Database Develop

compiling error kernel

hi guys....i've downloaded the last kernel sources of franciscofranco 2.35.10 kernel....i've added the sippolo's workaround and then increase the HZ value of the kernel(set at 500) for my custom rom.....after i've done
Code:
make thunderg-perf_defconfig
make zImage -j4
it stuck at this error
Code:
drivers/input/touchscreen/mcs6000_ts.c: In function ‘mcs6000_ts_ioctl_down’:
drivers/input/touchscreen/mcs6000_ts.c:430:6: error: implicit declaration of function ‘gpio_configure’
drivers/input/touchscreen/mcs6000_ts.c: In function ‘mcs6000_ts_probe’:
drivers/input/touchscreen/mcs6000_ts.c:733:25: warning: unused variable ‘hw_ver’
drivers/input/touchscreen/mcs6000_ts.c:733:17: warning: unused variable ‘fw_ver’
make[3]: *** [drivers/input/touchscreen/mcs6000_ts.o] Errore 1
i don't understand why......i think it is a lib problem.....but seems that all libraries are imported!!!can anyone help me??? thnx in advice

Which toolchain?
Sent from my LG-P500 using XDA Premium App

lupohirp said:
hi guys....i've downloaded the last kernel sources of franciscofranco 2.35.10 kernel....i've added the sippolo's workaround and then increase the HZ value of the kernel(set at 500) for my custom rom.....after i've done
Code:
make thunderg-perf_defconfig
make zImage -j4
it stuck at this error
Code:
drivers/input/touchscreen/mcs6000_ts.c: In function ‘mcs6000_ts_ioctl_down’:
drivers/input/touchscreen/mcs6000_ts.c:430:6: error: implicit declaration of function ‘gpio_configure’
drivers/input/touchscreen/mcs6000_ts.c: In function ‘mcs6000_ts_probe’:
drivers/input/touchscreen/mcs6000_ts.c:733:25: warning: unused variable ‘hw_ver’
drivers/input/touchscreen/mcs6000_ts.c:733:17: warning: unused variable ‘fw_ver’
make[3]: *** [drivers/input/touchscreen/mcs6000_ts.o] Errore 1
i don't understand why......i think it is a lib problem.....but seems that all libraries are imported!!!can anyone help me??? thnx in advice
Click to expand...
Click to collapse
.35 kernel have new GPIO library, you should code sippolo's workaround into new driver, do not replace whole file.

mik_os said:
.35 kernel have new GPIO library, you should code sippolo's workaround into new driver, do not replace whole file.
Click to expand...
Click to collapse
Aaaaa i don't know that thanks now it works ;-) now the next step i'm ttying to port the acer liquid multitouch fix thanks mik
Sent from my LG-P500 using XDA App

If you can fix also the tethering and the random bugs

use pdanet for tethering

now i'm stucking at this error...((
arch/arm/mm/built-in.o: In function `__flush_icache_all':
/home/pasquale/Scrivania/Franco.Kernel/2.6.35/arch/arm/include/asm/cacheflush.h:402: undefined reference to `swapper_space'
arch/arm/mm/built-in.o: In function `____atomic_set_bit':
/home/pasquale/Scrivania/Franco.Kernel/2.6.35/arch/arm/include/asm/bitops.h:46: undefined reference to `swapper_space'
arch/arm/mm/built-in.o: In function `spin_unlock':
/home/pasquale/Scrivania/Franco.Kernel/2.6.35/include/linux/spinlock.h:324: undefined reference to `swapper_space'
mm/built-in.o: In function `sync_page':
/home/pasquale/Scrivania/Franco.Kernel/2.6.35/mm/filemap.c:190: undefined reference to `swapper_space'
mm/built-in.o: In function `set_page_dirty':
/home/pasquale/Scrivania/Franco.Kernel/2.6.35/mm/page-writeback.c:1181: undefined reference to `swapper_space'
mm/built-in.o:/home/pasquale/Scrivania/Franco.Kernel/2.6.35/mm/page-writeback.c:1335: more undefined references to `swapper_space' follow
make: *** [.tmp_vmlinux1] Errore 1
Click to expand...
Click to collapse
seems another time a library problem... anyone can help me??

lupohirp said:
now i'm stucking at this error...((
seems another time a library problem... anyone can help me??
Click to expand...
Click to collapse
What have you patched?
It just says it's missing the swapper_space function header, it's not defined.

franciscofranco said:
What have you patched?
Click to expand...
Click to collapse
nothing franco....i've only added sippolo's workaround for base to port acer liquid multitouch fix and set kernel to 500hz i've worked only on mcs6000_ts......but it stuck at this error i don't know why!!! i think it is a library problem....something that is missing...but i don't know what are the libraries that i must install if it is this the problem!(

You downloaded the latest sources? Because I haven't compiled it, I just merged the patches, I might have forgot to apply something. In this case it says it misses the swapper_space reference.
http://lxr.free-electrons.com/source/mm/swap_state.c#L40
http://lxr.free-electrons.com/source/include/linux/mm.h#L771
http://lxr.free-electrons.com/source/include/linux/swap.h#L319
Add the swapper_space code to those files and compile again.

franciscofranco said:
You downloaded the latest sources? Because I haven't compiled it, I just merged the patches, I might have forgot to apply something. In this case it says it misses the swapper_space reference.
http://lxr.free-electrons.com/source/mm/swap_state.c#L40
http://lxr.free-electrons.com/source/include/linux/mm.h#L771
http://lxr.free-electrons.com/source/include/linux/swap.h#L319
Add the swapper_space code to those files and compile again.
Click to expand...
Click to collapse
yes franco i've downloaded latest sources! thanks now i'm going to add it and then try!

franco the swapper_space code that you posted is already there in the three files((

I dunno, I have to leave now, I'll look at it later.

I'm sorry to disappoint you, but there ain't a problem with the kernel, I just compiled it without any modification, just like it is in the sources and it compiles fine. Try to make clean and make distclean, then compile again.

franciscofranco said:
I'm sorry to disappoint you, but there ain't a problem with the kernel, I just compiled it without any modification, just like it is in the sources and it compiles fine. Try to make clean and make distclean, then compile again.
Click to expand...
Click to collapse
yeah franco i know.....i think it is a libraries problem on my linux distro.....but i have installe all the binutils and the header files yesterday also i have tried a dirty hack i tried to set CONFIG_SWAP=n in the thunderg_perf_defconfig configuration file....compiled succesfully but the phone doesn't started ...thanks anyway franco!

Kernel Compiling Help

Hello, I am trying to compile a stock KK4 kernel from source (obtained from the Samsung open source website), and I'm running into some errors. I downloaded the latest toolchain from Codesourcery and edited the makefile's cross compile to point to its installed directory. But when I get to this step,
$ cd kernel
$ make kepler_rev02_defconfig
$ make
Click to expand...
Click to collapse
Trying the first make gets me this error, and a similar one for the second make.
make: *** No rule to make target `defconfig'. Stop.
Click to expand...
Click to collapse
I have searched everywhere and nothing seems to be relevant. Can somebody point me in the right direction?

It looks like you are missing the makefile for deconfig, as far as a direction to point, you can try one of these great kernel devs on the board, they might be able to help you.

crt60 said:
It looks like you are missing the makefile for deconfig
Click to expand...
Click to collapse
It does that when you try to make the whole thing too though...

[celoxhd][SGH-I757M][CWM Advanced Edition] PhilZ Touch

Main thread + features + install instructions + dev support
http://forum.xda-developers.com/showthread.php?t=2201860
PhilZ Touch is a CWM Advanced Edition that adds all the features you could ever miss in CWM
It is a well proven recovery for many phones
It also adds a full touch interface a completely configurable GUI
Please give your feedback, what works, and any bug you could encounter
Read the features, and check if you are missing something
Also, do not forget to read about the powerful aroma file manager integration and double tap shortcut
Download links
Last version can be found here:
celoxhd
http://goo.im/devs/philz_touch/CWM_Advanced_Edition
Original dev supporter:
http://www.androidfilehost.com/?w=files&flid=12461
Click to expand...
Click to collapse
Huge thanks and credits to @titanic_fanatic
He ported the sources and provided all needed stuff to build it

Excellent, many thanks for all the work you put into this for us. We very much appreciate it
On the topic of bugs, I only have two (but I really don't mind at all). The screenshot works except for the resulting image is wavy coloured lines and no vibration regardless of the setting.
We've never had either, so I'm not missing them at all, but wanted to let you know.
Take care Phil
Sent from my SGH-I757M using XDA Premium 4 mobile app

titanic_fanatic said:
Excellent, many thanks for all the work you put into this for us. We very much appreciate it
On the topic of bugs, I only have two (but I really don't mind at all). The screenshot works except for the resulting image is wavy coloured lines and no vibration regardless of the setting.
We've never had either, so I'm not missing them at all, but wanted to let you know.
Take care Phil
Sent from my SGH-I757M using XDA Premium 4 mobile app
Click to expand...
Click to collapse
Added you to credits on the device page in main thread and here in OP :good:
screen shots: forget it, I should disable it once for all on new devices. Sadly, we have no working thing for most new devices. Only some Sammy and the mako did work
vibration: can you provide the path for vibrator?, it should be in kernel drivers I guess. Currently I use:
Code:
/sys/class/timed_output/vibrator/enable
We echo some msec value there and it Bzzzzz for that msec

Phil3759 said:
Added you to credits on the device page in main thread and here in OP :good:
screen shots: forget it, I should disable it once for all on new devices. Sadly, we have no working thing for most new devices. Only some Sammy and the mako did work
vibration: can you provide the path for vibrator?, it should be in kernel drivers I guess. Currently I use:
Code:
/sys/class/timed_output/vibrator/enable
We echo some msec value there and it Bzzzzz for that msec
Click to expand...
Click to collapse
That's the correct path Maybe the driver isn't loading properly... I'll see if I can get a dmesg Log when booting recovery to see if the driver actually loads
Sent from my SGH-I757M using XDA Premium 4 mobile app

Vibrator Fixed
Hey @Phil3759, I did a whole lot of investigating tonight and have figured out and fixed the vibration issue. I discovered that it was trying to load the module from /system/lib/modules/vibrator.ko. This is my roms ko file and is currently for cm10.2. The kerenel I setup for this recovery is our cm11 kernel and uses a different module (compiled against that kernel). It was trying to load the cm10.2 module from my rom into the cm11 kernel which obviously doesn't work, so I split the recovery.img added the correct module at /tmp/system/lib/modules added the necessary lines to init.recovery.qcom.rc and repackaged the kernel and new ramdisk.
I have flashed this and tested it. it works like a charm now Can you replace the version in your repo with this new one?

titanic_fanatic said:
Hey @Phil3759, I did a whole lot of investigating tonight and have figured out and fixed the vibration issue. I discovered that it was trying to load the module from /system/lib/modules/vibrator.ko. This is my roms ko file and is currently for cm10.2. The kerenel I setup for this recovery is our cm11 kernel and uses a different module (compiled against that kernel). It was trying to load the cm10.2 module from my rom into the cm11 kernel which obviously doesn't work, so I split the recovery.img added the correct module at /tmp/system/lib/modules added the necessary lines to init.recovery.qcom.rc and repackaged the kernel and new ramdisk.
I have flashed this and tested it. it works like a charm now Can you replace the version in your repo with this new one?
Click to expand...
Click to collapse
If you update your device tree, it would be even better. That way, my next updates won't be affected again

Phil3759 said:
If you update your device tree, it would be even better. That way, my next updates won't be affected again
Click to expand...
Click to collapse
I just finished updating my device tree with the necessary fixes for the vibrator to work in recovery
https://github.com/titanic-fanatic/android_device_samsung_celoxhd/commit/6f005f4045551e48adfa0ac228641bf8d6a12128
Also, not that I care, but Amarullz was able to get screenshots working in the latest AromaInstaller (Flamboyan I think). If your interested, maybe ask him what he did to get it to work. He could probably point you to the commits
Take Care

titanic_fanatic said:
I just finished updating my device tree with the necessary fixes for the vibrator to work in recovery
https://github.com/titanic-fanatic/...mmit/6f005f4045551e48adfa0ac228641bf8d6a12128
Also, not that I care, but Amarullz was able to get screenshots working in the latest AromaInstaller (Flamboyan I think). If your interested, maybe ask him what he did to get it to work. He could probably point you to the commits
Take Care
Click to expand...
Click to collapse
Looking at it quickly: why that first extra / at end of copy files?
Also, won't be cleaner to use standard path like /lib/modules?
/tmp can be wiped and modified by installers and recovery

Phil3759 said:
Looking at it quickly: why that first extra / at end of copy files?
Also, won't be cleaner to use standard path like /lib/modules?
/tmp can be wiped and modified by installers and recovery
Click to expand...
Click to collapse
The first / is because it's part of a set of chained product copy files directives:
# Ramdisk
PRODUCT_COPY_FILES += \
device/samsung/celoxhd/ramdisk/init.qcom.usb.rc:root/init.qcom.usb.rc \
# BT firmware
PRODUCT_COPY_FILES += \
device/samsung/celoxhd/firmware/bcm4330B1.hcd:system/etc/firmware/bcm4330B1.hcd \
# Needed for vibrator to work in recovery
PRODUCT_COPY_FILES += \
device/samsung/celoxhd/recovery/root/lib/modules/vibrator.ko:recovery/root/lib/modules/vibrator.ko \
device/samsung/celoxhd/recovery/root/init.recovery.qcom.rc:recovery/root/init.recovery.qcom.rc
Click to expand...
Click to collapse
I did this only to adhere to the existing pattern in my device.mk file, the one appended to the first PRODUCT_COPY_FILES was already there from when I forked the project and took that as being a set of chained commands. If this is wrong, I will change it
I definitely see your point on the /tmp dir. I here the commit that fixes that
https://github.com/titanic-fanatic/android_device_samsung_celoxhd/commit/61da6be2d60a03797ddc31c0187cc5c62bb3a97a

titanic_fanatic said:
The first / is because it's part of a set of chained product copy files directives:
I did this only to adhere to the existing pattern in my device.mk file, the one appended to the first PRODUCT_COPY_FILES was already there from when I forked the project and took that as being a set of chained commands. If this is wrong, I will change it
I definitely see your point on the /tmp dir. I here the commit that fixes that
https://github.com/titanic-fanatic/android_device_samsung_celoxhd/commit/61da6be2d60a03797ddc31c0187cc5c62bb3a97a
Click to expand...
Click to collapse
I uploaded 6.01.2 with your fixed vibrator :good:
PRODUCT_COPY_FILES:
Maybe I am wrong, but usually, in make files, only the last element has no trailing /
Code:
# Ramdisk
PRODUCT_COPY_FILES += \
device/samsung/celoxhd/ramdisk/init.qcom.usb.rc:root/init.qcom.usb.rc
xxx-yyy \
zzz-pppp\
last_one
# BT firmware
PRODUCT_COPY_FILES += \
device/samsung/celoxhd/firmware/bcm4330B1.hcd:system/etc/firmware/bcm4330B1.hcd
# Needed for vibrator to work in recovery
PRODUCT_COPY_FILES += \
device/samsung/celoxhd/recovery/root/tmp/system/lib/modules/vibrator.ko:recovery/root/tmp/system/lib/modules/vibrator.ko \
device/samsung/celoxhd/recovery/root/init.recovery.qcom.rc:recovery/root/init.recovery.qcom.rc
Also, just for info, my full gui builds need these, often in cm.mk:
Code:
TARGET_SCREEN_HEIGHT := 1280
TARGET_SCREEN_WIDTH := 720

Phil3759 said:
I uploaded 6.01.2 with your fixed vibrator :good:
PRODUCT_COPY_FILES:
Maybe I am wrong, but usually, in make files, only the last element has no trailing /
Code:
# Ramdisk
PRODUCT_COPY_FILES += \
device/samsung/celoxhd/ramdisk/init.qcom.usb.rc:root/init.qcom.usb.rc
xxx-yyy \
zzz-pppp\
last_one
# BT firmware
PRODUCT_COPY_FILES += \
device/samsung/celoxhd/firmware/bcm4330B1.hcd:system/etc/firmware/bcm4330B1.hcd
# Needed for vibrator to work in recovery
PRODUCT_COPY_FILES += \
device/samsung/celoxhd/recovery/root/tmp/system/lib/modules/vibrator.ko:recovery/root/tmp/system/lib/modules/vibrator.ko \
device/samsung/celoxhd/recovery/root/init.recovery.qcom.rc:recovery/root/init.recovery.qcom.rc
Also, just for info, my full gui builds need these, often in cm.mk:
Code:
TARGET_SCREEN_HEIGHT := 1280
TARGET_SCREEN_WIDTH := 720
Click to expand...
Click to collapse
It seems that the extra \ was introduced in our JB branch and has never been fixed. Various examples in googles repo confirm that you are correct. I have removed the extra \ and added the properties you mentioned above to the cm.mk file. Here is the relevant commit:
https://github.com/titanic-fanatic/android_device_samsung_celoxhd/commit/6ce5f482d63cd70cf64a3d1428ad2428844e23cb

Hey @Phil3759, I just downloaded and flashed the latest version. Thanks kindly for the update
I see that I made a boo boo and forgot to update the path to the vibrator in the init.recovery.qcom.rc file and so the vibrator isn't loading. I've fixed that with the following commit:
https://github.com/titanic-fanatic/android_device_samsung_celoxhd/commit/b87b3267d75a0de1bb94f68850ae6986ac4dde92
One other thing, for some reason the background is tiled 4 times as if my devices dimensions were not set correctly. I do have that set in my cm.mk file. Here was the commit for that:
https://github.com/titanic-fanatic/android_device_samsung_celoxhd/commit/6ce5f482d63cd70cf64a3d1428ad2428844e23cb
Is there somewhere else I'm supposed to specify the device screen dimensions?
Thanks again for the new version and I apologize for missing that path in the init.recovery.qcom.rc.

titanic_fanatic said:
Hey @Phil3759, I just downloaded and flashed the latest version. Thanks kindly for the update
I see that I made a boo boo and forgot to update the path to the vibrator in the init.recovery.qcom.rc file and so the vibrator isn't loading. I've fixed that with the following commit:
https://github.com/titanic-fanatic/android_device_samsung_celoxhd/commit/b87b3267d75a0de1bb94f68850ae6986ac4dde92
One other thing, for some reason the background is tiled 4 times as if my devices dimensions were not set correctly. I do have that set in my cm.mk file. Here was the commit for that:
https://github.com/titanic-fanatic/android_device_samsung_celoxhd/commit/6ce5f482d63cd70cf64a3d1428ad2428844e23cb
Is there somewhere else I'm supposed to specify the device screen dimensions?
Thanks again for the new version and I apologize for missing that path in the init.recovery.qcom.rc.
Click to expand...
Click to collapse
About background image:
I think this is the cause:
https://github.com/CyanogenMod/android_device_samsung_celox-common/blob/cm-11.0/celox-common.mk#L45
called here:
https://github.com/titanic-fanatic/...xhd/blob/philz-recovery-cm-11.0/device.mk#L37
Your cm.mk settings are overridden
Maybe just move device resolution from cm.mk to device.mk after Line 37
About the vibrator: do you need exfat through kernel modules? If yes, let me know. It is much faster than fuse. For that, you'll need to compile the kernel with samsung exfat sources. I can guide you for that
Else, link me to what exactly is needed in your kernel for the vibrator to work. If it is just a change in defconfig, I can add it and compile along exfat

Phil3759 said:
About background image:
I think this is the cause:
https://github.com/CyanogenMod/android_device_samsung_celox-common/blob/cm-11.0/celox-common.mk#L45
called here:
https://github.com/titanic-fanatic/...xhd/blob/philz-recovery-cm-11.0/device.mk#L37
Your cm.mk settings are overridden
Maybe just move device resolution from cm.mk to device.mk after Line 37
About the vibrator: do you need exfat through kernel modules? If yes, let me know. It is much faster than fuse. For that, you'll need to compile the kernel with samsung exfat sources. I can guide you for that
Else, link me to what exactly is needed in your kernel for the vibrator to work. If it is just a change in defconfig, I can add it and compile along exfat
Click to expand...
Click to collapse
I see it, and now as you've recommended I reset the values after calling celox-common.mk. The commit is here:
https://github.com/titanic-fanatic/android_device_samsung_celoxhd/commit/cd44b44b7cbb0ddb3b649ac5c4f8d5d1f51d7998
As for exfat, I will be adding this when I have some time to rebuild this kernel. We can worry about exfat for the next time you do device updates and hopefully I'll have it ready by that time. I have your main thread in my subscriptions, so I'll know when your ready to release a new version and will remind you to sync with my repo at that time
Thanks for catching that issue with screen size

titanic_fanatic said:
I see it, and now as you've recommended I reset the values after calling celox-common.mk. The commit is here:
https://github.com/titanic-fanatic/android_device_samsung_celoxhd/commit/cd44b44b7cbb0ddb3b649ac5c4f8d5d1f51d7998
As for exfat, I will be adding this when I have some time to rebuild this kernel. We can worry about exfat for the next time you do device updates and hopefully I'll have it ready by that time. I have your main thread in my subscriptions, so I'll know when your ready to release a new version and will remind you to sync with my repo at that time
Thanks for catching that issue with screen size
Click to expand...
Click to collapse
np
Let me know if you need info to add exfat
I will update with the current fixes asap

Phil3759 said:
np
Let me know if you need info to add exfat
I will update with the current fixes asap
Click to expand...
Click to collapse
Can I get a link to these fixes please when you have it ready

titanic_fanatic said:
Can I get a link to these fixes please when you have it ready
Click to expand...
Click to collapse
https://github.com/PhilZ-cwm6/GE-Ed...mmit/a8963ca853cfd88466716f8a25434d5c469a2bb9
It is just adding the exfat folder in drivers and edit the makefile with -y to have it built as static. No need to separate modules

Phil3759 said:
https://github.com/PhilZ-cwm6/GE-Ed...mmit/a8963ca853cfd88466716f8a25434d5c469a2bb9
It is just adding the exfat folder in drivers and edit the makefile with -y to have it built as static. No need to separate modules
Click to expand...
Click to collapse
Excellent. I'll add it soon, it should be a pretty quick process.
Quick question though since we are currently having some issues with random reboots in all of my latest builds of cm-11.0. My kernel was updated at some point, and that's when my builds started having random reboots. They are actually not random, but seem rather consistent.
My question is this, do you think this will affect operation in recovery? I believe the reboots are due to kernel panic from a graphics related issue but am not sure if that will only affect the rom and not recovery.
If this might affect recovery, I think I'll build it as a module so I can just add it to my device tree instead of having to update the entire zImage. This will at least give us the old working kernel with the exfat module included.
What do you think?
Sent from my SGH-I757M using XDA Premium 4 mobile app

titanic_fanatic said:
Excellent. I'll add it soon, it should be a pretty quick process.
Quick question though since we are currently having some issues with random reboots in all of my latest builds of cm-11.0. My kernel was updated at some point, and that's when my builds started having random reboots. They are actually not random, but seem rather consistent.
My question is this, do you think this will affect operation in recovery? I believe the reboots are due to kernel panic from a graphics related issue but am not sure if that will only affect the rom and not recovery.
If this might affect recovery, I think I'll build it as a module so I can just add it to my device tree instead of having to update the entire zImage. This will at least give us the old working kernel with the exfat module included.
What do you think?
Sent from my SGH-I757M using XDA Premium 4 mobile app
Click to expand...
Click to collapse
you'll end up with magic number mismatch errors
There are some hacks to force loading on mismatch, but, in my previous experience, they failed on many devices
the drivers must be compiled along the zImage
Now, why don't we drop to celox-common kernel which seems to work fine on skyrocket and other phones?

Phil3759 said:
you'll end up with magic number mismatch errors
There are some hacks to force loading on mismatch, but, in my previous experience, they failed on many devices
the drivers must be compiled along the zImage
Now, why don't we drop to celox-common kernel which seems to work fine on skyrocket and other phones?
Click to expand...
Click to collapse
It is for the most part the same kernel with the few quirks of my device. More notably is my devices display drivers are mipi s6e8aa0_720hd and had to fork the kernel to apply some fixes to that driver as it's not maintained by cm.
Hmmm. I'll be the guinea pig. I will build with the new kernel and I'll test the recovery before you upload. I just think that the issue might be a memory leak in android causing the kernel to panic (overlay issue) and might not have an effect in recovery since its graphics are very basic.
I'll let you know when it's ready
Sent from my SGH-I757M using XDA Premium 4 mobile app

CM11 Kenel compiling trouble !!

Hi. i have tried to compile cm11 kernel by using kernel source from http://github.com/cm-nokia-x2 and gcc toolchain 4.8 from here http://github.com/huyparody/gcc_linaro
i get config.gz from /proc/
and renamed it to ara_defconfig
and when i:
make ara_defconfig
make
after a few minute, it show me a error:
CC drivers/input/touchscreen/cyttsp4_device_access.o
drivers/input/touchscreen/cyttsp4_device_access.c: In function 'cyttsp4_grpdata_store_touch_params':
drivers/input/touchscreen/cyttsp4_device_access.c:1447:19: warning: 'rc3' may be used uninitialized in this function [-Wuninitialized]
error, forbidden warning: cyttsp4_device_access.c:1447
make[3]: *** [drivers/input/touchscreen/cyttsp4_device_access.o] Error 1
make[2]: *** [drivers/input/touchscreen] Error 2
make[1]: *** [drivers/input] Error 2
make: *** [drivers] Error 2
[email protected]:~/android/kernel/nokia/msm8610$
Click to expand...
Click to collapse
how to solve it!! thanks much

kaotd99 said:
how to solve it!! thanks much
Click to expand...
Click to collapse
Hello,
Use toolchain from CyanogenMod prebuilt repo, here is link: github.com/CyanogenMod/android_prebuilt/tree/cm-11.0/linux-x86/toolchain/arm-eabi-4.4.3
You also don't need to copy config from /proc, it is included in source at arch/arm/configs/msm8610-nokia-defconfig.

zombah said:
Hello,
Use toolchain from CyanogenMod prebuilt repo, here is link: github.com/CyanogenMod/android_prebuilt/tree/cm-11.0/linux-x86/toolchain/arm-eabi-4.4.3
You also don't need to copy config from /proc, it is included in source at arch/arm/configs/msm8610-nokia-defconfig.
Click to expand...
Click to collapse
thanks dev

oh god... i got new error:
i type:
export PATH=/home/kaotd/android/prebuilt/linux-x86/toolchain/arm-eabi-4.4.3/bin:$PATH
export ARCH=arm
export CROSS_COMPILE=arm-eabi-
and:
[email protected]:/home/kaotd/android/kernel/nokia/msm8610# make msm8610-nokia_defconfig
#
# configuration written to .config
#
[email protected]:/home/kaotd/android/kernel/nokia/msm8610# make
scripts/kconfig/conf --silentoldconfig Kconfig
CHK include/linux/version.h
CHK include/generated/utsrelease.h
make[1]: `include/generated/mach-types.h' is up to date.
CC kernel/bounds.s
kernel/bounds.c:1: error: bad value (cortex-a7) for -mtune= switch
make[1]: *** [kernel/bounds.s] Error 1
make: *** [prepare0] Error 2
[email protected]:/home/kaotd/android/kernel/nokia/msm8610#
Click to expand...
Click to collapse
can you help me ? thanks.... because im a newbie and sorry about that

kaotd99 said:
can you help me ? thanks.... because im a newbie and sorry about that
Click to expand...
Click to collapse
Hello,
Ahh yep, i forgot switch to memcpy optimizations, here is correct link to newer toolchain android.googlesource.com/platform/prebuilts/gcc/linux-x86/arm/arm-eabi-4.7

zombah said:
Hello,
Ahh yep, i forgot switch to memcpy optimizations, here is correct link to newer toolchain android.googlesource.com/platform/prebuilts/gcc/linux-x86/arm/arm-eabi-4.7
Click to expand...
Click to collapse
thanks.. i will try and report to you later!! thanks again

@zombah hi again. i was compiled a kernel( zImage )
now i have "Kernel: arch/arm/boot/zImage is ready"
so how can i compare it with boot.img ?

Hello,
If you wont use inline kernel building inside CyanogenMod, you probably need to unpack boot.img content replace kernel and then pack boot.img again with replaced kernel. Remember that it is dangerous if your kernel is not bootable, you'll stuck to recovery only boot and if your recovery also lost only unbricking will help.
Use CyanogenMod mkbootimg.c/unpackbootimg.c for that, here is repo github.com/CyanogenMod/android_system_core/tree/cm-11.0/mkbootimg
Here is how to compile them separatly of whole system:
Code:
gcc -o mkbootimg mkbootimg.c ../libmincrypt/sha.c -I../include
gcc -o unpackbootimg unpackbootimg.c ../libmincrypt/sha.c -I../include
you'll need to download dependencies also.

zombah said:
Hello,
If you wont use inline kernel building inside CyanogenMod, you probably need to unpack boot.img content replace kernel and then pack boot.img again with replaced kernel. Remember that it is dangerous if your kernel is not bootable, you'll stuck to recovery only boot and if your recovery also lost only unbricking will help.
Use CyanogenMod mkbootimg.c/unpackbootimg.c for that, here is repo github.com/CyanogenMod/android_system_core/tree/cm-11.0/mkbootimg
Here is how to compile them separatly of whole system:
Code:
gcc -o mkbootimg mkbootimg.c ../libmincrypt/sha.c -I../include
gcc -o unpackbootimg unpackbootimg.c ../libmincrypt/sha.c -I../include
you'll need to download dependencies also.
Click to expand...
Click to collapse
thanks much

can you tell me which file contain OverClock number of nokia x2 ? and which kernelPatch i can apply with this kernel ?

Hello,
I don't know, i'm not interested in overclocking and never investigated this question.
And i don't recommend you to play with overclocking without deep embedded development skills, it is very easy to kill your device with wrong power modifiers.

i'm understand, thanks for your suggestion.
not about nokia x2ds but can i ask you a question?
-can i make a kernel source for a mtk device ? because i have a mtk china phone which not have kernel source from manufacturer !

kaotd99 said:
i'm understand, thanks for your suggestion.
not about nokia x2ds but can i ask you a question?
-can i make a kernel source for a mtk device ? because i have a mtk china phone which not have kernel source from manufacturer !
Click to expand...
Click to collapse
Hello,
Sure it is possible, why not. But you need at least serial/uart console on device, otherwise how you will debug kernel boot problems?

zombah said:
Hello,
Sure it is possible, why not. But you need at least serial/uart console on device, otherwise how you will debug kernel boot problems?
Click to expand...
Click to collapse
thanks... now i'm trying to port double-tap to wake feature from Nokia Stock Kernel. There is any suggestion ?? Thanks

i'm trying to add this: https://github.com/ryanaepple/android_kernel_motoe/commit/080003bc8e61349bf7c8da00fa6923f3c84d1927
and being stuck at editing synaptics_dsx_i2c.c and synaptics_dsx_i2c.h
can you tell me which files is Touchscreen driver of nokia x2 ?? i think it is synaptics_i2c_rmi4.c/.h

kaotd99 said:
can you tell me which files is Touchscreen driver of nokia x2 ?? i think it is synaptics_i2c_rmi4.c/.h
Click to expand...
Click to collapse
Hello,
Nokia X2 have Cypress TMA4XX touchscreen, driver location is drivers/input/touchscreen/cyttsp4*

LineageOS 17 for suzuran?

Does its existence for devices like the Z3C make it more or less likely for us to see on suzuran? Would love to see a LineageOS 17 ROM for this device. Makes me wish I was a dev. Thanks to all those who continue to grind hours into porting and developing ROMs for the platform.

beflythis said:
Would love to see a LineageOS 17 ROM for this device.
Click to expand...
Click to collapse
Oh! I have successfully built a LineageOS 17.0 rom for z5c! :good:
But.... As my 16.0 try this rom is not booting...
Don't know why

Berni-0815 said:
Oh! I have successfully built a LineageOS 17.0 rom for z5c! :good:
But.... As my 16.0 try this rom is not booting...
Don't know why
Click to expand...
Click to collapse
Are you making any kernel changes? If not, that's probably why. You can't use the LineageOS 15.1 kernel on the PIE. Correct me if I'm wrong but Pie requires linux kernel 4.4x and greater while, Android 10 requires 4.9 and greater.

Apart from fixing compile-errors, no.
But I don't think that the kernel version is important. This project uses kernel 3.x too, if I'm not wrong...
Sure, there have to be changes, but I can't do so...

I've tried another kernel (this one) and I changed/added some files to avoid compile errors. But the result is the same:
The splash screen apears, then a blank screen....
The splash screen apears, then a blank screen....
The splash screen apears, then a blank screen....
The splash screen apears, then a blank screen....
...and so on until end of battery...

Have you tried looking at any sort of logs, kernel logs? (I'm not sure if there would be one in the first place) It would be better to try by increments (PIE first with https://github.com/joel16/android_kernel_sony_msm8994). I might give it a try this weekend just for the sake of it.

Joel16 said:
Have you tried looking at any sort of logs, kernel logs? (I'm not sure if there would be one in the first place)
Click to expand...
Click to collapse
Where should these logs be?
I have no access to the device at this early stage...
An when I reflash a working rom /system will be overwritten.
It would be better to try by increments (PIE first with https://github.com/joel16/android_kernel_sony_msm8994). I might give it a try this weekend just for the sake of it.
Click to expand...
Click to collapse
I've tried to build pie with this kernel. Nearly the same result.

Berni-0815 said:
Where should these logs be?
I have no access to the device at this early stage...
An when I reflash a working rom /system will be overwritten.
I've tried to build pie with this kernel. Nearly the same result.
Click to expand...
Click to collapse
Weird, I ran into some issues building pie yesterday, but didn't have any time to fix it. I'll try again tonight and see how it goes. If I still have issues, I'll have to wait till the weekend.

Joel16 said:
Weird, I ran into some issues building pie yesterday
Click to expand...
Click to collapse
I'm sorry; I should have tell you about that... :angel:
I changed a lot of things to compile it successfully. Here are my changes (I hope this list is complete):
.bashrc:
Code:
export CPU_SSE42=false
to avoid crashes in hiddenapi and profman
device/sony/kitakami-common/SonyOtgSwitch/Android.mk:
Code:
LOCAL_SDK_VERSION := current
kernel/sony/msm8994/include/uapi/linux/rmnet_data.h; insert lines 226 to 236 (line 226 changed [a comma at the end of line]):
Code:
RMNET_NETLINK_DEL_VND_TC_FLOW,
/*
* RMNET_NETLINK_NEW_VND_WITH_NAME - Creates a new virtual network
* device node with the specified
* device name
* Args: int32_t node number
* char[] vnd_name - Use as name
* Returns: status code
*/
RMNET_NETLINK_NEW_VND_WITH_NAME
bionic/libc/include/bits/fortify/fcntl.h; insert lines 33 to 36:
Code:
#ifndef __O_TMPFILE
#define __O_TMPFILE 020000000
#endif
#define O_TMPFILE (__O_TMPFILE | O_DIRECTORY)
hardware/qcom/audio-caf/msm8994/hal/audio_hw.c; line 2457:
Code:
ALOGE("%s: error [B]%ld[/B] - %s", __func__, ret, pcm_get_error(out->pcm));
hardware/qcom/audio-caf/msm8994/hal/audio_extn/audio_extn.c; lines 523 and 524:
Code:
void audio_extn_set_parameters(struct audio_device *adev [B]__unused[/B],
struct str_parms *parms [B]__unused[/B])
and line 67:
Code:
static struct audio_extn_module aextnmod [B]__unused[/B] = {
hardware/qcom/display-caf/msm8994/libqdutils/idle_invalidator.cpp; line 127:
Code:
ALOGD_IF(II_DEBUG, "IdleInvalidator::%s Idle Timeout fired len [B]%ld[/B]",
hardware/qcom/display-caf/msm8994/libhdmi/hdmi.cpp: line 196:
Code:
ALOGD("%s: Scan Info string: %s length = [B]%ld[/B]",
and line 320:
Code:
ALOGD_IF(DEBUG, "%s: EDID string: %s length = [B]%ld[/B]",
device/sony/kitakami-common/SonyOtgSwitch/Android.mk: line 5:
Code:
LOCAL_PRIVATE_PLATFORM_APIS := true
Changes in bold

Berni-0815 said:
I'm sorry; I should have tell you about that... :angel:
I changed a lot of things to compile it successfully. Here are my changes (I hope this list is complete):
.bashrc:
Code:
export CPU_SSE42=false
to avoid crashes in hiddenapi and profman
device/sony/kitakami-common/SonyOtgSwitch/Android.mk:
Code:
LOCAL_SDK_VERSION := current
kernel/sony/msm8994/include/uapi/linux/rmnet_data.h; insert lines 226 to 236 (line 226 changed [a comma at the end of line]):
Code:
RMNET_NETLINK_DEL_VND_TC_FLOW,
/*
* RMNET_NETLINK_NEW_VND_WITH_NAME - Creates a new virtual network
* device node with the specified
* device name
* Args: int32_t node number
* char[] vnd_name - Use as name
* Returns: status code
*/
RMNET_NETLINK_NEW_VND_WITH_NAME
bionic/libc/include/bits/fortify/fcntl.h; insert lines 33 to 36:
Code:
#ifndef __O_TMPFILE
#define __O_TMPFILE 020000000
#endif
#define O_TMPFILE (__O_TMPFILE | O_DIRECTORY)
hardware/qcom/audio-caf/msm8994/hal/audio_hw.c; line 2457:
Code:
ALOGE("%s: error [B]%ld[/B] - %s", __func__, ret, pcm_get_error(out->pcm));
hardware/qcom/audio-caf/msm8994/hal/audio_extn/audio_extn.c; lines 523 and 524:
Code:
void audio_extn_set_parameters(struct audio_device *adev [B]__unused[/B],
struct str_parms *parms [B]__unused[/B])
and line 67:
Code:
static struct audio_extn_module aextnmod [B]__unused[/B] = {
hardware/qcom/display-caf/msm8994/libqdutils/idle_invalidator.cpp; line 127:
Code:
ALOGD_IF(II_DEBUG, "IdleInvalidator::%s Idle Timeout fired len [B]%ld[/B]",
hardware/qcom/display-caf/msm8994/libhdmi/hdmi.cpp: line 196:
Code:
ALOGD("%s: Scan Info string: %s length = [B]%ld[/B]",
and line 320:
Code:
ALOGD_IF(DEBUG, "%s: EDID string: %s length = [B]%ld[/B]",
device/sony/kitakami-common/SonyOtgSwitch/Android.mk: line 5:
Code:
LOCAL_PRIVATE_PLATFORM_APIS := true
Changes in bold
Click to expand...
Click to collapse
Yup I made those changes except for
Code:
export CPU_SSE42=false
I did not get any issues building but the device does bootloop. I'm gonna see if we can try and pull any sort of logs to see what's causing the issue.

Joel16 said:
Yup I made those changes except for
Code:
export CPU_SSE42=false
I did not get any issues building
Click to expand...
Click to collapse
That seems to be a problem with my cpu. The mentioned line is a cpu instruction. While trying to build this rom on my pc I ran into errors.
but the device does bootloop. I'm gonna see if we can try and pull any sort of logs to see what's causing the issue.
Click to expand...
Click to collapse
Maybe it's possible to backup the system onto a sd-card with an actual twrp and analyse these files?

Berni-0815 said:
That seems to be a problem with my cpu. The mentioned line is a cpu instruction. While trying to build this rom on my pc I ran into errors.
Maybe it's possible to backup the system onto a sd-card with an actual twrp and analyse these files?
Click to expand...
Click to collapse
I'm not entirely sure how it works, but I have heard people talking about pulling logs form bootloops. I mean there has to be a way to debug a bootloop, I highly doubt people just blindly fix them. Perhaps we can access last kernel message or logcat, someway. I'm going to try this weekend and see what I can come up with.

You can do It!!!

Joel16 said:
I'm not entirely sure how it works, but I have heard people talking about pulling logs form bootloops. I mean there has to be a way to debug a bootloop, I highly doubt people just blindly fix them. Perhaps we can access last kernel message or logcat, someway. I'm going to try this weekend and see what I can come up with.
Click to expand...
Click to collapse
Berni-0815 said:
That seems to be a problem with my cpu. The mentioned line is a cpu instruction. While trying to build this rom on my pc I ran into errors.
Maybe it's possible to backup the system onto a sd-card with an actual twrp and analyse these files?
Click to expand...
Click to collapse
What is about adbd, see here:
https://forum.xda-developers.com/showpost.php?p=67386761&postcount=13
The z5c is based on Qualcomm Snapdragon 810 MSM8994, like:
SONY XPERIA Z5 PREMIUM | MOTOROLA DROID TURBO | MICROSOFT LUMIA 950 XL | NUBIA Z9 MAX
Is there any source of a pie build availible where we can get some hints how get a pie build to run?
Here are some kernel sources from the Nubia Z9 Max (nx510j):
https://github.com/nubia-development/android_kernel_nubia_msm8994

I_did_it_just_tmrrow said:
What is about adbd, see here:
https://forum.xda-developers.com/showpost.php?p=67386761&postcount=13
Click to expand...
Click to collapse
I'll have a look at this. :good:

Joel16 said:
Have you tried looking at any sort of logs, kernel logs? (I'm not sure if there would be one in the first place) It would be better to try by increments (PIE first with https://github.com/joel16/android_kernel_sony_msm8994). I might give it a try this weekend just for the sake of it.
Click to expand...
Click to collapse
For my self I try slowly become deeper into that topic: building a rom.
So I did a little bit of research in github and gitlab and I found some interesting repos from this user:
https://github.com/Cyborg2017?tab=repositories
I ask him if he could help us a little bit and the first result is this.
@Joel16 & @Berni-0815 Perhaps this is useful:
https://github.com/Cyborg2017/android_kernel_sony_msm8994/tree/sony_msm8994_p-upstream
I hope it help us to come closer to an LineageOS 17 for suzuran.

I_did_it_just_tmrrow said:
I ask him if he could help us a little bit and the first result is this.
Click to expand...
Click to collapse
Error 404
A little time later: I think, you're talking about this kernel?
---------- Post added at 16:22 ---------- Previous post was at 16:03 ----------
I_did_it_just_tmrrow said:
What is about adbd, see here:
https://forum.xda-developers.com/showpost.php?p=67386761&postcount=13
Click to expand...
Click to collapse
Tried the mentioned tip in the 2nd post of this that thread without any success. I'm so sorry.
How can I include adbd into my build? I'll give it a try...

Berni-0815 said:
Error 404
A little time later: I think, you're talking about this kernel?
---------- Post added at 16:22 ---------- Previous post was at 16:03 ----------
Tried the mentioned tip in the 2nd post of this that thread without any success. I'm so sorry.
How can I include adbd into my build? I'll give it a try...
Click to expand...
Click to collapse
Yep, just try his source. He told that it could be useful to test this kernel with building twrp first.
To the logging problem. I understand that right that you have nothing in:
Code:
/proc/last_kmsg
Nor in:
Code:
/sys/fs/pstore/console-ramoops
and you check this with the twrp terminal?
You should take a look at this:
https://android.stackexchange.com/questions/213336/how-can-i-enable-last-kmsg

I_did_it_just_tmrrow said:
Yep, just try his source.
Click to expand...
Click to collapse
I'm actually building a rom with that kernel. I'll see later if it's working.
To the logging problem. I understand that right that you have nothing in:
Code:
/proc/last_kmsg
Nor in:
Code:
/sys/fs/pstore/console-ramoops
and you check this with the twrp terminal?
Click to expand...
Click to collapse
Right
You should take a look at this:
https://android.stackexchange.com/questions/213336/how-can-i-enable-last-kmsg
Click to expand...
Click to collapse
Ah! Nice. I'll try it later. Thank you very much! :good:

Berni-0815 said:
I'm actually building a rom with that kernel. I'll see later if it's working.
Right
Ah! Nice. I'll try it later. Thank you very much! :good:
Click to expand...
Click to collapse
Let me know how it goes, I'm more than certain kernel is what's preventing us from getting Pie/Q to boot up.