> From: Linus Torvalds <torvalds@transmeta.com>
>....
> I really don't see the point there. It's slower, more complex, and uses
> more memory. Somebody please tell me why I shouldn't just get rid of it
> and revert to the old kmalloc(), which we know is stable and has none of
> these misfeatures?
I was very sceptical about slab when it appeared too.
To be honest, I maniacally cut it from kernel for pretty long time
since 2.1.42 8)
When I heard Andy's idea about double alloc, I laughed until
decided to test his patch. I stopped to laugh then 8)
It sounds as paradox, hell knows why (apparently, it is some strange
cache effect), but Andy's double alloc takes less effective time
than NO ALLOC at all. (I mean using small finite pool of preallocted buffers
with zero allocation/free time).
BTW even dumb measuring alloc_skb()/kfree_skb() timings
double alloc is not slower than old one, working through
2.0 style kmalloc. It is also paradoxical if we will compare code
comlexity by eyes, but it is experimental fact.
Alexey Kuznetsov
-
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.altern.org/andrebalsa/doc/lkml-faq.html