Re: ✗ Xe.CI.FULL: failure for drm/xe: Make drm_driver const

[Michal Wajdeczko <michal.wajdeczko@intel.com>]

On 5/14/2026 3:01 AM, Patchwork wrote:
> *Patch Details*
> *Series:*	drm/xe: Make drm_driver const
> *URL:*	https://patchwork.freedesktop.org/series/166441/ <https://patchwork.freedesktop.org/series/166441/>
> *State:*	failure
> *Details:*	https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-166441v1/index.html <https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-166441v1/index.html>
> 
> 
>   CI Bug Log - changes from xe-5052-161329a551bb5ebd34524b781cd7fb3d84b994be_FULL -> xe-pw-166441v1_FULL
> 
> 
>     Summary
> 
> *FAILURE*
> 
> Serious unknown changes coming with xe-pw-166441v1_FULL absolutely need to be
> verified manually.
> 
> If you think the reported changes have nothing to do with the changes
> introduced in xe-pw-166441v1_FULL, please notify your bug team (I915-ci-infra@lists.freedesktop.org) to allow them
> to document this new failure mode, which will reduce false positives in CI.
> 
> 
>     Participating hosts (2 -> 2)
> 
> No changes in participating hosts
> 
> 
>     Possible new issues
> 
> Here are the unknown changes that may have been introduced in xe-pw-166441v1_FULL:
> 
> 
>       IGT changes
> 
> 
>         Possible regressions
> 
>   * igt@xe_sriov_vfio@bind-unbind-vfs@numvfs-16:
>       o shard-bmg: PASS <https://intel-gfx-ci.01.org/tree/intel-xe/xe-5052-161329a551bb5ebd34524b781cd7fb3d84b994be/shard-bmg-2/igt@xe_sriov_vfio@bind-unbind-vfs@numvfs-16.html> -> FAIL <https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-166441v1/shard-bmg-2/igt@xe_sriov_vfio@bind-unbind-vfs@numvfs-16.html> +19 other tests fail

it's a GuC crash

<7>[  103.825980] xe 0000:03:00.0: [drm:xe_pci_sriov_configure [xe]] PF: disabling 4 VFs
<7>[  104.867824] xe 0000:03:00.0: [drm:pf_update_policy_bool [xe]] PF: Tile0: GT0: policy key 0x8001 'sched_if_idle' updated to 'disabled'
<6>[  104.868464] xe 0000:03:00.0: [drm] PF: Tile0: GT0: PF..VF24 provisioned with 0(low) scheduling priority
<6>[  104.869202] xe 0000:03:00.0: [drm] PF: Tile0: GT0: PF provisioned with 0(infinity) execution quantum
<6>[  104.869904] xe 0000:03:00.0: [drm] PF: Tile0: GT0: PF provisioned with 0(infinity) preemption timeout
<7>[  104.870736] xe 0000:03:00.0: [drm:pf_update_policy_bool [xe]] PF: Tile0: GT1: policy key 0x8001 'sched_if_idle' updated to 'disabled'
<6>[  104.871455] xe 0000:03:00.0: [drm] PF: Tile0: GT1: PF..VF24 provisioned with 0(low) scheduling priority
<6>[  104.872228] xe 0000:03:00.0: [drm] PF: Tile0: GT1: PF provisioned with 0(infinity) execution quantum
<3>[  104.872934] xe 0000:03:00.0: [drm] *ERROR* Tile0: GT0: GuC Crash dump notification
<6>[  105.065584] xe 0000:03:00.0: [drm] Tile0: GT0: trying reset from dequeue_one_g2h [xe]
<6>[  105.065660] xe 0000:03:00.0: [drm] Tile0: GT0: Capture 1.1: **** Xe Device Coredump ****
<6>[  105.065666] xe 0000:03:00.0: [drm] Tile0: GT0: Capture 1.2: Reason: CTB is dead - 0x8002
<6>[  105.065682] xe 0000:03:00.0: [drm] Tile0: GT0: reset queued
<6>[  105.065682] xe 0000:03:00.0: [drm] Tile0: GT0: Capture 1.3: PCI ID: 0xe20b
<6>[  105.065686] xe 0000:03:00.0: [drm] Tile0: GT0: Capture 1.4: PCI revision: 0x00
<6>[  105.065688] xe 0000:03:00.0: [drm] Tile0: GT0: Capture 1.5: GT id: 0
<6>[  105.065690] xe 0000:03:00.0: [drm] Tile0: GT0: Capture 1.6: 	Tile: 0
<6>[  105.065692] xe 0000:03:00.0: [drm] Tile0: GT0: Capture 1.7: 	Type: main
<6>[  105.065718] xe 0000:03:00.0: [drm] Tile0: GT0: Capture 1.8: 	IP ver: 20.1.0
<6>[  105.065721] xe 0000:03:00.0: [drm] Tile0: GT0: Capture 1.9: 	CS reference clock: 19200000
<6>[  105.065723] xe 0000:03:00.0: [drm] Tile0: GT0: Capture 1.10: GT id: 1
<6>[  105.065726] xe 0000:03:00.0: [drm] Tile0: GT0: Capture 1.11: 	Tile: 0
<6>[  105.065728] xe 0000:03:00.0: [drm] Tile0: GT0: Capture 1.12: 	Type: media
<6>[  105.065731] xe 0000:03:00.0: [drm] Tile0: GT0: Capture 1.13: 	IP ver: 13.1.1
<6>[  105.065733] xe 0000:03:00.0: [drm] Tile0: GT0: Capture 1.14: 	CS reference clock: 19200000
<6>[  105.065736] xe 0000:03:00.0: [drm] Tile0: GT0: Capture 1.15: **** GT #0 ****
<6>[  105.065738] xe 0000:03:00.0: [drm] Tile0: GT0: Capture 1.16: 	Tile: 0
<6>[  105.065741] xe 0000:03:00.0: [drm] Tile0: GT0: Capture 1.17: **** GuC Log ****
<6>[  105.065744] xe 0000:03:00.0: [drm] Tile0: GT0: Capture 1.18: GuC firmware: xe/bmg_guc_70.bin
<6>[  105.065747] xe 0000:03:00.0: [drm] Tile0: GT0: Capture 1.19: GuC version: 70.65.0 (wanted 70.54.0)
<3>[  105.065747] xe 0000:03:00.0: [drm] *ERROR* Tile0: GT0: GuC Exception notification
<6>[  105.065750] xe 0000:03:00.0: [drm] Tile0: GT0: Capture 1.20: Kernel timestamp: 0x1C63EC3DC6 [121935510982]
<6>[  105.065851] xe 0000:03:00.0: [drm] Tile0: GT0: reset started
<6>[  105.073225] xe 0000:03:00.0: [drm] Tile0: GT0: trying reset from dequeue_one_g2h [xe]
<6>[  105.073321] xe 0000:03:00.0: [drm] Tile0: GT0: Capture 1.21: GuC timestamp: 0x16B43077 [380907639]

but it seems to be happening on this machine, see #7774