shunwang / numexpr

Automatically exported from code.google.com/p/numexpr
MIT License
0 stars 0 forks source link

Multi-threading code should be disabled for small arrays #36

Closed GoogleCodeExporter closed 9 years ago

GoogleCodeExporter commented 9 years ago
When working with small arrays, the threaded code should be avoided for getting 
better performance.  See this thread for more info:

http://mail.scipy.org/pipermail/scipy-user/2010-November/027263.html

Original issue reported on code.google.com by fal...@gmail.com on 1 Nov 2010 at 9:18

GoogleCodeExporter commented 9 years ago
Fixed in r270.  Turns out that the even point is 32 KB, that is, the L1 cache 
size for my processor (which should be a good default nowadays).

Original comment by fal...@gmail.com on 25 Jan 2011 at 10:45