Re: [RFC] LZO de/compression support - take 6

["Nitin Gupta" <nitingupta910@gmail.com>]

On 5/28/07, Adrian Bunk <bunk@stusta.de> wrote:
> On Mon, May 28, 2007 at 08:36:44PM +0530, Nitin Gupta wrote:
> >...
> > So, before this change, it will be good if it gets merged in mainline
> > and tested, at least for correctness, on all supported achs. All the
> > while, we will have a good feeling that there is still a good scope
> > for perf improvement :)
>
> The correct order is:
> - create one version with all the optimizations you have in mind

Already done. One more optimization is regarding use of memcpy() in
place of COPY4() macros and open byte-by-byte copying. There are some
places where it's very hard to get it correct without adding
additional checks on various values which casues futher overhead by
iteslf - even then I could not get them correct so I decided not to go
with this particular optimization by myself.

> - then ensure that it works correctly on all architectures and

Already tested on x86, amd64, ppc (by Bret). I do not have machines
from other archs available. Bret tested 'take 3' version but no
changes were introduced in further revisions that could affect
correctness - but still it will be good to have this version tested
too. Only with inclusion in -mm and testing by much wider user base
can make it to mainline (I suppose nobody uses -mm for production use
anyway).

>   document why your version is that much faster than the original
>   version and why you know your optimizations have no side effects

Replied in earlier mail regarding this.

> - then get it tested in -mm
>

This is what I am looking for :)

> After these steps, you can start considering getting it into mainline.
>


- Nitin