[cfe-dev] deque<Incomplete Type> support

[Zhihao Yuan]

On Tue, Nov 12, 2013 at 10:02 PM, Peter Collingbourne <peter at pcc.me.uk> wrote:
> Discussion of a similar patch, which was rejected due to ABI concerns:
>
> http://www.mail-archive.com/cfe-commits@cs.uiuc.edu/msg80909.html

This patch seems has a tiny(?) issue: _LIBCPP_CONSTEXPR
does not fallbacks to inline (constexpr does inline by default),
so if it's not defined, then `__deque_block_size` is going to
run into an odr violation.

BTW, LIBCPP_CONSTEXPR usage on __block_size is
also unneeded.

> http://lists.cs.uiuc.edu/pipermail/cfe-dev/2013-October/032316.html
>
> If you would like to make this change I would suggest doing it via an ABI
> migration.

Well...

BTW, I tried to workaround this issue by detecting incomplete
type.  And, as you can think of, I found no way to escape from an
odr violation.

-- 
Zhihao Yuan, ID lichray
The best way to predict the future is to invent it.
___________________________________________________
4BSD -- http://4bsd.biz/