arm multilib not building on head

[Joel Sherrill]

On 05/25/2011 06:39 AM, Ralf Corsepius wrote:
> On 05/24/2011 09:31 PM, Joel Sherrill wrote:
>> Hi,
>>
>> Can anyone build the arm-rtems4.11 target on
>> the CVS head using multilib?
> Yes - Without major problems.
I saw that after I let it build on another machine overnight
with the 4.6 based toolset.  This was on rtbf64a with a test
build of the gcc head.  I did a fresh check out of RTEMS,
re-bootstrapped and still didn't have any luck.

I won't get back to this until after the Munich class.
>> I used this configure
>> command:
>>
>> ../rtems/configure --target=arm-rtems4.11
>> --prefix=/home/joel/rtems-4.11-work/bsp-install/
>> --disable-multiprocessing --disable-smp --enable-cxx --disable-rdbg
>> --enable-maintainer-mode --enable-tests=samples --enable-networking
>> --enable-shttpd --enable-posix --disable-itron --disable-deprecated
>> --disable-ada --disable-expada --enable-multilib
> ... tons of unused options ...
>
>> I don't have time to investigate if this is
>> just me or a general breakage.
> You are not providing sufficinet information about what fails to be able
> to help.
>
Yeah.  I realized I should have at least said this was when
testing the gcc head.

Can you try to test this?  If it is something in gcc, we don't
want it to rot anymore.
> Ralf