view tests/test-dirstate-race.t @ 48068:bf8837e3d7ce

dirstate: Remove the flat Rust DirstateMap implementation Before this changeset we had two Rust implementations of `DirstateMap`. This removes the "flat" DirstateMap so that the "tree" DirstateMap is always used when Rust enabled. This simplifies the code a lot, and will enable (in the next changeset) further removal of a trait abstraction. This is a performance regression when: * Rust is enabled, and * The repository uses the legacy dirstate-v1 file format, and * For `hg status`, unknown files are not listed (such as with `-mard`) The regression is about 100 milliseconds for `hg status -mard` on a semi-large repository (mozilla-central), from ~320ms to ~420ms. We deem this to be small enough to be worth it. The new dirstate-v2 is still experimental at this point, but we aim to stabilize it (though not yet enable it by default for new repositories) in Mercurial 6.0. Eventually, upgrating repositories to dirsate-v2 will eliminate this regression (and enable other performance improvements). # Background The flat DirstateMap was introduced with the first Rust implementation of the status algorithm. It works similarly to the previous Python + C one, with a single `HashMap` that associates file paths to a `DirstateEntry` (where Python has a dict). We later added the tree DirstateMap where the root of the tree contains nodes for files and directories that are directly at the root of the repository, and nodes for directories can contain child nodes representing the files and directly that *they* contain directly. The shape of this tree mirrors that of the working directory in the filesystem. This enables the status algorithm to traverse this tree in tandem with traversing the filesystem tree, which in turns enables a more efficient algorithm. Furthermore, the new dirstate-v2 file format is also based on a tree of the same shape. The tree DirstateMap can access a dirstate-v2 file without parsing it: binary data in a single large (possibly memory-mapped) bytes buffer is traversed on demand. This allows `DirstateMap` creation to take `O(1)` time. (Mutation works by creating new in-memory nodes with copy-on-write semantics, and serialization is append-mostly.) The tradeoff is that for "legacy" repositories that use the dirstate-v1 file format, parsing that file into a tree DirstateMap takes more time. Profiling shows that this time is dominated by `HashMap`. For a dirstate containing `F` files with an average `D` directory depth, the flat DirstateMap does parsing in `O(F)` number of HashMap operations but the tree DirstateMap in `O(F × D)` operations, since each node has its own HashMap containing its child nodes. This slower costs ~140ms on an old snapshot of mozilla-central, and ~80ms on an old snapshot of the Netbeans repository. The status algorithm is faster, but with `-mard` (when not listing unknown files) it is typically not faster *enough* to compensate the slower parsing. Both Rust implementations are always faster than the Python + C implementation # Benchmark results All benchmarks are run on changeset 98c0408324e6, with repositories that use the dirstate-v1 file format, on a server with 4 CPU cores and 4 CPU threads (no HyperThreading). `hg status` benchmarks show wall clock times of the entire command as the average and standard deviation of serveral runs, collected by https://github.com/sharkdp/hyperfine and reformated. Parsing benchmarks are wall clock time of the Rust function that converts a bytes buffer of the dirstate file into the `DirstateMap` data structure as used by the status algorithm. A single run each, collected by running `hg status` this environment variable: RUST_LOG=hg::dirstate::dirstate_map=trace,hg::dirstate_tree::dirstate_map=trace Benchmark 1: Rust flat DirstateMap → Rust tree DirstateMap hg status mozilla-clean 562.3 ms ± 2.0 ms → 462.5 ms ± 0.6 ms 1.22 ± 0.00 times faster mozilla-dirty 859.6 ms ± 2.2 ms → 719.5 ms ± 3.2 ms 1.19 ± 0.01 times faster mozilla-ignored 558.2 ms ± 3.0 ms → 457.9 ms ± 2.9 ms 1.22 ± 0.01 times faster mozilla-unknowns 859.4 ms ± 5.7 ms → 716.0 ms ± 4.7 ms 1.20 ± 0.01 times faster netbeans-clean 336.5 ms ± 0.9 ms → 339.5 ms ± 0.4 ms 0.99 ± 0.00 times faster netbeans-dirty 491.4 ms ± 1.6 ms → 475.1 ms ± 1.2 ms 1.03 ± 0.00 times faster netbeans-ignored 343.7 ms ± 1.0 ms → 347.8 ms ± 0.4 ms 0.99 ± 0.00 times faster netbeans-unknowns 484.3 ms ± 1.0 ms → 466.0 ms ± 1.2 ms 1.04 ± 0.00 times faster hg status -mard mozilla-clean 317.3 ms ± 0.6 ms → 422.5 ms ± 1.2 ms 0.75 ± 0.00 times faster mozilla-dirty 315.4 ms ± 0.6 ms → 417.7 ms ± 1.1 ms 0.76 ± 0.00 times faster mozilla-ignored 314.6 ms ± 0.6 ms → 417.4 ms ± 1.0 ms 0.75 ± 0.00 times faster mozilla-unknowns 312.9 ms ± 0.9 ms → 417.3 ms ± 1.6 ms 0.75 ± 0.00 times faster netbeans-clean 212.0 ms ± 0.6 ms → 283.6 ms ± 0.8 ms 0.75 ± 0.00 times faster netbeans-dirty 211.4 ms ± 1.0 ms → 283.4 ms ± 1.6 ms 0.75 ± 0.01 times faster netbeans-ignored 211.4 ms ± 0.9 ms → 283.9 ms ± 0.8 ms 0.74 ± 0.01 times faster netbeans-unknowns 211.1 ms ± 0.6 ms → 283.4 ms ± 1.0 ms 0.74 ± 0.00 times faster Parsing mozilla-clean 38.4ms → 177.6ms mozilla-dirty 38.8ms → 177.0ms mozilla-ignored 38.8ms → 178.0ms mozilla-unknowns 38.7ms → 176.9ms netbeans-clean 16.5ms → 97.3ms netbeans-dirty 16.5ms → 98.4ms netbeans-ignored 16.9ms → 97.4ms netbeans-unknowns 16.9ms → 96.3ms Benchmark 2: Python + C dirstatemap → Rust tree DirstateMap hg status mozilla-clean 1261.0 ms ± 3.6 ms → 461.1 ms ± 0.5 ms 2.73 ± 0.00 times faster mozilla-dirty 2293.4 ms ± 9.1 ms → 719.6 ms ± 3.6 ms 3.19 ± 0.01 times faster mozilla-ignored 1240.4 ms ± 2.3 ms → 457.7 ms ± 1.9 ms 2.71 ± 0.00 times faster mozilla-unknowns 2283.3 ms ± 9.0 ms → 719.7 ms ± 3.8 ms 3.17 ± 0.01 times faster netbeans-clean 879.7 ms ± 3.5 ms → 339.9 ms ± 0.5 ms 2.59 ± 0.00 times faster netbeans-dirty 1257.3 ms ± 4.7 ms → 474.6 ms ± 1.6 ms 2.65 ± 0.01 times faster netbeans-ignored 943.9 ms ± 1.9 ms → 347.3 ms ± 1.1 ms 2.72 ± 0.00 times faster netbeans-unknowns 1188.1 ms ± 5.0 ms → 465.2 ms ± 2.3 ms 2.55 ± 0.01 times faster hg status -mard mozilla-clean 903.2 ms ± 3.6 ms → 423.4 ms ± 2.2 ms 2.13 ± 0.01 times faster mozilla-dirty 884.6 ms ± 4.5 ms → 417.3 ms ± 1.4 ms 2.12 ± 0.01 times faster mozilla-ignored 881.9 ms ± 1.3 ms → 417.3 ms ± 0.8 ms 2.11 ± 0.00 times faster mozilla-unknowns 878.5 ms ± 1.9 ms → 416.4 ms ± 0.9 ms 2.11 ± 0.00 times faster netbeans-clean 434.9 ms ± 1.8 ms → 284.0 ms ± 0.8 ms 1.53 ± 0.01 times faster netbeans-dirty 434.1 ms ± 0.8 ms → 283.1 ms ± 0.8 ms 1.53 ± 0.00 times faster netbeans-ignored 431.7 ms ± 1.1 ms → 283.6 ms ± 1.8 ms 1.52 ± 0.01 times faster netbeans-unknowns 433.0 ms ± 1.3 ms → 283.5 ms ± 0.7 ms 1.53 ± 0.00 times faster Differential Revision: https://phab.mercurial-scm.org/D11516
author Simon Sapin <simon.sapin@octobus.net>
date Mon, 27 Sep 2021 12:09:15 +0200
parents 6763913fa175
children b4f83c9e7905
line wrap: on
line source

#testcases dirstate-v1 dirstate-v2

#if dirstate-v2
#require rust
  $ echo '[format]' >> $HGRCPATH
  $ echo 'exp-dirstate-v2=1' >> $HGRCPATH
#endif

  $ hg init repo
  $ cd repo
  $ echo a > a
  $ hg add a
  $ hg commit -m test

Do we ever miss a sub-second change?:

  $ for i in 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20; do
  >     hg co -qC 0
  >     echo b > a
  >     hg st
  > done
  M a
  M a
  M a
  M a
  M a
  M a
  M a
  M a
  M a
  M a
  M a
  M a
  M a
  M a
  M a
  M a
  M a
  M a
  M a
  M a

  $ echo test > b
  $ mkdir dir1
  $ echo test > dir1/c
  $ echo test > d

  $ echo test > e
#if execbit
A directory will typically have the execute bit -- make sure it doesn't get
confused with a file with the exec bit set
  $ chmod +x e
#endif

  $ hg add b dir1 d e
  adding dir1/c
  $ hg commit -m test2

  $ cat >> $TESTTMP/dirstaterace.py << EOF
  > from mercurial import (
  >     context,
  >     extensions,
  > )
  > def extsetup(ui):
  >     extensions.wrapfunction(context.workingctx, '_checklookup', overridechecklookup)
  > def overridechecklookup(orig, self, files):
  >     # make an update that changes the dirstate from underneath
  >     self._repo.ui.system(br"sh '$TESTTMP/dirstaterace.sh'",
  >                          cwd=self._repo.root)
  >     return orig(self, files)
  > EOF

  $ hg debugrebuilddirstate
  $ hg debugdirstate
  n   0         -1 unset               a
  n   0         -1 unset               b
  n   0         -1 unset               d
  n   0         -1 unset               dir1/c
  n   0         -1 unset               e

XXX Note that this returns M for files that got replaced by directories. This is
definitely a bug, but the fix for that is hard and the next status run is fine
anyway.

  $ cat > $TESTTMP/dirstaterace.sh <<EOF
  > rm b && rm -r dir1 && rm d && mkdir d && rm e && mkdir e
  > EOF

  $ hg status --config extensions.dirstaterace=$TESTTMP/dirstaterace.py
  M d
  M e
  ! b
  ! dir1/c
  $ hg debugdirstate
  n 644          2 * a (glob)
  n   0         -1 unset               b
  n   0         -1 unset               d
  n   0         -1 unset               dir1/c
  n   0         -1 unset               e

  $ hg status
  ! b
  ! d
  ! dir1/c
  ! e

  $ rmdir d e
  $ hg update -C -q .

Test that dirstate changes aren't written out at the end of "hg
status", if .hg/dirstate is already changed simultaneously before
acquisition of wlock in workingctx._poststatusfixup().

This avoidance is important to keep consistency of dirstate in race
condition (see issue5584 for detail).

  $ hg parents -q
  1:* (glob)

  $ hg debugrebuilddirstate
  $ hg debugdirstate
  n   0         -1 unset               a
  n   0         -1 unset               b
  n   0         -1 unset               d
  n   0         -1 unset               dir1/c
  n   0         -1 unset               e

  $ cat > $TESTTMP/dirstaterace.sh <<EOF
  > # This script assumes timetable of typical issue5584 case below:
  > #
  > # 1. "hg status" loads .hg/dirstate
  > # 2. "hg status" confirms clean-ness of FILE
  > # 3. "hg update -C 0" updates the working directory simultaneously
  > #    (FILE is removed, and FILE is dropped from .hg/dirstate)
  > # 4. "hg status" acquires wlock
  > #    (.hg/dirstate is re-loaded = no FILE entry in dirstate)
  > # 5. "hg status" marks FILE in dirstate as clean
  > #    (FILE entry is added to in-memory dirstate)
  > # 6. "hg status" writes dirstate changes into .hg/dirstate
  > #    (FILE entry is written into .hg/dirstate)
  > #
  > # To reproduce similar situation easily and certainly, #2 and #3
  > # are swapped.  "hg cat" below ensures #2 on "hg status" side.
  > 
  > hg update -q -C 0
  > hg cat -r 1 b > b
  > EOF

"hg status" below should excludes "e", of which exec flag is set, for
portability of test scenario, because unsure but missing "e" is
treated differently in _checklookup() according to runtime platform.

- "missing(!)" on POSIX, "pctx[f].cmp(self[f])" raises ENOENT
- "modified(M)" on Windows, "self.flags(f) != pctx.flags(f)" is True

  $ hg status --config extensions.dirstaterace=$TESTTMP/dirstaterace.py --debug -X path:e
  skip updating dirstate: identity mismatch
  M a
  ! d
  ! dir1/c

  $ hg parents -q
  0:* (glob)
  $ hg files
  a
  $ hg debugdirstate
  n * * * a (glob)

  $ rm b

#if fsmonitor

Create fsmonitor state.

  $ hg status
  $ f --type .hg/fsmonitor.state
  .hg/fsmonitor.state: file

Test that invalidating fsmonitor state in the middle (which doesn't require the
wlock) causes the fsmonitor update to be skipped.
hg debugrebuilddirstate ensures that the dirstaterace hook will be called, but
it also invalidates the fsmonitor state. So back it up and restore it.

  $ mv .hg/fsmonitor.state .hg/fsmonitor.state.tmp
  $ hg debugrebuilddirstate
  $ mv .hg/fsmonitor.state.tmp .hg/fsmonitor.state

  $ cat > $TESTTMP/dirstaterace.sh <<EOF
  > rm .hg/fsmonitor.state
  > EOF

  $ hg status --config extensions.dirstaterace=$TESTTMP/dirstaterace.py --debug
  skip updating fsmonitor.state: identity mismatch
  $ f .hg/fsmonitor.state
  .hg/fsmonitor.state: file not found

#endif

Set up a rebase situation for issue5581.

  $ echo c2 > a
  $ echo c2 > b
  $ hg add b
  $ hg commit -m c2
  created new head
  $ echo c3 >> a
  $ hg commit -m c3
  $ hg update 2
  1 files updated, 0 files merged, 0 files removed, 0 files unresolved
  $ echo c4 >> a
  $ echo c4 >> b
  $ hg commit -m c4
  created new head

Configure a merge tool that runs status in the middle of the rebase. The goal of
the status call is to trigger a potential bug if fsmonitor's state is written
even though the wlock is held by another process. The output of 'hg status' in
the merge tool goes to /dev/null because we're more interested in the results of
'hg status' run after the rebase.

  $ cat >> $TESTTMP/mergetool-race.sh << EOF
  > echo "custom merge tool"
  > printf "c2\nc3\nc4\n" > \$1
  > hg --cwd "$TESTTMP/repo" status > /dev/null
  > echo "custom merge tool end"
  > EOF
  $ cat >> $HGRCPATH << EOF
  > [extensions]
  > rebase =
  > [merge-tools]
  > test.executable=sh
  > test.args=$TESTTMP/mergetool-race.sh \$output
  > EOF

  $ hg rebase -s . -d 3 --tool test
  rebasing 4:b08445fd6b2a tip "c4"
  merging a
  custom merge tool
  custom merge tool end
  saved backup bundle to $TESTTMP/repo/.hg/strip-backup/* (glob)

This hg status should be empty, whether or not fsmonitor is enabled (issue5581).

  $ hg status