Re: [edk2-devel] The principles of EDK2 module reconstruction for archs

["Sunil V L" <sunilvl@ventanamicro.com>]

On Wed, Sep 28, 2022 at 03:33:45AM +0000, Ni, Ray wrote:
Hi Ray,
> 
>   1.  When a new arch's implementation is introduced to the existing module which was developed for the specific arch:
> 
>   1.  The folder reconstruction:
> 
>   *   Create arch folder for the existing arch implementation
> [Ray] Do you move existing arch implementation to that arch folder? It will break existing platforms a lot.
> 
>   *   Create the arch folder for the new introduced arch
> [Ray] I agree. But if we don't create arch folder for existing arch implementation, the pkg layout will be a mess.
> 
> [Ray] Hard for me to understand all the principles here. Maybe we review existing code including to-be-upstreamed code and decide how to go.
> 

Could you please take a look below changes which is trying to add RISC-V
support for CpuDxe?
https://github.com/tianocore/edk2-staging/commit/bba1a11be47dd091734e185afbed73ea75708749
https://github.com/tianocore/edk2-staging/commit/7fccf92a97a6d0618a20f10622220e78b3687906

What do you suggest with above example?

1) Common INF for all architectures - but modify INF alone, no X86
folder creation.

This is what I have done in the commit above. May be of least impact to existing code
since it is only INF change. But like you mentioned this is bit weird that X86 files will
remain in root folder directly along with some common files.

2) Common INF (CpuDxe.inf) + create arch folders X86, X64, IA32, RiscV64 etc

IMO, this is probably the best approach. What would be the challenges
with this?

3) Separate INF for arch like CpuDxe.inf for x86, CpuDxeRiscV64.inf for
RISC-V.

This again probably is not a good idea. 

4) If the module/library is specific to one arch (ex: SMM(X86),
SBI(RISC-V)), then create separate INF.

Thanks!
Sunil