The primary purpose of this change is to clean up
machine.py's console_socket property to return a single type,
a ConsoleSocket.

ConsoleSocket now derives from a socket, which means that
in the default case (of not draining), machine.py
will see the same behavior as it did prior to ConsoleSocket.

Signed-off-by: Robert Foley <robert.foley@linaro.org>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20200717203041.9867-3-robert.foley@linaro.org>
Message-Id: <20200724064509.331-16-alex.bennee@linaro.org>

The changes to console_socket.py and machine.py are to
cleanup for pylint and flake8.

Signed-off-by: Robert Foley <robert.foley@linaro.org>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Message-Id: <20200717203041.9867-2-robert.foley@linaro.org>
Message-Id: <20200724064509.331-15-alex.bennee@linaro.org>

3 seconds is too short for some tests running inside busy VMs. Build it out to
a rather generous 30 seconds to find out conclusively if there are more severe
problems in the merge/CI tests.

Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Eduardo Habkost <ehabkost@redhat.com>
Message-id: 20200720160252.104139-2-jsnow@redhat.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>

In the case that we receive a reply but are unable to understand it,
use this exception name to indicate that case.

Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Message-Id: <20200710052220.3306-7-jsnow@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

mypy and python type hints are not powerful enough to properly describe
JSON messages in Python 3.6. The best we can do, generally, is describe
them as Dict[str, Any].

Add casts to coerce this type for static analysis; but do NOT enforce
this type at runtime in any way.

Note: Python 3.8 adds a TypedDict construct which allows for the
description of more arbitrary Dictionary shapes. There is a third-party
module, "Pydantic", which is compatible with 3.6 that can be used
instead of the JSON library that parses JSON messages to fully-typed
Python objects, and may be preferable in some cases.

(That is well beyond the scope of this commit or series.)

Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Message-Id: <20200710052220.3306-6-jsnow@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

This makes typing the qmp library difficult, as it necessitates wrapping
Optional[] around the type for every return type up the stack. At some
point, it becomes difficult to discern or remember why it's None instead
of the expected object.

Use the python exception system to tell us exactly why we didn't get an
object. Remove this special-cased return.

Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Message-Id: <20200710052220.3306-5-jsnow@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

When I initially split this out, I considered this more of a machine
error than a QMP protocol error, but I think that's misguided.

Move this back to qmp.py and name it QMPResponseError. Convert
qmp.command() to use this exception type.

Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Message-Id: <20200710052220.3306-4-jsnow@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

Define some common types that we'll need to annotate a lot of other
functions going forward.

Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Message-Id: <20200710052220.3306-2-jsnow@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

Machine.wait() does not appear to be used except in the acceptance tests,
and an infinite timeout by default in a test suite is not the most helpful.

Change it to 3 seconds, like the default shutdown timeout.

Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Cleber Rosa <crosa@redhat.com>
Tested-by: Cleber Rosa <crosa@redhat.com>
Message-Id: <20200710050649.32434-13-jsnow@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

If the user kills QEMU on purpose, we don't need to warn
them about that having happened: they know already.

Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Cleber Rosa <crosa@redhat.com>
Message-Id: <20200710050649.32434-12-jsnow@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

This is done primarily to avoid the 'bare except' pattern, which
suppresses all exceptions during shutdown and can obscure errors.

Replace this with a pattern that isolates the different kind of shutdown
paradigms (_hard_shutdown and _soft_shutdown), and a new fallback shutdown
handler (_do_shutdown) that gracefully attempts one before the other.

This split now also ensures that no matter what happens,
_post_shutdown() is always invoked.

shutdown() changes in behavior such that if it attempts to do a graceful
shutdown and is unable to, it will now always raise an exception to
indicate this. This can be avoided by the test writer in three ways:

1. If the VM is expected to have already exited or is in the process of
exiting, wait() can be used instead of shutdown() to clean up resources
instead. This helps avoid race conditions in shutdown.

2. If a test writer is expecting graceful shutdown to fail, shutdown
should be called in a try...except block.

3. If the test writer has no interest in performing a graceful shutdown
at all, kill() can be used instead.

Handling shutdown in this way makes it much more explicit which type of
shutdown we want and allows the library to report problems with this
process.

Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Cleber Rosa <crosa@redhat.com>
Tested-by: Cleber Rosa <crosa@redhat.com>
Message-Id: <20200710050649.32434-11-jsnow@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

At this point, shutdown(has_quit=True) and wait() do essentially the
same thing; they perform cleanup without actually instructing QEMU to
quit.

Define one in terms of the other.

Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Cleber Rosa <crosa@redhat.com>
Tested-by: Cleber Rosa <crosa@redhat.com>
Message-Id: <20200710050649.32434-8-jsnow@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

Three seconds is hardcoded. Use it as a default parameter instead, and use that
value for both waits that may occur in the function.

Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Cleber Rosa <crosa@redhat.com>
Message-Id: <20200710050649.32434-7-jsnow@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

If the VM is not launched, don't try to shut it down. As a change,
_post_shutdown now unconditionally also calls _early_cleanup in order to
offer comprehensive object cleanup in failure cases.

As a courtesy, treat it as a NOP instead of rejecting it as an
error. This is slightly nicer for acceptance tests where vm.shutdown()
is issued unconditionally in tearDown callbacks.

Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20200710050649.32434-6-jsnow@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

This is primarily for consistency, and is a step towards wait() and
shutdown() sharing the same implementation so that the two cleanup paths
cannot diverge.

Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Cleber Rosa <crosa@redhat.com>
Tested-by: Cleber Rosa <crosa@redhat.com>
Message-Id: <20200710050649.32434-5-jsnow@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

Some parts of cleanup need to occur prior to shutdown, otherwise
shutdown might break. Move this into a suitably named method/callback.

Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Cleber Rosa <crosa@redhat.com>
Tested-by: Cleber Rosa <crosa@redhat.com>
Message-Id: <20200710050649.32434-4-jsnow@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

It's not important to do this before waiting for the process to exit, so
it can be done during generic post-shutdown cleanup.

Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Cleber Rosa <crosa@redhat.com>
Tested-by: Cleber Rosa <crosa@redhat.com>
Message-Id: <20200710050649.32434-3-jsnow@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

Move more cleanup actions into _post_shutdown. As a change, if QEMU
should so happen to be terminated during a call to wait(), that event
will now be logged.

This is not likely to occur during normative use.

Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Cleber Rosa <crosa@redhat.com>
Tested-by: Cleber Rosa <crosa@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20200710050649.32434-2-jsnow@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

We add the ConsoleSocket object, which has a socket interface
and which will consume all arriving characters on the
socket, placing them into an in memory buffer.
This will also provide those chars via recv() as
would a regular socket.
ConsoleSocket also has the option of dumping
the console bytes to a log file.

We also give QEMUMachine the option of using ConsoleSocket
to drain and to use for logging console to a file.
By default QEMUMachine does not use ConsoleSocket.

This is added in preparation for use by basevm.py in a later commit.
This is a workaround we found was needed for basevm.py since
there is a known issue where QEMU will hang waiting
for console characters to be consumed.

Cc: Eduardo Habkost <ehabkost@redhat.com>
Cc: Cleber Rosa <crosa@redhat.com>
Signed-off-by: Robert Foley <robert.foley@linaro.org>
Reviewed-by: Peter Puhov <peter.puhov@linaro.org>
Acked-by: Alex Bennée <alex.bennee@linaro.org>
Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20200601211421.1277-9-robert.foley@linaro.org>
Message-Id: <20200701135652.1366-13-alex.bennee@linaro.org>

It can be None; so add assertions or exceptions where appropriate to
guard the access accordingly.

Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20200514055403.18902-30-jsnow@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

In truth, if you don't do this, you'll just get a TypeError
exception. Now, you'll get an AssertionError.

Is this tangibly better? No.
Does mypy complain less? Yes.

Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20200514055403.18902-21-jsnow@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

The type system doesn't want integers.

Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20200514055403.18902-15-jsnow@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

mypy considers it incorrect to use `bool` to statically return false,
because it will assume that it could conceivably return True, and gives
different analysis in that case. Use a None return to achieve the same
effect, but make mypy happy.

Note: Pylint considers function signatures as code that might trip the
duplicate-code checker. I'd rather not disable this as it does not
trigger often in practice, so I'm disabling it as a one-off and filed a
change request; see https://github.com/PyCQA/pylint/issues/3619



Signed-off-by: John Snow <jsnow@redhat.com>
Acked-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20200514055403.18902-14-jsnow@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

Note:

A bug in typeshed (https://github.com/python/typeshed/issues/3977

)
misinterprets the type of makefile(). Work around this by explicitly
stating that we are opening a text-mode file.

Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20200514055403.18902-13-jsnow@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

Use the Python3 style instead.

Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20200514055403.18902-12-jsnow@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

Mostly, ignore the "no bare except" rule, because flake8 is not
contextual and cannot determine if we re-raise. Pylint can, though, so
always prefer pylint for that.

Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20200528222129.23826-5-jsnow@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

Bring our these files up to speed with pylint 2.5.0.
Add a pylintrc file to formalize which pylint subset
we are targeting.

The similarity ignore is there to suppress similarity
reports across imports, which for typing constants,
are going to trigger this report erroneously.

Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20200528222129.23826-4-jsnow@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

Python 3.5 and above do not print a warning when logging is not
configured. As a library, it's best practice to leave logging
configuration to the client executable.

Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20200514055403.18902-22-jsnow@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

Add method to hard-kill vm, without any quit commands.

Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Reviewed-by: Andrey Shinkevich <andrey.shinkevich@virtuozzo.com>
Message-Id: <20200217150246.29180-19-vsementsov@virtuozzo.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

With a QEMU bug, it can happen that the QEMU process doesn't react to a
'quit' QMP command. If we got an exception during previous QMP
communication (e.g. iotests Timeout expiring), we could also be in an
inconsistent state where after sending 'quit' we immediately read an old
response and close the socket even though the 'quit' command wasn't
processed yet. Both cases would lead to a hanging test.

Fix this by waiting for the QEMU process to exit after sending 'quit'
with a timeout, and if it doesn't happen within three seconds, send
SIGKILL.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: John Snow <jsnow@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-id: 20200313083617.8326-3-kwolf@redhat.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>

QEMUMachine writes some messages to the default logger.
But it sometimes hard to read the output if we have requests to
more than one VM.
This patch adds a label to the logger in the debug mode.

Signed-off-by: Oksana Vohchana <ovoshcha@redhat.com>
Reviewed-by: John Snow <jsnow@redhat.com>
Reviewed-by: Wainer dos Santos Moschetta <wainersm@redhat.com>
Reviewed-by: Cleber Rosa <crosa@redhat.com>
Message-Id: <20200316103203.10046-1-ovoshcha@redhat.com>
Signed-off-by: Cleber Rosa <crosa@redhat.com>

Currently the QEMU Python module limits the QEMUMachine class to
use the first serial console.

Some machines/guest might use another console than the first one as
the 'boot console'. For example the Raspberry Pi uses the second
(AUX) console.

To be able to use the Nth console as default, we simply need to
connect all the N - 1 consoles to the null chardev.

Add an index argument, so we can use a specific serial console as
default.

Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Liam Merwick <liam.merwick@oracle.com>
Tested-by: Liam Merwick <liam.merwick@oracle.com>
Reviewed-by: Wainer dos Santos Moschetta <wainersm@redhat.com>
Message-Id: <20200120235159.18510-5-f4bug@amsat.org>
[PMD: zero-initialize _console_index in __init__()]
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

In case qemu process dies the "monitor.cmd" returns None which gets
passed to the "__negotiate_capabilities" and leads to unhandled
exception. Let's only check the resp in case it has a value.

Signed-off-by: Lukáš Doktor <ldoktor@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Wainer dos Santos Moschetta <wainersm@redhat.com>
Message-Id: <20200120071202.30646-1-ldoktor@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

On ppc64le, the accel.kvm_available() check may wrongly
return False because the host arch (as returned by os.uname[4])
and the target arch (ppc64) mismatch. In order to solve this
it is added an ppc64le -> ppc64 mapping which is used as an
fallback verification.

Fixes: 53a049d7
Signed-off-by: Wainer dos Santos Moschetta <wainersm@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20200205203250.30526-5-wainersm@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

The `error` and `timeout` attributes in QEMUMonitorProtocol are
not used, so this delete them.

Signed-off-by: Wainer dos Santos Moschetta <wainersm@redhat.com>
Reviewed-by: John Snow <jsnow@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20191227134101.244496-6-wainersm@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

This implement the __enter__ and __exit__ functions on
QEMUMonitorProtocol class so that it can be used on 'with'
statement and the resources will be free up on block end:

with QEMUMonitorProtocol(socket_path) as qmp:
    qmp.connect()
    qmp.command('query-status')

Signed-off-by: Wainer dos Santos Moschetta <wainersm@redhat.com>
Reviewed-by: John Snow <jsnow@redhat.com>
Message-Id: <20200204141111.3207-5-wainersm@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

Currently the timeout of QEMUMonitorProtocol.accept() is
hard-coded to 15.0 seconds. This added the parameter `timeout`
so the value can be configured by the user.

Signed-off-by: Wainer dos Santos Moschetta <wainersm@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: John Snow <jsnow@redhat.com>
Message-Id: <20200204141111.3207-4-wainersm@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

This clean up the pylint-3 report on qmp:

************* Module qemu.qmp
python/qemu/qmp.py:1:0: C0111: Missing module docstring (missing-docstring)
python/qemu/qmp.py:17:0: C0111: Missing class docstring (missing-docstring)
python/qemu/qmp.py:21:0: C0111: Missing class docstring (missing-docstring)
python/qemu/qmp.py:25:0: C0111: Missing class docstring (missing-docstring)
python/qemu/qmp.py:29:0: C0111: Missing class docstring (missing-docstring)
python/qemu/qmp.py:33:0: C0111: Missing class docstring (missing-docstring)
python/qemu/qmp.py:33:0: R0205: Class 'QEMUMonitorProtocol' inherits from object, can be safely removed from bases in python3 (useless-object-inheritance)
python/qemu/qmp.py:80:4: R1710: Either all return statements in a function should return an expression, or none of them should. (inconsistent-return-statements)
python/qemu/qmp.py:131:4: R1710: Either all return statements in a function should return an expression, or none of them should. (inconsistent-return-statements)
python/qemu/qmp.py:159:4: R1710: Either all return statements in a function should return an expression, or none of them should. (inconsistent-return-statements)
python/qemu/qmp.py:245:4: C0111: Missing method docstring (missing-docstring)
python/qemu/qmp.py:249:4: C0111: Missing method docstring (missing-docstring)
python/qemu/qmp.py:252:4: C0111: Missing method docstring (missing-docstring)
python/qemu/qmp.py:255:4: C0111: Missing method docstring (missing-docstring)

Signed-off-by: Wainer dos Santos Moschetta <wainersm@redhat.com>
Reviewed-by: John Snow <jsnow@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20191227134101.244496-3-wainersm@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

The socket.error is deprecated from Python 3.3, instead it is
made a link to OSError. This change replaces the occurences
of socket.error with OSError.

Signed-off-by: Wainer dos Santos Moschetta <wainersm@redhat.com>
Reviewed-by: John Snow <jsnow@redhat.com>
Message-Id: <20191227134101.244496-2-wainersm@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

__init_.py import some sub-modules unnecessarily. So let's
clean it up.

Signed-off-by: Wainer dos Santos Moschetta <wainersm@redhat.com>
Suggested-by: Cleber Rosa <crosa@redhat.com>
Reviewed-by: Cleber Rosa <crosa@redhat.com>
Tested-by: Cleber Rosa <crosa@redhat.com>
Message-Id: <20191216191438.93418-6-wainersm@redhat.com>
Signed-off-by: Cleber Rosa <crosa@redhat.com>