Re: stable/13 - amdgpu broken with n253286-d8a88ec38149

From: marco <>
Date: Fri, 06 Jan 2023 14:52:31 UTC
On Sun, Dec 18, 2022 at 08:55:09AM -0800, you (Jeff Gibbons) sent the following to [freebsd-stable] :
> I see exactly the same thing Jonathan does, but in my case with
> i915kms.ko (which also comes from the /usr/ports/graphics/drm-510-kmod/
> port, like his amdgpu.ko does).  My video device is Intel's
> 'WhiskeyLake-U GT2 [UHD Graphics 620]'.  This bug report may be
> related:

Same here on a ThinkPad Carbon X1 gen 7:

sysctl hw.model
hw.model: Intel(R) Core(TM) i7-8665U CPU @ 1.90GHz

My previous and okay BE was running 308d3d6be6da stable/13 and I've tried rebuilding using several newer commits all ending
with a kernel panic due to not being able to load i915kms (fatal trap 9:
general protection fault while in kernel mode).

I then decided to mount 1 of those newer BEs (based off of commit
50f61166f7b9) and remove i915kms from
kld_list and boot into that BE.
System came up, then reinstalled latest drm-510-kmod from ports,
kldloaded i915kms and then re-added i915kms to kld_list

[/usr/src] uname -aKU
FreeBSD 13.1-STABLE FreeBSD 13.1-STABLE #0 n253282-50f61166f7b9: Fri Jan  6 13:09:11 UTC 2023 amd64 1301510 1301510

I'm now rebuilding world using the latest commit 1149f0ec2b18 to see
what happens and if I need to perform this dance again or not.

Marco van Lienen -- FreeBSD enthusiast , GnuPG fingerprint: A025 D8AA AC1B D2FC 380D 4FC1 8EA0 0BA8 8580 E6CB
"The Tuck Pendleton defects."