From mboxrd@z Thu Jan 1 00:00:00 1970 From: Wanpeng Li Subject: Re: [PATCH] KVM: nVMX: Fix try to emulate "Acknowledge interrupt on exit" for interrupt which belongs to L1 Date: Mon, 31 Jul 2017 20:17:47 +0800 Message-ID: References: <1501399902-7325-1-git-send-email-wanpeng.li@hotmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable Cc: "linux-kernel@vger.kernel.org" , kvm , =?UTF-8?B?UmFkaW0gS3LEjW3DocWZ?= , Wanpeng Li To: Paolo Bonzini Return-path: In-Reply-To: Sender: linux-kernel-owner@vger.kernel.org List-Id: kvm.vger.kernel.org 2017-07-31 20:10 GMT+08:00 Paolo Bonzini : > On 30/07/2017 09:31, Wanpeng Li wrote: >> From: Wanpeng Li >> >> ------------[ cut here ]------------ >> WARNING: CPU: 5 PID: 2288 at arch/x86/kvm/vmx.c:11124 nested_vmx_vmexit= +0xd64/0xd70 [kvm_intel] >> CPU: 5 PID: 2288 Comm: qemu-system-x86 Not tainted 4.13.0-rc2+ #7 >> RIP: 0010:nested_vmx_vmexit+0xd64/0xd70 [kvm_intel] >> Call Trace: >> vmx_check_nested_events+0x131/0x1f0 [kvm_intel] >> ? vmx_check_nested_events+0x131/0x1f0 [kvm_intel] >> kvm_arch_vcpu_ioctl_run+0x5dd/0x1be0 [kvm] >> ? vmx_vcpu_load+0x1be/0x220 [kvm_intel] >> ? kvm_arch_vcpu_load+0x62/0x230 [kvm] >> kvm_vcpu_ioctl+0x340/0x700 [kvm] >> ? kvm_vcpu_ioctl+0x340/0x700 [kvm] >> ? __fget+0xfc/0x210 >> do_vfs_ioctl+0xa4/0x6a0 >> ? __fget+0x11d/0x210 >> SyS_ioctl+0x79/0x90 >> do_syscall_64+0x8f/0x750 >> ? trace_hardirqs_on_thunk+0x1a/0x1c >> entry_SYSCALL64_slow_path+0x25/0x25 >> >> This can be reproduced by booting L1 guest w/ 'noapic' grub parameter, w= hich >> means that tells the kernel to not make use of any IOAPICs that may be p= resent >> in the system. >> >> There is a scenario when !kvm_cpu_has_interrupt(vcpu) && external_intr i= n >> vmx_check_nested_events() which means that there is a pending interrupt = from >> L0's qemu device model which should be injected to L1, we should not try= to >> emualte "Acknowledge interrupt on exit" for this interrupt. >> >> This patch fixes it by trying to emulate "Acknowledge interrupt on exit"= if L1 >> asks us to. > > The commit message is not clear; can you rephrase? Actually external_intr here is the req_int_win passed from vcpu_enter_guest() which means that the L0's qemu request an irq window. I observed the secanrio !kvm_cpu_has_interrupt(vcpu) && L0's qemu request an irq window is true. Which means that there is a pending interrupt from L0's qemu device model which should be injected to L1, we should not try to emualte "Acknowledge interrupt on exit" for this interrupt. Regards, Wanpeng Li > >> Cc: Paolo Bonzini >> Cc: Radim Kr=C4=8Dm=C3=A1=C5=99 >> Signed-off-by: Wanpeng Li >> --- >> arch/x86/kvm/vmx.c | 5 +++-- >> 1 file changed, 3 insertions(+), 2 deletions(-) >> >> diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c >> index 2737343..92eb0f75 100644 >> --- a/arch/x86/kvm/vmx.c >> +++ b/arch/x86/kvm/vmx.c >> @@ -11118,8 +11118,9 @@ static void nested_vmx_vmexit(struct kvm_vcpu *v= cpu, u32 exit_reason, >> >> vmx_switch_vmcs(vcpu, &vmx->vmcs01); >> >> - if ((exit_reason =3D=3D EXIT_REASON_EXTERNAL_INTERRUPT) >> - && nested_exit_intr_ack_set(vcpu)) { >> + if (exit_reason =3D=3D EXIT_REASON_EXTERNAL_INTERRUPT && >> + kvm_cpu_has_interrupt(vcpu) && >> + nested_exit_intr_ack_set(vcpu)) { > > Maybe check nested_exit_intr_ack_set first. > > Paolo > >> int irq =3D kvm_cpu_get_interrupt(vcpu); >> WARN_ON(irq < 0); >> vmcs12->vm_exit_intr_info =3D irq | >> >