https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122380
--- Comment #19 from Jakub Jelinek <jakub at gcc dot gnu.org> --- (In reply to Henrik Lehtonen from comment #18) > (In reply to Richard Biener from comment #17) > > Was the original testcase reduced from a real-world program or was it > > created artificially? > > I ran into this issue as part of a real-world program (where we had to work > around it when upgrading to GCC 15). The attached testcase was manually > reduced from there. And note it is even more important for GCC 16, because we've switched to -std=gnu++20 by default (the testcase doesn't reproduce with -std=gnu++17 we've previously used as default, <variant> had some C++20 changes that affect this).
