FC3 Irq Conflicts on Laptop

Reshat Sabiq sabiq at csociety.org
Wed Apr 13 03:42:07 UTC 2005


James Mckenzie wrote:

>-----Original Message-----
>From: David Cary Hart <Fedora at TQMcube.com>
>Sent: Jan 30, 2005 9:45 PM
>To: For users of Fedora Core releases <fedora-list at redhat.com>
>Subject: Re: FC3 Irq Conflicts on Laptop
>
>On Sun, 2005-01-30 at 21:35 -0700, James McKenzie wrote:
>
>  
>
>>>You might have to enable the second PCI IRQ in your systems BIOS before 
>>>trying to divert it through software.  I'm wondering why you are getting 
>>>the error, as IRQ 11 should be your PCI bus interrupt and you should not 
>>>have an 8250 UART in your system.  I have as much as you do in my 
>>>system, and I do not get this error.  However, PCI IRQ movement is more 
>>>art than science.
>>>
>>>      
>>>
>>What happens if he recompiles with APIC and APIC IO (assuming his
>>machine supports this)? Are their other non-standard kernel options that
>><might help?
>>    
>>
>
>David:
>
>I don't know if this would help.  What I was commenting on is assignment of certain devices to a particular IRQ.  That could help, then again it may not solve the problem of an overloaded IRQ.  PCI is an extension of the old Extended ISA bus, which allowed level based IRQ handling which was supposed to prevent IRQ 'collisions'.  I'm guessing that is what is happening in this case.  Moving the device to another IRQ should eliminate collisions but may not eliminate the overflow problem that the device experiences.
>
>  
>
It would also be nice to know the answer to a more generic question:
how can an IRQ of a (any) device be changed, w/o changing the driver? I 
suppose most drivers are responsible for making the best pick, but it 
looks like sometimes they don't. Perhaps some of them are configurable? 
Or is there anything else like the pcmcia.opts mentioned to suggest/tell 
the best IRQ for a device?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listman.redhat.com/archives/fedora-list/attachments/20050412/e010814b/attachment-0001.htm>


More information about the fedora-list mailing list