[LV] Statistics numbers for LoopVectorize introduced: a number of analyzed loops & a number of vectorized loops.

[zinovy.nis]

Summary

Use -stats to see how many loops were analyzed for possible vectorization and how many of them were actually vectorized.

This patch is a splitted version of http://reviews.llvm.org/D3423.

[nadav]

LGTM!

[zinovy.nis]

Can you please commit it for me?

[amusman]

Committed, revision 206956.

[zinovy.nis]

Thanks, Alexander!

lib/Transforms/Vectorize/LoopVectorize.cpp

 #include "llvm/ADT/DenseMap.h"
 #include "llvm/ADT/EquivalenceClasses.h"
 #include "llvm/ADT/Hashing.h"
 #include "llvm/ADT/MapVector.h"
 #include "llvm/ADT/SetVector.h"
 #include "llvm/ADT/SmallPtrSet.h"
 #include "llvm/ADT/SmallSet.h"
 #include "llvm/ADT/SmallVector.h"
+#include "llvm/ADT/Statistic.h"
 #include "llvm/ADT/StringExtras.h"
 #include "llvm/Analysis/AliasAnalysis.h"
 #include "llvm/Analysis/BlockFrequencyInfo.h"
 #include "llvm/Analysis/LoopInfo.h"
 #include "llvm/Analysis/LoopIterator.h"
 #include "llvm/Analysis/LoopPass.h"
 #include "llvm/Analysis/ScalarEvolution.h"
 #include "llvm/Analysis/ScalarEvolutionExpander.h"

 #include "llvm/Transforms/Utils/Local.h"
 #include "llvm/Transforms/Utils/VectorUtils.h"
 #include <algorithm>
 #include <map>
 
 using namespace llvm;
 using namespace llvm::PatternMatch;
 
+STATISTIC(LoopsVectorized, "Number of loops vectorized");
+STATISTIC(LoopsAnalyzed, "Number of loops analyzed for vectorization");
+
 static cl::opt<unsigned>
 VectorizationFactor("force-vector-width", cl::init(0), cl::Hidden,
                     cl::desc("Sets the SIMD width. Zero is autoselect."));
 
 static cl::opt<unsigned>
 VectorizationUnroll("force-vector-unroll", cl::init(0), cl::Hidden,
                     cl::desc("Sets the vectorization unroll count. "
                              "Zero is autoselect."));
bool runOnFunction(Function &F) override {
     // Build up a worklist of inner-loops to vectorize. This is necessary as
     // the act of vectorizing or partially unrolling a loop creates new loops
     // and can invalidate iterators across the loops.
     SmallVector<Loop *, 8> Worklist;
 
     for (Loop *L : *LI)
       addInnerLoop(*L, Worklist);
 
+    LoopsAnalyzed += Worklist.size();
+
     // Now walk the identified inner loops.
     bool Changed = false;
     while (!Worklist.empty())
       Changed |= processLoop(Worklist.pop_back_val());
 
     // Process each loop nest in the function.
     return Changed;
   }
if (VF.Width == 1) {
       DEBUG(dbgs() << "LV: Trying to at least unroll the loops.\n");
       // We decided not to vectorize, but we may want to unroll.
       InnerLoopUnroller Unroller(L, SE, LI, DT, DL, TLI, UF);
       Unroller.vectorize(&LVL);
     } else {
       // If we decided that it is *legal* to vectorize the loop then do it.
       InnerLoopVectorizer LB(L, SE, LI, DT, DL, TLI, VF.Width, UF);
       LB.vectorize(&LVL);
+      ++LoopsVectorized;
     }
 
     // Mark the loop as already vectorized to avoid vectorizing again.
     Hints.setAlreadyVectorized(L);
 
     DEBUG(verifyFunction(*L->getHeader()->getParent()));
     return true;
   }

test/Transforms/LoopVectorize/vect.stats.ll

+; RUN: opt < %s  -loop-vectorize -force-vector-unroll=4 -force-vector-width=4 -debug-only=loop-vectorize -stats -S 2>&1 | FileCheck %s
+; REQUIRES: asserts
+
+;
+; We have 2 loops, one of them is vectorizable and the second one is not.
+;
+
+; CHECK: 2 loop-vectorize               - Number of loops analyzed for vectorization
+; CHECK: 1 loop-vectorize               - Number of loops vectorized
+
+target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64-S128"
+target triple = "x86_64-unknown-linux-gnu"
+
+define void @vectorized(float* nocapture %a, i64 %size) {
+entry:
+  %cmp1 = icmp sgt i64 %size, 0
+  br i1 %cmp1, label %for.header, label %for.end
+
+for.header:
+  %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
+  %cmp2 = icmp sgt i64 %indvars.iv, %size
+  br i1 %cmp2, label %for.end, label %for.body
+
+for.body:
+
+  %arrayidx = getelementptr inbounds float* %a, i64 %indvars.iv
+  %0 = load float* %arrayidx, align 4
+  %mul = fmul float %0, %0
+  store float %mul, float* %arrayidx, align 4
+
+  %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
+  br label %for.header
+
+for.end:
+  ret void
+}
+
+define void @not_vectorized(float* nocapture %a, i64 %size) {
+entry:
+  %cmp1 = icmp sgt i64 %size, 0
+  br i1 %cmp1, label %for.header, label %for.end
+
+for.header:
+  %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
+  %cmp2 = icmp sgt i64 %indvars.iv, %size
+  br i1 %cmp2, label %for.end, label %for.body
+
+for.body:
+
+  %0 = add nsw i64 %indvars.iv, -5
+  %arrayidx = getelementptr inbounds float* %a, i64 %0
+  %1 = load float* %arrayidx, align 4
+  %2 = add nsw i64 %indvars.iv, 2
+  %arrayidx2 = getelementptr inbounds float* %a, i64 %2
+  %3 = load float* %arrayidx2, align 4
+  %mul = fmul float %1, %3
+  %arrayidx4 = getelementptr inbounds float* %a, i64 %indvars.iv
+  store float %mul, float* %arrayidx4, align 4
+
+  %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
+  br label %for.header
+
+for.end:
+  ret void
+}
 No newline at end of file