Some improvement to adc_continuous and i2s_common (IDFGH-11855)

[mcapdeville]

Hi,
This is somme proposed little improvement to adc and i2s driver.

• In adc continuous driver, it(s not always a good idea to use ringbuffer. to avoid unnecessary copying of data. The first patch allow disabling ringbuff by setting max_store_buf_sieze member of adc_continuous_handle_cfg_t structure to 0.

• The second patch resolve some rounding problem in adc_hal_digi_sample_freq_config

• The third patch is for

  • clearing i2s transmit dma buffer before calling on_sent callback
  • directly setting the dma buffer address in event.data of callbacks instead of that of the finish_desc->buf member

---

Marc

[CLAassistant]

All committers have signed the CLA.

[github-actions]

	Warnings
⚠️	Please consider squashing your 8 commits (simplifying branch history).

👋 Hello mcapdeville, we appreciate your contribution to this project!

---

📘 Please review the project's Contributions Guide for key guidelines on code, documentation, testing, and more.

---

🖊️ Please also make sure you have read and signed the Contributor License Agreement for this project.

---

Click to see more instructions ...

This automated output is generated by the PR linter DangerJS, which checks if your Pull Request meets the project's requirements and helps you fix potential issues.

DangerJS is triggered with each push event to a Pull Request and modify the contents of this comment.

Please consider the following:
- Danger mainly focuses on the PR structure and formatting and can't understand the meaning behind your code or changes.
- Danger is not a substitute for human code reviews; it's still important to request a code review from your colleagues.
- Resolve all warnings (⚠️ ) before requesting a review from human reviewers - they will appreciate it.
- To manually retry these Danger checks, please navigate to the Actions tab and re-run last Danger workflow.

Review and merge process you can expect ...

We do welcome contributions in the form of bug reports, feature requests and pull requests via this public GitHub repository.

This GitHub project is public mirror of our internal git repository

1. An internal issue has been created for the PR, we assign it to the relevant engineer.
2. They review the PR and either approve it or ask you for changes or clarifications.
3. Once the GitHub PR is approved, we synchronize it into our internal git repository.
4. In the internal git repository we do the final review, collect approvals from core owners and make sure all the automated tests are passing.
- At this point we may do some adjustments to the proposed change, or extend it by adding tests or documentation.
5. If the change is approved and passes the tests it is merged into the default branch.
5. On next sync from the internal git repository merged change will appear in this public GitHub repository.

Generated by 🚫 dangerJS against c0c6921

[suda-morris]

we don't allow to use "esp_log.h" in the hal component, instead, we suggest using "hal/log.h" in the hal.

[L-KAYA]

Hi @mcapdeville , thanks for your contribution! I have one question, why we should clear the i2s tx buffer before callback? Some use cases would like to read back these sent data to check the data latency, clearing before callback will break these cases.

Another concern is that the pointer change of event.data might lead to breaking change, do you think this change is critical?

[mcapdeville]

Hi,

Hi @mcapdeville , thanks for your contribution! I have one question, why we should clear the i2s tx buffer before callback? Some use cases would like to read back these sent data to check the data latency, clearing before callback will break these cases.

In my application, I directly fill dma buffers in the callback function. I 'm trying to avoid unnecessary recopy of data by passing the buffer's pointer to the encoding function. If ther is no data to be sent, the output datas are allready zeroed.

• Zeroing the buffer after the callback will erase the encoded datas.
• The i2s_channel_write function add an unnecessary copy of data.

Another concern is that the pointer change of event.data might lead to breaking change, do you think this change is critical?

I think, it is more practical the the event struct contain the pointer to the dma buffer and size of the buffer instead of a pointer on the pointer to the dma buffer.

Regards.

---

Marc

[suda-morris]

sha=5e6a1e5aee8a446670f291d048ae24abf491f456

[L-KAYA]

Hi @mcapdeville, the sequence of buffer clear and callback may need more discussion.

If you want to copy data into the TX DMA buffer directly in the callback, you can just disable the auto_clear and operate the buffer in the on_sent callback. Or do you have any reason that auto_clear must be set?

And if the auto_clear first is necessary, do you think it's reasonable to have another flag to decide to execute auto_clear first or on_sent callback first?

As for the DMA buffer, indeed the first level pointer would be more practical, this should be worth a breaking change

[masterxq]

Just a user story:
I have very likely changes in my espressif, and I'm patching them in my pipelines into the espressif framework to avoid extra copy in i2s... This is so unnecessary... I understand why this can be needed, but I really don't understand why this is the only possible way, it is just super ugly :/

In my implementation I have an additional function that's avoid copying and the Original is unpatches...

This could be a way around possible breaking changes, find problems, learn more user stories and avoid that users needs to implement it again and again...