When a card has more than one GPC, the grid used by the compute
kernel which reads MP performance counters seems to be too small.
The consequence is that the kernel is not launched on all TPCs.
Increasing the grid size using the number of GPCs now launches
enough blocks and we can read MP performance counters of all TPCs.
Signed-off-by: Samuel Pitoiset <samuel.pitoi...@gmail.com>
---
src/gallium/drivers/nouveau/nvc0/nvc0_query_hw_sm.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/gallium/drivers/nouveau/nvc0/nvc0_query_hw_sm.c
b/src/gallium/drivers/nouveau/nvc0/nvc0_query_hw_sm.c
index 840f200..c22ad4b 100644
--- a/src/gallium/drivers/nouveau/nvc0/nvc0_query_hw_sm.c
+++ b/src/gallium/drivers/nouveau/nvc0/nvc0_query_hw_sm.c
@@ -528,7 +528,7 @@ nvc0_hw_sm_end_query(struct nvc0_context *nvc0, struct
nvc0_hw_query *hq)
uint32_t mask;
uint32_t input[3];
const uint block[3] = { 32, is_nve4 ? 4 : 1, 1 };
- const uint grid[3] = { screen->mp_count, 1, 1 };
+ const uint grid[3] = { screen->mp_count, screen->gpc_count, 1 };
unsigned c;
if (unlikely(!screen->pm.prog)) {
--
2.6.1
_______________________________________________
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
