search for: rmrrs

...ge please find attached the three files as requested. > I have used > > iommu=soft > > in the grub command line for the kernel in the domU as explained before. > many thanks (Please reply in-line, like this, rather than top-posting.) Thanks -- as I suspected, your USB device has RMRRs, which are what's causing the problem. In this case, your USB controller's RMRRs are shared with another device. Theoretically, having two devices with the same RMRRs assigned to different domains could be a security issue, so Xen by default refuses to do it. The options are: 1. Figure o...

...om0. The parameter is "iommu_include_reserved" and is off by default. A warning is also traced when incorrect RMRR to system memory map values are detected. This is being added to address some incorrect BIOS''s that do not report correctly the requied reserved memory ranges in the RMRRs. When this occurs it currently can cause early boot hangs and crashes. Signed-off-by: Ross Philipson <ross.philipson@citrix.com> Based on changeset 19238 Thanks Ross Ross Philipson Senior Software Engineer Citrix Systems, Inc 14 Crosby Drive Bedford, MA 01730 781-301-7949 ross.philipson@...

Dear George please find attached the three files as requested. I have used iommu=soft in the grub command line for the kernel in the domU as explained before. many thanks Francis From: "George Dunlap" <dunlapg at umich.edu> To: "francis" <francis at choughs.net>, "centos-virt" <centos-virt at centos.org> Sent: Monday, 16 May, 2016 10:29:09

...Linux 4.1. Data corruption in the guest. We could make the rule that *all* virtio-pci devices (except on PPC) respect the bus rules. We'd have to fix QEMU so that virtio devices on Q35 iommu=on systems set up a PCI topology where the devices *aren't* behind the IOMMU or are protected by RMRRs or whatever. Then old kernels would work correctly on new hosts, new kernels would work correctly except on old iommu-providing hosts, and Xen would work. In fact, on Xen, it's impossible without colossal hacks to support non-IOMMU-respecting virtio devices because Xen acts as an intermediate...

...Linux 4.1. Data corruption in the guest. We could make the rule that *all* virtio-pci devices (except on PPC) respect the bus rules. We'd have to fix QEMU so that virtio devices on Q35 iommu=on systems set up a PCI topology where the devices *aren't* behind the IOMMU or are protected by RMRRs or whatever. Then old kernels would work correctly on new hosts, new kernels would work correctly except on old iommu-providing hosts, and Xen would work. In fact, on Xen, it's impossible without colossal hacks to support non-IOMMU-respecting virtio devices because Xen acts as an intermediate...

...nclusive_mapping=1" and is off by default. In addition to the parameter, a warning is also traced when incorrect RMRR to system memory map values are detected. These are being added to address some incorrect BIOS''s that do not report correctly the requied reserved memory ranges in the RMRRs. When this occurs it currently can cause early boot hangs and crashes. Signed-off-by: Ross Philipson <ross.philipson@citrix.com<mailto:ross.philipson@citrix.com>> Based on changeset 19243 Thanks Ross Ross Philipson Senior Software Engineer Citrix Systems, Inc 14 Crosby Drive Bedfor...

...gt;> >> We could make the rule that *all* virtio-pci devices (except on PPC) >> respect the bus rules. We'd have to fix QEMU so that virtio devices >> on Q35 iommu=on systems set up a PCI topology where the devices >> *aren't* behind the IOMMU or are protected by RMRRs or whatever. Then >> old kernels would work correctly on new hosts, new kernels would work >> correctly except on old iommu-providing hosts, and Xen would work. > > I don't see a point in doing anything about old QEMU with IOMMU enabled > and virtio devices plugged except...

...gt;> >> We could make the rule that *all* virtio-pci devices (except on PPC) >> respect the bus rules. We'd have to fix QEMU so that virtio devices >> on Q35 iommu=on systems set up a PCI topology where the devices >> *aren't* behind the IOMMU or are protected by RMRRs or whatever. Then >> old kernels would work correctly on new hosts, new kernels would work >> correctly except on old iommu-providing hosts, and Xen would work. > > I don't see a point in doing anything about old QEMU with IOMMU enabled > and virtio devices plugged except...

...ed by the guest. > > We could make the rule that *all* virtio-pci devices (except on PPC) > respect the bus rules. We'd have to fix QEMU so that virtio devices > on Q35 iommu=on systems set up a PCI topology where the devices > *aren't* behind the IOMMU or are protected by RMRRs or whatever. Then > old kernels would work correctly on new hosts, new kernels would work > correctly except on old iommu-providing hosts, and Xen would work. I don't see a point in doing anything about old QEMU with IOMMU enabled and virtio devices plugged except declaring such setups...

...> We could make the rule that *all* virtio-pci devices (except on PPC) >>> respect the bus rules. We'd have to fix QEMU so that virtio devices >>> on Q35 iommu=on systems set up a PCI topology where the devices >>> *aren't* behind the IOMMU or are protected by RMRRs or whatever. Then >>> old kernels would work correctly on new hosts, new kernels would work >>> correctly except on old iommu-providing hosts, and Xen would work. >> >> I don't see a point in doing anything about old QEMU with IOMMU enabled >> and virtio dev...

On Tue, Jul 28, 2015 at 9:44 AM, Jan Kiszka <jan.kiszka at siemens.com> wrote: > The ability to have virtio on systems with IOMMU in place makes testing > much more efficient for us. Ideally, we would have it in non-identity > mapping scenarios as well, e.g. to start secondary Linux instances in > the test VMs, giving them their own virtio devices. And we will > eventually

On Tue, Jul 28, 2015 at 9:44 AM, Jan Kiszka <jan.kiszka at siemens.com> wrote: > The ability to have virtio on systems with IOMMU in place makes testing > much more efficient for us. Ideally, we would have it in non-identity > mapping scenarios as well, e.g. to start secondary Linux instances in > the test VMs, giving them their own virtio devices. And we will > eventually

...lich <jbeulich@suse.com> date: Thu Sep 22 18:32:34 2011 +0100 VT-d: fix off-by-one error in RMRR validation (base_addr,end_addr) is an inclusive range, and hence there shouldn''t be a subtraction of 1 in the second invocation of page_is_ram_type(). For RMRRs covering a single page that actually resulted in the immediately preceding page to get checked (which could have resulted in a false warning). Signed-off-by: Jan Beulich <jbeulich@suse.com> changeset: 23867:571b6e90dfb4 user: Jan Beulich <jbeulich@suse.c...