AuroraMiddleware/skia2
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity
acdbb1e397
skia2/tools/skpbench
History
Chris Dalton 57ab06c14e Delete mixed samples 
Mixed samples is no longer relevant for Ganesh. DMSAA and the new
Ganesh architecture both rely on full MSAA, and any platform where
mixed samples is supported will ultimately not use the old
architecture.

Change-Id: I5acc745010e090ef26310d92ec6240be2cd494cf
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/399837
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Chris Dalton <csmartdalton@google.com>
2021-04-22 20:11:34 +00:00
..
__init__.py skpbench 2016-09-19 11:03:58 -07:00
_adb_path.py Assert there are some sources when running skpbench 2019-09-11 20:20:33 +00:00
_adb.py skpbench: invoke 'adb wait-for-device' lazily 2019-07-29 16:49:59 +00:00
_benchresult.py skpbench: add "resultsfile" option 2016-11-10 16:08:12 +00:00
_hardware_android.py Move GPU locking logic into generic hardware profile 2021-02-12 23:01:08 +00:00
_hardware_nexus_6p.py [skpbench] experiment with not disabling services 2018-03-01 19:28:44 +00:00
_hardware_pixel2.py skpbench: Add support for Pixel 2 2018-04-30 15:29:13 +00:00
_hardware_pixel_c.py [skpbench] Use android's perf settings for Pixel C 2018-03-19 14:22:55 +00:00
_hardware_pixel.py skpbench: Add support for Pixel 2 2018-04-30 15:29:13 +00:00
_hardware.py skpbench: support pixel phone 2017-10-20 17:52:36 +00:00
_os_path.py Add "AllPathsVolatile" skpbench bots 2020-09-24 19:50:34 +00:00
README.md Move GPU locking logic into generic hardware profile 2021-02-12 23:01:08 +00:00
sheet.py skpbench: add utility to format results for Skia Perf 2016-11-10 20:00:56 +00:00
skiaperf.py find min and max, separared by comma. not minmax. 2019-11-06 16:35:23 +00:00
skpbench.cpp Convert dmsaa to an SkSurfaceProp 2021-04-20 16:24:56 +00:00
skpbench.py Delete mixed samples 2021-04-22 20:11:34 +00:00

README.md

skpbench

skpbench is a benchmarking tool for replaying skp or mksp files on android devices. it achieves a lower variance in framerate by controlling the clock speed and stopping all other processes that could cause interference.

Build

skpbench consists of the skpbench binary which must be built for the phone you intend to run on, and skpbench.py which runs on the machine the phone is connected to via ADB and is the entry point.

The to build skia for android are at https://skia.org/user/build#android and reproduced here.

Download the Android NDK

cipd auth-login
python2 infra/bots/assets/android_ndk_linux/download.py  -t /tmp/ndk

After this is set up once, build skpbench for your target cpu (assumed to be arm64 here for a Pixel 3)

bin/gn gen out/arm64 --args='ndk="/tmp/ndk" target_cpu="arm64" is_debug=false'
ninja -C out/arm64 skpbench

Benchmark an SKP on a connected device.

First, copy the built skpbench binary and an example skp file to the device. (or pull a skp corpus using instructions in the section below)

adb push out/arm64/skpbench /data/local/tmp
adb push /home/nifong/Downloads/foo.skp /data/local/tmp/skps/

Run skpbench.py

python tools/skpbench/skpbench.py \
  --adb \
  --config gles \
  /data/local/tmp/skpbench \
  /data/local/tmp/skps/foo.skp

--adb specifies that it should use adb to the only connected device and run skpbench there. --force is necessary because we don't yet have a configuration to monitor vitals on the Pixel 3. --config gles specifies Open GL ES is the backend GPU config to use.

Additional documentation of arguments is printed by python tools/skpbench/skpbench.py --help

Output appears in the following format

   accum    median       max       min   stddev  samples  sample_ms  clock  metric  config    bench
  0.1834    0.1832    0.1897    0.1707    1.59%      101         50  cpu    ms      gles      foo.skp

accum is the time taken to draw all frames, divided by the number of frames. metric specifies that the unit is ms (milliseconds per frame)

MSKP corpus

A manually collected corpus of MSKPs from around 30 top apps (using skia via HWUI) and of about 20 actions in RenderEngine exists in a google cloud storage folder managed by skia/infra/bots/assets/mskp/upload.py

To download the fileset, first determine the highest current version of the fileset

gsutil ls gs://skia-assets/assets/mskp/

Download the latest version.

gsutil cp gs://skia-assets/assets/mskp/5.zip ~/Downloads

Unzip the archive and adb push it to the device.

To upload a new version of the corpus, use the steps above to download and unzip the last version, change the content however you need, then Use the upload tool, passing the directory of the altered archive (not a zip file). Note that you must provide it as an absolute path.

python upload.py --target_dir=/home/nifong/scratch/new_mskps

The upload script should print a version number. Finally, submit something like https://skia-review.googlesource.com/c/skia/+/304376 to point jobs at the new version.

Production

skpbench is run as a tryjob from gerrit, where it uploads the results to perf.skia.org. Once such job name is Perf-Android-Clang-Pixel4XL-GPU-Adreno640-arm64-Release-All-Android_Skpbench

Perf results are available by querying with this or similar. extra_config = Android_Skpbench sub_result = accum_cpu_ms

Example perf query https://perf.skia.org/e/?queries=extra_config%3DAndroid_Skpbench%26sub_result%3Daccum_cpu_ms