is_upper_nde...aster_filter(9)

💖 is_upper_ndev_bond_master_filter(9) 💖

okay so bestie, this is the Kernel API Manual page for is_upper_ndev_bond_master_filter, and the header was out here doing the most with is_upper_nde...aster_filter(9) — truncated queen behavior, honestly. The full name is the long one, and she's an absolute icon.

✨ NAME ✨

is_upper_ndev_bond_master_filter — check if a given netdevice is bond master device of netdevice of the RDMA device of port. that's the whole mission, no cap.

🛰️ SYNOPSIS 🛰️

here's the signature exactly as it is, because we do not mess with the code:

bool is_upper_ndev_bond_master_filter (struct ib_device *ib_dev , u32
port , struct net_device *rdma_ndev , void *cookie );

🎀 ARGUMENTS 🎀

four arguments, bestie, each one doing its little main-character thing:

• ib_dev — IB device to check.
• port — Port to consider for adding default GID.
• rdma_ndev — Pointer to rdma netdevice.
• cookie — Netdevice to consider to form a default GID.

💅 DESCRIPTION 💅

so the vibe check is: is_upper_ndev_bond_master_filter returns true if a cookie_netdev is bond master device and rdma_ndev is its lower netdevice. It might not have been established as slave device yet — iykyk, that's the caveat, don't come for me.

🔗 SEE ALSO 🔗

• Kernel file ./drivers/infiniband/core/roce_gid_mgmt.c
• is_ndev_for_default_gid_filter(9)
• del_default_gids(9)
• rdma_roce_rescan_device(9)
• rdma_roce_rescan_port(9)

mwah 💋 this page comes from ./drivers/infiniband/core, dated October 2026, and the footer was calling it is_upper_nde...aster_filter(9) — love that for her. byeee 🫶