qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: "Daniel P. Berrangé" <berrange@redhat.com>
To: Peter Xu <peterx@redhat.com>
Cc: qemu-devel@nongnu.org, "Paolo Bonzini" <pbonzini@redhat.com>,
	"Juan Quintela" <quintela@redhat.com>,
	"Markus Armbruster" <armbru@redhat.com>,
	"Marc-André Lureau" <marcandre.lureau@redhat.com>,
	"Stefan Hajnoczi" <stefanha@redhat.com>,
	"Dr . David Alan Gilbert" <dgilbert@redhat.com>
Subject: Re: [Qemu-devel] [PATCH 07/14] qio/chardev: update net listener gcontext
Date: Wed, 28 Feb 2018 13:06:08 +0000	[thread overview]
Message-ID: <20180228130608.GH17774@redhat.com> (raw)
In-Reply-To: <20180228125216.GH27381@xz-mi>

On Wed, Feb 28, 2018 at 08:52:16PM +0800, Peter Xu wrote:
> On Wed, Feb 28, 2018 at 09:25:11AM +0000, Daniel P. Berrangé wrote:
> > On Wed, Feb 28, 2018 at 01:06:26PM +0800, Peter Xu wrote:
> > > TCP chardevs can be using QIO network listeners working in the
> > > background when in listening mode.  However the network listeners are
> > > always running in main context.  This can race with chardevs that are
> > > running in non-main contexts.
> > > 
> > > To solve this: firstly introduce qio_net_listener_set_context() to allow
> > > caller to set gcontext for network listeners.  Then call it in
> > > tcp_chr_update_read_handler(), with the newly cached gcontext.
> > > 
> > > It's fairly straightforward after we have introduced some net listener
> > > helper functions - basically we unregister the GSources and add them
> > > back with the correct context.
> > > 
> > > Signed-off-by: Peter Xu <peterx@redhat.com>
> > > ---
> > >  chardev/char-socket.c     |  9 +++++++++
> > >  include/io/net-listener.h | 12 ++++++++++++
> > >  io/net-listener.c         |  7 +++++++
> > >  3 files changed, 28 insertions(+)
> > > 
> > > diff --git a/chardev/char-socket.c b/chardev/char-socket.c
> > > index 43a2cc2c1c..8f0935cd15 100644
> > > --- a/chardev/char-socket.c
> > > +++ b/chardev/char-socket.c
> > > @@ -559,6 +559,15 @@ static void tcp_chr_update_read_handler(Chardev *chr)
> > >  {
> > >      SocketChardev *s = SOCKET_CHARDEV(chr);
> > >  
> > > +    if (s->listener) {
> > > +        /*
> > > +         * It's possible that chardev context is changed in
> > > +         * qemu_chr_be_update_read_handlers().  Reset it for QIO net
> > > +         * listener if there is.
> > > +         */
> > > +        qio_net_listener_set_context(s->listener, chr->gcontext);
> > > +    }
> > > +
> > >      if (!s->connected) {
> > >          return;
> > >      }
> > > diff --git a/include/io/net-listener.h b/include/io/net-listener.h
> > > index 566be283b3..39dede9d6f 100644
> > > --- a/include/io/net-listener.h
> > > +++ b/include/io/net-listener.h
> > > @@ -106,6 +106,18 @@ int qio_net_listener_open_sync(QIONetListener *listener,
> > >                                 SocketAddress *addr,
> > >                                 Error **errp);
> > >  
> > > +/**
> > > + * qio_net_listener_set_context:
> > > + * @listener: the net listener object
> > > + * @context: the context that we'd like to bind the sources to
> > > + *
> > > + * This helper does not do anything but moves existing net listener
> > > + * sources from the old one to the new one.  It can be seen as a
> > > + * no-operation if there is no listening source at all.
> > > + */
> > > +void qio_net_listener_set_context(QIONetListener *listener,
> > > +                                  GMainContext *context);
> > 
> > I don't think this is a good design. The GMainContext should be provided
> > to the qio_net_listener_set_client_func method immediately, not updated
> > after the fact
> 
> After the patches, this is qio_net_listener_set_client_func_full():
> 
> void qio_net_listener_set_client_func_full(QIONetListener *listener,
>                                            QIONetListenerClientFunc func,
>                                            gpointer data,
>                                            GDestroyNotify notify,
>                                            GMainContext *context)
> {
>     if (listener->io_notify) {
>         listener->io_notify(listener->io_data);
>     }
>     listener->io_func = func;
>     listener->io_data = data;
>     listener->io_notify = notify;
> 
>     qio_net_listener_sources_clear(listener);
>     qio_net_listener_sources_update(listener, context);
> }
> 
> This is qio_net_listener_set_context():
> 
> void qio_net_listener_set_context(QIONetListener *listener,
>                                   GMainContext *context)
> {
>     qio_net_listener_sources_clear(listener);
>     qio_net_listener_sources_update(listener, context);
> }
> 
> So... Basically qio_net_listener_set_context() is just a simplified
> version of qio_net_listener_set_client_func_full(), but without
> passing in the funcs again.  So do you mean that I can just avoid
> introducing this function and call
> qio_net_listener_set_client_func_full() directly?

Yes, I don't see any reason to allow GMainContext to be changed separately
from setting the callback functions. The caller can easily just call
qio_net_listener_set_client_func_full() directly with the new GMainContext

Regards,
Daniel
-- 
|: https://berrange.com      -o-    https://www.flickr.com/photos/dberrange :|
|: https://libvirt.org         -o-            https://fstop138.berrange.com :|
|: https://entangle-photo.org    -o-    https://www.instagram.com/dberrange :|

  reply	other threads:[~2018-02-28 13:06 UTC|newest]

Thread overview: 41+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-02-28  5:06 [Qemu-devel] [PATCH 00/14] qio: general non-default GMainContext support Peter Xu
2018-02-28  5:06 ` [Qemu-devel] [PATCH 01/14] chardev: fix leak in tcp_chr_telnet_init_io() Peter Xu
2018-02-28  9:26   ` Daniel P. Berrangé
2018-02-28  5:06 ` [Qemu-devel] [PATCH 02/14] qio: rename qio_task_thread_result Peter Xu
2018-02-28  9:26   ` Daniel P. Berrangé
2018-02-28  5:06 ` [Qemu-devel] [PATCH 03/14] qio: introduce qio_channel_add_watch_full() Peter Xu
2018-02-28  9:08   ` Daniel P. Berrangé
2018-02-28 12:44     ` Peter Xu
2018-02-28 12:47       ` Daniel P. Berrangé
2018-02-28 13:01         ` Peter Xu
2018-02-28  5:06 ` [Qemu-devel] [PATCH 04/14] migration: let incoming side use thread context Peter Xu
2018-02-28  9:10   ` Daniel P. Berrangé
2018-03-01  4:33     ` Peter Xu
2018-02-28 17:43   ` Dr. David Alan Gilbert
2018-03-01  2:53     ` Peter Xu
2018-03-01  9:58       ` Dr. David Alan Gilbert
2018-02-28  5:06 ` [Qemu-devel] [PATCH 05/14] qio: refactor net listener source operations Peter Xu
2018-02-28  5:06 ` [Qemu-devel] [PATCH 06/14] qio: store gsources for net listeners Peter Xu
2018-02-28  5:06 ` [Qemu-devel] [PATCH 07/14] qio/chardev: update net listener gcontext Peter Xu
2018-02-28  9:25   ` Daniel P. Berrangé
2018-02-28 12:52     ` Peter Xu
2018-02-28 13:06       ` Daniel P. Berrangé [this message]
2018-02-28  5:06 ` [Qemu-devel] [PATCH 08/14] chardev: allow telnet gsource to switch gcontext Peter Xu
2018-02-28  5:06 ` [Qemu-devel] [PATCH 09/14] qio: basic non-default context support for thread Peter Xu
2018-02-28  5:06 ` [Qemu-devel] [PATCH 10/14] qio: refcount QIOTask Peter Xu
2018-02-28  9:16   ` Daniel P. Berrangé
2018-02-28 12:54     ` Peter Xu
2018-02-28 13:07       ` Daniel P. Berrangé
2018-02-28 13:15         ` Peter Xu
2018-02-28  5:06 ` [Qemu-devel] [PATCH 11/14] qio/chardev: return QIOTask when connect async Peter Xu
2018-02-28  9:20   ` Daniel P. Berrangé
2018-02-28 13:07     ` Peter Xu
2018-02-28  5:06 ` [Qemu-devel] [PATCH 12/14] qio: move QIOTaskThreadData into QIOTask Peter Xu
2018-02-28  5:06 ` [Qemu-devel] [PATCH 13/14] qio: allow threaded qiotask to switch contexts Peter Xu
2018-02-28  9:23   ` Daniel P. Berrangé
2018-02-28 13:05     ` Peter Xu
2018-02-28 13:20       ` Daniel P. Berrangé
2018-03-01  8:49         ` Peter Xu
2018-02-28  5:06 ` [Qemu-devel] [PATCH 14/14] qio/chardev: specify gcontext for TLS handshake Peter Xu
2018-02-28 13:22   ` Daniel P. Berrangé
2018-03-01  6:28     ` Peter Xu

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=20180228130608.GH17774@redhat.com \
    --to=berrange@redhat.com \
    --cc=armbru@redhat.com \
    --cc=dgilbert@redhat.com \
    --cc=marcandre.lureau@redhat.com \
    --cc=pbonzini@redhat.com \
    --cc=peterx@redhat.com \
    --cc=qemu-devel@nongnu.org \
    --cc=quintela@redhat.com \
    --cc=stefanha@redhat.com \
    /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).