Date: Tue, 15 Jun 2004 19:55:48 +0800 From: Xin LI <delphij@frontfree.net> To: astesin@ukrtelecom.net Cc: current@FreeBSD.org Subject: Re: HA: [backtrace] kernel dump when shutting down Message-ID: <20040615115548.GA1250@frontfree.net> In-Reply-To: <1152675CA9EDD71187130002B3CE5ADA08A6F4F7@hoexc010.ukrtelecom.net> References: <1152675CA9EDD71187130002B3CE5ADA08A6F4F7@hoexc010.ukrtelecom.net>
next in thread | previous in thread | raw e-mail | index | archive | help
--ibTvN161/egqYuK8 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Tue, Jun 15, 2004 at 12:22:34PM +0300, astesin@ukrtelecom.net wrote: > Hello Xin, >=20 > > I got a strange problem when having HTT enabled on a P4 2.8E=20 > > processor. When rebooting the system, after the disk sync and=20 > > the system is going to restart, it crashes and generates a=20 > > crashdump. Strangely, the panic message seems to be corrupted. >=20 > That's exactly what I'm seeing on P4 3.0E (yes, also `E') and Intel > S875WP1-E motherboard > ftp://download.intel.com/support/motherboards/server/s875wp1-e/tps.pdf , = but > it even doesn't do crashdump, just freeses. And not each time but maybe 3= -4 > of 5 reboots. BTW running `halt' instead of reboot worked. Sorry I didn't > had time this night to try a kernel with debugging. Ah... Yes, I'm using Intel motherboard too, however, with a different chipset. Are you sure that the system is actually frozen? In my case, nothing appeared in console, however, pressing Ctrl+C will reboot immediately, which is the behavior when system is dumping core. In addition, I have successfully obtained crashdump. The backtrace itself seemed to be meaningless. I thought there might be some locking problem within the shutdown code. I will do some experiments after my competitive examination for bachelor's degree tomorrow. > > The system used to crash before sync'ing disks. It seems that=20 > > Nate's recent commit has removed that crash. More information=20 > > available upon request. >=20 > Exactly. 5.2.1-R and -CURRENT of late May were also tested, they were > crashing before syncing discs, 20040614 -CURRENT crashes (freeses) after > syncing. >=20 > People say that on older P4 CPUs (Northwood core) that do not have this > problem, only on `E'-s (??? Strange enough). Exactly. The P4 box at my lab does work correctly. Let's check the developers' manual from Intel and we may find something useful :-) Cheers, --=20 Xin LI <delphij frontfree net> http://www.delphij.net/ See complete headers for GPG key and other information. --ibTvN161/egqYuK8 Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) iD8DBQFAzuPEOfuToMruuMARAvMUAKCLTzTZUY0CYr8NglinRXLz5upjewCfbRC9 l6DMj0Ewag/huAFnTbsl8DA= =MxtJ -----END PGP SIGNATURE----- --ibTvN161/egqYuK8--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20040615115548.GA1250>