[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] Re: [Xen-devel] AMD GPU passthrough in Xen
On 22 September 2014 16:05:41 BST, Gordan Bobic <gordan@xxxxxxxxxx> wrote: >On 09/22/2014 03:31 PM, Peter Kay wrote: >Do you happen to know if the recent KVM/QEMU provides a feature to >limit >the amount of memory below 4GB to assign to the guest? Or provide a >custom e820 map? To work around a hardware bug I have to somehow make >sure that everything between 2.5GB and 4GB is reserved and untouchable. KVM/Qemu, I haven't done it myself, but Qemu supports the -smbios parameter to read ACPI tables. These may have to be packed. Both Qemu and KVM seem to have APIs to modify e820. It's probably worth trying without any modifications on the latest Linux kernel, as KVM is (to my mind) best thought of as a KVM accelerator and VFIO does all the neat stuff (I.e. qemu can be run with passthrough without kvm acceleration). It might just work... I see xc_reserved_device_memory_map has recently been submitted for xen? It's a very new set of patches though and I can't seem to see it in the source tree. It's slated for inclusion in 4.5 release, written by Tiejun Chen PK -- Sent from my Android device with K-9 Mail. Please excuse my brevity. _______________________________________________ Xen-devel mailing list Xen-devel@xxxxxxxxxxxxx http://lists.xen.org/xen-devel
|
Lists.xenproject.org is hosted with RackSpace, monitoring our |