Michael Vorburger [Fri, 27 Jan 2017 17:24:30 +0000 (18:24 +0100)]
mdsalutil-api clean up Checkstyle violations (not enforced yet)
org.opendaylight.genius.mdsalutil.packet package is almost CS clean now.
Change-Id: Id8b35d4f87560d0e1c8c4f2641ceb7eafefada2a
Signed-off-by: Michael Vorburger <vorburger@redhat.com>
gowri [Tue, 6 Dec 2016 08:21:27 +0000 (13:51 +0530)]
FcapsApplication module migration to blueprint
Change-Id: Iec2908d5aea02d9eb9e10c06b45179247b0a99b7
Signed-off-by: gowri <gowri.r@ericsson.com>
Signed-off-by: Gowri R <gowri.r@ericsson.com>
Michael Vorburger [Sat, 14 Jan 2017 02:56:37 +0000 (03:56 +0100)]
TestIMdsalApiManager with assertFlowsInAnyOrder()
Moved here from netvirt's AclServiceTestBase
Change-Id: Ia4410d2b9bab840ec63f08c8cef8a684291f32cc
Signed-off-by: Michael Vorburger <vorburger@redhat.com>
Michael Vorburger [Fri, 27 Jan 2017 16:23:26 +0000 (17:23 +0100)]
remove apparently un-used Class<K> eventClazz from AsyncListeners
Or is this useful in a very subtle (and undocumented ..) way that eludes
me? I can sort of see the idea may have been for the generic K extends
DataTreeChangeListener<T> ... but it seems weird, to me, to have a
constructor argument that is otherwise un-used (except for a catch &
log; which BTW should never be hit anyway; more on that in a separate PR
coming up next).
Change-Id: I8370dda3e59c80782b8c9dc870d1a684f20d5f18
Signed-off-by: Michael Vorburger <vorburger@redhat.com>
Faseela K [Tue, 31 Jan 2017 10:59:21 +0000 (10:59 +0000)]
Merge "Bug 7531 : Different ids allocated for same key"
Anil Belur [Tue, 31 Jan 2017 09:05:24 +0000 (19:05 +1000)]
InterfaceManagerConfigurationTest newTunnelInterface sleep 2s -> 4s
* Fix similar to change submitted for #51184
* To unblock a problem hit on autorelease Carbon build
Change-Id: I5a7df4866ffee0d6b7a694041e52a1f4cf7e2b36
Signed-off-by: Anil Belur <abelur@linuxfoundation.org>
Michael Vorburger [Fri, 27 Jan 2017 17:04:36 +0000 (18:04 +0100)]
mdsalutil-api clean up Checkstyle violations (not enforced yet)
org.opendaylight.genius.mdsalutil.actions package is CS clean now.
Change-Id: I04b8a17f0efee953bbdad0e30baaccc53f51d678
Signed-off-by: Michael Vorburger <vorburger@redhat.com>
Michael Vorburger [Fri, 27 Jan 2017 16:41:23 +0000 (17:41 +0100)]
Clean up un-neccessary exception handling from AsyncListeners
The catch in close() is pointless; as close() has, intentionally, a
throws Exception. It's up to the caller of close() to handle this.
Or was this useful in a very subtle (and undocumented ..) way that
eludes me?? I mean, registerDataChangeListener() does not have a throws
clause. Sure it could theoretically throw some *RuntimeException, but
so what? Even if it does, that would propagate to the caller of
registerListener() - just like that wrapping IllegalStateException as-is
now also would! So... I think all this is kind of pointless.
TaskRetryLooper used here is then, if above is valid, likewise not
required here, so also removed; does this make sense to everyone?
Class<K> eventClazz from AsyncListeners also because pointless then?
see separate https://git.opendaylight.org/gerrit/#/c/51136/ re. that.
Unless there is something I could learn, where a
registerDataChangeListener() can fail, but then pass on retry - but I'm
kind of struggling to see the scenario here.
I've run into this while looking more closely at the AsyncListeners for
Checktyle violations clean up. Of course, we could also just put a
@SuppressWarnings("checkstyle:IllegalCatch") - but it seemed wrong to
me, in this particular case, so I'm suggesting to change it.
Change-Id: I0feef33bcf5bfbf7329371aa6eaf2160a98c4528
Signed-off-by: Michael Vorburger <vorburger@redhat.com>
Michael Vorburger [Fri, 27 Jan 2017 17:32:32 +0000 (18:32 +0100)]
mdsalutil-api clean up Checkstyle violations (not enforced yet)
org.opendaylight.genius.mdsalutil.matches package is CS clean now.
Change-Id: Iea28d160dc90696a676a529630d51b7da95d6091
Signed-off-by: Michael Vorburger <vorburger@redhat.com>
Michael Vorburger [Mon, 30 Jan 2017 13:38:54 +0000 (14:38 +0100)]
InterfaceManagerConfigurationTest vlanInterfaceTests sleep 2s -> 4s
to unblock a problem hit by Anil Belur on autorelease
It works with 2s locally, but may be slower on buid VMs.
PS: This is a quick fix / hack / workaround; of course, the real long
term solution is to find ways to not need Thread.sleep() at all .. I'd
like like to complete earlier work in this regard, but haven't gotten to
it yet.
Change-Id: I01e0380952ba8c3ee906fae5469dc1b2970d0185
Signed-off-by: Michael Vorburger <vorburger@redhat.com>
Stephen Kitt [Fri, 27 Jan 2017 23:42:22 +0000 (00:42 +0100)]
XtendBeanGenerator support and MatchInfoHelper equals
This patch adds a few missing pieces in the ActionInfo/MatchInfo
redesigns, necessary for some of NetVirt's tests:
* ActionNxConntrack needs getters so that XtendBeanGenerator can be
used;
* MatchIcmpv4 and MatchIcmpv6 need builders (because
XtendBeanGenerator can't find the constructor);
* MatchInfoHelper needs to implement equals (and hashCode) so that its
subclasses' equals implementation will give the correct result.
Change-Id: Iea986dcf4129e778c9c7387fcfc9405dd0bdc877
Signed-off-by: Stephen Kitt <skitt@redhat.com>
David Suarez [Fri, 27 Jan 2017 17:43:43 +0000 (17:43 +0000)]
Merge "Enforce checkstyle in countermanager module"
Stephen Kitt [Mon, 2 Jan 2017 10:15:23 +0000 (11:15 +0100)]
Action redesign: drop obsolete code
Once netvirt and sfc have migrated to the new ActionInfo-based
classes, all the deprecated ActionType-based code can be deleted.
Change-Id: Ie98ab6078710a462df0275b40e22b303962c2482
Signed-off-by: Stephen Kitt <skitt@redhat.com>
Stephen Kitt [Wed, 18 Jan 2017 10:35:22 +0000 (11:35 +0100)]
MatchInfo clean-up
This removes all the obsolete code, and must not be merged before all
downstream users have migrated to the new classes.
Change-Id: I9800c88b80e490bf5c8505ccec1ca0b91d2acf7a
Signed-off-by: Stephen Kitt <skitt@redhat.com>
David Suarez [Mon, 23 Jan 2017 14:59:26 +0000 (15:59 +0100)]
Enforce checkstyle in countermanager module
Checkstyle enforcement for countermanager module:
- Fix exception logging
- Remove Exception catching
- Remove catching of NPE
- PollerThread was meant to be an inner class, now it is
- Formatting
Change-Id: I196dbc68b326f201402b84a08a3a95014571d563
Signed-off-by: David Suarez <david.suarez.fuentes@ericsson.com>
Stephen Kitt [Mon, 2 Jan 2017 15:49:07 +0000 (16:49 +0100)]
InstructionInfo redesign: clean up
Once downstream users have upgraded to the new InstructionInfo
descendants, we can remove InstructionType altogether, and turn
InstructionInfo into an interface.
This patch also changes the AbstractActionInfoList descendants to use
composition rather than inheritance.
Change-Id: Ibfe3cc841f6ad89b29d72e7a4243718059d1c1c5
Signed-off-by: Stephen Kitt <skitt@redhat.com>
Michael Vorburger [Wed, 25 Jan 2017 15:08:47 +0000 (16:08 +0100)]
Use standard static Logger in AsyncDataTreeChangeListenerBase & Co.
I thought it was a smart idea to "use non-static Logger so that logs
easily identifies actual listener class", but in hindsight it clearly
wasn't; this was stupid (of me), because when users configure logging
this is confusing.
Change-Id: I2aaa86cf1dc99f9b5f3e748cb6799f836b2663e0
Signed-off-by: Michael Vorburger <vorburger@redhat.com>
Michael Vorburger [Wed, 25 Jan 2017 13:34:09 +0000 (13:34 +0000)]
Merge "Enforce checkstyle in lockmanager-impl module"
Alon Kochba [Wed, 25 Jan 2017 10:42:07 +0000 (12:42 +0200)]
Remove debug prints for DataTreeChanges
Remove excessive debug prints for any DataTreeChange,
this caused logs to blow up in CSIT runs with a high
DEBUG level, since the actual changes printed to the
log are huge.
Change-Id: I76422b9abde319a84760e7a22c655fc3db822ab0
Signed-off-by: Alon Kochba <alonko@hpe.com>
Kency Kurian [Thu, 12 Jan 2017 11:19:05 +0000 (16:49 +0530)]
Bug 7531 : Different ids allocated for same key
Fixed the issue when allocate id for same idKey returns different idValue.
- Updates a map with a future object if the map is empty during allocate.
- Also remove from map once data is written in order to avoid stale
entries.
- If there is already an entry wait until the future completes.
Also while testing found out that the deleteIdPool RPC was not working
since the listener registration got missed after blueprint migration.
Fixed the JUnit failures.
Change-Id: I0ad4882603ef6472dcd3b780e31175a1a19b49d1
Signed-off-by: Kency Kurian <kency.kurian@ericsson.com>
David Suarez [Mon, 23 Jan 2017 11:54:14 +0000 (12:54 +0100)]
Enforce checkstyle in lockmanager-impl module
Checkstyle enforcement for lockmanager-impl module.
Change-Id: I75fa5c9f0baf4e3c10a269b593d7227238394cef
Signed-off-by: David Suarez <david.suarez.fuentes@ericsson.com>
Michael Vorburger [Mon, 23 Jan 2017 18:58:42 +0000 (19:58 +0100)]
Replace mockito-all by mockito-core (see Bug 7662)
Change-Id: If97b6260f8682c20352d71c9d00de959efc0a45d
Signed-off-by: Michael Vorburger <vorburger@redhat.com>
Michael Vorburger [Mon, 23 Jan 2017 15:19:47 +0000 (16:19 +0100)]
Must always use mockito-core instead of mockito-all
to prevent NoSuchMethodError: org.hamcrest.Matcher ...
Bug 7662 related
Change-Id: I6988023775e35ca3e0babb29aa90b9e1c4eb30fa
Signed-off-by: Michael Vorburger <vorburger@redhat.com>
Stephen Kitt [Thu, 5 Jan 2017 14:22:35 +0000 (15:22 +0100)]
MatchInfo redesign
Change-Id: I83cb7d9a0ab04116f8796dd1b53cc125e237d981
Signed-off-by: Stephen Kitt <skitt@redhat.com>
Michael Vorburger [Wed, 18 Jan 2017 18:07:18 +0000 (19:07 +0100)]
mdsalutil-api Organize Imports for Checkstyle compliance
Change-Id: If3b907cde8a56d4e2c1627c857d995f26b61eb18
Signed-off-by: Michael Vorburger <vorburger@redhat.com>
Michael Vorburger [Fri, 20 Jan 2017 17:31:16 +0000 (17:31 +0000)]
Merge "Enforce checkstyle in itm-api module"
David Suarez [Fri, 20 Jan 2017 13:41:52 +0000 (14:41 +0100)]
Enforce checkstyle in itm-api module
Checkstyle enforcement for itm-api module
There are two FIXMEs that will be removed in a separate patch when
itm-impl is checkstyle-enforced.
Change-Id: Ic94e01aa994965ccdaa39f7af52ee167c99395bd
Signed-off-by: David Suarez <david.suarez.fuentes@ericsson.com>
Michael Vorburger [Wed, 18 Jan 2017 18:26:19 +0000 (19:26 +0100)]
mdsalutil-api various minor cleanups later for Checkstyle compliance
Change-Id: If31889a6ff22d48519493f0cce8dfdf4b62183cd
Signed-off-by: Michael Vorburger <vorburger@redhat.com>
Vishal Thapar [Fri, 20 Jan 2017 13:57:40 +0000 (13:57 +0000)]
Merge "Fix problems with checkstyle in alivenessmonitor-impl module"
David Suarez [Tue, 17 Jan 2017 11:39:08 +0000 (12:39 +0100)]
Fix problems with checkstyle in alivenessmonitor-impl module
Fix several problems found by checkstyle:
- Line lengths
- Unchecked exceptions
- Exception logging
- Formatting
- Redundant public modifier in constructors
This patch is the last of two and it includes the enforcement section
in POM.
Change-Id: Ic556e797a4694d9c148fee20fa96d29d0d70d6c6
Signed-off-by: David Suarez <david.suarez.fuentes@ericsson.com>
Michael Vorburger [Thu, 19 Jan 2017 17:03:36 +0000 (17:03 +0000)]
Merge "InterfaceManagerConfigurationTest @Rule LogRule"
Faseela K [Thu, 19 Jan 2017 16:11:17 +0000 (16:11 +0000)]
Merge "BUG 7576 - print error only if not spurious update"
Vishal Thapar [Thu, 19 Jan 2017 15:53:20 +0000 (15:53 +0000)]
Merge "Enforce checkstyle in lockmanager-api module"
Vishal Thapar [Thu, 19 Jan 2017 15:52:08 +0000 (15:52 +0000)]
Merge "Enforce checkstyle in fcaps-api module"
Michael Vorburger [Tue, 10 Jan 2017 13:20:21 +0000 (14:20 +0100)]
InterfaceManagerConfigurationTest @Rule LogRule
see https://git.opendaylight.org/gerrit/#/c/49499/
Change-Id: I59f48f73794e309739d4a03b57bb9757a87aa451
Signed-off-by: Michael Vorburger <vorburger@redhat.com>
Vishal Thapar [Thu, 19 Jan 2017 15:40:49 +0000 (15:40 +0000)]
Merge "Fix problems with checkstyle in interfacemanager-api module"
Sam Hague [Thu, 19 Jan 2017 15:40:18 +0000 (15:40 +0000)]
Merge "Add dependency to infrautils-testutils to genius.testutils POM"
Michael Vorburger [Thu, 19 Jan 2017 15:35:09 +0000 (15:35 +0000)]
Merge "ITM new style component test (empty)"
Sam Hague [Thu, 19 Jan 2017 15:33:55 +0000 (15:33 +0000)]
Merge "ChainableDataTreeChangeListener with addBeforeListener()"
Vivek Srivastava [Thu, 19 Jan 2017 15:28:52 +0000 (15:28 +0000)]
Merge "Added ability to set mask for ethernet source/destination match"
David Suarez [Thu, 19 Jan 2017 12:50:12 +0000 (13:50 +0100)]
Enforce checkstyle in fcaps-api module
Checkstyle enforcement for fcaps-api module.
Change-Id: I7ec50fd5830f5cd78b379eab0ff31f4c08371f10
Signed-off-by: David Suarez <david.suarez.fuentes@ericsson.com>
David Suarez [Thu, 19 Jan 2017 12:31:06 +0000 (13:31 +0100)]
Enforce checkstyle in lockmanager-api module
Checkstyle enforcement for lockmanager-api module.
Change-Id: Ie345aa80e5afd0e90c086f562a8daa8804cc0767
Signed-off-by: David Suarez <david.suarez.fuentes@ericsson.com>
Michael Vorburger [Wed, 14 Dec 2016 14:33:34 +0000 (15:33 +0100)]
Fix concurrency issue TestIMdsalApiManager assertFlows()
This is step #1 of 3 to fixing occasionally seen (not repro; flaky)
ConcurrentModification in the AclServiceStatefulTest >
AclServiceTestBase.newInterfaceWithDstAllPorts >
newInterfaceWithDstAllPortsCheck:55 ->
AclServiceTestBase.assertFlowsInAnyOrder.
For code review, assume that getFlows() is, about to become, private.
Change-Id: I78c1208d777b825efc37be40d7d059a7a07cb4d6
Signed-off-by: Michael Vorburger <vorburger@redhat.com>
Guy Sela [Wed, 18 Jan 2017 11:23:53 +0000 (13:23 +0200)]
BUG 7576 - print error only if not spurious update
Change-Id: I03fa7e7e859d35cf5acbe21e49f093ac0ad89fe2
Signed-off-by: Guy Sela <guy.sela@hpe.com>
David Suarez [Wed, 18 Jan 2017 10:30:36 +0000 (11:30 +0100)]
Fix problems with checkstyle in interfacemanager-api module
Fix several problems found by checkstyle:
- Incomplete JavaDoc
- Formatting
- Line lengths
- Re-order methods and constructors
- Optimize imports
Change-Id: I52c5a63967406f03289750d3eca8931dc11e70d9
Signed-off-by: David Suarez <david.suarez.fuentes@ericsson.com>
Michael Vorburger [Tue, 13 Dec 2016 18:22:42 +0000 (19:22 +0100)]
ChainableDataTreeChangeListener with addBeforeListener()
in addition to addAfterListener(), to properly "track" asynchronous
event processing more reliably than with the current implementation.
The new approach also also to catch "forgotten" await() on close() at
the end of a @Test.
Verified that netvirt's AclServiceStatelessTest/AclServiceStatefulTest
(AclServiceTestBase), which is so far the only use of AsyncEventsWaiter
in master, still pass with this change.
This change has 0 impact on run-time production code performance; its
just test infrastructure which registered chained listeners, in test
wiring/configuration (Guice), only.
Change-Id: I8f3a6dd2bf2fd7d4e85655d8911d734e6d0089bf
Signed-off-by: Michael Vorburger <vorburger@redhat.com>
Michael Vorburger [Tue, 10 Jan 2017 23:40:41 +0000 (00:40 +0100)]
ITM impl Organize Imports for Checkstyle compliance
as well as converting some tabs to spaces etc.
Change-Id: Ie6ecab9595bacd81dfac4d033cec71589bd2fefb
Signed-off-by: Michael Vorburger <vorburger@redhat.com>
Change-Id: Ie6ecab9595bacd81dfac4d033cec71589bd2fefb
Signed-off-by: Michael Vorburger <vorburger@redhat.com>
Kiran N Upadhyaya [Wed, 18 Jan 2017 07:30:28 +0000 (13:00 +0530)]
Ignoring a Junit test case in Idmanager to unblock autorelease
Change-Id: Ia4ad1b479c6ab89235376c604753b2e91f61f8d1
Signed-off-by: Kiran N Upadhyaya <kiran.n.upadhyaya@ericsson.com>
Michael Vorburger [Fri, 16 Dec 2016 17:43:46 +0000 (18:43 +0100)]
ITM new style component test (empty)
It's empty, but still already very useful as-is, because even without
real test methods in it yet it will already ensure non-regression at
least of all the internal binding wiring start-up.
This change all introduces dumb hard-coded cache invalidation. This is
required because we've seen that without doing this @After ItmTest,
other tests start to fail, because there are isolation issues. The real
solution is to de-static-ify all this stuff, but that's a bit more work,
so for now, and in order to unblock and enable people to write new style
component tests in ITM, this "hack" seems a good short term solution.
Change-Id: Ie12e43d6097bc50ff95d2edbff1926e5d4520193
Signed-off-by: Michael Vorburger <vorburger@redhat.com>
Faseela K [Tue, 17 Jan 2017 16:01:17 +0000 (16:01 +0000)]
Merge "Order imports to comply with checkstyle"
David Suarez [Tue, 17 Jan 2017 13:13:33 +0000 (13:13 +0000)]
Merge "Action redesign: nx_resubmit backwards compatibility"
David Suarez [Mon, 16 Jan 2017 09:51:35 +0000 (10:51 +0100)]
Order imports to comply with checkstyle
Re-order imports to comply with checkstyle in alivenessmonitor-impl
module.
Change-Id: Id0b5b94015a268a5c9616c850a0296214ee9d7e7
Signed-off-by: David Suarez <david.suarez.fuentes@ericsson.com>
Stephen Kitt [Mon, 16 Jan 2017 15:13:55 +0000 (16:13 +0100)]
Instruction redesign: add missing equals/hashCode
Change-Id: I418b9bcd1e08c910fb8fa9bb83d44c7befee3ed1
Signed-off-by: Stephen Kitt <skitt@redhat.com>
Michael Vorburger [Mon, 16 Jan 2017 15:50:05 +0000 (16:50 +0100)]
Add dependency to infrautils-testutils to genius.testutils POM
This is just "for convenience" - projects can add a single dependency to
genius.testutils POM to get all commonly used testutils together,
instead of having to remember each one of them.
Change-Id: Ie58eb0d9f41a65742687bf6f0b1ad998622abd20
Signed-off-by: Michael Vorburger <vorburger@redhat.com>
Periyasamy Palanisamy [Mon, 19 Dec 2016 06:46:56 +0000 (12:16 +0530)]
DataStoreJobCoordinator API for pending tasks count (useful for testing)
* added counter for jobs_incomplete
* exposing API for getIncompleteTaskCount
Test Driven Development (TDD) so includes self unit test.
Change-Id: I6b1d824804ab73c0b57929d8fe057bc783231308
Signed-off-by: Periyasamy Palanisamy <periyasamy.palanisamy@ericsson.com>
Signed-off-by: Michael Vorburger <vorburger@redhat.com>
Stephen Kitt [Mon, 16 Jan 2017 15:17:00 +0000 (16:17 +0100)]
Action redesign: nx_resubmit backwards compatibility
ActionNxResubmit doesn't set the old tables up correctly, this patch
fixes this.
Change-Id: Ia5596b16118bea0dc5e6981dba19393f0311f486
Signed-off-by: Stephen Kitt <skitt@redhat.com>
Faseela K [Mon, 16 Jan 2017 14:37:37 +0000 (14:37 +0000)]
Merge "Bug 7466 - NPE thrown for interface without lport tag"
Faseela K [Mon, 16 Jan 2017 11:14:34 +0000 (11:14 +0000)]
Merge "Minor cleanup and formatting"
Yakir Dorani [Mon, 16 Jan 2017 09:47:09 +0000 (11:47 +0200)]
Added ability to set mask for ethernet source/destination match
Change-Id: I61143c5ff7c046d8d2d73012feee204d2b0bd4cb
Signed-off-by: Yakir Dorani <yakir.dorani@hpe.com>
Stephen Kitt [Mon, 2 Jan 2017 14:30:03 +0000 (15:30 +0100)]
InstructionInfo redesign
This follows the same principles as the ActionInfo redesign; given the
small number of classes involves, the redesign is done in a two
commits (this backwards-compatible change, introducing the new
InstructionInfo descendents, and another clean-up which will be
applicable once the downstream users of InstructionInfo have been
fixed).
Change-Id: Ibfa7a8dc65f36c5f599dbdeffc24508fd136e198
Signed-off-by: Stephen Kitt <skitt@redhat.com>
David Suarez [Sun, 15 Jan 2017 18:29:13 +0000 (19:29 +0100)]
Minor cleanup and formatting
- Minor clean and formatting.
- Speedup lastIndexOf changing to char.
Change-Id: I56f47b90ffb36a057b1b7b3e7faa3b4b154bd35c
Signed-off-by: David Suarez <david.suarez.fuentes@ericsson.com>
David Suarez [Sat, 14 Jan 2017 11:39:50 +0000 (11:39 +0000)]
Merge "BUG-5222: do not pull in odl-mdsal-xsql"
Vishal Thapar [Sat, 14 Jan 2017 11:36:03 +0000 (11:36 +0000)]
Merge "Fix Idmanager JUnit test case"
Robert Varga [Fri, 13 Jan 2017 10:23:31 +0000 (11:23 +0100)]
BUG-5222: do not pull in odl-mdsal-xsql
xsql is deprecated, buggy and never was meant for production
use. Do not pull it in.
Change-Id: I3ea0428b7e74c103d87df8cb108e29285647c335
Signed-off-by: Robert Varga <nite@hq.sk>
Miguel Perez [Fri, 13 Jan 2017 14:51:54 +0000 (15:51 +0100)]
Cleanup unnecessary 'toStringIpAddress'
+ 'toStringIpAddress' is centralized in NWUtil
+ projects should use that one instead of their own copy
Change-Id: I44db9c910d8de57a6f1e6070e9cc37f61c1bc9f2
Signed-off-by: Miguel Perez <francisco.miguel.perez@ericsson.com>
Kiran N Upadhyaya [Sat, 14 Jan 2017 09:31:35 +0000 (15:01 +0530)]
Fix Idmanager JUnit test case
The JUnit used to test concurrent Id requests from Released Ids has to
wait before assering that values in the IdPool are consistent, as the
ReleaseId allocator uses DJC(spawing yet other independent threads) to
update the ReleasedId container with the correct values on every
allocation.
Change-Id: Ib6e69dc58c7436d9eca6b7c58098f5c307bf5ea7
Signed-off-by: Kiran N Upadhyaya <kiran.n.upadhyaya@ericsson.com>
Faseela K [Sat, 14 Jan 2017 02:06:45 +0000 (02:06 +0000)]
Merge "Enforce checkstyle validations"
Michael Vorburger [Tue, 10 Jan 2017 13:27:16 +0000 (14:27 +0100)]
interfacemanager Organize Imports for Checkstyle compliance
Change-Id: I0b44d9f431065629b46987fc2f28caf8e89883bd
Signed-off-by: Michael Vorburger <vorburger@redhat.com>
David Suarez [Fri, 13 Jan 2017 22:50:48 +0000 (22:50 +0000)]
Merge "TestableDataTreeChangeListenerModule with logging"
David Suarez [Fri, 13 Jan 2017 22:28:05 +0000 (23:28 +0100)]
Enforce checkstyle validations
Checkstyle enforcement.
Change-Id: I061bb99afeabeefa887029aa67a72cf0bb1dcc64
Signed-off-by: David Suarez <david.suarez.fuentes@ericsson.com>
Michael Vorburger [Fri, 13 Jan 2017 21:46:52 +0000 (22:46 +0100)]
TestableDataTreeChangeListenerModule with logging
Change-Id: I700186eb135dfd641a6804838e9611547c679345
Signed-off-by: Michael Vorburger <vorburger@redhat.com>
David Suarez [Fri, 13 Jan 2017 20:46:56 +0000 (21:46 +0100)]
Enforce checkstyle validations
Checkstyle enforcement.
Change-Id: I8ceb5e261a0e925c75237900c6cbf2c583cf173c
Signed-off-by: David Suarez <david.suarez.fuentes@ericsson.com>
Faseela K [Fri, 13 Jan 2017 16:21:33 +0000 (16:21 +0000)]
Merge "BUG 7494 : Idmanager returns the same Id from the same pool for different threads with different id keys"
Kiran N Upadhyaya [Mon, 9 Jan 2017 11:32:45 +0000 (17:02 +0530)]
BUG 7494 : Idmanager returns the same Id from the same pool for different
threads with different id keys
Problem with concurrency in IdManager, such that two threads requesting Id
at the same time with different keys might end up getting the same key.
This is because of unsafe read of an in-memory cached local Id Pool.
Also, added IdManager test cases for multiple threads requesting
allocated/ released Ids concurrently.
Change-Id: I3d8c0e921461c4ecc5d3ca4ed41b9d18fee9a725
Signed-off-by: Kiran N Upadhyaya <kiran.n.upadhyaya@ericsson.com>
David Suarez [Fri, 13 Jan 2017 11:48:15 +0000 (11:48 +0000)]
Merge "interfacemanager Checkstyle compliance: variable names & line length"
Tomer Pearl [Thu, 12 Jan 2017 14:37:17 +0000 (16:37 +0200)]
Bug 7466 - NPE thrown for interface without lport tag
Code uses Integer that throws NPE when unboxing.
Now it checks for null beforehand
Change-Id: I9e6e513327c3231c0a98e73ac1738b06058e73c8
Signed-off-by: Tomer Pearl <tomer.pearl@hpe.com>
Stephen Kitt [Thu, 22 Dec 2016 14:04:43 +0000 (15:04 +0100)]
Action redesign: implement equals() and hashCode()
Change-Id: Ie6e86ebac788492220419ed6e6beae6b04c59b1e
Signed-off-by: Stephen Kitt <skitt@redhat.com>
Stephen Kitt [Fri, 16 Dec 2016 13:39:15 +0000 (14:39 +0100)]
Action redesign: migrate set_{destination,source}_ip
Change-Id: I557ad1d1b79a224c5d4adc9b71c2b0aa00652165
Signed-off-by: Stephen Kitt <skitt@redhat.com>
Stephen Kitt [Fri, 16 Dec 2016 15:06:41 +0000 (16:06 +0100)]
Action redesign: migrate NICIRA ARP extensions
Change-Id: I64583a3f57028af371ff1f0c7a8cb2359c384501
Signed-off-by: Stephen Kitt <skitt@redhat.com>
Stephen Kitt [Fri, 16 Dec 2016 10:18:18 +0000 (11:18 +0100)]
Action redesign: migrate move_src_dst_{eth,ip}
Change-Id: Idda8f0ee6b0ebb91514bb2562e483e47362fb973
Signed-off-by: Stephen Kitt <skitt@redhat.com>
Stephen Kitt [Fri, 16 Dec 2016 14:27:53 +0000 (15:27 +0100)]
Action redesign: migrate nx_load_in_port
Change-Id: Ia3975bf642439cc33166613a5c2bfa8e2c58cf52
Signed-off-by: Stephen Kitt <skitt@redhat.com>
Stephen Kitt [Wed, 14 Dec 2016 16:04:07 +0000 (17:04 +0100)]
Action redesign: migrate set_{tcp,udp}_{dest,src}_port
Change-Id: I45da9e1d0c847b0f9b40cf06636ae6a8db6596a6
Signed-off-by: Stephen Kitt <skitt@redhat.com>
Stephen Kitt [Fri, 16 Dec 2016 14:04:49 +0000 (15:04 +0100)]
Action redesign: migrate set_icmp_type
Change-Id: Id9c36b7967c33013b4e43c5f161f654d2ddd1a8a
Signed-off-by: Stephen Kitt <skitt@redhat.com>
Stephen Kitt [Wed, 14 Dec 2016 14:54:15 +0000 (15:54 +0100)]
Action redesign: migrate set_udp_protocol
Change-Id: Id5d73de28e83f48840e591cea7a3240ddffb9147
Signed-off-by: Stephen Kitt <skitt@redhat.com>
David Suarez [Thu, 12 Jan 2017 11:23:52 +0000 (11:23 +0000)]
Merge "Action redesign: migrate punt_to_controller"
David Suarez [Thu, 12 Jan 2017 11:19:39 +0000 (11:19 +0000)]
Merge "Action redesign: migrate set_field_eth_{dest,src}"
David Suarez [Thu, 12 Jan 2017 11:18:12 +0000 (11:18 +0000)]
Merge "Action redesign: convert drop_action and nx_resubmit"
David Suarez [Thu, 12 Jan 2017 11:07:11 +0000 (11:07 +0000)]
Merge "Action redesign: convert set_tunnel_{src,dest}_ip"
David Suarez [Thu, 12 Jan 2017 11:06:48 +0000 (11:06 +0000)]
Merge "Action redesign: add typed getters in NxmOfFieldType"
David Suarez [Thu, 12 Jan 2017 11:05:57 +0000 (11:05 +0000)]
Merge "Action redesign: always provide actionKey constructors"
Faseela K [Thu, 12 Jan 2017 09:47:56 +0000 (09:47 +0000)]
Merge "Fix some minor problems."
Stephen Kitt [Wed, 14 Dec 2016 14:43:14 +0000 (15:43 +0100)]
Action redesign: migrate punt_to_controller
Change-Id: I62a64db77432c3ed7174181f12404a4d5f8db214
Signed-off-by: Stephen Kitt <skitt@redhat.com>
Stephen Kitt [Wed, 14 Dec 2016 14:00:50 +0000 (15:00 +0100)]
Action redesign: migrate set_field_eth_{dest,src}
Change-Id: I77b1ed6a9e76d7b66b7cd43941ead50ee01482fb
Signed-off-by: Stephen Kitt <skitt@redhat.com>
Stephen Kitt [Thu, 8 Dec 2016 20:13:31 +0000 (21:13 +0100)]
Action redesign: convert drop_action and nx_resubmit
Change-Id: Ie70eae9c06f36b57dcfc92933b7d816d033d15d3
Signed-off-by: Stephen Kitt <skitt@redhat.com>
Stephen Kitt [Thu, 8 Dec 2016 19:36:32 +0000 (20:36 +0100)]
Action redesign: convert set_tunnel_{src,dest}_ip
Change-Id: I286d333ef38fb3b59b2a41af9e52c755c96776b8
Signed-off-by: Stephen Kitt <skitt@redhat.com>
Stephen Kitt [Thu, 8 Dec 2016 02:19:49 +0000 (03:19 +0100)]
Action redesign: add typed getters in NxmOfFieldType
This is slightly inelegant since we can't replace getHexType() and
getFlowModHeaderLen() wholesale; we need to deprecate them in Carbon,
then replace them in Argon and deprecate the new functions, and
finally replace the now-useless new functions in Oxygen.
Change-Id: If63f9f732832e946ca43390cc6157840601d6da2
Signed-off-by: Stephen Kitt <skitt@redhat.com>
Stephen Kitt [Wed, 7 Dec 2016 16:09:13 +0000 (17:09 +0100)]
Action redesign: always provide actionKey constructors
netvirt needs these...
Change-Id: I03da2ddfe18c4a2bdec3fba9fef33957743f7780
Signed-off-by: Stephen Kitt <skitt@redhat.com>
Stephen Kitt [Wed, 7 Dec 2016 10:45:03 +0000 (11:45 +0100)]
Action redesign: ActionType.set_field_...
This patch switches MPLS labels, PBB ISID, VLAN and tunnel id handling
to the new model.
Change-Id: I18b621016f490bf84b4f68d1bfb78e14addc6030
Signed-off-by: Stephen Kitt <skitt@redhat.com>
Stephen Kitt [Wed, 7 Dec 2016 07:42:34 +0000 (08:42 +0100)]
Action redesign: refactor ActionType.learn
Change-Id: I5ffe1d46dd1a88d25a466a976ab4cd03c532ea69
Signed-off-by: Stephen Kitt <skitt@redhat.com>
David Suarez [Tue, 10 Jan 2017 10:41:08 +0000 (11:41 +0100)]
Fix some minor problems.
* Logger name
* Fix exception log
* Name of attributes
Change-Id: I11e3d829fa735c8a804ce9df71fa91da312eb072
Signed-off-by: David Suarez <david.suarez.fuentes@ericsson.com>