Аппаратное ускорение декодирования отключено, потому что работало коряво.
Плеер стоит на паузе и свёрнут.
В произвольный момент времени в лог начинает сраться бесконечное
Aug 7 00:03:44 RERE-turk kernel: [76742.076355] amdgpu 0000:05:00.0: amdgpu: [mmhub0] no-retry page fault (src_id:0 ring:40 vmid:7 pasid:32784, for process vlc pid 1852235 thread vlc:cs0 pid 1852470)
Aug 7 00:03:44 RERE-turk kernel: [76742.076370] amdgpu 0000:05:00.0: amdgpu: in page starting at address 0x0000800101a10000 from IH client 0x12 (VMC)
Aug 7 00:03:44 RERE-turk kernel: [76742.076378] amdgpu 0000:05:00.0: amdgpu: VM_L2_PROTECTION_FAULT_STATUS:0x00740051
Aug 7 00:03:44 RERE-turk kernel: [76742.076381] amdgpu 0000:05:00.0: amdgpu: Faulty UTCL2 client ID: MP1 (0x0)
Aug 7 00:03:44 RERE-turk kernel: [76742.076384] amdgpu 0000:05:00.0: amdgpu: MORE_FAULTS: 0x1
Aug 7 00:03:44 RERE-turk kernel: [76742.076386] amdgpu 0000:05:00.0: amdgpu: WALKER_ERROR: 0x0
Aug 7 00:03:44 RERE-turk kernel: [76742.076388] amdgpu 0000:05:00.0: amdgpu: PERMISSION_FAULTS: 0x5
Aug 7 00:03:44 RERE-turk kernel: [76742.076390] amdgpu 0000:05:00.0: amdgpu: MAPPING_ERROR: 0x0
Aug 7 00:03:44 RERE-turk kernel: [76742.076392] amdgpu 0000:05:00.0: amdgpu: RW: 0x1
Через несколько минут ради пущего веселья сабж пытается починить то, что не сломано, и ломает графику напрочь:
Aug 7 00:03:54 RERE-turk kernel: [76752.173003] [drm:amdgpu_job_timedout [amdgpu]] *ERROR* ring sdma0 timeout, signaled seq=428330, emitted seq=428331
Aug 7 00:03:54 RERE-turk kernel: [76752.173156] [drm:amdgpu_job_timedout [amdgpu]] *ERROR* Process information: process vlc pid 1852235 thread vlc:cs0 pid 1852470
Aug 7 00:03:54 RERE-turk kernel: [76752.173275] amdgpu 0000:05:00.0: amdgpu: GPU reset begin!
Aug 7 00:03:54 RERE-turk kernel: [76752.405438] [drm] free PSP TMR buffer
Aug 7 00:03:54 RERE-turk kernel: [76752.431452] CPU: 14 PID: 1837166 Comm: kworker/u32:0 Tainted: P OE 5.19.0-50-generic #50-Ubuntu
Aug 7 00:03:54 RERE-turk kernel: [76752.431459] Hardware name: LENOVO 82L5/LNVNB161216, BIOS GSCN33WW 07/04/2022
Aug 7 00:03:54 RERE-turk kernel: [76752.431462] Workqueue: amdgpu-reset-dev drm_sched_job_timedout [gpu_sched]
Aug 7 00:03:54 RERE-turk kernel: [76752.431476] Call Trace:
Aug 7 00:03:54 RERE-turk kernel: [76752.431479] <TASK>
Aug 7 00:03:54 RERE-turk kernel: [76752.431482] show_stack+0x52/0x69
Aug 7 00:03:54 RERE-turk kernel: [76752.431491] dump_stack_lvl+0x49/0x6d
Aug 7 00:03:54 RERE-turk kernel: [76752.431498] dump_stack+0x10/0x18
Aug 7 00:03:54 RERE-turk kernel: [76752.431504] amdgpu_do_asic_reset+0x2b/0x441 [amdgpu]
Aug 7 00:03:54 RERE-turk kernel: [76752.431912] amdgpu_device_gpu_recover_imp.cold+0x4f6/0x805 [amdgpu]
Aug 7 00:03:54 RERE-turk kernel: [76752.432283] amdgpu_job_timedout+0x15e/0x190 [amdgpu]
Aug 7 00:03:54 RERE-turk kernel: [76752.432650] ? finish_task_switch.isra.0+0x84/0x290
Aug 7 00:03:54 RERE-turk kernel: [76752.432657] drm_sched_job_timedout+0x6a/0x120 [gpu_sched]
Aug 7 00:03:54 RERE-turk kernel: [76752.432664] process_one_work+0x21c/0x400
Aug 7 00:03:54 RERE-turk kernel: [76752.432669] worker_thread+0x50/0x3f0
Aug 7 00:03:54 RERE-turk kernel: [76752.432673] ? rescuer_thread+0x3a0/0x3a0
Aug 7 00:03:54 RERE-turk kernel: [76752.432675] kthread+0xeb/0x120
Aug 7 00:03:54 RERE-turk kernel: [76752.432680] ? kthread_complete_and_exit+0x20/0x20
Aug 7 00:03:54 RERE-turk kernel: [76752.432684] ret_from_fork+0x1f/0x30
Aug 7 00:03:54 RERE-turk kernel: [76752.432692] </TASK>
Aug 7 00:03:54 RERE-turk kernel: [76752.432707] amdgpu 0000:05:00.0: amdgpu: MODE2 reset
Aug 7 00:03:54 RERE-turk kernel: [76752.432792] amdgpu 0000:05:00.0: amdgpu: GPU reset succeeded, trying to resume
Aug 7 00:03:54 RERE-turk kernel: [76752.433013] [drm] PCIE GART of 1024M enabled.
Aug 7 00:03:54 RERE-turk kernel: [76752.433019] [drm] PTB located at 0x000000F400FA0000
Aug 7 00:03:54 RERE-turk kernel: [76752.433036] [drm] VRAM is lost due to GPU reset!
Aug 7 00:03:54 RERE-turk kernel: [76752.433039] [drm] PSP is resuming...
Aug 7 00:03:54 RERE-turk kernel: [76752.452919] [drm] reserve 0x400000 from 0xf47fb00000 for PSP TMR
Aug 7 00:03:55 RERE-turk kernel: [76752.712977] amdgpu 0000:05:00.0: amdgpu: RAS: optional ras ta ucode is not available
Aug 7 00:03:55 RERE-turk kernel: [76752.723675] amdgpu 0000:05:00.0: amdgpu: RAP: optional rap ta ucode is not available
Aug 7 00:03:55 RERE-turk kernel: [76752.723681] amdgpu 0000:05:00.0: amdgpu: SECUREDISPLAY: securedisplay ta ucode is not available
Aug 7 00:03:55 RERE-turk kernel: [76752.723686] amdgpu 0000:05:00.0: amdgpu: SMU is resuming...
Aug 7 00:03:55 RERE-turk kernel: [76752.724205] amdgpu 0000:05:00.0: amdgpu: SMU is resumed successfully!
Aug 7 00:03:55 RERE-turk kernel: [76752.724753] [drm] DMUB hardware initialized: version=0x0101001F
Aug 7 00:03:55 RERE-turk kernel: [76753.118193] [drm] kiq ring mec 2 pipe 1 q 0
Aug 7 00:03:55 RERE-turk kernel: [76753.344452] amdgpu 0000:05:00.0: [drm:amdgpu_ring_test_helper [amdgpu]] *ERROR* ring sdma0 test failed (-110)
Aug 7 00:03:55 RERE-turk kernel: [76753.344813] [drm:amdgpu_device_ip_resume_phase2 [amdgpu]] *ERROR* resume of IP block <sdma_v4_0> failed -110
Aug 7 00:03:55 RERE-turk kernel: [76753.345136] amdgpu 0000:05:00.0: amdgpu: GPU reset(1) failed
Aug 7 00:03:55 RERE-turk kernel: [76753.345202] amdgpu 0000:05:00.0: amdgpu: GPU reset end with ret = -110
Aug 7 00:03:55 RERE-turk kernel: [76753.345210] [drm:amdgpu_job_timedout [amdgpu]] *ERROR* GPU Recovery Failed: -110
В интернетах нашёл только советы вида «обнови ядро». Сейчас стоит 6.2.0-26-generic, более нового в репе нет.
Если в светлом будущем будут существовать компьютеры, я буду пользоваться чем угодно, кроме продукции AMD. Сколько же, блин, можно.