[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Qemu-devel] [RFC v2 00/22] QMP: out-of-band (OOB) execution support
From: |
Peter Xu |
Subject: |
[Qemu-devel] [RFC v2 00/22] QMP: out-of-band (OOB) execution support |
Date: |
Fri, 29 Sep 2017 11:38:22 +0800 |
This is RFC version 2 of monitor OOB series. It is based on following
series:
[PATCH v4 0/5] iothread: allow to create internal iothreads
(queued in Stefan's tree, pre-requisite for iothread usage)
[PATCH v2 0/4] chardev: support non-default gcontext
(queued in Paolo's tree, fix chardev reconnect issue)
Big changes from version 1:
* Use IOThread for monitor IO thread.
* Request queue switched from global queue to per-monitor queue, with
basic flow control (currently 8 outstanding requests per monitor).
* Moved all documents into a single patch.
* Start to use "capabilities" field in QMP greeting message, to
declare OOB support on server side.
* Added per-monitor response queue, so finally it looks like this:
queue/kick queue/kick
JSON Parser ======> QMP Dispatcher =====> Responder
/|\ | (3) /|\ | (4) | /|\
(1) | | (2) | | | |
| | | \|/ (6)| |(5)
| | main thread | |
| | | |
| +--------> monitor IO thread <-------+ |
+-----------/ \----------+
v2 changelog:
- use 10-char hash for git commit ID [Eric]
- instead of changing qstring_get_str(), add new
qstring_get_try_str(), and rename qobject_get_str() to
qobject_get_try_str() to follow the naming rule [Eric]
- add one more patch to let object_property_get_str() leverage the new
qobject_get_try_str(). [Eric]
- introduce JSONMessageEmitFunc in the JSON parser patch [Eric]
- use per-monitor request queue, rather than a global queue
- add queue flow control, when full (current queue depth: 8), send
event to notify client (two new patches added for this)
- stop monitor thread first before destroying monitor objects [Dan]
- document that client should drop responses with unknown IDs.
- move all the documents into a single standalone patch.
- use iothread for the monitor io thread.
- respond in IO thread, rather than the dispatcher
- new patch: in QMP greeting message, add 'oob' field in
'capabilities', so that client can explicitly know whether server
supports oob
TODO:
- qmp qtest for oob
Please review. Thanks,
========== below is original RFC cover letter ===============
This series was born from this one:
https://lists.gnu.org/archive/html/qemu-devel/2017-08/msg04310.html
The design comes from Markus, and also the whole-bunch-of discussions
in previous thread. My heartful thanks to Markus, Daniel, Dave,
Stefan, etc. on discussing the topic (...again!), providing shiny
ideas and suggestions. Finally we got such a solution that seems to
satisfy everyone.
I re-started the versioning since this series is totally different
from previous one. Now it's version 1.
In case new reviewers come along the way without reading previous
discussions, I will try to do a summary on what this is all about.
What is OOB execution?
======================
It's the shortcut of Out-Of-Band execution, its name is given by
Markus. It's a way to quickly execute a QMP request. Say, originally
QMP is going throw these steps:
JSON Parser --> QMP Dispatcher --> Respond
/|\ (2) (3) |
(1) | \|/ (4)
+--------- main thread --------+
The requests are executed by the so-called QMP-dispatcher after the
JSON is parsed. If OOB is on, we run the command directly in the
parser and quickly returns.
Yeah I know in current code the parser calls dispatcher directly
(please see handle_qmp_command()). However it's not true again after
this series (parser will has its own IO thread, and dispatcher will
still be run in main thread). So this OOB does brings something
different.
There are more details on why OOB and the difference/relationship
between OOB, async QMP, block/general jobs, etc.. but IMHO that's
slightly out of topic (and believe me, it's not easy for me to
summarize that). For more information, please refers to [1].
Summary ends here.
Some Implementation Details
===========================
Again, I mentioned that the old QMP workflow is this:
JSON Parser --> QMP Dispatcher --> Respond
/|\ (2) (3) |
(1) | \|/ (4)
+--------- main thread --------+
What this series does is, firstly:
JSON Parser QMP Dispatcher --> Respond
/|\ | /|\ (4) |
| | (2) | (3) | (5)
(1) | +-----> | \|/
+--------- main thread <-------+
And further:
queue/kick
JSON Parser ======> QMP Dispatcher --> Respond
/|\ | (3) /|\ (4) |
(1) | | (2) | | (5)
| \|/ | \|/
IO thread main thread <-------+
Then it introduced the "allow-oob" parameter in QAPI schema to define
commands, and "run-oob" flag to let oob-allowed command to run in the
parser.
The last patch enables this for "migrate-incoming" command.
Please review. Thanks.
[1] https://lists.gnu.org/archive/html/qemu-devel/2017-08/msg04310.html
Peter Xu (22):
char-io: fix possible race on IOWatchPoll
qobject: introduce qstring_get_try_str()
qobject: introduce qobject_get_try_str()
qobject: let object_property_get_str() use new API
monitor: move skip_flush into monitor_data_init
qjson: add "opaque" field to JSONMessageParser
monitor: move the cur_mon hack deeper for QMP
monitor: unify global init
monitor: create monitor dedicate iothread
monitor: allow to use IO thread for parsing
monitor: introduce monitor_qmp_respond()
monitor: let mon_list be tail queue
monitor: separate QMP parser and dispatcher
qmp: add new event "request-dropped"
monitor: send event when request queue full
monitor: enable IO thread for (qmp & !mux) typed
qapi: introduce new cmd option "allow-oob"
qmp: support out-of-band (oob) execution
qmp: let migrate-incoming allow out-of-band
qmp: isolate responses into io thread
qmp: introduce QMPCapability
docs: update QMP documents for OOB commands
chardev/char-io.c | 16 +-
docs/devel/qapi-code-gen.txt | 51 ++++-
docs/interop/qmp-spec.txt | 24 ++-
include/monitor/monitor.h | 2 +-
include/qapi/qmp/dispatch.h | 2 +
include/qapi/qmp/json-streamer.h | 10 +-
include/qapi/qmp/qstring.h | 2 +
monitor.c | 441 ++++++++++++++++++++++++++++++++-------
qapi-schema.json | 48 +++++
qapi/introspect.json | 6 +-
qapi/migration.json | 3 +-
qapi/qmp-dispatch.c | 34 +++
qga/main.c | 5 +-
qobject/json-streamer.c | 6 +-
qobject/qjson.c | 5 +-
qobject/qstring.c | 21 ++
qom/object.c | 9 +-
scripts/qapi-commands.py | 19 +-
scripts/qapi-introspect.py | 10 +-
scripts/qapi.py | 15 +-
scripts/qapi2texi.py | 2 +-
tests/libqtest.c | 5 +-
tests/qapi-schema/test-qapi.py | 2 +-
tests/qmp-test.c | 2 +-
trace-events | 2 +
vl.c | 3 +-
26 files changed, 630 insertions(+), 115 deletions(-)
--
2.13.5
- [Qemu-devel] [RFC v2 00/22] QMP: out-of-band (OOB) execution support,
Peter Xu <=
- [Qemu-devel] [RFC v2 01/22] char-io: fix possible race on IOWatchPoll, Peter Xu, 2017/09/28
- [Qemu-devel] [RFC v2 02/22] qobject: introduce qstring_get_try_str(), Peter Xu, 2017/09/28
- [Qemu-devel] [RFC v2 03/22] qobject: introduce qobject_get_try_str(), Peter Xu, 2017/09/28
- [Qemu-devel] [RFC v2 04/22] qobject: let object_property_get_str() use new API, Peter Xu, 2017/09/28
- [Qemu-devel] [RFC v2 05/22] monitor: move skip_flush into monitor_data_init, Peter Xu, 2017/09/28
- [Qemu-devel] [RFC v2 06/22] qjson: add "opaque" field to JSONMessageParser, Peter Xu, 2017/09/28
- [Qemu-devel] [RFC v2 08/22] monitor: unify global init, Peter Xu, 2017/09/28
- [Qemu-devel] [RFC v2 07/22] monitor: move the cur_mon hack deeper for QMP, Peter Xu, 2017/09/28
- [Qemu-devel] [RFC v2 09/22] monitor: create monitor dedicate iothread, Peter Xu, 2017/09/28
- [Qemu-devel] [RFC v2 10/22] monitor: allow to use IO thread for parsing, Peter Xu, 2017/09/28