Scott Mabin

I didn't end up reviewing #836, so I don't have much context here, but the PR states that the calibration was being done in millivolts so the extra resolution was...

Good idea! Let's do that instead, I've updated the title.

[embedded-svc](https://github.com/esp-rs/embedded-svc) is meant to be completely unaware of esp-idf, so I don't think we should be changing this struct to fit esp-idf's alignment needs unless there is another valid reason...

> To be more concrete about my problem. I would like to use the I2S in DAC DMA mode using the LCD flag and with my current skills I would...

https://github.com/esp-rs/esp-hal/pull/1375 should solve your 2nd issue.

I think this is a good idea, the only down side is nightly is that is needed :(

When we look into this, we should also fix #954 as removing the 'static bound will make it more likely that the buffer could be in cache (which the DMA...

The heap is now configurable via the config so I think we can close this now.

This looks promising. To keep things simpler, what about just returning each half of the buffer based on the half given as a param? E.g if you pass Half::Second you...

I think option A is probably the right call, if users are using IDF versions that are EOL, they need to upgrade. In hindsight, we probably should have had an...