As the kernel has no way of disallowing the start of a huge page
backed VM, we can migrate a running huge backed VM to a host that has
no huge page KVM support.
Let's glue huge page support support to the 3.1 machine, so we do not
migrate to a destination host that doesn't have QEMU huge page support
and can stop migration if KVM doesn't indicate support.
Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Message-Id: <20180928093435.198573-1-frankja@linux.ibm.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Cornelia Huck <cohuck@redhat.com>
This option has been deprecated for two releases; remove it.
Acked-by: Christian Borntraeger <borntraeger@de.ibm.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Acked-by: Halil Pasic <pasic@linux.ibm.com>
Signed-off-by: Cornelia Huck <cohuck@redhat.com>
Calling pause_all_vcpus()/resume_all_vcpus() from a VCPU thread might
not be the best idea. As pause_all_vcpus() temporarily drops the qemu
mutex, two parallel calls to pause_all_vcpus() can be active at a time,
resulting in a deadlock. (either by two VCPUs or by the main thread and a
VCPU)
Let's handle it via the main loop instead, as suggested by Paolo. If we
would have two parallel reset requests by two different VCPUs at the
same time, the last one would win.
We use the existing ipl device to handle it. The nice side effect is
that we can get rid of reipl_requested.
This change implies that all reset handling now goes via the common
path, so "no-reboot" handling is now active for all kinds of reboots.
Let's execute any CPU initialization code on the target CPU using
run_on_cpu.
Signed-off-by: David Hildenbrand <david@redhat.com>
Message-Id: <20180424101859.10239-1-david@redhat.com>
Acked-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Cornelia Huck <cohuck@redhat.com>
Starting a guest with
<os>
<type arch='s390x' machine='s390-ccw-virtio-2.9'>hvm</type>
</os>
<cpu mode='host-model'/>
on an IBM z14 results in
"qemu-system-s390x: Some features requested in the CPU model are not
available in the configuration: gs"
This is because guarded storage is fenced for compat machines that did
not have guarded storage support. While this prevents future migration
abort (by not starting the guest at all), not being able to start a
"host-model" guest is very much unexpected. As it turns out, even if we
would modify libvirt to not expand the cpu model to contain "gs" for
compat machines, it cannot guarantee that a migration will succeed. For
example if the kernel changes its features (or the user has nested=1 on
one host but not on the other) the migration will fail nevertheless. So
instead of fencing "gs" for machines <= 2.9 lets allow it for all
machine types that support the CPU model. This will make "host-model"
runnable all the time, while relying on the CPU model to reject invalid
migration attempts. We also need to change the migration for guarded
storage.
Additional discussions about host-model are still pending but are out
of scope of this patch.
Suggested-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Acked-by: David Hildenbrand <david@redhat.com>
Acked-by: Cornelia Huck <Cornelia Huck <cohuck@redhat.com>
Acked-by: Halil Pasic <pasic@linux.vnet.ibm.com>
Implemented in s390-virtio-ccw.c, so move it to the right header.
We can also drop the extern. Fix up one include.
Reviewed-by: Thomas Huth <thuth@redhat.com>
Acked-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: David Hildenbrand <david@redhat.com>
Message-Id: <20170913132417.24384-7-david@redhat.com>
Signed-off-by: Cornelia Huck <cohuck@redhat.com>
Introduce guarded storage support for KVM guests on s390.
We need to enable the capability, extend machine check validity,
sigp store-additional-status-at-address, and migration.
The feature is fenced for older machine type versions.
Signed-off-by: Fan Zhang <zhangfan@linux.vnet.ibm.com>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Currently the migration of the channel subsystem (css) is only partial
and is done by the virtio ccw proxies -- the only migratable css devices
existing at the moment.
With the current work on emulated and passthrough devices we need to
decouple the migration of the channel subsystem state from virtio ccw,
and have a separate section for it. A new section however necessarily
breaks the migration compatibility.
So let us introduce a switch at the machine class, and put it in 'off'
state for now. We will turn the switch 'on' for future machines once all
preparations are met. For compatibility machines the switch will stay
'off'.
Signed-off-by: Halil Pasic <pasic@linux.vnet.ibm.com>
Acked-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Message-Id: <20170711145441.33925-3-pasic@linux.vnet.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
We want to support real (i.e. not virtual) channel devices
even for guests that do not support MCSS-E (where guests may
see devices from any channel subsystem image at once). As all
virtio-ccw devices are in css 0xfe (and show up in the default
css 0 for guests not activating MCSS-E), we need an option to
squash both the virtio subchannels and e.g. passed-through
subchannels from their real css (0-3, or 0 for hosts not
activating MCSS-E) into the default css. This will be
exploited in a later patch.
Signed-off-by: Xiao Feng Ren <renxiaof@linux.vnet.ibm.com>
Signed-off-by: Dong Jia Shi <bjsdjshi@linux.vnet.ibm.com>
Message-Id: <20170517004813.58227-4-bjsdjshi@linux.vnet.ibm.com>
Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
In order to specify the LOADPARM value one may now add ",loadparm=xxx"
parameter to the "-machine s390-ccw-virtio" option.
The property setter will normalize and check the value provided much
like the way the HMC does.
The value is stored, but not used at the moment.
Initial patch from Eugene (jno) Dvurechenski.
Signed-off-by: Eugene (jno) Dvurechenski <jno@linux.vnet.ibm.com>
Signed-off-by: Farhan Ali <alifm@linux.vnet.ibm.com>
Reviewed-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
cpu model was merged with 2.8, it is wrong to abuse ri_allowed which
was enabled with 2.7.
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Introduce run-time-instrumentation support when running under kvm for
virtio-ccw 2.7 machine and make sure older machines can not enable it.
The new ri_allowed field in the s390MachineClass serves as an indicator
whether the feature can be used by the machine and should therefore be
activated if available.
riccb_needed() is used to check whether riccb is needed or not in live
migration.
Signed-off-by: Fan Zhang <zhangfan@linux.vnet.ibm.com>
Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
As we now have the new machine definitions, that let us disable/enable
machine options more easily, we need a way to save them and make them
publicly available.
The new s390-virtio-ccw.h header exports the s390 ccw machine state
and class, so they can be easily used in other C files.
Signed-off-by: Janosch Frank <frankja@linux.vnet.ibm.com>
Acked-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>