Re: Compiler bugs

["Martin v. Loewis" <martin at loewis dot home dot cs dot tu-berlin dot de>]

> What are the guys at gnu thinking about this? Maybe I should tell them?
> Interested?

There are not really the "guys at gnu" which would be interested as a
group. Instead, individual developers might be interested or
desinterested. I personally agree that a debugger is a very important
tool, and I couldn't live (rather, work) without one.

I don't need a debugger for Windows, though, because I use gdb on
Solaris and Linux most of my time, and it works fine. If I happen to
compile on Windows, I usually use VC++, which also comes with a very
good debugger (IMHO). So I'd be interested in your project only at an
abstract level.

> And gdb?
> Well, I suspect that there isn't any development there.

I'm not sure this is entirely true, see

http://sourceware.cygnus.com/gdb/

As a freeware project, there is as much development as there are
people willing to put effort into it. I think it is unfortunate if the
debugger has deficiencies on Windows, but then, I don't need to use it
on Windows. On Unix, it is the best thing I've ever seen. It also
integrates with Emacs nicely.

> My debugger tries to find out what the program is doing and show
> automatically the changes to the program environment.
>
> This means less typing for the debugger user, the programmer.
> 
> I would propose this improvement to gdb.

Please explain what you mean by "program environment". Do you think a
compiler should track calls to setenv(3)? I can see that might be
useful in some cases, but I never had a need for that so far, and I
think it would be easy just to set a breakpoint on setenv. Or do you
mean something else?

Anyway, I think it is good to have choices, so having another debugger
for Windows is a good thing - especially if it works better than gdb.
Please consider releasing it as free software.

Regards,
Martin