Date: Mon, 30 Nov 1998 08:53:47 +1030 From: Matthew Thyer <Matthew.Thyer@dsto.defence.gov.au> To: Warner Losh <imp@village.org> Cc: CURRENT@FreeBSD.ORG Subject: Re: aha delaying patch works for genuine 1542B Message-ID: <3661C973.6EA47509@dsto.defence.gov.au> References: <365CE867.62F68C8A@dsto.defence.gov.au> <199811262135.OAA88116@harmony.village.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Sorry about the long delay in replying.
(This is a work machine and I dont have access over the weekend).
Dmesg output for my card is:
aha0: at 0x234-0x237 irq 11 drq 6 on isa
aha0: AHA-1540/1542 64 head BIOS FW Rev. 0.5 (ID=41) SCSI Host Adapter, SCSI ID 7, 16 CCBs
I know its not at the standard I/O port. I had to move it due to
a conflict. I have got my kernel config file right for this.
Also you might want to know that I have 3 very old (quite old) full
height SCSI drives on the bus... dmesg output is:
da1 at aha0 bus 0 target 1 lun 0
da1: <SEAGATE ST41650 6230> Fixed Direct Access SCSI2 device
da1: 3.300MB/s transfers
da1: 1351MB (2767245 512 byte sectors: 64H 32S/T 1351C)
da3 at aha0 bus 0 target 3 lun 0
da3: <SEAGATE ST42100 7544> Fixed Direct Access SCSI2 device
da3: 3.300MB/s transfers
da3: 1813MB (3713130 512 byte sectors: 64H 32S/T 1813C)
da4 at aha0 bus 0 target 4 lun 0
da4: <SEAGATE ST42100 8224> Fixed Direct Access SCSI2 device
da4: 3.300MB/s transfers
da4: 1812MB (3711990 512 byte sectors: 64H 32S/T 1812C)
(these are wired down to their SCSI id. I had a da0 but it finally
blew up with all the thrashing I gave it).
Warner Losh wrote:
>
> In message <365CE867.62F68C8A@dsto.defence.gov.au> Matthew Thyer writes:
> : The card in question is a genuine Adaptec 1542B and I really
> : need it to work as it contains my /usr/src !!
>
> Which revision? The 1542B that I have doesn't need this delay. The
> probe string will tell you the revision.
>
> : As Graham Menhennitt said: "Could somebody please commit this change"
>
> Been there, done that. Justin and I have been talking about a better
> way to deal with this problem than putting in a magic delay. There
> may be another round of patches for the folks with the problem 1542B
> cards to test.
>
> : fuzz: {1} diff /sys/dev/aha/aha.c.orig /sys/dev/aha/aha.c
>
> diff -u generally produces better results. Also, I have a better
> comment stating that a delay to 1000 works, but 10mS is very long in
> card time, but very short in probe time. There was a similar delay in
> the old driver in places, but I've not looked closely enough to know
> if this was one of the places or not.
>
> Warner
>
> To Unsubscribe: send mail to majordomo@FreeBSD.org
> with "unsubscribe freebsd-current" in the body of the message
--
Matthew Thyer Phone: +61 8 8259 7249
Corporate Information Systems Fax: +61 8 8259 5537
Defence Science and Technology Organisation, Salisbury
PO Box 1500 Salisbury South Australia 5108
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-current" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3661C973.6EA47509>