_rvt_reset_qp(9)

πŸ“¦ version 7.3.1, in nixpkgs unstable

πŸ’… _rvt_reset_qp β€” the Kernel API Manual glow-up πŸŽ€

okay so bestie, welcome to the _rvt_reset_qp(9) page from the Kernel API Manual β€” yes, that one. she's short, she's specific, and she's lowkey a control freak about locks. let's get into it ✨


🌸 NAME 🌸

_rvt_reset_qp β€” set the QP state back to the reset state. like a factory reset but for a queue pair, no cap πŸ’…


πŸ¦‹ SYNOPSIS πŸ¦‹

void _rvt_reset_qp (struct rvt_dev_info *rdi , struct rvt_qp *qp , enum
ib_qp_type type );

that's the whole signature, bestie β€” copy it exactly, don't get creative πŸ™ˆ


⭐ ARGUMENTS ⭐

  • rdi β€” the rvt dev struct, aka the device info. she's the context, she's the main character πŸ‘‘
  • qp β€” the QP you're resetting. literally the thing getting the reset treatment 🫢
  • type β€” the QP type. yes, there are types. iykyk 🌈

πŸ”₯ DESCRIPTION πŸ”₯

okay so here's the one (1) sentence of lore and it's kind of a big deal:

r_lock, s_hlock, and s_lock are required to be held by the caller

that's it, that's the description. no cap. if you're calling this without all three of those locks held, that's on you and it's giving "undefined consequences" energy πŸ’€ so like… hold the locks. all three. periodt.


πŸŽ€ SEE ALSO πŸŽ€

the friend group! all of these live in the Kernel file ./drivers/infiniband/sw/rdmavt/qp.c and honestly they're all obsessed with each other:


and that's the page, bestie! 🧚 this little gem comes from ./drivers/infiniband/sw/rdmavt and it's dated October 2026 β€” so like, future vibes. stay locked, stay reset πŸ’œπŸ’–βœ¨

πŸ“¦ where this page comes from