Date: Mon, 19 Jul 2004 10:13:07 -0700 (PDT) From: Doug White <dwhite@gumbysoft.com> To: "Bjoern A. Zeeb" <bzeeb-lists@lists.zabbadoz.net> Cc: FreeBSD current mailing list <freebsd-current@freebsd.org> Subject: Re: kldunload if_sk -> LOR & panic Message-ID: <20040719100952.H59702@carver.gumbysoft.com> In-Reply-To: <Pine.BSF.4.53.0407181955250.43309@e0-0.zab2.int.zabbadoz.net> References: <Pine.BSF.4.53.0407181443330.43309@e0-0.zab2.int.zabbadoz.net> <Pine.BSF.4.53.0407181955250.43309@e0-0.zab2.int.zabbadoz.net>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sun, 18 Jul 2004, Bjoern A. Zeeb wrote: > > At the point where it's actually detaching, once it's detached the methods > > that access its softc, it should drop its softc lock before calling the > > network-level detach routines. > > I think this is handled by dwhite's patch so I should be safe now. > Perhaps someone could work towards committing a cleaned up version of > this patch ? The LOR is the harmless part. The panic is caused by device_delete_child() recursing from skc to sk, and sk_detach() calls device_delete_child() in itself again after freeing its structures (or something like that). My patch should fix this, if not then I have it in my working sources. -- Doug White | FreeBSD: The Power to Serve dwhite@gumbysoft.com | www.FreeBSD.org
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20040719100952.H59702>