Deep question for the REAL C GURU's out there ...

[Kurt Wall]

On Friday 25 February 2005 01:37, Roger Oberholtzer enlightened us 
thusly:
> On Thu, 2005-02-24 at 20:29 -0500, Kurt Wall wrote:
> > True, I suppose, but, as a minor point, precedence and sequence
> > points are determined by the standard. Any compiler worth the name
> > should get that right.
>
> We have been burned by C compilers from TI (DSP programming) and
> Microware (embedded OS/9). In the latter, the C compilers originally
> were explicitly not ANSI compliant. That was OK as it was what it
> was. The problem was the first few releases as the complier claimed
> ANSI compatibility. Let's just say those were interesting times. And

That's unkind, to say the least. I have a friend who used to work
at Microware.

> one reason we have moved from OS/9 to Gentoo Linux in these systems.
> We are now sorting out a few responsiveness issues, with serious
> plans to add RTAI (www.rtai.org) to Gentoo.

Blech. RTAI? Why use a dual-kernel approach when you can get there with
one? <employer_flogging>TimeSys (http://www.timesys.com/) seels 
single-kernel real-time Linux.</employer_flogging>

Kurt