source: icGREP/icgrep-devel/llvm-3.8.0.src/test/CodeGen/AMDGPU/i8-to-double-to-float.ll @ 5027

Last change on this file since 5027 was 5027, checked in by cameron, 3 years ago

Upgrade to llvm 3.8

File size: 342 bytes
Line 
1;RUN: llc < %s -march=r600 -mcpu=redwood | FileCheck %s
2
3;CHECK: UINT_TO_FLT * T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
4
5define void @test(float addrspace(1)* %out, i8 addrspace(1)* %in) {
6  %1 = load i8, i8 addrspace(1)* %in
7  %2 = uitofp i8 %1 to double
8  %3 = fptrunc double %2 to float
9  store float %3, float addrspace(1)* %out
10  ret void
11}
Note: See TracBrowser for help on using the repository browser.