On Mon, Aug 19, 2013 at 11:00 AM, Nicolas B. Pierron <nicolas.b.pier...@mozilla.com> wrote: > > So having hooks on the string allocation sounds like a terrible idea.
A couple of highly predictable branches? Maybe. Fortunately it should be easy to measure. Jim and Nicolas clearly have different ideas about the right way forward here. I haven't been following this thread particularly closely, but it sounds like the analysis Nicolas has mentioned is potentially more powerful, but also a lot more complex and further from being landable. If the taint analysis is (a) useful, and (b) doesn't impact performance adversely when not on, and (c) close to being landable, perhaps it's worth moving forward with it? It wouldn't preclude the more powerful system being done later, should somebody find the time to incorporate it. I claim no great insight on this matter, but this thread has definitely gotten bogged down, and it would be a shame if we ended up with no analysis of any kind as a result. Nick _______________________________________________ dev-tech-js-engine-internals mailing list dev-tech-js-engine-internals@lists.mozilla.org https://lists.mozilla.org/listinfo/dev-tech-js-engine-internals
