Re: [boost] [fiber] ready for next review

Subject: Re: [boost] [fiber] ready for next review
From: Oliver Kowalke (oliver.kowalke_at_[hidden])
Date: 2015-12-04 02:47:32

2015-12-03 22:21 GMT+01:00 Agustín K-ballo Bergé <kaballo86_at_[hidden]>:

> Yes, in the documentation.

documentation might need some updates - my announcement was primarly
focused to the source code

> During the re-review process I raised three conceptual mistakes in the
> documentation, important things and not simply just typos.

which one - I can't remember your concerns. maybe 'this_fiber::yield() (et
al.) are valid from main()' issue?

> I see they are still there in the documentation that corresponds to the
> develop branch, where review feedback has supposedly been addressed.
>
> So I went looking at the code, and while I do see some problems I reported
> have been addressed, others have not. Could you please just let us know
> exactly which pieces of feedback did you address and which did you decide
> to ignore,
>

* addressed:
- cross-thread fiber migration
- suggestions related to scheduler structure
- wait_until() accepts any supported time_point
- mark_ready_and_notify_() accepts std::unique_lock<>
- release mutex before condition_variable signals
- async() does not return by std::move()
- use of intrusive-lists
- re-factoring of future<> and related classes
- use of predicate-based condition_variable::wait() internally

* ignored:
- C++11 equivalent for deferred calls (was impossible to get right for all
use cases in boost.fiber)
- nested schedulers

* todo:
- update documentation