nfs_request_..._commit_list(9)

πŸ“¦ version 7.3.1, in nixpkgs unstable

πŸ’… nfs_request_..._commit_list(9) πŸ’…

okay bestie so this is a page from the Kernel API Manual and it's about one tiny little function that does one tiny little thing, and honestly? iconic of it. her whole job is un-committing a request. that's it. that's the vibe. πŸŽ€


✨ NAME ✨

nfs_request_remove_commit_list β€” removes a request from a commit list. she's a remover. a list-leaver. a "you don't have to go home but you can't stay here" girlie. πŸ’…


🌸 SYNOPSIS 🌸

void nfs_request_remove_commit_list (struct nfs_page *req , struct
       nfs_commit_info *cinfo );

like, she returns void because she's not here to give you answers, she's here to do the thing. no cap. ✨


πŸ¦‹ ARGUMENTS πŸ¦‹

  • req β€” a pointer to a nfs_page. the actual request in question, the main character of this whole situation 🫢
  • cinfo β€” holds the list lock and the accounting info. she's got the receipts, the lock, and the numbers. the responsible friend. πŸ’–

πŸ”₯ DESCRIPTION πŸ”₯

so the way this works is: she clears the PG_CLEAN bit β€” yes, literally un-clean-ing it β€” and then updates cinfo's count of outstanding requests that require a commit. so the accounting stays honest. very demure, very mindful. 🧚

that said, she does not update the MM page stats. not her department. not her problem. don't come for her about the MM page stats. πŸ’€

⚠️ ok this part is serious so listen up bestie: the caller _must_ hold the cinfo->lock and the nfs_page lock. not "should". not "probably". _must_. no lockies = no touchy. fr fr. πŸ”’

🌈 SEE ALSO 🌈

if you're obsessed with this one (and why wouldn't you be), the kernel file is ./fs/nfs/write.c and the whole friend group is:

go read them. be a well-rounded kernel girlie. πŸ‘‘


and that's the whole page, bestie, no crumbs left behind. this little gem lives in ./fs/nfs and it's dated October 2026, served up fresh from the Kernel API Manual. mwah, go hold those locks πŸ’œπŸ’…βœ¨

πŸ“¦ where this page comes from