Re: ORCA/C

[mikew50@aol.com (MikeW50)]

> Anyway, my biggest problem with Orca C wasn't ever the small deviations
> from ANSI ...

I guess I missed the original thread because I sort of tuned out on yet another
defense of indefensable piracy.

Surprisingly enough, ORCA/C has actually been updated relatively recently.
While I have nothing planned in the near future, I won't rule out another
upgrade. For that reason as well as personal edification, I'm curious about
your comments.

First, the obvious: There are some bugs in ORCA/C. All of the ones I'm aware of
are relatively minor in the sense that they are both obvious and easy to get
around. Some of these are in code generation for the loop invariant removal and
common subexpression elimination optimizations. While I don't have a
photographic memory, I do not remember any problems with code generation for
function pointers in the most recent release. If you know of some, I'd like to
know. If you're just repeating old rumors... well, get an update and try for
yourself.

With some documented exceptions in the area of the libraries, though, and a
very small number of bugs, ORCA/C is ANSI C as it was documented and understood
when ORCA/C was still being actively developed. There are some occasional
discussions about some of the finer points of the standard--including one
recent one where different people contacting people on the standards committee
got different answers--but when you get to that level of discussion, you're
well beyond the compliance of any other C compilers I've ever used, then or
now. :) Still, if you know of problems, I'd be interested in hearing about
them. I won't promise to do anything at this late date, but I am curious.

Mike Westerfield