Re: [ovs-dev] [PATCH V2] dpif-netdev: Do RCU synchronization at fixed interval in PMD main loop.

2019-10-03 Thread Nitin Katiyar via dev
Hi, Can you please review the following patch and provide the feedback? Regards, Nitin > -Original Message- > From: Nitin Katiyar > Sent: Wednesday, August 28, 2019 2:00 PM > To: ovs-dev@openvswitch.org > Cc: Anju Thomas ; Ilya Maximets > > Subject: RE: [PATCH V2] dpif-netdev: Do RCU syn

Re: [ovs-dev] [PATCH v1 3/3 ovn] OVN: Use replace_src_ip and replace_dst_ip for NAT rules

2019-10-03 Thread 0-day Robot
Bleep bloop. Greetings Ankur Sharma, I am a robot and I have tried out your patch. Thanks for your contribution. I encountered some error that I wasn't expecting. See the details below. checkpatch: WARNING: Line is 80 characters long (recommended limit is 79) #30 FILE: northd/ovn-northd.8.xml

Re: [ovs-dev] [PATCH v1 2/3 ovn] OVN: ADD nbctl cli to mark a dnat_and_snat rule as stateless

2019-10-03 Thread 0-day Robot
Bleep bloop. Greetings Ankur Sharma, I am a robot and I have tried out your patch. Thanks for your contribution. I encountered some error that I wasn't expecting. See the details below. checkpatch: WARNING: Line is 220 characters long (recommended limit is 79) #121 FILE: utilities/ovn-nbctl.8

Re: [ovs-dev] [PATCH v2] tests: Add check for correct l3l4 conntrack frag reassembly

2019-10-03 Thread Darrell Ball
On Thu, Oct 3, 2019 at 6:56 PM Darrell Ball wrote: > Thanks for the patch > > This approach will not work for the userspace datapath > > Few issues off the top of my head: > > 1/ packet-out frees the packet (which is a fragment in this case) after > completion >hence multiple packet-outs need

Re: [ovs-dev] [PATCH v2] tests: Add check for correct l3l4 conntrack frag reassembly

2019-10-03 Thread Darrell Ball
Thanks for the patch This approach will not work for the userspace datapath Few issues off the top of my head: 1/ packet-out frees the packet (which is a fragment in this case) after completion hence multiple packet-outs need to be part of a single Openflow bundle command as in other similar

[ovs-dev] [PATCH v1 2/3 ovn] OVN: ADD nbctl cli to mark a dnat_and_snat rule as stateless

2019-10-03 Thread Ankur Sharma
Adding ovn-nbctl to mark a dnat_and_snat rule as stateless. This configuration will added to "options" column of NAT table. Signed-off-by: Ankur Sharma --- ovn-nb.ovsschema | 6 -- ovn-nb.xml| 5 + tests/ovn-nbctl.at| 29 + u

[ovs-dev] [PATCH v1 3/3 ovn] OVN: Use replace_src_ip and replace_dst_ip for NAT rules

2019-10-03 Thread Ankur Sharma
For dnat_and_snat rules which are meant to be stateless instead of using ct_snat/dnat OVN actions, we will use replace_src/dst_ip actions. This actions will do 1:1 mapping to inner ip to dest ip, while recalculating the checksums. Signed-off-by: Ankur Sharma --- northd/ovn-northd.8.xml | 34 +++

[ovs-dev] [PATCH v1 1/3 ovn] OVN: ADD new ovn actions to replace source and destination ip

2019-10-03 Thread Ankur Sharma
This patch adds 2 new ovn actions, "replace_src_ip" and "replace_dst_ip" These actions map to mod_nw_src and mod_nw_dst OVS actions respectively. These actions simply replace the source and destination IPs of the packets, while changing the L3 and L4 checksums. Signed-off-by: Ankur Sharma --- i

[ovs-dev] [PATCH v1 0/3] ALLOW Stateless NAT operations

2019-10-03 Thread Ankur Sharma
NAT implementation in OVN uses connection tracker to replace source and dest ips. This implementation works fine and is the right approach for cases where external ips are shared (i.e. SNAT) or where we replace ip only when relevant flow is there (i.e. DNAT). However, it opens the possibility of D

Re: [ovs-dev] [PATCH] travis: Fix 32-bit libunwind system build.

2019-10-03 Thread William Tu
On Thu, Oct 3, 2019 at 10:15 AM Ilya Maximets wrote: > > On 03.10.2019 18:13, Ilya Maximets wrote: > > On 02.10.2019 20:15, William Tu wrote: > >> 32-bit and 64-bit libunwind can not be installed at the same time. > >> For 32-bit build, this patch removes the 64-bit libunwind and install > >> 32-b

Re: [ovs-dev] [PATCH v1 ovn] ovn-nb/sbctl.c: Set no-leader-only as default for clustered dbs

2019-10-03 Thread aginwala
Thanks Han and ben for the suggestions. However, one more reason for using this approach is because ovn-controller uses no-leader-only by default so that all the chassis can be randomly distributed to talk to any node in the cluster to avoid overloading leader node in a large scale env "ovsdb_idl_s

Re: [ovs-dev] [PATCH v1 ovn] ovn-nb/sbctl.c: Set no-leader-only as default for clustered dbs

2019-10-03 Thread Han Zhou
On Thu, Oct 3, 2019 at 11:35 AM Ben Pfaff wrote: > > On Thu, Oct 03, 2019 at 10:19:35AM -0700, Han Zhou wrote: > > On Thu, Oct 3, 2019 at 9:09 AM Ben Pfaff wrote: > > > > > > On Tue, Oct 01, 2019 at 05:05:28PM -0700, amgin...@gmail.com wrote: > > > > From: Aliasgar Ginwala > > > > > > > > When u

Re: [ovs-dev] [PATCH v2 1/2] sparse: Get rid of obsolete rte_flow header.

2019-10-03 Thread 0-day Robot
Bleep bloop. Greetings David Marchand, I am a robot and I have tried out your patch. Thanks for your contribution. I encountered some error that I wasn't expecting. See the details below. git-am: Failed to merge in the changes. Patch failed at 0001 sparse: Get rid of obsolete rte_flow header.

[ovs-dev] [PATCH] doc: Fix incorrect reference for dpdk-testpmd.

2019-10-03 Thread David Marchand
Move back the dpdk-testpmd reference to the right section of this document so that the link in howto/dpdk does not point to "vhost-user-client tx retries config". Fixes: 080f080c3bc1 ("netdev-dpdk: Enable tx-retries-max config.") Signed-off-by: David Marchand --- The problem can be checked by lo

Re: [ovs-dev] [PATCH v2] tests: Add check for correct l3l4 conntrack frag reassembly

2019-10-03 Thread Ben Pfaff
On Thu, Oct 03, 2019 at 11:09:35AM -0700, Gregory Rose wrote: > I had a lot of fun writing that little test - figuring out the correct > placement of those '[]' brackets caused some > consternation for me to say the least.  But it gave me a chance to > read up on the autom4te documentation. m4 and

Re: [ovs-dev] [PATCH v1 ovn] ovn-nb/sbctl.c: Set no-leader-only as default for clustered dbs

2019-10-03 Thread Ben Pfaff
On Thu, Oct 03, 2019 at 10:19:35AM -0700, Han Zhou wrote: > On Thu, Oct 3, 2019 at 9:09 AM Ben Pfaff wrote: > > > > On Tue, Oct 01, 2019 at 05:05:28PM -0700, amgin...@gmail.com wrote: > > > From: Aliasgar Ginwala > > > > > > When using ovn-nb/sbctl running in cluster, one can use local > > > sock

Re: [ovs-dev] [PATCH v2 1/2] sparse: Get rid of obsolete rte_flow header.

2019-10-03 Thread David Marchand
On Thu, Oct 3, 2019 at 8:12 PM David Marchand wrote: > > This header had been copied to cope with issues on the dpdk side. > Now that the problems have been fixed [1], let's drop this file as it is > now out of sync with dpdk. > > 1: https://git.dpdk.org/dpdk/commit/?id=fbb25a3878cc > > Signed-off

[ovs-dev] [PATCH v2 1/2] sparse: Get rid of obsolete rte_flow header.

2019-10-03 Thread David Marchand
This header had been copied to cope with issues on the dpdk side. Now that the problems have been fixed [1], let's drop this file as it is now out of sync with dpdk. 1: https://git.dpdk.org/dpdk/commit/?id=fbb25a3878cc Signed-off-by: David Marchand --- include/sparse/automake.mk |1 - inclu

[ovs-dev] [PATCH v2 2/2] travis: Switch to dpdk master branch.

2019-10-03 Thread David Marchand
Make this branch point to current dpdk master branch so that we can track API breakage. Signed-off-by: David Marchand --- Changelog since v1: - since building DPDK is controlled by either setting DPDK or DPDK_SHARED variables, we can set DPDK_VER globally (Ilya), --- .travis.yml | 4 +++- 1 f

Re: [ovs-dev] [PATCH v2] tests: Add check for correct l3l4 conntrack frag reassembly

2019-10-03 Thread Gregory Rose
On 10/3/2019 11:07 AM, Ben Pfaff wrote: On Thu, Oct 03, 2019 at 10:46:09AM -0700, Greg Rose wrote: Two commits recently fixed an issue with setting the corrrect l3 and l4 flow information when conntrack reassembles packet fragments. c98f776 datapath: Clear the L4 portion of the key for "later"

Re: [ovs-dev] [PATCH v2] tests: Add check for correct l3l4 conntrack frag reassembly

2019-10-03 Thread Ben Pfaff
On Thu, Oct 03, 2019 at 10:46:09AM -0700, Greg Rose wrote: > Two commits recently fixed an issue with setting the corrrect l3 and l4 > flow information when conntrack reassembles packet fragments. > > c98f776 datapath: Clear the L4 portion of the key for "later" fragments > 2609173 datapath: Prope

Re: [ovs-dev] [PATCH] ci: Get rid of OVS_CFLAGS in CI scripts.

2019-10-03 Thread Ben Pfaff
On Thu, Oct 03, 2019 at 07:22:56PM +0200, Ilya Maximets wrote: > On 03.10.2019 15:36, Ben Pfaff wrote: > > On Thu, Oct 03, 2019 at 05:47:29PM +0200, Ilya Maximets wrote: > > > Our CI scripts uses OVS_CFLAGS variable that is intended for internal > > > usage by 'configure' script only. Usual CFLAGS

Re: [ovs-dev] [PATCH] tests: Add check for correct l3l4 conntrack frag reassembly

2019-10-03 Thread Gregory Rose
On 10/3/2019 6:08 AM, Ben Pfaff wrote: On Thu, Oct 03, 2019 at 08:35:42AM -0700, Greg Rose wrote: Two commits recently fixed an issue with setting the corrrect l3 and l4 flow information when conntrack reassembles packet fragments. c98f776 datapath: Clear the L4 portion of the key for "later"

[ovs-dev] [PATCH v2] tests: Add check for correct l3l4 conntrack frag reassembly

2019-10-03 Thread Greg Rose
Two commits recently fixed an issue with setting the corrrect l3 and l4 flow information when conntrack reassembles packet fragments. c98f776 datapath: Clear the L4 portion of the key for "later" fragments 2609173 datapath: Properly set L4 keys on "later" IP fragments This test checks for regress

Re: [ovs-dev] [PATCH] ci: Get rid of OVS_CFLAGS in CI scripts.

2019-10-03 Thread Ilya Maximets
On 03.10.2019 15:36, Ben Pfaff wrote: On Thu, Oct 03, 2019 at 05:47:29PM +0200, Ilya Maximets wrote: Our CI scripts uses OVS_CFLAGS variable that is intended for internal usage by 'configure' script only. Usual CFLAGS should be used instead to avoid giving bad example to users. Additionally, '

Re: [ovs-dev] [PATCH v1 ovn] ovn-nb/sbctl.c: Set no-leader-only as default for clustered dbs

2019-10-03 Thread Han Zhou
On Thu, Oct 3, 2019 at 9:09 AM Ben Pfaff wrote: > > On Tue, Oct 01, 2019 at 05:05:28PM -0700, amgin...@gmail.com wrote: > > From: Aliasgar Ginwala > > > > When using ovn-nb/sbctl running in cluster, one can use local > > socket to run different commands. It is very inconvenient to pass > > no-lea

Re: [ovs-dev] [PATCH] travis: Fix 32-bit libunwind system build.

2019-10-03 Thread Ilya Maximets
On 03.10.2019 18:13, Ilya Maximets wrote: On 02.10.2019 20:15, William Tu wrote: 32-bit and 64-bit libunwind can not be installed at the same time. For 32-bit build, this patch removes the 64-bit libunwind and install 32-bit version. Tested-at: https://travis-ci.org/williamtu/ovs-travis/builds/

Re: [ovs-dev] [PATCH net-next 0/9] optimize openvswitch flow looking up

2019-10-03 Thread William Tu
Hi Tonghao, Thanks for the patch. > On 29 Sep 2019, at 19:09, xiangxia.m@gmail.com wrote: > > > From: Tonghao Zhang > > > > This series patch optimize openvswitch. > > > > Patch 1, 2, 4: Port Pravin B Shelar patches to > > linux upstream with little changes. > > I thought the idea of adding

Re: [ovs-dev] [PATCH] ci: Get rid of OVS_CFLAGS in CI scripts.

2019-10-03 Thread Ben Pfaff
On Thu, Oct 03, 2019 at 05:47:29PM +0200, Ilya Maximets wrote: > Our CI scripts uses OVS_CFLAGS variable that is intended for internal > usage by 'configure' script only. Usual CFLAGS should be used instead > to avoid giving bad example to users. > > Additionally, '-m32' flag passed directly to C

Re: [ovs-dev] [PATCH 1/1] ofproto-dpif-xlate: fixed memory leak in is_neighbor_reply_correct() function

2019-10-03 Thread Ben Pfaff
On Thu, Oct 03, 2019 at 10:15:59AM +0200, Damijan Skvarc wrote: > memory leak happens while calling netdev_get_addr_list() function. This > function allocates memory for ip_addr and mask output arguments, but > this memory is never freed. > > Signed-off-by: Damijan Skvarc Thank you. I applied t

Re: [ovs-dev] [dpdk-latest PATCH 2/2] travis: Switch to dpdk master branch.

2019-10-03 Thread David Marchand
On Thu, Oct 3, 2019 at 6:00 PM Ilya Maximets wrote: > > On 01.10.2019 14:47, David Marchand wrote: > > Make this branch point to current dpdk master branch so that we can > > track API breakage and benefit from the latest enhancements. > > > > Signed-off-by: David Marchand > > --- > > .travis.y

Re: [ovs-dev] [PATCH] tests: Add check for correct l3l4 conntrack frag reassembly

2019-10-03 Thread Gregory Rose
On 10/3/2019 6:08 AM, Ben Pfaff wrote: On Thu, Oct 03, 2019 at 08:35:42AM -0700, Greg Rose wrote: Two commits recently fixed an issue with setting the corrrect l3 and l4 flow information when conntrack reassembles packet fragments. c98f776 datapath: Clear the L4 portion of the key for "later"

[ovs-dev] Invitation to Quote/Tender

2019-10-03 Thread IcmGrouping
GoodDay, We have an urgent requirement as per attached samples & specifications. Kindly quote your price and inquiry at the earliest .Also confirm the Terms & Condition included, Delivery schedule for supply. Looking forward to your valuable confirmation. sincerely yours -- Thanks Ar

Re: [ovs-dev] [PATCH] tests: Add check for correct l3l4 conntrack frag reassembly

2019-10-03 Thread Ben Pfaff
On Thu, Oct 03, 2019 at 08:35:42AM -0700, Greg Rose wrote: > Two commits recently fixed an issue with setting the corrrect l3 and l4 > flow information when conntrack reassembles packet fragments. > > c98f776 datapath: Clear the L4 portion of the key for "later" fragments > 2609173 datapath: Prope

Re: [ovs-dev] [PATCH] travis: Fix 32-bit libunwind system build.

2019-10-03 Thread Ilya Maximets
On 02.10.2019 20:15, William Tu wrote: 32-bit and 64-bit libunwind can not be installed at the same time. For 32-bit build, this patch removes the 64-bit libunwind and install 32-bit version. Tested-at: https://travis-ci.org/williamtu/ovs-travis/builds/592649311 Fixes: e2ed6fbeb18c ("fatal-signa

Re: [ovs-dev] [PATCH v1 ovn] ovn-nb/sbctl.c: Set no-leader-only as default for clustered dbs

2019-10-03 Thread Ben Pfaff
On Tue, Oct 01, 2019 at 05:05:28PM -0700, amgin...@gmail.com wrote: > From: Aliasgar Ginwala > > When using ovn-nb/sbctl running in cluster, one can use local > socket to run different commands. It is very inconvenient to pass > no-leader-only in different tools using ovn-nb/sbctl instead of > al

Re: [ovs-dev] [dpdk-latest PATCH 2/2] travis: Switch to dpdk master branch.

2019-10-03 Thread Ilya Maximets
On 01.10.2019 14:47, David Marchand wrote: Make this branch point to current dpdk master branch so that we can track API breakage and benefit from the latest enhancements. Signed-off-by: David Marchand --- .travis.yml | 8 1 file changed, 4 insertions(+), 4 deletions(-) diff --git

Re: [ovs-dev] [PATCH] tests: Add check for correct l3l4 conntrack frag reassembly

2019-10-03 Thread 0-day Robot
Bleep bloop. Greetings Greg Rose, I am a robot and I have tried out your patch. Thanks for your contribution. I encountered some error that I wasn't expecting. See the details below. git-am: fatal: corrupt patch at line 43 Repository lacks necessary blobs to fall back on 3-way merge. Cannot fa

[ovs-dev] [PATCH] ci: Get rid of OVS_CFLAGS in CI scripts.

2019-10-03 Thread Ilya Maximets
Our CI scripts uses OVS_CFLAGS variable that is intended for internal usage by 'configure' script only. Usual CFLAGS should be used instead to avoid giving bad example to users. Additionally, '-m32' flag passed directly to CC variable to avoid splitting it from the compiler invocations and force

[ovs-dev] [PATCH] tests: Add check for correct l3l4 conntrack frag reassembly

2019-10-03 Thread Greg Rose
Two commits recently fixed an issue with setting the corrrect l3 and l4 flow information when conntrack reassembles packet fragments. c98f776 datapath: Clear the L4 portion of the key for "later" fragments 2609173 datapath: Properly set L4 keys on "later" IP fragments This test checks for regress

[ovs-dev] Herramientas que optimicen el ciclo de venta

2019-10-03 Thread Entrenamiento para vendedores 4.0
08 de Noviembre | Horario de 10 A 14 Y DE 15:30 A 19:30 HRS | (hora del centro de México) | PRESENCIAL - CDMX: Entrenamiento para vendedores 4.0 - Conoceremos una metodología eficaz de ventas para manejarnos de manera efectiva utilizando comunicación, información y herramientas que se adecúen

[ovs-dev] Invitation to Quote/Tender

2019-10-03 Thread Pernood Ricard
GoodDay, We have an urgent requirement as per attached samples & specifications. Kindly quote your price and inquiry at the earliest .Also confirm the Terms & Condition included, Delivery schedule for supply. Looking forward to your valuable confirmation. sincerely yours -- Thanks Ar

[ovs-dev] Радиоспектакли, рассказы, повести и пьесы – весь золотой фонд в одной уникальной коллекции и в отличном качестве. 13_05_2019 02_03 198707

2019-10-03 Thread Роман Колунов via dev
РАДИОСПЕКТАКЛИ, РАССКАЗЫ, ПОВЕСТИ, ПЬЕСЫ Литературные чтения на Радио России Мы предлагаем Вашему вниманию действительно уникальную коллекцию спектаклей и радиоспектаклей, которая может стать прекрасным подарком. Нам удалось собрать практически всю русскую и зарубежную классику, а также много со

Re: [ovs-dev] [PATCH 1/1] vlog: fix memory leak in vlog_set_log_file() function

2019-10-03 Thread Damijan Skvarc
Perfect! Thanks. On Wed, Oct 2, 2019 at 5:39 PM Ben Pfaff wrote: > On Wed, Oct 02, 2019 at 01:44:49PM +0200, Damijan Skvarc wrote: > > memory leak happens in case previously closed log file was reopened > again, > > for example: > > > > ovs-appctl vlog/close > > ovs-appctl vlog/reopen > > Thank

Re: [ovs-dev] [PATCH v2] ovsdb-client: fix memory leak to prevent valgrind reporting memory leaks while running test suite

2019-10-03 Thread Damijan Skvarc
Super, thanks On Wed, Oct 2, 2019 at 5:31 PM Ben Pfaff wrote: > On Wed, Oct 02, 2019 at 11:37:52AM +0200, Damijan Skvarc wrote: > > memory leaks are reported in several tests and are expressed in a > following way: > > > > ==29840== 208 (48 direct, 160 indirect) bytes in 1 blocks are definitely

[ovs-dev] [PATCH 1/1] ofproto-dpif-xlate: fixed memory leak in is_neighbor_reply_correct() function

2019-10-03 Thread Damijan Skvarc
memory leak happens while calling netdev_get_addr_list() function. This function allocates memory for ip_addr and mask output arguments, but this memory is never freed. Signed-off-by: Damijan Skvarc --- ofproto/ofproto-dpif-xlate.c | 12 1 file changed, 8 insertions(+), 4 deletions(