Re: [ovs-dev] datapath/linux/compat question

On Wed, Apr 2, 2014 at 9:56 AM, Zoltan Kiss wrote: > On 01/04/14 20:26, Jesse Gross wrote: >> >> On Tue, Apr 1, 2014 at 11:41 AM, Kyle Mestery >> wrote: >>> >>> On Tue, Apr 1, 2014 at 1:27 PM, Zoltan Kiss >>> wrote: Hi, I have a recent patch on net-next which affects OVS as w

[ovs-dev] [PATCH] ofproto: per-table statistics

Add per-table counters. This resolves some short-comings in the data provided in a table stats reply message. * Lookups and matches are calculated based on table accesses rather than datapath flow hits and misses. * Lookups and matches are credited to the table where they occurred rather than

Re: [ovs-dev] Subject: [PATCH] Add support for 802.1ad (qinq)

Andy, I will send out a v2 patch for your review soon. Also, thanks for reviewing the patch in detail. My comments are inserted below. On 4/2/2014 3:54 PM, Andy Zhou wrote: Thomas, thanks a lot for working on this. The patch seems to be incomplete. Is this intentional? If yes, then we may w

Re: [ovs-dev] [megaflow bond rebase] ofproto/bond: Implement bond megaflow using recirculation

On Tue, Apr 01, 2014 at 03:22:54PM -0700, Andy Zhou wrote: > Infrastructure to enable megaflow support for bond ports using > recirculation. This patch adds the following features: > * Generate RECIRC action when bond can benefit from recirculation. > * Populate post recirculation rules in a hidden

Re: [ovs-dev] [PATCH] ovsdb-client: Support all vlog options.

Thanks, I'll push this in a minute. On Wed, Apr 02, 2014 at 11:02:06AM -0700, Andy Zhou wrote: > LGTM > > Acked-by: Andy Zhou > > On Wed, Apr 2, 2014 at 10:20 AM, Ben Pfaff wrote: > > Signed-off-by: Ben Pfaff > > --- > > ovsdb/ovsdb-client.c | 12 > > 1 file changed, 4 insertion

Re: [ovs-dev] [PATCH] debian: Allow kmod or module-init-tools for backward compatibility.

Thanks! On Wed, Apr 02, 2014 at 02:53:40PM -0700, Justin Pettit wrote: > Acked-by: Justin Pettit > > --Justin > > > On April 2, 2014 at 2:47:34 PM, Ben Pfaff (b...@nicira.com) wrote: > > Commit d473844693 (debian: Depend on 'kmod' instead of module-init-tools.) > > switched from depending on m

Re: [ovs-dev] [PATCH] debian: Allow kmod or module-init-tools for backward compatibility.

On Wed, Apr 02, 2014 at 02:52:09PM -0700, Gurucharan Shetty wrote: > On Wed, Apr 2, 2014 at 2:46 PM, Ben Pfaff wrote: > > Commit d473844693 (debian: Depend on 'kmod' instead of module-init-tools.) > > switched from depending on module-init-tools to depending on kmod, which > > is the new name of t

Re: [ovs-dev] [PATCH] debian: Allow kmod or module-init-tools for backward compatibility.

Acked-by: Justin Pettit --Justin On April 2, 2014 at 2:47:34 PM, Ben Pfaff (b...@nicira.com) wrote: > Commit d473844693 (debian: Depend on 'kmod' instead of module-init-tools.) > switched from depending on module-init-tools to depending on kmod, which > is the new name of the appropriate packag

Re: [ovs-dev] [PATCH] ofproto-dpif-upcall: Expire netflow flow when revalidate_ukey failed

On 3 April 2014 10:16, Joe Stringer wrote: > I had noticed that this logic was in the codepath for when we delete a > flow from the datapath, but not in this codepath. Thanks for picking this > up! > On second thought, I suppose that this means that we will expire the netflow immediately when we

Re: [ovs-dev] [PATCH] debian: Allow kmod or module-init-tools for backward compatibility.

On Wed, Apr 2, 2014 at 2:46 PM, Ben Pfaff wrote: > Commit d473844693 (debian: Depend on 'kmod' instead of module-init-tools.) > switched from depending on module-init-tools to depending on kmod, which > is the new name of the appropriate package in Debian. Unfortunately, > while kmod is the right

[ovs-dev] [PATCH] debian: Allow kmod or module-init-tools for backward compatibility.

Commit d473844693 (debian: Depend on 'kmod' instead of module-init-tools.) switched from depending on module-init-tools to depending on kmod, which is the new name of the appropriate package in Debian. Unfortunately, while kmod is the right name for the latest Debian distribution, it doesn't have

Re: [ovs-dev] [PATCH] ofpbuf: Abstract 'l2' pointer and document usage conventions.

I’ll also change all the “getter” functions to be spelled without the “get”, as done by Pravin for his patch earlier today on ofpbufs. Unless there are other comments, I’ll not send a new patch for review, as this is not affecting functionality in any way. Jarno On Apr 2, 2014, at 11:33 AM,

Re: [ovs-dev] [PATCH 5/5] netdev-dpdk: Remove alloc from packet recv.

On Wed, Apr 2, 2014 at 1:00 PM, Jarno Rajahalme wrote: > > On Mar 31, 2014, at 9:43 PM, Pravin wrote: > >> From: Pravin Shelar >> >> On DPDK packet recv, ovs is given pointer to mbuf which has >> information about a packet, for example pointer to data and size. >> By moving mbuf to ofpbuf we can

Re: [ovs-dev] [PATCH 2/5] ofpbuf: Introduce access api for base, data and size.

On Wed, Apr 2, 2014 at 12:43 PM, Jarno Rajahalme wrote: > LGTM, > > Some lines are too long now. > > One more thought: Maybe the code would look cleaner, if "getters" were > spelled without the "get", like "ofpbuf_size()", "ofpbuf_l2()". I know it > would not rhyme with the "setters", but as we do

[ovs-dev] [ovsdb 2/7] Revert "jsonrpc-server: Add test for disconnecting connections with too long queues."

This reverts commit 631583739f9aec55d4cbe25fb856143ccde48ab6. Connections that queue up too much data, because they are monitoring a table that is changing quickly and failing to keep up with the updates, cause problems with buffer management. Since commit 60533a405b2e (jsonrpc-server: Disconnect

[ovs-dev] [ovsdb 6/7] jsonrpc-server: Track monitor updates separately from sending them.

This will make combining monitor updates easier in the next commit. Signed-off-by: Ben Pfaff --- ovsdb/jsonrpc-server.c | 339 +--- 1 file changed, 264 insertions(+), 75 deletions(-) diff --git a/ovsdb/jsonrpc-server.c b/ovsdb/jsonrpc-server.c index

[ovs-dev] [ovsdb 4/7] Revert "ovsdb-data: New functions for predicting serialized length of data."

This reverts commit 0ea7bec76d804a2c4efccd3dbdaa3e30cf536a5c. Connections that queue up too much data, because they are monitoring a table that is changing quickly and failing to keep up with the updates, cause problems with buffer management. Since commit 60533a405b2e (jsonrpc-server: Disconnect

[ovs-dev] [ovsdb 5/7] Revert "json: New function json_serialized_length()."

This reverts commit 1600fa6853872e16130366351a2c14f6fa8b547c. Connections that queue up too much data, because they are monitoring a table that is changing quickly and failing to keep up with the updates, cause problems with buffer management. Since commit 60533a405b2e (jsonrpc-server: Disconnect

[ovs-dev] [ovsdb 1/7] ovsdb-server: Send update for _version for changes due to weak ref removal.

When a row was deleted, and that caused the transaction manager to remove a remaining weak reference to the row, and no other change had been made to the row in that transaction, and a client was monitoring modifications of the _version column in the row, then the monitor update for the column did

[ovs-dev] [ovsdb 3/7] Revert "jsonrpc-server: Disconnect connections that queue too much data."

This reverts commit 60533a405b2eb77214b767767fe143c8645f82d5. Connections that queue up too much data, because they are monitoring a table that is changing quickly and failing to keep up with the updates, cause problems with buffer management. Since commit 60533a405b2e (jsonrpc-server: Disconnect

[ovs-dev] [ovsdb 0/7] combine ovsdb updates when connections become backed up

Connections that queue up too much data, because they are monitoring a table that is changing quickly and failing to keep up with the updates, cause problems with buffer management. Since commit 60533a405b2e (jsonrpc-server: Disconnect connections that queue too much data.), ovsdb-server has dealt

[ovs-dev] [ovsdb 7/7] jsonrpc-server: Combine notifications when connection becomes backlogged.

Connections that queue up too much data, because they are monitoring a table that is changing quickly and failing to keep up with the updates, cause problems with buffer management. Since commit 60533a405b2e (jsonrpc-server: Disconnect connections that queue too much data.), ovsdb-server has dealt

Re: [ovs-dev] [PATCH] ofproto-dpif-upcall: Expire netflow flow when revalidate_ukey failed

I had noticed that this logic was in the codepath for when we delete a flow from the datapath, but not in this codepath. Thanks for picking this up! Acked-by: Joe Stringer On 2 April 2014 17:18, YAMAMOTO Takashi wrote: > This fixes missing netflow flows in > "ofproto-dpif - NetFlow flow expir

[ovs-dev] [PATCH] datapath: Add support for kernels 3.13

From: Pravin Shelar Add support for building the in-tree kernel datapath for Linux kernels up to 3.13. There were some changes in the netlink area which required adding new compatibility code for this layer. Also, some new per-cpu stats initialization code was added. Based on patch from Kyle Mes

Re: [ovs-dev] Subject: [PATCH] Add support for 802.1ad (qinq)

Thomas, thanks a lot for working on this. The patch seems to be incomplete. Is this intentional? If yes, then we may want delay mentioning it in the news. 1) User space open flow PUSH_VLAN currently blocking 802.1ad header. That needs to be fixed for open flow versions that supports it. 2) Us

Re: [ovs-dev] [PATCH 5/5] netdev-dpdk: Remove alloc from packet recv.

On Mar 31, 2014, at 9:43 PM, Pravin wrote: > From: Pravin Shelar > > On DPDK packet recv, ovs is given pointer to mbuf which has > information about a packet, for example pointer to data and size. > By moving mbuf to ofpbuf we can let dpdk allocate ofpbuf and > pass that to ovs for processing

Re: [ovs-dev] [PATCH 4/5] ofpbuf: Add DPDK mbuf to ofpbuf.

LGTM, Jarno Acked-by: Jarno Rajahalme On Mar 31, 2014, at 9:43 PM, Pravin wrote: > From: Pravin Shelar > > Define data, base and size access APIs for DPDK. > > Signed-off-by: Pravin B Shelar > --- > lib/ofpbuf.h | 47 ++- > 1 file changed, 42

Re: [ovs-dev] [PATCH 3/5] ofpbuf: Add ofpbuf_init_dpdk()

On Mar 31, 2014, at 9:43 PM, Pravin wrote: > From: Pravin Shelar > > Signed-off-by: Pravin B Shelar > --- > lib/ofpbuf.c | 27 ++- > lib/ofpbuf.h |2 ++ > 2 files changed, 24 insertions(+), 5 deletions(-) > > diff --git a/lib/ofpbuf.c b/lib/ofpbuf.c > index 8763e4

Re: [ovs-dev] [PATCH 2/5] ofpbuf: Introduce access api for base, data and size.

LGTM, Some lines are too long now. One more thought: Maybe the code would look cleaner, if “getters” were spelled without the “get”, like “ofpbuf_size()”, “ofpbuf_l2()”. I know it would not rhyme with the “setters”, but as we do much less setting than getting it would still be cleaner? Thought

Re: [ovs-dev] [megaflow bond rebase] ofproto/bond: Implement bond megaflow using recirculation

On Wed, Apr 2, 2014 at 12:44 AM, YAMAMOTO Takashi wrote: >> @@ -2021,6 +2025,17 @@ dp_netdev_port_input(struct dp_netdev *dp, struct >> ofpbuf *packet, >> } >> } >> >> +static void >> +dp_netdev_port_input(struct dp_netdev *dp, struct ofpbuf *packet, >> + struct pkt_meta

Re: [ovs-dev] [PATCH] datapath: Add support for kernels 3.13 and 3.14

Hi Kyle, I think we can better use new genl api in datapath. I am incorporating some changes I suggested before. I will send out patch soon. Thanks, Pravin. On Tue, Apr 1, 2014 at 6:32 AM, Kyle Mestery wrote: > Add support for building the in-tree kernel datapath for Linux kernels up to > 3.14.

Re: [ovs-dev] [PATCH 1/5] ofpbuf: Add private pointer for dpdk

LGTM, Acked-by: Jarno Rajahalme On Mar 31, 2014, at 9:43 PM, Pravin wrote: > From: Pravin Shelar > > netdev-dpdk uses this pointer to store dpdk mbuf. This patch fixes > compilation error in dpdk. > > Signed-off-by: Pravin B Shelar > --- > lib/ofpbuf.h |4 > 1 file changed, 4 inser

Re: [ovs-dev] [PATCH] ofpbuf: Abstract 'l2' pointer and document usage conventions.

On Apr 2, 2014, at 11:23 AM, Lori Jakab wrote: > On 4/2/14, 8:59 PM, Jarno Rajahalme wrote: > > [...] > >> diff --git a/lib/ofpbuf.h b/lib/ofpbuf.h >> index 8d1cb11..fa8a3f0 100644 >> --- a/lib/ofpbuf.h >> +++ b/lib/ofpbuf.h >> @@ -36,26 +36,46 @@ enum OVS_PACKED_ENUM ofpbuf_source { >> }; >>

Re: [ovs-dev] [PATCH] ofpbuf: Abstract 'l2' pointer and document usage conventions.

On 4/2/14, 8:59 PM, Jarno Rajahalme wrote: [...] diff --git a/lib/ofpbuf.h b/lib/ofpbuf.h index 8d1cb11..fa8a3f0 100644 --- a/lib/ofpbuf.h +++ b/lib/ofpbuf.h @@ -36,26 +36,46 @@ enum OVS_PACKED_ENUM ofpbuf_source { }; /* Buffer for holding arbitrary data. An ofpbuf is automatically real

Re: [ovs-dev] [PATCH v6 6/6] datapath: Minimize ovs_flow_cmd_new|set critical sections.

Thanks for the reviews! All six pushed to master, Jarno On Apr 1, 2014, at 3:20 PM, Pravin Shelar wrote: > On Mon, Mar 31, 2014 at 11:18 AM, Jarno Rajahalme > wrote: >> Signed-off-by: Jarno Rajahalme ... >> @@ -871,40 +885,45 @@ static int ovs_flow_cmd_new(struct sk_buff *skb, >> struct

Re: [ovs-dev] [RFC PATCH 4/4] lib/ofpbuf: Compact

On 4/2/14, 9:09 PM, Jarno Rajahalme wrote: On Apr 1, 2014, at 1:28 PM, Lori Jakab > wrote: On 4/1/14, 5:43 PM, Jarno Rajahalme wrote: Sent from my iPhone On Apr 1, 2014, at 7:27 AM, Ben Pfaff > wrote: On Tue, Apr 01, 2014 at 03:14:16PM +

Re: [ovs-dev] [RFC PATCH 4/4] lib/ofpbuf: Compact

On Apr 1, 2014, at 1:28 PM, Lori Jakab wrote: > On 4/1/14, 5:43 PM, Jarno Rajahalme wrote: >> >> Sent from my iPhone >> >>> On Apr 1, 2014, at 7:27 AM, Ben Pfaff wrote: >>> On Tue, Apr 01, 2014 at 03:14:16PM +0300, Lori Jakab wrote: >>> Can we do the following? >>>3) Replace 'l2' by

Re: [ovs-dev] [PATCH] ovsdb-client: Support all vlog options.

LGTM Acked-by: Andy Zhou On Wed, Apr 2, 2014 at 10:20 AM, Ben Pfaff wrote: > Signed-off-by: Ben Pfaff > --- > ovsdb/ovsdb-client.c | 12 > 1 file changed, 4 insertions(+), 8 deletions(-) > > diff --git a/ovsdb/ovsdb-client.c b/ovsdb/ovsdb-client.c > index d60d7ca..2942953 100644

[ovs-dev] [PATCH] ofpbuf: Abstract 'l2' pointer and document usage conventions.

Rename 'l2' to 'frame' and add new ofpbuf_set_frame() and ofpbuf_get_l2(). ofpbuf_set_frame() alse resets all the layer offsets. ofpbuf_get_l2() returns NULL if the packet has no Ethernet header, as indicated either by unset l3 offset or NULL frame pointer. Caller's of ofpbuf_get_l2() are suppose

[ovs-dev] [PATCH] ovsdb-client: Support all vlog options.

Signed-off-by: Ben Pfaff --- ovsdb/ovsdb-client.c | 12 1 file changed, 4 insertions(+), 8 deletions(-) diff --git a/ovsdb/ovsdb-client.c b/ovsdb/ovsdb-client.c index d60d7ca..2942953 100644 --- a/ovsdb/ovsdb-client.c +++ b/ovsdb/ovsdb-client.c @@ -1,5 +1,5 @@ /* - * Copyright (c)

Re: [ovs-dev] datapath/linux/compat question

On 01/04/14 20:26, Jesse Gross wrote: On Tue, Apr 1, 2014 at 11:41 AM, Kyle Mestery wrote: On Tue, Apr 1, 2014 at 1:27 PM, Zoltan Kiss wrote: Hi, I have a recent patch on net-next which affects OVS as well: http://git.kernel.org/cgit/linux/kernel/git/davem/net-next.git/commit/?id=36d5fe6a00

Re: [ovs-dev] [PATCH] datapath: Add support for kernels 3.13 and 3.14

Hi, It seems some parts of the changes from my original patch are missing. On 01/04/14 14:32, Kyle Mestery wrote: diff --git a/datapath/linux/compat/skbuff-openvswitch.c b/datapath/linux/compat/skbuff-openvswitch.c index ddd7bc8..bcba930 100644 --- a/datapath/linux/compat/skbuff-openvswitch.c

Re: [ovs-dev] [PATCH] vtep: add IP address configuration for bfd

Applied, thx, On Wed, Apr 2, 2014 at 9:07 AM, Alex Wang wrote: > Acked-by: Alex Wang > > > > On Tue, Apr 1, 2014 at 3:01 PM, Bruce Davie wrote: > >> The OVS implementation of BFD allows configuration of the source and >> destination IP addresses of BFD packets. This patch adds the same >> con

[ovs-dev] Subject: [PATCH] Add support for 802.1ad (qinq)

From c7521a74f0e8f2673d4f542b885f6c82a0302f45 Mon Sep 17 00:00:00 2001 From: Tom Herbert Date: Wed, 2 Apr 2014 06:03:23 + Add support for 802.1ad (qinq) Allows 0x88a8 for TPID is S-VLAN (outer) tag. Signed-off-by: Tom Herbert --- NEWS|1 + datapath/actions.c

Re: [ovs-dev] [PATCH] vtep: add IP address configuration for bfd

Acked-by: Alex Wang On Tue, Apr 1, 2014 at 3:01 PM, Bruce Davie wrote: > The OVS implementation of BFD allows configuration of the source and > destination IP addresses of BFD packets. This patch adds the same > configuration option to the VTEP schema. > > Signed-off-by: Bruce Davie > --- >

Benachrichtigung von Sparkasse‏

This is HTML source of message you composed. Do not modify here. To modify this message press HTML Messages Editor button. Sehr geehrter Kunde,   wir möchten Sie darauf hinweisen, dass der Zugang zu Ihrem Online-Konto in Kurze abläuft. Um dieses weiterhin nützen zu können, bit

[ovs-dev] Bug#733696: marked as done (please do not depend on module-init-tools)

Your message dated Wed, 02 Apr 2014 11:00:15 + with message-id and subject line Bug#733696: fixed in openvswitch 2.1.0+git20140325-1 has caused the Debian Bug report #733696, regarding please do not depend on module-init-tools to be marked as done. This means that you claim that the problem h

[ovs-dev] Bug#701760: marked as done (openvswitch-switch: problem interacting with libvirt)

Your message dated Wed, 02 Apr 2014 11:00:15 + with message-id and subject line Bug#701760: fixed in openvswitch 2.1.0+git20140325-1 has caused the Debian Bug report #701760, regarding openvswitch-switch: problem interacting with libvirt to be marked as done. This means that you claim that th

[ovs-dev] Bug#729304: marked as done (openvswitch-switch: logrotate error if directory "/var/run/openvswitch" does not exist)

Your message dated Wed, 02 Apr 2014 11:00:15 + with message-id and subject line Bug#729304: fixed in openvswitch 2.1.0+git20140325-1 has caused the Debian Bug report #729304, regarding openvswitch-switch: logrotate error if directory "/var/run/openvswitch" does not exist to be marked as done.

[ovs-dev] Bug#732260: marked as done (openvswitch: datapath kernel module fails to compile with linux kernel version 3.11)

Your message dated Wed, 02 Apr 2014 11:00:15 + with message-id and subject line Bug#732260: fixed in openvswitch 2.1.0+git20140325-1 has caused the Debian Bug report #732260, regarding openvswitch: datapath kernel module fails to compile with linux kernel version 3.11 to be marked as done. T

[ovs-dev] Bug#731133: marked as done (openvswitch-controller: logging is disabled by default however it conflicts with logrotate configuration)

Your message dated Wed, 02 Apr 2014 11:00:15 + with message-id and subject line Bug#731133: fixed in openvswitch 2.1.0+git20140325-1 has caused the Debian Bug report #731133, regarding openvswitch-controller: logging is disabled by default however it conflicts with logrotate configuration to

[ovs-dev] Bug#680538: marked as done (openvswitch - brcompat unable to overtake bridge configured by ifupdown)

Your message dated Wed, 02 Apr 2014 11:00:15 + with message-id and subject line Bug#680538: fixed in openvswitch 2.1.0+git20140325-1 has caused the Debian Bug report #680538, regarding openvswitch - brcompat unable to overtake bridge configured by ifupdown to be marked as done. This means th

[ovs-dev] Bug#681954: marked as done (openvswitch-brcompat - Init script switch and different package)

Your message dated Wed, 02 Apr 2014 11:00:15 + with message-id and subject line Bug#681954: fixed in openvswitch 2.1.0+git20140325-1 has caused the Debian Bug report #681954, regarding openvswitch-brcompat - Init script switch and different package to be marked as done. This means that you cl

[ovs-dev] openvswitch_2.1.0+git20140325-1_i386.changes ACCEPTED into unstable, unstable

Accepted: -BEGIN PGP SIGNED MESSAGE- Hash: SHA256 Format: 1.8 Date: Mon, 31 Mar 2014 09:31:10 -0700 Source: openvswitch Binary: openvswitch-datapath-source openvswitch-datapath-dkms openvswitch-common openvswitch-switch openvswitch-ipsec openvswitch-pki openvswitch-dbg python-openvswi

[ovs-dev] [PATCH v3 repost] ofproto: Support OF version-specific table-miss behaviours

OpenFlow 1.1 and 1.2 specify that if a table-miss occurs then the default behaviour is to forward the packet the controller using a packet-in message. And until this patch this is the default behaviour that Open vSwitch uses for all OpenFlow versions. OpenFlow1.3+ specifies that if a table-miss oc

Re: [ovs-dev] [megaflow bond rebase] ofproto/bond: Implement bond megaflow using recirculation

> @@ -2021,6 +2025,17 @@ dp_netdev_port_input(struct dp_netdev *dp, struct > ofpbuf *packet, > } > } > > +static void > +dp_netdev_port_input(struct dp_netdev *dp, struct ofpbuf *packet, > + struct pkt_metadata *md) > +OVS_REQ_RDLOCK(dp->port_rwlock) > +{ > +uin