Date: Tue, 7 Dec 1999 21:01:39 -0500 From: Greg Lehey <grog@mojave.sitaranetworks.com> To: Julian Elischer <julian@whistle.com>, Matthew Dillon <dillon@apollo.backplane.com> Cc: Mike Smith <msmith@FreeBSD.ORG>, cvs-committers@FreeBSD.ORG, cvs-all@FreeBSD.ORG Subject: Re: cvs commit: src/sys/kern kern_shutdown.c Message-ID: <19991207210139.51228@mojave.sitaranetworks.com> In-Reply-To: <Pine.BSF.4.10.9912070107000.21432-100000@current1.whistle.com>; from Julian Elischer on Tue, Dec 07, 1999 at 01:11:41AM -0800 References: <199912070829.AAA75674@apollo.backplane.com> <Pine.BSF.4.10.9912070107000.21432-100000@current1.whistle.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Tuesday, 7 December 1999 at 1:11:41 -0800, Julian Elischer wrote: > > On Tue, 7 Dec 1999, Matthew Dillon wrote: >> >> The disk can lie about whether it has written data to the disk without >> messing softupdates up, but it cannot reorder writes that the system has >> told it must be ordered. SCSI has a tag to guarentee ordering (or not), >> but I don't know if IDE has the same sort of feature. > > Softupdates ensures the ordering by not isuing a dependent write until > it is told that the dependency write has completed. > > Thus, if the drive says a block is written but it is really in cache, and > the system issues another write that depended on the first, the second > must not be able to 'overtake' the first. THe safest way to guarantee this > is to just not allow the situation to occur. Even this is not enough for Vinum. Take one drive with a long request queue and another with no queue at all. Issue a command to the first drive, get the data into cache and issue the command to the second drive. The second drive writes immediately, before the first, then you have a power failure. Bang goes your order. Greg -- Finger grog@lemis.com for PGP public key See complete headers for address and phone numbers 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?19991207210139.51228>