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

[Tobias Burnus]

  Steve Kargl wrote:
> I'll look over the patch in more detail in a few hours.

Thanks! Here is an updated version. With the help of Daniel K, I have 
found a solution to the constructor issue; JN now works as it should. YN 
also works, though for some of the examples, the result differs quite a 
bit from the result of mpfr_yn. I do not know whether I have chosen 
particularly difficult values or whether the algorithm has problems. 
Please check. (Cf. bessel_5.f90 - the ULP values given are the 
differences which maximally occur [though 1 ULP could be 0 ULP in some 
cases].)

Build and currently regtesting on x86-64-linux.
OK for the trunk?

Tobias
-------------- next part --------------
A non-text attachment was scrubbed...
Name: bessel_transf-2.diff
Type: text/x-patch
Size: 18294 bytes
Desc: not available
URL: <http://gcc.gnu.org/pipermail/fortran/attachments/20100817/dbabe9b8/attachment.bin>