cebcd8c51b
This reverts commit ebd933037e
.
Reason for revert: Breaks a test: https://logs.chromium.org/logs/v8/buildbucket/cr-buildbucket/8792462319927467985/+/u/OptimizeForSize/CreateIsolateFromReadOnlySnapshot
Original change's description:
> [flags,testrunner] Consider readonly flags for conflict detection
>
> Flag conflict detection 1) bails out on incompatible flag values (e.g.
> --jitless and --turbofan) and 2) handles such bailouts transparently in
> the test runner by marking affected tests as OUTCOMES_FAIL.
>
> This CL adds full support for readonly flags to this system, together
> with required additional annotations in variants.py.
>
> Drive-by: assert proper use of v8_enable_slow_dchecks, and add
> support when dcheck_always_on is set.
> Drive-by: introduce has_maglev build variable detection based on
> v8_enable_maglev and use that for .status file annotations.
> Drive-by: protect against unintended overwrites of build variables
> in statusfile.py.
>
> Cq-Include-Trybots: luci.v8.try:v8_linux64_fyi_rel
> Bug: v8:13629,v8:10577
> Change-Id: I04de399139a0490806df8bfee7e75e2ec767b4b5
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/4135879
> Reviewed-by: Tobias Tebbi <tebbi@chromium.org>
> Reviewed-by: Victor Gomes <victorgomes@chromium.org>
> Commit-Queue: Jakob Linke <jgruber@chromium.org>
> Cr-Commit-Position: refs/heads/main@{#85130}
Bug: v8:13629,v8:10577
Change-Id: I0cb072c6c9f05d92894cc0af83c4d1a28df100d5
Cq-Include-Trybots: luci.v8.try:v8_linux64_fyi_rel
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/4147098
Bot-Commit: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com>
Commit-Queue: Victor Gomes <victorgomes@chromium.org>
Reviewed-by: Victor Gomes <victorgomes@chromium.org>
Auto-Submit: Leszek Swirski <leszeks@chromium.org>
Reviewed-by: Michael Achenbach <machenbach@chromium.org>
Cr-Commit-Position: refs/heads/main@{#85133}
342 lines
11 KiB
Python
342 lines
11 KiB
Python
# Copyright 2012 the V8 project authors. All rights reserved.
|
|
# Redistribution and use in source and binary forms, with or without
|
|
# modification, are permitted provided that the following conditions are
|
|
# met:
|
|
#
|
|
# * Redistributions of source code must retain the above copyright
|
|
# notice, this list of conditions and the following disclaimer.
|
|
# * Redistributions in binary form must reproduce the above
|
|
# copyright notice, this list of conditions and the following
|
|
# disclaimer in the documentation and/or other materials provided
|
|
# with the distribution.
|
|
# * Neither the name of Google Inc. nor the names of its
|
|
# contributors may be used to endorse or promote products derived
|
|
# from this software without specific prior written permission.
|
|
#
|
|
# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
|
|
# "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
|
|
# LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
|
|
# A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
|
|
# OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
|
|
# SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
|
|
# LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
|
|
# DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
|
|
# THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
|
|
# (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
|
|
# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
|
|
|
|
import os
|
|
import re
|
|
|
|
from testrunner.local.variants import ALL_VARIANTS
|
|
from testrunner.local.utils import Freeze
|
|
|
|
# Possible outcomes
|
|
FAIL = "FAIL"
|
|
PASS = "PASS"
|
|
TIMEOUT = "TIMEOUT"
|
|
CRASH = "CRASH"
|
|
|
|
# Outcomes only for status file, need special handling
|
|
FAIL_OK = "FAIL_OK"
|
|
FAIL_SLOPPY = "FAIL_SLOPPY"
|
|
|
|
# Modifiers
|
|
HEAVY = "HEAVY"
|
|
SKIP = "SKIP"
|
|
SLOW = "SLOW"
|
|
NO_VARIANTS = "NO_VARIANTS"
|
|
FAIL_PHASE_ONLY = "FAIL_PHASE_ONLY"
|
|
|
|
ALWAYS = "ALWAYS"
|
|
|
|
KEYWORDS = {}
|
|
for key in [SKIP, FAIL, PASS, CRASH, HEAVY, SLOW, FAIL_OK, NO_VARIANTS,
|
|
FAIL_SLOPPY, ALWAYS, FAIL_PHASE_ONLY]:
|
|
KEYWORDS[key] = key
|
|
|
|
# Support arches, modes to be written as keywords instead of strings.
|
|
VARIABLES = {ALWAYS: True}
|
|
for var in [
|
|
"debug", "release", "big", "little", "android", "arm", "arm64", "ia32",
|
|
"mips64", "mips64el", "x64", "ppc", "ppc64", "s390", "s390x", "macos",
|
|
"windows", "linux", "aix", "r1", "r2", "r3", "r5", "r6", "riscv32",
|
|
"riscv64", "loong64"
|
|
]:
|
|
VARIABLES[var] = var
|
|
|
|
# Allow using variants as keywords.
|
|
for var in ALL_VARIANTS:
|
|
VARIABLES[var] = var
|
|
|
|
class StatusFile(object):
|
|
def __init__(self, path, variables):
|
|
"""
|
|
_rules: {variant: {test name: [rule]}}
|
|
_prefix_rules: {variant: {test name prefix: [rule]}}
|
|
"""
|
|
self.variables = variables
|
|
with open(path) as f:
|
|
self._rules, self._prefix_rules = ReadStatusFile(f.read(), variables)
|
|
|
|
def get_outcomes(self, testname, variant=None):
|
|
"""Merges variant dependent and independent rules."""
|
|
outcomes = frozenset()
|
|
|
|
for key in set([variant or '', '']):
|
|
rules = self._rules.get(key, {})
|
|
prefix_rules = self._prefix_rules.get(key, {})
|
|
|
|
if testname in rules:
|
|
outcomes |= rules[testname]
|
|
|
|
for prefix in prefix_rules:
|
|
if testname.startswith(prefix):
|
|
outcomes |= prefix_rules[prefix]
|
|
|
|
return outcomes
|
|
|
|
def warn_unused_rules(self, tests, check_variant_rules=False):
|
|
"""Finds and prints unused rules in status file.
|
|
|
|
Rule X is unused when it doesn't apply to any tests, which can also mean
|
|
that all matching tests were skipped by another rule before evaluating X.
|
|
|
|
Args:
|
|
tests: list of pairs (testname, variant)
|
|
check_variant_rules: if set variant dependent rules are checked
|
|
"""
|
|
|
|
if check_variant_rules:
|
|
variants = list(ALL_VARIANTS)
|
|
else:
|
|
variants = ['']
|
|
used_rules = set()
|
|
|
|
for testname, variant in tests:
|
|
variant = variant or ''
|
|
|
|
if testname in self._rules.get(variant, {}):
|
|
used_rules.add((testname, variant))
|
|
if SKIP in self._rules[variant][testname]:
|
|
continue
|
|
|
|
for prefix in self._prefix_rules.get(variant, {}):
|
|
if testname.startswith(prefix):
|
|
used_rules.add((prefix, variant))
|
|
if SKIP in self._prefix_rules[variant][prefix]:
|
|
break
|
|
|
|
for variant in variants:
|
|
for rule, value in (
|
|
list(self._rules.get(variant, {}).items()) +
|
|
list(self._prefix_rules.get(variant, {}).items())):
|
|
if (rule, variant) not in used_rules:
|
|
if variant == '':
|
|
variant_desc = 'variant independent'
|
|
else:
|
|
variant_desc = 'variant: %s' % variant
|
|
print('Unused rule: %s -> %s (%s)' % (rule, value, variant_desc))
|
|
|
|
|
|
def _JoinsPassAndFail(outcomes1, outcomes2):
|
|
"""Indicates if we join PASS and FAIL from two different outcome sets and
|
|
the first doesn't already contain both.
|
|
"""
|
|
return (
|
|
PASS in outcomes1 and
|
|
not (FAIL in outcomes1 or FAIL_OK in outcomes1) and
|
|
(FAIL in outcomes2 or FAIL_OK in outcomes2)
|
|
)
|
|
|
|
VARIANT_EXPRESSION = object()
|
|
|
|
def _EvalExpression(exp, variables):
|
|
"""Evaluates expression and returns its result. In case of NameError caused by
|
|
undefined "variant" identifier returns VARIANT_EXPRESSION marker.
|
|
"""
|
|
|
|
try:
|
|
return eval(exp, variables)
|
|
except NameError as e:
|
|
identifier = re.match("name '(.*)' is not defined", str(e)).group(1)
|
|
assert identifier == "variant", "Unknown identifier: %s" % identifier
|
|
return VARIANT_EXPRESSION
|
|
|
|
|
|
def _EvalVariantExpression(
|
|
condition, section, variables, variant, rules, prefix_rules):
|
|
variables_with_variant = dict(variables)
|
|
variables_with_variant["variant"] = variant
|
|
result = _EvalExpression(condition, variables_with_variant)
|
|
assert result != VARIANT_EXPRESSION
|
|
if result is True:
|
|
_ReadSection(
|
|
section,
|
|
variables_with_variant,
|
|
rules[variant],
|
|
prefix_rules[variant],
|
|
)
|
|
else:
|
|
assert result is False, "Make sure expressions evaluate to boolean values"
|
|
|
|
|
|
def _ParseOutcomeList(rule, outcomes, variables, target_dict):
|
|
"""Outcome list format: [condition, outcome, outcome, ...]"""
|
|
|
|
result = set([])
|
|
if type(outcomes) == str:
|
|
outcomes = [outcomes]
|
|
for item in outcomes:
|
|
if type(item) == str:
|
|
result.add(item)
|
|
elif type(item) == list:
|
|
condition = item[0]
|
|
exp = _EvalExpression(condition, variables)
|
|
assert exp != VARIANT_EXPRESSION, (
|
|
"Nested variant expressions are not supported")
|
|
if exp is False:
|
|
continue
|
|
|
|
# Ensure nobody uses an identifier by mistake, like "default",
|
|
# which would evaluate to true here otherwise.
|
|
assert exp is True, "Make sure expressions evaluate to boolean values"
|
|
|
|
for outcome in item[1:]:
|
|
assert type(outcome) == str
|
|
result.add(outcome)
|
|
else:
|
|
assert False
|
|
if len(result) == 0:
|
|
return
|
|
if rule in target_dict:
|
|
# A FAIL without PASS in one rule has always precedence over a single
|
|
# PASS (without FAIL) in another. Otherwise the default PASS expectation
|
|
# in a rule with a modifier (e.g. PASS, SLOW) would be joined to a FAIL
|
|
# from another rule (which intended to mark a test as FAIL and not as
|
|
# PASS and FAIL).
|
|
if _JoinsPassAndFail(target_dict[rule], result):
|
|
target_dict[rule] -= set([PASS])
|
|
if _JoinsPassAndFail(result, target_dict[rule]):
|
|
result -= set([PASS])
|
|
target_dict[rule] |= result
|
|
else:
|
|
target_dict[rule] = result
|
|
|
|
|
|
def ReadContent(content):
|
|
return eval(content, KEYWORDS)
|
|
|
|
|
|
def ReadStatusFile(content, variables):
|
|
"""Status file format
|
|
Status file := [section]
|
|
section = [CONDITION, section_rules]
|
|
section_rules := {path: outcomes}
|
|
outcomes := outcome | [outcome, ...]
|
|
outcome := SINGLE_OUTCOME | [CONDITION, SINGLE_OUTCOME, SINGLE_OUTCOME, ...]
|
|
"""
|
|
|
|
# Empty defaults for rules and prefix_rules. Variant-independent
|
|
# rules are mapped by "", others by the variant name.
|
|
rules = {variant: {} for variant in ALL_VARIANTS}
|
|
rules[""] = {}
|
|
prefix_rules = {variant: {} for variant in ALL_VARIANTS}
|
|
prefix_rules[""] = {}
|
|
|
|
variables.update(VARIABLES)
|
|
for conditional_section in ReadContent(content):
|
|
assert type(conditional_section) == list
|
|
assert len(conditional_section) == 2
|
|
condition, section = conditional_section
|
|
exp = _EvalExpression(condition, variables)
|
|
|
|
# The expression is variant-independent and evaluates to False.
|
|
if exp is False:
|
|
continue
|
|
|
|
# The expression is variant-independent and evaluates to True.
|
|
if exp is True:
|
|
_ReadSection(
|
|
section,
|
|
variables,
|
|
rules[''],
|
|
prefix_rules[''],
|
|
)
|
|
continue
|
|
|
|
# The expression is variant-dependent (contains "variant" keyword)
|
|
if exp == VARIANT_EXPRESSION:
|
|
# If the expression contains one or more "variant" keywords, we evaluate
|
|
# it for all possible variants and create rules for those that apply.
|
|
for variant in ALL_VARIANTS:
|
|
_EvalVariantExpression(
|
|
condition, section, variables, variant, rules, prefix_rules)
|
|
continue
|
|
|
|
assert False, "Make sure expressions evaluate to boolean values"
|
|
|
|
return Freeze(rules), Freeze(prefix_rules)
|
|
|
|
|
|
def _ReadSection(section, variables, rules, prefix_rules):
|
|
assert type(section) == dict
|
|
for rule, outcome_list in list(section.items()):
|
|
assert type(rule) == str
|
|
|
|
if rule[-1] == '*':
|
|
_ParseOutcomeList(rule[:-1], outcome_list, variables, prefix_rules)
|
|
else:
|
|
_ParseOutcomeList(rule, outcome_list, variables, rules)
|
|
|
|
JS_TEST_PATHS = {
|
|
'debugger': [[]],
|
|
'inspector': [[]],
|
|
'intl': [[]],
|
|
'message': [[]],
|
|
'mjsunit': [[]],
|
|
'mozilla': [['data']],
|
|
'test262': [['data', 'test'], ['local-tests', 'test']],
|
|
'webkit': [[]],
|
|
}
|
|
|
|
FILE_EXTENSIONS = [".js", ".mjs"]
|
|
|
|
def PresubmitCheck(path):
|
|
with open(path) as f:
|
|
contents = ReadContent(f.read())
|
|
basename = os.path.basename(os.path.dirname(path))
|
|
root_prefix = basename + "/"
|
|
status = {"success": True}
|
|
def _assert(check, message): # Like "assert", but doesn't throw.
|
|
if not check:
|
|
print("%s: Error: %s" % (path, message))
|
|
status["success"] = False
|
|
try:
|
|
for section in contents:
|
|
_assert(type(section) == list, "Section must be a list")
|
|
_assert(len(section) == 2, "Section list must have exactly 2 entries")
|
|
section = section[1]
|
|
_assert(type(section) == dict,
|
|
"Second entry of section must be a dictionary")
|
|
for rule in section:
|
|
_assert(type(rule) == str, "Rule key must be a string")
|
|
_assert(not rule.startswith(root_prefix),
|
|
"Suite name prefix must not be used in rule keys")
|
|
_assert(not rule.endswith('.js'),
|
|
".js extension must not be used in rule keys.")
|
|
_assert('*' not in rule or (rule.count('*') == 1 and rule[-1] == '*'),
|
|
"Only the last character of a rule key can be a wildcard")
|
|
if basename in JS_TEST_PATHS and '*' not in rule:
|
|
def _any_exist(paths):
|
|
return any(os.path.exists(os.path.join(os.path.dirname(path),
|
|
*(paths + [rule + ext])))
|
|
for ext in FILE_EXTENSIONS)
|
|
_assert(any(_any_exist(paths)
|
|
for paths in JS_TEST_PATHS[basename]),
|
|
"missing file for %s test %s" % (basename, rule))
|
|
return status["success"]
|
|
except Exception as e:
|
|
print(e)
|
|
return False
|