diff options
author | Eli Friedman <efriedma@quicinc.com> | 2022-03-18 15:36:20 -0700 |
---|---|---|
committer | Eli Friedman <efriedma@quicinc.com> | 2022-03-18 15:54:38 -0700 |
commit | ddca66622c62721f2d05ffebefda17fc567d2cfd (patch) | |
tree | 076029b544060b88e2066bb9452e03464746f9c2 /flang/lib/Parser/executable-parsers.cpp | |
parent | f10f16a6a9906a062ea1f478a3623d5627eaba65 (diff) | |
download | llvm-ddca66622c62721f2d05ffebefda17fc567d2cfd.zip llvm-ddca66622c62721f2d05ffebefda17fc567d2cfd.tar.gz llvm-ddca66622c62721f2d05ffebefda17fc567d2cfd.tar.bz2 |
[ARM] Fix shouldExpandAtomicLoadInIR for subtargets without ldrexd.
Regression from 2f497ec3; we should not try to generate ldrexd on
targets that don't have it.
Also, while I'm here, fix shouldExpandAtomicStoreInIR, for consistency.
That doesn't really have any practical effect, though. On Thumb targets
where we need to use __sync_* libcalls, there is no libcall for stores,
so SelectionDAG calls __sync_lock_test_and_set_8 anyway.
Diffstat (limited to 'flang/lib/Parser/executable-parsers.cpp')
0 files changed, 0 insertions, 0 deletions