From: "Dr. David Alan Gilbert" <dgilbert@redhat.com>
To: Bechir Bani <bechir.bani@gmail.com>
Cc: qemu-devel@nongnu.org
Subject: Re: [Qemu-devel] Qemu live migration code
Date: Tue, 13 May 2014 09:05:29 +0100 [thread overview]
Message-ID: <20140513080528.GB2566@work-vm> (raw)
In-Reply-To: <CACpvL09NSXCqOf1E-TiT1UWpZp1ZOcaGJumMGaQG6Mia6XfXrw@mail.gmail.com>
* Bechir Bani (bechir.bani@gmail.com) wrote:
> Hi ,
>
> There is someone who can explain to me the role of two trace points in the
> file* Migration.c *
>
> The trace points are :
>
> * migrate_pending* : what are the attributes *pending size *and *max* ?
'pending_size' is an estimate of the amount of data left to be transferred
in the iterative part.
'max_size' is an estimate of the amount of data that can be transferred in the
'max_downtime' period
Thus if pending_size < max_size it can stop iterating and transfer the last
part.
> * migrate_transferred*: what are the attributes *transferred* ,*
> time_spent* , *bandwidth* and *max_size* ?
Every so often it recalculates bandwidths etc:
transferred_bytes - actual bytes transferred during migration since the last
time it did that set of calculations
time_spent - time since it last did that set of calculations
bandwidth - transferred_bytes / time_spent
max_size = as in the previous trace point.
Dave
>
>
> Thank you !
>
>
> 2014-03-25 6:03 GMT-04:00 Sanidhya Kashyap <sanidhya.iiith@gmail.com>:
>
> >
> >
> >
> > On Tue, Mar 25, 2014 at 2:10 AM, Bechir Bani <bechir.bani@gmail.com>wrote:
> >
> >> Hi Sanidhya,
> >>
> >>
> >> Which function in savevm.c can tell me about the stop time ?
> >>
> >>
> > the migration thread function in migration.c file which has function name
> > - qemu_savevm_state_complete will get executed in the stop and copy phase.
> >
> >
> >>
> >> 2014-03-24 13:46 GMT-04:00 Sanidhya Kashyap <sanidhya.iiith@gmail.com>:
> >>
> >> savevm.c will tell you about the stop time.
> >>>
> >>> arch_init.c (ram_save_block) will tell about the number of pages
> >>> transferred.
> >>>
> >>>
> >>> On Mon, Mar 24, 2014 at 10:51 PM, Bechir Bani <bechir.bani@gmail.com>wrote:
> >>>
> >>>> I have a task to add trace points in the source code of Qemu. The goal
> >>>> is to know the number of pages transferred at each iteration and stop time
> >>>> of the machine as well.
> >>>>
> >>>>
> >>>> 2014-03-24 12:50 GMT-04:00 Dr. David Alan Gilbert <dgilbert@redhat.com>
> >>>> :
> >>>>
> >>>> * Bechir Bani (bechir.bani@gmail.com) wrote:
> >>>>> > Hi,
> >>>>> >
> >>>>> > I want to know the source code of qemu which is responsible for the
> >>>>> > migration of virtual machines, more precisely where the part of the
> >>>>> code
> >>>>> > that describes the stages of memory transfer. is that you can help
> >>>>> me?
> >>>>>
> >>>>> It's split around a few files; memory is mostly in arch_init.c;
> >>>>> It's something like:
> >>>>>
> >>>>> migration.c Overall management
> >>>>> savevm.c
> >>>>> qemu-file.c File buffering/bytes on the wire
> >>>>> vmstate.c Structured saving of individual devices
> >>>>> arch_init.c RAM special code, and a few other things
> >>>>>
> >>>>> What are you trying to do/change?
> >>>>>
> >>>>> Dave
> >>>>>
> >>>>> --
> >>>>> Dr. David Alan Gilbert / dgilbert@redhat.com / Manchester, UK
> >>>>>
> >>>>
> >>>>
> >>>>
> >>>> --
> >>>> *Béchir Bani *
> >>>> ****************************************************************
> >>>> *Ecole Polytechnique de Montréal *
> >>>> ****************************************************************
> >>>> *Laboratoire DORSAL*
> >>>> *****************************************************************
> >>>> *Montréal - Canada*
> >>>>
> >>>
> >>>
> >>
> >>
> >> --
> >> *Béchir Bani *
> >> ****************************************************************
> >> *Ecole Polytechnique de Montréal *
> >> ****************************************************************
> >> *Laboratoire DORSAL*
> >> *****************************************************************
> >> *Montréal - Canada*
> >>
> >
> >
>
>
> --
> *Béchir Bani *
> ****************************************************************
> *Ecole Polytechnique de Montréal *
> ****************************************************************
> *Laboratoire DORSAL*
> *****************************************************************
> *Montréal - Canada*
--
Dr. David Alan Gilbert / dgilbert@redhat.com / Manchester, UK
prev parent reply other threads:[~2014-05-13 8:05 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-03-24 15:29 [Qemu-devel] Qemu live migration code Bechir Bani
2014-03-24 15:33 ` 陈梁
2014-03-24 16:50 ` Dr. David Alan Gilbert
2014-03-24 17:21 ` Bechir Bani
2014-03-24 17:46 ` Sanidhya Kashyap
2014-03-24 20:40 ` Bechir Bani
2014-03-25 10:03 ` Sanidhya Kashyap
2014-05-12 16:03 ` Bechir Bani
2014-05-13 8:05 ` Dr. David Alan Gilbert [this message]
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=20140513080528.GB2566@work-vm \
--to=dgilbert@redhat.com \
--cc=bechir.bani@gmail.com \
--cc=qemu-devel@nongnu.org \
/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).