WEBVTT 00:00.000 --> 00:06.000 I think we can start. 00:06.000 --> 00:10.000 So, hello, my name is Fonselli Varian, and today I'm going to present to 00:10.000 --> 00:13.000 a risk-five hardware is here, what about software. 00:13.000 --> 00:17.000 And as a first disclaimer, this will be a very, very opinionated talk. 00:17.000 --> 00:20.000 So, there's not much science in this slide. 00:20.000 --> 00:24.000 Just my personal opinion from my empirical work on risk-five. 00:24.000 --> 00:29.000 And so, to start on, I'm a senior software engineer at RQ5 CPU. 00:29.000 --> 00:34.000 So, in the last two years, I've been creating the risk-five-unified DB project, 00:34.000 --> 00:41.000 and I've contributed to Zyfer, including the AIA driver that's actually health-merge, 00:41.000 --> 00:43.000 health-not-merge, but we'll get to it. 00:43.000 --> 00:47.000 So, yeah, and I've been bringing up the software stock for a lot of different 00:47.000 --> 00:53.000 risk-five cars inside the RQ5 family, so I had to play with all these different projects. 00:53.000 --> 00:56.000 And so, risk-five hardware is here. 00:56.000 --> 01:01.000 I think a lot of thoughts already went by that today, so I'll quickly go through there. 01:01.000 --> 01:05.000 So, like, we have a lot of products coming out in RQ5. 01:05.000 --> 01:09.000 I don't know if you've seen deep computing laptops, but those are pretty awesome, 01:09.000 --> 01:11.000 and they do run, right? 01:11.000 --> 01:16.000 So, we have products, and Western digital and video, they both already acknowledge 01:16.000 --> 01:19.000 that they are using risk-five cars inside their products. 01:19.000 --> 01:24.000 In Finian also came out and showed us an automotive platform with RQ5 products, 01:24.000 --> 01:26.000 which is pretty nice. 01:26.000 --> 01:30.000 And on the other hand, there's a lot of open cars, a lot of startups working on 01:30.000 --> 01:35.000 risk-five, and well, I have a couple of boards here, but in reality, there's many, many more. 01:35.000 --> 01:38.000 So, this is actually very, very, very small. 01:38.000 --> 01:42.000 I went to risk-five landscape, and I got the full list. 01:42.000 --> 01:44.000 It looks kind of like this. 01:44.000 --> 01:46.000 So, yeah, it's pretty extensive. 01:46.000 --> 01:51.000 I don't think I can go over all the names, but this really, 01:52.000 --> 01:55.000 like, this really showcases how big risk-five is growing. 01:55.000 --> 01:58.000 And if you're like me and you leave briefly, 01:58.000 --> 02:02.000 risk-five is always on your mind, and you're algorithm-sketched to it. 02:02.000 --> 02:06.000 At least for me, like on the way right now, what I feel like is that 02:06.000 --> 02:09.000 I'm constantly getting born with different news involving risk-five, 02:09.000 --> 02:12.000 a lot of different things happening in risk-five. 02:12.000 --> 02:17.000 And this is not just an impression, so it seems like 02:18.000 --> 02:22.000 some smart guys went and did like a market research on risk-five, 02:22.000 --> 02:27.000 and risk-five is expected to be in over 16 billion SLCs in by 2030, 02:27.000 --> 02:32.000 which now is like four years ago, so, well, four years in the future. 02:32.000 --> 02:34.000 So, it's pretty close by. 02:34.000 --> 02:37.000 And so, it's supposed to be in all these different markets, 02:37.000 --> 02:41.000 we'd actually like a very fed by in the chart. 02:41.000 --> 02:43.000 Well, part of the chart, right? 02:43.000 --> 02:47.000 And so, yeah, in very, very, very different markets, 02:47.000 --> 02:51.000 and with a really big presence. 02:51.000 --> 02:54.000 And so, like, I think as other people have told today, 02:54.000 --> 02:58.000 and what this data really shows is that risk-five hardware is here. 02:58.000 --> 03:02.000 Like, this is a statement that there's not much you can say against it. 03:02.000 --> 03:05.000 And the question I'm going through today is, 03:05.000 --> 03:07.000 what about software, right? 03:07.000 --> 03:10.000 We've seen couple talks talking about software in risk-five, 03:10.000 --> 03:14.000 and they were more in debt than I'll go here. 03:14.000 --> 03:17.000 But what I really want to strengthen is that, 03:17.000 --> 03:20.000 the ecosystem will at some point start deciding the adoption, 03:20.000 --> 03:23.000 because, well, there are some industries that care about the ISA, 03:23.000 --> 03:26.000 and so, they'll just move because, well, it's risk-five, 03:26.000 --> 03:30.000 it's open, and like, if you're building a space station, well, maybe. 03:30.000 --> 03:32.000 You prefer to have some openness. 03:32.000 --> 03:36.000 But in reality, most developers, and especially most consumers, 03:36.000 --> 03:39.000 will not really care about whatever ISA you're running. 03:39.000 --> 03:42.000 So, your ecosystem really needs to be ready 03:42.000 --> 03:44.000 to accommodate all these people into it. 03:44.000 --> 03:48.000 Otherwise, developers, like, face without an operating system, 03:48.000 --> 03:52.000 ready will be very against going to risk-five. 03:52.000 --> 03:56.000 And consumers will just see what final software exists 03:56.000 --> 03:58.000 and how well that software runs. 03:58.000 --> 04:01.000 So, we have to have all those layers ready, 04:01.000 --> 04:03.000 so that we get mass adoption in risk-five. 04:03.000 --> 04:05.000 And, like, here I have a fun story. 04:05.000 --> 04:06.000 This is a personal story. 04:06.000 --> 04:08.000 Two years ago, I had to buy laptop, 04:08.000 --> 04:11.000 and I had an X86, the good old X86 laptop, 04:11.000 --> 04:13.000 and the normal laptop. 04:13.000 --> 04:19.000 And even now that RM has, like, 30 years or more of a market share 04:19.000 --> 04:22.000 or in presence, I actually chose X86, 04:22.000 --> 04:25.000 because I do a lot of weird things in laptops, 04:25.000 --> 04:27.000 and I knew that at some point in my journey, 04:27.000 --> 04:29.000 I will find some package that was not built, 04:29.000 --> 04:33.000 and I will have to go and build it myself or find some extra 04:33.000 --> 04:36.000 two hours to debug whatever was going on. 04:36.000 --> 04:38.000 So, if we don't want to wait for next, 04:38.000 --> 04:40.000 I think risk-five is 11 now, 04:40.000 --> 04:42.000 if we don't want to wait for next 19 years, 04:42.000 --> 04:46.000 to not follow these mistakes, I think we should be prepared. 04:46.000 --> 04:51.000 And so, there is also a point in risk-five, 04:51.000 --> 04:55.000 that is very against this software ecosystem maturity, 04:55.000 --> 04:57.000 to what I call hardware dispersion here, 04:57.000 --> 05:00.000 but well, you can call it a lot of different things. 05:00.000 --> 05:03.000 This file is very, very modular and extensible, 05:03.000 --> 05:04.000 and that's a good thing. 05:04.000 --> 05:06.000 That's why everyone is adopting risk-five. 05:06.000 --> 05:10.000 But, as you know, like, if you're not familiar, 05:10.000 --> 05:12.000 risk-five is actually three different ISAs. 05:12.000 --> 05:16.000 It has, like, R32E and I, and R64E, 05:16.000 --> 05:19.000 which is normal, and there's an effort that, 05:19.000 --> 05:22.000 but with those three basic ISAs, 05:22.000 --> 05:25.000 there's, like, more than 70 risk-favings extensions, 05:25.000 --> 05:27.000 and then there's a lot of implementation, 05:27.000 --> 05:29.000 defined parameters that you have to accommodate 05:29.000 --> 05:30.000 in your software as well, 05:30.000 --> 05:32.000 then you have custom extensions 05:32.000 --> 05:34.000 that companies create for very good reasons, 05:34.000 --> 05:36.000 but it's also something you have to accommodate 05:36.000 --> 05:38.000 in your software, 05:38.000 --> 05:40.000 and then it's very open and very extensible. 05:40.000 --> 05:42.000 So, you'll see an endless creativity 05:42.000 --> 05:44.000 of people doing different things, 05:44.000 --> 05:46.000 and then you'll need to try to accommodate 05:46.000 --> 05:48.000 them all in the same software projects, 05:48.000 --> 05:50.000 which brings us to the question, 05:50.000 --> 05:52.000 what about software? 05:52.000 --> 05:53.000 And, well, earlier this year, 05:53.000 --> 05:58.000 we saw some controversy in the kernel mailing list, 05:58.000 --> 06:00.000 right? And, unfortunately, 06:00.000 --> 06:02.000 this is not the most fortunate event. 06:02.000 --> 06:04.000 I'm not going for it, but it happens, 06:04.000 --> 06:08.000 and we saw, why not going crazy with R64, right? 06:08.000 --> 06:11.000 And I think, although this is somewhat controversial, 06:11.000 --> 06:14.000 I think if you go down and you look, 06:14.000 --> 06:16.000 why is this reaction happening? 06:16.000 --> 06:19.000 I kind of understand where it's coming from, right? 06:19.000 --> 06:22.000 And this is what I'm going to talk about next. 06:22.000 --> 06:24.000 My own negative opinion on R64, 06:24.000 --> 06:28.000 well, not negative, but the word that it takes. 06:28.000 --> 06:31.000 And so, I think where that really comes from, 06:31.000 --> 06:35.000 it's R65 means supporting combinations. 06:35.000 --> 06:38.000 As we saw before, it's at least the three-base, 06:38.000 --> 06:41.000 well, the three-base, as a times, 06:41.000 --> 06:44.000 70-plus extensions, because they're always growing. 06:44.000 --> 06:46.000 And then you have to define number, 06:46.000 --> 06:48.000 which is implementation parameters, 06:48.000 --> 06:50.000 that I don't think anyone has ever counted. 06:50.000 --> 06:52.000 But I think there's more than 1,000. 06:52.000 --> 06:54.000 So, if you multiply this, 06:54.000 --> 06:57.000 it's like 200,000 different combinations 06:57.000 --> 07:00.000 that you can have to accommodate in your software. 07:00.000 --> 07:04.000 And this, as I showed time and effort, 07:04.000 --> 07:07.000 and then we go and we see that, 07:07.000 --> 07:10.000 well, why does this agree with these in some, 07:10.000 --> 07:12.000 some extensive, because beforehand, 07:12.000 --> 07:15.000 he had told about the millions of silly configuration issues already. 07:15.000 --> 07:18.000 And this really stem from these hundreds of thousands 07:18.000 --> 07:20.000 of configurations that we can have with R65, 07:20.000 --> 07:23.000 which is awesome, but comes with a cost. 07:24.000 --> 07:28.000 And so, R65's trans truly is its flexibility, 07:28.000 --> 07:30.000 and this flexibility fragments the ecosystem, 07:30.000 --> 07:33.000 but obviously, I mean, most of you guys here already know about this, 07:33.000 --> 07:36.000 there is an answer to this portion. 07:36.000 --> 07:39.000 And it is the risk file of profiles that we've been hearing about, 07:39.000 --> 07:41.000 and in case you don't know what these are, 07:41.000 --> 07:43.000 R65 profiles define a baseline, 07:43.000 --> 07:45.000 like a fixed-base either, 07:45.000 --> 07:47.000 a required set of extensions, 07:47.000 --> 07:50.000 and fixed choices for ambiguous implementation options. 07:50.000 --> 07:53.000 So, there is no that end was extensiveness, 07:53.000 --> 07:55.000 or at least there is a baseline 07:55.000 --> 07:58.000 that everyone has to follow for software compatibility. 07:58.000 --> 08:02.000 And so, this reduces a lot of reliability across implementations, 08:02.000 --> 08:05.000 and this enables portable software. 08:05.000 --> 08:11.000 I mean, this is kind of a term that we've tried to reach many, 08:11.000 --> 08:13.000 many times, but at least it will allow you 08:13.000 --> 08:16.000 to download the compliant binary to ERCPU and run it. 08:16.000 --> 08:19.000 And this allows consistency to two shades clear hardware targets, 08:19.000 --> 08:22.000 and essentially, you can trust your hardware, 08:22.000 --> 08:25.000 and you can trust that it will run X or Y project. 08:25.000 --> 08:28.000 And so, profiles really enable compatibility, 08:28.000 --> 08:31.000 and this is one of the very, very good things, 08:31.000 --> 08:33.000 I think, is going on in risk file, 08:33.000 --> 08:35.000 and I'll go over what these profiles are. 08:35.000 --> 08:37.000 So, there is mostly three profiles, 08:37.000 --> 08:40.000 I think only RVA and RVB have been ratified by now, 08:40.000 --> 08:44.000 but RVA is your Linux reach software, 08:44.000 --> 08:46.000 so where you run your Ubuntu, 08:46.000 --> 08:49.000 which is probably for server and even for laptops 08:49.000 --> 08:51.000 that will run this RVA-23, 08:51.000 --> 08:54.000 then there's RVB for a little bit smaller cores 08:54.000 --> 08:57.000 that will run the build router of the Linux, 08:57.000 --> 09:01.000 and then microcontrollers, obviously they also need to agree. 09:01.000 --> 09:03.000 Again, depending on the markets, 09:03.000 --> 09:06.000 there's a lot of different things, 09:06.000 --> 09:08.000 so there's a lot of different intricacies, 09:08.000 --> 09:09.000 for example, automotive. 09:09.000 --> 09:11.000 They really almost do everything by themselves, 09:11.000 --> 09:13.000 and part of software is stuck entirely, 09:13.000 --> 09:15.000 but if you want the laptop, 09:15.000 --> 09:19.000 you really need this software compatibility to happen. 09:19.000 --> 09:21.000 And so, here I'm going through what I believe 09:21.000 --> 09:24.000 are the most basic parts of the software stock 09:24.000 --> 09:26.000 that we need to have ready in risk file, 09:26.000 --> 09:28.000 so that we can then build on top of them. 09:28.000 --> 09:32.000 So, I think first one is obviously two chains and simulators, 09:32.000 --> 09:34.000 so like your team needs to be ready, 09:34.000 --> 09:36.000 so that people can actually try and write software, 09:36.000 --> 09:39.000 and obviously you need GCC and LVM otherwise, 09:39.000 --> 09:40.000 how will you get code? 09:40.000 --> 09:42.000 I mean, maybe for chemo, you won't need it, 09:42.000 --> 09:45.000 but for everything else, you will need it. 09:45.000 --> 09:47.000 And so then you need your firmware in risk file, 09:47.000 --> 09:49.000 it is open SPI, it's pretty standardized, 09:49.000 --> 09:51.000 everyone is using open SPI, 09:51.000 --> 09:53.000 and then for operating system, 09:53.000 --> 09:55.000 and real time operating system, 09:55.000 --> 09:57.000 I got Linux and Zefer here, 09:57.000 --> 09:59.000 I believe they are the most supportive, 09:59.000 --> 10:01.000 they are the ones that support with five pairs, 10:01.000 --> 10:02.000 there's of now. 10:02.000 --> 10:03.000 And well, on applications, 10:03.000 --> 10:06.000 well, I don't know a lot of applications about applications, 10:06.000 --> 10:07.000 but we may never have Adobe, 10:07.000 --> 10:09.000 that's what I feel like, 10:09.000 --> 10:11.000 which may not be so bad. 10:11.000 --> 10:14.000 And so, I created a tier list, 10:14.000 --> 10:16.000 this is a very opinionated tier list, 10:16.000 --> 10:20.000 and this is based entirely on my bringing up 10:20.000 --> 10:22.000 of different risk file hardware, 10:22.000 --> 10:24.000 and problems I faced, 10:24.000 --> 10:27.000 and the big downs I had to do for some of these. 10:27.000 --> 10:30.000 So, first, the only perfect project in risk file 10:30.000 --> 10:32.000 is risk file unify rate of base, 10:32.000 --> 10:34.000 as you should have known by now, 10:34.000 --> 10:36.000 this is a project I created, 10:36.000 --> 10:37.000 it's obviously not perfect, 10:38.000 --> 10:39.000 it's just a venture. 10:39.000 --> 10:41.000 But the ready projects, 10:41.000 --> 10:44.000 I think from all the projects I've seen and used, 10:44.000 --> 10:47.000 I never had any problems with neither key, 10:47.000 --> 10:48.000 neither tool shins, 10:48.000 --> 10:50.000 and even the Linux kernel, 10:50.000 --> 10:52.000 it can always have what I need it. 10:52.000 --> 10:54.000 On the other end, Zefer, 10:54.000 --> 10:55.000 and open SPI, 10:55.000 --> 10:58.000 I had to fix Zefer multiple times. 10:58.000 --> 10:59.000 And so, 10:59.000 --> 11:01.000 I think they do really, 11:01.000 --> 11:02.000 they are indeed lacking, 11:02.000 --> 11:04.000 a bit of the risk file support, 11:04.000 --> 11:05.000 and for microcontrollers, 11:05.000 --> 11:06.000 I definitely need Zefer, 11:06.000 --> 11:08.000 and so I think this is now a lacking point 11:08.000 --> 11:10.000 in this ecosystem. 11:10.000 --> 11:12.000 So, 11:12.000 --> 11:17.000 this next part of my talk is a bit of a practical beginner guide 11:17.000 --> 11:18.000 towards risk file, 11:18.000 --> 11:22.000 probably majority of people here already have contributed, 11:22.000 --> 11:23.000 maybe some not, 11:23.000 --> 11:26.000 and so I'll share the lessons I learned in the last couple years, 11:26.000 --> 11:28.000 while creating an open source project, 11:28.000 --> 11:30.000 and also contributing to Zefer mostly. 11:30.000 --> 11:34.000 So, let's start with my own three contribution commandments, 11:34.000 --> 11:35.000 things I learned, 11:35.000 --> 11:39.000 I hard learned myself by hitting my face against the wall multiple times, 11:39.000 --> 11:41.000 and so, 11:41.000 --> 11:44.000 first one is to sheep early and small. 11:44.000 --> 11:46.000 Second one is to make maintainers like easier. 11:46.000 --> 11:48.000 I know this seemed like very basic things, 11:48.000 --> 11:51.000 but I'll show some examples later on on how I messed up 11:51.000 --> 11:55.000 or how I saw people messing this one really badly. 11:55.000 --> 11:57.000 And then it is to solve real problems 11:57.000 --> 11:59.000 to not drive any maintainer crazy, 11:59.000 --> 12:00.000 because yeah, 12:00.000 --> 12:03.000 solving hypothetical problems with a lot of everyone's time, right? 12:04.000 --> 12:06.000 And so, starting with the first one, 12:06.000 --> 12:09.000 I won't really suggest that if you're starting to contribute 12:09.000 --> 12:13.000 to risk file or to a new project to release the minimum useful thing, 12:13.000 --> 12:16.000 and not your grand vision of how it will all go in, 12:16.000 --> 12:18.000 because well, you're reducing your risk, 12:18.000 --> 12:20.000 it may never be merged, 12:20.000 --> 12:25.000 and at least you wasted the instead of wasting 30 or 60 days. 12:25.000 --> 12:27.000 And so, an example of this is, 12:27.000 --> 12:30.000 my first commits to risk file actually were, 12:30.000 --> 12:31.000 and how good you may read them, 12:31.000 --> 12:33.000 but I fixed some typos, 12:33.000 --> 12:36.000 which was of my same win for me at the time. 12:36.000 --> 12:38.000 I got a lot of momentum, 12:38.000 --> 12:40.000 and actually some of the typos were reverted, 12:40.000 --> 12:43.000 because they were not typos, 12:43.000 --> 12:45.000 so, 12:45.000 --> 12:48.000 this was my first contribution to risk file. 12:48.000 --> 12:50.000 But, well, then I joined Zyfer, 12:50.000 --> 12:52.000 and I started contributing to Zyfer, 12:52.000 --> 12:54.000 and I had the word in Zyfer, 12:54.000 --> 12:56.000 because I was really struggling to understand it, 12:56.000 --> 12:59.000 and this was like a Twister output, 12:59.000 --> 13:01.000 and everyone that worked in Zyfer, 13:01.000 --> 13:03.000 like for 10 years, newbies, 13:03.000 --> 13:07.000 but I didn't know that the output path was changing, 13:07.000 --> 13:10.000 and it was like a meaningful contribution, 13:10.000 --> 13:12.000 very small, but it got merged. 13:12.000 --> 13:13.000 And so, 13:13.000 --> 13:17.000 now I'll show you examples of people that didn't follow this, 13:17.000 --> 13:18.000 including myself. 13:18.000 --> 13:20.000 Then, I parted the risk file, 13:20.000 --> 13:22.000 advanced interruptor architecture to Zyfer, 13:22.000 --> 13:24.000 2500 lines of call, 13:24.000 --> 13:27.000 36 comments and conversation with maintainers, 13:27.000 --> 13:30.000 67 times that I first pushed into that branch, 13:30.000 --> 13:32.000 and I quit, 13:32.000 --> 13:33.000 and I said, 13:33.000 --> 13:35.000 well, I'll upstream these in chunks, 13:35.000 --> 13:37.000 because this isn't ever going anywhere. 13:37.000 --> 13:39.000 So, yeah, I faced these problems, 13:39.000 --> 13:40.000 and as you may see, 13:40.000 --> 13:42.000 these were three weeks ago, 13:42.000 --> 13:44.000 not that long ago. 13:44.000 --> 13:47.000 So, the next one is to make maintenance easier. 13:47.000 --> 13:48.000 I think this will, 13:48.000 --> 13:49.000 this is pretty obvious, 13:49.000 --> 13:51.000 they will be the ones helping you, 13:51.000 --> 13:53.000 and be kind to them. 13:53.000 --> 13:55.000 Most of them are not even paid to maintain the project, 13:55.000 --> 13:56.000 probably they were at some point, 13:56.000 --> 13:57.000 maybe they still are, 13:57.000 --> 14:00.000 but you can't really expect much from them. 14:00.000 --> 14:01.000 Sometimes they're very fast, 14:01.000 --> 14:02.000 but sometimes they're not, 14:02.000 --> 14:04.000 and that's life. 14:04.000 --> 14:06.000 And one very important thing I have to say is, 14:06.000 --> 14:07.000 please don't push a, 14:07.000 --> 14:08.000 I saw, 14:08.000 --> 14:09.000 like, 14:09.000 --> 14:10.000 I created the unified database, 14:10.000 --> 14:12.000 and at one point there were 14:12.000 --> 14:13.000 re-five mentorships open, 14:13.000 --> 14:14.000 and so obviously, 14:14.000 --> 14:17.000 a lot of those mentees started to push things 14:17.000 --> 14:19.000 before getting accepted 14:19.000 --> 14:20.000 to enhased their chances, 14:20.000 --> 14:24.000 but this guy did it in quite the inverse, 14:25.000 --> 14:26.000 I believe. 14:26.000 --> 14:27.000 So, 14:27.000 --> 14:30.000 he pushed several of these poor requests, 14:30.000 --> 14:32.000 which, if you're familiar with CGPT, 14:32.000 --> 14:35.000 this is basically what CGPT answered to him, 14:35.000 --> 14:37.000 and he copied paste of them here, 14:37.000 --> 14:39.000 and then, 14:39.000 --> 14:42.000 he pushed 24,000 lines of code, 14:42.000 --> 14:46.000 a hundred and sixty-eight files changed, 14:46.000 --> 14:48.000 and we tried to help him, 14:48.000 --> 14:51.000 like, there's a hundred and ten comments, 14:51.000 --> 14:52.000 we tried, 14:52.000 --> 14:53.000 but obviously, 14:53.000 --> 14:54.000 there's not much, 14:54.000 --> 14:55.000 we can do, 14:55.000 --> 14:56.000 because he, 14:56.000 --> 14:57.000 like, 14:57.000 --> 14:58.000 and to give you context, 14:58.000 --> 14:59.000 he was peaking our comments, 14:59.000 --> 15:01.000 pasting them back onto CGPT, 15:01.000 --> 15:04.000 getting all that code re-factor, 15:04.000 --> 15:06.000 and then pushing a commit, 15:06.000 --> 15:07.000 and you may see, 15:07.000 --> 15:08.000 there's sixty-one commit, 15:08.000 --> 15:09.000 so like, 15:09.000 --> 15:10.000 it's going back and forth with him. 15:10.000 --> 15:11.000 So, yeah, 15:11.000 --> 15:14.000 please be mindful that you're wasting people's time, 15:14.000 --> 15:17.000 and at least try to make something meaningful, 15:17.000 --> 15:19.000 and it's my last point, 15:19.000 --> 15:20.000 and yeah, we're back and lines. 15:20.000 --> 15:23.000 So I think what wine is was saying with this one, 15:23.000 --> 15:24.000 is, 15:24.000 --> 15:26.000 and maybe not in the nicest way, 15:26.000 --> 15:28.000 there's already so much going on in Wies5, 15:28.000 --> 15:30.000 we have tens of thousands of configurations, 15:30.000 --> 15:32.000 and we're trying to preemptively support something 15:32.000 --> 15:35.000 that doesn't really exist anywhere, 15:35.000 --> 15:36.000 and I think, 15:36.000 --> 15:37.000 I mean, this bendooks, 15:37.000 --> 15:38.000 I think, 15:38.000 --> 15:39.000 hence, 15:39.000 --> 15:40.000 that this exists, 15:40.000 --> 15:42.000 meets some cars that have begun, 15:42.000 --> 15:43.000 but realities, 15:43.000 --> 15:44.000 there's not much, 15:44.000 --> 15:47.000 like there's no one showcasing that they are using those cars, 15:47.000 --> 15:49.000 so why are we all maintaining them open source? 15:49.000 --> 15:50.000 And so, 15:50.000 --> 15:53.000 this kind of shows a hypothetical use case, 15:53.000 --> 15:54.000 that maybe was not hypothetical, 15:54.000 --> 15:56.000 but was treated as such, 15:56.000 --> 15:58.000 and so I think a much nicer way to, 15:58.000 --> 15:59.000 so obviously, 15:59.000 --> 16:00.000 this was by, 16:00.000 --> 16:02.000 when Duke sees much more experience than me, 16:02.000 --> 16:05.000 but if you're starting to work in open source, 16:05.000 --> 16:06.000 I would say, 16:06.000 --> 16:07.000 and suggest that, 16:07.000 --> 16:09.000 like, a much nicer way to start contributing, 16:09.000 --> 16:11.000 would be to check very small things 16:11.000 --> 16:13.000 that you may contribute to, 16:13.000 --> 16:14.000 and like, 16:14.000 --> 16:15.000 check CI failures, 16:15.000 --> 16:16.000 everywhere, 16:16.000 --> 16:17.000 in all open source projects, 16:17.000 --> 16:18.000 you'll see hundreds or thousands of CI, 16:18.000 --> 16:21.000 warnings that no one has time to look after. 16:21.000 --> 16:22.000 Go look after them, 16:22.000 --> 16:24.000 I think that's like a very nice start, 16:24.000 --> 16:26.000 and our soft problems, 16:26.000 --> 16:27.000 you face as a user, 16:27.000 --> 16:28.000 so like, there's all these boards that are, 16:28.000 --> 16:29.000 like, I mean, 16:29.000 --> 16:30.000 out in risk, 16:30.000 --> 16:31.000 probably, 16:31.000 --> 16:32.000 if you run any type of software there, 16:32.000 --> 16:33.000 you'll find some things, 16:33.000 --> 16:35.000 and try to understand what are the issues 16:35.000 --> 16:37.000 that you're facing in contributing, 16:37.000 --> 16:38.000 contribute them back, 16:38.000 --> 16:41.000 and you'll probably help a lot of other people, 16:41.000 --> 16:43.000 and people will like you a lot. 16:43.000 --> 16:44.000 And so, 16:44.000 --> 16:47.000 if I can feast you to contribute to risk, 16:47.000 --> 16:50.000 this is a very curated list of tasks, 16:50.000 --> 16:52.000 and this is just in Zephyr, 16:52.000 --> 16:54.000 and in need of database, 16:54.000 --> 16:56.000 and these are things we can work together on, 16:56.000 --> 16:58.000 if you're interested in that. 16:58.000 --> 16:59.000 So, 16:59.000 --> 17:00.000 I think Zephyr right now, 17:00.000 --> 17:01.000 in the risk of cold, 17:01.000 --> 17:04.000 has a lot of CI warnings about different things, 17:04.000 --> 17:06.000 if you want to go check them, 17:06.000 --> 17:08.000 that will be very helpful. 17:08.000 --> 17:09.000 Or if you're more experienced, 17:09.000 --> 17:12.000 and you want to part extensions that are ratified, 17:12.000 --> 17:14.000 and don't exist in Zephyr, 17:14.000 --> 17:16.000 there's a whole bunch of them. 17:16.000 --> 17:17.000 So, 17:17.000 --> 17:19.000 so feel free to just hop in, 17:19.000 --> 17:21.000 and definitely a lot of people will be thankful, 17:21.000 --> 17:23.000 and will help you in that work. 17:23.000 --> 17:25.000 And in the risk of NFIDB, 17:25.000 --> 17:26.000 you never heard of it, 17:26.000 --> 17:27.000 most likely, 17:27.000 --> 17:30.000 it's a softer project to describe risk five ISA. 17:30.000 --> 17:32.000 We have a lot of open issues, 17:32.000 --> 17:33.000 you may start there, 17:33.000 --> 17:37.000 but I also have this personal journey of trying to develop data 17:37.000 --> 17:40.000 from UDB to upstream projects, 17:40.000 --> 17:41.000 like the Linux kernel, 17:41.000 --> 17:42.000 like being Newtills, 17:42.000 --> 17:43.000 like LLVM, 17:43.000 --> 17:44.000 like you know, 17:44.000 --> 17:46.000 that's a new tool for everyone this morning. 17:46.000 --> 17:48.000 And if you're interested in 17:48.000 --> 17:51.000 automate some of these definitions in upstream projects, 17:51.000 --> 17:52.000 yeah, let's talk, 17:52.000 --> 17:54.000 and let's make this happen. 17:54.000 --> 17:55.000 And so, 17:55.000 --> 17:58.000 there's also a whole world of contributions to be made. 17:58.000 --> 18:00.000 I think there's still a lot of time in this presentation, 18:00.000 --> 18:01.000 but I'm done, 18:01.000 --> 18:03.000 and thank you. 18:04.000 --> 18:05.000 Thank you. 18:11.000 --> 18:12.000 If you have any questions, 18:12.000 --> 18:13.000 feel free. 18:13.000 --> 18:14.000 Hi. 18:14.000 --> 18:15.000 I wouldn't do. 18:15.000 --> 18:17.000 I would not give too much in your presentation.