OpenWrt Forum Archive

Topic: Marvell sky2 Driver Bug- length reported by DMA does not match PHY

The content of this topic has been archived on 29 Apr 2018. There are no obvious gaps in this topic, but there may still be some posts missing at the end.

Post #1
Jackly
12 Jul 2016, 16:07

Hello,

We are using in-build OpenWrt packages (16.02 Version) on Ventana GW5410 hardware. The Ethernet Port 2 of Ventana Board (Eth1) is supported by Marvell sky2 driver. Default MTU size of Eth1 is 1500 and let say IP address is 192.168.0.100. We try to Ping Port2 of board from 192.168.0.110 with Data size 2000
#ping 192.168.0.100 -s 2000 . The ping was successful

Later we tried to change the MTU of Eth1 to 1450 and try to ping from 192.168.0.100 machine with Data size 2000
#ping 192.168.0.100 -s 2000. But this time Driver code drop the packets

Error is - sky2 0000:08:00.0 eth1: rx error, status 0x5ea0100 length 1470

On Debugging found that there is a mismatch in received packet length at PHY and on DMA . Packet size is 1514 at DMA and 1470 at PHY. Suspecting it is a Driver Bug . Any patch available to resolve this problem ?

Thank you

Post #2
elektroman
13 Jul 2016, 16:24

The latest update for Marvell sky is

commit a188222b6ed29404ac2d4232d35d1fe0e77af370
Author: Tom Herbert <tom@herbertland.com>
Date:   Mon Dec 14 11:19:43 2015 -0800

    net: Rename NETIF_F_ALL_CSUM to NETIF_F_CSUM_MASK
    
    The name NETIF_F_ALL_CSUM is a misnomer. This does not correspond to the
    set of features for offloading all checksums. This is a mask of the
    checksum offload related features bits. It is incorrect to set both
    NETIF_F_HW_CSUM and NETIF_F_IP_CSUM or NETIF_F_IPV6 at the same time for
    features of a device.
    
    This patch:
      - Changes instances of NETIF_F_ALL_CSUM to NETIF_F_CSUM_MASK (where
        NETIF_F_ALL_CSUM is being used as a mask).
      - Changes bonding, sfc/efx, ipvlan, macvlan, vlan, and team drivers to
        use NEITF_F_HW_CSUM in features list instead of NETIF_F_ALL_CSUM.
    
    Signed-off-by: Tom Herbert <tom@herbertland.com>
    Signed-off-by: David S. Miller <davem@davemloft.net>

So your problem may not be solved ..

The discussion might have continued from here.