Geoff Davis wrote:
>
> Well, active termination didn't do anything for me. I don't think it's a
> termination issue, since the BIOS does see it at startup, no matter
> whether I have the card, the drive, or the active terminators doing the
> termination.
>
> I went and dug through our pile of SCSI crap at work for a terminator
> for an internal 50 pin ribbon. That was a real pain. I also dug up a
> terminal for the HD50 connector on the outside.
>
> I disconnected everything else from the scsi chain, slapped the internal
> terminator on the internal cable, disconnected my zip drive and put the
> external terminator on the hd50, powered it up, BIOS detected the drive,
> and linux hung.
>
> I then took the card out of automatic termination mode, and tried again,
> with the same results.
>
> I then remembered that the termination jumper was still on the drive, so
> I took that off. Still nothing.
>
> I also tried a new cable, just for the hell of it. Nothing.
>
> A further observation: on the BUS reset, the HPs light flashes and the
> mechanics move, so the bus reset is affecting the drive in some manner.
> I guess that just shows that it's alive.
>
> Any other ideas about the hanging?
>
> As for the earlier data corruption problem, I have no clue what caused
> it. I'm ready to attribute it to gremlins and leave it, as long as it
> doesn't happen again. Right now I'm playing mp3s off the drive without
> any problems other than it sounding like a friken helicopter, but that's
> a function of my 40X toshiba's crappy mechanics.
>
> /Geoff
>
> bvermeul@devel.blackstar.nl wrote:
> >
> > Hi Geoff,
> >
> > >Well, after several weeks of searching newsgroups and mailing list
> > >archives, hacking, and slashing, I still cannot get my shiny new Iwill
> > >SIDE 2935UW controller to work properly with linux. My problem is that
> > >when I have my HP 6020i cd burner attached to the chain, the system
> > >freezes upon device detection.
> > >
> > >The same drive works with my crusty old Adaptec 152x card, with the same
> > >other devices on the chain, and in the same order. On the Iwill/Initio
> > >card, everything works fine under Windows 95 OSR2 (can access files and
> > >write a cd), but linux will not work with the drive attached.
> >
> > Could you try adding active termination at the end of the scsi chain?
> > I've just helped someone with about the same problems as you, and they
> > got solved by adding active termination to the chain. The Linux driver
> > has a whee bit of a problem with faulty termination.
> >
> > >When I have all of the devices attached, the screen output of modprobe
> > >initio looks like this:
> > >i91u: PCI Base=0xA400, IRQ=11, BIOS=0xFF000, SCSI ID=7
> > >i91u: Reset SCSI Bus ...
> > >
> > >(picks it's butt for a few seconds, then... )
> >
> > That's normal. It'll pause for about 5 seconds on a SCSI bus reset.
> >
> > >scsi0 : Initio INI-9X00U/UW SCSI device driver; Revision: 1.03g
> > >scsi : 1 host.
> > > Vendor: TOSHIBA Model: CD-ROM XM-6401TA Rev: 1009
> > > Type: CD-ROM ANSI SCSI revision: 02
> > >Detected scsi CD-ROM sr0 at scsi0, channel 0, id 2, lun 0
> > > Vendor: IOMEGA Model: ZIP 100 PLUS Rev: J.66
> > > Type: Direct-Access ANSI SCSI revision: 02
> > >Detected scsi removable disk sda at scsi0, channel 0, id 5, lun 0
> > >
> > >...and then the system locks. No numlock lights, no magic sysrq. I have
> > >to powercycle and wait for the fsck. Note that the Burner is set as ID
> > >4, but doesn't show up in the list.
I went through the routine again, just for ducks, and realized that the
output given above is erroneous. It only shows the Toshiba drive on SCSI
ID 2, and hangs when it probes the HP 6020i at ID 4. The Zip Drive at
ID5 is not displayed.
The test was done immediately after booting to windows and playing some
mp3s off of the same 6020i drive that causes the hang in linux.
Is there a set of debug options I can turn on in the driver to figure
out where it is hanging at, or does anyone know anything about the
Chipset on the card I have? It's an IWILL 2935UW, which I believe uses
the same chipset as the Initio 950, but this is just a guess from
looking at the chip on the card. Is the ini91xx driver the correct
module to load?
> >
> > Does the BIOS find it during bootup?
As mentioned in a previous post to this list, yes it is found at boot by
the bios, and win95 works just fine with it.
> >
> > >However, when the burner is disconnected from the chain, everything else
> > >is detected and the system runs fine, although I did have a wierd
> > >problem with the Toshiba CD-ROM garbling data that was solved by a
> > >reboot.
> >
> > I've never seen that here before. Were you able to reproduce that
> > particular problem?
> >
> > >I have attempted to isolate the problem by doing the following:
> > > played with the termination on the drive and on the card itself,
> > > order of devices on the chain,
> > > changed scsi IDs,
> > > Async transfer mode,
> > > Removed all other devices on the chain,
> > > removed all other cards in the system other than the video card,
> > > played "Musical PCI slots" to see if another pci slot will make the
> > >sucker work,
> > > recompiled the kernel with different PCI options (buggy bridge or
> > >whatever),
> > > compiled the drivers into the kernel instead of using them as a module,
> > > tried a Uni-processor kernel
> > > played with the probe all LUNs option,
> > > tried disabling vendor-specific extensions.
> > >
> > >However, none of these efforts got me anywhere. The fact that everything
> > >works fine under (gasp!!) windows leads me to believe that it is
> > >something with the kernel drivers.
> >
> > They're definately pecky about termination. I'm not sure if that's your
> > problem, but it may just be...
> >
> > <snip machine configuration>
> >
> > >The Scsi chain has the following attached, in the order listed:
> > >internal 50 pin:
> > > Toshiba CD-ROM XM-6401TA on ID #2
> > > HP SureStore 6020i on ID #4
> > >external 50 pin:
> > > Iomega Zip 100 Plus on ID #5
> >
> > How do you terminate the chain? The CDROM only has passive termination,
> > and that may not be enough. The external chain needs termination as
> > well.
> >
> > >Please let me know if you need any more information. I have basic C/C++
> > >skills, but know nothing about device drivers.
> >
> > Could you try using active termination, and if it works, (or doesn't work)
> > let me know? I've been using the driver for some time, with two drives and
> > a CD, without problems myself.
> >
> > Regards,
> >
> > Bas Vermeulen (Who maintains the initio drivers (sorta anyways))
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/