On Mon, Mar 28, 2022 at 12:40 PM Greg Stark <stark@mit.edu> wrote:
> I'm seeing a recovery test failure. Not sure if this represents an
> actual bug or just a test that needs to be adjusted for the new
> behaviour.
>
> https://cirrus-ci.com/task/5711008294502400
I doubt that the patch's use of pg_memory_barrier() in places like
_bt_killitems() is correct. There is no way to know for sure if this
novel new lockless algorithm is correct or not, since it isn't
explained anywhere.
The existing use of memory barriers is pretty much limited to a
handful of performance critical code paths, none of which are in
access method code that reads from shared_buffers. So this is not a
minor oversight.
--
Peter Geoghegan