Subject: Shared lib weirdness on Sep.27 mac68k
To: None <current-users@NetBSD.ORG, port-mac68k@NetBSD.ORG>
From: Ken Nakata <kenn@remus.rutgers.edu>
List: current-users
Date: 09/29/1995 10:58:29
I clean-built a GENERIC kernel current as of 27th, and it fails to run
any dynamically linked binaries.  Ktrace shows all failing binaries
get SIGSEGV after closing shared libc file, but it tells me nothing
about what the failing program was doing between closing shared libc
and getting signal.

So, I picked hexdump (for no particular reason), compiled it with -g
flag, and tried it under the new kernel.  It dumped a core just like
before, but later when I tried to see what was going on with gdb on a
good kernel (9/23), gdb refused to load the core complaining it's not
a core file... (file, OTOH, says the said core file is a netbsd-core
file)

Does anyone have any idea what might be going on?  This is the mac68k
port I'm using.  I need to compile a working kernel in order to test
my floating point emulator...

TIA,

ken n.
--
        Ken Nakata        | An unofficial MacBSD anon-ftp maintainer at
                          | ftp://remus.rutgers.edu/pub/NetBSD.  See also
    Rutgers University    | ftp://remus.rutgers.edu/pub/NetBSD/index.html.
New Brunswick, New Jersey | Home Page: http://remus.rutgers.edu/~kenn/