[Mesa-dev] [PATCH 2/3] i965: Use LOAD_PAYLOAD for SIMD8 TES input loads, not MOV.

[Kenneth Graunke]

We need a MOV to replicate g0.0<0,1,0> to all 8 channels.  Since the
message payload is a single register, MOV seemed more sensible than
LOAD_PAYLOAD.  However, MOV cannot be CSE'd, while LOAD_PAYLOAD can.

All input loads can use the same header - we don't need to re-expand
g0 every time.  CSE accomplishes this, saving instructions.

shader-db statistics for files containing tessellation shaders:

total instructions in shared programs: 186923 -> 184358 (-1.37%)
instructions in affected programs: 30536 -> 27971 (-8.40%)
helped: 226
HURT: 0

total cycles in shared programs: 1009850 -> 1005356 (-0.45%)
cycles in affected programs: 168206 -> 163712 (-2.67%)
helped: 226
HURT: 0

Signed-off-by: Kenneth Graunke <kenneth at whitecape.org>
---
 src/mesa/drivers/dri/i965/brw_fs_nir.cpp | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/src/mesa/drivers/dri/i965/brw_fs_nir.cpp b/src/mesa/drivers/dri/i965/brw_fs_nir.cpp
index 827dbee..788315f 100644
--- a/src/mesa/drivers/dri/i965/brw_fs_nir.cpp
+++ b/src/mesa/drivers/dri/i965/brw_fs_nir.cpp
@@ -1850,8 +1850,11 @@ fs_visitor::nir_emit_tes_intrinsic(const fs_builder &bld,
       fs_inst *inst;
       if (indirect_offset.file == BAD_FILE) {
          /* Replicate the patch handle to all enabled channels */
+         const fs_reg srcs[] = {
+            retype(brw_vec1_grf(0, 0), BRW_REGISTER_TYPE_UD)
+         };
          fs_reg patch_handle = bld.vgrf(BRW_REGISTER_TYPE_UD, 1);
-         bld.MOV(patch_handle, retype(brw_vec1_grf(0, 0), BRW_REGISTER_TYPE_UD));
+         bld.LOAD_PAYLOAD(patch_handle, srcs, ARRAY_SIZE(srcs), 0);
 
          inst = bld.emit(SHADER_OPCODE_URB_READ_SIMD8, dest, patch_handle);
          inst->mlen = 1;
-- 
2.6.4