<?xml version="1.0" encoding="UTF-8"?><!-- generator="wordpress/2.0.5" -->
<rss version="2.0" 
	xmlns:content="http://purl.org/rss/1.0/modules/content/">
<channel>
	<title>Comments on: Exploiting Flash Reliably</title>
	<link>http://taossa.com/index.php/2008/04/12/exploiting-flash-reliably/</link>
	<description>Continued ramblings on software security and code auditing</description>
	<pubDate>Fri, 30 Jul 2010 13:37:38 +0000</pubDate>
	<generator>http://wordpress.org/?v=2.0.5</generator>

	<item>
		<title>by: mark</title>
		<link>http://taossa.com/index.php/2008/04/12/exploiting-flash-reliably/#comment-22758</link>
		<pubDate>Mon, 21 Apr 2008 07:31:00 +0000</pubDate>
		<guid>http://taossa.com/index.php/2008/04/12/exploiting-flash-reliably/#comment-22758</guid>
					<description>@dmm: As the paper mentions, you need a ShowFrame tag.</description>
		<content:encoded><![CDATA[<p>@dmm: As the paper mentions, you need a ShowFrame tag.
</p>
]]></content:encoded>
				</item>
	<item>
		<title>by: dmm</title>
		<link>http://taossa.com/index.php/2008/04/12/exploiting-flash-reliably/#comment-22615</link>
		<pubDate>Fri, 18 Apr 2008 16:30:00 +0000</pubDate>
		<guid>http://taossa.com/index.php/2008/04/12/exploiting-flash-reliably/#comment-22615</guid>
					<description>Mark,

It looks if simply modify &#226;&#128;&#152;SceneCount&#226;&#128;&#153; will cause DoABC tag not executed ? What else should I considered when create such SWF file? 

Thanks.</description>
		<content:encoded><![CDATA[<p>Mark,</p>
<p>It looks if simply modify &#226;&#8364;&#732;SceneCount&#226;&#8364;&#8482; will cause DoABC tag not executed ? What else should I considered when create such SWF file? </p>
<p>Thanks.
</p>
]]></content:encoded>
				</item>
	<item>
		<title>by: mark</title>
		<link>http://taossa.com/index.php/2008/04/12/exploiting-flash-reliably/#comment-22530</link>
		<pubDate>Thu, 17 Apr 2008 12:57:00 +0000</pubDate>
		<guid>http://taossa.com/index.php/2008/04/12/exploiting-flash-reliably/#comment-22530</guid>
					<description>@Benjamin: Yes, if you do multiple overwrites, you overwrite some other data in each of the browsers.. but it turned out not to matter. (Also, your shellcode can return the corrupted address to their original state.)

@Nathan: Cool, thanks :). Yeah, you can email me if you have any questions..</description>
		<content:encoded><![CDATA[<p>@Benjamin: Yes, if you do multiple overwrites, you overwrite some other data in each of the browsers.. but it turned out not to matter. (Also, your shellcode can return the corrupted address to their original state.)</p>
<p>@Nathan: Cool, thanks :). Yeah, you can email me if you have any questions..
</p>
]]></content:encoded>
				</item>
	<item>
		<title>by: Nathan McFeters</title>
		<link>http://taossa.com/index.php/2008/04/12/exploiting-flash-reliably/#comment-22511</link>
		<pubDate>Thu, 17 Apr 2008 04:38:00 +0000</pubDate>
		<guid>http://taossa.com/index.php/2008/04/12/exploiting-flash-reliably/#comment-22511</guid>
					<description>Mark,

Covered this on the ZDNet Zero Day Security blog.  This vulnerability is absurdly cool.  Hell of a find, would like to chat with you about it sometime, as I wonder if the application of your exploit technique is limited to Flash, or if some of the concepts could be extended to other null pointer dereference issues.

-Nate</description>
		<content:encoded><![CDATA[<p>Mark,</p>
<p>Covered this on the ZDNet Zero Day Security blog.  This vulnerability is absurdly cool.  Hell of a find, would like to chat with you about it sometime, as I wonder if the application of your exploit technique is limited to Flash, or if some of the concepts could be extended to other null pointer dereference issues.</p>
<p>-Nate
</p>
]]></content:encoded>
				</item>
	<item>
		<title>by: Benjamin</title>
		<link>http://taossa.com/index.php/2008/04/12/exploiting-flash-reliably/#comment-22487</link>
		<pubDate>Wed, 16 Apr 2008 23:53:00 +0000</pubDate>
		<guid>http://taossa.com/index.php/2008/04/12/exploiting-flash-reliably/#comment-22487</guid>
					<description>Interesting read.  I&#8217;m not entirely clear how you handled the cross-browser case; you mentioned using the same technique to write two different words in memory.  If you write to an address that hits AS3_argmask in IE, aren&#8217;t you corrupting memory at some (possibly unrelated) location in FireFox?

I know Opera uses the NS plugin API, but I don&#8217;t know if that means that it loads the Flash plugin at the same address.  If it does then the exploit would work the same as for FireFox.</description>
		<content:encoded><![CDATA[<p>Interesting read.  I&#8217;m not entirely clear how you handled the cross-browser case; you mentioned using the same technique to write two different words in memory.  If you write to an address that hits AS3_argmask in IE, aren&#8217;t you corrupting memory at some (possibly unrelated) location in FireFox?</p>
<p>I know Opera uses the NS plugin API, but I don&#8217;t know if that means that it loads the Flash plugin at the same address.  If it does then the exploit would work the same as for FireFox.
</p>
]]></content:encoded>
				</item>
	<item>
		<title>by: Rhys</title>
		<link>http://taossa.com/index.php/2008/04/12/exploiting-flash-reliably/#comment-22480</link>
		<pubDate>Wed, 16 Apr 2008 14:46:00 +0000</pubDate>
		<guid>http://taossa.com/index.php/2008/04/12/exploiting-flash-reliably/#comment-22480</guid>
					<description>Whoa, Thomas Ptacek hit it on the head. Mark you really are a robot from the future sent to kill the mother of the person who will grow up to challenge SkyNet!

&lt;a href="http://www.matasano.com/log/1032/this-new-vulnerability-dowds-inhuman-flash-exploit/" rel="nofollow"&gt;http://www.matasano.com/log/1032/this-new-vulnerability-dowds-inhuman-flash-exploit/&lt;/a&gt;

Quite a few hurdles surpassed to get reliable remote code execution here, ultimately with a unique use of the ActionScript bytecode. I hope it&#8217;s a wake up call for the many prevalent script (Ruby, Python, ECMAScript etc) interpreters that absolved themselves of all security issues because of a view memory corruption only happens in C/C++ code.</description>
		<content:encoded><![CDATA[<p>Whoa, Thomas Ptacek hit it on the head. Mark you really are a robot from the future sent to kill the mother of the person who will grow up to challenge SkyNet!</p>
<p><a href="http://www.matasano.com/log/1032/this-new-vulnerability-dowds-inhuman-flash-exploit/" rel="nofollow" onclick="javascript:urchinTracker ('/outbound/www.matasano.com');">http://www.matasano.com/log/1032/this-new-vulnerability-dowds-inhuman-flash-exploit/</a></p>
<p>Quite a few hurdles surpassed to get reliable remote code execution here, ultimately with a unique use of the ActionScript bytecode. I hope it&#8217;s a wake up call for the many prevalent script (Ruby, Python, ECMAScript etc) interpreters that absolved themselves of all security issues because of a view memory corruption only happens in C/C++ code.
</p>
]]></content:encoded>
				</item>
	<item>
		<title>by: Juliano</title>
		<link>http://taossa.com/index.php/2008/04/12/exploiting-flash-reliably/#comment-22425</link>
		<pubDate>Tue, 15 Apr 2008 01:08:00 +0000</pubDate>
		<guid>http://taossa.com/index.php/2008/04/12/exploiting-flash-reliably/#comment-22425</guid>
					<description>Awesome! I really enjoyed  the paper. Congratulations.</description>
		<content:encoded><![CDATA[<p>Awesome! I really enjoyed  the paper. Congratulations.
</p>
]]></content:encoded>
				</item>
</channel>
</rss>

<!-- Dynamic Page Served (once) in 0.084 seconds -->
