diff options
author | mingmingl <mingmingl@google.com> | 2024-02-07 15:12:36 -0800 |
---|---|---|
committer | mingmingl <mingmingl@google.com> | 2024-02-07 15:12:36 -0800 |
commit | a746c3c5ad1ddd15cf167922d990af6af375f715 (patch) | |
tree | 655ef7f55250d73926552bed140d78677242e517 | |
parent | a7bc9cb6ffa91ff0ebabc45c0c7263c7c2c3a4de (diff) | |
parent | 48adcf1a142de6abeeb16596c5087fe83e8f422b (diff) | |
download | llvm-origin/users/minglotus-6/spr/instrprof.zip llvm-origin/users/minglotus-6/spr/instrprof.tar.gz llvm-origin/users/minglotus-6/spr/instrprof.tar.bz2 |
[𝘀𝗽𝗿] initial versionorigin/users/minglotus-6/spr/instrprof
Created using spr 1.3.4
49 files changed, 2215 insertions, 330 deletions
diff --git a/compiler-rt/include/profile/InstrProfData.inc b/compiler-rt/include/profile/InstrProfData.inc index 25df899..f0bc2d9 100644 --- a/compiler-rt/include/profile/InstrProfData.inc +++ b/compiler-rt/include/profile/InstrProfData.inc @@ -94,6 +94,26 @@ INSTR_PROF_DATA(const uint32_t, llvm::Type::getInt32Ty(Ctx), NumBitmapBytes, \ #undef INSTR_PROF_DATA /* INSTR_PROF_DATA end. */ +/* For a virtual table object, record the name hash to associate profiled + * addresses with global variables, and record {starting address, size in bytes} + * to map the profiled virtual table (which usually have an offset from the + * starting address) back to a virtual table object. */ +#ifndef INSTR_PROF_VTABLE_DATA +#define INSTR_PROF_VTABLE_DATA(Type, LLVMType, Name, Initializer) +#else +#define INSTR_PROF_VTABLE_DATA_DEFINED +#endif +INSTR_PROF_VTABLE_DATA( + const uint64_t, llvm::Type::getInt64Ty(Ctx), VTableNameHash, + ConstantInt::get(llvm::Type::getInt64Ty(Ctx), + IndexedInstrProf::ComputeHash(PGOVTableName))) +INSTR_PROF_VTABLE_DATA(const IntPtrT, llvm::Type::getInt8PtrTy(Ctx), + VTablePointer, VTableAddr) +INSTR_PROF_VTABLE_DATA(const uint32_t, llvm::Type::getInt32Ty(Ctx), VTableSize, + ConstantInt::get(llvm::Type::getInt32Ty(Ctx), + VTableSizeVal)) +#undef INSTR_PROF_VTABLE_DATA +/* INSTR_PROF_VTABLE_DATA end. */ /* This is an internal data structure used by value profiler. It * is defined here to allow serialization code sharing by LLVM @@ -145,6 +165,8 @@ INSTR_PROF_RAW_HEADER(uint64_t, CountersDelta, INSTR_PROF_RAW_HEADER(uint64_t, BitmapDelta, (uintptr_t)BitmapBegin - (uintptr_t)DataBegin) INSTR_PROF_RAW_HEADER(uint64_t, NamesDelta, (uintptr_t)NamesBegin) +INSTR_PROF_RAW_HEADER(uint64_t, VNamesSize, VNamesSize) +INSTR_PROF_RAW_HEADER(uint64_t, NumVTables, NumVTables) INSTR_PROF_RAW_HEADER(uint64_t, ValueKindLast, IPVK_Last) #undef INSTR_PROF_RAW_HEADER /* INSTR_PROF_RAW_HEADER end */ @@ -186,13 +208,28 @@ VALUE_PROF_FUNC_PARAM(uint32_t, CounterIndex, Type::getInt32Ty(Ctx)) VALUE_PROF_KIND(IPVK_IndirectCallTarget, 0, "indirect call target") /* For memory intrinsic functions size profiling. */ VALUE_PROF_KIND(IPVK_MemOPSize, 1, "memory intrinsic functions size") +/* For virtual table address profiling, the addresses of the virtual table + * (i.e., the address contained in objects pointing to a virtual table) are + * profiled. Note this may not be the address of the per C++ class virtual table + * object (i.e., there is an offset). + * + * The profiled addresses are stored in raw profile, together with the following + * two types of information. + * 1. The (beginning and ending) addresses of per C++ class virtual table objects. + * 2. The (compressed) virtual table object names. + * RawInstrProfReader converts profiled virtual table addresses to virtual table + * objects' MD5 hash. + */ +VALUE_PROF_KIND(IPVK_VTableTarget, 2, "The address of the compatible vtable (i.e., " + "there is an offset from this address to per C++ " + "class virtual table global variable.)") /* These two kinds must be the last to be * declared. This is to make sure the string * array created with the template can be * indexed with the kind value. */ VALUE_PROF_KIND(IPVK_First, IPVK_IndirectCallTarget, "first") -VALUE_PROF_KIND(IPVK_Last, IPVK_MemOPSize, "last") +VALUE_PROF_KIND(IPVK_Last, IPVK_VTableTarget, "last") #undef VALUE_PROF_KIND /* VALUE_PROF_KIND end */ @@ -267,9 +304,9 @@ COVMAP_HEADER(uint32_t, Int32Ty, Version, \ #undef COVMAP_HEADER /* COVMAP_HEADER end. */ - #ifdef INSTR_PROF_SECT_ENTRY #define INSTR_PROF_DATA_DEFINED + INSTR_PROF_SECT_ENTRY(IPSK_data, \ INSTR_PROF_QUOTE(INSTR_PROF_DATA_COMMON), \ INSTR_PROF_DATA_COFF, "__DATA,") @@ -282,12 +319,18 @@ INSTR_PROF_SECT_ENTRY(IPSK_bitmap, \ INSTR_PROF_SECT_ENTRY(IPSK_name, \ INSTR_PROF_QUOTE(INSTR_PROF_NAME_COMMON), \ INSTR_PROF_NAME_COFF, "__DATA,") +INSTR_PROF_SECT_ENTRY(IPSK_vname, \ + INSTR_PROF_QUOTE(INSTR_PROF_VNAME_COMMON), \ + INSTR_PROF_VNAME_COFF, "__DATA,") INSTR_PROF_SECT_ENTRY(IPSK_vals, \ INSTR_PROF_QUOTE(INSTR_PROF_VALS_COMMON), \ INSTR_PROF_VALS_COFF, "__DATA,") INSTR_PROF_SECT_ENTRY(IPSK_vnodes, \ INSTR_PROF_QUOTE(INSTR_PROF_VNODES_COMMON), \ INSTR_PROF_VNODES_COFF, "__DATA,") +INSTR_PROF_SECT_ENTRY(IPSK_vtab, \ + INSTR_PROF_QUOTE(INSTR_PROF_VTAB_COMMON), \ + INSTR_PROF_VTAB_COFF, "__DATA,") INSTR_PROF_SECT_ENTRY(IPSK_covmap, \ INSTR_PROF_QUOTE(INSTR_PROF_COVMAP_COMMON), \ INSTR_PROF_COVMAP_COFF, "__LLVM_COV,") @@ -307,7 +350,6 @@ INSTR_PROF_SECT_ENTRY(IPSK_covname, \ #undef INSTR_PROF_SECT_ENTRY #endif - #ifdef INSTR_PROF_VALUE_PROF_DATA #define INSTR_PROF_DATA_DEFINED @@ -479,7 +521,6 @@ getValueProfRecordHeaderSize(uint32_t NumValueSites); #undef INSTR_PROF_VALUE_PROF_DATA #endif /* INSTR_PROF_VALUE_PROF_DATA */ - #ifdef INSTR_PROF_COMMON_API_IMPL #define INSTR_PROF_DATA_DEFINED #ifdef __cplusplus @@ -663,9 +704,9 @@ serializeValueProfDataFrom(ValueProfRecordClosure *Closure, (uint64_t)'f' << 16 | (uint64_t)'R' << 8 | (uint64_t)129 /* Raw profile format version (start from 1). */ -#define INSTR_PROF_RAW_VERSION 9 +#define INSTR_PROF_RAW_VERSION 10 /* Indexed profile format version (start from 1). */ -#define INSTR_PROF_INDEX_VERSION 11 +#define INSTR_PROF_INDEX_VERSION 12 /* Coverage mapping format version (start from 0). */ #define INSTR_PROF_COVMAP_VERSION 6 @@ -703,10 +744,12 @@ serializeValueProfDataFrom(ValueProfRecordClosure *Closure, than WIN32 */ #define INSTR_PROF_DATA_COMMON __llvm_prf_data #define INSTR_PROF_NAME_COMMON __llvm_prf_names +#define INSTR_PROF_VNAME_COMMON __llvm_prf_vtabnames #define INSTR_PROF_CNTS_COMMON __llvm_prf_cnts #define INSTR_PROF_BITS_COMMON __llvm_prf_bits #define INSTR_PROF_VALS_COMMON __llvm_prf_vals #define INSTR_PROF_VNODES_COMMON __llvm_prf_vnds +#define INSTR_PROF_VTAB_COMMON __llvm_prf_vtab #define INSTR_PROF_COVMAP_COMMON __llvm_covmap #define INSTR_PROF_COVFUN_COMMON __llvm_covfun #define INSTR_PROF_COVDATA_COMMON __llvm_covdata @@ -717,10 +760,12 @@ serializeValueProfDataFrom(ValueProfRecordClosure *Closure, */ #define INSTR_PROF_DATA_COFF ".lprfd$M" #define INSTR_PROF_NAME_COFF ".lprfn$M" +#define INSTR_PROF_VNAME_COFF ".lprfn$M" #define INSTR_PROF_CNTS_COFF ".lprfc$M" #define INSTR_PROF_BITS_COFF ".lprfb$M" #define INSTR_PROF_VALS_COFF ".lprfv$M" #define INSTR_PROF_VNODES_COFF ".lprfnd$M" +#define INSTR_PROF_VTAB_COFF ".lprfvt$M" #define INSTR_PROF_COVMAP_COFF ".lcovmap$M" #define INSTR_PROF_COVFUN_COFF ".lcovfun$M" /* Since cov data and cov names sections are not allocated, we don't need to @@ -938,3 +983,4 @@ InstrProfIsSingleValRange(uint64_t Value) { } #endif /* INSTR_PROF_VALUE_PROF_MEMOP_API */ + diff --git a/compiler-rt/lib/profile/InstrProfiling.h b/compiler-rt/lib/profile/InstrProfiling.h index 0123908..9e6306a 100644 --- a/compiler-rt/lib/profile/InstrProfiling.h +++ b/compiler-rt/lib/profile/InstrProfiling.h @@ -49,6 +49,12 @@ typedef struct ValueProfNode { #include "profile/InstrProfData.inc" } ValueProfNode; +typedef void *IntPtrT; +typedef struct VTableProfData { +#define INSTR_PROF_VTABLE_DATA(Type, LLVMType, Name, Initializer) Type Name; +#include "profile/InstrProfData.inc" +} VTableProfData; + /*! * \brief Return 1 if profile counters are continuously synced to the raw * profile via an mmap(). This is in contrast to the default mode, in which @@ -103,12 +109,16 @@ const __llvm_profile_data *__llvm_profile_begin_data(void); const __llvm_profile_data *__llvm_profile_end_data(void); const char *__llvm_profile_begin_names(void); const char *__llvm_profile_end_names(void); +const char *__llvm_profile_begin_vtabnames(void); +const char *__llvm_profile_end_vtabnames(void); char *__llvm_profile_begin_counters(void); char *__llvm_profile_end_counters(void); char *__llvm_profile_begin_bitmap(void); char *__llvm_profile_end_bitmap(void); ValueProfNode *__llvm_profile_begin_vnodes(); ValueProfNode *__llvm_profile_end_vnodes(); +VTableProfData *__llvm_profile_begin_vtables(); +VTableProfData *__llvm_profile_end_vtables(); uint32_t *__llvm_profile_begin_orderfile(); /*! @@ -252,20 +262,31 @@ uint64_t __llvm_profile_get_num_bitmap_bytes(const char *Begin, /*! \brief Get the size of the profile name section in bytes. */ uint64_t __llvm_profile_get_name_size(const char *Begin, const char *End); -/* ! \brief Given the sizes of the data and counter information, return the - * number of padding bytes before and after the counters, and after the names, - * in the raw profile. +/*! \brief Get the number of virtual table profile data entries */ +uint64_t __llvm_profile_get_num_vtable(const VTableProfData *Begin, + const VTableProfData *End); + +/*! \brief Get the size of virtual table profile data in bytes. */ +uint64_t __llvm_profile_get_vtable_section_size(const VTableProfData *Begin, + const VTableProfData *End); + +/* ! \brief Given the sizes of the data and counter information, computes the + * number of padding bytes before and after the counter section, as well as the + * number of padding bytes after other setions in the raw profile. + * Returns -1 upon errors and 0 upon success. Output parameters should be used + * iff return value is 0. * * Note: When mmap() mode is disabled, no padding bytes before/after counters * are needed. However, in mmap() mode, the counter section in the raw profile * must be page-aligned: this API computes the number of padding bytes * needed to achieve that. */ -void __llvm_profile_get_padding_sizes_for_counters( +int __llvm_profile_get_padding_sizes_for_counters( uint64_t DataSize, uint64_t CountersSize, uint64_t NumBitmapBytes, - uint64_t NamesSize, uint64_t *PaddingBytesBeforeCounters, - uint64_t *PaddingBytesAfterCounters, uint64_t *PaddingBytesAfterBitmap, - uint64_t *PaddingBytesAfterNames); + uint64_t NamesSize, uint64_t VTableSize, uint64_t VNameSize, + uint64_t *PaddingBytesBeforeCounters, uint64_t *PaddingBytesAfterCounters, + uint64_t *PaddingBytesAfterBitmap, uint64_t *PaddingBytesAfterNames, + uint64_t *PaddingBytesAfterVTable, uint64_t *PaddingBytesAfterVNames); /*! * \brief Set the flag that profile data has been dumped to the file. diff --git a/compiler-rt/lib/profile/InstrProfilingBuffer.c b/compiler-rt/lib/profile/InstrProfilingBuffer.c index af52804..f31dc7d 100644 --- a/compiler-rt/lib/profile/InstrProfilingBuffer.c +++ b/compiler-rt/lib/profile/InstrProfilingBuffer.c @@ -70,6 +70,18 @@ uint64_t __llvm_profile_get_data_size(const __llvm_profile_data *Begin, const __llvm_profile_data *End) { return __llvm_profile_get_num_data(Begin, End) * sizeof(__llvm_profile_data); } +COMPILER_RT_VISIBILITY +uint64_t __llvm_profile_get_num_vtable(const VTableProfData *Begin, + const VTableProfData *End) { + intptr_t EndI = (intptr_t)End, BeginI = (intptr_t)Begin; + return (EndI + sizeof(VTableProfData) - 1 - BeginI) / sizeof(VTableProfData); +} + +COMPILER_RT_VISIBILITY +uint64_t __llvm_profile_get_vtable_section_size(const VTableProfData *Begin, + const VTableProfData *End) { + return __llvm_profile_get_num_vtable(Begin, End) * sizeof(VTableProfData); +} COMPILER_RT_VISIBILITY size_t __llvm_profile_counter_entry_size(void) { if (__llvm_profile_get_version() & VARIANT_MASK_BYTE_COVERAGE) @@ -119,11 +131,13 @@ static int needsCounterPadding(void) { } COMPILER_RT_VISIBILITY -void __llvm_profile_get_padding_sizes_for_counters( +int __llvm_profile_get_padding_sizes_for_counters( uint64_t DataSize, uint64_t CountersSize, uint64_t NumBitmapBytes, - uint64_t NamesSize, uint64_t *PaddingBytesBeforeCounters, - uint64_t *PaddingBytesAfterCounters, uint64_t *PaddingBytesAfterBitmapBytes, - uint64_t *PaddingBytesAfterNames) { + uint64_t NamesSize, uint64_t VTableSize, uint64_t VNameSize, + uint64_t *PaddingBytesBeforeCounters, uint64_t *PaddingBytesAfterCounters, + uint64_t *PaddingBytesAfterBitmapBytes, uint64_t *PaddingBytesAfterNames, + uint64_t *PaddingBytesAfterVTable, uint64_t *PaddingBytesAfterVName) { + // Counter padding is needed only if continuous mode is enabled. if (!needsCounterPadding()) { *PaddingBytesBeforeCounters = 0; *PaddingBytesAfterCounters = @@ -131,9 +145,19 @@ void __llvm_profile_get_padding_sizes_for_counters( *PaddingBytesAfterBitmapBytes = __llvm_profile_get_num_padding_bytes(NumBitmapBytes); *PaddingBytesAfterNames = __llvm_profile_get_num_padding_bytes(NamesSize); - return; + if (PaddingBytesAfterVTable != NULL) + *PaddingBytesAfterVTable = + __llvm_profile_get_num_padding_bytes(VTableSize); + if (PaddingBytesAfterVName != NULL) + *PaddingBytesAfterVName = __llvm_profile_get_num_padding_bytes(VNameSize); + return 0; } + // Value profiling not supported in continuous mode at profile-write time. + // Return -1 to alert the incompatibility. + if (VTableSize != 0 || VNameSize != 0) + return -1; + // In continuous mode, the file offsets for headers and for the start of // counter sections need to be page-aligned. *PaddingBytesBeforeCounters = @@ -142,6 +166,13 @@ void __llvm_profile_get_padding_sizes_for_counters( *PaddingBytesAfterBitmapBytes = calculateBytesNeededToPageAlign(NumBitmapBytes); *PaddingBytesAfterNames = calculateBytesNeededToPageAlign(NamesSize); + // Set these two variables to zero to avoid uninitialized variables + // even if VTableSize and VNameSize are known to be zero. + if (PaddingBytesAfterVTable != NULL) + *PaddingBytesAfterVTable = 0; + if (PaddingBytesAfterVName != NULL) + *PaddingBytesAfterVName = 0; + return 0; } COMPILER_RT_VISIBILITY @@ -162,9 +193,11 @@ uint64_t __llvm_profile_get_size_for_buffer_internal( uint64_t PaddingBytesBeforeCounters, PaddingBytesAfterCounters, PaddingBytesAfterNames, PaddingBytesAfterBitmapBytes; __llvm_profile_get_padding_sizes_for_counters( - DataSize, CountersSize, NumBitmapBytes, NamesSize, - &PaddingBytesBeforeCounters, &PaddingBytesAfterCounters, - &PaddingBytesAfterBitmapBytes, &PaddingBytesAfterNames); + DataSize, CountersSize, NumBitmapBytes, NamesSize, 0 /* VTableSize */, + 0 /* VNameSize */, &PaddingBytesBeforeCounters, + &PaddingBytesAfterCounters, &PaddingBytesAfterBitmapBytes, + &PaddingBytesAfterNames, NULL /* PaddingBytesAfterVTable */, + NULL /* PaddingbytesAfterVNames */); return sizeof(__llvm_profile_header) + __llvm_write_binary_ids(NULL) + DataSize + PaddingBytesBeforeCounters + CountersSize + @@ -191,7 +224,10 @@ COMPILER_RT_VISIBILITY int __llvm_profile_write_buffer_internal( const char *NamesBegin, const char *NamesEnd) { ProfDataWriter BufferWriter; initBufferWriter(&BufferWriter, Buffer); - return lprofWriteDataImpl(&BufferWriter, DataBegin, DataEnd, CountersBegin, - CountersEnd, BitmapBegin, BitmapEnd, 0, NamesBegin, - NamesEnd, 0); + // Set virtual table arguments to NULL since they are not supported yet. + return lprofWriteDataImpl( + &BufferWriter, DataBegin, DataEnd, CountersBegin, CountersEnd, + BitmapBegin, BitmapEnd, 0 /* VPDataReader */, NamesBegin, NamesEnd, + NULL /* VTableBegin */, NULL /* VTableEnd */, NULL /* VNamesBegin */, + NULL /* VNamesEnd */, 0 /* SkipNameDataWrite */); } diff --git a/compiler-rt/lib/profile/InstrProfilingInternal.h b/compiler-rt/lib/profile/InstrProfilingInternal.h index 03ed67f..38159b6 100644 --- a/compiler-rt/lib/profile/InstrProfilingInternal.h +++ b/compiler-rt/lib/profile/InstrProfilingInternal.h @@ -156,7 +156,9 @@ int lprofWriteDataImpl(ProfDataWriter *Writer, const char *CountersBegin, const char *CountersEnd, const char *BitmapBegin, const char *BitmapEnd, VPDataReaderType *VPDataReader, const char *NamesBegin, - const char *NamesEnd, int SkipNameDataWrite); + const char *NamesEnd, const VTableProfData *VTableBegin, + const VTableProfData *VTableEnd, const char *VNamesBegin, + const char *VNamesEnd, int SkipNameDataWrite); /* Merge value profile data pointed to by SrcValueProfData into * in-memory profile counters pointed by to DstData. */ diff --git a/compiler-rt/lib/profile/InstrProfilingMerge.c b/compiler-rt/lib/profile/InstrProfilingMerge.c index b5850e9..ad7a50dc 100644 --- a/compiler-rt/lib/profile/InstrProfilingMerge.c +++ b/compiler-rt/lib/profile/InstrProfilingMerge.c @@ -107,6 +107,27 @@ static uintptr_t signextIfWin64(void *V) { #endif } +static uint64_t +getDistanceFromCounterToValueProf(const __llvm_profile_header *const Header) { + // Skip names section, vtable profile data section and vtable names section + // for runtime profile merge. To merge runtime addresses from multiple + // profiles collected from the same instrumented binary, the binary should be + // loaded at fixed base address (e.g., build with -no-pie, or run with ASLR + // disabled). + // In this set-up these three sections remain unchanged. + const uint64_t VTableSectionSize = + Header->NumVTables * sizeof(VTableProfData); + const uint64_t PaddingBytesAfterVTableSection = + __llvm_profile_get_num_padding_bytes(VTableSectionSize); + const uint64_t VNamesSize = Header->VNamesSize; + const uint64_t PaddingBytesAfterVNamesSize = + __llvm_profile_get_num_padding_bytes(VNamesSize); + return Header->NamesSize + + __llvm_profile_get_num_padding_bytes(Header->NamesSize) + + VTableSectionSize + PaddingBytesAfterVTableSection + VNamesSize + + PaddingBytesAfterVNamesSize; +} + COMPILER_RT_VISIBILITY int __llvm_profile_merge_from_buffer(const char *ProfileData, uint64_t ProfileSize) { @@ -136,9 +157,9 @@ int __llvm_profile_merge_from_buffer(const char *ProfileData, Header->NumCounters * __llvm_profile_counter_entry_size(); SrcBitmapStart = SrcCountersEnd; SrcNameStart = SrcBitmapStart + Header->NumBitmapBytes; + SrcValueProfDataStart = - SrcNameStart + Header->NamesSize + - __llvm_profile_get_num_padding_bytes(Header->NamesSize); + SrcNameStart + getDistanceFromCounterToValueProf(Header); if (SrcNameStart < SrcCountersStart || SrcNameStart < SrcBitmapStart) return 1; diff --git a/compiler-rt/lib/profile/InstrProfilingPlatformLinux.c b/compiler-rt/lib/profile/InstrProfilingPlatformLinux.c index 19266ab..d2554a2 100644 --- a/compiler-rt/lib/profile/InstrProfilingPlatformLinux.c +++ b/compiler-rt/lib/profile/InstrProfilingPlatformLinux.c @@ -24,8 +24,12 @@ #define PROF_DATA_STOP INSTR_PROF_SECT_STOP(INSTR_PROF_DATA_COMMON) #define PROF_NAME_START INSTR_PROF_SECT_START(INSTR_PROF_NAME_COMMON) #define PROF_NAME_STOP INSTR_PROF_SECT_STOP(INSTR_PROF_NAME_COMMON) +#define PROF_VNAME_START INSTR_PROF_SECT_START(INSTR_PROF_VNAME_COMMON) +#define PROF_VNAME_STOP INSTR_PROF_SECT_STOP(INSTR_PROF_VNAME_COMMON) #define PROF_CNTS_START INSTR_PROF_SECT_START(INSTR_PROF_CNTS_COMMON) #define PROF_CNTS_STOP INSTR_PROF_SECT_STOP(INSTR_PROF_CNTS_COMMON) +#define PROF_VTABLE_START INSTR_PROF_SECT_START(INSTR_PROF_VTAB_COMMON) +#define PROF_VTABLE_STOP INSTR_PROF_SECT_STOP(INSTR_PROF_VTAB_COMMON) #define PROF_BITS_START INSTR_PROF_SECT_START(INSTR_PROF_BITS_COMMON) #define PROF_BITS_STOP INSTR_PROF_SECT_STOP(INSTR_PROF_BITS_COMMON) #define PROF_ORDERFILE_START INSTR_PROF_SECT_START(INSTR_PROF_ORDERFILE_COMMON) @@ -41,6 +45,10 @@ extern __llvm_profile_data PROF_DATA_STOP COMPILER_RT_VISIBILITY COMPILER_RT_WEAK; extern char PROF_CNTS_START COMPILER_RT_VISIBILITY COMPILER_RT_WEAK; extern char PROF_CNTS_STOP COMPILER_RT_VISIBILITY COMPILER_RT_WEAK; +extern VTableProfData PROF_VTABLE_START COMPILER_RT_VISIBILITY COMPILER_RT_WEAK; +extern VTableProfData PROF_VTABLE_STOP COMPILER_RT_VISIBILITY COMPILER_RT_WEAK; +extern char PROF_VNAME_START COMPILER_RT_VISIBILITY COMPILER_RT_WEAK; +extern char PROF_VNAME_STOP COMPILER_RT_VISIBILITY COMPILER_RT_WEAK; extern char PROF_BITS_START COMPILER_RT_VISIBILITY COMPILER_RT_WEAK; extern char PROF_BITS_STOP COMPILER_RT_VISIBILITY COMPILER_RT_WEAK; extern uint32_t PROF_ORDERFILE_START COMPILER_RT_VISIBILITY COMPILER_RT_WEAK; @@ -63,6 +71,18 @@ COMPILER_RT_VISIBILITY const char *__llvm_profile_begin_names(void) { COMPILER_RT_VISIBILITY const char *__llvm_profile_end_names(void) { return &PROF_NAME_STOP; } +COMPILER_RT_VISIBILITY const char *__llvm_profile_begin_vtabnames(void) { + return &PROF_VNAME_START; +} +COMPILER_RT_VISIBILITY const char *__llvm_profile_end_vtabnames(void) { + return &PROF_VNAME_STOP; +} +COMPILER_RT_VISIBILITY VTableProfData *__llvm_profile_begin_vtables(void) { + return &PROF_VTABLE_START; +} +COMPILER_RT_VISIBILITY VTableProfData *__llvm_profile_end_vtables(void) { + return &PROF_VTABLE_STOP; +} COMPILER_RT_VISIBILITY char *__llvm_profile_begin_counters(void) { return &PROF_CNTS_START; } diff --git a/compiler-rt/lib/profile/InstrProfilingWriter.c b/compiler-rt/lib/profile/InstrProfilingWriter.c index 4d767d1..8816a71 100644 --- a/compiler-rt/lib/profile/InstrProfilingWriter.c +++ b/compiler-rt/lib/profile/InstrProfilingWriter.c @@ -250,9 +250,14 @@ COMPILER_RT_VISIBILITY int lprofWriteData(ProfDataWriter *Writer, const char *BitmapEnd = __llvm_profile_end_bitmap(); const char *NamesBegin = __llvm_profile_begin_names(); const char *NamesEnd = __llvm_profile_end_names(); + const VTableProfData *VTableBegin = __llvm_profile_begin_vtables(); + const VTableProfData *VTableEnd = __llvm_profile_end_vtables(); + const char *VNamesBegin = __llvm_profile_begin_vtabnames(); + const char *VNamesEnd = __llvm_profile_end_vtabnames(); return lprofWriteDataImpl(Writer, DataBegin, DataEnd, CountersBegin, CountersEnd, BitmapBegin, BitmapEnd, VPDataReader, - NamesBegin, NamesEnd, SkipNameDataWrite); + NamesBegin, NamesEnd, VTableBegin, VTableEnd, + VNamesBegin, VNamesEnd, SkipNameDataWrite); } COMPILER_RT_VISIBILITY int @@ -261,7 +266,9 @@ lprofWriteDataImpl(ProfDataWriter *Writer, const __llvm_profile_data *DataBegin, const char *CountersBegin, const char *CountersEnd, const char *BitmapBegin, const char *BitmapEnd, VPDataReaderType *VPDataReader, const char *NamesBegin, - const char *NamesEnd, int SkipNameDataWrite) { + const char *NamesEnd, const VTableProfData *VTableBegin, + const VTableProfData *VTableEnd, const char *VNamesBegin, + const char *VNamesEnd, int SkipNameDataWrite) { /* Calculate size of sections. */ const uint64_t DataSectionSize = __llvm_profile_get_data_size(DataBegin, DataEnd); @@ -273,6 +280,12 @@ lprofWriteDataImpl(ProfDataWriter *Writer, const __llvm_profile_data *DataBegin, const uint64_t NumBitmapBytes = __llvm_profile_get_num_bitmap_bytes(BitmapBegin, BitmapEnd); const uint64_t NamesSize = __llvm_profile_get_name_size(NamesBegin, NamesEnd); + const uint64_t NumVTables = + __llvm_profile_get_num_vtable(VTableBegin, VTableEnd); + const uint64_t VTableSectionSize = + __llvm_profile_get_vtable_section_size(VTableBegin, VTableEnd); + const uint64_t VNamesSize = + __llvm_profile_get_name_size(VNamesBegin, VNamesEnd); /* Create the header. */ __llvm_profile_header Header; @@ -280,11 +293,15 @@ lprofWriteDataImpl(ProfDataWriter *Writer, const __llvm_profile_data *DataBegin, /* Determine how much padding is needed before/after the counters and after * the names. */ uint64_t PaddingBytesBeforeCounters, PaddingBytesAfterCounters, - PaddingBytesAfterNames, PaddingBytesAfterBitmapBytes; - __llvm_profile_get_padding_sizes_for_counters( - DataSectionSize, CountersSectionSize, NumBitmapBytes, NamesSize, - &PaddingBytesBeforeCounters, &PaddingBytesAfterCounters, - &PaddingBytesAfterBitmapBytes, &PaddingBytesAfterNames); + PaddingBytesAfterBitmapBytes, PaddingBytesAfterNames, + PaddingBytesAfterVTable, PaddingBytesAfterVNames; + if (__llvm_profile_get_padding_sizes_for_counters( + DataSectionSize, CountersSectionSize, NumBitmapBytes, NamesSize, + VTableSectionSize, VNamesSize, &PaddingBytesBeforeCounters, + &PaddingBytesAfterCounters, &PaddingBytesAfterBitmapBytes, + &PaddingBytesAfterNames, &PaddingBytesAfterVTable, + &PaddingBytesAfterVNames) == -1) + return -1; { /* Initialize header structure. */ @@ -323,7 +340,11 @@ lprofWriteDataImpl(ProfDataWriter *Writer, const __llvm_profile_data *DataBegin, {BitmapBegin, sizeof(uint8_t), NumBitmapBytes, 0}, {NULL, sizeof(uint8_t), PaddingBytesAfterBitmapBytes, 1}, {SkipNameDataWrite ? NULL : NamesBegin, sizeof(uint8_t), NamesSize, 0}, - {NULL, sizeof(uint8_t), PaddingBytesAfterNames, 1}}; + {NULL, sizeof(uint8_t), PaddingBytesAfterNames, 1}, + {VTableBegin, sizeof(uint8_t), VTableSectionSize, 0}, + {NULL, sizeof(uint8_t), PaddingBytesAfterVTable, 1}, + {SkipNameDataWrite ? NULL : VNamesBegin, sizeof(uint8_t), VNamesSize, 0}, + {NULL, sizeof(uint8_t), PaddingBytesAfterVNames, 1}}; if (Writer->Write(Writer, IOVecData, sizeof(IOVecData) / sizeof(*IOVecData))) return -1; diff --git a/llvm/include/llvm/Analysis/IndirectCallVisitor.h b/llvm/include/llvm/Analysis/IndirectCallVisitor.h index 0825e19e..5969241 100644 --- a/llvm/include/llvm/Analysis/IndirectCallVisitor.h +++ b/llvm/include/llvm/Analysis/IndirectCallVisitor.h @@ -12,27 +12,87 @@ #ifndef LLVM_ANALYSIS_INDIRECTCALLVISITOR_H #define LLVM_ANALYSIS_INDIRECTCALLVISITOR_H +#include "llvm/ADT/SetVector.h" #include "llvm/IR/InstVisitor.h" #include <vector> namespace llvm { -// Visitor class that finds all indirect call. +// Visitor class that finds indirect calls or instructions that gives vtable +// value, depending on Type. struct PGOIndirectCallVisitor : public InstVisitor<PGOIndirectCallVisitor> { + enum class InstructionType { + kIndirectCall = 0, + kVTableVal = 1, + }; std::vector<CallBase *> IndirectCalls; - PGOIndirectCallVisitor() = default; + std::vector<Instruction *> ProfiledAddresses; + PGOIndirectCallVisitor(InstructionType Type) : Type(Type) {} + + // Given an indirect call instruction, try to find the the following pattern + // + // %vtable = load ptr, ptr %obj + // %vfn = getelementptr inbounds ptr, ptr %vtable, i64 1 + // %2 = load ptr, ptr %vfn + // $call = tail call i32 %2 + // + // A heuristic is used to find the address feeding instructions. + static Instruction *tryGetVTableInstruction(CallBase *CB) { + assert(CB != nullptr && "Caller guaranteed"); + LoadInst *LI = dyn_cast<LoadInst>(CB->getCalledOperand()); + + if (LI != nullptr) { + Value *FuncPtr = LI->getPointerOperand(); // GEP (or bitcast) + Value *VTablePtr = FuncPtr->stripInBoundsConstantOffsets(); + // FIXME: Add support in the frontend so LLVM type intrinsics are + // emitted without LTO. This way, added intrinsics could filter + // non-vtable instructions and reduce instrumentation overhead. + // Since a non-vtable profiled address is not within the address + // range of vtable objects, it's stored as zero in indexed profiles. + // A pass that looks up symbol with an zero hash will (almost) always + // find nullptr and skip the actual transformation (e.g., comparison + // of symbols). So the performance overhead from non-vtable profiled + // address is negligible if exists at all. Comparing loaded address + // with symbol address guarantees correctness. + if (VTablePtr != nullptr && isa<Instruction>(VTablePtr)) { + return cast<Instruction>(VTablePtr); + } + } + return nullptr; + } void visitCallBase(CallBase &Call) { - if (Call.isIndirectCall()) + if (Call.isIndirectCall()) { IndirectCalls.push_back(&Call); + + if (Type != InstructionType::kVTableVal) + return; + + Instruction *VPtr = + PGOIndirectCallVisitor::tryGetVTableInstruction(&Call); + if (VPtr) { + ProfiledAddresses.push_back(VPtr); + } + } } + +private: + InstructionType Type; }; -// Helper function that finds all indirect call sites. inline std::vector<CallBase *> findIndirectCalls(Function &F) { - PGOIndirectCallVisitor ICV; + PGOIndirectCallVisitor ICV( + PGOIndirectCallVisitor::InstructionType::kIndirectCall); ICV.visit(F); return ICV.IndirectCalls; } + +inline std::vector<Instruction *> findVTableAddrs(Function &F) { + PGOIndirectCallVisitor ICV( + PGOIndirectCallVisitor::InstructionType::kVTableVal); + ICV.visit(F); + return ICV.ProfiledAddresses; +} + } // namespace llvm #endif diff --git a/llvm/include/llvm/ProfileData/InstrProf.h b/llvm/include/llvm/ProfileData/InstrProf.h index 87e7bbb..d5d3c7f 100644 --- a/llvm/include/llvm/ProfileData/InstrProf.h +++ b/llvm/include/llvm/ProfileData/InstrProf.h @@ -89,6 +89,9 @@ inline StringRef getInstrProfValueProfMemOpFuncName() { /// Return the name prefix of variables containing instrumented function names. inline StringRef getInstrProfNameVarPrefix() { return "__profn_"; } +/// Return the name prefix of variables containing virtual table profile data. +inline StringRef getInstrProfVTableVarPrefix() { return "__profvt_"; } + /// Return the name prefix of variables containing per-function control data. inline StringRef getInstrProfDataVarPrefix() { return "__profd_"; } @@ -110,6 +113,8 @@ inline StringRef getInstrProfNamesVarName() { return "__llvm_prf_nm"; } +inline StringRef getInstrProfVTableNamesVarName() { return "__llvm_prf_vnm"; } + /// Return the name of a covarage mapping variable (internal linkage) /// for each instrumented source module. Such variables are allocated /// in the __llvm_covmap section. @@ -193,6 +198,15 @@ std::string getPGOFuncName(StringRef RawFuncName, /// called from LTO optimization passes. std::string getIRPGOFuncName(const Function &F, bool InLTO = false); +/// Returns the PGO object name. This function has some special handling +/// when called in LTO optimization: +/// - In LTO mode, returns the name in `PGONameMetadata` if exists, otherwise +/// returns the IRPGO name as if the GO has non-local linkage. +/// - In non-LTO mode, returns the IRPGO name as it is. +/// More rationale in the comment of function implementation. +std::string getIRPGOObjectName(const GlobalObject &GO, bool InLTO = false, + MDNode *PGONameMetadata = nullptr); + /// \return the filename and the function name parsed from the output of /// \c getIRPGOFuncName() std::pair<StringRef, StringRef> getParsedIRPGOFuncName(StringRef IRPGOFuncName); @@ -246,6 +260,9 @@ Error collectGlobalObjectNameStrings(ArrayRef<std::string> NameStrs, Error collectPGOFuncNameStrings(ArrayRef<GlobalVariable *> NameVars, std::string &Result, bool doCompression = true); +Error collectVTableStrings(ArrayRef<GlobalVariable *> VTables, + std::string &Result, bool doCompression); + /// Check if INSTR_PROF_RAW_VERSION_VAR is defined. This global is only being /// set in IR PGO compilation. bool isIRPGOFlagSet(const Module *M); @@ -269,13 +286,15 @@ void annotateValueSite(Module &M, Instruction &Inst, uint32_t MaxMDCount = 3); /// Same as the above interface but using an ArrayRef, as well as \p Sum. +/// This function will not annotate !prof metadata on the instruction if the +/// referenced array is empty. void annotateValueSite(Module &M, Instruction &Inst, ArrayRef<InstrProfValueData> VDs, uint64_t Sum, InstrProfValueKind ValueKind, uint32_t MaxMDCount); /// Extract the value profile data from \p Inst which is annotated with /// value profile meta data. Return false if there is no value data annotated, -/// otherwise return true. +/// otherwise return true. bool getValueProfDataFromInst(const Instruction &Inst, InstrProfValueKind ValueKind, uint32_t MaxNumValueData, @@ -283,11 +302,23 @@ bool getValueProfDataFromInst(const Instruction &Inst, uint32_t &ActualNumValueData, uint64_t &TotalC, bool GetNoICPValue = false); +/// Extract the value profile data from \p Inst and returns them if \p Inst is +/// annotated with value profile data. Returns nullptr otherwise. It's similar +/// to `getValueProfDataFromInst` above except that an array is allocated only +/// after a preliminary checking that the value profiles of kind `ValueKind` +/// exist. +std::unique_ptr<InstrProfValueData[]> +getValueProfDataFromInst(const Instruction &Inst, InstrProfValueKind ValueKind, + uint32_t MaxNumValueData, uint32_t &ActualNumValueData, + uint64_t &TotalC, bool GetNoICPValue = false); + inline StringRef getPGOFuncNameMetadataName() { return "PGOFuncName"; } /// Return the PGOFuncName meta data associated with a function. MDNode *getPGOFuncNameMetadata(const Function &F); +std::string getPGOName(const GlobalVariable &V, bool InLTO = false); + /// Create the PGOFuncName meta data if PGOFuncName is different from /// function's raw name. This should only apply to internal linkage functions /// declared by users only. @@ -295,7 +326,7 @@ void createPGOFuncNameMetadata(Function &F, StringRef PGOFuncName); /// Check if we can use Comdat for profile variables. This will eliminate /// the duplicated profile variables for Comdat functions. -bool needsComdatForCounter(const Function &F, const Module &M); +bool needsComdatForCounter(const GlobalValue &GV, const Module &M); /// An enum describing the attributes of an instrumented profile. enum class InstrProfKind { @@ -429,28 +460,61 @@ uint64_t ComputeHash(StringRef K); class InstrProfSymtab { public: using AddrHashMap = std::vector<std::pair<uint64_t, uint64_t>>; + using RangeHashMap = + std::vector<std::pair<std::pair<uint64_t, uint64_t>, uint64_t>>; private: StringRef Data; uint64_t Address = 0; - // Unique name strings. + // Unique name strings. Used to ensure entries in MD5NameMap (a vector that's + // going to be sorted) has unique MD5 keys in the first place. StringSet<> NameTab; + // Records the unique virtual table names. This is used by InstrProfWriter to + // write out an on-disk chained hash table of virtual table names. + // InstrProfWriter stores per function profile data (keyed by function names) + // so it doesn't use a StringSet for function names. + StringSet<> VTableNames; // A map from MD5 keys to function name strings. std::vector<std::pair<uint64_t, StringRef>> MD5NameMap; + // A map from MD5 keys to virtual table definitions. Only populated when + // building the Symtab from a module. + std::vector<std::pair<uint64_t, GlobalVariable *>> MD5VTableMap; // A map from MD5 keys to function define. We only populate this map // when build the Symtab from a Module. std::vector<std::pair<uint64_t, Function *>> MD5FuncMap; // A map from function runtime address to function name MD5 hash. // This map is only populated and used by raw instr profile reader. AddrHashMap AddrToMD5Map; + // A map from virtual table runtime address to function name MD5 hash. + // This map is only populated and used by raw instr profile reader. + // This is a different map from 'AddrToMD5Map' for readability and + // debuggability. + RangeHashMap VTableAddrRangeToMD5Map; bool Sorted = false; static StringRef getExternalSymbol() { return "** External Symbol **"; } + // Returns the canonical name of the given PGOName by stripping the names + // suffixes that begins with ".". If MayHaveUniqueSuffix is true, ".__uniq." + // suffix is kept in the canonical name. + StringRef getCanonicalName(StringRef PGOName, bool MayHaveUniqueSuffix); + + // Add the function into the symbol table, by creating the following + // map entries: + // - <MD5Hash(PGOFuncName), PGOFuncName> + // - <MD5Hash(PGOFuncName), F> + // - <MD5Hash(getCanonicalName(PGOFuncName), F) Error addFuncWithName(Function &F, StringRef PGOFuncName); + // Add the vtable into the symbol table, by creating the following + // map entries: + // - <MD5Hash(PGOVTableName), PGOVTableName> + // - <MD5Hash(PGOVTableName), V> + // - <MD5Hash(getCanonicalName(PGOVTableName), V) + Error addVTableWithName(GlobalVariable &V, StringRef PGOVTableName); + // If the symtab is created by a series of calls to \c addFuncName, \c // finalizeSymtab needs to be called before looking up function names. // This is required because the underlying map is a vector (for space @@ -470,9 +534,19 @@ public: /// \c NameStrings is a string composed of one of more sub-strings /// encoded in the format described in \c collectPGOFuncNameStrings. - /// This method is a wrapper to \c readPGOFuncNameStrings method. + /// This method is a wrapper to \c readAndDecodeStrings method. Error create(StringRef NameStrings); + /// \c FuncNameStrings is a string composed of one or more encoded function + /// name strings, and \c VTableNameStrings composes of one or more encoded + /// vtable names. This function is a wrapper to \c readAndDecodeStrings + /// method. + Error create(StringRef FuncNameStrings, StringRef VTableNameStrings); + + /// Initialize 'this' with the set of vtable names encoded in + /// \c CompressedVTableNames. + Error initVTableNamesFromCompressedStrings(StringRef CompressedVTableNames); + /// This interface is used by reader of CoverageMapping test /// format. inline Error create(StringRef D, uint64_t BaseAddr); @@ -485,32 +559,71 @@ public: /// Create InstrProfSymtab from a set of names iteratable from /// \p IterRange. This interface is used by IndexedProfReader. - template <typename NameIterRange> Error create(const NameIterRange &IterRange); - - /// Update the symtab by adding \p FuncName to the table. This interface - /// is used by the raw and text profile readers. - Error addFuncName(StringRef FuncName) { - if (FuncName.empty()) + template <typename NameIterRange> + Error create(const NameIterRange &IterRange); + + /// Create InstrProfSymtab from a set of function names and vtable + /// names iteratable from \p IterRange. This interface is used by + /// IndexedProfReader. + template <typename FuncNameIterRange, typename VTableNameIterRange> + Error create(const FuncNameIterRange &FuncIterRange, + const VTableNameIterRange &VTableIterRange); + + // Map the MD5 of the symbol name to the name. + Error addSymbolName(StringRef SymbolName) { + if (SymbolName.empty()) return make_error<InstrProfError>(instrprof_error::malformed, - "function name is empty"); - auto Ins = NameTab.insert(FuncName); + "symbol name is empty"); + + // Insert into NameTab so that MD5NameMap (a vector that will be sorted) + // won't have duplicated entries in the first place. + auto Ins = NameTab.insert(SymbolName); if (Ins.second) { MD5NameMap.push_back(std::make_pair( - IndexedInstrProf::ComputeHash(FuncName), Ins.first->getKey())); + IndexedInstrProf::ComputeHash(SymbolName), Ins.first->getKey())); Sorted = false; } return Error::success(); } + /// The method name is kept since there are many callers. + /// It just forwards to 'addSymbolName'. + Error addFuncName(StringRef FuncName) { return addSymbolName(FuncName); } + + /// Adds VTableName as a known symbol, and inserts it to a map that + /// tracks all vtable names. + Error addVTableName(StringRef VTableName) { + if (Error E = addSymbolName(VTableName)) + return E; + + // Record VTableName. InstrProfWriter uses this map. The comment around + // class member explains why. + VTableNames.insert(VTableName); + return Error::success(); + } + + const StringSet<> &getVTableNames() const { return VTableNames; } + /// Map a function address to its name's MD5 hash. This interface /// is only used by the raw profiler reader. void mapAddress(uint64_t Addr, uint64_t MD5Val) { AddrToMD5Map.push_back(std::make_pair(Addr, MD5Val)); } + /// Map the address range (i.e., [start_address, end_address]) of a variable + /// to its names' MD5 hash. This interface is only used by the raw profile + /// reader. + void mapVTableAddress(uint64_t StartAddr, uint64_t EndAddr, uint64_t MD5Val) { + VTableAddrRangeToMD5Map.push_back( + std::make_pair(std::make_pair(StartAddr, EndAddr), MD5Val)); + } + /// Return a function's hash, or 0, if the function isn't in this SymTab. uint64_t getFunctionHashFromAddress(uint64_t Address); + /// Return a vtable's hash, or 0 if the vtable doesn't exist in this SymTab. + uint64_t getVTableHashFromAddress(uint64_t Address); + /// Return function's PGO name from the function name's symbol /// address in the object file. If an error occurs, return /// an empty string. @@ -532,6 +645,8 @@ public: /// Return function from the name's md5 hash. Return nullptr if not found. inline Function *getFunction(uint64_t FuncMD5Hash); + // Return vtable from the name's MD5 hash. Return nullptr if not found. + inline GlobalVariable *getGlobalVariable(uint64_t GlobalVariableMD5Hash); /// Return the name section data. inline StringRef getNameData() const { return Data; } @@ -556,14 +671,39 @@ Error InstrProfSymtab::create(const NameIterRange &IterRange) { return Error::success(); } +template <typename FuncNameIterRange, typename VTableNameIterRange> +Error InstrProfSymtab::create(const FuncNameIterRange &FuncIterRange, + const VTableNameIterRange &VTableIterRange) { + for (auto Name : FuncIterRange) + if (Error E = addFuncName(Name)) + return E; + + for (auto VTableName : VTableIterRange) { + if (Error E = addVTableName(VTableName)) { + return E; + } + } + + finalizeSymtab(); + return Error::success(); +} + void InstrProfSymtab::finalizeSymtab() { if (Sorted) return; llvm::sort(MD5NameMap, less_first()); + llvm::sort(MD5VTableMap, less_first()); llvm::sort(MD5FuncMap, less_first()); llvm::sort(AddrToMD5Map, less_first()); AddrToMD5Map.erase(std::unique(AddrToMD5Map.begin(), AddrToMD5Map.end()), AddrToMD5Map.end()); + // VTable object address ranges should not overlap; so sort by either + // beginning address or end address is fine. + llvm::sort(VTableAddrRangeToMD5Map, less_first()); + // std::unique uses == operator for std::pair. + VTableAddrRangeToMD5Map.erase(std::unique(VTableAddrRangeToMD5Map.begin(), + VTableAddrRangeToMD5Map.end()), + VTableAddrRangeToMD5Map.end()); Sorted = true; } @@ -594,6 +734,19 @@ Function* InstrProfSymtab::getFunction(uint64_t FuncMD5Hash) { return nullptr; } +GlobalVariable * +InstrProfSymtab::getGlobalVariable(uint64_t GlobalVariableMD5Hash) { + finalizeSymtab(); + auto Result = + llvm::lower_bound(MD5VTableMap, GlobalVariableMD5Hash, + [](const std::pair<uint64_t, GlobalVariable *> &LHS, + uint64_t RHS) { return LHS.first < RHS; }); + + if (Result != MD5VTableMap.end() && Result->first == GlobalVariableMD5Hash) + return Result->second; + return nullptr; +} + // To store the sums of profile count values, or the percentage of // the sums of the total count values. struct CountSumOrPercent { @@ -820,6 +973,7 @@ private: struct ValueProfData { std::vector<InstrProfValueSiteRecord> IndirectCallSites; std::vector<InstrProfValueSiteRecord> MemOPSizes; + std::vector<InstrProfValueSiteRecord> VTableTargets; }; std::unique_ptr<ValueProfData> ValueData; @@ -842,6 +996,8 @@ private: return ValueData->IndirectCallSites; case IPVK_MemOPSize: return ValueData->MemOPSizes; + case IPVK_VTableTarget: + return ValueData->VTableTargets; default: llvm_unreachable("Unknown value kind!"); } @@ -856,6 +1012,8 @@ private: return ValueData->IndirectCallSites; case IPVK_MemOPSize: return ValueData->MemOPSizes; + case IPVK_VTableTarget: + return ValueData->VTableTargets; default: llvm_unreachable("Unknown value kind!"); } @@ -1025,7 +1183,9 @@ enum ProfVersion { Version10 = 10, // An additional field is used for bitmap bytes. Version11 = 11, - // The current version is 11. + // VTable profiling, + Version12 = 12, + // The current version is 12. CurrentVersion = INSTR_PROF_INDEX_VERSION }; const uint64_t Version = ProfVersion::CurrentVersion; @@ -1046,6 +1206,7 @@ struct Header { uint64_t MemProfOffset; uint64_t BinaryIdOffset; uint64_t TemporalProfTracesOffset; + uint64_t VTableNamesOffset; // Organize virtual table names. // New fields should only be added at the end to ensure that the size // computation is correct. The methods below need to be updated to ensure that // the new field is read correctly. @@ -1182,8 +1343,13 @@ template <> inline uint64_t getMagic<uint32_t>() { // It should also match the synthesized type in // Transforms/Instrumentation/InstrProfiling.cpp:getOrCreateRegionCounters. template <class IntPtrT> struct alignas(8) ProfileData { - #define INSTR_PROF_DATA(Type, LLVMType, Name, Init) Type Name; - #include "llvm/ProfileData/InstrProfData.inc" +#define INSTR_PROF_DATA(Type, LLVMType, Name, Init) Type Name; +#include "llvm/ProfileData/InstrProfData.inc" +}; + +template <class IntPtrT> struct alignas(8) VTableProfileData { +#define INSTR_PROF_VTABLE_DATA(Type, LLVMType, Name, Init) Type Name; +#include "llvm/ProfileData/InstrProfData.inc" }; // File header structure of the LLVM profile data in raw format. diff --git a/llvm/include/llvm/ProfileData/InstrProfData.inc b/llvm/include/llvm/ProfileData/InstrProfData.inc index 25df899..77720ab 100644 --- a/llvm/include/llvm/ProfileData/InstrProfData.inc +++ b/llvm/include/llvm/ProfileData/InstrProfData.inc @@ -94,6 +94,22 @@ INSTR_PROF_DATA(const uint32_t, llvm::Type::getInt32Ty(Ctx), NumBitmapBytes, \ #undef INSTR_PROF_DATA /* INSTR_PROF_DATA end. */ +#ifndef INSTR_PROF_VTABLE_DATA +#define INSTR_PROF_VTABLE_DATA(Type, LLVMType, Name, Initializer) +#else +#define INSTR_PROF_VTABLE_DATA_DEFINED +#endif +INSTR_PROF_VTABLE_DATA( + const uint64_t, llvm::Type::getInt64Ty(Ctx), VTableNameHash, + ConstantInt::get(llvm::Type::getInt64Ty(Ctx), + IndexedInstrProf::ComputeHash(PGOVTableName))) +INSTR_PROF_VTABLE_DATA(const IntPtrT, llvm::PointerType::getUnqual(Ctx), + VTablePointer, VTableAddr) +INSTR_PROF_VTABLE_DATA(const uint32_t, llvm::Type::getInt32Ty(Ctx), VTableSize, + ConstantInt::get(llvm::Type::getInt32Ty(Ctx), + VTableSizeVal)) +#undef INSTR_PROF_VTABLE_DATA +/* INSTR_PROF_VTABLE_DATA end. */ /* This is an internal data structure used by value profiler. It * is defined here to allow serialization code sharing by LLVM @@ -145,6 +161,8 @@ INSTR_PROF_RAW_HEADER(uint64_t, CountersDelta, INSTR_PROF_RAW_HEADER(uint64_t, BitmapDelta, (uintptr_t)BitmapBegin - (uintptr_t)DataBegin) INSTR_PROF_RAW_HEADER(uint64_t, NamesDelta, (uintptr_t)NamesBegin) +INSTR_PROF_RAW_HEADER(uint64_t, VNamesSize, VNamesSize) +INSTR_PROF_RAW_HEADER(uint64_t, NumVTables, NumVTables) INSTR_PROF_RAW_HEADER(uint64_t, ValueKindLast, IPVK_Last) #undef INSTR_PROF_RAW_HEADER /* INSTR_PROF_RAW_HEADER end */ @@ -186,13 +204,14 @@ VALUE_PROF_FUNC_PARAM(uint32_t, CounterIndex, Type::getInt32Ty(Ctx)) VALUE_PROF_KIND(IPVK_IndirectCallTarget, 0, "indirect call target") /* For memory intrinsic functions size profiling. */ VALUE_PROF_KIND(IPVK_MemOPSize, 1, "memory intrinsic functions size") +VALUE_PROF_KIND(IPVK_VTableTarget, 2, "vtable target") /* These two kinds must be the last to be * declared. This is to make sure the string * array created with the template can be * indexed with the kind value. */ VALUE_PROF_KIND(IPVK_First, IPVK_IndirectCallTarget, "first") -VALUE_PROF_KIND(IPVK_Last, IPVK_MemOPSize, "last") +VALUE_PROF_KIND(IPVK_Last, IPVK_VTableTarget, "last") #undef VALUE_PROF_KIND /* VALUE_PROF_KIND end */ @@ -267,7 +286,6 @@ COVMAP_HEADER(uint32_t, Int32Ty, Version, \ #undef COVMAP_HEADER /* COVMAP_HEADER end. */ - #ifdef INSTR_PROF_SECT_ENTRY #define INSTR_PROF_DATA_DEFINED INSTR_PROF_SECT_ENTRY(IPSK_data, \ @@ -282,12 +300,18 @@ INSTR_PROF_SECT_ENTRY(IPSK_bitmap, \ INSTR_PROF_SECT_ENTRY(IPSK_name, \ INSTR_PROF_QUOTE(INSTR_PROF_NAME_COMMON), \ INSTR_PROF_NAME_COFF, "__DATA,") +INSTR_PROF_SECT_ENTRY(IPSK_vname, \ + INSTR_PROF_QUOTE(INSTR_PROF_VNAME_COMMON), \ + INSTR_PROF_VNAME_COFF, "__DATA,") INSTR_PROF_SECT_ENTRY(IPSK_vals, \ INSTR_PROF_QUOTE(INSTR_PROF_VALS_COMMON), \ INSTR_PROF_VALS_COFF, "__DATA,") INSTR_PROF_SECT_ENTRY(IPSK_vnodes, \ INSTR_PROF_QUOTE(INSTR_PROF_VNODES_COMMON), \ INSTR_PROF_VNODES_COFF, "__DATA,") +INSTR_PROF_SECT_ENTRY(IPSK_vtab, \ + INSTR_PROF_QUOTE(INSTR_PROF_VTAB_COMMON), \ + INSTR_PROF_VTAB_COFF, "__DATA,") INSTR_PROF_SECT_ENTRY(IPSK_covmap, \ INSTR_PROF_QUOTE(INSTR_PROF_COVMAP_COMMON), \ INSTR_PROF_COVMAP_COFF, "__LLVM_COV,") @@ -307,7 +331,6 @@ INSTR_PROF_SECT_ENTRY(IPSK_covname, \ #undef INSTR_PROF_SECT_ENTRY #endif - #ifdef INSTR_PROF_VALUE_PROF_DATA #define INSTR_PROF_DATA_DEFINED @@ -347,7 +370,7 @@ typedef struct ValueProfRecord { /*! * Return the number of value sites. */ - uint32_t getNumValueSites() const { return NumValueSites; } + uint32_t getNumValueSites() const { return NumValueSites; } /*! * Read data from this record and save it to Record. */ @@ -479,7 +502,6 @@ getValueProfRecordHeaderSize(uint32_t NumValueSites); #undef INSTR_PROF_VALUE_PROF_DATA #endif /* INSTR_PROF_VALUE_PROF_DATA */ - #ifdef INSTR_PROF_COMMON_API_IMPL #define INSTR_PROF_DATA_DEFINED #ifdef __cplusplus @@ -663,9 +685,9 @@ serializeValueProfDataFrom(ValueProfRecordClosure *Closure, (uint64_t)'f' << 16 | (uint64_t)'R' << 8 | (uint64_t)129 /* Raw profile format version (start from 1). */ -#define INSTR_PROF_RAW_VERSION 9 +#define INSTR_PROF_RAW_VERSION 10 /* Indexed profile format version (start from 1). */ -#define INSTR_PROF_INDEX_VERSION 11 +#define INSTR_PROF_INDEX_VERSION 12 /* Coverage mapping format version (start from 0). */ #define INSTR_PROF_COVMAP_VERSION 6 @@ -703,10 +725,12 @@ serializeValueProfDataFrom(ValueProfRecordClosure *Closure, than WIN32 */ #define INSTR_PROF_DATA_COMMON __llvm_prf_data #define INSTR_PROF_NAME_COMMON __llvm_prf_names +#define INSTR_PROF_VNAME_COMMON __llvm_prf_vtabnames #define INSTR_PROF_CNTS_COMMON __llvm_prf_cnts #define INSTR_PROF_BITS_COMMON __llvm_prf_bits #define INSTR_PROF_VALS_COMMON __llvm_prf_vals #define INSTR_PROF_VNODES_COMMON __llvm_prf_vnds +#define INSTR_PROF_VTAB_COMMON __llvm_prf_vtab #define INSTR_PROF_COVMAP_COMMON __llvm_covmap #define INSTR_PROF_COVFUN_COMMON __llvm_covfun #define INSTR_PROF_COVDATA_COMMON __llvm_covdata @@ -717,10 +741,12 @@ serializeValueProfDataFrom(ValueProfRecordClosure *Closure, */ #define INSTR_PROF_DATA_COFF ".lprfd$M" #define INSTR_PROF_NAME_COFF ".lprfn$M" +#define INSTR_PROF_VNAME_COFF ".lprfvn$M" #define INSTR_PROF_CNTS_COFF ".lprfc$M" #define INSTR_PROF_BITS_COFF ".lprfb$M" #define INSTR_PROF_VALS_COFF ".lprfv$M" #define INSTR_PROF_VNODES_COFF ".lprfnd$M" +#define INSTR_PROF_VTAB_COFF ".lprfvt$M" #define INSTR_PROF_COVMAP_COFF ".lcovmap$M" #define INSTR_PROF_COVFUN_COFF ".lcovfun$M" /* Since cov data and cov names sections are not allocated, we don't need to diff --git a/llvm/include/llvm/ProfileData/InstrProfReader.h b/llvm/include/llvm/ProfileData/InstrProfReader.h index 87f1563..c1edd7af 100644 --- a/llvm/include/llvm/ProfileData/InstrProfReader.h +++ b/llvm/include/llvm/ProfileData/InstrProfReader.h @@ -326,12 +326,16 @@ private: uint64_t NamesDelta; const RawInstrProf::ProfileData<IntPtrT> *Data; const RawInstrProf::ProfileData<IntPtrT> *DataEnd; + const RawInstrProf::VTableProfileData<IntPtrT> *VTableBegin = nullptr; + const RawInstrProf::VTableProfileData<IntPtrT> *VTableEnd = nullptr; const char *CountersStart; const char *CountersEnd; const char *BitmapStart; const char *BitmapEnd; const char *NamesStart; const char *NamesEnd; + const char *VNamesStart = nullptr; + const char *VNamesEnd = nullptr; // After value profile is all read, this pointer points to // the header of next profile data (if exists) const uint8_t *ValueDataStart; @@ -622,6 +626,12 @@ public: InstrProfKind getProfileKind() const override; Error populateSymtab(InstrProfSymtab &Symtab) override { + // FIXME: the create method calls 'finalizeSymtab' and sorts a bunch of + // arrays/maps. Since there are other data sources other than 'HashTable' to + // populate a symtab, it might make sense to have something like this + // 1. Let each data source populate Symtab and init the arrays/maps without + // calling 'finalizeSymtab' + // 2. Call 'finalizeSymtab' once to get all arrays/maps sorted if needed. return Symtab.create(HashTable->keys()); } }; @@ -656,6 +666,16 @@ private: std::unique_ptr<MemProfRecordHashTable> MemProfRecordTable; /// MemProf frame profile data on-disk indexed via frame id. std::unique_ptr<MemProfFrameHashTable> MemProfFrameTable; + /// The reader itself doesn't decompress vtable names. A compiler that reads + /// indexed profiles could construct symtab from module IR so it doesn't need + /// the decompressed names. + /// When a symtab is constructed from profiles by llvm-profdata, the list of + /// names could be decompressed based on `VTableNamePtr` and + /// `CompressedVTableNamesLen`. + /// VTableNamePtr points to the beginning of compressed vtable names. + const char *VTableNamePtr = nullptr; + /// The length of compressed vtable names. + uint64_t CompressedVTableNamesLen = 0; /// Total size of binary ids. uint64_t BinaryIdsSize{0}; /// Start address of binary id length and data pairs. diff --git a/llvm/include/llvm/ProfileData/InstrProfWriter.h b/llvm/include/llvm/ProfileData/InstrProfWriter.h index 047b14f..049fa36 100644 --- a/llvm/include/llvm/ProfileData/InstrProfWriter.h +++ b/llvm/include/llvm/ProfileData/InstrProfWriter.h @@ -63,6 +63,9 @@ private: // List of binary ids. std::vector<llvm::object::BuildID> BinaryIds; + // Read the vtable names from raw instr profile reader. + StringSet<> VTableNames; + // An enum describing the attributes of the profile. InstrProfKind ProfileKind = InstrProfKind::Unknown; // Use raw pointer here for the incomplete type object. @@ -84,6 +87,7 @@ public: void addRecord(NamedInstrProfRecord &&I, function_ref<void(Error)> Warn) { addRecord(std::move(I), 1, Warn); } + void addVTableName(StringRef VTableName) { VTableNames.insert(VTableName); } /// Add \p SrcTraces using reservoir sampling where \p SrcStreamSize is the /// total number of temporal profiling traces the source has seen. diff --git a/llvm/lib/Analysis/IndirectCallPromotionAnalysis.cpp b/llvm/lib/Analysis/IndirectCallPromotionAnalysis.cpp index ebfa1c8..ab53717 100644 --- a/llvm/lib/Analysis/IndirectCallPromotionAnalysis.cpp +++ b/llvm/lib/Analysis/IndirectCallPromotionAnalysis.cpp @@ -45,6 +45,10 @@ static cl::opt<unsigned> cl::desc("Max number of promotions for a single indirect " "call callsite")); +cl::opt<unsigned> MaxNumVTableAnnotations( + "icp-max-num-vtables", cl::init(6), cl::Hidden, + cl::desc("Max number of vtables annotated for a vtable load instruction.")); + ICallPromotionAnalysis::ICallPromotionAnalysis() { ValueDataArray = std::make_unique<InstrProfValueData[]>(MaxNumPromotions); } diff --git a/llvm/lib/Analysis/ModuleSummaryAnalysis.cpp b/llvm/lib/Analysis/ModuleSummaryAnalysis.cpp index 1f15e94..3ad0bab 100644 --- a/llvm/lib/Analysis/ModuleSummaryAnalysis.cpp +++ b/llvm/lib/Analysis/ModuleSummaryAnalysis.cpp @@ -82,6 +82,8 @@ static cl::opt<std::string> ModuleSummaryDotFile( extern cl::opt<bool> ScalePartialSampleProfileWorkingSetSize; +extern cl::opt<unsigned> MaxNumVTableAnnotations; + // Walk through the operands of a given User via worklist iteration and populate // the set of GlobalValue references encountered. Invoked either on an // Instruction or a GlobalVariable (which walks its initializer). @@ -124,6 +126,24 @@ static bool findRefEdges(ModuleSummaryIndex &Index, const User *CurUser, Worklist.push_back(Operand); } } + + const Instruction *I = dyn_cast<Instruction>(CurUser); + if (I) { + uint32_t ActualNumValueData = 0; + uint64_t TotalCount = 0; + // MaxNumVTableAnnotations is the maximum number of vtables annotated on + // the instruction. + auto ValueDataArray = + getValueProfDataFromInst(*I, IPVK_VTableTarget, MaxNumVTableAnnotations, + ActualNumValueData, TotalCount); + + if (ValueDataArray.get()) { + for (uint32_t j = 0; j < ActualNumValueData; j++) { + RefEdges.insert(Index.getOrInsertValueInfo(/* VTableGUID = */ + ValueDataArray[j].Value)); + } + } + } return HasBlockAddress; } diff --git a/llvm/lib/Bitcode/Writer/BitcodeWriter.cpp b/llvm/lib/Bitcode/Writer/BitcodeWriter.cpp index 13be0b0..7686e32 100644 --- a/llvm/lib/Bitcode/Writer/BitcodeWriter.cpp +++ b/llvm/lib/Bitcode/Writer/BitcodeWriter.cpp @@ -199,7 +199,7 @@ public: for (const auto &GUIDSummaryLists : *Index) // Examine all summaries for this GUID. for (auto &Summary : GUIDSummaryLists.second.SummaryList) - if (auto FS = dyn_cast<FunctionSummary>(Summary.get())) + if (auto FS = dyn_cast<FunctionSummary>(Summary.get())) { // For each call in the function summary, see if the call // is to a GUID (which means it is for an indirect call, // otherwise we would have a Value for it). If so, synthesize @@ -207,6 +207,15 @@ public: for (auto &CallEdge : FS->calls()) if (!CallEdge.first.haveGVs() || !CallEdge.first.getValue()) assignValueId(CallEdge.first.getGUID()); + + // For each referenced variables in the function summary, see if the + // variable is represented by a GUID (as opposed to a symbol to + // declarations or definitions in the module). If so, synthesize a + // value id. + for (auto &RefEdge : FS->refs()) + if ((!RefEdge.haveGVs() || !RefEdge.getValue())) + assignValueId(RefEdge.getGUID()); + } } protected: @@ -4071,7 +4080,7 @@ void ModuleBitcodeWriterBase::writePerModuleFunctionSummaryRecord( NameVals.push_back(SpecialRefCnts.second); // worefcnt for (auto &RI : FS->refs()) - NameVals.push_back(VE.getValueID(RI.getValue())); + NameVals.push_back(getValueId(RI)); const bool UseRelBFRecord = WriteRelBFToSummary && !F.hasProfileData() && diff --git a/llvm/lib/ProfileData/InstrProf.cpp b/llvm/lib/ProfileData/InstrProf.cpp index 2640027..3b05faf 100644 --- a/llvm/lib/ProfileData/InstrProf.cpp +++ b/llvm/lib/ProfileData/InstrProf.cpp @@ -219,6 +219,12 @@ cl::opt<bool> DoInstrProfNameCompression( "enable-name-compression", cl::desc("Enable name/filename string compression"), cl::init(true)); +cl::opt<bool> EnableVTableValueProfiling( + "enable-vtable-value-profiling", cl::init(false), + cl::desc("If true, the virtual table address will be instrumented to know " + "the types of a C++ pointer. The information is used in indirect " + "call promotion to do selective vtable-based comparison.")); + std::string getInstrProfSectionName(InstrProfSectKind IPSK, Triple::ObjectFormatType OF, bool AddSegmentInfo) { @@ -319,21 +325,20 @@ static std::optional<std::string> lookupPGONameFromMetadata(MDNode *MD) { return {}; } -// Returns the PGO object name. This function has some special handling -// when called in LTO optimization. The following only applies when calling in -// LTO passes (when \c InLTO is true): LTO's internalization privatizes many -// global linkage symbols. This happens after value profile annotation, but -// those internal linkage functions should not have a source prefix. -// Additionally, for ThinLTO mode, exported internal functions are promoted -// and renamed. We need to ensure that the original internal PGO name is -// used when computing the GUID that is compared against the profiled GUIDs. -// To differentiate compiler generated internal symbols from original ones, -// PGOFuncName meta data are created and attached to the original internal -// symbols in the value profile annotation step -// (PGOUseFunc::annotateIndirectCallSites). If a symbol does not have the meta -// data, its original linkage must be non-internal. -static std::string getIRPGOObjectName(const GlobalObject &GO, bool InLTO, - MDNode *PGONameMetadata) { +std::string getIRPGOObjectName(const GlobalObject &GO, bool InLTO, + MDNode *PGONameMetadata) { + // The following only applies when calling in + // LTO passes (when \c InLTO is true): LTO's internalization privatizes many + // global linkage symbols. This happens after value profile annotation, but + // those internal linkage functions should not have a source prefix. + // Additionally, for ThinLTO mode, exported internal functions are promoted + // and renamed. We need to ensure that the original internal PGO name is + // used when computing the GUID that is compared against the profiled GUIDs. + // To differentiate compiler generated internal symbols from original ones, + // PGOFuncName meta data are created and attached to the original internal + // symbols in the value profile annotation step + // (PGOUseFunc::annotateIndirectCallSites). If a symbol does not have the meta + // data, its original linkage must be non-internal. if (!InLTO) { auto FileName = getStrippedSourceFileName(GO); return getIRPGONameForGlobalObject(GO, GO.getLinkage(), FileName); @@ -378,6 +383,13 @@ std::string getPGOFuncName(const Function &F, bool InLTO, uint64_t Version) { return getPGOFuncName(F.getName(), GlobalValue::ExternalLinkage, ""); } +std::string getPGOName(const GlobalVariable &V, bool InLTO) { + // PGONameMetadata should be set by compiler at profile use time + // and read by symtab creation to look up symbols corresponding to + // a MD5 hash. + return getIRPGOObjectName(V, InLTO, nullptr /* PGONameMetadata */); +} + // See getIRPGOFuncName() for a discription of the format. std::pair<StringRef, StringRef> getParsedIRPGOFuncName(StringRef IRPGOFuncName) { @@ -460,11 +472,48 @@ Error InstrProfSymtab::create(Module &M, bool InLTO) { if (Error E = addFuncWithName(F, getPGOFuncName(F, InLTO))) return E; } + + SmallVector<MDNode *, 2> Types; + for (GlobalVariable &G : M.globals()) { + if (!G.hasName()) + continue; + Types.clear(); + G.getMetadata(LLVMContext::MD_type, Types); + if (!Types.empty()) { + if (Error E = addVTableWithName( + G, getIRPGOObjectName(G, InLTO, /* PGONameMetadata */ nullptr))) + return E; + } + } Sorted = false; finalizeSymtab(); return Error::success(); } +Error InstrProfSymtab::addVTableWithName(GlobalVariable &VTable, + StringRef VTablePGOName) { + if (Error E = addVTableName(VTablePGOName)) + return E; + + MD5VTableMap.emplace_back(GlobalValue::getGUID(VTablePGOName), &VTable); + + // NOTE: `-funique-internal-linkage-names` doesn't uniqufy vtables, so no + // need to check ".__uniq." + + // If a local-linkage vtable is promoted to have external linkage in ThinLTO, + // it will have `.llvm.` in its name. Use the name before externalization. + StringRef CanonicalName = + getCanonicalName(VTablePGOName, /* MayHaveUniqueSuffix= */ false); + if (CanonicalName != VTablePGOName) { + if (Error E = addVTableName(CanonicalName)) + return E; + + MD5VTableMap.emplace_back(GlobalValue::getGUID(CanonicalName), &VTable); + } + + return Error::success(); +} + /// \c NameStrings is a string composed of one of more possibly encoded /// sub-strings. The substrings are separated by 0 or more zero bytes. This /// method decodes the string and calls `NameCallback` for each substring. @@ -518,38 +567,95 @@ Error InstrProfSymtab::create(StringRef NameStrings) { std::bind(&InstrProfSymtab::addFuncName, this, std::placeholders::_1)); } -Error InstrProfSymtab::addFuncWithName(Function &F, StringRef PGOFuncName) { - if (Error E = addFuncName(PGOFuncName)) +Error InstrProfSymtab::create(StringRef FuncNameStrings, + StringRef VTableNameStrings) { + if (Error E = readAndDecodeStrings(FuncNameStrings, + std::bind(&InstrProfSymtab::addFuncName, + this, std::placeholders::_1))) return E; - MD5FuncMap.emplace_back(Function::getGUID(PGOFuncName), &F); + + return readAndDecodeStrings( + VTableNameStrings, + std::bind(&InstrProfSymtab::addVTableName, this, std::placeholders::_1)); +} + +Error InstrProfSymtab::initVTableNamesFromCompressedStrings( + StringRef CompressedVTableStrings) { + return readAndDecodeStrings( + CompressedVTableStrings, + std::bind(&InstrProfSymtab::addVTableName, this, std::placeholders::_1)); +} + +StringRef InstrProfSymtab::getCanonicalName(StringRef PGOName, + bool MayHaveUniqueSuffix) { + size_t pos = 0; // In ThinLTO, local function may have been promoted to global and have // suffix ".llvm." added to the function name. We need to add the // stripped function name to the symbol table so that we can find a match // from profile. // - // We may have other suffixes similar as ".llvm." which are needed to - // be stripped before the matching, but ".__uniq." suffix which is used - // to differentiate internal linkage functions in different modules - // should be kept. Now this is the only suffix with the pattern ".xxx" - // which is kept before matching. - const std::string UniqSuffix = ".__uniq."; - auto pos = PGOFuncName.find(UniqSuffix); + // ".__uniq." suffix is used to differentiate internal linkage functions in + // different modules and should be kept. Now this is the only suffix with the + // pattern ".xxx" which is kept before matching. In other words, other + // suffixes similar as + // ".llvm." will be stripped. + if (MayHaveUniqueSuffix) { + const std::string UniqSuffix = ".__uniq."; + pos = PGOName.find(UniqSuffix); + if (pos != StringRef::npos) + pos += UniqSuffix.length(); + else + pos = 0; + } + // Search '.' after ".__uniq." if ".__uniq." exists, otherwise // search '.' from the beginning. - if (pos != std::string::npos) - pos += UniqSuffix.length(); - else - pos = 0; - pos = PGOFuncName.find('.', pos); - if (pos != std::string::npos && pos != 0) { - StringRef OtherFuncName = PGOFuncName.substr(0, pos); - if (Error E = addFuncName(OtherFuncName)) + pos = PGOName.find('.', pos); + if (pos != StringRef::npos && pos != 0) + return PGOName.substr(0, pos); + + return PGOName; +} + +Error InstrProfSymtab::addFuncWithName(Function &F, StringRef PGOFuncName) { + if (Error E = addFuncName(PGOFuncName)) + return E; + MD5FuncMap.emplace_back(Function::getGUID(PGOFuncName), &F); + + StringRef CanonicalName = + getCanonicalName(PGOFuncName, /* MayHaveUniqueSuffix= */ true); + + if (CanonicalName != PGOFuncName) { + if (Error E = addFuncName(CanonicalName)) return E; - MD5FuncMap.emplace_back(Function::getGUID(OtherFuncName), &F); + MD5FuncMap.emplace_back(Function::getGUID(CanonicalName), &F); } + return Error::success(); } +uint64_t InstrProfSymtab::getVTableHashFromAddress(uint64_t Address) { + finalizeSymtab(); + auto It = lower_bound( + VTableAddrRangeToMD5Map, Address, + [](std::pair<std::pair<uint64_t, uint64_t>, uint64_t> VTableRangeAddr, + uint64_t Addr) { + // Find the first address range of which end address is larger than + // `Addr`. Smaller-than-or-equal-to is used because the profiled address + // within a vtable should be [start-address, end-address). + return VTableRangeAddr.first.second <= Addr; + }); + + // Returns the MD5 hash if Address is within the address range of an entry. + if (It != VTableAddrRangeToMD5Map.end() && It->first.first <= Address) { + return It->second; + } + // The virtual table address collected from value profiler could be defined + // in another module that is not instrumented. Force the value to be 0 in + // this case. + return 0; +} + uint64_t InstrProfSymtab::getFunctionHashFromAddress(uint64_t Address) { finalizeSymtab(); auto It = partition_point(AddrToMD5Map, [=](std::pair<uint64_t, uint64_t> A) { @@ -626,6 +732,17 @@ Error collectPGOFuncNameStrings(ArrayRef<GlobalVariable *> NameVars, NameStrs, compression::zlib::isAvailable() && doCompression, Result); } +Error collectVTableStrings(ArrayRef<GlobalVariable *> VTables, + std::string &Result, bool doCompression) { + std::vector<std::string> VTableNameStrs; + for (auto *VTable : VTables) { + VTableNameStrs.push_back(getPGOName(*VTable)); + } + return collectGlobalObjectNameStrings( + VTableNameStrs, compression::zlib::isAvailable() && doCompression, + Result); +} + void InstrProfRecord::accumulateCounts(CountSumOrPercent &Sum) const { uint64_t FuncSum = 0; Sum.NumEntries += Counts.size(); @@ -888,6 +1005,9 @@ uint64_t InstrProfRecord::remapValue(uint64_t Value, uint32_t ValueKind, if (ValueKind == IPVK_IndirectCallTarget) return SymTab->getFunctionHashFromAddress(Value); + if (ValueKind == IPVK_VTableTarget) + return SymTab->getVTableHashFromAddress(Value); + return Value; } @@ -1178,9 +1298,8 @@ void annotateValueSite(Module &M, Instruction &Inst, } void annotateValueSite(Module &M, Instruction &Inst, - ArrayRef<InstrProfValueData> VDs, - uint64_t Sum, InstrProfValueKind ValueKind, - uint32_t MaxMDCount) { + ArrayRef<InstrProfValueData> VDs, uint64_t Sum, + InstrProfValueKind ValueKind, uint32_t MaxMDCount) { LLVMContext &Ctx = M.getContext(); MDBuilder MDHelper(Ctx); SmallVector<Metadata *, 3> Vals; @@ -1206,46 +1325,44 @@ void annotateValueSite(Module &M, Instruction &Inst, Inst.setMetadata(LLVMContext::MD_prof, MDNode::get(Ctx, Vals)); } -bool getValueProfDataFromInst(const Instruction &Inst, - InstrProfValueKind ValueKind, - uint32_t MaxNumValueData, - InstrProfValueData ValueData[], - uint32_t &ActualNumValueData, uint64_t &TotalC, - bool GetNoICPValue) { +MDNode *mayHaveValueProfileOfKind(const Instruction &Inst, + InstrProfValueKind ValueKind) { MDNode *MD = Inst.getMetadata(LLVMContext::MD_prof); if (!MD) - return false; - - unsigned NOps = MD->getNumOperands(); + return nullptr; - if (NOps < 5) - return false; + if (MD->getNumOperands() < 5) + return nullptr; - // Operand 0 is a string tag "VP": MDString *Tag = cast<MDString>(MD->getOperand(0)); - if (!Tag) - return false; - - if (!Tag->getString().equals("VP")) - return false; + if (!Tag || !Tag->getString().equals("VP")) + return nullptr; // Now check kind: ConstantInt *KindInt = mdconst::dyn_extract<ConstantInt>(MD->getOperand(1)); if (!KindInt) - return false; + return nullptr; if (KindInt->getZExtValue() != ValueKind) - return false; + return nullptr; + return MD; +} + +static bool getValueProfDataFromInstImpl(const MDNode *const MD, + const uint32_t MaxNumDataWant, + InstrProfValueData ValueData[], + uint32_t &ActualNumValueData, + uint64_t &TotalC, bool GetNoICPValue) { + const unsigned NOps = MD->getNumOperands(); // Get total count ConstantInt *TotalCInt = mdconst::dyn_extract<ConstantInt>(MD->getOperand(2)); if (!TotalCInt) return false; TotalC = TotalCInt->getZExtValue(); - ActualNumValueData = 0; for (unsigned I = 3; I < NOps; I += 2) { - if (ActualNumValueData >= MaxNumValueData) + if (ActualNumValueData >= MaxNumDataWant) break; ConstantInt *Value = mdconst::dyn_extract<ConstantInt>(MD->getOperand(I)); ConstantInt *Count = @@ -1262,6 +1379,36 @@ bool getValueProfDataFromInst(const Instruction &Inst, return true; } +std::unique_ptr<InstrProfValueData[]> +getValueProfDataFromInst(const Instruction &Inst, InstrProfValueKind ValueKind, + uint32_t MaxNumValueData, uint32_t &ActualNumValueData, + uint64_t &TotalC, bool GetNoICPValue) { + MDNode *MD = mayHaveValueProfileOfKind(Inst, ValueKind); + if (!MD) + return nullptr; + auto ValueDataArray = std::make_unique<InstrProfValueData[]>(MaxNumValueData); + if (!getValueProfDataFromInstImpl(MD, MaxNumValueData, ValueDataArray.get(), + ActualNumValueData, TotalC, GetNoICPValue)) + return nullptr; + return ValueDataArray; +} + +// FIXME: Migrate existing callers to the function above that returns an +// array. +bool getValueProfDataFromInst(const Instruction &Inst, + InstrProfValueKind ValueKind, + uint32_t MaxNumValueData, + InstrProfValueData ValueData[], + uint32_t &ActualNumValueData, uint64_t &TotalC, + bool GetNoICPValue) { + MDNode *MD = mayHaveValueProfileOfKind(Inst, ValueKind); + if (!MD) + return false; + return getValueProfDataFromInstImpl(MD, MaxNumValueData, ValueData, + ActualNumValueData, TotalC, + GetNoICPValue); +} + MDNode *getPGOFuncNameMetadata(const Function &F) { return F.getMetadata(getPGOFuncNameMetadataName()); } @@ -1278,8 +1425,8 @@ void createPGOFuncNameMetadata(Function &F, StringRef PGOFuncName) { F.setMetadata(getPGOFuncNameMetadataName(), N); } -bool needsComdatForCounter(const Function &F, const Module &M) { - if (F.hasComdat()) +bool needsComdatForCounter(const GlobalValue &GV, const Module &M) { + if (GV.hasComdat()) return true; if (!Triple(M.getTargetTriple()).supportsCOMDAT()) @@ -1295,7 +1442,7 @@ bool needsComdatForCounter(const Function &F, const Module &M) { // available_externally functions will end up being duplicated in raw profile // data. This can result in distorted profile as the counts of those dups // will be accumulated by the profile merger. - GlobalValue::LinkageTypes Linkage = F.getLinkage(); + GlobalValue::LinkageTypes Linkage = GV.getLinkage(); if (Linkage != GlobalValue::ExternalWeakLinkage && Linkage != GlobalValue::AvailableExternallyLinkage) return false; @@ -1451,7 +1598,7 @@ void OverlapStats::dump(raw_fd_ostream &OS) const { for (unsigned I = 0; I < IPVK_Last - IPVK_First + 1; I++) { if (Base.ValueCounts[I] < 1.0f && Test.ValueCounts[I] < 1.0f) continue; - char ProfileKindName[20]; + char ProfileKindName[20] = {0}; switch (I) { case IPVK_IndirectCallTarget: strncpy(ProfileKindName, "IndirectCall", 19); @@ -1459,6 +1606,9 @@ void OverlapStats::dump(raw_fd_ostream &OS) const { case IPVK_MemOPSize: strncpy(ProfileKindName, "MemOP", 19); break; + case IPVK_VTableTarget: + strncpy(ProfileKindName, "VTable", 19); + break; default: snprintf(ProfileKindName, 19, "VP[%d]", I); break; @@ -1523,9 +1673,12 @@ Expected<Header> Header::readFromBuffer(const unsigned char *Buffer) { // When a new field is added in the header add a case statement here to // populate it. static_assert( - IndexedInstrProf::ProfVersion::CurrentVersion == Version11, + IndexedInstrProf::ProfVersion::CurrentVersion == Version12, "Please update the reading code below if a new field has been added, " "if not add a case statement to fall through to the latest version."); + case 12ull: + H.VTableNamesOffset = read(Buffer, offsetOf(&Header::VTableNamesOffset)); + [[fallthrough]]; case 11ull: [[fallthrough]]; case 10ull: @@ -1551,10 +1704,13 @@ size_t Header::size() const { // When a new field is added to the header add a case statement here to // compute the size as offset of the new field + size of the new field. This // relies on the field being added to the end of the list. - static_assert(IndexedInstrProf::ProfVersion::CurrentVersion == Version11, + static_assert(IndexedInstrProf::ProfVersion::CurrentVersion == Version12, "Please update the size computation below if a new field has " "been added to the header, if not add a case statement to " "fall through to the latest version."); + case 12ull: + return offsetOf(&Header::VTableNamesOffset) + + sizeof(Header::VTableNamesOffset); case 11ull: [[fallthrough]]; case 10ull: diff --git a/llvm/lib/ProfileData/InstrProfReader.cpp b/llvm/lib/ProfileData/InstrProfReader.cpp index 0d8d43d..4ef6823 100644 --- a/llvm/lib/ProfileData/InstrProfReader.cpp +++ b/llvm/lib/ProfileData/InstrProfReader.cpp @@ -366,6 +366,14 @@ TextInstrProfReader::readValueProfileData(InstrProfRecord &Record) { return E; Value = IndexedInstrProf::ComputeHash(VD.first); } + } else if (ValueKind == IPVK_VTableTarget) { + if (InstrProfSymtab::isExternalSymbol(VD.first)) { + Value = 0; + } else { + if (Error E = Symtab->addVTableName(VD.first)) + return E; + Value = IndexedInstrProf::ComputeHash(VD.first); + } } else { READ_NUM(VD.first, Value); } @@ -533,7 +541,8 @@ Error RawInstrProfReader<IntPtrT>::readNextHeader(const char *CurrentPos) { template <class IntPtrT> Error RawInstrProfReader<IntPtrT>::createSymtab(InstrProfSymtab &Symtab) { - if (Error E = Symtab.create(StringRef(NamesStart, NamesEnd - NamesStart))) + if (Error E = Symtab.create(StringRef(NamesStart, NamesEnd - NamesStart), + StringRef(VNamesStart, VNamesEnd - VNamesStart))) return error(std::move(E)); for (const RawInstrProf::ProfileData<IntPtrT> *I = Data; I != DataEnd; ++I) { const IntPtrT FPtr = swap(I->FunctionPointer); @@ -541,6 +550,21 @@ Error RawInstrProfReader<IntPtrT>::createSymtab(InstrProfSymtab &Symtab) { continue; Symtab.mapAddress(FPtr, swap(I->NameRef)); } + + if (VTableBegin != nullptr && VTableEnd != nullptr) { + for (const RawInstrProf::VTableProfileData<IntPtrT> *I = VTableBegin; + I != VTableEnd; ++I) { + const IntPtrT VPtr = swap(I->VTablePointer); + if (!VPtr) + continue; + // Map both begin and end address to the name hash, since the instrumented + // address could be somewhere in the middle. + // VPtr is of type uint32_t or uint64_t so 'VPtr + I->VTableSize' marks + // the end of vtable address. + Symtab.mapVTableAddress(VPtr, VPtr + swap(I->VTableSize), + swap(I->VTableNameHash)); + } + } return success(); } @@ -582,10 +606,17 @@ Error RawInstrProfReader<IntPtrT>::readHeader( auto NumBitmapBytes = swap(Header.NumBitmapBytes); auto PaddingBytesAfterBitmapBytes = swap(Header.PaddingBytesAfterBitmapBytes); auto NamesSize = swap(Header.NamesSize); + auto VTableNameSize = swap(Header.VNamesSize); + auto NumVTables = swap(Header.NumVTables); ValueKindLast = swap(Header.ValueKindLast); auto DataSize = NumData * sizeof(RawInstrProf::ProfileData<IntPtrT>); - auto PaddingSize = getNumPaddingBytes(NamesSize); + auto PaddingBytesAfterNames = getNumPaddingBytes(NamesSize); + auto PaddingBytesAfterVTableNames = getNumPaddingBytes(VTableNameSize); + + auto VTableSectionSize = + NumVTables * sizeof(RawInstrProf::VTableProfileData<IntPtrT>); + auto PaddingBytesAfterVTableProfData = getNumPaddingBytes(VTableSectionSize); // Profile data starts after profile header and binary ids if exist. ptrdiff_t DataOffset = sizeof(RawInstrProf::Header) + BinaryIdSize; @@ -594,7 +625,12 @@ Error RawInstrProfReader<IntPtrT>::readHeader( CountersOffset + CountersSize + PaddingBytesAfterCounters; ptrdiff_t NamesOffset = BitmapOffset + NumBitmapBytes + PaddingBytesAfterBitmapBytes; - ptrdiff_t ValueDataOffset = NamesOffset + NamesSize + PaddingSize; + ptrdiff_t VTableProfDataOffset = + NamesOffset + NamesSize + PaddingBytesAfterNames; + ptrdiff_t VTableNameOffset = VTableProfDataOffset + VTableSectionSize + + PaddingBytesAfterVTableProfData; + ptrdiff_t ValueDataOffset = + VTableNameOffset + VTableNameSize + PaddingBytesAfterVTableNames; auto *Start = reinterpret_cast<const char *>(&Header); if (Start + ValueDataOffset > DataBuffer->getBufferEnd()) @@ -614,8 +650,14 @@ Error RawInstrProfReader<IntPtrT>::readHeader( Data = reinterpret_cast<const RawInstrProf::ProfileData<IntPtrT> *>( Start + DataOffset); DataEnd = Data + NumData; + VTableBegin = + reinterpret_cast<const RawInstrProf::VTableProfileData<IntPtrT> *>( + Start + VTableProfDataOffset); + VTableEnd = VTableBegin + NumVTables; NamesStart = Start + NamesOffset; NamesEnd = NamesStart + NamesSize; + VNamesStart = Start + VTableNameOffset; + VNamesEnd = VNamesStart + VTableNameSize; } CountersStart = Start + CountersOffset; @@ -1260,6 +1302,19 @@ Error IndexedInstrProfReader::readHeader() { "corrupted binary ids"); } + if (GET_VERSION(Header->formatVersion()) >= 12) { + uint64_t VTableNamesOffset = + endian::byte_swap<uint64_t, llvm::endianness::little>( + Header->VTableNamesOffset); + const unsigned char *Ptr = Start + VTableNamesOffset; + + CompressedVTableNamesLen = + support::endian::readNext<uint64_t, llvm::endianness::little, + unaligned>(Ptr); + + VTableNamePtr = (const char *)Ptr; + } + if (GET_VERSION(Header->formatVersion()) >= 10 && Header->formatVersion() & VARIANT_MASK_TEMPORAL_PROF) { uint64_t TemporalProfTracesOffset = @@ -1319,7 +1374,16 @@ InstrProfSymtab &IndexedInstrProfReader::getSymtab() { if (Symtab) return *Symtab; - std::unique_ptr<InstrProfSymtab> NewSymtab = std::make_unique<InstrProfSymtab>(); + std::unique_ptr<InstrProfSymtab> NewSymtab = + std::make_unique<InstrProfSymtab>(); + + if (Error E = NewSymtab->initVTableNamesFromCompressedStrings( + StringRef(VTableNamePtr, CompressedVTableNamesLen))) { + auto [ErrCode, Msg] = InstrProfError::take(std::move(E)); + consumeError(error(ErrCode, Msg)); + } + + // finalizeSymtab is called inside populateSymtab. if (Error E = Index->populateSymtab(*NewSymtab)) { auto [ErrCode, Msg] = InstrProfError::take(std::move(E)); consumeError(error(ErrCode, Msg)); diff --git a/llvm/lib/ProfileData/InstrProfWriter.cpp b/llvm/lib/ProfileData/InstrProfWriter.cpp index d65f8fe..7592c0f 100644 --- a/llvm/lib/ProfileData/InstrProfWriter.cpp +++ b/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/Compression.h" #include "llvm/Support/Endian.h" #include "llvm/Support/EndianStream.h" #include "llvm/Support/Error.h" @@ -455,12 +456,13 @@ Error InstrProfWriter::writeImpl(ProfOStream &OS) { Header.MemProfOffset = 0; Header.BinaryIdOffset = 0; Header.TemporalProfTracesOffset = 0; + Header.VTableNamesOffset = 0; int N = sizeof(IndexedInstrProf::Header) / sizeof(uint64_t); // Only write out all the fields except 'HashOffset', 'MemProfOffset', - // 'BinaryIdOffset' and `TemporalProfTracesOffset`. We need to remember the - // offset of these fields to allow back patching later. - for (int I = 0; I < N - 4; I++) + // 'BinaryIdOffset', `TemporalProfTracesOffset` and `VTableNamesOffset`. We + // need to remember the offset of these fields to allow back patching later. + for (int I = 0; I < N - 5; I++) OS.write(reinterpret_cast<uint64_t *>(&Header)[I]); // Save the location of Header.HashOffset field in \c OS. @@ -484,6 +486,9 @@ Error InstrProfWriter::writeImpl(ProfOStream &OS) { uint64_t TemporalProfTracesOffset = OS.tell(); OS.write(0); + uint64_t VTableNamesOffset = OS.tell(); + OS.write(0); + // Reserve space to write profile summary data. uint32_t NumEntries = ProfileSummaryBuilder::DefaultCutoffs.size(); uint32_t SummarySize = Summary::getSize(Summary::NumKinds, NumEntries); @@ -604,6 +609,43 @@ Error InstrProfWriter::writeImpl(ProfOStream &OS) { OS.writeByte(0); } + // if version >= the version with vtable profile metadata. + uint64_t VTableNamesSectionStart = 0; + if (IndexedInstrProf::ProfVersion::CurrentVersion >= 12) { + VTableNamesSectionStart = OS.tell(); + + std::string CompressedVTableNames; + + std::vector<std::string> VTableNameStrs; + for (const auto &VTableName : VTableNames.keys()) { + VTableNameStrs.push_back(VTableName.str()); + } + + if (!VTableNameStrs.empty()) { + if (Error E = collectGlobalObjectNameStrings( + VTableNameStrs, compression::zlib::isAvailable(), + CompressedVTableNames)) + return E; + } + + uint64_t CompressedStringLen = CompressedVTableNames.length(); + + // Record the length of compressed string. + OS.write(CompressedStringLen); + + // Write the chars in compressed strings. + for (auto &c : CompressedVTableNames) + OS.writeByte(static_cast<uint8_t>(c)); + + // Pad up to a multiple of 8. + // InstrProfReader could read bytes according to 'CompressedStringLen'. + uint64_t PaddedLength = alignTo(CompressedStringLen, 8); + + for (uint64_t K = CompressedStringLen; K < PaddedLength; K++) { + OS.writeByte(0); + } + } + uint64_t TemporalProfTracesSectionStart = 0; if (static_cast<bool>(ProfileKind & InstrProfKind::TemporalProfile)) { TemporalProfTracesSectionStart = OS.tell(); @@ -647,6 +689,7 @@ Error InstrProfWriter::writeImpl(ProfOStream &OS) { // Patch the Header.TemporalProfTracesOffset (=0 for profiles without // traces). {TemporalProfTracesOffset, &TemporalProfTracesSectionStart, 1}, + {VTableNamesOffset, &VTableNamesSectionStart, 1}, // Patch the summary data. {SummaryOffset, reinterpret_cast<uint64_t *>(TheSummary.get()), (int)(SummarySize / sizeof(uint64_t))}, @@ -699,7 +742,8 @@ Error InstrProfWriter::validateRecord(const InstrProfRecord &Func) { std::unique_ptr<InstrProfValueData[]> VD = Func.getValueForSite(VK, S); DenseSet<uint64_t> SeenValues; for (uint32_t I = 0; I < ND; I++) - if ((VK != IPVK_IndirectCallTarget) && !SeenValues.insert(VD[I].Value).second) + if ((VK != IPVK_IndirectCallTarget && VK != IPVK_VTableTarget) && + !SeenValues.insert(VD[I].Value).second) return make_error<InstrProfError>(instrprof_error::invalid_prof); } } @@ -747,7 +791,7 @@ void InstrProfWriter::writeRecordInText(StringRef Name, uint64_t Hash, OS << ND << "\n"; std::unique_ptr<InstrProfValueData[]> VD = Func.getValueForSite(VK, S); for (uint32_t I = 0; I < ND; I++) { - if (VK == IPVK_IndirectCallTarget) + if (VK == IPVK_IndirectCallTarget || VK == IPVK_VTableTarget) OS << Symtab.getFuncOrVarNameIfDefined(VD[I].Value) << ":" << VD[I].Count << "\n"; else @@ -786,6 +830,11 @@ Error InstrProfWriter::writeText(raw_fd_ostream &OS) { } } + for (const auto &VTableName : VTableNames) { + if (Error E = Symtab.addVTableName(VTableName.getKey())) + return E; + } + if (static_cast<bool>(ProfileKind & InstrProfKind::TemporalProfile)) writeTextTemporalProfTraceData(OS, Symtab); diff --git a/llvm/lib/Transforms/Instrumentation/IndirectCallPromotion.cpp b/llvm/lib/Transforms/Instrumentation/IndirectCallPromotion.cpp index 7344fea..6a44a32 100644 --- a/llvm/lib/Transforms/Instrumentation/IndirectCallPromotion.cpp +++ b/llvm/lib/Transforms/Instrumentation/IndirectCallPromotion.cpp @@ -136,11 +136,13 @@ private: const CallBase &CB, const ArrayRef<InstrProfValueData> &ValueDataRef, uint64_t TotalCount, uint32_t NumCandidates); - // Promote a list of targets for one indirect-call callsite. Return - // the number of promotions. - uint32_t tryToPromote(CallBase &CB, - const std::vector<PromotionCandidate> &Candidates, - uint64_t &TotalCount); + // Promote a list of targets for one indirect-call callsite by comparing + // indirect callee with functions. Returns true if there are IR + // transformations and false otherwise. + bool tryToPromoteWithFuncCmp( + CallBase &CB, const std::vector<PromotionCandidate> &Candidates, + uint64_t TotalCount, ArrayRef<InstrProfValueData> ICallProfDataRef, + uint32_t NumCandidates); public: IndirectCallPromoter(Function &Func, InstrProfSymtab *Symtab, bool SamplePGO, @@ -273,9 +275,10 @@ CallBase &llvm::pgo::promoteIndirectCall(CallBase &CB, Function *DirectCallee, } // Promote indirect-call to conditional direct-call for one callsite. -uint32_t IndirectCallPromoter::tryToPromote( +bool IndirectCallPromoter::tryToPromoteWithFuncCmp( CallBase &CB, const std::vector<PromotionCandidate> &Candidates, - uint64_t &TotalCount) { + uint64_t TotalCount, ArrayRef<InstrProfValueData> ICallProfDataRef, + uint32_t NumCandidates) { uint32_t NumPromoted = 0; for (const auto &C : Candidates) { @@ -287,7 +290,18 @@ uint32_t IndirectCallPromoter::tryToPromote( NumOfPGOICallPromotion++; NumPromoted++; } - return NumPromoted; + + const bool Changed = (NumPromoted != 0); + + if (Changed) { + CB.setMetadata(LLVMContext::MD_prof, nullptr); + + if (TotalCount != 0) + annotateValueSite(*F.getParent(), CB, ICallProfDataRef.slice(NumPromoted), + TotalCount, IPVK_IndirectCallTarget, NumCandidates); + } + + return Changed; } // Traverse all the indirect-call callsite and get the value profile @@ -305,19 +319,8 @@ bool IndirectCallPromoter::processFunction(ProfileSummaryInfo *PSI) { continue; auto PromotionCandidates = getPromotionCandidatesForCallSite( *CB, ICallProfDataRef, TotalCount, NumCandidates); - uint32_t NumPromoted = tryToPromote(*CB, PromotionCandidates, TotalCount); - if (NumPromoted == 0) - continue; - - Changed = true; - // Adjust the MD.prof metadata. First delete the old one. - CB->setMetadata(LLVMContext::MD_prof, nullptr); - // If all promoted, we don't need the MD.prof metadata. - if (TotalCount == 0 || NumPromoted == NumVals) - continue; - // Otherwise we need update with the un-promoted records back. - annotateValueSite(*F.getParent(), *CB, ICallProfDataRef.slice(NumPromoted), - TotalCount, IPVK_IndirectCallTarget, NumCandidates); + Changed |= tryToPromoteWithFuncCmp(*CB, PromotionCandidates, TotalCount, + ICallProfDataRef, NumCandidates); } return Changed; } diff --git a/llvm/lib/Transforms/Instrumentation/InstrProfiling.cpp b/llvm/lib/Transforms/Instrumentation/InstrProfiling.cpp index a19b140..49978da 100644 --- a/llvm/lib/Transforms/Instrumentation/InstrProfiling.cpp +++ b/llvm/lib/Transforms/Instrumentation/InstrProfiling.cpp @@ -64,6 +64,9 @@ using namespace llvm; #define DEBUG_TYPE "instrprof" namespace llvm { +// Command line option to enable vtable value profiling. Defined in +// ProfileData/InstrProf.cpp: -enable-vtable-value-profiling= +extern cl::opt<bool> EnableVTableValueProfiling; // TODO: Remove -debug-info-correlate in next LLVM release, in favor of // -profile-correlate=debug-info. cl::opt<bool> DebugInfoCorrelate( @@ -196,12 +199,18 @@ private: PerFunctionProfileData() = default; }; DenseMap<GlobalVariable *, PerFunctionProfileData> ProfileDataMap; + // Key is virtual table variable, value is 'VTableProfData' in the form of + // GlobalVariable. + DenseMap<GlobalVariable *, GlobalVariable *> VTableDataMap; /// If runtime relocation is enabled, this maps functions to the load /// instruction that produces the profile relocation bias. DenseMap<const Function *, LoadInst *> FunctionToProfileBiasMap; std::vector<GlobalValue *> CompilerUsedVars; std::vector<GlobalValue *> UsedVars; std::vector<GlobalVariable *> ReferencedNames; + // The list of virtual table variables of which the VTableProfData is + // collected. + std::vector<GlobalVariable *> ReferencedVTables; GlobalVariable *NamesVar = nullptr; size_t NamesSize = 0; @@ -294,9 +303,15 @@ private: /// Create INSTR_PROF_DATA variable for counters and bitmaps. void createDataVariable(InstrProfCntrInstBase *Inc); + /// Get the counters for virtual table values, creating them if necessary. + void getOrCreateVTableProfData(GlobalVariable *GV); + /// Emit the section with compressed function names. void emitNameData(); + /// Emit the section with compressed vtable names. + void emitVTableNames(); + /// Emit value nodes section for value profiling. void emitVNodes(); @@ -740,6 +755,15 @@ bool InstrLowerer::lower() { } } + if (EnableVTableValueProfiling) { + for (GlobalVariable &GV : M.globals()) { + // Global variables with type metadata are virtual table variables. + if (GV.hasMetadata(LLVMContext::MD_type)) { + getOrCreateVTableProfData(&GV); + } + } + } + for (Function &F : M) MadeChange |= lowerIntrinsics(&F); @@ -753,6 +777,7 @@ bool InstrLowerer::lower() { emitVNodes(); emitNameData(); + emitVTableNames(); // Emit runtime hook for the cases where the target does not unconditionally // require pulling in profile runtime, and coverage is enabled on code that is @@ -1220,6 +1245,129 @@ void InstrLowerer::maybeSetComdat(GlobalVariable *GV, Function *Fn, GV->setLinkage(GlobalValue::InternalLinkage); } +static inline bool shouldRecordVTableAddr(GlobalVariable *GV) { + if (!profDataReferencedByCode(*GV->getParent())) + return false; + + if (!GV->hasLinkOnceLinkage() && !GV->hasLocalLinkage() && + !GV->hasAvailableExternallyLinkage()) + return true; + + // This avoids the profile data from referencing internal symbols in + // COMDAT. + if (GV->hasLocalLinkage() && GV->hasComdat()) + return false; + + return true; +} + +// FIXME: Does symbolic relocation from 'getFuncAddrForProfData' matter here? +static inline Constant *getVTableAddrForProfData(GlobalVariable *GV) { + auto *Int8PtrTy = PointerType::getUnqual(GV->getContext()); + + // Store a nullptr in __profvt_ if a real address shouldn't be used. + if (!shouldRecordVTableAddr(GV)) + return ConstantPointerNull::get(Int8PtrTy); + + return ConstantExpr::getBitCast(GV, Int8PtrTy); +} + +void InstrLowerer::getOrCreateVTableProfData(GlobalVariable *GV) { + assert(!DebugInfoCorrelate && + "Value profiling is not supported with lightweight instrumentation"); + if (GV->isDeclaration() || GV->hasAvailableExternallyLinkage()) + return; + + if (GV->getName().starts_with("llvm.") || + GV->getName().starts_with("__llvm") || + GV->getName().starts_with("__prof")) + return; + + // VTableProfData already created + auto It = VTableDataMap.find(GV); + if (It != VTableDataMap.end() && It->second) + return; + + GlobalValue::LinkageTypes Linkage = GV->getLinkage(); + GlobalValue::VisibilityTypes Visibility = GV->getVisibility(); + + // This is to keep consistent with per-function profile data + // for correctness. + if (TT.isOSBinFormatXCOFF()) { + Linkage = GlobalValue::InternalLinkage; + Visibility = GlobalValue::DefaultVisibility; + } + + LLVMContext &Ctx = M.getContext(); + Type *DataTypes[] = { +#define INSTR_PROF_VTABLE_DATA(Type, LLVMType, Name, Init) LLVMType, +#include "llvm/ProfileData/InstrProfData.inc" + }; + + auto *DataTy = StructType::get(Ctx, ArrayRef(DataTypes)); + + // Used by INSTR_PROF_VTABLE_DATA MACRO + Constant *VTableAddr = getVTableAddrForProfData(GV); + const std::string PGOVTableName = getPGOName(*GV); + // Record the length of the vtable. This is needed since vtable pointers + // loaded from C++ objects might be from the middle of a vtable definition. + uint32_t VTableSizeVal = + M.getDataLayout().getTypeAllocSize(GV->getValueType()); + + Constant *DataVals[] = { +#define INSTR_PROF_VTABLE_DATA(Type, LLVMType, Name, Init) Init, +#include "llvm/ProfileData/InstrProfData.inc" + }; + + std::string VarName = getInstrProfVTableVarPrefix().str() + PGOVTableName; + auto *Data = + new GlobalVariable(M, DataTy, false /* constant */, Linkage, + ConstantStruct::get(DataTy, DataVals), VarName); + + Data->setVisibility(Visibility); + Data->setSection(getInstrProfSectionName(IPSK_vtab, TT.getObjectFormat())); + Data->setAlignment(Align(8)); + + const bool NeedComdat = needsComdatForCounter(*GV, M); + + // GV is the data structure to record vtable information. + // Place the global variable for per-vtable profile data in a comdat group + // if the associated vtable definition is a COMDAT. This makes sure only one + // copy of the variable for the vtable will be emitted after linking. + auto MaybeSetComdat = [&](GlobalVariable *GV, StringRef GroupName) { + bool UseComdat = (NeedComdat || TT.isOSBinFormatELF()); + if (UseComdat) { + // Create a new comdat group using the name of the global variable as + // opposed to using the comdat group of the vtable. + Comdat *C = M.getOrInsertComdat(GroupName); + // For ELF, when not using COMDAT, put the vtable profile data into a + // nodeduplicate COMDAT which is lowered to a zero-flag zero group. + // This allows -z -start-stop-gc to discard the entire group when the + // vtable def is discarded. + if (!NeedComdat) + C->setSelectionKind(Comdat::NoDeduplicate); + GV->setComdat(C); + // COFF doesn't allow the comdat group leader to have private linkage, so + // upgrade private linkage to internal linkage to produce a symbol table + // entry. + if (TT.isOSBinFormatCOFF() && GV->hasPrivateLinkage()) { + GV->setLinkage(GlobalValue::InternalLinkage); + } + return; + } + }; + + MaybeSetComdat(Data, Data->getName()); + + VTableDataMap[GV] = Data; + + ReferencedVTables.push_back(GV); + + // VTable <Hash, Addr> is used by runtime but not referenced by other + // sections. Conservatively mark it linker retained. + UsedVars.push_back(Data); +} + GlobalVariable *InstrLowerer::setupProfileSection(InstrProfInstBase *Inc, InstrProfSectKind IPSK) { GlobalVariable *NamePtr = Inc->getName(); @@ -1633,6 +1781,31 @@ void InstrLowerer::emitNameData() { NamePtr->eraseFromParent(); } +void InstrLowerer::emitVTableNames() { + if (!EnableVTableValueProfiling || ReferencedVTables.empty()) + return; + + // Collect the PGO names of referenced vtables and compress them. + std::string CompressedVTableNames; + if (Error E = collectVTableStrings(ReferencedVTables, CompressedVTableNames, + DoInstrProfNameCompression)) { + report_fatal_error(Twine(toString(std::move(E))), false); + } + + auto &Ctx = M.getContext(); + auto *VTableNamesVal = ConstantDataArray::getString( + Ctx, StringRef(CompressedVTableNames), false /* AddNull */); + GlobalVariable *VTableNamesVar = + new GlobalVariable(M, VTableNamesVal->getType(), true /* constant */, + GlobalValue::PrivateLinkage, VTableNamesVal, + getInstrProfVTableNamesVarName()); + VTableNamesVar->setSection( + getInstrProfSectionName(IPSK_vname, TT.getObjectFormat())); + VTableNamesVar->setAlignment(Align(1)); + // Make VTableNames linker retained. + UsedVars.push_back(VTableNamesVar); +} + void InstrLowerer::emitRegistration() { if (!needsRuntimeRegistrationOfSectionRange(TT)) return; diff --git a/llvm/lib/Transforms/Instrumentation/PGOInstrumentation.cpp b/llvm/lib/Transforms/Instrumentation/PGOInstrumentation.cpp index c20fc94..f1aa17d 100644 --- a/llvm/lib/Transforms/Instrumentation/PGOInstrumentation.cpp +++ b/llvm/lib/Transforms/Instrumentation/PGOInstrumentation.cpp @@ -327,6 +327,11 @@ extern cl::opt<PGOViewCountsType> PGOViewCounts; // Defined in Analysis/BlockFrequencyInfo.cpp: -view-bfi-func-name= extern cl::opt<std::string> ViewBlockFreqFuncName; +extern cl::opt<bool> DebugInfoCorrelate; + +// Command line option to enable vtable value profiling. Defined in +// ProfileData/InstrProf.cpp: -enable-vtable-value-profiling= +extern cl::opt<bool> EnableVTableValueProfiling; extern cl::opt<InstrProfCorrelator::ProfCorrelatorKind> ProfileCorrelate; } // namespace llvm @@ -581,6 +586,8 @@ public: NumOfPGOMemIntrinsics += ValueSites[IPVK_MemOPSize].size(); NumOfPGOBB += MST.bbInfoSize(); ValueSites[IPVK_IndirectCallTarget] = VPC.get(IPVK_IndirectCallTarget); + if (EnableVTableValueProfiling) + ValueSites[IPVK_VTableTarget] = VPC.get(IPVK_VTableTarget); } else { NumOfCSPGOSelectInsts += SIVisitor.getNumOfSelectInsts(); NumOfCSPGOMemIntrinsics += ValueSites[IPVK_MemOPSize].size(); diff --git a/llvm/lib/Transforms/Instrumentation/ValueProfilePlugins.inc b/llvm/lib/Transforms/Instrumentation/ValueProfilePlugins.inc index 3a129de..96b2130 100644 --- a/llvm/lib/Transforms/Instrumentation/ValueProfilePlugins.inc +++ b/llvm/lib/Transforms/Instrumentation/ValueProfilePlugins.inc @@ -90,9 +90,39 @@ public: } }; +///------------------------ VirtualTableValueProfilingPlugin +///------------------------ +class VTableProfilingPlugin { + Function &F; + +public: + static constexpr InstrProfValueKind Kind = IPVK_VTableTarget; + + VTableProfilingPlugin(Function &Fn, TargetLibraryInfo &TLI) : F(Fn) {} + + void run(std::vector<CandidateInfo> &Candidates) { + std::vector<Instruction *> Result = findVTableAddrs(F); + for (Instruction *I : Result) { + Instruction *InsertPt = I->getNextNonDebugInstruction(); + // When finding an insertion point, keep PHI and EH pad instructions + // before vp intrinsics. This is similar to + // `BasicBlock::getFirstInsertionPt`. + while (InsertPt && (dyn_cast<PHINode>(InsertPt) || InsertPt->isEHPad())) + InsertPt = InsertPt->getNextNonDebugInstruction(); + // Skip instrumentating the value if InsertPt is the last instruction. + // FIXME: Set InsertPt to the end of basic block to instrument the value + // if InsertPt is the last instruction. + if (InsertPt == nullptr) + continue; + + Instruction *AnnotatedInst = I; + Candidates.emplace_back(CandidateInfo{I, InsertPt, AnnotatedInst}); + } + } +}; + ///----------------------- Registration of the plugins ------------------------- /// For now, registering a plugin with the ValueProfileCollector is done by /// adding the plugin type to the VP_PLUGIN_LIST macro. -#define VP_PLUGIN_LIST \ - MemIntrinsicPlugin, \ - IndirectCallPromotionPlugin +#define VP_PLUGIN_LIST \ + MemIntrinsicPlugin, IndirectCallPromotionPlugin, VTableProfilingPlugin diff --git a/llvm/test/Bitcode/thinlto-func-summary-vtableref-pgo.ll b/llvm/test/Bitcode/thinlto-func-summary-vtableref-pgo.ll new file mode 100644 index 0000000..ba3ce9a --- /dev/null +++ b/llvm/test/Bitcode/thinlto-func-summary-vtableref-pgo.ll @@ -0,0 +1,74 @@ +; Promote at most one function and annotate at most one vtable. +; As a result, only one value (of each relevant kind) shows up in the function +; summary. + +; RUN: opt -module-summary -icp-max-num-vtables=1 -icp-max-prom=1 %s -o %t.o + +; RUN: llvm-bcanalyzer -dump %t.o | FileCheck %s + +; RUN: llvm-dis -o - %t.o | FileCheck %s --check-prefix=DIS +; Round trip it through llvm-as +; RUN: llvm-dis -o - %t.o | llvm-as -o - | llvm-dis -o - | FileCheck %s --check-prefix=DIS + +; CHECK: <GLOBALVAL_SUMMARY_BLOCK +; CHECK-NEXT: <VERSION op0=9/> +; CHECK-NEXT: <FLAGS op0=0/> +; The `VALUE_GUID` below represents the "_ZTV4Base" referenced by the instruction +; that loads vtable pointers. +; CHECK-NEXT: <VALUE_GUID op0=21 op1=1960855528937986108/> +; The `VALUE_GUID` below represents the "_ZN4Base4funcEv" referenced by the +; indirect call instruction. +; CHECK-NEXT: <VALUE_GUID op0=20 op1=5459407273543877811/> +; NOTE vtables and functions from Derived class is dropped because +; `-icp-max-num-vtables` and `-icp-max-prom` are both set to one. +; <PERMODULE_PROFILE> has the format [valueid, flags, instcount, funcflags, +; numrefs, rorefcnt, worefcnt, +; m x valueid, +; n x (valueid, hotness+tailcall)] +; CHECK-NEXT: <PERMODULE_PROFILE abbrevid=4 op0=0 op1=0 op2=4 op3=256 op4=1 op5=1 op6=0 op7=21 op8=20 op9=3/> +; CHECK-NEXT: </GLOBALVAL_SUMMARY_BLOCK> + +target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128" +target triple = "x86_64-unknown-linux-gnu" + +; Function has one BB and an entry count of 150, so the BB is hot according to +; ProfileSummary and reflected so in the bitcode (see llvm-dis output). +define i32 @_Z4testP4Base(ptr %0) !prof !15 { + %2 = load ptr, ptr %0, !prof !16 + %3 = load ptr, ptr %2 + %4 = tail call i32 %3(ptr %0), !prof !17 + ret i32 %4 +} + +!llvm.module.flags = !{!1} + +!1 = !{i32 1, !"ProfileSummary", !2} +!2 = !{!3, !4, !5, !6, !7, !8, !9, !10} +!3 = !{!"ProfileFormat", !"InstrProf"} +!4 = !{!"TotalCount", i64 10000} +!5 = !{!"MaxCount", i64 200} +!6 = !{!"MaxInternalCount", i64 200} +!7 = !{!"MaxFunctionCount", i64 200} +!8 = !{!"NumCounts", i64 3} +!9 = !{!"NumFunctions", i64 3} +!10 = !{!"DetailedSummary", !11} +!11 = !{!12, !13, !14} +!12 = !{i32 10000, i64 100, i32 1} +!13 = !{i32 990000, i64 100, i32 1} +!14 = !{i32 999999, i64 1, i32 2} + +!15 = !{!"function_entry_count", i32 150} +; 1960855528937986108 is the MD5 hash of _ZTV4Base, and +; 13870436605473471591 is the MD5 hash of _ZTV7Derived +!16 = !{!"VP", i32 2, i64 150, i64 1960855528937986108, i64 100, i64 13870436605473471591, i64 50} +; 5459407273543877811 is the MD5 hash of _ZN4Base4funcEv, and +; 6174874150489409711 is the MD5 hash of _ZN7Derived4funcEv +!17 = !{!"VP", i32 0, i64 150, i64 5459407273543877811, i64 100, i64 6174874150489409711, i64 50} + +; ModuleSummaryIndex stores <guid, global-value summary> map in std::map; so +; global value summares are printed out in the order that gv's guid increases. +; DIS: ^0 = module: (path: "{{.*}}", hash: (0, 0, 0, 0, 0)) +; DIS: ^1 = gv: (guid: 1960855528937986108) +; DIS: ^2 = gv: (guid: 5459407273543877811) +; DIS: ^3 = gv: (name: "_Z4testP4Base", summaries: (function: (module: ^0, flags: (linkage: external, visibility: default, notEligibleToImport: 0, live: 0, dsoLocal: 0, canAutoHide: 0), insts: 4, funcFlags: (readNone: 0, readOnly: 0, noRecurse: 0, returnDoesNotAlias: 0, noInline: 0, alwaysInline: 0, noUnwind: 0, mayThrow: 0, hasUnknownCall: 1, mustBeUnreachable: 0), calls: ((callee: ^2, hotness: hot)), refs: (readonly ^1)))) ; guid = 15857150948103218965 +; DIS: ^4 = blockcount: 0 diff --git a/llvm/test/Instrumentation/InstrProfiling/coverage.ll b/llvm/test/Instrumentation/InstrProfiling/coverage.ll index bbf895e..08cbcaa 100644 --- a/llvm/test/Instrumentation/InstrProfiling/coverage.ll +++ b/llvm/test/Instrumentation/InstrProfiling/coverage.ll @@ -5,12 +5,12 @@ target triple = "aarch64-unknown-linux-gnu" @__profn_foo = private constant [3 x i8] c"foo" ; CHECK: @__profc_foo = private global [1 x i8] c"\FF", section "__llvm_prf_cnts", comdat, align 1 -; CHECK: @__profd_foo = private global { i64, i64, i64, i64, ptr, ptr, i32, [2 x i16], i32 } { i64 {{.*}}, i64 {{.*}}, i64 sub (i64 ptrtoint (ptr @__profc_foo to i64) -; BINARY: @__profd_foo = private global { i64, i64, i64, i64, ptr, ptr, i32, [2 x i16], i32 } { i64 {{.*}}, i64 {{.*}}, i64 ptrtoint (ptr @__profc_foo to i64), +; CHECK: @__profd_foo = private global { i64, i64, i64, i64, ptr, ptr, i32, [3 x i16], i32 } { i64 {{.*}}, i64 {{.*}}, i64 sub (i64 ptrtoint (ptr @__profc_foo to i64) +; BINARY: @__profd_foo = private global { i64, i64, i64, i64, ptr, ptr, i32, [3 x i16], i32 } { i64 {{.*}}, i64 {{.*}}, i64 ptrtoint (ptr @__profc_foo to i64), @__profn_bar = private constant [3 x i8] c"bar" ; CHECK: @__profc_bar = private global [1 x i8] c"\FF", section "__llvm_prf_cnts", comdat, align 1 -; CHECK: @__profd_bar = private global { i64, i64, i64, i64, ptr, ptr, i32, [2 x i16], i32 } { i64 {{.*}}, i64 {{.*}}, i64 sub (i64 ptrtoint (ptr @__profc_bar to i64) -; BINARY: @__profd_bar = private global { i64, i64, i64, i64, ptr, ptr, i32, [2 x i16], i32 } { i64 {{.*}}, i64 {{.*}}, i64 ptrtoint (ptr @__profc_bar to i64), +; CHECK: @__profd_bar = private global { i64, i64, i64, i64, ptr, ptr, i32, [3 x i16], i32 } { i64 {{.*}}, i64 {{.*}}, i64 sub (i64 ptrtoint (ptr @__profc_bar to i64) +; BINARY: @__profd_bar = private global { i64, i64, i64, i64, ptr, ptr, i32, [3 x i16], i32 } { i64 {{.*}}, i64 {{.*}}, i64 ptrtoint (ptr @__profc_bar to i64), ; CHECK: @__llvm_prf_nm = {{.*}} section "__llvm_prf_names" ; BINARY: @__llvm_prf_nm ={{.*}} section "__llvm_covnames" diff --git a/llvm/test/Transforms/PGOProfile/Inputs/thinlto_indirect_call_promotion.profraw b/llvm/test/Transforms/PGOProfile/Inputs/thinlto_indirect_call_promotion.profraw Binary files differindex 5efda10..5d96ba8a 100644 --- a/llvm/test/Transforms/PGOProfile/Inputs/thinlto_indirect_call_promotion.profraw +++ b/llvm/test/Transforms/PGOProfile/Inputs/thinlto_indirect_call_promotion.profraw diff --git a/llvm/test/Transforms/PGOProfile/Inputs/vtable_prof.profraw b/llvm/test/Transforms/PGOProfile/Inputs/vtable_prof.profraw Binary files differnew file mode 100644 index 0000000..5adeb77 --- /dev/null +++ b/llvm/test/Transforms/PGOProfile/Inputs/vtable_prof.profraw diff --git a/llvm/test/Transforms/PGOProfile/comdat_internal.ll b/llvm/test/Transforms/PGOProfile/comdat_internal.ll index 8c6942c..1bad0db 100644 --- a/llvm/test/Transforms/PGOProfile/comdat_internal.ll +++ b/llvm/test/Transforms/PGOProfile/comdat_internal.ll @@ -13,9 +13,9 @@ $foo = comdat any ; CHECK: @__llvm_profile_raw_version = hidden constant i64 {{[0-9]+}}, comdat ; CHECK-NOT: __profn__stdin__foo ; CHECK: @__profc__stdin__foo.[[#FOO_HASH]] = private global [1 x i64] zeroinitializer, section "__llvm_prf_cnts", comdat, align 8 -; CHECK: @__profd__stdin__foo.[[#FOO_HASH]] = private global { i64, i64, i64, i64, ptr, ptr, i32, [2 x i16], i32 } { i64 {{.*}}, i64 [[#FOO_HASH]], i64 sub (i64 ptrtoint (ptr @__profc__stdin__foo.742261418966908927 to i64), i64 ptrtoint (ptr @__profd__stdin__foo.742261418966908927 to i64)), i64 0, ptr null +; CHECK: @__profd__stdin__foo.[[#FOO_HASH]] = private global { i64, i64, i64, i64, ptr, ptr, i32, [3 x i16], i32 } { i64 {{.*}}, i64 [[#FOO_HASH]], i64 sub (i64 ptrtoint (ptr @__profc__stdin__foo.742261418966908927 to i64), i64 ptrtoint (ptr @__profd__stdin__foo.742261418966908927 to i64)), i64 0, ptr null ; CHECK-NOT: @foo -; CHECK-SAME: , ptr null, i32 1, [2 x i16] zeroinitializer, i32 0 }, section "__llvm_prf_data", comdat($__profc__stdin__foo.[[#FOO_HASH]]), align 8 +; CHECK-SAME: , ptr null, i32 1, [3 x i16] zeroinitializer, i32 0 }, section "__llvm_prf_data", comdat($__profc__stdin__foo.[[#FOO_HASH]]), align 8 ; CHECK: @__llvm_prf_nm ; CHECK: @llvm.compiler.used diff --git a/llvm/test/Transforms/PGOProfile/vtable_profile.ll b/llvm/test/Transforms/PGOProfile/vtable_profile.ll new file mode 100644 index 0000000..edc866e --- /dev/null +++ b/llvm/test/Transforms/PGOProfile/vtable_profile.ll @@ -0,0 +1,98 @@ +; RUN: opt < %s -passes=pgo-instr-gen -enable-vtable-value-profiling -S | FileCheck %s --check-prefix=GEN +; RUN: opt < %s -passes=pgo-instr-gen,instrprof -enable-vtable-value-profiling -S | FileCheck %s --check-prefix=LOWER + +; __llvm_prf_vnm stores zlib-compressed vtable names. +; REQUIRES: zlib + +source_filename = "vtable_local.ll" +target datalayout = "e-m:e-p270:32:32-p271:32:32-p272:64:64-i64:64-f80:128-n8:16:32:64-S128" +target triple = "x86_64-unknown-linux-gnu" + +; The test IR is generated based on the following C++ program. +; Base1 has external linkage and Base2 has local linkage. +; class Derived uses multiple inheritance so its virtual table +; global variable contains two vtables. func1 is loaded from +; the vtable compatible with class Base1, and func2 is loaded +; from the vtable compatible with class Base2. + +; class Base1 { +; public: +; virtual int func1(int a) ; +; }; +; +; namespace { +; class Base2 { +; public: +; __attribute__((noinline)) virtual int func2(int a) { +; return a; +; } +; }; +; } + +; class Derived : public Base1, public Base2 { +; public: +; Derived(int c) : v(c) {} +; private: +; int v; +; }; +; +; Derived* createType(); + +; int func(int a) { +; Derived* d = createType(); +; return d->func2(a) + d->func1(a); +; } + +target datalayout = "e-m:e-p270:32:32-p271:32:32-p272:64:64-i64:64-i128:128-f80:128-n8:16:32:64-S128" +target triple = "x86_64-unknown-linux-gnu" + +@_ZTV7Derived = constant { [3 x ptr], [3 x ptr] } { [3 x ptr] [ptr null, ptr null, ptr @_ZN5Base15func1Ei], [3 x ptr] [ptr inttoptr (i64 -8 to ptr), ptr null, ptr @_ZN12_GLOBAL__N_15Base25func2Ei] }, !type !0, !type !3, !type !6, !type !8, !type !10 +@_ZTV5Base1 = available_externally constant { [3 x ptr] } { [3 x ptr] [ptr null, ptr null, ptr @_ZN5Base15func1Ei] }, !type !0 +@_ZTVN12_GLOBAL__N_15Base2E = internal constant { [3 x ptr] } { [3 x ptr] [ptr null, ptr null, ptr @_ZN12_GLOBAL__N_15Base25func2Ei] }, !type !11, !type !8; !vcall_visibility !12 +@llvm.compiler.used = appending global [1 x ptr] [ptr @_ZTV5Base1], section "llvm.metadata" + +; GEN: __llvm_profile_raw_version = comdat any +; GEN: __llvm_profile_raw_version = hidden constant i64 72057594037927946, comdat +; GEN: __profn__Z4funci = private constant [8 x i8] c"_Z4funci" + +; LOWER: $__profvt__ZTV7Derived = comdat nodeduplicate +; LOWER: $"__profvt_vtable_local.ll;_ZTVN12_GLOBAL__N_15Base2E" = comdat nodeduplicate +; LOWER: @__profvt__ZTV7Derived = global { i64, ptr, i32 } { i64 -4576307468236080025, ptr @_ZTV7Derived, i32 48 }, section "__llvm_prf_vtab", comdat, align 8 +; LOWER: @"__profvt_vtable_local.ll;_ZTVN12_GLOBAL__N_15Base2E" = internal global { i64, ptr, i32 } { i64 1419990121885302679, ptr @_ZTVN12_GLOBAL__N_15Base2E, i32 24 }, section "__llvm_prf_vtab", comdat, align 8 +; LOWER: @__llvm_prf_vnm = private constant [64 x i8] c"7>x\DA\8B\8F\0A\093wI-\CA,KMa,+IL\CAI\8D\CF\C9ON\CC\D1\CB\C9\B1\8E\07J\FA\19\1A\C5\BB\FB\F8;9\FA\C4\C7\FB\C5\1B\9A:%\16\A7\1A\B9\02\00\19:\12o", section "__llvm_prf_vtabnames", align 1 +; LOWER: @llvm.used = appending global [5 x ptr] [ptr @__profvt__ZTV7Derived, ptr @"__profvt_vtable_local.ll;_ZTVN12_GLOBAL__N_15Base2E", ptr @__llvm_prf_vnodes, ptr @__llvm_prf_nm, ptr @__llvm_prf_vnm], section "llvm.metadata" + +define i32 @_Z4funci(i32 %a) { +entry: + %call = call ptr @_Z10createTypev() + %add.ptr = getelementptr inbounds i8, ptr %call, i64 8 + %vtable = load ptr, ptr %add.ptr +; GEN: [[P1:%[0-9]+]] = ptrtoint ptr %vtable to i64 +; GEN: call void @llvm.instrprof.value.profile(ptr @__profn__Z4funci, i64 [[CFGHash:[0-9]+]], i64 [[P1]], i32 2, i32 0) +; LOWER: [[P1:%[0-9]+]] = ptrtoint ptr %vtable to i64 +; LOWER: call void @__llvm_profile_instrument_target(i64 [[P1]], ptr @__profd__Z4funci, i32 2) + %vfunc1 = load ptr, ptr %vtable + %call1 = call i32 %vfunc1(ptr %add.ptr, i32 %a) + %vtable2 = load ptr, ptr %call +; GEN: [[P2:%[0-9]+]] = ptrtoint ptr %vtable2 to i64 +; GEN: call void @llvm.instrprof.value.profile(ptr @__profn__Z4funci, i64 [[CFGHash]], i64 [[P2]], i32 2, i32 1) +; LOWER: [[P2:%[0-9]+]] = ptrtoint ptr %vtable2 to i64 +; LOWER: call void @__llvm_profile_instrument_target(i64 [[P2]], ptr @__profd__Z4funci, i32 3) + %vfunc2 = load ptr, ptr %vtable2 + %call4 = call i32 %vfunc2(ptr %call, i32 %a) + %add = add nsw i32 %call1, %call4 + ret i32 %add +} + +declare ptr @_Z10createTypev() +declare i32 @_ZN12_GLOBAL__N_15Base25func2Ei(ptr %this, i32 %a) +declare i32 @_ZN5Base15func1Ei(ptr, i32) + +!0 = !{i64 16, !"_ZTS5Base1"} +!3 = !{i64 16, !"_ZTS7Derived"} +!6 = !{i64 40, !7} +!7 = distinct !{} +!8 = !{i64 16, !9} +!9 = distinct !{} +!10 = !{i64 40, !9} +!11 = !{i64 16, !7} diff --git a/llvm/test/tools/llvm-profdata/Inputs/c-general.profraw b/llvm/test/tools/llvm-profdata/Inputs/c-general.profraw Binary files differindex 9cd2255..a5dcc9f 100644 --- a/llvm/test/tools/llvm-profdata/Inputs/c-general.profraw +++ b/llvm/test/tools/llvm-profdata/Inputs/c-general.profraw diff --git a/llvm/test/tools/llvm-profdata/Inputs/compressed.profraw b/llvm/test/tools/llvm-profdata/Inputs/compressed.profraw Binary files differindex 9966729..4d36ffc 100644 --- a/llvm/test/tools/llvm-profdata/Inputs/compressed.profraw +++ b/llvm/test/tools/llvm-profdata/Inputs/compressed.profraw diff --git a/llvm/test/tools/llvm-profdata/Inputs/update_vtable_value_prof_inputs.sh b/llvm/test/tools/llvm-profdata/Inputs/update_vtable_value_prof_inputs.sh new file mode 100755 index 0000000..89c3e64 --- /dev/null +++ b/llvm/test/tools/llvm-profdata/Inputs/update_vtable_value_prof_inputs.sh @@ -0,0 +1,102 @@ +#!/bin/bash + +if [ $# -lt 1 ]; then + echo "Path to clang++ required!" + echo "Usage: update_vtable_value_prof_inputs.sh /path/to/updated/clang++" + exit 1 +else + CLANG=$1 +fi + + +# Remember current directory. +CURDIR=$PWD + +# Allows the script to be invoked from other directories. +OUTDIR=$(dirname $(realpath -s $0)) +echo $OUTDIR + +cd $OUTDIR + +# vtable_prof.cc has the following class hierarchy: +# class Base +# ├── class Derived1 +# └── class Derived2 +# Derived1 is a class in the global namespace and Derived2 is in anonymous +# namespace for test coverage. Overridden virtual methods are annotated as +# `noinline` so the callsite remains indirect calls for testing purposes. +cat > vtable_prof.cc << EOF +#include <cstdlib> +#include <cstdio> + +class Base { + public: + virtual int func1(int a, int b) = 0; + virtual int func2(int a, int b) = 0; +}; + +class Derived1 : public Base { + public: + __attribute__((noinline)) + int func1(int a, int b) override + { + return a + b; + } + + __attribute__((noinline)) + int func2(int a, int b) override { + return a * b; + } +}; + +namespace { +class Derived2 : public Base { + public: + __attribute__((noinline)) + int func1(int a, int b) override { + return a - b; + } + + __attribute__((noinline)) + int func2(int a, int b) override { + return a * (a - b); + } +}; +} // namespace + +__attribute__((noinline)) Base* createType(int a) { + Base* base = nullptr; + if (a % 4 == 0) + base = new Derived1(); + else + base = new Derived2(); + return base; +} + + +int main(int argc, char** argv) { + int sum = 0; + for (int i = 0; i < 1000; i++) { + int a = rand(); + int b = rand(); + Base* ptr = createType(i); + sum += ptr->func1(a, b) + ptr->func2(b, a); + } + printf("sum is %d\n", sum); + return 0; +} +EOF + + +# Clean up temporary files on exit and return to original directory. +cleanup() { + rm -f vtable_prof + rm -f vtable_prof.cc + cd $CURDIR +} +trap cleanup EXIT + +FLAGS="-fuse-ld=lld -O2 -g -fprofile-generate=. -mllvm -enable-vtable-value-profiling" + +${CLANG} ${FLAGS} vtable_prof.cc -o vtable_prof +env LLVM_PROFILE_FILE=vtable-value-prof-basic.profraw ./vtable_prof diff --git a/llvm/test/tools/llvm-profdata/Inputs/vtable-value-prof-basic.profraw b/llvm/test/tools/llvm-profdata/Inputs/vtable-value-prof-basic.profraw Binary files differnew file mode 100644 index 0000000..322c8dc --- /dev/null +++ b/llvm/test/tools/llvm-profdata/Inputs/vtable-value-prof-basic.profraw diff --git a/llvm/test/tools/llvm-profdata/Inputs/vtable-value-prof.proftext b/llvm/test/tools/llvm-profdata/Inputs/vtable-value-prof.proftext new file mode 100644 index 0000000..ec85dc4 --- /dev/null +++ b/llvm/test/tools/llvm-profdata/Inputs/vtable-value-prof.proftext @@ -0,0 +1,73 @@ +# IR level Instrumentation Flag +:ir +/path/to/vtable_prof.cc;_ZN12_GLOBAL__N_18Derived25func1Eii +# Func Hash: +742261418966908927 +# Num Counters: +1 +# Counter Values: +750 + +/path/to/vtable_prof.cc;_ZN12_GLOBAL__N_18Derived25func2Eii +# Func Hash: +742261418966908927 +# Num Counters: +1 +# Counter Values: +750 + +_Z10createTypei +# Func Hash: +146835647075900052 +# Num Counters: +2 +# Counter Values: +750 +250 + +_ZN8Derived15func1Eii +# Func Hash: +742261418966908927 +# Num Counters: +1 +# Counter Values: +250 + +_ZN8Derived15func2Eii +# Func Hash: +742261418966908927 +# Num Counters: +1 +# Counter Values: +250 + +main +# Func Hash: +1124236338992350536 +# Num Counters: +2 +# Counter Values: +1000 +1 +# Num Value Kinds: +2 +# ValueKind = IPVK_IndirectCallTarget: +0 +# NumValueSites: +2 +2 +/path/to/vtable_prof.cc;_ZN12_GLOBAL__N_18Derived25func1Eii:750 +_ZN8Derived15func1Eii:250 +2 +/path/to/vtable_prof.cc;_ZN12_GLOBAL__N_18Derived25func2Eii:750 +_ZN8Derived15func2Eii:250 +# ValueKind = IPVK_VTableTarget: +2 +# NumValueSites: +2 +2 +/path/to/vtable_prof.cc;_ZTVN12_GLOBAL__N_18Derived2E:750 +_ZTV8Derived1:250 +2 +/path/to/vtable_prof.cc;_ZTVN12_GLOBAL__N_18Derived2E:750 +_ZTV8Derived1:250 diff --git a/llvm/test/tools/llvm-profdata/binary-ids-padding.test b/llvm/test/tools/llvm-profdata/binary-ids-padding.test index eda6320..61881b6 100644 --- a/llvm/test/tools/llvm-profdata/binary-ids-padding.test +++ b/llvm/test/tools/llvm-profdata/binary-ids-padding.test @@ -10,10 +10,12 @@ // INSTR_PROF_RAW_HEADER(uint64_t, CountersDelta, (uintptr_t)CountersBegin) // INSTR_PROF_RAW_HEADER(uint64_t, BitmaskDelta, (uintptr_t)BitmaskBegin) // INSTR_PROF_RAW_HEADER(uint64_t, NamesDelta, (uintptr_t)NamesBegin) +// INSTR_PROF_RAW_HEADER(uint64_t, VNamesSize, VNamesSize) +// INSTR_PROF_RAW_HEADER(uint64_t, NumVTables, NumVTables) // INSTR_PROF_RAW_HEADER(uint64_t, ValueKindLast, IPVK_Last) RUN: printf '\201rforpl\377' > %t.profraw -RUN: printf '\11\0\0\0\0\0\0\0' >> %t.profraw +RUN: printf '\12\0\0\0\0\0\0\0' >> %t.profraw // There will be 2 20-byte binary IDs, so the total Binary IDs size will be 64 bytes. // 2 * 8 binary ID sizes // + 2 * 20 binary IDs (of size 20) @@ -32,6 +34,8 @@ RUN: printf '\0\0\4\0\1\0\0\0' >> %t.profraw RUN: printf '\0\0\0\0\0\0\0\0' >> %t.profraw RUN: printf '\0\0\4\0\2\0\0\0' >> %t.profraw RUN: printf '\0\0\0\0\0\0\0\0' >> %t.profraw +RUN: printf '\0\0\0\0\0\0\0\0' >> %t.profraw +RUN: printf '\0\0\0\0\0\0\0\0' >> %t.profraw // Binary IDs - There are only two in this case that are 20 bytes. RUN: printf '\24\0\0\0\0\0\0\0' >> %t.profraw diff --git a/llvm/test/tools/llvm-profdata/large-binary-id-size.test b/llvm/test/tools/llvm-profdata/large-binary-id-size.test index 38b838e..316a9a4 100644 --- a/llvm/test/tools/llvm-profdata/large-binary-id-size.test +++ b/llvm/test/tools/llvm-profdata/large-binary-id-size.test @@ -1,5 +1,5 @@ RUN: printf '\201rforpl\377' > %t.profraw -RUN: printf '\11\0\0\0\0\0\0\0' >> %t.profraw +RUN: printf '\12\0\0\0\0\0\0\0' >> %t.profraw RUN: printf '\40\0\0\0\0\0\0\0' >> %t.profraw RUN: printf '\0\0\0\0\0\0\0\0' >> %t.profraw RUN: printf '\0\0\0\0\0\0\0\0' >> %t.profraw @@ -12,6 +12,8 @@ RUN: printf '\0\0\0\0\0\0\0\0' >> %t.profraw RUN: printf '\0\0\0\0\0\0\0\0' >> %t.profraw RUN: printf '\0\0\0\0\0\0\0\0' >> %t.profraw RUN: printf '\0\0\0\0\0\0\0\0' >> %t.profraw +RUN: printf '\0\0\0\0\0\0\0\0' >> %t.profraw +RUN: printf '\0\0\0\0\0\0\0\0' >> %t.profraw // Check for a corrupted size being too large past the end of the file. RUN: printf '\7\7\7\7\7\7\7\7' >> %t.profraw diff --git a/llvm/test/tools/llvm-profdata/malformed-not-space-for-another-header.test b/llvm/test/tools/llvm-profdata/malformed-not-space-for-another-header.test index c967e85..8b686d5 100644 --- a/llvm/test/tools/llvm-profdata/malformed-not-space-for-another-header.test +++ b/llvm/test/tools/llvm-profdata/malformed-not-space-for-another-header.test @@ -10,10 +10,12 @@ // INSTR_PROF_RAW_HEADER(uint64_t, CountersDelta, (uintptr_t)CountersBegin) // INSTR_PROF_RAW_HEADER(uint64_t, BitmaskDelta, (uintptr_t)BitmaskBegin) // INSTR_PROF_RAW_HEADER(uint64_t, NamesDelta, (uintptr_t)NamesBegin) +// INSTR_PROF_RAW_HEADER(uint64_t, VNamesSize, VNamesSize) +// INSTR_PROF_RAW_HEADER(uint64_t, NumVTables, NumVTables) // INSTR_PROF_RAW_HEADER(uint64_t, ValueKindLast, IPVK_Last) RUN: printf '\201rforpl\377' > %t.profraw -RUN: printf '\11\0\0\0\0\0\0\0' >> %t.profraw +RUN: printf '\12\0\0\0\0\0\0\0' >> %t.profraw RUN: printf '\0\0\0\0\0\0\0\0' >> %t.profraw RUN: printf '\1\0\0\0\0\0\0\0' >> %t.profraw RUN: printf '\0\0\0\0\0\0\0\0' >> %t.profraw @@ -26,6 +28,8 @@ RUN: printf '\0\0\4\0\1\0\0\0' >> %t.profraw RUN: printf '\0\0\0\0\0\0\0\0' >> %t.profraw RUN: printf '\0\0\4\0\2\0\0\0' >> %t.profraw RUN: printf '\0\0\0\0\0\0\0\0' >> %t.profraw +RUN: printf '\0\0\0\0\0\0\0\0' >> %t.profraw +RUN: printf '\0\0\0\0\0\0\0\0' >> %t.profraw // Data Section // diff --git a/llvm/test/tools/llvm-profdata/malformed-num-counters-zero.test b/llvm/test/tools/llvm-profdata/malformed-num-counters-zero.test index 2e747f8..089afad 100644 --- a/llvm/test/tools/llvm-profdata/malformed-num-counters-zero.test +++ b/llvm/test/tools/llvm-profdata/malformed-num-counters-zero.test @@ -10,10 +10,12 @@ // INSTR_PROF_RAW_HEADER(uint64_t, CountersDelta, (uintptr_t)CountersBegin) // INSTR_PROF_RAW_HEADER(uint64_t, BitmaskDelta, (uintptr_t)BitmaskBegin) // INSTR_PROF_RAW_HEADER(uint64_t, NamesDelta, (uintptr_t)NamesBegin) +// INSTR_PROF_RAW_HEADER(uint64_t, VNamesSize, VNamesSize) +// INSTR_PROF_RAW_HEADER(uint64_t, NumVTables, NumVTables) // INSTR_PROF_RAW_HEADER(uint64_t, ValueKindLast, IPVK_Last) RUN: printf '\201rforpl\377' > %t.profraw -RUN: printf '\11\0\0\0\0\0\0\0' >> %t.profraw +RUN: printf '\12\0\0\0\0\0\0\0' >> %t.profraw RUN: printf '\0\0\0\0\0\0\0\0' >> %t.profraw RUN: printf '\1\0\0\0\0\0\0\0' >> %t.profraw RUN: printf '\0\0\0\0\0\0\0\0' >> %t.profraw @@ -26,6 +28,8 @@ RUN: printf '\0\0\4\0\1\0\0\0' >> %t.profraw RUN: printf '\0\0\0\0\0\0\0\0' >> %t.profraw RUN: printf '\0\0\4\0\2\0\0\0' >> %t.profraw RUN: printf '\0\0\0\0\0\0\0\0' >> %t.profraw +RUN: printf '\0\0\0\0\0\0\0\0' >> %t.profraw +RUN: printf '\0\0\0\0\0\0\0\0' >> %t.profraw // Data Section // diff --git a/llvm/test/tools/llvm-profdata/malformed-ptr-to-counter-array.test b/llvm/test/tools/llvm-profdata/malformed-ptr-to-counter-array.test index 3c23bc7..e404ba4 100644 --- a/llvm/test/tools/llvm-profdata/malformed-ptr-to-counter-array.test +++ b/llvm/test/tools/llvm-profdata/malformed-ptr-to-counter-array.test @@ -10,10 +10,12 @@ // INSTR_PROF_RAW_HEADER(uint64_t, CountersDelta, (uintptr_t)CountersBegin) // INSTR_PROF_RAW_HEADER(uint64_t, BitmaskDelta, (uintptr_t)BitmaskBegin) // INSTR_PROF_RAW_HEADER(uint64_t, NamesDelta, (uintptr_t)NamesBegin) +// INSTR_PROF_RAW_HEADER(uint64_t, VNamesSize, VNamesSize) +// INSTR_PROF_RAW_HEADER(uint64_t, NumVTables, NumVTables) // INSTR_PROF_RAW_HEADER(uint64_t, ValueKindLast, IPVK_Last) RUN: printf '\201rforpl\377' > %t.profraw -RUN: printf '\11\0\0\0\0\0\0\0' >> %t.profraw +RUN: printf '\12\0\0\0\0\0\0\0' >> %t.profraw RUN: printf '\0\0\0\0\0\0\0\0' >> %t.profraw RUN: printf '\1\0\0\0\0\0\0\0' >> %t.profraw RUN: printf '\0\0\0\0\0\0\0\0' >> %t.profraw @@ -26,6 +28,8 @@ RUN: printf '\0\0\6\0\1\0\0\0' >> %t.profraw RUN: printf '\0\0\0\0\0\0\0\0' >> %t.profraw RUN: printf '\0\0\6\0\2\0\0\0' >> %t.profraw RUN: printf '\0\0\0\0\0\0\0\0' >> %t.profraw +RUN: printf '\0\0\0\0\0\0\0\0' >> %t.profraw +RUN: printf '\0\0\0\0\0\0\0\0' >> %t.profraw // Data Section // diff --git a/llvm/test/tools/llvm-profdata/misaligned-binary-ids-size.test b/llvm/test/tools/llvm-profdata/misaligned-binary-ids-size.test index 4a5c428..ee54bfb 100644 --- a/llvm/test/tools/llvm-profdata/misaligned-binary-ids-size.test +++ b/llvm/test/tools/llvm-profdata/misaligned-binary-ids-size.test @@ -1,5 +1,5 @@ RUN: printf '\201rforpl\377' > %t.profraw -RUN: printf '\11\0\0\0\0\0\0\0' >> %t.profraw +RUN: printf '\12\0\0\0\0\0\0\0' >> %t.profraw // We should fail on this because the binary IDs is not a multiple of 8 bytes. RUN: printf '\77\0\0\0\0\0\0\0' >> %t.profraw RUN: printf '\0\0\0\0\0\0\0\0' >> %t.profraw @@ -10,6 +10,8 @@ RUN: printf '\0\0\0\0\0\0\0\0' >> %t.profraw RUN: printf '\0\0\0\0\0\0\0\0' >> %t.profraw RUN: printf '\0\0\0\0\0\0\0\0' >> %t.profraw RUN: printf '\0\0\0\0\0\0\0\0' >> %t.profraw +RUN: printf '\0\0\0\0\0\0\0\0' >> %t.profraw +RUN: printf '\0\0\0\0\0\0\0\0' >> %t.profraw // Binary IDs - There are only two in this case that are 20 bytes. RUN: printf '\24\0\0\0\0\0\0\0' >> %t.profraw diff --git a/llvm/test/tools/llvm-profdata/mismatched-raw-profile-header.test b/llvm/test/tools/llvm-profdata/mismatched-raw-profile-header.test index 2a92575..dfa163f 100644 --- a/llvm/test/tools/llvm-profdata/mismatched-raw-profile-header.test +++ b/llvm/test/tools/llvm-profdata/mismatched-raw-profile-header.test @@ -15,6 +15,8 @@ RUN: printf '\0\0\0\0\0\0\0\20' >> %t RUN: printf '\0\0\0\1\0\4\0\0' >> %t RUN: printf '\0\0\0\2\0\4\0\0' >> %t RUN: printf '\0\0\0\0\0\0\0\0' >> %t +RUN: printf '\0\0\0\0\0\0\0\0' >> %t +RUN: printf '\0\0\0\0\0\0\0\0' >> %t RUN: not llvm-profdata show %t -o /dev/null 2>&1 | FileCheck %s diff --git a/llvm/test/tools/llvm-profdata/raw-32-bits-be.test b/llvm/test/tools/llvm-profdata/raw-32-bits-be.test index 8220361..63782c8 100644 --- a/llvm/test/tools/llvm-profdata/raw-32-bits-be.test +++ b/llvm/test/tools/llvm-profdata/raw-32-bits-be.test @@ -1,5 +1,6 @@ +// Header RUN: printf '\377lprofR\201' > %t -RUN: printf '\0\0\0\0\0\0\0\11' >> %t +RUN: printf '\0\0\0\0\0\0\0\12' >> %t RUN: printf '\0\0\0\0\0\0\0\0' >> %t RUN: printf '\0\0\0\0\0\0\0\2' >> %t RUN: printf '\0\0\0\0\0\0\0\0' >> %t @@ -12,6 +13,8 @@ RUN: printf '\0\0\0\0\1\0\0\0' >> %t RUN: printf '\0\0\0\0\3\0\0\0' >> %t RUN: printf '\0\0\0\0\2\0\0\0' >> %t RUN: printf '\0\0\0\0\0\0\0\0' >> %t +RUN: printf '\0\0\0\0\0\0\0\0' >> %t +RUN: printf '\0\0\0\0\0\0\0\0' >> %t RUN: printf '\134\370\302\114\333\030\275\254' >> %t RUN: printf '\0\0\0\0\0\0\0\1' >> %t @@ -20,9 +23,8 @@ RUN: printf '\3\0\0\0' >> %t RUN: printf '\0\0\0\0' >> %t RUN: printf '\0\0\0\0' >> %t RUN: printf '\0\0\0\1' >> %t -RUN: printf '\0\0\0\0' >> %t +RUN: printf '\0\0\0\0\0\0\0\0' >> %t RUN: printf '\0\0\0\3' >> %t -RUN: printf '\0\0\0\0' >> %t RUN: printf '\344\023\165\112\031\035\265\067' >> %t RUN: printf '\0\0\0\0\0\0\0\2' >> %t @@ -31,9 +33,8 @@ RUN: printf '\2\xff\xff\xd3' >> %t RUN: printf '\0\0\0\0' >> %t RUN: printf '\0\0\0\0' >> %t RUN: printf '\0\0\0\2' >> %t -RUN: printf '\0\0\0\0' >> %t +RUN: printf '\0\0\0\0\0\0\0\0' >> %t RUN: printf '\0\0\0\1' >> %t -RUN: printf '\0\0\0\0' >> %t RUN: printf '\0\0\0\0\0\0\0\023' >> %t RUN: printf '\0\0\0\0\0\0\0\067' >> %t diff --git a/llvm/test/tools/llvm-profdata/raw-32-bits-le.test b/llvm/test/tools/llvm-profdata/raw-32-bits-le.test index 9352ae1..e9569be 100644 --- a/llvm/test/tools/llvm-profdata/raw-32-bits-le.test +++ b/llvm/test/tools/llvm-profdata/raw-32-bits-le.test @@ -1,5 +1,5 @@ RUN: printf '\201Rforpl\377' > %t -RUN: printf '\11\0\0\0\0\0\0\0' >> %t +RUN: printf '\12\0\0\0\0\0\0\0' >> %t RUN: printf '\0\0\0\0\0\0\0\0' >> %t RUN: printf '\2\0\0\0\0\0\0\0' >> %t RUN: printf '\0\0\0\0\0\0\0\0' >> %t @@ -12,6 +12,8 @@ RUN: printf '\0\0\0\1\0\0\0\0' >> %t RUN: printf '\0\0\0\3\0\0\0\0' >> %t RUN: printf '\0\0\0\2\0\0\0\0' >> %t RUN: printf '\0\0\0\0\0\0\0\0' >> %t +RUN: printf '\0\0\0\0\0\0\0\0' >> %t +RUN: printf '\0\0\0\0\0\0\0\0' >> %t RUN: printf '\254\275\030\333\114\302\370\134' >> %t RUN: printf '\1\0\0\0\0\0\0\0' >> %t @@ -20,9 +22,8 @@ RUN: printf '\0\0\0\3' >> %t RUN: printf '\0\0\0\0' >> %t RUN: printf '\0\0\0\0' >> %t RUN: printf '\1\0\0\0' >> %t -RUN: printf '\0\0\0\0' >> %t +RUN: printf '\0\0\0\0\0\0\0\0' >> %t RUN: printf '\3\0\0\0' >> %t -RUN: printf '\0\0\0\0' >> %t RUN: printf '\067\265\035\031\112\165\023\344' >> %t RUN: printf '\02\0\0\0\0\0\0\0' >> %t @@ -31,9 +32,8 @@ RUN: printf '\xd3\xff\xff\2' >> %t RUN: printf '\0\0\0\0' >> %t RUN: printf '\0\0\0\0' >> %t RUN: printf '\2\0\0\0' >> %t -RUN: printf '\0\0\0\0' >> %t +RUN: printf '\0\0\0\0\0\0\0\0' >> %t RUN: printf '\1\0\0\0' >> %t -RUN: printf '\0\0\0\0' >> %t RUN: printf '\023\0\0\0\0\0\0\0' >> %t RUN: printf '\067\0\0\0\0\0\0\0' >> %t diff --git a/llvm/test/tools/llvm-profdata/raw-64-bits-be.test b/llvm/test/tools/llvm-profdata/raw-64-bits-be.test index c3e995a..0bc579e 100644 --- a/llvm/test/tools/llvm-profdata/raw-64-bits-be.test +++ b/llvm/test/tools/llvm-profdata/raw-64-bits-be.test @@ -1,5 +1,5 @@ RUN: printf '\377lprofr\201' > %t -RUN: printf '\0\0\0\0\0\0\0\11' >> %t +RUN: printf '\0\0\0\0\0\0\0\12' >> %t RUN: printf '\0\0\0\0\0\0\0\0' >> %t RUN: printf '\0\0\0\0\0\0\0\2' >> %t RUN: printf '\0\0\0\0\0\0\0\0' >> %t @@ -12,6 +12,8 @@ RUN: printf '\0\0\0\1\0\4\0\0' >> %t RUN: printf '\0\0\0\3\0\4\0\0' >> %t RUN: printf '\0\0\0\2\0\4\0\0' >> %t RUN: printf '\0\0\0\0\0\0\0\0' >> %t +RUN: printf '\0\0\0\0\0\0\0\0' >> %t +RUN: printf '\0\0\0\0\0\0\0\0' >> %t RUN: printf '\134\370\302\114\333\030\275\254' >> %t RUN: printf '\0\0\0\0\0\0\0\1' >> %t @@ -20,9 +22,8 @@ RUN: printf '\0\0\0\3\0\4\0\0' >> %t RUN: printf '\0\0\0\0\0\0\0\0' >> %t RUN: printf '\0\0\0\0\0\0\0\0' >> %t RUN: printf '\0\0\0\1' >> %t -RUN: printf '\0\0\0\0' >> %t +RUN: printf '\0\0\0\0\0\0\0\0' >> %t RUN: printf '\0\0\0\3' >> %t -RUN: printf '\0\0\0\0' >> %t RUN: printf '\344\023\165\112\031\035\265\067' >> %t RUN: printf '\0\0\0\0\0\0\0\02' >> %t @@ -31,9 +32,8 @@ RUN: printf '\0\0\0\3\0\3\xff\xc3' >> %t RUN: printf '\0\0\0\0\0\0\0\0' >> %t RUN: printf '\0\0\0\0\0\0\0\0' >> %t RUN: printf '\0\0\0\02' >> %t -RUN: printf '\0\0\0\0' >> %t +RUN: printf '\0\0\0\0\0\0\0\0' >> %t RUN: printf '\0\0\0\1' >> %t -RUN: printf '\0\0\0\0' >> %t RUN: printf '\0\0\0\0\0\0\0\023' >> %t RUN: printf '\0\0\0\0\0\0\0\067' >> %t diff --git a/llvm/test/tools/llvm-profdata/raw-64-bits-le.test b/llvm/test/tools/llvm-profdata/raw-64-bits-le.test index 0b3ef2a..ca9ea54 100644 --- a/llvm/test/tools/llvm-profdata/raw-64-bits-le.test +++ b/llvm/test/tools/llvm-profdata/raw-64-bits-le.test @@ -1,5 +1,5 @@ RUN: printf '\201rforpl\377' > %t -RUN: printf '\11\0\0\0\0\0\0\0' >> %t +RUN: printf '\12\0\0\0\0\0\0\0' >> %t RUN: printf '\0\0\0\0\0\0\0\0' >> %t RUN: printf '\2\0\0\0\0\0\0\0' >> %t RUN: printf '\0\0\0\0\0\0\0\0' >> %t @@ -12,6 +12,8 @@ RUN: printf '\0\0\4\0\1\0\0\0' >> %t RUN: printf '\0\0\4\0\3\0\0\0' >> %t RUN: printf '\0\0\4\0\2\0\0\0' >> %t RUN: printf '\0\0\0\0\0\0\0\0' >> %t +RUN: printf '\0\0\0\0\0\0\0\0' >> %t +RUN: printf '\0\0\0\0\0\0\0\0' >> %t RUN: printf '\254\275\030\333\114\302\370\134' >> %t RUN: printf '\1\0\0\0\0\0\0\0' >> %t @@ -20,9 +22,8 @@ RUN: printf '\0\0\4\0\3\0\0\0' >> %t RUN: printf '\0\0\0\0\0\0\0\0' >> %t RUN: printf '\0\0\0\0\0\0\0\0' >> %t RUN: printf '\1\0\0\0' >> %t -RUN: printf '\0\0\0\0' >> %t +RUN: printf '\0\0\0\0\0\0\0\0' >> %t RUN: printf '\3\0\0\0' >> %t -RUN: printf '\0\0\0\0' >> %t RUN: printf '\067\265\035\031\112\165\023\344' >> %t RUN: printf '\02\0\0\0\0\0\0\0' >> %t @@ -31,9 +32,8 @@ RUN: printf '\xc3\xff\3\0\3\0\0\0' >> %t RUN: printf '\0\0\0\0\0\0\0\0' >> %t RUN: printf '\0\0\0\0\0\0\0\0' >> %t RUN: printf '\02\0\0\0' >> %t -RUN: printf '\0\0\0\0' >> %t +RUN: printf '\0\0\0\0\0\0\0\0' >> %t RUN: printf '\1\0\0\0' >> %t -RUN: printf '\0\0\0\0' >> %t RUN: printf '\023\0\0\0\0\0\0\0' >> %t RUN: printf '\067\0\0\0\0\0\0\0' >> %t diff --git a/llvm/test/tools/llvm-profdata/raw-two-profiles.test b/llvm/test/tools/llvm-profdata/raw-two-profiles.test index f4a9aa8..70a4210 100644 --- a/llvm/test/tools/llvm-profdata/raw-two-profiles.test +++ b/llvm/test/tools/llvm-profdata/raw-two-profiles.test @@ -1,5 +1,5 @@ RUN: printf '\201rforpl\377' > %t-foo.profraw -RUN: printf '\11\0\0\0\0\0\0\0' >> %t-foo.profraw +RUN: printf '\12\0\0\0\0\0\0\0' >> %t-foo.profraw RUN: printf '\0\0\0\0\0\0\0\0' >> %t-foo.profraw RUN: printf '\1\0\0\0\0\0\0\0' >> %t-foo.profraw RUN: printf '\0\0\0\0\0\0\0\0' >> %t-foo.profraw @@ -12,6 +12,8 @@ RUN: printf '\0\0\4\0\1\0\0\0' >> %t-foo.profraw RUN: printf '\0\0\0\0\0\0\0\0' >> %t-foo.profraw RUN: printf '\0\0\4\0\2\0\0\0' >> %t-foo.profraw RUN: printf '\0\0\0\0\0\0\0\0' >> %t-foo.profraw +RUN: printf '\0\0\0\0\0\0\0\0' >> %t-foo.profraw +RUN: printf '\0\0\0\0\0\0\0\0' >> %t-foo.profraw RUN: printf '\254\275\030\333\114\302\370\134' >> %t-foo.profraw RUN: printf '\1\0\0\0\0\0\0\0' >> %t-foo.profraw @@ -26,7 +28,7 @@ RUN: printf '\023\0\0\0\0\0\0\0' >> %t-foo.profraw RUN: printf '\3\0foo\0\0\0' >> %t-foo.profraw RUN: printf '\201rforpl\377' > %t-bar.profraw -RUN: printf '\11\0\0\0\0\0\0\0' >> %t-bar.profraw +RUN: printf '\12\0\0\0\0\0\0\0' >> %t-bar.profraw RUN: printf '\0\0\0\0\0\0\0\0' >> %t-bar.profraw RUN: printf '\1\0\0\0\0\0\0\0' >> %t-bar.profraw RUN: printf '\0\0\0\0\0\0\0\0' >> %t-bar.profraw @@ -39,6 +41,8 @@ RUN: printf '\0\0\6\0\1\0\0\0' >> %t-bar.profraw RUN: printf '\0\0\0\0\0\0\0\0' >> %t-bar.profraw RUN: printf '\0\0\6\0\2\0\0\0' >> %t-bar.profraw RUN: printf '\0\0\0\0\0\0\0\0' >> %t-bar.profraw +RUN: printf '\0\0\0\0\0\0\0\0' >> %t-bar.profraw +RUN: printf '\0\0\0\0\0\0\0\0' >> %t-bar.profraw RUN: printf '\067\265\035\031\112\165\023\344' >> %t-bar.profraw RUN: printf '\02\0\0\0\0\0\0\0' >> %t-bar.profraw diff --git a/llvm/test/tools/llvm-profdata/vtable-value-prof-basic.test b/llvm/test/tools/llvm-profdata/vtable-value-prof-basic.test new file mode 100644 index 0000000..fb070dc --- /dev/null +++ b/llvm/test/tools/llvm-profdata/vtable-value-prof-basic.test @@ -0,0 +1,124 @@ +To update the inputs used below, run +Inputs/update_vtable_value_prof_inputs.sh /path/to/updated/clang++ + +; Raw profiles stores zlib-compressed vtable names. Raw profile reader needs +; to decompress them. +; REQUIRES: zlib + +; RUN: rm -rf %t && mkdir %t && cd %t + +Show profile data from raw profiles. +RUN: llvm-profdata show --function=main --ic-targets --show-vtables %p/Inputs/vtable-value-prof-basic.profraw | FileCheck %s --check-prefix=RAW + +Generate indexed profile from raw profile and show the data. +RUN: llvm-profdata merge %p/Inputs/vtable-value-prof-basic.profraw -o indexed.profdata +RUN: llvm-profdata show --function=main --ic-targets --show-vtables indexed.profdata | FileCheck %s --check-prefix=INDEXED + +Generate text profile from raw profile and show the data. +RUN: llvm-profdata merge --text %p/Inputs/vtable-value-prof-basic.profraw -o vtable-value-prof-basic.proftext +RUN: llvm-profdata show --function=main --ic-targets --show-vtables --text vtable-value-prof-basic.proftext | FileCheck %s --check-prefix=ICTEXT + +RAW: Counters: +RAW-NEXT: main: +RAW-NEXT: Hash: 0x0f9a16fe6d398548 +RAW-NEXT: Counters: 2 +RAW-NEXT: Indirect Call Site Count: 2 +RAW-NEXT: Number of instrumented vtables: 2 +RAW-NEXT: Indirect Target Results: +RAW-NEXT: [ 0, _ZN8Derived15func1Eii, 250 ] (25.00%) +RAW-NEXT: [ 0, {{.*}}vtable_prof.cc;_ZN12_GLOBAL__N_18Derived25func1Eii, 750 ] (75.00%) +RAW-NEXT: [ 1, _ZN8Derived15func2Eii, 250 ] (25.00%) +RAW-NEXT: [ 1, {{.*}}vtable_prof.cc;_ZN12_GLOBAL__N_18Derived25func2Eii, 750 ] (75.00%) +RAW-NEXT: VTable Results: +RAW-NEXT: [ 0, _ZTV8Derived1, 250 ] (25.00%) +RAW-NEXT: [ 0, {{.*}}vtable_prof.cc;_ZTVN12_GLOBAL__N_18Derived2E, 750 ] (75.00%) +RAW-NEXT: [ 1, _ZTV8Derived1, 250 ] (25.00%) +RAW-NEXT: [ 1, {{.*}}vtable_prof.cc;_ZTVN12_GLOBAL__N_18Derived2E, 750 ] (75.00%) +RAW-NEXT: Instrumentation level: IR entry_first = 0 +RAW-NEXT: Functions shown: 1 +RAW-NEXT: Total functions: 6 +RAW-NEXT: Maximum function count: 1000 +RAW-NEXT: Maximum internal block count: 250 +RAW-NEXT: Statistics for indirect call sites profile: +RAW-NEXT: Total number of sites: 2 +RAW-NEXT: Total number of sites with values: 2 +RAW-NEXT: Total number of profiled values: 4 +RAW-NEXT: Value sites histogram: +RAW-NEXT: NumTargets, SiteCount +RAW-NEXT: 2, 2 +RAW-NEXT: Statistics for vtable profile: +RAW-NEXT: Total number of sites: 2 +RAW-NEXT: Total number of sites with values: 2 +RAW-NEXT: Total number of profiled values: 4 +RAW-NEXT: Value sites histogram: +RAW-NEXT: NumTargets, SiteCount +RAW-NEXT: 2, 2 + + +INDEXED: Counters: +INDEXED-NEXT: main: +INDEXED-NEXT: Hash: 0x0f9a16fe6d398548 +INDEXED-NEXT: Counters: 2 +INDEXED-NEXT: Indirect Call Site Count: 2 +INDEXED-NEXT: Number of instrumented vtables: 2 +INDEXED-NEXT: Indirect Target Results: +INDEXED-NEXT: [ 0, {{.*}}vtable_prof.cc;_ZN12_GLOBAL__N_18Derived25func1Eii, 750 ] (75.00%) +INDEXED-NEXT: [ 0, _ZN8Derived15func1Eii, 250 ] (25.00%) +INDEXED-NEXT: [ 1, {{.*}}vtable_prof.cc;_ZN12_GLOBAL__N_18Derived25func2Eii, 750 ] (75.00%) +INDEXED-NEXT: [ 1, _ZN8Derived15func2Eii, 250 ] (25.00%) +INDEXED-NEXT: VTable Results: +INDEXED-NEXT: [ 0, {{.*}}vtable_prof.cc;_ZTVN12_GLOBAL__N_18Derived2E, 750 ] (75.00%) +INDEXED-NEXT: [ 0, _ZTV8Derived1, 250 ] (25.00%) +INDEXED-NEXT: [ 1, {{.*}}vtable_prof.cc;_ZTVN12_GLOBAL__N_18Derived2E, 750 ] (75.00%) +INDEXED-NEXT: [ 1, _ZTV8Derived1, 250 ] (25.00%) +INDEXED-NEXT: Instrumentation level: IR entry_first = 0 +INDEXED-NEXT: Functions shown: 1 +INDEXED-NEXT: Total functions: 6 +INDEXED-NEXT: Maximum function count: 1000 +INDEXED-NEXT: Maximum internal block count: 250 +INDEXED-NEXT: Statistics for indirect call sites profile: +INDEXED-NEXT: Total number of sites: 2 +INDEXED-NEXT: Total number of sites with values: 2 +INDEXED-NEXT: Total number of profiled values: 4 +INDEXED-NEXT: Value sites histogram: +INDEXED-NEXT: NumTargets, SiteCount +INDEXED-NEXT: 2, 2 +INDEXED-NEXT: Statistics for vtable profile: +INDEXED-NEXT: Total number of sites: 2 +INDEXED-NEXT: Total number of sites with values: 2 +INDEXED-NEXT: Total number of profiled values: 4 +INDEXED-NEXT: Value sites histogram: +INDEXED-NEXT: NumTargets, SiteCount +INDEXED-NEXT: 2, 2 + +ICTEXT: :ir +ICTEXT: main +ICTEXT: # Func Hash: +ICTEXT: 1124236338992350536 +ICTEXT: # Num Counters: +ICTEXT: 2 +ICTEXT: # Counter Values: +ICTEXT: 1000 +ICTEXT: 1 +ICTEXT: # Num Value Kinds: +ICTEXT: 2 +ICTEXT: # ValueKind = IPVK_IndirectCallTarget: +ICTEXT: 0 +ICTEXT: # NumValueSites: +ICTEXT: 2 +ICTEXT: 2 +ICTEXT: {{.*}}vtable_prof.cc;_ZN12_GLOBAL__N_18Derived25func1Eii:750 +ICTEXT: _ZN8Derived15func1Eii:250 +ICTEXT: 2 +ICTEXT: {{.*}}vtable_prof.cc;_ZN12_GLOBAL__N_18Derived25func2Eii:750 +ICTEXT: _ZN8Derived15func2Eii:250 +ICTEXT: # ValueKind = IPVK_VTableTarget: +ICTEXT: 2 +ICTEXT: # NumValueSites: +ICTEXT: 2 +ICTEXT: 2 +ICTEXT: {{.*}}vtable_prof.cc;_ZTVN12_GLOBAL__N_18Derived2E:750 +ICTEXT: _ZTV8Derived1:250 +ICTEXT: 2 +ICTEXT: {{.*}}vtable_prof.cc;_ZTVN12_GLOBAL__N_18Derived2E:750 +ICTEXT: _ZTV8Derived1:250 diff --git a/llvm/test/tools/llvm-profdata/vtable-value-prof.proftext b/llvm/test/tools/llvm-profdata/vtable-value-prof.proftext new file mode 100644 index 0000000..3807391 --- /dev/null +++ b/llvm/test/tools/llvm-profdata/vtable-value-prof.proftext @@ -0,0 +1,16 @@ +# RUN: llvm-profdata show --function=main --show-vtables %p/Inputs/vtable-value-prof.proftext | FileCheck %s + +# CHECK: Counters: +# CHECK: main: +# CHECK: Hash: 0x0f9a16fe6d398548 +# CHECK: Counters: 2 +# CHECK: VTable Results: +# CHECK: [ 0, /path/to/vtable_prof.cc;_ZTVN12_GLOBAL__N_18Derived2E, 750 ] (75.00%) +# CHECK: [ 0, _ZTV8Derived1, 250 ] (25.00%) +# CHECK: [ 1, /path/to/vtable_prof.cc;_ZTVN12_GLOBAL__N_18Derived2E, 750 ] (75.00%) +# CHECK: [ 1, _ZTV8Derived1, 250 ] (25.00%) +# CHECK: Instrumentation level: IR entry_first = 0 +# CHECK: Functions shown: 1 +# CHECK: Total functions: 6 +# CHECK: Maximum function count: 1000 +# CHECK: Maximum internal block count: 250 diff --git a/llvm/tools/llvm-profdata/llvm-profdata.cpp b/llvm/tools/llvm-profdata/llvm-profdata.cpp index 7754ca3..9fb56b8 100644 --- a/llvm/tools/llvm-profdata/llvm-profdata.cpp +++ b/llvm/tools/llvm-profdata/llvm-profdata.cpp @@ -345,6 +345,9 @@ cl::opt<bool> ShowIndirectCallTargets( "ic-targets", cl::init(false), cl::desc("Show indirect call site target values for shown functions"), cl::sub(ShowSubcommand)); +cl::opt<bool> ShowVTables("show-vtables", cl::init(false), + cl::desc("Show vtable names for shown functions"), + cl::sub(ShowSubcommand)); cl::opt<bool> ShowMemOPSizes( "memop-sizes", cl::init(false), cl::desc("Show the profiled sizes of the memory intrinsic calls " @@ -722,6 +725,13 @@ static void loadInput(const WeightedFile &Input, SymbolRemapper *Remapper, }); } + const InstrProfSymtab &symtab = Reader->getSymtab(); + const auto &VTableNames = symtab.getVTableNames(); + + for (const auto &kv : VTableNames) { + WC->Writer.addVTableName(kv.getKey()); + } + if (Reader->hasTemporalProfile()) { auto &Traces = Reader->getTemporalProfTraces(Input.Weight); if (!Traces.empty()) @@ -1353,8 +1363,8 @@ remapSamples(const sampleprof::FunctionSamples &Samples, BodySample.second.getSamples()); for (const auto &Target : BodySample.second.getCallTargets()) { Result.addCalledTargetSamples(BodySample.first.LineOffset, - MaskedDiscriminator, - Remapper(Target.first), Target.second); + MaskedDiscriminator, Remapper(Target.first), + Target.second); } } for (const auto &CallsiteSamples : Samples.getCallsiteSamples()) { @@ -2817,6 +2827,10 @@ static int showInstrProfile(ShowFormat SFormat, raw_fd_ostream &OS) { OS << " Indirect Call Site Count: " << Func.getNumValueSites(IPVK_IndirectCallTarget) << "\n"; + if (ShowVTables) + OS << " Number of instrumented vtables: " + << Func.getNumValueSites(IPVK_VTableTarget) << "\n"; + uint32_t NumMemOPCalls = Func.getNumValueSites(IPVK_MemOPSize); if (ShowMemOPSizes && NumMemOPCalls > 0) OS << " Number of Memory Intrinsics Calls: " << NumMemOPCalls @@ -2838,6 +2852,13 @@ static int showInstrProfile(ShowFormat SFormat, raw_fd_ostream &OS) { &(Reader->getSymtab())); } + if (ShowVTables) { + OS << " VTable Results:\n"; + traverseAllValueSites(Func, IPVK_VTableTarget, + VPStats[IPVK_VTableTarget], OS, + &(Reader->getSymtab())); + } + if (ShowMemOPSizes && NumMemOPCalls > 0) { OS << " Memory Intrinsic Size Results:\n"; traverseAllValueSites(Func, IPVK_MemOPSize, VPStats[IPVK_MemOPSize], OS, @@ -2886,6 +2907,11 @@ static int showInstrProfile(ShowFormat SFormat, raw_fd_ostream &OS) { VPStats[IPVK_IndirectCallTarget]); } + if (ShownFunctions && ShowVTables) { + OS << "Statistics for vtable profile:\n"; + showValueSitesStats(OS, IPVK_VTableTarget, VPStats[IPVK_VTableTarget]); + } + if (ShownFunctions && ShowMemOPSizes) { OS << "Statistics for memory intrinsic calls sizes profile:\n"; showValueSitesStats(OS, IPVK_MemOPSize, VPStats[IPVK_MemOPSize]); diff --git a/llvm/unittests/ProfileData/InstrProfTest.cpp b/llvm/unittests/ProfileData/InstrProfTest.cpp index 8ffb68d..f0281bd 100644 --- a/llvm/unittests/ProfileData/InstrProfTest.cpp +++ b/llvm/unittests/ProfileData/InstrProfTest.cpp @@ -6,6 +6,7 @@ // //===----------------------------------------------------------------------===// +#include "llvm/IR/DerivedTypes.h" #include "llvm/IR/Function.h" #include "llvm/IR/IRBuilder.h" #include "llvm/IR/LLVMContext.h" @@ -638,32 +639,78 @@ TEST_F(InstrProfTest, test_irpgo_read_deprecated_names) { Succeeded()); } +// callee1 to callee6 are from vtable1 to vtable6 respectively. static const char callee1[] = "callee1"; static const char callee2[] = "callee2"; static const char callee3[] = "callee3"; static const char callee4[] = "callee4"; static const char callee5[] = "callee5"; static const char callee6[] = "callee6"; +// callee7 and callee8 are not from any vtables. +static const char callee7[] = "callee7"; +static const char callee8[] = "callee8"; +// 'callee' is primarily used to create multiple-element vtables. +static const char callee[] = "callee"; +static const uint64_t vtable1[] = {uint64_t(callee), uint64_t(callee1)}; +static const uint64_t vtable2[] = {uint64_t(callee2), uint64_t(callee)}; +static const uint64_t vtable3[] = { + uint64_t(callee), + uint64_t(callee3), +}; +static const uint64_t vtable4[] = {uint64_t(callee4), uint64_t(callee)}; +static const uint64_t vtable5[] = {uint64_t(callee5), uint64_t(callee)}; +static const uint64_t vtable6[] = {uint64_t(callee6), uint64_t(callee)}; + +// Returns the address of callee with a numbered suffix in vtable. +static uint64_t getCalleeAddress(const uint64_t *vtableAddr) { + uint64_t CalleeAddr; + // Callee with a numbered suffix is the 2nd element in vtable1 and vtable3, + // and the 1st element in the rest of vtables. + if (vtableAddr == vtable1 || vtableAddr == vtable3) + CalleeAddr = uint64_t(vtableAddr) + 8; + else + CalleeAddr = uint64_t(vtableAddr); + return CalleeAddr; +} -TEST_P(InstrProfReaderWriterTest, icall_data_read_write) { +TEST_P(InstrProfReaderWriterTest, icall_and_vtable_data_read_write) { NamedInstrProfRecord Record1("caller", 0x1234, {1, 2}); - // 4 value sites. - Record1.reserveSites(IPVK_IndirectCallTarget, 4); - InstrProfValueData VD0[] = { - {(uint64_t)callee1, 1}, {(uint64_t)callee2, 2}, {(uint64_t)callee3, 3}}; - Record1.addValueData(IPVK_IndirectCallTarget, 0, VD0, 3, nullptr); - // No value profile data at the second site. - Record1.addValueData(IPVK_IndirectCallTarget, 1, nullptr, 0, nullptr); - InstrProfValueData VD2[] = {{(uint64_t)callee1, 1}, {(uint64_t)callee2, 2}}; - Record1.addValueData(IPVK_IndirectCallTarget, 2, VD2, 2, nullptr); - InstrProfValueData VD3[] = {{(uint64_t)callee1, 1}}; - Record1.addValueData(IPVK_IndirectCallTarget, 3, VD3, 1, nullptr); + // 4 indirect call value sites. + { + Record1.reserveSites(IPVK_IndirectCallTarget, 4); + InstrProfValueData VD0[] = { + {(uint64_t)callee1, 1}, {(uint64_t)callee2, 2}, {(uint64_t)callee3, 3}}; + Record1.addValueData(IPVK_IndirectCallTarget, 0, VD0, 3, nullptr); + // No value profile data at the second site. + Record1.addValueData(IPVK_IndirectCallTarget, 1, nullptr, 0, nullptr); + InstrProfValueData VD2[] = {{(uint64_t)callee1, 1}, {(uint64_t)callee2, 2}}; + Record1.addValueData(IPVK_IndirectCallTarget, 2, VD2, 2, nullptr); + InstrProfValueData VD3[] = {{(uint64_t)callee7, 1}, {(uint64_t)callee8, 2}}; + Record1.addValueData(IPVK_IndirectCallTarget, 3, VD3, 2, nullptr); + } + + // 2 vtable value sites. + { + InstrProfValueData VD0[] = { + {getCalleeAddress(vtable1), 1}, + {getCalleeAddress(vtable2), 2}, + {getCalleeAddress(vtable3), 3}, + }; + InstrProfValueData VD2[] = { + {getCalleeAddress(vtable1), 1}, + {getCalleeAddress(vtable2), 2}, + }; + Record1.addValueData(IPVK_VTableTarget, 0, VD0, 3, nullptr); + Record1.addValueData(IPVK_VTableTarget, 2, VD2, 2, nullptr); + } Writer.addRecord(std::move(Record1), getProfWeight(), Err); Writer.addRecord({"callee1", 0x1235, {3, 4}}, Err); Writer.addRecord({"callee2", 0x1235, {3, 4}}, Err); Writer.addRecord({"callee3", 0x1235, {3, 4}}, Err); + Writer.addRecord({"callee7", 0x1235, {3, 4}}, Err); + Writer.addRecord({"callee8", 0x1235, {3, 4}}, Err); // Set writer value prof data endianness. Writer.setValueProfDataEndianness(getEndianness()); @@ -676,24 +723,66 @@ TEST_P(InstrProfReaderWriterTest, icall_data_read_write) { Expected<InstrProfRecord> R = Reader->getInstrProfRecord("caller", 0x1234); ASSERT_THAT_ERROR(R.takeError(), Succeeded()); + + // Test the number of instrumented indirect call sites and the number of + // profiled values at each site. ASSERT_EQ(4U, R->getNumValueSites(IPVK_IndirectCallTarget)); EXPECT_EQ(3U, R->getNumValueDataForSite(IPVK_IndirectCallTarget, 0)); EXPECT_EQ(0U, R->getNumValueDataForSite(IPVK_IndirectCallTarget, 1)); EXPECT_EQ(2U, R->getNumValueDataForSite(IPVK_IndirectCallTarget, 2)); - EXPECT_EQ(1U, R->getNumValueDataForSite(IPVK_IndirectCallTarget, 3)); + EXPECT_EQ(2U, R->getNumValueDataForSite(IPVK_IndirectCallTarget, 3)); + + // Test the number of instrumented vtable sites and the number of profiled + // values at each site. + ASSERT_EQ(2U, R->getNumValueSites(IPVK_VTableTarget)); + EXPECT_EQ(3U, R->getNumValueDataForSite(IPVK_VTableTarget, 0)); + EXPECT_EQ(2U, R->getNumValueDataForSite(IPVK_VTableTarget, 1)); + + // First indirect site. + { + uint64_t TotalC; + std::unique_ptr<InstrProfValueData[]> VD = + R->getValueForSite(IPVK_IndirectCallTarget, 0, &TotalC); + + EXPECT_EQ(3U * getProfWeight(), VD[0].Count); + EXPECT_EQ(2U * getProfWeight(), VD[1].Count); + EXPECT_EQ(1U * getProfWeight(), VD[2].Count); + EXPECT_EQ(6U * getProfWeight(), TotalC); + + EXPECT_EQ(StringRef((const char *)VD[0].Value, 7), StringRef("callee3")); + EXPECT_EQ(StringRef((const char *)VD[1].Value, 7), StringRef("callee2")); + EXPECT_EQ(StringRef((const char *)VD[2].Value, 7), StringRef("callee1")); + } - uint64_t TotalC; - std::unique_ptr<InstrProfValueData[]> VD = - R->getValueForSite(IPVK_IndirectCallTarget, 0, &TotalC); + // First vtable site. + { + uint64_t TotalC; + std::unique_ptr<InstrProfValueData[]> VD = + R->getValueForSite(IPVK_VTableTarget, 0, &TotalC); + + EXPECT_EQ(3U * getProfWeight(), VD[0].Count); + EXPECT_EQ(2U * getProfWeight(), VD[1].Count); + EXPECT_EQ(1U * getProfWeight(), VD[2].Count); + EXPECT_EQ(6U * getProfWeight(), TotalC); + + EXPECT_EQ(VD[0].Value, getCalleeAddress(vtable3)); + EXPECT_EQ(VD[1].Value, getCalleeAddress(vtable2)); + EXPECT_EQ(VD[2].Value, getCalleeAddress(vtable1)); + } - EXPECT_EQ(3U * getProfWeight(), VD[0].Count); - EXPECT_EQ(2U * getProfWeight(), VD[1].Count); - EXPECT_EQ(1U * getProfWeight(), VD[2].Count); - EXPECT_EQ(6U * getProfWeight(), TotalC); + // Second vtable site. + { + uint64_t TotalC; + std::unique_ptr<InstrProfValueData[]> VD = + R->getValueForSite(IPVK_VTableTarget, 1, &TotalC); - EXPECT_EQ(StringRef((const char *)VD[0].Value, 7), StringRef("callee3")); - EXPECT_EQ(StringRef((const char *)VD[1].Value, 7), StringRef("callee2")); - EXPECT_EQ(StringRef((const char *)VD[2].Value, 7), StringRef("callee1")); + EXPECT_EQ(2U * getProfWeight(), VD[0].Count); + EXPECT_EQ(1U * getProfWeight(), VD[1].Count); + EXPECT_EQ(3U * getProfWeight(), TotalC); + + EXPECT_EQ(VD[0].Value, getCalleeAddress(vtable2)); + EXPECT_EQ(VD[1].Value, getCalleeAddress(vtable1)); + } } INSTANTIATE_TEST_SUITE_P( @@ -801,33 +890,53 @@ TEST_P(MaybeSparseInstrProfTest, annotate_vp_data) { ASSERT_EQ(1U, ValueData[3].Count); } -TEST_P(MaybeSparseInstrProfTest, icall_data_merge) { +TEST_P(MaybeSparseInstrProfTest, icall_and_vtable_data_merge) { static const char caller[] = "caller"; NamedInstrProfRecord Record11(caller, 0x1234, {1, 2}); NamedInstrProfRecord Record12(caller, 0x1234, {1, 2}); - // 5 value sites. - Record11.reserveSites(IPVK_IndirectCallTarget, 5); - InstrProfValueData VD0[] = {{uint64_t(callee1), 1}, - {uint64_t(callee2), 2}, - {uint64_t(callee3), 3}, - {uint64_t(callee4), 4}}; - Record11.addValueData(IPVK_IndirectCallTarget, 0, VD0, 4, nullptr); + // 5 value sites for indirect calls. + { + Record11.reserveSites(IPVK_IndirectCallTarget, 5); + InstrProfValueData VD0[] = {{uint64_t(callee1), 1}, + {uint64_t(callee2), 2}, + {uint64_t(callee3), 3}, + {uint64_t(callee4), 4}}; + Record11.addValueData(IPVK_IndirectCallTarget, 0, VD0, 4, nullptr); - // No value profile data at the second site. - Record11.addValueData(IPVK_IndirectCallTarget, 1, nullptr, 0, nullptr); + // No value profile data at the second site. + Record11.addValueData(IPVK_IndirectCallTarget, 1, nullptr, 0, nullptr); - InstrProfValueData VD2[] = { - {uint64_t(callee1), 1}, {uint64_t(callee2), 2}, {uint64_t(callee3), 3}}; - Record11.addValueData(IPVK_IndirectCallTarget, 2, VD2, 3, nullptr); + InstrProfValueData VD2[] = { + {uint64_t(callee1), 1}, {uint64_t(callee2), 2}, {uint64_t(callee3), 3}}; + Record11.addValueData(IPVK_IndirectCallTarget, 2, VD2, 3, nullptr); - InstrProfValueData VD3[] = {{uint64_t(callee1), 1}}; - Record11.addValueData(IPVK_IndirectCallTarget, 3, VD3, 1, nullptr); + InstrProfValueData VD3[] = {{uint64_t(callee7), 1}, {uint64_t(callee8), 2}}; + Record11.addValueData(IPVK_IndirectCallTarget, 3, VD3, 2, nullptr); - InstrProfValueData VD4[] = {{uint64_t(callee1), 1}, - {uint64_t(callee2), 2}, - {uint64_t(callee3), 3}}; - Record11.addValueData(IPVK_IndirectCallTarget, 4, VD4, 3, nullptr); + InstrProfValueData VD4[] = { + {uint64_t(callee1), 1}, {uint64_t(callee2), 2}, {uint64_t(callee3), 3}}; + Record11.addValueData(IPVK_IndirectCallTarget, 4, VD4, 3, nullptr); + } + // 3 value sites for vtables. + { + Record11.reserveSites(IPVK_VTableTarget, 3); + InstrProfValueData VD0[] = {{getCalleeAddress(vtable1), 1}, + {getCalleeAddress(vtable2), 2}, + {getCalleeAddress(vtable3), 3}, + {getCalleeAddress(vtable4), 4}}; + Record11.addValueData(IPVK_VTableTarget, 0, VD0, 4, nullptr); + + InstrProfValueData VD2[] = {{getCalleeAddress(vtable1), 1}, + {getCalleeAddress(vtable2), 2}, + {getCalleeAddress(vtable3), 3}}; + Record11.addValueData(IPVK_VTableTarget, 1, VD2, 3, nullptr); + + InstrProfValueData VD4[] = {{getCalleeAddress(vtable1), 1}, + {getCalleeAddress(vtable2), 2}, + {getCalleeAddress(vtable3), 3}}; + Record11.addValueData(IPVK_VTableTarget, 3, VD4, 3, nullptr); + } // A different record for the same caller. Record12.reserveSites(IPVK_IndirectCallTarget, 5); @@ -843,11 +952,28 @@ TEST_P(MaybeSparseInstrProfTest, icall_data_merge) { Record12.addValueData(IPVK_IndirectCallTarget, 3, nullptr, 0, nullptr); - InstrProfValueData VD42[] = {{uint64_t(callee1), 1}, - {uint64_t(callee2), 2}, - {uint64_t(callee3), 3}}; + InstrProfValueData VD42[] = { + {uint64_t(callee1), 1}, {uint64_t(callee2), 2}, {uint64_t(callee3), 3}}; Record12.addValueData(IPVK_IndirectCallTarget, 4, VD42, 3, nullptr); + // 3 value sites for vtables. + { + Record12.reserveSites(IPVK_VTableTarget, 3); + InstrProfValueData VD0[] = {{getCalleeAddress(vtable2), 5}, + {getCalleeAddress(vtable3), 3}}; + Record12.addValueData(IPVK_VTableTarget, 0, VD0, 2, nullptr); + + InstrProfValueData VD2[] = {{getCalleeAddress(vtable2), 1}, + {getCalleeAddress(vtable3), 3}, + {getCalleeAddress(vtable4), 4}}; + Record12.addValueData(IPVK_VTableTarget, 1, VD2, 3, nullptr); + + InstrProfValueData VD4[] = {{getCalleeAddress(vtable1), 1}, + {getCalleeAddress(vtable2), 2}, + {getCalleeAddress(vtable3), 3}}; + Record12.addValueData(IPVK_VTableTarget, 3, VD4, 3, nullptr); + } + Writer.addRecord(std::move(Record11), Err); // Merge profile data. Writer.addRecord(std::move(Record12), Err); @@ -857,53 +983,99 @@ TEST_P(MaybeSparseInstrProfTest, icall_data_merge) { Writer.addRecord({callee3, 0x1235, {3, 4}}, Err); Writer.addRecord({callee3, 0x1235, {3, 4}}, Err); Writer.addRecord({callee4, 0x1235, {3, 5}}, Err); + Writer.addRecord({callee7, 0x1235, {3, 5}}, Err); + Writer.addRecord({callee8, 0x1235, {3, 5}}, Err); auto Profile = Writer.writeBuffer(); readProfile(std::move(Profile)); + // Test the number of instrumented value sites and the number of profiled + // values for each site. Expected<InstrProfRecord> R = Reader->getInstrProfRecord("caller", 0x1234); EXPECT_THAT_ERROR(R.takeError(), Succeeded()); + // For indirect calls. ASSERT_EQ(5U, R->getNumValueSites(IPVK_IndirectCallTarget)); ASSERT_EQ(4U, R->getNumValueDataForSite(IPVK_IndirectCallTarget, 0)); ASSERT_EQ(0U, R->getNumValueDataForSite(IPVK_IndirectCallTarget, 1)); ASSERT_EQ(4U, R->getNumValueDataForSite(IPVK_IndirectCallTarget, 2)); - ASSERT_EQ(1U, R->getNumValueDataForSite(IPVK_IndirectCallTarget, 3)); + ASSERT_EQ(2U, R->getNumValueDataForSite(IPVK_IndirectCallTarget, 3)); ASSERT_EQ(3U, R->getNumValueDataForSite(IPVK_IndirectCallTarget, 4)); + // For vtables. + ASSERT_EQ(3U, R->getNumValueSites(IPVK_VTableTarget)); + ASSERT_EQ(4U, R->getNumValueDataForSite(IPVK_VTableTarget, 0)); + ASSERT_EQ(4U, R->getNumValueDataForSite(IPVK_VTableTarget, 1)); + ASSERT_EQ(3U, R->getNumValueDataForSite(IPVK_VTableTarget, 2)); + + // Test the merged values for indirect calls. + { + std::unique_ptr<InstrProfValueData[]> VD = + R->getValueForSite(IPVK_IndirectCallTarget, 0); + EXPECT_EQ(StringRef((const char *)VD[0].Value, 7), StringRef("callee2")); + EXPECT_EQ(7U, VD[0].Count); + EXPECT_EQ(StringRef((const char *)VD[1].Value, 7), StringRef("callee3")); + EXPECT_EQ(6U, VD[1].Count); + EXPECT_EQ(StringRef((const char *)VD[2].Value, 7), StringRef("callee4")); + EXPECT_EQ(4U, VD[2].Count); + EXPECT_EQ(StringRef((const char *)VD[3].Value, 7), StringRef("callee1")); + EXPECT_EQ(1U, VD[3].Count); + + std::unique_ptr<InstrProfValueData[]> VD_2( + R->getValueForSite(IPVK_IndirectCallTarget, 2)); + EXPECT_EQ(StringRef((const char *)VD_2[0].Value, 7), StringRef("callee3")); + EXPECT_EQ(6U, VD_2[0].Count); + EXPECT_EQ(StringRef((const char *)VD_2[1].Value, 7), StringRef("callee4")); + EXPECT_EQ(4U, VD_2[1].Count); + EXPECT_EQ(StringRef((const char *)VD_2[2].Value, 7), StringRef("callee2")); + EXPECT_EQ(3U, VD_2[2].Count); + EXPECT_EQ(StringRef((const char *)VD_2[3].Value, 7), StringRef("callee1")); + EXPECT_EQ(1U, VD_2[3].Count); + + std::unique_ptr<InstrProfValueData[]> VD_3( + R->getValueForSite(IPVK_IndirectCallTarget, 3)); + EXPECT_EQ(StringRef((const char *)VD_3[0].Value, 7), StringRef("callee8")); + EXPECT_EQ(2U, VD_3[0].Count); + EXPECT_EQ(StringRef((const char *)VD_3[1].Value, 7), StringRef("callee7")); + EXPECT_EQ(1U, VD_3[1].Count); + + std::unique_ptr<InstrProfValueData[]> VD_4( + R->getValueForSite(IPVK_IndirectCallTarget, 4)); + EXPECT_EQ(StringRef((const char *)VD_4[0].Value, 7), StringRef("callee3")); + EXPECT_EQ(6U, VD_4[0].Count); + EXPECT_EQ(StringRef((const char *)VD_4[1].Value, 7), StringRef("callee2")); + EXPECT_EQ(4U, VD_4[1].Count); + EXPECT_EQ(StringRef((const char *)VD_4[2].Value, 7), StringRef("callee1")); + EXPECT_EQ(2U, VD_4[2].Count); + } - std::unique_ptr<InstrProfValueData[]> VD = - R->getValueForSite(IPVK_IndirectCallTarget, 0); - ASSERT_EQ(StringRef((const char *)VD[0].Value, 7), StringRef("callee2")); - ASSERT_EQ(7U, VD[0].Count); - ASSERT_EQ(StringRef((const char *)VD[1].Value, 7), StringRef("callee3")); - ASSERT_EQ(6U, VD[1].Count); - ASSERT_EQ(StringRef((const char *)VD[2].Value, 7), StringRef("callee4")); - ASSERT_EQ(4U, VD[2].Count); - ASSERT_EQ(StringRef((const char *)VD[3].Value, 7), StringRef("callee1")); - ASSERT_EQ(1U, VD[3].Count); - - std::unique_ptr<InstrProfValueData[]> VD_2( - R->getValueForSite(IPVK_IndirectCallTarget, 2)); - ASSERT_EQ(StringRef((const char *)VD_2[0].Value, 7), StringRef("callee3")); - ASSERT_EQ(6U, VD_2[0].Count); - ASSERT_EQ(StringRef((const char *)VD_2[1].Value, 7), StringRef("callee4")); - ASSERT_EQ(4U, VD_2[1].Count); - ASSERT_EQ(StringRef((const char *)VD_2[2].Value, 7), StringRef("callee2")); - ASSERT_EQ(3U, VD_2[2].Count); - ASSERT_EQ(StringRef((const char *)VD_2[3].Value, 7), StringRef("callee1")); - ASSERT_EQ(1U, VD_2[3].Count); - - std::unique_ptr<InstrProfValueData[]> VD_3( - R->getValueForSite(IPVK_IndirectCallTarget, 3)); - ASSERT_EQ(StringRef((const char *)VD_3[0].Value, 7), StringRef("callee1")); - ASSERT_EQ(1U, VD_3[0].Count); - - std::unique_ptr<InstrProfValueData[]> VD_4( - R->getValueForSite(IPVK_IndirectCallTarget, 4)); - ASSERT_EQ(StringRef((const char *)VD_4[0].Value, 7), StringRef("callee3")); - ASSERT_EQ(6U, VD_4[0].Count); - ASSERT_EQ(StringRef((const char *)VD_4[1].Value, 7), StringRef("callee2")); - ASSERT_EQ(4U, VD_4[1].Count); - ASSERT_EQ(StringRef((const char *)VD_4[2].Value, 7), StringRef("callee1")); - ASSERT_EQ(2U, VD_4[2].Count); + // Test the merged values for vtables + { + auto VD0 = R->getValueForSite(IPVK_VTableTarget, 0); + EXPECT_EQ(VD0[0].Value, getCalleeAddress(vtable2)); + EXPECT_EQ(VD0[0].Count, 7U); + EXPECT_EQ(VD0[1].Value, getCalleeAddress(vtable3)); + EXPECT_EQ(VD0[1].Count, 6U); + EXPECT_EQ(VD0[2].Value, getCalleeAddress(vtable4)); + EXPECT_EQ(VD0[2].Count, 4U); + EXPECT_EQ(VD0[3].Value, getCalleeAddress(vtable1)); + EXPECT_EQ(VD0[3].Count, 1U); + + auto VD1 = R->getValueForSite(IPVK_VTableTarget, 1); + EXPECT_EQ(VD1[0].Value, getCalleeAddress(vtable3)); + EXPECT_EQ(VD1[0].Count, 6U); + EXPECT_EQ(VD1[1].Value, getCalleeAddress(vtable4)); + EXPECT_EQ(VD1[1].Count, 4U); + EXPECT_EQ(VD1[2].Value, getCalleeAddress(vtable2)); + EXPECT_EQ(VD1[2].Count, 3U); + EXPECT_EQ(VD1[3].Value, getCalleeAddress(vtable1)); + EXPECT_EQ(VD1[3].Count, 1U); + + auto VD2 = R->getValueForSite(IPVK_VTableTarget, 2); + EXPECT_EQ(VD2[0].Value, getCalleeAddress(vtable3)); + EXPECT_EQ(VD2[0].Count, 6U); + EXPECT_EQ(VD2[1].Value, getCalleeAddress(vtable2)); + EXPECT_EQ(VD2[1].Count, 4U); + EXPECT_EQ(VD2[2].Value, getCalleeAddress(vtable1)); + EXPECT_EQ(VD2[2].Count, 2U); + } } struct ValueProfileMergeEdgeCaseTest @@ -1027,30 +1199,62 @@ INSTANTIATE_TEST_SUITE_P( EdgeCaseTest, ValueProfileMergeEdgeCaseTest, ::testing::Combine(::testing::Bool(), /* Sparse */ ::testing::Values(IPVK_IndirectCallTarget, - IPVK_MemOPSize) /* ValueKind */ + IPVK_MemOPSize, + IPVK_VTableTarget) /* ValueKind */ )); static void addValueProfData(InstrProfRecord &Record) { - Record.reserveSites(IPVK_IndirectCallTarget, 5); - InstrProfValueData VD0[] = {{uint64_t(callee1), 400}, - {uint64_t(callee2), 1000}, - {uint64_t(callee3), 500}, - {uint64_t(callee4), 300}, - {uint64_t(callee5), 100}}; - Record.addValueData(IPVK_IndirectCallTarget, 0, VD0, 5, nullptr); - InstrProfValueData VD1[] = {{uint64_t(callee5), 800}, - {uint64_t(callee3), 1000}, - {uint64_t(callee2), 2500}, - {uint64_t(callee1), 1300}}; - Record.addValueData(IPVK_IndirectCallTarget, 1, VD1, 4, nullptr); - InstrProfValueData VD2[] = {{uint64_t(callee6), 800}, - {uint64_t(callee3), 1000}, - {uint64_t(callee4), 5500}}; - Record.addValueData(IPVK_IndirectCallTarget, 2, VD2, 3, nullptr); - InstrProfValueData VD3[] = {{uint64_t(callee2), 1800}, - {uint64_t(callee3), 2000}}; - Record.addValueData(IPVK_IndirectCallTarget, 3, VD3, 2, nullptr); - Record.addValueData(IPVK_IndirectCallTarget, 4, nullptr, 0, nullptr); + // Add test data for indirect calls. + { + Record.reserveSites(IPVK_IndirectCallTarget, 6); + InstrProfValueData VD0[] = {{uint64_t(callee1), 400}, + {uint64_t(callee2), 1000}, + {uint64_t(callee3), 500}, + {uint64_t(callee4), 300}, + {uint64_t(callee5), 100}}; + Record.addValueData(IPVK_IndirectCallTarget, 0, VD0, 5, nullptr); + InstrProfValueData VD1[] = {{uint64_t(callee5), 800}, + {uint64_t(callee3), 1000}, + {uint64_t(callee2), 2500}, + {uint64_t(callee1), 1300}}; + Record.addValueData(IPVK_IndirectCallTarget, 1, VD1, 4, nullptr); + InstrProfValueData VD2[] = {{uint64_t(callee6), 800}, + {uint64_t(callee3), 1000}, + {uint64_t(callee4), 5500}}; + Record.addValueData(IPVK_IndirectCallTarget, 2, VD2, 3, nullptr); + InstrProfValueData VD3[] = {{uint64_t(callee2), 1800}, + {uint64_t(callee3), 2000}}; + Record.addValueData(IPVK_IndirectCallTarget, 3, VD3, 2, nullptr); + Record.addValueData(IPVK_IndirectCallTarget, 4, nullptr, 0, nullptr); + InstrProfValueData VD5[] = {{uint64_t(callee7), 1234}, + {uint64_t(callee8), 5678}}; + Record.addValueData(IPVK_IndirectCallTarget, 5, VD5, 2, nullptr); + } + + // Add test data for vtables + { + Record.reserveSites(IPVK_VTableTarget, 4); + InstrProfValueData VD0[] = { + {getCalleeAddress(vtable1), 400}, {getCalleeAddress(vtable2), 1000}, + {getCalleeAddress(vtable3), 500}, {getCalleeAddress(vtable4), 300}, + {getCalleeAddress(vtable5), 100}, + }; + InstrProfValueData VD1[] = {{getCalleeAddress(vtable5), 800}, + {getCalleeAddress(vtable3), 1000}, + {getCalleeAddress(vtable2), 2500}, + {getCalleeAddress(vtable1), 1300}}; + InstrProfValueData VD2[] = { + {getCalleeAddress(vtable6), 800}, + {getCalleeAddress(vtable3), 1000}, + {getCalleeAddress(vtable4), 5500}, + }; + InstrProfValueData VD3[] = {{getCalleeAddress(vtable2), 1800}, + {getCalleeAddress(vtable3), 2000}}; + Record.addValueData(IPVK_VTableTarget, 0, VD0, 5, nullptr); + Record.addValueData(IPVK_VTableTarget, 1, VD1, 4, nullptr); + Record.addValueData(IPVK_VTableTarget, 2, VD2, 3, nullptr); + Record.addValueData(IPVK_VTableTarget, 3, VD3, 2, nullptr); + } } TEST(ValueProfileReadWriteTest, value_prof_data_read_write) { @@ -1063,59 +1267,111 @@ TEST(ValueProfileReadWriteTest, value_prof_data_read_write) { VPData->deserializeTo(Record, nullptr); // Now read data from Record and sanity check the data - ASSERT_EQ(5U, Record.getNumValueSites(IPVK_IndirectCallTarget)); + ASSERT_EQ(6U, Record.getNumValueSites(IPVK_IndirectCallTarget)); ASSERT_EQ(5U, Record.getNumValueDataForSite(IPVK_IndirectCallTarget, 0)); ASSERT_EQ(4U, Record.getNumValueDataForSite(IPVK_IndirectCallTarget, 1)); ASSERT_EQ(3U, Record.getNumValueDataForSite(IPVK_IndirectCallTarget, 2)); ASSERT_EQ(2U, Record.getNumValueDataForSite(IPVK_IndirectCallTarget, 3)); ASSERT_EQ(0U, Record.getNumValueDataForSite(IPVK_IndirectCallTarget, 4)); + ASSERT_EQ(2U, Record.getNumValueDataForSite(IPVK_IndirectCallTarget, 5)); auto Cmp = [](const InstrProfValueData &VD1, const InstrProfValueData &VD2) { return VD1.Count > VD2.Count; }; + std::unique_ptr<InstrProfValueData[]> VD_0( Record.getValueForSite(IPVK_IndirectCallTarget, 0)); llvm::sort(&VD_0[0], &VD_0[5], Cmp); - ASSERT_EQ(StringRef((const char *)VD_0[0].Value, 7), StringRef("callee2")); - ASSERT_EQ(1000U, VD_0[0].Count); - ASSERT_EQ(StringRef((const char *)VD_0[1].Value, 7), StringRef("callee3")); - ASSERT_EQ(500U, VD_0[1].Count); - ASSERT_EQ(StringRef((const char *)VD_0[2].Value, 7), StringRef("callee1")); - ASSERT_EQ(400U, VD_0[2].Count); - ASSERT_EQ(StringRef((const char *)VD_0[3].Value, 7), StringRef("callee4")); - ASSERT_EQ(300U, VD_0[3].Count); - ASSERT_EQ(StringRef((const char *)VD_0[4].Value, 7), StringRef("callee5")); - ASSERT_EQ(100U, VD_0[4].Count); + EXPECT_EQ(StringRef((const char *)VD_0[0].Value, 7), StringRef("callee2")); + EXPECT_EQ(1000U, VD_0[0].Count); + EXPECT_EQ(StringRef((const char *)VD_0[1].Value, 7), StringRef("callee3")); + EXPECT_EQ(500U, VD_0[1].Count); + EXPECT_EQ(StringRef((const char *)VD_0[2].Value, 7), StringRef("callee1")); + EXPECT_EQ(400U, VD_0[2].Count); + EXPECT_EQ(StringRef((const char *)VD_0[3].Value, 7), StringRef("callee4")); + EXPECT_EQ(300U, VD_0[3].Count); + EXPECT_EQ(StringRef((const char *)VD_0[4].Value, 7), StringRef("callee5")); + EXPECT_EQ(100U, VD_0[4].Count); std::unique_ptr<InstrProfValueData[]> VD_1( Record.getValueForSite(IPVK_IndirectCallTarget, 1)); llvm::sort(&VD_1[0], &VD_1[4], Cmp); - ASSERT_EQ(StringRef((const char *)VD_1[0].Value, 7), StringRef("callee2")); - ASSERT_EQ(2500U, VD_1[0].Count); - ASSERT_EQ(StringRef((const char *)VD_1[1].Value, 7), StringRef("callee1")); - ASSERT_EQ(1300U, VD_1[1].Count); - ASSERT_EQ(StringRef((const char *)VD_1[2].Value, 7), StringRef("callee3")); - ASSERT_EQ(1000U, VD_1[2].Count); - ASSERT_EQ(StringRef((const char *)VD_1[3].Value, 7), StringRef("callee5")); - ASSERT_EQ(800U, VD_1[3].Count); + EXPECT_EQ(StringRef((const char *)VD_1[0].Value, 7), StringRef("callee2")); + EXPECT_EQ(2500U, VD_1[0].Count); + EXPECT_EQ(StringRef((const char *)VD_1[1].Value, 7), StringRef("callee1")); + EXPECT_EQ(1300U, VD_1[1].Count); + EXPECT_EQ(StringRef((const char *)VD_1[2].Value, 7), StringRef("callee3")); + EXPECT_EQ(1000U, VD_1[2].Count); + EXPECT_EQ(StringRef((const char *)VD_1[3].Value, 7), StringRef("callee5")); + EXPECT_EQ(800U, VD_1[3].Count); std::unique_ptr<InstrProfValueData[]> VD_2( Record.getValueForSite(IPVK_IndirectCallTarget, 2)); llvm::sort(&VD_2[0], &VD_2[3], Cmp); - ASSERT_EQ(StringRef((const char *)VD_2[0].Value, 7), StringRef("callee4")); - ASSERT_EQ(5500U, VD_2[0].Count); - ASSERT_EQ(StringRef((const char *)VD_2[1].Value, 7), StringRef("callee3")); - ASSERT_EQ(1000U, VD_2[1].Count); - ASSERT_EQ(StringRef((const char *)VD_2[2].Value, 7), StringRef("callee6")); - ASSERT_EQ(800U, VD_2[2].Count); + EXPECT_EQ(StringRef((const char *)VD_2[0].Value, 7), StringRef("callee4")); + EXPECT_EQ(5500U, VD_2[0].Count); + EXPECT_EQ(StringRef((const char *)VD_2[1].Value, 7), StringRef("callee3")); + EXPECT_EQ(1000U, VD_2[1].Count); + EXPECT_EQ(StringRef((const char *)VD_2[2].Value, 7), StringRef("callee6")); + EXPECT_EQ(800U, VD_2[2].Count); std::unique_ptr<InstrProfValueData[]> VD_3( Record.getValueForSite(IPVK_IndirectCallTarget, 3)); llvm::sort(&VD_3[0], &VD_3[2], Cmp); - ASSERT_EQ(StringRef((const char *)VD_3[0].Value, 7), StringRef("callee3")); - ASSERT_EQ(2000U, VD_3[0].Count); - ASSERT_EQ(StringRef((const char *)VD_3[1].Value, 7), StringRef("callee2")); - ASSERT_EQ(1800U, VD_3[1].Count); + EXPECT_EQ(StringRef((const char *)VD_3[0].Value, 7), StringRef("callee3")); + EXPECT_EQ(2000U, VD_3[0].Count); + EXPECT_EQ(StringRef((const char *)VD_3[1].Value, 7), StringRef("callee2")); + EXPECT_EQ(1800U, VD_3[1].Count); + + ASSERT_EQ(4U, Record.getNumValueSites(IPVK_VTableTarget)); + ASSERT_EQ(5U, Record.getNumValueDataForSite(IPVK_VTableTarget, 0)); + ASSERT_EQ(4U, Record.getNumValueDataForSite(IPVK_VTableTarget, 1)); + ASSERT_EQ(3U, Record.getNumValueDataForSite(IPVK_VTableTarget, 2)); + ASSERT_EQ(2U, Record.getNumValueDataForSite(IPVK_VTableTarget, 3)); + + std::unique_ptr<InstrProfValueData[]> VD0( + Record.getValueForSite(IPVK_VTableTarget, 0)); + llvm::sort(&VD0[0], &VD0[5], Cmp); + EXPECT_EQ(VD0[0].Value, getCalleeAddress(vtable2)); + EXPECT_EQ(VD0[0].Count, 1000U); + EXPECT_EQ(VD0[1].Value, getCalleeAddress(vtable3)); + EXPECT_EQ(VD0[1].Count, 500U); + EXPECT_EQ(VD0[2].Value, getCalleeAddress(vtable1)); + EXPECT_EQ(VD0[2].Count, 400U); + EXPECT_EQ(VD0[3].Value, getCalleeAddress(vtable4)); + EXPECT_EQ(VD0[3].Count, 300U); + EXPECT_EQ(VD0[4].Value, getCalleeAddress(vtable5)); + EXPECT_EQ(VD0[4].Count, 100U); + + std::unique_ptr<InstrProfValueData[]> VD1( + Record.getValueForSite(IPVK_VTableTarget, 1)); + llvm::sort(&VD1[0], &VD1[4], Cmp); + EXPECT_EQ(VD1[0].Value, getCalleeAddress(vtable2)); + EXPECT_EQ(VD1[0].Count, 2500U); + EXPECT_EQ(VD1[1].Value, getCalleeAddress(vtable1)); + EXPECT_EQ(VD1[1].Count, 1300U); + EXPECT_EQ(VD1[2].Value, getCalleeAddress(vtable3)); + EXPECT_EQ(VD1[2].Count, 1000U); + EXPECT_EQ(VD1[3].Value, getCalleeAddress(vtable5)); + EXPECT_EQ(VD1[3].Count, 800U); + + std::unique_ptr<InstrProfValueData[]> VD2( + Record.getValueForSite(IPVK_VTableTarget, 2)); + llvm::sort(&VD2[0], &VD2[3], Cmp); + EXPECT_EQ(VD2[0].Value, getCalleeAddress(vtable4)); + EXPECT_EQ(VD2[0].Count, 5500U); + EXPECT_EQ(VD2[1].Value, getCalleeAddress(vtable3)); + EXPECT_EQ(VD2[1].Count, 1000U); + EXPECT_EQ(VD2[2].Value, getCalleeAddress(vtable6)); + EXPECT_EQ(VD2[2].Count, 800U); + + std::unique_ptr<InstrProfValueData[]> VD3( + Record.getValueForSite(IPVK_VTableTarget, 3)); + llvm::sort(&VD3[0], &VD3[2], Cmp); + EXPECT_EQ(VD3[0].Value, getCalleeAddress(vtable3)); + EXPECT_EQ(VD3[0].Count, 2000U); + EXPECT_EQ(VD3[1].Value, getCalleeAddress(vtable2)); + EXPECT_EQ(VD3[1].Count, 1800U); } TEST(ValueProfileReadWriteTest, symtab_mapping) { @@ -1132,10 +1388,27 @@ TEST(ValueProfileReadWriteTest, symtab_mapping) { Symtab.mapAddress(uint64_t(callee4), 0x4000ULL); // Missing mapping for callee5 + auto getVTableStartAddr = [](const uint64_t *vtable) -> uint64_t { + return uint64_t(vtable); + }; + auto getVTableEndAddr = [](const uint64_t *vtable) -> uint64_t { + return uint64_t(vtable) + 16; + }; + // vtable1, vtable2, vtable3, vtable4 get mapped; vtable5, vtable6 are not + // mapped. + Symtab.mapVTableAddress(getVTableStartAddr(vtable1), + getVTableEndAddr(vtable1), MD5Hash("vtable1")); + Symtab.mapVTableAddress(getVTableStartAddr(vtable2), + getVTableEndAddr(vtable2), MD5Hash("vtable2")); + Symtab.mapVTableAddress(getVTableStartAddr(vtable3), + getVTableEndAddr(vtable3), MD5Hash("vtable3")); + Symtab.mapVTableAddress(getVTableStartAddr(vtable4), + getVTableEndAddr(vtable4), MD5Hash("vtable4")); + VPData->deserializeTo(Record, &Symtab); // Now read data from Record and sanity check the data - ASSERT_EQ(5U, Record.getNumValueSites(IPVK_IndirectCallTarget)); + ASSERT_EQ(6U, Record.getNumValueSites(IPVK_IndirectCallTarget)); ASSERT_EQ(5U, Record.getNumValueDataForSite(IPVK_IndirectCallTarget, 0)); auto Cmp = [](const InstrProfValueData &VD1, const InstrProfValueData &VD2) { @@ -1153,6 +1426,74 @@ TEST(ValueProfileReadWriteTest, symtab_mapping) { // callee5 does not have a mapped value -- default to 0. ASSERT_EQ(VD_0[4].Value, 0ULL); + + // Sanity check the vtable value data + ASSERT_EQ(4U, Record.getNumValueSites(IPVK_VTableTarget)); + + { + // The first vtable site. + std::unique_ptr<InstrProfValueData[]> VD( + Record.getValueForSite(IPVK_VTableTarget, 0)); + ASSERT_EQ(5U, Record.getNumValueDataForSite(IPVK_VTableTarget, 0)); + llvm::sort(&VD[0], &VD[5], Cmp); + EXPECT_EQ(1000U, VD[0].Count); + EXPECT_EQ(VD[0].Value, MD5Hash("vtable2")); + EXPECT_EQ(500U, VD[1].Count); + EXPECT_EQ(VD[1].Value, MD5Hash("vtable3")); + EXPECT_EQ(VD[2].Value, MD5Hash("vtable1")); + EXPECT_EQ(400U, VD[2].Count); + EXPECT_EQ(VD[3].Value, MD5Hash("vtable4")); + EXPECT_EQ(300U, VD[3].Count); + + // vtable5 isn't mapped -- default to 0. + EXPECT_EQ(VD[4].Value, 0U); + EXPECT_EQ(VD[4].Count, 100U); + } + + { + // The second vtable site. + std::unique_ptr<InstrProfValueData[]> VD( + Record.getValueForSite(IPVK_VTableTarget, 1)); + ASSERT_EQ(4, Record.getNumValueDataForSite(IPVK_VTableTarget, 1)); + llvm::sort(&VD[0], &VD[4], Cmp); + EXPECT_EQ(VD[0].Value, MD5Hash("vtable2")); + EXPECT_EQ(2500U, VD[0].Count); + EXPECT_EQ(VD[1].Value, MD5Hash("vtable1")); + EXPECT_EQ(1300U, VD[1].Count); + + EXPECT_EQ(VD[2].Value, MD5Hash("vtable3")); + EXPECT_EQ(1000U, VD[2].Count); + // vtable5 isn't mapped -- default to 0. + EXPECT_EQ(VD[3].Value, 0U); + EXPECT_EQ(800U, VD[3].Count); + } + + { + // The third vtable site. + std::unique_ptr<InstrProfValueData[]> VD( + Record.getValueForSite(IPVK_VTableTarget, 2)); + ASSERT_EQ(3, Record.getNumValueDataForSite(IPVK_VTableTarget, 2)); + llvm::sort(&VD[0], &VD[3], Cmp); + EXPECT_EQ(5500U, VD[0].Count); + EXPECT_EQ(VD[0].Value, MD5Hash("vtable4")); + EXPECT_EQ(1000U, VD[1].Count); + EXPECT_EQ(VD[1].Value, MD5Hash("vtable3")); + // vtable6 isn't mapped -- default to 0. + EXPECT_EQ(VD[2].Value, 0U); + EXPECT_EQ(800U, VD[2].Count); + } + + { + // The fourth vtable site. + std::unique_ptr<InstrProfValueData[]> VD( + Record.getValueForSite(IPVK_VTableTarget, 3)); + ASSERT_EQ(2, Record.getNumValueDataForSite(IPVK_VTableTarget, 3)); + llvm::sort(&VD[0], &VD[2], Cmp); + EXPECT_EQ(2000U, VD[0].Count); + EXPECT_EQ(VD[0].Value, MD5Hash("vtable3")); + EXPECT_EQ(1800U, VD[1].Count); + EXPECT_EQ(VD[1].Value, MD5Hash("vtable2")); + } } TEST_P(MaybeSparseInstrProfTest, get_max_function_count) { @@ -1266,6 +1607,41 @@ TEST(SymtabTest, instr_prof_symtab_module_test) { Function::Create(FTy, Function::WeakODRLinkage, "Wblah", M.get()); Function::Create(FTy, Function::WeakODRLinkage, "Wbar", M.get()); + ArrayType *VTableArrayType = ArrayType::get( + PointerType::get(Ctx, M->getDataLayout().getDefaultGlobalsAddressSpace()), + 3); + Constant *Int32TyNull = + llvm::ConstantExpr::getNullValue(PointerType::getUnqual(Ctx)); + SmallVector<llvm::Type *, 1> tys; + tys.push_back(VTableArrayType); + StructType *VTableType = llvm::StructType::get(Ctx, tys); + + GlobalVariable *ExternalGV = new llvm::GlobalVariable( + *M, VTableType, /* isConstant= */ true, + llvm::GlobalValue::ExternalLinkage, + llvm::ConstantStruct::get( + VTableType, {llvm::ConstantArray::get( + VTableArrayType, + {Int32TyNull, Int32TyNull, + Function::Create(FTy, Function::ExternalLinkage, + "VFuncInExternalGV", M.get())})}), + "ExternalGV"); + + GlobalVariable *PrivateGV = new llvm::GlobalVariable( + *M, VTableType, /* isConstant= */ true, + llvm::GlobalValue::InternalLinkage, + llvm::ConstantStruct::get( + VTableType, {llvm::ConstantArray::get( + VTableArrayType, + {Int32TyNull, Int32TyNull, + Function::Create(FTy, Function::ExternalLinkage, + "VFuncInPrivateGV", M.get())})}), + "PrivateGV"); + + // Only vtables with type metadata are added to symtab. + ExternalGV->addTypeMetadata(16, MDString::get(Ctx, "ExternalGV")); + PrivateGV->addTypeMetadata(16, MDString::get(Ctx, "PrivateGV")); + InstrProfSymtab ProfSymtab; EXPECT_THAT_ERROR(ProfSymtab.create(*M), Succeeded()); @@ -1288,6 +1664,17 @@ TEST(SymtabTest, instr_prof_symtab_module_test) { EXPECT_EQ(StringRef(PGOName), PGOFuncName); EXPECT_THAT(PGOFuncName.str(), EndsWith(Funcs[I].str())); } + + StringRef VTables[] = {"ExternalGV", "PrivateGV"}; + for (size_t I = 0; I < std::size(VTables); I++) { + GlobalVariable *GV = + M->getGlobalVariable(VTables[I], /* AllowInternal=*/true); + + std::string IRPGOName = getIRPGOObjectName(*GV); + EXPECT_EQ(IRPGOName, ProfSymtab.getFuncOrVarName( + IndexedInstrProf::ComputeHash(IRPGOName))); + EXPECT_EQ(VTables[I], getParsedIRPGOFuncName(IRPGOName).second); + } } // Testing symtab serialization and creator/deserialization interface |