So the given benchmark shows the speed is the same. And looking at the spew of IonMonkey of LICM we do hoist the array.length.
But like said, small changes to the microbenchmark could change the speed due to different things: - compiled with OSR entry - register allocation that require a move - caches - alignment - ... In all these cases more iterations won't help. grtz Hannes On Mon, Jan 20, 2014 at 9:43 PM, Brendan Eich <[email protected]> wrote: > hv1989 wrote: > >> This is really micro micro benchmarking. I mean the difference you >> measure possibly isn't because we output different code, but due to >> processor handling of the code. >> > > Would more iterations help? > > /be > _______________________________________________ dev-tech-js-engine-internals mailing list [email protected] https://lists.mozilla.org/listinfo/dev-tech-js-engine-internals
