svc_rdma_send_error_msg(9)

📦 version 7.3.1, in nixpkgs unstable

✨ svc_rdma_send_error_msg(9) ✨

okay so bestie, welcome to the kernel API manual for svc_rdma_send_error_msg 💅 this one's a function in svc_rdma_send_error_msg(9) and honestly it's giving "sorry, can't parse that" energy but make it RDMA 💖

🌸 SYNOPSIS 🌸

void svc_rdma_send_error_msg (struct svcxprt_rdma *rdma , struct
svc_rdma_send_ctxt *sctxt , struct svc_rdma_recv_ctxt *rctxt , int
status );

like, yes, that's the whole signature, no cap ✨

🎀 ARGUMENTS 🎀

  • rdma — the controlling transport context, aka the main character of this whole transport situation 👑
  • sctxt — send context for the response, literally the lil envelope the reply goes in 💌
  • rctxt — receive context for the incoming bad message, aka the messy incoming vibe we're responding to 🙈
  • status — a negative errno indicating the error that occurred, so like, the numeric reason things went left 😭

💖 DESCRIPTION 💖

okay so picture this: given the client-provided Read, Write, and Reply chunks, the server was not able to parse the Call or form a complete Reply 🫠 and at that point you can't just leave the client hanging, so you return an RDMA_ERROR message so the client can retire the RPC transaction ✨ periodt.

and here's the iconic part: the caller does not have to release sctxt. it is released by Send completion, or by this function on error 💅 so like, don't be a hero and free it yourself, bestie, the machinery's got it 🫶

🌈 SEE ALSO 🌈

Kernel file ./net/sunrpc/xprtrdma/svc_rdma_sendto.c

and that's the whole page, bestie 💖🦋 this lil baddie comes from ./net/sunrpc/xprtrdma, dated October 2026, and it's filed under svc_rdma_send_error_msg(9) — slay, go read the source, iykyk 💅✨

📦 where this page comes from