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
  54   RE = re.compile(r'^FunctionDecl=(\w+):(\d+):\d+ \(Definition\) \[mangled=([^]]+)\]')
  55   for line in output.splitlines():
  56     m = RE.match(line)
  57     if not m: continue
  58     spell, line, mangled = m.groups()
  59     if mangled == '_' + spell:
  60       # HACK for MacOS (where the mangled name includes an _ for C but the IR won't):
  61       mangled = spell
  62     # Note -test-print-mangle does not print file names so if #include is used,
  63     # the line number may come from an included file.
  64     ret[int(line)-1] = (spell, mangled)
  65   if args.verbose:
  66     for line, func_name in sorted(ret.items()):
  67       print('line {}: found function {}'.format(line+1, func_name), file=sys.stderr)
  68   return ret
  69
  70
  71 def config():
  72   parser = argparse.ArgumentParser(
  73       description=__doc__,
  74       formatter_class=argparse.RawTextHelpFormatter)
  75   parser.add_argument('-v', '--verbose', action='store_true')
  76   parser.add_argument('--llvm-bin', help='llvm $prefix/bin path')
  77   parser.add_argument('--clang',
  78                       help='"clang" executable, defaults to $llvm_bin/clang')
  79   parser.add_argument('--clang-args',
  80                       help='Space-separated extra args to clang, e.g. --clang-args=-v')
  81   parser.add_argument('--c-index-test',
  82                       help='"c-index-test" executable, defaults to $llvm_bin/c-index-test')
  83   parser.add_argument(
  84       '--functions', nargs='+', help='A list of function name regexes. '
  85       'If specified, update CHECK lines for functions matching at least one regex')
  86   parser.add_argument(
  87       '--x86_extra_scrub', action='store_true',
  88       help='Use more regex for x86 matching to reduce diffs between various subtargets')
  89   parser.add_argument('tests', nargs='+')
  90   args = parser.parse_args()
  91   args.clang_args = shlex.split(args.clang_args or '')
  92
  93   if args.clang is None:
  94     if args.llvm_bin is None:
  95       args.clang = 'clang'
  96     else:
  97       args.clang = os.path.join(args.llvm_bin, 'clang')
  98   if not distutils.spawn.find_executable(args.clang):
  99     print('Please specify --llvm-bin or --clang', file=sys.stderr)
 100     sys.exit(1)
 101   if args.c_index_test is None:
 102     if args.llvm_bin is None:
 103       args.c_index_test = 'c-index-test'
 104     else:
 105       args.c_index_test = os.path.join(args.llvm_bin, 'c-index-test')
 106   if not distutils.spawn.find_executable(args.c_index_test):
 107     print('Please specify --llvm-bin or --c-index-test', file=sys.stderr)
 108     sys.exit(1)
 109
 110   return args
 111
 112
 113 def get_function_body(args, filename, clang_args, prefixes, triple_in_cmd, func_dict):
 114   # TODO Clean up duplication of asm/common build_function_body_dictionary
 115   # Invoke external tool and extract function bodies.
 116   raw_tool_output = common.invoke_tool(args.clang, clang_args, filename)
 117   if '-emit-llvm' in clang_args:
 118     common.build_function_body_dictionary(
 119             common.OPT_FUNCTION_RE, common.scrub_body, [],
 120             raw_tool_output, prefixes, func_dict, args.verbose)
 121   else:
 122     print('The clang command line should include -emit-llvm as asm tests '
 123           'are discouraged in Clang testsuite.', file=sys.stderr)
 124     sys.exit(1)
 125
 126
 127 def main():
 128   args = config()
 129   autogenerated_note = (ADVERT + 'utils/' + os.path.basename(__file__))
 130
 131   for filename in args.tests:
 132     with open(filename) as f:
 133       input_lines = [l.rstrip() for l in f]
 134
 135     # Extract RUN lines.
 136     raw_lines = [m.group(1)
 137                  for m in [RUN_LINE_RE.match(l) for l in input_lines] if m]
 138     run_lines = [raw_lines[0]] if len(raw_lines) > 0 else []
 139     for l in raw_lines[1:]:
 140       if run_lines[-1].endswith("\\"):
 141         run_lines[-1] = run_lines[-1].rstrip("\\") + " " + l
 142       else:
 143         run_lines.append(l)
 144
 145     if args.verbose:
 146       print('Found {} RUN lines:'.format(len(run_lines)), file=sys.stderr)
 147       for l in run_lines:
 148         print('  RUN: ' + l, file=sys.stderr)
 149
 150     # Build a list of clang command lines and check prefixes from RUN lines.
 151     run_list = []
 152     line2spell_and_mangled_list = collections.defaultdict(list)
 153     for l in run_lines:
 154       commands = [cmd.strip() for cmd in l.split('|', 1)]
 155
 156       triple_in_cmd = None
 157       m = common.TRIPLE_ARG_RE.search(commands[0])
 158       if m:
 159         triple_in_cmd = m.groups()[0]
 160
 161       # Apply %clang substitution rule, replace %s by `filename`, and append args.clang_args
 162       clang_args = shlex.split(commands[0])
 163       if clang_args[0] not in SUBST:
 164         print('WARNING: Skipping non-clang RUN line: ' + l, file=sys.stderr)
 165         continue
 166       clang_args[0:1] = SUBST[clang_args[0]]
 167       clang_args = [filename if i == '%s' else i for i in clang_args] + args.clang_args
 168
 169       # Extract -check-prefix in FileCheck args
 170       filecheck_cmd = commands[-1]
 171       common.verify_filecheck_prefixes(filecheck_cmd)
 172       if not filecheck_cmd.startswith('FileCheck '):
 173         print('WARNING: Skipping non-FileChecked RUN line: ' + l, file=sys.stderr)
 174         continue
 175       check_prefixes = [item for m in common.CHECK_PREFIX_RE.finditer(filecheck_cmd)
 176                                for item in m.group(1).split(',')]
 177       if not check_prefixes:
 178         check_prefixes = ['CHECK']
 179       run_list.append((check_prefixes, clang_args, triple_in_cmd))
 180
 181     # Strip CHECK lines which are in `prefix_set`, update test file.
 182     prefix_set = set([prefix for p in run_list for prefix in p[0]])
 183     input_lines = []
 184     with open(filename, 'r+') as f:
 185       for line in f:
 186         m = CHECK_RE.match(line)
 187         if not (m and m.group(1) in prefix_set) and line != '//\n':
 188           input_lines.append(line)
 189       f.seek(0)
 190       f.writelines(input_lines)
 191       f.truncate()
 192
 193     # Execute clang, generate LLVM IR, and extract functions.
 194     func_dict = {}
 195     for p in run_list:
 196       prefixes = p[0]
 197       for prefix in prefixes:
 198         func_dict.update({prefix: dict()})
 199     for prefixes, clang_args, triple_in_cmd in run_list:
 200       if args.verbose:
 201         print('Extracted clang cmd: clang {}'.format(clang_args), file=sys.stderr)
 202         print('Extracted FileCheck prefixes: {}'.format(prefixes), file=sys.stderr)
 203
 204       get_function_body(args, filename, clang_args, prefixes, triple_in_cmd, func_dict)
 205
 206       # Invoke c-index-test to get mapping from start lines to mangled names.
 207       # Forward all clang args for now.
 208       for k, v in get_line2spell_and_mangled(args, clang_args).items():
 209         line2spell_and_mangled_list[k].append(v)
 210
 211     output_lines = [autogenerated_note]
 212     for idx, line in enumerate(input_lines):
 213       # Discard any previous script advertising.
 214       if line.startswith(ADVERT):
 215         continue
 216       if idx in line2spell_and_mangled_list:
 217         added = set()
 218         for spell, mangled in line2spell_and_mangled_list[idx]:
 219           # One line may contain multiple function declarations.
 220           # Skip if the mangled name has been added before.
 221           # The line number may come from an included file,
 222           # we simply require the spelling name to appear on the line
 223           # to exclude functions from other files.
 224           if mangled in added or spell not in line:
 225             continue
 226           if args.functions is None or any(re.search(regex, spell) for regex in args.functions):
 227             if added:
 228               output_lines.append('//')
 229             added.add(mangled)
 230             common.add_ir_checks(output_lines, '//', run_list, func_dict, mangled)
 231       output_lines.append(line.rstrip('\n'))
 232
 233     # Update the test file.
 234     with open(filename, 'w') as f:
 235       for line in output_lines:
 236         f.write(line + '\n')
 237
 238   return 0
 239
 240
 241 if __name__ == '__main__':
 242   sys.exit(main())