linux-kernel - Re: [PATCH] KVM: VMX: Don't register posted interrupt wakeup handler if alloc_kvm

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <aWZwE1QukfjYDB_Q@google.com>
Date: Tue, 13 Jan 2026 08:17:23 -0800
From: Sean Christopherson <seanjc@...gle.com>
To: Hou Wenlong <houwenlong.hwl@...group.com>
Cc: kvm@...r.kernel.org, Paolo Bonzini <pbonzini@...hat.com>, 
	Thomas Gleixner <tglx@...utronix.de>, Ingo Molnar <mingo@...hat.com>, Borislav Petkov <bp@...en8.de>, 
	Dave Hansen <dave.hansen@...ux.intel.com>, x86@...nel.org, 
	"H. Peter Anvin" <hpa@...or.com>, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] KVM: VMX: Don't register posted interrupt wakeup handler
 if alloc_kvm_area() fails

On Tue, Jan 13, 2026, Hou Wenlong wrote:
> Unregistering the posted interrupt wakeup handler only happens during
> hardware unsetup. Therefore, if alloc_kvm_area() fails and continue to
> register the posted interrupt wakeup handler, this will leave the global
> posted interrupt wakeup handler pointer in an incorrect state. Although
> it should not be an issue, it's still better to change it.

Ouch, yeah, that's ugly.  It's not entirely benign, as a failed allocation followed
by a spurious notification vector IRQ would trigger UAF.  So it's probably worth
adding:

  Fixes: ec5a4919fa7b ("KVM: VMX: Unregister posted interrupt wakeup handler on hardware unsetup")
  Cc: stable@...r.kernel.org

even though I agree it's extremely unlikely to be an issue in practice.

> Signed-off-by: Hou Wenlong <houwenlong.hwl@...group.com>
> ---
>  arch/x86/kvm/vmx/vmx.c | 7 +++++--
>  1 file changed, 5 insertions(+), 2 deletions(-)
> 
> diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c
> index 9b92f672ccfe..676f32aa72bb 100644
> --- a/arch/x86/kvm/vmx/vmx.c
> +++ b/arch/x86/kvm/vmx/vmx.c
> @@ -8829,8 +8829,11 @@ __init int vmx_hardware_setup(void)
>  	}
>  
>  	r = alloc_kvm_area();
> -	if (r && nested)
> -		nested_vmx_hardware_unsetup();
> +	if (r) {
> +		if (nested)
> +			nested_vmx_hardware_unsetup();
> +		return r;
> +	}

I'm leaning towards using a goto with an explicit "return 0" in the happy case,
to make it less likely that a similar bug is introduced in the future.  Any
preference on your end?

E.g. (untested)

diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c
index 9b92f672ccfe..cecaaeb3f82a 100644
--- a/arch/x86/kvm/vmx/vmx.c
+++ b/arch/x86/kvm/vmx/vmx.c
@@ -8829,8 +8829,8 @@ __init int vmx_hardware_setup(void)
        }
 
        r = alloc_kvm_area();
-       if (r && nested)
-               nested_vmx_hardware_unsetup();
+       if (r)
+               goto err_kvm_area;
 
        kvm_set_posted_intr_wakeup_handler(pi_wakeup_handler);
 
@@ -8857,6 +8857,11 @@ __init int vmx_hardware_setup(void)
 
        kvm_caps.inapplicable_quirks &= ~KVM_X86_QUIRK_IGNORE_GUEST_PAT;
 
+       return 0;
+
+err_kvm_area:
+       if (nested)
+               nested_vmx_hardware_unsetup();
        return r;
 }