Hello Gerard,
> [ ... ]
>
> > ncr53c8xx_abort: pid=14 serial_number=29 serial_number_at timeout=29
> >
> > And then it just does nothing at least in the 20 minutes it i needed to
> > type this over from the other computer..
> >
> > Is it posable that on the ncr53c810a-2 irq 13 not the best irq because
> > on the other kernels it runs up 9,10,11 with the 2.1.130 kernelstrange
>
> The driver reads the IRQ number from the PCI device configuration space on
> linux-2.0 kernels, but is provided by the PCI driver code with this
> information in linux-2.2 kernels. In both case, the IRQ number is assumed
> to be the one that corresponds to the actual IRQ routing of the device on
> your system. If for some obscure reason, it is not so, then the driver
> interrupt routine will never be called and no IO will complete.
>
> > After looking around in the readme from the scsi driver that it is not
> > posable to set irq's enz..
>
> That's the case for all PCI device drivers. The PCI config space is
> filled in by the POST code and shall be correct. For non-intel, Linux
> is performing some write to device PCI config space and they also
> shall corresponds to the reality of the actual hardware settings.
> Failing to do so, may just let your system not work at all.
>
Wel i think i am going to open up this machine to and reorganize the scsi
cards so that all the information is directly one after the other because
i had a look behind that machine the current cards (pci) is scsi,vga,scsi
and i wil try to make it scsi,scsi,vga but ik wil be in the weekend when
i wil have the time to test this out.
> > > What exactlyis happening on the second machine? I didn't see any errois
> > > in the log you included...
> > >
> > Wel see above but on the extra 2 scsi controlers there is no hardware.
> >
> >
> > Gr. Ronald Meijerink
>
> Gérard.
Gr. Ronald Meijerink
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/