Re: [boost] a new release of the Join asynchronous concurrency library, please schedule a review

From: Mathias Gaunard (mathias.gaunard_at_[hidden])
Date: 2007-07-19 10:35:56

Yigong Liu wrote:

> to do next:
> . better, earlier error reporting (compile time).
>   now many errors related to chord definitions are thrown at runtime as
> exceptions. since the type info is already available, we should be able to
> report them during compilation using some template programming. In boost,
> what should i use similar to Andrei Alexandrescu's STATIC_CHECK()?

There is a BOOST_STATIC_ASSERT

> . more optimization, profiling and benchmarking
>   with the same source code and simple Jamfile, the tutorials are much
> slower in Windows/VC++ than in Linux/g++ (my box has Centrino Duo).

You might want to directly look at the assembler output, it might simply 
have to do with some inlining that isn't done.

> More info can be found:
> Source code: http://sourceforge.net/project/showfiles.php?group_id=157583
> Documentation:
> http://channel.sourceforge.net/boost_join/libs/join/doc/boost_join_design.html
> Website: http://channel.sourceforge.net

I see that the string is taken by value in your small examples.
Shouldn't it be taken by const-reference instead?

Same for async_o etc.

> I'd really like (need) suggestions and corrections from more experts on this
> small library to make sure i am on the right track. Could a initial review
> be arranged for the Join library?

I have a few questions:

- why are you using boost.function? Aren't regular function objects enough?
- how is concurrency handled? Locks? What kind? Is the scheduling fair?