On 2023-12-13 15:08, Paul Koning wrote:
On Dec 13, 2023, at 6:17 AM, Josh Moyer <JMoyer%nodomain.net@localhost> wrote:...See attached PNG. The values in the CPU and WCPU columns don't sum up to the percentages in the CPU states line at the top of the display. And the sum of the CPU status != 100. This was happening in 9, too, but it might be worse now.Floating point percentages not summing to exactly 100 is perfectly normal. That's what you would expect given the approximate nature of floating point arithmetic, especially if you're looking at displayed values that are rounded to a smaller number of significant digits.
Well, to be honest, 106% on one process is more than an FP rounding error. :D
But one have to understand that the CPU % per process are not completely accurate to start with. I seem to remember they are somehow computed/extrapolated, unlike the numbers at the top ot top, which are actual measurements.
But I was hoping someone else would comment on that, since I think that is not related to any time drift, or anything VAX specific, but it just a result of how CPU is computed for a process.
But maybe I'm somewhat wrong on that one...
Do you have measurements that show clock drift under load?
I was assuming he had. But in a sense, it also wouldn't surprise me. When doing things, the amount of time in system mode these days for NetBSD/VAX really bothers me. Sometimes I can observe it being at over 50% system time for longer periods. And in Josh case, it's reporting over 50% in interrupt context, which *really* worries me, and suggests that would could easily be missing interrupts...
Johnny
--
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip
email: bqt%softjar.se@localhost || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol