srpt_cm_req_recv(9)

πŸ“¦ version 7.3.1, in nixpkgs unstable

πŸ’– NAME πŸ’–

okay so bestie, meet srpt_cm_req_recv(9) from the Kernel API Manual ✨. its whole assignment is processing the event IB_CM_REQ_RECEIVED πŸŽ€ β€” that's the name, that's the job, and it's serving exactly that.


πŸ“œ SYNOPSIS πŸ“œ

int srpt_cm_req_recv (struct srpt_device *const sdev , struct ib_cm_id
       *ib_cm_id , struct rdma_cm_id *rdma_cm_id , u8 port_num , __be16 pkey ,
       const struct srp_login_req *req , const char *src_addr );

🫢 ARGUMENTS 🫢

  • sdev β€” the HCA through which the login request was received πŸ›°οΈ
  • ib_cm_id β€” the IB/CM connection identifier, for the IB/CM case πŸͺͺ
  • rdma_cm_id β€” the RDMA/CM connection identifier, for the RDMA/CM case πŸͺͺ
  • port_num β€” the port through which the REQ message was received πŸ”Œ
  • pkey β€” the P_Key of the incoming connection πŸ”‘
  • req β€” the SRP login request πŸ“¨
  • src_addr β€” the GID (IB/CM) or IP address (RDMA/CM) of the port that submitted the login request 🌐

πŸ’… DESCRIPTION πŸ’…

okay so here's the ownership tea β˜•: if this function returns zero, ownership of the cm_id is transferred to the target session. otherwise, the caller remains the owner of cm_id. periodt πŸ’…

this is the main character detail, bestie: zero return = ownership of the cm_id moves to the target session; not-zero return = the caller stays the owner of cm_id. no cap πŸ”₯

πŸ‘‘ SEE ALSO πŸ‘‘

the group chat of related kernel friends πŸ’¬:


xoxo, your kernel bestie πŸ’– β€” source/version: ./drivers/infiniband/ulp/srpt; date: October 2026; page: srpt_cm_req_recv(9). go read the source and slay 🫢✨

πŸ“¦ where this page comes from