workflows: update lava-test-plans to 3544131#1821
workflows: update lava-test-plans to 3544131#1821qcom-anilyada wants to merge 1 commit intoqualcomm-linux:masterfrom
Conversation
qcom-anilyada
requested review from
anujm1,
koenkooi,
lumag,
ndechesne,
ricardosalveti,
sbanerjee-quic and
vkraleti
as code owners
|
Unit testing for all these commits has already been completed and merged. However, integration testing is still pending. I would like to request the |
|
I'll push it to |
|
I ran the update through If it was my call I would NAK this PR until all the issues with tests are resolved. A word of clarification - if the test result is missing, it's marked as failed. |
mwasilew
left a comment
mwasilew
left a comment
There was a problem hiding this comment.
@qcom-anilyada please fix the reported issues:
"qcs9100"
"GStreamer_Video_Encode_h264_480p",
"core_auth",
"AudioRecord_Config03",
"AudioRecord_Config09",
"OpenCV",
"weston-simple-egl",
"Interrupts",
"hotplug",
"CPUFreq_Validation",
"GStreamer_Video_Decode_h264_480p",
"AudioRecord",
"GStreamer_Video_Decode_h265_480p",
"GStreamer_Video_Encode_h265_480p",
"BT_ON_OFF",
"Ethernet",
"irq",
"WiFi_Firmware_Driver",
"AudioRecord_Config01",
"AudioRecord_Config07",
"KMSCube",
"AudioRecord_Config05",
"fastrpc_test",
"BT_SCAN",
"cdsp_remoteproc",
"DSP_AudioPD",
"BT_FW_KMD_Service",
"WiFi_OnOff",
"adsp_remoteproc"
"qrb2210-rb1"
"GStreamer_Video_Encode_h264_480p",
"AudioRecord_Config07",
"AudioRecord_Config01",
"GStreamer_Video_Encode_h265_480p",
"GStreamer_Video_Decode_h265_480p",
"AudioRecord_Config05",
"AudioRecord_Config03",
"AudioRecord_Config09",
"irq",
"Interrupts",
"GStreamer_Video_Decode_h264_480p"
"qcs9100-rb8"
"BT_SCAN",
"GStreamer_Video_Encode_h264_480p",
"AudioRecord_Config07",
"AudioRecord_Config01",
"GStreamer_Video_Encode_h265_480p",
"GStreamer_Video_Decode_h265_480p",
"AudioRecord_Config05",
"AudioRecord_Config03",
"AudioRecord_Config09",
"GStreamer_Video_Decode_h264_480p"
"qcs615-adp-air"
"GStreamer_Video_Encode_h264_480p",
"GStreamer_Video_Decode_h265_480p",
"GStreamer_Video_Encode_h265_480p",
"AudioRecord_Config01",
"AudioRecord_Config07",
"AudioRecord_Config05",
"core_auth",
"AudioRecord_Config03",
"AudioRecord_Config09",
"GStreamer_Video_Decode_h264_480p"
"qcs8300"
"GStreamer_Video_Encode_h264_480p",
"AudioRecord_Config01",
"GStreamer_Video_Encode_h265_480p",
"GStreamer_Video_Decode_h265_480p",
"AudioRecord_Config07",
"AudioRecord_Config05",
"AudioRecord_Config03",
"AudioRecord_Config09",
"GStreamer_Video_Decode_h264_480p"
"qcs6490"
"GStreamer_Video_Encode_h264_480p",
"AudioRecord_Config01",
"GStreamer_Video_Encode_h265_480p",
"GStreamer_Video_Decode_h265_480p",
"AudioRecord_Config07",
"KMSCube",
"AudioRecord_Config05",
"core_auth",
"AudioRecord_Config03",
"AudioRecord_Config09",
"weston-simple-egl",
"hotplug",
"irq",
"GStreamer_Video_Decode_h264_480p"
0003e2a to
96f19ce
Compare
qcom-anilyada
changed the title
qcom-anilyada
force-pushed
the
test/update-ref
branch
from
96f19ce to
133c664
Compare
qcom-anilyada
changed the title
Test run workflowTest jobs for commit 133c664
All jobs summary
|
Test Results 28 files - 33 28 suites - 236 1h 5m 51s ⏱️ - 2h 4m 12s For more details on these failures, see this check. Results for commit 7d718b3. ± Comparison against base commit f5875af. This pull request removes 48 and adds 3 tests. Note that renamed tests count towards both.♻️ This comment has been updated with latest results. |
|
All BT tests are disabled on qcs615 and it results in a "boot test" job. This job doesn't have any overlay produced by LAVA so it fails on the postprocess step: https://lava.infra.foundries.io/scheduler/job/171231 BT_SCAN test still fails in Hyd. Possibly the Hyd devices need to have the "has-bt" tag removed. @qcom-anilyada please address these issues. |
|
BT_SCAN issue is fixed now in hyd lab, we have kept some BT devices around the setup. Regarding qcs615, with existing setup, there's no way to deal with this. I'm trying to implement testplan exclusion, if that works it will handle such scenarios. PR#27 |
So we need qualcomm-linux/lava-test-plans#46 then, right? I would really like to get these changes in, we have way too many known failures atm. |
yes, pretty much. It fails unit tests at the moment. I'm waiting for the version with unit tests fixed. |
Got this fixed. All unit tests are passing now. |
Please update this pr then, as the pending change was merged. |
|
Pushed an updated version of this PR to next: https://github.com/qualcomm-linux/meta-qcom/actions/runs/24155885285: https://github.com/qualcomm-linux/meta-qcom/actions/runs/24155885285/job/70516463716#step:3:136 (seems an infra issue?) All infra related? |
Several BT failures still :-( |
BT Failures looks valid from build side [WARN] 2026-04-11 23:46:21 - bluetoothd not running, retrying in 5s... |
BT failures have been noticed on RB4, and CI does not currently include RB4. Please check this as a priority and fix the environment if necessary. |
this is new issue which is random and seeing on all targets and tracking with #1938 systemd failed while setting up Linux namespaces / sandboxing for the Bluetooth service, systemd kills the launch before bluetoothd properly starts. Further debug is in progress. [[0;1;31mFAILED[0m] Failed to start [0;1;39mBluetooth service[0m. systemctl status bluetooth -l |
Thank you for bringing up the issue. Since it occurs randomly, should we proceed with the next steps? |
Yes. Once the PR is merged please rebase and retest in next. |
@qcom-anilyada Could you please rebase since #1919 has been merged? |
qcom-anilyada
force-pushed
the
test/update-ref
branch
from
fba992d to
862cf9e
Compare
did rebase and push |
Test run workflowTest jobs for commit 862cf9e
All jobs summary
|
|
I pushed to next https://github.com/qualcomm-linux/meta-qcom/actions/runs/24502167506 |
qcom-anilyada
force-pushed
the
test/update-ref
branch
from
862cf9e to
942a576
Compare
qcom-anilyada
changed the title
I've updated the ref as well in this PR. |
Test run workflowTest jobs for commit 942a576
All jobs summary
|
|
Hamoa fails to boot. I'll exclude it from the list in |
here's the list of changes that comes with this ref update: - fix: correct device path for EXCLUDED_TESTPLANS extraction - workflows: upgrade actions/checkout to v6 - devices: re-enable AudioRecord test for iq-9075-evk - refractor: remove test_name override capability - testcases: Make camera postprocessing conditional on camera testplan - devices: Support per-device testplan exclusion using Jinja2 template engine - testcases: Pass LAVA_TESTCASE_ID to prevent result collisions - projects: meta-qcom: fix flasher_device_type for qcs9100 - devices: Use extensible postprocess steps for iq-x7181-evk - Add support for adding steps in postprocess.jinja2 - projects: meta-qcom: disable fastrpc_test for qcs615 - testcases: Add GStreamer video encode/decode test plan - templates: Add iq-x7181-evk device support - Add pre-merge-audio test for LAVA audio validation - meta-qcom-distro: add more devices - Fix "black" formatting warning - meta-qcom: disable BT tests for qcs615-ride - testcases: Add support for excluding all testcases in a set - Enable jinja2 loopcontrols extension - Add missing submit_to_lava function - projects: add flasher_device_type to devices - include: move flasher_device_type to internal variable Signed-off-by: Anil Yadav <anilyada@qti.qualcomm.com>
qcom-anilyada
force-pushed
the
test/update-ref
branch
from
942a576 to
7d718b3
Compare
|
did rebase.
the workflow failed on build step for some targets: |
Test run workflowTest jobs for commit 7d718b3
All jobs summary
|
rb3gen2 bootup issue https://lava.infra.foundries.io/results/180655 but other things went well |
The nightly jobs do not display this crash for Kodiak.
|
yes, correct we are not seeing this issue in nightly builds |
|
Here are the current issues:
My call would be to disable all failing tests (or fix issues) before merging. The problem with insufficient number of devices is hard to crack. The easiest solution is to limit the number of tests. Second one is to increase the timeout a bit. We need to find a balance between test feedback and test execution time. This problem has no good solution. |
This PR qualcomm-linux/qcom-linux-testkit#412 resolves the BT issue on RB1. @bhargav0610, please remember to report similar issues for all SoCs whenever there are such failures.
https://github.com/qualcomm-linux/meta-qcom/actions/runs/24511752498/job/71850334280#step:3:4619 This is not an issue with the test script. The failure occurred because core 3 did not increment the events, and the final results were updated to FAIL.
|
|
Here's the updated summary of the failed workflows along with their fixes, infra and script side issues have already been resolved. Workflow Summary: 13 Failed Jobs1. Device Crash (Build Issue)
2. Passed but Timed Out
3. Infrastructure / Script Issues (Resolved)
a) Bluetooth Dongle Issue —
|
@mwasilew can you pls review the PR now as most of the things has been addressed now |
6 participants
here's the list of changes that comes with this ref update:
Signed-off-by: Anil Yadav anilyada@qti.qualcomm.com