Re: linux-2.4.12 / linux-2.4.13 parallel port problem

>>>>> "TW" == Tim Waugh <[e-mail]> writes:

>> Question: is this intended behaviour? I would think that you would
>> normally want to just say irq=auto and let the driver find the io
>> address just as it does normally.

TW> It is intended behaviour.  'irq=auto' in this case didn't help because
TW> the ECP chipset would not tell us what IRQ it was assigned (it just
TW> said "it's set by jumpers, or alternatively I'm not telling you".

This part I do not quite understand.  I have an old laptop that
was working with parport=auto up to 2.4.10 and then stopped
working, just like the original poster's problem description.

detected both address and irq for parport0, while 2.4.12,
according to the your response, could not tell that IRQ=7.  Do
you mean that the logic which made 2.4.10 to claime to have
detected IRQ=7 was faulty and the logic in 2.4.12 is being
careful not to misdetect?

    Message-ID: <[e-mail]>
    Date: 	Wed, 24 Oct 2001 14:16:51 +0100
    From: Dave Garry <[e-mail]>
    Subject: linux-2.4.12 / linux-2.4.13 parallel port problem

    With kernel 2.4.12 and 2.4.13 the parallel port on
    my machine looks like this according to dmesg:

    parport0: PC-style at 0x378 [PCSPP,TRISTATE]
    parport0: cpp_daisy: aa5500ff(98)
    parport0: assign_addrs: aa5500ff(98)
    parport0: faking semi-colon
    parport0: Printer, Hewlett-Packard HP LaserJet 1100

    Under 2.4.10 is looks like this:

    ...
    parport0: PC-style at 0x378 (0x778) [PCSPP,TRISTATE,COMPAT,ECP]
    parport0: irq 7 detected
    ...