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/shared-op-cycle.ll b/llvm/test/CodeGen/AMDGPU/shared-op-cycle.ll
index f52a9ba..8c2cbf1 100644
--- a/llvm/test/CodeGen/AMDGPU/shared-op-cycle.ll
+++ b/llvm/test/CodeGen/AMDGPU/shared-op-cycle.ll
@@ -4,7 +4,7 @@
 ; CHECK: MULADD_IEEE *
 ; CHECK-NOT: MULADD_IEEE *
 
-define void @main(<4 x float> inreg %reg0, <4 x float> inreg %reg1, <4 x float> inreg %reg2) #0 {
+define amdgpu_vs void @main(<4 x float> inreg %reg0, <4 x float> inreg %reg1, <4 x float> inreg %reg2) {
    %w0 = extractelement <4 x float> %reg0, i32 3
    %w1 = extractelement <4 x float> %reg1, i32 3
    %w2 = extractelement <4 x float> %reg2, i32 3
@@ -28,5 +28,4 @@
 
 declare void @llvm.R600.store.swizzle(<4 x float>, i32, i32)
 
-attributes #0 = { "ShaderType"="1" }
-attributes #1 = { readnone }
\ No newline at end of file
+attributes #1 = { readnone }