Android P development

[Spookcity]

Place to discuss development of Android Pie for the G800F/M/Y

[andrea224]

Yes, I did every time "mka clean" just for being sure. However, even with "mka bootimage" it gaves me errors:

[ 10% 899/8201] Building Kernel Config
FAILED: /home/andrea/android/los16/out/target/product/kminilte/obj/KERNEL_OBJ/.config 
/bin/bash -c "(make -j4  CFLAGS_MODULE=\"-fno-pic\" -C kernel/samsung/kminilte O=/home/andrea/android/los16/out/target/product/kminilte/obj/KERNEL_OBJ ARCH=arm CROSS_COMPILE=\" arm-linux-androidkernel-\"   VARIANT_DEFCONFIG= SELINUX_DEFCONFIG= kminilte_00_defconfig ) && (if [ ! -z \"\" ]; then          echo \"Overriding kernel config with ''\";          echo  >> /home/andrea/android/los16/out/target/product/kminilte/obj/KERNEL_OBJ/.config;          make -j4 -C kernel/samsung/kminilte O=/home/andrea/android/los16/out/target/product/kminilte/obj/KERNEL_OBJ ARCH=arm CROSS_COMPILE=\" arm-linux-androidkernel-\"   oldconfig; fi ) && (make -j4 -C kernel/samsung/kminilte O=/home/andrea/android/los16/out/target/product/kminilte/obj/KERNEL_OBJ ARCH=arm CROSS_COMPILE=\" arm-linux-androidkernel-\"   savedefconfig ) && (if [ ! -z \"\" ]; then       echo \"Using additional config ''\";            kernel/samsung/kminilte/scripts/kconfig/merge_config.sh -m -O /home/andrea/android/los16/out/target/product/kminilte/obj/KERNEL_OBJ /home/andrea/android/los16/out/target/product/kminilte/obj/KERNEL_OBJ/.config kernel/samsung/kminilte/arch/arm/configs/;         make -j4 -C kernel/samsung/kminilte O=/home/andrea/android/los16/out/target/product/kminilte/obj/KERNEL_OBJ ARCH=arm CROSS_COMPILE=\" arm-linux-androidkernel-\"   KCONFIG_ALLCONFIG=/home/andrea/android/los16/out/target/product/kminilte/obj/KERNEL_OBJ/.config alldefconfig; fi )"
make: ingresso nella directory "/home/andrea/android/los16/kernel/samsung/kminilte"
  GEN     /home/andrea/android/los16/out/target/product/kminilte/obj/KERNEL_OBJ/Makefile
  HOSTCC  scripts/basic/fixdep
/home/andrea/android/los16/kernel/samsung/kminilte/scripts/basic/fixdep.c:106:10: fatal error: sys/types.h: File o directory non esistente
 #include <sys/types.h>
          ^~~~~~~~~~~~~
compilation terminated.
scripts/Makefile.host:118: recipe for target 'scripts/basic/fixdep' failed
make[2]: *** [scripts/basic/fixdep] Error 1
/home/andrea/android/los16/kernel/samsung/kminilte/Makefile:414: recipe for target 'scripts_basic' failed
make[1]: *** [scripts_basic] Error 2
Makefile:130: recipe for target 'sub-make' failed
make: *** [sub-make] Error 2
make: uscita dalla directory "/home/andrea/android/los16/kernel/samsung/kminilte"
[ 11% 908/8201] //external/protobuf:ap...piler/cpp/cpp_message.cc [linux_glibc]
ninja: build stopped: subcommand failed.
18:09:09 ninja failed with: exit status 1

#### failed to build some targets (01:06 (mm:ss)) ####

[Spookcity]

Yeah, its the fixdep error. I've had it before. It is because of a dirty kernel directory. It really should be solved with a "make mrproper" or a clean kernel sync. I'm not sure what else the cause could be, sorry.

[andrea224]

Don't worry, I'll do it again and see if I'll end somewhere good. If not then I'll wait. (luckily it isn't my main phone from years)

[andrea224]

@Spookcity I had to update the GCC, now it works...

[Spookcity]

@andrea224 Ahh,that's something I never would have thought about. Glad you got it working!

[rodman01]

Big activities here the last view hours :-) Great and gongrats that you solved the problem. Just for your information @andrea224 do not forget to run the patch script in device common, because I sometimes forgot it :-) and without them the device most probably won't boot or run into issues...just my thoughts :-)

[andrea224]

@rodman01 Yes, usually I don't forget it, but thanks for the tip.

[andrea224]

@Spookcity @rodman01 finally it's built and it works as it should so far, here's the link if you want to use/test or whatever do you want to do with it: https://drive.google.com/open?id=16RbXNnkrHCaME6I9jXB1bfjuAtguR5fb

Thanks again both for helping me.

[rodman01]

Hi @Spookcity good morning I have cherry picked the netfilter commits from haggertys klte kernel repo and now wifi tethering can be enabled and a device can be connected. But with enabled wifi tethering internet is now not working :-(. Neither on the tethering device nor with the connected device. Maybe I did something wrong or missed a pick or picked to much? Hope this helps anyways to find the right solution. Also because I am not really sure how it is supposed to work: with enabled mobile data, or without? I also read (keyword: enable DUN tethering "tether_dun_required"?!?) that dun has to be put to AP point entries? If so I haven't found the right place in sources to include it already in the build. The binder pick (found in review lineageos) in the repo what I have read could probably solve the video issue with instagram and whatsup, but can also effect and do nothing :-) not sure: https://github.com/rodman01/android_kernel_samsung_kminilte/tree/P_Tether_KLTE

btw I found a typo in your vendor repo, doubled // next to vendor, but correcting and from my testing, it changes nothing anyways :-)

[Spookcity]

Hey @rodman01 thats great news! I appreciate you working on this.Hopefully I'll have some time this week and I'll look through your kernel repo and pick some commits. I don't think I'v ever used the hotspot feature..or tethering in general really. So I am rather unsure how it is suppose to work as well. Either way,big step forward. Thanks a lot! Also DUN tethering can be added via the overlay/frameworks/base/core/res/res/values/config.xml in kminilte repo. I actually made This commit the other day. I dont think its needed any longer,but it may be worth looking into adding back to see if it helps matters. Thanks also for pointing out the mistake in the vendor repo. I will fix it.

[rodman01]

cool thanks think I will test the settings in your commit tomorrow. Problem is that i do have any mbs or credits left on my prepaid test sim card :-). Ah forgot to mention that for tethering I also added the hostapd hidl to hidl manifest, although I am not sure that this has an effect :-) According to bluetooth: In my tstes with my linux notebook, connecting and sending a file works. Receiving a file end in an error. Message: storage problem. I guess its a permission problem or a mount/connecting problem with internal and/or external storage. The file I tried to receive had been already "created" in bluetooth folder, but with 0 bytes. file name and .txt or.pdf is shown. I tried some changes in intit.universal3470.rc (changed symlink and permissions), but without success. Currently i am out of ideas, what I can try additionally. Maybe you can point me to a new direction? I would guess that it is a permission problem, but only a guess. I also added the selinux from msm8974-common repo, but from my point of view without any noticeable changes. And last nfc: I have also tried a lot without success or any changes beside the current state broke again :-). Also here I am unfortunately out of ideas. What I have not tried yet, is to use the prebuilt nfc blob same which haggerty did for the klte. Maybe this could bring us a step forward. Other than that do you have some more topics or bugs, I can try to find a solution for? Because to be honest, for the 3 current ones, I am a bit out of ideas and need some other things to do in between :-). Maybe you have something else to check or to test?

[jimmy999x]

since I sold the s5 mini, I have gone through a motorola moto g4 plus, galaxy note 4, and now I am in a oneplus 2 (I love it but it overheats) it is possible to sell it next month and buy some xiaomi, however to the end of January I will try to buy a s5 mini again as a second device, I miss you guys

[Spookcity]

@rodman01 Thanks a lot for the detailed info. I am curious if receiving files via bluetooth works in my test build It seems like this may very possibly also be caused by sdcardfs problems. I reverted back to fuse file system for that build just to see if sdcardfs was causing some of the storage/permission issues. It seems that it is indeed. So we are missing some commits relating to sdcardfs in the kernel. I will look at @friedensfurz kernel repo when I have time.They may be in there.

I am pretty much where you are with NFC. I have no idea. I think our only option is going to be to try and use a prebuilt from LOS 15.1. I am not sure when (or if) our NFC chipset will work properly again.

The only other major issue that I am aware of is the HW overlays. This is another problem I am about out of ideas for. The log shows SurfaceFlinger errors,but caused by libdisplay.so if I remember correctly. So perhaps the only fix is going to be a hex edit of our libdisplay.so,but I am not really sure where to start with that. I have not come across any other devices that have this issue though.

@jimmy999x Wow you are def going through some devices. I hope you are able to get an S5 mini again and come back and help us out and build some ROMS! Hope all is well!

Wishing a very happy 2019 to @rodman01 @jimmy999x @friedensfurz @IamEld3st @aelfinn @Mikahlo and anyone else I may have forgotten who is around these parts.

[jimmy999x]

equally! have a happy end of the year and that this 2019 is better than 2018, success in all

[rodman01]

Happy new year to all of you from my side too :-) 👍 hope your start today night had been good and the new year will bring you health, success and luck :-).

@Spookcity I have just tested your new test build and yes receiving a file via bluetooth is now possible with it. Cool and Gongrats! So the cause of the bluetooth/file problem seems to be located :-) Great!

Regarding tethering I have tested your commit for overlay yesterday, but it changed nothing. I guess, its supposed to work (only?) with mobile data on. Maybe you can ask this in the xda thread, if you want to add the netfilter commits to your kernel too and release a new test build or even only a new test kernel, somewhen....?

About surfaceflinger: haven't noticed the error in log yet, so I will search for it and can try to find a solution, but if hex editing is necessary my knowledge is very limited :-). I tried it once for my i9305, I think it was for ril, and just to verify the already done changes from a dev and see how this would/could be done. That was my only experience with this as for now :-)

[Spookcity]

@rodman01 Thanks a ton for trying out Bluetooth on the test build. Thats super helpful. It pretty much lets us know that all of the storage/permission problems are related to sdcardfs. This should make it a lot easier to fix. As opposed to trying to figure out exactly what the issue is. It must be some commits relating to sdcardfs in the kernel. If I have some time this week I will look over @friedensfurz kernel repo and see if we are missing some related commits. I don't want to revert back to fuse permanently if avoidable..and it hopefully should be now with this new info.

If I have time this week I will pick the netfilter commits from your kernel repo and do a build.

I'm pretty much the same when it comes to hex editing. I have little experience with it. I opened up our libdisplay.so a couple of months ago and searched for what I believed the errors were,but I didn't really know where to go from there. I would have no clue what to change them to. I guess I could just try to delete all of the offending lines. There were a lot though. Worse case it just doesn't work though,so I will try that when I have the time just to see.

[friedensfurz]

@Spookcity These should be the missing sdcardfs commits

Maybe there are some more, indirectly related and/or not named after "sdcardfs", but the ones above should definetly be included. Enjoy ;)

[rodman01]

Thanks Paul for the info and the commits. Some of them had been already included in Spooks Kernel, others not. So I have filtered the lineageos msm8974 kernel repo according to ANDROID: sdcardfs and cherry picked all commits since middle of 2016. Receiving files via bluetooth is now also possible. Whether the camera problem is still there after choosing the sd card as storage in camera setting is still there, I have to check, currently I cannot find the commit which makes that setting in the camera app visible. What else need or could be tested, to check whether all storage problems are solved now? Beside the whatsupp and other social media thingis I am not using with my test devise and sim card :-). And last: whats with the gps fix problem? Is there a fix in your minds? Or is testing, researching necessary? Also switching to KLTEs sources? Here is my kernel repo with tether and sdcardfs cherries: https://github.com/rodman01/android_kernel_samsung_kminilte/tree/P_SDcardfs_KLTE. I guess not all of them are necessary, but I hope they won't hurt anyways :-)

Edit: ok found the patches and error message: insert sd card before using the setting is still there and camera doesn't work anymore. Only after deleting cache and settings of the app is disabling the external sd card as storage setting. Not sure whether it is still a sdcard kernel problem or an app problem. # packages/apps/Camera2 repopick 224752 # Use mCameraAgentNg for getting camera info when available repopick 225265 # Add Storage preference (1/2)

Edit2: have I mixed something up? Or is still something sdcard related missing? 01-01 01:02:49.558 1972 1972 V vold : Waiting for FUSE to spin up... 01-01 01:02:49.592 2445 2445 D SystemUIBootTiming: StartServicescom.android.systemui.stackdivider.Divider took to complete: 172ms 01-01 01:02:49.609 1972 1972 V vold : Waiting for FUSE to spin up... 01-01 01:02:49.652 2551 2551 W SystemConfig: No directory /product/etc/sysconfig, skipping 01-01 01:02:49.652 2551 2551 W SystemConfig: No directory /product/etc/permissions, skipping 01-01 01:02:49.659 2746 2746 W sdcard : Device explicitly enabled sdcardfs 01-01 01:02:49.660 1972 1972 V vold : Waiting for FUSE to spin up...

[rodman01]

Here is a parted logcat with camera logs, while I "played" with the storage settings in camera app. Cannot see any related entries?!? logcat.txt

[Spookcity]

@friedensfurz Thanks a lot for that info. i checked yesterday and unfortunately I have all of those commits in my repo as far as I could tell. So I guess we will need to dig a bit deeper.

@rodman01 Are you using my "test build" for these sdcardfs changes? It seems not judging by the "Device explicitly enabled sdcardfs" message,but there are also messages about waiting for fuse to spin up,so I am a bit confused. I disable sdcardfs on the test build and re-enabled fuse. I will have a look at your sdcardfs repo hopefully a bit later because I have to assume if receiving files via Bluetooth is working for you then you are def on the right path! Also,I picked your netfilter commits yesterday. I only built the kernel. I didn't do a clean build,but indeed the hotspot now stays on. I connected to adb shell and there is still some logcat spam regarding WiFiVendorHal though,so I am not sure its fully working. Either way,this is a huge step forward so many thanks!

The only message in your log that stands out to me that seems to be new is "E ExynosCameraFrameFactory: Instance On wait Frame count(1) numFrames(14)" I cant really say whats thats about though honestly.

[rodman01]

First I only compiled the kernel with the sdcardfs commits to see whether it builds and is working. I tested it on your test rom, but haven't taken a log file. After that I added some overlay commits and the camera commits mentioned above and compile a new rom with sources approx 1 week old (29.12). So there should have your latest repo updates included. I guess you haven't updated your repos with the changes done for your test build? From this rom I took the log file and I was also confused about the fuse messages ...

[Spookcity]

@rodman01 Yeah, I didn't push the changes for the test build as I hope it to be temporary. All I did in that build was disable sdcardfs in the kminilte_00_defconfig and applied This patch. So your build is using sdcardfs. The fuse messages are a bit confusing indeed.

Edit: Also,just reading over your previous message again....I am not using whatsapp or any other app really to test the storage issues. I have to assume though that if recieving files via BT works now that these things prob will also. As it all seems related. And regarding GPS. I have not had any issues in limited use. I was reading through the XDA thread,it seems to work for some and not others,so I am a bit unsure where to look.

[rodman01]

OK fine many thanks for the info. Then we have that libdisplay and nfc topic left ... maybe I will take another look at it and if you have more or other things to try or to test, let me know and I will assisit :-)

[Spookcity]

@rodman01 Thanks a ton for helping with these issues!

Yeah,NFC and the HW overlays are the 2 remaining big issues as I see it.

NFC - Going to have to use a prebuilt I guess. I have not looked into this at all,so I'm not really sure what files would need to be copied over from an Oreo ROM.

HW Overlays - This script creates it so that they are always disable upon boot. The only way to test this is to either manually enable overlays in dev options or to do a build without the script included. I found a log from a couple of months ago regarding this issue. Again I am by no means certain its strictly a libdisplay error. If I had to guess on where the fix is,it would be in our android_hardware_samsung_slsi_exynos3470 repo. More specifically,prob the hwc.cpp file. These are just guess though honestly. I will include the log in case you have some time and feel like looking into it. I've been unable to determine a fix.

11-15 15:22:13.981 5440 5440 F libc : Fatal signal 11 (SIGSEGV), code 1 (SEGV_MAPERR), fault addr 0xff000000 in tid 5440 (surfaceflinger), pid 5440 (surfaceflinger) 11-15 15:22:13.886 5523 6769 I chatty : uid=1000(system) BackgroundDexOp identical 3 lines 11-15 15:22:13.976 5523 6769 I system_server: The ClassLoaderContext is a special shared library. 11-15 15:22:14.033 6771 6771 I crash_dump32: type=1400 audit(0.0:1197): avc: denied { read } for name="online" dev="sysfs" ino=34 scontext=u:r:crash_dump:s0 tcontext=u:object_r:sysfs:s0 tclass=file permissive=1 11-15 15:22:14.033 6771 6771 I crash_dump32: type=1400 audit(0.0:1197): avc: denied { open } for name="online" dev="sysfs" ino=34 scontext=u:r:crash_dump:s0 tcontext=u:object_r:sysfs:s0 tclass=file permissive=1 11-15 15:22:14.034 5678 5678 D OpenGLRenderer: Skia GL Pipeline 11-15 15:22:14.033 6771 6771 W crash_dump32: type=1300 audit(0.0:1197): arch=40000028 syscall=322 per=800008 success=yes exit=3 a0=ffffff9c a1=a8760e50 a2=a0000 a3=0 items=1 ppid=5440 auid=4294967295 uid=1000 gid=1003 euid=1000 suid=1000 fsuid=1000 egid=1003 sgid=1003 fsgid=1003 tty=(none) ses=4294967295 exe="/system/bin/crash_dump32" subj=u:r:crash_dump:s0 key=(null) 11-15 15:22:14.055 5523 6769 I system_server: The ClassLoaderContext is a special shared library. 11-15 15:22:14.107 5523 6769 I chatty : uid=1000(system) BackgroundDexOp identical 1 line 11-15 15:22:14.120 5523 6769 I system_server: The ClassLoaderContext is a special shared library. 11-15 15:22:14.125 6772 6772 W crash_dump32: failed to fetch registers for thread 5445: I/O error 11-15 15:22:14.126 6772 6772 W crash_dump32: failed to fetch registers for thread 5446: I/O error 11-15 15:22:14.126 6772 6772 W crash_dump32: failed to fetch registers for thread 5447: I/O error 11-15 15:22:14.127 6772 6772 W crash_dump32: failed to fetch registers for thread 5448: I/O error 11-15 15:22:14.127 6772 6772 W crash_dump32: failed to fetch registers for thread 5449: I/O error 11-15 15:22:14.127 6772 6772 W crash_dump32: failed to fetch registers for thread 5467: I/O error 11-15 15:22:14.128 6772 6772 W crash_dump32: failed to fetch registers for thread 5468: I/O error 11-15 15:22:14.128 6772 6772 W crash_dump32: failed to fetch registers for thread 5469: I/O error 11-15 15:22:14.128 6772 6772 W crash_dump32: failed to fetch registers for thread 5485: I/O error 11-15 15:22:14.129 6772 6772 W crash_dump32: failed to fetch registers for thread 5487: I/O error 11-15 15:22:14.129 6772 6772 W crash_dump32: failed to fetch registers for thread 5488: I/O error 11-15 15:22:14.130 6772 6772 W crash_dump32: failed to fetch registers for thread 5489: I/O error 11-15 15:22:14.130 6772 6772 W crash_dump32: failed to fetch registers for thread 5490: I/O error 11-15 15:22:14.131 6772 6772 W crash_dump32: failed to fetch registers for thread 5491: I/O error 11-15 15:22:14.132 6772 6772 W crash_dump32: failed to fetch registers for thread 5492: I/O error 11-15 15:22:14.132 6772 6772 W crash_dump32: failed to fetch registers for thread 5493: I/O error 11-15 15:22:14.132 6772 6772 W crash_dump32: failed to fetch registers for thread 5494: I/O error 11-15 15:22:14.133 6772 6772 W crash_dump32: failed to fetch registers for thread 5502: I/O error 11-15 15:22:14.133 6772 6772 W crash_dump32: failed to fetch registers for thread 5504: I/O error 11-15 15:22:14.134 6772 6772 W crash_dump32: failed to fetch registers for thread 5598: I/O error 11-15 15:22:14.134 5523 6769 I system_server: The ClassLoaderContext is a special shared library. 11-15 15:22:14.134 6772 6772 W crash_dump32: failed to fetch registers for thread 5851: I/O error 11-15 15:22:14.135 6772 6772 W crash_dump32: failed to fetch registers for thread 6724: I/O error 11-15 15:22:14.145 5523 6769 I system_server: The ClassLoaderContext is a special shared library. 11-15 15:22:14.151 6772 6772 I crash_dump32: obtaining output fd from tombstoned, type: kDebuggerdTombstone 11-15 15:22:14.151 2093 2093 I /system/bin/tombstoned: received crash request for pid 5440 11-15 15:22:14.152 6772 6772 I crash_dump32: performing dump of process 5440 (target tid = 5440) 11-15 15:22:14.168 6772 6772 F DEBUG : 11-15 15:22:14.168 6772 6772 F DEBUG : LineageOS Version: '16.0-20181115-UNOFFICIAL-kminilte' 11-15 15:22:14.169 6772 6772 F DEBUG : Build fingerprint: 'samsung/kminiltexx/kminilte:6.0.1/MMB29K/G800FXXU1CPK5:user/release-keys' 11-15 15:22:14.169 6772 6772 F DEBUG : Revision: '0' 11-15 15:22:14.169 6772 6772 F DEBUG : ABI: 'arm' 11-15 15:22:14.169 6772 6772 F DEBUG : pid: 5440, tid: 5440, name: surfaceflinger >>> /system/bin/surfaceflinger <<< 11-15 15:22:14.169 6772 6772 F DEBUG : signal 11 (SIGSEGV), code 1 (SEGV_MAPERR), fault addr 0xff000000 11-15 15:22:14.169 6772 6772 F DEBUG : r0 ff000000 r1 b4a47014 r2 b4a47000 r3 ff000000 11-15 15:22:14.169 6772 6772 F DEBUG : r4 b419e000 r5 b4a47074 r6 00000001 r7 b4a47000 11-15 15:22:14.169 6772 6772 F DEBUG : r8 b4d9f620 r9 00000060 r10 00000003 r11 b52caa94 11-15 15:22:14.169 6772 6772 F DEBUG : ip b3bf76b7 sp bec27118 lr b3bf899d pc b3bf7df6 11-15 15:22:14.173 5678 5678 W SubtypeLocaleUtils: KeyboardLayoutSet not found, use QWERTY: locale=en_AU extraValue=TrySuppressingImeSwitcher,AsciiCapable,SupportTouchPositionCorrection,EmojiCapable 11-15 15:22:14.176 5678 5678 W RichInputMethodSubtype: Can't find emoji subtype 11-15 15:22:14.176 5678 5678 W RichInputMethodSubtype: No input method subtype found; returning dummy subtype: Multi-lingual subtype: android.view.inputmethod.InputMethodSubtype@d78b2ed0, zz 11-15 15:22:14.192 5523 6769 I system_server: The ClassLoaderContext is a special shared library. 11-15 15:22:14.224 6772 6772 F DEBUG : 11-15 15:22:14.224 6772 6772 F DEBUG : backtrace: 11-15 15:22:14.224 6772 6772 F DEBUG : #00 pc 00002df6 /system/lib/libdisplay.so 11-15 15:22:14.224 6772 6772 F DEBUG : #01 pc 00003999 /system/lib/libdisplay.so (ExynosOverlayDisplay::determineYuvOverlay(hwc_display_contents_1)+64) 11-15 15:22:14.224 6772 6772 F DEBUG : #02 pc 00004a33 /system/lib/libdisplay.so (ExynosOverlayDisplay::prepare(hwc_display_contents_1)+82) 11-15 15:22:14.225 6772 6772 F DEBUG : #03 pc 00002269 /system/lib/hw/hwcomposer.exynos3.so (exynos5_prepare(hwc_composer_device_1*, unsigned int, hwc_display_contents_1)+96) 11-15 15:22:14.225 6772 6772 F DEBUG : #04 pc 0000e12b /system/vendor/lib/libhwc2on1adapter.so (android::HWC2On1Adapter::prepareAllDisplays()+818) 11-15 15:22:14.225 6772 6772 F DEBUG : #05 pc 0000ddbb /system/vendor/lib/libhwc2on1adapter.so (android::HWC2On1Adapter::Display::validate(unsigned int, unsigned int)+42) 11-15 15:22:14.225 6772 6772 F DEBUG : #06 pc 00005c8b /system/vendor/lib/hw/android.hardware.graphics.composer@2.1-impl.so (android::hardware::graphics::composer::V2_1::passthrough::detail::HwcHalImpl::validateDisplay(unsigned long long, std::1::vector<unsigned long long, std::1::allocator>, std::__1::vector<android::hardware::graphics::composer::V2_1::IComposerClient::Composition, std::__1::allocator<android::hardware::graphics::composer::V2_1::ICompo 11-15 15:22:14.225 6772 6772 F DEBUG : #07 pc 0000b727 /system/vendor/lib/hw/android.hardware.graphics.composer@2.1-impl.so (android::hardware::graphics::composer::V2_1::hal::ComposerCommandEngine::executePresentOrValidateDisplay(unsigned short)+190) 11-15 15:22:14.225 6772 6772 F DEBUG : #08 pc 0000a21f /system/vendor/lib/hw/android.hardware.graphics.composer@2.1-impl.so (android::hardware::graphics::composer::V2_1::hal::ComposerCommandEngine::execute(unsigned int, android::hardware::hidl_vec const&, bool, unsigned int, android::hardware::hidl_vec)+90) 11-15 15:22:14.225 6772 6772 F DEBUG : #09 pc 00008269 /system/vendor/lib/hw/android.hardware.graphics.composer@2.1-impl.so (android::hardware::graphics::composer::V2_1::hal::detail::ComposerClientImpl<android::hardware::graphics::composer::V2_1::IComposerClient, android::hardware::graphics::composer::V2_1::hal::ComposerHal>::executeCommands(unsigned int, android::hardware::hidl_vec const&, std::1::function<void (android::hardware::graphics::composer::V2_1::Error, bool, unsigned int, android::hardware::hid 11-15 15:22:14.225 6772 6772 F DEBUG : #10 pc 0002dba5 /system/lib/android.hardware.graphics.composer@2.1.so (android::hardware::graphics::composer::V2_1::BsComposerClient::executeCommands(unsigned int, android::hardware::hidl_vec const&, std::1::function<void (android::hardware::graphics::composer::V2_1::Error, bool, unsigned int, android::hardware::hidl_vec const&)>)+128) 11-15 15:22:14.225 6772 6772 F DEBUG : #11 pc 00046c53 /system/lib/libsurfaceflinger.so (android::Hwc2::impl::Composer::execute()+1310) 11-15 15:22:14.226 6772 6772 F DEBUG : #12 pc 00048267 /system/lib/libsurfaceflinger.so (android::Hwc2::impl::Composer::presentOrValidateDisplay(unsigned long long, unsigned int, unsigned int, int, unsigned int)+136) 11-15 15:22:14.226 6772 6772 F DEBUG : #13 pc 0004ee31 /system/lib/libsurfaceflinger.so (HWC2::Display::presentOrValidate(unsigned int, unsigned int, android::sp, unsigned int)+60) 11-15 15:22:14.226 6772 6772 F DEBUG : #14 pc 000529df /system/lib/libsurfaceflinger.so (android::HWComposer::prepare(android::DisplayDevice&)+242) 11-15 15:22:14.226 6772 6772 F DEBUG : #15 pc 00044e3b /system/lib/libsurfaceflinger.so (android::DisplayDevice::prepareFrame(android::HWComposer&)+10) 11-15 15:22:14.226 6772 6772 F DEBUG : #16 pc 00074229 /system/lib/libsurfaceflinger.so (android::SurfaceFlinger::handleMessageRefresh()+1304) 11-15 15:22:14.226 6772 6772 F DEBUG : #17 pc 0007303b /system/lib/libsurfaceflinger.so (android::SurfaceFlinger::onMessageReceived(int)+202) 11-15 15:22:14.226 6772 6772 F DEBUG : #18 pc 0000f171 /system/lib/libutils.so (android::Looper::pollInner(int)+284) 11-15 15:22:14.226 6772 6772 F DEBUG : #19 pc 0000efd3 /system/lib/libutils.so (android::Looper::pollOnce(int, int, int, void)+26) 11-15 15:22:14.226 6772 6772 F DEBUG : #20 pc 000678ff /system/lib/libsurfaceflinger.so (android::impl::MessageQueue::waitMessage()+62) 11-15 15:22:14.226 6772 6772 F DEBUG : #21 pc 00072825 /system/lib/libsurfaceflinger.so (android::SurfaceFlinger::run()+8) 11-15 15:22:14.226 6772 6772 F DEBUG : #22 pc 00002815 /system/bin/surfaceflinger (main+644) 11-15 15:22:14.227 6772 6772 F DEBUG : #23 pc 0008c865 /system/lib/libc.so (libc_init+48) 11-15 15:22:14.227 6772 6772 F DEBUG : #24 pc 0000254f /system/bin/surfaceflinger (_start_main+46) 11-15 15:22:14.227 6772 6772 F DEBUG : #25 pc 0000031e 11-15 15:22:14.502 5523 5685 W NativeCrashListener: Couldn't find ProcessRecord for pid 5440 11-15 15:22:14.449 5523 6769 I chatty : uid=1000(system) BackgroundDexOp identical 13 lines 11-15 15:22:14.457 5523 6769 I system_server: The ClassLoaderContext is a special shared library. 11-15 15:22:14.509 2093 2093 E /system/bin/tombstoned: Tombstone written to: /data/tombstones/tombstone_05 11-15 15:22:14.513 5523 6769 I system_server: The ClassLoaderContext is a special shared library. 11-15 15:22:14.525 1964 1964 I ServiceManager: service 'gpu' died 11-15 15:22:14.526 5691 5722 W SurfaceComposerClient: ComposerService remote (surfaceflinger) died [0x9d77f1c0] 11-15 15:22:14.526 5691 5999 E [EGL-ERROR]: void egl_platform_dequeue_buffer(egl_surface):1722: failed to dequeue buffer from native window 0x8abee808; err = -32, buf = 0x0,max_allowed_dequeued_buffers 3 11-15 15:22:14.526 1964 1964 I ServiceManager: service 'SurfaceFlinger' died 11-15 15:22:14.526 5841 5900 E [EGL-ERROR]: void __egl_platform_dequeue_buffer(egl_surface):1722: failed to dequeue buffer from native window 0x88a7f808; err = -32, buf = 0x0,max_allowed_dequeued_buffers 3 11-15 15:22:14.526 1964 1964 I ServiceManager: service 'Exynos.IPService' died 11-15 15:22:14.526 5691 5999 E [EGL-ERROR]: void egl_platform_dequeue_buffer(egl_surface):1722: failed to dequeue buffer from native window 0x8abee808; err = -32, buf = 0x0,max_allowed_dequeued_buffers 3 11-15 15:22:14.526 1964 1964 I ServiceManager: service 'Exynos.HWCService' died 11-15 15:22:14.526 5841 5900 E [EGL-ERROR]: void __egl_platform_dequeue_buffer(egl_surface):1722: failed to dequeue buffer from native window 0x88a7f808; err = -32, buf = 0x0,max_allowed_dequeued_buffers 3 11-15 15:22:14.526 1964 1964 I ServiceManager: service 'media.wifi_display' died 11-15 15:22:14.526 5841 6685 W SurfaceComposerClient: ComposerService remote (surfaceflinger) died [0x9c555500] 11-15 15:22:14.526 5523 5877 W SurfaceComposerClient: ComposerService remote (surfaceflinger) died [0x9f1df200] 11-15 15:22:14.526 5691 5999 E [EGL-ERROR]: void __egl_platform_dequeue_buffer(egl_surface*):1722: failed to dequeue buffer from native window 0x8abee808; err = -32, buf = 0x0,max_allowed_dequeued_buffers 3 11-15 15:22:14.527 5691 5999 E [EGL-ERROR]: void egl_platform_dequeue_buffer(egl_surface):1722: failed to dequeue buffer from native window 0x8abee808; err = -32, buf = 0x0,max_allowed_dequeued_buffers 3 11-15 15:22:14.527 5841 5900 E BpDisplayEventConnection: Failed to transact (-32) 11-15 15:22:14.527 5691 5999 E [EGL-ERROR]: void __egl_platform_dequeue_buffer(egl_surface):1722: failed to dequeue buffer from native window 0x8abee808; err = -32, buf = 0x0,max_allowed_dequeued_buffers 3 11-15 15:22:14.527 5691 5999 E [EGL-ERROR]: void egl_platform_dequeue_buffer(egl_surface):1722: failed to dequeue buffer from native window 0x8abee808; err = -32, buf = 0x0,max_allowed_dequeued_buffers 3 11-15 15:22:14.527 5691 5999 W OpenGLRenderer: swapBuffers encountered EGL error 12301 on 0x9d09fd60, halting rendering... 11-15 15:22:14.528 5841 5900 E [EGL-ERROR]: void __egl_platform_dequeue_buffer(egl_surface):1722: failed to dequeue buffer from native window 0x88a7f808; err = -32, buf = 0x0,max_allowed_dequeued_buffers 3 11-15 15:22:14.528 5523 5577 I Choreographer: Skipped 31 frames! The application may be doing too much work on its main thread. 11-15 15:22:14.528 5841 5900 E [EGL-ERROR]: void egl_platform_dequeue_buffer(egl_surface):1722: failed to dequeue buffer from native window 0x88a7f808; err = -32, buf = 0x0,max_allowed_dequeued_buffers 3 11-15 15:22:14.528 5841 5841 I Choreographer: Skipped 31 frames! The application may be doing too much work on its main thread. 11-15 15:22:14.528 5841 5900 W OpenGLRenderer: Dirty 0.00 272.00 720.00 1172.00 doesn't intersect with 0 0 1 0 ? 11-15 15:22:14.528 5523 5577 E BpDisplayEventConnection: Failed to transact (-32) 11-15 15:22:14.529 5841 5841 E BpDisplayEventConnection: Failed to transact (-32) 11-15 15:22:14.530 5822 6609 W SurfaceComposerClient: ComposerService remote (surfaceflinger) died [0x8b865540] 11-15 15:22:14.530 6204 6220 W SurfaceComposerClient: ComposerService remote (surfaceflinger) died [0x8c8c8340] 11-15 15:22:14.531 5523 5575 E BpDisplayEventConnection: Failed to transact (-32) 11-15 15:22:14.533 5691 5999 W libEGL : EGLNativeWindowType 0x8abee808 disconnect failed 11-15 15:22:14.534 5841 5900 E [EGL-ERROR]: void __egl_platform_dequeue_buffer(egl_surface):1722: failed to dequeue buffer from native window 0x88a7f808; err = -32, buf = 0x0,max_allowed_dequeued_buffers 3 11-15 15:22:14.534 5841 5900 E [EGL-ERROR]: void __egl_platform_dequeue_buffer(egl_surface*):1722: failed to dequeue buffer from native window 0x88a7f808; err = -32, buf = 0x0,max_allowed_dequeued_buffers 3 11-15 15:22:14.534 5841 5900 W OpenGLRenderer: Dirty 0.00 272.00 720.00 1172.00 doesn't intersect with 0 0 1 0 ? 11-15 15:22:14.535 5523 5549 E DisplayEventDispatcher: Display event receiver pipe was closed or an error occurred. events=0x9 11-15 15:22:14.536 6204 6204 E DisplayEventDispatcher: Display event receiver pipe was closed or an error occurred. events=0x9 11-15 15:22:14.536 6204 6266 E DisplayEventDispatcher: Display event receiver pipe was closed or an error occurred. events=0x9 11-15 15:22:14.536 5523 5547 E DisplayEventDispatcher: Display event receiver pipe was closed or an error occurred. events=0x9 11-15 15:22:14.536 5691 5999 E OpenGLRenderer: Display event receiver pipe was closed or an error occurred. events=0x9 11-15 15:22:14.536 5841 5841 E DisplayEventDispatcher: Display event receiver pipe was closed or an error occurred. events=0x9 11-15 15:22:14.536 5841 5900 E OpenGLRenderer: Display event receiver pipe was closed or an error occurred. events=0x9 11-15 15:22:14.537 5691 5691 E DisplayEventDispatcher: Display event receiver pipe was closed or an error occurred. events=0x9 11-15 15:22:14.543 5523 5544 I BootReceiver: Copying /data/tombstones/tombstone_05 to DropBox (SYSTEM_TOMBSTONE) 11-15 15:22:14.537 5691 5691 E DisplayEventDispatcher: Display event receiver pipe was closed or an error occurred. events=0x9 11-15 15:22:14.547 6204 6265 E OpenGLRenderer: Display event receiver pipe was closed or an error occurred. events=0x9 11-15 15:22:14.549 5523 6769 I system_server: The ClassLoaderContext is a special shared library. 11-15 15:22:14.571 5523 6769 I system_server: The ClassLoaderContext is a special shared library. 11-15 15:22:14.573 5523 6735 E OpenGLRenderer: Display event receiver pipe was closed or an error occurred. events=0x9 11-15 15:22:14.574 5822 6757 E OpenGLRenderer: Display event receiver pipe was closed or an error occurred. events=0x9 11-15 15:22:14.574 5822 5822 E DisplayEventDispatcher: Display event receiver pipe was closed or an error occurred. events=0x9

[friedensfurz]

@spookcity I think some of these might be missing for sdcardfs. It would be a shame to abandon it and fall back to FUSE.

[Spookcity]

Thanks @friedensfurz I cherry picked a bunch of sdcardfs related commits from @rodman01's repo. I have been busy,but I will look over these commits and see if we are missing any of those. Thanks again!

[rodman01]

Hi @Spookcity thanks for updating your repos. I synched sources today and added selinux from klte common repo to kminilte-common. Build finished without errors and rom is booting and working fine so far. Next I want to change the kernel from permissive to enforcing, will see what happens :-) Hi @friedensfurz I also synched RR sources and compiled my first RR pie :-). Build finished also without errors and rom is running well too. If someone what to test the one or the other, I can upload them to android file host. Just for your interest and sorry for OT and OT off :-)

[aelfinn]

@rodman01 – I’d love to test the RR Pie build. :)

[rodman01]

okidoki and here it is :-) https://www.androidfilehost.com/?fid=11410963190603914558 Happy testing 👍

[andrea224]

@rodman01 I would rather try your LOS 16.0 build, instead.

[rodman01]

Spooky announced in his thread that he is about to release a new version soon, so I don't not want to forestall and foreground me or my build. I would guess that the new version is around the corner soon @Spookcity ?!?

[Spookcity]

@rodman01 Thats a good idea using the SELinux policies from the S5. That may be a good starting point. Perhaps this along with comparing our SELinux policies from our Lineage-14.1 branch we may be able to get somewhere.Also,thanks for sharing your RR build. I will be testing this one out as well.

I plan to release a new build of LOS 16 this weekend on XDA if all goes well. I was going to last weekend,but am hoping maybe I can find a little time this week to work on a few things. If you have a newer build you are certainly welcome to share it here. I would be curios to hear from someone who uses the WiFi hotspot if it is fixed or if we still have a bit more to do.

[andrea224]

@rodman01 don't worry man, I'll wait @Spookcity 's build, Im not surely in a hurry. Btw, thanks both for your time on thi device.

[aelfinn]

@rodman01 – RR Pie is looking pretty good. Smooth and reasonably fast, no obvious bugs (can’t test RIL at the moment, though). The only negative point I noticed is high standby battery drain (display off, airplane mode, Universal GMS Doze 1.6) of more than 1% per hour—which is 3-4 times as much as an identically configured RR Pie Samsung S4 I installed at the same time.

[rodman01]

Thanks @Spookcity and I just uploaded my build from Sunday so everyone is welcome to download it for tests. I am also curios whether wifi hotspot is working now. Maybe someone who is using it before and with other devices, who is familiar with the feature can test and report back. The selinux enforcing kernel is just ready and if its booting I can also upload it for tests with netflix and how ever where it is necessary :-)

[rodman01]

@aelfinn thanks for your report. I noticed the battery drain too although I haven't tested much. I would guess its related to ambient display and when always on is enabled. If doing so dt2w seems to work and I think the device never go in sleep mode. Maybe I there are some specific overlays necessary which I haven't checked till now. If I remember right in previous RR versions and RR settings there had been th possibility to activate or deactivate the dt2w, but cannot find it at the moment. I think its not yet included or not included anymore. Maybe you can test the battery live with disabled ambient display always on too?

[polaurent]

@rodman01 : seems that your LOS upload on Androidfilehost is corrupted ( 328 Mo / 408 ) can you re-upload or verify ? cheers

[rodman01]

ok sorry....please wait I will recompile it with the "old" permissive kernel....

[polaurent]

@rodman01 : it's live !!!

[rodman01]

now it is: https://www.androidfilehost.com/?fid=11410963190603915898 Sorry for the confusion.

[polaurent]

yes , the previous does not boot ;-)

[rodman01]

Yes sorry I mistakenly added the test kernel to the wrong rom.zip. Wanted to include it in RR for my own tries, but mixed u the folders :-)

[polaurent]

no problems , this last seems to work for the moment as fine as possible, by the way i do not use things like Whatsapp or Instagram, so i couldn't consider myself as a tester. The Phone is my professional phone, and the wfi Hotspot was essential , and it is back !!

[aelfinn]

@rodman01 – Your LOS16 build feels very nice: smooth, fast, no obvious bugs AFAICT. On rare occasions, fingerprint unlock doesn’t work. And battery drain is pretty bad here too (settings as mentioned, Ambient display off of course).

[Mikahlo]

A late happy new year to everyone around here! Work is getting less busy and I will have some time to hang around a bit :)

[polaurent]

hello @rodman01 , little proble with A2DP profile with my car, no sound , phone HSP and PBAP works fine...

[Mikahlo]

Tested the RR-P-v7 and it looks great so far. The settings for hw keys don't apply and that's about it. Found some avc:denied messages in the logs and something seems to be wrong with wifi. "getWifiLinkLayerStats failed" is shown in the logs every couple of secs. The problem with modifying files, which could be an selinux thing, is there resulting in whatsapp not being able to make a chat backup, games not being able to save progress and other stuff. But overall it is usable and crazy to have android 9 on such an "old" device so thanks!

[Mikahlo]

dmesg.txt from dmesg: [ 768.101753] [1] 4668 WARNING: at /media/rodman01/SSD_1TB/rr-p/kernel/samsung/kminilte/fs/sdcardfs/inode.c:630 sdcardfs_setattr_wrn+0x48/0x60() [ 768.102041] [1] 4668 sdcardfs does not support setattr. User setattr2. [ 768.102166] [1] 4668 Backtrace: [ 768.102425] [1] 4668 [<c0012f50>] (dump_backtrace+0x0/0x128) from [<c07fa370>] (dump_stack+0x18/0x1c) [ 768.102573] [1] 4668 r6:c0954908 r5:00000276 r4:c0227f84 r3:60070013 [ 768.103108] [1] 4668 [<c07fa358>] (dump_stack+0x0/0x1c) from [<c003d698>] (warn_slowpath_fmt+0x64/0x80) [ 768.103323] [1] 4668 [<c003d634>] (warn_slowpath_fmt+0x0/0x80) from [<c0227f84>] (sdcardfs_setattr_wrn+0x48/0x60) [ 768.103532] [1] 4668 r3:c0227f3c r2:c0954954 [ 768.103798] [1] 4668 r6:c41f1d70 r5:d1bff330 r4:0000a068 [ 768.104292] [1] 4668 [<c0227f3c>] (sdcardfs_setattr_wrn+0x0/0x60) from [<c0148da8>] (notify_change2+0x2fc/0x314) [ 768.104555] [1] 4668 [<c0148aac>] (notify_change2+0x0/0x314) from [<c012c06c>] (do_truncate2+0x8c/0xb0) [ 768.104736] [1] 4668 [<c012bfe0>] (do_truncate2+0x0/0xb0) from [<c012c41c>] (do_truncate+0x2c/0x34) [ 768.104981] [1] 4668 r7:ede1a1a0 r6:ede1a26c r5:dfaca180 r4:c41f1ed0 [ 768.105474] [1] 4668 [<c012c3f0>] (do_truncate+0x0/0x34) from [<c013b544>] (do_last+0x590/0x890) [ 768.105710] [1] 4668 [<c013afb4>] (do_last+0x0/0x890) from [<c013b908>] (path_openat+0xc4/0x420) [ 768.105873] [1] 4668 [<c013b844>] (path_openat+0x0/0x420) from [<c013bcd4>] (do_filp_open+0x38/0x94) [ 768.106108] [1] 4668 [<c013bc9c>] (do_filp_open+0x0/0x94) from [<c012d308>] (do_sys_open+0xec/0x184) [ 768.106328] [1] 4668 r8:c000e8a4 r7:00000001 r6:ee09b000 r5:ffffff9c r4:0000007b [ 768.108795] [1] 4668 [<c012d21c>] (do_sys_open+0x0/0x184) from [<c012d3d8>] (sys_openat+0x10/0x14) [ 768.109008] [1] 4668 r9:c41f0000 r8:c000e8a4 r7:00000142 r6:00000241 r5:000001b6 [ 768.109465] [1] 4668 r4:8b5fe8f0 [ 768.109737] [1] 4668 [<c012d3c8>] (sys_openat+0x0/0x14) from [<c000e86c>] (__sys_trace_return+0x0/0x34) [ 768.109886] [1] 4668 ---[ end trace 8e205b89b83dd002 ]---

[Spookcity]

Thanks for the feedback on @rodman01's build @Mikahlo @polaurent @aelfinn. It is appreciated. I am getting ready to do a build in a minute to release on XDA either late today or tomorrow,so I will not have time to look into any of these issues for this build. I should however have some more free time to look into these in the coming weeks. I kinda figured that the sdcardfs (storage issues) were not fully solved. Its a bit confusing. I'm not sure what we are missing. I noticed the "sdcardfs does not support setattr. User setattr2" errors in the log as well @Mikahlo. I had forgotten about this. Thanks for the reminder. It is something to look into. I also noticed the WiFi errors,but have had no issues with WiFi (aside from hotspot which @polaurent mentioned is working now...thanks to @rodman01) So I'm not real sure what the log spam is due to. I will also try to look into the A2DP profile issues when i have time.

I also tested @rodman01's RR build. It is very nice and I had no issues with it. My battery life is very good on the Pie builds with AOD off and ambient display for new notifications only. My battery is as old as my phone and I am very happy with my battery life on these ROMS.

Also,I opened a new issue for development discussion as this has gotten fairly long. Lets continue these discussions Here

[ModestRepo]

finally: #### build completed successfully (01:05:54 (hh:mm:ss)) #### And guess what? I only changed the build command to mka bacon, after breakfast, then picks.sh, then brunch. Now I need to check your patches, because I have no sound, sim and signal is working, but no call possible and waking up the device after screen turns off (with home button and with power button) is problematic, will say extrem slow. But I think this might be solved (partly?) with applying a couple of your patches. Otherwise will I will try too find solutions :-).

Edit: Sound and calls are working now. Will test and check other things now...

Hey, I just wanted to say thanks for reporting on how you fixed that 9.txt error. This appears to be one of the few references to that issue on the internet and the only one that has a solution attached to it. I've grinding on this one for days and was close to just smashing everything to bits. Running breakfast and then brunch solved my problem too. Thanks!