Re: [ovs-dev] [PATCH] stream-ssl: Fix memory leak on error path.

On Fri, Oct 14, 2016 at 11:12 AM, Ben Pfaff wrote: > The commit that this fixes is from 2009. > > Reported-by: Kai-Wei Fan > Fixes: 9467fe624698 ("Add SSL support to "stream" library and OVSDB.") > Signed-off-by: Ben Pfaff > LGTM Acked-by: Andy

Re: [ovs-dev] [PATCH v3 1/4] ovn: ovn-ctl support for HA ovn DB servers

On Fri, Oct 14, 2016 at 2:52 AM, Babu Shanmugam wrote: > > > On Friday 14 October 2016 04:00 AM, Andy Zhou wrote: > > > > Done. Now it shows the following. > > [root@h2 ovs]# crm configure show >> >> node 1: h1 \ >> >> attributes >> >> node 2: h2 >> >> primitive ClusterIP

Re: [ovs-dev] The Patch netdev-dpdk: add TSO support for vhost-user ports

> > >Hi, Mark > > >OK, once it's ready， pls let me know and I'm glade to help to test it. > > >BTW， you mentioned the gap, is that the TSO and CSUM of tunnel offload? >I'm familiar with it, and I hope I could do something on it if possible. Yes, that's what it was. Commits have recently been

Re: [ovs-dev] [PATCH v3 1/4] ovn: ovn-ctl support for HA ovn DB servers

On Friday 14 October 2016 04:00 AM, Andy Zhou wrote: Done. Now it shows the following. [root@h2 ovs]# crm configure show node1: h1 \ attributes node2: h2 primitiveClusterIP IPaddr2 \ paramsip=10.33.75.200cidr_netmask=32\ opstart interval=0stimeout=20s\

Re: [ovs-dev] The Patch netdev-dpdk: add TSO support for vhost-user ports

> >Hi, Mark > > >I have reviewed your RFC patch, I found one place could have issue, see >below: > > >+if (v > ETHER_MAX_LEN) { >+b->mbuf.data_len = >+(uint16_t)(b->mbuf.buf_len - b->mbuf.data_off); >+} else { >+b->mbuf.data_len = (uint16_t)v; >+} >+ >

Re: [ovs-dev] The Patch netdev-dpdk: add TSO support for vhost-user ports

OK, Thanks for your info. BTW, you mentioned the gap, is it the tunneling tso can't work correctly for multi-segments mbuf, and after some packets, guest is unable to xmit packets(dpdk driver failed to work any more)? Thanks, Michael 2016年10月14日，下午5:15，Kavanagh, Mark B

Re: [ovs-dev] [PATCH v2 01/12] dpcls: Use 32 packet batches for lookups.

Thanks for your comments Daniele, my replies inline. Antonio > -Original Message- > From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Daniele Di > Proietto > Sent: Friday, October 14, 2016 3:08 AM > To: Bodireddy, Bhanuprakash > Cc:

Re: [ovs-dev] [PATCH v3 1/4] ovn: ovn-ctl support for HA ovn DB servers

On Saturday 15 October 2016 02:25 AM, Andy Zhou wrote: On Fri, Oct 14, 2016 at 2:52 AM, Babu Shanmugam > wrote: On Friday 14 October 2016 04:00 AM, Andy Zhou wrote: Done. Now it shows the following. [root@h2 ovs]# crm

Re: [ovs-dev] [PATCH] rhel-systemd: Delay shutting down the services

On Fri, Oct 14, 2016 at 9:36 PM, Russell Bryant wrote: > > On Thu, Oct 13, 2016 at 9:44 AM, Flavio Leitner wrote: > >> On Wed, Oct 12, 2016 at 11:56:51AM -0400, Russell Bryant wrote: >> > On Fri, Oct 7, 2016 at 1:36 PM, Aaron Conole >>

Re: [ovs-dev] [PATCH v2] Introducing support for P4 in OVS

Hi Ben, That's great! Thanks for the new set of changes to the new approach. I have looked at the changes and, I agree, it makes more sense to have template files inline with theOVS source code rather than having them under a separate folder. I like the new approach for P4, much cleaner. I will

Re: [ovs-dev] [Backport Request] For branch-2.5, backport rhel-systemd integration

On Wed, Oct 12, 2016 at 12:03 PM, Russell Bryant wrote: > > > On Fri, Oct 7, 2016 at 1:48 PM, Flavio Leitner wrote: > >> On Thu, Sep 08, 2016 at 07:58:21PM +0100, Markos Chandras wrote: >> > On 09/08/2016 05:50 PM, Aaron Conole wrote: >> > > >> > >> It sounds

Re: [ovs-dev] [PATCH v1] Python-IDL: getattr after mutate fix

On Wed, Oct 12, 2016 at 5:36 PM, Amitabha Biswas wrote: > This commit returns the updated column value when getattr is done > after a mutate operation is performed (but before the commit). > > Signed-off-by: Amitabha Biswas > > Reported-by: Richard Theis

Re: [ovs-dev] [RFC] [PATCH] ovn: Support sample action in logical datapath

Looks like Thunderbird messed up my patch when sending it. Sorry for that - although I hope it shouldn't be a problem for a high-level review. Best, Valentine On 14.10.2016 16:35, Valentine Sinitsyn wrote: Hi all, This is a quick attempt to implement sample action at logical port level.The

[ovs-dev] Bug#840770: openvswitch: FTBFS on almost all big-endian architectures: 954: ofproto-dpif - in place modification (vlan) FAILED (ofproto-dpif.at:7237)

Source: openvswitch Version: 2.5.1~pre+git20160626-2 Severity: serious Hi, openvswitch failed to build on almost all of our big-endian architectures: https://buildd.debian.org/status/package.php?p=openvswitch As you can see it failed on mips, powerpc, s390x, m68k, ppc64 and sparc64, all of

Re: [ovs-dev] [PATCH v3 03/12] flow: Skip invoking expensive count_1bits() with zero input.

> On Oct 14, 2016, at 7:37 AM, Bhanuprakash Bodireddy > wrote: > > This patch checks if trash is non-zero and only then resets the flowmap > bit and increment the pointer by set bits as found in trash. > > Signed-off-by: Bhanuprakash Bodireddy

[ovs-dev] [PATCH RFC v2 0/2] Add DPDK 16.11 & vHost PMD Support

This RFC patchset adds support for DPDK 16.11 (rc1) and the vHost PMD. Thanks to a lack of API changes that affect OVS, the DPDK 16.11 support patch requires just one minimal code change and is mostly documentation updates. The vHost PMD patch removes direct calls to the DPDK vHost User library,

[ovs-dev] [PATCH RFC v2 2/2] netdev-dpdk: Add vHost User PMD

The vHost PMD allows vHost User ports to be controlled by the librte_ether API, like physical 'dpdk' ports and IVSHM 'dpdkr' ports. This commit integrates this PMD into OVS and removes direct calls to the librte_vhost DPDK library. This commit requires DPDK v16.11 functionality that isn't

[ovs-dev] [PATCH RFC v2 1/2] netdev-dpdk: Add support for DPDK 16.11

This commit announces support for DPDK 16.11. Compaitibilty with DPDK v16.07 is not broken yet thanks to only minor code changes being needed for the upgrade. This may change soon however with the addition of subsequent commits that integrate 16.11-only features. Signed-off-by: Ciara Loftus

[ovs-dev] [PATCH v3 10/12] ovsdb: Reorder elements in ovsdb_table_schema structure.

By reordering the elements in ovsdb_table_schema structure, pad bytes can be reduced and also a cache line is saved. Before: structure size:72, holes:2, sum padbytes:10, cachelines:2 After: structure size:64, holes:1, sum padbytes:2, cachelines:1 Signed-off-by: Bhanuprakash Bodireddy

[ovs-dev] [PATCH RFC 0/2] Add DPDK 16.11 & vHost PMD Support

This RFC patchset adds support for DPDK 16.11 (rc1) and the vHost PMD. Thanks to a lack of API changes that affect OVS, the DPDK 16.11 support patch doesn't require any code changes, just documentation updates. The vHost PMD patch removes direct calls to the DPDK vHost User library, and instead

[ovs-dev] [PATCH RFC 1/2] netdev-dpdk: Add support for DPDK 16.11

This commit announces support for DPDK 16.11. No code changes are required, so compaitibilty with DPDK v16.07 is not broken yet. This may change soon however with the addition of subsequent commits that integrate 16.11-only features. Signed-off-by: Ciara Loftus ---

[ovs-dev] [PATCH RFC 2/2] netdev-dpdk: Add vHost User PMD

The vHost PMD allows vHost User ports to be controlled by the librte_ether API, like physical 'dpdk' ports and IVSHM 'dpdkr' ports. This commit integrates this PMD into OVS and removes direct calls to the librte_vhost DPDK library. This commit requires DPDK v16.11 functionality that isn't

[ovs-dev] [PATCH v3 01/12] dpcls: Use 32 packet batches for lookups.

This patch increases the number of packets processed in a batch during a lookup from 16 to 32. Processing batches of 32 packets improves performance and also one of the internal loops can be avoided here. Signed-off-by: Antonio Fischetti Co-authored-by: Bhanuprakash

[ovs-dev] [PATCH v3 06/12] cmap: Remove prefetching in cmap_find_batch().

prefetching the data in to the caches isn't improving the performance in cmap_find_batch(). Moreover its found that there is slight improvement in performance with out prefetching. This patch removes prefetching from cmap_find_batch(). Signed-off-by: Bhanuprakash Bodireddy

[ovs-dev] [PATCH v3 05/12] dpif-netdev: Add comments to dp_netdev_input__().

Add comments in dp_netdev_input__() to explain the reason behind clearing the flow batches before packet_batch_execute(). Signed-off-by: Bhanuprakash Bodireddy Co-authored-by: Antonio Fischetti Signed-off-by: Antonio Fischetti

[ovs-dev] [PATCH v3 03/12] flow: Skip invoking expensive count_1bits() with zero input.

This patch checks if trash is non-zero and only then resets the flowmap bit and increment the pointer by set bits as found in trash. Signed-off-by: Bhanuprakash Bodireddy Co-authored-by: Antonio Fischetti Signed-off-by: Antonio

[ovs-dev] [PATCH v3 04/12] hash: Skip Invoking mhash_add__() with zero input.

mhash_add__() is expensive and should be only called with valid input. zero-valued 'data' will not affect the 'hash' value and expensive hash computation can be skipped when input is zero. This patch will validate the input in mhash_add__ to save some cpu cycles. Signed-off-by: Bhanuprakash

[ovs-dev] [PATCH v3 00/12] Improve performance of OVS-DPDK classifier.

This patch series is aimed at improving the performance of OVS-DPDK dpcls. With few thousand flows installed, the EMC becomes inefficient due to thrashing and the bottleneck moves to the dpcls. In EMC disabled case, through VTune we found that significant performance degradation is due to LLC

[ovs-dev] [PATCH v3 02/12] flow: Add comments to mf_get_next_in_map().

This patch adds comments to mf_get_next_in_map() to make it more comprehensible. Signed-off-by: Jarno Rajahalme Acked-by: Bhanuprakash Bodireddy Acked-by: Antonio Fischetti --- lib/flow.h | 32

[ovs-dev] [PATCH] tnl-ports: fix missing netdev_close

There were leaks on netdev_close references. Whenever an address was removed from an interface, it was possible that a race would make the netlink message be processed and no address be found on the interface, causing insert_ipdev to bail out. When that happened, the netdev would be left opened

[ovs-dev] [PATCH v3 11/12] netlink-socket: Reorder elements in nl_dump structure.

By reordering the elements in nl_dump structure, pad bytes can be reduced there by saving a cache line. Before: structure size:72, holes:1, sum padbytes:4, cachelines:2 After: structure size:64, holes:0, sum padbytes:0, cachelines:1 Signed-off-by: Bhanuprakash Bodireddy

[ovs-dev] [PATCH v3 08/12] dpif-netdev: Reorder elements in dp_netdev_port structure.

By reordering the data elements in dp_netdev_port structure, pad bytes can be reduced and there by saving a cache line. Before: structure size:136, holes:3, sum padbytes:15, cachelines:3 After: structure size:128, holes:2, sum padbytes:7, cachelines:2 Signed-off-by: Bhanuprakash Bodireddy

[ovs-dev] [PATCH v3 09/12] dpif: Reorder elements in dpif_upcall structure.

By reordering the data elements in dpif_upcall structure, pad bytes can be reduced and also a cache line. Also dp_packet should be the first member of the structure because rte_mbuf, the first member of dp_packet should be aligned atleast on a 64-byte boundary. Before: structure size:768,

[ovs-dev] [PATCH v3 07/12] dpif-netdev: Cache align netdev_flow_keys.

Aligning the 'keys' array seems to have positive performance impact. Signed-off-by: Bhanuprakash Bodireddy Co-authored-by: Antonio Fischetti Signed-off-by: Antonio Fischetti Acked-by: Daniele Di

[ovs-dev] [PATCH v3 12/12] timeval: Reorder elements in clock structure.

By reordering the elements in clock structure, pad bytes can be reduced and also a cache line is saved. Before: structure size:136, holes:3, sum padbytes:18, cachelines:3 After: structure size:120, holes:1, sum padbytes:2, cachelines:2 Signed-off-by: Bhanuprakash Bodireddy

Re: [ovs-dev] [PATCH] rhel-systemd: Delay shutting down the services

On Thu, Oct 13, 2016 at 9:44 AM, Flavio Leitner wrote: > On Wed, Oct 12, 2016 at 11:56:51AM -0400, Russell Bryant wrote: > > On Fri, Oct 7, 2016 at 1:36 PM, Aaron Conole wrote: > > > > > During testing it was found that systemd would consider the openvswitch

[ovs-dev] [PATCH] ovsdb: implement read-only remote connection type

This change set adds a new optional access-type specifier to remote connection descriptors for ovsdb-server. Examples: --remote=ptcp:ro:0:192.168.0.10 --remote=punix:ro:asocket.sock --remote=pssl:ro:0:192.168.0.10 --remote=tcp:ro:192.168.0.99: --remote=unix:ro:asocket.sock

[ovs-dev] [RFC] [PATCH] ovn: Support sample action in logical datapath

Hi all, This is a quick attempt to implement sample action at logical port level.The goal is to export IPFIX flows for logical ports, yet it is easy to extend this approach to logical switches as well. Nothing is done to provision OVS instances with required Flow_Sample_Collector_Set and

[ovs-dev] [PATCH v2] ovsdb: implement read-only remote connection type

This change set adds a new optional access-type specifier to remote connection descriptors for ovsdb-server. Examples: --remote=ptcp:ro:0:192.168.0.10 --remote=punix:ro:asocket.sock --remote=pssl:ro:0:192.168.0.10 --remote=tcp:ro:192.168.0.99: --remote=unix:ro:asocket.sock

[ovs-dev] [PATCH] stream-ssl: Fix memory leak on error path.

The commit that this fixes is from 2009. Reported-by: Kai-Wei Fan Fixes: 9467fe624698 ("Add SSL support to "stream" library and OVSDB.") Signed-off-by: Ben Pfaff --- lib/stream-ssl.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git