[Patch,Fortran] PR36158 - Add transformational version of BESSEL_JN intrinsic

[Toon Moene]

Steve Kargl wrote:

> The rather limited range of i before one hits an INF is
> why I was surprised J3 included the transformational form
> in F2008.  IIRC, you raised a concern with (or request to 
> removei from?) F2008 over the transformational form and
> J3 rejected it saying something about recursion is a
> common method to compute besseli functions.  I think J3
> underestimated the care that is needed and the fact that
> the range is rather limited without specialized algorithms.

I think you are right in your assumptions about J3's BESSEL function 
prowess.  The only reason for requesting transformational BESSEL 
functions was the fact that they could use the recursive method to 
compute the function result (this is all AFAICRecall, BTW).

AFAICRecall, Tobias' concern was about a transformational *intrinsic* 
function, as that wasn't possible within the gfortran framework back 
then (but he can probably explain that better than I).

Kind regards,

-- 
Toon Moene - e-mail: toon@moene.org - phone: +31 346 214290
Saturnushof 14, 3738 XG  Maartensdijk, The Netherlands
At home: http://moene.org/~toon/; weather: http://moene.org/~hirlam/
Progress of GNU Fortran: http://gcc.gnu.org/gcc-4.5/changes.html#Fortran