From: Marcelo Tosatti <mtosatti@redhat.com>
To: Alexander Graf <agraf@suse.de>
Cc: qemu-devel@nongnu.org, kvm@vger.kernel.org
Subject: Re: [PATCH] kvmclock: Ensure time in migration never goes backward
Date: Wed, 7 May 2014 22:33:25 -0300 [thread overview]
Message-ID: <20140508013325.GA5474@amt.cnet> (raw)
In-Reply-To: <1399297882-3444-1-git-send-email-agraf@suse.de>
On Mon, May 05, 2014 at 03:51:22PM +0200, Alexander Graf wrote:
> When we migrate we ask the kernel about its current belief on what the guest
> time would be. However, I've seen cases where the kvmclock guest structure
> indicates a time more recent than the kvm returned time.
>
> To make sure we never go backwards, calculate what the guest would have seen
> as time at the point of migration and use that value instead of the kernel
> returned one when it's more recent.
>
> While this doesn't fix the underlying issue that the kernel's view of time
> is skewed, it allows us to safely migrate guests even from sources that are
> known broken.
>
> Signed-off-by: Alexander Graf <agraf@suse.de>
OK Alexander better move this logic to the kernel, in KVM_GET_CLOCK.
Otherwise every user of KVM_GET_CLOCK would have to apply the
workaround.
WARNING: multiple messages have this Message-ID (diff)
From: Marcelo Tosatti <mtosatti@redhat.com>
To: Alexander Graf <agraf@suse.de>
Cc: qemu-devel@nongnu.org, kvm@vger.kernel.org
Subject: Re: [Qemu-devel] [PATCH] kvmclock: Ensure time in migration never goes backward
Date: Wed, 7 May 2014 22:33:25 -0300 [thread overview]
Message-ID: <20140508013325.GA5474@amt.cnet> (raw)
In-Reply-To: <1399297882-3444-1-git-send-email-agraf@suse.de>
On Mon, May 05, 2014 at 03:51:22PM +0200, Alexander Graf wrote:
> When we migrate we ask the kernel about its current belief on what the guest
> time would be. However, I've seen cases where the kvmclock guest structure
> indicates a time more recent than the kvm returned time.
>
> To make sure we never go backwards, calculate what the guest would have seen
> as time at the point of migration and use that value instead of the kernel
> returned one when it's more recent.
>
> While this doesn't fix the underlying issue that the kernel's view of time
> is skewed, it allows us to safely migrate guests even from sources that are
> known broken.
>
> Signed-off-by: Alexander Graf <agraf@suse.de>
OK Alexander better move this logic to the kernel, in KVM_GET_CLOCK.
Otherwise every user of KVM_GET_CLOCK would have to apply the
workaround.
next prev parent reply other threads:[~2014-05-08 1:38 UTC|newest]
Thread overview: 44+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-05-05 13:51 [PATCH] kvmclock: Ensure time in migration never goes backward Alexander Graf
2014-05-05 13:51 ` [Qemu-devel] " Alexander Graf
2014-05-05 17:46 ` Marcin Gibuła
2014-05-05 18:05 ` Alexander Graf
2014-05-05 18:26 ` Marcin Gibuła
2014-05-05 23:27 ` Marcelo Tosatti
2014-05-05 23:27 ` Marcelo Tosatti
2014-05-05 23:31 ` Marcelo Tosatti
2014-05-05 23:31 ` Marcelo Tosatti
2014-05-06 8:07 ` Marcin Gibuła
2014-05-06 8:07 ` [Qemu-devel] " Marcin Gibuła
2014-05-06 7:11 ` Alexander Graf
2014-05-06 7:37 ` Marcin Gibuła
2014-05-05 23:23 ` Marcelo Tosatti
2014-05-05 23:23 ` [Qemu-devel] " Marcelo Tosatti
2014-05-05 23:31 ` Marcelo Tosatti
2014-05-05 23:31 ` [Qemu-devel] " Marcelo Tosatti
2014-05-06 7:18 ` Alexander Graf
2014-05-06 7:18 ` [Qemu-devel] " Alexander Graf
2014-05-06 19:54 ` Marcin Gibuła
2014-05-07 23:23 ` Marcelo Tosatti
2014-05-07 23:23 ` Marcelo Tosatti
2014-05-07 23:21 ` Marcelo Tosatti
2014-05-07 23:21 ` [Qemu-devel] " Marcelo Tosatti
2014-05-07 23:29 ` Alexander Graf
2014-05-07 23:29 ` [Qemu-devel] " Alexander Graf
2014-05-06 7:16 ` Alexander Graf
2014-05-06 7:16 ` [Qemu-devel] " Alexander Graf
2014-05-07 10:04 ` Nick Thomas
2014-05-07 10:04 ` [Qemu-devel] " Nick Thomas
2014-05-08 1:33 ` Marcelo Tosatti [this message]
2014-05-08 1:33 ` Marcelo Tosatti
2014-05-08 7:21 ` Alexander Graf
2014-05-08 7:21 ` [Qemu-devel] " Alexander Graf
2014-05-09 2:28 ` Marcelo Tosatti
2014-05-09 2:28 ` [Qemu-devel] " Marcelo Tosatti
2014-05-09 11:53 ` Paolo Bonzini
2014-05-09 11:53 ` [Qemu-devel] " Paolo Bonzini
2014-05-12 20:26 ` Alexander Graf
2014-05-12 20:26 ` [Qemu-devel] " Alexander Graf
2014-05-14 7:26 ` Marcelo Tosatti
2014-05-14 7:26 ` [Qemu-devel] " Marcelo Tosatti
2014-05-14 6:47 ` Marcelo Tosatti
2014-05-14 6:47 ` [Qemu-devel] " Marcelo Tosatti
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=20140508013325.GA5474@amt.cnet \
--to=mtosatti@redhat.com \
--cc=agraf@suse.de \
--cc=kvm@vger.kernel.org \
--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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.