Re: DPAA Ethernet traffice troubles with Linux kernel

Hi,

I just found out that something goes wrong within the ARP table (well thats
what i think). I hope someone has a clue..

When i bootup the AmigaOne X5000 with the Ethernet cable connected, it just
never senses the presence of the UTP cable.
I must run the following command as root: mii-tool -R eth0 ... it Resets
the transceiver and the ethernet connection is ready to go.
But then... it randomly dies....

After some digging i found that in a working situation the ARP table is
filled with the correct info. Router address, and corresponding MAC, C mask
and Interface.

(Working)
skateman@X5000LNX:~$ arp -n
Address                  HWtype  HWaddress           Flags Mask
Iface
192.168.22.66            ether   08:5b:0e:fd:db:6a   C
eth0

No more traffic..it just suddenly dies...
skateman@X5000LNX:~$ ping www.google.com
ping: unknown host www.google.com

Rechecked the ARP... and found out it has lost the necessary info.
skateman@X5000LNX:~$ arp -n
Address                  HWtype  HWaddress           Flags Mask
Iface
192.168.22.66                    (incomplete)
eth0

Anyone??? :-)


On Tue, Feb 6, 2018 at 12:20 PM, Christian Zigotzky [off-list ref]
wrote:

Hello,

I have tried to figure out why there is a problem with the buffer space
but unfortunately without any success. Any ideas? Could you please watch
Skateman's video? [1]

Thanks,
Christian

[1] https://drive.google.com/file/d/18RhksfcavRJPr86asQDTzrmsN20D0Xim/view


On 03 February 2018 at 12:54PM, mad skateman wrote:

quoted

For those interested... i have recorded a video of my X5000 DPAA

Ethernet, and the weird problems..

quoted

In this Video i am also transfering hundereds of megabytes from my NAS

to the X5000.

quoted

You will also see pings die... mostly after the 12th packet and giving

the no buffer space error..

quoted

Hopefully someone might have a clue about what is happening.

https://drive.google.com/file/d/18RhksfcavRJPr86asQDTzrmsN20D0Xim/view

On Wed, Jan 17, 2018 at 3:43 PM, Madalin-cristian Bucur <

madalin.bucur@nxp.com> wrote:

quoted

    > -----Original Message-----
    > From: Madalin-cristian Bucur
    > Sent: Wednesday, January 17, 2018 4:25 PM
    > To: David S . Miller [off-list ref]
    > Cc: linuxppc-dev@lists.ozlabs.org; netdev@vger.kernel.org;
    > madskateman@gmail.com; 'Madalin-cristian Bucur' <

madalin.bucur@nxp.com>;

quoted

    > Andrew Lunn [off-list ref]; Joakim Tjernlund
    > [off-list ref]
    > Subject: RE: DPAA Ethernet traffice troubles with Linux kernel
    >
    > > -----Original Message-----
    > > From: netdev-owner@vger.kernel.org [mailto:

netdev-owner@vger.kernel.org]

quoted

    > > On Behalf Of Madalin-cristian Bucur
    > > Sent: Wednesday, January 17, 2018 4:16 PM
    > > To: Andrew Lunn [off-list ref]; Joakim Tjernlund
    > > [off-list ref]
    > > Cc: linuxppc-dev@lists.ozlabs.org; netdev@vger.kernel.org;
    > > madskateman@gmail.com; David S . Miller [off-list ref]
    > > Subject: RE: DPAA Ethernet traffice troubles with Linux kernel
    > >
    > > > -----Original Message-----
    > > > From: Andrew Lunn [mailto:andrew@lunn.ch]
    > > > Sent: Wednesday, January 17, 2018 3:44 PM
    > > > To: Joakim Tjernlund [off-list ref]
    > > > Subject: Re: DPAA Ethernet traffice troubles with Linux kernel
    > > >
    > > > > That doesn't work really, having users to hit the bug, debug

it, fix

quoted

    > > it
    > > > and then
    > > > > find it fixed already in upstream, then specifically request

it to

quoted

    > be
    > > > backported to stable.
    > > > > I don't need this fix to be backported, already got it.

Someone else

quoted

    > > > might though.
    > > >
    > > > The "someone else might though" is a big point of asking for

it to

quoted

    > > > added to stable. The other reason is it means one less patch

you need

quoted

    > > > to maintain in your build.
    > >
    > > I've sent that patch [1] for net but I guess the timing was

wrong and

quoted

    > > it was merged to net-next.
    > >
    > > > > I would be interested in bug fixes upstream which fixes:
    > > >
    > > > Did you try upstream? Does it give the same errors?
    > > >
    > > >     Andrew
    > >
    > > [1] https://patchwork.kernel.org/patch/10146119/
    > >
    > > Madalin
    >
    > Hi Dave,
    >
    > Can you please add the fix [1] to stable?
    >
    > Thank you,
    > Madalin

    Sorry,

    I've provided the wrong link towards the patch (v1 instead of v3),
    here's the correct one:

    https://patchwork.kernel.org/patch/10151969/

    Madalin