Messages in this thread | | | Date | Thu, 16 Mar 2000 11:24:36 -0800 (PST) | From | Linus Torvalds <> | Subject | Re: Cardbus and kb/mouse lockup |
| |
On Thu, 16 Mar 2000, Wakko Warner wrote: > > I happen to think about it this morning. Someone told me (private email I > think) that his ps2 mouse locked if something was using irq 12. Mine > doesn't lock if the cdrom is missing. So what I thought thismorning is the > fact that the cardbus driver is walking over the first available IRQ (in my > case, cdrom installed, irq15 is used, 14 is always used, 13 reserved and 12 > doesn't show up in the pci irq list). But with the cdrom missing, it walks > over irq 15. > > Does this make any sense?
Yes. The change that fixed the mouse for others was to not probe for irq12.
I suspect that we could even probe for irq12 as long as we after the probe make sure that we don't leave it parked at the last irq we probed. It _shouldn't_ matter, but it obviously does.
You can test with one of the broken kernels whether the following one-liner makes a difference for you: add the line
exca_writeb(socket, I365_CSCINT, 0);
to the end of the yenta_probe_irq() function.
Linus
- 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/
| |