In order to simplify pageloader, let's use js files everywhere when loading framescript

the run is over and looks clean, besides one error I am not sure if it's related: https://treeherder.mozilla.org/logviewer.html#?job_id=149960323&repo=try&lineNumber=5798 what do you think ?

(In reply to Tarek Ziadé (:tarek) from comment #2) > the run is over and looks clean, besides one error I am not sure if it's > related: > > https://treeherder.mozilla.org/logviewer. > html#?job_id=149960323&repo=try&lineNumber=5798 > > what do you think ? No I don't believe the h2 win7 failure is related. The perfherder comparision is showing win7 regressions in perf-reftest, perf-reftest-singletons, and tp5n: https://treeherder.mozilla.org/perf.html#/compare?originalProject=mozilla-central&newProject=try&newRevision=27a7ce2a27a8d9a573d8a8d9620981ea23f99089&framework=1&selectedTimeRange=172800 I've added win 10 jobs to the run, let's wait and see how they look.

Perfherder compare [1] with moz-central shows significant perf regressions in perf-reftest (& singletons), tresize, and a very large regression (91%) in tp5n main_normal_fileio :( [1] https://treeherder.mozilla.org/perf.html#/compare?originalProject=mozilla-central&newProject=try&newRevision=27a7ce2a27a8d9a573d8a8d9620981ea23f99089&framework=1&selectedTimeRange=172800 I've pushed the patch's parent revision to try also and will compare to see if it's consistent with this.

tp5n fileio is an issue with try vs production- please ignore that. tresize and perf-reftest are actually artifacts of an improvement a few days ago- I assume the base revision of this is from a few days ago, not today :) You can see if you view the graph and then select 14 days.

(In reply to Joel Maher ( :jmaher) (UTC-5) from comment #5) > tp5n fileio is an issue with try vs production- please ignore that. Oh awesome ok good to know thanks! > tresize and perf-reftest are actually artifacts of an improvement a few days > ago- I assume the base revision of this is from a few days ago, not today :) > You can see if you view the graph and then select 14 days. Ahhh, I should have checked that, ok that's good news then thanks (and patch numbers look good).

Comment on attachment 8934646 [details] Bug 1423295 - Make pageloader use plain .js files - https://reviewboard.mozilla.org/r/205520/#review211580 Awesome - so much cleaner, thanks for this!

> Ahhh, I should have checked that, ok that's good news then thanks (and patch numbers look good). So, should I land this or do we need more comparisons at this point?

I believe we are ok to land if rwood thinks the changes are a benefit to pageloader (which from what I understand is a yes)

Pushed by tziade@mozilla.com: https://hg.mozilla.org/integration/autoland/rev/455daafcc7b6 Make pageloader use plain .js files - r=rwood