Mesa (master): radeonsi: fix the VGT performance tweak for small instances
Marek Olšák
mareko at kemper.freedesktop.org
Sat Sep 10 09:55:46 UTC 2016
Module: Mesa
Branch: master
Commit: a5a2cc530c1f493557c232557ad1910e607712c2
URL: http://cgit.freedesktop.org/mesa/mesa/commit/?id=a5a2cc530c1f493557c232557ad1910e607712c2
Author: Marek Olšák <marek.olsak at amd.com>
Date: Thu Sep 8 01:42:06 2016 +0200
radeonsi: fix the VGT performance tweak for small instances
Based on the VGT spec.
The Vulkan driver doesn't do it optimally and they plan to fix it.
Reviewed-by: Nicolai Hähnle <nicolai.haehnle at amd.com>
---
src/gallium/drivers/radeonsi/si_state_draw.c | 11 ++++++-----
1 file changed, 6 insertions(+), 5 deletions(-)
diff --git a/src/gallium/drivers/radeonsi/si_state_draw.c b/src/gallium/drivers/radeonsi/si_state_draw.c
index d3e6e1a..e44147f 100644
--- a/src/gallium/drivers/radeonsi/si_state_draw.c
+++ b/src/gallium/drivers/radeonsi/si_state_draw.c
@@ -318,14 +318,15 @@ static unsigned si_get_ia_multi_vgt_param(struct si_context *sctx,
wd_switch_on_eop = true;
/* Performance recommendation for 4 SE Gfx7-8 parts if
- * instances are smaller than a primgroup. Ignore the fact
- * primgroup_size is a primitive count, not vertex count.
- * Don't do anything for indirect draws.
+ * instances are smaller than a primgroup.
+ * Assume indirect draws always use small instances.
+ * This is needed for good VS wave utilization.
*/
if (sctx->b.chip_class <= VI &&
sctx->b.screen->info.max_se >= 4 &&
- !info->indirect &&
- info->instance_count > 1 && info->count < primgroup_size)
+ (info->indirect ||
+ (info->instance_count > 1 &&
+ si_num_prims_for_vertices(info) < primgroup_size)))
wd_switch_on_eop = true;
/* Required on CIK and later. */
More information about the mesa-commit
mailing list