tests/test-rename-merge1.t

revlog: use an LRU cache for delta chain bases

Profiling using statprof revealed a hotspot during changegroup
application calculating delta chain bases on generaldelta repos.
Essentially, revlog._addrevision() was performing a lot of redundant
work tracing the delta chain as part of determining when the chain
distance was acceptable. This was most pronounced when adding
revisions to manifests, which can have delta chains thousands of
revisions long.

There was a delta chain base cache on revlogs before, but it only
captured a single revision. This was acceptable before generaldelta,
when _addrevision would build deltas from the previous revision and
thus we'd pretty much guarantee a cache hit when resolving the delta
chain base on a subsequent _addrevision call. However, it isn't
suitable for generaldelta because parent revisions aren't necessarily
the last processed revision.

This patch converts the delta chain base cache to an LRU dict cache.
The cache can hold multiple entries, so generaldelta repos have a
higher chance of getting a cache hit.

The impact of this change when processing changegroup additions is
significant. On a generaldelta conversion of the "mozilla-unified"
repo (which contains heads of the main Firefox repositories in
chronological order - this means there are lots of transitions between
heads in revlog order), this change has the following impact when
performing an `hg unbundle` of an uncompressed bundle of the repo:

before: 5:42 CPU time
after:  4:34 CPU time

Most of this time is saved when applying the changelog and manifest
revlogs:

before: 2:30 CPU time
after:  1:17 CPU time

That nearly a 50% reduction in CPU time applying changesets and
manifests!

Applying a gzipped bundle of the same repo (effectively simulating a
`hg clone` over HTTP) showed a similar speedup:

before: 5:53 CPU time
after:  4:46 CPU time

Wall time improvements were basically the same as CPU time.

I didn't measure explicitly, but it feels like most of the time
is saved when processing manifests. This makes sense, as large
manifests tend to have very long delta chains and thus benefit the
most from this cache.

So, this change effectively makes changegroup application (which is
used by `hg unbundle`, `hg clone`, `hg pull`, `hg unshelve`, and
various other commands) significantly faster when delta chains are
long (which can happen on repos with large numbers of files and thus
large manifests).

In theory, this change can result in more memory utilization. However,
we're caching a dict of ints. At most we have 200 ints + Python object
overhead per revlog. And, the cache is really only populated when
performing read-heavy operations, such as adding changegroups or
scanning an individual revlog. For memory bloat to be an issue, we'd
need to scan/read several revisions from several revlogs all while
having active references to several revlogs. I don't think there are
many operations that do this, so I don't think memory bloat from the
cache will be an issue.

  $ hg init

  $ echo "[merge]" >> .hg/hgrc
  $ echo "followcopies = 1" >> .hg/hgrc

  $ echo foo > a
  $ echo foo > a2
  $ hg add a a2
  $ hg ci -m "start"

  $ hg mv a b
  $ hg mv a2 b2
  $ hg ci -m "rename"

  $ hg co 0
  2 files updated, 0 files merged, 2 files removed, 0 files unresolved

  $ echo blahblah > a
  $ echo blahblah > a2
  $ hg mv a2 c2
  $ hg ci -m "modify"
  created new head

  $ hg merge -y --debug
    searching for copies back to rev 1
    unmatched files in local:
     c2
    unmatched files in other:
     b
     b2
    all copies found (* = to merge, ! = divergent, % = renamed and deleted):
     src: 'a' -> dst: 'b' *
     src: 'a2' -> dst: 'b2' !
     src: 'a2' -> dst: 'c2' !
    checking for directory renames
  resolving manifests
   branchmerge: True, force: False, partial: False
   ancestor: af1939970a1c, local: 044f8520aeeb+, remote: 85c198ef2f6c
  note: possible conflict - a2 was renamed multiple times to:
   c2
   b2
   preserving a for resolve of b
  removing a
   b2: remote created -> g
  getting b2
   b: remote moved from a -> m (premerge)
  picked tool ':merge' for b (binary False symlink False changedelete False)
  merging a and b to b
  my b@044f8520aeeb+ other b@85c198ef2f6c ancestor a@af1939970a1c
   premerge successful
  1 files updated, 1 files merged, 0 files removed, 0 files unresolved
  (branch merge, don't forget to commit)

  $ hg status -AC
  M b
    a
  M b2
  R a
  C c2

  $ cat b
  blahblah

  $ hg ci -m "merge"

  $ hg debugindex b
     rev    offset  length  ..... linkrev nodeid       p1           p2 (re)
       0         0      67  .....       1 57eacc201a7f 000000000000 000000000000 (re)
       1        67      72  .....       3 4727ba907962 000000000000 57eacc201a7f (re)

  $ hg debugrename b
  b renamed from a:dd03b83622e78778b403775d0d074b9ac7387a66

This used to trigger a "divergent renames" warning, despite no renames

  $ hg cp b b3
  $ hg cp b b4
  $ hg ci -A -m 'copy b twice'
  $ hg up eb92d88a9712
  0 files updated, 0 files merged, 2 files removed, 0 files unresolved
  $ hg up
  2 files updated, 0 files merged, 0 files removed, 0 files unresolved
  $ hg rm b3 b4
  $ hg ci -m 'clean up a bit of our mess'

We'd rather not warn on divergent renames done in the same changeset (issue2113)

  $ hg cp b b3
  $ hg mv b b4
  $ hg ci -A -m 'divergent renames in same changeset'
  $ hg up c761c6948de0
  1 files updated, 0 files merged, 2 files removed, 0 files unresolved
  $ hg up
  2 files updated, 0 files merged, 1 files removed, 0 files unresolved

Check for issue2642

  $ hg init t
  $ cd t

  $ echo c0 > f1
  $ hg ci -Aqm0

  $ hg up null -q
  $ echo c1 > f1 # backport
  $ hg ci -Aqm1
  $ hg mv f1 f2
  $ hg ci -qm2

  $ hg up 0 -q
  $ hg merge 1 -q --tool internal:local
  $ hg ci -qm3

  $ hg merge 2
  merging f1 and f2 to f2
  0 files updated, 1 files merged, 0 files removed, 0 files unresolved
  (branch merge, don't forget to commit)

  $ cat f2
  c0

  $ cd ..

Check for issue2089

  $ hg init repo2089
  $ cd repo2089

  $ echo c0 > f1
  $ hg ci -Aqm0

  $ hg up null -q
  $ echo c1 > f1
  $ hg ci -Aqm1

  $ hg up 0 -q
  $ hg merge 1 -q --tool internal:local
  $ echo c2 > f1
  $ hg ci -qm2

  $ hg up 1 -q
  $ hg mv f1 f2
  $ hg ci -Aqm3

  $ hg up 2 -q
  $ hg merge 3
  merging f1 and f2 to f2
  0 files updated, 1 files merged, 0 files removed, 0 files unresolved
  (branch merge, don't forget to commit)

  $ cat f2
  c2

  $ cd ..

Check for issue3074

  $ hg init repo3074
  $ cd repo3074
  $ echo foo > file
  $ hg add file
  $ hg commit -m "added file"
  $ hg mv file newfile
  $ hg commit -m "renamed file"
  $ hg update 0
  1 files updated, 0 files merged, 1 files removed, 0 files unresolved
  $ hg rm file
  $ hg commit -m "deleted file"
  created new head
  $ hg merge --debug
    searching for copies back to rev 1
    unmatched files in other:
     newfile
    all copies found (* = to merge, ! = divergent, % = renamed and deleted):
     src: 'file' -> dst: 'newfile' %
    checking for directory renames
  resolving manifests
   branchmerge: True, force: False, partial: False
   ancestor: 19d7f95df299, local: 0084274f6b67+, remote: 5d32493049f0
  note: possible conflict - file was deleted and renamed to:
   newfile
   newfile: remote created -> g
  getting newfile
  1 files updated, 0 files merged, 0 files removed, 0 files unresolved
  (branch merge, don't forget to commit)
  $ hg status
  M newfile
  $ cd ..