Skip to content
Snippets Groups Projects
Commit 4bde7d3d authored by Marek Olšák's avatar Marek Olšák
Browse files

radeonsi: skip an unnecessary mutex lock for L2 prefetches


the mutex lock is inside util_range_add.

Reviewed-by: default avatarNicolai Hähnle <nicolai.haehnle@amd.com>
parent d93b0eac
No related branches found
No related tags found
No related merge requests found
......@@ -315,11 +315,13 @@ void si_copy_buffer(struct si_context *sctx,
if (!size)
return;
/* Mark the buffer range of destination as valid (initialized),
* so that transfer_map knows it should wait for the GPU when mapping
* that range. */
util_range_add(&r600_resource(dst)->valid_buffer_range, dst_offset,
dst_offset + size);
if (dst != src || dst_offset != src_offset) {
/* Mark the buffer range of destination as valid (initialized),
* so that transfer_map knows it should wait for the GPU when mapping
* that range. */
util_range_add(&r600_resource(dst)->valid_buffer_range, dst_offset,
dst_offset + size);
}
dst_offset += r600_resource(dst)->gpu_address;
src_offset += r600_resource(src)->gpu_address;
......
0% Loading or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment