Overview
Examples
Screenshots
Comparisons
Applications
Download
Documentation
Tutorials
Bazaar
Status & Roadmap
FAQ
Authors & License
Forums
Funding Ultimate++
Search on this site
Search in forums












SourceForge.net Logo
Home » Developing U++ » U++ Developers corner » memcpy() and memmove() optimization
memcpy() and memmove() optimization [message #53766] Tue, 28 April 2020 15:05
Tom1
Messages: 1212
Registered: March 2007
Senior Contributor
Hi Mirek,

This is just a reminder here to look at after releasing 2020.1.. Wink

But: Can we use this? There's no licensing information as far as I can see.

https://github.com/rukkal/apex-memmove

At least it is faster than anything else I've seen so far and compiles well on both CLANGx64 and MSBT19x64 on Windows.

MSBT19x64 on Core i7:

TIMING apex_memcpy    :  3.55 s  -  3.55 ms ( 3.55 s  / 1000 ), min:  3.28 ms, max:  7.80 ms, nesting: 0 - 1000
TIMING std::copy_n    :  4.46 s  -  4.46 ms ( 4.46 s  / 1000 ), min:  4.20 ms, max:  8.06 ms, nesting: 0 - 1000
TIMING memcpy         :  4.37 s  -  4.37 ms ( 4.37 s  / 1000 ), min:  4.08 ms, max:  7.98 ms, nesting: 0 - 1000

CLANGx64 on Core i7:

TIMING apex_memcpy    :  3.54 s  -  3.54 ms ( 3.54 s  / 1000 ), min:  3.23 ms, max:  7.85 ms, nesting: 0 - 1000
TIMING std::copy_n    :  4.45 s  -  4.45 ms ( 4.45 s  / 1000 ), min:  4.19 ms, max:  8.40 ms, nesting: 0 - 1000
TIMING memcpy         :  4.33 s  -  4.33 ms ( 4.33 s  / 1000 ), min:  4.05 ms, max:  8.07 ms, nesting: 0 - 1000


Best regards,

Tom
 
Read Message
Previous Topic: Skylark - Render 2 tables -- [SOLVED]
Next Topic: Rapsberry PI - cpu dilemma
Goto Forum:
  


Current Time: Fri Mar 29 01:34:32 CET 2024

Total time taken to generate the page: 0.01525 seconds