klibc - Mar 2005 - Cross compiling klibc for arm and linking shared stuff

Hi,

I am getting this:

-----
armv4-unknown-linux-gnu-ld  -o tests/fcntl.shared -e main interp.o tests/fcntl.o
-R libc.so /usr/lib/gcc-lib/armv4-unknown-linux-gnu/3.3.5/libgcc.a
armv4-unknown-linux-gnu-ld: internal error: aborting at
/space/var/tmp/portage/binutils-2.15.90.0.3-r4/work/binutils-2.15.90.0.3/ld/ldlang.c
line 3835 in lang_place_orphans
armv4-unknown-linux-gnu-ld: please report this bug
-----

When trying to cross compile for armv4, and was wondering if anybody
know of a patch to fix that ... I have tried different versions of
binutils as well.


Thanks,

-- 
Martin Schlemmer



-- 
---------------------[ Ciphire Signature ]----------------------
From: azarah@nosferatu.za.org signed email body (523 characters)
Date: on 09 March 2005 at 15:08:28 UTC
To:   klibc@zytor.com
----------------------------------------------------------------
: Ciphire has secured this email against identity theft.
: Free download at www.ciphire.com. The garbled lines
: below are the sender's verifiable digital signature.
----------------------------------------------------------------
00fAAAAAEAAABsES9CCwIAAHgBAAIAAgACACAjYe/qeWoHRrdDE3mOifT0WPgp92
kbkclxgX1lQsV9VwEAJOQiG+Djg3e7Ai63e8AMF0Hl/gle8q4Ud88Lecx5fcURUv
cwce4lmQpHCKQxabhybJnsCQccnbTb9otB1Pb0Sw=------------------[ End Ciphire Signed
Message ]----------------

Martin Schlemmer wrote:
> Hi,
> 
> I am getting this:
> 
> -----
> armv4-unknown-linux-gnu-ld  -o tests/fcntl.shared -e main interp.o
tests/fcntl.o -R libc.so /usr/lib/gcc-lib/armv4-unknown-linux-gnu/3.3.5/libgcc.a
> armv4-unknown-linux-gnu-ld: internal error: aborting at
/space/var/tmp/portage/binutils-2.15.90.0.3-r4/work/binutils-2.15.90.0.3/ld/ldlang.c
line 3835 in lang_place_orphans
> armv4-unknown-linux-gnu-ld: please report this bug
> -----
> 
> When trying to cross compile for armv4, and was wondering if anybody
> know of a patch to fix that ... I have tried different versions of
> binutils as well.
> 
Odds are they've broken ld -R again; it seems to break often and 
inconsistently among architectures.  I don't know why; apparently they 
need a test for it in their test suite.  Since it's an internal error in 
ld, it's pretty much impossible for anyone here to debug it, and you 
should be checking with the binutils people.

	-hpa