<p dir="ltr"><br>
On Aug 22, 2015 9:02 AM, "Patrascu, Alecsandru" <<a href="mailto:alecsandru.patrascu@intel.com">alecsandru.patrascu@intel.com</a>> wrote:<br>
[snip] <br>
> For instance, as shown from attached sample performance results from the Grand Unified Python Benchmark, >20% speed up was observed.</p>
<p dir="ltr">Are you referring to the tests in the benchmarks repo? [1]</p>
<p dir="ltr">How does the real-world performance improvement compare with other languages you are targeting for optimization?</p>
<p dir="ltr">And thanks for working on this! I have several more questions:</p>
<p dir="ltr">What sorts of future changes in CPython's code might interfere with your optimizations?</p>
<p dir="ltr">What future additions might stand to benefit?</p>
<p dir="ltr">What changes in existing code might improve optimization opportunities?</p>
<p dir="ltr">What is the added maintenance burden of the optimizations on CPython, if any?</p>
<p dir="ltr">What is the performance impact on non-Intel architectures? What about older Intel architectures? ...and future ones?</p>
<p dir="ltr">What is Intel's commitment to supporting these (or other) optimizations in the future? How is the practical EOL of the optimizations managed?</p>
<p dir="ltr">Finally, +1 on adding an opt-in Makefile target rather than enabling the optimizations by default.</p>
<p dir="ltr">Thanks again!</p>
<p dir="ltr">-eric</p>
<p dir="ltr">[1] <a href="https://hg.python.org/benchmarks/">https://hg.python.org/benchmarks/</a></p>