problem of compile matrixMul error of compilation

smeil · August 11, 2010, 5:10pm

hello, every body

i try to create my own code of multiplication of square matrix ,but when i would compile it

the message appear :

matrixMul.cu(96): error: a host function call can not be configured

1 error detected in the compilation of “/tmp/tmpxft_0000211c_00000000-4_matrixMul.cpp1.ii”.

you can view the link attached to my post

so if somebody can help me

tera · August 11, 2010, 5:37pm

sorry this is my code

[codebox]

// includes, system

include <stdlib.h>

include <stdio.h>

include <string.h>

include <math.h>

// includes, project

include <cutil_inline.h>

////////////////////////////////////////////////////////////////////////////////

// declaration, forward

void MatrixMulOnDevice(float * M, float * N, float * P, int Width);

void MatrixMulKernel(float * Md, float * Nd, float * Pd, int Width);

float* Md;

float* Nd;
float* Pd;
////////////////////////////////////////////////////////////////////////////////

// Program main

////////////////////////////////////////////////////////////////////////////////

int

main(int argc, char** argv)

{

MatrixMulOnDevice(Md, Nd, Pd, 10);

}

////////////////////////////////////////////////////////////////////////////////

//! Run a simple test for CUDA

////////////////////////////////////////////////////////////////////////////////

void MatrixMulKernel(float * Md, float * Nd, float * Pd, int Width)

{

int tx = threadIdx.x;
int ty = threadIdx.y;
float Pvaleur = 0;
for (int i = 0; i < Width; ++i)

{

    float MdElement = Md[ty * Width + i];

    float NdElement = Nd[i  * Width + tx];

    Pvaleur        += MdElement * NdElement;

}
Pd[ty * Width + tx] = Pvaleur;

}

void MatrixMulOnDevice(float * M, float * N, float * P, int Width)

{
int size = Width*Width * sizeof(float);
// allocate arrays on host
float* Ma = (float*) malloc(size);

float* Na = (float*) malloc(size);

 float* Pa = (float*) malloc(size);
// initialize host memory

for (int i=0; i<size; i++) Ma[i] = (float)i;

for (int j=0; j<size; j++) Na[j] = (float)j;

// allocate array on device

cudaMalloc((void **) &Md, size);
cudaMemcpy(Md, Ma, size, cudaMemcpyHostToDevice) ;

 cudaMalloc((void **) &Nd, size);

cudaMemcpy(Nd, Na, size, cudaMemcpyHostToDevice);

 cudaMalloc((void **) &Pd, size);
dim3 dimGrid(1, 1);

dim3 dimBlock(Width, Width);

//Call of MatrixMulKernel
MatrixMulKernel<<<dimGrid, dimBlock>>>(Md, Nd, Pd, Width);
// copy data from host to device
cudaMemcpy(P, Pd, size, cudaMemcpyDeviceToHost);
//destruction des matrices, dÃ©sormais inutilisÃ©es
free(Ma);

free(Na);

free(Pa);

cudaFree(Md);

cudaFree(Nd);

cudaFree(Pd);
}

[/codebox]

Declare your kernel as [font=“Courier New”]global void MatrixMulKernel(float * Md, float * Nd, float * Pd, int Width)[/font].

tera · August 11, 2010, 5:37pm

sorry this is my code

[codebox]

// includes, system

include <stdlib.h>

include <stdio.h>

include <string.h>

include <math.h>

// includes, project

include <cutil_inline.h>

////////////////////////////////////////////////////////////////////////////////

// declaration, forward

void MatrixMulOnDevice(float * M, float * N, float * P, int Width);

void MatrixMulKernel(float * Md, float * Nd, float * Pd, int Width);

float* Md;

float* Nd;
float* Pd;
////////////////////////////////////////////////////////////////////////////////

// Program main

////////////////////////////////////////////////////////////////////////////////

int

main(int argc, char** argv)

{

MatrixMulOnDevice(Md, Nd, Pd, 10);

}

////////////////////////////////////////////////////////////////////////////////

//! Run a simple test for CUDA

////////////////////////////////////////////////////////////////////////////////

void MatrixMulKernel(float * Md, float * Nd, float * Pd, int Width)

{

int tx = threadIdx.x;
int ty = threadIdx.y;
float Pvaleur = 0;
for (int i = 0; i < Width; ++i)

{

    float MdElement = Md[ty * Width + i];

    float NdElement = Nd[i  * Width + tx];

    Pvaleur        += MdElement * NdElement;

}
Pd[ty * Width + tx] = Pvaleur;

}

void MatrixMulOnDevice(float * M, float * N, float * P, int Width)

{
int size = Width*Width * sizeof(float);
// allocate arrays on host
float* Ma = (float*) malloc(size);

float* Na = (float*) malloc(size);

 float* Pa = (float*) malloc(size);
// initialize host memory

for (int i=0; i<size; i++) Ma[i] = (float)i;

for (int j=0; j<size; j++) Na[j] = (float)j;

// allocate array on device

cudaMalloc((void **) &Md, size);
cudaMemcpy(Md, Ma, size, cudaMemcpyHostToDevice) ;

 cudaMalloc((void **) &Nd, size);

cudaMemcpy(Nd, Na, size, cudaMemcpyHostToDevice);

 cudaMalloc((void **) &Pd, size);
dim3 dimGrid(1, 1);

dim3 dimBlock(Width, Width);

//Call of MatrixMulKernel
MatrixMulKernel<<<dimGrid, dimBlock>>>(Md, Nd, Pd, Width);
// copy data from host to device
cudaMemcpy(P, Pd, size, cudaMemcpyDeviceToHost);
//destruction des matrices, dÃ©sormais inutilisÃ©es
free(Ma);

free(Na);

free(Pa);

cudaFree(Md);

cudaFree(Nd);

cudaFree(Pd);
}

[/codebox]

Declare your kernel as [font=“Courier New”]global void MatrixMulKernel(float * Md, float * Nd, float * Pd, int Width)[/font].

smeil · August 11, 2010, 6:22pm

thank’s now there are no error of complitaion, but I have an other error when i run my program :

*** glibc detected *** ./vectMat: free(): invalid next size (normal): 0x0834f958 ***

======= Backtrace: =========

/lib/tls/i686/cmov/libc.so.6[0xb7d213f4]

/lib/tls/i686/cmov/libc.so.6(cfree+0x96)[0xb7d23456]

/lib/tls/i686/cmov/libc.so.6(fclose+0x144)[0xb7d11004]

/lib/tls/i686/cmov/libc.so.6(get_nprocs+0x16e)[0xb7d92e1e]

/usr/lib/libcuda.so[0xb78e16a0]

/usr/lib/libcuda.so[0xb78d75c1]

/usr/lib/libcuda.so(cuInit+0x41)[0xb793adc1]

/usr/lib/libcudart.so.2[0xb7f6a6a3]

/usr/lib/libcudart.so.2[0xb7f6ab2f]

/usr/lib/libcudart.so.2(cudaMalloc+0x29)[0xb7f4f289]

./vectMat[0x8048ce1]

./vectMat[0x8048ea8]

/lib/tls/i686/cmov/libc.so.6(__libc_start_main+0xe5)[0xb7cc8685]

./vectMat[0x80487c1]

======= Memory map: ========

08048000-0804b000 r-xp 00000000 08:07 876635 /root/NVIDIA_GPU_Computing_SDK/C/bin/linux/release/vectMat

0804b000-0804c000 r–p 00002000 08:07 876635 /root/NVIDIA_GPU_Computing_SDK/C/bin/linux/release/vectMat

0804c000-0804d000 rw-p 00003000 08:07 876635 /root/NVIDIA_GPU_Computing_SDK/C/bin/linux/release/vectMat

0834f000-08370000 rw-p 0834f000 00:00 0 [heap]

b7700000-b7721000 rw-p b7700000 00:00 0

b7721000-b7800000 —p b7721000 00:00 0

b7862000-b7876000 r-xp 00000000 08:07 1419763 /usr/lib/libz.so.1.2.3.3

b7876000-b7878000 rw-p 00013000 08:07 1419763 /usr/lib/libz.so.1.2.3.3

b7878000-b7c5f000 r-xp 00000000 08:07 1418413 /usr/lib/libcuda.so.190.53

b7c5f000-b7c64000 rw-p 003e6000 08:07 1418413 /usr/lib/libcuda.so.190.53

b7c64000-b7c8c000 rw-p b7c64000 00:00 0

b7c8c000-b7c93000 r-xp 00000000 08:07 1025049 /lib/tls/i686/cmov/librt-2.8.90.so

b7c93000-b7c94000 r–p 00007000 08:07 1025049 /lib/tls/i686/cmov/librt-2.8.90.so

b7c94000-b7c95000 rw-p 00008000 08:07 1025049 /lib/tls/i686/cmov/librt-2.8.90.so

b7c95000-b7caa000 r-xp 00000000 08:07 1025045 /lib/tls/i686/cmov/libpthread-2.8.90.so

b7caa000-b7cab000 r–p 00014000 08:07 1025045 /lib/tls/i686/cmov/libpthread-2.8.90.so

b7cab000-b7cac000 rw-p 00015000 08:07 1025045 /lib/tls/i686/cmov/libpthread-2.8.90.so

b7cac000-b7cae000 rw-p b7cac000 00:00 0

b7cae000-b7cb0000 r-xp 00000000 08:07 1025025 /lib/tls/i686/cmov/libdl-2.8.90.so

b7cb0000-b7cb1000 r–p 00001000 08:07 1025025 /lib/tls/i686/cmov/libdl-2.8.90.so

b7cb1000-b7cb2000 rw-p 00002000 08:07 1025025 /lib/tls/i686/cmov/libdl-2.8.90.so

b7cb2000-b7e0a000 r-xp 00000000 08:07 1025019 /lib/tls/i686/cmov/libc-2.8.90.so

b7e0a000-b7e0c000 r–p 00158000 08:07 1025019 /lib/tls/i686/cmov/libc-2.8.90.so

b7e0c000-b7e0d000 rw-p 0015a000 08:07 1025019 /lib/tls/i686/cmov/libc-2.8.90.so

b7e0d000-b7e10000 rw-p b7e0d000 00:00 0

b7e10000-b7e1d000 r-xp 00000000 08:07 1008157 /lib/libgcc_s.so.1

b7e1d000-b7e1e000 r–p 0000c000 08:07 1008157 /lib/libgcc_s.so.1

b7e1e000-b7e1f000 rw-p 0000d000 08:07 1008157 /lib/libgcc_s.so.1

b7e1f000-b7e43000 r-xp 00000000 08:07 1025027 /lib/tls/i686/cmov/libm-2.8.90.so

b7e43000-b7e44000 r–p 00023000 08:07 1025027 /lib/tls/i686/cmov/libm-2.8.90.so

b7e44000-b7e45000 rw-p 00024000 08:07 1025027 /lib/tls/i686/cmov/libm-2.8.90.so

b7e45000-b7e46000 rw-p b7e45000 00:00 0

b7e46000-b7f29000 r-xp 00000000 08:07 1420060 /usr/lib/libstdc++.so.6.0.10

b7f29000-b7f2d000 r–p 000e3000 08:07 1420060 /usr/lib/libstdc++.so.6.0.10

b7f2d000-b7f2e000 rw-p 000e7000 08:07 1420060 /usr/lib/libstdc++.so.6.0.10

b7f2e000-b7f34000 rw-p b7f2e000 00:00 0

b7f34000-b7f78000 r-xp 00000000 08:07 1420231 /usr/lib/libcudart.so.2

b7f78000-b7f79000 r–p 00043000 08:07 1420231 /usr/lib/libcudart.so.2

b7f79000-b7f7a000 rw-p 00044000 08:07 1420231 /usr/lib/libcudart.so.2

b7f87000-b7f89000 rw-p b7f87000 00:00 0

b7f89000-b7fa3000 r-xp 00000000 08:07 1007635 /lib/ld-2.8.90.so

b7fa3000-b7fa4000 r-xp b7fa3000 00:00 0 [vdso]

b7fa4000-b7fa5000 r–p 0001a000 08:07 1007635 /lib/ld-2.8.90.so

b7fa5000-b7fa6000 rw-p 0001b000 08:07 1007635 /lib/ld-2.8.90.so

bf890000-bf8a3000 rwxp bffeb000 00:00 0 [stack]

bf8a3000-bf8a5000 rw-p bfffe000 00:00 0

Aborted

smeil · August 11, 2010, 6:22pm

thank’s now there are no error of complitaion, but I have an other error when i run my program :

*** glibc detected *** ./vectMat: free(): invalid next size (normal): 0x0834f958 ***

======= Backtrace: =========

/lib/tls/i686/cmov/libc.so.6[0xb7d213f4]

/lib/tls/i686/cmov/libc.so.6(cfree+0x96)[0xb7d23456]

/lib/tls/i686/cmov/libc.so.6(fclose+0x144)[0xb7d11004]

/lib/tls/i686/cmov/libc.so.6(get_nprocs+0x16e)[0xb7d92e1e]

/usr/lib/libcuda.so[0xb78e16a0]

/usr/lib/libcuda.so[0xb78d75c1]

/usr/lib/libcuda.so(cuInit+0x41)[0xb793adc1]

/usr/lib/libcudart.so.2[0xb7f6a6a3]

/usr/lib/libcudart.so.2[0xb7f6ab2f]

/usr/lib/libcudart.so.2(cudaMalloc+0x29)[0xb7f4f289]

./vectMat[0x8048ce1]

./vectMat[0x8048ea8]

/lib/tls/i686/cmov/libc.so.6(__libc_start_main+0xe5)[0xb7cc8685]

./vectMat[0x80487c1]

======= Memory map: ========

08048000-0804b000 r-xp 00000000 08:07 876635 /root/NVIDIA_GPU_Computing_SDK/C/bin/linux/release/vectMat

0804b000-0804c000 r–p 00002000 08:07 876635 /root/NVIDIA_GPU_Computing_SDK/C/bin/linux/release/vectMat

0804c000-0804d000 rw-p 00003000 08:07 876635 /root/NVIDIA_GPU_Computing_SDK/C/bin/linux/release/vectMat

0834f000-08370000 rw-p 0834f000 00:00 0 [heap]

b7700000-b7721000 rw-p b7700000 00:00 0

b7721000-b7800000 —p b7721000 00:00 0

b7862000-b7876000 r-xp 00000000 08:07 1419763 /usr/lib/libz.so.1.2.3.3

b7876000-b7878000 rw-p 00013000 08:07 1419763 /usr/lib/libz.so.1.2.3.3

b7878000-b7c5f000 r-xp 00000000 08:07 1418413 /usr/lib/libcuda.so.190.53

b7c5f000-b7c64000 rw-p 003e6000 08:07 1418413 /usr/lib/libcuda.so.190.53

b7c64000-b7c8c000 rw-p b7c64000 00:00 0

b7c8c000-b7c93000 r-xp 00000000 08:07 1025049 /lib/tls/i686/cmov/librt-2.8.90.so

b7c93000-b7c94000 r–p 00007000 08:07 1025049 /lib/tls/i686/cmov/librt-2.8.90.so

b7c94000-b7c95000 rw-p 00008000 08:07 1025049 /lib/tls/i686/cmov/librt-2.8.90.so

b7c95000-b7caa000 r-xp 00000000 08:07 1025045 /lib/tls/i686/cmov/libpthread-2.8.90.so

b7caa000-b7cab000 r–p 00014000 08:07 1025045 /lib/tls/i686/cmov/libpthread-2.8.90.so

b7cab000-b7cac000 rw-p 00015000 08:07 1025045 /lib/tls/i686/cmov/libpthread-2.8.90.so

b7cac000-b7cae000 rw-p b7cac000 00:00 0

b7cae000-b7cb0000 r-xp 00000000 08:07 1025025 /lib/tls/i686/cmov/libdl-2.8.90.so

b7cb0000-b7cb1000 r–p 00001000 08:07 1025025 /lib/tls/i686/cmov/libdl-2.8.90.so

b7cb1000-b7cb2000 rw-p 00002000 08:07 1025025 /lib/tls/i686/cmov/libdl-2.8.90.so

b7cb2000-b7e0a000 r-xp 00000000 08:07 1025019 /lib/tls/i686/cmov/libc-2.8.90.so

b7e0a000-b7e0c000 r–p 00158000 08:07 1025019 /lib/tls/i686/cmov/libc-2.8.90.so

b7e0c000-b7e0d000 rw-p 0015a000 08:07 1025019 /lib/tls/i686/cmov/libc-2.8.90.so

b7e0d000-b7e10000 rw-p b7e0d000 00:00 0

b7e10000-b7e1d000 r-xp 00000000 08:07 1008157 /lib/libgcc_s.so.1

b7e1d000-b7e1e000 r–p 0000c000 08:07 1008157 /lib/libgcc_s.so.1

b7e1e000-b7e1f000 rw-p 0000d000 08:07 1008157 /lib/libgcc_s.so.1

b7e1f000-b7e43000 r-xp 00000000 08:07 1025027 /lib/tls/i686/cmov/libm-2.8.90.so

b7e43000-b7e44000 r–p 00023000 08:07 1025027 /lib/tls/i686/cmov/libm-2.8.90.so

b7e44000-b7e45000 rw-p 00024000 08:07 1025027 /lib/tls/i686/cmov/libm-2.8.90.so

b7e45000-b7e46000 rw-p b7e45000 00:00 0

b7e46000-b7f29000 r-xp 00000000 08:07 1420060 /usr/lib/libstdc++.so.6.0.10

b7f29000-b7f2d000 r–p 000e3000 08:07 1420060 /usr/lib/libstdc++.so.6.0.10

b7f2d000-b7f2e000 rw-p 000e7000 08:07 1420060 /usr/lib/libstdc++.so.6.0.10

b7f2e000-b7f34000 rw-p b7f2e000 00:00 0

b7f34000-b7f78000 r-xp 00000000 08:07 1420231 /usr/lib/libcudart.so.2

b7f78000-b7f79000 r–p 00043000 08:07 1420231 /usr/lib/libcudart.so.2

b7f79000-b7f7a000 rw-p 00044000 08:07 1420231 /usr/lib/libcudart.so.2

b7f87000-b7f89000 rw-p b7f87000 00:00 0

b7f89000-b7fa3000 r-xp 00000000 08:07 1007635 /lib/ld-2.8.90.so

b7fa3000-b7fa4000 r-xp b7fa3000 00:00 0 [vdso]

b7fa4000-b7fa5000 r–p 0001a000 08:07 1007635 /lib/ld-2.8.90.so

b7fa5000-b7fa6000 rw-p 0001b000 08:07 1007635 /lib/ld-2.8.90.so

bf890000-bf8a3000 rwxp bffeb000 00:00 0 [stack]

bf8a3000-bf8a5000 rw-p bfffe000 00:00 0

Aborted

tera · August 11, 2010, 6:30pm

You are scribbling past the end of the host arrays. Change the initialization to
[font=“Courier New”]// initialize host memory
for (int i=0; i<WidthWidth; i++) Ma[i] = (float)i;
for (int j=0; j<WidthWidth; j++) Na[j] = (float)j;
[/font]

tera · August 11, 2010, 6:30pm

You are scribbling past the end of the host arrays. Change the initialization to
[font=“Courier New”]// initialize host memory
for (int i=0; i<WidthWidth; i++) Ma[i] = (float)i;
for (int j=0; j<WidthWidth; j++) Na[j] = (float)j;
[/font]

smeil · August 12, 2010, 1:04pm

thank you very much tera

now i have no problem

smeil · August 12, 2010, 1:04pm

thank you very much tera

now i have no problem

Topic		Replies	Views
Noob question: Kernel configuration failed, invalid configuration argument CUDA Programming and Performance	7	17690	July 31, 2009
Problems of matrix multiplication With and without CUDA CUDA Programming and Performance	15	10017	January 18, 2012
Matrix Multiplication Inconsistency Different values output in every run of the matixMul program CUDA Programming and Performance	29	8760	December 16, 2009
Matrix Addition Failing Simple matrix addition acting up... CUDA Programming and Performance	6	2773	July 16, 2010
matrixMul.cu can't be preprocess then be compiled separately CUDA NVCC Compiler	0	459	December 5, 2022
what's wrong with my code? CUDA Programming and Performance	6	3962	July 7, 2009
Compiling error in Ubuntu 14.04, CUDA Toolkit v7.5 CUDA Programming and Performance	5	2370	November 13, 2015
Subtle problem with matrix-matrix multiplication CUDA Programming and Performance	10	1211	December 12, 2020
Matrix Multiplication Garbage value :( CUDA Programming and Performance	10	3416	July 25, 2009
CUDA 2.1 on VS 2008 Queries regarding compiler settings CUDA Programming and Performance	0	2188	May 20, 2009

problem of compile matrixMul error of compilation

Related topics