Re: 2.6.9 page allocation failure. order:0, mode:0x20

[Michael Clark]

On 10/28/04 16:29, Denis Vlasenko wrote:
> On Thursday 28 October 2004 08:31, Michael Clark wrote:
>
> > BTW - 2.6 is much more responsive than 2.4 while this is all
> > going on - i'm just worried about these messages.
>
>
> Which one was faster, and by how much?

Both tests compiling 2.6.9 tree with make -j192 bzImage modules
(.config posted earlier) from clean source after a reboot.
2CPUs, 2GB RAM, 2GB swap

2.4.27
real    15m38.504s
user    21m5.720s
sys     3m28.990s
peaked at about 1.7GB swap usage

2.6.9
real    14m50.360s
user    21m9.008s
sys     2m40.580s
peaked at 2.0GB swap usage - top said 0K swap free and it survived ;)

2.6.9 was 5% faster (although subjectively almost a magnitude more
responsive ie. sshing into the box in the middle of this took
about a minute with 2.4.27 and only about 10 or so seconds with 2.6.9,
although i didn't time this).

Seems 2.6's more proactive swapping helps a bit ie. swap more of
the right stuff so as to swap less overall as 2.6 went about 20%
deeper into swap.

~mc
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/