[ROM][Android9.0] Android P firmware released

You need install it first.
https://apkpure.com/amazon-prime-video/com.amazon.avod.thirdpartyclient

Hi,
Can I get a root? And how can I do this?

Radxa can built firmware with Magisk enabled.
Then you can patch your boot.img with MagiskManager(or a patched boot.img can be directly used for firmware image) to have root.

This is not working properly on my device. It is happening due to the router issue. I have visited Linksys Router Error 2118 page and get some useful ideas but I am looking for some proper ideas.

你好,请问编译的话如何编译?
这样编译出来的是TV版的
lunch rk3399_box-userdebug
make -j$(nproc)

如果我想编译成Tablet版的,是按下面这个方法么
lunch rk3399-userdebug
make -j$(nproc)
但是我编译之后,貌似启动不了,一直卡Android,是哪里做错了么

Unfortunately the 9.0 P image does not seems to support the RasPi Cam (the 7.1 image did!). Hope, this will be added soon?

Has anyone tried to use an USB (UVC) camera with the 9.0 image?


I will check.Thanks.


lunch rk3399-userdebug, 我已经找到问题了,等我更新吧.


现在没问题了,你更新试试

多谢多谢多谢多谢,问题已解决🙏

New version released.

Hello,

why the support for 2560x1440 resolution is disabled? I have it on first start on the “erase” screen but it switch back to 1920x1080 after reboot. The May image was fine.

Regards.

Hello,

another question. I must say the cpu runs very hot (with big heatsink) on android 9 compared to the linaro image. Is there is no power saving or something an the android 9 image?

Regards.

Android 9 RK filters 2560x1440 by default.
Reference: /system/usr/share/resolution_white.xml
We will try to add in later versions.

linaro image? Can you elaborate on it?
Do you think Android 9 is hotter than Android7(all other things being equal)?

I was noticing some increased CPU load (and heat) from com.rockchips.mediacenter. Since I don’t use it, I just disabled it pm disable com.rockchips.mediacenter and temperature went down. Now idling at;

# cat /sys/devices/virtual/thermal/thermal_zone?/temp
36875
36250
2600

Edit for those trying to compare temperatures… that’s a Rock960C with a modified AB heat sink case, so it won’t be directly comparable to RockPi4: https://store.vamrs.com/collections/rock960/products/metal-case-for-rock960

@Lili :

I’ve been finding it to be a problem that the console disconnects early in the boot.

The kernel is configured to put the console on ttyFIQ0;

# cat /proc/cmdline                                                                                                                                      
storagemedia=sd androidboot.mode=normal androidboot.dtbo_idx=1 root=PARTUUID=dea56848-7334-409d-a484-aed61e98819a androidboot.verifiedbootstate=orange androidboot.slot_suffix= androidboot.serialno=d1a806d913c70dbe console=ttyFIQ0 androidboot.baseband=N/A androidboot.wificountrycode=US androidboot.veritymode=enforcing androidboot.hardware=rk30board androidboot.console=ttyFIQ0 firmware_class.path=/vendor/etc/firmware init=/init rootwait ro init=/init loop.max_part=7 androidboot.selinux=permissive earlycon=uart8250,mmio32,0xff1a0000 swiotlb=1 coherent_pool=1m

But there is no such device;

# ls /dev
__properties__ cg2_bpf         dri            graphics    iep         loop-control     network_latency    ptmx   rtc0      tty   uhid          vpu_service 
ashmem         console         event-log-tags hdmi_hdcp1x iio:device0 mali0            network_throughput pts    rtkbt_dev ttyS0 uinput        xt_qtaguid  
binder         cpu_dma_latency fd             hwbinder    input       mem              null               random snd       ttyS2 urandom       zero        
block          cpuctl          fscklogs       i2c-0       keychord    memcg            opteearmtz00       rfkill socket    ttyS3 usb-ffs       
bus            cpuset          full           i2c-4       kmsg        memory_bandwidth pmsg0              rga    stune     ttyS4 usb_accessory 
cec0           device-mapper   fuse           i2c-9       kmsg_debug  mtp_usb          ppp                rkvdec sw_sync   tun   vndbinder     

Based on the u-boot configuration, it looks like it should be on 0xFF1A0000, which is ttyS2, so probably should be parameters as;
console=ttyS2,1500000n8 androidboot.console=ttyS2

no worry, i will check this with some future power measurements because the feeling of heatsink says nothing but here are some snapshots of the temperature outputs:

linaro image

43888
42777

android 9

47500
45625
2600

image
It USES tty2.
rk3399-android.dtsi

	fiq_debugger: fiq-debugger {
		compatible = "rockchip,fiq-debugger";
		rockchip,serial-id = <2>;
		rockchip,wake-irq = <0>;
		rockchip,irq-mode-enable = <0>;  /* If enable uart uses irq instead of fiq */
		rockchip,baudrate = <1500000>;  /* Only 115200 and 1500000 */
		pinctrl-names = "default";
		pinctrl-0 = <&uart2c_xfer>;
		interrupts = <GIC_SPI 150 IRQ_TYPE_LEVEL_HIGH 0>;
	};

@Lili : If that’s the case, then why am I getting this;

rk3399_box:/ # ls -l /dev/tty*
crw-rw-rw- 1 root      root   5,   0 2013-01-18 03:50 /dev/tty
crw-rw---- 1 bluetooth net_bt 4,  64 2019-07-26 07:09 /dev/ttyS0
crw------- 1 root      root   4,  66 2013-01-18 03:50 /dev/ttyS2
crw------- 1 root      root   4,  67 2013-01-18 03:50 /dev/ttyS3
crw------- 1 root      root   4,  68 2013-01-18 03:50 /dev/ttyS4
rk3399_box:/ #

EDIT: I didn’t notice you made a new build, with a change in it regarding ttyS2.
THANK YOU!