open-gpu-kernel-modules icon indicating copy to clipboard operation
open-gpu-kernel-modules copied to clipboard

GPU cant go to D3cold even when no process is running

Open tr1xem opened this issue 11 months ago • 5 comments

NVIDIA Open GPU Kernel Modules Version

570.133.07

Please confirm this issue does not happen with the proprietary driver (of the same version). This issue tracker is only for bugs specific to the open kernel driver.

  • [x] I confirm that this does not happen with the proprietary driver package.

Operating System and Version

CachyOS

Kernel Release

Linux cachyos 6.13.7-3-cachyos #1 SMP PREEMPT_DYNAMIC Wed, 19 Mar 2025 19:07:38 +0000 x86_64 GNU/Linux

Please confirm you are running a stable release kernel (e.g. not a -rc). We do not accept bug reports for unreleased kernels.

  • [x] I am running on a stable kernel release.

Hardware: GPU

GPU 0: NVIDIA GeForce RTX 4050 Laptop GPU (UUID: GPU-230091c2-98fc-0529-86de-0f02014e5b36)

Describe the bug

Im using rtd3 power management earlier on 0x03 that means the gpu should go to d3cold when not in use earlier (before 570) it was working fine gpu was sleeping good but now gpu is always no matter what i do its always D0

To Reproduce

Just update and monitor

Bug Incidence

Always

nvidia-bug-report.log.gz

nvidia-bug-report.log.gz

More Info

I cant test propitory as my system wont even boot using them

tr1xem avatar Mar 24 '25 12:03 tr1xem

@tr1xem Hi. Can you check out my comment here https://github.com/NVIDIA/open-gpu-kernel-modules/issues/905#issuecomment-3196986429 and see if your issue is related to mine (GPU oscillates between d0 and d3cold)?

imaGuru avatar Aug 25 '25 09:08 imaGuru

@tr1xem Hi. Can you check out my comment here #905 (comment) and see if your issue is related to mine (GPU oscillates between d0 and d3cold)?

it got fixed in 575

other than that tlp was issue because of it gpu cant sleep i removed tlp

Now gpu sleeps it just turns on for a bit while opening a xwayland app (hyprland mis config from myside as my pervious config is broken due to update) but yep above 575 everything works good for me

options nvidia NVreg_UsePageAttributeTable=1 \
    NVreg_InitializeSystemMemoryAllocations=0 \
    NVreg_DynamicPowerManagement=0x02 \
    NVreg_RegistryDwords=RMIntrLockingMode=1 \
    NVreg_EnableS0ixPowerManagement=1
    

my config

tr1xem avatar Aug 25 '25 10:08 tr1xem

Thanks

imaGuru avatar Aug 25 '25 13:08 imaGuru

@imaGuru Unfortunately this solution didn't work for me. However, I found out that for my issue, the GPU can go in and out of d3cold when nvidia_drm.modeset=0 is true. Unfortunately, when modeset is set to 0 my external monitor doesn't work because the HDMI port is directly connected to the NVIDIA GPU.

lonelyfloat avatar Oct 05 '25 05:10 lonelyfloat

@lonelyfloat thats weird see if ur display is conected ur gpu cant go to d3cold state anyways

tr1xem avatar Oct 07 '25 12:10 tr1xem