[llvm] r177432 - IndVarSimplify: do not recompute an IV value outside of the loop if :

[Arnaud A. de Grandmaison]

Author: aadg
Date: Tue Mar 19 15:00:22 2013
New Revision: 177432

URL: http://llvm.org/viewvc/llvm-project?rev=177432&view=rev
Log:
IndVarSimplify: do not recompute an IV value outside of the loop if :

- it is trivially known to be used inside the loop in a way that can not be optimized away
- there is no use outside of the loop which can take advantage of the computation hoisting

Added:
    llvm/trunk/test/Transforms/IndVarSimplify/dont-recompute.ll
Modified:
    llvm/trunk/lib/Transforms/Scalar/IndVarSimplify.cpp

Modified: llvm/trunk/lib/Transforms/Scalar/IndVarSimplify.cpp
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Transforms/Scalar/IndVarSimplify.cpp?rev=177432&r1=177431&r2=177432&view=diff
==============================================================================
--- llvm/trunk/lib/Transforms/Scalar/IndVarSimplify.cpp (original)
+++ llvm/trunk/lib/Transforms/Scalar/IndVarSimplify.cpp Tue Mar 19 15:00:22 2013
@@ -535,6 +535,45 @@ void IndVarSimplify::RewriteLoopExitValu
         if (!SE->isLoopInvariant(ExitValue, L))
           continue;
 
+        // Computing the value outside of the loop brings no benefit if :
+        //  - it is definitely used inside the loop in a way which can not be
+        //    optimized away.
+        //  - no use outside of the loop can take advantage of hoisting the
+        //    computation out of the loop
+        if (ExitValue->getSCEVType()>=scMulExpr) {
+          unsigned NumHardInternalUses = 0;
+          unsigned NumSoftExternalUses = 0;
+          unsigned NumUses = 0;
+          for (Value::use_iterator IB=Inst->use_begin(), IE=Inst->use_end();
+               IB!=IE && NumUses<=6 ; ++IB) {
+            Instruction *UseInstr = cast<Instruction>(*IB);
+            unsigned Opc = UseInstr->getOpcode();
+            NumUses++;
+            if (L->contains(UseInstr)) {
+              if (Opc == Instruction::Call || Opc == Instruction::Ret)
+                NumHardInternalUses++;
+            } else {
+              if (Opc == Instruction::PHI) {
+                // Do not count the Phi as a use. LCSSA may have inserted
+                // plenty of trivial ones.
+                NumUses--;
+                for (Value::use_iterator PB=UseInstr->use_begin(),
+                                         PE=UseInstr->use_end();
+                     PB!=PE && NumUses<=6 ; ++PB, ++NumUses) {
+                  unsigned PhiOpc = cast<Instruction>(*PB)->getOpcode();
+                  if (PhiOpc != Instruction::Call && PhiOpc != Instruction::Ret)
+                    NumSoftExternalUses++;
+                }
+                continue;
+              }
+              if (Opc != Instruction::Call && Opc != Instruction::Ret)
+                NumSoftExternalUses++;
+            }
+          }
+          if (NumUses <= 6 && NumHardInternalUses && !NumSoftExternalUses)
+            continue;
+        }
+
         Value *ExitVal = Rewriter.expandCodeFor(ExitValue, PN->getType(), Inst);
 
         DEBUG(dbgs() << "INDVARS: RLEV: AfterLoopVal = " << *ExitVal << '\n'

Added: llvm/trunk/test/Transforms/IndVarSimplify/dont-recompute.ll
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/test/Transforms/IndVarSimplify/dont-recompute.ll?rev=177432&view=auto
==============================================================================
--- llvm/trunk/test/Transforms/IndVarSimplify/dont-recompute.ll (added)
+++ llvm/trunk/test/Transforms/IndVarSimplify/dont-recompute.ll Tue Mar 19 15:00:22 2013
@@ -0,0 +1,69 @@
+; RUN: opt < %s -indvars -S | FileCheck %s
+
+; This tests that the IV is not recomputed outside of the loop when it is known
+; to be computed by the loop and used in the loop any way. In the example below
+; although a's value can be computed outside of the loop, there is no benefit
+; in doing so as it has to be computed by the loop anyway.
+;
+; extern void func(unsigned val);
+;
+; void test(unsigned m)
+; {
+;   unsigned a = 0;
+;
+;   for (int i=0; i<186; i++) {
+;     a += m;
+;     func(a);
+;   }
+;
+;   func(a);
+; }
+
+declare void @func(i32)
+
+; CHECK: @test
+define void @test(i32 %m) nounwind uwtable {
+entry:
+  br label %for.body
+
+for.body:                                         ; preds = %for.body, %entry
+  %i.06 = phi i32 [ 0, %entry ], [ %inc, %for.body ]
+  %a.05 = phi i32 [ 0, %entry ], [ %add, %for.body ]
+  %add = add i32 %a.05, %m
+; CHECK: tail call void @func(i32 %add)
+  tail call void @func(i32 %add)
+  %inc = add nsw i32 %i.06, 1
+  %exitcond = icmp eq i32 %inc, 186
+  br i1 %exitcond, label %for.end, label %for.body
+
+for.end:                                          ; preds = %for.body
+; CHECK: for.end:
+; CHECK-NOT: mul i32 %m, 186
+; CHECK:%add.lcssa = phi i32 [ %add, %for.body ]
+; CHECK-NEXT: tail call void @func(i32 %add.lcssa)
+  tail call void @func(i32 %add)
+  ret void
+}
+
+; CHECK: @test2
+define i32 @test2(i32 %m) nounwind uwtable {
+entry:
+  br label %for.body
+
+for.body:                                         ; preds = %for.body, %entry
+  %i.06 = phi i32 [ 0, %entry ], [ %inc, %for.body ]
+  %a.05 = phi i32 [ 0, %entry ], [ %add, %for.body ]
+  %add = add i32 %a.05, %m
+; CHECK: tail call void @func(i32 %add)
+  tail call void @func(i32 %add)
+  %inc = add nsw i32 %i.06, 1
+  %exitcond = icmp eq i32 %inc, 186
+  br i1 %exitcond, label %for.end, label %for.body
+
+for.end:                                          ; preds = %for.body
+; CHECK: for.end:
+; CHECK-NOT: mul i32 %m, 186
+; CHECK:%add.lcssa = phi i32 [ %add, %for.body ]
+; CHECK-NEXT: ret i32 %add.lcssa
+  ret i32 %add
+}