multithreading - Average latency of atomics cmpxchg instructions on Intel Cpus -



looking reference on average latencies lock cmpxchg instruction various intel processors. not able locate reference on topic , reference help.

thanks.

there few, if any, references on because there variation. depends on including bus speed, memory speed, processor speed, processor count, surrounding instructions, memory fencing , quite possibly angle between moon , mt everest...

if have specific application, in, known (fixed) hardware, operating environment, real-time operating system , exclusive control, maybe matter. in case, benchmark. if don't have level of control on software running, measurements meaningless.

as discussed in these answers, locks implemented using cas, if can away cas instead of lock (which need @ least 2 operations) faster (noticeably? maybe).

the best references find intel software developer's manuals, though since there variation won't give actual number. will, however, describe how best performance possible. possibly processor datasheet (such here i7 extreme edition, under "technical documents") give actual numbers (or @ least range).


Comments

Popular posts from this blog

android - Spacing between the stars of a rating bar? -

aspxgridview - Devexpress grid - header filter does not work if column is initially hidden -

c# - How to execute a particular part of code asynchronously in a class -