From: mrhines@linux.vnet.ibm.com
To: qemu-devel@nongnu.org
Cc: aliguori@us.ibm.com, quintela@redhat.com, knoel@redhat.com,
owasserm@redhat.com, abali@us.ibm.com, mrhines@us.ibm.com,
gokul@us.ibm.com, pbonzini@redhat.com, chegu_vinod@hp.com
Subject: [Qemu-devel] [PATCH v12 10/15] rdma: introduce capability x-rdma-pin-all
Date: Tue, 25 Jun 2013 21:35:36 -0400 [thread overview]
Message-ID: <1372210541-28092-11-git-send-email-mrhines@linux.vnet.ibm.com> (raw)
In-Reply-To: <1372210541-28092-1-git-send-email-mrhines@linux.vnet.ibm.com>
From: "Michael R. Hines" <mrhines@us.ibm.com>
This capability allows you to disable dynamic chunk registration
for better throughput on high-performance links.
For example, using an 8GB RAM virtual machine with all 8GB of memory in
active use and the VM itself is completely idle using a 40 gbps infiniband link:
1. x-rdma-pin-all disabled total time: approximately 7.5 seconds @ 9.5 Gbps
2. x-rdma-pin-all enabled total time: approximately 4 seconds @ 26 Gbps
These numbers would of course scale up to whatever size virtual machine
you have to migrate using RDMA.
Enabling this feature does *not* have any measurable affect on
migration *downtime*. This is because, without this feature, all of the
memory will have already been registered already in advance during
the bulk round and does not need to be re-registered during the successive
iteration rounds.
Reviewed-by: Juan Quintela <quintela@redhat.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Chegu Vinod <chegu_vinod@hp.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Tested-by: Chegu Vinod <chegu_vinod@hp.com>
Tested-by: Michael R. Hines <mrhines@us.ibm.com>
Signed-off-by: Michael R. Hines <mrhines@us.ibm.com>
---
include/migration/migration.h | 2 ++
migration.c | 9 +++++++++
qapi-schema.json | 7 ++++++-
3 files changed, 17 insertions(+), 1 deletion(-)
diff --git a/include/migration/migration.h b/include/migration/migration.h
index 7b153d7..9d3cc85 100644
--- a/include/migration/migration.h
+++ b/include/migration/migration.h
@@ -125,6 +125,8 @@ void migrate_add_blocker(Error *reason);
*/
void migrate_del_blocker(Error *reason);
+bool migrate_rdma_pin_all(void);
+
int xbzrle_encode_buffer(uint8_t *old_buf, uint8_t *new_buf, int slen,
uint8_t *dst, int dlen);
int xbzrle_decode_buffer(uint8_t *src, int slen, uint8_t *dst, int dlen);
diff --git a/migration.c b/migration.c
index 441a3b2..a704d48 100644
--- a/migration.c
+++ b/migration.c
@@ -476,6 +476,15 @@ void qmp_migrate_set_downtime(double value, Error **errp)
max_downtime = (uint64_t)value;
}
+bool migrate_rdma_pin_all(void)
+{
+ MigrationState *s;
+
+ s = migrate_get_current();
+
+ return s->enabled_capabilities[MIGRATION_CAPABILITY_X_RDMA_PIN_ALL];
+}
+
int migrate_use_xbzrle(void)
{
MigrationState *s;
diff --git a/qapi-schema.json b/qapi-schema.json
index 78da557..a30a728 100644
--- a/qapi-schema.json
+++ b/qapi-schema.json
@@ -608,10 +608,15 @@
# This feature allows us to minimize migration traffic for certain work
# loads, by sending compressed difference of the pages
#
+# @x-rdma-pin-all: Controls whether or not the entire VM memory footprint is
+# mlock()'d on demand or all at once. Refer to docs/rdma.txt for usage.
+# Disabled by default. Experimental: may (or may not) be renamed after
+# further testing is complete. (since 1.6)
+#
# Since: 1.2
##
{ 'enum': 'MigrationCapability',
- 'data': ['xbzrle'] }
+ 'data': ['xbzrle', 'x-rdma-pin-all'] }
##
# @MigrationCapabilityStatus
--
1.7.10.4
next prev parent reply other threads:[~2013-06-26 1:36 UTC|newest]
Thread overview: 28+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-06-26 1:35 [Qemu-devel] [PATCH v12 00/15] rdma: migration support mrhines
2013-06-26 1:35 ` [Qemu-devel] [PATCH v12 01/15] rdma: add documentation mrhines
2013-06-26 1:48 ` [Qemu-devel] [PATCH v12 00/15] rdma: migration support Michael R. Hines
2013-06-27 22:41 ` [Qemu-devel] [PATCH v12 01/15] rdma: add documentation Eric Blake
2013-06-27 22:48 ` Michael R. Hines
2013-06-26 1:35 ` [Qemu-devel] [PATCH v12 02/15] rdma: introduce qemu_update_position() mrhines
2013-06-26 1:35 ` [Qemu-devel] [PATCH v12 03/15] rdma: export yield_until_fd_readable() mrhines
2013-06-26 1:35 ` [Qemu-devel] [PATCH v12 04/15] rdma: export throughput w/ MigrationStats QMP mrhines
2013-06-27 23:02 ` Eric Blake
2013-06-28 13:13 ` Michael R. Hines
2013-06-26 1:35 ` [Qemu-devel] [PATCH v12 05/15] rdma: introduce qemu_file_mode_is_not_valid() mrhines
2013-06-26 1:35 ` [Qemu-devel] [PATCH v12 06/15] rdma: export qemu_fflush() mrhines
2013-06-26 1:35 ` [Qemu-devel] [PATCH v12 08/15] rdma: introduce qemu_ram_foreach_block() mrhines
2013-06-27 19:24 ` Peter Maydell
2013-06-27 19:56 ` Michael R. Hines
2013-06-26 1:35 ` [Qemu-devel] [PATCH v12 09/15] rdma: new QEMUFileOps hooks mrhines
2013-06-26 1:35 ` mrhines [this message]
2013-06-26 1:35 ` [Qemu-devel] [PATCH v12 11/15] rdma: core logic mrhines
2013-06-26 1:35 ` [Qemu-devel] [PATCH v12 12/15] rdma: send pc.ram mrhines
2013-06-26 1:35 ` [Qemu-devel] [PATCH v12 13/15] rdma: allow state transitions between other states besides ACTIVE mrhines
2013-06-26 1:35 ` [Qemu-devel] [PATCH v12 14/15] rdma: introduce MIG_STATE_NONE and change MIG_STATE_SETUP state transition mrhines
2013-06-26 1:35 ` [Qemu-devel] [PATCH v12 15/15] rdma: account for the time spent in MIG_STATE_SETUP through QMP mrhines
2013-06-27 22:58 ` Eric Blake
2013-06-28 7:15 ` Paolo Bonzini
2013-06-28 13:15 ` Michael R. Hines
2013-06-28 13:32 ` Eric Blake
2013-06-28 14:45 ` Michael R. Hines
2013-06-28 13:14 ` Michael R. Hines
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=1372210541-28092-11-git-send-email-mrhines@linux.vnet.ibm.com \
--to=mrhines@linux.vnet.ibm.com \
--cc=abali@us.ibm.com \
--cc=aliguori@us.ibm.com \
--cc=chegu_vinod@hp.com \
--cc=gokul@us.ibm.com \
--cc=knoel@redhat.com \
--cc=mrhines@us.ibm.com \
--cc=owasserm@redhat.com \
--cc=pbonzini@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).