 Thank you, everyone. So just a correction. I don't work for the open source technology center I still work for open source right now. I am in actually client computing, which actually is Something still doing open source across Intel not in a specialized group. But yeah, so thank you very much I will be talking. I mean, thank you Luca first for This topic that we have introduced. Yeah, I was not aware that was really great and Unfortunately, you will not see nice pictures as in my presentation as with Lucas, but actually I was thinking this is this is exactly what I'm trying to cut to cover here I mean that I mean once I go into the topic we were able to see what what we are trying to achieve with that I was trying to help You know going up from just providing an OS with OS services or you know Supporting device drivers and hardware and and keeping it at a low level to something a little bit more You know providing frameworks in this case for tracking animals or you know anything that you can imagine So basically the topic here is that in Zephyr these days. We are moving really away from just providing something basic like basic layers for firmware development or you know Solution development into into something a little bit more expanded in the context of this I mean what I'm when I'm talking about frameworks. I'm talking about middleware or You know application or as close as you can get to an application based on top of subsystems or low-level components accessing hardware Through an abstraction and using multiple OS services to provide basic firmware functionality That's not something that we have done Over the years since the project has started But we have been noticing a trend in the Zephyr project that there is more and more specialized Liars being added to support this solution or that solution this firmware this type of Product and that's you know, it's a double-edged sword because you know when you introduce something that is specialized that you only You are the only user of that becomes a problem That's where we try as much as possible to stay generic and and and make sure that you know Whatever you are providing in the project is is reusable by others and useful for others so Going through the history of how things Have been developing. I mean and this is usually the common case that you get with traditional artists environment is you get Basically an off-the-shelf colonel scheduler Usually you've got that the house or a custom piece of the house from your vendor or you build all your own custom BSP and you do basically a whole lot of firmware development basically all the way from you know logging to you know power management to You know shell networking Bluetooth in some cases and and so on and so on so you basically build everything and that's actually the problem we are trying to solve with Zephyr these days and The problem with that is that if you have multiple firmwares multiple IPs in this case, that's what we have inside Intel Usually you repeat the same process for each IP or fear for each firmware solution that you are trying to provide and that's a lot of duplication, right and When the Zephyr project has started it actually was it was like this I mean we had the basic kernel a few services here and there some device drivers of both and You could take it start implementing your application and As I said earlier, this is what you find in traditional artists environments. This is how you know free artists Works this is how when you take other commercial Artuses you take basically the you know the scheduler or the the the kernel parts and you build on top of them so they will not You will not get any OS services. You have to come up with this Zephyr. However over the years and Zephyr now has been out since 2016 has Developed or added a lot of interfaces and and this is You know to be honest here is a very old slide. So there are actually many many more Services there beside the obvious ones that are listed here But the whole idea was all is that We wanted to make application or firmware writing Simbler easier and we wanted to encourage reuse all the way. So what's happening is that we So basically We we we tried anybody who participates in the project anybody who Contributes or a member of the work that started pushing Whatever they had Managed for probably a long time internally into Zephyr Making in a generic way making that useful for all users of Zephyr and there are many examples of that Yeah Just to name a few for example when the project has started we had a very simple Redumentary logging subsystem when Nordic joined the project they actually Had a lot of experience Coming from their products and is decayed. They contributed that shell as well the power management And a few other services that a lot of the members of the project have been adding or all of the year make Making these components Useful for everybody participating in the project so that an application writer or if you are trying to build a firmware solution you actually just focus on the specifics of your product and not trying to reinvent the wheel Building up things from scratch or developing things from scratch Which which takes a lot of time a lot of resources has can have a lot of issues Doesn't have enough, you know probably coverage in terms of security and and and and and testing and so on as we have in the Zephyr project itself and Most likely it would be tied to your hardware solution as well. So it's not agnostic One thing that you would notice here is that a lot of these subsystems would work never mind if you are on arm x86 Risk five and so on and that's really the beauty of how we have been doing that so collaboration and reuse From a rich catalog of hardware and firmware agnostic features Applications in this case adds the missing pieces and implements the logic on top of that and Right now, this is really the common case for for firmware products based on Zephyr and this is why lots of Product makers have been actually attracted to the project and start using the project because they didn't have to worry about all of these things And this is simplified Representation but we have really mommy Laura was mentioned obviously Bluetooth and Networking and and so on come come to that which which is usually very very difficult to develop and maintain When you are doing things behind closed doors or on your own So just to go into Examples on a little bit more of what we are trying to do if this clicker starts working Okay Here we go So just just an example and that's something that is happening right now Basically, what we are trying to do is Harness the the the power of all of these OS services that we have as a far making them all work together Probably introduce additional some sometimes specialized subsystems to create frameworks so that basically You you get closer to a firmware Product or or final product but not quite there because it is not really specialized So he are talking about sensing and there are different ways how something like that can be used It's definitely not going to be tied to one product or another But it can be enhanced A on the application side to actually become more specific and and and more You know more unique or more differentiated from the competitor or like a similar product on the market providing interfaces and Building blocks to allow something like that becomes really a powerful Area in Zephyr and we are noticing that in a few Segments of the project I'm listing a few items here. Yeah, so sensing I mentioned sensing I mean recently we have been collaborating with Google and the and other parties in the project and to introducing a Framework in this case or a subsystem that utilizes existing components in the project like sensible drivers and so on This is still in the early stages But it will provide more than just being able to talk to sensors and and tries also to address different use cases and Utilizing many of the subsystems that we have in Zephyr Audio is another example in this case that the list of Examples here can be actually in three or out of three. So we have collaboration Across the open source ecosystem. So in this case audio for example, there's another Leung Foundation project is to have sound open for firmware which moved away from a custom Arthos to to use Zephyr and We see a lot of development and a lot of contributions happens to to enable audio both in SOF and and and Zephyr itself. So this collaboration is very productive and and it We are we are moving into A full framework or a solution that is in this case, which is which is really nice is not Vendor specific. So it's not something that would work only on Intel hardware or or an XP's hardware It's actually we have a whole lot of vendors participating in this and joining this effort and since actually the SOF project moved to Zephyr that enabled other Other others to jump because it was only, you know, possible on extensor architecture But now you can actually because Zephyr supports multiple architectures. You can start doing audio firmware on top of other actually in this case arm and and others So that's another example. One more, which is also something we at Intel interested interested in is the embedded controller So over the last few years, we have seen a lot of vendors Doing and embedded controller for those who don't know it's like it's one of the most important IPs in a in a in a laptop or in a in a desktop or in a computer system that actually initiates and controls the whole boot process and and so on. So it's something that is very important and We have been working on Adding all of the needed subsystems and components to allow vendors in this case, you know OEMs like HB Lenovo, etc to take that as a reference card and Implement their own embedded controller firmware And this is something that is composed of different components that are built in Zephyr Making writing the final application of firmware much easier. So you don't have to go and Reinvent the wheel every time you add a new feature and A few other items there as well. Yeah, but I mean the main You know outcome or take away from all of this that we want to see people upstreaming generic components as You know weird or you know non-standard they might be into Zephyr instead of maintaining these things into in their environments and And Basically, obviously on the project level we have to define You know the line between what is generic and what is custom and that's something we are doing We have been doing really well on the on the you know Zephyr TSC level and and by reviewing contributing in general and We also want to look at existing subsystems and components that we have in Zephyr and Enhance the interfaces there to allow adding additional liars on top of Zephyr and Improve the integration of subsystems. So we want always to look at the full solution instead of like, you know Looking at one system one subsystem at a time and testing things In isolation, we want to look at how can I use? Like a few components that we have in Zephyr and how they work in orchestration and how we can you know Make it possible for somebody to add something on top of that and in the most optimal way possible So you don't have to worry about it in your application. Zephyr will be already taking care of that and Obviously, we want to do all of this and still Allow and embrace innovation and differentiations on the application side Because if Zephyr becomes where the solution is found in that becomes really not interesting for for vendors because everybody wants to You know, I mean there is the competitive side and everybody wants to do things a little bit differently And that's actually we have seen that being done like on the EC So I mean we on the embedded controller side where All of the generic stuff is upstreamed and made available But there are always like also proprietary and you know add value features that you know Depending on who you are you get access to and so on so this is basically where we want to get to into enabling additional layer now that we are mature enough to actually Also Get get more involvement from vendors and get more contributions into Zephyr and make all of that Available for everybody in terms of OS services and generic components. So with that, thank you very much Yeah, and enjoy the rest of the event