https://gcc.gnu.org/bugzilla/show_bug.cgi?id=119508
--- Comment #19 from ro at CeBiTec dot Uni-Bielefeld.DE <ro at CeBiTec dot Uni-Bielefeld.DE> --- Solaris results (64-bit-default only; there's no 32-bit Solaris target support in rustc/cargo) are pretty weird (all on 20250411). In all cases, I'm testing both 64 and 32-bit multilibs: * amd64-pc-solaris2.11 using as: no XPASSes 6 FAILs * amd64-pc-solaris2.11,using gas: 38 XPASSes total, some for both 32 and 64-bit, others only for either 4 FAILs * sparcv9-sun-solaris2.11 using as: 36 XPASSes, simlarly either or both multilibs 72 FAILs * sparcv9-sun-solaris2.11 using gas: 40 XPASSes, same pattern 89 FAILs I haven't yet checked which of the FAILs above are known/reported or new. At least some issues on big-endian hosts like sparc almost certainly remain on every BE host. The difference between the Solaris/amd64 as and gas cases strongly suggest that the nr2 tests still aren't reliable. I have a very hard time believing that the different assemblers cause this.
