> I seem to remember something relating to early in the 2.0 series
> (around 2.0.11 IIRR) where a bug was found to exist relating to swap
> areas with equal priorities, and I don't think it was ever tracked
> down, but the advice given then was NOT to use swap areas with equal
> priority...
Hi Riley,
yep, looks like the problem's still there in 2.0.31.
Seeing as Netscape is where the problem is most evident, is a strace on
Netscape the best approach to solving the problem? Should be simple to
write a short program to keep just the last meg of strace output, run
Netscape, wait for the box to commence its grinding behaviour, then examine
the strace dump?
Of course, dumping the strace may interact with swapping in
wierd ways, and conceal the problem.
Any suggestions?
Cheers
Paul
---------------------------------------------------------------------------
,---
Paul Wilkins | o o |
Internet Operations Manager ` \_/ ' Linux
TMP Worldwide /\___/\ The Choice of
paulw@tmpw.com.au |_/ . \_| A GNU Generation
http://www.monsterboard.com.au \_/___\_/
---------------------------------------------------------------------------
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu