Add elements to enum ofp_table_config for table mod bits
defined in OpenFlow1.4.
Also tidy up the indentation of existing elements.
Signed-off-by: Simon Horman
---
include/openflow/openflow-common.h | 15 +++
1 file changed, 11 insertions(+), 4 deletions(-)
diff --git a/include/ope
Last patch got sent word wrapped, trying again.
This change allows vlog to export to a specified local udp syslog sink.
Signed-off-by: Henry Mai
---
NEWS | 1 +
lib/vlog.c| 98 +--
lib/vlog.h| 26 ++
The skb_priority, pkt_mark and tunl parameters dp_netdev_execute_actions
are always passed as 0, 0 and NULL respectively. So rather than
passing these values to dp_netdev_execute_actions just use them directly.
Signed-off-by: Simon Horman
---
lib/dpif-netdev.c | 10 --
1 file changed, 4
On Tue, Nov 26, 2013 at 03:46:23PM +0900, Simon Horman wrote:
> On Mon, Nov 25, 2013 at 05:24:23PM -0800, Ben Pfaff wrote:
> > On Mon, Nov 25, 2013 at 05:22:47PM -0800, Jesse Gross wrote:
> > > On Mon, Nov 25, 2013 at 4:04 PM, Simon Horman wrote:
> > > > On Mon, Nov 25, 2013 at 10:09:30AM -0800, B
This change allows vlog to export to a specified local udp syslog sink.
Signed-off-by: Henry Mai
---
NEWS | 1 +
lib/vlog.c| 98
+--
lib/vlog.h| 26 ++---
utilities/ovs-appctl.8.in | 6 +++
On Mon, Nov 25, 2013 at 05:24:23PM -0800, Ben Pfaff wrote:
> On Mon, Nov 25, 2013 at 05:22:47PM -0800, Jesse Gross wrote:
> > On Mon, Nov 25, 2013 at 4:04 PM, Simon Horman wrote:
> > > On Mon, Nov 25, 2013 at 10:09:30AM -0800, Ben Pfaff wrote:
> > >> On Mon, Nov 25, 2013 at 10:09:55PM +0900, Simon
Since you're OK with manual updates, I'm happy in principle with having
IDE-related files in the repository as long as they are not unreasonably
large. But there's something weird going on. Why would special files
would be needed for syntax highlighting or Git integration or even
integrated debug
Of course I do. :-)
A CI gate might be very helpful for this purpose as a further step to keep
those files aligned by avoiding regressions IMO but for the time being we'd
perfectly fine with manual updates.
> On 26/nov/2013, at 03:19, "Ben Pfaff" wrote:
>
> You realize that no one else is goi
On Mon, Nov 25, 2013 at 05:22:47PM -0800, Jesse Gross wrote:
> On Mon, Nov 25, 2013 at 4:04 PM, Simon Horman wrote:
> > On Mon, Nov 25, 2013 at 10:09:30AM -0800, Ben Pfaff wrote:
> >> On Mon, Nov 25, 2013 at 10:09:55PM +0900, Simon Horman wrote:
> >> > as far as I can tell no one is actively worki
On Mon, Nov 25, 2013 at 4:04 PM, Simon Horman wrote:
> On Mon, Nov 25, 2013 at 10:09:30AM -0800, Ben Pfaff wrote:
>> On Mon, Nov 25, 2013 at 10:09:55PM +0900, Simon Horman wrote:
>> > as far as I can tell no one is actively working on the following item in
>> > OPENFLOW-1.1+. So I have made a star
You realize that no one else is going to update them, right?
On Tue, Nov 26, 2013 at 12:03:02AM +, Alessandro Pilotti wrote:
> What if we simply add a folder with the Visual Studio build files to begin
> with?
>
> > On 26/nov/2013, at 01:29, "Ben Pfaff" wrote:
> >
> > We're not switching t
On Mon, Nov 25, 2013 at 10:09:30AM -0800, Ben Pfaff wrote:
> On Mon, Nov 25, 2013 at 10:09:55PM +0900, Simon Horman wrote:
> > as far as I can tell no one is actively working on the following item in
> > OPENFLOW-1.1+. So I have made a start it.
> >
> > * The new in_phy_port field in OFPT_PACK
What if we simply add a folder with the Visual Studio build files to begin with?
> On 26/nov/2013, at 01:29, "Ben Pfaff" wrote:
>
> We're not switching to CMake. If you have something to generate the
> XML files you need, we'll check that in.
>
>> On Mon, Nov 25, 2013 at 11:23:37PM +, Ales
On Thu, Nov 21, 2013 at 10:15 AM, Thomas Graf wrote:
> diff --git a/lib/dpif-linux.c b/lib/dpif-linux.c
> index 1dcf321..5c952bc 100644
> --- a/lib/dpif-linux.c
> +++ b/lib/dpif-linux.c
> @@ -232,6 +233,7 @@ dpif_linux_open(const struct dpif_class *class
> OVS_UNUSED, const char *name,
>
On Fri, Nov 22, 2013 at 8:56 AM, Thomas Graf wrote:
> diff --git a/include/uapi/linux/openvswitch.h
> b/include/uapi/linux/openvswitch.h
> index 07ef2c3..a6863e2 100644
> --- a/include/uapi/linux/openvswitch.h
> +++ b/include/uapi/linux/openvswitch.h
> @@ -40,7 +40,15 @@ struct ovs_header {
>
>
We're not switching to CMake. If you have something to generate the
XML files you need, we'll check that in.
On Mon, Nov 25, 2013 at 11:23:37PM +, Alessandro Pilotti wrote:
> Visual Studio is the "de facto" IDE for Windows development. It provides all
> the features you'd expect from a moder
On Tue, Nov 19, 2013 at 04:43:47PM -0800, Jarno Rajahalme wrote:
> Allow TCP flags match specification with symbolic flag names. TCP
> flags are optionally specified as a string of flag names, each
> preceded by '+' when the flag must be one, or '-' when the flag must
> be zero. Any flags not exp
Visual Studio is the "de facto" IDE for Windows development. It provides all
the features you'd expect from a modern environment (integrated debugger,
refactoring tools, Git integration, syntax highlighting and a gazillion
additional features) and in general it allows to be a few orders of magni
This patch series fixes the test breakage encountered with the tunnel
scalability series. Thanks!
On 25 November 2013 14:26, Alex Wang wrote:
> Commit (9cdc68a1 bfd: Add forwarding flag to struct bfd.) allows
> several functions to call bfd_forwarding__() and update the
> forwarding flag. When f
On Thu, Nov 21, 2013 at 07:16:54PM +0100, Thomas Graf wrote:
> Based on the initial patch by Cong Wang posted a couple of months
> ago.
>
> This is the user space counterpart needed for the kernel patch
> '[PATCH net-next 3/8] openvswitch: Enable memory mapped Netlink i/o'
>
> Allows the kernel t
The patch series posted below fixes the cause of these test failures:-
http://openvswitch.org/pipermail/dev/2013-November/034359.html
On 25 November 2013 13:44, Joe Stringer wrote:
> Currently, as part of ofproto-dpif run() processing, we loop through all
> ports and poll corresponding devices f
Commit (9cdc68a1 bfd: Add forwarding flag to struct bfd.) allows
several functions to call bfd_forwarding__() and update the
forwarding flag. When forwarding_if_rx feature is enabled, this
introduces a race condition among threads calling these functions
and trying to update the flag. And this ma
This commit refines the unit tests for forwarding_if_rx feature
and flap_count. Comments are added to explain each test case.
Signed-off-by: Alex Wang
---
tests/bfd.at | 84 ++
1 file changed, 32 insertions(+), 52 deletions(-)
diff --gi
Sure :).
Signed-off-by: Alin Serdean
---
CodingStyle |2 +-
Makefile.am | 13
lib/bundle.c |4 +-
lib/coverage.c|2 +-
lib/hmap.c|2 +-
lib/multipath.c |2 +-
Sent from my iPhone
> On Nov 25, 2013, at 5:05 PM, Jesse Gross wrote:
>
> They're the equivalent of makefiles for Visual Studio. Without them
> you can't use the Windows-native development tools so while it's not
> impossible to work it certainly makes life more difficult.
One can still edit
Thanks for the details. I found the bug and posted a fix:
http://openvswitch.org/pipermail/dev/2013-November/034354.html
On Fri, Nov 22, 2013 at 11:44:41AM +0800, ZhengLingyun wrote:
> Oh, sorry. I updated the code from master a few days ago.
>
> Now I updae the code again. The OVS works
These would have found the problem fixed in the previous patch.
Signed-off-by: Ben Pfaff
---
ofproto/connmgr.c |3 ++-
ofproto/fail-open.c |3 +++
ofproto/fail-open.h |6 +++---
3 files changed, 8 insertions(+), 4 deletions(-)
diff --git a/ofproto/connmgr.c b/ofproto/connmgr.c
ind
If one configured a controller which does not exist, waited for the switch
to enter fail-open mode, and then deleted the controller, then
ofproto_set_controllers() would take ofproto_mutex and call
update_fail_open(), which would call fail_open_destroy(), which would call
fail_open_recover(), which
Coming from linux, all I looked for while testing was whether I can create
executables that can be run and debugged. I could do both. (For debugging, I
had to add an additional option to cccl to create the symbol file).
Alessandro,
As Ben mentioned, please explain what is that the project file
They're the equivalent of makefiles for Visual Studio. Without them
you can't use the Windows-native development tools so while it's not
impossible to work it certainly makes life more difficult.
On Mon, Nov 25, 2013 at 1:53 PM, Ben Pfaff wrote:
> What I'm trying to get at is, what are the "solut
Yeah I don't really know, everyone who devs on window seems to think
it matters, but I haven't had any experience with it personally.
Ethan
On Mon, Nov 25, 2013 at 4:53 PM, Ben Pfaff wrote:
> What I'm trying to get at is, what are the "solution and related
> projects" good for? The non-Windows
What I'm trying to get at is, what are the "solution and related
projects" good for? The non-Windows world does fine without them, so
if "make" can work on Windows then why is the result "basically
useless for any practical development purpose"?
On Mon, Nov 25, 2013 at 04:50:40PM -0500, Ethan Jac
My understanding is that Guru is working on a solution to this
problem. What were your thoughts?
Ethan
On Mon, Nov 25, 2013 at 3:52 PM, Ben Pfaff wrote:
> On Mon, Nov 25, 2013 at 08:11:00PM +, Alessandro Pilotti wrote:
>> We did some testing with autoconf / automake on Windows. Makefiles
>>
Currently, we track status changes for ofports on a per-device basis. Each time
in the main thread's loop, we loop through all ofports and manually check
whether the status has changed for corresponding devices. This check for
changes is quite expensive, even if we don't need to update ofports.
Th
Currently, we iterat through all interfaces in instant_stats_run(), grabbing
up-to-date information about device status. After assembling all of this
information for all interfaces, we determine whether anything changed and only
send an update to ovsdb-server if something changed.
This patch uses
Currently, we refresh STP status (id, state, role) alongside statistics (rx,
tx, errors), all within instant_stats_run(). This patch splits statistics out,
and refreshes them with the 5 second stats instead. This paves the way to
reducing execution of instant_stats_run().
Signed-off-by: Joe String
Currently, as part of ofproto-dpif run() processing, we loop through all
ports and poll corresponding devices for changes in carrier, cfm and bfd
status. This allows us to determine how it may affect bundles, and cause
revalidation on bridges when necessary. For the average case where devices are
n
When configured with a large number of ports, ovs-vswitchd has ongoing high CPU
utilization even when there is little going on; For example, configuring
current master with more than 5,000 tunnel ports running bfd causes the average
CPU utilization of the main thread to hit 100% in a test environme
When dealing with a large number of ports, bundle_run() and bundle_wait() add
significant unnecessary processing to the main run loop, even when there are no
bonds and lacp is not configured. This patch skips such execution if it is
unneeded, reducing average CPU usage from around 25%->20% in a tes
On Mon, Nov 25, 2013 at 02:15:49PM +0800, Alexander Wu wrote:
> On 22/11/2013 06:19, Ben Pfaff wrote:
> >On Thu, Nov 21, 2013 at 05:04:29PM +0800, Alexander Wu wrote:
> >>V3:
> >> 1. Update names of functions/macros to make them meaningful.
> >> 2. Fix codingstyle.
> >> 3. Remove useless logi
On Fri, Nov 22, 2013 at 8:56 AM, Thomas Graf wrote:
> diff --git a/net/openvswitch/datapath.c b/net/openvswitch/datapath.c
> index 95d4424..3f1fb87 100644
> --- a/net/openvswitch/datapath.c
> +++ b/net/openvswitch/datapath.c
I'm a little worried that this introduces some quirks to the interface.
On Mon, Nov 25, 2013 at 12:41:33PM +0900, Simon Horman wrote:
> Include stddef.h in include/sparse/sys/socket.h to ensure
> that NULL is defined and thus avoid the following sparse warning.
>
> ./include/sparse/sys/socket.h:74:15: error: undefined identifier 'NULL'
>
> Signed-off-by: Simon Horman
On Mon, Nov 25, 2013 at 10:17:01AM +0100, Martino Fornasa wrote:
> One of the main differences between STP and RSTP is that RSTP reacts
> quickly to link failures by checking the link physical status,
> without having to rely on long timeout expirations. So, we need an
> entry point from which we c
On Sun, Nov 24, 2013 at 09:39:12PM +, Alin Serdean wrote:
> Bellow is the patch for the %z modifier.
>
> Please take a closer look over the Makefile and comments (I hope they are
> inline with your previous ones :) ).
That's much better, thanks.
Can you provide a Signed-off-by line? Submi
On Mon, Nov 25, 2013 at 08:11:00PM +, Alessandro Pilotti wrote:
> We did some testing with autoconf / automake on Windows. Makefiles
> are getting generated correctly although we cannot obviously verify
> the result with a full build since we didn?t port all the patches to
> the master branch y
LGTM, only one thing,
I did the following test:
"""
root@server329:~/ben-reviews# ovs-vsctl --version
ovs-vsctl (Open vSwitch) 2.0.90
Compiled Nov 25 2013 11:44:29
root@server329:~/ben-reviews# lsof -p `pidof ovs-vswitchd` | wc -l
31
*root@server329:~/ben-reviews# ovs-vsctl add-br test*
root@ser
Hi guys,
We did some testing with autoconf / automake on Windows. Makefiles are getting
generated correctly although we cannot obviously verify the result with a full
build since we didn’t port all the patches to the master branch yet.
There’s anyway a huge limitation: it does not generate a Vis
Signed-off-by: Ben Pfaff
---
FAQ | 44
1 file changed, 44 insertions(+)
diff --git a/FAQ b/FAQ
index 2912ae3..df7b6ef 100644
--- a/FAQ
+++ b/FAQ
@@ -935,6 +935,50 @@ A: Yes. Use an "internal port" configured as an access
port. For
ovs-vsc
LGTM, thx
On Sat, Nov 2, 2013 at 10:23 PM, Ben Pfaff wrote:
> These helpers simplify new code to be added in an upcoming commit.
>
> Signed-off-by: Ben Pfaff
> ---
> lib/dpif-linux.c | 46 +-
> 1 file changed, 33 insertions(+), 13 deletions(-)
>
>
LGTM, one comment below,
> -if ((dpif->epoll_fd >= 0) == enable) {
> -return 0;
> -}
> -
> -if (!enable) {
> -destroy_channels(dpif);
> -} else {
> -struct dpif_port_dump port_dump;
> -struct dpif_port port;
> -
> +/* To start with, we need an
On Mon, Nov 25, 2013 at 02:16:23PM +0800, Alexander Wu wrote:
> On 22/11/2013 06:21, Ben Pfaff wrote:
> >On Thu, Nov 21, 2013 at 05:04:30PM +0800, Alexander Wu wrote:
> >>V3:
> >> Add description for table-features get/set struct.
> >>
> >>V2:
> >> No change.
> >>
> >>V1:
> >> Add table_featu
On Mon, Nov 25, 2013 at 10:09:55PM +0900, Simon Horman wrote:
> as far as I can tell no one is actively working on the following item in
> OPENFLOW-1.1+. So I have made a start it.
>
> * The new in_phy_port field in OFPT_PACKET_IN needs some kind of
> implementation. It has a sensible i
Add two new tables to the VTEP schema in support of distributed L3.
Each table contains MAC addresses to be used by VTEPs (both hardware
and software) when issuing ARP requests on behalf of a logical router.
Signed-off-by: Bruce Davie
---
vtep/vtep.ovsschema | 18 --
vtep/vtep.xm
Dear Gold Buyer
we are suppliers of Gold Dory Bars/ Gold Dust and we mine our Au Gold
from our family mine in Cote D'lviore. We supply top quality Au Gold to
any interested Buyer. Our procedures are simple and we work close with
our buyer to create long term business relationship.
Product Speci
Hi,
as far as I can tell no one is actively working on the following item in
OPENFLOW-1.1+. So I have made a start it.
* The new in_phy_port field in OFPT_PACKET_IN needs some kind of
implementation. It has a sensible interpretation for tunnels
but in general the physical port is
Ben Pfaff wrote:
Yes, this makes sense. Please feel free to ask any questions.
Hi.
One of the main differences between STP and RSTP is that RSTP reacts
quickly to link failures by checking the link physical status, without
having to rely on long timeout expirations. So, we need an entry poi
56 matches
Mail list logo
