> Well it is a bug worth fixing. Who knows it may have something
> to do with the disable_irq problems the forcedeth driver was seeing.
I agree completely, the fix looks obvious and there's no point in
leaving the bug there. I just don't have a test case that I can point
to as being fixed by this, since I don't have a working 32-bit MSI
device...
Do you know whether the forcedeth devices have a 32-bit MSI address?
- R.
-