Google uses special crawlers and index for mobile websites (this index is used at
www.google.com/m/). So, there is two possible situations:
1. You don't use subdomains at all. In this case MJ will correctly determine type of the crawler and show corresponding content to it. As a result, there is no duplicating content per crawler.
2. You use subdomain(s) for different site version(s). In this case mobile crawlers will be correctly redirected to the subdomain, but there is a chance that general crawler will start to index subdomain too. Google is able to determine type of website on the basis of http headers and DTD markup type, but to prevent indexing for sure, I'd recommend to use the following substitution in .htaccess:
RewriteCond %{HTTP_HOST} ^m\.yoursitename\.com$ [NC]
RewriteRule ^robots\.txt$ m.robots.txt [L]
with corresponding m.robots.txt:
User-agent: Googlebot/
Disallow: /
It will prevent general google bot (with "Googlebot/" in User-agent string) from indexing the subdomain, but allows to do it by mobile crawler (with "Googlebot-Mobile/" in UA).
Also I'd recommend to replace in modules/mod_mj_markupchooser/tmpl/default.php
$parts[] = '<a class="markupchooser" href="'.$link['url'].'">'.$link['text'].'</a>';
by
$parts[] = '<a rel="nofollow" class="markupchooser" href="'.$link['url'].'">'.$link['text'].'</a>';
(but anyway google isn't able to visit this links correctly because it doesn't support cookies, and anyway indexing of subdomain is disabled in m.robots.txt).