Is there any framebuffer testcases?

[Joel Sherrill]

Attached is a patch to newlib to add the symbols I think are needed.

Please let me know if this helps.

--joel

On Wed, Apr 13, 2016 at 9:14 AM, Joel Sherrill <joel at rtems.org> wrote:

>
>
> On Wed, Apr 13, 2016 at 8:43 AM, Sebastian Huber <
> sebastian.huber at embedded-brains.de> wrote:
>
>>
>>
>> On 13/04/16 15:36, Peng Fan wrote:
>>
>>> On Wed, Apr 13, 2016 at 12:51:54PM +0200, Sebastian Huber wrote:
>>>
>>>> >On 13/04/16 12:33, Peng Fan wrote:
>>>>
>>>>> >>Ok. I see it. It was finally archived in to libscore.a. Please
>>>>> kindly guide
>>>>> >>me to fix this, I do not know where to begin.
>>>>>
>>>> >
>>>> >Sorry, I didn't follow the thread. What is the problem?
>>>>
>>> As Chris posted,
>>>
>>> "
>>> The test is simple:
>>>
>>>    $ cat hw.c
>>>    #include <stdio.h>
>>>    int main(void)
>>>    {
>>>      printf("hello world\n");
>>>      return 0;
>>>    }
>>>    $ arm-rtems4.12-gcc -o hw hw.c
>>>
>>>
>>> The actual failing test case used by configure is attached. It would seem
>>> changes to newlib or gcc have broken building 3rd party packages.
>>>
>>> $ i386-rtems4.12-gcc -o conftest conftest.c
>>>
>>> /opt/work/rtems/4.12/lib/gcc/i386-rtems4.12/6.0.0/../../../../i386-rtems4.12/bin/ld:
>>> warning: cannot find entry symbol _start; defaulting to 0000000008048074
>>>
>>> /opt/work/rtems/4.12/lib/gcc/i386-rtems4.12/6.0.0/../../../../i386-rtems4.12/lib/libc.a(lib_a-fclose.o):
>>> In function `_fclose_r':
>>>
>>> /opt/work/chris/rtems/rsb/rtems-source-builder.master/rtems/build/i386-rtems4.12-gcc-6-20160327-newlib-2.4.0-x86_64-freebsd10.1-1/build/i386-rtems4.12/newlib/libc/stdio/../../../../../gcc-6-20160327/newlib/libc/stdio/fclose.c:117:
>>> undefined reference to `_Mutex_recursive_Release'
>>>
>>
>> Ok, we need probably some hacks in Newlib to make Autoconf happy in case
>> you don't build against the proper RTEMS libraries.
>>
>>
> As I posted a few messages back, based on what symbols were missing when I
> did  a link on "whole library" libc.a and crt0.o, these were undefined
> that wouldn't
> come from libgcc*
>
> _Mutex_Acquire
> _Mutex_recursive_Acquire
> _Mutex_recursive_Release
> _Mutex_Release
> posix_memalign
>
> They need to be in newlib/libc/sys/rtems/crt0.c.
>
> It would be nice to have a repeatable way to check for missing symbols in
> the
> dummy crt0.c. I think my procedure was close but not 100%.
>
>
>
>>
>> --
>> Sebastian Huber, embedded brains GmbH
>>
>> Address : Dornierstr. 4, D-82178 Puchheim, Germany
>> Phone   : +49 89 189 47 41-16
>> Fax     : +49 89 189 47 41-09
>> E-Mail  : sebastian.huber at embedded-brains.de
>> PGP     : Public key available on request.
>>
>> Diese Nachricht ist keine geschäftliche Mitteilung im Sinne des EHUG.
>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.rtems.org/pipermail/users/attachments/20160413/9de4e2d5/attachment-0002.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-libc-sys-rtems-crt0.c-Add-symbols-required-to-link-a.patch
Type: application/octet-stream
Size: 1511 bytes
Desc: not available
URL: <http://lists.rtems.org/pipermail/users/attachments/20160413/9de4e2d5/attachment-0002.obj>