> I'm sure, you got rid of the Ugh message, but I'm not sure, whether
> you got rid of the problem causing these messages. It is possible,
> that the bug is still present, you just don't see any messages about
> it.
>
> I think it would be a good idea to apply the patches I have sent to
> the list and try to find the real problem for the Ugh messages. You
> seem to have a reliable way to trigger them, so why don't you try to
> find the reason for them, so that someone can fix the kernel?
The cause(s) for the Ugh messages are usually quite easy to find: it's
usually a copy_xx_user() call with interrupts turned off. So if you grep
for occurrences of copy_*_user and look for a nearby cli(), you'll
probably find an Ugh.
At one point I had a list of about a dozen places where this was being
done, mostly in ISDN drivers derived from an earlier serial driver.
Fixing the problem might be a little tricky, as you'd need to see what
the driver was trying to do, and then rearrange it so that no pag faults
are possible with interrupts turned off.
Regards,
Bill
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu