08:18 <+bridge> any reason ddnet doesn't use submodules on a per-dependency basis? 08:36 <+bridge> Why should it 09:00 <+bridge> For one, if in the future this dependency repo doesn't exist, you won't be able to fetch them & will have to have them installed 09:00 <+bridge> And it's just another link in the potential chain of failure, and I'm not sure what the benefit of having them in their own repo is other than observed tidiness 09:00 <+bridge> If you need a particular configuration of snapshot commits in each submodule, I'm not sure why you can't just do that with each commit in the ddnet repo 09:00 <+bridge> It would be more useful to use cmake sub directories 09:00 <+bridge> So it Auto compiles the dependencies 09:00 <+bridge> I didn't realize it doesn't do that already. But that's not strictly related to how they're fetched 09:00 <+bridge> Just having a deps subdir or something makes more sense, in my opinion 09:01 <+bridge> our dependencies are handled by package managers like apt 09:01 <+bridge> i doubt they would disappear, they are not on a github repo 09:01 <+bridge> actually, i would be more worried about github disappearing 09:02 <+bridge> Anyway I don't see the breaking thing u mention. Old versions have old submodule commits 09:02 <+bridge> I suppose this is something you couldn't do without having it the way it is now 09:02 <+bridge> https://cdn.discordapp.com/attachments/293493549758939136/1065541720235851786/image.png 09:02 <+bridge> Ah yeah SDL is always fun on windows and macos 09:03 <+bridge> this is my q answered then 09:04 <+bridge> i was just thinking about how I would have done it; by getting the deps directly in their own submodules 09:04 <+bridge> and i doubt having them in their own repo was the first choice/how it was from the start. there was probably something that made this necessary, right? 09:05 <+bridge> I think the best is to use a packet manager. There is for example vcpkg 09:05 <+bridge> https://vcpkg.io/en/index.html 09:05 <+bridge> Or Conan but I dunno 09:12 <+bridge> If that's the case, I would be kind of concerned 09:12 <+bridge> - Releases of libraries that get published to package managers are more second-class instances than those on the upstream repo's main branch, which (if the library has competent maintainers) shouldn't break. Just because they are less common 09:12 <+bridge> - Releases of libraries that get published to package managers may be much more prone to disappear or stop being maintained, because often times the demand isn't high enough or such deployment is handled by a third party 09:12 <+bridge> - A break in your package-fetching CI or the package's local install scripts, or the package manager itself leaves you open to problems where submodules are more sound -- not perfect, but less prone to disappearance at any given moment 09:12 <+bridge> And if you're using a library that doesn't have a git instance somewhere (GH or GitLab), either host it yourself or find something more modern/better maintained 😃 09:13 <+bridge> In current year that's pretty much a necessity 09:14 <+bridge> I'm pretty much positive everything here has an official repository 09:14 <+bridge> https://cdn.discordapp.com/attachments/293493549758939136/1065544668194017350/image.png 09:14 <+bridge> Build steps are trivial 09:16 <+bridge> Anyway. U'll need to accept almost all devs here are Linux devs and simply use the os package manager 09:16 <+bridge> Yeah fair enough 09:16 <+bridge> Depending on system local libs is good practice if a solution like what you have in place is around 09:16 <+bridge> I just wonder why it's been executed this way... 09:17 <+bridge> Probably BCS teeworlds did it like this 09:17 <+bridge> And the code base is over 13 years old 09:17 <+bridge> Well that would have been a good answer from the start xD 09:20 <+bridge> Good thing is, we added rust support which comes with cargo. Which handles dependencies cleanly 09:20 <+bridge> Yeah... rust dependencies, maybe 09:20 <+bridge> So maybe it changes in future anyway 09:34 <+bridge> it's to not baloon the size of the ddnet repo 09:35 <+bridge> it's to not balloon the size of the ddnet repo 09:36 <+bridge> discord likely doesn't. sqlite3 doesn't want to be compiled from the repo. curl likely has one, not sure if they want to be compiled from the repo. rest I don't know about 10:10 <+bridge> They don't want to be compiled from repo? That sounds kinda stupid xd 10:11 <+bridge> Is our SQLite built not self built? 10:13 <+bridge> but from a release, I'd hope 10:14 <+bridge> https://sqlite.org/howtocompile.html 10:14 <+bridge> > Building SQLite directly from individual source code files is certainly possible, but it is not recommended. For some specialized applications, it might be necessary to modify the build process in ways that cannot be done using just the prebuilt amalgamation source file downloaded from the website. For those situations, it is recommended that a customized amalgamation be built (as described below) and used. In other words, even if a project 10:14 <+bridge> > The use of the amalgamation is recommended for all applications. 10:14 <+bridge> > 10:14 <+bridge> > Building SQLite directly from individual source code files is certainly possible, but it is not recommended. For some specialized applications, it might be necessary to modify the build process in ways that cannot be done using just the prebuilt amalgamation source file downloaded from the website. For those situations, it is recommended that a customized amalgamation be built (as described below) and used. In other words, even if a project 10:19 <+bridge> Mh ok. SQLite is weird anyway 10:26 <+bridge> Repo size 10:26 <+bridge> Submodules 10:27 <+bridge> Ah heinrich answered before me 10:29 <+bridge> Hm, isn't that what we did? 10:29 <+bridge> My question was more why they’re not in their own submodules 10:31 <+bridge> You mean 1 per dependency? I guess we didn't really see a benefit to it 10:31 <+bridge> Yeah. I was told that’s how it was done in teeworlds 10:32 <+bridge> We don't mind straying from tw in those kinds of things, if we'd seen a reason to do that we would have I think 10:32 <+bridge> Why do you think it'd be better to have them in separate repos? 10:34 <+bridge> I explained a fair amount above. If you’re interested you can scroll, it’s not too far 10:36 <+bridge> I think main reason Ewan pointed out was those modules going extinct 10:36 <+bridge> If 1 repo with all dependencies can disappear, so can more than one. I don't see why amount of repos changes anything. 10:36 <+bridge> I think main reason Ewan pointed out was those modules going extinct and being less stable than if they would be taken from a repository. 10:38 <+bridge> Or maybe you mean have their entire sources instead of built dependencies? 10:39 <+bridge> Yeah, but that’s not entirely feasible with the discord/curl/sqlite developments 10:39 <+bridge> But, some prebuilts is better than all, IMO 10:41 <+bridge> I guess that is doable, we do already build some of these on our own 10:47 <+bridge> I’d be happy to stage some changes 10:48 <+bridge> >DDNet will outlive curl & sqlite 10:49 <+bridge> Huh? 10:49 <+bridge> Isn't that why you want them to build it themself? 10:50 <+bridge> Because I don't know how this world works. But their repos and modules will exist for a long time. No reason to delete it as long people use it. 10:50 <+bridge> Because I don't know how this world works. But their repos and modules will exist for a long time. No reason for them to delete it as long people use it. 10:51 <+bridge> Building with dependency sources vs. dynamic libs can decrease package size with LTOs. My point wasn’t that the libraries would croak, but the ways they’re deployed & used by ddnet 11:01 <+bridge> I wouldn't start anything without getting some agreement in place. We have many scripts building stuff that'd need changing. If linkage is changing we have some interesting linking quirks (like the curl stub). Our CI times(or build times in general) are also a concern, so some sort of prebuilt dependency will need to be there. We have a flatpak release that'll need those too 11:02 <+bridge> It's not a trivial task and I wouldn't want you to waste your time just for something that would be rejected 11:02 <+bridge> Right 11:03 <+bridge> CI times would change only if using the repo’s dependencies, which I wouldn’t expect to be default behavior for CI 11:04 <+bridge> If I could hear some more about the scripts & the curl stub, that’d be great 11:07 <+bridge> im against adding them as submodules 11:07 <+bridge> even windows ones, ddnet-libs should be removed 11:07 <+bridge> and use a package manager 11:07 <+bridge> its c++/cmake fault they dont do it like cargo 11:13 <+bridge> > Releases of libraries that get published to package managers are more second-class instances than those on the upstream repo's main branch, which (if the library has competent maintainers) shouldn't break 11:13 <+bridge> I would say otherwise, im sure libraries you get from debian stable are way more first-class (with custom patches!) and fit more into the ecosystem than upstream 11:13 <+bridge> > Releases of libraries that get published to package managers may be much more prone to disappear or stop being maintained 11:13 <+bridge> This depends on what package manager u use, on the linux world, i would say, debian, nixos, arch have quite the reputation, i would see it first that upstream disappears and debian still has the library in its ecosystem tbh. 11:13 <+bridge> > And if you're using a library that doesn't have a git instance somewhere (GH or GitLab), either host it yourself or find something more modern/better maintained 11:13 <+bridge> ??? 11:13 <+bridge> Could a package manager accomodate our needs? Could it facilitate a source build of the dep like Ewan wanted 11:13 <+bridge> I don't think vcpkg can e.g. 11:14 <+bridge> it does on linux 11:14 <+bridge> Good point on the second one. God knows how many packages I've submitted & abandoned 11:14 <+bridge> i couldnt care less about windows tbh 11:14 <+bridge> xd 11:15 <+bridge> You and Jupstar should form a game studio 😄 11:15 <+bridge> I didn't realize we were talking about package manager's on the user's system. I thought you were saying you used something like apt with custom repos for the binaries in ddnet-libs 11:15 <+bridge> Package manager is totally irrelevant now because we're talking about the scenario where system libs are not being used 11:16 <+bridge> in c++? i cant see that being realistic 11:16 <+bridge> we should use rust then 11:16 <+bridge> Hmm? 11:16 <+bridge> It's not hard 11:16 <+bridge> He is talking about one of those new package managers like vcpkg that is for C++ dependencies 11:16 <+bridge> Harder in C++ but not hard lol 11:17 <+bridge> submodules w/ cmakelists or (god forbid) cmake fetch content 11:17 <+bridge> Yeah I meant way back in the convo 11:17 <+bridge> When apt was brought up 11:17 <+bridge> vcpkg is not a bad idea. just a righteous pain in the ass to use 11:17 <+bridge> Let's go 11:18 <+bridge> :monkaS: 11:18 <+bridge> This conversation is drifting all over the place 😄 11:18 <+bridge> Let's static link drivers 11:18 <+bridge> Only 1 generation of gpu supported. Linux 6.1 required 11:19 <+bridge> Sounds amazing 11:19 <+bridge> Glibc must be rebuilt before launching the game, just to have the latest one 😛 11:21 <+bridge> Imagine the world could drop support for old stuff every 4 years. Would be amazing and nice adventure. Much more fun than it is rn 11:21 <+bridge> Plus no backward compatibility 11:23 <+bridge> Min support vulkan 1.1 11:23 <+bridge> C++22/rustc 1.66.1 11:23 <+bridge> Min support 32G ram 11:23 <+bridge> C++22 LOL 11:23 <+bridge> I seen projects do it 11:23 <+bridge> Do C++26, there is a proposition out already 11:24 <+bridge> Minimum should be C++26. No compromises 11:24 <+bridge> Damn 11:24 <+bridge> Not crazy 11:24 <+bridge> you stole my line 11:24 <+bridge> Ohh 11:24 <+bridge> Well i prefer rust 11:24 <+bridge> It's indeed less crazy than many ppl here imply 11:24 <+bridge> cpp22 is not a thing 11:24 <+bridge> Nightly? 11:24 <+bridge> I also seen many projects simply do it 11:24 <+bridge> Nightly has rly interesting stuff, but with stable havent had problems 11:25 <+bridge> Polybar did it before distros had compilers ready. It was supremely annoying to use for a while ngl 11:25 <+bridge> I defs miss features in rust 11:25 <+bridge> Ah its ,23 11:26 <+bridge> I could imagine blender will sooner or later also only allow new graphics apis. They already dropped opencl 11:26 <+bridge> Me ❤️ `Vec::drain_filter` 11:26 <+bridge> One always wants more 11:26 <+bridge> U can actually do it in a crate, or make a pr to rust 11:26 <+bridge> Core language features tho 11:27 <+bridge> Itertools often has features that get added to std 11:27 <+bridge> Didn't they also work like shit for like a year there on amd + linux? 11:27 <+bridge> I recommend that crate 11:27 <+bridge> Oh like what 11:27 <+bridge> Yeah and now it's nice 11:27 <+bridge> I want them to finish polonius 11:28 <+bridge> The better borrow checker, that accepts more valid programs 11:28 <+bridge> https://github.com/rust-lang/polonius 11:28 <+bridge> A project like blender can rip the bandaid and have their userbase pressure the correct people to get it working. There are actual companies with capital using blender 11:28 <+bridge> I doubt a game has similar pull 11:29 <+bridge> Cleaner default initialization. More constexpr. Some generic stuff. Many small things 11:29 <+bridge> @Learath2 valve 11:29 <+bridge> The const is on going 11:30 <+bridge> Last I checked this fixed some very odd edge cases, do you actually hit those often enough? 11:30 <+bridge> @Jupeyy_Keks they always add first a MVP so they dont shoot their foot with back compat in case smth was wrong 11:30 <+bridge> Mvp = min viable product 11:30 <+bridge> So not many features 11:30 <+bridge> Thats why const is limited 11:31 <+bridge> And even they gave up and shipped a bundle of deps, restricted to a specific version of ubuntu and called it a day 😛 11:31 <+bridge> Xd 11:31 <+bridge> Personally i want better async 11:32 <+bridge> Its not that bad but its hard to get right i guess 11:32 <+bridge> Thats why GATs need to have more features 11:32 <+bridge> Currently GATs are MVP 11:32 <+bridge> Simpler stuff in rust async is actually pretty nice. It makes sense, works well, looks pretty 11:33 <+bridge> Anyway atleast we have a systems lang with first class async 11:33 <+bridge> Idk what c++ has 11:33 <+bridge> But im sure its not that pretty 11:33 <+bridge> C++ has async too 😄 11:33 <+bridge> With the keyword? 11:33 <+bridge> Nothing is pretty in C++ tho 11:34 <+bridge> Tru xd 11:34 <+bridge> When do u even want async? 11:34 <+bridge> I think we are finally getting the keyword in C++26 😛 11:34 <+bridge> I mean for io yes 11:34 <+bridge> But else it also makes code much harder 11:34 <+bridge> Xd 11:34 <+bridge> https://cdn.discordapp.com/attachments/293493549758939136/1065580054584770580/Screenshot_2023-01-19-11-34-28-033_org.mozilla.firefox.jpg 11:34 <+bridge> @Jupeyy_Keks http makes lot of sense async 11:34 <+bridge> Lots of waiting etc 11:35 <+bridge> Yeah, but e.g. when I sometimes see JavaScript devs using async for all shit 11:35 <+bridge> Just to make everything more complicated 11:37 <+bridge> Ah ye 11:37 <+bridge> Js is special kid 11:37 <+bridge> I want to abolish js 11:37 <+bridge> Thats why we need wasm 11:37 <+bridge> With direct DOM access 11:37 <+bridge> Not through js 11:37 <+bridge> Asynchronously running code in general is too complex for you? 11:38 <+bridge> I mean there is async and async. If u use async and the async code is completely seperated code.. it's ofc smth different than using it in a stateful code, where async also means Ur state isn't updated until the code ran 11:38 <+bridge> Ppl use nodejs for server side with the excuse to use the same lang as frontend 11:38 <+bridge> And call them fullstack deca 11:38 <+bridge> It's by nature more complex 11:38 <+bridge> And call them fullstack devs 11:38 <+bridge> Its true its more complex 11:38 <+bridge> He is not saying he doesnt have the skills for it lel 11:38 <+bridge> .NET async is worse than JS’s IMO 11:38 <+bridge> Time race 11:39 <+bridge> Tasks vs promises 11:39 <+bridge> Urgh 11:40 <+bridge> Tokio gets it rly right along with the rust std 11:40 <+bridge> As in 11:40 <+bridge> task::spawn is identical to thread::spawn 11:41 <+bridge> Sux to be on mobilr 11:41 <+bridge> https://cdn.discordapp.com/attachments/293493549758939136/1065581815584276540/Screenshot_2023-01-19-11-40-55-530_org.mozilla.firefox.jpg 11:41 <+bridge> This metro has such bad connection in some stations 11:42 <+bridge> Oh that is so easy 11:42 <+bridge> https://docs.rs/tokio/latest/tokio/task/ 11:59 <+bridge> I guess async appears a lot in web presentation because of the prevalance of long running things. Api requests, animations, etc 12:00 <+bridge> Do you get a connection while in the tunnel btw? As in not at a station yet? 12:01 <+bridge> s\/?$// 12:01 <+bridge> Yes i do 12:01 <+bridge> But some tunnels have worse 12:02 <+bridge> Istanbul metro still doesn't have that, this is so sadge 12:02 <+bridge> The ones nearer poorer zones, casuality? Doubt 12:02 <+bridge> Xd 12:02 <+bridge> Everyone else figured it out, but not istanbul 12:02 <+bridge> syntax error 12:02 <+bridge> Tokio got it too, first class 12:03 <+bridge> did you mean `s/\?$//` 12:03 <+bridge> Nooooo, discord ate one of my \\ 12:03 <+bridge> Yep. But everywhere, where the code will not block I wouldn't use it I think. But I understand that it makes sense in the ecosystem^^ 12:04 <+bridge> Sry where the code will definitely block 12:04 <+bridge> As in no io 12:04 <+bridge> Yeah, I've seen some questionable async use 12:06 <+bridge> Async is good for waiting 12:06 <+bridge> Xd 12:19 <+bridge> Could get the new insider word for ppl that always wait. Asyncs 12:40 <+bridge> https://pandoc.org/releases.html 12:40 <+bridge> pandoc 3.0 released 12:40 <+bridge> one of the best software 12:40 <+bridge> make a markdown document, and get the beauty of latex 12:41 <+bridge> ofc not the full latex power but its smth 13:24 <+bridge> I recently got math notes from some guy and apparently he does them all in LaTeX, his notes look so cool 13:24 <+bridge> I tried doing latex notes once but I'm just not fast enough in writing it to take notes during a class 13:57 <+bridge> the hard parts of doing latex notes in class are formulas 13:57 <+bridge> they're easy to write by hand but hard in latex 13:57 <+bridge> also commutative diagrams, but I got https://q.uiver.app/ for that nowadays 14:06 <+bridge> one day, ddnet coded in latex 😄 14:07 <+ChillerDragon> i watched some dudes latex/vim maffs notes setup on youtube once eventho i dont do maffs nor latex and still was kinda cool 14:10 <+bridge> a site with a lot of cool stuff made in latex 14:10 <+bridge> https://texample.net/tikz/examples/ 14:12 <+bridge> some 3D graphes i made with latex for an internship 14:12 <+bridge> https://cdn.discordapp.com/attachments/293493549758939136/1065619731433930783/image.png 14:15 <+bridge> lol, crazy guy drew indian map 14:15 <+bridge> https://texample.net/tikz/examples/india-map/ 14:29 <+bridge> Didn't used MinGw64 in my life what am I supposed to do lol 14:31 <+bridge> does anybody know if a server i have with hetzner (german company) in helsinki, finland, needs to abide to finnish law or german law? 14:31 <+bridge> btw soon ill migrate stuff i have in 2 servers to this new one, including the ddnet wiki 14:31 <+bridge> and fix the email for chinese qqq which for some reason rejects me for dmarc 14:32 <+bridge> cheap hoster ips are often blacklisted for email 14:32 <+bridge> maybe i can just use the gandhi provided smtp 14:32 <+bridge> they give 3 free emails per domain 14:32 <+bridge> xd 14:33 <+bridge> some chinese ppl tried to register but the email wont reach them and i get the rejection sadly 14:34 <+bridge> self-hosting email is pretty broken 14:34 <+bridge> works fine for me 14:34 <+bridge> You won't reach anyone from Microsoft or Google to unblock your ip address 14:34 <+bridge> well my emails reached google emails 14:35 <+bridge> but ye 14:35 <+bridge> Then you're lucky that your ip was not previously used by a spammer 14:35 <+bridge> but why when i have the technical prowess to self host a email and use the proper protocols get rejected 14:35 <+bridge> smh humanity at its best 14:37 <+bridge> the technical part is solved and easy, making sure your emails actually reach every other email provider is impossible 14:39 <+bridge> yeah 14:39 <+bridge> anyway ill use gandhi 14:39 <+bridge> i hope they use smth reputable 14:40 <+bridge> https://www.gandi.net this i mean xd 14:40 <+bridge> i believe you need to abide to local server law 14:41 <+bridge> ie finnish one here 14:41 <+bridge> thats good 14:41 <+bridge> germany is stricter 14:41 <+bridge> iirc 14:41 <+bridge> with copyright for example 14:41 <+bridge> you need to abide by finnish, german, and laws of countries who you target 14:42 <+bridge> oof xd 14:42 <+bridge> what if u target the world 14:42 <+bridge> thats odd 14:42 <+bridge> i target earthlings 14:42 <+bridge> :poggers2: 14:42 <+bridge> de facto, you're probably fine if you don't do too illegal things 14:42 <+bridge> yeah i dont plan to 14:42 <+bridge> but sometimes for example 14:42 <+bridge> user generated content 14:42 <+bridge> hetzner null routed my ip 14:43 <+bridge> till i deleted it 14:43 <+bridge> idk if i missed their notice or if they did it instantly 14:43 <+bridge> but damn 14:43 <+bridge> Is there any wiki to compile ddnet with MinGw64? 14:43 <+bridge> and it was all cuz my url shortener had a sus link someone added 14:43 <+bridge> should be like linux 14:43 <+bridge> iirc 14:43 <+bridge> mingw is linux like 14:44 <+bridge> just use cmake like this 14:44 <+bridge> ```bash 14:44 <+bridge> mkdir build && cd build 14:44 <+bridge> cmake -G "MinGW Makefiles" .. 14:44 <+bridge> make 14:44 <+bridge> ``` 14:44 <+bridge> you can download the binaries here, btw: https://github.com/AllTheHaxx/AllTheHaxx/releases/tag/1.0.0 14:46 <+bridge> What would i do with them...? 14:47 <+bridge> idk, what are you trying to compile? ^^ 14:50 <+bridge> Ath but i modified it 14:52 <+bridge> this smells like bot client 15:02 <+bridge> Nope, I'm just playing with it lol 15:02 <+bridge> bot 15:03 <+bridge> :think_bot: :botmorning: 15:03 <+bridge> new user flag + asks about compiling on ***windows*** 15:03 <+bridge> big red flag 15:03 <+bridge> Why 15:03 <+bridge> u are not the first to come here xD 15:03 <+bridge> Playing on windows is a problem now ? 15:04 <+bridge> Duh 15:04 <+bridge> Crucial question do u prefer C++ or Rust @HestiaSpider ? 15:05 <+bridge> what do you expect from me regardless of my answer? 15:05 <+bridge> wanted to see Ryozuki's anwser xD 15:06 <+bridge> anyway I agree that it seems sus that you "modified" ath and don't know how to compile and it's not a bot 15:06 <+bridge> Because that's the first modded client I'm using lol 15:06 <+bridge> And also 15:07 <+bridge> I tried compiling normal teeworlds 15:07 <+bridge> But i couldn't lmao 15:07 <+bridge> teeworlds 0.7 or ddnet? 15:07 <+bridge> Teeworlds 0.6 15:08 <+bridge> Ddnet is even worse 15:08 <+bridge> ddnet/teeworlds 0.7 should be the easiest 15:11 <+bridge> Last time I remember being on windows I didn't have troubles compiling ddnet. Though for 16.6 upwards you will need rust I think 15:14 <+bridge> ah yea, that's a bit annoying 15:19 <+bridge> So 15:20 <+bridge> How do I compile my client lol 15:20 <+bridge> I can help but first is it a bot client? 15:20 <+bridge> Nope 15:20 <+bridge> Ofc you can add Lua scripts but that's your problem if u wanna cheat with it 15:21 <+bridge> Do you think you would be banned for server for the modifications you've done? 15:21 <+bridge> Nope, doesn't affect server side 15:22 <+bridge> @Reteah 15:25 <+bridge> yeah I'm back rebooted to windows 15:25 <+bridge> Kk 15:26 <+bridge> I suggest to DM to talk 15:26 <+bridge> I suggest to go DM to talk 15:29 <+bridge> I'll answer once I've compiled DDNet 15:31 <+bridge> K 15:45 <+bridge> cleanest htop i seen 15:45 <+bridge> https://cdn.discordapp.com/attachments/293493549758939136/1065643267565289482/image.png 15:46 <+bridge> wont last long tho xd 15:59 <+bridge> Does it take really that long ? 16:00 <+bridge> I have problem with well windows and mingw but no. Last time was 4 mins 16:07 <+bridge> @Jupeyy_Keks https://github.com/ddnet/ddnet/blob/4a4d6c54834465f7c9726b1c02cc8ccaf1a00ae6/src/engine/client/backend_sdl.cpp#L147-L151 16:07 <+bridge> this lock makes my mac hang 16:07 <+bridge> when changing for example windowed to fullscreen 16:07 <+bridge> well, when i change it hangs, and then i control-c on lldb and its there 16:07 <+bridge> backtrace 16:08 <+ChillerDragon> ryo on mac? 16:08 <+bridge> ``` 16:08 <+bridge> frame #2: 0x000000018fe16ac4 libc++.1.dylib`std::__1::condition_variable::wait(std::__1::unique_lock&) + 28 16:08 <+bridge> frame #3: 0x000000010001a3f4 DDNet`CGraphicsBackend_Threaded::WaitForIdle() + 80 16:08 <+bridge> ``` 16:08 <+bridge> got it from work ye 16:08 <+ChillerDragon> wowo 16:08 <+ChillerDragon> when? 16:08 <+bridge> since december 16:08 <+ChillerDragon> old news 16:08 <+ChillerDragon> how u like it? 16:08 <+bridge> ye 16:09 <+bridge> its like a bad joke of linux 16:09 <+ChillerDragon> xd 16:09 <+bridge> Yeh got almost the same problem when compiling ddnet 16:12 <+bridge> Stack trace would be more interesting 16:12 <+bridge> I assume it's on opengl 16:16 <+bridge> ogl 3.3 16:21 <+bridge> ``` 16:21 <+bridge> * thread #1, queue = 'com.apple.main-thread', stop reason = signal SIGSTOP 16:21 <+bridge> * frame #0: 0x000000018fe8d564 libsystem_kernel.dylib`__psynch_cvwait + 8 16:21 <+bridge> frame #1: 0x000000018fec9638 libsystem_pthread.dylib`_pthread_cond_wait + 1232 16:21 <+bridge> frame #2: 0x000000018fe16ac4 libc++.1.dylib`std::__1::condition_variable::wait(std::__1::unique_lock&) + 28 16:21 <+bridge> frame #3: 0x0000000100048934 DDNet`void std::__1::condition_variable::wait(this=0x0000000105826690, __lk=0x000000016fdfc748, __pred=(unnamed class) @ 0x000000016fdfc700)::$_2) at __mutex_base:404:9 16:21 <+bridge> frame #4: 0x00000001000488b0 DDNet`CGraphicsBackend_Threaded::WaitForIdle(this=0x0000000105826600) at backend_sdl.cpp:150:19 16:21 <+bridge> frame #5: 0x00000001000b5478 DDNet`CGraphics_Threaded::WaitForIdle(this=0x00000001085d8000) at graphics_threaded.cpp:3205:14 16:21 <+bridge> frame #6: 0x00000001000b4784 DDNet`CGraphics_Threaded::Swap(this=0x00000001085d8000) at graphics_threaded.cpp:3134:3 16:21 <+bridge> frame #7: 0x000000010006e738 DDNet`CClient::Run(this=0x0000000108020000) at client.cpp:3300:19 16:21 <+bridge> frame #8: 0x0000000100077910 DDNet`::TWMain(argc=1, argv=0x000000016fdff7e8) at client.cpp:4726:11 16:22 <+bridge> frame #9: 0x0000000100166524 DDNet`main(argc=1, argv=0x000000016fdff7e8) at client.mm:21:10 16:22 <+bridge> frame #10: 0x000000018fb9fe50 dyld`start + 2544 16:22 <+bridge> ``` 16:22 <+bridge> @Jupeyy_Keks this? 16:23 <+bridge> ``` 16:23 <+bridge> (lldb) thread list 16:23 <+bridge> Process 12560 stopped 16:23 <+bridge> * thread #1: tid = 0x25c78, 0x000000018fe8d564 libsystem_kernel.dylib`__psynch_cvwait + 8, queue = 'com.apple.main-thread', stop reason = signal SIGSTOP 16:23 <+bridge> thread #2: tid = 0x25cd2, 0x000000018fe9885c libsystem_kernel.dylib`sem_wait + 8 16:23 <+bridge> thread #3: tid = 0x25cd3, 0x000000018fe9885c libsystem_kernel.dylib`sem_wait + 8 16:23 <+bridge> thread #4: tid = 0x25cd4, 0x000000018fe9885c libsystem_kernel.dylib`sem_wait + 8 16:23 <+bridge> thread #6: tid = 0x25cd6, 0x000000018fe8ba1c libsystem_kernel.dylib`__workq_kernreturn + 8 16:23 <+bridge> thread #9: tid = 0x25cdf, 0x000000018fe8bb24 libsystem_kernel.dylib`__ulock_wait + 8 16:23 <+bridge> thread #10: tid = 0x25ce0, 0x000000018fe8ba1c libsystem_kernel.dylib`__workq_kernreturn + 8 16:23 <+bridge> thread #11: tid = 0x25ce1, 0x000000018fe8ba1c libsystem_kernel.dylib`__workq_kernreturn + 8 16:23 <+bridge> thread #12: tid = 0x25ce2, 0x000000018fe89cec libsystem_kernel.dylib`semaphore_wait_trap + 8, name = 'caulk.messenger.shared:17' 16:23 <+bridge> thread #13: tid = 0x25cfa, 0x000000018fe89d70 libsystem_kernel.dylib`mach_msg2_trap + 8, name = 'AudioQueue thread' 16:24 <+bridge> thread #14: tid = 0x25cfb, 0x000000018fe89cec libsystem_kernel.dylib`semaphore_wait_trap + 8, name = 'caulk.messenger.shared:high' 16:24 <+bridge>