Re: [Xen-devel] [PATCH v4 --for 4.6 COLOPre 20/25] tools/libx{l, c}: add back channel to libxc

[Yang Hongyang <yanghy@xxxxxxxxxxxxxx>]

On 07/15/2015 09:21 PM, Andrew Cooper wrote:
> 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 However, the set difference B - A (lets call this C) is out-of-date on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@xxxxxxxxxxxxxx>
> > commit message:
> > Signed-off-by: Andrew Cooper <andrew.cooper3@xxxxxxxxxx>
> > CC: Ian Campbell <Ian.Campbell@xxxxxxxxxx>
> > CC: Ian Jackson <Ian.Jackson@xxxxxxxxxxxxx>
> > CC: Wei Liu <wei.liu2@xxxxxxxxxx>
> > ---
> >   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).
Thank you for pointing this out, I will look at that commit.

> 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,
That would be great, thanks!

although I am focusing on some hypervisor side fixes
> right at the moment.
>
> ~Andrew
> .
--
Thanks,
Yang.

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel