[MemProf] Support for random hotness when writing profile (#113998)

Add support for generating random hotness in the memprof profile writer,
to be used for testing. The random seed is printed to stderr, and an
additional option enables providing a specific seed in order to
reproduce a particular random profile.

Author: teresajohnson

llvm/include/llvm/ProfileData/InstrProfWriter.h

@@ -78,12 +78,20 @@ class InstrProfWriter {
   // Whether to serialize the full schema.
   bool MemProfFullSchema;
 
+  // Whether to generated random memprof hotness for testing.
+  bool MemprofGenerateRandomHotness;
+
 public:
+  // For memprof testing, random hotness can be assigned to the contexts if
+  // MemprofGenerateRandomHotness is enabled. The random seed can be either
+  // provided by MemprofGenerateRandomHotnessSeed, or if that is 0, one will be
+  // generated in the writer using the current time.
   InstrProfWriter(
       bool Sparse = false, uint64_t TemporalProfTraceReservoirSize = 0,
       uint64_t MaxTemporalProfTraceLength = 0, bool WritePrevVersion = false,
       memprof::IndexedVersion MemProfVersionRequested = memprof::Version0,
-      bool MemProfFullSchema = false);
+      bool MemProfFullSchema = false, bool MemprofGenerateRandomHotness = false,
+      unsigned MemprofGenerateRandomHotnessSeed = 0);
   ~InstrProfWriter();
 
   StringMap<ProfilingData> &getProfileData() { return FunctionData; }

llvm/include/llvm/ProfileData/MemProf.h

@@ -147,6 +147,15 @@ struct PortableMemInfoBlock {
     return Name;                                                               \
   }
 #include "llvm/ProfileData/MIBEntryDef.inc"
+#undef MIBEntryDef
+
+  // Define setters for each type which can be called by the writer.
+#define MIBEntryDef(NameTag, Name, Type)                                       \
+  void set##Name(Type NewVal) {                                                \
+    assert(Schema[llvm::to_underlying(Meta::Name)]);                           \
+    Name = NewVal;                                                             \
+  }
+#include "llvm/ProfileData/MIBEntryDef.inc"
 #undef MIBEntryDef
 
   void clear() { *this = PortableMemInfoBlock(); }

llvm/lib/ProfileData/InstrProfWriter.cpp

@@ -19,6 +19,7 @@
 #include "llvm/ProfileData/InstrProf.h"
 #include "llvm/ProfileData/MemProf.h"
 #include "llvm/ProfileData/ProfileCommon.h"
+#include "llvm/Support/CommandLine.h"
 #include "llvm/Support/Compression.h"
 #include "llvm/Support/Endian.h"
 #include "llvm/Support/EndianStream.h"
@@ -184,13 +185,25 @@ class InstrProfRecordWriterTrait {
 InstrProfWriter::InstrProfWriter(
     bool Sparse, uint64_t TemporalProfTraceReservoirSize,
     uint64_t MaxTemporalProfTraceLength, bool WritePrevVersion,
-    memprof::IndexedVersion MemProfVersionRequested, bool MemProfFullSchema)
+    memprof::IndexedVersion MemProfVersionRequested, bool MemProfFullSchema,
+    bool MemprofGenerateRandomHotness,
+    unsigned MemprofGenerateRandomHotnessSeed)
     : Sparse(Sparse), MaxTemporalProfTraceLength(MaxTemporalProfTraceLength),
       TemporalProfTraceReservoirSize(TemporalProfTraceReservoirSize),
       InfoObj(new InstrProfRecordWriterTrait()),
       WritePrevVersion(WritePrevVersion),
       MemProfVersionRequested(MemProfVersionRequested),
-      MemProfFullSchema(MemProfFullSchema) {}
+      MemProfFullSchema(MemProfFullSchema),
+      MemprofGenerateRandomHotness(MemprofGenerateRandomHotness) {
+  // Set up the random number seed if requested.
+  if (MemprofGenerateRandomHotness) {
+    unsigned seed = MemprofGenerateRandomHotnessSeed
+                        ? MemprofGenerateRandomHotnessSeed
+                        : std::time(nullptr);
+    errs() << "random hotness seed = " << seed << "\n";
+    std::srand(seed);
+  }
+}
 
 InstrProfWriter::~InstrProfWriter() { delete InfoObj; }
 
@@ -273,13 +286,34 @@ void InstrProfWriter::addRecord(StringRef Name, uint64_t Hash,
 
 void InstrProfWriter::addMemProfRecord(
     const Function::GUID Id, const memprof::IndexedMemProfRecord &Record) {
-  auto [Iter, Inserted] = MemProfData.Records.insert({Id, Record});
+  auto NewRecord = Record;
+  // Provoke random hotness values if requested. We specify the lifetime access
+  // density and lifetime length that will result in a cold or not cold hotness.
+  // See the logic in getAllocType() in Analysis/MemoryProfileInfo.cpp.
+  if (MemprofGenerateRandomHotness) {
+    for (auto &Alloc : NewRecord.AllocSites) {
+      // To get a not cold context, set the lifetime access density to the
+      // maximum value and the lifetime to 0.
+      uint64_t NewTLAD = std::numeric_limits<uint64_t>::max();
+      uint64_t NewTL = 0;
+      bool IsCold = std::rand() % 2;
+      if (IsCold) {
+        // To get a cold context, set the lifetime access density to 0 and the
+        // lifetime to the maximum value.
+        NewTLAD = 0;
+        NewTL = std::numeric_limits<uint64_t>::max();
+      }
+      Alloc.Info.setTotalLifetimeAccessDensity(NewTLAD);
+      Alloc.Info.setTotalLifetime(NewTL);
+    }
+  }
+  auto [Iter, Inserted] = MemProfData.Records.insert({Id, NewRecord});
   // If we inserted a new record then we are done.
   if (Inserted) {
     return;
   }
   memprof::IndexedMemProfRecord &Existing = Iter->second;
-  Existing.merge(Record);
+  Existing.merge(NewRecord);
 }
 
 bool InstrProfWriter::addMemProfFrame(const memprof::FrameId Id,

llvm/test/Transforms/PGOProfile/memprof.ll

@@ -66,6 +66,18 @@
 ;; Check that the total sizes are reported if requested.
 ; RUN: opt < %s -passes='memprof-use<profile-filename=%t.memprofdata>' -pgo-warn-missing-function -S -memprof-report-hinted-sizes 2>&1 | FileCheck %s --check-prefixes=TOTALSIZES
 
+;; Make sure we emit a random hotness seed if requested.
+; RUN: llvm-profdata merge -memprof-random-hotness %S/Inputs/memprof.memprofraw --profiled-binary %S/Inputs/memprof.exe -o %t.memprofdatarand 2>&1 | FileCheck %s --check-prefix=RAND
+; RAND: random hotness seed =
+;; Can't check the exact values, but make sure applying the random profile
+;; succeeds with the same stats
+; RUN: opt < %s -passes='memprof-use<profile-filename=%t.memprofdatarand>' -pgo-warn-missing-function -S -stats 2>&1 | FileCheck %s --check-prefixes=ALL,MEMPROFONLY,MEMPROFSTATS
+
+;; Make sure we use a specific random hotness seed if requested.
+; RUN: llvm-profdata merge -memprof-random-hotness -memprof-random-hotness-seed=1730170724 %S/Inputs/memprof.memprofraw --profiled-binary %S/Inputs/memprof.exe -o %t.memprofdatarand2 2>&1 | FileCheck %s --check-prefix=RAND2
+; RAND2: random hotness seed = 1730170724
+; RUN: opt < %s -passes='memprof-use<profile-filename=%t.memprofdatarand2>' -pgo-warn-missing-function -S -stats 2>&1 | FileCheck %s --check-prefixes=MEMPROFRAND2,ALL,MEMPROFONLY,MEMPROFSTATS
+
 ; MEMPROFMATCHINFO: MemProf notcold context with id 1093248920606587996 has total profiled size 10 is matched
 ; MEMPROFMATCHINFO: MemProf notcold context with id 5725971306423925017 has total profiled size 10 is matched
 ; MEMPROFMATCHINFO: MemProf notcold context with id 6792096022461663180 has total profiled size 10 is matched
@@ -372,6 +384,13 @@ for.end:                                          ; preds = %for.cond
 ; MEMPROFNOCOLINFO: ![[C10]] = !{i64 -4535090212904553409}
 ; MEMPROFNOCOLINFO: ![[C11]] = !{i64 3577763375057267810}
 
+;; For the specific random seed, this is the expected order of hotness
+; MEMPROFRAND2: !"cold"
+; MEMPROFRAND2: !"cold"
+; MEMPROFRAND2: !"cold"
+; MEMPROFRAND2: !"hot"
+; MEMPROFRAND2: !"hot"
+
 ; MEMPROFSTATS:  8 memprof - Number of alloc contexts in memory profile.
 ; MEMPROFSTATS: 10 memprof - Number of callsites in memory profile.
 ; MEMPROFSTATS:  6 memprof - Number of functions having valid memory profile.

llvm/tools/llvm-profdata/llvm-profdata.cpp

@@ -342,6 +342,15 @@ cl::opt<bool> MemProfFullSchema(
     "memprof-full-schema", cl::Hidden, cl::sub(MergeSubcommand),
     cl::desc("Use the full schema for serialization"), cl::init(false));
 
+static cl::opt<bool>
+    MemprofGenerateRandomHotness("memprof-random-hotness", cl::init(false),
+                                 cl::Hidden, cl::sub(MergeSubcommand),
+                                 cl::desc("Generate random hotness values"));
+static cl::opt<unsigned> MemprofGenerateRandomHotnessSeed(
+    "memprof-random-hotness-seed", cl::init(0), cl::Hidden,
+    cl::sub(MergeSubcommand),
+    cl::desc("Random hotness seed to use (0 to generate new seed)"));
+
 // Options specific to overlap subcommand.
 cl::opt<std::string> BaseFilename(cl::Positional, cl::Required,
                                   cl::desc("<base profile file>"),
@@ -641,7 +650,8 @@ struct WriterContext {
                 SmallSet<instrprof_error, 4> &WriterErrorCodes,
                 uint64_t ReservoirSize = 0, uint64_t MaxTraceLength = 0)
       : Writer(IsSparse, ReservoirSize, MaxTraceLength, DoWritePrevVersion,
-               MemProfVersionRequested, MemProfFullSchema),
+               MemProfVersionRequested, MemProfFullSchema,
+               MemprofGenerateRandomHotness, MemprofGenerateRandomHotnessSeed),
         ErrLock(ErrLock), WriterErrorCodes(WriterErrorCodes) {}
 };