CHECK failure: object.is_null() || *object == scope_site->transition_info() in allocation-site- |
||||||
Issue descriptionDetailed report: https://clusterfuzz.com/testcase?key=5505985834909696 Fuzzer: decoder_langfuzz Job Type: linux_asan_d8_dbg Platform Id: linux Crash Type: CHECK failure Crash Address: Crash State: object.is_null() || *object == scope_site->transition_info() in allocation-site- Sanitizer: address (ASAN) Regressed: V8: 44423:44424 Reproducer Testcase: https://clusterfuzz.com/download/AMIfv962FFPDAzHkSqpCXr2sAHVxAshLCHri1Fp6yvQmIaQunJD6PN_Ep3mElcHyzSj4YZgvLSOz1gf_Gy2E9P-E9g43kmexm86sqkrD6LM33vvPf9xMnCxTAvRDW5FTpBga0pZ1FW8gR80ZFnC7JOr8Mk-iLLZCCeLor1bdY3fBD6yfu96VnXTJNWUbqt0VaJzPl3K8eEfvmox4QENtrinxFbWv1daI_L9uq-DPYajmy_yzQzJRGksp7NQ1JzrXfZIGhiJf8oXO8nuAWbNEyXsbu4CU7C7xMDRb4E67BUM4lH0FG826zL6GBcziReqGlg5uP0Z6Ttz0rFvaCHf_CFiNlJD94U84dyRzI3xorBITn9-lP0iEsq8?testcase_id=5505985834909696 Issue filed automatically. See https://dev.chromium.org/Home/chromium-security/bugs/reproducing-clusterfuzz-bugs for more information.
,
Apr 10 2017
Regression range points to bff3b266d23cf0054e687b150f949acc6ce2e1a9. Probably just flushes out an existing issue. Reproduces nicely though.
,
Apr 10 2017
Simplified repro:
---------------------------------------------------------------
// Flags: --allow-natives-syntax
function foo() {
return {0xd: {}, b: {}, c: {}, d: {}, e: {}, f: {}, g: {}};
}
foo();
foo();
%OptimizeFunctionOnNextCall(foo);
foo();
---------------------------------------------------------------
Also explodes in Crankshaft (with --noturbo), which has the old limit of 8. So it's not my CL. Problem is fixed by replacing 0xd with a, so probably there's some mismatch between the runtime and the compiler.
,
Apr 10 2017
,
Apr 11 2017
I'll take a look at this.
,
Apr 11 2017
The following revision refers to this bug: https://chromium.googlesource.com/v8/v8.git/+/1f3a863bbdb972f3314c9424ff6939152d4dd9ac commit 1f3a863bbdb972f3314c9424ff6939152d4dd9ac Author: Michael Starzinger <mstarzinger@chromium.org> Date: Tue Apr 11 11:42:52 2017 [turbofan] Fix traversal order of boilerplate objects. This fixes {JSCreateLowering} to traverse boilerplate objects in the same order the runtime uses (i.e. properties first, elements second). That order is hard-coded in the nesting of {AllocationSite} objects. R=bmeurer@chromium.org TEST=mjsunit/regress/regress-crbug-709537 BUG= chromium:709537 Change-Id: I8f446a0880448ea88a3e242e92d11d611581a42b Reviewed-on: https://chromium-review.googlesource.com/474028 Reviewed-by: Benedikt Meurer <bmeurer@chromium.org> Commit-Queue: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#44563} [modify] https://crrev.com/1f3a863bbdb972f3314c9424ff6939152d4dd9ac/src/compiler/js-create-lowering.cc [modify] https://crrev.com/1f3a863bbdb972f3314c9424ff6939152d4dd9ac/src/crankshaft/hydrogen.cc [add] https://crrev.com/1f3a863bbdb972f3314c9424ff6939152d4dd9ac/test/mjsunit/regress/regress-crbug-709537.js
,
Apr 11 2017
,
Apr 12 2017
ClusterFuzz has detected this issue as fixed in range 44562:44563. Detailed report: https://clusterfuzz.com/testcase?key=5505985834909696 Fuzzer: decoder_langfuzz Job Type: linux_asan_d8_dbg Platform Id: linux Crash Type: CHECK failure Crash Address: Crash State: object.is_null() || *object == scope_site->transition_info() in allocation-site- Sanitizer: address (ASAN) Regressed: V8: 44423:44424 Fixed: V8: 44562:44563 Reproducer Testcase: https://clusterfuzz.com/download/AMIfv962FFPDAzHkSqpCXr2sAHVxAshLCHri1Fp6yvQmIaQunJD6PN_Ep3mElcHyzSj4YZgvLSOz1gf_Gy2E9P-E9g43kmexm86sqkrD6LM33vvPf9xMnCxTAvRDW5FTpBga0pZ1FW8gR80ZFnC7JOr8Mk-iLLZCCeLor1bdY3fBD6yfu96VnXTJNWUbqt0VaJzPl3K8eEfvmox4QENtrinxFbWv1daI_L9uq-DPYajmy_yzQzJRGksp7NQ1JzrXfZIGhiJf8oXO8nuAWbNEyXsbu4CU7C7xMDRb4E67BUM4lH0FG826zL6GBcziReqGlg5uP0Z6Ttz0rFvaCHf_CFiNlJD94U84dyRzI3xorBITn9-lP0iEsq8?testcase_id=5505985834909696 See https://dev.chromium.org/Home/chromium-security/bugs/reproducing-clusterfuzz-bugs for more information. If you suspect that the result above is incorrect, try re-doing that job on the test case report page. |
||||||
►
Sign in to add a comment |
||||||
Comment 1 by mummare...@chromium.org
, Apr 7 2017