On Oct 6, 4:36 pm, Joshua Maurice<joshuamaur...@gmail.com> wrote:
I don't know how such myths survive
You don't? See the OP.
despite the community's best efforts to kill them.
Which community do you have in mind?
I believe comp.programming.threads community has done its part by
educating the author about volatile in C and C++.
Java community came up with the DCLP paper, which the page you
provide the link to lists in References.
The C++ community has mostly neglected the problem.
I do understand, that in 2001 multithreaded programming was not
taken seriously by most anybody in C++ community, or else how could
one announce, that their multithreaded code suffers from deadlocks
and still find it appropriate to dispense advice on multithreaded
programming.
Well, it is 2010, and Modern C++ Desing is at its 17-th printing.
I tried to contact the author by email, except most of the links on
his web page are broken, in particular "Input new erratum" is broken,
and the link to his e-mail address is broken.
Moreover, according to author's web page, C++ is not his main research
interest, and the author is "extremely busy with his research", which
I
understand not to be C++ related.
So, in short, it does not look likely, that this thing is going to be
fixed any time soon, and hence my question.
Is it ever going to be fixed? Anyone knows?
At this point fixing the 18th printing book will not fix the copies already sold
and will not effectively dispel possible misconceptions. You may want to point
article http://www.aristeia.com/Papers/DDJ_Jul_Aug_2004_revised.pdf which (a) is
coauthored by Modern C++ Design's author, (b) is more recent than the book, and
P.S. Sorry, I didn't receive email from you.
[ comp.lang.c++.moderated. First time posters: Do this! ]