Mesa (master): intel/vec4: Reswizzle VF immediates too

[GitLab Mirror]

Module: Mesa
Branch: master
Commit: 47c2aa5b481389bfc68d688f02aa1fd4f095713a
URL:    http://cgit.freedesktop.org/mesa/mesa/commit/?id=47c2aa5b481389bfc68d688f02aa1fd4f095713a

Author: Ian Romanick <ian.d.romanick at intel.com>
Date:   Wed Jun 12 13:19:25 2019 -0700

intel/vec4: Reswizzle VF immediates too

Previously, an instruction like

mul(8) vgrf29.xy:F, vgrf25.yxxx:F, [-1F, 1F, 0F, 0F]

would get rewritten as

mul(8) vgrf0.yz:F, vgrf25.yyxx:F, [-1F, 1F, 0F, 0F]

The latter does not produce the correct result.  The VF immediate in the
second should be either [-1F, -1F, 1F, 1F] or [0F, -1F, 1F, 0F].  This
commit produces the former.

Fixes: 1ee1d8ab468 ("i965/vec4: Reswizzle sources when necessary.")
Reviewed-by: Matt Turner <mattst88 at gmail.com>

---

 src/intel/compiler/brw_vec4.cpp | 24 +++++++++++++++++++++++-
 1 file changed, 23 insertions(+), 1 deletion(-)

diff --git a/src/intel/compiler/brw_vec4.cpp b/src/intel/compiler/brw_vec4.cpp
index 971439fa472..26e2a0ecf5b 100644
--- a/src/intel/compiler/brw_vec4.cpp
+++ b/src/intel/compiler/brw_vec4.cpp
@@ -1204,9 +1204,31 @@ vec4_instruction::reswizzle(int dst_writemask, int swizzle)
        opcode != BRW_OPCODE_DP3 && opcode != BRW_OPCODE_DP2 &&
        opcode != VEC4_OPCODE_PACK_BYTES) {
       for (int i = 0; i < 3; i++) {
-         if (src[i].file == BAD_FILE || src[i].file == IMM)
+         if (src[i].file == BAD_FILE)
             continue;
 
+         if (src[i].file == IMM) {
+            assert(src[i].type != BRW_REGISTER_TYPE_V &&
+                   src[i].type != BRW_REGISTER_TYPE_UV);
+
+            /* Vector immediate types need to be reswizzled. */
+            if (src[i].type == BRW_REGISTER_TYPE_VF) {
+               const unsigned imm[] = {
+                  (src[i].ud >>  0) & 0x0ff,
+                  (src[i].ud >>  8) & 0x0ff,
+                  (src[i].ud >> 16) & 0x0ff,
+                  (src[i].ud >> 24) & 0x0ff,
+               };
+
+               src[i] = brw_imm_vf4(imm[BRW_GET_SWZ(swizzle, 0)],
+                                    imm[BRW_GET_SWZ(swizzle, 1)],
+                                    imm[BRW_GET_SWZ(swizzle, 2)],
+                                    imm[BRW_GET_SWZ(swizzle, 3)]);
+            }
+
+            continue;
+         }
+
          src[i].swizzle = brw_compose_swizzle(swizzle, src[i].swizzle);
       }
    }