lookup_sign...y_partial_id(9)

πŸ“¦ version 7.3.1, in nixpkgs unstable

πŸ’… lookup_signaled_event_by_partial_id β€” from the Kernel API Manual, bestie

okay so this is a Kernel API Manual page and it's genuinely so useful, stay with me 🫢


✨ NAME ✨

lookup_signaled_event_by_partial_id β€” literally looks up a signaled event from a partial ID. that's the whole vibe πŸ’…


🌸 SYNOPSIS 🌸

struct kfd_event * lookup_signaled_event_by_partial_id (struct
kfd_process *p , uint32_t id , uint32_t bits , bool
signal_mailbox_updated );

πŸŽ€ ARGUMENTS πŸŽ€

  • p β€” pointer to struct kfd_process, the main character of this whole situation πŸ‘‘
  • id β€” the ID we're out here looking up πŸ”
  • bits β€” how many bits in id are actually valid, no cap
  • signal_mailbox_updated β€” a flag that says whether the FW updates the signal mailbox entry, iykyk πŸ¦‹

πŸ’– DESCRIPTION πŸ’–

so the way this works is: it finds the first signaled event that has a matching partial ID. slay. ✨

but if there's, like, no matching signaled event? it returns NULL. 😭 and in that case the caller should assume the partial ID is invalid and go do an exhaustive search of all signaled events β€” like a full grid search, bestie, no shortcuts.

and here's the spicy part 🌢️: if multiple events with the same partial ID signal at the same time, they get found one interrupt at a time, and not necessarily in the same order the interrupts actually happened. it's giving chaos but it's fine. as long as the number of interrupts is correct, all signaled events will be seen by the driver. periodt. πŸ’€


πŸ”₯ SEE ALSO πŸ”₯

  • Kernel file ./drivers/gpu/drm/amd/amdkfd/kfd_events.c β€” go read it, it ate πŸ’…

and that's the page!! brought to you by ./drivers/gpu/drm/amd/amdkfd, dated October 2026 β€” mwah, go touch some kernel code bestie πŸ§šπŸ’œ

πŸ“¦ where this page comes from