svc_rdma_post_send(9)
π NAME π
okay so this is the Kernel API Manual page for svc_rdma_post_send(9), and it is literally here to Post a WR chain to the Send Queue. That's the whole vibe: you got a WR chain, you got a Send Queue, and this function is the matchmaker. π
β¨ SYNOPSIS β¨
int svc_rdma_post_send (struct svcxprt_rdma *rdma , struct
svc_rdma_send_ctxt *ctxt );
π ARGUMENTS π
rdmaβ transport contextctxtβ WR chain to post
πΈ DESCRIPTION πΈ
So what's the tea? svc_rdma_post_send copies the fields in ctxt into stack variables. Why? To guarantee that those values remain available after the ib_post_send call. That's the protective bestie behavior we love to see. π«Ά
Heads up, bestie: in some error flow cases,svc_rdma_wc_sendreleasesctxt. That's not a drill, that's just the documented behavior.
Return values, spelled out exactly:
0:ctxt's WR chain was posted successfully-ENOTCONN: The connection was lost
No cap, those are the only return values listed here. β
π¦ SEE ALSO π¦
- Kernel file
./net/sunrpc/xprtrdma/svc_rdma_sendto.c svc_rdma_send_ctxts_destroy(9)svc_rdma_send_ctxt_get(9)svc_rdma_send_ctxts_drain(9)svc_rdma_send_ctxt_put(9)svc_rdma_wake_send_waiters(9)svc_rdma_sq_wait(9)svc_rdma_post_send_err(9)svc_rdma_wc_send(9)svc_rdma_encode_read_list(9)svc_rdma_encode_write_segment(9)svc_rdma_encode_write_chunk(9)svc_rdma_encode_write_list(9)svc_rdma_encode_reply_chunk(9)svc_rdma_page_dma_map(9)svc_rdma_iov_dma_map(9)svc_rdma_xb_dma_map(9)svc_rdma_xb_count_sges(9)svc_rdma_check_pull_up(9)svc_rdma_xb_linearize(9)svc_rdma_pull_up_reply_msg(9)svc_rdma_send_error_msg(9)svc_rdma_sendto(9)svc_rdma_result_payload(9)
xoxo, your svc_rdma_post_send(9) girlie π
β sourced from ./net/sunrpc/xprtrdma, dated October 2026, version tag svc_rdma_post_send(9).
π¦ where this page comes from
- nixpkgs unstable: linux-manual 7.3.1
- π girlified by deepseek-ai/DeepSeek-V4.1-Flash-1017 on : a language model's rewrite of the original, under its terms