radeonsi: reduce the number of FMASK input coordinates

Before:
  image_load v3, v[0:3] ...
After:
  image_load v3, v[0:1] ...

Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>

diff --git a/src/gallium/drivers/radeonsi/si_shader.c b/src/gallium/drivers/radeonsi/si_shader.c
index d898afd..eb2efbb 100644 (file)
--- a/src/gallium/drivers/radeonsi/si_shader.c
+++ b/src/gallium/drivers/radeonsi/si_shader.c
@@ -4395,16 +4395,12 @@ static void tex_fetch_args(
                struct lp_build_context *uint_bld = &bld_base->uint_bld;
                struct lp_build_emit_data txf_emit_data = *emit_data;
                LLVMValueRef txf_address[4];
-               unsigned txf_count = count;
+               /* We only need .xy for non-arrays, and .xyz for arrays. */
+               unsigned txf_count = target == TGSI_TEXTURE_2D_MSAA ? 2 : 3;
                struct tgsi_full_instruction inst = {};
 
                memcpy(txf_address, address, sizeof(txf_address));
 
-               if (target == TGSI_TEXTURE_2D_MSAA) {
-                       txf_address[2] = bld_base->uint_bld.zero;
-               }
-               txf_address[3] = bld_base->uint_bld.zero;
-
                /* Read FMASK using TXF. */
                inst.Instruction.Opcode = TGSI_OPCODE_TXF;
                inst.Texture.Texture = target;
@@ -4425,7 +4421,7 @@ static void tex_fetch_args(
                                                txf_emit_data.output[0],
                                                uint_bld->zero, "");
 
-               unsigned sample_chan = target == TGSI_TEXTURE_2D_MSAA ? 2 : 3;
+               unsigned sample_chan = txf_count; /* the sample index is last */
 
                LLVMValueRef sample_index4 =
                        LLVMBuildMul(gallivm->builder, address[sample_chan], four, "");