The corresponding version of the old "structural hacking" (?)

Question

I believe that I have found a way to achieve something like the famous “hack structure” in the portable C89. I am curious if this really strictly matches the C89.

Main idea: I allocate a large enough memory to contain the initial structure and elements of the array. The exact size is (K + N) * sizeof(array_base_type) , where K is chosen so that K * sizeof(array_base_type) >= sizeof(the_struct) and N number of elements in the array.

First, I will play out the pointer that malloc() returned to store the_struct , then I use pointer arithmetic to get a pointer to the beginning of the array following the structure.

One line of code costs more than a thousand words, so here is a minimal implementation:

 typedef struct Header { size_t length; /* other members follow */ } Header; typedef struct Value { int type; union { int intval; double fltval; } v; } Value; /* round up to nearest multiple of sizeof(Value) so that a Header struct fits in */ size_t n_hdr = (sizeof(Header) + sizeof(Value) - 1) / sizeof(Value); size_t n_arr = 42; /* arbitrary array size here */ void *frame = malloc((n_hdr + n_arr) * sizeof(Value)); if (!frame) return NULL; Header *hdr = frame; Value *stack_bottom = (Value *)frame + n_hdr; 

My main problem is that the last two assignments (using frame as a pointer to a header and a pointer to a value) may violate the strict anti-aliasing rule. However, I am not redefining hdr as a pointer to Value - it is just pointer arithmetic that is executed on frame to access the first element of the array of values, so I cannot effectively access the same object using pointers of different types.

So, is this approach better than the classic hacker (which is officially considered UB), or is he also UB?

Answer 1

The "obvious" (well ... not entirely obvious, but this is what comes to my mind :-)), to make it break, use the vectorization compiler, which somehow decides to load it normally, say 64 Header to the vector register from the 42-rounded up-64 + region in hdr , which comes from malloc , which always allocates enough for vectorization. Saving the vector register back to memory may overwrite one of Value s.

I think this vector compiler can point to a standard (well, if the compiler has fingers ...) and require compliance.

In practice, however, I expect this code to work. If you run into a vectorized compiler, add even more space (round up with a machine-specific macro that can insert a minimum) and charge .:-)