When software writes to the GICD_SGIR with the TargetListFilter field set to 0, we should use the target_cpus mask as the VCPU destination mask for the SGI. However, because we were falling through to the next case due to a missing break, we would always send the SGI to all other cores than ourselves. This does not change anything on dual-core system (unless a core is IPI'ing itself), but would look quite bad on systems with more cores. Cc: Haibin Wang <wanghaibin202@xxxxxxxxx> Reported-by: Haibin Wang <wanghaibin202@xxxxxxxxx> Signed-off-by: Christoffer Dall <christoffer.dall@xxxxxxxxxx> --- virt/kvm/arm/vgic.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/virt/kvm/arm/vgic.c b/virt/kvm/arm/vgic.c index 6699ed9..ecee766 100644 --- a/virt/kvm/arm/vgic.c +++ b/virt/kvm/arm/vgic.c @@ -751,7 +751,7 @@ static void vgic_dispatch_sgi(struct kvm_vcpu *vcpu, u32 reg) case 0: if (!target_cpus) return; - + break; case 1: target_cpus = ((1 << nrcpus) - 1) & ~(1 << vcpu_id) & 0xff; break; -- 1.8.4.3 -- To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html