unsigned char variable is not incremented - c++

I came across one strange behaviour. In my code one variable is decremented, but not incremented and as a result my algorithm does not work. The variable name is blocksAvailable, it is defined in Chunk class, initiated with Chunk::init method, decremented with Chunk::allocate method and must be incremented with Chunk::deallocate method. So, there are just two places where this variable is mentioned - allocate and deallocate methods. In one place it gets decremented (and it works) and in other place it gets incremented and it does not work. This is the completely minimized and reproducible code:
#include <cstddef>
#include <iostream>
#include <vector>
using uchar = unsigned char;
class Chunk
{
private:
friend class FixedAllocator;
void init(size_t blockSize, uchar blocks);
void release();
void* allocate(size_t blockSize);
void deallocate(void* p, size_t blockSize);
inline bool hasBlock(void* p, size_t chunkLen) const
{
uchar * pc = static_cast<uchar*>(p);
return (pData <= pc) && (pc <= (pData + chunkLen));
}
inline bool releasable(uchar numBlocks) const
{
return blocksAvailable == numBlocks;
}
uchar* pData;
uchar firstAvailableBlock, blocksAvailable;
};
void Chunk::init(size_t blockSize, uchar blocks)
{
// for n of Ts it will allocate n * sizeof(T) memory
pData = new uchar[blockSize * blocks];
firstAvailableBlock = 0;
blocksAvailable = blocks;
uchar i = 0;
uchar* p = pData;
// used by allocate method to move forward firstAvailableBlock
for (; i != blocks; p += blockSize)
{
*p = ++i;
}
}
void Chunk::release()
{
::operator delete(pData);
}
void* Chunk::allocate(size_t blockSize)
{
if (!blocksAvailable) return 0;
// move firstAvailableBlock one block ahead
uchar* pResult = pData + firstAvailableBlock * blockSize;
firstAvailableBlock = *pResult;
--blocksAvailable;
std::cout << "blocksAvailable after allocate " << blocksAvailable << std::endl;
return pResult;
}
void Chunk::deallocate(void* p, size_t blockSize)
{
uchar* toRelease = static_cast<uchar*>(p);
// find last but one available block
firstAvailableBlock = static_cast<uchar>((toRelease - pData) / blockSize);
++blocksAvailable;
std::cout << "blocksAvailable after deallocate " << blocksAvailable << std::endl;
}
class FixedAllocator
{
private:
size_t blockSize;
uchar blocks;
using Chunks = std::vector<Chunk>;
Chunks chunks;
Chunk* allocChunk;
public:
FixedAllocator();
~FixedAllocator();
void init(size_t blockSize, size_t pageSize);
const int blockOwner(void* p) const;
void * allocate();
void deallocate(void* p);
};
FixedAllocator::FixedAllocator()
:blockSize(0),
blocks(0),
chunks(0),
allocChunk(nullptr)
{
}
FixedAllocator::~FixedAllocator()
{
Chunks::iterator it;
for (it = chunks.begin(); it != chunks.end(); ++it)
{
it->release();
}
}
void FixedAllocator::init(size_t blockSize_, size_t pageSize)
{
blockSize = blockSize_;
size_t numBlocks = pageSize / blockSize;
blocks = static_cast<uchar>(numBlocks);
}
const int FixedAllocator::blockOwner(void* p) const
{
size_t chunkLen = blocks * blockSize;
std::vector<int>::size_type i = 0, sz = chunks.size();
for (; i < sz; i++)
{
if (chunks[i].hasBlock(p, chunkLen))
{
return i;
}
}
return -1;
}
void* FixedAllocator::allocate()
{
if (!allocChunk || allocChunk->blocksAvailable == 0)
{
Chunks::iterator i = chunks.begin();
for (;;++i)
{
if (i == chunks.end())
{
// allocate memory for one more chunk
chunks.reserve(chunks.size() + 1);
Chunk newChunk;
newChunk.init(blockSize, blocks);
// add new chunk to memory pool
chunks.push_back(newChunk);
// points to new just initiated chunk
allocChunk = &chunks.back();
break;
}
if (i->blocksAvailable > 0)
{
// points to chunk with available blocks
allocChunk = &*i;
break;
}
}
}
return allocChunk->allocate(blockSize);
}
void FixedAllocator::deallocate(void* p)
{
// TODO. Optimize. Now very bruteforce and non-efficient
const int chunkPos = blockOwner(p);
if (chunkPos != -1)
{
Chunk chunk = chunks[chunkPos];
chunk.deallocate(p, blockSize);
// if chunk is releasable, release memory
if (chunk.releasable(blocks))
{
chunk.release();
chunks.erase(chunks.begin() + chunkPos);
// allocChunk may point to deleted chunk
// so, reset it
allocChunk = &chunks.back();
} else {
// there are free blocks in chunk
// so, reset allocChunk for faster future allocation
allocChunk = &chunk;
}
}
}
int main() {
FixedAllocator* alloc = new FixedAllocator();
alloc->init(4, 12);
void* p = alloc->allocate();
void* q = alloc->allocate();
void* r = alloc->allocate();
alloc->deallocate(p);
alloc->deallocate(q);
alloc->deallocate(r);
return 0;
}
As you can see, I have two debug statements in my code. One which prints blocksAvailable value after increment and one which prints its value after decrement.
But this is what I have on my screen, when I compile and run my code:
As you can see, blocksAvailable is initiated with value 3, then it gets decremented three times (three calls to allocate method), but after each decrement (call to deallocate) its value stays the same - 1. It really drives me crazy and looks like some ghost in my code. You can easily reproduce it, compile and run as simply as:
$ g++ main.cpp
$ ./a.out
I hope, someone can help me to find where this ghost appeared from.

Here is the only place in your code where you call Chunk::deallocate:
Chunk chunk = chunks[chunkPos];
chunk.deallocate(p, blockSize);
The first line makes a copy of your Chunk; the second line calls deallocate on it, which increments chunk.blocksAvailable. But chunk is just a copy of the data. Modifying it has no lasting effect.
In particular, chunks[chunkPos] is unaffected and still contains blocksAvailable = 0.

Related

How can I create a destructor to check if the pointer is NULL? If not, delete the memory that has been allocated for the pointer

CheckedArray::CheckedArray(int size) :mSize(size){
int *mArray = new int[size];
for(int i = 0; i < size; i++)
mArray[i] = 0;
}
CheckedArray::~CheckedArray() {
if (mArray == NULL){
return;
}
else {
delete[] mArray;
}
}
I'm using dynamic memory allocation to create a new array. I want to check if the pointer is null, then return. If not, then delete. I'm getting these error messages, but I have no idea what's wrong.
(9094,0x100094600) malloc: *** error for object 0x10001e7b3: pointer being freed was not allocated
(9094,0x100094600) malloc: *** set a breakpoint in malloc_error_break to debug
To be completely clear
CheckedArray::CheckedArray(int size) :mSize(size){
int *mArray = new int[size];
for(int i = 0; i < size; i++)
mArray[i] = 0;
}
should be
CheckedArray::CheckedArray(int size) :mSize(size), mArray(new int[size]){
for(int i = 0; i < size; i++)
mArray[i] = 0;
}
Your version creates a local variable mArray which shadows the class variable of the same name.
Here is an example of what std::unique_ptr can do for you :
#include <iostream>
#include <algorithm>
#include <memory>
// https://isocpp.github.io/CppCoreGuidelines/CppCoreGuidelines#r11-avoid-calling-new-and-delete-explicitly
// so use https://en.cppreference.com/w/cpp/memory/unique_ptr/make_unique
// prefer range based for loops, they can't go out of bounds : https://en.cppreference.com/w/cpp/language/range-for
class dynamic_int_array_t final
{
public:
// creates an "empty array" with enough memory for 4 ints.
dynamic_int_array_t() :
m_capacity{ 4ul }, // start with a capacity for 4 ints.
m_size{ 0ul }, // but actually none are stored yet
m_values{ std::make_unique<int[]>(m_capacity) }
{
}
// allows you to construct an array from a list of integers
dynamic_int_array_t(std::initializer_list<int>&& values) :
m_capacity{ values.size() },
m_size{ values.size() },
m_values{ std::make_unique<int[]>(m_capacity) }
{
std::copy(values.begin(), values.end(), m_values.get());
}
~dynamic_int_array_t() = default; // destructor will destruct unique_ptr and free memory
// non-copyable non-movable (simplifies things for now)
dynamic_int_array_t(const dynamic_int_array_t&) = delete;
dynamic_int_array_t& operator=(const dynamic_int_array_t&) = delete;
dynamic_int_array_t(dynamic_int_array_t&&) = delete;
dynamic_int_array_t& operator=(dynamic_int_array_t&&) = delete;
// begin and end allow range based for loops to work
// range based for loops don't allow you to go out of bounds.
auto begin() const
{
return m_values.get();
}
// end should point "one past" the array (that's how end works)
auto end() const
{
int* ptr = begin();
ptr += m_size;
return ptr;
}
std::size_t size() const
{
return m_size;
}
void add(const int value)
{
// if not enough memory then allocate more
if (m_size == m_capacity) grow_capacity();
// add new value at the end
m_values[m_size] = value;
m_size++;
}
// add another array to this one
void append(const dynamic_int_array_t& rhs)
{
for (int value : rhs)
{
add(value);
}
}
private:
void grow_capacity()
{
m_capacity *= 2;
// allocate new memory
auto tmp = std::make_unique<int[]>(m_capacity);
// copy content to new memory
std::copy(begin(), end(), tmp.get());
// swap new memory with tmp so m_values will now be the newly allocated memory and tmp will hold the previously allocated memory
std::swap(tmp, m_values);
// tmp will go out of scope and delete old buffer
}
std::size_t m_capacity;
std::size_t m_size;
std::unique_ptr<int[]> m_values;
};
int main()
{
dynamic_int_array_t array{ 4,5 };
for (int n = 10; n < 20; ++n)
{
array.add(n);
}
for (const int value : array)
{
std::cout << value << " ";
}
return 0;
}

GCC "AddressSanitizer: heap-buffer-overflow" when initializing struct

I've been writing an VM/Interpreter combination thingy, I don't know how to exactly describe it.
Everything behaved as it should, now before I have hundreds of lines of code, I wanted to go into Garba Collection, because there were some pointers which somehow got lost, in some way. Not that I didn't delete pointers, I created, but they somehow got lost in the proccess of interpreting/running the code.
So, I wanted to track them. I wrote my own "Memory Manager" in some way, it's just a std::vector, where I collect all pointers in.
To track and allocate pointers, I have following code:
struct MemBlock {
bool free;
void* ptr;
size_t size;
};
std::vector<MemBlock*> mem;
size_t max_size;
size_t mem_size;
int count = 0;
void mem_init(size_t maxSize) {
max_size = size/sizeof(MemBlock*);
}
void* mem_alloc(size_t size) {
for (int i = 0; i < count; i++) {
MemBlock* block = mem[i];
if (block->free) {
mem_size -= block->size;
mem_size += size;
block->free = false;
block->ptr = malloc(size);
block->size = size;
if (block->ptr == nullptr) {
throw std::exception();
}
return block->ptr;
}
}
void* ptr = malloc(sizeof(size));
if (ptr == nullptr) {
throw PointerNullException();
}
MemBlock* block = (MemBlock*) malloc(sizeof(MemBlock));
*block = (MemBlock) {
false,
ptr,
size
};
mem_size += size;
count++;
mem.push_back(block);
return block->ptr;
}
But as soon, as I use mem_alloc() and initialize the object inside of the pointer:
Int* i = (Int*) mem_alloc(sizeof(Int));
*i = (Int) {}; // -- Here
i->value = atoi(advance().c_str());
The GCC AdressSanitizer shows following error:
==5939==ERROR: AddressSanitizer: heap-buffer-overflow on address 0x602000000098 at pc 0x555963d82fc5 bp 0x7fff4ec39070 sp 0x7fff4ec39060
WRITE of size 4 at 0x602000000098 thread T0
If I remove said line, then it just occurs on the nex line. The the pointer does point to a valid memory location, if not it should've had thrown an exception.
I'm sure that I missed something/did something wrong, of course.
But I don't know what. This is how I learned it, or at least what I understood...
Edit:
This would be a minimal reproducible Example:
#include <iostream>
#include <stdlib.h>
#include <vector>
struct Object {
const char* type;
};
template <typename T>
struct Primitive : Object {
T value;
};
struct Int : Primitive<int> {
const char* type = "int";
};
struct MemBlock {
bool free;
void* ptr;
size_t size;
};
std::vector<MemBlock*> mem;
size_t mem_size = 0;
int count = 0;
void* mem_alloc(size_t size) {
for (int i = 0; i < count; i++) {
MemBlock* block = mem[i];
if (block->free) {
mem_size -= block->size;
mem_size += size;
block->free = false;
block->ptr = malloc(size);
block->size = size;
if (block->ptr == nullptr) {
throw std::exception();
}
return block->ptr;
}
}
void* ptr = malloc(sizeof(size));
MemBlock* block = (MemBlock*) malloc(sizeof(MemBlock));
*block = (MemBlock) {
false,
ptr,
size
};
mem_size += size;
count++;
mem.push_back(block);
std::cout << "HI" << std::endl;
return block->ptr;
}
void mem_free(void* ptr) {
for (int i = 0; i < count; i++) {
MemBlock* block = mem[i];
if (block->ptr == ptr) {
free(ptr);
mem_size -= block->size;
block->size = 0;
block->ptr = nullptr;
block->free = true;
}
}
}
int main() {
// Create new Integer-Object
Int* i = (Int*) mem_alloc(sizeof(Int));
std::cout << "[Pointer]: " << i << std::endl;
*i = (Int) {};
i->value = 5;
std::cout << "[Value]: " << i->value << std::endl;
}
Well, thanks to Retired Ninja and Richar Critten, I've got the solution.
In mem_alloc() I've used sizeof(size) to allocate memory to the pointer, which of course is wrong. I guess my head was pretty much off after hours of coding.
But I guess this problem is now solved.

Pool allocator implementation

I have two questions about pool allocator implementation:
In given pool allocator implementation how can I check that void*
pointer that I pass to deallocate function is exactly one of those
that I allocated previously?
What should I do to allocate size
of memory bigger than size of block? Just calculate amount of blocks
that would be enough to fit given size and move my next free element pointer x blocks forward, where x is amount of blocks I need ?
class Pool_c { // Basic type define
typedef unsigned int uint;
typedef unsigned char uchar;
uint m_numOfBlocks; // Num of blocks
uint m_sizeOfEachBlock; // Size of each block
uint m_numFreeBlocks; // Num of remaining blocks
uint m_numInitialized; // Num of initialized blocks
uchar* m_memStart; // Beginning of memory pool
uchar* m_next; // Num of next free block
public:
Pool_c()
{
m_numOfBlocks = 0;
m_sizeOfEachBlock = 0;
m_numFreeBlocks = 0;
m_numInitialized = 0;
m_memStart = NULL;
m_next = 0;
}
~Pool_c() { DestroyPool(); }
void CreatePool(size_t sizeOfEachBlock,
uint numOfBlocks)
{
m_numOfBlocks = numOfBlocks;
m_sizeOfEachBlock = sizeOfEachBlock;
m_memStart = new uchar[m_sizeOfEachBlock * m_numOfBlocks];
m_numFreeBlocks = numOfBlocks;
m_next = m_memStart;
}
void DestroyPool()
{
delete[] m_memStart;
m_memStart = NULL;
}
uchar* AddrFromIndex(uint i) const
{
return m_memStart + (i * m_sizeOfEachBlock);
}
uint IndexFromAddr(const uchar* p) const
{
return (((uint)(p - m_memStart)) / m_sizeOfEachBlock);
}
void* Allocate()
{
if (m_numInitialized < m_numOfBlocks) {
uint* p = (uint*)AddrFromIndex(m_numInitialized);
*p = m_numInitialized + 1;
m_numInitialized++;
}
void* ret = NULL;
if (m_numFreeBlocks > 0) {
ret = (void*)m_next;
--m_numFreeBlocks;
if (m_numFreeBlocks != 0) {
m_next = AddrFromIndex(*((uint*)m_next));
}
else {
m_next = NULL;
}
}
return ret;
}
void DeAllocate(void* p)
{
if (m_next != NULL) {
(*(uint*)p) = IndexFromAddr(m_next);
m_next = (uchar*)p;
}
else {
*((uint*)p) = m_numOfBlocks;
m_next = (uchar*)p;
}
++m_numFreeBlocks;
}

C++ Debugging Help Memory Leak

I am currently working on an assignment and I am struggling greatly with debugging my memory leak. My program runs and passes however, when I upload it for grading my code has a memory leak. If someone could point me in the right direction on how to fix it I would greatly appreciate!
I've tried everything I could think of!
#ifndef A2_HPP
#define A2_HPP
#include <algorithm>
class sorted_sc_array {
public:
/*
* return: none
* constructor with no argument assign size_ = 0 and ptr_ to null pointer
*/
sorted_sc_array() : size_(0), ptr_(nullptr) {
}
/*
* return: none
* destructor delete the pointer ptr_
*/
~sorted_sc_array() {
delete[] ptr_;
}
/*
* return: none
* when assign an object to new object
*/
sorted_sc_array(const sorted_sc_array& A){
const signed char* str = A.data();
int sz = A.size_;
this->size_ = 0;
for(int i = 0; i < sz; i++) this->insert(str[i]);
delete[] str;
}
/*
* return: sorted_sc_array
* overloading of operator =
*/
sorted_sc_array& operator=(const sorted_sc_array& A){
const signed char* str = A.data();
int sz = A.size_;
this->size_ = 0;
for(int i = 0; i < sz; i++) this->insert(str[i]);
}
/*
* return int
* return the size of the ptr_
*/
int size() const {
return size_;
}
/*
* return char*
* return the deta stored in the pointer ptr_
*/
const signed char* data() const {
return ptr_;
}
/*
* return void
* add new char to the pointer ptr_ and sort the the new string after the addition
*/
void insert(signed char c) {
signed char *str = (signed char*)malloc((size_ + 1)*sizeof(char));
for(int i = 0; i < size_; i++) str[i] = ptr_[i];
str[size_++] = c;
ptr_ = (signed char*)malloc((size_)*sizeof(char));
for(int i = 0; i < size_; i++) ptr_[i] = str[i];
std::sort(ptr_, ptr_ + size_);
delete[] str;
}
private:
int size_; // size of the array
signed char* ptr_; // pointer to the array
}; // class sorted_sc_array
#endif // A2_HPP
This is the testing class:
/*
* File: a2.pp
* Description: testing class a2.hpp
*/
#include <iostream>
#include "a2.hpp"
int main(int argc, char* argv[]) {
sorted_sc_array A;
{
sorted_sc_array T;
for (signed char c = -128; c < 127; ++c) T.insert(c);
T = T;
sorted_sc_array V = T;
A = V;
}
const auto first = A.data();
const auto last = first + A.size();
auto size = A.size();
bool res = std::is_sorted(first, last);
if (!res || (A.size() != 255)) std::cout << "fail";
else std::cout << "pass";
std::cout << std::endl;
return 0;
} // main
The code compiles and executes, with a "pass" however there is a memory leak somewhere ! :(
There are multiple bugs in the shown code.
In the copy constructor:
delete[] str;
This deletes the other object's buffer. The other object's destructor will attempt to delete[] its own buffer again, when it gets destroyed. This will result in memory corruption, and undefined behavior.
The obvious memory leak is in your insert():
ptr_ = (signed char*)malloc((size_)*sizeof(char));
There are two simultaneous bugs here.
Using malloc for a buffer that the destructor will eventually delete[]. Only new-ed objects can be delete[]d. Using delete for a malloc-ed content is undefined behavior.
The previous contents of ptr_ are not explicitly deleted, thus leaking memory.
The insert(), overall, is doing unnecessary allocation. There's no need to allocate a buffer twice, there. Only one allocation is sufficient: allocate, copy over the content to the new buffer, delete the old ptr_, and set ptr_ to the newly-allocated buffer.

Minimal C++ STL Vector Implementation Problems

I have a technical problem and it's really confusing me. I apologise in advance because I may not be giving the relevant details; I don't yet why it's going wrong and it would be excessive to include all the code I'm working with.
I'm working with a large program that uses the C++ STL. I'm moving this code to a very sensitive environment without a standard clib nor STL implementaton; it will redefine malloc/free/new/delete etc... For that, I need to replace the std:: parts with my own simplified implementations. I've started with std::vector. Right now it's running in the standard ecosystem so it's the GNU libc and STL. The only thing that's changed is this vector class.
When I execute the program with the replaced class, it segfaults. I've put this through GDB and found that the program will request an object from the vector using the subscript operator. When the object reference is returned, a method is invoked and the program segfaults. It seems it can't find this method and ends up in main_arena() in GDB. The type of the object is an inherited class.
I'm really not sure at all what the problem is here. I would love to provide additional details, but I'm not sure what more I can give. I can only assume something is wrong with my vector implementation because nothing else in the program has been changed. Maybe there's something obvious that I'm doing wrong here that I'm not seeing at all.
I'm using: g++ (GCC) 4.4.5 20110214 (Red Hat 4.4.5-6)
I'd really appreciate any feedback/advice!
#ifndef _MYSTL_VECTOR_H_
#define _MYSTL_VECTOR_H_
#include <stdlib.h>
#include <assert.h>
typedef unsigned int uint;
namespace mystl
{
/******************
VECTOR
********************/
template <typename T>
class vector
{
private:
uint _size;
uint _reserved;
T *storage;
void init_vector(uint reserve)
{
if (reserve == 0)
{
_reserved = 0;
return;
}
storage = (T*)malloc(sizeof(T)*reserve);
assert(storage);
_reserved = reserve;
}
public:
vector()
{
// std::cerr << "default constructor " << this << std::endl;
storage = NULL;
_size = 0;
_reserved = 0;
}
vector(const vector<T> &other)
{
// std::cerr << "copy constructor " << this << std::endl;
storage = NULL;
_size = 0;
_reserved = 0;
init_vector(other.size());
_size = other.size();
for (uint i=0; i<other.size(); i++)
{
storage[i] = T(other[i]);
}
}
vector(uint init_num, const T& init_value)
{
// std::cerr << "special constructor1 " << this << std::endl;
storage = NULL;
_size = 0;
_reserved = 0;
init_vector(init_num);
for (size_t i=0; i<init_num; i++)
{
push_back(init_value);
}
}
vector(uint init_num)
{
// std::cerr << "special constructor2 " << this << std::endl;
storage = NULL;
_size = 0;
_reserved = 0;
init_vector(init_num);
}
void reserve(uint new_size)
{
if (new_size > _reserved)
{
storage = (T*)realloc(storage, sizeof(T)*new_size);
assert(storage);
_reserved = new_size;
}
}
void push_back(const T &item)
{
if (_size >= _reserved)
{
if (_reserved == 0) _reserved=1;
reserve(_reserved*2);
}
storage[_size] = T(item);
_size++;
}
uint size() const
{
return _size;
}
~vector()
{
if (_reserved)
{
free(storage);
storage = NULL;
_reserved = 0;
_size = 0;
}
}
// this is for read only
const T& operator[] (unsigned i) const
{
// do bounds check...
if (i >= _size || i < 0)
{
assert(false);
}
return storage[i];
}
T& operator[] (unsigned i)
{
// do bounds check...
if (i >= _size || i < 0)
{
assert(false);
}
return storage[i];
}
// overload = operator
const vector<T>& operator= (const vector<T>& x)
{
// check for self
if (this != &x)
{
_reserved = 0;
_size = 0;
storage = NULL;
init_vector( x.size() );
for(uint i=0; i<x.size(); i++)
{
storage[i] = T(x[i]);
}
_size = x.size();
}
return *this;
}
uint begin() const
{
return 0;
}
void insert(uint pos, const T& value)
{
push_back(value);
if (size() == 1)
{
return;
}
for (size_t i=size()-2; i>=pos&& i>=0 ; i--)
{
storage[i+1] = storage[i];
}
storage[pos] = value;
}
void erase(uint erase_index)
{
if (erase_index >= _size)
{
return;
}
//scoot everyone down by one
for (uint i=erase_index; i<_size; i++)
{
storage[i] = storage[i+1];
}
_size--;
}
void erase(uint start, uint end)
{
if (start > end)
{
assert(false);
}
if (end > _size)
end = _size;
for (uint i=start; i<end; i++)
{
erase(start);
}
assert(false);
}
void clear()
{
erase(0,_size);
}
bool empty() const
{
return _size == 0;
}
}; //class vector
}
#endif // _MYSTL_VECTOR_H_
Wow!
Your assignment operator also leaks memory.
Becuause you are using malloc/release the constructor to your type T will will not be called and thus you can not use your vector for anything except the most trivial of objects.
Edit:
I am bit bored this morning: Try this
#include <stdlib.h> // For NULL
#include <new> // Because you need placement new
// Because you are avoiding std::
// An implementation of swap
template<typename T>
void swap(T& lhs,T& rhs)
{
T tmp = lhs;
lhs = rhs;
rhs = tmp;
}
template <typename T>
class vector
{
private:
unsigned int dataSize;
unsigned int reserved;
T* data;
public:
~vector()
{
for(unsigned int loop = 0; loop < dataSize; ++loop)
{
// Because we use placement new we must explicitly destroy all members.
data[loop].~T();
}
free(data);
}
vector()
: dataSize(0)
, reserved(10)
, data(NULL)
{
reserve(reserved);
}
vector(const vector<T> &other)
: dataSize(0)
, reserved(other.dataSize)
, data(NULL)
{
reserve(reserved);
dataSize = reserved;
for(unsigned int loop;loop < dataSize;++loop)
{
// Because we are using malloc/free
// We need to use placement new to add items to the data
// This way they are constructed in place
new (&data[loop]) T(other.data[loop]);
}
}
vector(unsigned int init_num)
: dataSize(0)
, reserved(init_num)
, data(NULL)
{
reserve(reserved);
dataSize = reserved;
for(unsigned int loop;loop < dataSize;++loop)
{
// See above
new (&data[loop]) T();
}
}
const vector<T>& operator= (vector<T> x)
{
// use copy and swap idiom.
// Note the pass by value to initiate the copy
swap(dataSize, x.dataSize);
swap(reserved, x.rserved);
swap(data, x.data);
return *this;
}
void reserve(unsigned int new_size)
{
if (new_size < reserved)
{ return;
}
T* newData = (T*)malloc(sizeof(T) * new_size);
if (!newData)
{ throw int(2);
}
for(unsigned int loop = 0; loop < dataSize; ++loop)
{
// Use placement new to copy the data
new (&newData[loop]) T(data[loop]);
}
swap(data, newData);
reserved = new_size;
for(unsigned int loop = 0; loop < dataSize; ++loop)
{
// Call the destructor on old data before freeing the container.
// Remember we just did a swap.
newData[loop].~T();
}
free(newData);
}
void push_back(const T &item)
{
if (dataSize == reserved)
{
reserve(reserved * 2);
}
// Place the item in the container
new (&data[dataSize++]) T(item);
}
unsigned int size() const {return dataSize;}
bool empty() const {return dataSize == 0;}
// Operator[] should NOT check the value of i
// Add a method called at() that does check i
const T& operator[] (unsigned i) const {return data[i];}
T& operator[] (unsigned i) {return data[i];}
void insert(unsigned int pos, const T& value)
{
if (pos >= dataSize) { throw int(1);}
if (dataSize == reserved)
{
reserve(reserved * 2);
}
// Move the last item (which needs to be constructed correctly)
if (dataSize != 0)
{
new (&data[dataSize]) T(data[dataSize-1]);
}
for(unsigned int loop = dataSize - 1; loop > pos; --loop)
{
data[loop] = data[loop-1];
}
++dataSize;
// All items have been moved up.
// Put value in its place
data[pos] = value;
}
void clear() { erase(0, dataSize);}
void erase(unsigned int erase_index) { erase(erase_index,erase_index+1);}
void erase(unsigned int start, unsigned int end) /* end NOT inclusive so => [start, end) */
{
if (end > dataSize)
{ end = dataSize;
}
if (start > end)
{ start = end;
}
unsigned int dst = start;
unsigned int src = end;
for(;(src < dataSize) && (dst < end);++dst, ++src)
{
// Move Elements down;
data[dst] = data[src];
}
unsigned int count = start - end;
for(;count != 0; --count)
{
// Remove old Elements
--dataSize;
// Remember we need to manually call the destructor
data[dataSize].~T();
}
}
unsigned int begin() const {return 0;}
}; //class vector
With your current memory handling, this vector would only work with plain old data types.
To handle all types, it must ensure that objects
are actually created (malloc doesn't do that),
destroyed (free doesn't do that),
and you can't reallocate memory with realloc, because complex objects are not guaranteed to remain valid if they are byte-wise copied to another location.
Looks like the answer can be found in your question: "When the object reference is returned, a method is invoked and the program segfaults. It seems it can't find this method and ends up in main_arena() in GDB. The type of the object is an inherited class."
You probably store base class instance T in the vector, but make push_back for the instance of the class inherited from T. In push_back {storage[_size] = T(item);} you cast (actually make copy constructor T:T(const T&)) item to T (this probably named 'type cut'), then get reference to T and invoke a method of the class inherited from T using virtual table of T where the method is not defined yet/abstract. Am I right?
To make it properly work you should put T* in the vector or shared_ptr/unique_ptr depending on the ownership terms you apply to vector elements.
Generally in vector you can store only POD (Plain Old Data) types.