LS1021A Ethernet Receive Carrier Sense Error Counter counts up!

[MatthiasFuchs]

Hi,

I have the LS1021A running on a custom board that is very similiar to the TWR board.

In some situations - I cannot say in which one exactly - I loose some ping packets and I see

ethtool counting rx-carrier-sense-errors.

I figured out that this counter counter comes directly from the eTSEC RMON MIB register eTSECx_RCSE.

What does this register really count? Where should I start diggin'

Thanks,

Matthias

root@sdk2-0-105:~# ethtool -S eth2 | grep err      rx-allocation-errors: 0      rx-large-frame-errors: 0      rx-short-frame-errors: 0      rx-non-octet-errors: 0      rx-crc-errors: 0      rx-overrun-errors: 0      rx-busy-errors: 0      rx-babbling-errors: 0      ethernet-bus-error: 0      tx-babbling-errors: 0      tx-underrun-errors: 0      tx-timeout-errors: 0      rx-fcs-errors: 0      rx-alignment-error: 0      rx-frame-length-error: 0      rx-code-error: 0      rx-carrier-sense-error: 24      tx-deferral-packets: 0      tx-excessive-deferral-packets: 0  ..... 

[bpe]

In general, this counter increments when the controller sees unexpected
behaviour of the MAC/PHY interface signals. The exact conditions
depend on the interface type. For MII it can be random floating of
CRS. For RGMII it counts when a certain combination of Rx_DV/ER
and RxD[0:3] is observed, see RGMII specification, Table 4, False
Carrier row.  The suggestion is to monitor the MAC/PHY interface
and diagnose the PHY.

Have a great day,
Platon

-----------------------------------------------------------------------------------------------------------------------
Note: If this post answers your question, please click the Correct Answer button. Thank you!
-----------------------------------------------------------------------------------------------------------------------