From: Paolo Bonzini <pbonzini@redhat.com>
To: qemu-devel@nongnu.org
Cc: armbru@redhat.com, kwolf@redhat.com
Subject: [PATCH 2/5] monitor: cleanup fetching of QMP requests
Date: Thu, 18 May 2023 12:18:20 +0200 [thread overview]
Message-ID: <20230518101823.992158-3-pbonzini@redhat.com> (raw)
In-Reply-To: <20230518101823.992158-1-pbonzini@redhat.com>
Use a continue statement so that "after going to sleep" is treated the same
way as "after processing a request". Pull the monitor_lock critical
section out of monitor_qmp_requests_pop_any_with_lock() and protect
qmp_dispatcher_co_shutdown with the monitor_lock.
The two changes are complex to separate because monitor_qmp_dispatcher_co()
previously had a complicated logic to check for shutdown both before
and after going to sleep.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
monitor/monitor.c | 9 +++++++--
monitor/qmp.c | 38 ++++++++++++++------------------------
2 files changed, 21 insertions(+), 26 deletions(-)
diff --git a/monitor/monitor.c b/monitor/monitor.c
index c4ed2547c25f..042a1ab918f9 100644
--- a/monitor/monitor.c
+++ b/monitor/monitor.c
@@ -56,7 +56,10 @@ IOThread *mon_iothread;
/* Coroutine to dispatch the requests received from I/O thread */
Coroutine *qmp_dispatcher_co;
-/* Set to true when the dispatcher coroutine should terminate */
+/*
+ * Set to true when the dispatcher coroutine should terminate. Protected
+ * by monitor_lock.
+ */
bool qmp_dispatcher_co_shutdown;
/*
@@ -679,7 +682,9 @@ void monitor_cleanup(void)
* we'll just leave them in the queue without sending a response
* and monitor_data_destroy() will free them.
*/
- qmp_dispatcher_co_shutdown = true;
+ WITH_QEMU_LOCK_GUARD(&monitor_lock) {
+ qmp_dispatcher_co_shutdown = true;
+ }
if (!qatomic_xchg(&qmp_dispatcher_co_busy, true)) {
aio_co_wake(qmp_dispatcher_co);
}
diff --git a/monitor/qmp.c b/monitor/qmp.c
index f0cc6dc886f8..7516b92a4d3e 100644
--- a/monitor/qmp.c
+++ b/monitor/qmp.c
@@ -178,7 +178,10 @@ static QMPRequest *monitor_qmp_requests_pop_any_with_lock(void)
Monitor *mon;
MonitorQMP *qmp_mon;
- QEMU_LOCK_GUARD(&monitor_lock);
+ /* On shutdown, don't take any more requests from the queue */
+ if (qmp_dispatcher_co_shutdown) {
+ return NULL;
+ }
QTAILQ_FOREACH(mon, &mon_list, entry) {
if (!monitor_is_qmp(mon)) {
@@ -215,6 +218,10 @@ void coroutine_fn monitor_qmp_dispatcher_co(void *data)
MonitorQMP *mon;
while (true) {
+ /*
+ * busy must be set to true again by whoever
+ * rescheduled us to avoid double scheduling
+ */
assert(qatomic_mb_read(&qmp_dispatcher_co_busy) == true);
/*
@@ -224,36 +231,18 @@ void coroutine_fn monitor_qmp_dispatcher_co(void *data)
*/
qatomic_mb_set(&qmp_dispatcher_co_busy, false);
- /* On shutdown, don't take any more requests from the queue */
- if (qmp_dispatcher_co_shutdown) {
- qatomic_set(&qmp_dispatcher_co, NULL);
- return;
+ WITH_QEMU_LOCK_GUARD(&monitor_lock) {
+ req_obj = monitor_qmp_requests_pop_any_with_lock();
}
- while (!(req_obj = monitor_qmp_requests_pop_any_with_lock())) {
+ if (!req_obj) {
/*
* No more requests to process. Wait to be reentered from
* handle_qmp_command() when it pushes more requests, or
* from monitor_cleanup() when it requests shutdown.
*/
- if (!qmp_dispatcher_co_shutdown) {
- qemu_coroutine_yield();
-
- /*
- * busy must be set to true again by whoever
- * rescheduled us to avoid double scheduling
- */
- assert(qatomic_xchg(&qmp_dispatcher_co_busy, false) == true);
- }
-
- /*
- * qmp_dispatcher_co_shutdown may have changed if we
- * yielded and were reentered from monitor_cleanup()
- */
- if (qmp_dispatcher_co_shutdown) {
- qatomic_set(&qmp_dispatcher_co, NULL);
- return;
- }
+ qemu_coroutine_yield();
+ continue;
}
trace_monitor_qmp_in_band_dequeue(req_obj,
@@ -342,6 +331,7 @@ void coroutine_fn monitor_qmp_dispatcher_co(void *data)
aio_co_schedule(iohandler_get_aio_context(), qmp_dispatcher_co);
qemu_coroutine_yield();
}
+ qatomic_set(&qmp_dispatcher_co, NULL);
}
static void handle_qmp_command(void *opaque, QObject *req, Error *err)
--
2.40.1
next prev parent reply other threads:[~2023-05-18 10:18 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-05-18 10:18 [PATCH 0/5] qmp: cleanup QMP dispatcher coroutine Paolo Bonzini
2023-05-18 10:18 ` [PATCH 1/5] monitor: cleanup detection of qmp_dispatcher_co shutting down Paolo Bonzini
2023-05-18 10:18 ` Paolo Bonzini [this message]
2023-05-18 10:18 ` [PATCH 3/5] monitor: introduce qmp_dispatcher_co_wake Paolo Bonzini
2023-05-18 10:18 ` [PATCH 4/5] monitor: extract request dequeuing to a new function Paolo Bonzini
2023-05-18 10:18 ` [PATCH 5/5] monitor: do not use mb_read/mb_set Paolo Bonzini
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20230518101823.992158-3-pbonzini@redhat.com \
--to=pbonzini@redhat.com \
--cc=armbru@redhat.com \
--cc=kwolf@redhat.com \
--cc=qemu-devel@nongnu.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).