[pypy-dev] Static bytecode instruction optimization and pypy (JIT?) optimization

[Rocky Bernstein]

Yes, that's true, but I believe that such decisions should offered to the
programmers rather than dictated.  There is also a whole other discussion on

* how/where optimizations/transformations could  be recorded
* how tracebacks could be handled
* debugging optimized code

But here I wanted to focus on just the optimization interaction. If you are
implying that pypy doesn't do changes that would make such behaviorial
changes, then that would suggest there may be a lot of benefit static
compiler optimization.

On Sun, Jul 9, 2017 at 9:41 PM, rymg19 at gmail.com <rymg19 at gmail.com> wrote:

> Part of the problem is behavioral changes. If you implement tail call
> recursion, your tracebacks change. Common subexpression elimination will
> also have subtle behavioral changes.
>
>
> --
> Ryan (ライアン)
> Yoko Shimomura, ryo (supercell/EGOIST), Hiroyuki Sawano >> everyone elsehttp://refi64.com
>
> On Jul 9, 2017 at 8:11 PM, <Rocky Bernstein <rb at dustyfeet.com>> wrote:
>
> In looking at Python bytecode over the years, I've noticed that it does
> very little to no traditional static-compiler optimization. Specifically:
>
> * Dead code elmination (most of the time)
> * Jumps to Jumps or Jumps to the next instruction
> * Constant propagation (most of the time)
> * Common subexpression elimination
> * Tail recursion
> * Code hoisting/sinking
>
> Yes, over the years more compiler optimization has been done but it's
> still pretty sparse.
>
> The little that I've looked at pypy, it is pretty much the same thing at
> the Python bytecode level. That's why a python decompiler for say 2.7 will
> work largely unmodified for he corresponding pypy 2.7 version. Same is true
> 3.2 versus pypy 3.2.
>
> I understand though that pypy can do and probably does some of the
> optimization when it JITs. But my question is: if traditional compiler
> optimization were done would this hinder, be neutral, or help pypy's
> optimization.
>
> Of course, if there were static compiler optimization of the type
> described above, that might be a win when JITing doesn't kick in. (And
> perhaps then who cares)
>
> But I'm interested in the other situation where both are in play.
> _______________________________________________ pypy-dev mailing list
> pypy-dev at python.org https://mail.python.org/mailman/listinfo/pypy-dev
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/pypy-dev/attachments/20170711/1a913788/attachment.html>