dblaikie added a comment. I imagine, again, the idea is that if you're generating that many warnings that the performance of printing warnings matters, perhaps you're not paying attention to those warnings? you could disable the ones you aren't interested in? Especially if they're being printed to a terminal (which, since teh output seems to be coloured/including ansi escape codes, it looks like they are going to a terminal) - if you're getting thousands of pages of warnings, hardly seems likely that you can see/act on them?
But sure, in general, probably would be fine to buffer stdout too, might be nice. Same sort of solution - enabling buffering with some reasonable buffer size. (can't be buffered on some platforms if coloured output is being used, like windows, where colouring requires API calls so it's not embedded in the stream like ansi escape codes) Might be worth checking what clang or other LLVM tools do about this sort of thing, maybe there's an existing solution that can be ported over. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D90010/new/ https://reviews.llvm.org/D90010 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
