svc_rdma_cma_handler(9)
π svc_rdma_cma_handler(9) β like, the Kernel API Manual one, bestie π
okay so picture this: you're deep in the kernel, someone's connecting over RDMA, and the connection manager is out here throwing events at you. this little function is the one catching them on client connections. it's giving main character energy in svc_rdma_transport.c and honestly? we stan. β¨
πΈ SYNOPSIS πΈ
int svc_rdma_cma_handler (struct rdma_cm_id *cma_id , struct
rdma_cm_event *event );
no cap, that's the whole signature. copy it exactly, don't get creative with it π«Ά
π¦ ARGUMENTS π¦
cma_idβ the server's listenerrdma_cm_id. yes, the listener's. that's the whole vibe. πeventβ the details of the event, aka the actual tea of what just happened β
π DESCRIPTION π
so what does it do? the return value is the entire personality here, so let's get into it:
0β do not destroycma_id. keep it alive, it's thriving, let it cook π₯1β destroycma_id. bye bestie π
and listen β that 1 is never returned here. like, the code literally doesn't do that. don't come for me, that's just what the page says π
periodt. that's the contract. lowkey iconic for how short it is.
β SEE ALSO β
- the kernel file
./net/sunrpc/xprtrdma/svc_rdma_transport.cβ that's where the magic lives, iykyk π svc_rdma_xprt_deferred_close(9)svc_rdma_listen_handler(9)svc_rdma_free(9)
go read those, they're all in the group chat π
π§ sign-off π§
this whole masterpiece comes to you from ./net/sunrpc/xprtrdma, dated October 2026 β so yes, it's from the future, and it's still slaying. stream svc_rdma_cma_handler(9) ππΈπ
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