On Wed, Apr 4, 2018 at 11:35 AM, Brad Hubbard <bhubbard@xxxxxxxxxx> wrote: > On Wed, Apr 4, 2018 at 12:27 PM, Sage Weil <sweil@xxxxxxxxxx> wrote: >> C++ question: is this patch necessary? >> >> https://github.com/liewegas/ceph/commit/116ea627b4dabc1b2339e4e02450487c86433f41 >> >> We want to avoid the OSDShardPGSlot allocation in the common case (where >> the entry already exists). > > Geordi says it is necessary. > > badone | geordi: { using tracked::B; unordered_map<int, unique_ptr<B>> > um; um.insert(make_pair(1, make_unique<B>())); um.emplace(make_pair(1, > nullptr));} > geordi | new(B) B0* B0~ delete(B0) > badone | geordi: { using tracked::B; unordered_map<int, unique_ptr<B>> > um; um.insert(make_pair(1, make_unique<B>())); um.emplace(make_pair(1, > make_unique<B>()));} > geordi | new(B) B0* new(B) B1* B1~ delete(B1) B0~ delete(B0) agreed, i think that's why try_emplace() was added in C++17. -- Regards Kefu Chai -- To unsubscribe from this list: send the line "unsubscribe ceph-devel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
