[Bug 216177] kvm-unit-tests vmx has about 60% of failure chance

[bugzilla-daemon@kernel.org]

https://bugzilla.kernel.org/show_bug.cgi?id=216177

--- Comment #11 from Yang Lixiao (lixiao.yang@intel.com) ---
(In reply to Jim Mattson from comment #10)
> On Mon, Jun 27, 2022 at 11:32 PM <bugzilla-daemon@kernel.org> wrote:
> >
> > https://bugzilla.kernel.org/show_bug.cgi?id=216177
> >
> > --- Comment #9 from Yang Lixiao (lixiao.yang@intel.com) ---
> > (In reply to Jim Mattson from comment #8)
> > > On Mon, Jun 27, 2022 at 8:54 PM Nadav Amit <nadav.amit@gmail.com> wrote:
> > >
> > > > The failure on bare-metal that I experienced hints that this is either
> a
> > > test
> > > > bug or (much less likely) a hardware bug. But I do not think it is
> likely
> > > to
> > > > be
> > > > a KVM bug.
> > >
> > > KVM does not use the VMX-preemption timer to virtualize L1's
> > > VMX-preemption timer (and that is why KVM is broken). The KVM bug was
> > > introduced with commit f4124500c2c1 ("KVM: nVMX: Fully emulate
> > > preemption timer"), which uses an L0 CLOCK_MONOTONIC hrtimer to
> > > emulate L1's VMX-preemption timer. There are many reasons that this
> > > cannot possibly work, not the least of which is that the
> > > CLOCK_MONOTONIC timer is subject to time slew.
> > >
> > > Currently, KVM reserves L0's VMX-preemption timer for emulating L1's
> > > APIC timer. Better would be to determine whether L1's APIC timer or
> > > L1's VMX-preemption timer is scheduled to fire first, and use L0's
> > > VMX-preemption timer to trigger a VM-exit on the nearest alarm.
> > > Alternatively, as Sean noted, one could perhaps arrange for the
> > > hrtimer to fire early enough that it won't fire late, but I don't
> > > really think that's a viable solution.
> > >
> > > I can't explain the bare-metal failures, but I will note that the test
> > > assumes the default treatment of SMIs and SMM. The test will likely
> > > fail with the dual-monitor treatment of SMIs and SMM. Aside from the
> > > older CPUs with broken VMX-preemption timers, I don't know of any
> > > relevant errata.
> > >
> > > Of course, it is possible that the test itself is buggy. For the
> > > person who reported bare-metal failures on Ice Lake and Cooper Lake,
> > > how long was the test in VMX non-root mode past the VMX-preemption
> > > timer deadline?
> >
> > On the first Ice lake:
> > Test suite: vmx_preemption_timer_expiry_test
> > FAIL: Last stored guest TSC (28067103426) < TSC deadline (28067086048)
> >
> > On the second Ice lake:
> > Test suite: vmx_preemption_timer_expiry_test
> > FAIL: Last stored guest TSC (27014488614) < TSC deadline (27014469152)
> >
> > On Cooper lake:
> > Test suite: vmx_preemption_timer_expiry_test
> > FAIL: Last stored guest TSC (29030585690) < TSC deadline (29030565024)
> 
> Wow! Those are *huge* overruns. What is the value of MSR 0x9B on these hosts?

All of the values of MSR 0x9B on the three hosts are 0.

-- 
You may reply to this email to add a comment.

You are receiving this mail because:
You are watching the assignee of the bug.