This patch-set includes various bug fixes.
The most notable bug is the wrong decoding of guest instructions which cross
page boundary. I am surprised it does not happen in real-systems.
Please review carefully the changes the DR4/5 bug-fix introduces to SVM part.
Since I do not have AMD machine I could not check the patch. Anyhow, I suspect
SVM dr-interception is buggy since in certain cases instructions are skipped
while fault is injected.
Thanks for reviewing the patches.
Nadav Amit (5):
KVM: x86: Clear DR7.LE during task-switch
KVM: x86: Emulator performs code segment checks on read access
KVM: x86: Decoding guest instructions which cross page boundary may
fail
KVM: vmx: Unavailable DR4/5 is checked before CPL
KVM: x86: Using TSC deadline may cause multiple interrupts by user
writes
arch/x86/include/asm/kvm_host.h | 1 +
arch/x86/kvm/emulate.c | 10 +++++----
arch/x86/kvm/lapic.c | 7 +++++-
arch/x86/kvm/lapic.h | 3 ++-
arch/x86/kvm/svm.c | 10 +++++----
arch/x86/kvm/vmx.c | 19 ++++++++++------
arch/x86/kvm/x86.c | 48 ++++++++++++++++-------------------------
7 files changed, 51 insertions(+), 47 deletions(-)
--
1.9.1
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
