Hello Andres,
24.10.2025 17:20, Andres Freund wrote:
I wonder if we should just write these off as "probably an LLVM bug".
I'm not sure that's really convincing, given that REL_16_STABLE seems to not
have an issue?
FWIW, petalure with llvm 4.0.1 has produced the first crash due to
"double free or corruption (!prev)" in REL_16_STABLE too:
https://buildfarm.postgresql.org/cgi-bin/show_log.pl?nm=petalura&dt=2026-03-14%2002%3A03%3A06 Best regards,
Alexander