labrinea wrote: > A motivating example might be target_version("simd+dotprod") and > target_version("sve"): when sve is available we should pick that version, but > the current rule prioritizes the number of attribute-specified features over > their overall weight.
I agree about this. As I said we will adress the priorities and selection algorithm separately. However in the example I am mentioning the problem arises once we mix target-features from the cmdline with fmv-features. So we need separate metadata. I was thinking whether it makes sense to expand the fmv dependecies before we store them in metadata but I think it's not necessary. We can perform this step later. https://github.com/llvm/llvm-project/pull/118544 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits