Re: [ovs-dev] [PATCH v6 0/1] Balance-tcp bond mode optimization

On Tue, Sep 10, 2019 at 11:53 AM Vishal Deep Ajmera wrote: > > v5->v6: > Addressed comments from Ilya Maximets. > https://mail.openvswitch.org/pipermail/ovs-dev/2019-August/362001.html > Rebased to OVS master. > > v4->v5: > Support for stats per hash bucket. > Support for dynamic load

[ovs-dev] [PATCH ovn] Learn the mac binding only if required

From: Numan Siddique OVN has the actions - put_arp and put_nd to learn the mac bindings from the ARP/ND packets. These actions update the Southbound MAC_Binding table. These actions translates to controller actions. Whenever pinctrl thread receives such packets, it wakes up the main

[ovs-dev] Poor performance OVS_DPDK 2.12.90 and 18.11.2

Hello OVS/DPDK team, I am using OVS+DPDK for some of my work and using 2.12.90 and 18.11.2 versions respectively. I ran some performance tests (iperf3 for TCP and UDP) on a c5 4xlarge AWS instance. I am supposed to get 10Gbps, however, for TCP streams I get 2.4Gbps and UDP < 1Gbps. Ethernet

[ovs-dev] [PATCH 10/10] conntrack: Validate accessing of conntrack data in pkt_metadata

Valgrind reported: 1305: ofproto-dpif - conntrack - ipv6 ==26942== Conditional jump or move depends on uninitialised value(s) ==26942==at 0x587C00: check_orig_tuple (conntrack.c:1006) ==26942==by 0x587C00: process_one (conntrack.c:1141) ==26942==by 0x587C00: conntrack_execute

[ovs-dev] [PATCH 07/10] dns-resolve: Free 'struct ub_result' when callback returns error results

Valgrind reported: 1074: ofproto - flush flows, groups, and meters for controller change ==5499== 695 (288 direct, 407 indirect) bytes in 3 blocks are definitely lost in loss record 344 of 355 ==5499==at 0x4C2FB55: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==5499==

[ovs-dev] [PATCH 06/10] ovsdb-client: Free ovsdb_schema

Valgrind reported: 1925: schema conversion online - standalone ==10727== 689 (56 direct, 633 indirect) bytes in 1 blocks are definitely lost in loss record 64 of 66 ==10727==at 0x4C2FB55: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==10727==by 0x449D42: xcalloc

[ovs-dev] [PATCH 08/10] ofproto-dpif: Free leaked 'webster'

Valgrind reported: 1122: ofproto-dpif - select group with explicit dp_hash selection method ==16884== 64 bytes in 1 blocks are definitely lost in loss record 320 of 346 ==16884==at 0x4C2FB55: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==16884==by 0x532512: xcalloc

[ovs-dev] [PATCH 09/10] db-ctl-base: Free leaked ovsdb_datum

Valgrind reported: 2491: database commands -- negative checks ==19245== 36 (32 direct, 4 indirect) bytes in 1 blocks are definitely lost in loss record 36 of 53 ==19245==at 0x4C2FD5F: realloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==19245==by 0x431AB4: xrealloc

[ovs-dev] [RFC ovn 1/3] Add egress QoS mapping for non-tunnel interfaces

Introduce add_egress_interface_mappings routine in order to collect as egress interfaces all ovs bridge interfaces marked with ovn-egress-iface in the external_ids column of ovs interface table. ovn-egress-iface is used to indicate to which localnet ports QoS egress shaping has to be applied.

[ovs-dev] [RFC ovn 2/3] northd: add the possibility to define localnet as qos capable port

Refactor allocate_chassis_queueid and free_chassis_queueid in order to get an unused queue_id even for localnet ports and add the possibility to define localnet as qos capable port Signed-off-by: Lorenzo Bianconi --- northd/ovn-northd.c | 45 ++--- 1 file

[ovs-dev] [RFC ovn 3/3] northd: interoduce logical flow for localnet egress shaping

Add set_queue() action for qos capable localnet port in S_SWITCH_OUT_PORT_SEC_L2 stage of logical switch pipeline Signed-off-by: Lorenzo Bianconi --- northd/ovn-northd.8.xml | 7 ++- northd/ovn-northd.c | 12 +++- 2 files changed, 17 insertions(+), 2 deletions(-) diff --git

[ovs-dev] [PATCH 03/10] dpif-netdev: Handle uninitialized value error for 'match.wc'

Valgrind reported that match.wc was not initialized, as below: 1176: ofproto-dpif - fragment handling - actions ==21214== Conditional jump or move depends on uninitialised value(s) ==21214==at 0x4B77C1: odp_flow_key_from_flow__ (odp-util.c:6143) ==21214==by 0x46DB58: dp_netdev_upcall

[ovs-dev] [PATCH 02/10] ofproto-dpif: Uninitialize 'xlate_cache' to free resources

Valgrind reported: 1210: ofproto-dpif - continuation after clone ==32205== 4,392 (1,440 direct, 2,952 indirect) bytes in 12 blocks are definitely lost in loss record 359 of 362 ==32205==at 0x4C2DB8F: malloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==32205==by 0x532574:

[ovs-dev] [PATCH 01/10] raft: Free leaked json data

Valgrind reported: 1924: compacting online - cluster ==29312== 2,886 (240 direct, 2,646 indirect) bytes in 6 blocks are definitely lost in loss record 406 of 413 ==29312==at 0x4C2DB8F: malloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==29312==by 0x44A5F4: xmalloc

Re: [ovs-dev] [PATCH 10/10] conntrack: Validate accessing of conntrack data in pkt_metadata

Bleep bloop. Greetings Yifeng Sun, 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) #93 FILE: lib/conntrack.c:1142:

[ovs-dev] [RFC ovn 0/3] Introduce localnet egress QoS support

OVN applies logical switch QoS settings to egress interfaces. It currently works by analyzing each br-int interface to see what the remote-ip is on it, and then adding qdiscs to the tunnel-egress-iface associated with this br-int interface. This doesn't work as well when working with VLAN

Re: [ovs-dev] [PATCH ovn v2] OVN: Fix learning of neighbors from ARP/ND packets.

On Thu, Aug 15, 2019 at 12:24 PM Han Zhou wrote: > On Wed, Aug 14, 2019 at 11:11 PM Dumitru Ceara wrote: > > > > On Thu, Aug 8, 2019 at 1:52 AM Han Zhou wrote: > > > > > > > > > > > > On Wed, Aug 7, 2019 at 8:12 AM Dumitru Ceara > wrote: > > >> > > >> On Mon, Aug 5, 2019 at 5:34 PM Han Zhou

[ovs-dev] [PATCHv3] netdev-afxdp: Add need_wakeup supprt.

The patch adds support for using need_wakeup flag in AF_XDP rings. A new option, use_need_wakeup, is added. When this option is used, it means that OVS has to explicitly wake up the kernel RX, using poll() syscall and wake up TX, using sendto() syscall. This feature improves the performance by

Re: [ovs-dev] [PATCH] MAINTAINERS: Update email for Ilya Maximets.

On Wed, Sep 11, 2019 at 06:51:44PM +0300, Ilya Maximets wrote: > CC: Ilya Maximets > Signed-off-by: Ilya Maximets > --- > > As someone could already know, my samsung email will be deactivated > soon. So, the ovn.org address should be used instead. > > BTW, I'll continue working with upstream

[ovs-dev] [PATCH 05/10] trigger: Free leaked ovsdb_schema

Valgrind reported: 1925: schema conversion online - standalone ==10884== 689 (56 direct, 633 indirect) bytes in 1 blocks are definitely lost in loss record 384 of 420 ==10884==at 0x4C2FB55: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==10884==by 0x44A592: xcalloc

[ovs-dev] [PATCH 04/10] ovs-ofctl: Free leaked minimatch

Valgrind reported: 1056: ofproto - bundle with multiple flow mods (OpenFlow 1.4) ==19220== 160 bytes in 2 blocks are definitely lost in loss record 24 of 34 ==19220==at 0x4C2DB8F: malloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==19220==by 0x4979A4: xmalloc (util.c:138)

[ovs-dev] [PATCH 2/2 ovn] Support starting containers with unix socket

From: Aliasgar Ginwala --- Documentation/intro/install/general.rst | 14 ++ utilities/docker/start-ovn | 9 + 2 files changed, 23 insertions(+) diff --git a/Documentation/intro/install/general.rst b/Documentation/intro/install/general.rst index

[ovs-dev] [PATCH 1/2 ovn] OVN: Fix container build.

From: Aliasgar Ginwala Current code changes in ovs/ovn split replaced openvswitch with ovn and caused container patches to break. Also, as per ff4439d, one should build ovn specifying ovs source. This patch fixes the same. --- utilities/docker/create_ovn_dbs.sh | 18 --

Re: [ovs-dev] [PATCH 2/2 ovn] Support starting containers with unix socket

Signed-off-by: Aliasgar Ginwala On Wed, Sep 11, 2019 at 6:04 PM 0-day Robot wrote: > Bleep bloop. Greetings aginwala aginwala, 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. > > >

Re: [ovs-dev] [PATCH 1/1] stream_ssl: fix important memory leak in ssl_connect() function

Kindly reminder to look over this patch. It is quite important since after a while (after certain number of ssl reconnections) the application is crashed by OOM killer. Thanks. On Fri, Jul 26, 2019 at 10:11 AM Damijan Skvarc wrote: > > While checking valgrind reports after running "make

Re: [ovs-dev] [PATCH 2/2 ovn] Support starting containers with unix socket

Bleep bloop. Greetings aginwala aginwala, 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: ERROR: Author Aliasgar Ginwala needs to sign off. Lines checked: 58, Warnings: 0,

Re: [ovs-dev] [PATCH 1/2 ovn] OVN: Fix container build.

Signed-off-by: Aliasgar Ginwala On Wed, Sep 11, 2019 at 6:00 PM 0-day Robot wrote: > Bleep bloop. Greetings aginwala aginwala, 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. > > >

[ovs-dev] [PATCH] netdev-offload-dpdk : add ipv6 rte flow item support

Nowadays some Nics support hw offloading via dpdk rte_flow lib. Many layer2-layer4 fields can be offloaded to nics, including smac/dmac ipv4 sip/dip etc. Also some nics(including intel X710) supports ipv6 header offloading, but when we execute netdev_offload_dpdk_add_flow, there is no IPV6

Re: [ovs-dev] [PATCH 1/2 ovn] OVN: Fix container build.

Bleep bloop. Greetings aginwala aginwala, 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: ERROR: Author Aliasgar Ginwala needs to sign off. Lines checked: 157, Warnings: 0,

[ovs-dev] [PATCH v4 4/7] Performance improvements and native tnl fixes.

From: Tiago Lam Make miniflow_extract() perform by checking if the header fits in the first mbuf only in a single place, since the multiple checks would bring some performance penalties. Furthermore, in some functions, such as dp_packet_set_size(), to distinguish between multi-segment mbufs and

[ovs-dev] [PATCH v4 3/7] netdev-dpdk: Enable TSO when using multi-seg mbufs

From: Tiago Lam TCP Segmentation Offload (TSO) is a feature which enables the TCP/IP network stack to delegate segmentation of a TCP segment to the hardware NIC, thus saving compute resources. This may improve performance significantly for TCP workload in virtualized environments. While a

Re: [ovs-dev] [PATCH ovn v2 1/2] Add ovn-appctl utility

On Tue, Sep 10, 2019 at 11:37 PM Mark Michelson wrote: > For the series, > Acked-by: Mark Michelson > Thanks. I applied this series to master. Numan > > On 9/10/19 3:42 AM, nusid...@redhat.com wrote: > > From: Numan Siddique > > > > Now that OVN has it's own rundir, "ovs-appctl -t >

[ovs-dev] [PATCH v15 00/15] Support multi-segment mbufs

From: Michal Obrembski 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

[ovs-dev] [PATCH v4 2/7] netdev-dpdk: Consider packets marked for TSO.

From: Tiago Lam Previously, TSO was being explicity disabled on vhost interfaces, meaning the guests wouldn't have TSO support negotiated in. With TSO negotiated and enabled, packets are now marked for TSO, through the PKT_TX_TCP_SEG flag. In order to deal with this type of packets, a new

[ovs-dev] [PATCH v4 1/7] netdev-dpdk: Validate packets burst before Tx.

From: Tiago Lam Given that multi-segment mbufs might be sent between interfaces that support different capabilities, and may even support different layouts of mbufs, outgoing packets should be validated before sent on the egress interface. Thus, netdev_dpdk_eth_tx_burst() now calls DPDK's

[ovs-dev] [PATCH v15 14/15] dp-packet: Fix invalid size of ICMPv6 header

From: Artur Twardowski Signed-off-by: Artur Twardowski Signed-off-by: Michal Obrembski --- lib/flow.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/lib/flow.c b/lib/flow.c index 94cfd62..e6019bf 100644 --- a/lib/flow.c +++ b/lib/flow.c @@ -1105,8 +1105,7 @@

[ovs-dev] [PATCH v15 15/15] Fix DPDK MBUF tests compilation on some compilers

From: Michal Obrembski Signed-off-by: Michal Obrembski --- tests/test-dpdk-mbufs.c | 12 ++-- 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/tests/test-dpdk-mbufs.c b/tests/test-dpdk-mbufs.c index 0c152bf..c8dd155 100644 --- a/tests/test-dpdk-mbufs.c +++

[ovs-dev] [PATCH v4 0/7] dpdk: Add support for TSO

From: Michal Obrembski Enabling TSO offload allows a host stack to delegate the segmentation of oversized TCP packets to the underlying physical NIC, if supported. In the case of a VM this means that the segmentation of the packets is not performed by the guest kernel, but by the host NIC

[ovs-dev] [PATCH v15 01/15] netdev-dpdk: Serialise non-pmds mbufs' alloc/free.

From: Tiago Lam A new mutex, 'nonpmd_mp_mutex', has been introduced to serialise allocation and free operations by non-pmd threads on a given mempool. free_dpdk_buf() has been modified to make use of the introduced mutex. Signed-off-by: Tiago Lam Acked-by: Eelco Chaudron Acked-by: Flavio

[ovs-dev] [PATCH v15 02/15] dp-packet: Fix data_len handling multi-seg mbufs.

From: Tiago Lam When a dp_packet is from a DPDK source, and it contains multi-segment mbufs, the data_len is not equal to the packet size, pkt_len. Instead, the data_len of each mbuf in the chain should be considered while distributing the new (provided) size. To account for the above

[ovs-dev] [PATCH v15 03/15] dp-packet: Handle multi-seg mbufs in helper funcs.

From: Tiago Lam Most helper functions in dp-packet assume that the data held by a dp_packet is contiguous, and perform operations such as pointer arithmetic under that assumption. However, with the introduction of multi-segment mbufs, where data is non-contiguous, such assumptions are no longer

[ovs-dev] [PATCH v15 11/15] dpdk-tests: End-to-end tests for multi-seg mbufs.

From: Tiago Lam The following tests are added to the DPDK testsuite to add some coverage for the multi-segment mbufs: - Check that multi-segment mbufs are disabled by default; - Check that providing `other_config:dpdk-multi-seg-mbufs=true` indeed enables mbufs; - Using a DPDK port, send a

[ovs-dev] [PATCH v15 12/15] dpdk-tests: Fix Multi-segment DPDK Unittests

From: Michal Obrembski Signed-off-by: Michal Obrembski --- tests/system-dpdk.at | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/tests/system-dpdk.at b/tests/system-dpdk.at index ca66135..abfe6e6 100644 --- a/tests/system-dpdk.at +++ b/tests/system-dpdk.at @@ -42,6 +42,7

[ovs-dev] [PATCH v15 13/15] Fix build without DPDK

From: Michal Obrembski It looks like in dp_packet_clone_with_headroom() was used a DPDK functions when DPDK was not enabled. Signed-off-by: Michal Obrembski --- lib/dp-packet.c | 19 +++ 1 file changed, 11 insertions(+), 8 deletions(-) diff --git a/lib/dp-packet.c

Re: [ovs-dev] [PATCH ovn] ovn-openstack.rst: Add check for Gateway_Chassis table

On Thu, Sep 5, 2019 at 7:57 PM Flavio Fernandes wrote: > This is related to the section called "Adding a Gateway". > > Added workaround command for a known issue in networking-ovn. > Even after the issue is resolved, it may be useful to have this > in the tutorial, so folks have a feel for how

[ovs-dev] [PATCH v15 06/15] dp-packet: Add support for data "linearization".

From: Tiago Lam Previous commits have added support to the dp_packet API to handle multi-segmented packets, where data is not stored contiguously in memory. However, in some cases, it is inevitable and data must be provided contiguously. Examples of such cases are when performing csums over the

[ovs-dev] [PATCH v15 08/15] netdev-dpdk: support multi-segment jumbo frames.

From: Mark Kavanagh 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

[ovs-dev] [PATCH v15 07/15] netdev-dpdk: copy large packet to multi-seg. mbufs

From: Mark Kavanagh 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 and copies the

[ovs-dev] [PATCH v15 05/15] 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. Also, lots of DPDK mbuf's info is missed during a copy, like packet type, ol_flags, etc. That information is very important for DPDK to do

[ovs-dev] [PATCH v15 04/15] dp-packet: Handle multi-seg mubfs in shift() func.

From: Tiago Lam In its current implementation dp_packet_shift() is also unaware of multi-seg mbufs (that holds data in memory non-contiguously) and assumes that data exists contiguously in memory, memmove'ing data to perform the shift. To add support for multi-seg mbufs a new set of functions

[ovs-dev] [PATCH v4 7/7] Too long line split into two

From: Artur Twardowski Signed-off-by: Artur Twardowski --- lib/netdev-native-tnl.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/lib/netdev-native-tnl.c b/lib/netdev-native-tnl.c index d0ea3d5..f542858 100644 --- a/lib/netdev-native-tnl.c +++ b/lib/netdev-native-tnl.c

[ovs-dev] [PATCH v4 6/7] Give the variables more meaningful names

From: Artur Twardowski Replaced variable names "hlen", "ulen", "gnh" with longer ones to make the code analysis easier. Additionally, UDP header len is no longer accumulated into GENEVE header length, additions are used in the places where the sum of both headers is used. Signed-off-by: Artur

[ovs-dev] [PATCH] vswitchd: update bond-rebalance-interval maxInteger in vswtich.xml

According to the documentation in the vswtich.xml the maximum configurable bond-rebalance-interval is 1ms. However, this is not enforced anywhere in the code and people are using larger values in the field. To avoid confusion this change changes the maximum value the 2147483647 which will fit

Re: [ovs-dev] [PATCH v15 03/15] dp-packet: Handle multi-seg mbufs in helper funcs.

Bleep bloop. Greetings Obrembski, 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: Unexpected sign-offs from developers who are not authors or co-authors or committers:

Re: [ovs-dev] [PATCH v15 09/15] dpdk-tests: Add unit-tests for multi-seg mbufs.

Bleep bloop. Greetings Obrembski, 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 dpdk-tests: Add unit-tests for multi-seg mbufs.

Re: [ovs-dev] [PATCH v4 1/7] netdev-dpdk: Validate packets burst before Tx.

Bleep bloop. Greetings Obrembski, 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: Unexpected sign-offs from developers who are not authors or co-authors or committers:

Re: [ovs-dev] [PATCH v1] netdev-dpdk: Fix flow control configuration.

>-Original Message- >From: Ilya Maximets >Sent: 10 September 2019 14:29 >To: Konieczny, TomaszX ; d...@openvswitch.org >Cc: Stokes, Ian >Subject: Re: [ovs-dev] [PATCH v1] netdev-dpdk: Fix flow control configuration. > > >I looked at the code once more and to the dpdk drivers'

Re: [ovs-dev] [PATCH v1] netdev-dpdk: Fix flow control configuration.

>-Original Message- >From: Ilya Maximets >Sent: 11 September 2019 12:08 >To: Konieczny, TomaszX ; d...@openvswitch.org >Cc: Stokes, Ian >Subject: Re: [ovs-dev] [PATCH v1] netdev-dpdk: Fix flow control configuration. > >On 11.09.2019 12:53, Konieczny, TomaszX wrote: >> >> Hi Ilya, >>

Re: [ovs-dev] [PATCH] ovn: Fix broken dist* targets.

Kind reminder. This fix is important, because the issue breaks all the testsuite targets in TravisCI. It'll be good to make it work. Best regards, Ilya Maximets. On 10.09.2019 11:01, Ilya Maximets wrote: > Redundant line continuation was missed while removing OVN > that broke distdir target and

Re: [ovs-dev] [PATCH v15 02/15] dp-packet: Fix data_len handling multi-seg mbufs.

Bleep bloop. Greetings Obrembski, 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: Unexpected sign-offs from developers who are not authors or co-authors or committers:

Re: [ovs-dev] [PATCH v15 05/15] dp-packet: copy data from multi-seg. DPDK mbuf

Bleep bloop. Greetings Obrembski, 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: Unexpected sign-offs from developers who are not authors or co-authors or committers:

Re: [ovs-dev] [PATCH v4 2/7] netdev-dpdk: Consider packets marked for TSO.

Bleep bloop. Greetings Obrembski, 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: sha1 information is lacking or useless (lib/dp-packet.h). Repository lacks necessary blobs to

Re: [ovs-dev] [PATCH v1] netdev-dpdk: Fix flow control configuration.

On 11.09.2019 12:53, Konieczny, TomaszX wrote: >> -Original Message- >> From: Ilya Maximets >> Sent: 10 September 2019 14:29 >> To: Konieczny, TomaszX ; d...@openvswitch.org >> Cc: Stokes, Ian >> Subject: Re: [ovs-dev] [PATCH v1] netdev-dpdk: Fix flow control >> configuration. >> >> >>

Re: [ovs-dev] [PATCH v5 ovn] Replace chassis mac with router port mac on destination chassis

On Wed, Sep 11, 2019 at 2:53 AM Ankur Sharma wrote: > During E-W routing for vlan backed networks, we replace router port > mac with chassis mac, when packet leaves the source hypervisor. > > As a result, the destination VM (on remote hypervisor) will see > chassis mac as source mac in received

Re: [ovs-dev] [PATCH v1] netdev-dpdk: Fix flow control configuration.

On 11.09.2019 13:50, Konieczny, TomaszX wrote: >> -Original Message- >> From: Ilya Maximets >> Sent: 11 September 2019 12:26 >> To: Konieczny, TomaszX ; d...@openvswitch.org >> Cc: Stokes, Ian >> Subject: Re: [ovs-dev] [PATCH v1] netdev-dpdk: Fix flow control >> configuration. >> >> On

Re: [ovs-dev] [PATCH v15 01/15] netdev-dpdk: Serialise non-pmds mbufs' alloc/free.

Bleep bloop. Greetings Obrembski, 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: Unexpected sign-offs from developers who are not authors or co-authors or committers:

Re: [ovs-dev] [PATCH v15 04/15] dp-packet: Handle multi-seg mubfs in shift() func.

Bleep bloop. Greetings Obrembski, 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: Unexpected sign-offs from developers who are not authors or co-authors or committers:

Re: [ovs-dev] [PATCH v1] netdev-dpdk: Fix flow control configuration.

>-Original Message- >From: Ilya Maximets >Sent: 11 September 2019 12:26 >To: Konieczny, TomaszX ; d...@openvswitch.org >Cc: Stokes, Ian >Subject: Re: [ovs-dev] [PATCH v1] netdev-dpdk: Fix flow control configuration. > >On 11.09.2019 13:16, Konieczny, TomaszX wrote: >> >> >>>

[ovs-dev] [PATCH v16 00/14] 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] [PATCH v16 01/14] netdev-dpdk: Serialise non-pmds mbufs' alloc/free.

From: Tiago Lam A new mutex, 'nonpmd_mp_mutex', has been introduced to serialise allocation and free operations by non-pmd threads on a given mempool. free_dpdk_buf() has been modified to make use of the introduced mutex. Signed-off-by: Tiago Lam Acked-by: Eelco Chaudron Acked-by: Flavio

[ovs-dev] [PATCH v16 07/14] netdev-dpdk: copy large packet to multi-seg. mbufs

From: Mark Kavanagh 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 and copies the

[ovs-dev] [PATCH v16 06/14] dp-packet: Add support for data "linearization".

From: Tiago Lam Previous commits have added support to the dp_packet API to handle multi-segmented packets, where data is not stored contiguously in memory. However, in some cases, it is inevitable and data must be provided contiguously. Examples of such cases are when performing csums over the

[ovs-dev] [PATCH v16 10/14] dpdk-tests: Accept other configs in OVS_DPDK_START

From: Tiago Lam As it stands, OVS_DPDK_START() won't allow other configs to be set before starting the ovs-vswitchd daemon. This is a problem since some configs, such as the "dpdk-multi-seg-mbufs=true" for enabling the multi-segment mbufs, need to be set prior to start OvS. To support other

[ovs-dev] [PATCH v16 08/14] netdev-dpdk: support multi-segment jumbo frames.

From: Mark Kavanagh 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

Re: [ovs-dev] [PATCH] ovn: Fix broken dist* targets.

On 11.09.2019 14:31, Numan Siddique wrote: > > > On Tue, Sep 10, 2019 at 1:31 PM Ilya Maximets > wrote: > > Redundant line continuation was missed while removing OVN > that broke distdir target and subsequently all the dependent > targets, e.g.

[ovs-dev] [PATCH v3] netdev-dpdk: add support for the RTE_ETH_EVENT_INTR_RESET event

Currently, OVS does not register and therefore not handle the interface reset event from the DPDK framework. This would cause a problem in cases where a VF is used as an interface, and its configuration changes. As an example in the following scenario the MAC change is not detected/acted upon

[ovs-dev] [PATCH v16 05/14] 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. Also, lots of DPDK mbuf's info is missed during a copy, like packet type, ol_flags, etc. That information is very important for DPDK to do

[ovs-dev] [PATCH v16 13/14] dp-packet: Fix invalid size of ICMPv6 header

From: Artur Twardowski Signed-off-by: Artur Twardowski Signed-off-by: Michal Obrembski --- lib/flow.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/lib/flow.c b/lib/flow.c index 94cfd62..e6019bf 100644 --- a/lib/flow.c +++ b/lib/flow.c @@ -1105,8 +1105,7 @@

[ovs-dev] [PATCH v16 12/14] dpdk-tests: Fix Multi-segment DPDK Unittests

Signed-off-by: Michal Obrembski --- tests/system-dpdk.at | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/tests/system-dpdk.at b/tests/system-dpdk.at index ca66135..abfe6e6 100644 --- a/tests/system-dpdk.at +++ b/tests/system-dpdk.at @@ -42,6 +42,7 @@

[ovs-dev] [PATCH v16 14/14] Fix DPDK MBUF tests compilation on some compilers

Signed-off-by: Michal Obrembski --- tests/test-dpdk-mbufs.c | 12 ++-- 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/tests/test-dpdk-mbufs.c b/tests/test-dpdk-mbufs.c index 0c152bf..c8dd155 100644 --- a/tests/test-dpdk-mbufs.c +++ b/tests/test-dpdk-mbufs.c @@ -601,9

Re: [ovs-dev] [PATCH] ovn: Fix broken dist* targets.

On Tue, Sep 10, 2019 at 1:31 PM Ilya Maximets wrote: > Redundant line continuation was missed while removing OVN > that broke distdir target and subsequently all the dependent > targets, e.g. distcheck: > > make[1]: > *** No rule to make target 'nodist_ovn_lib_libovn_la_SOURCES', >

Re: [ovs-dev] [PATCH v2] netdev-dpdk: add support for the RTE_ETH_EVENT_INTR_RESET event

On 10 Sep 2019, at 16:02, Ilya Maximets wrote: On 10.09.2019 15:12, Eelco Chaudron wrote: Currently, OVS does not register and therefore not handle the interface reset event from the DPDK framework. This would cause a problem in cases where a VF is used as an interface, and its

Re: [ovs-dev] [PATCH v1] netdev-dpdk: Fix flow control configuration.

>-Original Message- >From: Ilya Maximets >Sent: 11 September 2019 13:04 >To: Konieczny, TomaszX ; d...@openvswitch.org >Cc: Stokes, Ian >Subject: Re: [ovs-dev] [PATCH v1] netdev-dpdk: Fix flow control configuration. > >On 11.09.2019 13:50, Konieczny, TomaszX wrote: >>> -Original

Re: [ovs-dev] [PATCH v1] netdev-dpdk: Fix flow control configuration.

On 11.09.2019 15:21, Konieczny, TomaszX wrote: >> -Original Message- >> From: Ilya Maximets >> Sent: 11 September 2019 13:04 >> To: Konieczny, TomaszX ; d...@openvswitch.org >> Cc: Stokes, Ian >> Subject: Re: [ovs-dev] [PATCH v1] netdev-dpdk: Fix flow control >> configuration. >> >> On

[ovs-dev] [PATCH v16 03/14] dp-packet: Handle multi-seg mbufs in helper funcs.

From: Tiago Lam Most helper functions in dp-packet assume that the data held by a dp_packet is contiguous, and perform operations such as pointer arithmetic under that assumption. However, with the introduction of multi-segment mbufs, where data is non-contiguous, such assumptions are no longer

[ovs-dev] [PATCH v16 02/14] dp-packet: Fix data_len handling multi-seg mbufs.

From: Tiago Lam When a dp_packet is from a DPDK source, and it contains multi-segment mbufs, the data_len is not equal to the packet size, pkt_len. Instead, the data_len of each mbuf in the chain should be considered while distributing the new (provided) size. To account for the above

[ovs-dev] [PATCH v16 09/14] dpdk-tests: Add unit-tests for multi-seg mbufs.

From: Tiago Lam In order to create a minimal environment that allows the tests to get mbufs from an existing mempool, the following approach is taken: - EAL is initialised (by using the main dpdk_init()) and a (very) small mempool is instantiated (mimicking the logic in dpdk_mp_create()).

[ovs-dev] [PATCH v16 11/14] dpdk-tests: End-to-end tests for multi-seg mbufs.

From: Tiago Lam The following tests are added to the DPDK testsuite to add some coverage for the multi-segment mbufs: - Check that multi-segment mbufs are disabled by default; - Check that providing `other_config:dpdk-multi-seg-mbufs=true` indeed enables mbufs; - Using a DPDK port, send a

[ovs-dev] [PATCH v4 1/7] netdev-dpdk: Validate packets burst before Tx.

From: Tiago Lam Given that multi-segment mbufs might be sent between interfaces that support different capabilities, and may even support different layouts of mbufs, outgoing packets should be validated before sent on the egress interface. Thus, netdev_dpdk_eth_tx_burst() now calls DPDK's

[ovs-dev] [PATCH v5 5/7] Fix OVN failing tests.

An unnecessary offset has been added while introducing dp_packet_may_pull in netdev-native-tnl.c Signed-off-by: Michal Obrembski --- lib/netdev-native-tnl.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/lib/netdev-native-tnl.c b/lib/netdev-native-tnl.c index

[ovs-dev] [PATCH v5 6/7] Give the variables more meaningful names

From: Artur Twardowski Replaced variable names "hlen", "ulen", "gnh" with longer ones to make the code analysis easier. Additionally, UDP header len is no longer accumulated into GENEVE header length, additions are used in the places where the sum of both headers is used. Signed-off-by: Artur

[ovs-dev] [PATCH v5 0/7] dpdk: Add support for TSO

Enabling TSO offload allows a host stack to delegate the segmentation of oversized TCP packets to the underlying physical NIC, if supported. In the case of a VM this means that the segmentation of the packets is not performed by the guest kernel, but by the host NIC itself. In turn, since the

[ovs-dev] [PATCH v5 7/7] Too long line split into two

From: Artur Twardowski Signed-off-by: Artur Twardowski --- lib/netdev-native-tnl.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/lib/netdev-native-tnl.c b/lib/netdev-native-tnl.c index d0ea3d5..f542858 100644 --- a/lib/netdev-native-tnl.c +++ b/lib/netdev-native-tnl.c

Re: [ovs-dev] [PATCH v16 06/14] dp-packet: Add support for data "linearization".

Bleep bloop. Greetings Michal Obrembski, 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: Unexpected sign-offs from developers who are not authors or co-authors or

Re: [ovs-dev] [PATCH v16 07/14] netdev-dpdk: copy large packet to multi-seg. mbufs

Bleep bloop. Greetings Michal Obrembski, 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: Unexpected sign-offs from developers who are not authors or co-authors or

Re: [ovs-dev] [PATCH v16 08/14] netdev-dpdk: support multi-segment jumbo frames.

Bleep bloop. Greetings Michal Obrembski, 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: Unexpected sign-offs from developers who are not authors or co-authors or

[ovs-dev] [PATCH v5 3/7] netdev-dpdk: Enable TSO when using multi-seg mbufs

From: Tiago Lam TCP Segmentation Offload (TSO) is a feature which enables the TCP/IP network stack to delegate segmentation of a TCP segment to the hardware NIC, thus saving compute resources. This may improve performance significantly for TCP workload in virtualized environments. While a

[ovs-dev] [PATCH v5 4/7] Performance improvements and native tnl fixes.

From: Tiago Lam Make miniflow_extract() perform by checking if the header fits in the first mbuf only in a single place, since the multiple checks would bring some performance penalties. Furthermore, in some functions, such as dp_packet_set_size(), to distinguish between multi-segment mbufs and

Re: [ovs-dev] [PATCH v16 09/14] dpdk-tests: Add unit-tests for multi-seg mbufs.

Bleep bloop. Greetings Michal Obrembski, 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 dpdk-tests: Add unit-tests for multi-seg

Re: [ovs-dev] [PATCH v5 2/7] netdev-dpdk: Consider packets marked for TSO.

Bleep bloop. Greetings Michal Obrembski, 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: sha1 information is lacking or useless (lib/netdev-dpdk.c). Repository lacks necessary

  1   2   >