Intermittent Middle Fan on Sapphire PULSE RX 7900 XT

blabla52 · 2025-04-27 12:03:12

Hi everyone,

I'm experiencing a strange and intermittent issue with the middle fan on my Sapphire PULSE AMD Radeon RX 7900 XT 20GB graphics card and I'm hoping someone might have some insights.

The Problem:

The middle fan on the GPU will intermittently spin up to what sounds like maximum speed, while the other two fans remain off or at low speed (as expected when idle).

This happens unpredictably. Sometimes it spins up for a few seconds and then returns to normal speed (or stops, matching the other fans at idle).
Other times, it gets stuck at maximum speed and stays there for minutes or indefinitely.
Crucially, when the fan is stuck at high speed, this behavior persists even through a reboot and into the BIOS/UEFI setup screen. The only way to get it back to normal then is to completely shut down the PC for several minutes and power it back on.
When the middle fan is behaving normally, it seems to respond correctly to fan control software (LACT).

System Information:

GPU: Sapphire PULSE AMD Radeon RX 7900 XT 20GB
CPU: AMD Ryzen 9 7950X3D
Motherboard: ASRock B650E Taichi Lite
OS: Arch Linux (up-to-date)
Kernel: 6.14.3-arch1-1
Mesa Drivers: mesa 1:25.0.4-1, lib32-mesa 1:25.0.4-1, vulkan-radeon 1:25.0.4-1, lib32-vulkan-radeon 1:25.0.4-1
Display Server: Wayland
Fan Control Tool: lact 0.7.3-1
Boot Parameter: amdgpu.ppfeaturemask=0xffffffff (added for LACT control)

Relevant Command Outputs:

lspci -k | grep -EA3 'VGA|3D':

03:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] Navi 31 [Radeon RX 7900 XT/7900 XTX/7900 GRE/7900M] (rev cc)
    Subsystem: Sapphire Technology Limited PULSE RX 7900 XTX
    Kernel driver in use: amdgpu
    Kernel modules: amdgpu
--
50:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] Raphael (rev c9)
    Subsystem: Advanced Micro Devices, Inc. [AMD/ATI] Raphael
    Kernel driver in use: amdgpu
    Kernel modules: amdgpu

sensors (GPU section only):

amdgpu-pci-0300
Adapter: PCI adapter
vddgfx:      721.00 mV 
fan1:           0 RPM  (min =    0 RPM, max = 3600 RPM)
edge:         +44.0°C  (crit = +100.0°C, hyst = -273.1°C)
                       (emerg = +105.0°C)
junction:     +52.0°C  (crit = +110.0°C, hyst = -273.1°C)
                       (emerg = +115.0°C)
mem:          +60.0°C  (crit = +108.0°C, hyst = -273.1°C)
                       (emerg = +113.0°C)
PPT:          58.00 W  (cap = 265.00 W)

(Note: sensors only shows fan1)

hwmon interface for the GPU (card1):
Path:

/sys/class/drm/card1/device/hwmon/hwmon8/

Relevant files found:

.rw-r--r-- root root 4.0 KB Sun Apr 27 13:27:05 2025 fan1_enable
.r--r--r-- root root 4.0 KB Sun Apr 27 11:36:27 2025 fan1_input
.r--r--r-- root root 4.0 KB Sun Apr 27 11:36:27 2025 fan1_max
.r--r--r-- root root 4.0 KB Sun Apr 27 11:36:27 2025 fan1_min
.rw-r--r-- root root 4.0 KB Sun Apr 27 13:27:05 2025 fan1_target
.rw-r--r-- root root 4.0 KB Sun Apr 27 11:36:27 2025 pwm1
.rw-r--r-- root root 4.0 KB Sun Apr 27 11:35:37 2025 pwm1_enable
.r--r--r-- root root 4.0 KB Sun Apr 27 13:27:05 2025 pwm1_max
.r--r--r-- root root 4.0 KB Sun Apr 27 13:27:05 2025 pwm1_min

sudo dmesg | grep -i amdgpu:

[    0.000000] Command line: BOOT_IMAGE=/@/boot/vmlinuz-linux root=UUID=6b8f6d86-5209-411b-a783-43a57c116751 rw rootflags=subvol=@ nowatchdog nvme_load=YES resume=UUID=9b260dbf-3643-407a-a924-2123976a42ef loglevel=3 amdgpu.ppfeaturemask=0xffffffff
[    0.047205] Kernel command line: BOOT_IMAGE=/@/boot/vmlinuz-linux root=UUID=6b8f6d86-5209-411b-a783-43a57c116751 rw rootflags=subvol=@ nowatchdog nvme_load=YES resume=UUID=9b260dbf-3643-407a-a924-2123976a42ef loglevel=3 amdgpu.ppfeaturemask=0xffffffff
[    7.960597] [drm] amdgpu kernel modesetting enabled.
[    7.960615] amdgpu: vga_switcheroo: detected switching method \_SB_.PCI0.GP17.VGA_.ATPX handle
[    7.960749] amdgpu: ATPX version 1, functions 0x00000000
[    7.969179] amdgpu: Virtual CRAT table created for CPU
[    7.969188] amdgpu: Topology: Add CPU node
[    7.969212] amdgpu: Overdrive is enabled, please disable it before reporting any bugs unrelated to overdrive.
[    7.969573] amdgpu 0000:03:00.0: enabling device (0006 -> 0007)
[    7.973819] amdgpu 0000:03:00.0: amdgpu: detected ip block number 0 <soc21_common>
[    7.973821] amdgpu 0000:03:00.0: amdgpu: detected ip block number 1 <gmc_v11_0>
[    7.973822] amdgpu 0000:03:00.0: amdgpu: detected ip block number 2 <ih_v6_0>
[    7.973823] amdgpu 0000:03:00.0: amdgpu: detected ip block number 3 <psp>
[    7.973825] amdgpu 0000:03:00.0: amdgpu: detected ip block number 4 <smu>
[    7.973826] amdgpu 0000:03:00.0: amdgpu: detected ip block number 5 <dm>
[    7.973827] amdgpu 0000:03:00.0: amdgpu: detected ip block number 6 <gfx_v11_0>
[    7.973828] amdgpu 0000:03:00.0: amdgpu: detected ip block number 7 <sdma_v6_0>
[    7.973829] amdgpu 0000:03:00.0: amdgpu: detected ip block number 8 <vcn_v4_0>
[    7.973831] amdgpu 0000:03:00.0: amdgpu: detected ip block number 9 <jpeg_v4_0>
[    7.973832] amdgpu 0000:03:00.0: amdgpu: detected ip block number 10 <mes_v11_0>
[    7.973848] amdgpu 0000:03:00.0: amdgpu: Fetched VBIOS from VFCT
[    7.973849] amdgpu: ATOM BIOS: 113-D70401XT-P11
[    7.978040] amdgpu 0000:03:00.0: amdgpu: CP RS64 enable
[    8.030618] amdgpu 0000:03:00.0: vgaarb: deactivate vga console
[    8.030622] amdgpu 0000:03:00.0: amdgpu: Trusted Memory Zone (TMZ) feature not supported
[    8.030654] amdgpu 0000:03:00.0: amdgpu: MEM ECC is not presented.
[    8.030656] amdgpu 0000:03:00.0: amdgpu: SRAM ECC is not presented.
[    8.030696] amdgpu 0000:03:00.0: amdgpu: VRAM: 20464M 0x0000008000000000 - 0x00000084FEFFFFFF (20464M used)
[    8.030698] amdgpu 0000:03:00.0: amdgpu: GART: 512M 0x00007FFF00000000 - 0x00007FFF1FFFFFFF
[    8.030780] [drm] amdgpu: 20464M of VRAM memory ready
[    8.030782] [drm] amdgpu: 31699M of GTT memory ready.
[    8.101274] amdgpu 0000:03:00.0: amdgpu: reserve 0x1300000 from 0x84fc000000 for PSP TMR
[    8.248956] amdgpu 0000:03:00.0: amdgpu: RAP: optional rap ta ucode is not available
[    8.248961] amdgpu 0000:03:00.0: amdgpu: SECUREDISPLAY: securedisplay ta ucode is not available
[    8.249013] amdgpu 0000:03:00.0: amdgpu: smu driver if version = 0x0000003d, smu fw if version = 0x00000040, smu fw program = 0, smu fw version = 0x004e8000 (78.128.0)
[    8.249016] amdgpu 0000:03:00.0: amdgpu: SMU driver if version not matched
[    8.408519] amdgpu 0000:03:00.0: amdgpu: SMU is initialized successfully!
[    8.481930] snd_hda_intel 0000:03:00.1: bound 0000:03:00.0 (ops amdgpu_dm_audio_component_bind_ops [amdgpu])
[    8.952440] amdgpu: HMM registered 20464MB device memory
[    8.953691] kfd kfd: amdgpu: Allocated 3969056 bytes on gart
[    8.953702] kfd kfd: amdgpu: Total number of KFD nodes to be created: 1
[    8.953736] amdgpu: Virtual CRAT table created for GPU
[    8.953894] amdgpu: Topology: Add dGPU node [0x744c:0x1002]
[    8.953896] kfd kfd: amdgpu: added device 1002:744c
[    8.953907] amdgpu 0000:03:00.0: amdgpu: SE 6, SH per SE 2, CU per SH 8, active_cu_number 84
[    8.953910] amdgpu 0000:03:00.0: amdgpu: ring gfx_0.0.0 uses VM inv eng 0 on hub 0
[    8.953911] amdgpu 0000:03:00.0: amdgpu: ring comp_1.0.0 uses VM inv eng 1 on hub 0
[    8.953912] amdgpu 0000:03:00.0: amdgpu: ring comp_1.1.0 uses VM inv eng 4 on hub 0
[    8.953913] amdgpu 0000:03:00.0: amdgpu: ring comp_1.2.0 uses VM inv eng 6 on hub 0
[    8.953914] amdgpu 0000:03:00.0: amdgpu: ring comp_1.3.0 uses VM inv eng 7 on hub 0
[    8.953915] amdgpu 0000:03:00.0: amdgpu: ring comp_1.0.1 uses VM inv eng 8 on hub 0
[    8.953916] amdgpu 0000:03:00.0: amdgpu: ring comp_1.1.1 uses VM inv eng 9 on hub 0
[    8.953917] amdgpu 0000:03:00.0: amdgpu: ring comp_1.2.1 uses VM inv eng 10 on hub 0
[    8.953918] amdgpu 0000:03:00.0: amdgpu: ring comp_1.3.1 uses VM inv eng 11 on hub 0
[    8.953919] amdgpu 0000:03:00.0: amdgpu: ring sdma0 uses VM inv eng 12 on hub 0
[    8.953920] amdgpu 0000:03:00.0: amdgpu: ring sdma1 uses VM inv eng 13 on hub 0
[    8.953921] amdgpu 0000:03:00.0: amdgpu: ring vcn_unified_0 uses VM inv eng 0 on hub 8
[    8.953923] amdgpu 0000:03:00.0: amdgpu: ring vcn_unified_1 uses VM inv eng 1 on hub 8
[    8.953924] amdgpu 0000:03:00.0: amdgpu: ring jpeg_dec uses VM inv eng 4 on hub 8
[    8.953925] amdgpu 0000:03:00.0: amdgpu: ring mes_kiq_3.1.0 uses VM inv eng 14 on hub 0
[    8.957444] amdgpu 0000:03:00.0: amdgpu: Using BACO for runtime pm
[    8.957819] amdgpu 0000:03:00.0: [drm] Registered 4 planes with drm panic
[    8.957820] [drm] Initialized amdgpu 3.61.0 for 0000:03:00.0 on minor 1
[    8.964435] fbcon: amdgpudrmfb (fb0) is primary device
[    9.180628] amdgpu 0000:03:00.0: [drm] fb0: amdgpudrmfb frame buffer device
[    9.192302] amdgpu 0000:50:00.0: enabling device (0006 -> 0007)
[    9.193941] amdgpu 0000:50:00.0: amdgpu: detected ip block number 0 <nv_common>
[    9.193943] amdgpu 0000:50:00.0: amdgpu: detected ip block number 1 <gmc_v10_0>
[    9.193944] amdgpu 0000:50:00.0: amdgpu: detected ip block number 2 <navi10_ih>
[    9.193945] amdgpu 0000:50:00.0: amdgpu: detected ip block number 3 <psp>
[    9.193946] amdgpu 0000:50:00.0: amdgpu: detected ip block number 4 <smu>
[    9.193947] amdgpu 0000:50:00.0: amdgpu: detected ip block number 5 <dm>
[    9.193949] amdgpu 0000:50:00.0: amdgpu: detected ip block number 6 <gfx_v10_0>
[    9.193950] amdgpu 0000:50:00.0: amdgpu: detected ip block number 7 <sdma_v5_2>
[    9.193951] amdgpu 0000:50:00.0: amdgpu: detected ip block number 8 <vcn_v3_0>
[    9.193953] amdgpu 0000:50:00.0: amdgpu: detected ip block number 9 <jpeg_v3_0>
[    9.193964] amdgpu 0000:50:00.0: amdgpu: Fetched VBIOS from VFCT
[    9.193966] amdgpu: ATOM BIOS: 102-RAPHAEL-008
[    9.201297] amdgpu 0000:50:00.0: amdgpu: Trusted Memory Zone (TMZ) feature disabled as experimental (default)
[    9.201325] amdgpu 0000:50:00.0: amdgpu: VRAM: 512M 0x000000F400000000 - 0x000000F41FFFFFFF (512M used)
[    9.201326] amdgpu 0000:50:00.0: amdgpu: GART: 1024M 0x0000000000000000 - 0x000000003FFFFFFF
[    9.201395] [drm] amdgpu: 512M of VRAM memory ready
[    9.201396] [drm] amdgpu: 31699M of GTT memory ready.
[    9.224130] amdgpu 0000:50:00.0: amdgpu: reserve 0xa00000 from 0xf41e000000 for PSP TMR
[    9.291131] amdgpu 0000:50:00.0: amdgpu: RAS: optional ras ta ucode is not available
[    9.296926] amdgpu 0000:50:00.0: amdgpu: RAP: optional rap ta ucode is not available
[    9.296928] amdgpu 0000:50:00.0: amdgpu: SECUREDISPLAY: securedisplay ta ucode is not available
[    9.298628] amdgpu 0000:50:00.0: amdgpu: SMU is initialized successfully!
[    9.300218] snd_hda_intel 0000:50:00.1: bound 0000:50:00.0 (ops amdgpu_dm_audio_component_bind_ops [amdgpu])
[    9.304081] kfd kfd: amdgpu: Allocated 3969056 bytes on gart
[    9.304091] kfd kfd: amdgpu: Total number of KFD nodes to be created: 1
[    9.304650] amdgpu: Virtual CRAT table created for GPU
[    9.305544] amdgpu: Topology: Add dGPU node [0x164e:0x1002]
[    9.305546] kfd kfd: amdgpu: added device 1002:164e
[    9.305553] amdgpu 0000:50:00.0: amdgpu: SE 1, SH per SE 1, CU per SH 2, active_cu_number 2
[    9.305556] amdgpu 0000:50:00.0: amdgpu: ring gfx_0.0.0 uses VM inv eng 0 on hub 0
[    9.305557] amdgpu 0000:50:00.0: amdgpu: ring gfx_0.1.0 uses VM inv eng 1 on hub 0
[    9.305558] amdgpu 0000:50:00.0: amdgpu: ring comp_1.0.0 uses VM inv eng 4 on hub 0
[    9.305559] amdgpu 0000:50:00.0: amdgpu: ring comp_1.1.0 uses VM inv eng 5 on hub 0
[    9.305560] amdgpu 0000:50:00.0: amdgpu: ring comp_1.2.0 uses VM inv eng 6 on hub 0
[    9.305561] amdgpu 0000:50:00.0: amdgpu: ring comp_1.3.0 uses VM inv eng 7 on hub 0
[    9.305562] amdgpu 0000:50:00.0: amdgpu: ring comp_1.0.1 uses VM inv eng 8 on hub 0
[    9.305563] amdgpu 0000:50:00.0: amdgpu: ring comp_1.1.1 uses VM inv eng 9 on hub 0
[    9.305564] amdgpu 0000:50:00.0: amdgpu: ring comp_1.2.1 uses VM inv eng 10 on hub 0
[    9.305565] amdgpu 0000:50:00.0: amdgpu: ring comp_1.3.1 uses VM inv eng 11 on hub 0
[    9.305566] amdgpu 0000:50:00.0: amdgpu: ring kiq_0.2.1.0 uses VM inv eng 12 on hub 0
[    9.305567] amdgpu 0000:50:00.0: amdgpu: ring sdma0 uses VM inv eng 13 on hub 0
[    9.305568] amdgpu 0000:50:00.0: amdgpu: ring vcn_dec_0 uses VM inv eng 0 on hub 8
[    9.305568] amdgpu 0000:50:00.0: amdgpu: ring vcn_enc_0.0 uses VM inv eng 1 on hub 8
[    9.305569] amdgpu 0000:50:00.0: amdgpu: ring vcn_enc_0.1 uses VM inv eng 4 on hub 8
[    9.305570] amdgpu 0000:50:00.0: amdgpu: ring jpeg_dec uses VM inv eng 5 on hub 8
[    9.305812] amdgpu 0000:50:00.0: amdgpu: Runtime PM not available
[    9.306167] amdgpu 0000:50:00.0: [drm] Registered 4 planes with drm panic
[    9.306169] [drm] Initialized amdgpu 3.61.0 for 0000:50:00.0 on minor 0
[    9.308008] amdgpu 0000:50:00.0: [drm] Cannot find any crtc or sizes
[   13.467839] amdgpu 0000:03:00.0: vgaarb: VGA decodes changed: olddecodes=io+mem,decodes=none:owns=none
[   13.467842] amdgpu 0000:50:00.0: vgaarb: VGA decodes changed: olddecodes=io+mem,decodes=none:owns=none

Troubleshooting Steps Taken:

Observation: Confirmed the issue occurs randomly and persists in BIOS when stuck.
Fan Control Software (LACT):
- When the middle fan is behaving normally, LACT can control all fans (including the middle one, presumably via pwm1).
- When the middle fan is stuck at high speed, LACT commands have no effect on the middle fan (it stays at max speed), although LACT can still control the other two fans.
hwmon Interface: Confirmed only pwm1/fan1 are exposed. Haven't tried direct echo to pwm1 during the high-speed event, but LACT's failure to control it during the event (while normally being able to) suggests the fan is ignoring the pwm1 signal when malfunctioning.
Visual Inspection: No obvious obstructions or loose cables visible on the GPU fans.

Has anyone encountered similar intermittent fan behavior on this card or other AMD GPUs? Is there anything else I could check software-wise?

Thanks for any help or suggestions!

jl2 · 2025-04-28 19:04:59

Has anyone encountered similar intermittent fan behavior on this card or other AMD GPUs? Is there anything else I could check software-wise?

Did you try googling your issue? Otherwise, this isn't normal, I think your card is broken. If you still have waranty, make good use of it, otherwise, you will need to replace or remove (cutting it's wire should do the job) the fan (DON'T DO THIS YET, maybe somebody else has anything better to add).

Last edited by jl2 (2025-04-28 19:05:33)

Arch Linux

#1 2025-04-27 12:03:12

Intermittent Middle Fan on Sapphire PULSE RX 7900 XT

#2 2025-04-28 19:04:59

Re: Intermittent Middle Fan on Sapphire PULSE RX 7900 XT

Board footer