This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
[Bug bootstrap/47215] [4.6 Regression] Failed to bootstrap
- From: "ktietz at gcc dot gnu.org" <gcc-bugzilla at gcc dot gnu dot org>
- To: gcc-bugs at gcc dot gnu dot org
- Date: Sun, 16 Jan 2011 08:34:58 +0000
- Subject: [Bug bootstrap/47215] [4.6 Regression] Failed to bootstrap
- Auto-submitted: auto-generated
- References: <bug-47215-4@http.gcc.gnu.org/bugzilla/>
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47215
--- Comment #15 from Kai Tietz <ktietz at gcc dot gnu.org> 2011-01-16 08:34:56 UTC ---
I will prepare today a patch, which removes the part in java's decl.c, and adds
to i386 the additional check for initialized va_list_type_node in alignment
check.
Nevertheless, I see this more as a working hack, then as a solution. By looking
through the ChangeLog's of java FE, I see that in 1999 va_list_type_node was
added, but there is no line to be found in ChangeLog that this type
initialization was removed. So, it is in my understanding an regression.
That java's FE should use base type-machinery from tree.c would be something
good, but this design change should be done (or at least triggered) by one of
the java FE maintainers and looks to me in stage 4 a bit risky.