xen-devel.lists.xenproject.org archive mirror
 help / color / mirror / Atom feed
From: Anthony PERARD <anthony.perard@citrix.com>
To: Wei Liu <wei.liu2@citrix.com>
Cc: Xen-devel <xen-devel@lists.xenproject.org>,
	Stefano Stabellini <sstabellini@kernel.org>,
	Ian Jackson <Ian.Jackson@eu.citrix.com>,
	Jan Beulich <JBeulich@suse.com>
Subject: Re: [PATCH 3/3] libxl: update vcpus bitmap in retrieved guest config
Date: Tue, 7 Jun 2016 17:36:05 +0100	[thread overview]
Message-ID: <20160607163605.GM1184@perard.uk.xensource.com> (raw)
In-Reply-To: <20160607160307.GN25922@citrix.com>

On Tue, Jun 07, 2016 at 05:03:07PM +0100, Wei Liu wrote:
> On Tue, Jun 07, 2016 at 03:45:29PM +0100, Anthony PERARD wrote:
> > On Tue, Jun 07, 2016 at 12:24:02PM +0100, Wei Liu wrote:
> > > ... because the available vcpu bitmap can change during domain life time
> > > due to cpu hotplug and unplug.
> > > 
> > > For QEMU upstream, we interrogate QEMU for the number of vcpus. For
> > > others, we look directly into xenstore for information.
> > > 
> > > Reported-by: Jan Beulich <jbeulich@suse.com>
> > > Signed-off-by: Wei Liu <wei.liu2@citrix.com>
> > > ---
> > >  tools/libxl/libxl.c | 91 +++++++++++++++++++++++++++++++++++++++++++++++++++++
> > >  1 file changed, 91 insertions(+)
> > > 
> > > diff --git a/tools/libxl/libxl.c b/tools/libxl/libxl.c
> > > index 006b83f..4f8b663 100644
> > > --- a/tools/libxl/libxl.c
> > > +++ b/tools/libxl/libxl.c
> > > @@ -7222,6 +7222,57 @@ void libxl_mac_copy(libxl_ctx *ctx, libxl_mac *dst, libxl_mac *src)
> > >          (*dst)[i] = (*src)[i];
> > >  }
> > >  
> > > +static int libxl__update_avail_vcpus_qmp(libxl__gc *gc, uint32_t domid,
> > > +                                         unsigned int max_vcpus,
> > > +                                         libxl_bitmap *map)
> > > +{
> > > +    unsigned int count, i;
> > > +    int rc;
> > > +
> > > +    /* For QEMU upstream we always need to return the number
> > > +     * of cpus present to QEMU whether they are online or not;
> > > +     * otherwise QEMU won't accept the saved state.
> > > +     */
> > > +    rc = libxl__qmp_query_cpus(gc, domid, &count);
> > > +    if (rc) {
> > > +        LOG(ERROR, "fail to get number of cpus for domain %d", domid);
> > > +        goto out;
> > > +    }
> > > +
> > > +    for (i = 0; i < count; i++)
> > > +        libxl_bitmap_set(map, i);
> > 
> > What if I have cpu 1, 7 and 42 online, but all the other offline?
> > 
> 
> I have more or less the same question when I wrote this patch. At that
> time I thought the avail_vcpus was only used for generating -smp option
> to QEMU.

Right, and -smp does not let you set a bitmap... I guess having a count
for this purpose is enough.

> In your example, you will have -smp 3,maxvcpus=$Y. I think the migration
> should still succeed. Furthermore, the cpu-add operation doesn't care,
> so it probably won't have visible effect.

What does cpu-add not care about?

> I agree it would be good to set the exact bits though -- if you can tell
> me which field to test. 
> 

From the example below, I guess it would be "CPU".

> 
> -> { "execute": "query-cpus" }
> <- {
>       "return":[
>          {
>             "CPU":0,
>             "current":true,
>             "halted":false,
>             "qom_path":"/machine/unattached/device[0]",
>             "arch":"x86",
>             "pc":3227107138,
>             "thread_id":3134
>          },
>          {
>             "CPU":1,
>             "current":false,
>             "halted":true,
>             "qom_path":"/machine/unattached/device[2]",
>             "arch":"x86",
>             "pc":7108165,
>             "thread_id":3135
>          }
>       ]
>    }
> EQMP

-- 
Anthony PERARD

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

  parent reply	other threads:[~2016-06-07 16:36 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-06-07 11:23 [PATCH 0/3] libxl: update available vcpus map in retrieve configuration function Wei Liu
2016-06-07 11:24 ` [PATCH 1/3] libxl: introduce libxl__json_array_count Wei Liu
2016-06-07 11:24 ` [PATCH 2/3] libxl: introduce libxl__qmp_query_cpus Wei Liu
2016-06-07 11:24 ` [PATCH 3/3] libxl: update vcpus bitmap in retrieved guest config Wei Liu
2016-06-07 14:45   ` Anthony PERARD
2016-06-07 16:03     ` Wei Liu
2016-06-07 16:22       ` Dario Faggioli
2016-06-07 16:46         ` Wei Liu
2016-06-07 17:05           ` Dario Faggioli
2016-06-07 16:36       ` Anthony PERARD [this message]
2016-06-07 16:39         ` Wei Liu
2016-06-14 16:31   ` Ian Jackson
2016-06-14 16:39     ` Wei Liu

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=20160607163605.GM1184@perard.uk.xensource.com \
    --to=anthony.perard@citrix.com \
    --cc=Ian.Jackson@eu.citrix.com \
    --cc=JBeulich@suse.com \
    --cc=sstabellini@kernel.org \
    --cc=wei.liu2@citrix.com \
    --cc=xen-devel@lists.xenproject.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).