[patch,wwwdata] Update Fortran section of gcc-4.2/changes.html

[Tobias Burnus]

Hi Thomas,

Thomas Koenig wrote:
> this is OK for checkin.
>   
Thanks, but Jerry already approved it 5 days ago ;-)


But you might be interested in looking at the other fortran patches, see:

http://www.dberlin.org/patches/patches/list

1401    Paul Thomas     Sat Nov 25 07:04:33 2006
    [Patch, fortran] PR29464 - problem with duplicate USE, ONLY of
procedure in INTERFACE

1405    Tobias Burnus   Tue Nov 28 06:43:09 2006
    [patch,fortran] Allow "IMPORT :: list-of-symbols" in interface bodys
in modules (PR27546)

1409    Tobias Burnus   Wed Nov 29 11:54:07 2006
    [patch,fortran] Fix border case where consecutive calls to
date_and_time can return times out of sequence (PR30015)    

1410    Paul Thomas     Wed Nov 29 11:59:00 2006
    [Patch, fortran] PR27996 and PR29978 - warnings and errors on
character lengths

(There is also Tobias Schlüter's "Re: [gfortran, testsuite] Fix bounds
checking for substrings, enable bounds chekcing in dg testsuite,
assorted fixes", but I lost the overview what was already checked in,
what approved and what not. At least the dg testsuite changes are not
yet checked in.)


Tobias

PS: If one vectorizes the sin and cos calls in fatigue.f90, the
execution time drops from 21s to 8.6s, which is faster than ifort (and
slightly slower than sunf95). However, applying the great sincos patches
by Richard is not enough as the tree optimizer does not realize that
sin() and cos() have the same arguments; I had to tweak fatigue.f90
manually. [PR30038]