Hi Crista:

again on page 158.

I was assuming that the DJ and Java program had the same behavior regarding
notifications.

If we would like this, could we change in the method manager for put:

empty = false;

to

if (usedSlots==1) empty=false;

Then the COOL code should only notify if a condition variable has actually
changed. Does your implementation do this?
Josh, does your implementation do it.

-- Karl


================
From lopes@parc.xerox.com Wed Oct 29 19:00:05 1997
Sender: Cristina Lopes <lopes@parc.xerox.com>
From: Cristina Lopes <lopes@parc.xerox.com>
To: Karl Lieberherr <lieber@ccs.neu.edu>
Subject: Re: typos


Karl Lieberherr writes:
> on page 158: in Java: you check whether the buffer is empty after a put
> and whether it is full after a take. Does not sound right.

Karl,

When I went back to chapter 5 to correct the bug, I realized that it
is not a bug - it's correct.

When you put an element, and just before you return, you check if the
buffer _was_ empty. (Note that the test takes precedence over the
increment) If the buffer _was_ empty, there may be threads suspended
waiting to take an element, and that's why you notify. If the buffer
_was_not_ empty, then no threads could possibly be suspended.
Same for take().

-Crista


============

Hi Crista:

I see; that makes sense.

But then you should write, I think,

if (usedSlots++ == 0)

(NOT if (++usedSlots == 0))

so that for the test the old value of usedSlots is used, or better:

if (usedSlots == 0) notifyAll();
usedSlots++;

-- Karl


From lopes@parc.xerox.com Thu Oct 30 21:03:30 1997
From: Cristina Lopes <lopes@parc.xerox.com>
To: Karl Lieberherr <lieber@ccs.neu.edu>
Subject: Re: minimize notifies


Karl Lieberherr writes:
> Hi Crista:
> 
> again on page 158.
> 
> I was assuming that the DJ and Java program had the same behavior regarding
> notifications.

They have the same behavior, but not necessarily the same implementation.


> If we would like this, could we change in the method manager for put:
> 
> empty = false;
> 
> to
> 
> if (usedSlots==1) empty=false;
> 
> Then the COOL code should only notify if a condition variable has actually
> changed. Does your implementation do this?

Yes - ideally notifications should be specific to condition variables.
But that's a bit more complex to implement - although not impossible.