Update to 6.15 failed due to kernel null pointer dereference

Ask Fedora
,
1

Hopefully, this is evidence enough: [1]

23:55:34 systemd[1]: systemd-udevd.service: State 'stop-sigterm' timed out. Killing.
23:55:34 systemd[1]: systemd-udevd.service: Killing process 550 (systemd-udevd) with signal SIGKILL.
23:55:34 systemd[1]: systemd-udevd.service: Main process exited, code=killed, status=9/KILL
23:55:51 kernel: amdgpu 0000:59:00.0: [drm] Alt mode has timed out after 57004 ms
23:55:51 kernel: [drm:amdgpu_dm_init [amdgpu]] *ERROR* amdgpu: failed to initialize sw for display support.
23:55:51 kernel: [drm:amdgpu_device_ip_init [amdgpu]] *ERROR* hw_init of IP block <dm> failed -22
23:55:51 kernel: amdgpu 0000:59:00.0: amdgpu: amdgpu_device_ip_init failed
23:55:51 kernel: amdgpu 0000:59:00.0: amdgpu: Fatal error during GPU init
23:55:51 kernel: amdgpu 0000:59:00.0: amdgpu: amdgpu: finishing device.
23:55:51 kernel: BUG: kernel NULL pointer dereference, address: 00000000000005b0
23:55:51 kernel: #PF: supervisor read access in kernel mode
23:55:51 kernel: #PF: error_code(0x0000) - not-present page
23:55:51 kernel: PGD 0 P4D 0 
23:55:51 kernel: Oops: Oops: 0000 [#1] SMP NOPTI
23:55:51 kernel: CPU: 3 UID: 0 PID: 568 Comm: (udev-worker) Not tainted 6.15.3-200.fc42.x86_64 #1 PREEMPT(lazy) 
23:55:51 kernel: Hardware name: ASRock X670E Taichi/X670E Taichi, BIOS 3.08 09/19/2024
23:55:51 kernel: RIP: 0010:dc_resource_is_dsc_encoding_supported+0x9/0x30 [amdgpu]
23:55:51 kernel: Code: 8b 14 24 e9 69 fb ff ff e9 a7 d2 06 00 0f 1f 40 00 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 66 0f 1f 00 0f 1f 44 00 00 <48> 8b 87 b0 05 00 00 48 85 c0 74 14 48 8b 80 b8 04 00 00 8b 40 24
23:55:51 kernel: RSP: 0018:ffffc9cf80827648 EFLAGS: 00010246
23:55:51 kernel: RAX: 0000000000000000 RBX: ffff896529700010 RCX: 0000000000000000
23:55:51 kernel: RDX: ffff896529700010 RSI: ffff896504c2da00 RDI: 0000000000000000
23:55:51 kernel: RBP: ffff896515bfa900 R08: 0000000000000000 R09: 0000000000000000
23:55:51 kernel: R10: ffffc9cf808277b8 R11: 0000000000000000 R12: 0000000000000000
23:55:51 kernel: R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000000000
23:55:51 kernel: FS:  00007f7465d78040(0000) GS:ffff896c739f1000(0000) knlGS:0000000000000000
23:55:51 kernel: CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
23:55:51 kernel: CR2: 00000000000005b0 CR3: 000000011309b000 CR4: 0000000000f50ef0
23:55:51 kernel: PKRU: 55555554
23:55:51 kernel: Call Trace:
23:55:51 kernel:  <TASK>
23:55:51 kernel:  amdgpu_dm_atomic_check+0xf5/0x1530 [amdgpu]
23:55:51 kernel:  drm_atomic_check_only+0x18d/0x3f0
23:55:51 kernel:  ? srso_alias_return_thunk+0x5/0xfbef5
23:55:51 kernel:  drm_atomic_commit+0x71/0xe0
23:55:51 kernel:  ? __pfx___drm_printfn_info+0x10/0x10
23:55:51 kernel:  drm_atomic_helper_disable_all+0x187/0x1a0
23:55:51 kernel:  drm_atomic_helper_shutdown+0x8a/0x120
23:55:51 kernel:  ? srso_alias_return_thunk+0x5/0xfbef5
23:55:51 kernel:  amdgpu_device_fini_hw+0xad/0x1a6 [amdgpu]
23:55:51 kernel:  ? srso_alias_return_thunk+0x5/0xfbef5
23:55:51 kernel:  amdgpu_driver_load_kms.cold+0x19/0x2f [amdgpu]
23:55:51 kernel:  amdgpu_pci_probe+0x1df/0x540 [amdgpu]
23:55:51 kernel:  ? srso_alias_return_thunk+0x5/0xfbef5
23:55:51 kernel:  local_pci_probe+0x3f/0x90
23:55:51 kernel:  pci_call_probe+0x5b/0x190
23:55:51 kernel:  ? srso_alias_return_thunk+0x5/0xfbef5
23:55:51 kernel:  ? kernfs_create_link+0x61/0xb0
23:55:51 kernel:  pci_device_probe+0x95/0x140
23:55:51 kernel:  ? srso_alias_return_thunk+0x5/0xfbef5
23:55:51 kernel:  really_probe+0xdb/0x340
23:55:51 kernel:  ? pm_runtime_barrier+0x55/0x90
23:55:51 kernel:  __driver_probe_device+0x78/0x140
23:55:51 kernel:  driver_probe_device+0x1f/0xa0
23:55:51 kernel:  ? __pfx___driver_attach+0x10/0x10
23:55:51 kernel:  __driver_attach+0xcb/0x1e0
23:55:51 kernel:  bus_for_each_dev+0x82/0xd0
23:55:51 kernel:  bus_add_driver+0x12f/0x210
23:55:51 kernel:  ? __pfx_amdgpu_init+0x10/0x10 [amdgpu]
23:55:51 kernel:  driver_register+0x75/0xe0
23:55:51 kernel:  ? amdgpu_init+0x42/0xff0 [amdgpu]
23:55:51 kernel:  do_one_initcall+0x58/0x300
23:55:51 kernel:  do_init_module+0x84/0x260
23:55:51 kernel:  init_module_from_file+0x8a/0xe0
23:55:51 kernel:  idempotent_init_module+0x114/0x310
23:55:51 kernel:  __x64_sys_finit_module+0x67/0xc0
23:55:51 kernel:  do_syscall_64+0x7b/0x160
23:55:51 kernel:  ? srso_alias_return_thunk+0x5/0xfbef5
23:55:51 kernel:  ? ksys_read+0x73/0xe0
23:55:51 kernel:  ? srso_alias_return_thunk+0x5/0xfbef5
23:55:51 kernel:  ? syscall_exit_to_user_mode+0x10/0x210
23:55:51 kernel:  ? srso_alias_return_thunk+0x5/0xfbef5
23:55:51 kernel:  ? do_syscall_64+0x87/0x160
23:55:51 kernel:  ? __pfx_page_put_link+0x10/0x10
23:55:51 kernel:  ? alloc_fd+0xbd/0x110
23:55:51 kernel:  ? srso_alias_return_thunk+0x5/0xfbef5
23:55:51 kernel:  ? do_sys_openat2+0xa2/0xe0
23:55:51 kernel:  ? srso_alias_return_thunk+0x5/0xfbef5
23:55:51 kernel:  ? srso_alias_return_thunk+0x5/0xfbef5
23:55:51 kernel:  ? syscall_exit_to_user_mode+0x10/0x210
23:55:51 kernel:  ? srso_alias_return_thunk+0x5/0xfbef5
23:55:51 kernel:  ? do_syscall_64+0x87/0x160
23:55:51 kernel:  ? srso_alias_return_thunk+0x5/0xfbef5
23:55:51 kernel:  ? do_syscall_64+0x87/0x160
23:55:51 kernel:  ? srso_alias_return_thunk+0x5/0xfbef5
23:55:51 kernel:  ? do_syscall_64+0x87/0x160
23:55:51 kernel:  ? exc_page_fault+0x7e/0x1a0
23:55:51 kernel:  entry_SYSCALL_64_after_hwframe+0x76/0x7e
23:55:51 kernel: RIP: 0033:0x7f7466637a8d
23:55:51 kernel: Code: ff c3 66 2e 0f 1f 84 00 00 00 00 00 90 f3 0f 1e fa 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d 4b 63 0f 00 f7 d8 64 89 01 48
23:55:51 kernel: RSP: 002b:00007fffe39c79f8 EFLAGS: 00000246 ORIG_RAX: 0000000000000139
23:55:51 kernel: RAX: ffffffffffffffda RBX: 000055fa47842b70 RCX: 00007f7466637a8d
23:55:51 kernel: RDX: 0000000000000004 RSI: 00007f7465d6e965 RDI: 000000000000001d
23:55:51 kernel: RBP: 00007fffe39c7ab0 R08: 0000000000000000 R09: 00007fffe39c7a60
23:55:51 kernel: R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000020000
23:55:51 kernel: R13: 000055fa47808400 R14: 00007f7465d6e965 R15: 0000000000000000
23:55:51 kernel:  </TASK>
23:55:51 kernel: Modules linked in: amdgpu(+) amdxcp i2c_algo_bit drm_ttm_helper ttm drm_exec gpu_sched polyval_clmulni drm_suballoc_helper polyval_generic drm_panel_backlight_quirks drm_buddy ghash_clmulni_intel nvme uas sha512_ssse3 drm_display_helper usb_storage nvme_core sha256_ssse3 sha1_ssse3 sp5100_tco nvme_keyring cec nvme_auth video wmi fuse i2c_dev
23:55:51 kernel: CR2: 00000000000005b0
23:55:51 kernel: ---[ end trace 0000000000000000 ]---
23:55:51 kernel: RIP: 0010:dc_resource_is_dsc_encoding_supported+0x9/0x30 [amdgpu]
23:55:51 kernel: Code: 8b 14 24 e9 69 fb ff ff e9 a7 d2 06 00 0f 1f 40 00 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 66 0f 1f 00 0f 1f 44 00 00 <48> 8b 87 b0 05 00 00 48 85 c0 74 14 48 8b 80 b8 04 00 00 8b 40 24
23:55:51 kernel: RSP: 0018:ffffc9cf80827648 EFLAGS: 00010246
23:55:51 kernel: RAX: 0000000000000000 RBX: ffff896529700010 RCX: 0000000000000000
23:55:51 kernel: RDX: ffff896529700010 RSI: ffff896504c2da00 RDI: 0000000000000000
23:55:51 kernel: RBP: ffff896515bfa900 R08: 0000000000000000 R09: 0000000000000000
23:55:51 kernel: R10: ffffc9cf808277b8 R11: 0000000000000000 R12: 0000000000000000
23:55:51 kernel: R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000000000
23:55:51 kernel: FS:  00007f7465d78040(0000) GS:ffff896c739f1000(0000) knlGS:0000000000000000
23:55:51 kernel: CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
23:55:51 kernel: CR2: 00000000000005b0 CR3: 000000011309b000 CR4: 0000000000f50ef0
23:55:51 kernel: PKRU: 55555554
23:55:51 kernel: note: (udev-worker)[568] exited with irqs disabled

Damned strange that this isn’t in gnome-abrt.

I also see the errors from gitlab.freedesktop.org/drm/amd/-/issues/3647 appear beforehand, which I’ve never seen otherwise.

  1. journalctl -b 76f7df1e0b7a4a47998a5c51353e90e2 ↩︎

1 Like
2

AMD issues have become a real problem recently :frowning: I’m currently testing kernels for #4141 and broke my btrfs root fs with an experimental kernel as I cannot reproduce the issue anywhere else than on my laptop natively :laughing:

You might add your data upstream, and maybe add a bugzilla kernel ticket, elaborating the issue+behavior, and linking it to the upstream ticket (and if useful data comes up there, also link the bugzilla ticket upstream). Add a full kernel log of a broken boot if doing so (sudo journalctl -k --no-hostname --boot=<broken boot number>).

Btw, which 6.15 is it? Before opening anything, test at least 6.15.3. I would be careful with testing the current 6.16rc kernel, as it is not yet clear if it was the kernel that broke my btrfs (BZ#2374607) → I mention this as the kernel template might be read to encourage people to test current rc’s.

1 Like
3

@Chirts, dnf5 info $(rpm -qa kernel | tail -n 1) returns:

Name            : kernel
Epoch           : 0
Version         : 6.15.3
Release         : 200.fc42
Architecture    : x86_64
Installed size  : 0.0   B
Source          : kernel-6.15.3-200.fc42.src.rpm
From repository : <unknown>
Vendor          : Fedora Project

If you’ve reason think it’s upstream, or that’s the normal way to report potentially downstream problems in the kernel, I shall. I’ve always been wary of doing so without explicitly receiving confirmation from a downstream report, because I was under the impression that Fedora building the kernel themselves could cause downstream problems that upstream has no interest in. Consequently, I’ve preliminarily submitted bugzilla.redhat.com/show_bug.cgi?id=2374742. KDE accepts downstream issues, though, so it’s not without precedent.

4

Doubled cursors and flickering in SDDM also occur on my brother’s machine:

5

I’ve written about the flicker issue here: System is unstable after Kernel update to 6.15.3-200.fc42 - #4 by guiltydoggy

Try turning off Freesync or revert to 6.14.

1 Like
6

I’ve summarised both of these at Bodhi. [1]

  1. bodhi.fedoraproject.org/updates/FEDORA-2025-af128cd558 ↩︎

7

Yeah, I’ll say. I replaced all my AMD 7600s with NVIDIA 5060s and have been much happier ever since.

I had Zen5 kernel issues and amdgpu issues regularly during the last 9 months / year. I gave up and went with what generally, usually works… :sad_but_relieved_face:

8

It looks like it’s been reported

2 Likes
9

@leigh123linux, thanks. I’ve CLOSED the RHBZ ticket as UPSTREAM, and added the information to it at #note_2979355. It seems like my trace differs slightly, as do all of the submissions to its thread thus far.

1 Like