MO-Disk on PPC with 53C94 SCSI-chip

MO-Disk on PPC with 53C94 SCSI-chip

[Andreas Tobler]

Hi,

I have a LinuxPPC R5 system running on a PowerMac 7200 with the
2.2.12pre8 kernel. 
My problem is my MO-drive from HP see below for type and SCSI-Spec (it's
not SCSI-2)
I got it to work when I preformatted it under MacOS ->  partition-table, partition.
When I boot with the medium inserted the kernel recognizes the MO
without any problem.
I can create a fs on the partiton, I can read/write from/to it. BUT,
when I try to change the side of the media I get the same fs tree as on
the first side.

So I tried booting without media in to verify if it works on the clean
way. This what the dmesg says:

----
[snip]
(scsi0) <Adaptec AHA-294X Ultra SCSI host adapter> found at PCI 14/0
(scsi0) Wide Channel, SCSI ID=7, 16/255 SCBs
(scsi0) Warning - detected auto-termination
(scsi0) Please verify driver detected settings are correct.
(scsi0) If not, then please properly set the device termination
(scsi0) in the Adaptec SCSI BIOS by hitting CTRL-A when prompted
(scsi0) during machine bootup.
(scsi0) Cables present (Int-50 NO, Int-68 YES, Ext-68 NO)
(scsi0) Downloading sequencer code... 413 instructions downloaded
scsi0 : Adaptec AHA274x/284x/294x (EISA/VLB/PCI-Fast SCSI) 5.1.19/3.2.4
       <Adaptec AHA-294X Ultra SCSI host adapter>
scsi1 : 53C94
scsi : 2 hosts.
(scsi0:0:0:0) Synchronous at 20.0 Mbyte/sec, offset 8.
  Vendor: Quantum   Model: XP32150W          Rev: L912
  Type:   Direct-Access                      ANSI SCSI revision: 02
Detected scsi disk sda at scsi0, channel 0, id 0, lun 0
  Vendor: HP        Model: C3324A            Rev: 5144
  Type:   Direct-Access                      ANSI SCSI revision: 02
Detected scsi disk sdb at scsi1, channel 0, id 1, lun 0
  Vendor: QUANTUM   Model: LIGHTNING 730S    Rev: 241E
  Type:   Direct-Access                      ANSI SCSI revision: 02
Detected scsi disk sdc at scsi1, channel 0, id 2, lun 0
  Vendor: SONY      Model: CD-ROM CDU-8005   Rev: 1.0j
  Type:   CD-ROM                             ANSI SCSI revision: 02
Detected scsi CD-ROM sr0 at scsi1, channel 0, id 3, lun 0
  Vendor: QUANTUM   Model: FIREBALL1280S     Rev: 630G
  Type:   Direct-Access                      ANSI SCSI revision: 02
Detected scsi disk sdd at scsi1, channel 0, id 4, lun 0
  Vendor: HP        Model: C1716T            Rev: 3404
  Type:   Optical Device                     ANSI SCSI revision: 01 CCS
Detected scsi disk sde at scsi1, channel 0, id 5, lun 0
scsi : detected 1 SCSI cdrom 5 SCSI disks total.
Uniform CDROM driver Revision: 2.55
SCSI device sda: hdwr sector= 512 bytes. Sectors= 4406960 [2151 MB] [2.2 GB]
SCSI device sdb: hdwr sector= 512 bytes. Sectors= 2354660 [1149 MB] [1.1 GB]
SCSI device sdc: hdwr sector= 512 bytes. Sectors= 1431760 [699 MB] [0.7 GB]
SCSI device sdd: hdwr sector= 512 bytes. Sectors= 2503872 [1222 MB] [1.2 GB]
sde: Spinning up
disk.......................................................................................................not responding...
sde : READ CAPACITY failed.
sde : status = 0, message = 00, host = 0, driver = 28 
sde : extended sense code = 2 
sde : block size assumed to be 512 bytes, disk size 1GB.  
eth0: MACE at 00:05:02:44:21:ad, chip revision 9.64
eth1: DE450-CA at 0x0400 (PCI bus 0, device 15), h/w address 00:00:f8:30:76:00,
      and requires IRQ25 (provided by PCI BIOS).
de4x5.c:V0.544 1999/5/8 davies@maniac.ultranet.com
Partition check:
 sda: sda1 sda2 sda3 sda4 sda5 sda6 sda7
 sdb: sdb1 sdb2 sdb3 sdb4 sdb5 sdb6
 sdc: sdc1 sdc2 sdc3 sdc4
 sdd: sdd1 sdd2 sdd3
 sde:SCSI disk error : host 1 channel 0 id 5 lun 0 return code = 28000000
[valid=0] Info fld=0x0, Current sd08:40: sense key Not Ready
Additional sense indicates Medium not present
scsidisk I/O error: dev 08:40, sector 0
 unable to read partition table
VFS: Mounted root (ext2 filesystem) readonly.
[snip]
------
 
The MO is attached to the 53c94, the behavior on the AHA is the same.
(as I tried last time)
After logging in I inserted a medium and I tried to prepare a fs. It
gave tons of errors on 53c94 which forced me to reboot due to SCSI-bus resetting.
I used pdisk to try to format the media, this is what it complained:
--
53c94: illegal cmd, intr=40 stat=83 seq=83 phase=1
scsi1 channel 0: resetting for second half of retries.
SCSI bus is being reset for host one channel 0.
53c94: illegal cmd, intr=40 stat=83 seq=83 phase=1
SCSI disk error : host 1 channel 0 id 1 lun 0 return code 27070000
scsidisk I/O error: dev 08:14, sector 2
--repeating
--

This repeats and I have to do the reboot (hard way) becomming stable
without the MO.

Has anybody of you same experiences with this type of MO under x86 Linux?
Is it an unsolvable problem due to the fact that it is a SCSI revision
01:CCS device?
Or can I hope?

Thanks for any hints, and sorry for the long post but I felt to include
this msgs to explain.


Andreas

P.S: this is the line when the media is inserted on booting

SCSI device sde: hdwr sector= 1024 bytes. Sectors= 637041 [622 MB] [0.6 GB]


[[ This message was sent via the linuxppc-dev mailing list.  Replies are ]]
[[ not  forced  back  to the list, so be sure to Cc linuxppc-dev if your ]]
[[ reply is of general interest. Please check http://lists.linuxppc.org/ ]]
[[ and http://www.linuxppc.org/ for useful information before posting.   ]]

Re: MO-Disk on PPC with 53C94 SCSI-chip

[Giuliano Pochini]

>I have a LinuxPPC R5 system running on a PowerMac 7200 with the
>2.2.12pre8 kernel. 
[...]
>I can create a fs on the partiton, I can read/write from/to it. BUT,
>when I try to change the side of the media I get the same fs tree as 
on
>the first side.

Some removable medias don't tell Linux when you change the disk. The 
easies way to make it work is to try to mount the device before 
reinserting a new disk (you'll get some orrors of course... ignore 
them), so Linux notices you changed it.

>So I tried booting without media in to verify if it works on the clean
>way. This what the dmesg says:

[...]

Strange. I get a "READ CAPACITY failed, blah blah blah", but then all 
works fine.

[PMac 7300, MO Fujitsu 640MB on ext. SCSI bus]

Bye.

[[ This message was sent via the linuxppc-dev mailing list.  Replies are ]]
[[ not  forced  back  to the list, so be sure to Cc linuxppc-dev if your ]]
[[ reply is of general interest. Please check http://lists.linuxppc.org/ ]]
[[ and http://www.linuxppc.org/ for useful information before posting.   ]]

Re: MO-Disk on PPC with 53C94 SCSI-chip

[Sriranga Veeraraghavan]

Hi,

I had the exact same problem when trying to use by ORB drive with my
7600. The error messages were almost identical. 

I wasn't able to get it to work with under LinuxPPC, I also had
problems with the drive under Linux x86 when connected to a
AHA-2906. It works fine Linux x86 when connected to a AHA-1540.

I think that the problem has to do with the SCSI revision. 

----ranga <ranga@soda.berkeley.edu>

[[ This message was sent via the linuxppc-dev mailing list.  Replies are ]]
[[ not  forced  back  to the list, so be sure to Cc linuxppc-dev if your ]]
[[ reply is of general interest. Please check http://lists.linuxppc.org/ ]]
[[ and http://www.linuxppc.org/ for useful information before posting.   ]]

MO-Disk on PPC with 53C94 SCSI-chip

[Jerry Quinn]

Andreas <toa@pop.agri.ch> writes -->:
Andreas> Hi,
Andreas> 
Andreas> I have a LinuxPPC R5 system running on a PowerMac 7200 with the
Andreas> 2.2.12pre8 kernel. 
Andreas> My problem is my MO-drive from HP see below for type and SCSI-Spec (it's
Andreas> not SCSI-2)

[other descriptions snipped]

Andreas> The MO is attached to the 53c94, the behavior on the AHA is the same.
Andreas> (as I tried last time)
Andreas> After logging in I inserted a medium and I tried to prepare a fs. It
Andreas> gave tons of errors on 53c94 which forced me to reboot due to SCSI-bus resetting.
Andreas> I used pdisk to try to format the media, this is what it complained:
Andreas> --
Andreas> 53c94: illegal cmd, intr=40 stat=83 seq=83 phase=1
Andreas> scsi1 channel 0: resetting for second half of retries.
Andreas> SCSI bus is being reset for host one channel 0.
Andreas> 53c94: illegal cmd, intr=40 stat=83 seq=83 phase=1
Andreas> SCSI disk error : host 1 channel 0 id 1 lun 0 return code 27070000
Andreas> scsidisk I/O error: dev 08:14, sector 2
Andreas> --repeating
Andreas> --
Andreas> 
Andreas> This repeats and I have to do the reboot (hard way) becomming stable
Andreas> without the MO.

I have had a similar problem with a repeating SCSI error that didn't clear
until reboot.  I'm running stock 2.2.10 (not vger) on a PowerCenter - also the
53c94.  In my case, I tried to mount a blank CD on my panasonic cw-7502 cdr
drive.

I suspect we're seeing a slightly more generic problem in changeable media,
not just MO.

-- 
Jerry Quinn                             Tel: (514) 761-8737
jquinn@nortelnetworks.com               Fax: (514) 761-8505
Speech Recognition Research


[[ This message was sent via the linuxppc-dev mailing list.  Replies are ]]
[[ not  forced  back  to the list, so be sure to Cc linuxppc-dev if your ]]
[[ reply is of general interest. Please check http://lists.linuxppc.org/ ]]
[[ and http://www.linuxppc.org/ for useful information before posting.   ]]

Re: MO-Disk on PPC with 53C94 SCSI-chip

[Andreas Tobler]

Jerry Quinn wrote:
> 
> Andreas <toa@pop.agri.ch> writes -->:
> Andreas> Hi,
> Andreas>
> Andreas> I have a LinuxPPC R5 system running on a PowerMac 7200 with the
> Andreas> 2.2.12pre8 kernel.
> Andreas> My problem is my MO-drive from HP see below for type and SCSI-Spec (it's
> Andreas> not SCSI-2)
> 
> [other descriptions snipped]
> 
> Andreas> The MO is attached to the 53c94, the behavior on the AHA is the same.
> Andreas> (as I tried last time)
> Andreas> After logging in I inserted a medium and I tried to prepare a fs. It
> Andreas> gave tons of errors on 53c94 which forced me to reboot due to SCSI-bus resetting.
> Andreas> I used pdisk to try to format the media, this is what it complained:
> Andreas> --
> Andreas> 53c94: illegal cmd, intr=40 stat=83 seq=83 phase=1
> Andreas> scsi1 channel 0: resetting for second half of retries.
> Andreas> SCSI bus is being reset for host one channel 0.
> Andreas> 53c94: illegal cmd, intr=40 stat=83 seq=83 phase=1
> Andreas> SCSI disk error : host 1 channel 0 id 1 lun 0 return code 27070000
> Andreas> scsidisk I/O error: dev 08:14, sector 2
> Andreas> --repeating
> Andreas> --
> Andreas>
> Andreas> This repeats and I have to do the reboot (hard way) becomming stable
> Andreas> without the MO.
> 
> I have had a similar problem with a repeating SCSI error that didn't clear
> until reboot.  I'm running stock 2.2.10 (not vger) on a PowerCenter - also the
> 53c94.  In my case, I tried to mount a blank CD on my panasonic cw-7502 cdr
> drive.
> 
> I suspect we're seeing a slightly more generic problem in changeable media,
> not just MO.

I made some further tests with my drive. At first I added a SMO-F541 to
the chain to verify if it is a scsi problem or something else.
With the sony all went ok on the 53C94 and on the Adaptec-2940UW.
As stated above the resets on the scsi bus occurs only when no cartridge
is inserted and I try to mount one. But this only happens on the 53C94
bus, ont the adaptec it only says bus reseting. And then it does so and
comes back while on the 53C94 I have to hardboot the machine due to
repeating scsi resets.

 
Summary, I see two problems here, first a 'damned' silly HP-optical
(C1716T) which really seems to be to old to run perfectly under Linux
and second, a mac53c94 driver which may have some bugs.

My knowledge seems to be at the end, I don't know how to go further.....
If anyone has an idea or something else I'd like to try it.

Thanks for any hint.


Andreas


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: MO-Disk on PPC with 53C94 SCSI-chip

[Simon Piette]

Andreas Tobler wrote:
> 
> Jerry Quinn wrote:
> >
> > Andreas <toa@pop.agri.ch> writes -->:
> > Andreas> Hi,
> > Andreas>
> > Andreas> I have a LinuxPPC R5 system running on a PowerMac 7200 with the
> > Andreas> 2.2.12pre8 kernel.
> > Andreas> My problem is my MO-drive from HP see below for type and SCSI-Spec (it's
> > Andreas> not SCSI-2)
> >
> > [other descriptions snipped]
> >
> > Andreas> The MO is attached to the 53c94, the behavior on the AHA is the same.
> > Andreas> (as I tried last time)
> > Andreas> After logging in I inserted a medium and I tried to prepare a fs. It
> > Andreas> gave tons of errors on 53c94 which forced me to reboot due to SCSI-bus resetting.
> > Andreas> I used pdisk to try to format the media, this is what it complained:
> > Andreas> --
> > Andreas> 53c94: illegal cmd, intr=40 stat=83 seq=83 phase=1
> > Andreas> scsi1 channel 0: resetting for second half of retries.
> > Andreas> SCSI bus is being reset for host one channel 0.
> > Andreas> 53c94: illegal cmd, intr=40 stat=83 seq=83 phase=1
> > Andreas> SCSI disk error : host 1 channel 0 id 1 lun 0 return code 27070000
> > Andreas> scsidisk I/O error: dev 08:14, sector 2
> > Andreas> --repeating
> > Andreas> --
> > Andreas>
> > Andreas> This repeats and I have to do the reboot (hard way) becomming stable
> > Andreas> without the MO.
> >
> > I have had a similar problem with a repeating SCSI error that didn't clear
> > until reboot.  I'm running stock 2.2.10 (not vger) on a PowerCenter - also the
> > 53c94.  In my case, I tried to mount a blank CD on my panasonic cw-7502 cdr
> > drive.
> >
> > I suspect we're seeing a slightly more generic problem in changeable media,
> > not just MO.
> 
> I made some further tests with my drive. At first I added a SMO-F541 to
> the chain to verify if it is a scsi problem or something else.
> With the sony all went ok on the 53C94 and on the Adaptec-2940UW.
> As stated above the resets on the scsi bus occurs only when no cartridge
> is inserted and I try to mount one. But this only happens on the 53C94
> bus, ont the adaptec it only says bus reseting. And then it does so and
> comes back while on the 53C94 I have to hardboot the machine due to
> repeating scsi resets.
> 
> 
> Summary, I see two problems here, first a 'damned' silly HP-optical
> (C1716T) which really seems to be to old to run perfectly under Linux
> and second, a mac53c94 driver which may have some bugs.
> 
I would tend to say that the 53c94 driver is not perfect, because it's still
impossible to use SANE (scanning software) with it. Anybody succesfully using
the 53c94 to scan, please manifest yourself.

Simon Piette

** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: MO-Disk on PPC with 53C94 SCSI-chip

[Paul Mackerras]

> and second, a mac53c94 driver which may have some bugs.

I wrote the mac53c94 driver a long time ago and I haven't spent
anything like as much time on it as on the mesh driver (which isn't
perfect either).

The opportunity is sitting there, waiting for someone to pick it up,
to stress-test, debug and optimize the mac53c94 driver, including
doing things like implementing disconnect/reconnect.  In the Open
Source tradition, this should be someone for whom this is an itch that
they want to scratch. :-)

Paul.

** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: MO-Disk on PPC with 53C94 SCSI-chip

[Jerry Quinn]

>> "Paul" == Paul Mackerras <paulus@cs.anu.edu.au> writes:

 >> and second, a mac53c94 driver which may have some bugs.

 Paul> I wrote the mac53c94 driver a long time ago and I haven't spent
 Paul> anything like as much time on it as on the mesh driver (which isn't
 Paul> perfect either).

 Paul> The opportunity is sitting there, waiting for someone to pick it up, to
 Paul> stress-test, debug and optimize the mac53c94 driver, including doing
 Paul> things like implementing disconnect/reconnect.  In the Open Source
 Paul> tradition, this should be someone for whom this is an itch that they
 Paul> want to scratch. :-)

Can you tell us what the difference is between this driver and the generic
53c94 driver in the same directory?  Also, is there a source of docs other
than the code there (for us SCSI newbies :-)?

Thanks,
Jerry

-- 
Jerry Quinn                             Tel: (514) 761-8737
jquinn@nortelnetworks.com               Fax: (514) 761-8505
Speech Recognition Research


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: MO-Disk on PPC with 53C94 SCSI-chip

[Paul Mackerras]

> Can you tell us what the difference is between this driver and the generic
> 53c94 driver in the same directory?  Also, is there a source of docs other
> than the code there (for us SCSI newbies :-)?

I assume you're talking about the mca_53c9x.c driver?  I don't think
that existed when I started doing mac53c94.c.  I would expect that the
main differences are in the areas of DMA and interrupt handling.  The
mca_53c9x driver is probably worth a look.  I used the esp.c driver as
a reference as the ESP chip in Suns is apparently a 53c96 and has
marked similarities to the 53c94.  There appear to be no fewer than 4
esp drivers there. :-)

I never found a definitive manual or data sheet on the 53c94,
unfortunately.  For documentation on the SCSI-2 standard, I have a
late draft that I got by ftp from somewhere.  The actual final
standard isn't available on the web as it costs real money. :-(

Paul.

** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: MO-Disk on PPC with 53C94 SCSI-chip

[Joshua M. Thompson]

On Wed, 8 Sep 1999, Paul Mackerras wrote:

> mca_53c9x driver is probably worth a look.  I used the esp.c driver as
> a reference as the ESP chip in Suns is apparently a 53c96 and has
> marked similarities to the 53c94.  There appear to be no fewer than 4
> esp drivers there. :-)

You should be able to do something similar to what mac_esp.c (the driver
for the ESP chip on the Quadra 660AV and 840AV) does...that is, just
enough work to find the chip and get the interrupts and then pass the buck
to the NCR53C9x.c driver.

-- 
funaho@jurai.org             | Voodoo is a very interesting religion for the
http://www.jurai.org/~funaho | whole family, even those members of it who are
                             | dead.
                             |  -- (Terry Pratchett & Neil Gaiman, Good Omens)


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/