boot hang in 2.6.16-rt[7-10]

[Clark Williams <williams@redhat.com>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

I'm seeing a boot hang in the latest -rt series (from -rt7 to current
- -rt10), on an Athlon64x2 (3800+). The kernel loads and boots up to
calibrate_migration_costs then hangs. I'd give some console output,
except that in their infinite wisdom, Gateway decided that this system
didn't need a serial port. Gah...

I've prink'ed my way down into where measure_one tries to migrate a
thread from cpu0 to cpu1 and calls set_cpus_allowed, where it then
tries to wake the migration thread by calling wake_up_process, which
just calls try_to_wake_up. Strangely, I see multiple printks from
try_to_wake_up (like it's running multiple times) and the last print I
see is that I'm returning a 1 from wake_up_process. Nothing after that.

I'd really like to see the boot messages before all those printk's
I've scattered in sched.c, so I'm investigating using the parallel
port as a console and just spewing that output into an xterm on the
other end. Never done it before so I'm bumping my head against the
wall a bit. Any advice would be appreciated.

Clark

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2.2 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iD8DBQFEKay+Hyuj/+TTEp0RAi8YAJ92U1KHM0BW5nWZWl7lCBBQAZWCQQCg1FWy
8dGHlzkPQBn22Y1pK6lOen8=
=Hr9n
-----END PGP SIGNATURE-----

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/