From owner-freebsd-current Sun Aug 22 12:12:23 1999 Delivered-To: freebsd-current@freebsd.org Received: from alpo.whistle.com (alpo.whistle.com [207.76.204.38]) by hub.freebsd.org (Postfix) with ESMTP id 9D82E14C19 for ; Sun, 22 Aug 1999 12:11:43 -0700 (PDT) (envelope-from julian@whistle.com) Received: from current1.whistle.com (current1.whistle.com [207.76.205.22]) by alpo.whistle.com (8.9.1a/8.9.1) with SMTP id LAA71629; Sun, 22 Aug 1999 11:46:28 -0700 (PDT) Date: Sun, 22 Aug 1999 11:48:02 -0700 (PDT) From: Julian Elischer To: Poul-Henning Kamp Cc: Matthew Dillon , Maxim Sobolev , current@FreeBSD.ORG Subject: Re: Sync(8) doesn't have any effect on softupdates-enabled filesystem In-Reply-To: <4301.935346213@critter.freebsd.dk> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG yes and no. the sync wheel indicates buffers to wrrite in an order that will probably ensure that early dependencies are satisfied before later ones, however even after being written, a buffer might have remaining unsatisfied dependencies that required that teh on disk and in momory version sstill do not match. In this case it will have been rewritten back to the sync-wheel further in the future. What one can do is acellerate teh speed that teh daemon goes around the wheel, but you must never go faster than the disks can write, because if you queue a block that depends on something that hasn't got to disk yet, you will be queueing the old version. On Sun, 22 Aug 1999, Poul-Henning Kamp wrote: > In message <199908221758.KAA80382@apollo.backplane.com>, Matthew Dillon writes: > >:> structures used internally by softupdates are not condusive to doing a > >:> hard-sync. > >: > >:I gues sync needs to set a flag which makes the sync'er go through all > >:buckets with no delay and then wake the sync'ing process afterwards... > >: > >:-- > >:Poul-Henning Kamp FreeBSD coreteam member > > > > It won't help. What needs to happen is for the VOP_FSYNC in ffs to > > figure out buffer<->buffer dependancies > > But the buffer to buffer dependencies are already recorded in the > sequence on the "sync-wheel" which the syncer daemon runs through, > isn't it ? > > -- > Poul-Henning Kamp FreeBSD coreteam member > phk@FreeBSD.ORG "Real hackers run -current on their laptop." > FreeBSD -- It will take a long time before progress goes too far! > > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-current" in the body of the message > To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message