[llvm] de34b54 - [SCEV] Swap guards estimation sequence. NFC

[via llvm-commits]

Author: dfukalov
Date: 2020-01-20T16:41:16+03:00
New Revision: de34b54edce4b7b4e4a68a02fae10283b3e2d7ea

URL: https://github.com/llvm/llvm-project/commit/de34b54edce4b7b4e4a68a02fae10283b3e2d7ea
DIFF: https://github.com/llvm/llvm-project/commit/de34b54edce4b7b4e4a68a02fae10283b3e2d7ea.diff

LOG: [SCEV] Swap guards estimation sequence. NFC

Summary:
Loop unroll spends a lot of time in SCEVs processing in case when a function
contains hundreds of simple 'for' loops with a quite complex arrays indexes like

  for (int i = 0; i < 8; ++i) {
    for (int j = 0; j < 32; ++j) {
      C[j*8+i] = B[j*32+i+128] + A[i*64+128];
    }
  }
  for (int i = 0; i < 8; ++i) {
    for (int j = 0; j < 8; ++j) {
      for (int k = 0; k < 32; ++k) {
        D[k*64+i*8+j] = D[k*64+i*8+j] + E[i+16] * C[k*8+j+256];
      }
    }
  }

The patch improves loop unroll speed since isLoopBackedgeGuardedByCond takes
much less time than isLoopEntryGuardedByCond in the edge case.

Reviewers: skatkov, sanjoy, mkazantsev

Reviewed By: sanjoy

Subscribers: fhahn, hiraditya, javed.absar, llvm-commits

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D72929

Added: 
    

Modified: 
    llvm/lib/Analysis/ScalarEvolution.cpp

Removed: 
    


################################################################################
diff  --git a/llvm/lib/Analysis/ScalarEvolution.cpp b/llvm/lib/Analysis/ScalarEvolution.cpp
index 89dcc4793d72..36a9e7d09882 100644
--- a/llvm/lib/Analysis/ScalarEvolution.cpp
+++ b/llvm/lib/Analysis/ScalarEvolution.cpp
@@ -9222,9 +9222,11 @@ bool ScalarEvolution::isKnownViaInduction(ICmpInst::Predicate Pred,
       !isAvailableAtLoopEntry(SplitRHS.first, MDL))
     return false;
 
-  return isLoopEntryGuardedByCond(MDL, Pred, SplitLHS.first, SplitRHS.first) &&
-         isLoopBackedgeGuardedByCond(MDL, Pred, SplitLHS.second,
-                                     SplitRHS.second);
+  // It seems backedge guard check is faster than entry one so in some cases
+  // it can speed up whole estimation by short circuit
+  return isLoopBackedgeGuardedByCond(MDL, Pred, SplitLHS.second,
+                                     SplitRHS.second) &&
+         isLoopEntryGuardedByCond(MDL, Pred, SplitLHS.first, SplitRHS.first);
 }
 
 bool ScalarEvolution::isKnownPredicate(ICmpInst::Predicate Pred,