On Tue, Jul 28, 2015 at 8:44 AM, Ben Pfaff wrote:
> This will be used in documentation for an upcoming change, to document
> how Geneve OVN options are encoded.
>
> The code in this change is from a series (not yet submitted) that makes
> much more extensive use of it for documenting protocol hea
Until now, OVS has parsed all OF1.1+ group buckets that lack a weight
as having weight 1. Unfortunately, OpenFlow says that only "select"
groups may have a nonzero weight, and requires reporting an error for
other kinds of groups that have a nonzero weight. This commit fixes
the problem by parsin
On Tue, Jul 28, 2015 at 09:42:45PM -0300, Flavio Leitner wrote:
> On Tue, Jul 28, 2015 at 03:21:14PM -0700, Ben Pfaff wrote:
> > This complies with OpenFlow 1.1 and later.
> >
> > Found by OFTest.
>
> OF1.1 specs says on page 42 that weight is only defined for
> select groups.
>
> and on page 23
Thanks Ethan, applied to master.
On Tue, Jul 28, 2015 at 06:54:57PM -0700, Ethan Jackson wrote:
> Ah my bad sorry.
>
> Acked-by: Ethan Jackson
>
>
> On Tue, Jul 28, 2015 at 6:53 PM, Ben Pfaff wrote:
> > Fixes the following error:
> >
> > The following files are in git but not the distribu
Ah my bad sorry.
Acked-by: Ethan Jackson
On Tue, Jul 28, 2015 at 6:53 PM, Ben Pfaff wrote:
> Fixes the following error:
>
> The following files are in git but not the distribution:
> ovn/OVN-GW-HA.md
>
> CC: Ethan Jackson
> Signed-off-by: Ben Pfaff
> ---
> ovn/automake.mk | 3 ++-
>
Fixes the following error:
The following files are in git but not the distribution:
ovn/OVN-GW-HA.md
CC: Ethan Jackson
Signed-off-by: Ben Pfaff
---
ovn/automake.mk | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/ovn/automake.mk b/ovn/automake.mk
index 459ee36..117
Thanks Alex, I applied this to master.
On Tue, Jul 28, 2015 at 04:01:13PM -0700, Alex Wang wrote:
> Thx for fixing this~!
>
> Acked-by: Alex Wang
>
> On Tue, Jul 28, 2015 at 1:41 PM, Ben Pfaff wrote:
>
> > A commit to the database takes multiple trips through the main loop. In
> > that time,
On Tue, Jul 28, 2015 at 03:22:46PM -0700, Ben Pfaff wrote:
> This is implied by the list of ports that are valid for output in the
> various versions of the OpenFlow specification.
>
> Found by OFTest.
Looks good to me.
I wonder how one could set the output port to none.
Acked-by: Flavio Leitner
On Tue, Jul 28, 2015 at 03:21:49PM -0700, Ben Pfaff wrote:
> The OpenFlow specification says that only role changes to slave or master
> check the generation ID, so this is a bug fix.
>
> Found by OFTest.
Found in OF1.5 page 45 which says that switch must ignore
generation_id if the role in the R
On Tue, Jul 28, 2015 at 4:40 PM, Justin Pettit wrote:
> Do you think we should print a warning (at least once) to indicate that
> the integration bridge is missing? Otherwise, it could be confusing to
> someone why nothing's happening if they typo-ed their bridge name.
>
> Acked-by: Justin Petti
On Tue, Jul 28, 2015 at 03:21:14PM -0700, Ben Pfaff wrote:
> This complies with OpenFlow 1.1 and later.
>
> Found by OFTest.
OF1.1 specs says on page 42 that weight is only defined for
select groups.
and on page 23 says that switch must refuse to add the group
entry and must send an error msg wi
Acked-by: Justin Pettit
--Justin
> On Jul 28, 2015, at 8:44 AM, Ben Pfaff wrote:
>
> The generic *_init() names for these functions made it sounds like they
> do something more than just register IDL columns, even though that's all
> they do. Also, the controller_ctx that was passed into eac
Acked-by: Justin Pettit
--Justin
> On Jul 28, 2015, at 8:44 AM, Ben Pfaff wrote:
>
> Until now, if the chassis id was missing, ovn-controller exited. This
> commit makes ovn-controller wait for it to return.
>
> Signed-off-by: Ben Pfaff
> ---
> ovn/controller/binding.c| 3 +++
> ov
On Tue, Jul 28, 2015 at 04:36:29PM -0700, Justin Pettit wrote:
> This adds whitespace to the end of the line. (I've always wanted to
> say that to you.)
No worries. (I'll get my revenge when I implement the emergency flow
cache.)
___
dev mailing list
d
Thanks. I'm sure it needs a rebase, that's true pretty much all the
time ;-)
On Tue, Jul 28, 2015 at 04:50:54PM -0700, Gurucharan Shetty wrote:
> Okay. (I already committed this patch, so you probably need a rebase)
>
> On Tue, Jul 28, 2015 at 3:42 PM, Ben Pfaff wrote:
> > On Tue, Jul 28, 2015
> On Jul 28, 2015, at 4:18 PM, Justin Pettit wrote:
>
> Acked-by: Justin Pettit
I should amend my Ack. I think this introduces a problem.
>
>> On Jul 28, 2015, at 8:44 AM, Ben Pfaff wrote:
>>
>>
>> /* Returns true if the database is all cleaned up, false if more work is
>> * required. *
Okay. (I already committed this patch, so you probably need a rebase)
On Tue, Jul 28, 2015 at 3:42 PM, Ben Pfaff wrote:
> On Tue, Jul 28, 2015 at 01:44:42PM -0700, Ben Pfaff wrote:
>> On Fri, Jul 24, 2015 at 04:31:03PM -0700, Gurucharan Shetty wrote:
>> > In table 64, when a vlan tag is set on a
Do you think we should print a warning (at least once) to indicate that the
integration bridge is missing? Otherwise, it could be confusing to someone why
nothing's happening if they typo-ed their bridge name.
Acked-by: Justin Pettit
--Justin
> On Jul 28, 2015, at 8:44 AM, Ben Pfaff wrote:
I forgot to add my Ack:
Acked-by: Justin Pettit
--Justin
> On Jul 28, 2015, at 4:36 PM, Justin Pettit wrote:
>
>
>> On Jul 28, 2015, at 8:44 AM, Ben Pfaff wrote:
>>
>> I found it otherwise difficult to see what code depended on this.
>
> It might be nice to explain that you care about th
> On Jul 28, 2015, at 8:44 AM, Ben Pfaff wrote:
>
> I found it otherwise difficult to see what code depended on this.
It might be nice to explain that you care about the dependencies due to
upcoming patches, since it wasn't immediately obvious to me why you'd care.
> diff --git a/ovn/controll
Acked-by: Justin Pettit
--Justin
> On Jul 28, 2015, at 8:44 AM, Ben Pfaff wrote:
>
> These two pieces of code have different requirements and hardly anything in
> common, and I found it easier to understand and reasonably modify the
> overall structure of the program when they were separated.
Thx for fixing this~!
Acked-by: Alex Wang
On Tue, Jul 28, 2015 at 1:41 PM, Ben Pfaff wrote:
> A commit to the database takes multiple trips through the main loop. In
> that time, the database could change, but ovn-controller can't properly
> react to the changes until the commit has succeeded
On Tue, Jul 28, 2015 at 01:44:42PM -0700, Ben Pfaff wrote:
> On Fri, Jul 24, 2015 at 04:31:03PM -0700, Gurucharan Shetty wrote:
> > In table 64, when a vlan tag is set on a packet
> > destined to a container running inside a VM, we currently
> > don't revert it. This has an unintended consequence f
On Tue, Jul 28, 2015 at 3:30 PM, Ben Pfaff wrote:
> On Tue, Jul 28, 2015 at 03:30:05PM -0700, Ben Pfaff wrote:
>> On Tue, Jul 28, 2015 at 03:20:52PM -0700, Jesse Gross wrote:
>> > It's possible to imagine that a switch might want to report additional
>> > capabilities related to Geneve beyond just
On Tue, Jul 28, 2015 at 03:30:05PM -0700, Ben Pfaff wrote:
> On Tue, Jul 28, 2015 at 03:20:52PM -0700, Jesse Gross wrote:
> > It's possible to imagine that a switch might want to report additional
> > capabilities related to Geneve beyond just the number of options and
> > how much space they can c
On Tue, Jul 28, 2015 at 03:20:52PM -0700, Jesse Gross wrote:
> It's possible to imagine that a switch might want to report additional
> capabilities related to Geneve beyond just the number of options and
> how much space they can consume. Some examples include additional
> restrictions on parsing
This is implied by the list of ports that are valid for output in the
various versions of the OpenFlow specification.
Found by OFTest.
Signed-off-by: Ben Pfaff
---
lib/ofp-actions.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/lib/ofp-actions.c b/lib/ofp-actions.c
inde
The OpenFlow specification says that only role changes to slave or master
check the generation ID, so this is a bug fix.
Found by OFTest.
Signed-off-by: Ben Pfaff
---
ofproto/ofproto.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/ofproto/ofproto.c b/ofproto/ofproto.c
in
This complies with OpenFlow 1.1 and later.
Found by OFTest.
Signed-off-by: Ben Pfaff
---
lib/ofp-util.c | 4
1 file changed, 4 insertions(+)
diff --git a/lib/ofp-util.c b/lib/ofp-util.c
index 9996e84..28f6d8f 100644
--- a/lib/ofp-util.c
+++ b/lib/ofp-util.c
@@ -8592,6 +8592,10 @@ ofputil_
It's possible to imagine that a switch might want to report additional
capabilities related to Geneve beyond just the number of options and
how much space they can consume. Some examples include additional
restrictions on parsing (if this command is used for non-OVS implementations
or OVS changes h
On Tue, Jul 28, 2015 at 02:33:11PM -0700, Andy Zhou wrote:
> On Mon, Jul 27, 2015 at 1:58 PM, Andy Zhou wrote:
> > Ben, thanks for rebasing. I will review it soon.
> >
> > On Mon, Jul 27, 2015 at 1:56 PM, Ben Pfaff wrote:
> >> I spent some time recently looking at the results of "ovs-ofctl
> >> d
On Mon, Jul 27, 2015 at 1:58 PM, Andy Zhou wrote:
> Ben, thanks for rebasing. I will review it soon.
>
> On Mon, Jul 27, 2015 at 1:56 PM, Ben Pfaff wrote:
>> I spent some time recently looking at the results of "ovs-ofctl
>> dump-table-features". It was really distressing because of the volume o
On Mon, Jul 27, 2015 at 11:25 AM, Jesse Gross wrote:
> On Tue, Jul 21, 2015 at 9:52 PM, Pravin B Shelar wrote:
>> v2.3.0 - 14 Aug 2014
>> diff --git a/datapath/vport-gre.c b/datapath/vport-gre.c
>> index 7bbcf57..8a47ab2 100644
>> --- a/datapath/vport-gre.c
>> +++ b/datapath/vport-gre.c
>> @@ -3
On Thu, Jul 23, 2015 at 10:48 AM, Gary Mussar wrote:
> Dpdk allows users to create a config that includes other config files and
> then override values.
>
> Eg.
> defconfig_x86_64-native_vhost_cuse-linuxapp-gcc:
>
> CONFIG_RTE_BUILD_COMBINE_LIBS=y
> CONFIG_RTE_BUILD_SHARED_LIB=n
> CONFIG_RTE_LIBRT
On Tue, Jul 28, 2015 at 09:55:52AM +0300, Ilya Maximets wrote:
> Currently pmd threads select queues in pmd_load_queues() according to
> get_n_pmd_threads_on_numa(). This behavior leads to race between pmds,
> beacause dp_netdev_set_pmds_on_numa() starts them one by one and
> current number of thre
On Fri, Jul 24, 2015 at 04:31:03PM -0700, Gurucharan Shetty wrote:
> In table 64, when a vlan tag is set on a packet
> destined to a container running inside a VM, we currently
> don't revert it. This has an unintended consequence for
> broadcast traffic when one endpoint of the braodcast
> traffic
On Tue, Jul 28, 2015 at 11:56:05AM -0700, Alex Wang wrote:
> Curious here, if we could run into the following race (in main function):
>
> 1. first run of while loop: we have some change from ovnsb~
> 2. ovn-controller execute all *_run() functions to update~
> 3. the txn commit to ovsdb is INCOMP
A commit to the database takes multiple trips through the main loop. In
that time, the database could change, but ovn-controller can't properly
react to the changes until the commit has succeeded or failed. Since
commit f1fd765733 (ovn-controller: Avoid blocking to commit OVSDB
transactions), Ope
On 07/28/2015 04:12 PM, Ben Pfaff wrote:
> On Tue, Jul 21, 2015 at 12:59:19PM -0400, Russell Bryant wrote:
>> A recent patch added "type" and "options" columns to the Logical_Port
>> table in OVN_Northbound. This patch allows you to get and set those
>> columns with ovn-nbctl.
>>
>> Signed-off-by:
On Tue, Jul 21, 2015 at 12:59:20PM -0400, Russell Bryant wrote:
> Introduce a new logical port type called "localnet". A logical port
> with this type also has an option called "network_name". A "localnet"
> logical port represents a connection to a locally accessible network.
> ovn-controller wi
On Tue, Jul 21, 2015 at 12:59:19PM -0400, Russell Bryant wrote:
> A recent patch added "type" and "options" columns to the Logical_Port
> table in OVN_Northbound. This patch allows you to get and set those
> columns with ovn-nbctl.
>
> Signed-off-by: Russell Bryant
This seems reasonable, althou
On Tue, Jul 21, 2015 at 12:59:15PM -0400, Russell Bryant wrote:
> Add a new OVN configuration entry in the Open_vSwitch database called
> "ovn-bridge-mappings". This allows the configuration of mappings
> between a physical network name and an OVS bridge that provides
> connectivity to that networ
On Tue, Jul 21, 2015 at 12:59:18PM -0400, Russell Bryant wrote:
> We have started discussing the use of the logical port abstraction in
> OVN to represent special types of connections into an OVN logical
> switch. This patch proposes some schema updates to reflect these
> special types of logical
On 07/28/2015 04:06 PM, Ben Pfaff wrote:
> On Tue, Jul 21, 2015 at 12:59:16PM -0400, Russell Bryant wrote:
>> While parsing the OVN bridge mapping configuration, ensure that patch
>> ports exist between the OVN integration bridge and the physical
>> network bridge. If they do not exist, create the
On 07/28/2015 04:02 PM, Ben Pfaff wrote:
> On Tue, Jul 21, 2015 at 12:59:15PM -0400, Russell Bryant wrote:
>> Add a new OVN configuration entry in the Open_vSwitch database called
>> "ovn-bridge-mappings". This allows the configuration of mappings
>> between a physical network name and an OVS brid
On Tue, Jul 21, 2015 at 12:59:16PM -0400, Russell Bryant wrote:
> While parsing the OVN bridge mapping configuration, ensure that patch
> ports exist between the OVN integration bridge and the physical
> network bridge. If they do not exist, create them automatically.
>
> Signed-off-by: Russell B
On 07/28/2015 03:55 PM, Ben Pfaff wrote:
> On Tue, Jul 21, 2015 at 12:59:14PM -0400, Russell Bryant wrote:
>> When getting the initial configuration, go ahead and store br_int in
>> the controller context. This is already done in the main loop, but a
>> future patch wants to use br_int from the co
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
On Tue, Jul 21, 2015 at 12:59:14PM -0400, Russell Bryant wrote:
> When getting the initial configuration, go ahead and store br_int in
> the controller context. This is already done in the main loop, but a
> future patch wants to use br_int from the controller context sooner in
> the code.
>
> Si
On Tue, Jul 28, 2015 at 12:24 PM, Russell Bryant wrote:
> On 07/28/2015 02:26 PM, Ben Pfaff wrote:
>> When ofctrl_add_flow() called ovn_flow_lookup(), the latter used the hash
>> from the flow's hmap_node, but the former hadn't initialized it at that
>> point. This commit fixes the problem.
>>
>>
On Tue, Jul 21, 2015 at 12:59:13PM -0400, Russell Bryant wrote:
> The logic in ovn-northd's parents_equal() and tags_equal() is useful
> for other columns, so convert them into more generic functions that
> can be reused.
>
> Signed-off-by: Russell Bryant
Applied to master, thanks!
_
On 07/28/2015 02:13 PM, Ben Pfaff wrote:
> On Tue, Jul 21, 2015 at 12:59:11PM -0400, Russell Bryant wrote:
>> Add a method to determine of two smaps are equal (have the exact same
>> set of key-value pairs).
>>
>> Signed-off-by: Russell Bryant
>
> Can't this be done in O(n) instead of O(n log n)?
On 07/28/2015 02:26 PM, Ben Pfaff wrote:
> When ofctrl_add_flow() called ovn_flow_lookup(), the latter used the hash
> from the flow's hmap_node, but the former hadn't initialized it at that
> point. This commit fixes the problem.
>
> Reported-by: Russell Bryant
> Reported-at: http://openvswitch
On Tue, Jul 28, 2015 at 8:44 AM, Ben Pfaff wrote:
> Until now, ovn-controller has been full of loops that commit a transaction
> to the OVS or OVN Southbound database. These blocking loops delay other
> work within ovn-controller. They also make it unsafe to keep pointers to
> database records
Hi,
Currently, the ovs netdev-dpdk ring type dpdkr, restricts dpdk ring name
size to 10 bytes (includes the _rx/tx suffix).
This only gives us space to accommodate 10 rings as truncation happens
with dpdkr10_rx/tx.
The enclosed patch below ensures that we respect the ring name size
constraints
This is interesting, thanks for the patch.
I'm definitely not a IVSHMEM expert, but I have a concern: what
happens if the secondary OVS process allocates or frees some
mbufs? (e.g because a packet is sent to multiple destinations or is
dropped)
I found this in the DPDK documentation:
http://dpdk.
On Tue, Jul 21, 2015 at 12:59:12PM -0400, Russell Bryant wrote:
> While running ovn-controller under valgrind, I noticed several
> variations of the same warning. Switching this allocation from malloc
> to zalloc resolves it.
>
> For reference, some examples from valgrind are:
>
> ==8487== Condi
When ofctrl_add_flow() called ovn_flow_lookup(), the latter used the hash
from the flow's hmap_node, but the former hadn't initialized it at that
point. This commit fixes the problem.
Reported-by: Russell Bryant
Reported-at: http://openvswitch.org/pipermail/dev/2015-July/057851.html
Signed-off-b
On Tue, Jul 21, 2015 at 12:59:11PM -0400, Russell Bryant wrote:
> Add a method to determine of two smaps are equal (have the exact same
> set of key-value pairs).
>
> Signed-off-by: Russell Bryant
Can't this be done in O(n) instead of O(n log n)?
bool
smap_equal(const struct smap *smap1, const
Acked-by: Daniele Di Proietto
Thanks!
On 28/07/2015 07:55, "Ilya Maximets" wrote:
>Currently pmd threads select queues in pmd_load_queues() according to
>get_n_pmd_threads_on_numa(). This behavior leads to race between pmds,
>beacause dp_netdev_set_pmds_on_numa() starts them one by one and
>cu
On Tue, Jul 28, 2015 at 10:54:30AM -0700, Alex Wang wrote:
> On Tue, Jul 28, 2015 at 8:44 AM, Ben Pfaff wrote:
>
> >
> > Signed-off-by: Ben Pfaff
> > Acked-by: Russell Bryant
> > ---
> > v1->v2: Call ovsdb_idl_run() before instead of after accessing IDL.
> > ---
> > ovn/controller/ovn-controll
On Tue, Jul 28, 2015 at 8:44 AM, Ben Pfaff wrote:
>
> Signed-off-by: Ben Pfaff
> Acked-by: Russell Bryant
> ---
> v1->v2: Call ovsdb_idl_run() before instead of after accessing IDL.
> ---
> ovn/controller/ovn-controller.c | 19 +--
> 1 file changed, 9 insertions(+), 10 deletion
On Tue, Jul 28, 2015 at 02:43:52PM -0300, Thadeu Lima de Souza Cascardo wrote:
> On Tue, Jul 28, 2015 at 10:22:47AM -0700, Ben Pfaff wrote:
> > On Tue, Jul 28, 2015 at 09:48:56AM -0300, Thadeu Lima de Souza Cascardo
> > wrote:
> > > On Mon, Jul 27, 2015 at 03:46:28PM -0700, Ben Pfaff wrote:
> > >
On Tue, Jul 28, 2015 at 10:22:47AM -0700, Ben Pfaff wrote:
> On Tue, Jul 28, 2015 at 09:48:56AM -0300, Thadeu Lima de Souza Cascardo wrote:
> > On Mon, Jul 27, 2015 at 03:46:28PM -0700, Ben Pfaff wrote:
> > > On Mon, Jul 27, 2015 at 02:24:19PM -0300, Thadeu Lima de Souza Cascardo
> > > wrote:
> >
On Tue, Jul 28, 2015 at 10:32:27AM -0700, Alex Wang wrote:
> When ovs cleans up unused vlan sub interfaces created by vlan splinter
> feature, it will not touch interface that has ip address assigned.
> However, the get_in6() function implemented in netdev-linux module
> is buggy and always returns
When ovs cleans up unused vlan sub interfaces created by vlan splinter
feature, it will not touch interface that has ip address assigned.
However, the get_in6() function implemented in netdev-linux module
is buggy and always returns 0 (found ipv6 address). This causes the
vlan splinter created sub
On Tue, Jul 28, 2015 at 09:48:56AM -0300, Thadeu Lima de Souza Cascardo wrote:
> On Mon, Jul 27, 2015 at 03:46:28PM -0700, Ben Pfaff wrote:
> > On Mon, Jul 27, 2015 at 02:24:19PM -0300, Thadeu Lima de Souza Cascardo
> > wrote:
> > > When devices are created, they are announced using RTM_IFANNOUNCE
On Tue, Jul 28, 2015 at 11:50:07AM -0400, Russell Bryant wrote:
> On 07/28/2015 11:44 AM, Ben Pfaff wrote:
> > As I was working in ovn-controller, I found it hard to tell what code
> > produced and what code consumed the OpenFlow flow table, because it was
> > all implicit. This commit makes the d
On Tue, Jul 28, 2015 at 9:30 AM, Alex Wang wrote:
> Thx for the review, adopt the suggestion,
>
> applied to branch2.3, 2.4, master~
>
When backporting to branch-2.3, found few conflicts... also, considering
that, get_in4,
is only used for vlan-splinter check... we should fix it the simple wa
OK, done.
On Tue, Jul 28, 2015 at 04:58:23PM +, Sorin Vinturis wrote:
> Ben, could you also apply this patch to the 2.4 branch?
>
> Thanks,
> Sorin
>
> -Original Message-
> From: Ben Pfaff [mailto:b...@nicira.com]
> Sent: Thursday, 16 July, 2015 21:43
> To: Sorin Vinturis
> Cc: dev@
Ben, could you also apply this patch to the 2.4 branch?
Thanks,
Sorin
-Original Message-
From: Ben Pfaff [mailto:b...@nicira.com]
Sent: Thursday, 16 July, 2015 21:43
To: Sorin Vinturis
Cc: dev@openvswitch.org
Subject: Re: [ovs-dev] [PATCH v2] datapath-windows: Fixed spelling errors in OV
Thx for the review, adopt the suggestion,
applied to branch2.3, 2.4, master~
On Tue, Jul 28, 2015 at 8:06 AM, Ben Pfaff wrote:
> On Mon, Jul 27, 2015 at 11:56:28PM -0700, Alex Wang wrote:
> > Currently the netdev_linux_notify_sock only joins multicast group
> > RTNLGRP_LINK for link status chan
On Tue, Jul 28, 2015 at 8:26 AM, Anoob Soman wrote:
> On 27/07/15 22:53, Jesse Gross wrote:
>> On Mon, Jul 27, 2015 at 9:59 AM, Anoob Soman
>> wrote:
>> #2: Add some mechanism where these daemons can talk to OVS and have
>> them cooperate to get the appropriate packets, while still keeping the
On 07/28/2015 11:44 AM, Ben Pfaff wrote:
> As I was working in ovn-controller, I found it hard to tell what code
> produced and what code consumed the OpenFlow flow table, because it was
> all implicit. This commit makes the data structure an explicit variable
> in the main loop, which makes it ea
Until now, if the integration bridge was missing, ovn-controller exited.
This commit makes it wait until the integration bridge is created.
Signed-off-by: Ben Pfaff
---
ovn/controller/binding.c| 4 +++-
ovn/controller/encaps.c | 2 +-
ovn/controller/ofctrl.c | 16 ++
On Tue, Jul 21, 2015 at 12:04:03PM -0700, Ethan Jackson wrote:
> High availability for gateways in network virtualization deployments
> is fairly difficult to get right. There are a ton of options, most of
> which are too complicated or perform badly. To help solve this
> problem, this patch prop
The other logical fields have their own macros, so the logical datapath
field might as well have one.
Signed-off-by: Ben Pfaff
---
ovn/controller/physical.c | 10 +-
ovn/controller/rule.h | 7 ---
2 files changed, 9 insertions(+), 8 deletions(-)
diff --git a/ovn/controller/phys
This won't really get used until the next commit.
Signed-off-by: Ben Pfaff
---
ovn/controller/ofctrl.c | 470
ovn/controller/ofctrl.h | 5 +-
ovn/controller/ovn-controller.c | 6 +-
ovn/controller/physical.h | 7 +
4 files chan
The OVN pipeline is being split into two phases, which are most naturally
called "pipelines". I kept getting very confused trying to call them
anything else, and in the end it seems to make more sense to just rename
the Pipeline table.
It would be even better to call this table Flow or Logical_Fl
When an upcoming commit divides the pipeline up into ingress and egress
pipeline, it will become necessary to resubmit to different tables from
each of those pipelines to implement output. This commit makes that
possible.
Signed-off-by: Ben Pfaff
---
ovn/controller/rule.c | 2 +-
ovn/lib/actio
An upcoming patch will add a Datapath_Binding table, so clarifying the
name seems useful.
Signed-off-by: Ben Pfaff
---
ovn/controller/binding.c | 28 ++---
ovn/controller/physical.c | 4 +-
ovn/controller/pipeline.c | 4 +-
ovn/northd/ovn-northd.c | 105 --
This drops an unused parameter and groups the calls to these functions
with ofctrl_destroy() in each case.
Signed-off-by: Ben Pfaff
---
ovn/controller/ovn-controller.c | 5 ++---
ovn/controller/pipeline.c | 2 +-
ovn/controller/pipeline.h | 2 +-
3 files changed, 4 insertions(+), 5 d
This will be used in documentation for an upcoming change, to document
how Geneve OVN options are encoded.
The code in this change is from a series (not yet submitted) that makes
much more extensive use of it for documenting protocol headers.
Signed-off-by: Ben Pfaff
---
python/build/nroff.py |
To be used in an upcoming commit.
Signed-off-by: Ben Pfaff
Acked-by: Russell Bryant
---
lib/smap.c | 13 -
lib/smap.h | 4 +++-
2 files changed, 15 insertions(+), 2 deletions(-)
diff --git a/lib/smap.c b/lib/smap.c
index 7fe3ce4..9badedd 100644
--- a/lib/smap.c
+++ b/lib/smap.c
@@
Until now, controller_ctx has been a store of common state (although
the amount of data stored in it has declined to just database state).
I think it's clearer if we just use it as a way to pass data to
functions. This commit makes that change.
Signed-off-by: Ben Pfaff
---
ovn/controller/ovn-co
Until now, if the chassis id was missing, ovn-controller exited. This
commit makes ovn-controller wait for it to return.
Signed-off-by: Ben Pfaff
---
ovn/controller/binding.c| 3 +++
ovn/controller/chassis.c| 4
ovn/controller/ovn-controller.c | 38 +--
Until now, if external-ids:ovn-bridge changed, ovn-controller ignored
the change. With this commit, ovn-controller uses the new bridge.
Signed-off-by: Ben Pfaff
---
ovn/controller/ovn-controller.c | 81 +
1 file changed, 33 insertions(+), 48 deletions(-)
The generic *_init() names for these functions made it sounds like they
do something more than just register IDL columns, even though that's all
they do. Also, the controller_ctx that was passed into each of them was
only used to get the IDL handle. This commit renames them and changes
their para
I found it hard otherwise to see what code depended on this.
Signed-off-by: Ben Pfaff
Acked-by: Russell Bryant
---
ovn/controller/binding.c| 12 ++--
ovn/controller/binding.h| 3 ++-
ovn/controller/chassis.c| 22 +++---
ovn/controller/chassis.h
I found it otherwise difficult to see what code depended on this.
Signed-off-by: Ben Pfaff
---
ovn/controller/binding.c| 13 +++--
ovn/controller/binding.h| 5 +++--
ovn/controller/chassis.c| 14 +++---
ovn/controller/chassis.h| 4 ++--
ovn/contr
Until now, ovn-controller has been full of loops that commit a transaction
to the OVS or OVN Southbound database. These blocking loops delay other
work within ovn-controller. They also make it unsafe to keep pointers to
database records within a single ovn-controller main loop, since calls
to ovs
These two pieces of code have different requirements and hardly anything in
common, and I found it easier to understand and reasonably modify the
overall structure of the program when they were separated.
Signed-off-by: Ben Pfaff
---
ovn/controller/automake.mk | 2 +
ovn/controller
The IDLs as created by ovn-controller always retry failed connections,
which means that ovsdb_idl_is_alive() will always report that they are
alive.
Signed-off-by: Ben Pfaff
Acked-by: Russell Bryant
---
ovn/controller/ovn-controller.c | 16
1 file changed, 16 deletions(-)
diff
As I was working in ovn-controller, I found it hard to tell what code
produced and what code consumed the OpenFlow flow table, because it was
all implicit. This commit makes the data structure an explicit variable
in the main loop, which makes it easier to see.
Signed-off-by: Ben Pfaff
---
ovn/
This is mostly a rebase versus v1. A few patches have been acked,
and patch 1 includes a bug fix.
Ben Pfaff (21):
ovn-controller: Fix potential use-after-free in get_core_config().
ovn-controller: Drop unnecessary checks for ovsdb_idl_is_alive().
ovn-controller: Avoid blocking to commit OVS
It's unsafe to hold a pointer to a row in the IDL across calls to
ovsdb_idl_run() for that IDL.
Signed-off-by: Ben Pfaff
Acked-by: Russell Bryant
---
v1->v2: Call ovsdb_idl_run() before instead of after accessing IDL.
---
ovn/controller/ovn-controller.c | 19 +--
1 file changed,
On Wed, Jul 22, 2015 at 05:17:22PM +, Chandran, Sugesh wrote:
> I am just trying out this new patches in my OVN setup.
> This setup has two logical network span across two nodes/Chassis.
> I found two issues with the new patch set as below
> 1) The controller cannot start properly due to the
On Sat, Jul 25, 2015 at 11:31:48PM -0700, Alex Wang wrote:
> Also, could we add 'commit' to function name like
> idl_loop_commit_and_wait()? I just assumed that the idl_loop_wait() just
> waits... but it actually commits as well,
OK, I'll make that change for v2.
On Wed, Jul 22, 2015 at 04:26:48PM -0400, Russell Bryant wrote:
> On 07/19/2015 06:44 PM, Ben Pfaff wrote:
> > Until now, ovn-controller has been full of loops that commit a transaction
> > to the OVS or OVN Southbound database. These blocking loops delay other
> > work within ovn-controller. The
On Sat, Jul 25, 2015 at 11:04:13PM -0700, Alex Wang wrote:
> On Sun, Jul 19, 2015 at 3:44 PM, Ben Pfaff wrote:
>
> > It's unsafe to hold a pointer to a row in the IDL across calls to
> > ovsdb_idl_run() for that IDL.
> >
> > Signed-off-by: Ben Pfaff
> > ---
> > ovn/controller/ovn-controller.c |
1 - 100 of 122 matches
Mail list logo
