[PATCH 4/4] KVM: Nullify async #PF worker's "apf" pointer as soon as it might be freed

Sean Christopherson posted 4 patches 1 year, 11 months ago
[PATCH 4/4] KVM: Nullify async #PF worker's "apf" pointer as soon as it might be freed
Posted by Sean Christopherson 1 year, 11 months ago
Nullify the async #PF worker's local "apf" pointer immediately after the
point where the structure can be freed by the vCPU.  The existing comment
is helpful, but easy to overlook as there is no associated code.

Update the comment to clarify that it can be freed by as soon as the lock
is dropped, as "after this point" isn't strictly accurate, nor does it
help understand what prevents the structure from being freed earlier.

Signed-off-by: Sean Christopherson <seanjc@google.com>
---
 virt/kvm/async_pf.c | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)

diff --git a/virt/kvm/async_pf.c b/virt/kvm/async_pf.c
index c3f4f351a2ae..1088c6628de9 100644
--- a/virt/kvm/async_pf.c
+++ b/virt/kvm/async_pf.c
@@ -83,13 +83,14 @@ static void async_pf_execute(struct work_struct *work)
 	apf->vcpu = NULL;
 	spin_unlock(&vcpu->async_pf.lock);
 
-	if (!IS_ENABLED(CONFIG_KVM_ASYNC_PF_SYNC) && first)
-		kvm_arch_async_page_present_queued(vcpu);
-
 	/*
-	 * apf may be freed by kvm_check_async_pf_completion() after
-	 * this point
+	 * The apf struct may freed by kvm_check_async_pf_completion() as soon
+	 * as the lock is dropped.  Nullify it to prevent improper usage.
 	 */
+	apf = NULL;
+
+	if (!IS_ENABLED(CONFIG_KVM_ASYNC_PF_SYNC) && first)
+		kvm_arch_async_page_present_queued(vcpu);
 
 	trace_kvm_async_pf_completed(addr, cr2_or_gpa);
 
-- 
2.43.0.472.g3155946c3a-goog
Re: [PATCH 4/4] KVM: Nullify async #PF worker's "apf" pointer as soon as it might be freed
Posted by Vitaly Kuznetsov 1 year, 11 months ago
Sean Christopherson <seanjc@google.com> writes:

> Nullify the async #PF worker's local "apf" pointer immediately after the
> point where the structure can be freed by the vCPU.  The existing comment
> is helpful, but easy to overlook as there is no associated code.
>
> Update the comment to clarify that it can be freed by as soon as the lock
> is dropped, as "after this point" isn't strictly accurate, nor does it
> help understand what prevents the structure from being freed earlier.
>

"No functional change intended." must be made a requirement, especially
for those who made it their trademark)

> Signed-off-by: Sean Christopherson <seanjc@google.com>
> ---
>  virt/kvm/async_pf.c | 11 ++++++-----
>  1 file changed, 6 insertions(+), 5 deletions(-)
>
> diff --git a/virt/kvm/async_pf.c b/virt/kvm/async_pf.c
> index c3f4f351a2ae..1088c6628de9 100644
> --- a/virt/kvm/async_pf.c
> +++ b/virt/kvm/async_pf.c
> @@ -83,13 +83,14 @@ static void async_pf_execute(struct work_struct *work)
>  	apf->vcpu = NULL;
>  	spin_unlock(&vcpu->async_pf.lock);
>  
> -	if (!IS_ENABLED(CONFIG_KVM_ASYNC_PF_SYNC) && first)
> -		kvm_arch_async_page_present_queued(vcpu);
> -
>  	/*
> -	 * apf may be freed by kvm_check_async_pf_completion() after
> -	 * this point
> +	 * The apf struct may freed by kvm_check_async_pf_completion() as soon

Nit: "may be freed"/"may get freed" maybe?

> +	 * as the lock is dropped.  Nullify it to prevent improper usage.
>  	 */
> +	apf = NULL;
> +
> +	if (!IS_ENABLED(CONFIG_KVM_ASYNC_PF_SYNC) && first)
> +		kvm_arch_async_page_present_queued(vcpu);
>  
>  	trace_kvm_async_pf_completed(addr, cr2_or_gpa);

Reviewed-by: Vitaly Kuznetsov <vkuznets@redhat.com>

-- 
Vitaly
Re: [PATCH 4/4] KVM: Nullify async #PF worker's "apf" pointer as soon as it might be freed
Posted by Xu Yilun 1 year, 11 months ago
On Tue, Jan 09, 2024 at 05:15:33PM -0800, Sean Christopherson wrote:
> Nullify the async #PF worker's local "apf" pointer immediately after the
> point where the structure can be freed by the vCPU.  The existing comment
> is helpful, but easy to overlook as there is no associated code.
> 
> Update the comment to clarify that it can be freed by as soon as the lock
> is dropped, as "after this point" isn't strictly accurate, nor does it
> help understand what prevents the structure from being freed earlier.
> 
> Signed-off-by: Sean Christopherson <seanjc@google.com>

Reviewed-by: Xu Yilun <yilun.xu@intel.com>

> ---
>  virt/kvm/async_pf.c | 11 ++++++-----
>  1 file changed, 6 insertions(+), 5 deletions(-)
> 
> diff --git a/virt/kvm/async_pf.c b/virt/kvm/async_pf.c
> index c3f4f351a2ae..1088c6628de9 100644
> --- a/virt/kvm/async_pf.c
> +++ b/virt/kvm/async_pf.c
> @@ -83,13 +83,14 @@ static void async_pf_execute(struct work_struct *work)
>  	apf->vcpu = NULL;
>  	spin_unlock(&vcpu->async_pf.lock);
>  
> -	if (!IS_ENABLED(CONFIG_KVM_ASYNC_PF_SYNC) && first)
> -		kvm_arch_async_page_present_queued(vcpu);
> -
>  	/*
> -	 * apf may be freed by kvm_check_async_pf_completion() after
> -	 * this point
> +	 * The apf struct may freed by kvm_check_async_pf_completion() as soon
> +	 * as the lock is dropped.  Nullify it to prevent improper usage.
>  	 */
> +	apf = NULL;
> +
> +	if (!IS_ENABLED(CONFIG_KVM_ASYNC_PF_SYNC) && first)
> +		kvm_arch_async_page_present_queued(vcpu);
>  
>  	trace_kvm_async_pf_completed(addr, cr2_or_gpa);
>  
> -- 
> 2.43.0.472.g3155946c3a-goog
>