A Possible Workaround For AMD APUs With Stability Issues On Recent Kernels

[phoronix]

Phoronix: A Possible Workaround For AMD APUs With Stability Issues On Recent Kernels

While we have found recent Linux kernels paired with latest motherboard BIOS releases to work out generally well for recent AMD APUs, not everyone has been having a trouble-free experience on recent kernels. But an affected user has discovered a possible workaround if hitting stability issues...

http://www.phoronix.com/scan.php?page=news_item&px=AMDGPU-APU-noretry

[amedeo]

I have a laptop with amd apu 2500U and since the kernel 5.4.0 through 5.4.6 I was unable to start plasma DE (and for this I kept latest 5.3), but with the latest kernel 5.4.7 I saw a lot of patches on kernel changelogs regarding amdgpu driver, and building that I was able to boot kernel without issues (since now ).

Anyone still has problems with the kernel 5.4.7 and amd gpu?

$ cat /sys/module/amdgpu/parameters/noretry
1

[josh_walrath]

"APU"? W-what's that??

[S.Pam]

Did not work for my AMD Athlon 3000G APU =(

[ 7766.244363] [drm] amdgpu kernel modesetting enabled.
[ 7766.244398] Parsing CRAT table with 1 nodes
[ 7766.244400] Creating topology SYSFS entries
[ 7766.244413] Topology: Add APU node [0x0:0x0]
[ 7766.244413] Finished initializing topology
[ 7766.244467] amdgpu 0000:06:00.0: remove_conflicting_pci_framebuffers: bar 0: 0xe0000000 -> 0xefffffff
[ 7766.244468] amdgpu 0000:06:00.0: remove_conflicting_pci_framebuffers: bar 2: 0xf0000000 -> 0xf01fffff
[ 7766.244469] amdgpu 0000:06:00.0: remove_conflicting_pci_framebuffers: bar 5: 0xfce00000 -> 0xfce7ffff
[ 7766.244470] checking generic (e0000000 7f0000) vs hw (e0000000 10000000)
[ 7766.244471] fb0: switching to amdgpudrmfb from VESA VGA
[ 7766.244554] Console: switching to colour dummy device 80x25
[ 7766.244584] amdgpu 0000:06:00.0: vgaarb: deactivate vga console
[ 7766.244813] [drm] initializing kernel modesetting (RAVEN 0x1002:0x15D8 0x1002:0x15D8 0xCC).
[ 7766.244827] [drm] register mmio base: 0xFCE00000
[ 7766.244827] [drm] register mmio size: 524288
[ 7766.244846] [drm] add ip block number 0 <soc15_common>
[ 7766.244847] [drm] add ip block number 1 <gmc_v9_0>
[ 7766.244847] [drm] add ip block number 2 <vega10_ih>
[ 7766.244848] [drm] add ip block number 3 <psp>
[ 7766.244848] [drm] add ip block number 4 <gfx_v9_0>
[ 7766.244849] [drm] add ip block number 5 <sdma_v4_0>
[ 7766.244850] [drm] add ip block number 6 <powerplay>
[ 7766.244850] [drm] add ip block number 7 <dm>
[ 7766.244851] [drm] add ip block number 8 <vcn_v1_0>
[ 7766.270326] [drm] BIOS signature incorrect 0 0
[ 7766.270350] ATOM BIOS: 113-RAVEN2-115
[ 7766.271234] [drm] VCN decode is enabled in VM mode
[ 7766.271235] [drm] VCN encode is enabled in VM mode
[ 7766.271235] [drm] VCN jpeg decode is enabled in VM mode
[ 7766.271267] [drm] vm size is 262144 GB, 3 levels, block size is 9-bit, fragment size is 9-bit
[ 7766.271273] amdgpu 0000:06:00.0: VRAM: 2048M 0x000000F400000000 - 0x000000F47FFFFFFF (2048M used)
[ 7766.271274] amdgpu 0000:06:00.0: GART: 1024M 0x0000000000000000 - 0x000000003FFFFFFF
[ 7766.271275] amdgpu 0000:06:00.0: AGP: 267419648M 0x000000F800000000 - 0x0000FFFFFFFFFFFF
[ 7766.271278] [drm] Detected VRAM RAM=2048M, BAR=2048M
[ 7766.271278] [drm] RAM width 128bits DDR4
[ 7766.271325] [TTM] Zone  kernel: Available graphics memory: 3052426 KiB
[ 7766.271326] [TTM] Zone   dma32: Available graphics memory: 2097152 KiB
[ 7766.271326] [TTM] Initializing pool allocator
[ 7766.271331] [TTM] Initializing DMA pool allocator
[ 7766.271382] [drm] amdgpu: 2048M of VRAM memory ready
[ 7766.271384] [drm] amdgpu: 3072M of GTT memory ready.
[ 7766.271392] [drm] GART: num cpu pages 262144, num gpu pages 262144
[ 7766.271526] [drm] PCIE GART of 1024M enabled (table at 0x000000F400900000).
[ 7766.278715] [drm] use_doorbell being set to: [true]
[ 7766.278768] amdgpu: [powerplay] hwmgr_sw_init smu backed is smu10_smu
[ 7766.281086] [drm] Found VCN firmware Version: 1.86 Family ID: 18
[ 7766.281091] [drm] PSP loading VCN firmware
[ 7766.301791] [drm] reserve 0x400000 from 0xf47fc00000 for PSP TMR
[ 7766.358099] [drm] failed to load ucode id (18)
[ 7766.358101] [drm] psp command failed and response status is (0x300F)
[ 7766.361093] [drm] failed to load ucode id (19)
[ 7766.361094] [drm] psp command failed and response status is (0xF)
[ 7766.364093] [drm] failed to load ucode id (20)
[ 7766.364095] [drm] psp command failed and response status is (0xF)
[ 7766.384556] [drm] DM_PPLIB: values for F clock
[ 7766.384558] [drm] DM_PPLIB:   0 in kHz, 3649 in mV
[ 7766.384559] [drm] DM_PPLIB:   0 in kHz, 0 in mV
[ 7766.384559] [drm] DM_PPLIB:   0 in kHz, 0 in mV
[ 7766.384560] [drm] DM_PPLIB:   0 in kHz, 0 in mV
[ 7766.384577] ------------[ cut here ]------------
[ 7766.384733] WARNING: CPU: 0 PID: 11164 at drivers/gpu/drm/amd/amdgpu/../display/dc/calcs/dcn_calcs.c:1464 dcn_bw_update_from_pplib+0x94/0x2c0 [amdgpu]
[ 7766.384734] Modules linked in: amdgpu(+) gpu_sched ttm l2tp_netlink l2tp_core ip_set_hash_ip xt_state ip6t_REJECT nf_reject_ipv6 ip6table_filter ip6table_raw ip6table_mangle xt_multiport ip6table_nat nfnetlink_log xt_limit xt_NFLOG ipt_REJECT nf_reject_ipv4 xt_conntrack iptable_filter iptable_mangle xt_nat iptable_nat xt_CT iptable_raw ip_set_bitmap_port ip_set_hash_net nf_nat_pptp nf_conntrack_pptp nf_nat xt_sctp nf_conntrack_sip nf_conntrack_irc nf_conntrack_ftp nf_conntrack_h323 nf_conntrack_netbios_ns nf_conntrack_broadcast nf_conntrack_bridge nf_conntrack nf_defrag_ipv6 ip6_tables ip_tables xt_recent xt_set ip_set nfnetlink nf_defrag_ipv4 nf_socket_ipv4 uas pinctrl_amd
[ 7766.384760] CPU: 0 PID: 11164 Comm: modprobe Not tainted 5.4.7-gentoo-test2 #3
[ 7766.384761] Hardware name: Gigabyte Technology Co., Ltd. B450M DS3H/B450M DS3H-CF, BIOS F50 11/27/2019
[ 7766.384901] RIP: 0010:dcn_bw_update_from_pplib+0x94/0x2c0 [amdgpu]
[ 7766.384904] Code: 0c 24 85 c9 74 24 8d 71 ff 48 8d 44 24 04 48 8d 54 f4 0c eb 0d 48 83 c0 08 48 39 d0 0f 84 13 01 00 00 44 8b 00 45 85 c0 75 eb <0f> 0b e8 65 3e d3 e0 4c 89 e2 be 04 00 00 00 4c 89 ef e8 a5 9b fe
[ 7766.384905] RSP: 0018:ffffc90003f5b700 EFLAGS: 00010246
[ 7766.384907] RAX: ffffc90003f5b704 RBX: ffff888205c20000 RCX: 0000000000000004
[ 7766.384908] RDX: ffffc90003f5b724 RSI: 0000000000000003 RDI: ffff888218816350
[ 7766.384909] RBP: ffffc90003f5b840 R08: 0000000000000000 R09: 00000000000003e3
[ 7766.384910] R10: 0000000000000001 R11: 0000000000000000 R12: ffffc90003f5b790
[ 7766.384911] R13: ffff888207637d80 R14: 0000000000000001 R15: 000000000000000b
[ 7766.384913] FS:  00007f2ba4206b80(0000) GS:ffff888218800000(0000) knlGS:0000000000000000
[ 7766.384914] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 7766.384915] CR2: 00005555bc5824f0 CR3: 0000000205d0a000 CR4: 00000000003406f0
[ 7766.384916] Call Trace:
[ 7766.384924]  ? kmem_cache_alloc+0xe6/0x180
[ 7766.385061]  dcn10_create_resource_pool+0x7d9/0xb10 [amdgpu]
[ 7766.385197]  ? firmware_parser_create+0x6fb/0x720 [amdgpu]
[ 7766.385326]  dc_create_resource_pool+0x21/0x100 [amdgpu]
[ 7766.385454]  dc_create+0x206/0x680 [amdgpu]
[ 7766.385457]  ? kmem_cache_alloc+0xe6/0x180
[ 7766.385591]  amdgpu_dm_init+0x138/0x1c0 [amdgpu]
[ 7766.385722]  ? phm_wait_for_register_unequal.part.0+0x44/0x70 [amdgpu]
[ 7766.385854]  dm_hw_init+0x9/0x20 [amdgpu]
[ 7766.385986]  amdgpu_device_init.cold+0xf47/0x129e [amdgpu]
[ 7766.385989]  ? __alloc_pages_nodemask+0x128/0x240
[ 7766.386096]  amdgpu_driver_load_kms+0x44/0xe0 [amdgpu]
[ 7766.386101]  drm_dev_register+0x109/0x150
[ 7766.386206]  amdgpu_pci_probe+0xe9/0x150 [amdgpu]
[ 7766.386210]  ? __pm_runtime_resume+0x44/0x50
[ 7766.386214]  local_pci_probe+0x38/0x70
[ 7766.386216]  ? pci_match_device+0xd2/0x100
[ 7766.386218]  pci_device_probe+0xe4/0x190
[ 7766.386221]  really_probe+0xdf/0x290
[ 7766.386224]  driver_probe_device+0x4b/0xc0
[ 7766.386226]  device_driver_attach+0x4e/0x60
[ 7766.386229]  __driver_attach+0x44/0xb0
[ 7766.386231]  ? device_driver_attach+0x60/0x60
[ 7766.386232]  bus_for_each_dev+0x5c/0x90
[ 7766.386235]  bus_add_driver+0x16d/0x1c0
[ 7766.386237]  driver_register+0x67/0xb0
[ 7766.386238]  ? 0xffffffffa05a7000
[ 7766.386241]  do_one_initcall+0x44/0x16f
[ 7766.386244]  ? __vunmap+0x223/0x260
[ 7766.386246]  ? kmem_cache_alloc+0xe6/0x180
[ 7766.386250]  do_init_module+0x51/0x200
[ 7766.386252]  load_module+0x20d6/0x23d0
[ 7766.386256]  ? vfs_read+0x117/0x140
[ 7766.386259]  ? __do_sys_finit_module+0x9b/0xb0
[ 7766.386261]  __do_sys_finit_module+0x9b/0xb0
[ 7766.386263]  do_syscall_64+0x3d/0x100
[ 7766.386267]  entry_SYSCALL_64_after_hwframe+0x44/0xa9
[ 7766.386269] RIP: 0033:0x7f2ba4327289
[ 7766.386272] Code: 00 00 00 75 05 48 83 c4 18 c3 e8 c2 5f 01 00 66 90 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 d7 4b 09 00 f7 d8 64 89 01 48
[ 7766.386273] RSP: 002b:00007ffd8bd5b718 EFLAGS: 00000246 ORIG_RAX: 0000000000000139
[ 7766.386275] RAX: ffffffffffffffda RBX: 00005558b1ba0ad0 RCX: 00007f2ba4327289
[ 7766.386276] RDX: 0000000000000000 RSI: 00005558b1ba10f0 RDI: 0000000000000005
[ 7766.386277] RBP: 0000000000040000 R08: 0000000000000000 R09: 00005558b1ba0cc0
[ 7766.386278] R10: 0000000000000005 R11: 0000000000000246 R12: 00005558b1ba10f0
[ 7766.386278] R13: 0000000000000000 R14: 00005558b1ba0c00 R15: 00005558b1ba0ad0
[ 7766.386280] ---[ end trace 0b1d2ec02ad9ff2e ]---
[ 7766.386294] [drm] DM_PPLIB: values for DCF clock
[ 7766.386295] [drm] DM_PPLIB:   300000 in kHz, 3649 in mV
[ 7766.386296] [drm] DM_PPLIB:   600000 in kHz, 3974 in mV
[ 7766.386297] [drm] DM_PPLIB:   626000 in kHz, 4174 in mV
[ 7766.386298] [drm] DM_PPLIB:   654000 in kHz, 4325 in mV
[ 7766.388514] [drm] Display Core initialized with v3.2.48!
[ 7766.388884] [drm] Supports vblank timestamp caching Rev 2 (21.10.2013).
[ 7766.388885] [drm] Driver supports precise vblank timestamp query.
[ 7766.400748] [drm] VCN decode and encode initialized successfully(under SPG Mode).
[ 7766.402438] kfd kfd: Allocated 3969056 bytes on gart
[ 7766.561624] AMD-Vi: Completion-Wait loop timed out
[ 7766.702125] AMD-Vi: Completion-Wait loop timed out
[ 7766.842496] AMD-Vi: Completion-Wait loop timed out
[ 7766.982830] AMD-Vi: Completion-Wait loop timed out
[ 7767.109235] AMD-Vi: Completion-Wait loop timed out
[ 7767.235633] AMD-Vi: Completion-Wait loop timed out
[ 7767.361947] AMD-Vi: Completion-Wait loop timed out
[ 7767.405232] iommu ivhd0: AMD-Vi: Event logged [IOTLB_INV_TIMEOUT device=06:00.0 address=0x217878ea0]
[ 7767.405289] kfd kfd: Failed to resume IOMMU for device 1002:15d8
[ 7767.542717] AMD-Vi: Completion-Wait loop timed out
[ 7767.669016] AMD-Vi: Completion-Wait loop timed out
[ 7767.806977] AMD-Vi: Completion-Wait loop timed out
[ 7767.938609] AMD-Vi: Completion-Wait loop timed out
[ 7768.069527] AMD-Vi: Completion-Wait loop timed out
[ 7768.069654] e1000e 0000:03:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0001 address=0xfb93cbc0 flags=0x0050]
[ 7768.069657] e1000e 0000:03:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0001 address=0xfb93cc00 flags=0x0050]
[ 7768.069659] e1000e 0000:03:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0001 address=0xfdb487c0 flags=0x0050]
[ 7768.069662] e1000e 0000:03:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0001 address=0xfdb48800 flags=0x0050]
[ 7768.069664] e1000e 0000:03:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0001 address=0xfc1d2340 flags=0x0050]
[ 7768.069666] e1000e 0000:03:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0001 address=0xfc1d2380 flags=0x0050]
[ 7768.208963] AMD-Vi: Completion-Wait loop timed out
[ 7768.345312] AMD-Vi: Completion-Wait loop timed out
[ 7768.345339] e1000e 0000:03:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0001 address=0xfdb487c0 flags=0x0050]
[ 7768.345342] e1000e 0000:03:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0001 address=0xfdb48800 flags=0x0050]
[ 7768.345345] e1000e 0000:03:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0001 address=0xfc1d2340 flags=0x0050]
[ 7768.345348] e1000e 0000:03:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0001 address=0xfc1d2380 flags=0x0050]
[ 7768.345350] AMD-Vi: Event logged [IO_PAGE_FAULT device=03:00.0 domain=0x0001 address=0xfb93cbc0 flags=0x0050]
[ 7768.345354] AMD-Vi: Event logged [IO_PAGE_FAULT device=03:00.0 domain=0x0001 address=0xfb93cc00 flags=0x0050]
[ 7768.345357] AMD-Vi: Event logged [IO_PAGE_FAULT device=03:00.0 domain=0x0001 address=0xfdb487c0 flags=0x0050]
[ 7768.345359] AMD-Vi: Event logged [IO_PAGE_FAULT device=03:00.0 domain=0x0001 address=0xfdb48800 flags=0x0050]
[ 7768.345361] AMD-Vi: Event logged [IO_PAGE_FAULT device=03:00.0 domain=0x0001 address=0xfc1d2340 flags=0x0050]
[ 7768.345364] AMD-Vi: Event logged [IO_PAGE_FAULT device=03:00.0 domain=0x0001 address=0xfc1d2380 flags=0x0050]
[ 7768.345366] AMD-Vi: Event logged [IO_PAGE_FAULT device=03:00.0 domain=0x0001 address=0xfb93cbc0 flags=0x0050]
[ 7768.345368] AMD-Vi: Event logged [IO_PAGE_FAULT device=03:00.0 domain=0x0001 address=0xfb93cc00 flags=0x0050]
[ 7768.345370] AMD-Vi: Event logged [IO_PAGE_FAULT device=03:00.0 domain=0x0001 address=0xfdb487c0 flags=0x0050]
[ 7768.345373] AMD-Vi: Event logged [IO_PAGE_FAULT device=03:00.0 domain=0x0001 address=0xfdb48800 flags=0x0050]
[ 7768.345573] e1000e 0000:05:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0001 address=0xfd2581f0 flags=0x0050]
[ 7768.345576] e1000e 0000:05:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0001 address=0xfd258200 flags=0x0050]
[ 7768.345578] e1000e 0000:05:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0001 address=0xfd258280 flags=0x0050]
[ 7768.345581] e1000e 0000:05:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0001 address=0xfd258300 flags=0x0050]
[ 7768.345584] e1000e 0000:05:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0001 address=0xfd258380 flags=0x0050]
[ 7768.345604] e1000e 0000:05:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0001 address=0xfd2583f0 flags=0x0050]
[ 7768.345607] e1000e 0000:05:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0001 address=0xfd258400 flags=0x0050]
[ 7768.345609] e1000e 0000:05:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0001 address=0xfd258480 flags=0x0050]
[ 7768.417988] e1000e 0000:05:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0001 address=0xfd2581f0 flags=0x0050]
[ 7768.417991] e1000e 0000:05:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0001 address=0xfd258280 flags=0x0050]
[ 7768.418025] iommu ivhd0: AMD-Vi: Event logged [IOTLB_INV_TIMEOUT device=06:00.0 address=0x217878fd0]
[ 7768.628973] failed to write reg 28b4 wait reg 28c6
[ 7769.639244] failed to write reg 1a6f4 wait reg 1a706
[ 7769.797215] AMD-Vi: Completion-Wait loop timed out
[ 7769.937621] AMD-Vi: Completion-Wait loop timed out
[ 7770.077813] AMD-Vi: Completion-Wait loop timed out
[ 7770.213245] AMD-Vi: Completion-Wait loop timed out
[ 7770.346275] AMD-Vi: Completion-Wait loop timed out
[ 7770.472762] AMD-Vi: Completion-Wait loop timed out
[ 7770.472877] kfd kfd: device 1002:15d8 NOT added due to errors
[ 7770.472927] [drm] Cannot find any crtc or sizes
[ 7770.473152] amdgpu 0000:06:00.0: ring gfx uses VM inv eng 0 on hub 0
[ 7770.473155] amdgpu 0000:06:00.0: ring comp_1.0.0 uses VM inv eng 1 on hub 0
[ 7770.473158] amdgpu 0000:06:00.0: ring comp_1.1.0 uses VM inv eng 4 on hub 0
[ 7770.473160] amdgpu 0000:06:00.0: ring comp_1.2.0 uses VM inv eng 5 on hub 0
[ 7770.473162] amdgpu 0000:06:00.0: ring comp_1.3.0 uses VM inv eng 6 on hub 0
[ 7770.473164] amdgpu 0000:06:00.0: ring comp_1.0.1 uses VM inv eng 7 on hub 0
[ 7770.473166] amdgpu 0000:06:00.0: ring comp_1.1.1 uses VM inv eng 8 on hub 0
[ 7770.473167] amdgpu 0000:06:00.0: ring comp_1.2.1 uses VM inv eng 9 on hub 0
[ 7770.473169] amdgpu 0000:06:00.0: ring comp_1.3.1 uses VM inv eng 10 on hub 0
[ 7770.473171] amdgpu 0000:06:00.0: ring kiq_2.1.0 uses VM inv eng 11 on hub 0
[ 7770.473173] amdgpu 0000:06:00.0: ring sdma0 uses VM inv eng 0 on hub 1
[ 7770.473175] amdgpu 0000:06:00.0: ring vcn_dec uses VM inv eng 1 on hub 1
[ 7770.473177] amdgpu 0000:06:00.0: ring vcn_enc0 uses VM inv eng 4 on hub 1
[ 7770.473179] amdgpu 0000:06:00.0: ring vcn_enc1 uses VM inv eng 5 on hub 1
[ 7770.473180] amdgpu 0000:06:00.0: ring vcn_jpeg uses VM inv eng 6 on hub 1
[ 7770.874100] AMD-Vi: Completion-Wait loop timed out
[ 7771.146463] AMD-Vi: Completion-Wait loop timed out
[ 7771.657229] hrtimer: interrupt took 819731944 ns
[ 7771.748326] iommu ivhd0: AMD-Vi: Event logged [IOTLB_INV_TIMEOUT device=06:00.0 address=0x217879600]
[ 7771.748327] iommu ivhd0: AMD-Vi: Event logged [IOTLB_INV_TIMEOUT device=06:00.0 address=0x217879650]
[ 7772.056622] AMD-Vi: Completion-Wait loop timed out
[ 7772.183439] AMD-Vi: Completion-Wait loop timed out
[ 7772.385872] AMD-Vi: Completion-Wait loop timed out
[ 7772.476956] clocksource: timekeeping watchdog on CPU2: Marking clocksource 'tsc' as unstable because the skew is too large:
[ 7772.476957] clocksource:                       'hpet' wd_now: e920488c wd_last: e8c474f2 mask: ffffffff
[ 7772.476959] clocksource:                       'tsc' cs_now: 18c12e78ba1a cs_last: 18c0e9eae84b mask: ffffffffffffffff
[ 7772.476960] tsc: Marking TSC unstable due to clocksource watchdog
[ 7772.603205] AMD-Vi: Completion-Wait loop timed out
[ 7772.785356] AMD-Vi: Completion-Wait loop timed out
[ 7773.205601] iommu ivhd0: AMD-Vi: Event logged [IOTLB_INV_TIMEOUT device=06:00.0 address=0x217879720]
[ 7773.296693] TSC found unstable after boot, most likely due to broken BIOS. Use 'tsc=unstable'.
[ 7773.296694] sched_clock: Marking unstable (7772776598458, 520105623)<-(7773300209551, -3517379)
[ 7773.661021] clocksource: Switched to clocksource hpet
[ 7829.857898] hpet_rtc_timer_reinit: 62 callbacks suppressed
[ 7829.857899] hpet: Lost 4 RTC interrupts
[ 7829.949053] hpet: Lost 5 RTC interrupts
[ 7830.040075] hpet: Lost 5 RTC interrupts
[ 7830.131139] hpet: Lost 5 RTC interrupts
[ 7830.223174] hpet: Lost 10 RTC interrupts
[ 7830.404374] hpet: Lost 4 RTC interrupts
[ 7830.495496] hpet: Lost 5 RTC interrupts
[ 7830.950856] hpet: Lost 4 RTC interrupts
[ 7831.042896] hpet: Lost 11 RTC interrupts
[ 7831.315236] hpet: Lost 11 RTC interrupts

[bezirg]

With my Amd ryzen 3400G APU, I run very often into problems of freezing after resuming from suspension. (`systemctl suspend`). I will try this workaround. Unfortunately I am not running the latest AGESA firmware BIOS, because my motherboard manufacturer is being slow in providing an up-to-date BIOS version.

[caligula]

Did not work for my AMD Athlon 3000G APU =(

[ 7766.244363]
[ 7769.797215] AMD-Vi: Completion-Wait loop timed out

Re ''AMD-Vi: Completion-Wait loop timed out'' -- try kernel parameter: iommu=pt

[r_a_trip]

"APU"? W-what's that??

An Accelerated Processing Unit. Also known as CPU with integrated GPU.

[S.Pam]

Re ''AMD-Vi: Completion-Wait loop timed out'' -- try kernel parameter: iommu=pt

Yea. I tried that. No change. Same problem. Also tried amdgpu.noretry=0.

[gukin]

I was having issues with 5.4 and Mesa 19.2, just intermittent hangs for a few seconds but once I went to Mesa 19.3 the first hang would hang the GPU, I could get my desktop back by logging in remotely and doing an sddm restart.

Happily this fix worked for me, well my 2400g, and I've had no hangs since adding amdgpu.noretry=0 to my kernel boot line and rebooting.

It's odd that these processors have been so temperamental, and this issue carried on with the 3400g. I'm looking forward to getting my Renior APU processor.

[R41N3R]

Thanks Michael for reporting about this workaround. It works quite well after doing some tests, no immediate crash like before on Plasma Wayland. It's quite unfortunate how this could happen though, I would expect some testing from AMD! We talk here about a released kernel and not about some development stuff. AMD really needs to invest some time for QC, none of the current AMD APU's worked without major issues for me.