Doing a mke2fs on a 12G device, buffer cache eats up my main memory without
recognizing buffer_mem.max_percent as it did before.
The system (2*PII/350, 256M) is absolutly not interactive for at least a minute
friendly swaping at that time.
A look into the code fs/buffer.c showed refill_freelist() to be very
slim now ;*)
If i put in an artivicial limit for the buffercache, i run into
a MM deadlock, because the freeing code seems to be gone with
refill_freelist()'s emaciation and waking up bdflush is not very successfull
in this case.
BTW: a while ago in the 2.1.* time, buffer_mem.max_percent wasn't
recognized either.
But at that time refill_freelist() code was smart enough to refill
with my artificial limit hack giving a much better interactive feeling.
Please bring us back the limited buffer cache with refilling of the feelist! 8*)
Any comments by the MM gurus?
Regards,
Heinz
--=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Systemmanagement Entwicklungsbereich 2 Deutsche Telekom AG Entwicklungszentrum Darmstadt Heinz Mauelshagen Otto-Roehm-Strasse 71c Postfach 10 05 41 mge@ez-darmstadt.telekom.de 64205 Darmstadt Germany +49 6151 886-425 FAX-386 =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
- 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/