At 07:45 AM 9/23/2006, [EMAIL PROTECTED] wrote:
>  nice compromise : allocate smaller bunches of blocks : for
>instance allocate 8 blocks to each thread (thread 1 gets blocks 0-7, thread
>2 gets 8-15...), each thread to finish takes the first free bunch of blocks
>when it finishes. Would the number of carries to be stored and to be
>processed afterwards be the drawback of this solution ?

Yes you have identified the drawback.  When you finish blocks 0-7 you must
add the carries into block 8 (and the carries from that into block 
9).  So if you
divide it into too many chunks the carry propagation overhead negates any
benefits.

That doesn't mean there isn't a good middle ground somewhere. 

_______________________________________________
Prime mailing list
[email protected]
http://hogranch.com/mailman/listinfo/prime

Reply via email to