556350de37
We have a similar ingestion strategy to Gold. I tried to use something off the shelf like benchmark.js but passing the PathKit context into the benchmarks was non-trivial. Plus, making a basic benchmarking tool ended up being not too hard. We should be able to re-use the docker container/aggregator for CanvasKit too. Bug: skia: Change-Id: I613dfc58ea57c31cf71566a8ac55f8df9272ad25 Reviewed-on: https://skia-review.googlesource.com/c/161620 Commit-Queue: Kevin Lubick <kjlubick@google.com> Reviewed-by: Joe Gregorio <jcgregorio@google.com> Reviewed-by: Stephan Altmueller <stephana@google.com>
34 lines
1.2 KiB
Bash
Executable File
34 lines
1.2 KiB
Bash
Executable File
#!/bin/bash
|
|
# Copyright 2018 Google LLC
|
|
#
|
|
# Use of this source code is governed by a BSD-style license that can be
|
|
# found in the LICENSE file.
|
|
|
|
# This assumes it is being run inside a docker container of perf-karma-chrome-tests
|
|
# and a Skia checkout has been mounted at /SRC and the output directory
|
|
# is mounted at /OUT
|
|
|
|
# For example:
|
|
# docker run -v $SKIA_ROOT:/SRC -v /tmp/dockerout:/OUT gcr.io/skia-public/perf-karma-chrome-tests:68.0.3440.106_v1 /SRC/infra/pathkit/perf_pathkit.sh
|
|
|
|
set -ex
|
|
|
|
#BASE_DIR is the dir this script is in ($SKIA_ROOT/infra/pathkit)
|
|
BASE_DIR=`cd $(dirname ${BASH_SOURCE[0]}) && pwd`
|
|
PATHKIT_DIR=$BASE_DIR/../../modules/pathkit
|
|
|
|
# Start the aggregator in the background
|
|
/opt/perf-aggregator $@ &
|
|
# Run the tests 10 times to get a wide set of data
|
|
for i in `seq 1 10`;
|
|
do
|
|
npx karma start $PATHKIT_DIR/karma.bench.conf.js --single-run
|
|
done
|
|
# Tell the aggregator to dump the json
|
|
# This curl command gets the HTTP code and stores it into $CODE
|
|
CODE=`curl -s -o /dev/null -I -w "%{http_code}" -X POST localhost:8081/dump_json`
|
|
if [ $CODE -ne 200 ]; then
|
|
# If we don't get 200 back, something is wrong with writing to disk, so exit with error
|
|
exit 1
|
|
fi
|