Compiling the RTEMS HEAD

[Ralf Corsepius]

On 01/26/2011 08:04 AM, Sebastian Huber wrote:
> On 01/26/2011 03:09 AM, Ralf Corsepius wrote:
>> On 01/25/2011 11:15 PM, Joel Sherrill wrote:
> [...]
>>> RTEMS uses C99,
>> Wrong ... RTEMS uses gnu99.
> [...]
>
> It would be really nice if we can document somewhere which C standard uses
> RTEMS.  This also affects a PR:

As you probably know, we are using "whatever gcc uses as default".

I.e. this is a moving target, which is widely out of our control and 
hence doesn't make much sense.

This is worse with natively compiled parts of the source tree which are 
compiled by vender-supplied toolchains: What they are doing is entirely 
out of our control, esp. when taking into account that esp. under Linux 
each vendor, has custom changes applied, which may change "their 
understanding of and level of compliance with" standards.

In other words, in general, the only way to escape this is to write 
portable code and/or to apply configure-checks to cope with changing 
standards and different levels of standard-compliance.

Also note that wrt. RTEMS itself, we escape a lot of the nastiness 
lurking inside by "tying" specific RTEMS releases to specific 
GNU-toolchain releases. The breakdown this thread started with clearly 
exposes this.

Ralf