[PATCH v4 3/3] PCI/AER: Report fatal errors of RCiEP and EP if link recoverd

Shuai Xue xueshuai at linux.alibaba.com
Thu Apr 24 21:48:57 AEST 2025



在 2025/3/17 14:02, Shuai Xue 写道:
> 
> 
> 在 2025/3/3 12:33, Shuai Xue 写道:
>>
>>
>> 在 2025/3/3 11:43, Sathyanarayanan Kuppuswamy 写道:
>>>
>>> On 2/16/25 6:42 PM, Shuai Xue wrote:
>>>> The AER driver has historically avoided reading the configuration space of
>>>> an endpoint or RCiEP that reported a fatal error, considering the link to
>>>> that device unreliable. Consequently, when a fatal error occurs, the AER
>>>> and DPC drivers do not report specific error types, resulting in logs like:
>>>>
>>>>    pcieport 0000:30:03.0: EDR: EDR event received
>>>>    pcieport 0000:30:03.0: DPC: containment event, status:0x0005 source:0x3400
>>>>    pcieport 0000:30:03.0: DPC: ERR_FATAL detected
>>>>    pcieport 0000:30:03.0: AER: broadcast error_detected message
>>>>    nvme nvme0: frozen state error detected, reset controller
>>>>    nvme 0000:34:00.0: ready 0ms after DPC
>>>>    pcieport 0000:30:03.0: AER: broadcast slot_reset message
>>>>
>>>> AER status registers are sticky and Write-1-to-clear. If the link recovered
>>>> after hot reset, we can still safely access AER status of the error device.
>>>> In such case, report fatal errors which helps to figure out the error root
>>>> case.
>>>>
>>>> After this patch, the logs like:
>>>>
>>>>    pcieport 0000:30:03.0: EDR: EDR event received
>>>>    pcieport 0000:30:03.0: DPC: containment event, status:0x0005 source:0x3400
>>>>    pcieport 0000:30:03.0: DPC: ERR_FATAL detected
>>>>    pcieport 0000:30:03.0: AER: broadcast error_detected message
>>>>    nvme nvme0: frozen state error detected, reset controller
>>>>    pcieport 0000:30:03.0: waiting 100 ms for downstream link, after activation
>>>>    nvme 0000:34:00.0: ready 0ms after DPC
>>>>    nvme 0000:34:00.0: PCIe Bus Error: severity=Uncorrectable (Fatal), type=Data Link Layer, (Receiver ID)
>>>>    nvme 0000:34:00.0:   device [144d:a804] error status/mask=00000010/00504000
>>>>    nvme 0000:34:00.0:    [ 4] DLP                    (First)
>>>>    pcieport 0000:30:03.0: AER: broadcast slot_reset message
>>>
>>> IMO, above info about device error details is more of a debug info. Since the
>>> main use of this info use to understand more details about the recovered
>>> DPC error. So I think is better to print with debug tag. Lets see what others
>>> think.
>>>
>>> Code wise, looks fine to me.
>>
>> thanks, looking forward to more feedback.
>>>
>>>
> 
> Hi, all,
> 
> Gentle ping.
> 
> Thanks.
> Shuai
> 


Hi, all,
  
Gentle ping.
  
Thanks.
Shuai


More information about the Linuxppc-dev mailing list