aminer
unread,Sep 24, 2013, 5:48:10 PM9/24/13You do not have permission to delete messages in this group
Either email addresses are anonymous for this group or you need the view member email addresses permission to view the original message
to
Hello,
I have wrote this about my distributed priority and non priority LOCK :
"So if you have noticed, in fact i am using 2 CASes , so my algorithm is
good."
You will say that i will in fact use 3 atomic operations , 2 CASes and one
inside the waitfree queue, but be smart please, when there is
threads waiting and spining , one CAS among the CASes will be very
cheap cause the variable will be inside the L2 cache, and the other CAS
will be
expensive cause the threads must load the variable from the
remote caches and this is expensive, so it is as we would
have one CAS cause one amongst the second is very cheap,
so it will make this a total of 2 CASes if we count the CAS for
the waitfree queue.
This was my algorithm for a distributed priority and non priority LOCK
that is efficient, and i will code it as soon as possible.
Thank you,
Amine Moulay Ramdane.