On Fri, 30 Mar 2007, Christopher D. Quenelle wrote:
What *are* the issues with keeping warlock closed? I was under the
impression it was a Sun developed tool..
I believe that warlock was the basis for the Sun Studio tool called "lock lint",
which is fully supported and available externally to Sun. But I believe it's
not a 100% drop-in replacement. So someone would need to evaluate what's needed
to transition from using warlock to using Lock Lint. I'd be glad to help
with that from the Sun Studio side, but I've never used warlock.
lock_lint and warlock are essentially the same, except that
part of warlock is witten in perl. And that part uses some perlish
features which when repeated in C look awful and work in a strange way.
The only crucial difference between warlock and lock_lint is
the .wlcmd files, though.
I believe that warlock is based on a modified version of the C front end
from Sun Studio, which is not open source (yet).
Warlock is a backend processing .ll files. The .ll files are produced
by our C compiler when given -Zll flag. The flag is present in all the
recent Studio compilers on SPARC. I have a fix for -Zll to be present
on x84/x64, and hope to backport it to Studio11, since it is the version
used for the kernel compilations.
It is also probably based
on an older version of the C front end? So even if we did open source
a new Sun Studio, that wouldn't open up the code underlying warlock.
Aaron Zang is the guy to ask about opensourcing warlock.
Of course, maybe the "warlock" referred to by the bug is referring to some
completely other tool besides the "warlock" that I'm talking about. In which
case, never mind. ;-)
--chris
This message posted from opensolaris.org
_______________________________________________
tools-discuss mailing list
[email protected]
_______________________________________________
tools-discuss mailing list
[email protected]
