VlaoMao Posted March 8, 2018 Posted March 8, 2018 Hello! I followed build instruction from github, builded full OS image to flash to microSD and get kernel oops on boot (selected orangepiwin->dev(4.14.x)->xenial->server) log: https://pastebin.com/G4UtkGPR Possible to fix?
Igor Posted March 8, 2018 Posted March 8, 2018 On 3/8/2018 at 11:58 AM, VlaoMao said: Possible to fix? Expand By who? This year? Today? What for? Have you noticed this?
VlaoMao Posted March 8, 2018 Author Posted March 8, 2018 On 3/8/2018 at 12:37 PM, Igor said: Have you noticed this? Expand Yes, I can read. can there be conjectures or thoughts that caused this?
Xalius Posted March 8, 2018 Posted March 8, 2018 Can you post a boot log with earlyprintk enabled? There are about 7 seconds missing from it....
VlaoMao Posted March 8, 2018 Author Posted March 8, 2018 On 3/8/2018 at 12:52 PM, Xalius said: Can you post a boot log with earlyprintk enabled? There are about 7 seconds missing from it.... Expand yes, i will try rebuild and post output here
VlaoMao Posted March 8, 2018 Author Posted March 8, 2018 On 3/8/2018 at 12:52 PM, Xalius said: Can you post a boot log with earlyprintk enabled? There are about 7 seconds missing from it.... Expand I rebuilded with debian, but log is still the same https://pastebin.com/rWLnrT28
zador.blood.stained Posted March 8, 2018 Posted March 8, 2018 [ 1.583064] WARNING: CPU: 0 PID: 29 at drivers/clocksource/arm_arch_timer.c:222 fsl_a008585_read_cntvct_el0+0x28/0x30 ... [ 1.595818] PC is at fsl_a008585_read_cntvct_el0+0x28/0x30 This is the "famous" A64 timer errata/bug: https://lists.denx.de/pipermail/u-boot/2016-November/271836.html Currently there are no 100% reliable ways to fix this, and it's not up to us to fix. [ 7.588166] Internal error: undefined instruction: 0 [#1] SMP Possibly DRAM instability or CPU underpowering. So DVFS tables and DRAM frequency need to be checked first by anyone who wants to investigate and fix this, assuming there are no HW issues.
Recommended Posts