This is the mail archive of the gcc-bugs@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

[Bug fortran/42736] [4.3/4.4/4.5 Regression] Wrong-code with allocatable compounds



------- Comment #6 from burnus at gcc dot gnu dot org  2010-01-14 11:18 -------
The problem seems to be that for the tmp array, one has a RANK 1 array:
  "realdata(:,1)"
thus gfc_trans_create_temp_array only gets info->dimen == 1 and thus it only
fills the data for rank 1. However, the stride for rank 2 is later used (in
this case unnecessarily as (:,1) is contiguous.) I do not see ad hoc how this
can be solved.

Paul, do you have an idea? (BTW: 4.4.3 is supposed to get its RC tomorrow thus
it will be rather a 4.4.4/4.5 item, unless we are really fast.)


-- 

burnus at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |pault at gcc dot gnu dot org


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=42736


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]