Bugzilla

I see, thank you Serge. I think I understand.

I suspect based on how obscure the feature is, the population of users using userContent.css to customize web pages is quite small.

I suppose for now, the most prudent course of action would be to back this set of patches out, and try to encourage users that only want to style web pages to use something like Stylus[1] instead if they want to avoid this extra main-thread IO during start-up. The vast, vast majority of users that don't use userContent.css won't see much benefit from this patch anyways, and we should probably focus our efforts elsewhere to improve their lives rather than trying to make userContent.css performant.

[1]: https://addons.mozilla.org/en-US/firefox/addon/styl-us/

I see, thank you Serge. I think I understand.

I suspect based on how obscure the feature is, the population of users using userContent.css to customize web pages is quite small.

I suppose for now, the most prudent course of action would be to back this set of patches out, and try to encourage users that only want to style web pages to use something like Stylus([1]) instead if they want to avoid this extra main-thread IO during start-up. The vast, vast majority of users that don't use userContent.css won't see much benefit from this patch anyways, and we should probably focus our efforts elsewhere to improve their lives rather than trying to make userContent.css performant.

[1]: https://addons.mozilla.org/en-US/firefox/addon/styl-us/