From: Lidong Chen <jemmy858585@gmail.com>
To: zhang.zhanghailiang@huawei.com, quintela@redhat.com, dgilbert@redhat.com
Cc: qemu-devel@nongnu.org, Lidong Chen <lidongchen@tencent.com>
Subject: [Qemu-devel] [PATCH v6 12/12] migration: create a dedicated thread to release rdma resource
Date: Fri, 3 Aug 2018 17:13:50 +0800 [thread overview]
Message-ID: <1533287630-4221-13-git-send-email-lidongchen@tencent.com> (raw)
In-Reply-To: <1533287630-4221-1-git-send-email-lidongchen@tencent.com>
ibv_dereg_mr wait for a long time for big memory size virtual server.
The test result is:
10GB 326ms
20GB 699ms
30GB 1021ms
40GB 1387ms
50GB 1712ms
60GB 2034ms
70GB 2457ms
80GB 2807ms
90GB 3107ms
100GB 3474ms
110GB 3735ms
120GB 4064ms
130GB 4567ms
140GB 4886ms
this will cause the guest os hang for a while when migration finished.
So create a dedicated thread to release rdma resource.
Signed-off-by: Lidong Chen <lidongchen@tencent.com>
---
migration/migration.c | 6 ++++++
migration/migration.h | 3 +++
migration/rdma.c | 47 +++++++++++++++++++++++++++++++----------------
3 files changed, 40 insertions(+), 16 deletions(-)
diff --git a/migration/migration.c b/migration/migration.c
index 360ee94..f58fe45 100644
--- a/migration/migration.c
+++ b/migration/migration.c
@@ -1499,6 +1499,7 @@ void migrate_init(MigrationState *s)
s->vm_was_running = false;
s->iteration_initial_bytes = 0;
s->threshold_size = 0;
+ s->rdma_cleanup_thread_quit = true;
}
static GSList *migration_blockers;
@@ -1660,6 +1661,10 @@ static bool migrate_prepare(MigrationState *s, bool blk, bool blk_inc,
return false;
}
+ if (s->rdma_cleanup_thread_quit != true) {
+ return false;
+ }
+
if (runstate_check(RUN_STATE_INMIGRATE)) {
error_setg(errp, "Guest is waiting for an incoming migration");
return false;
@@ -3214,6 +3219,7 @@ static void migration_instance_init(Object *obj)
ms->state = MIGRATION_STATUS_NONE;
ms->mbps = -1;
+ ms->rdma_cleanup_thread_quit = true;
qemu_sem_init(&ms->pause_sem, 0);
qemu_mutex_init(&ms->error_mutex);
diff --git a/migration/migration.h b/migration/migration.h
index a50c2de..4d1be08 100644
--- a/migration/migration.h
+++ b/migration/migration.h
@@ -231,6 +231,9 @@ struct MigrationState
* do not trigger spurious decompression errors.
*/
bool decompress_error_check;
+
+ /* Set this when rdma resource have released */
+ bool rdma_cleanup_thread_quit;
};
void migrate_set_state(int *state, int old_state, int new_state);
diff --git a/migration/rdma.c b/migration/rdma.c
index 14cdf82..3d1a4ad 100644
--- a/migration/rdma.c
+++ b/migration/rdma.c
@@ -2995,35 +2995,50 @@ static void qio_channel_rdma_set_aio_fd_handler(QIOChannel *ioc,
}
}
-static int qio_channel_rdma_close(QIOChannel *ioc,
- Error **errp)
+static void *qio_channel_rdma_close_thread(void *arg)
{
- QIOChannelRDMA *rioc = QIO_CHANNEL_RDMA(ioc);
- RDMAContext *rdmain, *rdmaout;
- trace_qemu_rdma_close();
-
- rdmain = rioc->rdmain;
- if (rdmain) {
- atomic_rcu_set(&rioc->rdmain, NULL);
- }
+ RDMAContext **rdma = arg;
+ RDMAContext *rdmain = rdma[0];
+ RDMAContext *rdmaout = rdma[1];
+ MigrationState *s = migrate_get_current();
- rdmaout = rioc->rdmaout;
- if (rdmaout) {
- atomic_rcu_set(&rioc->rdmaout, NULL);
- }
+ rcu_register_thread();
synchronize_rcu();
-
if (rdmain) {
qemu_rdma_cleanup(rdmain);
}
-
if (rdmaout) {
qemu_rdma_cleanup(rdmaout);
}
g_free(rdmain);
g_free(rdmaout);
+ g_free(rdma);
+
+ rcu_unregister_thread();
+ s->rdma_cleanup_thread_quit = true;
+ return NULL;
+}
+
+static int qio_channel_rdma_close(QIOChannel *ioc,
+ Error **errp)
+{
+ QemuThread t;
+ QIOChannelRDMA *rioc = QIO_CHANNEL_RDMA(ioc);
+ RDMAContext **rdma = g_new0(RDMAContext*, 2);
+ MigrationState *s = migrate_get_current();
+
+ trace_qemu_rdma_close();
+ if (rioc->rdmain || rioc->rdmaout) {
+ rdma[0] = rioc->rdmain;
+ rdma[1] = rioc->rdmaout;
+ atomic_rcu_set(&rioc->rdmain, NULL);
+ atomic_rcu_set(&rioc->rdmaout, NULL);
+ s->rdma_cleanup_thread_quit = false;
+ qemu_thread_create(&t, "rdma cleanup", qio_channel_rdma_close_thread,
+ rdma, QEMU_THREAD_DETACHED);
+ }
return 0;
}
--
1.8.3.1
next prev parent reply other threads:[~2018-08-03 9:14 UTC|newest]
Thread overview: 16+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-08-03 9:13 [Qemu-devel] [PATCH v6 00/12] Enable postcopy RDMA live migration Lidong Chen
2018-08-03 9:13 ` [Qemu-devel] [PATCH v6 01/12] migration: disable RDMA WRITE after postcopy started Lidong Chen
2018-08-03 9:13 ` [Qemu-devel] [PATCH v6 02/12] migration: create a dedicated connection for rdma return path Lidong Chen
2018-08-03 9:13 ` [Qemu-devel] [PATCH v6 03/12] migration: avoid concurrent invoke channel_close by different threads Lidong Chen
2018-08-06 9:07 ` 858585 jemmy
2018-08-03 9:13 ` [Qemu-devel] [PATCH v6 04/12] migration: implement bi-directional RDMA QIOChannel Lidong Chen
2018-08-03 9:13 ` [Qemu-devel] [PATCH v6 05/12] migration: Stop rdma yielding during incoming postcopy Lidong Chen
2018-08-03 9:13 ` [Qemu-devel] [PATCH v6 06/12] migration: implement io_set_aio_fd_handler function for RDMA QIOChannel Lidong Chen
2018-08-03 9:13 ` [Qemu-devel] [PATCH v6 07/12] migration: invoke qio_channel_yield only when qemu_in_coroutine() Lidong Chen
2018-08-03 9:13 ` [Qemu-devel] [PATCH v6 08/12] migration: poll the cm event while wait RDMA work request completion Lidong Chen
2018-08-03 9:13 ` [Qemu-devel] [PATCH v6 09/12] migration: implement the shutdown for RDMA QIOChannel Lidong Chen
2018-08-03 9:13 ` [Qemu-devel] [PATCH v6 10/12] migration: poll the cm event for destination qemu Lidong Chen
2018-08-03 9:13 ` [Qemu-devel] [PATCH v6 11/12] migration: remove the unnecessary RDMA_CONTROL_ERROR message Lidong Chen
2018-08-03 9:13 ` Lidong Chen [this message]
2018-08-03 9:42 ` [Qemu-devel] [PATCH v6 00/12] Enable postcopy RDMA live migration no-reply
2018-08-06 6:09 ` 858585 jemmy
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=1533287630-4221-13-git-send-email-lidongchen@tencent.com \
--to=jemmy858585@gmail.com \
--cc=dgilbert@redhat.com \
--cc=lidongchen@tencent.com \
--cc=qemu-devel@nongnu.org \
--cc=quintela@redhat.com \
--cc=zhang.zhanghailiang@huawei.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).