zeyi2 wrote: > I think maybe the way forward is to add only the assembly info, and leave out > the codegen info (i.e. `mayLoad`, `Uses`, `Defs`, `isBarrier`, `isReturn`, > `isTerminator`), which we can come back to when we implement code generation > for these instructions.
That makes sense. I've removed these properties, leaving only the instruction encodings and assembler predicates for now. Thanks! https://github.com/llvm/llvm-project/pull/215095 _______________________________________________ cfe-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
