https://bugs.exim.org/show_bug.cgi?id=2618
--- Comment #10 from Carlo Marcelo Arenas Belón <carenas@???> ---
If our guesses are correct something as simple as the proposed pull[1] might be
all that is needed to get this working again in macOS 11 arm64.
but a traceback of the original crash would help, and the final solution will
obviously need to be more careful to avoid impacting any other cases, with the
following likely requiring access to that platform for testing :
* how is the code supposed to react when NOT threaded?
* is `if (__builtin_available(macOS 11, *))` enough of a warranty for
availability of this new interface?, how is it supposed to work with other
compilers like gcc?
* how do the constrains change with notarization or the hardened runtime?
* are there any other considerations to take for the other sub-platforms (iOS,
watchOS, ipadOS, ..)?, for example there was an undocumented 1 mmap with
MAP_JIT constrain that affected them and required all the extra runtime checks
to protect against also in macOS
[1]
https://github.com/zherczeg/sljit/pull/52
--
You are receiving this mail because:
You are on the CC list for the bug.
