v8/test/mjsunit/mjsunit.status
mvstanton@chromium.org cec8548d0e Correct handling of arrays with callbacks in the prototype chain.
Our generic KeyedStoreIC doesn't handle the case when a callback is
set on array elements in the prototype chain of the object, nor do
we recognize that we need to avoid the monomorphic case if these
callbacks exist.

This CL addresses the issue by looking for dictionary elements in
the prototype chain on IC misses and crankshaft element store
instructions. When found, the generic IC is used. The generic IC is
changed to go to the runtime in this case too.

In general, keyed loads are immune from this problem because they
won't return the hole: discovery of the hole goes to the runtime where
the callback will be found in the prototype chain. Double array loads
in crankshaft can return the hole but only if the prototype chain is
unaltered (we will catch such alterations).

Includes the following patch as well (already reviewed by bmeurer):
Performance regression found in test regress-2185-2.js. The problem was
that the bailout method for TransitionAndStoreStub was not performing
the appropriate transition.

(Review URL for the ElementsTransitionAndStoreIC_Miss change:
https://codereview.chromium.org/26911007)

R=danno@chromium.org

Review URL: https://codereview.chromium.org/35413006

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17525 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-11-06 15:45:43 +00:00

277 lines
11 KiB
Plaintext

# 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.
[
[ALWAYS, {
# All tests in the bug directory are expected to fail.
'bugs/*': [FAIL],
##############################################################################
# Flaky tests.
# BUG(v8:2921).
'debug-step-4-in-frame': [PASS, FLAKY],
##############################################################################
# Fails.
'regress/regress-1119': [FAIL],
# Issue 1719: Slow to collect arrays over several contexts.
'regress/regress-524': [SKIP],
# When that bug is fixed, revert the expectation to:
# Skip long running test in debug and allow it to timeout in release mode.
# regress/regress-524: [PASS, TIMEOUT, ['mode == debug', SKIP]],
# This test non-deterministically runs out of memory on Windows ia32.
'regress/regress-crbug-160010': [SKIP],
##############################################################################
# Too slow in debug mode with --stress-opt mode.
'compiler/regress-stacktrace-methods': [PASS, ['mode == debug', SKIP]],
'compiler/regress-funcaller': [PASS, ['mode == debug', SKIP]],
'regress/regress-2318': [PASS, ['mode == debug', SKIP]],
'regress/regress-create-exception': [PASS, ['mode == debug', SKIP]],
##############################################################################
# Too slow in debug mode for GC stress mode.
'regress/regress-crbug-217858': [PASS, ['mode == debug', SKIP]],
##############################################################################
# Only regexp stuff tested, no need for extensive Crankshaft tests.
'regexp-global': [PASS, NO_VARIANTS],
##############################################################################
# No need to waste time for this test.
'd8-performance-now': [PASS, NO_VARIANTS],
##############################################################################
# These use a built-in that's only present in debug mode. They take
# too long to run in debug mode on ARM and MIPS.
'fuzz-natives-part*': [PASS, ['mode == release or arch == arm or arch == android_arm or arch == mipsel', SKIP]],
'big-object-literal': [PASS, ['arch == arm or arch == android_arm', SKIP]],
# Issue 488: this test sometimes times out.
'array-constructor': [PASS, TIMEOUT],
# Very slow on ARM and MIPS, contains no architecture dependent code.
'unicode-case-overoptimization': [PASS, NO_VARIANTS, ['arch == arm or arch == android_arm or arch == mipsel', TIMEOUT]],
##############################################################################
# This test expects to reach a certain recursion depth, which may not work
# for debug mode.
'json-recursive': [PASS, ['mode == debug', PASS, FAIL]],
##############################################################################
# Skip long running tests that time out in debug mode.
'generated-transition-stub': [PASS, ['mode == debug', SKIP]],
##############################################################################
# This test sets the umask on a per-process basis and hence cannot be
# used in multi-threaded runs.
# On android there is no /tmp directory.
'd8-os': [PASS, ['isolates or arch == android_arm or arch == android_ia32', SKIP]],
'tools/tickprocessor': [PASS, ['arch == android_arm or arch == android_ia32', SKIP]],
##############################################################################
# Long running test that reproduces memory leak and should be run manually.
'regress/regress-2073': [SKIP],
}], # ALWAYS
##############################################################################
['arch == arm or arch == android_arm', {
# Slow tests which times out in debug mode.
'try': [PASS, ['mode == debug', SKIP]],
'debug-scripts-request': [PASS, ['mode == debug', SKIP]],
'array-constructor': [PASS, ['mode == debug', SKIP]],
'regress/regress-1122': [PASS, ['mode == debug and arch == android_arm', SKIP]],
# Flaky test that can hit compilation-time stack overflow in debug mode.
'unicode-test': [PASS, ['mode == debug', PASS, FAIL]],
# Times out often in release mode on ARM.
'compiler/regress-stacktrace-methods': [PASS, PASS, ['mode == release', TIMEOUT]],
'array-splice': [PASS, TIMEOUT],
# Long running test.
'string-indexof-2': [PASS, TIMEOUT],
'mirror-object': [PASS, TIMEOUT],
# BUG(3251035): Timeouts in long looping crankshaft optimization
# tests. Skipping because having them timeout takes too long on the
# buildbot.
'compiler/alloc-number': [SKIP],
'compiler/array-length': [SKIP],
'compiler/assignment-deopt': [SKIP],
'compiler/deopt-args': [SKIP],
'compiler/inline-compare': [SKIP],
'compiler/inline-global-access': [SKIP],
'compiler/optimized-function-calls': [SKIP],
'compiler/pic': [SKIP],
'compiler/property-calls': [SKIP],
'compiler/recursive-deopt': [SKIP],
'compiler/regress-4': [SKIP],
'compiler/regress-funcaller': [SKIP],
'compiler/regress-rep-change': [SKIP],
'compiler/regress-arguments': [SKIP],
'compiler/regress-funarguments': [SKIP],
'compiler/regress-3249650': [SKIP],
'compiler/simple-deopt': [SKIP],
'regress/regress-490': [SKIP],
'regress/regress-634': [SKIP],
'regress/regress-create-exception': [SKIP],
'regress/regress-3218915': [SKIP],
'regress/regress-3247124': [SKIP],
# Requires bigger stack size in the Genesis and if stack size is increased,
# the test requires too much time to run. However, the problem test covers
# should be platform-independent.
'regress/regress-1132': [SKIP],
# Stack manipulations in LiveEdit is not implemented for this arch.
'debug-liveedit-check-stack': [SKIP],
'debug-liveedit-stack-padding': [SKIP],
'debug-liveedit-restart-frame': [SKIP],
'debug-liveedit-double-call': [SKIP],
# Currently always deopt on minus zero
'math-floor-of-div-minus-zero': [SKIP],
}], # 'arch == arm or arch == android_arm'
##############################################################################
['arch == mipsel', {
# Slow tests which times out in debug mode.
'try': [PASS, ['mode == debug', SKIP]],
'debug-scripts-request': [PASS, ['mode == debug', SKIP]],
'array-constructor': [PASS, ['mode == debug', SKIP]],
# Times out often in release mode on MIPS.
'compiler/regress-stacktrace-methods': [PASS, PASS, ['mode == release', TIMEOUT]],
'array-splice': [PASS, TIMEOUT],
# Long running test.
'mirror-object': [PASS, TIMEOUT],
'string-indexof-2': [PASS, TIMEOUT],
# BUG(3251035): Timeouts in long looping crankshaft optimization
# tests. Skipping because having them timeout takes too long on the
# buildbot.
'compiler/alloc-number': [SKIP],
'compiler/array-length': [SKIP],
'compiler/assignment-deopt': [SKIP],
'compiler/deopt-args': [SKIP],
'compiler/inline-compare': [SKIP],
'compiler/inline-global-access': [SKIP],
'compiler/optimized-function-calls': [SKIP],
'compiler/pic': [SKIP],
'compiler/property-calls': [SKIP],
'compiler/recursive-deopt': [SKIP],
'compiler/regress-4': [SKIP],
'compiler/regress-funcaller': [SKIP],
'compiler/regress-rep-change': [SKIP],
'compiler/regress-arguments': [SKIP],
'compiler/regress-funarguments': [SKIP],
'compiler/regress-3249650': [SKIP],
'compiler/simple-deopt': [SKIP],
'regress/regress-490': [SKIP],
'regress/regress-634': [SKIP],
'regress/regress-create-exception': [SKIP],
'regress/regress-3218915': [SKIP],
'regress/regress-3247124': [SKIP],
# Requires bigger stack size in the Genesis and if stack size is increased,
# the test requires too much time to run. However, the problem test covers
# should be platform-independent.
'regress/regress-1132': [SKIP],
# Stack manipulations in LiveEdit is not implemented for this arch.
'debug-liveedit-check-stack': [SKIP],
'debug-liveedit-stack-padding': [SKIP],
'debug-liveedit-restart-frame': [SKIP],
'debug-liveedit-double-call': [SKIP],
# Currently always deopt on minus zero
'math-floor-of-div-minus-zero': [SKIP],
}], # 'arch == mipsel'
##############################################################################
# Native Client uses the ARM simulator so will behave similarly to arm
# on mjsunit tests.
# TODO(bradchen): enable more tests for NaCl V8 when it stops using
# the ARM simulator.
##############################################################################
['arch == nacl_ia32 or arch == nacl_x64', {
# There is no /tmp directory for NaCl runs
'd8-os': [SKIP],
# Stack manipulations in LiveEdit is not implemented for this arch.
'debug-liveedit-check-stack': [SKIP],
'debug-liveedit-stack-padding': [SKIP],
'debug-liveedit-restart-frame': [SKIP],
'debug-liveedit-double-call': [SKIP],
# This test dumps core for arm.debug, so no reason to expect it to work
# for NaCl. The other three fuzz-natives tests seem to run fine.
# As noted above none of them are run in the arm.debug case.
'fuzz-natives-part4': [SKIP],
# NaCl builds have problems with this test since Pepper_28.
# V8 Issue 2786
'math-exp-precision': [SKIP],
# Requires bigger stack size in the Genesis and if stack size is increased,
# the test requires too much time to run. However, the problem test covers
# should be platform-independent.
'regress/regress-1132': [SKIP],
# Poor performance for NaCl V8 causes an assertion failure for this test.
'regress/regress-165637': [SKIP],
# Skip long running test that times out in debug mode and goes OOM on NaCl.
'regress/regress-crbug-160010': [SKIP],
# Bug(v8:2978).
'lithium/MathExp': [PASS, FAIL],
}], # 'arch == nacl_ia32 or arch == nacl_x64'
##############################################################################
['deopt_fuzzer == True', {
# Skip tests that are not suitable for deoptimization fuzzing.
'assert-opt-and-deopt': [SKIP],
'never-optimize': [SKIP],
'regress/regress-2185-2': [SKIP],
'harmony/object-observe': [SKIP],
'readonly': [SKIP],
'array-feedback': [SKIP],
# Deopt every n garbage collections collides with deopt every n times.
'regress/regress-2653': [SKIP],
}], # 'deopt_fuzzer == True'
]