XE DEVICE COREDUMP(9)

💖 Xe device coredump (9) 💖

okay so bestie, this one's from the Kernel API Manual, section 9 — it's the page about Xe device coredump, and honestly it's kinda iconic once you get what it's doing. Let's get into it 🫶

---

✨ DESCRIPTION ✨

Xe uses the dev_coredump infrastructure to expose crash errors in a standardized way. no cap, that's the whole vibe — crashes, but organized.

Once a crash occurs, devcoredump exposes a temporary node under:

/sys/class/devcoredump/devcd<m>/

The same node is also accessible in:

/sys/class/drm/card<n>/device/devcoredump/

and the failing_device symlink points to the device that crashed and created the coredump. so you can literally just follow the link to find the culprit 💅

---

Xe observes the following characteristics when creating a device coredump:

🦋 Snapshot at hang

The 'data' file contains a snapshot of the HW and driver states at the time the hang happened. but lowkey important: because the driver is recovering from resets/crashes, it may not correspond to the state of the system when the file is read by userspace. so like… it's a snapshot of the moment, not a livestream 😭

🎀 Coredump release

After a coredump is generated, it stays in kernel memory until it's released by userspace — you do that by writing anything to it — or after an internal timer expires. the exact timeout may vary and should not be relied upon. fr fr, don't build your whole plan around that timer.

Example to release a coredump:

$ > /sys/class/drm/card0/device/devcoredump/data

that's it, that's the whole ritual ✨

👑 First failure only

In general, the first hang is the most critical one, since the following hangs can be a consequence of the initial hang. main character energy, honestly. for this reason a snapshot is taken only for the first failure.

Until the devcoredump is released by userspace or kernel, all subsequent hangs do not override the snapshot, nor do they create new ones. Devcoredump has a delayed work queue that will eventually delete the file node and free all the dump information. so it cleans up after itself 🧚

💀 heads up bestie: first crash gets the whole photoshoot. everything after that is just… not getting a snapshot. don't come for me, that's literally what the page says.

---

🌈 SEE ALSO 🌈

• Kernel file ./drivers/gpu/drm/xe/xe_devcoredump.c
• xe_devcoredump_read(9)
• xe_devcoredump(9)
• xe_print_blob_ascii85(9)

go read those too, iykyk 🍓

---

and that's the page!! this lil gem lives at ./drivers/gpu/drm/xe, dated October 2026, and it's literally titled Xe device coredump(9) — slay, go debug something 💜