Hi, Billy Araujo <billyaraujo@xxxxxxxxx> writes: > Hi Felipe and Minas, > > > Firstly, thank you very much for your responses. > > I work as a software engineer at Fluke. > We compile our own version of the kernel but I am not the Linux kernel > maintainer at Fluke, I work mostly on the application side using Qt and > C++. > > I am still learning about the Linux kernel so sorry if this is a dumb > question but can these SOF interrupts be disabled in the kernel/controller? possibly, but it may be that SOF interrupts are needed for whatever Erratum implementation. Check the sources. > We are using Linux based on Angstrom (v2016.06), built using Yocto recipes > (I think) and using a USB analyser I see that are millions of NAKs even on > a relatively short capture.The problem is that when we initialise a 3rd > party kernel module/library the system slows down significantly. I think > they tested it on Raspberry PI and it worked fine. So not sure where the > problem is. But now I know where to look. I'd ask for support from the vendor giving you the kernel and the other vendor giving you the 3rd party driver that slows down the system. Keep in mind you're already paying both suppliers for support, might as well use it. good luck -- balbi
Attachment:
signature.asc
Description: PGP signature
