Mike C

tonight as I was about to head to bed, I noticed low buzz coming from my Tinker board as if my fan was running. Sure enough it was running, I checked top and a couple processes were trying to read from the cifs share and had the cores pinned using 100%... I did some testing, I could read files from the cifs share but whatever I used would hang at the end of the file, cat, cp, etc... I also tried almost everything, nothing fixed the issue until I rolled back my kernel from Dev (4.14.0-rc1) to Next (4.13.2) then all the problems were gone. I also noticed on the 4.14.0-rc1 kernel booting and shutdown times were much longer... at least double of whats normal, this was true with or without any cifs shares mounted. digging threw logs hasn't amounted to anything... Anyone else running 4.14.0-rc1 having these or other issues out of the blue?

FYI: The Asus guys changed this in TinkerOS 2.0.1 Beta, I'm willing to bet is was pretty hackish lol. Their git repository has the change to zone0, as I compiled from their sources and it was moved to Zone0 in them as well. https://github.com/TinkerBoard/debian_kernel

@Tido, sounds like something I might be interested. I've had a bit of a rough night so I will look at it a bit more in the morning. Cheers

Cool thank you for the update mate. One thing that concerns me is the cpu throttling likely is not happening since I believe its kernel based. all thou the kernel seams to throttle at kind of low temps, anything over 70c, .Datasheet on the rk3288 states a max of 85c for operating temp, and absolute max of 125c. To start throttling 70c seams to be bit overly conservative. I think I saw a patch from the rockchip guys that lowered all throttling levels 5 degrees C, unsure why, might be a very good reason for it.

Just started using the thinker board, and got fed up with the OEM builds which brought me here. So I would like to thank everyone for all the hard work, Im sure it can't be easy fixing things blind! Cheers mates. As the Tinkerboard runs hot under load, i run a program that adjusts pwm to adjust fan speed. so I noticed right away when upgrading to 4.13.0 that both soc and gpu thermal zones had broken. I've tried 4.13.1 and 4.13.2 as well they are broken as well. 4.13.x all kick error messages into dmesg as well. root@Asus:/sys/class/thermal# uname -r 4.12.0-rockchip root@Asus:/sys/class/thermal# cat /sys/class/thermal/thermal_zone[1-2]/temp 32272 32727 root@Asus:/sys/class/thermal# uname -r 4.13.0-rockchip root@Asus:/sys/class/thermal# cat /sys/class/thermal/thermal_zone[0-1]/temp cat: /sys/class/thermal/thermal_zone0/temp: Invalid argument cat: /sys/class/thermal/thermal_zone1/temp: Invalid argument nothermal.log