On 16.01.20 13:26, Thomas Huth wrote: > On 16/01/2020 13.23, David Hildenbrand wrote: >> On 16.01.20 13:20, Thomas Huth wrote: >>> The AIS feature has been disabled late in the v2.10 development >>> cycle since there were some issues with migration (see commit >>> 3f2d07b3b01ea61126b - "s390x/ais: for 2.10 stable: disable ais >>> facility"). We originally wanted to enable it again for newer >>> machine types, but apparently we forgot to do this so far. Let's >>> do it for the new s390-ccw-virtio-5.0 machine now. >>> >>> Buglink: https://bugzilla.redhat.com/show_bug.cgi?id=1756946 >>> Signed-off-by: Thomas Huth <th...@redhat.com> >>> --- >>> hw/s390x/s390-virtio-ccw.c | 4 ++++ >>> include/hw/s390x/s390-virtio-ccw.h | 4 ++++ >>> target/s390x/kvm.c | 11 ++++++++--- >>> 3 files changed, 16 insertions(+), 3 deletions(-) >>> >>> diff --git a/hw/s390x/s390-virtio-ccw.c b/hw/s390x/s390-virtio-ccw.c >>> index e7eadd14e8..6f43136396 100644 >>> --- a/hw/s390x/s390-virtio-ccw.c >>> +++ b/hw/s390x/s390-virtio-ccw.c >>> @@ -456,6 +456,7 @@ static void ccw_machine_class_init(ObjectClass *oc, >>> void *data) >>> s390mc->cpu_model_allowed = true; >>> s390mc->css_migration_enabled = true; >>> s390mc->hpage_1m_allowed = true; >>> + s390mc->kvm_ais_allowed = true; >>> mc->init = ccw_init; >>> mc->reset = s390_machine_reset; >>> mc->hot_add_cpu = s390_hot_add_cpu; >>> @@ -662,6 +663,9 @@ static void >>> ccw_machine_4_2_instance_options(MachineState *machine) >>> >>> static void ccw_machine_4_2_class_options(MachineClass *mc) >>> { >>> + S390CcwMachineClass *s390mc = S390_MACHINE_CLASS(mc); >>> + >>> + s390mc->kvm_ais_allowed = false; >>> ccw_machine_5_0_class_options(mc); >>> compat_props_add(mc->compat_props, hw_compat_4_2, hw_compat_4_2_len); >>> } >>> diff --git a/include/hw/s390x/s390-virtio-ccw.h >>> b/include/hw/s390x/s390-virtio-ccw.h >>> index 8aa27199c9..f142d379c6 100644 >>> --- a/include/hw/s390x/s390-virtio-ccw.h >>> +++ b/include/hw/s390x/s390-virtio-ccw.h >>> @@ -21,6 +21,9 @@ >>> #define S390_MACHINE_CLASS(klass) \ >>> OBJECT_CLASS_CHECK(S390CcwMachineClass, (klass), TYPE_S390_CCW_MACHINE) >>> >>> +#define S390_CCW_MACHINE_OBJ_GET_CLASS(obj) \ >>> + OBJECT_GET_CLASS(S390CcwMachineClass, obj, TYPE_S390_CCW_MACHINE) >>> + >>> typedef struct S390CcwMachineState { >>> /*< private >*/ >>> MachineState parent_obj; >>> @@ -40,6 +43,7 @@ typedef struct S390CcwMachineClass { >>> bool cpu_model_allowed; >>> bool css_migration_enabled; >>> bool hpage_1m_allowed; >>> + bool kvm_ais_allowed; >>> } S390CcwMachineClass; >>> >>> /* runtime-instrumentation allowed by the machine */ >>> diff --git a/target/s390x/kvm.c b/target/s390x/kvm.c >>> index 15260aeb9a..4c1c8c0208 100644 >>> --- a/target/s390x/kvm.c >>> +++ b/target/s390x/kvm.c >>> @@ -329,6 +329,8 @@ static void ccw_machine_class_foreach(ObjectClass *oc, >>> void *opaque) >>> >>> int kvm_arch_init(MachineState *ms, KVMState *s) >>> { >>> + S390CcwMachineClass *smc = S390_CCW_MACHINE_OBJ_GET_CLASS(ms); >>> + >>> object_class_foreach(ccw_machine_class_foreach, TYPE_S390_CCW_MACHINE, >>> false, NULL); >>> >>> @@ -365,10 +367,13 @@ int kvm_arch_init(MachineState *ms, KVMState *s) >>> /* >>> * The migration interface for ais was introduced with kernel 4.13 >>> * but the capability itself had been active since 4.12. As migration >>> - * support is considered necessary let's disable ais in the 2.10 >>> - * machine. >>> + * support is considered necessary we only enable this for newer >>> + * machine types and if KVM_CAP_S390_AIS_MIGRATION is available. >>> */ >>> - /* kvm_vm_enable_cap(s, KVM_CAP_S390_AIS, 0); */ >>> + if (smc->kvm_ais_allowed && >>> + kvm_check_extension(s, KVM_CAP_S390_AIS_MIGRATION)) { >>> + kvm_vm_enable_cap(s, KVM_CAP_S390_AIS, 0); >>> + } >>> >>> kvm_set_max_memslot_size(KVM_SLOT_MAX_BYTES); >>> return 0; >>> >> >> We have ri_allowed(), cpu_model_allowed(), hpage_1m_allowed(). >> >> Care to create a similar wrapper? > > Honestly, why do we need these wrappers at all? They look cumbersome to > me. I'd rather remove them in case they are not urgently needed (so far > I don't see the point... could someone enlighten me why we have them?).
I assume to minimize the number of places you have to lookup the machine/machine class. -- Thanks, David / dhildenb