grant@torque.net
Sat, 22 May 1999 09:55:23 -0400 (EDT)
> FYI, I'm using a printer daisy-chained to a backpack CDROM drive
> and pass-through sort of works in 2.2.5-ac6. Sort of because I
> can access the CDROM or the printer, but not simultaneously. If
> I access the CDROM while printing, for example, the printer
> immediately stops.
>
> I'm just relating this as anecdotal evidence that the
> pass-through code is not entirely broken. Heck, maybe even
> simultaneous access works and I'm only doing something wrong
> (e.g. the lp driver uses irq 7 while the paride/bpck/pcd modules
> appear to be polling in 4bit mode...)
NB: none of the PARIDE modules use interrupts, so the IRQ is a non-issue
for pcd.
The 4-bit mode is more serious. Do try to change your CMOS settings
so the port is in EPP mode. The lower the mode number, the longer it
takes to move data from the CD-ROM and the longer the printing pauses.
But, there may also be a problem with the sharing layer in recent
kernels - I think Tim is looking into that.
The trouble is that the reports are all ambiguous - including yours.
Are you saying that printing stops _forever_ or that printing stops
until the CD-ROM is finished ? Pauses while the CD is busy are to be
expected. But, if the printer dies altogether, this is more serious.
Does this happen if you revert to polling with the lp driver ?
While the interrupt is irrelevant for PARIDE, there may be a bug in
lp where it loses control if it is preempted when it is expecting an
interrupt.
--------------------------------------------------------------------------
Grant R. Guenther grant@torque.net
--------------------------------------------------------------------------
-- 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 Sat 22 May 1999 - 09:56:57 EDT