platform_build_kati/affinity.cc, branch master Unnamed repository; edit this file 'description' to name the repository. Refactor source tree into directories 2020-06-27T01:52:06+00:00 Dan Willemsen dwillemsen@google.com 2020-06-27T01:46:21+00:00 979e7ae6e417ae4ee45e835104b66191ae16a14c Now instead of almost every file in the top level, move the old go code into its own directory 'golang', and the C++ code into it's own 'src' Also removes a few obsolete scripts that were used to work on Android before Android fully switched to Kati. 
Now instead of almost every file in the top level, move the old go code
into its own directory 'golang', and the C++ code into it's own 'src'

Also removes a few obsolete scripts that were used to work on Android
before Android fully switched to Kati.
Reformat with clang-format-9 2020-06-12T04:20:36+00:00 Dan Willemsen dwillemsen@google.com 2020-06-12T00:49:36+00:00 546261a74b75581fb35721b8def124821f2edb53 Travis was using clang-format-7 Change-Id: Ia9b6e2490927ebc20ec6549d39ee70c11bfbb766 
Travis was using clang-format-7

Change-Id: Ia9b6e2490927ebc20ec6549d39ee70c11bfbb766
Use std::random_device instead of initializing via getpid 2019-01-11T22:27:33+00:00 Dan Willemsen dwillemsen@google.com 2019-01-11T22:27:33+00:00 59e7b3e05f1b8553f2c6e9fa329543d8b42fca8e getpid worked fine for this until we started trying to run Kati within a pid namespace (via nsjail). Then it consistently got the same pid, which resulted in multiple copies of Kati all running on the same CPU. Change-Id: If50846f742c3898ca92e21e0d2c78ad8e34d0128 
getpid worked fine for this until we started trying to run Kati within a
pid namespace (via nsjail). Then it consistently got the same pid, which
resulted in multiple copies of Kati all running on the same CPU.

Change-Id: If50846f742c3898ca92e21e0d2c78ad8e34d0128
`clang-format -i -style=file *.cc *.h` 2017-10-12T05:20:49+00:00 Dan Willemsen dwillemsen@google.com 2017-10-12T05:17:48+00:00 3ce083f01d6e62e9fb9b328e312d97a074fe6128 Change-Id: I62a87c5d8309b21265e904c0aeb9b3e094c9024a 
Change-Id: I62a87c5d8309b21265e904c0aeb9b3e094c9024a
Randomize cpu affinity 2017-02-24T02:41:50+00:00 Dan Willemsen dwillemsen@google.com 2017-02-23T07:59:06+00:00 5db7f7c46b94424be56db9aa5af4ae537252a703 Continue to lock kati to one or two CPUs, but pick one CPU at random, and pick another CPU next to it. This dramatically speeds up cases where more than one Kati instance is running at a time. There's a multiproduct_build tool in Android that attempts to run Kati on every build configuration present in the tree. My machine has 48 (including hyperthreading) cpu cores, with 64GB of ram. Before this change, it would take ~12 minutes to build all 59 configurations, 12 at a time. After this change it takes 3 minutes. There doesn't seem to be any significant change in timings for standalone builds. It looks like there was 1-2% difference with my previous change that chose two completely random CPUs, but choosing two that are likely hyperthreaded, or at least on the same chip helps. 
Continue to lock kati to one or two CPUs, but pick one CPU at random,
and pick another CPU next to it. This dramatically speeds up cases where
more than one Kati instance is running at a time.

There's a multiproduct_build tool in Android that attempts to run Kati
on every build configuration present in the tree. My machine has 48
(including hyperthreading) cpu cores, with 64GB of ram. Before this
change, it would take ~12 minutes to build all 59 configurations, 12 at
a time. After this change it takes 3 minutes.

There doesn't seem to be any significant change in timings for
standalone builds. It looks like there was 1-2% difference with my
previous change that chose two completely random CPUs, but choosing two
that are likely hyperthreaded, or at least on the same chip helps.
[C++] add +build ignore magic comment for go build 2016-11-28T17:52:11+00:00 Koichi Shiraishi zchee.io@gmail.com 2016-09-06T06:05:35+00:00 df8cd0596bc8b7a92767d9290fdce59fd4355460 Signed-off-by: Koichi Shiraishi <zchee.io@gmail.com> 
Signed-off-by: Koichi Shiraishi <zchee.io@gmail.com>
[C++] Set CPU affinity 2016-02-16T04:56:27+00:00 Shinichiro Hamaji shinichiro.hamaji@gmail.com 2016-02-16T04:49:49+00:00 1a444a870705b263cb4ee080ddfafde42c0f5563 Sticking to a single processor improves the performance while we are running only a single thread. 
Sticking to a single processor improves the performance while
we are running only a single thread.