Thanks,
Daniele
On Jul 21, 2014, at 9:43 PM, Ben Pfaff wrote:
> On Mon, Jul 21, 2014 at 05:11:49PM -0700, Daniele Di Proietto wrote:
>> Signed-off-by: Daniele Di Proietto
>> ---
>> I've noticed a testcase (ofproto-dpif - controller, 758) failure on master.
>> I believe that the problem is in t
On Mon, Jul 21, 2014 at 5:11 PM, Ansis Atteka wrote:
> On Thu, Jul 17, 2014 at 4:08 PM, Jesse Gross wrote:
>> One thing that I worry about is that this has the possibility to
>> change how the flow is reported - before the flow deletion it has
>> Geneve options but immediately after the flow stil
After some pondering, I've concluded that the initial Windows datapath
should use Netlink for userspace-kernel datapath communication,
primarily because Netlink is extensible in a forward- and
backward-compatible fashion. That is, as we add new features to the
kernel module, existing userspace con
>On Fri, Jul 18, 2014 at 5:27 PM, Saurabh Shah wrote:
>> Ben Pfaff writes:
>> > The logging in dpif_windows_flow_del() seems like a debugging stray.
>>
>> Do you mean in do_put?
>
>I mean the one here:
>
>+static int
>+dpif_windows_flow_del(struct dpif *dpif, const struct dpif_flow_del *del)
>+{
On Thu, Jul 17, 2014 at 07:33:03PM +, Alin Serdean wrote:
> Add defines needed to compile netlink-socket.c and netlink.c.
>
> Add wrappers for the functions sendmsg and recvmsg. They can also be added to
> another file instead of defining them inlined.
>
> Add a wrapper and the functionality
On Thu, Jul 17, 2014 at 07:33:43PM +, Alin Serdean wrote:
> For our kernel extension we do not take into consideration the padding
> thus remove it.
>
> Add MAX_LINKS define needed for nl_pool.
>
> Add NETLINK_ROUTE needed to compile rtnetlink-link.
>
> Signed-off-by: Alin Gabriel Serdean
On Thu, Jul 17, 2014 at 07:34:14PM +, Alin Serdean wrote:
> Bypass the error compilation when compiling under MSVC.
>
> Signed-off-by: Alin Gabriel Serdean
This seems fine, thanks.
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mai
On Thu, Jul 17, 2014 at 07:34:51PM +, Alin Serdean wrote:
> Add the structure sockaddr_nl to netlink-socket.c when compiling under
> MSVC.
>
> Add two functions set_sock_pid_in_kernel and portid_next. This will allow
> the channel identification for the kernel extension to send back messages.
On Thu, Jul 17, 2014 at 07:35:26PM +, Alin Serdean wrote:
> Add netlink related files to the windows build.
>
>
> Signed-off-by: Alin Gabriel Serdean
Seems fine, thanks.
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listi
Debian puts an extra "-Wformat" in the CFLAGS following OVS's own
"-Wformat -Wno-format-zero-length", which therefore overrides
-Wno-format-zero-length, so this commit adds an extra
-Wno-format-zero-length to avoid those false positives.
Signed-off-by: Ben Pfaff
---
debian/rules |7 +++
Hi folks,
Currently the branch-2.1 has 40 commits since the last 2.1.2 tag.
Do we have a schedule to release a new version?
Thanks,
fbl
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
On Mon, Jul 21, 2014 at 8:35 PM, Wenyu Zhang wrote:
> Thanks for the review, my comments inline.
>
> Bests,
> Wenyu
> -Original Message-
> From: Pravin Shelar [mailto:pshe...@nicira.com]
> Sent: Tuesday, July 22, 2014 6:56 AM
> To: Wenyu Zhang
> Cc: Romain Lenglet; Ben Pfaff; Jesse Gross;
On Tue, Jul 22, 2014 at 12:55 PM, Flavio Leitner wrote:
>
> Currently the branch-2.1 has 40 commits since the last 2.1.2 tag.
> Do we have a schedule to release a new version?
>
Since our last LTS is so old, let's release 2.1.3 along with 2.3.0. The
2.3.0 release is still in QA, but I expect we
On Tue, Jul 22, 2014 at 01:13:57PM -0700, Justin Pettit wrote:
> On Tue, Jul 22, 2014 at 12:55 PM, Flavio Leitner wrote:
>
> >
> > Currently the branch-2.1 has 40 commits since the last 2.1.2 tag.
> > Do we have a schedule to release a new version?
>
> Since our last LTS is so old, let's release
Hi,
RHEL-7 kernel package comes with openvswitch module ready, so why are you
compiling the external module?
I am willing to fix the issue anyway, so feel free to open a bugzilla
at http://bugzilla.redhat.com.
Thanks,
fbl
On Fri, Jul 18, 2014 at 11:26:02AM -0700, Jesse Gross wrote:
> You will l
I have seen multiple coding style errors and warning from kernel
checkpatch.pl on kernel part of the patch:-
WARNING: Missing a blank line after declarations
#88: FILE: datapath/actions.c:530:
+ struct vport *vport;
+ vport = ovs_vport_ovsl_rcu(dp, nla_get_u32(a));
ERROR: space required before th
Thanks for the reply. I have already found a workaround.
JZ
> Date: Tue, 22 Jul 2014 18:29:33 -0300
> From: f...@redhat.com
> To: jgzh...@hotmail.com
> CC: dev@openvswitch.org; je...@nicira.com
> Subject: Re: issue on building openvswitch 2.1.2 on RHEL 7.0
>
> Hi,
>
> RHEL-7 kernel package come
On Mon, Jul 21, 2014 at 1:39 PM, Andy Zhou wrote:
> o get_dp_rcu() requires its caller to hold rcu lock.
> o get_dp() requires its caller to hold either ovs lock or
> rcu lock.
Can you add why we are doing this change to the commit msg.
> Signed-off-by: Andy
On Tue, Jul 22, 2014 at 02:32:16PM -0700, JIANGANG Zhang wrote:
> Thanks for the reply. I have already found a workaround.
Could you share the workaround?
> > Date: Tue, 22 Jul 2014 18:29:33 -0300
> > From: f...@redhat.com
> > To: jgzh...@hotmail.com
> > CC: dev@openvswitch.org; je...@nicira.com
Hi Ben,
On 22 Jul 2014, at 20:10, Ben Pfaff wrote:
> After some pondering, I've concluded that the initial Windows datapath
> should use Netlink for userspace-kernel datapath communication,
> primarily because Netlink is extensible in a forward- and
> backward-compatible fashion. That is, as we
Signed-off-by: Ben Pfaff
---
lib/daemon-windows.c |2 +-
lib/stream-tcp.c |4 ++--
lib/unixctl.c|6 +++---
tests/test-odp.c |2 +-
tests/test-vconn.c |2 +-
5 files changed, 8 insertions(+), 8 deletions(-)
diff --git a/lib/daemon-windows.c b/lib/daemon-windo
Acked-by: Justin Pettit
On Tue, Jul 22, 2014 at 3:48 PM, Ben Pfaff wrote:
> Signed-off-by: Ben Pfaff
> ---
> lib/daemon-windows.c |2 +-
> lib/stream-tcp.c |4 ++--
> lib/unixctl.c|6 +++---
> tests/test-odp.c |2 +-
> tests/test-vconn.c |2 +-
> 5 file
Thanks. Pushed with changes suggested.
On Tue, Jul 22, 2014 at 3:00 PM, Pravin Shelar wrote:
> On Mon, Jul 21, 2014 at 1:39 PM, Andy Zhou wrote:
>> o get_dp_rcu() requires its caller to hold rcu lock.
>> o get_dp() requires its caller to hold either ovs lock or
>>
Acked-by: Justin Pettit
On Tue, Jul 22, 2014 at 11:50 AM, Ben Pfaff wrote:
> Debian puts an extra "-Wformat" in the CFLAGS following OVS's own
> "-Wformat -Wno-format-zero-length", which therefore overrides
> -Wno-format-zero-length, so this commit adds an extra
> -Wno-format-zero-length to av
This matches the option offered by some other Open vSwitch daemons. I
intend to use it in tests in an upcoming commit.
Signed-off-by: Ben Pfaff
---
utilities/ovs-ofctl.c | 13 -
1 file changed, 12 insertions(+), 1 deletion(-)
diff --git a/utilities/ovs-ofctl.c b/utilities/ovs-ofc
Service controllers can set their roles, so it's necessary to demote them
to slaves if another controller becomes master. Unfortunately the
'controllers' hmap only contains primary controllers, so this was omitted.
Signed-off-by: Ben Pfaff
---
ofproto/connmgr.c |2 +-
1 file changed, 1 inse
Only OpenFlow 1.4 and later support role status messages, but this code
tried to send them to all controllers, which caused an assertion failure.
Also, add tests to check that role status messages work, and that they
don't cause trouble with OF1.2.
Reported-by: Anup Khadka
Signed-off-by: Ben Pfa
Thanks, applied.
On Tue, Jul 22, 2014 at 03:51:33PM -0700, Justin Pettit wrote:
> Acked-by: Justin Pettit
>
>
>
> On Tue, Jul 22, 2014 at 3:48 PM, Ben Pfaff wrote:
>
> > Signed-off-by: Ben Pfaff
> > ---
> > lib/daemon-windows.c |2 +-
> > lib/stream-tcp.c |4 ++--
> > lib/unixc
Thanks, applied to master.
On Tue, Jul 22, 2014 at 03:54:47PM -0700, Justin Pettit wrote:
> Acked-by: Justin Pettit
>
>
> On Tue, Jul 22, 2014 at 11:50 AM, Ben Pfaff wrote:
>
> > Debian puts an extra "-Wformat" in the CFLAGS following OVS's own
> > "-Wformat -Wno-format-zero-length", which th
We've had a couple of questions about this lately, including one suggestion
that we should always save the OpenFlow port numbers. This explains why
the behavior is as it is.
Signed-off-by: Ben Pfaff
---
utilities/ovs-ctl.in |7 +--
1 file changed, 5 insertions(+), 2 deletions(-)
diff -
Acked-by: Justin Pettit
On Tue, Jul 22, 2014 at 4:09 PM, Ben Pfaff wrote:
> We've had a couple of questions about this lately, including one suggestion
> that we should always save the OpenFlow port numbers. This explains why
> the behavior is as it is.
>
> Signed-off-by: Ben Pfaff
> ---
>
Acked-by: Justin Pettit
On Mon, Jul 21, 2014 at 5:09 PM, Ben Pfaff wrote:
> Reported-by: Madhu Venugopal
> Signed-off-by: Ben Pfaff
> ---
> ovsdb/ovsdb-server.1.in | 16
> 1 file changed, 16 insertions(+)
>
> diff --git a/ovsdb/ovsdb-server.1.in b/ovsdb/ovsdb-server.1.in
Thanks, applied to master.
On Tue, Jul 22, 2014 at 04:11:42PM -0700, Justin Pettit wrote:
> Acked-by: Justin Pettit
>
>
> On Tue, Jul 22, 2014 at 4:09 PM, Ben Pfaff wrote:
>
> > We've had a couple of questions about this lately, including one suggestion
> > that we should always save the Open
My comments inline.
发自我的 iPhone
在 2014-7-23，4:02，"Pravin Shelar" 写道：
> On Mon, Jul 21, 2014 at 8:35 PM, Wenyu Zhang wrote:
>> Thanks for the review, my comments inline.
>>
>> Bests,
>> Wenyu
>> -Original Message-
>> From: Pravin Shelar [mailto:pshe...@nicira.com]
>> Sent: Tuesday, Jul
Thanks, applied to master.
On Tue, Jul 22, 2014 at 04:14:51PM -0700, Justin Pettit wrote:
> Acked-by: Justin Pettit
>
>
> On Mon, Jul 21, 2014 at 5:09 PM, Ben Pfaff wrote:
>
> > Reported-by: Madhu Venugopal
> > Signed-off-by: Ben Pfaff
> > ---
> > ovsdb/ovsdb-server.1.in | 16 +++
Thanks for the checking.
I will correct them. Do you know any similar script for user space codes?
Thanks a lot.
Wenyu
在 2014-7-23，5:29，"Pravin Shelar" 写道：
> I have seen multiple coding style errors and warning from kernel
> checkpatch.pl on kernel part of the patch:-
>
> WARNING: Missing a
Commit 825da1c6d1c7 (leak-checker: Remove because it cannot be made
thread-safe.) removed the only uses of these hooks but neglected to remove
the test for them.
Signed-off-by: Ben Pfaff
---
configure.ac |1 -
m4/openvswitch.m4 | 20
2 files changed, 21 deletions(
Thanks, applied to master.
On Tue, Jul 22, 2014 at 04:25:12PM -0700, Justin Pettit wrote:
> Acked-by: Justin Pettit
>
>
>
> On Tue, Jul 22, 2014 at 4:21 PM, Ben Pfaff wrote:
>
> > Commit 825da1c6d1c7 (leak-checker: Remove because it cannot be made
> > thread-safe.) removed the only uses of t
Acked-by: Justin Pettit
On Tue, Jul 22, 2014 at 4:21 PM, Ben Pfaff wrote:
> Commit 825da1c6d1c7 (leak-checker: Remove because it cannot be made
> thread-safe.) removed the only uses of these hooks but neglected to remove
> the test for them.
>
> Signed-off-by: Ben Pfaff
> ---
> configure.ac
Thanks, for the fix!!! applied to master
Acked-by: Alex Wang
On Mon, Jul 21, 2014 at 4:00 PM, Ben Pfaff wrote:
> v2 seems fine to me too.
>
> I'll leave it to Alex to apply this in case he has further comments.
>
> On Mon, Jul 21, 2014 at 03:48:14PM -0700, Daniele Di Proietto wrote:
> > Thank
On Tue, Jul 22, 2014 at 4:19 PM, Wenyu Zhang wrote:
> Thanks for the checking.
>
> I will correct them. Do you know any similar script for user space codes?
>
I don't think there is such script for userspace.
___
dev mailing list
dev@openvswitch.org
htt
>
> This is not safe against simultaneous calls for the same port, but it
> would be if the assignment to 'old' happened with the lock held:
> > +void
> > +stp_port_set_name(struct stp_port *p, const char *name)
> > +{
> > +char *old = p->port_name;
> > +
> > +ovs_mutex_lock(&mutex);
> > +
Signed-off-by: Ben Pfaff
---
ovs-vswitchd typically runs as root, but it doesn't need most of the powers
of root. This commit makes it drop all the capabilities that it doesn't
typically need.
I don't know how capabilities are really used in practice. Something
equivalent could be achieved thro
This series introduces in dpif-netdev an exact match cache for frequently
used flows. It uses the RSS hash (currently implemented only for netdev-dpdk)
to search the miniflow in an hash table.
While there might be still some tuning left to do, these patch provides
significant performance improveme
struct dp_netdev_flow used to have a reference counter. It has been replaced by
RCU. Unfortunately RCU is not enough if we plan to hold a reference to the
dp_netdev_flow for a long time. So this commit reintroduces reference counting
for struct dp_netdev_flow
Subsequent commits make use of it.
Si
ofpbuf_rss() is introduced to get the RSS hash from the NIC. It works only for
packets received by DPDK (otherwise it returns 0, which tells the caller to
compute the hash in another way)
This commit also configure DPDK devices to compute RSS hash for UDP packets
Signed-off-by: Daniele Di Proiett
Since lookups in the classifier can be pretty expensive, we introduce this
(thread local) cache which simply compares the miniflows of the packets
Signed-off-by: Daniele Di Proietto
---
lib/dpif-netdev.c | 419 +-
1 file changed, 353 insertions
struct 'miniflow' already contains MINI_N_INLINE values, therefore we can save
few bytes in netdev_flow_key.
Signed-off-by: Daniele Di Proietto
---
lib/dpif-netdev.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/lib/dpif-netdev.c b/lib/dpif-netdev.c
index cfd7539..cad8c7a 1
These values has been found to give the best throughput in simple cases (1 flow
64 bytes UDP packets)
Signed-off-by: Daniele Di Proietto
---
lib/netdev-dpdk.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/lib/netdev-dpdk.c b/lib/netdev-dpdk.c
index 62c9a0c..b45e367 1006
May you share checkpatch.pl? I try to look up it in ovs but not find.
Bests,
Wenyu
-Original Message-
From: Pravin Shelar [mailto:pshe...@nicira.com]
Sent: Wednesday, July 23, 2014 7:50 AM
To: Wenyu Zhang
Cc: Romain Lenglet; Ben Pfaff; Jesse Gross; dev@openvswitch.org
Subject: Re: [PATCH
It's not really a workaround - I just updated the kernel to 3.14.13 and used
the latest ovs source code from git - I am going to customize it anyways.
JZ
> Date: Tue, 22 Jul 2014 19:24:02 -0300
> From: f...@redhat.com
> To: jgzh...@hotmail.com
> CC: dev@openvswitch.org
> Subject: Re: [ovs-dev]
Attached is the script.
It is also available in kernel source under script directory.
Thanks,
Pravin.
On Tue, Jul 22, 2014 at 5:23 PM, Wenyu Zhang wrote:
> May you share checkpatch.pl? I try to look up it in ovs but not find.
>
> Bests,
> Wenyu
> -Original Message-
> From: Pravin Shelar
On Tue, Jul 22, 2014 at 5:09 PM, Daniele Di Proietto
wrote:
> These values has been found to give the best throughput in simple cases (1
> flow
> 64 bytes UDP packets)
>
> Signed-off-by: Daniele Di Proietto
Pushed to master. But we need to make these parameter configurable.
> ---
> lib/netdev
On Tue, Jul 22, 2014 at 5:06 PM, Daniele Di Proietto
wrote:
> struct 'miniflow' already contains MINI_N_INLINE values, therefore we can save
> few bytes in netdev_flow_key.
>
> Signed-off-by: Daniele Di Proietto
Pushed to master.
Thanks.
Pravin.
> ---
> lib/dpif-netdev.c | 2 +-
> 1 file chan
Commit db73f7166a6 (netdev-dpdk: Fix race condition with DPDK mempools in
non pmd threads) switched to a new way of setting up 'upcall->packet', but
only initialized two of the fields in the packet. This could cause
core dumps and other strange behavior. In particular it caused failures in
severa
Acked-by: Justin Pettit
On July 22, 2014 at 9:13:58 PM, Ben Pfaff (b...@nicira.com) wrote:
> Commit db73f7166a6 (netdev-dpdk: Fix race condition with DPDK mempools in
> non pmd threads) switched to a new way of setting up 'upcall->packet', but
> only initialized two of the fields in the packet.
Thanks, applied to master.
On Tue, Jul 22, 2014 at 10:22:30PM -0700, Justin Pettit wrote:
> Acked-by: Justin Pettit
>
>
>
> On July 22, 2014 at 9:13:58 PM, Ben Pfaff (b...@nicira.com) wrote:
> > Commit db73f7166a6 (netdev-dpdk: Fix race condition with DPDK mempools in
> > non pmd threads) swit
Pravin, the code in dpif-netdev.c modified by this patch is similar but
not identical on branch-2.3. Do you have an idea whether differences
elsewhere in the tree make this safe on branch-2.3 without change?
Thanks,
Ben.
On Tue, Jul 22, 2014 at 10:30:27PM -0700, Ben Pfaff wrote:
> Thanks, appli
Commit 62974663fe (datapath/flow_netlink: Create right mask with
disabled megaflows) introduced the bug which caused
ovs_nla_get_match() returns immediately after parsing the flow
mask for OVS_KEY_ATTR_ENCAP. Consequently, when vlan encapsulated
packets are present, the corresponding datapath flow
59 matches
Mail list logo
