Re: andrea buffer code (2.2.9-C.gz)

Manfred Spraul (masp0008@stud.uni-sb.de)
Fri, 21 May 1999 13:50:57 +0200 (CST)

Andrea wrote:
> >number of spinlocks and choose which to use with some reductive function?
> I think that if you want to SMP scale better it worth to
> pay _only_ with memory wastage and _not_ with wasted CPU cycles.
This is not always true: as Ingo pointed out, more memory
means more cache misses, and that costs CPU cycles as well.
The sometimes the reductive function could be as simple as
(x&0xFF).

The advantage of a reductive function is that - if you have a suitable
parameter - you can achieve parallel execution even without a memory
structure (e.g. cache lookup's: instead of one fully associated
cache with one lock you implement an array of 4(16,...)-way set
associate caches with one lock for each cache)

> My latest code is placed here. masp0008 feel free to give it a review.

I've downloaded it yesterday, but I don't know when I'll have enough
time for a review.

--
Manfred


-
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/