diff options
| author | Craig Topper <craig.topper@intel.com> | 2020-10-09 11:48:10 -0700 |
|---|---|---|
| committer | Craig Topper <craig.topper@intel.com> | 2020-10-09 11:55:24 -0700 |
| commit | f34bb06935aa3bab353d70d515b767fdd2f5625c (patch) | |
| tree | 832b247172354f39b2e35ea782c4f246d37f9f0d /llvm/lib/Bitcode/Reader/MetadataLoader.cpp | |
| parent | 662024df331bd1f1a206678435e51232683e3cf6 (diff) | |
| download | llvm-f34bb06935aa3bab353d70d515b767fdd2f5625c.zip llvm-f34bb06935aa3bab353d70d515b767fdd2f5625c.tar.gz llvm-f34bb06935aa3bab353d70d515b767fdd2f5625c.tar.bz2 | |
[X86] When expanding LCMPXCHG16B_NO_RBX in EmitInstrWithCustomInserter, directly copy address operands instead of going through X86AddressMode.
I suspect getAddressFromInstr and addFullAddress are not handling
all addresses cases properly based on a report from MaskRay.
So just copy the operands directly. This should be more efficient
anyway.
Diffstat (limited to 'llvm/lib/Bitcode/Reader/MetadataLoader.cpp')
0 files changed, 0 insertions, 0 deletions