As a datapoint, I've had 2.4.0-test9pre6 up over 1.5 days doing
some serious compiling for a piece of that (probably a good 6-8 hours,
but I haven't timed it with my new AMD processor lately).
I'm running X 4.0.1. I don't know where the `c' and `d' bits you
guys are talking about is coming from. This is the source and patches
I grabbed off of XFree86's website.
I haven't seen any fatal lockups. In the past I've apparently been
lucky since I'm running a 2.4.0.x kernel with 256MB of RAM and SMP on
a non-SMP machine, but there is a reason I do that. (:
I was running 2.4.0-test9pre2 (no X at all) for 4 DAYS before that
with no known problems.
Unfortunately, I'm running syslogd-ng-1.4.5 with seems to stop syslogging
output after some period of time, so I can't tell if I'm catching all
the warning messages that I might, even if they're not deadlocking me.
On Sun, Sep 24, 2000 at 01:21:40AM -0400, Mohammad A. Haque wrote:
> I think there's problem in both the kernel and in X. Why?
>
> Work machine:
> * test9-pre2 w/ X 4.0d ... uptime 5 days nothing reporting
> shm errors like my other boxes. This machine gets hit pretty
> hard when I'm on it.
>
> Co-worker's machine:
> * 2.2.16 w/ X 4.0.1c ... getting reports of shm errors but no
> lockups. This guy tortures his machine.
>
> Home machine:
> * test7 w/ X 4.0.1c ... getting reports of shm errors but no
> lockups
>
> * test9-pre* w/ X 4.0.1c ... can't stay up for more than a
> couple hours before locking up.
>
> To do:
> * Try test 8
> * Try X from about a week or two ago (or futher back if needed)
>
>
> Not very scientific I know. But so far the thing I see in common is that
> using kernels before test9-pre* with or without X 4.0.1c and your
> machine wont lock up in a relatively short period of time. use
> test9-pre* w/ X 4.0.1c and you're not gonna be up for more than a couple
> hours. i may be lumping all the test9-pre versions in here when I
> shouldnt be.
>
>
> Like someone stated earlier, I think this is like the truncate bug.
> Something has been changed that just brought this bug out to the front.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Sat Sep 30 2000 - 21:00:12 EST