[Mesa-dev] [PATCH] llvmpipe: fix timer query if there's no bins

sroland at vmware.com sroland at vmware.com
Fri Jun 28 07:54:58 PDT 2013


From: Roland Scheidegger <sroland at vmware.com>

b04a295a4a0cd2defe352b3193b5fa79ca8fc9fc removed seemingly unnecessary
code in get_query. Turns out this code could in fact be reached - while
timestamps are always binned, if there are no bins (which happens if fb
size is 0) then the rasterization query code filling this in is still
never executed.
So fix this up by filling in some timestamp, but do it at EndQuery time
not GetQuery time which should be more appropriate.
Makes piglit arb_timer_query-timestamp-get happy again.
---
 src/gallium/drivers/llvmpipe/lp_setup.c |   10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/src/gallium/drivers/llvmpipe/lp_setup.c b/src/gallium/drivers/llvmpipe/lp_setup.c
index 49b61c3..65f61ed 100644
--- a/src/gallium/drivers/llvmpipe/lp_setup.c
+++ b/src/gallium/drivers/llvmpipe/lp_setup.c
@@ -40,6 +40,7 @@
 #include "util/u_memory.h"
 #include "util/u_pack_color.h"
 #include "draw/draw_pipe.h"
+#include "os/os_time.h"
 #include "lp_context.h"
 #include "lp_memory.h"
 #include "lp_scene.h"
@@ -1263,6 +1264,15 @@ lp_setup_end_query(struct lp_setup_context *setup, struct llvmpipe_query *pq)
           pq->type == PIPE_QUERY_OCCLUSION_PREDICATE ||
           pq->type == PIPE_QUERY_PIPELINE_STATISTICS ||
           pq->type == PIPE_QUERY_TIMESTAMP) {
+         if (pq->type == PIPE_QUERY_TIMESTAMP &&
+               !(setup->scene->tiles_x | setup->scene->tiles_y)) {
+            /*
+             * If there's a zero width/height framebuffer, there's no bins and
+             * hence no rast task is ever run. So fill in something here instead.
+             */
+            pq->end[0] = os_time_get_nano();
+         }
+
          if (!lp_scene_bin_everywhere(setup->scene,
                                       LP_RAST_OP_END_QUERY,
                                       lp_rast_arg_query(pq))) {
-- 
1.7.9.5


More information about the mesa-dev mailing list