This is the mail archive of the gcc-patches@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]

Re: [tree-ssa] Remove LOOP_EXPR


Hello,

> They're being generated by stmt.c:expand_start_loop.  I am quite lost
> around this code :)  So, could I ask one of you folks to take a look?
> 
> The test case I was looking at is
> gfortran.fortran-torture/compile/ambig.f90

the problem is that fortran frontend does not call gimplification at all.
I am not really sure whether this is intentional (is it trying to
produce gimple directly?).  Anyway, this patch fixes the problem.

Zdenek

	* tree-optimize.c (tree_rest_of_compilation): Call
	gimplification.

Index: tree-optimize.c
===================================================================
RCS file: /cvs/gcc/gcc/gcc/tree-optimize.c,v
retrieving revision 1.1.4.45
diff -c -3 -p -r1.1.4.45 tree-optimize.c
*** tree-optimize.c	4 Sep 2003 13:52:38 -0000	1.1.4.45
--- tree-optimize.c	6 Sep 2003 20:10:02 -0000
*************** tree_rest_of_compilation (tree fndecl)
*** 219,226 ****
    /* Gimplify the function.  Don't try to optimize the function if
       gimplification failed.  */
    if (!flag_disable_gimple
!       && (keep_function_tree_in_gimple_form (fndecl)
!           || gimplify_function_tree (fndecl)))
      {
        /* Debugging dump after gimplification.  */
        dump_function (TDI_gimple, fndecl);
--- 220,227 ----
    /* Gimplify the function.  Don't try to optimize the function if
       gimplification failed.  */
    if (!flag_disable_gimple
!       && keep_function_tree_in_gimple_form (fndecl)
!       && gimplify_function_tree (fndecl))
      {
        /* Debugging dump after gimplification.  */
        dump_function (TDI_gimple, fndecl);


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