KVM: Initialize IRQ FD after arch hardware setup

Move initialization of KVM's IRQ FD workqueue below arch hardware setup
as a step towards consolidating arch "init" and "hardware setup", and
eventually towards dropping the hooks entirely.  There is no dependency
on the workqueue being created before hardware setup, the workqueue is
used only when destroying VMs, i.e. only needs to be created before
/dev/kvm is exposed to userspace.

Move the destruction of the workqueue before the arch hooks to maintain
symmetry, and so that arch code can move away from the hooks without
having to worry about ordering changes.

Reword the comment about kvm_irqfd_init() needing to come after
kvm_arch_init() to call out that kvm_arch_init() must come before common
KVM does _anything_, as x86 very subtly relies on that behavior to deal
with multiple calls to kvm_init(), e.g. if userspace attempts to load
kvm_amd.ko and kvm_intel.ko.  Tag the code with a FIXME, as x86's subtle
requirement is gross, and invoking an arch callback as the very first
action in a helper that is called only from arch code is silly.

Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20221130230934.1014142-3-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index 28a1a02f522860fe9d6170e55ff09045fc7cfb67..6793a7e3525addf76e7ffc99846be7530566134e 100644 (file)
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -5921,24 +5921,19 @@ int kvm_init(void *opaque, unsigned vcpu_size, unsigned vcpu_align,
        int r;
        int cpu;
 
-       r = kvm_arch_init(opaque);
-       if (r)
-               goto out_fail;
-
        /*
-        * kvm_arch_init makes sure there's at most one caller
-        * for architectures that support multiple implementations,
-        * like intel and amd on x86.
-        * kvm_arch_init must be called before kvm_irqfd_init to avoid creating
-        * conflicts in case kvm is already setup for another implementation.
+        * FIXME: Get rid of kvm_arch_init(), vendor code should call arch code
+        * directly.  Note, kvm_arch_init() _must_ be called before anything
+        * else as x86 relies on checks buried in kvm_arch_init() to guard
+        * against multiple calls to kvm_init().
         */
-       r = kvm_irqfd_init();
+       r = kvm_arch_init(opaque);
        if (r)
-               goto out_irqfd;
+               return r;
 
        if (!zalloc_cpumask_var(&cpus_hardware_enabled, GFP_KERNEL)) {
                r = -ENOMEM;
-               goto out_free_0;
+               goto err_hw_enabled;
        }
 
        r = kvm_arch_hardware_setup(opaque);
@@ -5982,9 +5977,13 @@ int kvm_init(void *opaque, unsigned vcpu_size, unsigned vcpu_align,
                }
        }
 
+       r = kvm_irqfd_init();
+       if (r)
+               goto err_irqfd;
+
        r = kvm_async_pf_init();
        if (r)
-               goto out_free_4;
+               goto err_async_pf;
 
        kvm_chardev_ops.owner = module;
 
@@ -6015,6 +6014,9 @@ err_register:
        kvm_vfio_ops_exit();
 err_vfio:
        kvm_async_pf_deinit();
+err_async_pf:
+       kvm_irqfd_exit();
+err_irqfd:
 out_free_4:
        for_each_possible_cpu(cpu)
                free_cpumask_var(per_cpu(cpu_kick_mask, cpu));
@@ -6026,11 +6028,8 @@ out_free_2:
        kvm_arch_hardware_unsetup();
 out_free_1:
        free_cpumask_var(cpus_hardware_enabled);
-out_free_0:
-       kvm_irqfd_exit();
-out_irqfd:
+err_hw_enabled:
        kvm_arch_exit();
-out_fail:
        return r;
 }
 EXPORT_SYMBOL_GPL(kvm_init);
@@ -6055,9 +6054,9 @@ void kvm_exit(void)
        unregister_reboot_notifier(&kvm_reboot_notifier);
        cpuhp_remove_state_nocalls(CPUHP_AP_KVM_STARTING);
        on_each_cpu(hardware_disable_nolock, NULL, 1);
+       kvm_irqfd_exit();
        kvm_arch_hardware_unsetup();
        kvm_arch_exit();
-       kvm_irqfd_exit();
        free_cpumask_var(cpus_hardware_enabled);
        kvm_vfio_ops_exit();
 }