utils/update_cc_test_checks.py

   1 #!/usr/bin/env python3
   2 '''A utility to update LLVM IR CHECK lines in C/C++ FileCheck test files.
   3
   4 Example RUN lines in .c/.cc test files:
   5
   6 // RUN: %clang -emit-llvm -S %s -o - -O2 | FileCheck %s
   7 // RUN: %clangxx -emit-llvm -S %s -o - -O2 | FileCheck -check-prefix=CHECK-A %s
   8
   9 Usage:
  10
  11 % utils/update_cc_test_checks.py --llvm-bin=release/bin test/a.cc
  12 % utils/update_cc_test_checks.py --c-index-test=release/bin/c-index-test \
  13   --clang=release/bin/clang /tmp/c/a.cc
  14 '''
  15
  16 import argparse
  17 import collections
  18 import distutils.spawn
  19 import os
  20 import shlex
  21 import string
  22 import subprocess
  23 import sys
  24 import re
  25 import tempfile
  26
  27 from UpdateTestChecks import asm, common
  28
  29 ADVERT = '// NOTE: Assertions have been autogenerated by '
  30
  31 CHECK_RE = re.compile(r'^\s*//\s*([^:]+?)(?:-NEXT|-NOT|-DAG|-LABEL)?:')
  32 RUN_LINE_RE = re.compile('^//\s*RUN:\s*(.*)$')
  33
  34 SUBST = {
  35     '%clang': [],
  36     '%clang_cc1': ['-cc1'],
  37     '%clangxx': ['--driver-mode=g++'],
  38 }
  39
  40 def get_line2spell_and_mangled(args, clang_args):
  41   ret = {}
  42   with tempfile.NamedTemporaryFile() as f:
  43     # TODO Make c-index-test print mangled names without circumventing through precompiled headers
  44     status = subprocess.run([args.c_index_test, '-write-pch', f.name, *clang_args],
  45                             stdout=subprocess.PIPE, stderr=subprocess.STDOUT)
  46     if status.returncode:
  47       sys.stderr.write(status.stdout.decode())
  48       sys.exit(2)
  49     output = subprocess.check_output([args.c_index_test,
  50         '-test-print-mangle', f.name])
  51     if sys.version_info[0] > 2:
  52       output = output.decode()
  53   DeclRE = re.compile(r'^FunctionDecl=(\w+):(\d+):\d+ \(Definition\)')
  54   MangleRE = re.compile(r'.*\[mangled=([^]]+)\]')
  55   MatchedDecl = False
  56   for line in output.splitlines():
  57     # Get the function source name, line number and mangled name.  Sometimes
  58     # c-index-test outputs the mangled name on a separate line (this can happen
  59     # with block comments in front of functions).  Keep scanning until we see
  60     # the mangled name.
  61     decl_m = DeclRE.match(line)
  62     mangle_m = MangleRE.match(line)
  63
  64     if decl_m:
  65       MatchedDecl = True
  66       spell, lineno = decl_m.groups()
  67     if MatchedDecl and mangle_m:
  68       mangled = mangle_m.group(1)
  69       MatchedDecl = False
  70     else:
  71       continue
  72
  73     if mangled == '_' + spell:
  74       # HACK for MacOS (where the mangled name includes an _ for C but the IR won't):
  75       mangled = spell
  76     # Note -test-print-mangle does not print file names so if #include is used,
  77     # the line number may come from an included file.
  78     ret[int(lineno)-1] = (spell, mangled)
  79   if args.verbose:
  80     for line, func_name in sorted(ret.items()):
  81       print('line {}: found function {}'.format(line+1, func_name), file=sys.stderr)
  82   return ret
  83
  84
  85 def config():
  86   parser = argparse.ArgumentParser(
  87       description=__doc__,
  88       formatter_class=argparse.RawTextHelpFormatter)
  89   parser.add_argument('-v', '--verbose', action='store_true')
  90   parser.add_argument('--llvm-bin', help='llvm $prefix/bin path')
  91   parser.add_argument('--clang',
  92                       help='"clang" executable, defaults to $llvm_bin/clang')
  93   parser.add_argument('--clang-args',
  94                       help='Space-separated extra args to clang, e.g. --clang-args=-v')
  95   parser.add_argument('--c-index-test',
  96                       help='"c-index-test" executable, defaults to $llvm_bin/c-index-test')
  97   parser.add_argument('--opt',
  98                       help='"opt" executable, defaults to $llvm_bin/opt')
  99   parser.add_argument(
 100       '--functions', nargs='+', help='A list of function name regexes. '
 101       'If specified, update CHECK lines for functions matching at least one regex')
 102   parser.add_argument(
 103       '--x86_extra_scrub', action='store_true',
 104       help='Use more regex for x86 matching to reduce diffs between various subtargets')
 105   parser.add_argument('-u', '--update-only', action='store_true',
 106                       help='Only update test if it was already autogened')
 107   parser.add_argument('tests', nargs='+')
 108   args = parser.parse_args()
 109   args.clang_args = shlex.split(args.clang_args or '')
 110
 111   if args.clang is None:
 112     if args.llvm_bin is None:
 113       args.clang = 'clang'
 114     else:
 115       args.clang = os.path.join(args.llvm_bin, 'clang')
 116   if not distutils.spawn.find_executable(args.clang):
 117     print('Please specify --llvm-bin or --clang', file=sys.stderr)
 118     sys.exit(1)
 119
 120   if args.opt is None:
 121     if args.llvm_bin is None:
 122       args.opt = 'opt'
 123     else:
 124       args.opt = os.path.join(args.llvm_bin, 'opt')
 125   if not distutils.spawn.find_executable(args.opt):
 126     # Many uses of this tool will not need an opt binary, because it's only
 127     # needed for updating a test that runs clang | opt | FileCheck. So we
 128     # defer this error message until we find that opt is actually needed.
 129     args.opt = None
 130
 131   if args.c_index_test is None:
 132     if args.llvm_bin is None:
 133       args.c_index_test = 'c-index-test'
 134     else:
 135       args.c_index_test = os.path.join(args.llvm_bin, 'c-index-test')
 136   if not distutils.spawn.find_executable(args.c_index_test):
 137     print('Please specify --llvm-bin or --c-index-test', file=sys.stderr)
 138     sys.exit(1)
 139
 140   return args
 141
 142
 143 def get_function_body(args, filename, clang_args, extra_commands, prefixes, triple_in_cmd, func_dict):
 144   # TODO Clean up duplication of asm/common build_function_body_dictionary
 145   # Invoke external tool and extract function bodies.
 146   raw_tool_output = common.invoke_tool(args.clang, clang_args, filename)
 147   for extra_command in extra_commands:
 148     extra_args = shlex.split(extra_command)
 149     with tempfile.NamedTemporaryFile() as f:
 150       f.write(raw_tool_output.encode())
 151       f.flush()
 152       if extra_args[0] == 'opt':
 153         if args.opt is None:
 154           print(filename, 'needs to run opt. '
 155                 'Please specify --llvm-bin or --opt', file=sys.stderr)
 156           sys.exit(1)
 157         extra_args[0] = args.opt
 158       raw_tool_output = common.invoke_tool(extra_args[0],
 159                                            extra_args[1:], f.name)
 160   if '-emit-llvm' in clang_args:
 161     common.build_function_body_dictionary(
 162             common.OPT_FUNCTION_RE, common.scrub_body, [],
 163             raw_tool_output, prefixes, func_dict, args.verbose)
 164   else:
 165     print('The clang command line should include -emit-llvm as asm tests '
 166           'are discouraged in Clang testsuite.', file=sys.stderr)
 167     sys.exit(1)
 168
 169
 170 def main():
 171   args = config()
 172   script_name = os.path.basename(__file__)
 173   autogenerated_note = (ADVERT + 'utils/' + script_name)
 174
 175   for filename in args.tests:
 176     with open(filename) as f:
 177       input_lines = [l.rstrip() for l in f]
 178
 179     first_line = input_lines[0] if input_lines else ""
 180     if 'autogenerated' in first_line and script_name not in first_line:
 181       common.warn("Skipping test which wasn't autogenerated by " + script_name, filename)
 182       continue
 183
 184     if args.update_only:
 185       if not first_line or 'autogenerated' not in first_line:
 186         common.warn("Skipping test which isn't autogenerated: " + filename)
 187         continue
 188
 189     # Extract RUN lines.
 190     raw_lines = [m.group(1)
 191                  for m in [RUN_LINE_RE.match(l) for l in input_lines] if m]
 192     run_lines = [raw_lines[0]] if len(raw_lines) > 0 else []
 193     for l in raw_lines[1:]:
 194       if run_lines[-1].endswith("\\"):
 195         run_lines[-1] = run_lines[-1].rstrip("\\") + " " + l
 196       else:
 197         run_lines.append(l)
 198
 199     if args.verbose:
 200       print('Found {} RUN lines:'.format(len(run_lines)), file=sys.stderr)
 201       for l in run_lines:
 202         print('  RUN: ' + l, file=sys.stderr)
 203
 204     # Build a list of clang command lines and check prefixes from RUN lines.
 205     run_list = []
 206     line2spell_and_mangled_list = collections.defaultdict(list)
 207     for l in run_lines:
 208       commands = [cmd.strip() for cmd in l.split('|')]
 209
 210       triple_in_cmd = None
 211       m = common.TRIPLE_ARG_RE.search(commands[0])
 212       if m:
 213         triple_in_cmd = m.groups()[0]
 214
 215       # Apply %clang substitution rule, replace %s by `filename`, and append args.clang_args
 216       clang_args = shlex.split(commands[0])
 217       if clang_args[0] not in SUBST:
 218         print('WARNING: Skipping non-clang RUN line: ' + l, file=sys.stderr)
 219         continue
 220       clang_args[0:1] = SUBST[clang_args[0]]
 221       clang_args = [filename if i == '%s' else i for i in clang_args] + args.clang_args
 222
 223       # Permit piping the output through opt
 224       if not (len(commands) == 2 or
 225               (len(commands) == 3 and commands[1].startswith('opt'))):
 226         print('WARNING: Skipping non-clang RUN line: ' + l, file=sys.stderr)
 227
 228       # Extract -check-prefix in FileCheck args
 229       filecheck_cmd = commands[-1]
 230       common.verify_filecheck_prefixes(filecheck_cmd)
 231       if not filecheck_cmd.startswith('FileCheck '):
 232         print('WARNING: Skipping non-FileChecked RUN line: ' + l, file=sys.stderr)
 233         continue
 234       check_prefixes = [item for m in common.CHECK_PREFIX_RE.finditer(filecheck_cmd)
 235                                for item in m.group(1).split(',')]
 236       if not check_prefixes:
 237         check_prefixes = ['CHECK']
 238       run_list.append((check_prefixes, clang_args, commands[1:-1], triple_in_cmd))
 239
 240     # Strip CHECK lines which are in `prefix_set`, update test file.
 241     prefix_set = set([prefix for p in run_list for prefix in p[0]])
 242     input_lines = []
 243     with open(filename, 'r+') as f:
 244       for line in f:
 245         m = CHECK_RE.match(line)
 246         if not (m and m.group(1) in prefix_set) and line != '//\n':
 247           input_lines.append(line)
 248       f.seek(0)
 249       f.writelines(input_lines)
 250       f.truncate()
 251
 252     # Execute clang, generate LLVM IR, and extract functions.
 253     func_dict = {}
 254     for p in run_list:
 255       prefixes = p[0]
 256       for prefix in prefixes:
 257         func_dict.update({prefix: dict()})
 258     for prefixes, clang_args, extra_commands, triple_in_cmd in run_list:
 259       if args.verbose:
 260         print('Extracted clang cmd: clang {}'.format(clang_args), file=sys.stderr)
 261         print('Extracted FileCheck prefixes: {}'.format(prefixes), file=sys.stderr)
 262
 263       get_function_body(args, filename, clang_args, extra_commands, prefixes, triple_in_cmd, func_dict)
 264
 265       # Invoke c-index-test to get mapping from start lines to mangled names.
 266       # Forward all clang args for now.
 267       for k, v in get_line2spell_and_mangled(args, clang_args).items():
 268         line2spell_and_mangled_list[k].append(v)
 269
 270     output_lines = [autogenerated_note]
 271     for idx, line in enumerate(input_lines):
 272       # Discard any previous script advertising.
 273       if line.startswith(ADVERT):
 274         continue
 275       if idx in line2spell_and_mangled_list:
 276         added = set()
 277         for spell, mangled in line2spell_and_mangled_list[idx]:
 278           # One line may contain multiple function declarations.
 279           # Skip if the mangled name has been added before.
 280           # The line number may come from an included file,
 281           # we simply require the spelling name to appear on the line
 282           # to exclude functions from other files.
 283           if mangled in added or spell not in line:
 284             continue
 285           if args.functions is None or any(re.search(regex, spell) for regex in args.functions):
 286             if added:
 287               output_lines.append('//')
 288             added.add(mangled)
 289             common.add_ir_checks(output_lines, '//', run_list, func_dict, mangled, False)
 290       output_lines.append(line.rstrip('\n'))
 291
 292     # Update the test file.
 293     with open(filename, 'w') as f:
 294       for line in output_lines:
 295         f.write(line + '\n')
 296
 297   return 0
 298
 299
 300 if __name__ == '__main__':
 301   sys.exit(main())