trace_gpu_job_irq(9)

Okay bestie, welcome to trace_gpu_job_irq(9) from the Kernel API Manual! ✨ This page is about the tracepoint that’s called after a job interrupt from firmware completes. Buckle up, it’s short and iconic 💅

🎀 NAME 🎀

trace_gpu_job_irq — called after a job interrupt from firmware completes 🎉

✨ SYNOPSIS ✨

void trace_gpu_job_irq (const struct device *dev , u32 events , u32 duration_ns );

🌸 ARGUMENTS 🌸

• dev — pointer to the struct device, for printing the device name 💻
• events — bitmask of BIT(CSG id) | BIT(31) for a global event 🎛️
• duration_ns — Nanoseconds between job IRQ handler entry and exit ⏱️

💜 DESCRIPTION 💜

Okay so here’s the timing tea ☕: the panthor_job_irq_handler function instrumented by this tracepoint exits once it has queued the firmware interrupts for processing — not when the firmware interrupts are fully processed. Like, it is not waiting around for the whole glow-up to finish, bestie 😭

⚠️ Important vibe check: exits once it has queued the firmware interrupts for processing, not when the firmware interrupts are fully processed.

This tracepoint allows for debugging issues with delays in the workqueue's processing of events. No cap, that’s the whole point 🫶

🦋 SEE ALSO 🦋

• Kernel file ./drivers/gpu/drm/panthor/panthor_trace.h 📁

That’s the whole page, bestie! 🫶 Source/version: ./drivers/gpu/drm/panthor. Date: October 2026. Page energy: trace_gpu_job_irq(9). Stay slay, bye 💅✨