[llvm-commits] [llvm] r47569 - /llvm/trunk/lib/Target/X86/X86ISelDAGToDAG.cpp

[Chris Lattner]

On Feb 25, 2008, at 2:15 PM, Dan Gohman wrote:

> Author: djg
> Date: Mon Feb 25 16:15:55 2008
> New Revision: 47569
>
> URL: http://llvm.org/viewvc/llvm-project?rev=47569&view=rev
> Log:
> Add an assert to verify that we don't see an
> {S,U}MUL_LOHI with an unused high value.

Hey Dan,

Is this a correctness or a performance thing?  If we chose to turn of  
the dag combiner at -O0, could this assertion get triggered?

-Chris

>
>
> Modified:
>    llvm/trunk/lib/Target/X86/X86ISelDAGToDAG.cpp
>
> Modified: llvm/trunk/lib/Target/X86/X86ISelDAGToDAG.cpp
> URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Target/X86/X86ISelDAGToDAG.cpp?rev=47569&r1=47568&r2=47569&view=diff
>
> = 
> = 
> = 
> = 
> = 
> = 
> = 
> = 
> ======================================================================
> --- llvm/trunk/lib/Target/X86/X86ISelDAGToDAG.cpp (original)
> +++ llvm/trunk/lib/Target/X86/X86ISelDAGToDAG.cpp Mon Feb 25  
> 16:15:55 2008
> @@ -1221,6 +1221,9 @@
>
>     case ISD::SMUL_LOHI:
>     case ISD::UMUL_LOHI: {
> +      assert(!N.getValue(1).use_empty() &&
> +             "A MUL_LOHI with an unused high result should fold to  
> a MUL!");
> +
>       SDOperand N0 = Node->getOperand(0);
>       SDOperand N1 = Node->getOperand(1);
>
>
>
> _______________________________________________
> llvm-commits mailing list
> llvm-commits at cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits