Okay, excuse me, my C knowlege is not great. I don't even understand how that patch worked (it didn't), but I've figured out the reason: I have two instances of netspeed, for two interfaces. So the rxbytes is getting overwritten for the two different network interfaces. Sorry for the noise.
On 13/10/2018 20:22, John Soros wrote: > Hi. > On my system (Linux z.y 4.18.12-200.fc28.x86_64 #1 SMP Thu Oct 4 > 15:46:35 UTC 2018 x86_64 x86_64 x86_64 GNU/Linux) the netspeeds > component of slstatus doesn't work at all. I get some Pib results for my > wireless network card and just 0 for my ethernet card. > I don't exactly understand how and where slstatus handles calling this > component twice, but from my debugging it is obvious it is called twice > for every slstatus update. > The issue seems to be that often, the read from the linux sys file > yields 0, and this makes the result of this component incoherent. > I really am not sure why this works, but with this patch I get reliable > and seemingly correct results. If someone can explain me how this works, > I would be very thankful. > Regards > John >