[PATCH] [LAA] Fix estimation of number of memchecks

[Adam Nemet]

With this change I don't think there is a reason to cache NumComparisons anymore.  What do you think?

Specifically I mean:


================
Comment at: include/llvm/Analysis/LoopAccessAnalysis.h:390-392
@@ -385,5 +389,5 @@
 
   /// \brief Number of memchecks required to prove independence of otherwise
   /// may-alias pointers.
   unsigned getNumRuntimePointerChecks() const { return NumComparisons; }
 
----------------
I think that this guy should now forward to getNumberOfChecks() and probably take a partition (by default nullptr) parameter.

================
Comment at: include/llvm/Analysis/LoopAccessAnalysis.h:467-469
@@ -462,5 +466,5 @@
 
   /// \brief Number of memchecks required to prove independence of otherwise
   /// may-alias pointers
   unsigned NumComparisons;
 
----------------
To avoid confusion we should probably remove this.

================
Comment at: lib/Analysis/LoopAccessAnalysis.cpp:228-234
@@ -221,8 +227,9 @@
 
   /// \brief Check whether we can check the pointers at runtime for
-  /// non-intersection.
+  /// non-intersection. Returns true when we have 0 pointers
+  /// (a check on 0 pointers for non-intersection will always return true).
   bool canCheckPtrAtRT(LoopAccessInfo::RuntimePointerCheck &RtCheck,
                        unsigned &NumComparisons, ScalarEvolution *SE,
                        Loop *TheLoop, const ValueToValueMap &Strides,
                        bool ShouldCheckStride = false);
 
----------------
I don't think there is a reason for this function to return NumComparisons now.  It should probably return NeedRTCheck instead (used in the caller).

http://reviews.llvm.org/D10217

EMAIL PREFERENCES
  http://reviews.llvm.org/settings/panel/emailpreferences/