Old Dell AMD business laptop AMDGPU crashing with 6.18 kerne

Message

RayDude · Post by **RayDude** » Fri Jan 02, 2026 1:30 am

I'm using an old Dell laptop as a TV on my kid's TV. Running Gentoo of course.

Thing has been flawless until lately. I think it's the latest kernel 6.18.2. I'm going to roll back to see if that fixes the issue.

Plasma is crashing and the task bar is disappearing leaving alt-tab as the only way to switch windows and no way to open new apps.

Here is a core dump from dmesg:

Code: Select all

[35999.083620] WARNING: CPU: 5 PID: 4221 at drivers/gpu/drm/amd/amdgpu/../display/dc/clk_mgr/dcn10/rv1_clk_mgr_vbios_smu.c:119 rv1_vbios_smu_set_dispclk+0x119/0x120 [amdgpu]
[35999.083959] Modules linked in: cifs nls_ucs2_utils cifs_md4 netfs algif_hash algif_skcipher af_alg bnep amdgpu binfmt_misc uvcvideo videobuf2_vmalloc videobuf2_memops uvc videobuf2_v4l2 videodev videobuf2_common mc usbmouse usbkbd btusb btrtl btintel joydev btbcm rtsx_usb_sdmmc btmtk bluetooth ath10k_pci rtsx_usb i2c_algo_bit ath10k_core usbhid drm_ttm_helper ttm snd_hda_codec_atihdmi drm_exec snd_hda_codec_generic drm_suballoc_helper snd_hda_codec_hdmi kvm_amd snd_hda_intel ath drm_buddy hid_multitouch snd_intel_dspcfg hid_generic snd_hda_codec drm_panel_backlight_quirks mac80211 wmi_bmof cpufreq_ondemand kvm gpu_sched snd_hda_core amdxcp snd_hwdep snd_pcm drm_display_helper drm_client_lib video r8169 snd_timer drm_kms_helper input_leds snd sp5100_tco ghash_clmulni_intel aesni_intel pcspkr cfg80211 serio_raw efi_pstore ccp cec soundcore i2c_piix4 i2c_hid_acpi backlight realtek libarc4 i2c_hid i2c_smbus wmi
[35999.084016] CPU: 5 UID: 0 PID: 4221 Comm: Xorg Not tainted 6.18.2-gentoo #1 PREEMPTLAZY 
[35999.084019] Hardware name: Dell Inc. Inspiron 5575/0DRVH0, BIOS 1.6.0 03/21/2022
[35999.084021] RIP: 0010:rv1_vbios_smu_set_dispclk+0x119/0x120 [amdgpu]
[35999.084337] Code: ef 48 69 f6 93 24 49 92 c1 f8 1f 48 c1 ee 20 01 de c1 fe 02 29 c6 49 8b 45 08 ff 50 28 69 c3 e8 03 00 00 5b 41 5c 41 5d 5d c3 <0f> 0b eb 97 0f 1f 00 55 48 89 e5 53 48 89 f3 e8 13 fe ff ff 48 c7
[35999.084339] RSP: 0018:ffffa31941683600 EFLAGS: 00010246
[35999.084342] RAX: 000041636fee8a40 RBX: 0000000000000000 RCX: 0000000000000005
[35999.084343] RDX: 0000000000005244 RSI: 0000000000004dfc RDI: 000041636fee37fc
[35999.084344] RBP: ffffa31941683618 R08: ffffa319487d8a6c R09: ffffa319487d8a0c
[35999.084346] R10: ffffa319416835c8 R11: 0000000000000000 R12: ffff9106f5cac800
[35999.084347] R13: ffff91068f32bde0 R14: ffff9107cd580000 R15: ffff9106f5cac800
[35999.084348] FS:  00007f7e126c6900(0000) GS:ffff9109f3930000(0000) knlGS:0000000000000000
[35999.084350] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[35999.084351] CR2: 00007fb5f9d45fe0 CR3: 000000014797e000 CR4: 00000000003506f0
[35999.084352] Call Trace:
[35999.084354]  <TASK>
[35999.084356]  rv1_update_clocks+0x227/0x470 [amdgpu]
[35999.084668]  dcn10_prepare_bandwidth+0x69/0x170 [amdgpu]
[35999.084984]  dc_commit_state_no_check+0x1d4/0xe10 [amdgpu]
[35999.085271]  dc_commit_streams+0x324/0x540 [amdgpu]
[35999.085553]  amdgpu_dm_atomic_commit_tail+0x53d/0x36b0 [amdgpu]
[35999.085857]  ? free_large_kmalloc+0x67/0xb0
[35999.085863]  ? kfree+0x2d0/0x3c0
[35999.085866]  ? kvfree+0x21/0x30
[35999.085869]  ? __set_output_tf.constprop.0+0xe9/0x170 [amdgpu]
[35999.086169]  ? dma_resv_iter_first_unlocked+0x26/0x80
[35999.086174]  ? dma_resv_get_fences+0x48/0x1f0
[35999.086177]  ? amdgpu_dm_plane_fill_plane_buffer_attributes+0x2af/0x520 [amdgpu]
[35999.086475]  ? wait_for_completion_timeout+0xfe/0x130
[35999.086481]  commit_tail+0xc9/0x160 [drm_kms_helper]
[35999.086496]  drm_atomic_helper_commit+0x12e/0x170 [drm_kms_helper]
[35999.086506]  drm_atomic_commit+0xa2/0xd0
[35999.086511]  ? drm_plane_get_damage_clips.cold+0x2a/0x2a
[35999.086515]  drm_atomic_helper_set_config+0x71/0xb0 [drm_kms_helper]
[35999.086525]  drm_mode_setcrtc+0x33d/0x930
[35999.086530]  ? drm_mode_getcrtc+0x180/0x180
[35999.086532]  drm_ioctl_kernel+0x9a/0xf0
[35999.086535]  drm_ioctl+0x29c/0x5f0
[35999.086537]  ? drm_mode_getcrtc+0x180/0x180
[35999.086541]  amdgpu_drm_ioctl+0x47/0x90 [amdgpu]
[35999.086778]  __x64_sys_ioctl+0x8f/0xe0
[35999.086782]  x64_sys_call+0x99e/0xa20
[35999.086785]  do_syscall_64+0x4c/0x1f0
[35999.086787]  entry_SYSCALL_64_after_hwframe+0x4b/0x53
[35999.086790] RIP: 0033:0x7f7e1292896f
[35999.086792] Code: 00 48 89 44 24 18 31 c0 48 8d 44 24 60 c7 04 24 10 00 00 00 48 89 44 24 08 48 8d 44 24 20 48 89 44 24 10 b8 10 00 00 00 0f 05 <89> c2 3d 00 f0 ff ff 77 18 48 8b 44 24 18 64 48 2b 04 25 28 00 00
[35999.086793] RSP: 002b:00007ffea0ba46d0 EFLAGS: 00000246 ORIG_RAX: 0000000000000010
[35999.086796] RAX: ffffffffffffffda RBX: 000055b1eabd07d0 RCX: 00007f7e1292896f
[35999.086797] RDX: 00007ffea0ba4760 RSI: 00000000c06864a2 RDI: 000000000000000c
[35999.086798] RBP: 00007ffea0ba4760 R08: 0000000000000000 R09: 000055b1ebb38480
[35999.086799] R10: 0000000000000058 R11: 0000000000000246 R12: 00000000c06864a2
[35999.086800] R13: 000000000000000c R14: 000055b1ebb38480 R15: 000055b1eaec3b80
[35999.086802]  </TASK>
[35999.086803] ---[ end trace 0000000000000000 ]---

Here is the video card in use:

Code: Select all

04:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] Raven Ridge [Radeon Vega Series / Radeon Vega Mobile Series] (rev c4)
        Subsystem: Dell Device 0812
        Kernel driver in use: amdgpu
        Kernel modules: amdgpu

I've done a little googling, and I doubt you guys will know anything, but just in case and so I can document my findings later, I started this thread.

Navar · Post by **Navar** » Mon Jan 05, 2026 12:31 pm

I'm on this kernel and the highest available linux firmware (20251125_p20251203)::gentoo, since I've been working on getting my new Framework 13 laptop up to speed with Gentoo and maybe document since existing information seems very out of date. It has been a pia in some regards vs every other kernel config on hardware I've manually done over the years. The pains of obscure "bleeding edge" in Linux. My aging Lenovo thinkpad was so much easier.

Anyway, that said, 6.18.0-2 has been stable for me on multiple AMD hardware here (desktop is at 6.18.0, fw13 aforementioned at .2). Suggest trying to align much newer firmware with that stage of the kernel and your issue may go away. YMMV.

RayDude · Post by **RayDude** » Mon Jan 05, 2026 7:37 pm

Navar wrote:I'm on this kernel and the highest available linux firmware (20251125_p20251203)::gentoo, since I've been working on getting my new Framework 13 laptop up to speed with Gentoo and maybe document since existing information seems very out of date. It has been a pia in some regards vs every other kernel config on hardware I've manually done over the years. The pains of obscure "bleeding edge" in Linux. My aging Lenovo thinkpad was so much easier.

Anyway, that said, 6.18.0-2 has been stable for me on multiple AMD hardware here (desktop is at 6.18.0, fw13 aforementioned at .2). Suggest trying to align much newer firmware with that stage of the kernel and your issue may go away. YMMV.

Thanks. I'll check the firmware version.

RayDude · Post by **RayDude** » Sat Jan 24, 2026 11:53 pm

RayDude wrote:
Navar wrote:I'm on this kernel and the highest available linux firmware (20251125_p20251203)::gentoo, since I've been working on getting my new Framework 13 laptop up to speed with Gentoo and maybe document since existing information seems very out of date. It has been a pia in some regards vs every other kernel config on hardware I've manually done over the years. The pains of obscure "bleeding edge" in Linux. My aging Lenovo thinkpad was so much easier.

Anyway, that said, 6.18.0-2 has been stable for me on multiple AMD hardware here (desktop is at 6.18.0, fw13 aforementioned at .2). Suggest trying to align much newer firmware with that stage of the kernel and your issue may go away. YMMV.
Thanks. I'll check the firmware version.

I don't think the latest firmware helped. I'm running 6.18.6 and will report back if the plasma bar is still crashing.

Update, removing the unmask from boost and from cryfs is currently compiling, so yeah, that's the right fix.