joker-eph wrote: > I think you are assuming guarantees here that doesn't exist.
I'm just going by you're writing :) You wrote: "To be able to flag incompatible bytecode files rather than have it fail later in mysterious ways" But my take is that you're not actually providing the ability to flag incompatible bytecode, instead I believe what you're providing is the ability to explicitly invalidate every existing serialized bytecode at some arbitrary points in time by bumping the version. In between these even where the version is bumped: you haven't changed anything and we can't flag incompatible bytecode. > So this is a mechanism with policy TBD. Right, and I'd like a better story around how we'd handle this policy before landing this kind of code in-tree actually: this is a bit too much "YOLO" to me right now. https://github.com/llvm/llvm-project/pull/79514 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
