andykaylor wrote: > @Andres-Salamanca thanks for working on this. We usually write tests that > exercise any possible different paths codegen can take. For next rounds: it's > fine if you add less things in one go and make it more incremental with the > use of (a) errorNYI and (b) assert on missing features.
We should probably make it a standard practice when upstreaming new ops like this that if the change goes above 500 lines or so, we should leave the lowering to LLVM IR and related transformations for a follow-up PR. https://github.com/llvm/llvm-project/pull/134333 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
