Date: Sat, 11 Nov 2000 11:18:06 -0800 From: Alfred Perlstein <bright@wintelcom.net> To: Andre Albsmeier <andre.albsmeier@mchp.siemens.de> Cc: Mike Batchelor <smujohnson@home.com>, questions@FreeBSD.ORG Subject: Re: Memory Caching Message-ID: <20001111111806.S11449@fw.wintelcom.net> In-Reply-To: <20001111184104.A13422@curry.mchp.siemens.de>; from andre.albsmeier@mchp.siemens.de on Sat, Nov 11, 2000 at 06:41:04PM %2B0100 References: <003201c04c8a$9209fda0$8ded4518@kldt1.bc.wave.home.com> <20001111021530.M11449@fw.wintelcom.net> <20001111184104.A13422@curry.mchp.siemens.de>
next in thread | previous in thread | raw e-mail | index | archive | help
* Andre Albsmeier <andre.albsmeier@mchp.siemens.de> [001111 09:41] wrote: > On Sat, 11-Nov-2000 at 02:15:30 -0800, Alfred Perlstein wrote: > > > > ... > > > > So either: > > 1) read /usr/src/sys/ufs/ffs/README.softupdates and enable softupdates > > -or- > > 2) mount your filesystems 'async' which can cause serious problems > > if you happen to crash while it's 'async' > > > > option 1 is a better idea. :) > > > > you may also want to try this setting: > > > > sysctl -w vfs.vmiodirenable=1 > > Is somewhere documented what this does and if it is dangerous? > It is not in the manpages... There was a single bug report over a year ago that has never to my knowledge resurfaced. I use it on all of my machines when I remeber to turn it on. What it does it allow the kernel to use the pagecache rather than malloc to store directory blocks, on machines with more than 64 megs of a RAM it's a big win although it wastes signifigant amounts of memory per directory cached. -- -Alfred Perlstein - [bright@wintelcom.net|alfred@freebsd.org] "I have the heart of a child; I keep it in a jar on my desk." To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20001111111806.S11449>