Date: Fri, 29 Oct 2004 11:06:33 -0700 (PDT) From: Don Lewis <truckman@FreeBSD.org> To: phk@critter.freebsd.dk Cc: current@FreeBSD.org Subject: Re: HEADSUP: UFS on GEOM now. Message-ID: <200410291806.i9TI6XQr026868@gw.catspoiler.org> In-Reply-To: <16180.1099045813@critter.freebsd.dk>
next in thread | previous in thread | raw e-mail | index | archive | help
On 29 Oct, Poul-Henning Kamp wrote: > > This is hopefully a no-op for all purposes than the increased access > flexibility. > > I am very interested if anybody can measure the performance difference > better than I could myself: 1.7% +/- 0.5%. > > Poul-Henning > > phk 2004-10-29 10:15:56 UTC > > FreeBSD src repository > > Modified files: > sys/ufs/ffs ffs_alloc.c ffs_extern.h ffs_rawread.c > ffs_snapshot.c ffs_softdep.c ffs_vfsops.c > ffs_vnops.c > sys/ufs/ufs inode.h ufs_vnops.c ufsmount.h > Log: > Move UFS from DEVFS backing to GEOM backing. > > This eliminates a bunch of vnode overhead (approx 1-2 % speed > improvement) and gives us more control over the access to the storage > device. > > Access counts on the underlying device are not correctly tracked and > therefore it is possible to read-only mount the same disk device multiple > times: > syv# mount -p > /dev/md0 /var ufs rw 2 2 > /dev/ad0 /mnt ufs ro 1 1 > /dev/ad0 /mnt2 ufs ro 1 1 > /dev/ad0 /mnt3 ufs ro 1 1 > > Since UFS/FFS is not a synchrousely consistent filesystem (ie: it caches > things in RAM) this is not possible with read-write mounts, and the system > will correctly reject this. Very cool!
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200410291806.i9TI6XQr026868>