Unfortunately that is the end of my testing. My APU2D4 was damaged by lightning today and no longer boots.
Posts by rahlquist
-
-
So honestly, with the success I have had with the "Intel Corporation Ethernet Controller I225-V (rev 03)" card in my APU2 and also the RTL8125B card, and the fact that with IRQ balancing these can easily continue the life of these, I wonder if maybe the I225 should be the next network chip they consider. I dont know what design challenges that would present other than likely needing more power.
-
Have you considered maybe a CMOS reset as described in the manual? Or this https://pcengines.ch/spi1a.htm
Possibly that might get you back up and running?
-
And with this adapter; https://smile.amazon.com/dp/B07DZ8SB8X
I got the Intel I225 M.2 B + M card working. https://smile.amazon.com/IO-CR…SY-PEX24075/dp/B09SRW3RCV
This adapter worked with the intel nic also https://smile.amazon.com/dp/B07ZKGWNDQ
Code- 59: 0 3 0 0 PCI-MSI 2621440-edge eth3
- 60: 2628 7370 45862 139694 PCI-MSI 2621441-edge eth3-TxRx-0
- 61: 79258 3333 13089 70307 PCI-MSI 2621442-edge eth3-TxRx-1
- 62: 72587 12703 5020 4520 PCI-MSI 2621443-edge eth3-TxRx-2
- 63: 246592 83325 6261 4015 PCI-MSI 2621444-edge eth3-TxRx-3
- [ ID] Interval Transfer Bitrate Retr
- [ 5] 0.00-10.00 sec 1.58 GBytes 1.36 Gbits/sec 31 sender
- [ 5] 0.00-10.00 sec 1.58 GBytes 1.36 Gbits/sec receiver
You can see the both of the mini PCIe adapters balance their load across the cores which the xhci_hcd driver in openwrt does not try to do for the USB based adapters.
Performance is crisp and stable. So 2.5Gbps/sec connection, with Wifi 6 on APU2, yes, it can work.
-
Success!
So with this adapter; https://www.amazon.com/dp/B094YY6KX6
I can get a 2.5Gbps uplink to my MB8611 cable modem. Which is connecting at just over 1.2Gbps to my broadband connection. The APU can handle the load with a PCIe Mini card because unlike with the usb based solution the IRQ load is spread across all 4 cores of the CPU. This is all with the standard timings. My cores are running around 600mhz at idle.
Next step is bonding two of the I210's to my Mikrotek switch as a single interface.
Code- [Sat Jul 9 04:16:09 2022] r8169 0000:05:00.0 eth3: RTL8125B, {REDACTED}, XID 641, IRQ 59
- [Sat Jul 9 04:16:09 2022] r8169 0000:05:00.0 eth3: jumbo features [frames: 9194 bytes, tx checksumming: ko]
- [Sat Jul 9 04:16:22 2022] RTL8226B_RTL8221B 2.5Gbps PHY r8169-0-500:00: attached PHY driver [RTL8226B_RTL8221B 2.5Gbps PHY] (mii_bus:phy_addr=r8169-0-500:00, irq=IGNORE)
- [Sat Jul 9 04:16:22 2022] r8169 0000:05:00.0 eth3: Link is Down
- [Sat Jul 9 04:16:26 2022] r8169 0000:05:00.0 eth3: Link is Up - 2.5Gbps/Full - flow control rx/tx
- root@ghost:~# iperf3 -c atl.speedtest.clouvider.net -p 5201 -R -P 4 -t 30
- - - - - - - - - - - - - - - - - - - - - - - - - -
- [ ID] Interval Transfer Bitrate Retr
- [ 5] 0.00-30.00 sec 1.38 GBytes 397 Mbits/sec 490 sender
- [ 5] 0.00-30.00 sec 1.38 GBytes 396 Mbits/sec receiver
- [ 7] 0.00-30.00 sec 1.06 GBytes 303 Mbits/sec 569 sender
- [ 7] 0.00-30.00 sec 1.05 GBytes 302 Mbits/sec receiver
- [ 9] 0.00-30.00 sec 872 MBytes 244 Mbits/sec 662 sender
- [ 9] 0.00-30.00 sec 868 MBytes 243 Mbits/sec receiver
- [ 11] 0.00-30.00 sec 1.08 GBytes 308 Mbits/sec 342 sender
- [ 11] 0.00-30.00 sec 1.07 GBytes 307 Mbits/sec receiver
- [SUM] 0.00-30.00 sec 4.37 GBytes 1.25 Gbits/sec 2063 sender
- [SUM] 0.00-30.00 sec 4.36 GBytes 1.25 Gbits/sec receiver
- iperf Done.
- root@ghost:~# ethtool eth3
- Settings for eth3:
- Supported ports: [ TP MII ]
- Supported link modes: 10baseT/Half 10baseT/Full
- 100baseT/Half 100baseT/Full
- 1000baseT/Full
- 2500baseT/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/Full
- 2500baseT/Full
- Advertised pause frame use: Symmetric Receive-only
- Advertised auto-negotiation: Yes
- Advertised FEC modes: Not reported
- Link partner advertised link modes: 100baseT/Half 100baseT/Full
- 1000baseT/Half 1000baseT/Full
- 2500baseT/Full
- Link partner advertised pause frame use: Symmetric Receive-only
- Link partner advertised auto-negotiation: Yes
- Link partner advertised FEC modes: Not reported
- Speed: 2500Mb/s
- Duplex: Full
- Port: Twisted Pair
- PHYAD: 0
- Transceiver: external
- Auto-negotiation: on
- MDI-X: Unknown
- Supports Wake-on: pumbg
- Wake-on: d
- Link detected: yes
- root@ghost:~# ip a
- 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
- link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
- inet 127.0.0.1/8 scope host lo
- valid_lft forever preferred_lft forever
- inet6 ::1/128 scope host
- valid_lft forever preferred_lft forever
- 2: eth0: <BROADCAST,MULTICAST> mtu 1500 qdisc mq state DOWN group default qlen 1000
- link/ether 00:0d:b9:4e:85:bc brd ff:ff:ff:ff:ff:ff
- 3: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq master br-lan state UP group default qlen 1000
- link/ether 00:0d:b9:4e:85:bd brd ff:ff:ff:ff:ff:ff
- inet 10.10.10.1/24 brd 10.10.10.255 scope global eth1
- valid_lft forever preferred_lft forever
- 4: eth2: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc mq master br-lan state DOWN group default qlen 1000
- link/ether 00:0d:b9:4e:85:be brd ff:ff:ff:ff:ff:ff
- inet 10.5.5.1/24 brd 10.5.5.255 scope global eth2
- valid_lft forever preferred_lft forever
- 5: eth3: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc fq_codel state UP group default qlen 1000
- link/ether {REDACTED} brd ff:ff:ff:ff:ff:ff
- inet 71.199.152.53/24 brd 71.199.152.255 scope global eth3
- valid_lft forever preferred_lft forever
- inet6 2001:558:6011:8:c8c7:e46e:1261:4a60/128 scope global dynamic noprefixroute
- valid_lft 3223sec preferred_lft 3223sec
- inet6 fe80::8ac9:b3ff:feb0:b392/64 scope link
- valid_lft forever preferred_lft forever
- 8: br-lan: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 1000
- link/ether {REDACTED} brd ff:ff:ff:ff:ff:ff
- inet 192.168.1.1/24 brd 192.168.1.255 scope global br-lan
- valid_lft forever preferred_lft forever
- inet6 2601:cc:8100:1f20::1/64 scope global dynamic noprefixroute
- valid_lft 3223sec preferred_lft 3223sec
- inet6 fdf0:a0e2:6db5::1/60 scope global noprefixroute
- valid_lft forever preferred_lft forever
- inet6 fe80::20d:b9ff:fe4e:85bd/64 scope link
- valid_lft forever preferred_lft forever
- 9: eth2.5@eth2: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc noqueue state LOWERLAYERDOWN group default qlen 1000
- link/ether 00:0d:b9:4e:85:be brd ff:ff:ff:ff:ff:ff
- inet 10.0.5.1/24 brd 10.0.5.255 scope global eth2.5
- valid_lft forever preferred_lft forever
- 10: eth1.7@eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 1000
- link/ether 00:0d:b9:4e:85:bd brd ff:ff:ff:ff:ff:ff
- inet 10.0.10.1/24 brd 10.0.10.255 scope global eth1.7
- valid_lft forever preferred_lft forever
- inet6 fe80::20d:b9ff:fe4e:85bd/64 scope link
- valid_lft forever preferred_lft forever
- 11: wlan0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue master br-lan state UP group default qlen 1000
- link/ether 00:0a:52:06:fe:ed brd ff:ff:ff:ff:ff:ff
- inet 192.168.1.1/24 brd 192.168.1.255 scope global wlan0
- valid_lft forever preferred_lft forever
- inet6 fdf0:a0e2:6db5:10::1/60 scope global noprefixroute
- valid_lft forever preferred_lft forever
- inet6 fe80::20a:52ff:fe06:feed/64 scope link
- valid_lft forever preferred_lft forever
- 12: wlan1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue master br-lan state UP group default qlen 1000
- link/ether 00:0a:52:06:df:8f brd ff:ff:ff:ff:ff:ff
- inet6 fe80::20a:52ff:fe06:df8f/64 scope link
- valid_lft forever preferred_lft forever
- root@ghost:/# cat /proc/interrupts
- CPU0 CPU1 CPU2 CPU3
- 0: 27 0 0 0 IO-APIC 2-edge timer
- 4: 555 0 0 0 IO-APIC 4-edge ttyS0
- 8: 0 0 1 0 IO-APIC 8-edge rtc0
- 9: 0 0 0 0 IO-APIC 9-fasteoi acpi
- 16: 0 0 0 2051 IO-APIC 16-fasteoi mmc0
- 18: 27 0 0 0 IO-APIC 18-fasteoi ehci_hcd:usb1
- 19: 635 0 0 2059 IO-APIC 19-fasteoi ahci[0000:00:11.0]
- 45: 0 0 0 1 PCI-MSI 1572864-edge eth1
- 46: 588 363 269 736 PCI-MSI 1572865-edge eth1-TxRx-0
- 47: 1025 465 67 590 PCI-MSI 1572866-edge eth1-TxRx-1
- 48: 460 318 344 568 PCI-MSI 1572867-edge eth1-TxRx-2
- 49: 435 480 6933 1124 PCI-MSI 1572868-edge eth1-TxRx-3
- 51: 0 0 0 0 PCI-MSI 2097152-edge eth2
- 52: 36 58 0 46 PCI-MSI 2097153-edge eth2-TxRx-0
- 53: 20 52 13 55 PCI-MSI 2097154-edge eth2-TxRx-1
- 54: 39 49 3 49 PCI-MSI 2097155-edge eth2-TxRx-2
- 55: 44 49 2 45 PCI-MSI 2097156-edge eth2-TxRx-3
- 57: 0 0 0 0 PCI-MSI 131073-edge ccp-1
- 59: 4265 16196 994 590 PCI-MSI 2621440-edge eth3
- 60: 0 0 33626 0 PCI-MSI 524288-edge mt7915e
- LOC: 37156 34434 39894 38399 Local timer interrupts
- RES: 23318 21339 13739 21105 Rescheduling interrupts
- CAL: 1561 1552 2123 1852 Function call interrupts
- TLB: 1455 1731 1878 1091 TLB shootdowns
- root@ghost:/# lspci -tv
- -[0000:00]-+-00.0 Advanced Micro Devices, Inc. [AMD] Family 16h (Models 30h-3fh) Processor Root Complex
- +-00.2 Advanced Micro Devices, Inc. [AMD] Mullins IOMMU
- +-02.0 Advanced Micro Devices, Inc. [AMD] Family 16h (Models 30h-3fh) Host Bridge
- +-02.1-[01]----00.0 MEDIATEK Corp. MT7915E 802.11ax PCI Express Wireless Network Adapter
- +-02.2-[02]----00.0 Intel Corporation I210 Gigabit Network Connection
- +-02.3-[03]----00.0 Intel Corporation I210 Gigabit Network Connection
- +-02.4-[04]----00.0 Intel Corporation I210 Gigabit Network Connection
- +-02.5-[05]----00.0 Realtek Semiconductor Co., Ltd. RTL8125 2.5GbE Controller
-
So I was able to get both USB Ethernet adapters working. I am currently running without using the onboard Ethernet ports.
Code- root@ghost:~# ip a
- 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
- link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
- inet 127.0.0.1/8 scope host lo
- valid_lft forever preferred_lft forever
- inet6 ::1/128 scope host
- valid_lft forever preferred_lft forever
- 2: eth0: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN group default qlen 1000
- link/ether brd ff:ff:ff:ff:ff:ff
- 3: eth1: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc mq master br-lan state DOWN group default qlen 1000
- link/ether brd ff:ff:ff:ff:ff:ff
- inet 10.10.10.1/24 brd 10.10.10.255 scope global eth1
- valid_lft forever preferred_lft forever
- 4: eth2: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc mq state DOWN group default qlen 1000
- link/ether brd ff:ff:ff:ff:ff:ff
- inet 10.5.5.1/24 brd 10.5.5.255 scope global eth2
- valid_lft forever preferred_lft forever
- 5: eth3: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc fq_codel state UP group default qlen 1000
- link/ether brd ff:ff:ff:ff:ff:ff
- inet scope global eth3
- valid_lft forever preferred_lft forever
- inet6 2scope global dynamic noprefixroute
- valid_lft 334949sec preferred_lft 334949sec
- inet6 scope link
- valid_lft forever preferred_lft forever
- 6: eth4: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc fq_codel master br-lan state UP group default qlen 1000
- link/ether brd ff:ff:ff:ff:ff:ff
- inet 192.168.1.1/24 brd 192.168.1.255 scope global eth4
- valid_lft forever preferred_lft forever
- 9: br-lan: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 1000
- link/ether brd ff:ff:ff:ff:ff:ff
- inet 192.168.1.1/24 brd 192.168.1.255 scope global br-lan
- valid_lft forever preferred_lft forever
- inet6 4 scope global dynamic noprefixroute
- valid_lft 334949sec preferred_lft 334949sec
- inet6 scope global noprefixroute
- valid_lft forever preferred_lft forever
- inet6 scope link
- valid_lft forever preferred_lft forever
- 12: wlan0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue master br-lan state UP group default qlen 1000
- link/ether brd ff:ff:ff:ff:ff:ff
- inet 192.168.1.1/24 brd 192.168.1.255 scope global wlan0
- valid_lft forever preferred_lft forever
- inet6 scope global noprefixroute
- valid_lft forever preferred_lft forever
- inet6 scope link
- valid_lft forever preferred_lft forever
and
Unfortunately I am having issues with my broadband so still not fully able to test my 1.2Gbps service.
UPDATE
Wont have broadband fixed till next week (ugh) but I can say that I am hitting a CPU core max out. If I run a speed test they are maxing out at 240Mbps. Core 0 is at 100% cpu.
I tried stopping and restarting the irqbalance service, futzing with the settings, etc but no matter what I do I cant seem to get irq31 to spread across cores.
So at the end of the day unless I can get a working 2.5Gbps mini PCIe card it looks like the hope of breaking the 1Gbps barrier on the apu is a lost cause for the average user. Frankly I am starting to look to build a low power Alder lake based router and may just setup my APU as an AP only.
Code- root@ghost:~# cat /proc/interrupts
- CPU0 CPU1 CPU2 CPU3
- 0: 27 0 0 0 IO-APIC 2-edge timer
- 4: 0 0 0 15 IO-APIC 4-edge ttyS0
- 8: 0 1 0 0 IO-APIC 8-edge rtc0
- 9: 0 0 0 0 IO-APIC 9-fasteoi acpi
- 16: 0 0 2051 0 IO-APIC 16-fasteoi mmc0
- 18: 0 0 0 26 IO-APIC 18-fasteoi ehci_hcd:usb1
- 19: 0 0 6753 52281 IO-APIC 19-fasteoi ahci[0000:00:11.0]
- 31: 18609911 0 0 0 PCI-MSI 262144-edge xhci_hcd
- 32: 0 0 0 0 PCI-MSI 262145-edge xhci_hcd
- 33: 0 0 0 0 PCI-MSI 262146-edge xhci_hcd
- 34: 0 0 0 0 PCI-MSI 262147-edge xhci_hcd
- 35: 0 0 0 0 PCI-MSI 262148-edge xhci_hcd
- 42: 0 0 0 0 PCI-MSI 1048576-edge eth1
- 43: 0 15908 0 12 PCI-MSI 1048577-edge eth1-TxRx-0
- 44: 88 71 5370 10391 PCI-MSI 1048578-edge eth1-TxRx-1
- 45: 17 112 5597 10194 PCI-MSI 1048579-edge eth1-TxRx-2
- 46: 22 50 5142 10706 PCI-MSI 1048580-edge eth1-TxRx-3
- 48: 0 0 0 0 PCI-MSI 1572864-edge eth2
- 49: 32 59 5085 10742 PCI-MSI 1572865-edge eth2-TxRx-0
- 50: 76 156 5677 10009 PCI-MSI 1572866-edge eth2-TxRx-1
- 51: 67 77 5244 10530 PCI-MSI 1572867-edge eth2-TxRx-2
- 52: 20 51 5726 10121 PCI-MSI 1572868-edge eth2-TxRx-3
- 54: 0 0 0 0 PCI-MSI 131073-edge ccp-1
- 56: 0 3359482 0 0 PCI-MSI 2097152-edge mt7915e
- NMI: 0 0 0 0 Non-maskable interrupts
- LOC: 18173502 4551790 4343410 4208556 Local timer interrupts
- SPU: 0 0 0 0 Spurious interrupts
- PMI: 0 0 0 0 Performance monitoring interrupts
- IWI: 0 0 0 0 IRQ work interrupts
- RTR: 0 0 0 0 APIC ICR read retries
- RES: 1272660 977475 3178870 1944499 Rescheduling interrupts
- CAL: 32740 32946 62828 32473 Function call interrupts
- TLB: 59062 80087 78514 83796 TLB shootdowns
- TRM: 0 0 0 0 Thermal event interrupts
- THR: 0 0 0 0 Threshold APIC interrupts
- DFR: 0 0 0 0 Deferred Error APIC interrupts
- MCE: 0 0 0 0 Machine check exceptions
- MCP: 106 106 106 106 Machine check polls
- ERR: 91822
- MIS: 0
- PIN: 0 0 0 0 Posted-interrupt notification event
- NPI: 0 0 0 0 Nested posted-interrupt event
- PIW: 0 0 0 0 Posted-interrupt wakeup event
- root@ghost:~#
-
It should not be a problem, devices without 160MHz support will just fall back to a narrower channel width that they support. I have some non-160MHz clients connected to mine for example.
So iIt shouldnt but with 160Mhz on it turns out the interface doesnt come up. It's likely the card is not DFC certified for use in the US. So if I select Us region it wont tenable if I choose 160Mhz? Maybe?
ref: https://www.smallnetbuilder.co…fi-channels-friend-or-foe
My heatsink mod/update netted me about a 12% drop in temps with the cover on. Not bad, but not great. The layers are the problem I am sure. I may see if I can get a larger copper billet style heatsink that could replace the RF cover and then machine the bottom face so that its just big enough to sit on the IC's and then attach it with thermal epoxy directly. For me that would be a lot of hand filing. Some more details here; https://twitter.com/drahcir_rahl/status/1540722244883808256 -
I've not tested the MT7915 with iperf because I am always happy with anything over 30Mbps on wifi. I cant run at 160Mhz only 2 of my 5 5Ghz clients support it currently. It runs very hot with my case closed (59-74c) so I have ordered a heatsink for it. https://www.amazon.com/dp/B07X…dt_b_product_details&th=1 just waiting to replace the cheap TIM inside the RF lids tonight before putting it all together. [Blocked Image: https://i.imgur.com/mJX0rro.jpg]
Not attached yet just there for the pic.
[Blocked Image: https://i.imgur.com/HSOO4Vm.png]
Since obviously they knew it needed to be cooled.
As for the CPU scaling yeah I run the following script after boot;
Code- echo 1000000 > /sys/devices/system/cpu/cpufreq/policy0/scaling_max_freq
- echo 1000000 > /sys/devices/system/cpu/cpufreq/policy0/scaling_min_freq
- echo 1000000 > /sys/devices/system/cpu/cpufreq/policy1/scaling_max_freq
- echo 1000000 > /sys/devices/system/cpu/cpufreq/policy1/scaling_min_freq
- echo 1000000 > /sys/devices/system/cpu/cpufreq/policy2/scaling_max_freq
- echo 1000000 > /sys/devices/system/cpu/cpufreq/policy2/scaling_min_freq
- echo 1000000 > /sys/devices/system/cpu/cpufreq/policy3/scaling_max_freq
- echo 1000000 > /sys/devices/system/cpu/cpufreq/policy3/scaling_min_freq
- echo 1000000 > /sys/devices/system/cpu/cpufreq/policy3/scaling_max_freq
- echo 1000000 > /sys/devices/system/cpu/cpufreq/policy3/scaling_min_freq
- for CPUFREQ in /sys/devices/system/cpu/cpu*/cpufreq/scaling_governor; do [ -f $CPUFREQ ] || continue; echo -n performance > $CPUFREQ; done
That seems to let the adapter work at 2.5Gbps for data transfer but iperf and iperf3 are still too cpu hungry, also the cdc_ncm driver that its picking by default doesnt provide any data with ethtool, have not figured out how to properly blacklist that driver so it picks the R8152 one instead. The clock speeds shown in htop still drift in the 900Mhz range.
-
No matter what comes next, I gotta say, PCEngines is the only real player in the mid-cost range for this type of hardware. The closest competition feature wise are all significantly higher cost or weaker platforms.
I dont want to name names but the cheapest device I can see that I would willing replace my APU2 with is 549 euro.
-
Appears the MT7915 has heat issues. With no added cooling some folks are reporting temps in excess of 100c I have heatsinks coming and will report back. Card still working well, just gets hot.
Mine is currently clocking in at 74c, CPU is at 53c and my cooling for the CPU is stock with the exception of using a copper plate instead of the aluminum ones that come with our cases normally.
root@ghost:~# sensors
mt7915_phy1-pci-0400
Adapter: PCI adapter
temp1: +74.0°C (high = +120.0°C, crit = +110.0°C)
k10temp-pci-00c3
Adapter: PCI adapter
temp1: +53.2°C (high = +70.0°C)
(crit = +105.0°C, hyst = +104.0°C)
acpitz-acpi-0
Adapter: ACPI interface
temp1: +53.2°C (crit = +115.0°C)
mt7915_phy0-pci-0400
Adapter: PCI adapter
temp1: +74.0°C (high = +120.0°C, crit = +110.0°C)
fam15h_power-pci-00c4
Adapter: PCI adapter
power1: 2.77 W (interval = 0.01 s, crit = 6.00 W)
-
Ok so it wasnt final verdict.
The CPU scaling is not working well enough in openwrt. Once I set it to performance and set them to 1ghz then I could rsync files at 2.5Gbs over the wire with the ASUS 2.5G Ethernet USB Adapter (USB-C2500) Realtek R8152 based adapter on the APU2 and the other on my windows laptop. Some more info here. https://forum.openwrt.org/t/re…2-2/125102/23?u=rahlquist
-
Final verdict on the 2.5Gbps usb adapters. If locked to 1Gbps, they function fine if you get the right drivers in place. Let them negotiate to 2.5Gbps and iperf tests only work up to just under 700Mbps. It uses up 100% of the core its running on and limits the bandwidth. I may still try a 2.5Gbps Intel I225 based mini pcie but I am afraid we have just reached some limitations of the platform currently.
Going to roll back to their current RC and drop this back into production to test the MT7915 for reliability (will be weird to only need 2 antennas vs the 6 I had before. Will update this thread later with results on the MT7915.
Not sure why I could never get QCA6391 to work. Unless there is broken or funky about its pinout.
-
Ok so here is where I am with this.
Two 2.5Gbps usb 3 adapters
RTL8152 has driver issues in both my openwrt 5.15 based kernel and 5.15 and higher manjaro kernels.
RTL8169 can work in openwrt and manjaro 5.15 and later, though no extensive testing.
Two Wifi 6 mini PCIe adapters
MT7915 "works" with the current OpenWRT OpenWrt 22.03.0-rc1 and my custom roll of 5.15.x but on reboot it is not automatically starting.
QCA6391 Cant even get it recognized as a device. Since its a pre-production product on a custom PCB, it may have been DOA too. Even so, this would require Ath11k drivers that are not in mainline OpenWRT yet and I had not gotten to getting them working in my build yet.I'll be honest, I haven't had this many network issues since win 95 dropped LOL.
I think of all of it the MT7915 shows a LOT of promise, and should be fine after a few more driver revisions.
-
Yes the MT7915 works. Some caveats, the speed in AC vs AX modes, avg AC speed is 7-800's
In AX I cant get it to break 400, yet.Code- 04:00.0 Unclassified device [0002]: MEDIATEK Corp. MT7915E 802.11ax PCI Express Wireless Network Adapter (prog-if 80)
- Subsystem: MEDIATEK Corp. MT7915E 802.11ax PCI Express Wireless Network Adapter
- Control: I/O- Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx+
- Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
- Latency: 0, Cache Line Size: 64 bytes
- Interrupt: pin A routed to IRQ 58
- Region 0: Memory at d0000000 (64-bit, prefetchable) [size=1M]
- Region 2: Memory at d0100000 (64-bit, prefetchable) [size=16K]
- Region 4: Memory at d0104000 (64-bit, prefetchable) [size=4K]
- Capabilities: [80] Express (v2) Endpoint, MSI 00
- DevCap: MaxPayload 128 bytes, PhantFunc 0, Latency L0s <1us, L1 <4us
- ExtTag+ AttnBtn- AttnInd- PwrInd- RBE+ FLReset- SlotPowerLimit 0.000W
- DevCtl: CorrErr- NonFatalErr- FatalErr- UnsupReq-
- RlxdOrd+ ExtTag+ PhantFunc- AuxPwr- NoSnoop+
- MaxPayload 128 bytes, MaxReadReq 512 bytes
- DevSta: CorrErr- NonFatalErr- FatalErr- UnsupReq- AuxPwr- TransPend-
- LnkCap: Port #1, Speed 5GT/s, Width x1, ASPM L0s L1, Exit Latency L0s <2us, L1 <8us
- ClockPM+ Surprise- LLActRep- BwNot- ASPMOptComp+
- LnkCtl: ASPM Disabled; RCB 64 bytes, Disabled- CommClk+
- ExtSynch- ClockPM- AutWidDis- BWInt- AutBWInt-
- LnkSta: Speed 5GT/s (ok), Width x1 (ok)
- TrErr- Train- SlotClk+ DLActive- BWMgmt- ABWMgmt-
- DevCap2: Completion Timeout: Range ABCD, TimeoutDis+ NROPrPrP- LTR+
- 10BitTagComp- 10BitTagReq- OBFF Not Supported, ExtFmt+ EETLPPrefix-
- EmergencyPowerReduction Not Supported, EmergencyPowerReductionInit-
- FRS- TPHComp- ExtTPHComp-
- AtomicOpsCap: 32bit- 64bit- 128bitCAS-
- DevCtl2: Completion Timeout: 50us to 50ms, TimeoutDis- LTR- OBFF Disabled,
- AtomicOpsCtl: ReqEn-
- LnkCap2: Supported Link Speeds: 2.5-5GT/s, Crosslink- Retimer- 2Retimers- DRS-
- LnkCtl2: Target Link Speed: 5GT/s, EnterCompliance- SpeedDis-
- Transmit Margin: Normal Operating Range, EnterModifiedCompliance- ComplianceSOS-
- Compliance De-emphasis: -6dB
- LnkSta2: Current De-emphasis Level: -6dB, EqualizationComplete- EqualizationPhase1-
- EqualizationPhase2- EqualizationPhase3- LinkEqualizationRequest-
- Retimer- 2Retimers- CrosslinkRes: unsupported
- Capabilities: [e0] MSI: Enable+ Count=1/32 Maskable+ 64bit+
- Address: 00000000fee02004 Data: 4025
- Masking: fffffffe Pending: 00000000
- Capabilities: [f8] Power Management version 3
- Flags: PMEClk- DSI+ D1+ D2+ AuxCurrent=0mA PME(D0+,D1+,D2+,D3hot+,D3cold+)
- Status: D0 NoSoftRst+ PME-Enable- DSel=0 DScale=0 PME-
- Capabilities: [100 v1] Vendor Specific Information: ID=1556 Rev=1 Len=008 <?>
- Capabilities: [108 v1] Latency Tolerance Reporting
- Max snoop latency: 0ns
- Max no snoop latency: 0ns
- Capabilities: [110 v1] L1 PM Substates
- L1SubCap: PCI-PM_L1.2+ PCI-PM_L1.1+ ASPM_L1.2+ ASPM_L1.1+ L1_PM_Substates+
- PortCommonModeRestoreTime=3us PortTPowerOnTime=28us
- L1SubCtl1: PCI-PM_L1.2- PCI-PM_L1.1- ASPM_L1.2- ASPM_L1.1-
- T_CommonMode=0us LTR1.2_Threshold=0ns
- L1SubCtl2: T_PwrOn=10us
- Capabilities: [200 v2] Advanced Error Reporting
- UESta: DLP- SDES- TLP- FCP- CmpltTO- CmpltAbrt- UnxCmplt- RxOF- MalfTLP- ECRC- UnsupReq- ACSViol-
- UEMsk: DLP- SDES- TLP- FCP- CmpltTO- CmpltAbrt- UnxCmplt- RxOF- MalfTLP- ECRC- UnsupReq- ACSViol-
- UESvrt: DLP+ SDES+ TLP- FCP+ CmpltTO- CmpltAbrt- UnxCmplt- RxOF- MalfTLP+ ECRC- UnsupReq- ACSViol-
- CESta: RxErr- BadTLP- BadDLLP- Rollover- Timeout- AdvNonFatalErr-
- CEMsk: RxErr- BadTLP- BadDLLP- Rollover- Timeout- AdvNonFatalErr+
- AERCap: First Error Pointer: 00, ECRCGenCap- ECRCGenEn- ECRCChkCap- ECRCChkEn-
- MultHdrRecCap- MultHdrRecEn- TLPPfxPres- HdrLogCap-
- HeaderLog: 00000000 00000000 00000000 00000000
- Kernel driver in use: mt7915e
-
So, did it work alright?
If it works fine, and if you provided some details, we could add it to the list of supported Wi-Fi cards in the official documentation
Life got busy, I plan on working with it tomorrow. I had a lot of trouble getting a custom openwrt compile done that would include the proper drivers and firmware for the QCA6391 I will give it a shot though. I also picked up a "WiFi 6 Mediatek MT7915 2T2R" card that the drivers should already be in the main branch.
Also will be testing two usb3 to 2.5Gbps Ethernet adapters.
Killer E3100 USB-C 3.1 to RJ-45 2.5Gbps Ethernet Adapter (100123) Realtek R8169 based
ASUS 2.5G Ethernet USB Adapter (USB-C2500) Realtek R8152 based
My broadband is 1.2Gbps speed so would like to use one of the above for uplink to my APU2D4. If I can get 1.2Gbps to it, I can live with 1Gbps out over Ethernet and having wifi. Or I may just bond two of the APU2's ports to my switch. Or maybe 1 usb adapter and one of these https://www.amazon.com/IO-CRES…-Converter/dp/B094YY6KX6/
Will post some results tomorrow.
-
Hello,
Are the PCI Express side slot connectors on all APU3 enabled and able to be used with proper connection, or do they have to be special ordered? Considering picking up one.
If I read the system specs right, that should be equivalent to a PCIe x1 2.0 slot so limited to 500MB/s ?
-
While I agree/disagree with some of the thoughts here, since this is a wish list I am going to stick to that.
CPU is a struggle here, not a ton of good SOC from AMD recently.
https://www.cpubenchmark.net/c…A6-9220C/4156vs2473vs3455
Compares the GX-412hc vs the 9220c and a low power 5700U (15w) ryzen. With the Ryzen costs would be more due to needing more support components I would think?
Ram ECC please. 4GB should be fine
Network Config 1 (likely cheaper)
1 SFP+ cage 82599EN
3x Intel I225 2.5G
Network Config 2 (higher power(13.6 W), expensive)
Intel X557-AT4 (4 10Gbps ports)
Network Config 3 (compromise of above)
Intel X557-AT2 (2 10 Gbps ports)
2x Intel I225 2.5G
2 Mini PCIe w/ 1 connected to sim and the other connected with UART support for cards like QCA6391 so BT could be supported. Enhance power handling to spec max for these slots
1x SATA
1x M.2
Optional side connect like APU3 for PCIe expansion with a PCIe x4 slot.
2 USB 2.x internal header
2 USB 3.x external ports
TPM
RTC
Hardware Watchdog
Serial console
-
Some devices are more tolerant. With the shortage of parts for these boards and them being generally hard to get now..
Something along this line may do what you need. https://www.amazon.com/Convert…ransformer/dp/B08D8L79TS/
That's not a referral or a recommendation, just a suggested starting point. With buck to lower voltage and boost to raise it when it drops (like during cold cranking) it may protect the device from;
Key on (+12v), start boot-->engine crank, voltage(+10v) drop, APU shuts back off>post crank(+13.8-14v) power up again.
Good luck! These apu would make a great mobile vehicle AP/Router.
-
So currently I am running 2 AEX-QCA9880-NX, one for 2.4ghz bands and one for 5.0ghx bands in my apu2d4.
I just received a AEX-QCA6391-NX1 that I want to see if I can get working under OpenWRT. The BT portion of the card wont work at all in the APU2 slots because the needed UART connections either run to the sim card or to the unpopulated sim. I may try an adapter later that will allow it just for grins.
Hoping this isn't too power hungry. Looks like max power draw is under 700ma.
Would be so nice to free up a slot.
-
I would not advise exceeding 12v. That said based on p14-15 of the schematic.
The 3.3v reg APW8713A by spec can take up to 28v in
The 5v reg APW8715a by spec can take up to 28v in
But that is just 2 components.
Without digging more I am willing to bet it was designed with parts with a bit of range. The problem is will it produce the proper output with a higher input? Not sure.
If it were me I would make certain to keep it at 12. I am assuming you have an application where it would be easier to feed it more. A cheap $10-20 buck/boos converter could take 8-40v in and turn it into 12V within 1.5% or tighter if you dig for a good one and pay a smidge more. If you were looking to do POE then there are splitters that will get it down to 12v/2a just like the original AC adapters.