Messages in this thread | | | Date | Thu, 15 Jul 2004 19:44:13 -0700 (PDT) | From | Christoph Lameter <> | Subject | Re: gettimeofday nanoseconds patch (makes it possible for the posix-timer functions to return higher accuracy) |
| |
On Thu, 15 Jul 2004, George Anzinger wrote:
> As to accuracy, the more "accurate" way is to change get_offset to return > nanoseconds. This way there is only one round off (by the divide) instead of > two (the get_offset and the divide). I ran into this problem in the latest HRT > patch. One of my tests is to do a gettimeofday and a clock_gettime and make > sure there is no "backward" stuff happening. Test failed by 1 micro second > "some" of the time because of this double round off.
Well yes this is what the interpolator etc does on IA64. time_interpolator_get_offset returns nanoseconds. I have done tests to insure that no backward stuff happens. Tests were done with the mentioned patch and clock_gettime was returning nanosecond accuracy.
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
| |