[Patch, Fortran, 4.5 Regression] PR 40822: Internal compiler error when Fortran intrinsic LEN referenced before explicit declaration

Janus Weil janus@gcc.gnu.org
Fri Jul 24 12:28:00 GMT 2009


2009/7/24 Tobias Burnus <burnus@net-b.de>:
> Janus Weil wrote:
>> The new patch is again regtested on x86_64-unknown-linux-gnu. Ok now?
>>
> Looks OK to me. Thanks for the patch.

Committed as r150047.

Cheers,
Janus



>> 2009-07-23  Janus Weil  <janus@gcc.gnu.org>
>>
>>       PR fortran/40822
>>       * array.c (gfc_resolve_character_array_constructor): Use new function
>>       gfc_new_charlen.
>>       * decl.c (add_init_expr_to_sym,variable_decl,match_char_spec,
>>       gfc_match_implicit): Ditto.
>>       * expr.c (gfc_simplify_expr): Ditto.
>>       * gfortran.h (gfc_new_charlen): New prototype.
>>       * iresolve.c (check_charlen_present,gfc_resolve_char_achar): Use new
>>       function gfc_new_charlen.
>>       * module.c (mio_charlen): Ditto.
>>       * resolve.c (gfc_resolve_substring_charlen,
>>       gfc_resolve_character_operator,fixup_charlen,resolve_fl_derived,
>>       resolve_symbol): Ditto.
>>       * symbol.c (gfc_new_charlen): New function to create a new gfc_charlen
>>       structure and add it to a namespace.
>>       (gfc_copy_formal_args_intr): Make sure ts.cl is present
>>       for CHARACTER variables.
>>
>>
>> 2009-07-23  Janus Weil  <janus@gcc.gnu.org>
>>
>>       PR fortran/40822
>>       * gfortran.dg/char_length_16.f90: New.
>>
>



More information about the Gcc-patches mailing list