All of lore.kernel.org
 help / color / mirror / Atom feed
* Who maintains the atp870u driver? (ACARD PCI SCSI)
@ 2004-04-04 15:36 Stuart Longland
  2004-04-07 14:02 ` Marcelo Tosatti
  0 siblings, 1 reply; 3+ messages in thread
From: Stuart Longland @ 2004-04-04 15:36 UTC (permalink / raw)
  To: linux-kernel

[-- Attachment #1: Type: text/plain, Size: 3230 bytes --]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi All,
	Does anyone still maintain the atp870u SCSI driver, and if so, is it
possible for them to contact me?  I had a look in the atp870u.c file --
if I read this correctly, this driver hasn't changed much in the last
few years.  (Last log entry was in 2001).

	Basically, I recently acquired a PCI ACARD SCSI card (can't recall what
chipset exactly) and decided to use that in a Gateway Microserver server
appliance I had laying around.  The intent was to plug it into an
external tower of hard drives to make a nice compact fileserver for lans.

	The machine is a clone of the Cobalt Qube 2 (2800) server appliance,
running a 250MHz (?) MIPS "Nervada" R5200 CPU, 64MB EDO RAM, and 10GB
IBM IDE HDD internal.  It has one PCI slot, attached to a Galileo PCI
controller.  When first bought, it had a Lucent Winmodem installed
(Gateway branded).  All I've done is simply replaced the modem with this
SCSI card.  The box is running Debian 3.0 with Linux Kernel 2.4.24-pre2
(from linux-mips CVS).

	If I fire the box up, then run 'modprobe atp870u' -- with no devices
attached (or switched off), the driver loads, no worries.  However, if I
try doing the same with a device attached (and switched on), I get SCSI
timeout errors, and the driver initialisation stalls. -- I have to
reboot to unload the driver. (See attachment)

	The device in question is the tower of hard drives I mentioned above
(has 3x 18.2GB and 1x 9.1GB HDDs).  I know from previous testing that
neither the tower, the cable nor the terminator are at fault -- they
work just fine on both an SGI Indy and my main machine (with an Advansys
SCSI card) without any hassles.  I have also tried plugging in an
external tape drive, and got the same results.

	The only thing that I haven't tested is whether the card works (it was
sold second hand with a CDRW burner -- I presumed it did, and from the
tests tonight, there's at least a little bit of life), and what its
compatability is with Linux (in particular Linux/MIPS).  I've also tried
an Advansys SCSI card (the one in my main box) in the same machine with
no joy (it didn't work _at all_ -- so the ACARD host is currently doing
better right now)

	I'm not able to hack the driver myself, as I'm a newbie when it comes
to kernel hacking. (Otherwise I'd be looking into it right
now...unfortunately I'm a C newbie, and therefore the code is right over
my head.)  I might also later try the Linux-MIPS crew if this turns out
to be a MIPS-specific thing (I can't rule that one out).

Any assistance would be appreciated,
- --
+-------------------------------------------------------------+
| Stuart Longland           stuartl at longlandclan.hopto.org |
| Brisbane Mesh Node: 719             http://stuartl.cjb.net/ |
| I haven't lost my mind - it's backed up on a tape somewhere |
| Atomic Linux Project    <--->    http://atomicl.berlios.de/ |
+-------------------------------------------------------------+
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.2 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org

iD8DBQFAcCuHuarJ1mMmSrkRAsUHAJ9gyP5C/rGCes54tKVY4PC2V155fwCcCWMT
nDLFI1+sAvydOfNoRQVT2J8=
=L26G
-----END PGP SIGNATURE-----

[-- Attachment #2: qube_scsi_log.txt.gz --]
[-- Type: application/gzip, Size: 1234 bytes --]

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: Who maintains the atp870u driver? (ACARD PCI SCSI)
  2004-04-04 15:36 Who maintains the atp870u driver? (ACARD PCI SCSI) Stuart Longland
@ 2004-04-07 14:02 ` Marcelo Tosatti
  2004-04-12 10:49   ` Stuart Longland
  0 siblings, 1 reply; 3+ messages in thread
From: Marcelo Tosatti @ 2004-04-07 14:02 UTC (permalink / raw)
  To: Stuart Longland; +Cc: linux-kernel

On Mon, Apr 05, 2004 at 01:36:39AM +1000, Stuart Longland wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> Hi All,
> 	Does anyone still maintain the atp870u SCSI driver, and if so, is it
> possible for them to contact me?  I had a look in the atp870u.c file --
> if I read this correctly, this driver hasn't changed much in the last
> few years.  (Last log entry was in 2001).
> 
> 	Basically, I recently acquired a PCI ACARD SCSI card (can't recall 
> 	what
> chipset exactly) and decided to use that in a Gateway Microserver server
> appliance I had laying around.  The intent was to plug it into an
> external tower of hard drives to make a nice compact fileserver for lans.
> 
> 	The machine is a clone of the Cobalt Qube 2 (2800) server appliance,
> running a 250MHz (?) MIPS "Nervada" R5200 CPU, 64MB EDO RAM, and 10GB
> IBM IDE HDD internal.  It has one PCI slot, attached to a Galileo PCI
> controller.  When first bought, it had a Lucent Winmodem installed
> (Gateway branded).  All I've done is simply replaced the modem with this
> SCSI card.  The box is running Debian 3.0 with Linux Kernel 2.4.24-pre2
> (from linux-mips CVS).
> 
> 	If I fire the box up, then run 'modprobe atp870u' -- with no devices
> attached (or switched off), the driver loads, no worries.  However, if I
> try doing the same with a device attached (and switched on), I get SCSI
> timeout errors, and the driver initialisation stalls. -- I have to
> reboot to unload the driver. (See attachment)
> 
> 	The device in question is the tower of hard drives I mentioned above
> (has 3x 18.2GB and 1x 9.1GB HDDs).  I know from previous testing that
> neither the tower, the cable nor the terminator are at fault -- they
> work just fine on both an SGI Indy and my main machine (with an Advansys
> SCSI card) without any hassles.  I have also tried plugging in an
> external tape drive, and got the same results.
> 
> 	The only thing that I haven't tested is whether the card works (it 
> 	was
> sold second hand with a CDRW burner -- I presumed it did, and from the
> tests tonight, there's at least a little bit of life), and what its
> compatability is with Linux (in particular Linux/MIPS).  I've also tried
> an Advansys SCSI card (the one in my main box) in the same machine with
> no joy (it didn't work _at all_ -- so the ACARD host is currently doing
> better right now)
> 
> 	I'm not able to hack the driver myself, as I'm a newbie when it comes
> to kernel hacking. (Otherwise I'd be looking into it right
> now...unfortunately I'm a C newbie, and therefore the code is right over
> my head.)  I might also later try the Linux-MIPS crew if this turns out
> to be a MIPS-specific thing (I can't rule that one out).
> 
> Any assistance would be appreciated,

No one really maintains it officially.

James Bottomley and Doug Ledford have done some fixes for it on v2.6, you might want 
trying to ask them directly.

Sorry.

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: Who maintains the atp870u driver? (ACARD PCI SCSI)
  2004-04-07 14:02 ` Marcelo Tosatti
@ 2004-04-12 10:49   ` Stuart Longland
  0 siblings, 0 replies; 3+ messages in thread
From: Stuart Longland @ 2004-04-12 10:49 UTC (permalink / raw)
  To: Marcelo Tosatti; +Cc: linux-kernel

Marcelo Tosatti wrote:

> 
> No one really maintains it officially.
> 
> James Bottomley and Doug Ledford have done some fixes for it on v2.6, you might want 
> trying to ask them directly.
> 
> Sorry.

	Ahh okay.  I might try Linux 2.6 on this box, however after testing 
this card in an x86 machine I had, and watching it fail to complete 
POST, I'm guessing my problem is that the card is in fact, dead.

	I merely pulled it out of the Microserver, and plonked in my dedicated 
games server (P4 1.4GHz, 256MB RDRAM, MSI Motherboard -- Gentoo Linux 
1.4) and watched it boot.  I heard no beeps, saw nothing on screen, and 
the whole system just sat there.

	Near the sound card sockets, there's a set of 4 two-colour LEDs 
arranged vertically, which indicate system status -- I don't know what 
the meaning is.  (It'll probably be in the handbook I don't have -- the 
board is second hand off Ebay).  Normally these are all glowing green 
when the system is fully up.  Looking at it now, it halts with them 
(going towards board), green, red, red, red.  On removing the card, the 
machine comes up just fine.  I take it this means the card is dead.

	Out of interest, the card itself is an "ACARD AEC-6710S" with an 
ATP870IU-C chipset.

	Odd that the Microserver should work with it though -- even if it gets 
SCSI parity errors -- but I spose this explains the odd error messages. 
  Hrmm, looks like I'm back on the search for another SCSI card that 
fits... :-/ (Anyone managed to get a SCSI card working in a Cobalt Qube 
or Gateway Microserver?)

-- 
+-------------------------------------------------------------+
| Stuart Longland           stuartl at longlandclan.hopto.org |
| Brisbane Mesh Node: 719             http://stuartl.cjb.net/ |
| I haven't lost my mind - it's backed up on a tape somewhere |
| Atomic Linux Project    <--->    http://atomicl.berlios.de/ |
+-------------------------------------------------------------+

^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2004-04-12 10:49 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2004-04-04 15:36 Who maintains the atp870u driver? (ACARD PCI SCSI) Stuart Longland
2004-04-07 14:02 ` Marcelo Tosatti
2004-04-12 10:49   ` Stuart Longland

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.