Greg Ward (gward@ase.com)
Fri, 15 Jan 1999 23:27:36 -0500
Hi all --
I'm having a problem with the ppa driver in 2.2.0-pre7 (same thing
happened with pre6; didn't disappear when I patched up, so I'm asking
here). I have a plain vanilla two-year-old parallel port ZIP drive that
worked just fine (albeit reaaaaally slowly) with ppa under the 2.0
kernels.
Here's the situation: 2.2.0-pre7 kernel, patched up from 2.2.0-pre6.
Everything relevant to the ZIP drive (parport, parport_pc, scsi_mod,
ppa) compiled as modules and sitting in /lib/modules/2.2.0-pre7 as
expected. I enabled the "parport_probe" stuff the second time I built
the kernel, but it doesn't seem to help.
A bit wary of modprobe, I do things the old-fashioned way and load
modules one at at time:
# insmod scsi_mod
# insmod parport
# lsmod
Module Size Used by
parport 6596 0 (unused)
scsi_mod 46148 0 (unused)
ppp 17692 2 (autoclean)
slhc 4196 1 (autoclean) [ppp]
(ppp and slhc are loaded because I'm online typing this email!)
But then I try to load parport_pc, and it fails:
# insmod parport_pc
/lib/modules/2.2.0-pre7/misc/parport_pc.o: init_module: Device or resource busy
*Which* device or resource is busy? Is there any way to find out?
A couple of possible things occur to me, but I don't really know how to
try them out. First, is parport or parport_pc trying to initialize a
bogus parallel port? I'll have to reboot my machine and check where the
parallel port is (port and IRQ), but why have the defaults change since
the 2.0 kernels? If this is the case, I'm guessing I'll have to supply
some options to the driver when I load it: what are the options? are
they documented anywhere? (not in drivers/scsi/README.ppa, not on the
web pages referenced by them, and there's no Documentation/ppa.txt as I
was hoping).
And now I will reveal the stunning depth of my ignorance of kernel
matters: could /dev/lp* have anything to do with it? I vaguely realize
that something changed with the device files for parallel devices
somewhere in 2.1.x ... is this a concern at the module-loading point, or
does that only matter later, when applications are trying to access the
drivers through /dev/lp*? (For the record, I have *not* changed my
/dev/lp* files; they're major 6, minor 0, 1, and 2 for lp[012].)
Thanks for any help you can supply...
Greg
-- Greg Ward, programmer-at-large email: gward@ase.com-- To unsubscribe, send mail to: linux-parport-request@torque.net -- -- with the single word "unsubscribe" in the body of the message. --
This archive was generated by hypermail 2.0b3 on Fri 15 Jan 1999 - 23:30:04 EST