Re: [ovs-dev] [code] OFTest for OVS

On Thu, Oct 25, 2012 at 05:12:21PM -0700, Ben Pfaff wrote: > It passes a few tests; I've only tried a few. > > Now that I look at it, there are some bugs here that will prevent > traffic from actually passing through, but they are not fundamental to > the approach. I'm mostly passing this along i

Re: [ovs-dev] [openflow-dev] [code] OFTest for OVS

Like I said, it's buggy. I'm doing some after-hours hacking on it tonight. Perhaps I'll get it working. I got my copy of oftest from that repo. I'm basing on commit b371430baebd8c3b130e418e0c6f5a5a2144ece2 "Merge branch 'kenc'". On Thu, Oct 25, 2012 at 10:25:30PM -0700, Rob Sherwood wrote: > H

Re: [ovs-dev] [openflow-dev] [code] OFTest for OVS

Hi Ben, Thanks for the work! I haven't worked on OFTest in a bit, but I'll forward this on to people who are actively working on it and see if we can merge your code or at least give comments, etc. For my sanity, where did you get oftest? The most current version lives at g...@github.com:floodl

[ovs-dev] [PATCH v2] OF11: push_vlan support

This implementes push_vlan with 802.1Q. NOTE: 802.1AD (QinQ) is not supported. It requires another effort. Signed-off-by: Isaku Yamahata --- Changes v1 -> v2: - don't carrying around ethertype - move ethertype check from ofp_check__() to ofpact_from_openflow11() - drop non-standard vlan ethertype

[ovs-dev] [PATCH 17/18] test: add ofproto OpenFlow1.2 tests

These were useful in isolating a number of bugs. Signed-off-by: Simon Horman --- The "ofproto - flow table configuration (OpenFlow 1.2)]" test will need to be updated for the MPLS changes that are outstanding as they change the following: wild=0xf -> wild=0x1f match=0xf

[ovs-dev] [PATCH 18/18] ofp-util: Optimize ofputil_version_bitmap_scanr

Make use of __builtin_clz if available which should optimize ofputil_version_bitmap_scanr() by replacing a loop with a single CLZ instruction when available. I'm unsure if this approach is worth it or not. But a similar approach could be taken to use ffs() in bitmap_scan(). Signed-off-by: Simon H

[ovs-dev] [PATCH 15/18] ofp-util: Open Flow 1.1 and 1.2 flow format capabilities

Open Flow and 1.1 and 1.2 support some of the same flow format capabilities as NXM. Signed-off-by: Simon Horman --- lib/ofp-util.c | 30 +++--- lib/ofp-util.h |3 +++ tests/ovs-ofctl.at |2 +- 3 files changed, 19 insertions(+), 16 deletions(-) diff --gi

[ovs-dev] [PATCH 16/18] ofp-util: Flow Dump Protocol for OpenFlow 12

Allow only OpenFlow 12 as a flow dump protocol for OpenFlow12. Allow OpenFlow10 and NXM as flow dump protocols for other OpenFlow versions. This only changes the behaviour prior to this patch in the case of OpenFlow12 which seemed broken as an attempt would be made to set NXM as the flow dump forma

[ovs-dev] [PATCH 13/18] ovs-controller: Allow setting of allowed OpenFlow versions

--allowed-ofp-versions allows configuration of the versions that may be used when establishing an OpenFlow connection. The default is 'OpenFlow10' which is consistent with the behaviour prior to this patch. The useful values at this time are: 'OpenFlow10', 'OpenFlow12' and 'OpenFlow10,OpenFlow12'

[ovs-dev] [PATCH 12/18] ovs-ofctl: Add option to set allowed OpenFlow versions

--allowed-ofp-versions allows configuration of the versions that may be used when establishing an OpenFlow connection. The default is 'OpenFlow10' which is consistent with the behaviour prior to this patch. The useful values at this time are: 'OpenFlow10', 'OpenFlow12' and 'OpenFlow10,OpenFlow12'

[ovs-dev] [PATCH 11/18] lib: Add helpers for OpenFlow version command line options

Signed-off-by: Simon Horman --- lib/automake.mk |3 +++ lib/ofp-version-opt.c | 42 ++ lib/ofp-version-opt.h | 28 lib/ofp-version.man | 26 ++ 4 files changed, 99 insertions(+) create

[ovs-dev] [PATCH 09/18] connmgr: Use version of underlying rconn

Signed-off-by: Simon Horman --- lib/ofp-util.c| 14 -- lib/ofp-util.h|2 ++ ofproto/connmgr.c |9 - 3 files changed, 22 insertions(+), 3 deletions(-) diff --git a/lib/ofp-util.c b/lib/ofp-util.c index bf453e2..3c85185 100644 --- a/lib/ofp-util.c +++ b/lib/of

[ovs-dev] [PATCH 07/18] ofproto, connmgr: Parameterise OpenFlow versions for adding controller

Allow allowed Open Flow versions to be passed rather than relying on hard-coded defaults. This is in preparation for allowing configuration of the allowed OpenFlow versions. Signed-off-by: Simon Horman --- ofproto/connmgr.c | 23 ++- ofproto/connmgr.h |3 ++- ofproto/o

[ovs-dev] [PATCH 06/18] rconn: Add allowed OpenFlow versions

Add allowed OpenFlow versions to struct rconn to allow reconnect to use these parameters rather than hard-coded defaults. This is in preparation for allowing configuration of the allowed OpenFlow versions. Signed-off-by: Simon Horman --- lib/rconn.c|8 ++-- lib/rconn.h

[ovs-dev] [PATCH 05/18] ofproto: Make set_pvconns() aware of OpenFlow versions

Paramatise the allowed OpenFlow verion for snoops. This is in preparation for making the allow OpenFlow verions to be configured. Signed-off-by: Simon Horman --- ofproto/connmgr.c | 19 ++- ofproto/connmgr.h |2 +- ofproto/ofproto.c |2 +- ofproto/ofproto.h |7 +

[ovs-dev] [PATCH 14/18] ovs-ofctl: By default, do not set flow format for OpenFlow 1.1+

Only set the default format for ovs-ofctl monitor if OpenFlow 1.0 is the prevailing version. IMHO that is the only case where it makes sense. Signed-off-by: Simon Horman --- utilities/ovs-ofctl.c | 36 +++- 1 file changed, 23 insertions(+), 13 deletions(-) diff

[ovs-dev] [PATCH 10/18] vswitchd: Configuration of allowed OpenFlow versions

Versions may be configured using a comma delimited list as the value for 'openflow-versions' the 'other-config' column of the Bridge table. Currently list elements other than 'OpenFlow10' and 'OpenFlow12' will be ignored. If the list is empty, then OpenFlow10 is used. This default is consistent w

[ovs-dev] [PATCH 08/18] ofp-util: Allow use of OpenFlow 12 flow format

This enables the use of the OpenFlow 12 flow format. Signed-off-by: Simon Horman --- lib/ofp-util.c |2 +- lib/ofp-util.h |4 ++-- tests/learn.at |2 +- 3 files changed, 4 insertions(+), 4 deletions(-) diff --git a/lib/ofp-util.c b/lib/ofp-util.c index b415128..bf453e2 100644 --- a/

[ovs-dev] [PATCH 02/18] vconn: Allowed OpenFlow versions

Replace minimum version with bitmap of allowed versions. This is in preparation for allowing the range of allowed OpenFlow versions to be configured. As part of this change pvconn_open() is now paramatised over the allowed versions. this is to avoid avoids needing to provide version information

[ovs-dev] [PATCH 03/18] ofp-util: Add version bitmap support to hello messages

Allow encoding and decoding of version bitmap in hello messages as specified in Open Flow 1.3.1. Signed-off-by: Simon Horman --- v3 * As suggested by Ben Pfaff - Skip unknown hello elements - Use bitmap directly in the form of a uint32_t instead of creating a struct ofp_hello v2 * Init

[ovs-dev] [PATCH 01/18] vconn: Add ofputil_version_bitmap

Functions to manipulate a bitmap of allowed versions. The bitmap is simple a uint32_t, allowing for versions up until wire-protocol number 31. Bit offsets correspond to ofp_version numbers which in turn correspond to wire-protocol numbers for Open Flow versions.. E.g. (1 << OFP10_VERSION) is the

[ovs-dev] [PATCH v3 00/18] Run-Time Open Flow Version Configuration

Hi, This series adds run-time configuration of allowed OpenFlow versions to ovs-vswtichd, ovs-ofctl and ovs-controller; and adds tests for Open Flow 1.2 messages that utilise the run-time configuration of allowed OpenFlow versions. I believe that this addresses the issues raised by Ben Pfaff in h

[ovs-dev] [PATCH 04/18] connmgr: Paramatise ofservice_create() over allowed OpenFlow versions

This is in preparation for allowing configuration of the allowed OpenFlow versions. Signed-off-by: Simon Horman --- ofproto/connmgr.c | 13 - 1 file changed, 8 insertions(+), 5 deletions(-) diff --git a/ofproto/connmgr.c b/ofproto/connmgr.c index f1e6487..b3c94db 100644 --- a/ofpr

Re: [ovs-dev] [PATCHv2] netdev-vport: Disable Path MTU Discovery by default

On Thu, Oct 25, 2012 at 3:46 PM, Ansis Atteka wrote: > On Thu, Oct 25, 2012 at 9:24 PM, Jesse Gross wrote: >> On Thu, Oct 25, 2012 at 4:50 AM, Ansis Atteka wrote: >>> In near future we will switch to the flow-based tunneling and drop path >>> MTU discovery support. This patch prepares for that b

[ovs-dev] [code] OFTest for OVS

It's somewhat difficult to use OFTest, in my opinion, because you need to be root, work with OS network devices, and so on. Today I played around with some code to avoid those two needs. First, I have a patch series starting here: http://openvswitch.org/pipermail/dev/2012-October/022300.h

[ovs-dev] [tests 22/22] bond: Add test for active-backup bond.

Signed-off-by: Ben Pfaff --- tests/automake.mk |1 + tests/bond.at | 349 +++ tests/lacp.at |4 +- tests/ofproto-macros.at |2 +- tests/testsuite.at |1 + 5 files changed, 353 insertions(+), 4 deletions(-)

[ovs-dev] [tests 19/22] ofproto-dpif: Report port number consistently in "fdb/show" output.

Previously, the port number was nondeterministic for bonds (it could be any one of the port's interfaces). Signed-off-by: Ben Pfaff --- ofproto/ofproto-dpif.c | 11 +-- 1 files changed, 9 insertions(+), 2 deletions(-) diff --git a/ofproto/ofproto-dpif.c b/ofproto/ofproto-dpif.c index

[ovs-dev] [tests 21/22] netdev-linux: Improve log message.

Signed-off-by: Ben Pfaff --- lib/netdev-linux.c |4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) diff --git a/lib/netdev-linux.c b/lib/netdev-linux.c index 412a92d..6e2570f 100644 --- a/lib/netdev-linux.c +++ b/lib/netdev-linux.c @@ -1365,8 +1365,8 @@ netdev_linux_sys_get_stats(cons

[ovs-dev] [tests 20/22] ofproto-dpif: Fix "ofproto/trace" of flow for local port.

The 'in_port' variable is an OpenFlow port, not an ODP port, so we should not translate it to ODP. Signed-off-by: Ben Pfaff --- ofproto/ofproto-dpif.c |2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/ofproto/ofproto-dpif.c b/ofproto/ofproto-dpif.c index 3e1b7b3..7d914c8

[ovs-dev] [tests 18/22] ofproto-dpif: Report more detailed reason for flooding a packet.

Signed-off-by: Ben Pfaff --- ofproto/ofproto-dpif.c |4 +++- 1 files changed, 3 insertions(+), 1 deletions(-) diff --git a/ofproto/ofproto-dpif.c b/ofproto/ofproto-dpif.c index 63ce1c2..86a5361 100644 --- a/ofproto/ofproto-dpif.c +++ b/ofproto/ofproto-dpif.c @@ -6333,7 +6333,9 @@ xlate_norma

[ovs-dev] [tests 17/22] ofproto-dpif: Factor duplicated code into new function.

Signed-off-by: Ben Pfaff --- ofproto/ofproto-dpif.c | 36 1 files changed, 24 insertions(+), 12 deletions(-) diff --git a/ofproto/ofproto-dpif.c b/ofproto/ofproto-dpif.c index c800386..63ce1c2 100644 --- a/ofproto/ofproto-dpif.c +++ b/ofproto/ofproto-dpif.c

[ovs-dev] [tests 16/22] netdev-dummy: Add "pstream" option for connecting a dummy to a process.

Signed-off-by: Ben Pfaff --- lib/netdev-dummy.c | 171 +++- 1 files changed, 170 insertions(+), 1 deletions(-) diff --git a/lib/netdev-dummy.c b/lib/netdev-dummy.c index 4790492..e9d9a9d 100644 --- a/lib/netdev-dummy.c +++ b/lib/netdev-dummy.c @@

[ovs-dev] [tests 15/22] netdev-dummy: Add pcap feature.

Signed-off-by: Ben Pfaff --- lib/netdev-dummy.c | 46 +- lib/pcap.c | 31 +-- 2 files changed, 70 insertions(+), 7 deletions(-) diff --git a/lib/netdev-dummy.c b/lib/netdev-dummy.c index a466358..4790492 100644 -

[ovs-dev] [tests 14/22] netdev-dummy: Remove unneeded forward declaration.

Signed-off-by: Ben Pfaff --- lib/netdev-dummy.c |1 - 1 files changed, 0 insertions(+), 1 deletions(-) diff --git a/lib/netdev-dummy.c b/lib/netdev-dummy.c index 6722dcb..a466358 100644 --- a/lib/netdev-dummy.c +++ b/lib/netdev-dummy.c @@ -77,7 +77,6 @@ struct netdev_dummy { static struct

[ovs-dev] [tests 13/22] netdev-dummy: Allow "netdev-dummy/set-admin-state" accept multiple interfaces.

Signed-off-by: Ben Pfaff --- lib/netdev-dummy.c | 20 1 files changed, 12 insertions(+), 8 deletions(-) diff --git a/lib/netdev-dummy.c b/lib/netdev-dummy.c index cd19789..6722dcb 100644 --- a/lib/netdev-dummy.c +++ b/lib/netdev-dummy.c @@ -730,14 +730,18 @@ netdev_dummy_s

[ovs-dev] [tests 12/22] netdev-dummy: Correctly maintain port statistics.

Signed-off-by: Ben Pfaff --- lib/netdev-dummy.c |6 ++ 1 files changed, 6 insertions(+), 0 deletions(-) diff --git a/lib/netdev-dummy.c b/lib/netdev-dummy.c index 9776edc..cd19789 100644 --- a/lib/netdev-dummy.c +++ b/lib/netdev-dummy.c @@ -327,6 +327,9 @@ netdev_dummy_send(struct netdev

[ovs-dev] [tests 11/22] netdev-dummy: Add a packet generator.

This is useful for unit tests. Signed-off-by: Ben Pfaff --- lib/netdev-dummy.c | 328 +--- lib/packets.h |3 + 2 files changed, 317 insertions(+), 14 deletions(-) diff --git a/lib/netdev-dummy.c b/lib/netdev-dummy.c index 5636854..9776ed

[ovs-dev] [tests 10/22] netdev-dummy: Add "patch port"-like feature.

This allows testing functionality that requires connectivity between multiple bridges, such as bonding, LACP, and CFM. This commit adds an example use to test basic LACP negotiation. Signed-off-by: Ben Pfaff --- lib/netdev-dummy.c | 81 ++- tests/lacp.at | 428 ++

[ovs-dev] [tests 09/22] netdev-dummy: Drop "nobody listened" reply from netdev-dummy/receive.

Ethan pointed out that this wasn't very useful. Signed-off-by: Ben Pfaff --- lib/netdev-dummy.c| 16 +++- tests/ofproto-dpif.at |6 ++ 2 files changed, 5 insertions(+), 17 deletions(-) diff --git a/lib/netdev-dummy.c b/lib/netdev-dummy.c index 3aaace3..9b9e251 100644 -

[ovs-dev] [tests 08/22] netdev-dummy: Factor some netdev_dummy_receive() code out into new function.

An upcoming patch will add another user. Signed-off-by: Ben Pfaff --- lib/netdev-dummy.c | 33 ++--- 1 files changed, 22 insertions(+), 11 deletions(-) diff --git a/lib/netdev-dummy.c b/lib/netdev-dummy.c index bfc2997..3aaace3 100644 --- a/lib/netdev-dummy.c +++ b

[ovs-dev] [tests 07/22] netdev-dummy: Limit receive queue length to 100 packets.

It doesn't seem like a good idea to allow the queue length to grow without bound, even for a test-only device. Signed-off-by: Ben Pfaff --- lib/netdev-dummy.c | 11 ++- 1 files changed, 10 insertions(+), 1 deletions(-) diff --git a/lib/netdev-dummy.c b/lib/netdev-dummy.c index 6aa4084

[ovs-dev] [tests 06/22] ofproto-macros: Fail a test that logs a WARN or higher level message.

Signed-off-by: Ben Pfaff --- tests/ofproto-macros.at |7 ++- 1 files changed, 6 insertions(+), 1 deletions(-) diff --git a/tests/ofproto-macros.at b/tests/ofproto-macros.at index 52f19fc..585bcae 100644 --- a/tests/ofproto-macros.at +++ b/tests/ofproto-macros.at @@ -64,5 +64,10 @@ m4_def

[ovs-dev] [tests 05/22] bridge: Drop warning about the default bridge Ethernet address.

This information is also available via ovs-ofctl and through other means, and it's not really anything we need to warn about anyhow. Signed-off-by: Ben Pfaff --- vswitchd/bridge.c | 10 -- 1 files changed, 0 insertions(+), 10 deletions(-) diff --git a/vswitchd/bridge.c b/vswitchd/brid

[ovs-dev] [tests 04/22] tests: Set explicit bond mode in LACP test.

This avoids a log warning: bridge|WARN|port bond: Using the default bond_mode active-backup. Note that in previous versions, the default bond_mode was balance-slb This warning is harmless, but I'm trying to add checks for "warn" and higher severity log messages to the tests, so it makes s

[ovs-dev] [tests 03/22] bond: Reduce log level from WARN to INFO for interface status updates.

An interface coming up or going down isn't a big deal. Signed-off-by: Ben Pfaff --- lib/bond.c |6 +++--- 1 files changed, 3 insertions(+), 3 deletions(-) diff --git a/lib/bond.c b/lib/bond.c index 25a0fa1..0f91413 100644 --- a/lib/bond.c +++ b/lib/bond.c @@ -1296,12 +1296,12 @@ bond_enable

[ovs-dev] [tests 02/22] poll-loop: Reduce high-CPU log messages from WARN to INFO.

These can happen occasionally in normal circumstances. Signed-off-by: Ben Pfaff --- lib/poll-loop.c |4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) diff --git a/lib/poll-loop.c b/lib/poll-loop.c index 516cf13..c7c9952 100644 --- a/lib/poll-loop.c +++ b/lib/poll-loop.c @@ -165,8 +1

[ovs-dev] [tests 01/22] netlink-socket: Don't bother logging SO_RCVBUFFORCE failure as non-root.

Some Open vSwitch utilities can do useful work when they are not run as root. Without this commit, these utilities will log a warning on failure to use the SO_RCVBUFFORCE socket option if they open any Netlink sockets. This will always happen, it does not report anything unexpected or fixable as n

[ovs-dev] [tests 00/22] bond tests, oftest support

This series reflects two different development chains that both involve improving the netdev-dummy implementation. First, months ago I spent a week or so writing unit tests for the bonding code. The tests themselves, which are in the final patch of this series, are half-baked. But the patches on

Re: [ovs-dev] [PATCHv2] netdev-vport: Disable Path MTU Discovery by default

On Thu, Oct 25, 2012 at 9:24 PM, Jesse Gross wrote: > On Thu, Oct 25, 2012 at 4:50 AM, Ansis Atteka wrote: >> In near future we will switch to the flow-based tunneling and drop path >> MTU discovery support. This patch prepares for that by changing the default >> PMTUD value from 'enabled' to 'di

Re: [ovs-dev] [PATCH] datapath: Deprecated CAPWAP support.

On Thu, Oct 25, 2012 at 1:28 PM, Kyle Mestery (kmestery) wrote: > On Oct 27, 2012, at 3:07 PM, Pravin B Shelar wrote: >> Signed-off-by: Pravin B Shelar > > > This looks ok. Is this error going to be around for a release, at which point > CAPWAP support will then be removed from the code? Just cu

Re: [ovs-dev] [PATCH] datapath: Add support for 3.6 kernel.

On Thu, Oct 25, 2012 at 01:33:32PM -0700, Pravin Shelar wrote: > On Thu, Oct 25, 2012 at 1:22 PM, Ben Pfaff wrote: > > On Sat, Oct 27, 2012 at 01:07:35PM -0700, Pravin B Shelar wrote: > >> Signed-off-by: Pravin B Shelar > > > > ... > > > >> +#if LINUX_VERSION_CODE >= KERNEL_VERSION(3,6,0) > >> +#

Re: [ovs-dev] [PATCH] datapath: Add support for 3.6 kernel.

On Thu, Oct 25, 2012 at 1:22 PM, Ben Pfaff wrote: > On Sat, Oct 27, 2012 at 01:07:35PM -0700, Pravin B Shelar wrote: >> Signed-off-by: Pravin B Shelar > > ... > >> +#if LINUX_VERSION_CODE >= KERNEL_VERSION(3,6,0) >> +#define NETLINK_CB_PORT_ID(skb) (NETLINK_CB((skb)).portid) >> +#else >> +#define

Re: [ovs-dev] [PATCH] datapath: enable encap for capwap.

On Oct 27, 2012, at 3:07 PM, Pravin B Shelar wrote: > kernel 3.5 added a switch to turn on UDP encap, capwap needs > to enable it. > > Signed-off-by: Pravin B Shelar Acked-by: Kyle Mestery ___ dev mailing list dev@openvswitch.org http://openvswitch.

Re: [ovs-dev] [PATCH] datapath: Add Upstream id for GRE type.

On Oct 27, 2012, at 3:06 PM, Pravin B Shelar wrote: > Once GRE is upstream it will have new type to have continuous sequence > of ids for vport type. Following patch adds this ID to have > compatibility with it. > > Signed-off-by: Pravin B Shelar Acked-by: Kyle Mestery ___

Re: [ovs-dev] [PATCH] datapath: Fix sparse warning for symbol 'BUILD_BUG_ON_NOT_POWER_OF_2'

On Thu, Oct 25, 2012 at 01:28:02PM -0700, Pravin Shelar wrote: > On Thu, Oct 25, 2012 at 1:19 PM, Ben Pfaff wrote: > > On Sat, Oct 27, 2012 at 01:07:24PM -0700, Pravin B Shelar wrote: > >> Signed-off-by: Pravin B Shelar > > > > I'm curious, what warning does sparse give? (You could add this to >

Re: [ovs-dev] [PATCH] datapath: Deprecated CAPWAP support.

On Oct 27, 2012, at 3:07 PM, Pravin B Shelar wrote: > Signed-off-by: Pravin B Shelar This looks ok. Is this error going to be around for a release, at which point CAPWAP support will then be removed from the code? Just curious about the plan there. Acked-by: Kyle Mestery _

Re: [ovs-dev] [PATCH] datapath: Fix sparse warning for symbol 'BUILD_BUG_ON_NOT_POWER_OF_2'

On Thu, Oct 25, 2012 at 1:19 PM, Ben Pfaff wrote: > On Sat, Oct 27, 2012 at 01:07:24PM -0700, Pravin B Shelar wrote: >> Signed-off-by: Pravin B Shelar > > I'm curious, what warning does sparse give? (You could add this to > the commit message.) > This symbol was moved from kernel.h to bug.h in 3

Re: [ovs-dev] [PATCH] datapath: Add support for 3.6 kernel.

On Sat, Oct 27, 2012 at 01:07:35PM -0700, Pravin B Shelar wrote: > Signed-off-by: Pravin B Shelar ... > +#if LINUX_VERSION_CODE >= KERNEL_VERSION(3,6,0) > +#define NETLINK_CB_PORT_ID(skb) (NETLINK_CB((skb)).portid) > +#else > +#define NETLINK_CB_PORT_ID(skb) (NETLINK_CB((skb)).pid) > +#endif I'

Re: [ovs-dev] [PATCH] datapath: Fix sparse warning for symbol 'BUILD_BUG_ON_NOT_POWER_OF_2'

On Sat, Oct 27, 2012 at 01:07:24PM -0700, Pravin B Shelar wrote: > Signed-off-by: Pravin B Shelar I'm curious, what warning does sparse give? (You could add this to the commit message.) Thanks, Ben. ___ dev mailing list dev@openvswitch.org http://ope

Re: [ovs-dev] [1.4 backports 0/6] backports of datapaths fixes to 1.4

On Wed, Oct 24, 2012 at 06:16:55PM -0700, Jesse Gross wrote: > On Wed, Oct 24, 2012 at 12:47 PM, Ben Pfaff wrote: > > Yesterday, while going through datapath changes on master, I > > noticed that some bugfixes seem to be relevant and missing on > > master. Here is a series of backports for review

Re: [ovs-dev] [1.4 backports 2/6] datapath: Move CSUM_MANGLED_0 definition to net checksum header.

On Wed, Oct 24, 2012 at 06:13:57PM -0700, Jesse Gross wrote: > On Wed, Oct 24, 2012 at 12:47 PM, Ben Pfaff wrote: > > From: Pravin B Shelar > > > > Following patch fixes compilation error on older kernel. > > > > This is a crossport of commit 08d19ca9fef29b23826f1fb52e2368a9077783ca > > from mast

[ovs-dev] [PATCH] datapath: Add support for 3.6 kernel.

Signed-off-by: Pravin B Shelar --- datapath/datapath.c | 50 - datapath/linux/compat/include/linux/netlink.h |6 +++ datapath/linux/compat/include/net/genetlink.h |8 datapath/tunnel.c | 35 +---

[ovs-dev] [PATCH] datapath: Remove tunnel header caching.

Tunnel caching was added to reduce CPU utilization on TX path by caching packet header, So performance gain is directly proportional to number of skbs transferred. But with help of offloads skb are getting larger. So there are less number of skbs. Therefore header caching does not shows similar gai

[ovs-dev] [PATCH] datapath: Fix sparse warning for symbol 'BUILD_BUG_ON_NOT_POWER_OF_2'

Signed-off-by: Pravin B Shelar --- datapath/linux/compat/include/linux/kernel.h |2 ++ 1 file changed, 2 insertions(+) diff --git a/datapath/linux/compat/include/linux/kernel.h b/datapath/linux/compat/include/linux/kernel.h index 812f213..ff354ad 100644 --- a/datapath/linux/compat/include/l

[ovs-dev] [PATCH] datapath: enable encap for capwap.

kernel 3.5 added a switch to turn on UDP encap, capwap needs to enable it. Signed-off-by: Pravin B Shelar --- datapath/linux/compat/include/linux/udp.h |5 + datapath/vport-capwap.c |2 +- 2 files changed, 6 insertions(+), 1 deletion(-) diff --git a/datapath/linux/

[ovs-dev] [PATCH] datapath: Deprecated CAPWAP support.

Signed-off-by: Pravin B Shelar --- NEWS|1 + datapath/vport-capwap.c |1 + 2 files changed, 2 insertions(+) diff --git a/NEWS b/NEWS index 3cb1bd3..3f99054 100644 --- a/NEWS +++ b/NEWS @@ -42,6 +42,7 @@ v1.9.0 - xx xxx - The autopath action. - I

[ovs-dev] [PATCH] datapath: Add Upstream id for GRE type.

Once GRE is upstream it will have new type to have continuous sequence of ids for vport type. Following patch adds this ID to have compatibility with it. Signed-off-by: Pravin B Shelar --- datapath/vport-gre.c| 23 +++ datapath/vport.c|1 + datapath/

[ovs-dev] Bug#691438: openvswitch packages do not work well together with upstream's kernel module

On Thu, Oct 25, 2012 at 07:20:48PM +0200, Arno Töll wrote: > Package: openvswitch > Severity: important > > As you are certainly aware, Open vSwitch is included upstream since > kernel 3.3. Post-Wheezy kernels also enable CONFIG_OPENVSWITCH in > pristine Debian kernel builds (see #675010). Yes.

Re: [ovs-dev] [PATCH] OF11: push_vlan support

On Thu, Oct 25, 2012 at 02:42:23PM +0900, Isaku Yamahata wrote: > This implementes push_vlan with 802.1Q. > NOTE: 802.1AD (QinQ) is not supported. It requires another effort. > > Signed-off-by: Isaku Yamahata The check that this patch does in ofpact_check__() should be moved into ofpact_from_ope

Re: [ovs-dev] [PATCHv2] netdev-vport: Disable Path MTU Discovery by default

On Thu, Oct 25, 2012 at 4:50 AM, Ansis Atteka wrote: > In near future we will switch to the flow-based tunneling and drop path > MTU discovery support. This patch prepares for that by changing the default > PMTUD value from 'enabled' to 'disabled'. > > Signed-off-by: Ansis Atteka Since the goal

Re: [ovs-dev] [PATCH 2/3] timeval: Coalesce backtraces with counts.

Thanks for the reviews, I'll merge this series shortly. Ethan On Thu, Oct 25, 2012 at 1:13 PM, Ben Pfaff wrote: > On Thu, Oct 25, 2012 at 10:06:26AM -0700, Ethan Jackson wrote: >> Making stubs turned out to be cleanest. Here's an incremental. > > Looks good to me, thanks. __

Re: [ovs-dev] [PATCH 2/3] timeval: Coalesce backtraces with counts.

On Thu, Oct 25, 2012 at 10:06:26AM -0700, Ethan Jackson wrote: > Making stubs turned out to be cleanest. Here's an incremental. Looks good to me, thanks. ___ dev mailing list dev@openvswitch.org http://openvswitch.org/mailman/listinfo/dev

[ovs-dev] Bug#691438: openvswitch packages do not work well together with upstream's kernel module

Package: openvswitch Severity: important As you are certainly aware, Open vSwitch is included upstream since kernel 3.3. Post-Wheezy kernels also enable CONFIG_OPENVSWITCH in pristine Debian kernel builds (see #675010). However, the Debian packages do not work well together with this upstream mo

Re: [ovs-dev] [PATCH 3/3] poll-loop: Log backtraces when CPU usage is high.

Good idea, I've folded that in. Ethan On Wed, Oct 24, 2012 at 6:58 PM, Ben Pfaff wrote: > On Wed, Oct 24, 2012 at 01:49:09PM -0700, Ethan Jackson wrote: >> Often when debugging Open vSwitch, one will see in the logs that >> CPU usage has been high for some period of time, but it's totally >> unc

Re: [ovs-dev] [PATCH 2/3] timeval: Coalesce backtraces with counts.

Making stubs turned out to be cleanest. Here's an incremental. --- lib/timeval.c | 28 ++-- 1 file changed, 18 insertions(+), 10 deletions(-) diff --git a/lib/timeval.c b/lib/timeval.c index bd2a84d..b36203a 100644 --- a/lib/timeval.c +++ b/lib/timeval.c @@ -39,10 +39,

Re: [ovs-dev] [PATCHv2] netdev-vport: Disable Path MTU Discovery by default

On Thu, Oct 25, 2012 at 6:22 PM, Ben Pfaff wrote: > On Thu, Oct 25, 2012 at 02:50:10PM +0300, Ansis Atteka wrote: >> In near future we will switch to the flow-based tunneling and drop path >> MTU discovery support. This patch prepares for that by changing the default >> PMTUD value from 'enabled'

Re: [ovs-dev] [PATCHv2] netdev-vport: Disable Path MTU Discovery by default

On Thu, Oct 25, 2012 at 02:50:10PM +0300, Ansis Atteka wrote: > In near future we will switch to the flow-based tunneling and drop path > MTU discovery support. This patch prepares for that by changing the default > PMTUD value from 'enabled' to 'disabled'. > > Signed-off-by: Ansis Atteka Thanks

Re: [ovs-dev] [PATCH] ovs-vsctl: check if the device name is valid

On Thu, Oct 25, 2012 at 04:24:37PM +0800, Cong Wang wrote: > On Wed, Oct 24, 2012 at 11:54 PM, Ben Pfaff wrote: > > On Wed, Oct 24, 2012 at 05:45:36PM +0800, Cong Wang wrote: > >> Before waiting for the kernel to reject an invalid name, we > >> can actually check it before going into the kernel. T

[ovs-dev] [PATCHv2] netdev-vport: Disable Path MTU Discovery by default

In near future we will switch to the flow-based tunneling and drop path MTU discovery support. This patch prepares for that by changing the default PMTUD value from 'enabled' to 'disabled'. Signed-off-by: Ansis Atteka --- NEWS |1 + lib/netdev-vport.c | 10 +- vsw

Re: [ovs-dev] [PATCH] ovs-vsctl: check if the device name is valid

On Wed, Oct 24, 2012 at 11:54 PM, Ben Pfaff wrote: > On Wed, Oct 24, 2012 at 05:45:36PM +0800, Cong Wang wrote: >> Before waiting for the kernel to reject an invalid name, we >> can actually check it before going into the kernel. The code >> is stolen from linux kernel function dev_valid_name(), >