On Fri, 24 Feb 2006, Christoph Hellwig wrote: > On Thu, Feb 23, 2006 at 11:59:50AM +0000, Alan Cox wrote: > > On Iau, 2006-02-23 at 17:50 +0800, erich wrote: > > > But unfortunately I found some mainboards will hang up if I always enable > > > this function in my lab. > > > To avoid this issue, I do an option for this case. > > > > > > But Christoph Hellwig give me comment with it. > > > > > > Another thing you can also do for many of these cases is to use either > > the PCI or DMI interfaces to identify the problem board and > > automatically set the option as well. > > > > There are two ways to do this. One is > > Please avoid that unless really nessecary. I doubt there's boards where > MSI would only be broken with the areca card but not with other MSI-capable > ones. If a board or chipset is generally broken vs MSI it should be > added to the global MSI blacklist. It's probably be nice to have a global > nomsi boot option instead of one in every driver aswell.. Jeff G. added an "msi" option to the sata_mv driver recently. But yes, I expect it to be more of a platform issue than a driver issue. s2io network people report MSI interrupt problems on various platforms (on netdev mailing list). There are other reports (that I have at home, not visible to me now). It would be good to have an MSI expert around. http://www.xenotime.net/linux/patches/pci_nomsi.patch adds a global boot option to disable MSI interrupt assignments. -- ~Randy - : send the line "unsubscribe linux-scsi" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
