575 BETA release feedback & discussion

[birdie]

NVIDIA has released Linux x64 (AMD64/EM64T) Display Driver 575.51.02 BETA.

Changelog:

• Extended the __NV_DISABLE_EXPLICIT_SYNC environment variable, which was available to EGL applications, to also apply to GLX and Vulkan applications.
• Fixed a bug that could cause Marvel Rivals to crash on startup or when loading levels.
• Fixed a bug that could cause the applications that use the VK_KHR_present_wait extension to hang on Wayland.
• Added support for GLX front buffer rendering on Xwayland.
• Fixed a bug that could cause Minecraft to crash on Xwayland.
• Fixed a bug that prevented PRIME Render Offload from working correctly when using NVIDIA GPUs as both the render offload source and the render offload sink.
• Fixed a bug which prevented VRR from working when overriding an EDID through the /sys/kernel/debug/dri/*/edid_override interface.
• Added support for the DRM plane properties COLOR_ENCODING and COLOR_RANGE.
• Fixed a bug that prevented the Default TGP and Max TGP values from being reported in the nvidia-settings control panel while running notebook systems on battery power.
• Fixed a bug that could lead to display freezes on some systems when toggling Night Mode with GNOME on Wayland.
• Fixed a bug that could cause graphics applications to not render correctly after a system suspend/resume cycle, if using the nvidia.ko kernel module parameter NVreg_PreserveVideoMemoryAllocations=1.
• Added a new kernel module parameter, conceal_vrr_caps, to the nvidia-modeset kernel module. This parameter may be used to enable usage of features on some displays such as ULMB (Ultra Low Motion Blur) which are incompatible with VRR. See the “Direct Rendering Manager Kernel Modesetting” (DRM KMS) chapter of the README for further information.
• Added support for NVIDIA Smooth Motion. See the “NVIDIA Smooth Motion” chapter in the README for details.
• Extended the nvidia-powerd daemon to also support Dynamic Boost while a notebook is running on battery power. See the “Dynamic Boost on Linux” chapter in the README for details.
• Updated the nvidia-modeset driver to trim trailing whitespace from the product name passed to the GPU’s audio device as part of the EDID-Like Data (ELD).
• Dropped support for NV_PLANE_BLEND_CTM, NV_PLANE_DEGAMMA_TF, NV_PLANE_DEGAMMA_LUT, NV_PLANE_DEGAMMA_LUT_SIZE, and NV_PLANE_DEGAMMA_MULTIPLIER DRM plane properties on Linux kernels earlier than 6.8 to avoid exceeding DRM_OBJECT_MAX_PROPERTY.
• Fixed an issue that could cause render-offloaded applications using KDE Frameworks 6 to crash.

Not in the changelog:

• Added support for the RTX 5060 Ti

Downloads: Driver Details | NVIDIA

If you have new problems, please make sure you’ve uploaded an nvidia bug report: run sudo nvidia-bug-report.sh in a terminal and upload the resulting file.

[agarplayerarlon]

I hope they are referring to my Minecraft bug report

Big Minecraft issues with latest stable driver 565.77 on Wayland

You’re brilliant. Thanks a lot. This helps me from those annoying crashes on exit. They persist even in newer library - Java VM: OpenJDK 64-Bit Server VM (Red_Hat-21.0.6.0.7-1) (21.0.6+7, mixed mode, sharing, tiered, compressed oops, compressed class ptrs, g1 gc, linux-amd64) Problematic frame: C [libnvidia-glcore.so.570.124.04+0xa5911c] I just add export __GL_THREADED_OPTIMIZATIONS=0 before starting TLauncher and voila ))) And there are no any other issues with MC on My Fedora 41 (KDE…

I’ve not tested it yet but it seems like a great update!

Also I think they are gonna release some minor driver versions like they did with the 570 driver to enable rtx 5060 support

[birdie]

Let’s be constructive:

• Install the driver
• Make sure the issue is reproducible
• If so, upload nvidia-bug-report and explain your issue or link to the existing post.

Thanks!

Being toxic literally helps no one. I’ve had issues with NVIDIA that took the company up to three (!) years to resolve. The company is now first and foremost about AI. Windows drivers are second, and even there the 572 branch was truly horrible. Linux drivers come dead last.

[BlueGoliath]

Linux trolls are out in full force.

No, the 575 driver does not fix any VRAM issues. 1.6GB at the desktop from a fresh boot and initial program open. Very nice.

[matt-schwartz]

I just had a pretty serious crash where I was dumped into TTY while testing gamescope on Plasma 6 Wayland. After repeatedly fullscreening then windowing gamescope -- vkcube ~10 times with Super + F, suddenly my PC locked up and I was forced to reboot with the power button.

Logs seem to show it was a full GPU reset:

Apr 16 11:10:15 blackwell kernel: NVRM: _issueRpcAndWait: rpcRecvPoll failed with status 0x00000062 for fn 76!
Apr 16 11:10:15 blackwell kernel: NVRM: nvCheckOkFailedNoLog: Check failed: Reset required [NV_ERR_RESET_REQUIRED] (0x00000062) returned from pRmApi->Control(pRmApi, pGpu->hInternalClient, pGpu->hInternalSubdevice, NV2080_CTRL_CMD_INTERNAL_LOG_OOB_XID, &params, sizeof(params)) @ gpu.c:6469
Apr 16 11:10:15 blackwell kernel: NVRM: Xid (PCI:0000:01:00): 154, GPU recovery action changed from 0x0 (None) to 0x1 (GPU Reset Required)

I’ve never seen this bug before trying 575.51.02. It also seems reproducible here.

nvidia-bug-report.log.gz (627.5 KB)

journal log from that boot:
journal-crash.log (2.1 MB)

[42n4]

Suspend is not working (only sometimes under special conditions):

RTX 5070 Ti with 570.124.04 won’t resume monitor from “suspend to RAM”

I installed a kernel 6.14.2 and drivers 575.51.02. Again I have succeded several times to wake up the monitor and system with e.g. systemctl suspend; nvidia-bug-report.sh in one line. I attach a nvidia bug report after such rare successful event. nvidia-bug-report.log.gz (886.0 KB)

[luisalvaradox]

What they are mentioning here about GCC version also impacts anyone with the latest distro, or even in my case Ubuntu 25. 04

NVIDIA 575.51.02 Linux Driver Beta Released With Smooth Motion Support -...

Basically was about to test and found no default easy way (without falling back to gcc 14) to install this driver to see if it solves all those performance issues on 570.

[adolfotregosa]

@birdie

Fixed a bug which prevented VRR from working when overriding an EDID through the /sys/kernel/debug/dri/*/edid_override interface.

Confirmed fixed and kernel module nvidia-modeset parameter, conceal_vrr_capsis now working.

[DanielCeregatti]

The VRAM exhaustion issue is still present. Proton games go to single digit FPS once VRAM reaches 8G in Mangohud. This can be mitigated a bit by closing other programs that are using VRAM, but eventually it reaches 8G again, and we’re back to single digit FPS.

Also, exiting a game will still some times crash the entire system when running under gamescope. I can alt-sysrq to reboot, but no other means of recovery works. This has been happening with all driver versions going back several months.

nvidia-bug-report.log.gz (1.6 MB)

[UrbenLegend]

575 beta is still showing weird issues with Ambient Shadows in The Last of Us Part 2. Completely disabling them under Shadow Quality works around the issue, but it is a shame we can’t see all the graphical goodness.

[matt-schwartz]

note - I ended up reinstalling and it stopped happening for a bit. then, after changing the KDE setting for Color accuracy from “Prefer Efficiency” to “Prefer color accuracy”, the crash started to occur again. it stops happening if I switch it back to “Prefer Efficiency” and then reboot.

[agarplayerarlon]

I have tested the new drivers and the Minecraft crashes are all fixed, than you Nvidia!

Big Minecraft issues with latest stable driver 565.77 on Wayland

So far I’ve found 4 issues facing Minecraft with the new 565.77 drivers (on Wayland, I haven’t tested on x11) 1 - When setting minecraft to fullscreen with the F11 key, sometimes it will fullscreen with a black bar below it [Screenshot From 2024-12-12 16-54-31] 2 - Minecraft versions below 1.13 don’t open anymore and crash instantly 3 - When closing Minecraft it doesen’t close, but crashes after around 4 seconds with this message: [Screenshot From 2024-12-12 16-52-01] 4 - If I press F11 …

[faz]

open kernel modules don’t seem to have released yet. Are we expecting them soon? Can’t really package the driver update on most distros without the open kernel modules too. Plus, don’t 50 series need the open modules?

[leigh123linux]

The open driver is included in the .run file, that’s what I package for rpmfusion nvidia packages.

$ sudo dmesg |grep Open
[    4.952344] NVRM: loading NVIDIA UNIX Open Kernel Module for x86_64  575.51.02  Release Build  (dvs-builder@U22-I3-G01-3-2)  Thu Apr 10 15:55:07 UTC 2025
[    5.181100] nvidia-modeset: Loading NVIDIA UNIX Open Kernel Mode Setting Driver for x86_64  575.51.02  Release Build  (dvs-builder@U22-I3-G01-3-2)  Thu Apr 10 15:40:53 UTC 2025

[ptr1337]

They can be found here:

Index of /XFree86/NVIDIA-kernel-module-source

Here I pushed them also to github, extracted from the above link:

GitHub - CachyOS/open-gpu-kernel-modules at 575

[taco-bell-5-layer-burrito]

Just mentioning: 570.133.07 and 575.51.02 aren’t on nvidia’s Debian 12 repo

EDIT: 570.133.20 is available on the Debian 12 repo

[Xpander]

So. reporting here also that the issue with 32bit applications showing wrong power usage. Was an issue with 570 drivers and carried over to 575, but now it just reports very small numbers in mW. while with 570 it reported huge numbers. 64bit doesn’t have that issue.

570 release feedback & discussion

With 575.51.02 drivers it reports 8mW with 32bit applications. 64bit unaffected [2025-04-16 22:15:23.005] [MANGOHUD] [debug] Raw power usage (mW): 8 [2025-04-16 22:15:23.031] [MANGOHUD] [debug] Raw power usage (mW): 8 [2025-04-16 22:15:23.058] [MANGOHUD] [debug] Raw power usage (mW): 8 [2025-04-16 22:15:23.085] [MANGOHUD] [debug] Raw power usage (mW): 8 [2025-04-16 22:15:23.111] [MANGOHUD] [debug] Raw power usage (mW): 8

[v19930312]

Installed open version of drivers from CachyOS archive for testing on a dual-GPU laptop (AMD Radeon 680M + NVIDIA GeForce RTX 4060 Laptop GPU). System completely locks up when it’s time to show the login screen and won’t let me do anything, including switching to another TTY. Somehow managed to get past the login screen one time, but it locked up about one second later anyways.

System journal ends with

kernel: BUG: unable to handle page fault for address: ffff995f4d680008
kernel: #PF: supervisor read access in kernel mode
kernel: #PF: error_code(0x0000) - not-present page