Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > comp.programming > #3789 > unrolled thread
| Started by | aminer <aminer@toto.net> |
|---|---|
| First post | 2013-09-24 14:48 -0700 |
| Last post | 2013-09-24 14:48 -0700 |
| Articles | 1 — 1 participant |
Back to article view | Back to comp.programming
About my distributed priority and non priority LOCK... aminer <aminer@toto.net> - 2013-09-24 14:48 -0700
| From | aminer <aminer@toto.net> |
|---|---|
| Date | 2013-09-24 14:48 -0700 |
| Subject | About my distributed priority and non priority LOCK... |
| Message-ID | <l1smpk$1op$1@news.albasani.net> |
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.
Back to top | Article view | comp.programming
csiph-web