utils/update_analyze_test_checks.py

   1 #!/usr/bin/env python2.7
   2
   3 """A script to generate FileCheck statements for 'opt' analysis tests.
   4
   5 This script is a utility to update LLVM opt analysis test cases with new
   6 FileCheck patterns. It can either update all of the tests in the file or
   7 a single test function.
   8
   9 Example usage:
  10 $ update_analyze_test_checks.py --opt=../bin/opt test/foo.ll
  11
  12 Workflow:
  13 1. Make a compiler patch that requires updating some number of FileCheck lines
  14    in regression test files.
  15 2. Save the patch and revert it from your local work area.
  16 3. Update the RUN-lines in the affected regression tests to look canonical.
  17    Example: "; RUN: opt < %s -analyze -cost-model -S | FileCheck %s"
  18 4. Refresh the FileCheck lines for either the entire file or select functions by
  19    running this script.
  20 5. Commit the fresh baseline of checks.
  21 6. Apply your patch from step 1 and rebuild your local binaries.
  22 7. Re-run this script on affected regression tests.
  23 8. Check the diffs to ensure the script has done something reasonable.
  24 9. Submit a patch including the regression test diffs for review.
  25
  26 A common pattern is to have the script insert complete checking of every
  27 instruction. Then, edit it down to only check the relevant instructions.
  28 The script is designed to make adding checks to a test case fast, it is *not*
  29 designed to be authoratitive about what constitutes a good test!
  30 """
  31
  32 import argparse
  33 import itertools
  34 import os         # Used to advertise this file's name ("autogenerated_note").
  35 import string
  36 import subprocess
  37 import sys
  38 import tempfile
  39 import re
  40
  41 from UpdateTestChecks import common
  42
  43 ADVERT = '; NOTE: Assertions have been autogenerated by '
  44
  45 # RegEx: this is where the magic happens.
  46
  47 IR_FUNCTION_RE = re.compile('^\s*define\s+(?:internal\s+)?[^@]*@([\w-]+)\s*\(')
  48
  49
  50
  51
  52
  53 def main():
  54   from argparse import RawTextHelpFormatter
  55   parser = argparse.ArgumentParser(description=__doc__, formatter_class=RawTextHelpFormatter)
  56   parser.add_argument('-v', '--verbose', action='store_true',
  57                       help='Show verbose output')
  58   parser.add_argument('--opt-binary', default='opt',
  59                       help='The opt binary used to generate the test case')
  60   parser.add_argument(
  61       '--function', help='The function in the test file to update')
  62   parser.add_argument('tests', nargs='+')
  63   args = parser.parse_args()
  64
  65   autogenerated_note = (ADVERT + 'utils/' + os.path.basename(__file__))
  66
  67   opt_basename = os.path.basename(args.opt_binary)
  68   if (opt_basename != "opt"):
  69     print >>sys.stderr, 'ERROR: Unexpected opt name: ' + opt_basename
  70     sys.exit(1)
  71
  72   for test in args.tests:
  73     if args.verbose:
  74       print >>sys.stderr, 'Scanning for RUN lines in test file: %s' % (test,)
  75     with open(test) as f:
  76       input_lines = [l.rstrip() for l in f]
  77
  78     raw_lines = [m.group(1)
  79                  for m in [common.RUN_LINE_RE.match(l) for l in input_lines] if m]
  80     run_lines = [raw_lines[0]] if len(raw_lines) > 0 else []
  81     for l in raw_lines[1:]:
  82       if run_lines[-1].endswith("\\"):
  83         run_lines[-1] = run_lines[-1].rstrip("\\") + " " + l
  84       else:
  85         run_lines.append(l)
  86
  87     if args.verbose:
  88       print >>sys.stderr, 'Found %d RUN lines:' % (len(run_lines),)
  89       for l in run_lines:
  90         print >>sys.stderr, '  RUN: ' + l
  91
  92     prefix_list = []
  93     for l in run_lines:
  94       (tool_cmd, filecheck_cmd) = tuple([cmd.strip() for cmd in l.split('|', 1)])
  95
  96       if not tool_cmd.startswith(opt_basename + ' '):
  97         print >>sys.stderr, 'WARNING: Skipping non-%s RUN line: %s' % (opt_basename, l)
  98         continue
  99
 100       if not filecheck_cmd.startswith('FileCheck '):
 101         print >>sys.stderr, 'WARNING: Skipping non-FileChecked RUN line: ' + l
 102         continue
 103
 104       tool_cmd_args = tool_cmd[len(opt_basename):].strip()
 105       tool_cmd_args = tool_cmd_args.replace('< %s', '').replace('%s', '').strip()
 106
 107       check_prefixes = [item for m in common.CHECK_PREFIX_RE.finditer(filecheck_cmd)
 108                                for item in m.group(1).split(',')]
 109       if not check_prefixes:
 110         check_prefixes = ['CHECK']
 111
 112       # FIXME: We should use multiple check prefixes to common check lines. For
 113       # now, we just ignore all but the last.
 114       prefix_list.append((check_prefixes, tool_cmd_args))
 115
 116     func_dict = {}
 117     for prefixes, _ in prefix_list:
 118       for prefix in prefixes:
 119         func_dict.update({prefix: dict()})
 120     for prefixes, opt_args in prefix_list:
 121       if args.verbose:
 122         print >>sys.stderr, 'Extracted opt cmd: ' + opt_basename + ' ' + opt_args
 123         print >>sys.stderr, 'Extracted FileCheck prefixes: ' + str(prefixes)
 124
 125       raw_tool_outputs = common.invoke_tool(args.opt_binary, opt_args, test)
 126
 127       # Split analysis outputs by "Printing analysis " declarations.
 128       for raw_tool_output in re.split(r'Printing analysis ', raw_tool_outputs):
 129         common.build_function_body_dictionary(
 130           common.ANALYZE_FUNCTION_RE, common.scrub_body, [],
 131           raw_tool_output, prefixes, func_dict, args.verbose)
 132
 133     is_in_function = False
 134     is_in_function_start = False
 135     prefix_set = set([prefix for prefixes, _ in prefix_list for prefix in prefixes])
 136     if args.verbose:
 137       print >>sys.stderr, 'Rewriting FileCheck prefixes: %s' % (prefix_set,)
 138     output_lines = []
 139     output_lines.append(autogenerated_note)
 140
 141     for input_line in input_lines:
 142       if is_in_function_start:
 143         if input_line == '':
 144           continue
 145         if input_line.lstrip().startswith(';'):
 146           m = common.CHECK_RE.match(input_line)
 147           if not m or m.group(1) not in prefix_set:
 148             output_lines.append(input_line)
 149             continue
 150
 151         # Print out the various check lines here.
 152         common.add_analyze_checks(output_lines, ';', prefix_list, func_dict, func_name)
 153         is_in_function_start = False
 154
 155       if is_in_function:
 156         if common.should_add_line_to_output(input_line, prefix_set):
 157           # This input line of the function body will go as-is into the output.
 158           # Except make leading whitespace uniform: 2 spaces.
 159           input_line = common.SCRUB_LEADING_WHITESPACE_RE.sub(r'  ', input_line)
 160           output_lines.append(input_line)
 161         else:
 162           continue
 163         if input_line.strip() == '}':
 164           is_in_function = False
 165         continue
 166
 167       # Discard any previous script advertising.
 168       if input_line.startswith(ADVERT):
 169         continue
 170
 171       # If it's outside a function, it just gets copied to the output.
 172       output_lines.append(input_line)
 173
 174       m = IR_FUNCTION_RE.match(input_line)
 175       if not m:
 176         continue
 177       func_name = m.group(1)
 178       if args.function is not None and func_name != args.function:
 179         # When filtering on a specific function, skip all others.
 180         continue
 181       is_in_function = is_in_function_start = True
 182
 183     if args.verbose:
 184       print>>sys.stderr, 'Writing %d lines to %s...' % (len(output_lines), test)
 185
 186     with open(test, 'wb') as f:
 187       f.writelines([l + '\n' for l in output_lines])
 188
 189
 190 if __name__ == '__main__':
 191   main()