From: Wei Liu <wei.liu2@citrix.com>
To: Ian Jackson <ian.jackson@eu.citrix.com>
Cc: Xen-devel <xen-devel@lists.xenproject.org>,
Wei Liu <wei.liu2@citrix.com>,
andrew.cooper3@citrix.com
Subject: Re: [XTF PATCH] xtf-runner: fix two synchronisation issues
Date: Fri, 29 Jul 2016 15:46:11 +0100 [thread overview]
Message-ID: <20160729144611.GH22419@citrix.com> (raw)
In-Reply-To: <22427.27034.410515.535834@mariner.uk.xensource.com>
On Fri, Jul 29, 2016 at 03:35:06PM +0100, Ian Jackson wrote:
> Wei Liu writes ("Re: [XTF PATCH] xtf-runner: fix two synchronisation issues"):
> > On Fri, Jul 29, 2016 at 03:21:52PM +0100, Ian Jackson wrote:
> > > Wei Liu writes ("[XTF PATCH] xtf-runner: fix two synchronisation issues"):
> > > > There were two synchronisation issues for the old code:
> > > >
> > > > 1. There was no guarantee that guest console was ready before "xl
> > > > console" invocation.
> > >
> > > Is this not a bug in xl console ?
> >
> > I don't think so. It gives up when it can't get tty from xenstore. I
> > think that's reasonable.
>
> If you say
> xl create /etc/xen/foo.cfg
> xl console foo
> then surely you shouldn't find that xl console fails because of some
> race.
>
What if the guest just doesn't have / want a console? In that case there
will never be a tty node in xenstore, so xl console should wait forever
for one? What is the expectation for this case?
> > > > 2. Poll xenstore guest console node to make sure console is available
> > > > before "xl console" invocation.
> > >
> > > Users of things like xl shouldn't need to prat about in xenstore too.
> >
> > What does this mean?
>
> I mean that xenstore is part of the implementation of libxl, not part
> of its public interface. In this sense it is a bit like libxc.
> libxl callers should (in general) not look at xenstore - and of
> course they should therefore;2~ not need to do so.
>
In this particular case, all nodes that xtf-runner code reles on are
documented, so I already consider them public interfaces.
> If the xtf runner needs to look in xenstore (other than perhaps if
> it's doing strange things there as part of its test cases) then that
> means there is some interface or capability missing in libxl.
>
I don't disagree.
But the missing capability won't be backported to older version of Xen.
I don't want to eliminate the possibility for running xtf on older
versions of Xen.
> OTOH I'm slightly confused because I was under the impression that
> there was a polling loop in xl console (xenconsole) already.
>
> Maybe I don't understand your problem.
>
The issue is that when it tries to get tty from xenstore and there isn't
one, it gives up. Your polling loop is something else.
http://osstest.xs.citrite.net/~osstest/testlogs/logs/66855/test-xtf-amd64-amd64-1/10.ts-xtf-fep.log
Wei.
> Ian.
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel
next prev parent reply other threads:[~2016-07-29 14:46 UTC|newest]
Thread overview: 45+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-07-29 12:07 [XTF PATCH] xtf-runner: fix two synchronisation issues Wei Liu
2016-07-29 12:43 ` Andrew Cooper
2016-07-29 12:58 ` Wei Liu
2016-07-29 13:06 ` Andrew Cooper
2016-07-29 13:12 ` Wei Liu
2016-07-29 13:23 ` Andrew Cooper
2016-07-29 13:26 ` Wei Liu
2016-07-29 14:31 ` Ian Jackson
2016-07-29 14:55 ` Wei Liu
2016-07-29 16:18 ` Ian Jackson
2016-07-29 16:35 ` Andrew Cooper
2016-07-29 16:41 ` Wei Liu
2016-07-29 15:05 ` Andrew Cooper
2016-08-01 13:16 ` [RFC PATCH 0/8] Fix console " Wei Liu
2016-08-01 13:16 ` [RFC PATCH 1/8] tools/console: fix help string in client Wei Liu
2016-08-05 15:40 ` Ian Jackson
2016-08-01 13:16 ` [RFC PATCH 2/8] tools/console: introduce --start-notify-fd option for console client Wei Liu
2016-08-05 15:43 ` Ian Jackson
2016-08-01 13:16 ` [RFC PATCH 3/8] libxl: factor out libxl__console_tty_path Wei Liu
2016-08-05 15:44 ` Ian Jackson
2016-08-01 13:16 ` [RFC PATCH 4/8] libxl: wait up to 5s in libxl_console_exec for xenconsoled Wei Liu
2016-08-05 15:48 ` Ian Jackson
2016-08-01 13:16 ` [RFC PATCH 5/8] libxl: libxl_{primary_, }console_exec now take notify_fd argument Wei Liu
2016-08-05 15:49 ` Ian Jackson
2016-08-05 15:50 ` Ian Jackson
2016-08-01 13:16 ` [RFC PATCH 6/8] docs: document xenconsole startup protocol Wei Liu
2016-08-05 15:52 ` Ian Jackson
2016-08-01 13:16 ` [RFC PATCH 7/8] xl: use " Wei Liu
2016-08-05 15:55 ` Ian Jackson
2016-08-01 13:16 ` [RFC PATCH 8/8] tools/console: remove 5s bodge in console client Wei Liu
2016-08-05 15:57 ` Ian Jackson
2016-08-05 16:16 ` Wei Liu
2016-08-05 16:18 ` Ian Jackson
2016-08-05 16:28 ` Wei Liu
2016-08-05 16:32 ` Ian Jackson
2016-08-05 16:36 ` Wei Liu
2016-08-05 17:23 ` Wei Liu
2016-08-08 10:07 ` Ian Jackson
2016-08-01 14:04 ` [XTF PATCH] xtf-runner: use xl create -Fc directly Wei Liu
2016-07-29 13:27 ` [XTF PATCH] xtf-runner: fix two synchronisation issues Andrew Cooper
2016-07-29 14:21 ` Ian Jackson
2016-07-29 14:25 ` Wei Liu
2016-07-29 14:35 ` Ian Jackson
2016-07-29 14:46 ` Wei Liu [this message]
2016-07-29 14:26 ` Andrew Cooper
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=20160729144611.GH22419@citrix.com \
--to=wei.liu2@citrix.com \
--cc=andrew.cooper3@citrix.com \
--cc=ian.jackson@eu.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).