rdma_find_gid(9)

πŸ“¦ version 7.3.1, in nixpkgs unstable

πŸ’– rdma_find_gid(9) β€” Kernel API Manual πŸ’–

okay bestie, welcome to the rdma_find_gid(9) page from the Kernel API Manual. This one is a little GID-lookup situation, and we’re gonna read it together. πŸ’…


✨ NAME ✨

rdma_find_gid β€” returns SGID attributes if the matching GID is found. That’s the whole mission. πŸŽ€

🌸 SYNOPSIS 🌸

const struct ib_gid_attr * rdma_find_gid (struct ib_device *device ,
const union ib_gid *gid , enum ib_gid_type gid_type , struct net_device
*ndev );

πŸ¦‹ ARGUMENTS πŸ¦‹

  • device β€” the device to query. Like, which RDMA device are we asking? πŸ‘€
  • gid β€” the GID value to search for. The actual thing we’re hunting.
  • gid_type β€” the GID type to search for. Gotta match the type too, no cap.
  • ndev β€” in RoCE, the net device of the device. NULL means ignore. So if it’s NULL, we’re not checking that.

πŸ’œ DESCRIPTION πŸ’œ

rdma_find_gid searches for the specified GID value in the software cache. It’s giving β€œlook it up in the cache, bestie” energy. ✨

It returns GID attributes if a valid GID is found, or returns ERR_PTR for the error. The caller must invoke rdma_put_gid_attr to release the reference. Do not forget this part β€” not me being dramatic, it’s literally the cleanup rule. 🫢

lowkey important: the caller has to call rdma_put_gid_attr to release the reference.

🌈 SEE ALSO 🌈


okay that’s the page, bestie. This one’s from source/version ./drivers/infiniband/core, dated October 2026, and its title is rdma_find_gid(9). byeee, go release your references πŸ’–πŸŽ€βœ¨

πŸ“¦ where this page comes from