[comp.sys.nsc.32k] gcc stage3 braindeath.

markm@markm.intel.COM (Mark Mason) (06/15/91)

Hmm, part of the problem was also that gas and gcc did not quite agree on the
form of assembly syntax to use, and then gdb would mis-dis-assembled some
of the addressing modes. Wierd.

Well, as soon as I get my system up (any day now... ;-) I will probably
start looking into it.

BTW: Is there still any interest in a display controller by people out there???
I remember something in passing about someone doing one on the 80960. This
would be of some interest to me (see the line below my e-mail address).

mason
markm@markm.intel.com
80960 Compiler Development Team

jkp@cs.hut.fi (Jyrki Kuoppala) (06/15/91)

Jordan says:
>Quite possible that mind doesn't either, to tell the truth.  Not having
>a DEBUGGER UNDER MINIX (whoa boy, calm down!) makes it difficult to
>ascertain just what is going on..

Oh, by the way, I noticed that setting breakpoints with gdb quite
often crashes the machine.  Strange, since at the beginning (just
after I wrote the ptrace support code) it worked just fine.  But then,
I don't even invalidate the cache so there might be other problems
also.

It's been a long while since I did anything worth mentioning with the
pc532 - recently I compiled make 3.60 & cpio 1.2, no big diffs there.

//Jyrki