i965/blorp: Align vertex buffers to 64B
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Cc: "13.0 17.0" <mesa-stable@lists.freedesktop.org>
(cherry picked from commit f938354362655a378d474c5f79c52cea9852ab91)
[Emil Velikov: brw_state_batch has different signature]
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Conflicts:
src/mesa/drivers/dri/i965/genX_blorp_exec.c
diff --git a/src/mesa/drivers/dri/i965/genX_blorp_exec.c b/src/mesa/drivers/dri/i965/genX_blorp_exec.c
index 8e011e9..adcce92 100644
--- a/src/mesa/drivers/dri/i965/genX_blorp_exec.c
+++ b/src/mesa/drivers/dri/i965/genX_blorp_exec.c
@@ -130,9 +130,21 @@
assert(batch->blorp->driver_ctx == batch->driver_batch);
struct brw_context *brw = batch->driver_batch;
+ /* From the Skylake PRM, 3DSTATE_VERTEX_BUFFERS:
+ *
+ * "The VF cache needs to be invalidated before binding and then using
+ * Vertex Buffers that overlap with any previously bound Vertex Buffer
+ * (at a 64B granularity) since the last invalidation. A VF cache
+ * invalidate is performed by setting the "VF Cache Invalidation Enable"
+ * bit in PIPE_CONTROL."
+ *
+ * This restriction first appears in the Skylake PRM but the internal docs
+ * also list it as being an issue on Broadwell. In order to avoid this
+ * problem, we align all vertex buffer allocations to 64 bytes.
+ */
uint32_t offset;
void *data = brw_state_batch(brw, AUB_TRACE_VERTEX_BUFFER,
- size, 32, &offset);
+ size, 64, &offset);
*addr = (struct blorp_address) {
.buffer = brw->batch.bo,