DavidSpickett wrote: > Ohh, there is so many -none-unknown- in current code. I guess it may be > widely used. Do we really want to change all of them?
In current test cases, which were likely generated using clang. So they are likely not relying on that exact form of the triple, it's just what clang put out at the time. So they don't prove anything either way, unless they fail after this change of course. Thank you Peter for testing this PR, I've also asked another downstream team to test their bare metal builds with this. That'll be our best indication for now if this makes sense. https://github.com/llvm/llvm-project/pull/89638 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
