6ba9f702ba
To make the atomic rules a bit easier to work with, in many of the folders, this adds in cc_library rules to group together the sources from that folder (and subfolders where prudent). We only needs sources because those atoms should have their headers as deps. One issue that was pointed out is that there is currently no way to restrict the inclusion of certain packages, a la, `gn check`. For example, there is no mechanism from stopping a dev from adding #include "modules/canvaskit/WasmCommon.h" to something in //src/core (except circular dependencies). We can probably address that using Bazel's visibility rules as needed: https://docs.bazel.build/versions/main/visibility.html https://docs.bazel.build/versions/main/be/functions.html#package_group It is recommended to look at this CL patchset by patchset. PS1: Update gazelle command to generate rules in more folders. PS2: A few changes to make generation work better. PS3: The result of running make generate in //bazel PS4: Adding the rules to build sksllex, the simplest binary I could find in the Skia repo. PS5: Adding the rules to build skdiff, a more complex binary. I tried a few approaches, but ended up gravitating back towards the layout where we have each folder/package group up the sources. I imagine at some point, we'll have skdiff depend on skia_core or something, which will have things like //src/core, //src/codecs, //src/pathops all bundled together. PS7: Added in the groupings of sources, similar to what we had earlier. I liked these for readability. These helped fix up the //:skia_core build, and by extension, the CanvasKit build. Change-Id: I3faa7c4e821c876b243617aacf0246efa524cbde Bug: skia:12541 Reviewed-on: https://skia-review.googlesource.com/c/skia/+/476219 Reviewed-by: Ben Wagner <bungeman@google.com> Reviewed-by: Leandro Lovisolo <lovisolo@google.com> |
||
---|---|---|
.. | ||
__init__.py | ||
_adb_path.py | ||
_adb.py | ||
_benchresult.py | ||
_hardware_android.py | ||
_hardware_nexus_6p.py | ||
_hardware_pixel2.py | ||
_hardware_pixel_c.py | ||
_hardware_pixel.py | ||
_hardware.py | ||
_os_path.py | ||
BUILD.bazel | ||
README.md | ||
sheet.py | ||
skiaperf.py | ||
skpbench.cpp | ||
skpbench.py |
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
./bin/sk asset download android_ndk_linux /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