R600: support f16 -> f64 conversion intrinsic.

Unfortunately, we don't seem to have a direct truncation, but the
extension can be legally split into two operations so we should
support that.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@213357 91177308-0d34-0410-b5e6-96231b3b80d8

diff --git a/lib/Target/R600/AMDGPUISelLowering.cpp b/lib/Target/R600/AMDGPUISelLowering.cpp
index 89d5b08af416347b0d4a138fce153077b0932be8..42d2a13a398a2e93b8ce0c69adef5f2b8ae51c63 100644 (file)
--- a/lib/Target/R600/AMDGPUISelLowering.cpp
+++ b/lib/Target/R600/AMDGPUISelLowering.cpp
@@ -242,6 +242,8 @@ AMDGPUTargetLowering::AMDGPUTargetLowering(TargetMachine &TM) :
     setOperationAction(ISD::FCOPYSIGN, MVT::f64, Expand);
   }
 
+  setOperationAction(ISD::FP16_TO_FP, MVT::f64, Expand);
+
   const MVT ScalarIntVTs[] = { MVT::i32, MVT::i64 };
   for (MVT VT : ScalarIntVTs) {
     setOperationAction(ISD::SREM, VT, Expand);

diff --git a/test/CodeGen/R600/fp16_to_fp.ll b/test/CodeGen/R600/fp16_to_fp.ll
new file mode 100644 (file)
index 0000000..777eadc
--- /dev/null
+++ b/test/CodeGen/R600/fp16_to_fp.ll
@@ -0,0 +1,28 @@
+; RUN: llc -march=r600 -mcpu=SI -verify-machineinstrs < %s | FileCheck -check-prefix=SI %s
+
+declare float @llvm.convert.from.fp16.f32(i16) nounwind readnone
+declare double @llvm.convert.from.fp16.f64(i16) nounwind readnone
+
+; SI-LABEL: @test_convert_fp16_to_fp32:
+; SI: BUFFER_LOAD_USHORT [[VAL:v[0-9]+]]
+; SI: V_CVT_F32_F16_e32 [[RESULT:v[0-9]+]], [[VAL]]
+; SI: BUFFER_STORE_DWORD [[RESULT]]
+define void @test_convert_fp16_to_fp32(float addrspace(1)* noalias %out, i16 addrspace(1)* noalias %in) nounwind {
+  %val = load i16 addrspace(1)* %in, align 2
+  %cvt = call float @llvm.convert.from.fp16.f32(i16 %val) nounwind readnone
+  store float %cvt, float addrspace(1)* %out, align 4
+  ret void
+}
+
+
+; SI-LABEL: @test_convert_fp16_to_fp64:
+; SI: BUFFER_LOAD_USHORT [[VAL:v[0-9]+]]
+; SI: V_CVT_F32_F16_e32 [[RESULT32:v[0-9]+]], [[VAL]]
+; SI: V_CVT_F64_F32_e32 [[RESULT:v\[[0-9]+:[0-9]+\]]], [[RESULT32]]
+; SI: BUFFER_STORE_DWORDX2 [[RESULT]]
+define void @test_convert_fp16_to_fp64(double addrspace(1)* noalias %out, i16 addrspace(1)* noalias %in) nounwind {
+  %val = load i16 addrspace(1)* %in, align 2
+  %cvt = call double @llvm.convert.from.fp16.f64(i16 %val) nounwind readnone
+  store double %cvt, double addrspace(1)* %out, align 4
+  ret void
+}

diff --git a/test/CodeGen/R600/fp32_to_fp16.ll b/test/CodeGen/R600/fp32_to_fp16.ll
deleted file mode 100644 (file)
index 3a051f8..0000000
--- a/test/CodeGen/R600/fp32_to_fp16.ll
+++ /dev/null
@@ -1,14 +0,0 @@
-; RUN: llc -march=r600 -mcpu=SI -verify-machineinstrs < %s | FileCheck -check-prefix=SI %s
-
-declare float @llvm.convert.from.fp16.f32(i16) nounwind readnone
-
-; SI-LABEL: @test_convert_fp16_to_fp32:
-; SI: BUFFER_LOAD_USHORT [[VAL:v[0-9]+]]
-; SI: V_CVT_F32_F16_e32 [[RESULT:v[0-9]+]], [[VAL]]
-; SI: BUFFER_STORE_DWORD [[RESULT]]
-define void @test_convert_fp16_to_fp32(float addrspace(1)* noalias %out, i16 addrspace(1)* noalias %in) nounwind {
-  %val = load i16 addrspace(1)* %in, align 2
-  %cvt = call float @llvm.convert.from.fp16.f32(i16 %val) nounwind readnone
-  store float %cvt, float addrspace(1)* %out, align 4
-  ret void
-}