https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=265974

--- Comment #3 from Konstantin Belousov <[email protected]> ---
I believe you are right, but I would wait for an opinion of people who
spent a time with smr code.

That said, atomic_add_acq on x86 has the sequentially consistent semantic
already (which is why you said that it works on Intel, right?).  So the
#ifdef from the patch in smr_enter() is not needed, use seq_cst fence
for all arches.

-- 
You are receiving this mail because:
You are the assignee for the bug.

Reply via email to