[LLVMdev] Question

Patrick Meredith pmeredit at uiuc.edu
Wed Mar 9 14:23:37 PST 2005


If you want to be particularly lazy you can just type M.dump() at the very end of your pass too.  Then you don't even had to dissamble any bytecode files. (Assuming M is the name of your Module, and you are doing a ModulePass, otherwise F.dump() for a FunctionPass with F as the name of the function).
  ----- Original Message ----- 
  From: Tanu Sharma 
  To: LLVM Developers Mailing List 
  Sent: Wednesday, March 09, 2005 2:18 PM
  Subject: Re: [LLVMdev] Question


  This is exactly what i m trying :

  opt -load /home/tsharma/ankur/llvm/Debug/lib/LLVMHello.so -hello <helloprog.bc> /dev/null

  Tanu


  Chris Lattner <sabre at nondot.org> wrote:
    On Wed, 9 Mar 2005, Tanu Sharma wrote:

    > I wrote a pass which randomizes basic blocks and insert new block.But 
    > when i run another pass over it which simply lists all basic blocks I 
    > don't get to see the changes.The bytecode file is not changed.Why is 
    > that ?

    Maybe your random number generator sucks? :-)

    Can you tell me exactly what options your using to run your pass? I 
    assume you're running it with 'opt' like this?

    opt in.bc -randomize | llvm-dis

    ?

    -Chris

    -- 
    http://nondot.org/sabre/
    http://llvm.cs.uiuc.edu/

    _______________________________________________
    LLVM Developers mailing list
    LLVMdev at cs.uiuc.edu http://llvm.cs.uiuc.edu
    http://mail.cs.uiuc.edu/mailman/listinfo/llvmdev



------------------------------------------------------------------------------
  Celebrate Yahoo!'s 10th Birthday! 
  Yahoo! Netrospective: 100 Moments of the Web 


------------------------------------------------------------------------------


  _______________________________________________
  LLVM Developers mailing list
  LLVMdev at cs.uiuc.edu         http://llvm.cs.uiuc.edu
  http://mail.cs.uiuc.edu/mailman/listinfo/llvmdev
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20050309/fca41c4b/attachment.html>


More information about the llvm-dev mailing list