[NOMERGE] Test macOS CI workflow without Roswell #53
Conversation
ak-coram
force-pushed
the
macos-no-roswell
branch
4 times, most recently
from
2512816 to
5f7ba4d
Compare
ak-coram
force-pushed
the
macos-no-roswell
branch
18 times, most recently
from
08a097c to
67bc91c
Compare
|
FYI: ivmai/bdwgc#576 |
catap
force-pushed
the
macos-no-roswell
branch
from
4758404 to
62a1f8c
Compare
|
The fix inside ECL: https://gitlab.com/embeddable-common-lisp/ecl/-/merge_requests/303 |
|
This fix allows to pass CI a few time in the row |
Co-authored-by: Kirill A. Korinsky <kirill@korins.ky>
catap
force-pushed
the
macos-no-roswell
branch
from
62a1f8c to
9c60ea6
Compare
|
Fix inside ecl |
|
Seems quite stable now. Thanks to discover how to reproduce it. I do have encountered it before but haven't able to reproduce it. |
|
@catap thanks for the investigation and the fixes! Let me know if you still need this PR, otherwise I'll close it. |
|
The fix were merged into upstream => nothing to do with that on my side. but it is sadly that such issue were discovered after release and to be fixed in the released version we must wait quite a long time :( |
@catap I agree, hopefully it's not two and a half years this time :( |
|
who knows. BTW I've created patch which was merged on 23.9.9 branch which allows me to easy backport it to MacPorts: macports/macports-ports#20581 So, technically you may use https://gitlab.com/embeddable-common-lisp/ecl/-/commit/e8c9355d31e15d15900416e9925db56c54c9f265 as root and enjoy ECL-23.9.9 with that patch at your CI. |
No description provided.