We need to perform cross-batch flushing if any batch writes to a BO
while others refer to it. We checked this case when recording a new
BO in the list which we'd never seen before. However, we neglected to
handle the case when we already read from a BO, but then began writing
to it. That new write may provoke a conflict between existing reads
in other batches, so we need to re-check the cross-batch flushing.
Caught by Piglit's copyteximage when forcing blits and copies to use
a new IRIS_BATCH_BLITTER that isn't upstream yet. But this bug could
be provoked by render/compute work today...we just hadn't noticed it.
Fixes:
b21e916a628 ("iris: Combine iris_use_pinned_bo and add_exec_bo")
Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13808>
if (batch->measure && bo == batch->measure->bo)
return;
- /* This is the first time our batch has seen this BO. Before we use it,
- * we may need to flush and synchronize with other batches.
+ /* When a batch uses a buffer for the first time, or newly writes a buffer
+ * it had already referenced, we may need to flush other batches in order
+ * to correctly synchronize them.
*/
for (int b = 0; b < ARRAY_SIZE(batch->other_batches); b++) {
struct iris_batch *other_batch = batch->other_batches[b];
if (existing_index != -1) {
/* The BO is already in the list; mark it writable */
- if (writable)
+ if (writable && !BITSET_TEST(batch->bos_written, existing_index)) {
BITSET_SET(batch->bos_written, existing_index);
+ flush_for_cross_batch_dependencies(batch, bo, writable);
+ }
return;
}