Hi Igor R,I have one answer to your question and one new question for you and other people.>> Don't the following notes aswer your question?
1. I am sorry but it is no because there is no statement whether the epoll is edge-triggered or level-triggered. It only said like the one shown below.Linux Kernel 2.6
Demultiplexing mechanism:
Usesepoll
for demultiplexing.I downloaded latest boost 1.53.0 and searched for epoll_create, epoll_wait, and many socket APIs for linux and windows but failed to find api calls for those.I expected that I would find api calls for those.2. If it - boost.asio - doesn't call these api, how boost.asio implemens it's functions?Thank you very much.SincerelyJourneyer----------------------------------------
Journeyer J. Joh
o o s a p r o g r a m m e r
a t
g m a i l d o t c o m
----------------------------------------2013/5/29 Igor R <boost.lists@gmail.com>> I have some questions about boost's async IO implementation for linux.Don't the following notes aswer your question?
>
> 1. Does it use edge-triggered epoll to implement the Proactor (in terms of
> Reactor)?
>
> http://www.boost.org/doc/libs/1_53_0/doc/html/boost_asio/overview/core/async.html
>
> The guide above says that boost's Proactor is implemented in terms of
> Reactor when it comes to Linux.
>
> If it is based on Reactor, I guess it should be level-triggered epoll.
> 2. Am I right?
>
> Then my question is
> 3. Does boost use edge-triggered epoll in any kind of async IO on linux?
> 4. Isn't edge-triggered epoll a good method to implement async IO on linux?
http://www.boost.org/doc/libs/1_53_0/doc/html/boost_asio/overview/implementation.html
_______________________________________________
Boost-users mailing list
Boost-users@lists.boost.org
http://lists.boost.org/mailman/listinfo.cgi/boost-users
_______________________________________________
Boost-users mailing list
Boost-users@lists.boost.org
http://lists.boost.org/mailman/listinfo.cgi/boost-users