diff options
| author | Matt Arsenault <Matthew.Arsenault@amd.com> | 2023-01-26 13:20:12 -0400 |
|---|---|---|
| committer | Matt Arsenault <arsenm2@gmail.com> | 2023-03-06 05:54:25 -0400 |
| commit | 9f4746b65f9fdea39975abeea37fd1f96475d369 (patch) | |
| tree | f72a20b571a82b8ffd48eb6054048e944a69a06e /llvm/lib/ExecutionEngine/Interpreter/ExternalFunctions.cpp | |
| parent | 9f493be7e695bcd915087d3f86032a9387bac605 (diff) | |
| download | llvm-9f4746b65f9fdea39975abeea37fd1f96475d369.zip llvm-9f4746b65f9fdea39975abeea37fd1f96475d369.tar.gz llvm-9f4746b65f9fdea39975abeea37fd1f96475d369.tar.bz2 | |
AMDGPU: Combine down fcopysign f64 magnitude
Copy through the low bits and only apply an f32
copysign to the high half. This is effectively
what we do for codegen anyway, but this provides
some combine benefits. The cases involving constants
show some small improvements.
https://reviews.llvm.org/D142682
Diffstat (limited to 'llvm/lib/ExecutionEngine/Interpreter/ExternalFunctions.cpp')
0 files changed, 0 insertions, 0 deletions