Joel Jacobs wrote: > Ok, found something interesting. I left the isolcpus=1 in grub and > re-enabled hyper-threading and latency topped out around 11.2us with > acceptable performance. I thought it should be fixed so ran EMC2 and > after a few hours the error tripped again. Here is the details from > dmesg: > [snip] > At first glance it looks like the 1058805 is the outlier here, not the > 1273653 it's complaining about. Anyway I'm trying to make sense of > these numbers. Earlier in the dmesg output is this: > > [ 62.467909] RTAI[sched]: Linux timer freq = 250 (Hz), TimeBase freq > = 1800221000 hz. > > So, f=1800221000, p=1/f, servo thread = p*n (n=numbers from dmesg) > > clocks servo thread jitter > 1259487 = 699.629us -0.371us -0.0535% > 1262655 = 701.389us +1.389us +1.984% > 1058805 = 588.153us -111.847us -15.978% > 1265220 = 702.814us +2.814us +0.402% > 1248228 = 693.375us -6.625us -0.946% > 1273653 = 707.498us +7.498us +1.071% > 1273653 / 1058805 = 1.20291555. This is slightly more than the 20% threshold. It may be true that the detection method should be changed, since it's asymmetric (downward deviations are more significant than upward, since the calculation is more or less max/min). In practice though, it's not a deviation of 20% from "normal", it's a range which is more than 20% of the lowest measurement in the buffer. > It looks to me that the 1058805 is the problem child here and it's not > an Unexpected realtime delay at all. Looks like an Unexpected > premature execution of the thread! Any idea what could cause this? > Sure, maybe the message could be changed to "Unexpectedly high jitter in realtime thread", but I don't know if that makes any more sense to anyone.
- Steve ------------------------------------------------------------------------------ Enable your software for Intel(R) Active Management Technology to meet the growing manageability and security demands of your customers. Businesses are taking advantage of Intel(R) vPro (TM) technology - will your software be a part of the solution? Download the Intel(R) Manageability Checker today! http://p.sf.net/sfu/intel-dev2devmar _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
