KVM: X86: always reset st->preempted in record_steal_time()
PATCH 1/1 KVM: X86: always reset st->preempted in record_steal_time()
From: Lai Jiangshan <[email protected]>
st->preempted needs to be reset in record_steal_time() to clear the KVM_VCPU_PREEMPTED bit.
But the commit 66570e966dd9 ("kvm: x86: only provide PV features if enabled in guest's CPUID") made it cleared conditionally and KVM_VCPU_PREEMPTED might not be cleared when entering into the guest.
Also make st->preempted be only read once, so that trace_kvm_pv_tlb_flush() and kvm_vcpu_flush_tlb_guest() is consistent with same value of st->preempted.
--- arch/x86/kvm/x86.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-)
arch/x86/kvm/x86.c +6 -2
@@ -3101,10 +3101,14 @@ static void record_steal_time(struct kvm_vcpu *vcpu)* expensive IPIs. */ if (guest_pv_has (vcpu,KVM_FEATURE_PV_TLB_FLUSH )) { +u8 st_preempted =xchg (&st->preempted,0 ); +trace_kvm_pv_tlb_flush (vcpu->vcpu_id, - st->preempted &KVM_VCPU_FLUSH_TLB ); -if (xchg (&st->preempted,0 ) &KVM_VCPU_FLUSH_TLB ) + st_preempted &KVM_VCPU_FLUSH_TLB ); +if (st_preempted &KVM_VCPU_FLUSH_TLB )kvm_vcpu_flush_tlb_guest (vcpu); + }else { + st->preempted =0 ; } vcpu->arch.st.preempted =0 ;
Signature · 1 line
2.19.1.6.gb485710b
Quoted 10 lines · Lai Jiangshan
From: Lai Jiangshan <[email protected]>
st->preemptedneeds to be reset inrecord_steal_time()to clear theKVM_VCPU_PREEMPTEDbit.But the commit
66570e966dd9("kvm: x86: only provide PV features if enabled in guest's CPUID") made it cleared conditionally andKVM_VCPU_PREEMPTEDmight not be cleared when entering into the guest.
Quoted 5 lines
Also make
st->preemptedbe only read once, so thattrace_kvm_pv_tlb_flush()andkvm_vcpu_flush_tlb_guest()is consistent with same value ofst->preempted.Cc Oliver Upton <[email protected]>Fixes 66570e966dd9 ("kvm: x86: only provide PV features if enabled in guest's CPUID")
It has already been fixed by commit 1eff0ada88b48 (KVM: X86: Fix vCPU preempted state from guest's point of view) in kvm/master.
Wanpeng
PATCH 1/1 KVM: X86: reset and read st->preempted in atomic way
From: Lai Jiangshan <[email protected]>
In record_steal_time(), st->preempted is read twice, and trace_kvm_pv_tlb_flush() might output result inconsistent if kvm_vcpu_flush_tlb_guest() see a different st->preempted later.
It is a very trivial problem and hardly has actual harm and can be avoided by reseting and reading st->preempted in atomic way via xchg().
arch/x86/kvm/x86.c +4 -2
@@ -3117,9 +3117,11 @@ static void record_steal_time(struct kvm_vcpu *vcpu)* expensive IPIs. */ if (guest_pv_has (vcpu,KVM_FEATURE_PV_TLB_FLUSH )) { +u8 st_preempted =xchg (&st->preempted,0 ); +trace_kvm_pv_tlb_flush (vcpu->vcpu_id, - st->preempted &KVM_VCPU_FLUSH_TLB ); -if (xchg (&st->preempted,0 ) &KVM_VCPU_FLUSH_TLB ) + st_preempted &KVM_VCPU_FLUSH_TLB ); +if (st_preempted &KVM_VCPU_FLUSH_TLB )kvm_vcpu_flush_tlb_guest (vcpu); }else { st->preempted =0 ;
Signature · 1 line
2.19.1.6.gb485710b
Re: [PATCH] KVM: X86: reset and read st->preempted in atomic way
Nit: the shortlog is somewhat inaccurate now, maybe just:
KVM: x86: Ensure PV TLB flush tracepoint reflects KVM behavior
or something along those lines. Not sure what the best wording is :-/
Quoted 10 lines · Lai Jiangshan
From: Lai Jiangshan <[email protected]>
In
record_steal_time(),st->preemptedis read twice, andtrace_kvm_pv_tlb_flush()might output result inconsistent ifkvm_vcpu_flush_tlb_guest()see a differentst->preemptedlater.It is a very trivial problem and hardly has actual harm and can be avoided by reseting and reading
st->preemptedin atomic way viaxchg().Signed-off-by Lai Jiangshan <[email protected]>
I saw this quirk too, but couldn't quite bring myself to care enought to test a patch :-)