News for hackers. Stuff that matters.

A comment on Branch Prediction and the Performance of Interpreters – Don’t Trust Folklore. Parent comment

Re:Branch Prediction and the Performance of Interpreters – D... by sklogic on Tuesday August 11, 2015 @03:44AM (#10039584)
The problem with the OoO is that it's a way too expensive (in terms of power and area) in many cases. You won't ever see OoO in GPUs and DSPs, unlikely in the microcontrollers. So, VLIW and the other "stupid core, smart compiler" approaches are still legitimate and will always remain valuable.
[ Parent | on HN ]
Re:Branch Prediction and the Performance of Interpreters – D... by cfallin on Tuesday August 11, 2015 @12:40PM (#10042426)
Yup, in some domains it definitely still makes sense. GPUs work well for highly-data-parallel applications (they're essentially vector machines, modulo branch divergence) and VLIW-style DSPs work because the code is numerical and easy to schedule at compile time.

I've worked mostly in the "need performance as high as possible for general-purpose code" domain, so I may be biased!

[ Parent | on HN ]