Jan Midtgaard

I just pushed a few changes, partly to address some of the red lights. I still expect a 32-bit 0xdeadbeef out-of-range though.

I've taken a stab at this. I realized the same indexing issue with shrinking as you, and have (in parallel) `Util.protect`ed all indexed `Dynarray` calls. Furthermore I observed this taking...

CI summary: all 33 workflows green.

CI summary for merge to `main`: - Bytecode 5.3 timed out due to excessive shrinking in STM Weak test parallel #498 Out of 34 workflows, 1 failed with a false...

Saw this on ppc64le 5.2 too in #469 https://ocaml-multicoretests.ci.dev:8100/job/2024-09-03/111022-ci-ocluster-build-4bdab6 ``` random seed: 461619666 generated error fail pass / total time test name [ ] 0 0 0 0 / 1000...

This is failing CI spectacularly, despite I've only changed a platform not tested on the CI: - Arch, Fedora and macOS are failing because they install Python 3.12, but the...

The latter 3 issues should be handled by 4fdd349, 34c2625, and d340759. Fixing the former requires changes to the https://raw.githubusercontent.com/ocaml/opam-source-archives/main/patches/conf-python-3-7/configure.sh - which is now located elsewhere. What's the policy regarding...

I've now added a patch PR in opam-source-archives here: https://github.com/ocaml/opam-source-archives/pull/32

I've updated https://github.com/ocaml/opam-source-archives/pull/32 as requested last week. Once that's merged, this PR needs updating to point to `conf-python-3-7.1/configure.sh`

Modulo a curious opam lint error ~~and risc-v runners still waiting for a worker~~, this is all green :tada: