Add quotes for values in the diagnostic remark YAML outputs where necessary

[hfinkel]

Summary

Some of the strings used for the YAML diagnostic remark outputs can be parsed as raw string values in YAML, but many cannot (i.e. anything that has a ':' character, as many of the vectorizer outputs do). Quoting these strings is necessary: otherwise they won't parse correctly. We don't want everything to be a quoted string: numbers should be numbers, etc. but we do need to quote where necessary. This has the side effect of preserving leading/trailing spaces in the output strings, which I think also makes sense.

[hfinkel]

Not needed after r283231.

include/llvm/ADT/StringExtras.h

switch (Val % 10) {
       case 2: return "nd";
       case 3: return "rd";
       default: return "th";
     }
   }
 }
 
 /// PrintEscapedString - Print each character of the specified string, escaping
-/// it if it is not printable or if it is an escape char.
-void PrintEscapedString(StringRef Name, raw_ostream &Out);
+/// it if it is not printable or if it is an escape char. The escaping is done
+/// using a backslash, followed by the optional HexPrefix, followed by the
+/// character value in hexadecimal.
+void PrintEscapedString(StringRef Name, raw_ostream &Out,
+                        const char *HexPrefix = "");
 
 namespace detail {
 
 template <typename IteratorT>
 inline std::string join_impl(IteratorT Begin, IteratorT End,
                              StringRef Separator, std::input_iterator_tag) {
   std::string S;
   if (Begin == End)

lib/Analysis/OptimizationDiagnosticInfo.cpp

 //===- OptimizationDiagnosticInfo.cpp - Optimization Diagnostic -*- C++ -*-===//
 //
 //                     The LLVM Compiler Infrastructure
 //
 // This file is distributed under the University of Illinois Open Source
 // License. See LICENSE.TXT for details.
 //
 //===----------------------------------------------------------------------===//
 //
 // Optimization diagnostic interfaces.  It's packaged as an analysis pass so
 // that by using this service passes become dependent on BFI as well.  BFI is
 // used to compute the "hotness" of the diagnostic message.
 //===----------------------------------------------------------------------===//
 
+#include "llvm/ADT/StringExtras.h"
 #include "llvm/Analysis/OptimizationDiagnosticInfo.h"
 #include "llvm/Analysis/BranchProbabilityInfo.h"
 #include "llvm/Analysis/LazyBlockFrequencyInfo.h"
 #include "llvm/Analysis/LoopInfo.h"
 #include "llvm/IR/DebugInfo.h"
 #include "llvm/IR/DiagnosticInfo.h"
 #include "llvm/IR/Dominators.h"
 #include "llvm/IR/LLVMContext.h"
template <> struct MappingTraits<DebugLoc> {
   }
 
   static const bool flow = true;
 };
 
 template <> struct ScalarTraits<DiagnosticInfoOptimizationBase::Argument> {
   static void output(const DiagnosticInfoOptimizationBase::Argument &Arg,
                      void *, llvm::raw_ostream &out) {
-    out << Arg.Key << ": " << Arg.Val;
+    out << Arg.Key << ": ";
+
+    // We quote values that need quotes, but we write out numbers, booleans,
+    // etc. in raw form so that they don't parse as strings.
+    if (needsQuotes(Arg.Val) && !isNull(Arg.Val) && !isBool(Arg.Val) &&
+        !isNumeric(Arg.Val)) {
+      out << '"';
+      PrintEscapedString(Arg.Val, out, "x");
+      out << '"';
+    } else {
+      out << Arg.Val;
+    }
   }
 
   static StringRef input(StringRef scalar, void *,
                          DiagnosticInfoOptimizationBase::Argument &Arg) {
     llvm_unreachable("input not yet implemented");
   }
 
   static bool mustQuote(StringRef) { return false; }

lib/IR/AsmWriter.cpp

static void PrintCallingConv(unsigned cc, raw_ostream &Out) {
   case CallingConv::AMDGPU_VS:     Out << "amdgpu_vs"; break;
   case CallingConv::AMDGPU_GS:     Out << "amdgpu_gs"; break;
   case CallingConv::AMDGPU_PS:     Out << "amdgpu_ps"; break;
   case CallingConv::AMDGPU_CS:     Out << "amdgpu_cs"; break;
   case CallingConv::AMDGPU_KERNEL: Out << "amdgpu_kernel"; break;
   }
 }
 
-void llvm::PrintEscapedString(StringRef Name, raw_ostream &Out) {
+void llvm::PrintEscapedString(StringRef Name, raw_ostream &Out,
+                              const char *HexPrefix) {
   for (unsigned i = 0, e = Name.size(); i != e; ++i) {
     unsigned char C = Name[i];
     if (isprint(C) && C != '\\' && C != '"')
       Out << C;
     else
-      Out << '\\' << hexdigit(C >> 4) << hexdigit(C & 0x0F);
+      Out << '\\' << HexPrefix << hexdigit(C >> 4) << hexdigit(C & 0x0F);
   }
 }
 
 enum PrefixType {
   GlobalPrefix,
   ComdatPrefix,
   LabelPrefix,
   LocalPrefix,

test/Transforms/Inline/optimization-remarks-passed-yaml.ll

 ; YAML:      --- !Analysis
 ; YAML-NEXT: Pass:            inline
 ; YAML-NEXT: Name:            CanBeInlined
 ; YAML-NEXT: DebugLoc:        { File: /tmp/s.c, Line: 4, Column: 10 }
 ; YAML-NEXT: Function:        bar
 ; YAML-NEXT: Hotness:         30
 ; YAML-NEXT: Args:
 ; YAML-NEXT:   - Callee: foo
-; YAML-NEXT:   - String:  can be inlined into
+; YAML-NEXT:   - String: " can be inlined into "
 ; YAML-NEXT:   - Caller: bar
-; YAML-NEXT:   - String:  with cost=
+; YAML-NEXT:   - String: " with cost="
 ; YAML-NEXT:   - Cost: {{[0-9]+}}
-; YAML-NEXT:   - String:  (threshold=
+; YAML-NEXT:   - String: " (threshold="
 ; YAML-NEXT:   - Threshold: {{[0-9]+}}
-; YAML-NEXT:   - String: )
+; YAML-NEXT:   - String: ")"
 ; YAML-NEXT: ...
 ; YAML-NEXT: --- !Passed
 ; YAML-NEXT: Pass:            inline
 ; YAML-NEXT: Name:            Inlined
 ; YAML-NEXT: DebugLoc:        { File: /tmp/s.c, Line: 4, Column: 10 }
 ; YAML-NEXT: Function:        bar
 ; YAML-NEXT: Hotness:         30
 ; YAML-NEXT: Args:
 ; YAML-NEXT:   - Callee: foo
-; YAML-NEXT:   - String:  inlined into
+; YAML-NEXT:   - String: " inlined into "
 ; YAML-NEXT:   - Caller: bar
 ; YAML-NEXT: ...
 
 ; ModuleID = '/tmp/s.c'
 source_filename = "/tmp/s.c"
 target datalayout = "e-m:o-i64:64-f80:128-n8:16:32:64-S128"
 target triple = "x86_64-apple-macosx10.11.0"
 

test/Transforms/Inline/optimization-remarks-yaml.ll

 ; YAML:      --- !Missed
 ; YAML-NEXT: Pass:            inline
 ; YAML-NEXT: Name:            NoDefinition
 ; YAML-NEXT: DebugLoc:        { File: /tmp/s.c, Line: 5, Column: 10 }
 ; YAML-NEXT: Function:        baz
 ; YAML-NEXT: Hotness:         30
 ; YAML-NEXT: Args:
 ; YAML-NEXT:   - Callee: foo
-; YAML-NEXT:   - String:  will not be inlined into
+; YAML-NEXT:   - String: " will not be inlined into "
 ; YAML-NEXT:   - Caller: baz
-; YAML-NEXT:   - String:  because its definition is unavailable
+; YAML-NEXT:   - String: " because its definition is unavailable"
 ; YAML-NEXT: ...
 ; YAML-NEXT: --- !Missed
 ; YAML-NEXT: Pass:            inline
 ; YAML-NEXT: Name:            NoDefinition
 ; YAML-NEXT: DebugLoc:        { File: /tmp/s.c, Line: 5, Column: 18 }
 ; YAML-NEXT: Function:        baz
 ; YAML-NEXT: Hotness:         30
 ; YAML-NEXT: Args:
 ; YAML-NEXT:   - Callee: bar
-; YAML-NEXT:   - String:  will not be inlined into
+; YAML-NEXT:   - String: " will not be inlined into "
 ; YAML-NEXT:   - Caller: baz
-; YAML-NEXT:   - String:  because its definition is unavailable
+; YAML-NEXT:   - String: " because its definition is unavailable"
 ; YAML-NEXT: ...
 
 ; ModuleID = '/tmp/s.c'
 source_filename = "/tmp/s.c"
 target datalayout = "e-m:o-i64:64-f80:128-n8:16:32:64-S128"
 target triple = "x86_64-apple-macosx10.11.0"
 
 ; Function Attrs: nounwind ssp uwtable

test/Transforms/LoopVectorize/optimization-remarks-yaml.ll

+; RUN: opt -S -loop-vectorize -pass-remarks-analysis=loop-vectorize -pass-remarks-output=%t < %s 2>&1 | FileCheck %s
+; RUN: cat %t | FileCheck -check-prefix=YAML %s
+
+;  1     extern int map[];
+;  2     extern int out[];
+;  3
+;  4     void f(int a, int n) {
+;  5       for (int i = 0; i < n; ++i) {
+;  6         out[i] = a;
+;  7         a = map[a];
+;  8       }
+;  9     }
+
+; CHECK: remark: /tmp/s.c:5:3: loop not vectorized: value that could not be identified as reduction is used outside the loop
+
+; YAML: --- !Analysis
+; YAML: Pass:            loop-vectorize
+; YAML: Name:            NonReductionValueUsedOutsideLoop
+; YAML: DebugLoc:        { File: /tmp/s.c, Line: 5, Column: 3 }
+; YAML: Function:        f
+; YAML: Args:            
+; YAML:   - String: "loop not vectorized: "
+; YAML:   - String: value that could not be identified as reduction is used outside the loop
+
+target datalayout = "e-m:o-i64:64-f80:128-n8:16:32:64-S128"
+target triple = "x86_64-apple-macosx10.11.0"
+
+@out = external local_unnamed_addr global [0 x i32], align 4
+@map = external local_unnamed_addr global [0 x i32], align 4
+
+; Function Attrs: norecurse nounwind ssp uwtable
+define void @f(i32 %a, i32 %n) local_unnamed_addr #0 !dbg !6 {
+entry:
+  %cmp7 = icmp sgt i32 %n, 0, !dbg !8
+  br i1 %cmp7, label %for.body.preheader, label %for.cond.cleanup, !dbg !9
+
+for.body.preheader:                               ; preds = %entry
+  %wide.trip.count = zext i32 %n to i64, !dbg !9
+  br label %for.body, !dbg !10
+
+for.cond.cleanup:                                 ; preds = %for.body, %entry
+  ret void, !dbg !11
+
+for.body:                                         ; preds = %for.body.preheader, %for.body
+  %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %for.body.preheader ]
+  %a.addr.08 = phi i32 [ %0, %for.body ], [ %a, %for.body.preheader ]
+
+  %arrayidx = getelementptr inbounds [0 x i32], [0 x i32]* @out, i64 0, i64 %indvars.iv, !dbg !10
+  store i32 %a.addr.08, i32* %arrayidx, align 4, !dbg !12, !tbaa !13
+  %idxprom1 = sext i32 %a.addr.08 to i64, !dbg !17
+  %arrayidx2 = getelementptr inbounds [0 x i32], [0 x i32]* @map, i64 0, i64 %idxprom1, !dbg !17
+  %0 = load i32, i32* %arrayidx2, align 4, !dbg !17, !tbaa !13
+  %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1, !dbg !9
+  %exitcond = icmp eq i64 %indvars.iv.next, %wide.trip.count, !dbg !9
+  br i1 %exitcond, label %for.cond.cleanup, label %for.body, !dbg !9, !llvm.loop !18
+}
+
+attributes #0 = { norecurse nounwind ssp uwtable "correctly-rounded-divide-sqrt-fp-math"="false" "disable-tail-calls"="false" "less-precise-fpmad"="false" "no-frame-pointer-elim"="true" "no-frame-pointer-elim-non-leaf" "no-infs-fp-math"="false" "no-jump-tables"="false" "no-nans-fp-math"="false" "no-signed-zeros-fp-math"="false" "no-trapping-math"="false" "stack-protector-buffer-size"="8" "target-cpu"="core2" "target-features"="+cx16,+fxsr,+mmx,+sse,+sse2,+sse3,+ssse3,+x87" "unsafe-fp-math"="false" "use-soft-float"="false" }
+
+!llvm.dbg.cu = !{!0}
+!llvm.module.flags = !{!3, !4}
+!llvm.ident = !{!5}
+
+!0 = distinct !DICompileUnit(language: DW_LANG_C99, file: !1, producer: "clang version 4.0.0 (trunk 281293) (llvm/trunk 281290)", isOptimized: true, runtimeVersion: 0, emissionKind: NoDebug, enums: !2)
+!1 = !DIFile(filename: "/tmp/s.c", directory: "/tmp")
+!2 = !{}
+!3 = !{i32 2, !"Debug Info Version", i32 3}
+!4 = !{i32 1, !"PIC Level", i32 2}
+!5 = !{!"clang version 4.0.0 (trunk 281293) (llvm/trunk 281290)"}
+!6 = distinct !DISubprogram(name: "f", scope: !1, file: !1, line: 4, type: !7, isLocal: false, isDefinition: true, scopeLine: 4, flags: DIFlagPrototyped, isOptimized: true, unit: !0, variables: !2)
+!7 = !DISubroutineType(types: !2)
+!8 = !DILocation(line: 5, column: 21, scope: !6)
+!9 = !DILocation(line: 5, column: 3, scope: !6)
+!10 = !DILocation(line: 6, column: 5, scope: !6)
+!11 = !DILocation(line: 9, column: 1, scope: !6)
+!12 = !DILocation(line: 6, column: 12, scope: !6)
+!13 = !{!14, !14, i64 0}
+!14 = !{!"int", !15, i64 0}
+!15 = !{!"omnipotent char", !16, i64 0}
+!16 = !{!"Simple C/C++ TBAA"}
+!17 = !DILocation(line: 7, column: 9, scope: !6)
+!18 = distinct !{!18, !9}