forked from OpenDDS/OpenDDS
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathChangeLog-2.1.3
1026 lines (662 loc) · 40 KB
/
ChangeLog-2.1.3
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
Fri Jun 18 20:15:06 UTC 2010 Adam Mitz <[email protected]>
* OpenDDS version 2.1.3 released.
Fri Jun 18 20:08:49 UTC 2010 Adam Mitz <[email protected]>
* README:
Updated TAO and compiler versions for forthcoming release.
Thu Jun 17 20:09:07 UTC 2010 Scott Harris <[email protected]>
* NEWS:
Added TTL and interupt suppression.
Thu Jun 17 20:05:08 UTC 2010 Adam Mitz <[email protected]>
* ChangeLog:
Attempted to normalize the formatting and whitespace of the recent
ChangeLog messages. Please use cle.pl and zap trailing whitespace.
* NEWS:
Formatted the NEWS to look like all the other releases.
Thu Jun 17 19:59:05 UTC 2010 Yan Dai <[email protected]>
* performance-tests/Bench/src/Process.cpp:
Extended the sub timeout on waiting for pub termination.
* NEWS:
Added changes after last release.
Thu Jun 17 19:50:31 UTC 2010 Don Hudson <[email protected]>
* dds/DCPS/transport/framework/TransportReceiveStrategy.h:
Increased RECEIVED_BUFFERS from 2 to 16 and BUFFER_LOW_WATER from 1500 to 4096.
This is to better insure that the total available space in the receive buffers
can hold the max size message (64K).
* dds/DCPS/transport/framework/TransportReceiveStrategy.cpp:
Changed receive_bytes error from debug message to warning message because some
errors may be non-recoverable.
* dds/DCPS/transport/framework/TransportSendStrategy.cpp:
Changed OUTCOME_SEND_ERROR from debug message to warning message because some
errors may be non-recoverable.
* dds/DCPS/transport/multicast/MulticastDataLink.cpp:
Removed unused method name parameter from ACE_ERROR usages
(not needed as method is described in error messages).
* dds/DCPS/transport/multicast/ReliableSession.cpp:
Changed send_naks() to do nothing if the session hasn't been acknowledged.
Changed nak_received() to send nakack only if the send buffer no longer has
the lowest sequence number requested.
Thu Jun 17 19:47:17 UTC 2010 Scott Harris <[email protected]>
* dds/DCPS/Service_Participant.cpp:
Avoid OpenDDS failures because of interupts while in system calls.
Thu Jun 17 19:44:59 UTC 2010 Mike Martinez <[email protected]>
* NEWS:
Added OpenDDS-Bench entry for point release.
Thu Jun 17 15:51:26 UTC 2010 Don Hudson <[email protected]>
* dds/DCPS/Definitions.h:
Corrected constructor of SequenceNumber to modulo by SHRT_MAX + 1 as SHRT_MAX is a valid SequenceNumber.
Thu Jun 17 15:41:41 UTC 2010 Adam Mitz <[email protected]>
* dds/DCPS/Service_Participant.cpp:
Fixed a typo in command-line argument processing. Thanks to Shawn
Mitchell <shawn dot mitchell at Q1Labs dot com> for reporting it.
Wed Jun 16 06:25:37 UTC 2010 Yan Dai <[email protected]>
* dds/DCPS/transport/framework/DataLinkWatchdog_T.h:
Fixed threading issue for internal state setting which causes
access violation during shutdown.
Tue Jun 15 19:35:29 UTC 2010 Don Hudson <[email protected]>
* dds/DCPS/transport/multicast/MulticastConfiguration.cpp:
* dds/DCPS/transport/multicast/MulticastDataLink.cpp:
Removed support for IP_MULTICAST_LOOP as it isn't supported by TAO 1.5 and it
isn't a required feature for OpenDDS multicast.
Tue Jun 15 17:01:42 UTC 2010 Yan Dai <[email protected]>
* dds/DCPS/transport/framework/DataLinkWatchdog_T.h:
Fixed a warning.
Tue Jun 15 16:52:06 UTC 2010 Yan Dai <[email protected]>
* dds/DCPS/transport/framework/DataLinkWatchdog_T.h:
Fixed a deadlock situation during shutdown. One thread calls
the timer cancellation while the reactor thread handling timeout.
The DataLinkWatchdog internal lock and reactor lock are acquired
reversely. Added reverse lock to solve the issue.
* dds/DCPS/transport/multicast/ReliableSession.cpp:
- Set send_naks_ flag based on active/passive role of the session
instead of setting upon send_syn() call and not schedule nak timer
if session is for pub.
- Added debug outpt to help detect lost message.
* dds/DCPS/ZeroCopySeq_T.inl:
* dds/DCPS/transport/multicast/MulticastDataLink.cpp:
* dds/DCPS/transport/multicast/MulticastTransport.cpp:
* performance-tests/DCPS/MulticastListenerTest/DataReaderListener.cpp:
Added missing include.
Mon Jun 14 19:01:17 UTC 2010 Yan Dai <[email protected]>
* dds/DCPS/ZeroCopySeq_T.inl:
* dds/DCPS/transport/multicast/MulticastDataLink.cpp:
* dds/DCPS/transport/multicast/MulticastTransport.cpp:
* dds/DCPS/transport/multicast/ReliableSession.cpp:
* performance-tests/DCPS/MulticastListenerTest/DataReaderListener.cpp:
Fixed warnings.
Mon Jun 14 06:28:29 UTC 2010 Yan Dai <[email protected]>
* dds/DCPS/transport/multicast/MulticastDataLink.cpp:
Corrected session_lock_ scope in sample_received()
missed in last commit.
Mon Jun 14 04:26:23 UTC 2010 Yan Dai <[email protected]>
* dds/DCPS/transport/multicast/MulticastDataLink.h:
* dds/DCPS/transport/multicast/MulticastDataLink.cpp:
* dds/DCPS/transport/multicast/ReliableSession.h:
* dds/DCPS/transport/multicast/ReliableSession.cpp:
- Resolved a deadlock situation due to reverse locking during
pub and sub association.
One thread calls add_associations which acquires TransportImpl
reservation_lock_ and then MulticastDataLink session_lock_.
The reactor thread handle the control message which acquires
MulticastDataLink session_lock_ and then TransportImpl
reservation_lock_ due to check_fully_association() call to
TransportImpl. To solve the deadlock, the check_fully_association()
is called outside of session_lock_ locking.
- Updated to send one nak message per DisjointSequence with all ranges
instead of one per each range. This would reduce nak message overflow.
* performance-tests/DCPS/MulticastListenerTest/DataReaderListener.h:
* performance-tests/DCPS/MulticastListenerTest/DataReaderListener.cpp:
* performance-tests/DCPS/MulticastListenerTest/publisher.cpp:
* performance-tests/DCPS/MulticastListenerTest/run_test-16p4s.pl:
* performance-tests/DCPS/MulticastListenerTest/subscriber.cpp:
Added test result verify using DisjointSequence and added 16p4s test
script.
Sat Jun 12 15:21:00 UTC 2010 Mike Martinez <[email protected]>
* dds/DCPS/transport/framework/TransportSendBuffer.inl:
Added ACE_INLINE qualifier.
* dds/DCPS/transport/multicast/MulticastConfiguration.cpp:
Changed an empty #elif to #else to avoid a compilation
complaint.
Fri Jun 11 12:27:24 UTC 2010 Don Hudson <[email protected]>
* dds/DCPS/DataReaderImpl.cpp:
Moved deletion of writer association info within check that iterator isn't at end.
* dds/DCPS/DataWriterImpl.cpp:
Added removal of pending associations when removing all.
* dds/DCPS/Definitions.h:
Corrected <operator.
* dds/DCPS/DisjointSequence.inl:
Corrected DisjointSequence::range_iterator when there is no gap between two elements in the sequence.
* dds/DCPS/DisjointSequence.cpp:
Corrected shifting of low water mark.
* dds/DCPS/SubscriberImpl.cpp:
Added nil check before using listener.
* dds/DCPS/transport/framework/DataLink.cpp:
Added debug warning to ack_received() for publication not found as the ack may be for
another link (ack for a publication that doesn't have a listener is still an error.)
* dds/DCPS/transport/framework/TransportImpl.cpp:
Added scope around guard in OpenDDS::DCPS::TransportImpl::demarshal_acks(ACE_Message_Block* acks, bool byte_order)
and added guard to void OpenDDS::DCPS::TransportImpl::check_fully_association() so guard isn't skipped when
check_fully_association() is called from ReliableSession::synack_received().
* dds/DCPS/transport/framework/TransportQueueElement.cpp:
Correcting cloning of messages that have continuation blocks
* dds/DCPS/transport/framework/TransportReceiveStrategy.h:
* dds/DCPS/transport/framework/TransportReceiveStrategy.cpp:
Corrected computation of pdu_remaining to use buffer->length() instead of buffer->space().
Added checking of sample header and skipping of samples that fail the check.
* dds/DCPS/transport/framework/TransportSendBuffer.h:
* dds/DCPS/transport/framework/TransportSendBuffer.inl:
Added empty().
* dds/DCPS/transport/multicast/BestEffortSession.h:
* dds/DCPS/transport/multicast/BestEffortSession.cpp:
Changed header_received(const TransportHeader& header) to check_header(const TransportHeader& header)
which better reflects its purpose.
Added virtual bool check_header(const DataSampleHeader& header) that returns true
with comment as to why it returns true (this mirrors existing method check_header(const TransportHeader& header)).
* dds/DCPS/transport/multicast/MulticastConfiguration.h:
* dds/DCPS/transport/multicast/MulticastConfiguration.cpp:
* dds/DCPS/transport/multicast/MulticastDataLink.cpp:
Added support for configuring the socket receive buffer size by transport through new configuration option: rcv_buffer_size.
Increasing the receive buffer size can reduce dropped samples (which contribute to the nak storms), especially
in a many-to-one multi-cast topology.
* dds/DCPS/transport/multicast/MulticastDataLink.cpp:
Added support for configuring multicast loop back through use of ACE option -ORBIPMulticastLoop.
When multicast loop back is on (by default), everything in the send buffer is copied to the recieve buffer.
Multicast loop back can be turned off when producers are on separate machines from subscribers; this can improve performance.
* dds/DCPS/transport/multicast/MulticastDataLink.h:
Changed header_received(const TransportHeader& header) to check_header(const TransportHeader& header)
which better reflects its purpose.
Added bool MulticastDataLink::check_header(const DataSampleHeader& header).
* dds/DCPS/transport/multicast/MulticastDataLink.cpp:
Changed to Skip a data sample unless there is a session for it (control samples aren't skipped).
Reset rd_ptr of control message when looping through sessions.
* dds/DCPS/transport/multicast/MulticastReceiveStrategy.h:
Added virtual bool check_header(const DataSampleHeader& header).
* dds/DCPS/transport/multicast/MulticastReceiveStrategy.cpp:
Changed to skip samples that were sent by itself.
Changed to skip samples that have already been processed.
* dds/DCPS/transport/multicast/MulticastSession.h:
Changed header_received(const TransportHeader& header) to check_header(const TransportHeader& header)
which better reflects its purpose.
* dds/DCPS/transport/multicast/ReliableSession.h:
* dds/DCPS/transport/multicast/ReliableSession.cpp:
Recorded control messages in header received list even if the session hasn't been acknowedged.
Added additional filtering out of control messages that aren't to the session.
Added additional nakacks to reduce naks from remotes.
Limited nakack warning messages to DCPS_debug_level 0 and made warning message more generic.
Disabled sending naks from producers and receiving naks by subscribers since only producers
send data samples, control samples don't need to be resent, and implementation doesn't
support a session being both a producer and subscriber.
Changed header_received(const TransportHeader& header) to check_header(const TransportHeader& header)
which better reflects its purpose.
* performance-tests/Bench/etc/transport-multi-rel.ini:
Tweaked configuration of reliable multicast transports to reduce nak storms.
* tests/DCPS/FooTest3_0/SimpleSubscriber.h:
* tests/DCPS/FooTest3_0/SimpleSubscriber.cpp:
* tests/DCPS/FooTest3_0/SubDriver.cpp:
* tests/DCPS/FooTest3_1/SimpleSubscriber.h:
* tests/DCPS/FooTest3_1/SimpleSubscriber.cpp:
* tests/DCPS/FooTest3_1/SubDriver.cpp:
* tests/DCPS/FooTest3_2/SimpleSubscriber.h:
* tests/DCPS/FooTest3_2/SimpleSubscriber.cpp:
* tests/DCPS/FooTest3_2/SubDriver.cpp:
* tests/transport/simple/PubDriver.cpp:
* tests/transport/simple/SimplePublisher.h:
* tests/transport/simple/SimplePublisher.cpp:
* tests/transport/simple/SimpleSubscriber.h:
* tests/transport/simple/SimpleSubscriber.cpp:
* tests/transport/simple/SubDriver.cpp:
* tests/transport/simple_bp/PubDriver.cpp:
* tests/transport/simple_bp/SimplePublisher.h:
* tests/transport/simple_bp/SimplePublisher.cpp:
* tests/transport/simple_bp/SimpleSubscriber.h:
* tests/transport/simple_bp/SimpleSubscriber.cpp:
* tests/transport/simple_bp/SubDriver.cpp:
* tests/transport/simple_bp_rm/PubDriver.cpp:
* tests/transport/simple_bp_rm/SimplePublisher.h:
* tests/transport/simple_bp_rm/SimplePublisher.cpp:
* tests/transport/simple_bp_rm/SimpleSubscriber.h:
* tests/transport/simple_bp_rm/SimpleSubscriber.cpp:
* tests/transport/simple_bp_rm/SubDriver.cpp:
* tests/transport/simple_n/PubDriver.cpp:
* tests/transport/simple_n/SimplePublisher.h:
* tests/transport/simple_n/SimplePublisher.cpp:
* tests/transport/simple_n/SimpleSubscriber.h:
* tests/transport/simple_n/SimpleSubscriber.cpp:
* tests/transport/simple_n/SubDriver.cpp:
Added remove association before shutting down tests that don't go through the normal shutdown steps.
This stops warning from ~DataLink() that link is still in use.
Tue Jun 1 21:40:48 UTC 2010 Yan Dai <[email protected]>
* dds/DCPS/RcHandle_T.h:
The operator=(T* p) needs remove reference from current
handle if it's same object as provided, otherwise, the
object will be leaked. This fixed RepoIdSet and
ReceiveListenerSet leaks.
* dds/DCPS/transport/framework/TransportImpl.cpp:
Fixed association info leak.
* dds/DCPS/transport/framework/TransportControlElement.h:
* dds/DCPS/transport/framework/TransportControlElement.cpp:
* dds/DCPS/transport/framework/TransportQueueElement.h:
* dds/DCPS/transport/framework/TransportReplacedElement.h:
* dds/DCPS/transport/framework/TransportReplacedElement.cpp:
* dds/DCPS/transport/framework/TransportRetainedElement.h:
* dds/DCPS/transport/framework/TransportRetainedElement.inl:
* dds/DCPS/transport/framework/TransportSendControlElement.h:
* dds/DCPS/transport/framework/TransportSendControlElement.inl:
* dds/DCPS/transport/framework/TransportSendElement.h:
* dds/DCPS/transport/framework/TransportSendElement.inl:
* dds/DCPS/transport/framework/TransportSendStrategy.cpp:
* dds/DCPS/transport/simpleTCP/SimpleTcpDataLink.cpp:
Fixed the TransportControl message leaks.
* tests/DCPS/Messenger/publisher.cpp:
* tests/DCPS/Messenger/subscriber.cpp:
Fixed leaks due to get_type_name() not assigned to String_var.
Fri May 28 16:29:14 UTC 2010 Adam Mitz <[email protected]>
* java/build_scripts/jni_check.pl:
Merged 64-bit Windows support from the sbir branch.
* java/idl2jni/codegen/im_java.cpp:
* java/idl2jni/codegen/im_jni.cpp:
Handle AST_Decl::NT_interface_fwd (forward declarations) as if they
were full declarations. I don't think we should be seeing these,
but something has changed in the TAO_IDL_FE.
Mon May 24 21:41:41 UTC 2010 Scott Harris <[email protected]>
* dds/DCPS/transport/framework/PerConnectionSynch.cpp:
* dds/DCPS/transport/framework/ThreadPerConnectionSendTask.cpp:
* dds/DCPS/transport/framework/TransportReactorTask.cpp:
* dds/DCPS/transport/simpleTCP/SimpleTcpReconnectTask.cpp:
Avoid OpenDDS failures because of interupts while in system calls.
Example: ERROR: ACE::handle_write_ready return -1 while waiting
to unclog. handle_write_ready: Interrupted system call
Fri May 21 10:38:10 UTC 2010 Brian Johnson <[email protected]>
* tests/DCPS/Crossplatform/test_list.txt:
Changing Crossplatform nightly build to be on cicada2.
Wed May 19 16:26:44 UTC 2010 Adam Mitz <[email protected]>
* MPC/config/dcps_ts_defaults.mpb:
* MPC/config/java.mpb:
* java/dds/dcps_java_optional.mpb:
Changed "command" to just the executable (perl) and put the
arguments in "commandflags". This is for compatibility with the
latest gnuace changes in the DOC group repository.
* java/INSTALL:
Added vc9 projects to documentation.
* tools/odds_dissector/README:
Added notes on the difference between the wireshark install dir
and WIRESHARK_ROOT.
Wed May 12 21:02:47 UTC 2010 Yan Dai <[email protected]>
* dds/DCPS/transport/framework/PacketRemoveVisitor.cpp:
* dds/DCPS/transport/framework/QueueRemoveVisitor.cpp:
Fixed bugs in QueueRemoveVisitor::visit_element_remove and
PacketRemoveVisitor::visit_element_ref that access samples
after it's released.
Wed May 12 03:50:38 UTC 2010 Mike Martinez <[email protected]>
* performance-tests/Bench/bin/lj-plots.gpi:
* performance-tests/Bench/bin/plot-density.gpi:
* performance-tests/Bench/bin/plot-jitter.gpi:
* performance-tests/Bench/bin/plot-quantiles.gpi:
* performance-tests/Bench/bin/plot-throughput-testformats.gpi:
* performance-tests/Bench/bin/plot-throughput-transports.gpi:
* performance-tests/Bench/bin/plot-transports-scaling.gpi:
* performance-tests/Bench/bin/plot-transports.gpi:
* performance-tests/Bench/tools/plot-latency.gpi:
* performance-tests/Bench/tools/plot-scaling.gpi:
* performance-tests/Bench/tools/plot-throughput.gpi:
Moved files to accepted GNUPlot file extension name.
* performance-tests/Bench/bin/plot-test-results.sh:
Fixed references to the files.
* performance-tests/Bench/bin/lj-plots.gp:
* performance-tests/Bench/bin/plot-density.gp:
* performance-tests/Bench/bin/plot-jitter.gp:
* performance-tests/Bench/bin/plot-quantiles.gp:
* performance-tests/Bench/bin/plot-throughput-testformats.gp:
* performance-tests/Bench/bin/plot-throughput-transports.gp:
* performance-tests/Bench/bin/plot-transports-scaling.gp:
* performance-tests/Bench/bin/plot-transports.gp:
* performance-tests/Bench/tools/plot-latency.gp:
* performance-tests/Bench/tools/plot-scaling.gp:
* performance-tests/Bench/tools/plot-throughput.gp:
Removed these files.
Mon May 10 06:37:48 UTC 2010 Trevor Fields <[email protected]>
* performance-tests/Bench/bin/plot-throughput-transports.gp:
Adjusted name of mulitcast plot files to be similar to latency
and to fit the existing html.
Fri May 07 22:59:20 UTC 2010 Trevor Fields <[email protected]>
* performance-tests/Bench/bin/generate-perf-html.sh:
Corrected throughput graph title and added pub and sub bounded graphs.
Fri May 07 22:48:44 UTC 2010 Trevor Fields <[email protected]>
* performance-tests/Bench/bin/plot-test-results.sh:
* performance-tests/Bench/bin/plot-throughput-transports.gp:
Corrected graph mapping and changed to graph with simple lines by
default and try to graph with smooth on a second pass to ensure
that graphs are drawn even when one set of data is missing.
Fri May 7 16:03:44 UTC 2010 Yan Dai <[email protected]>
* dds/DCPS/transport/framework/PacketRemoveVisitor.cpp:
* dds/DCPS/transport/framework/QueueRemoveVisitor.cpp:
* dds/DCPS/transport/framework/RemoveAllVisitor.cpp:
Reverted changes made in last commit and added comments
for dropped_by_transport flag used. These changes in last
commit do not affect the duplicate message issue.
Fri May 7 01:40:13 UTC 2010 Yan Dai <[email protected]>
* dds/DCPS/transport/framework/TransportQueueElement.inl:
Fixed a bug that incorrectly callback
data_dropped(dropped_by_transport=false) when one link
drops sample and other links deliver successfully. It was
corrected to call data_dropped(true). This fixed duplicate
message issue caused by a lost link.
* dds/DCPS/transport/framework/PacketRemoveVisitor.cpp:
* dds/DCPS/transport/framework/QueueRemoveVisitor.cpp:
Corrected dropped_by_transport flag passed to data_dropped
callback.
Thu May 06 23:28:13 UTC 2010 Trevor Fields <[email protected]>
* performance-tests/Bench/tests/thru/run_test.pl:
Added remove of repo ior file before spawning.
Wed May 05 23:16:41 UTC 2010 Trevor Fields <[email protected]>
* performance-tests/Bench/bin/gen-latency-tables.pl:
* performance-tests/Bench/bin/gen-throughput-tables.pl:
Corrected transport name recognition.
Wed May 05 20:15:51 UTC 2010 Trevor Fields <[email protected]>
* performance-tests/Bench/bin/generate-perf-html.sh:
Corrected order of commands to generate tables _after_ the data
was processed. Added in missing test run information.
Wed May 5 17:14:34 UTC 2010 Yan Dai <[email protected]>
* dds/DCPS/DataWriterImpl.cpp:
Moved the publisher->remove_associations call out of DataWriter
locking scope to fix a deadlock issue due to reverse locking with
DataWriter lock and TransportImpl lock when remove/add associations
are handled concurrently.
* dds/DCPS/transport/framework/DataLink.cpp:
Added locking for pub_map_.release_subscriber to make pub_map_
update synchronized. This is a potential cause for a SEGV due
to ReceiveListenerSet destructed during remove_associations.
Tue May 05 06:31:35 UTC 2010 Trevor Fields <[email protected]>
* performance-tests/Bench/tests/thru/run_test.pl:
Corrected typo that loaded incorrect configuration for multicast
best effort.
Tue May 04 23:31:01 UTC 2010 Trevor Fields <[email protected]>
* performance-tests/Bench/tests/thru/run_test.pl:
Adjusted so the run_test.pl timeout is longer than the run_test
timeout. This should stop stranding testprocesses waiting on the
multicast transports.
Tue May 04 20:42:29 UTC 2010 Trevor Fields <[email protected]>
* performance-tests/Bench/bin/plot-density.gp:
* performance-tests/Bench/bin/plot-quantiles.gp:
* performance-tests/Bench/bin/plot-test-results.sh:
* performance-tests/Bench/bin/generate-perf-html.sh:
* performance-tests/Bench/bin/generate-test-results.sh:
Adding scripts to generate html files from results.
* performance-tests/Bench/tests/latency/run_test.pl:
Corrected name of the multicast best effort output directory.
* performance-tests/Bench/tests/thru/run_test.pl:
Removed extra time added for the client side.
* performance-tests/Bench/tools/plot-latency.gp:
Changed order so gnuplot 4.4 plot is called last.
Sun May 02 06:00:41 UTC 2010 Trevor Fields <[email protected]>
* performance-tests/Bench/bin/gen-run-info-table.pl:
* performance-tests/Bench/bin/generate-perf-html.sh:
Fixed typos and adjusted archive to not fail on missing directory.
Sat May 01 00:59:12 UTC 2010 Trevor Fields <[email protected]>
* performance-tests/Bench/bin/gen-latency-tables.pl:
* performance-tests/Bench/bin/gen-run-info-table.pl:
* performance-tests/Bench/bin/gen-throughput-tables.pl:
* performance-tests/Bench/bin/generate-perf-html.sh:
* performance-tests/Bench/tools/formatting.css:
Adding scripts for generating html pages from results.
* performance-tests/Bench/bin/extract-throughput.pl:
* performance-tests/Bench/bin/plot-throughput-testformats.gp:
* performance-tests/Bench/bin/plot-throughput-transports.gp:
Changed the output of extract-throughput.pl to use commas and added
test type and transport fields.
Wed Apr 28 21:15:01 UTC 2010 Mike Martinez <[email protected]>
* performance-tests/Bench/bin/gen-latency-stats.pl:
Use zero-based array indices correctly.
* performance-tests/Bench/bin/plot-density.gp:
Removed incorrect hop adjustment.
* performance-tests/Bench/bin/plot-jitter.gp:
* performance-tests/Bench/bin/plot-transports.gp:
Migrated to robust statistics.
* performance-tests/Bench/bin/plot-quantiles.gp:
Removed incorrect hop adjustment.
* performance-tests/Bench/bin/reduce-latency-data.pl:
Corrected a minor formatting annoyance.
* performance-tests/Bench/tools/plot-latency.gp:
Added kernel density and quantile plots to generated results.
Wed Apr 28 19:50:16 UTC 2010 Mike Martinez <[email protected]>
* performance-tests/Bench/bin/extract-latency.pl:
Updated documentation to be correct after adding the Median and
Median absolute deviation statistical data.
* performance-tests/Bench/bin/gen-latency-stats.pl:
Updated generated GNUPlot statistical legend data to contain
the Median and Median absolute deviation instead of the Mean
and Standard Deviation. This is part of the move towards use
of robust statistics in the data reduction and visualization.
* performance-tests/Bench/bin/run_test:
Updated the 'netstat' command arguments to provide all inet
information instead of just active TCP connections.
Wed Apr 28 00:21:25 UTC 2010 Trevor Fields <[email protected]>
* performance-tests/Bench/tests/latency/run_test.pl:
* performance-tests/Bench/tests/thru/run_test.pl:
Adjusted running time handling.
Tue Apr 27 22:53:33 UTC 2010 Mike Martinez <[email protected]>
* performance-tests/Bench/bin/run_test:
Changed statistics command termination from SIGQUIT (which
produces a core file) to SIGINT (which does not).
Updated POD documentation.
* performance-tests/Bench/doc/userguide.html:
Updated documentation.
Tue Apr 27 18:39:41 UTC 2010 Mike Martinez <[email protected]>
* performance-tests/Bench/bin/reduce-latency-data.pl:
Simplified the MAD determination. Added a Pearson 2 parameter
estimation of skewness for investigative purposes.
* performance-tests/Bench/bin/run_test:
* performance-tests/Bench/doc/userguide.html:
Incorporated some more of Brian's comments. Updated run_test
man page with new statistics output filename decorator
specification feature documentation.
* performance-tests/Bench/src/Process.cpp:
Extended signal handling to be a bit more robust. Control can
now fall out of the publication wait as well as allowing the
subscription wait to be gracefully terminated as well.
Mon Apr 26 20:33:01 UTC 2010 Scott Harris <[email protected]>
* dds/DCPS/transport/multicast/MulticastConfiguration.cpp:
* dds/DCPS/transport/multicast/MulticastConfiguration.h:
* dds/DCPS/transport/multicast/MulticastDataLink.cpp:
* tests/DCPS/Messenger/pub_multicast.ini:
* docs/html/multicast/multicast.html:
Added support to set TTL for multicast transport.
Sun Apr 26 00:35:23 UTC 2010 Trevor Fields <[email protected]>
* performance-tests/Bench/tests/thru/run_test.pl:
Changed to add time to the run time of the client side to compensate
for server side delay in starting repo.
Sat Apr 24 00:00:38 UTC 2010 Trevor Fields <[email protected]>
* performance-tests/Bench/bin/plot-throughput-testformats.gp:
* performance-tests/Bench/bin/plot-throughput-transports.gp:
* performance-tests/Bench/bin/plot-throughput.gp:
Removed plot-throughput.gp because all because graphs are made through
the more specific plots.
* performance-tests/Bench/doc/userguide.html:
Updated information about changed files.
* performance-tests/Bench/tests/latency/test_list.txt:
* performance-tests/Bench/tests/thru/test_list.txt:
Swapped the client and server for castor and pollux.
Fri Apr 23 20:39:23 UTC 2010 Yan Dai <[email protected]>
* dds/DCPS/transport/framework/CopyChainVisitor.h:
* dds/DCPS/transport/framework/CopyChainVisitor.inl:
* dds/DCPS/transport/framework/CopyChainVisitor.cpp:
* dds/DCPS/transport/framework/PacketRemoveVisitor.cpp:
* dds/DCPS/transport/framework/QueueRemoveVisitor.cpp:
* dds/DCPS/transport/framework/TransportQueueElement.h:
* dds/DCPS/transport/framework/TransportQueueElement.inl:
* dds/DCPS/transport/framework/TransportQueueElement.cpp:
* dds/DCPS/transport/framework/TransportReplacedElement.h:
* dds/DCPS/transport/framework/TransportReplacedElement.inl:
* dds/DCPS/transport/framework/TransportReplacedElement.cpp:
* dds/DCPS/transport/framework/TransportRetainedElement.h:
* dds/DCPS/transport/framework/TransportRetainedElement.inl:
* dds/DCPS/transport/framework/TransportSendBuffer.h:
* dds/DCPS/transport/framework/TransportSendBuffer.cpp:
* dds/DCPS/transport/framework/TransportSendStrategy.cpp:
- Made TransportSendBuffer clone the transport element and
packet with its own MessageBlock and DataBlock cached
allocators and heap allocation for the sample data.
- Fixed a potential issue in TransportSendBuffer due to buffers_
not mutex protected. The bug may cause invalid iterator(or
corruption) while iterating sequence buffers for resending
from reactor thread and the insert calls to release aged
buffer due to capacity limit from another thread (e.g. write
thread).
- Changed TransportQueueElement::data_dropped and
TransportQueueElement::data_delivered to have return value
to indicate if the element is released. This would fix
memory corruption due to access the release flag after
TransportQueueElement is released.
- Moved clone() from TransportReplacedElement to
TransportQueueElement so the TransportRetainElement
can use it.
Fri Apr 23 14:34:11 UTC 2010 Brian Johnson <[email protected]>
* performance-tests/Bench/doc/userguide.html:
minor edits
Thu Apr 22 22:55:09 UTC 2010 Trevor Fields <[email protected]>
* performance-tests/Bench/bin/genstats.pl:
* performance-tests/Bench/bin/gen-latency-stats.pl:
* performance-tests/Bench/tools/convert-latency:
Renamed latency stat generating script. Changed stat output file
names to include latency prefix.
Thu Apr 22 20:46:25 UTC 2010 Trevor Fields <[email protected]>
* bin/performance_tests.lst:
* performance-tests/Bench/tests/latency/run_test.pl:
* performance-tests/Bench/tests/thru/run_test.pl:
Added running of throughput tests and corrected directory names.
Thu Apr 22 18:16:27 UTC 2010 Mike Martinez <[email protected]>
* performance-tests/Bench/bin/extract-latency.pl:
* performance-tests/Bench/bin/reduce-latency-data.pl:
Added median and median absolute deviation (MAD) statistics
gathering.
* performance-tests/Bench/bin/run_test:
Factored the filtering closures out and made them selectable by
OS type.
Wed Apr 21 19:25:35 UTC 2010 Mike Martinez <[email protected]>
* performance-tests/Bench/src/testprocess.cpp:
Added static initialization feature for static builds.
Mon Apr 19 23:50:09 UTC 2010 Trevor Fields <[email protected]>
* performance-tests/Bench/tests/latency/run_test.pl:
* performance-tests/Bench/tests/thru/run_test.pl:
Adjusted to have the run_test on the server side handle the repo.
Running the repo as a separate run_test was leaving cases of the repo
still running after terminating the run_test.
Mon Apr 19 19:19:09 UTC 2010 Mike Martinez <[email protected]>
* performance-tests/Bench/bin/extract-throughput.pl:
Adjusted to produce a single data file with the test and
transport data partitioned into separate indices.
* performance-tests/Bench/bin/plot-density.gp:
* performance-tests/Bench/bin/plot-quantiles.gp:
Candidate plotting commands for Quantile and Kernel Density
Estimation charts. These currently assume specific file names
and locations.
* performance-tests/Bench/tests/thru/test-commands.txt:
Updated to generate output files useful for the current
throughput extraction script.
Mon Apr 19 05:29:23 UTC 2010 Yan Dai <[email protected]>
* dds/DCPS/transport/framework/PacketRemoveVisitor.h:
* dds/DCPS/transport/framework/PacketRemoveVisitor.cpp:
* dds/DCPS/transport/framework/TransportReplacedElement.h:
* dds/DCPS/transport/framework/TransportReplacedElement.inl:
* dds/DCPS/transport/framework/TransportReplacedElement.cpp:
* dds/DCPS/transport/framework/TransportSendBuffer.h:
* dds/DCPS/transport/framework/TransportSendBuffer.cpp:
* dds/DCPS/transport/framework/TransportSendStrategy.h:
* dds/DCPS/transport/framework/TransportSendStrategy.cpp:
Fixed bugs that TransportReplacedElement clones samples owned
by DataWriter using the cached allocators from DataWriter.This
causes SEGV when releasing TransportReplacedElement after
DataWrite is deleted.
Tue Apr 13 00:06:06 UTC 2010 Trevor Fields <[email protected]>
* performance-tests/Bench/tools/mkpkg:
Added files for running scripted tests in packaged environment.
* performance-tests/Bench/bin/extract-throughput.pl:
* performance-tests/Bench/bin/plot-throughput-testformats.gp:
* performance-tests/Bench/bin/plot-throughput-transports.gp:
* performance-tests/Bench/tools/plot-throughput.gp:
Added more graphing scripts for throughput results.
Fri Apr 09 21:27:04 UTC 2010 Trevor Fields <[email protected]>
* performance-tests/Bench/tests/latency/run_test.pl:
* performance-tests/Bench/tests/thru/run_test.pl:
Fixed repo option to Info Repo.
Fri Apr 09 18:37:50 UTC 2010 Trevor Fields <[email protected]>
* performance-tests/Bench/tests/latency/run_test.pl:
* performance-tests/Bench/tests/thru/run_test.pl:
Added some basic POD documentation.
Thu Apr 8 19:17:27 UTC 2010 Mike Martinez <[email protected]>
* performance-tests/Bench/bin/reduce-latency-data.pl:
Added plot data for quantile charts to the output data.
Tue Apr 06 18:45:55 UTC 2010 Trevor Fields <[email protected]>
* bin/PerlDDS/Cross_Sync_Common.pm:
Fix typo and separated output of failing to read file and
failing to find the required instance.
* performance-tests/Bench/bin/run_test:
Improved handling of the Info Repo.
* performance-tests/Bench/tests/latency/run_test.pl:
* performance-tests/Bench/tests/thru/run_test.pl:
Fixed issues with running.
Thu Apr 01 22:36:25 UTC 2010 Trevor Fields <[email protected]>
* performance-tests/Bench/bin/extract-throughput.pl:
* performance-tests/Bench/tests/latency/run_test.pl:
* performance-tests/Bench/tests/thru/run_test.pl:
* performance-tests/Bench/tests/thru/test-commands.txt:
* performance-tests/Bench/tests/thru/test_list.txt:
Minor fixes to run tests and data extraction script.
Thu Apr 1 15:20:09 UTC 2010 Mike Martinez <[email protected]>
* performance-tests/Bench/bin/run_test:
Explicitly search the environment PATH variable for the
repository and test commands to execute.
Wed Mar 31 23:59:23 UTC 2010 Trevor Fields <[email protected]>
* bin/performance_tests.lst:
* performance-tests/Bench/bin/extract-latency.pl:
* performance-tests/Bench/bin/extract-throughput.pl:
* performance-tests/Bench/bin/plot-throughput.gp:
* performance-tests/Bench/bin/reduce-latency-data.pl:
* performance-tests/Bench/tests/latency/run_test.pl:
* performance-tests/Bench/tests/latency/test_list.txt:
* performance-tests/Bench/tests/thru/test-commands.txt:
* performance-tests/Bench/tools/convert-all:
* performance-tests/Bench/tools/convert-latency:
* performance-tests/Bench/tools/plot-latency.gp:
* performance-tests/Bench/tools/plot-throughput.gp:
* performance-tests/Bench/bin/reduce.pl:
* performance-tests/Bench/bin/extract.pl:
* performance-tests/Bench/tools/plot-all.gp:
Added initial scripts for plotting throughput data and renamed
the genericly name latency scripts.
Wed Mar 31 19:51:26 UTC 2010 Mike Martinez <[email protected]>
* performance-tests/Bench/bin/run_test:
Prepend the script directory to the command search path and
then use that path to find commands. This allows installations
other than the 'mkpkg' ones to work as well.
Wed Mar 31 19:02:50 UTC 2010 Mike Martinez <[email protected]>
* performance-tests/Bench/bin/run_test:
Previous debug file specification change is now extended to the
testprocess commands as well.
* performance-tests/Bench/doc/userguide.html:
* performance-tests/Bench/tests/latency/latency.mm:
Adding and updating documentation.
Wed Mar 31 13:55:53 UTC 2010 Mike Martinez <[email protected]>
* performance-tests/Bench/bin/run_test:
Allow debug file to be specified in the absence of a debug
level. This allows the verbose test output to reach a logfile
instead of the standard output.
Tue Mar 30 22:39:41 UTC 2010 Mike Martinez <[email protected]>
* performance-tests/Bench/bin/run_test: