[llvm-commits] [llvm] r82923 - /llvm/trunk/lib/Transforms/Scalar/InstructionCombining.cpp

[Chris Lattner]

Author: lattner
Date: Sun Sep 27 15:46:36 2009
New Revision: 82923

URL: http://llvm.org/viewvc/llvm-project?rev=82923&view=rev
Log:
allow pushing icmps through phis with multiple uses and across critical edges.
These are important to push up to encourage jump threading.  This shrinks 176.gcc a bit.

Modified:
    llvm/trunk/lib/Transforms/Scalar/InstructionCombining.cpp

Modified: llvm/trunk/lib/Transforms/Scalar/InstructionCombining.cpp
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Transforms/Scalar/InstructionCombining.cpp?rev=82923&r1=82922&r2=82923&view=diff

==============================================================================
--- llvm/trunk/lib/Transforms/Scalar/InstructionCombining.cpp (original)
+++ llvm/trunk/lib/Transforms/Scalar/InstructionCombining.cpp Sun Sep 27 15:46:36 2009
@@ -388,7 +388,11 @@
     // which has a PHI node as operand #0, see if we can fold the instruction
     // into the PHI (which is only possible if all operands to the PHI are
     // constants).
-    Instruction *FoldOpIntoPhi(Instruction &I);
+    //
+    // If AllowAggressive is true, FoldOpIntoPhi will allow certain transforms
+    // that would normally be unprofitable because they strongly encourage jump
+    // threading.
+    Instruction *FoldOpIntoPhi(Instruction &I, bool AllowAggressive = false);
 
     // FoldPHIArgOpIntoPHI - If all operands to a PHI node are the same "unary"
     // operator and they all are only used by the PHI, PHI together their
@@ -1942,11 +1946,22 @@
 /// FoldOpIntoPhi - Given a binary operator, cast instruction, or select which
 /// has a PHI node as operand #0, see if we can fold the instruction into the
 /// PHI (which is only possible if all operands to the PHI are constants).
-Instruction *InstCombiner::FoldOpIntoPhi(Instruction &I) {
+///
+/// If AllowAggressive is true, FoldOpIntoPhi will allow certain transforms
+/// that would normally be unprofitable because they strongly encourage jump
+/// threading.
+Instruction *InstCombiner::FoldOpIntoPhi(Instruction &I,
+                                         bool AllowAggressive) {
+  AllowAggressive = false;
   PHINode *PN = cast<PHINode>(I.getOperand(0));
   unsigned NumPHIValues = PN->getNumIncomingValues();
-  if (!PN->hasOneUse() || NumPHIValues == 0) return 0;
-
+  if (NumPHIValues == 0 ||
+      // We normally only transform phis with a single use, unless we're trying
+      // hard to make jump threading happen.
+      (!PN->hasOneUse() && !AllowAggressive))
+    return 0;
+  
+  
   // Check to see if all of the operands of the PHI are simple constants
   // (constantint/constantfp/undef).  If there is one non-constant value,
   // remember the BB it is in.  If there is more than one or if *it* is a PHI,
@@ -1970,7 +1985,7 @@
   // operation in that block.  However, if this is a critical edge, we would be
   // inserting the computation one some other paths (e.g. inside a loop).  Only
   // do this if the pred block is unconditionally branching into the phi block.
-  if (NonConstBB) {
+  if (NonConstBB != 0 && !AllowAggressive) {
     BranchInst *BI = dyn_cast<BranchInst>(NonConstBB->getTerminator());
     if (!BI || !BI->isUnconditional()) return 0;
   }
@@ -5865,7 +5880,7 @@
         // block.  If in the same block, we're encouraging jump threading.  If
         // not, we are just pessimizing the code by making an i1 phi.
         if (LHSI->getParent() == I.getParent())
-          if (Instruction *NV = FoldOpIntoPhi(I))
+          if (Instruction *NV = FoldOpIntoPhi(I, true))
             return NV;
         break;
       case Instruction::SIToFP:
@@ -6221,11 +6236,11 @@
         break;
 
       case Instruction::PHI:
-        // Only fold icmp into the PHI if the phi and fcmp are in the same
+        // Only fold icmp into the PHI if the phi and icmp are in the same
         // block.  If in the same block, we're encouraging jump threading.  If
         // not, we are just pessimizing the code by making an i1 phi.
         if (LHSI->getParent() == I.getParent())
-          if (Instruction *NV = FoldOpIntoPhi(I))
+          if (Instruction *NV = FoldOpIntoPhi(I, true))
             return NV;
         break;
       case Instruction::Select: {