AMDGPU random crashes on Ryzen AI HX 370 (6.14.4-arch1-1)

LS_15B1ADE5 · April 30, 2025, 10:02pm

Hi,
just wanted to know if anyone else has similar issue with amdgpu crashes on Linux:

I’m using Arch, up to date as of 29.04.2025, kernel 6.14.4-arch1-1 on FW 13 with HX 370.
Usually, while using Firefox (especially when starting videos) or Kitty terminal, the amdgpu module may crash at random moment. Sometimes the affected process continue to work after amdgpu resumed, sometimes process crashes as well.

Typical crash log:

amdgpu 0000:c1:00.0: amdgpu: Dumping IP State
amdgpu 0000:c1:00.0: amdgpu: Dumping IP State Completed
amdgpu 0000:c1:00.0: amdgpu: ring vcn_unified_0 timeout, signaled seq=77533, emitted seq=77535
amdgpu 0000:c1:00.0: amdgpu: Process information: process RDD Process pid 61318 thread firefox:cs0 pid 62586
amdgpu 0000:c1:00.0: amdgpu: GPU reset begin!
amdgpu 0000:c1:00.0: amdgpu: MODE2 reset
amdgpu 0000:c1:00.0: amdgpu: GPU reset succeeded, trying to resume
[drm] PCIE GART of 512M enabled (table at 0x0000008001700000).
amdgpu 0000:c1:00.0: amdgpu: SMU is resuming...
amdgpu 0000:c1:00.0: amdgpu: SMU is resumed successfully!
[drm] DMUB hardware initialized: version=0x09001B00
amdgpu 0000:c1:00.0: amdgpu: ring gfx_0.0.0 uses VM inv eng 0 on hub 0
amdgpu 0000:c1:00.0: amdgpu: ring comp_1.0.0 uses VM inv eng 1 on hub 0
amdgpu 0000:c1:00.0: amdgpu: ring comp_1.1.0 uses VM inv eng 4 on hub 0
amdgpu 0000:c1:00.0: amdgpu: ring comp_1.2.0 uses VM inv eng 6 on hub 0
amdgpu 0000:c1:00.0: amdgpu: ring comp_1.3.0 uses VM inv eng 7 on hub 0
amdgpu 0000:c1:00.0: amdgpu: ring comp_1.0.1 uses VM inv eng 8 on hub 0
amdgpu 0000:c1:00.0: amdgpu: ring comp_1.1.1 uses VM inv eng 9 on hub 0
amdgpu 0000:c1:00.0: amdgpu: ring comp_1.2.1 uses VM inv eng 10 on hub 0
amdgpu 0000:c1:00.0: amdgpu: ring comp_1.3.1 uses VM inv eng 11 on hub 0
amdgpu 0000:c1:00.0: amdgpu: ring sdma0 uses VM inv eng 12 on hub 0
amdgpu 0000:c1:00.0: amdgpu: ring vcn_unified_0 uses VM inv eng 0 on hub 8
amdgpu 0000:c1:00.0: amdgpu: ring jpeg_dec_0 uses VM inv eng 1 on hub 8
amdgpu 0000:c1:00.0: amdgpu: ring mes_kiq_3.1.0 uses VM inv eng 13 on hub 0
amdgpu 0000:c1:00.0: amdgpu: ring vpe uses VM inv eng 4 on hub 8
amdgpu 0000:c1:00.0: amdgpu: GPU reset(3) succeeded!

Also during load amdgpu module says Optional firmware ... was not found:

amdgpu: ATOM BIOS: 113-STRIXEMU-001
amdgpu 0000:c1:00.0: amdgpu: VPE: collaborate mode false
amdgpu 0000:c1:00.0: amdgpu: [drm] Optional firmware "amdgpu/isp_4_1_0.bin" was not found
amdgpu 0000:c1:00.0: amdgpu: Trusted Memory Zone (TMZ) feature disabled as experimental (default)
amdgpu 0000:c1:00.0: amdgpu: VRAM: 512M 0x0000008000000000 - 0x000000801FFFFFFF (512M used)
amdgpu 0000:c1:00.0: amdgpu: GART: 512M 0x00007FFF00000000 - 0x00007FFF1FFFFFFF
[drm] amdgpu: 512M of VRAM memory ready
[drm] amdgpu: 31787M of GTT memory ready.

Probably unrelated problem, but I’m also getting USB C errors from time to time:

ucsi_acpi USBC000:00: unknown error 256
or
ucsi_acpi USBC000:00: unknown error 0

System Info:

Motherboard: Framework Laptop 13 AMD Ryzen 9 HX 370
Memory: 64.0 GiB
BIOS Version: 3.03 (latest)
OS: Arch Linux x86_64
Kernel: 6.14.4-arch1-1
Window Manager: Hyprland (Wayland)

Thanks!

UPD:

I see at least one person reporting the same problems:

ava1ar · May 1, 2025, 12:35am

I confirm I have both problems as stated in my posts. For gpu issue we need to wait for new kernel and/or firmware. For PD issue we probably need new BIOS/firmware update from Framework. Hardware is still new, but software will catch up pretty soon.

Mario_Limonciello · May 1, 2025, 1:59pm

The VCN issue is tracked upstream here: https://gitlab.freedesktop.org/mesa/mesa/-/issues/12528

It is still being investigated, not yet sure if it will be a kernel patch or FW update to fix it.

The ISP firmware is optional, it is only needed for designs that a camera sensor is connected directly to the APU (IE not a USB camera).

mordisko · May 6, 2025, 3:32pm

Just sharing that I’m on the same boat. Details:

FW 13", AI H370 (batch 2)
Ubuntu 24 (6.11.0-25-generic)
Latest firmware
Mostly breaks when using Firefox
I’m using an extra monitor connected to an HDMI expansion card

amdgpu crashes, then resumes. This causes the extra monitor goes black, the system to become unresponsive, and the audio to still keep playing.

My crash log:

may 06 17:00:48 cheshire kernel: amdgpu 0000:c1:00.0: amdgpu: ring vcn_unified_0 timeout, signaled seq=206212>
may 06 17:00:48 cheshire kernel: amdgpu 0000:c1:00.0: amdgpu: Process information: process RDD Process pid 34>
may 06 17:00:48 cheshire kernel: amdgpu 0000:c1:00.0: amdgpu: GPU reset begin!
may 06 17:00:48 cheshire kernel: amdgpu 0000:c1:00.0: amdgpu: Dumping IP State
may 06 17:00:48 cheshire kernel: amdgpu 0000:c1:00.0: amdgpu: Dumping IP State Completed
may 06 17:00:48 cheshire kernel: amdgpu 0000:c1:00.0: amdgpu: MODE2 reset
may 06 17:00:48 cheshire kernel: amdgpu 0000:c1:00.0: amdgpu: GPU reset succeeded, trying to resume
may 06 17:00:48 cheshire kernel: [drm] PCIE GART of 512M enabled (table at 0x0000008001700000).
may 06 17:00:48 cheshire kernel: amdgpu 0000:c1:00.0: amdgpu: SMU is resuming...
may 06 17:00:48 cheshire kernel: amdgpu 0000:c1:00.0: amdgpu: SMU is resumed successfully!
may 06 17:00:48 cheshire kernel: [drm] DMUB hardware initialized: version=0x0000D500
may 06 17:00:49 cheshire kernel: amdgpu 0000:c1:00.0: amdgpu: ring gfx_0.0.0 uses VM inv eng 0 on hub 0
may 06 17:00:49 cheshire kernel: amdgpu 0000:c1:00.0: amdgpu: ring comp_1.0.0 uses VM inv eng 1 on hub 0
may 06 17:00:49 cheshire kernel: amdgpu 0000:c1:00.0: amdgpu: ring comp_1.1.0 uses VM inv eng 4 on hub 0
may 06 17:00:49 cheshire kernel: amdgpu 0000:c1:00.0: amdgpu: ring comp_1.2.0 uses VM inv eng 6 on hub 0
may 06 17:00:49 cheshire kernel: amdgpu 0000:c1:00.0: amdgpu: ring comp_1.3.0 uses VM inv eng 7 on hub 0
may 06 17:00:49 cheshire kernel: amdgpu 0000:c1:00.0: amdgpu: ring comp_1.0.1 uses VM inv eng 8 on hub 0
may 06 17:00:49 cheshire kernel: amdgpu 0000:c1:00.0: amdgpu: ring comp_1.1.1 uses VM inv eng 9 on hub 0
may 06 17:00:49 cheshire kernel: amdgpu 0000:c1:00.0: amdgpu: ring comp_1.2.1 uses VM inv eng 10 on hub 0
may 06 17:00:49 cheshire kernel: amdgpu 0000:c1:00.0: amdgpu: ring comp_1.3.1 uses VM inv eng 11 on hub 0
may 06 17:00:49 cheshire kernel: amdgpu 0000:c1:00.0: amdgpu: ring sdma0 uses VM inv eng 12 on hub 0
may 06 17:00:49 cheshire kernel: amdgpu 0000:c1:00.0: amdgpu: ring vcn_unified_0 uses VM inv eng 0 on hub 8
may 06 17:00:49 cheshire kernel: amdgpu 0000:c1:00.0: amdgpu: ring jpeg_dec_0 uses VM inv eng 1 on hub 8
may 06 17:00:49 cheshire kernel: amdgpu 0000:c1:00.0: amdgpu: ring mes_kiq_3.1.0 uses VM inv eng 13 on hub 0
may 06 17:00:49 cheshire kernel: amdgpu 0000:c1:00.0: amdgpu: ring vpe uses VM inv eng 4 on hub 8
may 06 17:00:49 cheshire kernel: amdgpu 0000:c1:00.0: amdgpu: recover vram bo from shadow start
may 06 17:00:49 cheshire kernel: amdgpu 0000:c1:00.0: amdgpu: recover vram bo from shadow done
may 06 17:00:49 cheshire kernel: amdgpu 0000:c1:00.0: amdgpu: GPU reset(1) succeeded!

Der_Dackel · May 23, 2025, 11:18am

I think, I am in the same boat. I could however tie my GPU driver instability specifically to Firefox hardware acceleration after its recent 138.0.3 update.

FW 13 Ryzen AI HX 370
96 GB RAM
CachyOS (experienced on Kernel 6.14.4 through 6.14.8)
Firmware 3.03
Running KDE 6.3 on Wayland
External monitors do not seem to play a role.

[18551.996374] amdgpu 0000:c1:00.0: amdgpu: Dumping IP State
[18551.998145] amdgpu 0000:c1:00.0: amdgpu: Dumping IP State Completed
[18551.998196] amdgpu 0000:c1:00.0: amdgpu: ring vcn_unified_0 timeout, signaled seq=3602701, emitted seq=3602703
[18551.998198] amdgpu 0000:c1:00.0: amdgpu: Process information: process RDD Process pid 3568 thread firefox:cs0 pid 10821
[18551.998201] amdgpu 0000:c1:00.0: amdgpu: GPU reset begin!
[18552.165466] amdgpu 0000:c1:00.0: amdgpu: MODE2 reset
[18552.183668] amdgpu 0000:c1:00.0: amdgpu: GPU reset succeeded, trying to resume
[18552.184212] [drm] PCIE GART of 512M enabled (table at 0x0000008001700000).
[18552.184282] amdgpu 0000:c1:00.0: amdgpu: SMU is resuming...
[18552.188578] amdgpu 0000:c1:00.0: amdgpu: SMU is resumed successfully!
[18552.196621] [drm] DMUB hardware initialized: version=0x09001B00
[18552.322599] amdgpu 0000:c1:00.0: amdgpu: ring gfx_0.0.0 uses VM inv eng 0 on hub 0
[18552.322605] amdgpu 0000:c1:00.0: amdgpu: ring comp_1.0.0 uses VM inv eng 1 on hub 0
[18552.322607] amdgpu 0000:c1:00.0: amdgpu: ring comp_1.1.0 uses VM inv eng 4 on hub 0
[18552.322608] amdgpu 0000:c1:00.0: amdgpu: ring comp_1.2.0 uses VM inv eng 6 on hub 0
[18552.322608] amdgpu 0000:c1:00.0: amdgpu: ring comp_1.3.0 uses VM inv eng 7 on hub 0
[18552.322609] amdgpu 0000:c1:00.0: amdgpu: ring comp_1.0.1 uses VM inv eng 8 on hub 0
[18552.322610] amdgpu 0000:c1:00.0: amdgpu: ring comp_1.1.1 uses VM inv eng 9 on hub 0
[18552.322611] amdgpu 0000:c1:00.0: amdgpu: ring comp_1.2.1 uses VM inv eng 10 on hub 0
[18552.322612] amdgpu 0000:c1:00.0: amdgpu: ring comp_1.3.1 uses VM inv eng 11 on hub 0
[18552.322612] amdgpu 0000:c1:00.0: amdgpu: ring sdma0 uses VM inv eng 12 on hub 0
[18552.322613] amdgpu 0000:c1:00.0: amdgpu: ring vcn_unified_0 uses VM inv eng 0 on hub 8
[18552.322614] amdgpu 0000:c1:00.0: amdgpu: ring jpeg_dec_0 uses VM inv eng 1 on hub 8
[18552.322615] amdgpu 0000:c1:00.0: amdgpu: ring mes_kiq_3.1.0 uses VM inv eng 13 on hub 0
[18552.322616] amdgpu 0000:c1:00.0: amdgpu: ring vpe uses VM inv eng 4 on hub 8
[18552.326663] amdgpu 0000:c1:00.0: amdgpu: GPU reset(1) succeeded!

I have confirmed that downgrading Firefox to 138.0.1 resolves the issue for me completely. The 138.0.3 update seems to have addressd some bugs with both WebGL and video acceleration (changelog). “vcn_unified_0” being the source of the driver crash points to the video decoder. I was close to opening a bug on Mozilla’s bugtracker, but the fact that I couldn’t find anyone else with the problem outside of this forum thread, makes me reconsider whether it’s an issue specific to the new FW13 mainboard.

Were you folks also running Firefox, or one of its forks/redistributions (e.g. LIbreWolf, Zen, Tor Browser…) when the crashes happened?

EDIT: I forgot: Problem frequency went way down on newer Firefox versiosn when disabling hardware video acceleration, another indicator pointing towards VCN.

Mario_Limonciello · May 23, 2025, 1:20pm

There is a bug with racing on VCN. It’s fixed in 6.15-rc7. it will come back to the next 6.14.y.

Here’s the patches for 6.14 if you want to build early.

Der_Dackel · May 23, 2025, 1:44pm

I was just looking through the thread on Mesa’s bugtracker. I had completely overlooked your first post when scrolling through this one here. I just switched to 6.15.rc7 and will report whether that fixes the issue. Thanks for your work on getting to the bottom of this!

Rahat_Ahmed · May 28, 2025, 4:24am

Did this end up fixing the issue for you? Curious if I should bother switching or just deal with it for a few days until arch gets the update.

I have similar logs, but I’ll add another symptom for other folks searching for this: I’m running Hyprland and while I do have the same occasional flashes of a black screen like the others mentioned, every now and then Hyprland itself will crash, too. The last line in this crash report seems like it might be related (but I’m not really sure):

--------------------------------------------
   Hyprland Crash Report
--------------------------------------------
*thud*

Hyprland received signal 6(ABRT)
Version: 9958d297641b5c84dcff93f9039d80a5ad37ab00
Tag: v0.49.0
Date: Thu May 8 20:15:18 2025
Flags:

System info:
	System name: Linux
	Node name: hopeitframeworks
	Release: 6.14.7-arch2-1
	Version: #1 SMP PREEMPT_DYNAMIC Thu, 22 May 2025 05:37:49 +0000

GPU:
	c1:00.0 Display controller [0380]: Advanced Micro Devices, Inc. [AMD/ATI] Krackan [Radeon 840M / 860M Graphics] [1002:1114] (rev c2)


os-release:
	NAME="Arch Linux"
	PRETTY_NAME="Arch Linux"
	ID=arch
	BUILD_ID=rolling
	ANSI_COLOR="38;2;23;147;209"
	HOME_URL="https://archlinux.org/"
	DOCUMENTATION_URL="https://wiki.archlinux.org/"
	SUPPORT_URL="https://bbs.archlinux.org/"
	BUG_REPORT_URL="https://gitlab.archlinux.org/groups/archlinux/-/issues"
	PRIVACY_POLICY_URL="https://terms.archlinux.org/docs/privacy-policy/"
	LOGO=archlinux-logo
Backtrace:
	# | Hyprland(_Z12getBacktracev+0x61) [0x55fbedef15c1]
		getBacktrace()
		??:?
	#1 | Hyprland(_ZN14NCrashReporter18createAndSaveCrashEi+0x11b7) [0x55fbede47557]
		NCrashReporter::createAndSaveCrash(int)
		??:?
	#2 | Hyprland(+0x210bd8) [0x55fbeddbabd8]
		CCompositor::getMonitorFromCursor()
		??:?
	#3 | /usr/lib/libc.so.6(+0x3def0) [0x7fd416620ef0]
		??
		??:0
	#4 | /usr/lib/libc.so.6(+0x9774c) [0x7fd41667a74c]
		??
		??:0
	#5 | /usr/lib/libc.so.6(gsignal+0x20) [0x7fd416620dc0]
		??
		??:0
	#6 | Hyprland(_ZN15CHyprOpenGLImpl5beginEN9Hyprutils6Memory14CSharedPointerI8CMonitorEERKNS0_4Math7CRegionEP12CFramebufferSt8optionalIS6_E+0x9be) [0x55fbee1d8fae]
		CHyprOpenGLImpl::begin(Hyprutils::Memory::CSharedPointer<CMonitor>, Hyprutils::Math::CRegion const&, CFramebuffer*, std::optional<Hyprutils::Math::CRegion>)
		??:?
	#7 | Hyprland(_ZN13CHyprRenderer11beginRenderEN9Hyprutils6Memory14CSharedPointerI8CMonitorEERNS0_4Math7CRegionE11eRenderModeNS2_I9IHLBufferEEP12CFramebufferb+0x6c5) [0x55fbee206b85]
		CHyprRenderer::beginRender(Hyprutils::Memory::CSharedPointer<CMonitor>, Hyprutils::Math::CRegion&, eRenderMode, Hyprutils::Memory::CSharedPointer<IHLBuffer>, CFramebuffer*, bool)
		??:?
	#8 | Hyprland(_ZN13CHyprRenderer13renderMonitorEN9Hyprutils6Memory14CSharedPointerI8CMonitorEE+0x7de) [0x55fbee20742e]
		CHyprRenderer::renderMonitor(Hyprutils::Memory::CSharedPointer<CMonitor>)
		??:?
	#9 | Hyprland(_ZN8CMonitor14onMonitorFrameEv+0x604) [0x55fbedf08c24]
		CMonitor::onMonitorFrame()
		??:?
	#1 | Hyprland(+0x35ed98) [0x55fbedf08d98]
		CMonitor::onMonitorFrame()
		??:?
	#11 | /usr/lib/libhyprutils.so.6(_ZN9Hyprutils6Signal15CSignalListener4emitESt3any+0x67) [0x7fd417261fd7]
		??
		??:0
	#12 | /usr/lib/libhyprutils.so.6(_ZN9Hyprutils6Signal7CSignal4emitESt3any+0x293) [0x7fd4172623f3]
		??
		??:0
	#13 | /usr/lib/libaquamarine.so.7(+0xb398e) [0x7fd41736998e]
		??
		??:0
	#14 | /usr/lib/libdrm.so.2(drmHandleEvent+0x130) [0x7fd416ec1f30]
		??
		??:0
	#15 | /usr/lib/libaquamarine.so.7(_ZN10Aquamarine11CDRMBackend14dispatchEventsEv+0x7a) [0x7fd4173688da]
		??
		??:0
	#16 | Hyprland(+0x428efa) [0x55fbedfd2efa]
		CHyprXWaylandManager::sendCloseWindow(Hyprutils::Memory::CSharedPointer<CWindow>)
		??:?
	#17 | /usr/lib/libwayland-server.so.0(wl_event_loop_dispatch+0x1f2) [0x7fd417149112]
		??
		??:0
	#18 | /usr/lib/libwayland-server.so.0(wl_display_run+0x37) [0x7fd41714b1f7]
		??
		??:0
	#19 | Hyprland(_ZN17CEventLoopManager9enterLoopEv+0x1aa) [0x55fbedfd612a]
		CEventLoopManager::enterLoop()
		??:?
	#2 | Hyprland(main+0x1002) [0x55fbedd63172]
		main
		??:?
	#21 | /usr/lib/libc.so.6(+0x276b5) [0x7fd41660a6b5]
		??
		??:0
	#22 | /usr/lib/libc.so.6(__libc_start_main+0x89) [0x7fd41660a769]
		??
		??:0
	#23 | Hyprland(_start+0x25) [0x55fbeddb5da5]
		_start
		??:?


Log tail:
[LOG] [CLinuxDMABUFParamsResource]  | plane 0: mod 144115188214496004 fd 136 stride 11776 offset 0
[LOG] [CLinuxDMABUFParamsResource]  | plane 1: mod 144115188214496004 fd 147 stride 3072 offset 22609920
[LOG] [CLinuxDMABUFParamsResource]  | plane 2: mod 144115188214496004 fd 151 stride 3072 offset 22708224
[LOG] [CLinuxDMABUFParamsResource] Creating a dmabuf, with id 94: size [Vector2D: x: 2852, y: 1824], fmt AR24, planes 3
[LOG] [CLinuxDMABUFParamsResource]  | plane 0: mod 144115188214496004 fd 111 stride 11776 offset 0
[LOG] [CLinuxDMABUFParamsResource]  | plane 1: mod 144115188214496004 fd 143 stride 3072 offset 22609920
[LOG] [CLinuxDMABUFParamsResource]  | plane 2: mod 144115188214496004 fd 159 stride 3072 offset 22708224
[LOG] [CLinuxDMABUFParamsResource] Creating a dmabuf, with id 76: size [Vector2D: x: 2852, y: 1824], fmt AR24, planes 3
[LOG] [CLinuxDMABUFParamsResource]  | plane 0: mod 144115188214496004 fd 136 stride 11776 offset 0
[LOG] [CLinuxDMABUFParamsResource]  | plane 1: mod 144115188214496004 fd 138 stride 3072 offset 22609920
[LOG] [CLinuxDMABUFParamsResource]  | plane 2: mod 144115188214496004 fd 151 stride 3072 offset 22708224
[LOG] [CLinuxDMABUFParamsResource] Creating a dmabuf, with id 94: size [Vector2D: x: 2852, y: 1824], fmt AR24, planes 3
[LOG] [CLinuxDMABUFParamsResource]  | plane 0: mod 144115188214496004 fd 111 stride 11776 offset 0
[LOG] [CLinuxDMABUFParamsResource]  | plane 1: mod 144115188214496004 fd 143 stride 3072 offset 22609920
[LOG] [CLinuxDMABUFParamsResource]  | plane 2: mod 144115188214496004 fd 147 stride 3072 offset 22708224
[LOG] [AQ] [libinput] event8  - button state: touch 1 from BUTTON_STATE_AREA    event BUTTON_EVENT_UP          to BUTTON_STATE_NONE   

[LOG] [AQ] [libinput] event8  - button state: touch 0 from BUTTON_STATE_AREA    event BUTTON_EVENT_UP          to BUTTON_STATE_NONE   

[LOG] [AQ] [libinput] event8  - gesture: [2fg] event GESTURE_STATE_SCROLL → GESTURE_EVENT_END → GESTURE_STATE_NONE

[LOG] [AQ] [libinput] event8  - tap: touch 0 (TOUCH_END), tap state TAP_STATE_DEAD → TAP_EVENT_RELEASE → TAP_STATE_IDLE

[LOG] [AQ] [libinput] event8  - button state: touch 0 from BUTTON_STATE_NONE    event BUTTON_EVENT_IN_AREA     to BUTTON_STATE_AREA   

[LOG] [AQ] [libinput] event8  - button state: touch 1 from BUTTON_STATE_NONE    event BUTTON_EVENT_IN_AREA     to BUTTON_STATE_AREA   

[LOG] [AQ] [libinput] event8  - tap: touch 0 (TOUCH_BEGIN), tap state TAP_STATE_IDLE → TAP_EVENT_TOUCH → TAP_STATE_TOUCH

[LOG] [AQ] [libinput] event8  - tap: touch 1 (TOUCH_BEGIN), tap state TAP_STATE_TOUCH → TAP_EVENT_TOUCH → TAP_STATE_TOUCH_2

[LOG] [AQ] [libinput] event8  - gesture: [2fg] event GESTURE_STATE_NONE → GESTURE_EVENT_FINGER_DETECTED → GESTURE_STATE_UNKNOWN

[LOG] [AQ] [libinput] event8  - gesture: [2fg] state GESTURE_STATE_NONE → GESTURE_STATE_UNKNOWN

[LOG] [AQ] [libinput] event8  - gesture: [2fg] event GESTURE_STATE_UNKNOWN → GESTURE_EVENT_HOLD_TIMEOUT → GESTURE_STATE_HOLD

[LOG] [AQ] [libinput] event8  - tap: touch 1 (TOUCH_UPDATE), tap state TAP_STATE_TOUCH_2 → TAP_EVENT_MOTION → TAP_STATE_DEAD

[LOG] [AQ] [libinput] event8  - gesture: [2fg] event GESTURE_STATE_HOLD → GESTURE_EVENT_CANCEL → GESTURE_STATE_NONE

[LOG] [AQ] [libinput] event8  - gesture: [2fg] event GESTURE_STATE_HOLD → GESTURE_EVENT_SCROLL_START → GESTURE_STATE_SCROLL_START

[LOG] [AQ] [libinput] event8  - gesture: [2fg] state GESTURE_STATE_HOLD → GESTURE_STATE_SCROLL_START → GESTURE_STATE_SCROLL

[ERR] [EGL] Command eglDupNativeFenceFDANDROID errored out with EGL_BAD_PARAMETER (0x12300): eglDupNativeFenceFDANDROID
[ERR] eglDupNativeFenceFDANDROID failed
[ERR] renderer: couldn't use EGLSync for explicit gpu synchronization
[CRITICAL] 
==========================================================================================
ASSERTION FAILED! 

Aborting, glGetGraphicsResetStatus returned GL_INNOCENT_CONTEXT_RESET. Cannot continue until proper GPU reset handling is implemented.

at: line 734 in OpenGL.cpp

Der_Dackel · May 28, 2025, 5:57am

I wanted to make sure and used it for two full work days. I can say that at least for myself, on linux-cachyos-rc 6.15-rc7 and then linux-cachyos 6.15.0, the problem completely went away and I haven’t had a single GPU reset in four days.

mordisko · May 28, 2025, 9:17pm

It did. Nevertheless, there are more amdgpu issues that affect older kernels so, while this is much more stable now, I sometimes get random freezes caused by another issue.

Adam_AS_Strauch · May 29, 2025, 2:54pm

I can confirm, that my Fedora Silverblue with 6.15.0 have had zero GPU crashes since I updated.

The patch mentioned by @Mario_Limonciello is now in 6.14.9.

badlydrawnface · May 30, 2025, 5:32am

My Ryzen AI 7 350 Framework 13 has had the same problems. I will try to load a video on say, Twitter, and then the video buffers, followed by a second-or-so-long GPU crash and the video suddenly plays. Sometimes two or three hangs occur, and I am bumped out of my desktop to either the display manager or TTY (depending on whether or not I am using a DM). In addition, these issues are present in every single browser and Linux distro I’ve used, Fedora, Bluefin, NixOS, everything.

I hope this gets fixed soon, I’ve heard that AMD graphics on Linux are very unstable for around the first year before changes come upstream. I have not had these issues on Windows.

badlydrawnface · May 30, 2025, 5:45am

I gotta learn how to read…

Installed 6.15, I’ll let y’all know if I get another hang.

Topic		Replies	Views
[RESPONDED] Crashing amdgpu on AMD Ryzen 7040 13-inch (Ubuntu 22.04) Linux ubuntu	16	4083	August 1, 2024
[RESPONDED] VRAM is lost due to GPU reset! (followed by a crash) Linux	28	2527	August 16, 2024
Amdgpu crashes on FW 13 HX 370 Linux ubuntu	1	383	May 9, 2025
Youtube + Firefox + AMD = Display Crash Linux fedora	36	2340	July 26, 2024
AMD Framework 13 with Debian still crashing a year later Linux debian	19	1442	May 18, 2025

AMDGPU random crashes on Ryzen AI HX 370 (6.14.4-arch1-1)

System Info:

UPD:

Related topics