From: Marc Zyngier <maz@kernel.org> mainline inclusion from mainline-v6.8-rc1 commit 7b95382f965133ef61ce44aaabc518c16eb46909 category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/ID2P1J CVE: NA Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?i... -------------------------------- When the VMM writes to ISPENDR0 to set the state pending state of an SGI, we fail to convey this to the HW if this SGI is already backed by a GICv4.1 vSGI. This is a bit of a corner case, as this would only occur if the vgic state is changed on an already running VM, but this can apparently happen across a guest reset driven by the VMM. Fix this by always writing out the pending_latch value to the HW, and reseting it to false. Reported-by: Kunkun Jiang <jiangkunkun@huawei.com> Signed-off-by: Marc Zyngier <maz@kernel.org> Reviewed-by: Zenghui Yu <yuzenghui@huawei.com> Cc: stable@vger.kernel.org # 5.10+ Link: https://lore.kernel.org/r/7e7f2c0c-448b-10a9-8929-4b8f4f6e2a32@huawei.com Signed-off-by: Jinqian Yang <yangjinqian1@huawei.com> --- arch/arm64/kvm/vgic/vgic-mmio-v3.c | 49 ++++++++++++++++-------------- 1 file changed, 26 insertions(+), 23 deletions(-) diff --git a/arch/arm64/kvm/vgic/vgic-mmio-v3.c b/arch/arm64/kvm/vgic/vgic-mmio-v3.c index b6bdf22abab1..fd29102a5388 100644 --- a/arch/arm64/kvm/vgic/vgic-mmio-v3.c +++ b/arch/arm64/kvm/vgic/vgic-mmio-v3.c @@ -377,32 +377,35 @@ static int vgic_v3_uaccess_write_pending(struct kvm_vcpu *vcpu, struct vgic_irq *irq = vgic_get_irq(vcpu->kvm, vcpu, intid + i); raw_spin_lock_irqsave(&irq->irq_lock, flags); - if (test_bit(i, &val)) { - /* - * pending_latch is set irrespective of irq type - * (level or edge) to avoid dependency that VM should - * restore irq config before pending info. - */ - irq->pending_latch = true; - vgic_queue_irq_unlock(vcpu->kvm, irq, flags); - } else { - /** - * workaround: On reset, userspace clears pending status - * for all PPIs and SGIs by writing all 0's to - * GICR_ISPENDR0. The pending state of vtimer interrupt - * is somehow staying in redistributor and we have to - * explicitly clear it... - * - * P.S., irq->vtimer_info is NULL on restore. - */ - if (irq->vtimer_info) - WARN_ON_ONCE(irq_set_irqchip_state(irq->host_irq, - IRQCHIP_STATE_PENDING, - false)); + + /* + * pending_latch is set irrespective of irq type + * (level or edge) to avoid dependency that VM should + * restore irq config before pending info. + */ + irq->pending_latch = test_bit(i, &val); + + /* + * workaround: On reset, userspace clears pending status + * for all PPIs and SGIs by writing all 0's to + * GICR_ISPENDR0. The pending state of vtimer interrupt + * is somehow staying in redistributor and we have to + * explicitly clear it... + * + * P.S., irq->vtimer_info is NULL on restore. + */ + if ((irq->hw && vgic_irq_is_sgi(irq->intid)) || irq->vtimer_info) { + irq_set_irqchip_state(irq->host_irq, + IRQCHIP_STATE_PENDING, + irq->pending_latch); irq->pending_latch = false; - raw_spin_unlock_irqrestore(&irq->irq_lock, flags); } + if (irq->pending_latch) + vgic_queue_irq_unlock(vcpu->kvm, irq, flags); + else + raw_spin_unlock_irqrestore(&irq->irq_lock, flags); + vgic_put_irq(vcpu->kvm, irq); } -- 2.33.0