Hi David, 🤖 This review was drafted with assistance from Claude (Anthropic) and reviewed by me before posting.
No functional concerns with this patch in isolation. Flagging for context going into the rest of the series: by the end of it, the fixed MLX5_MAX_MAC_ADDRESSES / MLX5_MAX_UC_MAC_ADDRESSES bound is replaced by a device-reported dynamic limit, but mlx5_flow_hw.c's HWS control-flow paths are never updated to match it (see my reply on patch 5/5, "net/mlx5: accept more unicast MAC addresses"). -- Raslan Darawsheh
