On 15/01/2025 08:42, Jon Pan-Doh wrote:
Info logged is duplicated when either the source device is processed. If
no source device is found, than an error is logged.
Nit: s/than/then/
Code flow:
aer_isr_one_error()
-> aer_print_port_info()
-> find_source_device()
-> return/pci_info() if no device found else continue
-> aer_process_err_devices()
-> aer_print_error()
aer_print_port_info():
[ 21.596150] pcieport 0000:00:04.0: Correctable error message received
from 0000:01:00.0
I agree that the bus, device and function info is repeated later, but
isn't this line also about the fact we deal with one or multiple errors
in a message? The question is how valuable this information, in itself, is.
All the best,
Karolina
aer_print_error():
[ 21.596163] e1000e 0000:01:00.0: PCIe Bus Error: severity=Correctable, type=Data Link Layer, (Receiver ID)
[ 21.600575] e1000e 0000:01:00.0: device [8086:10d3] error status/mask=00000040/0000e000
[ 21.604707] e1000e 0000:01:00.0: [ 6] BadTLP
Tested using aer-inject[1] tool. No more root port log on dmesg.
>
[1] https://git.kernel.org/pub/scm/linux/kernel/git/gong.chen/aer-inject.git
Signed-off-by: Jon Pan-Doh <pandoh@xxxxxxxxxx>
