Related
i read something in the net while i was surfing yesterday, you can look here: http://www.androidcentral.com/motorola-atrix-hd-gets-official-cyanogenmod-support
and then reading these lines made me surprised(maybe getting official cyanogenmod support for ancora is impossible after this moment but who kowns?)
If an Android device has enough of a following, and there are developer(s) willing to commit (no pun intended) to maintaining the device in the official CyanogenMod source tree, this device becomes officially supported -- and receives official builds of the CyanogenMod ROM.
if someone be maintainer for our device can we get official support from CyanogenMod?
is this really possible?
ı mean our device source code are available in github(kernel source device source and vendor files)
maybe forking our device source into CyanogenMod source tree can make this possible or am ı wrong??
maybe this is a bulsh*it after this moment but just wondering
for example galaxy s is getting official support from Cyanogenmod still if im not mistaken mr. pawit is their maintainer and developer
ım waiting your answers brothers what do you think about that? @arco68 @arsradu @LeroViten @honeyx and all ancora devs
with my best regards :good:
The SGW was for more than 3 years without official CM support and we didn´t miss anything, so I don´t think this is that much important for the rest of it´s counted days.
honeyx said:
The SGW was for more than 3 years without official CM support and we didn´t miss anything, so I don´t think this is that much important for the rest of it´s counted days.
Click to expand...
Click to collapse
ı understood bro but is there any hope for galaxy w? return of ancora :good: and also ı understood that from your comment : technically this is possible,galaxy w can get official support from CM Team but this is unnecessary for this dead device isnt it?
I simply don´t think this will happen because you have to look from their point of view.
1. this device is a low activity device in Amercian terms. 2. So never was that popular to get their attention. 3. is already old and not up to date. 4. It´s hype time is almost over and many earlier SGW users are going to replace it. 5. CM is going a different route now and will be focusing on their "device" they are going to launch soon. And finally 6. Why should they change this now after already 3 years?
So if I were CM I would look for the future not looking back to the past with it´s outdated devices and not even waste a single tought about devices like the SGW.
Is this bad for the SGW owners? As long we are having Devs like Radu who are porting the latest ROMs to this device it really doesn´t matter CM is supporting it officially or not supporting it.
So what do you expect from being supported by CM? You only would be able to install the latest nightlys launched by the CM team with all the bugs in the CM code included and maybe some day fixed. A Dev who is porting their CM sources of course has to fight with these bugs too, but also cares to fix the most important ones.
I asked arco68 a while ago, and he said it's because of legacy camera drivers that we are unsupported.
Sent from my Nexus 5 using Tapatalk
SGWUser said:
I asked arco68 a while ago, and he said it's because of legacy camera drivers that we are unsupported.
Sent from my Nexus 5 using Tapatalk
Click to expand...
Click to collapse
auhh thats true sir arne is right as always anyway thanx man for informing us
regards :good:
I don't think so...anyway, we already have the stable n smooth ROM so far. Without Officail, it's does not matter at all.
Hello sir @arco68
First of all huge thanx for your great work on our Wonders
And secondly i wonder something
We are using s5k4ecgx sensor for back camera
And if we add the following patches and if we use v4l2 for camera driver, Can we use CAF based qcom cameraHAL?
-http://www.spinics.net/lists/linux-media/msg50380.html
And probably if we use new qcom cameraHAL, then we wont need legacy stuff the CM doesnt like(MemoryHeapBase,Overlay)
Then maybe we can get cm nightlies
Maybe porting new camera Hal to our phone can be hard but we must not give up. We are so close to official CM after ION i think. We can do it and we can success :good:
Thanx
Regards
hack4er said:
Hello sir @arco68
First of all huge thanx for your great work on our Wonders
And secondly i wonder something
We are using s5k4ecgx sensor for back camera
And if we add the following patches and if we use v4l2 for camera driver, Can we use CAF based qcom cameraHAL?
-http://www.spinics.net/lists/linux-media/msg50380.html
And probably if we use new qcom cameraHAL, then we wont need legacy stuff the CM doesnt like(MemoryHeapBase,Overlay)
Then maybe we can get cm nightlies
Maybe porting new camera Hal to our phone can be hard but we must not give up. We are so close to official CM after ION i think. We can do it and we can success :good:
Thanx
Regards
Click to expand...
Click to collapse
That would be so f... Awesome *-*
Please we need someone who knows how to do this
if cm11 out i will buy one
Sent from my Moto G using Tapatalk
I'm currently working on unofficial build CM10.2.
Kernel is done
device tree is done
vendor setup is done
Android sources currently downloading
testing not yet.
Riyal said:
I'm currently working on unofficial build CM10.2.
Kernel is done
device tree is done
vendor setup is done
Android sources currently downloading
testing not yet.
Click to expand...
Click to collapse
Not CM11, but still very exciting =)
TheQwertiest said:
Not CM11, but still very exciting =)
Click to expand...
Click to collapse
I will try cm11 once i get cm10 running. Thing is we don't have official android 4.4 roms yet I don't have any sources for 4.4 binaries and kernel
Riyal said:
I will try cm11 once i get cm10 running. Thing is we don't have official android 4.4 roms yet I don't have any sources for 4.4 binaries and kernel
Click to expand...
Click to collapse
We wait for CM11!
Xperia V, T doesn't have official 4.4 ROM too, but they have CM11.
Also, I think it's possibile developing CM11 or other CM11 based (KitKat) AOSP ROM.
Unfortunatelly I can't develop it...
I just will use it :fingers-crossed:
u guyz are lucky...u have munjeni wid u now. just be patient.. his hardwork brought cm11 novathor devices... which were abondoned on ics.
two reason i see z1compact as my next phone is that i dont like large phones and master munjeni has z1c also...
We can use CM11 from Z1 since Z1 CM11 on Z1C is working but there is no display available since CM kernel on MSM8974 need some changes related to the Z1C display Bad news is we can not use stock kernel since gralloc, copybit ..etc is not working than. I started working on CM kernel but hmmm CM kernel and stock kernel is a lot diferent so hard work is needed If we get CM kernel working we will get CM11 working with no big modifications
Edit:
Compared now stock Sony kernel source code from Z1 and stock ekrnel source code from Z1C, there is absolutly no diferencie, only one line is diferent in defconfig, line is CONFIG_MACH_SONY_*****_ROW, source is totaly identic. I'm confused now since have no idea why CM kernel working on Z1 but no on Z1C. Maybe I have downloaded wrong kernel for Z1 ??? I have downloaded https://github.com/CyanogenMod/android_kernel_sony_msm8974 branch cm-11.0 . Maybe these kernel not working on Z1 too ???
Can't wait to try CM
kernel config from boot.img
munjeni said:
Compared now stock Sony kernel source code from Z1 and stock ekrnel source code from Z1C, there is absolutly no diferencie, only one line is diferent in defconfig, line is CONFIG_MACH_SONY_*****_ROW, source is totaly identic. I'm confused now since have no idea why CM kernel working on Z1 but no on Z1C. Maybe I have downloaded wrong kernel for Z1 ??? I have downloaded{removed} branch cm-11.0 . Maybe these kernel not working on Z1 too ???
Click to expand...
Click to collapse
Hello,
I am not sure if this is helpfull, but I'll give it a try: You are right - there is absolut no difference between the two kernel configurations, beside that one line. Since there are already nightlies available for the Z1(honami) I thought there might some helpfull informations inside that nightly. Of course there was a boot.img inside the nightly and I found that script to extract the kernel.config from that boot.img. Using the extract-ikconfig from my xperia z(yuga) kernel located in ./kernel/sony/apq8064/scripts/extract-ikconfig.
If the extracted config is reliable, there are indeed many differences.
I attached the extracted config. Maybe you remove the Lines starting with # and the blank lines for a better compare.
Let me know if I can help in some way.
Kind Regards
I don't know whats going on or why CM kernel is working on Z1 and not on Z1C but I think these fails on Z1C is related to the baseband and bootloader? Just asking, DO NOT FLASH THEM!!!
And another thing is maybe dtbs is a thing related to the fail since I can see some things in dmesg related to the pingpong command in mdss_mdp_intf_cmd.
Error log -> http://pastebin.com/9zgcWd4a
Thanks to the developers who are trying to make CM11 working on this phone. I will buy this phone if CM11 is available
@munjeni
Guys I have compiled a CM10.2 kernel here. Who's willing to try it against Z1's CM10.2 ROM?
http://forum.xda-developers.com/showthread.php?t=2461282
I'll provide the source if it works.
Riyal said:
@munjeni
Guys I have compiled a CM10.2 kernel here. Who's willing to try it against Z1's CM10.2 ROM?
http://forum.xda-developers.com/showthread.php?t=2461282
I'll provide the source if it works.
Click to expand...
Click to collapse
You can try! You no need to install, simple boot them without installing! On next reboot your phone will boot normaly! So boot these image temporary by command:
fastboot boot boot.img
Nope, doesn't seem to boot :/
munjeni said:
You can try! You no need to install, simple boot them without installing! On next reboot your phone will boot normaly! So boot these image temporary by command:
fastboot boot boot.img
Click to expand...
Click to collapse
Yeah but I just relocked my bootloader yesterday. I won't be unlocking it till I get more time in development.
crytic said:
Nope, doesn't seem to boot :/
Click to expand...
Click to collapse
Weird then... The patch I applied to port z1 kernel to z1 compact made a fully working stock kernel. But when I applied it to CM kernel it doesn't even boot. Can you by chance get a copy of /proc/last_kmesg and upload it somewhere or here?
* daemon not running. starting it now on port 5037 *
* daemon started successfully *
~ # ←[6nuname -r
uname -r
3.4.0cyanogenmod-g77098c0-dirty
~ # ←[6n
Click to expand...
Click to collapse
munjeni said:
Click to expand...
Click to collapse
Woot! There it goes Now I don't have to port CM haha!
Sweet, with or without display?
crytic said:
Sweet, with or without display?
Click to expand...
Click to collapse
With! Investigating between commits on git since using old working commits, need to find which commit broke display init. There is ~100 commits between but I will try every 20.th When commits get located than will try to fix upstream kernel . There is a lot of work but thats a only way by now.
munjeni said:
Click to expand...
Click to collapse
munjeni said:
With! Investigating between commits on git since using old working commits, need to find which commit broke display init. There is ~100 commits between but I will try every 20.th When commits get located than will try to fix upstream kernel . There is a lot of work but thats a only way by now.
Click to expand...
Click to collapse
git bisect to the rescue Well, if you need any help, just drop me a line!
Hallo i watch all this roms... They are all 4.2.2. In the sort time adroid L its out... And i have the big question when we have one real update ??? I have this phone 9 months until now i live in eu
gkm8123 said:
Hallo i watch all this roms... They are all 4.2.2. In the sort time adroid L its out... And i have the big question when we have one real update ??? I have this phone 9 months until now i live in eu
Click to expand...
Click to collapse
100% agree but unfortunately duo to mate 2 is out they have no plan to update mate1 to 4.3 or 4.4:crying:
i really think ham is good phablet and deserve better support from manufacturer.
mazi_v said:
100% agree but unfortunately duo to mate 2 is out they have no plan to update mate1 to 4.3 or 4.4:crying:
i really think ham is good phablet and deserve better support from manufacturer.
Click to expand...
Click to collapse
This kind of support let me turn away from Huawei. A pitty. But i want decent support. A phone schould get 2 year regular updates. Thats wat a customer deserves
Well you'll have to dive deeply into it to do the job of building a custom Android ROM. I just downloaded the kernel sources from the official JB (4.1) release, but I'm not sure how far I can get with that. Some experienced devs here posted that they would need Huawei's sources for KitKat to build a KitKat ROM, but I also have an Ascend P1 and I'm running a custom KitKat CyanogenMod build from the forums here although Huawei only released ICS (4.0) Kernel sources. I will have a try with a custom cross compiling toolchain of mine that I use for my Radxa Rock, but it's quite generic. I'll let you know how far I could get if at all.
Cheers!
The kernel build was successful, but I've simply used Google's latest toolchain (arm-linux-androideabi-4.6-android-4.4.4_r1). I'm now syncing the most recent Android repo, that will take some time. I will try to set everything up to start compiling tomorrow morning or something...
Just in case you want to build the kernel yourself as well: There is an indentation issue in one of Huawei's Makefiles for the touchscreen, you'll see an error message about it if you run the first make according to the README_kernel.txt. You can just add tabs and your file should look like this:
[drivers/huawei/device/touchscreen_Mate/Makefile]
Code:
[COLOR="DarkOrange"]#
# Makefile for the touchscreen drivers.
#
# Each configuration option enables a list of files.
obj-y += synaptics_i2c_rmi.o
obj-y += rmi_dev_6.o
obj-y += rmi_f54_6.o
obj-y += rmi_fw_update_6.o
all:
make -C /lib/modules/$(shell uname -r)/build M=$(PWD) modules
clean:
make -C /lib/modules/$(shell uname -r)/build M=$(PWD) clean[/COLOR]
CyReVolt said:
The kernel build was successful, but I've simply used Google's latest toolchain (arm-linux-androideabi-4.6-android-4.4.4_r1). I'm now syncing the most recent Android repo, that will take some time. I will try to set everything up to start compiling tomorrow morning or something...
Just in case you want to build the kernel yourself as well: There is an indentation issue in one of Huawei's Makefiles for the touchscreen, you'll see an error message about it if you run the first make according to the README_kernel.txt. You can just add tabs and your file should look like this:
[drivers/huawei/device/touchscreen_Mate/Makefile]
Code:
[COLOR="DarkOrange"]#
# Makefile for the touchscreen drivers.
#
# Each configuration option enables a list of files.
obj-y += synaptics_i2c_rmi.o
obj-y += rmi_dev_6.o
obj-y += rmi_f54_6.o
obj-y += rmi_fw_update_6.o
all:
make -C /lib/modules/$(shell uname -r)/build M=$(PWD) modules
clean:
make -C /lib/modules/$(shell uname -r)/build M=$(PWD) clean[/COLOR]
Click to expand...
Click to collapse
A kitkat kernel that would make you in my opinion almost immortal for Mate users
henk_j said:
A kitkat kernel that would make you in my opinion almost immortal for Mate users
Click to expand...
Click to collapse
We'll see, I'm still learning about the stuff, but the chances look quite fair. I've set up the rest of the environment and the build is running now. But I guess it will take some more hours. I usually need about 3.5 hours to build chrome, firefox or libreoffice. AOSP should be similar. I'll see how far it got today evening/night. But for now it looks good, I'm running with -j4 as recommended. ^^
CyReVolt said:
We'll see, I'm still learning about the stuff, but the chances look quite fair. I've set up the rest of the environment and the build is running now. But I guess it will take some more hours. I usually need about 3.5 hours to build chrome, firefox or libreoffice. AOSP should be similar. I'll see how far it got today evening/night. But for now it looks good, I'm running with -j4 as recommended. ^^
Click to expand...
Click to collapse
if you have more cores processor why dont increase the thread value so that build process will complete more quickly
svvv said:
if you have more cores processor why dont increase the thread value so that build process will complete more quickly
Click to expand...
Click to collapse
I have 2 cores, so 4 is quite fine. The next option would have been 16.
Take your time! we need more developers
Any news ?
Really appreciate the works here.
so many had forgotten Mate 1 and P6 since 2 & 7 came out...
For me the first editions are always the best although the the 2nd might have little differences or improvements (but always have flaws or bugs more than the 1st editions)..
so very much grateful and thanks for devs that spent their time looking at mate 1 and p6...
Will donate if theres any need for it..count on me for that..
[email protected]
gkm8123 said:
Any news ?
Click to expand...
Click to collapse
Come on, I'm writing every couple of hours...
Well, the news is: v8 failed in some point. I've just resynced the AOSP repo to start over. ^^ I'll let you know how it went today evening. I also have a job, you know.
CyReVolt said:
Come on, I'm writing every couple of hours...
Well, the news is: v8 failed in some point. I've just resynced the AOSP repo to start over. ^^ I'll let you know how it went today evening. I also have a job, you know.
Click to expand...
Click to collapse
Ok no problem, thank so much for your help
Well, the newer patches didn't resolve the issue so I decided to try the normal user build (I tried a full debug build first). For now it looks very promising. I'm not sure if I can figure out how to to fix the issue I had, so if this one fails at the same point (although as far as I understood it was one of the debug parts so it shouldn't), it might get a little harder. We'll see ^^
thank you mate.. i mean CyReVolt.
i hope ur doing ok with ur job also,, and really hope u give us a good news..
cumanisengdoang said:
thank you mate.. i mean CyReVolt.
i hope ur doing ok with ur job also,, and really hope u give us a good news..
Click to expand...
Click to collapse
Sure, np... actually, it failed again due to the same problem, and I quickly figured out the cause (I already had suspected that): I'm on a strongly hardened Linux system, and the
Code:
mksnapshot
tool was killed due to NX protection (see http://wiki.gentoo.org/wiki/Hardened/PaX_Quickstart#Understanding_PaX). So if you're running a PaX protected kernel or similar you might run into the same issue. Duh... -.-
It's up and running again, so let's wait for a result!
While waiting you can try the Android kernel I have attached. Don't ask me how to do that. I have no clue. This is just for fun. And for you!
Considering work: It was actually quite funny today because I had to work on a Mac and OSX feels just horrible to me. My boss asked me to take a glance at Xcode and especially the new beta and Swift. I'm so glad to be back at my Linux machine now.
CyReVolt said:
Well you'll have to dive deeply into it to do the job of building a custom Android ROM. I just downloaded the kernel sources from the official JB (4.1) release, but I'm not sure how far I can get with that. Some experienced devs here posted that they would need Huawei's sources for KitKat to build a KitKat ROM, but I also have an Ascend P1 and I'm running a custom KitKat CyanogenMod build from the forums here although Huawei only released ICS (4.0) Kernel sources. I will have a try with a custom cross compiling toolchain of mine that I use for my Radxa Rock, but it's quite generic. I'll let you know how far I could get if at all.
Cheers!
Click to expand...
Click to collapse
The main problem is not the kernel. We can build a kitkat rom with the 3.0.8 kernel , but we have no source for display part , audio part. We have to use the prebuilt hwcomposer , grallloc etc. . These prebuilt libs are compatible only with the android version of the stock rom they are coming from and with the kernel they are patched for.
With sources of display driver we can patch the kernel and the driver to work on KK . Like I did JB on ZTE V11,V71,V55 tablet with 2.6.35 kernel , but that device was QC based device with available display driver sources. (http://forum.xda-developers.com/showthread.php?t=1629630)
I have compiled a 4.3 rom for mate , but I am stuck at display. There are changed routines in framework which are not working with 4.2's display drivers . One solution can be to get these sources , or to change the framework to work with the 4.2' display driver.
The Mate is my first non QC chipset based device , but also the last too. This is my fourth device from Huawei and until now I was satisfied , they simply do not care Mate customers. The release of the KK for Mate is just a few hours job , but they wont do it.
joe.stone said:
The main problem is not the kernel. We can build a kitkat rom with the 3.0.8 kernel , but we have no source for display part , audio part. We have to use the prebuilt hwcomposer , grallloc etc. . These prebuilt libs are compatible only with the android version of the stock rom they are coming from and with the kernel they are patched for.
With sources of display driver we can patch the kernel and the driver to work on KK . Like I did JB on ZTE V11,V71,V55 tablet with 2.6.35 kernel , but that device was QC based device with available display driver sources. (http://forum.xda-developers.com/showthread.php?t=1629630)
I have compiled a 4.3 rom for mate , but I am stuck at display. There are changed routines in framework which are not working with 4.2's display drivers . One solution can be to get these sources , or to change the framework to work with the 4.2' display driver.
The Mate is my first non QC chipset based device , but also the last too. This is my fourth device from Huawei and until now I was satisfied , they simply do not care Mate customers. The release of the KK for Mate is just a few hours job , but they wont do it.
Click to expand...
Click to collapse
Thats why we have to get some sense into their heads thats trough they're Wallet. The only way tot make them do this strangely by hurting them where it counts.
Alright, I'm getting the point now... so I did some research.
The GPU is a Vivante GC4000, so might this help https://github.com/laanwj/etna_viv ?
I wouldn't mind the sound at first. I would love to work on this somehow.
Can you point out which files I should look at? I know a bit about the ARM/Thumb ISA, so I'd try to reverse engineer a little to either adjust the pre-built drivers or the framework using them. =) I don't give up very quickly, you know! Finally, I'd just contact Huawei and talk straight to them. Thanks in advance for all your efforts.
Cheers
CyReVolt
Due to several reasons I have stopped the deployment of the "monthly Update.Zip packages" !
Thanks for your understanding.
This thread will be used for the unified tegra-3.18 kernel only.
Twelve:
tegra-3.18_unified_dragon_kernel_12_v211213.zip
Eleven:
tegra-3.18_unified_dragon_kernel_11_v201218.zip
Ten:
tegra-3.18_unified_dragon_kernel_10_v201010.zip
Pie:
tegra-3.18_unified_dragon_kernel_p_v200310.zip
Oreo:
tegra-3.18_unified_dragon_kernel_o_v200310.zip
Changes/Features:
- Upstream to kernel version 3.18.140
- Various patches and security updates
- Added Wireguard support
- Added/Enabled various file systems (exFat, NTFS, Network Filesystems .. etc.)
- Changed kernel config to enable/improve support for varioius services (tethering .. etc.)
- Added "fix cpu frequency limitation"
- etc.
Installation:
1. Boot into TWRP
2. Install the Kernel Update.Zip file
3. Reboot into system .. no need for SuperSU / Magisk !
The package contains an Image.fit kernel binary only - and no ramdisk part !
This way the kernel should run on all available Android Oreo/Pie/Ten ROMs for the dragon device.
And there is no need for a new SuperSu/Magisk installation afterwards, unlike to "full boot.img" installation.
Sources:
https://github.com/followmsi/android_kernel_tegra
https://github.com/followmsi/lazyflasher/tree/followmsi-dragon
https://github.com/followmsi/AnyKernel2/tree/dragon
Download:
https://drive.google.com/drive/folders/18cqluw7lp1S720QdV0p-OtHfY9MNpSwn
Big thanks to @NYCHitman1 (Dirty Unicorn), @cheep5k8 (xceed), @Vartom and updateing from Team Mokee for the nice kernel sources !
Big thanks to @osm0sis (Anykernel2) and @jcadduono (lazyflasher) for the nice tools !
Enjoy
flashed your system image
devices does not boot
flashed back to stock
boots
edit: further details
every boot just dumps me into recovery.
dmitrygr said:
flashed your system image
devices does not boot
flashed back to stock
boots
edit: further details
every boot just dumps me into recovery.
Click to expand...
Click to collapse
Thanks for feedback .. no good news.
Sorry, I can not reproduce your problem.
Tested the system_vendor udapte.zip several times on my Pixel C.
No problem so far ...
You flashed the complete .zip via TWRP ?
What was your previous firmware level ?
Did you upgrade bootloader as well ?
What kernel do you have in use ?
EDIT:
Which Version of TWRP ?
Pls use xceed version for now.
The NYChitman1 version seems to have problems.
http://forum.xda-developers.com/pix...ock-bootloader-install-custom-t3307183/page14
followmsi said:
The NYChitman1 version seems to have problems.
http://forum.xda-developers.com/pix...ock-bootloader-install-custom-t3307183/page14
Click to expand...
Click to collapse
Only the latest one I released because I didn't put the proper flag in place -- the previous one works perfectly fine as indicated by myself and others.
Also, I find it interesting that you've created a development thread while you aren't actually developing anything.
Thanks for the update.
Still like to test your version.
Ok, the version from here is recommended .. http://forum.xda-developers.com/showpost.php?p=66557703&postcount=101
In the one from yesterday you fixed vendor image flashing via TWRP.. sounds nice.
When do you re-release this version ?
You may open a new thread for your TWRP releases ..
Thanks a lot for all your work!
New kernel released ... pls test
Update.ZIP_Nexus_Ryu_chromeos-3.18-kernel_v160519_followmsi.zip
There are lot of changes in compare to latest stock kernel .. https://chromium.googlesource.com/chromiumos/third_party/kernel/+log/chromeos-3.18/
Cheers
followmsi said:
New kernel released ... pls test
Update.ZIP_Nexus_Ryu_chromeos-3.18-kernel_v160519_followmsi.zip
There are lot of changes in compare to latest stock kernel .. https://chromium.googlesource.com/chromiumos/third_party/kernel/+log/chromeos-3.18/
Cheers
Click to expand...
Click to collapse
Many thanks for your work The kernel is for stock / AOSP roms? Running Dirty Unicorns and I think it´s not compatible :/
redukt said:
Many thanks for your work The kernel is for stock / AOSP roms? Running Dirty Unicorns and I think it´s not compatible :/
Click to expand...
Click to collapse
There are so many fixes since MXC89F kernel .. which was commited on Arpil 5th .. even it's the May release. Ok, was tagged on May, 4th.
https://chromium.googlesource.com/chromiumos/third_party/kernel/+/MXC89F
Last changes .. 6 weeks ..
https://chromium.googlesource.com/c...+log/9c052efad6e651d33334b84c8f1feaf8945e8e7a
I am running the new kernel on latest M stock release .. MXC89F.
No problems so far .. really stable.
The kernel should run on all stock M releases for PixelC.
Used latest stock ramdisk to build the kernel .. maybe the dirty unicorn ramdisk is different.
But the actual kernel should work as well ..it's a dragon kernel.
Repack DU ramdisk with my kernel and it will most probably work as well .. maybe even with the same ramdisk, without repacking. Needs to be tested ..
Give it a try .. it will not hurt.
Cheers
Any chance of getting a port of multirom? Let me know if there's any interest, and what I can do to help.
nic. said:
Any chance of getting a port of multirom? Let me know if there's any interest, and what I can do to help.
Click to expand...
Click to collapse
Sounds good .. have it running on Nexus 7 flo .. really like it.
MultiROM for PixelC would be really nice .. the chromeos is part is maybe tricky.
But like SuperSU .. automatic chromeos image unpack and repack, if required at all.
It seems Tassadar is not around anymore .. not sure if he will come back.
Some new guys have taken over .. they just made a new update release for TWRP and MultiROM..
Works fine
Are you able to port it to Pixel C ?
Thanks
followmsi said:
Give it a try .. it will not hurt.
Cheers
Click to expand...
Click to collapse
Bootloop
Restored my Nandroid and device is up
DU is CM based and unfortunately incompatible :/
followmsi said:
Sounds good .. have it running on Nexus 7 flo .. really like it.
MultiROM for PixelC would be really nice .. the chromeos is part is maybe tricky.
But like SuperSU .. automatic chromeos image unpack and repack, if required at all.
It seems Tassadar is not around anymore .. not sure if he will come back.
Some new guys have taken over .. they just made a new update release for TWRP and MultiROM..
Works fine
Are you able to port it to Pixel C ?
Thanks
Click to expand...
Click to collapse
I am wanting to dig into the Linux4Tegra side of this, and was hoping that some of the work for multirom could or had been done. I don't know anything about the chromeos bootloader. Looking into it, Chrubuntu might be a viable path for this. I feel like this device is gimped with Android.
redukt said:
Bootloop
Restored my Nandroid and device is up
DU is CM based and unfortunately incompatible :/
Click to expand...
Click to collapse
Thanks for the test ..
Restore DU boot.img and you are back to normal ..yep
I will unpack the DU boot.img and have short look .. just to check it.
nic. said:
I am wanting to dig into the Linux4Tegra side of this, and was hoping that some of the work for multirom could or had been done. I don't know anything about the chromeos bootloader. Looking into it, Chrubuntu might be a viable path for this. I feel like this device is gimped with Android.
Click to expand...
Click to collapse
Sounds like a lot of work .. but if you have the right people involved ..
Maybe xceed or nychitman will know better how to "manage" the bootloader ..
As I know, until today, "nobody" knows how to remove the 30 second warning time beep sound ****.
Right now I am happy to "manage" the layer above .. boot.img and recovery.img
redukt said:
DU is CM based and unfortunately incompatible :/
Click to expand...
Click to collapse
What are you talking about? We're not CM based.
followmsi said:
There are so many fixes since MXC89F kernel .. which was commited on Arpil 5th .. even it's the May release. Ok, was tagged on May, 4th.
https://chromium.googlesource.com/chromiumos/third_party/kernel/+/MXC89F
Last changes .. 6 weeks ..
https://chromium.googlesource.com/c...+log/9c052efad6e651d33334b84c8f1feaf8945e8e7a
I am running the new kernel on latest M stock release .. MXC89F.
No problems so far .. really stable.
The kernel should run on all stock M releases for PixelC.
Used latest stock ramdisk to build the kernel .. maybe the dirty unicorn ramdisk is different.
But the actual kernel should work as well ..it's a dragon kernel.
Repack DU ramdisk with my kernel and it will most probably work as well .. maybe even with the same ramdisk, without repacking. Needs to be tested ..
Give it a try .. it will not hurt.
Cheers
Click to expand...
Click to collapse
I had been keeping up with the chromeos-3.18 branch upstream, but I realized that a majority of the commits do not even apply to Dragon -- so it wasn't all that worthwhile to have in the majority of cases. With that said, I decided to go back to the MXC89F source until the next batch of source is released in June.
NYCHitman1 said:
What are you talking about? We're not CM based.
Click to expand...
Click to collapse
Well my bad...DU is AOSP based. I get mixed up with all several custom roms on my devices :silly:
NYCHitman1 said:
What are you talking about? We're not CM based.
I had been keeping up with the chromeos-3.18 branch upstream, but I realized that a majority of the commits do not even apply to Dragon -- so it wasn't all that worthwhile to have in the majority of cases. With that said, I decided to go back to the MXC89F source until the next batch of source is released in June.
Click to expand...
Click to collapse
You are right .. there also many fixed not relevant for dragon at all.
But it´s a fresh kernel .. upstream
I was so free .. and made a new DU kernel now .. used your ramdisk and my fresh chomeos-3.18 build from today.
@redukt, please test this kernel again .. on DU
Update.ZIP_PixelC_Ryu_chromeos-3.18-kernel_DU_v160519_followmsi.zip
https://onedrive.live.com/?authkey=!ABLdSOKIeXgg_XU&id=479EB75F5DD9DA59!117&cid=479EB75F5DD9DA59
Thanks .. still bootloop ?
redukt said:
Well my bad...DU is AOSP based. I get mixed up with all several custom roms on my devices :silly:
Click to expand...
Click to collapse
Please test this one ... Update.ZIP_PixelC_Ryu_chromeos-3.18-kernel_DU_v160519_followmsi.zip
https://onedrive.live.com/?authkey=!ABLdSOKIeXgg_XU&id=479EB75F5DD9DA59!117&cid=479EB75F5DD9DA59
Thanks !
followmsi said:
Please test this one ... Update.ZIP_PixelC_Ryu_chromeos-3.18-kernel_DU_v160519_followmsi.zip
https://onedrive.live.com/?authkey=!ABLdSOKIeXgg_XU&id=479EB75F5DD9DA59!117&cid=479EB75F5DD9DA59
Thanks !
Click to expand...
Click to collapse
!!! Device is booting!!!
Meh... forget to install SU lol.... Will have a look in kernel auditor
Many thanks
edit: I have to flash the latest supersu (beta v2.74-2) to get root again. Will test your kernel til the next DU weeklie as dd. I´ll have a look at the frequency overview and kept stock settings / governor. System runs fluid with no hickups or so. Great work!
Looks also very interessting ..stabilize-smaug-7897.B-chromeos-3.18
https://chromium.googlesource.com/c...kernel/+/stabilize-smaug-7897.B-chromeos-3.18
Will see .. today they released the 3rd N preview ... maybe it´s time for some N testing.
As I understand correct .. using N ramdisk and the same kernel from today should work on N as well.
We will see ...
Cheers
Hi guys, I have been looking at / working on allowing the 32GB variant Mi5 to run at the full speeds (2.15/1.6/624) on the msm8996 SOC as its 64/128GB brothers.
I was able to patch the kernel frequency table to run my 64GB at the reduced frequency of the 32GB but cannot test this in reverse as I don't have a 32GB device
I am hoping a 32GB user can test a boot.img for me. Can fastboot boot/flash it, or flash via TWRP (install -> install image -> select img) and just backup your existing img.
I do suspect it will not boot or work (it does boot/otherwise work on my 64GB though, but there's a few more code differences I haven't changed), but it'd be nice to know
boot_rev1.img this revision does not work, yet
boot_rev2.img a few users report that this one works.
What I would like to know is overall stability for 32GB users, if they could test their phone in games etc as voltage tables may need more adjustment
Edit: please note this kernel is for CM13/AOSP based ROM. please don't flash it over MIUI8
Project moved to http://forum.xda-developers.com/mi-...d-32gb-mi5-t3482455/post69177557#post69177557
Source here: https://github.com/harrynowl/falconkernel-gemini
awesome. did you post in the miui forum?
rochee94 said:
awesome. did you post in the miui forum?
Click to expand...
Click to collapse
Not yet, hoping to find feedback here first
As you are using CM kernel as base, you must inform that this will only work on devices running MIUI 7.x firmware and a CM based ROM.
On top of that, don't forget to publish the link to the kernel source in order to follow XDA's GPL regulations. I know you have the source in Github.
Could you build a kernel to overclock the 64 &128GB versions too? That would be great.
Sorry out of topic but, does snapdragon 820 have mpdecision or not? (by default)
razibrizal said:
Sorry out of topic but, does snapdragon 820 have mpdecision or not? (by default)
Click to expand...
Click to collapse
No: https://source.codeaurora.org/quic/...init.qcom.post_boot.sh?h=LA.HB.1.3.1.c1#n1729.
GalaxyS4_Lollipop said:
Could you build a kernel to overclock the 64 &128GB versions too? That would be great.
Click to expand...
Click to collapse
First focus is on giving 32GB users the full power of the hardware they paid for
I've tried it and it fails to boot
It goes into the bootanimation but after like one second it freeze and reboot
And when it happen, it kinda overheat
Edit
Using AOKP and Cm13 same result
sys01 said:
I've tried it and it fails to boot
It goes into the bootanimation but after like one second it freeze and reboot
And when it happen, it kinda overheat
Click to expand...
Click to collapse
Thanks for info, I think I know why this is..
Harrynowl said:
First focus is on giving 32GB users the full power of the hardware they paid for
Click to expand...
Click to collapse
That's cool.
Anyway, if it is any method to power up the beast to the SnapDragon821 power, I wait for it. (on 64gb)
Good luck and thanks for your support.
4 days ago bought a 32gb Mi5 -> upgraded to latest MIUI 8.0.12
Unlocked bootloader -> Flashed Ressurection Remix's latest ROM via TWRP
Looking forward in testing this !
Alex4455 said:
4 days ago bought a 32gb Mi5 -> upgraded to latest MIUI 8.0.12
Unlocked bootloader -> Flashed Ressurection Remix's latest ROM via TWRP
Looking forward in testing this !
Click to expand...
Click to collapse
Well, please don't use the one I posted originally as it won't work. But I'm looking at a fix now :good:
Looking forward towards the fix.
Many people will be happy if this comes out with no issues !
Harrynowl said:
Well, please don't use the one I posted originally as it won't work. But I'm looking at a fix now :good:
Click to expand...
Click to collapse
thank u for the initiative will wait for the next build :good:
pls after making it work, build one for MIUI 8 if it's possible.
catalizgar said:
pls after making it work, build one for MIUI 8 if it's possible.
Click to expand...
Click to collapse
Miui8 firmware blobs are not currently compatible with the kernel sources we have.
A member got a kernel booting on miui8 but stability was a big problem.
First of all, thank you. I think all 32GB users will thank you for that.
Obviously when it's done, it will consume more power, and would have less SoT.
Can you make predictions of what it will take when optimized and fully working?
After changing my ROM to AOSP based I will use this. thanks for your hard working. really appreciated
I added a new boot.img to the OP with some updates to the source. Hopefully this one has more luck. Specifically changes to the voltage table(s).
Please test 32GB users