• Introducing XDA Computing: Discussion zones for Hardware, Software, and more!    Check it out!

Development [Kernel] (Anakin) Skywalker ZS673KS - Stable Gaming [10/16 - 2109.176]

Search This thread
First I would like to apologize as I'm a newbie. I would like to know whether is there a manager app or where I can access the Boeffla Wakelock Blocker v1.1.0 and Wireguard v1.0.20210606 ?

I've successfully installed the kernel via TWRP > Magisk(TWRP) > Kernel Flash(TWRP).

Thanks for the effort!

Any kernel manager should work for the wakelock blocker, but it's not something you should be using on stock firmware. It is a legacy feature that originated when wakelocks were poorly managed and currently exists only as a last resort for broken functionality that comes from poorly designed or modified drivers.

As for Wireguard, there is an official app
 

JazonX

Senior Member
Dec 16, 2009
1,833
728
Xiaomi Mi 11 Ultra
Flashed the latest as of last night, and everything went well. Used the device in the evening, no fuss whatsoever.

Woke up in the morning with an error message "Waiting for full ram dump"

Forced an exit with the 3 key combination, and the battery had drained from 96% - 72%.
It was in this state probably for about 2 hours, noticed only when I woke up.

What can be the cause?
Do you need logs from somewhere, if yes - how to pull?
 
Flashed the latest as of last night, and everything went well. Used the device in the evening, no fuss whatsoever.

Woke up in the morning with an error message "Waiting for full ram dump"

Forced an exit with the 3 key combination, and the battery had drained from 96% - 72%.
It was in this state probably for about 2 hours, noticed only when I woke up.

What can be the cause?
Do you need logs from somewhere, if yes - how to pull?

I've seen it once during testing and never again, even on that same build.

In trying to debug kernels, I realized that Asus has rerouted the location of some of the more important logs to make them "easily" accessible.

Disclaimer: This is not a complete list of every file included, but lists the more "useful" information for debugging and research.

[X] values represent numbers equal to the number of X's included
[XYZ] values represent a variable word(s) for multiple similar files

/asdf

18.[XXXX].[XXXX].[XX]-0
Records of each firmware update (may include skipped)

ASDF
Collection of encrypted / binary LastShutdown archives

ASUSEvtLog(_old).txt
System log that contains boot / shutdown details

boot_sound_[XYZ]
These are the details about boot sound / vibration

boot_sound_vol_index
Boot sound volume, but changes are overwritten

last_kmsg_16K
This appears to be a slightly corrupted last_kmsg

logcat-crash.txt
This is a record of system crashes logged by logcat

versionHistory.txt
Installed firmware updates by date / time since last restore

Those are all the logs, but the only one that has any decent detail is ASUSEvtLog
 
18.0840.2109.176 was a CN exclusive firmware, but appears to have updated some drivers, so we're going to go ahead and update, too.

At least if the firmware doesn't make it outside China, the kernel will...

Isn't the idea with Blocked Touch Edges that the set edge areas are blocked to prevent accidental touches? It does nothing no my device. On the latest WW firmware.


View attachment 5363925


Code:
static ssize_t fts_xy_resize_store(
    struct device *dev, struct device_attribute *attr, const char *buf, size_t count)
{
    struct fts_ts_data *ts_data = fts_data;

    if (FTS_SYSFS_ECHO_ON(buf)) {
        ts_data->resize = ENABLE;
    } else if (FTS_SYSFS_ECHO_OFF(buf)) {
        ts_data->resize = DISABLE;
    }

    FTS_DEBUG("touch area resize:%d", ts_data->resize);
    return count;
}

Looks like they finally realized there was a problem
 
Last edited:
Haha! Awesome ...

I thought a WW variant might come shortly after.
Guess not ...

I'm guessing they assumed that since the changelog items were directed at CN apps, there was no need to release it as WW. If the code is what it looks like, though, it also fixes an issue that has been around since July.

 

gouen1995

Member
May 12, 2021
27
3
Hi I'm sorry to trouble you but I need a little help, I was listening to Spotify today and unplugged my earphone. The 'Waiting for flashing full ramdump...' warning screen appeared.

After that I search for the internet abit and tried the following :
1. Third finger reboot method -> Which results in a very long waiting boot then fallback to recovery automatically
2. Enter recovery mode (TWRP) -> Restore previously backed kernel boot -> Which results in the 'Your device is corrupt' screen.
3. Enter recovery mode (TWRP) -> Wipe -> Which results in a shorter bootloop

I needed your professional advice on this, since my bootloader is unlocked I guess warranty isn't an option. Is there anyway to save this?

Untitled (1).jpg
 

JazonX

Senior Member
Dec 16, 2009
1,833
728
Xiaomi Mi 11 Ultra
Hi I'm sorry to trouble you but I need a little help, I was listening to Spotify today and unplugged my earphone. The 'Waiting for flashing full ramdump...' warning screen appeared.

After that I search for the internet abit and tried the following :
1. Third finger reboot method -> Which results in a very long waiting boot then fallback to recovery automatically
2. Enter recovery mode (TWRP) -> Restore previously backed kernel boot -> Which results in the 'Your device is corrupt' screen.
3. Enter recovery mode (TWRP) -> Wipe -> Which results in a shorter bootloop

I needed your professional advice on this, since my bootloader is unlocked I guess warranty isn't an option. Is there anyway to save this?

View attachment 5420027


I was in the exact same situation, Happened about 3 times yesterday.
Just keep the 3 buttons (Volume Up + Down + Power) pressed for 5 seconds, Phone will shut off leaving this ram dump mode.

Then give a 2 - 3 min break and power on normally.

Since I had an important day ahead, I flashed the Kirisakura kernel for now.

If you flashed stock, It's probably trying to force the encryption again.
Flash the Kirisakura kernel from TWRP, and you'll be back up ...
 

gouen1995

Member
May 12, 2021
27
3
I was in the exact same situation, Happened about 3 times yesterday.
Just keep the 3 buttons (Volume Up + Down + Power) pressed for 5 seconds, Phone will shut off leaving this ram dump mode.

Then give a 2 - 3 min break and power on normally.

Since I had an important day ahead, I flashed the Kirisakura kernel for now.

If you flashed stock, It's probably trying to force the encryption again.
Flash the Kirisakura kernel from TWRP, and you'll be back up ...
I very much appreciate for your advice. It seems like now i'm stuck in bootloader mode and can't even boot into TWRP. Guess I had to flash the stock ROM first. Anyway thanks for the help.
 
I very much appreciate for your advice. It seems like now i'm stuck in bootloader mode and can't even boot into TWRP. Guess I had to flash the stock ROM first. Anyway thanks for the help.

All you ever need to restore is to flash the stock boot, vendor boot, and dtbo. You'll be back up without any issues or data loss. Flashing another kernel without restoring to stock is risky when suspecting an issue. You may still be able to recover with just those three, though.


It is recommended you patch the boot image with Magisk using your own device, so it is best to flash the boot image that does NOT end in "-magisk" to restore.


I haven't experienced the ramdump since one of the early tests and @JazonX didn't post logs. My best guess was something we had both tried at one point that cleaned itself up after the reboot. I wasn't aware it was still happening, but the only similarity I could come up with was installing in TWRP.

TWRP being built on Kirisakura could explain why it seems to be exempt from the issue. As for the root cause, I can only guess until I can manage to force another ramdump to diagnose it.

Since TRWP kernel source points back to the main Kirisakura repository, Kirisakura hasn't been updated, and there is no new version of TWRP... my best guess would be that the issues with kernel install are still unresolved. Until an update, it's probably best not to do kernel install there anyway.
 
Last edited:
  • Like
Reactions: mohammed510

gouen1995

Member
May 12, 2021
27
3
All you ever need to restore is to flash the stock boot, vendor boot, and dtbo. You'll be back up without any issues or data loss. Flashing a different custom kernel without restoring to stock when there is an issue is risky. You may still be able to recover with just those three, though.


It is still recommended you patch the boot image with Magisk using your own device, so it is best to flash the boot image that does NOT end in "-magisk" to restore.

I haven't experienced the ramdump since one of the early tests, but @JazonX never came back with any logs and my best guess was something we had both tried at one point that cleaned itself up after the reboot.

Now I have a feeling I know what the issue is, at least. The one thing we had in common was installing with TWRP. I typically use EX and haven't had the issue with that. TWRP being built on Kirisakura would most likely be why it seems to be exempt from the issue. As for the root cause, I can only guess until I can manage to force another ramdump notice to diagnose it.

Since TRWP kernel source points back to the Kirisakura repository with no record of any changes for TWRP, Kirisakura hasn't been updated, and there is no new version of TWRP... my best guess would be that the issues with kernel install are still pending. Until they update it, it's probably best not to do kernel install there.
Hi, thanks for the help. I'm kinda in an awkward and senseless position right now.

First off, I tried to flash boot.img, vendor_boot.img, and dtbo.img using 'fastboot flash boot boot.img' but it seems to be flashing into the 'a' partition. I tried to manually flashing into both partition a and b and had no luck..

2. Everytime i tried to 'fastboot reboot fastboot', I seemed to be trapped in a fastbootd bootloop which consist of this error 'Failed to boot into userspace fastboot; one or more components might be unbootable.'

I wanna get advice from you guys before I wipe my userdata via 'fastboot -w' and I'm not even sure if it will do any good. :/ I hate to be such a nuisance but I hope you guys could give me a chance to learn.
 
Hi, thanks for the help. I'm kinda in an awkward and senseless position right now.

First off, I tried to flash boot.img, vendor_boot.img, and dtbo.img using 'fastboot flash boot boot.img' but it seems to be flashing into the 'a' partition. I tried to manually flashing into both partition a and b and had no luck..

2. Everytime i tried to 'fastboot reboot fastboot', I seemed to be trapped in a fastbootd bootloop which consist of this error 'Failed to boot into userspace fastboot; one or more components might be unbootable.'

I wanna get advice from you guys before I wipe my userdata via 'fastboot -w' and I'm not even sure if it will do any good. :/ I hate to be such a nuisance but I hope you guys could give me a chance to learn.

Each image should be flashed to the partition with the same name. For example "fastboot flash vendor_boot vendor_boot.img"

It sounds like a safer alternative would be to flash raw firmware without wiping before resorting to a wipe, though.


You want the flash raw command that doesn't wipe.

I really need to grab the commands from that one to be used with the others that didn't include the option, but that's unrelated to this.
 

gouen1995

Member
May 12, 2021
27
3
Each image should be flashed to the partition with the same name. For example "fastboot flash vendor_boot vendor_boot.img"

It sounds like a safer alternative would be to flash raw firmware without wiping before resorting to a wipe, though.


You want the flash raw command that doesn't wipe.

I really need to grab the commands from that one to be used with the others that didn't include the option, but that's unrelated to this.
Thanks for the reply and help sir. But do you happen to have the flash_all command script for Mac OS?
 
Code:
#!/bin/sh

#===== DEFINE =====#
Image=`find . -type f -name "*.raw"`
#==================#

echo =============================================================
echo raw=${Image}
echo =============================================================

echo Flashing now ...
echo

./fastboot flash all ${Image}

echo
echo =============================================================
echo Flashing raw successfully
echo =============================================================
echo
./fastboot reboot
 
I haven't gotten a ramdump yet, but I let it sit a while to see if it's an idle thing. The worst part of an issue like that is not being able to push it to happen to trace it up the line.

That or it was accidentally fixed already. How lucky would that be...

Edit: After sitting idle for hours, it hasn't crashed out. I honestly don't know.
 
Last edited:

Optimize LTO kernel binary compression
Fix a few trace address inconsistencies
General cleanup for extended kernel test

This build was cleaned up to do some longer testing in the hope that if an issue is still causing the dump, it is encountered to trace the cause.
 
Last edited:

gouen1995

Member
May 12, 2021
27
3
Code:
#!/bin/sh

#===== DEFINE =====#
Image=`find . -type f -name "*.raw"`
#==================#

echo =============================================================
echo raw=${Image}
echo =============================================================

echo Flashing now ...
echo

./fastboot flash all ${Image}

echo
echo =============================================================
echo Flashing raw successfully
echo =============================================================
echo
./fastboot reboot
Thank you very much for your help, I successfully unbricked my phone but sadly a factory wipe is needed issued by fastbootd. But it is totally ok for me.

This script doesn't seem to be working on macOS, returning me some sort of error unable to execute executable. Whip out of old windows machine to flash it instead.

As for the ramdump issue, I wish I could be of any help. But I will keep an eye on your kernel. Thanks again.
 
  • Like
Reactions: twistedumbrella

Top Liked Posts

  • There are no posts matching your filters.
  • 1
    I very much appreciate for your advice. It seems like now i'm stuck in bootloader mode and can't even boot into TWRP. Guess I had to flash the stock ROM first. Anyway thanks for the help.

    All you ever need to restore is to flash the stock boot, vendor boot, and dtbo. You'll be back up without any issues or data loss. Flashing another kernel without restoring to stock is risky when suspecting an issue. You may still be able to recover with just those three, though.


    It is recommended you patch the boot image with Magisk using your own device, so it is best to flash the boot image that does NOT end in "-magisk" to restore.


    I haven't experienced the ramdump since one of the early tests and @JazonX didn't post logs. My best guess was something we had both tried at one point that cleaned itself up after the reboot. I wasn't aware it was still happening, but the only similarity I could come up with was installing in TWRP.

    TWRP being built on Kirisakura could explain why it seems to be exempt from the issue. As for the root cause, I can only guess until I can manage to force another ramdump to diagnose it.

    Since TRWP kernel source points back to the main Kirisakura repository, Kirisakura hasn't been updated, and there is no new version of TWRP... my best guess would be that the issues with kernel install are still unresolved. Until an update, it's probably best not to do kernel install there anyway.
    1
    Code:
    #!/bin/sh
    
    #===== DEFINE =====#
    Image=`find . -type f -name "*.raw"`
    #==================#
    
    echo =============================================================
    echo raw=${Image}
    echo =============================================================
    
    echo Flashing now ...
    echo
    
    ./fastboot flash all ${Image}
    
    echo
    echo =============================================================
    echo Flashing raw successfully
    echo =============================================================
    echo
    ./fastboot reboot
    Thank you very much for your help, I successfully unbricked my phone but sadly a factory wipe is needed issued by fastbootd. But it is totally ok for me.

    This script doesn't seem to be working on macOS, returning me some sort of error unable to execute executable. Whip out of old windows machine to flash it instead.

    As for the ramdump issue, I wish I could be of any help. But I will keep an eye on your kernel. Thanks again.
    1
    I've re-flashed the latest kernel to give it a run, it's been a 1+ days for now and so far there's no issue. And I noticed there's a new WW firmware version, it is safe for me for update the version with the kernel?

    Glad to hear the issue hasn't come back. The kernel was made for the newer version before it reached WW, so it is definitely compatible.

    If you don't already know the update process, it is:

    Download the firmware from Asus and put it on the root of the sdcard
    Reboot the phone so it detects the update but do NOT reboot after the install reaches 100%
    Open Magisk and click the first install, then select the inactive slot option
    Reboot after Magisk has completed it's install, then you can reinstall the kernel
    1
    I've re-flashed the latest kernel to give it a run, it's been a 1+ days for now and so far there's no issue. And I noticed there's a new WW firmware version, it is safe for me for update the version with the kernel?

    Sorry for not replying in your other thread, but it's being trolled and I was curious if the drain started before or after the kernel. There is no such thing as a "system cache" anymore, so don't waste your time.

    You may want to look into some of the dial codes people have used to unlock volte. There may be options in there to lock certain settings, too.


    If it started after the kernel, I will definitely look into things on my end.
  • 6
    Obsolete post. Please disregard.
    5
    Vegeta9000.gif


    This kernel is (just barely a little) experimental. Use at your own risk.
    Feel free to skip past the first 5 pages when reading over this thread.​


    This kernel assumes a few things about you, the user:

    1. You have a ROG Phone 5
    2. You have read all warnings and instructions
    3. You are rooted (implying an unlocked bootloader)
    4. You have an installation method
    5. You have a backup boot image


    Releases will be labeled with the commit hash of their source. The newest firmware is recommended for this kernel, but not necessarily required. I am not responsible for any issues or damage caused by installing this kernel, so have a backup ready before proceeding.



    Archived Releases (Skywalker)​



    Thank you to everyone that has contributed to kernel development.
    Specific credit is provided by maintaining accurate commit history.


    4
    [Thread Cleaned]

    Please keep the conversation respectful and on-topic. Think to yourself: "Will this contribute to the development of this kernel?" before posting and we'll all be much happier. Any questions or concerns about this announcement can be voiced in my PM box. Just please let our devs work their magic and support them; it's the only way forward.

    Stay safe!
    Timmy
    Forum Moderator
    2
    Formatted my hard drive and installed Mojave. Hopefully whatever issues were on my end should get resolved by doing everything fresh.

    Asus has also begun diagnosing the issue. Their support is horribly slow, but has at least agreed to investigate.
    2

    Asus has just released new firmware and source for V18.0840.2104.47

    The next kernel release will require updating to V18.0840.2104.47