view tests/test-merge-combination.t @ 43198:c16fe77e340a

pathcopies: give up any optimization based on `introrev` Between 8a0136f69027 and d98fb3f42f33, we sped up the search for the introduction revision during path copies. However, further checking show that finding the introduction revision is still expensive and that we are better off without it. So we simply drop it and only rely on the linkrev optimisation. I ran `perfpathcopies` on 6989 pair of revision in the pypy repository (`hg perfhelper-pathcopies`. The result is massively in favor of dropping this condition. The result of the copy tracing are unchanged. Attempt to use a smaller changes preserving linkrev usage were unsuccessful, it can return wrong result. The following changesets broke test-mv-cp-st-diff.t - if not f.isintroducedafter(limit): + if limit >= 0 and f.linkrev() < limit: return None Here are various numbers (before this changeset/after this changesets) source destination before after saved-time ratio worth cases e66f24650daf 695dfb0f493b 1.062843 1.246369 -0.183526 1.172675 c979853a3b6a 8d60fe293e79 1.036985 1.196414 -0.159429 1.153743 22349fa2fc33 fbb1c9fd86c0 0.879926 1.038682 -0.158756 1.180420 682b98f3e672 a4878080a536 0.909952 1.063801 -0.153849 1.169074 5adabc9b9848 920958a93997 0.993622 1.147452 -0.153830 1.154817 worse 1% dbfbfcf077e9 aea8f2fd3593 1.016595 1.082999 -0.066404 1.065320 worse 5% c95f1ced15f2 7d29d5e39734 0.453694 0.471156 -0.017462 1.038488 worse 10% 3e144ed1d5b7 2aef0e942480 0.035140 0.037535 -0.002395 1.068156 worse 25% 321fc60db035 801748ba582a 0.009267 0.009325 -0.000058 1.006259 median 2088ce763fc2 e6991321d78b 0.000665 0.000651 0.000014 0.978947 best 25% 915631a97de6 385b31354be6 0.040743 0.040363 0.000380 0.990673 best 10% ad495c36a765 19c10384d3e7 0.431658 0.411490 0.020168 0.953278 best 5% d13ae7d283ae 813c99f810ac 1.141404 1.075346 0.066058 0.942126 best 1% 81593cb4a496 99ae11866969 1.833297 0.063823 1.769474 0.034813 best cases c3b14617fbd7 743a0fcaa4eb 1101.811740 2.735970 1099.075770 0.002483 c3b14617fbd7 9ba6ab77fd29 1116.753953 2.800729 1113.953224 0.002508 058b99d6e81f 57e249b7a3ea 1246.128485 3.042762 1243.085723 0.002442 9a8c361aab49 0354a250d371 1253.111894 3.085796 1250.026098 0.002463 442dbbc53c68 3ec1002a818c 1261.786294 3.138607 1258.647687 0.002487 As one can see, the average case is not really impacted. However, the worth case we get after this changeset are much better than the one we had before it. We have 30 pairs where improvements are above 10 minutes. This reflect in the combined time for all pairs before: 26256s after: 1300s (-95%) If we remove these pathological 30 cases, we still see a significant improvements: before: 1631s after: 1245s (-24%)
author Pierre-Yves David <pierre-yves.david@octobus.net>
date Thu, 10 Oct 2019 03:49:33 +0200
parents 302dbc9d52be
children 8045e4aa366b
line wrap: on
line source

This file shows what hg says are "modified" files for a merge commit
(hg log -T {files}), somewhat exhaustively.
It shows merges that involves files contents changing, and merges that
involve executable bit changing, but not merges with multiple or zero
merge ancestors, nor copies/renames, and nor identical file contents
with different filelog revisions.

genmerges is the workhorse. Given:
- a range function describing the possible values for file a
- a isgood function to filter out uninteresting combination
- a createfile function to actually write the values for file a on the
filesystem
it print a series of lines that look like: abcd C: output of -T {files}
describing the file a at respectively the base, p2, p1, merge
revision. "C" indicates that hg merge had conflicts.
  $ genmerges () {
  >   for base in `range` -; do
  >     for r1 in `range $base` -; do
  >       for r2 in `range $base $r1` -; do
  >         for m in `range $base $r1 $r2` -; do
  >           line="$base$r1$r2$m"
  >           isgood $line || continue
  >           hg init repo
  >           cd repo
  >           make_commit () {
  >             v=$1; msg=$2; file=$3;
  >             if [ $v != - ]; then
  >               createfile $v
  >             else
  >               if [ -f a ]
  >               then rm a
  >               else touch $file
  >               fi
  >             fi
  >             hg commit -q -Am $msg || exit 123
  >           }
  >           echo foo > foo
  >           make_commit $base base b
  >           make_commit $r1 r1 c
  >           hg up -r 0 -q
  >           make_commit $r2 r2 d
  >           hg merge -q -r 1 > ../output 2>&1
  >           if [ $? -ne 0 ]; then rm -f *.orig; hg resolve -m --all -q; fi
  >           if [ -s ../output ]; then conflicts=" C"; else conflicts="  "; fi
  >           make_commit $m m e
  >           if [ $m = $r1 ] && [ $m = $r2 ]
  >           then expected=
  >           elif [ $m = $r1 ]
  >           then if [ $base = $r2 ]
  >                then expected=
  >                else expected=a
  >                fi
  >           elif [ $m = $r2 ]
  >           then if [ $base = $r1 ]
  >                then expected=
  >                else expected=a
  >                fi
  >           else expected=a
  >           fi
  >           got=`hg log -r 3 --template '{files}\n' | tr -d 'e '`
  >           if [ "$got" = "$expected" ]
  >           then echo "$line$conflicts: agree on \"$got\""
  >           else echo "$line$conflicts: hg said \"$got\", expected \"$expected\""
  >           fi
  >           cd ../
  >           rm -rf repo
  >         done
  >       done
  >     done
  >   done
  > }

All the merges of various file contents.

  $ range () {
  >   max=0
  >   for i in $@; do
  >     if [ $i = - ]; then continue; fi
  >     if [ $i -gt $max ]; then max=$i; fi
  >   done
  >   $TESTDIR/seq.py `expr $max + 1`
  > }
  $ isgood () { true; }
  $ createfile () {
  >   if [ -f a ] && [ "`cat a`" = $1 ]
  >   then touch $file
  >   else echo $v > a
  >   fi
  > }

  $ genmerges
  1111  : agree on ""
  1112  : agree on "a"
  111-  : agree on "a"
  1121  : agree on "a"
  1122  : agree on ""
  1123  : agree on "a"
  112-  : agree on "a"
  11-1  : hg said "", expected "a"
  11-2  : agree on "a"
  11--  : agree on ""
  1211  : agree on "a"
  1212  : agree on ""
  1213  : agree on "a"
  121-  : agree on "a"
  1221  : agree on "a"
  1222  : agree on ""
  1223  : agree on "a"
  122-  : agree on "a"
  1231 C: agree on "a"
  1232 C: agree on "a"
  1233 C: agree on "a"
  1234 C: agree on "a"
  123- C: agree on "a"
  12-1 C: agree on "a"
  12-2 C: hg said "", expected "a"
  12-3 C: agree on "a"
  12-- C: agree on "a"
  1-11  : hg said "", expected "a"
  1-12  : agree on "a"
  1-1-  : agree on ""
  1-21 C: agree on "a"
  1-22 C: hg said "", expected "a"
  1-23 C: agree on "a"
  1-2- C: agree on "a"
  1--1  : agree on "a"
  1--2  : agree on "a"
  1---  : agree on ""
  -111  : agree on ""
  -112  : agree on "a"
  -11-  : agree on "a"
  -121 C: agree on "a"
  -122 C: agree on "a"
  -123 C: agree on "a"
  -12- C: agree on "a"
  -1-1  : agree on ""
  -1-2  : agree on "a"
  -1--  : agree on "a"
  --11  : agree on ""
  --12  : agree on "a"
  --1-  : agree on "a"
  ---1  : agree on "a"
  ----  : agree on ""

All the merges of executable bit.

  $ range () {
  >   max=a
  >   for i in $@; do
  >     if [ $i = - ]; then continue; fi
  >     if [ $i > $max ]; then max=$i; fi
  >   done
  >   if [ $max = a ]; then echo f; else echo f x; fi
  > }
  $ isgood () { case $line in *f*x*) true;; *) false;; esac; }
  $ createfile () {
  >   if [ -f a ] && (([ -x a ] && [ $v = x ]) || (! [ -x a ] && [ $v != x ]))
  >   then touch $file
  >   else touch a; if [ $v = x ]; then chmod +x a; else chmod -x a; fi
  >   fi
  > }

#if execbit
  $ genmerges
  fffx  : agree on "a"
  ffxf  : agree on "a"
  ffxx  : agree on ""
  ffx-  : agree on "a"
  ff-x  : hg said "", expected "a"
  fxff  : hg said "", expected "a"
  fxfx  : hg said "a", expected ""
  fxf-  : agree on "a"
  fxxf  : agree on "a"
  fxxx  : agree on ""
  fxx-  : agree on "a"
  fx-f  : hg said "", expected "a"
  fx-x  : hg said "", expected "a"
  fx--  : hg said "", expected "a"
  f-fx  : agree on "a"
  f-xf  : agree on "a"
  f-xx  : hg said "", expected "a"
  f-x-  : agree on "a"
  f--x  : agree on "a"
  -ffx  : agree on "a"
  -fxf C: agree on "a"
  -fxx C: hg said "", expected "a"
  -fx- C: agree on "a"
  -f-x  : hg said "", expected "a"
  --fx  : agree on "a"
#endif

Files modified or cleanly merged, with no greatest common ancestors:

  $ hg init repo; cd repo
  $ touch a0 b0; hg commit -qAm 0
  $ hg up -qr null; touch a1 b1; hg commit -qAm 1
  $ hg merge -qr 0; rm b*; hg commit -qAm 2
  $ hg log -r . -T '{files}\n'
  b0 b1
  $ cd ../
  $ rm -rf repo

A few cases of criss-cross merges involving deletions (listing all
such merges is probably too much). Both gcas contain $files, so we
expect the final merge to behave like a merge with a single gca
containing $files.

  $ hg init repo; cd repo
  $ files="c1 u1 c2 u2"
  $ touch $files; hg commit -qAm '0 root'
  $ for f in $files; do echo f > $f; done; hg commit -qAm '1 gca1'
  $ hg up -qr0; hg revert -qr 1 --all; hg commit -qAm '2 gca2'
  $ hg up -qr 1; hg merge -qr 2; rm *1; hg commit -qAm '3 p1'
  $ hg up -qr 2; hg merge -qr 1; rm *2; hg commit -qAm '4 p2'
  $ hg merge -qr 3; echo f > u1; echo f > u2; rm -f c1 c2
  $ hg commit -qAm '5 merge with two gcas'
  $ hg log -r . -T '{files}\n' # expecting u1 u2
  
  $ cd ../
  $ rm -rf repo