swap_vma_readahead(9)

📦 version 7.3.1, in nixpkgs unstable

💅 swap_vma_readahead — kernel API manual, bestie edition

okay so welcome to swap_vma_readahead(9), the lil function that goes "hmm what if we need these pages soon" and just… grabs them. iconic behavior tbh. 🦋

✨ NAME ✨

swap_vma_readahead — swap in pages in hope we need them soon 💖

like, manifesting pages into memory. the audacity. the vision. 🔮

🌸 SYNOPSIS 🌸

struct folio * swap_vma_readahead (swp_entry_t targ_entry , gfp_t
gfp_mask , struct mempolicy *mpol , pgoff_t targ_ilx , struct vm_fault
*vmf );

she returns a struct folio *, she takes five whole arguments, she's booked and busy 💅

🎀 ARGUMENTS 🎀

  • targ_entry — the swap entry of the targeted memory. like, the actual thing we're going after 🎯
  • gfp_mask — memory allocation flags. the vibes for how we allocate ✨
  • mpol — the NUMA memory allocation policy to be applied. NUMA said "let me handle this" 👑
  • targ_ilx — the NUMA interleave index, and heads up bestie, it's for use only when MPOL_INTERLEAVE — don't be using it otherwise 🙈
  • vmf — fault information. the lore. the context. 📋

🦋 DESCRIPTION 🦋

So what she does: she returns the struct folio for entry and addr, after queueing swapin. ✨

This is the primitive swap readahead code, no cap. We simply read in a few pages whose virtual addresses are around the fault address in the same vma. that's it. that's the whole bit. surrounding pages, same vma, grab a few, hope for the best 🫶

heads up bestie 💀 — the caller must hold read mmap_lock if vmf->vma is not NULL. like, mandatory. non-negotiable. do not skip this step fr fr.

💜 SEE ALSO 💜

kernel file ./mm/swap_state.c and honestly the whole friend group:

🌸 okay that's the whole page, ily, go read some swap cache code 💅

sign-off era: this one lives in ./mm, dated October 2026, and it's swap_vma_readahead(9) — that's a wrap bestie 🎀✨🫶

📦 where this page comes from