Re: KVM regression due to commit 28a1f3ac1d0... (Set highest physical address bits in, non-present/reserved SPTEs)

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Mon, 2018-10-08 at 21:40 +0200, tinyusbboard .matrixstorm wrote:
> Hi
> 
> Since commit 28a1f3ac1d0c8558ee4453d9634dad891a6e922e ([PATCH] kvm: x86: Set highest physical address bits in non-present/reserved SPTEs) I have no IO (INPUT and network) on old qemu-VMs using
> -enable-kvm. For example Windows XP-VMs.
> 
> My hardware (core2duo E8500 @ 3.16GHz) is quite old, but worked until this patch was committed.
> 
> Does anyone have the same issues? What is the reason and how to fix this?

It's a known bug in 4.18.8+ kernels, fixed by commit daa07cbc9ae3
("KVM: x86: fix L1TF's MMIO GFN calculation"), which is available
in v4.19-rc7.  I'm not sure when it will propagate to 4.18 kernels,
I assume sooner rather than later.

https://bugzilla.redhat.com/show_bug.cgi?id=1632426
https://patchwork.kernel.org/patch/10614795/



[Index of Archives]     [KVM ARM]     [KVM ia64]     [KVM ppc]     [Virtualization Tools]     [Spice Development]     [Libvirt]     [Libvirt Users]     [Linux USB Devel]     [Linux Audio Users]     [Yosemite Questions]     [Linux Kernel]     [Linux SCSI]     [XFree86]

  Powered by Linux