GPU HEALTH INDICATOR(9)

💖 GPU HEALTH INDICATOR (9) 💖

ok so real talk, this one lives in section 9 of the Kernel API Manual, and it's genuinely such a cute little interface bestie. no cap.

---

✨ DESCRIPTION ✨

okay so picture this: you're on an Intel Xe platform that supports the gpu health indicator interface, and the driver is just standing there, holding up this sysfs attribute so you can get in-band access to the gpu health state. iconic behavior honestly:

/sys/bus/pci/devices/<device>/gpu_health

🫶 Reading it is available to all users — literally everyone, come one come all — and it hands you back a single line containing the current gpu health state. so simple, so clean, we love her.

✋ Writing it though? that's restricted to administrative users (sorry bestie, rules are rules), and writing updates the state to one of the valid values.

Management tools and administrators vibe with this interface to do two things:

• query the current gpu health state (e.g. for telemetry/monitoring — very main character surveillance energy 👀)
• update it — for example, to mark the gpu as warning or critical after diagnostics, or reset it back to ok once it's been remediated 🎀

---

🌈 the valid values for the gpu health state 🌈

• ok — the gpu is healthy and operating within normal parameters. she's thriving. slay. 💅
• warning — the gpu is experiencing minor issues but remains operational. lowkey going through it but still showing up. 🫶
• critical — the gpu is in a critical state and may not be operational. 🚨 not me, not her, oh no 😭

heads up bestie: for the full ABI specification, go check out Documentation/ABI/testing/sysfs-driver-intel-xe-ras. that's the receipts. iykyk. 📄

---

👀 SEE ALSO 👀

• Kernel file ./drivers/gpu/drm/xe/xe_ras.c
• xe_ras_process_errors(9)
• xe_ras_get_counter(9)
• xe_ras_clear_counter(9)
• xe_ras_init(9)

---

ok that's the whole page, no notes 💜

this little beauty ships out of ./drivers/gpu/drm/xe, dated October 2026, and she's titled GPU Health Indicator(9) — periodt. 🦋✨