Date: Mon, 10 Jul 2006 19:04:40 +0200 From: Fabian Keil <freebsd-listen@fabiankeil.de> To: John Baldwin <jhb@freebsd.org> Cc: freebsd-acpi@freebsd.org Subject: Re: Selectively disabling acpi sub-systems Message-ID: <20060710190440.7632135b@localhost> In-Reply-To: <200607100938.10108.jhb@freebsd.org> References: <20060710143810.61705f74@localhost> <200607100938.10108.jhb@freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
--Sig_PI/0QyBgQocsOlDfIy9bt8H Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: quoted-printable John Baldwin <jhb@freebsd.org> wrote: > On Monday 10 July 2006 08:38, Fabian Keil wrote: > > As described in: > > http://lists.freebsd.org/pipermail/freebsd-stable/2006-June/026554.html > > I'm currently experiencing strange hangs or crashes with a FreeBSD > > RELENG_6 system running Tor. > > > > ... > >=20 > > Could someone please tell me how to stop acpi0 from > > grabbing sio0, without affecting fxp0? >=20 > You most likely can't. You can't disable ACPI from managing devices > and then expect it to manage PCI interrupts. Thanks for the clarification, I misunderstood the man page then. =20 > As far as your actual panic. I've looked at the PR's, etc. and there > are few if any details. For example, what is a "hard crash?" Does > the machine turn off? Is it locked up such that it won't respond to > pings? In general, even if it's deadlocked you can break into DDB via > a serial console (trust me, I've seen this far too many times at my > day job in the past few weeks). I don't have physical access to the system and have no way to tell if it turned it self off or just hangs. All I can tell is that it suddenly stops responding on the network and on the serial console, which makes manually entering ddb impossible. If the system is running the serial console is working as expected: I can enter ddb with <cr> ~ ^b, sysctl debug.kdb.enter=3D1 (or by typing kldload /dev/mem instead of mount_procfs ...). Another sio related thing where I'm not sure if it means anything: according to top -S, "swi4: clock sio" hardly ever leaves the state *Giant. top's output is logged every five minute and at the moment I have: [fk@tor ~]$ sudo grep "clock sio" /var/log/stats.log.bak | wc -l 1022 [fk@tor ~]$ sudo grep "clock sio" /var/log/stats.log.bak | grep Giant | wc = -l 1006 [fk@tor ~]$ sudo grep "clock sio" /var/log/stats.log.bak | grep WAIT | wc -l 16 Fabian --=20 http://www.fabiankeil.de/ --Sig_PI/0QyBgQocsOlDfIy9bt8H Content-Type: application/pgp-signature; name=signature.asc Content-Disposition: attachment; filename=signature.asc -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.4 (FreeBSD) iD8DBQFEsoitjV8GA4rMKUQRAt8BAJ46WNp5ueYtFGmDE9Sr3ytLRM2AKACgp6r5 UenWaQjSW2oZ0yw/RGJgIqQ= =YniF -----END PGP SIGNATURE----- --Sig_PI/0QyBgQocsOlDfIy9bt8H--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20060710190440.7632135b>