eddyz87 wrote:
@peilin-ye ,
> However, I didn't want to use `lock` because I feel like it's too similar to
> the x86 `LOCK` prefix (implies a full memory barrier, which could be
> confusing here). WDYT? Cc: @yonghong-song @4ast
Well, we already confuse people in a way, since existing `lock *(u64 *)(r1 +
0x0) += r2` does not imply full barrier (it is translated to `stadd` by arm jit
in kernel, which is documented as "has no memory ordering semantics"). So,
`lock ... {acquire,release}` shouldn't make things worse.
https://github.com/llvm/llvm-project/pull/108636
_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
