On Tue, 5 Sep 2000, Alan Cox wrote:
> Then why are we still finding obscure lock ordering bugs in 2.2 ?
> Finding them when they bite you is easy, finding the obscure ones when
> they dont generally bite is much harder - thats where debugging
> kernels that trap and dump lock lists for order violations are a big
> win
i stand corrected - lock ordering indeed is a problem, although i think
2.4 is much cleaner in this regard, we have less lock hierarchies, mainly
due to getting rid of much of the BKL.
i think another, natural debugging help here will be to make the UP kernel
preemptibe in 2.5. That will simulate most of the SMP locking scenarios on
UP as well, and we have much more UP users than SMP users.
Ingo
-
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 : Thu Sep 07 2000 - 21:00:23 EST