You are right. How about this one? Thanks
Shuxin
Index: gcc/tree.c
===================================================================
--- gcc/tree.c (revision 3413)
+++ gcc/tree.c (working copy)
@@ -10522,6 +10522,23 @@
if (CPR()) { /* C++ */
switch (TREE_CODE(t)) {
case FUNCTION_DECL:
+
+ if (!DECL_LANG_SPECIFIC (t)) {
+ /* Following C++ snippet necessitates the need of condition
+ * "if (!DECL_LANG_SPECIFIC(t))"
+ *
+ * #pragma weak bar1 = foo1
+ * extern "C" void foo1 (void) { }
+ *
+ * The front end creates a decl tree both for bar1() and
foo1().
+ * The decl.lang_specific field of the decl tree for bar1()
is NULL,
+ * as evidenced in maybe_apply_pending_pragma_weaks().
+ * If the <t> being procesed is the decl tree for bar1(),
calling
+ * macros like DECL_GLOBAL_CTOR_P will incur segamentation
fault.
+ */
+ break;
+ }
+
_gs_bv(cp_decl_flags, GS_DECL_GLOBAL_CTOR_P,
DECL_GLOBAL_CTOR_P(t));
_gs_bv(cp_decl_flags, GS_DECL_GLOBAL_DTOR_P,
On Thu, Dec 2, 2010 at 2:41 PM, Gautam Chakrabarti <gautam.c...@yahoo.com>wrote:
> Hi,
>
> Could you use the conditional to guard code at a smaller granularity, i.e.,
> maybe guard only the code under the FUNCTION_DECL case statement (ideally
> enclose the smallest chunk of code depending on this)? The reason is there
> is some other code under the CPR() conditional that needs
> DECL_LANG_SPECIFIC, and they are already guarded as such. There is some code
> under CPR() which is valid even if DECL_LANG_SPECIFIC is NULL. So we
> probably don't want to disable these codes just because this field is null,
> unless we can prove it's safe to do so.
>
> Thanks,
> Gautam
>
> ------------------------------
>
>
------------------------------------------------------------------------------
Increase Visibility of Your 3D Game App & Earn a Chance To Win $500!
Tap into the largest installed PC base & get more eyes on your game by
optimizing for Intel(R) Graphics Technology. Get started today with the
Intel(R) Software Partner Program. Five $500 cash prizes are up for grabs.
http://p.sf.net/sfu/intelisp-dev2dev
_______________________________________________
Open64-devel mailing list
Open64-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/open64-devel
