[ovs-dev] [ovsdb monitor fix v2 3/3] ovsdb: avoid unnecessary call to ovsdb_monitor_get_update()

[Ben Pfaff]

On Wed, Feb 24, 2016 at 02:58:25PM -0800, Andy Zhou wrote:
> On Wed, Feb 24, 2016 at 2:44 PM, Ben Pfaff <blp at ovn.org> wrote:
> 
> > On Mon, Feb 22, 2016 at 04:28:14PM -0800, Andy Zhou wrote:
> > > Avoid calling ovsdb_monitor_get_update() monitor has been flushed
> > > already.
> > >
> > > Reported-by: Liran Schour <lirans at il.ibm.com>
> > > Signed-off-by: Andy Zhou <azhou at ovn.org>
> > >
> > > ----
> > > v1 -> v2: no change.
> >
> > What are the visible consequences of this change?  That is, does it
> > suppress sending something to clients, or does it just suppress work in
> > the monitor, or...?  This information is probably suitable to include in
> > the commit message.
> >
> 
> Make sense, how about the following?
> 
> Optimizing ovsdb_jsonrpc_mintor_flush_all() by avoiding calling
> ovsdb_monitor_get_update() on monitors that do not have any
> unflushed updates.  This change saves CPU cycles on ovsdb-server's
> main loop, but should not introduce any client visible changes.

Thanks.

Acked-by: Ben Pfaff <blp at ovn.org>