OCaml 5.1.0 announce and release page #1470
Conversation
|
Might be worth mentioning the significant number of new functions added to the Stdlib too, since we mention existing functions have been augmented with TMRC. |
| - Restored support for RISC-V and s390x architectures. | ||
| - Restored Cygwin port. | ||
| - Reduced installation size (50% reduction) | ||
| - Compressed compilation artefacts (.cmi, .cmt, .cmti, .cmo, .cma files) |
There was a problem hiding this comment.
I would note that all OCaml packages, not just the compiler, benefit from this.
There was a problem hiding this comment.
I have the impression that it would be a bit too heavy-handed for the brief list of highlights.
| - There are a number of known runtime concurrency bugs (that triggers under | ||
| rare circumstances) | ||
|
|
||
| We are planning to address those regressions, hopefully in the next release OCaml 5.2.0. Meanwhile, the OCaml 4.14 branch will be maintained and the next release on the OCaml 4.14 branch, OCaml 4.14.2, should follow this release in the upcoming months. |
There was a problem hiding this comment.
I would prudently say that we are planning to address some of those regressions.
There was a problem hiding this comment.
We are planning to address some of those regressions can be read as We already know that some of those regression will never be fixed. If the issue is the implicit 5.2 timeframe, I would propose:
We are planning to address those regressions, hopefully in time for the next 5.2.0 release for some of them.
?
There was a problem hiding this comment.
Have we reached consensus on all of these changes?
There was a problem hiding this comment.
Which part do you find non-consensual?
|
|
||
| Some of the highlights in OCaml 5.1.0 are: | ||
|
|
||
| - Many runtime performance regression fixes (dynlinking, weak array, weak hash sets, GC with idle domains, GC prefetching). |
There was a problem hiding this comment.
- Performance improvements: fix regressions compared to 4.x (dynlink, weak arrays and hash sets, restore GC prefetching), improve the GC scheduling to reduce memory usage in some cases (for example GC with idle domains).
There was a problem hiding this comment.
Taking in account @kit-ty-kate and your remark, it is probably better to split the item in two between the GC fix and the performance fixes.
Octachron
force-pushed
the
OCaml_5.1.0_release
branch
from
d334a0f
to
4dc0e9c
Compare
|
Just a head-up that the release for OCaml 5.1.0 is tomorrow. |
Octachron
changed the title
|
Is this ready to merge? |
|
The announce is ready to merge from my point of view. But we may wish to wait for ocaml/opam-repository#24433 to be merged. (I will ping this PR once this is the case). |
merged. The official opam repository should take < 4 hours to build so that it is available for everyone. I’d recommend waiting for that |
|
It looks like I somehow missed including the changelog, this is fixed now. |
|
all good to go |
|
Congrats on the release! |
* WIP: OCaml 5.1.0 announce * One latest release only * typo * Review: highlights * set OCaml 5.1.0 release date * review * minor grammar fixes * Minor grammar formatting fixes * OCaml 5.1.0: add missing changelog * minor formatting / grammar fixes * minor formatting change --------- Co-authored-by: Christine Rose <christinerose@users.noreply.github.com>
* WIP: OCaml 5.1.0 announce * One latest release only * typo * Review: highlights * set OCaml 5.1.0 release date * review * minor grammar fixes * Minor grammar formatting fixes * OCaml 5.1.0: add missing changelog * minor formatting / grammar fixes * minor formatting change --------- Co-authored-by: Christine Rose <christinerose@users.noreply.github.com>
This is the work-in-progress version of the announce for the upcoming release of OCaml 5.1.0.
Comments are welcome. I am particularly interested on remarks on the highlights for this version, since those highlights tend to be colored by my own interests.