Re: [net-next V4 5/5] eth: mlx5: Move pause storm errors to pause stats

[Mohsin Bashir]

On 3/5/26 6:12 AM, Tariq Toukan wrote:
> If we return here, tx_pause_storm_events remains unset, while rx/ 
> tx_pause_frames are already assigned.
> Is that acceptable? I'm probably fine with it.

Right. Since we are aggregating across all prio and if we fail 
mid-flight, IMO, we should not report partial sum which maybe inaccurate.