From mirageos-devel-bounces@lists.xenproject.org Wed Jun 08 06:05:11 2022 Return-path: Envelope-to: archives@lists.xenproject.org Delivery-date: Wed, 08 Jun 2022 06:05:11 +0000 Received: from list by lists.xenproject.org with outflank-mailman.343692.569129 (Exim 4.92) (envelope-from ) id 1nyooN-0002wB-Rs; Wed, 08 Jun 2022 06:04:59 +0000 X-Outflank-Mailman: Message body and most headers restored to incoming version Received: by outflank-mailman (output) from mailman id 343692.569129; Wed, 08 Jun 2022 06:04:59 +0000 Received: from localhost ([127.0.0.1] helo=lists.xenproject.org) by lists.xenproject.org with esmtp (Exim 4.92) (envelope-from ) id 1nyooN-0002w4-Ox; Wed, 08 Jun 2022 06:04:59 +0000 Received: by outflank-mailman (input) for mailman id 343692; Wed, 08 Jun 2022 06:04:58 +0000 Received: from se1-gles-flk1-in.inumbo.com ([94.247.172.50] helo=se1-gles-flk1.inumbo.com) by lists.xenproject.org with esmtp (Exim 4.92) (envelope-from ) id 1nyooM-0002vy-CH for mirageos-devel@lists.xenproject.org; Wed, 08 Jun 2022 06:04:58 +0000 Received: from mail.mehnert.org (mail.mehnert.org [213.73.89.200]) by se1-gles-flk1.inumbo.com (Halon) with ESMTPS id eaedfb83-e6f0-11ec-b605-df0040e90b76; Wed, 08 Jun 2022 08:04:55 +0200 (CEST) Received: from [192.168.42.80] (dslb-188-102-156-145.188.102.pools.vodafone-ip.de [188.102.156.145]) (using TLSv1 with cipher AES128-SHA (128/128 bits)) (Client CN "hannes@mehnert.org", Issuer "mehnert root CA" (not verified)) by mail.mehnert.org (Postfix) with ESMTPS id 0B97E3BF for ; Wed, 8 Jun 2022 08:04:54 +0200 (CEST) X-BeenThere: mirageos-devel@lists.xenproject.org List-Id: Developer list for MirageOS List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Errors-To: mirageos-devel-bounces@lists.xenproject.org Precedence: list Sender: "MirageOS-devel" X-Inumbo-ID: eaedfb83-e6f0-11ec-b605-df0040e90b76 Content-Type: multipart/mixed; boundary="------------BGxTSBSjO8HyVW7c0lthHq9h" Message-ID: Date: Wed, 8 Jun 2022 08:04:53 +0200 MIME-Version: 1.0 User-Agent: Mozilla/5.0 (X11; FreeBSD amd64; rv:91.0) Gecko/20100101 Thunderbird/91.8.0 Subject: Re: no meeting today; but in two weeks Content-Language: en-US To: mirageos-devel@lists.xenproject.org References: <73b5057c-c0ec-7a20-d29f-258ad27526f2@mehnert.org> <133723f5-3673-4c7e-d13e-882603834eff@mehnert.org> From: Hannes Mehnert In-Reply-To: <133723f5-3673-4c7e-d13e-882603834eff@mehnert.org> This is a multi-part message in MIME format. --------------BGxTSBSjO8HyVW7c0lthHq9h Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit Hi, sorry I'm not able to make it today. So the next meeting will be in two weeks, June 22nd 14:00 CEST (ics attached). I'll write a reminder before that takes place. Best, Hannes On 27/05/2022 12:08, Hannes Mehnert wrote: > The next meeting will be on June 8th at 14:00 CEST (again on > https://whereby.com/ocamllabs). The agenda and minutes at > https://pad.data.coop/jgZ3NnyETWaaIT-5FM8ceA# (ics attached) --------------BGxTSBSjO8HyVW7c0lthHq9h Content-Type: text/calendar; charset=UTF-8; name="2022-06-22-mirage.ics" Content-Disposition: attachment; filename="2022-06-22-mirage.ics" Content-Transfer-Encoding: base64 QkVHSU46VkNBTEVOREFSDQpQUk9ESUQ6LS8vTW96aWxsYS5vcmcvTk9OU0dNTCBNb3ppbGxh IENhbGVuZGFyIFYxLjEvL0VODQpWRVJTSU9OOjIuMA0KQkVHSU46VlRJTUVaT05FDQpUWklE OkV1cm9wZS9QYXJpcw0KQkVHSU46REFZTElHSFQNClRaT0ZGU0VURlJPTTorMDEwMA0KVFpP RkZTRVRUTzorMDIwMA0KVFpOQU1FOkNFU1QNCkRUU1RBUlQ6MTk3MDAzMjlUMDIwMDAwDQpS UlVMRTpGUkVRPVlFQVJMWTtCWURBWT0tMVNVO0JZTU9OVEg9Mw0KRU5EOkRBWUxJR0hUDQpC RUdJTjpTVEFOREFSRA0KVFpPRkZTRVRGUk9NOiswMjAwDQpUWk9GRlNFVFRPOiswMTAwDQpU Wk5BTUU6Q0VUDQpEVFNUQVJUOjE5NzAxMDI1VDAzMDAwMA0KUlJVTEU6RlJFUT1ZRUFSTFk7 QllEQVk9LTFTVTtCWU1PTlRIPTEwDQpFTkQ6U1RBTkRBUkQNCkVORDpWVElNRVpPTkUNCkJF R0lOOlZFVkVOVA0KVUlEOjQzOWU0MDgzLWFiZTMtNDRiZC1iODU2LTJlMmQ5NTIyMTE4Ng0K RFRTVEFNUDoyMDIyMDYwOFQwNjAzMDhaDQpEVFNUQVJUO1RaSUQ9RXVyb3BlL1BhcmlzOjIw MjIwNjIyVDE0MDAwMA0KRFRFTkQ7VFpJRD1FdXJvcGUvUGFyaXM6MjAyMjA2MjJUMTUwMDAw DQpDUkVBVEVEOjIwMjIwNTI3VDEwMDMzMFoNCkxBU1QtTU9ESUZJRUQ6MjAyMjA2MDhUMDYw MzA4Wg0KU1VNTUFSWTptaXJhZ2VvcyBtZWV0aW5nDQpUUkFOU1A6T1BBUVVFDQpMT0NBVElP TjpodHRwczovL3doZXJlYnkuY29tL29jYW1sbGFicw0KREVTQ1JJUFRJT047QUxUUkVQPSJk YXRhOnRleHQvaHRtbCwlM0Nib2R5JTNFaHR0cHM6JTJGJTJGcGFkLmRhdGEuY29vcCUyRmpn WjNObnlFVFdhYUlULTVGTThjZUElMjMlM0MlMkZib2R5JTNFIjpodHRwczovL3BhZC5kYXRh LmNvb3AvamdaM05ueUVUV2FhSVQtNUZNOGNlQSMNClNFUVVFTkNFOjENClgtTU9aLUdFTkVS QVRJT046MQ0KQkVHSU46VkFMQVJNDQpBQ1RJT046RElTUExBWQ0KVFJJR0dFUjtWQUxVRT1E VVJBVElPTjotUFQ1TQ0KREVTQ1JJUFRJT046RGVmYXVsdCBNb3ppbGxhIERlc2NyaXB0aW9u DQpFTkQ6VkFMQVJNDQpFTkQ6VkVWRU5UDQpFTkQ6VkNBTEVOREFSDQo= --------------BGxTSBSjO8HyVW7c0lthHq9h-- From mirageos-devel-bounces@lists.xenproject.org Wed Jun 22 13:23:19 2022 Return-path: Envelope-to: archives@lists.xenproject.org Delivery-date: Wed, 22 Jun 2022 13:23:19 +0000 Received: from list by lists.xenproject.org with outflank-mailman.353879.580877 (Exim 4.92) (envelope-from ) id 1o40K5-0000Ia-FB; Wed, 22 Jun 2022 13:23:09 +0000 X-Outflank-Mailman: Message body and most headers restored to incoming version Received: by outflank-mailman (output) from mailman id 353879.580877; Wed, 22 Jun 2022 13:23:09 +0000 Received: from localhost ([127.0.0.1] helo=lists.xenproject.org) by lists.xenproject.org with esmtp (Exim 4.92) (envelope-from ) id 1o40K5-0000IT-Bv; Wed, 22 Jun 2022 13:23:09 +0000 Received: by outflank-mailman (input) for mailman id 353879; Wed, 22 Jun 2022 13:23:07 +0000 Received: from se1-gles-flk1-in.inumbo.com ([94.247.172.50] helo=se1-gles-flk1.inumbo.com) by lists.xenproject.org with esmtp (Exim 4.92) (envelope-from ) id 1o40K3-0000IJ-Lz for mirageos-devel@lists.xenproject.org; Wed, 22 Jun 2022 13:23:07 +0000 Received: from mail-io1-xd2d.google.com (mail-io1-xd2d.google.com [2607:f8b0:4864:20::d2d]) by se1-gles-flk1.inumbo.com (Halon) with ESMTPS id 725fd4fe-f22e-11ec-b725-ed86ccbb4733; Wed, 22 Jun 2022 15:23:06 +0200 (CEST) Received: by mail-io1-xd2d.google.com with SMTP id s17so11246185iob.7 for ; Wed, 22 Jun 2022 06:23:04 -0700 (PDT) X-BeenThere: mirageos-devel@lists.xenproject.org List-Id: Developer list for MirageOS List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Errors-To: mirageos-devel-bounces@lists.xenproject.org Precedence: list Sender: "MirageOS-devel" X-Inumbo-ID: 725fd4fe-f22e-11ec-b725-ed86ccbb4733 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20210112; h=mime-version:references:in-reply-to:from:date:message-id:subject:to :cc; bh=PhLlbpkQ4T+lWaLu4R+r9hem3IYO6XyR4/NsJjmreSY=; b=SroCDT0VorquT1/fgDtdhpvxljnKMLC67+4tUiUYlxpTjkbgkhmmBR31rnXpdAgIJY MhuxbkJqwSbAiFWgiq12u/MxMlT0ZDWzmNpKUF2d/v5HlwfYCCiWZcTtvQmGGRcqMcVI d15Ztp1IFbNzqfpsacKOa0nugO/nWfCOh2scIJk/FPZ4iA3aRHfJ1OKxq/u1kyWuqL5O PGEl+f1haOu78MWZrruEuH18mzKGfTalu/pqxbQ79Ewpf2qAYCoo1Et/EdK9wDTxdyDo Qvrs5uDlUEvXNDOld6CmQDjWjzXS9z4WNSbUXUl072y5jjMGCv9FtXD2PLCih+1OcVTb +zAg== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=x-gm-message-state:mime-version:references:in-reply-to:from:date :message-id:subject:to:cc; bh=PhLlbpkQ4T+lWaLu4R+r9hem3IYO6XyR4/NsJjmreSY=; b=oUD34g6iQMYragCT+vOtIEaXL+6lUVnUiUuPmfJZmvU9u1JtDH/ACyuT+iv+jtuosp YF/U21GSmf0E7PIVexGYfoBD83ao5RBisfYTP6E4/xMWQ3CtVCM6miWgg5KXFa/NqmN7 +ihVpwuY9qCp6zcrXKx+2C8g+HqonLk2nUTyzsYk1xojsYB5rOfr0yU1TMSqfjxRtHnO oAgQOddXlUhHpIAXcadraakK1ae1opJPOwECp4x5KpYybed/OsEUEjXv14k7u4tGaM0h yjjsGA6f+FoMOD6xyzJHyomJ+S/s6BP7eGkOk9/T/AXk3Ez3KGW5H5F185//FdW6m+Jq Oadg== X-Gm-Message-State: AJIora90YSXfztpzEDHr1387gV093g6opsB6JscQ6Gp85l2E791hko9l 5wTaShkNwVwtWxtFt1A9c0XsHFA1oocgl249qv/Je5VxFTA= X-Google-Smtp-Source: AGRyM1ui4xpiVV4xVBREXFAUAXxrLYOJPElMDzEiFsKMJ+YmRfldfB+JcpD7zEYrAkXd8VVIHSpNrWLvUJ6JDBHSj2w= X-Received: by 2002:a02:caa9:0:b0:339:e1a6:387e with SMTP id e9-20020a02caa9000000b00339e1a6387emr705130jap.97.1655904183660; Wed, 22 Jun 2022 06:23:03 -0700 (PDT) MIME-Version: 1.0 References: <73b5057c-c0ec-7a20-d29f-258ad27526f2@mehnert.org> <133723f5-3673-4c7e-d13e-882603834eff@mehnert.org> In-Reply-To: From: Romain Calascibetta Date: Wed, 22 Jun 2022 15:22:52 +0200 Message-ID: Subject: Re: no meeting today; but in two weeks To: Hannes Mehnert Cc: mirageos-devel@lists.xenproject.org Content-Type: multipart/alternative; boundary="000000000000a636f605e2093c6f" --000000000000a636f605e2093c6f Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable Hi, this is what we said at the meeting (2022-06-22) and we scheduled another meeting in 2 weeks (2022-07-06 14:00 - 15:00 (CEST)). Thanks for your participation. Sorry for my english and if I missed something, please tell us through this email :) ! Enjoy. PS: I will let Hannes send another email to schedule the next meeting. # Mirage 2022-06-22 ## Presentation round reynir: I'm in robur :) pierre: I'm Pierre, CS teacher (in France, Bretagne), I recently joined MirageOS. I want to use it and teach new things to my students. thomas: continue to work on eio and OCaml 5 stuff. I plugged eio into lwt and async. It should be possible to plug into MirageOS timada: from Japan, working on Solo5 and microcontrollers, I finished implementing 40% percent of tenders. I look into details. ## What is going on about mirage romain: Hannes did a PR about reproducible builds and added few meta-informations into the generated OPAM files to let `orb` to build unikernels: https://github.com/mirage/mirage/pull/1335 Some works was done about `ocaml-solo5` (thx to Pierre and Hannes) to provides some metrics about allocations on the C heap (with `dlmalloc`): https://github.com/mirage/ocaml-solo5/pull/120 A release of `ocaml-solo5` was done and a pending PR exists about `mirage-xen` to be able to get these informations: https://github.com/mirage/mirage-xen/pull/48 All of that is related to some issues found about `qubes-firewall` and Solo5, you can see the CHANGES.md of Solo5 here: https://github.com/Solo5/solo5/pull/518 Waiting to merge the reproducible build PR on MirageOS 4 and be able to: 1) do a release of `mirage-xen` (which includes allocation metrics) 2) do a release of `mirage` with right bounds about `mirage-xen`/`ocaml-solo5` (because the API was updated) This release will be done probably in a few weeks. ## Job rounds ### Pierre round pierre: question about memory consumption and mirage-xen PR ( https://github.com/mirage/mirage-xen/pull/48) should we merge what we did on `mirage-xen` into `mirage-solo5` (because they use the same things, eg. solo5) romain: for me it's good to do pierre:I did an unikernel which implements a NAT router into the Tor network, https://github.com/palainp/torrefacteur thomas: it's like an unix gateway? pierre: I would like to replace the unix gateway by a VM, so yes be able to route some traffic on default gateway and route some traffic on tor thomas: will be a useful unikernel reynir: Hannes made a PR (https://github.com/mirage/ocaml-conduit/pull/414) into conduit which should fix the issue about the Tor unikernel pierre: I saw the fix but I'm not sure how to use this fix. I think I need to work on something on what Hannes wrote on this PR pierre: I also have a question. I wrote a code to work with ESP32 to be able to use it as a wifi card with the PPP protocol. Then, I can use this controller to process the wifi. I have a question about Gilbraltar (RPi4) which does not yet have a network. PPP interesting? timada: if you want to use TCP/IP with MirageOS on a micro-controller, you need to have both a TCP/IP handling layer on MirageOS and also basically a packet transmitting/receiving layer which is provided on the C layer. For instance, the packet transmitting/receiving layer is implemented by the Linux operating system for unix, seccomp. For a microcontroller, you must have something which handles that. romain: for RPi4, the difficulty is about the Ethernet driver which is missing from Gilbraltar. But if someone wants to implement the Ethernet driver, it's just a matter of plugging `mirage-tcpip` then and you have a full network stack. More concretely, you need to deal with *mbox* and interrupts to be able to implement the ethernet driver, a PR exists on Gilbraltar which handles interrupts from the CPU to the OCaml side ( https://github.com/dinosaure/gilbraltar/pull/19). From it, you should be able to implement an Ethernet driver in OCaml may be (with few C codes) timada: https://mcuxpresso.nxp.com/api_doc/dev/329/group__lpc__enet.html this is an ethernet driver target to bare-metal and it does not depends to highly complexe Linux things, it can be interesting. I think it's difficult to port the ethernet driver from Linux to bare-metal. romain: Yeah. the ethernet driver for RPi4 provided by Linux is too complex for us and require too many things but I found a bare-metal ethernet driver too in C++ (https://github.com/rsta2/circle/blob/master/lib/bcm54213.cpp) (BCM54213PE PHY). From a quick look, it should not be super difficult to implement it. timada: if we use a small microcontroller, in many cases the microcontroller provides something for ethernet/wifi. And you probably can use them. Like ESP32 provides that. ### Romain romain: try to implement a peer-to-peer file-transfer with a simple tool and a unikernel as a relay. Implement the protocol with SPAKE2+EE ( https://github.com/dinosaure/spoke) and the tool ( https://github.com/dinosaure/bob) # Thomas thomas: Look into the eio supports on MirageOS 4, I would like to upgrade mirage-skeleton with eio. Be able to use the direct-style on some of them. # Reynir reynir: Caldav code, otherwise I work on builder/reproducible build romain: maybe you can talk about caldav reynir: It's an unikernel which implements a CalDAV server with the CalDAV protocol. Git is used as the store. You can create users and multiple calendars. romain: we currently use it (4 peoples) to find some bugs by usage and improve it. # Timada timada: Solo5 on microcontrollers, I also need to understand the muen Solo5. Especially the muen framework. The new building framework is different from the previous one. Look into `ocaml-solo5`, `solo5` romain: yeah the new layout of Solo5 is about _toolchains_. Solo5 provides a new toolchain (like `cc`, `ld` and `objcopy`) which is a wrapper with a compiler (like `gcc` or `clang`) or a cross-compiler (like `aarch64-gcc`). Then, `ocaml-solo5` uses this C toolchain to compile the _caml runtime_ and provide a new OCaml _toolchain_. At the end, for the mirage tool and depending on what the user wants, we choose the right OCaml _toolchain_. timada: What about `opam monorepo`? romain: Yeah, it's mostly about (transitive?) dependencies which include C stubs. We must compile them with certain flags (like `-ffreestanding` or `-nostdlib`). These flags will be provided by the OCaml toolchain which use the correct C toolchain (the `cc`/`ld`/`objcopy` provided by Solo5 first). By this way, we compile everything needed by the unikernel into a coherent `dune`'s context. timada: I'm currently testing my test compilation with a very simple Makefile. It does not care about `opam-monorepo`. I will look into that. Le mer. 8 juin 2022 =C3=A0 08:05, Hannes Mehnert a =C3= =A9crit : > Hi, > > sorry I'm not able to make it today. So the next meeting will be in two > weeks, June 22nd 14:00 CEST (ics attached). I'll write a reminder before > that takes place. > > > Best, > > Hannes > > > On 27/05/2022 12:08, Hannes Mehnert wrote: > > The next meeting will be on June 8th at 14:00 CEST (again on > > https://whereby.com/ocamllabs). The agenda and minutes at > > https://pad.data.coop/jgZ3NnyETWaaIT-5FM8ceA# (ics attached) --=20 Romain Calascibetta - http://din.osau.re/ --000000000000a636f605e2093c6f Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable
Hi, this is what we said at the meeting (2022-06-22) = and we scheduled another meeting in 2 weeks (2022-07-06 14:00 - 15:00 (CEST= )). Thanks for your participation.
Sorry for my english and if I = missed something, please tell us through this email :) ! Enjoy.
<= br>
PS: I will let Hannes send another email to schedule the next= meeting.

# Mirage 2022-06-22

## Presen= tation round

reynir: I'm in robur :)

pierre: I'm Pier= re, CS teacher (in France, Bretagne), I recently joined MirageOS. I want to= use it and teach new things to my students.

thomas: continue to wor= k on eio and OCaml 5 stuff. I plugged eio into lwt and async. It should be = possible to plug into MirageOS

timada: from Japan, working on Solo5 = and microcontrollers, I finished implementing 40% percent of tenders. I loo= k into details.

## What is going on about mirage

romain: Hann= es did a PR about reproducible builds and added few meta-informations into = the generated OPAM files to let `orb` to build unikernels: https://github.com/mirage/mirage/pul= l/1335
Some works was done about `ocaml-solo5` (thx to Pierre and Ha= nnes) to provides some metrics about allocations on the C heap (with `dlmal= loc`): https://g= ithub.com/mirage/ocaml-solo5/pull/120
A release of `ocaml-solo5` was= done and a pending PR exists about `mirage-xen` to be able to get these in= formations: https:= //github.com/mirage/mirage-xen/pull/48
All of that is related to som= e issues found about `qubes-firewall` and Solo5, you can see the CHANGES.md= of Solo5 here: https:/= /github.com/Solo5/solo5/pull/518
Waiting to merge the reproducible b= uild PR on MirageOS 4 and be able to:
1) do a release of `mirage-xen` (w= hich includes allocation metrics)
2) do a release of `mirage` with right= bounds about `mirage-xen`/`ocaml-solo5` (because the API was updated)
<= br>This release will be done probably in a few weeks.

## Job rounds

### Pierre round

pierre: que= stion about memory consumption and mirage-xen PR (https://github.com/mirage/mirage-xen/pull/4= 8) should we merge what we did on `mirage-xen` into `mirage-solo5` (bec= ause they use the same things, eg. solo5)
romain: for me it's good t= o do
pierre:I did an unikernel which implements a NAT router into the To= r network, https://gith= ub.com/palainp/torrefacteur
thomas: it's like an unix gateway?pierre: I would like to replace the unix gateway by a VM, so yes be able = to route some traffic on default gateway and route some traffic on tor
t= homas: will be a useful unikernel
reynir: Hannes made a PR (https://github.com/mirage/= ocaml-conduit/pull/414) into conduit which should fix the issue about t= he Tor unikernel
pierre: I saw the fix but I'm not sure how to use t= his fix. I think I need to work on something on what Hannes wrote on this P= R
pierre: I also have a question. I wrote a code to work with ESP32 to b= e able to use it as a wifi card with the PPP protocol. Then, I can use this= controller to process the wifi. I have a question about Gilbraltar (RPi4) = which does not yet have a network. PPP interesting?
timada: if you want = to use TCP/IP with MirageOS on a micro-controller, you need to have both a = TCP/IP handling layer on MirageOS and also basically a packet transmitting/= receiving layer which is provided on the C layer. For instance, the packet = transmitting/receiving layer is implemented by the Linux operating system f= or unix, seccomp. For a microcontroller, you must have something which hand= les that.
romain: for RPi4, the difficulty is about the Ethernet driver = which is missing from Gilbraltar. But if someone wants to implement the Eth= ernet driver, it's just a matter of plugging `mirage-tcpip` then and yo= u have a full network stack. More concretely, you need to deal with *mbox* = and interrupts to be able to implement the ethernet driver, a PR exists on = Gilbraltar which handles interrupts from the CPU to the OCaml side (https://github.com/din= osaure/gilbraltar/pull/19). From it, you should be able to implement an= Ethernet driver in OCaml may be (with few C codes)
timada: https://mc= uxpresso.nxp.com/api_doc/dev/329/group__lpc__enet.html this is an ether= net driver target to bare-metal and it does not depends to highly complexe = Linux things, it can be interesting. I think it's difficult to port the= ethernet driver from Linux to bare-metal.
romain: Yeah. the ethernet dr= iver for RPi4 provided by Linux is too complex for us and require too many = things but I found a bare-metal ethernet driver too in C++ (https://github.co= m/rsta2/circle/blob/master/lib/bcm54213.cpp) (BCM54213PE PHY). From a q= uick look, it should not be super difficult to implement it.
timada: if = we use a small microcontroller, in many cases the microcontroller provides = something for ethernet/wifi. And you probably can use them. Like ESP32 prov= ides that.

### Romain

romain: try to implement a peer-to-peer= file-transfer with a simple tool and a unikernel as a relay. Implement the= protocol with SPAKE2+EE (ht= tps://github.com/dinosaure/spoke) and the tool (https://github.com/dinosaure/bob)

# Thomas=

thomas: Look into the eio supports on MirageOS 4, I would like to u= pgrade mirage-skeleton with eio. Be able to use the direct-style on some of= them.

# Reynir

reynir: Caldav code, otherwise I work on buil= der/reproducible build
romain: maybe you can talk about caldav
reynir= : It's an unikernel which implements a CalDAV server with the CalDAV pr= otocol. Git is used as the store. You can create users and multiple calenda= rs.
romain: we currently use it (4 peoples) to find some bugs by usage a= nd improve it.

# Timada

timada: Solo5 on microcontrollers, I = also need to understand the muen Solo5. Especially the muen framework. The = new building framework is different from the previous one. Look into `ocaml= -solo5`, `solo5`
romain: yeah the new layout of Solo5 is about _toolchai= ns_. Solo5 provides a new toolchain (like `cc`, `ld` and `objcopy`) which i= s a wrapper with a compiler (like `gcc` or `clang`) or a cross-compiler (li= ke `aarch64-gcc`). Then, `ocaml-solo5` uses this C toolchain to compile the= _caml runtime_ and provide a new OCaml _toolchain_. At the end, for the mi= rage tool and depending on what the user wants, we choose the right OCaml _= toolchain_.
timada: What about `opam monorepo`?
romain: Yeah, it'= s mostly about (transitive?) dependencies which include C stubs. We must co= mpile them with certain flags (like `-ffreestanding` or `-nostdlib`). These= flags will be provided by the OCaml toolchain which use the correct C tool= chain (the `cc`/`ld`/`objcopy` provided by Solo5 first). By this way, we co= mpile everything needed by the unikernel into a coherent `dune`'s conte= xt.
timada: I'm currently testing my test compilation with a very si= mple Makefile. It does not care about `opam-monorepo`. I will look into tha= t.

Le=C2=A0mer. 8 juin 2022 =C3=A0=C2=A008:05, Hannes Mehnert <hannes@mehnert.org> a =C3=A9crit= =C2=A0:
Hi,

sorry I'm not able to make it today. So the next meeting will be in two=
weeks, June 22nd 14:00 CEST (ics attached). I'll write a reminder befor= e
that takes place.


Best,

Hannes


On 27/05/2022 12:08, Hannes Mehnert wrote:
> The next meeting will be on June 8th at 14:00 CEST (again on
> https://whereby.com/ocamllabs). The agenda and minutes at
> https://pad.data.coop/jgZ3NnyETWaaIT-5FM8ceA# = (ics attached)


--
Romain Calascibetta - http://din.osau.re/
--000000000000a636f605e2093c6f--