##// END OF EJS Templates
upstream » mercurial-mirror
RSS

Clone from https://www.mercurial-scm.org/repo/hg-all

py3: make stdout line-buffered if connected to a TTY...
py3: make stdout line-buffered if connected to a TTY Status messages that are to be shown on the terminal should be written to the file descriptor before anything further is done, to keep the user updated. One common way to achieve this is to make stdout line-buffered if it is connected to a TTY. This is done on Python 2 (except on Windows, where libc, which the CPython 2 streams depend on, does not properly support this). Python 3 rolls it own I/O streams. On Python 3, buffered binary streams can't be set line-buffered. The previous code (added in 227ba1afcb65) incorrectly assumed that on Python 3, pycompat.stdout (sys.stdout.buffer) is already line-buffered. However the interpreter initializes it with a block-buffered stream or an unbuffered stream (when the -u option or the PYTHONUNBUFFERED environment variable is set), never with a line-buffered stream. One example where the current behavior is unacceptable is when running `hg pull https://www.mercurial-scm.org/repo/hg` on Python 3, where the line "pulling from https://www.mercurial-scm.org/repo/hg" does not appear on the terminal before the hg process blocks while waiting for the server. Various approaches to fix this problem are possible, including: 1. Weaken the contract of procutil.stdout to not give any guarantees about buffering behavior. In this case, users of procutil.stdout need to be changed to do enough flushes. In particular, 1. either ui must insert enough flushes for ui.write() and friends, or 2. ui.write() and friends get split into flushing and fully buffered methods, or 3. users of ui.write() and friends must flush explicitly. 2. Make stdout unbuffered. 3. Make stdout line-buffered. Since Python 3 does not natively support that for binary streams, we must implement it ourselves. (2.) is problematic because using unbuffered I/O changes the performance characteristics significantly compared to line-buffered (which is used on Python 2) and this would be a regression. (1.2.) and (1.3) are a substantial amount of work. It’s unclear whether the added complexity would be justified, given that raw performance doesn’t matter that much when writing to a terminal much faster than the user could read it. (1.1.) pushes complexity into the ui class instead of separating the concern of how stdout is buffered. Other users of procutil.stdout would still need to take care of the flushes. This patch implements (3.). The general performance considerations are very similar to (1.1.). The extra method invocation and method forwarding add a little more overhead if the class is used. In exchange, it doesn’t add overhead if not used. For the benchmarks, I compared the previous implementation (incorrect on Python 3), (1.1.), (3.) and (2.). The command was chosen so that the streams were configured as if they were writing to a TTY, but actually write to a pager, which is also the default: HGRCPATH=/dev/null python3 ./hg --cwd ~/vcs/mozilla-central --time --pager yes --config pager.pager='cat > /dev/null' status --all previous: time: real 7.880 secs (user 7.290+0.050 sys 0.580+0.170) time: real 7.830 secs (user 7.220+0.070 sys 0.590+0.140) time: real 7.800 secs (user 7.210+0.050 sys 0.570+0.170) (1.1.) using Yuya Nishihara’s patch: time: real 9.860 secs (user 8.670+0.350 sys 1.160+0.830) time: real 9.540 secs (user 8.430+0.370 sys 1.100+0.770) time: real 9.830 secs (user 8.630+0.370 sys 1.180+0.840) (3.) using this patch: time: real 9.580 secs (user 8.480+0.350 sys 1.090+0.770) time: real 9.670 secs (user 8.480+0.330 sys 1.170+0.860) time: real 9.640 secs (user 8.500+0.350 sys 1.130+0.810) (2.) using a previous patch by me: time: real 10.480 secs (user 8.850+0.720 sys 1.590+1.500) time: real 10.490 secs (user 8.750+0.750 sys 1.710+1.470) time: real 10.240 secs (user 8.600+0.700 sys 1.590+1.510) As expected, there’s no difference on Python 2, as exactly the same code paths are used: previous: time: real 6.950 secs (user 5.870+0.330 sys 1.070+0.770) time: real 7.040 secs (user 6.040+0.360 sys 0.980+0.750) time: real 7.070 secs (user 5.950+0.360 sys 1.100+0.760) this patch: time: real 7.010 secs (user 5.900+0.390 sys 1.070+0.730) time: real 7.000 secs (user 5.850+0.350 sys 1.120+0.760) time: real 7.000 secs (user 5.790+0.380 sys 1.170+0.710)
Manuel Jacob -
r45477:f9734b2d default
Show More
Name Size Modified Last Commit Author
/ rust
.cargo
chg
hg-core
hg-cpython
hgcli
Cargo.lock Loading ...
Cargo.toml Loading ...
README.rst Loading ...

Mercurial Rust Code

This directory contains various Rust code for the Mercurial project. Rust is not required to use (or build) Mercurial, but using it improves performance in some areas.

There are currently three independent rust projects: - chg. An implementation of chg, in rust instead of C. - hgcli. A project that provide a (mostly) self-contained "hg" binary,

for ease of deployment and a bit of speed, using PyOxidizer. See hgcli/README.md.
  • hg-core (and hg-cpython): implementation of some functionality of mercurial in rust, e.g. ancestry computations in revision graphs, status or pull discovery. The top-level Cargo.toml file defines a workspace containing these crates.

Using Rust code

Local use (you need to clean previous build artifacts if you have built without rust previously):

$ make PURE=--rust local # to use ./hg
$ ./tests/run-tests.py --rust # to run all tests
$ ./hg debuginstall | grep -i rust # to validate rust is in use
checking Rust extensions (installed)
checking module policy (rust+c-allow)

If the environment variable HGWITHRUSTEXT=cpython is set, the Rust extension will be used by default unless --no-rust.

One day we may use this environment variable to switch to new experimental binding crates like a hypothetical HGWITHRUSTEXT=hpy.

Profiling

Setting the environment variable RUST_LOG=trace will make hg print a few high level rust-related performance numbers. It can also indicate why the rust code cannot be used (say, using lookarounds in hgignore).

py-spy (https://github.com/benfred/py-spy) can be used to construct a single profile with rust functions and python functions (as opposed to hg --profile, which attributes time spent in rust to some unlucky python code running shortly after the rust code, and as opposed to tools for native code like perf, which attribute time to the python interpreter instead of python functions).

Developing Rust

The current version of Rust in use is 1.34.2, because it's what Debian stable has. You can use rustup override set 1.34.2 at the root of the repo to make it easier on you.

Go to the hg-cpython folder:

$ cd rust/hg-cpython

Or, only the hg-core folder. Be careful not to break compatibility:

$ cd rust/hg-core

Simply run:

$ cargo build --release

It is possible to build without --release, but it is not recommended if performance is of any interest: there can be an order of magnitude of degradation when removing --release.

For faster builds, you may want to skip code generation:

$ cargo check

For even faster typing:

$ cargo c

You can run only the rust-specific tests (as opposed to tests of mercurial as a whole) with:

$ cargo test --all

Formatting the code

We use rustfmt to keep the code formatted at all times. For now, we are using the nightly version because it has been stable enough and provides comment folding.

To format the entire Rust workspace:

$ cargo +nightly fmt

This requires you to have the nightly toolchain installed.