This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
[tree-profiling] tree-inline cleanup 2/many
- From: Jan Hubicka <jh at suse dot cz>
- To: gcc-patches at gcc dot gnu dot org
- Date: Sat, 16 Apr 2005 14:53:44 +0200
- Subject: [tree-profiling] tree-inline cleanup 2/many
Hi,
another if 0 hunk that should be unnecesary - we warn about returns
early in the pre-inline optimization block and thus we should not need
take any special care of functions being inlined.
Honza
Index: tree-inline.c
===================================================================
RCS file: /cvs/gcc/gcc/gcc/tree-inline.c,v
retrieving revision 1.26.2.83.2.38
diff -c -3 -p -r1.26.2.83.2.38 tree-inline.c
*** tree-inline.c 15 Apr 2005 13:27:36 -0000 1.26.2.83.2.38
--- tree-inline.c 16 Apr 2005 12:51:59 -0000
*************** expand_call_inline (tree *tp, int *walk_
*** 2378,2398 ****
a self-referential call; if we're calling ourselves, we need to
duplicate our body before altering anything. */
dup_fndecl = copy_body (id, id->oic_basic_block->count, id->oic_basic_block->frequency);
- #if 0
- /* If the function uses a return slot, then it may legitimately
- fall through while still returning a value, so we have to skip
- the warning here. */
- if (warn_return_type
- && !TREE_NO_WARNING (fn)
- && !VOID_TYPE_P (TREE_TYPE (TREE_TYPE (fn)))
- && return_slot_addr == NULL_TREE
- && block_may_fallthru (dup_fndecl))
- {
- warning ("control may reach end of non-void function %qD being inlined",
- fn);
- TREE_NO_WARNING (fn) = 1;
- }
- #endif
id->current_node = old_node;
my_cfun = DECL_STRUCT_FUNCTION (dup_fndecl);
/* Append copied EH information to that of current function. */
--- 2358,2363 ----