and please fix! On DE GNOME and Kde frametime curve.
disabling GSP does not solve the problem.
on AMD cards smooth and smooth.
6 Likes
Installed the 560 drivers and after rebooting it seems they werenât loaded. This is because 560 switched to the open kernel modules, which do not support my GTX 1060. Shouldnât there be a flag when installing 560 to use the propietary modules for older GPUs? Seems like a weird oversight.
hi I want to report issue I have with drivers :
1- canât launch EA app after install it with bottles/lutris on 550.107.02/560.31.02
however on 560.28.03 and other older drivers wine will detect my both GPU/iGPU and launch EA app successfully
2- VRAM have high usage issue on xwayland/wayland apps and will lead to system freeze/lockup and hard reboot is a must
3- KDE Plasma on xorg/x11 240hz feel like 60hz tried both arch/fedora same issue very laggy
Im on arch, KDE plasma, up to date, RTX 2070 SUPER / iGPU 630 UHD, dual monitor, with this settings:
options nvidia_drm modeset=1
options nvidia_drm fbdev=1
options nvidia NVreg_EnableGpuFirmware=0
options nvidia NVreg_PreserveVideoMemoryAllocations=1
1 Like
Unfortunately the 560.31.02 fix for flip event timeout doesnât seem to help at all on serious sam modding tools, however I was able to generate a bug report hopefully this will bring insights on what may be causing the issue.
nvidia-bug-report.log.gz (1.2 MB)
It could be a memory leak but Iâm not sure, Xwayland stops allocating memory around 1.3GB thoâ and seems to never go above that limit.
I filed a bug report for Xwayland but the conclusion there was that it might just be how Nvidia currently handles vram management, it needs rework by Nvidia if thatâs the case,
I also tried different Nvidia kernel settings on my end but nothing changed the behavior.
1 Like
lenovo legion 7i intel igpu, nvidia 4080 dgpu. kernel 6.10.3 , nvidia driver 560.31.02
having an external monitor connected over usb-c and calling reboot causes kernel NULL pointer dereference.
aug 08 13:51:02 tom-legion /opt/bin/nvidia-powerd[1026]: Failed to initialize Dynamic Boost
aug 08 13:51:02 tom-legion /opt/bin/nvidia-powerd[1026]: Failed to detach GPU id 256
aug 08 13:51:02 tom-legion nvidia-persistenced[1030]: Shutdown (1030)
aug 08 13:51:02 tom-legion kernel: BUG: kernel NULL pointer dereference, address: 0000000000000030
aug 08 13:51:02 tom-legion kernel: #PF: supervisor read access in kernel mode
aug 08 13:51:02 tom-legion kernel: #PF: error_code(0x0000) - not-present page
aug 08 13:51:02 tom-legion kernel: PGD 0 P4D 0
aug 08 13:51:02 tom-legion kernel: Oops: Oops: 0000 [#1] PREEMPT SMP NOPTI
aug 08 13:51:02 tom-legion kernel: CPU: 14 PID: 1259 Comm: Hyprland Tainted: P S U W O 6.10.3-gentoo-tom #1
aug 08 13:51:02 tom-legion kernel: Hardware name: LENOVO 82WQ/LNVNB161216, BIOS KWCN42WW 09/15/2023
aug 08 13:51:02 tom-legion kernel: RIP: 0010:nv_dma_release_sgt+0x2a/0x540 [nvidia]
aug 08 13:51:02 tom-legion kernel: Code: 0f 1f 44 00 00 48 85 f6 74 3d 48 85 ff 53 48 89 f3 74 48 b8 ff ff ff ff f0 0f c1 03 83 f8 01 74 1c 85 c0 7e 27 48 8b 43 08 5b <48> 8b 40 30 48 8b 80 c8 00 00 00 48 8b 38 e9 b3 33 90 f4 48 89 df
aug 08 13:51:02 tom-legion kernel: RSP: 0018:ffffa60e04f0b7a8 EFLAGS: 00010286
aug 08 13:51:02 tom-legion kernel: RAX: 0000000000000000 RBX: ffff925d95778008 RCX: 000000000020001e
aug 08 13:51:02 tom-legion kernel: RDX: 0000000000200001 RSI: ffff925d92b67c00 RDI: 0000000000000000
aug 08 13:51:02 tom-legion kernel: RBP: ffff925d9851ac70 R08: 0000000000000000 R09: ffffffffc08e0a15
aug 08 13:51:02 tom-legion kernel: R10: 0000000000000000 R11: 0000000000000000 R12: ffff925dbfa45408
aug 08 13:51:02 tom-legion kernel: R13: ffff925da7c70cc0 R14: 00000000caf00004 R15: 00000000fade0001
aug 08 13:51:02 tom-legion kernel: FS: 0000000000000000(0000) GS:ffff9274cd300000(0000) knlGS:0000000000000000
aug 08 13:51:02 tom-legion kernel: CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
aug 08 13:51:02 tom-legion kernel: CR2: 0000000000000030 CR3: 0000001165020000 CR4: 0000000000750ef0
aug 08 13:51:02 tom-legion kernel: PKRU: 55555554
aug 08 13:51:02 tom-legion kernel: Call Trace:
aug 08 13:51:02 tom-legion kernel: <TASK>
aug 08 13:51:02 tom-legion kernel: ? __die+0x52/0x93
aug 08 13:51:02 tom-legion kernel: ? page_fault_oops+0x130/0x2b0
aug 08 13:51:02 tom-legion kernel: ? exc_page_fault+0x5dd/0x9d0
aug 08 13:51:02 tom-legion kernel: ? asm_exc_page_fault+0x22/0x30
aug 08 13:51:02 tom-legion kernel: ? nv_dma_release_sgt+0x45/0x540 [nvidia]
aug 08 13:51:02 tom-legion kernel: ? nv_dma_release_sgt+0x2a/0x540 [nvidia]
aug 08 13:51:02 tom-legion kernel: _nv000703rm+0x627/0x6a0 [nvidia]
aug 08 13:51:02 tom-legion kernel: ? _nv038494rm+0x72a/0xa90 [nvidia]
aug 08 13:51:02 tom-legion kernel: ? _nv038455rm+0x6b/0x130 [nvidia]
aug 08 13:51:02 tom-legion kernel: ? _nv004292rm+0xd/0x20 [nvidia]
aug 08 13:51:02 tom-legion kernel: ? _nv006122rm+0x1e/0xb0 [nvidia]
aug 08 13:51:02 tom-legion kernel: ? _nv018077rm+0x59c/0x680 [nvidia]
aug 08 13:51:02 tom-legion kernel: ? _nv048178rm+0xb3/0xe0 [nvidia]
aug 08 13:51:02 tom-legion kernel: ? _nv049953rm+0xb3/0x180 [nvidia]
aug 08 13:51:02 tom-legion kernel: ? _nv049952rm+0x4ae/0x660 [nvidia]
aug 08 13:51:02 tom-legion kernel: ? _nv049948rm+0xa7/0x1a0 [nvidia]
aug 08 13:51:02 tom-legion kernel: ? rm_cleanup_file_private+0x1db/0x200 [nvidia]
aug 08 13:51:02 tom-legion kernel: ? nv_free_pages+0x229/0x470 [nvidia]
aug 08 13:51:02 tom-legion kernel: ? __fput+0xe5/0x2c0
aug 08 13:51:02 tom-legion kernel: ? task_work_run+0x55/0x90
aug 08 13:51:02 tom-legion kernel: ? do_exit+0x2f2/0xaf0
aug 08 13:51:02 tom-legion kernel: ? do_group_exit+0x2c/0x80
aug 08 13:51:02 tom-legion kernel: ? get_signal+0x875/0x8f0
aug 08 13:51:02 tom-legion kernel: ? futex_wait+0xab/0x140
aug 08 13:51:02 tom-legion kernel: ? arch_do_signal_or_restart+0x7e/0x290
aug 08 13:51:02 tom-legion kernel: ? syscall_exit_to_user_mode+0xc4/0x170
aug 08 13:51:02 tom-legion kernel: ? do_syscall_64+0x8a/0x190
aug 08 13:51:02 tom-legion kernel: ? __alloc_pages_noprof+0x15e/0x1070
aug 08 13:51:02 tom-legion kernel: ? try_charge_memcg+0x7f/0xaa0
aug 08 13:51:02 tom-legion kernel: ? __anon_vma_prepare+0xf4/0x190
aug 08 13:51:02 tom-legion kernel: ? __mod_memcg_lruvec_state+0xa6/0x150
aug 08 13:51:02 tom-legion kernel: ? do_anonymous_page+0x185/0x7f0
aug 08 13:51:02 tom-legion kernel: ? __count_memcg_events+0x53/0xf0
aug 08 13:51:02 tom-legion kernel: ? handle_mm_fault+0x77a/0xf30
aug 08 13:51:02 tom-legion kernel: ? flush_tlb_func+0x1af/0x1f0
aug 08 13:51:02 tom-legion kernel: ? __pfx_flush_tlb_func+0x10/0x10
aug 08 13:51:02 tom-legion kernel: ? sched_clock+0xc/0x30
aug 08 13:51:02 tom-legion kernel: ? sched_clock_cpu+0x7/0x20
aug 08 13:51:02 tom-legion kernel: ? irqtime_account_irq+0x3a/0xc0
aug 08 13:51:02 tom-legion kernel: ? irq_exit_rcu+0x4c/0x90
nvidia-bug-report.log.gz (2.0 MB)
Doubt it. Itâs existed since 555 and Nvidia wonât even acknowledge that itâs an issue.
560 (with Gnome + Wayland) has broken screen streaming / recording for me in OBS. The âScreen Capture (PipeWire)â source shows up 100% black, except for the mouse cursor, after giving it permission (via the Gnome modal window that pops up) to share my entire desktop - and alternately, with specific windows marked ok to share. At no point do any pixels from my desktop appear in OBS.
This definitely worked correctly with 550. I donât think I tried it with 555, so I canât say how recently this regression began.
Seems like there is still GSP related frame stuttering and slowdowns on the 560.31.02 driver in Gnome 46 when navigating around the desktop.
1 Like
So far this driver seems stable, though Iâve noticed that Far Cry 5 is stuck on a black screen when launching and killing the process in this state makes the system freeze for a few seconds. This has already been an issue with the 555 driver. According to ProtonDB downgrading to 550 works around this issue, which I can confirm after testing it myself, so I think this is a driver issue after all.
560.31.02 fixed the problem I was having running games in KDE.
This is annoying as hell! I have the same problem here, whatever DE I use.
2 Likes
Hi all,
Multi-monitor VRR support on Wayland with Explicit sync is being tracked internally at NVBug 4545537. We can reproduce the issue on our systems. The issue is under investigation.
Thanks!
11 Likes
I experience 2 issues related to VRR. With the new 560 beta driver I experience lag spikes in dxvk games like Rocket League and GTA V when VRR disabled.
With VRR enabled, when I turn the display off and then back on switching to fullscren causes the screen to turn off for a few seconds. It also turns off randomly while in fullscreen. This issue occurs with the 560 beta and 555 drivers. I am not sure about older versions.
GPU: RTX 3060
Display: ASUS VP28U 4k 60hz DisplayPort 1.2
Memory: 16 GiB
CPU: i5-12400F
OS: Arch Linux
Session: Wayland
DE: KDE Plasma
560 is no longer able to work with both my video cards:
Aug 09 11:48:45 test kernel: NVRM: The NVIDIA GPU 0000:03:00.0 (PCI ID: 10de:1b81)
NVRM: installed in this system is not supported by open
NVRM: nvidia.ko because it does not include the required GPU
NVRM: System Processor (GSP).
NVRM: Please see the âOpen Linux Kernel Modulesâ and âGSP
NVRM: Firmwareâ sections in the driver README, available on
NVRM: the Linux graphics driver download page at
NVRM: www.nvidia.com.
Aug 09 11:48:45 test kernel: nvidia 0000:03:00.0: probe with driver nvidia failed with error -1
Aug 09 11:48:45 test kernel: nvidia 0000:04:00.0: vgaarb: VGA decodes changed: olddecodes=io+mem,decodes=none:owns=io+mem
Aug 09 11:48:45 test kernel: NVRM: The NVIDIA probe routine failed for 1 device(s).
Aug 09 11:48:45 test kernel: NVRM: loading NVIDIA UNIX Open Kernel Module for x86_64 560.28.03 Release Build (dvs-builder@U16-A24-27-4)
Everything was working fine before this last update. I am using Fedora 39 with the latest driver from the nvidia repo.
03:00.0 VGA compatible controller: NVIDIA Corporation GP104 [GeForce GTX 1070] (rev a1)
04:00.0 VGA compatible controller: NVIDIA Corporation AD106 [GeForce RTX 4060 Ti 16GB] (rev a1)
Do you have anything running, which calls nvidia-smi?
Some example applications, which are using that:
- coolercontrol (fixed in latest version and should use now nvml)
- Plasma System Monitor
There could be might even more applications.
That is because, you have a 1070 and 4060 Ti and using the open source modules.
You should either remove the 1070 or use the closed source modules.
Yep, apparently the update didnât include any warning, notice that it will render some video cards unusable. I switched to legacy branch and both cards are working again.
Please fix, 555-560 drivers problematic.
let the developers know
No nvidia-smi proccess spawns here. I am still getting the same GSP related issues