Re: [ovs-dev] [PATCH] ovs-router: Define stub for ovs_router_unixctl_register()

Acked-by: Eitan Eliahu Thanks! Eitan -Original Message- From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Pravin B Shelar Sent: Wednesday, November 05, 2014 3:22 PM To: dev@openvswitch.org Subject: [ovs-dev] [PATCH] ovs-router: Define stub for ovs_router_unixctl_register() o

[ovs-dev] [PATCH 08/17] rstp-state-machines: fix TCN reception.

Set rcdv_tcn and return OTHER_INFO when a TOPOLOGY_CHANGE_NOTIFICATION_BPDU is received. Signed-off-by: Daniele Venturino --- lib/rstp-state-machines.c | 8 1 file changed, 8 insertions(+) diff --git a/lib/rstp-state-machines.c b/lib/rstp-state-machines.c index 5398f12.

[ovs-dev] [PATCH 12/17] rstp: Refactor rstp_port_set_administrative_bridge_port__().

Invoke move_rstp__() in rstp_port_set_administrative_bridge_port__() if port is not initializing. Signed-off-by: Daniele Venturino --- lib/rstp.c | 16 1 file changed, 12 insertions(+), 4 deletions(-) diff --git a/lib/rstp.c b/lib/rstp.c index 1ab4938..fd42a7d 10064

[ovs-dev] [PATCH 00/17] RSTP validation tests.

Hi. We tested the RSTP implementation against the IXIA IxANVL validation software. This allowed us to fix some small bugs, and the implementation now passes such tests with success! Please find attached the patch series. It appears that patches 12 and 17 break the "RSTP dummy" test, while all oth

[ovs-dev] [PATCH 07/17] rstp-state-machines: Rename ALTERNATE_AGREED state in ALTERNATE_AGREED_EXEC.

All other similar port_role_transition_sm states are named like this. Signed-off-by: Daniele Venturino --- lib/rstp-common.h | 2 +- lib/rstp-state-machines.c | 4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/lib/rstp-common.h b/lib/rstp-common.h index cd43

[ovs-dev] [PATCH 06/17] rstp-state-machines: Fix updt_roles_tree().

If the DesignatedBridgeID Bridge Address component is equal to that component of the Bridge's own bridge priority vector skip to the next port. Signed-off-by: Daniele Venturino --- lib/rstp-state-machines.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git

[ovs-dev] [PATCH 05/17] rstp-state-machines: Fix topology_change_sm transition transition.

Signed-off-by: Daniele Venturino --- lib/rstp-state-machines.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/lib/rstp-state-machines.c b/lib/rstp-state-machines.c index 1912d7e..5c03385 100644 --- a/lib/rstp-state-machines.c +++ b/lib/rstp-state-machines.c @@ -1948,7 +1948,7

[ovs-dev] [PATCH 03/17] rstp-state-machines: Fix reception of BPDUs conveying an unknown port role.

Signed-off-by: Daniele Venturino --- lib/rstp-state-machines.c | 12 1 file changed, 12 insertions(+) diff --git a/lib/rstp-state-machines.c b/lib/rstp-state-machines.c index e03f720..ade2d61 100644 --- a/lib/rstp-state-machines.c +++ b/lib/rstp-state-machines.c @@ -117,6 +117,18 @@

[ovs-dev] [PATCH 01/17] rstp: Show some useful rstp port fields.

designated_bridge_id, designated_port_id and designated_path_cost are now displayed in rstp_status when using 'ovs-vsctl list port'. Signed-off-by: Daniele Venturino --- lib/rstp.c | 10 -- lib/rstp.h | 6 -- ofproto/ofproto-dpif.c | 4 +++- ofpr

[ovs-dev] [PATCH 04/17] rstp-state-machines: Fix compare_rstp_priority_vector().

The bridge_port_id is never conveyed in Configuration Messages, but is used as a tie-breaker within a Bridge. This patch extends the comparison to this fifth field in a rstp_priority_vector. Signed-off-by: Daniele Venturino --- lib/rstp-state-machines.c | 18 +

[ovs-dev] [PATCH 10/17] rstp: disable learning and forwarding in STP/RSTP disabled state.

There is a difference between a port with STP/RSTP protocol enabled and a disabled role and a port which has a disabled role because STP/RSTP is not active. This commit ensure to make such distinction. Standard 802.1D claims that the Topology Change state machine (17.31)

[ovs-dev] [PATCH 09/17] rstp: Invoke updt_roles_tree__() in max_age and forward_delay setters.

Signed-off-by: Daniele Venturino --- lib/rstp.c | 14 ++ 1 file changed, 10 insertions(+), 4 deletions(-) diff --git a/lib/rstp.c b/lib/rstp.c index e3007e2..55c43c0 100644 --- a/lib/rstp.c +++ b/lib/rstp.c @@ -523,8 +523,9 @@ static void rstp_set_bridge_max_age__(struct rstp *rstp,

[ovs-dev] [PATCH 15/17] rstp: Fix global transitions.

When the condition associated with a global transition is met, it supersedes all other exit conditions including UCT. Signed-off-by: Daniele Venturino --- lib/rstp-state-machines.c | 139 +++--- 1 file changed, 130 insertions(+), 9 deletions(-)

[ovs-dev] [PATCH 02/17] rstp: add admin-point-to-point and admin-port-state setters.

Signed-off-by: Daniele Venturino --- lib/rstp-common.h| 6 -- lib/rstp.c | 31 ++- lib/rstp.h | 9 - ofproto/ofproto-dpif.c | 4 +++- ofproto/ofproto.h| 2 ++ utilities/ovs-vsctl.8.in | 9 + vswitch

[ovs-dev] [PATCH 17/17] rstp: refactor RSTP setters.

With this patch setters invoke procedures only if values have changed. Also rstp_set_bridge_address__() keeps the existing priority in the bridge_identifier. Signed-off-by: Daniele Venturino --- lib/rstp.c | 80 -- 1 f

[ovs-dev] [PATCH 11/17] rstp: Refactor rstp_check_and_reset_fdb_flush().

With this commit, RSTP is able to flush from the MAC learning table entries pertaining to a single port. Before this commit the whole table was flushed every time a port requested flushing actions. Signed-off-by: Daniele Venturino --- lib/rstp.c | 52 +++

[ovs-dev] [PATCH 16/17] rstp-state-machines: fix proposal reception behaviour.

Other ports should stop forwarding and learning when a port receives a superior BPDU carrying a Proposal flag. Without this patch this does not happen and other ports keep executing the learning and forwarding processes. This patch contains some fixes reported in the 8

[ovs-dev] [PATCH 13/17] rstp-state-machines: fix rcv_info().

A Config BPDU always conveys a Designated Port Role. Signed-off-by: Daniele Venturino --- lib/rstp-state-machines.c | 13 - 1 file changed, 8 insertions(+), 5 deletions(-) diff --git a/lib/rstp-state-machines.c b/lib/rstp-state-machines.c index 13abec0..516093f 100644 --- a/l

[ovs-dev] [PATCH 14/17] rstp: shift learned MAC addresses to new Root port.

All MAC addresses previously learned on a Root Port can be moved to an Alternate Port that becomes the new Root Port; i.e., Dynamic Filtering Entries for those addresses may be modified to show the new Root Port as their source, reducing the need to flood frames when recover

Re: [ovs-dev] [PATCH v6 3/3] datapath: add layer 3 flow/port support

On Wed, Nov 5, 2014 at 6:06 PM, Pravin Shelar wrote: > On Wed, Nov 5, 2014 at 10:19 AM, Lori Jakab wrote: >> On 11/5/14 12:16 AM, Pravin Shelar wrote: >>> On Mon, Nov 3, 2014 at 1:36 PM, Lorand Jakab wrote: +static int pop_eth(struct sk_buff *skb) +{ + skb_pull_rcsum(skb, ET

Re: [ovs-dev] [PATCH v6 3/3] datapath: add layer 3 flow/port support

On Thu, Nov 6, 2014 at 7:53 AM, Jesse Gross wrote: > On Wed, Nov 5, 2014 at 6:06 PM, Pravin Shelar wrote: >> On Wed, Nov 5, 2014 at 10:19 AM, Lori Jakab wrote: >>> On 11/5/14 12:16 AM, Pravin Shelar wrote: On Mon, Nov 3, 2014 at 1:36 PM, Lorand Jakab wrote: > +static int pop_eth(struct

Re: [ovs-dev] [PATCH 00/17] RSTP validation tests.

Thanks for testing, I'll review this and check out the dummy test case. Jarno > On Nov 6, 2014, at 7:30 AM, Daniele Venturino > wrote: > > Hi. > We tested the RSTP implementation against the IXIA IxANVL validation software. > This allowed us to fix some small bugs, and the implementation no

Re: [ovs-dev] [PATCH v2] openvswitch: Userspace tunneling.

On Wed, Nov 5, 2014 at 4:44 PM, Ben Pfaff wrote: > On Sun, Nov 02, 2014 at 09:29:28PM -0800, Pravin B Shelar wrote: >> Following patch adds support for userspace tunneling. Tunneling >> needs three more component first is routing table which is configured by >> caching kernel routes and second is

[ovs-dev] [PATCH] ovs-ofctl: Fix memory exhaustion failures in corner cases.

Inserting or removing a sequence of flows with different wildcard patterns causes temporary use of O(n**2) memory due to pvector modifications inside the classifier. This commit fixes the problem in two easy cases. There is at least one more difficult case inside ovs-vswitchd that this does not f

Re: [ovs-dev] [PATCH v2] openvswitch: Userspace tunneling.

On Nov 6, 2014, at 9:36 AM, Pravin Shelar wrote: >> I think that all of the flows added to the tnl-ports classifier have >> the same wildcard pattern. I don't know why a classifier is the best >> choice; wouldn't an hmap work? >> The lookup should be RCU, so hmap would not work. cmap would th

Re: [ovs-dev] [PATCH v2] openvswitch: Userspace tunneling.

On Thu, Nov 06, 2014 at 10:19:14AM -0800, Jarno Rajahalme wrote: > On Nov 6, 2014, at 9:36 AM, Pravin Shelar wrote: > >> I think that all of the flows added to the tnl-ports classifier have > >> the same wildcard pattern. I don't know why a classifier is the best > >> choice; wouldn't an hmap wor

Re: [ovs-dev] [dp version v6] bridge: Store datapath version into ovsdb

On Wed, Nov 05, 2014 at 09:57:25PM -0200, Flavio Leitner wrote: > On Wed, Nov 05, 2014 at 03:51:30PM -0800, Andy Zhou wrote: > > On Wed, Nov 5, 2014 at 3:47 PM, Flavio Leitner wrote: > > > On Wed, Nov 05, 2014 at 03:23:24PM -0800, Andy Zhou wrote: > > >> On Wed, Nov 5, 2014 at 2:56 PM, Flavio Leit

Re: [ovs-dev] [PATCH v2] openvswitch: Userspace tunneling.

On Thu, Nov 06, 2014 at 09:36:19AM -0800, Pravin Shelar wrote: > On Wed, Nov 5, 2014 at 4:44 PM, Ben Pfaff wrote: > > On Sun, Nov 02, 2014 at 09:29:28PM -0800, Pravin B Shelar wrote: > > I don't understand why struct udp_header is now marked as packed, or > > why struct gre_base_hdr is marked as a

Re: [ovs-dev] [dp version v6] bridge: Store datapath version into ovsdb

On Thu, Nov 06, 2014 at 10:33:54AM -0800, Ben Pfaff wrote: > On Wed, Nov 05, 2014 at 09:57:25PM -0200, Flavio Leitner wrote: > > On Wed, Nov 05, 2014 at 03:51:30PM -0800, Andy Zhou wrote: > > > On Wed, Nov 5, 2014 at 3:47 PM, Flavio Leitner wrote: > > > > On Wed, Nov 05, 2014 at 03:23:24PM -0800,

Re: [ovs-dev] [PATCH] ovs-ofctl: Fix memory exhaustion failures in corner cases.

With notes below: Acked-by: Jarno Rajahalme On Nov 6, 2014, at 10:03 AM, Ben Pfaff wrote: > Inserting or removing a sequence of flows with different wildcard patterns > causes temporary use of O(n**2) memory due to pvector modifications inside > the classifier. This commit fixes the problem i

[ovs-dev] [PATCH 01/33] fedora.spec: added license details

Signed-off-by: Flavio Leitner --- rhel/openvswitch-fedora.spec.in | 7 +-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/rhel/openvswitch-fedora.spec.in b/rhel/openvswitch-fedora.spec.in index 0a1fe00..dd61449 100644 --- a/rhel/openvswitch-fedora.spec.in +++ b/rhel/openvswitch

[ovs-dev] [PATCH 00/33] Update openvswitch fedora spec file

This patchset updates the fedora spec file to be compliant with the distro guidelines and includes some fixes as well. Flavio Leitner (33): fedora.spec: added license details fedora-spec: use default buildroot fedora-spec: remove stale comment fedora-spec: source line points to upstream ta

[ovs-dev] [PATCH 02/33] fedora-spec: use default buildroot

Signed-off-by: Flavio Leitner --- rhel/openvswitch-fedora.spec.in | 1 - 1 file changed, 1 deletion(-) diff --git a/rhel/openvswitch-fedora.spec.in b/rhel/openvswitch-fedora.spec.in index dd61449..d036acb 100644 --- a/rhel/openvswitch-fedora.spec.in +++ b/rhel/openvswitch-fedora.spec.in @@ -27,7

[ovs-dev] [PATCH 03/33] fedora-spec: remove stale comment

Signed-off-by: Flavio Leitner --- rhel/openvswitch-fedora.spec.in | 1 - 1 file changed, 1 deletion(-) diff --git a/rhel/openvswitch-fedora.spec.in b/rhel/openvswitch-fedora.spec.in index d036acb..1ee08fc 100644 --- a/rhel/openvswitch-fedora.spec.in +++ b/rhel/openvswitch-fedora.spec.in @@ -26,7

[ovs-dev] [PATCH 04/33] fedora-spec: source line points to upstream tarball

Signed-off-by: Flavio Leitner --- rhel/openvswitch-fedora.spec.in | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/rhel/openvswitch-fedora.spec.in b/rhel/openvswitch-fedora.spec.in index 1ee08fc..776cfbf 100644 --- a/rhel/openvswitch-fedora.spec.in +++ b/rhel/openvswitch-fedora

[ovs-dev] [PATCH 05/33] fedora-spec: Added buildrequires

This is needed to ensure all the packages are installed on the system to be able to build Open vSwitch. This also permits to use yum-builddep(1) to automatically install the dependencies on the system. Signed-off-by: Flavio Leitner --- rhel/openvswitch-fedora.spec.in | 6 ++ 1 file changed,

[ovs-dev] [PATCH 06/33] fedora-spec: removed extra space

Signed-off-by: Flavio Leitner --- rhel/openvswitch-fedora.spec.in | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/rhel/openvswitch-fedora.spec.in b/rhel/openvswitch-fedora.spec.in index 42a816e..0413cf1 100644 --- a/rhel/openvswitch-fedora.spec.in +++ b/rhel/openvswitch-fedora

[ovs-dev] [PATCH 07/33] fedora-spec: added more requires

The minimum recommended kernel for fedora is 3.15.0-0 due to be the first one including the upstream kernel commit below: commit 4f647e0a3c37b8d5086214128614a136064110c3 Author: Flavio Leitner Date: Thu Mar 27 11:05:34 2014 -0300 openvswitch: fix a possible deadlock and lockdep warning

[ovs-dev] [PATCH 08/33] fedora-spec: more simple description

Signed-off-by: Flavio Leitner --- rhel/openvswitch-fedora.spec.in | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/rhel/openvswitch-fedora.spec.in b/rhel/openvswitch-fedora.spec.in index 26d4b37..db62455 100644 --- a/rhel/openvswitch-fedora.spec.in +++ b/rhel/openvswitch-fedora

[ovs-dev] [PATCH 09/33] fedora-spec: use %configure macro

Signed-off-by: Flavio Leitner --- rhel/openvswitch-fedora.spec.in | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/rhel/openvswitch-fedora.spec.in b/rhel/openvswitch-fedora.spec.in index db62455..3e19c91 100644 --- a/rhel/openvswitch-fedora.spec.in +++ b/rhel/openvswitch-fedora

[ovs-dev] [PATCH 10/33] fedora-spec: add python-openvswitch subpackage

This creates a subpackage called python-openvswitch providing python bidings for the Open vSwitch database. Signed-off-by: Flavio Leitner --- rhel/openvswitch-fedora.spec.in | 55 + 1 file changed, 17 insertions(+), 38 deletions(-) diff --git a/rhel/openv

[ovs-dev] [PATCH 11/33] fedora-spec: add openvswitch-test package

Put the utilities that are useful to diagnose performance and connectivity issues in Open vSwitch setup into another package since they are not needed in most cases. Signed-off-by: Flavio Leitner --- rhel/openvswitch-fedora.spec.in | 27 --- 1 file changed, 20 insertions(

[ovs-dev] [PATCH 13/33] fedora-spec: Enable PIE

Package guidelines request to enable PIE. Signed-off-by: Flavio Leitner --- rhel/openvswitch-fedora.spec.in | 3 +++ 1 file changed, 3 insertions(+) diff --git a/rhel/openvswitch-fedora.spec.in b/rhel/openvswitch-fedora.spec.in index 9dc607c..401698d 100644 --- a/rhel/openvswitch-fedora.spec.in

[ovs-dev] [PATCH 17/33] fedora-spec: the db and id must be owned

Both conf.db and system-id.conf needs to be owned, so spec now list them in the %files section. However, they are not shipped with the rpm (%ghost) Signed-off-by: Flavio Leitner --- rhel/openvswitch-fedora.spec.in | 11 --- 1 file changed, 8 insertions(+), 3 deletions(-) diff --git a/r

[ovs-dev] [PATCH 12/33] fedora-spec: add openvswitch-devel subpackage

This provides static library, libopenswitch.a and the openvswitch header files needed to build an external application in a separate RPM package. Signed-off-by: Flavio Leitner --- rhel/openvswitch-fedora.spec.in | 30 -- 1 file changed, 28 insertions(+), 2 deletions(-

[ovs-dev] [PATCH 20/33] fedora-spec: use macro for /usr/sbin

Signed-off-by: Flavio Leitner --- rhel/openvswitch-fedora.spec.in | 8 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/rhel/openvswitch-fedora.spec.in b/rhel/openvswitch-fedora.spec.in index c97d0ef..f1f283c 100644 --- a/rhel/openvswitch-fedora.spec.in +++ b/rhel/openvswitc

[ovs-dev] [PATCH 14/33] fedora-spec: obsolete controller subpackage

Once upon a time there was a controller subpackage that was obsoleted Signed-off-by: Flavio Leitner --- rhel/openvswitch-fedora.spec.in | 1 + 1 file changed, 1 insertion(+) diff --git a/rhel/openvswitch-fedora.spec.in b/rhel/openvswitch-fedora.spec.in index 401698d..b226218 100644 --- a/rhel/o

[ovs-dev] [PATCH 15/33] fedora-spec: _smp_mflags macro must be conditional

Signed-off-by: Flavio Leitner --- rhel/openvswitch-fedora.spec.in | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/rhel/openvswitch-fedora.spec.in b/rhel/openvswitch-fedora.spec.in index b226218..456e5d5 100644 --- a/rhel/openvswitch-fedora.spec.in +++ b/rhel/openvswitch-fedora

[ovs-dev] [PATCH 18/33] fedora-spec: sort man-pages and use macros

Signed-off-by: Flavio Leitner --- rhel/openvswitch-fedora.spec.in | 37 - 1 file changed, 20 insertions(+), 17 deletions(-) diff --git a/rhel/openvswitch-fedora.spec.in b/rhel/openvswitch-fedora.spec.in index 3834339..a75175a 100644 --- a/rhel/openvswitch-fedo

[ovs-dev] [PATCH 19/33] fedora-spec: break %doc long line

Signed-off-by: Flavio Leitner --- rhel/openvswitch-fedora.spec.in | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/rhel/openvswitch-fedora.spec.in b/rhel/openvswitch-fedora.spec.in index a75175a..c97d0ef 100644 --- a/rhel/openvswitch-fedora.spec.in +++ b/rhel/openvswitch-fedo

[ovs-dev] [PATCH 21/33] fedora-spec: use macro for /usr/bin

Signed-off-by: Flavio Leitner --- rhel/openvswitch-fedora.spec.in | 30 +++--- 1 file changed, 15 insertions(+), 15 deletions(-) diff --git a/rhel/openvswitch-fedora.spec.in b/rhel/openvswitch-fedora.spec.in index f1f283c..6d0aca4 100644 --- a/rhel/openvswitch-fedora.spec

[ovs-dev] [PATCH 22/33] fedora-spec: use macro for /usr/share

Signed-off-by: Flavio Leitner --- rhel/openvswitch-fedora.spec.in | 18 +- 1 file changed, 9 insertions(+), 9 deletions(-) diff --git a/rhel/openvswitch-fedora.spec.in b/rhel/openvswitch-fedora.spec.in index 6d0aca4..daa5d73 100644 --- a/rhel/openvswitch-fedora.spec.in +++ b/rhel

[ovs-dev] [PATCH 16/33] fedora-spec: use default tarball dirs name

The default tarball directory follows the standard. Signed-off-by: Flavio Leitner --- rhel/openvswitch-fedora.spec.in | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/rhel/openvswitch-fedora.spec.in b/rhel/openvswitch-fedora.spec.in index 456e5d5..03bf0b2 100644 --- a/rhel/ope

[ovs-dev] [PATCH 23/33] fedora-spec: convert missing man-pages to use macro

Signed-off-by: Flavio Leitner --- rhel/openvswitch-fedora.spec.in | 8 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/rhel/openvswitch-fedora.spec.in b/rhel/openvswitch-fedora.spec.in index daa5d73..8ccfc2c 100644 --- a/rhel/openvswitch-fedora.spec.in +++ b/rhel/openvswitc

[ovs-dev] [PATCH 24/33] fedora-spec: do not replace logrotate cfg if modified

Don't overwrite if the logrotate configuration file is modified. Signed-off-by: Flavio Leitner --- rhel/openvswitch-fedora.spec.in | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/rhel/openvswitch-fedora.spec.in b/rhel/openvswitch-fedora.spec.in index 8ccfc2c..96b808e 100644 -

[ovs-dev] [PATCH 25/33] fedora-spec: use RPM macros in %install section

Signed-off-by: Flavio Leitner --- rhel/openvswitch-fedora.spec.in | 26 +- 1 file changed, 13 insertions(+), 13 deletions(-) diff --git a/rhel/openvswitch-fedora.spec.in b/rhel/openvswitch-fedora.spec.in index 96b808e..080a460 100644 --- a/rhel/openvswitch-fedora.spec.in

[ovs-dev] [PATCH 26/33] fedora-spec: fix indentation in %install section

Use spaces instead of tabs as the other lines Signed-off-by: Flavio Leitner --- rhel/openvswitch-fedora.spec.in | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/rhel/openvswitch-fedora.spec.in b/rhel/openvswitch-fedora.spec.in index 080a460..1d73435 100644 --- a/rhel/ope

[ovs-dev] [PATCH 28/33] fedora-spec: use empty lines to split blocks in %install

Use empty lines to split blocks to improve readability. Signed-off-by: Flavio Leitner --- rhel/openvswitch-fedora.spec.in | 11 +++ 1 file changed, 11 insertions(+) diff --git a/rhel/openvswitch-fedora.spec.in b/rhel/openvswitch-fedora.spec.in index 1d81567..b48e971 100644 --- a/rhel/op

[ovs-dev] [PATCH 29/33] fedora-spec: remove uneeded lines

Signed-off-by: Flavio Leitner --- rhel/openvswitch-fedora.spec.in | 12 +--- 1 file changed, 1 insertion(+), 11 deletions(-) diff --git a/rhel/openvswitch-fedora.spec.in b/rhel/openvswitch-fedora.spec.in index b48e971..a51365d 100644 --- a/rhel/openvswitch-fedora.spec.in +++ b/rhel/openv

[ovs-dev] [PATCH 30/33] fedora-spec: break long lines in %install

Signed-off-by: Flavio Leitner --- rhel/openvswitch-fedora.spec.in | 15 ++- 1 file changed, 10 insertions(+), 5 deletions(-) diff --git a/rhel/openvswitch-fedora.spec.in b/rhel/openvswitch-fedora.spec.in index a51365d..6e169b4 100644 --- a/rhel/openvswitch-fedora.spec.in +++ b/rhel/o

[ovs-dev] [PATCH 31/33] fedora-spec: move sysconfig template

Move the sysconfig template close to other systemd lines. Signed-off-by: Flavio Leitner --- rhel/openvswitch-fedora.spec.in | 7 +++ 1 file changed, 3 insertions(+), 4 deletions(-) diff --git a/rhel/openvswitch-fedora.spec.in b/rhel/openvswitch-fedora.spec.in index 6e169b4..67b7ca9 100644 -

[ovs-dev] [PATCH 32/33] fedora-spec: added systemd post/postun/pre/preun sections

The systemd requires some actions when the package is installed, upgraded or removed. This patch adds the needed RPM sections with the missing systemd actions. There is a change in behavior - the service is not enabled or started by default. Signed-off-by: Flavio Leitner --- rhel/openvswitch-fe

[ovs-dev] [PATCH 33/33] fedora-spec: remove unneeded db initialization

The DB initialization is done by ovs-ctl script as well, so remove the duplicated code from the spec. Signed-off-by: Flavio Leitner --- rhel/openvswitch-fedora.spec.in | 13 - 1 file changed, 13 deletions(-) diff --git a/rhel/openvswitch-fedora.spec.in b/rhel/openvswitch-fedora.spec

[ovs-dev] [PATCH] rhel: fix tunnel port ifup/ifdown failure

The tunnel port is invisible to the OS, so there is no reason to call OTHERSCRIPT in the ifup/ifdown scripts. Signed-off-by: Flavio Leitner --- rhel/etc_sysconfig_network-scripts_ifdown-ovs | 4 ++-- rhel/etc_sysconfig_network-scripts_ifup-ovs | 1 - 2 files changed, 2 insertions(+), 3 deletio

[ovs-dev] [PATCH] meta-flow: Make it simpler to expand mf_values in the future

From: Madhu Challa Remove hard coded array index and make it dependent on the array size. Signed-off-by: Madhu Challa Co-authored-by: Ben Pfaff Signed-off-by: Ben Pfaff --- lib/learn.c | 6 -- lib/meta-flow.h | 2 ++ 2 files changed, 6 insertions(+), 2 deletions(-) diff --git a/lib/

Re: [ovs-dev] [PATCH v2] Build: shared libraries and versioning

Thanks. On Tue, Nov 04, 2014 at 06:45:02PM -0700, Scott Mann wrote: > Ben, > Yes, I will. But it will likely be this weekend before I am able to do that > (in traveling). > Scott > > On Nov 4, 2014 3:51 PM, "Ben Pfaff" wrote: > > > > On Wed, Oct 29, 2014 at 02:55:39PM -0600, Scott Mann wrote: >

Re: [ovs-dev] [PATCH v2 1/5] classifier: Constify RCU pointers.

On Mon, Nov 03, 2014 at 11:39:00AM -0800, Jarno Rajahalme wrote: > Returning const struct cls_rule pointers from the classifier API helps > callers to remember that they should not modify the rules returned. > > Signed-off-by: Jarno Rajahalme I don't think it has much practical effect since most

Re: [ovs-dev] [PATCH] meta-flow: Make it simpler to expand mf_values in the future

On Thu, Nov 06, 2014 at 10:57:13AM -0800, Ben Pfaff wrote: > From: Madhu Challa > > Remove hard coded array index and make it dependent on the array size. > > Signed-off-by: Madhu Challa > Co-authored-by: Ben Pfaff > Signed-off-by: Ben Pfaff Madhu looked at this, so I applied it. Thanks, B

Re: [ovs-dev] [PATCH v2 1/5] classifier: Constify RCU pointers.

On Thu, Nov 06, 2014 at 11:02:56AM -0800, Ben Pfaff wrote: > On Mon, Nov 03, 2014 at 11:39:00AM -0800, Jarno Rajahalme wrote: > > Returning const struct cls_rule pointers from the classifier API helps > > callers to remember that they should not modify the rules returned. > > > > Signed-off-by: Ja

Re: [ovs-dev] [PATCH v2 2/5] classifier: Make classifier_find_rule_exactly() lockless.

On Mon, Nov 03, 2014 at 11:39:01AM -0800, Jarno Rajahalme wrote: > struct cls_match 'list' member was recently changed to an rculist. > This allows classifier_find_rule_exactly() to be made lockless. > > Since subtable's 'max_priority' member would still require a lock, we > no longer check it bef

Re: [ovs-dev] [PATCH v2 3/5] classifier: Constify fields.

On Mon, Nov 03, 2014 at 11:39:02AM -0800, Jarno Rajahalme wrote: > Some struct cls_match and cls_subtable fields were already documented > of being const. Make them const and use CONST_CAST where appropriate > to initialize them. > > This will help catch future errors modifying those fields after

Re: [ovs-dev] [PATCH v2 1/5] classifier: Constify RCU pointers.

On Nov 6, 2014, at 11:08 AM, Ben Pfaff wrote: > On Thu, Nov 06, 2014 at 11:06:59AM -0800, Ben Pfaff wrote: >> On Thu, Nov 06, 2014 at 11:02:56AM -0800, Ben Pfaff wrote: >>> On Mon, Nov 03, 2014 at 11:39:00AM -0800, Jarno Rajahalme wrote: Returning const struct cls_rule pointers from the cla

Re: [ovs-dev] [PATCH v2 4/5] classifier: Clean up destroy_subtable.

On Mon, Nov 03, 2014 at 11:39:03AM -0800, Jarno Rajahalme wrote: > Add asserts to make sure the containers within are already empty. > > Signed-off-by: Jarno Rajahalme Acked-by: Ben Pfaff ___ dev mailing list dev@openvswitch.org http://openvswitch.org

Re: [ovs-dev] [PATCH] ovs-ofctl: Fix memory exhaustion failures in corner cases.

On Nov 6, 2014, at 10:50 AM, Jarno Rajahalme wrote: >> +for i in `seq 1 5`; do echo "reg0=$i/$i actions=drop"; done > flows Running this test case alone with ‘time’ added to the preceding line: real1m12.531s user0m2.492s sys 0m6.928s >> +AT_CHECK([ovs-ofctl diff-flows flows fl

Re: [ovs-dev] [PATCH v2 1/5] classifier: Constify RCU pointers.

On Thu, Nov 06, 2014 at 11:06:59AM -0800, Ben Pfaff wrote: > On Thu, Nov 06, 2014 at 11:02:56AM -0800, Ben Pfaff wrote: > > On Mon, Nov 03, 2014 at 11:39:00AM -0800, Jarno Rajahalme wrote: > > > Returning const struct cls_rule pointers from the classifier API helps > > > callers to remember that th

Re: [ovs-dev] [PATCH v2 1/5] classifier: Constify RCU pointers.

On Thu, Nov 06, 2014 at 11:29:26AM -0800, Jarno Rajahalme wrote: > > On Nov 6, 2014, at 11:08 AM, Ben Pfaff wrote: > > > On Thu, Nov 06, 2014 at 11:06:59AM -0800, Ben Pfaff wrote: > >> On Thu, Nov 06, 2014 at 11:02:56AM -0800, Ben Pfaff wrote: > >>> On Mon, Nov 03, 2014 at 11:39:00AM -0800, Jarn

Re: [ovs-dev] [PATCH] ovs-ofctl: Fix memory exhaustion failures in corner cases.

On Thu, Nov 06, 2014 at 11:34:41AM -0800, Jarno Rajahalme wrote: > > On Nov 6, 2014, at 10:50 AM, Jarno Rajahalme wrote: > > >> +for i in `seq 1 5`; do echo "reg0=$i/$i actions=drop"; done > flows > > Running this test case alone with ‘time’ added to the preceding line: > > real 1m12.531s

[ovs-dev] classifier mutex

The classifier has an internal mutex, but in ofproto that mutex is redundant with the external ofproto_mutex. If the classifier mutex were removed, to be replaced by documentation that certain routines need mutual exclusion, then possibly the reduced locking could yield a performance improvement.

Re: [ovs-dev] [PATCH v6 3/3] datapath: add layer 3 flow/port support

On 11/6/14 4:06 AM, Pravin Shelar wrote: +static int push_eth(struct sk_buff *skb, const struct ovs_action_push_eth *ethh) +{ + /* De-accelerate any hardware accelerated VLAN tag added to a previous +* Ethernet header */ + if (unlikely(vlan_tx_tag_present(skb))) { +

Re: [ovs-dev] [PATCH v6 3/3] datapath: add layer 3 flow/port support

On 11/6/14 7:17 PM, Pravin Shelar wrote: On Thu, Nov 6, 2014 at 7:53 AM, Jesse Gross wrote: On Wed, Nov 5, 2014 at 6:06 PM, Pravin Shelar wrote: On Wed, Nov 5, 2014 at 10:19 AM, Lori Jakab wrote: On 11/5/14 12:16 AM, Pravin Shelar wrote: On Mon, Nov 3, 2014 at 1:36 PM, Lorand Jakab wrote:

Re: [ovs-dev] [PATCH] ovs-ofctl: Fix memory exhaustion failures in corner cases.

On Thu, Nov 06, 2014 at 10:50:09AM -0800, Jarno Rajahalme wrote: > With notes below: > > Acked-by: Jarno Rajahalme > > On Nov 6, 2014, at 10:03 AM, Ben Pfaff wrote: > > > Inserting or removing a sequence of flows with different wildcard patterns > > causes temporary use of O(n**2) memory due t

Re: [ovs-dev] [PATCH v2 5/5] classifier: Lockless and robust classifier iteration.

On Mon, Nov 03, 2014 at 11:39:04AM -0800, Jarno Rajahalme wrote: > Previously, accurate iteration required writers to be excluded during > iteration. This patch adds an rculist to struct cls_subtable, and a > corresponding list node to struct cls_rule, which makes iteration more > straightforward,

Re: [ovs-dev] [PATCH v2 5/5] classifier: Lockless and robust classifier iteration.

On Nov 6, 2014, at 12:34 PM, Ben Pfaff wrote: > On Mon, Nov 03, 2014 at 11:39:04AM -0800, Jarno Rajahalme wrote: >> Previously, accurate iteration required writers to be excluded during >> iteration. This patch adds an rculist to struct cls_subtable, and a >> corresponding list node to struct c

Re: [ovs-dev] classifier mutex

On Nov 6, 2014, at 12:21 PM, Ben Pfaff wrote: > The classifier has an internal mutex, but in ofproto that mutex is > redundant with the external ofproto_mutex. If the classifier mutex were > removed, to be replaced by documentation that certain routines need > mutual exclusion, then possibly th

Re: [ovs-dev] [PATCH] ovs-ofctl: Fix memory exhaustion failures in corner cases.

On Nov 6, 2014, at 12:28 PM, Ben Pfaff wrote: > On Thu, Nov 06, 2014 at 10:50:09AM -0800, Jarno Rajahalme wrote: >> With notes below: >> >> Acked-by: Jarno Rajahalme >> >> On Nov 6, 2014, at 10:03 AM, Ben Pfaff wrote: >> >>> Inserting or removing a sequence of flows with different wildcard

Re: [ovs-dev] [PATCH] FAQ: Describe how to add new fields and new actions.

I am in the process adding a new action. The steps described are good starting points. Thanks for documenting them. Acked-by: Andy Zhou On Tue, Nov 4, 2014 at 10:59 AM, Ben Pfaff wrote: > We get these questions from time to time and it would be nice to just be > able to cut and paste the answ

Re: [ovs-dev] [PATCH] ovs-ofctl: Fix memory exhaustion failures in corner cases.

On Thu, Nov 06, 2014 at 01:16:39PM -0800, Jarno Rajahalme wrote: > > On Nov 6, 2014, at 12:28 PM, Ben Pfaff wrote: > > > On Thu, Nov 06, 2014 at 10:50:09AM -0800, Jarno Rajahalme wrote: > >> With notes below: > >> > >> Acked-by: Jarno Rajahalme > >> > >> On Nov 6, 2014, at 10:03 AM, Ben Pfaff

Re: [ovs-dev] classifier mutex

On Thu, Nov 06, 2014 at 01:09:20PM -0800, Jarno Rajahalme wrote: > > On Nov 6, 2014, at 12:21 PM, Ben Pfaff wrote: > > > The classifier has an internal mutex, but in ofproto that mutex is > > redundant with the external ofproto_mutex. If the classifier mutex were > > removed, to be replaced by

Re: [ovs-dev] [PATCH] FAQ: Describe how to add new fields and new actions.

Thanks! Applied. On Thu, Nov 06, 2014 at 01:20:41PM -0800, Andy Zhou wrote: > I am in the process adding a new action. The steps described are good > starting points. Thanks for documenting them. > > Acked-by: Andy Zhou > > > On Tue, Nov 4, 2014 at 10:59 AM, Ben Pfaff wrote: > > We get thes

Re: [ovs-dev] [PATCH 00/33] Update openvswitch fedora spec file

On Thu, Nov 06, 2014 at 04:45:18PM -0200, Flavio Leitner wrote: > This patchset updates the fedora spec file to be > compliant with the distro guidelines and includes > some fixes as well. I can tell when I'm outclassed ;-) Applied. ___ dev mailing list

Re: [ovs-dev] [PATCH] rhel: fix tunnel port ifup/ifdown failure

On Thu, Nov 06, 2014 at 04:50:06PM -0200, Flavio Leitner wrote: > The tunnel port is invisible to the OS, so there is > no reason to call OTHERSCRIPT in the ifup/ifdown scripts. > > Signed-off-by: Flavio Leitner Applied, thanks! Also backported to branch-2.3. ___

Re: [ovs-dev] [PATCH v2 5/5] classifier: Lockless and robust classifier iteration.

On Thu, Nov 06, 2014 at 01:04:39PM -0800, Jarno Rajahalme wrote: > > On Nov 6, 2014, at 12:34 PM, Ben Pfaff wrote: > > > On Mon, Nov 03, 2014 at 11:39:04AM -0800, Jarno Rajahalme wrote: > >> Previously, accurate iteration required writers to be excluded during > >> iteration. This patch adds an

Re: [ovs-dev] [PATCH] ovs-ofctl: Fix memory exhaustion failures in corner cases.

On Nov 6, 2014, at 1:40 PM, Ben Pfaff wrote: > On Thu, Nov 06, 2014 at 01:16:39PM -0800, Jarno Rajahalme wrote: >> >> On Nov 6, 2014, at 12:28 PM, Ben Pfaff wrote: >> >>> On Thu, Nov 06, 2014 at 10:50:09AM -0800, Jarno Rajahalme wrote: With notes below: Acked-by: Jarno Rajahal

[ovs-dev] [PATCH] ovs-vsctl: Prevent creating duplicate VLAN bridges.

ovs-vsctl has the concept of a VLAN (or "fake") bridge, which is a sort of a sub-bridge that receives only packets on a particular VLAN. There is no way to distinguish two VLAN bridges with the same parent on the same VLAN, but until now ovs-vsctl did not prevent creating duplicates or report them.

Re: [ovs-dev] [PATCH] ovs-ofctl: Only allow usable protocols for group commands

On Wed, Nov 05, 2014 at 01:57:39PM +0900, Simon Horman wrote: > parse_ofp_group_mod_str() may limit the usable protocols according > to the group and in particular its actions. However, without this > change ovs-ofctl ignores this calculation. > > Signed-off-by: Simon Horman It looks useful, but

Re: [ovs-dev] [PATCH v2 1/5] classifier: Constify RCU pointers.

Thanks for the review! Pushed with the change that classifier_remove takes a const struct cls_rule pointer as an argument. Jarno On Nov 6, 2014, at 11:58 AM, Ben Pfaff wrote: > On Thu, Nov 06, 2014 at 11:29:26AM -0800, Jarno Rajahalme wrote: >> >> On Nov 6, 2014, at 11:08 AM, Ben Pfaff wr

Re: [ovs-dev] [RFC PATCH v2 1/2] miniflow_extract: Properly handle small IP packets.

On Wed, Nov 05, 2014 at 10:37:18AM -0800, Jarno Rajahalme wrote: > Ethernet frames may contain padding after the IP payload. When > parsing IP packets, check the IP total size (IPv4) or IP payload size > (IPv6) to detect the size of l2 padding. The l2 padding size is > stored in the ofpbuf to pre

Re: [ovs-dev] [PATCH v2 2/5] classifier: Make classifier_find_rule_exactly() lockless.

Updated the commit message according to your comment and pushed to master, thanks for the review! Jarno On Nov 6, 2014, at 11:24 AM, Ben Pfaff wrote: > On Mon, Nov 03, 2014 at 11:39:01AM -0800, Jarno Rajahalme wrote: >> struct cls_match 'list' member was recently changed to an rculist. >> Th

Re: [ovs-dev] [PATCH v2 3/5] classifier: Constify fields.

On Nov 6, 2014, at 11:25 AM, Ben Pfaff wrote: > On Mon, Nov 03, 2014 at 11:39:02AM -0800, Jarno Rajahalme wrote: >> Some struct cls_match and cls_subtable fields were already documented >> of being const. Make them const and use CONST_CAST where appropriate >> to initialize them. >> >> This wi

Re: [ovs-dev] [PATCH v2 4/5] classifier: Clean up destroy_subtable.

On Nov 6, 2014, at 11:28 AM, Ben Pfaff wrote: > On Mon, Nov 03, 2014 at 11:39:03AM -0800, Jarno Rajahalme wrote: >> Add asserts to make sure the containers within are already empty. >> >> Signed-off-by: Jarno Rajahalme > > Acked-by: Ben Pfaff Pushed, thanks! Jarno __

  1   2   >