svc_rdma_pull_up_reply_msg(9)

πŸ“¦ version 7.3.1, in nixpkgs unstable

πŸŽ€ NAME πŸŽ€

okay so this is the svc_rdma_pull_up_reply_msg(9) Kernel API Manual page, and the name is svc_rdma_pull_up_reply_msg β€” Copy Reply into a single buffer. That's the whole name section and it's lowkey iconic. πŸ’…

✨ SYNOPSIS ✨

       int svc_rdma_pull_up_reply_msg (const struct svcxprt_rdma *rdma , struct
       svc_rdma_send_ctxt *sctxt , const struct svc_rdma_pcl *write_pcl , const
       struct xdr_buf *xdr );

🫢 ARGUMENTS 🫢

  • rdma β€” the controlling transport. it's giving boss of the RDMA situation.
  • sctxt β€” send_ctxt for the Send WR; xprt hdr is already prepared. so don't worry about that part.
  • write_pcl β€” Write chunk list provided by client. ty bestie client.
  • xdr β€” prepared xdr_buf containing RPC message.

πŸ’… DESCRIPTION πŸ’…

The device is not capable of sending the reply directly. like, it literally cannot. So we assemble the elements of xdr into the transport header buffer.

Assumptions: check_pull_up has determined that xdr will fit in the buffer. no cap, that check is doing the pre-work so this function can vibe.

🌈 RETURN 🌈

  • 0 β€” if pull-up was successful. slay. ✨
  • -EMSGSIZE β€” if a buffer manipulation problem occurred. 😭

πŸ”₯ SEE ALSO πŸ”₯

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


okay byeee πŸ’– β€” source/version: ./net/sunrpc/xprtrdma, date: October 2026, page: svc_rdma_pull_up_reply_msg(9). mwah 🫢

πŸ“¦ where this page comes from