apply_link_downgrade_policy(9)

📦 version 7.3.1, in nixpkgs unstable

okay so bestie, we're in the apply_link_downgrade_policy(9) page of the Kernel API Manual, and honestly? this one's lowkey a vibe check for your link widths. no cap, let's get into it 🫶

✨ NAME ✨

apply_link_downgrade_policy — literally applies the link width downgrade enabled policy against the current active link widths. that's it, that's the whole thing 💅

🌸 SYNOPSIS 🌸

bool apply_link_downgrade_policy (struct hfi1_pportdata *ppd , bool
refresh_widths );

🎀 ARGUMENTS 🎀

  • ppd — the info of the physical Hfi port. like, the whole personality of the port goes in here 💖
  • refresh_widths — True means a link downgrade event happened. it's giving "yes queen, something changed" ✨

🦋 RETURN 🦋

  • True — slay, the link downgrade worked 🎉
  • False — nope, the link downgrade event failed, and the link will bounce back to default link width. it's giving "we're so back", but like, sadly 😭

👑 DESCRIPTION 👑

so this gets called whenever the enabled policy changes or the active link widths change — either one, she's on it 🌈

refresh_widths is the one that says "hey, a link downgrade occurred." and then the link_downgraded variable gets set by refresh_widths, and that is what determines the success/failure of the policy application. so like, refresh_widths is the main character here, fr fr 👑

no big dramatic warnings on this one, bestie — just know that the return value is the whole story: True = the downgrade stuck, False = the link bounces back to default link width. 🫶

💜 SEE ALSO 💜

the kernel file lives right here, iconic:

./drivers/infiniband/hw/hfi1/chip.c

and the whole friend group 🫂:

okay that's the whole page, you ate that 💅 this lil gem comes to you straight from ./drivers/infiniband/hw/hfi1, dated October 2026 — apply_link_downgrade_policy(9), periodt. 🎀💖✨

📦 where this page comes from