Re: [boost] [AFIO] Formal review

Subject: Re: [boost] [AFIO] Formal review
From: Michael Marcin (mike.marcin_at_[hidden])
Date: 2015-09-12 16:39:52

On 9/2/2015 4:19 AM, Manu343726 wrote:
> Niall Douglas wrote
>> On 31 Aug 2015 at 20:39, Andrey Semashev wrote:
>> Please do correct me if I am wrong, but I had thought that this is
>> defined behaviour:
>>
>> int a=5;
>> void *b=(void *)(size_t) a;
>> int c=(int)(size_t) b;
>> assert(c==a);
>>
>> This is certainly a very common pattern in C.
>
> Although it's a common pattern, it's not correct. It has UB since the
> standard doesn't guarantee that sizeof(void*) == sizeof(int). The issues
> arise in 64 bit systems where memory addresses are 64 bits wide and int is
> 32 bits wide. The safe way to handle integer conversion of pointers is
> through std::intptr_t, which only purpose is to give an integer type
> guaranteed to have same width as ptrs.
>
>

I don't see how a 64bit pointer can't store a 32bit integer.
He's not converting some arbitrary void* to an int in the third line.
He knows it is actually a 32bit int stored in a void*.