-
Notifications
You must be signed in to change notification settings - Fork 7
/
ChangeLog
14812 lines (9419 loc) · 495 KB
/
ChangeLog
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
Tue Oct 29 12:08:05 2013 Tanaka Akira <akr@fsij.org>
* ext/readline/readline.c (readline_getc): Consider
NULL as input.
Tue Oct 29 11:10:08 2013 Aman Gupta <ruby@tmm1.net>
* gc.c (gc_profile_total_time): fix off-by-one error in
GC::Profiler.total_time.
* test/ruby/test_gc.rb (class TestGc): test for above.
Tue Oct 29 09:53:00 2013 Charlie Somerville <charliesome@ruby-lang.org>
* insns.def, vm.c, vm_insnhelper.c, vm_insnhelper.h, vm_method.c: split
ruby_vm_global_state_version into two separate counters - one for the
global method state and one for the global constant state. This means
changes to constants do not affect method caches, and changes to
methods do not affect constant caches. In particular, this means
inclusions of modules containing constants no longer globally
invalidate the method cache.
* class.c, eval.c, include/ruby/intern.h, insns.def, vm.c, vm_method.c:
rename rb_clear_cache_by_class to rb_clear_method_cache_by_class
* class.c, include/ruby/intern.h, variable.c, vm_method.c: add
rb_clear_constant_cache
* compile.c, vm_core.h, vm_insnhelper.c: rename vmstat field in
rb_call_info_struct to method_state
* vm_method.c: rename vmstat field in struct cache_entry to method_state
Mon Oct 28 23:26:04 2013 Tanaka Akira <akr@fsij.org>
* test/readline/test_readline.rb (teardown): Clear Readline.input and
Readline.output.
Mon Oct 28 21:35:31 2013 Tanaka Akira <akr@fsij.org>
* ext/-test-/file/depend, ext/-test-/postponed_job/depend,
ext/-test-/tracepoint/depend: New files for dependencies.
Mon Oct 28 15:32:18 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
* ext/openssl/depend (ossl.o): work around of dependency of
thread_native.h, which depends on headers by THREAD_MODEL.
[ruby-dev:47777]
* ext/openssl/extconf.rb: need THREAD_MODEL.
Mon Oct 28 14:57:01 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
* load.c (ruby_init_ext): share feature names between frame name and
provided features.
Mon Oct 28 14:41:27 2013 Akinori MUSHA <knu@iDaemons.org>
* misc/ruby-electric.el: Import ruby-electric.el 2.1 from
https://github.com/knu/ruby-electric.el.
* Hitting the newline-and-indent key within a comment fires
comment-indent-new-line.
* Introduce a new feature
`ruby-electric-autoindent-on-closing-char`.
* Fix fallback behavior of ruby-electric-space/return that
caused error with auto-complete.
Mon Oct 28 13:17:17 2013 Or Cohen <orc@fewbytes.com>
* error.c (name_err_to_s): remove no longer needed overriding, since
r30455 which made exc_to_s almost same. Fixes [GH-413].
Mon Oct 28 12:42:11 2013 Tanaka Akira <akr@fsij.org>
* common.mk, ext/objspace/depend, ext/coverage/depend,
ext/-test-/debug/depend, ext/date/depend: Update dependencies.
Mon Oct 28 09:29:00 2013 Charlie Somerville <charliesome@ruby-lang.org>
* vm.c: vm_clear_all_cache is not necessary now we use a 64 bit counter
for global state version.
* vm_insnhelper.h: ruby_vm_global_state_version overflow is unnecessary
Mon Oct 28 07:47:32 2013 Aman Gupta <ruby@tmm1.net>
* vm_backtrace.c (rb_profile_frame_classpath): do not use rb_inspect
directly, since it might have a custom implementation or show ivars.
Mon Oct 28 04:10:41 2013 Aman Gupta <ruby@tmm1.net>
* vm_backtrace.c (rb_profile_frame_classpath): handle singleton
methods defined directly on an object.
* test/-ext-/debug/test_profile_frames.rb: test for above.
Mon Oct 28 00:52:36 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
* struct.c (new_struct): fix warning message, class name and encoding.
Sun Oct 27 20:53:08 2013 Tanaka Akira <akr@fsij.org>
* ext/readline/readline.c: Include ruby/thread.h for
rb_thread_call_without_gvl2.
(readline_rl_instream, readline_rl_outstream): Record FILE
structures allocated by this extension.
(getc_body): New function extracted from readline_getc.
(getc_func): New function.
(readline_getc): Use rb_thread_call_without_gvl2 to invoke getc_func.
[ruby-dev:47033] [Bug #8749]
(clear_rl_instream, clear_rl_outstream): Close FILE structure
allocated by this extension reliably. [ruby-core:57951] [Bug #9040]
(readline_readline): Use clear_rl_instream and clear_rl_outstream.
(readline_s_set_input): Set readline_rl_instream.
(readline_s_set_output): Set readline_rl_outstream.
(Init_readline): Don't call readline_s_set_input because
readline_getc doesn't block other threads for any FILE structure now.
[ruby-dev:47033] [Bug #8749] reported by Nobuhiro IMAI.
[ruby-core:57951] [Bug #9040] reported by Eamonn Webster.
Sat Oct 26 19:31:28 2013 Kazuki Tsujimoto <kazuki@callcc.net>
* gc.c: catch up recent changes to compile on GC_DEBUG,
RGENGC_CHECK_MODE.
Sat Oct 26 19:08:00 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
* range.c (range_initialize_copy): disallow to modify after
initialized.
Sat Oct 26 17:48:54 2013 Tanaka Akira <akr@fsij.org>
* lib/open-uri.rb (meta_add_field): : Re-implemented.
[ruby-core:58017] [Bug #9051] patch by Eamonn Webster.
Sat Oct 26 14:35:09 2013 Koichi Sasada <ko1@atdot.net>
* gc.c (gc_profile_dump_on): use "Page" terminology.
Sat Oct 26 13:25:45 2013 Koichi Sasada <ko1@atdot.net>
* gc.c (gc_sweep, gc_heap_lazy_sweep): fix measurement code.
We only need one sweep time measurement without lazy sweep.
Sat Oct 26 11:59:13 2013 Tanaka Akira <akr@fsij.org>
* addr2line.c: Include ELF header after system headers (especially
sys/types.h) to avoid compilation failure,
"usr/include/sh3/elf_machdep.h:4:2: error: #error Define _BYTE_ORDER!",
on NetBSD/sh3 (dreamcast, hpcsh, landisk, mmeye).
Sat Oct 26 11:35:22 2013 Koichi Sasada <ko1@atdot.net>
* gc.c: tuning parameters.
* gc.c (GC_MALLOC_LIMIT): change default value to 16MB.
* gc.c (GC_MALLOC_LIMIT_GROWTH_FACTOR): change default value to 2.0.
* gc.c (gc_before_sweep): change decrease ratio of `malloc_limit'
from 1/4 to 1/10.
Sat Oct 26 11:30:07 2013 Koichi Sasada <ko1@atdot.net>
* gc.c (vm_malloc_increase): do gc_rest_sweep() before GC.
gc_rest_sweep() can reduce malloc_increase, so try it before GC.
Otherwise, malloc_increase can be less than malloc_limit at
gc_before_sweep(). This means that re-calculation of malloc_limit
may be wrong value.
Sat Oct 26 06:35:41 2013 Masaya Tarui <tarui@ruby-lang.org>
* gc.c (gc_before_heap_sweep): Restructure code to mean clearly.
heap->freelist is connected to end of list.
Sat Oct 26 04:01:35 2013 Koichi Sasada <ko1@atdot.net>
* gc.c (gc_before_heap_sweep): fix freelist management.
After rb_gc_force_recycle() for a object belonging to heap->freelist,
`heap->using_page->freelist' is not null.
Thu Oct 24 21:57:24 2013 Marc-Andre Lafortune <ruby-core@marc-andre.ca>
* parse.y: Remove +(binary) and -(binary) special cases
[Feature #9048]
Thu Oct 24 12:45:53 2013 Zachary Scott <e@zzak.io>
* object.c: [DOC] Document first argument also takes string for:
rb_mod_const_get, rb_mod_const_set, rb_mod_const_defined
Also added note about NameError exception for invalid constant name
Thu Oct 24 12:23:58 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
* thread.c (rb_thread_terminate_all): add a comment why we need
state check and call terminate_i again.
Thu Oct 24 12:15:02 2013 KOSAKI Motohiro <kosaki.motohiro@gmail.com>
* thread.c (rb_thread_terminate_all): add a comment why infinite
sleep is safe.
Thu Oct 24 07:41:42 2013 Aman Gupta <ruby@tmm1.net>
* gc.c: add new initial_growth_max tuning parameter.
[ruby-core:57928] [Bug #9035]
* gc.c (heap_set_increment): when initial_growth_max is set,
do not grow number of slots by more than growth_max at a time.
* gc.c (rb_gc_set_params): load optional new tuning value from
RUBY_HEAP_SLOTS_GROWTH_MAX environment variable.
* test/ruby/test_gc.rb (class TestGc): test for above.
Thu Oct 24 01:34:12 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
* include/ruby/win32.h (rb_infinity_float): suppress overflow in
constant arithmetic warnings. [ruby-core:57981] [Bug #9044]
Thu Oct 24 00:11:24 2013 Marc-Andre Lafortune <ruby-core@marc-andre.ca>
* lib/ostruct.rb: raise NoMethodError with a #name and #args.
Raise RuntimeError when modifying frozen instances
instead of TypeError.
(OpenStruct#each_pair): Return an enumerator with size
(OpenStruct#delete): Use the converted argument.
Patches by Kenichi Kamiya. [Fixes GH-383]
* test/ostruct/test_ostruct.rb: Added tests for above.
Thu Oct 24 00:10:22 2013 Marc-Andre Lafortune <ruby-core@marc-andre.ca>
* array.c: Add Array#to_h [Feature #7292]
* enum.c: Add Enumerable#to_h
Wed Oct 23 23:48:28 2013 Aman Gupta <ruby@tmm1.net>
* gc.c: Rename free_min to min_free_slots and free_min_page to
max_free_slots. The algorithm for heap growth is:
if (swept_slots < min_free_slots) pages++
if (swept_slots > max_free_slots) pages--
Wed Oct 23 22:51:03 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
* win32/Makefile.sub (config.h): VC 2013 supports C99 mathematics
functions. [ruby-core:57981] [Bug #9044]
Wed Oct 23 19:13:18 2013 Koichi Sasada <ko1@atdot.net>
* gc.c: move increment from heap to heap_pages.
Share `increment' information with heaps.
* gc.c: change ratio of heap_pages_free_min_page
to 0.80.
This change means slow down page freeing speed.
Wed Oct 23 17:52:03 2013 Koichi Sasada <ko1@atdot.net>
* gc.c (heap_pages_free_unused_pages): cast to (int) for size_t
variable `i'.
Wed Oct 23 17:39:35 2013 Koichi Sasada <ko1@atdot.net>
* gc.c: introduce tomb heap.
Tomb heap is where zombie objects and ghost (freed slot) lived in.
Separate from other heaps (now there is only eden heap) at sweeping
helps freeing pages more efficiently.
Before this patch, even if there is an empty page at former phase
of sweeping, we can't free it.
Algorithm:
(1) Sweeping all pages in a heap and move empty pages from the
heap to tomb_heap.
(2) Check all existing pages and free a page
if all slots of this page are empty and
there is enough empty slots (checking by swept_num)
To introduce this patch, there are several tuning of GC parameters.
Wed Oct 23 14:20:56 2013 Koichi Sasada <ko1@atdot.net>
* gc.c (gc_prof_sweep_timer_stop): catch up recent changes
to compile on GC_PROFILE_MORE_DETAIL=1.
Wed Oct 23 11:43:27 2013 Zachary Scott <e@zzak.io>
* file.c: [DOC] fix rdoc format of File#expand_path from r43386
Tue Oct 22 21:58:28 2013 URABE Shyouhei <shyouhei@ruby-lang.org>
* vm_core.h (enum): avoid syntax error.
* method.h: ditto.
* internal.h: ditto.
Tue Oct 22 19:53:16 2013 Koichi Sasada <ko1@atdot.net>
* gc.c (Init_heap): move logics from heap_pages_init() and remove
heap_pages_init().
Tue Oct 22 19:19:05 2013 Koichi Sasada <ko1@atdot.net>
* gc.c: allow multiple heaps.
Now, objects are managed by page. And a set of pages is called heap.
This commit supports multiple heaps in the object space.
* Functions heap_* and rb_heap_t manages heap data structure.
* Functions heap_page_* and struct heap_page manage page data
structure.
* Functions heap_pages_* and struct rb_objspace_t::heap_pages
maintains all pages.
For example, pages are allocated from the heap_pages.
See https://bugs.ruby-lang.org/projects/ruby-trunk/wiki/GC_design
and https://bugs.ruby-lang.org/attachments/4015/data-heap_structure_with_multiple_heaps.png
for more details.
Now, there is only one heap called `eden', which is a space for all
new generated objects.
Tue Oct 22 18:26:12 2013 Tanaka Akira <akr@fsij.org>
* lib/pp.rb (object_address_group): Use Kernel#to_s to obtain the class
name and object address.
This fix a problem caused by %p in C generates variable length
address.
Reported by ko1 via IRC.
Tue Oct 22 16:57:48 2013 Benoit Daloze <eregontp@gmail.com>
* file.c (File#expand_path): [DOC] improve documentation of File#expand_path.
Based on patch by Prathamesh Sonpatki. [ruby-core:57734] [Bug #9002]
Tue Oct 22 15:59:51 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
* dir.c (glob_helper): don't skip current directories if FNM_DOTMATCH
is given. [ruby-core:53108] [Bug #8006]
Tue Oct 22 14:53:11 2013 Koichi Sasada <ko1@atdot.net>
* vm_trace.c: exterminate Zombies.
There is a bug that T_ZOMBIE objects are not collected.
Because there is a pass to miss finalizer postponed job
with multi-threading. This patch solve this issue.
* vm_trace.c (rb_postponed_job_register_one): set
RUBY_VM_SET_POSTPONED_JOB_INTERRUPT(th) if another same job
is registered.
There is a possibility to remain a postponed job without
interrupt flag.
* vm_trace.c (rb_postponed_job_register_one): check interrupt
carefully.
* vm_trace.c (rb_postponed_job_register_one): use additional space
to avoid buffer full.
* gc.c (gc_finalize_deferred_register): check failure.
* thread.c (rb_threadptr_execute_interrupts): check
`postponed_job_interrupt' immediately. There is a possibility
to miss this flag.
Tue Oct 22 12:11:16 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
* configure.in: check if the given CFLAGS and LDFLAGS are working, and
bail out early if not.
Tue Oct 22 00:06:57 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
* file.c (rb_file_exists_p): warn deprecated name. [Bug #9041]
Mon Oct 21 23:57:53 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
* encoding.c (load_encoding): should preserve outer errinfo, so that
expected exception may not be lost. [ruby-core:57949] [Bug #9038]
Sun Oct 20 15:41:22 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
* io.c (rb_io_reopen): create a new, temporary FD via rb_sysopen and
call rb_cloexec_dup2 on it to atomically replace the file fptr->fd
points to. This leaves no possible window where fptr->fd is invalid
to userspace (even for any threads running w/o GVL). based on the
patch by Eric Wong <normalperson@yhbt.net> at [ruby-core:57943].
[Bug #9036]
Sun Oct 20 15:29:05 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
* error.c (rb_syserr_fail_path_in): new function split from
rb_sys_fail_path_in to raise SystemCallError without errno.
* internal.h (rb_syserr_fail_path): like rb_sys_fail_path but without
errno.
Sun Oct 20 13:58:47 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
* include/ruby/ruby.h (rb_obj_wb_unprotect, rb_obj_written),
(rb_obj_write): suppress unused-parameter warnings.
Sun Oct 20 10:32:48 2013 Eric Hodel <drbrain@segment7.net>
* lib/rubygems: Update RubyGems to master 0886307. This commit
improves documentation and should bring ruby above 75% documented on
rubyci.
Sun Oct 20 09:30:56 2013 Eric Hodel <drbrain@segment7.net>
* lib/rubygems: Update to RubyGems master 3de7e0f. Changes:
Only attempt to build extensions for newly-installed gems. This
prevents compilation attempts at gem activation time for gems that
already have extensions built.
Fix crash in the dependency resolver for dependencies that cannot be
resolved.
* test/rubygems: ditto.
Sun Oct 20 05:24:29 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
* variable.c (rb_class2name): should return real class name, not
singleton class or iclass.
Sun Oct 20 04:18:48 2013 Aman Gupta <ruby@tmm1.net>
* variable.c (rb_class2name): call rb_tmp_class_path() directly to
avoid extra rb_str_dup() from rb_class_name().
Sat Oct 19 19:59:02 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
* win32/file.c (code_page): use simple array instead of st_table.
* encoding.c (rb_locale_encindex): defer initialization of win32 code
page table until encoding db loaded.
Sat Oct 19 08:25:05 2013 Koichi Sasada <ko1@atdot.net>
* gc.c: fix rb_objspace_t.
* make "struct heap" and move most of variables
in rb_objspace_t::heap.
* rename rb_objspace_t::heap::sorted to
rb_objspace_t::heap_sorted_pages
and make a macro heap_sorted_pages.
* rename rb_objspace_t::heap::range to
rb_objspace_t::heap_range and rename macros
lomem/himem to heap_lomem/heap_himem.
Sat Oct 19 07:14:40 2013 Eric Hodel <drbrain@segment7.net>
* lib/rubygems: Update to RubyGems master 42543b6. Changes:
Fix `gem update` for gems with multiple platforms.
* test/rubygems: ditto.
Sat Oct 19 06:55:52 2013 Eric Hodel <drbrain@segment7.net>
* lib/rubygems: Update to RubyGems master 0a3814b. Changes:
Fixed extension directory in Gem::Specification#require_paths.
Allow installation of gems when $HOME is nonexistent or unwritable.
Use proper API in InstallCommand.
Improve support for path option in gem dependency files.
Remove warnings.
* test/rubygems: ditto.
Fri Oct 18 15:23:34 2013 Koichi Sasada <ko1@atdot.net>
* gc.c: change terminology of heap.
Change "slot" to "page". "Slot" is a space of RVALUE.
1. "Heap" consists of a set of "heap_page"s (pages).
2. Each "heap_page" has "heap_page_body".
3. "heap_page_body" has RVALUE (a.k.a. "slot") spaces.
4. "sorted" is a sorted array of "heap_page"s, sorted
by address of heap_page_body (for "is_pointer_to_heap").
See https://bugs.ruby-lang.org/attachments/4008/data-heap_structure.png.
Fri Oct 18 09:40:43 2013 Eric Hodel <drbrain@segment7.net>
* lib/rubygems: Update to RubyGems master cee6788. Changes:
Fix test failure on vc10-x64 Server on rubyci.org due to attempting
to File.chmod where it is not supported.
Continuing work on improved gem dependencies file (Gemfile) support.
* test: ditto.
Fri Oct 18 06:02:49 2013 Eric Hodel <drbrain@segment7.net>
* lib/rubygems: Update to RubyGems master f738c67. Changes:
Fixed test bug for ruby with ENABLE_SHARED = no
* test/rubygems: ditto.
Fri Oct 18 00:57:07 2013 Tanaka Akira <akr@fsij.org>
* lib/tsort.rb (TSort.tsort): Extracted from TSort#tsort.
(TSort.tsort_each): Extracted from TSort#tsort_each.
(TSort.strongly_connected_components): Extracted from
TSort#strongly_connected_components.
(TSort.each_strongly_connected_component): Extracted from
TSort#each_strongly_connected_component.
Thu Oct 17 18:50:08 2013 Koichi Sasada <ko1@atdot.net>
* gc.c (CALC_EXACT_MALLOC_SIZE_CHECK_OLD_SIZE): introduced.
This macro enable checker compare with allocated memory and
declared old_size of sized_xfree and sized_xrealloc.
Thu Oct 17 18:45:41 2013 Koichi Sasada <ko1@atdot.net>
* string.c (STR_HEAP_SIZE): includes TERM_LEN(str).
* string.c (rb_str_memsize): use STR_HEAP_SIZE().
Thu Oct 17 17:43:00 2013 Shugo Maeda <shugo@ruby-lang.org>
* vm_insnhelper.c (vm_call_method): set ci->me to 0 when the
original method of a refined method is undef to avoid SEGV.
* vm_method.c (rb_method_entry_without_refinements): return 0 when
the original method of a refined method is undef to avoid SEGV.
* test/ruby/test_refinement.rb: related test.
Thu Oct 17 17:38:36 2013 Koichi Sasada <ko1@atdot.net>
* gc.c, internal.h: rename ruby_xsizefree/realloc to
rb_sized_free/realloc.
* array.c: catch up these changes.
* string.c: ditto.
Thu Oct 17 17:32:51 2013 Koichi Sasada <ko1@atdot.net>
* array.c, string.c: use ruby_xsizedfree() and ruby_xsizedrealloc().
* internal.h (SIZED_REALLOC_N): define a macro as REALLOC_N().
Thu Oct 17 17:11:17 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
* win32/win32.c (console_emulator_p): check by comparison between
module handle of WriteConsoleW and kernel32.dll.
* configure.in, win32/Makefile.sub, win32/setup.mak: no longer need
psapi.lib.
Thu Oct 17 16:53:30 2013 Koichi Sasada <ko1@atdot.net>
* gc.c, internal.h: add new internal memory management functions.
* void *ruby_xsizedrealloc(void *ptr, size_t new_size, size_t old_size)
* void ruby_xsizedfree(void *x, size_t size)
These functions accept additional size parameter to calculate more
accurate malloc_increase parameter which control GC timing.
[Feature #8985]
Thu Oct 17 14:21:34 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
* win32/file.c (rb_file_expand_path_internal): fix memory leaks at
a non-absolute home exception.
Thu Oct 17 14:06:39 2013 Koichi Sasada <ko1@atdot.net>
* ext/objspace/object_tracing.c (newobj_i): fix memory leak.
There is possibility to remain info due to missing FREEOBJ event.
FREEOBJ events are skipped while suppress_tracing state, for example,
during trace events are invoking.
Thu Oct 17 12:30:16 2013 Tanaka Akira <akr@fsij.org>
* lib/tsort.rb (TSort.each_strongly_connected_component_from):
Extracted from TSort#each_strongly_connected_component_from.
Thu Oct 17 11:07:06 2013 Eric Hodel <drbrain@segment7.net>
* lib/rubygems: Update to RubyGems master 941c21a. Changes:
Restored method bundler wants to remove for compatibility.
Improvements to Gemfile compatibility.
* test/rubygems: ditto.
Thu Oct 17 08:08:11 2013 Koichi Sasada <ko1@atdot.net>
* ext/objspace/object_tracing.c (newobj_i): add workaround.
some bugs hits this check.
* ext/objspace/object_tracing.c (object_allocations_reporter_i): cast as pointer.
Thu Oct 17 07:36:53 2013 Eric Hodel <drbrain@segment7.net>
* lib/rubygems: Update to RubyGems master 2abce58. Changes:
Fixed documentation generation when sdoc and json are installed as
gems.
Added some missing documentation.
Thu Oct 17 07:10:26 2013 Zachary Scott <e@zzak.io>
* ext/curses/curses.c: [DOC] Cleaned up formatting consistency of rdoc
comments for Curses, including period spacing and column width.
This patch also fixed some typos. Thanks to @postmodern for the patch!
[Fixes GH-420] https://github.com/ruby/ruby/pull/420
Thu Oct 17 06:58:42 2013 Zachary Scott <e@zzak.io>
* ext/date/date_core.c: [DOC] plural grammar fixed by @scott113341
Contributed via documenting-ruby.org: documenting-ruby/ruby#16
https://github.com/documenting-ruby/ruby/pull/16
Thu Oct 17 05:52:31 2013 Zachary Scott <e@zzak.io>
* ext/io/nonblock/nonblock.c: [DOC] Document io/nonblock by reprah
[Fixes GH-418] https://github.com/ruby/ruby/pull/418 based on the
original discussion from documenting-ruby/ruby#18
Thu Oct 17 05:40:33 2013 Koichi Sasada <ko1@atdot.net>
* gc.c (objspace_each_objects): do not skip empty RVALUEs.
Thu Oct 17 05:31:31 2013 Koichi Sasada <ko1@atdot.net>
* error.c (rb_bug_reporter_add): return simply 0 if failed.
Please check return value.
Thu Oct 17 05:17:33 2013 Koichi Sasada <ko1@atdot.net>
* ext/objspace/object_tracing.c: add new method
ObjectSpace.trace_object_allocations_debug_start for GC debugging.
If you encounter the BUG "... is T_NONE" (and so on) on your
application, please try this method at the beginning of your app.
Wed Oct 16 22:35:27 2013 Zachary Scott <e@zzak.io>
* ext/io/nonblock/nonblock.c: use rb_cIO instead of VALUE
Wed Oct 16 17:45:13 2013 Koichi Sasada <ko1@atdot.net>
* bootstraptest/runner.rb: check nil before calling `signal?'
for a process status.
Wed Oct 16 17:37:17 2013 Koichi Sasada <ko1@atdot.net>
* error.c, internal.h (rb_bug_reporter_add): add a new C-API.
rb_bug_reporter_add() allows to register a function which
is called at rb_bug() called.
* ext/-test-/bug_reporter/bug_reporter.c: add a test for this C-API.
* ext/-test-/bug_reporter/extconf.rb: ditto.
* test/-ext-/bug_reporter/test_bug_reporter.rb: ditto.
Wed Oct 16 15:14:21 2013 Koichi Sasada <ko1@atdot.net>
* NEWS: add a line into NEWS for last commit.
Wed Oct 16 15:09:14 2013 Koichi Sasada <ko1@atdot.net>
* ext/objspace/objspace.c: add a new method `reachable_objects_from_root'.
ObjectSpace.reachable_objects_from_root returns all objects referred
from root (called "root objects").
This feature is for deep object analysis.
* test/objspace/test_objspace.rb: add a test.
Wed Oct 16 15:00:21 2013 Eric Hodel <drbrain@segment7.net>
* lib/rubygems: Update to RubyGems master b955554. Changes:
Fixed NameError for Gem::Ext due to re-entering file lookup in
RubyGems' overridden require. Bug by Koichi Sasada.
Fixed possible circular require warning in tests.
Used existing constant for `gem install -g` dependency file list.
* test/rubygems: ditto.
Wed Oct 16 09:42:42 2013 Eric Hodel <drbrain@segment7.net>
* lib/rubygems: Update to RubyGems master 278d00d. Changes:
Fixes building extensions without a "clean" make rule
Adds gem dependency file autodetection to "gem install -g"
* test/rubygems: Tests for the above.
Wed Oct 16 09:12:23 2013 Eric Hodel <drbrain@segment7.net>
* lib/rubygems: Update to RubyGems master commit 2a74263. This fixes
several bugs in RubyGems 2.2.0.preview.1.
* test/rubygems: ditto.
Wed Oct 16 07:25:02 2013 Aman Gupta <ruby@tmm1.net>
* gc.c (gc_mark_roots): rename roots to be categories
instead of function names.
Tue Oct 15 19:18:13 2013 Koichi Sasada <ko1@atdot.net>
* gc.h (rb_objspace_reachable_objects_from_root): added.
This API provides information which objects are root objects.
`category' shows what kind of root objects.
* gc.c (gc_mark_roots): separate from gc_marks_body().
Tue Oct 15 17:47:59 2013 Tanaka Akira <akr@fsij.org>
* process.c: Fix a typo. MacOS X doesn't have ENOTSUPP.
Mon Oct 14 12:32:52 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
* ruby.c (process_options): load statically linked extensions before
rubygems, because of ext/thread.
* ruby.c (process_options): use gem_prelude instead of requiring
rubygems directly when --enable=gems is given.
* Makefile.in (DEFAULT_PRELUDES): always use gem_prelude regardless of
--disable-rubygems.
Mon Oct 14 11:07:51 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
* lib/mkmf.rb (have_framework): should append framework options to
$LIBS, not $LDFLAGS. The former is propagated to exts.mk when
enable-static-linked-ext.
* lib/mkmf.rb (create_makefile): ranlib on static library, not DLLIB.
Sun Oct 13 23:53:40 2013 Andrew Grimm <andrew.j.grimm@gmail.com>
* vsnprintf.c: Fix spelling from compliment to complement.
Patch by @agrimm.
* include/ruby/ruby.h: ditto
Sun Oct 13 20:59:27 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
* vm.c (Init_BareVM): initialize defined_module_hash here,
Init_top_self() is too late to register core classes/modules.
* compile.c (compile_array_): no hash to merge if it is empty.
* vm.c (m_core_hash_merge_kwd): just check keys if only one argument
is given, without merging.
Sat Oct 12 06:35:01 2013-10-11 Eric Hodel <drbrain@segment7.net>
* lib/rake: Update to rake 10.1.0
* bin/rake: ditto.
* test/rake: ditto.
* NEWS: Update NEWS to include rake 10.1.0 and links to release notes.
Sat Oct 12 03:26:04 2013 Koichi Sasada <ko1@atdot.net>
* class.c, variable.c, gc.c (rb_class_tbl): removed.
* vm.c, vm_core.h (rb_vm_add_root_module): added to register as a
defined root module or class.
This guard helps mark miss from defined classes/modules they are
only referred from C's global variables in C-exts.
Basically, it is extension's bug.
Register to hash object VM has.
Marking a hash objects allows generational GC supports.
* gc.c (RGENGC_PRINT_TICK): disable (revert).
Sat Oct 12 03:24:49 2013 Koichi Sasada <ko1@atdot.net>
* vm_method.c (rb_gc_mark_unlinked_live_method_entries):
revert last commit to introduce debug prints.
Fri Oct 11 21:05:19 2013 Koichi Sasada <ko1@atdot.net>
* internal.h, parse.y: use `full_mark' instead of `full_marking'.
Fri Oct 11 20:58:16 2013 Koichi Sasada <ko1@atdot.net>
* gc.c: use terminology `full_mark' instead of `minor_gc'
in mark functions.
Fri Oct 11 20:46:09 2013 Koichi Sasada <ko1@atdot.net>
* gc.c: use __GNUC__ instead of __GCC__.
Fri Oct 11 20:35:59 2013 Koichi Sasada <ko1@atdot.net>
* gc.c, parse.y: support generational Symbol related marking.
Each symbols has String objects respectively to represent
Symbols.
These objects are marked only when:
* full marking
* new symbols are added
This hack reduce symbols (related strings) marking time.
For example, on my Linux environment, the following code
"20_000_000.times{''}"
with 40k symbols (similar symbol number on Rails 3.2.14 app,
@jugyo tells me) boosts, from 7.3sec to 4.2sec.
* internal.h: change prototype of rb_gc_mark_symbols().
Fri Oct 11 19:27:22 2013 Akinori MUSHA <knu@iDaemons.org>
* misc/ruby-electric.el: Import ruby-electric.el 2.0.1 which fixes
a bug and a flaw with auto-end introduced in the revamp.
* ruby-forward-sexp is inappropriate here because it moves the
cursor past the keyword.
* Fix a reversed looking-back check in
ruby-electric--block-beg-keyword-at-point-p.
* Do not add end again if space or return is hit repeatedly
after a block beginning keyword.
Fri Oct 11 18:12:47 2013 Koichi Sasada <ko1@atdot.net>
* ext/objspace/gc_hook.c: prohibit reentrant.
Fri Oct 11 18:11:34 2013 Koichi Sasada <ko1@atdot.net>
* vm_trace.c (rb_postponed_job_flush): fix bit operation.
Fri Oct 11 17:33:24 2013 Akinori MUSHA <knu@iDaemons.org>
* misc/ruby-electric.el: Import ruby-electric.el 2.0 from
https://github.com/knu/ruby-electric.el which integrates changes
from another fork by @qoobaa.
* Allow ruby-electric-mode to be disabled by introducing a
dedicated key map. Electric key bindings are now defined in
ruby-electric-mode-map instead of overwriting ruby-mode-map.
* Add ruby-electric-mode-hook.
* Use a remap in binding ruby-electric-delete-backward-char.
* Totally revamp electric keywords and then introduce electric
return. Modifier keywords are now properly detected making
use of ruby-mode's indentation level calculator, and
* block-mid keywords (then, else, elsif, when, rescue and
ensure) also become electric with automatic reindentation.
* Add standardized comments for ELPA integration.
* Fix interaction with smartparens-mode by disabling its end
keyword completion, since ruby-electric has become more clever
at it.
* The custom variable `ruby-electric-keywords` is changed to
`ruby-electric-keywords-alist`, allowing user to fine-grained
configuration.
Fri Oct 11 16:53:28 2013 Koichi Sasada <ko1@atdot.net>
* vm_trace.c (rb_postponed_job_flush): simplify.
Fri Oct 11 03:36:49 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
* thread.c (rb_threadptr_execute_interrupts): flush postponed job only
once at last.
* vm_trace.c (rb_postponed_job_flush): defer calling postponed jobs
registered while flushing to get rid of infinite reentrance of
ObjectSpace.after_gc_start_hook. [ruby-dev:47400] [Bug #8492]
Thu Oct 10 23:04:00 2013 Masaki Matsushita <glass.saga@gmail.com>
* array.c (rb_ary_or): remove unused variables.
Thu Oct 10 23:01:16 2013 Masaki Matsushita <glass.saga@gmail.com>
* array.c (rb_ary_or): use rb_hash_keys().
Thu Oct 10 21:36:16 2013 Masaki Matsushita <glass.saga@gmail.com>
* array.c (rb_ary_compact_bang): use ary_resize_smaller().
Thu Oct 10 17:25:28 2013 Koichi Sasada <ko1@atdot.net>
* vm.c (vm_exec): support :b_return event for "lambda{return}.call".
[Bug #8622]
* test/ruby/test_settracefunc.rb: add a test.
Thu Oct 10 13:52:37 2013 Koichi Sasada <ko1@atdot.net>
* vm_trace.c (postponed_job): use preallocated buffer.
Pre-allocate MAX_POSTPONED_JOB (1024) sized buffer
and use it.
If rb_postponed_job_register() cause overflow, simply it
fails and returns 0.
And maybe rb_postponed_job_register() is signal safe.
* vm_core.h: change data structure.
Thu Oct 10 11:11:33 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
* vm.c (Init_VM): hide also the singleton class of frozen-core, not
only frozen-core itself.
Thu Oct 10 06:02:08 2013 Koichi Sasada <ko1@atdot.net>
* test/ruby/test_rand.rb: fix r43224. local variable `e' is
no longer available.
Thu Oct 10 00:02:35 2013 Yusuke Endoh <mame@tsg.ne.jp>
* numeric.c (fix_aref): avoid a possible undefined behavior.
1L << 63 on 64-bit platform is undefined, at least, according to
ISO/IEC 9899 (C99) 6.5.7.
Wed Oct 9 23:57:02 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
* object.c (id_for_attr): avoid inadvertent symbol creation.
Wed Oct 9 18:03:01 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
* vm_method.c (rb_attr): preserve encoding of the attribute ID in
error message.
Wed Oct 9 17:40:16 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
* string.c (rb_fstring): because of lazy sweep, str may be unmarked
already and swept at next time, so mark it for the time being.
[ruby-core:57756]
Wed Oct 9 13:53:14 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
* compar.c (cmp_eq): fail if recursion. [ruby-core:57736] [Bug #9003]
* thread.c (rb_exec_recursive_paired_outer): new function which is
combination of paired and outer variants.
Wed Oct 9 09:18:14 2013 Koichi Sasada <ko1@atdot.net>
* include/ruby/debug.h,
vm_backtrace.c (rb_profile_frame_full_label): add new C API
rb_profile_frame_full_label() which returns label with
qualified method name.
Note that in future version of Ruby label() may return
same return value of full_label().
* ext/-test-/debug/profile_frames.c,
test/-ext-/debug/test_profile_frames.rb: fix a test for this change.
Wed Oct 9 00:55:51 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
* load.c (load_lock): display backtrace to $stderr at circular
require.
* vm_backtrace.c (rb_backtrace_print_to): new function to print
backtrace to the given output.
Tue Oct 8 21:03:35 2013 Koichi Sasada <ko1@atdot.net>
* vm_backtrace.c, include/ruby/debug.h: add new APIs
* VALUE rb_profile_frame_method_name(VALUE frame)
* VALUE rb_profile_frame_qualified_method_name(VALUE frame)
* iseq.c (rb_iseq_klass), internal.h: add new internal function
rb_iseq_method_name().
* ext/-test-/debug/profile_frames.c (profile_frames),
test/-ext-/debug/test_profile_frames.rb: add a test.
Tue Oct 8 16:11:11 2013 Nobuyoshi Nakada <nobu@ruby-lang.org>
* array.c (rb_ary_uniq): use rb_hash_values(), as well as the case no
block is given.
* internal.h: define rb_hash_values() as internal API.
Tue Oct 8 13:53:21 2013 Masaki Matsushita <glass.saga@gmail.com>
* array.c (rb_ary_uniq): use rb_hash_keys().
* internal.h: define rb_hash_keys() as internal API.