[SGVLUG] New kernel 2.6.21 supports my video card

[Miguel Hernandez]

It sounds like you've done the best you can do, David. This (older hardware
support) is what linux is really known for, amongst the the more tech-savvy
folks but there's got to be items/scenarios that "fall through the cracks."
I don't think that the kernel devs are apathetic to similar situations but,
more likely, you can get hardware that's a couple years old for pretty cheap
nowadays so their time is probably best served elsewhere (that's what I'd
imagine, anyway).

All I know is this discussion totally reminded me of the good old days of
autoexec.bat & config.sys on MS-DOS....ahh, /reminisce.

--miguel

On 8/3/07, David Lawyer <dave at lafn.org> wrote:
>
> On Fri, Aug 03, 2007 at 01:18:26PM -0700, John  Jefferson Lowry IV wrote:
> > Isnzt there supposed to an auto-balancer for IRQs?
>
> Yes but that's for the PCI bus where IRQ's may be shared.  My PC has
> both a PCI bus and an old ISA bus where interrupts may not be shared
> nor may the same interrupt be used on the PCI and ISA busses.  On the
> ISA bus one may have cards where the IRQs are set by jumpers.  I set
> them to avoid conflicts but the Linux driver for my sound card changes
> the IRQ from what I set, resulting in a conflict.  This is a card
> where proprietary software (and not standard ISAPNP software) can
> override IRQs set by jumpers.  Perhaps the reason why the Linux driver
> can't deal with the IRQs is that it doesn't know this code used to set
> the IRQs in the hardware, but it's doing something to change the IRQ
> from what I've set by jumpers to what is called the default on the
> Internet: IRQ=5.  Anyway, the Debian maintainer is supposedly looking
> into it.  If the software can't be fixed, at least it should be
> documented regarding the IRQ behavior.
>
>                         David Lawyer
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.sgvlug.net/pipermail/sgvlug/attachments/20070804/b682849d/attachment.html