Changes in directory llvm/lib/Target:

TargetMachine.cpp updated: 1.44 -> 1.45
---
Log message:

Added option -enable-finite-only-fp-math. When on, the codegen can assume that
FP arithmetic arguments and results are never NaNs or +=Infs. This includes
ignoring parity flag (PF) when checking for FP equality.


---
Diffs of the changes:  (+6 -0)

 TargetMachine.cpp |    6 ++++++
 1 files changed, 6 insertions(+)


Index: llvm/lib/Target/TargetMachine.cpp
diff -u llvm/lib/Target/TargetMachine.cpp:1.44 
llvm/lib/Target/TargetMachine.cpp:1.45
--- llvm/lib/Target/TargetMachine.cpp:1.44      Tue May  2 20:29:56 2006
+++ llvm/lib/Target/TargetMachine.cpp   Tue May 23 01:39:12 2006
@@ -26,6 +26,7 @@
   bool NoFramePointerElim;
   bool NoExcessFPPrecision;
   bool UnsafeFPMath;
+  bool FiniteOnlyFPMath;
   Reloc::Model RelocationModel;
 };
 namespace {
@@ -48,6 +49,11 @@
                cl::desc("Enable optimizations that may decrease FP precision"),
                cl::location(UnsafeFPMath),
                cl::init(false));
+  cl::opt<bool, true>
+  EnableFiniteOnltFPMath("enable-finite-only-fp-math",
+               cl::desc("Enable optimizations that assumes non- NaNs / 
+-Infs"),
+               cl::location(FiniteOnlyFPMath),
+               cl::init(false));
   cl::opt<llvm::Reloc::Model, true>
   DefRelocationModel(
     "relocation-model",



_______________________________________________
llvm-commits mailing list
llvm-commits@cs.uiuc.edu
http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits

Reply via email to