On Mon, 07 May 2018 06:00:56 +0100, Bockholdt Arne wrote: > > > > On Thu, 2018-05-03 at 10:41 +0100, Marc Zyngier wrote: > > I'm talking about making the whole workaround dependent on the USB > > controller being behind an iommu. No iommu, no workaround (because it > > is > > likely that there is no problem in that case). > > > My server doesn't have an IOMMU at all and is affected by the bug. I'm > the original reporter of the problem. No, you are just reporting a regression caused by the workaround for the original issue, which has to do with transitioning from 64bit addresses to 32bit addresses. Resting the controller solves that problem, but generates other issues (yours and a few others). The current approach is to drop the mandatory reset and to use a different sequence, only if the device is behind an IOMMU. So in your case, we won't do anything at all, which is what seem to work for most people. Only a subset of the platforms that would be affected will have another workaround. Machines that don't have an IOMMU *and* perform an 64/32 address range transition may see memory corruption or other side effects, but that's nothing new (the HW looks terminally broken). Hope this makes it clear. M. -- Jazz is not dead, it just smell funny. -- To unsubscribe from this list: send the line "unsubscribe linux-usb" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
