From: Peter Xu <peterx@redhat.com>
To: qemu-devel@nongnu.org
Cc: Alexey Perevalov <a.perevalov@samsung.com>,
"Daniel P . Berrange" <berrange@redhat.com>,
Juan Quintela <quintela@redhat.com>,
Andrea Arcangeli <aarcange@redhat.com>,
"Dr . David Alan Gilbert" <dgilbert@redhat.com>,
peterx@redhat.com
Subject: [Qemu-devel] [PATCH v6 06/28] migration: allow dst vm pause on postcopy
Date: Thu, 8 Feb 2018 18:31:10 +0800 [thread overview]
Message-ID: <20180208103132.28452-7-peterx@redhat.com> (raw)
In-Reply-To: <20180208103132.28452-1-peterx@redhat.com>
When there is IO error on the incoming channel (e.g., network down),
instead of bailing out immediately, we allow the dst vm to switch to the
new POSTCOPY_PAUSE state. Currently it is still simple - it waits the
new semaphore, until someone poke it for another attempt.
One note is that here on ram loading thread we cannot detect the
POSTCOPY_ACTIVE state, but we need to detect the more specific
POSTCOPY_INCOMING_RUNNING state, to make sure we have already loaded all
the device states.
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: Peter Xu <peterx@redhat.com>
---
migration/migration.c | 1 +
migration/migration.h | 3 +++
migration/savevm.c | 63 ++++++++++++++++++++++++++++++++++++++++++++++++--
migration/trace-events | 2 ++
4 files changed, 67 insertions(+), 2 deletions(-)
diff --git a/migration/migration.c b/migration/migration.c
index de5ee9c1d4..fc0d0e0e31 100644
--- a/migration/migration.c
+++ b/migration/migration.c
@@ -155,6 +155,7 @@ MigrationIncomingState *migration_incoming_get_current(void)
memset(&mis_current, 0, sizeof(MigrationIncomingState));
qemu_mutex_init(&mis_current.rp_mutex);
qemu_event_init(&mis_current.main_thread_load_event, false);
+ qemu_sem_init(&mis_current.postcopy_pause_sem_dst, 0);
once = true;
}
return &mis_current;
diff --git a/migration/migration.h b/migration/migration.h
index 3e4af0a831..27ba256c40 100644
--- a/migration/migration.h
+++ b/migration/migration.h
@@ -61,6 +61,9 @@ struct MigrationIncomingState {
/* The coroutine we should enter (back) after failover */
Coroutine *migration_incoming_co;
QemuSemaphore colo_incoming_sem;
+
+ /* notify PAUSED postcopy incoming migrations to try to continue */
+ QemuSemaphore postcopy_pause_sem_dst;
};
MigrationIncomingState *migration_incoming_get_current(void);
diff --git a/migration/savevm.c b/migration/savevm.c
index 76fcc2b1c3..d83c1e846a 100644
--- a/migration/savevm.c
+++ b/migration/savevm.c
@@ -1544,8 +1544,8 @@ static int loadvm_postcopy_ram_handle_discard(MigrationIncomingState *mis,
*/
static void *postcopy_ram_listen_thread(void *opaque)
{
- QEMUFile *f = opaque;
MigrationIncomingState *mis = migration_incoming_get_current();
+ QEMUFile *f = mis->from_src_file;
int load_res;
migrate_set_state(&mis->state, MIGRATION_STATUS_ACTIVE,
@@ -1559,6 +1559,14 @@ static void *postcopy_ram_listen_thread(void *opaque)
*/
qemu_file_set_blocking(f, true);
load_res = qemu_loadvm_state_main(f, mis);
+
+ /*
+ * This is tricky, but, mis->from_src_file can change after it
+ * returns, when postcopy recovery happened. In the future, we may
+ * want a wrapper for the QEMUFile handle.
+ */
+ f = mis->from_src_file;
+
/* And non-blocking again so we don't block in any cleanup */
qemu_file_set_blocking(f, false);
@@ -1641,7 +1649,7 @@ static int loadvm_postcopy_handle_listen(MigrationIncomingState *mis)
/* Start up the listening thread and wait for it to signal ready */
qemu_sem_init(&mis->listen_thread_sem, 0);
qemu_thread_create(&mis->listen_thread, "postcopy/listen",
- postcopy_ram_listen_thread, mis->from_src_file,
+ postcopy_ram_listen_thread, NULL,
QEMU_THREAD_DETACHED);
qemu_sem_wait(&mis->listen_thread_sem);
qemu_sem_destroy(&mis->listen_thread_sem);
@@ -2026,11 +2034,44 @@ void qemu_loadvm_state_cleanup(void)
}
}
+/* Return true if we should continue the migration, or false. */
+static bool postcopy_pause_incoming(MigrationIncomingState *mis)
+{
+ trace_postcopy_pause_incoming();
+
+ migrate_set_state(&mis->state, MIGRATION_STATUS_POSTCOPY_ACTIVE,
+ MIGRATION_STATUS_POSTCOPY_PAUSED);
+
+ assert(mis->from_src_file);
+ qemu_file_shutdown(mis->from_src_file);
+ qemu_fclose(mis->from_src_file);
+ mis->from_src_file = NULL;
+
+ assert(mis->to_src_file);
+ qemu_file_shutdown(mis->to_src_file);
+ qemu_mutex_lock(&mis->rp_mutex);
+ qemu_fclose(mis->to_src_file);
+ mis->to_src_file = NULL;
+ qemu_mutex_unlock(&mis->rp_mutex);
+
+ error_report("Detected IO failure for postcopy. "
+ "Migration paused.");
+
+ while (mis->state == MIGRATION_STATUS_POSTCOPY_PAUSED) {
+ qemu_sem_wait(&mis->postcopy_pause_sem_dst);
+ }
+
+ trace_postcopy_pause_incoming_continued();
+
+ return true;
+}
+
static int qemu_loadvm_state_main(QEMUFile *f, MigrationIncomingState *mis)
{
uint8_t section_type;
int ret = 0;
+retry:
while (true) {
section_type = qemu_get_byte(f);
@@ -2075,6 +2116,24 @@ static int qemu_loadvm_state_main(QEMUFile *f, MigrationIncomingState *mis)
out:
if (ret < 0) {
qemu_file_set_error(f, ret);
+
+ /*
+ * Detect whether it is:
+ *
+ * 1. postcopy running (after receiving all device data, which
+ * must be in POSTCOPY_INCOMING_RUNNING state. Note that
+ * POSTCOPY_INCOMING_LISTENING is still not enough, it's
+ * still receiving device states).
+ * 2. network failure (-EIO)
+ *
+ * If so, we try to wait for a recovery.
+ */
+ if (postcopy_state_get() == POSTCOPY_INCOMING_RUNNING &&
+ ret == -EIO && postcopy_pause_incoming(mis)) {
+ /* Reset f to point to the newly created channel */
+ f = mis->from_src_file;
+ goto retry;
+ }
}
return ret;
}
diff --git a/migration/trace-events b/migration/trace-events
index 7afbb3e791..8685a62c98 100644
--- a/migration/trace-events
+++ b/migration/trace-events
@@ -100,6 +100,8 @@ open_return_path_on_source(void) ""
open_return_path_on_source_continue(void) ""
postcopy_start(void) ""
postcopy_pause_continued(void) ""
+postcopy_pause_incoming(void) ""
+postcopy_pause_incoming_continued(void) ""
postcopy_start_set_run(void) ""
source_return_path_thread_bad_end(void) ""
source_return_path_thread_end(void) ""
--
2.14.3
next prev parent reply other threads:[~2018-02-08 10:32 UTC|newest]
Thread overview: 56+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-02-08 10:31 [Qemu-devel] [PATCH v6 00/28] Migration: postcopy failure recovery Peter Xu
2018-02-08 10:31 ` [Qemu-devel] [PATCH v6 01/28] migration: better error handling with QEMUFile Peter Xu
2018-02-08 10:31 ` [Qemu-devel] [PATCH v6 02/28] migration: reuse mis->userfault_quit_fd Peter Xu
2018-02-08 10:31 ` [Qemu-devel] [PATCH v6 03/28] migration: provide postcopy_fault_thread_notify() Peter Xu
2018-02-08 10:31 ` [Qemu-devel] [PATCH v6 04/28] migration: new postcopy-pause state Peter Xu
2018-02-08 10:31 ` [Qemu-devel] [PATCH v6 05/28] migration: implement "postcopy-pause" src logic Peter Xu
2018-02-08 10:31 ` Peter Xu [this message]
2018-02-08 10:31 ` [Qemu-devel] [PATCH v6 07/28] migration: allow src return path to pause Peter Xu
2018-02-08 10:31 ` [Qemu-devel] [PATCH v6 08/28] migration: allow send_rq to fail Peter Xu
2018-02-08 10:31 ` [Qemu-devel] [PATCH v6 09/28] migration: allow fault thread to pause Peter Xu
2018-02-08 10:31 ` [Qemu-devel] [PATCH v6 10/28] qmp: hmp: add migrate "resume" option Peter Xu
2018-02-08 10:31 ` [Qemu-devel] [PATCH v6 11/28] migration: pass MigrationState to migrate_init() Peter Xu
2018-02-08 10:31 ` [Qemu-devel] [PATCH v6 12/28] migration: rebuild channel on source Peter Xu
2018-02-08 10:31 ` [Qemu-devel] [PATCH v6 13/28] migration: new state "postcopy-recover" Peter Xu
2018-02-08 10:31 ` [Qemu-devel] [PATCH v6 14/28] migration: wakeup dst ram-load-thread for recover Peter Xu
2018-02-08 10:31 ` [Qemu-devel] [PATCH v6 15/28] migration: new cmd MIG_CMD_RECV_BITMAP Peter Xu
2018-02-08 10:31 ` [Qemu-devel] [PATCH v6 16/28] migration: new message MIG_RP_MSG_RECV_BITMAP Peter Xu
2018-02-08 10:31 ` [Qemu-devel] [PATCH v6 17/28] migration: new cmd MIG_CMD_POSTCOPY_RESUME Peter Xu
2018-02-08 10:31 ` [Qemu-devel] [PATCH v6 18/28] migration: new message MIG_RP_MSG_RESUME_ACK Peter Xu
2018-02-08 10:31 ` [Qemu-devel] [PATCH v6 19/28] migration: introduce SaveVMHandlers.resume_prepare Peter Xu
2018-02-08 10:31 ` [Qemu-devel] [PATCH v6 20/28] migration: synchronize dirty bitmap for resume Peter Xu
2018-02-08 10:31 ` [Qemu-devel] [PATCH v6 21/28] migration: setup ramstate " Peter Xu
2018-02-13 18:17 ` Dr. David Alan Gilbert
2018-02-14 4:20 ` Peter Xu
2018-02-14 18:40 ` Dr. David Alan Gilbert
2018-02-22 7:34 ` Peter Xu
2018-02-08 10:31 ` [Qemu-devel] [PATCH v6 22/28] migration: final handshake for the resume Peter Xu
2018-02-08 10:31 ` [Qemu-devel] [PATCH v6 23/28] migration: init dst in migration_object_init too Peter Xu
2018-02-08 10:31 ` [Qemu-devel] [PATCH v6 24/28] io: let watcher of the channel run in same ctx Peter Xu
2018-02-08 10:31 ` [Qemu-devel] [PATCH v6 25/28] qmp/migration: new command migrate-recover Peter Xu
2018-02-13 18:56 ` Dr. David Alan Gilbert
2018-02-14 4:30 ` Peter Xu
2018-02-08 10:31 ` [Qemu-devel] [PATCH v6 26/28] hmp/migration: add migrate_recover command Peter Xu
2018-02-13 19:45 ` Dr. David Alan Gilbert
2018-02-14 4:48 ` Peter Xu
2018-02-08 10:31 ` [Qemu-devel] [PATCH v6 27/28] migration/qmp: add command migrate-pause Peter Xu
2018-02-13 20:11 ` Dr. David Alan Gilbert
2018-02-14 4:56 ` Peter Xu
2018-02-14 18:56 ` Dr. David Alan Gilbert
2018-02-22 7:43 ` Peter Xu
2018-02-28 20:14 ` Dr. David Alan Gilbert
2018-03-13 9:13 ` Peter Xu
2018-03-13 12:35 ` Dr. David Alan Gilbert
2018-03-13 15:57 ` Peter Xu
2018-03-13 16:05 ` Dr. David Alan Gilbert
2018-02-08 10:31 ` [Qemu-devel] [PATCH v6 28/28] migration/hmp: add migrate_pause command Peter Xu
2018-02-13 20:14 ` Dr. David Alan Gilbert
2018-02-08 10:38 ` [Qemu-devel] [PATCH v6 00/28] Migration: postcopy failure recovery Peter Xu
2018-02-08 11:25 ` no-reply
2018-02-09 6:07 ` Peter Xu
2018-02-09 6:19 ` Fam Zheng
2018-02-08 11:27 ` no-reply
2018-02-08 11:28 ` no-reply
2018-02-08 11:30 ` no-reply
2018-02-08 12:18 ` no-reply
2018-02-09 6:28 ` 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=20180208103132.28452-7-peterx@redhat.com \
--to=peterx@redhat.com \
--cc=a.perevalov@samsung.com \
--cc=aarcange@redhat.com \
--cc=berrange@redhat.com \
--cc=dgilbert@redhat.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).