radeonsi: don't flush in fence_server_sync

This reverts commit 50b06cbc10 and fixes
an Android performance regression.

Fixes: 50b06cbc10 "radeonsi: fix fence_server_sync() holding up extra work v2"

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5602>
(cherry picked from commit 1401fc055c)
This commit is contained in:
Marek Olšák 2020-06-22 23:54:59 -04:00 committed by Eric Engestrom
parent facfe6c0d8
commit 06c2e4f1cd
2 changed files with 9 additions and 6 deletions

View file

@ -2920,7 +2920,7 @@
"description": "radeonsi: don't flush in fence_server_sync",
"nominated": true,
"nomination_type": 1,
"resolution": 0,
"resolution": 1,
"master_sha": null,
"because_sha": "50b06cbc10dbca1dfee89b529ba9b564cc4ea6f6"
},

View file

@ -589,13 +589,16 @@ static void si_fence_server_sync(struct pipe_context *ctx, struct pipe_fence_han
if (sfence->gfx_unflushed.ctx && sfence->gfx_unflushed.ctx == sctx)
return;
/* All unflushed commands will not start execution before
* this fence dependency is signalled.
/* All unflushed commands will not start execution before this fence
* dependency is signalled. That's fine. Flushing is very expensive
* if we get fence_server_sync after every draw call. (which happens
* with Android/SurfaceFlinger)
*
* Therefore we must flush before inserting the dependency
* In a nutshell, when CPU overhead is greater than GPU overhead,
* or when the time it takes to execute an IB on the GPU is less than
* the time it takes to create and submit that IB, flushing decreases
* performance. Therefore, DO NOT FLUSH.
*/
si_flush_from_st(ctx, NULL, PIPE_FLUSH_ASYNC);
if (sfence->sdma)
si_add_fence_dependency(sctx, sfence->sdma);
if (sfence->gfx)