[gclist] Re: MP synchonization without atomic instructions

Henry Baker hbaker@netcom.com
Tue, 28 Jan 1997 23:54:22 -0800

hbaker:  good stuff in anderson's web site listed below.

In article <5ckvvn$ehe@anderson.cs.unc.edu>, anderson@cs.unc.edu (James H.
Anderson) wrote:

> In article <aegl.854143257@stratus>, Tony Luck <aegl@swdc.stratus.com> wrote:
> >I know about Dekker's algorithm, and read the excellent description of
> >algorithm in Tannenbaum's "Operating Systems: Design and Implementation."
> >
> >Can anyone give me some references to other algorithms to implement
> >primitives on machines with no atomic operations.
> >
> >-Tony Luck <aegl@swdc.stratus.com>
> You might be interested in:
> J.-H. Yang and J. Anderson, "A Fast, Scalable Mutual Exclusion Algorithm",
> Distributed Computing, Vol. 9, No. 1, pp. 51-60, August 1995.
> Also at: http://www.cs.unc.edu/~anderson/papers.html
> (Please excuse the long proof.  It was forced on us by a referee.  Despite
> the length of the proof, the algorithm really isn't too hard to understand.)
> This paper presents a new algorithm, and also includes references to more
> recent work.  My web page also includes a number of papers on synchronization
> without mutual exclusion (i.e., lock-free and wait-free sycnhronization).  I
> believe someone else mentioned this topic in response to your post.
> Hope this helps.
> -- 
> ----------**********----------**********----------**********----------
> Jim Anderson    UNC Chapel Hill    anderson@cs.unc.edu    919 962-1757