fdo-mirrors/mesa
1
0
Fork 0
mirror of https://gitlab.freedesktop.org/mesa/mesa.git synced 2025-12-24 19:40:10 +01:00
Code Issues Projects Releases Packages Wiki Activity Actions 1

radeonsi: do not use staging texture for APU

Browse source 
This is to fix serious performance drop of texture_upload/
texture_resue relative items in chromeos glbench test.
Staging texture is not efficient for CPU uploading.

Signed-off-by: Jasber Chen <yipeng.chen@amd.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13306>
This commit is contained in:
Yipeng Chen (Jasber) 2021-10-12 16:37:24 +08:00 • committed by Marge Bot
parent 80efe661d7
commit 33484a327e
1 changed files with 2 additions and 2 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

4
src/gallium/drivers/radeonsi/si_texture.c
View file

@ -1771,7 +1771,7 @@ static void *si_texture_transfer_map(struct pipe_context *ctx, struct pipe_resou
/* Tiled textures need to be converted into a linear texture for CPU
* access. The staging texture is always linear and is placed in GART.
*
* Always use a staging texture for VRAM, so that we don't map it and
* dGPU use a staging texture for VRAM, so that we don't map it and
* don't relocate it to GTT.
*
* Reading from VRAM or GTT WC is slow, always use the staging
@ -1781,7 +1781,7 @@ static void *si_texture_transfer_map(struct pipe_context *ctx, struct pipe_resou
* is busy.
*/
if (!tex->surface.is_linear || (tex->buffer.flags & RADEON_FLAG_ENCRYPTED) ||
(tex->buffer.domains & RADEON_DOMAIN_VRAM &&
(tex->buffer.domains & RADEON_DOMAIN_VRAM && sctx->screen->info.has_dedicated_vram &&
!sctx->screen->info.smart_access_memory))
use_staging_texture = true;
else if (usage & PIPE_MAP_READ)