Screen tearing returned after latest updates

DNI_R

Today i installed latest batch of updates and after restart screen tearing returned.
Method that was working before, doesn't work anymore. Repeating commands below doesn't help:
echo "nvidia-drm.modeset=1" | sudo tee /etc/kernel/cmdline.d/50-nvidia-drm.conf sudo clr-boot-manager update
I used nvidia developers driver before (because of vulcan support for games) without any problems, but maybe my card became unsupported because the laptop is quite old (GT 740M, optimus laptop), if so, how do i reinstall to standard drivers safely (without running into black screens and so on)? Or is there a different (new?) method to deal with screen tearing efficiently?

Solarmass

Yeah, I have noticed a tearing in Firefox, so I enabled Force Full Composition Pipline in the NVIDIA settings today

DNI_R

Since it's optimus laptop, i don't have such option (Force Full Composition Pipeline).

I guess i should return to 390 drivers just to be safe and then see what's the situation.
Would running:
sudo eopkg install nvidia-390-glx-driver-common nvidia-390-glx-driver-32bit nvidia-390-glx-driver-current reinstall the driver safely removing unneeded developers driver components?
Or should i just use doflicky, will it remove developers driver when installing the 390 one?

P.S. I use current kernel.
P.P.S. Though checking in https://developer.nvidia.com/vulkan-driver my graphics card is still there, then maybe i can still use the driver, but how can i resolve screen tearing then?
Looking at changelog in solus for the driver, it wasn't updated this week, so updates in kernel caused this?

SOLUSfiddler

DNI_R
Just use "hardware drivers" (= DoFlicky) to find the best driver for your setup.
Install it, reboot.

Scotty-Trees

It's rather late so I will make a more detailed follow up post with screenshots and system info to this sometime tomorrow on Sunday, but under kernel 5.4.1 I too have screen tearing issues dragging windows, watching videos, and when scrolling using any type of web browser, it's a system wide issue. I do not have this issue on kernel 5.3 so for myself for the time being I will stick with that kernel since it works. Some users on the Arch side of things are reporting memory leaks and artifact issues with kernel 5.4.1 but some say er suggest rather that 5.4.2 "may" fix these issues. Until more is discovered, I will use the 5.3 kernel.

In the meantime hope everyone is having a lovely weekend!

DNI_R

I returned to 390 drivers just in case with doflicky, thankfully all went well with installation, but the problem with tearing still remains.
So it's probably exactly what Scotty-Trees wrote - issues with newest kernel.

jastisriradheshyam

I am on newest kernel 5.4, I am able to observe screen tearing when I play videos on Firefox.

Scotty-Trees

Just to add to my post from yesterday, the 5.4 kernel is currently causing screen tearing and artifact issues system wide. In the meantime, although not an ideal option, I'm going to remain on the 5.3 kernel which does not have these issues. Now normally I wouldn't advise doing this, I think it's always best to remain up to date, however issues happen, and hopefully sooner rather than later this issue can be fixed here within this lovely community or perhaps a fix may be around the corner for 5.4.2 who knows!

Others within the linux community are reporting similar issues here: https://www.reddit.com/r/archlinux/comments/e6z7lu/linux_541_kernel_causes_screen_artifacts_while/

The laptop I am using is an Acer Aspire E5 576g 5762 model, and it does have Intel graphics as well as a Nvidia Geforce MX150 graphics card as well.

Here's my hardware info for those curious:

scott@solus ~ $ inxi -Fxxxza --no-host
System:
  Kernel: 5.4.1-136.current x86_64 bits: 64 compiler: gcc v: 9.2.0 
  parameters: initrd=\EFI\com.solus-project\initrd-com.solus-project.current.5.4.1-136 
  root=PARTUUID=c846a705-b415-458a-bde0-ee639c61e82b quiet loglevel=3 splash 
  systemd.show_status=false rw radeon.si_support=0 radeon.cik_support=0 
  amdgpu.si_support=1 amdgpu.cik_support=1 
  resume=UUID=fba32bd8-3043-490c-b340-21abd1936dc6 nvidia-drm.modeset=1 
  Desktop: Gnome 3.34.1 wm: gnome-shell dm: GDM 3.34.1 Distro: Solus 4.0 
Machine:
  Type: Laptop System: Acer product: Aspire E5-576G v: V1.32 serial: <filter> 
  Mobo: KBL model: Ironman_SK v: V1.32 serial: <filter> UEFI: Insyde v: 1.32 
  date: 10/24/2017 
Battery:
  ID-1: BAT1 charge: 36.6 Wh condition: 36.6/62.2 Wh (59%) volts: 12.7/11.1 
  model: PANASONIC AS16B5J type: Li-ion serial: <filter> status: Full 
CPU:
  Topology: Quad Core model: Intel Core i5-8250U bits: 64 type: MT MCP 
  arch: Kaby Lake family: 6 model-id: 8E (142) stepping: A (10) microcode: CA 
  L2 cache: 6144 KiB 
  flags: avx avx2 lm nx pae sse sse2 sse3 sse4_1 sse4_2 ssse3 vmx 
  bogomips: 28800 
  Speed: 800 MHz min/max: 400/3400 MHz Core speeds (MHz): 1: 800 2: 801 3: 800 
  4: 800 5: 800 6: 800 7: 800 8: 801 
  Vulnerabilities: Type: itlb_multihit status: KVM: Split huge pages 
  Type: l1tf 
  mitigation: PTE Inversion; VMX: conditional cache flushes, SMT vulnerable 
  Type: mds mitigation: Clear CPU buffers; SMT vulnerable 
  Type: meltdown mitigation: PTI 
  Type: spec_store_bypass 
  mitigation: Speculative Store Bypass disabled via prctl and seccomp 
  Type: spectre_v1 
  mitigation: usercopy/swapgs barriers and __user pointer sanitization 
  Type: spectre_v2 mitigation: Full generic retpoline, IBPB: conditional, 
  IBRS_FW, STIBP: conditional, RSB filling 
  Type: tsx_async_abort status: Not affected 
Graphics:
  Device-1: Intel UHD Graphics 620 vendor: Acer Incorporated ALI driver: i915 
  v: kernel bus ID: 00:02.0 chip ID: 8086:5917 
  Device-2: NVIDIA GP108M [GeForce MX150] vendor: Acer Incorporated ALI 
  driver: nvidia v: 440.36 bus ID: 01:00.0 chip ID: 10de:1d10 
  Display: x11 server: X.Org 1.20.5 driver: modesetting,nvidia 
  compositor: gnome-shell resolution: 1920x1080~60Hz 
  OpenGL: renderer: GeForce MX150/PCIe/SSE2 v: 4.6.0 NVIDIA 440.36 
  direct render: Yes 
Audio:
  Device-1: Intel Sunrise Point-LP HD Audio vendor: Acer Incorporated ALI 
  driver: snd_hda_intel v: kernel bus ID: 00:1f.3 chip ID: 8086:9d71 
  Sound Server: ALSA v: k5.4.1-136.current 
Network:
  Device-1: Intel Dual Band Wireless-AC 3168NGW [Stone Peak] driver: iwlwifi 
  v: kernel port: 4000 bus ID: 03:00.0 chip ID: 8086:24fb 
  IF: wlp3s0 state: up mac: <filter> 
  Device-2: Realtek RTL8111/8168/8411 PCI Express Gigabit Ethernet 
  vendor: Acer Incorporated ALI driver: r8169 v: kernel port: 3000 
  bus ID: 04:00.1 chip ID: 10ec:8168 
  IF: enp4s0f1 state: down mac: <filter> 
Drives:
  Local Storage: total: 238.47 GiB used: 73.77 GiB (30.9%) 
  ID-1: /dev/sda vendor: SK Hynix model: HFS256G39TND-N210A size: 238.47 GiB 
  block size: physical: 4096 B logical: 512 B speed: 6.0 Gb/s serial: <filter> 
  rev: 1P10 scheme: GPT 
Partition:
  ID-1: / raw size: 234.27 GiB size: 229.59 GiB (98.00%) 
  used: 73.77 GiB (32.1%) fs: ext4 dev: /dev/sda3 
  ID-2: swap-1 size: 3.73 GiB used: 0 KiB (0.0%) fs: swap 
  swappiness: 10 (default 60) cache pressure: 100 (default) dev: /dev/sda2 
Sensors:
  System Temperatures: cpu: 54.0 C mobo: N/A gpu: nvidia temp: 48 C 
  Fan Speeds (RPM): N/A 
Info:
  Processes: 280 Uptime: N/A Memory: 15.58 GiB used: 1.02 GiB (6.6%) 
  Init: systemd v: 239 runlevel: 5 Compilers: gcc: N/A Shell: bash v: 5.0.11 
  running in: tilix inxi: 3.0.37

Here's just some additional info on graphics from the application CPU-X which shows system hardware in a GUI

Here are the nvidia settings I have. Sadly for display config. I don't have an option to switch between basic or advanced settings, so I am unable to change any sync settings or tweak anything really. It's always been this way, I assume that's just the nature of having a hybrid set up of Intel hd graphics and graphics card.

If anyone has any thoughts, comments, or helpful tips feel free to share them! And if you are also someone experiencing this issue as well, please let us know. Thank you for anyone that took the time to read through this I appreciate it!

Scotty-Trees

Scotty-Trees Piggy backing off my recent post, when I boot up in the 5.3 kernel (the one with no screen tearing), my nvidia settings are a bit different as shown below.
The option "Synchronization" is ON in 5.3, but mysteriously it is OFF in 5.4
So I guess my fix would be to re-enable synchronization to ON, but how can I do that if I don't have the option in the nvidia settings to do so? Any thoughts?

DNI_R

Scotty-Trees How do you boot into previous kernel? Is it through the boot menu before booting into Solus (hitting space bar if not enabled)? Sad that there is one option for me (Solus 4.0 Fortitude) when i get to boot menu, though i remember having multiple kernels available some time ago with previous install, wondering why not this time.

I also have Synchronization off in nvidia driver settings, no idea what that means though.

Scotty-Trees

DNI_R Well, I may be a bit wrong on this, but I believe Solus defaults to having two versions of the kernel on your machine. The most up to date kernel that you've updated to and the version previous to that one. So in my case I have 5.3 and 5.4 when I boot up. Also, if you have the LTS (long term support) kernel, then that would show up too.

To see what kernels you have when you boot up your machine use these commands one at a time (note this will always show when you boot since the value is set to 5 seconds):

sudo clr-boot-manager set-timeout 5
sudo clr-boot-manager update

What this means is you'll see what kernel(s) are on your machine for 5 seconds counting down before it will automatically pick the most recent kernel to boot to, you can just use your UP and DOWN arrow keys to select which one you want.

One way that is meant to prevent screen tearing is having Synchronization to ON, but for whatever reason the latest kernel and/or nvidia drivers have seemed to have borked that. Having it on is a good thing and the fact that it's off might just be a bug that upstream needs to fix. Solus has one amazing dev that works on the kernel updates and updating the kernel isn't a quick and easy task, so expect this may take some more time to fix. In the meantime we'll just have to be patient more than likely for the next Friday sync if a newer kernel gets pushed we'll have to test it out.

According to a post I found over at nvidia it "may" be their bug which they are currently tracking:
https://devtalk.nvidia.com/default/topic/1044496/linux/hangs-freezes-when-vulkan-v-sync-vk_present_mode_fifo_khr-is-enabled/