napi_skb_cache_get_bulk(9)

📦 version 7.3.1, in nixpkgs unstable

💖 napi_skb_cache_get_bulk(9) 💖

okay so bestie, this is a page straight out of the Kernel API Manual (yes, that Kernel API Manual, the one we all live in) and it's all about grabbing a whole little squad of zeroed skb heads out of the cache. literally so iconic of her. 🎀

✨ NAME ✨

napi_skb_cache_get_bulk — obtain a number of zeroed skb heads from the cache 🦋

🌸 SYNOPSIS 🌸

u32 napi_skb_cache_get_bulk (void **skbs , u32 n );

💅 ARGUMENTS 💅

  • skbs — pointer to an array that's at least n-sized, and you're gonna fill it up with skb pointers. like, she needs room to work, give her the room. 👑
  • n — the number of entries to provide, i.e. how many you're asking for. no cap, this is just the count.

🌈 DESCRIPTION 🌈

So what she does is: she tries to obtain n sk_buff entries from the NAPI percpu cache and writes those pointers into the array you handed her, skbs. ✨

If there are less entries available than you asked for, she tries to replenish the cache and bulk-allocates the diff from the MM layer if needed. the way she just handles it. 🔥

Now this next part is the main character detail: the heads are being zeroed with either memset or __GFP_ZERO, so they're ready for {,__}build_skb_around and they don't have any data buffers attached. clean slate, fresh vibe, zero baggage. 🫶

⚠️ big warning, bestie — this must be called only from the BH context. like, only. don't get cute about it. 💀

⭐ RETURN ⭐

Returns the number of successfully allocated skbs — that's n if no actual allocation was needed, or if kmem_cache_alloc_bulk didn't fail. slay. 💜

🧚 SEE ALSO 🧚

Kernel file ./net/core/skbuff.c, plus this whole extended friend group:

and that's the whole page, bestie 🍓 — brought to you by ./net/core, dated October 2026, and filed under the manual name napi_skb_cache_get_bulk(9). go forth and allocate responsibly 💖✨

📦 where this page comes from