Date: Mon, 26 Jul 1999 23:10:01 +0200 (CEST) From: Arjan de Vet <Arjan.deVet@adv.iae.nl> To: cvs-all@freebsd.org Subject: Re: cvs commit: src/sys/kern vfs_bio.c vfs_lookup.c vfs_subr.c vfs_syscalls.c vfs_vnops.c src/sys/sys vnode.h Message-ID: <199907262110.XAA00399@adv.iae.nl> In-Reply-To: <199907261551.IAA46895@apollo.backplane.com> References: <Pine.BSF.4.10.9907260326500.15081-100000@janus.syracuse.net>
next in thread | previous in thread | raw e-mail | index | archive | help
In article <199907261551.IAA46895@apollo.backplane.com> you write: > Yes. Anyone who wants to play with VMIO'd directories can now. I ported the patch to 3.2-stable and am now running it on my home machine. No problems so far, but hadn't any problems with Matt's original patch either. Are there still problems on 3.X with this patch regarding NFS (not that I'm using NFS myself but other people may want to use it too)? > * First, the effect VMIOing directories has on the VM cache. We know for > a fact that in a directory-intensive environment directories will > stick around in the VM cache much longer then they previously stuck > around as B_MALLOC buffers in the buffer cache. The real question is Only access time updates seem to generate disk I/O now and then on my lightly loaded home machine. > whether the additional 'waste in space' creates a problem or not. > Smaller (< 1K) directories ate 1K as B_MALLOC buffers but will eat 4K > as VMIO-backed buffers. I don't care, I like the new silence of my disks :-). > * Second, the last time we tried to turn VMIO'd directories on > softupdates started panicing occassionally. We need to determine > if this still occurs and, if it does, try to collect information that > will enable Kirk to track the problem down after he gets back from > his vacation. I'm running with softupdates enabled and DDB in the 3.2-stable kernel. Not sure whether this could help solving possible problems on current given the big differences... > The best environment for both tests is a relatively busy server running > on a large dataset. If the server pages normallly the effect of #1 can I don't have a busy FreeBSD server running at the moment with Squid or a news server with a 'traditional' storage system, so I cannot help with this at the moment. We do have a dedicated FreeBSD-3.2 based CVS-server at work but I'm not sure whether we want to test this patch there; it sure would help speeding up the CVS operations I guess. > be seen most easily by determining whether the server pages more or less > due to the enabling of VMIO'd directories. Arjan 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?199907262110.XAA00399>