Enhanced schema to represent conditional extension implications?

[ThinkOpenly]

Some extensions don't fall neatly into using the existing "implies" and "requires" schema.

For example, the "C" extension has some somewhat complex implications (from ISA 20240411, chapter 27.2):

The C extension is the superset of the following extensions:

• Zca
• Zcf if F is specified (RV32 only)
• Zcd if D is specified

As C defines the same instructions as Zca, Zcf and Zcd, the rule is that:

• C always implies Zca
• C+F implies Zcf (RV32 only)
• C+D implies Zcd

I don't think there is a way, currently, to represent the sort of conditional "implies" of the latter two rules.

Currently, the C extension doesn't imply anything, which is a bug, but based on the above:

• C should clearly imply Zca (we can implement this now, of course)
• C should imply Zcf if F (RV32 only)
• C should imply Zcd if D

I believe this would be required if instructions claimed to only be definedBy, say "Zca", but a configuration only defines "C"?

Similarly "Zce" (same page as above):

Specifying Zce on RV32 without F includes Zca, Zcb, Zcmp, Zcmt
Specifying Zce on RV32 with F includes Zca, Zcb, Zcmp, Zcmt and Zcf
Specifying Zce on RV64 always includes Zca, Zcb, Zcmp, Zcmt

Oh! I see a TODO in arch/ext/Zce.yaml that describes this problem.

[dhower-qc]

Thanks for pulling together this nice summary. I think we have the ability to specify these arbitrary dependencies for instruction definedBy, but not for implications/requirements. I think if we change the schema of implies/requires to match definedBy, then we can capture this.