https://gcc.gnu.org/bugzilla/show_bug.cgi?id=124048
Jeffrey A. Law <law at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Ever confirmed|0 |1
Last reconfirmed| |2026-02-10
Status|UNCONFIRMED |WAITING
--- Comment #3 from Jeffrey A. Law <law at gcc dot gnu.org> ---
FWIW, I've got a clean bootstrap on the following recent hashes:
08a17b96c31f78bf7d030f5550ef801580a178ed (2/9/2026)
3350e383b057fb7816f0f51ed6c3d36b7ca690d0 (2/9/2026)
1a3f27ab4d3f0d4549b0e1724d5297dc2d393a77 (2/7/2026)
15ca69befd0e0a5408284a532affe9085742fd86 (2/7/2026)
7dda46111eb3ebc390c0287a00d97a69567b33f1 (2/7/2026)
It's not an exact match for Matthias's builds as it's just generic rv64gc and
the in-play theory is the failure is vectorizer related.
Give the slowness of these systems, it's probably best if Matthias gives us an
exact hash and we configure in an equivalent manner. As others have also
noted, knowing if it's stage1, stage2 or stage3 in the build process would be
helpful.
