On 11/08/2015 13:25, Mika Kuoppala wrote:
Don't set the size of bindless surface state on rendercopy. And as of doing so, take into account the workaround for setting the command size. This was tried during hunting for https://bugs.freedesktop.org/show_bug.cgi?id=89959. But no impact was found. Cc: Arun Siluvery <arun.siluvery@xxxxxxxxxxxxxxx> Signed-off-by: Mika Kuoppala <mika.kuoppala@xxxxxxxxx> --- lib/rendercopy_gen9.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) diff --git a/lib/rendercopy_gen9.c b/lib/rendercopy_gen9.c index 0766192..4a4a604 100644 --- a/lib/rendercopy_gen9.c +++ b/lib/rendercopy_gen9.c @@ -511,7 +511,11 @@ gen7_emit_push_constants(struct intel_batchbuffer *batch) { static void gen9_emit_state_base_address(struct intel_batchbuffer *batch) { - OUT_BATCH(GEN6_STATE_BASE_ADDRESS | (19 - 2)); + + /* WaBindlessSurfaceStateModifyEnable:skl,bxt */ + /* The length has to be one less if we dont modify + bindless state */ + OUT_BATCH(GEN6_STATE_BASE_ADDRESS | (19 - 1 - 2)); /* general */ OUT_BATCH(0 | BASE_ADDRESS_MODIFY); @@ -544,9 +548,9 @@ gen9_emit_state_base_address(struct intel_batchbuffer *batch) { OUT_BATCH(1 << 12 | 1); /* Bindless surface state base address */ - OUT_BATCH(0 | BASE_ADDRESS_MODIFY); OUT_BATCH(0); - OUT_BATCH(0xfffff000); + OUT_BATCH(0); + OUT_BATCH(0); } static void
Agrees with spec and looks good to me. No impact observed with gem_concurrent_blit subtests.
Reviewed-by: Arun Siluvery <arun.siluvery@xxxxxxxxxxxxxxx> regards Arun _______________________________________________ Intel-gfx mailing list Intel-gfx@xxxxxxxxxxxxxxxxxxxxx http://lists.freedesktop.org/mailman/listinfo/intel-gfx
