Limit template parameters in hashtable
Jonathan Wakely
jwakely.gcc@gmail.com
Sun Jul 21 17:56:00 GMT 2013
On 17 July 2013 21:53, François Dumont wrote:
> On 07/16/2013 11:03 AM, Jonathan Wakely wrote:
>>
>> On 15 July 2013 21:36, François Dumont wrote:
>>>
>>> The drawback is of course that it introduces a binary breaking
>>> change,
>>> is it fine ?
>>
>> Did we already do that for 4.9 anyway?
>
> AFAIK no, no binary incompatibility so far. But I can't see how to do this
> simplification without this breaking change. _M_bbegin could be restore in
> its previous place but the allocator instance must be managed by
> _Hashtable_alloc so it will necessarily generate this breaking change.
Ah OK. I like this change, I think it should produce smaller code
when multiple hash tables are used in the same program.
Benjamin, Paolo, do you have any opinion on the ABI change for 4.9? It
happens because the _M_bbegin member moves to a base class, so
_hashtable changes layout, with members in a different order.
More information about the Libstdc++
mailing list