On Wed, Mar 11, 2015 at 2:34 PM, David Bruant <bruan...@gmail.com> wrote: > Le 10/03/2015 23:48, Jonas Sicking a écrit : >> >> Even more importantly, if http://site-a.com opens an <iframe> to >> http://site-b.com/widget.html, it's expected that the iframe is loaded using >> site-b's cookies. While we theoretically could use a different child process >> to load the <iframe>, Gecko isn't able to do this yet. >> >> What I'm hoping we can do eventually, is just that though. I.e. make each >> origin load in a separate process. Even if the origin is loaded inside an >> <iframe> inside another origin. > > Note that this is tedious as two contexts with originally different origins > may end up with the same origin if they set document.domain on both sides. > However, this question is taken care of for @sandbox-ed iframes on the > standards side as well as in Firefox and Chrome. (See > https://www.w3.org/Bugs/Public/show_bug.cgi?id=23040 )
Yeah. In practice it likely wouldn't be a process per origin, but rather a process per eTLD+1. Possibly there could be some way (http header) for a page to promise that it won't document.domain, which would allow us to load that page in an origin-specific process. >> That way we could use process boundaries to ensure that a given origin is >> not able to read data that belongs to another origin. >> >> Hopefully this is something that the platform team will start looking at >> once the initial process separation work is done for desktop. > > I am looking forward to this moment. > And I believe this relates to the question at hand about vendor-specific > APIs. Mozilla has expressed the intention to implement new API to have > <canvas> handled in a Worker (and I haven't read interest for that from > other browsers, please correct me if I'm mistaken). However, it appears that > process-isolated iframes should lead to the same performance benefits > without needing any new API (but perhaps strongly encouraging other vendors > to improve their iframe@sandbox implementation). It's not at all obvious to me that process-isolated <iframe>s would be a performance gain at all. The overhead of running more processes, could offset any separation of threads. / Jonas _______________________________________________ dev-b2g mailing list dev-b2g@lists.mozilla.org https://lists.mozilla.org/listinfo/dev-b2g
