gfs2_update_reply_times(9)

πŸ“¦ version 7.3.1, in nixpkgs unstable

πŸ’… gfs2_update_reply_times(9) β€” like, the Kernel API Manual era ✨

okay so bestie, welcome to gfs2_update_reply_times β€” the little function whose whole personality is Update locking statistics πŸŽ€ no cap, that's the vibe.


✨ SYNOPSIS ✨

void gfs2_update_reply_times (struct gfs2_glock *gl , bool blocking );

that's it, that's the whole signature, she's compact and she knows it πŸ’…


🫢 ARGUMENTS 🫢

  • gl β€” the glock we're updating, bestie. that's the one. the main character πŸ‘‘
  • blocking β€” a bool that says whether the operation may have been blocking πŸ™ˆ note the "may have been", she's not making promises

🌸 DESCRIPTION 🌸

okay so first thing's first: this function assumes that gl->gl_dstamp has been set earlier ⭐ like, that's a prerequisite, not a suggestion β€” she's not checking, she's just trusting you. iykyk 😭

now, the rtt (lock round trip time, and yes that's the acronym, keep up πŸ¦‹) is an estimate of the time taken to perform a dlm lock request. we update it on each reply from the dlm β€” every single reply, no skips, fr fr πŸ”₯

and then there's the blocking flag πŸ‘€ it gets set on the glock for all dlm requests which may potentially block due to lock requests from other nodes. so which ones count as non-blocking, you ask?? let me put you on:

  • DLM requests where the current lock state is exclusive
  • DLM requests where the requested state is null (or unlocked)
  • DLM requests where the TRY or TRY_1CB flags are set

those get classified as non-blocking ✨ and literally all other DLM requests are counted as (potentially) blocking β€” potentially, bestie, the parenthesis is doing so much work there πŸ’€


πŸ’œ SEE ALSO πŸ’œ


and that's the page!! πŸ“ this whole thing lives in ./fs/gfs2 and the version vibes are dated October 2026 β€” so like, future girlie energy, slay 🌈 periodt, go update those locking statistics bestie πŸ’–πŸ§š

πŸ“¦ where this page comes from