Date: Wed, 23 Sep 1998 13:50:02 -0400 (EDT) From: Simon Shapiro <shimon@simon-shapiro.org> To: Eivind Eklund <eivind@yes.no> Cc: scsi@FreeBSD.ORG Subject: Re: DPT boot Message-ID: <XFMail.980923135002.shimon@simon-shapiro.org> In-Reply-To: <19980923131222.21236@follo.net>
next in thread | previous in thread | raw e-mail | index | archive | help
Eivind Eklund, On 23-Sep-98 you wrote: > On Tue, Sep 22, 1998 at 11:06:48PM -0400, Simon Shapiro wrote: > > Eivind Eklund, On 22-Sep-98 you wrote: > > > Any clues? Anything I should test? (I can hook up a pretty > > > complete > > > debugging environment here, including doing remote kernel debugging > > > if > > > necessary - at the moment I also have 'scratch' disks I can use for > > > testing, though not having this controller operational means I my > > > main > > > workstation is not usable, so I'd prefer to avoid tests that take a > > > long time). > > > > I assume this controller works on the pre-CAM driver... :-) > > Yes, perfectly. The only problem is that the DPT is attempting to > outsmart > the BIOS, but once it is in FreeBSD, everything works perfectly. I talked to someone who knows at DPT; This is a knwon problem on certain motherboards. I suspect that due to age, etc., they will not fix this minor problem anytime soon. Anyway, this is a BIOS geometry, if I remember correctly. > > > Oh, and while this was happening, the DPT kept a steady light #3 (as > > > counted from the connector-less end of the DPT). > > > > On a 3224 it is bus transfer from adapter, or adapter reset (depends on > > how > > you look at it, and assuming first LED is 1). > > First LED is 1, and I counted from the end of the board that hasn't got > a > connector to the outside world (ie, not the end where the PCI-bus and > the > disk-connectors are, but the end with the little black plastic thing > that > support the board. The end I'm talking about is where usual PCI boards > are > no more (due to being smaller) :-) Yup. This is the Data Transfer From Adaper LED. It means a DMA trasfer starts and then the card hangs. Either Justin is correct and you walked on a firmware bug, or the new driver is missing something in the initialization, that was done correctly in the precambian days. I will look at it, once Justin has saturated his changes to the driver. > > If it is a transfer LED, then the DMA engine is stuck. If it is a > > reset, > > that means the controller was reset and never released. In either case > > I > > would suspect the initialization code does not work correctly. I have > > some work-work emergencies to handle right now, but will start > > debugging > > this driver any day soon. > > OK, great. Tell me if you want me to do anything. If you start wiping > disks, a pre-alpha version of my testbox building scripts are at > http://www.freebsd.org/~eivind/testbox-0.1.tar.gz I am going to test on Bras, and ONLY on brass. IF anyone (Justin?) needs access to it, just say so. Once Brass is passing regression, I will test on Nomis. Then Sendero, then performance... Sincerely Yours, Shimon@Simon-Shapiro.ORG 770.265.7340 Simon Shapiro Unwritten code has no bugs and executes at twice the speed of mouth To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-scsi" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?XFMail.980923135002.shimon>