Re: mm 2.2.17pre6

From: Andrea Arcangeli (andrea@suse.de)
Date: Sun Jun 25 2000 - 20:29:02 EST


On Mon, 26 Jun 2000, Alan Cox wrote:

>There are big database test sets that fail in all 2.2 kernels tried. So I
>dont think we can ignore the problem

Is it possible for me to reproduce here? (is there a testcase or a
downloadable thing?, URL?) I never got this reports thus I wasn't aware of
any problem. (I may have missed spare reports on l-k on May due holydays)

I definitely see the race in the allocator but we can fix it right without
hurting performance significantly. And unless there's swap going on it's
unlikely to trigger due the big kernel locking all over the place.

BTW, I'd be curious if people could test such thing also in 2.2.17pre6aa2
and tell me if they can break it too.

Thanks for the info.

Andrea

-
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 : Mon Jun 26 2000 - 21:00:07 EST