[wasm] disabled - `JIT.Performance` taking more than 30mins, and timing out

[radical]

JIT.Performance failing:

[16:58:21] info: Using js engine V8 from path /home/helixbot/.jsvu/bin/v8
[16:58:21] info: V8 version: 11.5.151

...

[17:27:09] info: 17:27:09.145 Passed test: JIT/Performance/CodeQuality/Benchstones/MDBenchI/MDNDhrystone/MDNDhrystone/MDNDhrystone.dll
[17:27:09] info: 17:27:09.147 Running test: JIT/Performance/CodeQuality/Benchstones/MDBenchI/MDPuzzle/MDPuzzle/MDPuzzle.dll
[17:28:21] info: Process 44 didn't exit within 00:30:00 and will be killed

[17:28:21] info: Killing process tree of 44...

[17:28:21] fail: Tests timed out after 1800secs
[17:28:21] info: Pids to kill: 44, 45

[17:28:21] info: Running lldb diagnostics for pid 44

[17:28:21] info: Printing backtrace for pid=44

[17:28:21] info: lldb was not found, skipping diagnosis..

[17:28:21] info: Running lldb diagnostics for pid 45

[17:28:21] info: Printing backtrace for pid=45

[17:28:21] info: lldb was not found, skipping diagnosis..

[17:28:21] info: 44 Execution timed out after 1800 seconds and the process was killed.

[17:28:21] info: Process v8 exited with 137

XHarness exit code: 70 (TIMED_OUT)

https://github.com/dotnet/runtime/pull/86747 re-enables runtime tests that have not been running for almost a month. Enabling them hit this issue.

cc @radekdoulik

[ghost]

Tagging subscribers to 'arch-wasm': @lewing See info in area-owners.md if you want to be subscribed.

Issue Details

---

[JIT.Performance failing](https://helixre107v0xdcypoyl9e7f.blob.core.windows.net/dotnet-runtime-refs-pull-86747-merge-871e237cc3ae4a1ea2/JIT.performance/1/console.46ebec28.log?helixlogtype=result): ``` [17:27:09] info: 17:27:09.145 Passed test: JIT/Performance/CodeQuality/Benchstones/MDBenchI/MDNDhrystone/MDNDhrystone/MDNDhrystone.dll [17:27:09] info: 17:27:09.147 Running test: JIT/Performance/CodeQuality/Benchstones/MDBenchI/MDPuzzle/MDPuzzle/MDPuzzle.dll [17:28:21] info: Process 44 didn't exit within 00:30:00 and will be killed [17:28:21] info: Killing process tree of 44... [17:28:21] fail: Tests timed out after 1800secs [17:28:21] info: Pids to kill: 44, 45 [17:28:21] info: Running lldb diagnostics for pid 44 [17:28:21] info: Printing backtrace for pid=44 [17:28:21] info: lldb was not found, skipping diagnosis.. [17:28:21] info: Running lldb diagnostics for pid 45 [17:28:21] info: Printing backtrace for pid=45 [17:28:21] info: lldb was not found, skipping diagnosis.. [17:28:21] info: 44 Execution timed out after 1800 seconds and the process was killed. [17:28:21] info: Process v8 exited with 137 XHarness exit code: 70 (TIMED_OUT) ``` https://github.com/dotnet/runtime/pull/86747 re-enables runtime tests that have not been running for almost a month. Enabling them hit this issue. cc @radekdoulik

Author:	radical
Assignees:	-
Labels:	`arch-wasm`, `area-VM-meta-mono`, `test-failure`
Milestone:	-

[radical]

cc @kg

[radical]

The log shows:

+ export __TestTimeout=600000
+ export __TestCollectionTimeoutMins=200

So, maybe just one test is running too long?

[kg]

From the timestamps in the log it looks like a lot of these tests just take a while to run. How do I run this suite? In the past when I tried to run it, it didn't work.

[radical]

JIT.Performance is completely disabled now with https://github.com/dotnet/runtime/pull/86747 . We can try to run this with an increased timeout.

[kg]

We may just want to reduce the iteration count on all of the slow tests inside the suite, since we have dotnet/performance for actual ongoing benchmarks.

[lewing]

when did this start and was there a new test or a regression?

[kg]

Is there a reason why we need to do this for 8.0? These benchmarks all run in the perf infrastructure and are autofiled, iirc.

[lewing]

Only if there is a regression