Perier Posted December 3, 2018 Posted December 3, 2018 Hello Community, Now I encountered the following problem when testing my "orange pi zero plus": I tested the processor load (stress-ng) at a working frequency of 624mhz for 25 minutes. Settings: mainline sunxi64-next (4.14.y) Spoiler armbianmonitor -m 10:53:44: 624MHz 4.50 100% 0% 99% 0% 0% 0% 100.1°C 1/1 [ 3314.299373] thermal thermal_zone0: critical temperature reached (105 n Message from syslogd@localhost at Dec 3 10:53:51 ... kernel: [ 3314.299373] thermal thermal_zone0: critical temperature reached (105 C), shutting down Stopping Session 1 of user root.. [ OK ] Stopped Session 1 of user root. [ OK ] Removed slice User Slice of root. Stopping Login Service... Stopping Permit User Sessions... [ OK ] Stopped Login Service. [ OK ] Stopped Permit User Sessions. [ OK ] Stopped target Remote File Systems. [ OK ] Stopped Unattended Upgrades Shutdown. [ OK ] Stopped target Network. Stopping Network Manager... Stopping Raise network interfaces... [ OK ] Stopped Network Manager. Stopping D-Bus System Message Bus... [ OK ] Stopped D-Bus System Message Bus. [ OK ] Stopped target Basic System. [ OK ] Stopped target Slices. [ OK ] Removed slice User and Session Slice. [ OK ] Stopped target Paths. [ OK ] Stopped target Sockets. [ OK ] Closed Syslog Socket. [ OK ] Closed D-Bus System Message Bus Socket. [ OK ] Stopped target System Initialization. Stopping Armbian memory supported logging... Stopping Restore / save the current clock... Stopping Update UTMP about System Boot/Shutdown... [ OK ] Stopped target Swap. Deactivating swap /dev/zram4... Deactivating swap /dev/zram1... Deactivating swap /dev/zram2... Deactivating swap /dev/zram3... Stopping Entropy daemon using the HAVEGE algorithm... [ OK ] Stopped target Encrypted Volumes. [ OK ] Stopped Forward Password Requests to Wall Directory Watch. [ OK ] Stopped Dispatch Password Requests to Console Directory Watch. [ OK ] Stopped Entropy daemon using the HAVEGE algorithm. [ OK ] Stopped Raise network interfaces. [ OK ] Stopped Restore / save the current clock. [ OK ] Stopped Apply Kernel Variables. Stopping Load/Save Random Seed... [ OK ] Stopped Update UTMP about System Boot/Shutdown. [ OK ] Stopped Load/Save Random Seed. [ OK ] Stopped Create Volatile Files and Directories. [ OK ] Deactivated swap /dev/zram1. [ OK ] Deactivated swap /dev/zram4. [ OK ] Deactivated swap /dev/zram2. [ OK ] Deactivated swap /dev/zram3. [ OK ] Unmounted /var/log. [ OK ] Unmounted /var/log.hdd. [ OK ] Stopped Armbian memory supported logging. Stopping Armbian ZRAM config... [ OK ] Stopped Armbian ZRAM config. [ OK ] Stopped target Local File Systems. Unmounting /run/user/0... Unmounting /tmp... [ OK ] Unmounted /run/user/0. [ OK ] Unmounted /tmp. [ OK ] Reached target Unmount All Filesystems. [ OK ] Stopped target Local File Systems (Pre). [ OK ] Stopped Remount Root and Kernel File Systems. [ OK ] Stopped Create Static Device Nodes in /dev. [ OK ] Reached target Shutdown. [ 3325.807677] reboot: Power down INFO: PSCI Affinity Map: INFO: AffInst: Level 0, MPID 0x0, State ON INFO: AffInst: Level 0, MPID 0x1, State ON INFO: AffInst: Level 0, MPID 0x2, State ON INFO: AffInst: Level 0, MPID 0x3, State ON ERROR: PSCI system shutdown: still alive ... "ERROR: PSCI system shutdown: still alive ..." The development scenario is predictable, but an error at the end prevents the device from rebooting. How to fix it? Update.1: Or should a reboot not occur, but only a shutdown? Update.2: Still, the device did not turn off and did not restart (just hung), since the processor temperature is still high and the Ethernet LED flashes.
Tido Posted December 3, 2018 Posted December 3, 2018 shutdown'ed H3 without proper voltageregulator, heat up to 60°C when off search the Forum for: Overheating
Igor Posted December 3, 2018 Posted December 3, 2018 1 hour ago, Evgeny said: How to fix it? By studying: http://linux-sunxi.org/AR100 https://forum.armbian.com/search/?q=AR100 and implementing.
Perier Posted December 3, 2018 Author Posted December 3, 2018 1 hour ago, Igor said: By studying: http://linux-sunxi.org/AR100 https://forum.armbian.com/search/?q=AR100 and implementing. As I understand it, this is my case - https://github.com/crust-firmware/crust/tree/master Did I try to figure it out myself, maybe I used this repository? P.S. Thanks for the tip.
Igor Posted December 3, 2018 Posted December 3, 2018 40 minutes ago, Evgeny said: P.S. Thanks for the tip. Community will be helpful for the implementation ... if you managed to get it working as expected. We are slowly moving to 4.19.y kernel so it has to work there as well. (Its expected)
Perier Posted December 5, 2018 Author Posted December 5, 2018 On 12/3/2018 at 5:42 PM, Igor said: Community will be helpful for the implementation ... if you managed to get it working as expected. We are slowly moving to 4.19.y kernel so it has to work there as well. (Its expected) https://github.com/armbian/build/pull/1176 Glad to help people who have encountered this problem. P.S. For the tip, thanks to Andre Przywara. 1
Igor Posted December 5, 2018 Posted December 5, 2018 4 hours ago, Evgeny said: Glad to help people who have encountered this problem. Great, it works. I still have 230mA draft in "powered down" state but I guess we solve the overheating problem.
Tido Posted December 5, 2018 Posted December 5, 2018 P = U x I 5V x 0,23A = 1,15 Watt for the whole board I guess ? (sounds better, did you do the 'hot finger' test? ) Does this also affect H3 or only H5 ?
Perier Posted December 5, 2018 Author Posted December 5, 2018 1 hour ago, Igor said: Great, it works. I still have 230mA draft in "powered down" state but I guess we solve the overheating problem. Thank you. Yes, overheating and error messages eliminated. 17 minutes ago, Tido said: P = U x I 5V x 0,23A = 1,15 Watt for the whole board I guess ? (sounds better, did you do the 'hot finger' test? ) Does this also affect H3 or only H5 ? Not tested on H3 (sunxi), since this fix covers mainly A64, H5, H6 (sunxi64). "Hot finger" test was performed on sun50i-h5-orangepi-zero-plus.
Igor Posted December 5, 2018 Posted December 5, 2018 28 minutes ago, Tido said: P = U x I 5V x 0,23A = 1,15 Watt for the whole board I guess ? (sounds better, did you do the 'hot finger' test? ) Does this also affect H3 or only H5 ? I did a quick hot finger test also on Orangepi Zero+ and measurement. The consumption was before more than twice higher.
Recommended Posts