All of lore.kernel.org
 help / color / mirror / Atom feed
From: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
To: Gleb Natapov <gleb@redhat.com>
Cc: Avi Kivity <avi@redhat.com>,
	Marcelo Tosatti <mtosatti@redhat.com>,
	LKML <linux-kernel@vger.kernel.org>, KVM <kvm@vger.kernel.org>
Subject: Re: [RFC PATCH v2 7/7] KVM: KVM: don't break vcpu 'halt' state due to apfs
Date: Tue, 02 Nov 2010 10:30:10 +0800	[thread overview]
Message-ID: <4CCF77B2.9080604@cn.fujitsu.com> (raw)
In-Reply-To: <20101101125551.GD31722@redhat.com>

On 11/01/2010 08:55 PM, Gleb Natapov wrote:

> diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
> index 2cfdf2d..f7aed95 100644
> --- a/arch/x86/kvm/x86.c
> +++ b/arch/x86/kvm/x86.c
> @@ -5295,8 +5295,9 @@ static int __vcpu_run(struct kvm_vcpu *vcpu)
>  			{
>  				switch(vcpu->arch.mp_state) {
>  				case KVM_MP_STATE_HALTED:
> -					vcpu->arch.mp_state =
> -						KVM_MP_STATE_RUNNABLE;
> +					if (list_empty_careful(&vcpu->async_pf.done))
> +						vcpu->arch.mp_state =
> +							KVM_MP_STATE_RUNNABLE;

if nmi/interrupt and apfs completed event occur at the same time, we will miss to
exit halt sate. Maybe we can check the pending event here, see below patch please.

>  				case KVM_MP_STATE_RUNNABLE:
>  					vcpu->arch.apf.halted = false;
>  					break;
> @@ -6279,6 +6280,7 @@ void kvm_arch_async_page_present(struct kvm_vcpu *vcpu,
>  		vcpu->arch.fault.error_code = 0;
>  		vcpu->arch.fault.address = work->arch.token;
>  		kvm_inject_page_fault(vcpu);
> +		vcpu->arch.mp_state = KVM_MP_STATE_RUNNABLE;
>  	}
>  }

Have a stupid question, why we make the vcpu runnable here? Sorry i don't know
kvm pv guest to much. :-(

diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 2044302..27a712b 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -5267,6 +5267,12 @@ out:
 	return r;
 }
 
+static bool kvm_vcpu_leave_halt_state(struct kvm_vcpu *vcpu)
+{
+	return vcpu->arch.nmi_pending ||
+		(kvm_arch_interrupt_allowed(vcpu) &&
+		 kvm_cpu_has_interrupt(vcpu));
+}
 
 static int __vcpu_run(struct kvm_vcpu *vcpu)
 {
@@ -5299,8 +5305,9 @@ static int __vcpu_run(struct kvm_vcpu *vcpu)
 			{
 				switch(vcpu->arch.mp_state) {
 				case KVM_MP_STATE_HALTED:
-					vcpu->arch.mp_state =
-						KVM_MP_STATE_RUNNABLE;
+					if (kvm_vcpu_leave_halt_state(vcpu))
+						vcpu->arch.mp_state =
+							KVM_MP_STATE_RUNNABLE;
 				case KVM_MP_STATE_RUNNABLE:
 					vcpu->arch.apf.halted = false;
 					break;
@@ -6113,9 +6120,7 @@ int kvm_arch_vcpu_runnable(struct kvm_vcpu *vcpu)
 		!vcpu->arch.apf.halted)
 		|| !list_empty_careful(&vcpu->async_pf.done)
 		|| vcpu->arch.mp_state == KVM_MP_STATE_SIPI_RECEIVED
-		|| vcpu->arch.nmi_pending ||
-		(kvm_arch_interrupt_allowed(vcpu) &&
-		 kvm_cpu_has_interrupt(vcpu));
+		|| kvm_vcpu_leave_halt_state(vcpu);
 }
 
 void kvm_vcpu_kick(struct kvm_vcpu *vcpu)

  reply	other threads:[~2010-11-02  2:25 UTC|newest]

Thread overview: 26+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-11-01  8:58 [PATCH v2 1/7] KVM: fix tracing kvm_try_async_get_page Xiao Guangrong
2010-11-01  8:59 ` [PATCH v2 2/7] KVM: cleanup aysnc_pf tracepoints Xiao Guangrong
2010-11-01  9:00 ` [PATCH v2 3/7] KVM: fix searching async gfn in kvm_async_pf_gfn_slot Xiao Guangrong
2010-11-01  9:01 ` [PATCH v2 4/7] KVM: avoid unnecessary wait for a async pf Xiao Guangrong
2010-11-01  9:25   ` Gleb Natapov
2010-11-01  9:02 ` [PATCH v2 5/7] KVM: handle more completed apfs if possible Xiao Guangrong
2010-11-01  9:24   ` Gleb Natapov
2010-11-01  9:34     ` Xiao Guangrong
2010-11-02  9:35     ` [PATCH v3 " Xiao Guangrong
2010-11-02 12:38       ` Gleb Natapov
2010-11-01  9:03 ` [RFC PATCH v2 6/7] KVM: fix the race while wakeup all pv guest Xiao Guangrong
2010-11-01 12:58   ` Gleb Natapov
2010-11-01  9:05 ` [RFC PATCH v2 7/7] KVM: KVM: don't break vcpu 'halt' state due to apfs Xiao Guangrong
2010-11-01 12:55   ` Gleb Natapov
2010-11-02  2:30     ` Xiao Guangrong [this message]
2010-11-02  6:56       ` Gleb Natapov
2010-11-02  7:31         ` Xiao Guangrong
2010-11-02  7:45           ` Gleb Natapov
2010-11-02  9:09             ` Xiao Guangrong
2010-11-02  9:14               ` Gleb Natapov
2010-11-02  9:30                 ` Xiao Guangrong
2010-11-02 12:39                   ` Gleb Natapov
2010-11-03  9:47     ` Xiao Guangrong
2010-11-03  9:45       ` Gleb Natapov
2010-11-03 13:43         ` Marcelo Tosatti
2010-11-01 13:09 ` [PATCH v2 1/7] KVM: fix tracing kvm_try_async_get_page Gleb Natapov

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=4CCF77B2.9080604@cn.fujitsu.com \
    --to=xiaoguangrong@cn.fujitsu.com \
    --cc=avi@redhat.com \
    --cc=gleb@redhat.com \
    --cc=kvm@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mtosatti@redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body. 
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.