[Mesa-dev] [PATCH 36/47] glsl: Add a lowering pass for 64-bit float rcp()

[Elie Tournier]

Signed-off-by: Elie Tournier <elie.tournier at collabora.com>
---
 src/compiler/glsl/ir_optimization.h        | 1 +
 src/compiler/glsl/lower_64bit.cpp          | 7 +++++++
 src/mesa/state_tracker/st_glsl_to_tgsi.cpp | 3 ++-
 3 files changed, 10 insertions(+), 1 deletion(-)

diff --git a/src/compiler/glsl/ir_optimization.h b/src/compiler/glsl/ir_optimization.h
index 4a44ee7660..e18924d62b 100644
--- a/src/compiler/glsl/ir_optimization.h
+++ b/src/compiler/glsl/ir_optimization.h
@@ -77,6 +77,7 @@
 #define SQRT64                    (1U << 16)
 #define TRUNC64                   (1U << 17)
 #define ROUND64                   (1U << 18)
+#define RCP64                     (1U << 19)
 
 /**
  * \see class lower_packing_builtins_visitor
diff --git a/src/compiler/glsl/lower_64bit.cpp b/src/compiler/glsl/lower_64bit.cpp
index 894503a14f..f8fcf9fd0a 100644
--- a/src/compiler/glsl/lower_64bit.cpp
+++ b/src/compiler/glsl/lower_64bit.cpp
@@ -478,6 +478,13 @@ lower_64bit_visitor::handle_rvalue(ir_rvalue **rvalue)
       }
       break;
 
+   case ir_unop_rcp:
+      if (lowering(RCP64)) {
+         if (ir->type->base_type == GLSL_TYPE_DOUBLE)
+            *rvalue = handle_op(ir, "__builtin_frcp64", generate_ir::frcp64);
+      }
+      break;
+
    case ir_unop_round_even:
       if (lowering(ROUND64)) {
          if (ir->type->base_type == GLSL_TYPE_DOUBLE)
diff --git a/src/mesa/state_tracker/st_glsl_to_tgsi.cpp b/src/mesa/state_tracker/st_glsl_to_tgsi.cpp
index 760ffd594e..dc88a881f6 100644
--- a/src/mesa/state_tracker/st_glsl_to_tgsi.cpp
+++ b/src/mesa/state_tracker/st_glsl_to_tgsi.cpp
@@ -7074,7 +7074,8 @@ st_link_shader(struct gl_context *ctx, struct gl_shader_program *prog)
                                F2D |
                                SQRT64 |
                                TRUNC64 |
-                               ROUND64;
+                               ROUND64 |
+                               RCP64;
          lower_64bit_double_instructions(ir, lower_inst);
       }
 
-- 
2.14.1