WEBVTT 00:00.000 --> 00:11.200 Okay, so thanks everyone, so welcome to our talk, can't focus any conductor, hold my 00:11.200 --> 00:16.440 beer, we didn't bring beer because we drank it, we're really sorry, it's kind of what 00:16.440 --> 00:24.200 we do at Open Hardware, but just for some instructions, I'm Charlie Man, so I lead communications 00:24.200 --> 00:28.680 and dissemination for the European Tristan Project, which is all about getting risk 00:28.680 --> 00:34.680 5 out of the lab and into industry, and I also get to lead technical content at the Open Hardware 00:34.680 --> 00:35.680 Foundation. 00:35.680 --> 00:44.120 So, yes, my name is Cairo Kaplan, I am a verification engineer at Open Hardware Foundation 00:44.120 --> 00:51.120 and I also work on managing European projects under the Planet Foundation framework. 00:51.120 --> 00:58.640 Thank you, and just a quick overview of what we'll take 00:58.640 --> 01:05.960 you through today on today's testing menu, why we need a revolution in European silicon, 01:05.960 --> 01:08.080 which we really do. 01:08.080 --> 01:14.480 The Master recipe, a little bit about our Tristan Project, and also about the unified risk 01:14.480 --> 01:20.080 5 IP access platform, which will henceforth be referred to as the UAP because it's 01:20.080 --> 01:25.360 mouthful, and a little bit about how we built it, what it's for, the quality control that's 01:25.360 --> 01:30.320 behind it, and also the tasting room, so you get to see a couple of the cool risk 5 projects 01:30.320 --> 01:34.440 that have already been donated. 01:34.440 --> 01:39.240 Just a few disclaimers, everything we're going to show you is licensed under the Open 01:39.240 --> 01:41.920 Source Eclipse License. 01:41.920 --> 01:47.280 It is a work in progress, so please don't expect this to be perfect, but we hope by the 01:47.280 --> 01:52.000 end you'll be interested enough that you'd also like to contribute, maybe donate some 01:52.000 --> 01:59.000 of the risk 5 IP you're working on, and a lot of what we are doing this for is supporting 01:59.000 --> 02:03.160 European technical sovereignty, underneath the ship's act. 02:03.160 --> 02:09.040 You've probably heard so much about this in recent months, so we need more European sovereignty. 02:09.040 --> 02:16.440 This is our way of contributing towards that, making risk 5 a reality in Europe. 02:16.440 --> 02:20.240 Just a little bit of context too, don't worry, we'll get to the fun stuff soon, all 02:20.240 --> 02:25.800 the work that we're talking about today was developed and funded under the ChipsJU Project 02:25.800 --> 02:31.880 Tristan, our sister or wife project, depending on how you look at it, is older, so they 02:31.880 --> 02:34.480 do the software side of things. 02:34.480 --> 02:42.680 They've also joined, which is fantastic, scale for edge, Rebecca and Rigoletto, also projects 02:42.680 --> 02:47.120 researching risk 5 applications in Europe, but we were so great, they joined too, which 02:47.120 --> 02:52.840 is fantastic, so you can now find IP from every single one of these projects in the 02:52.840 --> 02:55.840 UAP. 02:55.840 --> 02:59.040 On to the fun stuff. 02:59.040 --> 03:05.040 I apologize for my chat GPT graphics, I'm not a designer. 03:05.040 --> 03:10.240 But we're here to talk about today is, well, an issue that's very close to my heart is, 03:10.240 --> 03:17.040 why is it so damn hard for us hard, where folks to find any verified risk 5 components, 03:17.040 --> 03:19.560 that are fully open sourced. 03:19.560 --> 03:23.440 On the side, we've got our software counterparts and they just get to tap up the free 03:23.440 --> 03:28.200 bar whenever they want to, because they can just fork their project straight from get hub. 03:28.200 --> 03:33.640 And it's great, they're contributing back, it's a nice cycle, but it's not fair. 03:33.640 --> 03:37.000 We get stuck in a long queue for expensive beer. 03:37.000 --> 03:39.200 We don't know how it's been brewed. 03:39.200 --> 03:40.240 We don't know what's in it. 03:40.240 --> 03:44.360 We don't know if it's going to go with the other beer that we've already got. 03:44.360 --> 03:49.000 Quite frankly, why should we have to queue for expensive beer? 03:49.000 --> 03:51.840 Now, I'm sure we're talking to the choir here. 03:51.840 --> 03:58.880 It is very possible to find your recipes for free open sourced beer online. 03:58.880 --> 04:05.480 It is going to be difficult, though, because quite frankly, we have a silo problem in Europe 04:05.480 --> 04:07.640 when it comes to risk 5 IP. 04:07.640 --> 04:14.120 We've got world-class projects, absolutely top of the range, the freaking hard to find, though, 04:14.160 --> 04:19.760 the repositories are isolated, the servers are really, really specific. 04:19.760 --> 04:24.640 We've got non-standardised metadata, we've got no long-term maintenance, and this is a real 04:24.640 --> 04:27.040 issue with European Research Projects. 04:27.040 --> 04:32.600 We produce outstanding IP, and then we just walk away and leave it, and nothing happens, 04:32.600 --> 04:37.480 and we really want to change that. 04:37.480 --> 04:43.160 And of course, maybe you do find the ingredients, the IP that you need. 04:43.480 --> 04:47.160 It's going to be really fricking expensive, though, and a lot of them are covered by 04:47.160 --> 04:48.160 NDAs. 04:48.160 --> 04:52.400 So, let's just say that we have a very motivated benefactor. 04:52.400 --> 04:58.160 They've dropped us a couple of million, you know, to burrow up our beer silicon. 04:58.160 --> 05:02.280 You've got a test-each into the German ingredient and make sure it works together. 05:02.280 --> 05:05.280 Like, not all of that is going to be done for you. 05:05.280 --> 05:08.760 And after all of this hard work, what if it doesn't work in the end? 05:08.760 --> 05:12.480 What if you end up with nothing? 05:12.480 --> 05:16.560 And that's a genuine issue in hardware. 05:16.560 --> 05:20.240 And quite frankly, we're thirsty, and we don't believe that engineers should have to 05:20.240 --> 05:24.000 wait for beer. 05:24.000 --> 05:27.320 And us waiting for beer is a massive flipping problem for Europe. 05:27.320 --> 05:31.560 Like, there is a 700 billion global semiconductor market. 05:31.560 --> 05:35.160 We've captured 10% of that, that's nothing. 05:35.160 --> 05:38.240 We have some of the best IP in the world, we can do more. 05:38.480 --> 05:43.440 The chips-chair U project is trying to get us up to 20% by 2030. 05:43.440 --> 05:48.720 We're making great strides, but it's going to take us all contributing to make sure 05:48.720 --> 05:52.960 that we reach the goal that we want to reach. 05:52.960 --> 05:54.800 And this is one of the reasons we exist. 05:54.800 --> 05:59.360 The open hardware foundation, we are the champions of beer brewing. 05:59.360 --> 06:02.800 We are an industry partnership with the eclipse foundation. 06:02.880 --> 06:09.200 We basically, well, we like to think, we are the leader in industrial grade risk five 06:09.200 --> 06:10.560 cores. 06:10.560 --> 06:12.720 Everything that we produce is open source. 06:12.720 --> 06:16.480 It's permissively licensed at its production ready. 06:16.480 --> 06:20.680 And we are already used in projects across Europe for some very large names, including 06:20.680 --> 06:27.600 Talis, and a bunch of others, which I can't tell you about, but I'm working on it. 06:27.600 --> 06:30.320 I won't stop too long on the cores that we have. 06:30.320 --> 06:33.680 We produce the core five family. 06:33.680 --> 06:39.280 Everything that we have is either rated at TRL 4 or TRL 5, ready for adoption. 06:39.280 --> 06:43.520 If you want to talk to us or get your hands on any of this IP, come and see us afterwards 06:43.520 --> 06:48.600 or talk to us at the Eclipse booth. 06:48.600 --> 06:51.600 And again, we've mentioned the Tristan project. 06:51.600 --> 06:57.560 We've come to the Tristan project through Eclipse, because Eclipse is based here in Brussels. 06:57.960 --> 07:04.760 The Tristan project is a $50 million project to mature and industrialize European risk five. 07:04.760 --> 07:10.360 46 partners, including Talis who I've mentioned, ETH Zurich, and XP, some massive 07:10.360 --> 07:14.680 names in research and development of risk five components. 07:14.680 --> 07:20.600 We like to think that we provide the brewery equipment, so you can make your own risk five. 07:20.600 --> 07:25.400 Oh, and actually just on a side note, if anyone is interested, this project is wrapping up 07:25.480 --> 07:27.160 in six months in the summer. 07:27.160 --> 07:30.920 We're going to be at the risk five summer in Italy, and we are bringing all of our 07:30.920 --> 07:32.120 demonstrated models. 07:32.120 --> 07:35.800 So if you want to get a closer look at what we've produced, come talk to us. 07:37.400 --> 07:40.920 And with that, I'm going to hand over to my colleague Cairo to actually take you through the 07:40.920 --> 07:41.640 UAP. 07:41.640 --> 07:45.400 Thank you for the introduction, Charlie. 07:45.400 --> 07:48.920 So yeah, I'll try to keep up with it if you're a presentation. 07:48.920 --> 07:52.040 So yeah, the master's type. 07:52.040 --> 07:59.480 So we are open hardware trying to build the infrastructure of European risk five collaboration. 08:02.200 --> 08:07.720 So this is why Tristan and the OpenRF Foundation, we have launched the 08:07.720 --> 08:12.440 Unified risk five access platform for now on quality UAP, 08:12.680 --> 08:20.520 which is one of the outcomes from the three years of work of the Tristan project. 08:21.080 --> 08:28.840 So the UAP, in fact, is a first of its kind and in this sense that we are trying to provide a 08:28.840 --> 08:36.360 single collection of verified industry-ready risk five IPs or not only cores, but also the 08:37.160 --> 08:43.800 peripherals and low levels of trends and the society projects. 08:44.520 --> 08:46.440 And I'll develop here in Europe. 08:48.440 --> 08:56.360 So yeah, it's a consolidation of a hardware, software components with a very clear view of each item's 08:56.360 --> 09:04.760 maturity level, visibility, licensing, which is also very important and also the integration workflow. 09:05.720 --> 09:13.320 So much of what's developed on the Tristan and what's available on the UAP is open source, 09:13.320 --> 09:14.920 but it's not restricted to that. 09:17.080 --> 09:23.480 So yeah, critically, it's interested that IP from multiple EU research project is captured 09:23.480 --> 09:29.160 and maintained in a way that an organization can benefit from it, not only the organization, 09:29.160 --> 09:36.520 but also students and engineers, we're trying to driving a sustainable long-term collaboration. 09:37.240 --> 09:44.040 So yeah, in fact, right now, as Charlie said, we have IP from Tristan and Israel, 09:44.040 --> 09:50.760 which were the first two projects that were involved on this, but also recently also from Rebekah, 09:50.760 --> 10:03.080 SK for Edge, and yeah, and more to come. So yeah, let's see what features the UAP has. 10:05.320 --> 10:11.080 So let's say the core of the UAP is what we call the vitriopository, which was the initial name 10:11.160 --> 10:19.560 of the UAP as some of the concept characteristics, I can say that it's centralized, 10:20.360 --> 10:29.960 the sense that the IPs that we show with catalogue, they remain on the partner servers, 10:29.960 --> 10:36.200 that means it can be on GitHub, on Project SkitLab, or even on other companies. 10:36.520 --> 10:48.680 So the index, the metadata we don't host is actually IP itself. Of course, for open-holder IPs, 10:48.680 --> 10:59.400 it's GitHub on the organization. The UAP is also a tool for discovery. Actually, it's more 11:00.360 --> 11:12.520 oriented towards people, humans, we have other projects that's more integrated with software tools, 11:12.520 --> 11:19.800 let's talk about later. So it really tries to be and propose to be a single port of what you can 11:19.800 --> 11:31.080 find the risk-5 IPs and slurators and so on, and you got this sovereignty. We follow the 11:31.080 --> 11:40.040 fair data principles, which means a findable, accessible, interoperable, and reusable information. 11:40.280 --> 11:48.120 And we back up this infrastructure and European servers. 11:51.640 --> 11:55.560 And now about the IP card, which is, that's it's a more 11:58.280 --> 12:06.440 machine level or software-integrated part, it's run by the IP container task group, 12:06.440 --> 12:20.520 with this initiative. We are still on the conception part on the prototype. So we build a 12:20.520 --> 12:30.520 JSON file that has the metadata about an IP regarding terrestrial license, I say, the profile 12:31.320 --> 12:41.800 and core configuration, so that let's say once you run your data tool or setup, you can get 12:41.800 --> 12:51.720 information automatically, you don't need to access an upside for that. And yeah, it's a working 12:51.720 --> 13:02.600 progress. We plan to generate this information automatically. So we can get dependency 13:03.640 --> 13:10.920 IP, which they'll choose like a bender or fused as you see. So yeah, on this, we are really 13:11.960 --> 13:14.840 building this up, so if you have a contribution on that, you're welcome. 13:15.080 --> 13:23.080 And yeah, the testing IP compatibility matrix, we are trying to, we are also building 13:23.800 --> 13:29.800 data base where you can each, we can have information about each IP, regarding their compatibility 13:29.800 --> 13:39.000 of other IPs, to get interface and profile. So now the testing room will explain the 13:39.000 --> 13:51.560 current type list for a reputable IP. So this is how the UAP looks currently. It's a variable 13:51.560 --> 13:59.320 both at the open-harder group GitHub, it's server design GitHub page, actually. And it can be 13:59.320 --> 14:06.760 embeddable easily on other projects website, including Tristan. It's composed now, currently, 14:06.760 --> 14:12.760 of four pages, the welcome page, that's what you see on the right, where you can see the 14:12.760 --> 14:19.480 list of the basic information and list of projects that listed the platform page that I show 14:19.480 --> 14:26.280 in the next slides, which the what you really want to see when you access this, it's the dynamic 14:26.280 --> 14:34.840 table with the permission of IPs available, enter page. For now, we have been a really 14:35.160 --> 14:46.280 project centric in the sense that this was a tool initially used more to, so that you 14:46.280 --> 14:51.400 research project can add their own IPs and know the status of development across time, but 14:51.400 --> 14:57.720 trying to make it more open so that I need to be associated with a project. So this is the 14:57.880 --> 15:03.960 purpose where you can add information about it and the help page that has the information 15:03.960 --> 15:17.960 how you can help or contribute. So this is the platform page that says it's the main page that you 15:18.040 --> 15:28.440 will see when you want the first column is the name of an IP or the name of a contribution 15:28.440 --> 15:35.400 from an existing IP. You can see that valid figures the openholder CVC, since CVA true, 15:37.240 --> 15:43.400 then a category here I filter for course because I think that's more in terms of many of you, 15:43.880 --> 15:57.160 but we can also add, we also have peripherals, real-time OS, and other regulators and so on. 15:57.160 --> 16:05.000 This list of categories still being viewed, being viewed. The license, this is very important, 16:05.000 --> 16:14.680 so I'll talk about how we prefer to use the soldepal license for hardware IP, so it's a business 16:14.680 --> 16:27.320 friendly. This status, this is something more interesting for user projects, but for you guys outside 16:28.280 --> 16:33.480 maybe interested in that, basically the description, and the associated project, 16:34.520 --> 16:41.320 if we're using the future, as I said, when we transfer it from the project to IP centric, 16:43.160 --> 16:48.600 of course it can also be seen on a mobile phone or small screens. 16:49.080 --> 16:59.960 Yeah, this is the architecture of that, so it's basically a post of two layers. 17:02.360 --> 17:10.040 So the first part, let's say it's the UAP websites, it has a list of several views. 17:10.360 --> 17:20.040 You can see that as an engineer, project manager, or project reviewer, this shows, for instance, 17:20.040 --> 17:27.800 if you are a project manager, the information about IPs and which work item it has related 17:27.800 --> 17:35.480 in a project reviewer, if they want to see during the window of time, where the project has been reviewed, 17:36.440 --> 17:42.680 they create status of a single project, and for the public, we are not interested in 17:42.680 --> 17:53.640 project, we're interested in IPs and their status. So yeah, this is completely a frontend code, 17:53.640 --> 17:59.960 so you can host it not only on GitHub page, you can mirror it and back up everywhere, 17:59.960 --> 18:09.160 this makes it very easy to be transferred. The database path let's say it's the 18:09.160 --> 18:15.000 future repository, it's where we store, currently, the information about the each IP in the future 18:15.000 --> 18:24.040 this will be obtained automatically through the IP cards, and yeah, there is a, since it's 18:24.200 --> 18:32.040 hosting GitHub, we have some say validation scripts for when we have new contributions, 18:32.040 --> 18:40.520 of course, we are to open how the as part of the system will review it, it's the governance 18:40.520 --> 18:46.920 of that, we will make changes in the future, since when it's an answer, but this is how it's 18:46.920 --> 18:57.080 done now, and then at the very right is where actually the IP sits on, it's where also you 18:57.080 --> 19:05.400 the JSON IP cards will be stored, so it can either be open source IP from open how to 19:05.400 --> 19:15.160 it's on GitHub or it can be hosted on a partner repository, be on GitHub or if it's a proprietary 19:15.240 --> 19:29.640 project, you can provide assets or information on such an IP. So now I show some of the IPs that 19:29.640 --> 19:35.160 are included on such a list, as for example, the first one that's, let's say it's the most popular 19:36.120 --> 19:47.160 penader IPs, the CVA6, so yeah, it's a, it's a industrial grade application class family 19:47.160 --> 19:57.000 of course, it's a, it's a, it's quite a family because CVA6 is highly configurable, so 19:58.600 --> 20:03.560 under the framework of a Tristan project, one of the configurations, the CVA2, 20:03.560 --> 20:11.320 E6, X, has reached a chart of five, the level of verification, thanks to the effort that 20:11.320 --> 20:23.400 by Tales, but also for contribution by us, open hardware and CNA, as I said, it's highly 20:23.400 --> 20:31.160 configurable, you can not leave them so that, and you are not, if you use the CVA6 20:31.160 --> 20:38.040 CNA to brush, you use it as it is, you can modify it, many, some companies they, they customize it 20:40.600 --> 20:46.120 and let's say they patch it when they use on their own, you know, projects, but yeah, that's all I can say for that. 20:47.080 --> 20:59.480 The CVA2A4TP, the SessionA2, it's an embedded class CPU, we use it on the, on the 20:59.560 --> 21:10.760 counter, you need projects, it's also has many configurations, it's, it's built through, 21:10.760 --> 21:19.480 we use the on efficient power efficient, but also on this build operations, I guess that's the 21:19.480 --> 21:26.680 CVA2, E40P is one of the configurations that we choose, TLF5 on previous years, 21:29.240 --> 21:40.440 you have the CVA2, E40P, VQ, that's for noise TLF4, also the PX, the F40X, which has the 21:40.520 --> 21:49.400 accession interface, that's a co-processor interface, so yeah, and then the, let's say there's 21:49.400 --> 21:58.040 malice score that has been working on this project, it's on the CVA2, it's a two-stage pipeline 21:58.040 --> 22:06.600 core, initially it was, it was developed at the zero risk, donated to low risk, as a lab 22:06.600 --> 22:15.640 X project, we took some configuration of that to make it right to be in this upgrade, it's still 22:15.640 --> 22:24.440 TLF4, TLF3, by now, but we are making it working to make it TLF5, so yeah, these are the 22:24.440 --> 22:32.520 examples of cores that you can find on this platform, what you can also find now talking about 22:32.920 --> 22:39.800 verification security, continuously subject, you can also find the open-hunter core V, very 22:39.800 --> 22:52.360 if, which is our function verification framework using a UVM for the core 5 family, and for the 22:52.360 --> 22:59.960 CVP, it's not the case right now, but we have a roadmap that we add metrics on regarding 23:01.960 --> 23:13.560 users, also from its insights about each other's IP and status badges, for other IPs we show, 23:13.560 --> 23:19.320 as I said, the current TLF level, which is very important if you want to use it on a prototype 23:19.560 --> 23:29.080 and finally, some interesting characteristics of the UAP are ready, most of it's already implemented 23:29.080 --> 23:39.720 right now is that the UAP as an other open-hunter IPs, they are hosted on GitHub, but they are 23:39.720 --> 23:49.240 backed up on a crypto foundation on GitHub instance, it's doing weekly, we are working to make 23:49.560 --> 24:02.040 it possible to us to mirror in case GitHub goes down, let's say, and also supporting 24:03.320 --> 24:16.440 supporting a user warranty, we hosted on vaccination EU servers in France, and yeah, that's it 24:20.200 --> 24:28.680 yeah, before finishing it, it's just a benefit, sustainability, as I said, it actually 24:28.680 --> 24:33.880 requires low maintenance because it's basically for now it's only front-end code, 24:37.400 --> 24:44.440 the governance from it, it's this initiative is led by OpenHunter and an XP, there is a 24:44.440 --> 24:50.760 three-stern on the virtual repository task group, this may change the future, but the 24:50.760 --> 24:59.320 virtual repository task group was continuous, and so yeah, and we are ready to onboard new projects 24:59.320 --> 25:06.520 and trying to make it more friendly for general IPs, of course, trying to be always 25:06.760 --> 25:20.600 in fact as an open source of contributions to be added on this platform, that's how a final message 25:21.640 --> 25:30.520 we tend to build sustainable, sovereign European ecosystem for the third carries results 25:30.520 --> 25:37.480 for contributing to the blue, and yeah, the where you can have information about that until 25:37.480 --> 25:41.480 access the actual platform, that's it, thank you