diff options
| author | Ahmed Bougacha <ahmed.bougacha@gmail.com> | 2016-01-14 19:45:36 +0000 |
|---|---|---|
| committer | Ahmed Bougacha <ahmed.bougacha@gmail.com> | 2016-01-14 19:45:36 +0000 |
| commit | 60b201b662789fb96f63170afe0131d7d0bd08dd (patch) | |
| tree | a67ff774e5f3eb0fd35f69f43ae27e5183137085 /llvm/lib/CodeGen/DFAPacketizer.cpp | |
| parent | 3e8d8c7d26684109c115b26da93256a37445cce8 (diff) | |
| download | llvm-60b201b662789fb96f63170afe0131d7d0bd08dd.zip llvm-60b201b662789fb96f63170afe0131d7d0bd08dd.tar.gz llvm-60b201b662789fb96f63170afe0131d7d0bd08dd.tar.bz2 | |
[CodeGen] Don't assume fp_to_fp16 produces i16 when legalizing it.
Since r230276, we support an improved legalization for f64->f16,
which goes through a temporary f32, improving codegen when
f32->f16 is legal but not f64->f16. This requires unsafe-fp-math.
However, that legalization assumed that the second step, producing
a pseudo-softened f16, had type i16. That's not true on targets
with illegal i16, such as ARM.
Use the initial f64->f16 result type instead.
llvm-svn: 257794
Diffstat (limited to 'llvm/lib/CodeGen/DFAPacketizer.cpp')
0 files changed, 0 insertions, 0 deletions