This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
Re: Java broken after i386 PIC change
- To: Bryce McKinlay <bryce at albatross dot co dot nz>
- Subject: Re: Java broken after i386 PIC change
- From: Richard Henderson <rth at redhat dot com>
- Date: Sat, 20 Jan 2001 20:19:08 -0800
- Cc: apbianco at cygnus dot com, gcc-bugs at gcc dot gnu dot org, martin dot kahlert at infineon dot com, java-discuss at sourceware dot cygnus dot com
- References: <20010116123822.A16053@keksy.muc.infineon.com> <3A6675A9.745842BC@albatross.co.nz> <14950.34116.664233.394672@deliverance.cygnus.com> <3A66CD61.9E2D7772@albatross.co.nz> <20010120181423.A4865@redhat.com> <3A6A5F36.81CD863D@albatross.co.nz>
On Sun, Jan 21, 2001 at 05:01:58PM +1300, Bryce McKinlay wrote:
> So, are you saying that Java is working for you on x86,
> with the current tree? Interesting.
Yes. With cvs as of 3:30 PST, I get
=== libjava Summary ===
# of expected passes 1478
# of unexpected failures 64
# of unexpected successes 26
# of expected failures 164
This is on i686-pc-linux-gnu with rh7 plus errata,
ie glibc-2.2-and binutils-2.10.0.18-1.
> 0x804aa99 <__register_frame_info+9>: call 0x0 <---- eh?
Interesting. Looks suspiciously like a binutils problem.
What version do you have installed?
In libgcc.a, that insn plus relocation should look like
604: e8 fc ff ff ff call 605 <__register_frame_info+0x5>
605: R_386_PC32 .gnu.linkonce.t.i686.get_pc_thunk
correct? I can't imagine how that could get relocated to
zero unless your binutils decided to put that section at
vma zero. Would you send me objdump -h of the failing
executable?
What happens when you run the libstdc++ testsuite? Does
that fail too? (Or do you not have bash2 installed so it
just doesn't run at all...)
r~