Problem with checksum offloading on RPi3 (PF + Jails involved)

[Hans Petter Selasky]

On 2020-11-04 07:36, Carsten Bäcker wrote:
> Am 03.11.2020 um 09:46 schrieb Hans Petter Selasky:
>> On 2020-11-03 09:19, Hans Petter Selasky wrote:
>>> It looks like smsc(4) uses the following RX format but I don't
>>> know actual RX format of H/W(no access to datasheet).
>>>
>>> <---------------------------- actlen
>>> -------------------------------------------------->
>>>                  <------------- pktlen ------------------------>
>>> rxhdr(4 bytes) | padding (2 bytes) | RX frame | FCS(4 bytes) |
>>> partial checksum(2 bytes)
>>
>> Hi,
>>
>> I wonder if the checksum is zero, when not valid, and that we should
>> check for this in the driver!
>>
>> Can you try this patch?
>>
>> Also enabling debugging in the SMSC driver would be useful.
>>
>> --HPS
>>
>> _______________________________________________
>> freebsd-hackers at freebsd.org mailing list
>> https://lists.freebsd.org/mailman/listinfo/freebsd-hackers
>> To unsubscribe, send any mail to 
>> "freebsd-hackers-unsubscribe at freebsd.org"
> Hi,
> 
> i applied the patch on -CURRENT but got a panic right after loading the
> kernel. Most likely an unrelated problem.
> 
> But i was able to apply the patch on releng/12.2 (with an offset).
> Unfortunately it doesn't change the previously described behavior with
> rxcsum and i didn't manage to get any reasonable debug-output.
> 
> Since i can easily reproduce the problem. How else can i help?
> 

Can you enable the debug knob for SMSC:

sysctl hw.usb.smsc.debug=16

It will print the RX checksum per frame received. Just do this with the 
unpatched driver while doing some traffic. Capture log messages from 
dmesg or /var/log/messages.

--HPS