This question returns every so often. If I recall correctly: - the JIT-compiled code is much, much, much larger than the JS source code, and just reading it from the cache may actually slow down execution of the page; - in many pages, JIT-compiled code actually depends on the interactions between the user and the page, since it optimizes the code that is actively used, so this will result in suboptimal code; - JIT-compiled code actually hardwires into the code a number of in-memory addresses, which are different between successive launches of Firefox, so this will result in code that will crash.
Now, it might be possible to save some information (e.g. shapes, type information, hot code paths, etc.) Cheers, David On 16/10/14 22:45, Just Fill Bugs wrote: > Since the html pages are already cached, why not also cache the JIT > compiled javascript while leaving a page? Shouldn't use too much space > than the text content of the embedding page. Much less space than the > image files embedded in a page. > > _______________________________________________ > dev-platform mailing list > dev-platform@lists.mozilla.org > https://lists.mozilla.org/listinfo/dev-platform -- David Rajchenbach-Teller, PhD Performance Team, Mozilla
signature.asc
Description: OpenPGP digital signature
_______________________________________________ dev-platform mailing list dev-platform@lists.mozilla.org https://lists.mozilla.org/listinfo/dev-platform
