Konstantin Boyarinov

@lin72h @knoepfel Thank you for expressing your interest into this PR. Current status of this change for us is a low priority enhancement of the FG implementation and there are...

Hi @LXYan2333, Sorry for replying after closing the issue. Unfortunately, for `parallel_reduce` algorithm, we have to copy the identity element into each task to ensure each task has it's own...

@astrojuanlu, could you please check if the newer GLIBCXX ABI can be used as described [here](https://github.com/oneapi-src/oneTBB/issues/383#issuecomment-1308681786)?

@astrojuanlu, There was a typo in the comment I linked above. Setting `_GLIBCXX_USE_CXX11_ABI` to `1`, not `0` should force libstdc++ to use the newest version of `std::string`. Sorry for this...

@astrojuanlu, I am not sure that setting the `_GLIBCXX_USE_CXX11_ABI` environment variable will fix the issue. I guess it should necessarily be a macro. It can be defined by - `#define...

Hi @bayueqiankong, For now `tbb::concurrent_vector` does not allow concurrent erasure of single of multiple elements. And unfortunately, we don't have any "unsafe" versions for such an API. We can consider...

> We doesn't call destructors in these algorithms for already created objects if some constructor throw any exception. For example, implementation at https://en.cppreference.com/w/cpp/memory/uninitialized_fill doing destructor calls int this case. Regarding...

@WalterDehnen, thank you for highlighting this interesting use case. We will consider making some changes in that regard. We will need to discuss this in details because this change introduces...

Hi, @chris-monardo Many thanks for highlighting this and describing your use-case. We will definitely consider extending the priorities API for join_node and other non-functional nodes. As a temporary solution for...

I am not aware (or not remember:)) on what initially caused this issue and why the additional testing was required. If you remind me, I would be able to advice...