On Wed, 2022-11-02 at 23:19 +0000, Sean Christopherson wrote:
> From: Chao Gao <chao.gao@xxxxxxxxx>
>
> The CPU STARTING section doesn't allow callbacks to fail. Move KVM's
> hotplug callback to ONLINE section so that it can abort onlining a
> CPU in
> certain cases to avoid potentially breaking VMs running on existing
> CPUs.
> For example, when KVM fails to enable hardware virtualization on the
> hotplugged CPU.
>
> Place KVM's hotplug state before CPUHP_AP_SCHED_WAIT_EMPTY as it
> ensures
> when offlining a CPU, all user tasks and non-pinned kernel tasks have
> left
> the CPU, i.e. there cannot be a vCPU task around. So, it is safe for
> KVM's
> CPU offline callback to disable hardware virtualization at that
> point.
> Likewise, KVM's online callback can enable hardware virtualization
> before
> any vCPU task gets a chance to run on hotplugged CPUs.
>
> Rename KVM's CPU hotplug callbacks accordingly.
>
> Suggested-by: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
> Signed-off-by: Chao Gao <chao.gao@xxxxxxxxx>
> Reviewed-by: Sean Christopherson <seanjc@xxxxxxxxxx>
> Signed-off-by: Isaku Yamahata <isaku.yamahata@xxxxxxxxx>
> Reviewed-by: Yuan Yao <yuan.yao@xxxxxxxxx>
> Signed-off-by: Sean Christopherson <seanjc@xxxxxxxxxx>
> ---
> include/linux/cpuhotplug.h | 2 +-
> virt/kvm/kvm_main.c | 30 ++++++++++++++++++++++--------
> 2 files changed, 23 insertions(+), 9 deletions(-)
>
> diff --git a/include/linux/cpuhotplug.h b/include/linux/cpuhotplug.h
> index 7337414e4947..de45be38dd27 100644
> --- a/include/linux/cpuhotplug.h
> +++ b/include/linux/cpuhotplug.h
> @@ -185,7 +185,6 @@ enum cpuhp_state {
> CPUHP_AP_CSKY_TIMER_STARTING,
> CPUHP_AP_TI_GP_TIMER_STARTING,
> CPUHP_AP_HYPERV_TIMER_STARTING,
> - CPUHP_AP_KVM_STARTING,
> /* Must be the last timer callback */
> CPUHP_AP_DUMMY_TIMER_STARTING,
> CPUHP_AP_ARM_XEN_STARTING,
> @@ -200,6 +199,7 @@ enum cpuhp_state {
>
> /* Online section invoked on the hotplugged CPU from the
> hotplug thread */
> CPUHP_AP_ONLINE_IDLE,
> + CPUHP_AP_KVM_ONLINE,
> CPUHP_AP_SCHED_WAIT_EMPTY,
> CPUHP_AP_SMPBOOT_THREADS,
> CPUHP_AP_X86_VDSO_VMA_ONLINE,
> diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
> index dd13af9f06d5..fd9e39c85549 100644
> --- a/virt/kvm/kvm_main.c
> +++ b/virt/kvm/kvm_main.c
> @@ -5026,13 +5026,27 @@ static void hardware_enable_nolock(void
> *junk)
> }
> }
>
> -static int kvm_starting_cpu(unsigned int cpu)
> +static int kvm_online_cpu(unsigned int cpu)
> {
> + int ret = 0;
> +
> raw_spin_lock(&kvm_count_lock);
> - if (kvm_usage_count)
> + /*
> + * Abort the CPU online process if hardware virtualization
> cannot
> + * be enabled. Otherwise running VMs would encounter
> unrecoverable
> + * errors when scheduled to this CPU.
> + */
> + if (kvm_usage_count) {
> + WARN_ON_ONCE(atomic_read(&hardware_enable_failed));
> +
> hardware_enable_nolock(NULL);
> + if (atomic_read(&hardware_enable_failed)) {
> + atomic_set(&hardware_enable_failed, 0);
I see other places using this hardware_enable_failed with atomic_inc(),
should here use atomic_dec() instead of straightly set to 0?
Though here is embraced by spin_lock, hardware_enable_nolock() can be
invoked in other places in parallel?
Fortunately in the end of this patch set, global hardware_enable_failed
is get rid of.
> + ret = -EIO;
> + }
> + }
> raw_spin_unlock(&kvm_count_lock);
> - return 0;
> + return ret;
> }
>
> static void hardware_disable_nolock(void *junk)
> @@ -5045,7 +5059,7 @@ static void hardware_disable_nolock(void *junk)
> kvm_arch_hardware_disable();
> }
>
> -static int kvm_dying_cpu(unsigned int cpu)
> +static int kvm_offline_cpu(unsigned int cpu)
> {
> raw_spin_lock(&kvm_count_lock);
> if (kvm_usage_count)
> @@ -5822,8 +5836,8 @@ int kvm_init(unsigned vcpu_size, unsigned
> vcpu_align, struct module *module)
> if (!zalloc_cpumask_var(&cpus_hardware_enabled, GFP_KERNEL))
> return -ENOMEM;
>
> - r = cpuhp_setup_state_nocalls(CPUHP_AP_KVM_STARTING,
> "kvm/cpu:starting",
> - kvm_starting_cpu, kvm_dying_cpu);
> + r = cpuhp_setup_state_nocalls(CPUHP_AP_KVM_ONLINE,
> "kvm/cpu:online",
> + kvm_online_cpu, kvm_offline_cpu);
> if (r)
> goto out_free_2;
> register_reboot_notifier(&kvm_reboot_notifier);
> @@ -5897,7 +5911,7 @@ int kvm_init(unsigned vcpu_size, unsigned
> vcpu_align, struct module *module)
> kmem_cache_destroy(kvm_vcpu_cache);
> out_free_3:
> unregister_reboot_notifier(&kvm_reboot_notifier);
> - cpuhp_remove_state_nocalls(CPUHP_AP_KVM_STARTING);
> + cpuhp_remove_state_nocalls(CPUHP_AP_KVM_ONLINE);
> out_free_2:
> free_cpumask_var(cpus_hardware_enabled);
> return r;
> @@ -5923,7 +5937,7 @@ void kvm_exit(void)
> kvm_async_pf_deinit();
> unregister_syscore_ops(&kvm_syscore_ops);
> unregister_reboot_notifier(&kvm_reboot_notifier);
> - cpuhp_remove_state_nocalls(CPUHP_AP_KVM_STARTING);
> + cpuhp_remove_state_nocalls(CPUHP_AP_KVM_ONLINE);
> on_each_cpu(hardware_disable_nolock, NULL, 1);
> kvm_irqfd_exit();
> free_cpumask_var(cpus_hardware_enabled);
