I2C: sample SDA earlier

[bugadani]

Thank you for your contribution!

We appreciate the time and effort you've put into this pull request.
To help us review it efficiently, please ensure you've gone through the following checklist:

Submission Checklist 📝

• I have updated existing examples or added new ones (if applicable).
• I have used cargo xtask fmt-packages command to ensure that all changed code is formatted correctly.
• My changes were added to the CHANGELOG.md in the proper section.
• I have added necessary changes to user code to the Migration Guide.
• My changes are in accordance to the esp-rs API guidelines

Extra:

• I have read the CONTRIBUTING.md guide and followed its instructions.

Pull Request Details 📖

Description

Please provide a clear and concise description of your changes, including the motivation behind these changes. The context is crucial for the reviewers.

Testing

Describe how you tested your changes.

[bugadani]

To be fair, I should probably remove the half_cycle/2 term instead. Technically, we can sample at any point after the rising edge, and scl_wait_high is supposed to be somewhat dependent on the hardware (i.e. it should be tunable, maybe). scl_wait_high represents the rising edge of the clock signal, after which the clock signal is high for scl_high. We should probably aim at the start/middle of the interval covered by scl_high.

Still, this should be good enough to test.

[armandas]

@bugadani I just ran into this issue and can confirm that the proposed change does resolve it. Any chance we can revive this PR?

As far as I can see, the esp_hal timing differs from that used in esp-idf, and it appears that the change was introduced in #263 without any comment.

Background

I have a ESP32-C6 board from Waveshare and was trying to read the touch driver. The provided ESP-IDF code works flawlessly, however, when using esp_hal the data is garbled.

I started probing the bus with my oscilloscope and to my surprise, the readout started working. I went to see if anyone has reported such an issue and ended up here.

[bugadani]

I completely forgot about this. This PR was abandoned because it failed to fix an issue where I thought this would be relevant. We can certainly revisit this

[armandas]

May I help with any additional testing or can we just treat esp-idf as a source of truth?

[bugadani]

Of course, any testing you do, any issues you can find, are welcome.

[armandas]

I mean in order to be able merge this PR.

[bugadani]

Ah, this has been already done in #4268

[armandas]

Thanks a lot, I missed that!