Two paths, one peak: a view on high-performance language implementations

AoT, JIT, bytecode, none of this matters if you can&#x27;t control your data layout and access patterns. All this talk of performance and still not a mention of cache misses anywhere.<p>There seems to be a desire for this magic inliner and compiler that fixes all your performance problems when it just doesn&#x27;t work like that.<p>Until you understand your data and have complete control over it none of this other stuff matters.

If &quot;template JIT&quot; means what I think it does, this is basically what you get from Cython[1] or, later, Nuitka[2], only they&#x27;re &quot;template AOT&quot;s.<p>Sadly, it buys you sorely little. CPython&#x27;s bytecodes are &quot;big&quot;, so very little time is spent between them relative to that spent inside them. The motivation to do anything smarter here is lacking.<p>[1] <a href="http:&#x2F;&#x2F;cython.org&#x2F;" rel="nofollow">http:&#x2F;&#x2F;cython.org&#x2F;</a><p>[2] <a href="http:&#x2F;&#x2F;nuitka.net&#x2F;" rel="nofollow">http:&#x2F;&#x2F;nuitka.net&#x2F;</a>

The Leroy paper in the bottom of the comments was worth it.