[email protected] writes:

> On Tue, Dec 10, 2013 at 4:55 PM, Dima Kogan <[email protected]> wrote:
>> [email protected] writes:
>>
>>> Any further status on this, Dima?
>>>
>>> <snip>

Chris:

OK. You are right. I am being too hard-line about this. I will look into
changing PDL::FFTW3 to work both ways. Can we also make a parallel
effort to align memory inside PDL itself? To evaluate the viability of
my tree that does this, we need

- some windows users to confirm that _aligned_free() must indeed be used
  instead of free(), and that this is the only way to get a
  memory-aligned malloc()

- some test code for mmapped piddles. This should probably be added to
  the test suite, if it's something we care about

I'll need a few days to find time to look at the PDL::FFTW3 piece of
this.

dima

_______________________________________________
Perldl mailing list
[email protected]
http://mailman.jach.hawaii.edu/mailman/listinfo/perldl

Reply via email to