Chris Evich
Chris Evich
Looking at the suggested policy details, `CMP0022` seems to suggest a fix which will also be compatible with older cmake versions (re: `cmake/Platform/Arduino.cmake` line 983): ``` Warning-free future-compatible code which...
Just by total guessing and some poking around, I found this makes all the warnings go away (though **I don't know** if it is a correct fix, or will work...
Here's a pull-request with this change if by chance I **guessed correctly** (please double-check): #94
Dunno, but I've been running with #94 applied and it seems to address the issue. But again, I'm not a CMAKE expert, just know enough to be dangerous :smile:
Oh wow, this has been around for quite a while. I thought it was a new thing. Thanks for the data Ed.
[Another Ubuntu example from last night's cirrus-cron run.](https://cirrus-ci.com/task/5938293828222976?logs=integration_test#L1198)
[Last night on 'main' in cirrus-cron](https://cirrus-ci.com/task/4712885250162688?logs=integration_test#L1078) - also F34 + overlay.
This continues to flake fairly regularly in daily branch runs. [Latest example.](https://cirrus-ci.com/task/4720256957546496?logs=integration_test#L1080)
> we can have a global runtime lock to prevent this. I don't know the code, but that sounds like it might prevent parallel pulling, no? IMHO, pulling in parallel...
This has been a problem for a very long time. I seem to remember Miloslav or Nalin mentioning the underlying name/pull race problem being in c/storage or c/image. At the...