Code mistaken as divide by zero

[Lou Cyphre]

Given the following code, and a generated warning

static int64_t  floor_divn(int64_t x_in, int64_t y_in)
{
    int64_t x = x_in;
    int64_t y = y_in;
    int64_t quot = INT64_MAX;

    if (y != 0)
    {
        if ((x > 0) && (y < 0))
        {
            x = -x;
            y = -y;
        }

        if ((x < 0) && (y > 0))
        {
            x = x - y + 1;
        }

        quot = x / y;   // warning: Either the condition 'y>0' is redundant or 
                        // there is division by zero at line ... [zerodivcond]
    }
    return  quot;
}

to me it looks like none of the two hypothesis for a warning is true

[CHR]

Thanks for reporting, ticket is here: https://trac.cppcheck.net/ticket/13292