JonChesterfield accepted this revision. JonChesterfield added a comment. This revision is now accepted and ready to land.
I agree this is an improvement on the current status and should land. Reading through it has made me nervous about our current semantic checking. In particular, I'm not sure suppressing error messages from code that ultimately proves to be discarded as the same as not parsing code that ultimately was not required, but that design choice seems to have been made a long time ago. Presumably lazy diagnostics had better engineering properties than lazy parsing and codegen - I'd be interested to see docs/emails about the early design choices for single source target offloading if someone has them on hand. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D95912/new/ https://reviews.llvm.org/D95912 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
