[8479] in linux-scsi channel archive
Re: ncr53c400a
daemon@ATHENA.MIT.EDU (Jordi Mallach)
Fri Mar 24 07:07:50 2000
Date: Fri, 24 Mar 2000 13:10:43 +0100
From: Jordi Mallach <jordi@sindominio.net>
To: Karl-Heinz Herrmann <k.-h.herrmann@fz-juelich.de>
Cc: vorlon-kosh <vorlon-kosh@bol.com.br>, linux-scsi@vger.rutgers.edu
Message-ID: <20000324131043.C710@sindominio.net>
Mail-Followup-To: Karl-Heinz Herrmann <k.-h.herrmann@fz-juelich.de>,
vorlon-kosh <vorlon-kosh@bol.com.br>, linux-scsi@vger.rutgers.edu
Mime-Version: 1.0
Content-Type: multipart/signed; micalg=pgp-sha1;
protocol="application/pgp-signature"; boundary="7gGkHNMELEOhSGF6"
In-Reply-To: <XFMail.000324120836.k.-h.herrmann@fz-juelich.de>; from k.-h.herrmann@fz-juelich.de on Fri, Mar 24, 2000 at 12:08:36PM +0100
--7gGkHNMELEOhSGF6
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: quoted-printable
On Fri, Mar 24, 2000 at 12:08:36PM +0100, Karl-Heinz Herrmann wrote:
[snip snip]
> Then try to set the card to a valid combination. The insmod ... irq=3D255
> probably means "don't use irq" ? Then still you would have to tell the
> card a iobase it is willing to accept.=20
I managed to get the card detected some months ago without the irq option,
but Alan said it was better to use irq=3D255, although it worked without
specifying.
> Try to get the information to which it is set under winXX (you mentioned
> it's working there). Try the same values with Linux. Try a warmboot from
> win into Linux, maybe the cardsettings survive. If the card shows up
> then, try too figure out how to set the same without winXX (read
> /proc/ioports /proc/interrupts etc. again).
The warmboot is the only way I have managed to make that old card work. I'm
not currently using that box a whole lot, so I can't really test further.
The worst is that box does not dualboot anymore, so the warmboot is a bit
more complicated now.
> Then you could try to find some more info on that card under Linux.=20
> Try the hardware and support data base on www.suse.com or look at RedHat
> or Debian. Also the /usr/src/linux/Documentation could hold some valuable
> information to get the card recognised.=20
The documentation for this driver is a bit incomplete. I should try to
complete it with the info I collected in my previous queries to this list.
If I did, should I mail it to you for revision, Alan?
> [Alan]
> > > insmod g_NCR5380 ncr_addr=3D0x348 ncr_irq=3D255 ncr_53c400a=3D1
>=20
> [vorlon-kosh]
> > I already tried this command, but it says:
> > device or resource busy (or something like this).
>=20
> Could this also happen when the module is not loadable by some
> kernel-mismatch? That would usually print some other error.
Umh, nah, it's probably wrong params. I suffered that.
Searching for old 53C400a messages I find the following:
On Tue, 16 Nov 1999 23:21:01 +0100 I asked:
> In short, how do I know which of the two cards I have (port or memory
> mapped), and how do I know my io port?
Alan replied:
NCR53c400a is I/O mapped. I've only ever used it as a loadable module and in
my case that works
More, I said and Alan replied:
> boira:~# modprobe g_NCR5380 ncr_addr=3D0x290 ncr_53c400a=3D1
> scsi0 : IRQ0 not free, interrupts disabled
Specify irq=3D255 if you want to say "do this polled"
> boira:~# modprobe g_NCR5380 ncr_addr=3D0x290 ncr_53c400a=3D1 ncr_irq=3D5
> scsi0 : at port 0x290 irq 5 options CAN_QUEUE=3D16 CMD_PER_LUN=3D2
> release=3D1
> generic options AUTOPROBE_IRQ AUTOSENSE PSEUDO DMA USLEEP, USLEEP_POLL=3D1
> USLEEP_SLEEP=3D20 generic release=3D7
Thats perfect, IRQ and all. It'll now merely suck horribly being an 8bit
ISA controller 8)
But this was after a warm boot. I don't remember being able to detect the
card without pre-booting windows, but as I said, I'll get that box back to
me this weekend to test.
I hope this helps someone.
Jordi
--=20
Jordi Mallach P=E9rez || jordi@pusa.informat.uv.es || Rediscovering Freed=
om,
ka Oskuro in RL-MUD || jordi@sindominio.net || Using Debian GNU/Linux
http://sindominio.net GnuPG public information: pub 1024D/917A225E=
=20
telnet pusa.uv.es 23 73ED 4244 FD43 5886 20AC 2644 2584 94BA 917A 225E
--7gGkHNMELEOhSGF6
Content-Type: application/pgp-signature
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.1 (GNU/Linux)
Comment: For info see http://www.gnupg.org
iD8DBQE421tDJYSUupF6Il4RAquMAKDIKgUwbNdPDIdCXpKGRV+Fta1RjgCgtNtR
5tZsXpjoVKC/EKA5/Vqk9S8=
=SIWq
-----END PGP SIGNATURE-----
--7gGkHNMELEOhSGF6--
-
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@vger.rutgers.edu
