Notizen zu meinem AI/KI Lab System (Minisforum MS-S1 Ryzen AI MAX 395+)

Debian Trixie (13.3) kommt (Stand Januar 2026) mit folgender Kernel- und Firmwareversion:

benny@ailab:~$ uname -a
Linux ailab 6.12.63+deb13-amd64 #1 SMP PREEMPT_DYNAMIC Debian 6.12.63-1 (2025-12-30) x86_64 GNU/Linux

benny@ailab:~$ dpkg -l | grep firmware
ii  amd64-microcode                       3.20250311.1                         amd64        Platform firmware and microcode for AMD CPUs and SoCs
ii  firmware-amd-graphics                 20250410-2                           all          Binary firmware for AMD/ATI graphics and NPU chips
ii  firmware-mediatek                     20250410-2                           all          Binary firmware for MediaTek and Ralink chips for networking, SoCs and media
ii  firmware-realtek                      20250410-2                           all          Binary firmware for Realtek network and audio chips

In seinem Video auf Youtube, teilt Donato Capitella mit uns welche Kernel- und Firmwareversion die Schwierigkeiten der „Strix Halo“-Plattform beheben:

linux-image-amd64 >= 6.18.4
firmware-amd-graphics >= 2026010
ROCm nightly (TheRock) or ROCm-7.2+ (when released)

Ich nutze ein Minisforum MS-S1 (Ryzen AI MAX 395+) und dieses System verwendet eine auf dem RTL8127 basierende onboard 10G-Netzwerkkarte. Für diese Karte ist ein Treiber ab Linux Kernel >= v6.16 verfügbar.

Eine Lösung für die Netzwerkkarte ist der Einsatz von „Testing“, da dort der Kernel 6.17 zur Verfügung steht. Es ist mir derzeit aber nicht bekannt ob der notwendige Patch aus 6.18.4 in 6.17.x zurückportiert wurde.

Mein Ansatz ist daher erst einmal auf den „Unstable“ Zweig für den Kernel zu setzen. Die Firmware muss derzeit wohl manuell ersetzt werden, da das Paket in „Unstable“ zu alt ist (Stand Januar 2026: firmware-amd-graphics (20251111-1))

Codename der GPU: gfx1151

Ausgabe von dmesg für amdgpu mit 6.12.63+deb13-amd64 & Firmware 20250410-2

root@ailab:~# dmesg |grep amdgpu
[    2.814500] [drm] amdgpu kernel modesetting enabled.
[    2.816621] amdgpu: Virtual CRAT table created for CPU
[    2.816627] amdgpu: Topology: Add CPU node
[    2.816723] amdgpu 0000:f4:00.0: enabling device (0006 -> 0007)
[    2.819967] amdgpu 0000:f4:00.0: amdgpu: Fetched VBIOS from VFCT
[    2.819968] amdgpu: ATOM BIOS: 113-STRXLGEN-001
[    2.825368] amdgpu 0000:f4:00.0: amdgpu: VPE: collaborate mode true
[    2.825499] amdgpu 0000:f4:00.0: amdgpu: Trusted Memory Zone (TMZ) feature disabled as experimental (default)
[    2.825558] amdgpu 0000:f4:00.0: amdgpu: VRAM: 2048M 0x0000008000000000 - 0x000000807FFFFFFF (2048M used)
[    2.825559] amdgpu 0000:f4:00.0: amdgpu: GART: 512M 0x00007FFF00000000 - 0x00007FFF1FFFFFFF
[    2.825689] [drm] amdgpu: 2048M of VRAM memory ready
[    2.825691] [drm] amdgpu: 63219M of GTT memory ready.
[    2.850304] amdgpu 0000:f4:00.0: amdgpu: reserve 0x8c00000 from 0x8060000000 for PSP TMR
[    3.184679] amdgpu 0000:f4:00.0: amdgpu: RAS: optional ras ta ucode is not available
[    3.188577] amdgpu 0000:f4:00.0: amdgpu: RAP: optional rap ta ucode is not available
[    3.188578] amdgpu 0000:f4:00.0: amdgpu: SECUREDISPLAY: securedisplay ta ucode is not available
[    3.231322] amdgpu 0000:f4:00.0: amdgpu: SMU is initialized successfully!
[    3.298733] amdgpu 0000:f4:00.0: amdgpu: MES FW version must be >= 0x7f to enable LR compute workaround.
[    3.312271] kfd kfd: amdgpu: Allocated 3969056 bytes on gart
[    3.312284] kfd kfd: amdgpu: Total number of KFD nodes to be created: 1
[    3.315282] amdgpu: Virtual CRAT table created for GPU
[    3.315396] amdgpu: Topology: Add dGPU node [0x1586:0x1002]
[    3.315396] kfd kfd: amdgpu: added device 1002:1586
[    3.315406] amdgpu 0000:f4:00.0: amdgpu: SE 2, SH per SE 2, CU per SH 10, active_cu_number 40
[    3.315409] amdgpu 0000:f4:00.0: amdgpu: ring gfx_0.0.0 uses VM inv eng 0 on hub 0
[    3.315410] amdgpu 0000:f4:00.0: amdgpu: ring comp_1.0.0 uses VM inv eng 1 on hub 0
[    3.315411] amdgpu 0000:f4:00.0: amdgpu: ring comp_1.1.0 uses VM inv eng 4 on hub 0
[    3.315411] amdgpu 0000:f4:00.0: amdgpu: ring comp_1.2.0 uses VM inv eng 6 on hub 0
[    3.315412] amdgpu 0000:f4:00.0: amdgpu: ring comp_1.3.0 uses VM inv eng 7 on hub 0
[    3.315412] amdgpu 0000:f4:00.0: amdgpu: ring comp_1.0.1 uses VM inv eng 8 on hub 0
[    3.315412] amdgpu 0000:f4:00.0: amdgpu: ring comp_1.1.1 uses VM inv eng 9 on hub 0
[    3.315413] amdgpu 0000:f4:00.0: amdgpu: ring comp_1.2.1 uses VM inv eng 10 on hub 0
[    3.315413] amdgpu 0000:f4:00.0: amdgpu: ring comp_1.3.1 uses VM inv eng 11 on hub 0
[    3.315414] amdgpu 0000:f4:00.0: amdgpu: ring sdma0 uses VM inv eng 12 on hub 0
[    3.315414] amdgpu 0000:f4:00.0: amdgpu: ring vcn_unified_0 uses VM inv eng 0 on hub 8
[    3.315414] amdgpu 0000:f4:00.0: amdgpu: ring vcn_unified_1 uses VM inv eng 1 on hub 8
[    3.315415] amdgpu 0000:f4:00.0: amdgpu: ring jpeg_dec_0 uses VM inv eng 4 on hub 8
[    3.315415] amdgpu 0000:f4:00.0: amdgpu: ring jpeg_dec_1 uses VM inv eng 6 on hub 8
[    3.315416] amdgpu 0000:f4:00.0: amdgpu: ring mes_kiq_3.1.0 uses VM inv eng 13 on hub 0
[    3.315416] amdgpu 0000:f4:00.0: amdgpu: ring vpe uses VM inv eng 7 on hub 8
[    3.322406] amdgpu 0000:f4:00.0: amdgpu: Runtime PM not available
[    3.322845] [drm] Initialized amdgpu 3.61.0 for 0000:f4:00.0 on minor 0
[    3.421753] amdgpu 0000:f4:00.0: [drm] fb0: amdgpudrmfb frame buffer device
[    4.668771] snd_hda_intel 0000:f4:00.1: bound 0000:f4:00.0 (ops amdgpu_dm_audio_component_bind_ops [amdgpu])

root@ailab:~# lspci
00:00.0 Host bridge: Advanced Micro Devices, Inc. [AMD] Strix/Strix Halo Root Complex (rev 02)
00:00.2 IOMMU: Advanced Micro Devices, Inc. [AMD] Strix/Strix Halo IOMMU (rev 02)
00:01.0 Host bridge: Advanced Micro Devices, Inc. [AMD] Strix/Strix Halo Dummy Host Bridge
00:01.1 PCI bridge: Advanced Micro Devices, Inc. [AMD] Strix/Strix Halo PCIe USB4 Bridge (rev 02)
00:01.2 PCI bridge: Advanced Micro Devices, Inc. [AMD] Strix/Strix Halo PCIe USB4 Bridge (rev 02)
00:02.0 Host bridge: Advanced Micro Devices, Inc. [AMD] Strix/Strix Halo Dummy Host Bridge
00:02.1 PCI bridge: Advanced Micro Devices, Inc. [AMD] Strix/Strix Halo GPP Bridge (rev 02)
00:02.2 PCI bridge: Advanced Micro Devices, Inc. [AMD] Strix/Strix Halo GPP Bridge (rev 02)
00:02.3 PCI bridge: Advanced Micro Devices, Inc. [AMD] Strix/Strix Halo GPP Bridge (rev 02)
00:02.5 PCI bridge: Advanced Micro Devices, Inc. [AMD] Strix/Strix Halo GPP Bridge
00:03.0 Host bridge: Advanced Micro Devices, Inc. [AMD] Strix/Strix Halo Dummy Host Bridge
00:03.2 PCI bridge: Advanced Micro Devices, Inc. [AMD] Strix/Strix Halo GPP Bridge (rev 02)
00:08.0 Host bridge: Advanced Micro Devices, Inc. [AMD] Strix/Strix Halo Dummy Host Bridge
00:08.1 PCI bridge: Advanced Micro Devices, Inc. [AMD] Strix/Strix Halo Internal GPP Bridge to Bus [C:A]
00:08.2 PCI bridge: Advanced Micro Devices, Inc. [AMD] Strix/Strix Halo Internal GPP Bridge to Bus [C:A]
00:08.3 PCI bridge: Advanced Micro Devices, Inc. [AMD] Strix/Strix Halo Internal GPP Bridge to Bus [C:A]
00:14.0 SMBus: Advanced Micro Devices, Inc. [AMD] FCH SMBus Controller (rev 71)
00:14.3 ISA bridge: Advanced Micro Devices, Inc. [AMD] FCH LPC Bridge (rev 51)
00:18.0 Host bridge: Advanced Micro Devices, Inc. [AMD] Strix Halo Data Fabric; Function 0
00:18.1 Host bridge: Advanced Micro Devices, Inc. [AMD] Strix Halo Data Fabric; Function 1
00:18.2 Host bridge: Advanced Micro Devices, Inc. [AMD] Strix Halo Data Fabric; Function 2
00:18.3 Host bridge: Advanced Micro Devices, Inc. [AMD] Strix Halo Data Fabric; Function 3
00:18.4 Host bridge: Advanced Micro Devices, Inc. [AMD] Strix Halo Data Fabric; Function 4
00:18.5 Host bridge: Advanced Micro Devices, Inc. [AMD] Strix Halo Data Fabric; Function 5
00:18.6 Host bridge: Advanced Micro Devices, Inc. [AMD] Strix Halo Data Fabric; Function 6
00:18.7 Host bridge: Advanced Micro Devices, Inc. [AMD] Strix Halo Data Fabric; Function 7
c1:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. Device 8127 (rev 05)
c2:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. Device 8127 (rev 05)
c3:00.0 Network controller: MEDIATEK Corp. Device 7925
c4:00.0 Non-Volatile memory controller: Micron/Crucial Technology Device 5427 (rev 01)
c5:00.0 PCI bridge: Intel Corporation JHL9580 Thunderbolt 5 80/120G Bridge [Barlow Ridge Host 80G 2023] (rev 84)
c6:00.0 PCI bridge: Intel Corporation JHL9580 Thunderbolt 5 80/120G Bridge [Barlow Ridge Host 80G 2023] (rev 84)
c6:01.0 PCI bridge: Intel Corporation JHL9580 Thunderbolt 5 80/120G Bridge [Barlow Ridge Host 80G 2023] (rev 84)
c6:02.0 PCI bridge: Intel Corporation JHL9580 Thunderbolt 5 80/120G Bridge [Barlow Ridge Host 80G 2023] (rev 84)
c6:03.0 PCI bridge: Intel Corporation JHL9580 Thunderbolt 5 80/120G Bridge [Barlow Ridge Host 80G 2023] (rev 84)
c7:00.0 USB controller: Intel Corporation JHL9580 Thunderbolt 5 80/120G NHI [Barlow Ridge Host 80G 2023] (rev 84)
f2:00.0 USB controller: Intel Corporation JHL9580 Thunderbolt 5 80/120G USB Controller [Barlow Ridge Host 80G 2023]
f4:00.0 Display controller: Advanced Micro Devices, Inc. [AMD/ATI] Strix Halo [Radeon Graphics / Radeon 8050S Graphics / Radeon 8060S Graphics] (rev c1)
f4:00.1 Audio device: Advanced Micro Devices, Inc. [AMD/ATI] Radeon High Definition Audio Controller [Rembrandt/Strix]
f4:00.2 Encryption controller: Advanced Micro Devices, Inc. [AMD] Strix/Krackan/Strix Halo CCP/ASP
f4:00.4 USB controller: Advanced Micro Devices, Inc. [AMD] Strix Halo USB 3.1 xHCI
f4:00.6 Audio device: Advanced Micro Devices, Inc. [AMD] Family 17h/19h/1ah HD Audio Controller
f4:00.7 Signal processing controller: Advanced Micro Devices, Inc. [AMD] Sensor Fusion Hub
f5:00.0 Non-Essential Instrumentation [1300]: Advanced Micro Devices, Inc. [AMD] Strix/Strix Halo PCIe Dummy Function
f5:00.1 Signal processing controller: Advanced Micro Devices, Inc. [AMD] Strix/Krackan/Strix Halo Neural Processing Unit (rev 11)
f6:00.0 USB controller: Advanced Micro Devices, Inc. [AMD] Strix Halo USB 3.1 xHCI
f6:00.3 USB controller: Advanced Micro Devices, Inc. [AMD] Strix Halo USB 3.1 xHCI
f6:00.4 USB controller: Advanced Micro Devices, Inc. [AMD] Strix Halo USB 3.1 xHCI
f6:00.5 USB controller: Advanced Micro Devices, Inc. [AMD] Strix Halo USB4 Host Router
f6:00.6 USB controller: Advanced Micro Devices, Inc. [AMD] Strix Halo USB4 Host Router
root@ailab:~# lspci|grep -i real
c1:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. Device 8127 (rev 05)
c2:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. Device 8127 (rev 05)
root@ailab:~# lsusb
Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 002 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub
Bus 003 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 003 Device 002: ID 05e3:0610 Genesys Logic, Inc. Hub
Bus 003 Device 003: ID 0bda:5409 Realtek Semiconductor Corp. 4-Port USB 2.0 Hub
Bus 003 Device 004: ID 093a:2510 Pixart Imaging, Inc. Optical Mouse
Bus 003 Device 005: ID 046a:0023 CHERRY Keyboard
Bus 004 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub
Bus 004 Device 002: ID 05e3:0625 Genesys Logic, Inc. USB3.2 Hub
Bus 004 Device 003: ID 0bda:8153 Realtek Semiconductor Corp. RTL8153 Gigabit Ethernet Adapter
Bus 005 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 005 Device 002: ID 0e8d:7925 MediaTek Inc. Wireless_Device
Bus 006 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub
Bus 007 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 008 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub
Bus 009 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 010 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub

TODO

Anpassung VRAM für GPU von 2G auf 96G
ROCm Nightly/7.2+
vLLM

Upgrade auf 6.18.x Kernel

Quellen für „sid“ hinzufügen in /etc/apt/sources.list

# Unstable for Kernel
deb http://deb.debian.org/debian sid main contrib non-free non-free-firmware

Präferenz einstellen, unstable nur auf Anforderung installieren via /etc/apt/preferences.d/kernel-from-sid

Package: *
Pin: release a=unstable
Pin-Priority: 100

Installation & Neustart

apt update
apt install linux-image-6.18.5+deb14-amd64
reboot

Firmware sichern und aktualisieren (Quelle):

Bitte beachten dass es dafür sicher bessere Wege gibt, allerdings ging es so erstmal „flott“ für mich ;))

Bitte beachten dass Updates für die Firmware aus Debian-Quellen evtl. diesen Schritt überschreiben!

wget https://launchpad.net/linux/incoming/20260110/+download/linux-firmware-20260110.tar.gz
cp -R /lib/firmware .
mv firmware/ firmware-20250410-2-backup
tar xzf linux-firmware-20260110.tar.gz 
cd /lib/firmware/
cp -R ~/linux-firmware-20260110/* .
reboot

Nach dem Neustart werden sowohl die Netzwerkkarte als auch die AMDGPU mit aktuellem Firmwarestand erkannt.

sudo dmesg | grep amdgpu
[    2.762088] [drm] amdgpu kernel modesetting enabled.
[    2.764199] amdgpu: Virtual CRAT table created for CPU
[    2.764205] amdgpu: Topology: Add CPU node
[    2.764299] amdgpu 0000:f4:00.0: enabling device (0006 -> 0007)
[    2.764339] amdgpu 0000:f4:00.0: amdgpu: initializing kernel modesetting (IP DISCOVERY 0x1002:0x1586 0x1F4C:0xB026 0xC1).
[    2.764346] amdgpu 0000:f4:00.0: amdgpu: register mmio base: 0xA0300000
[    2.764347] amdgpu 0000:f4:00.0: amdgpu: register mmio size: 1048576
[    2.767227] amdgpu 0000:f4:00.0: amdgpu: detected ip block number 0 <common_v1_0_0> (soc21_common)
[    2.767228] amdgpu 0000:f4:00.0: amdgpu: detected ip block number 1 <gmc_v11_0_0> (gmc_v11_0)
[    2.767229] amdgpu 0000:f4:00.0: amdgpu: detected ip block number 2 <ih_v6_0_0> (ih_v6_1)
[    2.767230] amdgpu 0000:f4:00.0: amdgpu: detected ip block number 3 <psp_v13_0_0> (psp)
[    2.767231] amdgpu 0000:f4:00.0: amdgpu: detected ip block number 4 <smu_v14_0_0> (smu)
[    2.767232] amdgpu 0000:f4:00.0: amdgpu: detected ip block number 5 <dce_v1_0_0> (dm)
[    2.767233] amdgpu 0000:f4:00.0: amdgpu: detected ip block number 6 <gfx_v11_0_0> (gfx_v11_0)
[    2.767234] amdgpu 0000:f4:00.0: amdgpu: detected ip block number 7 <sdma_v6_0_0> (sdma_v6_0)
[    2.767235] amdgpu 0000:f4:00.0: amdgpu: detected ip block number 8 <vcn_v4_0_5> (vcn_v4_0_5)
[    2.767235] amdgpu 0000:f4:00.0: amdgpu: detected ip block number 9 <jpeg_v4_0_5> (jpeg_v4_0_5)
[    2.767236] amdgpu 0000:f4:00.0: amdgpu: detected ip block number 10 <mes_v11_0_0> (mes_v11_0)
[    2.767237] amdgpu 0000:f4:00.0: amdgpu: detected ip block number 11 <vpe_v6_1_0> (vpe_v6_1)
[    2.767258] amdgpu 0000:f4:00.0: amdgpu: Fetched VBIOS from VFCT
[    2.767259] amdgpu: ATOM BIOS: 113-STRXLGEN-001
[    2.768783] amdgpu 0000:f4:00.0: amdgpu: VPE: collaborate mode true
[    2.768895] amdgpu 0000:f4:00.0: amdgpu: Trusted Memory Zone (TMZ) feature disabled as experimental (default)
[    2.768927] amdgpu 0000:f4:00.0: amdgpu: vm size is 262144 GB, 4 levels, block size is 9-bit, fragment size is 9-bit
[    2.768947] amdgpu 0000:f4:00.0: amdgpu: VRAM: 98304M 0x0000008000000000 - 0x00000097FFFFFFFF (98304M used)
[    2.768948] amdgpu 0000:f4:00.0: amdgpu: GART: 512M 0x00007FFF00000000 - 0x00007FFF1FFFFFFF
[    2.769348] amdgpu 0000:f4:00.0: amdgpu: amdgpu: 98304M of VRAM memory ready
[    2.769350] amdgpu 0000:f4:00.0: amdgpu: amdgpu: 15866M of GTT memory ready.
[    2.770425] amdgpu 0000:f4:00.0: amdgpu: [drm] Loading DMUB firmware via PSP: version=0x09001B00
[    2.770785] amdgpu 0000:f4:00.0: amdgpu: [VCN instance 0] Found VCN firmware Version ENC: 1.23 DEC: 9 VEP: 0 Revision: 3
[    2.770826] amdgpu 0000:f4:00.0: amdgpu: [VCN instance 1] Found VCN firmware Version ENC: 1.23 DEC: 9 VEP: 0 Revision: 3
[    2.793976] amdgpu 0000:f4:00.0: amdgpu: reserve 0x8c00000 from 0x97e0000000 for PSP TMR
[    3.128962] amdgpu 0000:f4:00.0: amdgpu: RAS: optional ras ta ucode is not available
[    3.132860] amdgpu 0000:f4:00.0: amdgpu: RAP: optional rap ta ucode is not available
[    3.132861] amdgpu 0000:f4:00.0: amdgpu: SECUREDISPLAY: optional securedisplay ta ucode is not available
[    3.167027] amdgpu 0000:f4:00.0: amdgpu: SMU is initialized successfully!
[    3.168439] amdgpu 0000:f4:00.0: amdgpu: [drm] Display Core v3.2.351 initialized on DCN 3.5.1
[    3.168440] amdgpu 0000:f4:00.0: amdgpu: [drm] DP-HDMI FRL PCON supported
[    3.171296] amdgpu 0000:f4:00.0: amdgpu: [drm] DMUB hardware initialized: version=0x09001B00
[    3.175222] amdgpu 0000:f4:00.0: amdgpu: [drm] PSR support 0, DC PSR ver -1, sink PSR ver 0 DPCD caps 0x0 su_y_granularity 0
[    3.175308] amdgpu 0000:f4:00.0: amdgpu: [drm] PSR support 0, DC PSR ver -1, sink PSR ver 0 DPCD caps 0x0 su_y_granularity 0
[    3.175423] amdgpu 0000:f4:00.0: amdgpu: [drm] PSR support 0, DC PSR ver -1, sink PSR ver 0 DPCD caps 0x0 su_y_granularity 0
[    3.175533] amdgpu 0000:f4:00.0: amdgpu: [drm] PSR support 0, DC PSR ver -1, sink PSR ver 0 DPCD caps 0x0 su_y_granularity 0
[    3.205183] amdgpu 0000:f4:00.0: amdgpu: [drm] PSR support 0, DC PSR ver -1, sink PSR ver 0 DPCD caps 0x0 su_y_granularity 0
[    3.205243] amdgpu 0000:f4:00.0: amdgpu: [drm] PSR support 0, DC PSR ver -1, sink PSR ver 0 DPCD caps 0x0 su_y_granularity 0
[    3.205279] amdgpu 0000:f4:00.0: amdgpu: [drm] PSR support 0, DC PSR ver -1, sink PSR ver 0 DPCD caps 0x0 su_y_granularity 0
[    3.205313] amdgpu 0000:f4:00.0: amdgpu: [drm] PSR support 0, DC PSR ver -1, sink PSR ver 0 DPCD caps 0x0 su_y_granularity 0
[    3.205347] amdgpu 0000:f4:00.0: amdgpu: [drm] PSR support 0, DC PSR ver -1, sink PSR ver 0 DPCD caps 0x0 su_y_granularity 0
[    3.206907] amdgpu 0000:f4:00.0: amdgpu: MES FW version must be >= 0x7f to enable LR compute workaround.
[    3.351143] amdgpu: HMM registered 98304MB device memory
[    3.352104] kfd kfd: amdgpu: Allocated 3969056 bytes on gart
[    3.352125] kfd kfd: amdgpu: Total number of KFD nodes to be created: 1
[    3.355138] amdgpu: Virtual CRAT table created for GPU
[    3.355955] amdgpu: Topology: Add dGPU node [0x1586:0x1002]
[    3.355956] kfd kfd: amdgpu: added device 1002:1586
[    3.355967] amdgpu 0000:f4:00.0: amdgpu: SE 2, SH per SE 2, CU per SH 10, active_cu_number 40
[    3.355973] amdgpu 0000:f4:00.0: amdgpu: ring gfx_0.0.0 uses VM inv eng 0 on hub 0
[    3.355974] amdgpu 0000:f4:00.0: amdgpu: ring comp_1.0.0 uses VM inv eng 1 on hub 0
[    3.355975] amdgpu 0000:f4:00.0: amdgpu: ring comp_1.1.0 uses VM inv eng 4 on hub 0
[    3.355975] amdgpu 0000:f4:00.0: amdgpu: ring comp_1.2.0 uses VM inv eng 6 on hub 0
[    3.355976] amdgpu 0000:f4:00.0: amdgpu: ring comp_1.3.0 uses VM inv eng 7 on hub 0
[    3.355976] amdgpu 0000:f4:00.0: amdgpu: ring comp_1.0.1 uses VM inv eng 8 on hub 0
[    3.355977] amdgpu 0000:f4:00.0: amdgpu: ring comp_1.1.1 uses VM inv eng 9 on hub 0
[    3.355977] amdgpu 0000:f4:00.0: amdgpu: ring comp_1.2.1 uses VM inv eng 10 on hub 0
[    3.355978] amdgpu 0000:f4:00.0: amdgpu: ring comp_1.3.1 uses VM inv eng 11 on hub 0
[    3.355978] amdgpu 0000:f4:00.0: amdgpu: ring sdma0 uses VM inv eng 12 on hub 0
[    3.355979] amdgpu 0000:f4:00.0: amdgpu: ring vcn_unified_0 uses VM inv eng 0 on hub 8
[    3.355980] amdgpu 0000:f4:00.0: amdgpu: ring vcn_unified_1 uses VM inv eng 1 on hub 8
[    3.355980] amdgpu 0000:f4:00.0: amdgpu: ring jpeg_dec_0 uses VM inv eng 4 on hub 8
[    3.355981] amdgpu 0000:f4:00.0: amdgpu: ring jpeg_dec_1 uses VM inv eng 6 on hub 8
[    3.355981] amdgpu 0000:f4:00.0: amdgpu: ring mes_kiq_3.1.0 uses VM inv eng 13 on hub 0
[    3.355982] amdgpu 0000:f4:00.0: amdgpu: ring vpe uses VM inv eng 7 on hub 8
[    3.357223] amdgpu 0000:f4:00.0: amdgpu: Runtime PM not available
[    3.357642] amdgpu 0000:f4:00.0: [drm] Registered 4 planes with drm panic
[    3.357643] [drm] Initialized amdgpu 3.64.0 for 0000:f4:00.0 on minor 0
[    3.361354] amdgpu 0000:f4:00.0: amdgpu: [drm] Failed to setup vendor infoframe on connector HDMI-A-1: -22
[    3.362562] fbcon: amdgpudrmfb (fb0) is primary device
[    3.541052] amdgpu 0000:f4:00.0: [drm] fb0: amdgpudrmfb frame buffer device
[    4.510349] snd_hda_intel 0000:f4:00.1: bound 0000:f4:00.0 (ops amdgpu_dm_audio_component_bind_ops [amdgpu])