[Changeset] Re: Faster Array transpose

[Jaroslav Hajek]

On Wed, May 21, 2008 at 12:08 PM, John W. Eaton <jwe at bevo.che.wisc.edu> wrote:
> On 21-May-2008, Jaroslav Hajek wrote:
>
> | What a huge update :) I see this is interferring non-trivially with
> | all my proposed changes (the compound operators repo and two pending
> | patches.) Shall I merge them? (And re-send the patches, probably).
>
> I was planning to look at those changes today.  If you can update the
> changeset in the next few hours so that it applies cleanly, that would
> help.  Otherwise, I will try to do it.
>

I have two pending changesets:
the improved matrix_type SPD check:
http://www.nabble.com/matrix_type-check-tt16883771.html#a16883771
and the (quite old) 2D pchip interpolation
http://www.nabble.com/pchip-method-for-interp2-tt16049957.html#a16049957

The first one actually applies cleanly onto current tip, but obviously
leaves the old code
for FloatMatrix and FoatComplexMatrix. A copy would suffice, but I
think I can easily remake this to use templates to avoid duplicating
code.

The second patch should be more trivial, as most of it is m-file code.

As for the compound operators repo, I intend to import changes from
your main repo, do a merge, then add code to support single precision
where appropriate. I.e., that will be two changesets more. You can
then simply pull from the repo. Is that OK or is there a more
desirable way?

> Thanks,
>
> jwe
>



-- 
RNDr. Jaroslav Hajek
computing expert
Aeronautical Research and Test Institute (VZLU)
Prague, Czech Republic
url: www.highegg.matfyz.cz