OSDN Git Service

[PPC] Avoid incorrect fp-i128-fp lowering.
authorTim Shen <timshen91@gmail.com>
Tue, 23 Jan 2018 22:06:57 +0000 (22:06 +0000)
committerTim Shen <timshen91@gmail.com>
Tue, 23 Jan 2018 22:06:57 +0000 (22:06 +0000)
commit04b580d0637147d190765064d6eb8049a06c470b
tree47b6a89388ee693c1fc9c0ff606b2de716bf2a7c
parentbf4a8b9bd6216e6200bd2c657cd5358f33e3d56e
[PPC] Avoid incorrect fp-i128-fp lowering.

Summary:
Fix an issue that's similar to what D41411 fixed:
  float(__int128(float_var)) shouldn't be optimized to xscvdpsxds +
  xscvsxdsp, as they mean (float)(int64_t)float_var.

Reviewers: jtony, hfinkel, echristo

Subscribers: sanjoy, nemanjai, hiraditya, llvm-commits, kbarton

Differential Revision: https://reviews.llvm.org/D42400

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@323270 91177308-0d34-0410-b5e6-96231b3b80d8
lib/Target/PowerPC/PPCISelLowering.cpp
test/CodeGen/PowerPC/fp-int128-fp-combine.ll [new file with mode: 0644]