Author: Eugene Nalimov
Date: 11:13:10 10/30/02
Go up one level in this thread
ftp://download.intel.com/design/perftool/cbts/appnotes/sse2/w_spinlock.pdf Thanks, Eugene On October 30, 2002 at 13:10:59, Robert Hyatt wrote: >On October 30, 2002 at 12:33:00, Eugene Nalimov wrote: > >>On October 30, 2002 at 10:53:58, Robert Hyatt wrote: >> >>>On October 30, 2002 at 06:59:21, Terje Vagle wrote: >>> >>>>Hi all, >>>> >>>>The new cpu from intel will have a new function called >>>>hyper-threading. >>>> >>>>This will make the operating system able to recognize the cpu as if it was >>>>2 cpu's. >>>> >>>>Could the programs with smp-support make use of this? >>>> >>>>Regards, >>>> >>>>Terje Vagle >>> >>> >>>Yes, but it is going to cause some problems. You don't _really_ have two cpus. >>>You >>>have a single cpu executing two threads, and when one thread "blocks" waiting on >>>a >>>memory access or anything else (FP result) then the other thread will run. >>> >>>Spinlocks will wreck this however... >> >>There is special new instruction PAUSE for such a spinlocks. It is handled as >>NOP on all existing (Intel and non-Intel) x86 CPUs prior to P4. It suspends a >>(hyper)thread on P4 CPU for undisclosed amount of clock cycles. >> >>Thanks, >>Eugene > > >basically like sched_yield() does in the linux process scheduler? Simply tells >the >scheduler "if you got anything else to run, run it, otherwise come back to me >and I'll >continue."???
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.