search

lion03 / thrust

Automatically exported from code.google.com/p/thrust
Apache License 2.0
0 stars 0 forks source link

selectively enable use of fast_scan for non-pod types #203

Closed GoogleCodeExporter closed 8 years ago

GoogleCodeExporter commented 8 years ago 
CUDA 3.1 and higher permit the use of non-pod types in statically-allocated 
__shared__ memory.  As a result cuda::detail::fast_scan can be dispatched in 
more cases than before.

This ought to significantly improve performance of the reduce_by_key and 
*_scan_by_key algorithms

Original issue reported on code.google.com by wnbell on 3 Sep 2010 at 1:36

GoogleCodeExporter commented 8 years ago 
This issue was closed by revision 1179c8e3b1.

Original comment by wnbell on 3 Sep 2010 at 2:15