[Mesa-dev] [PATCH 2/7] i965: Perform basic optimizations on the BROADCAST opcode.

[Francisco Jerez]

---
 src/mesa/drivers/dri/i965/brw_fs.cpp                    | 15 +++++++++++++++
 src/mesa/drivers/dri/i965/brw_fs_copy_propagation.cpp   |  1 +
 src/mesa/drivers/dri/i965/brw_fs_cse.cpp                |  1 +
 src/mesa/drivers/dri/i965/brw_ir_fs.h                   |  7 +++++++
 src/mesa/drivers/dri/i965/brw_ir_vec4.h                 |  7 +++++++
 src/mesa/drivers/dri/i965/brw_vec4.cpp                  | 10 ++++++++++
 src/mesa/drivers/dri/i965/brw_vec4_copy_propagation.cpp |  1 +
 src/mesa/drivers/dri/i965/brw_vec4_cse.cpp              |  1 +
 8 files changed, 43 insertions(+)

diff --git a/src/mesa/drivers/dri/i965/brw_fs.cpp b/src/mesa/drivers/dri/i965/brw_fs.cpp
index a2a5234..d567c2b 100644
--- a/src/mesa/drivers/dri/i965/brw_fs.cpp
+++ b/src/mesa/drivers/dri/i965/brw_fs.cpp
@@ -2484,6 +2484,21 @@ fs_visitor::opt_algebraic()
          }
          break;
       }
+      case SHADER_OPCODE_BROADCAST:
+         if (is_uniform(inst->src[0])) {
+            inst->opcode = BRW_OPCODE_MOV;
+            inst->sources = 1;
+            progress = true;
+
+         } else if (inst->src[1].file == IMM) {
+            inst->opcode = BRW_OPCODE_MOV;
+            inst->src[0] = component(inst->src[0],
+                                     inst->src[1].fixed_hw_reg.dw1.ud);
+            inst->sources = 1;
+            progress = true;
+         }
+         break;
+
       default:
 	 break;
       }
diff --git a/src/mesa/drivers/dri/i965/brw_fs_copy_propagation.cpp b/src/mesa/drivers/dri/i965/brw_fs_copy_propagation.cpp
index e265ce0..cc4d9d0 100644
--- a/src/mesa/drivers/dri/i965/brw_fs_copy_propagation.cpp
+++ b/src/mesa/drivers/dri/i965/brw_fs_copy_propagation.cpp
@@ -572,6 +572,7 @@ fs_visitor::try_constant_propagate(fs_inst *inst, acp_entry *entry)
          break;
 
       case FS_OPCODE_UNIFORM_PULL_CONSTANT_LOAD:
+      case SHADER_OPCODE_BROADCAST:
          inst->src[i] = val;
          progress = true;
          break;
diff --git a/src/mesa/drivers/dri/i965/brw_fs_cse.cpp b/src/mesa/drivers/dri/i965/brw_fs_cse.cpp
index ae069bb..bbe0747 100644
--- a/src/mesa/drivers/dri/i965/brw_fs_cse.cpp
+++ b/src/mesa/drivers/dri/i965/brw_fs_cse.cpp
@@ -89,6 +89,7 @@ is_expression(const fs_inst *const inst)
    case FS_OPCODE_VARYING_PULL_CONSTANT_LOAD:
    case FS_OPCODE_CINTERP:
    case FS_OPCODE_LINTERP:
+   case SHADER_OPCODE_BROADCAST:
       return true;
    case SHADER_OPCODE_RCP:
    case SHADER_OPCODE_RSQ:
diff --git a/src/mesa/drivers/dri/i965/brw_ir_fs.h b/src/mesa/drivers/dri/i965/brw_ir_fs.h
index 9ef1261..dc1e041 100644
--- a/src/mesa/drivers/dri/i965/brw_ir_fs.h
+++ b/src/mesa/drivers/dri/i965/brw_ir_fs.h
@@ -166,6 +166,13 @@ component(fs_reg reg, unsigned idx)
    return reg;
 }
 
+static inline bool
+is_uniform(const fs_reg &reg)
+{
+   return (reg.width == 1 || reg.stride == 0 || reg.is_null()) &&
+      (!reg.reladdr || is_uniform(*reg.reladdr));
+}
+
 /**
  * Get either of the 8-component halves of a 16-component register.
  *
diff --git a/src/mesa/drivers/dri/i965/brw_ir_vec4.h b/src/mesa/drivers/dri/i965/brw_ir_vec4.h
index 941086f..066240e 100644
--- a/src/mesa/drivers/dri/i965/brw_ir_vec4.h
+++ b/src/mesa/drivers/dri/i965/brw_ir_vec4.h
@@ -102,6 +102,13 @@ negate(src_reg reg)
    return reg;
 }
 
+static inline bool
+is_uniform(const src_reg &reg)
+{
+   return (reg.file == IMM || reg.file == UNIFORM || reg.is_null()) &&
+      (!reg.reladdr || is_uniform(*reg.reladdr));
+}
+
 class dst_reg : public backend_reg
 {
 public:
diff --git a/src/mesa/drivers/dri/i965/brw_vec4.cpp b/src/mesa/drivers/dri/i965/brw_vec4.cpp
index 0a68413..601b8c1 100644
--- a/src/mesa/drivers/dri/i965/brw_vec4.cpp
+++ b/src/mesa/drivers/dri/i965/brw_vec4.cpp
@@ -756,6 +756,16 @@ vec4_visitor::opt_algebraic()
          }
          break;
       }
+      case SHADER_OPCODE_BROADCAST:
+         if (is_uniform(inst->src[0]) ||
+             (inst->src[1].file == IMM &&
+              inst->src[1].fixed_hw_reg.dw1.ud == 0)) {
+            inst->opcode = BRW_OPCODE_MOV;
+            inst->src[1] = src_reg();
+            progress = true;
+         }
+         break;
+
       default:
 	 break;
       }
diff --git a/src/mesa/drivers/dri/i965/brw_vec4_copy_propagation.cpp b/src/mesa/drivers/dri/i965/brw_vec4_copy_propagation.cpp
index 679867c..c65f23d 100644
--- a/src/mesa/drivers/dri/i965/brw_vec4_copy_propagation.cpp
+++ b/src/mesa/drivers/dri/i965/brw_vec4_copy_propagation.cpp
@@ -152,6 +152,7 @@ try_constant_propagate(struct brw_context *brw, vec4_instruction *inst,
 
    switch (inst->opcode) {
    case BRW_OPCODE_MOV:
+   case SHADER_OPCODE_BROADCAST:
       inst->src[arg] = value;
       return true;
 
diff --git a/src/mesa/drivers/dri/i965/brw_vec4_cse.cpp b/src/mesa/drivers/dri/i965/brw_vec4_cse.cpp
index 5fb8f31..dbfea92 100644
--- a/src/mesa/drivers/dri/i965/brw_vec4_cse.cpp
+++ b/src/mesa/drivers/dri/i965/brw_vec4_cse.cpp
@@ -71,6 +71,7 @@ is_expression(const vec4_instruction *const inst)
    case BRW_OPCODE_MAD:
    case BRW_OPCODE_LRP:
    case VEC4_OPCODE_UNPACK_UNIFORM:
+   case SHADER_OPCODE_BROADCAST:
       return true;
    case SHADER_OPCODE_RCP:
    case SHADER_OPCODE_RSQ:
-- 
2.1.3