//===----------------------------------------------------------------------===//
//
// This file implements the class that reads LLVM sample profiles. It
-// supports two file formats: text and binary. The textual representation
-// is useful for debugging and testing purposes. The binary representation
-// is more compact, resulting in smaller file sizes. However, they can
-// both be used interchangeably.
+// supports three file formats: text, binary and gcov.
//
-// NOTE: If you are making changes to the file format, please remember
-// to document them in the Clang documentation at
-// tools/clang/docs/UsersManual.rst.
+// The textual representation is useful for debugging and testing purposes. The
+// binary representation is more compact, resulting in smaller file sizes.
//
-// Text format
-// -----------
+// The gcov encoding is the one generated by GCC's AutoFDO profile creation
+// tool (https://github.com/google/autofdo)
//
-// Sample profiles are written as ASCII text. The file is divided into
-// sections, which correspond to each of the functions executed at runtime.
-// Each section has the following format
+// All three encodings can be used interchangeably as an input sample profile.
//
-// function1:total_samples:total_head_samples
-// offset1[.discriminator]: number_of_samples [fn1:num fn2:num ... ]
-// offset2[.discriminator]: number_of_samples [fn3:num fn4:num ... ]
-// ...
-// offsetN[.discriminator]: number_of_samples [fn5:num fn6:num ... ]
-// offsetA[.discriminator]: fnA:num_of_total_samples
-// offsetA1[.discriminator]: number_of_samples [fn7:num fn8:num ... ]
-// ...
-//
-// This is a nested tree in which the identation represent the nest level
-// of the inline stack. There is no blank line in the file. And the spacing
-// within a single line is fixed. Additional spaces will result in an error
-// while reading the file.
-//
-// Inline stack is a stack of source locations in which the top of the stack
-// represents the leaf function, and the bottom of the stack represents the
-// actual symbol in which the instruction belongs.
-//
-// Function names must be mangled in order for the profile loader to
-// match them in the current translation unit. The two numbers in the
-// function header specify how many total samples were accumulated in the
-// function (first number), and the total number of samples accumulated
-// in the prologue of the function (second number). This head sample
-// count provides an indicator of how frequently the function is invoked.
-//
-// There are two types of lines in the function body.
-//
-// * Sampled line represents the profile information of a source location.
-// * Callsite line represents the profile inofrmation of a callsite.
-//
-// Each sampled line may contain several items. Some are optional (marked
-// below):
-//
-// a. Source line offset. This number represents the line number
-// in the function where the sample was collected. The line number is
-// always relative to the line where symbol of the function is
-// defined. So, if the function has its header at line 280, the offset
-// 13 is at line 293 in the file.
-//
-// Note that this offset should never be a negative number. This could
-// happen in cases like macros. The debug machinery will register the
-// line number at the point of macro expansion. So, if the macro was
-// expanded in a line before the start of the function, the profile
-// converter should emit a 0 as the offset (this means that the optimizers
-// will not be able to associate a meaningful weight to the instructions
-// in the macro).
-//
-// b. [OPTIONAL] Discriminator. This is used if the sampled program
-// was compiled with DWARF discriminator support
-// (http://wiki.dwarfstd.org/index.php?title=Path_Discriminators).
-// DWARF discriminators are unsigned integer values that allow the
-// compiler to distinguish between multiple execution paths on the
-// same source line location.
-//
-// For example, consider the line of code ``if (cond) foo(); else bar();``.
-// If the predicate ``cond`` is true 80% of the time, then the edge
-// into function ``foo`` should be considered to be taken most of the
-// time. But both calls to ``foo`` and ``bar`` are at the same source
-// line, so a sample count at that line is not sufficient. The
-// compiler needs to know which part of that line is taken more
-// frequently.
-//
-// This is what discriminators provide. In this case, the calls to
-// ``foo`` and ``bar`` will be at the same line, but will have
-// different discriminator values. This allows the compiler to correctly
-// set edge weights into ``foo`` and ``bar``.
-//
-// c. Number of samples. This is an integer quantity representing the
-// number of samples collected by the profiler at this source
-// location.
-//
-// d. [OPTIONAL] Potential call targets and samples. If present, this
-// line contains a call instruction. This models both direct and
-// number of samples. For example,
-//
-// 130: 7 foo:3 bar:2 baz:7
-//
-// The above means that at relative line offset 130 there is a call
-// instruction that calls one of ``foo()``, ``bar()`` and ``baz()``,
-// with ``baz()`` being the relatively more frequently called target.
-//
-// Each callsite line may contain several items. Some are optional.
-//
-// a. Source line offset. This number represents the line number of the
-// callsite that is inlined in the profiled binary.
-//
-// b. [OPTIONAL] Discriminator. Same as the discriminator for sampled line.
-//
-// c. Number of samples. This is an integer quantity representing the
-// total number of samples collected for the inlined instance at this
-// callsite
//===----------------------------------------------------------------------===//
#include "llvm/ProfileData/SampleProfReader.h"
+#include "llvm/ADT/DenseMap.h"
+#include "llvm/ADT/SmallVector.h"
#include "llvm/Support/Debug.h"
#include "llvm/Support/ErrorOr.h"
#include "llvm/Support/LEB128.h"
#include "llvm/Support/LineIterator.h"
#include "llvm/Support/MemoryBuffer.h"
-#include "llvm/ADT/DenseMap.h"
-#include "llvm/ADT/SmallVector.h"
using namespace llvm::sampleprof;
using namespace llvm;
-/// \brief Print the samples collected for a function on stream \p OS.
-///
-/// \param OS Stream to emit the output to.
-void FunctionSamples::print(raw_ostream &OS, unsigned Indent) const {
- OS << TotalSamples << ", " << TotalHeadSamples << ", " << BodySamples.size()
- << " sampled lines\n";
- for (const auto &SI : BodySamples) {
- LineLocation Loc = SI.first;
- const SampleRecord &Sample = SI.second;
- OS.indent(Indent);
- OS << "line offset: " << Loc.LineOffset
- << ", discriminator: " << Loc.Discriminator
- << ", number of samples: " << Sample.getSamples();
- if (Sample.hasCalls()) {
- OS << ", calls:";
- for (const auto &I : Sample.getCallTargets())
- OS << " " << I.first() << ":" << I.second;
- }
- OS << "\n";
- }
- for (const auto &CS : CallsiteSamples) {
- CallsiteLocation Loc = CS.first;
- const FunctionSamples &CalleeSamples = CS.second;
- OS.indent(Indent);
- OS << "line offset: " << Loc.LineOffset
- << ", discriminator: " << Loc.Discriminator
- << ", inlined callee: " << Loc.CalleeName << ": ";
- CalleeSamples.print(OS, Indent + 2);
- }
-}
-
/// \brief Dump the function profile for \p FName.
///
/// \param FName Name of the function to print.
/// \param OS Stream to emit the output to.
void SampleProfileReader::dumpFunctionProfile(StringRef FName,
raw_ostream &OS) {
- OS << "Function: " << FName << ": ";
- Profiles[FName].print(OS);
+ OS << "Function: " << FName << ": " << Profiles[FName];
}
/// \brief Dump all the function profiles found on stream \p OS.
///
/// \returns true if parsing is successful.
static bool ParseHead(const StringRef &Input, StringRef &FName,
- unsigned &NumSamples, unsigned &NumHeadSamples) {
+ uint64_t &NumSamples, uint64_t &NumHeadSamples) {
if (Input[0] == ' ')
return false;
size_t n2 = Input.rfind(':');
return true;
}
+
+/// \brief Returns true if line offset \p L is legal (only has 16 bits).
+static bool isOffsetLegal(unsigned L) {
+ return (L & 0xffff) == L;
+}
+
/// \brief Parse \p Input as line sample.
///
/// \param Input input line.
/// \param TargetCountMap map from indirect call target to count.
///
/// returns true if parsing is successful.
-static bool ParseLine(const StringRef &Input, bool &IsCallsite, unsigned &Depth,
- unsigned &NumSamples, unsigned &LineOffset,
- unsigned &Discriminator, StringRef &CalleeName,
- DenseMap<StringRef, unsigned> &TargetCountMap) {
+static bool ParseLine(const StringRef &Input, bool &IsCallsite, uint32_t &Depth,
+ uint64_t &NumSamples, uint32_t &LineOffset,
+ uint32_t &Discriminator, StringRef &CalleeName,
+ DenseMap<StringRef, uint64_t> &TargetCountMap) {
for (Depth = 0; Input[Depth] == ' '; Depth++)
;
if (Depth == 0)
StringRef Loc = Input.substr(Depth, n1 - Depth);
size_t n2 = Loc.find('.');
if (n2 == StringRef::npos) {
- if (Loc.getAsInteger(10, LineOffset))
+ if (Loc.getAsInteger(10, LineOffset) || !isOffsetLegal(LineOffset))
return false;
Discriminator = 0;
} else {
if (n3 != StringRef::npos) {
pair = Rest.substr(0, n3);
}
- int n4 = pair.find(':');
- unsigned count;
+ size_t n4 = pair.find(':');
+ uint64_t count;
if (pair.substr(n4 + 1).getAsInteger(10, count))
return false;
TargetCountMap[pair.substr(0, n4)] = count;
}
} else {
IsCallsite = true;
- int n3 = Rest.find_last_of(':');
+ size_t n3 = Rest.find_last_of(':');
CalleeName = Rest.substr(0, n3);
if (Rest.substr(n3 + 1).getAsInteger(10, NumSamples))
return false;
// The only requirement we place on the identifier, then, is that it
// should not begin with a number.
if ((*LineIt)[0] != ' ') {
- unsigned NumSamples, NumHeadSamples;
+ uint64_t NumSamples, NumHeadSamples;
StringRef FName;
if (!ParseHead(*LineIt, FName, NumSamples, NumHeadSamples)) {
reportError(LineIt.line_number(),
InlineStack.clear();
InlineStack.push_back(&FProfile);
} else {
- unsigned NumSamples;
+ uint64_t NumSamples;
StringRef FName;
- DenseMap<StringRef, unsigned> TargetCountMap;
+ DenseMap<StringRef, uint64_t> TargetCountMap;
bool IsCallsite;
- unsigned Depth, LineOffset, Discriminator;
+ uint32_t Depth, LineOffset, Discriminator;
if (!ParseLine(*LineIt, IsCallsite, Depth, NumSamples, LineOffset,
Discriminator, FName, TargetCountMap)) {
reportError(LineIt.line_number(),
return sampleprof_error::success;
}
+bool SampleProfileReaderText::hasFormat(const MemoryBuffer &Buffer) {
+ bool result = false;
+
+ // Check that the first non-comment line is a valid function header.
+ line_iterator LineIt(Buffer, /*SkipBlanks=*/true, '#');
+ if (!LineIt.is_at_eof()) {
+ if ((*LineIt)[0] != ' ') {
+ uint64_t NumSamples, NumHeadSamples;
+ StringRef FName;
+ result = ParseHead(*LineIt, FName, NumSamples, NumHeadSamples);
+ }
+ }
+
+ return result;
+}
+
template <typename T> ErrorOr<T> SampleProfileReaderBinary::readNumber() {
unsigned NumBytesRead = 0;
std::error_code EC;
return Str;
}
-std::error_code
-SampleProfileReaderBinary::readProfile(FunctionSamples &FProfile) {
- auto Val = readNumber<unsigned>();
- if (std::error_code EC = Val.getError())
+ErrorOr<StringRef> SampleProfileReaderBinary::readStringFromTable() {
+ std::error_code EC;
+ auto Idx = readNumber<uint32_t>();
+ if (std::error_code EC = Idx.getError())
return EC;
- FProfile.addTotalSamples(*Val);
+ if (*Idx >= NameTable.size())
+ return sampleprof_error::truncated_name_table;
+ return NameTable[*Idx];
+}
- Val = readNumber<unsigned>();
- if (std::error_code EC = Val.getError())
+std::error_code
+SampleProfileReaderBinary::readProfile(FunctionSamples &FProfile) {
+ auto NumSamples = readNumber<uint64_t>();
+ if (std::error_code EC = NumSamples.getError())
return EC;
- FProfile.addHeadSamples(*Val);
+ FProfile.addTotalSamples(*NumSamples);
// Read the samples in the body.
- auto NumRecords = readNumber<unsigned>();
+ auto NumRecords = readNumber<uint32_t>();
if (std::error_code EC = NumRecords.getError())
return EC;
- for (unsigned I = 0; I < *NumRecords; ++I) {
+ for (uint32_t I = 0; I < *NumRecords; ++I) {
auto LineOffset = readNumber<uint64_t>();
if (std::error_code EC = LineOffset.getError())
return EC;
+ if (!isOffsetLegal(*LineOffset)) {
+ return std::error_code();
+ }
+
auto Discriminator = readNumber<uint64_t>();
if (std::error_code EC = Discriminator.getError())
return EC;
if (std::error_code EC = NumSamples.getError())
return EC;
- auto NumCalls = readNumber<unsigned>();
+ auto NumCalls = readNumber<uint32_t>();
if (std::error_code EC = NumCalls.getError())
return EC;
- for (unsigned J = 0; J < *NumCalls; ++J) {
- auto CalledFunction(readString());
+ for (uint32_t J = 0; J < *NumCalls; ++J) {
+ auto CalledFunction(readStringFromTable());
if (std::error_code EC = CalledFunction.getError())
return EC;
}
// Read all the samples for inlined function calls.
- auto NumCallsites = readNumber<unsigned>();
+ auto NumCallsites = readNumber<uint32_t>();
if (std::error_code EC = NumCallsites.getError())
return EC;
- for (unsigned J = 0; J < *NumCallsites; ++J) {
+ for (uint32_t J = 0; J < *NumCallsites; ++J) {
auto LineOffset = readNumber<uint64_t>();
if (std::error_code EC = LineOffset.getError())
return EC;
if (std::error_code EC = Discriminator.getError())
return EC;
- auto FName(readString());
+ auto FName(readStringFromTable());
if (std::error_code EC = FName.getError())
return EC;
std::error_code SampleProfileReaderBinary::read() {
while (!at_eof()) {
- auto FName(readString());
+ auto NumHeadSamples = readNumber<uint64_t>();
+ if (std::error_code EC = NumHeadSamples.getError())
+ return EC;
+
+ auto FName(readStringFromTable());
if (std::error_code EC = FName.getError())
return EC;
Profiles[*FName] = FunctionSamples();
FunctionSamples &FProfile = Profiles[*FName];
+ FProfile.addHeadSamples(*NumHeadSamples);
+
if (std::error_code EC = readProfile(FProfile))
return EC;
}
else if (*Version != SPVersion())
return sampleprof_error::unsupported_version;
+ // Read the name table.
+ auto Size = readNumber<uint32_t>();
+ if (std::error_code EC = Size.getError())
+ return EC;
+ NameTable.reserve(*Size);
+ for (uint32_t I = 0; I < *Size; ++I) {
+ auto Name(readString());
+ if (std::error_code EC = Name.getError())
+ return EC;
+ NameTable.push_back(*Name);
+ }
+
return sampleprof_error::success;
}
return Magic == SPMagic();
}
-bool SourceInfo::operator<(const SourceInfo &P) const {
- if (Line != P.Line)
- return Line < P.Line;
- if (StartLine != P.StartLine)
- return StartLine < P.StartLine;
- if (Discriminator != P.Discriminator)
- return Discriminator < P.Discriminator;
- return FuncName < P.FuncName;
-}
-
std::error_code SampleProfileReaderGCC::skipNextWord() {
uint32_t dummy;
if (!GcovBuffer.readInt(dummy))
auto Buffer = std::move(BufferOrErr.get());
// Sanity check the file.
- if (Buffer->getBufferSize() > std::numeric_limits<unsigned>::max())
+ if (Buffer->getBufferSize() > std::numeric_limits<uint32_t>::max())
return sampleprof_error::too_large;
return std::move(Buffer);
Reader.reset(new SampleProfileReaderBinary(std::move(Buffer), C));
else if (SampleProfileReaderGCC::hasFormat(*Buffer))
Reader.reset(new SampleProfileReaderGCC(std::move(Buffer), C));
- else
+ else if (SampleProfileReaderText::hasFormat(*Buffer))
Reader.reset(new SampleProfileReaderText(std::move(Buffer), C));
+ else
+ return sampleprof_error::unrecognized_format;
if (std::error_code EC = Reader->readHeader())
return EC;
projects / oota-llvm.git / blobdiff