Comment on attachment 8984458 [details] [diff] [review] release-64.patch Review of attachment 8984458 [details] [diff] [review]: ----------------------------------------------------------------- Looks good but needs sync with https://github.com/devtools-html/debugger.html/pull/6490 first. ::: devtools/client/debugger/new/README.mozilla @@ -5,2 @@ > > -Comparison: https://github.com/devtools-html/debugger.html/compare/release-61...release-62 Damn, I landed the version without readme update last time ... ::: devtools/client/debugger/new/test/mochitest/browser_dbg-layout-changes.js @@ -16,4 @@ > const layouts = [ > ["vertical", "window:small"], > ["horizontal", "bottom"], > - ["vertical", "right"], Needs synchronization from https://bugzilla.mozilla.org/show_bug.cgi?id=1192642

Comment on attachment 8984527 [details] [diff] [review] release-64-3.patch Review of attachment 8984527 [details] [diff] [review]: ----------------------------------------------------------------- thanks for the update!

Comment on attachment 8984527 [details] [diff] [review] release-64-3.patch Review of attachment 8984527 [details] [diff] [review]: ----------------------------------------------------------------- Will amend myself, but the commit message is missing the bug number and has a "@" before the reviewer name

Pushed by jdescottes@mozilla.com: https://hg.mozilla.org/integration/mozilla-inbound/rev/0485c263b1d3 Debugger Release v64;r=jdescottes

v64+shouldComponentUpdate vs v64: https://treeherder.mozilla.org/perf.html#/compare?originalProject=try&originalRevision=e80b589512697306f61f858eee17f47029064f04&newProject=try&newRevision=4a8ebb0f85e8bf3c9b0babff83d8bf8fbf02eade&framework=1 v64 vs v63: https://treeherder.mozilla.org/perf.html#/compare?originalProject=try&originalRevision=17c62b0758f7d47077f1e72df021d6db3640b226&newProject=try&newRevision=e80b589512697306f61f858eee17f47029064f04&framework=1 v64+shouldComponentUpdate vs v63: https://treeherder.mozilla.org/perf.html#/compare?originalProject=try&originalRevision=17c62b0758f7d47077f1e72df021d6db3640b226&newProject=try&newRevision=4a8ebb0f85e8bf3c9b0babff83d8bf8fbf02eade&framework=1

v64+shouldComponentUpdate vs v63 : no performance difference. v64 vs v63: -12% on pause, -5% on stepin v64+shouldComponentUpdate vs v64: +11% on pause, +5% on stepin This confirms that removing the shouldComponentUpdate is responsible for the performance regression.