On Mon, 28 Apr 2008, Tore Anderson wrote: > * Andrew Vasquez > > > Hmm, MSI is enabled: > > Indeed - seems to be the default behaviour. > > > could you try disabling MSI via 'pci=nomsi' (I believe), we've dealt > > with a large number of problem reports where customers reported 'odd' > > behaviours (no interrupt routining) with several motherboard > > chipsets. At least it could be another useful datapoint... > > The system has now been up for 58 hours straight with MSI disabled per > your suggestion. This is almost a new record, usually it has crashed > once per day or more. I won't open the champagne before a week more has > passed, but this looks _very_ promising! That's good news... > Thank you very much for your help in tracking this down! Be sure to let > me know if you want any more information and I'll see what I can do. > The system isn't yet back in full production, so I have a small window > of opportunity to reproduce the problem, test patches, and so on. In general, most MSI/MSI-X machine/chipset quirks (global/bus-level disabling) are handled by the PCI layer. I'd be curious to see if any of the 30+ updates to drivers/pci/quirk.c since 2.6.22 have addressed this issue via DECLARE_PCI_FIXUP_FINAL()??? Could you send the output of 'lspci -vvx'? -- To unsubscribe from this list: 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
