Re: Adventures with bonnie and linux-2.3.7

Linus Torvalds (torvalds@transmeta.com)
22 Jun 1999 09:23:34 GMT


In article <99Jun22.081906gmt+0100.20610@fw.mol.hu>,
BOSZORMENYI Zoltan <zboszor@mol.hu> wrote:
>
>The other operations give almost the same throughput results but 2.3.7
>does it with less CPU time, in one case with 50% less! I did these
>tests some times, and the results were just like the above with some
>(~1 - 2%) difference

I would hope that most of them were already maxing out the disk with the
2.2.x setup, so 2.3.x can't improve much on that.

There are certainly cases where we do much worse, simply because some of
the 2.3.x code changes behaviour a _lot_, and no real tuning has taken
place, so I'm not surprised if some of the initial numbers aren't as
good as they should be.

You could try pre-2.3.8-2, which has at least some problems in plain
2.3.7 fixed.

>The above was the end of the story. :) I had strange experiences with
>2.3.7. After compilation, I rebooted with it and it gave oops when it
>attempted "swapon", and fsck found uncorrectable errors.

These are hopefully fixed.

Linus

-
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/