On Wed, Oct 03, 2018 at 07:44:03PM +0200, Jakub Jelinek wrote: > On Wed, Oct 03, 2018 at 12:30:28PM -0400, Nathan Sidwell wrote: > > On 10/1/18 7:47 PM, Jason Merrill wrote: > > > On Mon, Oct 1, 2018 at 6:41 PM Marek Polacek <pola...@redhat.com> wrote: > > > > > > > > This patch implements C++20 explicit(bool), as described in: > > > > <http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2018/p0892r2.html>. > > > > > > > > I tried to follow the noexcept specifier implementation where I could, > > > > which > > > > made the non-template parts of this fairly easy. To make > > > > explicit(expr) work > > > > with dependent expressions, I had to add DECL_EXPLICIT_SPEC to > > > > lang_decl_fn, > > > > which serves as a vessel to get the explicit-specifier to > > > > tsubst_function_decl > > > > where I substitute the dependent arguments. > > > > > > What's the impact of that on memory consumption? I'm nervous about > > > adding another word to most functions when it's not useful to most of > > > them. For several similar things we've been using hash tables on the > > > side. > > > > Iain had the same question about coroutine state. For the moment we're > > doing the simple thing of growing lang_decl_fn. It seems the better > > approach is a flag in the function_decl (or it's decl_lang_specific > > extension) to say 'go look over there', and a hash table you know will > > contain something. > > Yeah, that is what is used e.g. for DECL_VALUE_EXPR, DECL_DEBUG_EXPR and > many other things.
Thanks. We have 13 spare bits in lang_decl_fn so taking one of them should be fair game. I'll rework my patch as per above. Marek
