https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114505
--- Comment #12 from Eli Schwartz <eschwartz93 at gmail dot com> --- (In reply to Xi Ruoyao from comment #1) > Confirmed as I've seen it before. At least we should turn it into a > "normal" error message instead of ICE. > > But note that even "downgrading the compression algorithm" is supported, it > won't really help rolling-release distros. When GCC itself is upgraded with > a LTO API version change, all previous LTO static archives will be unusable > anyway. Generally the distros should not ship LTO static archives at all. Agreed, figuring out how to solve that is an open ticket too: https://bugs.gentoo.org/926120 Still would like to solve the zstd case :D especially since I am sure given the flexibility of a from-source distro there will be users who insist they'd rather rebuild the entire OS on a GCC upgrade, so reducing the compression algorithm edge case will come in use. (In reply to Andrew Pinski from comment #2) > On the input side: > Wait this should not happen. > > lto_end_uncompression (stream, > file_data->lto_section_header.get_compression > ()); I cannot swear this was the case but the error very specifically occurred as part of a Gentoo profile change that caused zstd to be enabled. The version of zstd should be the same, at least...
