Arch Linux

Markix

Member

Registered: 2025-02-23

Posts: 2

AMD Graphics dying seemingly at random

I'm running a laptop with an AMD CPU and integrated graphics (Ryzen 7 8840HS + Radeon 780M) and every so often, input locks up for a second and then the graphics just die, without being able to come back up on their own. Digging through journnalctl logs it seems like the GPU drivers are resetting on their own, however I have no idea what could be causing this in the first place.

Mar 27 09:10:52 EeveeArch kernel: amdgpu 0000:c2:00.0: amdgpu: Dumping IP State
Mar 27 09:10:57 EeveeArch kernel: amdgpu 0000:c2:00.0: amdgpu: SMU: I'm not done with your previous command: SMN_C2PMSG_66:0x0000001A SMN_C2PMSG_82:0x00000000
Mar 27 09:10:57 EeveeArch kernel: amdgpu 0000:c2:00.0: amdgpu: Failed to disable gfxoff!
Mar 27 09:11:02 EeveeArch kernel: amdgpu 0000:c2:00.0: amdgpu: SMU: I'm not done with your previous command: SMN_C2PMSG_66:0x0000001A SMN_C2PMSG_82:0x00000000
Mar 27 09:11:02 EeveeArch kernel: amdgpu 0000:c2:00.0: amdgpu: Failed to disable gfxoff!
Mar 27 09:11:07 EeveeArch kernel: amdgpu 0000:c2:00.0: amdgpu: SMU: I'm not done with your previous command: SMN_C2PMSG_66:0x0000001A SMN_C2PMSG_82:0x00000000
Mar 27 09:11:07 EeveeArch kernel: amdgpu 0000:c2:00.0: amdgpu: Failed to disable gfxoff!
Mar 27 09:11:11 EeveeArch kernel: ACPI Error: Aborting method \_SB.A018 due to previous error (AE_AML_LOOP_TIMEOUT) (20240827/psparse-529)
Mar 27 09:11:11 EeveeArch kernel: ACPI Error: Aborting method \_SB.A011 due to previous error (AE_AML_LOOP_TIMEOUT) (20240827/psparse-529)
Mar 27 09:11:11 EeveeArch kernel: ACPI Error: Aborting method \_SB.A026 due to previous error (AE_AML_LOOP_TIMEOUT) (20240827/psparse-529)
Mar 27 09:11:11 EeveeArch kernel: ACPI Error: Aborting method \_SB.ALIB due to previous error (AE_AML_LOOP_TIMEOUT) (20240827/psparse-529)
Mar 27 09:11:11 EeveeArch kernel: ACPI Error: Aborting method \_SB.PCI0.LPC0.EC0.ACAD._PSR due to previous error (AE_AML_LOOP_TIMEOUT) (20240827/psparse-529)
Mar 27 09:11:11 EeveeArch kernel: ACPI: \_SB_.PCI0.LPC0.EC0_.ACAD: Error reading AC Adapter state: AE_AML_LOOP_TIMEOUT
Mar 27 09:11:12 EeveeArch kernel: amdgpu 0000:c2:00.0: amdgpu: SMU: I'm not done with your previous command: SMN_C2PMSG_66:0x0000001A SMN_C2PMSG_82:0x00000000
Mar 27 09:11:12 EeveeArch kernel: amdgpu 0000:c2:00.0: amdgpu: Failed to disable gfxoff!
Mar 27 09:11:12 EeveeArch kernel: amdgpu 0000:c2:00.0: amdgpu: Dumping IP State Completed
Mar 27 09:11:12 EeveeArch kernel: amdgpu 0000:c2:00.0: amdgpu: ring gfx_0.0.0 timeout, signaled seq=12965, emitted seq=12967
Mar 27 09:11:12 EeveeArch kernel: amdgpu 0000:c2:00.0: amdgpu: Process information: process codium pid 6261 thread codium:cs0 pid 6277
Mar 27 09:11:12 EeveeArch kernel: amdgpu 0000:c2:00.0: amdgpu: Starting gfx_0.0.0 ring reset
Mar 27 09:11:15 EeveeArch kernel: amdgpu 0000:c2:00.0: amdgpu: MES failed to respond to msg=RESET
Mar 27 09:11:15 EeveeArch kernel: [drm:amdgpu_mes_reset_legacy_queue [amdgpu]] *ERROR* failed to reset legacy queue
Mar 27 09:11:15 EeveeArch kernel: amdgpu 0000:c2:00.0: amdgpu: Ring gfx_0.0.0 reset failure
Mar 27 09:11:15 EeveeArch kernel: amdgpu 0000:c2:00.0: amdgpu: GPU reset begin!
Mar 27 09:11:20 EeveeArch kernel: amdgpu 0000:c2:00.0: amdgpu: SMU: I'm not done with your previous command: SMN_C2PMSG_66:0x0000001A SMN_C2PMSG_82:0x00000000
Mar 27 09:11:20 EeveeArch kernel: amdgpu 0000:c2:00.0: amdgpu: Failed to disable gfxoff!
Mar 27 09:11:25 EeveeArch kernel: amdgpu 0000:c2:00.0: amdgpu: SMU: I'm not done with your previous command: SMN_C2PMSG_66:0x0000001A SMN_C2PMSG_82:0x00000000
Mar 27 09:11:25 EeveeArch kernel: amdgpu 0000:c2:00.0: amdgpu: Failed to disable gfxoff!
Mar 27 09:11:27 EeveeArch kernel: ------------[ cut here ]------------
Mar 27 09:11:27 EeveeArch kernel: WARNING: CPU: 6 PID: 12 at drivers/gpu/drm/amd/amdgpu/../display/dc/clk_mgr/dcn314/dcn314_smu.c:159 dcn314_smu_send_msg_with>
Mar 27 09:11:27 EeveeArch kernel: Modules linked in: ccm snd_seq_dummy snd_hrtimer snd_seq snd_seq_device snd_ctl_led snd_soc_dmic snd_soc_ps_mach snd_ps_pdm_>
Mar 27 09:11:27 EeveeArch kernel:  polyval_clmulni hid_multitouch soundcore snd_pci_acp3x polyval_generic ghash_clmulni_intel sha512_ssse3 mac80211 sha256_sss>
Mar 27 09:11:27 EeveeArch kernel: CPU: 6 UID: 0 PID: 12 Comm: kworker/u64:0 Tainted: G        W          6.14.0-rt3-arch1-7-rt #1 PREEMPT_{RT,(lazy)}  cfedb87>
Mar 27 09:11:27 EeveeArch kernel: Tainted: [W]=WARN
Mar 27 09:11:27 EeveeArch kernel: Hardware name: Acer Aspire A15-61M/Kinder2_HKU, BIOS V1.03 03/22/2025
Mar 27 09:11:27 EeveeArch kernel: Workqueue: amdgpu-reset-dev drm_sched_job_timedout [gpu_sched]
Mar 27 09:11:27 EeveeArch kernel: RIP: 0010:dcn314_smu_send_msg_with_param+0x109/0x190 [amdgpu]
Mar 27 09:11:27 EeveeArch kernel: Code: c7 c2 10 6d ff c0 5d 41 5c 41 5d e9 81 18 ef ff 89 da 48 c7 c6 60 ff 14 c1 48 c7 c7 68 dc c5 c0 e8 bc 8b 6b db e9 46 f>
Mar 27 09:11:27 EeveeArch kernel: RSP: 0018:ffffb2ad4019f848 EFLAGS: 00010246
Mar 27 09:11:27 EeveeArch kernel: RAX: 000000cc7cf72a0a RBX: 0000000000000000 RCX: 0000000000000006
Mar 27 09:11:27 EeveeArch kernel: RDX: 000000000000877b RSI: 00000000000080a9 RDI: 000000cc7cf6a28f
Mar 27 09:11:27 EeveeArch kernel: RBP: ffff9c608d503c00 R08: ffff9c608fe6df80 R09: 0000000000000000
Mar 27 09:11:27 EeveeArch kernel: R10: 0000000000000000 R11: 0000000000000001 R12: 000000000000000d
Mar 27 09:11:27 EeveeArch kernel: R13: 0000000000000000 R14: ffff9c60857f8228 R15: ffff9c62d7780fb8
Mar 27 09:11:27 EeveeArch kernel: FS:  0000000000000000(0000) GS:ffff9c63ee500000(0000) knlGS:0000000000000000
Mar 27 09:11:27 EeveeArch kernel: CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
Mar 27 09:11:27 EeveeArch kernel: CR2: 000076ca5dac3060 CR3: 000000011f430000 CR4: 0000000000f50ef0
Mar 27 09:11:27 EeveeArch kernel: PKRU: 55555554
Mar 27 09:11:27 EeveeArch kernel: Call Trace:
Mar 27 09:11:27 EeveeArch kernel:  <TASK>
Mar 27 09:11:27 EeveeArch kernel:  ? dcn314_smu_send_msg_with_param+0x109/0x190 [amdgpu afcd62455856506c868a8f438816a0d6ab1da2d1]
Mar 27 09:11:27 EeveeArch kernel:  ? __warn.cold+0x93/0xfa
Mar 27 09:11:27 EeveeArch kernel:  ? dcn314_smu_send_msg_with_param+0x109/0x190 [amdgpu afcd62455856506c868a8f438816a0d6ab1da2d1]
Mar 27 09:11:27 EeveeArch kernel:  ? __warn.cold+0x93/0xfa
Mar 27 09:11:27 EeveeArch kernel:  ? dcn314_smu_send_msg_with_param+0x109/0x190 [amdgpu afcd62455856506c868a8f438816a0d6ab1da2d1]
Mar 27 09:11:27 EeveeArch kernel:  ? report_bug+0xe4/0x180
Mar 27 09:11:27 EeveeArch kernel:  ? handle_bug+0x5e/0xa0
Mar 27 09:11:27 EeveeArch kernel:  ? exc_invalid_op+0x18/0x70
Mar 27 09:11:27 EeveeArch kernel:  ? asm_exc_invalid_op+0x1a/0x20
Mar 27 09:11:27 EeveeArch kernel:  ? dcn314_smu_send_msg_with_param+0x109/0x190 [amdgpu afcd62455856506c868a8f438816a0d6ab1da2d1]
Mar 27 09:11:27 EeveeArch kernel:  ? dcn314_smu_send_msg_with_param+0xaf/0x190 [amdgpu afcd62455856506c868a8f438816a0d6ab1da2d1]
Mar 27 09:11:27 EeveeArch kernel:  link_set_dpms_off+0x113/0x780 [amdgpu afcd62455856506c868a8f438816a0d6ab1da2d1]
Mar 27 09:11:27 EeveeArch kernel:  ? optc31_set_drr+0x12b/0x1e0 [amdgpu afcd62455856506c868a8f438816a0d6ab1da2d1]
Mar 27 09:11:27 EeveeArch kernel:  dcn31_reset_hw_ctx_wrap+0x2a2/0x490 [amdgpu afcd62455856506c868a8f438816a0d6ab1da2d1]
Mar 27 09:11:27 EeveeArch kernel:  dce110_apply_ctx_to_hw+0x63/0x2d0 [amdgpu afcd62455856506c868a8f438816a0d6ab1da2d1]
Mar 27 09:11:27 EeveeArch kernel:  ? __entry_text_end+0x101e45/0x101e49
Mar 27 09:11:27 EeveeArch kernel:  dc_commit_state_no_check+0x60e/0xe50 [amdgpu afcd62455856506c868a8f438816a0d6ab1da2d1]
Mar 27 09:11:27 EeveeArch kernel:  dc_commit_streams+0x359/0x470 [amdgpu afcd62455856506c868a8f438816a0d6ab1da2d1]
Mar 27 09:11:27 EeveeArch kernel:  dm_suspend+0x244/0x260 [amdgpu afcd62455856506c868a8f438816a0d6ab1da2d1]
Mar 27 09:11:27 EeveeArch kernel:  amdgpu_ip_block_suspend+0x24/0x50 [amdgpu afcd62455856506c868a8f438816a0d6ab1da2d1]
Mar 27 09:11:27 EeveeArch kernel:  amdgpu_device_ip_suspend_phase1+0xaa/0xc0 [amdgpu afcd62455856506c868a8f438816a0d6ab1da2d1]
Mar 27 09:11:27 EeveeArch kernel:  amdgpu_device_ip_suspend+0x2c/0x80 [amdgpu afcd62455856506c868a8f438816a0d6ab1da2d1]
Mar 27 09:11:27 EeveeArch kernel:  ? amdgpu_device_ip_need_full_reset+0x16/0x80 [amdgpu afcd62455856506c868a8f438816a0d6ab1da2d1]
Mar 27 09:11:27 EeveeArch kernel:  amdgpu_device_pre_asic_reset+0xe9/0x2c0 [amdgpu afcd62455856506c868a8f438816a0d6ab1da2d1]
Mar 27 09:11:27 EeveeArch kernel:  amdgpu_device_gpu_recover.cold+0x52d/0xcee [amdgpu afcd62455856506c868a8f438816a0d6ab1da2d1]
Mar 27 09:11:27 EeveeArch kernel:  amdgpu_job_timedout.cold+0x345/0x38a [amdgpu afcd62455856506c868a8f438816a0d6ab1da2d1]
Mar 27 09:11:27 EeveeArch kernel:  drm_sched_job_timedout+0x85/0x120 [gpu_sched ea51dd608ff6ba2fcc0e280b16c72577998705d4]
Mar 27 09:11:27 EeveeArch kernel:  process_one_work+0x17c/0x340
Mar 27 09:11:27 EeveeArch kernel:  worker_thread+0x2d2/0x400
Mar 27 09:11:27 EeveeArch kernel:  ? _raw_spin_lock_irqsave+0x27/0x60
Mar 27 09:11:27 EeveeArch kernel:  ? __pfx_worker_thread+0x10/0x10
Mar 27 09:11:27 EeveeArch kernel:  kthread+0xfd/0x260
Mar 27 09:11:27 EeveeArch kernel:  ? srso_alias_return_thunk+0x5/0xfbef5
Mar 27 09:11:27 EeveeArch kernel:  ? srso_alias_return_thunk+0x5/0xfbef5
Mar 27 09:11:27 EeveeArch kernel:  ? preempt_count_add+0x55/0xd0
Mar 27 09:11:27 EeveeArch kernel:  ? srso_alias_return_thunk+0x5/0xfbef5
Mar 27 09:11:27 EeveeArch kernel:  ? migrate_enable+0xf6/0x120
Mar 27 09:11:27 EeveeArch kernel:  ? __pfx_kthread+0x10/0x10
Mar 27 09:11:27 EeveeArch kernel:  ret_from_fork+0x31/0x50
Mar 27 09:11:27 EeveeArch kernel:  ? __pfx_kthread+0x10/0x10
Mar 27 09:11:27 EeveeArch kernel:  ret_from_fork_asm+0x1a/0x30
Mar 27 09:11:27 EeveeArch kernel:  </TASK>
Mar 27 09:11:27 EeveeArch kernel: ---[ end trace 0000000000000000 ]---

System specs: https://pastebin.com/6X6C2Y3s
Other system specs: https://pastebin.com/uXiRzatd

Last edited by Markix (2026-03-29 10:31:08)

5hridhyan

Member

From: Asia

Registered: 2025-12-25

Posts: 537

Re: AMD Graphics dying seemingly at random

may I ask you any specific reason, using rt kernel? Can you check the behavior on a LTS/Normal Kernel??

EeveeArch kernel: amdgpu 0000:c2:00.0: amdgpu: SMU: I'm not done with your previous command: SMN_C2PMSG_66:0x0000001A SMN_C2PMSG_82:0x00000000
EeveeArch kernel: amdgpu 0000:c2:00.0: amdgpu: Failed to disable gfxoff!

if still the same in normal/LTS Kernels then boot with `amdgpu.gfxoff=0` as https://wiki.archlinux.org/title/Kernel_parameters, this disables the GFXOFF power-saving feature, I believe this prevents GPU resets caused by failed SMU/GFXOFF commands. Any changes in behavior? if not boot with `amdgpu.ppfeaturemask=0xffffbfff`, This disables a specific power-play feature which I guess sometimes causes instability on newer AMD APUs... basically you’re masking out the feature that may trigger the reset. use this only if gfxoff=0 isn’t enough...  also please post full journal logs, for such runs

---

---

Markix

Member

Registered: 2025-02-23

Posts: 2

Re: AMD Graphics dying seemingly at random

may I ask you any specific reason, using rt kernel?

I'm a music artist/producer! And some of the things I use my laptop for include sketching out musical ideas in Ardour for when I'm not at home or running the PA system at my local theater club.

I' m gonna try the the stock kernel for a few days and give an update if anything changes!.

Last edited by Markix (2026-04-10 21:04:41)