Jan Midtgaard
Jan Midtgaard
That makes sense to me, thanks! :+1: I'll build the compiler with your proposed fix to confirm that I can no longer reproduce.
Fixed in #13408 (I also ran the reproducer steps for several hours on Linux/macOS without triggering it)
Thanks for the suggestion! I can try it with the native code compiler to see if anything pops up :+1: 32-bit is bytecode only and `ocaml-option-tsan` has `conflicts: "ocaml-option-bytecode-only"` (among...
> For bytecode it should be sufficient to compile the runtime with TSan, though that will give you C-level backtraces instead of OCaml-level ones. I've spent a bit of time...
> I have not been able to reproduce the issue on a 32-bit PowerPC system under OpenBSD yet (it's been hours). Have you been able to reproduce this on a...
> Have you been able to reproduce this on a musl-based Linux system? The `musl` keyword was indeed the magic spell to get this to reproduce locally @dustanddreams! Interestingly, I've...
I'm on opam.2.2.0
On Ubuntu `conf-gmp` should install `libgmp-dev` with your system package manager (`apt` on Ubuntu) and then try to compile an example `test.c` that depends on `gmp.h`. You can find the...
CI summary for ca8cf93: - 11 5.2 workflows failed as expected due to the Gc test - 32bit 5.2 failed with a segfault - Bytecode aborted with `Fatal error: allocation...
I removed the stress test, as the positive parallel test is already stress testing. I've also rebased based on #471 which should take care of the MinGW+Cygwin version mismatching. I...