Phil Cohen <phillco@fb.com> [Thu, 14 Sep 2017 13:14:32 -0700] rev 34302
merge: allow a custom working context to be passed to update
This will allow anyone to enable the first in-menmory merge milestone
by wrapping merge.update in an extension and creating an overlayworkingctx.
Differential Revision: https://phab.mercurial-scm.org/D682
Augie Fackler <augie@google.com> [Wed, 14 Jun 2017 11:13:57 -0400] rev 34301
patchbomb: add test that shows --to and --cc override matching config item
As far as I know this has always been true and is intentional (it's in
line with many other behaviors), but it wasn't tested.
Jun Wu <quark@fb.com> [Thu, 21 Sep 2017 11:10:20 -0700] rev 34300
blackbox: move _openlogfile to a separate method
This removes several `stat` syscalls for accessing `_bbvfs` and makes `ui`
object cleaner.
Differential Revision: https://phab.mercurial-scm.org/D769
Jun Wu <quark@fb.com> [Thu, 21 Sep 2017 11:03:37 -0700] rev 34299
blackbox: do not prevent 'chg init' from working
Previously, blackbox always appends to blackbox.log and creates the
directory for that file on demand. That could be an issue if:
1. chg starts from `$REPO` directory, so `ui._bbrepo` is set.
2. `rm -rf $REPO`.
3. `chg init $REPO`, blackbox writes something and `init` will fail
because `$REPO` directory is non-empty.
This patch fixes that by verifying whether vfs exists before re-using it.
Differential Revision: https://phab.mercurial-scm.org/D768
Augie Fackler <augie@google.com> [Mon, 11 Sep 2017 20:07:41 -0400] rev 34298
tests: add a test for blackbox with nested alias configurations
I've observed some weirdness around this, and needed to rule some
things out. There aren't any bugs in core around this, but it was nice
to have confirmation.
Augie Fackler <augie@google.com> [Mon, 11 Sep 2017 20:06:52 -0400] rev 34297
tests: clean up blackbox test around aliases a little bit
Mark Thomas <mbthomas@fb.com> [Wed, 13 Sep 2017 17:26:26 +0000] rev 34296
revlog: add option to mmap revlog index
Following on from Jun Wu's patch last October[1], we have found that using mmap
for the revlog index in repos with large revlogs gives a noticable performance
improvment (~110ms on each hg invocation), particularly for commands that don't
touch the index very much.
This changeset adds this as an option, activated by a new experimental config
option so that it can be enabled on a per-repo basis. The configuration option
specifies an index size threshold at which Mercurial will switch to using mmap
to access the index.
If the configuration option is not specified, the default remains to load the
full file, which seems to be the best option for smaller repos.
Some initial performance numbers for average of 5 invocations of `hg log -l 5`
for different cache states:
| Repo: | HG | FB |
|---|---|---|
| Index size: | 2.3MB | much bigger |
| read (warm): | 237ms | 432ms |
| mmap (warm): | 227ms | 321ms |
| | (-3%) | (-26%) |
| read (cold): | 397ms | 696ms |
| mmap (cold): | 410ms | 888ms |
| | (+3%) | (+28%) |
[1] https://www.mercurial-scm.org/pipermail/mercurial-devel/2016-October/088737.html
Test Plan:
`hg log --config experimental.mmapindex=true`
Differential Revision: https://phab.mercurial-scm.org/D477
Mark Thomas <mbthomas@fb.com> [Thu, 21 Sep 2017 05:54:34 -0700] rev 34295
util: add an mmapread method
This is useful for large files that are only partly touched.
Test Plan:
Will be used and tested in a later patch.
Differential Revision: https://phab.mercurial-scm.org/D476