MDIO drivers were moved into their own sub directory of networking drivers.
This has caused the AR40xx driver to probe before MDIO drivers and that wont
work as it depends on the MDIO bus to be up so it can be fetched.
Lets solve it by moving the AR40xx into MDIO folder so they get probed like
before.
Signed-off-by: Robert Marko <robert.marko@sartura.hr>
The target uses 5.4 as default kernel since 03/2020.
Kernel 4.19 support is not really maintained anymore, it does not
seem to be needed, and removing it will make upcoming driver
updates easier. Thus, remove it.
Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
There is no point in keeping the AR40xx driver as a patch as its
not pending merge or backport.
To allow for easier maintenance until DSA is ready move it into
files like EDMA is.
Signed-off-by: Robert Marko <robert.marko@sartura.hr>
[combine with removal from patches-5.4]
Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
Enable the VLAN tag offloading mechanism for RGMII single-port devices.
This allows those devices to use 802.1Q VLANs on the ethernet port.
Previously, RX frames were double tagged, as the RX TAG removal flag was
not enabled and an additional 802.1Q header was inserted elsewhere in
the code.
On the TX side, tagging was completely not present for single-port
devices. Enable tagging if an 802.1Q frame should be transmitted and
disable the default tagging mechanism for single-port devices.
Tested on Aruba AP-303
Signed-off-by: David Bauer <mail@david-bauer.net>
As the the SoC uses implicit vlan tagging for dual MAC support, the
offload feature breaks when using double tagging.
Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: John Crispin <john@phrozen.org>
It was noticed that the the whole MAC can hang when transferring data from
one ar40xx port (WAN ports) to the CPU and from the CPU back to another
ar40xx port (LAN ports). The CPU was doing only NATing in that process.
Usually, the problem first starts with a simple data corruption:
$ wget https://cdimage.debian.org/debian-cd/current/amd64/iso-cd/debian-10.4.0-amd64-netinst.iso -O /dev/null
...
Connecting to saimei.ftp.acc.umu.se (saimei.ftp.acc.umu.se)|2001:6b0:19::138|:443... connected.
...
Read error at byte 48807936/352321536 (Decryption has failed.). Retrying.
But after a short while, the whole MAC will stop to react. No traffic can
be transported anymore from the CPU port from/to the AR40xx PHY/switch and
the MAC has to be resetted.
Signed-off-by: Sven Eckelmann <sven@narfation.org>
Acked-by: John Crispin <john@phrozen.org>
In 5.4 kernel old u32 array way of setting network features was dropped and linkmode is now the only way.
So lets migrate the EDMA driver to support linkmode.
Also, old get/set settings API for ethtool is also dropped so lets convert to new ksettings API while at it as it demands linkmode.
Now, gigabit works properly as well as ethtool.
Previously you would get this in ethtool:
root@OpenWrt:/# ethtool eth1
Settings for eth1:
Supports Wake-on: d
Wake-on: d
Current message level: 0x00000000 (0)
Link detected: yes
Now, features are properly advertised:
root@OpenWrt:/# ethtool eth1
Settings for eth1:
Supported ports: [ TP MII ]
Supported link modes: 10baseT/Half 10baseT/Full
100baseT/Half 100baseT/Full
1000baseT/Half 1000baseT/Full
Supported pause frame use: Symmetric Receive-only
Supports auto-negotiation: Yes
Supported FEC modes: Not reported
Advertised link modes: 10baseT/Half 10baseT/Full
100baseT/Half 100baseT/Full
1000baseT/Half 1000baseT/Full
Advertised pause frame use: Symmetric Receive-only
Advertised auto-negotiation: Yes
Advertised FEC modes: Not reported
Link partner advertised link modes: 10baseT/Half 10baseT/Full
100baseT/Half 100baseT/Full
1000baseT/Full
Link partner advertised pause frame use: No
Link partner advertised auto-negotiation: No
Link partner advertised FEC modes: Not reported
Speed: 1000Mb/s
Duplex: Full
Port: Twisted Pair
PHYAD: 4
Transceiver: internal
Auto-negotiation: on
MDI-X: Unknown
Supports Wake-on: d
Wake-on: d
Current message level: 0x00000000 (0)
Link detected: yes
Signed-off-by: Robert Marko <robert.marko@sartura.hr>
of_get_mac_address returns valid pointer or ERR_PTR since 5.2 via commit
d01f449c008a ("of_net: add NVMEM support to of_get_mac_address") so the
patch fixes following OOPs on nbg6617:
Unable to handle kernel paging request at virtual address ffffffed
CPU: 1 PID: 1 Comm: swapper/0 Not tainted 5.4.24 #0
PC is at edma_axi_probe+0x444/0x1114
LR is at bus_find_device+0x88/0x9c
Where the PC can be resolved to:
>>> l *edma_axi_probe+0x444
0xc067be5c is in edma_axi_probe (./include/linux/string.h:378).
>>> l *edma_axi_probe+0x43f
0xc067be57 is in edma_axi_probe (drivers/net/ethernet/qualcomm/essedma/edma_axi.c:936)
Which leads to the following code fragment:
935 mac_addr = of_get_mac_address(pnp);
936 if (mac_addr)
937 memcpy(edma_netdev[idx_mac]->dev_addr, mac_addr, ETH_ALEN);
Where using mac_addr=0xffffffed (-ENODEV) as source address in memcpy()
is causing the OOPs.
Acked-by: John Crispin <john@phrozen.org>
Signed-off-by: Petr Štetiar <ynezz@true.cz>
