From: marcandre.lureau@redhat.com
To: qemu-devel@nongnu.org
Cc: haifeng.lin@huawei.com, mst@redhat.com, thibaut.collet@6wind.com,
jasowang@redhat.com, pbonzini@redhat.com,
"Marc-André Lureau" <marcandre.lureau@redhat.com>
Subject: [Qemu-devel] [PATCH v8 15/27] vhost-user: document migration log
Date: Fri, 9 Oct 2015 17:17:29 +0200 [thread overview]
Message-ID: <1444403861-3240-16-git-send-email-marcandre.lureau@redhat.com> (raw)
In-Reply-To: <1444403861-3240-1-git-send-email-marcandre.lureau@redhat.com>
From: Marc-André Lureau <marcandre.lureau@redhat.com>
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
---
docs/specs/vhost-user.txt | 48 +++++++++++++++++++++++++++++++++++++++++++++--
1 file changed, 46 insertions(+), 2 deletions(-)
diff --git a/docs/specs/vhost-user.txt b/docs/specs/vhost-user.txt
index 4eadad1..e0292a0 100644
--- a/docs/specs/vhost-user.txt
+++ b/docs/specs/vhost-user.txt
@@ -115,11 +115,13 @@ the ones that do:
* VHOST_GET_FEATURES
* VHOST_GET_PROTOCOL_FEATURES
* VHOST_GET_VRING_BASE
+ * VHOST_SET_LOG_BASE (if VHOST_USER_PROTOCOL_F_LOG_SHMFD)
There are several messages that the master sends with file descriptors passed
in the ancillary data:
* VHOST_SET_MEM_TABLE
+ * VHOST_SET_LOG_BASE (if VHOST_USER_PROTOCOL_F_LOG_SHMFD)
* VHOST_SET_LOG_FD
* VHOST_SET_VRING_KICK
* VHOST_SET_VRING_CALL
@@ -140,8 +142,7 @@ Multiple queue support
Multiple queue is treated as a protocol extension, hence the slave has to
implement protocol features first. The multiple queues feature is supported
-only when the protocol feature VHOST_USER_PROTOCOL_F_MQ (bit 0) is set:
-#define VHOST_USER_PROTOCOL_F_MQ 0
+only when the protocol feature VHOST_USER_PROTOCOL_F_MQ (bit 0) is set.
The max number of queues the slave supports can be queried with message
VHOST_USER_GET_PROTOCOL_FEATURES. Master should stop when the number of
@@ -152,6 +153,48 @@ queue in the sent message to identify a specified queue. One queue pair
is enabled initially. More queues are enabled dynamically, by sending
message VHOST_USER_SET_VRING_ENABLE.
+Migration
+---------
+
+During live migration, the master may need to track the modifications
+the slave makes to the memory mapped regions. The client should mark
+the dirty pages in a log. Once it complies to this logging, it may
+declare the VHOST_F_LOG_ALL vhost feature.
+
+All the modifications to memory pointed by vring "descriptor" should
+be marked. Modifications to "used" vring should be marked if
+VHOST_VRING_F_LOG is part of ring's features.
+
+Dirty pages are of size:
+#define VHOST_LOG_PAGE 0x1000
+
+The log memory fd is provided in the ancillary data of
+VHOST_USER_SET_LOG_BASE message when the slave has
+VHOST_USER_PROTOCOL_F_LOG_SHMFD protocol feature.
+
+The size of the log may be computed by using all the known guest
+addresses. The log covers from address 0 to the maximum of guest
+regions. In pseudo-code, to mark page at "addr" as dirty:
+
+page = addr / VHOST_LOG_PAGE
+log[page / 8] |= 1 << page % 8
+
+Use atomic operations, as the log may be concurrently manipulated.
+
+VHOST_USER_SET_LOG_FD is an optional message with an eventfd in
+ancillary data, it may be used to inform the master that the log has
+been modified.
+
+Once the source has finished migration, VHOST_USER_RESET_OWNER message
+will be sent by the source. No further update must be done before the
+destination takes over with new regions & rings.
+
+Protocol features
+-----------------
+
+#define VHOST_USER_PROTOCOL_F_MQ 0
+#define VHOST_USER_PROTOCOL_F_LOG_SHMFD 1
+
Message types
-------------
@@ -236,6 +279,7 @@ Message types
Id: 6
Equivalent ioctl: VHOST_SET_LOG_BASE
Master payload: u64
+ Slave payload: N/A
Sets the logging base address.
--
2.4.3
next prev parent reply other threads:[~2015-10-09 15:18 UTC|newest]
Thread overview: 39+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-10-09 15:17 [Qemu-devel] [PATCH v8 00/27] vhost-user: add migration support marcandre.lureau
2015-10-09 15:17 ` [Qemu-devel] [PATCH v8 01/27] exec: factor out duplicate mmap code marcandre.lureau
2015-10-09 15:17 ` [Qemu-devel] [PATCH v8 02/27] configure: probe for memfd marcandre.lureau
2015-10-09 15:17 ` [Qemu-devel] [PATCH v8 03/27] linux-headers: add unistd.h marcandre.lureau
2015-10-09 15:17 ` [Qemu-devel] [PATCH v8 04/27] build-sys: split util-obj- on multi-lines marcandre.lureau
2015-10-09 15:17 ` [Qemu-devel] [PATCH v8 05/27] util: add linux-only memfd fallback marcandre.lureau
2015-10-09 15:17 ` [Qemu-devel] [PATCH v8 06/27] util: add memfd helpers marcandre.lureau
2015-10-09 15:17 ` [Qemu-devel] [PATCH v8 07/27] util: add fallback for qemu_memfd_alloc() marcandre.lureau
2015-10-09 15:17 ` [Qemu-devel] [PATCH v8 08/27] vhost: document log resizing marcandre.lureau
2015-10-09 15:17 ` [Qemu-devel] [PATCH v8 09/27] vhost: add vhost_set_log_base op marcandre.lureau
2015-10-09 15:17 ` [Qemu-devel] [PATCH v8 10/27] vhost-user: add vhost_user_requires_shm_log() marcandre.lureau
2015-10-09 15:17 ` [Qemu-devel] [PATCH v8 11/27] vhost: alloc shareable log marcandre.lureau
2015-10-09 15:17 ` [Qemu-devel] [PATCH v8 12/27] vhost-user: send log shm fd along with log_base marcandre.lureau
2015-10-09 15:17 ` [Qemu-devel] [PATCH v8 13/27] vhost-user: add a migration blocker marcandre.lureau
2015-10-09 15:17 ` [Qemu-devel] [PATCH v8 14/27] vhost: use a function for each call marcandre.lureau
2015-10-09 15:17 ` marcandre.lureau [this message]
2015-10-09 15:17 ` [Qemu-devel] [PATCH v8 16/27] net: add trace_vhost_user_event marcandre.lureau
2015-10-09 15:17 ` [Qemu-devel] [PATCH v8 17/27] vhost user: add support of live migration marcandre.lureau
2015-10-09 15:17 ` [Qemu-devel] [PATCH v8 18/27] vhost user: add rarp sending after live migration for legacy guest marcandre.lureau
2015-10-09 15:17 ` [Qemu-devel] [PATCH v8 19/27] vhost-user: use an enum helper for features mask marcandre.lureau
2015-10-09 15:17 ` [Qemu-devel] [PATCH v8 20/27] vhost: add migration block if memfd failed marcandre.lureau
2015-10-09 15:17 ` [Qemu-devel] [PATCH v8 21/27] vhost-user-test: move wait_for_fds() out marcandre.lureau
2015-10-09 15:17 ` [Qemu-devel] [PATCH v8 22/27] vhost-user-test: remove useless static check marcandre.lureau
2015-10-09 15:17 ` [Qemu-devel] [PATCH v8 23/27] vhost-user-test: wrap server in TestServer struct marcandre.lureau
2015-10-09 15:17 ` [Qemu-devel] [PATCH v8 24/27] vhost-user-test: learn to tweak various qemu arguments marcandre.lureau
2015-10-09 15:17 ` [Qemu-devel] [PATCH v8 25/27] vhost-user-test: add live-migration test marcandre.lureau
2015-10-09 15:17 ` [Qemu-devel] [PATCH v8 26/27] vhost-user-test: check ownership during migration marcandre.lureau
2015-10-09 15:17 ` [Qemu-devel] [PATCH v8 27/27] seccomp: add memfd_create to whitelist marcandre.lureau
2015-10-09 15:21 ` Eduardo Otubo
2015-10-12 15:56 ` [Qemu-devel] [PATCH v8 00/27] vhost-user: add migration support Thibaut Collet
2015-10-13 12:19 ` Thibaut Collet
2015-10-18 8:21 ` Michael S. Tsirkin
2015-10-19 13:22 ` Thibaut Collet
2015-10-19 15:39 ` Michael S. Tsirkin
2015-10-19 16:42 ` Thibaut Collet
2015-10-19 21:12 ` Michael S. Tsirkin
2015-10-20 6:30 ` Thibaut Collet
2015-10-20 10:21 ` Michael S. Tsirkin
2015-10-20 11:47 ` Thibaut Collet
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=1444403861-3240-16-git-send-email-marcandre.lureau@redhat.com \
--to=marcandre.lureau@redhat.com \
--cc=haifeng.lin@huawei.com \
--cc=jasowang@redhat.com \
--cc=mst@redhat.com \
--cc=pbonzini@redhat.com \
--cc=qemu-devel@nongnu.org \
--cc=thibaut.collet@6wind.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).