##// END OF EJS Templates
mmap: populate the mapping by default...
mmap: populate the mapping by default Without pre-population, accessing all data through a mmap can result in many pagefault, reducing performance significantly. If the mmap is prepopulated, the performance can no longer get slower than a full read. (See benchmark number below) In some cases were very few data is read, prepopulating can be overkill and slower than populating on access (through page fault). So that behavior can be controlled when the caller can pre-determine the best behavior. (See benchmark number below) In addition, testing with populating in a secondary thread yield great result combining the best of each approach. This might be implemented in later changesets. In all cases, using mmap has a great effect on memory usage when many processes run in parallel on the same machine. ### Benchmarks # What did I run A couple of month back I ran a large benchmark campaign to assess the impact of various approach for using mmap with the revlog (and other files), it highlighted a few benchmarks that capture the impact of the changes well. So to validate this change I checked the following: - log command displaying various revisions (read the changelog index) - log command displaying the patch of listed revisions (read the changelog index, the manifest index and a few files indexes) - unbundling a few revisions (read and write changelog, manifest and few files indexes, and walk the graph to update some cache) - pushing a few revisions (read and write changelog, manifest and few files indexes, walk the graph to update some cache, performs various accesses locally and remotely during discovery) Benchmarks were run using the default module policy (c+py) and the rust one. No significant difference were found between the two implementation, so we will present result using the default policy (unless otherwise specified). I ran them on a few repositories : - mercurial: a "public changeset only" copy of mercurial from 2018-08-01 using zstd compression and sparse-revlog - pypy: a copy of pypy from 2018-08-01 using zstd compression and sparse-revlog - netbeans: a copy of netbeans from 2018-08-01 using zstd compression and sparse-revlog - mozilla-try: a copy of mozilla-try from 2019-02-18 using zstd compression and sparse-revlog - mozilla-try persistent-nodemap: Same as the above but with a persistent nodemap. Used for the log --patch benchmark only # Results For the smaller repositories (mercurial, pypy), the impact of mmap is almost imperceptible, other cost dominating the operation. The impact of prepopulating is undiscernible in the benchmark we ran. For larger repositories the benchmark support explanation given above: On netbeans, the log can be about 1% faster without repopulation (for a difference < 100ms) but unbundle becomes a bit slower, even when small. ### data-env-vars.name = netbeans-2018-08-01-zstd-sparse-revlog # benchmark.name = hg.command.unbundle # benchmark.variants.issue6528 = disabled # benchmark.variants.reuse-external-delta-parent = yes # benchmark.variants.revs = any-1-extra-rev # benchmark.variants.source = unbundle # benchmark.variants.verbosity = quiet with-populate: 0.240157 no-populate: 0.265087 (+10.38%, +0.02) # benchmark.variants.revs = any-100-extra-rev with-populate: 1.459518 no-populate: 1.481290 (+1.49%, +0.02) ## benchmark.name = hg.command.push # benchmark.variants.explicit-rev = none # benchmark.variants.issue6528 = disabled # benchmark.variants.protocol = ssh # benchmark.variants.reuse-external-delta-parent = yes # benchmark.variants.revs = any-1-extra-rev with-populate: 0.771919 no-populate: 0.792025 (+2.60%, +0.02) # benchmark.variants.revs = any-100-extra-rev with-populate: 1.459518 no-populate: 1.481290 (+1.49%, +0.02) For mozilla-try, the "slow down" from pre-populate for small `hg log` is more visible, but still small in absolute time. (using rust value for the persistent nodemap value to be relevant). ### data-env-vars.name = mozilla-try-2019-02-18-ds2-pnm # benchmark.name = hg.command.log # bin-env-vars.hg.flavor = rust # benchmark.variants.patch = yes # benchmark.variants.limit-rev = 1 with-populate: 0.237813 no-populate: 0.229452 (-3.52%, -0.01) # benchmark.variants.limit-rev = 10 # benchmark.variants.patch = yes with-populate: 1.213578 no-populate: 1.205189 ### data-env-vars.name = mozilla-try-2019-02-18-zstd-sparse-revlog # benchmark.variants.limit-rev = 1000 # benchmark.variants.patch = no # benchmark.variants.rev = tip with-populate: 0.198607 no-populate: 0.195038 (-1.80%, -0.00) However pre-populating provide a significant boost on more complex operations like unbundle or push: ### data-env-vars.name = mozilla-try-2019-02-18-zstd-sparse-revlog # benchmark.name = hg.command.push # benchmark.variants.explicit-rev = none # benchmark.variants.issue6528 = disabled # benchmark.variants.protocol = ssh # benchmark.variants.reuse-external-delta-parent = yes # benchmark.variants.revs = any-1-extra-rev with-populate: 4.798632 no-populate: 4.953295 (+3.22%, +0.15) # benchmark.variants.revs = any-100-extra-rev with-populate: 4.903618 no-populate: 5.014963 (+2.27%, +0.11) ## benchmark.name = hg.command.unbundle # benchmark.variants.revs = any-1-extra-rev with-populate: 1.423411 no-populate: 1.585365 (+11.38%, +0.16) # benchmark.variants.revs = any-100-extra-rev with-populate: 1.537909 no-populate: 1.688489 (+9.79%, +0.15)

File last commit:

r44583:d56a2d6f default
r52574:522b4d72 default
Show More
environment.txt
124 lines | 4.3 KiB | text/plain | TextLexer
HG
Path to the 'hg' executable, automatically passed when running
hooks, extensions or external tools. If unset or empty, this is
the hg executable's name if it's frozen, or an executable named
'hg' (with %PATHEXT% [defaulting to COM/EXE/BAT/CMD] extensions on
Windows) is searched.
HGEDITOR
This is the name of the editor to run when committing. See EDITOR.
(deprecated, see :hg:`help config.ui.editor`)
HGENCODING
This overrides the default locale setting detected by Mercurial.
This setting is used to convert data including usernames,
changeset descriptions, tag names, and branches. This setting can
be overridden with the --encoding command-line option.
HGENCODINGMODE
This sets Mercurial's behavior for handling unknown characters
while transcoding user input. The default is "strict", which
causes Mercurial to abort if it can't map a character. Other
settings include "replace", which replaces unknown characters, and
"ignore", which drops them. This setting can be overridden with
the --encodingmode command-line option.
HGENCODINGAMBIGUOUS
This sets Mercurial's behavior for handling characters with
"ambiguous" widths like accented Latin characters with East Asian
fonts. By default, Mercurial assumes ambiguous characters are
narrow, set this variable to "wide" if such characters cause
formatting problems.
HGMERGE
An executable to use for resolving merge conflicts. The program
will be executed with three arguments: local file, remote file,
ancestor file.
(deprecated, see :hg:`help config.ui.merge`)
HGRCPATH
A list of files or directories to search for configuration
files. Item separator is ":" on Unix, ";" on Windows. If HGRCPATH
is not set, platform default search path is used. If empty, only
the .hg/hgrc from the current repository is read.
For each element in HGRCPATH:
- if it's a directory, all files ending with .rc are added
- otherwise, the file itself will be added
HGRCSKIPREPO
When set, the .hg/hgrc from repositories are not read.
HGPLAIN
When set, this disables any configuration settings that might
change Mercurial's default output. This includes encoding,
defaults, verbose mode, debug mode, quiet mode, tracebacks, and
localization. This can be useful when scripting against Mercurial
in the face of existing user configuration.
In addition to the features disabled by ``HGPLAIN=``, the following
values can be specified to adjust behavior:
``+strictflags``
Restrict parsing of command line flags.
Equivalent options set via command line flags or environment
variables are not overridden.
See :hg:`help scripting` for details.
HGPLAINEXCEPT
This is a comma-separated list of features to preserve when
HGPLAIN is enabled. Currently the following values are supported:
``alias``
Don't remove aliases.
``color``
Don't disable colored output.
``i18n``
Preserve internationalization.
``revsetalias``
Don't remove revset aliases.
``templatealias``
Don't remove template aliases.
``progress``
Don't hide progress output.
Setting HGPLAINEXCEPT to anything (even an empty string) will
enable plain mode.
HGUSER
This is the string used as the author of a commit. If not set,
available values will be considered in this order:
- HGUSER (deprecated)
- configuration files from the HGRCPATH
- EMAIL
- interactive prompt
- LOGNAME (with ``@hostname`` appended)
(deprecated, see :hg:`help config.ui.username`)
EMAIL
May be used as the author of a commit; see HGUSER.
LOGNAME
May be used as the author of a commit; see HGUSER.
VISUAL
This is the name of the editor to use when committing. See EDITOR.
EDITOR
Sometimes Mercurial needs to open a text file in an editor for a
user to modify, for example when writing commit messages. The
editor it uses is determined by looking at the environment
variables HGEDITOR, VISUAL and EDITOR, in that order. The first
non-empty one is chosen. If all of them are empty, the editor
defaults to 'vi'.
PYTHONPATH
This is used by Python to find imported modules and may need to be
set appropriately if this Mercurial is not installed system-wide.