diff options
| author | Stephen Hines <srhines@google.com> | 2015-03-23 12:10:34 -0700 |
|---|---|---|
| committer | Stephen Hines <srhines@google.com> | 2015-03-23 12:10:34 -0700 |
| commit | ebe69fe11e48d322045d5949c83283927a0d790b (patch) | |
| tree | c92f1907a6b8006628a4b01615f38264d29834ea /test/CodeGen/R600/llvm.AMDGPU.rsq.clamped.f64.ll | |
| parent | b7d2e72b02a4cb8034f32f8247a2558d2434e121 (diff) | |
| download | external_llvm-ebe69fe11e48d322045d5949c83283927a0d790b.zip external_llvm-ebe69fe11e48d322045d5949c83283927a0d790b.tar.gz external_llvm-ebe69fe11e48d322045d5949c83283927a0d790b.tar.bz2 | |
Update aosp/master LLVM for rebase to r230699.
Change-Id: I2b5be30509658cb8266be782de0ab24f9099f9b9
Diffstat (limited to 'test/CodeGen/R600/llvm.AMDGPU.rsq.clamped.f64.ll')
| -rw-r--r-- | test/CodeGen/R600/llvm.AMDGPU.rsq.clamped.f64.ll | 14 |
1 files changed, 13 insertions, 1 deletions
diff --git a/test/CodeGen/R600/llvm.AMDGPU.rsq.clamped.f64.ll b/test/CodeGen/R600/llvm.AMDGPU.rsq.clamped.f64.ll index 18854be..67f1d22 100644 --- a/test/CodeGen/R600/llvm.AMDGPU.rsq.clamped.f64.ll +++ b/test/CodeGen/R600/llvm.AMDGPU.rsq.clamped.f64.ll @@ -1,9 +1,21 @@ -; RUN: llc -march=r600 -mcpu=SI -verify-machineinstrs < %s | FileCheck -check-prefix=SI -check-prefix=FUNC %s +; RUN: llc -march=amdgcn -mcpu=SI -verify-machineinstrs < %s | FileCheck -check-prefix=SI -check-prefix=FUNC %s +; RUN: llc -march=amdgcn -mcpu=tonga -verify-machineinstrs < %s | FileCheck -check-prefix=VI -check-prefix=FUNC %s declare double @llvm.AMDGPU.rsq.clamped.f64(double) nounwind readnone ; FUNC-LABEL: {{^}}rsq_clamped_f64: ; SI: v_rsq_clamp_f64_e32 + +; VI: v_rsq_f64_e32 [[RSQ:v\[[0-9]+:[0-9]+\]]], s[2:3] +; TODO: this constant should be folded: +; VI: s_mov_b32 s[[ALLBITS:[0-9+]]], -1 +; VI: s_mov_b32 s[[HIGH1:[0-9+]]], 0x7fefffff +; VI: s_mov_b32 s[[LOW1:[0-9+]]], s[[ALLBITS]] +; VI: v_min_f64 v[0:1], [[RSQ]], s{{\[}}[[LOW1]]:[[HIGH1]]] +; VI: s_mov_b32 s[[HIGH2:[0-9+]]], 0xffefffff +; VI: s_mov_b32 s[[LOW2:[0-9+]]], s[[ALLBITS]] +; VI: v_max_f64 v[0:1], v[0:1], s{{\[}}[[LOW2]]:[[HIGH2]]] + define void @rsq_clamped_f64(double addrspace(1)* %out, double %src) nounwind { %rsq_clamped = call double @llvm.AMDGPU.rsq.clamped.f64(double %src) nounwind readnone store double %rsq_clamped, double addrspace(1)* %out, align 8 |