Hi Joerg, The method I described is not making any guess, we need a table to write down the correct ioapic value for those machines, we can add the BIOS version check(DMI_BIOS_VERSION) if you think the behavior may change with different version of BIOSes.
If you still feel uncomfortable, I'm okay with any solutions, but I don't get it how to do it in your way. We can found those machines from bug reports(I have some machines on my hand, too), so we need a table to record those machines? And how to disable interrupt remapping, will it lead to any side effect? Best regards, AceLan Kao. 2015-05-05 23:13 GMT+08:00 Joerg Roedel <j...@8bytes.org>: > Hi, > > On Tue, May 05, 2015 at 03:30:41PM +0800, AceLan Kao wrote: >> There is no easy way to check whether this BIOS bug is present, so we >> have to list them explicitly. >> I think we can do just like what platform drivers(ex. asus_quirks[1]) >> do, check the system vendor and product name to identify those >> machines, and give them a working ioapic value. > > Problem here is that system vendor and platform name is not enough, the > version of the BIOS is also important. To make things worse, the problem > could be introduced into one BIOS version and fixes in another later > version. > > My feeling is that trying to make guesses here will open a can of > worms we better keep closed. The strategy for now has been to search for > broken configurations and disable interrupt remapping when they are > detected. > > > Joerg > _______________________________________________ iommu mailing list iommu@lists.linux-foundation.org https://lists.linuxfoundation.org/mailman/listinfo/iommu
