summaryrefslogtreecommitdiff
path: root/gnu/llvm/lib/XRay
diff options
context:
space:
mode:
authorPatrick Wildt <patrick@cvs.openbsd.org>2017-10-04 20:28:12 +0000
committerPatrick Wildt <patrick@cvs.openbsd.org>2017-10-04 20:28:12 +0000
commitce4942c67154008535160eec8b913fb008038b6b (patch)
treea80e1a407d28d61e22be5dd84b73c4cc51ade54a /gnu/llvm/lib/XRay
parent92440caccb2098ea0dd638dc607065155340d940 (diff)
Import LLVM 5.0.0 release including clang, lld and lldb.
Diffstat (limited to 'gnu/llvm/lib/XRay')
-rw-r--r--gnu/llvm/lib/XRay/CMakeLists.txt3
-rw-r--r--gnu/llvm/lib/XRay/InstrumentationMap.cpp198
-rw-r--r--gnu/llvm/lib/XRay/Trace.cpp403
3 files changed, 584 insertions, 20 deletions
diff --git a/gnu/llvm/lib/XRay/CMakeLists.txt b/gnu/llvm/lib/XRay/CMakeLists.txt
index 6c1acba79bf..8d558209d8e 100644
--- a/gnu/llvm/lib/XRay/CMakeLists.txt
+++ b/gnu/llvm/lib/XRay/CMakeLists.txt
@@ -1,4 +1,5 @@
add_llvm_library(LLVMXRay
+ InstrumentationMap.cpp
Trace.cpp
ADDITIONAL_HEADER_DIRS
@@ -7,7 +8,9 @@ add_llvm_library(LLVMXRay
DEPENDS
LLVMSupport
+ LLVMObject
LINK_LIBS
LLVMSupport
+ LLVMObject
)
diff --git a/gnu/llvm/lib/XRay/InstrumentationMap.cpp b/gnu/llvm/lib/XRay/InstrumentationMap.cpp
new file mode 100644
index 00000000000..d9ce255bc68
--- /dev/null
+++ b/gnu/llvm/lib/XRay/InstrumentationMap.cpp
@@ -0,0 +1,198 @@
+//===- InstrumentationMap.cpp - XRay Instrumentation Map ------------------===//
+//
+// The LLVM Compiler Infrastructure
+//
+// This file is distributed under the University of Illinois Open Source
+// License. See LICENSE.TXT for details.
+//
+//===----------------------------------------------------------------------===//
+//
+// Implementation of the InstrumentationMap type for XRay sleds.
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/XRay/InstrumentationMap.h"
+#include "llvm/ADT/None.h"
+#include "llvm/ADT/STLExtras.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/ADT/Triple.h"
+#include "llvm/ADT/Twine.h"
+#include "llvm/Object/Binary.h"
+#include "llvm/Object/ObjectFile.h"
+#include "llvm/Support/DataExtractor.h"
+#include "llvm/Support/Error.h"
+#include "llvm/Support/FileSystem.h"
+#include "llvm/Support/YAMLTraits.h"
+#include <algorithm>
+#include <cstddef>
+#include <cstdint>
+#include <system_error>
+#include <vector>
+
+using namespace llvm;
+using namespace xray;
+
+Optional<int32_t> InstrumentationMap::getFunctionId(uint64_t Addr) const {
+ auto I = FunctionIds.find(Addr);
+ if (I != FunctionIds.end())
+ return I->second;
+ return None;
+}
+
+Optional<uint64_t> InstrumentationMap::getFunctionAddr(int32_t FuncId) const {
+ auto I = FunctionAddresses.find(FuncId);
+ if (I != FunctionAddresses.end())
+ return I->second;
+ return None;
+}
+
+static Error
+loadELF64(StringRef Filename, object::OwningBinary<object::ObjectFile> &ObjFile,
+ InstrumentationMap::SledContainer &Sleds,
+ InstrumentationMap::FunctionAddressMap &FunctionAddresses,
+ InstrumentationMap::FunctionAddressReverseMap &FunctionIds) {
+ InstrumentationMap Map;
+
+ // Find the section named "xray_instr_map".
+ if (!ObjFile.getBinary()->isELF() ||
+ !(ObjFile.getBinary()->getArch() == Triple::x86_64 ||
+ ObjFile.getBinary()->getArch() == Triple::ppc64le))
+ return make_error<StringError>(
+ "File format not supported (only does ELF little endian 64-bit).",
+ std::make_error_code(std::errc::not_supported));
+
+ StringRef Contents = "";
+ const auto &Sections = ObjFile.getBinary()->sections();
+ auto I = llvm::find_if(Sections, [&](object::SectionRef Section) {
+ StringRef Name = "";
+ if (Section.getName(Name))
+ return false;
+ return Name == "xray_instr_map";
+ });
+
+ if (I == Sections.end())
+ return make_error<StringError>(
+ "Failed to find XRay instrumentation map.",
+ std::make_error_code(std::errc::executable_format_error));
+
+ if (I->getContents(Contents))
+ return errorCodeToError(
+ std::make_error_code(std::errc::executable_format_error));
+
+ // Copy the instrumentation map data into the Sleds data structure.
+ auto C = Contents.bytes_begin();
+ static constexpr size_t ELF64SledEntrySize = 32;
+
+ if ((C - Contents.bytes_end()) % ELF64SledEntrySize != 0)
+ return make_error<StringError>(
+ Twine("Instrumentation map entries not evenly divisible by size of "
+ "an XRay sled entry in ELF64."),
+ std::make_error_code(std::errc::executable_format_error));
+
+ int32_t FuncId = 1;
+ uint64_t CurFn = 0;
+ for (; C != Contents.bytes_end(); C += ELF64SledEntrySize) {
+ DataExtractor Extractor(
+ StringRef(reinterpret_cast<const char *>(C), ELF64SledEntrySize), true,
+ 8);
+ Sleds.push_back({});
+ auto &Entry = Sleds.back();
+ uint32_t OffsetPtr = 0;
+ Entry.Address = Extractor.getU64(&OffsetPtr);
+ Entry.Function = Extractor.getU64(&OffsetPtr);
+ auto Kind = Extractor.getU8(&OffsetPtr);
+ static constexpr SledEntry::FunctionKinds Kinds[] = {
+ SledEntry::FunctionKinds::ENTRY, SledEntry::FunctionKinds::EXIT,
+ SledEntry::FunctionKinds::TAIL,
+ };
+ if (Kind >= sizeof(Kinds))
+ return errorCodeToError(
+ std::make_error_code(std::errc::executable_format_error));
+ Entry.Kind = Kinds[Kind];
+ Entry.AlwaysInstrument = Extractor.getU8(&OffsetPtr) != 0;
+
+ // We do replicate the function id generation scheme implemented in the
+ // XRay runtime.
+ // FIXME: Figure out how to keep this consistent with the XRay runtime.
+ if (CurFn == 0) {
+ CurFn = Entry.Function;
+ FunctionAddresses[FuncId] = Entry.Function;
+ FunctionIds[Entry.Function] = FuncId;
+ }
+ if (Entry.Function != CurFn) {
+ ++FuncId;
+ CurFn = Entry.Function;
+ FunctionAddresses[FuncId] = Entry.Function;
+ FunctionIds[Entry.Function] = FuncId;
+ }
+ }
+ return Error::success();
+}
+
+static Error
+loadYAML(int Fd, size_t FileSize, StringRef Filename,
+ InstrumentationMap::SledContainer &Sleds,
+ InstrumentationMap::FunctionAddressMap &FunctionAddresses,
+ InstrumentationMap::FunctionAddressReverseMap &FunctionIds) {
+ std::error_code EC;
+ sys::fs::mapped_file_region MappedFile(
+ Fd, sys::fs::mapped_file_region::mapmode::readonly, FileSize, 0, EC);
+ if (EC)
+ return make_error<StringError>(
+ Twine("Failed memory-mapping file '") + Filename + "'.", EC);
+
+ std::vector<YAMLXRaySledEntry> YAMLSleds;
+ yaml::Input In(StringRef(MappedFile.data(), MappedFile.size()));
+ In >> YAMLSleds;
+ if (In.error())
+ return make_error<StringError>(
+ Twine("Failed loading YAML document from '") + Filename + "'.",
+ In.error());
+
+ Sleds.reserve(YAMLSleds.size());
+ for (const auto &Y : YAMLSleds) {
+ FunctionAddresses[Y.FuncId] = Y.Function;
+ FunctionIds[Y.Function] = Y.FuncId;
+ Sleds.push_back(
+ SledEntry{Y.Address, Y.Function, Y.Kind, Y.AlwaysInstrument});
+ }
+ return Error::success();
+}
+
+// FIXME: Create error types that encapsulate a bit more information than what
+// StringError instances contain.
+Expected<InstrumentationMap>
+llvm::xray::loadInstrumentationMap(StringRef Filename) {
+ // At this point we assume the file is an object file -- and if that doesn't
+ // work, we treat it as YAML.
+ // FIXME: Extend to support non-ELF and non-x86_64 binaries.
+
+ InstrumentationMap Map;
+ auto ObjectFileOrError = object::ObjectFile::createObjectFile(Filename);
+ if (!ObjectFileOrError) {
+ auto E = ObjectFileOrError.takeError();
+ // We try to load it as YAML if the ELF load didn't work.
+ int Fd;
+ if (sys::fs::openFileForRead(Filename, Fd))
+ return std::move(E);
+
+ uint64_t FileSize;
+ if (sys::fs::file_size(Filename, FileSize))
+ return std::move(E);
+
+ // If the file is empty, we return the original error.
+ if (FileSize == 0)
+ return std::move(E);
+
+ // From this point on the errors will be only for the YAML parts, so we
+ // consume the errors at this point.
+ consumeError(std::move(E));
+ if (auto E = loadYAML(Fd, FileSize, Filename, Map.Sleds,
+ Map.FunctionAddresses, Map.FunctionIds))
+ return std::move(E);
+ } else if (auto E = loadELF64(Filename, *ObjectFileOrError, Map.Sleds,
+ Map.FunctionAddresses, Map.FunctionIds)) {
+ return std::move(E);
+ }
+ return Map;
+}
diff --git a/gnu/llvm/lib/XRay/Trace.cpp b/gnu/llvm/lib/XRay/Trace.cpp
index 51000c777de..6677063f944 100644
--- a/gnu/llvm/lib/XRay/Trace.cpp
+++ b/gnu/llvm/lib/XRay/Trace.cpp
@@ -24,8 +24,8 @@ using llvm::yaml::Input;
using XRayRecordStorage =
std::aligned_storage<sizeof(XRayRecord), alignof(XRayRecord)>::type;
-Error NaiveLogLoader(StringRef Data, XRayFileHeader &FileHeader,
- std::vector<XRayRecord> &Records) {
+// Populates the FileHeader reference by reading the first 32 bytes of the file.
+Error readBinaryFormatHeader(StringRef Data, XRayFileHeader &FileHeader) {
// FIXME: Maybe deduce whether the data is little or big-endian using some
// magic bytes in the beginning of the file?
@@ -37,16 +37,6 @@ Error NaiveLogLoader(StringRef Data, XRayFileHeader &FileHeader,
// (4) uint32 : bitfield
// (8) uint64 : cycle frequency
// (16) - : padding
- //
- if (Data.size() < 32)
- return make_error<StringError>(
- "Not enough bytes for an XRay log.",
- std::make_error_code(std::errc::invalid_argument));
-
- if (Data.size() - 32 == 0 || Data.size() % 32 != 0)
- return make_error<StringError>(
- "Invalid-sized XRay data.",
- std::make_error_code(std::errc::invalid_argument));
DataExtractor HeaderExtractor(Data, true, 8);
uint32_t OffsetPtr = 0;
@@ -56,11 +46,29 @@ Error NaiveLogLoader(StringRef Data, XRayFileHeader &FileHeader,
FileHeader.ConstantTSC = Bitfield & 1uL;
FileHeader.NonstopTSC = Bitfield & 1uL << 1;
FileHeader.CycleFrequency = HeaderExtractor.getU64(&OffsetPtr);
-
+ std::memcpy(&FileHeader.FreeFormData, Data.bytes_begin() + OffsetPtr, 16);
if (FileHeader.Version != 1)
return make_error<StringError>(
Twine("Unsupported XRay file version: ") + Twine(FileHeader.Version),
std::make_error_code(std::errc::invalid_argument));
+ return Error::success();
+}
+
+Error loadNaiveFormatLog(StringRef Data, XRayFileHeader &FileHeader,
+ std::vector<XRayRecord> &Records) {
+ // Check that there is at least a header
+ if (Data.size() < 32)
+ return make_error<StringError>(
+ "Not enough bytes for an XRay log.",
+ std::make_error_code(std::errc::invalid_argument));
+
+ if (Data.size() - 32 == 0 || Data.size() % 32 != 0)
+ return make_error<StringError>(
+ "Invalid-sized XRay data.",
+ std::make_error_code(std::errc::invalid_argument));
+
+ if (auto E = readBinaryFormatHeader(Data, FileHeader))
+ return E;
// Each record after the header will be 32 bytes, in the following format:
//
@@ -98,9 +106,357 @@ Error NaiveLogLoader(StringRef Data, XRayFileHeader &FileHeader,
return Error::success();
}
-Error YAMLLogLoader(StringRef Data, XRayFileHeader &FileHeader,
- std::vector<XRayRecord> &Records) {
+/// When reading from a Flight Data Recorder mode log, metadata records are
+/// sparse compared to packed function records, so we must maintain state as we
+/// read through the sequence of entries. This allows the reader to denormalize
+/// the CPUId and Thread Id onto each Function Record and transform delta
+/// encoded TSC values into absolute encodings on each record.
+struct FDRState {
+ uint16_t CPUId;
+ uint16_t ThreadId;
+ uint64_t BaseTSC;
+
+ /// Encode some of the state transitions for the FDR log reader as explicit
+ /// checks. These are expectations for the next Record in the stream.
+ enum class Token {
+ NEW_BUFFER_RECORD_OR_EOF,
+ WALLCLOCK_RECORD,
+ NEW_CPU_ID_RECORD,
+ FUNCTION_SEQUENCE,
+ SCAN_TO_END_OF_THREAD_BUF,
+ CUSTOM_EVENT_DATA,
+ };
+ Token Expects;
+
+ // Each threads buffer may have trailing garbage to scan over, so we track our
+ // progress.
+ uint64_t CurrentBufferSize;
+ uint64_t CurrentBufferConsumed;
+};
+
+Twine fdrStateToTwine(const FDRState::Token &state) {
+ switch (state) {
+ case FDRState::Token::NEW_BUFFER_RECORD_OR_EOF:
+ return "NEW_BUFFER_RECORD_OR_EOF";
+ case FDRState::Token::WALLCLOCK_RECORD:
+ return "WALLCLOCK_RECORD";
+ case FDRState::Token::NEW_CPU_ID_RECORD:
+ return "NEW_CPU_ID_RECORD";
+ case FDRState::Token::FUNCTION_SEQUENCE:
+ return "FUNCTION_SEQUENCE";
+ case FDRState::Token::SCAN_TO_END_OF_THREAD_BUF:
+ return "SCAN_TO_END_OF_THREAD_BUF";
+ case FDRState::Token::CUSTOM_EVENT_DATA:
+ return "CUSTOM_EVENT_DATA";
+ }
+ return "UNKNOWN";
+}
+
+/// State transition when a NewBufferRecord is encountered.
+Error processFDRNewBufferRecord(FDRState &State, uint8_t RecordFirstByte,
+ DataExtractor &RecordExtractor) {
+
+ if (State.Expects != FDRState::Token::NEW_BUFFER_RECORD_OR_EOF)
+ return make_error<StringError>(
+ "Malformed log. Read New Buffer record kind out of sequence",
+ std::make_error_code(std::errc::executable_format_error));
+ uint32_t OffsetPtr = 1; // 1 byte into record.
+ State.ThreadId = RecordExtractor.getU16(&OffsetPtr);
+ State.Expects = FDRState::Token::WALLCLOCK_RECORD;
+ return Error::success();
+}
+
+/// State transition when an EndOfBufferRecord is encountered.
+Error processFDREndOfBufferRecord(FDRState &State, uint8_t RecordFirstByte,
+ DataExtractor &RecordExtractor) {
+ if (State.Expects == FDRState::Token::NEW_BUFFER_RECORD_OR_EOF)
+ return make_error<StringError>(
+ "Malformed log. Received EOB message without current buffer.",
+ std::make_error_code(std::errc::executable_format_error));
+ State.Expects = FDRState::Token::SCAN_TO_END_OF_THREAD_BUF;
+ return Error::success();
+}
+
+/// State transition when a NewCPUIdRecord is encountered.
+Error processFDRNewCPUIdRecord(FDRState &State, uint8_t RecordFirstByte,
+ DataExtractor &RecordExtractor) {
+ if (State.Expects != FDRState::Token::FUNCTION_SEQUENCE &&
+ State.Expects != FDRState::Token::NEW_CPU_ID_RECORD)
+ return make_error<StringError>(
+ "Malformed log. Read NewCPUId record kind out of sequence",
+ std::make_error_code(std::errc::executable_format_error));
+ uint32_t OffsetPtr = 1; // Read starting after the first byte.
+ State.CPUId = RecordExtractor.getU16(&OffsetPtr);
+ State.BaseTSC = RecordExtractor.getU64(&OffsetPtr);
+ State.Expects = FDRState::Token::FUNCTION_SEQUENCE;
+ return Error::success();
+}
+
+/// State transition when a TSCWrapRecord (overflow detection) is encountered.
+Error processFDRTSCWrapRecord(FDRState &State, uint8_t RecordFirstByte,
+ DataExtractor &RecordExtractor) {
+ if (State.Expects != FDRState::Token::FUNCTION_SEQUENCE)
+ return make_error<StringError>(
+ "Malformed log. Read TSCWrap record kind out of sequence",
+ std::make_error_code(std::errc::executable_format_error));
+ uint32_t OffsetPtr = 1; // Read starting after the first byte.
+ State.BaseTSC = RecordExtractor.getU64(&OffsetPtr);
+ return Error::success();
+}
+
+/// State transition when a WallTimeMarkerRecord is encountered.
+Error processFDRWallTimeRecord(FDRState &State, uint8_t RecordFirstByte,
+ DataExtractor &RecordExtractor) {
+ if (State.Expects != FDRState::Token::WALLCLOCK_RECORD)
+ return make_error<StringError>(
+ "Malformed log. Read Wallclock record kind out of sequence",
+ std::make_error_code(std::errc::executable_format_error));
+ // We don't encode the wall time into any of the records.
+ // XRayRecords are concerned with the TSC instead.
+ State.Expects = FDRState::Token::NEW_CPU_ID_RECORD;
+ return Error::success();
+}
+
+/// State transition when a CustomEventMarker is encountered.
+Error processCustomEventMarker(FDRState &State, uint8_t RecordFirstByte,
+ DataExtractor &RecordExtractor,
+ size_t &RecordSize) {
+ // We can encounter a CustomEventMarker anywhere in the log, so we can handle
+ // it regardless of the expectation. However, we do se the expectation to read
+ // a set number of fixed bytes, as described in the metadata.
+ uint32_t OffsetPtr = 1; // Read after the first byte.
+ uint32_t DataSize = RecordExtractor.getU32(&OffsetPtr);
+ uint64_t TSC = RecordExtractor.getU64(&OffsetPtr);
+
+ // FIXME: Actually represent the record through the API. For now we only skip
+ // through the data.
+ (void)TSC;
+ RecordSize = 16 + DataSize;
+ return Error::success();
+}
+
+/// Advances the state machine for reading the FDR record type by reading one
+/// Metadata Record and updating the State appropriately based on the kind of
+/// record encountered. The RecordKind is encoded in the first byte of the
+/// Record, which the caller should pass in because they have already read it
+/// to determine that this is a metadata record as opposed to a function record.
+Error processFDRMetadataRecord(FDRState &State, uint8_t RecordFirstByte,
+ DataExtractor &RecordExtractor,
+ size_t &RecordSize) {
+ // The remaining 7 bits are the RecordKind enum.
+ uint8_t RecordKind = RecordFirstByte >> 1;
+ switch (RecordKind) {
+ case 0: // NewBuffer
+ if (auto E =
+ processFDRNewBufferRecord(State, RecordFirstByte, RecordExtractor))
+ return E;
+ break;
+ case 1: // EndOfBuffer
+ if (auto E = processFDREndOfBufferRecord(State, RecordFirstByte,
+ RecordExtractor))
+ return E;
+ break;
+ case 2: // NewCPUId
+ if (auto E =
+ processFDRNewCPUIdRecord(State, RecordFirstByte, RecordExtractor))
+ return E;
+ break;
+ case 3: // TSCWrap
+ if (auto E =
+ processFDRTSCWrapRecord(State, RecordFirstByte, RecordExtractor))
+ return E;
+ break;
+ case 4: // WallTimeMarker
+ if (auto E =
+ processFDRWallTimeRecord(State, RecordFirstByte, RecordExtractor))
+ return E;
+ break;
+ case 5: // CustomEventMarker
+ if (auto E = processCustomEventMarker(State, RecordFirstByte,
+ RecordExtractor, RecordSize))
+ return E;
+ break;
+ default:
+ // Widen the record type to uint16_t to prevent conversion to char.
+ return make_error<StringError>(
+ Twine("Illegal metadata record type: ")
+ .concat(Twine(static_cast<unsigned>(RecordKind))),
+ std::make_error_code(std::errc::executable_format_error));
+ }
+ return Error::success();
+}
+
+/// Reads a function record from an FDR format log, appending a new XRayRecord
+/// to the vector being populated and updating the State with a new value
+/// reference value to interpret TSC deltas.
+///
+/// The XRayRecord constructed includes information from the function record
+/// processed here as well as Thread ID and CPU ID formerly extracted into
+/// State.
+Error processFDRFunctionRecord(FDRState &State, uint8_t RecordFirstByte,
+ DataExtractor &RecordExtractor,
+ std::vector<XRayRecord> &Records) {
+ switch (State.Expects) {
+ case FDRState::Token::NEW_BUFFER_RECORD_OR_EOF:
+ return make_error<StringError>(
+ "Malformed log. Received Function Record before new buffer setup.",
+ std::make_error_code(std::errc::executable_format_error));
+ case FDRState::Token::WALLCLOCK_RECORD:
+ return make_error<StringError>(
+ "Malformed log. Received Function Record when expecting wallclock.",
+ std::make_error_code(std::errc::executable_format_error));
+ case FDRState::Token::NEW_CPU_ID_RECORD:
+ return make_error<StringError>(
+ "Malformed log. Received Function Record before first CPU record.",
+ std::make_error_code(std::errc::executable_format_error));
+ default:
+ Records.emplace_back();
+ auto &Record = Records.back();
+ Record.RecordType = 0; // Record is type NORMAL.
+ // Strip off record type bit and use the next three bits.
+ uint8_t RecordType = (RecordFirstByte >> 1) & 0x07;
+ switch (RecordType) {
+ case static_cast<uint8_t>(RecordTypes::ENTER):
+ Record.Type = RecordTypes::ENTER;
+ break;
+ case static_cast<uint8_t>(RecordTypes::EXIT):
+ case 2: // TAIL_EXIT is not yet defined in RecordTypes.
+ Record.Type = RecordTypes::EXIT;
+ break;
+ default:
+ // When initializing the error, convert to uint16_t so that the record
+ // type isn't interpreted as a char.
+ return make_error<StringError>(
+ Twine("Illegal function record type: ")
+ .concat(Twine(static_cast<unsigned>(RecordType))),
+ std::make_error_code(std::errc::executable_format_error));
+ }
+ Record.CPU = State.CPUId;
+ Record.TId = State.ThreadId;
+ // Back up to read first 32 bits, including the 8 we pulled RecordType
+ // and RecordKind out of. The remaining 28 are FunctionId.
+ uint32_t OffsetPtr = 0;
+ // Despite function Id being a signed int on XRayRecord,
+ // when it is written to an FDR format, the top bits are truncated,
+ // so it is effectively an unsigned value. When we shift off the
+ // top four bits, we want the shift to be logical, so we read as
+ // uint32_t.
+ uint32_t FuncIdBitField = RecordExtractor.getU32(&OffsetPtr);
+ Record.FuncId = FuncIdBitField >> 4;
+ // FunctionRecords have a 32 bit delta from the previous absolute TSC
+ // or TSC delta. If this would overflow, we should read a TSCWrap record
+ // with an absolute TSC reading.
+ uint64_t new_tsc = State.BaseTSC + RecordExtractor.getU32(&OffsetPtr);
+ State.BaseTSC = new_tsc;
+ Record.TSC = new_tsc;
+ }
+ return Error::success();
+}
+
+/// Reads a log in FDR mode for version 1 of this binary format. FDR mode is
+/// defined as part of the compiler-rt project in xray_fdr_logging.h, and such
+/// a log consists of the familiar 32 bit XRayHeader, followed by sequences of
+/// of interspersed 16 byte Metadata Records and 8 byte Function Records.
+///
+/// The following is an attempt to document the grammar of the format, which is
+/// parsed by this function for little-endian machines. Since the format makes
+/// use of BitFields, when we support big-Endian architectures, we will need to
+/// adjust not only the endianness parameter to llvm's RecordExtractor, but also
+/// the bit twiddling logic, which is consistent with the little-endian
+/// convention that BitFields within a struct will first be packed into the
+/// least significant bits the address they belong to.
+///
+/// We expect a format complying with the grammar in the following pseudo-EBNF.
+///
+/// FDRLog: XRayFileHeader ThreadBuffer*
+/// XRayFileHeader: 32 bits to identify the log as FDR with machine metadata.
+/// ThreadBuffer: BufSize NewBuffer WallClockTime NewCPUId FunctionSequence EOB
+/// BufSize: 8 byte unsigned integer indicating how large the buffer is.
+/// NewBuffer: 16 byte metadata record with Thread Id.
+/// WallClockTime: 16 byte metadata record with human readable time.
+/// NewCPUId: 16 byte metadata record with CPUId and a 64 bit TSC reading.
+/// EOB: 16 byte record in a thread buffer plus mem garbage to fill BufSize.
+/// FunctionSequence: NewCPUId | TSCWrap | FunctionRecord
+/// TSCWrap: 16 byte metadata record with a full 64 bit TSC reading.
+/// FunctionRecord: 8 byte record with FunctionId, entry/exit, and TSC delta.
+Error loadFDRLog(StringRef Data, XRayFileHeader &FileHeader,
+ std::vector<XRayRecord> &Records) {
+ if (Data.size() < 32)
+ return make_error<StringError>(
+ "Not enough bytes for an XRay log.",
+ std::make_error_code(std::errc::invalid_argument));
+
+ // For an FDR log, there are records sized 16 and 8 bytes.
+ // There actually may be no records if no non-trivial functions are
+ // instrumented.
+ if (Data.size() % 8 != 0)
+ return make_error<StringError>(
+ "Invalid-sized XRay data.",
+ std::make_error_code(std::errc::invalid_argument));
+
+ if (auto E = readBinaryFormatHeader(Data, FileHeader))
+ return E;
+
+ uint64_t BufferSize = 0;
+ {
+ StringRef ExtraDataRef(FileHeader.FreeFormData, 16);
+ DataExtractor ExtraDataExtractor(ExtraDataRef, true, 8);
+ uint32_t ExtraDataOffset = 0;
+ BufferSize = ExtraDataExtractor.getU64(&ExtraDataOffset);
+ }
+ FDRState State{0, 0, 0, FDRState::Token::NEW_BUFFER_RECORD_OR_EOF,
+ BufferSize, 0};
+ // RecordSize will tell the loop how far to seek ahead based on the record
+ // type that we have just read.
+ size_t RecordSize = 0;
+ for (auto S = Data.drop_front(32); !S.empty(); S = S.drop_front(RecordSize)) {
+ DataExtractor RecordExtractor(S, true, 8);
+ uint32_t OffsetPtr = 0;
+ if (State.Expects == FDRState::Token::SCAN_TO_END_OF_THREAD_BUF) {
+ RecordSize = State.CurrentBufferSize - State.CurrentBufferConsumed;
+ if (S.size() < State.CurrentBufferSize - State.CurrentBufferConsumed) {
+ return make_error<StringError>(
+ Twine("Incomplete thread buffer. Expected ") +
+ Twine(State.CurrentBufferSize - State.CurrentBufferConsumed) +
+ " remaining bytes but found " + Twine(S.size()),
+ make_error_code(std::errc::invalid_argument));
+ }
+ State.CurrentBufferConsumed = 0;
+ State.Expects = FDRState::Token::NEW_BUFFER_RECORD_OR_EOF;
+ continue;
+ }
+ uint8_t BitField = RecordExtractor.getU8(&OffsetPtr);
+ bool isMetadataRecord = BitField & 0x01uL;
+ if (isMetadataRecord) {
+ RecordSize = 16;
+ if (auto E = processFDRMetadataRecord(State, BitField, RecordExtractor,
+ RecordSize))
+ return E;
+ State.CurrentBufferConsumed += RecordSize;
+ } else { // Process Function Record
+ RecordSize = 8;
+ if (auto E = processFDRFunctionRecord(State, BitField, RecordExtractor,
+ Records))
+ return E;
+ State.CurrentBufferConsumed += RecordSize;
+ }
+ }
+ // There are two conditions
+ if (State.Expects != FDRState::Token::NEW_BUFFER_RECORD_OR_EOF &&
+ !(State.Expects == FDRState::Token::SCAN_TO_END_OF_THREAD_BUF &&
+ State.CurrentBufferSize == State.CurrentBufferConsumed))
+ return make_error<StringError>(
+ Twine("Encountered EOF with unexpected state expectation ") +
+ fdrStateToTwine(State.Expects) +
+ ". Remaining expected bytes in thread buffer total " +
+ Twine(State.CurrentBufferSize - State.CurrentBufferConsumed),
+ std::make_error_code(std::errc::executable_format_error));
+
+ return Error::success();
+}
+Error loadYAMLLog(StringRef Data, XRayFileHeader &FileHeader,
+ std::vector<XRayRecord> &Records) {
// Load the documents from the MappedFile.
YAMLXRayTrace Trace;
Input In(Data);
@@ -175,14 +531,21 @@ Expected<Trace> llvm::xray::loadTraceFile(StringRef Filename, bool Sort) {
uint16_t Version = HeaderExtractor.getU16(&OffsetPtr);
uint16_t Type = HeaderExtractor.getU16(&OffsetPtr);
+ enum BinaryFormatType { NAIVE_FORMAT = 0, FLIGHT_DATA_RECORDER_FORMAT = 1 };
+
Trace T;
- if (Version == 1 && (Type == 0 || Type == 1)) {
- if (auto E = NaiveLogLoader(StringRef(MappedFile.data(), MappedFile.size()),
- T.FileHeader, T.Records))
+ if (Version == 1 && Type == NAIVE_FORMAT) {
+ if (auto E =
+ loadNaiveFormatLog(StringRef(MappedFile.data(), MappedFile.size()),
+ T.FileHeader, T.Records))
+ return std::move(E);
+ } else if (Version == 1 && Type == FLIGHT_DATA_RECORDER_FORMAT) {
+ if (auto E = loadFDRLog(StringRef(MappedFile.data(), MappedFile.size()),
+ T.FileHeader, T.Records))
return std::move(E);
} else {
- if (auto E = YAMLLogLoader(StringRef(MappedFile.data(), MappedFile.size()),
- T.FileHeader, T.Records))
+ if (auto E = loadYAMLLog(StringRef(MappedFile.data(), MappedFile.size()),
+ T.FileHeader, T.Records))
return std::move(E);
}