linux/posix port compiling problem

[Joel Sherrill]

Wei-Tsun Sun wrote:
> Hi,
>
> Sorry for the late reply.
>
> On Tue, 14 Apr 2009 16:45:20 +0200
> Ralf Corsepius <ralf.corsepius at rtems.org> wrote:
>
>   
>> Joel Sherrill wrote:
>>     
>>> Looking at the patch, it looks like it boils down to 3 things:
>>>
>>> + disable shell for posix port (easier to drop it from building at all).
>>>       
>> Well, rtems shell got infected with with newlib-proprietary, 
>> non-portable construct.
>>
>>     
>
> I am not very familiar with libmisc yet, but `shell` looks an important thing to me, or it does not look like its name to be...as a proper shell ?
>
>   
It is a small but more or less proper shell.  The issue it not
with the code you are trying to compile.  But the fact that the
unix port is to a "synthetic" target -- not real hardware.  It is
using the native unix libc.  Much of RTEMS is not buildable in
this environment -- including the shell, BSD TCP/IP stack and the
entire POSIX thread interface.

If you want a more proper RTEMS experience switch to
a real port (e.g. i386-rtems4.10) and run the RTEMS executables
on a simulator or real hardware.
>> IMO, we should get rid of them and either re-write the code or abandon it.
>>
>>     
>>> + use *.o on command line instead of from file with ar (why?)
>>>       
>> IMO, this patch doesn't make much sense.
>>
>>     
>
> When I built rtems, sometimes rel files were not generated for some wrapups, and "ls" generated an error code which stopped the building process
> due to "file (*.rel) not found".
>
>   
>>> + a small patch to provide a malloc initialize stub.
>>>       
>> This is a real bug in RTEMS.
>>
>> Ralf
>>     
>
>   


-- 
Joel Sherrill, Ph.D.             Director of Research & Development
joel.sherrill at OARcorp.com        On-Line Applications Research
Ask me about RTEMS: a free RTOS  Huntsville AL 35805
   Support Available             (256) 722-9985