From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Thu, 14 Nov 2002 07:01:59 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Thu, 14 Nov 2002 07:01:59 -0500 Received: from port-213-148-149-130.reverse.qsc.de ([213.148.149.130]:24467 "EHLO eumucln01.mscsoftware.com") by vger.kernel.org with ESMTP id ; Thu, 14 Nov 2002 07:01:57 -0500 From: Martin Knoblauch Reply-To: martin.knoblauch@mscsoftware.com Organization: MSC.Software GmbH To: linux-kernel@vger.kernel.org Subject: [Patch] Prevent crash when loading "cpqfc" with Tachyon XL2 cards (against 2.4.20-rc1-ac1) Date: Thu, 14 Nov 2002 13:07:49 +0100 User-Agent: KMail/1.4.2 Cc: fibrechannel@compaq.com, ehm@cris.com, grif@cs.ucr.edu MIME-Version: 1.0 Message-Id: <200211141307.49206.martin.knoblauch@mscsoftware.com> X-AntiVirus: OK! AntiVir MailGate Version 2.0.1.2; AVE: 6.16.0.0; VDF: 6.16.0.17 at mailmuc has not found any known virus in this email. X-MIMETrack: Itemize by SMTP Server on EUMUCLN01/MSCsoftware(Release 5.0.10 |March 22, 2002) at 11/14/2002 01:04:15 PM, Serialize by Router on EUMUCLN01/MSCsoftware(Release 5.0.10 |March 22, 2002) at 11/14/2002 01:04:24 PM, Serialize complete at 11/14/2002 01:04:24 PM Content-Type: Multipart/Mixed; boundary="------------Boundary-00=_11FKYUZU8UVFVF8XEIJW" Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org --------------Boundary-00=_11FKYUZU8UVFVF8XEIJW Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="us-ascii" Hi, please apply the appended patch to "drivers/scsi/cpqfcTSinit.c". It=20 adds a call to "pci_enable_device" which prevents a hard crash when=20 loading the driver on a Tachyon XL2 based HP card (IA64 platform). The=20 patch is against 2.4.20-rc1-ac1, but should apply to mainline. In general, is someone working making the Tachyon XL2 cards really work=20 with the cpqfc driver? I am very interested in that. It is really sad=20 to see those 60 FC disks idle when Linux is booted on our rx5670 :-( I=20 would love to help out with testing and experimenting. The current problems (after applying this patch are): - unable to read WWN from NVRAM - no GBIC found - no link state - no disks found Thanks Martin --=20 Martin Knoblauch Senior System Architect MSC.software GmbH Am Moosfeld 13 D-81829 Muenchen, Germany e-mail: martin.knoblauch@mscsoftware.com http://www.mscsoftware.com Phone/Fax: +49-89-431987-189 / -7189 Mobile: +49-174-3069245 --------------Boundary-00=_11FKYUZU8UVFVF8XEIJW Content-Transfer-Encoding: 7bit Content-Type: text/x-diff; charset="us-ascii"; name="cpqfcTSinit.c.patch" Content-Disposition: attachment; filename="cpqfcTSinit.c.patch" --- cpqfcTSinit.c-orig Wed Nov 13 11:05:10 2002 +++ cpqfcTSinit.c Thu Nov 14 12:48:48 2002 @@ -242,6 +242,10 @@ while ((PciDev = pci_find_device(cpqfc_boards[i].vendor_id, cpqfc_boards[i].device_id, PciDev))) { + if (pci_enable_device(PciDev) != 0) { + printk(KERN_WARNING "cpqfc: pci_enable_devive failed, skipping.\n"); + continue; + } if (pci_set_dma_mask(PciDev, CPQFCTS_DMA_MASK) != 0) { printk(KERN_WARNING "cpqfc: HBA cannot support required DMA mask, skipping.\n"); continue; @@ -411,6 +415,7 @@ // can we find an FC device mapping to this SCSI target? DumCmnd.channel = ScsiDev->channel; // For searching DumCmnd.target = ScsiDev->id; + DumCmnd.lun = ScsiDev->lun; pLoggedInPort = fcFindLoggedInPort(fcChip, &DumCmnd, // search Scsi Nexus 0, // DON'T search linked list for FC port id NULL, // DON'T search linked list for FC WWN --------------Boundary-00=_11FKYUZU8UVFVF8XEIJW--