1530033 - windows/aarch64 - disable build/tests/reftest/tests/layout/reftests/bugs/256180-1.html, 256180-2.html, 256180-3.html

Status: RESOLVED FIXED

(Firefox Build System :: General, defect)

Description

[Treeherder Bug Filer]

#[markdown(off)]
Filed by: egao [at] mozilla.com

https://treeherder.mozilla.org/logviewer.html#?job_id=229891968&repo=try

https://queue.taskcluster.net/v1/task/EuOcrzLaSr66GujjsQKNQw/runs/0/artifacts/public/logs/live_backing.log

https://hg.mozilla.org/mozilla-central/raw-file/tip/layout/tools/reftest/reftest-analyzer.xhtml#logurl=https://queue.taskcluster.net/v1/task/EuOcrzLaSr66GujjsQKNQw/runs/0/artifacts/public/logs/live_backing.log&only_show_unexpected=1

Raised during reftest-no-accel-e10s-2 on win10/aarch64.

Context:

14:47:20 INFO - REFTEST PROCESS-CRASH | file:///C:/tasks/task_1550846210/build/tests/reftest/tests/layout/reftests/bugs/256180-1.html | application crashed [@ static union core::option::Option<servo_arc::Arc<indexmap::map::IndexMap<style::gecko_string_cache::Atom, servo_arc::Arc<style::custom_properties::VariableValue>, core::hash::BuildHasherDefault<style::selector_map::PrecomputedHasher>>>> style::custom_properties::CustomPropertiesBuilder::build()]
14:47:20 INFO - Crash dump filename: c:\users\testdr~1\appdata\local\temp\tmp1u_n7p.mozrunner\minidumps\a7f7d801-397f-4fa6-93db-54a41cf13002.dmp
14:47:20 INFO - Operating system: Windows NT
14:47:20 INFO - 10.0.17134
14:47:20 INFO - CPU: 0x000c
14:47:20 INFO - 8 CPUs
14:47:20 INFO -
14:47:20 INFO - GPU: UNKNOWN
14:47:20 INFO -
14:47:20 INFO - Crash reason: EXCEPTION_STACK_OVERFLOW
14:47:20 INFO - Crash address: 0x7ffca0b59144
14:47:20 INFO - Assertion: Unknown assertion type 0x00000000
14:47:20 INFO - Process uptime: 6 seconds
14:47:20 INFO -
14:47:20 INFO - Thread 0 (crashed)
14:47:20 INFO - 0 xul.dll!static union core::option::Option<servo_arc::Arc<indexmap::map::IndexMap<style::gecko_string_cache::Atom, servo_arc::Arc<style::custom_properties::VariableValue>, core::hash::BuildHasherDefault<style::selector_map::PrecomputedHasher>>>> style::custom_properties::CustomPropertiesBuilder::build() [custom_properties.rs:e92ff56d2be21676b447c6fbb87b4c4479539bc9 : 610 + 0x10]
14:47:20 INFO -
14:47:20 INFO - Found by: given as instruction pointer in context
14:47:20 INFO - 1 xul.dll!static struct servo_arc::Arc<style::gecko_properties::ComputedValues> style::properties::cascade::cascade_rules<style::gecko::wrapper::GeckoElement>(struct style::gecko::media_queries::Device *, union core::option::Option<style::gecko::pseudo_element::PseudoElement*>, struct style::rule_tree::StrongRuleNode *, struct style::shared_lock::StylesheetGuards *, union core::option::Option<style::gecko_properties::ComputedValues*>, union core::option::Option<style::gecko_properties::ComputedValues*>, union core::option::Option<style::gecko_properties::ComputedValues*>, struct style::font_metrics::FontMetricsProvider*, union style::properties::cascade::CascadeMode, selectors::context::QuirksMode, union core::option::Option<style::rule_cache::RuleCache*>, struct style::rule_cache::RuleCacheConditions *, union core::option::Option<style::gecko::wrapper::GeckoElement>) [cascade.rs:e92ff56d2be21676b447c6fbb87b4c4479539bc9 : 170 + 0x408]
14:47:20 INFO -
14:47:20 INFO - Found by: previous frame's frame pointer
14:47:20 INFO -
14:47:20 INFO - Thread 1
14:47:20 INFO - 0 ntdll.dll!NtRemoveIoCompletion + 0x4
14:47:20 INFO -
14:47:20 INFO - Found by: given as instruction pointer in context
14:47:20 INFO - 1 KERNELBASE.dll + 0x17e9e8
14:47:20 INFO -
14:47:20 INFO - Found by: previous frame's frame pointer

<rest truncated for brevity>

Comment 2

[Edwin Takahashi (:egao | infrequent contributor)]

I have a sneaking suspicion that this may be due to lack of crashreporter on win10/aarch64.

:svoisen could you point me to the correct person to handle this issue? It does not appear to be failing in windows7/opt-Ru4, so I think this is an windows10-aarch64 issue.

Comment 3

[Sean Voisen (:svoisen)]

@egao hsivonen authored the test and fixed the related bug, so maybe he might know why this is crashing on aarch64?

Comment 4

[Henri Sivonen (:hsivonen)]

The test relies on Firefox having a non-default runtime stack size on Windows. https://searchfox.org/mozilla-central/rev/dbddac86aadf1d4871fb350bbe66db43728a9f81/config/config.mk#293 needs to be adjusted to set the stack size to 8 MB on aarch64 Windows, too.

Comment 5

[Henri Sivonen (:hsivonen)]

Moving over to build system, since this is build config problem.

Comment 6

[Intermittent Failures Robot]

2 failures in 2981 pushes (0.001 failures/push) were associated with this bug in the last 7 days.

Repository breakdown:

• try: 2

Platform breakdown:

• windows10-aarch64: 2

For more details, see:
https://treeherder.mozilla.org/intermittent-failures.html#/bugdetails?bug=1530033&startday=2019-02-25&endday=2019-03-03&tree=all

Comment 7

[Intermittent Failures Robot]

3 failures in 2818 pushes (0.001 failures/push) were associated with this bug in the last 7 days.

Repository breakdown:

• try: 3

Platform breakdown:

• windows10-aarch64: 3

For more details, see:
https://treeherder.mozilla.org/intermittent-failures.html#/bugdetails?bug=1530033&startday=2019-03-04&endday=2019-03-10&tree=all

Comment 8

[Edwin Takahashi (:egao | infrequent contributor)]

Attachment: Bug 1530033 - disable build/tests/reftest/tests/layout/reftests/bugs/256180-1.html, 256180-2.html, 256180-3.html

Comment 9

[Edwin Takahashi (:egao | infrequent contributor)]

Bug is being disabled with attached patch for a greener run on windows10-aarch64 platform.

Comment 10

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/26192d9f0bca
disable build/tests/reftest/tests/layout/reftests/bugs/256180-1.html, 256180-2.html, 256180-3.html r=jmaher

Comment 11

[Andrei Ciure[:aciure]]

https://hg.mozilla.org/mozilla-central/rev/26192d9f0bca

Comment 12

[Henri Sivonen (:hsivonen)]

Is there another bug on file for actually fixing this as described in comment 4?

Comment 13

[Edwin Takahashi (:egao | infrequent contributor)]

(In reply to Henri Sivonen (:hsivonen) from comment #12)

Is there another bug on file for actually fixing this as described in comment 4?

I used this bug to disable the tests for the time being, I could file a new bug to implement the fix if reusing this bug isn't preferable.

Comment 14

[Henri Sivonen (:hsivonen)]

(In reply to Edwin Gao (:egao) from comment #13)

(In reply to Henri Sivonen (:hsivonen) from comment #12)

Is there another bug on file for actually fixing this as described in comment 4?

I used this bug to disable the tests for the time being, I could file a new bug to implement the fix if reusing this bug isn't preferable.

Considering that this has been marked FIXED, it would be clearer to file a separate Bugzilla bug for fixing the actual build system bug and re-enabling the test.