Re: [Corelinux-develop] [Fwd: [Bug #115535] Deadlock on example22]

[Frank V. C. <fr...@co...>]

Hans - Dulimarta wrote:
> 
> On Thu, 28 Sep 2000, Frank V. Castellucci wrote:
> 
> > Date: Thu, 28 Sep 2000 07:17:58 -0400
> > From: Frank V. Castellucci <fr...@co...>
> > Reply-To: cor...@li...
> > To: CoreLinux Development <cor...@li...>
> > Subject: [Corelinux-develop] [Fwd: [Bug #115535] Deadlock on example22]
> >
> > Hans,
> >
> > Was this like a note to yourself?
> 
> Well, kind of.
> 
> > I assume you changed the behavior of examp22 to test the state change
> > issue we discussed previously. No?
> >
> 
> I guess the two bugs are different. Maybe I did not write the bug text
> specific enough. In bug 115535 I wanted to see how the code behaves when
> the each of the two "listeners" enters a loop but the "trigger" does not
> and tries to destroy the semaphore. Apparently in this case the trigger
> cannot run until completion. I think the bug is the destroySemaphore
> method. When it sees that the "share count" is non-zero it executes
> nothing.
> 
> In bug 113589 both the listeners and the trigger will run in a loop.
> 
> The two bugs may be related. I am not sure now.

Idle musing here but, maybe, they are the same. To me the situation is
one where the behavior of the client may be:

1. Single shot - wait for one event and then disappear.
2. Continuous - wait for event, do some processing, come back for next
event
3. Non-blocking variations - interested to see if event has fired but
doesn't want to block

Where the "owner" wants to post, insure that all listeners for that
event "post instance" have been released, and then get the semaphore
back into the (1) state.

> 
> >
> >
> 
> --
> Hans Dulimarta, Ph.D.                  dul...@co...
> P: 517-432-7589             http://www.egr.msu.edu/~dulimart
> F: 760-281-7691             http://corelinux.sourceforge.net
> Elec. & Comp. Engg., Mich. State Univ., E. Lansing, MI 48824
> 
>   ------------------------------------------------------------------------
> 
> Subject: [Bug #115535] Deadlock on example22
> Date: Wed, 27 Sep 2000 22:23:36 -0700
> From: no...@so...
> To: dul...@co..., no...@so...,
>      fr...@us...
> 
> Bug #115535, was updated on 2000-Sep-27 22:23
> Here is a current snapshot of the bug.
> 
> Project: CoreLinux++
> Category: libcorelinux++
> Status: Open
> Resolution: None
> Bug Group: Defects
> Priority: 5
> Summary: Deadlock on example22
> 
> Details: Example22 was changed to have a loop on the listener task.
> As a result a deadlock occured.
> 
> For detailed info, follow this link:
> http://sourceforge.net/bugs/?func=detailbug&bug_id=115535&group_id=308

-- 
Frank V. Castellucci
http://corelinux.sourceforge.net
OOA/OOD/C++ Standards and Guidelines for Linux
http://PythPat.sourceforge.net
Pythons Pattern Package