i965/vec4: Compare full register offsets in cmod propagation.

[mesa.git] / src / mesa / drivers / dri / i965 / brw_fs.cpp

diff --git a/src/mesa/drivers/dri/i965/brw_fs.cpp b/src/mesa/drivers/dri/i965/brw_fs.cpp
index ed80a54bf3653754f5c98da17289241ba94fd03b..c858f449c8f0642548a964dd26099e222640428c 100644 (file)
--- a/src/mesa/drivers/dri/i965/brw_fs.cpp
+++ b/src/mesa/drivers/dri/i965/brw_fs.cpp
@@ -172,12 +172,12 @@ fs_visitor::VARYING_PULL_CONSTANT_LOAD(const fs_builder &bld,
     * be any component of a vector, and then we load 4 contiguous
     * components starting from that.
     *
-    * We break down the const_offset to a portion added to the variable
-    * offset and a portion done using reg_offset, which means that if you
-    * have GLSL using something like "uniform vec4 a[20]; gl_FragColor =
-    * a[i]", we'll temporarily generate 4 vec4 loads from offset i * 4, and
-    * CSE can later notice that those loads are all the same and eliminate
-    * the redundant ones.
+    * We break down the const_offset to a portion added to the variable offset
+    * and a portion done using fs_reg::offset, which means that if you have
+    * GLSL using something like "uniform vec4 a[20]; gl_FragColor = a[i]",
+    * we'll temporarily generate 4 vec4 loads from offset i * 4, and CSE can
+    * later notice that those loads are all the same and eliminate the
+    * redundant ones.
     */
    fs_reg vec4_offset = vgrf(glsl_type::uint_type);
    bld.ADD(vec4_offset, varying_offset, brw_imm_ud(const_offset & ~0xf));
@@ -2062,7 +2062,7 @@ fs_visitor::assign_constant_locations()
    stage_prog_data->nr_params = num_push_constants;
    stage_prog_data->nr_pull_params = num_pull_constants;
 
-   /* Up until now, the param[] array has been indexed by reg + reg_offset
+   /* Up until now, the param[] array has been indexed by reg + offset
     * of UNIFORM registers.  Move pull constants into pull_param[] and
     * condense param[] to only contain the uniforms we chose to push.
     *
@@ -3166,10 +3166,6 @@ fs_visitor::insert_gen4_send_dependency_workarounds()
 
    bool progress = false;
 
-   /* Note that we're done with register allocation, so GRF fs_regs always
-    * have a .reg_offset of 0.
-    */
-
    foreach_block_and_inst(block, fs_inst, inst, cfg) {
       if (inst->mlen != 0 && inst->dst.file == VGRF) {
          insert_gen4_pre_send_dependency_workarounds(block, inst);
@@ -5233,10 +5229,6 @@ fs_visitor::dump_instruction(backend_instruction *be_inst, FILE *file)
    switch (inst->dst.file) {
    case VGRF:
       fprintf(file, "vgrf%d", inst->dst.nr);
-      if (alloc.sizes[inst->dst.nr] * REG_SIZE != inst->size_written ||
-          inst->dst.offset % REG_SIZE)
-         fprintf(file, "+%d.%d",
-                 inst->dst.offset / REG_SIZE, inst->dst.offset % REG_SIZE);
       break;
    case FIXED_GRF:
       fprintf(file, "g%d", inst->dst.nr);
@@ -5248,10 +5240,10 @@ fs_visitor::dump_instruction(backend_instruction *be_inst, FILE *file)
       fprintf(file, "(null)");
       break;
    case UNIFORM:
-      fprintf(file, "***u%d***", inst->dst.nr + inst->dst.offset / 4);
+      fprintf(file, "***u%d***", inst->dst.nr);
       break;
    case ATTR:
-      fprintf(file, "***attr%d***", inst->dst.nr + inst->dst.offset / REG_SIZE);
+      fprintf(file, "***attr%d***", inst->dst.nr);
       break;
    case ARF:
       switch (inst->dst.nr) {
@@ -5271,12 +5263,19 @@ fs_visitor::dump_instruction(backend_instruction *be_inst, FILE *file)
          fprintf(file, "arf%d.%d", inst->dst.nr & 0xf, inst->dst.subnr);
          break;
       }
-      if (inst->dst.subnr)
-         fprintf(file, "+%d", inst->dst.subnr);
       break;
    case IMM:
       unreachable("not reached");
    }
+
+   if (inst->dst.offset ||
+       (inst->dst.file == VGRF &&
+        alloc.sizes[inst->dst.nr] * REG_SIZE != inst->size_written)) {
+      const unsigned reg_size = (inst->dst.file == UNIFORM ? 4 : REG_SIZE);
+      fprintf(file, "+%d.%d", inst->dst.offset / reg_size,
+              inst->dst.offset % reg_size);
+   }
+
    if (inst->dst.stride != 1)
       fprintf(file, "<%u>", inst->dst.stride);
    fprintf(file, ":%s, ", brw_reg_type_letters(inst->dst.type));
@@ -5289,10 +5288,6 @@ fs_visitor::dump_instruction(backend_instruction *be_inst, FILE *file)
       switch (inst->src[i].file) {
       case VGRF:
          fprintf(file, "vgrf%d", inst->src[i].nr);
-         if (alloc.sizes[inst->src[i].nr] * REG_SIZE != inst->size_read(i) ||
-             inst->src[i].offset % REG_SIZE != 0)
-            fprintf(file, "+%d.%d", inst->src[i].offset / REG_SIZE,
-                    inst->src[i].offset % REG_SIZE);
          break;
       case FIXED_GRF:
          fprintf(file, "g%d", inst->src[i].nr);
@@ -5301,14 +5296,10 @@ fs_visitor::dump_instruction(backend_instruction *be_inst, FILE *file)
          fprintf(file, "***m%d***", inst->src[i].nr);
          break;
       case ATTR:
-         fprintf(file, "attr%d+%d", inst->src[i].nr, inst->src[i].offset / REG_SIZE);
+         fprintf(file, "attr%d", inst->src[i].nr);
          break;
       case UNIFORM:
-         fprintf(file, "u%d", inst->src[i].nr + inst->src[i].offset / 4);
-         if (inst->src[i].offset % 4 != 0) {
-            fprintf(file, "+%d.%d", inst->src[i].offset / 4,
-                    inst->src[i].offset % 4);
-         }
+         fprintf(file, "u%d", inst->src[i].nr);
          break;
       case BAD_FILE:
          fprintf(file, "(null)");
@@ -5359,10 +5350,17 @@ fs_visitor::dump_instruction(backend_instruction *be_inst, FILE *file)
             fprintf(file, "arf%d.%d", inst->src[i].nr & 0xf, inst->src[i].subnr);
             break;
          }
-         if (inst->src[i].subnr)
-            fprintf(file, "+%d", inst->src[i].subnr);
          break;
       }
+
+      if (inst->src[i].offset ||
+          (inst->src[i].file == VGRF &&
+           alloc.sizes[inst->src[i].nr] * REG_SIZE != inst->size_read(i))) {
+         const unsigned reg_size = (inst->src[i].file == UNIFORM ? 4 : REG_SIZE);
+         fprintf(file, "+%d.%d", inst->src[i].offset / reg_size,
+                 inst->src[i].offset % reg_size);
+      }
+
       if (inst->src[i].abs)
          fprintf(file, "|");