Re: [ovs-dev] [PATCH v2 0/3]: dpif-netdev: Detailed PMD performance metrics and supervision

I submitted a rebased version of this series yesterday. Please look at the new v3 series instead. /Jan > -Original Message- > From: Jan Scheurich > Sent: Wednesday, 18 October, 2017 12:46 > To: 'ovs-dev@openvswitch.org' > Cc: Jan Scheurich

Re: [ovs-dev] [PATCH V2 3/4] tc: Add header rewrite using tc pedit action

On Tue, Nov 21, 2017 at 07:50:24AM +0200, Roi Dayan wrote: > > > On 20/11/2017 14:30, Simon Horman wrote: > > On Sun, Nov 19, 2017 at 08:45:19AM +0200, Roi Dayan wrote: > > > > > > > > > On 16/11/2017 18:29, Simon Horman wrote: > > > > On Wed, Oct 25, 2017 at 02:24:15PM +0300, Roi Dayan wrote:

Re: [ovs-dev] [PATCH v2] odp-execute: Skip processing actions when batch is emptied

Do anyone sees any issue with this change ? Though it does not address any existing bug, but will make the code less error prone. Regards, Vishal -Original Message- From: ovs-dev-boun...@openvswitch.org [mailto:ovs-dev-boun...@openvswitch.org] On Behalf Of Vishal Deep Ajmera Sent:

Re: [ovs-dev] [PATCH] dpif-netdev: Refactor datapath flow cache

Hi Jan, Thanks, that's a really interesting patch. Currently does not apply to head of master - what rev can I apply it to? Some more below - including one way down in the code. Thanks, /Billy. > -Original Message- > From: ovs-dev-boun...@openvswitch.org [mailto:ovs-dev- >

Re: [ovs-dev] [PATCH] dpif-netdev: Refactor datapath flow cache

Hi Billy, Thanks for your feedback. I continue to have problems with sending out patches via mail from our company network. This time it seems it is the "Page Feed" (^L) character in the dpif-netdev.c file that got corrupted in the emailed version of the patch. I wish there was a way to

[ovs-dev] [PATCH 4/7] netdev-tc-offloads: Remove redundant brackets

From: Paul Blakey Signed-off-by: Paul Blakey Reviewed-by: Roi Dayan --- lib/tc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/lib/tc.c b/lib/tc.c index 360a770..4f82623 100644 --- a/lib/tc.c +++ b/lib/tc.c @@

[ovs-dev] [PATCH 7/7] tc: Send csum action only if we need to update csum

From: Paul Blakey Currently we send the tc csum action even if it's not needed. Fix that by sending it only if csum update flags isn't zero. Signed-off-by: Paul Blakey Reviewed-by: Roi Dayan --- lib/tc.c | 8 +--- 1 file changed,

[ovs-dev] [PATCH 2/7] netdev-tc-offloads: Fix accidental skipping of extended pedit keys

From: Paul Blakey We only support extended pedit keys for now, so it's the type we expect. Skip the legacy ones instead. Signed-off-by: Paul Blakey Reviewed-by: Roi Dayan --- lib/tc.c | 2 +- 1 file changed, 1 insertion(+), 1

[ovs-dev] [PATCH 6/7] tc: Fix wrong struct variable order

From: Paul Blakey Fix the struct variable order to corrospond with it's usage. Signed-off-by: Paul Blakey Reviewed-by: Roi Dayan --- lib/tc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/lib/tc.c b/lib/tc.c

[ovs-dev] [PATCH 3/7] netdev-tc-offloads: Verify csum flags on dump from tc

From: Paul Blakey On dump, parse and verify the tc csum action update flags in the same way as we put them. Signed-off-by: Paul Blakey Reviewed-by: Roi Dayan --- lib/tc.c | 117

Re: [ovs-dev] OVS DPDK: dpdk_merge pull request

> On Fri, Nov 17, 2017 at 07:27:51PM +, Stokes, Ian wrote: > > The following changes since commit > 1ae83bb20677b42d63dbb2140fa8ed3144c6260f: > > > > netdev-tc-offloads: Add support for action set (2017-11-16 08:10:29 > > -0800) > > > > are available in the git repository at: > > > >

Re: [ovs-dev] [PATCH 1/2] ovn-ctl: Add -vfile:info option to OVN_NB/SB_LOG options

Thank you! On Nov 20, 2017 10:07 PM, "Ben Pfaff" wrote: > Yes, of course. I didn't get much work done last week with ovscon > going. > > On Mon, Nov 20, 2017 at 09:11:41PM +0530, Numan Siddique wrote: > > A gentle ping. Ben, is it possible to have a look at this patch set - > >

[ovs-dev] [PATCH 1/7] netdev-tc-offloads: Fix travis compilation error

From: Paul Blakey Travis complains about variable length array usage, use a fixed size instead. Signed-off-by: Paul Blakey Reviewed-by: Roi Dayan --- lib/netdev-tc-offloads.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)

[ovs-dev] [PATCH 0/7] Fixes for header rewrite feature

Hi, The following series fixes some issues with header rewrite and some small refactoring. Thanks, Roi Paul Blakey (7): netdev-tc-offloads: Fix travis compilation error netdev-tc-offloads: Fix accidental skipping of extended pedit keys netdev-tc-offloads: Verify csum flags on dump from

[ovs-dev] [RFC PATCH v3 0/8] netdev-dpdk: support multi-segment mbufs

Overview This patchset introduces support for multi-segment mbufs to OvS-DPDK. Multi-segment mbufs are typically used when the size of an mbuf is insufficient to contain the entirety of a packet's data. Instead, the data is split across numerous mbufs, each carrying a portion, or

[ovs-dev] [RFC PATCH v3 3/8] lib/dp-packet: copy mbuf info for packet copy

From: Michael Qiu Currently, when doing packet copy, lots of DPDK mbuf's info will be missed, like packet type, ol_flags, etc. Those information is very important for DPDK to do packets processing. Signed-off-by: Michael Qiu [mark.b.kavan...@intel.com

Re: [ovs-dev] [PATCH v2 1/9] tests: Convert uuidfilt utility from Perl to Python.

Ben Pfaff writes: > Perl is unfashionable and Python is more widely available and understood, > so this commit converts one of the OVS uses of Perl into Python. > > Signed-off-by: Ben Pfaff > --- > tests/automake.mk| 5 +- > tests/ofproto-macros.at

[ovs-dev] [RFC PATCH v3 8/8] netdev-dpdk: support multi-segment jumbo frames

Currently, jumbo frame support for OvS-DPDK is implemented by increasing the size of mbufs within a mempool, such that each mbuf within the pool is large enough to contain an entire jumbo frame of a user-defined size. Typically, for each user-defined MTU, 'requested_mtu', a new mempool is created,

Re: [ovs-dev] [PATCH] datapath-windows: Account for VLAN tag in tunnel Decap

Acked-by: Anand Kumar Thanks, Anand Kumar On 11/20/17, 3:06 PM, "ovs-dev-boun...@openvswitch.org on behalf of Shashank Ram" wrote: Decap functions for tunneling protocols do not compute the packet

[ovs-dev] [RFC PATCH v3 6/8] lib/dp-packet: copy data from multi-seg. DPDK mbuf

From: Michael Qiu When doing packet clone, if packet source is from DPDK driver, multi-segment must be considered, and copy the segment's data one by one. Signed-off-by: Michael Qiu Signed-off-by: Mark Kavanagh ---

[ovs-dev] [RFC PATCH v3 7/8] netdev-dpdk: copy large packet to multi-seg. mbufs

From: Michael Qiu Currently, packets are only copied to a single segment in the function dpdk_do_tx_copy(). This could be an issue in the case of jumbo frames, particularly when multi-segment mbufs are involved. This patch calculates the number of segments needed by a packet

[ovs-dev] [RFC PATCH v3 5/8] lib/dp-packet: fix dp_packet_put_uninit for multi-seg mbufs

dp_packet_put_uninit(dp_packet, size) appends 'size' bytes to the tail of a dp_packet. In the case of multi-segment mbufs, it is the data length of the last mbuf in the mbuf chain that should be adjusted by 'size' bytes. In its current implementation, dp_packet_put_uninit() adjusts the

Re: [ovs-dev] [PATCH v2 0/9] convert Perl to Python

Ben Pfaff writes: > This series gets rid of all usage of Perl in the tree, converting > it to Python instead. > > v1->v2: > * Minor fixes from Aaron Conole. > * Break soexpand and sodepends common code into library (suggested by > Aaron Conole) > > Ben Pfaff (9): >

[ovs-dev] [RFC PATCH v3 4/8] lib/dp-packet: Fix data_len issue with multi-segs

From: Michael Qiu When a packet is from DPDK source, and it contains multiple segments, data_len is not equal to the packet size. This patch fixes this issue. Co-authored-by: Mark Kavanagh Co-authored-by: Przemyslaw Lal

[ovs-dev] Gestión de Tesorería

Liquidez en equilibrio para hacer frente a los retos financieros Gestión de Tesorería: Importancia ante una crisis financiera 24 de noviembre - C.P. y MAN. Fernando García Zárate 9am-6pm Obtendrá las herramientas necesarias que le permitirán hacer frente a los retos en caso de una crisis

Re: [ovs-dev] [PATCH v2 0/5] dpif-netdev: Cuckoo-Distributor implementation

Hi, Jan, Thanks for sharing the patch and nice talking with you in the conference. Could you tell us more about the test case you used such as the rule set and traffic pattern? Since we have some other work at hand to be done soon, we will definitely test the patch later and get back to you in

[ovs-dev] [PATCH 5/7] netdev-tc-offloads: Remove redundant loop handling ovs action set

From: Paul Blakey OVS action set always has a single nested OVS_KEY_ATTR_* so there is no need to iterate it. Signed-off-by: Paul Blakey Reviewed-by: Roi Dayan --- lib/netdev-tc-offloads.c | 90

Re: [ovs-dev] [PATCH 2/2] dpctl: Support flush conntrack by conntrack 5-tuple

On Wed, Nov 15, 2017 at 5:28 PM, Justin Pettit wrote > > > > With this patch, ovs-dpctl flush-conntrack accepts a conntrack 5-tuple > > I'd also mention ovs-appctl, since ovs-dpctl doesn't work on all platforms. > Sure. > diff --git a/NEWS b/NEWS > > index

[ovs-dev] [PATCH v2 0/3] Support conntrack flush by ct 5-tuple

This patch series add support to delete a connection tracking entry specified by a 5-tuple. v1 -> v2: * Incoporate suggestions from Justin * Add dpif-netdev implementation * Improve error reporting logic in dpctl.c * Support ICMP 5-tuple * Rebase to master Yi-Hung Wei (3):

[ovs-dev] [PATCH v2 3/3] dpctl: Support flush conntrack by conntrack 5-tuple

With this patch, "flush-conntrack" in ovs-dpctl and ovs-appctl accept a conntrack 5-tuple to delete the conntrack entry specified by the 5-tuple. For example, user can use the following command to flush a conntrack entry in zone 5. $ ovs-dpctl flush-conntrack zone=5 \

[ovs-dev] [PATCH v1] datapath-windows: Add support for deleting conntrack entry by 5-tuple.

To delete a conntrack entry specified by 5-tuple pass an additional conntrack 5-tuple parameter to flush-conntrack. Signed-off-by: Anand Kumar --- datapath-windows/ovsext/Conntrack.c | 146 +--- 1 file changed, 134 insertions(+), 12

[ovs-dev] [PATCH v1] Add support to delete a conntrack entry by 5-tuple

This patch adds support for deleting conntrack entry by conntrack 5-tuple in windows kernel. The related userspace changes are currrently being reviewed https://patchwork.ozlabs.org/project/openvswitch/list/?series=13345 The windows kernel changes can go in independent of the userspace changes

Re: [ovs-dev] [PATCH 1/2] ct-dpif, dpif-netlink: Support conntrack flush by ct 5-tuple

Thanks Justin for reviewing this series. I will repin v2 based on your comments. On Wed, Nov 15, 2017 at 4:28 PM, Justin Pettit wrote: > > --- a/lib/ct-dpif.c > > +++ b/lib/ct-dpif.c > > @@ -111,19 +111,30 @@ ct_dpif_dump_done(struct ct_dpif_dump_state *dump) > > } > > > > /*

[ovs-dev] [PATCH v2 1/3] ct-dpif, dpif-netlink: Support conntrack flush by ct 5-tuple

This patch adds support of flushing a conntrack entry specified by the conntrack 5-tuple, and provides the implementation in dpif-netlink. The implementation of dpif-netlink in the linux datapath utilizes the NFNL_SUBSYS_CTNETLINK netlink subsystem to delete a conntrack entry in nf_conntrack.

[ovs-dev] [PATCH v2 2/3] conntrack: Support conntrack flush by ct 5-tuple

This patch adds support of flushing a conntrack entry specified by the conntrack 5-tuple in dpif-netdev. Signed-off-by: Yi-Hung Wei --- lib/conntrack.c | 64 +++ lib/conntrack.h | 3 +++ lib/dpif-netdev.c | 2 +- 3