Message-ID: <CALG0vJt1D_yK89cRT=m=1uH=cffgkzZ4f69M3MUgp2fP15P1Cw@mail.gmail.com>
Date: Sun, 5 Oct 2025 20:16:25 +0200
From: Michael Guntsche <michael.guntsche@...loops.com>
To: linux-kernel@...r.kernel.org, regressions@...ts.linux.dev
Subject: [Regression] kernel panic after resume from hibernate, with
 thunderbolt dock connected

Good Evening,

After upgrading from 6.14 to a 6.17 kernel I noticed that my system
had a kernel panic after resuming from hibernation with the
thunderbolt dock connected. The caps-lock led kept blinking and the
screen was blank.
I bisected this down to a change between 6.14 and 6.15-rc1 but since
there were a lot of non booting configurations between these two I was
not able to find offending commit for it.

I compiled a minimum kernel image, with the main required drivers only
(nvme, fs, lvm) but no usb and thunderbolt and was able to reproduce
this issue as well.

What I did see in some debug sessions though, was that reading efi
variables was not possible after a return from hibernation
Okt 05 13:33:05 arthur kernel: ------------[ cut here ]------------
Okt 05 13:33:05 arthur kernel: [Firmware Bug]: Page fault caused by
firmware at PA: 0xfffffffee7e64514
Okt 05 13:33:05 arthur kernel: WARNING: CPU: 2 PID: 12902 at
arch/x86/platform/efi/quirks.c:747
efi_crash_gracefully_on_page_fault+0x64/0xd0
Okt 05 13:33:05 arthur kernel: Modules linked in: snd_seq_dummy
snd_hrtimer xt_conntrack nft_chain_nat xt_MASQUERADE nf_nat
nf_conntrack nf_defrag_ipv6 nf_defrag_ipv4 bridge stp llc xfrm_user>
Okt 05 13:33:05 arthur kernel:  snd_soc_acpi_intel_match
intel_tcc_cooling snd_soc_acpi_intel_sdca_quirks
soundwire_generic_allocation snd_soc_acpi x86_pkg_temp_thermal
intel_powerclamp sound>
Okt 05 13:33:05 arthur kernel:  cec intel_rapl_common rc_core
processor_thermal_wt_req processor_thermal_power_floor i2c_algo_bit
processor_thermal_mbox igen6_edac thinkpad_acpi intel_pmc_cor>
Okt 05 13:33:05 arthur kernel: CPU: 2 UID: 0 PID: 12902 Comm:
kworker/u48:159 Tainted: G        W          6.14.0-33-generic
#33-Ubuntu
Okt 05 13:33:05 arthur kernel: Tainted: [W]=WARN
Okt 05 13:33:05 arthur kernel: Hardware name: LENOVO
21HMCTO1WW/21HMCTO1WW, BIOS N3XET61W (1.36 ) 06/09/2025
Okt 05 13:33:05 arthur kernel: Workqueue: efi_rts_wq efi_call_rts
Okt 05 13:33:05 arthur kernel: RIP:
0010:efi_crash_gracefully_on_page_fault+0x64/0xd0
Okt 05 13:33:05 arthur kernel: Code: 00 00 76 08 48 3d 50 2c 7a 9d 74
0f 5b 41 5c 5d 31 c0 31 f6 31 ff c3 cc cc cc cc 48 89 de 48 c7 c7 a0
72 28 9c e8 ac 51 01 00 <0f> 0b 83 3d 83 a1 db 02 0a>
Okt 05 13:33:05 arthur kernel: RSP: 0018:ffffd2f14b52bbc0 EFLAGS: 00010046
Okt 05 13:33:05 arthur kernel: RAX: 0000000000000000 RBX:
fffffffee7e64514 RCX: 0000000000000000
Okt 05 13:33:05 arthur kernel: RDX: 0000000000000000 RSI:
0000000000000000 RDI: 0000000000000000
Okt 05 13:33:05 arthur kernel: RBP: ffffd2f14b52bbd0 R08:
0000000000000000 R09: 0000000000000000
Okt 05 13:33:05 arthur kernel: R10: 0000000000000000 R11:
0000000000000000 R12: ffffd2f14b52bd28
Okt 05 13:33:05 arthur kernel: R13: 0000000000000011 R14:
0000000000000000 R15: 0000000000000000
Okt 05 13:33:05 arthur kernel: FS:  0000000000000000(0000)
GS:ffff8e7fff300000(0000) knlGS:0000000000000000
Okt 05 13:33:05 arthur kernel: CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
Okt 05 13:33:05 arthur kernel: CR2: fffffffee7e64514 CR3:
0000000100333005 CR4: 0000000000f72ef0
Okt 05 13:33:05 arthur kernel: PKRU: 55555554
Okt 05 13:33:05 arthur kernel: Call Trace:
Okt 05 13:33:05 arthur kernel:  <TASK>
Okt 05 13:33:05 arthur kernel:  page_fault_oops+0x6b/0x180
Okt 05 13:33:05 arthur kernel:  kernelmode_fixup_or_oops.isra.0+0x69/0x90
Okt 05 13:33:05 arthur kernel:  __bad_area_nosemaphore+0x18e/0x1e0
Okt 05 13:33:05 arthur kernel:  bad_area_nosemaphore+0x16/0x30
Okt 05 13:33:05 arthur kernel:  do_kern_addr_fault.part.0+0x62/0x80
Okt 05 13:33:05 arthur kernel:  exc_page_fault+0x1c9/0x1e0
Okt 05 13:33:05 arthur kernel:  asm_exc_page_fault+0x27/0x30
Okt 05 13:33:05 arthur kernel: RIP: 0010:0xfffffffee7e64514
Okt 05 13:33:05 arthur kernel: Code: ff ff ff ff ff ff ff ff ff ff ff
ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
ff ff ff ff ff ff ff ff <ff> ff ff ff ff ff ff ff ff>
Okt 05 13:33:05 arthur kernel: RSP: 0018:ffffd2f14b52bdd8 EFLAGS: 00010282
Okt 05 13:33:05 arthur kernel: RAX: ffff8e78c29f7e00 RBX:
0000000000000246 RCX: ffffffffc1291ef8
Okt 05 13:33:05 arthur kernel: RDX: ffffd2f140ed72a0 RSI:
ffffffffc1291ef8 RDI: fffffffee7e64514
Okt 05 13:33:05 arthur kernel: RBP: ffffd2f14b52be10 R08:
0000000000000000 R09: ffffd2f140ed7258
Okt 05 13:33:05 arthur kernel: R10: 0000000000000000 R11:
0000000000000000 R12: ffffd2f140ed71c0
Okt 05 13:33:05 arthur kernel: R13: ffffffff9d7a2c50 R14:
ffff8e78ce70b180 R15: 0000000000000000
Okt 05 13:33:05 arthur kernel:  ? __efi_call+0x25/0x30
Okt 05 13:33:05 arthur kernel:  ? switch_mm+0x2c/0x60
Okt 05 13:33:05 arthur kernel:  ? efi_call_rts+0x11f/0x350
Okt 05 13:33:05 arthur kernel:  ? process_one_work+0x174/0x350
Okt 05 13:33:05 arthur kernel:  ? worker_thread+0x34a/0x480
Okt 05 13:33:05 arthur kernel:  ? __pfx_worker_thread+0x10/0x10
Okt 05 13:33:05 arthur kernel:  ? kthread+0xf9/0x230
Okt 05 13:33:05 arthur kernel:  ? __pfx_kthread+0x10/0x10
Okt 05 13:33:05 arthur kernel:  ? ret_from_fork+0x44/0x70
Okt 05 13:33:05 arthur kernel:  ? __pfx_kthread+0x10/0x10
Okt 05 13:33:05 arthur kernel:  ? ret_from_fork_asm+0x1a/0x30
Okt 05 13:33:05 arthur kernel:  </TASK>
Okt 05 13:33:05 arthur kernel: ---[ end trace 0000000000000000 ]---

To catch the real panic entry I set-up kdump and while it triggered
during testing by manually creating a kernel oops I did not start in
this specific case, so sadly there are no real logs of the kernel
panic available.

The hardware being used:
Hardware: Lenovo ThinkPad X1 Carbon Gen 11
Thunderbolt docks tested: Caldigit T3, Lenovo ThinkPad Thunderbold 3
Dock, Dell WD19TB

Since I am able to reproduce this with both thunderbolt and usb
removed, I think the problem lies somewhere else, but right now I have
no idea how to debug/bisect this further.

For testing I tried the current MASTER branch with the same results.
Please note the issue also happens if you hibernate with the dock
connected, remove the dock and boot-up again. This works fine in 6.14
as well.
I even tried downgrading the firmware to an earlier version, this did
fix the issue either.

Kind regards,
Michael

Powered by blists - more mailing lists