On Jun 14, 2006, at 4:56 AM, John Maddock wrote:

Gottlob Frege wrote:

...

...

if unknown_struct is an incomplete type (which is what I meant, and should have said), alignment_of doesn't compile.

If the type is incomplete then you can't use aligned_storage anyway, since you don't know how much storage you'll need! Sorry but you've lost me now.

There are applications such as boost::function where aligned storage for an unknown type is useful. In boost::function's case, aligned storage could be used to implement the "short string optimization". The general pattern is: class Container { private: const unsigned buf_size = ...; aligned_storage::type buffer; public: template <class T> Container(const T& t) {/* might stick t into buffer instead of on heap, based on sizeof(T) */} }; On Jun 14, 2006, at 6:55 AM, David Abrahams wrote:

I think what the OP is looking for is called boost::shared_ptr ;-)

shared_ptr can't compete with this application performance-wise. -Howard

On Jun 14, 2006, at 10:22 AM, David Abrahams wrote:

Howard Hinnant writes:

...

There are applications such as boost::function where aligned storage for an unknown type is useful. In boost::function's case, aligned storage could be used to implement the "short string optimization".

And so it is, AFAIK.

...

The general pattern is:

class Container { private: const unsigned buf_size = ...; aligned_storage::type buffer; public: template <class T> Container(const T& t) {/* might stick t into buffer instead of on heap, based on sizeof(T) */} };

On Jun 14, 2006, at 6:55 AM, David Abrahams wrote:

...

I think what the OP is looking for is called boost::shared_ptr ;-)

shared_ptr can't compete with this application performance-wise.

Well, it depends how expensive it is to copy your small object. If it turns out that T is a vector, and it fits in your local buffer, when Container is copied it might be a lot cheaper to use shared_ptr.

Sure. I'm responding to the assertion that you need a known complete type to effectively use aligned_storage. My point is that there exist valuable use cases where this assertion is false. The existence of these cases is what makes the defaulted alignment template parameter on aligned_storage valuable. Tony's observation is that this default value is even more valuable if it is dependent upon the buffer size, which again I fully agree with. If you want a 5 byte buffer that is aligned to the "most stringent alignment requirements" (definition found in 5.3.4p10), an alignment of 16 (for example) is likely to be correct but overkill. I had personally settled on 8 for this particular example but Tony makes a convincing argument for 4 being the optimum (all numbers assuming current popular desktop processors and quoted just for example). -Howard