Hi Khem, This is regarding the weston systemd watchdog timeout of 20 seconds:
https://github.com/openembedded/openembedded-core/commit/c21fa5a291ab207a084285935ab73a0b4225c965#diff-ac8a4b56ade4a43a070c93486fb1d7606573da5d44de96bc1529d15b0b216660R36 For GPU-limited parts, we see watchdog timeouts during intense testing, though the part is not hung, it's just going to take longer than 20 seconds. I stumbled across the following systemd commit, which looks like the watchdog used to default to 180 seconds. Yet even then was under scrutiny because it was causing un-necessary failures more often than not: https://github.com/systemd/systemd/commit/21d0dd5a89fe0ef259ca51ebea9f39dd79a341c2#diff-9d3c1cd071b0f173aaf53a7799ac5765bb3a58d0d07125f075b6b49112f0a425 Do you think the default should be raised? Or should we just make a special case in our layer? Tom
<<attachment: winmail.dat>>
-=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#155193): https://lists.openembedded.org/g/openembedded-core/message/155193 Mute This Topic: https://lists.openembedded.org/mt/85096163/21656 Group Owner: openembedded-core+ow...@lists.openembedded.org Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-