gdb? HELP ME PLEASE!

rfg@nsc.nsc.com (Ron Guilmette) (06/15/88)

I have just today (finally) gotten a tape in the mail from Michael Tiemann
with a pre-release of g++/gcc (1.22) and related tools on it.

Unfortunately, Michael is now out of the country, and there is one important
thing which seems to have been left off of the tape, probably due to Michael's
rush to leave the country.

The software which I am desperately seeking is alternate code for gcc/g++/gdb
which enable these tools to support & use COFF format binaries.  I specifically
need COFF support for the NS32000 series processors.

Michael told me before he left that someone at MCC had done the code for COFF
support, but he failed to mention any names. Now I am stuck with no contact
to get this stuff from.

If you know who this person is, or if you are this person, PLEASE HELP ME!
I am already the office laughing stock because I bragged 4 weeks ago that
I would get g++ in house and up & running on our System V based systems.
How was I to know that Michael would have hardware problems which would
delay his sending me a tape for three weeks?

If you can help me, please E-mail to the address below, or, better yet,
call me (collect is fine) at 408-721-8172.  The sooner the better.

Ron Guilmette
National SemiConductor
Internet: rfg@nsc.nsc.com
Uucp: ...{pyramid,sun,amdahl,apple}!nsc!rfg
-- 
Ron Guilmette
National SemiConductor
Internet: rfg@nsc.nsc.com
Uucp: ...{pyramid,sun,amdahl,apple}!nsc!rfg

rich@oxtrap (K. Richard Magill) (06/16/88)

In article <5169@nsc.nsc.com>, rfg@nsc (Ron Guilmette) writes:
>The software which I am desperately seeking is alternate code for gcc/g++/gdb
>which enable these tools to support & use COFF format binaries.  I specifically
>need COFF support for the NS32000 series processors.

Doesn't exist.  (to my knowledge and certainly not in a "public" form). gcc is
running on several machines, including multimax, that use coff, but in these
configurations the compiler produces native assembler and thus no gdb support.
(although there appears to be gdb support specific to the multimax).

There was a lot of talk about this recently in the gnu mailing lists and the
bottom line seems to be that the best approach is for gnu ld to produce, on
demand, bsd objects in a COFF wrapper.  To my knowledge, neither this, nor the
COFF to COFF'd bsd translator that would be needed for system libraries, has
been done.

rich.
--