freedreno: proper locking for iterating dependent batches

In transfer_map(), when we need to flush batches that read from a
resource, we should be holding screen->lock to guard against race
conditions.  Somehow deferred flush seems to make this existing
race more obvious.

Signed-off-by: Rob Clark <robdclark@gmail.com>
This commit is contained in:
Rob Clark 2017-11-20 09:52:04 -05:00
parent ef6313ffd3
commit deb57fb237
2 changed files with 20 additions and 8 deletions

View File

@ -66,7 +66,7 @@ struct fd_hw_sample;
struct fd_batch {
struct pipe_reference reference;
unsigned seqno;
unsigned idx;
unsigned idx; /* index into cache->batches[] */
int in_fence_fd;
bool needs_out_fence_fd;

View File

@ -531,14 +531,26 @@ fd_resource_transfer_map(struct pipe_context *pctx,
if (needs_flush) {
if (usage & PIPE_TRANSFER_WRITE) {
struct fd_batch *batch, *last_batch = NULL;
foreach_batch(batch, &ctx->screen->batch_cache, rsc->batch_mask) {
fd_batch_reference(&last_batch, batch);
struct fd_batch *batch, *batches[32] = {0};
uint32_t batch_mask;
/* This is a bit awkward, probably a fd_batch_flush_locked()
* would make things simpler.. but we need to hold the lock
* to iterate the batches which reference this resource. So
* we must first grab references under a lock, then flush.
*/
mtx_lock(&ctx->screen->lock);
batch_mask = rsc->batch_mask;
foreach_batch(batch, &ctx->screen->batch_cache, batch_mask)
fd_batch_reference(&batches[batch->idx], batch);
mtx_unlock(&ctx->screen->lock);
foreach_batch(batch, &ctx->screen->batch_cache, batch_mask)
fd_batch_flush(batch, false);
}
if (last_batch) {
fd_batch_sync(last_batch);
fd_batch_reference(&last_batch, NULL);
foreach_batch(batch, &ctx->screen->batch_cache, batch_mask) {
fd_batch_sync(batch);
fd_batch_reference(&batches[batch->idx], NULL);
}
assert(rsc->batch_mask == 0);
} else {