irdma_invalidate_pf_hmc_pd(9)

πŸ“¦ version 7.3.1, in nixpkgs unstable

πŸ’… irdma_invalidate_pf_hmc_pd(9) β€” from the Kernel API Manual, bestie

okay so picture this: it's a Kernel API Manual page, the star of the show is irdma_invalidate_pf_hmc_pd(9), and she is here to do exactly one thing. no cap. πŸŽ€


✨ NAME ✨

irdma_invalidate_pf_hmc_pd β€” invalidates the pd cache in the hardware for PF. that's it, that's the whole vibe. she invalidates, she leaves, she doesn't text back. πŸ’…


🌸 SYNOPSIS 🌸

void irdma_invalidate_pf_hmc_pd (struct irdma_sc_dev *dev , u32 sd_idx ,
       u32 pd_idx );

literally the entire signature, character for character, no crumbs left behind 🫢. returns void β€” she gives nothing back and honestly? iconic.


πŸ¦‹ ARGUMENTS πŸ¦‹

bestie, there are exactly three (3) things she needs from you, and she needs all of them:

  • dev β€” pointer to our device struct. yes our device struct. it's giving main character energy, it's the whole device, hand her the pointer πŸ”₯
  • sd_idx β€” segment descriptor index. the SD one. don't mix it up with the pd one, i'm begging πŸ™ˆ
  • pd_idx β€” page descriptor index. the PD one. a different number. a different vibe. keep them straight 😭

🧚 SEE ALSO 🧚

the whole friend group 🫢 β€” go say hi to the Kernel file at ./drivers/infiniband/hw/irdma/hmc.c, and then these absolute legends:

iykyk πŸ’œ


ngl the whole squad lives under ./drivers/infiniband/hw/irdma, so if you get lost just go home to that directory. 🏠

and that's the page!! 🌈 sourced straight from ./drivers/infiniband/hw/irdma, dated October 2026, and she signs off as irdma_invalidate_pf_hmc_pd(9) β€” periodt. πŸ’–βœ¨πŸŽ€

πŸ“¦ where this page comes from