Re: [beagleboard] Beaglebone RT test

2016-03-31 Thread Micka
u want real time behaviour use the pru ! Le ven. 1 avr. 2016 à 03:15, Victor WANG a écrit : > Hi John, > > Thanks so much for your response, as you said, it is true that there is no > load for the CPU. It is well of getting this result, thanks again. > > Regards, > >

Re: [beagleboard] Beaglebone RT test

2016-03-31 Thread Victor WANG
Hi John, Thanks so much for your response, as you said, it is true that there is no load for the CPU. It is well of getting this result, thanks again. Regards, Victor 在 2016年3月31日星期四 UTC+8下午2:49:24,john3909写道: > > I would say that avg latency of 32uS is very good and I would be skeptical >

[beagleboard] Beaglebone RT test

2016-03-31 Thread wangvictor2...@163.com
Hello, I use the beaglebone black and I've installed Linux debian on it, whitch is a Real Time system (*Linux beaglebone 4.4.6-bone-rt-r6 #1 PREEMPT RT Thu Mar 17 04:47:32 UTC 2016 armv7l GNU/Linux*), and I want to test the characteristic of RT with cyclictest, after using " *cyclictest -p

Re: [beagleboard] Beaglebone RT test

2016-03-31 Thread John Syne
I would say that avg latency of 32uS is very good and I would be skeptical of this result. I don’t know what cyclictest does, but I’m wondering what the CPU load is during this test. My guess is the CPU load is low and that is why you are getting 32uS. If the CPU load was high, I would expect

[beagleboard] Beaglebone RT test

2016-03-31 Thread Victor WANG
hello, I use beaglebone black and install Linux Debian (*Linux beaglebone 4.4.6-bone-rt-r6 #1 PREEMPT RT Thu Mar 17 04:47:32 UTC 2016 armv7l GNU/Linux*), I want to test the characteristic of Real Time, so I use cyclictest: After "*cyclictest -p 90 - m -c 0 -i 200 -n -h 100 -q -l 1*", I