OCaml Weekly News

Previous Week<= /a> Up Next Week

Hello

Here is the latest OCaml Weekly News, for the week of November 05 to 12, 20= 24.

Picos =E2=80=94 Interoperable effects based concurrency<= /a>

Picos =E2=80=94 Interoperable effects based concurrency

Archive: https://discuss.ocaml.org/t/ann-picos-inter= operable-effects-based-concurrency/14507/3

polytypic announced

I'm happy to announce that Picos version 0.6.0 has been released!

Picos is a systems programming interface between effects based schedulers a= nd concurrent abstractions.

A lot of work has been done on Picos since previous announcements.

You might start on the new minimalist landing page for Picos, which, among other things, allo= ws you to access the documentation of all the released Picos versions.

Also, in case you missed it, a recording of the talk

Picos =E2=80=94 Interoperable effects= based concurrency

can be found here.

We also held a workshop on concurrency and parallelism at Fun OCaml. You might enjoy trying out the exercise we develope= d for the workshop.

As, for reasons of dependencies, Picos now comes in no less than 8 packages and multipl= e libraries, here is a summary of the packages and the libraries inside eac= h package:

picos =E2=80=94 Picos =E2=80=94 Interoperable effects = based concurrency
- picos =E2=80=94 A systems programming interface = between effects based schedulers and concurrent abstractions
- picos.domain =E2=80=94 Minimalistic domai= n API available both on OCaml 5 and on OCaml 4
- picos.thread =E2=80=94 Minimalistic threa= d API available with or without threads.posix
picos_mux =E2=80=94 Sample schedulers for Picos
- picos_mux.fifo =E2=80=94 Basic sing= le-threaded effects based Picos compatible scheduler for OCaml 5
- picos_mux.multififo =E2=80=94 = Basic multi-threaded effects based Picos compatible scheduler for OCaml 5
- picos_mux.random =E2=80=94 Random= ized multi-threaded effects based Picos compatible scheduler for OCaml 5
- picos_mux.thread =E2=80=94 Basic = Thread based Picos compatible scheduler for OCaml 4
picos_std =E2=80=94 Sample libraries for Picos
- picos_std.finally =E2=80=94 Synt= ax for avoiding resource leaks for Picos
- picos_std.awaitable =E2=80=94 = Basic futex-like awaitable atomic location for Picos
- picos_std.event =E2=80=94 Basic ev= ent abstraction for Picos
- picos_std.structured =E2=80= =94 Basic structured concurrency primitives for Picos
- picos_std.sync =E2=80=94 Basic comm= unication and synchronization primitives for Picos
picos_io =E2=80=94 Asynchronous IO system for Picos
- picos_io =E2=80=94 Basic IO facilities bas= ed on OCaml standard libraries for Picos
- picos_io.select =E2=80=94 Basic Uni= x.select based IO event loop for Picos
- picos_io.fd =E2=80=94 Externally refere= nce counted file descriptors
picos_io_cohttp =E2=80=94 Cohttp running on = Picos IO
- picos_io_cohttp =E2=80=94 Mi= nimalistic Cohttp implementation using Picos_io for Picos
picos_lwt =E2=80=94 Lwt interface for Picos
- picos_lwt =E2=80=94 Direct style Picos c= ompatible interface to Lwt for OCaml 5
- picos_lwt.unix =E2=80=94 Direct sty= le Picos compatible interface to Lwt with Lwt_unix for OCaml 5
picos_aux =E2=80=94 Auxiliary libraries for Picos
- picos_aux.htbl =E2=80=94 Lock-free = hash table
- picos_aux.mpmcq =E2=80=94 Lock-fre= e multi-producer, multi-consumer queue
- picos_aux.mpscq =E2=80=94 Lock-fre= e multi-producer, single-consumer queue
- picos_aux.rc =E2=80=94 External refer= ence counting tables for disposable resources
picos_meta =E2=80=94 Integration tests for Picos = packages

In addition to the above, Mo= onpool now uses Picos underneath.

And, I almost forgot, there is a ready to be merged PR for Kcas to change it to use Picos= . You should be able to try it with an opam pin-depends.

findlib-1.9.7

Archive: https://sympa.inria.fr/sympa/arc/caml-list/2024-11/msg00004.html=

Gerd Stolpmann announced

findlib-1.9.7 is out. This is mostly a bugfix release. There is now also some support for relocability (driven by environment variables), contributed by Marek Kubica.

For manual, download, manuals, etc. see here:

http://proje= cts.camlcity.org/projects/findlib.html

An updated OPAM package will follow soon.

First release candidate for OCaml 5.2.1

Archive: https://discuss.ocaml.org/t/first-release-candidate-f= or-ocaml-5-2-1/15578/1

octachron announced

The release of OCaml version 5.2.1 is imminent.

OCaml 5.2.1 is a collection of safe but import runtime time bug fixes backp= orted from the 5.3 branch of OCaml. The full list of bug fixes is available= below.

In order to ensure that the future release works as expected, we are planni= ng to test a release candidate during the upcoming week.

If you find any bugs, please report them here on GitHub.

Installation Instructions

The base compiler can be installed as an opam switch with the following com= mands on opam 2.1:

opam update
opam switch create 5.2.1~rc1

The source code for the release candidate is available on

Fine-Tuned Compiler Configuration
If you want to tweak the configuration of the compiler, you can switch to t= he option variant with:
```
opam update
opam switch create <switch_name> ocaml-variants.5.2.1~rc1+options <=
;option_list>
```
where `<option_list>` is a space-separated list of `ocaml-option-*` p= ackages. For instance, for a `flambda` and `no-flat-float-array` switch:
```
opam switch create 5.2.1~rc1+flambda+nffa ocaml-variants.5.2.1~rc1+options =
ocaml-option-flambda ocaml-option-no-flat-float-array
```
All available options can be listed with `opam search ocaml-option`.

Changes Since OCaml 5.2.0

Runtime System:

#13207: Be = sure to reload the register caching the exception handler in caml_c_call and caml_c_call_stack_args, as its value may have been changed if the OCaml stack is expanded during a callback. (Miod Vallat, report by Vesa Karvonen, review by Gabriel Scherer and Xavier Leroy)
#13252: Rew= ork register assignment in the interpreter code on m68k on Linux, due to the %a5 register being used by Glibc. (Miod Vallat, report by St=C3=A9phane Glondu, review by Gabriel Scherer and Xavier Leroy)
#13268: Fix= a call to test in configure.ac that was causing errors when LDFLAGS contains several words. (St=C3=A9phane Glondu, review by Miod Vallat)
#13234, #13267: Open runti= me events file in read-write mode on armel (armv5) systems due to atomic operations limitations on that platform. (St=C3=A9phane Glondu, review by Miod Vallat and Vincent Laviron)
#13188: fix= races in the FFI code coming from the use of Int_val(…) on rooted values inside blocking questions / without the runtime lock. (Calling Int_val(…) on non-rooted immediates is fine, but any access to rooted values must be done outside blocking sections / with the runtime lock.) (Etienne Millon, review by Gabriel Scherer, Jan Midtgaard, Olivier Nicole)<= /li>
#13318: Fix= regression in GC alarms, and fix them for flambda. (Guillaume Munch-Maccagnoni, report by Benjamin Monate, review by Vincent Laviron and Gabriel Scherer)
#13140: POW= ER back-end: fix issue with call to `caml_call_realloc_stack` from a DLL (Xavier Leroy, review by Miod Vallat)
#13370: Fix= a low-probability crash when calling Gc.counters. (Demi Marie Obenour, review by Gabriel Scherer)
#13402, #13512, #13549, #13553: Revise bytecode implemen= tation of callbacks so that it no longer produces dangling registered bytecode fragments. (Xavier Leroy, report by Jan Midtgaard, analysis by Stephen Dolan, review by Miod Vallat)
#13502: Fix= misindexing related to `Gc.finalise_last` that could prevent finalisers from being run. (Nick Roberts, review by Mark Shinwell)
#13520: Fix= compilation of native-code version of systhreads. Bytecode fields were being included in the thread descriptors. (David Allsopp, review by S=C3=A9bastien Hinderer and Miod Vallat)

mirage-swapfs

Archive: = https://discuss.ocaml.org/t/ann-mirage-swapfs/15583/1

Reynir Bj=C3=B6rnsson announced

I am pleased to announce the first release of mirage-swapfs (swapfs on opam). It is an exp= erimental library to use a mirage block device for ephemeral, append-only, = anonymous "files". It was developed for use cases such as in opam-mirror where opam package= source archives are downloaded. The files are first downloaded to "swap" a= nd if the download succeeds and the checksum is as expected the data is the= n copied over to the tar filesystem.

Internally it uses a weak pointer array (Weak.t) to map "block= " offsets to handles. The idea is the garbage collector can help us free up= "blocks" if the user forgets to explicitly free the handle. A "block" is (= configurable, see blocking_factor) multiple of sectors in orde= r to reduce bookkeeping overhead. With a sector size of 512 bytes the defau= lt is 1 MiB per block.

I would be interested to hear about other ideas or approaches.

Dune dev meeting

Archive: https://discuss.ocaml.org/t/ann-dune-dev-meeting/14994/16

Etienne Marais announced

We will hold our regular Dune dev meeting tomorrow, on Wednesday, November,= 13th at 9:00 CET. :warning: Note that the session has been moved one hour earlier. As usual, the session will be one hour long.

Whether you are a maintainer, a regular contributor, a new joiner or just c= urious, you are welcome to join: these discussions are opened! The goal of = these meetings is to provide a place to discuss the ongoing work together a= nd synchronise between the Dune developers !:smile:

:calendar: Agenda

The agenda is available on the meeting dedicated page. Feel free to ask if y= ou want to add more items in it.

:computer: Links

Meeting link: zoom
Calendar event: google calendar
Wiki with information and previous notes: GitHub Wiki

Other OCaml News

From the ocaml.org blog

Here are links from many OCaml blogs aggregated at the ocaml.org blog.

Old CWN

If you happen to miss a CWN, you can send me a message and I'll mail it to you, or go take a loo= k at the archive or the <= a href=3D"https://alan.petitepomme.net/cwn/cwn.rss">RSS feed of the archive= s.

If you also wish to receive it every week by mail, you may subscribe to the= caml-list.

Alan Schmitt