How to identify problematic hardware by ID or PCIe connection

English Hardware
1

Hi,
I’m getting hundreds of error logs per minute from some device. I’ve used lspci and dmidecode and hunted through my BIOS but I can’t find any way to link the data they spit out to my actual hardware. I can see the problem seems to be with PCI device 00:1c.4 and 8086:7ab . How am I supposed to track a device down?

9/10/22 4:16 PM    kernel    thunderbolt 0000:09:00.0: AER: can't recover (no error_detected callback)
9/10/22 4:16 PM    kernel    xhci_hcd 0000:3d:00.0: AER: can't recover (no error_detected callback)
9/10/22 4:16 PM    kernel    pcieport 0000:00:1c.4: AER: device recovery failed
9/10/22 4:16 PM    kernel    pcieport 0000:00:1c.4: AER: Corrected error received: 0000:00:1c.4
9/10/22 4:16 PM    kernel    pcieport 0000:00:1c.4: AER: can't find device of ID00e4
9/10/22 4:16 PM    kernel    pcieport 0000:00:1c.4: AER: Uncorrected (Non-Fatal) error received: 0000:00:1c.4
9/10/22 4:16 PM    kernel    pcieport 0000:00:1c.4: PCIe Bus Error: severity=Uncorrected (Non-Fatal), type=Transaction Layer, (Requester ID)
9/10/22 4:16 PM    kernel    pcieport 0000:00:1c.4:   device [8086:7abc] error status/mask=00100000/00004000
9/10/22 4:16 PM    kernel    pcieport 0000:00:1c.4:    [20] UnsupReq               (First)
9/10/22 4:16 PM    kernel    pcieport 0000:00:1c.4: AER:   TLP Header: 34000000 07000052 00000000 00000000


9/10/22 4:16 PM    kernel    pcieport 0000:00:1c.4:    [20] UnsupReq               (First)
9/10/22 4:16 PM    kernel    pcieport 0000:00:1c.4: AER:   TLP Header: 34000000 07000052 00000000 00000000
9/10/22 4:16 PM    kernel    pcieport 0000:00:1c.4: PCIe Bus Error: severity=Uncorrected (Non-Fatal), type=Transaction Layer, (Requester ID)
9/10/22 4:16 PM    kernel    pcieport 0000:00:1c.4:   device [8086:7abc] error status/mask=00100000/00004000



lspci -v -s 00:1c.4
00:1c.4 PCI bridge: Intel Corporation Alder Lake-S PCH PCI Express Root Port #5 (rev 11) (prog-if 00 [Normal decode])
        Flags: bus master, fast devsel, latency 0, IRQ 143
        Bus: primary=00, secondary=07, subordinate=70, sec-latency=0
        I/O behind bridge: 00006000-00008fff [size=12]
        Memory behind bridge: 54000000-820fffff [size=737]
        Prefetchable memory behind bridge: 0000004000000000-000000404a0fffff [size=1185]
        Capabilities: [40] Express Root Port (Slot+), MSI 00
        Capabilities: [80] MSI: Enable+ Count=1/1 Maskable- 64bit+
        Capabilities: [98] Subsystem: ASUSTeK Computer Inc. Device 8694
        Capabilities: [a0] Power Management version 3
        Capabilities: [100] Advanced Error Reporting
        Capabilities: [220] Access Control Services
        Capabilities: [200] L1 PM Substates
        Capabilities: [150] Precision Time Measurement
        Capabilities: [a30] #19
        Capabilities: [a90] #25
        Kernel driver in use: pcieport

[/size][/size][/size]

2

Hi
Thunderbolt ports by the look: See https://bugzilla.kernel.org/show_bug.cgi?id=215453 are you using thunderbolt? I would disable temporarily in the BIOS and see if that helps.

3

8086:7abc, not 8086:7ab. Intel PCI Express Root Port, Alder Lake-S PCH PCI Express Root Port #5. Many users complain about it. Try to update BIOS. New CPU firmware may help.
Check dmesg.
PCIe error may be a result of bending motherboard, which is typical with Alder Lake. Try to dissemble and carefully assemble PC - if you can. Or ask manufacturer about warranty case.
Try to use lower PCIe version - 3.0 or 2.x.

4

Thanks malcolm. Indeed, I have TB though nothing is plugged into it. Will test after disabling.

Thanks, I have the newest BIOS and Intel firmware from OBS kernel stable backport.

I don’t think it would be a bent board because this one has a steel plate on the entire backside – very stiff. Good suggestions though. I can try dropping PCIe version but I still need to identify which device/slot/lane is causing issues.

I still want to know how a user is supposed to identify hardware from the logs!
8086:7abc = ???
PCH PCI Express Root Port #5 = Do I need to ask the manufacturer for help identifying this?

5

Hi
It should be the thunderbolt port…