[Bug 21568] advapi32/crypt tests crash with +heap

http://bugs.winehq.org/show_bug.cgi?id=21568

--- Comment #7 from Andrew Nguyen   2010-07-13 03:33:13 ---
It seems reasonable to believe that an HCRYPTKEY handle that CryptGenKey
outputs is not a direct pointer to a data structure. Given that it's typedefed
as a ULONG_PTR, and doing something (admittedly crude) like:

HCRYPTKEY p;
BOOL result;
for (p = 0; p < 0x00400000; p++)
{
    result = pCryptDestroyKey(p);
    ok (!result && GetLastError() == ERROR_INVALID_PARAMETER, "%d\n",
GetLastError());
}

does not induce a crash, nor is there any evidence of the process heap being
corrupted (HeapValidate), this seems to suggest that some sort of handle table
should be employed.