gc_sync_wbufs(9)

💅 gc_sync_wbufs — from the gc_sync_wbufs(9) kernel API manual, bestie

okay so picture this: it's a kernel API manual page, it's section 9, and it lives in ./fs/ubifs. and the star of the show is gc_sync_wbufs. buckle up 🎀

---

✨ NAME ✨

gc_sync_wbufs — sync write-buffers for GC. that's it, that's the whole vibe. literally just syncing write-buffers for GC 🦋

---

🌸 SYNOPSIS 🌸

int gc_sync_wbufs (struct ubifs_info *c );

---

🫶 ARGUMENTS 🫶

• c — the UBIFS file-system description object. that's your girl, she's carrying all the info ✨

---

💖 DESCRIPTION 💖

okay so here's the situation, and it's lowkey serious. we must guarantee that obsoleting nodes are on flash. that's non-negotiable. but unfortunately 😭 they may be hanging out in a write-buffer instead. rude of them honestly.

like, picture a node getting written to a write-buffer, and it's obsoleting another node that lives in a LEB that's getting GC'd. the way that's a whole problem is: if that LEB gets erased before the write-buffer is sync'd, and then there's an unclean unmount? then an existing node is just... lost. gone. not the vibe. 💀

so to avoid alllll of that drama, we sync all write-buffers. every single one. no cap 🙈

this function returns 0 on success, or a negative error code on failure. simple as that, periodt 💅

bestie the stakes are real here — lost nodes after an unclean unmount are not it. sync first, always. 🔥

---

👑 SEE ALSO 👑

kernel file ./fs/ubifs/gc.c, plus all these iconic friends:

• switch_gc_head(9)
• data_nodes_cmp(9)
• sort_nodes(9)
• move_node(9)
• move_nodes(9)
• ubifs_garbage_collect_leb(9)
• ubifs_garbage_collect(9)
• ubifs_gc_start_commit(9)
• ubifs_gc_end_commit(9)
• ubifs_destroy_idx_gc(9)
• ubifs_get_idx_gc_leb(9)

---

✨ okay that's the whole page, mwah 💋 this cutie came from ./fs/ubifs and it's dated October 2026 — go off, calendar icon 🗓️💜