pdc_irq_handler(9)

📦 version 7.3.1, in nixpkgs unstable

💅 pdc_irq_handler(9) — bestie edition 🎀

okay so this is a page from the Kernel API Manual, and honestly? it's a little one but it's important. sit down, grab a boba, we're talking about the interrupt handler that gets called in interrupt context. yes, the spicy context. 🔥

✨ NAME ✨

pdc_irq_handler — the interrupt handler that gets called in interrupt context. that's it. that's the whole vibe. 🫶

🌸 SYNOPSIS 🌸

irqreturn_t pdc_irq_handler (int irq , void *data );

look at her. two args, one return type. iconic behavior only. 👑

🦋 ARGUMENTS 🦋

  • irq — the interrupt number that has fired. like, the one that actually went off. no cap, that's the whole thing. ✨
  • data — the device struct for the DMA engine that generated the interrupt. so basically the "who did this" of the situation 💀

🎀 DESCRIPTION 🎀

okay so here's what we have to do: we gotta clear the device interrupt status flags right here in this function. no skipping it, bestie, it's literally the assignment. 💅

so we cache the status for later use in the thread function — yes, cache it, because the thread function is gonna need it and we're being responsible about it. main character energy is being prepared. 🌈

other than that? that's it. that's the whole job. just return WAKE_THREAD to invoke the thread function. slay, done, periodt. 🧚

⭐ RETURN ⭐

  • IRQ_WAKE_THREAD — if the interrupt is ours 🫶 (yay, it's ours, it's literally ours)
  • IRQ_NONE — otherwise 🙈 (not ours, not our problem)

💜 SEE ALSO 💜

the kernel file lives at ./drivers/mailbox/bcm-pdc-mailbox.c — go look at her, she's the source of truth, iykyk.

and then the whole friend group, no one left out, fr fr:

and that's the whole page, bestie 💖 this little baddie ships out of ./drivers/mailbox and the version on the tin says October 2026 — so like, future vibes. go forth and handle those interrupts responsibly. mwah 🫶✨

📦 where this page comes from