Subj : Re: STL Containers w/ Virtually Zero-Overhead Lock-Free Reads...
To   : (Usenet)
From : Chris Thomasson
Date : Thu Sep 22 2005 06:16 am

> I could be missing something important, but it seems to me that if you
> want to "actually apply some algorithms to it" then a lock-free
> container is of little use, since you get synchronization at the wrong
> level of granularity.  You'd need algorithm-level synchronization,
> rather than container-level synchronization, in that case.  In other
> words, the SGI argument against putting mutexes in the containers
> applies just as well to the question of whether lock-free container
> operations are a good idea.

Per-container mutexs? That would be a nightmare! Think of an applications 
that happens to create extremely high numbers of collection objects...

;)


You could use groups of global "static" arrays of mutexs. Hash the this 
pointer of a collection into an index and do your business. There is a 
catch, you can deadlock real fast, like this:

http://groups.google.com/group/comp.programming.threads/msg/56406d367ac85dcb?hl=en


A scaleable solution involves keeping order/track of the array of mutexs an 
application thread has locked on a per-thread basis... 



      [ See http://www.gotw.ca/resources/clcm.htm for info about ]
      [ comp.lang.c++.moderated.    First time posters: Do this! ]

.