I am currently working on a contiki os port for a custom platform based on the TI CC2650 chipset (and srf06-cc26xx sample plattform).
When trying to call the cc26xxware function ti_lib_timer_int_register(GPT1_BASE, TIMER_A, trf_timer_fxn);
I am getting a linker error.
The exact error message is:
cc2650.c:(.text.trf_startClock+0x4a): undefined reference to `NOROM_TimerIntRegister'
collect2: error: ld returned 1 exit status
However, other function calls from the same library (e.g. ti_lib_timer_int_enable(), etc.) are linked without any issues.
The function trf_timer_fxn is declared as static void trf_timer_fxn(void). Removing the static did not change anything.
I am using the latest contiki os version from github and have updated to the current cc26xxware's. Building is performed using a modified version of the sample Makefile that usually works very well.
Can anybody reproduce this behaviour? Any ideas what else I could try?
Thank you very much!
In case anyone stumbles upon this, the answer actually lies within the README of the contiki plattform.
You have to rename the cc26xxware timer.c into driverlib-timer.c as contiki already uses a file called timer.c
Related
I know there is a lot of similar errors out there, but i tried doing exactly what the answers said, and nothing helped.
I have now even created a completely new DemoProj (React-Native app), linked the repository the error is in, and i STILL get the same error.
The repository can be found here: https://github.com/Renetnielsen/DemoProj
If one fetches it and runs xcode, and runs that, you would get the same error.
I have no solution to this what so ever, especially since im no Objective C programmer.
Please please help me!!
The module react-native-appswitch is a small 2 function module, that just exports some IOS SDK functionality (https://developer.mobilepay.dk/appswitch-integration)
Complete Stacktrace: https://github.com/Renetnielsen/DemoProj/blob/master/README.md
Solutions tried
- Create completely new app, refered repo, installed, got error
- Ran on device, same error.
- Cleaned build folder, pods and so on - still error
- Linking binary files and so on
It needs to be fixed using 64 bit architecture, so using 32 is not an option, since ios does not support that any more as far as i know.
while building an upnp based project on xcode, it is compiling and building successfully and the app is working fine but in logs it is generating warning : NPT_CHECK failed over and over.
So my question is : what exactly NPT_CHECK is and also what exactly NPT_ERROR_CANCELLED is ??
I found that NPT_CHECK is in fact a function defined in platinum framework source code which is being used for upnp connections and if the output is coming like this because of conditions mentioned in functions are not met. If you are facing same problems then just trace the error to its definition.
I have a code base which under Xcode get the error "bad instruction" for every NEON instruction I call. It basically seems like NEON is not detected.
I am attempting to build a static library, I went to New Project, selected Cocoa Touch Static Library, then added my existing files.
Everything I'm reading indicates that NEON should be already enabled. I removed all references to armv6, and am targeting iOS 5.1
Also the code in question is all contained as routines defined in ".s" files -- pure assembly. I am not using the intrinsic method calls.
this is the error which I get whenever I try to run the code:
unknown directive .fpu neon
Command /Applications/Xcode.app/Contents/Developer/Toolchains/
XcodeDefault.xctoolchain /usr/bin/clang failed with exit code 1
also when I delete .fpu neon command from my code, it compiles and I get the .o file but then it fails to link as I still am not able to use the programs defined in the code file.
Try the answer on my similar question https://stackoverflow.com/a/10507325/571778
Sort answer, in my case, was I was porting assembly from another compiler. A few points:
Xcode requires all lower case instructions
The pseudo-ops are different (try http://www.shervinemami.info/armAssembly.html#template)
you must start your assembly function names with "_", because that's how the linker finds them (in C, call "foo()", but in ASM name your function "_foo")
I've spent the last hour or two trying to hunt this problem down and can't seem to get it right. I am using OpenCV with MS VS 2012 RC, and to this point it has worked quite well (one bug dealing with the newer Mat container, but fine otherwise). I have a previously written application I am trying to port (from GCC/G++), which used the cvblob.dll library from here. I have recompiled the cvblob library with VS2012 (my previous version was compiled with MinGW and worked flawlessly), and managed to create both a DLL and an import library .LIB.
The application compiles properly, but the one spot where I use the cvblob library is now my source of error, because the linker cannot find the symbol for the cvLabel function in the library. This is the complete error message I get from VS2012:
ConsoleApplication1.cpp
1> Generating Code...
1>imageinput.obj : error LNK2019: unresolved external symbol _cvLabel referenced in function "public: class std::vector,class std::allocator > > _thiscall ImageInput::getROI(class cv::Mat,class cv::Mat)" (?getROI#ImageInput##QAE?AV?$vector#V?$Rect#H#cv##V?$allocator#V?$Rect_#H#cv###std###std##VMat#cv##0#Z)
1>C:\Users\Jake\Documents\Visual Studio 2012\Projects\ConsoleApplication1\Debug\ConsoleApplication1.exe : fatal error LNK1120: 1 unresolved externals
I'm not certain whether this is a problem with the cvblob library itself, my current binary for it, or just the process I used to compile it. I will gladly furnish any information I can upon request. Thanks in advance for your advice!
You obviously didn't link needed library in project settings. Try to add all OpenCV/CvBlob libs in linker properties of your prooject. See docs.
Also look at this discussion.
I have a make file from an OpenCV port for Android. Since this project seems fairly dead, there haven't been any updates and the last supported NDK version is 4b. With this NDK version everything compiles just find and the code also is working. However If I build it using a newer version I get the following error.
C:/Android/NDK/toolchains/arm-linux-androideabi-4.4.3/prebuilt/windows/bin/../lib/gcc/arm-linux-androideabi/4.4.3/../../../../arm-linux-androideabi/bin/ld.exe: cannot find -lcxcore
collect2: ld returned 1 exit status
make: *** [/cygdrive/c/Users/<Name>/workspace/<Project>/obj/local/armeabi/libopencv.so] Error 1
I think it is caused by this line:
LOCAL_LDLIBS := -L$(SYSROOT)/usr/lib -ldl -llog \
-L$(TARGET_OUT) -lcxcore -lcv -lcvaux -lcvml -lcvhighgui
However when I check where $(TARGET_OUT) points and see if the library is there, it is available. I think the problem is that I am getting a libcxcore.a file and might need a libcscore.so file in order for this to work. But I am not really sure.
From what I have found on the internet it seems like the linker might be causing problems for many reasons and I really wonder how this could be the case since the code itself works just fine. I hope someone can tell me how to either change my make files so that it will also work with newer NDKs or point me into a direction of where to look for the problem.
For me ld.exe could not find libs under path specified, because I was using cygwin-style path format:
-L/cygdrive/c/Users/tkozak/dev/Android/libs
After changing this to normal windows path:
-LC:/Users/tkozak/dev/Android/libs
Linker worked properly.