New issue
Advanced search Search tips

Issue 684856 link

Starred by 2 users
Status: Verified
Owner:
keishi@chromium.org
Closed: Feb 2017
Cc:
tzik@chromium.org
keishi@chromium.org
Components:
EstimatedDays: ----
NextAction: ----
OS: Linux
Pri: 1
Type: Bug



Sign in to add a comment

Lock-order-inversion in pthread_mutex_lock

Project Member Reported by ClusterFuzz, Jan 24 2017 
Detailed report: https://cluster-fuzz.appspot.com/testcase?key=4856224299614208

Fuzzer: inferno_twister
Job Type: linux_tsan_chrome_mp
Platform Id: linux

Crash Type: Lock-order-inversion
Crash Address: 
Crash State:
  pthread_mutex_lock
  blink::AudioBufferSourceHandler::setBuffer
  blink::AudioBufferSourceNode::setBuffer
  
Sanitizer: thread (TSAN)

Regressed: https://cluster-fuzz.appspot.com/revisions?job=linux_tsan_chrome_mp&range=371187:371266

Minimized Testcase (0.64 Kb):
Download: https://cluster-fuzz.appspot.com/download/AMIfv94JoXNuTWOCGewrLqNbQelBBgVuUHW4eMFLqIeD78_YMkl-XULq2tktOjlbEOYOt4Kp767XcohWRKJXoPp6jXSvQwGQrFTfpD86QFU0qhARlS7HOoHnYBLx_zHMlKXbSsaW863oNYNQbqK1iLewQZxY-556KneCUc7Aok9lFsDggkT3mW-lET_SpUA7Ul5acvFUy4xssSFbcwkUU_djyBiZef24qxLrV82NaqmQttBusIyQ8vEFn3XvpmCa97bn95hZYMNTH4BEHXBbOZp-7uCu-3CkWP_gVTYEsxNAnapdZwds5PnNBy44_kMNLtp-jAmlbHlP1yOqqw8LF0E6AKzCrBYi1vlFbJ0wHA4d0BvW6IPkDvA?testcase_id=4856224299614208
<script>
		var sampleRate=44100.0;
		(function() {
		})();
		/*
		*/
		(function() {
			var zeroElementCurve=[];
			executeTest(zeroElementCurve, "Testing zero-element curve (unspecified result)");
		})();
		function executeTest(curveData, inputData) {
			var ac=new OfflineAudioContext(1, inputData.length, sampleRate);
			var waveShaper=ac.createWaveShaper();
			waveShaper.connect(ac.destination);
			var inputBuffer=ac.createBuffer(1, 2, sampleRate);
			var src=ac.createBufferSource();
			src.buffer=inputBuffer;
			src.connect(waveShaper);
			src.start();
			ac.startRendering();
		}
setTimeout("tCFcrash()");
function tCFcrash() {
gc()
}</script>


Issue filed automatically.

See https://dev.chromium.org/Home/chromium-security/bugs/reproducing-clusterfuzz-bugs for more information.

Comment 1 by mummare...@chromium.org, Jan 25 2017

Cc: tzik@chromium.org
Components: Blink>WebAudio
Labels: Test-Predator-Wrong M-56
Owner: keishi@chromium.org
Status: Assigned (was: Untriaged)
As per  issue 681329 , assigning to keishi@. could you please take a look and help us to find correct owner if it is not related your changes.
Thank you.

Comment 2 by rtoy@chromium.org, Jan 25 2017

Cc: keishi@chromium.org
Owner: rtoy@chromium.org
This looks like a webaudio problem. I'll take it.

And for the record, I can reproduce, but only if I keep reloading the minimized test case.

Comment 3 by rtoy@chromium.org, Jan 25 2017

Owner: keishi@chromium.org
Could this be a GC/oilpan problem?  Possibly related to issue 681527?

Assigning to keishi again, since you're also looking at 681527.
Project Member

Comment 4 by bugdroid1@chromium.org, Feb 14 2017 

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/52b19c8f68c222c08944af937a01f3ff7e46705c

commit 52b19c8f68c222c08944af937a01f3ff7e46705c
Author: keishi <keishi@chromium.org>
Date: Tue Feb 14 04:47:46 2017

Move postGC out of CrossThreadPersistentRegion::LockScope

Some finalizers/prefinalizers acquire locks, causing a dead lock when CrossThreadPersistent tries to acquire the CrossThreadPersistentRegion lock.

This CL moves the postGC step outside of the CrossThreadPersistentRegion::LockScope. This should be okay as the CrossThreadPersistentRegion::LockScope only exists to avoid add/deletetion of CrossThreadPersistentNodes during marking.

BUG= 684856 , 685624 , 685624 

Review-Url: https://codereview.chromium.org/2686533003
Cr-Commit-Position: refs/heads/master@{#450251}

[modify] https://crrev.com/52b19c8f68c222c08944af937a01f3ff7e46705c/third_party/WebKit/Source/platform/heap/Heap.cpp
[modify] https://crrev.com/52b19c8f68c222c08944af937a01f3ff7e46705c/third_party/WebKit/Source/platform/heap/Heap.h
[modify] https://crrev.com/52b19c8f68c222c08944af937a01f3ff7e46705c/third_party/WebKit/Source/platform/heap/HeapTest.cpp
[modify] https://crrev.com/52b19c8f68c222c08944af937a01f3ff7e46705c/third_party/WebKit/Source/platform/heap/ThreadState.cpp
[modify] https://crrev.com/52b19c8f68c222c08944af937a01f3ff7e46705c/third_party/WebKit/Source/platform/heap/ThreadState.h
Project Member

Comment 5 by ClusterFuzz, Feb 14 2017 

ClusterFuzz has detected this issue as fixed in range 450202:450256.

Detailed report: https://cluster-fuzz.appspot.com/testcase?key=4856224299614208

Fuzzer: inferno_twister
Job Type: linux_tsan_chrome_mp
Platform Id: linux

Crash Type: Lock-order-inversion
Crash Address: 
Crash State:
  pthread_mutex_lock
  blink::AudioBufferSourceHandler::setBuffer
  blink::AudioBufferSourceNode::setBuffer
  
Sanitizer: thread (TSAN)

Regressed: https://cluster-fuzz.appspot.com/revisions?job=linux_tsan_chrome_mp&range=371187:371266
Fixed: https://cluster-fuzz.appspot.com/revisions?job=linux_tsan_chrome_mp&range=450202:450256

Reproducer Testcase: https://cluster-fuzz.appspot.com/download/AMIfv94JoXNuTWOCGewrLqNbQelBBgVuUHW4eMFLqIeD78_YMkl-XULq2tktOjlbEOYOt4Kp767XcohWRKJXoPp6jXSvQwGQrFTfpD86QFU0qhARlS7HOoHnYBLx_zHMlKXbSsaW863oNYNQbqK1iLewQZxY-556KneCUc7Aok9lFsDggkT3mW-lET_SpUA7Ul5acvFUy4xssSFbcwkUU_djyBiZef24qxLrV82NaqmQttBusIyQ8vEFn3XvpmCa97bn95hZYMNTH4BEHXBbOZp-7uCu-3CkWP_gVTYEsxNAnapdZwds5PnNBy44_kMNLtp-jAmlbHlP1yOqqw8LF0E6AKzCrBYi1vlFbJ0wHA4d0BvW6IPkDvA?testcase_id=4856224299614208


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.
Project Member

Comment 6 by ClusterFuzz, Feb 14 2017

Labels: ClusterFuzz-Verified
Status: Verified (was: Assigned)
ClusterFuzz testcase 4856224299614208 is verified as fixed, so closing issue.

If this is incorrect, please add ClusterFuzz-Wrong label and re-open the issue.
Project Member

Comment 7 by bugdroid1@chromium.org, Feb 16 2017

Labels: merge-merged-2987
The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/590a7c6ceb9bea92bbce8e7636406ed6ea39bfe8

commit 590a7c6ceb9bea92bbce8e7636406ed6ea39bfe8
Author: Keishi Hattori <keishi@chromium.org>
Date: Thu Feb 16 08:00:38 2017

Move postGC out of CrossThreadPersistentRegion::LockScope

Some finalizers/prefinalizers acquire locks, causing a dead lock when CrossThreadPersistent tries to acquire the CrossThreadPersistentRegion lock.

This CL moves the postGC step outside of the CrossThreadPersistentRegion::LockScope. This should be okay as the CrossThreadPersistentRegion::LockScope only exists to avoid add/deletetion of CrossThreadPersistentNodes during marking.

BUG=681527, 684856 , 685624 , 685624 

Review-Url: https://codereview.chromium.org/2686533003
Cr-Commit-Position: refs/heads/master@{#450251}
(cherry picked from commit 52b19c8f68c222c08944af937a01f3ff7e46705c)

Review-Url: https://codereview.chromium.org/2695063005 .
Cr-Commit-Position: refs/branch-heads/2987@{#539}
Cr-Branched-From: ad51088c0e8776e8dcd963dbe752c4035ba6dab6-refs/heads/master@{#444943}

[modify] https://crrev.com/590a7c6ceb9bea92bbce8e7636406ed6ea39bfe8/third_party/WebKit/Source/platform/heap/Heap.cpp
[modify] https://crrev.com/590a7c6ceb9bea92bbce8e7636406ed6ea39bfe8/third_party/WebKit/Source/platform/heap/Heap.h
[modify] https://crrev.com/590a7c6ceb9bea92bbce8e7636406ed6ea39bfe8/third_party/WebKit/Source/platform/heap/HeapTest.cpp
[modify] https://crrev.com/590a7c6ceb9bea92bbce8e7636406ed6ea39bfe8/third_party/WebKit/Source/platform/heap/ThreadState.cpp
[modify] https://crrev.com/590a7c6ceb9bea92bbce8e7636406ed6ea39bfe8/third_party/WebKit/Source/platform/heap/ThreadState.h
Project Member

Comment 8 by bugdroid1@chromium.org, Feb 16 2017 

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/3d80b9d4de29bc69458e1c27eb8bce181d7dcb27

commit 3d80b9d4de29bc69458e1c27eb8bce181d7dcb27
Author: keishi <keishi@chromium.org>
Date: Thu Feb 16 09:50:46 2017

Revert of Move postGC out of CrossThreadPersistentRegion::LockScope (patchset #2 id:20001 of https://codereview.chromium.org/2695063005/ )

Reason for revert:
Broke build.
crbug.com/692982

Original issue's description:
> Move postGC out of CrossThreadPersistentRegion::LockScope
>
> Some finalizers/prefinalizers acquire locks, causing a dead lock when CrossThreadPersistent tries to acquire the CrossThreadPersistentRegion lock.
>
> This CL moves the postGC step outside of the CrossThreadPersistentRegion::LockScope. This should be okay as the CrossThreadPersistentRegion::LockScope only exists to avoid add/deletetion of CrossThreadPersistentNodes during marking.
>
> BUG=681527, 684856 , 685624 , 685624 
>
> Review-Url: https://codereview.chromium.org/2686533003
> Cr-Commit-Position: refs/heads/master@{#450251}
> (cherry picked from commit 52b19c8f68c222c08944af937a01f3ff7e46705c)
>
> Review-Url: https://codereview.chromium.org/2695063005 .
> Cr-Commit-Position: refs/branch-heads/2987@{#539}
> Cr-Branched-From: ad51088c0e8776e8dcd963dbe752c4035ba6dab6-refs/heads/master@{#444943}
> Committed: https://chromium.googlesource.com/chromium/src/+/590a7c6ceb9bea92bbce8e7636406ed6ea39bfe8

TBR=
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=681527, 684856 , 685624 , 685624 

Review-Url: https://codereview.chromium.org/2701513003
Cr-Commit-Position: refs/branch-heads/2987@{#541}
Cr-Branched-From: ad51088c0e8776e8dcd963dbe752c4035ba6dab6-refs/heads/master@{#444943}

[modify] https://crrev.com/3d80b9d4de29bc69458e1c27eb8bce181d7dcb27/third_party/WebKit/Source/platform/heap/Heap.cpp
[modify] https://crrev.com/3d80b9d4de29bc69458e1c27eb8bce181d7dcb27/third_party/WebKit/Source/platform/heap/Heap.h
[modify] https://crrev.com/3d80b9d4de29bc69458e1c27eb8bce181d7dcb27/third_party/WebKit/Source/platform/heap/HeapTest.cpp
[modify] https://crrev.com/3d80b9d4de29bc69458e1c27eb8bce181d7dcb27/third_party/WebKit/Source/platform/heap/ThreadState.cpp
[modify] https://crrev.com/3d80b9d4de29bc69458e1c27eb8bce181d7dcb27/third_party/WebKit/Source/platform/heap/ThreadState.h
Project Member

Comment 9 by bugdroid1@chromium.org, Feb 16 2017 

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/04062423c93acc7bcd6d313c480d3f29bbf382e2

commit 04062423c93acc7bcd6d313c480d3f29bbf382e2
Author: Keishi Hattori <keishi@chromium.org>
Date: Thu Feb 16 11:09:42 2017

Move postGC out of CrossThreadPersistentRegion::LockScope

Some finalizers/prefinalizers acquire locks, causing a dead lock when CrossThreadPersistent tries to acquire the CrossThreadPersistentRegion lock.

This CL moves the postGC step outside of the CrossThreadPersistentRegion::LockScope. This should be okay as the CrossThreadPersistentRegion::LockScope only exists to avoid add/deletetion of CrossThreadPersistentNodes during marking.

BUG= 684856 , 685624 , 685624 

Review-Url: https://codereview.chromium.org/2686533003
Cr-Commit-Position: refs/heads/master@{#450251}
(cherry picked from commit 52b19c8f68c222c08944af937a01f3ff7e46705c)

Review-Url: https://codereview.chromium.org/2690943009 .
Cr-Commit-Position: refs/branch-heads/2987@{#543}
Cr-Branched-From: ad51088c0e8776e8dcd963dbe752c4035ba6dab6-refs/heads/master@{#444943}

[modify] https://crrev.com/04062423c93acc7bcd6d313c480d3f29bbf382e2/third_party/WebKit/Source/platform/heap/Heap.cpp
[modify] https://crrev.com/04062423c93acc7bcd6d313c480d3f29bbf382e2/third_party/WebKit/Source/platform/heap/Heap.h
[modify] https://crrev.com/04062423c93acc7bcd6d313c480d3f29bbf382e2/third_party/WebKit/Source/platform/heap/HeapTest.cpp
[modify] https://crrev.com/04062423c93acc7bcd6d313c480d3f29bbf382e2/third_party/WebKit/Source/platform/heap/ThreadState.cpp
[modify] https://crrev.com/04062423c93acc7bcd6d313c480d3f29bbf382e2/third_party/WebKit/Source/platform/heap/ThreadState.h
►

Sign in to add a comment