[etherlab-users] skipped/unmatched datagrams - at startup - safe to ignore?

[Matthieu Bec]

I'm now able to get reliable clean starts (no more startup warnings) by 
ramping up my cycle time in increments; like 1, 2, 4, 8, 10 kHz over a 
short period.

Is anyone using similar "warm-up" technique?

The work around seems to work fine, I was thinking I might be able to 
get rid of it, maybe something non-optimal in my kernel config?

Regards,
Matthieu



On 6/11/12 2:45 PM, Matthieu Bec wrote:
>
> Hello all,
>
> I'm experimenting with different NIC to better understand the pros, cons
> and tradeoff of different solution. My system has on-board e1000e, and a
> PCIe r8169, my test application runs in kernel cycled by hr-timer. I'm
> cranking up the frame rate looking how far it will reliably go and
> checking dmesg logs.
>
> In all cases, there seems to be an intermediate frame rate where I get
> skipped/unmatched datagrams right after inserting my app. module, but
> quickly; once the master has found all the slaves (?) no more warnings
> show in the logs
>
> I was wondering if it was safe to ignore those skipped/unmatched
> warnings? Should I plan some checks in my application to verify the pdo
> mapping, sdo config, etc have all been properly applied?
>
> Regards,
>