From 2240a50e6294214de791729e9dcba6880fa7e44e Mon Sep 17 00:00:00 2001 From: Zenghui Yu Date: Sat, 17 Aug 2024 18:15:41 +0800 Subject: [PATCH 1/4] KVM: arm64: vgic-debug: Don't put unmarked LPIs If there were LPIs being mapped behind our back (i.e., between .start() and .stop()), we would put them at iter_unmark_lpis() without checking if they were actually *marked*, which is obviously not good. Switch to use the xa_for_each_marked() iterator to fix it. Cc: stable@vger.kernel.org Fixes: 85d3ccc8b75b ("KVM: arm64: vgic-debug: Use an xarray mark for debug iterator") Signed-off-by: Zenghui Yu Reviewed-by: Marc Zyngier Link: https://lore.kernel.org/r/20240817101541.1664-1-yuzenghui@huawei.com Signed-off-by: Oliver Upton --- arch/arm64/kvm/vgic/vgic-debug.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm64/kvm/vgic/vgic-debug.c b/arch/arm64/kvm/vgic/vgic-debug.c index bc74d06398ef..e1397ab2072a 100644 --- a/arch/arm64/kvm/vgic/vgic-debug.c +++ b/arch/arm64/kvm/vgic/vgic-debug.c @@ -85,7 +85,7 @@ static void iter_unmark_lpis(struct kvm *kvm) struct vgic_irq *irq; unsigned long intid; - xa_for_each(&dist->lpi_xa, intid, irq) { + xa_for_each_marked(&dist->lpi_xa, intid, irq, LPI_XA_MARK_DEBUG_ITER) { xa_clear_mark(&dist->lpi_xa, intid, LPI_XA_MARK_DEBUG_ITER); vgic_put_irq(kvm, irq); } From f616506754d34bcfdbfbc7508b562e5c98461e9a Mon Sep 17 00:00:00 2001 From: Marc Zyngier Date: Mon, 19 Aug 2024 13:50:45 +0100 Subject: [PATCH 2/4] KVM: arm64: vgic: Don't hold config_lock while unregistering redistributors We recently moved the teardown of the vgic part of a vcpu inside a critical section guarded by the config_lock. This teardown phase involves calling into kvm_io_bus_unregister_dev(), which takes the kvm->srcu lock. However, this violates the established order where kvm->srcu is taken on a memory fault (such as an MMIO access), possibly followed by taking the config_lock if the GIC emulation requires mutual exclusion from the other vcpus. It therefore results in a bad lockdep splat, as reported by Zenghui. Fix this by moving the call to kvm_io_bus_unregister_dev() outside of the config_lock critical section. At this stage, there shouln't be any need to hold the config_lock. As an additional bonus, document the ordering between kvm->slots_lock, kvm->srcu and kvm->arch.config_lock so that I cannot pretend I didn't know about those anymore. Fixes: 9eb18136af9f ("KVM: arm64: vgic: Hold config_lock while tearing down a CPU interface") Reported-by: Zenghui Yu Signed-off-by: Marc Zyngier Reviewed-by: Zenghui Yu Tested-by: Zenghui Yu Link: https://lore.kernel.org/r/20240819125045.3474845-1-maz@kernel.org Signed-off-by: Oliver Upton --- arch/arm64/kvm/vgic/vgic-init.c | 9 ++++++--- arch/arm64/kvm/vgic/vgic.c | 5 +++++ 2 files changed, 11 insertions(+), 3 deletions(-) diff --git a/arch/arm64/kvm/vgic/vgic-init.c b/arch/arm64/kvm/vgic/vgic-init.c index 41feb858ff9a..e7c53e8af3d1 100644 --- a/arch/arm64/kvm/vgic/vgic-init.c +++ b/arch/arm64/kvm/vgic/vgic-init.c @@ -417,10 +417,8 @@ static void __kvm_vgic_vcpu_destroy(struct kvm_vcpu *vcpu) kfree(vgic_cpu->private_irqs); vgic_cpu->private_irqs = NULL; - if (vcpu->kvm->arch.vgic.vgic_model == KVM_DEV_TYPE_ARM_VGIC_V3) { - vgic_unregister_redist_iodev(vcpu); + if (vcpu->kvm->arch.vgic.vgic_model == KVM_DEV_TYPE_ARM_VGIC_V3) vgic_cpu->rd_iodev.base_addr = VGIC_ADDR_UNDEF; - } } void kvm_vgic_vcpu_destroy(struct kvm_vcpu *vcpu) @@ -448,6 +446,11 @@ void kvm_vgic_destroy(struct kvm *kvm) kvm_vgic_dist_destroy(kvm); mutex_unlock(&kvm->arch.config_lock); + + if (kvm->arch.vgic.vgic_model == KVM_DEV_TYPE_ARM_VGIC_V3) + kvm_for_each_vcpu(i, vcpu, kvm) + vgic_unregister_redist_iodev(vcpu); + mutex_unlock(&kvm->slots_lock); } diff --git a/arch/arm64/kvm/vgic/vgic.c b/arch/arm64/kvm/vgic/vgic.c index 974849ea7101..abe29c7d85d0 100644 --- a/arch/arm64/kvm/vgic/vgic.c +++ b/arch/arm64/kvm/vgic/vgic.c @@ -36,6 +36,11 @@ struct vgic_global kvm_vgic_global_state __ro_after_init = { * we have to disable IRQs before taking this lock and everything lower * than it. * + * The config_lock has additional ordering requirements: + * kvm->slots_lock + * kvm->srcu + * kvm->arch.config_lock + * * If you need to take multiple locks, always take the upper lock first, * then the lower ones, e.g. first take the its_lock, then the irq_lock. * If you are already holding a lock and need to take a higher one, you From 1d8c3c23a6bc1527e253b305b4b68c03d833b824 Mon Sep 17 00:00:00 2001 From: Oliver Upton Date: Thu, 22 Aug 2024 07:17:09 +0000 Subject: [PATCH 3/4] KVM: arm64: Ensure canonical IPA is hugepage-aligned when handling fault Zenghui reports that VMs backed by hugetlb pages are no longer booting after commit fd276e71d1e7 ("KVM: arm64: nv: Handle shadow stage 2 page faults"). Support for shadow stage-2 MMUs introduced the concept of a fault IPA and canonical IPA to stage-2 fault handling. These are identical in the non-nested case, as the hardware stage-2 context is always that of the canonical IPA space. Both addresses need to be hugepage-aligned when preparing to install a hugepage mapping to ensure that KVM uses the correct GFN->PFN translation and installs that at the correct IPA for the current stage-2. And now I'm feeling thirsty after all this talk of IPAs... Fixes: fd276e71d1e7 ("KVM: arm64: nv: Handle shadow stage 2 page faults") Reported-by: Zenghui Yu Reviewed-by: Marc Zyngier Link: https://lore.kernel.org/r/20240822071710.2291690-1-oliver.upton@linux.dev Signed-off-by: Oliver Upton --- arch/arm64/kvm/mmu.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/arch/arm64/kvm/mmu.c b/arch/arm64/kvm/mmu.c index 6981b1bc0946..a509b63bd4dd 100644 --- a/arch/arm64/kvm/mmu.c +++ b/arch/arm64/kvm/mmu.c @@ -1540,8 +1540,15 @@ static int user_mem_abort(struct kvm_vcpu *vcpu, phys_addr_t fault_ipa, vma_pagesize = min(vma_pagesize, (long)max_map_size); } - if (vma_pagesize == PMD_SIZE || vma_pagesize == PUD_SIZE) + /* + * Both the canonical IPA and fault IPA must be hugepage-aligned to + * ensure we find the right PFN and lay down the mapping in the right + * place. + */ + if (vma_pagesize == PMD_SIZE || vma_pagesize == PUD_SIZE) { fault_ipa &= ~(vma_pagesize - 1); + ipa &= ~(vma_pagesize - 1); + } gfn = ipa >> PAGE_SHIFT; mte_allowed = kvm_vma_mte_allowed(vma); From 3e6245ebe7ef341639e9a7e402b3ade8ad45a19f Mon Sep 17 00:00:00 2001 From: Marc Zyngier Date: Tue, 20 Aug 2024 11:03:38 +0100 Subject: [PATCH 4/4] KVM: arm64: Make ICC_*SGI*_EL1 undef in the absence of a vGICv3 On a system with a GICv3, if a guest hasn't been configured with GICv3 and that the host is not capable of GICv2 emulation, a write to any of the ICC_*SGI*_EL1 registers is trapped to EL2. We therefore try to emulate the SGI access, only to hit a NULL pointer as no private interrupt is allocated (no GIC, remember?). The obvious fix is to give the guest what it deserves, in the shape of a UNDEF exception. Reported-by: Alexander Potapenko Signed-off-by: Marc Zyngier Cc: stable@vger.kernel.org Link: https://lore.kernel.org/r/20240820100349.3544850-2-maz@kernel.org Signed-off-by: Oliver Upton --- arch/arm64/kvm/sys_regs.c | 6 ++++++ arch/arm64/kvm/vgic/vgic.h | 7 +++++++ 2 files changed, 13 insertions(+) diff --git a/arch/arm64/kvm/sys_regs.c b/arch/arm64/kvm/sys_regs.c index c90324060436..31e49da867ff 100644 --- a/arch/arm64/kvm/sys_regs.c +++ b/arch/arm64/kvm/sys_regs.c @@ -33,6 +33,7 @@ #include #include "sys_regs.h" +#include "vgic/vgic.h" #include "trace.h" @@ -435,6 +436,11 @@ static bool access_gic_sgi(struct kvm_vcpu *vcpu, { bool g1; + if (!kvm_has_gicv3(vcpu->kvm)) { + kvm_inject_undefined(vcpu); + return false; + } + if (!p->is_write) return read_from_write_only(vcpu, p, r); diff --git a/arch/arm64/kvm/vgic/vgic.h b/arch/arm64/kvm/vgic/vgic.h index ba8f790431bd..8532bfe3fed4 100644 --- a/arch/arm64/kvm/vgic/vgic.h +++ b/arch/arm64/kvm/vgic/vgic.h @@ -346,4 +346,11 @@ void vgic_v4_configure_vsgis(struct kvm *kvm); void vgic_v4_get_vlpi_state(struct vgic_irq *irq, bool *val); int vgic_v4_request_vpe_irq(struct kvm_vcpu *vcpu, int irq); +static inline bool kvm_has_gicv3(struct kvm *kvm) +{ + return (static_branch_unlikely(&kvm_vgic_global_state.gicv3_cpuif) && + irqchip_in_kernel(kvm) && + kvm->arch.vgic.vgic_model == KVM_DEV_TYPE_ARM_VGIC_V3); +} + #endif