Re: [ovs-dev] [PATCH v2 1/1] netdev-dpdk: Fix egress policer error detection bug.

2016-08-09 10:20 GMT-07:00 Ian Stokes : > When egress policer is set as a QoS type for a port, an error may occur > during > setup if incorrect parameters are used for the rte_meter. If this occurs > the egress policer construct and set functions should free any allocated >

Re: [ovs-dev] [PATCH] INSTALL.DPDK-ADVANCED: Add vhost multiqueue loopback testcase.

Applied to master, thanks 2016-07-28 5:48 GMT-07:00 Bhanuprakash Bodireddy < bhanuprakash.bodire...@intel.com>: > Add steps for loopback test using vhost-user configured with multiqueue > doing packet forwarding in kernel. > > Signed-off-by: Bhanuprakash Bodireddy

Re: [ovs-dev] [PATCH] Revert "pvector: Expose non-concurrent priority vector."

Simple revert, looks good to me, thanks Acked-by: Daniele Di Proietto <diproiet...@vmware.com> 2016-08-09 13:59 GMT-07:00 Jarno Rajahalme <ja...@ovn.org>: > This reverts commit 8bdfe1313894047d44349fa4cf4402970865950f. > > I failed to see that lib/dpif-netdev.c actually

Re: [ovs-dev] netdev-dpdk: Fix deadlock in destroy_device().

again for >resolving, Daniele. > >Acked-by: mark.b.kavan...@intel.com Thanks for the reviews, I applied this to master > >>On 05.08.2016 23:57, Daniele Di Proietto wrote: >>> netdev_dpdk_vhost_destruct() calls rte_vhost_driver_unregister(), which >>> can trigger

Re: [ovs-dev] [PATCH V7 7/7] netdev-dpdk: add support for jumbo frames

h <mark.b.kavan...@intel.com> > Signed-off-by: Ilya Maximets <i.maxim...@samsung.com> > [diproiet...@vmware.com rebased] > Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> > --- > > v7: > - add 'Signed-off-by' for Ilya Maximets (i.maxim...@samsung

Re: [ovs-dev] [PATCH V7 1/7] ofproto: Consider datapath_type when looking for internal ports.

On 09/08/2016 09:08, "Thadeu Lima de Souza Cascardo" <casca...@redhat.com> wrote: >On Tue, Aug 09, 2016 at 05:01:14PM +0100, Mark Kavanagh wrote: >> From: Daniele Di Proietto <diproiet...@vmware.com> >> >> Interfaces with type "internal"

Re: [ovs-dev] [PATCH v2 3/3] netdev-dpdk: vHost client mode and reconnect

The patch mostly looks good to me, thanks. I'm not 100% sure about the interface. Can we make the flag interface specific? If I'm not mistaken we currently limit vhost-sock-dir to be under OVS rundir. With client mode this is not necessary anymore. I hope that client will be made the default

Re: [ovs-dev] [PATCH 1/3] system-userspace-macros: Check the exit code of ethtool.

On 05/08/2016 11:16, "Joe Stringer" <j...@ovn.org> wrote: >On 4 August 2016 at 18:40, Daniele Di Proietto <diproiet...@vmware.com> wrote: >> If the ethtool command is not available on the system we should fail, >> since the userspace testsuite c

Re: [ovs-dev] [PATCH 3/3] check-kernel: Remove '-d' from TESTSUITEFLAGS.

On 05/08/2016 10:18, "Andy Zhou" <az...@ovn.org> wrote: > > >On Thu, Aug 4, 2016 at 6:43 PM, Daniele Di Proietto ><diproiet...@vmware.com> wrote: > >The '-d' flag tells autotest to always keep the testcase output, but >prevents '--recheck' fr

Re: [ovs-dev] [ovs-dev,V2] netdev-dpdk: fix memory leak

Thanks for the report, I didn't realize that the callback could come in the same thread. I sent a patch that I believe should fix the deadlock here: http://openvswitch.org/pipermail/dev/2016-August/077315.html 2016-08-05 7:48 GMT-07:00 Ilya Maximets : > On 04.08.2016

[ovs-dev] [PATCH] netdev-dpdk: Fix deadlock in destroy_device().

dev-dpdk: fix memory leak") Reported-by: Ilya Maximets <i.maxim...@samsung.com> Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> --- lib/netdev-dpdk.c | 34 -- 1 file changed, 24 insertions(+), 10 deletions(-) diff --git a/lib/netdev-dpdk.c

Re: [ovs-dev] [ovs-discuss] [openvswitch 2.5.90] testsuite: 2224 failed

I can reproduce this too With -march=native, if the CPU has CRC32 extensions we use a different hash function. I suspect the dhcp options are output on the packet in a different order because of this. Perhaps we should make the test agnostic of the order, or order the options on the DHCP

[ovs-dev] [PATCH 3/3] check-kernel: Remove '-d' from TESTSUITEFLAGS.

: Andy Zhou <az...@ovn.org> Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> --- tests/automake.mk | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/automake.mk b/tests/automake.mk index a9ebf91..5d12ae5 100644 --- a/tests/automake.mk +++ b/tests/automake.

[ovs-dev] [PATCH 2/3] system-traffic: Flush conntrack after debug ping6.

test.") Reported-by: Joe Stringer <j...@ovn.org> Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> --- tests/system-traffic.at | 4 1 file changed, 4 insertions(+) diff --git a/tests/system-traffic.at b/tests/system-traffic.at index 666a14d..5b0a1ce 100644 --- a/t

[ovs-dev] [PATCH 1/3] system-userspace-macros: Check the exit code of ethtool.

tests.") Reported-by: Joe Stringer <j...@ovn.org> Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> --- Vagrantfile | 2 +- tests/system-userspace-macros.at | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/Vagrantfile b/Vagrantfile

Re: [ovs-dev] [PATCH] system-traffic: Make ping6 vlan test more reliable.

LGTM, thanks Acked-by: 2016-08-04 17:40 GMT-07:00 Joe Stringer : > Previously we checked on the underlying interfaces rather than the vlan > interfaces to verify whether IPv6 connectivity is available; > occasionally this would fail on some systems. Wait

Re: [ovs-dev] [PATCH] bridge: No QoS configured is not an error

Applied to master, thanks 2016-08-02 8:27 GMT-07:00 Stokes, Ian : > > If no QoS is configured, type value is likely to be an empty string. > > > > This is not an error though, so use the regular command reply function, > > not the error one. > > > > For example, before this

Re: [ovs-dev] [PATCH v1 1/1] netdev-dpdk: Fix egress policer error detection bug.

Thanks for the patch, comments inline 2016-08-02 9:37 GMT-07:00 Ian Stokes : > When egress policer is set as a QoS type for a port, an error may occur > during > setup if incorrect parameters are used for the rte_meter. If this occurs > the egress policer construct and set

Re: [ovs-dev] [PATCH] netdev-dpdk: When no QoS set, set type to empty string

Thanks for the fix! I added you name to AUTHORS and applied this to master 2016-08-02 9:52 GMT-07:00 Maxime Coquelin : > > > On 08/02/2016 05:19 PM, Stokes, Ian wrote: > >> This patch sets *typep to an empty string instead of letting it >>> uninitialized when no QoS

Re: [ovs-dev] [PATCH] netdev-dpdk: Make libnuma dependencies optional

LGTM, thanks Pushed to master 2016-08-04 3:44 GMT-07:00 Ciara Loftus : > Prior to this patch, OVS with DPDK required the libnuma packages to > build. This patch removes this dependency, making it only a requirement > when the CONFIG_RTE_LIBRTE_VHOST_NUMA option is

Re: [ovs-dev] [PATCH V2] netdev-dpdk: fix memory leak

I'm glad we can finally uncomment this code! The patch looks good to me. I made a few style changes and pushed this to master Thanks 2016-08-04 2:49 GMT-07:00 Mark Kavanagh : > DPDK v16.07 introduces the ability to free memzones. > Up until this point, DPDK memory

Re: [ovs-dev] [PATCH] ovs-rcu: Add new ovsrcu_index type.

Applied to master, thanks! On 03/08/2016 15:00, "Jarno Rajahalme" <ja...@ovn.org> wrote: >Looks good to me, > >Acked-by: Jarno Rajahalme <ja...@ovn.org> > >> On Aug 2, 2016, at 5:03 PM, Daniele Di Proietto <diproiet...@vmware.com> >> w

Re: [ovs-dev] [PATCH] test-netlink-conntrack: Fix sparse warning.

On 03/08/2016 10:54, "Joe Stringer" <j...@ovn.org> wrote: >On 3 August 2016 at 10:50, Daniele Di Proietto <diproiet...@vmware.com> wrote: >> On some systems I get a sparse warning when compiling >> tests/test-netlink-conntrack.c >> >> /usr/inc

[ovs-dev] [PATCH] test-netlink-conntrack: Fix sparse warning.

rks around the issue by including a glibc header before a kernel header. Fixes: 2c06d9a927c5("ovstest: Add test-netlink-conntrack command.") Reported-by: Joe Stringer <j...@ovn.org> Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> --- tests/test-netlink-conntrack.c | 1 +

Re: [ovs-dev] [PATCH RFC v3 1/1] netdev-dpdk: Add support for DPDK 16.07

Given that using vhost PMD doesn't seem viable in the very short term, I think we should stick with the vhost lib. I sent a patch for ovsrcu to add a new RCU protected array index. http://openvswitch.org/pipermail/dev/2016-August/077097.html Thanks, Daniele 2016-07-28 6:26 GMT-07:00 Loftus,

Re: [ovs-dev] [PATCH RFC v3 1/1] netdev-dpdk: Add support for DPDK 16.07

2016-07-24 11:06 GMT-07:00 Ben Pfaff <b...@ovn.org>: > On Sun, Jul 24, 2016 at 10:17:13AM -0400, Aaron Conole wrote: > > Daniele Di Proietto <diproiet...@ovn.org> writes: > > > > > Thanks for the patch. > > > > > > I have another c

[ovs-dev] [PATCH] ovs-rcu: Add new ovsrcu_index type.

*() functions should be used to access the type. Even though we say "Do not, in general, declare a typedef for a struct, union, or enum.", I think we're not in the "general" case. Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com&g

Re: [ovs-dev] [PATCH] tests: Fix conntrack tests on windows.

On 02/08/2016 13:19, "Joe Stringer" <j...@ovn.org> wrote: >On 2 August 2016 at 11:58, Daniele Di Proietto <diproiet...@vmware.com> wrote: >> The conntrack unit tests seem to generate different megaflow masks on >> Windows. The megaflow masks depend on the

Re: [ovs-dev] FAILING UNIT ofproto-dpif.at:8215: testing ofproto-dpif - conntrack - output action

Hi Alin, Thanks for the report! I couldn't reproduce this on Linux or BSD, but Guru (thanks!) gave me a Windows setup, which showed the problem. I sent a patch to fix this and the other issue you reported here: http://openvswitch.org/pipermail/dev/2016-August/077062.html Thanks, Daniele

[ovs-dev] [PATCH] tests: Fix conntrack tests on windows.

erd...@cloudbasesolutions.com> Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> --- tests/ofproto-dpif.at | 263 +- 1 file changed, 174 insertions(+), 89 deletions(-) diff --git a/tests/ofproto-dpif.at b/tests/ofproto-dpif.at index 5ce6

[ovs-dev] [PATCH 2/7] vswitchd: Introduce 'mtu_request' column in Interface.

. The current implementation of set_mtu() in netdev-dpdk is removed because it's broken. It will be reintroduced by a subsequent commit on this series. Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> --- NEWS | 2 ++ lib/netdev-dpdk.c

[ovs-dev] [PATCH 7/7] netdev-dpdk: add support for Jumbo Frames

th ovs-vsctl, via the 'mtu_request' parameter. Signed-off-by: Mark Kavanagh <mark.b.kavan...@intel.com> [diproiet...@vmware.com rebased] Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> --- INSTALL.DPDK-ADVANCED.md | 59 +- INSTALL.DPDK.md

[ovs-dev] [PATCH 6/7] netdev: Make netdev_set_mtu() netdev parameter non-const.

Every provider silently drops the const attribute when converting the parameter to the appropriate subclass. Might as well drop the const attribute from the parameter, since this is a "set" function. --- lib/netdev-dummy.c| 2 +- lib/netdev-linux.c| 2 +- lib/netdev-provider.h | 2 +-

[ovs-dev] [PATCH 5/7] tests: Add a new MTU test.

Also, netdev-dummy needs to call netdev_change_seq_changed() in set_mtu(). Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> --- lib/netdev-dummy.c| 5 - tests/ofproto-dpif.at | 30 ++ 2 files changed, 34 insertions(+), 1 deletion(-) diff

[ovs-dev] [PATCH 4/7] netdev-dummy: Add dummy-internal class.

netdev type ("dummy"), is harder to implement, because in so many places only the netdev object is available. Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> --- lib/dpif-netdev.c | 2 +- lib/netdev-dummy.c| 14 -- tests/bridge.at

[ovs-dev] [PATCH 1/7] ofproto: Consider datapath_type when looking for internal ports.

internal interfaces are considered non-internal in the userspace datapath for the purpose of adjusting the MTU. Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> --- ofproto/ofproto.c | 16 +--- 1 file changed, 9 insertions(+), 7 deletions(-) diff --git a/ofproto/ofp

[ovs-dev] [PATCH 3/7] netdev: Pass 'netdev_class' to ->run() and ->wait().

This will allow run() and wait() methods to be shared between different classes and still perform class-specific work. Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> --- lib/netdev-bsd.c | 6 +++--- lib/netdev-dummy.c| 4 ++-- lib/netdev-linux.c| 6 +++--- lib/

Re: [ovs-dev] tests: Add new pmd test for pmd-rxq-affinity.

Thanks for the review, applied to master On 28/07/2016 01:59, "Ilya Maximets" <i.maxim...@samsung.com> wrote: >Thanks for making this. > >Acked-by: Ilya Maximets <i.maxim...@samsung.com> > >On 27.07.2016 23:12, Daniele Di Proietto wrote: >> T

Re: [ovs-dev] [PATCH] dpif-netdev: Fix xps revalidation.

Thanks for the fix, applied to master! On 29/07/2016 01:07, "Ilya Maximets" wrote: >Revalidation should work in case of 'dynamic_txqs == true'. > >Fixes: 324c8374852a ("dpif-netdev: XPS (Transmit Packet Steering) >implementation.") >Signed-off-by: Ilya Maximets

Re: [ovs-dev] [PATCH v2] netdev-dpdk: Add Flow Control support.

I changed the fc_conf initialization to use memset, because clang was complaining about the initializer. I moved the description in vswitch.xml to avoid nesting inside the "Common Columns" group. I changed slightly the wording in vswitch.xml, in case we want to implement this even for non DPDK

Re: [ovs-dev] [PATCH] ovs-numa: fixed cmask parse with 0x prefix

so thanks for the styling reminder… I haven’t gone through those in much > detail. Let me send another patch that complies with those. > > > > -- > > Best, > > Wei Shen. > > > > *From: *Daniele Di Proietto <diproiet...@ovn.org> > *Date: *Wednesday, July

Re: [ovs-dev] [PATCH 2/4] tests: Remove trim_zeros() from ovn tests.

On 28/07/2016 22:22, "Ben Pfaff" <b...@ovn.org> wrote: >On Thu, Jul 28, 2016 at 07:58:04PM -0700, Daniele Di Proietto wrote: >> trim_zeros() is not necessary anymore, since now we don't pad packets in >> the userspace datapath. >> >> S

Re: [ovs-dev] [PATCH v5 13/16] system-tests: Run conntrack tests with userspace.

On 27/07/2016 14:18, "Ben Pfaff" <b...@ovn.org> wrote: >On Wed, Jul 27, 2016 at 01:51:00PM -0700, Jesse Gross wrote: >> On Wed, Jul 27, 2016 at 1:40 PM, Daniele Di Proietto >> <diproiet...@vmware.com> wrote: >> > On 27/07/2016 13:12, "Joe S

[ovs-dev] [PATCH 4/4] system-tests: Add ping through conntrack test.

Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> Acked-by: Joe Stringer <j...@ovn.org> --- tests/system-traffic.at | 84 + 1 file changed, 84 insertions(+) diff --git a/tests/system-traffic.at b/tests/system-traffic.at i

[ovs-dev] [PATCH 2/4] tests: Remove trim_zeros() from ovn tests.

trim_zeros() is not necessary anymore, since now we don't pad packets in the userspace datapath. Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> --- tests/ovn.at | 135 +++ 1 file changed, 43 insertions(+), 92 deletions(-)

[ovs-dev] [PATCH 3/4] system-tests: Run conntrack tests with userspace.

. Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> Acked-by: Joe Stringer <j...@ovn.org> Acked-by: Flavio Leitner <f...@sysclose.org> --- tests/system-kmod-macros.at | 28 + tests/system-ovn.at | 3 +++ tests/system-traffi

Re: [ovs-dev] [PATCH v5 00/16] Userspace (DPDK) connection tracker

I pushed the intended version except that I forgot to squash that commit with the previous one. Sorry about this On 28/07/2016 00:40, "Ilya Maximets" <i.maxim...@samsung.com> wrote: >Sorry. >TO: Daniele Di Proietto <diproiet...@vmware.com> > >On 28.07.201

Re: [ovs-dev] Mea culpa - rewound master branch

2016-07-28 10:52 GMT-07:00 Flavio Leitner : > On Thu, Jul 28, 2016 at 09:32:54AM -0700, Ben Pfaff wrote: > > On Thu, Jul 28, 2016 at 12:49:37PM -0300, Flavio Leitner wrote: > > > On Wed, Jul 27, 2016 at 11:49:18AM -0700, Ben Pfaff wrote: > > > > Some time ago this morning, I

Re: [ovs-dev] [PATCH v5 00/16] Userspace (DPDK) connection tracker

Thanks for the reviews, I pushed this to master except for the system tests part. On 26/07/2016 17:58, "Daniele Di Proietto" <diproiet...@vmware.com> wrote: >This series aims to implement the ct() action for the dpif-netdev datapath. >The bulk of the code is in th

Re: [ovs-dev] [PATCH v5 05/16] conntrack: Periodically delete expired connections.

On 27/07/2016 21:13, "Joe Stringer" <j...@ovn.org> wrote: >On 27 July 2016 at 19:01, Daniele Di Proietto <diproiet...@vmware.com> wrote: >> >> >> >> >> >> On 27/07/2016 17:14, "Joe Stringer" <j...@ovn.org> wrote

Re: [ovs-dev] [PATCH v5 05/16] conntrack: Periodically delete expired connections.

On 27/07/2016 17:14, "Joe Stringer" <j...@ovn.org> wrote: >On 26 July 2016 at 17:58, Daniele Di Proietto <diproiet...@vmware.com> wrote: >> This commit adds a thread that periodically removes expired connections. >> >> The expiration time of a conne

Re: [ovs-dev] [PATCH] netdev-dpdk: Add Flow Control support.

Thanks for the patch. I agree, I'd be nice to document this in INSTALL.DPDK-ADVANCED.md We should also document the new fields in vswitchd/vswitch.xml. Probably it's better to use "true" and "false" rather that "on" and "off", for consistency with other configuration options and so that we can

Re: [ovs-dev] [PATCH 0/5] netdev_open conflicting types

Looks good to me, applied to master, thanks! 2016-07-27 8:06 GMT-07:00 Thadeu Lima de Souza Cascardo : > Fix some uses of type on netdev_open. > > We established that there are two types: the database type and the netdev > type. > And that ofproto and dpif layers use the

Re: [ovs-dev] [PATCH] ovs-numa: fixed cmask parse with 0x prefix

Thanks for the patch. We never accepted the 0x prefix for pmd-cpu-mask, but I guess there's no harm in doing it and it might make user's life easier. We always use braces, even for single statement, please read CodingStyle.md

Re: [ovs-dev] [PATCH v5] netdev-dpdk: Set pmd thread priority

Thanks for the patch, the implementation looks good to me too. During testing I kept noticing that it's way too easy to make OVS completely unresponsive. As you point out in the documentation by having dpdk-lcore-mask the same as pmd-cpu-mask, OVS cannot even be killed (a kill -9 is required).

Re: [ovs-dev] [PATCH v5 13/16] system-tests: Run conntrack tests with userspace.

On 27/07/2016 13:12, "Joe Stringer" <j...@ovn.org> wrote: >On 26 July 2016 at 17:58, Daniele Di Proietto <diproiet...@vmware.com> wrote: >> The userspace connection tracker doesn't support ALGs, frag reassembly >> or NAT yet, so skip those tests. >&g

[ovs-dev] [PATCH] tests: Add new pmd test for pmd-rxq-affinity.

This tests that the newly introduced pmd-rxq-affinity option works as intended, at least for a single port. Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> --- tests/pmd.at | 53 + 1 file changed, 53 insertions(+) diff

Re: [ovs-dev] [PATCH v5 1/4] dpif-netdev: XPS (Transmit Packet Steering) implementation.

I don't think dynamic_txqs should be atomic, since we change it when the pmd threads are stopped. Also, in port_create() we should check for 'netdev_n_txq(netdev) < n_cores + 1' after we reconfigure the device. Other than that this looks good to me, so I applied the following incremental and

Re: [ovs-dev] [PATCH v5 4/4] dpif-netdev: Introduce pmd-rxq-affinity.

ofputil_parse_key_value() to parse the affinity seems like a good idea, thanks! I got a compiler warning on an unused variable. I fixed that and applied the series to master. Thanks, Daniele On 27/07/2016 07:44, "Ilya Maximets" wrote: >New

[ovs-dev] [PATCH v5 04/16] conntrack: New userspace connection tracker.

and to allow the pickup of already established connections. Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> Acked-by: Antonio Fischetti <antonio.fische...@intel.com> Acked-by: Joe Stringer <j...@ovn.org> --- COPYING | 1 + debian/copyrigh

[ovs-dev] [PATCH v5 12/16] tests: Add conntrack ofproto-dpif tests.

While the system testsuite already has connection tracking tests, it will be still useful to add some to the standard testsuite because: * They're run more often by developers. * Some of them are more interesting for the userspace datapath. Signed-off-by: Daniele Di Proietto <dipro

[ovs-dev] [PATCH v5 15/16] conntrack: Track ICMP type and code.

Reported-by: Subramani Paramasivam <subramani.paramasi...@wipro.com> Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> Acked-by: Joe Stringer <j...@ovn.org> --- lib/automake.mk | 1 + lib/conntrack-icmp.c| 105 l

[ovs-dev] [PATCH v5 06/16] tests: Add very simple conntrack benchmark.

different batching and locking strategies. E.g. the line: `./tests/ovstest test-conntrack benchmark 1 1488 32` starts 1 thread that will send 1488 packets to the connection tracker, 32 at a time. It will print the time taken to process them. Signed-off-by: Daniele Di Proietto <dipro

[ovs-dev] [PATCH v5 16/16] conntrack: Add 'dl_type' parameter to conntrack_execute().

have to complicate sightly test-conntrack to group the packets by dl_type before passing them to the connection tracker. Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> Acked-by: Joe Stringer <j...@ovn.org> --- lib/conntrack.c

[ovs-dev] [PATCH v5 13/16] system-tests: Run conntrack tests with userspace.

datapath checks for the IPv4 header checksum, so fix those in the hardcoded packets. Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> Acked-by: Joe Stringer <j...@ovn.org> Acked-by: Flavio Leitner <f...@sysclose.org> --- tests/system-kmod-macros.at | 28 ++

[ovs-dev] [PATCH v5 03/16] flow: Introduce parse_dl_type().

The function simply returns the ethernet type of the packet (after eventually discarding the VLAN tag). It will be used by a following commit. Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> Acked-by: Flavio Leitner <f...@sysclose.org> --- lib/flow.c | 14 --

[ovs-dev] [PATCH v5 14/16] system-tests: Add ping through conntrack test.

Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> Acked-by: Joe Stringer <j...@ovn.org> --- tests/system-traffic.at | 84 + 1 file changed, 84 insertions(+) diff --git a/tests/system-traffic.at b/tests/system-traffic.at i

[ovs-dev] [PATCH v5 09/16] dpif-netdev: Implement conntrack dump functions.

New functions are implemented in the conntrack module to support this. Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> Acked-by: Flavio Leitner <f...@sysclose.org> --- lib/conntrack-private.h | 3 ++ lib/conntrack-tcp.c | 34 + lib/conntrack.c

[ovs-dev] [PATCH v5 11/16] flow: Generate checksum and udp_len in flow_compose().

This is useful to test the connection tracker, which performs checksum and udp length verification. Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> Acked-by: Joe Stringer <j...@ovn.org> --- lib/flow.c| 62 ++-- tests/ofproto-d

[ovs-dev] [PATCH v5 10/16] dpif-netdev: Implement conntrack flush interface.

New functions are implemented in the conntrack module to support this. Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> Acked-by: Flavio Leitner <f...@sysclose.org> --- lib/conntrack.c | 23 +++ lib/conntrack.h | 2 ++ lib/dpif-netdev.c | 10 +

[ovs-dev] [PATCH v5 08/16] dpif-netdev: Execute conntrack action.

This commit implements the OVS_ACTION_ATTR_CT action in dpif-netdev. To allow ofproto-dpif to detect the conntrack feature, flow_put will not discard anymore flows with ct_* fields set. We still shouldn't allow flows with NAT bits set, since there is no support for NAT. Signed-off-by: Daniele Di

[ovs-dev] [PATCH v5 05/16] conntrack: Periodically delete expired connections.

of the corresponding 'timeout' list. This ways, the list is always ordered by 'expiration'. When the cleanup thread iterates through the lists for expired connections, it can stop at the first non expired connection. Suggested-by: Joe Stringer <j...@ovn.org> Signed-off-by: Daniele Di Pr

[ovs-dev] [PATCH v5 02/16] flow: Export parse_ipv6_ext_hdrs().

This will be used by a future commit. Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> Acked-by: Joe Stringer <j...@ovn.org> Acked-by: Flavio Leitner <f...@sysclose.org> --- lib/flow.c | 140 ++--- lib/flow.h

[ovs-dev] [PATCH v5 07/16] tests: Add test-conntrack pcap test.

-off-by: Daniele Di Proietto <diproiet...@vmware.com> Acked-by: Flavio Leitner <f...@sysclose.org> --- tests/test-conntrack.c | 73 ++ 1 file changed, 73 insertions(+) diff --git a/tests/test-conntrack.c b/tests/test-conntrack.c index 37c7

[ovs-dev] [PATCH v5 01/16] packets: Define ICMP types.

Linux and FreeBSD have slightly different names for these constants. Windows doesn't define them. It is simpler to redefine them from scratch for OVS. The new names are different than those used in Linux and FreeBSD. These definitions will be used by a future commit. Signed-off-by: Daniele Di

[ovs-dev] [PATCH v5 00/16] Userspace (DPDK) connection tracker

ing strategy in conntrack_execute() to allow a newly created connection to be picked up by packets in the same batch. * Added an ovs-test module to throw pcap files at the connection tracker. * Added a workaround for the userspace testsuite on new kernels and a tcp non-conntrack test. Daniele Di

Re: [ovs-dev] [PATCH v4] dpif-netdev: XPS (Transmit Packet Steering) implementation.

Thanks for the patch I think the caller of dp_netdev_execute_actions() should always pass a valid timestamp. We can pass it from aux->now to dp_execute_userspace_actions(), we can add it to fast_path_processing() so that it can be passed down to handle_packet_upcall(). In the other cases

Re: [ovs-dev] [PATCH v3 3/3] dpif-netdev: Introduce pmd-rxq-affinity.

Looks mostly good to me, a couple more comments inline Thanks, Daniele On 26/07/2016 06:48, "Ilya Maximets" <i.maxim...@samsung.com> wrote: >On 26.07.2016 04:46, Daniele Di Proietto wrote: >> Thanks for the patch. >> >> I haven't been abl

Re: [ovs-dev] [RFC 4/5] dpctl: uses open_type when calling netdev_open

2016-07-26 11:29 GMT-07:00 Thadeu Lima de Souza Cascardo < casca...@redhat.com>: > On Tue, Jul 26, 2016 at 11:20:38AM -0700, Daniele Di Proietto wrote: > > Hi Cascardo, > > > > thanks for your input on this. It's quite messy right now, but I believe > > we

Re: [ovs-dev] [PATCH v3 1/3] bridge: Pass interface's configuration to datapath.

On 26/07/2016 06:19, "Ilya Maximets" <i.maxim...@samsung.com> wrote: >On 26.07.2016 04:45, Daniele Di Proietto wrote: >> Thanks for the patch >> >> It looks good to me, a few minor comments inline >> >> >> On 15/07/201

Re: [ovs-dev] [RFC 4/5] dpctl: uses open_type when calling netdev_open

Hi Cascardo, thanks for your input on this. It's quite messy right now, but I believe we have a chance to fix this up. Replies inline 2016-07-26 7:33 GMT-07:00 Thadeu Lima de Souza Cascardo <casca...@redhat.com >: > On Mon, Jul 25, 2016 at 11:03:29AM -0700, Daniele Di Proietto wrote:

Re: [ovs-dev] [PATCH v3 3/3] dpif-netdev: Introduce pmd-rxq-affinity.

Thanks for the patch. I haven't been able to apply this without the XPS patch. This looks like a perfect chance to add more tests to pmd.at. I can do it if you want I started taking a look at this patch and I have a few comments inline. I'll keep looking at it tomorrow Thanks, Daniele

Re: [ovs-dev] [PATCH v3 2/3] util: Expose function nullable_string_is_equal.

Nice! Applied to master, thanks On 15/07/2016 04:54, "Ilya Maximets" wrote: >Implementation of 'nullable_string_is_equal()' moved to util.c and >reused inside dpif-netdev. > >Signed-off-by: Ilya Maximets >--- > lib/dpif-netdev.c|

Re: [ovs-dev] [PATCH v3 1/3] bridge: Pass interface's configuration to datapath.

Thanks for the patch It looks good to me, a few minor comments inline On 15/07/2016 04:54, "Ilya Maximets" wrote: >This commit adds functionality to pass value of 'other_config' column >of 'Interface' table to datapath. > >This may be used to pass not directly

Re: [ovs-dev] [PATCH v3] netdev-dpdk: Set pmd thread priority

I agree with Mark's comments, other than that this looks good to me. If you agree with the comments would you mind sending an updates version? Thanks, Daniele 2016-07-19 7:04 GMT-07:00 Kavanagh, Mark B : > > > > Hi Bhanu, > > Thanks for the patch - some comments

Re: [ovs-dev] [PATCH 3/4] netdev-dpdk: Add vHost User PMD

Thanks for the patch This needs a little bit of rebasing, I did it myself to review, but it'd be nice to have an updated version. I like the simplification that this brings especially to the fast path. If we merge this before we merge the DPDK 16.07 we won't have to deal with the vid change.

Re: [ovs-dev] [PATCH 2/4] netdev-dpdk: Consistent naming for vhost functions

Looks like every caller of NETDEV_DPDK_CLASS() passes NULL as INIT param. Maybe we can remove that? Acked-by: Daniele Di Proietto <diproiet...@vmware.com> 2016-07-15 7:26 GMT-07:00 Ciara Loftus <ciara.lof...@intel.com>: > A mix of vhost_user_ and vhost_ is used when naming

Re: [ovs-dev] [PATCH 1/4] netdev-dpdk: Remove dpdkvhostcuse ports

Acked-by: Daniele Di Proietto <diproiet...@vmware.com> 2016-07-15 7:26 GMT-07:00 Ciara Loftus <ciara.lof...@intel.com>: > This commit removes the 'dpdkvhostcuse' port type from the userspace > datapath. vhost-cuse ports are quickly becoming obsolete as the > vhost-

Re: [ovs-dev] [PATACH]netdev-dpdk: remove duplicated code in netdev_dpdk_get_status

Thanks for the patch! I had to manually unwrap some long lines. Could you maybe use git send-email next time? Anyway, I added you to AUTHORS and applied this to master Thanks, Daniele 2016-07-25 1:37 GMT-07:00 : > From caeb84217c38ccd0b2076689fd36b578c00678ad Mon Sep

Re: [ovs-dev] [PATCH] netdev-dpdk: Apply batch truncation API.

Applied to master, thanks 2016-07-25 8:14 GMT-07:00 William Tu : > Instead of looping into each packet and check whether to truncate, the > patch moves it out of the loop and uses batch API. If truncation is > not set, checking 'trunc' in 'struct dp_packet_batch' at

Re: [ovs-dev] Backport Request: dpif-netdev: Remove PMD latency on seq_mutex

>But I understand there may be some risk associated with due to the nature of >the changes. > >Thanks, >Ciara > >> >> Thanks >> Vinod >> >> >> >> -Original Message- >> From: Daniele Di Proietto [mailto:diproiet...@vmware.com]

Re: [ovs-dev] [RFC 4/5] dpctl: uses open_type when calling netdev_open

2016-07-25 9:57 GMT-07:00 Thadeu Lima de Souza Cascardo <casca...@redhat.com >: > On Fri, Jul 22, 2016 at 02:49:39PM -0700, Daniele Di Proietto wrote: > > I would prefer if dpctl kept using the datapath types. The translation > > from database types to datapath type sho

Re: [ovs-dev] releasing 2.6: branch Aug 1, release Sep 15

On 25/07/2016 09:16, "Ben Pfaff" wrote: >On Mon, Jul 25, 2016 at 03:44:01PM +, Gray, Mark D wrote: >> >> >> > -Original Message- >> > From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Ben Pfaff >> > Sent: Saturday, July 23, 2016 5:00 PM >> > To:

Re: [ovs-dev] [PATCH v2] ovs-lib: Keep internal interface ip during upgrade.

On 22/07/2016 12:54, "Ben Pfaff" <b...@ovn.org> wrote: >On Tue, Jun 21, 2016 at 07:27:30PM -0700, Daniele Di Proietto wrote: >> Commit 9b5422a98f81("ovs-lib: Try to call exit before killing.") >> introduced a problem where internal interfaces are de

Re: [ovs-dev] [PATCH v2] ovs-lib: Keep internal interface ip during upgrade.

2016-07-22 16:02 GMT-07:00 Daniele Di Proietto <diproiet...@ovn.org>: > > > 2016-06-22 9:52 GMT-07:00 Darrell Ball <dlu...@gmail.com>: > >> On Tue, Jun 21, 2016 at 7:27 PM, Daniele Di Proietto < >> diproiet...@vmware.com >> > wrote: >> >

Re: [ovs-dev] [PATCH v2] ovs-lib: Keep internal interface ip during upgrade.

2016-06-22 9:52 GMT-07:00 Darrell Ball <dlu...@gmail.com>: > On Tue, Jun 21, 2016 at 7:27 PM, Daniele Di Proietto < > diproiet...@vmware.com > > wrote: > > > Commit 9b5422a98f81("ovs-lib: Try to call exit before killing.") > > introduced a p

Re: [ovs-dev] Backport Request: dpif-netdev: Remove PMD latency on seq_mutex

I'm not sure I'm 100% comfortable back porting this to branch-2.5 I see the change more as a feature rather than a bugfix. Also it touches some core modules (seq and rcu) in a non trivial way. What do you guys think? Thanks, Daniele On 22/07/2016 15:03, "Flavio Leitner"

Re: [ovs-dev] [PATCH RFC v3 1/1] netdev-dpdk: Add support for DPDK 16.07

Thanks for the patch. I have another concern with this. If we're still going to rely on RCU to protect the vhost device (and as pointed out by Ilya, I think we should) we need to use RCU-like semantics on the vid array index. I'm not sure a boolean flag is going to be enough. CCing Jarno: We

Re: [ovs-dev] [RFC 4/5] dpctl: uses open_type when calling netdev_open

I would prefer if dpctl kept using the datapath types. The translation from database types to datapath type should happen in ofproto, dpctl is supposed to be used to interact with the datapath directly. What do you guys think? The rest of the series looks good to me as well. Thanks, Daniele

Re: [ovs-dev] [RFC PATCH v2 1/1] netdev-dpdk: Add support for DPDK 16.07

[...] > @@ -1776,7 +1764,8 @@ netdev_dpdk_get_stats(const struct netdev > > *netdev, struct netdev_stats *stats) > > netdev_dpdk_get_carrier(netdev, ); > > ovs_mutex_lock(>mutex); > > > > -struct rte_eth_xstats *rte_xstats; > > +struct rte_eth_xstat *rte_xstats; > > +struct

<    1   2   3   4   5   6   7   8   9   10   >