Re: [PATCH v3 3/4] KVM: arm64: GICv4.1: Restore VLPI's pending state to physical side

From: Marc Zyngier <maz@kernel.org>
To: Shenming Lu <lushenming@huawei.com>
Cc: Eric Auger <eric.auger@redhat.com>, Will Deacon <will@kernel.org>,
	<linux-arm-kernel@lists.infradead.org>,
	<kvmarm@lists.cs.columbia.edu>, <kvm@vger.kernel.org>,
	<linux-kernel@vger.kernel.org>,
	Alex Williamson <alex.williamson@redhat.com>,
	Cornelia Huck <cohuck@redhat.com>,
	"Lorenzo\ Pieralisi" <lorenzo.pieralisi@arm.com>,
	<wanghaibin.wang@huawei.com>, <yuzenghui@huawei.com>
Subject: Re: [PATCH v3 3/4] KVM: arm64: GICv4.1: Restore VLPI's pending state to physical side
Date: Fri, 12 Mar 2021 09:05:45 +0000	[thread overview]
Message-ID: <87k0qcg2s6.wl-maz@kernel.org> (raw)
In-Reply-To: <0820f429-4c29-acd6-d9e0-af9f6deb68e4@huawei.com>

On Thu, 11 Mar 2021 12:32:07 +0000,
Shenming Lu <lushenming@huawei.com> wrote:
> 
> On 2021/3/11 17:14, Marc Zyngier wrote:
> > On Wed, 27 Jan 2021 12:13:36 +0000,
> > Shenming Lu <lushenming@huawei.com> wrote:
> >>
> >> From: Zenghui Yu <yuzenghui@huawei.com>
> >>
> >> When setting the forwarding path of a VLPI (switch to the HW mode),
> >> we could also transfer the pending state from irq->pending_latch to
> >> VPT (especially in migration, the pending states of VLPIs are restored
> >> into kvm’s vgic first). And we currently send "INT+VSYNC" to trigger
> >> a VLPI to pending.
> >>
> >> Signed-off-by: Zenghui Yu <yuzenghui@huawei.com>
> >> Signed-off-by: Shenming Lu <lushenming@huawei.com>
> >> ---
> >>  arch/arm64/kvm/vgic/vgic-v4.c | 14 ++++++++++++++
> >>  1 file changed, 14 insertions(+)
> >>
> >> diff --git a/arch/arm64/kvm/vgic/vgic-v4.c b/arch/arm64/kvm/vgic/vgic-v4.c
> >> index ac029ba3d337..a3542af6f04a 100644
> >> --- a/arch/arm64/kvm/vgic/vgic-v4.c
> >> +++ b/arch/arm64/kvm/vgic/vgic-v4.c
> >> @@ -449,6 +449,20 @@ int kvm_vgic_v4_set_forwarding(struct kvm *kvm, int virq,
> >>  	irq->host_irq	= virq;
> >>  	atomic_inc(&map.vpe->vlpi_count);
> >>  
> >> +	/* Transfer pending state */
> >> +	if (irq->pending_latch) {
> >> +		ret = irq_set_irqchip_state(irq->host_irq,
> >> +					    IRQCHIP_STATE_PENDING,
> >> +					    irq->pending_latch);
> >> +		WARN_RATELIMIT(ret, "IRQ %d", irq->host_irq);
> >> +
> >> +		/*
> >> +		 * Let it be pruned from ap_list later and don't bother
> >> +		 * the List Register.
> >> +		 */
> >> +		irq->pending_latch = false;
> > 
> > NAK. If the interrupt is on the AP list, it must be pruned from it
> > *immediately*. The only case where it can be !pending and still on the
> > AP list is in interval between sync and prune. If we start messing
> > with this, we can't reason about the state of this list anymore.
> > 
> > Consider calling vgic_queue_irq_unlock() here.
> 
> Thanks for giving a hint, but it seems that vgic_queue_irq_unlock() only
> queues an IRQ after checking, did you mean vgic_prune_ap_list() instead?

No, I really mean vgic_queue_irq_unlock(). It can be used to remove
the pending state from an interrupt, and drop it from the AP
list. This is exactly what happens when clearing the pending state of
a level interrupt, for example.

	M.

-- 
Without deviation from the norm, progress is not possible.

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel