Re: Should I use mutex in this context?
On Fri, 31 Oct 2008 03:11:34 -0400, Tommy <tommy767@gmail.com> wrote:
With all your side-stepping, you still can't fail the code
Oh good grief. I have done zero "side-stepping", and until you lose the
attitude, you're never going to learn anything. I talked about how to "fail
the code" in just about every message I've posted in this thread. Others
have, too. You're just not getting it. Let me reiterate that I couldn't
care less whether or not some random piece of code works when it uses
techniques that commonly do not work.
The point was proven - a synchronization kernel object is not required.
If you can read and understand my original post to the OP, I explicitly
stated that under some conditions, a volatile bool can work. Your flawed
"proof" was neither requested nor required. As I explained in my first post
in this thread, for what you posted to have a prayer of working in general,
the bool needs to be volatile. I told you that again in direct reply to
your flawed "proof".
You may have the last word, if that helps you. - I'm done.
You would be doing everyone a great favor to honor that statement. You need
to slow down and study what you've been told, not pontificate and judge,
which is mainly what you've taken to doing. Believe it or not, I mean that
kindly.
--
Doug Harrison
Visual C++ MVP