Re: [boost] Re: RFC: Multithreading design constraints

From: Branko Èibej (branko.cibej_at_[hidden])
Date: 2000-06-09 03:59:07

Beman Dawes wrote:
> >(With appropriate locks to protect the class' internal data.)
> 
> Yes, modulo details, that looks promising.  Would it be a bit safer
> if start_thread were a non-member function?  Even though it is
> static, it still has to be called by what is probably a C function
> like pthread_create in create_thread, and that strikes me as a
> possible source of undefined behavior (although I didn't try to
> puzzle through what the standard says about it).

I've no idea. But it's possible it may even have to be an
extern "C" function; C and C++ functions could have different
calling conventions.

    Brane

-- 
Branko Èibej                 <branko.cibej_at_[hidden]>
HERMES SoftLab, Litijska 51, 1000 Ljubljana, Slovenia
voice: ([phone]     fax: [phone]