cascardo/ovs.git
9 years agoFix build break in ofproto/tunnel.c for windows platform.
Saurabh Shah [Wed, 19 Nov 2014 11:28:41 +0000 (03:28 -0800)]
Fix build break in ofproto/tunnel.c for windows platform.

The breakage was introduced by commit: a36de779
("openvswitch: Userspace tunneling.").

Reported-by: Edwin Chiu <echiu@nicira.com>
Signed-off-by: Saurabh Shah <ssaurabh@vmware.com>
Acked-by: Nithin Raju <nithin@vmware.com>
Signed-off-by: Gurucharan Shetty <gshetty@nicira.com>
9 years agodatapath: compat: Fix build on RHEL 6.6
Pravin B Shelar [Wed, 12 Nov 2014 02:56:34 +0000 (18:56 -0800)]
datapath: compat: Fix build on RHEL 6.6

RHEL 6.6 kernel percpu APIs are broken, so following patch is using OVS
backported version.

Reported-by: Wang Sheng-Hui <shhuiw@gmail.com>
Signed-off-by: Pravin B Shelar <pshelar@nicira.com>
9 years agoofproto-dpif: fix bundle_move() call in rstp_run().
Daniele Venturino [Wed, 19 Nov 2014 21:16:17 +0000 (22:16 +0100)]
ofproto-dpif: fix bundle_move() call in rstp_run().

bundle_move() expects two struct ofbundle pointers as arguments, but
void pointers hide the error.

Signed-off-by: Daniele Venturino <daniele.venturino@m3s.it>
Acked-by: Jarno Rajahalme <jrajahalme@nicira.com>
9 years agorstp: refactor RSTP setters.
Daniele Venturino [Thu, 6 Nov 2014 15:31:15 +0000 (16:31 +0100)]
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 <daniele.venturino@m3s.it>
Acked-by: Jarno Rajahalme <jrajahalme@nicira.com>
9 years agorstp-state-machines: fix proposal reception behaviour.
Jarno Rajahalme [Fri, 14 Nov 2014 21:58:25 +0000 (13:58 -0800)]
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 802.1q-2008 standard.

Signed-off-by: Daniele Venturino <daniele.venturino@m3s.it>
Acked-by: Jarno Rajahalme <jrajahalme@nicira.com>
9 years agorstp: Fix global transitions.
Jarno Rajahalme [Wed, 19 Nov 2014 17:27:56 +0000 (09:27 -0800)]
rstp: Fix global transitions.

Global transitions are highest priority transitions.  When the
condition associated with a global transition is met, it supersedes
all other exit conditions including UCT.

Extracted from 802.1D-2004 standard (17.16):

A transition that is global in nature (i.e., a transition that occurs
from any of the possible states if the condition attached to the arrow
is met) is denoted by an open arrow, i.e., no specific state is
identified as the origin of the transition.  When the condition
associated with a global transition is met, it supersedes all other
exit conditions including UCT.  The special global condition BEGIN
supersedes all other global conditions, and once asserted remains
asserted until all state blocks have executed to the point that
variable assignments and other consequences of their execution remain
unchanged.

Signed-off-by: Daniele Venturino <daniele.venturino@m3s.it>
Acked-by: Jarno Rajahalme <jrajahalme@nicira.com>
9 years agorstp: shift learned MAC addresses to new Root port.
Jarno Rajahalme [Wed, 19 Nov 2014 17:24:52 +0000 (09:24 -0800)]
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 recovering from
some component failures.

Signed-off-by: Daniele Venturino <daniele.venturino@m3s.it>
Signed-off-by: Jarno Rajahalme <jrajahalme@nicira.com>
Acked-by: Jarno Rajahalme <jrajahalme@nicira.com>
9 years agoclassifier: Defer pvector publication.
Jarno Rajahalme [Thu, 13 Nov 2014 19:54:31 +0000 (11:54 -0800)]
classifier: Defer pvector publication.

This patch adds a new functions classifier_defer() and
classifier_publish(), which control when the classifier modifications
are made available to lookups.  By default, all modifications are made
available to lookups immediately.  Modifications made after a
classifier_defer() call MAY be 'deferred' for later 'publication'.  A
call to classifier_publish() will both publish any deferred
modifications, and cause subsequent changes to to be published
immediately.

Currently any deferring is limited to the visibility of the subtable
vector changes.  pvector now processes modifications mostly in a
working copy, which needs to be explicitly published with
pvector_publish().  pvector_publish() sorts the working copy and
removes gaps before publishing it.

This change helps avoiding O(n**2) memory behavior in corner cases,
where large number of rules with different masks are inserted or
deleted.

VMware-BZ: #1322017
Signed-off-by: Jarno Rajahalme <jrajahalme@nicira.com>
Acked-by: Ben Pfaff <blp@nicira.com>
9 years agoclassifier: Make insert and replace take a const rule, too.
Jarno Rajahalme [Thu, 13 Nov 2014 19:54:31 +0000 (11:54 -0800)]
classifier: Make insert and replace take a const rule, too.

classifier_remove() was recently changed to take a const struct
cls_rule *.  Make the corresponding change to classifier_replace() and
classifier_insert().  This simplifies existing calling sites in
ofproto.

Signed-off-by: Jarno Rajahalme <jrajahalme@nicira.com>
Acked-by: Ben Pfaff <blp@nicira.com>
9 years agoclassifier: Remove internal mutex.
Jarno Rajahalme [Fri, 14 Nov 2014 23:58:09 +0000 (15:58 -0800)]
classifier: Remove internal mutex.

Almost all classifier users already exclude concurrent modifications,
or are single-threaded, hence the classifier internal mutex can be
removed.  Due to this change, ovs-router.c and tnl-ports.c need new
mutexes, which are added.

As noted by Ben in review, ovs_router_flush() should also free the
entries it removes from the classifier.  It now calls
ovsrcu_postpone() to that effect.

Suggested-by: Ben Pfaff <blp@nicira.com>
Signed-off-by: Jarno Rajahalme <jrajahalme@nicira.com>
Acked-by: Ben Pfaff <blp@nicira.com>
9 years agoclassifier: Lockless and robust classifier iteration.
Jarno Rajahalme [Thu, 13 Nov 2014 19:54:31 +0000 (11:54 -0800)]
classifier: Lockless and robust classifier iteration.

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, and allows the iterators to remain ignorant of the
internals of the cls_match.  This new list allows iteration of rules
in the classifier by traversing the RCU-friendly subtables vector, and
the rculist of rules in each subtable.

Classifier modifications may be performed concurrently, but whether or
not the concurrent iterator sees those changes depends on the timing
of change.  More specifically, an concurrent iterator:

- May or may not see a rule that is being inserted or removed.
- Will see either the new or the old version of a rule that is replaced.
- Will see all the other rules (that are not being modified).

Finally, The subtable's rculist also allows to make
classifier_rule_overlaps() lockless, which this patch also does.

Signed-off-by: Jarno Rajahalme <jrajahalme@nicira.com>
Acked-by: Ben Pfaff <blp@nicira.com>
9 years agoclassifier: Do not insert duplicate rules in indices.
Jarno Rajahalme [Fri, 14 Nov 2014 22:47:03 +0000 (14:47 -0800)]
classifier: Do not insert duplicate rules in indices.

There is no point in adding duplicate information into prefix tries.

Also, since the lower-priority duplicate rules are not visible to
lookups, they do not need to be in staged lookup indices directly
either (the head rule is).

Finally, now that cmap operations return the number of elements in the
cmap, subtable's 'n_rules' member is not needed any more.

Signed-off-by: Jarno Rajahalme <jrajahalme@nicira.com>
Acked-by: Ben Pfaff <blp@nicira.com>
9 years agodatapath-windows: Remove all duplicate checks for NULL.
Sorin Vinturis [Fri, 14 Nov 2014 11:35:21 +0000 (11:35 +0000)]
datapath-windows: Remove all duplicate checks for NULL.

Right now the gOvsSwitchContext pointer is checked against NULL
in a lot of places of the OVS extension code. This check should
be done only once to avoid wasteful checks. Thus I have added the
check in the dispatch routine, before doing any processing, and
removed all other checks from the rest of the code.

Signed-off-by: Sorin Vinturis <svinturis@cloudbasesolutions.com>
Acked-by: Eitan Eliahu <eliahue@vmware.com>
Signed-off-by: Ben Pfaff <blp@nicira.com>
9 years agorstp-state-machines: fix rcv_info().
Daniele Venturino [Fri, 14 Nov 2014 22:07:51 +0000 (14:07 -0800)]
rstp-state-machines: fix rcv_info().

A Config BPDU always conveys a Designated Port Role.

Signed-off-by: Daniele Venturino <daniele.venturino@m3s.it>
Acked-by: Jarno Rajahalme <jrajahalme@nicira.com>
9 years agorstp: Refactor rstp_port_set_administrative_bridge_port__().
Jarno Rajahalme [Fri, 14 Nov 2014 22:07:51 +0000 (14:07 -0800)]
rstp: Refactor rstp_port_set_administrative_bridge_port__().

Invoke move_rstp__() in rstp_port_set_administrative_bridge_port__()
if port is not initializing.  This is necessary in a test that checks
that a Port becoming alternate/backup/disabled stops learning and
forwarding.  The move_rstp__() call is necessary to immediatly disable
learning and forwarding on that port.  Without this, the test fails.

Signed-off-by: Daniele Venturino <daniele.venturino@m3s.it>
Signed-off-by: Jarno Rajahalme <jrajahalme@nicira.com>
Acked-by: Jarno Rajahalme <jrajahalme@nicira.com>
9 years agorstp: Refactor rstp_check_and_reset_fdb_flush().
Jarno Rajahalme [Fri, 14 Nov 2014 22:07:50 +0000 (14:07 -0800)]
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 <daniele.venturino@m3s.it>
Acked-by: Jarno Rajahalme <jrajahalme@nicira.com>
9 years agostp,rstp: disable learning and forwarding in STP/RSTP disabled state.
Daniele Venturino [Fri, 14 Nov 2014 22:07:50 +0000 (14:07 -0800)]
stp,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)
treats a Port as no longer active when it becomes an Alternate, Backup,
or Disabled Port and stops learning from received frames.

Signed-off-by: Daniele Venturino <daniele.venturino@m3s.it>
Acked-by: Jarno Rajahalme <jrajahalme@nicira.com>
9 years agorstp: Invoke updt_roles_tree__() in max_age and forward_delay setters.
Jarno Rajahalme [Fri, 14 Nov 2014 22:07:50 +0000 (14:07 -0800)]
rstp: Invoke updt_roles_tree__() in max_age and forward_delay setters.

This patch was necessary in a test on the acceptable range values of
max_age and forward_delay.  Since rstp_set_bridge_max_age__() and
rstp_set_bridge_forward_delay__() set the rstp->bridge_max_age and
rstp->bridge_forward_delay variables, it was necessary to call
updt_roles_tree__() to immediately update p->designated_times, used in
tx_rstp() (in lib/rstp-state-machines.c).  After this change the
validation software received the expected maxAge and forwardDelay
values.  Otherwise, such test failed.

Signed-off-by: Daniele Venturino <daniele.venturino@m3s.it>
Acked-by: Jarno Rajahalme <jrajahalme@nicira.com>
9 years agorstp-state-machines: fix TCN reception.
Daniele Venturino [Fri, 14 Nov 2014 22:07:50 +0000 (14:07 -0800)]
rstp-state-machines: fix TCN reception.

Set rcdv_tcn and return OTHER_INFO when a
TOPOLOGY_CHANGE_NOTIFICATION_BPDU is received, as required by
802.1Q-2008.

Signed-off-by: Daniele Venturino <daniele.venturino@m3s.it>
Acked-by: Jarno Rajahalme <jrajahalme@nicira.com>
9 years agorstp-state-machines: Rename ALTERNATE_AGREED state as ALTERNATE_AGREED_EXEC.
Daniele Venturino [Fri, 14 Nov 2014 22:07:50 +0000 (14:07 -0800)]
rstp-state-machines: Rename ALTERNATE_AGREED state as ALTERNATE_AGREED_EXEC.

All other similar port_role_transition_sm states are named according
to this pattern.

Signed-off-by: Daniele Venturino <daniele.venturino@m3s.it>
Acked-by: Jarno Rajahalme <jrajahalme@nicira.com>
9 years agorstp-state-machines: Fix updt_roles_tree().
Daniele Venturino [Fri, 14 Nov 2014 22:07:50 +0000 (14:07 -0800)]
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 <daniele.venturino@m3s.it>
Acked-by: Jarno Rajahalme <jrajahalme@nicira.com>
9 years agorstp-state-machines: Fix Topology Change State Machine transition.
Daniele Venturino [Fri, 14 Nov 2014 22:07:50 +0000 (14:07 -0800)]
rstp-state-machines: Fix Topology Change State Machine transition.

Accordingly to the 802.1D-2004 standard, the transition from the
TOPOLOGY_CHANGE_SM_NOTIFIED_TCN_EXEC state should go to the
TOPOLOGY_CHANGE_SM_NOTIFIED_TC_EXEC state and not to the
TOPOLOGY_CHANGE_SM_ACTIVE.

See section 17.31 of 802.1D-2004 standard.

Signed-off-by: Daniele Venturino <daniele.venturino@m3s.it>
Acked-by: Jarno Rajahalme <jrajahalme@nicira.com>
9 years agorstp-state-machines: Fix compare_rstp_priority_vector().
Daniele Venturino [Fri, 14 Nov 2014 22:07:50 +0000 (14:07 -0800)]
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 <daniele.venturino@m3s.it>
Acked-by: Jarno Rajahalme <jrajahalme@nicira.com>
9 years agorstp-state-machines: Fix reception of BPDUs conveying an unknown port role.
Jarno Rajahalme [Fri, 14 Nov 2014 22:07:50 +0000 (14:07 -0800)]
rstp-state-machines: Fix reception of BPDUs conveying an unknown port role.

If the Unknown value of the Port Role parameter is received, the state
machines should treat the RST BPDU as if it were a Configuration BPDU.

Signed-off-by: Daniele Venturino <daniele.venturino@m3s.it>
Acked-by: Jarno Rajahalme <jrajahalme@nicira.com>
9 years agorstp: Add 'rstp-admin-p2p-mac' and 'rstp-admin-port-state' setters.
Jarno Rajahalme [Fri, 14 Nov 2014 22:07:50 +0000 (14:07 -0800)]
rstp: Add 'rstp-admin-p2p-mac' and 'rstp-admin-port-state' setters.

'rstp-admin-port-state' is the Administrative Bridge Port state
variable defined in the 802.1D-2004 standard.  It can be set to
include or exclude a port from the active topology by management
(section 7.4).

operPointToPointMAC and 'rstp-admin-p2p-mac' are a pair of parameters
that permit inspection of, and control over, the administrative and
operational state of the point-to-point status of the MAC entity by
the MAC Relay Entity.  adminPointToPointMAC can be set by management
and its value is reflected on operPointToPointMAC.

Signed-off-by: Daniele Venturino <daniele.venturino@m3s.it>
Acked-by: Jarno Rajahalme <jrajahalme@nicira.com>
9 years agorstp: Show some useful rstp port fields.
Jarno Rajahalme [Fri, 14 Nov 2014 22:07:50 +0000 (14:07 -0800)]
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 <daniele.venturino@m3s.it>
Acked-by: Jarno Rajahalme <jrajahalme@nicira.com>
9 years agotests: Skip tunnel-push-pop test on Windows and BSD.
Pravin B Shelar [Thu, 13 Nov 2014 22:52:51 +0000 (14:52 -0800)]
tests: Skip tunnel-push-pop test on Windows and BSD.

Also updates ovs-router README documentation.

Signed-off-by: Pravin B Shelar <pshelar@nicira.com>
Acked-by: YAMAMOTO Takashi <yamamoto@valinux.co.jp>
9 years agorstp: Show some useful rstp port fields.
Jarno Rajahalme [Thu, 13 Nov 2014 22:55:25 +0000 (14:55 -0800)]
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 <daniele.venturino@m3s.it>
Acked-by: Jarno Rajahalme <jrajahalme@nicira.com>
9 years agoovs-ofctl: Support bucket commands
Simon Horman [Thu, 13 Nov 2014 02:53:29 +0000 (11:53 +0900)]
ovs-ofctl: Support bucket commands

(Draft) OpenFlow 1.5 flow mod commands include commands to manipulate the
buckets of existing groups. This patch add support to ovs-ofctl for these
commands. It also adds documentation and tests for them.

ONF-JIRA: EXT-350
Signed-off-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: Ben Pfaff <blp@nicira.com>
9 years agoofproto: Handle (draft) OpenFlow 1.5 insert and remove group commands.
Simon Horman [Thu, 13 Nov 2014 02:53:28 +0000 (11:53 +0900)]
ofproto: Handle (draft) OpenFlow 1.5 insert and remove group commands.

Handle (draft) OpenFlow 1.5 insert and remove group commands
of group mod messages.

ONF-JIRA: EXT-350
Signed-off-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: Ben Pfaff <blp@nicira.com>
9 years agoopenflow: Add OpenFlow 1.4 packet-in reasons.
Shu Shen [Thu, 13 Nov 2014 00:29:59 +0000 (16:29 -0800)]
openflow: Add OpenFlow 1.4 packet-in reasons.

Signed-off-by: Shu Shen <shu.shen@radisys.com>
Signed-off-by: Ben Pfaff <blp@nicira.com>
9 years agorhel: Include pkgconfig and update header install in Fedora/RHEL spec.
Thomas Graf [Thu, 13 Nov 2014 11:28:45 +0000 (12:28 +0100)]
rhel: Include pkgconfig and update header install in Fedora/RHEL spec.

As headers are installed by 'make install' now, the explicit install
instructions can be be removed from the spec file.

The spec file has been including <config.h> plus all headers in lib/ so
far which is dangerous:

 * config.h is a description of the capabilities of the build machine
   where the package was built. Although it may hint on what is
   included in the library, the defines do not necessarily hold true on
   another build machine.

 * Some headers in lib/ do not contain proper prefixes and pollute the
   global header name space. Do not include them anymore. Properly
   cleaned up headers will be exposed in openvswitch/ and openflow/

For the RHEL spec, both pkgconfig and headers are omitted as no -devel
package exists yet.

Cc: Flavio Leitner <fbl@redhat.com>
Suggested-by: Flavio Leitner <fbl@redhat.com>
Signed-off-by: Thomas Graf <tgraf@noironetworks.com>
Acked-by: Flavio Leitner <fbl@redhat.com>
Signed-off-by: Ben Pfaff <blp@nicira.com>
9 years agolib: Add support for pkgconfig for libsflow.
Thomas Graf [Thu, 13 Nov 2014 11:28:44 +0000 (12:28 +0100)]
lib: Add support for pkgconfig for libsflow.

Signed-off-by: Thomas Graf <tgraf@noironetworks.com>
Acked-by: Flavio Leitner <fbl@redhat.com>
Signed-off-by: Ben Pfaff <blp@nicira.com>
9 years agolib: Add support for pkgconfig for libovsdb.
Thomas Graf [Thu, 13 Nov 2014 11:28:43 +0000 (12:28 +0100)]
lib: Add support for pkgconfig for libovsdb.

Signed-off-by: Thomas Graf <tgraf@noironetworks.com>
Acked-by: Flavio Leitner <fbl@redhat.com>
Signed-off-by: Ben Pfaff <blp@nicira.com>
9 years agolib: Add support for pkgconfig for libofproto.
Thomas Graf [Thu, 13 Nov 2014 11:28:42 +0000 (12:28 +0100)]
lib: Add support for pkgconfig for libofproto.

Signed-off-by: Thomas Graf <tgraf@noironetworks.com>
Acked-by: Flavio Leitner <fbl@redhat.com>
Signed-off-by: Ben Pfaff <blp@nicira.com>
9 years agolib: Add support for pkgconfig for libopenvswitch.
Rob Adams [Thu, 13 Nov 2014 11:28:41 +0000 (12:28 +0100)]
lib: Add support for pkgconfig for libopenvswitch.

Add rule to generate pkgconfig .pc file from configure.
Install pkg-config file to $(libdir)/pkgconfig

Signed-off-by: Rob Adams <readams@readams.net>
Acked-by: Flavio Leitner <fbl@redhat.com>
Signed-off-by: Ben Pfaff <blp@nicira.com>
9 years agoinclude: Install openflow/ and openvswitch/ headers.
Thomas Graf [Thu, 13 Nov 2014 11:28:40 +0000 (12:28 +0100)]
include: Install openflow/ and openvswitch/ headers.

Suggested-by: Ben Pfaff <blp@nicira.com>
Signed-off-by: Thomas Graf <tgraf@noironetworks.com>
Acked-by: Flavio Leitner <fbl@redhat.com>
Signed-off-by: Ben Pfaff <blp@nicira.com>
9 years agodpif-netdev: Allow direct destroy of 'struct dp_netdev_port'.
Alex Wang [Fri, 7 Nov 2014 21:27:05 +0000 (13:27 -0800)]
dpif-netdev: Allow direct destroy of 'struct dp_netdev_port'.

Before this commit, when 'struct dp_netdev_port' is deleted from
'dpif-netdev' datapath, if there is pmd thread, the pmd thread
will release the last reference to the port and ovs-rcu postpone
the destroy.  However, the delayed close of object like 'struct
netdev' could cause failure in immediate re-add or reconfigure of
the same device.

To fix the above issue, this commit uses condition variable and
makes the main thread wait for pmd thread to release the reference
when deleting port.  Then, the main thread can directly destroy the
port.

Reported-by: Cian Ferriter <cian.ferriter@intel.com>
Signed-off-by: Alex Wang <alexw@nicira.com>
Acked-by: Pravin B Shelar <pshelar@nicira.com>
9 years agoofproto-dpif-xlate: Allow direct destroy of previous config.
Alex Wang [Fri, 7 Nov 2014 21:02:05 +0000 (13:02 -0800)]
ofproto-dpif-xlate: Allow direct destroy of previous config.

Before this commit, the ofproto-dpif-xlate module uses ovs-rcu to
postpone the destroy of previous configuration.  However, the delayed
close of object like 'struct netdev' could cause failure in immediate
re-add or reconfigure of the same device.

To fix the above issue, this commit makes the ofproto-dpif-xlate
module call ovsrcu_synchronize(), which waits for all threads
to finish the use of reference to previous config.  Then, the
module can just directly destroy the previous config.

Reported-by: Cian Ferriter <cian.ferriter@intel.com>
Signed-off-by: Alex Wang <alexw@nicira.com>
Acked-by: Ben Pfaff <blp@nicira.com>
9 years agodpif-netdev: Move 'struct dp_netdev_port' initialization before use.
Alex Wang [Sat, 8 Nov 2014 01:11:07 +0000 (17:11 -0800)]
dpif-netdev: Move 'struct dp_netdev_port' initialization before use.

There is a portion of the 'struct dp_netdev_port' initialization
that is placed after the reload of pmd threads.  This means in
theory, there could be a race where pmd threads access half-
initialized struct.  Although such race has not been seen, it
makes sense to fully initialize the struct before use.

Found by code inspection.

Signed-off-by: Alex Wang <alexw@nicira.com>
Acked-by: Pravin B Shelar <pshelar@nicira.com>
9 years agoclassifier: Integrate insert_rule() into classifier_replace().
Jarno Rajahalme [Wed, 12 Nov 2014 23:22:35 +0000 (15:22 -0800)]
classifier: Integrate insert_rule() into classifier_replace().

insert_rule() only had one caller and this makes the code easier to
understand.

Signed-off-by: Ben Pfaff <blp@nicira.com>
Acked-by: Jarno Rajahalme <jrajahalme@nicira.com>
9 years agotests: Fix ovs-ofctl rule with importance test case.
Jarno Rajahalme [Wed, 12 Nov 2014 23:22:35 +0000 (15:22 -0800)]
tests: Fix ovs-ofctl rule with importance test case.

Sort the flow dump results to always have them in consistent order.

Signed-off-by: Jarno Rajahalme <jrajahalme@nicira.com>
Acked-by: Ben Pfaff <blp@nicira.com>
9 years agoopenvswitch: Userspace tunneling.
Pravin B Shelar [Tue, 11 Nov 2014 19:53:47 +0000 (11:53 -0800)]
openvswitch: Userspace tunneling.

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 ARP cache which build automatically
by snooping arp. And third is tunnel protocol table which list all
listening protocols which is populated by vswitchd as tunnel ports
are added. GRE and VXLAN protocol support is added in this patch.

Tunneling works as follows:
On packet receive vswitchd check if this packet is targeted to tunnel
port. If it is then vswitchd inserts tunnel pop action which pops
header and sends packet to tunnel port.
On packet xmit rather than generating Set tunnel action it generate
tunnel push action which has tunnel header data. datapath can use
tunnel-push action data to generate header for each packet and
forward this packet to output port. Since tunnel-push action
contains most of packet header vswitchd needs to lookup routing
table and arp table to build this action.

Signed-off-by: Pravin B Shelar <pshelar@nicira.com>
Acked-by: Jarno Rajahalme <jrajahalme@nicira.com>
Acked-by: Thomas Graf <tgraf@noironetworks.com>
Acked-by: Ben Pfaff <blp@nicira.com>
9 years agoFix misspellings of "OpenFlow".
Ben Pfaff [Wed, 12 Nov 2014 00:42:24 +0000 (16:42 -0800)]
Fix misspellings of "OpenFlow".

Signed-off-by: Ben Pfaff <blp@nicira.com>
Acked-by: Flavio Leitner <fbl@redhat.com>
9 years agoutil: Introduce ovs_scan_len()
Pravin B Shelar [Mon, 10 Nov 2014 20:46:11 +0000 (12:46 -0800)]
util: Introduce ovs_scan_len()

This is similar to ovs_scan but takes int pointer as extra
parameter, this pointer point to starting index of the string.
On successful scan this API stores number of characters
scanned.  This API is useful for parsing complex odp actions
e.g. tun_push action.

Signed-off-by: Pravin B Shelar <pshelar@nicira.com>
Acked-by: Thomas Graf <tgraf@noironetworks.com>
Acked-by: Ben Pfaff <blp@nicira.com>
9 years agotunnel: Convert rwlock to fat-rwlock.
Pravin B Shelar [Mon, 10 Nov 2014 20:45:57 +0000 (12:45 -0800)]
tunnel: Convert rwlock to fat-rwlock.

Tunnel lookup is done in slow path but DPDK tunneling
slow path is called in PMD context. Therefore it is
better to convert rwlock to fat rwlock.

Signed-off-by: Pravin B Shelar <pshelar@nicira.com>
Acked-by: Thomas Graf <tgraf@noironetworks.com>
Acked-by: Ben Pfaff <blp@nicira.com>
9 years agolib: Avoid using C++ keywords as variable names in header files.
Amit Bose [Wed, 12 Nov 2014 14:06:08 +0000 (15:06 +0100)]
lib: Avoid using C++ keywords as variable names in header files.

Otherwise C++ programs can't #include the headers.

Signed-off-by: Amit Bose <bose@noironetworks.com>
Signed-off-by: Ben Pfaff <blp@nicira.com>
9 years agoinclude: Use #include <.*> in public headers.
Thomas Graf [Wed, 12 Nov 2014 14:06:02 +0000 (15:06 +0100)]
include: Use #include <.*> in public headers.

Required to make the headers installable.

Signed-off-by: Thomas Graf <tgraf@noironetworks.com>
Acked-by: Flavio Leitner <fbl@redhat.com>
Signed-off-by: Ben Pfaff <blp@nicira.com>
9 years agolib: Document --enable-shared configure switch
Thomas Graf [Wed, 12 Nov 2014 14:06:00 +0000 (15:06 +0100)]
lib: Document --enable-shared configure switch

Signed-off-by: Thomas Graf <tgraf@noironetworks.com>
Acked-by: Flavio Leitner <fbl@redhat.com>
Signed-off-by: Ben Pfaff <blp@nicira.com>
9 years agolib: Correctly pass collected AM_LDFLAGS to linker
Thomas Graf [Wed, 12 Nov 2014 14:05:59 +0000 (15:05 +0100)]
lib: Correctly pass collected AM_LDFLAGS to linker

AM_LDFLAGS is not automatically passed to individual library LDFLAGS
and must be passed in explicitly.

Signed-off-by: Thomas Graf <tgraf@noironetworks.com>
Acked-by: Flavio Leitner <fbl@redhat.com>
Signed-off-by: Ben Pfaff <blp@nicira.com>
9 years agoofp-parse: Parse bucket commands of group mod.
Simon Horman [Wed, 12 Nov 2014 06:24:12 +0000 (15:24 +0900)]
ofp-parse: Parse bucket commands of group mod.

(Draft) OpenFlow 1.5 adds bucket commands to group mod.
This patch allows parsing of them as a step towards supporting them.

ONF-JIRA: EXT-350
Signed-off-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: Ben Pfaff <blp@nicira.com>
9 years agoofp-util: Provide bucket list helper functions
Simon Horman [Wed, 12 Nov 2014 06:24:10 +0000 (15:24 +0900)]
ofp-util: Provide bucket list helper functions

This is in preparation for supporting the bucket commands of
(draft) Open Flow 1.5 group mod messages.

Also document ofputil_bucket_check_duplicate_id() for good measure.

ONF-JIRA: EXT-350
Signed-off-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: Ben Pfaff <blp@nicira.com>
9 years agoofp-util: Enhance error logging during encoding group mod messages.
Simon Horman [Wed, 12 Nov 2014 06:24:09 +0000 (15:24 +0900)]
ofp-util: Enhance error logging during encoding group mod messages.

In preparation for supporting (draft) OpenFlow 1.5
group mod commands enhance the error logging of them.

ONF-JIRA: EXT-350
Signed-off-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: Ben Pfaff <blp@nicira.com>
9 years agoofp-util: Do not allow buckets for OFPGC15_REMOVE_BUCKET.
Simon Horman [Wed, 12 Nov 2014 06:24:08 +0000 (15:24 +0900)]
ofp-util: Do not allow buckets for OFPGC15_REMOVE_BUCKET.

ONF-JIRA: EXT-350
Signed-off-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: Ben Pfaff <blp@nicira.com>
9 years agotests: Skip a sflow test for Windows.
Gurucharan Shetty [Wed, 12 Nov 2014 15:23:00 +0000 (07:23 -0800)]
tests: Skip a sflow test for Windows.

AT_XFAIL_IF cannot be used here because it lets the tests
run first and then say that the failure is expected. In this
particular test, the utilities simply hang blocking all
other tests. So use AT_SKIP_IF instead.

Signed-off-by: Gurucharan Shetty <gshetty@nicira.com>
Acked-by: Ben Pfaff <blp@nicira.com>
Acked-by: Nithin Raju <nithin@vmware.com>
9 years agoINSTALL.Windows: Add a TODO for QOS.
Gurucharan Shetty [Wed, 12 Nov 2014 14:56:10 +0000 (06:56 -0800)]
INSTALL.Windows: Add a TODO for QOS.

Also, remove the entry that talks about the need for atomic support
on Windows as that has already been added.

Signed-off-by: Gurucharan Shetty <gshetty@nicira.com>
Acked-by: Ben Pfaff <blp@nicira.com>
Acked-by: Nithin Raju <nithin@vmware.com>
9 years agoofproto: Test that group commands fail for OpenFlow 1.0
Simon Horman [Wed, 12 Nov 2014 06:45:17 +0000 (15:45 +0900)]
ofproto: Test that group commands fail for OpenFlow 1.0

Groups are not supported by OpenFlow 1.0

Signed-off-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: Ben Pfaff <blp@nicira.com>
9 years agodatapath-windows: Update port property.
Eitan Eliahu [Wed, 12 Nov 2014 09:31:27 +0000 (01:31 -0800)]
datapath-windows: Update port property.

Update Hyper-V port properties on NDIS property port set callback.
Driver update the port friendly name in its internal table.
Since the NIC Create callback won't be called after we need to maintain the
NIC and THE OVS states of the Vport.

Setting a new friendly name was tested in the following states:
[1] The port is disconnected
[2] The port is connected but not OVS user mode counter port
[3] The port is connected and an OCS port associated with it

Signed-off-by: Eitan Eliahu <eliahue@vmware.com>
Acked-by: Nithin Raju <nithin@vmware.com>
Signed-off-by: Ben Pfaff <blp@nicira.com>
9 years agodatapath-windows: Add functionality to return OVSPortName given a VM
Alin Serdean [Tue, 11 Nov 2014 18:45:32 +0000 (18:45 +0000)]
datapath-windows: Add functionality to return OVSPortName given a VM

This commit adds functionality to return the ElementName of all the VNIC of a
given virtual machine name.

The function returns all the allocated ports including the ones which the OVS
port name was not set.

The default OVS port name that is set is the following:
"ElementName             : Dynamic Ethernet Switch Port".

Example of usage:
Import-Module OVS.psm1
Get-VMNetworkAdapterWithOvsPort test_ubuntu_1
$a = Get-VMNetworkAdapterWithOvsPort test_2_1
$a | findstr ElementName

Requested-by: Nithin Raju <nithin@vmware.com>
Reported-at: https://github.com/openvswitch/ovs-issues/issues/51
Signed-off-by: Alin Gabriel Serdean <aserdean@cloudbasesolutions.com>
Tested-by: Nithin Raju <nithin@vmware.com>
Acked-by: Nithin Raju <nithin@vmware.com>
Signed-off-by: Ben Pfaff <blp@nicira.com>
9 years agotravis: Add building of shared library to CI matrix.
Thomas Graf [Wed, 12 Nov 2014 09:22:02 +0000 (10:22 +0100)]
travis: Add building of shared library to CI matrix.

Signed-off-by: Thomas Graf <tgraf@noironetworks.com>
Signed-off-by: Ben Pfaff <blp@nicira.com>
9 years agoBuild: Add support for shared libraries and versioning.
Scott Mann [Sat, 8 Nov 2014 03:02:09 +0000 (19:02 -0800)]
Build: Add support for shared libraries and versioning.

These changes allow for the building of shared libraries by providing
the --enable-shared option to configure. In particular, lib/libopenvwitch.so,
lib/libsflow.so, ofproto/libofproto.so, and ovsdb/libovsdb.so will be built.
Original behavior of building static remains the same.

Additionally, versioning is introduced to each of the libraries objects
paving the way for APIs to be built around them. A detailed comment
outlining the rules for changing a version number is provided in
configure.ac. Note that at this time, the version number is set to
1.0.0, no API is specified yet, and there are no requirements to
maintain any sort of compatibility in any of the libraries.

Signed-off-by: Scott Mann <smann@noironetworks.com>
Signed-off-by: Ben Pfaff <blp@nicira.com>
9 years agoutil: Add be32_prefix_mask().
Jarno Rajahalme [Tue, 11 Nov 2014 23:50:51 +0000 (15:50 -0800)]
util: Add be32_prefix_mask().

Shifting a 32-bit entity by 32 bits is undefined behavior.  As we have 3
cases where we may hit this, it is a time to introduce a helper for
this.

VMware-BZ: #1355026
Signed-off-by: Jarno Rajahalme <jrajahalme@nicira.com>
Acked-by: Pravin B Shelar <pshelar@nicira.com>
9 years agoovs-atomic: Prefer Clang intrinsics over <stdatomic.h>.
Ben Pfaff [Tue, 11 Nov 2014 16:19:27 +0000 (08:19 -0800)]
ovs-atomic: Prefer Clang intrinsics over <stdatomic.h>.

On my Debian "jessie" system, <stdatomic.h> provided by GCC 4.9 is busted
when Clang 3.5 tries to use it.  Even a trivial program like this:

    #include <stdatomic.h>

    void
    foo(void)
    {
         _Atomic(int) x;
         atomic_fetch_add(&x, 1);
}

yields:

     atomic.c:7:5: error: address argument to atomic operation must be a
        pointer to integer or pointer ('_Atomic(int) *' invalid)

The Clang-specific version of ovs-atomic.h stills works, though, so this
commit works around the problem.

Signed-off-by: Ben Pfaff <blp@nicira.com>
Acked-by: Jarno Rajahalme <jrajahalme@nicira.com>
9 years agosflow: Export LAG, PORTNAME, and OPENFLOWPORT information also.
Neil McKee [Fri, 27 Jun 2014 18:19:59 +0000 (11:19 -0700)]
sflow: Export LAG, PORTNAME, and OPENFLOWPORT information also.

Export standard sFlow LAG, PORTNAME and OPENFLOWPORT structures with each
counter-sample. Add unit-test for sFlow-LAG. Adjust other unit-tests to
accommodate these new annotations.

The sFlow-LAG structures are important for topology discovery, for
troubleshooting LAG instability,  and for correctly combining
sFlow feeds from multiple sources.

The OPENFLOWPORT and PORTNAME structures are important for systems that
aim to combine sFlow monitoring with OpenFlow controls,  as they
provide straightforward mapping (1) between sFlow agent IP and OpenFlow
datapath-id,  and (2) between interface name,ifIndex and OpenFlow
port number.

Signed-off-by: Neil McKee <neil.mckee@inmon.com>
Signed-off-by: Ben Pfaff <blp@nicira.com>
9 years agoofproto: Add OpenFlow 1.5 group tests.
Simon Horman [Tue, 11 Nov 2014 03:39:24 +0000 (12:39 +0900)]
ofproto: Add OpenFlow 1.5 group tests.

This motivation for this change is to exercise
(draft) OpenFlow 1.5 group messages.

ONF-JIRA: EXT-350
Signed-off-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: Ben Pfaff <blp@nicira.com>
9 years agoofproto: Add OpenFlow 1.3 group stats test.
Simon Horman [Tue, 11 Nov 2014 03:39:23 +0000 (12:39 +0900)]
ofproto: Add OpenFlow 1.3 group stats test.

This is to exercise OpenFlow 1.3 group stats replies
messages which include a duration.

ONF-JIRA: EXT-350
Signed-off-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: Ben Pfaff <blp@nicira.com>
9 years agoofp-parse: Parse (draft) OpenFlow 1.5 bucket ids.
Simon Horman [Tue, 11 Nov 2014 03:39:22 +0000 (12:39 +0900)]
ofp-parse: Parse (draft) OpenFlow 1.5 bucket ids.

This is part of support for (draft) OpenFlow 1.5 flow mod messages.

This adds support for specifying the bucket_id of buckets of groups in
ovs-ofctl and documents that accordingly.

ONF-JIRA: EXT-350
Signed-off-by: Simon Horman <simon.horman@netronome.com>
[blp@nicira.com tweaked the documentation]
Signed-off-by: Ben Pfaff <blp@nicira.com>
9 years agoofp-print: Support printing of (draft) OpenFlow 1.5 group messages with bucket actions
Simon Horman [Tue, 11 Nov 2014 03:39:21 +0000 (12:39 +0900)]
ofp-print: Support printing of (draft) OpenFlow 1.5 group messages with bucket actions

ONF-JIRA: EXT-350
Signed-off-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: Ben Pfaff <blp@nicira.com>
9 years agoofp-print: Print bucket ids of OpenFlow 1.5 group messages.
Simon Horman [Tue, 11 Nov 2014 03:39:20 +0000 (12:39 +0900)]
ofp-print: Print bucket ids of OpenFlow 1.5 group messages.

ONF-JIRA: EXT-350
Signed-off-by: Simon Horman <simon.horman@netronome.com>
[blp@nicira.com changed ofp_print_bucket_id()]
Signed-off-by: Ben Pfaff <blp@nicira.com>
9 years agoofp-util: Encoding and decoding of (draft) OpenFlow 1.5 group messages.
Simon Horman [Tue, 11 Nov 2014 03:39:19 +0000 (12:39 +0900)]
ofp-util: Encoding and decoding of (draft) OpenFlow 1.5 group messages.

This provides the bulk of the ofproto side of support for
OpenFlow 1.5 group messages. It provides for encoding and decoding
of updated group mod and group desc reply messages. This includes
a new bucket format and their properties.

Open Flow 1.5 Groups also have properties but as no non-experimenter
properties are defined this patch does not provide parsing or encoding
of group properties.

ONF-JIRA: EXT-350
Signed-off-by: Simon Horman <simon.horman@netronome.com>
[blp@nicira.com fixed minor bugs and style issues]
Signed-off-by: Ben Pfaff <blp@nicira.com>
9 years agoopenflow: Add types for (draft) OpenFlow 1.5 group mod.
Simon Horman [Tue, 11 Nov 2014 03:39:18 +0000 (12:39 +0900)]
openflow: Add types for (draft) OpenFlow 1.5 group mod.

ONF-JIRA: EXT-350
Signed-off-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: Ben Pfaff <blp@nicira.com>
9 years agotravis: Enable -Wsparse-error and fail on new sparse warnings
Thomas Graf [Tue, 11 Nov 2014 13:51:46 +0000 (14:51 +0100)]
travis: Enable -Wsparse-error and fail on new sparse warnings

A new option -Wsparse-error has been introduced to sparse in the
commits:

4d8811 sparse: Make -Werror turn warnigns into errors
fe57af rename -Werror to -Wsparse-error

which makes sparse fail on warnings. Start using the sparse git tree
for verification and make use of the new flag.

Signed-off-by: Thomas Graf <tgraf@noironetworks.com>
Signed-off-by: Ben Pfaff <blp@nicira.com>
9 years agotests: Avoid nonportable use of bash-specific {a..b..c} construct.
Ben Pfaff [Tue, 11 Nov 2014 04:27:26 +0000 (20:27 -0800)]
tests: Avoid nonportable use of bash-specific {a..b..c} construct.

Reported-by: Alex Wang <alexw@nicira.com>
Reported-by: YAMAMOTO Takashi <yamamoto@valinux.co.jp>
Signed-off-by: Ben Pfaff <blp@nicira.com>
Acked-by: YAMAMOTO Takashi <yamamoto@valinux.co.jp>
9 years agoofp-errors: Add (draft) OpenFlow 1.5 group mod errors.
Simon Horman [Tue, 11 Nov 2014 03:39:17 +0000 (12:39 +0900)]
ofp-errors: Add (draft) OpenFlow 1.5 group mod errors.

Add OFPERR_OFPGMFC_UNKNOWN_BUCKET and OFPERR_OFPGMFC_BUCKET_EXISTS
which are part of (draft) OpenFlow 1.5 groups.

ONF-JIRA: EXT-350
Signed-off-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: Ben Pfaff <blp@nicira.com>
9 years agoovs-router: Fix build on NetBSD
YAMAMOTO Takashi [Mon, 10 Nov 2014 07:31:47 +0000 (16:31 +0900)]
ovs-router: Fix build on NetBSD

Split the linux-specific part of ovs-router.h
into ovs-router-linux.h.

The breakage was introduced by
commit d9b4ebc5d15202bedad06969353435a4c1951c71
("route-table: Use classifier to store routing table.")

Signed-off-by: YAMAMOTO Takashi <yamamoto@valinux.co.jp>
Acked-by: Ben Pfaff <blp@nicira.com>
9 years agoAUTHORS: Add Shu Shen.
Ben Pfaff [Tue, 11 Nov 2014 01:39:25 +0000 (17:39 -0800)]
AUTHORS: Add Shu Shen.

Signed-off-by: Ben Pfaff <blp@nicira.com>
9 years agoofproto: Fix supported group types.
Shu Shen [Tue, 11 Nov 2014 00:35:08 +0000 (16:35 -0800)]
ofproto: Fix supported group types.

Previously the types field is uninited and leds to no supported type
being reported in OFPMP_GROUP_FEATURES message.

Signed-off-by: Shu Shen <shu.shen@radisys.com>
Signed-off-by: Ben Pfaff <blp@nicira.com>
9 years agoFix setting transport ports with frags.
Jarno Rajahalme [Wed, 5 Nov 2014 18:10:13 +0000 (10:10 -0800)]
Fix setting transport ports with frags.

Packets with 'LATER' fragment do not have a transport header, so it is
not possible to either match on or set transport ports on such
packets.  Matching is prevented by augmenting mf_are_prereqs_ok() with
a nw_frag 'LATER' bit check.  Setting the transport headers on such
packets is prevented in three ways:

1. Flows with an explicit match on nw_frag, where the LATER bit is 1:
   existing calls to the modified mf_are_prereqs_ok() prohibit using
   transport header fields (port numbers) in OXM/NXM actions
   (set_field, move).  SET_TP_* actions need a new check on the LATER
   bit.

2. Flows that wildcard the nw_frag LATER bit: At flow translation
   time, add calls to mf_are_prereqs_ok() to make sure that we do not
   use transport ports in flows that do not have them.

3. At action execution time, do not set transport ports, if the packet
   does not have a full transport header.  This ensures that we never
   call the packet_set functions, that require a valid transport
   header, with packets that do not have them.  For example, if the
   flow was created with a IPv6 first fragment that had the full TCP
   header, but the next packet's first fragment is missing them.

3 alone would suffice for correct behavior, but 1 and 2 seem like a
right thing to do, anyway.

Currently, if we are setting port numbers, we will also match them,
due to us tracking the set fields with the same flow_wildcards as the
matched fields.  Hence, if the incoming port number was not zero, the
flow would not match any packets with missing or truncated transport
headers.  However, relying on no packets having zero port numbers
would not be very robust.  Also, we may separate the tracking of set
and matched fields in the future, which would allow some flows that
blindly set port numbers to not match on them at all.

For TCP in case 3 we use ofpbuf_get_tcp_payload() that requires the
whole (potentially variable size) TCP header to be present.  However,
when parsing a flow, we only require the fixed size portion of the TCP
header to be present, which would be enough to set the port numbers
and fix the TCP checksum.

Finally, we add tests testing the new behavior.

Signed-off-by: Jarno Rajahalme <jrajahalme@nicira.com>
Acked-by: Ben Pfaff <blp@nicira.com>
9 years agominiflow_extract: Properly handle small IP packets.
Jarno Rajahalme [Wed, 5 Nov 2014 18:10:13 +0000 (10:10 -0800)]
miniflow_extract: Properly handle small IP packets.

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 prevent ofpbuf_pull from entering the padding,
as well as to allow ofpbuf_l4_size() to return the size of the IP
payload without the l2 padding.

This helps avoiding parsing truncated transport headers, for example.

Signed-off-by: Jarno Rajahalme <jrajahalme@nicira.com>
Acked-by: Ben Pfaff <blp@nicira.com>
9 years agoofproto-dpif: Clean up table lookup code.
Jarno Rajahalme [Mon, 10 Nov 2014 21:14:29 +0000 (13:14 -0800)]
ofproto-dpif: Clean up table lookup code.

Keep knowledge about ofproto-dpif internal rules within ofproto-dpif.
This removes duplication of code (accross multiple files), and allows
making the ofproto-dpif-xlate interface a bit smaller.

'may_packet_in' was previously not honored for the
'LOOKUP_VERDICT_DEFAULT' case.  Now it is, but this change did not
break any unit tests.

Also, previously in the 'LOOKUP_VERDICT_CONTROLLER' case packet-in
messages would have been generated when the in_port could not be found
('config' was set to zero in this case, causing 'OFPUTIL_PC_NO_PACKET_IN'
flag to be zero as well).  This patch changes this behavior and never
issues packet-in messages when the in_port cannot be found.  This did
not break any unit tests either.

Signed-off-by: Jarno Rajahalme <jrajahalme@nicira.com>
Acked-by: Ben Pfaff <blp@nicira.com>
9 years agoofproto-dpif-xlate: Remove xbridge->frag.
Jarno Rajahalme [Mon, 10 Nov 2014 21:14:29 +0000 (13:14 -0800)]
ofproto-dpif-xlate: Remove xbridge->frag.

It it no longer used, so better remove it.

Signed-off-by: Jarno Rajahalme <jrajahalme@nicira.com>
Acked-by: Ben Pfaff <blp@nicira.com>
9 years agorule_dpif_lookup_from_table: Check frags only once.
Jarno Rajahalme [Mon, 10 Nov 2014 21:14:29 +0000 (13:14 -0800)]
rule_dpif_lookup_from_table: Check frags only once.

Move the frags handling check up in the call chain, so that it is done
once for each rule_dpif_lookup_from_table() call.

Signed-off-by: Jarno Rajahalme <jrajahalme@nicira.com>
Acked-by: Ben Pfaff <blp@nicira.com>
9 years agoofp-util: Refactor decoding of OpenFlow 1.1 group desc reply messages.
Simon Horman [Mon, 10 Nov 2014 04:47:58 +0000 (13:47 +0900)]
ofp-util: Refactor decoding of OpenFlow 1.1 group desc reply messages.

This refactoring is in preparation for supporting encoding
of (draft) OpenFlow 1.5 group mod messages.

ONF-JIRA: EXT-350
Signed-off-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: Ben Pfaff <blp@nicira.com>
9 years agoofp-util: Refactor encoding of OpenFlow 1.1 group desc reply messages.
Simon Horman [Mon, 10 Nov 2014 04:47:57 +0000 (13:47 +0900)]
ofp-util: Refactor encoding of OpenFlow 1.1 group desc reply messages.

This refactoring is in preparation for supporting encoding
of (draft) OpenFlow 1.5 group mod messages.

ONF-JIRA: EXT-350
Signed-off-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: Ben Pfaff <blp@nicira.com>
9 years agoofp-util: Refactor decoding of OpenFlow 1.1 group mod messages.
Simon Horman [Mon, 10 Nov 2014 04:47:56 +0000 (13:47 +0900)]
ofp-util: Refactor decoding of OpenFlow 1.1 group mod messages.

This refactoring is in preparation for supporting encoding
of (draft) OpenFlow 1.5 group mod messages.

ONF-JIRA: EXT-350
Signed-off-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: Ben Pfaff <blp@nicira.com>
9 years agoofp-util: Rename OpenFlow 1.1 buckets decoder.
Simon Horman [Mon, 10 Nov 2014 04:47:55 +0000 (13:47 +0900)]
ofp-util: Rename OpenFlow 1.1 buckets decoder.

Rename ofputil_pull_buckets() as ofputil_pull_buckets11()
to denote that it decodes OpenFlow 1.1 buckets (also used in
OpenFlow 1.2 to 1.4).

This refactoring is in preparation for supporting decoding
of (draft) OpenFlow 1.5 group mod messages and their
buckets.

ONF-JIRA: EXT-350
Signed-off-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: Ben Pfaff <blp@nicira.com>
9 years agoofp-util: Break out encoding of OpenFlow 1.1 group mod messages.
Simon Horman [Mon, 10 Nov 2014 04:47:54 +0000 (13:47 +0900)]
ofp-util: Break out encoding of OpenFlow 1.1 group mod messages.

This refactoring is in preparation for supporting encoding
of (draft) OpenFlow 1.5 group mod messages.

ONF-JIRA: EXT-350
Signed-off-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: Ben Pfaff <blp@nicira.com>
9 years agoofp-util: Provide helper to encode OpenFlow 1.1 group buckets.
Simon Horman [Mon, 10 Nov 2014 04:47:53 +0000 (13:47 +0900)]
ofp-util: Provide helper to encode OpenFlow 1.1 group buckets.

The group bucket encoding code appears in two places
in almost identical form.  Consolidate this into
a helper function.

The helper name includes ofp11 as later patches
will add an ofp15 version to support (draft) OpenFlow 1.5 buckets.

ONF-JIRA: EXT-350
Signed-off-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: Ben Pfaff <blp@nicira.com>
9 years agoAdd support for OpenFlow 1.4+ "importance" values.
Rishi Bamba [Fri, 7 Nov 2014 12:48:48 +0000 (18:18 +0530)]
Add support for OpenFlow 1.4+ "importance" values.

This patch enables a user to set importance for a new rule via add-flow
OF1.4+ in the OVS and display the same via dump-flows command OF1.4+.
The changes are made in accordance with OpenFlow 1.4 specs to implement
eviction on the basis of "importance".  This patch also enhances the
diff-flows & replace-flows CLI for addition of importance parameter in
a rule.

This doesn't actually implement eviction on the basis of importance, which
will happen in a later patch.

Signed-off-by: Rishi Bamba <rishi.bamba@tcs.com>
Signed-off-by: Ben Pfaff <blp@nicira.com>
9 years agoid-pool: Fix memory leak.
Ben Pfaff [Mon, 10 Nov 2014 16:43:31 +0000 (08:43 -0800)]
id-pool: Fix memory leak.

This dates back to the previous location of the code in ofproto-dpif-rid.
I will backport.

Signed-off-by: Ben Pfaff <blp@nicira.com>
9 years agoid-pool: Refactor id_pool_alloc_id to allow any 32 bit value to be an id
Simon Horman [Mon, 10 Nov 2014 04:47:49 +0000 (13:47 +0900)]
id-pool: Refactor id_pool_alloc_id to allow any 32 bit value to be an id

id_pool_alloc_id() was created by breaking out the recirculation
allocation code. As it is now a library call it makes sense to remove
the restriction that id 0 is reserved.

Signed-off-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: Ben Pfaff <blp@nicira.com>
9 years agoid-pool: Re-factor recirculation id allocator into standalone id pool.
Simon Horman [Mon, 10 Nov 2014 04:47:48 +0000 (13:47 +0900)]
id-pool: Re-factor recirculation id allocator into standalone id pool.

Refactor the lock-free portion of the recirculation id allocator
into stand-alone id pool. This is in preparation for re-using
that portion to allocate bucket ids which are part of (draft)
OpenFlow 1.5 groups.

ONF-JIRA: EXT-350
Signed-off-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: Ben Pfaff <blp@nicira.com>
9 years agodatapath: fix coding style.
Pravin B Shelar [Sat, 8 Nov 2014 15:24:42 +0000 (07:24 -0800)]
datapath: fix coding style.

Kernel datapath code has diverged from upstream code.  This
makes porting patches between these two code bases harder
than it needs to be. Following patch fixes this by fixing
coding style issues on this branch.

Signed-off-by: Pravin B Shelar <pshelar@nicira.com>
9 years agodatapath: Fix few mpls issues.
Pravin B Shelar [Sat, 8 Nov 2014 16:14:21 +0000 (08:14 -0800)]
datapath: Fix few mpls issues.

Found during MPLS upstreaming.  Also sync-up MPLS header files
with upstream code.

Signed-off-by: Pravin B Shelar <pshelar@nicira.com>
9 years agodatapath: Rename last_action() as nla_is_last() and move to netlink.h
Simon Horman [Sat, 8 Nov 2014 21:53:52 +0000 (13:53 -0800)]
datapath: Rename last_action() as nla_is_last() and move to netlink.h

The original motivation for this change was to allow the helper to be used
in files other than actions.c as part of work on an odp select group
action.

It was as pointed out by Thomas Graf that this helper would be best off
living in netlink.h. Furthermore, I think that the generic nature of this
helper means it is best off in netlink.h regardless of if it is used more
than one .c file or not. Thus, I would like it considered independent of
the work on an odp select group action.

Cc: Thomas Graf <tgraf@suug.ch>
Cc: Pravin Shelar <pshelar@nicira.com>
Cc: Andy Zhou <azhou@nicira.com>
Signed-off-by: Simon Horman <simon.horman@netronome.com>
Acked-by: Thomas Graf <tgraf@noironetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Acked-by: Pravin B Shelar <pshelar@nicira.com>
9 years agodatapath-windows: Remove a not needed debug message from Flow.c.
Ankur Sharma [Wed, 5 Nov 2014 20:12:03 +0000 (12:12 -0800)]
datapath-windows: Remove a not needed debug message from Flow.c.

Signed-off-by: Ankur Sharma <ankursharma@vmware.com>
Acked-by: Nithin Raju <nithin@vmware.com>
Signed-off-by: Ben Pfaff <blp@nicira.com>
9 years agoclassifier: Clean up destroy_subtable.
Jarno Rajahalme [Mon, 3 Nov 2014 19:23:11 +0000 (11:23 -0800)]
classifier: Clean up destroy_subtable.

Add asserts to make sure the containers within are already empty.

Signed-off-by: Jarno Rajahalme <jrajahalme@nicira.com>
Acked-by: Ben Pfaff <blp@nicira.com>
9 years agoclassifier: Constify fields.
Jarno Rajahalme [Mon, 27 Oct 2014 17:57:28 +0000 (10:57 -0700)]
classifier: Constify fields.

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
initialization.

Signed-off-by: Jarno Rajahalme <jrajahalme@nicira.com>
Acked-by: Ben Pfaff <blp@nicira.com>
9 years agoclassifier: Make classifier_find_rule_exactly() lockless.
Jarno Rajahalme [Mon, 3 Nov 2014 17:56:54 +0000 (09:56 -0800)]
classifier: Make classifier_find_rule_exactly() lockless.

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 before calling find_equal().  This adds a hash
table lookup in cases where the subtable may already be known to not
contain any rule of the target priority.
classifier_find_rule_exactly() is never called on the fastpath, so
this should not be significant.

Signed-off-by: Jarno Rajahalme <jrajahalme@nicira.com>
Acked-by: Ben Pfaff <blp@nicira.com>
9 years agoclassifier: Constify RCU pointers.
Jarno Rajahalme [Thu, 6 Nov 2014 22:55:29 +0000 (14:55 -0800)]
classifier: Constify RCU pointers.

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 <jrajahalme@nicira.com>
Acked-by: Ben Pfaff <blp@nicira.com>