Hi,
On Mon, Apr 18, 2011 at 2:26 PM, kevin diggs <diggskevin38@xxxxxxxxx> wrote:
> Hi,
>
> On Mon, Apr 18, 2011 at 2:23 PM, kevin diggs <diggskevin38@xxxxxxxxx> wrote:
>> Hi,
>>
>> On Mon, Apr 18, 2011 at 1:59 PM, Ian Lance Taylor <iant@xxxxxxxxxx> wrote:
>>>
>>> Your gcc is passing --sysroot to the linker but the linker was not
>>> configured with --with-sysroot.
>>>
>>> I think it's a bug that the linker has to be configured with
>>> --with-sysroot in order to pass the --sysroot option to the linker. I
>>> don't know why it works that way.
>>>
>>> Ian
>>>
>>
>> Uh ... I'm not really smart enough to argue with you ... but ... are
>> you sure? I use centrally located shell scripts to run configure so
>> that I can easily recreate stuff if I need to. The same script was
>> used to create binutils for both:
>>
>> #!/bin/bash
>>
>> # --prefix=/opt/cross/q700/binutils-2.16.1 \
>> CFLAGS="-O2 -march=pentium4 -fomit-frame-pointer" `dirname ${0}`/configure \
>> --target=m68k-netbsdelf1.6.1 \
>> --prefix=/opt/cross/q700/binutils-2.16.1 \
>> --with-mpfr=/opt/MPFR-2.4.2 \
>> --with-gmp=/opt/GMP-4.3.2 \
>> --enable-shared
>>
>> The only difference was pentium3 vs pentium4?
>>
>> Thanks!
>>
>> kevin
>>
>
> One other thing. I have not tried to build 4.2.4 on the pentium 3
> laptop. A 4.2.4 build could fail on it to. So could this sysroot thing
> be something unique to the 4.2 era?
>
> Thanks!
>
> kevin
>
And still one other thing. Maybe I have not seen this with 4.3.5 yet
since I can't get it to build ...
Thanks!
kevin
