System: cuda 11.3, gcc 7.5, boost 1.65.1, pcl 1.8.0
When I compile code that uses PCL library, it shows the following error
/usr/include/pcl-1.8/pcl/io/file_io.h(264): error: namespace "boost" has no member "numeric_cast"
/usr/include/pcl-1.8/pcl/io/file_io.h(264): error: type name is not allowed
/usr/include/pcl-1.8/pcl/io/file_io.h(280): error: namespace "boost" has no member "numeric_cast"
/usr/include/pcl-1.8/pcl/io/file_io.h(280): error: type name is not allowed
/usr/include/pcl-1.8/pcl/io/file_io.h(346): error: namespace "boost" has no member "iequals"
/usr/include/pcl-1.8/pcl/io/file_io.h(372): error: namespace "boost" has no member "iequals"
/usr/include/pcl-1.8/pcl/io/pcd_io.h(485): error: name followed by "::" must be a class or namespace name
/usr/include/pcl-1.8/pcl/io/pcd_io.h(493): error: name followed by "::" must be a class or namespace name
I looked into the file_io.h, found that the related code is <pcl/io/boost.h>. In this file, the header <boost/numeric/conversion/cast.hpp> contains the numeria_cast function, obviously this header is not included. Is this error related to the macro __CUDACC__? How do I solve this issue? /usr/include/pcl-1.8/pcl/io/boost.h is:
#ifndef _PCL_IO_BOOST_H_
#define _PCL_IO_BOOST_H_
#if defined __GNUC__
# pragma GCC system_header
#endif
#ifndef __CUDACC__
//https://bugreports.qt-project.org/browse/QTBUG-22829
#ifndef Q_MOC_RUN
#include <boost/version.hpp>
#include <boost/numeric/conversion/cast.hpp>
#include <boost/thread/mutex.hpp>
#include <boost/thread/condition.hpp>
#include <boost/thread.hpp>
#include <boost/thread/thread.hpp>
#include <boost/filesystem.hpp>
#include <boost/bind.hpp>
#include <boost/cstdint.hpp>
#include <boost/function.hpp>
#include <boost/tuple/tuple.hpp>
#include <boost/shared_ptr.hpp>
#include <boost/weak_ptr.hpp>
#include <boost/mpl/fold.hpp>
#include <boost/mpl/inherit.hpp>
#include <boost/mpl/inherit_linearly.hpp>
#include <boost/mpl/joint_view.hpp>
#include <boost/mpl/transform.hpp>
#include <boost/mpl/vector.hpp>
#include <boost/algorithm/string.hpp>
#ifndef Q_MOC_RUN
#include <boost/date_time/posix_time/posix_time.hpp>
#endif
#if BOOST_VERSION >= 104700
#include <boost/chrono.hpp>
#endif
#include <boost/tokenizer.hpp>
#include <boost/foreach.hpp>
#include <boost/shared_array.hpp>
#include <boost/interprocess/sync/file_lock.hpp>
#if BOOST_VERSION >= 104900
#include <boost/interprocess/permissions.hpp>
#endif
#include <boost/iostreams/device/mapped_file.hpp>
#define BOOST_PARAMETER_MAX_ARITY 7
#include <boost/signals2.hpp>
#include <boost/signals2/slot.hpp>
#endif
#endif
#endif // _PCL_IO_BOOST_H_
Right, I solved it by using normal gcc compiler... DONT USE NVCC!
Related
I have 2 files that form a small CUDA library from my previous program (which works well, btw) written on C++.
The header for this library is:
#ifndef __cudaLU__
#define __cudaLU__
#include <assert.h>
#include <cuda_runtime.h>
#include <cusolverDn.h>
#include <cusolverSp.h>
#include <cusparse.h>
#include <cuComplex.h>
#include <stdlib.h>
void denseLS(int dim,
std::complex<float> * A,
std::complex<float> * b );
void sparseLS(int dim,
std::complex<float> *csrVal,
int *csrRowPtr,
int *csrColInd,
std::complex<float> *vecVal);
#endif
And I want to use this library in my old-as-the-hills C program just by setting procedure in the head of my main.c file:
extern void denseLS(int dim, float complex *A, float complex *b);
And it fails with a bunch of similar errors. Few of them are:
..NA/cudaLS.cu(115): error: namespace "std" has no member "complex"
..NA/cudaLS.cu(115): error: expected a ")"
..NA/cudaLS.cu(137): error: identifier "csrRowPtr" is undefined
..NA/cudaLS.cu(169): error: identifier "csrColInd" is undefined
..NA/cudaLS.cu(170): error: identifier "csrVal" is undefined
..NA/cudaLS.cu(171): error: identifier "vecVal" is undefined
I tried to make a change std::complex -> float complex and nothing works. Still same errors (without std error, ofc).
The cmake instructions file
cmake_minimum_required(VERSION 3.8)
project(NA)
set(CMAKE_C_STANDARD 11)
find_package(GSL REQUIRED)
find_package(CUDA REQUIRED)
include_directories("${CUDA_INCLUDE_DIRS}")
cuda_add_library(solvers STATIC
cudaLS.cu
cudaLS.h)
target_link_libraries(solvers ${CUDA_LIBRARIES} ${CUDA_cusparse_LIBRARY} ${CUDA_cusolver_LIBRARY})
target_compile_features(solvers PUBLIC cxx_std_11)
set_target_properties( solvers
PROPERTIES CUDA_SEPARABLE_COMPILATION ON)
add_executable(NA main.c)
set_target_properties(NA PROPERTIES CUDA_SEPARABLE_COMPILATION ON)
target_link_libraries(NA PRIVATE GSL::gsl m solvers)
What am I doing wrong pals?
UPD:
g++/gcc - 7.3
Linux
Well, I found what exactly I did in a wrong way.
Cmake is OK. But headers in the .h file have to be modified to
extern "C" void denseLS(int dim, cuComplex *A, cuComplex *b );
The cuda functions in .c have to be decleared in the head (or separate .h-file) as
void denseLS(int dim, float complex *A, float complex *b);
I have a plain C code (running on Linux) and I would like to implement it in Octave, so I thought I could use a mex-file for handling the memory mapping and send the information I received (or send) back and forth to my script in Octave and my sensors. The C code looks like this:
#include <stdio.h>
#include <stdint.h>
#include <stdlib.h>
#include <poll.h>
#include <fcntl.h>
#include <errno.h>
#include <unistd.h>
#include <sys/mman.h>
#define CUSTOM_IP_MAP_SIZE 0x10000
#define CUSTOM_IP_BASEADDR 0x43C00000
#define CUSTOM_IP_S00_AXI_SLV_REG0_OFFSET 0
#define CUSTOM_IP_S00_AXI_SLV_REG1_OFFSET 4
int main(void)
{
uint32_t leds=0x0;
int fd = open("/dev/uio0", O_RDWR);
void *ptr;
if (fd < 0) {
perror("open");
exit(EXIT_FAILURE);
}
ptr = mmap(NULL, CUSTOM_IP_MAP_SIZE, PROT_READ|PROT_WRITE, MAP_SHARED, fd, 0);
while (1) {
leds = *((unsigned *)(ptr + CUSTOM_IP_S00_AXI_SLV_REG1_OFFSET)); //Read from the IP (slv_reg1).
*((unsigned *)(ptr + CUSTOM_IP_S00_AXI_SLV_REG0_OFFSET)) = leds; //Write to the IP (slv_reg0).
}
close(fd);
exit(EXIT_SUCCESS);
}
I compiled the code with no errors and the following command:
mkoctfile --mex mmap.c
I get the following error when I run it in Octave:
error: failed to install .mex file function 'mmap'
Should I keep trying to do this with a mex-function or there is other option better for this?
Thank you for any help.
Is there any way to find the rank of element in a matrix row separately using CUDA or any functions for the same provided by NVidia?
I don't know of a built-in ranking or argsort function in CUDA or any of the libraries I am familiar with.
You could certainly build such a function out of lower-level operations using thrust for example.
Here is a (non-optimized) outline of a possible solution approach using thrust:
$ cat t84.cu
#include <thrust/device_vector.h>
#include <thrust/copy.h>
#include <thrust/sort.h>
#include <thrust/sequence.h>
#include <thrust/functional.h>
#include <thrust/adjacent_difference.h>
#include <thrust/transform.h>
#include <thrust/iterator/permutation_iterator.h>
#include <iostream>
typedef int mytype;
struct clamp
{
template <typename T>
__host__ __device__
T operator()(T data){
if (data == 0) return 0;
return 1;}
};
int main(){
mytype data[] = {4,1,7,1};
int dsize = sizeof(data)/sizeof(data[0]);
thrust::device_vector<mytype> d_data(data, data+dsize);
thrust::device_vector<int> d_idx(dsize);
thrust::device_vector<int> d_result(dsize);
thrust::sequence(d_idx.begin(), d_idx.end());
thrust::sort_by_key(d_data.begin(), d_data.end(), d_idx.begin(), thrust::less<mytype>());
thrust::device_vector<int> d_diff(dsize);
thrust::adjacent_difference(d_data.begin(), d_data.end(), d_diff.begin());
d_diff[0] = 0;
thrust::transform(d_diff.begin(), d_diff.end(), d_diff.begin(), clamp());
thrust::inclusive_scan(d_diff.begin(), d_diff.end(), d_diff.begin());
thrust::copy(d_diff.begin(), d_diff.end(), thrust::make_permutation_iterator(d_result.begin(), d_idx.begin()));
thrust::copy(d_result.begin(), d_result.end(), std::ostream_iterator<int>(std::cout, ","));
std::cout << std::endl;
}
$ nvcc -arch=sm_61 -o t84 t84.cu
$ ./t84
1,0,2,0,
$
If you are in CUDA, the concept rank is not the same as the one on other languages as openmp or mpi. On that case you will need to go on a global block of the code you need to work with threadIdx.x and blockIdx.x parameters
I'm trying to compile the following piece of code:
#include <stdio.h>
#include <time.h>
#include <cuda.h>
#include <cuda_runtime_api.h>
texture<float, 2, cudaReadModeElementType> tex;
int main () { ... }
yet, nvcc gives me the following error:
main.c:6:8: error: expected ‘=’, ‘,’, ‘;’, ‘asm’ or ‘__attribute__’ before ‘<’ token
I'm pretty new to CUDA, so I suppose I'm missing something here.
You can only use CUDA syntax in .cu files.
Has anybody tried something as simple as #include <vector> in an application for the Mac compiled with XE2?
Is something so basic broken in XE2 update 1 for C++ Builder or is my install broken?
Simply adding #include <vector> to a new fire monkey HD app, I get build failures:
[BCC32 Error] cstdlib(43): E2015 Ambiguity between 'ldiv_t' and 'Posix::Stdlib::ldiv_t'
Full parser context
Unit1.cpp(7): #include C:\Program Files\Embarcadero\RAD Studio\9.0\include\boost_1_39\boost\tr1\tr1\vector
vector(16): #include C:\Program Files\Embarcadero\RAD Studio\9.0\include\boost_1_39\boost/tr1/detail/config_all.hpp
config_all.hpp(48): #include c:\program files\embarcadero\rad studio\9.0\include\dinkumware\cstdlib
cstdlib(32): namespace std
From the code:
//---------------------------------------------------------------------------
#include <fmx.h>
#pragma hdrstop
#include "Unit1.h"
#include <vector>
//---------------------------------------------------------------------------
#pragma package(smart_init)
#pragma resource "*.fmx"
TForm1 *Form1;
//---------------------------------------------------------------------------
__fastcall TForm1::TForm1(TComponent* Owner)
: TForm(Owner)
{
}
//---------------------------------------------------------------------------
Solution from Embarcadero (this is for update 2 only):
Open cstdlib in the [RADStudioXE2]\include\dinkumware directory
Take line 49: using _CSTD size_t; using _CSTD div_t; using _CSTD ldiv_t;
Move it above the "#if defined" block right above it (line 33)
The STL can now be compiled into an FMX C++ application
try removing $(CG_BOOST_ROOT) from your include paths.