pagecache_isize_extended(9)

🎀 NAME 🎀

pagecache_isize_extended — update pagecache after extension of i_size 💖

Okay bestie, welcome to pagecache_isize_extended(9) from the Kernel API Manual ✨ This one is all about what happens to the pagecache when an inode’s size gets bigger.

---

✨ SYNOPSIS ✨

void pagecache_isize_extended (struct inode *inode , loff_t from ,
       loff_t to );

---

💅 ARGUMENTS 💅

• inode — the inode for which i_size was extended 👑
• from — the original inode size, aka where it was before 🌸
• to — the new inode size, aka the glow-up value 🦋

---

🌈 DESCRIPTION 🌈

So this function handles the extension of an inode’s size — either because of an extending truncate, or because of a write that starts after the current i_size 🫶 We mark the page straddling the current i_size as RO so that page_mkwrite gets called on the first write access to that page. That way, the filesystem will update its per-block information before the user writes to the page via mmap after the i_size has been changed 💖

⚠️ Timing check, bestie: this function must be called after i_size is updated, so that a page fault coming after we unlock the folio will already see the new i_size 😭 And it must be called while we still hold i_rwsem — that not only makes sure i_size is stable, but also makes sure userspace cannot observe the new i_size value before we are prepared to store mmap writes at the new inode size 💀

---

🔥 SEE ALSO 🔥

Kernel file ./mm/truncate.c 👀 and these iconic friends:

• folio_invalidate(9)
• mapping_evict_folio(9)
• truncate_inode_pages_range(9)
• truncate_inode_pages(9)
• truncate_inode_pages_final(9)
• mapping_try_invalidate(9)
• invalidate_mapping_pages(9)
• invalidate_inode_pages2_range(9)
• invalidate_inode_pages2(9)
• truncate_pagecache(9)
• truncate_setsize(9)
• truncate_pagecache_range(9)

---

Made with 💖 from source/version ./mm, dated October 2026, and this whole page is pagecache_isize_extended(9) — bye bestie, slay your pagecache 🫶✨