Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Tue, 16 Oct 2001 15:22:00 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Tue, 16 Oct 2001 15:21:50 -0400 Received: from postfix1-2.free.fr ([213.228.0.130]:33544 "HELO postfix1-2.free.fr") by vger.kernel.org with SMTP id convert rfc822-to-8bit; Tue, 16 Oct 2001 15:21:41 -0400 Date: Tue, 16 Oct 2001 21:16:30 +0200 (CEST) From: =?ISO-8859-1?Q?G=E9rard_Roudier?= X-X-Sender: To: "David S. Miller" Cc: , , Linus Torvalds Subject: Re: Oops while inserting sym53c8xx.o In-Reply-To: <20011016.022902.74752070.davem@redhat.com> Message-ID: <20011016205934.M356-100000@gerard> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=ISO-8859-1 Content-Transfer-Encoding: 8BIT Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Hi David, I missed this one, but, btw, it wasn't from me as you know. :-) OTOH, I use sym-2 instead since months. :) Btw, sym-2 has a new PCI 64 bit DMA addressing mode using 16 segment registers. 4GB * 16 = 64 GB. The change is simple and can be back-ported to sym-1. It also support 40 bit addressing, but not both at the same time (compilation optionnable). Note that I would rather prefer to push sym-2 to kernel main-stream. This is planned, but condition was to prove sym-2 in kernel 2.5 first, but kernel 2.5 is very long to come. The new 'up to 64 GB addressing DMA support' has no limitation regarding 64 bit addressing and should better fit existing 64 bit machines. It is obviously untested since, for some obscure reasons, both SUN and COMPAQ missed to offer me a machine suitable for the testings. :-) sym-2.1.15 is available for download from ftp.tux.org. If you have time for giving it a try on some large memory machine, you may let me know the results. Bye for now, G?rard. PS: The enabling of the new 64 GB DMA support requires a simple define changes in the source. You cannot miss it. On Tue, 16 Oct 2001, David S. Miller wrote: > > This should fix it, Linux please apply. > > --- drivers/scsi/sym53c8xx.c.~1~ Mon Oct 8 21:04:56 2001 > +++ drivers/scsi/sym53c8xx.c Tue Oct 16 02:27:44 2001 > @@ -13168,14 +13168,14 @@ > ** in the size field. We use normal 32-bit PCI addresses for > ** descriptors. > */ > - if (chip->features & FE_DAC) { > + if (chip && (chip->features & FE_DAC)) { > if (pci_set_dma_mask(pdev, (u64) 0xffffffffff)) > chip->features &= ~FE_DAC_IN_USE; > else > chip->features |= FE_DAC_IN_USE; > } > > - if (!(chip->features & FE_DAC_IN_USE)) { > + if (chip && !(chip->features & FE_DAC_IN_USE)) { > if (pci_set_dma_mask(pdev, (u64) 0xffffffff)) { > printk(KERN_WARNING NAME53C8XX > "32 BIT PCI BUS DMA ADDRESSING NOT SUPPORTED\n"); > - > To unsubscribe from this list: send the line "unsubscribe linux-kernel" in > the body of a message to majordomo@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html > Please read the FAQ at http://www.tux.org/lkml/ > > - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/