gaudi_handl..._pqes_on_err(9)

πŸ“¦ version 7.3.1, in nixpkgs unstable

πŸ’… gaudi_handl..._pqes_on_err(9) β€” Kernel API Manual

okay so bestie, welcome to the kernel API manual for gaudi_handle_last_pqes_on_err β€” yes the title gets truncated in the header, that's just how it vibes, the full name is right down there in NAME. iykyk πŸŽ€


✨ NAME ✨

gaudi_handle_last_pqes_on_err β€” literally just print the last PQEs when an error happens. that's the whole job. iconic, honestly πŸ¦‹


🌸 SYNOPSIS 🌸

void gaudi_handle_last_pqes_on_err (struct hl_device *hdev , u32
qid_base , u32 stream , u64 qman_base , u64 event_mask , bool pr_sw_conf
);

no cap, that's the whole signature. copy it exactly, don't be cute with it πŸ’–


🧚 ARGUMENTS 🧚

okay so here's the cast of characters, main character energy only:

  • hdev β€” pointer to the habanalabs device structure. the device. the moment. the whole context, fr.
  • qid_base β€” first QID of the QMAN (out of 4 streams). yes, out of 4 streams, keep that detail, it matters πŸ’…
  • stream β€” the QMAN's stream. just the one. the stream in question.
  • qman_base β€” base address of the QMAN registers block. where the registers live 🏠
  • event_mask β€” mask of the last events that occurred. like, what just happened, but as a mask ✨
  • pr_sw_conf β€” if true, print the SW config stream data (CQ PTR and SIZE). if it's false, it simply does not. obsessed with a boolean that tells the truth 🫢

πŸ”₯ SEE ALSO πŸ”₯

go read those next bestie, they're all right there in the same little gaudi family 🌈


πŸ’€ friendly reminder that this page is literally just the function docs β€” no extra lore, no secret behaviour, nothing invented. what you see is what you get, fr fr.

and that's the whole page! slay πŸ’œ this came from the ./drivers/accel/habanalabs/gaudi source, dated October 2026 β€” love that for us, mwah πŸ“πŸ‘‘

πŸ“¦ where this page comes from