i965/vec4: Fix swizzles on atomic sources.

Atomic operation sources are scalar values, but we were failing to
select the .x component of the second operand.  For example,

   atomicCounterCompSwapARB(counter, 5u, 10u)

would generate

   mov(8) vgrf4.x:D, 5D
   mov(8) vgrf5.x:D, 10D

   mov(8) vgrf9.x:UD, vgrf4.xyzw:D
   mov(8) vgrf9.y:UD, vgrf5.xyzw:D

which wrongly selects the .y component of vgrf5, so the actual 10u value
would get dead code eliminated.  The swizzle works for the other source,
but both of them ought to be .xxxx.

Fixes the compare and swap CTS tests in:
KHR-GL45.shader_atomic_counter_ops_tests.ShaderAtomicCounterOpsExchangeTestCase

Cc: "17.2 17.1 17.0 13.0" <mesa-stable@lists.freedesktop.org>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>

diff --git a/src/intel/compiler/brw_vec4_surface_builder.cpp b/src/intel/compiler/brw_vec4_surface_builder.cpp
index 5029cdce558bd825ea1ffef4ff5c8bc98ed66bf3..0e02aaf933a196194757bec8018ada7d4312139c 100644 (file)
--- a/src/intel/compiler/brw_vec4_surface_builder.cpp
+++ b/src/intel/compiler/brw_vec4_surface_builder.cpp
@@ -212,10 +212,15 @@ namespace brw {
          const unsigned size = (src0.file != BAD_FILE) + (src1.file != BAD_FILE);
          const dst_reg srcs = bld.vgrf(BRW_REGISTER_TYPE_UD);
 
-         if (size >= 1)
-            bld.MOV(writemask(srcs, WRITEMASK_X), src0);
-         if (size >= 2)
-            bld.MOV(writemask(srcs, WRITEMASK_Y), src1);
+         if (size >= 1) {
+            bld.MOV(writemask(srcs, WRITEMASK_X),
+                    swizzle(src0, BRW_SWIZZLE_XXXX));
+         }
+
+         if (size >= 2) {
+            bld.MOV(writemask(srcs, WRITEMASK_Y),
+                    swizzle(src1, BRW_SWIZZLE_XXXX));
+         }
 
          return emit_send(bld, SHADER_OPCODE_UNTYPED_ATOMIC, src_reg(),
                           emit_insert(bld, addr, dims, has_simd4x2),