LoopInfo: introduce Loop::getLocStr; unify debug output (llvm#93051)

Introduce a Loop::getLocStr stolen from LoopVectorize's static function
getDebugLocString in order to have uniform debug output headers across
LoopVectorize, LoopAccessAnalysis, and LoopDistribute. The motivation
for this change is to have UpdateTestChecks recognize the headers and
automatically generate CHECK lines for debug output, with minimal
special-casing.

Author: artagnon

llvm/include/llvm/Analysis/LoopInfo.h

@@ -385,6 +385,11 @@ class LLVM_EXTERNAL_VISIBILITY Loop : public LoopBase<BasicBlock, Loop> {
   /// Return the source code span of the loop.
   LocRange getLocRange() const;
 
+  /// Return a string containing the debug location of the loop (file name +
+  /// line number if present, otherwise module name). Meant to be used for debug
+  /// printing within LLVM_DEBUG.
+  std::string getLocStr() const;
+
   StringRef getName() const {
     if (BasicBlock *Header = getHeader())
       if (Header->hasName())
@@ -693,7 +698,6 @@ llvm::MDNode *
 makePostTransformationMetadata(llvm::LLVMContext &Context, MDNode *OrigLoopID,
                                llvm::ArrayRef<llvm::StringRef> RemovePrefixes,
                                llvm::ArrayRef<llvm::MDNode *> AddAttrs);
-
 } // namespace llvm
 
 #endif

llvm/lib/Analysis/LoopAccessAnalysis.cpp

@@ -2355,9 +2355,9 @@ void MemoryDepChecker::Dependence::print(
 
 bool LoopAccessInfo::canAnalyzeLoop() {
   // We need to have a loop header.
-  LLVM_DEBUG(dbgs() << "LAA: Found a loop in "
-                    << TheLoop->getHeader()->getParent()->getName() << ": "
-                    << TheLoop->getHeader()->getName() << '\n');
+  LLVM_DEBUG(dbgs() << "\nLAA: Checking a loop in '"
+                    << TheLoop->getHeader()->getParent()->getName() << "' from "
+                    << TheLoop->getLocStr() << "\n");
 
   // We can only analyze innermost loops.
   if (!TheLoop->isInnermost()) {
@@ -2386,6 +2386,8 @@ bool LoopAccessInfo::canAnalyzeLoop() {
     return false;
   }
 
+  LLVM_DEBUG(dbgs() << "LAA: Found an analyzable loop: "
+                    << TheLoop->getHeader()->getName() << "\n");
   return true;
 }
 

llvm/lib/Analysis/LoopInfo.cpp

@@ -663,6 +663,17 @@ Loop::LocRange Loop::getLocRange() const {
   return LocRange();
 }
 
+std::string Loop::getLocStr() const {
+  std::string Result;
+  raw_string_ostream OS(Result);
+  if (const DebugLoc LoopDbgLoc = getStartLoc())
+    LoopDbgLoc.print(OS);
+  else
+    // Just print the module name.
+    OS << getHeader()->getParent()->getParent()->getModuleIdentifier();
+  return Result;
+}
+
 #if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
 LLVM_DUMP_METHOD void Loop::dump() const { print(dbgs()); }
 

llvm/lib/Transforms/Scalar/LoopDistribute.cpp

@@ -659,9 +659,9 @@ class LoopDistributeForLoop {
   bool processLoop() {
     assert(L->isInnermost() && "Only process inner loops.");
 
-    LLVM_DEBUG(dbgs() << "\nLDist: In \""
-                      << L->getHeader()->getParent()->getName()
-                      << "\" checking " << *L << "\n");
+    LLVM_DEBUG(dbgs() << "\nLDist: Checking a loop in '"
+                      << L->getHeader()->getParent()->getName() << "' from "
+                      << L->getLocStr() << "\n");
 
     // Having a single exit block implies there's also one exiting block.
     if (!L->getExitBlock())
@@ -686,6 +686,9 @@ class LoopDistributeForLoop {
     if (!Dependences || Dependences->empty())
       return fail("NoUnsafeDeps", "no unsafe dependences to isolate");
 
+    LLVM_DEBUG(dbgs() << "LDist: Found a candidate loop: "
+                      << L->getHeader()->getName() << "\n");
+
     InstPartitionContainer Partitions(L, LI, DT);
 
     // First, go through each memory operation and assign them to consecutive

llvm/lib/Transforms/Vectorize/LoopVectorize.cpp

@@ -1027,23 +1027,6 @@ static void reportVectorization(OptimizationRemarkEmitter *ORE, Loop *TheLoop,
 
 } // end namespace llvm
 
-#ifndef NDEBUG
-/// \return string containing a file name and a line # for the given loop.
-static std::string getDebugLocString(const Loop *L) {
-  std::string Result;
-  if (L) {
-    raw_string_ostream OS(Result);
-    if (const DebugLoc LoopDbgLoc = L->getStartLoc())
-      LoopDbgLoc.print(OS);
-    else
-      // Just print the module name.
-      OS << L->getHeader()->getParent()->getParent()->getModuleIdentifier();
-    OS.flush();
-  }
-  return Result;
-}
-#endif
-
 namespace llvm {
 
 // Loop vectorization cost-model hints how the scalar epilogue loop should be
@@ -9758,13 +9741,9 @@ bool LoopVectorizePass::processLoop(Loop *L) {
   assert((EnableVPlanNativePath || L->isInnermost()) &&
          "VPlan-native path is not enabled. Only process inner loops.");
 
-#ifndef NDEBUG
-  const std::string DebugLocStr = getDebugLocString(L);
-#endif /* NDEBUG */
-
   LLVM_DEBUG(dbgs() << "\nLV: Checking a loop in '"
                     << L->getHeader()->getParent()->getName() << "' from "
-                    << DebugLocStr << "\n");
+                    << L->getLocStr() << "\n");
 
   LoopVectorizeHints Hints(L, InterleaveOnlyWhenForced, *ORE, TTI);
 
@@ -10034,15 +10013,15 @@ bool LoopVectorizePass::processLoop(Loop *L) {
     });
   } else if (VectorizeLoop && !InterleaveLoop) {
     LLVM_DEBUG(dbgs() << "LV: Found a vectorizable loop (" << VF.Width
-                      << ") in " << DebugLocStr << '\n');
+                      << ") in " << L->getLocStr() << '\n');
     ORE->emit([&]() {
       return OptimizationRemarkAnalysis(LV_NAME, IntDiagMsg.first,
                                         L->getStartLoc(), L->getHeader())
              << IntDiagMsg.second;
     });
   } else if (VectorizeLoop && InterleaveLoop) {
     LLVM_DEBUG(dbgs() << "LV: Found a vectorizable loop (" << VF.Width
-                      << ") in " << DebugLocStr << '\n');
+                      << ") in " << L->getLocStr() << '\n');
     LLVM_DEBUG(dbgs() << "LV: Interleave Count is " << IC << '\n');
   }
 

llvm/test/Analysis/LoopAccessAnalysis/debug-loc.ll

@@ -0,0 +1,34 @@
+; REQUIRES: asserts
+; RUN: opt -passes='print<access-info>' -debug-only=loop-accesses \
+; RUN:   -disable-output -S %s 2>&1 | FileCheck %s
+
+define void @negative_step(ptr nocapture %A) {
+; CHECK-LABEL: LAA: Checking a loop in 'negative_step' from negative_step.c:5:2
+entry:
+  %A.plus.1 = getelementptr i32, ptr %A, i64 1
+  br label %loop
+
+loop:
+  %iv = phi i64 [ 1022, %entry ], [ %iv.next, %loop ]
+  %gep.A = getelementptr inbounds i32, ptr %A, i64 %iv
+  %l = load i32, ptr %gep.A, align 4
+  %add = add nsw i32 %l, 1
+  %gep.A.plus.1 = getelementptr i32, ptr %A.plus.1, i64 %iv
+  store i32 %add, ptr %gep.A.plus.1, align 4
+  %iv.next = add nsw i64 %iv, -1
+  %cmp.not = icmp eq i64 %iv, 0
+  br i1 %cmp.not, label %exit, label %loop, !dbg !2
+
+exit:
+  ret void
+}
+
+!llvm.module.flags = !{!5, !6, !7}
+
+!0 = !DIFile(filename: "negative_step.c", directory: "/")
+!1 = distinct !DISubprogram(name: "negative_step", scope: !0, file: !0, unit: !4)
+!2 = !DILocation(line: 5, column: 2, scope: !1)
+!4 = distinct !DICompileUnit(language: DW_LANG_C99, file: !0, producer: "clang")
+!5 = !{i32 1, !"Debug Info Version", i32 3}
+!6 = !{i32 2, !"Dwarf Version", i32 2}
+!7 = !{i32 1, !"PIC Level", i32 2}

llvm/test/Analysis/LoopAccessAnalysis/print-order.ll

@@ -6,7 +6,8 @@
 ;    A[i+1] = A[i] + 1;
 ; }
 
-; CHECK: LAA: Found a loop in negative_step: loop
+; CHECK-LABEL: 'negative_step'
+; CHECK: LAA: Found an analyzable loop: loop
 ; CHECK: LAA: Checking memory dependencies
 ; CHECK-NEXT: LAA: Src Scev: {(4092 + %A),+,-4}<nw><%loop>Sink Scev: {(4088 + %A)<nuw>,+,-4}<nw><%loop>(Induction step: -1)
 ; CHECK-NEXT: LAA: Distance for   store i32 %add, ptr %gep.A.plus.1, align 4 to   %l = load i32, ptr %gep.A, align 4: -4
@@ -37,7 +38,8 @@ exit:
 ;    A[i-1] = A[i] + 1;
 ; }
 
-; CHECK: LAA: Found a loop in positive_step: loop
+; CHECK-LABEL: 'positive_step'
+; CHECK: LAA: Found an analyzable loop: loop
 ; CHECK: LAA: Checking memory dependencies
 ; CHECK-NEXT: LAA: Src Scev: {(4 + %A)<nuw>,+,4}<nuw><%loop>Sink Scev: {%A,+,4}<nw><%loop>(Induction step: 1)
 ; CHECK-NEXT: LAA: Distance for   %l = load i32, ptr %gep.A, align 4 to   store i32 %add, ptr %gep.A.minus.1, align 4: -4

llvm/test/Analysis/LoopAccessAnalysis/scalable-vector-regression-tests.ll

@@ -24,7 +24,8 @@ loop.end:
   ret void
 }
 
-; CHECK-LABEL: LAA: Found a loop in regression_test_loop_access_scalable_typesize
+; CHECK-LABEL: 'regression_test_loop_access_scalable_typesize'
+; CHECK: LAA: Found an analyzable loop: vector.body
 ; CHECK: LAA: Bad stride - Scalable object:
 define void @regression_test_loop_access_scalable_typesize(ptr %input_ptr) {
 entry:
@@ -42,7 +43,8 @@ end:
   ret void
 }
 
-; CHECK-LABEL: LAA: Found a loop in regression_test_loop_access_scalable_typesize_nonscalable_object
+; CHECK-LABEL: 'regression_test_loop_access_scalable_typesize_nonscalable_object'
+; CHECK: LAA: Found an analyzable loop: vector.body
 ; CHECK: LAA: Bad stride - Scalable object:
 define void @regression_test_loop_access_scalable_typesize_nonscalable_object(ptr %input_ptr) {
 entry:

llvm/test/Transforms/LoopDistribute/debug-loc.ll

@@ -0,0 +1,43 @@
+; REQUIRES: asserts
+; RUN: opt -passes=loop-distribute -enable-loop-distribute \
+; RUN:   -debug-only=loop-distribute -disable-output 2>&1 %s | FileCheck %s
+
+define void @f(ptr noalias %a, ptr noalias %b, ptr noalias %c, ptr noalias %d, i64 %stride) {
+; CHECK-LABEL: LDist: Checking a loop in 'f' from f.c:5:2
+entry:
+  br label %for.body
+
+for.body:                                         ; preds = %for.body, %entry
+  %ind = phi i64 [ 0, %entry ], [ %add, %for.body ]
+  %gep.a = getelementptr inbounds i32, ptr %a, i64 %ind
+  %load.a = load i32, ptr %gep.a, align 4
+  %gep.b = getelementptr inbounds i32, ptr %b, i64 %ind
+  %load.b = load i32, ptr %gep.b, align 4
+  %mul.a = mul i32 %load.b, %load.a
+  %add = add nuw nsw i64 %ind, 1
+  %gep.a.plus4 = getelementptr inbounds i32, ptr %a, i64 %add
+  store i32 %mul.a, ptr %gep.a.plus4, align 4
+  %gep.d = getelementptr inbounds i32, ptr %d, i64 %ind
+  %loadD = load i32, ptr %gep.d, align 4
+  %mul = mul i64 %ind, %stride
+  %gep.strided.a = getelementptr inbounds i32, ptr %a, i64 %mul
+  %load.strided.a = load i32, ptr %gep.strided.a, align 4
+  %mul.c = mul i32 %loadD, %load.strided.a
+  %gep.c = getelementptr inbounds i32, ptr %c, i64 %ind
+  store i32 %mul.c, ptr %gep.c, align 4
+  %exitcond = icmp eq i64 %add, 20
+  br i1 %exitcond, label %exit, label %for.body, !dbg !2
+
+exit:                                             ; preds = %for.body
+  ret void
+}
+
+!llvm.module.flags = !{!5, !6, !7}
+
+!0 = !DIFile(filename: "f.c", directory: "/")
+!1 = distinct !DISubprogram(name: "f", scope: !0, file: !0, unit: !4)
+!2 = !DILocation(line: 5, column: 2, scope: !1)
+!4 = distinct !DICompileUnit(language: DW_LANG_C99, file: !0, producer: "clang")
+!5 = !{i32 1, !"Debug Info Version", i32 3}
+!6 = !{i32 2, !"Dwarf Version", i32 2}
+!7 = !{i32 1, !"PIC Level", i32 2}

llvm/test/Transforms/LoopDistribute/debug-print.ll

@@ -0,0 +1,44 @@
+; REQUIRES: asserts
+; RUN: opt -passes=loop-distribute -enable-loop-distribute \
+; RUN:   -debug-only=loop-distribute -disable-output 2>&1 %s | FileCheck %s
+
+define void @f(ptr noalias %a, ptr noalias %b, ptr noalias %c, ptr noalias %d, i64 %stride) {
+; CHECK-LABEL: 'f'
+; CHECK:        LDist: Found a candidate loop: for.body
+; CHECK:        Backward dependences:
+; CHECK-NEXT:     Backward:
+; CHECK-NEXT:         %load.a = load i32, ptr %gep.a, align 4 ->
+; CHECK-NEXT:         store i32 %mul.a, ptr %gep.a.plus4, align 4
+; CHECK:        Seeded partitions:
+; CHECK:        Partition 0
+; CHECK:        Partition 1
+; CHECK:        Partition 2
+; CHECK:        Partition 3
+; CHECK:        Distributing loop
+entry:
+  br label %for.body
+
+for.body:                                         ; preds = %for.body, %entry
+  %ind = phi i64 [ 0, %entry ], [ %add, %for.body ]
+  %gep.a = getelementptr inbounds i32, ptr %a, i64 %ind
+  %load.a = load i32, ptr %gep.a, align 4
+  %gep.b = getelementptr inbounds i32, ptr %b, i64 %ind
+  %load.b = load i32, ptr %gep.b, align 4
+  %mul.a = mul i32 %load.b, %load.a
+  %add = add nuw nsw i64 %ind, 1
+  %gep.a.plus4 = getelementptr inbounds i32, ptr %a, i64 %add
+  store i32 %mul.a, ptr %gep.a.plus4, align 4
+  %gep.d = getelementptr inbounds i32, ptr %d, i64 %ind
+  %loadD = load i32, ptr %gep.d, align 4
+  %mul = mul i64 %ind, %stride
+  %gep.strided.a = getelementptr inbounds i32, ptr %a, i64 %mul
+  %load.strided.a = load i32, ptr %gep.strided.a, align 4
+  %mul.c = mul i32 %loadD, %load.strided.a
+  %gep.c = getelementptr inbounds i32, ptr %c, i64 %ind
+  store i32 %mul.c, ptr %gep.c, align 4
+  %exitcond = icmp eq i64 %add, 20
+  br i1 %exitcond, label %exit, label %for.body
+
+exit:                                             ; preds = %for.body
+  ret void
+}

llvm/test/Transforms/LoopVectorize/ARM/mve-hoist-runtime-checks.ll

@@ -18,7 +18,8 @@ target triple = "thumbv8.1m.main-none-unknown-eabi"
 ; NOTE: The strides of the starting address values in the inner loop differ, i.e.
 ; '(i * (n + 1))' vs '(i * n)'.
 
-; DEBUG-LABEL: LAA: Found a loop in diff_checks:
+; DEBUG-LABEL: 'diff_checks'
+; DEBUG:      LAA: Found an analyzable loop: inner.loop
 ; DEBUG:      LAA: Not creating diff runtime check, since these  cannot be hoisted out of the outer loop
 ; DEBUG:      LAA: Adding RT check for range:
 ; DEBUG-NEXT: LAA: Expanded RT check for range to include outer loop in order to permit hoisting