Re: [boost] [FSM] Review library version update

From: Steven Watanabe (steven_at_[hidden])
Date: 2007-02-05 16:31:36

AMDG

Andreas Huber <ahd6974-spamgroupstrap <at> yahoo.com> writes:

> I should have been clearer. The standard sais that *initializing* a POD 
> is not a problem because that is done before any threads have a chance 
> to race against each other (typically when the binary is loaded into 
> memory).
> <snip>
> 
>     static int & GetPodInstance()
>     {
>       static int podInstance = 42;
>       return podInstance;
>     }
> <snip>
> When multiple threads are present, there's no race condition when more 
> than one of them call GetPodInstance(). Clearly there is one when they 
> call GetNonPodInstance() because NonPod::NonPod() will non run before 
> the first call...
> 

GetPodInstance() is probably safe, but the standard does
not require it. "A local object of POD type (3.9) with static
storage duration initialized with constant-expressions is
initialized before its block is first entered." (6.7/4)

In Christ,
Steven Watanabe