Sometimes, there is a need to see how DevTools performance has changed over a longer period of time, for example half a year. During such a time frame, it is possible that our performance tests have been updated and refined, and we cannot do a comparison between a Talos run from 6 months ago with a new release, simply because the two tests were testing for different data. To solve this, we backport the talos tests from the more modern Firefox, and try to approximate as closely as possible so that we get a more accurate result.
- two revisions that are being tested. If you are backporting between FF versions, a useful list of commit hashes can be