AMDGPU: Add a shader calling convention

This makes it possible to distinguish between mesa shaders
and other kernels even in the presence of compute shaders.

Patch By: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>

Differential Revision: http://reviews.llvm.org/D18559

llvm-svn: 265589
diff --git a/llvm/test/CodeGen/AMDGPU/reciprocal.ll b/llvm/test/CodeGen/AMDGPU/reciprocal.ll
index b4ac47a..17fee93 100644
--- a/llvm/test/CodeGen/AMDGPU/reciprocal.ll
+++ b/llvm/test/CodeGen/AMDGPU/reciprocal.ll
@@ -2,7 +2,7 @@
 
 ;CHECK: RECIP_IEEE * T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
 
-define void @test(<4 x float> inreg %reg0) #0  {
+define amdgpu_ps void @test(<4 x float> inreg %reg0) {
    %r0 = extractelement <4 x float> %reg0, i32 0
    %r1 = fdiv float 1.0, %r0
    %vec = insertelement <4 x float> undef, float %r1, i32 0
@@ -11,5 +11,3 @@
 }
 
 declare void @llvm.R600.store.swizzle(<4 x float>, i32, i32)
-
-attributes #0 = { "ShaderType"="0" }