@bascule word's on street that STM approaches the perf of HTM. I'm not really sure about it, though.
Funny thing about HTM: even though heavily multithreaded/multicore Java programs really needed it, Sun never managed to ship a SPARC with it
-
-
-
@whitequark no, there's a 30% perf hit to do it in software since you need indirection around each pointer access. That's free with HTM -
@bascule I imagine the behavior is more complex (wrt/ perf) than that, the HTM being routed through cache coherency circuitry -
@whitequark I'm more or less paraphrasing Cliff Click... I can link you the video if you're interested -
@bascule please do, I am. - End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.