| From: | Neil Conway <neilc(at)samurai(dot)com> |
|---|---|
| To: | Bruce Momjian <pgman(at)candle(dot)pha(dot)pa(dot)us> |
| Cc: | pgsql-hackers <pgsql-hackers(at)postgresql(dot)org> |
| Subject: | Re: lwlocks and starvation |
| Date: | 2004-11-24 12:18:23 |
| Message-ID: | 41A47C0F.5030000@samurai.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
Bruce Momjian wrote:
> My guess is the existing behavior was designed to allow waking of
> multiple waiters _sometimes_ without starving of exclusive waiters.
Well, I think the current algorithm *does* allow starvation, at least in
some situations. Consider a workload in which a new shared reader
arrives every 50 ms, and holds the lock for, say, 500 ms. If an
exclusive waiter arrives, they will starve with the current algorithm.
> There should be a comment in the code explaining this usage and I bet it
> was intentional.
Oh, I bet it was intentional as well :) I'm mostly curious to see
exactly what the reasoning was, and whether it is necessary that we
preserve the FIFO behavior while considering optimizations.
-Neil
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Bruce Momjian | 2004-11-24 12:34:26 | Re: lwlocks and starvation |
| Previous Message | Bruce Momjian | 2004-11-24 11:55:03 | Re: lwlocks and starvation |