Re: 48 day uptime problem?

Jakub Jelinek (jj@sunsite.ms.mff.cuni.cz)
Thu, 29 Apr 1999 16:54:58 +0200 (CEST)


> Conveniently, my alpha hit 48 days uptime a day after you guys found this.
> Unfortunately, it seems the problem is more complicated than just the TCP
> code. I suspect that in many places in the kernel, time (in ms) is taken as
> a 32-bit integer. I experienced the following symptoms (kernel 2.2.2-ac7)
> when I hit 48 days:
>
> 1) No network connections succeeded.
> 2) processes that I "visited" (i.e. went to xterm, typed something...)
> spammed the CPU. Load hit 10 very quickly, and it took about 10 minutes for
> 'shutdown -r now' to succeed.
>
> I'll install 2.2.7 with your patch today, but in 48 days I'll be out of the
> country and away from my computer. I hope someone else can investigate this
> further sometime around the middle of June. ;)
>
> Maybe some greps of the source tree for time calls?

Or you can do it yourself. Just initialize jiffies to some other value, say
if you set it (in kernel/sched.c) to
unsigned long volatile jiffies=0xfff00000;

you should see things going in 17 minutes.

Cheers,
Jakub
___________________________________________________________________
Jakub Jelinek | jj@sunsite.mff.cuni.cz | http://sunsite.mff.cuni.cz
Administrator of SunSITE Czech Republic, MFF, Charles University
___________________________________________________________________
UltraLinux | http://ultra.linux.cz/ | http://ultra.penguin.cz/
Linux version 2.2.5 on a sparc64 machine (3958.37 BogoMips)
___________________________________________________________________

-
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/