Segmentation fault occurs when I'm accessing a struct's member - c

A segmentation fault occurs when I want to access or modify the value of a struct member in C programming language.
I actually write a battleship networked game in C language (it is a student project). I have a structure named Player:
enum state{PLAYING = 0, WINNER = 1, LOOSER = 2};
enum action{ATTACK = 1, WAIT = 0, NOTHING = -1};
typedef struct Player Player;
struct Player
{
enum action action;
enum state state;
char name[25];
int isFirstPlayer;
Client* client;
Server* server;
Boards* boards;
};
the main function.
int main(int argc, char *argv[])
{
if(argc < 2)
{
errorUsage();
}
// Initialise player
Player player = newPlayer();
// I removed the network initialization here.
// play game
play_game(&player);
}
the segmentation fault occurs in the function play_game(Player* p):
void attack(Player* p)
{
char msg[2];
bzero(msg, strlen(msg));
printf("Where do you want to fire?");
scanf("%s", msg);
while(verifyEntryAttack(msg) != 1)
{
printf("ERROR Entry:\nFormat = A4, C9, ...\nColumns = A B C D E F G H I J\nRows = 0 1 2 3 4 5 6 7 8 9\n");
printf("please try again: ");
scanf("%s", msg);
}
sendData(p, msg);
Box my_attack_box;
my_attack_box.abs = msg[0];
my_attack_box.ord = msg[1]-'0';
updateMarkBoard(p, my_attack_box);
}
void receiveAttack(Player* p)
{
char received_attack[2];
receiveData(p, received_attack);
printf("I have received the attack: %s\n", received_attack);
Box my_attack_box;
my_attack_box.abs = received_attack[0];
my_attack_box.ord = received_attack[1]-'0';
endureAttack(my_attack_box, p);
}
void play_game(Player* p)
{
while(p->state == PLAYING){
printPlayer(p); //print the game boards.
if(p->action == ATTACK){
printf("Your turn to attack.\n");
attack(p); //attack the ennemy
printf("end of attack turn.\n");
} else if(p->action == WAIT){
printf("wait for an attack...\n");
receiveAttack(p); //receive the attack from the ennemy.
printf("end of waiting turn.\n");
}
if(p->action == WAIT){
p->action = ATTACK;
}
else{
p->action = WAIT;
}
}
}
After one turn in the while loop, the program says "segmentation fault". My tests show that is the p->action which is the problem.
See below the program output that shows the problem: outputs screenshot.
The entire code is available here: GitLab repo link.
It's like my program can't access the action member of my player.
Does anyone have any ideas?

char msg[2];
bzero(msg, strlen(msg));
Okay, so here you try to calculate the length of an uninitialized string, a typical source of page faults (and kernel vulnerabilities). strlen does not know your array's size, it simply looks for a zero character and is likely to get out of your page very soon. Even if the program does not crash at this point, bzero will do the trick as it is likely to write memory out of your reach.
Note that a cell is usually described with at least two characters ("C4" or so), so scanf will write the terminating zero character past the end of msg (again, this could be even out of your available memory). Increase msg's size so that it can accommodate at least valid inputs (at least three chars, the more the better). And you don't need to call strlen here; if you want to pre-clear your array, the simplest way would be to initialize it properly: char msg[256] = {'\0'};.
Next, you haven't shown your NewPlayer()'s code, but your struct contains three pointer fields, please double-check they are allocated, and initialized, properly.

Related

Segmentation fault while function call

I got a struct Chat
struct Chat
{
int m_FD;
int m_BindPort;
char m_NameLength;
char* m_Name;
char m_PeerCount;
char** m_PeerList;
} typedef Chat_t;
i'm initializing it with this function:
int chat_init(Chat_t* this, unsigned int nameLen, char* name, unsigned short int bindPort, unsigned int peerCount, char** peerList)
{
this->m_NameLength = nameLen;
this->m_Name = malloc(sizeof(char) * (nameLen+1));
strcpy(this->m_Name, name);
this->m_BindPort = bindPort;
this->m_PeerCount = peerCount;
this->m_PeerList = malloc(sizeof(char*) * peerCount);
for(int i=0; i<peerCount; i++)
{
this->m_PeerList[i] = malloc(sizeof(char) * 16); // enough for xxx.xxx.xxx.xxx\0
strcpy(this->m_PeerList[i], peerList[i]);
}
//Socket initialization for TCP connection...
//Commenting this out doesn't change anything so i'm hiding it for simplification
return 0;
}
After that i'm calling a second function
int chat_communicate(Chat_t* this)
{
printf("2\n");
fflush(stdout);
//Some stuff that doesn't matter because it isn't even called
return retVar;
}
in main like this
void main(void)
{
char* peerList[1];
char username[USERNAME_MAX_LEN];
int initRet;
int loopRet;
Chat_t chat;
peerList[0] = "192.168.2.2";
memset(username, 0, USERNAME_MAX_LEN);
printf("Please enter your user name: ");
scanf("%s", username);
username[USERNAME_MAX_LEN-1] = 0;
initRet = chat_init(&chat, strlen(username), username, 1234, 1, peerList);
printf("File Descriptor: %d\n", chat.m_FD);
printf("Binding Port: %d\n", chat.m_BindPort);
printf("Name Length: %d\n", chat.m_NameLength);
printf("Name: %s\n", chat.m_Name);
printf("Peer Count: %d\n", chat.m_PeerCount);
for(int i=0; i< chat.m_PeerCount; i++)
{
printf("Peer[%d]: %s\n", i, chat.m_PeerList[i]);
}
printf("1");
ret = chat_communicate(&chat);
//Even more Stuff that isn't even called
}
My program outputs the following
File Descriptor: 3
Binding Port: 1234
Name Length: 4
Name: User
Peer Count: 1
Peer[0]: 192.168.2.2
1
Segmentation Fault
It compiles without errors or even warnings.
You can also assume that every string is null-Terminated The stuff i replaced with comments itn't that complicated but just too much to show.
Every value inside the struct is printed with printf right before but when passing this very struct per reference the application crashes.
What i want to know is why i'm getting this Segmentation Fault. Since it appeared while calling a function i thought it is some kind of layout problem but i havn't find anything like that.
Addition:
Because some people weren't able to believe me that the code i hid behind "some stuff" comments doesn't change anything i want to state this here once again. This code just contains a tcp socket communication and only performs read-operations. I also am able to reproduce the error mentioned above without this code so please don't get stuck with it. Parts does not influence the object under observation at all.
Among other potential problems,
this->m_PeerList = malloc(sizeof(char)*peerCount);
is clearly wrong.
m_PeerList is a char **, yet you're only allocating peerCount bytes, which only works if a char * pointer is one byte on your system - not likely.
Replace it with something like
this->m_PeerList = malloc(peerCount * sizeof( *( this->m_peerList ) ) );
Note that sizeof( char ) is always one - by definition.
You're not allocating enough memory for the this->m_Name. It should be on more than this if you want it to store the null-terminated string of the name.
That, or we need more information about the peerList.
Now that you have posted an almost complete code, I was able to spot two problems next to each other:
int chat_init(Chat_t* this, unsigned int nameLen, char* name, unsigned short int bindPort, unsigned int peerCount, char** peerList)
{
this->m_NameLength = nameLen;
this->m_Name = malloc(sizeof(char) * (nameLen + 1)); // correct
//< this->m_Name = malloc(sizeof(char) * nameLen); // wrong
strcpy(this->m_Name, name); // correct
//< memcpy(this->m_Name, name, nameLen); // wrong
...
The lines starting with //< is your original code:
Here you don't allocate enough space, you need to account for the NUL terminator:
this->m_Name = malloc(sizeof(char) * nameLen);
And here you don't copy the NUL terminator:
memcpy(this->m_Name, name, nameLen);
You really need to be aware how strings work in C.
Why don't you debug it yourself. If using GCC, compile your code with options -g -O0. Then run it with gdb:
gdb ./a.out
...
(gdb) r
If it crashes do:
(gdb) bt
This will give exactly where it crashes.
Update: There may be potential problems with your code as found by other users. However, memory allocation related issues will not crash your application just on calling function chat_communicate. There could be different reasons for this behaviour ranging from stack overflow to improper compilation. Without seeing the whole code it is very difficult to tell. Best advice is to consider review comments by other users and debug it yourself.

C - Linked List Segmentation Fault During Display

Edit 2: I realized that I did not have a "Not found" result for any query not in the database. Changes have been made to introduce this feature. Here is the current test and test output:
Input:
3
sam
99912222
tom
11122222
harry
12299933
sam
edward
harry
Output:
Not found
=0
Not found
=0
Not found
=0
Not found
=0
sam
=99912222
Not found
=0
Not found
=0
Not found
[Infinite loop continues]
Edit: I have changed a few things in the while loop in display(). I am now getting an infinite loop printing "=0" except for the third or fourth cycle through the search. Hmmm...
By the way, thanks for the reminder of testing strings with ==. Seems like a no-brainer now.
I have done some searching and have yet to be able to understand where I have gone wrong with my code. I am working on a challenge which will result in a simple phone-book program. It will take input of a number (the number of entries to be added) then the names and associated phone numbers (no dashes or periods). After the entries have been added then the user can search for entries by name and have the number displayed in the format "name=number".
The code throws a segmentation fault with the while loop in the display() function. I assume that I am attempting to print something assigned as NULL, but I cannot figure out where I have gone wrong. Any help would be very appreciated.
Lastly, the challenge calls for me to read queries until EOF; however, this confuses me since I am to accept user input from stdin. What does EOF look like with stdin, just a register return (\n)?
(PS: This is my first attempt at linked lists, so any pointers would be greatly appreciated.)
#include <stdio.h>
#include <string.h>
#include <math.h>
#include <stdlib.h>
void add_entry(void);
void display(void);
struct phonebook {
char name[50];
int number;
struct phonebook *next;
};
struct phonebook *firstp, *currentp, *newp;
char tempname[50];
int main() {
int N;
firstp = NULL;
scanf("%d", &N);
for (int i = 0; i < N; i++) {
add_entry();
}
display();
return 0;
}
void add_entry(void) {
newp = (struct phonebook*)malloc(sizeof(struct phonebook));
if (firstp == NULL) firstp = currentp = newp;
else {
currentp = firstp;
while (currentp->next != NULL)
currentp = currentp->next;
currentp->next = newp;
currentp = newp;
}
fgets(currentp->name, 50, stdin);
scanf("%d", &currentp->number);
currentp->next = NULL;
}
void display(void) {
while (strcmp(tempname, "\n") != 0) {
currentp = firstp;
fgets(tempname, 50, stdin);
while (strcmp(currentp->name, tempname) != 0) {
if (currentp->next == NULL) {
printf("Not found\n");
break;
}
currentp = currentp->next;
}
printf("%s=%d\n", currentp->name, currentp->number);
}
}
Your problem is that you never find the entry you're looking for. The expression currentp->name != tempname will always be true, since those are always unequal. In C, this equality test will not compile into a character-by-character comparison, but into a comparison of pointers to currentp->name and tempname. Since those are never at the same addresses, they will never be equal.
Try !strcmp(currentp->name, tempname) instead.
The reason you crash, then, is because you reach the end of the list, so that currentp will be NULL after your loop, and then you try to print NULL->name and NULL->number, actually causing the crash.
Also, on another note, you may want to start using local variables instead of using global variables for everything.
Not sure if this solves the problem, but you can't directly compare strings with != in C. You need to use if( strcmp( string1, string2 ) == 0 ) to check.
fgets doesn't take EOF (= -1) like getchar does, but it does include '\n' and pad the rest with NULL (= 0) so checking for EOF is not really helpful, but yes you can stop after \n or NULL.

syscall read acting weird

c lang, ubuntu
so i have a task - write a menu with these 3 options:
1. close program
2. show user id
3. show current working directory
i can only use 3 libraries - unistd.h, sys/syscall.h, sys/sysinfo.h.
so no printf/scanf
i need to use an array of a struct im given, that has a function pointer,
to call the function the user wants to use.
problem is on options 2 & 3;
when i pick 2, on the first time it works fine (i think)
second time i pick 2, it works, but then when going to the third iteration,
it doesn't wait for an input, it takes '\n' as an input for some reason, then it says invalid input. (i checked what it takes as input with printf, i printed index after recalculating it and it because -39, so it means selection[0] = 10 = '\n')
that's the first problem, that i just cant find the solution for.
second problem is on the current working directory function;
the SYS_getcwd returns -1 for some reason, which means there's an error, but i cant figure it out.
any explanations for these things?
(also - slen and __itoa are functions i am given - slen returns the length of a string,
__itoa returns a char*, that was the string representation of an integer)
helper.h:
typedef struct func_desc {
char *name;
void (*func)(void);
} fun_desc;
main.c:
#include <unistd.h>
#include "helper.h"
#include <sys/syscall.h>
#include <sys/sysinfo.h>
void exitProgram();
void printID();
void currDir();
int main() {
fun_desc arrFuncs[3];
arrFuncs[0].name = "exitProgram";
arrFuncs[0].func = &exitProgram;
arrFuncs[1].name = "printID";
arrFuncs[1].func = &printID;
arrFuncs[2].name = "currDir";
arrFuncs[2].func = &currDir;
char selection[2];
int index;
const char* menu = "Welcome to the menu. Please pick one of the following actions:\n1.Close the program\n2.Print the current user's id\n3.Print the current directory's id\n";
while(1 == 1) {
syscall(SYS_write, 0, menu, slen(menu));
syscall(SYS_write, 0, "Your selection: ", slen("Your selection: "));
syscall(SYS_read, 1, selection, slen(selection)); //might be a problem
selection[1] = '\0';
index = selection[0] - '0' - 1;
if(index > 2)
syscall(SYS_write, 0, "Invalid input\n", slen("Invalid input\n"));
else
arrFuncs[index].func();
}
return(0);
}
void exitProgram() {
syscall(SYS_write, 0, "The program will close\n", slen("The program will close\n"));
syscall(SYS_exit);
}
void printID() { //problem
char* uid = __itoa(syscall(SYS_getuid));
syscall(SYS_write, 0, uid, slen(uid));
syscall(SYS_write, 0, "\n", slen("\n"));
}
void currDir() { //????
char* buf = __itoa(syscall(SYS_getcwd));
syscall(SYS_write, 0, buf, slen(buf));
syscall(SYS_write, 0, "\n", slen("\n"));
}
You're passing the wrong number of arguments to some of these system calls. In particular:
syscall(SYS_exit);
_exit() takes one argument: the exit code.
char* buf = __itoa(syscall(SYS_getcwd));
getcwd() takes two arguments: a pointer to a buffer to write the string to, and the length of that buffer. In practice, this probably looks something like:
char pathbuf[PATH_MAX];
syscall(SYS_getcwd, pathbuf, sizeof(pathbuf));
If you don't have the header which defines PATH_MAX, define it yourself. 4096 is an appropriate value.
Note that getcwd() writes a string into the buffer passed to it — it does not return a numeric identifier.
As an aside, you may want to save yourself some time by implementing a wrapper to write a string, e.g.
void putstring(const char *str) {
syscall(SYS_write, 0, str, slen(str));
}
since you seem to be doing that a lot.

Pointers getting junk values outside of functions, but regular values inside them

Full disclosure: This is my first time doing any significant programming in C, and my first post on Stack Overflow.
I'm working on code that will eventually be used with Bison to implement a small subset of the Scheme/Racket language. All of this code is in a single C file. I have three structs: Binding, Lambda, and SymbolEntry. I'm not using the Lambda struct yet, it's just there for completeness. I also have a symbol table that holds symbol entries. printSymbolTable() does exactly what the name implies:
typedef struct
{
char* name;
char* value;
} Binding;
typedef struct
{
int numBindings;
Binding** bindings;
char* functionBody;
} Lambda;
typedef struct
{
Binding* binding;
Lambda* function;
} SymbolEntry;
SymbolEntry* symbolTable = NULL;
int numSymbols = 0;
void printSymbolTable()
{
if (symbolTable)
{
int i = 0;
for (i; i < numSymbols; i++)
{
printf("\tsymbolTable[%i]: %s = %s\n", i, symbolTable[i].binding->name, symbolTable[i].binding->value);
}
}
}
I'm currently trying to work out the logic for defining and looking up variables. The 2 relevant functions:
// Takes a name and an exprssion and stores the result in the symbol table
void defineVar(char* name, char* expr)
{
printf("\nSetting %s = %s\n", name, expr);
printf("Previous number of symbols: %i\n", numSymbols);
Binding props;
props.name = name;
props.value = expr;
SymbolEntry entry;
entry.binding = &props;
entry.function = NULL;
symbolTable = realloc(symbolTable, sizeof(SymbolEntry) * ++numSymbols);
if (!symbolTable)
{
printf("Memory allocation failed. Exiting.\n");
exit(1);
}
symbolTable[numSymbols - 1] = entry;
printf("New number of symbols: %i\n", numSymbols);
printf("defineVar result:\n");
printSymbolTable();
}
// Test storing and looking up at least 4 variables, including one that is undefined
void testVars()
{
printf("Variable tests\n");
defineVar("foo", "0");
printf("After returning from defineVar:\n");
printSymbolTable();
defineVar("bar", "20");
printf("After returning from defineVar:\n");
printSymbolTable();
}
main() calls testVars(). I get no warnings or errors when compiling, and the program executes successfully. However, this is the result:
Variable tests
Setting foo = 0
Previous number of symbols: 0
New number of symbols: 1
defineVar result:
symbolTable[0]: foo = 0
After returning from defineVar:
symbolTable[0]: 1�I��^H��H���PTI��# = �E
Setting bar = 20
Previous number of symbols: 1
New number of symbols: 2
defineVar result:
symbolTable[0]: bar = 20
symbolTable[1]: bar = 20
After returning from defineVar:
symbolTable[0]: 1�I��^H��H���PTI��# = �E
symbolTable[1]: 1�I��^H��H���PTI��# = �E���
Not only am I getting junk values when outside of the defineVar() function, but the call to define bar shows incorrect non-junk values as well. I'm not sure what I'm doing wrong, but I assume it's probably something with realloc(). However, a similar strategy worked when parsing a string into individual tokens, so that's what I was trying to emulate. What am I doing wrong?
Because it's pointing to variables (or variable — at least props, haven't read further) local to functions and the stack frame is discarded (and soon overwritten) after you return.

C structs strange behaviour

I have some long source code that involves a struct definition:
struct exec_env {
cl_program* cpPrograms;
cl_context cxGPUContext;
int cpProgramCount;
int cpKernelCount;
int nvidia_platform_index;
int num_cl_mem_buffs_used;
int total;
cl_platform_id cpPlatform;
cl_uint ciDeviceCount;
cl_int ciErrNum;
cl_command_queue commandQueue;
cl_kernel* cpKernels;
cl_device_id *cdDevices;
cl_mem* cmMem;
};
The strange thing, is that the output of my program is dependent on the order in which I declare the components of this struct. Why might this be?
EDIT:
Some more code:
int HandleClient(int sock) {
struct exec_env my_env;
int err, cl_err;
int rec_buff [sizeof(int)];
log("[LOG]: In HandleClient. \n");
my_env.total = 0;
//in anticipation of some cl_mem buffers, we pre-emtively init some. Later, we should have these
//grow/shrink dynamically.
my_env.num_cl_mem_buffs_used = 0;
if ((my_env.cmMem = (cl_mem*)malloc(MAX_CL_BUFFS * sizeof(cl_mem))) == NULL)
{
log("[ERROR]:Failed to allocate memory for cl_mem structures\n");
//let the client know
replyHeader(sock, MALLOC_FAIL, UNKNOWN, 0, 0);
return EXIT_FAILURE;
}
my_env.cpPlatform = NULL;
my_env.ciDeviceCount = 0;
my_env.cdDevices = NULL;
my_env.commandQueue = NULL;
my_env.cxGPUContext = NULL;
while(1){
log("[LOG]: Awaiting next packet header... \n");
//read the first 4 bytes of the header 1st, which signify the function id. We later switch on this value
//so we can read the rest of the header which is function dependent.
if((err = receiveAll(sock,(char*) &rec_buff, sizeof(int))) != EXIT_SUCCESS){
return err;
}
log("[LOG]: Got function id %d \n", rec_buff[0]);
log("[LOG]: Total Function count: %d \n", my_env.total);
my_env.total++;
//now we switch based on the function_id
switch (rec_buff[0]) {
case CREATE_BUFFER:;
{
//first define a client packet to hold the header
struct clCreateBuffer_client_packet my_client_packet_hdr;
int client_hdr_size_bytes = CLI_PKT_HDR_SIZE + CRE_BUFF_CLI_PKT_HDR_EXTRA_SIZE;
//buffer for the rest of the header (except the size_t)
int header_rec_buff [(client_hdr_size_bytes - sizeof(my_client_packet_hdr.buff_size))];
//size_t header_rec_buff_size_t [sizeof(my_client_packet_hdr.buff_size)];
size_t header_rec_buff_size_t [1];
//set the first field
my_client_packet_hdr.std_header.function_id = rec_buff[0];
//read the rest of the header
if((err = receiveAll(sock,(char*) &header_rec_buff, (client_hdr_size_bytes - sizeof(my_client_packet_hdr.std_header.function_id) - sizeof(my_client_packet_hdr.buff_size)))) != EXIT_SUCCESS){
//signal the client that something went wrong. Note we let the client know it was a socket read error at the server end.
err = replyHeader(sock, err, CREATE_BUFFER, 0, 0);
cleanUpAllOpenCL(&my_env);
return err;
}
//read the rest of the header (size_t)
if((err = receiveAll(sock, (char*)&header_rec_buff_size_t, sizeof(my_client_packet_hdr.buff_size))) != EXIT_SUCCESS){
//signal the client that something went wrong. Note we let the client know it was a socket read error at the server end.
err = replyHeader(sock, err, CREATE_BUFFER, 0, 0);
cleanUpAllOpenCL(&my_env);
return err;
}
log("[LOG]: Got the rest of the header, packet size is %d \n", header_rec_buff[0]);
log("[LOG]: Got the rest of the header, flags are %d \n", header_rec_buff[1]);
log("[LOG]: Buff size is %d \n", header_rec_buff_size_t[0]);
//set the remaining fields
my_client_packet_hdr.std_header.packet_size = header_rec_buff[0];
my_client_packet_hdr.flags = header_rec_buff[1];
my_client_packet_hdr.buff_size = header_rec_buff_size_t[0];
//get the payload (if one exists)
int payload_size = (my_client_packet_hdr.std_header.packet_size - client_hdr_size_bytes);
log("[LOG]: payload_size is %d \n", payload_size);
char* payload = NULL;
if(payload_size != 0){
if ((payload = malloc(my_client_packet_hdr.buff_size)) == NULL){
log("[ERROR]:Failed to allocate memory for payload!\n");
replyHeader(sock, MALLOC_FAIL, UNKNOWN, 0, 0);
cleanUpAllOpenCL(&my_env);
return EXIT_FAILURE;
}
if((err = receiveAllSizet(sock, payload, my_client_packet_hdr.buff_size)) != EXIT_SUCCESS){
//signal the client that something went wrong. Note we let the client know it was a socket read error at the server end.
err = replyHeader(sock, err, CREATE_BUFFER, 0, 0);
free(payload);
cleanUpAllOpenCL(&my_env);
return err;
}
}
//make the opencl call
log("[LOG]: ***num_cl_mem_buffs_used before***: %d \n", my_env.num_cl_mem_buffs_used);
cl_err = h_clCreateBuffer(&my_env, my_client_packet_hdr.flags, my_client_packet_hdr.buff_size, payload, &my_env.cmMem);
my_env.num_cl_mem_buffs_used = (my_env.num_cl_mem_buffs_used+1);
log("[LOG]: ***num_cl_mem_buffs_used after***: %d \n", my_env.num_cl_mem_buffs_used);
//send back the reply with the error code to the client
log("[LOG]: Sending back reply header \n");
if((err = replyHeader(sock, cl_err, CREATE_BUFFER, 0, (my_env.num_cl_mem_buffs_used -1))) != EXIT_SUCCESS){
//send the header failed, so we exit
log("[ERROR]: Failed to send reply header to client, %d \n", err);
log("[LOG]: OpenCL function result was %d \n", cl_err);
if(payload != NULL) free(payload);
cleanUpAllOpenCL(&my_env);
return err;
}
//now exit if failed
if(cl_err != CL_SUCCESS){
log("[ERROR]: Error executing OpenCL function clCreateBuffer %d \n", cl_err);
if(payload != NULL) free(payload);
cleanUpAllOpenCL(&my_env);
return EXIT_FAILURE;
}
}
break;
Now what's really interesting is the call to h_clCreateBuffer. This function is as follows
int h_clCreateBuffer(struct exec_env* my_env, int flags, size_t size, void* buff, cl_mem* all_mems){
/*
* TODO:
* Sort out the flags.
* How do we store cl_mem objects persistantly? In the my_env struct? Can we have a pointer int the my_env
* struct that points to a mallocd area of mem. Each malloc entry is a pointer to a cl_mem object. Then we
* can update the malloced area, growing it as we have more and more cl_mem objects.
*/
//check that we have enough pointers to cl_mem. TODO, dynamically expand if not
if(my_env->num_cl_mem_buffs_used == MAX_CL_BUFFS){
return CL_MEM_OUT_OF_RANGE;
}
int ciErrNum;
cl_mem_flags flag;
if(flags == CL_MEM_READ_WRITE_ONLY){
flag = CL_MEM_READ_WRITE;
}
if(flags == CL_MEM_READ_WRITE_OR_CL_MEM_COPY_HOST_PTR){
flag = CL_MEM_READ_WRITE | CL_MEM_COPY_HOST_PTR;
}
log("[LOG]: Got flags. Calling clCreateBuffer\n");
log("[LOG]: ***num_cl_mem_buffs_used before in function***: %d \n", my_env->num_cl_mem_buffs_used);
all_mems[my_env->num_cl_mem_buffs_used] = clCreateBuffer(my_env->cxGPUContext, flag , size, buff, &ciErrNum);
log("[LOG]: ***num_cl_mem_buffs_used after in function***: %d \n", my_env->num_cl_mem_buffs_used);
log("[LOG]: Finished clCreateBuffer with id: %d \n", my_env->num_cl_mem_buffs_used);
//log("[LOG]: Finished clCreateBuffer with id: %d \n", buff_counter);
return ciErrNum;
}
The first time round the while loop, my_env->num_cl_mem_buffs_used is increased by 1. However, the next time round the loop, after the call to clCreateBuffer, the value of my_env->num_cl_mem_buffs_used gets reset to 0. This does not happen when I change the order in which I declare the members of the struct! Thoughts? Note that I've omitted the other case statements, all of which so similar things, i.e. updating the structs members.
Well, if your program dumps the raw memory content of the object of your struct type, then the output will obviously depend on the ordering of the fields inside your struct. So, here's one obvious scenario that will create such a dependency. There are many others.
Why are you surprised that the output of your program depends on that order? In general, there's nothing strange in that dependency. If you base your verdict of on the knowledge of the rest of the code, then I understand. But people here have no such knowledge and we are not telepathic.
It's hard to tell. Maybe you can post some code. If I had to guess, I'd say you were casting some input file (made of bytes) into this struct. In that case, you must have the proper order declared (usually standardized by some protocol) for your struct in order to properly cast or else risk invalidating your data.
For example, if you have file that is made of two bytes and you are casting the file to a struct, you need to be sure that your struct has properly defined the order to ensure correct data.
struct example1
{
byte foo;
byte bar;
};
struct example2
{
byte bar;
byte foo;
};
//...
char buffer[];
//fill buffer with some bits
(example1)buffer;
(example2)buffer;
//those two casted structs will have different data because of the way they are defined.
In this case, "buffer" will always be filled in the same manner, as per some standard.
Of course the output depends on the order. Order of fields in a struct matters.
An additional explanation to the other answers posted here: the compiler may be adding padding between fields in the struct, especially if you are on a 64 bit platform.
If you are not using binary serialization, then your best bet is an invalid pointer issue. Like +1 error, or some invalid pointer arithmetics can cause this. But it is hard to know without code. And it is still hard to know, even with the code. You may try to use some kind of pointer validation/tracking system to be sure.
other guesses
by changing the order you are having different uninitialized values in the struct. A pointer being zero or not zero for example
you somehow manage to overrun an item (by casting ) and blast later items. Different items get blasted depending on order
This may happen if your code uses on "old style" initializer as from C89. For a simple example
struct toto {
unsigned a;
double b;
};
.
.
toto A = { 0, 1 };
If you interchange the fields in the definition this remains a valid initializer, but your fields are initialized completely different. Modern C, AKA C99, has designated initializers for that:
toto A = { .a = 0, .b = 1 };
Now, even when reordering your fields or inserting a new one, your initialization remains valid.
This is a common error which is perhaps at the origin of the initializerphobia that I observe in many C89 programs.
You have 14 fields in your struct, so there is 14! possible ways the compiler and/or standard C library can order them during the output.
If you think from the compiler designer's point of view, what should the order be? Random is certainly not useful. The only useful order is the order in which the struct fields were declared by you.

Resources