diff options
| author | Michael Kuperstein <michael.m.kuperstein@intel.com> | 2015-11-04 11:17:53 +0000 |
|---|---|---|
| committer | Michael Kuperstein <michael.m.kuperstein@intel.com> | 2015-11-04 11:17:53 +0000 |
| commit | b34de72269ee8bd92917465328c17aa7ece9f581 (patch) | |
| tree | 8ee149e1402766b8c74b7a6ca84849d24069cb97 /llvm/lib/Object/ELF.cpp | |
| parent | 3d7d92283274d835463d35829c3e6523f153612a (diff) | |
| download | bcm5719-llvm-b34de72269ee8bd92917465328c17aa7ece9f581.tar.gz bcm5719-llvm-b34de72269ee8bd92917465328c17aa7ece9f581.zip | |
[X86] DAGCombine should not introduce FILD in soft-float mode
The x86 "sitofp i64 to double" dag combine, in 32-bit mode, lowers sitofp
directly to X86ISD::FILD (or FILD_FLAG). This should not be done in soft-float mode.
llvm-svn: 252042
Diffstat (limited to 'llvm/lib/Object/ELF.cpp')
0 files changed, 0 insertions, 0 deletions
