Skip to content

Linux arm32 GCStress=3 timeout test failures #2231

Open
@BruceForstall

Description

@BruceForstall

The following tests time out in GCStress=3 test runs (after an hour?).

Interop/DllImportAttribute/DllImportPath/DllImportPathTest/DllImportPathTest.sh
baseservices/threading/interlocked/compareexchange/CompareExchangeTString/CompareExchangeTString.sh
JIT/HardwareIntrinsics/General/Vector256_1/Vector256_1_ro/Vector256_1_ro.sh
JIT/HardwareIntrinsics/General/Vector128_1/Vector128_1_ro/Vector128_1_ro.sh
JIT/Methodical/tailcall/_il_dbgreference_i/_il_dbgreference_i.sh
baseservices/threading/regressions/beta2/437044/437044.sh
JIT/Generics/Exceptions/general_class_static01/general_class_static01.sh
JIT/HardwareIntrinsics/General/Vector128_1/Vector128_1_r/Vector128_1_r.sh
JIT/HardwareIntrinsics/General/Vector256_1/Vector256_1_r/Vector256_1_r.sh
JIT/Generics/Exceptions/general_struct_instance01/general_struct_instance01.sh
JIT/Generics/Exceptions/general_struct_static01/general_struct_static01.sh
JIT/Generics/Exceptions/general_class_instance01/general_class_instance01.sh

https://dev.azure.com/dnceng/public/_build/results?buildId=497294&view=ms.vss-test-web.build-test-results-tab

So are they really that slow? Are the machines overloaded or underpowered? Is there a GCStress=3 bug, or a VM bug exposed by GCStress=3 leading to this?

Should these be disabled for GCStress=3 until such time as they can be investigated?

Metadata

Metadata

Assignees

Type

No type

Projects

No projects

Relationships

None yet

Development

No branches or pull requests

Issue actions