[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <20190729190810.125640760@linuxfoundation.org>
Date: Mon, 29 Jul 2019 21:22:50 +0200
From: Greg Kroah-Hartman <gregkh@...uxfoundation.org>
To: linux-kernel@...r.kernel.org
Cc: Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
stable@...r.kernel.org, Thomas Lambertz <mail@...maslambertz.de>,
anthony <antdev66@...il.com>,
Paolo Bonzini <pbonzini@...hat.com>,
Radim Krčmář <rkrcmar@...hat.com>,
Wanpeng Li <wanpengli@...cent.com>
Subject: [PATCH 5.2 174/215] KVM: X86: Fix fpu state crash in kvm guest
From: Wanpeng Li <wanpengli@...cent.com>
commit e751732486eb3f159089a64d1901992b1357e7cc upstream.
The idea before commit 240c35a37 (which has just been reverted)
was that we have the following FPU states:
userspace (QEMU) guest
---------------------------------------------------------------------------
processor vcpu->arch.guest_fpu
>>> KVM_RUN: kvm_load_guest_fpu
vcpu->arch.user_fpu processor
>>> preempt out
vcpu->arch.user_fpu current->thread.fpu
>>> preempt in
vcpu->arch.user_fpu processor
>>> back to userspace
>>> kvm_put_guest_fpu
processor vcpu->arch.guest_fpu
---------------------------------------------------------------------------
With the new lazy model we want to get the state back to the processor
when schedule in from current->thread.fpu.
Reported-by: Thomas Lambertz <mail@...maslambertz.de>
Reported-by: anthony <antdev66@...il.com>
Tested-by: anthony <antdev66@...il.com>
Cc: Paolo Bonzini <pbonzini@...hat.com>
Cc: Radim Krčmář <rkrcmar@...hat.com>
Cc: Thomas Lambertz <mail@...maslambertz.de>
Cc: anthony <antdev66@...il.com>
Cc: stable@...r.kernel.org
Fixes: 5f409e20b (x86/fpu: Defer FPU state load until return to userspace)
Signed-off-by: Wanpeng Li <wanpengli@...cent.com>
[Add a comment in front of the warning. - Paolo]
Signed-off-by: Paolo Bonzini <pbonzini@...hat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@...uxfoundation.org>
---
arch/x86/kvm/x86.c | 9 ++++++---
1 file changed, 6 insertions(+), 3 deletions(-)
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -3264,6 +3264,10 @@ void kvm_arch_vcpu_load(struct kvm_vcpu
kvm_x86_ops->vcpu_load(vcpu, cpu);
+ fpregs_assert_state_consistent();
+ if (test_thread_flag(TIF_NEED_FPU_LOAD))
+ switch_fpu_return();
+
/* Apply any externally detected TSC adjustments (due to suspend) */
if (unlikely(vcpu->arch.tsc_offset_adjustment)) {
adjust_tsc_offset_host(vcpu, vcpu->arch.tsc_offset_adjustment);
@@ -7955,9 +7959,8 @@ static int vcpu_enter_guest(struct kvm_v
wait_lapic_expire(vcpu);
guest_enter_irqoff();
- fpregs_assert_state_consistent();
- if (test_thread_flag(TIF_NEED_FPU_LOAD))
- switch_fpu_return();
+ /* The preempt notifier should have taken care of the FPU already. */
+ WARN_ON_ONCE(test_thread_flag(TIF_NEED_FPU_LOAD));
if (unlikely(vcpu->arch.switch_db_regs)) {
set_debugreg(0, 7);
Powered by blists - more mailing lists