News for hackers. Stuff that matters.

A comment on A tool to make Windows builds reproducible. Parent comment

Re:A tool to make Windows builds reproducible by CJefferson on Wednesday December 07, 2016 @03:25PM (#13125892)
It's common for small changes in code, or even the alignment of code (moving it all up/down memory) to have surprising results in the performance of the resulting code, due to things like page alignment.

Here is a random example (you can find lots of others):

    http://stackoverflow.com/questions/17896714/why-would-introducing-useless-mov-instructions-speed-up-a-tight-loop-in-x86-64-a
[ Parent | on HN ]
Re:A tool to make Windows builds reproducible by metaphor on Thursday December 08, 2016 @02:16AM (#13128910)
Hmmm...perhaps a rephrasing of the question given this particular example.

Which would you consider to be the surprising part: ~0.75% average execution speed slowdown, or on the order of tens of milliseconds slowdown?

W.r.t. page alignment, thought you may find this to be an interesting counter example[1].

[1] http://danluu.com/3c-conflict/

[ Parent | on HN ]