linux-kernel - Re: [patch V4 15/15] x86/kvm: Use generic exit to guest work function

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 for Android: free password hash cracker in your pocket

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

Date:   Wed, 22 Jul 2020 09:40:13 +0200
From:   Thomas Gleixner <tglx@...utronix.de>
To:     Sean Christopherson <sean.j.christopherson@...el.com>
Cc:     LKML <linux-kernel@...r.kernel.org>, x86@...nel.org,
        linux-arch@...r.kernel.org, Will Deacon <will@...nel.org>,
        Arnd Bergmann <arnd@...db.de>,
        Mark Rutland <mark.rutland@....com>,
        Kees Cook <keescook@...omium.org>,
        Keno Fischer <keno@...iacomputing.com>,
        Paolo Bonzini <pbonzini@...hat.com>, kvm@...r.kernel.org,
        Gabriel Krisman Bertazi <krisman@...labora.com>
Subject: Re: [patch V4 15/15] x86/kvm: Use generic exit to guest work function

Sean Christopherson <sean.j.christopherson@...el.com> writes:
> On Tue, Jul 21, 2020 at 12:57:21PM +0200, Thomas Gleixner wrote:
>> -		if (signal_pending(current))
>> +		if (__exit_to_guest_mode_work_pending())
>
> I whined about this back in v2[*].  "exit to guest mode" is confusing becuase
> virt terminology is "enter to guest" and "exit from guest", whereas the
> kernel's terminology here is "exit from kernel to guest".
>
> My past and current self still like XFER_TO_GUEST as the base terminology.
>
> [*] https://lkml.kernel.org/r/20191023144848.GH329@linux.intel.com

Forgot about that, sorry. Will fix.

>> @@ -8676,15 +8677,11 @@ static int vcpu_run(struct kvm_vcpu *vcp
>>  			break;
>>  		}
>>  
>> -		if (signal_pending(current)) {
>> -			r = -EINTR;
>> -			vcpu->run->exit_reason = KVM_EXIT_INTR;
>> -			++vcpu->stat.signal_exits;
>> -			break;
>> -		}
>> -		if (need_resched()) {
>> +		if (exit_to_guest_mode_work_pending()) {
>>  			srcu_read_unlock(&kvm->srcu, vcpu->srcu_idx);
>> -			cond_resched();
>> +			r = exit_to_guest_mode(vcpu);
>> +			if (r)
>
> This loses the stat.signal_exits accounting.  Maybe this?

No, it does not:

+		if (ti_work & _TIF_SIGPENDING) {
+			kvm_handle_signal_exit(vcpu);
+			return -EINTR;
+		}

and

+#ifdef CONFIG_KVM_EXIT_TO_GUEST_WORK
+static inline void kvm_handle_signal_exit(struct kvm_vcpu *vcpu)
+{
+	vcpu->run->exit_reason = KVM_EXIT_INTR;
+	vcpu->stat.signal_exits++;
+}
+#endif /* CONFIG_KVM_EXIT_TO_GUEST_WORK */

in patch 5/15

Thanks,

        tglx