From: Andrew Cooper <andrew.cooper3@citrix.com>
To: Yang Hongyang <yanghy@cn.fujitsu.com>, xen-devel@lists.xen.org
Cc: wei.liu2@citrix.com, ian.campbell@citrix.com,
wency@cn.fujitsu.com, ian.jackson@eu.citrix.com,
yunhong.jiang@intel.com, eddie.dong@intel.com,
guijianfeng@cn.fujitsu.com, rshriram@cs.ubc.ca
Subject: Re: [PATCH v4 --for 4.6 COLOPre 20/25] tools/libx{l, c}: add back channel to libxc
Date: Wed, 15 Jul 2015 14:21:51 +0100 [thread overview]
Message-ID: <55A65E6F.4090400@citrix.com> (raw)
In-Reply-To: <1436946351-21118-21-git-send-email-yanghy@cn.fujitsu.com>
On 15/07/15 08:45, Yang Hongyang wrote:
> In COLO mode, both VMs are running, and are considered in sync if the
> visible network traffic is identical. After some time, they fall out of
> sync.
>
> At this point, the two VMs have definitely diverged. Lets call the
> primary dirty bitmap set A, while the secondary dirty bitmap set B.
>
> Sets A and B are different.
>
> Under normal migration, the page data for set A will be sent form the
> primary to the secondary.
>
> However, the set difference B - A (lets call this C) is out-of-date on
> the secondary (with respect to the primary) and will not be sent by the
> primary, as it was not memory dirtied by the primary. The secondary
> needs the page data for C to reconstruct an exact copy of the primary at
> the checkpoint.
>
> The secondary cannot calculate C as it doesn't know A. Instead, the
> secondary must send B to the primary, at which point the primary
> calculates the union of A and B (lets call this D) which is all the
> pages dirtied by both the primary and the secondary, and sends all page
> data covered by D.
>
> In the general case, D is a superset of both A and B. Without the
> backchannel dirty bitmap, a COLO checkpoint can't reconstruct a valid
> copy of the primary.
>
> We transfer the dirty bitmap on libxc side, so we need to introduce back
> channel to libxc.
>
> Signed-off-by: Yang Hongyang <yanghy@cn.fujitsu.com>
> commit message:
> Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
> CC: Ian Campbell <Ian.Campbell@citrix.com>
> CC: Ian Jackson <Ian.Jackson@eu.citrix.com>
> CC: Wei Liu <wei.liu2@citrix.com>
> ---
> tools/libxc/include/xenguest.h | 8 ++++----
> tools/libxc/xc_domain_restore.c | 4 ++--
> tools/libxc/xc_domain_save.c | 4 ++--
> tools/libxc/xc_sr_restore.c | 2 +-
> tools/libxc/xc_sr_save.c | 2 +-
> tools/libxl/libxl_save_callout.c | 39 ++++++++++++++++++++++++++-------------
> tools/libxl/libxl_save_helper.c | 8 ++++++--
> 7 files changed, 42 insertions(+), 25 deletions(-)
You have not patched xc_nomigrate.c, which means this will break the ARM
build. (I fell into the same trap, requiring c/s f50fe3a5 as a fixup).
Having said that, I plan to throw together some cleanup patches removing
files like xc_domain_{save,restore}.c and dropping most of the
parameters from the parameter list, as they are superfluous.
I will try to get my cleanup done shortly, which should make this prereq
series easier, although I am focusing on some hypervisor side fixes
right at the moment.
~Andrew
next prev parent reply other threads:[~2015-07-15 13:21 UTC|newest]
Thread overview: 101+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-07-15 7:45 [PATCH v4 --for 4.6 COLOPre 00/25] Prerequisite patches for COLO Yang Hongyang
2015-07-15 7:45 ` [PATCH v4 --for 4.6 COLOPre 01/25] tools/libxl: rename libxl__domain_suspend to libxl__domain_save Yang Hongyang
2015-07-15 11:16 ` Ian Campbell
2015-07-15 7:45 ` [PATCH v4 --for 4.6 COLOPre 02/25] tools/libxl: move domain suspend code into libxl_dom_suspend.c Yang Hongyang
2015-07-15 7:45 ` [PATCH v4 --for 4.6 COLOPre 03/25] tools/libxl: move domain resume " Yang Hongyang
2015-07-15 7:45 ` [PATCH v4 --for 4.6 COLOPre 04/25] tools/libxl: rename remus checkpoint callbacks Yang Hongyang
2015-07-15 11:17 ` Ian Campbell
2015-07-16 1:43 ` Yang Hongyang
2015-07-15 7:45 ` [PATCH v4 --for 4.6 COLOPre 05/25] libxl/remus: introduce libxl__remus_setup Yang Hongyang
2015-07-15 11:26 ` Ian Campbell
2015-07-16 5:32 ` Yang Hongyang
2015-07-16 10:40 ` Ian Campbell
2015-07-15 7:45 ` [PATCH v4 --for 4.6 COLOPre 06/25] libxl/remus: introduce libxl__remus_teardown Yang Hongyang
2015-07-15 11:59 ` Ian Campbell
2015-07-16 1:43 ` Yang Hongyang
2015-07-15 7:45 ` [PATCH v4 --for 4.6 COLOPre 07/25] libxl/remus: init checkpoint_callback in Remus checkpoint callback Yang Hongyang
2015-07-15 12:02 ` Ian Campbell
2015-07-15 12:35 ` Yang Hongyang
2015-07-16 10:32 ` Ian Campbell
2015-07-16 11:00 ` Yang Hongyang
2015-07-16 11:16 ` Ian Campbell
2015-07-15 7:45 ` [PATCH v4 --for 4.6 COLOPre 08/25] tools/libxl: move remus code into libxl_remus.c Yang Hongyang
2015-07-15 12:05 ` Ian Campbell
2015-07-15 7:45 ` [PATCH v4 --for 4.6 COLOPre 09/25] tools/libxl: move save/restore code into libxl_dom_save.c Yang Hongyang
2015-07-15 7:45 ` [PATCH v4 --for 4.6 COLOPre 10/25] libxl/save: Refactor libxl__domain_suspend_state Yang Hongyang
2015-07-15 12:10 ` Ian Campbell
2015-07-15 7:45 ` [PATCH v4 --for 4.6 COLOPre 11/25] tools/libxc: support to resume uncooperative HVM guests Yang Hongyang
2015-07-15 12:26 ` Ian Campbell
2015-07-16 5:57 ` Yang Hongyang
2015-07-16 15:40 ` Ian Jackson
2015-07-16 16:15 ` Yang Hongyang
2015-07-16 16:27 ` Ian Jackson
2015-12-15 2:05 ` Wen Congyang
2016-01-04 16:33 ` Ian Jackson
2015-07-15 7:45 ` [PATCH v4 --for 4.6 COLOPre 12/25] tools/libxl: introduce enum type libxl_checkpointed_stream Yang Hongyang
2015-07-15 12:34 ` Ian Campbell
2015-07-15 13:58 ` Yang Hongyang
2015-07-16 10:34 ` Ian Campbell
2015-07-16 10:47 ` Yang Hongyang
2015-07-15 7:45 ` [PATCH v4 --for 4.6 COLOPre 13/25] migration/save: pass checkpointed_stream from libxl to libxc Yang Hongyang
2015-07-15 12:38 ` Ian Campbell
2015-07-16 6:05 ` Yang Hongyang
2015-07-16 10:47 ` Ian Campbell
2015-07-16 16:13 ` Wei Liu
2015-07-16 16:21 ` Yang Hongyang
2015-07-16 16:39 ` Wei Liu
2015-07-16 16:10 ` Wei Liu
2015-07-16 16:24 ` Yang Hongyang
2015-07-16 16:37 ` Wei Liu
2015-07-15 7:45 ` [PATCH v4 --for 4.6 COLOPre 14/25] tools/libxl: introduce libxl__domain_restore_device_model to load qemu state Yang Hongyang
2015-07-15 12:45 ` Ian Campbell
2015-07-15 13:42 ` Yang Hongyang
2015-07-15 7:45 ` [PATCH v4 --for 4.6 COLOPre 15/25] tools/libxl: check QEMU state before resume dm Yang Hongyang
2015-07-15 12:48 ` Ian Campbell
2015-07-15 12:54 ` Ian Campbell
2015-07-15 13:00 ` Wei Liu
2015-07-15 13:48 ` Ian Campbell
2015-07-15 13:49 ` Ian Campbell
2015-07-16 14:43 ` Wei Liu
2015-07-16 15:43 ` Yang Hongyang
2015-07-15 7:45 ` [PATCH v4 --for 4.6 COLOPre 16/25] tools/libxl: Update libxl_domain_unpause() to support qemu-xen Yang Hongyang
2015-07-15 12:50 ` Ian Campbell
2015-07-16 3:49 ` Yang Hongyang
2015-07-16 10:39 ` Ian Campbell
2015-07-16 10:51 ` Yang Hongyang
2015-07-16 16:26 ` Wei Liu
2015-07-15 7:45 ` [PATCH v4 --for 4.6 COLOPre 17/25] tools/libxl: introduce libxl__domain_common_switch_qemu_logdirty() Yang Hongyang
2015-07-15 7:45 ` [PATCH v4 --for 4.6 COLOPre 18/25] tools/libxl: export logdirty_init Yang Hongyang
2015-07-15 7:45 ` [PATCH v4 --for 4.6 COLOPre 19/25] tools/libxl: Add back channel to allow migration target send data back Yang Hongyang
2015-07-15 7:45 ` [PATCH v4 --for 4.6 COLOPre 20/25] tools/libx{l, c}: add back channel to libxc Yang Hongyang
2015-07-15 13:13 ` Ian Campbell
2015-07-16 6:29 ` Yang Hongyang
2015-07-16 11:01 ` Ian Campbell
2015-07-15 13:21 ` Andrew Cooper [this message]
2015-07-16 6:07 ` Yang Hongyang
2015-07-15 7:45 ` [PATCH v4 --for 4.6 COLOPre 21/25] tools/libxl: rename remus device to checkpoint device Yang Hongyang
2015-07-15 13:15 ` Ian Campbell
2015-07-15 13:34 ` Yang Hongyang
2015-07-16 9:26 ` Andrew Cooper
2015-07-16 9:29 ` Yang Hongyang
2015-07-15 13:32 ` Ian Campbell
2015-07-15 13:38 ` Yang Hongyang
2015-07-16 9:23 ` Yang Hongyang
2015-07-16 9:31 ` Ian Campbell
2015-07-16 9:36 ` Yang Hongyang
2015-07-16 10:14 ` Ian Campbell
2015-07-16 10:22 ` Yang Hongyang
2015-07-15 7:45 ` [PATCH v4 --for 4.6 COLOPre 22/25] tools/libxl: adjust the indentation Yang Hongyang
2015-07-15 7:45 ` [PATCH v4 --for 4.6 COLOPre 23/25] tools/libxl: store remus_ops in checkpoint device state Yang Hongyang
2015-07-15 13:21 ` Ian Campbell
2015-07-15 7:45 ` [PATCH v4 --for 4.6 COLOPre 24/25] tools/libxl: move remus state into a seperate structure Yang Hongyang
2015-07-15 13:28 ` Ian Campbell
2015-07-15 13:50 ` Yang Hongyang
2015-07-16 10:37 ` Ian Campbell
2015-07-16 11:10 ` Ian Jackson
2015-07-16 11:19 ` Ian Campbell
2015-07-15 15:08 ` Ian Jackson
2015-07-15 15:18 ` Yang Hongyang
2015-07-15 7:45 ` [PATCH v4 --for 4.6 COLOPre 25/25] tools/libxl: seperate device init/cleanup from checkpoint device layer Yang Hongyang
2015-07-15 13:37 ` Ian Campbell
2015-07-16 1:37 ` [PATCH v4 --for 4.6 COLOPre 00/25] Prerequisite patches for COLO Yang Hongyang
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=55A65E6F.4090400@citrix.com \
--to=andrew.cooper3@citrix.com \
--cc=eddie.dong@intel.com \
--cc=guijianfeng@cn.fujitsu.com \
--cc=ian.campbell@citrix.com \
--cc=ian.jackson@eu.citrix.com \
--cc=rshriram@cs.ubc.ca \
--cc=wei.liu2@citrix.com \
--cc=wency@cn.fujitsu.com \
--cc=xen-devel@lists.xen.org \
--cc=yanghy@cn.fujitsu.com \
--cc=yunhong.jiang@intel.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).