Re: 000217: status of DJGPP support

[Eli Zaretskii <eliz at delorie dot com>]

> >   - There are lots of warnings about comparison of signed with
> >     unsigned and unused arguments.  I find it hard to believe that
> >     this is specific to DJGPP: doesn't anyone else see these warnings?
> >     No doubt they are due to -Wall, but I understand we want to go
> >     into production with these switches, right?
> 
> Not for 5.0.  The only warnings I consider significant for 5.0 are those
> from:
> 
> ,-Wimplicit\
> ,-Wreturn-type\
> ,-Wcomment\
> ,-Wtrigraphs\
> ,-Wformat\
> ,-Wparentheses\
> ,-Wpointer-arith\
> ,-Woverloaded-virtual\

Does 000217 use those or something else?  Sorry, I assumed it was
using -Wall, but I didn't actually check that (and the sources are far
away from the place I'm typing this).  If the snapshot uses the above
options, then the problems I saw happen with them as well.

> I'd suggest:
> 
> 	gdb/config/djgpp/*

I couldn't ask for more ;-).

> Unfortunatly, this also means that if BINUTILS pushes back on a patch,
> GDB can't accept it :-(
> 
> How to solve the situtation you're in? Off hand, I don't know - what
> exactly is the status of those patches?

Ian accepted them a long time ago, but they are not yet in the
Binutils CVS tree, last time I looked (about a week ago).  I asked Ian
once or twice about them, he said they will be installed once he had
time.

It's not very urgent for Binutils, since the next release seems to be
nowhere in sight, but I understand GDB 5.0 *is* going to be released
soon...

The changes are DJGPP- and/or DOS/Windows-specific, they are
conditioned on __MSDOS__ or some such, so they cannot possibly hurt
any other platforms.  Any help in getting these committed will be
highly appreciated ;-)