From: | Stephen Frost <sfrost(at)snowman(dot)net> |
---|---|
To: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
Cc: | Marko Kreen <marko(at)l-t(dot)ee>, pgsql-hackers(at)postgresql(dot)org |
Subject: | Re: Spinlocks, yet again: analysis and proposed patches |
Date: | 2005-09-13 23:49:01 |
Message-ID: | 20050913234901.GN6026@ns.snowman.net |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
* Tom Lane (tgl(at)sss(dot)pgh(dot)pa(dot)us) wrote:
> I'm starting to think that we might have to succumb to having a compile
> option "optimize for multiprocessor" or "optimize for single processor".
> It's pretty hard to see how we'd alter a data structure decision like
> this on the fly.
I'd really hate to see this happen. In this case I don't think the
change you're proposing would have much of an impact on a uniprocessor
machine. Having seperate compile-time options for uniprocessor and
multiprocessor would open the gates for potentially other changes which
*would* have a more serious impact on one or the other when compiled for
the opposite. I think this would be a serious problem for binary
distributions and correspondingly their users.
Happy to test these changes, btw.
Also, I'm redoing my tests from the other patches with all the various
combinations; I'm a bit concerned that the recompiles I did (which were
just 'make', not 'make clean', 'make') for them previously didn't
actually recompile everything necessary. Sorry if the results differ.
Hope to have that all done this evening.
Thanks,
Stephen
From | Date | Subject | |
---|---|---|---|
Next Message | Gavin Sherry | 2005-09-14 00:05:59 | Re: Spinlocks, yet again: analysis and proposed patches |
Previous Message | Min Xu (Hsu) | 2005-09-13 23:32:02 | Re: Spinlocks, yet again: analysis and proposed patches |