[Bug ada/14538] *-rtems broken for gnat

["joel dot sherrill at oarcorp dot com" <gcc-bugzilla at gcc dot gnu dot org>]

------- Additional Comments From joel dot sherrill at oarcorp dot com  2004-03-12 13:29 -------
Subject: Re:  *-rtems broken for gnat

charlet at act-europe dot fr wrote:

>------- Additional Comments From charlet at act-europe dot fr  2004-03-12 06:24 -------
>Subject: Re:  *-rtems broken for gnat
>
>  
>
>>Please review so I can commit this before 3.4.0 is cut.
>>    
>>
>
>I'd suggest you post your patch with the missing ChangeLog on
>gcc-patches@, following the usual patch submission rules.
>
>The init.c change seems to be a no-op, so I'd rather not integrate it.
>
>  
>
Actually it isn't a nop.  It puts the simple __rtems__ check ahead of  a 
check that
is something like the following that I believe indicates the cpp 
predefines for one of the
vxworks targets is broken:

if (AIX or unix) and !vxworks

At least some of the embedded sparc targets were built using solaris 
config files and
ended up with unix defined.  I can see where this could happen on 
another embedded
target cpu configuration and wanted to avoid it accidentally being 
tripped.  In fact,
I submitted a fix for sparc to address this for sparc-rtems.  I simply 
moved a very
straight forward check on an OS specific predefine that is guaranteed to 
be there
ahead of one that is questionable and could be unintentally tripped.

This same problem also resulted in sys/termios.h being included from the 
<ctrl>c
code since it ended up thinking it was on a unix system when it wasn't.

I will also post the patch to the bugs list once I get in.

>Arno
>
>
>  
>




-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=14538