Hi, This recently hit our tree and was deployed, and our operations team is quite upset because this results in a massive (orders of magnitude) increase in our kernel log messages due to our monitoring scripts probing each port.
Can we please move this under bootverbose, or have some other way to suppress these prints? I can back this out in our tree, but I don't want to create yet another local patch. Thank you, Drew On Sun, May 5, 2024, at 11:15 PM, Konstantin Belousov wrote: > The branch main has been updated by kib: > > URL: > https://cgit.FreeBSD.org/src/commit/?id=c097967b9af946cf46246cce5adde53f7e955ecc > > commit c097967b9af946cf46246cce5adde53f7e955ecc > Author: Konstantin Belousov <[email protected]> > AuthorDate: 2024-05-05 07:09:07 +0000 > Commit: Konstantin Belousov <[email protected]> > CommitDate: 2024-05-06 03:15:35 +0000 > > mlx5en: add diagnostic in one more case of failed eeprom read preparation > > Sponsored by: Nvidia Networking > MFC after: 1 week > --- > sys/dev/mlx5/mlx5_en/mlx5_en_main.c | 3 +++ > 1 file changed, 3 insertions(+) > > diff --git a/sys/dev/mlx5/mlx5_en/mlx5_en_main.c > b/sys/dev/mlx5/mlx5_en/mlx5_en_main.c > index e1dbd02fcf3a..962705e6d258 100644 > --- a/sys/dev/mlx5/mlx5_en/mlx5_en_main.c > +++ b/sys/dev/mlx5/mlx5_en/mlx5_en_main.c > @@ -3674,6 +3674,9 @@ out: > /* Check if module is present before doing an access */ > module_status = mlx5_query_module_status(priv->mdev, module_num); > if (module_status != MLX5_MODULE_STATUS_PLUGGED_ENABLED) { > + mlx5_en_err(ifp, > + "Query module %d status: not plugged (%d), eeprom reading is not > supported\n", > + module_num, module_status); > error = EINVAL; > goto err_i2c; > } >
