ojhunt wrote: @AaronBallman I think this is likely to be a real memory use regression - a source location for every comma in a paren expression seems excessive for the diagnostics.
@nvetrini I suspect it's probably better to find the comma location by traversing the from the end of each expression to find the next comma. You'll need to do some checking for macro expansions, for which I'd just say give up and not bother with accurate comma placement because at that point comma location becomes irrelevant. https://github.com/llvm/llvm-project/pull/199411 _______________________________________________ cfe-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
