Re: [2.4.17/18pre] VM and swap - it's really unusable

Daniel Phillips (phillips@bonn-fries.net)
Tue, 22 Jan 2002 03:10:42 +0100


On January 22, 2002 12:56 am, yodaiken@fsmlabs.com wrote:
> I have not seen that argued - certainly I have not argued it myself.
> My argument is:
> It makes the kernel _much_ more complex

The patch itself is simple, so this must be an extended interpretation of the
word 'complex'.

> It has known costs e.g. by making the lockless
> per-processor caching more difficult if not impossible

Not at all, the lazy man's way of dealing with this is to disable preemption
around that code, an efficient operation.

> It seems to lead to a requirement for inheritance

I don't know about that. From the (long) thread above, it looks like you
haven't successfully proved the assertion that -preempt introduces any new
inheritance requirement.

> It has no demonstrated benefits.

Demonstrated to who? I have certainly demonstrated the benefits to myself,
and others have attested to doing the same.

As far as arguments go, your main points don't seem to be rooted in firm
ground at all. On the other hand, the proponents of this patch have
compelling arguments: it makes Linux feel smoother, it makes certain tests
run faster, it doesn't slow anything down measurably, it's stable and so on.
I even explained why it does what it does. I don't understand why you're so
vehemently opposed to this, especially as it's a config option.

--
Daniel
-
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/