Mailinglist Archive: opensuse-bugs (13459 mails)

[bugzilla_noreply@xxxxxxxxxx]

https://bugzilla.novell.com/show_bug.cgi?id=335676#c20





--- Comment #20 from Arjen de Korte <suse+build@xxxxxxxxxxxx>  2007-10-26 
05:30:44 MST ---
(In reply to comment #19 from Stanislav Brabec)

> I have not seen any crash of nut driver (even before patch), both standalone 
> or
> nut. The crashing side is the kernel and memory corruption occurs inside
> kernel. Nothing from user space (even incorrectly written) should cause kernel
> OOPS.

I agree.

> If this crash is limited to this UPS model and x86_64, I would prefer to
> release fix for bug 331749 now. If anybody else is able to reproduce this
> crash, I'll postpone the (apparently correct) fix, which actually causes 
> kernel
> to crash.

For me the fixes (both to udev and hidparser.c) obviously work. But I don't
have this UPS model and neither do I run x86_64. Does this also happen with the
default (i586) kernel?

> Note that this UPS also communicates somehow with GNOME Power Manager and it
> does not cause OOPS.

You don't run the standard and HAL drivers at the same time, do you? That
doesn't fly, since only one driver can claim an USB device at the same time.
NUT 2.2.0 is still broken in this respect, after successfully claiming the
device, it won't notice that another driver may have claimed it afterwards
(fixed in nut-2.2.1). This leads to the kernel complaining about the driver
using the interface without claiming it.


-- 
Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.