view tests/test-rebase-bookmarks.t @ 38732:be4984261611

merge: mark file gets as not thread safe (issue5933)

In default installs, this has the effect of disabling the thread-based
worker on Windows when manifesting files in the working directory. My
measurements have shown that with revlog-based repositories, Mercurial
spends a lot of CPU time in revlog code resolving file data. This ends
up incurring a lot of context switching across threads and slows down
`hg update` operations when going from an empty working directory to
the tip of the repo.

On mozilla-unified (246,351 files) on an i7-6700K (4+4 CPUs):

before: 487s wall
after:  360s wall (equivalent to worker.enabled=false)
cpus=2: 379s wall

Even with only 2 threads, the thread pool is still slower.

The introduction of the thread-based worker (02b36e860e0b) states that
it resulted in a "~50%" speedup for `hg sparse --enable-profile` and
`hg sparse --disable-profile`. This disagrees with my measurement
above. I theorize a few reasons for this:

1) Removal of files from the working directory is I/O - not CPU - bound
   and should benefit from a thread pool (unless I/O is insanely fast
   and the GIL release is near instantaneous). So tests like `hg sparse
   --enable-profile` may exercise deletion throughput and aren't good
   benchmarks for worker tasks that are CPU heavy.
2) The patch was authored by someone at Facebook. The results were
   likely measured against a repository using remotefilelog. And I
   believe that revision retrieval during working directory updates with
   remotefilelog will often use a remote store, thus being I/O and not
   CPU bound. This probably resulted in an overstated performance gain.

Since there appears to be a need to enable the thread-based worker with
some stores, I've made the flagging of file gets as thread safe
configurable. I've made it experimental because I don't want to formalize
a boolean flag for this option and because this attribute is best
captured against the store implementation. But we don't have a proper
store API for this yet. I'd rather cross this bridge later.

It is possible there are revlog-based repositories that do benefit from
a thread-based worker. I didn't do very comprehensive testing. If there
are, we may want to devise a more proper algorithm for whether to use
the thread-based worker, including possibly config options to limit the
number of threads to use. But until I see evidence that justifies
complexity, simplicity wins.

Differential Revision: https://phab.mercurial-scm.org/D3963

  $ cat >> $HGRCPATH <<EOF
  > [extensions]
  > rebase=
  > drawdag=$TESTDIR/drawdag.py
  > 
  > [phases]
  > publish=False
  > 
  > [alias]
  > tglog = log -G --template "{rev}: {node|short} '{desc}' bookmarks: {bookmarks}\n"
  > EOF

Create a repo with several bookmarks
  $ hg init a
  $ cd a

  $ echo a > a
  $ hg ci -Am A
  adding a

  $ echo b > b
  $ hg ci -Am B
  adding b
  $ hg book 'X'
  $ hg book 'Y'

  $ echo c > c
  $ hg ci -Am C
  adding c
  $ hg book 'Z'

  $ hg up -q 0

  $ echo d > d
  $ hg ci -Am D
  adding d
  created new head

  $ hg book W

  $ hg tglog
  @  3: 41acb9dca9eb 'D' bookmarks: W
  |
  | o  2: 49cb3485fa0c 'C' bookmarks: Y Z
  | |
  | o  1: 6c81ed0049f8 'B' bookmarks: X
  |/
  o  0: 1994f17a630e 'A' bookmarks:
  

Move only rebased bookmarks

  $ cd ..
  $ hg clone -q a a1

  $ cd a1
  $ hg up -q Z

Test deleting divergent bookmarks from dest (issue3685)

  $ hg book -r 3 Z@diverge

... and also test that bookmarks not on dest or not being moved aren't deleted

  $ hg book -r 3 X@diverge
  $ hg book -r 0 Y@diverge

  $ hg tglog
  o  3: 41acb9dca9eb 'D' bookmarks: W X@diverge Z@diverge
  |
  | @  2: 49cb3485fa0c 'C' bookmarks: Y Z
  | |
  | o  1: 6c81ed0049f8 'B' bookmarks: X
  |/
  o  0: 1994f17a630e 'A' bookmarks: Y@diverge
  
  $ hg rebase -s Y -d 3
  rebasing 2:49cb3485fa0c "C" (Y Z)
  saved backup bundle to $TESTTMP/a1/.hg/strip-backup/49cb3485fa0c-126f3e97-rebase.hg

  $ hg tglog
  @  3: 17fb3faba63c 'C' bookmarks: Y Z
  |
  o  2: 41acb9dca9eb 'D' bookmarks: W X@diverge
  |
  | o  1: 6c81ed0049f8 'B' bookmarks: X
  |/
  o  0: 1994f17a630e 'A' bookmarks: Y@diverge
  
Do not try to keep active but deleted divergent bookmark

  $ cd ..
  $ hg clone -q a a4

  $ cd a4
  $ hg up -q 2
  $ hg book W@diverge

  $ hg rebase -s W -d .
  rebasing 3:41acb9dca9eb "D" (W tip)
  saved backup bundle to $TESTTMP/a4/.hg/strip-backup/41acb9dca9eb-b35a6a63-rebase.hg

  $ hg bookmarks
     W                         3:0d3554f74897
     X                         1:6c81ed0049f8
     Y                         2:49cb3485fa0c
     Z                         2:49cb3485fa0c

Keep bookmarks to the correct rebased changeset

  $ cd ..
  $ hg clone -q a a2

  $ cd a2
  $ hg up -q Z

  $ hg rebase -s 1 -d 3
  rebasing 1:6c81ed0049f8 "B" (X)
  rebasing 2:49cb3485fa0c "C" (Y Z)
  saved backup bundle to $TESTTMP/a2/.hg/strip-backup/6c81ed0049f8-a687065f-rebase.hg

  $ hg tglog
  @  3: 3d5fa227f4b5 'C' bookmarks: Y Z
  |
  o  2: e926fccfa8ec 'B' bookmarks: X
  |
  o  1: 41acb9dca9eb 'D' bookmarks: W
  |
  o  0: 1994f17a630e 'A' bookmarks:
  

Keep active bookmark on the correct changeset

  $ cd ..
  $ hg clone -q a a3

  $ cd a3
  $ hg up -q X

  $ hg rebase -d W
  rebasing 1:6c81ed0049f8 "B" (X)
  rebasing 2:49cb3485fa0c "C" (Y Z)
  saved backup bundle to $TESTTMP/a3/.hg/strip-backup/6c81ed0049f8-a687065f-rebase.hg

  $ hg tglog
  o  3: 3d5fa227f4b5 'C' bookmarks: Y Z
  |
  @  2: e926fccfa8ec 'B' bookmarks: X
  |
  o  1: 41acb9dca9eb 'D' bookmarks: W
  |
  o  0: 1994f17a630e 'A' bookmarks:
  
  $ hg bookmarks
     W                         1:41acb9dca9eb
   * X                         2:e926fccfa8ec
     Y                         3:3d5fa227f4b5
     Z                         3:3d5fa227f4b5

rebase --continue with bookmarks present (issue3802)

  $ hg up 2
  0 files updated, 0 files merged, 0 files removed, 0 files unresolved
  (leaving bookmark X)
  $ echo 'C' > c
  $ hg add c
  $ hg ci -m 'other C'
  created new head
  $ hg up 3
  1 files updated, 0 files merged, 0 files removed, 0 files unresolved
  $ hg rebase --dest 4
  rebasing 3:3d5fa227f4b5 "C" (Y Z)
  merging c
  warning: conflicts while merging c! (edit, then use 'hg resolve --mark')
  unresolved conflicts (see hg resolve, then hg rebase --continue)
  [1]
  $ echo 'c' > c
  $ hg resolve --mark c
  (no more unresolved files)
  continue: hg rebase --continue
  $ hg rebase --continue
  rebasing 3:3d5fa227f4b5 "C" (Y Z)
  saved backup bundle to $TESTTMP/a3/.hg/strip-backup/3d5fa227f4b5-c6ea2371-rebase.hg
  $ hg tglog
  @  4: 45c0f0ec1203 'C' bookmarks: Y Z
  |
  o  3: b0e10b7175fd 'other C' bookmarks:
  |
  o  2: e926fccfa8ec 'B' bookmarks: X
  |
  o  1: 41acb9dca9eb 'D' bookmarks: W
  |
  o  0: 1994f17a630e 'A' bookmarks:
  

ensure that bookmarks given the names of revset functions can be used
as --rev arguments (issue3950)

  $ hg update -q 3
  $ echo bimble > bimble
  $ hg add bimble
  $ hg commit -q -m 'bisect'
  $ echo e >> bimble
  $ hg ci -m bisect2
  $ echo e >> bimble
  $ hg ci -m bisect3
  $ hg book bisect
  $ hg update -q Y
  $ hg rebase -r '"bisect"^^::"bisect"^' -r bisect -d Z
  rebasing 5:345c90f326a4 "bisect"
  rebasing 6:f677a2907404 "bisect2"
  rebasing 7:325c16001345 "bisect3" (bisect tip)
  saved backup bundle to $TESTTMP/a3/.hg/strip-backup/345c90f326a4-b4840586-rebase.hg

Bookmark and working parent get moved even if --keep is set (issue5682)

  $ hg init $TESTTMP/book-keep
  $ cd $TESTTMP/book-keep
  $ hg debugdrawdag <<'EOS'
  > B C
  > |/
  > A
  > EOS
  $ eval `hg tags -T 'hg bookmark -ir {node} {tag};\n' | grep -v tip`
  $ rm .hg/localtags
  $ hg up -q B
  $ hg tglog
  o  2: dc0947a82db8 'C' bookmarks: C
  |
  | @  1: 112478962961 'B' bookmarks: B
  |/
  o  0: 426bada5c675 'A' bookmarks: A
  
  $ hg rebase -r B -d C --keep
  rebasing 1:112478962961 "B" (B)
  $ hg tglog
  @  3: 9769fc65c4c5 'B' bookmarks: B
  |
  o  2: dc0947a82db8 'C' bookmarks: C
  |
  | o  1: 112478962961 'B' bookmarks:
  |/
  o  0: 426bada5c675 'A' bookmarks: A