Date: Sun, 10 Nov 1996 07:16:40 +0300 (MSK) From: "=?KOI8-R?Q?=E1=CE=C4=D2=C5=CA_=FE=C5=D2=CE=CF=D7?=" (Andrey A. Chernov) <ache@nagual.ru> To: gibbs@freefall.freebsd.org (Justin T. Gibbs) Cc: current@freebsd.org, scsi@freebsd.org Subject: Re: SCSI system becomes broken in -current? Message-ID: <199611100416.HAA00208@nagual.ru> In-Reply-To: <199611100304.TAA07061@freefall.freebsd.org> from "Justin T. Gibbs" at "Nov 9, 96 07:04:22 pm"
next in thread | previous in thread | raw e-mail | index | archive | help
> >Tagged queuing always work with it before recent days. I don't know about > >any firmware upgrade needed for this drive. BTW, it is almost impossible > >to get firmware upgrade in Russia or snail mail will costs so much money as > >compared to new device price. > >How do you know that firmware upgrade needed for this drive? > >Could you please resend me your sources? > > Drive firmware upgrades are usually ftp-able and are downloaded to the > drive on the user's system. As for my sources, Gary Palmer mentioned some > time ago that his 1060 would not work with tagged queueing and there is an > entry in one of the Linux hardware FAQs about this drive. What I find in Linux FAQ: ---------------------------------------------------------------------------------------------- Some Conner CFP1060S drives may have problems with Linux and ext2fs. The symptoms are inode errors during e2fsck and corrupt file systems. Conner has released a firmware upgrade to fix this problem, contact Conner at 1-800-4CONNER (US) or +44-1294-315333 (Europe). Have the microcode version (found on the drive label, 9WA1.6x) handy when you call. ---------------------------------------------------------------------------------------------- I have 9WA1.68 label, so it seems my drive is affected. BTW, I can't call any of this phones, International calls costs too much here. If you got this upgrade somehow, please don't forget me... > They probably have a quirk entry for it. Either that, or there is a new > bug in the driver. The changes I made shouldn't have effected tagged > queueing, so the bug should be reproducible with tagged queueing disabled. I turn tagged queue off about two hours ago and extensively test my drives under hard load. And still not have this bug, so it seems it _is_ tagged queue related. > >Could you please add device flags to turn tagged queuing on per-drive basis > >instead on for whole controller? My second drive is new IBM disk which > >works well with tagged queueing and I not want to turn it off for this drive > >too. > > Don't you read your commit mail? I added this feature to the new SCSI > branch code last weekend. Sorry, I not pay attention on SCSI branch, I track -current only. My home SCSI drives is valuable enough for me to run experimental code on them. -- Andrey A. Chernov <ache@nagual.ru> http://www.nagual.ru/~ache/
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199611100416.HAA00208>