update_qp_retry_state(9)

📦 version 7.3.1, in nixpkgs unstable

okay so bestie, welcome to update_qp_retry_state(9) from the Kernel API Manual — yes, section 9, we're deep in kernel docs now and it's giving tiny but important helper 💅✨

💖 NAME 💖

update_qp_retry_state — update qp retry state. That's the whole assignment, no cap 🎀

✨ SYNOPSIS ✨

void update_qp_retry_state (struct rvt_qp *qp , u32 psn , u32 spsn , u32 lpsn );

🎀 ARGUMENTS 🎀

  • qp — the QP. periodt 🫶
  • psn — the packet sequence number of the TID RDMA WRITE RESP.
  • spsn — the start psn for the given TID RDMA WRITE swqe.
  • lpsn — the last psn for the given TID RDMA WRITE swqe.

🌸 DESCRIPTION 🌸

This function is called to update the qp retry state upon receiving a TID WRITE RESP after the qp is scheduled to retry a request. So: retry-state update time, triggered by that TID WRITE RESP, after the qp is scheduled to retry — that's the whole plot, bestie 😭🔥

🦋 SEE ALSO 🦋

Kernel file ./drivers/infiniband/hw/hfi1/rc.c is in the mix, plus the whole reference crew: make_rc_ack(9), hfi1_make_rc_req(9), update_num_rd_atomic(9), reset_psn(9), hfi1_rc_verbs_aborted(9), rc_rcv_resp(9), rc_rcv_error(9), hfi1_rc_rcv(9). Go click through those next, iykyk 🌈💜

Signing off with the source/version ./drivers/infiniband/hw/hfi1 and the date October 2026 — and yes, this whole cute little page is update_qp_retry_state(9), mwah 💋🎀

📦 where this page comes from