Yrurtj

I just bought a new desktop PC and had a freeze a week ago. Looking into the kernel log reveals a soft lockup (from journalctl -xa):

Mar 03 23:48:55 xxx kernel: watchdog: BUG: soft lockup - CPU#6 stuck for 23s! [systemd:1]

Mar 03 23:48:55 xxx kernel: Modules linked in: nls_iso8859_1 nls_cp437 vfat fat amdgpu chash amd_iommu_v2 snd_hda_codec_realtek gpu_sched snd_hda_codec_generic snd_hda_codec_hdmi ttm snd_hda_intel drm_kms_helper snd_hda_codec edac_mce_amd eeepc_wmi asus_wmi snd_hda_core sparse_keymap drm snd_hwdep kvm_amd rfkill mxm_wmi wmi_bmof snd_pcm kvm igb agpgart syscopyarea snd_timer sysfillrect sysimgblt sp5100_tco i2c_algo_bit mousedev joydev input_leds fb_sys_fops dca i2c_piix4 pcspkr snd irqbypass k10temp soundcore pinctrl_amd wmi evdev gpio_amdpt pcc_cpufreq mac_hid acpi_cpufreq ip_tables x_tables ext4 crc32c_generic crc16 mbcache jbd2 fscrypto algif_skcipher af_alg dm_crypt dm_mod sr_mod cdrom hid_holtek_mouse hid_cherry hid_generic usbhid hid sd_mod crct10dif_pclmul crc32_pclmul crc32c_intel ghash_clmulni_intel ahci libahci aesni_intel libata aes_x86_64 crypto_simd cryptd xhci_pci ccp glue_helper xhci_hcd scsi_mod rng_core

Mar 03 23:48:55 xxx kernel: CPU: 6 PID: 1 Comm: systemd Tainted: G             L    4.20.13-arch1-1-ARCH #1

Mar 03 23:48:55 xxx kernel: Hardware name: System manufacturer System Product Name/PRIME X370-PRO, BIOS 4207 12/08/2018

Mar 03 23:48:55 xxx kernel: RIP: 0010:smp_call_function_many+0x1f4/0x250

Mar 03 23:48:55 xxx kernel: Code: c7 e8 60 b1 6f 00 3b 05 6e 3d 21 01 0f 83 8f fe ff ff 48 63 d0 48 8b 0b 48 03 0c d5 20 c8 53 91 8b 51 18 83 e2 01 74 0a f3 90 <8b> 51 18 83 e2 01 75 f6 eb c8 48 c7 c2 60 47 73 91 4c 89 e6 89 df

Mar 03 23:48:55 xxx kernel: RSP: 0018:ffff99e08004bb40 EFLAGS: 00000202 ORIG_RAX: ffffffffffffff13

Mar 03 23:48:55 xxx kernel: RAX: 0000000000000002 RBX: ffff9481ceba2bc0 RCX: ffff9481ceaa7ac0

Mar 03 23:48:55 xxx kernel: RDX: 0000000000000001 RSI: 0000000000000000 RDI: ffff9481ceba2bc8

Mar 03 23:48:55 xxx kernel: RBP: ffff9481ceba2bc8 R08: 0000000000000000 R09: 0000000000000000

Mar 03 23:48:55 xxx kernel: R10: ffff9481ceba2bc8 R11: 0000000000000005 R12: ffff9481ceba2bf0

Mar 03 23:48:55 xxx kernel: R13: ffffffff90476530 R14: 0000000000000000 R15: 0000000000000140

Mar 03 23:48:55 xxx kernel: FS:  00007f7111232e80(0000) GS:ffff9481ceb80000(0000) knlGS:0000000000000000

Mar 03 23:48:55 xxx kernel: CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033

Mar 03 23:48:55 xxx kernel: CR2: 00005574b01882c8 CR3: 0000000405a68000 CR4: 00000000003406e0

Mar 03 23:48:55 xxx kernel: Call Trace:

Mar 03 23:48:55 xxx kernel:  ? tlbflush_read_file+0x70/0x70

Mar 03 23:48:55 xxx kernel:  smp_call_function+0x36/0x60

Mar 03 23:48:55 xxx kernel:  ? tlbflush_read_file+0x70/0x70

Mar 03 23:48:55 xxx kernel:  on_each_cpu+0x2a/0x80

Mar 03 23:48:55 xxx kernel:  flush_tlb_kernel_range+0x48/0x90

Mar 03 23:48:55 xxx kernel:  ? cpumask_next+0x16/0x20

Mar 03 23:48:55 xxx kernel:  ? purge_fragmented_blocks_allcpus+0x45/0x1d0

Mar 03 23:48:55 xxx kernel:  __purge_vmap_area_lazy+0x4d/0xc0

Mar 03 23:48:55 xxx kernel:  vm_unmap_aliases+0xf0/0x120

Mar 03 23:48:55 xxx kernel:  change_page_attr_set_clr+0xc8/0x2d0

Mar 03 23:48:55 xxx kernel:  ? bpf_jit_binary_alloc+0x60/0xe0

Mar 03 23:48:55 xxx kernel:  set_memory_ro+0x26/0x30

Mar 03 23:48:55 xxx kernel:  bpf_int_jit_compile+0x237/0x2fe

Mar 03 23:48:55 xxx kernel:  bpf_prog_select_runtime+0xa5/0xf0

Mar 03 23:48:55 xxx kernel:  bpf_prog_load+0x37a/0x5a0

Mar 03 23:48:55 xxx kernel:  ? lpm_trie_node_alloc+0x32/0x90

Mar 03 23:48:55 xxx kernel:  ? lpm_trie_node_alloc+0x58/0x90

Mar 03 23:48:55 xxx kernel:  ? _raw_spin_unlock_irqrestore+0x20/0x40

Mar 03 23:48:55 xxx kernel:  ? trie_update_elem+0x20b/0x340

Mar 03 23:48:55 xxx kernel:  __se_sys_bpf+0x5c6/0x16d0

Mar 03 23:48:55 xxx kernel:  do_syscall_64+0x5b/0x170

Mar 03 23:48:55 xxx kernel:  entry_SYSCALL_64_after_hwframe+0x44/0xa9

Mar 03 23:48:55 xxx kernel: RIP: 0033:0x7f71132534ed

Mar 03 23:48:55 xxx kernel: Code: 00 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 43 79 0c 00 f7 d8 64 89 01 48

Mar 03 23:48:55 xxx kernel: RSP: 002b:00007fff04946e98 EFLAGS: 00000246 ORIG_RAX: 0000000000000141

Mar 03 23:48:55 xxx kernel: RAX: ffffffffffffffda RBX: 00005574b01c95d0 RCX: 00007f71132534ed

Mar 03 23:48:55 xxx kernel: RDX: 0000000000000048 RSI: 00007fff04946ea0 RDI: 0000000000000005

Mar 03 23:48:55 xxx kernel: RBP: 0000000000000000 R08: 656369767265732e R09: 0000002500000008

Mar 03 23:48:55 xxx kernel: R10: 0000000000000000 R11: 0000000000000246 R12: 00005574b01ca7c0

Mar 03 23:48:55 xxx kernel: R13: 0000000000000001 R14: 00005574afa8412a R15: 0000000000000000

All necessary info should be in there, but feel free to ask for more details.

I think this easily should not happen, and I don't know how to deal with it. Since arch is running on kernel 5.0 by now, I was unsure whether it's worth to investigate. A freeze occurred again just a few minutes ago (sadly, this time without any useful kernel logs for the last 1.5h of running). I greped for "watchdog" after the reboot and found the following (don't know if it might help) entries after reboot:

Mar 10 22:03:48 xxx kernel: NMI watchdog: Enabled. Permanently consumes one hw-PMU counter.

Mar 10 22:03:48 xxx kernel: sp5100_tco: SP5100/SB800 TCO WatchDog Timer Driver

Mar 10 22:03:48 xxx kernel: sp5100-tco sp5100-tco: Using 0xfed80b00 for watchdog MMIO address

Mar 10 22:03:48 xxx kernel: sp5100-tco sp5100-tco: Watchdog hardware is disabled

Mar 10 22:05:10 xxx rtkit-daemon[1392]: Watchdog thread running.

Is this related to the strange artifacts I get when switching to another desktop or opening a new window (see my other question)?