Dear All,
I was intending to backport this patch to 6-branch. However, I see
that this is closed to all but regressions and documentation at
present.
The patch applies cleanly, except for two small chunks in resolve.c
which were easily inserted by hand, and bootstraps and regtests OK. I
will set aside the patch and wait for the release of 6.2 unless there
is demand for it to be applied now. I am somewhat nervous about doing
this, however, since it is a rather radical change to select type and
has been in trunk for less than two weeks.
Cheers
Paul
On 24 October 2016 at 12:18, Paul Richard Thomas
<paul.richard.tho...@gmail.com> wrote:
> Hi Andre,
>
> It was the comment for the function that you eliminated. I'll get rid
> of the comment. I was in a bit of a hurry and did indeed only look
> where you suggested. Sorry about that.
>
> Thanks
>
> Paul
>
> On 24 October 2016 at 11:07, Andre Vehreschild <ve...@gmx.de> wrote:
>> Hi Paul,
>>
>> concerning this:
>>
>>> >> Index: gcc/fortran/trans-stmt.c
>>> >> ===================================================================
>>> >> *** gcc/fortran/trans-stmt.c (revision 241439)
>>> >> --- gcc/fortran/trans-stmt.c (working copy)
>>> >> *************** gfc_trans_do_while (gfc_code * code)
>>> >> *** 2331,2336 ****
>>> >> --- 2331,2455 ----
>>> > <snipp>
>>> >> +
>>> >> + /* Translate an assignment to a CLASS object
>>> >> + (pointer or ordinary assignment). */
>>> >> +
>>> >> +
>>> >
>>> > Here is no routine the above comment could document. Left over from prior
>>> > version?
>>>
>>> This is in your tree, not mine :-)
>>
>> Well, now it is in everyone's tree. But I am guilty of pasting the wrong
>> location. The correct one is:
>>
>>> Index: gcc/fortran/trans-expr.c
>>> ===================================================================
>>> *** gcc/fortran/trans-expr.c (revision 241439)
>>> --- gcc/fortran/trans-expr.c (working copy)
>>> *************** gfc_trans_class_init_assign (gfc_code *c
>>> *** 1508,1513 ****
>>> }
>>>
>>>
>>> + /* Return the backend_decl for the vtable of an arbitrary typespec
>>> + and the vtable symbol. */
>>> +
>>> + tree
>>> + gfc_get_vtable_decl (gfc_typespec *ts, gfc_symbol **vtab)
>>> + {
>>> + gfc_symbol *vtable = gfc_find_vtab (ts);
>>> + gcc_assert (vtable != NULL);
>>> + if (vtab != NULL)
>>> + *vtab = vtable;
>>> + if (vtable->backend_decl == NULL_TREE)
>>> + return gfc_get_symbol_decl (vtable);
>>> + else
>>> + return vtable->backend_decl;
>>> + }
>>> +
>>> +
>>> + /* Translate an assignment to a CLASS object
>>> + (pointer or ordinary assignment). */
>>> +
>>
>> Still I do not see what the above documents. The next function is
>> realloc_lhs_warning () and the comment makes no sense to it.
>>
>>> +
>>> /* End of prototype trans-class.c */
>>
>> Regards,
>> Andre
>> --
>> Andre Vehreschild * Email: vehre ad gmx dot de
>
>
>
> --
> The difference between genius and stupidity is; genius has its limits.
>
> Albert Einstein
--
The difference between genius and stupidity is; genius has its limits.
Albert Einstein
