Hi Everyone
I am fairly new to android customisation, rooting, etc. and have been playing around with creating a custom CyanogenMod kernel to support a TP-Link USB wifi dongle.
The primary reference article for this can be found here - http://forum.xda-developers.com/showthread.php?t=2338179. However I have made some changes as listed below.
Keen to hack this further and any advice or guidance will be greatly appreciated.
Technical Spec:
Ubuntu 12.04 (32 Bit)
Samsung S3 (Cyanogenmod ROM)
Kernel = https://github.com/CyanogenMod/android_kernel_samsung_smdk4412
.config = https://github.com/CyanogenMod/andr.../arch/arm/configs/cyanogenmod_i9300_defconfig
Note - There is no /proc/config.gz on my device, hence the above use of the config.
gcc-arm-linux-gnueabi libstdc++6 libncurses5 libncurses5-dev - exports correctly setup.
Objective
Getting an TP-Link TL-WN821N USB Wifi Dongle to work.
The relevant drivers are already available in mac80211 - http://www.aircrack-ng.org/doku.php?id=mac80211
Error Message
CC net/mac80211/agg-rx.o
CC net/mac80211/ibss.o
CC net/mac80211/mlme.o
net/mac80211/mlme.c: In function ‘ieee80211_mgd_auth’:
net/mac80211/mlme.c:2317:9: error: ‘struct cfg80211_auth_request’ has no member named ‘local_state_change’
net/mac80211/mlme.c: In function ‘ieee80211_assoc_done’:
net/mac80211/mlme.c:2419:2: warning: passing argument 2 of ‘cfg80211_send_rx_assoc’ from incompatible pointer type [enabled by default]
include/net/cfg80211.h:2802:6: note: expected ‘struct cfg80211_bss *’ but argument is of type ‘unsigned char *’
net/mac80211/mlme.c:2419:2: warning: passing argument 3 of ‘cfg80211_send_rx_assoc’ makes pointer from integer without a cast [enabled by default]
include/net/cfg80211.h:2802:6: note: expected ‘const u8 *’ but argument is of type ‘unsigned int’
net/mac80211/mlme.c:2419:2: error: too few arguments to function ‘cfg80211_send_rx_assoc’
include/net/cfg80211.h:2802:6: note: declared here
net/mac80211/mlme.c: In function ‘ieee80211_mgd_deauth’:
net/mac80211/mlme.c:2560:19: error: ‘struct cfg80211_deauth_request’ has no member named ‘bss’
net/mac80211/mlme.c:2561:30: error: ‘struct cfg80211_deauth_request’ has no member named ‘bss’
net/mac80211/mlme.c:2581:18: error: ‘struct cfg80211_deauth_request’ has no member named ‘bss’
net/mac80211/mlme.c:2600:4: error: implicit declaration of function ‘__cfg80211_auth_canceled’ [-Werror=implicit-function-declaration]
net/mac80211/mlme.c:2610:16: error: ‘struct cfg80211_deauth_request’ has no member named ‘local_state_change’
cc1: some warnings being treated as errors
make[2]: *** [net/mac80211/mlme.o] Error 1
make[1]: *** [net/mac80211] Error 2
make: *** [net] Error 2
franna81 said:
Hi Everyone
I am fairly new to android customisation, rooting, etc. and have been playing around with creating a custom CyanogenMod kernel to support a TP-Link USB wifi dongle.
The primary reference article for this can be found here - http://forum.xda-developers.com/showthread.php?t=2338179. However I have made some changes as listed below.
Keen to hack this further and any advice or guidance will be greatly appreciated.
Technical Spec:
Ubuntu 12.04 (32 Bit)
Samsung S3 (Cyanogenmod ROM)
Kernel = https://github.com/CyanogenMod/android_kernel_samsung_smdk4412
.config = https://github.com/CyanogenMod/andr.../arch/arm/configs/cyanogenmod_i9300_defconfig
Note - There is no /proc/config.gz on my device, hence the above use of the config.
gcc-arm-linux-gnueabi libstdc++6 libncurses5 libncurses5-dev - exports correctly setup.
Objective
Getting an TP-Link TL-WN821N USB Wifi Dongle to work.
The relevant drivers are already available in mac80211 - http://www.aircrack-ng.org/doku.php?id=mac80211
Error Message
CC net/mac80211/agg-rx.o
CC net/mac80211/ibss.o
CC net/mac80211/mlme.o
net/mac80211/mlme.c: In function ‘ieee80211_mgd_auth’:
net/mac80211/mlme.c:2317:9: error: ‘struct cfg80211_auth_request’ has no member named ‘local_state_change’
net/mac80211/mlme.c: In function ‘ieee80211_assoc_done’:
net/mac80211/mlme.c:2419:2: warning: passing argument 2 of ‘cfg80211_send_rx_assoc’ from incompatible pointer type [enabled by default]
include/net/cfg80211.h:2802:6: note: expected ‘struct cfg80211_bss *’ but argument is of type ‘unsigned char *’
net/mac80211/mlme.c:2419:2: warning: passing argument 3 of ‘cfg80211_send_rx_assoc’ makes pointer from integer without a cast [enabled by default]
include/net/cfg80211.h:2802:6: note: expected ‘const u8 *’ but argument is of type ‘unsigned int’
net/mac80211/mlme.c:2419:2: error: too few arguments to function ‘cfg80211_send_rx_assoc’
include/net/cfg80211.h:2802:6: note: declared here
net/mac80211/mlme.c: In function ‘ieee80211_mgd_deauth’:
net/mac80211/mlme.c:2560:19: error: ‘struct cfg80211_deauth_request’ has no member named ‘bss’
net/mac80211/mlme.c:2561:30: error: ‘struct cfg80211_deauth_request’ has no member named ‘bss’
net/mac80211/mlme.c:2581:18: error: ‘struct cfg80211_deauth_request’ has no member named ‘bss’
net/mac80211/mlme.c:2600:4: error: implicit declaration of function ‘__cfg80211_auth_canceled’ [-Werror=implicit-function-declaration]
net/mac80211/mlme.c:2610:16: error: ‘struct cfg80211_deauth_request’ has no member named ‘local_state_change’
cc1: some warnings being treated as errors
make[2]: *** [net/mac80211/mlme.o] Error 1
make[1]: *** [net/mac80211] Error 2
make: *** [net] Error 2
Click to expand...
Click to collapse
Exactly the same issue here, bump :/
When I had to build a mac80211 driver (rt2800usb), I had to apply the following changes on top of cm-10.2.0 (d6f2f4ae4e593067dc59e2ac63f022434199f673):
Code:
# Depends on 3.4 backport that breaks mac80211.
git revert d6f2f4ae4e593067dc59e2ac63f022434199f673 # "Add support for Note 8 (N5100 and N5110)"
git revert c421809918b7106b40a81134f9fb5103146fc715 # "update bcmdhd driver from GT-9505 Source"
git revert 44655588a0b361afa15761b5c58f5a558c1c181c # "WIFI: Use kernel version 3.4 ifdef's."
git revert 3cedbff33e9884fdf3f7aaaa83ab9774d8145d1f # "wireless backport from 3.4"
# Another build fix:
Date: Sun Nov 24 12:04:33 2013 +0100
lib: fix compile of lib/genalloc.c
Taken from commit 3cedbff ("wireless backport from 3.4").
lib/genalloc.c | 54 ++++++++++++++++++++++++++++++++----------------------
1 file changed, 32 insertions(+), 22 deletions(-)
After that I was able to build rt2x00 from 3.4.70 (after enabling MAC80211 in config).. Note: these patches do not seem to revert cleanly on cm-11.0
For cm-11.0 (HEAD was 22cbf2cfb211cef2e493c5984d237d0509bf98ce, "otg: when removing ED from readyQ also set flag"), I now do:
Code:
# v--- 2014-11-06 (CM 11.0)
- revert c421809 # update bcmdhd driver from GT-9505 Source
Conflict in drivers/net/wireless/bcmdhd/Makefile, remove -DPASS_IPV4_SUSPEND
and DEBUGFS_CFG80211 (unused), all that remains is:
DHDCFLAGS += -DBLOCK_IPV6_PACKET
Conflict in drivers/net/wireless/bcmdhd/dhd_sec_feature.h, remove:
CONFIG_MACH_KONA
.config:
BCMDHD=M (was =Y, but this allows debugging)
DHD_USE_STATIC_BUF=N
DHD_USE_SCHED_SCAN=N
- revert 4465558 # WIFI: Use kernel version 3.4 ifdef's.
- revert 3cedbff # wireless backport from 3.4
- git checkout 3cedbff -- include/linux/{export,module}.h lib/genalloc.c
git commit -m 'Build fixes from 3cedbff'
Compile fixes, also used by d90b43b "mali: add r3p1 for devices that need it"
This compiles (with CONFIG_MAC80211=M), I have yet to try the newer rt2x00 for my purposes.
Related
I have a build environment that works fine when building Froyo, but when I pull down the Eclair repo from Cyanogen and try to build it for my Passion/NexusOne, it dies on BusyBox. Even if I fix the BusyBox errors below by hand modifying the code, my build still ends at "kernel-msm"... am I missing something? I have followed the instructions from the Cyanogen Wiki to a tee multiple times now, and no dice.
Thanks!
debug output:
host Executable: yuv420sp2rgb (out/host/linux-x86/obj/EXECUTABLES/yuv420sp2rgb_intermediates/yuv420sp2rgb)
true
host Executable: zipalign (out/host/linux-x86/obj/EXECUTABLES/zipalign_intermediates/zipalign)
target StaticLib: libc (out/target/product/passion/obj/STATIC_LIBRARIES/libc_intermediates/libc.a)
true
host SharedLib: libacc (out/host/linux-x86/obj/lib/libacc.so)
target thumb C: libbusybox <= external/busybox/shell/ash.c
target thumb C: libbusybox <= external/busybox/shell/math.c
external/busybox/shell/ash.c: In function 'subevalvar':
external/busybox/shell/ash.c:6163: warning: pointer of type 'void *' used in arithmetic
external/busybox/shell/ash.c:6324: warning: pointer of type 'void *' used in arithmetic
external/busybox/shell/ash.c:6326: warning: pointer of type 'void *' used in arithmetic
external/busybox/shell/ash.c: In function 'printlim':
external/busybox/shell/ash.c:12714: error: 'rlim_t' undeclared (first use in this function)
external/busybox/shell/ash.c:12714: error: (Each undeclared identifier is reported only once
external/busybox/shell/ash.c:12714: error: for each function it appears in.)
external/busybox/shell/ash.c:12714: error: expected ';' before 'val'
external/busybox/shell/ash.c:12716: error: 'val' undeclared (first use in this function)
external/busybox/shell/ash.c: In function 'ulimitcmd':
external/busybox/shell/ash.c:12731: error: 'rlim_t' undeclared (first use in this function)
external/busybox/shell/ash.c:12731: error: expected ';' before 'val'
external/busybox/shell/ash.c:12792: error: 'val' undeclared (first use in this function)
make: *** [out/target/product/passion/obj/STATIC_LIBRARIES/libbusybox_intermediates/shell/ash.o] Error 1
make: *** Waiting for unfinished jobs....
cd external/busybox
git checkout d0194cfc47a71a65e049
Much obliged, Maxisma. That did it!
No problem ;-)
it should be fixed now. We made a Froyo branch.
Hi i'm trying to build ICS Source but i'm getting this error:
HTML:
make: *** No rule to make target `out/target/product/ace/obj/STATIC_LIBRARIES/libaudiointerface_intermediates/libaudiointerface.a', needed by `out/target/product/ace/obj/SHARED_LIBRARIES/libaudio_intermediates/LINKED/libaudio.so'. Stop.
make: *** Waiting for unfinished jobs....
target thumb C++: libaudio <= hardware/msm7k/libaudio-qdsp5v2/AudioHardware.cpp
In file included from hardware/msm7k/libaudio-qdsp5v2/AudioHardware.cpp:34:
hardware/msm7k/libaudio-qdsp5v2/AudioHardware.h:51: error: expected class-name before '{' token
hardware/msm7k/libaudio-qdsp5v2/AudioHardware.h:70: error: ISO C++ forbids declaration of 'AudioStreamOut' with no type
hardware/msm7k/libaudio-qdsp5v2/AudioHardware.h:70: error: 'AudioStreamOut' declared as a 'virtual' field
hardware/msm7k/libaudio-qdsp5v2/AudioHardware.h:70: error: expected ';' before '*' token
hardware/msm7k/libaudio-qdsp5v2/AudioHardware.h:77: error: ISO C++ forbids declaration of 'AudioStreamIn' with no type
hardware/msm7k/libaudio-qdsp5v2/AudioHardware.h:77: error: 'AudioStreamIn' declared as a 'virtual' field
hardware/msm7k/libaudio-qdsp5v2/AudioHardware.h:77: error: expected ';' before '*' token
hardware/msm7k/libaudio-qdsp5v2/AudioHardware.h:85: error: 'AudioStreamOut' has not been declared
hardware/msm7k/libaudio-qdsp5v2/AudioHardware.h:86: error: 'AudioStreamIn' has not been declared
hardware/msm7k/libaudio-qdsp5v2/AudioHardware.h:97: error: expected class-name before '{' token
hardware/msm7k/libaudio-qdsp5v2/AudioHardware.h: In member function 'virtual int android::AudioHardware::AudioStreamOutQ5V2::format() const':
hardware/msm7k/libaudio-qdsp5v2/AudioHardware.h:110: error: 'AudioSystem' has not been declared
hardware/msm7k/libaudio-qdsp5v2/AudioHardware.h: In member function 'virtual uint32_t android::AudioHardware::AudioStreamOutQ5V2::latency() const':
hardware/msm7k/libaudio-qdsp5v2/AudioHardware.h:111: error: 'frameSize' was not declared in this scope
hardware/msm7k/libaudio-qdsp5v2/AudioHardware.cpp: In destructor 'virtual android::AudioHardware::~AudioHardware()':
hardware/msm7k/libaudio-qdsp5v2/AudioHardware.cpp:53: error: 'AudioStreamOut' was not declared in this scope
hardware/msm7k/libaudio-qdsp5v2/AudioHardware.cpp:53: error: expected primary-expression before ')' token
hardware/msm7k/libaudio-qdsp5v2/AudioHardware.cpp: At global scope:
hardware/msm7k/libaudio-qdsp5v2/AudioHardware.cpp:62: error: expected constructor, destructor, or type conversion before '*' token
hardware/msm7k/libaudio-qdsp5v2/AudioHardware.cpp:341: error: expected '}' at end of input
make: *** [out/target/product/ace/obj/SHARED_LIBRARIES/libaudio_intermediates/AudioHardware.o] Error 1
i have tryed many things to fix this erro but no success
i hope some one can help me
bump
Sent from my SGH-T959 using xda premium
Many devs face this error and they are fighting with this.
I just started compiling android but did finish a successful build at 05:00am this morning and didn't get (or see) this error but, no i'm expert by any means..
Considering the error:
No rule to make target `out/target/product/ace/obj/STATIC_LIBRARIES/libaudiointerface_intermediates/libaudiointerface.a', needed by....
Click to expand...
Click to collapse
It literally says its missing something, the rule, so would it be an idea to resync repo perhaps ?
I resynced last night and there were no changes. But, yes, resyncing is important... You never know when Google will drop more source, and they may or may not announce it via mailinglist.
Inspire 4G // BlackICE
Try compiling for tuna/maguro/panda first
Also, try removing libaudio folder from device folder, and if your device has aosp then try using libaudio from there(assuming you are using device from cm7)
I also face with this error ... Maybe something is wrong in include/media
I'm getting this error for my passion.
Ive gotten a little farther by changing the namespace from android to android_audio_legacy as they did it hardware/libhardware_legacy/include/hardware_legacy/
Has anyone managed to fix this problem yet?
But it will still conpiling. Not a big problem. We can fix this later.
Sent from Xperia X10 Self-modded
I can't even seen to compile for tuna keep getting errors on llvm. I compile cm7 so my builds setup is good. Confused
texasice said:
I'm getting this error for my passion.
Ive gotten a little farther by changing the namespace from android to android_audio_legacy as they did it hardware/libhardware_legacy/include/hardware_legacy/
Click to expand...
Click to collapse
Could you pastebin the modded file or generate an .patch file for your change ?
@up I also face with llvm error Do someone have fixed that issue ?
mikegapinski said:
Could you pastebin the modded file or generate an .patch file for your change ?
@up I also face with llvm error Do someone have fixed that issue ?
Click to expand...
Click to collapse
in AudioHardware.h line 28 change :
Code:
namespace android {
as mentioned to :
Code:
namespace android_audio_legacy {
I tried it myself just by curiosity and it leads to these errors :
Code:
target thumb C++: libaudio <= hardware/msm7k/libaudio-qsd8k/AudioHardware.cpp
In file included from hardware/msm7k/libaudio-qsd8k/AudioHardware.cpp:37:
hardware/msm7k/libaudio-qsd8k/AudioHardware.h:318: error: 'Mutex' does not name a type
hardware/msm7k/libaudio-qsd8k/AudioHardware.h:328: error: ISO C++ forbids declaration of 'SortedVector' with no type
hardware/msm7k/libaudio-qsd8k/AudioHardware.h:328: error: expected ';' before '<' token
hardware/msm7k/libaudio-qsd8k/AudioHardware.h:337: error: 'Mutex' does not name a type
hardware/msm7k/libaudio-qsd8k/AudioHardware.cpp:92: error: 'AudioHardware' has not been declared
hardware/msm7k/libaudio-qsd8k/AudioHardware.cpp:102: error: 'AudioHardware' has not been declared
hardware/msm7k/libaudio-qsd8k/AudioHardware.cpp:102: error: ISO C++ forbids declaration of 'AudioHardware' with no type
hardware/msm7k/libaudio-qsd8k/AudioHardware.cpp: In function 'int android::AudioHardware()':
hardware/msm7k/libaudio-qsd8k/AudioHardware.cpp:103: error: only constructors take base initializers
hardware/msm7k/libaudio-qsd8k/AudioHardware.cpp:109: error: 'TTY_MODE_OFF' was not declared in this scope
hardware/msm7k/libaudio-qsd8k/AudioHardware.cpp:112: error: 'msm_bt_endpoint' was not declared in this scope
hardware/msm7k/libaudio-qsd8k/AudioHardware.cpp:112: error: expected primary-expression before ')' token
hardware/msm7k/libaudio-qsd8k/AudioHardware.cpp:116: error: invalid type in declaration before ';' token
hardware/msm7k/libaudio-qsd8k/AudioHardware.cpp:118: error: 'doA1026_init' was not declared in this scope
hardware/msm7k/libaudio-qsd8k/AudioHardware.cpp:124: error: 'mNumBTEndpoints' was not declared in this scope
hardware/msm7k/libaudio-qsd8k/AudioHardware.cpp:125: error: 'mInit' was not declared in this scope
hardware/msm7k/libaudio-qsd8k/AudioHardware.cpp:126: error: return-statement with no value, in function returning 'int'
hardware/msm7k/libaudio-qsd8k/AudioHardware.cpp:131: error: return-statement with no value, in function returning 'int'
hardware/msm7k/libaudio-qsd8k/AudioHardware.cpp:158: error: 'mNumBTEndpoints' was not declared in this scope
hardware/msm7k/libaudio-qsd8k/AudioHardware.cpp:160: error: 'mBTEndpoints' was not declared in this scope
hardware/msm7k/libaudio-qsd8k/AudioHardware.cpp:160: error: expected type-specifier before 'msm_bt_endpoint'
hardware/msm7k/libaudio-qsd8k/AudioHardware.cpp:160: error: expected ';' before 'msm_bt_endpoint'
hardware/msm7k/libaudio-qsd8k/AudioHardware.cpp:161: error: 'mInit' was not declared in this scope
hardware/msm7k/libaudio-qsd8k/AudioHardware.cpp:164: error: expected primary-expression before 'int'
hardware/msm7k/libaudio-qsd8k/AudioHardware.cpp:164: error: expected ')' before 'int'
hardware/msm7k/libaudio-qsd8k/AudioHardware.cpp:167: error: return-statement with no value, in function returning 'int'
hardware/msm7k/libaudio-qsd8k/AudioHardware.cpp:169: error: 'snd_get_bt_endpoint' cannot be used as a function
hardware/msm7k/libaudio-qsd8k/AudioHardware.cpp:206: error: no return statement in function returning non-void
hardware/msm7k/libaudio-qsd8k/AudioHardware.cpp: At global scope:
hardware/msm7k/libaudio-qsd8k/AudioHardware.cpp:208: error: expected constructor, destructor, or type conversion before '::' token
hardware/msm7k/libaudio-qsd8k/AudioHardware.cpp:2025: error: expected '}' at end of input
make: *** [out/target/product/passion/obj/SHARED_LIBRARIES/libaudio_intermediates/AudioHardware.o] Error 1
Any update on this? I had to add LOCAL_MODULE_TAGS := optional to some of the Android.mk to even reach that far. Faulty code from Google or am I doing something wrong myself?
Ive gotten past this by just removing the libaudio* folders from the Android.mk
It at least results in a system.img for the passion
In hardwark/msm7k/Android.mk
for mine qsd8k
from
Code:
common_msm_dirs := libcopybit liblights libopencorehw librpc libstagefrighthw
msm7k_dirs := $(common_msm_dirs) boot libgralloc libaudio
qsd8k_dirs := $(common_msm_dirs) libgralloc-qsd8k libaudio-qsd8k dspcrashd
msm7x30_dirs := liblights libgralloc-qsd8k librpc libaudio-qdsp5v2
to
Code:
common_msm_dirs := libcopybit liblights libopencorehw librpc libstagefrighthw
msm7k_dirs := $(common_msm_dirs) boot libgralloc libaudio
qsd8k_dirs := $(common_msm_dirs) dspcrashd
msm7x30_dirs := liblights libgralloc-qsd8k librpc libaudio-qdsp5v2
i also removed libgralloc cause it causes the screen to not work.
mikegapinski said:
Could you pastebin the modded file or generate an .patch file for your change ?
@up I also face with llvm error Do someone have fixed that issue ?
Click to expand...
Click to collapse
for go past llvm errors i've excluded -Werror in frameworks/compile/slang/Android.mk (local_cflags_for_slang line)
Fix, er, well, lets it build...
On the libaudio issue in msm7k for ICS:
Okay, after mucking around with it for a bit, it looks like it was an issue with namespaces, and things having the same name. So what I did to get it to compile is I went through and specified everything to use the android_audio_legacy namespace except for the things that needed to be on the android namespace.
I also slapped on a definition for 2 methods that were undefined. I do not know if it will work, but it does build.
My diff is here: http://pastebin.com/iP7FSa0C
Disclaimers:
C++ is not my primary language, actually, its not a language I know much of at all
I have not been able to test it, I can only verify that it builds, which I assume is better than having to remove the makefile.
I pretty much did the same thing as jbcode42
It builds but doesnt work.
http://pastebin.com/JGYBKfSs
I sow that libaudio.so is no more used in ICS i think ... Now it is an module under system/hw ex. audio.primary.goldfish ... We will need to wait for new soruces from CM or CA to get rid of the errors ...
lupohirp said:
for go past llvm errors i've excluded -Werror in frameworks/compile/slang/Android.mk (local_cflags_for_slang line)
Click to expand...
Click to collapse
Thanks for this. Any ideas why that error is coming up? It seems like others have got it to compile without modifications. What is everyone compiling on? I am on Ubuntu 11.10 with GCC 4.6.1
Hey guys, sorry if i wasn't supposed to create a thread for this but i am getting this error
Code:
system/core/libion/ion.c: In function 'ion_alloc':
system/core/libion/ion.c:65:3: error: unknown field 'heap_mask' specified in initializer
system/core/libion/ion.c:67:17: warning: initialized field overwritten [-Woverride-init]
system/core/libion/ion.c:67:17: warning: (near initialization for 'data.flags') [-Woverride-init]
system/core/libion/ion.c: In function 'ion_sync_fd':
system/core/libion/ion.c:160:26: error: 'ION_IOC_SYNC' undeclared (first use in this function)
system/core/libion/ion.c:160:26: note: each undeclared identifier is reported only once for each function it appears in
system/core/libion/ion.c:162:1: error: control reaches end of non-void function [-Werror=return-type]
cc1: some warnings being treated as errors
make: *** [out/target/product/glacier/obj/EXECUTABLES/iontest_intermediates/ion.o] Error 1
Thats while compiling CM10.1
I have no idea what is up, i am tearing my hair out over it lol. Thanks in advanced for the help!
THEindian said:
Hey guys, sorry if i wasn't supposed to create a thread for this but i am getting this error
Code:
system/core/libion/ion.c: In function 'ion_alloc':
system/core/libion/ion.c:65:3: error: unknown field 'heap_mask' specified in initializer
system/core/libion/ion.c:67:17: warning: initialized field overwritten [-Woverride-init]
system/core/libion/ion.c:67:17: warning: (near initialization for 'data.flags') [-Woverride-init]
system/core/libion/ion.c: In function 'ion_sync_fd':
system/core/libion/ion.c:160:26: error: 'ION_IOC_SYNC' undeclared (first use in this function)
system/core/libion/ion.c:160:26: note: each undeclared identifier is reported only once for each function it appears in
system/core/libion/ion.c:162:1: error: control reaches end of non-void function [-Werror=return-type]
cc1: some warnings being treated as errors
make: *** [out/target/product/glacier/obj/EXECUTABLES/iontest_intermediates/ion.o] Error 1
Thats while compiling CM10.1
I have no idea what is up, i am tearing my hair out over it lol. Thanks in advanced for the help!
Click to expand...
Click to collapse
bump
anybody?
Q&A for [SOURCES][DEVONLY] CyanogenMod 11.0
Some developers prefer that questions remain separate from their main development thread to help keep things organized. Placing your question within this thread will increase its chances of being answered by a member of the community or by the developer.
Before posting, please use the forum search and read through the discussion thread for [SOURCES][DEVONLY] CyanogenMod 11.0. If you can't find an answer, post it here, being sure to give as much information as possible (firmware version, steps to reproduce, logcat if available) so that you can get help.
Thanks for understanding and for helping to keep XDA neat and tidy!
CM11
Thank you for your efforts and hardwork. I look forward to running cm 11 on the alpha.
Congo from India. Pl. Check it and tell if all features working or not..
what about CM12 any news on that?
HELLO?
Hi. Nice to catch this thread.
I just wanna know... Can korean alpha get this cm11? Or...can get g850f only?
And i wanna talk to you. Cheer up!
Thank you for developing this alpha, modfunk!
Finally, good luck for my ss g alpha....
(i just wanna 'plus alpha' yeah!)
We will only know if you try. I don't know your hardware I only own a G850F
Hi there,
I can hardly wait to get CM11 on my device. Otherwise using this phone is pointelss for me. I would throw the whole Samsung stock ROM to somewhere that it can burn slowly and flawlessly to the ground.
Thanks guys for your efforts!
Can anybody show video hows cyano 11 on alpha ?
thnx
Simona Simmy said:
Can anybody show video hows cyano 11 on alpha ?
thnx
Click to expand...
Click to collapse
CM11 isn't ready for consuming yet on the alpha.
Yet this is called "Devonly" (developers only) and "source".
This means: get github access, get the source, get the tools, compile it, debug compilation process, fix compilation process, get a image.
Pray, push it on your device, hopefully you won't have softbricked your device.
Debug the features that are missing, fix the bugs, find solutions, recompile.
At the moment, modpunkt still has the following (major) issues:
http://forum.xda-developers.com/showpost.php?p=57233732&postcount=46
https://www.youtube.com/watch?v=1_H4AlQaNa0 seems to be a video that explains how to compile from source, it is not mine, but looks good from a first glance at google.
Simona Simmy said:
Can anybody show video hows cyano 11 on alpha ?
thnx
Click to expand...
Click to collapse
It's not useable. No network connection is working yet ...
Thank youu
Thanks to dev(s) from me too. It's interesting to follow your progress, it reminds me of the good old zte blade 1 ages where few devs hacked the **** out of that device.
How can I help ?
Hello guys, I have a Galaxy Alpha for over a week now... And just can't stand that touchwiz stock rom anymore..
So what can I do to help you guys ? I have no experience in porting roms, but I have SM-G850F and just want to help you guys.. For example I can test everything you give me and other things Just let me know
will try to help
Hello, @modpunk
Although I have no CM porting experience I know C and Java. I have set up my CM environment, built and deployed my own unofficial ROM to one of my other devices.
My SM-G850F should arrive this coming week and I just may try help with the TODO. I'm a little confused trying to find out what are your repos to place in my local manifests, can you please list and describe them?
TIA,
Nuno
nuno.f.ferreira said:
Hello, @modpunk
Although I have no CM porting experience I know C and Java. I have set up my CM environment, built and deployed my own unofficial ROM to one of my other devices.
My SM-G850F should arrive this coming week and I just may try help with the TODO. I'm a little confused trying to find out what are your repos to place in my local manifests, can you please list and describe them?
TIA,
Nuno
Click to expand...
Click to collapse
Everything is in the readme of the cm-11.0 rep
http://git.cryptomilk.org/projects/android/android_device_samsung_slte.git/tree/README?h=cm-11.0
Building CM11 for Samsung Galaxy Alpha
Hello,
Was trying to post in http://forum.xda-developers.com/galaxy-alpha/orig-development/sources-cyanogenmod-11-0-t2933305 but apparently need to do it here.
Building with this instructions http://git.cryptomilk.org/projects/android/android_device_samsung_slte.git/tree/README?h=cm-11.0 have the following error. Thought I ask before starting to dig...
<CONSOLE>
target thumb C++: hwcomposer.exynos5 <= hardware/samsung_slsi/exynos5-insignal/libhwc/ExynosHWC.cpp
hardware/samsung_slsi/exynos5-insignal/libhwc/ExynosHWC.cpp: In function 'bool hdmi_is_preset_supported(exynos5_hwc_composer_device_1_t*, int)':
hardware/samsung_slsi/exynos5-insignal/libhwc/ExynosHWC.cpp:462:32: error: aggregate 'hdmi_is_preset_supported(exynos5_hwc_composer_device_1_t*, int)::v4l2_dv_enum_preset enum_preset' has incomplete type and cannot be defined
hardware/samsung_slsi/exynos5-insignal/libhwc/ExynosHWC.cpp:469:45: error: 'VIDIOC_ENUM_DV_PRESETS' was not declared in this scope
hardware/samsung_slsi/exynos5-insignal/libhwc/ExynosHWC.cpp: In function 'int hdmi_get_config(exynos5_hwc_composer_device_1_t*)':
hardware/samsung_slsi/exynos5-insignal/libhwc/ExynosHWC.cpp:577:27: error: aggregate 'hdmi_get_config(exynos5_hwc_composer_device_1_t*)::v4l2_dv_preset preset' has incomplete type and cannot be defined
hardware/samsung_slsi/exynos5-insignal/libhwc/ExynosHWC.cpp:578:32: error: aggregate 'hdmi_get_config(exynos5_hwc_composer_device_1_t*)::v4l2_dv_enum_preset enum_preset' has incomplete type and cannot be defined
hardware/samsung_slsi/exynos5-insignal/libhwc/ExynosHWC.cpp:586:39: error: 'VIDIOC_G_DV_PRESET' was not declared in this scope
hardware/samsung_slsi/exynos5-insignal/libhwc/ExynosHWC.cpp: In function 'int hdmi_output(exynos5_hwc_composer_device_1_t*, hdmi_layer_t&, hwc_layer_1_t&, private_handle_t*, int, int*)':
hardware/samsung_slsi/exynos5-insignal/libhwc/ExynosHWC.cpp:1159:52: error: invalid conversion from '__u32 {aka unsigned int}' to 'v4l2_buf_type' [-fpermissive]
In file included from hardware/samsung_slsi/exynos5-insignal/libhwc/ExynosHWC.h:52:0,
from hardware/samsung_slsi/exynos5-insignal/libhwc/ExynosHWC.cpp:27:
./hardware/samsung_slsi/exynos/include/exynos_v4l2.h:76:5: error: initializing argument 2 of 'int exynos_v4l2_streamon(int, v4l2_buf_type)' [-fpermissive]
hardware/samsung_slsi/exynos5-insignal/libhwc/ExynosHWC.cpp: At global scope:
hardware/samsung_slsi/exynos5-insignal/libhwc/ExynosHWC.cpp:4586:1: warning: missing initializer for member 'hw_module_t::dso' [-Wmissing-field-initializers]
hardware/samsung_slsi/exynos5-insignal/libhwc/ExynosHWC.cpp:4586:1: warning: missing initializer for member 'hw_module_t::reserved' [-Wmissing-field-initializers]
hardware/samsung_slsi/exynos5-insignal/libhwc/ExynosHWC.cpp: In function 'int hdmi_get_config(exynos5_hwc_composer_device_1_t*)':
hardware/samsung_slsi/exynos5-insignal/libhwc/ExynosHWC.cpp:592:1: error: control reaches end of non-void function [-Werror=return-type]
cc1plus: some warnings being treated as errors
make: *** [/opt/android/system/out/target/product/slte/obj/SHARED_LIBRARIES/hwcomposer.exynos5_intermediates/ExynosHWC.o] Error 1
</CONSOLE>
TIA,
Nuno
nuno.f.ferreira said:
Hello,
Was trying to post in http://forum.xda-developers.com/galaxy-alpha/orig-development/sources-cyanogenmod-11-0-t2933305 but apparently need to do it here.
Building with this instructions http://git.cryptomilk.org/projects/android/android_device_samsung_slte.git/tree/README?h=cm-11.0 have the following error. Thought I ask before starting to dig...
<CONSOLE>
target thumb C++: hwcomposer.exynos5 <= hardware/samsung_slsi/exynos5-insignal/libhwc/ExynosHWC.cpp
hardware/samsung_slsi/exynos5-insignal/libhwc/ExynosHWC.cpp: In function 'bool hdmi_is_preset_supported(exynos5_hwc_composer_device_1_t*, int)':
hardware/samsung_slsi/exynos5-insignal/libhwc/ExynosHWC.cpp:462:32: error: aggregate 'hdmi_is_preset_supported(exynos5_hwc_composer_device_1_t*, int)::v4l2_dv_enum_preset enum_preset' has incomplete type and cannot be defined
hardware/samsung_slsi/exynos5-insignal/libhwc/ExynosHWC.cpp:469:45: error: 'VIDIOC_ENUM_DV_PRESETS' was not declared in this scope
hardware/samsung_slsi/exynos5-insignal/libhwc/ExynosHWC.cpp: In function 'int hdmi_get_config(exynos5_hwc_composer_device_1_t*)':
hardware/samsung_slsi/exynos5-insignal/libhwc/ExynosHWC.cpp:577:27: error: aggregate 'hdmi_get_config(exynos5_hwc_composer_device_1_t*)::v4l2_dv_preset preset' has incomplete type and cannot be defined
hardware/samsung_slsi/exynos5-insignal/libhwc/ExynosHWC.cpp:578:32: error: aggregate 'hdmi_get_config(exynos5_hwc_composer_device_1_t*)::v4l2_dv_enum_preset enum_preset' has incomplete type and cannot be defined
hardware/samsung_slsi/exynos5-insignal/libhwc/ExynosHWC.cpp:586:39: error: 'VIDIOC_G_DV_PRESET' was not declared in this scope
hardware/samsung_slsi/exynos5-insignal/libhwc/ExynosHWC.cpp: In function 'int hdmi_output(exynos5_hwc_composer_device_1_t*, hdmi_layer_t&, hwc_layer_1_t&, private_handle_t*, int, int*)':
hardware/samsung_slsi/exynos5-insignal/libhwc/ExynosHWC.cpp:1159:52: error: invalid conversion from '__u32 {aka unsigned int}' to 'v4l2_buf_type' [-fpermissive]
In file included from hardware/samsung_slsi/exynos5-insignal/libhwc/ExynosHWC.h:52:0,
from hardware/samsung_slsi/exynos5-insignal/libhwc/ExynosHWC.cpp:27:
./hardware/samsung_slsi/exynos/include/exynos_v4l2.h:76:5: error: initializing argument 2 of 'int exynos_v4l2_streamon(int, v4l2_buf_type)' [-fpermissive]
hardware/samsung_slsi/exynos5-insignal/libhwc/ExynosHWC.cpp: At global scope:
hardware/samsung_slsi/exynos5-insignal/libhwc/ExynosHWC.cpp:4586:1: warning: missing initializer for member 'hw_module_t::dso' [-Wmissing-field-initializers]
hardware/samsung_slsi/exynos5-insignal/libhwc/ExynosHWC.cpp:4586:1: warning: missing initializer for member 'hw_module_t::reserved' [-Wmissing-field-initializers]
hardware/samsung_slsi/exynos5-insignal/libhwc/ExynosHWC.cpp: In function 'int hdmi_get_config(exynos5_hwc_composer_device_1_t*)':
hardware/samsung_slsi/exynos5-insignal/libhwc/ExynosHWC.cpp:592:1: error: control reaches end of non-void function [-Werror=return-type]
cc1plus: some warnings being treated as errors
make: *** [/opt/android/system/out/target/product/slte/obj/SHARED_LIBRARIES/hwcomposer.exynos5_intermediates/ExynosHWC.o] Error 1
</CONSOLE>
TIA,
Nuno
Click to expand...
Click to collapse
seems you haven't applied this patch: http://review.cyanogenmod.org/#/c/78565/
Thanks for the reply.
Hmm, I have CyanogenMod/android_hardware_samsung_slsi_exynos5-insignal in my local manifest and repos are synced. Do you mean there are patch sets that have to be applied manually? If so, do you know if anymore besides this one?
nuno.f.ferreira said:
Thanks for the reply.
Hmm, I have CyanogenMod/android_hardware_samsung_slsi_exynos5-insignal in my local manifest and repos are synced. Do you mean there are patch sets that have to be applied manually? If so, do you know if anymore besides this one?
Click to expand...
Click to collapse
yes, you have to apply that patch manually. No more patches are needed to build cm at this moment.
Thanks. Did that.
Now I have this. Is this known?
<CONSOLE>
Package OTA: /opt/android/system/out/target/product/slte/cm_slte-ota-67a8ffa7a4.zip
MKBOOTIMG= \
./build/tools/releasetools/ota_from_target_files -v \
-p /opt/android/system/out/host/linux-x86 \
-k build/target/product/security/testkey \
--backup=true \
--override_device=auto \
\
\
/opt/android/system/out/target/product/slte/obj/PACKAGING/target_files_intermediates/cm_slte-target_files-67a8ffa7a4.zip /opt/android/system/out/target/product/slte/cm_slte-ota-67a8ffa7a4.zip
unzipping target target-files...
running: unzip -o -q /opt/android/system/out/target/product/slte/obj/PACKAGING/target_files_intermediates/cm_slte-target_files-67a8ffa7a4.zip -d /tmp/targetfiles-6wcH3r
Traceback (most recent call last):
File "./build/tools/releasetools/ota_from_target_files", line 1132, in <module>
main(sys.argv[1:])
File "./build/tools/releasetools/ota_from_target_files", line 1075, in main
OPTIONS.info_dict = common.LoadInfoDict(input_zip)
File "/opt/android/system/build/tools/releasetools/common.py", line 155, in LoadInfoDict
d["fstab"] = LoadRecoveryFSTab(zip, d["fstab_version"])
File "/opt/android/system/build/tools/releasetools/common.py", line 227, in LoadRecoveryFSTab
raise ValueError("malformed recovery.fstab line: "%s"" % (line,))
ValueError: malformed recovery.fstab line: "/efs ext4 /dev/block/mmcblk0p3 flags=backup=1"
make: *** [/opt/android/system/out/target/product/slte/cm_slte-ota-67a8ffa7a4.zip] Error 1
</CONSOLE>
Hi!
I have been scratching my head over this for 3 days now, no amount of googling and searching here has given me any results.
Build environment is MAC OS X Yosemite with xcode 6 and 5.1.1 command tools installed, 100 gb sparse image file to work on.
Here is the error:
vendor/sony/system/mkqcdtbootimg/mkqcdtbootimg.c: In function ‘load_dtqc_block’:
vendor/sony/system/mkqcdtbootimg/mkqcdtbootimg.c:170: error: ‘fname’ undeclared (first use in this function)
vendor/sony/system/mkqcdtbootimg/mkqcdtbootimg.c:170: error: (Each undeclared identifier is reported only once
vendor/sony/system/mkqcdtbootimg/mkqcdtbootimg.c:170: error: for each function it appears in.)
vendor/sony/system/mkqcdtbootimg/mkqcdtbootimg.c: In function ‘write_padding’:
vendor/sony/system/mkqcdtbootimg/mkqcdtbootimg.c:382: warning: comparison between signed and unsigned
vendor/sony/system/mkqcdtbootimg/mkqcdtbootimg.c: In function ‘main’:
vendor/sony/system/mkqcdtbootimg/mkqcdtbootimg.c:491: warning: pointer targets in passing argument 1 of ‘__builtin___strcpy_chk’ differ in signedness
vendor/sony/system/mkqcdtbootimg/mkqcdtbootimg.c:491: warning: pointer targets in passing argument 1 of ‘__inline_strcpy_chk’ differ in signedness
vendor/sony/system/mkqcdtbootimg/mkqcdtbootimg.c:559: warning: comparison between signed and unsigned
vendor/sony/system/mkqcdtbootimg/mkqcdtbootimg.c:562: warning: comparison between signed and unsigned
vendor/sony/system/mkqcdtbootimg/mkqcdtbootimg.c:566: warning: comparison between signed and unsigned
vendor/sony/system/mkqcdtbootimg/mkqcdtbootimg.c:571: warning: comparison between signed and unsigned
make: *** [out/host/darwin-x86/obj32/EXECUTABLES/mkqcdtbootimg_intermediates/mkqcdtbootimg.o] Error 1
make: *** Waiting for unfinished jobs....
Any help is appretiated
I'm still completely lost on this matter and is driving me crazy
I hit a similar error, but with PATH_MAX as undefined. Basically what it is complaining is that fname is not defined anywhere, which is weird. Try installing Brew (for Mac) and install automake (brew install automake).
To solve my error, I had to edit the file mkqcdtbootimg.c and add these lines after all the C variable defines (right after #include "bootimg.h", in new lines):
#ifndef PATH_MAX
#define PATH_MAX 4096
I used this as a reference: http://serverfault.com/questions/9546/filename-length-limits-on-linux
dferreira said:
I hit a similar error, but with PATH_MAX as undefined. Basically what it is complaining is that fname is not defined anywhere, which is weird. Try installing Brew (for Mac) and install automake (brew install automake).
To solve my error, I had to edit the file mkqcdtbootimg.c and add these lines after all the C variable defines (right after #include "bootimg.h", in new lines):
#ifndef PATH_MAX
#define PATH_MAX 4096
I used this as a reference: http://serverfault.com/questions/9546/filename-length-limits-on-linux
Click to expand...
Click to collapse
Thanks, I´m going to test that brew thing