[llvm-commits] [PATCH] More constant-folding for PPCDoubleDouble

[Hal Finkel]

----- Original Message -----
> From: "Ulrich Weigand" <Ulrich.Weigand at de.ibm.com>
> To: llvm-commits at cs.uiuc.edu
> Sent: Monday, October 29, 2012 4:32:04 PM
> Subject: [llvm-commits] [PATCH] More constant-folding for PPCDoubleDouble
> 
> 
> Hello,
> 
> I noticed that the PPDoubleDouble patch set I just committed missed a
> couple of places where we now can do constant-folding on
> PPCDoubleDouble
> types.  The attached patch below enables this (and also updates a
> test case
> that specifically verified that constant folding does *not* take
> place ...).
> 
> This fixes the Clang :: CodeGen/complex-builtints.c test case on
> PowerPC.

Great!

> 
> OK to commit?

-      APFloat apf(APInt::getNullValue(DestTy->getPrimitiveSizeInBits()), true);
+      APFloat apf(APInt::getNullValue(DestTy->getPrimitiveSizeInBits()),
+                  !DestTy->isPPC_FP128Ty());

I recommend putting a comment like /* isIEEE */ next to the !DestTy->isPPC_FP128Ty()
just to remind people (like me) what's going on.

LGTM.

 -Hal

> 
> Bye,
> Ulrich
> 
> (See attached file: diff-llvm-ppcdoubledouble-enable2)
> _______________________________________________
> llvm-commits mailing list
> llvm-commits at cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits
> 

-- 
Hal Finkel
Postdoctoral Appointee
Leadership Computing Facility
Argonne National Laboratory