Re: [ovs-dev] [patch_v7 0/9] Userspace Datapath: Introduce NAT support.

Hi Darrell, I took another look at the series and provided a few comments inline. Other than those the patches look good to me, but I haven't looked at every possible corner case :-) Thanks, Daniele 2017-03-24 2:15 GMT-07:00 Darrell Ball : > This patch series introduces NAT

Re: [ovs-dev] [patch_v7 6/9] System Tests: Enhance NAT tests.

Not sure this is very important: so far we managed to avoid using tcpdump in the tests. Would it be possible to use ovs-ofctl monitor? In any case, maybe, it shouldn't be prefixed by sudo 2017-03-24 2:15 GMT-07:00 Darrell Ball : > Two new tests are added and two other tests

Re: [ovs-dev] [patch_v7 5/9] dpdk: Add orig tuple context recovery.

2017-03-24 2:15 GMT-07:00 Darrell Ball : > This patch adds orig tuple checking and context > recovery; NAT interactions are factored in. > Orig tuple support exists to better handle policy > changes. > > Signed-off-by: Darrell Ball > --- > lib/conntrack.c | 69

Re: [ovs-dev] [patch_v7 2/9] dpdk: Remove batch sorting in userspace conntrack.

Thanks for doing this Acked-by: Daniele Di Proietto <diproiet...@ovn.org> 2017-03-24 2:15 GMT-07:00 Darrell Ball <dlu...@gmail.com>: > Signed-off-by: Darrell Ball <dlu...@gmail.com> > Acked-by: Flavio Leitner <f...@sysclose.org>

Re: [ovs-dev] [PATCH] tests/pmd.at: Fix race in "PMD - change numa node" test

monitor) and verifying > any single packet instead of checking the 2 packets at the same time. > > CC: Daniele Di Proietto <daniele.di.proie...@gmail.com> > Fixes: a12e2a88d672 ("test: Add more pmd tests.") > Signed-off-by: Timothy Redaelli <tredae...@redhat.com>

[ovs-dev] [RFC] packets: Do not initialize ct_orig_tuple.

ove commit (it was 14.88 Mpps before, it is 13 Mpps after). This patch skips initializing ct_orig_tuple in pkt_metadata_init(). It should be enough to initialize ct_state, because nobody should look at ct_orig_tuple unless ct_state is != 0. CC: Jarno Rajahalme <ja...@ovn.org> Signed-off-by: Dan

Re: [ovs-dev] [PATCH v2 branch-2.6] docs: Use DPDK 16.07.2 stable release

2017-03-10 3:47 GMT-08:00 Ian Stokes : > DPDK now provides a stable release branch. Modify dpdk docs and travis > linux build script to use the DPDK 16.07.2 stable branch to benefit from > most recent bug fixes. > > Signed-off-by: Ian Stokes Thanks,

Re: [ovs-dev] [PATCH v2] docs: Use DPDK 16.11.1 stable release.

2017-03-10 3:47 GMT-08:00 Ian Stokes : > DPDK now provides a stable release branch. Modify dpdk docs and travis linux > build script to use the DPDK 16.11.1 stable branch to benefit from most > recent bug fixes. > > Signed-off-by: Ian Stokes Thanks,

[ovs-dev] [PATCH v3 3/3] ofp-actions: Add limit to learn action.

delete the ukey, the next upcall will be a miss upcall and that will immediatedly install the correct datapath flow. Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> --- include/openvswitch/ofp-actions.h | 12 +++ lib/learn.c| 24 + lib/ofp-act

[ovs-dev] [PATCH v3 2/3] ofp-actions: Factor out decode_LEARN_{common, spec}().

No functional change, they will be used by next commit. Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> --- lib/ofp-actions.c | 58 ++- 1 file changed, 40 insertions(+), 18 deletions(-) diff --git a/lib/ofp-actions.c b/l

[ovs-dev] [PATCH v3 1/3] ofproto-dpif-xlate: Create XC_LEARN entry after learning.

This will be useful in a separate commit, because learning can fail. Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> Acked-by: Ben Pfaff <b...@ovn.org> --- ofproto/ofproto-dpif-xlate.c | 24 +--- 1 file changed, 17 insertions(+), 7 deletions(-)

[ovs-dev] [PATCH v3 0/3] Learn action limit

o call ofproto_flow_mod_uninit() if we don't call ofproto_flow_mod_learn_start(). * Simplify ofp-actions changes according to Ben comments(thanks!) Daniele Di Proietto (3): ofproto-dpif-xlate: Create XC_LEARN entry after learning. ofp-actions: Factor out decode_LEARN_{common,spec}(). ofp-actions: Add limit to

Re: [ovs-dev] [PATCH RFC 4/4] dpif-netdev: Don't uninit emc on reload.

2017-02-21 6:49 GMT-08:00 Ilya Maximets : > There are many reasons for reloading of pmd threads: > * reconfiguration of one of the ports. > * Adjusting of static_tx_qid. > * Adding new tx/rx ports. > > In many cases EMC is still useful after reload

Re: [ovs-dev] [PATCH 3/4] dpif-netdev: Avoid port's reconfiguration on pmd-cpu-mask changes.

2017-02-21 6:49 GMT-08:00 Ilya Maximets : > Reconfiguration of HW NICs may lead to packet drops. > In current model all physical ports will be reconfigured each > time number of PMD threads changed. Since we not stopping > threads on pmd-cpu-mask changes, this patch will

Re: [ovs-dev] [PATCH 2/4] dpif-netdev: Incremental addition/deletion of PMD threads.

2017-02-21 6:49 GMT-08:00 Ilya Maximets : > Currently, change of 'pmd-cpu-mask' is very heavy operation. > It requires destroying of all the PMD threads and creating > them back. After that, all the threads will sleep until > ports' redistribution finished. > > This patch

Re: [ovs-dev] [PATCH] docs: Use DPDK 16.11.1 stable release.

2017-03-09 13:15 GMT-08:00 Ian Stokes : > DPDK now provides a stable release branch. Modify dpdk docs and travis linux > build script to use the DPDK 16.11.1 stable branch to benefit from most > recent bug fixes. > > Signed-off-by: Ian Stokes Thanks

Re: [ovs-dev] [PATCH v2] dpdk: Redirect DPDK log to OVS logging subsystem.

2017-03-06 11:28 GMT-08:00 Aaron Conole : > Ilya Maximets writes: > >> This should be helpful for have all the logs in one place. >> 'ovs-appctl vlog' commands for 'dpdk' module can be used >> to configure the log level. Lower bound for DPDK logging >>

Re: [ovs-dev] [PATCH v3] netdev-dpdk: Fix mempool segfault.

lper(). > > Signed-off-by: Ian Stokes <ian.sto...@intel.com> > Fixes: 2ae3d542 ("netdev-dpdk: Refactor dpdk_mp_get().") > Fixes: 0072e931 ("netdev-dpdk: add support for jumbo frames") > CC: Daniele Di Proietto <diproiet...@vmware.com> > CC: Mark

Re: [ovs-dev] [patch_v6 6/8] dpdk: Add missing CHECK_CONNTRACK_ALG guards.

2017-02-16 0:47 GMT-08:00 Darrell Ball : > Signed-off-by: Darrell Ball > Acked-by: Flavio Leitner Thanks, I applied this to master > --- > tests/system-traffic.at | 3 +++ > 1 file changed, 3 insertions(+) > > diff --git

Re: [ovs-dev] [patch_v6 4/8] dpdk: Userspace Datapath: Introduce NAT Support.

2017-03-09 10:21 GMT-08:00 Darrell Ball <db...@vmware.com>: > > > On 3/8/17, 6:14 PM, "ovs-dev-boun...@openvswitch.org on behalf of Daniele Di > Proietto" <ovs-dev-boun...@openvswitch.org on behalf of diproiet...@ovn.org> > wrote: > > 2017-02-16 0:

Re: [ovs-dev] [patch_v6 3/8] dpdk: Remove batch sorting in userspace conntrack.

Thanks for posting this as a separate patch, it makes the review easier. The idea and the patch look good to me. One comment: with this code we don't need the ctxs array, we can just have a single ctx in the for loop. Other than that: Acked-by: Daniele Di Proietto <diproiet...@vmware.

Re: [ovs-dev] [PATCH 4/7] ofproto: New cookie-counters module.

On 07/03/2017 10:35, "Ben Pfaff" <b...@ovn.org> wrote: >On Fri, Feb 24, 2017 at 06:57:58PM -0800, Daniele Di Proietto wrote: >> The new module will be used by ofproto to keep track of the number of >> learned flows with the same cookie in the same tabl

Re: [ovs-dev] [PATCH 6/7] ofproto: Implement learning limit.

On 07/03/2017 10:39, "Ben Pfaff" <b...@ovn.org> wrote: >On Fri, Feb 24, 2017 at 06:58:00PM -0800, Daniele Di Proietto wrote: >> With this commit we honor the newly introduced limit to the learn >> action. >> >> When learning a new rule (with the li

[ovs-dev] [PATCH v2 3/3] ofp-actions: Add limit to learn action.

execution, so the only option is to slowpath the ukey. I don't think this has big performance implications since it's done only for ukey with no traffic. Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> --- include/openvswitch/ofp-actions.h | 12 +++ lib/l

[ovs-dev] [PATCH v2 2/3] ofp-actions: Factor out decode_LEARN_{common, spec}().

No functional change, they will be used by next commit. Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> --- lib/ofp-actions.c | 58 ++- 1 file changed, 40 insertions(+), 18 deletions(-) diff --git a/lib/ofp-actions.c b/l

[ovs-dev] [PATCH v2 1/3] ofproto-dpif-xlate: Create XC_LEARN entry after learning.

This will be useful in a separate commit, because learning can fail. Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> Acked-by: Ben Pfaff <b...@ovn.org> --- ofproto/ofproto-dpif-xlate.c | 24 +--- 1 file changed, 17 insertions(+), 7 deletions(-)

[ovs-dev] [PATCH v2 0/3] Learn action limit

all ofproto_flow_mod_uninit() if we don't call ofproto_flow_mod_learn_start(). * Simplify ofp-actions changes according to Ben comments(thanks!) Daniele Di Proietto (3): ofproto-dpif-xlate: Create XC_LEARN entry after learning. ofp-actions: Factor out decode_LEARN_{common,spec}(). ofp-actions:

Re: [ovs-dev] [PATCH v2] conntrack: Fix checks for TCP, UDP, and IPv6 header sizes.

t;bshas...@sec.t-labs.tu-berlin.de> > Signed-off-by: Ben Pfaff <b...@ovn.org> Acked-by: Daniele Di Proietto <diproiet...@vmware.com> > --- > v1->v2: Eliminate duplicate check in extract_l3_ipv6(). Thanks Daniele! > > lib/conntrack.c | 16 +++- > 1 file ch

Re: [ovs-dev] [PATCH] conntrack: Fix checks for TCP, UDP, and IPv6 header sizes.

t;bshas...@sec.t-labs.tu-berlin.de> > Signed-off-by: Ben Pfaff <b...@ovn.org> Oops, thanks for the fix, Ben Fixes: a489b16854b5("conntrack: New userspace connection tracker.") One minor comment below, Acked-by: Daniele Di Proietto <diproiet...@vmware.com> > --- >

[ovs-dev] [PATCH 7/7] tests: Add learn action with limit tests.

Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> --- tests/learn.at | 175 + 1 file changed, 175 insertions(+) diff --git a/tests/learn.at b/tests/learn.at index 3f6fb5a7e..f91a662ad 100644 --- a/tests/learn.at +++ b

[ovs-dev] [PATCH 6/7] ofproto: Implement learning limit.

since it's done only for ukey with no traffic. Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> --- lib/odp-util.h | 6 -- ofproto/ofproto-dpif-upcall.c | 25 ++- ofproto/ofproto-dpif-xlate-cache.c | 6 +- ofproto/ofproto-dpif

[ovs-dev] [PATCH 1/7] ofp-actions: Factor out decode_LEARN_common().

No functional change, it will be used by next commit. Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> --- lib/ofp-actions.c | 77 +++ 1 file changed, 43 insertions(+), 34 deletions(-) diff --git a/lib/ofp-actions.c b/l

[ovs-dev] [PATCH 2/7] ofp-actions: Add limit to learn action.

and documentation. A later commit will implement the feature in ofproto-dpif. Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> --- include/openvswitch/ofp-actions.h | 12 lib/learn.c | 28 lib/ofp-actions.c

[ovs-dev] [PATCH 0/7] Learn action limit

implements the actual logic and the last commit has some basic tests. Daniele Di Proietto (7): ofp-actions: Factor out decode_LEARN_common(). ofp-actions: Add limit to learn action. ofproto-dpif-xlate: Create XC_LEARN entry after learning. ofproto: New cookie-counters module. ofproto-dpif-upcall

Re: [ovs-dev] [PATCH] netdev-dpdk: Fix rx_error stat for dpdk ports.

2017-02-16 7:31 GMT-08:00 Ian Stokes : > "rx_error" stat for a DPDK interface was calculated with the assumption that > dropped packets due to hardware buffer overload were counted as errors > in DPDK and the rte ierror stat included rte imissed packets i.e. > > rx_errors =

Re: [ovs-dev] [PATCH v10] dpif-netdev: Conditional EMC insert

2017-02-16 3:01 GMT-08:00 Kevin Traynor : > On 02/16/2017 10:22 AM, Ciara Loftus wrote: >> Unconditional insertion of EMC entries results in EMC thrashing at high >> numbers of parallel flows. When this occurs, the performance of the EMC >> often falls below that of the dpcls

Re: [ovs-dev] [PATCH] vswitchd: Move config_ofproto_types call before bridge_add_port

2017-02-15 10:02 GMT-08:00 Shashank Ram : > Currently, the call to config_ofproto_types() happens at the end > of bridge_reconfigure(), after missing ofprotos and ports are created. > However, it might be usefull to make this call before adding missing > ports through the dpif

Re: [ovs-dev] [PATCH] rhel/ifup: support vhost-user client mode

2017-02-10 9:49 GMT-08:00 Aaron Conole : > From: Aaron Conole > > This adds support for ifup to configure client-mode sockets by exposing > two new variables $OVS_PORT_MODE and $OVS_PORT_PATH to the ifcfg > scripts. When OVS_PORT_MODE is set to 'client',

Re: [ovs-dev] [PATCH 3/3] rhel: Fix ifup and ifdown after DPDK naming change.

On 02/02/2017 12:48, "Ben Pfaff" <b...@ovn.org> wrote: >On Tue, Jan 24, 2017 at 06:21:53PM -0800, Daniele Di Proietto wrote: >> Names like dpdk0 and dpdk1 are not enough to identify a DPDK interface. >> We could update README.RHEL.rst and add >>

Re: [ovs-dev] [PATCH v3] dpif-netdev: Conditional EMC insert

2017-01-31 9:55 GMT-08:00 Ciara Loftus : > Unconditional insertion of EMC entries results in EMC thrashing at high > numbers of parallel flows. When this occurs, the performance of the EMC > often falls below that of the dpcls classifier, rendering the EMC > practically

Re: [ovs-dev] [PATCH v2 1/1] doc: Remove "experimental" warning for userspace.

2017-02-03 9:38 GMT-08:00 Kevin Traynor : > On 02/02/2017 08:22 PM, Stokes, Ian wrote: >>> On 02/02/2017 04:44 PM, Ian Stokes wrote: Remove the experimental warning tag in documentation regarding OVS deployed via userspace. Signed-off-by: Ian Stokes

Re: [ovs-dev] [PATCH] doc: Update DPDK version for 2.7 release.

2017-02-02 8:59 GMT-08:00 Kevin Traynor : > On 02/02/2017 04:30 PM, Ian Stokes wrote: >> Add DPDK version required for the OVS 2.7 release in documentation. >> >> Signed-off-by: Ian Stokes > > Acked-by: Kevin Traynor Thanks, pushed

Re: [ovs-dev] [PATCH 2/3] rhel: Remove obsolete OVSDPDKVhostPort from ifdown script.

On 02/02/2017 11:49, "Ben Pfaff" <b...@ovn.org> wrote: >On Tue, Jan 24, 2017 at 06:21:52PM -0800, Daniele Di Proietto wrote: >> The support for vhost cuse port has been removed long ago. >> >> Fixes:419876444357("netdev-dpdk: Remove dpdkvhos

Re: [ovs-dev] [PATCH V2] netdev-dpdk: fix ifindex assignment for DPDK ports

On 31/01/2017 13:52, "Ben Pfaff" wrote: >On Thu, Dec 08, 2016 at 01:16:22PM +0100, Przemyslaw Lal wrote: >> diff --git a/lib/netdev-dpdk.c b/lib/netdev-dpdk.c >> index de78ddd..ef99eb3 100644 >> --- a/lib/netdev-dpdk.c >> +++ b/lib/netdev-dpdk.c >> @@ -2075,7 +2075,13 @@

Re: [ovs-dev] [PATCH 1/1] dpif-netdev: Conditional EMC insert

2017-01-26 9:55 GMT-08:00 Loftus, Ciara : >> >> 2017-01-25 7:52 GMT-08:00 Loftus, Ciara : >> >> 2017-01-22 11:45 GMT-08:00 Jan Scheurich : >> >> > >> >> >> It's not a big deal, since the most important use case we have for >> >>

Re: [ovs-dev] [PATCH 1/3] rhel: Fix ifdown for OVSDPDKBond.

On 26/01/2017 11:11, "Aaron Conole" <acon...@redhat.com> wrote: >Daniele Di Proietto <diproiet...@vmware.com> writes: > >> The OVSDPDKBond case wasn't handled in the rhel ifdown script. >> >> Fixes: f6bf8880613a ("rhel: Add support DPDK po

Re: [ovs-dev] [patch_v4 4/6] Unset CS_NEW for established connections.

2017-01-24 20:40 GMT-08:00 Darrell Ball : > Signed-off-by: Darrell Ball > --- > lib/conntrack.c | 1 + > 1 file changed, 1 insertion(+) > > diff --git a/lib/conntrack.c b/lib/conntrack.c > index 34728a6..aaecb00 100644 > --- a/lib/conntrack.c > +++

Re: [ovs-dev] [patch_v4 3/6] Userspace Datapath: Introduce NAT support.

2017-01-24 20:40 GMT-08:00 Darrell Ball : > This patch introduces NAT support for the userspace datapath. > The conntrack module changes are in this patch. > > The per packet scope of lookups for NAT and un_NAT is at > the bucket level rather than global. One hash table is >

Re: [ovs-dev] [patch_v4 2/6] Parse NAT netlink for userspace datapath.

2017-01-24 10:50 GMT-08:00 Darrell Ball : > Signed-off-by: Darrell Ball > --- > lib/conntrack-private.h | 9 -- > lib/conntrack.c | 3 +- > lib/conntrack.h | 31 +- > lib/dpif-netdev.c | 85 >

Re: [ovs-dev] [patch_v4 0/6] Userspace Datapath: Introduce NAT support.

2017-01-24 20:40 GMT-08:00 Darrell Ball : > This patch series introduces NAT support for the userspace datapath. > > The per packet scope of lookups for NAT and un_NAT is at > the bucket level rather than global. One hash table is > introduced to support create/delete handling.

Re: [ovs-dev] [PATCH v2] dpif-netdev: Conditional EMC insert

2017-01-26 9:51 GMT-08:00 Ciara Loftus : > Unconditional insertion of EMC entries results in EMC thrashing at high > numbers of parallel flows. When this occurs, the performance of the EMC > often falls below that of the dpcls classifier, rendering the EMC > practically

[ovs-dev] [PATCH] dpif-netdev: Pass Openvswitch other_config smap to dpif.

ofproto.c, ofproto-dpif.c and dpif.c. This patch sends the entire other_config column to dpif-netdev, so that dpif-netdev can extract the values it's interested in. No functional change. Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> --- I don't like that dpif-netdev receives the

Re: [ovs-dev] [PATCH] extract-ofp-fields: Define .TQ directive in nroff output.

2017-01-25 20:31 GMT-08:00 Ben Pfaff <b...@ovn.org>: > This missing directive caused groff warnings and probably some erroneous > output too. > > Fixes: 96fee5e0a2a0 ("ovs-fields: New manpage to document Open vSwitch and > OpenFlow fields.") > Reported-by: Daniel

Re: [ovs-dev] [PATCH] selinux: Allow creating tap devices.

On 25/01/2017 00:01, "Ansis Atteka" <ansisatt...@gmail.com> wrote: > > >On Jan 25, 2017 4:22 AM, "Daniele Di Proietto" <diproiet...@vmware.com> wrote: > >Current SELinux policy in RHEL and Fedora doesn't allow the creation of >TAP devices

[ovs-dev] [PATCH] selinux: Allow creating tap devices.

SELinux yet, but at least we can use the userspace datapath. Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> --- selinux/openvswitch-custom.te | 5 + 1 file changed, 5 insertions(+) diff --git a/selinux/openvswitch-custom.te b/selinux/openvswitch-custom.te index 47d

[ovs-dev] [PATCH 3/3] rhel: Fix ifup and ifdown after DPDK naming change.

. Fixes: 55e075e65ef9("netdev-dpdk: Arbitrary 'dpdk' port naming") Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> --- rhel/README.RHEL.rst| 13 + rhel/etc_sysconfig_network-scripts_ifup-ovs | 6 -- 2 files changed, 13 insertions

[ovs-dev] [PATCH 2/3] rhel: Remove obsolete OVSDPDKVhostPort from ifdown script.

The support for vhost cuse port has been removed long ago. Fixes:419876444357("netdev-dpdk: Remove dpdkvhostcuse ports") Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> --- rhel/etc_sysconfig_network-scripts_ifdown-ovs | 2 +- 1 file changed, 1 insertion(+), 1 deletio

Re: [ovs-dev] [PATCH 1/1] dpif-netdev: Conditional EMC insert

2017-01-22 11:45 GMT-08:00 Jan Scheurich : > >> It's not a big deal, since the most important use case we have for >> dpif-netdev is with dpdk, but I'd still like the code to behave >> similarly on different platforms. How about defining a function that >> uses random_uint32

Re: [ovs-dev] [PATCH v2] dpif-netdev: Change definitions of 'idle' & 'processing' cycles

2017-01-20 5:59 GMT-08:00 Jan Scheurich <jan.scheur...@web.de>: > > > On 2017-01-18 17:32, Kevin Traynor wrote: >> >> On 01/18/2017 01:34 AM, Daniele Di Proietto wrote: >>> >>> 2017-01-17 11:43 GMT-08:00 Kevin Traynor <ktray...@redhat.com>: &

Re: [ovs-dev] [PATCH] Documentation: Update DPDK doc after port naming change.

nstead of a port name. >> >> CC: Ciara Loftus <ciara.lof...@intel.com> >> Fixes: 55e075e65ef9("netdev-dpdk: Arbitrary 'dpdk' port naming") >> Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> > >Patch looks good.

Re: [ovs-dev] [PATCH] configuration.rst: Update the example of DPDK port's configuration

2017-01-18 15:18 GMT-08:00 Daniele Di Proietto <diproiet...@ovn.org>: > 2017-01-18 11:55 GMT-08:00 Binbin Xu <xu.binb...@zte.com.cn>: >> After the hotplug of DPDK ports, a valid dpdk-devargs must be >> specified. Otherwise, the DPDK device can't be available. >

[ovs-dev] [PATCH] Documentation: Update DPDK doc after port naming change.

ng") Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> --- Documentation/howto/dpdk.rst| 77 - Documentation/howto/userspace-tunneling.rst | 2 +- 2 files changed, 43 insertions(+), 36 deletions(-) diff --git a/Documentation/howto/dp

Re: [ovs-dev] [PATCH] configuration.rst: Update the example of DPDK port's configuration

2017-01-18 11:55 GMT-08:00 Binbin Xu : > After the hotplug of DPDK ports, a valid dpdk-devargs must be > specified. Otherwise, the DPDK device can't be available. > > Signed-off-by: Binbin Xu Thanks! Applied to master and branch-2.7 > --- >

Re: [ovs-dev] [PATCH v2] dpif-netdev: Change definitions of 'idle' & 'processing' cycles

2017-01-17 11:43 GMT-08:00 Kevin Traynor : > On 01/17/2017 05:43 PM, Ciara Loftus wrote: >> Instead of counting all polling cycles as processing cycles, only count >> the cycles where packets were received from the polling. > > This makes these stats much clearer. One minor

Re: [ovs-dev] [PATCH] dpif-netdev: Avoids repeated addition of DP_STAT_LOST.

if (OVS_UNLIKELY(!rules[i])) { > >Acked-by: Ben Pfaff <b...@ovn.org> > >I believe that this also should be tagged: > >CC: Daniele Di Proietto <diproiet...@vmware.com> >Fixes: 8aaa125dab66 ("dpif-netdev: Share emc and fast path output batches.") >

Re: [ovs-dev] [PATCH] netdev-vport: Do not log empty warnings on success.

On 12/01/2017 09:33, "Ben Pfaff" <b...@ovn.org> wrote: >On Thu, Jan 12, 2017 at 12:23:55AM -0800, Daniele Di Proietto wrote: >> set_tunnel_config() always logs a warning, even on success. This >> shouldn't happen. >> >> Without this, some unit tes

Re: [ovs-dev] broken tests

On 12/01/2017 09:24, "Ben Pfaff" wrote: >Commit 9fff138ec3a6dbe75073d16cba7fbe86ac273c36 "netdev: Add 'errp' to >set_config()." breaks the unit tests because netdev-vport now logs lots >of blank lines. I am unsure of the right fix--is it to just drop the >new VLOG_WARN call?

[ovs-dev] [PATCH] netdev-vport: Do not log empty warnings on success.

set_tunnel_config() always logs a warning, even on success. This shouldn't happen. Without this, some unit tests fail. Fixes: 9fff138ec3a6("netdev: Add 'errp' to set_config().") Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> --- lib/netdev-vport.c | 10 ++ 1

Re: [ovs-dev] [PATCH v2] netdev-dpdk: Assign socket id according to device's numa id

2017-01-12 6:18 GMT-08:00 Binbin Xu : > We can hotplug attach DPDK ports specified via the 'dpdk-devargs' > option now. > > But the socket id of DPDK ports can't be assigned correctly, > it is always 0. The socket id of DPDK ports should be assigned > according to the numa

Re: [ovs-dev] [PATCH v3] netdev-dummy: Limits the number of tx/rx queues.

2017-01-09 21:56 GMT-08:00 nickcooper-zhangtonghao : > This patch avoids the ovs_rcu to report WARN, caused by blocked > for a long time, when ovs-vswitchd processes a port with many > rx/tx queues. The number of tx/rx queues per port may be appropriate, > because the dpdk

[ovs-dev] [PATCH v3] dpdk: Late initialization.

the process (e.g. if there not enough memory), so the user must check the status of the process after setting dpdk-init to true. Nonetheless, I think this is an improvement, because it doesn't require restarting the whole unit. CC: Aaron Conole <acon...@redhat.com> Signed-off-by: Daniele Di Pr

Re: [ovs-dev] [PATCH 2/4] datapath: Limits the number of tx/rx queues for netdev-dummy.

()? I think for DPDK devices ultimately there's no way to check without actually setting up the queues, that's why it's done in reconfigure(). Thanks, Daniele > > Now we check it in dpdk_eth_dev_init(). > > Thanks. > > > > On Jan 9, 2017, at 11:22 AM, Daniele Di Proietto <

Re: [ovs-dev] [PATCH v2] dpdk: Late initialization.

mething more complicated and needs to be a separate patch, we should probably defer it, since we're on feature freeze now. Thanks, Daniele > > > >Thanks. >Nick > > > >On Jan 9, 2017, at 11:21 AM, Daniele Di Proietto <diproiet...@vmware.com> >wrote: > >Wit

Re: [ovs-dev] [PATCH v2] dpdk: Late initialization.

On 09/01/2017 07:14, "Aaron Conole" <acon...@redhat.com> wrote: >Daniele Di Proietto <diproiet...@vmware.com> writes: > >> With this commit, we allow the user to set other_config:dpdk-init=true >> after the process is started. This makes it easier

Re: [ovs-dev] [PATCH 4/4] datapath: Uses the OVS_CORE_UNSPEC instead of magic numbers.

2017-01-08 17:30 GMT-08:00 nickcooper-zhangtonghao : > This patch uses OVS_CORE_UNSPEC for the queue unpinned instead > of "-1". More important, the "-1" casted to unsigned int is > equal to NON_PMD_CORE_ID. We make the distinction between them. > > Signed-off-by:

Re: [ovs-dev] [PATCH 3/4] datapath: Uses the NR_QUEUE instead of magic numbers.

2017-01-08 17:30 GMT-08:00 nickcooper-zhangtonghao : > The NR_QUEUE is defined in "lib/dpif-netdev.h", netdev-dpdk > uses it instead of magic number. netdev-dummy should be > in the same case. > > Signed-off-by: nickcooper-zhangtonghao Thanks, I changed

Re: [ovs-dev] [PATCH 2/4] datapath: Limits the number of tx/rx queues for netdev-dummy.

2017-01-08 17:30 GMT-08:00 nickcooper-zhangtonghao : > This patch avoids the ovs_rcu to report WARN, caused by blocked > for a long time, when ovs-vswitchd processes a port with many > rx/tx queues. The number of tx/rx queues per port may be appropriate, > because the dpdk

[ovs-dev] [PATCH v2] dpdk: Late initialization.

the process (e.g. if there not enough memory), so the user must check the status of the process after setting dpdk-init to true. Nonetheless, I think this is an improvement, because it doesn't require restarting the whole unit. CC: Aaron Conole <acon...@redhat.com> Signed-off-by: Daniele Di Pr

Re: [ovs-dev] [PATCH 1/4] datapath: Fix formatting typo.

2017-01-08 17:30 GMT-08:00 nickcooper-zhangtonghao : > Signed-off-by: nickcooper-zhangtonghao Thanks, I changed the prefix to netdev-dpdk (instead of datapath) and pushed this to master > --- > lib/netdev-dpdk.c | 2 +- > 1 file changed, 1

[ovs-dev] [PATCH v3 16/18] dpif-netdev: Use hmap for poll_list in pmd threads.

, uses 'struct dp_netdev_rxq' from 'struct rxq_poll' and uses 'rx' for 'netdev_rxq' and 'rxq' for 'dp_netdev_rxq'. Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> --- lib/dpif-netdev.c | 168 -- 1 file changed, 112 insertions(

[ovs-dev] [PATCH v3 18/18] ovs-numa: Remove unused functions.

ovs-numa doesn't need to keep the state of the pmd threads, it is an implementation detail of dpif-netdev. Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> --- lib/ovs-numa.c | 175 - lib/ovs-numa.h | 7 --- 2 files c

[ovs-dev] [PATCH v3 11/18] dpctl: Avoid making assumptions on pmd threads.

is an implementation detail of dpif-netdev, dpctl shouldn't know anything about it. This commit changes the dpif_flow_put() and dpif_flow_del() calls to iterate over all the pmd threads, if pmd_id is PMD_ID_NULL. A simple test is added. Signed-off-by: Daniele Di Proietto <diproiet...@vmware.

[ovs-dev] [PATCH v3 15/18] ovs-numa: Add per numa and global counts in dump.

They will be used by a future commit. Suggested-by: Ilya Maximets <i.maxim...@samsung.com> Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> --- lib/ovs-numa.c | 96 +- lib/ovs-numa.h | 18 +-- 2 files

[ovs-dev] [PATCH v3 14/18] ovs-numa: Don't use hmap_first_with_hash().

I think it's better to iterate the hmap than to use hmap_first_with_hash(), because it handles hash collisions. Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> --- lib/ovs-numa.c | 26 ++ 1 file changed, 14 insertions(+), 12 deletions(-) diff --git a/l

[ovs-dev] [PATCH v3 12/18] ovs-numa: New ovs_numa_dump_contains_core() function.

It will be used by a future commit. struct ovs_numa_dump now uses an hmap instead of a list to make ovs_numa_dump_contains_core() more efficient. Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> --- lib/ovs-numa.c | 25 ++--- lib/ovs-numa.h | 10 ++-

[ovs-dev] [PATCH v3 10/18] dpif-netdev: Make 'static_tx_qid' const.

Since previous commit, 'static_tx_qid' doesn't need to be atomic and is actually never touched (except for initialization), so it can be made const. Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> --- lib/dpif-netdev.c | 11 +-- 1 file changed, 5 insertions(+), 6 del

[ovs-dev] [PATCH v3 09/18] dpif-netdev: Create pmd threads for every numa node.

will not consume any CPU, so it seems easier to create all the threads at once. This will also make future commits easier. Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> --- lib/dpif-netdev.c | 208 ++ tests/pmd.at | 2 +- 2

[ovs-dev] [PATCH v3 13/18] ovs-numa: Add new dump types.

They will be used by a future commit. This patch introduces some code duplication which will be removed in a future commit. Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> --- lib/ovs-numa.c | 78 ++ lib/ovs-numa.h | 4

[ovs-dev] [PATCH v3 08/18] dpif-netdev: Block pmd threads if there are no ports.

it's blocked) and set 'reload' to true. This is useful to avoid wasting CPU cycles and is also necessary for a future commit. Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> --- lib/dpif-netdev.c | 15 +++ 1 file changed, 15 insertions(+) diff --git a/lib/dpif-netd

[ovs-dev] [PATCH v3 07/18] dpif-netdev: Use a boolean instead of pmd->port_seq.

-by: Daniele Di Proietto <diproiet...@vmware.com> --- lib/dpif-netdev.c | 19 +++ 1 file changed, 7 insertions(+), 12 deletions(-) diff --git a/lib/dpif-netdev.c b/lib/dpif-netdev.c index 004b28dc8..0d47a3286 100644 --- a/lib/dpif-netdev.c +++ b/lib/dpif-netdev.c @@ -485,7

[ovs-dev] [PATCH v3 06/18] netdev-dpdk: Refactor construct and destruct.

common_destruct(). * Avoid taking 'dev->mutex' in construct and destruct: we're guaranteed to be the only thread with access to the object. Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> --- lib/netdev-dpdk.c | 86 ++- 1 file ch

[ovs-dev] [PATCH v3 03/18] dpif-netdev: Don't try to output on a device without txqs.

{push,pop}_header (tunnel devices), and ports that support netdev_send. With this we can also remove a branch in netdev_send(). This is also necessary for a future commit, which starts DPDK devices without txqs. Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> --- lib/dpif-netdev.

[ovs-dev] [PATCH v3 05/18] netdev-dpdk: Start also dpdkr devices only once on port-add.

patch also documents that, even though no change to the userspace datapath (the only user) is required. Lastly, this patch makes sure the errors returned by ofproto_port_add (which includes the first port reconfiguration) are reported back to the database. Signed-off-by: Daniele Di Proietto <diproiet.

[ovs-dev] [PATCH v3 04/18] netdev-dpdk: Don't call rte_dev_stop() in update_flags().

come from OpenFlow. Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> --- lib/netdev-dpdk.c | 28 1 file changed, 12 insertions(+), 16 deletions(-) diff --git a/lib/netdev-dpdk.c b/lib/netdev-dpdk.c index 8bb908691..2df3e220c 100644 --- a/lib/netdev-

[ovs-dev] [PATCH v3 01/18] dpif-netdev: Fix memory leak.

We keep all the per-port classifiers around, since they can be reused, but when a pmd thread is destroyed we should free them. Found using valgrind. Fixes: 3453b4d62a98("dpif-netdev: dpcls per in_port with sorted subtables") Signed-off-by: Daniele Di Proietto <diproiet...@vmware.

[ovs-dev] [PATCH v3 02/18] dpif-netdev: Take non_pmd_mutex to access tx cached ports.

As documented in dp_netdev_pmd_thread, we must take non_pmd_mutex to access the tx port caches for the non pmd thread. Found by inspection. Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> --- lib/dpif-netdev.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/lib/dpif-netd

[ovs-dev] [PATCH] netdev: Add 'errp' to set_config().

ory is "/var/log/openvswitch/". CC: Ciara Loftus <ciara.lof...@intel.com> CC: Kevin Traynor <ktray...@redhat.com> Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com> --- lib/netdev-dpdk.c | 27 ++ lib/

Re: [ovs-dev] [PATCH v3] dpif: Return ENODEV from dpif_port_query_by_*() if there's no port.

On 06/01/2017 13:01, "Ben Pfaff" <b...@ovn.org> wrote: >On Fri, Jan 06, 2017 at 12:42:35PM -0800, Daniele Di Proietto wrote: >> bridge_delete_or_reconfigure() deletes every interface that's not dumped >> by OFPROTO_PORT_FOR_EACH(). ofprot

Re: [ovs-dev] [PATCH v2] dpif: Return ENODEV from dpif_port_query_by_name() if there's no port.

On 06/01/2017 11:34, "Ben Pfaff" <b...@ovn.org> wrote: >On Fri, Jan 06, 2017 at 10:59:07AM -0800, Daniele Di Proietto wrote: >> bridge_delete_or_reconfigure() deletes every interface that's not dumped >> by OFPROTO_PORT_FOR_EACH(). ofprot

Re: [ovs-dev] [PATCH] ofproto-dpif: Continue port dump if a port is missing from dpif-netdev.

On 06/01/2017 09:28, "Ben Pfaff" <b...@ovn.org> wrote: >On Thu, Jan 05, 2017 at 08:37:26PM -0800, Daniele Di Proietto wrote: >> bridge_delete_or_reconfigure() deletes every interface that's not dumped >> by OFPROTO_PORT_FOR_EACH(). ofprot

  1   2   >