| sfiet_konstantin[m] | Who is in FOSDEM ? | 08:16 |
|---|---|---|
| *** rdr_ is now known as rdr | 10:02 | |
| rubdos[m] | me | 10:16 |
| rubdos[m] | I've been trying to find the linux-on-mobile stand for five times now, but I think I'm just walking past | 10:16 |
| rubdos[m] | are they just the ubports people atm? Because I don't see any sailfish | 10:17 |
| rubdos[m] | sfiet_konstantin: ^ | 10:17 |
| sfiet_konstantin[m] | <rubdos[m]> "I've been trying to find the..." <- K second floor | 10:27 |
| sfiet_konstantin[m] | near stairs | 10:27 |
| sfiet_konstantin[m] | Was very crowded like 1h ago | 10:27 |
| sfiet_konstantin[m] | Actually K second floor is crowded | 10:39 |
| sfiet_konstantin[m] | nextcloud firefox vlc | 10:39 |
| sfiet_konstantin[m] | quite a few big stands | 10:39 |
| rubdos[m] | > <@sfiet_konstantin:matrix.org> K second floor | 10:42 |
| rubdos[m] | > near stairs | 10:42 |
| rubdos[m] | uh, I was searching in H | 10:42 |
| rubdos[m] | why do I think it was H | 10:42 |
| rubdos[m] | I'll come by after https://fosdem.org/2023/schedule/event/security_elliptic_curves_in_foss/ | 10:42 |
| rubdos[m] | okay no, after Thaodan's talk :'-) | 10:43 |
| rubdos[m] | I'll come by* | 10:43 |
| sfiet_konstantin[m] | rubdos[m]: Because it's H in the planning | 10:45 |
| sfiet_konstantin[m] | even cybette told me that | 10:45 |
| rubdos[m] | It's even in the fosdem23 app. :'-) | 10:45 |
| sfiet_konstantin[m] | Finishing the talk about Asahi Linux. Time to grab some belgian fries ! | 11:23 |
| Tunanodra | hello guys! | 11:28 |
| sfiet_konstantin[m] | Hello | 11:41 |
| Nico | > <@sfiet_konstantin:matrix.org> K second floor | 12:00 |
| Nico | > near stairs | 12:00 |
| Nico | Oh, the UBPorts people were looking for you next to the Gentoo stand in the basement | 12:00 |
| x2s | .oO( you are lucky I'm not there, I'd bother you with questions about IV ;) | 12:04 |
| sfiet_konstantin[m] | Ruben De Smet: will you ever come to the Rust devroom ? | 13:13 |
| sfiet_konstantin[m] | if so, don't. Prolly the worst room (and a crowded one) | 13:13 |
| mal | rubdos[m]: did you see my recent PRs based on your llvm and rust PRs, I'm planning on merging those | 13:17 |
| sfiet_konstantin[m] | mal: mind sharing a link ? :) | 13:59 |
| sfiet_konstantin[m] | (I'm curious about them too) | 14:00 |
| mal | https://github.com/sailfishos/llvm/pull/2 and https://github.com/sailfishos/rust/pull/18 also mesa https://github.com/sailfishos/mesa/pull/2 | 14:04 |
| sfiet_konstantin[m] | nice ! | 14:28 |
| sfiet_konstantin[m] | interesting ! | 14:28 |
| mal | not the very latest version because of some issues but still newer | 14:29 |
| sfiet_konstantin[m] | what kind of issues: segfaults ? | 14:33 |
| mal | gecko build has issues with newer rust | 14:36 |
| mal | I tried to backport some patches but it started to become too big so I decided to update only to that for now | 14:36 |
| sfiet_konstantin[m] | ok ... I think we can live with the rust version compiling ghecko right | 14:37 |
| sfiet_konstantin[m] | s/ghecko/gecko/ | 14:37 |
| mal | also some cross compile issue with newer rust that I didn't yet figure out, some extra changes needed probably to the patches | 14:37 |
| sfiet_konstantin[m] | cross compiling ... something I never really understood properly :( | 14:39 |
| mal | yeah, it's complicated sometimes | 14:40 |
| sfiet_konstantin[m] | Is there a SFOS BOF today at fosdem ? | 15:07 |
| sfiet_konstantin[m] | (maybe I already missed it ?) | 15:07 |
| mal | sfiet_konstantin[m]: BOF? | 15:44 |
| rubdos[m] | <sfiet_konstantin[m]> "Is there a SFOS BOF today at..." <- abr told me it's tomorrow | 16:28 |
| rubdos[m] | but I could misremember | 16:28 |
| sfiet_konstantin[m] | <mal> "sfiet_konstantin: BOF?" <- Bird of feather | 16:29 |
| rubdos[m] | <Nico> "> <@sfiet_konstantin:matrix.org>..." <- Ah, that'll have to be tomorrow, then. I'll go say hi in the morning :-) | 16:29 |
| rubdos[m] | > <@sfiet_konstantin:matrix.org> Ruben De Smet: will you ever come to the Rust devroom ? | 16:29 |
| rubdos[m] | > if so, don't. Prolly the worst room (and a crowded one) | 16:29 |
| rubdos[m] | I tried in the morning, and I ran. | 16:29 |
| rubdos[m] | <mal> "Ruben De Smet: did you see my..." <- I saw them indeed, abr told me again today about their existence. I'm very very happy you took it on. | 16:30 |
| rubdos[m] | Is there anything in particular you want me to try out on those PRs, mal ? Just check whether it's recent enough for me? | 16:31 |
| rubdos[m] | Nico: Right, nanu-c told me about the UBports people, I should've remembered. | 16:35 |
| rubdos[m] | This is the first time for me that I'm meeting so many people, it's quite something. | 16:35 |
| sfiet_konstantin[m] | rubdos[m]: Too much maybe ? | 16:35 |
| rubdos[m] | Too much to remember, but I don't mind meeting people | 16:36 |
| sfiet_konstantin[m] | It's taxing for sure | 16:36 |
| rubdos[m] | I do mind the crowd, so I went home after my chat with the dev of Molly | 16:36 |
| rubdos[m] | it's impossible to get into rooms anyway :'-) | 16:36 |
| rubdos[m] | Q: about the dinner, will people bring "stuff"? Like, am I supposed to show my broken N9 and my broken Jolla in order to get in? ;-) | 16:37 |
| sfiet_konstantin[m] | Bring yourself, that should be fine ! | 16:58 |
| mal | rubdos[m]: nothing special to check, rust is the same version you used | 16:59 |
| mal | I just did some fixes to it | 16:59 |
| rubdos[m] | <mal> "Ruben De Smet: nothing special..." <- 1.61, that'sawesome | 17:49 |
| rubdos[m] | <sfiet_konstantin[m]> "Bring yourself, that should be..." <- Brought myself | 17:49 |
| Nico | We'll be a bit late for the dinner, but I think we are in the right bus | 17:49 |
| sfiet_konstantin[m] | Nico: :D in the 71 here | 17:51 |
| sfiet_konstantin[m] | are you in it too ? | 17:51 |
| Nico | Onr of them right at the front | 17:53 |
| sfiet_konstantin[m] | Onr ? :D | 17:54 |
| Nico | One | 17:54 |
| Nico | Onehand typing to not fall over! | 17:55 |
| sfiet_konstantin[m] | I wonder if you are near to me lol | 17:57 |
| sfiet_konstantin[m] | i'm seated, facing back of bus and wearing a blue mask | 17:57 |
| Nico | I can't see any face of someone facin backwards. I'm next to Chris at the front door, long hair | 18:00 |
| sfiet_konstantin[m] | Fine ... | 18:01 |
| rostam98[m] | Hello I need some help to understand Sailfishos Sensor framework, is this a right place to ask questions? | 19:16 |
| mal | rostam98[m]: what do you want to know? | 19:26 |
| rostam98[m] | Great thanks, I am reading the Sailfishos and trying to understand, how the sensor framework works. I am about to get one of the already supported phones that supports Sailfish, and I need to access the data from Accelerometer and Gyroscope sensors. From reading the document, it seems Sensor framework works with Android OS. I may be very wrong please help me ... | 19:29 |
| mal | sensors in Sailfish are usually accessed via QtSensors which uses sensorfw backend which then communicates to the android hal sensor process which is running in the background | 19:36 |
| mal | QtSensors can be accessed either via qml and c++ interfaces depending on application needs | 19:39 |
| rostam98[m] | @mal from what I have read, I understand Sailfish os supports Linux and Android kernel. So my assumption is if I want to use Sensor I need to have an Android image running on the device. The confusion on my side is, how Android and Generic Linux kernels run at the same time on the device. Please help me to clear up my mis understanding. | 19:42 |
| mal | well Sailfish can run on linux and android kernel, most just use android kernel but there some community ports which use linux kernel but only very few | 19:43 |
| rostam98[m] | In that case for my use case, I should use case (Sensor). Now Android has its own GUI framework, but we Sailfishos supports QT and does not require using the Android part, is this correct understanding? | 19:46 |
| mal | not sure what the thing is that you don't understand | 19:47 |
| mal | sensorfw can use either linux kernel interfaces directly (iio in this case) or use android hal, it depends on the device | 19:47 |
| rostam98[m] | This is what I got from the document for sensor call chain: | 19:52 |
| rostam98[m] | sensorw-qt5->sensorfw-qt5-hybris->libgbinder->Android BSP HAL->android.hardware.sensors | 19:52 |
| rostam98[m] | The above call chain seems it only supports Android that is the part I do not understand. | 19:52 |
| mal | like I said sensorfw can use also linux kernel iio interface in which case sensorfw-qt5-hybris is not even installed | 19:53 |
| mal | in that case it would just be sensorfw-qt5->sensorfw->linux kernel | 19:54 |
| rostam98[m] | oh okay, I see, thanks for the clarification. Is there documentation or sample implementation that I can access and leverage from it? | 19:55 |
| mal | which device are planning on using? | 19:56 |
| mal | you only need to think what happens below sensorfw (or even sensorfw) when porting to a new device | 19:56 |
| rostam98[m] | We are planning to use one of the latest Sony phone that is supported, I am not sure about model number at this time. | 19:57 |
| rostam98[m] | The HAL layer should have some support, such as API for what happens below sensorfw, is this correct assumption? | 19:58 |
| mal | those use the android hal via sensorfw binder plugin, you really don't need to think about that | 19:59 |
| rostam98[m] | Sorry by those do you mean, Sony phones? | 20:00 |
| mal | yes | 20:05 |
| rostam98[m] | Thanks so much | 20:05 |
| mal | what happens at sensorfw level is just hardware adaptation details and not relevant for normal use or app development | 20:05 |
| rostam98[m] | So we are trying to port our application to a phone that utilizes sensors. Our code is implemented purely in C (not C++), and we plan to use Sailfishos for some of the security features that we like. | 20:08 |
| mal | sensorfw can be accessed via dbus also | 20:09 |
| mal | if Qt is not an option | 20:09 |
| rostam98[m] | OT is okay as long as we stay with pure Linux. dbus API can be accessed by higher-layer library named, gio. Is that also supported on Sailfishos? | 20:10 |
| mal | any dbus library available in sailfish can be used if needed | 20:11 |
| mal | Qt is just much easier for sensor use | 20:11 |
| rostam98[m] | So I believe we can use QT on top of the dbus and make our life easier? | 20:12 |
| mal | qt does things directly to sensorfw and not dbus afaik | 20:13 |
| mal | so if you use QtSensors you really don't need to think about anything related to sensorfw, it just works | 20:14 |
| rostam98[m] | If I understand you correctly, if we use QTSensors then it should work regardless we are using Android or Linux kernels, if that is the case, it would be great, dealing with dbus directly is painful. | 20:16 |
| attah | If you mean *on Sailfish OS*, sure! That's what he has been trying top tell you this whole time! | 20:20 |
| attah | There is really no "Android kernel"... or it is just a tweaked Linux kernel anyway. It has nothing whatsoever to do with the Android App Support that runs *on top of* that, if that is what has you confused... | 20:23 |
| rostam98[m] | Okay I think I am getting to understand. thank you both. | 20:27 |
| rostam98[m] | Seems most of the complexity we need to deal with is at the HAL layer. I think... | 20:27 |
| attah | Unless you are porting to a new device, NO! | 20:27 |
| attah | As a user you simply don't have to care at all. | 20:27 |
| mal | like I said many times, for user or app developer there is no need to thing about what happens behind QtSensors (or sensorfw dbus api if that is used, it very rarely is if ever, everything uses QtSensors apis) | 20:29 |
| mal | *think | 20:30 |
| rostam98[m] | I think I am very clear now. | 20:33 |
| rostam98[m] | Just to give you a background, we are working with a mobile hardware vendor and the final plan is to port sailfish on their device and have our app work on that device. | 20:33 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!