From: Connor Abbott <connor.w.abb...@intel.com>

This can happen when the source of the compare was split by the SIMD
lowering pass. Potentially, we could allow the case where the exec size
of scan_inst is larger, and scan_inst has the right quarter selected,
but doing that seems a little more risky.

Reviewed-by: Iago Toral Quiroga <ito...@igalia.com>
---
 src/mesa/drivers/dri/i965/brw_fs_cmod_propagation.cpp | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/src/mesa/drivers/dri/i965/brw_fs_cmod_propagation.cpp 
b/src/mesa/drivers/dri/i965/brw_fs_cmod_propagation.cpp
index 8fdc959..93461f7 100644
--- a/src/mesa/drivers/dri/i965/brw_fs_cmod_propagation.cpp
+++ b/src/mesa/drivers/dri/i965/brw_fs_cmod_propagation.cpp
@@ -93,6 +93,9 @@ opt_cmod_propagation_local(bblock_t *block)
                 scan_inst->dst.reg_offset != inst->src[0].reg_offset)
                break;
 
+            if (scan_inst->exec_size != inst->exec_size)
+               break;
+
             /* CMP's result is the same regardless of dest type. */
             if (inst->conditional_mod == BRW_CONDITIONAL_NZ &&
                 scan_inst->opcode == BRW_OPCODE_CMP &&
-- 
1.9.1

_______________________________________________
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev

Reply via email to