Messages in this thread | | | Date | Mon, 20 Jul 1998 03:45:20 +0200 (CEST) | From | MOLNAR Ingo <> | Subject | Re: gettimeofday() vs. /proc/rtc |
| |
On Sun, 19 Jul 1998, Richard B. Johnson wrote:
> How would you get a system call from user mode to take less than 1 > microsecond, is really what you are asking. [...]
actually, if you look at the SYSENTER/SYSEXIT patch and testcode posted recently, with that stuff a PII does a getpid() in ~150 cycles. This means a 0.5 microseconds system call on a 300MHz PII, 0.33 microseconds on a 450MHz PII. So Intel has indeed fixed the latency of system entry.
with that stuff, gettimeofday() takes ~300 cycles, thats under 1 microsecond too. (unless you have CONFIG_APM turned on, but people have posted patches to fix that case too)
-- mingo
- 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.altern.org/andrebalsa/doc/lkml-faq.html
| |