This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
[Bug bootstrap/53963] [4.8 Regression] LTO bootstrap failed with bootstrap-profiled
- From: "hjl.tools at gmail dot com" <gcc-bugzilla at gcc dot gnu dot org>
- To: gcc-bugs at gcc dot gnu dot org
- Date: Sat, 14 Jul 2012 16:52:42 +0000
- Subject: [Bug bootstrap/53963] [4.8 Regression] LTO bootstrap failed with bootstrap-profiled
- Auto-submitted: auto-generated
- References: <bug-53963-4@http.gcc.gnu.org/bugzilla/>
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53963
--- Comment #2 from H.J. Lu <hjl.tools at gmail dot com> 2012-07-14 16:52:42 UTC ---
(In reply to comment #1)
> > "--prefix=/usr/local" is the key to trigger this regression.
> > The stage 3 compiler is miscompiled. Revision 189349 is OK.
>
> I don' think so. I have configured as
>
> ../gcc-svn/trunk/configure --with-build-config=bootstrap-lto
> --enable-languages=c
I believe --prefix=/usr/local is the default. I didn't get this
failure with --prefix=/usr/gcc-4.8.0.
> and fails as well. Backing out r189453 works OK.
I am testing reverting:
diff --git a/gcc/gimple.c b/gcc/gimple.c
index 398cb1f..b419591 100644
--- a/gcc/gimple.c
+++ b/gcc/gimple.c
@@ -3422,10 +3422,6 @@ gimple_types_compatible_p_1 (tree t1, tree t2,
type_pair_
t p,
goto different_types;
}
- /* If their attributes are not the same they can't be the same type. */
- if (!attribute_list_equal (TYPE_ATTRIBUTES (t1), TYPE_ATTRIBUTES (t2)))
- goto different_types;
-
/* Do type-specific comparisons. */
switch (TREE_CODE (t1))
{