global_dirtyable_memory(9)

💅 global_dirtyable_memory(9) — from the Kernel API Manual, bestie ✨

Okay so picture this: you're deep in the kernel, the Kernel API Manual is open on your other monitor, and you need to know how many pages the whole system could hypothetically let go dirty. That's the whole gig. This little function is the main character of page-writeback bookkeeping, and honestly? It ate. 🫶

---

🌸 NAME 🌸

global_dirtyable_memory — the number of globally dirtyable pages, periodt 💅

---

✨ SYNOPSIS ✨

unsigned long global_dirtyable_memory (void );

---

🎀 ARGUMENTS 🎀

• void — no arguments, babe. Zero. None. It takes nothing and it gives everything. 🦋

---

👑 RETURN 👑

It hands you back the global number of pages potentially available for dirty page cache. Read that again: potentially. It's the vibe-check number, not a promise.

Lowkey the important bit: this is the base value for the global dirty limits. So everything downstream keys off this. It's giving "foundation of the whole situation" energy. 🔥

Ngl, the word "potentially" is doing a lot of heavy lifting in that sentence, so don't come for this function when your page cache math looks off. 🙈

---

💖 SEE ALSO 💖

Okay bestie, buckle up, because the friend group here is large. Kernel file ./mm/page-writeback.c, plus the whole squad:

• node_dirtyable_memory(9)
• domain_dirty_limits(9)
• global_dirty_limits(9)
• node_dirty_limit(9)
• node_dirty_ok(9)
• __wb_calc_thresh(9)
• balance_dirty_pages_ratelimited_flags(9)
• balance_dirty_pages_ratelimited(9)
• wb_over_bg_thresh(9)
• tag_pages_for_writeback(9)
• writeback_iter(9)
• filemap_dirty_folio(9)
• folio_redirty_for_writepage(9)
• folio_mark_dirty(9)
• folio_wait_writeback(9)
• folio_wait_writeback_killable(9)
• folio_wait_stable(9)

Like, that's a full group chat right there. Go say hi. 💜

---

That's the whole page, no cap 💫 — straight out of ./mm, dated October 2026, and yes it's still called global_dirtyable_memory(9) the whole way through. Mwah! 🍓🧚💅