Speed up the tail duplication pass on the testcase below from 68.2s to 1.23s:

#define CL0(a) case a: f(); goto c;
 #define CL1(a) CL0(a##0) CL0(a##1) CL0(a##2) CL0(a##3) CL0(a##4) CL0(a##5) \
 CL0(a##6) CL0(a##7) CL0(a##8) CL0(a##9)
 #define CL2(a) CL1(a##0) CL1(a##1) CL1(a##2) CL1(a##3) CL1(a##4) CL1(a##5) \
 CL1(a##6) CL1(a##7) CL1(a##8) CL1(a##9)
 #define CL3(a) CL2(a##0) CL2(a##1) CL2(a##2) CL2(a##3) CL2(a##4) CL2(a##5) \
 CL2(a##6) CL2(a##7) CL2(a##8) CL2(a##9)
 #define CL4(a) CL3(a##0) CL3(a##1) CL3(a##2) CL3(a##3) CL3(a##4) CL3(a##5) \
 CL3(a##6) CL3(a##7) CL3(a##8) CL3(a##9)

 void f();

 void a() {
     int b;
  c: switch (b) {
         CL4(1)
     }
 }

This comes from GCC PR 15524

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@17390 91177308-0d34-0410-b5e6-96231b3b80d8

diff --git a/lib/Transforms/Scalar/TailDuplication.cpp b/lib/Transforms/Scalar/TailDuplication.cpp
index 85e1eac267e3b27012128a71d990fddd0bc2d417..545068b6dfb977e4718cab8fd1a356dc8c614744 100644 (file)
--- a/lib/Transforms/Scalar/TailDuplication.cpp
+++ b/lib/Transforms/Scalar/TailDuplication.cpp
@@ -114,9 +114,14 @@ bool TailDup::shouldEliminateUnconditionalBranch(TerminatorInst *TI) {
   // with a single successor if the block has many other predecessors.  This can
   // cause an N^2 explosion in CFG edges (and PHI node entries), as seen in
   // cases that have a large number of indirect gotos.
-  if (DTI->getNumSuccessors() > 8)
-    if (std::distance(PI, PE) * DTI->getNumSuccessors() > 128)
-      return false;
+  unsigned NumSuccs = DTI->getNumSuccessors();
+  if (NumSuccs > 8) {
+    unsigned TooMany = 128;
+    if (NumSuccs >= TooMany) return false;
+    TooMany = TooMany/NumSuccs;
+    for (; PI != PE; ++PI)
+      if (TooMany-- == 0) return false;
+  }
 
   return true;  
 }