Tested this in virtualbox, Gnome using Wayland



The following 3 packages are going to be installed:

- cpupower-5.12.6-1.mga8.x86_64
- kernel-desktop-5.12.6-1.mga8-1-1.mga8.x86_64
- kernel-desktop-latest-5.12.6-1.mga8.x86_64

------------

$ uname -a
Linux localhost 5.12.6-desktop-1.mga8 #1 SMP Sat May 22 10:59:56 UTC 2021 x86_64 x86_64 x86_64 GNU/Linux

------------

I don't have the hardware to really validate the new equipment functionality, but from a base system perspective, I'm not having any issues.  Sound and tools work.

CC: (none) => brtians1

Kernel: 5.12.6-desktop-1.mga8 x86_64
10-Core Intel Core i9-7900X
NVIDIA GP102 [GeForce GTX 1080 Ti]
No wifi.

Reboot without issues.  Mate desktop functioning normally.
Stress tests OK.  NFS shares mounted.
Virtualbox launches clients - NFS share mounted.

CC: (none) => tarazed25

Aorus X5 laptop
Kernel: 5.12.6-desktop-1.mga8 x86_64
Quad Core Intel Core i7-5700HQ
NVIDIA GM204M [GeForce GTX 965M] driver: nvidia v: 460.80
Intel Wireless 7265 driver: iwlwifi

Backported kernel installed, `drakboot v--boot` and rebooted OK.

However, although everything worked including recovery after shutting the lid, the whole system was subject to continual freezes which appeared to have something to do with temperature.  The fans kicked in again and again and  eveything locked for sometimes up to 20 seconds.  Looks like some kind of governor operating when temperatures hit 76°C.  No rogue processes that can be identified in htop or ksysguard.  Tried Mate and Plasma.

dmesg reports "martian source" 224.0.0.1 from 0.0.0.0 after messages about
"NVRM: Going over RM unhandled interrupt threshold for irq 43"

There is a clue there.  Need to investigate.  Reporting back later.

Installed desktop and server versions.
Kernel: 5.12.6-desktop-1.mga8 x86_64
Quad Core Intel Core i7-4790
NVIDIA GM204 [GeForce GTX 970] driver: nvidia v: 460.80
Realtek RTL8111/8168/8411 PCI Express Gigabit Ethernet - driver: r8169 

Mate running fine.  NFS shares mounted from fstab.
stress and graphics tests work without problems.

Before I test this on my main system;
Is there some problem to be expected about swithing back and forth between 5.12 and 5.10, i.e OK to have the same cpupower-5.12.6-1 ?

CC: (none) => fri

(In reply to Len Lawrence from comment #3)
> Aorus X5 laptop
> "NVRM: Going over RM unhandled interrupt threshold for irq 43"
> There is a clue there.  Need to investigate.  Reporting back later.

Check any batteries such as in a wireless mouse as one with a barely
adequate charge can cause too many intterupts (not cpu usage) for the
system to handle. Same with all hardware connections. Poor contact on
an ethernet or usb connection can do the same.

Excessive interrupts will cause poor response and increase the heat,
but not show in tools like top.

CC: (none) => davidwhodgins

Should have added, you can check the counts of hardware interrupts (irq) in
/proc/interrupts
If any of those counters are going up by more than a 1000 in a second or
less, then it's likely an indication of a problem.

Continuing from comment 3.  The "clue" led nowhere.  Other systems on the same laptop ran better, far less fan activity.  Thanks for the suggestions David.  Checked the mouse battery - still well in the green. 
Just logged in to the test system and now it runs normally; videos play smoothly, no fan noise, all is well.  ??

So, no problems with this kernel.

Continuing from comment 8.
However, the battery drains very rapidly so it is necessary to run off mains power.  After the power lead was plugged in the fans started to run harder and dmesg reported repeatedly:
"Going over RM unhandled interrupt threshold for irq 43"
and issued the "martian source from ........ <wifi device>"

One of the address elements mentioned is the final component of the NAS drive LAN address and since the mains power circuit is used for ethernet transmissions I am wondering if the modulation is some how getting filtered by the wifi device to cause the spurious interrupts.  I am not inclined to shut down my network just to test this possibility just now but might have a look at it in the future.  Anyway it looks like a local problem - noise as far as QA is concerned.

Mga8-64 OK here

$ uname -a
Linux svarten.tribun 5.12.6-desktop-1.mga8 #1 SMP Sat May 22 10:59:56 UTC 2021 x86_64 x86_64 x86_64 GNU/Linux

Everything updated to testing

Hardware: My workstation "svarten": Mainboard: Sabertooth P67, CPU: i7-3770, RAM 16G, GM107 [GeForce GTX 750] using nvidia-current; GeForce 635 series and later, 4k display.  Disk&Filesystem: SSD with /boot/EFI and ext4 /boot, LUKS{LVM {swap, ext4 /home & / } and a spinner at /mnt/spinner

dkms status tells me VirtualBox and nvidia-current are OK.
BOINC detects CUDA and OpenCL, and use Nvidia GPU successfully

Plasma desktop, using Thunderbird, LibreOffice, Ktorrent, Nextcloud client, flatpak Firefox...
Video with sound in Firefox
Stress test: While working with other things BOINC use all cores to 100%, videos do not stutter.

VirtualBox running MSW7 64 bit OK: graphics, window resize, bidirectional clipboard, drag file from Dolphin to Explorer, folder sharing write protected and not, folder sharing, USB2 with plugin from upstream, internet video playing in Firefox at high other CPU load - smoother (i.e perfect) than with 5.10 kernel. (but that may have been network issues before)

Good enough tests...

Keywords: (none) => validated_backport
Whiteboard: (none) => MGA8-64-OK

moved

Resolution: (none) => FIXED
Status: NEW => RESOLVED