Mercurial > hg > CbC > CbC_llvm

comparison test/CodeGen/AMDGPU/fp32_to_fp16.ll @ 95:afa8332a0e37 LLVM3.8

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
LLVM 3.8
author Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
date Tue, 13 Oct 2015 17:48:58 +0900
parents
children 803732b1fca8
comparison
equal deleted inserted replaced
84:f3e34b893a5f 95:afa8332a0e37
1 ; RUN: llc -march=amdgcn -mcpu=SI -verify-machineinstrs < %s | FileCheck -check-prefix=SI %s
2 ; RUN: llc -march=amdgcn -mcpu=tonga -verify-machineinstrs < %s | FileCheck -check-prefix=SI %s
3
4 declare i16 @llvm.convert.to.fp16.f32(float) nounwind readnone
5
6 ; SI-LABEL: {{^}}test_convert_fp32_to_fp16:
7 ; SI: buffer_load_dword [[VAL:v[0-9]+]]
8 ; SI: v_cvt_f16_f32_e32 [[RESULT:v[0-9]+]], [[VAL]]
9 ; SI: buffer_store_short [[RESULT]]
10 define void @test_convert_fp32_to_fp16(i16 addrspace(1)* noalias %out, float addrspace(1)* noalias %in) nounwind {
11 %val = load float, float addrspace(1)* %in, align 4
12 %cvt = call i16 @llvm.convert.to.fp16.f32(float %val) nounwind readnone
13 store i16 %cvt, i16 addrspace(1)* %out, align 2
14 ret void
15 }