From: Peter Xu <peterx@redhat.com>
To: qemu-devel@nongnu.org
Cc: peterx@redhat.com, Fabiano Rosas <farosas@suse.de>,
Juan Quintela <quintela@redhat.com>,
Joao Martins <joao.m.martins@oracle.com>
Subject: [PATCH 2/3] migration: Add migration_downtime_start|end() helpers
Date: Thu, 26 Oct 2023 11:53:36 -0400 [thread overview]
Message-ID: <20231026155337.596281-3-peterx@redhat.com> (raw)
In-Reply-To: <20231026155337.596281-1-peterx@redhat.com>
Unify the three users on recording downtimes with the same pair of helpers.
Signed-off-by: Peter Xu <peterx@redhat.com>
---
migration/migration.c | 37 ++++++++++++++++++++++++-------------
1 file changed, 24 insertions(+), 13 deletions(-)
diff --git a/migration/migration.c b/migration/migration.c
index f8a54ff4d1..70d775942a 100644
--- a/migration/migration.c
+++ b/migration/migration.c
@@ -101,6 +101,24 @@ static int migration_maybe_pause(MigrationState *s,
static void migrate_fd_cancel(MigrationState *s);
static int close_return_path_on_source(MigrationState *s);
+static void migration_downtime_start(MigrationState *s)
+{
+ s->downtime_start = qemu_clock_get_ms(QEMU_CLOCK_REALTIME);
+}
+
+static void migration_downtime_end(MigrationState *s)
+{
+ int64_t now = qemu_clock_get_ms(QEMU_CLOCK_REALTIME);
+
+ /*
+ * If downtime already set, should mean that postcopy already set it,
+ * then that should be the real downtime already.
+ */
+ if (!s->downtime) {
+ s->downtime = now - s->downtime_start;
+ }
+}
+
static bool migration_needs_multiple_sockets(void)
{
return migrate_multifd() || migrate_postcopy_preempt();
@@ -2142,7 +2160,7 @@ static int postcopy_start(MigrationState *ms, Error **errp)
qemu_mutex_lock_iothread();
trace_postcopy_start_set_run();
- ms->downtime_start = qemu_clock_get_ms(QEMU_CLOCK_REALTIME);
+ migration_downtime_start(ms);
qemu_system_wakeup_request(QEMU_WAKEUP_REASON_OTHER, NULL);
global_state_store();
@@ -2246,7 +2264,7 @@ static int postcopy_start(MigrationState *ms, Error **errp)
ms->postcopy_after_devices = true;
migration_call_notifiers(ms);
- ms->downtime = qemu_clock_get_ms(QEMU_CLOCK_REALTIME) - ms->downtime_start;
+ migration_downtime_end(ms);
qemu_mutex_unlock_iothread();
@@ -2342,7 +2360,7 @@ static int migration_completion_precopy(MigrationState *s,
int ret;
qemu_mutex_lock_iothread();
- s->downtime_start = qemu_clock_get_ms(QEMU_CLOCK_REALTIME);
+ migration_downtime_start(s);
qemu_system_wakeup_request(QEMU_WAKEUP_REASON_OTHER, NULL);
s->vm_old_state = runstate_get();
@@ -2699,15 +2717,8 @@ static void migration_calculate_complete(MigrationState *s)
int64_t end_time = qemu_clock_get_ms(QEMU_CLOCK_REALTIME);
int64_t transfer_time;
+ migration_downtime_end(s);
s->total_time = end_time - s->start_time;
- if (!s->downtime) {
- /*
- * It's still not set, so we are precopy migration. For
- * postcopy, downtime is calculated during postcopy_start().
- */
- s->downtime = end_time - s->downtime_start;
- }
-
transfer_time = s->total_time - s->setup_time;
if (transfer_time) {
s->mbps = ((double) bytes * 8.0) / transfer_time / 1000;
@@ -3126,7 +3137,7 @@ static void bg_migration_vm_start_bh(void *opaque)
s->vm_start_bh = NULL;
vm_start();
- s->downtime = qemu_clock_get_ms(QEMU_CLOCK_REALTIME) - s->downtime_start;
+ migration_downtime_end(s);
}
/**
@@ -3193,7 +3204,7 @@ static void *bg_migration_thread(void *opaque)
s->setup_time = qemu_clock_get_ms(QEMU_CLOCK_HOST) - setup_start;
trace_migration_thread_setup_complete();
- s->downtime_start = qemu_clock_get_ms(QEMU_CLOCK_REALTIME);
+ migration_downtime_start(s);
qemu_mutex_lock_iothread();
--
2.41.0
next prev parent reply other threads:[~2023-10-26 15:54 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-10-26 15:53 [PATCH 0/3] migration: Downtime tracepoints Peter Xu
2023-10-26 15:53 ` [PATCH 1/3] migration: Set downtime_start even for postcopy Peter Xu
2023-10-26 17:05 ` Fabiano Rosas
2023-10-26 15:53 ` Peter Xu [this message]
2023-10-26 17:11 ` [PATCH 2/3] migration: Add migration_downtime_start|end() helpers Fabiano Rosas
2023-10-26 15:53 ` [PATCH 3/3] migration: Add per vmstate downtime tracepoints Peter Xu
2023-10-26 16:06 ` [PATCH 0/3] migration: Downtime tracepoints Joao Martins
2023-10-26 17:03 ` Peter Xu
2023-10-26 18:18 ` Peter Xu
2023-10-26 19:33 ` Joao Martins
2023-10-26 20:07 ` Peter Xu
2023-10-27 8:58 ` Joao Martins
2023-10-27 14:41 ` Peter Xu
2023-10-27 22:17 ` Joao Martins
2023-10-30 15:13 ` Peter Xu
2023-10-30 16:09 ` Peter Xu
2023-10-30 16:11 ` Joao Martins
2023-10-26 19:01 ` [PATCH 4/3] migration: Add tracepoints for downtime checkpoints Peter Xu
2023-10-26 19:43 ` Joao Martins
2023-10-26 20:08 ` Peter Xu
2023-10-26 20:14 ` Peter Xu
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=20231026155337.596281-3-peterx@redhat.com \
--to=peterx@redhat.com \
--cc=farosas@suse.de \
--cc=joao.m.martins@oracle.com \
--cc=qemu-devel@nongnu.org \
--cc=quintela@redhat.com \
/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).