JFFS3 & performance

[Jörn Engel]

On Fri, 17 December 2004 10:56:22 +0000, Artem B. Bityuckiy wrote:
> 
> > Another idea, calculate the CRC backwards(from the end of the buffer to the beginning).
> > That give better L1 cache behaviour.
> >
> Why (may be some URL)? For all archs? 

Sounds rather odd indeed.  Most CPUs I know about either have no
hardware-prefetch, can only prefetch forward or can prefetch both
forward and backward.  In either case, calculating the CRC backward
should either be the same speed or slower than going forward.

But I like to learn something new.

Jörn

-- 
I've never met a human being who would want to read 17,000 pages of
documentation, and if there was, I'd kill him to get him out of the
gene pool.
-- Joseph Costello