jmorse wrote: Interesting -- I suppose the new variable recording where vtables are count as a source-variable, and in some circumstances the vtable address must be optimised away? (Whole program devirtualisation for example).
I'd suggest that this isn't a concerning reduction in coverage (because it's new coverage we're adding, but not 100% of the time). Is the 100% source-location-coverage for google a target that you want to maintain, or is it alright to accept the new lower coverage measurement? We could try to claw back some of the coverage (perhaps by not emitting vtable locations if they're likely to be optimised away?), but it'd be an uphill battle. https://github.com/llvm/llvm-project/pull/130255 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
