John Snow [Fri, 4 Jun 2021 15:55:31 +0000 (11:55 -0400)]
python/qemu-ga-client: add entry point
Remove the shebang, and add a package-defined entry point instead. Now,
it can be accessed using 'qemu-ga-client' from the command line after
installing the package.
The next commit adds a forwarder shim that allows the running of this
script without needing to install the package again.
Signed-off-by: John Snow <jsnow@redhat.com>
Message-id: 20210604155532.1499282-11-jsnow@redhat.com Signed-off-by: John Snow <jsnow@redhat.com>
John Snow [Fri, 4 Jun 2021 15:55:30 +0000 (11:55 -0400)]
scripts/qemu-ga-client: move to python/qemu/qmp/qemu_ga_client.py
The script itself will be unavailable for a few commits before being
restored, with no way to run it right after this commit. This helps move
git history into the new file. To prevent linter regressions, though, we
do need to immediately touch up the filename to remove dashes (to make
the module importable), and remove the executable bit.
Signed-off-by: John Snow <jsnow@redhat.com>
Message-id: 20210604155532.1499282-10-jsnow@redhat.com Signed-off-by: John Snow <jsnow@redhat.com>
John Snow [Fri, 4 Jun 2021 15:55:29 +0000 (11:55 -0400)]
scripts/qemu-ga-client: add mypy type hints
This script is in slightly rough shape, but it still works. A lot of
care went into its initial development. In good faith, I'm updating it
to the latest Python coding standards. If there is in interest in this
script, though, I'll be asking for a contributor to take care of it
further.
Signed-off-by: John Snow <jsnow@redhat.com>
Message-id: 20210604155532.1499282-9-jsnow@redhat.com Signed-off-by: John Snow <jsnow@redhat.com>
John Snow [Fri, 4 Jun 2021 15:55:28 +0000 (11:55 -0400)]
python/qmp: Correct type of QMPReturnValue
It's only a Dict[str, Any] most of the time. It's not actually
guaranteed to be anything in particular. Fix this type to be
more accurate to the reality we live in.
Signed-off-by: John Snow <jsnow@redhat.com>
Message-id: 20210604155532.1499282-8-jsnow@redhat.com Signed-off-by: John Snow <jsnow@redhat.com>
John Snow [Fri, 4 Jun 2021 15:55:27 +0000 (11:55 -0400)]
scripts/qemu-ga-client: apply (most) pylint rules
I'm only doing a very quick best-effort to preserve this script, to help
keep it from breaking further. I think there are pending ideas swirling
on the right way to implement better SDKs and better clients, and this
script might be a handy reference for those discussions. It presents
some interesting design problems, like static type safety when using a
dynamic RPC mechanism.
I believe it's worth preserving the effort and care that went into
making this script by updating it to work with our current
infrastructure. However, I am disabling the requirement for docstrings
in this file.
If you would like to help improve this script, please add docstrings
alongside any refactors or rejuvenations you might apply at that time.
Signed-off-by: John Snow <jsnow@redhat.com>
Message-id: 20210604155532.1499282-7-jsnow@redhat.com Signed-off-by: John Snow <jsnow@redhat.com>
John Snow [Thu, 3 Jun 2021 00:37:19 +0000 (20:37 -0400)]
python/qmp: add fuse command to 'qom' tools
The 'fuse' command will be unavailable if 'fusepy' is not installed. It
will simply not load and subsequently be unavailable as a subcommand.
Signed-off-by: John Snow <jsnow@redhat.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-id: 20210603003719.1321369-20-jsnow@redhat.com Signed-off-by: John Snow <jsnow@redhat.com>
John Snow [Thu, 3 Jun 2021 00:37:18 +0000 (20:37 -0400)]
scripts/qom-fuse: add redirection shim to python/qemu/qmp/qom-fuse.py
By leaving the script absent for a commit, git-blame travels to the new
file instead of staying on the shim.
Signed-off-by: John Snow <jsnow@redhat.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-id: 20210603003719.1321369-19-jsnow@redhat.com Signed-off-by: John Snow <jsnow@redhat.com>
John Snow [Thu, 3 Jun 2021 00:37:17 +0000 (20:37 -0400)]
scripts/qom-fuse: move to python/qemu/qmp/qom_fuse.py
Move qom-fuse over to the python package now that it passes the
linter. Update the import paradigms so that it continues to pass in the
context of the Python package.
Signed-off-by: John Snow <jsnow@redhat.com>
Message-id: 20210603003719.1321369-18-jsnow@redhat.com Signed-off-by: John Snow <jsnow@redhat.com>
John Snow [Thu, 3 Jun 2021 00:37:16 +0000 (20:37 -0400)]
python: add optional FUSE dependencies
In preparation for moving qom-fuse over to the python package, we need
some new dependencies to support it.
Add an optional 'fusepy' dependency that users of the package can opt
into with e.g. "pip install qemu[fuse]" which installs the requirements
necessary to obtain the additional functionality.
Add the same fusepy dependency to the 'devel' extras group --
unfortunately I do not see a way for optional groups to imply other
optional groups at present, so the dependency is repeated. The
development group needs to include the full set of dependencies for the
purpose of static analysis of all features offered by this library.
Lastly, add the [fuse] extras group to tox's configuration as a
workaround so that if a stale tox environment is found when running
`make check-tox`, tox will know to rebuild its environments.
Signed-off-by: John Snow <jsnow@redhat.com>
Message-id: 20210603003719.1321369-17-jsnow@redhat.com Signed-off-by: John Snow <jsnow@redhat.com>
John Snow [Thu, 3 Jun 2021 00:37:15 +0000 (20:37 -0400)]
scripts/qom-fuse: add static type hints
Because fusepy does not have type hints, add some targeted warning
suppressions.
Namely, we need to allow subclassing something of an unknown type (in
qom_fuse.py), and we need to allow missing imports (recorded against
fuse itself) because mypy will be unable to import fusepy (even when
installed) as it has no types nor type stubs available.
Note: Until now, it was possible to run invocations like 'mypy qemu/'
from ./python and have that work. However, these targeted suppressions
require that you run 'mypy -p qemu/' instead. The correct, canonical
invocation is recorded in ./python/tests/mypy.sh and all of the various
CI invocations always use this correct form.
Signed-off-by: John Snow <jsnow@redhat.com>
Message-id: 20210603003719.1321369-16-jsnow@redhat.com Signed-off-by: John Snow <jsnow@redhat.com>
- Use FuseOSError to signal ENOENT instead of returning it
- Wrap qom-get in str(), as we don't always know its type
- The empty return should be b'', not ''.
Signed-off-by: John Snow <jsnow@redhat.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-id: 20210603003719.1321369-15-jsnow@redhat.com Signed-off-by: John Snow <jsnow@redhat.com>
John Snow [Thu, 3 Jun 2021 00:37:13 +0000 (20:37 -0400)]
scripts/qom-fuse: use QOMCommand.qom_list()
the qom_list method provides a type-safe object that's easier to type
check, so switch to using it.
Signed-off-by: John Snow <jsnow@redhat.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-id: 20210603003719.1321369-14-jsnow@redhat.com Signed-off-by: John Snow <jsnow@redhat.com>
John Snow [Thu, 3 Jun 2021 00:37:12 +0000 (20:37 -0400)]
scripts/qom-fuse: Convert to QOMCommand
Move qom-fuse onto the QOMCommand base established in
python/qemu/qmp/qom_common.py. The interface doesn't change
incompatibly, "qom-fuse mountpoint" still works as an invocation, and
QMP_SOCKET is still used as the environment variable.
Signed-off-by: John Snow <jsnow@redhat.com>
Message-id: 20210603003719.1321369-13-jsnow@redhat.com Signed-off-by: John Snow <jsnow@redhat.com>
John Snow [Thu, 3 Jun 2021 00:37:10 +0000 (20:37 -0400)]
scripts/qom-fuse: Apply pylint rules
- Catch specific exceptions from QMP
- Reraise errors with explicit context
- method parameters should match parent's names
Signed-off-by: John Snow <jsnow@redhat.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-id: 20210603003719.1321369-11-jsnow@redhat.com Signed-off-by: John Snow <jsnow@redhat.com>
John Snow [Thu, 3 Jun 2021 00:37:09 +0000 (20:37 -0400)]
python: Add 'fh' to known-good variable names
fd and fh are fine: we often use these for "file descriptor" or "file
handle" accordingly. It is rarely the case that you need to enforce a
more semantically meaningful name beyond "This is the file we are using
right now."
While we're here: add comments for all of the non-standard pylint
names. (And the underscore.)
Signed-off-by: John Snow <jsnow@redhat.com>
Message-id: 20210603003719.1321369-10-jsnow@redhat.com Signed-off-by: John Snow <jsnow@redhat.com>
John Snow [Thu, 3 Jun 2021 00:37:08 +0000 (20:37 -0400)]
scripts/qom-fuse: apply flake8 rules
flake8 still has one warning because of the sys.path hack, but that will
be going away by the end of this patch series.
Signed-off-by: John Snow <jsnow@redhat.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-id: 20210603003719.1321369-9-jsnow@redhat.com Signed-off-by: John Snow <jsnow@redhat.com>
John Snow [Thu, 3 Jun 2021 00:37:07 +0000 (20:37 -0400)]
scripts/qom-fuse: apply isort rules
Hint: you can use symlinks to create qom_fuse.py in python/qemu/qmp/ and
point to scripts/qom-fuse to apply the standard linting rules to this
script.
Signed-off-by: John Snow <jsnow@redhat.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-id: 20210603003719.1321369-8-jsnow@redhat.com Signed-off-by: John Snow <jsnow@redhat.com>
John Snow [Thu, 3 Jun 2021 00:37:06 +0000 (20:37 -0400)]
scripts/qmp: redirect qom-xxx scripts to python/qemu/qmp/
Redirect to the new qom scripts. These forwarders can be deleted
eventually when there has been more time for the dust on the Python
packaging to settle and people understand how to find these commands.
Note: You can run these by setting $PYTHONPATH in your shell and then
running "python3 -m qemu.qmp.qom", or you can install the qemu namespace
package and use the "qom" or "qom-set" scripts.
I've written how to install the package elsewhere, but for the sake of
git-blame, cd to ./python, and then do:
- pip3 install [--user] [-e] .
--user will install to your local user install (will not work inside of
a venv), omitting this flag installs to your system-wide packages
(outside of a venv) or to your current virtual environment (inside the
venv).
When installing to a venv or to your system-wide packages, "qom"
should be in your $PATH already. If you do a user install, you may
need to add ~/.local/bin to your $PATH if you haven't already.
-e installs in editable mode: the installed package is effectively just
a symlink to this folder; so changes to your git working tree are
reflected in the installed package.
Note: installing these packages to an environment outside a venv can be
dangerous: Many QEMU scripts will begin to prefer the installed version
instead of the version directly in the tree. Use with caution. editable
mode is recommended when working outside of a venv.
Signed-off-by: John Snow <jsnow@redhat.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-id: 20210603003719.1321369-7-jsnow@redhat.com Signed-off-by: John Snow <jsnow@redhat.com>
John Snow [Thu, 3 Jun 2021 00:37:05 +0000 (20:37 -0400)]
python/qmp: add qom script entry points
Add the 'qom', 'qom-set', 'qom-get', 'qom-list', and 'qom-tree' scripts
to the qemu.qmp package. When you install this package, these scripts
will become available on your command line.
(e.g. when inside of a venv, `cd python && pip install .` will add
'qom', 'qom-set', etc to your $PATH.)
Signed-off-by: John Snow <jsnow@redhat.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-id: 20210603003719.1321369-6-jsnow@redhat.com Signed-off-by: John Snow <jsnow@redhat.com>
John Snow [Thu, 3 Jun 2021 00:37:03 +0000 (20:37 -0400)]
python/qmp: add parse_address classmethod
This takes the place of qmp-shell's __get_address function. It also
allows other utilities to share the same parser and syntax for
specifying QMP locations.
Signed-off-by: John Snow <jsnow@redhat.com>
Message-id: 20210603003719.1321369-4-jsnow@redhat.com Signed-off-by: John Snow <jsnow@redhat.com>
John Snow [Thu, 3 Jun 2021 00:37:02 +0000 (20:37 -0400)]
python/qmp: Fix type of SocketAddrT
In porting the qom tools, qmp-shell, etc; it becomes evident that this
type is wrong.
This is an integer, not a string. We didn't catch this before because
none of QEMUMonitorProtocol's *users* happen to be checked, and the
internal logic of this class is otherwise self-consistent. Additionally,
mypy was not introspecting into the socket() interface to realize we
were passing a bad type for AF_INET. Fixed now.
Signed-off-by: John Snow <jsnow@redhat.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-id: 20210603003719.1321369-3-jsnow@redhat.com Signed-off-by: John Snow <jsnow@redhat.com>
John Snow [Thu, 3 Jun 2021 00:37:01 +0000 (20:37 -0400)]
python/pipenv: Update Pipfile.lock
In a previous commit, I added tox to the development requirements of the
Python library. I never bothered to add them to the Pipfile, because
they aren't needed there. Here, I sync it anyway in its own commit so
that when we add new packages later that the diffstats will not
confusingly appear to pull in lots of extra packages.
Ideally I could tell Pipenv simply not to install these, but it doesn't
seem to support that, exactly. The alternative is removing Tox from the
development requires, which I'd rather not do.
The other alternative is re-specifying all of the dependencies of
setup.cfg in the Pipfile, which I'd also rather not do.
Picking what feels least-worst here.
Signed-off-by: John Snow <jsnow@redhat.com>
Message-id: 20210603003719.1321369-2-jsnow@redhat.com Signed-off-by: John Snow <jsnow@redhat.com>
req->receiving is a flag of request being in one concrete yield point
in nbd_co_do_receive_one_chunk().
Such kind of boolean flag is always better to unset before scheduling
the coroutine, to avoid double scheduling. So, let's be more careful.
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Reviewed-by: Eric Blake <eblake@redhat.com>
Message-Id: <20210610100802.5888-33-vsementsov@virtuozzo.com> Signed-off-by: Eric Blake <eblake@redhat.com>
We already have two similar helpers for other state. Let's add another
one for convenience.
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Reviewed-by: Eric Blake <eblake@redhat.com>
Message-Id: <20210610100802.5888-32-vsementsov@virtuozzo.com> Signed-off-by: Eric Blake <eblake@redhat.com>
block/nbd: reuse nbd_co_do_establish_connection() in nbd_open()
The only last step we need to reuse the function is coroutine-wrapper.
nbd_open() may be called from non-coroutine context. So, generate the
wrapper and use it.
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Reviewed-by: Eric Blake <eblake@redhat.com>
Message-Id: <20210610100802.5888-31-vsementsov@virtuozzo.com> Signed-off-by: Eric Blake <eblake@redhat.com>
nbd/client-connection: add option for non-blocking connection attempt
We'll need a possibility of non-blocking nbd_co_establish_connection(),
so that it returns immediately, and it returns success only if a
connections was previously established in background.
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Reviewed-by: Eric Blake <eblake@redhat.com>
Message-Id: <20210610100802.5888-30-vsementsov@virtuozzo.com> Signed-off-by: Eric Blake <eblake@redhat.com>
block/nbd: split nbd_co_do_establish_connection out of nbd_reconnect_attempt
Split out the part that we want to reuse for nbd_open().
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Message-Id: <20210610100802.5888-29-vsementsov@virtuozzo.com> Reviewed-by: Eric Blake <eblake@redhat.com> Signed-off-by: Eric Blake <eblake@redhat.com>
We are going to reuse the script to generate a nbd_ function in
further commit. Prepare the script now.
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Reviewed-by: Eric Blake <eblake@redhat.com>
Message-Id: <20210610100802.5888-28-vsementsov@virtuozzo.com> Signed-off-by: Eric Blake <eblake@redhat.com>
block/nbd doesn't need underlying sioc channel anymore. So, we can
update nbd/client-connection interface to return only one top-most io
channel, which is more straight forward.
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Reviewed-by: Eric Blake <eblake@redhat.com>
Message-Id: <20210610100802.5888-27-vsementsov@virtuozzo.com>
[eblake: squash in Vladimir's fixes for uninit usage caught by clang] Signed-off-by: Eric Blake <eblake@redhat.com>
Currently sioc pointer is used just to pass from socket-connection to
nbd negotiation. Drop the field, and use local variables instead. With
next commit we'll update nbd/client-connection.c to behave
appropriately (return only top-most ioc, not two channels).
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Reviewed-by: Eric Blake <eblake@redhat.com>
Message-Id: <20210610100802.5888-26-vsementsov@virtuozzo.com> Signed-off-by: Eric Blake <eblake@redhat.com>
block/nbd: don't touch s->sioc in nbd_teardown_connection()
Negotiation during reconnect is now done in a thread, and s->sioc is
not available during negotiation. Negotiation in thread will be
cancelled by nbd_client_connection_release() called from
nbd_clear_bdrvstate(). So, we don't need this code chunk anymore.
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Reviewed-by: Eric Blake <eblake@redhat.com>
Message-Id: <20210610100802.5888-25-vsementsov@virtuozzo.com> Signed-off-by: Eric Blake <eblake@redhat.com>
Now that we can opt in to negotiation as part of the client connection
thread, use that to simplify connection_co. This is another step on
the way to moving all reconnect code into NBDClientConnection.
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Reviewed-by: Eric Blake <eblake@redhat.com>
Message-Id: <20210610100802.5888-24-vsementsov@virtuozzo.com> Signed-off-by: Eric Blake <eblake@redhat.com>
block/nbd: split nbd_handle_updated_info out of nbd_client_handshake()
To be reused in the following patch.
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Reviewed-by: Roman Kagan <rvkagan@yandex-team.ru> Reviewed-by: Eric Blake <eblake@redhat.com>
Message-Id: <20210610100802.5888-23-vsementsov@virtuozzo.com> Signed-off-by: Eric Blake <eblake@redhat.com>
nbd/client-connection: shutdown connection on release
Now, when a thread can do negotiation and retry, it may run relatively
long. We need a mechanism to stop it, when the user is not interested
in a result any more. So, on nbd_client_connection_release() let's
shutdown the socket, and do not retry connection if thread is detached.
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Message-Id: <20210610100802.5888-22-vsementsov@virtuozzo.com> Reviewed-by: Eric Blake <eblake@redhat.com> Signed-off-by: Eric Blake <eblake@redhat.com>
Add an option for a thread to retry connecting until it succeeds. We'll
use nbd/client-connection both for reconnect and for initial connection
in nbd_open(), so we need a possibility to use same NBDClientConnection
instance to connect once in nbd_open() and then use retry semantics for
reconnect.
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Message-Id: <20210610100802.5888-21-vsementsov@virtuozzo.com> Reviewed-by: Eric Blake <eblake@redhat.com>
[eblake: grammar tweak] Signed-off-by: Eric Blake <eblake@redhat.com>
nbd/client-connection: add possibility of negotiation
Add arguments and logic to support nbd negotiation in the same thread
after successful connection.
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Message-Id: <20210610100802.5888-20-vsementsov@virtuozzo.com> Reviewed-by: Eric Blake <eblake@redhat.com> Signed-off-by: Eric Blake <eblake@redhat.com>
We don't update connect_thread_func() to use QEMU_LOCK_GUARD, as it
will get more complex critical sections logic in further commit, where
QEMU_LOCK_GUARD doesn't help.
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Message-Id: <20210610100802.5888-19-vsementsov@virtuozzo.com> Reviewed-by: Eric Blake <eblake@redhat.com> Signed-off-by: Eric Blake <eblake@redhat.com>
This is a last step of creating bs-independent nbd connection
interface. With next commit we can finally move it to separate file.
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Message-Id: <20210610100802.5888-17-vsementsov@virtuozzo.com> Reviewed-by: Eric Blake <eblake@redhat.com> Signed-off-by: Eric Blake <eblake@redhat.com>
This is a step of creating bs-independent nbd connection interface.
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Reviewed-by: Roman Kagan <rvkagan@yandex-team.ru> Reviewed-by: Eric Blake <eblake@redhat.com>
Message-Id: <20210610100802.5888-16-vsementsov@virtuozzo.com> Signed-off-by: Eric Blake <eblake@redhat.com>
block/nbd: rename NBDConnectThread to NBDClientConnection
We are going to move the connection code to its own file, and want
clear names and APIs first.
The structure is shared between user and (possibly) several runs of
connect-thread. So it's wrong to call it "thread". Let's rename to
something more generic.
Appropriately rename connect_thread and thr variables to conn.
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Reviewed-by: Roman Kagan <rvkagan@yandex-team.ru> Reviewed-by: Eric Blake <eblake@redhat.com>
Message-Id: <20210610100802.5888-15-vsementsov@virtuozzo.com> Signed-off-by: Eric Blake <eblake@redhat.com>
block/nbd: make nbd_co_establish_connection_cancel() bs-independent
nbd_co_establish_connection_cancel() actually needs only pointer to
NBDConnectThread. So, make it clean.
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Reviewed-by: Roman Kagan <rvkagan@yandex-team.ru> Reviewed-by: Eric Blake <eblake@redhat.com>
Message-Id: <20210610100802.5888-14-vsementsov@virtuozzo.com> Signed-off-by: Eric Blake <eblake@redhat.com>
block/nbd: bs-independent interface for nbd_co_establish_connection()
We are going to split connection code to a separate file. Now we are
ready to give nbd_co_establish_connection() clean and bs-independent
interface.
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Reviewed-by: Roman Kagan <rvkagan@yandex-team.ru> Reviewed-by: Eric Blake <eblake@redhat.com>
Message-Id: <20210610100802.5888-13-vsementsov@virtuozzo.com> Signed-off-by: Eric Blake <eblake@redhat.com>
We don't need all these states. The code refactored to use two boolean
variables looks simpler.
While moving the comment in nbd_co_establish_connection() rework it to
give better information. Also, we are going to move the connection code
to separate file and mentioning drained section would be confusing.
Improve also the comment in NBDConnectThread, while dropping removed
state names from it.
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Message-Id: <20210610100802.5888-12-vsementsov@virtuozzo.com> Reviewed-by: Eric Blake <eblake@redhat.com>
[eblake: comment tweak] Signed-off-by: Eric Blake <eblake@redhat.com>
block/nbd: simplify waking of nbd_co_establish_connection()
Instead of managing connect_bh, bh_ctx, and wait_connect fields, we
can use a single link to the waiting coroutine with proper mutex
protection.
So new logic is:
nbd_co_establish_connection() sets wait_co under the mutex, releases
the mutex, then yield()s. Note that wait_co may be scheduled by the
thread immediately after unlocking the mutex. Still, the main thread
(or iothread) will not reach the code for entering the coroutine until
the yield(), so we are safe.
connect_thread_func() and nbd_co_establish_connection_cancel() do
the following to handle wait_co:
Under the mutex, if thr->wait_co is not NULL, make it NULL and
schedule it. This way, we avoid scheduling the coroutine twice.
Still scheduling is a bit different:
In connect_thread_func() we can just call aio_co_wake under mutex,
after commit
[async: the main AioContext is only "current" if under the BQL]
we are sure that aio_co_wake() will not try to acquire the aio context
and do qemu_aio_coroutine_enter() but simply schedule the coroutine by
aio_co_schedule().
nbd_co_establish_connection_cancel() will be called from non-coroutine
context in further patch and will be able to go through
qemu_aio_coroutine_enter() path of aio_co_wake(). So keep current
behavior of waking the coroutine after the critical section.
Also, this commit reduces the dependence of
nbd_co_establish_connection() on the internals of bs (we now use a
generic pointer to the coroutine, instead of direct use of
s->connection_co). This is a step towards splitting the connection
API out of nbd.c.
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Message-Id: <20210610100802.5888-11-vsementsov@virtuozzo.com> Reviewied-by: Eric Blake <eblake@redhat.com> Signed-off-by: Eric Blake <eblake@redhat.com>
block/nbd: BDRVNBDState: drop unused connect_err and connect_status
These fields are write-only. Drop them.
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Reviewed-by: Roman Kagan <rvkagan@yandex-team.ru> Reviewed-by: Eric Blake <eblake@redhat.com>
Message-Id: <20210610100802.5888-10-vsementsov@virtuozzo.com> Signed-off-by: Eric Blake <eblake@redhat.com>
block/nbd: nbd_client_handshake(): fix leak of s->ioc
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Reviewed-by: Roman Kagan <rvkagan@yandex-team.ru>
Message-Id: <20210610100802.5888-9-vsementsov@virtuozzo.com> Signed-off-by: Eric Blake <eblake@redhat.com>
Roman Kagan [Thu, 10 Jun 2021 10:07:37 +0000 (13:07 +0300)]
block/nbd: ensure ->connection_thread is always valid
Simplify lifetime management of BDRVNBDState->connect_thread by
delaying the possible cleanup of it until the BDRVNBDState itself goes
away.
This also reverts 0267101af6 "block/nbd: fix possible use after free of s->connect_thread"
as now s->connect_thread can't be cleared until the very end.
Signed-off-by: Roman Kagan <rvkagan@yandex-team.ru>
[vsementsov: rebase, revert 0267101af6 changes] Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
[eblake: tweak comment] Reviewed-by: Eric Blake <eblake@redhat.com>
Message-Id: <20210610100802.5888-8-vsementsov@virtuozzo.com> Signed-off-by: Eric Blake <eblake@redhat.com>
block/nbd: call socket_address_parse_named_fd() in advance
Detecting monitor by current coroutine works bad when we are not in
coroutine context. And that's exactly so in nbd reconnect code, where
qio_channel_socket_connect_sync() is called from thread.
Monitor is needed only to parse named file descriptor. So, let's just
parse it during nbd_open(), so that all further users of s->saddr don't
need to access monitor.
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Message-Id: <20210610100802.5888-7-vsementsov@virtuozzo.com> Reviewed-by: Eric Blake <eblake@redhat.com> Signed-off-by: Eric Blake <eblake@redhat.com>
Add function that transforms named fd inside SocketAddress structure
into number representation. This way it may be then used in a context
where current monitor is not available.
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Message-Id: <20210610100802.5888-6-vsementsov@virtuozzo.com> Reviewed-by: Eric Blake <eblake@redhat.com>
[eblake: comment tweak] Signed-off-by: Eric Blake <eblake@redhat.com>
block/nbd: connect_thread_func(): do qio_channel_set_delay(false)
nbd_open() does it (through nbd_establish_connection()).
Actually we lost that call on reconnect path in 1dc4718d849e1a1fe
"block/nbd: use non-blocking connect: fix vm hang on connect()"
when we have introduced reconnect thread.
Fixes: 1dc4718d849e1a1fe665ce5241ed79048cfa2cfc Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Message-Id: <20210610100802.5888-5-vsementsov@virtuozzo.com> Reviewed-by: Eric Blake <eblake@redhat.com> Signed-off-by: Eric Blake <eblake@redhat.com>
block/nbd: fix how state is cleared on nbd_open() failure paths
We have two "return error" paths in nbd_open() after
nbd_process_options(). Actually we should call nbd_clear_bdrvstate()
on these paths. Interesting that nbd_process_options() calls
nbd_clear_bdrvstate() by itself.
Let's fix leaks and refactor things to be more obvious:
- intialize yank at top of nbd_open()
- move yank cleanup to nbd_clear_bdrvstate()
- refactor nbd_open() so that all failure paths except for
yank-register goes through nbd_clear_bdrvstate()
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Reviewed-by: Roman Kagan <rvkagan@yandex-team.ru> Reviewed-by: Eric Blake <eblake@redhat.com>
Message-Id: <20210610100802.5888-4-vsementsov@virtuozzo.com> Signed-off-by: Eric Blake <eblake@redhat.com>
Roman Kagan [Thu, 10 Jun 2021 10:07:32 +0000 (13:07 +0300)]
block/nbd: fix channel object leak
nbd_free_connect_thread leaks the channel object if it hasn't been
stolen.
Unref it and fix the leak.
Signed-off-by: Roman Kagan <rvkagan@yandex-team.ru> Reviewed-by: Eric Blake <eblake@redhat.com>
Message-Id: <20210610100802.5888-3-vsementsov@virtuozzo.com> Signed-off-by: Eric Blake <eblake@redhat.com>
qemu_co_queue_next() and qemu_co_queue_restart_all() just call
aio_co_wake() which works well in non-coroutine context. So these
functions can be called from non-coroutine context as well. And
actually qemu_co_queue_restart_all() is called from
nbd_cancel_in_flight(), which is called from non-coroutine context.
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Message-Id: <20210610100802.5888-2-vsementsov@virtuozzo.com> Reviewed-by: Eric Blake <eblake@redhat.com> Signed-off-by: Eric Blake <eblake@redhat.com>
Paolo Bonzini [Mon, 14 Jun 2021 11:02:14 +0000 (13:02 +0200)]
tests: cover aio_co_enter from a worker thread without BQL taken
Add a testcase for the test fixed by commit 'async: the main AioContext
is only "current" if under the BQL.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <20210614110214.726722-1-pbonzini@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Signed-off-by: Eric Blake <eblake@redhat.com>
Paolo Bonzini [Wed, 9 Jun 2021 12:22:34 +0000 (14:22 +0200)]
async: the main AioContext is only "current" if under the BQL
If we want to wake up a coroutine from a worker thread, aio_co_wake()
currently does not work. In that scenario, aio_co_wake() calls
aio_co_enter(), but there is no current AioContext and therefore
qemu_get_current_aio_context() returns the main thread. aio_co_wake()
then attempts to call aio_context_acquire() instead of going through
aio_co_schedule().
The default case of qemu_get_current_aio_context() was added to cover
synchronous I/O started from the vCPU thread, but the main and vCPU
threads are quite different. The main thread is an I/O thread itself,
only running a more complicated event loop; the vCPU thread instead
is essentially a worker thread that occasionally calls
qemu_mutex_lock_iothread(). It is only in those critical sections
that it acts as if it were the home thread of the main AioContext.
Therefore, this patch detaches qemu_get_current_aio_context() from
iothreads, which is a useless complication. The AioContext pointer
is stored directly in the thread-local variable, including for the
main loop. Worker threads (including vCPU threads) optionally behave
as temporary home threads if they have taken the big QEMU lock,
but if that is not the case they will always schedule coroutines
on remote threads via aio_co_schedule().
With this change, the stub qemu_mutex_iothread_locked() must be changed
from true to false. The previous value of true was needed because the
main thread did not have an AioContext in the thread-local variable,
but now it does have one.
Reported-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <20210609122234.544153-1-pbonzini@redhat.com> Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Tested-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
[eblake: tweak commit message per Vladimir's review] Signed-off-by: Eric Blake <eblake@redhat.com>
* remotes/awilliam/tags/vfio-update-20210618.0:
vfio/migration: Correct device state from vmstate change for savevm case
vfio: Fix unregister SaveVMHandler in vfio_migration_finalize
docs/devel: Add VFIO device migration documentation
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Kirti Wankhede [Tue, 8 Jun 2021 18:37:21 +0000 (00:07 +0530)]
vfio/migration: Correct device state from vmstate change for savevm case
Set _SAVING flag for device state from vmstate change handler when it
gets called from savevm.
Currently State transition savevm/suspend is seen as:
_RUNNING -> _STOP -> Stop-and-copy -> _STOP
State transition savevm/suspend should be:
_RUNNING -> Stop-and-copy -> _STOP
State transition from _RUNNING to _STOP occurs from
vfio_vmstate_change() where when vmstate changes from running to
!running, _RUNNING flag is reset but at the same time when
vfio_vmstate_change() is called for RUN_STATE_SAVE_VM, _SAVING bit
should be set.
Kunkun Jiang [Thu, 27 May 2021 12:31:01 +0000 (20:31 +0800)]
vfio: Fix unregister SaveVMHandler in vfio_migration_finalize
In the vfio_migration_init(), the SaveVMHandler is registered for
VFIO device. But it lacks the operation of 'unregister'. It will
lead to 'Segmentation fault (core dumped)' in
qemu_savevm_state_setup(), if performing live migration after a
VFIO device is hot deleted.
Fixes: 7c2f5f75f94 (vfio: Register SaveVMHandlers for VFIO device) Reported-by: Qixin Gan <ganqixin@huawei.com> Signed-off-by: Kunkun Jiang <jiangkunkun@huawei.com>
Message-Id: <20210527123101.289-1-jiangkunkun@huawei.com>
Reviewed by: Kirti Wankhede <kwankhede@nvidia.com> Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
* remotes/kraxel/tags/audio-20210617-pull-request:
coreaudio: Fix output stream format settings
audio: Fix format specifications of debug logs
hw/audio/sb16: Avoid assertion by restricting I/O sampling rate range
jackaudio: avoid that the client name contains the word (NULL)
audio: move code to audio/audio.c
paaudio: remove unused stream flags
alsaaudio: remove #ifdef DEBUG to avoid bit rot
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
docs: add a table showing x86-64 ABI compatibility levels
It is useful to know which CPUs satisfy each x86-64 ABI
compatibility level, when dealing with guest OS that require
something newer than the baseline ABI.
These ABI levels are defined in:
https://gitlab.com/x86-psABIs/x86-64-ABI/
and supported by GCC, Clang, glibc and more.
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20210607135843.196595-2-berrange@redhat.com> Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Tom Lendacky [Fri, 23 Apr 2021 20:08:19 +0000 (15:08 -0500)]
docs/interop/firmware.json: Add SEV-ES support
Create an enum definition, '@amd-sev-es', for SEV-ES and add documention
for the new enum. Add an example that shows some of the requirements for
SEV-ES, including not having SMM support and the requirement for an
X64-only build.
Tom Lendacky [Fri, 23 Apr 2021 20:08:18 +0000 (15:08 -0500)]
docs: Add SEV-ES documentation to amd-memory-encryption.txt
Update the amd-memory-encryption.txt file with information about SEV-ES,
including how to launch an SEV-ES guest and some of the differences
between SEV and SEV-ES guests in regards to launching and measuring the
guest.
Chenyi Qiang [Fri, 21 May 2021 04:38:20 +0000 (12:38 +0800)]
i386: Add ratelimit for bus locks acquired in guest
A bus lock is acquired through either split locked access to writeback
(WB) memory or any locked access to non-WB memory. It is typically >1000
cycles slower than an atomic operation within a cache and can also
disrupts performance on other cores.
Virtual Machines can exploit bus locks to degrade the performance of
system. To address this kind of performance DOS attack coming from the
VMs, bus lock VM exit is introduced in KVM and it can report the bus
locks detected in guest. If enabled in KVM, it would exit to the
userspace to let the user enforce throttling policies once bus locks
acquired in VMs.
The availability of bus lock VM exit can be detected through the
KVM_CAP_X86_BUS_LOCK_EXIT. The returned bitmap contains the potential
policies supported by KVM. The field KVM_BUS_LOCK_DETECTION_EXIT in
bitmap is the only supported strategy at present. It indicates that KVM
will exit to userspace to handle the bus locks.
This patch adds a ratelimit on the bus locks acquired in guest as a
mitigation policy.
Introduce a new field "bus_lock_ratelimit" to record the limited speed
of bus locks in the target VM. The user can specify it through the
"bus-lock-ratelimit" as a machine property. In current implementation,
the default value of the speed is 0 per second, which means no
restrictions on the bus locks.
As for ratelimit on detected bus locks, simply set the ratelimit
interval to 1s and restrict the quota of bus lock occurence to the value
of "bus_lock_ratelimit". A potential alternative is to introduce the
time slice as a property which can help the user achieve more precise
control.
The detail of bus lock VM exit can be found in spec:
https://software.intel.com/content/www/us/en/develop/download/intel-architecture-instruction-set-extensions-programming-reference.html
Peter Maydell [Thu, 17 Jun 2021 14:43:26 +0000 (15:43 +0100)]
Merge remote-tracking branch 'remotes/bonzini-gitlab/tags/for-upstream' into staging
* avoid deprecation warnings for SASL on macOS 10.11 or newer
* fix -readconfig when config blocks have an id (like [chardev "qmp"])
* Error* initialization fixes
* Improvements to ESP emulation (Mark)
* Allow creating noreserve memory backends (David)
* Improvements to query-memdev (David)
* Bump compiler to C11 (Richard)
* First round of SVM fixes from GSoC project (Lara)
* remotes/bonzini-gitlab/tags/for-upstream: (45 commits)
configure: Remove probe for _Static_assert
qemu/compiler: Remove QEMU_GENERIC
include/qemu/lockable: Use _Generic instead of QEMU_GENERIC
util: Use unique type for QemuRecMutex in thread-posix.h
util: Pass file+line to qemu_rec_mutex_unlock_impl
util: Use real functions for thread-posix QemuRecMutex
softfloat: Use _Generic instead of QEMU_GENERIC
configure: Use -std=gnu11
target/i386: Added Intercept CR0 writes check
target/i386: Added consistency checks for CR0
target/i386: Added consistency checks for VMRUN intercept and ASID
target/i386: Refactored intercept checks into cpu_svm_has_intercept
configure: map x32 to cpu_family x86_64 for meson
hmp: Print "reserve" property of memory backends with "info memdev"
qmp: Include "reserve" property of memory backends
hmp: Print "share" property of memory backends with "info memdev"
qmp: Include "share" property of memory backends
qmp: Clarify memory backend properties returned via query-memdev
hostmem: Wire up RAM_NORESERVE via "reserve" property
util/mmap-alloc: Support RAM_NORESERVE via MAP_NORESERVE under Linux
...
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Corey Minyard [Tue, 18 May 2021 21:08:03 +0000 (16:08 -0500)]
sensor: Move hardware sensors from misc to a sensor directory
Lots of this are expected to be coming in, create a directory for them.
Also move the tmp105.h file into the include directory where it
should be.
Cc: Cédric Le Goater <clg@kaod.org> Cc: Peter Maydell <peter.maydell@linaro.org> Cc: Andrew Jeffery <andrew@aj.id.au> Cc: Joel Stanley <joel@jms.id.au> Cc: Andrzej Zaborowski <balrogg@gmail.com> Cc: qemu-arm@nongnu.org Signed-off-by: Corey Minyard <cminyard@mvista.com> Acked-by: Cédric Le Goater <clg@kaod.org>
Corey Minyard [Tue, 18 May 2021 20:51:24 +0000 (15:51 -0500)]
adc: Move the zynq-xadc file to the adc directories
It's an ADC, put it where it belongs.
Cc: Edgar E. Iglesias <edgar.iglesias@gmail.com> Cc: Alistair Francis <alistair@alistair23.me> Cc: Peter Maydell <peter.maydell@linaro.org> Cc: qemu-arm@nongnu.org Signed-off-by: Corey Minyard <cminyard@mvista.com> Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Tested-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Akihiko Odaki [Wed, 16 Jun 2021 14:17:21 +0000 (23:17 +0900)]
coreaudio: Fix output stream format settings
Before commit 7d6948cd98cf5ad8a3458a4ce7fdbcb79bcd1212, it was coded to
retrieve the initial output stream format settings, modify the frame
rate, and set again. However, I removed a frame rate modification code by
mistake in the commit. It also assumes the initial output stream format
is consistent with what QEMU expects, but that expectation is not in the
code, which makes it harder to understand and will lead to breakage if
the initial settings change.
This change explicitly sets all of the output stream settings to solve
these problems.
hw/audio/sb16: Avoid assertion by restricting I/O sampling rate range
While the SB16 seems to work up to 48000 Hz, the "Sound Blaster Series
Hardware Programming Guide" limit the sampling range from 4000 Hz to
44100 Hz (Section 3-9, 3-10: Digitized Sound I/O Programming, tables
3-2 and 3-3).
Later, section 6-15 (DSP Commands) is more specific regarding the 41h /
42h registers (Set digitized sound output sampling rate):
Valid sampling rates range from 5000 to 45000 Hz inclusive.
There is no comment regarding error handling if the register is filled
with an out-of-range value. (See also section 3-28 "8-bit or 16-bit
Auto-initialize Transfer"). Assume limits are enforced in hardware.
This fixes triggering an assertion in audio_calloc():
Volker Rümelin [Mon, 17 May 2021 19:46:04 +0000 (21:46 +0200)]
jackaudio: avoid that the client name contains the word (NULL)
Currently with jackaudio client name and qemu guest name unset,
the JACK client names are out-(NULL) and in-(NULL). These names
are user visible in the patch bay. Replace the function call to
qemu_get_vm_name() with a call to audio_application_name() which
replaces NULL with "qemu" to have more descriptive names.
Volker Rümelin [Mon, 17 May 2021 19:46:03 +0000 (21:46 +0200)]
audio: move code to audio/audio.c
Move the code to generate the pa_context_new() application name
argument to a function in audio/audio.c. The new function
audio_application_name() will also be used in the jackaudio
backend.
Volker Rümelin [Mon, 17 May 2021 19:46:02 +0000 (21:46 +0200)]
paaudio: remove unused stream flags
In current code there are no calls to pa_stream_get_latency()
or pa_stream_get_time() to receive latency or time information.
Remove the flags PA_STREAM_INTERPOLATE_TIMING and
PA_STREAM_AUTO_TIMING_UPDATE which instruct PulseAudio to
calculate this information in regular intervals.
Peter Maydell [Thu, 17 Jun 2021 09:42:57 +0000 (10:42 +0100)]
Merge remote-tracking branch 'remotes/rth-gitlab/tags/pull-c11-20210615' into staging
Change to -std=gnu11.
Replace QEMU_GENERIC with _Generic.
Remove configure detect of _Static_assert.
# gpg: Signature made Wed 16 Jun 2021 02:32:32 BST
# gpg: using RSA key 7A481E78868B4DB6A85A05C064DF38E8AF7E215F
# gpg: issuer "richard.henderson@linaro.org"
# gpg: Good signature from "Richard Henderson <richard.henderson@linaro.org>" [full]
# Primary key fingerprint: 7A48 1E78 868B 4DB6 A85A 05C0 64DF 38E8 AF7E 215F
* remotes/rth-gitlab/tags/pull-c11-20210615:
configure: Remove probe for _Static_assert
qemu/compiler: Remove QEMU_GENERIC
include/qemu/lockable: Use _Generic instead of QEMU_GENERIC
util: Use unique type for QemuRecMutex in thread-posix.h
util: Pass file+line to qemu_rec_mutex_unlock_impl
util: Use real functions for thread-posix QemuRecMutex
softfloat: Use _Generic instead of QEMU_GENERIC
configure: Use -std=gnu11
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Peter Maydell [Mon, 14 Jun 2021 15:09:43 +0000 (16:09 +0100)]
include/qemu/int128.h: Add function to create Int128 from int64_t
int128_make64() creates an Int128 from an unsigned 64 bit value; add
a function int128_makes64() creating an Int128 from a signed 64 bit
value.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Message-id: 20210614151007.4545-34-peter.maydell@linaro.org
Peter Maydell [Mon, 14 Jun 2021 15:09:26 +0000 (16:09 +0100)]
bitops.h: Provide hswap32(), hswap64(), wswap64() swapping operations
Currently the ARM SVE helper code defines locally some utility
functions for swapping 16-bit halfwords within 32-bit or 64-bit
values and for swapping 32-bit words within 64-bit values,
parallel to the byte-swapping bswap16/32/64 functions.
We want these also for the ARM MVE code, and they're potentially
generally useful for other targets, so move them to bitops.h.
(We don't put them in bswap.h with the bswap* functions because
they are implemented in terms of the rotate operations also
defined in bitops.h, and including bitops.h from bswap.h seems
better avoided.)
Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Message-id: 20210614151007.4545-17-peter.maydell@linaro.org
Peter Maydell [Mon, 14 Jun 2021 15:09:23 +0000 (16:09 +0100)]
target/arm: Move expand_pred_b() data to vec_helper.c
For MVE, we want to re-use the large data table from expand_pred_b().
Move the data table to vec_helper.c so it is no longer in an SVE
specific source file.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20210614151007.4545-14-peter.maydell@linaro.org
Peter Maydell [Mon, 14 Jun 2021 15:09:20 +0000 (16:09 +0100)]
target/arm: Add framework for MVE decode
Add the framework for decoding MVE insns, with the necessary new
files and the meson.build rules, but no actual content yet.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20210614151007.4545-11-peter.maydell@linaro.org
Peter Maydell [Mon, 14 Jun 2021 15:09:19 +0000 (16:09 +0100)]
target/arm: Implement MVE LETP insn
Implement the MVE LETP insn. This is like the existing LE loop-end
insn, but it must perform an FPU-enabled check, and on loop-exit it
resets LTPSIZE to 4.
To accommodate the requirement to do something on loop-exit, we drop
the use of condlabel and instead manage both the TB exits manually,
in the same way we already do in trans_WLS().
The other MVE-specific change to the LE insn is that we must raise an
INVSTATE UsageFault insn if LTPSIZE is not 4.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20210614151007.4545-10-peter.maydell@linaro.org
projects / qemu.git / log