This is a meta bug to verify performance before shipping trusted types. This was done by two means: 1) "try perf", specifically Speedometer 3 on Windows, to check for any regression. 2) Profiling simple test case with "hot injection sinks" to detect any unwanted function call or extra work. For innerHTML we used https://mozilla.pettay.fi/moztests/innerhtml_nonsimple_nostyle_test.xhtml and attachment 9434166 [details] ; for setAttribute we used attachment 9437631 [details]. Both looked ok last time we checked (bug 1909168 and bug 1925468), but we may need to try again later. We also discussed other ideas for potential perf improvements, I'll elaborate later.
Bug 1936014 Comment 0 Edit History
Note: The actual edited comment in the bug view page will always show the original commenter’s name and original timestamp.
This is a meta bug to verify performance before shipping trusted types. This was done by two means: 1) "try perf", specifically Speedometer 3 on Windows, to check for any regression. 2) [Profiling](https://profiler.firefox.com/) with high sampling rate some simple test cases with "hot injection sinks" to detect any unwanted function call or extra work. For innerHTML we used https://mozilla.pettay.fi/moztests/innerhtml_nonsimple_nostyle_test.xhtml and attachment 9434166 [details] ; for setAttribute we used attachment 9437631 [details]. Both looked ok last time we checked (bug 1909168 and bug 1925468), but we may need to try again later. We also discussed other ideas for potential perf improvements, I'll elaborate later.