sparse-revlog: implement algorithm to write sparse delta chains (
issue5480)
The classic behavior of revlog._isgooddeltainfo is to consider the span size
of the whole delta chain, and limit it to 4 * textlen.
Once sparse-revlog writing is allowed (and enforced with a requirement),
revlog._isgooddeltainfo considers the span of the largest chunk as the
distance used in the verification, instead of using the span of the whole
delta chain.
In order to compute the span of the largest chunk, we need to slice into
chunks a chain with the new revision at the top of the revlog, and take the
maximal span of these chunks. The sparse read density is a parameter to the
slicing, as it will stop when the global read density reaches this threshold.
For instance, a density of 50% means that 2 of 4 read bytes are actually used
for the reconstruction of the revision (the others are part of other chains).
This allows a new revision to be potentially stored with a diff against
another revision anywhere in the history, instead of forcing it in the last 4
* textlen. The result is a much better compression on repositories that have
many concurrent branches. Here are a comparison between using deltas from
current upstream (aggressive-merge-deltas on by default) and deltas from a
sparse-revlog
Comparison of `.hg/store/` size:
mercurial (6.74% merges):
before: 46,831,873 bytes
after: 46,795,992 bytes (no relevant change)
pypy (8.30% merges):
before: 333,524,651 bytes
after: 308,417,511 bytes -8%
netbeans (34.21% merges):
before: 1,141,847,554 bytes
after: 1,131,093,161 bytes -1%
mozilla-central (4.84% merges):
before: 2,344,248,850 bytes
after: 2,328,459,258 bytes -1%
large-private-repo-A (merge 19.73%)
before: 41,510,550,163 bytes
after: 8,121,763,428 bytes -80%
large-private-repo-B (23.77%)
before: 58,702,221,709 bytes
after: 8,351,588,828 bytes -76%
Comparison of `00manifest.d` size:
mercurial (6.74% merges):
before: 6,143,044 bytes
after: 6,107,163 bytes
pypy (8.30% merges):
before: 52,941,780 bytes
after: 27,834,082 bytes -48%
netbeans (34.21% merges):
before: 130,088,982 bytes
after: 119,337,636 bytes -10%
mozilla-central (4.84% merges):
before: 215,096,339 bytes
after: 199,496,863 bytes -8%
large-private-repo-A (merge 19.73%)
before: 33,725,285,081 bytes
after: 390,302,545 bytes -99%
large-private-repo-B (23.77%)
before: 49,457,701,645 bytes
after: 1,366,752,187 bytes -97%
The better delta chains provide a performance boost in relevant repositories:
pypy, bundling 1000 revisions:
before: 1.670s
after: 1.149s -31%
Unbundling got a bit slower. probably because the sparse algorithm is still
pure
python.
pypy, unbundling 1000 revisions:
before: 4.062s
after: 4.507s +10%
Performance of bundle/unbundle in repository with few concurrent branches (eg:
mercurial) are unaffected.
No significant differences have been noticed then timing `hg push` and `hg
pull` locally. More state timings are being gathered.
Same as for aggressive-merge-delta, better delta comes with longer delta
chains. Longer chains have a performance impact. For example. The length of
the chain needed to get the manifest of pypy's tip moves from 82 item to 1929
items. This moves the restore time from 3.88ms to 11.3ms.
Delta chain length is an independent issue that affects repository without
this changes. It will be dealt with independently.
No significant differences have been observed on repositories where
`sparse-revlog` have not much effect (mercurial, unity, netbeans). On pypy,
small differences have been observed on some operation affected by delta chain
building and retrieval.
pypy, perfmanifest
before: 0.006162s
after: 0.017899s +190%
pypy, commit:
before: 0.382
after: 0.376 -1%
pypy, status:
before: 0.157
after: 0.168 +7%
More comprehensive and stable timing comparisons are in progress.
$ cat >> $HGRCPATH << EOF
> [extensions]
> show =
> EOF
$ hg init repo0
$ cd repo0
Empty repo / no checkout results in error
$ hg show stack
abort: stack view only available when there is a working directory
[255]
Stack displays single draft changeset as root revision
$ echo 0 > foo
$ hg -q commit -A -m 'commit 0'
$ hg show stack
@ 9f17 commit 0
Stack displays multiple draft changesets
$ echo 1 > foo
$ hg commit -m 'commit 1'
$ echo 2 > foo
$ hg commit -m 'commit 2'
$ echo 3 > foo
$ hg commit -m 'commit 3'
$ echo 4 > foo
$ hg commit -m 'commit 4'
$ hg show stack
@ 2737 commit 4
o d1a6 commit 3
o 128c commit 2
o 181c commit 1
o 9f17 commit 0
Public parent of draft base is displayed, separated from stack
$ hg phase --public -r 0
$ hg show stack
@ 2737 commit 4
o d1a6 commit 3
o 128c commit 2
o 181c commit 1
/ (stack base)
o 9f17 commit 0
$ hg phase --public -r 1
$ hg show stack
@ 2737 commit 4
o d1a6 commit 3
o 128c commit 2
/ (stack base)
o 181c commit 1
Draft descendants are shown
$ hg -q up 2
$ hg show stack
o 2737 commit 4
o d1a6 commit 3
@ 128c commit 2
/ (stack base)
o 181c commit 1
$ hg -q up 3
$ hg show stack
o 2737 commit 4
@ d1a6 commit 3
o 128c commit 2
/ (stack base)
o 181c commit 1
working dir on public changeset should display special message
$ hg -q up 1
$ hg show stack
(empty stack; working directory parent is a published changeset)
Branch point in descendants displayed at top of graph
$ hg -q up 3
$ echo b > foo
$ hg commit -m 'commit 5 (new dag branch)'
created new head
$ hg -q up 2
$ hg show stack
\ / (multiple children)
|
o d1a6 commit 3
@ 128c commit 2
/ (stack base)
o 181c commit 1
$ cd ..
Base is stopped at merges
$ hg init merge-base
$ cd merge-base
$ echo 0 > foo
$ hg -q commit -A -m initial
$ echo h1 > foo
$ hg commit -m 'head 1'
$ hg -q up 0
$ echo h2 > foo
$ hg -q commit -m 'head 2'
$ hg phase --public -r 0:tip
$ hg -q up 1
$ hg merge -t :local 2
0 files updated, 1 files merged, 0 files removed, 0 files unresolved
(branch merge, don't forget to commit)
$ hg commit -m 'merge heads'
TODO doesn't yet handle case where wdir is a draft merge
$ hg show stack
@ 8ee9 merge heads
/ (stack base)
o 5947 head 1
$ echo d1 > foo
$ hg commit -m 'draft 1'
$ echo d2 > foo
$ hg commit -m 'draft 2'
$ hg show stack
@ 430d draft 2
o 787b draft 1
/ (stack base)
o 8ee9 merge heads
$ cd ..
Now move on to stacks when there are more commits after the base branchpoint
$ hg init public-rebase
$ cd public-rebase
$ echo 0 > foo
$ hg -q commit -A -m 'base'
$ hg phase --public -r .
$ echo d1 > foo
$ hg commit -m 'draft 1'
$ echo d2 > foo
$ hg commit -m 'draft 2'
$ hg -q up 0
$ echo 1 > foo
$ hg commit -m 'new 1'
created new head
$ echo 2 > foo
$ hg commit -m 'new 2'
$ hg -q up 2
Newer draft heads don't impact output
$ hg show stack
@ eaff draft 2
o 2b21 draft 1
/ (stack base)
o b66b base
Newer public heads are rendered
$ hg phase --public -r '::tip'
$ hg show stack
o baa4 new 2
/ (2 commits ahead)
:
: (stack head)
: @ eaff draft 2
: o 2b21 draft 1
:/ (stack base)
o b66b base
If rebase is available, we show a hint how to rebase to that head
$ hg --config extensions.rebase= show stack
o baa4 new 2
/ (2 commits ahead; hg rebase --source 2b21 --dest baa4)
:
: (stack head)
: @ eaff draft 2
: o 2b21 draft 1
:/ (stack base)
o b66b base
Similar tests but for multiple heads
$ hg -q up 0
$ echo h2 > foo
$ hg -q commit -m 'new head 2'
$ hg phase --public -r .
$ hg -q up 2
$ hg show stack
o baa4 new 2
/ (2 commits ahead)
: o 9a84 new head 2
:/ (1 commits ahead)
:
: (stack head)
: @ eaff draft 2
: o 2b21 draft 1
:/ (stack base)
o b66b base
$ hg --config extensions.rebase= show stack
o baa4 new 2
/ (2 commits ahead; hg rebase --source 2b21 --dest baa4)
: o 9a84 new head 2
:/ (1 commits ahead; hg rebase --source 2b21 --dest 9a84)
:
: (stack head)
: @ eaff draft 2
: o 2b21 draft 1
:/ (stack base)
o b66b base