Date: Mon, 18 Mar 2002 09:54:17 -0500 From: "Brian F. Feldman" <green@FreeBSD.ORG> To: Matthew Dillon <dillon@apollo.backplane.com> Cc: Hiten Pandya <hitmaster2k@yahoo.com>, current@FreeBSD.ORG, cvs-all@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, tanimura@r.dl.itc.u-tokyo.ac.jp Subject: Re: cvs commit: src/sys/kern kern_mtxpool.c src/sys/sys kernel.h src/sys/vm vm_fault.c vm_glue.c vm_map.c vm_map.h vm_pageout.c vm_zone.c Message-ID: <200203181454.g2IEsHE88877@green.bikeshed.org> In-Reply-To: Your message of "Sun, 17 Mar 2002 16:48:24 PST." <200203180048.g2I0mOc40231@apollo.backplane.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Matthew Dillon <dillon@apollo.backplane.com> wrote:
>
> :
> :Hi,
> :
> :Sorry to unwantedly butt in, but the patch supplied by Seigo actually
> :solved the vm_map.c locking problems which used to come up on my system.
> :Just some info. :)
> :
> :Regards,
> :
> : -- Hiten Pandya
> : -- <hiten@uk.FreeBSD.org>
>
> It fixed some things, it broke some things. Pretty much standard
> fare for anyone who has ever done work on the vm_map lock, including
> yours truely. John Dyson couldn't get it right, David Greenman couldn't
> get it right, I couldn't get it completely right... getting it to do
> the right thing even under -stable is difficult, which is why I am
> suggesting that it simply be made into an exclusive lock in -current.
In addition to the fact it doesn't actually serialize the general
modification of the vm_map {} structure itself, just certain kinds of
changes, so is easily a primary reason that the VM system as it stands now
is inherently single-threaded.
--
Brian Fundakowski Feldman \'[ FreeBSD ]''''''''''\
<> green@FreeBSD.org <> bfeldman@tislabs.com \ The Power to Serve! \
Opinions expressed are my own. \,,,,,,,,,,,,,,,,,,,,,,\
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe cvs-all" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200203181454.g2IEsHE88877>