diff options
Diffstat (limited to 'llvm/utils')
7 files changed, 145 insertions, 45 deletions
diff --git a/llvm/utils/gn/secondary/libcxx/include/BUILD.gn b/llvm/utils/gn/secondary/libcxx/include/BUILD.gn index e747006..278c29c 100644 --- a/llvm/utils/gn/secondary/libcxx/include/BUILD.gn +++ b/llvm/utils/gn/secondary/libcxx/include/BUILD.gn @@ -911,7 +911,6 @@ if (current_toolchain == default_toolchain) { "__cxx03/cmath", "__cxx03/codecvt", "__cxx03/complex", - "__cxx03/complex.h", "__cxx03/condition_variable", "__cxx03/csetjmp", "__cxx03/csignal", @@ -924,25 +923,20 @@ if (current_toolchain == default_toolchain) { "__cxx03/cstring", "__cxx03/ctgmath", "__cxx03/ctime", - "__cxx03/ctype.h", "__cxx03/cuchar", "__cxx03/cwchar", "__cxx03/cwctype", "__cxx03/deque", - "__cxx03/errno.h", "__cxx03/exception", "__cxx03/experimental/__config", "__cxx03/experimental/utility", "__cxx03/ext/__hash", "__cxx03/ext/hash_map", "__cxx03/ext/hash_set", - "__cxx03/fenv.h", - "__cxx03/float.h", "__cxx03/forward_list", "__cxx03/fstream", "__cxx03/functional", "__cxx03/future", - "__cxx03/inttypes.h", "__cxx03/iomanip", "__cxx03/ios", "__cxx03/iosfwd", @@ -969,11 +963,8 @@ if (current_toolchain == default_toolchain) { "__cxx03/sstream", "__cxx03/stack", "__cxx03/stdatomic.h", - "__cxx03/stdbool.h", - "__cxx03/stddef.h", "__cxx03/stdexcept", "__cxx03/stdint.h", - "__cxx03/stdio.h", "__cxx03/stdlib.h", "__cxx03/streambuf", "__cxx03/string", @@ -981,7 +972,6 @@ if (current_toolchain == default_toolchain) { "__cxx03/string_view", "__cxx03/strstream", "__cxx03/system_error", - "__cxx03/tgmath.h", "__cxx03/thread", "__cxx03/type_traits", "__cxx03/typeindex", @@ -994,7 +984,6 @@ if (current_toolchain == default_toolchain) { "__cxx03/vector", "__cxx03/version", "__cxx03/wchar.h", - "__cxx03/wctype.h", "__debug_utils/randomize_range.h", "__debug_utils/sanitizers.h", "__debug_utils/strict_weak_ordering_check.h", diff --git a/llvm/utils/gn/secondary/llvm/lib/CodeGen/BUILD.gn b/llvm/utils/gn/secondary/llvm/lib/CodeGen/BUILD.gn index 39dacf7..4446702 100644 --- a/llvm/utils/gn/secondary/llvm/lib/CodeGen/BUILD.gn +++ b/llvm/utils/gn/secondary/llvm/lib/CodeGen/BUILD.gn @@ -122,6 +122,7 @@ static_library("CodeGen") { "MLRegAllocPriorityAdvisor.cpp", "MachineBasicBlock.cpp", "MachineBlockFrequencyInfo.cpp", + "MachineBlockHashInfo.cpp", "MachineBlockPlacement.cpp", "MachineBranchProbabilityInfo.cpp", "MachineCFGPrinter.cpp", diff --git a/llvm/utils/gn/secondary/llvm/lib/ExecutionEngine/Orc/TargetProcess/BUILD.gn b/llvm/utils/gn/secondary/llvm/lib/ExecutionEngine/Orc/TargetProcess/BUILD.gn index c4ce990..937e81b 100644 --- a/llvm/utils/gn/secondary/llvm/lib/ExecutionEngine/Orc/TargetProcess/BUILD.gn +++ b/llvm/utils/gn/secondary/llvm/lib/ExecutionEngine/Orc/TargetProcess/BUILD.gn @@ -12,6 +12,8 @@ static_library("TargetProcess") { "JITLoaderGDB.cpp", "JITLoaderPerf.cpp", "JITLoaderVTune.cpp", + "LibraryResolver.cpp", + "LibraryScanner.cpp", "OrcRTBootstrap.cpp", "RegisterEHFrames.cpp", "SimpleExecutorDylibManager.cpp", diff --git a/llvm/utils/gn/secondary/llvm/unittests/ExecutionEngine/Orc/BUILD.gn b/llvm/utils/gn/secondary/llvm/unittests/ExecutionEngine/Orc/BUILD.gn index dfe6d6d..111e4c9 100644 --- a/llvm/utils/gn/secondary/llvm/unittests/ExecutionEngine/Orc/BUILD.gn +++ b/llvm/utils/gn/secondary/llvm/unittests/ExecutionEngine/Orc/BUILD.gn @@ -24,6 +24,7 @@ unittest("OrcJITTests") { "JITLinkRedirectionManagerTest.cpp", "JITTargetMachineBuilderTest.cpp", "LazyCallThroughAndReexportsTest.cpp", + "LibraryResolverTest.cpp", "LookupAndRecordAddrsTest.cpp", "MachOPlatformTest.cpp", "MapperJITLinkMemoryManagerTest.cpp", diff --git a/llvm/utils/lldbDataFormatters.py b/llvm/utils/lldbDataFormatters.py index 5e553ca..a3e4ae1 100644 --- a/llvm/utils/lldbDataFormatters.py +++ b/llvm/utils/lldbDataFormatters.py @@ -197,6 +197,11 @@ def StringRefSummaryProvider(valobj, internal_dict): return '""' data = data_pointer.deref + # StringRef may be uninitialized with length exceeding available memory, + # potentially causing bad_alloc exceptions. Limit the length to max string summary setting. + limit_obj = valobj.target.debugger.GetSetting("target.max-string-summary-length") + if limit_obj: + length = min(length, limit_obj.GetUnsignedIntegerValue()) # Get a char[N] type, from the underlying char type. array_type = data.type.GetArrayType(length) # Cast the char* string data to a char[N] array. diff --git a/llvm/utils/mlgo-utils/IR2Vec/generateTriplets.py b/llvm/utils/mlgo-utils/IR2Vec/generateTriplets.py index 80ac4c6..dba9e2c 100644 --- a/llvm/utils/mlgo-utils/IR2Vec/generateTriplets.py +++ b/llvm/utils/mlgo-utils/IR2Vec/generateTriplets.py @@ -1,14 +1,19 @@ # Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions. # See https://llvm.org/LICENSE.txt for license information. # SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception -"""IR2Vec Triplet Generator +"""IR2Vec/MIR2Vec Triplet Generator -Generates IR2Vec triplets by applying random optimization levels to LLVM IR files -and extracting triplets using llvm-ir2vec. Automatically generates preprocessed -files: entity2id.txt, relation2id.txt, and train2id.txt. +Generates IR2Vec or MIR2Vec triplets by applying random optimization levels to +LLVM IR files (or processing MIR files) and extracting triplets using llvm-ir2vec. +Automatically generates preprocessed files (entity2id.txt, relation2id.txt, and +train2id.txt) necessary for training IR2Vec or MIR2Vec vocabularies. Usage: - python generateTriplets.py <llvm_build_dir> <num_optimizations> <ll_file_list> <output_dir> + For LLVM IR: + python generateTriplets.py <llvm_build_dir> <num_optimizations> <ll_file_list> <output_dir> + + For Machine IR: + python generateTriplets.py --mode=mir <llvm_build_dir> <mir_file_list> <output_dir> """ import argparse @@ -41,7 +46,7 @@ class TripletResult: class IR2VecTripletGenerator: - """Main class for generating IR2Vec triplets""" + """Main class for generating IR2Vec or MIR2Vec triplets""" def __init__( self, @@ -49,11 +54,13 @@ class IR2VecTripletGenerator: num_optimizations: int, output_dir: Path, max_workers: int = DEFAULT_MAX_WORKERS, + mode: str = "llvm", ): self.llvm_build_dir = llvm_build_dir self.num_optimizations = num_optimizations self.output_dir = output_dir self.max_workers = max_workers + self.mode = mode # "llvm" or "mir" # Tool paths self.opt_binary = os.path.join(llvm_build_dir, "bin", "opt") @@ -85,7 +92,11 @@ class IR2VecTripletGenerator: f"llvm-ir2vec binary not found or not executable: {self.ir2vec_binary}" ) - if not (1 <= self.num_optimizations <= len(OPT_LEVELS)): + if self.mode not in ["llvm", "mir"]: + raise ValueError(f"Mode must be 'llvm' or 'mir', got: {self.mode}") + + # For LLVM IR mode, validate optimization count + if self.mode == "llvm" and not (1 <= self.num_optimizations <= len(OPT_LEVELS)): raise ValueError( f"Number of optimizations must be between 1-{len(OPT_LEVELS)}" ) @@ -95,19 +106,28 @@ class IR2VecTripletGenerator: return random.sample(OPT_LEVELS, self.num_optimizations) def _process_single_file(self, input_file: Path) -> TripletResult: - """Process a single LLVM IR file with multiple optimization levels""" + """Process a single LLVM IR or MIR file""" all_triplets = set() max_relation = 1 - opt_levels = self._select_optimization_levels() - for opt_level in opt_levels: - triplets, file_max_relation = self._run_pipeline(input_file, opt_level) + if self.mode == "mir": + # For MIR files, process directly without optimization + triplets, file_max_relation = self._run_mir_pipeline(input_file) if triplets: all_triplets.update(triplets) max_relation = max(max_relation, file_max_relation) - logger.debug( - f"Generated {len(triplets)} triplets for {input_file} with {opt_level}" - ) + logger.debug(f"Generated {len(triplets)} triplets for {input_file}") + else: + # For LLVM IR files, apply multiple optimization levels + opt_levels = self._select_optimization_levels() + for opt_level in opt_levels: + triplets, file_max_relation = self._run_pipeline(input_file, opt_level) + if triplets: + all_triplets.update(triplets) + max_relation = max(max_relation, file_max_relation) + logger.debug( + f"Generated {len(triplets)} triplets for {input_file} with {opt_level}" + ) return TripletResult(all_triplets, max_relation) @@ -124,7 +144,7 @@ class IR2VecTripletGenerator: # Run llvm-ir2vec with opt's output as input ir2vec_proc = subprocess.Popen( - [self.ir2vec_binary, "triplets", "-", "-o", "-"], + [self.ir2vec_binary, "triplets", "--mode=llvm", "-", "-o", "-"], stdin=opt_proc.stdout, stdout=subprocess.PIPE, stderr=subprocess.PIPE, @@ -143,6 +163,32 @@ class IR2VecTripletGenerator: except (subprocess.SubprocessError, OSError): return set(), 1 + def _run_mir_pipeline(self, input_file: Path) -> Tuple[Set[str], int]: + """Run llvm-ir2vec pipeline for MIR files.""" + try: + # Run llvm-ir2vec directly on MIR file + result = subprocess.run( + [ + self.ir2vec_binary, + "triplets", + "--mode=mir", + str(input_file), + "-o", + "-", + ], + stdout=subprocess.PIPE, + stderr=subprocess.PIPE, + text=True, + check=False, + ) + + if result.returncode != 0: + return set(), 1 + + return self._parse_triplet_output(result.stdout) + except (subprocess.SubprocessError, OSError): + return set(), 1 + def _parse_triplet_output(self, output: str) -> Tuple[Set[str], int]: """Parse triplet output and extract max relation""" if not output.strip(): @@ -160,12 +206,21 @@ class IR2VecTripletGenerator: return set(lines), max_relation def generate_triplets(self, file_list: Path) -> None: - """Main method to generate triplets from a list of LLVM IR files""" + """Main method to generate triplets from a list of LLVM IR or MIR files""" + # Store file_list_path for later use in entity generation + self.file_list_path = file_list + input_files = self._read_file_list(file_list) - logger.info( - f"Processing {len(input_files)} files with {self.num_optimizations} " - f"optimization levels using {self.max_workers} workers" - ) + + if self.mode == "mir": + logger.info( + f"Processing {len(input_files)} MIR files using {self.max_workers} workers" + ) + else: + logger.info( + f"Processing {len(input_files)} files with {self.num_optimizations} " + f"optimization levels using {self.max_workers} workers" + ) all_triplets = set() global_max_relation = 1 @@ -222,28 +277,60 @@ class IR2VecTripletGenerator: def _generate_entity2id(self, output_file: Path) -> None: """Generate entity2id.txt using llvm-ir2vec""" - subprocess.run( - [str(self.ir2vec_binary), "entities", "-o", str(output_file)], - check=True, - capture_output=True, - ) + if self.mode == "mir": + # For MIR mode, we need to provide a sample MIR file to determine target + # Use the first file from the processed list + input_files = self._read_file_list(self.file_list_path) + if not input_files: + raise ValueError("No input files available for entity generation") + + subprocess.run( + [ + str(self.ir2vec_binary), + "entities", + "--mode=mir", + str(input_files[0]), + "-o", + str(output_file), + ], + check=True, + capture_output=True, + ) + else: + subprocess.run( + [ + str(self.ir2vec_binary), + "entities", + "--mode=llvm", + "-o", + str(output_file), + ], + check=True, + capture_output=True, + ) def _generate_relation2id(self, output_file: Path, max_relation: int) -> None: """Generate relation2id.txt from max relation""" - max_relation = max(max_relation, 1) # At least Type and Next relations + max_relation = max(max_relation, 1) # At least Next relation num_relations = max_relation + 1 with open(output_file, "w") as f: f.write(f"{num_relations}\n") - f.write("Type\t0\n") - f.write("Next\t1\n") - f.writelines(f"Arg{i-2}\t{i}\n" for i in range(2, num_relations)) + if self.mode == "llvm": + # LLVM IR has Type relation at 0 + f.write("Type\t0\n") + f.write("Next\t1\n") + f.writelines(f"Arg{i-2}\t{i}\n" for i in range(2, num_relations)) + else: + # MIR doesn't have Type relation, starts with Next at 0 + f.write("Next\t0\n") + f.writelines(f"Arg{i-1}\t{i}\n" for i in range(1, num_relations)) def main(): """Main entry point""" parser = argparse.ArgumentParser( - description="Generate IR2Vec triplets from LLVM IR files", + description="Generate IR2Vec or MIR2Vec triplets from LLVM IR or Machine IR files", formatter_class=argparse.RawDescriptionHelpFormatter, ) @@ -253,17 +340,26 @@ def main(): parser.add_argument( "num_optimizations", type=int, - help="Number of optimization levels to apply (1-6)", + nargs="?", + default=1, + help="Number of optimization levels to apply (1-6) for LLVM IR mode", ) parser.add_argument( - "ll_file_list", + "input_file_list", type=Path, - help="File containing list of LLVM IR files to process", + help="File containing list of LLVM IR or MIR files to process", ) parser.add_argument( "output_dir", type=Path, help="Output directory for generated files" ) parser.add_argument( + "--mode", + type=str, + choices=["llvm", "mir"], + default="llvm", + help="Operation mode: 'llvm' for LLVM IR (default) or 'mir' for Machine IR", + ) + parser.add_argument( "-j", "--max-workers", type=int, @@ -296,8 +392,9 @@ def main(): args.num_optimizations, args.output_dir, args.max_workers, + args.mode, ) - generator.generate_triplets(args.ll_file_list) + generator.generate_triplets(args.input_file_list) if __name__ == "__main__": diff --git a/llvm/utils/profcheck-xfail.txt b/llvm/utils/profcheck-xfail.txt index b570f8d..151b065 100644 --- a/llvm/utils/profcheck-xfail.txt +++ b/llvm/utils/profcheck-xfail.txt @@ -71,6 +71,7 @@ CodeGen/Hexagon/autohvx/interleave.ll CodeGen/Hexagon/loop-idiom/hexagon-memmove1.ll CodeGen/Hexagon/loop-idiom/hexagon-memmove2.ll CodeGen/Hexagon/loop-idiom/memmove-rt-check.ll +CodeGen/Hexagon/masked_gather.ll CodeGen/NVPTX/lower-ctor-dtor.ll CodeGen/RISCV/zmmul.ll CodeGen/WebAssembly/memory-interleave.ll @@ -331,6 +332,10 @@ Instrumentation/MemorySanitizer/AArch64/arm64-vshift.ll Instrumentation/MemorySanitizer/AArch64/module-flags-aarch64.ll Instrumentation/MemorySanitizer/AArch64/neon_vst_float.ll Instrumentation/MemorySanitizer/AArch64/qshrn.ll +Instrumentation/MemorySanitizer/AArch64/sme-aarch64-svcount-mini.ll +Instrumentation/MemorySanitizer/AArch64/sme-aarch64-svcount.ll +Instrumentation/MemorySanitizer/AArch64/sme2-intrinsics-add-mini.ll +Instrumentation/MemorySanitizer/AArch64/sme2-intrinsics-add.ll Instrumentation/MemorySanitizer/AArch64/vararg.ll Instrumentation/MemorySanitizer/AArch64/vararg_shadow.ll Instrumentation/MemorySanitizer/abs-vector.ll |