octave-maintainers
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [CHANGESET]: First attempt at a single precision type.


From: Jaroslav Hajek
Subject: Re: [CHANGESET]: First attempt at a single precision type.
Date: Mon, 28 Apr 2008 10:12:05 +0200

On Sun, Apr 27, 2008 at 10:55 PM, David Bateman <address@hidden> wrote:
> Its not attached to this message as the changeset is nearly 5MB in size,
>  but I have a working and reasonably complete adaptation of Octave to
>  allow the use of single precision types. It can be found in my mercurial
>  repository at
>
>  http://hg.dbateman.org/octave
>
>  and it is changeset number 8002 (1e224fa7f5bc).
>
>  Basically, as this is just a copy of the double precision type I'd
>  always thought it would be relatively easy to add this. With that idea I
>  didn't attempt to make template versions of the copied code to reuse it
>  and just copied it as the easier fashion to get this working. Therefore,
>  there might be some simplifications that might be done..
>
>  Also although I've converted most functions to allow floating point
>  types (or rather those that made sense to be modified), I haven't
>  converted some including some basic ones like find and quad.
>
>  I've assumed that the correct mixed double/single behavior is to promote
>  the single type to double, and so I've used the numeric conversion
>  function of the octave_value to do this rather than have to define all
>  of the mixed type opeators.
>
>  Finally it has had little testing, and in fact doesn't have any test
>  code for it at all yet. I suppose that almost all of the existing test
>  code  should be converted to allow floating versions. I'm sure there is
>  some easy way to do that but I'm not sure I see it. However, I'd rather
>  get the code out there before going further, to allow feedback on the
>  patch and for others to test the code out and if motivated to contribute.
>
>  Regards
>  David
>

Hi David,

attached is a changeset correcting the single precision versions of
qrupdate routines.

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

Attachment: qrupdate_single.diff
Description: Text Data


reply via email to

[Prev in Thread] Current Thread [Next in Thread]