diff options
| author | Philip Reames <listmail@philipreames.com> | 2019-05-14 04:43:37 +0000 |
|---|---|---|
| committer | Philip Reames <listmail@philipreames.com> | 2019-05-14 04:43:37 +0000 |
| commit | 3098e44daa764ed8c9d9bc94d1514f3e7e8c886e (patch) | |
| tree | 3dd97c8d7494b180d3cc29c4f243973bc2a159c4 /llvm/tools | |
| parent | e1cb2c0f404d9fe68f5f465a281be295ca24ec33 (diff) | |
| download | bcm5719-llvm-3098e44daa764ed8c9d9bc94d1514f3e7e8c886e.tar.gz bcm5719-llvm-3098e44daa764ed8c9d9bc94d1514f3e7e8c886e.zip | |
[X86] Prefer locked stack op over mfence for seq_cst 64-bit stores on 32-bit targets
This is a follow on to D58632, with the same logic. Given a memory operation which needs ordering, but doesn't need to modify any particular address, prefer to use a locked stack op over an mfence.
Differential Revision: https://reviews.llvm.org/D61863
llvm-svn: 360649
Diffstat (limited to 'llvm/tools')
0 files changed, 0 insertions, 0 deletions
