gfs2_flush_revokes(9)

πŸ“¦ version 7.3.1, in nixpkgs unstable

πŸ’… gfs2_flush_revokes β€” the kernel API manual, but make it make sense

okay so bestie, you're reading the page for gfs2_flush_revokes(9) from the Kernel API Manual, and honestly? she's a little baddie with a big job. let's get into it πŸŽ€


✨ NAME ✨

gfs2_flush_revokes β€” add as many revokes to the system transaction as we can 🫢


🌸 SYNOPSIS 🌸

void gfs2_flush_revokes (struct gfs2_sbd *sdp );

like, that's it. that's the whole call. no return value, no drama. πŸ’…


πŸ¦‹ ARGUMENTS πŸ¦‹

  • sdp β€” the GFS2 superblock πŸ‘‘ (yes, the superblock, main character energy fr fr)

πŸ’– DESCRIPTION πŸ’–

okay so here's the tea β˜• β€” our usual strategy is to defer writing revokes as much as we can, because we're lowkey manifesting that we'll eventually overwrite the journal, and that would make those revokes just… go away ✨ iconique behaviour, no notes.

but that changes when we flush the log. at that point, there will likely be some left-over space in the last revoke block of that transaction πŸ‘€ and we are not about to waste it.

so what do we do? we fill that space with additional revokes for blocks that have already been written back πŸ§šβ€β™€οΈ basically it's free real estate β€” comes at no cost now β€” and it saves us from having to keep track of those blocks on the AIL2 list later. ate and left no crumbs, honestly πŸ”₯

ngl this one's a "why didn't we always do this" moment and i'm obsessed πŸ’€

πŸ‘‘ SEE ALSO πŸ‘‘

the whole friend group, in the order the page says it 🫢:

Kernel file ./fs/gfs2/log.c, plus:


and that's the whole page, bestie 🌈 this cutie lives in ./fs/gfs2 and the version/date situation is October 2026 β€” so yeah, she's from the future, slay πŸ’œβœ¨

πŸ“¦ where this page comes from