[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Gwydion & threads



> Moreover you do not want to use atomic writes for each
> descriptor_t access, as atomic writes typically have
> memory barriers semantics that will kill the memory system
> performance.

Seems to me it would require 2 different representations, efficient & thread
local, vs slower & global?


-- 
Scott Ribe
scott_ribe@killerbytes.com
http://www.killerbytes.com/
(303) 665-7007 voice