Date: Mon, 03 Nov 2008 20:05:31 +0200 From: Giorgos Keramidas <keramida@freebsd.org> To: Peter Jeremy <peterjeremy@optushome.com.au> Cc: svn-src-head@freebsd.org, svn-src-all@freebsd.org, src-committers@freebsd.org, Ivan Voras <ivoras@freebsd.org> Subject: Re: svn commit: r184471 - in head/bin: cat cp Message-ID: <87y700it1w.fsf@kobe.laptop> In-Reply-To: <20081103075818.GK99398@server.vk2pj.dyndns.org> (Peter Jeremy's message of "Mon, 3 Nov 2008 18:58:18 %2B1100") References: <200810301405.m9UE5wwZ053250@svn.freebsd.org> <877i7pdzva.fsf@kobe.laptop> <9bbcef730811021534gb114db9x16a2d6c9a3ecee8b@mail.gmail.com> <20081103075818.GK99398@server.vk2pj.dyndns.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, 3 Nov 2008 18:58:18 +1100, Peter Jeremy <peterjeremy@optushome.com.au> wrote: > On 2008-Nov-03 00:34:47 +0100, Ivan Voras <ivoras@freebsd.org> wrote: >>2008/10/31 Giorgos Keramidas <keramida@freebsd.org>: >>The difference between the best and the worst is 247 context switches >>(total, not per second). >> >>> It looks like it's probably worth pushing the limit up to 8 MB when we >>> have lots of memory :) >> >>Yes, it's an improvement, but who will ever notice? :) > > Note that you will eventually hit a limit and start getting worse > performace: Once the buffer size exceeds the available free RAM > then cp/cat will start eating into inactive pages or (worse) active > pages for other processes. Yes, that's a possibility. Ivan is right that shaving 1-2 hundred more context _total_ context switches may be a small gain for the extra cost in buffer memory :)
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?87y700it1w.fsf>