Fixes higher-half address parsing for QMP commands
`[p]memsave`.
Signed-off-by: Josh Junon <junon@oro.sh>
Message-ID: <20240802140704.13591-1-junon@oro.sh>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
[Subject tweaked, and one PRId64 updated to PRIu64]
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Sweep the entire documentation again. Last done in commit
209e64d9ed (qapi: Refill doc comments to conform to current
conventions).
To check the generated documentation does not change, I compared the
generated HTML before and after this commit with "wdiff -3". Finds no
differences. Comparing with diff is not useful, as the reflown
paragraphs are visible there.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-ID: <20240729065220.860163-1-armbru@redhat.com>
[Straightforward conflict with commit 442110bc6f resolved]
CpuModelInfo is used both as command argument and in command
returns.
Its @deprecated-props array does not make any sense in arguments,
and is silently ignored. We actually want it only as return value
of query-cpu-model-expansion.
Move it from CpuModelInfo to CpuModelExpansionType, and document
its dependence on expansion type property.
This was identified late during review [1] and we have to fix it up
while it's not part of an official QEMU release yet.
[1] https://lore.kernel.org/qemu-devel/20240719181741.35146-1-walling@linux.ibm.com/
Message-ID: <20240726203646.20279-1-walling@linux.ibm.com>
Fixes: eed0e8ffa3 ("target/s390x: filter deprecated properties based on model expansion type")
Signed-off-by: Collin Walling <walling@linux.ibm.com>
[ david: - add "Fixes", adjust description, reference v3 instead
- make property s390x-only and non-optional
- fixup "populate" vs. "populated" ]
Signed-off-by: David Hildenbrand <david@redhat.com>
Some QOM properties are associated with ObjectTypes that already
depend on CONFIG_* switches. So to avoid generating dead code,
let's also make the definition of those properties dependent on
the corresponding CONFIG_*.
Suggested-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
Message-ID: <20240604135931.311709-1-sgarzare@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
[Make SecretKeyringProperties conditional, too]
Signed-off-by: Markus Armbruster <armbru@redhat.com>
The 'detached-header' field in QCryptoBlockCreateOptionsLUKS
was left over from earlier patch iterations.
Acked-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Currently, there is no way to execute the query-cpu-model-expansion
command to retrieve a comprehenisve list of deprecated properties, as
the result is dependent per-model. To enable this, the expansion output
is modified as such:
When reporting a "full" CPU model, show the *entire* list of deprecated
properties regardless if they are supported on the model. A full
expansion outputs all known CPU model properties anyway, so it makes
sense to report all deprecated properties here too.
This allows management apps to query a single model (e.g. host) to
acquire the full list of deprecated properties.
Additionally, when reporting a "static" CPU model, the command will
only show deprecated properties that are a subset of the model's
*enabled* properties. This is more accurate than how the query was
handled before, which blindly reported deprecated properties that
were never otherwise introduced for certain models.
Acked-by: David Hildenbrand <david@redhat.com>
Suggested-by: Jiri Denemark <jdenemar@redhat.com>
Signed-off-by: Collin Walling <walling@linux.ibm.com>
Message-ID: <20240719181741.35146-1-walling@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
These examples require longer explanations or have explanations that
require markup to look reasonable when rendered and so use the longer
form of the ".. qmp-example::" directive.
By using the :annotated: option, the content in the example block is
assumed *not* to be a code block literal and is instead parsed as normal
rST - with the exception that any code literal blocks after `::` will
assumed to be a QMP code literal block.
Note: There's one title-less conversion in this patch that comes along
for the ride because it's part of a larger "Examples" block that was
better to convert all at once.
See commit-5: "docs/qapidoc: create qmp-example directive", for a
detailed explanation of this custom directive syntax.
See commit+1: "qapi: remove "Example" doc section" for a detailed
explanation of why.
Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Message-ID: <20240717021312.606116-9-jsnow@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
When an Example section has a brief explanation, convert it to a
qmp-example:: section using the :title: option.
Rule of thumb: If the title can fit on a single line and requires no rST
markup, it's a good candidate for using the :title: option of
qmp-example.
In this patch, trailing punctuation is removed from the title section
for consistent headline aesthetics. In just one case, specifics of the
example are removed to make the title read better.
See commit-4: "docs/qapidoc: create qmp-example directive", for a
detailed explanation of this custom directive syntax.
See commit+2: "qapi: remove "Example" doc section" for a detailed
explanation of why.
Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Message-ID: <20240717021312.606116-8-jsnow@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Use the no-option form of ".. qmp-example::" to convert any Examples
that do not have any form of caption or explanation whatsoever. Note
that in a few cases, example sections are split into two or more
separate example blocks. This is only done stylistically to create a
delineation between two or more logically independent examples.
See commit-3: "docs/qapidoc: create qmp-example directive", for a
detailed explanation of this custom directive syntax.
See commit+3: "qapi: remove "Example" doc section" for a detailed
explanation of why.
Note: an empty "TODO" line was added to announce-self to keep the
example from floating up into the body; this will be addressed more
rigorously in the new qapidoc generator.
Signed-off-by: John Snow <jsnow@redhat.com>
Message-ID: <20240717021312.606116-7-jsnow@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
[Markup fixed in one place]
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Doc comments are reference documentation for users of QMP.
SpiceQueryMouseMode's doc comment contains a note explaining why it's
not named SpiceMouseMode: spice/enums.h has it already. Irrelevant
for users of QMP; delete the note.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-ID: <20240711112228.2140606-6-armbru@redhat.com>
Reviewed-by: John Snow <jsnow@redhat.com>
Doc comments are reference documentation for users of QMP.
SocketAddress's doc comment contains a deprecation note advising
developers to use SocketAddress for new code. Irrelevant for users of
QMP. Move the note out of the doc comment.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-ID: <20240711112228.2140606-5-armbru@redhat.com>
Reviewed-by: John Snow <jsnow@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
When no UUID has been specified, query-uuid returns
{"UUID": "00000000-0000-0000-0000-000000000000"}
The doc comment calls this "a null UUID", which I find less than
clear. RFC 9562 calls it "the nil UUID (all zeroes)", so use that
instead.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-ID: <20240711112228.2140606-4-armbru@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: John Snow <jsnow@redhat.com>
[Wording improved, commit message adjusted]
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
CpuInstanceProperties' doc comment describes its members as properties
to be passed to device_add when hot-plugging a CPU.
This was in fact the initial use of this type, with
query-hotpluggable-cpus: letting management applications find out what
properties need to be passed with device_add to hot-plug a CPU.
We've since added other uses: set-numa-node (commit 419fcdec3c and
f3be67812c), and query-cpus-fast (commit ce74ee3dea). These are not
about device-add.
query-hotpluggable-cpus uses CpuInstanceProperties within
HotpluggableCPU. Lift the documentation related to device-add from
CpuInstanceProperties to HotpluggableCPU.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-ID: <20240711112228.2140606-3-armbru@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: John Snow <jsnow@redhat.com>
PciDeviceInfo's doc comment has a note on PciDeviceClass member @desc.
Since the note applies always, not just within PciDeviceInfo, merge it
into PciDeviceClass's description of member @desc.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-ID: <20240711112228.2140606-2-armbru@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: John Snow <jsnow@redhat.com>
Commit 8f9a9259d3 added ObjectType member @x-vfio-user-server with
feature unstable, but neglected to explain why it is unstable. Do
that now.
Fixes: 8f9a9259d3 (vfio-user: define vfio-user-server object)
Cc: Elena Ufimtseva <elena.ufimtseva@oracle.com>
Cc: John G Johnson <john.g.johnson@oracle.com>
Cc: Jagannathan Raman <jag.raman@oracle.com>
Cc: qemu-stable@nongnu.org
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-ID: <20240703095310.1242102-1-armbru@redhat.com>
Reviewed-by: John Snow <jsnow@redhat.com>
[Indentation fixed]
Currently if the 'legacy-vm-type' property of the sev-guest object is
'on', QEMU will attempt to use the newer KVM_SEV_INIT2 kernel
interface in conjunction with the newer KVM_X86_SEV_VM and
KVM_X86_SEV_ES_VM KVM VM types.
This can lead to measurement changes if, for instance, an SEV guest was
created on a host that originally had an older kernel that didn't
support KVM_SEV_INIT2, but is booted on the same host later on after the
host kernel was upgraded.
Instead, if legacy-vm-type is 'off', QEMU should fail if the
KVM_SEV_INIT2 interface is not provided by the current host kernel.
Modify the fallback handling accordingly.
In the future, VMSA features and other flags might be added to QEMU
which will require legacy-vm-type to be 'off' because they will rely
on the newer KVM_SEV_INIT2 interface. It may be difficult to convey to
users what values of legacy-vm-type are compatible with which
features/options, so as part of this rework, switch legacy-vm-type to a
tri-state OnOffAuto option. 'auto' in this case will automatically
switch to using the newer KVM_SEV_INIT2, but only if it is required to
make use of new VMSA features or other options only available via
KVM_SEV_INIT2.
Defining 'auto' in this way would avoid inadvertantly breaking
compatibility with older kernels since it would only be used in cases
where users opt into newer features that are only available via
KVM_SEV_INIT2 and newer kernels, and provide better default behavior
than the legacy-vm-type=off behavior that was previously in place, so
make it the default for 9.1+ machine types.
Cc: Daniel P. Berrangé <berrange@redhat.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>
cc: kvm@vger.kernel.org
Signed-off-by: Michael Roth <michael.roth@amd.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Link: https://lore.kernel.org/r/20240710041005.83720-1-michael.roth@amd.com
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
When a command's arguments are specified as an explicit type T,
generated documentation points to the members of T.
Example:
##
# @announce-self:
#
# Trigger generation of broadcast RARP frames to update network
[...]
##
{ 'command': 'announce-self', 'boxed': true,
'data' : 'AnnounceParameters'}
generates
"announce-self" (Command)
-------------------------
Trigger generation of broadcast RARP frames to update network
[...]
Arguments
~~~~~~~~~
The members of "AnnounceParameters"
Except when the command takes its arguments unboxed , i.e. it doesn't
have 'boxed': true, we generate *nothing*. A few commands have a
reference in their doc comment to compensate, but most don't.
Example:
##
# @blockdev-snapshot-sync:
#
# Takes a synchronous snapshot of a block device.
#
# For the arguments, see the documentation of BlockdevSnapshotSync.
[...]
##
{ 'command': 'blockdev-snapshot-sync',
'data': 'BlockdevSnapshotSync',
'allow-preconfig': true }
generates
"blockdev-snapshot-sync" (Command)
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Takes a synchronous snapshot of a block device.
For the arguments, see the documentation of BlockdevSnapshotSync.
[...]
Same for event data.
Fix qapidoc.py to generate the reference regardless of boxing. Delete
now redundant references in the doc comments.
Fixes: 4078ee5469 (docs/sphinx: Add new qapi-doc Sphinx extension)
Cc: qemu-stable@nongnu.org
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-ID: <20240628112756.794237-1-armbru@redhat.com>
Reviewed-by: John Snow <jsnow@redhat.com>
Generally, surround command-line options with ``literal`` markup to help
it stand out from prose in rendered HTML, and add cross-references to
replace "see also" messages.
References to types, values, and other QAPI definitions are not yet
adjusted here; they will be converted en masse in a subsequent patch
after the new QAPI doc generator is merged.
Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Message-ID: <20240626222128.406106-13-jsnow@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Where I've noticed, rephrase the note to read more fluently.
Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Message-ID: <20240626222128.406106-12-jsnow@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
We do not need a dedicated section for notes. By eliminating a specially
parsed section, these notes can be treated as normal rST paragraphs in
the new QMP reference manual, and can be placed and styled much more
flexibly.
Convert all existing "Note" and "Notes" sections to pure rST. As part of
the conversion, capitalize the first letter of each sentence and add
trailing punctuation where appropriate to ensure notes look sensible and
consistent in rendered HTML documentation. Markup is also re-aligned to
the de-facto standard of 3 spaces for directives.
Update docs/devel/qapi-code-gen.rst to reflect the new paradigm, and
update the QAPI parser to prohibit "Note" sections while suggesting a
new syntax. The exact formatting to use is a matter of taste, but a good
candidate is simply:
.. note:: lorem ipsum ...
... dolor sit amet ...
... consectetur adipiscing elit ...
... but there are other choices, too. The Sphinx readthedocs theme
offers theming for the following forms (capitalization unimportant); all
are adorned with a (!) symbol () in the title bar for rendered HTML
docs.
See
https://sphinx-rtd-theme.readthedocs.io/en/stable/demo/demo.html#admonitions
for examples of each directive/admonition in use.
These are rendered in orange:
.. Attention:: ...
.. Caution:: ...
.. WARNING:: ...
These are rendered in red:
.. DANGER:: ...
.. Error:: ...
These are rendered in green:
.. Hint:: ...
.. Important:: ...
.. Tip:: ...
These are rendered in blue:
.. Note:: ...
.. admonition:: custom title
admonition body text
This patch uses ".. note::" almost everywhere, with just two "caution"
directives. Several instances of "Notes:" have been converted to
merely ".. note::", or multiple ".. note::" where appropriate.
".. admonition:: notes" is used in a few places where we had an
ordered list of multiple notes that would not make sense as
standalone/separate admonitions. Two "Note:" following "Example:"
have been turned into ordinary paragraphs within the example.
NOTE: Because qapidoc.py does not attempt to preserve source ordering of
sections, the conversion of Notes from a "tagged section" to an
"untagged section" means that rendering order for some notes *may
change* as a result of this patch. The forthcoming qapidoc.py rewrite
strictly preserves source ordering in the rendered documentation, so
this issue will be rectified in the new generator.
Signed-off-by: John Snow <jsnow@redhat.com>
Acked-by: Stefan Hajnoczi <stefanha@redhat.com> [for block*.json]
Message-ID: <20240626222128.406106-11-jsnow@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
[Commit message clarified slightly, period added to one more note]
Signed-off-by: Markus Armbruster <armbru@redhat.com>
By unstated convention, Errors sections are rST lists. Document the
convention, and make the one exception conform.
Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Message-ID: <20240626222128.406106-10-jsnow@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
The new QMP documentation generator wants to parse all examples as
"QMP". We have an existing QMP lexer in docs/sphinx/qmp_lexer.py (Seen
in-use here: https://qemu-project.gitlab.io/qemu/interop/bitmaps.html)
that allows the use of "->", "<-" and "..." tokens to denote QMP
protocol flow with elisions, but otherwise defers to the JSON lexer.
To utilize this lexer for the existing QAPI documentation, we need them
to conform to a standard so that they lex and render correctly. Once the
QMP lexer is active for examples, errant QMP/JSON will produce warning
messages and fail the build.
Fix any invalid JSON found in QAPI documentation (identified by
attempting to lex all examples as QMP; see subsequent
commits). Additionally, elisions must be standardized for the QMP lexer;
they must be represented as the value "...", so three examples have been
adjusted to support that format here.
Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Message-ID: <20240626222128.406106-9-jsnow@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Markus suggested that we make the unstable. I don't expect these
interfaces to change because of their tight coupling to the Compute
Express Link (CXL) Specification, Revision 3.1 Fabric Management API
definitions which can only be extended in backwards compatible way.
However, there seems little disadvantage in taking a cautious path
for now and marking them as unstable interfaces.
Suggested-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Message-Id: <20240625170805.359278-3-Jonathan.Cameron@huawei.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
New DCD command definitions updated in response to review comments
from Markus.
- Used CxlXXXX instead of CXLXXXXX for newly added types.
- Expanded some abreviations in type names to be easier to read.
- Additional documentation for some fields.
- Replace slightly vague cxl r3.1 references with
"Compute Express Link (CXL) Specification, Revision 3.1, XXXX"
to bring them inline with what it says on the specification cover.
Suggested-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Message-Id: <20240625170805.359278-2-Jonathan.Cameron@huawei.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
shm_open() creates and opens a new POSIX shared memory object.
A POSIX shared memory object allows creating memory backend with an
associated file descriptor that can be shared with external processes
(e.g. vhost-user).
The new `memory-backend-shm` can be used as an alternative when
`memory-backend-memfd` is not available (Linux only), since shm_open()
should be provided by any POSIX-compliant operating system.
This backend mimics memfd, allocating memory that is practically
anonymous. In theory shm_open() requires a name, but this is allocated
for a short time interval and shm_unlink() is called right after
shm_open(). After that, only fd is shared with external processes
(e.g., vhost-user) as if it were associated with anonymous memory.
In the future we may also allow the user to specify the name to be
passed to shm_open(), but for now we keep the backend simple, mimicking
anonymous memory such as memfd.
Acked-by: David Hildenbrand <david@redhat.com>
Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
Acked-by: Markus Armbruster <armbru@redhat.com> (QAPI schema)
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
Message-Id: <20240618100519.145853-1-sgarzare@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
The default value of the @share option of the @MemoryBackendProperties
really depends on the backend type, so let's document the default
values in the same place where we define the option to avoid
dispersing the information.
Cc: David Hildenbrand <david@redhat.com>
Suggested-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
Message-Id: <20240618100043.144657-2-sgarzare@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Emit a QMP event on receiving a PVPANIC_SHUTDOWN event. Even though a typical
SHUTDOWN event will be sent, it will be indistinguishable from a shutdown
originating from other cases (e.g. KVM exit due to KVM_SYSTEM_EVENT_SHUTDOWN)
that also issue the guest-shutdown cause.
A management layer application can detect the new GUEST_PVSHUTDOWN event to
determine if the guest is using the pvpanic interface to request shutdowns.
Signed-off-by: Alejandro Jimenez <alejandro.j.jimenez@oracle.com>
Message-Id: <20240527-pvpanic-shutdown-v8-6-5a28ec02558b@t-8ch.de>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
To simulate FM functionalities for initiating Dynamic Capacity Add
(Opcode 5604h) and Dynamic Capacity Release (Opcode 5605h) as in CXL spec
r3.1 7.6.7.6.5 and 7.6.7.6.6, we implemented two QMP interfaces to issue
add/release dynamic capacity extents requests.
With the change, we allow to release an extent only when its DPA range
is contained by a single accepted extent in the device. That is to say,
extent superset release is not supported yet.
1. Add dynamic capacity extents:
For example, the command to add two continuous extents (each 128MiB long)
to region 0 (starting at DPA offset 0) looks like below:
{ "execute": "qmp_capabilities" }
{ "execute": "cxl-add-dynamic-capacity",
"arguments": {
"path": "/machine/peripheral/cxl-dcd0",
"host-id": 0,
"selection-policy": "prescriptive",
"region": 0,
"extents": [
{
"offset": 0,
"len": 134217728
},
{
"offset": 134217728,
"len": 134217728
}
]
}
}
2. Release dynamic capacity extents:
For example, the command to release an extent of size 128MiB from region 0
(DPA offset 128MiB) looks like below:
{ "execute": "cxl-release-dynamic-capacity",
"arguments": {
"path": "/machine/peripheral/cxl-dcd0",
"host-id": 0,
"removal-policy":"prescriptive",
"region": 0,
"extents": [
{
"offset": 134217728,
"len": 134217728
}
]
}
}
Tested-by: Svetly Todorov <svetly.todorov@memverge.com>
Reviewed-by: Gregory Price <gregory.price@memverge.com>
Signed-off-by: Fan Ni <fan.ni@samsung.com>
Message-Id: <20240523174651.1089554-12-nifan.cxl@gmail.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
This patch adds a migration state on src called "postcopy-recover-setup".
The new state will describe the intermediate step starting from when the
src QEMU received a postcopy recovery request, until the migration channels
are properly established, but before the recovery process take place.
The request came from Libvirt where Libvirt currently rely on the migration
state events to detect migration state changes. That works for most of the
migration process but except postcopy recovery failures at the beginning.
Currently postcopy recovery only has two major states:
- postcopy-paused: this is the state that both sides of QEMU will be in
for a long time as long as the migration channel was interrupted.
- postcopy-recover: this is the state where both sides of QEMU handshake
with each other, preparing for a continuation of postcopy which used to
be interrupted.
The issue here is when the recovery port is invalid, the src QEMU will take
the URI/channels, noticing the ports are not valid, and it'll silently keep
in the postcopy-paused state, with no event sent to Libvirt. In this case,
the only thing Libvirt can do is to poll the migration status with a proper
interval, however that's less optimal.
Considering that this is the only case where Libvirt won't get a
notification from QEMU on such events, let's add postcopy-recover-setup
state to mimic what we have with the "setup" state of a newly initialized
migration, describing the phase of connection establishment.
With that, postcopy recovery will have two paths to go now, and either path
will guarantee an event generated. Now the events will look like this
during a recovery process on src QEMU:
- Initially when the recovery is initiated on src, QEMU will go from
"postcopy-paused" -> "postcopy-recover-setup". Old QEMUs don't have
this event.
- Depending on whether the channel re-establishment is succeeded:
- In succeeded case, src QEMU will move from "postcopy-recover-setup"
to "postcopy-recover". Old QEMUs also have this event.
- In failure case, src QEMU will move from "postcopy-recover-setup" to
"postcopy-paused" again. Old QEMUs don't have this event.
This guarantees that Libvirt will always receive a notification for
recovery process properly.
One thing to mention is, such new status is only needed on src QEMU not
both. On dest QEMU, the state machine doesn't change. Hence the events
don't change either. It's done like so because dest QEMU may not have an
explicit point of setup start. E.g., it can happen that when dest QEMUs
doesn't use migrate-recover command to use a new URI/channel, but the old
URI/channels can be reused in recovery, in which case the old ports simply
can work again after the network routes are fixed up.
Add a new helper postcopy_is_paused() detecting whether postcopy is still
paused, taking RECOVER_SETUP into account too. When using it on both
src/dst, a slight change is done altogether to always wait for the
semaphore before checking the status, because for both sides a sem_post()
will be required for a recovery.
Cc: Jiri Denemark <jdenemar@redhat.com>
Cc: Prasad Pandit <ppandit@redhat.com>
Reviewed-by: Fabiano Rosas <farosas@suse.de>
Buglink: https://issues.redhat.com/browse/RHEL-38485
Signed-off-by: Peter Xu <peterx@redhat.com>
Signed-off-by: Fabiano Rosas <farosas@suse.de>
Add the direct-io migration parameter that tells the migration code to
use O_DIRECT when opening the migration stream file whenever possible.
This is currently only used with the mapped-ram migration that has a
clear window guaranteed to perform aligned writes.
Acked-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
Signed-off-by: Fabiano Rosas <farosas@suse.de>
This is a counterpart to the HMP "info pic" command. It is being
added with an "x-" prefix because this QMP command is intended as an
adhoc debugging tool and will thus not be modelled in QAPI as fully
structured data, nor will it have long term guaranteed stability.
The existing HMP command is rewritten to call the QMP command.
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20240610063518.50680-3-philmd@linaro.org>
add the Query Processing Library (QPL) compression method
Introduce the qpl as a new multifd migration compression method, it can
use In-Memory Analytics Accelerator(IAA) to accelerate compression and
decompression, which can not only reduce network bandwidth requirement
but also reduce host compression and decompression CPU overhead.
How to enable qpl compression during migration:
migrate_set_parameter multifd-compression qpl
There is no qpl compression level parameter added since it only supports
level one, users do not need to specify the qpl compression level.
Signed-off-by: Yuan Liu <yuan1.liu@intel.com>
Reviewed-by: Nanhai Zou <nanhai.zou@intel.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Fabiano Rosas <farosas@suse.de>
[fixed docs spacing in migration.json]
Signed-off-by: Fabiano Rosas <farosas@suse.de>
Commit 1f25c172f8 ("monitor: use aio_co_reschedule_self()") was a code
cleanup that uses aio_co_reschedule_self() instead of open coding
coroutine rescheduling.
Bug RHEL-34618 was reported and Kevin Wolf <kwolf@redhat.com> identified
the root cause. I missed that aio_co_reschedule_self() ->
qemu_get_current_aio_context() only knows about
qemu_aio_context/IOThread AioContexts and not about iohandler_ctx. It
does not function correctly when going back from the iohandler_ctx to
qemu_aio_context.
Go back to open coding the AioContext transitions to avoid this bug.
This reverts commit 1f25c172f8.
Cc: qemu-stable@nongnu.org
Buglink: https://issues.redhat.com/browse/RHEL-34618
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-ID: <20240506190622.56095-2-stefanha@redhat.com>
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
* require x86-64-v2 baseline ISA
* SEV-SNP host support
* fix xsave.flat with TCG
* fixes for CPUID checks done by TCG
-----BEGIN PGP SIGNATURE-----
iQFIBAABCAAyFiEE8TM4V0tmI4mGbHaCv/vSX3jHroMFAmZgKVYUHHBib256aW5p
QHJlZGhhdC5jb20ACgkQv/vSX3jHroPKYgf/QkWrNXdjjD3yAsv5LbJFVTVyCYW3
b4Iax29kEDy8k9wbzfLxOfIk9jXIjmbOMO5ZN9LFiHK6VJxbXslsMh6hm50M3xKe
49X1Rvf9YuVA7KZX+dWkEuqLYI6Tlgj3HaCilYWfXrjyo6hY3CxzkPV/ChmaeYlV
Ad4Y8biifoUuuEK8OTeTlcDWLhOHlFXylG3AXqULsUsXp0XhWJ9juXQ60eATv/W4
eCEH7CSmRhYFu2/rV+IrWFYMnskLRTk1OC1/m6yXGPKOzgnOcthuvQfiUgPkbR/d
llY6Ni5Aaf7+XX3S7Avcyvoq8jXzaaMzOrzL98rxYGDR1sYBYO+4h4ZToA==
=qQeP
-----END PGP SIGNATURE-----
Merge tag 'for-upstream' of https://gitlab.com/bonzini/qemu into staging
* virtio-blk: remove SCSI passthrough functionality
* require x86-64-v2 baseline ISA
* SEV-SNP host support
* fix xsave.flat with TCG
* fixes for CPUID checks done by TCG
# -----BEGIN PGP SIGNATURE-----
#
# iQFIBAABCAAyFiEE8TM4V0tmI4mGbHaCv/vSX3jHroMFAmZgKVYUHHBib256aW5p
# QHJlZGhhdC5jb20ACgkQv/vSX3jHroPKYgf/QkWrNXdjjD3yAsv5LbJFVTVyCYW3
# b4Iax29kEDy8k9wbzfLxOfIk9jXIjmbOMO5ZN9LFiHK6VJxbXslsMh6hm50M3xKe
# 49X1Rvf9YuVA7KZX+dWkEuqLYI6Tlgj3HaCilYWfXrjyo6hY3CxzkPV/ChmaeYlV
# Ad4Y8biifoUuuEK8OTeTlcDWLhOHlFXylG3AXqULsUsXp0XhWJ9juXQ60eATv/W4
# eCEH7CSmRhYFu2/rV+IrWFYMnskLRTk1OC1/m6yXGPKOzgnOcthuvQfiUgPkbR/d
# llY6Ni5Aaf7+XX3S7Avcyvoq8jXzaaMzOrzL98rxYGDR1sYBYO+4h4ZToA==
# =qQeP
# -----END PGP SIGNATURE-----
# gpg: Signature made Wed 05 Jun 2024 02:01:10 AM PDT
# gpg: using RSA key F13338574B662389866C7682BFFBD25F78C7AE83
# gpg: issuer "pbonzini@redhat.com"
# gpg: Good signature from "Paolo Bonzini <bonzini@gnu.org>" [full]
# gpg: aka "Paolo Bonzini <pbonzini@redhat.com>" [full]
* tag 'for-upstream' of https://gitlab.com/bonzini/qemu: (46 commits)
hw/i386: Add support for loading BIOS using guest_memfd
hw/i386/sev: Use guest_memfd for legacy ROMs
memory: Introduce memory_region_init_ram_guest_memfd()
i386/sev: Allow measured direct kernel boot on SNP
i386/sev: Reorder struct declarations
i386/sev: Extract build_kernel_loader_hashes
i386/sev: Enable KVM_HC_MAP_GPA_RANGE hcall for SNP guests
i386/kvm: Add KVM_EXIT_HYPERCALL handling for KVM_HC_MAP_GPA_RANGE
i386/sev: Invoke launch_updata_data() for SNP class
i386/sev: Invoke launch_updata_data() for SEV class
hw/i386/sev: Add support to encrypt BIOS when SEV-SNP is enabled
i386/sev: Add support for SNP CPUID validation
i386/sev: Add support for populating OVMF metadata pages
hw/i386/sev: Add function to get SEV metadata from OVMF header
i386/sev: Set CPU state to protected once SNP guest payload is finalized
i386/sev: Add handling to encrypt/finalize guest launch data
i386/sev: Add the SNP launch start context
i386/sev: Update query-sev QAPI format to handle SEV-SNP
i386/sev: Add a class method to determine KVM VM type for SNP guests
i386/sev: Don't return launch measurements for SEV-SNP guests
...
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Most of the current 'query-sev' command is relevant to both legacy
SEV/SEV-ES guests and SEV-SNP guests, with 2 exceptions:
- 'policy' is a 64-bit field for SEV-SNP, not 32-bit, and
the meaning of the bit positions has changed
- 'handle' is not relevant to SEV-SNP
To address this, this patch adds a new 'sev-type' field that can be
used as a discriminator to select between SEV and SEV-SNP-specific
fields/formats without breaking compatibility for existing management
tools (so long as management tools that add support for launching
SEV-SNP guest update their handling of query-sev appropriately).
The corresponding HMP command has also been fixed up similarly.
Signed-off-by: Michael Roth <michael.roth@amd.com>
Co-developed-by:Pankaj Gupta <pankaj.gupta@amd.com>
Signed-off-by: Pankaj Gupta <pankaj.gupta@amd.com>
Message-ID: <20240530111643.1091816-15-pankaj.gupta@amd.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
SEV-SNP support relies on a different set of properties/state than the
existing 'sev-guest' object. This patch introduces the 'sev-snp-guest'
object, which can be used to configure an SEV-SNP guest. For example,
a default-configured SEV-SNP guest with no additional information
passed in for use with attestation:
-object sev-snp-guest,id=sev0
or a fully-specified SEV-SNP guest where all spec-defined binary
blobs are passed in as base64-encoded strings:
-object sev-snp-guest,id=sev0, \
policy=0x30000, \
init-flags=0, \
id-block=YWFhYWFhYWFhYWFhYWFhCg==, \
id-auth=CxHK/OKLkXGn/KpAC7Wl1FSiisWDbGTEKz..., \
author-key-enabled=on, \
host-data=LNkCWBRC5CcdGXirbNUV1OrsR28s..., \
guest-visible-workarounds=AA==, \
See the QAPI schema updates included in this patch for more usage
details.
In some cases these blobs may be up to 4096 characters, but this is
generally well below the default limit for linux hosts where
command-line sizes are defined by the sysconf-configurable ARG_MAX
value, which defaults to 2097152 characters for Ubuntu hosts, for
example.
Signed-off-by: Brijesh Singh <brijesh.singh@amd.com>
Co-developed-by: Michael Roth <michael.roth@amd.com>
Acked-by: Markus Armbruster <armbru@redhat.com> (for QAPI schema)
Signed-off-by: Michael Roth <michael.roth@amd.com>
Co-developed-by: Pankaj Gupta <pankaj.gupta@amd.com>
Signed-off-by: Pankaj Gupta <pankaj.gupta@amd.com>
Message-ID: <20240530111643.1091816-8-pankaj.gupta@amd.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Currently all SEV/SEV-ES functionality is managed through a single
'sev-guest' QOM type. With upcoming support for SEV-SNP, taking this
same approach won't work well since some of the properties/state
managed by 'sev-guest' is not applicable to SEV-SNP, which will instead
rely on a new QOM type with its own set of properties/state.
To prepare for this, this patch moves common state into an abstract
'sev-common' parent type to encapsulate properties/state that are
common to both SEV/SEV-ES and SEV-SNP, leaving only SEV/SEV-ES-specific
properties/state in the current 'sev-guest' type. This should not
affect current behavior or command-line options.
As part of this patch, some related changes are also made:
- a static 'sev_guest' variable is currently used to keep track of
the 'sev-guest' instance. SEV-SNP would similarly introduce an
'sev_snp_guest' static variable. But these instances are now
available via qdev_get_machine()->cgs, so switch to using that
instead and drop the static variable.
- 'sev_guest' is currently used as the name for the static variable
holding a pointer to the 'sev-guest' instance. Re-purpose the name
as a local variable referring the 'sev-guest' instance, and use
that consistently throughout the code so it can be easily
distinguished from sev-common/sev-snp-guest instances.
- 'sev' is generally used as the name for local variables holding a
pointer to the 'sev-guest' instance. In cases where that now points
to common state, use the name 'sev_common'; in cases where that now
points to state specific to 'sev-guest' instance, use the name
'sev_guest'
In order to enable kernel-hashes for SNP, pull it from
SevGuestProperties to its parent SevCommonProperties so
it will be available for both SEV and SNP.
Signed-off-by: Michael Roth <michael.roth@amd.com>
Co-developed-by: Dov Murik <dovmurik@linux.ibm.com>
Signed-off-by: Dov Murik <dovmurik@linux.ibm.com>
Acked-by: Markus Armbruster <armbru@redhat.com> (QAPI schema)
Co-developed-by: Pankaj Gupta <pankaj.gupta@amd.com>
Signed-off-by: Pankaj Gupta <pankaj.gupta@amd.com>
Message-ID: <20240530111643.1091816-5-pankaj.gupta@amd.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
'vcpu' fields are deprecated since commit 5485e52a33
("qapi: make the vcpu parameters deprecated for 8.1"),
time to remove them.
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <20240530071548.20074-3-philmd@linaro.org>
The MEM_UNPLUG_ERROR event is deprecated since commit d43f1670c7
("qapi/qdev.json: add DEVICE_UNPLUG_GUEST_ERROR QAPI event"),
time to remove it.
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Harsh Prateek Bora <harshpb@linux.ibm.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <20240530071548.20074-2-philmd@linaro.org>
Using -fsanitize=undefined with Clang v18 causes an error if function
pointers are casted:
qapi/qapi-clone-visitor.c:188:5: runtime error: call to function visit_type_SocketAddress through pointer to incorrect function type 'bool (*)(struct Visitor *, const char *, void **, struct Error **)'
/tmp/qemu-ubsan/qapi/qapi-visit-sockets.c:487: note: visit_type_SocketAddress defined here
#0 0x5642aa2f7f3b in qapi_clone qapi/qapi-clone-visitor.c:188:5
#1 0x5642aa2c8ce5 in qio_channel_socket_listen_async io/channel-socket.c:285:18
#2 0x5642aa2b8903 in test_io_channel_setup_async tests/unit/test-io-channel-socket.c:116:5
#3 0x5642aa2b8204 in test_io_channel tests/unit/test-io-channel-socket.c:179:9
#4 0x5642aa2b8129 in test_io_channel_ipv4 tests/unit/test-io-channel-socket.c:323:5
...
It also prevents enabling the strict mode of CFI which is currently
disabled with -fsanitize-cfi-icall-generalize-pointers.
The problematic casts are necessary to pass visit_type_T() and
visit_type_T_members() as callbacks to qapi_clone() and qapi_clone_members(),
respectively. Open-code these two functions to avoid the callbacks, and
thus the type casts.
Resolves: https://gitlab.com/qemu-project/qemu/-/issues/2346
Signed-off-by: Akihiko Odaki <akihiko.odaki@daynix.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Message-ID: <20240524-xkb-v4-3-2de564e5c859@daynix.com>
[thuth: Improve commit message according to Markus' suggestions]
Signed-off-by: Thomas Huth <thuth@redhat.com>
Add a parameter that enables discard-after-copy. That is mostly useful
in "push backup with fleecing" scheme, when source is snapshot-access
format driver node, based on copy-before-write filter snapshot-access
API:
[guest] [snapshot-access] ~~ blockdev-backup ~~> [backup target]
| |
| root | file
v v
[copy-before-write]
| |
| file | target
v v
[active disk] [temp.img]
In this case discard-after-copy does two things:
- discard data in temp.img to save disk space
- avoid further copy-before-write operation in discarded area
Note that we have to declare WRITE permission on source in
copy-before-write filter, for discard to work. Still we can't take it
unconditionally, as it will break normal backup from RO source. So, we
have to add a parameter and pass it thorough bdrv_open flags.
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@yandex-team.ru>
Reviewed-by: Fiona Ebner <f.ebner@proxmox.com>
Tested-by: Fiona Ebner <f.ebner@proxmox.com>
Acked-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <20240313152822.626493-5-vsementsov@yandex-team.ru>
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@yandex-team.ru>
Add a new QAPI event for VFIO migration. This event will be emitted when
a VFIO device changes its migration state, for example, during migration
or when stopping/starting the guest.
This event can be used by management applications to get updates on the
current state of the VFIO device for their own purposes.
Note that this new event is introduced since VFIO devices have a unique
set of migration states which cannot be described as accurately by other
existing events such as run state or migration status.
Signed-off-by: Avihai Horon <avihaih@nvidia.com>
Reviewed-by: Cédric Le Goater <clg@redhat.com>
Signed-off-by: Cédric Le Goater <clg@redhat.com>
Retain a list of deprecated features disjoint from any particular
CPU model. A query-cpu-model-expansion reply will now provide a list of
properties (i.e. features) that are flagged as deprecated. Example:
{
"return": {
"model": {
"name": "z14.2-base",
"deprecated-props": [
"bpb",
"csske"
],
"props": {
"pfmfi": false,
"exrl": true,
...a lot more props...
"skey": false,
"vxpdeh2": false
}
}
}
}
It is recommended that s390 guests operate with these features
explicitly disabled to ensure compatibility with future hardware.
Signed-off-by: Collin Walling <walling@linux.ibm.com>
Acked-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Message-ID: <20240429191059.11806-2-walling@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
The 'compress' migration capability enables the old compression code
which has shown issues over the years and is thought to be less stable
and tested than the more recent multifd-based compression. The old
compression code has been deprecated in 8.2 and now is time to remove
it.
Deprecation commit 864128df46 ("migration: Deprecate old compression
method").
Acked-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
Signed-off-by: Fabiano Rosas <farosas@suse.de>
The block migration has been considered obsolete since QEMU 8.2 in
favor of the more flexible storage migration provided by the
blockdev-mirror driver. Two releases have passed so now it's time to
remove it.
Deprecation commit 66db46ca83 ("migration: Deprecate block
migration").
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
Signed-off-by: Fabiano Rosas <farosas@suse.de>
