I am trying to use this (http://theboostcpplibraries.com/boost.mpi-simple-data-exchange#ex.mpi_01) example 47.2 to understand the valgrind outputs for my own boost mpi code ran with valgrind. This following code ran fine and I got "99" back. However, when I ran it with valgrind, I got "Syscall param write(buf) points to uninitialised byte(s)" complaints.
#include <boost/mpi.hpp>
#include <iostream>
int main(int argc, char *argv[])
{
boost::mpi::environment env{argc, argv};
boost::mpi::communicator world;
if (world.rank() == 0)
{
int i;
world.recv(1, 16, i);
std::cout << i << '\n';
}
else if (world.rank() == 1)
{
world.send(0, 16, 99);
}
}
Syscall param write(buf) points to uninitialised byte(s)
at 0x37BEA0E6FD: ??? (in /lib64/libpthread-2.12.so)
.......
by 0x409B70: main (e47.2.cpp:6)
Address 0x7feffec48 is on thread 1's stack
.......
Related
Using VS2013, in the C++ code below I get a debug assertion when accessing the vector with an out of range index, which is OK (but the catch block is not reached). When running the release version the program runs without the exception being caught. The output is 1 while it should be 0.
#include <iostream>
#include <vector>
#include <stdexcept>
using namespace std;
vector<int> Vector;
void GetException()
try{
int Sum{ 0 };
// Access an element beyond the end of the vector
for( int i = 0; i <= Vector.size(); i++ )
Sum += Vector[i];
}
catch( ... ){
Vector.clear();
}
int _tmain(int argc, _TCHAR* argv[])
{
Vector.push_back( 1 );
GetException();
cout << Vector.size() << endl;
return 0;
}
When I change it to access a nullptr I get in the debug version an error message for an "unhandled exception" and in the release version it crashes on the access. Are there parameters in VS that modify the behaviour of exceptions?
int Sum{ 0 };
void GetException()
try{
int *pSum{ nullptr };
Sum = *pSum;
}
catch( ... ){
Sum = 1;
}
int _tmain(int argc, _TCHAR* argv[])
{
GetException();
cout << Sum << endl;
return 0;
}
This expression Vector[i] (in your first code example) does not throw exceptions (in normal, release builds anyways), it's simply undefined behavior if i is out of bounds.
This expression Vector.at(i) does throw exceptions (in normal, release builds) if i is out of bounds.
If you want to have a std::vector do bounds checking for you, and throw exceptions if an access is out of bounds, then you should use that form. You can't catch the problem if a Vector[i] access is out of bounds.
I am following a source code on my documents, but I encounter an error when I try to use MPI_Send() and MPI_Recv() from Open MPI library.
I have googled and read some threads in this site but I can not find the solution to resolve my error.
This is my error:
mca_oob_tcp_msg_recv: readv faled : Unknown error (108)
Here is details image:
And this is the code that I'm following:
#include <stdio.h>
#include <string.h>
#include <conio.h>
#include <mpi.h>
int main(int argc, char **argv) {
int rank, size, mesg, tag = 123;
MPI_Status status;
MPI_Init(&argv, &argv);
MPI_Comm_size(MPI_COMM_WORLD, &size);
MPI_Comm_rank(MPI_COMM_WORLD, &rank);
if (size < 2) {
printf("Need at least 2 processes!\n");
} else if (rank == 0) {
mesg = 11;
MPI_Send(&mesg,1,MPI_INT,1,tag,MPI_COMM_WORLD);
MPI_Recv(&mesg,1,MPI_INT,1,tag,MPI_COMM_WORLD,&status);
printf("Rank 0 received %d from rank 1\n",mesg);
} else if (rank == 1) {
MPI_Recv(&mesg,1,MPI_INT,0,tag,MPI_COMM_WORLD,&status);
printf("Rank 1 received %d from rank 0/n",mesg);
mesg = 42;
MPI_Send(&mesg,1,MPI_INT,0,tag,MPI_COMM_WORLD);
}
MPI_Finalize();
return 0;
}
I commented all of MPI_Send(), and MPI_Recv(), and my program worked. In other hand, I commented either MPI_Send() or MPI_Recv(), and I still got that error. So I think the problem are MPI_Send() and MPI_Recv() functions.
P.S.: I'm using Open MPI v1.6 on Windows 8.1 OS.
You pass in the wrong arguments to MPI_Init (two times argv, instead of argc and argv once each).
The sends and receives actually look fine, I think. But there is also one typo in one of your prints with a /n instead of \n.
Here is what works for me (on MacOSX, though):
int main(int argc, char **argv) {
int rank, size, mesg, tag = 123;
MPI_Status status;
MPI_Init(&argc, &argv);
MPI_Comm_size(MPI_COMM_WORLD, &size);
MPI_Comm_rank(MPI_COMM_WORLD, &rank);
if (size < 2) {
printf("Need at least 2 processes!\n");
} else if (rank == 0) {
mesg = 11;
MPI_Send(&mesg,1,MPI_INT,1,tag,MPI_COMM_WORLD);
MPI_Recv(&mesg,1,MPI_INT,1,tag,MPI_COMM_WORLD,&status);
printf("Rank 0 received %d from rank 1\n",mesg);
} else if (rank == 1) {
MPI_Recv(&mesg,1,MPI_INT,0,tag,MPI_COMM_WORLD,&status);
printf("Rank 1 received %d from rank 0\n",mesg);
mesg = 42;
MPI_Send(&mesg,1,MPI_INT,0,tag,MPI_COMM_WORLD);
}
MPI_Finalize();
return 0;
}
If this does not work, I'd guess your OS does not let the processes communicate with each other via the method chosen by OpenMPI.
Set MPI_STATUS_IGNORED instead of &status in MPI_Recv in both places.
I'm trying to add a custom system call into the linux kernel. Here is a simple code:
#include <linux/mysyscall.h>
#include <linux/kernel.h>
#include <asm/uaccess.h>
#include <asm/system.h>
asmlinkage int sys_mysyscall(int *data){
int a = 3;
cli();
copy_to_user(data, &a, 1);
sti();
printk(KERN_EMERG "Called with %d\n", a);
return a;
}
I can compile a kernel with mysyscall added and when I try to access it with a user program like:
#include <linux/mysyscall.h>
int main(void){
int *data;
int r;
int a = 0;
data = &a;
r = mysyscall(data);
printf("r is %d and data is %d", r, *data);
}
*data does not equal to 3 it equals to 0.
How should I use copy_to_user to fix it?
The copy to user line of code copies only one byte from 'a'. In case of little endian systems it is going to be 0. Copy all the 4 bytes to get the correct result.
I am using gcc compiler to Implement a random-number generator using only getpid() and gettimeofday(). Here is my code
#include <stdio.h>
#include <sys/time.h>
#include <sys/time.h>
#include <time.h>
#include <stdlib.h>
int main(int argc, char **argv)
{
struct timeval tv;
int count;
int i;
int INPUT_MAX =10;
int NO_OF_SAMPLES =10;
gettimeofday(&tv, NULL);
printf("Enter Max: \n");
scanf("%d", &INPUT_MAX);
printf("Enter No. of samples needed: \n");
scanf("%d", &NO_OF_SAMPLES);
/*printf("%ld\n",tv.tv_usec);
printf("PID :%d\n", getpid());*/
for (count = 0; count< NO_OF_SAMPLES; count++) {
printf("%ld\n", (getpid() * tv.tv_usec) % INPUT_MAX + 1);
for (i = 0; i < 1000000; ++i)
{
/* code */
}
}
return 0;
}
I gave a inner for loop for delay purpose but the result what i am getting is always same no. like this
./a.out
Enter Max:
10
Enter No. of samples needed:
10
1
1
1
1
1
1
1
1
1
1
Plz correct me what am i doing wrong?
getpid() is constant during the programs execution, so you get constant values, too.
But even if you use gettimeofday() inside the loop, this likely won't help:
gcc will likely optimize away your delay loop.
even it it's not optimized away, the delays will be very similar and your values won't be very random.
I'd suggest you look up "linear congruential generator", for a simple way to generate more random numbers.
Put gettimeofday in the loop. Look if getpid() is divisible by INPUT_MAX + 1 you will get the same answer always. Instead you can add getpid() (not make any sense though()) to tv.tv_usec.
good afternoon.
I got the code below on a book. I'm trying to execute it, but I don't know what is the "first" and "last" parameters on the MakeCodeWritable function, or where I can find them. Someone can help? This code is about C obfuscation method. I'm using Xcode program and LLVM GCC 4.2 compiler.
#include <stdio.h>
#include <sys/mman.h>
#include <stdlib.h>
#include <unistd.h>
#include <sys/types.h>
typedef unsigned int uint32;
typedef char* caddr_t;
typedef uint32* waddr_t;
#define Tam_celula 64
#define ALIGN __attribute__((aligned(Tam_celula)))
void makeCodeWritable(char* first, char* last) {
char* firstpage = first - ((int)first % getpagesize());
char* lastpage = last - ((int)last % getpagesize());
int pages = (lastpage-firstpage)/getpagesize()+1;
if (mprotect(firstpage,pages*getpagesize(), PROT_READ|PROT_EXEC|PROT_WRITE)==-1) perror("mprotect");
}
void xor(caddr_t from, caddr_t to, int len){
int i;
for(i=0;i<len;i++){
*to ^= *from; from++; to++;
} }
void swap(caddr_t from, caddr_t to, int len){
int i;
for(i=0;i<len;i++){
char t = *from; *from = *to; *to = t; from++; to++;
} }
#define CELLSIZE 64
#define ALIGN asm volatile (".align 64\n");
void P() {
static int firsttime=1; if (firsttime) {
xor(&&cell5,&&cell2,CELLSIZE);
xor(&&cell0,&&cell3,CELLSIZE);
swap(&&cell1,&&cell4,CELLSIZE);
firsttime = 0; }
char* a[] = {&&align0,&&align1,&&align2,&&align3,&&align4,&&align5};
char*next[] ={&&cell0,&&cell1,&&cell2,&&cell3, &&cell4,&&cell5};
goto *next[0];
align0: ALIGN
cell0: printf("SPGM0\n");
xor(&&cell0,&&cell3,3*CELLSIZE);
goto *next[3];
align1: ALIGN
cell1: printf("SPGM2\n"); xor(&&cell0,&&cell3,3*CELLSIZE);
goto *next[4];
align2: ALIGN
cell2: printf("SPGM4\n"); xor(&&cell0,&&cell3,3*CELLSIZE);
goto *next[5];
align3: ALIGN
cell3: printf("SPGM1\n"); xor(&&cell3,&&cell0,3*CELLSIZE);
goto *next[1];
align4: ALIGN
cell4: printf("SPGM3\n"); xor(&&cell3,&&cell0,3*CELLSIZE);
goto *next[2];
align5: ALIGN
cell5: printf("SPGM5\n");
xor(&&cell3,&&cell0,3*CELLSIZE);
}
int main (int argc, char *argv[]) {
makeCodeWritable(...);
P(); P();
}
The first argument should be (char *)P, because it looks like you want to modify code inside function P. The second argument is the ending address of function P. You can first compile the code, and using objdump -d to see the address of beginning and end of P, then calculate the size of the function, SIZE, then manually specify in the makeCodeWritable( (char *)P, ((char *)P) + SIZE.
The second way is utilizing the as to get the size of function P, but it depends on the assembler language on your platform. This is code snipe I modified from your code, it should be able to compile and run in x86, x86_64 in GCC 4.x on Linux platform.
align5: ALIGN
cell5: printf("SPGM5\n");
xor(&&cell3,&&cell0,3*CELLSIZE);
// adding an label to the end of function P to assembly code
asm ("END_P: \n");
;
}
extern char __sizeof__myfunc[];
int main (int argc, char *argv[]) {
// calculate the code size, ending - starting address of P
asm (" __sizeof__myfunc = END_P-P \n");
// you can see the code size of P
printf("code size is %d\n", (unsigned)__sizeof__myfunc);
makeCodeWritable( (char*)P, ((char *)P) + (unsigned)__sizeof__myfunc);
P(); P();
}
With some modification to support LLVM GCC and as in Mac OS X
int main (int argc, char *argv[]) {
size_t sizeof__myfunc = 0;
asm volatile ("movq $(_END_P - _P),%0;"
: "=r" (sizeof__myfunc)
: );
printf("%d\n", sizeof__myfunc);