Re: FSM Corruption (was: Could not read block at end of the relation) - Mailing list pgsql-bugs

On Thu, Apr 11, 2024 at 12:01:09PM -0400, Peter Geoghegan wrote:
> On Wed, Mar 13, 2024 at 12:55 PM Noah Misch <noah@leadboat.com> wrote:
> > That's a reasonable thing to worry about.  We could do wrong by trying too
> > hard to use an FSM slot, and we could do wrong by not trying hard enough.
> 
> Although it's not related to the problem you're working on, it seems
> like a good opportunity to bring up a concern about the FSM that I
> don't believe was discussed at any point in the past few years: I
> wonder if the way that fsm_search_avail() sometimes updates
> fsmpage->fp_next_slot with only a shared lock on the page could cause
> problems. At the very least, it's weird that we allow it.

fsm_search_avail() treats an out-of-range fp_next_slot like zero, so I'm not
seeing a correctness issue.  I bet changing it under an exclusive lock
wouldn't deliver better-optimized searches to an extent that pays for the
synchronization overhead, but it might.