[PowerPC] Add profitablilty check for conversion to mtctr loops

Add profitability checks for modifying counted loops to use the mtctr instruction. The latency of mtctr is only justified if there are more than 4 comparisons that will be removed as a result. Usually counted loops are formed relatively early and before unrolling, so most low trip count loops often don't survive. However we want to ensure that if they do, we do not mistakenly update them to mtctr loops. Use CodeMetrics to ensure we are only doing this for small loops with small trip counts. Differential Revision: https://reviews.llvm.org/D38212 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@315592 91177308-0d34-0410-b5e6-96231b3b80d8
baberthal · Oct 12, 2017 · 818cdb5 · 818cdb5
1 parent fadd83b
commit 818cdb5
Show file tree

Hide file tree

Showing 3 changed files with 190 additions and 6 deletions.
diff --git a/lib/Target/PowerPC/PPCCTRLoops.cpp b/lib/Target/PowerPC/PPCCTRLoops.cpp
@@ -26,12 +26,17 @@
 #include "PPC.h"
 #include "PPCSubtarget.h"
 #include "PPCTargetMachine.h"
+#include "PPCTargetTransformInfo.h"
 #include "llvm/ADT/STLExtras.h"
 #include "llvm/ADT/Statistic.h"
+#include "llvm/Analysis/AssumptionCache.h"
+#include "llvm/Analysis/CodeMetrics.h"
 #include "llvm/Analysis/LoopInfo.h"
 #include "llvm/Analysis/ScalarEvolutionExpander.h"
 #include "llvm/Analysis/TargetLibraryInfo.h"
+#include "llvm/Analysis/TargetTransformInfo.h"
 #include "llvm/CodeGen/TargetPassConfig.h"
+#include "llvm/CodeGen/TargetSchedule.h"
 #include "llvm/IR/Constants.h"
 #include "llvm/IR/DerivedTypes.h"
 #include "llvm/IR/Dominators.h"
@@ -64,6 +69,13 @@ using namespace llvm;
 static cl::opt<int> CTRLoopLimit("ppc-max-ctrloop", cl::Hidden, cl::init(-1));
 #endif
 
+// The latency of mtctr is only justified if there are more than 4
+// comparisons that will be removed as a result.
+static cl::opt<unsigned>
+SmallCTRLoopThreshold("min-ctr-loop-threshold", cl::init(4), cl::Hidden,
+                      cl::desc("Loops with a constant trip count smaller than "
+                               "this value will not use the count register."));
+
 STATISTIC(NumCTRLoops, "Number of loops converted to CTR loops");
 
 namespace llvm {
@@ -95,6 +107,8 @@ namespace {
       AU.addRequired<DominatorTreeWrapperPass>();
       AU.addPreserved<DominatorTreeWrapperPass>();
       AU.addRequired<ScalarEvolutionWrapperPass>();
+      AU.addRequired<AssumptionCacheTracker>();
+      AU.addRequired<TargetTransformInfoWrapperPass>();
     }
 
   private:
@@ -107,10 +121,12 @@ namespace {
     const PPCTargetLowering *TLI;
     const DataLayout *DL;
     const TargetLibraryInfo *LibInfo;
+    const TargetTransformInfo *TTI;
     LoopInfo *LI;
     ScalarEvolution *SE;
     DominatorTree *DT;
     bool PreserveLCSSA;
+    TargetSchedModel SchedModel;
   };
 
   char PPCCTRLoops::ID = 0;
@@ -179,6 +195,7 @@ bool PPCCTRLoops::runOnFunction(Function &F) {
   LI = &getAnalysis<LoopInfoWrapperPass>().getLoopInfo();
   SE = &getAnalysis<ScalarEvolutionWrapperPass>().getSE();
   DT = &getAnalysis<DominatorTreeWrapperPass>().getDomTree();
+  TTI = &getAnalysis<TargetTransformInfoWrapperPass>().getTTI(F);
   DL = &F.getParent()->getDataLayout();
   auto *TLIP = getAnalysisIfAvailable<TargetLibraryInfoWrapperPass>();
   LibInfo = TLIP ? &TLIP->getTLI() : nullptr;
@@ -462,10 +479,24 @@ bool PPCCTRLoops::mightUseCTR(BasicBlock *BB) {
 
   return false;
 }
-
 bool PPCCTRLoops::convertToCTRLoop(Loop *L) {
   bool MadeChange = false;
 
+  // Do not convert small short loops to CTR loop.
+  unsigned ConstTripCount = SE->getSmallConstantTripCount(L);
+  if (ConstTripCount && ConstTripCount < SmallCTRLoopThreshold) {
+    SmallPtrSet<const Value *, 32> EphValues;
+    auto AC = getAnalysis<AssumptionCacheTracker>().getAssumptionCache(
+        *L->getHeader()->getParent());
+    CodeMetrics::collectEphemeralValues(L, &AC, EphValues);
+    CodeMetrics Metrics;
+    for (BasicBlock *BB : L->blocks())
+      Metrics.analyzeBasicBlock(BB, *TTI, EphValues);
+    // 6 is an approximate latency for the mtctr instruction.
+    if (Metrics.NumInsts <= (6 * SchedModel.getIssueWidth()))
+      return false;
+  }
+
   // Process nested loops first.
   for (Loop::iterator I = L->begin(), E = L->end(); I != E; ++I) {
     MadeChange |= convertToCTRLoop(*I);

diff --git a/test/CodeGen/PowerPC/ctr-minmaxnum.ll b/test/CodeGen/PowerPC/ctr-minmaxnum.ll
@@ -1,6 +1,5 @@
-; RUN: llc -verify-machineinstrs -mcpu=pwr7 < %s | FileCheck %s
-; RUN: llc -verify-machineinstrs -mcpu=a2q < %s | FileCheck %s --check-prefix=QPX
-target triple = "powerpc64-unknown-linux-gnu"
+; RUN: llc -mtriple=powerpc64-unknown-linux-gnu -verify-machineinstrs -mcpu=pwr7 < %s | FileCheck %s
+; RUN: llc -mtriple=powerpc64-unknown-linux-gnu -verify-machineinstrs -mcpu=a2q < %s | FileCheck %s --check-prefix=QPX
 
 declare float @fabsf(float)
 
@@ -38,6 +37,9 @@ loop_exit:
 ; CHECK-LABEL: test1:
 ; CHECK-NOT: mtctr
 ; CHECK: bl fminf
+; CHECK-NOT: bl fminf
+; CHECK-NOT: mtctr
+; CHECK: blr
 
 define void @test1v(<4 x float> %f, <4 x float>* %fp) {
 entry:
@@ -48,16 +50,23 @@ loop_body:
   %0 = call <4 x float> @llvm.minnum.v4f32(<4 x float> %f, <4 x float> <float 1.0, float 1.0, float 1.0, float 1.0>)
   store <4 x float> %0, <4 x float>* %fp, align 16
   %1 = add i64 %invar_address.dim.0.01, 1
-  %2 = icmp eq i64 %1, 2
+  %2 = icmp eq i64 %1, 4
   br i1 %2, label %loop_exit, label %loop_body
 
 loop_exit:
   ret void
 }
 
 ; CHECK-LABEL: test1v:
+; CHECK: bl fminf
+; CHECK-NOT: mtctr
+; CHECK: bl fminf
 ; CHECK-NOT: mtctr
 ; CHECK: bl fminf
+; CHECK-NOT: mtctr
+; CHECK: bl fminf
+; CHECK-NOT: bl fminf
+; CHECK: blr
 
 ; QPX-LABEL: test1v:
 ; QPX: mtctr
@@ -83,6 +92,9 @@ loop_exit:
 ; CHECK-LABEL: test1a:
 ; CHECK-NOT: mtctr
 ; CHECK: bl fminf
+; CHECK-NOT: bl fminf
+; CHECK-NOT: mtctr
+; CHECK: blr
 
 define void @test2(float %f, float* %fp) {
 entry:
@@ -103,6 +115,9 @@ loop_exit:
 ; CHECK-LABEL: test2:
 ; CHECK-NOT: mtctr
 ; CHECK: bl fmaxf
+; CHECK-NOT: bl fmaxf
+; CHECK-NOT: mtctr
+; CHECK: blr
 
 define void @test2v(<4 x double> %f, <4 x double>* %fp) {
 entry:
@@ -113,16 +128,23 @@ loop_body:
   %0 = call <4 x double> @llvm.maxnum.v4f64(<4 x double> %f, <4 x double> <double 1.0, double 1.0, double 1.0, double 1.0>)
   store <4 x double> %0, <4 x double>* %fp, align 16
   %1 = add i64 %invar_address.dim.0.01, 1
-  %2 = icmp eq i64 %1, 2
+  %2 = icmp eq i64 %1, 4
   br i1 %2, label %loop_exit, label %loop_body
 
 loop_exit:
   ret void
 }
 
 ; CHECK-LABEL: test2v:
+; CHECK: bl fmax
+; CHECK-NOT: mtctr
+; CHECK: bl fmax
+; CHECK-NOT: mtctr
+; CHECK: bl fmax
 ; CHECK-NOT: mtctr
 ; CHECK: bl fmax
+; CHECK-NOT: bl fmax
+; CHECK: blr
 
 ; QPX-LABEL: test2v:
 ; QPX: mtctr
@@ -148,6 +170,9 @@ loop_exit:
 ; CHECK-LABEL: test2a:
 ; CHECK-NOT: mtctr
 ; CHECK: bl fmaxf
+; CHECK-NOT: bl fmaxf
+; CHECK-NOT: mtctr
+; CHECK: blr
 
 define void @test3(double %f, double* %fp) {
 entry:
@@ -168,6 +193,9 @@ loop_exit:
 ; CHECK-LABEL: test3:
 ; CHECK-NOT: mtctr
 ; CHECK: bl fmin
+; CHECK-NOT: bl fmin
+; CHECK-NOT: mtctr
+; CHECK: blr
 
 define void @test3a(double %f, double* %fp) {
 entry:
@@ -188,6 +216,9 @@ loop_exit:
 ; CHECK-LABEL: test3a:
 ; CHECK-NOT: mtctr
 ; CHECK: bl fmin
+; CHECK-NOT: bl fmin
+; CHECK-NOT: mtctr
+; CHECK: blr
 
 define void @test4(double %f, double* %fp) {
 entry:
@@ -208,6 +239,9 @@ loop_exit:
 ; CHECK-LABEL: test4:
 ; CHECK-NOT: mtctr
 ; CHECK: bl fmax
+; CHECK-NOT: bl fmax
+; CHECK-NOT: mtctr
+; CHECK: blr
 
 define void @test4a(double %f, double* %fp) {
 entry:
@@ -228,4 +262,7 @@ loop_exit:
 ; CHECK-LABEL: test4a:
 ; CHECK-NOT: mtctr
 ; CHECK: bl fmax
+; CHECK-NOT: bl fmax
+; CHECK-NOT: mtctr
+; CHECK: blr
 
diff --git a/test/CodeGen/PowerPC/ctrloop-shortLoops.ll b/test/CodeGen/PowerPC/ctrloop-shortLoops.ll
@@ -0,0 +1,116 @@
+; RUN: llc < %s -mtriple=powerpc64le-unknown-linux-gnu -verify-machineinstrs -mcpu=pwr8 | FileCheck %s
+; RUN: llc < %s -mtriple=powerpc64le-unknown-linux-gnu -verify-machineinstrs -mcpu=a2q | FileCheck %s
+
+; Verify that we do NOT generate the mtctr instruction for loop trip counts < 4
+; The latency of the mtctr is only justified if there are more than 4 comparisons that are removed as a result.
+
+@a = common local_unnamed_addr global i32 0, align 4
+@arr = common local_unnamed_addr global [5 x i32] zeroinitializer, align 4
+
+; Function Attrs: norecurse nounwind readonly
+define signext i32 @testTripCount2(i32 signext %a) {
+
+; CHECK-LABEL: testTripCount2:
+; CHECK-NOT: mtctr
+; CHECK: blr
+
+entry:
+  br label %for.body
+
+for.cond.cleanup:                                 ; preds = %for.body
+  ret i32 %add
+
+for.body:                                         ; preds = %entry, %for.body
+  %indvars.iv = phi i64 [ 1, %entry ], [ %indvars.iv.next, %for.body ]
+  %Sum.05 = phi i32 [ 0, %entry ], [ %add, %for.body ]
+  %arrayidx = getelementptr inbounds [5 x i32], [5 x i32]* @arr, i64 0, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
+  %add = add nsw i32 %0, %Sum.05
+  %indvars.iv.next = add nsw i64 %indvars.iv, -1
+  %tobool = icmp eq i64 %indvars.iv, 0
+  br i1 %tobool, label %for.cond.cleanup, label %for.body
+}
+
+; Function Attrs: norecurse nounwind readonly
+define signext i32 @testTripCount3(i32 signext %a) {
+
+; CHECK-LABEL: testTripCount3:
+; CHECK-NOT: mtctr
+; CHECK: blr
+
+entry:
+  br label %for.body
+
+for.cond.cleanup:                                 ; preds = %for.body
+  ret i32 %add
+
+for.body:                                         ; preds = %entry, %for.body
+  %indvars.iv = phi i64 [ 2, %entry ], [ %indvars.iv.next, %for.body ]
+  %Sum.05 = phi i32 [ 0, %entry ], [ %add, %for.body ]
+  %arrayidx = getelementptr inbounds [5 x i32], [5 x i32]* @arr, i64 0, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
+  %add = add nsw i32 %0, %Sum.05
+  %indvars.iv.next = add nsw i64 %indvars.iv, -1
+  %tobool = icmp eq i64 %indvars.iv, 0
+  br i1 %tobool, label %for.cond.cleanup, label %for.body
+}
+
+; Function Attrs: norecurse nounwind readonly
+
+define signext i32 @testTripCount4(i32 signext %a) {
+
+; CHECK-LABEL: testTripCount4:
+; CHECK: mtctr
+; CHECK: bdnz
+
+entry:
+  br label %for.body
+
+for.cond.cleanup:                                 ; preds = %for.body
+  ret i32 %add
+
+for.body:                                         ; preds = %entry, %for.body
+  %indvars.iv = phi i64 [ 3, %entry ], [ %indvars.iv.next, %for.body ]
+  %Sum.05 = phi i32 [ 0, %entry ], [ %add, %for.body ]
+  %arrayidx = getelementptr inbounds [5 x i32], [5 x i32]* @arr, i64 0, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
+  %add = add nsw i32 %0, %Sum.05
+  %indvars.iv.next = add nsw i64 %indvars.iv, -1
+  %tobool = icmp eq i64 %indvars.iv, 0
+  br i1 %tobool, label %for.cond.cleanup, label %for.body
+}
+
+; Function Attrs: norecurse nounwind
+define signext i32 @testTripCount2NonSmallLoop() {
+
+; CHECK-LABEL: testTripCount2NonSmallLoop:
+; CHECK: mtctr
+; CHECK: blr
+
+entry:
+  %.pre = load i32, i32* @a, align 4
+  br label %for.body
+
+for.body:                                         ; preds = %entry, %if.end
+  %0 = phi i32 [ %.pre, %entry ], [ %1, %if.end ]
+  %dec4 = phi i32 [ 1, %entry ], [ %dec, %if.end ]
+  %b.03 = phi i8 [ 0, %entry ], [ %b.1, %if.end ]
+  %tobool1 = icmp eq i32 %0, 0
+  br i1 %tobool1, label %if.end, label %if.then
+
+if.then:                                          ; preds = %for.body
+  store i32 2, i32* @a, align 4
+  br label %if.end
+
+if.end:                                           ; preds = %for.body, %if.then
+  %1 = phi i32 [ 2, %if.then ], [ 0, %for.body ]
+  %b.1 = phi i8 [ 2, %if.then ], [ %b.03, %for.body ]
+  %dec = add nsw i32 %dec4, -1
+  %tobool = icmp eq i32 %dec4, 0
+  br i1 %tobool, label %for.end, label %for.body
+
+for.end:                                          ; preds = %if.end
+  %conv = zext i8 %b.1 to i32
+  ret i32 %conv
+}
+