[utils] remove ability to generate llc check lines from update_test_checks.py

The dream of a unified check-line auto-generator for all phases of compilation is dead.
The llc script has already diverged to be better at its goal, so having 2 scripts that
do almost the same thing just causes confusion. Now, this script will only work with 
opt to produce check lines for IR transforms.

llvm-svn: 305208
This commit is contained in:
Sanjay Patel 2017-06-12 17:44:30 +00:00
parent b5fe855dfb
commit cae64a0497
1 changed files with 27 additions and 69 deletions

View File

@ -1,13 +1,13 @@
#!/usr/bin/env python2.7
"""A script to generate FileCheck statements for regression tests.
"""A script to generate FileCheck statements for 'opt' regression tests.
This script is a utility to update LLVM opt or llc test cases with new
This script is a utility to update LLVM opt test cases with new
FileCheck patterns. It can either update all of the tests in the file or
a single test function.
Example usage:
$ update_test_checks.py --tool=../bin/opt test/foo.ll
$ update_test_checks.py --opt=../bin/opt test/foo.ll
Workflow:
1. Make a compiler patch that requires updating some number of FileCheck lines
@ -45,23 +45,11 @@ ADVERT = '; NOTE: Assertions have been autogenerated by '
SCRUB_LEADING_WHITESPACE_RE = re.compile(r'^(\s+)')
SCRUB_WHITESPACE_RE = re.compile(r'(?!^(| \w))[ \t]+', flags=re.M)
SCRUB_TRAILING_WHITESPACE_RE = re.compile(r'[ \t]+$', flags=re.M)
SCRUB_X86_SHUFFLES_RE = (
re.compile(
r'^(\s*\w+) [^#\n]+#+ ((?:[xyz]mm\d+|mem)( \{%k\d+\}( \{z\})?)? = .*)$',
flags=re.M))
SCRUB_X86_SP_RE = re.compile(r'\d+\(%(esp|rsp)\)')
SCRUB_X86_RIP_RE = re.compile(r'[.\w]+\(%rip\)')
SCRUB_X86_LCP_RE = re.compile(r'\.LCPI[0-9]+_[0-9]+')
SCRUB_KILL_COMMENT_RE = re.compile(r'^ *#+ +kill:.*\n')
SCRUB_IR_COMMENT_RE = re.compile(r'\s*;.*')
RUN_LINE_RE = re.compile('^\s*;\s*RUN:\s*(.*)$')
IR_FUNCTION_RE = re.compile('^\s*define\s+(?:internal\s+)?[^@]*@([\w-]+)\s*\(')
LLC_FUNCTION_RE = re.compile(
r'^_?(?P<func>[^:]+):[ \t]*#+[ \t]*@(?P=func)\n[^:]*?'
r'(?P<body>^##?[ \t]+[^:]+:.*?)\s*'
r'^\s*(?:[^:\n]+?:\s*\n\s*\.size|\.cfi_endproc|\.globl|\.comm|\.(?:sub)?section)',
flags=(re.M | re.S))
OPT_FUNCTION_RE = re.compile(
r'^\s*define\s+(?:internal\s+)?[^@]*@(?P<func>[\w-]+?)\s*\('
r'(\s+)?[^)]*[^{]*\{\n(?P<body>.*?)^\}$',
@ -76,29 +64,14 @@ IR_VALUE_RE = re.compile(r'(\s+)%([\w\.]+?)([,\s\(\)]|\Z)')
# Invoke the tool that is being tested.
def invoke_tool(args, cmd_args, ir):
with open(ir) as ir_file:
stdout = subprocess.check_output(args.tool_binary + ' ' + cmd_args,
stdout = subprocess.check_output(args.opt_binary + ' ' + cmd_args,
shell=True, stdin=ir_file)
# Fix line endings to unix CR style.
stdout = stdout.replace('\r\n', '\n')
return stdout
# FIXME: Separate the x86-specific scrubbers, so this can be used for other targets.
def scrub_asm(asm):
# Detect shuffle asm comments and hide the operands in favor of the comments.
asm = SCRUB_X86_SHUFFLES_RE.sub(r'\1 {{.*#+}} \2', asm)
# Generically match the stack offset of a memory operand.
asm = SCRUB_X86_SP_RE.sub(r'{{[0-9]+}}(%\1)', asm)
# Generically match a RIP-relative memory operand.
asm = SCRUB_X86_RIP_RE.sub(r'{{.*}}(%rip)', asm)
# Generically match a LCP symbol.
asm = SCRUB_X86_LCP_RE.sub(r'{{\.LCPI.*}}', asm)
# Strip kill operands inserted into the asm.
asm = SCRUB_KILL_COMMENT_RE.sub('', asm)
return asm
def scrub_body(body, tool_basename):
def scrub_body(body, opt_basename):
# Scrub runs of whitespace out of the assembly, but leave the leading
# whitespace in place.
body = SCRUB_WHITESPACE_RE.sub(r' ', body)
@ -106,22 +79,17 @@ def scrub_body(body, tool_basename):
body = string.expandtabs(body, 2)
# Strip trailing whitespace.
body = SCRUB_TRAILING_WHITESPACE_RE.sub(r'', body)
if tool_basename == "llc":
body = scrub_asm(body)
return body
# Build up a dictionary of all the function bodies.
def build_function_body_dictionary(raw_tool_output, prefixes, func_dict, verbose, tool_basename):
if tool_basename == "llc":
func_regex = LLC_FUNCTION_RE
else:
func_regex = OPT_FUNCTION_RE
def build_function_body_dictionary(raw_tool_output, prefixes, func_dict, verbose, opt_basename):
func_regex = OPT_FUNCTION_RE
for m in func_regex.finditer(raw_tool_output):
if not m:
continue
func = m.group('func')
scrubbed_body = scrub_body(m.group('body'), tool_basename)
scrubbed_body = scrub_body(m.group('body'), opt_basename)
if func.startswith('stress'):
# We only use the last line of the function body for stress tests.
scrubbed_body = '\n'.join(scrubbed_body.splitlines()[-1:])
@ -188,12 +156,9 @@ def genericize_check_lines(lines):
return lines
def add_checks(output_lines, prefix_list, func_dict, func_name, tool_basename):
# Select a label format based on the whether we're checking asm or IR.
if tool_basename == "llc":
check_label_format = "; %s-LABEL: %s:"
else:
check_label_format = "; %s-LABEL: @%s("
def add_checks(output_lines, prefix_list, func_dict, func_name, opt_basename):
# Label format is based on IR string.
check_label_format = "; %s-LABEL: @%s("
printed_prefixes = []
for checkprefixes, _ in prefix_list:
@ -212,8 +177,7 @@ def add_checks(output_lines, prefix_list, func_dict, func_name, tool_basename):
# For IR output, change all defs to FileCheck variables, so we're immune
# to variable naming fashions.
if tool_basename == "opt":
func_body = genericize_check_lines(func_body)
func_body = genericize_check_lines(func_body)
# This could be selectively enabled with an optional invocation argument.
# Disabled for now: better to check everything. Be safe rather than sorry.
@ -226,20 +190,14 @@ def add_checks(output_lines, prefix_list, func_dict, func_name, tool_basename):
# output_lines.append('; %s: %s' % (checkprefix, func_body[0]))
# is_blank_line = False
# For llc tests, there may be asm directives between the label and the
# first checked line (most likely that first checked line is "# BB#0").
if tool_basename == "opt":
is_blank_line = False
else:
is_blank_line = True;
is_blank_line = False
for func_line in func_body:
if func_line.strip() == '':
is_blank_line = True
continue
# Do not waste time checking IR comments.
if tool_basename == "opt":
func_line = SCRUB_IR_COMMENT_RE.sub(r'', func_line)
func_line = SCRUB_IR_COMMENT_RE.sub(r'', func_line)
# Skip blank lines instead of checking them.
if is_blank_line == True:
@ -275,8 +233,8 @@ def main():
parser = argparse.ArgumentParser(description=__doc__, formatter_class=RawTextHelpFormatter)
parser.add_argument('-v', '--verbose', action='store_true',
help='Show verbose output')
parser.add_argument('--tool-binary', default='llc',
help='The tool used to generate the test case')
parser.add_argument('--opt-binary', default='opt',
help='The opt binary used to generate the test case')
parser.add_argument(
'--function', help='The function in the test file to update')
parser.add_argument('tests', nargs='+')
@ -284,9 +242,9 @@ def main():
autogenerated_note = (ADVERT + 'utils/' + os.path.basename(__file__))
tool_basename = os.path.basename(args.tool_binary)
if (tool_basename != "llc" and tool_basename != "opt"):
print >>sys.stderr, 'ERROR: Unexpected tool name: ' + tool_basename
opt_basename = os.path.basename(args.opt_binary)
if (opt_basename != "opt"):
print >>sys.stderr, 'ERROR: Unexpected opt name: ' + opt_basename
sys.exit(1)
for test in args.tests:
@ -313,15 +271,15 @@ def main():
for l in run_lines:
(tool_cmd, filecheck_cmd) = tuple([cmd.strip() for cmd in l.split('|', 1)])
if not tool_cmd.startswith(tool_basename + ' '):
print >>sys.stderr, 'WARNING: Skipping non-%s RUN line: %s' % (tool_basename, l)
if not tool_cmd.startswith(opt_basename + ' '):
print >>sys.stderr, 'WARNING: Skipping non-%s RUN line: %s' % (opt_basename, l)
continue
if not filecheck_cmd.startswith('FileCheck '):
print >>sys.stderr, 'WARNING: Skipping non-FileChecked RUN line: ' + l
continue
tool_cmd_args = tool_cmd[len(tool_basename):].strip()
tool_cmd_args = tool_cmd[len(opt_basename):].strip()
tool_cmd_args = tool_cmd_args.replace('< %s', '').replace('%s', '').strip()
check_prefixes = [item for m in CHECK_PREFIX_RE.finditer(filecheck_cmd)
@ -337,13 +295,13 @@ def main():
for prefixes, _ in prefix_list:
for prefix in prefixes:
func_dict.update({prefix: dict()})
for prefixes, tool_args in prefix_list:
for prefixes, opt_args in prefix_list:
if args.verbose:
print >>sys.stderr, 'Extracted tool cmd: ' + tool_basename + ' ' + tool_args
print >>sys.stderr, 'Extracted opt cmd: ' + opt_basename + ' ' + opt_args
print >>sys.stderr, 'Extracted FileCheck prefixes: ' + str(prefixes)
raw_tool_output = invoke_tool(args, tool_args, test)
build_function_body_dictionary(raw_tool_output, prefixes, func_dict, args.verbose, tool_basename)
raw_tool_output = invoke_tool(args, opt_args, test)
build_function_body_dictionary(raw_tool_output, prefixes, func_dict, args.verbose, opt_basename)
is_in_function = False
is_in_function_start = False
@ -364,7 +322,7 @@ def main():
continue
# Print out the various check lines here.
output_lines = add_checks(output_lines, prefix_list, func_dict, name, tool_basename)
output_lines = add_checks(output_lines, prefix_list, func_dict, name, opt_basename)
is_in_function_start = False
if is_in_function: