Re: [boost] [thread] Bug sprint: #2100 thread fails to compile with -fno-exceptions

Subject: Re: [boost] [thread] Bug sprint: #2100 thread fails to compile with -fno-exceptions
From: Andrey Semashev (andrey.semashev_at_[hidden])
Date: 2009-11-25 11:47:36

Anthony Williams wrote:
> Vicente Botet Escriba <vicente.botet_at_[hidden]> writes:
> 
>> the use of   boost::throw_exception is welcome but more for the ability this
>> function provides to store a pointer to an exception than the ability to
>> disable exception. If in addition you use some ifdefs to mask features that
>> are not easy to provide without exceptions the reporter of the ticket could
>> use Boost.Thread on an environment that has or  requires no exceptions.
> 
> Thanks.

Perhaps, Boost.Thread could provide a callback for thread interruption, 
like throw_exception does. The user could release the resources 
associated with the thread and then call something like 
pthread_kill(pthread_self()). Just a thought.