efa_ah_cache_put(9)

πŸ“¦ version 7.3.1, in nixpkgs unstable

πŸ’… efa_ah_cache_put β€” the "okay bestie, you can let go now" function

So this is a page from the Kernel API Manual, and honestly? It's a short one, but it's iconic. Let's get into it. 🫢


🌸 NAME

efa_ah_cache_put β€” put a refcount of an AH cache entry

Like, you're releasing your grip on an entry. Politely. With manners. ✨


πŸ¦‹ SYNOPSIS

void efa_ah_cache_put (struct efa_ah_cache *ah_cache , struct
efa_ah_cache_entry *entry );

Returns void, so don't be out here expecting a value back, bestie β€” it gives nothing and that's the point. πŸ’…


πŸŽ€ ARGUMENTS

  • ah_cache β€” the AH cache itself. The whole vibe container. πŸ‘‘
  • entry β€” the AH cache entry you're putting. The specific little guy in question. 🐣

πŸ’– DESCRIPTION

Okay so here's what's happening: you drop the refcount. That's it. That's the whole job. 😌

But then β€” and this is the main character moment β€” if that refcount reaches zero, the entry gets removed from the hashtable and freed. So the last one to leave turns off the lights. No cap. πŸ”₯

ngl this is the kind of function where the order of operations actually matters, so like… respect it. πŸ™ˆ

⭐ SEE ALSO

  • Kernel file ./drivers/infiniband/hw/efa/efa_ah_cache.c β€” go read the source, it's giving lore. πŸ“–
  • efa_ah_cache_get(9) β€” the counterpart, the "actually hold on to this one" function. They're a matching set, iykyk. πŸ’œ

That's the whole page, bestie β€” short, sweet, and lowkey doing a lot of heavy lifting with that free-on-zero behaviour. Slay. 🌈

This gem comes to you from ./drivers/infiniband/hw/efa, dated October 2026 β€” go off, EFA drivers. πŸŽ€βœ¨

πŸ“¦ where this page comes from