diff options
| author | Sanjay Patel <spatel@rotateright.com> | 2020-02-23 11:26:28 -0500 |
|---|---|---|
| committer | Sanjay Patel <spatel@rotateright.com> | 2020-02-23 11:36:53 -0500 |
| commit | a253a2a793cda34d1f6421ee9b7ca76a03fdfc59 (patch) | |
| tree | 8348abd082bcf0136e616ad4ffa514e91132df33 /llvm/lib/CodeGen/CodeGenPrepare.cpp | |
| parent | af69d5e10eb7bef52521fc052795690598040dc7 (diff) | |
| download | llvm-a253a2a793cda34d1f6421ee9b7ca76a03fdfc59.zip llvm-a253a2a793cda34d1f6421ee9b7ca76a03fdfc59.tar.gz llvm-a253a2a793cda34d1f6421ee9b7ca76a03fdfc59.tar.bz2 | |
[SDAG] fold fsub -0.0, undef to undef rather than NaN
A question about this behavior came up on llvm-dev:
http://lists.llvm.org/pipermail/llvm-dev/2020-February/139003.html
...and as part of backend improvements in D73978.
We decided not to implement a more general change that would have
folded any FP binop with nearly arbitrary constant + undef operand
to undef because that is not theoretically correct (even if it is
practically correct).
This is the SDAG-equivalent to the IR change in D74713.
Diffstat (limited to 'llvm/lib/CodeGen/CodeGenPrepare.cpp')
0 files changed, 0 insertions, 0 deletions