JFFS3 & performance

Jörn Engel joern at wohnheim.fh-wedel.de
Wed Dec 22 12:33:13 EST 2004

On Wed, 22 December 2004 17:39:08 +0100, Joakim Tjernlund wrote:
> OK, then there might be worth considering a reverse checksum.

Looks like it.

> This generates a little better code on PPC.
> Not tested.

Thanks.  Imo this variant is less readable and doesn't help i386, so
I'll ignore it for now.  If an i386-optimized engel32 is in the
adler32 range, I'll get back to it.  If not, let's just drop the whole

> > BTW: I reject the name "Engler32".  Few names are worse than
> > "engel32", but that one is. ;)
> I am so sorry, I felt something was wrong when I wrote that but I didn't
> check, my bad.

No offense taken.  Btw, if you know a better name, that would be nice.
There should be quite a few, I'm just too lazy to think of one.


Fancy algorithms are slow when n is small, and n is usually small.
Fancy algorithms have big constants. Until you know that n is
frequently going to be big, don't get fancy.
-- Rob Pike

More information about the linux-mtd mailing list