commit | f3a4ced0a4a6a9a71e3d2247d5f3e2608da04df6 | [log] [tgz] |
---|---|---|
author | Dan Willemsen <dwillemsen@google.com> | Tue Aug 25 14:22:35 2015 -0700 |
committer | Dan Willemsen <dwillemsen@google.com> | Mon Aug 31 19:25:58 2015 -0700 |
tree | f383be9cec99b4ac4bbd95bd859ab0644bedb86d | |
parent | 3faa60f59806427411be456f8a71480f96fdfb40 [diff] |
[C++] Fix ninja EmitNode early exit Change-Id: I2b574a71472508515e9d06e55ff2c2e1a5292c2b
kati is an experimental GNU make clone. The main goal of this tool is to speed-up incremental build of Android.
Currently, kati does not offer a faster build by itself. It instead converts your Makefile to a ninja file.
Set up kati:
% cd ~/src % git clone https://github.com/google/kati % cd kati % make
Build Android:
% cd <android-directory> % source build/envsetup.sh % lunch <your-choice> % ~/src/kati/m2n --kati_stats # Use --goma if you are a Googler. % ./ninja.sh
You need ninja in your $PATH.
% ./ninja.sh -t clean
Note ./ninja.sh passes all parameters to ninja.
For example, the following is equivalent to "make cts":
% ./ninja.sh cts
Or, if you know the path you want, you can do:
% ./ninja.sh out/host/linux-x86/bin/adb
% ~/src/kati/m2n -j10 % ./ninja.sh
Or
% ./ninja.sh -j10
Note the latter kills the parallelism of goma.