view tests/test-merge-partial-tool.t @ 51681:522b4d729e89

mmap: populate the mapping by default

Without pre-population, accessing all data through a mmap can result in many
pagefault, reducing performance significantly. If the mmap is prepopulated, the
performance can no longer get slower than a full read.
(See benchmark number below)

In some cases were very few data is read, prepopulating can be overkill and
slower than populating on access (through page fault). So that behavior can be
controlled when the caller can pre-determine the best behavior.
(See benchmark number below)

In addition, testing with populating in a secondary thread yield great result
combining the best of each approach. This might be implemented in later
changesets.

In all cases, using mmap has a great effect on memory usage when many processes
run in parallel on the same machine.

### Benchmarks

# What did I run

A couple of month back I ran a large benchmark campaign to assess the impact of
various approach for using mmap with the revlog (and other files), it
highlighted a few benchmarks that capture the impact of the changes well. So to
validate this change I checked the following:

- log command displaying various revisions
  (read the changelog index)
- log command displaying the patch of listed revisions
  (read the changelog index, the manifest index and a few files indexes)
- unbundling a few revisions
  (read and write changelog, manifest and few files indexes, and walk the graph
  to update some cache)
- pushing a few revisions
  (read and write changelog, manifest and few files indexes, walk the graph to
  update some cache, performs various accesses locally and remotely during
  discovery)

Benchmarks were run using the default module policy (c+py) and the rust one. No
significant difference were found between the two implementation, so we will
present result using the default policy (unless otherwise specified).

I ran them on a few repositories :

- mercurial: a "public changeset only" copy of mercurial from 2018-08-01 using
  zstd compression and sparse-revlog

- pypy: a copy of pypy from 2018-08-01 using zstd compression and sparse-revlog

- netbeans: a copy of netbeans from 2018-08-01 using zstd compression and
  sparse-revlog

- mozilla-try: a copy of mozilla-try from 2019-02-18 using zstd compression and
  sparse-revlog

- mozilla-try persistent-nodemap: Same as the above but with a persistent
  nodemap. Used for the log --patch benchmark only

# Results

For the smaller repositories (mercurial, pypy), the impact of mmap is almost
imperceptible, other cost dominating the operation. The impact of prepopulating
is undiscernible in the benchmark we ran.

For larger repositories the benchmark support explanation given above:

On netbeans, the log can be about 1% faster without repopulation (for a
difference < 100ms) but unbundle becomes a bit slower, even when small.

### data-env-vars.name                             = netbeans-2018-08-01-zstd-sparse-revlog
  # benchmark.name                                 = hg.command.unbundle
  # benchmark.variants.issue6528                   = disabled
  # benchmark.variants.reuse-external-delta-parent = yes
  # benchmark.variants.revs                        = any-1-extra-rev
  # benchmark.variants.source                      = unbundle
  # benchmark.variants.verbosity                   = quiet
with-populate: 0.240157
no-populate:   0.265087  (+10.38%, +0.02)
  # benchmark.variants.revs                        = any-100-extra-rev
with-populate: 1.459518
no-populate:   1.481290   (+1.49%, +0.02)
 ## benchmark.name                                 = hg.command.push
  # benchmark.variants.explicit-rev                = none
  # benchmark.variants.issue6528                   = disabled
  # benchmark.variants.protocol                    = ssh
  # benchmark.variants.reuse-external-delta-parent = yes
  # benchmark.variants.revs                        = any-1-extra-rev
with-populate: 0.771919
no-populate:   0.792025   (+2.60%, +0.02)
  # benchmark.variants.revs                        = any-100-extra-rev
with-populate: 1.459518
no-populate:   1.481290   (+1.49%, +0.02)


For mozilla-try, the "slow down" from pre-populate for small `hg log` is more
visible, but still small in absolute time. (using rust value for the persistent
nodemap value to be relevant).

### data-env-vars.name            = mozilla-try-2019-02-18-ds2-pnm
  # benchmark.name                = hg.command.log
  # bin-env-vars.hg.flavor        = rust
  # benchmark.variants.patch      = yes
  # benchmark.variants.limit-rev  = 1
with-populate: 0.237813
no-populate:   0.229452   (-3.52%, -0.01)
  # benchmark.variants.limit-rev  = 10
  # benchmark.variants.patch      = yes
with-populate: 1.213578
no-populate:   1.205189
### data-env-vars.name            = mozilla-try-2019-02-18-zstd-sparse-revlog
  # benchmark.variants.limit-rev  = 1000
  # benchmark.variants.patch      = no
  # benchmark.variants.rev        = tip
with-populate: 0.198607
no-populate:   0.195038   (-1.80%, -0.00)

However pre-populating provide a significant boost on more complex operations
like unbundle or push:

### data-env-vars.name                             = mozilla-try-2019-02-18-zstd-sparse-revlog
  # benchmark.name                                 = hg.command.push
  # benchmark.variants.explicit-rev                = none
  # benchmark.variants.issue6528                   = disabled
  # benchmark.variants.protocol                    = ssh
  # benchmark.variants.reuse-external-delta-parent = yes
  # benchmark.variants.revs                        = any-1-extra-rev
with-populate: 4.798632
no-populate:   4.953295   (+3.22%, +0.15)
  # benchmark.variants.revs                        = any-100-extra-rev
with-populate: 4.903618
no-populate:   5.014963   (+2.27%, +0.11)
 ## benchmark.name                                 = hg.command.unbundle
  # benchmark.variants.revs                        = any-1-extra-rev
with-populate: 1.423411
no-populate:   1.585365  (+11.38%, +0.16)
  # benchmark.variants.revs                        = any-100-extra-rev
with-populate: 1.537909
no-populate:   1.688489   (+9.79%, +0.15)

Test support for partial-resolution tools

Create a tool that resolves conflicts after line 5 by simply dropping those
lines (even if there are no conflicts there)
  $ cat >> "$TESTTMP/head.sh" <<'EOF'
  > #!/bin/sh
  > for f in "$@"; do
  >   head -5 $f > tmp
  >   mv -f tmp $f
  > done
  > EOF
  $ chmod +x "$TESTTMP/head.sh"
...and another tool that keeps only the last 5 lines instead of the first 5.
  $ cat >> "$TESTTMP/tail.sh" <<'EOF'
  > #!/bin/sh
  > for f in "$@"; do
  >   tail -5 $f > tmp
  >   mv -f tmp $f
  > done
  > EOF
  $ chmod +x "$TESTTMP/tail.sh"

Set up both tools to run on all patterns (the default), and let the `tail` tool
run after the `head` tool, which means it will have no effect (we'll override it
to test order later)
  $ cat >> "$HGRCPATH" <<EOF
  > [partial-merge-tools]
  > head.executable=$TESTTMP/head.sh
  > tail.executable=$TESTTMP/tail.sh
  > tail.order=1
  > EOF

  $ make_commit() {
  >   echo "$@" | xargs -n1 > file
  >   hg add file 2> /dev/null
  >   hg ci -m "$*"
  > }


Let a partial-resolution tool resolve some conflicts and leave other conflicts
for the regular merge tool (:merge3 here)

  $ hg init repo
  $ cd repo
  $ make_commit a b c d e f
  $ make_commit a b2 c d e f2
  $ hg up 0
  1 files updated, 0 files merged, 0 files removed, 0 files unresolved
  $ make_commit a b3 c d e f3
  created new head
  $ hg merge 1 -t :merge3
  merging file
  warning: conflicts while merging file! (edit, then use 'hg resolve --mark')
  0 files updated, 0 files merged, 0 files removed, 1 files unresolved
  use 'hg resolve' to retry unresolved file merges or 'hg merge --abort' to abandon
  [1]
  $ cat file
  a
  <<<<<<< working copy:    e11a49d4b620 - test: a b3 c d e f3
  b3
  ||||||| common ancestor: 8ae8bb9cc43a - test: a b c d e f
  b
  =======
  b2
  >>>>>>> merge rev:       fbc096a40cc5 - test: a b2 c d e f2
  c
  d
  e


With premerge=keep, the partial-resolution tools runs before and doesn't see
the conflict markers

  $ hg up -C 2
  1 files updated, 0 files merged, 0 files removed, 0 files unresolved
  $ cat >> .hg/hgrc <<EOF
  > [merge-tools]
  > my-local.executable = cat
  > my-local.args = $local
  > my-local.premerge = keep-merge3
  > EOF
  $ hg merge 1 -t my-local
  merging file
  0 files updated, 1 files merged, 0 files removed, 0 files unresolved
  (branch merge, don't forget to commit)
  $ cat file
  a
  <<<<<<< working copy:    e11a49d4b620 - test: a b3 c d e f3
  b3
  ||||||| common ancestor: 8ae8bb9cc43a - test: a b c d e f
  b
  =======
  b2
  >>>>>>> merge rev:       fbc096a40cc5 - test: a b2 c d e f2
  c
  d
  e


When a partial-resolution tool resolves all conflicts, the resolution should
be recorded and the regular merge tool should not be invoked for the file.

  $ hg up -C 0
  1 files updated, 0 files merged, 0 files removed, 0 files unresolved
  $ make_commit a b c d e f2
  created new head
  $ hg up 0
  1 files updated, 0 files merged, 0 files removed, 0 files unresolved
  $ make_commit a b c d e f3
  created new head
  $ hg merge 3 -t false
  merging file
  0 files updated, 1 files merged, 0 files removed, 0 files unresolved
  (branch merge, don't forget to commit)
  $ cat file
  a
  b
  c
  d
  e


Can disable all partial merge tools (the `head` tool would have resolved this
conflict it had been enabled)

  $ hg up -C 4
  1 files updated, 0 files merged, 0 files removed, 0 files unresolved
  $ hg merge 3 -t :merge3 --config merge.disable-partial-tools=yes
  merging file
  warning: conflicts while merging file! (edit, then use 'hg resolve --mark')
  0 files updated, 0 files merged, 0 files removed, 1 files unresolved
  use 'hg resolve' to retry unresolved file merges or 'hg merge --abort' to abandon
  [1]
  $ cat file
  a
  b
  c
  d
  e
  <<<<<<< working copy:    d57edaa6e21a - test: a b c d e f3
  f3
  ||||||| common ancestor: 8ae8bb9cc43a - test: a b c d e f
  f
  =======
  f2
  >>>>>>> merge rev:       8c217da987be - test: a b c d e f2


Can disable one partial merge tool (the `head` tool would have resolved this
conflict it had been enabled)

  $ hg up -C 4
  1 files updated, 0 files merged, 0 files removed, 0 files unresolved
  $ hg merge 3 -t :merge3 --config partial-merge-tools.head.disable=yes
  merging file
  warning: conflicts while merging file! (edit, then use 'hg resolve --mark')
  0 files updated, 0 files merged, 0 files removed, 1 files unresolved
  use 'hg resolve' to retry unresolved file merges or 'hg merge --abort' to abandon
  [1]
  $ cat file
  b
  c
  d
  e
  <<<<<<< working copy:    d57edaa6e21a - test: a b c d e f3
  f3
  ||||||| common ancestor: 8ae8bb9cc43a - test: a b c d e f
  f
  =======
  f2
  >>>>>>> merge rev:       8c217da987be - test: a b c d e f2


Only tools whose patterns match are run. We make `head` not match here, so
only `tail` should run

  $ hg up -C 4
  1 files updated, 0 files merged, 0 files removed, 0 files unresolved
  $ hg merge 3 -t :merge3 --config partial-merge-tools.head.patterns=other
  merging file
  warning: conflicts while merging file! (edit, then use 'hg resolve --mark')
  0 files updated, 0 files merged, 0 files removed, 1 files unresolved
  use 'hg resolve' to retry unresolved file merges or 'hg merge --abort' to abandon
  [1]
  $ cat file
  b
  c
  d
  e
  <<<<<<< working copy:    d57edaa6e21a - test: a b c d e f3
  f3
  ||||||| common ancestor: 8ae8bb9cc43a - test: a b c d e f
  f
  =======
  f2
  >>>>>>> merge rev:       8c217da987be - test: a b c d e f2


Filesets can be used to select which files to run partial merge tools on.

  $ hg up -C 4
  1 files updated, 0 files merged, 0 files removed, 0 files unresolved
  $ hg merge 3 -t :merge3 --config partial-merge-tools.head.patterns=set:other
  merging file
  warning: conflicts while merging file! (edit, then use 'hg resolve --mark')
  0 files updated, 0 files merged, 0 files removed, 1 files unresolved
  use 'hg resolve' to retry unresolved file merges or 'hg merge --abort' to abandon
  [1]
  $ cat file
  b
  c
  d
  e
  <<<<<<< working copy:    d57edaa6e21a - test: a b c d e f3
  f3
  ||||||| common ancestor: 8ae8bb9cc43a - test: a b c d e f
  f
  =======
  f2
  >>>>>>> merge rev:       8c217da987be - test: a b c d e f2


If there are several matching tools, they are run in requested order. We move
`head` after `tail` in order here so it has no effect (the conflict in "f" thus
remains).

  $ hg up -C 4
  1 files updated, 0 files merged, 0 files removed, 0 files unresolved
  $ hg merge 3 -t :merge3 --config partial-merge-tools.head.order=2
  merging file
  warning: conflicts while merging file! (edit, then use 'hg resolve --mark')
  0 files updated, 0 files merged, 0 files removed, 1 files unresolved
  use 'hg resolve' to retry unresolved file merges or 'hg merge --abort' to abandon
  [1]
  $ cat file
  b
  c
  d
  e
  <<<<<<< working copy:    d57edaa6e21a - test: a b c d e f3
  f3
  ||||||| common ancestor: 8ae8bb9cc43a - test: a b c d e f
  f
  =======
  f2
  >>>>>>> merge rev:       8c217da987be - test: a b c d e f2


When using "nomerge" tools (e.g. `:other`), the partial-resolution tools
should not be run.

  $ hg up -C 4
  1 files updated, 0 files merged, 0 files removed, 0 files unresolved
  $ hg merge 3 -t :other
  0 files updated, 1 files merged, 0 files removed, 0 files unresolved
  (branch merge, don't forget to commit)
  $ cat file
  a
  b
  c
  d
  e
  f2


If a partial-resolution tool resolved some conflict and simplemerge can
merge the rest, then the regular merge tool should not be used. Here we merge
"a b c d e3 f3" with "a b2 c d e f2". The `head` tool resolves the conflict in
"f" and the internal simplemerge merges the remaining changes in "b" and "e".

  $ hg up -C 0
  1 files updated, 0 files merged, 0 files removed, 0 files unresolved
  $ make_commit a b c d e3 f3
  created new head
  $ hg merge 1 -t false
  merging file
  0 files updated, 1 files merged, 0 files removed, 0 files unresolved
  (branch merge, don't forget to commit)
  $ cat file
  a
  b2
  c
  d
  e3

Test that arguments get passed as expected.

  $ cat >> "$TESTTMP/log-args.sh" <<'EOF'
  > #!/bin/sh
  > echo "$@" > args.log
  > EOF
  $ chmod +x "$TESTTMP/log-args.sh"
  $ cat >> "$HGRCPATH" <<EOF
  > [partial-merge-tools]
  > log-args.executable=$TESTTMP/log-args.sh
  > EOF
  $ hg up -C 2
  1 files updated, 0 files merged, 0 files removed, 0 files unresolved
  $ hg merge 1
  merging file
  warning: conflicts while merging file! (edit, then use 'hg resolve --mark')
  0 files updated, 0 files merged, 0 files removed, 1 files unresolved
  use 'hg resolve' to retry unresolved file merges or 'hg merge --abort' to abandon
  [1]
  $ cat args.log
  */hgmerge-*/file~local */hgmerge-*/file~base */hgmerge-*/file~other (glob)
  $ hg up -C 2
  1 files updated, 0 files merged, 0 files removed, 0 files unresolved
  $ hg merge 1 --config partial-merge-tools.log-args.args='--other $other $base --foo --local $local --also-other $other'
  merging file
  warning: conflicts while merging file! (edit, then use 'hg resolve --mark')
  0 files updated, 0 files merged, 0 files removed, 1 files unresolved
  use 'hg resolve' to retry unresolved file merges or 'hg merge --abort' to abandon
  [1]
  $ cat args.log
  --other */hgmerge-*/file~other */hgmerge-*/file~base --foo --local */hgmerge-*/file~local --also-other */hgmerge-*/file~other (glob)