Bugzilla

Okay so looks like it times out on safari because of this variable being too large in `cpuBenchmark.js`?
https://github.com/sitespeedio/browsertime/blob/main/browserscripts/browser/cpuBenchmark.js#L5

I verified this by manually knocking off a couple zeros locally, and it stopped timing out and fully ran. (I wonder if this is the reason for the other benchmark timeout failures)

Not sure why this affects Safari matrix-react-bench, but not some of the other passing benchmarks.

So I plan to make a patch that adds a safari+mrb exception in `_kill_browsertime_process()`
https://searchfox.org/mozilla-central/source/testing/raptor/raptor/browsertime/base.py#622

 (this func was originally made to handle such exceptions during the line handling anyway e.g. the current xpath 2fa thing for login logic that is present there)

Okay so looks like it times out on safari because of this variable being too large in `cpuBenchmark.js`?
https://github.com/sitespeedio/browsertime/blob/main/browserscripts/browser/cpuBenchmark.js#L5

I verified this by manually knocking off a couple zeros locally, and it stopped timing out and is running. (I wonder if this is the reason for the other benchmark timeout failures)

Not sure why this affects Safari matrix-react-bench, but not some of the other passing benchmarks.

So I plan to make a patch that adds a safari+mrb exception in `_kill_browsertime_process()`
https://searchfox.org/mozilla-central/source/testing/raptor/raptor/browsertime/base.py#622

 (this func was originally made to handle such exceptions during the line handling anyway e.g. the current xpath 2fa thing for login logic that is present there)

Okay so looks like it times out on safari because of this variable being too large in `cpuBenchmark.js`?
https://github.com/sitespeedio/browsertime/blob/main/browserscripts/browser/cpuBenchmark.js#L5

I verified this by manually knocking off a couple zeros locally, and it stopped timing out and is running. (I wonder if this is the reason for the other benchmark timeout failures)

Not sure why this affects Safari matrix-react-bench, but not some of the other passing benchmarks.

So I plan to make a patch that adds a safari+mrb exception in `_kill_browsertime_process()`
https://searchfox.org/mozilla-central/source/testing/raptor/raptor/browsertime/base.py#622

 (this func was originally made to handle such exceptions during the line handling anyway e.g. the current xpath 2fa thing for login logic that is present there)


**edit**
never mind, that approach doesn't quite work, as ignoring the timeouterror from cpubenchmark in the proposed way still has seleniumrunner attempt to add undefined data to the browser which creates an error..

instead - do we even need cpuBenchmark.js script for matrix react bench? if not then adding something like `  delete commands.measure.scriptsByCategory.browser.cpuBenchmark;`  to [browsertime_benchmark.js](https://searchfox.org/mozilla-central/source/testing/raptor/browsertime/browsertime_benchmark.js) may suffice

Okay so looks like it times out on safari because of this variable being too large in `cpuBenchmark.js`?
https://github.com/sitespeedio/browsertime/blob/main/browserscripts/browser/cpuBenchmark.js#L5

I verified this by manually knocking off a couple zeros locally, and it stopped timing out and is running. (I wonder if this is the reason for the other benchmark timeout failures)

Not sure why this affects Safari matrix-react-bench, but not some of the other passing benchmarks.

So I plan to make a patch that adds a safari+mrb exception in `_kill_browsertime_process()`
https://searchfox.org/mozilla-central/source/testing/raptor/raptor/browsertime/base.py#622

 (this func was originally made to handle such exceptions during the line handling anyway e.g. the current xpath 2fa thing for login logic that is present there)


**edit**
never mind, that approach doesn't quite work, as ignoring the timeouterror from cpubenchmark in the proposed way still has seleniumrunner attempt to add undefined data to the browser which creates an error..