Date: Sun, 24 Feb 2002 12:59:55 +1100 (EST) From: Bruce Evans <bde@zeta.org.au> To: Matthew Dillon <dillon@apollo.backplane.com> Cc: Terry Lambert <tlambert2@mindspring.com>, Alfred Perlstein <alfred@FreeBSD.ORG>, Bosko Milekic <bmilekic@unixdaemons.com>, Seigo Tanimura <tanimura@r.dl.itc.u-tokyo.ac.jp>, <current@FreeBSD.ORG>, John Baldwin <jhb@FreeBSD.ORG> Subject: Re: malloc_bucket() idea (was Re: How to fix malloc.) Message-ID: <20020224125005.N31250-100000@gamplex.bde.org> In-Reply-To: <200202240049.g1O0noD15327@apollo.backplane.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sat, 23 Feb 2002, Matthew Dillon wrote:
> Bruce, I've looked at the vector assembly and it looks like cleaning
> up critical_enter() and critical_exit() for i386 ought to be simple.
>
> If you have a complete patch set I would like to see it posted for
> review or comitted, or if you don't want to deal with the commit I
> would love to have it so I can bang it into shape for commit.
Patches for most of my sys tree are in sys.tar.gz in my home directory
on freefall.
> Having a sleek critical_enter()/critical_exit() would double sched_lock's
> performance.
I'm not sure about that. Perhaps I've already optimized sched_lock
enough that I don't notice. My changes were actually directed at
avoiding sched_lock's interrupt latency and it wasn't until the
critical_enter()/ exit() calls pessimized mtx_{lock,unlock}_spin in
-current that I managed to share enough code (the critnest
increment/decrement) to avoid losing efficiency.
Bruce
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-current" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20020224125005.N31250-100000>