 All right. Good morning. Welcome to the first session of Automotive Linux Summit 2023. My name is Walt Minor. I'm the AGL community manager. And I'm going to give an update on what AGL has been doing this year and a little bit of what we have planned for next year. I know Dan Kaushy already gave some of the information during the keynote, but hopefully I've got some new information. So first off, who is this guy? So I've been the AGL community manager now since 2014, which is to me a really long time. I've worked in Linux mostly since 2005 or so when I started working at Motorola mobile devices, Linux Java phones. We were putting out the first Linux mobile phones in the world back then. And I worked at mentor graphics. I worked at some tier one suppliers and I did some defense electronics. And those are some pictures of me on some motorcycle trips, one to Yosemite and one with my new BMW motorcycle that I got this year. I had my motorcycles thing here on my lanyard, but I forgot it today. So if you want to talk about motorcycles, I'm here for you. So last year when we were in Japan, my colleague, John Simon and I, we took sushi classes. So now I'm an apprentice sushi master. The thing is, whenever I live in Asheville, North Carolina, and there's a great need for sushi masters there, but whenever I try to apply, they tell me I need advanced training. So I'm still looking for the advanced training. I was hoping to take another class this year, but I haven't had time yet. So this is kind of an overall timeline of what we accomplished this year. And we still have a few more milestones. So as you probably know, we name all of our releases after fish. And the beginning of the year, we made our optimistic octopus release. We did about five releases, five patch releases after that. We did our prickly pike release in the middle of the year. And then coming up at the beginning of next year in February, we'll have our prickly pike release, I'm sorry, quirky quillback release coming out. And we just are releasing our first milestone release, making our first milestone release on the way to making the final release in February. So prickly pike, which was again, it was the July release. I'm going to explain our relationship with Yachto, but basically the AGL is Yachto-based. All of our code is built on top of Yachto, Paki, and Open Embedded. And we've been using the Kirkstone, the long-term support release of Yachto now for a few years. And so prickly pike originally had the 4.0.10 release of Yachto, and we've updated it in later patch versions. We included Rust for the first time, because we needed that for the Cuxit.Val data broker. So that Cuxit.Val is an implementation of the CoVisa project's vehicle signal specification. Cuxit.Val was created and is maintained by Bosch. And so we use that now for our intra-vehicle communication and for our vehicle-to-cloud communication, as I'll talk about that in a few minutes. So we made a lot of general cleanup. We included some new VertIO features. We made some demo image changes. So it was a pretty busy release. You can get the full details on our release notes. And we added Chromium Embedded Framework. So we were using a WebOS open-source edition version of Chromium. And that was a very slow path of update from LG. So we decided to move on to Chromium Embedded Framework, which is a more modern implementation of using Chromium. And so that was now as a technology demo in the pike release. So this is just kind of a more detailed view of what we've done in pike. Every time we do a pike update, we have also updated the Yachto version, so that 4.0.10 has moved up as we've gone along here. And we have another 16.03 coming out in another couple weeks. That's pretty much ready to go. We just need to get it through QA. So Corky Quillback. So Corky Quillback is coming out in February. We are making the milestone one release. We had a little hang up with QA, and we found some issues when we wanted to get some extra features in. But milestone one is about to come out this week. As Dan mentioned, we're using Flutter. The Flutter Embedder has been greatly improved throughout this year. And the Flutter Embedder is now pretty up to date with baseline Flutter, upstream Flutter. It's now using version 3.13.9. We've made some, we made a lot of improvements to our Chromium Embedded Framework. Chromium implementation. There's a talk later today from, today or tomorrow, from Eagalia about talking about what they've done to the Chromium Embedded Framework. Again, we're much further up, so we're much further along on the upstream Chromium path. I think in the last six months, we've jumped from a version, a version in the 90s up to version 118. So we've made rapid improvement compared to where we used to be in terms of keeping up with upstream. So we're actually going to deprecate that old version of the WebOS Chromium and remove the support completely. In connectivity, we've updated to the latest version of Cuxa.Val, which is 0.42. We now use the Cuxa.Val data broker, and we're basically converting all of our APIs to use GRPC. And then a really cool thing that happened this year, something I've been asking for, for years, is for someone to create a demo, what we call a demo control panel. You can run it on a Raspberry Pi, you can run it on a PC, on a QMU emulator, but basically it allows you to control can signals or vehicle signals going into the vehicle from a remote device. And then you can either write a script or you can toggle switches on the can simulator. And that was all done by, we have a, we sponsor every year a few Google Summer of Code students. And one of our GSOC students, so Jimtin wrote that. Our other GSOC student wrote some new open source speech recognition plugins for us. So you'll be able to see that demo control panel. Tomorrow there'll be a GSOC update. So our GSOC students will be giving a talk. I really encourage you to take a look at what they've done. They've really did some really amazing work in a short amount of time this summer. And that control panel will be very useful for us for the next few years. I think everybody who's tried it out has really liked it. Audio, we've done a lot of audio work, basically updating to the new pipe wire and wire plumber 1.0. So wire plumber and pipe wire finally released their 1.0 version very, very recently. We've gotten that onto our main, our main branch. Our mixer API was now, is now being updated to use GRPC. And we added controls, specific controls for base, treble, balance, and fader. Previously, we only had very detailed controls for more like a graphic equalizer. But this allows people to think more in terms of, you know, the old school knobs you would see on our radio. We didn't really have that available in pipe wire and wire plumber before. For quillback, we're adding some new BSPs. We've got risk five supported in HGL now. Sci five has an unmatched board. And they're actually just built another 1000 of the unmatched boards. It's called unmatched RevB. Those are now available. You can go purchase those. We'll be supporting the Unrav, unmatched RevB as well as the RevA. I've got about eight of those coming in for HGL to use. The new Beagle Bone AI64 and the Beagle Play are also going to be available in quillback. The AI64 is pretty much there already and the play is also in progress. Raspberry Pi five was just released. You can pre-order them. You can try to buy them. They're really the stocks running short, but they'll I think become more widely available as we get into 2024. So we've got the BSP ready for that. And then the new Renaissance S4 starter kit is also available as the BSP is available there. I've already, John Simon and I already took a whack at creating the release notes for quillback so you can get a little more details there. We'll continue to fill that in the next few weeks as we get to the quillback release. So here's our schedule for quillback. We'll do the M1. Like I said, this month we're a little delayed. We originally planned for the 30th of November. We're now going to be this week, late this week. M2 will be before Christmas. Milestone 3 will be a few weeks after CES. What we try to do is anything that we do, we show at CES. We try to get into quillback so you can then download that and play with that. Do whatever you want with it. And the final release will be in the middle of February. And the slides are all uploaded already. I uploaded them before the talk so you can take a look at these at your leisure. So like I mentioned, AGL, we're Yachto based. We're trying to stick with Yachto LTS versions. Earlier this year, Yachto announced that they would be going to a standard four-year support model for their long-term support versions. We've been supporting Dunfell then for the last almost four years. It'll go end of life in April of next year. And our lucky lamprey release, which is our 12.x release, has been updated basically for every Dunfell release for the last few years. You can see as we go end of life here, I just kind of showed we'll get out to 12.1.19. So we'll actually have made, I think, 21 releases of lamprey by the time it's done, which is pretty amazing for what we've done the last few years. Right now the Yachto schedule shows that the last release of Dunfell is in mid-March. So what that means for us is our more later versions of AGL have been supporting Kirkstone, which is 4.0.x. Kirkstone support will continue through April of 2026. They will have a new Yachto LTS version 5.0, which is named Scarthcap. I'm sure I'm pronouncing that wrong. So Scarthcap will be released in April of 2024. AGL's Royal Ricefish release will basically move on to Scarthcap. And similar to what we did with Lucky Lamprey, Quirky Quilback will become our long-term release for the next few years as we continue to track Kirkstone. So what that looks like is, like I just showed you a minute ago, we finished the release in February, and then I tried to map the Kirkstone schedule for next year into releases for us. We tried to do them about two to three weeks after the Yachto release is done. So you see we have seven patch releases planned for Quilback next year. So this is the overall 2024 schedule. I think it's the first time I'm showing this. So pleased to announce that our late second half of the year release will be called Super Salmon, an easy one, most easy. Super Salmon. Every year our steering committee goes through a process where they vote on the most relevant features to work on for the year. And what this does is it doesn't necessarily determine all of the work that we'll do on AGL, but it does help focus our our paid development on particular efforts. And Jerry's probably laughing right now in the front here. He's going to have an SDV expert group update right after me. And he's going to tell me I'm all wrong. But what I tried to do was, what I tried to do was summarize where we are with each of the with each of the steering committee with each of the steering committee priority items, which were the top 10 or 11. So SDV, the top item, top three were all software defined vehicle related. Extending vert IO support. So virtual open systems is doing some work for the SDV expert group. The effort's about 80% complete. It should wrap up by CES. Now what Jerry needs to tell us in the next, you're going to, you're going to ask him this question. Okay. Cause I don't know what the answer is, but I've been able to talk to Jerry for a few weeks because he's been traveling. He's right there. He's been traveling. So what I don't know is how that 80% complete or how completing this actually finishes the vert IO support to decouple the UCB. I think there's still a few more things to be done to finish the decoupling, but I don't know how much Jerry's going to tell us in the next session. Okay. You ask him that on the Q and A. The container orchestration, to be honest, that didn't really get started. We had a guy working on it from another company and he kind of fell off the map. And so we're going to get that started again. Panasonic is making great upgrades strides on the unified HMI. Some initial commits were merged to Garrett last month. And we have some more commits that were actually made into Garrett this week that are in review. We were supposed to have a discussion about the unified HMI, but that one got canceled, right? Oh, that's not the one that got canceled. The vert IO one got canceled. Yes. See, that was the one that was canceled. He would have answered the question, but now Jerry has to answer that question. So there'll be a good talk on the unified HMI work that Panasonic has been doing and they are making great strides on this. Vehicle to cloud. So if some of the features were vehicle to cloud related, and again, vehicle to cloud is kind of part of adjacent to software defined vehicle. A lot of people will argue without vehicle to cloud, without OTA updates, without the cloud data component, you don't really have a software defined vehicle. So that's a work in progress to convert from those standard vehicle signal specification object names to something that talks that the cloud can understand. That's a work in progress. We expect to complete that in time for embedded world, which is in Nuremberg in April. So we'll be showing some new demos in embedded world as well as the new some new demos at CES this year. Additional reference hardware. From my point of view, I know Panasonic's been hard at work with this with Qualcomm. From my point of view, there's no progress. Flutter. So like I already said, we've made a lot of improvements in the Flutter workflow. One of and the Flutter and better continued evolution. So we've got those new reference apps that Dan talked about from ICS. One of the things that we need to work through, we're going to have a kind of a wrap up session next week with ICS and our Toyota Flutter guys. We're going to talk about kind of that workflow. So we had to bring those flutter. We had to bring the ICS guys on board, get them up and running on an AGL, just get the UCB up and running for them. So what lessons learned do we have from that and how easy or hard was it? And how can we improve that? How can we get them, how can we onboard new developers better and faster? So we should get some lessons learned and turn that into some activities for 2024. Numbers eight and 10 are more vehicle to cloud topics. So there will be, I'll have a couple slides on vehicle to cloud later, but there will be a vehicle to cloud expert group update by Megan Knight tomorrow. Our top support for AGL UCB. So again, in these mixed criticality use cases where you have the AGL with Linux running alongside an RTOS, we have, we've done a lot of requirements and design work or our requirements and architecture work within our system architecture team. And that work continues to evolve. So this is the list of the current, I broke it. This is the list of active expert groups that we have. Our system architecture team, we have an IVI team that's IVI production writing this team that's devoted mostly to IVI type topics, instrument cluster. So I'm just going to give you a quick update on what each of these groups has been doing. In the last year, last six months or so, we went through Confluence and created a much better, I think a much better landing page for AGL overall with very clearly marked links to each of the expert groups and the system architecture team. We're still, it's still a work in progress. We're still improving those, but we now have Confluence pages for each of our expert expert groups as well as that landing page. And the system architecture team has really been focused on that RTOS support for AGL. And they've been working with the SDV expert group on defining some of those use cases as well. The IVI expert group, again, I have these slides all uploaded to the schedule page. So you can click on all these hyperlinks. The flutter and better is now updated to 3.13.9. We've got the new IVI reference design ready for CES. And I had to tell my boss just now that we're not unveiling that this week. We're unveiling it at CES. But it looks really cool. And if you come to our old member meeting in January or February, rather, which he announced, we will definitely be able to show it to you there. So it looks really sharp. I think you'll really like it. It's a big improvement over our legacy UI. App framework and connectivity, Kooxa.doll data broker, the flutter and the web apps, as well as the Qt apps, the legacy Qt apps, are all adapted to use the Kooxa.doll data broker, so they're all using GRPC now. We did a major documentation update on our documentation site, created a Rust Mix-in layer. So now you can use Rust on AGL. And the reason we had to do that, that Rust Mix-in layer is a Yachto thing. Yachto Kirkstone had an older version of Rust, and we needed a newer version of Rust to be supported for the data broker. So now we can more rapidly update our Rust versions. That Yachto Mix-in layer for Rust was created by Scott Murray and in our connectivity group. And you can see there's the Confluence page for the app framework group. There's also a separate one for the connectivity group. So SDV, you'll learn all about it in the next talk. I'm not even going to talk about this, this is all great stuff they've been doing. We have three talks planned for this week, and then a boff session to close tomorrow out. So we can all talk about SDV. And I think one of the questions most of you are going to ask is, what is an SDV? And hopefully we can answer that. Vehicle to Cloud. So we're working on this reference implementation for messaging between connected vehicles and cloud providers. The work is being done by largely being done by AWS as well as our connectivity group for the VSS part. But intentionally the interface is not going to be, it's going to be cloud provider agnostic, but we will provide a AWS reference implementation. So the idea is the APIs and the specs, so to speak, will all be agnostic to the cloud provider. You can use Microsoft, you can use whoever you want. And again, we're not trying to reinvent the wheel here. We're not trying to create an AGL specific version of a vehicle to cloud service. What we want is to work with other groups like Covisa and other groups that are working on these types of activities, make sure that we're all focusing on the same specifications and that AGL is implementing those specifications and providing the reference architecture for that. So we don't want to write specs. There's nobody on AGL that I know of who likes writing specs. Getting the right documentation is hard enough. Writing specs is even harder. So Megan tomorrow has a session with a full update of the vehicle to cloud activities. Instrument cluster expert group. This afternoon, Kurokawasan will provide an update on the expert group activities. Basically our prickly pike release had our full CES 2023 demo equivalent with the containers solution that they have. And I think they're showing some of that upstairs in the AGL booth. And they're working on basically what they're going to provide in Quillback and for CES this year. Our continuous integration and test group expert group. So they're the guys who are focusing, the folks who are focusing on Jenkins and Lava and making sure that our Garrett CI works whenever you make a commit to Garrett. It goes through automated CI, some automated testing. We added risk five, are that risk five board to the Lava lab. So now everything that we're doing, all of our, any commit that you do now gets run through CI for risk five as well as ARM as well as Intel. We're still trying to acquire that minus us as four starter kit. We're going to get it soon. We were looking at, but I should have updated this. We're not looking at Beagle bone AI and Raspberry Pi five support. We've got to been there pretty much already now for Quillback. So you say, well, I want to get involved. How do I get involved? I'd love AGL. It's really wonderful. How can you help? How can you help me? Or how can I help you? So if you're interested in any of these topics, all these expert groups have a zoom call every other week in addition to the Confluence pages. So you can talk to the people who are actively working on it, learn what's going on. So there's a call, like I said, every other Monday for vehicle to cloud, every other Tuesday for the system architecture team and the SDV expert group. We have a weekly developer call. So every Thursday, if you have a question about AGL or how to do something or you're having trouble getting something to work, or you just want to talk about AGL or listen to me blabber on for an hour, you can call into the developer call. We've been really working on our documentation improvements there as well. So we've had some folks who very kindly volunteered their time to just do documentation updates. It's really helped our documentation. The gentleman who was working on that, he's moved on to some other things now. But that's a continual effort to get our documentation improved. And I think helping improve documentation, you will learn AGL. Just kind of the way it is. So here's some more information. If you want, we have a mailing list for our community, our developer community. We have IRC chat available on IRC available on hash automotive on Libra chat. And our weekly developer call. So that's it. Thank you. And if you have any questions, I'll be happy to make up an answer.