Welcome to Our Website!

Just for your information:

cy·an·o·gen
1. A colorless, flammable, pungent, highly poisonous gas, C2N2, used as a rocket propellant, an insecticide, and a chemical weapon.
2. A univalent CN group found in simple and complex cyanide compounds.

 June 17th, 2015 / 9:56pm
Lookin' at this again... The following was Originally posted by @aosp on our  CM11 Thread  :


@bunchies @Chucktr @Flyhalf205 @kkozma @REV3NT3CH @Snuzzo

Any of you working on lollipop for the vigor?

msm8660-lollipop - http://pastebin.com/raw.php?i=py1jRS1v

http://forum.xda-developers.com/show....php?t=3083272

[ROM][5.1.1] Unoffical Cyanogenmod 12.1 for Evo 3D - 28 May

http://forum.xda-developers.com/show....php?t=3102149

[DEV] Android L ROM and Kernel Development

http://forum.xda-developers.com/show....php?t=2911152

[ROM][Android 4.4.4/5.0] CyanogenMod11/12 by ivanich [3.4Kernel][Up-To-Date][Stable]

http://forum.xda-developers.com/show....php?t=2946570

[UnORom][Lollipop] CM 12 for Sensation [12 Jan]


 Vigor-Pyramid Board Compare    No Wire S-Off  
 github of MY Kernel    github of Keiths Kernel    Vigorous Revival  


Today we started on this newest version. Now it is later and there was a lot of scrambling by some of my team members in trying, and accomplishing, to get the 3.4 kernel to compile. Right now I have a mixture of notes and I need to get them straightened out. Will do that as I re-start my creation of CM12 for the Rez. 15Feb15
Ah hell, I can't make heads nor tails out of this stuff. Gona start all over!!!
19Feb15: 21Feb15: One more time... Hopefully, I have learned enough to do it better... but I really doubt it. This is not something that you do, or learn, overnight. Like everything else, it takes time, practice and application of what you learn. And the mistakes. Theoretically, this setup should only need to be performed once. -But- it needs to be done in a manner that updates can be performed and not mess things up. Annnddd... I need to make up my mind on what I am doing. Restarted -and- then went back to my first start -- and now I am restarting again!!

  1. Used the previously Modified initrepo_hybrid.sh script to bring over the source. The only mods were dir names. This was a script created by Neo -or- one of the 'other' previous 'OPs'. It does the 'init' and then fires off a "repo sync"... which initially takes a long, long time. Can be, and usually is, over an hour!! As you can see in the part I captured below. Well, this latest one only took 47mins.
      -- -- -- 9 Apr 15 / 5:45am -- -- --
    Fetching projects: 100% (546/546), done.  
    Checking out files: 100% (8507/8507), done.ut files:  16% (1403/8507)   
    Checking out files: 100% (10496/10496), done.t files:  22% (2413/10496)   
    Checking out files: 100% (59286/59286), done.t files:   9% (5638/59286)   
    Checking out files: 100% (10980/10980), done.
    Checking out files: 100% (4343/4343), done.out files:  12% (537/4343)   
    Checking out files: 100% (28621/28621), done.ut files:  27% (7997/28621)   
    Checking out files: 100% (54/54), done.king out files:   3% (2/54)   
    Checking out files: 100% (741/741), done.ng out files:   0% (6/741)   
    Checking out files: 100% (269/269), done.ng out files:  13% (37/269)   
    Checking out files: 100% (4197/4197), done. out files:   6% (279/4197)   
    Checking out files: 100% (194/194), done.
    Checking out files: 100% (235/235), done.ng out files:  43% (103/235)   
    Checking out files: 100% (244/244), done.
    Checking out files: 100% (69331/69331), done.ut files:  20% (14309/69331)   
    Checking out files: 100% (35/35), done.king out files:  45% (16/35)   
    Checking out files: 100% (1030/1030), done.
    Checking out files: 100% (13/13), done.king out files:  46% (6/13)   
    Checking out files: 100% (149/149), done.
    Syncing work tree: 100% (546/546), done.  
    
    ~/Android/cm12.0-rezound/vendor/cm ~/Android/cm12.0-rezound
    ./TR-12initrepo_hybrid.sh: line 43: ./get-prebuilts: No such file or directory
    ~/Android/cm12.0-rezound
    
    real    47m40.278s
    user    12m50.010s
    sys     4m4.888s
    
    Had prob with prebuilts. They were NOT included. Copied the KK get-prebuilts script from vendor/cm in CM11 to vendor/cm in CM12. Started init again... looks OK.
    get-prebuils creates a 'proprietary' dir in vendor/cm which contains some Vendor Libs.
     April 8, 2015 / 5:37am  Well... I'm doing it again! Hmmm... did it again.
      -- -- --
    Syncing work tree: 100% (542/542), done.  
    
    ~/Android/cm12.0-rezound/vendor/cm ~/Android/cm12.0-rezound
      % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                                     Dload  Upload   Total   Spent    Left  Speed
    100   178  100   178    0     0    779      0 --:--:-- --:--:-- --:--:--   780
    100  554k  100  554k    0     0   574k      0 --:--:-- --:--:-- --:--:--  574k
    Archive:  ./proprietary/Term.apk
      inflating: ./proprietary/lib/arm64-v8a/libjackpal-androidterm4.so  
      inflating: ./proprietary/lib/armeabi/libjackpal-androidterm4.so  
      inflating: ./proprietary/lib/armeabi-v7a/libjackpal-androidterm4.so  
      inflating: ./proprietary/lib/mips/libjackpal-androidterm4.so  
      inflating: ./proprietary/lib/mips64/libjackpal-androidterm4.so  
      inflating: ./proprietary/lib/x86/libjackpal-androidterm4.so  
      inflating: ./proprietary/lib/x86_64/libjackpal-androidterm4.so  
    ~/Android/cm12.0-rezound
    
    real    2m52.238s
    user    0m12.595s
    sys     0m10.259s
    
     April 9th, 2015 / 7:58am  Well now, this time we got different libraries. I could just copy over the ones from CM11 -but- we'll just continue for now. Maybe they have something new.
    Archive:  ./proprietary/Term.apk
      inflating: ./proprietary/lib/armeabi/libjackpal-androidterm5.so  
      inflating: ./proprietary/lib/mips/libjackpal-androidterm5.so  
      inflating: ./proprietary/lib/x86/libjackpal-androidterm5.so  
      inflating: ./proprietary/lib/armeabi/libjackpal-termexec2.so  
      inflating: ./proprietary/lib/mips/libjackpal-termexec2.so  
      inflating: ./proprietary/lib/x86/libjackpal-termexec2.so  
    
  2. Attempted compile. First modified the make.sh script.
    The following was on this second go around AFTER just doing the very first init. Nothing else. The get-prebuilts above was performed AFTER this try. This is here to show what happens IF you don't set things up
    > time ./TR-12make.sh 
    including device/generic/mini-emulator-x86/vendorsetup.sh
    including device/generic/mini-emulator-x86_64/vendorsetup.sh
    including device/generic/mini-emulator-arm64/vendorsetup.sh
    including device/generic/mini-emulator-armv7-a-neon/vendorsetup.sh
    including device/generic/mini-emulator-mips/vendorsetup.sh
    including vendor/cm/vendorsetup.sh
    including sdk/bash_completion/adb.bash
    including vendor/cm/bash_completion/git.bash
    including vendor/cm/bash_completion/repo.bash
    Set cache size limit to 20.0 Gbytes
    including vendor/cm/vendorsetup.sh
    Device vigor not found. Attempting to retrieve device repository from CyanogenMod Github (http://github.com/CyanogenMod).
    Found repository: android_device_htc_vigor
    Default revision: cm-12.0
    Checking branch info
    Default revision cm-12.0 not found in android_device_htc_vigor. Bailing.
    Branches found:
    cm-10.1
    jellybean
    Use the ROOMSERVICE_BRANCHES environment variable to specify a list of fallback branches.
    
    ** Don't have a product spec for: 'cm_vigor'
    ** Do you have the right repo manifest?
    
    No such item in brunch menu. Try 'breakfast'
    
    real    0m2.065s
    user    0m0.102s
    sys     0m0.024s
    
  3.  April 9th, 2015 / 6:07pm  Actually, another try at CM12 for my ReZound. This is above the previous remarks concerning the 'kernel' cause I am trying to do things the 'correct way' -or- at least as I can best understand the correct way. Not sure there is one for my Rez. They, CyanogenMod, dropped us long ago.
    Refer:  CyanogenMod Device  
    > breakfast vigor
    including vendor/cm/vendorsetup.sh
    ls: cannot access device/*/vigor/cm.mk: No such file or directory
    build/core/product_config.mk:234: *** Can not locate config makefile for product "cm_vigor".  Stop.
    Device vigor not found. Attempting to retrieve device repository from CyanogenMod Github (http://github.com/CyanogenMod).
    Found repository: android_device_htc_vigor
    Default revision: cm-12.0
    Checking branch info
    Default revision cm-12.0 not found in android_device_htc_vigor. Bailing.
    Branches found:
    cm-10.1
    jellybean
    Use the ROOMSERVICE_BRANCHES environment variable to specify a list of fallback branches.
    ls: cannot access device/*/vigor/cm.mk: No such file or directory
    build/core/product_config.mk:234: *** Can not locate config makefile for product "cm_vigor".  Stop.
    
    ** Don't have a product spec for: 'cm_vigor'
    ** Do you have the right repo manifest?
    
    Like I said earlier, they dropped us!! So, I will copy the files from CM11 and try to make it work.
  4. How's about that..?? I haven't mentioned that I copied over the 3.4 kernel to this build dir.!! Well, I am mentioning it now. The vimsm8960 kernel that Shaun helped me get successfully compiled was copied over to this Build.
    Had to create a 'kernel' dir, an 'htc' dir, a 'vigor-3.4' dir, and then copyied the folders/files from one to the other.
  5. Had to copy the HTC dir from device dir in CM11 to device dir in CM12.
  6. Had to change from Oracle Java to openJDK Java. Did it.
    Reason: They changed from the "Common Oracle Java" to the Open "openJDK" form. You get warnings in both CM11 and CM12 IF'n you use the wrong one.
  7. Then we searched for all the places where the "vigor-3.0" dir was referenced. Only two places: ../device/htc/msm8660-common/BoardConfigCommon.mk
    # Kernel
    TARGET_KERNEL_SOURCE := kernel/htc/vigor-3.4
    I changed the 3.0 to 3.4.
    and: ../device/htc/vigor/liquid.dependencies
    {
    "remote": "gh",
    "repository": "vigor/vigor_aosp_kernel",
    "target_path": "kernel/htc/vigor-3.4",
    "revision": "cm-11.0_hybrid"
    },
    Left this one alone for now. Nope... changed it. It was not finding my 3.4 and I could not find anywhere else where is specified the vigor-3.0 dir.
  8. We know that just copying the 3.4 kernel to our CM12 dir will NOT be cool. -But- ya gota start somewhere. So, we are copying the dir.
  9. Var info at the start of the make.
    ============================================
    PLATFORM_VERSION_CODENAME=REL
    PLATFORM_VERSION=5.0.2
    CM_VERSION=12-20150222-UNOFFICIAL-vigor
    TARGET_PRODUCT=cm_vigor
    TARGET_BUILD_VARIANT=userdebug
    TARGET_BUILD_TYPE=release
    TARGET_BUILD_APPS=
    TARGET_ARCH=arm
    TARGET_ARCH_VARIANT=armv7-a-neon
    TARGET_CPU_VARIANT=scorpion
    TARGET_2ND_ARCH=
    TARGET_2ND_ARCH_VARIANT=
    TARGET_2ND_CPU_VARIANT=
    HOST_ARCH=x86_64
    HOST_OS=linux
    HOST_OS_EXTRA=Linux-3.16.7-7-desktop-x86_64-with-SuSE-13.2-x86_64
    HOST_BUILD_TYPE=release
    BUILD_ID=LRX22G
    OUT_DIR=/home/ctaylor/Android/cm12.0-rezound/out
    ============================================
    
  10. Complained about sepolicy.
    external/sepolicy/Android.mk:56: 
      *** No sepolicy file found for files in device/htc/msm8660-common/sepolicy vendor/cm/sepolicy.  Stop.
    
    Now we found 3 different places where it is messing with this file so I don't know why it is not created. For now we are going to comment it our of the make. It is not important.
    Re-started make.
  11. Complained about no rules to make the kerenl. That was cause I had to make the change in liquid.dependencies.
    Well, that was not true. had to change ../device/htc/vigor/BoardConfig.mk
    # Kernel
    BOARD_KERNEL_BASE := 0x48800000
    BOARD_KERNEL_PAGE_SIZE := 2048
    BOARD_KERNEL_CMDLINE := console=ttyHSL3 androidboot.hardware=vigor no_console_suspend=1 androidboot.selinux=permissive
    TARGET_KERNEL_VERSION := 3.4
    TARGET_KERNEL_CONFIG := vigor_aosp_defconfig
    TARGET_KERNEL_SOURCE := kernel/htc/vigor-$(TARGET_KERNEL_VERSION)
    Then Re-started make.
  12. ../frameworks/native/services/surfaceflinger
    make: *** No rule to make target '/home/ctaylor/Android/cm12.0-rezound/out/target/product/vigor/obj/SHARED_LIBRARIES/libqdutils_intermediates/export_includes',
       needed by '/home/ctaylor/Android/cm12.0-rezound/out/target/product/vigor/obj/SHARED_LIBRARIES/libsurfaceflinger_intermediates/import_includes'.  Stop.
    
  13. We copied over the ../vendor/htc dir from CM11.
    That was what was needed to fix the above error. NO it wasn't!! -But- got further in the compile.. I think. I believe that this dir is needed, -but- not to fix this problem... at least not directly.
    ../frameworks/native/services/surfaceflinger/Android.mk
    ../build/core/binary.mk
    Well now, LOCAL_SHARED_LIBRARIES is used everywhere....
    .... -and- surfaceflinger seems to be just as popular.
    hmmmm...libqdutils_intermediates only shows up in the error file.
    However, libqdutils shows up in the Android.mk file above, and is combinded with a number of others into the LOCAL_SHARED_LIBRARIES! Which also appears to be where the import and export includes are created. I could not find an occurance of 'libqdutils_intermediates' anywhere except as mentioned above. It must be created by combing some variable names... and it becomes a dir name in the 'out' dir.
    Velllly intellesting... in CM11 'libqdutils' is in the name of 6 folders and 5 files... -but- in CM12 it is in the name of 9 folders and 77 files!! Now the other interesting part of this is... one of the dirs and all of the files in CM11 are in the 'out' dir. In CM12 all of them are in 'regular' folders and dirs cause I looked 'after' doing a make clean.
    Well, we gota perform a make on the CM12 so's we can see IF the 5 'libqdutils.so' appear.
    Darn!!! Got called into work!! I'll start a make whilst I am gone. Feb 16th, 2015 / 1:26pm
  14. Feb 17th, 2015 / 2:34am - Back and had a little snooze. The main reason I create these pages is for meself. Now hopefully we can take up where we left off. The make finished -and- it looked different than the previous one. Can't think of anything that I did 'other' than a 'Make Clean'. Ok first we'll check that folder and file count.
    Hmmmm... IT IS THE SAME. Which means NONE of the 'libqdutils.so' got created. Which is just what the error message says!! OK. I still have that CM11 that I was trying to put kernel 3.4 in. I think I did a make clean on it before I left it. We'll look at the folder and file count in it concerning 'libqdutils'.
    OH DARN!! OH DARN!! Messed up. Did the search by filename in CM11 and then by content in CM12!!! Trying again!!
    Allllrighty now... In the 'compiled' CM12 we still get 9 folders -but- zero files. In the 'UN-compiled' CM11 we only get 5 folders and zero files.
    A re-check on CM12 'UN-compiled' and we still have the 9 folders -but- no files. NOW we go back to the 'content' check so we can find out just where they are supposed to be created. Hmmmm... get 77 files -but- only 3 of them are NOT mk files. We gots 74 mk files to look thru!! -But- in searching on the Web it appears that libqdutils is associated with QCOM Display.
    It is: Now then, the TARGET_BOARD_PLATFORM is set to 'msm8660' and it wants to build a display-caf -but- it can not find an msm8660 in the display-caf dir soooo ... we will create one by copying the contents of ../hardware/qcom/display-caf dir from CM11 and put it in CM12 in a newly created ../hardware/qcom/display-caf/msm8660. Yep! It is a gamble -but- the only thing the Rez has is an 8660. Now we try the make again. OK! Got us thru to the next problem.
  15. Remote Procedure Call - hardware/qcom/gps   QCOM_BOARD_PLATFORMS
        ../build/core/qcom_utils.mk
    I do believe they error'd in their logic in the make file. Please look at:../harware/qcom/gps/Android.mk Don't think that they are getting what they wanted. I believe that the last ifneq should be ifeq... what do you think??
    # Changed from ifneq to ifeq 17Feb15 cmt

    Well, that didn't fix it either -but- TARGET_GPS_HAL_PATH is checked for and it is NOT defined anywhere. So I commented it out.

    Did not fix it either. At a suggestion from Shaun, I made a change to the manifest.xml file in .repo dir. revision="cm-11.0"; added to the end of the gps line
    <project path="hardware/qcom/gps" name="CyanogenMod/android_hardware_qcom_gps" groups="qcom" revision="cm-11.0" />
    and then did: repo sync hardware/qcom/gps
    Deleted the gps dir and:
    > repo sync hardware/qcom/gps
    Fetching project CyanogenMod/android_hardware_qcom_gps
    So that is what you mean by "branching the CM11 stuff"??
    Allllllrigthy... that got us thru ... to the next bug. Got a 12 minute compile... before it failed.
  16.  ** ** Begin cmaera_vendor_tags bug...  Shaun helped me on this part...
    In file included from hardware/libhardware/include/hardware/camera.h:20:0,
                     from device/htc/msm8660-common/camerawrapper/CameraWrapper.cpp:34:
    hardware/libhardware/include/hardware/camera_common.h:28:39: fatal error: system/camera_vendor_tags.h: No such file or directory
     #include <system/camera_vendor_tags.h>
    
    It is looking for <system/camera_vendor_tags.h> and it is in:
    <system/media/camera/include/system/camera_vendor_tags.h> I'll fix it and try again.
    My first try did not fix it. Need to find and fix the "include path".
    The main.mk file contains:
    include $(BUILD_SYSTEM)/config.mk
    Which contains a bunch of directives to set the include path(s). Further checking is needed.

  17.      hackers-guide-to-git    Git-Tools-Stashing    git-commit    git-fetch    git-cherry-pick  

  18. refer to manifest.xml for the list of projects. might be easier for you to read
    let's say you made a change to system/core/Android.mk
    system/core is it's local path
    "git stash"
    "git stash branch <name>"
    then "git add -A"
    "git commit -m "<insert a commit message here>" "
    safe to sync. ill show you how to get those changes back later
  19. So, what I did was; insert:
        $(TOPDIR)system/media/camera/include \
    into the list of SRC_HEADERS in ../build/core/config.mk
    Compiler now gets past this spot..
     ** ** End cmaera_vendor_tags bug... 
  20. Got a new problem in the media area. Problem is compounded by the fact that there are 6 of these QCMediaDefs.h files, in 6 differnt dirs. Which one is it looking for?? Only 3 in CM11.
    frameworks/av/media/libmediaplayerservice/StagefrightRecorder.cpp:65:25: fatal error: QCMediaDefs.h: No such file or directory
     #include <QCMediaDefs.h>
                             ^
    compilation terminated.
    make: *** [/home/ctaylor/Android/cm12.0-rezound/out/target/product/vigor/obj/SHARED_LIBRARIES/libmediaplayerservice_intermediates/StagefrightRecorder.o] Error 1
    make: *** Waiting for unfinished jobs....
    frameworks/av/media/libstagefright/ACodec.cpp:62:25: fatal error: QCMediaDefs.h: No such file or directory
     #include <QCMediaDefs.h>
                             ^
    compilation terminated.
    make: *** [/home/ctaylor/Android/cm12.0-rezound/out/target/product/vigor/obj/SHARED_LIBRARIES/libstagefright_intermediates/ACodec.o] Error 1
    
    The file is a Conditional Include; Condition?? is ENABLE_AV_ENHANCEMENTS.
    Added:
        $(TOP)/hardware/qcom/media-caf/mm-core/inc \
    to LOCAL_C_INCLUDES := in Android.mk in ../frameworks/av/media/libmediaplayerservice/
    Added:
        $(TOP)/hardware/qcom/media-caf/mm-core/inc \
    to LOCAL_C_INCLUDES := in Android.mk in ../frameworks/av/media/libstagefright/
    This appears to have satisfied the compiler ... we moved on.
  21. To the Camera!! It has stopped in what appears to be the section that Neo "cherrypic'd". Since I don't know all the ins and outs of this 'git' ... yet, I am gona do my "copy and paste thing". I'll save the 'new' and copy in the 'old' from CM11. It will either work -or- it won't.!!
    Well now, in this 2nd -or- 3rd go around of the re-start of the re-starts, I changed my mind again and again. This time I am trying to figure out 'just what this does'.
       git-fetch - Download objects and refs from another repository:
    
    > git fetch https://github.com/vigor/android_frameworks_av cm-11.0 && git cherry-pick 1dd29957a2292c0398923ffff8e2aca47c5931ac
    remote: Counting objects: 13, done.
    remote: Compressing objects: 100% (12/12), done.
    remote: Total 13 (delta 0), reused 4 (delta 0)
    Unpacking objects: 100% (13/13), done.
    From https://github.com/vigor/android_frameworks_av
     * branch            cm-11.0    -> FETCH_HEAD
    fatal: bad object 1dd29957a2292c0398923ffff8e2aca47c5931ac
    ctaylor@Tipper-i7:~/Android/cm12.0-rezound/frameworks/av
    > git status
    Not currently on any branch.
    Changes not staged for commit:
      (use "git add ..." to update what will be committed)
      (use "git checkout -- ..." to discard changes in working directory)
    
            modified:   media/libmediaplayerservice/Android.mk
            modified:   media/libstagefright/Android.mk
    
    no changes added to commit (use "git add" and/or "git commit -a")
    

    Well, as can be seen, the previous "cherry-pic" from CM11 did not bode well with this version. -BUT- it gives us some areas to look at.

    ./frameworks/av/camera/CameraParameters.cpp
    ./frameworks/av/camera/Android.mk
    ./frameworks/av/include/camera/CameraParameters.h

    Parameters - (any factor that defines a system and determines (or limits) its performance)
    KEY_AUTO_EXPOSURE, KEY_SCENE_DETECT, KEY_SKIN_TONE_ENHANCEMENT, KEY_SCENE_DETECT, KEY_TOUCH_AF_AEC
    Edited the make file and copied over the cpp and h files.
    CameraParameters.cpp include file was in quotes instead of angle brackets.

    ./frameworks/av/services/camera/libcameraservice/CameraService.cpp
    ./frameworks/av/services/camera/libcameraservice/device1/CameraHardwareInterface.h
    ./frameworks/av/services/camera/libcameraservice/Android.mk
    ./frameworks/av/media/libstagefright/Android.mk
    ./frameworks/av/media/libstagefright/CameraSource.cpp
    ./frameworks/av/media/libstagefright/OMXCodec.cpp
    Ooooopppss... overwrote my previous fix in libstagefright/Android.mk. Fixed it and Re-Starting compile.
  22. libconnectivitymanager is used in CM11 -but- not in CM12. The make file was part of my "chucktrpic". It was in the 'err' file and in ../frameworks/av/media/libstagefright/Android.mk. So I removed it from the make file... -and- restarted the compile.
  23. Well, it is heavy into the Camera now and it looks like it involves the area that Neo fixed the last time. A number of the files that I just copied into it are mentioned. Expected. But now to see why and to see if'n we can understand it!!
    target thumb C++: libcameraservice <= frameworks/av/services/camera/libcameraservice/api1/client2/StreamingProcessor.cpp
    build/core/binary.mk:617: recipe for target '/home/ctaylor/Android/cm12.0-rezound/out/target/product/vigor/obj/SHARED_LIBRARIES/libcameraservice_intermediates/CameraService.o' failed
    target thumb C++: libcameraservice <= frameworks/av/services/camera/libcameraservice/api1/client2/JpegProcessor.cpp
    build/core/binary.mk:617: recipe for target '/home/ctaylor/Android/cm12.0-rezound/out/target/product/vigor/obj/SHARED_LIBRARIES/libcameraservice_intermediates/api1/client2/Parameters.o' failed
    
    Ok, the "chucktrpic" didn't work. Gona put things back and then do what I shuda done in the first place. Look at what Neo changed and see how it will fit with the new stuff.
  24. Well, we have combined both types of changes, mostly just my copy and paste. I modified a couple of the "Android.mk" files and I am not sure that I documented them. Also got a little cornfused on what was what. Soooo... I got another copy of CM12 and it is just for reference. It will remain unmodified, unchanged ... nada ... nothing altered... so that I can keep things straight. Another hour of time wasted... but hopefully for a good purpose.
  25. Now then, what was all the concern?? "LPAPlayer.cpp"... -and- the "Android.mk" file in ../frameworks/av/media/libstagefright/. "LPAPlayer.cpp" is conditionally included -and- I'm not sure IF'n the 'conditional' include was in CM12 -or- not. I don't think so UNLESS we copy LPAPlayer from CM11 to CM12 ... which we may do -AFTER- we find out what it is!!
    And it is: LPAPlayer (Low Power Audio Player) in Android stagefright.
    Now then, there 'is' an LPAPlayerALSA.cpp in both CM11 and in CM12. I am gona copy LPAPlayer.cpp from CM11 ../frameworks/av/media/libstagefright/ to CM12 ../frameworks/av/media/libstagefright/. Done!! Compile Re-Started... 8:01:47am
  26. #ifdef QCOM_HARDWARE
    const char CameraParameters::KEY_SUPPORTED_HFR_SIZES[] = "hfr-size-values";
    #endif
    ====
    frameworks/av/camera/CameraParameters.cpp:31:54: error: 'const char android::CameraParameters::KEY_SUPPORTED_HFR_SIZES []' is not a static member of 'class android::CameraParameters'
     const char CameraParameters::KEY_SUPPORTED_HFR_SIZES[] = "hfr-size-values";
    
    Well, we are not passing on the "conditions"!! All of the vars complained about, concerning CameraParameters, are enclosed in a condition similar to the one above. The condition is set in qcom_flags in ../build/core/qcom_target.mk.
    ifeq ($(BOARD_USES_QCOM_HARDWARE),true)
    
        qcom_flags := -DQCOM_HARDWARE
        qcom_flags += -DQCOM_BSP
    
    Now then, the following flags get 'set' in: ../device/htc/msm8660-common/BoardConfigCommon.mk
    # Flags
    COMMON_GLOBAL_CFLAGS += -DQCOM_HARDWARE
    
    # QCOM hardware
    BOARD_USES_QCOM_HARDWARE := true
    TARGET_QCOM_DISPLAY_VARIANT := legacy
    TARGET_QCOM_MEDIA_VARIANT := legacy
    
    in both CM11 and CM12, which makes sense, since I copied from one to the other!! -BUT- we need to see how it gets called -or- accessed during the compile.
  27. ../build/core/combo/include/arch/linux-x86/AndroidConfig.h
    Lots of warnings about __GLIBC__ mainly cause it is NOT reading the following include for some reason..?.
        #include </usr/include/features.h>
    From  StackOverflow  : To find the include path. The standard way is as given below (for c and c++) include the grave accents:
        `gcc -print-prog-name=cc1` -v
            and
        `gcc -print-prog-name=cc1plus` -v
  28. February 24th, 2015 / 2:36am - In asking for help -and- trying to explain this to 'The Rev', I found that I did NOT copy over 'CameraParameters.h'. -But- I had copied over 'CameraParameters.cpp'.
  29. This was an attempt to get back the Wheel Animation.
    > git revert --no-edit b71b47d72b129915327f41653b06edaea5008534 || retval=1
    warning: Cannot merge binary files: prebuilt/common/bootanimation/800.zip (HEAD vs. parent of b71b47d... Bootanimation by Dima Skvarskyi)
    warning: Cannot merge binary files: prebuilt/common/bootanimation/768.zip (HEAD vs. parent of b71b47d... Bootanimation by Dima Skvarskyi)
    warning: Cannot merge binary files: prebuilt/common/bootanimation/720.zip (HEAD vs. parent of b71b47d... Bootanimation by Dima Skvarskyi)
    warning: Cannot merge binary files: prebuilt/common/bootanimation/600.zip (HEAD vs. parent of b71b47d... Bootanimation by Dima Skvarskyi)
    warning: Cannot merge binary files: prebuilt/common/bootanimation/540.zip (HEAD vs. parent of b71b47d... Bootanimation by Dima Skvarskyi)
    warning: Cannot merge binary files: prebuilt/common/bootanimation/480.zip (HEAD vs. parent of b71b47d... Bootanimation by Dima Skvarskyi)
    warning: Cannot merge binary files: prebuilt/common/bootanimation/360.zip (HEAD vs. parent of b71b47d... Bootanimation by Dima Skvarskyi)
    warning: Cannot merge binary files: prebuilt/common/bootanimation/320.zip (HEAD vs. parent of b71b47d... Bootanimation by Dima Skvarskyi)
    warning: Cannot merge binary files: prebuilt/common/bootanimation/240.zip (HEAD vs. parent of b71b47d... Bootanimation by Dima Skvarskyi)
    warning: Cannot merge binary files: prebuilt/common/bootanimation/1200.zip (HEAD vs. parent of b71b47d... Bootanimation by Dima Skvarskyi)
    warning: Cannot merge binary files: prebuilt/common/bootanimation/1080.zip (HEAD vs. parent of b71b47d... Bootanimation by Dima Skvarskyi)
    error: could not revert b71b47d... Bootanimation by Dima Skvarskyi
    hint: after resolving the conflicts, mark the corrected paths
    hint: with 'git add ' or 'git rm '
    hint: and commit the result with 'git commit'
    ctaylor@Tipper-i7:~/Android/cm12.0-rezound/vendor/cm
    > git status
    Not currently on any branch.
    You are currently reverting commit b71b47d.
      (fix conflicts and run "git revert --continue")
      (use "git revert --abort" to cancel the revert operation)
    
    Unmerged paths:
      (use "git reset HEAD ..." to unstage)
      (use "git add/rm ..." as appropriate to mark resolution)
    
            both modified:   prebuilt/common/bootanimation/1080.zip
            both modified:   prebuilt/common/bootanimation/1200.zip
            deleted by them: prebuilt/common/bootanimation/1600.zip
            both modified:   prebuilt/common/bootanimation/240.zip
            both modified:   prebuilt/common/bootanimation/320.zip
            both modified:   prebuilt/common/bootanimation/360.zip
            both modified:   prebuilt/common/bootanimation/480.zip
            both modified:   prebuilt/common/bootanimation/540.zip
            both modified:   prebuilt/common/bootanimation/600.zip
            both modified:   prebuilt/common/bootanimation/720.zip
            both modified:   prebuilt/common/bootanimation/768.zip
            both modified:   prebuilt/common/bootanimation/800.zip
    
    Untracked files:
      (use "git add ..." to include in what will be committed)
    
            get-prebuilts
    
    no changes added to commit (use "git add" and/or "git commit -a")
    


  1. Don't even remember what the following list was for -but- not gona delete it till I remember.
  2. https://github.com/chucktr/android_hardware_msm7k
  3. kernel/htc/vigor-3.0/include/asm-generic/Kbuild
        Fix missing asm/current.h error
      https://github.com/vigor/vigor_aosp_kernel/commit/a401a29dff48bb6bb31457f3d1534299e7461841
  4. repo sync -j1
  5. Makefile inside ../arch/arm/mach-msm/htc/vigor :
    obj-$(CONFIG_MACH_VIGOR) += board-vigor.o
    # 2/7/15 cmt obj-$(CONFIG_MACH_VIGOR) += board-vigor-gpiomux.o
    obj-$(CONFIG_MACH_VIGOR) += board-vigor-keypad.o
    obj-$(CONFIG_MACH_VIGOR) += board-vigor-mmc.o
    # 2/7/15 cmt obj-$(CONFIG_MACH_VIGOR) += board-vigor-storage.o
    obj-$(CONFIG_MACH_VIGOR) += board-vigor-wifi.o
    # 2/7/15 cmt obj-$(CONFIG_MACH_VIGOR) += board-vigor-regulator.o
    # 2/7/15 cmt obj-$(CONFIG_MACH_VIGOR) += board-vigor-pmic.o
    obj-$(CONFIG_MACH_VIGOR) += board-vigor-audio.o
    obj-$(CONFIG_MACH_VIGOR) += board-vigor-rfkill.o
    ifeq ($(CONFIG_MSM_CAMERA),y)
    # 2/7/15 cmt obj-$(CONFIG_MACH_VIGOR) += board-vigor-camera.o
    endif
    obj-$(CONFIG_MACH_VIGOR) += display/
    

  The 3.4 Kernel Build

Credits: Shaun Nuzzo ; Keith Kozma ; Bunchies XDA ; Beau (the Reven)
 Building a Kernel    building-the-android-kernel-on-linux    Doc:_integrated_kernel_building  

Today, with the help of Shaun Nuzzo, I was able to get the 3.4 kernel compiled on my Computer. It will not work without the drivers and HTC stuff -but- we got it compiled. Actually, Keith Kozma is the one that got it to work. Shaun just helped me bumble through the procedures to get it to compile on my machine.

 Pyramid vs Vigor    diff in board files  

The following is an edited, for clarification and some grammar, outline of what we did. There was lots of conversation before this and some may be injected later, but for now this is it.

  1. nvaigate in yoru web browser to Keith's 3.4 source
    you will want to fork that
    to your own account
  2. before you do any asking
    noticed keith has a new branch pushed
    WIP branch?
    yup
    Oh I see it in the arch.
  3. My fork:  https://github.com/chucktr/android_kernel_htc_msm8960  
    ok so let's add your fork as a new remote, were going to preserve the current one to pull any changes Keith does
    see the  HTTPS clone URL  on the right hand side
  4. > git remote -v
    origin https://github.com/vwmofo/android_kernel_htc_msm8960 (fetch)
    origin https://github.com/vwmofo/android_kernel_htc_msm8960 (push)
  5. please do but make sure you are documenting it as adding a forked tree
    and pull requests
    and pushing new branches
    and adding a new git remote
    and creating a local branch
    and git stash
  6. alright in your terminal, going to add a new git remote with the following its always going to be this method
    git remote add
    in this case we will call it this
    git remote add chuck https://github.com/chucktr/android_kernel_htc_msm8960.git
    but i want you to manually type all that and paste the clone url in
  7. wheres your 3.4 source at
    navigate to that, you remember /home/snuzzo/beer ?
    ~/vimsm8960/android_kernel_htc_msm8960/
  8. so where was I?
    getting a remote
  9. nope
    setting a new remote
    git remote add
  10. git remote add chuck https://github.com/chucktr/android_kernel_htc_msm8960.git
  11. did you do that?
    verify the above with git remote -v
  12. > git remote -v
    chuck https://github.com/chucktr/android_kernel_htc_msm8960.git (fetch)
    chuck https://github.com/chucktr/android_kernel_htc_msm8960.git (push)
    origin https://github.com/vwmofo/android_kernel_htc_msm8960 (fetch)
    origin https://github.com/vwmofo/android_kernel_htc_msm8960 (push)
  13. ok before you go any further were going to get
    the new branches
    only need to do this once
    but before you do your going to stash your changes
    with
    git stash
    dont worry we can get them back
  14. basically this saves any changes you made without staging them as a commit
    just a "git stash"
    but makes your git directory clean
    yup
    and what i mean by clean is it will revert back to no changes done to the latest commit
  15. > git stash
    Saved working directory and index state WIP on cm-12.0-34: cc5b8ad Vigor build script
    HEAD is now at cc5b8ad Vigor build script
  16. you got it
    now do git fetch chuck
  17. > git fetch chuck
    remote: Counting objects: 46, done.
    remote: Compressing objects: 100% (36/36), done.
    remote: Total 46 (delta 16), reused 25 (delta 10)
    Unpacking objects: 100% (46/46), done.
    From https://github.com/chucktr/android_kernel_htc_msm8960
    * [new branch] cm-12.0-34 -> chuck/cm-12.0-34
    * [new branch] cm-12.0WIP -> chuck/cm-12.0WIP
    * [new branch] lollipop -> chuck/lollipop
  18. say do git checkout cm-12.0-34
    next do
    git checkout cm-12.0WIP
    then git checkout lollipop
    to verify do "git branch"
  19. > git checkout cm-12.0-34
    Already on 'cm-12.0-34'
    Your branch is up-to-date with 'origin/cm-12.0-34'.
  20. > git checkout cm-12.0WIP
    Branch cm-12.0WIP set up to track remote branch cm-12.0WIP from chuck.
    Switched to a new branch 'cm-12.0WIP'
  21. > git checkout lollipop
    error: pathspec 'lollipop' did not match any file(s) known to git.
  22. wonder why that errored
    try
    git checkout chuck/lollipop
  23. > git checkout chuck/lollipop
    Note: checking out 'chuck/lollipop'.

    You are in 'detached HEAD' state. You can look around, make experimental
    changes and commit them, and you can discard any commits you make in this
    state without impacting any branches by performing another checkout.

    If you want to create a new branch to retain commits you create, you may
    do so (now or later) by using -b with the checkout command again. Example:

    git checkout -b new_branch_name

    HEAD is now at acac1ff... introduce SIZE_MAX
  24. ok do as it says
    git checkout -b lollipop
    there was a small mishap, that shouldnt have happenned
  25. > git checkout -b lollipop
    Switched to a new branch 'lollipop'
  26. once done veriy all branches are intact with "git branch"
  27. > git branch
    cm-12.0-34
    cm-12.0WIP
    * lollipop
  28. ok now to switch branches you would do the same
    switch to the cm-12.0WIP branch
    Shaun
    Sorry ... what do you mean to switch branches.
    remember, you can have different revisions of the same source, thats what branches are for
    here let me help clarify
    ok ... will I find a new dir -or- is it the same one and I am switching where I get the code??
    Now
    Chuck Taylor
    Izzae Roth
    Shaun Nuzzo
    no hold on let me clarify
    ok online you know how you forked
    chucktr/android_kernel_htc_msm8960
    yes...
    think of that as a "tree"
    k
    same "trunk" same "root"
    and it "branches" out
    good analogy?
    so when we say "root" of your source
    you now know what we mean
    right now your in the "root" hanging out on a "branch"
  29. say we want to swing from a different "branch"
  30. so lets "checkout" that "branch"
  31. Ok at the moment I am in "my" 3.4 dir on "my" PC. The outside world is whereever.
    I presume you are calling that my root
  32. you got it ??
    Shaun
    if i need to refer to the remotesies ill let you know
  33. i think you need to checkout to a WIP branch
  34. That why was not in the right place. It was for why we need to switch.
  35. the git fetch and the git checkouts were to set up your root source without
    having to fully init and pull it down all over again
    so we utilized your current source root that was initially a git clone
    of essentially the same code and revision
    saved you a little time downloading
  36. so again checkout to cm-12.0WIP
  37. > git checkout cm-12.0WIP
    Switched to branch 'cm-12.0WIP'
    Your branch is up-to-date with 'chuck/cm-12.0WIP'.
  38. git stash branch will apply them and checkout to that branch name without them staged
    I think we can just drop them and IF they are needed I can put them in this code. I have my notes.
    git stash drop will abandon them
    just a "git stash drop" ??
    yup
  39. if you keep them present tho you can create a new branch with them intact and stage them if you wish that was going to be next teaching anyways
    Shaun
    k I'll keep them for the moment...
    Chuck Taylor
    Shaun Nuzzo
    see youre ahead of me a little ??
  40. ok ... let me check the compile ... back in a little.
    3 mins
    Chuck Taylor
    Shaun Nuzzo
    ??
    have fun, you know prety much alot of the goodies with git
    if done properly you can do everything now
  41. yw
    here let me help you out further
    Shaun
    1 min
    Shaun Nuzzo
    you know when you sync ?
    Izzae Roth
    Shaun Nuzzo
    you are syncing specific branches of specific trees placed in specific dirs
    in theory, you could go in, create a branch
    in any of those dirs
    make some changes and build there
  42. let's say you only wish to build the Settings app
    at the top of your source you would lunch your device
  43. then "make -j# Settings"
    Shaun
    to elaborate further
    Shaun
    make -j#
    just throwing that out there for extra info
  44. for all intents and purposes "repo sync" basically just does a git pull for a specific project and for each project listed in manifest
  45. Hey guess what?? It compiled with only two warnings and I have a 4.3MB kernel that claims it was a successful build!!
    2 mins
    Chuck Taylor
    Shaun Nuzzo
    now to figure out what binaries to use to pair with it
    gotta have driver binaries aka vendor blobs to run the kernel and interface with the ramdisk and rom
  46. ok say Keith makes and uploads a few changes to cm12.0-WIP branch
    "git pull origin cm12.0-WIP" for example
    same item as you did the other day
    BUT here's where its tricky
    you can then push to your remote those changes if you wish
    with "git push chuck cm12.0-WIP"
  47. So these will be "out" on the github and I will then have to do a fetch or a checkout??
    "push" and "pull" Which is to my PC??
  48. no need to fetch nor checkout
    pull will do exactly that
    pull down changes of a specific branch
    push will do exactly what it says as well
    Shaun
    1 min
    Bunchies xda
    Ok ... so I take that answer as "pull" will change the code on my PC and "push" will change the code up on the github... correct??
    Now
    Chuck Taylor
    Shaun Nuzzo
    in theory yes
    theres also another way to bring over his changes
    and thats strictly online through a reverse pull request
    but i will show you that later since youre currently up to date
  49. Now that we have blubbered all over on another ... anyone got any suggestions on which device to swipe the drivers and HTC code from??
  50. pyramid/shooter
  51. probably get the current stuff that uses that same 3.4 source
  52. For the rest of you the stuff I am talking about is at:
    https://github.com/orgs/vigor/dashboard
  53. i found this device tree:
    https://github.com/ivanich/android_device_htc_pyramid
    Izzae
    Looks promising ... except I did not see a vendor or htc dir.
    Just a sec and I will clone it.
    3 mins
    Chuck Taylor
    Shaun Nuzzo
    https://github.com/ivanich/proprietary_vendor_htc
    https://github.com/ivanich/proprietary_vendor_htc/tree/cm-12.0/pyramid
    https://github.com/ivanich/android_kernel_htc_pyramid
  54. =========================================
  55. Copied htc dir from CM11/device to CM12/device
  56. Copied htc dir from CM11/vendor to CM12/device
  57. https://github.com/chucktr/android_kernel_htc_msm8960
  58. https://github.com/vwmofo/android_kernel_htc_msm8960/tree/cm-12.0WIP
  59. http://wiki.cyanogenmod.org/w/Devices#vendor=;
    http://wiki.cyanogenmod.org/w/Device_Status
    is that what you wanted?
  60. =========================================
    The following are changes I am making to the 3.4 kernel compile to see
    IF I can get it to work with the CM11 code.
    =========================================
  61. Cp from vigor ../arch/arm/plat-spear/include/plat/ to vimsm8960 ../arch/arm/plat-spear/include/plat/ :
        clkdev.h
        io.h
        memory.h
        system.h
        vmalloc.h
  62. Copied from vigor ../include/linux/input/ to vimsm8960 ../include/linux/input/ :
        pmic8058-keypad.h
  63. Created pmic8058-pwrkey.h in : ../include/linux/input/
  64. Moved the above to ../include/linux/
  65. Copied sysinfo-8x60.h from vigor ../arch/arm/mach-msm/ to vimsm8960 ../arch/arm/mach-msm
  66. Moved tps65200.h from ../include/linux/mfd/ up to ../include/linux/
  67. The following board files are in: ../arch/arm/mach-msm/htc/vigor/
  68. changed include in board-vigor-mmc.c from "proc_comm.h" to <mach/proc_comm.h>
  69. Changed include in board-vigor.c from <mach/tpa2051d3.h> to <linux/tpa2051d3.h>
  70. Changed include in board-vigor.c from "cpuidle.h" to <mach/cpuidle.h>
  71. Changed include in board-vigor.c from "mpm.h" to <mach/mpm.h>
  72. Copied htc_util.h from vigor ../arch/arm/mach-msm/include/mach/ to vimsm8960 ../arch/arm/mach-asm/include/mach/
  73. Copied update-binary from vigor_aosp_kernel to android_kernel_htc_msm8960
  74. Copied prebuilt-scripts dir from vigor_aosp_kernel to android_kernel_htc_msm8960
  75. Edited build-vigor.sh in android_kernel_htc_msm8960 to use my ToolChain.
  76. Executed build-vigor.sh and built the kernel with the Packages.
  77. Installed it on my phone inside CM11.
  78. These two warnings are cause of changes in the behavior of -Wall.
    The behavior of -Wall has changed and now includes the new warning flag -Wsizeof-pointer-memaccess. This may result in new warnings in code that compiled cleanly with previous versions of GCC.
      CC      drivers/video/msm/mipi_truly.o
    drivers/video/msm/mipi_novatek.c: In function 'mipi_dsi_3d_barrier_read':
    drivers/video/msm/mipi_novatek.c:569:37: warning: argument to 'sizeof' in 'snprintf' call
     is the same expression as the destination; did you mean to provide an explicit length? [-Wsizeof-pointer-memaccess]
      return snprintf((char *)buf, sizeof(buf), "%u\n", barrier_mode);
                                         ^
      CC      drivers/video/msm/mipi_NT35510.o
      CC      drivers/video/msm/mipi_truly_tft540960_1_e.o
      CC      drivers/video/msm/mipi_NT35590.o
      CC      drivers/video/msm/mipi_simulator.o
      CC      drivers/video/msm/mipi_tc358764_dsi2lvds.o
      CC      drivers/video/msm/msm_fb_panel.o
      CC      drivers/video/msm/mipi_toshiba_video_wvga_pt.o
      CC      drivers/video/msm/mipi_toshiba_video_wsvga_pt.o
      CC      drivers/video/msm/mipi_toshiba_video_wuxga.o
      CC      drivers/video/msm/mipi_novatek_video_qhd_pt.o
    drivers/video/msm/mipi_tc358764_dsi2lvds.c: In function 'mipi_d2l_enable_3d_read':
    drivers/video/msm/mipi_tc358764_dsi2lvds.c:746:37: warning: argument to 'sizeof' in 'snprintf' call
     is the same expression as the destination; did you mean to provide an explicit length? [-Wsizeof-pointer-memaccess]
      return snprintf((char *)buf, sizeof(buf), "%u\n", d2l_enable_3d);
                                         ^
      CC      drivers/video/msm/mipi_novatek_cmd_qhd_pt.o
    
    
  79. https://gitorious.org/ac100/abootimg/source/7e127fee6a3981f6b0a50ce9910267cd501e09d4:
  80. https://wiki.linaro.org/android/DisplayHotplug
  81. https://www.google.com/search?lr=&hl=en&as_qdr=all&tbm=isch&imgil=8lUoSPINRXRsZM%253A%253Bj0FXB8kSSHBYJM%253Bhttp%25253A%25252F%25252Fwww.programering.com%25252Fa%25252FMTNwQDMwATg.html&source=iu&pf=m&fir=8lUoSPINRXRsZM%253A%252Cj0FXB8kSSHBYJM%252C_&usg=__0UHM6ws83HIB0dsI4w-DerPIqII%3D&biw=1037&bih=679&ved=0CFwQyjc&ei=8yTaVID7EK20sAScnYHYAQ&q=surfaceflinger+flowchart#imgdii=_
3 commands to learn
You already know reset
log
status
And
Revert
If youre going to make code changes
branch
add and rm
And commit
repo sync external/sepolicy
That will get it back to original state
====
ctaylor@Tipper-i7:~/vimsm8960/android_kernel_htc_msm8960
> git remote -v
origin  https://github.com/vwmofo/android_kernel_htc_msm8960 (fetch)
origin  https://github.com/vwmofo/android_kernel_htc_msm8960 (push)
ctaylor@Tipper-i7:~/vimsm8960/android_kernel_htc_msm8960
> git branch
* cm-12.0-34
ctaylor@Tipper-i7:~/vimsm8960/android_kernel_htc_msm8960
> git branch -v
* cm-12.0-34 09d9957 Initial import of HTC Vigor
ctaylor@Tipper-i7:~/vimsm8960/android_kernel_htc_msm8960
> git pull origin  cm-12.0-34
remote: Counting objects: 19, done.
remote: Compressing objects: 100% (15/15), done.
remote: Total 19 (delta 5), reused 12 (delta 4)
Unpacking objects: 100% (19/19), done.
From https://github.com/vwmofo/android_kernel_htc_msm8960
 * branch            cm-12.0-34 -> FETCH_HEAD
   09d9957..cc5b8ad  cm-12.0-34 -> origin/cm-12.0-34
Updating 09d9957..cc5b8ad
error: Your local changes to the following files would be overwritten by merge:
        arch/arm/configs/vigor_aosp_defconfig
Please, commit your changes or stash them before you can merge.
Aborting
ctaylor@Tipper-i7:~/vimsm8960/android_kernel_htc_msm8960

  Installing 3.4 in CM11

 February 15th, 2015 /4:23am   Bad idea. It didn't work and there are to many changes. The better idea would be to try to get it to work in CM12. So I am switching to that.

Decided to build the kernel inside of CM11. Found some Cyanogen stuff that talks about installing a custom kernel, so we will try it. First thing was to create a new dir called:
    ../kernel/htc/vigor-3.4
the previous one, vigor-3.0 is still there... for now.

Then we searched for all the places where the "vigor-3.0" dir was referenced. Only two places: BoardConfigCommon.mk
# Kernel
TARGET_KERNEL_SOURCE := kernel/htc/vigor-3.4
I changed the 3.0 to 3.4.
and: liquid.dependencies
{
"remote": "gh",
"repository": "vigor/vigor_aosp_kernel",
"target_path": "kernel/htc/vigor-3.0",
"revision": "cm-11.0_hybrid"
},

First one I believe is important. The second, I think is only needed IF you are building for a ROM callled "Liquid". But, what do I know?? Not a lot yet. Because of that word "remote", I am leaving it alone for now.

Now then, the 'top level' dirs for these kernels are: ../kernel/htc/vigor-3.x. The 'x', naturally, will bounce between '0' and '4', depending on what we are talking about. Underneath the vigor-3.x dirs is a collection of dirs that are all the same in both. At least they all have the same name at this level.

I do believe that BoardConfigCommon.mk is what will link this kernel to Cyanogen. And I am not naieve enough to believe that changing the version number will make it all work. But, as I always say, ya gota start somewhere. So, we will make the change and fire off a build.

Didn't work. Looked like it might ... but then it failed miserably. Like the note at the top says, there are so many changes, that it is best to just jump into CM12 and be done with it. So that is what we will do.

   

  Blend

An H3 Followed by a Blockquote:

“Donec leo, vivamus nibh in augue at urna congue rutrum. Quisque dictum integer nisl risus, sagittis convallis, rutrum id, congue, and nibh.”

Bulleted List:

  • Lorem ipsum dolor sit amet, consectetuer adipiscing elit.
  • Phasellus nec erat sit amet nibh pellentesque congue.
  • Cras vitae metus aliquam risus pellentesque pharetra.

Numbered List:

  1. Lorem ipsum dolor sit amet, consectetuer adipiscing elit.
  2. Phasellus nec erat sit amet nibh pellentesque congue.
  3. Cras vitae metus aliquam risus pellentesque pharetra.

   

  ICS vs JB vs KitKat vs Lollipop - Feature comparison/Usage

Now then, I started this with great expectations in KitKat but did not get finished. Hopefully, this time we will... but don't hold your breath.

ICSJBKitKatLollipop
  1. ICS Col
  2. Camcorder works
  3. Theoretically Slowest
  1. JB Col
  2. Camcorder no worky
  3. introduced Google Now
  4. Faster than GB
  1. KitKat Col
  2. Camcorder no worky
  3. Theoretically Fastest
  1. Lollipop Col
  2. Hopefully faster ...