From mboxrd@z Thu Jan 1 00:00:00 1970 From: Mike Christie Subject: Re: SG_IO with >4k buffer size to iscsi sg device causes "Bad page" panic Date: Tue, 08 May 2007 14:30:16 -0500 Message-ID: <4640CFC8.9020004@cs.wisc.edu> References: <0F08E10B769EAF4EA2C43A573B8CC87FB106C0@NAMAIL3.ad.lsil.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Cc: netdev@vger.kernel.org, linux-scsi@vger.kernel.org, linux-iscsi-devel@lists.sourceforge.net, "Qi, Yanling" , dougg@torque.net, James Bottomley To: open-iscsi@googlegroups.com Return-path: In-Reply-To: <0F08E10B769EAF4EA2C43A573B8CC87FB106C0@NAMAIL3.ad.lsil.com> Sender: linux-scsi-owner@vger.kernel.org List-Id: netdev.vger.kernel.org Qi, Yanling wrote: > Hi All, > > This panic is related to the interactions between scsi/sg.c, iscsi > initiator and tcp on the RHEL 2.6.9-42 kernel. But we may also have the > similar problem with open-iscsi initiator. I will explain why we see the Yeah, this problem should occur in the upstream open-iscsi iscsi code. open-iscsi works very similar to linux-scsi where it just sends pages around with sock->ops-sendpage, and it looks like sg uses __get_free_pages in RHEL's kernel and upstream it uses alloc_pages so unless there was a change in those functions or the network layer then we should have a similar problem.