Looks good to me. Only a few minor comments, and questions. This
will really make it much easier to work on both the bridge and bonding
so I think it's a huge win.
Ethan
> + /* Tag set saved for next bond_run(). */
> + struct tag_set tags;
> +};
It wasn't entirely clear to me exactly wha
> static void
> bond_choose_active_iface(struct port *port)
> {
> static struct vlog_rate_limit rl = VLOG_RATE_LIMIT_INIT(5, 20);
> + struct iface *old_active_iface = port->active_iface;
>
> port->active_iface = bond_choose_iface(port);
> if (port->active_iface) {
> - VLOG_
Nothing uses these anymore. ofputil_decode_flow_stats_reply() is a better
alternative.
---
lib/ofp-util.c | 45 -
lib/ofp-util.h |7 ---
2 files changed, 0 insertions(+), 52 deletions(-)
diff --git a/lib/ofp-util.c b/lib/ofp-util.c
index d91f
---
ofproto/ofproto.c | 16 +---
1 files changed, 5 insertions(+), 11 deletions(-)
diff --git a/ofproto/ofproto.c b/ofproto/ofproto.c
index ce8c99b..dc5c915 100644
--- a/ofproto/ofproto.c
+++ b/ofproto/ofproto.c
@@ -1124,13 +1124,6 @@ ofport_equal(const struct ofport *a_, const stru
These patches follow up on the connmgr series with a few
more minor improvements.
Ben Pfaff (2):
ofp-util: Remove flow_stats_iterator, flows_stats_first(),
flow_stats_next()
ofproto: Get rid of send_port_status() trivial wrapper function.
lib/ofp-util.c| 45
On Tue, Mar 29, 2011 at 11:56:15AM -0700, Ethan Jackson wrote:
> > +/* Frees 'mgr' and all of its resources. */
> > +void
> > +connmgr_destroy(struct connmgr *mgr)
> > +{
> > + ? ?struct ofservice *ofservice, *next_ofservice;
> > + ? ?struct ofconn *ofconn, *next_ofconn;
> > + ? ?size_t i;
> > +
>
> +/* Frees 'mgr' and all of its resources. */
> +void
> +connmgr_destroy(struct connmgr *mgr)
> +{
> + struct ofservice *ofservice, *next_ofservice;
> + struct ofconn *ofconn, *next_ofconn;
> + size_t i;
> +
> + if (!mgr) {
> + return;
> + }
> +
> + LIST_FOR_EACH_SAFE (ofc
Oh, I misunderstood your statement. I thought you meant remove
that line of code entirely so the loop would be:
for (a;;c) {
I'm fine with moving it up. I'll do that and merge.
On Tue, Mar 29, 2011 at 11:11 AM, Ben Pfaff wrote:
> On Tue, Mar 29, 2011 at 11:08:18AM -0700, Ethan Jackson wro
On Tue, Mar 29, 2011 at 11:08:18AM -0700, Ethan Jackson wrote:
> > Optionally you could move "row != NULL" up a line, too.
>
> Can I? I can certainly change it to simply "row". If I remove it
> entirely won't it either loop forever or null dereference at some
> point?
The following are equivale
> Optionally you could move "row != NULL" up a line, too.
Can I? I can certainly change it to simply "row". If I remove it
entirely won't it either loop forever or null dereference at some
point?
Ethan
___
dev mailing list
dev@openvswitch.org
http://o
On Tue, Mar 29, 2011 at 10:59:59AM -0700, Ethan Jackson wrote:
> Noticed this last night while playing around with the clang static
> analyzer.
Looks good, thank you.
Optionally you could move "row != NULL" up a line, too.
___
dev mailing list
dev@openv
Noticed this last night while playing around with the clang static
analyzer.
---
utilities/ovs-vsctl.c |5 ++---
1 files changed, 2 insertions(+), 3 deletions(-)
diff --git a/utilities/ovs-vsctl.c b/utilities/ovs-vsctl.c
index 0068f64..b8c6d30 100644
--- a/utilities/ovs-vsctl.c
+++ b/utilitie
Thanks, I pushed this.
(That was fast!)
On Tue, Mar 29, 2011 at 10:10:37AM -0700, Justin Pettit wrote:
> Good catch. Makes sense to me.
>
> --Justin
>
>
> On Mar 29, 2011, at 10:08 AM, Ben Pfaff wrote:
>
> > The test ofproto_has_primary_controller() is meaningless, since OFPP_NORMAL
> > can
On Tue, Mar 29, 2011 at 09:56:49AM -0700, Justin Pettit wrote:
> On Mar 29, 2011, at 9:36 AM, Ben Pfaff wrote:
>
> > +pid = read_pidfile_if_exists(_pidfile)
> > +if pid > 0:
> > if not _overwrite_pidfile:
> > -msg = "%s: already running" % _pidfile
> > -logg
Good catch. Makes sense to me.
--Justin
On Mar 29, 2011, at 10:08 AM, Ben Pfaff wrote:
> The test ofproto_has_primary_controller() is meaningless, since OFPP_NORMAL
> can cause the MAC learning table and port bonding to be in use even when
> there is a controller.
>
> I see that this bug has
The test ofproto_has_primary_controller() is meaningless, since OFPP_NORMAL
can cause the MAC learning table and port bonding to be in use even when
there is a controller.
I see that this bug has been here since early 2009, when the OFPP_NORMAL
feature was introduced in the bridge. (Obviously it'
On Mar 29, 2011, at 9:36 AM, Ben Pfaff wrote:
> +pid = read_pidfile_if_exists(_pidfile)
> +if pid > 0:
> if not _overwrite_pidfile:
> -msg = "%s: already running" % _pidfile
> -logging.error(msg)
> +msg = "%s: already running as pid %d" % (_pidfi
It seems possible that "restart" or a quick application of "stop" then
"start" could kill ovs-xapi-sync without starting it again, if
ovs-xapi-sync takes a little while to die, long enough for the next
instance of it to see that its pidfile is still open and locked.
I hope that this fixes some odd
Otherwise it's hard to diagnose later if the daemon failed to start because
it thinks that it is already running.
---
lib/daemon.c |2 ++
python/ovs/daemon.py |6 --
2 files changed, 6 insertions(+), 2 deletions(-)
diff --git a/lib/daemon.c b/lib/daemon.c
index 2f6f081..17ec9f
This function substantially duplicated read_pidfile(), so reuse that
code instead.
---
lib/daemon.c | 59 -
lib/daemon.h |1 +
python/ovs/daemon.py | 25 -
3 files changed, 49 insertions(+), 36 deletions(-)
On Mon, Mar 28, 2011 at 10:38:00PM -0700, Justin Pettit wrote:
> Looks good.
>
> This wouldn't explain the restart issues we've seen that cause
> ovsdb-server or ovs-vswitchd to not start (or run multiple
> instances), would it?
It would definitely explain why ovs-xapi-sync wasn't running (I'm
pr
On Mon, Mar 28, 2011 at 10:27:58PM -0700, Justin Pettit wrote:
> On Mar 28, 2011, at 4:24 PM, Ben Pfaff wrote:
>
> > diff --git a/python/ovs/daemon.py b/python/ovs/daemon.py
> > index 6240e0d..c8af043 100644
> > --- a/python/ovs/daemon.py
> > +++ b/python/ovs/daemon.py
> > @@ -139,7 +139,10 @@ def
On Mon, Mar 28, 2011 at 10:18:28PM -0700, Justin Pettit wrote:
> On Mar 28, 2011, at 4:24 PM, Ben Pfaff wrote:
>
> > @@ -141,7 +141,9 @@ def die_if_already_running():
> > ignore_existing_pidfile() has been called, terminate the program."""
> > if _already_running():
> > if not _ove
On Mon, Mar 28, 2011 at 02:03:13PM -0700, Ethan Jackson wrote:
> On Fri, Mar 25, 2011 at 10:35 AM, Ben Pfaff wrote:
> > ---
> > ?vswitchd/vswitch.xml | ? 17 +
> > ?1 files changed, 13 insertions(+), 4 deletions(-)
> >
> > diff --git a/vswitchd/vswitch.xml b/vswitchd/vswitch.xml
> >
On Mon, Mar 28, 2011 at 01:43:41PM -0700, Ethan Jackson wrote:
> Looks Good to me.
>
> It might be worth commenting why we are allocating and reserving
> VLAN_HEADER_LEN extra bytes. That may confuse a future reader.
That's reasonable. I folded this in:
diff --git a/lib/packets.c b/lib/packets
On Mon, Mar 28, 2011 at 01:40:27PM -0700, Ethan Jackson wrote:
> Looks Good to me.
>
> Is there any reason not to simply always set packet->l2? I think it's
> fine how it is, just curious.
I guess there isn't. I changed it to do that.
___
dev mailing
Title: Paraiso
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
Greetings,
I am sorry to encroach into your privacy in this manner, I found your names
listed in the Trade Center Chambers of Commerce directory here in Japan.
I find it pleasurable to offer you my partnership in business, I only pray at
this time that your address is still valid.
I want to so
28 matches
Mail list logo
