[gpu] Use vector.reset() not resize(0) in hb_gpu_paint_clear

Three accumulator vectors -- ops, sub_blobs, transform_stack --
can enter error state via failed push_or_fail() under
failing-malloc.  Once in error state, resize(0) short-circuits
at the in_error() guard in alloc() and leaves length unchanged.

In hb_gpu_paint_clear that means the next iteration's clear
walks the stale entries again.  For sub_blobs this is a
double-destroy: the blob pointers were already freed in the
previous clear, and hb_blob_destroy reads the dead ref_count
field -- heap-use-after-free.

Fix by calling vector.reset(), which clears the error state
before resizing to 0.  Matches the pattern hb_gpu_draw_clear
already uses for draw->curves.

Found by oss-fuzz/CIFuzz on hb-gpu-fuzzer; minimized testcase
added to the fuzz corpus.

Assisted-by: Claude <noreply@anthropic.com>
This commit is contained in:
Behdad Esfahbod committed 2026-04-14 21:49:12 -06:00
1 parent 24aa15d39a
commit df4be0cec8
2 files changed
+8 -3

No files matched your search

+8 -3
View File
@@ -1441,15 +1441,20 @@ hb_gpu_paint_encode (hb_gpu_paint_t *paint,
void
hb_gpu_paint_clear (hb_gpu_paint_t *paint)
{
paint->ops.resize (0);
/* Use reset() not resize(0): if an earlier push_or_fail() set
* the vector's error state (common under failing-malloc),
* resize(0) short-circuits without clearing length. A
* subsequent clear would double-destroy the blob pointers left
* in the array, or keep stale ops across encodes. */
paint->ops.reset ();
paint->num_ops = 0;
for (hb_blob_t *b : paint->sub_blobs)
hb_blob_destroy (b);
paint->sub_blobs.resize (0);
paint->sub_blobs.reset ();
paint->clip_depth = 0;
paint->unsupported = false;
paint->cur_transform = {1, 0, 0, 1, 0, 0};
paint->transform_stack.resize (0);
paint->transform_stack.reset ();
paint->ext_min_x = 0x7fffffff;
paint->ext_min_y = 0x7fffffff;
paint->ext_max_x = -0x7fffffff;