On CI_DRM_3075 fi-kbl-7500u assert on I2C access error: (chamelium:1396) igt-chamelium-CRITICAL: Test assertion failure function chamelium_rpc, file igt_chamelium.c:301: (chamelium:1396) igt-chamelium-CRITICAL: Failed assertion: !chamelium->env.fault_occurred (chamelium:1396) igt-chamelium-CRITICAL: Chamelium RPC call failed: RPC failed at server. <class 'chameleond.utils.i2c.I2cBusError'>:I2C access error https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_3075/fi-kbl-7500u/igt@chamelium@dp-edid-read.html
History seen: https://intel-gfx-ci.01.org/cibuglog/index.html%3Faction_failures_history=272.html
https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_3111/fi-kbl-7500u/igt@chamelium@hdmi-hpd-fast.html Note, on all other runs on the fi-kbl-7500u the igt@chamelium@hdmi-hpd-fast is skipped. So, it appear that it is possible to get an assert from chamelium tests before checking that the test should be skipped. I.e. these tests need to be restructured otherwise they will produce noise.
For the igt@chamelium@dp-edid-read case we don't know if the I2C issue also happened before even starting test or during execution. Functions in igt_chamelium.c that are used to established if the tests can be run should not assert they should return an error code and let the actual test decide what to do.
Are these more Chamelium issues than driver / IGT issues?
(In reply to Jani Saarinen from comment #4) > Are these more Chamelium issues than driver / IGT issues? It should be fixed in igt/lib/igt_chamelium.c. But it is slightly tricky, since we can't know when instabilities on I2C could start occuring.
*** Bug 102514 has been marked as a duplicate of this bug. ***
https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_7136/fi-kbl-7500u/igt@chamelium@dp-crc-fast.html (chamelium:1361) igt-chamelium-CRITICAL: Test assertion failure function chamelium_rpc, file igt_chamelium.c:302: (chamelium:1361) igt-chamelium-CRITICAL: Failed assertion: !chamelium->env.fault_occurred (chamelium:1361) igt-chamelium-CRITICAL: Chamelium RPC call failed: RPC failed at server. <class 'chameleond.utils.i2c.I2cBusError'>:I2C access error Subtest dp-crc-fast failed.
Note, the chamelium test have changed name to kms_chamelium since CI_DRM_3369. Also, this issue was not reproduce after on the first 2 runs.
Reproduced: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_3612/fi-kbl-7500u/igt@kms_chamelium@hdmi-hpd-fast.html (kms_chamelium:3885) igt-chamelium-CRITICAL: Test assertion failure function chamelium_rpc, file igt_chamelium.c:302: (kms_chamelium:3885) igt-chamelium-CRITICAL: Failed assertion: !chamelium->env.fault_occurred (kms_chamelium:3885) igt-chamelium-CRITICAL: Chamelium RPC call failed: RPC failed at server. <class 'chameleond.utils.i2c.I2cBusError'>:I2C access error
Last seen CI_DRM_3612: 2018-01-09 / 211 runs ago
https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_3752/fi-kbl-7500u/igt@kms_chamelium@hdmi-hpd-fast.html
Still valid?
yes, last IGT run: https://intel-gfx-ci.01.org/tree/drm-tip/IGT_3995/fi-kbl-7500u/igt@chamelium@hdmi-hpd-fast.html
Now it just shows a warn: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_4969/fi-kbl-7500u/igt@kms_chamelium@dp-edid-read.html
https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_5090_113/fi-kbl-7500u/igt@kms_chamelium@dp-crc-fast.html Starting subtest: dp-crc-fast Subtest dp-crc-fast: SUCCESS (8.582s) (kms_chamelium:2846) igt_chamelium-CRITICAL: Test assertion failure function chamelium_rpc, file ../lib/igt_chamelium.c:303: (kms_chamelium:2846) igt_chamelium-CRITICAL: Failed assertion: !chamelium->env.fault_occurred (kms_chamelium:2846) igt_chamelium-CRITICAL: Chamelium RPC call failed: RPC failed at server. <class 'chameleond.utils.i2c.I2cBusError'>:I2C access error
Not seen lately: (1 month, 2 weeks / 614 runs ago). Resolving
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.