Re: [Chicken-users] recursive mutex-lock!

[Daishi Kato]

At Mon, 25 Feb 2008 00:52:25 -0800,
Vincent Manis wrote:
> 
> On 2008 Feb 25, at 00:19, Daishi Kato wrote:
> 
> > Hi,
> >
> > SRFI-18 states,
> >
> > The mutex primitives specified in this SRFI do not implement  
> > "recursive" mutex semantics; an attempt to lock a mutex that is  
> > locked implies that the current thread must wait even if the mutex  
> > is owned by the current thread
> >
> > so, I want a macro or procedure to support "recursive" mutex.
> > Would be glad someone can note on it.
> >
> > Here's my macro, not sure if it works:
> > Isn't there a similar code?
> >
> > (define my-mutex (make-mutex))
> > (define-macro (my-lock . body)
> >  `(let ([result #f])
> >     (if (eq? (mutex-state my-mutex) (current-thread))
> >         (set! result (begin ,@body))
> >         ;;else
> >         (begin
> >           (mutex-lock! my-mutex)
> >           (set! result (begin ,@body))
> >           (mutex-unlock! my-mutex)))
> >     result))
> >
> I have actually never used the Chicken threading capabilities, but  
> with almost
> any mutex code it's a safe bet that there's  race condition. In this  
> case, is
> it possible that the mutex could become locked between a false eq?  
> result and
> the call of mutex-lock! ?
> 
> I think recursive mutex locking almost always needs a conditional lock  
> operation.

Yes, that was my question, too.
I was writing with a conditional lock for this,
but wondering if it might be ok with this case
of eq?ing with (current-thread).

Any other comments?
--daishi