Banana Pi BPI-R4 Wifi 7 router board with MediaTek MT7988A (Filogic 880),4G RAM and 8G eMMC

[quartis]

hey. version with 1xstp and 1x2.5gbe with poe is avaible in shop? or must solder yourself?

[sinovoip (bpi team)]

you can contact [email protected] ask support

[fengaiyu]

How to setup or inplement the AFC(Automated Frequency Coordination) in 6GHz? Hardware:Banana pi R4+WiFi7:BPI-R4-NIC-BE14 We have already completed the AFC DUT test Harness on the computer(6 GHz AFC Resources | Wi-Fi Alliance). Is there any information about AFC?

[Avamander]

AFC requires, well, automated frequency coordination. This entails something like running afcd (having it connect to the coordinating organization) and then hostapd (or equivalent) with its support.

I’m not aware of any publicly usable implementations of the former just yet. You have to use the low limits that are allowed without AFC.

[fengaiyu]

Thanks,your answer is useful to me. Maybe the “afcd” is what i want. Like these features:[PATCH v7 1/4] hostapd: afcd: add AFC daemon support (Linux Host AP)

[kvic]

FYI. A different vendor’s board has similar, if not exactly same, traces between MT7988a and memory chips.

AsiaRF Wifi 7 module

[AA] One member asked about the ‘straight traces’ between SoC and memory chips. This board has the same pattern as BPI R4. Hence, I conclude it’s likely from reference design/recommended design by MediaTek. So indeed nothing to worry about.

[c3sk8]

Atheros AR5BXB63 recycled here…ok, It Is really old, but works fine… Any info about the official wifi7 nic realase?

[Lorem_amicus (Michael Fischer)]

OpenWrt One/AP-24.XY is an upcoming router board developed by OpenWrt and Banana Pi

Did you talking about the wifi 7 modul for BPI-R4?: https://de.aliexpress.com/item/1005007223137654.html But please, there are other chats for the wifi 7 NIC BE14 and BPI-R4: or:

[luke-meshii (Luke Slark)]

Just putting this out there as cannot find much information on it, other than SINOVOIP’s getting started wiki.

Has anyone managed to get a Quectel RM520N-GL module working in CN16 M.2 B slot?

Device does not appear to initialize upon boot, i’m not wanting to use an outdated verison of OpenWRT (sinivoip BPI-R4 images) due to not being able to install new drivers/packages etc…

[frank-w]

please create own thread for special 3rd party hardware…this thread is about 355 posts log…so the answer gets lost for all who have same problem

[nezar_taima]

I have two 5G modules from quectel (RM520N-GL & RM520N-EU), both works okay inside R4 board. BTW, using mainline openwrt not official image.

[Karasik]

Do SFPs (I have DFP-34X-2C3) work in OpenWRT snapshot? I can’t log into the web interface, build 21.02 works without problems.

ethtool is the same in both versions:

Settings for eth2:
        Supported ports: [ FIBRE ]
        Supported link modes:   1000baseX/Full
        Supported pause frame use: Symmetric Receive-only
        Supports auto-negotiation: Yes
        Supported FEC modes: Not reported
        Advertised link modes:  1000baseX/Full
        Advertised pause frame use: Symmetric Receive-only
        Advertised auto-negotiation: Yes
        Advertised FEC modes: Not reported
        Speed: Unknown!
        Duplex: Unknown! (255)
        Port: FIBRE
        PHYAD: 0
        Transceiver: internal
        Auto-negotiation: on
        Current message level: 0x000000ff (255)
                               drv probe link timer ifdown ifup rx_err tx_err                                                   
        Link detected: no

[frank-w]

please do specific extra hardware in separate threads…afaik there is already one for the DFP-34X-2C3 where someone created patches for (afair rejected upstream so far)

this thread goes very messy because of different topics mixed

[danpawlik (Dan)]

Does someone have an issue in recent builds (since 2 weeks), that device freeze with kernel panic:

Thu Aug 22 17:57:24 2024 daemon.notice hostapd: phy1-ap0: BEACON-REQ-TX-STATUS 8c:7a:3d:aa:d7:aa 30 ack=1
Thu Aug 22 17:57:25 2024 kern.alert kernel: [ 5520.777538] Unable to handle kernel paging request at virtual address ffffffffffffff98
Thu Aug 22 17:57:25 2024 kern.alert kernel: [ 5520.785480] Mem abort info:
Thu Aug 22 17:57:25 2024 kern.alert kernel: [ 5520.788264]   ESR = 0x0000000096000005
Thu Aug 22 17:57:25 2024 kern.alert kernel: [ 5520.792016]   EC = 0x25: DABT (current EL), IL = 32 bits
Thu Aug 22 17:57:25 2024 kern.alert kernel: [ 5520.797318]   SET = 0, FnV = 0
Thu Aug 22 17:57:25 2024 kern.alert kernel: [ 5520.800361]   EA = 0, S1PTW = 0
Thu Aug 22 17:57:25 2024 kern.alert kernel: [ 5520.803500]   FSC = 0x05: level 1 translation fault
Thu Aug 22 17:57:25 2024 kern.alert kernel: [ 5520.808367] Data abort info:
Thu Aug 22 17:57:25 2024 kern.alert kernel: [ 5520.811237]   ISV = 0, ISS = 0x00000005, ISS2 = 0x00000000
Thu Aug 22 17:57:25 2024 kern.alert kernel: [ 5520.816714]   CM = 0, WnR = 0, TnD = 0, TagAccess = 0
Thu Aug 22 17:57:25 2024 kern.alert kernel: [ 5520.821756]   GCS = 0, Overlay = 0, DirtyBit = 0, Xs = 0
Thu Aug 22 17:57:25 2024 kern.alert kernel: [ 5520.827056] swapper pgtable: 4k pages, 39-bit VAs, pgdp=0000000046b4c000
Thu Aug 22 17:57:25 2024 kern.alert kernel: [ 5520.833747] [ffffffffffffff98] pgd=0000000000000000, p4d=0000000000000000, pud=0000000000000000

and after reboot, it goes to the initramfs and only unplug power for few seconds and turn on works? I suspect that the microSD card is broken, but on second one (very old) issue comes again.

[frank-w]

Please create new topic with operation system and version. For me it looks like an error-pointer is used as memory address.

0xf…f98 should be -104 which is “connection reset”

[danpawlik (Dan)]

thanks @frank-w . I will do, but first I need to collect some logs Just leave a msg for others, if someone have similar issue- my guess is related to mt7925e and recent update for mt76 in OpenWRT (probably it starts since this: mt76: update to Git HEAD (2024-08-16) · openwrt/openwrt@20bf5f3 · GitHub )

EDIT: Issue created: mt7925: Kernel panic after authentication · Issue #909 · openwrt/mt76 · GitHub

[ljy26 (arkin jade)]

hope experts here can help me with this issue, currently using FM350GL / lastest snapshot , getting good speedtest around 600mbps but when upload starts , R4 will reboot due to kernel panic.

[   22.894881] mtk_t7xx 0003:01:00.0: Packet drop on channel 0x100a, port not found
[   27.698024] NOHZ tick-stop error: local softirq work is pending, handler #08!!!
[   28.856590] NOHZ tick-stop error: local softirq work is pending, handler #08!!!
[   28.894956] NOHZ tick-stop error: local softirq work is pending, handler #08!!!
[   28.894962] NOHZ tick-stop error: local softirq work is pending, handler #08!!!
[   29.267697] NOHZ tick-stop error: local softirq work is pending, handler #08!!!
[   29.407202] NOHZ tick-stop error: local softirq work is pending, handler #08!!!
[   29.407208] NOHZ tick-stop error: local softirq work is pending, handler #08!!!
[   29.493214] NOHZ tick-stop error: local softirq work is pending, handler #08!!!
[   29.603202] NOHZ tick-stop error: local softirq work is pending, handler #08!!!
[   29.688209] NOHZ tick-stop error: local softirq work is pending, handler #08!!!
[   88.920730] wwan wwan0 wwan0: TX queue 0 is full
[   88.925360] wwan wwan0 wwan0: TX queue 0 is full
[   88.929969] wwan wwan0 wwan0: TX queue 0 is full
[   88.934591] wwan wwan0 wwan0: TX queue 0 is full
[   88.939199] wwan wwan0 wwan0: TX queue 0 is full
[   88.943824] wwan wwan0 wwan0: TX queue 0 is full
[   88.948432] wwan wwan0 wwan0: TX queue 0 is full
[   88.953046] wwan wwan0 wwan0: TX queue 0 is full
[   88.957654] wwan wwan0 wwan0: TX queue 0 is full
[   89.990457] wwan wwan0 wwan0: NETDEV WATCHDOG: CPU: 2: transmit queue 0 timed out 1070 ms
[   91.030488] wwan wwan0 wwan0: NETDEV WATCHDOG: CPU: 2: transmit queue 0 timed out 2110 ms
[   91.990518] wwan wwan0 wwan0: NETDEV WATCHDOG: CPU: 2: transmit queue 0 timed out 3070 ms
[   93.030549] wwan wwan0 wwan0: NETDEV WATCHDOG: CPU: 2: transmit queue 0 timed out 4110 ms
[   93.990577] wwan wwan0 wwan0: NETDEV WATCHDOG: CPU: 2: transmit queue 0 timed out 5070 ms
[   95.030610] wwan wwan0 wwan0: NETDEV WATCHDOG: CPU: 2: transmit queue 0 timed out 6110 ms
[   95.990636] wwan wwan0 wwan0: NETDEV WATCHDOG: CPU: 2: transmit queue 0 timed out 7070 ms
[   97.030666] wwan wwan0 wwan0: NETDEV WATCHDOG: CPU: 2: transmit queue 0 timed out 8110 ms
[   97.990694] wwan wwan0 wwan0: NETDEV WATCHDOG: CPU: 2: transmit queue 0 timed out 9070 ms
[   99.030721] wwan wwan0 wwan0: NETDEV WATCHDOG: CPU: 2: transmit queue 0 timed out 10110 ms
[   99.990750] wwan wwan0 wwan0: NETDEV WATCHDOG: CPU: 2: transmit queue 0 timed out 11070 ms
[  101.030779] wwan wwan0 wwan0: NETDEV WATCHDOG: CPU: 2: transmit queue 0 timed out 12110 ms
[  101.990807] wwan wwan0 wwan0: NETDEV WATCHDOG: CPU: 2: transmit queue 0 timed out 13070 ms
[  102.303752] mtk_t7xx 0003:01:00.0: Could not unmask the UL interrupt. DPMAIF_AO_UL_AP_L2TIMR0 is 0xfffe1
[  102.313258] wwan wwan0 wwan0: TX queue 0 is full
[  102.492744] skbuff: skb_under_panic: text:ffffffc08077c45c len:-65269 put:149 head:ffffff80c805b000 data:ffffff7fc806afff tail:0xd3 end:0xec0 dev:wwan0

[M10 (M)]

So, wifi be14 module arrived, mounted it , and after i’ve connected power board is not booting, nothing on serial, straight after power red pwr led and next g led green lights up and thats all… Someone have an idea how to diagnose this dead men?

[Mathieulh]

Did you set SW4 to the “on” position? (to supply 12V to the BE14)

[M10 (M)]

Yes, i did, but now without module board is also dead…