We currently do a bunch of unnecessary things when we create bootstrap scopes, including (expensively) statting the extension file to check that it exists so that we can create a dummy scope if it doesn't (just creating an ordinary bootstrap scope and failing to load bootstrap.js works just as well), and adding a bunch of properties that no-one uses anymore. This can all go away.

Comment on attachment 8976404 [details] Bug 1462223: Remove unnecessary/unused bootstrap scope setup code. https://reviewboard.mozilla.org/r/244550/#review250932

For reasons that I don't understand yet, removing the workerbootstrap extension makes pretty much all mochitests fail. It apparently prevents the delayed framescripts they register from being loaded into new processes. The extension doesn't have to do anything. Just to exist and have a valid install.rdf manifest. I guess it has something to do with its presence causing the AOM to do something different at startup. Anyway, I'm going to leave the skeleton XPI there for now, and file a follow-up to figure out what's going on.

Never mind. Bug 1460610 apparently already exists.

https://hg.mozilla.org/integration/mozilla-inbound/rev/1bd01365736af6016a39f97f8dd2b68cb3ad491d Bug 1462223: Remove unnecessary/unused bootstrap scope setup code. r=aswan

https://hg.mozilla.org/integration/mozilla-inbound/rev/ccfe668f71ffded4b72485805b56fe8040bc0d9c Bug 1462223: Follow-up: Ignore worker bootstrap test extension in unrelated test. r=bustage,test-only CLOSED TREE

https://hg.mozilla.org/mozilla-central/rev/1bd01365736a https://hg.mozilla.org/mozilla-central/rev/ccfe668f71ff

Is manual testing required on this bug? If Yes, please provide some STR and the proper webextension(if required), if No set the “qe-verify-“ flag.

OK. I see the problem now.