ext4_free_branches(9)

📦 version 7.3.1, in nixpkgs unstable

✨ ext4_free_branches ✨

okay so this is a page from the Kernel API Manual, bestie — it's ext4_free_branches(9) and it does exactly what it says on the tin: it frees an array of branches 🎀

🦋 SYNOPSIS 🦋

void ext4_free_branches (handle_t *handle , struct inode *inode , struct buffer_head *parent_bh , __le32 *first , __le32 *last , int depth );

literally that's the whole signature, no cap 💅

🌸 ARGUMENTS 🌸

  • handle — the JBD handle for this transaction. your transaction's little passport, basically 🫶
  • inode — the inode we are dealing with. the main character of this whole operation 👑
  • parent_bh — the buffer_head which contains *first and *last. it's the container holding your array pointers, okurrr
  • first — array of block numbers. the starting point bestie 💖
  • last — pointer immediately past the end of array. yes, immediately past — classic half-open range vibe, we love a [first, last) moment ✨
  • depth — depth of the branches to free. how deep are we going? 🌈

💜 DESCRIPTION 💜

so here's the tea 🍓 — we are freeing all blocks referred from these branches (and yes, the numbers are stored as little-endian 32-bit, that detail matters, don't @ me), and we're updating inode->i_blocks appropriately while we're at it.

like, the blocks go poof and the block count stays honest. it's giving responsible filesystem citizen 🧚

🔥 SEE ALSO 🔥

the whole ext4 indirect-block friend group honestly, iykyk 😭💀

ngl this one's short but she is load-bearing. read the SEE ALSO list before you go poking at branches, bestie 🙈

okay bye ily, this page lives in ./fs/ext4 and the version stamp says October 2026, signed off as ext4_free_branches(9) — slay, see u next time 🎀✨

📦 where this page comes from