0

[ANN] chelydra v0.0.6: a snapcast client launcher

preview

chelydra is a GUI interface for starting and stopping snapclient.

It’s an MVP – it’s the bare minimum to do what I need it to do, which is provide a toggle to start and stop snapclient. It’s Go & gioui - gioui is a fabulous library, but it doesn’t provide drop-downs so until I hack up my own you have to manually enter the correct values in text fields.

This is meant for phones. It’s laid out assuming limited screen space in portrait mode. It’s a single binary, although the package contains a .desktop and icon so it shows up in a launcher. It uses almost no CPU, as you’d expect.

Importantly, I do build packages, and there are arm64 .debs that can be downloaded and installed – I think all of the mobile Linuxii are debian-based. You can find a link to the assets on the project page. I have only an FLX1s to test on, but I’m cross-compiling on an amd64 and there shouldn’t be anything particularly coupled with FuriOS.

I don’t use AI, and none was involved in chelydra. It’s, like, 81 LOC… why would there be?

An Aside

Mobile Linux doesn’t benefit much from the vast library of software available to most Linuxes; some of this is because a majority of GUI programs are designed for landscape windows with a lot of real estate for multiple panels, menus, and so on. A second reason is because a lot of stuff still doesn’t compile cleanly for arm64, which is what most mobile phones have in them. And oddly, when I do find projects, many of them don’t provide binaries for some reason, and it can be difficult to gather the requirements to compile binaries for arm. Consequently, I’m constantly running into situations where I need some little program for which there are dozens of options for the desktop and yet none for my phone. I’ve started putting them here – there are only two at the moment, but there will be more.

chelydra is a GUI interface for starting and stopping snapclient.

It’s an MVP – it’s the bare minimum to do what I need it to do, which is provide a toggle to start and stop snapclient. It’s Go & gioui - gioui is a fabulous library, but it doesn’t provide drop-downs so until I hack up my own you have to manually enter the correct values in text fields.

This is meant for phones. It’s laid out assuming limited screen space in portrait mode. It’s a single binary, although the package contains a .desktop and icon so it shows up in a launcher. It uses almost no CPU, as you’d expect.

Importantly, I do build packages, and there are arm64 .debs that can be downloaded and installed – I think all of the mobile Linuxii are debian-based. You can find a link to the assets on the project page. I have only an FLX1s to test on, but I’m cross-compiling on an amd64 and there shouldn’t be anything particularly coupled with FuriOS.

I don’t use AI, and none was involved in chelydra. It’s, like, 81 LOC… why would there be?

An Aside

Mobile Linux doesn’t benefit much from the vast library of software available to most Linuxes; some of this is because a majority of GUI programs are designed for landscape windows with a lot of real estate for multiple panels, menus, and so on. A second reason is because a lot of stuff still doesn’t compile cleanly for arm64, which is what most mobile phones have in them. And oddly, when I do find projects, many of them don’t provide binaries for some reason, and it can be difficult to gather the requirements to compile binaries for arm. Consequently, I’m constantly running into situations where I need some little program for which there are dozens of options for the desktop and yet none for my phone. I’ve started putting them here – there are only two at the moment, but there will be more.

0

RISC-V and micro kernel synergy

preview

I came across the post about Milk-V Titan, and there was a comment asking about the lack of the V extension would hinder running Ubuntu 25.10 which was targetting a particular RISC-V configuration, and it made me wonder if there were an opportunity for micro kernels to exploit.

Now, up-front: it’s been literally decades since I had an OS design class, and my knowledge of OS design is superficial; and while I’ve always been interested in RISC architectures, the depth of my knowledge of that also dates back to the 90’s. In particular (my knowledge of) RISC-V’s extension design approach is really, really shallow. It’s all at a lower level than I’ve concerned myself with for years and years. So I’m hoping for an ELI-16 conversation.

What I was thinking was that a challenge of RISC-V’s design is that operating systems can’t rely on extensions being available, which (in my mind) means either a lot of really specific kernel builds – like, potentially an exponential number – or a similar number of code paths in the kernel code, making for more complicated and consequently more buggy kernels (per the McConnell rule). It made me wonder if this is not, then, an opportunity for micro kernels to shine, by exploiting an ability to load extension-specific modules based on a given CPU capability set.

As I see it, the practicality of this depends on whether the extensions would be isolatable to kernel modules, or whether (like the FP extension) it’d just be so intrinsic that even the core kernel would need to vary. Even so, wouldn’t having a permutation of core kernel builds be smaller, more manageable, and less bug-prone than permutations of monolithic kernels?

Given the number of different possible RISC-V combinations, would a micro kernel design not have an intrinsic advantage over monolithic kernels, and be able to exploit the more modular nature of their design?

edit clarification

I came across the post about Milk-V Titan, and there was a comment asking about the lack of the V extension would hinder running Ubuntu 25.10 which was targetting a particular RISC-V configuration, and it made me wonder if there were an opportunity for micro kernels to exploit.

Now, up-front: it’s been literally decades since I had an OS design class, and my knowledge of OS design is superficial; and while I’ve always been interested in RISC architectures, the depth of my knowledge of that also dates back to the 90’s. In particular (my knowledge of) RISC-V’s extension design approach is really, really shallow. It’s all at a lower level than I’ve concerned myself with for years and years. So I’m hoping for an ELI-16 conversation.

What I was thinking was that a challenge of RISC-V’s design is that operating systems can’t rely on extensions being available, which (in my mind) means either a lot of really specific kernel builds – like, potentially an exponential number – or a similar number of code paths in the kernel code, making for more complicated and consequently more buggy kernels (per the McConnell rule). It made me wonder if this is not, then, an opportunity for micro kernels to shine, by exploiting an ability to load extension-specific modules based on a given CPU capability set.

As I see it, the practicality of this depends on whether the extensions would be isolatable to kernel modules, or whether (like the FP extension) it’d just be so intrinsic that even the core kernel would need to vary. Even so, wouldn’t having a permutation of core kernel builds be smaller, more manageable, and less bug-prone than permutations of monolithic kernels?

Given the number of different possible RISC-V combinations, would a micro kernel design not have an intrinsic advantage over monolithic kernels, and be able to exploit the more modular nature of their design?

edit clarification

0

What's your strategy for Fulgora

preview

Picture unrelated to question; built for style, not functionality. I’ve learned that the most efficient (and boring) platform design is a spear.

I think of all the planets, I like Fulgora the least. I have tried two or three approaches to dealing with the sushi belts of parts; my biggest issue is the monomanical logic of inserters, which – given a storage chest of a dozen different items, seem to get stuck on one item for extended periods. For instance, unloading a train of random stuff, I’ll get 15 inserters all pulling out gears for several minutes, ignoring everything else in the chests. It’s frustrating beyond belief, so I think I’m not looking at this the right way.

The biggest issue is buffering. I’ll get nothing but gears for a few hours, but then they’ll dry up and I’ll get nothing but low density structure for a few hours and almost no gears. If I were paranoid, I’d think the devs did this on purpose to maximize jamming.

What I’m going to try next is building a train system with per-item trains, or maybe cars, and see if that helps. Transport raw to an island, recycle it, and load up trains with product and send them to a big factory island. I suspect it’s simply going to get jammed up at the recycling center and I won’t be any further ahead, but maybe I can mitigate that by just having one, really long train that stops every 12 cars and every series of 12 cars has each car containing a single item.

I don’t know. Fulgora just feels stupid, or makes me feel stupid. How do you folks handle Fulgora?

Update edit

I completely redesigned everything on Fulgora, and it doesn’t get jammed, and I’m only having to “throw away” steel (the plate, not the girders; I’m not sure why the game calls “girders” “plate”, but whatever).

My solution, as it is, is based on the suggestion of having a recycling plant rather than recycling at the miners. I’m using trains: I mine ore into trains and ship it to a recycling island, where the output is filtered into train cars - one type per car. I run the trains full of product to wherever I have factories set up, and pull specific items needed there. I avoid sushi belts and sushi boxcars; my production islands are consequently much simpler, and as yet I’m not having to waste product.

I will point out that I initially used recyclers to reduce the amounts out items, but found I had to keep changing what I was recycling because what was overflowing kept changing, and by recycling I was regularly running out of items; this, again, is related to whatever the RNG is doing to me. Honestly, I have a mind to actually keep a counter, because it appears as if I’ll get a run of copper coil, where it’s the dominant item mined for dozens of minutes, then it’ll be blue chips, and so on, cycling through each of the items. It swamps me with one thing that I can’t even manage with belt weaving, then suddenly that’ll dry up and I’ll get only a trickle of that thing for the next couple of hours. I’ve frequently had launches completely stop because I had to recycle one component to built rocket parts, and then it dries up and I have to built an assembler to produce the product until the RNG deems me worthy to have it start being a mined item again. If the cycle happened over a period of less than hours, I’d record it. Although, the fact that the inserters (almost) all get stuck on the same item pulling from sushi boxes, I could easily show. If I, of anyone else, cared enough.

In any case, with trains and enough storage boxes I have it set up to weather the droughts and floods of specific items, and am assuming either it’s coded this way on purpose, or only doing it on my machine because I don’t hear anyone else complaining about it.

Picture unrelated to question; built for style, not functionality. I’ve learned that the most efficient (and boring) platform design is a spear.

I think of all the planets, I like Fulgora the least. I have tried two or three approaches to dealing with the sushi belts of parts; my biggest issue is the monomanical logic of inserters, which – given a storage chest of a dozen different items, seem to get stuck on one item for extended periods. For instance, unloading a train of random stuff, I’ll get 15 inserters all pulling out gears for several minutes, ignoring everything else in the chests. It’s frustrating beyond belief, so I think I’m not looking at this the right way.

The biggest issue is buffering. I’ll get nothing but gears for a few hours, but then they’ll dry up and I’ll get nothing but low density structure for a few hours and almost no gears. If I were paranoid, I’d think the devs did this on purpose to maximize jamming.

What I’m going to try next is building a train system with per-item trains, or maybe cars, and see if that helps. Transport raw to an island, recycle it, and load up trains with product and send them to a big factory island. I suspect it’s simply going to get jammed up at the recycling center and I won’t be any further ahead, but maybe I can mitigate that by just having one, really long train that stops every 12 cars and every series of 12 cars has each car containing a single item.

I don’t know. Fulgora just feels stupid, or makes me feel stupid. How do you folks handle Fulgora?

Update edit

I completely redesigned everything on Fulgora, and it doesn’t get jammed, and I’m only having to “throw away” steel (the plate, not the girders; I’m not sure why the game calls “girders” “plate”, but whatever).

My solution, as it is, is based on the suggestion of having a recycling plant rather than recycling at the miners. I’m using trains: I mine ore into trains and ship it to a recycling island, where the output is filtered into train cars - one type per car. I run the trains full of product to wherever I have factories set up, and pull specific items needed there. I avoid sushi belts and sushi boxcars; my production islands are consequently much simpler, and as yet I’m not having to waste product.

I will point out that I initially used recyclers to reduce the amounts out items, but found I had to keep changing what I was recycling because what was overflowing kept changing, and by recycling I was regularly running out of items; this, again, is related to whatever the RNG is doing to me. Honestly, I have a mind to actually keep a counter, because it appears as if I’ll get a run of copper coil, where it’s the dominant item mined for dozens of minutes, then it’ll be blue chips, and so on, cycling through each of the items. It swamps me with one thing that I can’t even manage with belt weaving, then suddenly that’ll dry up and I’ll get only a trickle of that thing for the next couple of hours. I’ve frequently had launches completely stop because I had to recycle one component to built rocket parts, and then it dries up and I have to built an assembler to produce the product until the RNG deems me worthy to have it start being a mined item again. If the cycle happened over a period of less than hours, I’d record it. Although, the fact that the inserters (almost) all get stuck on the same item pulling from sushi boxes, I could easily show. If I, of anyone else, cared enough.

In any case, with trains and enough storage boxes I have it set up to weather the droughts and floods of specific items, and am assuming either it’s coded this way on purpose, or only doing it on my machine because I don’t hear anyone else complaining about it.