[clang] [clang][CodeGen] Emit atomic IR instead of libcalls for misaligned po… (PR #73176)
via cfe-commits
cfe-commits at lists.llvm.org
Tue Nov 28 11:27:46 PST 2023
Logikable wrote:
I believe Clang currently does not properly propagate alignment information for certain IR (e.g. atomicrmw and cmpxchg). Look at the definitions of `CreateAtomic*` in Clang. The patches in https://bugs.llvm.org/show_bug.cgi?id=27168 did not address this (thanks @MaskRay for helping me find this).
https://github.com/llvm/llvm-project/pull/73176
More information about the cfe-commits
mailing list