[libc-commits] [clang] [compiler-rt] [libc] [libunwind] [lld] [lldb] [llvm] Add NXP EZH (SmartDMA) architecture (PR #201498)

[via libc-commits]

rpasek wrote:

> Hi, please read https://llvm.org/docs/DeveloperPolicy.html#adding-a-new-target.

Thanks! I believe the relevant section for EZH is "The basic rules for a back-end to be upstreamed in experimental mode are:"

- I've added myself to Maintainers.md.
- I don't believe there are contentious issues. I've avoided making changes anywhere in the project except in areas where I explicitly needed to add EZH
- I think I comply with the LLVM policies. I'm adhering to LLVM coding standards and I don't have any license and or patent issues. Even though it's not required because I used only publicly available information in my implementation, I sought out and received approval from NXP before proceeding.
- There is reasonable documentation. I've shared this in my commit messages "SmartDMA Cookbook (AN14650.pdf), fsl_smartdma_prv.h". There is also publicly available low cost hardware such as https://www.digikey.com/en/products/detail/nxp-usa-inc/LPC55S69-EVK/9865963

The area where I'm still a bit confused on is "There must be an active community behind the target." Considering this is a new target and it wasn't previously possible to write C code for this processor, a community doesn't really exist yet. What would be the best approach for this? Should I wait and try to form a community? Maybe a pledge from NXP to be a maintainer would be sufficient?



https://github.com/llvm/llvm-project/pull/201498