>
> On Wed, 18 Jul 2001, Marcelo Tosatti wrote:
> >
> > Ok, I understand and I agree with doing _unconditional_
> > "zone_inactive_plenty()" instead of conditional
> > "zone_inactive_shortage()".
> >
> > This way we do not get _strict_ zoned behaviour (with strict I mean only
> > doing scanning for zones which have a shortage), making the shortage
> > handling smoother and doing "fair" aging in cases where there are not
> > specific zones under pressure.
>
> Cool.
>
> Willing to write a patch and give it some preliminary testing?
Sure. However its not _that_ easy. We do have a global inactive target.
There is no perzone inactive shortage, which is needed to calculate
"zone_inactive_plenty()".
> I also agree with the patch Rik sent in about GFP_HIGHUSER, that's
> orthogonal though (even if I suspect it could also have made the
> problem _appear_ much much more clearly).
Right.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/