0

Platform: Linux version 3.10.0-693.el7.s390x (mockbuild@clefos-build-image08.bld.sinenomine.net) (gcc version 4.8.5 20150623 (Red Hat 4.8.5-11) (GCC) ) #1 SMP Sat Sep 16 05:21:59 EDT 2017

Postgres version: PostgreSQL 13.6 on s390x-ibm-linux-musl, compiled by gcc (Alpine 10.3.1_git20211027) 10.3.1 20211027, 64-bit

Postgres' docker image used: 13-alpine, hash ad9f6807f8a2

org.postgresql.util.PSQLException: ERROR: failed to JIT module: Added modules have incompatible data layouts: E-m:e-i1:8:16-i8:8:16-i64:64-f128:64-a:8:16-n32:64 (module) vs E-m:e-i1:8:16-i8:8:16-i64:64-f128:64-v128:64-a:8:16-n32:64 (jit)
P K
  • 69
  • 7
  • Why don't you disable `jit` then? –  May 04 '22 at 07:10
  • Is that a postgres setting? Would it have any performance implications? – P K May 04 '22 at 07:12
  • https://www.postgresql.org/docs/current/jit.html –  May 04 '22 at 07:13
  • That means, there could be some performance reductions that I'd need to measure. Any idea how to make JIT work? – P K May 04 '22 at 07:16
  • And looks like this gets enabled/ disabled during compilation. – P K May 04 '22 at 07:19
  • 1
    You can disable it globally in [postgresql.conf](https://www.postgresql.org/docs/current/runtime-config-query.html#GUC-JIT) or inside the current session using `set jit=off`. In my experience the cases where this actually provides a benefit are very rare (mainly in data warehouse/analytical environments) –  May 04 '22 at 07:21
  • I'm just assuming Postgres developers community and you are much smarter than me. If Postgres chose to use Jit for the queries I'm executing there should be some reason. Since your experience includes data warehouse knowledge you are also right. These two considerations have created a dilemma. – P K May 04 '22 at 07:36
  • Yes, and you should report it as a bug. Don't forget to include a reproducible test case, what exact version of PostgreSQL and LLVM you have installed, and how exactly you installed PostgreSQL. – Laurenz Albe May 04 '22 at 09:22
  • @LaurenzAlbe there's already a similar bug but not sure whether it got fixed - https://www.postgresql.org/message-id/16971-5d004d34742a3d35%40postgresql.org – P K May 04 '22 at 10:29
  • I found this; you could check if your version contains that patch: https://www.postgresql.org/message-id/E1kTBdj-0004T3-QD@gemulon.postgresql.org Based on the time, you should have it, so that's not the problem here. – Laurenz Albe May 04 '22 at 12:10

1 Answers1

0

@a_horse_with_no_name disabling jit fixed it

P K
  • 69
  • 7
  • As it’s currently written, your answer is unclear. Please [edit] to add additional details that will help others understand how this addresses the question asked. You can find more information on how to write good answers [in the help center](/help/how-to-answer). – Community Jun 20 '22 at 07:40