These jumps wouldn't occur if ptp4l was preempted from the CPU? -----Original Message----- From: Keller, Jacob E <jacob.e.kel...@intel.com> Sent: Thursday, February 11, 2021 9:53 AM To: Richard Cochran <richardcoch...@gmail.com>; Zuckerbrod, Daniel <dzuckerb...@textronsystems.com> Cc: linuxptp-users@lists.sourceforge.net Subject: RE: [Linuxptp-users] PTP4L Offset Jumps When Querying PMC
> -----Original Message----- > From: Richard Cochran <richardcoch...@gmail.com> > Sent: Wednesday, February 10, 2021 7:01 PM > To: Zuckerbrod, Daniel <dzuckerb...@textronsystems.com> > Cc: linuxptp-users@lists.sourceforge.net > Subject: Re: [Linuxptp-users] PTP4L Offset Jumps When Querying PMC > > On Thu, Feb 11, 2021 at 02:41:28AM +0000, Zuckerbrod, Daniel wrote: > > I have seen other Linux commands other than PMC causes jumps as well. > > Thinking a bit more, the PMC queries never cause ptp4l to actually > access the PHC device. So the issue is likely in the driver. Looks > like that link you shared has some custom, out-of-tree driver hacks. > I would start there. > > HTH, > Richard > > I agree with Richard's assessment. If you're seeing significant jumps in the clock like this, it smells of an issue in the driver. It is plausible that heat could change the frequency of the clock source, but this shouldn't cause an immediate clock jump like you're seeing in these logs. Thanks, Jake _______________________________________________ Linuxptp-users mailing list Linuxptp-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/linuxptp-users
