replica
/
hanchenye-llvm-project
2
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

R600: Update for div_fmas intrinsic change 

llvm-svn: 220339
This commit is contained in:
Matt Arsenault 2014-10-21 22:21:41 +00:00
parent 75c658e2cc
commit 2174a9dc28
3 changed files with 17 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
clang/include/clang/Basic/BuiltinsR600.def
View File

@ -16,8 +16,8 @@
BUILTIN(__builtin_amdgpu_div_scale, "dddbb*", "n")
BUILTIN(__builtin_amdgpu_div_scalef, "fffbb*", "n")
BUILTIN(__builtin_amdgpu_div_fmas, "dddd", "nc")
BUILTIN(__builtin_amdgpu_div_fmasf, "ffff", "nc")
BUILTIN(__builtin_amdgpu_div_fmas, "ddddb", "nc")
BUILTIN(__builtin_amdgpu_div_fmasf, "ffffb", "nc")
BUILTIN(__builtin_amdgpu_div_fixup, "dddd", "nc")
BUILTIN(__builtin_amdgpu_div_fixupf, "ffff", "nc")
BUILTIN(__builtin_amdgpu_trig_preop, "ddi", "nc")

13
clang/lib/CodeGen/CGBuiltin.cpp
View File

@ -6153,8 +6153,17 @@ Value *CodeGenFunction::EmitR600BuiltinExpr(unsigned BuiltinID,
return Result;
}
case R600::BI__builtin_amdgpu_div_fmas:
case R600::BI__builtin_amdgpu_div_fmasf:
return emitTernaryFPBuiltin(*this, E, Intrinsic::AMDGPU_div_fmas);
case R600::BI__builtin_amdgpu_div_fmasf: {
llvm::Value *Src0 = EmitScalarExpr(E->getArg(0));
llvm::Value *Src1 = EmitScalarExpr(E->getArg(1));
llvm::Value *Src2 = EmitScalarExpr(E->getArg(2));
llvm::Value *Src3 = EmitScalarExpr(E->getArg(3));
llvm::Value *F = CGM.getIntrinsic(Intrinsic::AMDGPU_div_fmas,
Src0->getType());
llvm::Value *Src3ToBool = Builder.CreateIsNotNull(Src3);
return Builder.CreateCall4(F, Src0, Src1, Src2, Src3ToBool);
}
case R600::BI__builtin_amdgpu_div_fixup:
case R600::BI__builtin_amdgpu_div_fixupf:
return emitTernaryFPBuiltin(*this, E, Intrinsic::AMDGPU_div_fixup);

8
clang/test/CodeGenOpenCL/builtins-r600.cl
View File

@ -31,16 +31,16 @@ void test_div_scale_f32(global float* out, global int* flagout, float a, float b
// CHECK-LABEL: @test_div_fmas_f32
// CHECK: call float @llvm.AMDGPU.div.fmas.f32
void test_div_fmas_f32(global float* out, float a, float b, float c)
void test_div_fmas_f32(global float* out, float a, float b, float c, int d)
{
*out = __builtin_amdgpu_div_fmasf(a, b, c);
*out = __builtin_amdgpu_div_fmasf(a, b, c, d);
}
// CHECK-LABEL: @test_div_fmas_f64
// CHECK: call double @llvm.AMDGPU.div.fmas.f64
void test_div_fmas_f64(global double* out, double a, double b, double c)
void test_div_fmas_f64(global double* out, double a, double b, double c, int d)
{
*out = __builtin_amdgpu_div_fmas(a, b, c);
*out = __builtin_amdgpu_div_fmas(a, b, c, d);
}
// CHECK-LABEL: @test_div_fixup_f32