Subject: Re: MPACPI fix (for those who had trouble with it, urgent!)
To: Takayoshi Kochi <kochi@NetBSD.org>
From: John Nemeth <jnemeth@victoria.tc.ca>
List: port-amd64
Date: 05/12/2004 14:30:03
On Aug 19, 11:39pm, Takayoshi Kochi wrote:
}
} I looked into the problem of MPACPI related panics and am trying to
} fix it. If you had any panic with the existing MPACPI code,
} please try the patch attached to this message ASAP, so that
} we can make it for 2.0-release.
I tried it on the weekend. I can report some degree of success.
First I upgraded the machine (ASUS P4P800 MB) to the latest BIOS
(1016). This didn't help. Then I upgraded it from 1.6ZL to 2.0E
(20040509) which didn't help. Finally, I applied your patch. That
worked. The machine is now happily running with two "CPUs"
(hyperthreading actually).
However, USB isn't functional with a multiprocessor kernel. I
have no idea if this is related to the MPAPCI stuff though. When I
plug in a USB device, it complains that there is a problem and it shuts
down the port. If I boot a non-multiprocessor GENERIC kernel, the
device works fine.
Also, I see some rather interesting output from 'ps auwwx':
root 1 0.0 0.0 72 220 ?? Ss 3:23PM 0:-37.33 init
root 4 0.0 10.1 0 158432 ?? DK 3:23PM 0:-1.95 [usb0]
root 6 0.0 10.1 0 158432 ?? DK 3:23PM 0:-1.69 [usb1]
root 7 0.0 10.1 0 158432 ?? DK 3:23PM 0:-1.54 [usb2]
root 8 0.0 10.1 0 158432 ?? DK 3:23PM 0:-1.33 [usb3]
root 596 0.0 0.0 236 508 ?? Is 3:24PM 0:-1.45 /usr/sbin/cron
root 29015 0.0 0.1 1144 1140 p2 I 6:19PM 0:-1.94 -tcsh -m
}-- End of excerpt from Takayoshi Kochi