nir: Allow vec2/vec3/vec4 instructions in the select peephole pass.

These are basically just moves, so they should be safe as well.

When disabling i965's GLSL IR level scalarizer (channel expressions)
pass, I started seeing NIR code like this:

        if ssa_21 {
                block block_1:
                /* preds: block_0 */
                vec4 ssa_120 = vec4 ssa_82, ssa_83, ssa_84, ssa_30
                /* succs: block_3 */
        } else {
                block block_2:
                /* preds: block_0 */
                /* succs: block_3 */
        }
        block block_3:
        /* preds: block_1 block_2 */
        vec4 ssa_33 = phi block_1: ssa_120, block_2: ssa_2

Previously, the GLSL IR scalarizer pass would break the vec4 into a
series of fmovs, which were allowed by the peephole pass.  But with
the vec4 operation, they were not.  We want to keep getting selects.

Normal i965 on Broadwell:
instructions in affected programs:     200 -> 176 (-12.00%)
helped:                                4

With brw_fs_channel_expressions() disabled:
instructions in affected programs:     1832 -> 1646 (-10.15%)
helped:                                30

Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>

diff --git a/src/glsl/nir/nir_opt_peephole_select.c b/src/glsl/nir/nir_opt_peephole_select.c
index 82c65bb442fa1724ff641e5a719ec3f05a45ff30..ef7c9775aa3071b3c4c4beb489a65007d6777f1c 100644 (file)
--- a/src/glsl/nir/nir_opt_peephole_select.c
+++ b/src/glsl/nir/nir_opt_peephole_select.c
@@ -86,7 +86,9 @@ block_check_for_allowed_instrs(nir_block *block)
          nir_alu_instr *mov = nir_instr_as_alu(instr);
          if (mov->op != nir_op_fmov && mov->op != nir_op_imov &&
              mov->op != nir_op_fneg && mov->op != nir_op_ineg &&
-             mov->op != nir_op_fabs && mov->op != nir_op_iabs)
+             mov->op != nir_op_fabs && mov->op != nir_op_iabs &&
+             mov->op != nir_op_vec2 && mov->op != nir_op_vec3 &&
+             mov->op != nir_op_vec4)
             return false;
 
          /* Can't handle saturate */