[LLVMdev] Accessing Dominator Tree

[John Criswell]

On 7/29/11 10:00 AM, david.dewey at comcast.net wrote:
>
> John,
>
> Thanks for the quick reply.  Using a PassManager object should work 
> fine.  As you said, my only objection is to exec'ing opt.
>
> I had actually tried instantiating a PassManager object before, but I 
> was definitely not doing it right.  I will take a look at how clang 
> and SAFECode use it, and see if I can get it working.
>

The SAFECode sc tool is based on opt, so you can just see what opt 
does.  Clang does something weird with the PassManager object (I think), 
so I recommend looking at opt or sc over clang.

-- John T.

> Thanks,
>
> David
>
>
> ------------------------------------------------------------------------
>
> *From: *"John Criswell" <criswell at illinois.edu>
> *To: *"david dewey" <david.dewey at comcast.net>
> *Cc: *llvmdev at cs.uiuc.edu
> *Sent: *Friday, July 29, 2011 10:56:20 AM
> *Subject: *Re: [LLVMdev] Accessing Dominator Tree
>
> On 7/29/11 9:43 AM, david.dewey at comcast.net wrote:
>
>     Hello,
>
>     I am working on a project that requires me to perform some
>     analysis on a bitcode file outside the existing LLVM analysis
>     frameworks (opt, etc.).  In what I am doing, I need to be able to
>     access the dominator tree for a given function.  Is there a way I
>     can instantiate a DominatorTree object outside the existing
>     analysis frameworks?
>
>
> If you use a PassManager object to run passes, then you can simply 
> have your analysis be an LLVM pass and use the standard 
> getAnalysis<>() template function to get access to the dominator tree 
> analysis.  Any program can create a PassManager to run LLVM passes to 
> perform analysis and transformation; the clang C/C++ front-end and the 
> sc tool in SAFECode do this.
>
> I don't know if using a PassManager is what you consider to be 
> "outside the existing LLVM framework."  I'm hoping your only objection 
> is to using the opt tool to run LLVM passes.
>
> -- John T.
>
>     I have tried several things all along the lines of:
>
>     DominatorTree DT;
>
>     DT.getBase().recalculate(F);
>
>     Everything I have tried results in an assertion failure when
>     trying to access the front() of what I assume is the DominatorTree:
>
>     Assertion failed: !empty() && "Called front() on empty list!"
>
>     If anyone has an example of how to do this, I would really
>     appreciate it.
>
>     Thanks,
>
>     David
>
>
>
>     _______________________________________________
>     LLVM Developers mailing list
>     LLVMdev at cs.uiuc.edu          http://llvm.cs.uiuc.edu
>     http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev
>
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20110729/20445425/attachment.html>