filemap_fda..._keep_errors(9)

📦 version 7.3.1, in nixpkgs unstable

hey bestie, welcome to the filemap_fda..._keep_errors(9) page from the Kernel API Manual 💖 this one is lowkey iconic if you're into writeback waiting, and yes we're gonna get through every single detail together ✨

🌸 NAME 🌸

filemap_fdatawait_range_keep_errors — wait for writeback to complete 🎀

✨ SYNOPSIS ✨

here's the exact signature, bestie — copy it character for character, no edits:

int filemap_fdatawait_range_keep_errors (struct address_space *mapping , loff_t start_byte , loff_t end_byte );

🎀 ARGUMENTS 🎀

  • mapping — the address space structure to wait for. like, this is the thing you're waiting on until writeback is done 💅
  • start_byte — offset in bytes where the range starts. beginning of the wait zone ✨
  • end_byte — offset in bytes where the range ends, and yes bestie it is inclusive 🫶 don't forget that part

💅 DESCRIPTION 💅

okay so what it actually does: it walks the list of under-writeback pages of the given address space in the given range, and waits for all of them. all of them, no cap 🔥

the key difference from filemap_fdatawait_range is that this function does not clear the error status of the address space. that's the whole “keep errors” moment, iykyk 😭

use this function if callers don't handle errors themselves. expected call sites are system-wide / filesystem-wide data flushers: e.g. sync(2), fsfreeze(8) — iconic flusher behavior only ✨

heads up bestie: it keeps the error status, so don't treat it like the version that clears it 💀

🦋 SEE ALSO 🦋

kernel file: ./mm/filemap.c

the whole reference squad, in order:

okay that's the whole page, bestie 💖 source/version says ./mm, date says October 2026, and the page is filemap_fda..._keep_errors(9) — go wait on those pages and slay ✨🦋

📦 where this page comes from