On Fri, 11 Feb 2000, Ingo Molnar wrote:
>
> the lowlatency patch has the potential to trigger various races within the
> kernel much more agressively. For the 'stuck process' case you first have
> to check out at which EIP the process is stuck:
>
> hell:~> ps -elnf | more
> F S UID PID PPID C PRI NI ADDR SZ WCHAN TTY TIME CMD
> 100 S 0 1 0 0 60 0 - 274 139331 ? 00:00:02 init
>
> take the 'WCHAN' field and add '0xc0' in front of it: 0xc0139331. Then
> look this EIP up in your System.map. Hopefully it will be some meaningful
> place, not a generic one like sleep_on(). If this does not help then you
> might need the IKD patch-collection's stuck-process-oopser feature.
>
I am attempting to reproduce and debug according to your suggestions.
Also, in the failed hunks when applying your patch to 2.2.14aa7, there
is a new function "mark_buffer_protected" in include/linux/fs.h. I think
it should also have a conditional_schedule but am not sure whether it
needs mark_buffer_highprio or not?
Wm
-
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/
This archive was generated by hypermail 2b29 : Tue Feb 15 2000 - 21:00:21 EST