d28bfcd570
This patch refactors the ClassFields benchmarks and makes the results detection work properly. Previously the errors weren't caught since the ClassFields benchmarks don't get run in the perf_integration step in the CI. - Instead of putting different configs (single/multiple fields, type of fields, etc.)in the JSON configuration, we now group the related benchmarks into the same script and run the different configurations in the scripts directly. Only the optimization status is now controlled in JSON. All the class fields definition benchmarks are merged into initialize-class.js. - Update the number of local iterations of evaluate-class.js to 100 (similar to most of other benchmarks) to keep the time spent on this benchmark similar to that of other benchmarks. In addition, copy the configs to JSTests3 so that the benchmarks gets run by the perf_integration step and we can see the graphs on http://chromeperf.appspot.com/report. These can be removed when the ClassFields benchmark results are generated there too. Bug: v8:10793, v8:9888 Change-Id: I4e677bdc7b582650f39cf6e9ec02775c57fd04ab Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3226550 Commit-Queue: Joyee Cheung <joyee@igalia.com> Reviewed-by: Shu-yu Guo <syg@chromium.org> Cr-Commit-Position: refs/heads/main@{#77640} |
||
---|---|---|
.. | ||
benchmarks | ||
bigint | ||
cctest | ||
common | ||
debugger | ||
debugging | ||
fuzzer | ||
fuzzilli | ||
inspector | ||
intl | ||
js-perf-test | ||
memory | ||
message | ||
mjsunit | ||
mkgrokdump | ||
mozilla | ||
test262 | ||
torque | ||
unittests | ||
wasm-api-tests | ||
wasm-js | ||
wasm-spec-tests | ||
webkit | ||
BUILD.gn | ||
OWNERS |