How to get exact result with unsigned long multiplication?

Question

I am struggling with strange result for basic program.

I checked tutorialspoint site and it says unsigned long is 8byte.

I am multiplying unsigned long and unsigned int.

And it seems that multiplying result does not exceeds the maximum value of unsigned long(18,446,744,073,709,551,615). But I get 56557 instead of 85,899,402,477.

I tried also

unsigned long gridTotalSize = (unsigned long)xGrids * yGrids;

but the same result.

I wanted to limit the memory allocation capacity by this multiplying value.

unsigned long gridTotalSize = ((unsigned long )xGrids) * yGrids;
if (gridTotalSize == 1)
{
    return false;
}
else if (gridTotalSize > 1000000)
{
    return false;
}

But because gridTotalSize = 56557, it passes this size limit check, and tries to allocate 85,899,402,477 bytes, this is nearly 85GB, and very frustrating. Every time I debug, I should press power off button and restart computer for completely stuck from memory overflow.

std::vector<Row> rows;
rows.resize(yGrids); //111,321
for (Row& row : rows)
{
    row.resize(xGrids); //771,637
}

I am wasting so many time for this. Any help will be really appreciated.

Answer 1

The width of unsigned long depends on the C or C++ implementation. The C standard and the C++ standard only require it to be at least 32 bits.

To calculate sizes of objects, include <stddef.h> in C or <cstddef> in C++ and use the size_t type in C or std::size_t in C++. It is intended to be a type suitable for working with sizes. (Although the C standard does not explicitly guarantee this. The C++ standard might not either.)

To print a size_t value with printf, use %zu. In C++, it can of course be inserted into a stream in the usual way.