580 GSP crash on EDID

Graphics / Linux Linux
1

Hello

I have used my laptop with an external (FHD@75) monitor for a while now, but decided to buy a DELL S2725QS. It features 4k@120fps, VRR, HDR and connection over HDMI 2.1 (which my laptop has). My laptop has an AMD iGPU and an Nvidia dGPU of the Ampere family, and I’m running nvidia-open-dkms drivers (the recommended variant). Everything worked like a charm with the old monitor, but now, upon connecting the new monitor my nvidia GPU crashes. According to dmesg the GPU goes into reset required state so I ran the generate bug report tool after a restart, so that all GPU parameters can be queried.

In general, everything is fine until I connect the HDMI cable, which is when the dmesg is updated with a an error of Xid: 120, which is similiar to one reported here:

Xid 120: GSP load access page fault during driver init (575.51.02) · Issue #829 · NVIDIA/open-gpu-kernel-modules

In my case, however, the case is that the GSP crashes at a very specific moment, which according to my greenhorn debugging skills seems to be the moment it fetches EDID from the monitor. It also correlates with the fact that everything works fine with the driver UNTIL I plug in the monitor. Same happens with a cheap, non 2.1 HDMI cable. What helps is using a USB-C hub I have, which seems to further hint at EDID being at fault.

Dmesg

[   54.030270] NVRM: GPU at PCI:0000:01:00: GPU-bf47e61d-0cf9-7732-d665-2b1dc01ce151[   54.030274] NVRM: GPU Board Serial Number: 0[   54.030275] NVRM: Xid (PCI:0000:01:00): 120, pid=625, name=kworker/0:2, GSP task exception: load access page fault (cause:0xd) @ pc:0x12b017c, partition:2#0, task:3[   54.030280] NVRM:     Reported by libos partition:2#4 kernel v3.1 [0] @ ts:1761077830[   54.030283] NVRM:     RISC-V CSR State:[   54.030286] NVRM:         sstatus:0x0000000200000020  sscratch:0xffffffffa30134d0     sie:0x0000000000000220  sip:0x0000000000000000[   54.030289] NVRM:         sepc:0x00000000012b017c     stval:0x0000000000000009  scause:0x000000000000000d[   54.030292] NVRM:     RISC-V GPR State:[   54.030294] NVRM:         ra:0x00000000012b011e   sp:0x0000000300a0c4c0   gp:0x0000000000000000   tp:0x0000000000000000[   54.030297] NVRM:         a0:0x0000000000000000   a1:0x0000000000000004   a2:0x0000000300a0c570   a3:0x0000000900000000[   54.030300] NVRM:         a4:0x0000000000001000   a5:0x0000000000000000   a6:0x0000000029292929   a7:0x0000000000000006[   54.030303] NVRM:         s0:0x0000000300a0c640   s1:0x00000002fa01be90   s2:0x00000002fa1a2380   s3:0x00000002fa116d50[   54.030305] NVRM:         s4:0x0000000004160fe0   s5:0x0000000000000000   s6:0x00000002fa1a2370   s7:0x00000002fa01be90[   54.030308] NVRM:         s8:0x00000002fa013dd0   s9:0x00000002fa01b750  s10:0x00000002fa1a3370  s11:0x00000002fa01b590[   54.030311] NVRM:         t0:0x0000000000000001   t1:0x0000000000000002   t2:0x000000000ba01db8   t3:0x0000000000000000[   54.030313] NVRM:         t4:0x00000002fa01ae90   t5:0x0000000020000000   t6:0x000000000419b570[   54.030316] NVRM:     Stack Trace:[   54.030318] NVRM:         0x00000000012b017c[   54.030321] NVRM:         0x00000000012ad968[   54.030323] NVRM:         0x000000000121d89c[   54.030325] NVRM:         0x00000000018e3866[   54.030328] NVRM:         0x00000000011cc028[   54.030330] NVRM:         0x0000000001adc01c[   54.030333] NVRM:         0x00000000010554a4[   54.030335] NVRM:         0x0000000001055e12[   54.030337] NVRM:         0x0000000001a487a6[   54.030340] NVRM:         0x0000000001a56a5c[   54.030342] NVRM:         0x0000000001b8b9f2[   54.030344] NVRM:         0x0000000001a89c54[   54.030347] NVRM:         0x0000000001bed3b8[   54.030359] NVRM:         0x0000000001a85ee6[   54.030361] NVRM:     PC Trace:[   54.030364] NVRM:         0x00000000012b017c  0x0000000001b20f76  0x00000000012b011a  0x00000000012a9c1c  0x0000000001a8640c[   54.030367] NVRM:         0xffffffff93002e4c  0x0000000001a86430  0x00000000012a9bec  0x0000000001439ff0  0x0000000001b3ecf0[   54.030370] NVRM:         0xffffffff93002e4c  0x0000000001b3eca2  0x000000000143a028  0x000000000142e210  0xffffffff93002e4c[   54.030372] NVRM:         0x000000000142e1e6  0x0000000001439ff8  0xffffffff93002e4c  0x000000000143a00a  0x00000000012a9bc6[   54.030375] NVRM:         0x00000000017d1412  0x00000000017c5160  0x0000000001468252  0x00000000016fcdea  0x0000000001730254[   54.030378] NVRM:         0x00000000016fcda4  0x00000000014681c4  0x00000000017c5114  0x00000000017b9fbe  0x00000000017bff1c[   54.030380] NVRM:         0x00000000017b9fca  0x00000000017c5324  0x00000000018db628  0x00000000017c50e0  0x0000000001730254[   54.030383] NVRM:         0x00000000017c5720[   54.030385] NVRM:     Local I/O Register State:[   54.030387] NVRM:         0x01281500:0x00000000   0x01281504:0xbadf5620   0x01281508:0x00849970   0x01281510:0x00000000[   54.030391] NVRM:         0x01281528:0x01049042   0x0128152c:0x01000040   0x01281530:0x00000000   0x01280008:0x00000000[   54.030394] NVRM:     ------------[ end crash report ]------------[   54.030418] NVRM: GPU0 GSP RPC buffer contains function 76 (GSP_RM_CONTROL) sequence 1866 and data 0x000000000073029a 0x0000000000000010.[   54.030421] NVRM: GPU0 RPC history (CPU → GSP):[   54.030424] NVRM:     entry function                     sequence data0              data1              ts_start           ts_end             duration actively_polling[   54.030426] NVRM:      0    76   GSP_RM_CONTROL              1866 0x000000000073029a 0x0000000000000010 0x000641b0e287b219 0x0000000000000000          y[   54.030431] NVRM:     -1    76   GSP_RM_CONTROL              1865 0x00000000007302a2 0x0000000000000008 0x000641b0e287b07e 0x000641b0e287b20e    400us[   54.030435] NVRM:     -2    76   GSP_RM_CONTROL              1864 0x0000000000731152 0x0000000000000050 0x000641b0e287acd1 0x000641b0e287b070    927us[   54.030438] NVRM:     -3    76   GSP_RM_CONTROL              1863 0x000000000073028a 0x0000000000000014 0x000641b0e287ab5d 0x000641b0e287acc6    361us[   54.030442] NVRM:     -4    76   GSP_RM_CONTROL              1862 0x0000000000730293 0x000000000000000c 0x000641b0e287a55e 0x000641b0e287ab52   1524us[   54.030445] NVRM:     -5    76   GSP_RM_CONTROL              1861 0x00000000007302ae 0x0000000000000006 0x000641b0e287a38c 0x000641b0e287a552    454us[   54.030448] NVRM:     -6    76   GSP_RM_CONTROL              1860 0x0000000000730245 0x0000000000000810 0x000641b0e286b92c 0x000641b0e287a2f5  59849us[   54.030452] NVRM:     -7    76   GSP_RM_CONTROL              1859 0x0000000000730246 0x000000000000080c 0x000641b0e286b7cc 0x000641b0e286b920    340us[   54.030454] NVRM: GPU0 RPC event history (CPU ← GSP):[   54.030457] NVRM:     entry function                     sequence data0              data1              ts_start           ts_end             duration during_incomplete_rpc[   54.030460] NVRM:      0    4099 POST_EVENT                     0 0x0000000000000001 0x0000000000000000 0x000641b0e286aa2b 0x000641b0e286aa36     11us[   54.030463] NVRM:     -1    4108 UCODE_LIBOS_PRINT              0 0x0000000000000000 0x0000000000000000 0x000641b0e0fc27cc 0x000641b0e0fc27cd      1us[   54.030467] NVRM:     -2    4108 UCODE_LIBOS_PRINT              0 0x0000000000000000 0x0000000000000000 0x000641b0e0fc26a2 0x000641b0e0fc26a5      3us[   54.030470] NVRM:     -3    4098 GSP_RUN_CPU_SEQUENCER          0 0x000000000000061c 0x0000000000003fe2 0x000641b0e0fb40c1 0x000641b0e0fb54a8   5095us[   54.030474] NVRM:     -4    4108 UCODE_LIBOS_PRINT              0 0x0000000000000000 0x0000000000000000 0x000641b0e0ce052d 0x000641b0e0ce052d[   54.030477] NVRM:     -5    4111 PERF_BRIDGELESS_INFO_          0 0x0000000000000000 0x0000000000000000 0x000641b0e0cd8a42 0x000641b0e0cd8a43      1us[   54.030480] NVRM:     -6    4108 UCODE_LIBOS_PRINT              0 0x0000000000000000 0x0000000000000000 0x000641b0df8bad62 0x000641b0df8bad63      1us[   54.030484] NVRM:     -7    4108 UCODE_LIBOS_PRINT              0 0x0000000000000000 0x0000000000000000 0x000641b0df8bac5f 0x000641b0df8bac61      2us[   54.030614] nvidia-modeset: ERROR: GPU:0: Failure reading DP dongle info for display device DELL S2725QS (HDMI-0).[   54.030625] nvidia-modeset: ERROR: GPU:0: Failed to query VR headset for DELL S2725QS (HDMI-0)[   54.030652] NVRM: Xid (PCI:0000:01:00): 154, GPU recovery action changed from 0x0 (None) to 0x1 (GPU Reset Required)[   54.040412] nvidia 0000:01:00.0: [drm] Cannot find any crtc or sizes[   54.040440] nvidia-modeset: ERROR: GPU:0: Failed detecting connected display devices[   54.040463] nvidia-modeset: ERROR: GPU:0: Failed detecting connected display devices[   54.040481] nvidia-modeset: ERROR: GPU:0: Failed detecting connected display devices[   54.040503] nvidia-modeset: ERROR: GPU:0: Failure reading maximum pixel clock value for display device HDMI-0.[   54.040515] nvidia 0000:01:00.0: [drm] Cannot find any crtc or sizes[   54.040555] nvidia-modeset: ERROR: GPU:0: Failed detecting connected display devices[   54.040580] nvidia-modeset: ERROR: GPU:0: Failed detecting connected display devices[   54.040602] nvidia-modeset: ERROR: GPU:0: Failed detecting connected display devices[   54.040619] nvidia-modeset: ERROR: GPU:0: Failed detecting connected display devices[   54.040628] nvidia 0000:01:00.0: [drm] Cannot find any crtc or sizes[   54.040685] nvidia-modeset: ERROR: GPU:0: Failed detecting connected display devices[   54.040717] nvidia-modeset: ERROR: GPU:0: Failed detecting connected display devices[   54.040892] nvidia-modeset: ERROR: GPU:0: Failed detecting connected display devices[   54.040939] nvidia-modeset: ERROR: GPU:0: Failed detecting connected display devices[   54.040966] nvidia-modeset: ERROR: GPU:0: Failed detecting connected display devices[   54.042165] nvidia-modeset: ERROR: GPU:0: Failed detecting connected display devices[   54.042205] nvidia-modeset: ERROR: GPU:0: Failed detecting connected display devices[   54.042233] nvidia-modeset: ERROR: GPU:0: Failed detecting connected display devices

EDIT: Sorry, I’m not sure how to format this on the forum to be multiline, but it’s also present in the attached log.

Other things I checked

  • monitor settings, but could not find anything like ECO mode, which is supposed to break Linux systems according to Arch Wiki.
  • I can also reproduce the error on both the closed and open drivers and even with GSP disabled, although in the latter case, my computer simply freezes and becomes non-functional.
  • tried disabling HDR with options nvidia-drm NVreg_EnableHDR=0
  • I get the same error when launched in TTY only mode, i.e. with systemd.unit=multi-user.target appended to my kernel cmdline, so I doubt it’s an issue with my DE
  • I have dual boot and the monitor works alright on Windows
  • it’s not an issue with hybrid graphic, as I disabled that in BIOS

The next steps for me are to borrow a USB-C to DisplayPort cable, but I would still like to get to the bottom of this, since I’ve sunk so much time into it. Right now I can use my monitor with the USB hub, but far from its max capacity as it is a shitty hub ;)

after-plug-nvidia-bug-report.log.gz (529.7 KB)

after-plug-and-restart-nvidia-bug-report.log.gz (1.6 MB)

EDIT: I extracted the EDID file in Windows. Curiously, read-edid crashes when parsing extension blocks

(change txt to bin)

dell_monitor_edid.txt (384 Bytes)

1 Like
2

same stack trace as No HDMI output on Nobara Linux 42, from Lenovo Legion 7i w/ RTX 3080
it’s a 580.95.05 regression, so downgrade back to 580.82.09 in the meanwhile. NVIDIA still hasn’t acknowledged it while I reported it a couple of weeks ago (trillion dollars company btw).

we all encounter this ONLY on legion laptops, so might be a manufacturer issue as well.

1 Like
3

Same here.

Reproduction:

  1. launch browser(librewolf or other)
  2. press F11 to go fullscreen.
  3. press F11 again to quit fullscreen
  4. Crash

Fedora 42 (kde) @ Kernel 6.17.4

GPU: 5090

Display: PG32UCDP

Tested driver:

NVIDIA UNIX Open Kernel Mode Setting Driver for x86_64 580.95.05 Release Build (dvs-builder@U22-I3-B17-02-5) Tue Sep 23 09:42:01 UTC 2025

NVIDIA UNIX Open Kernel Module for x86_64 580.82.09 Release Build (akmods@[hostname]) Sun Oct 26 07:13:23 EDT 2025

image
image1928×1879 832 KB

Some stacktrace related to nvidia lib

#4  0x000071e316971592 n/a (libnvidia-eglcore.so.580.95.05 + 0xb71592)
#5  0x000071e3169869fd n/a (libnvidia-eglcore.so.580.95.05 + 0xb869fd)
#6  0x000071e316982e3f n/a (libnvidia-eglcore.so.580.95.05 + 0xb82e3f)
#7  0x000071e316553b4e n/a (libnvidia-eglcore.so.580.95.05 + 0x753b4e)
#8  0x000071e31655d877 n/a (libnvidia-eglcore.so.580.95.05 + 0x75d877)
4

laptop? destkop? full nvidia-bug-report-log?

5

I’m on desktop (5090). Sorry I didn’t generate bug report and I’m unable to do so because I switch back to my old 7900XTX from team red. Here is another log that I was able to capture.

image
image1280×327 169 KB

6

Can confirm that downgrading the driver helps. However, I do get choppy framerates with little GPU utilization. GLXGears reports 40 FPS on my external monitor while the laptop screen works fine, but this may well be an issue with my Hyprland setup.