forked from git/git
-
Notifications
You must be signed in to change notification settings - Fork 1
Expand file tree
/
Copy pathgit-bisect-lk2009.html
More file actions
1479 lines (1428 loc) · 66.5 KB
/
git-bisect-lk2009.html
File metadata and controls
1479 lines (1428 loc) · 66.5 KB
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
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN"
"http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd">
<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en">
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8" />
<meta name="generator" content="AsciiDoc 8.2.5" />
<style type="text/css">
/* Debug borders */
p, li, dt, dd, div, pre, h1, h2, h3, h4, h5, h6 {
/*
border: 1px solid red;
*/
}
body {
margin: 1em 5% 1em 5%;
}
a {
color: blue;
text-decoration: underline;
}
a:visited {
color: fuchsia;
}
em {
font-style: italic;
}
strong {
font-weight: bold;
}
tt {
color: navy;
}
h1, h2, h3, h4, h5, h6 {
color: #527bbd;
font-family: sans-serif;
margin-top: 1.2em;
margin-bottom: 0.5em;
line-height: 1.3;
}
h1, h2, h3 {
border-bottom: 2px solid silver;
}
h2 {
padding-top: 0.5em;
}
h3 {
float: left;
}
h3 + * {
clear: left;
}
div.sectionbody {
font-family: serif;
margin-left: 0;
}
hr {
border: 1px solid silver;
}
p {
margin-top: 0.5em;
margin-bottom: 0.5em;
}
pre {
padding: 0;
margin: 0;
}
span#author {
color: #527bbd;
font-family: sans-serif;
font-weight: bold;
font-size: 1.1em;
}
span#email {
}
span#revision {
font-family: sans-serif;
}
div#footer {
font-family: sans-serif;
font-size: small;
border-top: 2px solid silver;
padding-top: 0.5em;
margin-top: 4.0em;
}
div#footer-text {
float: left;
padding-bottom: 0.5em;
}
div#footer-badges {
float: right;
padding-bottom: 0.5em;
}
div#preamble,
div.tableblock, div.imageblock, div.exampleblock, div.verseblock,
div.quoteblock, div.literalblock, div.listingblock, div.sidebarblock,
div.admonitionblock {
margin-right: 10%;
margin-top: 1.5em;
margin-bottom: 1.5em;
}
div.admonitionblock {
margin-top: 2.5em;
margin-bottom: 2.5em;
}
div.content { /* Block element content. */
padding: 0;
}
/* Block element titles. */
div.title, caption.title {
font-family: sans-serif;
font-weight: bold;
text-align: left;
margin-top: 1.0em;
margin-bottom: 0.5em;
}
div.title + * {
margin-top: 0;
}
td div.title:first-child {
margin-top: 0.0em;
}
div.content div.title:first-child {
margin-top: 0.0em;
}
div.content + div.title {
margin-top: 0.0em;
}
div.sidebarblock > div.content {
background: #ffffee;
border: 1px solid silver;
padding: 0.5em;
}
div.listingblock {
margin-right: 0%;
}
div.listingblock > div.content {
border: 1px solid silver;
background: #f4f4f4;
padding: 0.5em;
}
div.quoteblock > div.content {
padding-left: 2.0em;
}
div.attribution {
text-align: right;
}
div.verseblock + div.attribution {
text-align: left;
}
div.admonitionblock .icon {
vertical-align: top;
font-size: 1.1em;
font-weight: bold;
text-decoration: underline;
color: #527bbd;
padding-right: 0.5em;
}
div.admonitionblock td.content {
padding-left: 0.5em;
border-left: 2px solid silver;
}
div.exampleblock > div.content {
border-left: 2px solid silver;
padding: 0.5em;
}
div.verseblock div.content {
white-space: pre;
}
div.imageblock div.content { padding-left: 0; }
div.imageblock img { border: 1px solid silver; }
span.image img { border-style: none; }
dl {
margin-top: 0.8em;
margin-bottom: 0.8em;
}
dt {
margin-top: 0.5em;
margin-bottom: 0;
font-style: italic;
}
dd > *:first-child {
margin-top: 0;
}
ul, ol {
list-style-position: outside;
}
div.olist2 ol {
list-style-type: lower-alpha;
}
div.tableblock > table {
border: 3px solid #527bbd;
}
thead {
font-family: sans-serif;
font-weight: bold;
}
tfoot {
font-weight: bold;
}
div.hlist {
margin-top: 0.8em;
margin-bottom: 0.8em;
}
div.hlist td {
padding-bottom: 5px;
}
td.hlist1 {
vertical-align: top;
font-style: italic;
padding-right: 0.8em;
}
td.hlist2 {
vertical-align: top;
}
@media print {
div#footer-badges { display: none; }
}
div#toctitle {
color: #527bbd;
font-family: sans-serif;
font-size: 1.1em;
font-weight: bold;
margin-top: 1.0em;
margin-bottom: 0.1em;
}
div.toclevel1, div.toclevel2, div.toclevel3, div.toclevel4 {
margin-top: 0;
margin-bottom: 0;
}
div.toclevel2 {
margin-left: 2em;
font-size: 0.9em;
}
div.toclevel3 {
margin-left: 4em;
font-size: 0.9em;
}
div.toclevel4 {
margin-left: 6em;
font-size: 0.9em;
}
/* Workarounds for IE6's broken and incomplete CSS2. */
div.sidebar-content {
background: #ffffee;
border: 1px solid silver;
padding: 0.5em;
}
div.sidebar-title, div.image-title {
font-family: sans-serif;
font-weight: bold;
margin-top: 0.0em;
margin-bottom: 0.5em;
}
div.listingblock div.content {
border: 1px solid silver;
background: #f4f4f4;
padding: 0.5em;
}
div.quoteblock-content {
padding-left: 2.0em;
}
div.exampleblock-content {
border-left: 2px solid silver;
padding-left: 0.5em;
}
/* IE6 sets dynamically generated links as visited. */
div#toc a:visited { color: blue; }
</style>
<title>Fighting regressions with git bisect</title>
</head>
<body>
<div id="header">
<h1>Fighting regressions with git bisect</h1>
<span id="author">Christian Couder</span><br />
<span id="email"><tt><<a href="mailto:chriscool@tuxfamily.org">chriscool@tuxfamily.org</a>></tt></span><br />
2009/11/08
</div>
<h2 id="_abstract">Abstract</h2>
<div class="sectionbody">
<div class="para"><p>"git bisect" enables software users and developers to easily find the
commit that introduced a regression. We show why it is important to
have good tools to fight regressions. We describe how "git bisect"
works from the outside and the algorithms it uses inside. Then we
explain how to take advantage of "git bisect" to improve current
practices. And we discuss how "git bisect" could improve in the
future.</p></div>
</div>
<h2 id="_introduction_to_git_bisect">Introduction to "git bisect"</h2>
<div class="sectionbody">
<div class="para"><p>Git is a Distributed Version Control system (DVCS) created by Linus
Torvalds and maintained by Junio Hamano.</p></div>
<div class="para"><p>In Git like in many other Version Control Systems (VCS), the different
states of the data that is managed by the system are called
commits. And, as VCS are mostly used to manage software source code,
sometimes "interesting" changes of behavior in the software are
introduced in some commits.</p></div>
<div class="para"><p>In fact people are specially interested in commits that introduce a
"bad" behavior, called a bug or a regression. They are interested in
these commits because a commit (hopefully) contains a very small set
of source code changes. And it's much easier to understand and
properly fix a problem when you only need to check a very small set of
changes, than when you don't know where look in the first place.</p></div>
<div class="para"><p>So to help people find commits that introduce a "bad" behavior, the
"git bisect" set of commands was invented. And it follows of course
that in "git bisect" parlance, commits where the "interesting
behavior" is present are called "bad" commits, while other commits are
called "good" commits. And a commit that introduce the behavior we are
interested in is called a "first bad commit". Note that there could be
more than one "first bad commit" in the commit space we are searching.</p></div>
<div class="para"><p>So "git bisect" is designed to help find a "first bad commit". And to
be as efficient as possible, it tries to perform a binary search.</p></div>
</div>
<h2 id="_fighting_regressions_overview">Fighting regressions overview</h2>
<div class="sectionbody">
<h3 id="_regressions_a_big_problem">Regressions: a big problem</h3><div style="clear:left"></div>
<div class="para"><p>Regressions are a big problem in the software industry. But it's
difficult to put some real numbers behind that claim.</p></div>
<div class="para"><p>There are some numbers about bugs in general, like a NIST study in
2002 <a href="#1">[1]</a> that said:</p></div>
<div class="quoteblock">
<div class="quoteblock-content">
<div class="para"><p>Software bugs, or errors, are so prevalent and so detrimental that
they cost the U.S. economy an estimated $59.5 billion annually, or
about 0.6 percent of the gross domestic product, according to a newly
released study commissioned by the Department of Commerce's National
Institute of Standards and Technology (NIST). At the national level,
over half of the costs are borne by software users and the remainder
by software developers/vendors. The study also found that, although
all errors cannot be removed, more than a third of these costs, or an
estimated $22.2 billion, could be eliminated by an improved testing
infrastructure that enables earlier and more effective identification
and removal of software defects. These are the savings associated with
finding an increased percentage (but not 100 percent) of errors closer
to the development stages in which they are introduced. Currently,
over half of all errors are not found until "downstream" in the
development process or during post-sale software use.</p></div>
<div class="attribution">
</div></div></div>
<div class="para"><p>And then:</p></div>
<div class="quoteblock">
<div class="quoteblock-content">
<div class="para"><p>Software developers already spend approximately 80 percent of
development costs on identifying and correcting defects, and yet few
products of any type other than software are shipped with such high
levels of errors.</p></div>
<div class="attribution">
</div></div></div>
<div class="para"><p>Eventually the conclusion started with:</p></div>
<div class="quoteblock">
<div class="quoteblock-content">
<div class="para"><p>The path to higher software quality is significantly improved software
testing.</p></div>
<div class="attribution">
</div></div></div>
<div class="para"><p>There are other estimates saying that 80% of the cost related to
software is about maintenance <a href="#2">[2]</a>.</p></div>
<div class="para"><p>Though, according to Wikipedia <a href="#3">[3]</a>:</p></div>
<div class="quoteblock">
<div class="quoteblock-content">
<div class="para"><p>A common perception of maintenance is that it is merely fixing
bugs. However, studies and surveys over the years have indicated that
the majority, over 80%, of the maintenance effort is used for
non-corrective actions (Pigosky 1997). This perception is perpetuated
by users submitting problem reports that in reality are functionality
enhancements to the system.</p></div>
<div class="attribution">
</div></div></div>
<div class="para"><p>But we can guess that improving on existing software is very costly
because you have to watch out for regressions. At least this would
make the above studies consistent among themselves.</p></div>
<div class="para"><p>Of course some kind of software is developed, then used during some
time without being improved on much, and then finally thrown away. In
this case, of course, regressions may not be a big problem. But on the
other hand, there is a lot of big software that is continually
developed and maintained during years or even tens of years by a lot
of people. And as there are often many people who depend (sometimes
critically) on such software, regressions are a really big problem.</p></div>
<div class="para"><p>One such software is the linux kernel. And if we look at the linux
kernel, we can see that a lot of time and effort is spent to fight
regressions. The release cycle start with a 2 weeks long merge
window. Then the first release candidate (rc) version is tagged. And
after that about 7 or 8 more rc versions will appear with around one
week between each of them, before the final release.</p></div>
<div class="para"><p>The time between the first rc release and the final release is
supposed to be used to test rc versions and fight bugs and especially
regressions. And this time is more than 80% of the release cycle
time. But this is not the end of the fight yet, as of course it
continues after the release.</p></div>
<div class="para"><p>And then this is what Ingo Molnar (a well known linux kernel
developer) says about his use of git bisect:</p></div>
<div class="quoteblock">
<div class="quoteblock-content">
<div class="para"><p>I most actively use it during the merge window (when a lot of trees
get merged upstream and when the influx of bugs is the highest) - and
yes, there have been cases that i used it multiple times a day. My
average is roughly once a day.</p></div>
<div class="attribution">
</div></div></div>
<div class="para"><p>So regressions are fought all the time by developers, and indeed it is
well known that bugs should be fixed as soon as possible, so as soon
as they are found. That's why it is interesting to have good tools for
this purpose.</p></div>
<h3 id="_other_tools_to_fight_regressions">Other tools to fight regressions</h3><div style="clear:left"></div>
<div class="para"><p>So what are the tools used to fight regressions? They are nearly the
same as those used to fight regular bugs. The only specific tools are
test suites and tools similar as "git bisect".</p></div>
<div class="para"><p>Test suites are very nice. But when they are used alone, they are
supposed to be used so that all the tests are checked after each
commit. This means that they are not very efficient, because many
tests are run for no interesting result, and they suffer from
combinational explosion.</p></div>
<div class="para"><p>In fact the problem is that big software often has many different
configuration options and that each test case should pass for each
configuration after each commit. So if you have for each release: N
configurations, M commits and T test cases, you should perform:</p></div>
<div class="listingblock">
<div class="content">
<pre><tt>N * M * T tests</tt></pre>
</div></div>
<div class="para"><p>where N, M and T are all growing with the size your software.</p></div>
<div class="para"><p>So very soon it will not be possible to completely test everything.</p></div>
<div class="para"><p>And if some bugs slip through your test suite, then you can add a test
to your test suite. But if you want to use your new improved test
suite to find where the bug slipped in, then you will either have to
emulate a bisection process or you will perhaps bluntly test each
commit backward starting from the "bad" commit you have which may be
very wasteful.</p></div>
</div>
<h2 id="_git_bisect_overview">"git bisect" overview</h2>
<div class="sectionbody">
<h3 id="_starting_a_bisection">Starting a bisection</h3><div style="clear:left"></div>
<div class="para"><p>The first "git bisect" subcommand to use is "git bisect start" to
start the search. Then bounds must be set to limit the commit
space. This is done usually by giving one "bad" and at least one
"good" commit. They can be passed in the initial call to "git bisect
start" like this:</p></div>
<div class="listingblock">
<div class="content">
<pre><tt>$ git bisect start [BAD [GOOD...]]</tt></pre>
</div></div>
<div class="para"><p>or they can be set using:</p></div>
<div class="listingblock">
<div class="content">
<pre><tt>$ git bisect bad [COMMIT]</tt></pre>
</div></div>
<div class="para"><p>and:</p></div>
<div class="listingblock">
<div class="content">
<pre><tt>$ git bisect good [COMMIT...]</tt></pre>
</div></div>
<div class="para"><p>where BAD, GOOD and COMMIT are all names that can be resolved to a
commit.</p></div>
<div class="para"><p>Then "git bisect" will checkout a commit of its choosing and ask the
user to test it, like this:</p></div>
<div class="listingblock">
<div class="content">
<pre><tt>$ git bisect start v2.6.27 v2.6.25
Bisecting: 10928 revisions left to test after this (roughly 14 steps)
[2ec65f8b89ea003c27ff7723525a2ee335a2b393] x86: clean up using max_low_pfn on 32-bit</tt></pre>
</div></div>
<div class="para"><p>Note that the example that we will use is really a toy example, we
will be looking for the first commit that has a version like
"2.6.26-something", that is the commit that has a "SUBLEVEL = 26" line
in the top level Makefile. This is a toy example because there are
better ways to find this commit with git than using "git bisect" (for
example "git blame" or "git log -S<string>").</p></div>
<h3 id="_driving_a_bisection_manually">Driving a bisection manually</h3><div style="clear:left"></div>
<div class="para"><p>At this point there are basically 2 ways to drive the search. It can
be driven manually by the user or it can be driven automatically by a
script or a command.</p></div>
<div class="para"><p>If the user is driving it, then at each step of the search, the user
will have to test the current commit and say if it is "good" or "bad"
using the "git bisect good" or "git bisect bad" commands respectively
that have been described above. For example:</p></div>
<div class="listingblock">
<div class="content">
<pre><tt>$ git bisect bad
Bisecting: 5480 revisions left to test after this (roughly 13 steps)
[66c0b394f08fd89236515c1c84485ea712a157be] KVM: kill file->f_count abuse in kvm</tt></pre>
</div></div>
<div class="para"><p>And after a few more steps like that, "git bisect" will eventually
find a first bad commit:</p></div>
<div class="listingblock">
<div class="content">
<pre><tt>$ git bisect bad
2ddcca36c8bcfa251724fe342c8327451988be0d is the first bad commit
commit 2ddcca36c8bcfa251724fe342c8327451988be0d
Author: Linus Torvalds <torvalds@linux-foundation.org>
Date: Sat May 3 11:59:44 2008 -0700
Linux 2.6.26-rc1
:100644 100644 5cf8258195331a4dbdddff08b8d68642638eea57 4492984efc09ab72ff6219a7bc21fb6a957c4cd5 M Makefile</tt></pre>
</div></div>
<div class="para"><p>At this point we can see what the commit does, check it out (if it's
not already checked out) or tinker with it, for example:</p></div>
<div class="listingblock">
<div class="content">
<pre><tt>$ git show HEAD
commit 2ddcca36c8bcfa251724fe342c8327451988be0d
Author: Linus Torvalds <torvalds@linux-foundation.org>
Date: Sat May 3 11:59:44 2008 -0700
Linux 2.6.26-rc1
diff --git a/Makefile b/Makefile
index 5cf8258..4492984 100644
--- a/Makefile
+++ b/Makefile
@@ -1,7 +1,7 @@
VERSION = 2
PATCHLEVEL = 6
-SUBLEVEL = 25
-EXTRAVERSION =
+SUBLEVEL = 26
+EXTRAVERSION = -rc1
NAME = Funky Weasel is Jiggy wit it
# *DOCUMENTATION*</tt></pre>
</div></div>
<div class="para"><p>And when we are finished we can use "git bisect reset" to go back to
the branch we were in before we started bisecting:</p></div>
<div class="listingblock">
<div class="content">
<pre><tt>$ git bisect reset
Checking out files: 100% (21549/21549), done.
Previous HEAD position was 2ddcca3... Linux 2.6.26-rc1
Switched to branch 'master'</tt></pre>
</div></div>
<h3 id="_driving_a_bisection_automatically">Driving a bisection automatically</h3><div style="clear:left"></div>
<div class="para"><p>The other way to drive the bisection process is to tell "git bisect"
to launch a script or command at each bisection step to know if the
current commit is "good" or "bad". To do that, we use the "git bisect
run" command. For example:</p></div>
<div class="listingblock">
<div class="content">
<pre><tt>$ git bisect start v2.6.27 v2.6.25
Bisecting: 10928 revisions left to test after this (roughly 14 steps)
[2ec65f8b89ea003c27ff7723525a2ee335a2b393] x86: clean up using max_low_pfn on 32-bit
$
$ git bisect run grep '^SUBLEVEL = 25' Makefile
running grep ^SUBLEVEL = 25 Makefile
Bisecting: 5480 revisions left to test after this (roughly 13 steps)
[66c0b394f08fd89236515c1c84485ea712a157be] KVM: kill file->f_count abuse in kvm
running grep ^SUBLEVEL = 25 Makefile
SUBLEVEL = 25
Bisecting: 2740 revisions left to test after this (roughly 12 steps)
[671294719628f1671faefd4882764886f8ad08cb] V4L/DVB(7879): Adding cx18 Support for mxl5005s
...
...
running grep ^SUBLEVEL = 25 Makefile
Bisecting: 0 revisions left to test after this (roughly 0 steps)
[2ddcca36c8bcfa251724fe342c8327451988be0d] Linux 2.6.26-rc1
running grep ^SUBLEVEL = 25 Makefile
2ddcca36c8bcfa251724fe342c8327451988be0d is the first bad commit
commit 2ddcca36c8bcfa251724fe342c8327451988be0d
Author: Linus Torvalds <torvalds@linux-foundation.org>
Date: Sat May 3 11:59:44 2008 -0700
Linux 2.6.26-rc1
:100644 100644 5cf8258195331a4dbdddff08b8d68642638eea57 4492984efc09ab72ff6219a7bc21fb6a957c4cd5 M Makefile
bisect run success</tt></pre>
</div></div>
<div class="para"><p>In this example, we passed "grep <em>^SUBLEVEL = 25</em> Makefile" as
parameter to "git bisect run". This means that at each step, the grep
command we passed will be launched. And if it exits with code 0 (that
means success) then git bisect will mark the current state as
"good". If it exits with code 1 (or any code between 1 and 127
included, except the special code 125), then the current state will be
marked as "bad".</p></div>
<div class="para"><p>Exit code between 128 and 255 are special to "git bisect run". They
make it stop immediately the bisection process. This is useful for
example if the command passed takes too long to complete, because you
can kill it with a signal and it will stop the bisection process.</p></div>
<div class="para"><p>It can also be useful in scripts passed to "git bisect run" to "exit
255" if some very abnormal situation is detected.</p></div>
<h3 id="_avoiding_untestable_commits">Avoiding untestable commits</h3><div style="clear:left"></div>
<div class="para"><p>Sometimes it happens that the current state cannot be tested, for
example if it does not compile because there was a bug preventing it
at that time. This is what the special exit code 125 is for. It tells
"git bisect run" that the current commit should be marked as
untestable and that another one should be chosen and checked out.</p></div>
<div class="para"><p>If the bisection process is driven manually, you can use "git bisect
skip" to do the same thing. (In fact the special exit code 125 makes
"git bisect run" use "git bisect skip" in the background.)</p></div>
<div class="para"><p>Or if you want more control, you can inspect the current state using
for example "git bisect visualize". It will launch gitk (or "git log"
if the DISPLAY environment variable is not set) to help you find a
better bisection point.</p></div>
<div class="para"><p>Either way, if you have a string of untestable commits, it might
happen that the regression you are looking for has been introduced by
one of these untestable commits. In this case it's not possible to
tell for sure which commit introduced the regression.</p></div>
<div class="para"><p>So if you used "git bisect skip" (or the run script exited with
special code 125) you could get a result like this:</p></div>
<div class="listingblock">
<div class="content">
<pre><tt>There are only 'skip'ped commits left to test.
The first bad commit could be any of:
15722f2fa328eaba97022898a305ffc8172db6b1
78e86cf3e850bd755bb71831f42e200626fbd1e0
e15b73ad3db9b48d7d1ade32f8cd23a751fe0ace
070eab2303024706f2924822bfec8b9847e4ac1b
We cannot bisect more!</tt></pre>
</div></div>
<h3 id="_saving_a_log_and_replaying_it">Saving a log and replaying it</h3><div style="clear:left"></div>
<div class="para"><p>If you want to show other people your bisection process, you can get a
log using for example:</p></div>
<div class="listingblock">
<div class="content">
<pre><tt>$ git bisect log > bisect_log.txt</tt></pre>
</div></div>
<div class="para"><p>And it is possible to replay it using:</p></div>
<div class="listingblock">
<div class="content">
<pre><tt>$ git bisect replay bisect_log.txt</tt></pre>
</div></div>
</div>
<h2 id="_git_bisect_details">"git bisect" details</h2>
<div class="sectionbody">
<h3 id="_bisection_algorithm">Bisection algorithm</h3><div style="clear:left"></div>
<div class="para"><p>As the Git commits form a directed acyclic graph (DAG), finding the
best bisection commit to test at each step is not so simple. Anyway
Linus found and implemented a "truly stupid" algorithm, later improved
by Junio Hamano, that works quite well.</p></div>
<div class="para"><p>So the algorithm used by "git bisect" to find the best bisection
commit when there are no skipped commits is the following:</p></div>
<div class="para"><p>1) keep only the commits that:</p></div>
<div class="para"><p>a) are ancestor of the "bad" commit (including the "bad" commit itself),
b) are not ancestor of a "good" commit (excluding the "good" commits).</p></div>
<div class="para"><p>This means that we get rid of the uninteresting commits in the DAG.</p></div>
<div class="para"><p>For example if we start with a graph like this:</p></div>
<div class="listingblock">
<div class="content">
<pre><tt>G-Y-G-W-W-W-X-X-X-X
\ /
W-W-B
/
Y---G-W---W
\ / \
Y-Y X-X-X-X
-> time goes this way -></tt></pre>
</div></div>
<div class="para"><p>where B is the "bad" commit, "G" are "good" commits and W, X, and Y
are other commits, we will get the following graph after this first
step:</p></div>
<div class="listingblock">
<div class="content">
<pre><tt>W-W-W
\
W-W-B
/
W---W</tt></pre>
</div></div>
<div class="para"><p>So only the W and B commits will be kept. Because commits X and Y will
have been removed by rules a) and b) respectively, and because commits
G are removed by rule b) too.</p></div>
<div class="para"><p>Note for git users, that it is equivalent as keeping only the commit
given by:</p></div>
<div class="listingblock">
<div class="content">
<pre><tt>git rev-list BAD --not GOOD1 GOOD2...</tt></pre>
</div></div>
<div class="para"><p>Also note that we don't require the commits that are kept to be
descendants of a "good" commit. So in the following example, commits W
and Z will be kept:</p></div>
<div class="listingblock">
<div class="content">
<pre><tt>G-W-W-W-B
/
Z-Z</tt></pre>
</div></div>
<div class="para"><p>2) starting from the "good" ends of the graph, associate to each
commit the number of ancestors it has plus one</p></div>
<div class="para"><p>For example with the following graph where H is the "bad" commit and A
and D are some parents of some "good" commits:</p></div>
<div class="listingblock">
<div class="content">
<pre><tt>A-B-C
\
F-G-H
/
D---E</tt></pre>
</div></div>
<div class="para"><p>this will give:</p></div>
<div class="listingblock">
<div class="content">
<pre><tt>1 2 3
A-B-C
\6 7 8
F-G-H
1 2/
D---E</tt></pre>
</div></div>
<div class="para"><p>3) associate to each commit: min(X, N - X)</p></div>
<div class="para"><p>where X is the value associated to the commit in step 2) and N is the
total number of commits in the graph.</p></div>
<div class="para"><p>In the above example we have N = 8, so this will give:</p></div>
<div class="listingblock">
<div class="content">
<pre><tt>1 2 3
A-B-C
\2 1 0
F-G-H
1 2/
D---E</tt></pre>
</div></div>
<div class="para"><p>4) the best bisection point is the commit with the highest associated
number</p></div>
<div class="para"><p>So in the above example the best bisection point is commit C.</p></div>
<div class="para"><p>5) note that some shortcuts are implemented to speed up the algorithm</p></div>
<div class="para"><p>As we know N from the beginning, we know that min(X, N - X) can't be
greater than N/2. So during steps 2) and 3), if we would associate N/2
to a commit, then we know this is the best bisection point. So in this
case we can just stop processing any other commit and return the
current commit.</p></div>
<h3 id="_bisection_algorithm_debugging">Bisection algorithm debugging</h3><div style="clear:left"></div>
<div class="para"><p>For any commit graph, you can see the number associated with each
commit using "git rev-list —bisect-all".</p></div>
<div class="para"><p>For example, for the above graph, a command like:</p></div>
<div class="listingblock">
<div class="content">
<pre><tt>$ git rev-list --bisect-all BAD --not GOOD1 GOOD2</tt></pre>
</div></div>
<div class="para"><p>would output something like:</p></div>
<div class="listingblock">
<div class="content">
<pre><tt>e15b73ad3db9b48d7d1ade32f8cd23a751fe0ace (dist=3)
15722f2fa328eaba97022898a305ffc8172db6b1 (dist=2)
78e86cf3e850bd755bb71831f42e200626fbd1e0 (dist=2)
a1939d9a142de972094af4dde9a544e577ddef0e (dist=2)
070eab2303024706f2924822bfec8b9847e4ac1b (dist=1)
a3864d4f32a3bf5ed177ddef598490a08760b70d (dist=1)
a41baa717dd74f1180abf55e9341bc7a0bb9d556 (dist=1)
9e622a6dad403b71c40979743bb9d5be17b16bd6 (dist=0)</tt></pre>
</div></div>
<h3 id="_bisection_algorithm_discussed">Bisection algorithm discussed</h3><div style="clear:left"></div>
<div class="para"><p>First let's define "best bisection point". We will say that a commit X
is a best bisection point or a best bisection commit if knowing its
state ("good" or "bad") gives as much information as possible whether
the state of the commit happens to be "good" or "bad".</p></div>
<div class="para"><p>This means that the best bisection commits are the commits where the
following function is maximum:</p></div>
<div class="listingblock">
<div class="content">
<pre><tt>f(X) = min(information_if_good(X), information_if_bad(X))</tt></pre>
</div></div>
<div class="para"><p>where information_if_good(X) is the information we get if X is good
and information_if_bad(X) is the information we get if X is bad.</p></div>
<div class="para"><p>Now we will suppose that there is only one "first bad commit". This
means that all its descendants are "bad" and all the other commits are
"good". And we will suppose that all commits have an equal probability
of being good or bad, or of being the first bad commit, so knowing the
state of c commits gives always the same amount of information
wherever these c commits are on the graph and whatever c is. (So we
suppose that these commits being for example on a branch or near a
good or a bad commit does not give more or less information).</p></div>
<div class="para"><p>Let's also suppose that we have a cleaned up graph like one after step
1) in the bisection algorithm above. This means that we can measure
the information we get in terms of number of commit we can remove from
the graph..</p></div>
<div class="para"><p>And let's take a commit X in the graph.</p></div>
<div class="para"><p>If X is found to be "good", then we know that its ancestors are all
"good", so we want to say that:</p></div>
<div class="listingblock">
<div class="content">
<pre><tt>information_if_good(X) = number_of_ancestors(X) (TRUE)</tt></pre>
</div></div>
<div class="para"><p>And this is true because at step 1) b) we remove the ancestors of the
"good" commits.</p></div>
<div class="para"><p>If X is found to be "bad", then we know that its descendants are all
"bad", so we want to say that:</p></div>
<div class="listingblock">
<div class="content">
<pre><tt>information_if_bad(X) = number_of_descendants(X) (WRONG)</tt></pre>
</div></div>
<div class="para"><p>But this is wrong because at step 1) a) we keep only the ancestors of
the bad commit. So we get more information when a commit is marked as
"bad", because we also know that the ancestors of the previous "bad"
commit that are not ancestors of the new "bad" commit are not the
first bad commit. We don't know if they are good or bad, but we know
that they are not the first bad commit because they are not ancestor
of the new "bad" commit.</p></div>
<div class="para"><p>So when a commit is marked as "bad" we know we can remove all the
commits in the graph except those that are ancestors of the new "bad"
commit. This means that:</p></div>
<div class="listingblock">
<div class="content">
<pre><tt>information_if_bad(X) = N - number_of_ancestors(X) (TRUE)</tt></pre>
</div></div>
<div class="para"><p>where N is the number of commits in the (cleaned up) graph.</p></div>
<div class="para"><p>So in the end this means that to find the best bisection commits we
should maximize the function:</p></div>
<div class="listingblock">
<div class="content">
<pre><tt>f(X) = min(number_of_ancestors(X), N - number_of_ancestors(X))</tt></pre>
</div></div>
<div class="para"><p>And this is nice because at step 2) we compute number_of_ancestors(X)
and so at step 3) we compute f(X).</p></div>
<div class="para"><p>Let's take the following graph as an example:</p></div>
<div class="listingblock">
<div class="content">
<pre><tt> G-H-I-J
/ \
A-B-C-D-E-F O
\ /
K-L-M-N</tt></pre>
</div></div>
<div class="para"><p>If we compute the following non optimal function on it:</p></div>
<div class="listingblock">
<div class="content">
<pre><tt>g(X) = min(number_of_ancestors(X), number_of_descendants(X))</tt></pre>
</div></div>
<div class="para"><p>we get:</p></div>
<div class="listingblock">
<div class="content">
<pre><tt> 4 3 2 1
G-H-I-J
1 2 3 4 5 6/ \0
A-B-C-D-E-F O
\ /
K-L-M-N
4 3 2 1</tt></pre>
</div></div>
<div class="para"><p>but with the algorithm used by git bisect we get:</p></div>
<div class="listingblock">
<div class="content">
<pre><tt> 7 7 6 5
G-H-I-J
1 2 3 4 5 6/ \0
A-B-C-D-E-F O
\ /
K-L-M-N
7 7 6 5</tt></pre>
</div></div>
<div class="para"><p>So we chose G, H, K or L as the best bisection point, which is better
than F. Because if for example L is bad, then we will know not only
that L, M and N are bad but also that G, H, I and J are not the first
bad commit (since we suppose that there is only one first bad commit
and it must be an ancestor of L).</p></div>
<div class="para"><p>So the current algorithm seems to be the best possible given what we
initially supposed.</p></div>
<h3 id="_skip_algorithm">Skip algorithm</h3><div style="clear:left"></div>
<div class="para"><p>When some commits have been skipped (using "git bisect skip"), then
the bisection algorithm is the same for step 1) to 3). But then we use
roughly the following steps:</p></div>
<div class="para"><p>6) sort the commit by decreasing associated value</p></div>
<div class="para"><p>7) if the first commit has not been skipped, we can return it and stop
here</p></div>
<div class="para"><p>8) otherwise filter out all the skipped commits in the sorted list</p></div>
<div class="para"><p>9) use a pseudo random number generator (PRNG) to generate a random
number between 0 and 1</p></div>
<div class="para"><p>10) multiply this random number with its square root to bias it toward
0</p></div>
<div class="para"><p>11) multiply the result by the number of commits in the filtered list
to get an index into this list</p></div>
<div class="para"><p>12) return the commit at the computed index</p></div>
<h3 id="_skip_algorithm_discussed">Skip algorithm discussed</h3><div style="clear:left"></div>
<div class="para"><p>After step 7) (in the skip algorithm), we could check if the second
commit has been skipped and return it if it is not the case. And in
fact that was the algorithm we used from when "git bisect skip" was
developed in git version 1.5.4 (released on February 1st 2008) until
git version 1.6.4 (released July 29th 2009).</p></div>
<div class="para"><p>But Ingo Molnar and H. Peter Anvin (another well known linux kernel
developer) both complained that sometimes the best bisection points
all happened to be in an area where all the commits are
untestable. And in this case the user was asked to test many
untestable commits, which could be very inefficient.</p></div>
<div class="para"><p>Indeed untestable commits are often untestable because a breakage was
introduced at one time, and that breakage was fixed only after many
other commits were introduced.</p></div>
<div class="para"><p>This breakage is of course most of the time unrelated to the breakage
we are trying to locate in the commit graph. But it prevents us to
know if the interesting "bad behavior" is present or not.</p></div>
<div class="para"><p>So it is a fact that commits near an untestable commit have a high
probability of being untestable themselves. And the best bisection
commits are often found together too (due to the bisection algorithm).</p></div>
<div class="para"><p>This is why it is a bad idea to just chose the next best unskipped
bisection commit when the first one has been skipped.</p></div>
<div class="para"><p>We found that most commits on the graph may give quite a lot of
information when they are tested. And the commits that will not on
average give a lot of information are the one near the good and bad
commits.</p></div>
<div class="para"><p>So using a PRNG with a bias to favor commits away from the good and
bad commits looked like a good choice.</p></div>
<div class="para"><p>One obvious improvement to this algorithm would be to look for a
commit that has an associated value near the one of the best bisection
commit, and that is on another branch, before using the PRNG. Because
if such a commit exists, then it is not very likely to be untestable
too, so it will probably give more information than a nearly randomly
chosen one.</p></div>
<h3 id="_checking_merge_bases">Checking merge bases</h3><div style="clear:left"></div>
<div class="para"><p>There is another tweak in the bisection algorithm that has not been
described in the "bisection algorithm" above.</p></div>
<div class="para"><p>We supposed in the previous examples that the "good" commits were
ancestors of the "bad" commit. But this is not a requirement of "git
bisect".</p></div>
<div class="para"><p>Of course the "bad" commit cannot be an ancestor of a "good" commit,
because the ancestors of the good commits are supposed to be
"good". And all the "good" commits must be related to the bad commit.
They cannot be on a branch that has no link with the branch of the
"bad" commit. But it is possible for a good commit to be related to a
bad commit and yet not be neither one of its ancestor nor one of its
descendants.</p></div>
<div class="para"><p>For example, there can be a "main" branch, and a "dev" branch that was
forked of the main branch at a commit named "D" like this:</p></div>
<div class="listingblock">
<div class="content">
<pre><tt>A-B-C-D-E-F-G <--main
\
H-I-J <--dev</tt></pre>
</div></div>
<div class="para"><p>The commit "D" is called a "merge base" for branch "main" and "dev"
because it's the best common ancestor for these branches for a merge.</p></div>
<div class="para"><p>Now let's suppose that commit J is bad and commit G is good and that
we apply the bisection algorithm like it has been previously
described.</p></div>
<div class="para"><p>As described in step 1) b) of the bisection algorithm, we remove all
the ancestors of the good commits because they are supposed to be good
too.</p></div>
<div class="para"><p>So we would be left with only:</p></div>
<div class="listingblock">
<div class="content">
<pre><tt>H-I-J</tt></pre>
</div></div>
<div class="para"><p>But what happens if the first bad commit is "B" and if it has been
fixed in the "main" branch by commit "F"?</p></div>
<div class="para"><p>The result of such a bisection would be that we would find that H is
the first bad commit, when in fact it's B. So that would be wrong!</p></div>
<div class="para"><p>And yes it can happen in practice that people working on one branch
are not aware that people working on another branch fixed a bug! It
could also happen that F fixed more than one bug or that it is a
revert of some big development effort that was not ready to be
released.</p></div>
<div class="para"><p>In fact development teams often maintain both a development branch and
a maintenance branch, and it would be quite easy for them if "git
bisect" just worked when they want to bisect a regression on the
development branch that is not on the maintenance branch. They should
be able to start bisecting using:</p></div>
<div class="listingblock">
<div class="content">
<pre><tt>$ git bisect start dev main</tt></pre>
</div></div>
<div class="para"><p>To enable that additional nice feature, when a bisection is started
and when some good commits are not ancestors of the bad commit, we
first compute the merge bases between the bad and the good commits and
we chose these merge bases as the first commits that will be checked
out and tested.</p></div>
<div class="para"><p>If it happens that one merge base is bad, then the bisection process
is stopped with a message like:</p></div>
<div class="listingblock">
<div class="content">
<pre><tt>The merge base BBBBBB is bad.
This means the bug has been fixed between BBBBBB and [GGGGGG,...].</tt></pre>
</div></div>
<div class="para"><p>where BBBBBB is the sha1 hash of the bad merge base and [GGGGGG,…]
is a comma separated list of the sha1 of the good commits.</p></div>
<div class="para"><p>If some of the merge bases are skipped, then the bisection process
continues, but the following message is printed for each skipped merge
base:</p></div>
<div class="listingblock">