Received-SPF: pass (google.com: domain of linux-kernel+bounces-70474-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.80.249 as permitted sender) client-ip=147.75.80.249;
Date: Sun, 18 Feb 2024 10:05:37 -0800
From: Oliver Upton <oliver.upton@linux.dev>
To: Marc Zyngier <maz@kernel.org>
Cc: Zenghui Yu <yuzenghui@huawei.com>, kvmarm@lists.linux.dev,
	kvm@vger.kernel.org, James Morse <james.morse@arm.com>,
	Suzuki K Poulose <suzuki.poulose@arm.com>,
	linux-kernel@vger.kernel.org
Subject: Re: [PATCH v3 04/10] KVM: arm64: vgic-its: Walk the LPI xarray in
 vgic_copy_lpi_list()
Message-ID: <ZdJG8W-TLW8I6O07@linux.dev>
References: <20240216184153.2714504-1-oliver.upton@linux.dev>
 <20240216184153.2714504-5-oliver.upton@linux.dev>
 <beca07ad-833e-ca68-2fe7-a30a2cb9faef@huawei.com>
 <86frxq3w3g.wl-maz@kernel.org>
Precedence: bulk
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <86frxq3w3g.wl-maz@kernel.org>

On Sun, Feb 18, 2024 at 10:28:19AM +0000, Marc Zyngier wrote:
> On Sun, 18 Feb 2024 08:46:53 +0000,
> Zenghui Yu <yuzenghui@huawei.com> wrote:
> > 
> > On 2024/2/17 2:41, Oliver Upton wrote:
> > > Start iterating the LPI xarray in anticipation of removing the LPI
> > > linked-list.
> > > 
> > > Signed-off-by: Oliver Upton <oliver.upton@linux.dev>
> > > ---
> > >  arch/arm64/kvm/vgic/vgic-its.c | 7 ++++++-
> > >  1 file changed, 6 insertions(+), 1 deletion(-)
> > > 
> > > diff --git a/arch/arm64/kvm/vgic/vgic-its.c b/arch/arm64/kvm/vgic/vgic-its.c
> > > index fb2d3c356984..9ce2edfadd11 100644
> > > --- a/arch/arm64/kvm/vgic/vgic-its.c
> > > +++ b/arch/arm64/kvm/vgic/vgic-its.c
> > > @@ -335,6 +335,7 @@ static int update_lpi_config(struct kvm *kvm, struct vgic_irq *irq,
> > >  int vgic_copy_lpi_list(struct kvm *kvm, struct kvm_vcpu *vcpu, u32 **intid_ptr)
> > >  {
> > >  	struct vgic_dist *dist = &kvm->arch.vgic;
> > > +	XA_STATE(xas, &dist->lpi_xa, GIC_LPI_OFFSET);
> > >  	struct vgic_irq *irq;
> > >  	unsigned long flags;
> > >  	u32 *intids;
> > > @@ -353,7 +354,9 @@ int vgic_copy_lpi_list(struct kvm *kvm, struct kvm_vcpu *vcpu, u32 **intid_ptr)
> > >  		return -ENOMEM;
> > >   	raw_spin_lock_irqsave(&dist->lpi_list_lock, flags);
> > > -	list_for_each_entry(irq, &dist->lpi_list_head, lpi_list) {
> > > +	rcu_read_lock();
> > > +
> > > +	xas_for_each(&xas, irq, INTERRUPT_ID_BITS_ITS) {
> > 
> > We should use '1 << INTERRUPT_ID_BITS_ITS - 1' to represent the maximum
> > LPI interrupt ID.

/facepalm

Thanks Zenghui!

> Huh, well caught! I'm not even sure how it works, as that's way
> smaller than the start of the walk (8192). Probably doesn't.
> 
> An alternative would be to use max_lpis_propbaser(), but I'm not sure
> we always have a valid PROPBASER value set when we start using this
> function. Worth investigating though.

Given the plans to eventually replace this with xarray marks, I'd vote
for doing the lazy thing and deciding this at compile time.

I can squash this in when I apply the series if the rest of it isn't
offensive, otherwise respin with the change.

-- 
Thanks,
Oliver

diff --git a/arch/arm64/kvm/vgic/vgic-its.c b/arch/arm64/kvm/vgic/vgic-its.c
index d84cb7618c59..f6025886071c 100644
--- a/arch/arm64/kvm/vgic/vgic-its.c
+++ b/arch/arm64/kvm/vgic/vgic-its.c
@@ -316,6 +316,8 @@ static int update_lpi_config(struct kvm *kvm, struct vgic_irq *irq,
 	return 0;
 }
 
+#define GIC_LPI_MAX_INTID	((1 << INTERRUPT_ID_BITS_ITS) - 1)
+
 /*
  * Create a snapshot of the current LPIs targeting @vcpu, so that we can
  * enumerate those LPIs without holding any lock.
@@ -345,7 +347,7 @@ int vgic_copy_lpi_list(struct kvm *kvm, struct kvm_vcpu *vcpu, u32 **intid_ptr)
 	raw_spin_lock_irqsave(&dist->lpi_list_lock, flags);
 	rcu_read_lock();
 
-	xas_for_each(&xas, irq, INTERRUPT_ID_BITS_ITS) {
+	xas_for_each(&xas, irq, GIC_LPI_MAX_INTID) {
 		if (i == irq_count)
 			break;
 		/* We don't need to "get" the IRQ, as we hold the list lock. */