Closed Bug 1838695 Opened 2 years ago Closed 1 year ago

Frequent mozilla/tests/webgpu/chunked/<number>/cts.https.html | single tracking bug

Categories

(Core :: Graphics: WebGPU, defect, P2)

defect

Tracking

()

RESOLVED FIXED
Tracking Status
firefox-esr102 --- unaffected
firefox-esr115 --- unaffected
firefox114 --- unaffected
firefox115 --- unaffected
firefox116 --- wontfix
firefox117 --- wontfix
firefox118 --- wontfix
firefox119 --- wontfix

People

(Reporter: intermittent-bug-filer, Assigned: nical)

References

(Depends on 1 open bug, Regression)

Details

(Keywords: intermittent-failure, intermittent-testcase, regression, Whiteboard: [stockwell unknown])

Attachments

(1 file)

Filed by: smolnar [at] mozilla.com
Parsed log: https://treeherder.mozilla.org/logviewer?job_id=419445188&repo=autoland
Full log: https://firefox-ci-tc.services.mozilla.com/api/queue/v1/task/DXq8di7DR9yjfjxQDhmuPw/runs/0/artifacts/public/logs/live_backing.log


TEST-START | /_mozilla/webgpu/chunked/14/cts.https.html?q=webgpu:api,validation,encoding,cmds,clearBuffer:buffer,device_mismatch:*
[task 2023-06-15T16:38:41.639Z] 16:38:41     INFO - Closing window add187e0-e897-483c-a6b6-55407ccb2da5
[task 2023-06-15T16:39:06.634Z] 16:39:06     INFO - Got timeout in harness
[task 2023-06-15T16:39:06.645Z] 16:39:06     INFO - TEST-UNEXPECTED-TIMEOUT | /_mozilla/webgpu/chunked/14/cts.https.html?q=webgpu:api,validation,encoding,cmds,clearBuffer:buffer,device_mismatch:* | TestRunner hit external timeout (this may indicate a hang)
[task 2023-06-15T16:39:06.645Z] 16:39:06     INFO - TEST-INFO took 25024ms
[task 2023-06-15T16:40:17.988Z] 16:40:17     INFO - Browser exited with return code 572
[task 2023-06-15T16:40:28.002Z] 16:40:28  WARNING - Forcibly terminating runner process
[task 2023-06-15T16:40:28.050Z] 16:40:28     INFO - Application command: Z:\task_168684417078243\build\application\firefox\firefox.exe -marionette about:blank --wait-for-browser -profile C:\Users\task_168684417078243\AppData\Local\Temp\tmp03432srs
[task 2023-06-15T16:40:28.069Z] 16:40:28     INFO - PID 2140 | 1686847105357	Marionette	INFO	Marionette enabled
[task 2023-06-15T16:40:28.070Z] 16:40:28     INFO - PID 2140 | 1686847105441	Marionette	INFO	Listening on port 56440
[task 2023-06-15T16:40:28.071Z] 16:40:28     INFO - PID 2140 | JavaScript error: resource://gre/modules/XULStore.sys.mjs, line 60: Error: Can't find profile directory.
[task 2023-06-15T16:40:28.072Z] 16:40:28     INFO - PID 2140 | console.error: (new Error("Polling for changes failed: Unexpected content-type \"text/plain;charset=US-ASCII\".", "resource://services-settings/remote-settings.sys.mjs", 324))
[task 2023-06-15T16:40:28.073Z] 16:40:28     INFO - PID 2140 | console.error: (new Error("Unexpected content-type \"text/plain;charset=US-ASCII\"", "resource://services-settings/Utils.sys.mjs", 399))
[task 2023-06-15T16:40:28.073Z] 16:40:28     INFO - PID 2140 | console.error: (new Error("Unexpected content-type \"text/plain;charset=US-ASCII\"", "resource://services-settings/Utils.sys.mjs", 399))
[task 2023-06-15T16:40:28.074Z] 16:40:28     INFO - PID 2140 | console.error: (new Error("Unexpected content-type \"text/plain;charset=US-ASCII\"", "resource://services-settings/Utils.sys.mjs", 399))
[task 2023-06-15T16:40:28.075Z] 16:40:28     INFO - PID 2140 | console.error: (new Error("Unexpected content-type \"text/plain;charset=US-ASCII\"", "resource://services-settings/Utils.sys.mjs", 399))
[task 2023-06-15T16:40:28.076Z] 16:40:28     INFO - Starting runner
[task 2023-06-15T16:40:29.116Z] 16:40:29     INFO - Installing extension from Z:\task_168684417078243\build\tests\extensions\specialpowers@mozilla.org.xpi
[task 2023-06-15T16:40:29.352Z] 16:40:29     INFO - TEST-START | /_mozilla/webgpu/chunked/14/cts.https.html?q=webgpu:api,validation,debugMarker:push_pop_call_count_unbalance,command_encoder:*
Summary: Intermittent mozilla/tests/webgpu/chunked/14/cts.https.html | single tracking bug → Intermittent mozilla/tests/webgpu/chunked/<number>/cts.https.html | single tracking bug
Duplicate of this bug: 1839963
Duplicate of this bug: 1839962
Severity: S4 → --
Flags: needinfo?(egubler)
Keywords: regression
Priority: P5 → --
Regressed by: 1839768
Summary: Intermittent mozilla/tests/webgpu/chunked/<number>/cts.https.html | single tracking bug → Frequent mozilla/tests/webgpu/chunked/<number>/cts.https.html | single tracking bug
Regressed by: webgpu-v1-cts-windows
No longer regressed by: 1839768

Set release status flags based on info from the regressing bug 1839768

:CosmimS: I intend to look at this as part of my immediate work pipeline in bug 1836479. 👍🏻

Flags: needinfo?(egubler)
Assignee: nobody → egubler
Severity: -- → S3
Status: NEW → ASSIGNED
Priority: -- → P1

Set release status flags based on info from the regressing bug 1831263

In the past week we had 115 occurrences of this failures on:

  • windows11-32-2009-qr opt and debug
  • windows11-32-2009-shippable-qr opt
  • windows11-64-2009-qr opt and debug
  • windows11-64-2009-shippable-qr opt
    Recent failure log
[task 2023-07-07T10:45:45.586Z] 10:45:45     INFO - Closing window 80e89224-c3b2-418a-9709-ea2310c5af01
[task 2023-07-07T10:45:46.525Z] 10:45:46     INFO - PID 8728 | [GFX1-[]G: CompositorBridgFXeChil1-]: Cod mpositorBrreceidives geCh[GF[IPCiGldX[F GcFl1o[-XG]Xs11F-:X-e  1 wi]tr:]Ch  reaecC-]s::oe oomp oCCmopnossimipiotoves Ismo=PAiCtporBr crobntoBrilrismaidgodtgleSorBCohsuet drhwithi BrreowreaCslndi odngeh
[task 2023-07-07T10:45:46.526Z] 10:45:46     INFO - PID 8728 | ildire drCchgi=eldi erCevchAieeld irbvneosre ceecs eIImPiiCvPvC  ees cIslac PloIlPCCS houtsd oesccw n
[task 2023-07-07T10:45:46.527Z] 10:45:46     INFO - PID 8728 | lloeswit h ree woitwsaithsohn=e wit Area sonrbenasoh rno==reAabmaslShonn=AuotAbrnombalnrodmarlShSomahuutwlSnh
[task 2023-07-07T10:45:46.527Z] 10:45:46     INFO - PID 8728 | uttddownow
[task 2023-07-07T10:45:46.528Z] 10:45:46     INFO - PID 8728 | n
[task 2023-07-07T10:45:46.528Z] 10:45:46     INFO - PID 8728 | down
[task 2023-07-07T10:45:46.551Z] 10:45:46     INFO - PID 8728 | [GFX1-]: CompositorBridgeChild receives IPC close with reason=AbnormalShutdown
[task 2023-07-07T10:45:46.555Z] 10:45:46     INFO - 
[task 2023-07-07T10:45:46.555Z] 10:45:46     INFO - TEST-UNEXPECTED-FAIL | /_mozilla/webgpu/chunked/12/cts.https.html?q=webgpu:api,validation,createBindGroup:buffer,usage:* | :type="uniform" - assert_unreached: 
[task 2023-07-07T10:45:46.555Z] 10:45:46     INFO -   - EXCEPTION: WebGPU device failed to initialize with AbortError "Internal communication error!"; not retrying
[task 2023-07-07T10:45:46.555Z] 10:45:46     INFO -     assert@https://web-platform.test:8443/_mozilla/webgpu/common/util/util.js:37:11
[task 2023-07-07T10:45:46.555Z] 10:45:46     INFO -     acquire@https://web-platform.test:8443/_mozilla/webgpu/webgpu/util/device_pool.js:36:11
[task 2023-07-07T10:45:46.555Z] 10:45:46     INFO -     
[task 2023-07-07T10:45:46.555Z] 10:45:46     INFO -  Reached unreachable code
[task 2023-07-07T10:45:46.555Z] 10:45:46     INFO - wpt_fn@https://web-platform.test:8443/_mozilla/webgpu/common/runtime/wpt.js:65:25```
Whiteboard: [stockwell disable-recommended] → [stockwell needswork:owner]

Erich, is there an ETA to resolve these frequent test failures?

We're looking into this. These intermittents significantly reduce the value of the CTS to us, so it's a priority. But we do not have a diagnosis, so we don't have an ETA.

Since the intermittents aren't isolated to any particular WebGPU CTS tests, as far as we know, it's difficult to identify some subset of tests we could disable to quiet things down. If we still don't have a diagnosis after a few days, we can talk about broader measures to keep this from wasting people's time until we can get it fixed.

Depends on: 1843021
Depends on: 1843250
No longer depends on: 1843021
See Also: → 1843021
No longer depends on: 1843250
See Also: → 1843250

There were a significant number of failures associated with this bug that are because of this. I'm hoping that the number of failures here goes down significantly after fixing it. 🤞🏻

See Also: → 1844636

Update

There have been 207 total failures within the last 7 days:

  • 96 failures on Windows 11 x86 22H2 WebRender opt
  • 63 failures on Windows 11 x86 22H2 WebRender Shippable opt
  • 36 failures on Windows 11 x64 22H2 WebRender opt
  • 12 failures on Windows 11 x64 22H2 WebRender Shippable opt

Recent failure log: https://treeherder.mozilla.org/logviewer?job_id=424073238&repo=autoland&lineNumber=1758

Depends on: 1845857

It seems that the highest number of intermittent failures are on 32-bit builds, by a wide margin. I think the next priority for now is going to be tackling those.

Priority: P1 → P2
Assignee: egubler → nical.bugzilla

This failure turned into a perma failure and there are 269 failures in the last 7 days.
Nicolas, can you please take a look over ?

Flags: needinfo?(nical.bugzilla)
Summary: Frequent mozilla/tests/webgpu/chunked/<number>/cts.https.html | single tracking bug → Perma mozilla/tests/webgpu/chunked/<number>/cts.https.html | single tracking bug

:NarcisB: Could you elaborate on what but signatures turned perma? This bug aggregates numerous signatures, and part of our work has been to isolate them into separate well-understood bugs.

:ErichDonGubler, I did a few backfills & retriggers and all of them failed

This seems to be permafailing only with this failure line TEST-UNEXPECTED-FAIL | /_mozilla/webgpu/chunked/1/cts.https.html?q=webgpu:api,operation,adapter,requestDevice:limit,worse_than_default:* | :limit="maxBufferSize" - assert_unreached:
Here are some examples that show that this is not permafailing on all webgpu jobs or other signatures:

Snippet from permafailing job:

[task 2023-08-07T21:20:44.729Z] 21:20:44     INFO - TEST-START | /_mozilla/webgpu/chunked/1/cts.https.html?q=webgpu:api,operation,adapter,requestDevice:limit,worse_than_default:*
[task 2023-08-07T21:20:44.734Z] 21:20:44     INFO - Closing window 512136e6-ed72-4bd1-8c7d-4fe09779f824
[task 2023-08-07T21:20:51.636Z] 21:20:51     INFO - .......
[task 2023-08-07T21:20:51.636Z] 21:20:51     INFO - TEST-PASS | /_mozilla/webgpu/chunked/1/cts.https.html?q=webgpu:api,operation,adapter,requestDevice:limit,worse_than_default:* | :limit="maxSampledTexturesPerShaderStage" 
[task 2023-08-07T21:20:51.636Z] 21:20:51     INFO - TEST-PASS | /_mozilla/webgpu/chunked/1/cts.https.html?q=webgpu:api,operation,adapter,requestDevice:limit,worse_than_default:* | :limit="maxSamplersPerShaderStage" 
[task 2023-08-07T21:20:51.637Z] 21:20:51     INFO - TEST-PASS | /_mozilla/webgpu/chunked/1/cts.https.html?q=webgpu:api,operation,adapter,requestDevice:limit,worse_than_default:* | :limit="maxStorageBuffersPerShaderStage" 
[task 2023-08-07T21:20:51.637Z] 21:20:51     INFO - TEST-PASS | /_mozilla/webgpu/chunked/1/cts.https.html?q=webgpu:api,operation,adapter,requestDevice:limit,worse_than_default:* | :limit="maxStorageTexturesPerShaderStage" 
[task 2023-08-07T21:20:51.637Z] 21:20:51     INFO - TEST-PASS | /_mozilla/webgpu/chunked/1/cts.https.html?q=webgpu:api,operation,adapter,requestDevice:limit,worse_than_default:* | :limit="maxUniformBuffersPerShaderStage" 
[task 2023-08-07T21:20:51.637Z] 21:20:51     INFO - TEST-PASS | /_mozilla/webgpu/chunked/1/cts.https.html?q=webgpu:api,operation,adapter,requestDevice:limit,worse_than_default:* | :limit="maxUniformBufferBindingSize" 
[task 2023-08-07T21:20:51.637Z] 21:20:51     INFO - TEST-PASS | /_mozilla/webgpu/chunked/1/cts.https.html?q=webgpu:api,operation,adapter,requestDevice:limit,worse_than_default:* | :limit="maxStorageBufferBindingSize" 
[task 2023-08-07T21:20:51.637Z] 21:20:51     INFO - TEST-PASS | /_mozilla/webgpu/chunked/1/cts.https.html?q=webgpu:api,operation,adapter,requestDevice:limit,worse_than_default:* | :limit="minUniformBufferOffsetAlignment" 
[task 2023-08-07T21:20:51.637Z] 21:20:51     INFO - TEST-PASS | /_mozilla/webgpu/chunked/1/cts.https.html?q=webgpu:api,operation,adapter,requestDevice:limit,worse_than_default:* | :limit="minStorageBufferOffsetAlignment" 
[task 2023-08-07T21:20:51.637Z] 21:20:51     INFO - TEST-PASS | /_mozilla/webgpu/chunked/1/cts.https.html?q=webgpu:api,operation,adapter,requestDevice:limit,worse_than_default:* | :limit="maxVertexBuffers" 
[task 2023-08-07T21:20:51.637Z] 21:20:51     INFO - TEST-UNEXPECTED-FAIL | /_mozilla/webgpu/chunked/1/cts.https.html?q=webgpu:api,operation,adapter,requestDevice:limit,worse_than_default:* | :limit="maxBufferSize" - assert_unreached: 
[task 2023-08-07T21:20:51.637Z] 21:20:51     INFO -   - EXCEPTION: requestDevice: Request for limit 'maxBufferSize' must be <= supported 0, was 268435455.
[task 2023-08-07T21:20:51.637Z] 21:20:51     INFO -     subcase: mul=1;add=-1
[task 2023-08-07T21:20:51.637Z] 21:20:51     INFO -     @https://web-platform.test:8443/_mozilla/webgpu/webgpu/api/operation/adapter/requestDevice.spec.js:263:36
[task 2023-08-07T21:20:51.637Z] 21:20:51     INFO -   - INFO: subcase: mul=1;add=-1
[task 2023-08-07T21:20:51.637Z] 21:20:51     INFO -     OK
[task 2023-08-07T21:20:51.637Z] 21:20:51     INFO -   - EXCEPTION: requestDevice: Request for limit 'maxBufferSize' must be <= supported 0, was 268435356.
[task 2023-08-07T21:20:51.637Z] 21:20:51     INFO -     subcase: mul=1;add=-100
[task 2023-08-07T21:20:51.637Z] 21:20:51     INFO -     @https://web-platform.test:8443/_mozilla/webgpu/webgpu/api/operation/adapter/requestDevice.spec.js:263:36
[task 2023-08-07T21:20:51.638Z] 21:20:51     INFO -   - INFO: subcase: mul=1;add=-100
[task 2023-08-07T21:20:51.638Z] 21:20:51     INFO -     OK
[task 2023-08-07T21:20:51.638Z] 21:20:51     INFO -  Reached unreachable code
[task 2023-08-07T21:20:51.638Z] 21:20:51     INFO - wpt_fn@https://web-platform.test:8443/_mozilla/webgpu/common/runtime/wpt.js:65:25
[task 2023-08-07T21:20:51.643Z] 21:20:51     INFO - ...........
[task 2023-08-07T21:20:51.643Z] 21:20:51     INFO - TEST-OK | /_mozilla/webgpu/chunked/1/cts.https.html?q=webgpu:api,operation,adapter,requestDevice:limit,worse_than_default:* | took 6909ms
[task 2023-08-07T21:20:51.645Z] 21:20:51     INFO - PID 2172 | 1691443251640	Marionette	INFO	Stopped listening on port 53094
[task 2023-08-07T21:20:52.327Z] 21:20:52     INFO - PID 2172 | console.error: ({})
[task 2023-08-07T21:20:52.636Z] 21:20:52     INFO - Browser exited with return code 0
[task 2023-08-07T21:20:52.637Z] 21:20:52     INFO - Closing logging queue
[task 2023-08-07T21:20:52.637Z] 21:20:52     INFO - queue closed
[task 2023-08-07T21:20:52.841Z] 21:20:52     INFO - Application command: Z:\task_169144156238687\build\application\firefox\firefox.exe -marionette about:blank --wait-for-browser -profile C:\Users\task_169144156238687\AppData\Local\Temp\tmpxza982he
[task 2023-08-07T21:20:52.852Z] 21:20:52     INFO - PID 3652 | 1691443213094	Marionette	INFO	Marionette enabled
[task 2023-08-07T21:20:52.853Z] 21:20:52     INFO - PID 3652 | 1691443213190	Marionette	INFO	Listening on port 53095
[task 2023-08-07T21:20:52.853Z] 21:20:52     INFO - PID 3652 | JavaScript error: resource://gre/modules/XULStore.sys.mjs, line 60: Error: Can't find profile directory.
[task 2023-08-07T21:20:52.854Z] 21:20:52     INFO - PID 3652 | console.error: (new Error("Unexpected content-type \"text/plain;charset=US-ASCII\"", "resource://services-settings/Utils.sys.mjs", 399))
[task 2023-08-07T21:20:52.855Z] 21:20:52     INFO - PID 3652 | console.error: (new Error("Unexpected content-type \"text/plain;charset=US-ASCII\"", "resource://services-settings/Utils.sys.mjs", 399))
[task 2023-08-07T21:20:52.855Z] 21:20:52     INFO - PID 3652 | console.error: (new Error("Unexpected content-type \"text/plain;charset=US-ASCII\"", "resource://services-settings/Utils.sys.mjs", 399))
[task 2023-08-07T21:20:52.856Z] 21:20:52     INFO - PID 3652 | console.error: (new Error("Unexpected content-type \"text/plain;charset=US-ASCII\"", "resource://services-settings/Utils.sys.mjs", 399))
[task 2023-08-07T21:20:52.857Z] 21:20:52     INFO - PID 3652 | console.error: (new Error("Unexpected content-type \"text/plain;charset=US-ASCII\"", "resource://services-settings/Utils.sys.mjs", 399))
[task 2023-08-07T21:20:52.857Z] 21:20:52     INFO - PID 3652 | console.error: (new Error("Polling for changes failed: Unexpected content-type \"text/plain;charset=US-ASCII\".", "resource://services-settings/remote-settings.sys.mjs", 321))
[task 2023-08-07T21:20:52.857Z] 21:20:52     INFO - Starting runner
[task 2023-08-07T21:20:54.217Z] 21:20:54     INFO - Installing extension from Z:\task_169144156238687\build\tests\extensions\specialpowers@mozilla.org.xpi
[task 2023-08-07T21:20:54.473Z] 21:20:54     INFO - TEST-START | /_mozilla/webgpu/chunked/1/cts.https.html?q=webgpu:api,operation,buffers,map:mapAsync,write:*

Hi Erich! If you want to take a look at the details I've mentioned above, to get a more clearer picture.
And, should we file a separate bug for the permafailure?

Flags: needinfo?(egubler)
Flags: needinfo?(nical.bugzilla)
See Also: → 1847771

Filed Bug 1847771 to be easier to track the permafailure and not mess up the orange factor on this bug. I've managed to find what caused this.

Summary: Perma mozilla/tests/webgpu/chunked/<number>/cts.https.html | single tracking bug → Frequent mozilla/tests/webgpu/chunked/<number>/cts.https.html | single tracking bug
Flags: needinfo?(egubler)
See Also: → 1847724

Update

There have been 164 failures within the last 7 days:

  • 46 failures on Windows 11 x86 22H2 WebRender opt/debug
  • 33 failures on Windows 11 x86 22H2 WebRender Shippable opt
  • 50 failures on Windows 11 x64 22H2 WebRender debug
  • 35 failures on Windows 11 x64 22H2 WebRender Shippable opt

Recent failure log: https://treeherder.mozilla.org/logviewer?job_id=427155783&repo=mozilla-central&lineNumber=7277

Whiteboard: [stockwell disable-recommended] → [stockwell disable-recommended][stockwell needswork:owner]

Update:
There have been 626 failures within the last 7 days:

  • 207 failures on linux2204-64-wayland
  • 206 failures on windows11-64-2009
  • 205 failures on windows11-32-2009
  • 8 failures on linux1804-64-shippable-qr

Recent failure log: https://treeherder.mozilla.org/logviewer?job_id=429101013&repo=mozilla-central&lineNumber=2267

Eric, can you please take a look?

Flags: needinfo?(egubler)

These tests fail intermittently or permanently. The plan is to fix them in a
burndown. After they all pass, the tier can be changed to 2 or 1, and their
results will be monitored (sheriffed).

Status: ASSIGNED → RESOLVED
Closed: 2 years ago
Resolution: --- → FIXED
Target Milestone: --- → 119 Branch

Since nightly and release are affected, beta will likely be affected too.
For more information, please visit BugBot documentation.

Joel, some of the tests still run as part of the backlog task. How can the tests run as part of the tier 3 task and no webgpu tests run as part of the backlog task?

Status: RESOLVED → REOPENED
Flags: needinfo?(jmaher)
Resolution: FIXED → ---
Target Milestone: 119 Branch → ---

:NarcisB, :aryx, :jmaher: I have been attending to this for the last day or so after being on PTO last week, sorry for not updating the NI here.

The main problem we have with the backlog task is the issue I mention in comment 5 on bug 1852952 and comment 4 on bug 1852858; the webgpu-platform-tests-backlog task still runs canvas and webgpu tasks. These tasks can be sensitive to driver and hardware environment, which is why web-platform-test-webgpu was broken out in the first place (CC :jmaher) with their own runner configuration and set of supported platforms. In those comments I also mention patch D188387, which I believe should fix this issue.

I'm sad to downgrade to tier 2, but I have no objection (CC :jimb). I apologize that managing these failures on tier 2 has been enough work to warrant the downgrade.

Flags: needinfo?(egubler)
Flags: needinfo?(jmaher)

I have a patch up to fix the backlog tests ( https://phabricator.services.mozilla.com/D188413 ), but no real traction yet, let me nudge on that.

you are welcome to steal some of the backlog changes in web-platform.yml and I can remove them from my patch.

Whiteboard: [stockwell disable-recommended][stockwell needswork:owner] → [stockwell needswork:owner]
Whiteboard: [stockwell disable-recommended]
Whiteboard: [stockwell disable-recommended] → [stockwell needswork:owner]

Nicolas, could you please write a patch that would stop running mozilla/tests/webgpu/chunked/** tests as part of the backlog? There was one is https://phabricator.services.mozilla.com/D188387 but got abandoned. From what I gather from Erich Gubler these should not be running.
These are the jobs on mozilla-central that all have permafailures on tests from /_mozilla/webgpu/chunked/. Bugs filed because of these failures are: Bug 1852952, 1859202, 1859198, 1859178, 1847771, 1852858.

Flags: needinfo?(nical.bugzilla)
Duplicate of this bug: 1859202
Duplicate of this bug: 1859198
Duplicate of this bug: 1859178
Duplicate of this bug: 1859203

:CosminS: This bug is for general issues discovered in WebGPU CTS tests while executing in the expected job(s) and environment(s). AFAIK, the running of WebGPU CTS in WPT backlog tests is a recent regression in tests that the WebGPU team does not own. Therefore, the issues you've pinged :nical on are not relevant to this bug. Could we please combine those duplicate issues into a separate bug?

Flags: needinfo?(nical.bugzilla)
No longer duplicate of this bug: 1859178
No longer duplicate of this bug: 1859198
No longer duplicate of this bug: 1859203
No longer duplicate of this bug: 1859202

Left only Bug 1859178 to track those failures. Being a permafail on several jobs it would be nice to have that patch what would stop running those tests on backlog as stated in https://bugzilla.mozilla.org/show_bug.cgi?id=1852952#c10.

RE: comment 123: The test-coverage CI tasks that yielded the above should be running in the same environment as current (tier 3) web-platform-tests-webgpu tests, but they do not. This should be fixed. NI'ing :nataliaCs to get clarity on who I need to contact WRT ownership of those tests.

I don't believe it's useful to keep this bug open, as we (the WebGPU team) changed WebGPU CTS testing to tier 3 specifically to eliminate this bug as an ineffectual hotspot of sheriff and team attention. I'm going to close it, and accept whatever dialogue is necessary to converge stakeholder understanding on my decision post-hoc.

Status: REOPENED → RESOLVED
Closed: 2 years ago1 year ago
Flags: needinfo?(ncsoregi)
Resolution: --- → INCOMPLETE
Status: REOPENED → RESOLVED
Closed: 1 year ago1 year ago
Resolution: --- → INCOMPLETE
Status: RESOLVED → REOPENED
Resolution: INCOMPLETE → ---

Erich, I think is best to close this as worksforme or fixed so that the failures bot won't reopen it.

NI'ing :nataliaCs to get clarity on who I need to contact WRT ownership of those tests.

Joel, James, can we single out these TCw jobs that run the webgpu tests to be tier 3?

Status: REOPENED → RESOLVED
Closed: 1 year ago1 year ago
Flags: needinfo?(ncsoregi)
Flags: needinfo?(jmaher)
Flags: needinfo?(james)
Resolution: --- → FIXED

we could add --exclude-tag=webgpu here:
https://searchfox.org/mozilla-central/source/taskcluster/ci/test/web-platform.yml#589

we might want to consider as well:
--exclude-tag=canvas
--exclude-tag=privatebrowsing

:jgraham, do you have other thoughts?

Flags: needinfo?(jmaher)

I don't really know what that coverage job is used for (or what the outputs are used for), so I don't really know the impact of disabling specific tests. Assuming the job is useful at all it makes some sense that it would run all tests, so I don't see why we'd want to exclude canvas or privatebrowsing. Maybe webgpu is different if those tests really don't run at all without hardware support.

Flags: needinfo?(james)
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Created:
Updated:
Size: