patches awaiting approval

[Paul Thomas]

Dear All,

I have some patches waiting approval.  Whilst they are a long way from 
needing pinging, my schedule is a bit difficult this year and I am 
worried that not only will the patches get bit-rot but I will too.

They are:

(i) PR33554 - [4.3 regression]           
http://gcc.gnu.org/ml/fortran/2007-10/msg00005.html

Note that this suffered some tidying up before submission, with the 
result that there is an error :-)

+ /* Initialize INTENT(OUT) derived type dummies.  */
+ static tree
+ init_intent_out_dt (gfc_symbol * proc_sym, tree body)
+ {
+   stmtblock_t fnblock;
+   gfc_init_block (&fnblock); /*put this after the declarations!!!!  */
+   gfc_formal_arglist *f;
+   gfc_expr *tmpe;
+   tree tmp;
+   tree present;


(ii) PR33550 - http://gcc.gnu.org/ml/fortran/2007-09/msg00471.html      
okayed by Jerry - thanks
(iii) PR31154 , PR31229 and PR33334 - 
http://gcc.gnu.org/ml/fortran/2007-09/msg00457.html
(iv) PR33566 - http://gcc.gnu.org/ml/fortran/2007-09/msg00454.html

All these patches have been tested by Dominique, stand up to cp2k/tonto 
and so just need somebody to cast an eye over the style.  Note that 3 of 
these are amongst the remaining 8 on the first line of the Wiki's bug 
bashing list.

Another fix, which is rock solid but has not been posted is that to 
PR33542 - see the PR itself.  I am inclined to apply it as obvious but 
would appreciate somebody casting an eye over it.  The submission, with 
changelogs and testcase is stuck as a draft on a workstation that I 
cannot get to until tomorrow.

Cheers

Paul