Martin von Zweigbergk <martinvonz@google.com> [Fri, 21 Jun 2019 09:33:57 -0700] rev 42521
drawdag: don't crash when writing copy info to changesets
When writing copies to the changeset, localrepo.commitctx() will call
ctx.p1copies() and ctx.p2copies(). These crashed on simplecommitctx
because they ended up trying to access the manifest. drawdag doesn't
support copies at all, so we can simply override the methods to return
empty dicts.
Differential Revision: https://phab.mercurial-scm.org/D6565
Martin von Zweigbergk <martinvonz@google.com> [Fri, 21 Jun 2019 23:35:04 -0700] rev 42520
merge with stable
Martin von Zweigbergk <martinvonz@google.com> [Wed, 19 Jun 2019 10:19:32 -0700] rev 42519
log: pass getcopies() function instead of getrenamed() to displayer (API)
This reduces the duplication between the two displayer functions (and
between them and scmutil.getcopiesfn()). It's still more code than two
patches ago, but there's less duplication.
Differential Revision: https://phab.mercurial-scm.org/D6546
Martin von Zweigbergk <martinvonz@google.com> [Wed, 19 Jun 2019 09:59:45 -0700] rev 42518
copies: create helper for getting all copies for changeset
There are a few places where we get all the copies for a changeset (at
least the {file_copies} template and in two places in `hg log
--copies` code). These places currently call scmutil.getrenamedfn() to
get a caching "getrenamed" function. They all use it in a similar
way. We will be able to reuse more code by having a function for
getting all the copies for a changeset. This patch introduces such a
function. It uses it in the {file_copies} template to show that it
works. It relies on the existing scmutil.getrenamedfn() for caching in
the filelog-centric case.
Differential Revision: https://phab.mercurial-scm.org/D6545
Martin von Zweigbergk <martinvonz@google.com> [Tue, 18 Jun 2019 23:19:24 -0700] rev 42517
logcmdutil: also check for copies in null revision and working copy
It's safe (and fast) to look for copies in the null revision, and it's
incorrect not to look for them in the working copy, so let's look in
both places.
Differential Revision: https://phab.mercurial-scm.org/D6544
Martin von Zweigbergk <martinvonz@google.com> [Tue, 18 Jun 2019 23:23:30 -0700] rev 42516
tests: demonstrate missing copy information in working copy with graphlog
Differential Revision: https://phab.mercurial-scm.org/D6543
Martin von Zweigbergk <martinvonz@google.com> [Wed, 19 Jun 2019 10:33:13 -0700] rev 42515
remotefilelog: handle copies in changesets in getrenamedfn() override
E.g. the {file_copies} template keyword didn't work with copies in
changesets before this patch because remotefilelog overrides the
getrenamedfn() and didn't handle the changeset-centric case.
Differential Revision: https://phab.mercurial-scm.org/D6542
Martin von Zweigbergk <martinvonz@google.com> [Wed, 19 Jun 2019 11:12:06 -0700] rev 42514
remotefilelog: check if RFL is enabled in getrenamedfn() override
In 8a0e03f7baf4 (remotefilelog: move most setup from onetimesetup() to
uisetup(), 2019-05-01), I said:
All the wrappers moved in this patch check if remotefilelog is enabled
before they change behavior, so it's safe to always wrap.
That was clearly a lie, because getrenamedfn() didn't. That made
e.g. `hg log -T {file_copies}` unbearably slow. This patch fixes that.
Differential Revision: https://phab.mercurial-scm.org/D6541
Martin von Zweigbergk <martinvonz@google.com> [Tue, 18 Jun 2019 08:55:23 -0700] rev 42513
relnotes: document template support for `hg root`
Differential Revision: https://phab.mercurial-scm.org/D6540
Augie Fackler <augie@google.com> [Tue, 18 Jun 2019 09:57:06 -0400] rev 42512
remotefilelog: tell runbgcommand to not block on child process startup
These two invocations will always find a binary because they're
re-running hg. As a result, we can skip waiting for the subprocess to
start running and save a little bit of wall-time.
Differential Revision: https://phab.mercurial-scm.org/D6539
Augie Fackler <augie@google.com> [Tue, 18 Jun 2019 09:43:27 -0400] rev 42511
procutil: allow callers of runbgcommand to assume the process starts
Experimentally starting the subprocess can take as much as 40ms, and
for some of our use cases that's frivolous: we know the binary will
start, and if it doesn't we'd only ever ignore it and continue
anyway. This lets those use cases be faster.
Differential Revision: https://phab.mercurial-scm.org/D6537
Augie Fackler <augie@google.com> [Tue, 18 Jun 2019 09:58:01 -0400] rev 42510
shallowrepo: remove backwards compat code that predates in-tree remotefilelog
Differential Revision: https://phab.mercurial-scm.org/D6538
Sushil khanchi <sushilkhanchi97@gmail.com> [Tue, 16 Apr 2019 02:53:28 +0530] rev 42509
commit: make the error message more specific while aborting branch closing
Differential Revision: https://phab.mercurial-scm.org/D6493
Sushil khanchi <sushilkhanchi97@gmail.com> [Tue, 16 Apr 2019 02:33:54 +0530] rev 42508
commit: add a check if it is trying to close an already closed branch head
It would check if the revision we are going to close is already a
closed branch head and print the error message accordingly.
Differential Revision: https://phab.mercurial-scm.org/D6491
Martin von Zweigbergk <martinvonz@google.com> [Mon, 17 Jun 2019 10:53:00 -0700] rev 42507
strip: move checksubstate() to mq (its only caller)
Differential Revision: https://phab.mercurial-scm.org/D6536
Martin von Zweigbergk <martinvonz@google.com> [Mon, 17 Jun 2019 10:19:41 -0700] rev 42506
strip: use bailifchanged() instead of reimplementing it
This also means that we get the standard error messages (see changed
test cases).
Differential Revision: https://phab.mercurial-scm.org/D6535
Martin von Zweigbergk <martinvonz@google.com> [Mon, 17 Jun 2019 10:40:24 -0700] rev 42505
strip: remove unused excsuffix argument from checklocalchanges()
It was only used by mq, and mq now has its own copy of the function.
Differential Revision: https://phab.mercurial-scm.org/D6534
Martin von Zweigbergk <martinvonz@google.com> [Mon, 17 Jun 2019 10:38:50 -0700] rev 42504
mq: remove dependency on strip's checklocalchanges()
Some of the functionality in strip.checklocalchanges() was only used
by mq, so let's move it to mq so we can clean up strip.
Differential Revision: https://phab.mercurial-scm.org/D6533
Martin von Zweigbergk <martinvonz@google.com> [Thu, 02 May 2019 23:39:33 -0700] rev 42503
copies: avoid calling matcher if matcher.always()
When storing copy information in the changesets
(experimental.copies.read-from=changeset-only), this patch speeds up
hg debugpathcopies FENNEC_58_0_2_BUILD1 FIREFOX_59_0b8_BUILD2
from 5.9s to 4.7s. At the start of this series (b162229e), that
command took 18min.
Differential Revision: https://phab.mercurial-scm.org/D6422
Martin von Zweigbergk <martinvonz@google.com> [Thu, 18 Apr 2019 21:21:44 -0700] rev 42502
copies: avoid unnecessary copying of copy dict
When storing copy information in the changesets, this patch speeds up
hg debugpathcopies FENNEC_58_0_2_BUILD1 FIREFOX_59_0b8_BUILD2
from 11s to 5.9s. That command takes 6.2s when storing copy
information in filelogs.
Differential Revision: https://phab.mercurial-scm.org/D6421
Martin von Zweigbergk <martinvonz@google.com> [Thu, 18 Apr 2019 21:22:14 -0700] rev 42501
copies: don't filter out copy targets created on other side of merge commit
If file X is copied to Y on one side of merge and the other side
creates Y (no copy), we would not mark that as copy. In the
changeset-centric pathcopies() version, that was done by checking if
the copy target existed on the other branch. Even though merge commits
are pretty uncommon, it still turned out to be too expensive to load
the manifest of the parents of merge commits. In a repo of
mozilla-unified converted to storing copies in changesets, about 2m30s
of `hg debugpathcopies FIREFOX_BETA_59_END FIREFOX_BETA_60_BASE` is
spent on this check of merge commits.
I tried to think of a way of storing more information in the
changesets in order to cheaply detect these cases, but I couldn't
think of a solution. So this patch simply removes those checks.
For reference, these extra copies are reported from the aforementioned
command after this patch:
browser/base/content/sanitize.js -> browser/modules/Sanitizer.jsm
testing/mozbase/mozprocess/tests/process_normal_finish_python.ini -> testing/mozbase/mozprocess/tests/process_normal_finish.ini
testing/mozbase/mozprocess/tests/process_waittimeout_python.ini -> testing/mozbase/mozprocess/tests/process_waittimeout.ini
testing/mozbase/mozprocess/tests/process_waittimeout_10s_python.ini -> testing/mozbase/mozprocess/tests/process_waittimeout_10s.ini
Since these copies were created on one side of some merge, it still
seems reasonable to include them, so I'm not even sure it's worse than
filelog pathcopies(), just different.
Differential Revision: https://phab.mercurial-scm.org/D6420
Martin von Zweigbergk <martinvonz@google.com> [Thu, 18 Apr 2019 00:40:53 -0700] rev 42500
copies: do full filtering at end of _changesetforwardcopies()
As mentioned earlier, pathcopies() is very slow when copies are stored
in the changeset. Most of the cost comes from calling _chain() for
every changeset, which is slow because it needs to read manifests. It
needs to read manifests to be able to filter out copies that are were
created in one commit and then deleted. (It also filters out copies
that were created from a file that didn't exist in the starting
revision, but that's a fixed revision across calls to _chain(), so
it's much cheaper.)
This patch changes from _chainandfilter() to just _chain() in the main
loop in _changesetforwardcopies(). It instead removes copies that have
subsequently been removed by using ctx.filesremoved(). We thus rely on
that to be fast.
It timed this command in mozilla-unified:
hg debugpathcopies FIREFOX_59_0b3_BUILD2 FIREFOX_BETA_59_END
It took 18s before and 1.1s after. It's still faster when copy
information is stored in filelogs: 0.70s. It also still gets slow when
there are merge commits involved, because we read manifests there
too. We'll deal with that later.
Differential Revision: https://phab.mercurial-scm.org/D6419
Yuya Nishihara <yuya@tcha.org> [Sat, 15 Jun 2019 10:58:53 +0900] rev 42499
rust-filepatterns: add comment about Windows path handling
As I replied to the Phabricator message, this is wrong. And I even suspect
it wouldn't compile because of multiple type mismatches.
I think, in Rust where type system is rock solid, we can live with UTF-8
strings except for the bottom storage layer and the top UI/command layer.
We'll still have to get around undecodable characters not to be lost, but
I think it's okay to drop such filenames from match result if they don't
match in UTF-8 world, not in Latin-1 world.
Yuya Nishihara <yuya@tcha.org> [Sat, 15 Jun 2019 10:35:53 +0900] rev 42498
rust-filepatterns: silence warning of non_upper_case_globals
Yuya Nishihara <yuya@tcha.org> [Sat, 15 Jun 2019 10:35:03 +0900] rev 42497
rust: update Cargo.lock to include @generated comment
cargo 1.34.0 of Debian sid inserts this comment, and I'm tired of reverting
the change every time I do make local.
https://github.com/rust-lang/cargo/commit/bd0e4a08471b8bc7957829b4fd294b8985d4fa2d
Augie Fackler <augie@google.com> [Mon, 17 Jun 2019 13:21:41 -0400] rev 42496
merge with stable
Matt Harbison <matt_harbison@yahoo.com> [Fri, 14 Jun 2019 00:30:33 -0400] rev 42495
lfs: correct an error in the TODO file
Matt Harbison <matt_harbison@yahoo.com> [Thu, 04 Oct 2018 00:57:11 -0400] rev 42494
cat: don't prefetch files unless the output requires it
It's a waste to cache lfs blobs when cat'ing the raw data at best, but a hassle
debugging when the blob is missing. I'm not sure if there are other commands
that have '{data}' for output, and if there's a general way to prefetch on that
keyword.
It's interesting that the verbose output seems to leak into the JSON output, but
that seems like an existing bug.
Augie Fackler <augie@google.com> [Wed, 12 Jun 2019 19:01:49 -0400] rev 42493
tracing: add support for emitting counters
Differential Revision: https://phab.mercurial-scm.org/D6526
Augie Fackler <augie@google.com> [Wed, 12 Jun 2019 19:01:37 -0400] rev 42492
tracing: extract tracing-active logic to separate function
I'm about to add support for counters, and want to avoid duplicating this
logic.
Differential Revision: https://phab.mercurial-scm.org/D6525
Augie Fackler <augie@google.com> [Wed, 12 Jun 2019 19:00:46 -0400] rev 42491
catapipe: add support for COUNTER events
Differential Revision: https://phab.mercurial-scm.org/D6524
Augie Fackler <augie@google.com> [Wed, 12 Jun 2019 16:08:21 -0400] rev 42490
demandimport: add tracing coverage for Python 3
This makes things feel a little less mysterious when modules are being
imported.
Differential Revision: https://phab.mercurial-scm.org/D6523
Martin von Zweigbergk <martinvonz@google.com> [Fri, 14 Jun 2019 10:21:47 -0700] rev 42489
export: don't prefetch *all* files in manifest
`hg export` only shows changed files, not all files, but we still
prefetched all files in cmdutil.export(). The same is true for the
other commands calling cmdutil.exportfile(). That meant that `hg
export` with remotefilelog (or lfs, I assume) could take much longer
than expected because it would download all the files in the repo.
Differential Revision: https://phab.mercurial-scm.org/D6532
Martin von Zweigbergk <martinvonz@google.com> [Fri, 14 Jun 2019 13:50:06 -0700] rev 42488
remotefilelog: remove obsolete filtering of treemanifest directories
I think this has been obsolete since 2cf18f46a1ce (narrow: only walk
files within narrowspec also for committed revisions, 2018-09-28).
Differential Revision: https://phab.mercurial-scm.org/D6531
Pulkit Goyal <7895pulkit@gmail.com> [Fri, 14 Jun 2019 18:27:50 +0300] rev 42487
py3: add test-dirstate-race2.t to list of passing tests
This test was added new recently. The py3 buildbot found that it passes, so
let's add it to the list of passing tests.
Differential Revision: https://phab.mercurial-scm.org/D6530
Taapas Agrawal <taapas2897@gmail.com> [Fri, 14 Jun 2019 18:25:14 +0530] rev 42486
strip: during merge allow strip only when -f is used
This ensures to abort strip to `hg strip` when we have a merge
in progress and allow it only when a `--force` flag is used.
Differential Revision: https://phab.mercurial-scm.org/D6529
Pierre-Yves David <pierre-yves.david@octobus.net> [Fri, 26 Apr 2019 00:48:12 +0200] rev 42485
deltas: set estimated compression upper bound to "3x" instead of "10x"
In pratice, we very rarely observer compression better than "3x" on manifest
deltas. Having a more aggressive estimate significantly helps our pathological
use case on a private repository. Here are a comparison of timings using
different upper bound.
Estimated compression | ø | ×10 | ×5 | ×3 |
timing | 14.11 | 2.61 | 1.96 | 1.53 |
We also tested the impact of this series on an array of public repositories.
This shown no impact in either size nor timing.
Full data set below for those interested.
Size
----
Regarding size, not significant impact have been noticed on neither public nor
private repositories. Here are the number we gathered on public repositories:
zlib/upperbound | no | 10x | 5x | 3x
mercurial | 5 875 730 | 5 875 730 | 5 875 730 | 5 875 730
pypy | 27 782 913 | 27 782 913 | 27 782 913 | 27 782 913
netbeans | 159 161 207 | 159 161 207 | 159 161 207 | 159 959 879 (+0.5%)
mozilla-central | 323 841 642 | 323 841 642 | 323 841 642 | 319 867 519 (-2.5%)
mozilla-try | 746 649 123 | 746 649 123 | 746 649 123 | 741 155 568 (-0.7%)
private-repo | 1 485 287 294 | 1 485 287 294 | 1 485 287 294 | 1 409 248 382 (-5.1%)
zstd/upperbound | no | 10x | 5x | 3x
mercurial | 5 895 206 | 5 895 206 | 5 895 206 | 5 895 206
pypy | 28 689 230 | 28 689 230 | 28 689 230 | 28 689 230
netbeans | 157 636 387 | 157 636 387 | 157 636 387 | 159 692 678 (+1.3%)
mozilla-central | 317 650 281 | 317 650 281 | 317 650 281 | 319 613 603 (+0.6%)
mozilla-try | 737 555 275 | 737 555 275 | 737 555 275 | 738 079 473 (+0.1%)
private-repo | 1 352 362 982 | 1 352 362 982 | 1 346 961 880 | 1 361 327 384 (+0.7%)
Speed
------
Timing gathered using `hg perfrevlogwrite -m`. Value are in seconds.
mercurial
zlib | no | 10x | 5x | 3x |
total | 65.551783 | 65.388887 | 65.260658 | 65.321199 |
max | 0.034544 | 0.034571 | 0.034659 | 0.034521 |
99.99% | 0.034544 | 0.034571 | 0.034659 | 0.034521 |
zstd | no | 10x | 5x | 3x |
total | 49.118449 | 49.054062 | 48.753588 | 48.740230 |
max | 0.009338 | 0.009239 | 0.009202 | 0.009178 |
99.99% | 0.007618 | 0.007639 | 0.007626 | 0.007621 |
pypy
zlib | no | 10x | 5x | 3x |
total | 560.865984 | 558.983817 | 559.083815 | 559.349152 |
max | 0.219614 | 0.215922 | 0.218112 | 0.218107 |
99.99% | 0.219614 | 0.215922 | 0.218112 | 0.218107 |
zstd | no | 10x | 5x | 3x |
total | 349.393280 | 347.395819 | 347.185407 | 345.643985 |
max | 0.084143 | 0.083536 | 0.081834 | 0.082178 |
99.99% | 0.039445 | 0.039639 | 0.039612 | 0.039175 |
netbeans
zlib | no | 10x | 5x | 3x |
total | 33103.327727 | 33314.932260 | 33211.745233 | 33345.891778 |
max | 2.666852 | 2.672059 | 2.662453 | 2.662936 |
99.99% | 2.058772 | 2.070429 | 2.069569 | 2.064653 |
zstd | no | 10x | 5x | 3x |
total | 20112.102708 | 20095.879719 | 20083.390300 | 20123.221859 |
max | 2.063482 | 2.062851 | 2.065229 | 2.060147 |
99.99% | 1.146647 | 1.143794 | 1.142933 | 1.146529 |
mozilla
zlib | no | 10x | 5x | 3x |
total | 41374.102138 | 41418.816773 | 41381.956370 | 41334.280732 |
max | 3.383474 | 3.387400 | 3.405711 | 3.387316 |
99.99% | 1.006755 | 1.005954 | 1.007700 | 1.007373 |
zstd | no | 10x | 5x | 3x |
total | 24689.691520 | 24643.939662 | 24664.630027 | 24664.512714 |
max | 1.460822 | 1.449640 | 1.439747 | 1.465304 |
99.99% | 0.527111 | 0.527377 | 0.527807 | 0.527226 |
Valentin Gatien-Baron <vgatien-baron@janestreet.com> [Mon, 21 Jan 2019 22:46:31 +0100] rev 42484
deltas: skip if projected compressed size is bigger than previous snapshot
If we have a delta, we check constraints against a lower bound estimate of the
resulting compressed delta. We then checks this projected size against the
`size(snapshotⁿ) > size(snapshotⁿ⁺¹)` constraint. This allows to exclude
potential base candidates before doing any expensive computation.
This only apply to the intermediate-snapshot case since this constraint only
apply to them.
For some pathological cases of a private repository this step provide a
further performance boost (timing from `hg perfrevlogwrite`):
before: 3.010646 seconds
after: 2.609307 seconds
Valentin Gatien-Baron <vgatien-baron@janestreet.com> [Mon, 21 Jan 2019 22:46:18 +0100] rev 42483
deltas: skip if projected compressed size does not match text size constraint
If we have a delta, we check constraints against a lower bound estimate of the
resulting compressed delta. We then checks this projected size against the ½ⁿ
size constraints. This allows to exclude potential base candidates before doing
any expensive computation.
This only apply to the intermediate-snapshot case since this constraint only apply to
them.
For some pathological cases of a private repository this step provide a
further performance boost (timing from `hg perfrevlogwrite`):
before: 3.145906 seconds
after: 3.010646 seconds
Valentin Gatien-Baron <vgatien-baron@janestreet.com> [Mon, 21 Jan 2019 22:37:30 +0100] rev 42482
deltas: accept and skip None return for delta info
They are some extra computation that will shortcut the delta compression if the
delta seems hopeless, returning None.
Valentin Gatien-Baron <vgatien-baron@janestreet.com> [Mon, 21 Jan 2019 22:36:16 +0100] rev 42481
delta: move some delta chain related computation earlier in deltainfo
They are some more optimization change that will make use of this in the
function. So we retrieve the data earlier.
Valentin Gatien-Baron <vgatien-baron@janestreet.com> [Thu, 25 Apr 2019 22:50:33 +0200] rev 42480
deltas: skip if projected delta size is bigger than previous snapshot
Before computing any delta, we get a basic estimation of the delta size we can
expect and the resulted compressed value. We then checks this projected size
against the `size(snapshotⁿ) > size(snapshotⁿ⁺¹)` constraint. This allows to
exclude potential base candidates before doing any expensive computation.
This only apply to the intermediate-snapshot case since this constraint only
apply to them.
For some pathological cases of a private repository this step provide a
significant performance boost (timing from `hg perfrevlogwrite`):
before: 14.115908 seconds
after: 3.145906 seconds
Valentin Gatien-Baron <vgatien-baron@janestreet.com>, Pierre-Yves David <pierre-yves.david@octobus.net> [Thu, 25 Apr 2019 22:30:14 +0200] rev 42479
deltas: skip if projected delta size does not match text size constraint
Before computing any delta, we get a basic estimation of the delta size we can
expect and the resulted compressed value. We then checks this projected size
against the ½ⁿ size constraints. This allows to exclude potential base
candidates before doing any expensive computation.
This only apply to the intermediate-snapshot case since this constraint only
apply to them.
In practice we only perform this new checks for the manifestlog. Manifest log
combine two property: it is likely to have delta chain issue and its
diffing/compression is fairly predictable.
The initial author of this changeset is Valentin Gatien-Baron providing the
initial idea and initial testing, Pierre-Yves David later consolidated the code
in the right location and run more extensive testing.
Pierre-Yves David <pierre-yves.david@octobus.net> [Fri, 26 Apr 2019 00:28:22 +0200] rev 42478
revlog: add the option to track the expected compression upper bound
There are various optimization we can do if we can estimate the size of delta
before actually spending CPU compressing them. So we add a attributed dedicated
to tracking that.
We only use it on Manifest because (1) it structure is quite stable across all
Mercurial repository so its compression ratio is fairly universal. This is the
revlog with most extreme delta (cf the sparse-revlog optimization).
This will be put to use in later changesets.
Right now the compression upper bound is set to 10. This is a fairly
conservative value (observed value is more around 3), but I prefer to be safe
while introducing the optimization principles. We can tune the optimization
threshold later.
Pierre-Yves David <pierre-yves.david@octobus.net> [Wed, 12 Jun 2019 17:30:24 +0100] rev 42477
perf: clarify some of the custom behavior of `perfrevlogwrite`
This reduce the chance of developers being surprised by special cases.
Pierre-Yves David <pierre-yves.david@octobus.net> [Wed, 12 Jun 2019 16:56:41 +0100] rev 42476
perf: fix perfrevlogwrite --count documentation
The help text was copy pasted from the previous option.
Georges Racinet <georges.racinet@octobus.net> [Fri, 17 May 2019 00:17:43 +0200] rev 42475
rust: switched to 'cargo rustc' in setup.py
This is more flexible in the passing of additional flags, also
what setuptools_rust does, giving less uncertainty about non-Linux
platforms.
Georges Racinet <georges.racinet@octobus.net> [Fri, 14 Jun 2019 11:18:06 +0100] rev 42474
rust-cpython: fix build for MacOSX
MacOSX needs special link flags. Quoting the README of rust-cpython:
create a `.cargo/config` with the following content:
```
[target.x86_64-apple-darwin]
rustflags = [
"-C", "link-arg=-undefined",
"-C", "link-arg=dynamic_lookup",
]
```
This is tested with Python 2.7 (Anaconda install) and Python 3
(Homebrew install)
Georges Racinet <georges.racinet@octobus.net> [Fri, 14 Jun 2019 10:57:07 +0100] rev 42473
rust-cpython: management of shared libray suffix
Before this changeset, the shared library objects suffixes
were both (rustc output and Python input) hardcoded to '.so',
which is wrong for Python3 and non Linux targets.
Valentin Gatien-Baron <valentin.gatienbaron@gmail.com> [Mon, 27 May 2019 16:55:46 -0400] rev 42472
merge: fix race that could cause wrong size in dirstate
The problem is that hg merge/update/etc work the following way:
1. figure out what files to update
2. apply the update to disk
3. apply the update to in-memory dirstate
4. write dirstate
where step3 looks at the filesystem and assumes it sees the result of
step2. If a file is changed between step2 and step3, step3 will record
incorrect information in the dirstate.
I avoid this by passing the size step3 needs directly from step2, for
the common path (not implemented for change/delete conflicts for
instance).
I didn't fix the same race for the exec bit for now, because it's less
likely to be problematic and I had trouble due to the fact that the
dirstate stores the permissions differently from the manifest (st_mode
vs '' 'l' 'x'), in combination with tests that pretend that symlinks
are not supported.
However, I moved the lstat from step3 to step2, which should tighten
the race window markedly, both for the exec bit and for the mtime.
Differential Revision: https://phab.mercurial-scm.org/D6475
Valentin Gatien-Baron <vgatien-baron@janestreet.com> [Wed, 12 Jun 2019 13:10:52 -0400] rev 42471
worker: support parallelization of functions with return values
Currently worker supports running functions that return a progress
iterator. Generalize it to handle function that return a progress
iterator then a return value.
It's unused in this commit, but will be used in the next one.
Differential Revision: https://phab.mercurial-scm.org/D6515
Valentin Gatien-Baron <valentin.gatienbaron@gmail.com> [Sun, 19 May 2019 16:06:06 -0400] rev 42470
tests: show how the dirstate can end up containing wrong information
which can result in bad status output.
Concretely, this seems to be easily triggered by having a build system
watching the filesystem for changes, and rebuilding files that are
both tracked and generated while an update is happening.
Differential Revision: https://phab.mercurial-scm.org/D6474
Georges Racinet <georges.racinet@octobus.net> [Thu, 23 May 2019 02:05:32 +0200] rev 42469
rust: new rust options in setup.py
The --rust global option turns on usage (and by default compilation)
of the rust-cpython based mercurial.rustext.
Similarly to what's previously done for zstd, there is a --no-rust
option for the build_ext subcommand in order not to build
mercurial.rustext, allowing for an OS distribution to prebuild it.
The HGWITHRUSTEXT environment variable is still honored, and has
the same effect as before, but now it works mostly by making
the --rust global option defaulting to True, with some special
cases for the direct-ffi case (see more about that below)
Coincidentally, the --rust flag can also be passed from the make
commands, like actually all global options, in the PURE variable
make local PURE=--rust
This feels inappropriate, though, and we should follow up with
a proper make variable for that case.
Although the direct-ffi bindings aren't directly useful any more, we
keep them at this stage because
- they provide a short prototyping path for experiments in which a C extension
module has to call into a Rust extension. The proper way of doing that would
be to use capsules, and it's best to wait for our pull request onto
rust-cpython for that: https://github.com/dgrunwald/rust-cpython/pull/169
- Build support for capsules defined in Rust will probably need to reuse
some of what's currently in use for direct-ffi.
Georges Racinet <georges.racinet@octobus.net> [Thu, 30 May 2019 09:14:41 +0200] rev 42468
rust: using policy.importrust from Python callers
This commit converts all current Python callers of
mercurial.rustext to the new policy.importrust system.
After this point, going through policy.importrust
or policy.importmod (in some more distant future)
is mandatory for callers of Rust code outside of
Python tests.
We felt it to be appropriate to keep Rust-specific tests
run inconditionally if the Rust extensions are present.
Georges Racinet <georges.racinet@octobus.net> [Wed, 29 May 2019 13:27:56 +0200] rev 42467
rust: module policy with importrust
We introduce two rust+c module policies and a new
`policy.importrust()` that makes use of them.
This simple approach provides runtime switching of
implementations, which is crucial for the performance
measurements such as those Octobus does with ASV.
It can also be useful for bug analysis.
It also has the advantage of making conditionals in
Rust callers more uniform, in particular
abstracting over specifics like `demandimport`
At this point, the build stays unchanged, with the rust-cpython based
`rustext` module being built if HGWITHRUSTEXT=cpython.
More transparency for the callers, i.e., just using
`policy.importmod` would be a much longer term and riskier
effort for the following reasons:
1. It would require to define common module boundaries
for the three or four cases (pure, c, rust+ext, cffi) and that
is premature with the Rust extension currently under heavy
development in areas that are outside the scope of the C extensions.
2. It would imply internal API changes that are not currently wished,
as the case of ancestors demonstrates.
3. The lack of data or property-like attributes (tp_member
and tp_getset) in current `rust-cpython` makes it impossible to
achieve direct transparent replacement of pure Python classes by
Rust extension code, meaning that the caller sometimes has to be able
to make adjustments or provide additional wrapping.
Navaneeth Suresh <navaneeths1998@gmail.com> [Thu, 13 Jun 2019 23:28:31 +0300] rev 42466
help: add help entry for internals.mergestate
This patch adds an entry for `internals.mergestate` as suggested
by @marmoute. Most of the help text is taken from `merge.mergestate`.
Differential Revision: https://phab.mercurial-scm.org/D6448
Differential Revision: https://phab.mercurial-scm.org/D6528
Ian Moody <moz-ian@perix.co.uk> [Wed, 12 Jun 2019 17:22:37 +0100] rev 42465
phabricator: use parents.set to always set dependencies
Now that Mercurial's Phabricator instance has been updated to a version that
supports the parents.set transaction on revision.edit we can use that to set
dependency relationships in patch stacks instead of abusing the summary.
This has the advantage that we can use it on every `phabsend` so commit
reordering is picked up without spamming changes like abusing the summary would,
and using parents.set will clear previous parents unlike parents.add.
Differential Revision: https://phab.mercurial-scm.org/D6514
amalloy [Fri, 31 May 2019 10:12:56 -0700] rev 42464
help: remove repeated word in 'hg help rebase'
Specifically, the second 'with' in 'with which to merge with'.
Differential Revision: https://phab.mercurial-scm.org/D6483
Kyle Lippincott <spectral@google.com> [Mon, 10 Jun 2019 15:35:06 -0700] rev 42463
rebase: tweak description of inmemory working even w/ dirty working dir
One of our users was confused because they read this, and then attempted to run
`hg rebase` with a dirty working directory, and it still complained. The reason
was that they were attempting to rebase the commit they currently had checked
out, which (at least with evolve workflows enabled) involves updating the
working directory to be based on the newly rebased commit.
Differential Revision: https://phab.mercurial-scm.org/D6507
Valentin Gatien-Baron <vgatien-baron@janestreet.com> [Mon, 10 Jun 2019 13:23:14 -0400] rev 42462
revlog: speed up isancestor
Currently, it is implemented on top of commonancestorsheads.
Implement it on top of reachableroots instead, as reachableroots could
stop walking the graph much sooner than commonancestorsheads.
Measuring repo.changelog.isancestorrev on two revisions in a private repository:
before: ! wall 0.005175 comb 0.010000 user 0.010000 sys 0.000000 (best of 550)
after : ! wall 0.000072 comb 0.000000 user 0.000000 sys 0.000000 (best of 36199)
When hg does this kind of operations 1500 times in pull ->
bookmarks.comparebookmarks -> bookmarks.validdest, that's 11s that
drop from the --profile output.
Differential Revision: https://phab.mercurial-scm.org/D6506
Valentin Gatien-Baron <vgatien-baron@janestreet.com> [Mon, 10 Jun 2019 11:40:43 -0400] rev 42461
dagop: fix documentation of reachableroots
The previous revset couldn't be correct as it is symmetric in <roots>
and <heads>, but reachableroots has no such symmetry. It makes a
difference with for instance reachableroots(2, 3) where 2 and 3 are
both children of 1.
Differential Revision: https://phab.mercurial-scm.org/D6505
Ian Moody <moz-ian@perix.co.uk> [Tue, 11 Jun 2019 19:52:16 +0100] rev 42460
phabricator: add --blocker argument to phabsend to specify blocking reviewers
The way to signal to Conduit that a reviewer is considered blocking is just to
wrap their PHID in "blocking()" when including it in the list of PHIDs passed
to `reviewers.add`.
arc doesn't have a --blocker, instead one is supposed to append a '!' to the
end of reviewer names (I think reviewers are usually added in an editor rather
than the command line, where '!'s can be more hazardous).
moz-phab (Mozilla's arcanist wrapper) does have a --blocker argument, and being
explicit like this is also more discoverable. Even `arc diff`'s help doesn't
seem to mention the reviewer! syntax.
Differential Revision: https://phab.mercurial-scm.org/D6512
Ian Moody <moz-ian@perix.co.uk> [Tue, 11 Jun 2019 19:37:19 +0100] rev 42459
phabricator: auto-sanitise API tokens and HTTP cookies from VCR recordings
Currently when making VCR recordings one needs to manually sanitise sensitive
credentials before committing and submitting them as part of tests. It is easy
to imagine this being accidentally missed one time by a fallible human and said
credentials being leaked. It is also possible that it wouldn't be noticed to
alert the user to the leak since the recording files are so large and
practically unreviewable. Thus do so automatically, so the only place that needs
checking is in the test-phabricator.t file.
Differential Revision: https://phab.mercurial-scm.org/D6513
Pulkit Goyal <pulkit@yandex-team.ru> [Tue, 11 Jun 2019 15:46:07 +0300] rev 42458
py3: use .startswith() instead of bytes[0]
Doing bytes[0] will return the ascii value of that position which breaks
comparison with a bytechar.
This makes test-absorb.t work again on py3.
Differential Revision: https://phab.mercurial-scm.org/D6508
Yuya Nishihara <yuya@tcha.org> [Sun, 09 Jun 2019 22:23:41 +0900] rev 42457
revset: fix merge() to fall back to changectx API if wdir specified
I have a code which basically runs "0:wdir() & <user-revset>", and it crashed
if merge() were passed in.
Yuya Nishihara <yuya@tcha.org> [Sun, 09 Jun 2019 22:18:22 +0900] rev 42456
revset: use nullrev constant in merge()
Martin von Zweigbergk <martinvonz@google.com> [Fri, 31 May 2019 22:38:04 -0700] rev 42455
mixedrepostorecache: fix a silly redundant updating of set
Differential Revision: https://phab.mercurial-scm.org/D6470
Raphaël Gomès <rgomes@octobus.net> [Thu, 06 Jun 2019 18:37:21 +0200] rev 42454
rust-regex: fix shortcut for exact matches
The current shortcut for rootglobs that can be simplified to exact matches
does not work, it instead treats the pattern as a regex, which is not the
same thing.
This changes fixes the behavior and introduces a test for this behavior.
Differential Revision: https://phab.mercurial-scm.org/D6489
Raphaël Gomès <rgomes@octobus.net> [Thu, 06 Jun 2019 15:30:56 +0200] rev 42453
rust-filepatterns: use bytes instead of String
In my initial patch, I introduced an unnecessary hard constraint on UTF-8
filenames and patterns which I forgot to remove. Although the performance
penalty for using String might be negligible, we don't want to break
compatibility with non-UTF-8 encodings for no reason.
Moreover, this change allows for a cleaner Rust core API.
This patch introduces a new utils module that is used with this fix.
Finally, PatternError was not put inside the Python module generated by
Rust, which would have raised a NameError.
Differential Revision: https://phab.mercurial-scm.org/D6485
Joerg Sonnenberger <joerg@bec.de> [Sat, 01 Jun 2019 01:24:49 +0200] rev 42452
doc: fix description of "predecessors" to match reality
Differential Revision: https://phab.mercurial-scm.org/D6467
Pulkit Goyal <pulkit@yandex-team.ru> [Sat, 08 Jun 2019 18:48:06 +0300] rev 42451
phabricator: make `hg debugcallconduit` work outside a hg repo
I am trying to write some automations around phabricator and having
debugcallconduit work outside a hg repo will be nice!
Marking command as optionalrepo instead of norepo because we might to load
repo/.hg/hgrc.
Differential Revision: https://phab.mercurial-scm.org/D6499
Pulkit Goyal <pulkit@yandex-team.ru> [Sat, 08 Jun 2019 18:41:15 +0300] rev 42450
phabricator: pass ui instead of repo to callconduit
This will help us make `hg debugcallconduit` work outside a hg repo as next
patch will mark that command as no repo.
Differential Revision: https://phab.mercurial-scm.org/D6498
Pulkit Goyal <pulkit@yandex-team.ru> [Sat, 08 Jun 2019 18:32:12 +0300] rev 42449
phabricator: pass ui into readurltoken instead of passing repo
The goal of this series is to make `hg debugcallconduit` work outside of a hg
repo.
This patch, removes requirement of repo object from readurltoken as we only need
ui there. It also updates the callers to pass in ui instead of repo.
Differential Revision: https://phab.mercurial-scm.org/D6497
Pulkit Goyal <pulkit@yandex-team.ru> [Sat, 08 Jun 2019 19:20:31 +0300] rev 42448
py3: add test-contrib-emacs.t to passing tests list
I installed emacs on the server running buildbot and the test started passing on
Python 3. Lets add it to the list of passing test.
Differential Revision: https://phab.mercurial-scm.org/D6500
Ian Moody <moz-ian@perix.co.uk> [Fri, 07 Jun 2019 20:19:55 +0100] rev 42447
phabricator: add commenting to phabsend for new/updated Diffs
Especially useful when sending updates to existing Revisions so one can specify
the sort of changes e.g. "Address review comments" or "Rebase to tip"
If the diff content hasn't changed then it only needs a metadata update which
doesn't show in the Phabricator updates UI, so don't add a comment that will.
Differential Revision: https://phab.mercurial-scm.org/D6496
Pulkit Goyal <pulkit@yandex-team.ru> [Wed, 05 Jun 2019 22:09:26 +0300] rev 42446
py3: fix test-bookmarks-corner-case.t
For some reasons, the output of print was not going through. Replaced that
ui.status().
Differential Revision: https://phab.mercurial-scm.org/D6481
Pulkit Goyal <pulkit@yandex-team.ru> [Wed, 05 Jun 2019 22:02:57 +0300] rev 42445
py3: fix test-fix-metadata.t
# skip-blame as just b'' prefixes
Differential Revision: https://phab.mercurial-scm.org/D6480
Pulkit Goyal <pulkit@yandex-team.ru> [Wed, 05 Jun 2019 22:44:38 +0300] rev 42444
py3: add b'' prefix at one place in run-tests.py
#skip-blame because just b'' prefix
Differential Revision: https://phab.mercurial-scm.org/D6482
Martin von Zweigbergk <martinvonz@google.com> [Thu, 06 Jun 2019 10:07:14 -0700] rev 42443
copies: separate added/removed files by newline instead of null
This makes it more consistent with how we encode copies
(newline-separated lists of null-separated pairs). This perhaps makes
{extras} a little less readable (?) despite avoiding the escaping. I
don't know how I feel about this patch. I'm okay with it being queued
or dropped.
Differential Revision: https://phab.mercurial-scm.org/D6486
Martin von Zweigbergk <martinvonz@google.com> [Wed, 22 May 2019 09:54:00 -0700] rev 42442
copies: also encode p[12]copies destination as index into "files" list
This is mostly for consistency with the filesaddes/filesremoved
fields, but it should also save a bit of space.
Differential Revision: https://phab.mercurial-scm.org/D6431
Pierre-Yves David <pierre-yves.david@octobus.net> [Wed, 05 Jun 2019 11:23:25 +0200] rev 42441
discovery: be more conservative when adjusting the sample size
Since 5b34972a0094, the discovery will increase the sample size when it detect a
"complex" undecided set. However this detection focussed on the number of roots
only, this could regress discovery performance when the undecided set has many
roots that eventually get merged into a few heads.
To prevent such misbehavior, we adjust the logic to take in account both heads
and roots. The sample size will be increased only if both are especially large.
Performance testing on the same case as 5b34972a0094, does not show a
significant difference.
Raphaël Gomès <rgomes@octobus.net> [Thu, 16 May 2019 16:22:20 +0200] rev 42440
rust-dirstate: create dirstate submodule
This change is here to facilitate a future patch that is written in a
different file. I expect this module to grow a few different files.
Differential Revision: https://phab.mercurial-scm.org/D6389
Valentin Gatien-Baron <vgatien-baron@janestreet.com> [Wed, 05 Jun 2019 12:51:21 -0400] rev 42439
profiling: show actual time spent in hotpath display
To get, for instance:
...
\ 6.6% 4.08s lock.py: __exit__ line 1566: ...
| 6.5% 4.01s exchange.py: close line 1191: ...
| 6.5% 4.01s transaction.py: _active line 1443: ...
| 6.5% 4.01s transaction.py: close line 47: ...
| 6.2% 3.84s scmutil.py: wrapped line 529: ...
| 6.2% 3.81s localrepo.py: wrapper line 2114: ...
| 6.2% 3.81s localrepo.py: updatecaches line 177: ...
...
instead of:
...
\ 6.6% lock.py: __exit__ line 1566: ...
| 6.5% exchange.py: close line 1191: ...
| 6.5% transaction.py: _active line 1443: ...
| 6.5% transaction.py: close line 47: ...
| 6.2% scmutil.py: wrapped line 529: ...
| 6.2% localrepo.py: wrapper line 2114: ...
| 6.2% localrepo.py: updatecaches line 177: ...
...
I find that if it's not displayed, I frequently end up estimating the
numbers by hand.
Differential Revision: https://phab.mercurial-scm.org/D6477
Martin von Zweigbergk <martinvonz@google.com> [Wed, 05 Jun 2019 14:29:44 -0700] rev 42438
merge with stable
Augie Fackler <augie@google.com> [Wed, 05 Jun 2019 10:18:00 -0400] rev 42437
merge with stable
Yuya Nishihara <yuya@tcha.org> [Tue, 04 Jun 2019 21:13:35 +0900] rev 42436
root: add template variables pointing to repository directories
These paths are useful for GUI applications to detect changes. A GUI process
typically monitors .hg and .hg/store directories so that it will be notified
on lock/wlock deletion.
Alternatively, maybe we can add debugpaths command if we don't want to extend
the root command. I'm not sure which will be nicer.
Yuya Nishihara <yuya@tcha.org> [Tue, 04 Jun 2019 20:58:39 +0900] rev 42435
root: add support for -Tformatter option
It's useless right now, but it should just work and I want to add a few more
fields.
Pulkit Goyal <7895pulkit@gmail.com> [Thu, 23 May 2019 03:03:36 +0530] rev 42434
narrow: pass the bundle to bundle2.widen_bundle() instead of generating there
This will make the code in narrowwirepeer.py more better for further
refactoring.
Differential Revision: https://phab.mercurial-scm.org/D6438
Pulkit Goyal <7895pulkit@gmail.com> [Thu, 23 May 2019 02:48:25 +0530] rev 42433
narrow: refactor code around widening complicated by previous patch
Previous patch while adding support for using narrow_widen wireproto command,
complicated the code a bit. This patch refactors that.
Differential Revision: https://phab.mercurial-scm.org/D6437
Pulkit Goyal <7895pulkit@gmail.com> [Wed, 22 May 2019 02:59:48 +0530] rev 42432
narrow: use narrow_widen wireproto command to widen in case of ellipses
Few releases ago, we introduce narrow_widen wireproto command to be used to widen
narrow repositories. Before this patch, that was used in non-ellipses cases
only. In ellipses cases, we still do exchange.pull() which can pull more data
than required.
After this patch, the client will first check whether server supports doing
ellipses widening using wireproto command or not by checking server's wireproto
capability. If the server is upto date and support latest ellipses capability,
we call the wireproto command. Otherwise we fallback to exchange.pull() like
before.
The compat code make sure that things works even if one of the client or server
is old. The initial version of this patch does not had this compat code. It's
added to help Google release things smoothly internally. I plan to drop the
compat code before the upcoming major release.
Due to change to wireproto command, the code looks a bit dirty, next patches
will clean that up.
Differential Revision: https://phab.mercurial-scm.org/D6436
Anton Shestakov <av6@dwimlabs.net> [Tue, 04 Jun 2019 17:24:35 +0800] rev 42431
merge: correct argument name in docstring
Differential Revision: https://phab.mercurial-scm.org/D6476
Martin von Zweigbergk <martinvonz@google.com> [Fri, 31 May 2019 15:28:31 -0700] rev 42430
narrowspec: replace one recursion-avoidance hack with another
When updating the working copy narrowspec, we call context.walk() in
order to find which files to update the working copy
with. context.walk() calls repo.narrowmatch(). In order to avoid
infinite recursion in this case, we have a hack that assigns the new
values for repo.narrowpats and repo._narrowmatch. However, doing that
of course breaks future invalidation of those properties (they're
@storecache'd). Let's instead avoid the infinite recursion by setting
a flag on the repo instance when we're updating the working copy.
Differential Revision: https://phab.mercurial-scm.org/D6468
Martin von Zweigbergk <martinvonz@google.com> [Sat, 09 Mar 2019 22:13:06 -0800] rev 42429
merge: simplify initialization of "pas"
Differential Revision: https://phab.mercurial-scm.org/D6472
Martin von Zweigbergk <martinvonz@google.com> [Sat, 09 Mar 2019 22:11:27 -0800] rev 42428
merge: reorder some initialization to make more sense
This puts the closely related definitions of "pl", "p1", "p2", "pas"
close together, and moves the definition of "overwrite" away and
closer to where it's first used.
Differential Revision: https://phab.mercurial-scm.org/D6471
Georges Racinet <georges.racinet@octobus.net> [Wed, 22 May 2019 08:27:02 +0000] rev 42427
rust-dirstate: architecture independence fix
Apparently, c_char is u8 on ppc64le and i8 on amd64
Differential Revision: https://phab.mercurial-scm.org/D6473
Martin von Zweigbergk <martinvonz@google.com> [Tue, 14 May 2019 22:20:10 -0700] rev 42426
context: get filesadded() and filesremoved() from changeset if configured
This adds the read side for getting the sets of added and removed
files from the changeset extras. I timed this command on the hg repo:
hg log -T '{rev}\n {files}\n %:{file_mods}\n +{file_adds}\n -{file_dels}\n'
It took 1m21s before and 6.4s after. I also used that command to check
that the result didn't change compared to calculating the values from
the manifests on the fly (it didn't change).
In the mozilla-unified repo, the same command run on
FIREFOX_BETA_58_END::FIREFOX_BETA_59_END went from 29s to 0.67s.
Differential Revision: https://phab.mercurial-scm.org/D6417