filemap_flush(9)

πŸ“¦ version 7.3.1, in nixpkgs unstable

πŸ’… filemap_flush β€” like, the mostly non-blocking flush bestie

okay so this is filemap_flush(9) from the Kernel API Manual, section 9 (kernel API, the fun one). Buckle up πŸŽ€


✨ NAME ✨

filemap_flush β€” mostly a non-blocking flush 🌸

like yes it flushes, no it does not fully commit, it's giving "I'll get to it" energy 🫢


✨ SYNOPSIS ✨

int filemap_flush (struct address_space *mapping );

✨ ARGUMENTS ✨

  • mapping β€” the target address_space, aka the thing you're flushing. that's it, that's the whole argument πŸ’…

✨ DESCRIPTION ✨

So this is a mostly non-blocking flush. Note the mostly β€” that word is doing a LOT of heavy lifting here, bestie 😭

It is not suitable for data-integrity purposes, and I need you to say that back to me: the I/O may not be started against all dirty pages. So if you were about to use this to guarantee your data actually made it to disk… no. do not. that is not the vibe. 🚫

⚠️ real talk: if data integrity is your whole personality, this is not the function for you. it's the "fire and sort-of-forget" one. periodt.

✨ RETURN ✨

  • 0 on success πŸŽ‰
  • negative error code otherwise πŸ’€

✨ SEE ALSO ✨

and because the kernel is a group chat and nobody gets left on read, here's the whole squad πŸ‘‘ the kernel file ./mm/filemap.c plus:


and that's the page, bestie πŸ’– this cuteness was served up by the source directory ./mm (yep, the ./mm one) and it's dated October 2026 β€” so when you cite it, it's filemap_flush(9), ./mm, October 2026. slay πŸ¦‹βœ¨

πŸ“¦ where this page comes from