Mercurial > hg
view tests/test-histedit-fold.t @ 30155:b7a966ce89ed
changelog: disable delta chains
This patch disables delta chains on changelogs. After this patch, new
entries on changelogs - including existing changelogs - will be stored
as the fulltext of that data (likely compressed). No delta computation
will be performed.
An overview of delta chains and data justifying this change follows.
Revlogs try to store entries as a delta against a previous entry (either
a parent revision in the case of generaldelta or the previous physical
revision when not using generaldelta). Most of the time this is the
correct thing to do: it frequently results in less CPU usage and smaller
storage.
Delta chains are most effective when the base revision being deltad
against is similar to the current data. This tends to occur naturally
for manifests and file data, since only small parts of each tend to
change with each revision. Changelogs, however, are a different story.
Changelog entries represent changesets/commits. And unless commits in a
repository are homogonous (same author, changing same files, similar
commit messages, etc), a delta from one entry to the next tends to be
relatively large compared to the size of the entry. This means that
delta chains tend to be short. How short? Here is the full vs delta
revision breakdown on some real world repos:
Repo % Full % Delta Max Length
hg 45.8 54.2 6
mozilla-central 42.4 57.6 8
mozilla-unified 42.5 57.5 17
pypy 46.1 53.9 6
python-zstandard 46.1 53.9 3
(I threw in python-zstandard as an example of a repo that is homogonous.
It contains a small Python project with changes all from the same
author.)
Contrast this with the manifest revlog for these repos, where 99+% of
revisions are deltas and delta chains run into the thousands.
So delta chains aren't as useful on changelogs. But even a short delta
chain may provide benefits. Let's measure that.
Delta chains may require less CPU to read revisions if the CPU time
spent reading smaller deltas is less than the CPU time used to
decompress larger individual entries. We can measure this via
`hg perfrevlog -c -d 1` to iterate a revlog to resolve each revision's
fulltext. Here are the results of that command on a repo using delta
chains in its changelog and on a repo without delta chains:
hg (forward)
! wall 0.407008 comb 0.410000 user 0.410000 sys 0.000000 (best of 25)
! wall 0.390061 comb 0.390000 user 0.390000 sys 0.000000 (best of 26)
hg (reverse)
! wall 0.515221 comb 0.520000 user 0.520000 sys 0.000000 (best of 19)
! wall 0.400018 comb 0.400000 user 0.390000 sys 0.010000 (best of 25)
mozilla-central (forward)
! wall 4.508296 comb 4.490000 user 4.490000 sys 0.000000 (best of 3)
! wall 4.370222 comb 4.370000 user 4.350000 sys 0.020000 (best of 3)
mozilla-central (reverse)
! wall 5.758995 comb 5.760000 user 5.720000 sys 0.040000 (best of 3)
! wall 4.346503 comb 4.340000 user 4.320000 sys 0.020000 (best of 3)
mozilla-unified (forward)
! wall 4.957088 comb 4.950000 user 4.940000 sys 0.010000 (best of 3)
! wall 4.660528 comb 4.650000 user 4.630000 sys 0.020000 (best of 3)
mozilla-unified (reverse)
! wall 6.119827 comb 6.110000 user 6.090000 sys 0.020000 (best of 3)
! wall 4.675136 comb 4.670000 user 4.670000 sys 0.000000 (best of 3)
pypy (forward)
! wall 1.231122 comb 1.240000 user 1.230000 sys 0.010000 (best of 8)
! wall 1.164896 comb 1.160000 user 1.160000 sys 0.000000 (best of 9)
pypy (reverse)
! wall 1.467049 comb 1.460000 user 1.460000 sys 0.000000 (best of 7)
! wall 1.160200 comb 1.170000 user 1.160000 sys 0.010000 (best of 9)
The data clearly shows that it takes less wall and CPU time to resolve
revisions when there are no delta chains in the changelogs, regardless
of the direction of traversal. Furthermore, not using a delta chain
means that fulltext resolution in reverse is as fast as iterating
forward. So not using delta chains on the changelog is a clear CPU win
for reading operations.
An example of a user-visible operation showing this speed-up is revset
evaluation. Here are results for
`hg perfrevset 'author(gps) or author(mpm)'`:
hg
! wall 1.655506 comb 1.660000 user 1.650000 sys 0.010000 (best of 6)
! wall 1.612723 comb 1.610000 user 1.600000 sys 0.010000 (best of 7)
mozilla-central
! wall 17.629826 comb 17.640000 user 17.600000 sys 0.040000 (best of 3)
! wall 17.311033 comb 17.300000 user 17.260000 sys 0.040000 (best of 3)
What about 00changelog.i size?
Repo Delta Chains No Delta Chains
hg 7,033,250 6,976,771
mozilla-central 82,978,748 81,574,623
mozilla-unified 88,112,349 86,702,162
pypy 20,740,699 20,659,741
The data shows that removing delta chains from the changelog makes the
changelog smaller.
Delta chains are also used during changegroup generation. This
operation essentially converts a series of revisions to one large
delta chain. And changegroup generation is smart: if the delta in
the revlog matches what the changegroup is emitting, it will reuse
the delta instead of recalculating it. We can measure the impact
removing changelog delta chains has on changegroup generation via
`hg perfchangegroupchangelog`:
hg
! wall 1.589245 comb 1.590000 user 1.590000 sys 0.000000 (best of 7)
! wall 1.788060 comb 1.790000 user 1.790000 sys 0.000000 (best of 6)
mozilla-central
! wall 17.382585 comb 17.380000 user 17.340000 sys 0.040000 (best of 3)
! wall 20.161357 comb 20.160000 user 20.120000 sys 0.040000 (best of 3)
mozilla-unified
! wall 18.722839 comb 18.720000 user 18.680000 sys 0.040000 (best of 3)
! wall 21.168075 comb 21.170000 user 21.130000 sys 0.040000 (best of 3)
pypy
! wall 4.828317 comb 4.830000 user 4.820000 sys 0.010000 (best of 3)
! wall 5.415455 comb 5.420000 user 5.410000 sys 0.010000 (best of 3)
The data shows eliminating delta chains makes the changelog part of
changegroup generation slower. This is expected since we now have to
compute deltas for revisions where we could recycle the delta before.
It is worth putting this regression into context of overall changegroup
times. Here is the rough total CPU time spent in changegroup generation
for various repos while using delta chains on the changelog:
Repo CPU Time (s) CPU Time w/ compression
hg 4.50 7.05
mozilla-central 111.1 222.0
pypy 28.68 75.5
Before compression, removing delta chains from the changegroup adds
~4.4% overhead to hg changegroup generation, 1.3% to mozilla-central,
and 2.0% to pypy. When you factor in zlib compression, these percentages
are roughly divided by 2.
While the increased CPU usage for changegroup generation is unfortunate,
I think it is acceptable because the percentage is small, server
operators (those likely impacted most by this) have other mechanisms
to mitigate CPU consumption (namely reducing zlib compression level and
pre-generated clone bundles), and because there is room to optimize this
in the future. For example, we could use the nullid as the base revision,
effectively encoding the full revision for each entry in the changegroup.
When doing this, `hg perfchangegroupchangelog` nearly halves:
mozilla-unified
! wall 21.168075 comb 21.170000 user 21.130000 sys 0.040000 (best of 3)
! wall 11.196461 comb 11.200000 user 11.190000 sys 0.010000 (best of 3)
This looks very promising as a future optimization opportunity.
It's worth that the changes in test-acl.t to the changegroup part size.
This is because revision 6 in the changegroup had a delta chain of
length 2 before and after this patch the base revision is nullrev.
When the base revision is nullrev, cg2packer.deltaparent() hardcodes
the *previous* revision from the changegroup as the delta parent.
This caused the delta in the changegroup to switch base revisions,
the delta to change, and the size to change accordingly. While the
size increased in this case, I think sizes will remain the same
on average, as the delta base for changelog revisions doesn't matter
too much (as this patch shows). So, I don't consider this a regression.
author | Gregory Szorc <gregory.szorc@gmail.com> |
---|---|
date | Thu, 13 Oct 2016 12:50:27 +0200 |
parents | 34165875fa5d |
children | f1b63ec4b987 |
line wrap: on
line source
Test histedit extension: Fold commands ====================================== This test file is dedicated to testing the fold command in non conflicting case. Initialization --------------- $ . "$TESTDIR/histedit-helpers.sh" $ cat >> $HGRCPATH <<EOF > [alias] > logt = log --template '{rev}:{node|short} {desc|firstline}\n' > [extensions] > histedit= > EOF Simple folding -------------------- $ initrepo () > { > hg init r > cd r > for x in a b c d e f ; do > echo $x > $x > hg add $x > hg ci -m $x > done > } $ initrepo log before edit $ hg logt --graph @ 5:652413bf663e f | o 4:e860deea161a e | o 3:055a42cdd887 d | o 2:177f92b77385 c | o 1:d2ae7f538514 b | o 0:cb9a9f314b8b a $ hg histedit 177f92b77385 --commands - 2>&1 <<EOF | fixbundle > pick e860deea161a e > pick 652413bf663e f > fold 177f92b77385 c > pick 055a42cdd887 d > EOF log after edit $ hg logt --graph @ 4:9c277da72c9b d | o 3:6de59d13424a f | o 2:ee283cb5f2d5 e | o 1:d2ae7f538514 b | o 0:cb9a9f314b8b a post-fold manifest $ hg manifest a b c d e f check histedit_source $ hg log --debug --rev 3 changeset: 3:6de59d13424a8a13acd3e975514aed29dd0d9b2d phase: draft parent: 2:ee283cb5f2d5955443f23a27b697a04339e9a39a parent: -1:0000000000000000000000000000000000000000 manifest: 3:81eede616954057198ead0b2c73b41d1f392829a user: test date: Thu Jan 01 00:00:00 1970 +0000 files+: c f extra: branch=default extra: histedit_source=a4f7421b80f79fcc59fff01bcbf4a53d127dd6d3,177f92b773850b59254aa5e923436f921b55483b description: f *** c rollup will fold without preserving the folded commit's message $ OLDHGEDITOR=$HGEDITOR $ HGEDITOR=false $ hg histedit d2ae7f538514 --commands - 2>&1 <<EOF | fixbundle > pick d2ae7f538514 b > roll ee283cb5f2d5 e > pick 6de59d13424a f > pick 9c277da72c9b d > EOF $ HGEDITOR=$OLDHGEDITOR log after edit $ hg logt --graph @ 3:c4a9eb7989fc d | o 2:8e03a72b6f83 f | o 1:391ee782c689 b | o 0:cb9a9f314b8b a description is taken from rollup target commit $ hg log --debug --rev 1 changeset: 1:391ee782c68930be438ccf4c6a403daedbfbffa5 phase: draft parent: 0:cb9a9f314b8b07ba71012fcdbc544b5a4d82ff5b parent: -1:0000000000000000000000000000000000000000 manifest: 1:b5e112a3a8354e269b1524729f0918662d847c38 user: test date: Thu Jan 01 00:00:00 1970 +0000 files+: b e extra: branch=default extra: histedit_source=d2ae7f538514cd87c17547b0de4cea71fe1af9fb,ee283cb5f2d5955443f23a27b697a04339e9a39a description: b check saving last-message.txt $ cat > $TESTTMP/abortfolding.py <<EOF > from mercurial import util > def abortfolding(ui, repo, hooktype, **kwargs): > ctx = repo[kwargs.get('node')] > if set(ctx.files()) == set(['c', 'd', 'f']): > return True # abort folding commit only > ui.warn('allow non-folding commit\\n') > EOF $ cat > .hg/hgrc <<EOF > [hooks] > pretxncommit.abortfolding = python:$TESTTMP/abortfolding.py:abortfolding > EOF $ cat > $TESTTMP/editor.sh << EOF > echo "==== before editing" > cat \$1 > echo "====" > echo "check saving last-message.txt" >> \$1 > EOF $ rm -f .hg/last-message.txt $ hg status --rev '8e03a72b6f83^1::c4a9eb7989fc' A c A d A f $ HGEDITOR="sh $TESTTMP/editor.sh" hg histedit 8e03a72b6f83 --commands - 2>&1 <<EOF > pick 8e03a72b6f83 f > fold c4a9eb7989fc d > EOF allow non-folding commit ==== before editing f *** c *** d HG: Enter commit message. Lines beginning with 'HG:' are removed. HG: Leave message empty to abort commit. HG: -- HG: user: test HG: branch 'default' HG: added c HG: added d HG: added f ==== transaction abort! rollback completed abort: pretxncommit.abortfolding hook failed [255] $ cat .hg/last-message.txt f *** c *** d check saving last-message.txt $ cd .. $ rm -r r folding preserves initial author -------------------------------- $ initrepo $ hg ci --user "someone else" --amend --quiet tip before edit $ hg log --rev . changeset: 5:a00ad806cb55 tag: tip user: someone else date: Thu Jan 01 00:00:00 1970 +0000 summary: f $ hg --config progress.debug=1 --debug \ > histedit e860deea161a --commands - 2>&1 <<EOF | \ > egrep 'editing|unresolved' > pick e860deea161a e > fold a00ad806cb55 f > EOF editing: pick e860deea161a 4 e 1/2 changes (50.00%) editing: fold a00ad806cb55 5 f 2/2 changes (100.00%) tip after edit $ hg log --rev . changeset: 4:698d4e8040a1 tag: tip user: test date: Thu Jan 01 00:00:00 1970 +0000 summary: e $ cd .. $ rm -r r folding and creating no new change doesn't break: ------------------------------------------------- folded content is dropped during a merge. The folded commit should properly disappear. $ mkdir fold-to-empty-test $ cd fold-to-empty-test $ hg init $ printf "1\n2\n3\n" > file $ hg add file $ hg commit -m '1+2+3' $ echo 4 >> file $ hg commit -m '+4' $ echo 5 >> file $ hg commit -m '+5' $ echo 6 >> file $ hg commit -m '+6' $ hg logt --graph @ 3:251d831eeec5 +6 | o 2:888f9082bf99 +5 | o 1:617f94f13c0f +4 | o 0:0189ba417d34 1+2+3 $ hg histedit 1 --commands - << EOF > pick 617f94f13c0f 1 +4 > drop 888f9082bf99 2 +5 > fold 251d831eeec5 3 +6 > EOF 1 files updated, 0 files merged, 0 files removed, 0 files unresolved merging file warning: conflicts while merging file! (edit, then use 'hg resolve --mark') Fix up the change (fold 251d831eeec5) (hg histedit --continue to resume) [1] There were conflicts, we keep P1 content. This should effectively drop the changes from +6. $ hg status M file ? file.orig $ hg resolve -l U file $ hg revert -r 'p1()' file $ hg resolve --mark file (no more unresolved files) continue: hg histedit --continue $ hg histedit --continue 251d831eeec5: empty changeset saved backup bundle to $TESTTMP/*-backup.hg (glob) $ hg logt --graph @ 1:617f94f13c0f +4 | o 0:0189ba417d34 1+2+3 $ cd .. Test fold through dropped ------------------------- Test corner case where folded revision is separated from its parent by a dropped revision. $ hg init fold-with-dropped $ cd fold-with-dropped $ printf "1\n2\n3\n" > file $ hg commit -Am '1+2+3' adding file $ echo 4 >> file $ hg commit -m '+4' $ echo 5 >> file $ hg commit -m '+5' $ echo 6 >> file $ hg commit -m '+6' $ hg logt -G @ 3:251d831eeec5 +6 | o 2:888f9082bf99 +5 | o 1:617f94f13c0f +4 | o 0:0189ba417d34 1+2+3 $ hg histedit 1 --commands - << EOF > pick 617f94f13c0f 1 +4 > drop 888f9082bf99 2 +5 > fold 251d831eeec5 3 +6 > EOF 1 files updated, 0 files merged, 0 files removed, 0 files unresolved merging file warning: conflicts while merging file! (edit, then use 'hg resolve --mark') Fix up the change (fold 251d831eeec5) (hg histedit --continue to resume) [1] $ cat > file << EOF > 1 > 2 > 3 > 4 > 5 > EOF $ hg resolve --mark file (no more unresolved files) continue: hg histedit --continue $ hg commit -m '+5.2' created new head $ echo 6 >> file $ HGEDITOR=cat hg histedit --continue +4 *** +5.2 *** +6 HG: Enter commit message. Lines beginning with 'HG:' are removed. HG: Leave message empty to abort commit. HG: -- HG: user: test HG: branch 'default' HG: changed file saved backup bundle to $TESTTMP/fold-with-dropped/.hg/strip-backup/55c8d8dc79ce-4066cd98-backup.hg (glob) saved backup bundle to $TESTTMP/fold-with-dropped/.hg/strip-backup/617f94f13c0f-a35700fc-backup.hg (glob) $ hg logt -G @ 1:10c647b2cdd5 +4 | o 0:0189ba417d34 1+2+3 $ hg export tip # HG changeset patch # User test # Date 0 0 # Thu Jan 01 00:00:00 1970 +0000 # Node ID 10c647b2cdd54db0603ecb99b2ff5ce66d5a5323 # Parent 0189ba417d34df9dda55f88b637dcae9917b5964 +4 *** +5.2 *** +6 diff -r 0189ba417d34 -r 10c647b2cdd5 file --- a/file Thu Jan 01 00:00:00 1970 +0000 +++ b/file Thu Jan 01 00:00:00 1970 +0000 @@ -1,3 +1,6 @@ 1 2 3 +4 +5 +6 $ cd .. Folding with initial rename (issue3729) --------------------------------------- $ hg init fold-rename $ cd fold-rename $ echo a > a.txt $ hg add a.txt $ hg commit -m a $ hg rename a.txt b.txt $ hg commit -m rename $ echo b >> b.txt $ hg commit -m b $ hg logt --follow b.txt 2:e0371e0426bc b 1:1c4f440a8085 rename 0:6c795aa153cb a $ hg histedit 1c4f440a8085 --commands - 2>&1 << EOF | fixbundle > pick 1c4f440a8085 rename > fold e0371e0426bc b > EOF $ hg logt --follow b.txt 1:cf858d235c76 rename 0:6c795aa153cb a $ cd .. Folding with swapping --------------------- This is an excuse to test hook with histedit temporary commit (issue4422) $ hg init issue4422 $ cd issue4422 $ echo a > a.txt $ hg add a.txt $ hg commit -m a $ echo b > b.txt $ hg add b.txt $ hg commit -m b $ echo c > c.txt $ hg add c.txt $ hg commit -m c $ hg logt 2:a1a953ffb4b0 c 1:199b6bb90248 b 0:6c795aa153cb a Setup the proper environment variable symbol for the platform, to be subbed into the hook command. #if windows $ NODE="%HG_NODE%" #else $ NODE="\$HG_NODE" #endif $ hg histedit 6c795aa153cb --config hooks.commit="echo commit $NODE" --commands - 2>&1 << EOF | fixbundle > pick 199b6bb90248 b > fold a1a953ffb4b0 c > pick 6c795aa153cb a > EOF commit 9599899f62c05f4377548c32bf1c9f1a39634b0c $ hg logt 1:9599899f62c0 a 0:79b99e9c8e49 b $ echo "foo" > amended.txt $ hg add amended.txt $ hg ci -q --config extensions.largefiles= --amend -I amended.txt Test that folding multiple changes in a row doesn't show multiple editors. $ echo foo >> foo $ hg add foo $ hg ci -m foo1 $ echo foo >> foo $ hg ci -m foo2 $ echo foo >> foo $ hg ci -m foo3 $ hg logt 4:21679ff7675c foo3 3:b7389cc4d66e foo2 2:0e01aeef5fa8 foo1 1:578c7455730c a 0:79b99e9c8e49 b $ cat > "$TESTTMP/editor.sh" <<EOF > echo ran editor >> "$TESTTMP/editorlog.txt" > cat \$1 >> "$TESTTMP/editorlog.txt" > echo END >> "$TESTTMP/editorlog.txt" > echo merged foos > \$1 > EOF $ HGEDITOR="sh \"$TESTTMP/editor.sh\"" hg histedit 1 --commands - 2>&1 <<EOF | fixbundle > pick 578c7455730c 1 a > pick 0e01aeef5fa8 2 foo1 > fold b7389cc4d66e 3 foo2 > fold 21679ff7675c 4 foo3 > EOF $ hg logt 2:e8bedbda72c1 merged foos 1:578c7455730c a 0:79b99e9c8e49 b Editor should have run only once $ cat $TESTTMP/editorlog.txt ran editor foo1 *** foo2 *** foo3 HG: Enter commit message. Lines beginning with 'HG:' are removed. HG: Leave message empty to abort commit. HG: -- HG: user: test HG: branch 'default' HG: added foo END $ cd ..