The issue is that when I call the PostDominatorTree pass and I try to get the root of the tree, I get the seg fault. When I compile llvm in release mode everything works fine and dandy. <div><br></div><div>I removed the inline attribute from the functions in PostDominators.h and recompiled llvm in debug mode. Same problem!</div>
<div>The problems has to do with the PostDominators.h or PostDominatorTree. I am not even sure any longer.</div><div><br></div><div>--George<br><br><div class="gmail_quote">On Tue, Aug 21, 2012 at 10:57 AM, Marco Minutoli <span dir="ltr"><<a href="mailto:mminutoli@gmail.com" target="_blank">mminutoli@gmail.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div class="HOEnZb"><div class="h5">George Baah <<a href="mailto:georgebaah@gmail.com">georgebaah@gmail.com</a>> writes:<br>
<br>
> Hi Guys,<br>
><br>
> I saw in the API that DominanceFrontier has been deprecated. Is this<br>
> for real?<br>
> I have read the computational issues but shouldn't that be left to the<br>
> users of llvm?<br>
> My concern is I am writing a Control Dependence module that uses<br>
> Dominance-Frontier information.<br>
> Getting rid of Dominance Frontiers will put me in a tight spot.<br>
><br>
> George<br>
<br>
</div></div>Hi,<br>
<br>
I had the same issue implementing a pass that build PDGs for an exam two<br>
months ago.<br>
<br>
As an alternative to the (post)-dominance frontier method I used the<br>
algorithm described in [1], section 9.5 "Program Dependence Graph" where<br>
you can find a method to build the CDG using only the post-dominator<br>
tree.<br>
<br>
Happy hacking,<br>
Marco<br>
<br>
[1] Advanced Compiler Design & Implementation by Steven S. Muchnick<br>
</blockquote></div><br></div>