Re: patch to add QNX NTO i386 support

["Kris Warkentin" <kewarken@qnx.com>]

>    > > You'll also need to find a way to distinguish QNX binaries from
other
>    > > OS'es.  The hardware breakpoint stuff will have to be moved to the
>    > > nm.h.  Most of the other stuff shouldn't be necessary or should be
>    > > moved there too.
>    >
>    > Okay.  Not sure how but I'll look into it.  Once I've done that, do I
>    > register an osabi_sniffer to determine the binary?
>
>    I don't think this is going to work.  Our binaries are very generic elf
with
>    no special sections or anything to distinguish them.
>
> In that case you'll have to provide a default OS/ABI for your target.
> Add something like:
>
> *-*-nto*)     gdb_osabi=GDB_OSABI_QNX ;;
>
> to the appropriate case-statement at the end of configure.tgt.

This seems like a good way to do it.  I added GDB_OSABI_QNXNTO to defs.h.
I'm in general using 'nto' since that describes the kernel but people don't
always know that so I made sure to leave the 'qnx' in a few places like here
and config/tm-qnxnto.h.

>    The only way to tell that we're targetting a qnx binary will be when
>    someone types 'target qnx ...'.  Is it acceptable to have the
>    gdbarch stuff come into play when someone does that targetting?
>
> I'm not sure.  If the "qnx" protocol implies that the OS/ABI is QNX,
> this would make sense to me, but there are other people on this list
> that could judge better.

You're right about this.  It's unlikely that anyone else would use our qnx
pdebug protocol for remote debugging but one of my future projects is to
port the gdbserver to Neutrino.  People wouldn't necessarily need to use
pdebug but in this case, there would really be no way of knowing the ABI
other than having a default set.

cheers,

Kris