Re: [Bug 215644] New: "PCIe Bus Error: severity=Corrected, type=Data Link Layer" flood

Re: [Bug 215644] New: "PCIe Bus Error: severity=Corrected, type=Data Link Layer" flood

[Bjorn Helgaas]

On Sun, Feb 27, 2022 at 02:34:29AM +0000, bugzilla-daemon@kernel.org wrote:
> https://bugzilla.kernel.org/show_bug.cgi?id=215644
> 
>             Bug ID: 215644
>            Summary: "PCIe Bus Error: severity=Corrected, type=Data Link
>                     Layer" flood
>            Product: Drivers
>     Kernel Version: 5.13.0-30-generic(Ubuntu 20.04 LTS)
>           Hardware: Intel
>           Priority: P1
>           Reporter: wangzhihao9@hotmail.com
>         Regression: No
> 
> [36418.907739] pcieport 0000:00:1c.4: PCIe Bus Error: severity=Corrected, type=Data Link Layer, (Receiver ID)
> [36418.907759] pcieport 0000:00:1c.4:   device [8086:02bc] error status/mask=00000080/00002000
> [36418.907779] pcieport 0000:00:1c.4:    [ 7] BadDLLP
> [36422.383718] pcieport 0000:00:1c.4: AER: Corrected error received: 0000:00:1c.4
> [36422.383758] pcieport 0000:00:1c.4: PCIe Bus Error: severity=Corrected, type=Data Link Layer, (Transmitter ID)
> [36422.383770] pcieport 0000:00:1c.4:   device [8086:02bc] error status/mask=00001000/00002000
> [36422.383785] pcieport 0000:00:1c.4:    [12] Timeout
> [36423.383999] pcieport 0000:00:1c.4: AER: Corrected error received: 0000:00:1c.4
> [36423.384044] pcieport 0000:00:1c.4: PCIe Bus Error: severity=Corrected, type=Data Link Layer, (Transmitter ID)
> [36423.384057] pcieport 0000:00:1c.4:   device [8086:02bc] error status/mask=00001000/00002000
> [36423.384073] pcieport 0000:00:1c.4:    [12] Timeout
> [36423.384099] pcieport 0000:00:1c.4: AER: Corrected error received: 0000:00:1c.4
> [36423.384137] pcieport 0000:00:1c.4: AER: can't find device of ID00e4
> [36423.384145] pcieport 0000:00:1c.4: AER: Corrected error received: 0000:00:1c.4
> [36423.384167] pcieport 0000:00:1c.4: PCIe Bus Error: severity=Corrected, type=Data Link Layer, (Transmitter ID)
> [36423.384177] pcieport 0000:00:1c.4:   device [8086:02bc] error status/mask=00001000/00002000
> [36423.384188] pcieport 0000:00:1c.4:    [12] Timeout
> [36423.384218] pcieport 0000:00:1c.4: AER: Corrected error received: 0000:00:1c.4
> [36423.384244] pcieport 0000:00:1c.4: AER: can't find device of ID00e4
> [36425.385279] pcieport 0000:00:1c.4: AER: Corrected error received: 0000:00:1c.4
> 
> The Device is 00:1c.0 PCI bridge: Intel Corporation Device 02b8 (rev f0)
>               -- 02:00.0 3D controller: NVIDIA Corporation GP108M [GeForce MX250] (rev a1)
> 
> It happens just after I open the ASPM of the video card to make CPU to achieve
> the Package pc8 state for Power Saving. Normally only pc2.
> 
> Hewlett Packard disabled the ASPM of Video Card in the BIOS defaultly.

Thanks for the report!  I'm ccing it to the linux-pci since the
Bugzilla doesn't get much visibility.

How do you enable ASPM?

Please collect the complete dmesg log and "sudo lspci -vv" output
(before enabling ASPM, and also after if possible) and attach them to
the bugzilla.

Bjorn