srpt_zerolength_write(9)

πŸ“¦ version 7.3.1, in nixpkgs unstable

πŸŽ€ srpt_zerolength_write(9) πŸŽ€

okay so real talk, bestie β€” this one's a page out of the Kernel API Manual, section 9, and it's short but she's iconic. it's giving "tiny function, huge spec energy". let's get into it ✨


🌸 NAME 🌸

srpt_zerolength_write β€” perform a zero-length RDMA write πŸ’…


πŸ¦‹ SYNOPSIS πŸ¦‹

int srpt_zerolength_write (struct srpt_rdma_ch *ch );

yeah, that's the whole thing. one function, one argument, no drama 🫢


πŸŽ€ ARGUMENTS πŸŽ€

  • ch β€” the SRPT RDMA channel. that's it, that's the entire parameter list. low maintenance queen πŸ’–

✨ DESCRIPTION ✨

so here's the vibe: this one comes with a direct quote from the InfiniBand specification, and we're just gonna let the spec speak for itself, no cap:

C9-88: For an HCA responder using Reliable Connection service, for each zero-length RDMA READ or WRITE request, the R_Key shall not be validated, even if the request includes Immediate data.

that's the whole description, bestie. fr fr the manpage is not gonna elaborate. it said what it said and it moved on 😭 periodt.


πŸ‘‘ SEE ALSO πŸ‘‘

first up, the kernel file itself, in case you want to go read the actual source like a true main character:

./drivers/infiniband/ulp/srpt/ib_srpt.c

and then, buckle up buttercup, the entire friend group πŸŽ‰ β€” every single one of these:

like I said. the whole group chat. add them to the study playlist, they all ate πŸ’€πŸ”₯


and that's the page!! thank you for coming to my TED talk πŸ§šβ€β™€οΈ this little gem lives down in ./drivers/infiniband/ulp/srpt and it's dated October 2026, and yes it's still srpt_zerolength_write(9) all the way to the end because the title card never changes, bestie πŸ’œ byeee πŸ“

πŸ“¦ where this page comes from