boards/arm/rp2040/common: Add weak_function to SPI common logic

[keever50]

Summary

Board logic change.
This PR adds weak_function attributes to the RP2040 common SPI board logic.
This allows board developers to override and extend the SPI board logic.

Impact

This allows board developers to add custom SPI logic such as adding additional chip select pins.
Adding new SPI devices such as displays or custom SPI devices like external boards is now possible.

External custom boards will have the biggest impact, as these are typically not pushed.

Testing

This has been tested by building the code on linux with raspberrypi-pico:nsh config.
I also tested it on a custom board and it works.
However, my knowledge is not very big about compiler compatibility when weak_function is used.

Please let me know if there are issues.

[nuttxpr]

[Experimental Bot, please feedback here]

No, this PR does not fully meet the NuttX requirements. While it provides a good starting point, it lacks crucial details.

Missing/Insufficient Information:

• Summary: While it mentions the "what," it's missing a strong "why." Why is this change necessary? What specific problem does it solve beyond a general desire for extensibility? A related NuttX issue would greatly strengthen this section.
• Impact: The impact section is too vague.
  • User Impact: While it mentions custom boards, it doesn't specify how users would adapt. Would they need to change configurations? Write new drivers? Clearer examples are needed.
  • Build Impact: Does adding weak_function change the build process in any way? Even if the answer is "NO," explicitly state it.
  • Hardware Impact: Specify the affected architecture (ARM) and board (Raspberry Pi Pico, and any others). "Custom board" is not specific enough. List the specific hardware changes if any.
  • Documentation Impact: Does this require documentation updates? If so, have they been included in the PR?
  • Security Impact: Explicitly state "NO" if there's no security impact.
  • Compatibility Impact: Explicitly state "NO" or explain potential backward compatibility issues. Consider the impact on existing boards using the RP2040 SPI.
• Testing:
  • Build Host(s): Be specific about the Linux distribution used, the compiler version (e.g., GCC 12.2.0), and any relevant build tools.
  • Target(s): Be precise about the target configurations tested. "raspberrypi-pico:nsh config" isn't precise enough. Specify the full configuration name (e.g., configs/raspberrypi-pico/nsh). For the custom board, provide details about its configuration.
  • Logs: The "Testing logs" sections are empty. Include relevant log output demonstrating the functionality before and after the change. Show how the new weak functions are being overridden and what the results are. Demonstrate that existing functionality is not broken.

Recommendations:

1. Clarify the "why": Explain the specific problem this solves. Provide a concrete example of how the lack of this feature hindered development before. Link a related NuttX issue if one exists.
2. Detailed Impact Assessment: Be explicit about all impact categories, even if the answer is "NO." Provide specific details and examples where necessary.
3. Thorough Testing Information: Provide complete details about the build host and target environments. Include relevant log output demonstrating the change's functionality and the absence of regressions. Test and document the behavior when the weak functions are not overridden to ensure backward compatibility.
4. Compiler Compatibility: Research and address the compiler compatibility concerns regarding weak_function. Document any limitations or required compiler versions.

By addressing these points, the PR will better meet the NuttX requirements and be more likely to be accepted.

[lupyuen]

Please fill in the Commit Description (copy from the PR Summary). And remember to sign-off with git commit -s. Thanks! :-)

• 93e0db1

[jerpelea]

please add a commit message

[keever50]

Okay! Signed and added a message

[keever50]

I realized i copied the entire PR message and should have been only summery i believe. Hope this isnt a big issue.

[lupyuen]

Sorry could you run git commit -s to sign off? Thanks :-)

[keever50]

Signed

[lupyuen]

Hmmm I don't see the "Signed-off-by". Could you do git push? Thanks!

[keever50]

@lupyuen By just running commit -s i get "nothing to commit, working tree clean". I ran git commit --amend --no-edit -S which should in theory add a sign?

[keever50]

ahhh i see signed off now! However now i have issues pushing it because it does not let me enter my password anymore. Ill get back to this soon

[keever50]

oh okay do the same thing twice and it works.

[lupyuen]

Looks great, thanks! :-)