sammccall added a comment. Herald added a subscriber: wangpc. In D122529#3422158 <https://reviews.llvm.org/D122529#3422158>, @aaron.ballman wrote:
> This looks.... good?.... to me. :-) > > Despite this complicating things by a fair amount, I don't have a better > suggestion to offer. LGTM (sorry to turn up a year later, just want to see if I'm missing something) AFAICT there's only ever one CurMatchData object. In which case squeezing the union discriminator into the low bits saves... 8 bytes of stack over the whole AST traversal. I ran into this when I tried to add a ninth pointer type. We could use a second bit of the callback, but seems like we can just store the discriminator in an int instead? Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D122529/new/ https://reviews.llvm.org/D122529 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
