[Mesa-dev] [PATCH 1/2] tgsi/scan: record compute shader system value usage

[Marek Olšák]

From: Marek Olšák <marek.olsak at amd.com>

---
 src/gallium/auxiliary/tgsi/tgsi_scan.c | 47 ++++++++++++++++++++++++++++++++++
 src/gallium/auxiliary/tgsi/tgsi_scan.h |  4 +++
 2 files changed, 51 insertions(+)

diff --git a/src/gallium/auxiliary/tgsi/tgsi_scan.c b/src/gallium/auxiliary/tgsi/tgsi_scan.c
index bf614db..885a60d 100644
--- a/src/gallium/auxiliary/tgsi/tgsi_scan.c
+++ b/src/gallium/auxiliary/tgsi/tgsi_scan.c
@@ -108,20 +108,67 @@ static void
 scan_src_operand(struct tgsi_shader_info *info,
                  const struct tgsi_full_instruction *fullinst,
                  const struct tgsi_full_src_register *src,
                  unsigned src_index,
                  unsigned usage_mask,
                  bool is_interp_instruction,
                  bool *is_mem_inst)
 {
    int ind = src->Register.Index;
 
+   if (info->processor == PIPE_SHADER_COMPUTE &&
+       src->Register.File == TGSI_FILE_SYSTEM_VALUE) {
+      unsigned swizzle[4], i, name;
+
+      name = info->system_value_semantic_name[src->Register.Index];
+      swizzle[0] = src->Register.SwizzleX;
+      swizzle[1] = src->Register.SwizzleY;
+      swizzle[2] = src->Register.SwizzleZ;
+      swizzle[3] = src->Register.SwizzleW;
+
+      switch (name) {
+      case TGSI_SEMANTIC_THREAD_ID:
+      case TGSI_SEMANTIC_BLOCK_ID:
+         for (i = 0; i < 4; i++) {
+            switch (swizzle[i]) {
+            case TGSI_SWIZZLE_X:
+               if (name == TGSI_SEMANTIC_THREAD_ID)
+                  info->uses_thread_id[0] = true;
+               else
+                  info->uses_block_id[0] = true;
+               break;
+            case TGSI_SWIZZLE_Y:
+               if (name == TGSI_SEMANTIC_THREAD_ID)
+                  info->uses_thread_id[1] = true;
+               else
+                  info->uses_block_id[1] = true;
+               break;
+            case TGSI_SWIZZLE_Z:
+               if (name == TGSI_SEMANTIC_THREAD_ID)
+                  info->uses_thread_id[2] = true;
+               else
+                  info->uses_block_id[2] = true;
+               break;
+            }
+         }
+         break;
+      case TGSI_SEMANTIC_BLOCK_SIZE:
+         /* The block size is translated to IMM with a fixed block size. */
+         if (info->properties[TGSI_PROPERTY_CS_FIXED_BLOCK_WIDTH] == 0)
+            info->uses_block_size = true;
+         break;
+      case TGSI_SEMANTIC_GRID_SIZE:
+         info->uses_grid_size = true;
+         break;
+      }
+   }
+
    /* Mark which inputs are effectively used */
    if (src->Register.File == TGSI_FILE_INPUT) {
       if (src->Register.Indirect) {
          for (ind = 0; ind < info->num_inputs; ++ind) {
             info->input_usage_mask[ind] |= usage_mask;
          }
       } else {
          assert(ind >= 0);
          assert(ind < PIPE_MAX_SHADER_INPUTS);
          info->input_usage_mask[ind] |= usage_mask;
diff --git a/src/gallium/auxiliary/tgsi/tgsi_scan.h b/src/gallium/auxiliary/tgsi/tgsi_scan.h
index 3854827..98387c9 100644
--- a/src/gallium/auxiliary/tgsi/tgsi_scan.h
+++ b/src/gallium/auxiliary/tgsi/tgsi_scan.h
@@ -114,20 +114,24 @@ struct tgsi_shader_info
    boolean uses_linear_opcode_interp_centroid;
    boolean uses_linear_opcode_interp_offset;
    boolean uses_linear_opcode_interp_sample;
    boolean uses_instanceid;
    boolean uses_vertexid;
    boolean uses_vertexid_nobase;
    boolean uses_basevertex;
    boolean uses_primid;
    boolean uses_frontface;
    boolean uses_invocationid;
+   boolean uses_thread_id[3];
+   boolean uses_block_id[3];
+   boolean uses_block_size;
+   boolean uses_grid_size;
    boolean writes_position;
    boolean writes_psize;
    boolean writes_clipvertex;
    boolean writes_primid;
    boolean writes_viewport_index;
    boolean writes_layer;
    boolean writes_memory; /**< contains stores or atomics to buffers or images */
    boolean is_msaa_sampler[PIPE_MAX_SAMPLERS];
    boolean uses_doubles; /**< uses any of the double instructions */
    boolean uses_derivatives;
-- 
2.7.4