From: Oliver Upton <oliver.upton@xxxxxxxxx> Running an L2 guest with GICv4 enabled goes absolutely nowhere, and gets into a vicious cycle of nested ERET followed by nested exception entry into the L1. When KVM does a put on a runnable vCPU, it marks the vPE as nonresident but does not request a doorbell IRQ. Behind the scenes in the ITS driver's view of the vCPU, its_vpe::pending_last gets set to true to indicate that context is still runnable. This comes to a head when doing the nested ERET into L2. The vPE doesn't get scheduled on the redistributor as it is exclusively part of the L1's VGIC context. kvm_vgic_vcpu_pending_irq() returns true because the vPE appears runnable, and KVM does a nested exception entry into the L1 before L2 ever gets off the ground. This issue can be papered over by requesting a doorbell IRQ when descheduling a vPE as part of a nested ERET. KVM needs this anyway to kick the vCPU out of the L2 when an IRQ becomes pending for the L1. Signed-off-by: Oliver Upton <oliver.upton@xxxxxxxxx> Link: https://lore.kernel.org/r/20240823212703.3576061-4-oliver.upton@xxxxxxxxx Signed-off-by: Marc Zyngier <maz@xxxxxxxxxx> --- arch/arm64/include/asm/kvm_host.h | 2 ++ arch/arm64/kvm/emulate-nested.c | 2 ++ arch/arm64/kvm/vgic/vgic-v4.c | 18 +++++++++++++++++- 3 files changed, 21 insertions(+), 1 deletion(-) diff --git a/arch/arm64/include/asm/kvm_host.h b/arch/arm64/include/asm/kvm_host.h index cb969c096d7bd..18d9166761972 100644 --- a/arch/arm64/include/asm/kvm_host.h +++ b/arch/arm64/include/asm/kvm_host.h @@ -958,6 +958,8 @@ struct kvm_vcpu_arch { #define PMUSERENR_ON_CPU __vcpu_single_flag(sflags, BIT(4)) /* WFI instruction trapped */ #define IN_WFI __vcpu_single_flag(sflags, BIT(5)) +/* KVM is currently emulating a nested ERET */ +#define IN_NESTED_ERET __vcpu_single_flag(sflags, BIT(6)) /* Pointer to the vcpu's SVE FFR for sve_{save,load}_state() */ diff --git a/arch/arm64/kvm/emulate-nested.c b/arch/arm64/kvm/emulate-nested.c index c460b8403aec5..69233dcc81a46 100644 --- a/arch/arm64/kvm/emulate-nested.c +++ b/arch/arm64/kvm/emulate-nested.c @@ -2434,6 +2434,7 @@ void kvm_emulate_nested_eret(struct kvm_vcpu *vcpu) } preempt_disable(); + vcpu_set_flag(vcpu, IN_NESTED_ERET); kvm_arch_vcpu_put(vcpu); if (!esr_iss_is_eretax(esr)) @@ -2445,6 +2446,7 @@ void kvm_emulate_nested_eret(struct kvm_vcpu *vcpu) *vcpu_cpsr(vcpu) = spsr; kvm_arch_vcpu_load(vcpu, smp_processor_id()); + vcpu_clear_flag(vcpu, IN_NESTED_ERET); preempt_enable(); kvm_pmu_nested_transition(vcpu); diff --git a/arch/arm64/kvm/vgic/vgic-v4.c b/arch/arm64/kvm/vgic/vgic-v4.c index eedecbbbcf31b..0d9fb235c0180 100644 --- a/arch/arm64/kvm/vgic/vgic-v4.c +++ b/arch/arm64/kvm/vgic/vgic-v4.c @@ -336,6 +336,22 @@ void vgic_v4_teardown(struct kvm *kvm) its_vm->vpes = NULL; } +static inline bool vgic_v4_want_doorbell(struct kvm_vcpu *vcpu) +{ + if (vcpu_get_flag(vcpu, IN_WFI)) + return true; + + if (likely(!vcpu_has_nv(vcpu))) + return false; + + /* + * GICv4 hardware is only ever used for the L1. Mark the vPE (i.e. the + * L1 context) nonresident and request a doorbell to kick us out of the + * L2 when an IRQ becomes pending. + */ + return vcpu_get_flag(vcpu, IN_NESTED_ERET); +} + int vgic_v4_put(struct kvm_vcpu *vcpu) { struct its_vpe *vpe = &vcpu->arch.vgic_cpu.vgic_v3.its_vpe; @@ -343,7 +359,7 @@ int vgic_v4_put(struct kvm_vcpu *vcpu) if (!vgic_supports_direct_msis(vcpu->kvm) || !vpe->resident) return 0; - return its_make_vpe_non_resident(vpe, !!vcpu_get_flag(vcpu, IN_WFI)); + return its_make_vpe_non_resident(vpe, vgic_v4_want_doorbell(vcpu)); } int vgic_v4_load(struct kvm_vcpu *vcpu) -- 2.39.2
