ojhunt wrote: I have a very immediate fix, but I'd rather a slightly nicer one - at some point refactoring exposed a problem in access control diagnostics, in a way that is very very weird and implies something else broken in the existing tree as well.
If I can't work out the correct systemic fix in the next half hour or so I'll introduce a trivial behavioral fix and then work out the correct fix. Basically an access check that should pass fails, the pre-p2719 code path ignored that failure, so worked, p2719 unified a pile of diagnostics which means that the failing access check causes an error, however at the same time the failing access check does not actually report an error, so we get sadness. The underlying issues are: the access check fails erroneously, and the failed access check does not report an error, the trivial fix is restoring the "ignore the return value of the access check function" but if possible I'd like to just make sure that the access check does the right thing, I'll land "ignore the access check failure like we did in the past" in half an hour or so if I haven't worked out what the actual error is. https://github.com/llvm/llvm-project/pull/113510 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
