Date: Sat, 29 Jul 2006 16:02:14 -0700 From: Paul Allen <nospam@ugcs.caltech.edu> To: Michal Mertl <mime@traveller.cz> Cc: Peter Jeremy <peterjeremy@optushome.com.au>, freebsd-current@freebsd.org, Brian Candler <B.Candler@pobox.com> Subject: Re: vmstat's entries type Message-ID: <20060729230214.GI12597@groat.ugcs.caltech.edu> In-Reply-To: <1154212340.3609.18.camel@genius.i.cz> References: <200607251254.k6PCsBef092737@lurza.secnetix.de> <200607271058.13055.jhb@freebsd.org> <20060728121525.GA44917@uk.tiscali.com> <200607280928.36573.jhb@freebsd.org> <20060728134701.GA45273@uk.tiscali.com> <20060728210154.GC748@turion.vk2pj.dyndns.org> <1154189612.1565.10.camel@genius.i.cz> <20060729205655.GE748@turion.vk2pj.dyndns.org> <20060729211530.GA50342@uk.tiscali.com> <1154212340.3609.18.camel@genius.i.cz>
next in thread | previous in thread | raw e-mail | index | archive | help
Surely all you need to do is a cheap crit_enter,crit_exit
while updating the 64-bit per cpu counters. and on
a 64-bit arch you skip the crit_enter,crit_exit.
Seriously this is a bike shed. We can summarize it thus:
statistics should be maintained in 64-bit counters, these
counters should be per-cpu and consistent in that context,
nothing else should appear on the critical path.
**end of story**
Paul
p.s., and as a general of thumb, if a per-cpu solution
doesn't explode in complexity/memory usage it should be
preferred over a mutex by default.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20060729230214.GI12597>