Related
I assume no liability for any damage to your phone!
You can use my complete rom or parts of it to make your own rom. Only post a small thanks on your owm thread.
Big thanks to: iaio72, spear1403, tgzw, Quarx, Simondo, ein1002, santimanuel, epsylon3 and ...
Language support for final 4.0 --> da, de, el, en, es, fi, fr, hr, hu, id, it, nb, nl, pl, pt, pt-BR, sv, th, ti, vi, zh_CH, zh_TW + arabic* and ru*
MS2Ginger 4.0
- based on 4.5.2-109-DHT-25 - Android 2.3.6 blur lite rom
- support for Defy, Defy+, Milestone 2, Bravo
- MS2Ginger4 can used as 2nd-system (tested on GB kernel only) !! system image file have to be >= 350 MB
- multibootmanger app by royale1223
- new bootmenu 1.1.8, CWM 5.0.5/6, multiboot big thanks to m11kkaa
- 2 way to boot the rom: normal and via 2nd-init (default)
- new reboot option form cm7.2 --> trigger: all keys at the same time vol- , vol+ , power (only tested on Defy)
- disabled blurhome rotation
- button backlight can set from 10-100 % via build.prop
- mmcfix from CM7
- CPU settings default safe 30/300 38/600 {48/800} 58/1000
- Camera will work on SOC and Bayer camera (green* and red)
- better camera support for bayer camera
- new 720p record* and play (Milestone 2 - Android 2.3.6 files)
- 720p record in m4v (h264 can be enabled via build.prop)
- Milestone 2 froyo camara app for SOC camera (fixed green strips)
- BlurHome from Milestone 2 - 2.3.6 (no widget updating bug)
- Battery min. 2-5 mA/h
- no wifi, apn, vpn bugs
- wifi channels 12, 13, 14 enabled
- bootmenu + cwm (vsel4/clk4 added) "did not work for SOC camera with greenbreadMOD"
- bootmenu with Stable, Custom, Stock-Recovery
- call record libs
- init.d support
- zipalign at boot
- ramscript
- backlight button fix for Defy with bayer camera (can removed via optional patch)
- deepsleep support
- max 10 finger multitouch
- support for farsi-arabic, hebrew (thanks to MMarv)
- CyanogenMod-settings not finished (alpha) - sorry, will try to fix it if i have more time.
(* via greenbreadmod for MS2Ginger4.0 and if you never flashed a Android 2.3.6 full sbf with bl7)
(* 720p record only for Bayer camera Defy and Milestone 2)
Download: ROM, BlurApps, Greenbread: (MS2Ginger4.0.zip, BlurApps-MS2Ginger4.zip,Milestone2patchMS2Ginger4.0withbaseband.zip )
Mirror 1 (google)
Mirror 2 (Dev-Host)
More information for BlurApps --> LINK
Soft-sim-unlock for MS2Ginger 4.0
based on Dabb guide: http://forum.xda-developers.com/showthread.php?t=1447950
Install View attachment 1129769 via CWM.
If you like MS2Ginger feel free to donate! Donations will be used to purchase a Defy with SOC camera (green lens)
Bugs and problems Reporting: http://code.google.com/p/ms2ginger/issues/list?can=1&q
- wifi bug like all 2.3.6 Motorola roms http://forum.xda-developers.com/showthread.php?t=1502817
(fixed, disable unlimited lease time for IP adress on your rooter)
Timeline:
19.06.12 - MS2Ginger 4.0 released
05.02.12 - MS2Ginger 3.0 released
Patch and updates for current release
none for MS2Ginger 4.0
HOW to INSTALL:
To begin:
Put the Nandroid backup folder MS2Ginger on the SDCard in this folder: /sdcard/clockworkmod/backup
Put "Milestone2patchMS2Ginger4.0withbaseband.zip" to the root of your sdcard
Put "Milestone2-GB-kernel-update.zip*" to the root of your sdcard (only if you have a froyo kernel)
-------------------------------------------------------
Root your Milestone 2 and Install BOOTMENU - CWM for Milestone 2
REBOOT
run 2nd-init (voldown on blue LED)
Go to "Recovery"
Go to "Latest or Stable or Custom Recovery"
"wipe data and cache"
Go to "Backup and Restore"
Go to "Advance Restore"
Select "MS2Ginger4.0"
"restore system"
Go to "install zip from sdcard"
select and install "Milestone2patchMS2Ginger4.0withbaseband.zip"
"Install "Milestone2-GB-kernel-update.zip" via install zip" (only if you have froyo kernel)
reboot
Booting up can take a while, max is about 10 minutes.
Team:
- walter79
- spear1403 (theming)
- santiemanuel
- Defy red lens tests spear1403
- Defy greens lens tests luciferps and Cristi_10
- Defy plus tests Dollo73 and free2live
- Milestone 2 tests Carolinux and Motodoido
Translation Team: (each language max. 4 person) View attachment 806172
Brazillian Portuguese: pupiloalpha , Cassinander , filipeazev
Bulgarian Auris 1.6 vvt-i
Farsi: MMarv
Hungarian: szabof1 , balika01
Polish: yackoo , szacuken , Antszej
Serbian: atomskasarma
Spanish espaciosalter20
Spanish-Argentina YaPeL
walter79 said:
Welcome.
The idea:
Defy is running Milestone 2 Android 2.2.2 ROMs. (DACH retail based)
So maybe we could make a working Gingerbread for Milestone 2.
- Defy has rooted only Android 2.3.4 /system nandroid-backup http://forum.xda-developers.com/showpost.php?p=14793844&postcount=8
- it is also possible to run 2nd-init on milestone 2
- read this (Defy 2nd-int + froyo kernel + Gingerbread-system working) http://forum.xda-developers.com/showthread.php?t=1130656
- i think it could be possible to get gingerbread working on milestone 2
- basband should be change with original milestone 2 files
- /lib/hw/ from milestone 2 needed
- kernel from milestone 2
- fix for camera
Whats your opinion?
Click to expand...
Click to collapse
I'm sorry to kill your optimism, but we are unable to flash onto /system for now. Big files at least..
Did clockworkmod not work?
http://www.gamaze.com/blog/2010/11/installing-clockworkmod-recovery-on-motorola-milestone-2/
walter79 said:
Did clockworkmod not work?
http://www.gamaze.com/blog/2010/11/installing-clockworkmod-recovery-on-motorola-milestone-2/
Click to expand...
Click to collapse
It works, it's just that.. it doesn't flash big update.zip's, which overwrite the whole system. Problem could be caused by us not having a recovery build specifically for our devices.
But nandroid backup works, or?
Mikevhl said:
It works, it's just that.. it doesn't flash big update.zip's, which overwrite the whole system. Problem could be caused by us not having a recovery build specifically for our devices.
Click to expand...
Click to collapse
So build one: https://github.com/koush/Droid2Bootstrap
Edit: Check this out "use an update.zip that waits a bit longer for the system to settle then unmounts tmp and remounts /"
dangpzanco said:
So build one: https://github.com/koush/Droid2Bootstrap
Edit: Check this out "use an update.zip that waits a bit longer for the system to settle then unmounts tmp and remounts /"
Click to expand...
Click to collapse
Nice. Will try to get a clean nandroid backup of my /system, unyaffs it and create my own update.zip formatting /system and using this trick to see if I it works.
r2beta0 said:
Nice. Will try to get a clean nandroid backup of my /system, unyaffs it and create my own update.zip formatting /system and using this trick to see if I it works.
Click to expand...
Click to collapse
Lol, I think it's a lucky shot finding this...
Sent from my Milestone 2 XDA App
Hey, what happens if I flash Defy's SBF on M2? Besides other problems is it possible to enter bootloader mode and flash another sbf, or it will corrupt the device?
I didn't, but I saw it was a prerequisite for the Rom posted here, so if there's theorical trouble, then I won't try.
Also I found out that Droid 2 bootloader image has the same first and last bytes as M2 (but different check sums), and Defy's is little different.
dangpzanco said:
Hey, what happens if I flash Defy's SBF on M2? Besides other problems is it possible to enter bootloader mode and flash another sbf, or it will corrupt the device?
I didn't, but I saw it was a prerequisite for the Rom posted here, so if there's theorical trouble, then I won't try.
Also I found out that Droid 2 bootloader image has the same first and last bytes as M2 (but different check sums), and Defy's is little different.
Click to expand...
Click to collapse
Did not work.
dangpzanco said:
Hey, what happens if I flash Defy's SBF on M2? Besides other problems is it possible to enter bootloader mode and flash another sbf, or it will corrupt the device?
I didn't, but I saw it was a prerequisite for the Rom posted here, so if there's theorical trouble, then I won't try.
Also I found out that Droid 2 bootloader image has the same first and last bytes as M2 (but different check sums), and Defy's is little different.
Click to expand...
Click to collapse
I don't think RSD Lite will allow you to do that. Even if it does, I don't think it's a good idea since it may replace your bootloader potentially bricking your phone.
r2beta0 said:
I don't think RSD Lite will allow you to do that. Even if it does, I don't think it's a good idea since it may replace your bootloader potentially bricking your phone.
Click to expand...
Click to collapse
Yeah, I thought about that bricking stuff... (maybe linux/mac sbf_flash would flash it)
Did the "update.zip that unmount tmp and remounts /" work?
Sent from my Milestone 2 XDA App
dangpzanco said:
Yeah, I thought about that bricking stuff... (maybe linux/mac sbf_flash would flash it)
Did the "update.zip that unmount tmp and remounts /" work?
Sent from my Milestone 2 XDA App
Click to expand...
Click to collapse
Not yet, still working on it on my free time. Unfortunately 'free time' is a scarce resource
Maybe if you point the update-script to another script:
Code:
#!/system/bin/sh
#This was in Apex 1.4.1 /etc/init.rc folder (wait for system to settle?)
sleep 24s
#unmount tmp
busybox umount /tmp
#remount /
busybox mount -o remount -rw -t rootfs /
Will it work?
dangpzanco said:
Maybe if you point the update-script to another script:
Code:
#!/system/bin/sh
#This was in Apex 1.4.1 /etc/init.rc folder (wait for system to settle?)
sleep 24s
#unmount tmp
busybox umount /tmp
#remount /
busybox mount -o remount -rw -t rootfs /
Will it work?
Click to expand...
Click to collapse
First attempt failed. I had some problems formatting /system partition. The only way I could get it to format was through Recovery Menu. Maybe the problem is reformatting the system. The command mount("MTD", "system", "/system") also doesn't work. I'm starting to think that Motorola messed with the file system. Will try again using delete_recursive instead of format.
There is any d2 / d2g rom based on GB?
With all the problems developers were facing lately, I guess motorola add the same "weird" protections in m2 2.2 to GB d2 rom
It's very hard to make a custom ROM with all these protections and without knowing what happens when phone tries to boot. All we get is the damned M logo and no info at all. Without some output from the phone everything we can do is guessing.
I was reading that new fission roms based on GB (for d2), were facing the "motorola logo" stuck problem. That´s why the developer gave up.
But, like you say, there is no way to know if it´s the same problem.
I have no knowledge at all on android platform. I´d stuck on winmo 6.
If you need some testing, I can try, there is nothing more I can offer
Second attempt failed. It may have nothing to do with format or system file. I'm really clueless Jut the M logo again. Maybe the /system protection checks for some date/time on some random file? Will reflash with RSD and 'touch' everything in /system to see if the damned thing still boots.
Did you compare the hash of apex files to the moto original files?
This rom story has been moved development thread.
Follow link:
http://forum.xda-developers.com/showthread.php?t=1227835
--------------------------------------------------------------------
Ported CM7-RC12 to XT720(Korean Motoroi)!!
[Outline]
CM-7.1.0-RC12 customized.(default base is nadlabak's Milestone ROM)
Model : Korean Motoroi XT720
Based Rom : STSKT_N_79.33.50R (Android 2.2.1)
[Work]
3G fixed.
WiFi fixed.
Battery fixed.
USB fixed.
KeyPad fixed.
Sensor's fixed.
Touch fixed.
App2sd fixed.(ext3)
Etc.
[Not Work]
USB Connect icon(Status-bar)
overclock
[Install]
1) RSD for STSKT_N_79.33.50R (Android 2.2.1)
2) root
3) OpenRecovery install
4) Update-pack download and copy (/sdcard/OpenRecovery/updates)
5) OpenRecovery boot
6) Wipe Data / Factory reset
7) Apply Update (CM-7.1.0-RC12-XT720-J.Y.Daddy-Beta1.zip)
8) reboot
[Mirror]
http://www.multiupload.com/8Z6Y5O1HSW
[2011.08.15]
And MIUI Gingerbread 1.8.12 version, we also succeeded in porting.
Yet there remains a minor bug, but can be used.
If you succeed in porting MilestoneXT720 CM7, of course, a port will become MIUI Gingerbread!
Good Luck!!
Alpha-1 Version Download :
(MIUI-1.8.12-KO-XT720-J.Y.Daddy-alpha1.zip) http://www.multiupload.com/NOJWWVYF00
Thanks for your help.
Quick question: Is this compiled from source for XT720 or is it a binary port of nadlabak's A853 (or other) ROM?
We did not compiled.
Through a variety of modules that are compatible with the process of extraction is created by the ROM.
Of course, the default base is nadlabak's A853 ROM.
For European xt720 we get bootloop...
I believe you must 1st sbf to Korean 2.2 for it to boot. Just like the one we already have. And Mioz what about the 5 point touch module for milestone? You think that might make our touch work.
fjfalcon said:
For European xt720 we get bootloop...
Click to expand...
Click to collapse
When you update from a different ROM will be bootloop.
Based rom is Korea Froyo Rom(STSKT_N_79.33.50R).
easye420 said:
I believe you must 1st sbf to Korean 2.2 for it to boot. Just like the one we already have. And Mioz what about the 5 point touch module for milestone? You think that might make our touch work.
Click to expand...
Click to collapse
We are the "5 point touch module" tested. Is not active.
However, we have "3 point touch module" has to.
It is active only in Korea Motoroi XT720.
J.Y.Daddy said:
We are the "5 point touch module" tested. Is not active.
However, we have "3 point touch module" has to.
It is active only in Korea Motoroi XT720.
Click to expand...
Click to collapse
Does the 3 point touch module enable touch on cm7 on motoroi?
easye420 said:
Does the 3 point touch module enable touch on cm7 on motoroi?
Click to expand...
Click to collapse
CM7 touch does not work. So, can't test yet.
Possibilities are very high.
hello i'm members of androiders,too
I heard usb problem fixed by vista of androiders.
can you find vista's thread that fixed usb problems??
i want to this project will be successful.
(in korean)낭만로드님 제가 그전에 비스타님의 글로 터치제외한 모든버그를 고쳤다는 글을 읽은것 같습니다. 한번 비스타님의 글을 뒤져보심이......
Sent from my A853 using XDA App
siwkim said:
hello i'm members of androiders,too
I heard usb problem fixed by vista of androiders.
can you find vista's thread that fixed usb problems??
i want to this project will be successful.
(in korean)낭만로드님 제가 그전에 비스타님의 글로 터치제외한 모든버그를 고쳤다는 글을 읽은것 같습니다. 한번 비스타님의 글을 뒤져보심이......
Sent from my A853 using XDA App
Click to expand...
Click to collapse
Oh! nice to meet you.
This work will progress up with Vista's.
And some members are doing together.
(in korean)
여기서 한국사람을 뵈니 색다르네요. 한글도 ㅎㅎ;
지금 말씀하신게 바로 저와 비스타님이 진행해서 현재 여기까지 온거에요.
즉, 저와 비스타님 뿐만 아니라 로이포럼에서 연구소 쓰레드로 진행해 왔는데
터치문제는 지금 난관에 부딪혀서 이렇게 XDA에 올리게 된 것입니다.
have you tested XT720(milestone)'s original touch module?
i think that module make topuch work.
(in Korean)낭만로드님 답변을 해주시니 감사할 따름입니다.
혹시 유럽발 XT720의 기본 터치모듈이 작동하는지 확인해 보셨나요?? 아마도 모듈 문제 같은데 한번 테스트 해보심이...
Sent from my A853 using XDA App
siwkim said:
have you tested XT720(milestone)'s original touch module?
i think that module make topuch work.
(in Korean)낭만로드님 답변을 해주시니 감사할 따름입니다.
혹시 유럽발 XT720의 기본 터치모듈이 작동하는지 확인해 보셨나요?? 아마도 모듈 문제 같은데 한번 테스트 해보심이...
Sent from my A853 using XDA App
Click to expand...
Click to collapse
hmm.. good idea. i will try.
(in korean)
그건 미쳐 생각 못했네요 ㅎㅎ; 어차피 이것 저것 다 시도해보는터라 한번 해보아야겠네요.
아이디어 고맙습니다^^;
------------------------------------------------------------------------------
[Test Result]
Are the same. touch doesnt work. (in korean: 달라지는게 전혀 없네요 --)
Proprietary files comparison
Attached is a zip that contains a text file with md5 sums and two csv spreadsheets that can be used to compare some relevant ROMs. The ROMs included in the comparisons are:
1. STSKT_N_79.33.50R (Motorola's 2.2.1 for Motoroi XT720)
2. update-cm-7.1.0-RC12-11.07.14-Milestone-signed (Nadlabak's CM7 for A853)
3. android_vendor_motorola_umts_sholes (nadlabak's github)
4. CM-7.1.0-RC12-XT720-J.Y.Daddy-alpha-1 (port of #2/#3 for Motoroi XT720)
5. SC_2.2_XT720_Beta3a (Dexter's/WanHu's port of #1 for Milestone XT720)
The spreadsheet is limited to only the files included in android_vendor_motorola_umts_sholes/proprietary. These are the files that are copied verbatim from the A853 Motorola ROM and used by nadlabak's build. Some files needed on XT720 don't have the same name, but they are not included (i.e. cameraCalFileDef5M.bin and cameraCalFileDef8M.bin is used on A853 but cameraCalFileDef.bin used instead on both XT720s) The columns of the spreadsheet are:
1. file name
2. md5sum
3. path hint
Using the md5sum you can see if the file included in each one is different from the others. For example:
Code:
pvrsrvinit 579227b3cbaca8a67f24075d5b51d2d8 CM-7.1.0-RC12-XT720-J.Y.Daddy-alpha-1/system/bin/pvrsrvinit
pvrsrvinit 3c3d144ae96061476218d283e393c54e SC_2.2_XT720_Beta3a/system/bin/pvrsrvinit
pvrsrvinit 579227b3cbaca8a67f24075d5b51d2d8 STSKT_N_79.33.50R/system/bin/pvrsrvinit
pvrsrvinit 579227b3cbaca8a67f24075d5b51d2d8 update-cm-7.1.0-RC12-11.07.14-Milestone-signed/system/bin/pvrsrvinit
pvrsrvinit 579227b3cbaca8a67f24075d5b51d2d8 android_vendor_motorola_umts_sholes/proprietary/bin/pvrsrvinit
This indicates that the file /system/bin/pvrsrvinit is identical on STSKT_N_79.33.50R, update-cm-7.1.0-RC12-11.07.14-Milestone-signed, CM-7.1.0-RC12-XT720-J.Y.Daddy-alpha-1 because they all have md5sum=579227b3cbaca8a67f24075d5b51d2d8. But on Milestone XT720 (SC_2.2_XT720_Beta3a) the md5sum=3c3d144ae96061476218d283e393c54e. This means that this file was changed by Dexter/WanHu from the version in Motoroi XT720 version in order to get the ROM to work on Milestone XT720.
So, one way to go about this is to try and see if there are some files that might need to be updated. For example, files that are identical in STSKT_N_79.33.50R and SC_2.2_XT720_Beta3a very likely need to be identical in both CM-7.1.0-RC12-XT720-J.Y.Daddy-alpha-1 and a Milestone XT720 port. So one thing to do is to study these lists and identify files that seem suspicious.
The comparisons.md5 file has md5sums for all files in the ROMs (i.e. not just the ones in nadlabak's github).
Edit: added a shortlist.csv to the zip that removes files that are identical in all five ROMs
Or we can just debug the ROM via adb and see if any useful error-messages regarding the touch screen come up.
Vistaus said:
Or we can just debug the ROM via adb and see if any useful error-messages regarding the touch screen come up.
Click to expand...
Click to collapse
I wasn't able to find anything useful using logcat or dmesg when I was trying to binary port stlifey's, so I wasn't expecting to see anything useful here, either. The touch is quite strange on our phone. For example if you use a different bootanim binary, *poof* touch doesn't work (I was trying to get the live logcat/live dmesg working, chainfire seemed to think that was strange, too). Or if you delete /system/media/bootanim.zip to try and use the default android animation, *poof* no touch. Even in those cases, I couldn't find anything in the logs that was different.
Some questions for Motoroi owners
Can someone with a Motoroi XT720 post the output of "getevent" run over ADB?
Is there an OpenRecovery for Motoroi XT720 that works when holding volume-up at boot to enter OpenRecovery?
It looks like when I apply the 50R sbf on Milestone XT720, then the volume keys are on event3 (check this by "adb shell getevent /dev/input/event3" and pressing volume up/down):
Code:
/dev/input/event3: 0001 0073 00000001
/dev/input/event3: 0001 0073 00000000
But on Milestone XT720 running a Milestone XT720 kernel, the volume up keys are on event4 (check this by "adb shell getevent /dev/input/event4" and pressing volume up/down)::
Code:
/dev/input/event4: 0001 0073 00000001
/dev/input/event4: 0001 0073 00000000
This is what I get from getevent on Milestone XT720 when running 50R:
Code:
add device 1: /dev/input/event6
name: "sensors"
add device 2: /dev/input/event5
name: "lm3530_led"
add device 3: /dev/input/event4
name: "proximity"
add device 4: /dev/input/event3
name: "sholes-keypad"
add device 5: /dev/input/event2
name: "qtouch-touchscreen"
add device 6: /dev/input/event1
name: "cpcap-key"
add device 7: /dev/input/event0
name: "compass"
And this is what I get running a normal Milestone XT720 ROM:
Code:
add device 1: /dev/input/event7
name: "sensors"
add device 2: /dev/input/event6
name: "lm3530_led"
add device 3: /dev/input/event5
name: "proximity"
add device 4: /dev/input/event4
name: "sholes-keypad"
add device 5: /dev/input/event3
name: "qtouch-touchscreen"
add device 6: /dev/input/event2
name: "cpcap-key"
add device 7: /dev/input/event1
name: "accelerometer"
add device 8: /dev/input/event0
name: "compass"
So, it seems that the accelerometer device is missing when booting the Korean kernel on Milestone XT720 and this shifts the keyboard to event3? Or does accelerometer not exist on Motoroi?
Edit: Here's a modified sh binary for Milestone XT720 users running 50R. I'm pretty sure this will not work on Motoroi XT720 based on looking at the key mapping file in the stock 50R. My theory is this is a problem for Motoroi kernels on Milestone XT720, only. I just used a hex editor to change /dev/input/event4 to /dev/input/event3. This works on my Milestone XT720. Replace /system/bin/sh and /sdcard/OpenRecovery/orbootstrap/str/sh with this copy, but be very careful--if you flash a non-korean sbf the volume up will stop working. Eventually it would be better to have a single binary that checks both event4 and event3, but I haven't figure out how to recompile orbootstrap yet. And you should probably be religious about reapplying this after applying updates etc incase it gets replaced with a bad version, etc.
Mioze7Ae said:
Can someone with a Motoroi XT720 post the output of "getevent" run over ADB?
Is there an OpenRecovery for Motoroi XT720 that works when holding volume-up at boot to enter OpenRecovery?
Click to expand...
Click to collapse
Thank you very much for your help.
The same OpenRecovery Motoroi is being used.
Volume-up key at boot to enter the OpenRecovery available.
And acceleration sensors are also Motoroi.
I do not know why that does not appear in MilestoneXT720.
However, CM-7.1.0-RC12-XT720-JYDaddy-alpha-1 does not appear in.
[Question]
"sh binary" input device to control possible?
CM-7.1.0-RC12-XT720-JYDaddy-alpha-1 after install
other input devices has confirmed that the event occurred.
"/ dev/input/event3". In other words, the only thing can't see the qtouch-screen events.
Thanks.
I don't think the event renumbering is much of a problem for Android on MsXT720, because the 50R works fine (except for the missing accelerometer) and touch works there. It was just that the hijacked sh from open recovery is uses /dev/input/event4 to read the keys. Since they are not shifted on Motoroi and touch does not work, and Milestone is shifted and touch does work on stock 50R, I think the problem must be somewhere else.
Did you try using /system/bin/bootanimation and/or /system/bin/linker from 50R? Dexter changed those for MsXT720 when he ported 50R to the MsXT720. That may give clues about what files are kernel-specific (since this ROM wad compiled for A853).
Maybe I can try breaking the WanHu ROM on MsXT720 to see which files stop the touchscreen.
Sent from my XT720 using XDA Premium App
Mioze7Ae said:
Did you try using /system/bin/bootanimation and/or /system/bin/linker from 50R? Dexter changed those for MsXT720 when he ported 50R to the MsXT720. That may give clues about what files are kernel-specific (since this ROM wad compiled for A853).
Maybe I can try breaking the WanHu ROM on MsXT720 to see which files stop the touchscreen.
Sent from my XT720 using XDA Premium App
Click to expand...
Click to collapse
bootanimation change not try.
change bootanimation!! i will try it, tonight.
and then will shared result.
Hi,
currently we are try to port the unofficial CyanogenMod 10.1 build for Huawei G510 Daytona to the Huawei Ascend G330 U8825-1.
The rom runs out of the box but there are some things that won't work at this time. The baddest Problem is that telephony won't work.
There is no sim and no imei. After some search we found out that we have to use the libcm.so from the stock Firmware of Ascend G330 and now we ould connect to sim card and we have imei back and data Connection is working fully. But it's not possible to call someone, because we can't hear anything and the microphone doesn't work.
Even this tutorial: http://forum.xda-developers.com/showthread.php?t=1948560 isn't successful.
Could someone help?
Thanks
StefanV3
Hey,do you can send this cm10 via pm to me? Thanks,and,try this directly from g510 cm10.1 snapshot [quote name='joestone' timestamp='1376772208' post='2149556']Spanish guys : The troll is on again with the photoshopped screen as I read it on htcmania. You can check my previous jobs here on modaco or on xda (gt540, zte smarttab, g300).
You can download my build http://goo.im/devs/joestone/g510/cm-10.1-20130817-UNOFFICIAL-G510.zip it has a lot of non working parts , but ril and nfc are working.
Developers : This is my huaweiril2 whic was used to fix the ril. It logs a lot and not needed in production build.http://goo.im/devs/joestone/g510/HuaweiRIL2.java
How to fix ril .
First be sure you have ro.telephony.ril.v3=qcomdsds in your system prop . This is needed as our ril libs are made for multisim and have to send sub1 as main ril socket.
Qualcomsharedril is not suitable for our device as it has mQANElements=5 by default and we need 4. The main RIL has 4 as default.It is used by responseOperatorInfos(Parcel p) The second thing why it is not suitable it uses too many elements in responseIccCardStatus(Parcel p) by default (no sim pin request , invalid apptype in radiolog) . It can be switched off by using skippinpukcount oldril feature , but main RIL is fine with this too by default. Datacall routines are also fine in main ril.
The only reason why we should need a custom ril is to fix the shown baseband version. Of course if no other bug is found .
The non working state drops first this in radiolog :
W/RILC( 144): RIL_onUnsolicitedSendResponse called before RIL_register
E/RILD( 144): isMultiSimEnabled: prop_val = 0 enabled = 0
D/RILC( 144): QCRIL_RPC
D/RILC( 144): QCRIL_RPC
E/RILC( 144): RIL_register: RIL version 7
While with the working version contains :
E/RILD( 186): **RIL Daemon Started**
E/RILD( 186): **RILd param count=1**
E/RILD( 186): RIL_Init argc = 5 client = 0
E/RILD( 186): isMultiSimEnabled: prop_val = 0 enabled = 0
E/RILC( 186): RIL_register: RIL version 7
E/RILC( 186): s_registerCalled = 1, s_started = 1, RIL_getMaxNumClients = 1
No custom ril needed in framework , we can use the main ril.java.
Copy the ril libfiles from stock rom. I used these files from stock VF rom : Copy libqcci_legacy.so also as it used by atserver.
check your init files to have the following services:
in on boot section
pcm-bridge is not needed as pcm-bridge does not exist in stock rom (at least in my stock rom)
service atserver /system/bin/atserver
class core
oneshot
service netmgrd /system/bin/netmgrd
class main
service bridgemgrd /system/bin/bridgemgrd
class main
user radio
group radio
disabled
service port-bridge /system/bin/port-bridge /dev/smd0 /dev/ttyGS0
class main
user system
group system inet
disabled
service qmiproxy /system/bin/qmiproxy
class main
user radio
group radio
disabled
service qmuxd /system/bin/qmuxd
class main
user radio
group radio
In post-fs section
service ril-qmi /system/bin/sh /init.qcom.ril.path.sh
class main
user root
oneshot
in on emmc-fs section (you need BOARD_WANTS_EMMC_BOOT := true in boardconfig)
write /sys/module/block2mtd/parameters/block2mtd /dev/block/mmcblk0p14
on property:sys.radio.shutdown=true
exec /system/bin/sh /system/etc/init.qcom.efs.sync.sh
sorry for the long post.[/quote]
I have a friend devoloper,if you send to me link to cm10 i can help you,and.....are you embeddev on github
Tapatalkato via il mio gelato sfigato biscotto
Hi,
isee that you registered in german www.android-hilfe.de and joined the discussion at Huawei G330 Custom Rom section.
The CM10.1 rom for Huawei G510/Y300 is from this post at spanish htcmania Forum:
http://www.htcmania.com/showthread.php?t=613560
First of all we don't build CyanogenMod 10.1 from source for G330. We tried to modify the rom built from HERRERO7 for G510 to get it work on our G330 because the Rom does run on our G330 as it is but with the so called problems.. Our first succes was to get 3G work with the libcm.so from stock firmware. Telephoy was also prossible but without a working speaker and withou working mic. Maybe the new build of joestone is Fixing the most things...
But i am a Little bit confused about your words because this is CyanogenMod 10.1 and not CyanogenMod 10. And i am confused about the words of user joestone because of dual sim. Oour Huawei Ascend G330 (8825-1) isn't dual sim. It's the "one sim version" of Huawei Ascend G330D (U8825D)tha is dual sim.
But currently i am downloading the builds from joestone and kra1o5 an find out if it's working (even mic).
Also i am confused of the words of joestone too because i am not a developer so for me it seems that porting CyanogenMod is more that just modify some files for configuration and replacing some liberys with the ones from stock rom.
Thanks
StefanV3
Joestone explain how to fix sims problem,but id is already fixed just try joestone build,good luck
Ps:watch for modaco forum,its english and all developers are here,dont'use herrero build
Tapatalkato via il mio gelato sfigato biscotto
Hi,
thanks...
StefanV3
Hi,
Many guides suggest using the well known dd command to dump partitions from Android systems.
(For instance http://forum.xda-developers.com/showthread.php?t=2450045)
However, some people suggest using nanddump / nandwrite instead, because of the nature of the storage components.
(see : http://forum.xda-developers.com/showthread.php?t=1247908)
So, my two questions :
1) Should nandump be used instead of dd clone Android partitions in low level mode (i.e. filesystem unaware) ?
2) What would be the possible shortcomings of using the widely spread "dd" command instead ?
Thanks.
nanddump => with OOB metadata (dd without)
A quite interesting question indeed, I'm wondering about that too...
in my unsuccessfull case of trying to get my phone SMS data back it didn't seem to make any difference.
As far as I could find out is the Physical image by nanddump a little bit more rich of content: including the metadata Out-Of-Band (OOB)
... don't think this is relevant in forensics analysis of data recovery! ....??
However, more experted details can be found here:
w w w. nowsecure .com/blog/2011/05/23/afphysical-method/
--> w w w. nowsecure .com/blog/2011/05/23/android-dd-image/
--> w w w. nowsecure .com/blog/2011/05/23/android-oob-image/
----> w w w. nowsecure .com/blog/2011/05/23/oob/
This thread is meant to share informations about trying to build a custom rom (CM, Omnirom...) for Lenovo Tab2 A7-10F
I'm currently trying to build CM12.1 using source
Any comment, hint, link, collaboration is welcome
Previously...
Previously...
[email protected] said:
Now time for new rom , i think you should try out building any cm based rom , omni is 75% Aosp so there might be some issues.
Sent from my A311 using XDA-Developers mobile app
Click to expand...
Click to collapse
pix106 said:
How can AOSP be an issue ? I always choose opensource when possible.
Click to expand...
Click to collapse
[email protected] said:
For mediatek open sources are not totally compatible with aosp's . every mtk chipset has same issue. So you can try out other others.
Sent from my A311 using XDA-Developers mobile app
Click to expand...
Click to collapse
pix106 said:
Ok, i may try some other rom though.
Thanks for your advice.
Do you have any info/link to mediatek sources ?
Click to expand...
Click to collapse
MaRtYy01 said:
Yep,agreed.With the years cyanogenmod went on its own and its not really based on aosp right now,while omni is more aosp based rom.
Click to expand...
Click to collapse
[email protected] said:
You can try out official sources with stock kernel , i have only links for mt6582 and mt6592 sources for now.
Sent from my A311 using XDA-Developers mobile app
---------- Post added at 03:31 AM ---------- Previous post was at 02:52 AM ----------
Can you link me any device/vendor tree . I'll try to compile cm12.1.
Sent from my A311 using XDA-Developers mobile app
Click to expand...
Click to collapse
pix106 said:
A few month ago, with a friend we did this for omnirom, we got boot (prebuilt kernel), graphics, but no wifi etc... : https://github.com/PixNDom/android_device_lenovo_Tab2A710F https://github.com/PixNDom/android_vendor_lenovo_Tab2A710F (other repos https://github.com/PixNDom).
We stopped as we did not know how to go on.
Yesterday after your advice i started a cm12.1 branch on my github (https://github.com/pix106/android_device_lenovo_Tab2A710F/blob/cm12.1), and got TWRP running using cm12.1 sources (with wrong device id, but it does not matter now).
I tried building the kernel again using lenovo sources (https://github.com/pix106/android_kernel_lenovo_Tab2A710F/tree/lenovo_osc_201511), with cm12, i got the same result than i got with omnirom : no graphics, "GTP Reset", i2C errors...
Thanks for your help
Click to expand...
Click to collapse
pix106 said:
Previously...
Click to expand...
Click to collapse
Try compiling with stock kernel ,
Sent from my A311 using XDA-Developers mobile app
That's what i do, i use prebuilt stock kernel.
But i would like to use kernel sources to build kernel with rom
If cm12 has the same problems as omni its most likely a kernel issue.Or there is something wrong with the sources.
I have worked on building CM12.1 today, and updated the first post
System is booting, adb and logcat work.
I added a few mali and hw related blobs, and a script to extract blobs from a file system dump instead of using adb to extract them from a running stock rom.
Any comment, hint, link, collaboration is welcome
pix106 said:
I have worked on building CM12.1 today, and updated the first post
System is booting, adb and logcat work.
I added a few mali and hw related blobs, and a script to extract blobs from a file system dump instead of using adb to extract them from a running stock rom.
Any comment, hint, link, collaboration is welcome
Click to expand...
Click to collapse
Cm12.1 booted succesfully. Then what are issues.
Sent from my A311 using XDA-Developers mobile app
[email protected] said:
Cm12.1 booted succesfully. Then what are issues.
Click to expand...
Click to collapse
Yesterday only adb and logcat were working.
Today, with a few MTK related flags, and a fresh build while i was at work, bootanimation starts
I'm going to add the missing files i detect from logcat and see what happens next...
After adding libxlog to PRODUCT_PACKAGES and providing a source for it, to avoid : cannot locate symbol "__xlog_buf_printf" referenced by "libdpframework.so", I got bootanimation.
I then added a few missing files as blobs.
Here are some errors i get :
Code:
E/audit_rules( 135): Could not read audit rules /data/misc/audit/audit.rules: No such file or directory
I/Vold ( 131): Vold 2.1 (the revenge) firing up
D/Vold ( 131): Volume sdcard0 state changing -1 (Initializing) -> 0 (No-Media)
D/Vold ( 131): Volume sdcard1 state changing -1 (Initializing) -> 0 (No-Media)
D/Vold ( 131): Volume usbotg state changing -1 (Initializing) -> 0 (No-Media)
I/Cryptfs ( 131): Check if PFE is activated on Boot
E/Cryptfs ( 131): Bad magic for real block device /[email protected]
E/Cryptfs ( 131): Error getting crypt footer and key
W/DirectVolume( 131): Deprecated implied prefix pattern detected, please use '/devices/platform/mtk-msdc.0/mmc_host*' instead
I/[MALI][Gralloc]( 132): dlopen libtz_uree.so fail
E/HAL ( 132): dlopen failed: library "libtz_uree.so" not found
I/[MALI][Gralloc]( 207): dlopen libtz_uree.so fail
E/Netd ( 520): Failed to open /proc/sys/net/ipv6/conf/default/use_optimistic: No such file or directory
E/Netd ( 520): Failed to open /proc/sys/net/ipv6/conf/ifb0/use_optimistic: No such file or directory
E/Netd ( 520): Failed to open /proc/sys/net/ipv6/conf/ifb1/use_optimistic: No such file or directory
E/Netd ( 520): Failed to open /proc/sys/net/ipv6/conf/ip6tnl0/use_optimistic: No such file or directory
E/Netd ( 520): Failed to open /proc/sys/net/ipv6/conf/lo/use_optimistic: No such file or directory
E/Netd ( 520): Failed to open /proc/sys/net/ipv6/conf/sit0/use_optimistic: No such file or directory
E/android.os.Debug( 522): failed to load memtrack module: -2
E/MediaPlayerFactory(15733): calling dlopen on FACTORY_LIB
E/MediaPlayerFactory(15733): Failed to open FACTORY_LIB Error : dlopen failed: library "libdashplayer.so" not found
E/CameraService(15733): setUpVendorTags: Vendor tag operations not fully defined. Ignoring definitions.
E/HAL (15733): load: module=/system/lib/hw/audio.primary.mt8127.so
E/HAL (15733): dlopen failed: cannot locate symbol "_ZN7android11AudioSystem24getVoiceUnlockDLInstanceEv" referenced by "audio.primary.mt8127.so"...
E/AudioFlinger(15733): int android::load_audio_interface(const char*, audio_hw_device_t**) couldn't load audio hw module audio.primary (Invalid argument)
E/MonoPipe(15733): Failed to fetch local time frequency when constructing a MonoPipe (res = -32). getNextWriteTimestamp calls will be non-functional
E/AudioPolicyManager(15733): Default device 00000002 is unreachable
E/AudioPolicyManager(15733): Failed to open primary output
If add libtz_uree.so, hwcomposer (blob) fails, bootanimation does not start anymore.
I also tried adding libdashplayer to PRODUCT_PACKAGES, but it didn't help. And i have no source.
During a previous omnirom 5.1 build with a friend (https://github.com/PixNDom/android_device_lenovo_Tab2A710F), we used https://github.com/aomp/android_mediatek_frameworks
It allowed to build some libs, but it's rather old, i think.
Does anyone have a recent repository for mediatek sources, so hwcomposer (and other files that are blobs for now) could be built ?
pix106 said:
After adding libxlog to PRODUCT_PACKAGES and providing a source for it, to avoid : cannot locate symbol "__xlog_buf_printf" referenced by "libdpframework.so", I got bootanimation.
I then added a few missing files as blobs.
Here are some errors i get :
Code:
E/audit_rules( 135): Could not read audit rules /data/misc/audit/audit.rules: No such file or directory
I/Vold ( 131): Vold 2.1 (the revenge) firing up
D/Vold ( 131): Volume sdcard0 state changing -1 (Initializing) -> 0 (No-Media)
D/Vold ( 131): Volume sdcard1 state changing -1 (Initializing) -> 0 (No-Media)
D/Vold ( 131): Volume usbotg state changing -1 (Initializing) -> 0 (No-Media)
I/Cryptfs ( 131): Check if PFE is activated on Boot
E/Cryptfs ( 131): Bad magic for real block device /[email protected]
E/Cryptfs ( 131): Error getting crypt footer and key
W/DirectVolume( 131): Deprecated implied prefix pattern detected, please use '/devices/platform/mtk-msdc.0/mmc_host*' instead
I/[MALI][Gralloc]( 132): dlopen libtz_uree.so fail
E/HAL ( 132): dlopen failed: library "libtz_uree.so" not found
I/[MALI][Gralloc]( 207): dlopen libtz_uree.so fail
E/Netd ( 520): Failed to open /proc/sys/net/ipv6/conf/default/use_optimistic: No such file or directory
E/Netd ( 520): Failed to open /proc/sys/net/ipv6/conf/ifb0/use_optimistic: No such file or directory
E/Netd ( 520): Failed to open /proc/sys/net/ipv6/conf/ifb1/use_optimistic: No such file or directory
E/Netd ( 520): Failed to open /proc/sys/net/ipv6/conf/ip6tnl0/use_optimistic: No such file or directory
E/Netd ( 520): Failed to open /proc/sys/net/ipv6/conf/lo/use_optimistic: No such file or directory
E/Netd ( 520): Failed to open /proc/sys/net/ipv6/conf/sit0/use_optimistic: No such file or directory
E/android.os.Debug( 522): failed to load memtrack module: -2
E/MediaPlayerFactory(15733): calling dlopen on FACTORY_LIB
E/MediaPlayerFactory(15733): Failed to open FACTORY_LIB Error : dlopen failed: library "libdashplayer.so" not found
E/CameraService(15733): setUpVendorTags: Vendor tag operations not fully defined. Ignoring definitions.
E/HAL (15733): load: module=/system/lib/hw/audio.primary.mt8127.so
E/HAL (15733): dlopen failed: cannot locate symbol "_ZN7android11AudioSystem24getVoiceUnlockDLInstanceEv" referenced by "audio.primary.mt8127.so"...
E/AudioFlinger(15733): int android::load_audio_interface(const char*, audio_hw_device_t**) couldn't load audio hw module audio.primary (Invalid argument)
E/MonoPipe(15733): Failed to fetch local time frequency when constructing a MonoPipe (res = -32). getNextWriteTimestamp calls will be non-functional
E/AudioPolicyManager(15733): Default device 00000002 is unreachable
E/AudioPolicyManager(15733): Failed to open primary output
If add libtz_uree.so, hwcomposer (blob) fails, bootanimation does not start anymore.
I also tried adding libdashplayer to PRODUCT_PACKAGES, but it didn't help. And i have no source.
During a previous omnirom 5.1 build with a friend (https://github.com/PixNDom/android_device_lenovo_Tab2A710F), we used https://github.com/aomp/android_mediatek_frameworks
It allowed to build some libs, but it's rather old, i think.
Does anyone have a recent repository for mediatek sources, so hwcomposer (and other files that are blobs for now) could be built ?
Click to expand...
Click to collapse
Can't you use something from the fire 2015 device tree?Its fully complete,the cm12.1 for it is bugless.Not sure if building roms works like that,but if you are missing drivers or something since it has the same chipset and kernel the stuff from it might work,I guess.
https://bitbucket.org/amazonfire2015/cm_device_amazon_ford
BTW as far as I know mediatek has no official source like qualcomm,thats why building roms for it is an issue.
MaRtYy01 said:
Can't you use something from the fire 2015 device tree?Its fully complete,the cm12.1 for it is bugless.Not sure if building roms works like that,but if you are missing drivers or something since it has the same chipset and kernel the stuff from it might work,I guess.
https://bitbucket.org/amazonfire2015/cm_device_amazon_ford
BTW as far as I know mediatek has no official source like qualcomm,thats why building roms for it is an issue.
Click to expand...
Click to collapse
Thanks, i'll have a look at this.
Either try to build it changing hardware related stuff, or pick a few things from it... or both.
I had already looked on fire 2015 source on github in 2015, it has changed since then too.
At the moment, i'm trying to build with some other mt8127 github repo (https://github.com/neporood/android_device_viking_mt8127), changing hardware stuff.
Thanks
pix106 said:
During a previous omnirom 5.1 build with a friend (https://github.com/PixNDom/android_device_lenovo_Tab2A710F), we used https://github.com/aomp/android_mediatek_frameworks
It allowed to build some libs, but it's rather old, i think.
Click to expand...
Click to collapse
sorry, i actualy wanted to mention https://github.com/aomp/android_mediatek_hardware
pix106 said:
sorry, i actualy wanted to mention https://github.com/aomp/android_mediatek_hardware
Click to expand...
Click to collapse
I'm now trying to build using this hardware/mediatek repo, it should provide libs that are blobs for now.
After adding a few .h files, it builds fine, but i don't get all the libs that should be built from this repo, although i added them to PRODUCT_PACKAGES.
I'll push these changes to github.
Even with a linux and python dev experience, i'm quite new to building roms, i may have missed a few things.
I made some ROMs for my Nexus4 and even there, you needed some closed source libs from LG which was not so easy to obtain.. On a damn nexus device..
Some people mentioned that an Amazon Fire tablet is mostly the same tablet..
The CPU looks the same for the Fire HD 7" (4th gen - 2014) and the GPU is the same from FireHD 8.9" (2.5th Gen) and Fire HD 7 (3rd Gen).
Does CM and/or other Custom ROMs work with Amazon-Fire tablets?
thE_29 said:
I made some ROMs for my Nexus4 and even there, you needed some closed source libs from LG which was not so easy to obtain.. On a damn nexus device..
Some people mentioned that an Amazon Fire tablet is mostly the same tablet..
The CPU looks the same for the Fire HD 7" (4th gen - 2014) and the GPU is the same from FireHD 8.9" (2.5th Gen) and Fire HD 7 (3rd Gen).
Does CM and/or other Custom ROMs work with Amazon-Fire tablets?
Click to expand...
Click to collapse
apparently it does, so there's hope.
I tried building using a mt8127 device tree and vendor from github, with my device tree and his vendor.
I have also made a "noblob" build (before using the hardware/mediatek repo) so that blobs can be pushed.
pix106 said:
apparently it does, so there's hope.
I tried building using a mt8127 device tree and vendor from github, with my device tree and his vendor.
I have also made a "noblob" build (before using the hardware/mediatek repo) so that blobs can be pushed.
Click to expand...
Click to collapse
Can you provide cm12.1 meta inf folder and boot.img?I feel like the one from the amazon fire is superpatched because of fire os crap and cant be ported and Im trying to make a flashable version of your modified stock rom but I dont have meta inf folder(tried one from cm12.0 from one mt6582 device,it did boot but it got stuck at the first boot loading screen).
MaRtYy01 said:
Can you provide cm12.1 meta inf folder and boot.img?I feel like the one from the amazon fire is superpatched because of fire os crap and cant be ported and Im trying to make a flashable version of your modified stock rom but I dont have meta inf folder(tried one from cm12.0 from one mt6582 device,it did boot but it got stuck at the first boot loading screen).
Click to expand...
Click to collapse
I tried too a few months ago, neither did it boot.
The "noblobs" build i did a few days ago : https://mega.nz/#F!i5h23AKY!jnsxiU8Ob0WiDt1YnSabLg
I'm quite sure a few needed init scripts are missing from this build, and from the device tree in general.
Happy testing .
pix106 said:
I tried too a few months ago, neither did it boot.
The "noblobs" build i did a few days ago : https://mega.nz/#F!i5h23AKY!jnsxiU8Ob0WiDt1YnSabLg
I'm quite sure a few needed init scripts are missing from this build, and from the device tree in general.
Happy testing .
Click to expand...
Click to collapse
is this build boots or not . (just asking) by the way :good::good:
[email protected] said:
is this build boots or not . (just asking) by the way :good::good:
Click to expand...
Click to collapse
Its stuck on the bootlogo,since it has no blobs.
BTW the results of my "testing"
-changing the boot.img didnt help,the frecking amazon fire cm12.1 refuses to boot no matter what I do
-the updater script seemed useless,since the system folder in the roms built from source is in just one file
Anyway,got another idea.If I do a backup of system via twrp I probably can use the updater script with it.
[email protected] said:
is this build boots or not . (just asking) by the way :good::good:
Click to expand...
Click to collapse
MaRtYy01 said:
Its stuck on the bootlogo,since it has no blobs.
Click to expand...
Click to collapse
Exactly. It boots, adb and logcat work, but nothing more.
Missing files can be looking for errors in logcat, and be pushed "live" from stock rom dump.
MaRtYy01 said:
BTW the results of my "testing"
-changing the boot.img didnt help,the frecking amazon fire cm12.1 refuses to boot no matter what I do
Click to expand...
Click to collapse
no adb nor logcat ?
MaRtYy01 said:
-the updater script seemed useless,since the system folder in the roms built from source is in just one file
Click to expand...
Click to collapse
is /system mounted ?
MaRtYy01 said:
Anyway,got another idea.If I do a backup of system via twrp I probably can use the updater script with it.
Click to expand...
Click to collapse
i did not understand what you are planning to do... anyway good luck with it