Re: New Developer

[Eric Christopher <echristo cygnus com>]

> In multi-threaded code the 'mutex' calls give you mutually exclusive
> access to the variables -- when you call the mutex function before and
> after checking the sending_mail global you are, in effect, 'locking' it if
> this is what you mean.  Introducing another variable to lock sending_mail
> would just move the potential conflict one level further 'back'.
> 

I see what you mean. I was thinking of sending_mail as a semaphore so
that you could ensure that sending_mail is never changed from external
to the functions that already lock the mutex.

I might just be a bit paranoid.

-eric