Werner

Use armbian-config and switch to dev (or beta, not sure) branch. This ships 4.19 kernel. Do you miss anything special within 4.14? This is a maintained long term kernel btw.

After you run compile.sh the first time a userpatches directory is created where you can place own patches. Until now the only kernel known to work is 4.18.0-rc7. I do not know if pure Debian packages with this specific kernel image still existing somewhere, but AFAIK are the latest precompiled images for OPi 1+ and Lite2 are still shipped with this kernel. So if you have one of these boards you may start over and freeze the kernel from being upgraded using armbian-config

Images [edit: for boards with Allwinner H6 SoC] with kernel versions above 4.18.0-rc7 are unbootable anyways AFAIK. But feel free to update your kernel within the OS like Igor said. Please let me know if it still boots.

I suggest to not bother too much with the onboard WiFi as it is not very good. A better and more reliable solution would be to get a supported USB WiFi dongle.

Disconnect power

Known issue. Go back to 4.18.0-rc7 and freeze kernel until solved or build your own kernel with all the stuff needed and disable on board Ethernet and use USB-Ethernet adapter. Interesting to know though that other boards are affected too.

The reason for this seems to be a regression or a missing commit from @Icenowy sources (https://github.com/Icenowy/linux/commits/h6-integrate-2-ugly) which has not made the jump into mainline (yet). Therefore you either have to use the 4.18.0-rc7 kernel image (which I run two OPi 1+ perfectly fine on right now) or temporarily disable dwmac like @martinayotte did as workaround to successfully boot up the board with a kernel built from newer sources (and use a USB-Ethernet adapter). For confirmation you can utilize a UART serial adapter to check the kernel boot for errors like "sun8i_dwmac_probe+0x164/0x518". As stated mainline support has still a long way to go to better support the H6 SoC. Do not just discard the board. Just put it on the shelf and it may run fine in a couple of month.

If an image was generated successfully there is usually no need for cleanup by users hand.

Fixed...maybe. Just give it a shot. @lomady

Did just a quick look at 4.19 but did not come very far... I leave it that way as well and wait for @Igor to push sunxi64 to 4.19. Thank you anyways.

@martinayotte Did you find a solution for the unbootable OPi 1+ yet?

Known issue. Reboot stops the board.

Fun fact: The kernel did not freeze. When you remove the sd card you still receive output from the console, also when you re-insert it.

This answers the question if OPi One Plus is affected only I guess

This basically changed the kernel source from Icenowy sources to mainline. For me this results in an unbootable image. No clue if this affects Lite2 too or OPi One Plus only.

The build script will automatically update and pull the changes from github before the menu comes up. Simply run ./compile.sh Edit: Nope....mainline kernel is still an issue U-Boot SPL 2018.09-rc1-armbian (Sep 28 2018 - 18:39:47 +0200) DRAM: 1024 MiB Trying to boot from MMC1 NOTICE: BL31: v1.5(debug):5069c1c NOTICE: BL31: Built : 14:47:15, Sep 9 2018 NOTICE: BL31: Detected Allwinner H6 SoC (1728) INFO: ARM GICv2 driver initialized NOTICE: PMIC: Probing AXP805 NOTICE: PMIC: AXP805 detected INFO: BL31: Platform setup done INFO: BL31: Initializing runtime services INFO: BL31: cortex_a53: CPU workaround for 855873 was applied INFO: BL31: Preparing for EL3 exit to normal world INFO: Entry point address = 0x4a000000 INFO: SPSR = 0x3c9 U-Boot 2018.09-rc1-armbian (Sep 28 2018 - 18:39:47 +0200) Allwinner Technology CPU: Allwinner H6 (SUN50I) Model: OrangePi One Plus DRAM: 1 GiB MMC: SUNXI SD/MMC: 0 Loading Environment from FAT... Unable to use mmc 0:1... Failed (-5) Using device 'sun50i_dw_hdmi', disp_uc_priv=0000000079f33320 sunxi_de3_init: device 'sunxi_de3' display won't probe (ret=-1) Using device 'sun50i_dw_hdmi', disp_uc_priv=0000000079f33320 sunxi_de3_init: device 'sunxi_de3' display won't probe (ret=-1) In: serial Out: serial Err: serial Net: No ethernet found. starting USB... No controllers found Hit any key to stop autoboot: 0 switch to partitions #0, OK mmc0 is current device Scanning mmc 0:1... Found U-Boot script /boot/boot.scr 3090 bytes read in 13 ms (231.4 KiB/s) ## Executing script at 4fc00000 U-boot loaded from SD Boot script loaded from mmc 116 bytes read in 10 ms (10.7 KiB/s) 16689 bytes read in 30 ms (543 KiB/s) ** File not found /boot/dtb/allwinner/overlay/sun50i-h6-fixup.scr ** 5026188 bytes read in 524 ms (9.1 MiB/s) 13844488 bytes read in 1412 ms (9.3 MiB/s) ## Loading init Ramdisk from Legacy Image at 4fe00000 ... Image Name: uInitrd Image Type: AArch64 Linux RAMDisk Image (gzip compressed) Data Size: 5026124 Bytes = 4.8 MiB Load Address: 00000000 Entry Point: 00000000 Verifying Checksum ... OK ## Flattened Device Tree blob at 4fa00000 Booting using the fdt blob at 0x4fa00000 Loading Ramdisk to 49b34000, end 49fff14c ... OK reserving fdt memory region: addr=4fa00000 size=6a000 Loading Device Tree to 0000000049ac7000, end 0000000049b33fff ... OK Setting up simplefb DE3 present but not probed Starting kernel ... [ 0.880733] Internal error: Oops: 96000004 [#1] SMP [ 0.885616] Modules linked in: [ 0.888674] CPU: 3 PID: 36 Comm: kworker/3:1 Not tainted 4.18.10-sunxi64 #1 [ 0.895628] Hardware name: Orange Pi One+ (DT) [ 0.900079] Workqueue: events deferred_probe_work_func [ 0.905215] pstate: 80000005 (Nzcv daif -PAN -UAO) [ 0.910005] pc : sun8i_dwmac_probe+0x164/0x518 [ 0.914446] lr : sun8i_dwmac_probe+0x20c/0x518 [ 0.918885] sp : ffff00000907bb30 [ 0.922196] x29: ffff00000907bb30 x28: 0000000000000000 [ 0.927506] x27: 0000000000000000 x26: ffff000008b7d790 [ 0.932817] x25: ffff800037fe4558 x24: ffff800036ac4010 [ 0.938127] x23: ffff800036ac4000 x22: ffff800034e5d018 [ 0.943437] x21: ffff8000352fbe98 x20: ffff000008bcf000 [ 0.948746] x19: ffff000008d08000 x18: 0000000000008000 [ 0.954056] x17: ffff0000089f8c98 x16: ffff0000089f8c98 [ 0.959366] x15: ffff800037fdf600 x14: ffff000008c0d000 [ 0.964675] x13: 00000000000000e0 x12: 0000000000000030 [ 0.969985] x11: 0101010101010101 x10: 7f7f7f7f7f7f7f7f [ 0.975295] x9 : 6061686d68656d68 x8 : ffff00000907b808 [ 0.980604] x7 : 0000000000000000 x6 : 0000000000000003 [ 0.985914] x5 : 0000000000000002 x4 : 0000000000000003 [ 0.991223] x3 : 0000000000000000 x2 : ffff00000907bba0 [ 0.996533] x1 : ffff800034e5d400 x0 : ffff800036ac4010 [ 1.001844] Process kworker/3:1 (pid: 36, stack limit = 0x(____ptrval____)) [ 1.008799] Call trace: [ 1.011247] sun8i_dwmac_probe+0x164/0x518 [ 1.015343] platform_drv_probe+0x50/0xa0 [ 1.019352] driver_probe_device+0x21c/0x2f8 [ 1.023621] __device_attach_driver+0x98/0xf0 [ 1.027979] bus_for_each_drv+0x64/0xc8 [ 1.031814] __device_attach+0xd8/0x130 [ 1.035648] device_initial_probe+0x10/0x18 [ 1.039829] bus_probe_device+0x90/0x98 [ 1.043664] deferred_probe_work_func+0x9c/0x140 [ 1.048283] process_one_work+0x1e8/0x338 [ 1.052292] worker_thread+0x240/0x478 [ 1.056041] kthread+0x128/0x130 [ 1.059270] ret_from_fork+0x10/0x18 [ 1.062847] Code: f94012a3 9101c3a2 aa1803e0 f9400463 (a9401464) [ 1.068937] ---[ end trace bd9dc8f0489fd612 ]--- [ 1.121726] mmc0: host does not support reading read-only switch, assuming write-enable [ 1.133153] mmc0: new high speed SDHC card at address 0001 [ 1.139547] mmcblk0: mmc0:0001 00000 14.6 GiB [ 1.146240] mmcblk0: p1

Will test this against the OPi One Plus. From earlier nightlies we know that 4.18.x has some issues. 4.18.0-rc7 from Icenowy works though. Lets see what happens. As far for now build did not fail at ATF. That's a good sign.

https://github.com/armbian/build/issues/1104 Reverting the change from this commit fixed this for me: https://github.com/armbian/build/commit/cbcdf570f38498e567f84b1abe71edcb4bc3458f

The kernel is the same, just the OS on top is different.

swappiness=0 http://ix.io/1nDr swappiness=1 http://ix.io/1nDw swappiness=60 http://ix.io/1nDC Reboot between test 2 and 3 to clear memory and skip cooldown for the loadavg

I am ready to give it a shot whenever you are ready

Sure thing. I'll catch up later.

I did the benchmarking two more times with some non-essential services disabled to minimize impact, but no better results. No idea what the io wait is causing. No cooling this time. http://ix.io/1nwh The 2nd one with swap disabled as I noticed swap was used after the first run, so I disabled it: http://ix.io/1nwo ... not a good idea. I give up for now, last but not least because with the PineH64's results we know the results for the OPi One Plus as well...if it would rund flawless ¯\_(ツ)_/¯

Unbootable kernel with OPi One Plus confirmed, runs into panic, reading from UART. There is something wrong with the later 4.18.x kernel. 4.18.0-rc7 from @Icenowy sources working well. But this kind a runs off topic I guess

Should not have lots of background activity at all. The image I used was fresh without any packages added. I could not find activity of unattended-upgrades. I may or may not redo the benchmark though with some other processes like networkmanager, rsyslog or cron disabled.