6a5/shard-bmg-5/igt@kms_psr@fbc-pr-no-drrs.html
[103]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-bmg-2/igt@kms_psr@fbc-pr-no-drrs.html
* igt@kms_rotation_crc@primary-rotation-270:
- shard-bmg: [SKIP][104] ([Intel XE#3904] / [Intel XE#7342]) -> [SKIP][105] ([Intel XE#6703])
[104]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-4976-5e3e393d3eab6774e7aa49acd82f4bba2d8ef6a5/shard-bmg-5/igt@kms_rotation_crc@primary-rotation-270.html
[105]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-bmg-2/igt@kms_rotation_crc@primary-rotation-270.html
* igt@kms_tiled_display@basic-test-pattern:
- shard-bmg: [FAIL][106] ([Intel XE#1729] / [Intel XE#7424]) -> [SKIP][107] ([Intel XE#2426] / [Intel XE#5848])
[106]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-4976-5e3e393d3eab6774e7aa49acd82f4bba2d8ef6a5/shard-bmg-5/igt@kms_tiled_display@basic-test-pattern.html
[107]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-bmg-7/igt@kms_tiled_display@basic-test-pattern.html
* igt@xe_eudebug_online@writes-caching-sram-bb-sram-target-vram:
- shard-bmg: [SKIP][108] ([Intel XE#7636]) -> [SKIP][109] ([Intel XE#6703])
[108]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-4976-5e3e393d3eab6774e7aa49acd82f4bba2d8ef6a5/shard-bmg-5/igt@xe_eudebug_online@writes-caching-sram-bb-sram-target-vram.html
[109]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-bmg-2/igt@xe_eudebug_online@writes-caching-sram-bb-sram-target-vram.html
* igt@xe_exec_multi_queue@few-execs-preempt-mode-fault-userptr:
- shard-bmg: [SKIP][110] ([Intel XE#6874]) -> [SKIP][111] ([Intel XE#6703]) +2 other tests skip
[110]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-4976-5e3e393d3eab6774e7aa49acd82f4bba2d8ef6a5/shard-bmg-5/igt@xe_exec_multi_queue@few-execs-preempt-mode-fault-userptr.html
[111]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-bmg-2/igt@xe_exec_multi_queue@few-execs-preempt-mode-fault-userptr.html
* igt@xe_query@multigpu-query-engines:
- shard-bmg: [SKIP][112] ([Intel XE#944]) -> [SKIP][113] ([Intel XE#6703])
[112]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-4976-5e3e393d3eab6774e7aa49acd82f4bba2d8ef6a5/shard-bmg-5/igt@xe_query@multigpu-query-engines.html
[113]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/shard-bmg-2/igt@xe_query@multigpu-query-engines.html
[Intel XE#1124]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1124
[Intel XE#1392]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1392
[Intel XE#1397]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1397
[Intel XE#1406]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1406
[Intel XE#1407]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1407
[Intel XE#1421]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1421
[Intel XE#1424]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1424
[Intel XE#1435]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1435
[Intel XE#1477]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1477
[Intel XE#1489]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1489
[Intel XE#1499]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1499
[Intel XE#1503]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1503
[Intel XE#1729]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1729
[Intel XE#1745]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1745
[Intel XE#2229]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2229
[Intel XE#2231]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2231
[Intel XE#2234]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2234
[Intel XE#2244]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2244
[Intel XE#2252]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2252
[Intel XE#2311]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2311
[Intel XE#2313]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2313
[Intel XE#2320]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2320
[Intel XE#2322]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2322
[Intel XE#2325]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2325
[Intel XE#2327]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2327
[Intel XE#2373]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2373
[Intel XE#2426]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2426
[Intel XE#2652]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2652
[Intel XE#2850]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2850
[Intel XE#2887]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2887
[Intel XE#2893]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2893
[Intel XE#301]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/301
[Intel XE#306]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/306
[Intel XE#309]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/309
[Intel XE#3414]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/3414
[Intel XE#3432]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/3432
[Intel XE#352]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/352
[Intel XE#367]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/367
[Intel XE#373]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/373
[Intel XE#3904]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/3904
[Intel XE#4141]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/4141
[Intel XE#4331]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/4331
[Intel XE#4354]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/4354
[Intel XE#4609]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/4609
[Intel XE#4650]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/4650
[Intel XE#4733]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/4733
[Intel XE#4760]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/4760
[Intel XE#5100]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/5100
[Intel XE#5812]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/5812
[Intel XE#5848]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/5848
[Intel XE#5870]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/5870
[Intel XE#607]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/607
[Intel XE#6312]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/6312
[Intel XE#6321]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/6321
[Intel XE#6376]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/6376
[Intel XE#6503]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/6503
[Intel XE#651]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/651
[Intel XE#6540]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/6540
[Intel XE#656]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/656
[Intel XE#6703]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/6703
[Intel XE#6874]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/6874
[Intel XE#688]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/688
[Intel XE#6900]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/6900
[Intel XE#6912]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/6912
[Intel XE#6964]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/6964
[Intel XE#7061]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7061
[Intel XE#7084]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7084
[Intel XE#7130]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7130
[Intel XE#7136]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7136
[Intel XE#7138]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7138
[Intel XE#7140]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7140
[Intel XE#7174]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7174
[Intel XE#7178]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7178
[Intel XE#7227]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7227
[Intel XE#7283]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7283
[Intel XE#7304]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7304
[Intel XE#7322]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7322
[Intel XE#7330]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7330
[Intel XE#7342]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7342
[Intel XE#7343]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7343
[Intel XE#7345]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7345
[Intel XE#7347]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7347
[Intel XE#7351]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7351
[Intel XE#7356]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7356
[Intel XE#7358]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7358
[Intel XE#7361]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7361
[Intel XE#7362]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7362
[Intel XE#7372]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7372
[Intel XE#7375]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7375
[Intel XE#7385]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7385
[Intel XE#7408]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7408
[Intel XE#7417]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7417
[Intel XE#7422]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7422
[Intel XE#7424]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7424
[Intel XE#7448]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7448
[Intel XE#7482]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7482
[Intel XE#7636]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7636
[Intel XE#7642]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7642
[Intel XE#7793]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7793
[Intel XE#7866]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7866
[Intel XE#7892]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7892
[Intel XE#944]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/944
Build changes
-------------
* Linux: xe-4976-5e3e393d3eab6774e7aa49acd82f4bba2d8ef6a5 -> xe-pw-165893v1
IGT_8882: 3ef61dd87d23e6c8d7956b5a72ba01a6597a3fd1 @ https://gitlab.freedesktop.org/drm/igt-gpu-tools.git
xe-4976-5e3e393d3eab6774e7aa49acd82f4bba2d8ef6a5: 5e3e393d3eab6774e7aa49acd82f4bba2d8ef6a5
xe-pw-165893v1: 165893v1
== Logs ==
For more details see: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-165893v1/index.html
[-- Attachment #2: Type: text/html, Size: 43575 bytes --]
^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [PATCH v2 1/2] gpu/buddy: replace dual-tree/force_merge with decoupled clear tracker
2026-05-04 11:10 [PATCH v2 1/2] gpu/buddy: replace dual-tree/force_merge with decoupled clear tracker Arunpravin Paneer Selvam
` (4 preceding siblings ...)
2026-05-04 19:24 ` ✓ Xe.CI.FULL: " Patchwork
@ 2026-05-08 15:19 ` Matthew Auld
2026-05-08 16:26 ` Arunpravin Paneer Selvam
5 siblings, 1 reply; 8+ messages in thread
From: Matthew Auld @ 2026-05-08 15:19 UTC (permalink / raw)
To: Arunpravin Paneer Selvam, christian.koenig, dri-devel, intel-gfx,
intel-xe, amd-gfx
Cc: alexander.deucher
On 04/05/2026 12:10, Arunpravin Paneer Selvam wrote:
> The current buddy allocator maintains separate clear_tree[] and
> dirty_tree[] rbtrees per order, preventing coalescing between cleared
> and dirty buddies. Under mixed workloads, this creates a merge barrier:
> adjacent buddies frequently end up split across trees, forcing reliance
> on __force_merge() during allocation.
>
> __force_merge() performs an O(N x max_order) scan under the VRAM manager
> lock, leading to allocation stalls and failures for large contiguous
> requests even when sufficient total free memory is available.
>
> Solution
>
> Replace the dual-tree design with:
> - A single free_tree[order] rbtree for dirty and mixed free blocks
> (fully cleared free blocks float outside this tree)
> - A lightweight out-of-band clear tracker (gpu_clear_tracker)
>
> Fully cleared free blocks are tracked outside the buddy trees using an
> augmented interval rbtree, enabling O(log E) lookup of the largest
> cleared extents.
>
> Buddy coalescing is now unconditional in __gpu_buddy_free(), regardless
> of clear/dirty state. This removes the merge barrier and eliminates the
> need for __force_merge().
>
> Benefits
>
> - Correct high-order allocations after mixed clear/dirty workloads
> - Elimination of O(N x max_order) merge cost from the allocation path
> - O(log E) cleared-extent lookup replacing O(N) scans
> - Predictable allocation latency under fragmentation
> - Reduced complexity with a single tree per order
>
> Test:
> dEQP-VK.memory.allocation.basic.size_8KiB.reverse.count_4000
>
> Below data is from /sys/kernel/debug/dri/1/amdgpu_vram_mm:
>
> Base (dual-tree), before VKCTS test:
> order- 6 free: 6 MiB, blocks: 26
> order- 5 free: 1 MiB, blocks: 15
> order- 4 free: 960 KiB, blocks: 15
> order- 3 free: 5 MiB, blocks: 171
> order- 2 free: 2 MiB, blocks: 176
> order- 1 free: 1 MiB, blocks: 165
> order- 0 free: 16 KiB, blocks: 4
>
> Base (dual-tree), after VKCTS test:
> order- 6 free: 768 KiB, blocks: 3
> order- 5 free: 499 MiB, blocks: 3999
> order- 4 free: 250 MiB, blocks: 4001
> order- 3 free: 129 MiB, blocks: 4157
> order- 2 free: 65 MiB, blocks: 4161
> order- 1 free: 63 MiB, blocks: 8138
> order- 0 free: 20 KiB, blocks: 5
>
> Clear tracker, before VKCTS test:
> order- 6 free: 4 MiB, blocks: 19
> order- 5 free: 2 MiB, blocks: 18
> order- 4 free: 704 KiB, blocks: 11
> order- 3 free: 5 MiB, blocks: 168
> order- 2 free: 2 MiB, blocks: 174
> order- 1 free: 1 MiB, blocks: 167
> order- 0 free: 32 KiB, blocks: 8
>
> Clear tracker, after VKCTS test:
> order- 6 free: 4 MiB, blocks: 19
> order- 5 free: 2 MiB, blocks: 18
> order- 4 free: 704 KiB, blocks: 11
> order- 3 free: 5 MiB, blocks: 168
> order- 2 free: 2 MiB, blocks: 174
> order- 1 free: 1 MiB, blocks: 167
> order- 0 free: 28 KiB, blocks: 7
>
> v2:
> - Code-style cleanup and minor refactoring
> - Renamed locals for clarity
>
> Cc: Matthew Auld <matthew.auld@intel.com>
> Cc: Christian König <christian.koenig@amd.com>
> Signed-off-by: Arunpravin Paneer Selvam <Arunpravin.PaneerSelvam@amd.com>
Still need some more time to fully go over this, but in the meantime
there is some feedback here from sashiko, which might be worth a look:
https://sashiko.dev/#/patchset/20260504111055.262964-1-Arunpravin.PaneerSelvam%40amd.com
^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [PATCH v2 1/2] gpu/buddy: replace dual-tree/force_merge with decoupled clear tracker
2026-05-08 15:19 ` [PATCH v2 1/2] " Matthew Auld
@ 2026-05-08 16:26 ` Arunpravin Paneer Selvam
0 siblings, 0 replies; 8+ messages in thread
From: Arunpravin Paneer Selvam @ 2026-05-08 16:26 UTC (permalink / raw)
To: Matthew Auld, christian.koenig, dri-devel, intel-gfx, intel-xe,
amd-gfx
Cc: alexander.deucher
Hi Matthew,
On 5/8/2026 8:49 PM, Matthew Auld wrote:
> On 04/05/2026 12:10, Arunpravin Paneer Selvam wrote:
>> The current buddy allocator maintains separate clear_tree[] and
>> dirty_tree[] rbtrees per order, preventing coalescing between cleared
>> and dirty buddies. Under mixed workloads, this creates a merge barrier:
>> adjacent buddies frequently end up split across trees, forcing reliance
>> on __force_merge() during allocation.
>>
>> __force_merge() performs an O(N x max_order) scan under the VRAM manager
>> lock, leading to allocation stalls and failures for large contiguous
>> requests even when sufficient total free memory is available.
>>
>> Solution
>>
>> Replace the dual-tree design with:
>> - A single free_tree[order] rbtree for dirty and mixed free blocks
>> (fully cleared free blocks float outside this tree)
>> - A lightweight out-of-band clear tracker (gpu_clear_tracker)
>>
>> Fully cleared free blocks are tracked outside the buddy trees using an
>> augmented interval rbtree, enabling O(log E) lookup of the largest
>> cleared extents.
>>
>> Buddy coalescing is now unconditional in __gpu_buddy_free(), regardless
>> of clear/dirty state. This removes the merge barrier and eliminates the
>> need for __force_merge().
>>
>> Benefits
>>
>> - Correct high-order allocations after mixed clear/dirty workloads
>> - Elimination of O(N x max_order) merge cost from the allocation path
>> - O(log E) cleared-extent lookup replacing O(N) scans
>> - Predictable allocation latency under fragmentation
>> - Reduced complexity with a single tree per order
>>
>> Test:
>> dEQP-VK.memory.allocation.basic.size_8KiB.reverse.count_4000
>>
>> Below data is from /sys/kernel/debug/dri/1/amdgpu_vram_mm:
>>
>> Base (dual-tree), before VKCTS test:
>> order- 6 free: 6 MiB, blocks: 26
>> order- 5 free: 1 MiB, blocks: 15
>> order- 4 free: 960 KiB, blocks: 15
>> order- 3 free: 5 MiB, blocks: 171
>> order- 2 free: 2 MiB, blocks: 176
>> order- 1 free: 1 MiB, blocks: 165
>> order- 0 free: 16 KiB, blocks: 4
>>
>> Base (dual-tree), after VKCTS test:
>> order- 6 free: 768 KiB, blocks: 3
>> order- 5 free: 499 MiB, blocks: 3999
>> order- 4 free: 250 MiB, blocks: 4001
>> order- 3 free: 129 MiB, blocks: 4157
>> order- 2 free: 65 MiB, blocks: 4161
>> order- 1 free: 63 MiB, blocks: 8138
>> order- 0 free: 20 KiB, blocks: 5
>>
>> Clear tracker, before VKCTS test:
>> order- 6 free: 4 MiB, blocks: 19
>> order- 5 free: 2 MiB, blocks: 18
>> order- 4 free: 704 KiB, blocks: 11
>> order- 3 free: 5 MiB, blocks: 168
>> order- 2 free: 2 MiB, blocks: 174
>> order- 1 free: 1 MiB, blocks: 167
>> order- 0 free: 32 KiB, blocks: 8
>>
>> Clear tracker, after VKCTS test:
>> order- 6 free: 4 MiB, blocks: 19
>> order- 5 free: 2 MiB, blocks: 18
>> order- 4 free: 704 KiB, blocks: 11
>> order- 3 free: 5 MiB, blocks: 168
>> order- 2 free: 2 MiB, blocks: 174
>> order- 1 free: 1 MiB, blocks: 167
>> order- 0 free: 28 KiB, blocks: 7
>>
>> v2:
>> - Code-style cleanup and minor refactoring
>> - Renamed locals for clarity
>>
>> Cc: Matthew Auld <matthew.auld@intel.com>
>> Cc: Christian König <christian.koenig@amd.com>
>> Signed-off-by: Arunpravin Paneer Selvam
>> <Arunpravin.PaneerSelvam@amd.com>
>
> Still need some more time to fully go over this, but in the meantime
> there is some feedback here from sashiko, which might be worth a look:
>
> https://sashiko.dev/#/patchset/20260504111055.262964-1-Arunpravin.PaneerSelvam%40amd.com
>
I have sent the v3. Please go through it. I will check the Sashiko
review comments.
Regards,
Arun.
^ permalink raw reply [flat|nested] 8+ messages in thread
end of thread, other threads:[~2026-05-08 16:26 UTC | newest]
Thread overview: 8+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-05-04 11:10 [PATCH v2 1/2] gpu/buddy: replace dual-tree/force_merge with decoupled clear tracker Arunpravin Paneer Selvam
2026-05-04 11:10 ` [PATCH v2 2/2] gpu/tests/buddy: add clear-tracker allocation latency benchmarks Arunpravin Paneer Selvam
2026-05-04 14:48 ` ✗ CI.checkpatch: warning for series starting with [v2,1/2] gpu/buddy: replace dual-tree/force_merge with decoupled clear tracker Patchwork
2026-05-04 14:49 ` ✓ CI.KUnit: success " Patchwork
2026-05-04 15:43 ` ✓ Xe.CI.BAT: " Patchwork
2026-05-04 19:24 ` ✓ Xe.CI.FULL: " Patchwork
2026-05-08 15:19 ` [PATCH v2 1/2] " Matthew Auld
2026-05-08 16:26 ` Arunpravin Paneer Selvam
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox