The buffer is already at max for ARM (32) , you need to unpack boot, and in default.prop of ramdisk allow debugging, and set ro.secure to =0
Tell me what's the screen / LCM revision of your unit, I can quickly do that for you
Hi. First, thanks for your work @superdragonpt.
I've been running build 20210821 for a few days and, in general, it's been working really good. I haven't experienced any problems with wifi or bluetooth, no crashes, and it feels smooth (considering how old the hardware is). I've only found two issues:
1. Offline charging is not working for me. When the tablet is turned off, the moment I connect the charger, it turns on automatically. And if I shut it down while it's charging, it will simply reboot.
2. The battery indicator never reaches 100%. It stays at 99% even if I unplug it and plug it again. Not a big deal, but I'm curious if that's to be expected or if it's a bug. Nevermind about this. For the last couple of days it wouldn't get past 99%, even after an hour, but that seems to be fixed. Maybe I had a calibration problem or something.
If you have twrp installed, and you have made a backup of your stuff on internal sdcard ...I started messing around with this tablet again because I was bored and followed the steps to install the CWM-based recovery. The CWM recovery refused to mount /sdcard and the tablet wouldn't boot (stuck on asus splash), however I was then able to install TWRP via sideload, which now refuses to mount system and refuses to use sideload. Anyone got any ideas?
Trouble is that I can't mount system at all in twrp, this is my second attempt after fully going back to stock and trying againIf you have twrp installed, and you have made a backup of your stuff on internal sdcard ...
Then on twrp just format / data, and then reboot to twrp again, and do all wipes ( including /system)
In twrp you can then just connect via USB to PC ( mtp works fine on twrp, unlike the old CWM recovery)
And just flash your desire ROM
Its an kernel issueHello fellow platformer
I was surprised to see that someone else is picking at this old platform besides me
Is your accelerometer broken in the core or in the LOS firmware?
I can help if anything, I just recently raised a similar one, I have a platform 8389. I, like you, was able to build the kernel without the source code (only the camera does not work yet)
Sorry for my bad English, I'm from Russia
Its an kernel issue
This tablet uses an uncommon power gpios, and also custom chips.
I.e. : cameras are connected to a custom External ISP chip the "iCatch" chip, this is the first MediaTek device that I know ...up to this date, that uses this chip, the chip is usually used on QCOMM devices...
( hence main camera still not working , and on the selfie camera I was abled to bypass this chip by using the slightly modified power-on sequence with gated IRQ )
Same thing for many drivers inside the device, they have been extremely modified and hard to reverse engineered
The magnetometer driver was also quite time consuming to reverse, it has also an uncommon way to be powered up.
same thing for the Goodix Touch.Panel
( took some years to figured it out)...
Asus really did an impressive job here ..
The "regular" drivers just don't workThe sad thing is that asus did not provide the complete source code for your device.
I came across their sources a couple of times and thought that they were complete, but recently I needed one driver, and it turned out that they are fake.
As far as I understand, Axel uses gpio for you?
I had a custom axel using gpio, but the reverse didn't help me.
Through trial and error, I picked up the driver and, having slightly edited it, started it.
It differs from the drain, but it works)
With the standard Ktj2-1009 driver, it doesn't work?
Usually, there are no problems with launching with these axels, you like the address 0x1E like my kxtik1004, it seems like some kind of ktjf2 uses such an address, does not work with them?
Yes, you have a more specific device that you almost made it work (not a working magnetometer is not critical).The "regular" drivers just don't work
I've tried that years ago, on the first tests.
The board was assembled very differently, the gpio / power and the way the components where connected on the board is very uncommon.
Hence the i2c slave addresses of this drivers aren't the "standard ".
Another example :The magnetometer needs to be manually powered up by calling the right gpio number that it is connected to, with the specific mdelay while the driver is loading.
This device has been an tremendous headache.
We can only thank Asus by making such an weird board design
And by not complying with GPL, the kernel source they released is " dummy " , just an generic one, not containing any drivers used by the device
fastboot flash recovery recovery.img