I write Rust Black person and I just cleared out my target directories the other day to free up 100GB+ of space. If that shit was C/C++ it would have been like a few dozen MB of obj files.
Rust is one of the slowest language to compile. Compiling rust programs uses an insane amount of disk space for some reason. Your build directory can easily become 30GB in size for fairly small programs. Seriously go build an SDL example program for Rust and instantly use 10GB of disk space.
https://github.com/mTvare6/hello-world.rs
Why are you compiling rust itself? Do weird.
Rust is one of the slowest language to compile. Compiling rust programs uses an insane amount of disk space for some reason. Your build directory can easily become 30GB in size for fairly small programs. Seriously go build an SDL example program for Rust and instantly use 10GB of disk space.
https://github.com/mTvare6/hello-world.rs
What is that link supposed to prove?
2 years ago
Anonymous
Shut up bruce everyone in this thread is moronic and needs to die immediately.
Rust is one of the slowest language to compile. Compiling rust programs uses an insane amount of disk space for some reason. Your build directory can easily become 30GB in size for fairly small programs. Seriously go build an SDL example program for Rust and instantly use 10GB of disk space.
> https://github.com/mTvare6/hello-world.rs
That's a parody project, m8, you fell for a joke.
2 years ago
Anonymous
You clearly have never written any real Rust code. And yeah, no one would ever genuinely use dependencies like glfw, freetype and cairo in a program. Truly madness.
2 years ago
Anonymous
not in a hello world you ape
2 years ago
Anonymous
It's example code. A basically empty program demonstrating a very basic opengl/audio stack takes 30GB. This would take like a 100MB in C/C++ at most assuming you're compiling every last thing.
2 years ago
Anonymous
>a very basic opengl/audio stack
it's pulling 89 packages that include things like amethyst and TensorFlow. are you genuinely moronic?
2 years ago
Anonymous
>oh my god instead of 100 glue lines in the build system or going "works on my machine" you can just import working dependencies with one (1) line in a config file!
this is genuinely, unironically, a good thing
[...] >a "very basic opengl/audio stack" with a web framework, a game engine and tensorflow
>a "very basic opengl/audio stack" with a web framework, a game engine and tensorflow
yeah, not everyone is writing fizz buzz all day long.
2 years ago
Anonymous
>It's example code
Have you checked out it Cargo.toml? >x11 >Winapi >tensorflow
It's just got dependencies loaded out the ass. I believe it's what the kids call a "joke". Like Enterprise FizzBuzz.
https://github.com/EnterpriseQualityCoding/FizzBuzzEnterpriseEdition
2 years ago
Anonymous
>no one would ever genuinely use dependencies like glfw, freetype and cairo in a program
I know IQfy is mainly comprised of first year CompSci students nowadays but please try not to let it show.
2 years ago
Anonymous
>nobody would ever genuinely make a video game
Well not in Rust anyways, lmao
2 years ago
Anonymous
>You clearly have never written any real Rust code.
Neither have you. Nobody has.
2 years ago
Anonymous
>And yeah, no one would ever genuinely use dependencies like glfw, freetype
Curious, what should I use instead of freetype for videogames?
All games I've worked on for many years have used freetype (including AAA multiplatform with all major localizations). What's wrong with it? Never had any problems with it.
>oh my god instead of 100 glue lines in the build system or going "works on my machine" you can just import working dependencies with one (1) line in a config file!
this is genuinely, unironically, a good thing
It's example code. A basically empty program demonstrating a very basic opengl/audio stack takes 30GB. This would take like a 100MB in C/C++ at most assuming you're compiling every last thing.
>a "very basic opengl/audio stack" with a web framework, a game engine and tensorflow
it pulls in a shitload of code and leaves binary shit all over the place so that it builds faster next time you do another "cargo build". The release binaries are not that large.
According to an article I read a while back, rust has implemented the most complicated incremental build system ever devised.
2 years ago
Anonymous
>rust has implemented the most complicated incremental build system ever devised
I would not be surprised. Caching intermediate build artifacts is a natural result of using a properly compartmentalized build system. Why did rustrannies need to overcomplicate this shit instead of doing it properly?
A driver which uses 1 unsafe line for 99 of safe lines is 99% safer than the same driver written in C, because unsafe Rust is as safe as regular C. Limiting this "unsafeness" to a couple of unsafe blocks is a huge win.
you can use all those resources you never really needed. i wish that programmers would become good again... get rid of the bloat, be more like templeos
2 years ago
Anonymous
you mean more schizo?
2 years ago
Anonymous
>you mean more schizo?
If thats what it takes.Temple os is surprisingly simple compared to modern usles complexity.And holly c is simpler than c .
Because it's better to limit "unsafeness" to a couple of easily audited block than to have your entire project be inherently unsafe, as is the case with C/C++. Besides, most of the time you don't need unsafe at all.
Sorry I'm not a soi windev who needs a heckin solution to please do the needful.
> he doesn't know about the dependency hell
The kids these days...
just don't hire pajeet coders then. c is safe if you can write it well
> c is safe if you can write it well
Too bad no one can write it well, that's why C in CVE actually stands for C Programming Language.
2 years ago
Anonymous
>The kids these days...
they're jobless neets or basically spoiled little shits that don't even know what they have. lol
>unsafe Rust is as safe as regular C
It’s safer. Unsafe doesn’t disable all checks - it just allows a few operations that are unsafe but have legitimate uses.
I don't particularly like Rust, but you are moronic. Unsafe doesn't mean "don't use". It's just a way of segregating/marking the code that isn't guaranteed to be memory safe. You are meant to use "unsafe" code. You must use it, if you want to allocate memory. There is nothing wrong with it. It's just meant to help contain memory code.
so why did they release rust before all the safety problems were solved? if it was good there would be absolutely no need for any of the unsafe functionality
I don't know. But you completely misunderstand what "unsafe" is about. Expecting to write driver code without "unsafe" is moronic. You just have no idea what you are talking about.
2 years ago
Anonymous
just make it safe. this feels like game dev where they dump out the shit as soon as the program is able to start and finish it later if they feel like it
2 years ago
Anonymous
Impossible, code that is safe, but cannot be proven to be, will always exist
https://en.wikipedia.org/wiki/G%C3%B6del%27s_incompleteness_theorems
so why did they release rust before all the safety problems were solved? if it was good there would be absolutely no need for any of the unsafe functionality
>code that is safe, but cannot be proven to be, will always exist
You should try not to write that kind of code. If computers fundamentally can't prove that it's safe then neither can you.
There is useful code that can be formally proven to be correct.
larpers having no clue what unsafe even means in rust, and why it's necessary.
RP thread
>lul just dont use unsafe bro >just don't work with raw ptrs bro >just don't use FFI bro >just don't request data from outside rust bro >just don't use unions bro >just larp as a sys programmer by using rust for webdev bro
2 years ago
Anonymous
I know what unsafe code means and I've written my share, I just said it's just not because of anything to do with Gödel
Some things are inherently unsafe. If I call a C function then how is the compiler supposed to know what happens? Even assuming the C code isn't buggy, who is to say whether it tries to free() the pointer I give it or write to it or just read from it or ignore it? You need an unsafe block around it, it's the only sane way.
You can't actually verify everything. "Unsafe" marks things the compiler can't verify but you can. It would be great if that were unnecessary, but there's no solution in sight. This doesn't make Rust a failure, all other supposedly "safe" languages are the same. Python is basically memory-safe but you can still cause segfaults by calling to unsafe C code and much of the standard library is written in unsafe C code because it has to be.
Impossible, code that is safe, but cannot be proven to be, will always exist
https://en.wikipedia.org/wiki/G%C3%B6del%27s_incompleteness_theorems
2 years ago
Anonymous
>code that is safe, but cannot be proven to be, will always exist
You should try not to write that kind of code. If computers fundamentally can't prove that it's safe then neither can you.
There is useful code that can be formally proven to be correct.
Determining the safety of code that can use pointers arbitrarily is extremely intensive and generally undecidable. The seL4 kernel is the best example of formally verified, low level code and it required millions of lines of proofs to verify less than 10,000 lines of C.
The fact that safe Rust works without writing any proofs is pretty amazing, although there is an effort to develop a mechanized framework for verifying unsafe code with proofs.
Can't wait for the inevitable news of some rust troony going on a political crusade by changing something in the rust compiler or standard library and breaking the whole internet.
The Rust programming language was designed by the NSA. Its sole purpose is to backdoor *everything*. Rust and the Rust compiler are designed to use absurd amounts of time and resources, and to fail compilation randomly - compiling Veloren for example, a 3d voxel game written in rust, takes over 16GB of ram and 12 hours before compilation fails in in a system with 2 cores; this will de-incentivize compilation from source (specially on systems that are completely free, i.e: Libre/CoreBoot & no IME or AMD PSP), leading people to opt for simply downloading pre-compiled binaries - who will compile these binaries? of course ((they)) - that will be compromised. Even if you were to compile all your rust software from source, compiling the compiler itself is a paramount task, so anyone without modern backdoored hardware will be forced to download pre-compiled, which will of course, be compromised and compromise anything it compiles (for proof, just look at a simple helloworld in rust, it is 2.6MB and makes a web request every time its run). Rust is not about troonyes, Rust is the NSA's response to the community rejecting Systemd, they wont conform with backdooring your init anymore, now they will backdoor every piece of software in your system, starting with the kernel.
Backdoors are conspicuous and therefore risky. They can be detected just by analyzing the object code or the runtime behavior. You do not want to do this at scale.
If the NSA is backdooring things they're definitely backdooring the hardware, not the software. Backdooring software is much more risk for much less reward. >the community rejecting Systemd
Most of "the community" is on systemd, like it or not. >it is 2.6MB
Not if you strip the debug symbols. >and makes a web request every time its run
No.
The real NSA backdoor is C, single-handedly responsible for 90% of vulnerabilities out there. Without C they'd have nothing to do.
Also modern browsers require clusters to compile, so you trust google or mozilla to provide you with a binary without screeching and making up statistics.
The real NSA backdoor is C, single-handedly responsible for 90% of vulnerabilities out there. Without C they'd have nothing to do.
Also modern browsers require clusters to compile, so you trust google or mozilla to provide you with a binary without screeching and making up statistics.
I can't believe that the people that got filtered by pointers and memory allocations made a whole new programming language and is pushing for it to be the industry standard.
> the people that got filtered by pointers and memory allocations
That would be the entire industry, according to the CVE database. The only people who don't make memory-related errors in C are the fizzbuzzers from IQfy.
If you're too poor to compile shit then just don't compile shit. Spend that time getting a job. Spend those cpu cycles incrementally compiling a product to sell
2 years ago
Anonymous
Fine by me. I'm more bothered by buggy software than by software taking longer to compile.
Rust's compile times bother me when developing large programs but when installing them it hasn't been all that big a deal.
0.001% of the code taking 10x time to compile isn't a big deal. It will take another 20 years for Rust to make up a significant share of the kernel code. By that time we'll have a quantum-accelerated hardware Rust compilers.
>j-just get better hardware bigot! >no! it doesn't matter that the compiler is shit!
As expected, the rust homosexuals have to cope about their compiler being dysfunctional.
2 years ago
Anonymous
frick no. i still think its insane that theres one website that does not run well on my pc while everything else i could have wanted to run works perfectly
2 years ago
Anonymous
> being as slow as g++ is now dysfunctional
Gee, no wonder C++ has failed and no one uses Chrome and Firefox, written in it.
2 years ago
Anonymous
great examples of truly horrible software. it takes fricking 10 seconds to render the blocks that they use as interface and some text. the older systems were able to do this much faster
2 years ago
Anonymous
Have you used "the older systems" tho? Or is the nostalgia clouds your Cnile mind?
2 years ago
Anonymous
when i used firefox 1 and 2 on windows xp systems pages used to load very quickly. they also looked much nicer than the generic material/responsive garbage that everything is today
2 years ago
Anonymous
Now imagine the early 2000s web in modern browsers, with state-of-the-art JIT and GPU acceleration. Browsers are fine, they're getting faster and faster, it's the webshits who's to blame.
2 years ago
Anonymous
Those advancements hardly help when the underlying concepts are flawed. Javascript is designed so badly as a language I can't really blame webshits for doing what they do. HTML lets you get away with stupid bullshit that SGML allows rather than being strict like XML.
2 years ago
Anonymous
>Javascript is designed so badly as a language
What do you mean? Javascript is pretty much just a variant of scheme with built in hashmaps, and the syntax copied from java.
2 years ago
Anonymous
How is it like Scheme? Don't quote Brendan Eich, explain what parts of the language are similar to Scheme. It doesn't have tail call elimination, or lists (no, arrays are different), or symbols.
It's dynamically typed, I guess.
2 years ago
Anonymous
are you sure that it would be an improvement? can i make those things run smoothly on a pentium 4?
i believe a big problem these days is that no one cares about efficiency so you use a million times more resources to do the same task as you were doing 20 years ago. sure it looks more shiny now but in the end you are still doing the same job...
2 years ago
Anonymous
Yeah, I'm sure that JIT and GPU acceleration improve performance dramatically. Pre-JIT JS was barely practical.
2 years ago
Anonymous
we didn't have this kind of js back then. flash and java were the bloat of those days
2 years ago
Anonymous
>the time it takes for the two most well known browsers to compile is comparable to compiling a simple project
Also fair to mention that Firefox has parts of it written in Rust, arguably helping to make compile times longer, perfectly normal and fine of course.
Fine by me. I'm more bothered by buggy software than by software taking longer to compile.
Rust's compile times bother me when developing large programs but when installing them it hasn't been all that big a deal.
0.001% of the code taking 10x time to compile isn't a big deal. It will take another 20 years for Rust to make up a significant share of the kernel code. By that time we'll have a quantum-accelerated hardware Rust compilers.
2 years ago
Anonymous
>let's just assume there'll be better hardware in the future
Great excuse to make software increasingly bloated and heavy, like it is today with JS ecosystem.
So, Rust is a C with everything we learned from the past 40 years, including the mistakes?
i would like fixes to bugs that affect functionality over some tinfoil memory security patch. for example whatever makes the gpu driver generate one of those cut here dmesg crash logs has been there for who knows how long and no one seems to care
I've never written any kernel code but in the userland memory unsafety is a big source of crashes and Rust can help with correctness in general
2 years ago
Anonymous
Valgrind exists. If the only selling point to Rust is the borrow checker, there’s no reason it couldn’t have just been a linter for c code.
2 years ago
Anonymous
> paramedics exist, therefore we don't need safety regulations
2 years ago
Anonymous
>> paramedics exist, therefore we don't need safety regulations
What next, a food analogy?
2 years ago
Anonymous
Nobody has managed to make such a linter. People have tried, but it works much better if you design your whole language around it.
One language Rust took inspiration from is Cyclone, a research language that was a dialect of C, but that one never became production ready and needed garbage collection anyhow.
Generics are really important to Rust's safety. You can write one generic container or wrapper type with unchecked code and restrict which types can safely be used with it and then all code that uses the generic type can be checked automatically. C doesn't have generics, so it can't do this.
You'd also have to change the way you write code and throw out or wrap a lot of existing code, which is a burden comparable to switching to another language.
2 years ago
Anonymous
Rust has always been bloat, just write safe code, it's really easy like that.
>If it's popular it must be good
Cniles for you, everyone.
It's not good because it's popular, it's popular because it's good.
2 years ago
Anonymous
Actually it's neither of those
2 years ago
Anonymous
>It's not good because it's popular, it's popular because it's good.
Nope. It's popular because it was developed in the right place at the right time. If some one created C today, no one would use it.
What is it with this Rust support trend on here lately? Rust generals, Rust shillers, Rust trolls etc. I mean sure some of it's obviously trolling but it makes me actually afraid that some actually think Rust is going to replace C or something.
because rust is a corporate sponsored language with a marketing team that goes around the internet offering to donate to people's projects if they adopt the usage of rust. they think it's perfectly normal to come to websites like this, that have actual programmers, to lie about the language - like we're illiterate or we've never programmed before.
>that goes around the internet offering to donate to people's projects if they adopt the usage of rust
holy fricking shit the whole rust team needs to be purged
You're right, rust's macros are hygenic unlike C's dumb text copy-paste. The only reason people hold the >muh macro bad
opinion is because C's macros are too shit and too much of a footgun even for cniles.
but in all seriousness, rust supports all major platforms and will be mostly (only) used in drivers, probably for hardware that the snowflake platforms will not use anyway
>make open source kernel >constantly shill "muh freedoms" everywhere >tell everyone they can do what ever they want with it >they do what ever they want with it >NOOOOOOOOOOOOOOOO STOP DOING THAT
I can't wait until compiling the kernel takes a week and requires a TB of disk space.
This is a serious issue. Compiling the kernel right now is super easy. It's going to be a complete mess soon.
I don’t understand. Why would that be the case?
because the imaginary rust in their heads is really hard to build
I write Rust Black person and I just cleared out my target directories the other day to free up 100GB+ of space. If that shit was C/C++ it would have been like a few dozen MB of obj files.
Why are you compiling rust itself? Do weird.
What is that link supposed to prove?
Shut up bruce everyone in this thread is moronic and needs to die immediately.
Rust is one of the slowest language to compile. Compiling rust programs uses an insane amount of disk space for some reason. Your build directory can easily become 30GB in size for fairly small programs. Seriously go build an SDL example program for Rust and instantly use 10GB of disk space.
https://github.com/mTvare6/hello-world.rs
> https://github.com/mTvare6/hello-world.rs
That's a parody project, m8, you fell for a joke.
You clearly have never written any real Rust code. And yeah, no one would ever genuinely use dependencies like glfw, freetype and cairo in a program. Truly madness.
not in a hello world you ape
It's example code. A basically empty program demonstrating a very basic opengl/audio stack takes 30GB. This would take like a 100MB in C/C++ at most assuming you're compiling every last thing.
>a very basic opengl/audio stack
it's pulling 89 packages that include things like amethyst and TensorFlow. are you genuinely moronic?
>a "very basic opengl/audio stack" with a web framework, a game engine and tensorflow
yeah, not everyone is writing fizz buzz all day long.
>It's example code
Have you checked out it Cargo.toml?
>x11
>Winapi
>tensorflow
It's just got dependencies loaded out the ass. I believe it's what the kids call a "joke". Like Enterprise FizzBuzz.
https://github.com/EnterpriseQualityCoding/FizzBuzzEnterpriseEdition
>no one would ever genuinely use dependencies like glfw, freetype and cairo in a program
I know IQfy is mainly comprised of first year CompSci students nowadays but please try not to let it show.
>nobody would ever genuinely make a video game
Well not in Rust anyways, lmao
>You clearly have never written any real Rust code.
Neither have you. Nobody has.
>And yeah, no one would ever genuinely use dependencies like glfw, freetype
Curious, what should I use instead of freetype for videogames?
All games I've worked on for many years have used freetype (including AAA multiplatform with all major localizations). What's wrong with it? Never had any problems with it.
>oh my god instead of 100 glue lines in the build system or going "works on my machine" you can just import working dependencies with one (1) line in a config file!
this is genuinely, unironically, a good thing
>a "very basic opengl/audio stack" with a web framework, a game engine and tensorflow
Hey, kid, ever heard of "DLL hell"? That's how you deal with it.
Sorry I'm not a soi windev who needs a heckin solution to please do the needful.
No. This problem was solved in 2003, in a language-agnostic way. THAT is how you deal with DLL hell.
Based nix.
>https://github.com/mTvare6/hello-world.rs
>a spic
>a cnile
>a redditor
no wonder he seethes so hard, he lost at life in every regard.
when you do a:
cargo build
it pulls in a shitload of code and leaves binary shit all over the place so that it builds faster next time you do another "cargo build". The release binaries are not that large.
According to an article I read a while back, rust has implemented the most complicated incremental build system ever devised.
>rust has implemented the most complicated incremental build system ever devised
I would not be surprised. Caching intermediate build artifacts is a natural result of using a properly compartmentalized build system. Why did rustrannies need to overcomplicate this shit instead of doing it properly?
cant wait for the day its removed
Don't care, I don't use that piece of shit
rust is a meme. i want to see a driver that does not resort to using any of the unsafe functionality
A driver which uses 1 unsafe line for 99 of safe lines is 99% safer than the same driver written in C, because unsafe Rust is as safe as regular C. Limiting this "unsafeness" to a couple of unsafe blocks is a huge win.
Whats the fricking point of Rust if you can diseable all this "safeness"?
>ignore post entirely
>reply anyway
?
you can use all those resources you never really needed. i wish that programmers would become good again... get rid of the bloat, be more like templeos
you mean more schizo?
>you mean more schizo?
If thats what it takes.Temple os is surprisingly simple compared to modern usles complexity.And holly c is simpler than c .
Because it's better to limit "unsafeness" to a couple of easily audited block than to have your entire project be inherently unsafe, as is the case with C/C++. Besides, most of the time you don't need unsafe at all.
> he doesn't know about the dependency hell
The kids these days...
> c is safe if you can write it well
Too bad no one can write it well, that's why C in CVE actually stands for C Programming Language.
>The kids these days...
they're jobless neets or basically spoiled little shits that don't even know what they have. lol
it's very useful, for those of us that can read
just don't hire pajeet coders then. c is safe if you can write it well
>unsafe Rust is as safe as regular C
It’s safer. Unsafe doesn’t disable all checks - it just allows a few operations that are unsafe but have legitimate uses.
I don't particularly like Rust, but you are moronic. Unsafe doesn't mean "don't use". It's just a way of segregating/marking the code that isn't guaranteed to be memory safe. You are meant to use "unsafe" code. You must use it, if you want to allocate memory. There is nothing wrong with it. It's just meant to help contain memory code.
so why did they release rust before all the safety problems were solved? if it was good there would be absolutely no need for any of the unsafe functionality
I don't know. But you completely misunderstand what "unsafe" is about. Expecting to write driver code without "unsafe" is moronic. You just have no idea what you are talking about.
just make it safe. this feels like game dev where they dump out the shit as soon as the program is able to start and finish it later if they feel like it
larpers having no clue what unsafe even means in rust, and why it's necessary.
RP thread
>lul just dont use unsafe bro
>just don't work with raw ptrs bro
>just don't use FFI bro
>just don't request data from outside rust bro
>just don't use unions bro
>just larp as a sys programmer by using rust for webdev bro
I know what unsafe code means and I've written my share, I just said it's just not because of anything to do with Gödel
Some things are inherently unsafe. If I call a C function then how is the compiler supposed to know what happens? Even assuming the C code isn't buggy, who is to say whether it tries to free() the pointer I give it or write to it or just read from it or ignore it? You need an unsafe block around it, it's the only sane way.
You can't actually verify everything. "Unsafe" marks things the compiler can't verify but you can. It would be great if that were unnecessary, but there's no solution in sight. This doesn't make Rust a failure, all other supposedly "safe" languages are the same. Python is basically memory-safe but you can still cause segfaults by calling to unsafe C code and much of the standard library is written in unsafe C code because it has to be.
Impossible, code that is safe, but cannot be proven to be, will always exist
https://en.wikipedia.org/wiki/G%C3%B6del%27s_incompleteness_theorems
>code that is safe, but cannot be proven to be, will always exist
You should try not to write that kind of code. If computers fundamentally can't prove that it's safe then neither can you.
There is useful code that can be formally proven to be correct.
just use a non-Turing complete language
Determining the safety of code that can use pointers arbitrarily is extremely intensive and generally undecidable. The seL4 kernel is the best example of formally verified, low level code and it required millions of lines of proofs to verify less than 10,000 lines of C.
The fact that safe Rust works without writing any proofs is pretty amazing, although there is an effort to develop a mechanized framework for verifying unsafe code with proofs.
Can't wait for the inevitable news of some rust troony going on a political crusade by changing something in the rust compiler or standard library and breaking the whole internet.
I mean, Rust already has a huge change in their board of directors some months ago. This is a serious risk, these people aren't stable.
Imagine a board of directors full of trannies. The churn rate must be crazy
meh - sys programmers are far less worrisome than webdevs. Those fricks can't seem to separate their mouths from their ass holes
you're not a webdev, are you anon?
Chuds on suicide watch.
The Rust programming language was designed by the NSA. Its sole purpose is to backdoor *everything*. Rust and the Rust compiler are designed to use absurd amounts of time and resources, and to fail compilation randomly - compiling Veloren for example, a 3d voxel game written in rust, takes over 16GB of ram and 12 hours before compilation fails in in a system with 2 cores; this will de-incentivize compilation from source (specially on systems that are completely free, i.e: Libre/CoreBoot & no IME or AMD PSP), leading people to opt for simply downloading pre-compiled binaries - who will compile these binaries? of course ((they)) - that will be compromised. Even if you were to compile all your rust software from source, compiling the compiler itself is a paramount task, so anyone without modern backdoored hardware will be forced to download pre-compiled, which will of course, be compromised and compromise anything it compiles (for proof, just look at a simple helloworld in rust, it is 2.6MB and makes a web request every time its run). Rust is not about troonyes, Rust is the NSA's response to the community rejecting Systemd, they wont conform with backdooring your init anymore, now they will backdoor every piece of software in your system, starting with the kernel.
Backdoors are conspicuous and therefore risky. They can be detected just by analyzing the object code or the runtime behavior. You do not want to do this at scale.
If the NSA is backdooring things they're definitely backdooring the hardware, not the software. Backdooring software is much more risk for much less reward.
>the community rejecting Systemd
Most of "the community" is on systemd, like it or not.
>it is 2.6MB
Not if you strip the debug symbols.
>and makes a web request every time its run
No.
you just woke up, did you not?
It's 2PM where I live and time.is tells me it's 3PM in Tel Aviv. What are you talking about?
The real NSA backdoor is C, single-handedly responsible for 90% of vulnerabilities out there. Without C they'd have nothing to do.
Also modern browsers require clusters to compile, so you trust google or mozilla to provide you with a binary without screeching and making up statistics.
kys troon
giving furgays full rights to technology was a grave mistake you had it coming IQfyoyim install linux from scratch
I can't believe that the people that got filtered by pointers and memory allocations made a whole new programming language and is pushing for it to be the industry standard.
> the people that got filtered by pointers and memory allocations
That would be the entire industry, according to the CVE database. The only people who don't make memory-related errors in C are the fizzbuzzers from IQfy.
Linus, blink twice if you're in danger!
i don't feel so good c++ frens
Will this affect me at all?
Yes, but only if you try hard to get angry about it
your choice. some paranoid people probably still run old kernels because that wiener thing happened
Three years from now you'll install a driver that's a little less buggy than it would've been otherwise.
With the side effect of taking 10x more time to compile.
>inb4 b-but that's just for the first compile chud! t-trust the compiler!
If you're too poor to compile shit then just don't compile shit. Spend that time getting a job. Spend those cpu cycles incrementally compiling a product to sell
>j-just get better hardware bigot!
>no! it doesn't matter that the compiler is shit!
As expected, the rust homosexuals have to cope about their compiler being dysfunctional.
frick no. i still think its insane that theres one website that does not run well on my pc while everything else i could have wanted to run works perfectly
> being as slow as g++ is now dysfunctional
Gee, no wonder C++ has failed and no one uses Chrome and Firefox, written in it.
great examples of truly horrible software. it takes fricking 10 seconds to render the blocks that they use as interface and some text. the older systems were able to do this much faster
Have you used "the older systems" tho? Or is the nostalgia clouds your Cnile mind?
when i used firefox 1 and 2 on windows xp systems pages used to load very quickly. they also looked much nicer than the generic material/responsive garbage that everything is today
Now imagine the early 2000s web in modern browsers, with state-of-the-art JIT and GPU acceleration. Browsers are fine, they're getting faster and faster, it's the webshits who's to blame.
Those advancements hardly help when the underlying concepts are flawed. Javascript is designed so badly as a language I can't really blame webshits for doing what they do. HTML lets you get away with stupid bullshit that SGML allows rather than being strict like XML.
>Javascript is designed so badly as a language
What do you mean? Javascript is pretty much just a variant of scheme with built in hashmaps, and the syntax copied from java.
How is it like Scheme? Don't quote Brendan Eich, explain what parts of the language are similar to Scheme. It doesn't have tail call elimination, or lists (no, arrays are different), or symbols.
It's dynamically typed, I guess.
are you sure that it would be an improvement? can i make those things run smoothly on a pentium 4?
i believe a big problem these days is that no one cares about efficiency so you use a million times more resources to do the same task as you were doing 20 years ago. sure it looks more shiny now but in the end you are still doing the same job...
Yeah, I'm sure that JIT and GPU acceleration improve performance dramatically. Pre-JIT JS was barely practical.
we didn't have this kind of js back then. flash and java were the bloat of those days
>the time it takes for the two most well known browsers to compile is comparable to compiling a simple project
Also fair to mention that Firefox has parts of it written in Rust, arguably helping to make compile times longer, perfectly normal and fine of course.
Fine by me. I'm more bothered by buggy software than by software taking longer to compile.
Rust's compile times bother me when developing large programs but when installing them it hasn't been all that big a deal.
0.001% of the code taking 10x time to compile isn't a big deal. It will take another 20 years for Rust to make up a significant share of the kernel code. By that time we'll have a quantum-accelerated hardware Rust compilers.
>let's just assume there'll be better hardware in the future
Great excuse to make software increasingly bloated and heavy, like it is today with JS ecosystem.
So, Rust is a C with everything we learned from the past 40 years, including the mistakes?
i would like fixes to bugs that affect functionality over some tinfoil memory security patch. for example whatever makes the gpu driver generate one of those cut here dmesg crash logs has been there for who knows how long and no one seems to care
I've never written any kernel code but in the userland memory unsafety is a big source of crashes and Rust can help with correctness in general
Valgrind exists. If the only selling point to Rust is the borrow checker, there’s no reason it couldn’t have just been a linter for c code.
> paramedics exist, therefore we don't need safety regulations
>> paramedics exist, therefore we don't need safety regulations
What next, a food analogy?
Nobody has managed to make such a linter. People have tried, but it works much better if you design your whole language around it.
One language Rust took inspiration from is Cyclone, a research language that was a dialect of C, but that one never became production ready and needed garbage collection anyhow.
Generics are really important to Rust's safety. You can write one generic container or wrapper type with unchecked code and restrict which types can safely be used with it and then all code that uses the generic type can be checked automatically. C doesn't have generics, so it can't do this.
You'd also have to change the way you write code and throw out or wrap a lot of existing code, which is a burden comparable to switching to another language.
Rust has always been bloat, just write safe code, it's really easy like that.
when will they rewrite systemd in rust? i would like to see the threads about that
hare in the kernel when
https://github.com/Rust-GCC/gccrs/wiki/Frequently-Asked-Questions
We need to support this or else we will need 2 compilers to compile the kernel.
Rust is non-standardized cancer
C is standardized cancer.
Remind me what language do most microcontrollers use and what language is the most popular mobile OS kernel is written in.
Assembler.
Wrong.
It's not good because it's popular, it's popular because it's good.
Actually it's neither of those
>It's not good because it's popular, it's popular because it's good.
Nope. It's popular because it was developed in the right place at the right time. If some one created C today, no one would use it.
>If it's popular it must be good
Cniles for you, everyone.
thank god, time to leave codelets behind, they've been getting a free ride for too long
What is it with this Rust support trend on here lately? Rust generals, Rust shillers, Rust trolls etc. I mean sure some of it's obviously trolling but it makes me actually afraid that some actually think Rust is going to replace C or something.
because rust is a corporate sponsored language with a marketing team that goes around the internet offering to donate to people's projects if they adopt the usage of rust. they think it's perfectly normal to come to websites like this, that have actual programmers, to lie about the language - like we're illiterate or we've never programmed before.
>that goes around the internet offering to donate to people's projects if they adopt the usage of rust
please tell me more
>please tell me more
are there any others?
>0 blacks
Problematic.
>refuse converting project to rust
>they say you're misogynistic because you don't support rust
>suddenly blacklisted from the industry
t-thanks?
blacklisted from the industry
so no change?
>blacklisted
Don't come to work tomorrow, Josh.
>that goes around the internet offering to donate to people's projects if they adopt the usage of rust
holy fricking shit the whole rust team needs to be purged
Yes because C++ is obviously never used or sponsored by any corporations, mirite lads?
>Ponzi scam trash
Unironically the worst part about rust
>goes around the internet offering to donate to people's projects if they adopt the usage of rust
Never heard of this. How can I get in on it?
This appears to be sponsored by some blockchain charity rather than anyone affiliated with Rust
>When i talk about something i like its based
>When you talk about something i dont like its cringe and shilling
KYS.
> the image
Wew, imagine him learning that printf compiles to.
and this is the C++ hello world after (partially) expanding it
https://0x0.st/oSih.cpp
Apples to oranges.
You're right, rust's macros are hygenic unlike C's dumb text copy-paste. The only reason people hold the
>muh macro bad
opinion is because C's macros are too shit and too much of a footgun even for cniles.
All macros that aren't LISP macros are shit.
Lisp macros are shit, they aren't even hygienic.
>they aren't even hygienic.
No, that's your front hole.
How's this work for platforms that Rust doesn't support?
*Linux doesn't support ;^)
but in all seriousness, rust supports all major platforms and will be mostly (only) used in drivers, probably for hardware that the snowflake platforms will not use anyway
why are rustrannies so obnoxious? just make your own kernel
>make open source kernel
>constantly shill "muh freedoms" everywhere
>tell everyone they can do what ever they want with it
>they do what ever they want with it
>NOOOOOOOOOOOOOOOO STOP DOING THAT
fork it then troony
lmao remember glimpse?
we did, it's on kernel.org, if you don't like it fork it then
we did, it's also on kernel.org. it's called linux/torvalds. please keep your rust out of our repo.