[Bug c++/92654] internal compiler error: in lookup_template_class_1

2019-11-25 Thread marxin at gcc dot gnu.org
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92654 Martin Liška changed: What|Removed |Added Keywords||ice-on-invalid-code Status|UN

[Bug c++/92654] internal compiler error: in lookup_template_class_1

2019-11-25 Thread fiesh at zefix dot tv
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92654 --- Comment #2 from fiesh at zefix dot tv --- It's been made invalid by creduce, but the original code was valid. If necessary, we can try to produce valid code that leads to the same issue. But I'd only do so if necessary since it's somewhat cu

[Bug c++/92654] internal compiler error: in lookup_template_class_1

2019-11-25 Thread fiesh at zefix dot tv
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92654 --- Comment #3 from fiesh at zefix dot tv --- (Or is there some other trick to make it valid apart from extending the interestingness test of creduce to include a clang compilation step?)

[Bug c++/92654] internal compiler error: in lookup_template_class_1

2019-11-25 Thread marxin at gcc dot gnu.org
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92654 --- Comment #4 from Martin Liška --- (In reply to fiesh from comment #2) > It's been made invalid by creduce, but the original code was valid. > > If necessary, we can try to produce valid code that leads to the same issue. > But I'd only do so

[Bug c++/92654] internal compiler error: in lookup_template_class_1

2019-11-25 Thread fiesh at zefix dot tv
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92654 --- Comment #5 from fiesh at zefix dot tv --- Thank you for your offer. The original translation unit is a whopping 20MB and took about 3 days to reduce ;-) I changed the file and the interestingness test to make sure clang compiles it. It's ru