Messages in this thread | | | Date | Sat, 9 Jan 1999 02:00:46 -0500 | From | Paul Barton-Davis <> | Subject | using the pentium cycle counter on MP's |
| |
is the cycle counter on any one processor guaranteed to be within a certain distance of the value on all the other processors, or does a restart leave them pretty much randomly staggered by a few microseconds or more ?
i'm worried about a thread that uses rdtsc in user space migrating from cpu to cpu, and getting very different values from each counter. i looked at the intel docs that i could track down on the web, but couldn't find an answer to this. i don't understand the kernel code that plays with this stuff.
does linux do anything to prevent the use of the cycle counters from being used as extremely accurate real time clocks (i'm thinking of the idle stuff - does this in anyway stop the cycle counter from advancing ?)
--p
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |