LLVM ERROR: Error parsing inline asm, unexpected token - clang

I am new to LLVM and clang. I am working on a tool that uses LLVM/clang in the backend as a compiler. With the recent changes in LLVM tool, I have started getting following error:
<inline asm>:1:13: error: unexpected token in argument list
lock addb 2$2, 1048576
^
LLVM ERROR: Error parsing inline asm
I do not have any idea how to resolve this. Any input is appreciated.

Related

formatting c++ file with clang-format in vscode doesn´t work

I´m getting the error message:
YAML:58:21: error: unknown key 'IncludeIsMainRegex'
IncludeIsMainRegex: '$'
Error reading ~/path/to/c++/project/.clang-format: Invalid argument
in vscode when I try to format the cpp file with clang format does anybody how to fix it ?

Can't build OpenCV 3.1 on Windows (using MinGW)

I need to build OpenCV from source, because I want to make modifications to it. So using pre-built binaries is not an option for me. I'm doing this on Windows 8. I do not have MS Visual Studio, and I don't want to use it.
Here is how I reproduce the build error:
I git clone https://github.com/opencv/opencv
I use CMake 3.7.1, press Configure, then press Generate. It seems to work (no errors). I named the output folder opencv/release.
I go to the opencv/release folder, and type mingw32-make.
Eventually, after compiling for a while, it displays the following error message:
Building CXX object modules/core/CMakeFiles/opencv_core.dir/src/parallel.cpp.obj
modules\core\src\parallel.cpp:135:43: error: operator '&&' has no right operand
modules\core\src\parallel.cpp: In function 'void cv::parallel_for_(const cv::Range&, const cv::ParallelLoopBody&, double)':
modules\core\src\parallel.cpp:311:52: error: 'parallel_for_pthreads' was not declared in this scope parallel_for_pthreads(range, body, nstripes);
^
modules\core\src\parallel.cpp: In function 'int cv::getNumThreads()':
modules\core\src\parallel.cpp:370:50: error: 'parallel_pthreads_get_threads_num' was not declared in this scope return parallel_pthreads_get_threads_num();
^
modules\core\src\parallel.cpp: In function 'void cv::setNumThreads(int)':
modules\core\src\parallel.cpp:431:46: error: 'parallel_pthreads_set_threads_num' was not declared in this scope parallel_pthreads_set_threads_num(threads);
^
modules\core\CMakeFiles\opencv_core.dir\build.make:990: recipe for target 'modules/core/CMakeFiles/opencv_core.dir/src/parallel.cpp.obj' failed mingw32-make[2]: *** [modules/core/CMakeFiles/opencv_core.dir/src/parallel.cpp.obj] Error 1
Note: I tried to compile both with and without TPP (in the CMake config), and I get the same error message.
Also, I read the other similar questions on StackOverflow and the answers do not help.

Visual Studio opaque compilation error indication

I am trying to compile OpenCV with Cuda support using Microsoft Visual Studio 2013 and get the following compilation error message:
22> CMake Error at cuda_compile_generated_pyrlk.cu.obj.cmake:264 (message):
22> Error generating file
22> F:/Nir/dev/opencv/build/visualStudioA/modules/cudaoptflow/CMakeFiles/cuda_compile.dir/src/cuda/Debug/cuda_compile_generated_pyrlk.cu.obj
22>
22>
22>C:\Program Files\MSBuild\Microsoft.Cpp\v4.0\V120\Microsoft.CppCommon.targets(170,5): error MSB6006: "cmd.exe" exited with code 1.
Is there a way to get more information about the error?
Increasing log verbosity, or getting the nvcc command that failed could be useful
Cuda compile verbosity can be enabled at CMakeCache.txt file:
//Print out the commands run while compiling the CUDA source file.
// With the Makefile generator this defaults to VERBOSE variable
// specified on the command line, but can be forced on with this
// option.
CUDA_VERBOSE_BUILD:BOOL=ON

armv7 is successful but compile for arm64 throws Implicit Function declaration error

Have to compile security library (create static library) for IOS.
Below is the makefile snippet:
SDK_ROOT=/Applications/Xcode.app/Contents/Developer/Platforms/iPhoneOS.platform/Developer/SDKs/iPhoneOS9.0.sdk
#CC=/Applications/Xcode.app/Contents/Developer/Platforms/iPhoneOS.platform/Developer/usr/bin/llvm-gcc
CC=llvm-gcc
TARGET=security
ARCH=arm64
Now on changing the ARCH=armv7, compilation is successful but for ARCH=arm64
Compiling the same code gives error as below:
crypto.c:580:18: error: implicit declaration of function 'aes_hardware_cbc' is invalid in C99 [-Werror,-Wimplicit-function-declaration]
The above error can be resolved on adding declaration of the function in header file
But I wanted to know how architecture changes brings on Implicit Function Declaration errors or , change warning to errors.
Machine Details:
MAC OS: 10.11 EL Captain
Iphone SDK: iPhoneOS9.0.sdk
Compiler: llvm-gcc (/usr/bin)

Emscripten clang hello_world.cpp link error

I am trying to follow the tutorial on Emscripten setup.I have installed the SDK and now trying to test clang compilation.
Trying simple : clang tests/hello_world.cpp I am getting the following error:
c:\Program Files\Emscripten\emscripten\1.13.0>clang++
tests/hello_world.cpp hello_world-005255.o : error LNK2019:
unresolved external symbol printf referenc ed in function main LINK :
error LNK2001: unresolved external symbol mainCRTStartup hell.exe :
fatal error LNK1120: 2 unresolved externals clang++.exe: error: linker
command failed with exit code 1120 (use -v to see inv ocation)
I am executing the command in VS 2010 cmd and my system is 64bit (Emscripten version is 64bit too)

Resources