The case against a C alternative | Hacker News Reader

The case against a C alternative

c3.handmade.network

217 pointslerno379 comments

> Any C alternative will be expected to be on par with C in performance. The problem is that C have practically no checks, so any safety checks put into the competing language will have a runtime cost, which often is unacceptable. This leads to a strategy of only having checks in "safe" mode. Where the "fast" mode is just as "unsafe" as C.

I don't think this is true, in the general case: Rust has shown that languages can be safe in ways that improve runtime performance.

In particular, languages like Rust allow programmers to express stronger compile-time constraints on runtime behavior, meaning that the compiler can safely omit bounds and other checks that an ordinary C program would require for safety. Similarly, Rust's (lack of) mutable aliasing opens up entire classes of optimizations that are extremely difficult on C programs (to the extent that Rust regularly exposes bugs in LLVM's alias analysis, due to a lack of exercise on C/C++ inputs).

Edit: Other examples include ergonomic static dispatch (Rust makes things like `foo: impl Trait` look dynamic, but they're really static under the hood) and the entire notion of a "zero-cost abstraction" (Rust's abstractions are no worse than their "as if" equivalent, meaning that the programmer is restricted in their ability to create suboptimal implementations).

Not to mention that both C++ and Rust can specialise algorithms and containers for specific types, whereas in C most developers resort to void* and function pointers. It's not unusual to see C programs written in a "typical" C style become dramatically faster when rewritten in a more modern language.

For example, typical C programs also don't use hashtables even when this makes the most sense, causing weird performance cliffs due to O(n^2) algorithms all over the place. Why not hashtables? Because they're not generic, so they're a pain to use. Not impossible of course, it's just that C developers avoid them.

Similarly, "strongly typed" containers full of simple struct types enable compiler auto-vectorisation that's often unavailable in C for the same kind of reason.

Last but not least, you would have to be a masochist to write heavily multi-threaded code in C... so hardly anybody does. These days, that's throwing away over 90% of the computer power in a typical PC, let alone a server.

It's not unusual to see C programs written in a "typical" C style become dramatically faster when rewritten in a more modern language.

This was equally true back when C vs Fortran was the big debate, and something not easily captured in benchmarks. C, as written by an expert in high performance C, was equally fast as Fortran written by an expert in high performance Fortran. C, as written by a domain expert with limited programming skills, was often very much slower than Fortran written by a domain expert with limited programming skills.

This actually reminds me a bit of an old competition between two Microsoft MVPs comparing C++ and C#, where they went back and forth optimizing their respective versions of a model program, and discussing the optimizations they made.

The gist, as I recall it was: the initial, idiomatic, written-for-maintainability version of the C# program was significantly faster than the C++ equivalent. Up until the end, the C# version also generally needed to go through less heroics to keep up with the C++ version. Eventually, the final C++ version did end up being faster than the fastest C# one, but, considering what needed to be done to get there, it was a decidedly Pyrrhic victory.

One huge mitigating factor, though, is that the model program was doing something business-y. I doubt C++ would have had such a hard time of it if it had been a number crunching or systems program.

C# is closer to Java than C++. It’s a garbage collector language and nowhere near the performance of C++

Precisely. This is perhaps the strangest part of the original post: C++ has the same performance advantages as Rust! It has them not because it's more safe (although it is, in some regards), but because it allows programmers to express behaviors that the compiler can reason about statically.

C++ has a few of the performance advantages that Rust has.

And many that are negated due to a bad std API.

Interesting. Can you list an example of C++ type system allowing optimization Rust system doesn't?

Rust's assumption is that it's the compiler's job to reject all wrong programs (usually with a helpful diagnostic). In C++ the assumption is that it's the compiler's job to permit all correct programs.

You obviously ideally want both, but that's not actually possible when you have a language this powerful. So, Rest's choice means sometimes (more rarely these days but it can happen) you will write a program that is correct, but the compiler doesn't believe you and rejects your program, you will need to alter it, perhaps after alterations it's actually nicer, but equally perhaps you feel this made it uglier or slower, nevertheless you have no choice in Rust (well, you could try waiting a few years, the compiler gets smarter)

However the C++ choice means sometimes (maybe even often) you will write a program that isn't correct and the compiler gives you no indication whatsoever that there's a problem, you get an executable or object file or whatever out, but what it does is completely arbitrary. Maybe it works how you expected... until it doesn't.

The magic phrase in the C++ standard is "Ill-formed, no diagnostic required". For example suppose you try to sort some floats in C++ 20. That's ill-formed (floats aren't in fact Totally Ordered but the function signature says you promise they are) and no diagnostic is required for... whatever it is your program now does. Maybe it crashes, maybe it works fine, not their problem, good luck with that.

Now, probably if all your floats are like boring normal finite reals like -2.5 or something this will work fine, there's no practical reason it wouldn't, but who knows, the C++ language denies all responsibility. So it gets to be very "optimal" here since it can do whatever it wants and it's your fault.

To expand on your float sorting example, sorting a slice[1] in Rust requires the element type to implement the Ord trait, i.e. be totally ordered. Trying to sort a slice of floats will result in a compiler error, even though it might be totally fine as long as all your floats are "ordinary".

Instead, to sort a slice of floats, you have to explicitly specify what would happen for the non-ordinary cases; e.g. by using `.sort_by(f32::total_cmp)`, where f32::total_cmp()[2] is one possible interpretation of a total ordering of floats. This requires writing more code even for cases where it would be completely unnecessary.

[1]: https://doc.rust-lang.org/std/primitive.slice.html#method.so... [2]: https://doc.rust-lang.org/std/primitive.f32.html#method.tota...

I don't think you could do [0] in Rust, but would be very interested in finding out otherwise.

[0] https://capnproto.org/news/2015-03-02-security-advisory-and-...

Not on stable, but you can on nightly (though it's still quite wonky): https://play.rust-lang.org/?version=nightly&mode=debug&editi...

I don't think they suggested this.

template<size_t n> struct example { int e[n]; };

Rust:

  struct Example<const N: usize> {
      pub e: [i32; N],
  }

One cannot imagine C++ failing to emulate Rust's borrow checker by the end of the decade, even if compiler support is required.

Has C++ ever failed to snarf a feature?

Safety features. The committee are, perhaps unconsciously, biased against safety on the presumption (seen in many comments here on HN) that safer has to mean lower performance.

But part of the impetus for Carbon is that WG21 (the C++ Standards Committee) rejected proposals that C++ should focus on better performance and safety. So maybe performance is no longer important either. What's left?

Where they've taken things which might appear on the surface to be modelled on a safer Rust feature, usually the committee insists they be made unsafe. For example suppose I call a Rust function which might return a char or might not, it returns Option<char> and if I'm an idiot and I try to treat that as a char, it doesn't type check because it isn't one, I need to say what I'm going to do when it isn't or else that won't compile.

You can write that in modern C++... except it can automatically try to take the char (which isn't there) out of the empty optional structure and that's Undefined Behaviour. So whereas the Rust prevents programmers from making easy mistakes, the C++ turns those into unexploded bombs throughout your code.

Many on the C++ committee are interested in the borrow checking, but are not sure how to make it work in C++. The hard part is they cannot break compatibility with code that is legal with previous versions of C++. If there is even one pathological case where the borrow checker will reject code that doesn't have a memory leak then they will not accept it, and require whoever proposes this borrow checker to prove the absence of such a thing. (note if it rejects code that worked until the leaks mean you run out of memory they will accept that). I don't know if such a thing even exists, but if it does I'm confident that in Rust it is new code that you can write differently to avoid the bug, while with C++ that may be a very massive effort to figure out 25 year old code nobody understands anymore before you can rewrite it.

One obvious corner case: It is very common to allocate a buffer at startup and let the system clean it up when the program exits. (often this is embedded cases where the only way for the program to exit is power off). I don't know how you do this in rust (if you can - I'm not a rust expert)

Box::leak. Assuming the object you're leaking contains no references, it can then live for 'static (the lifetime of the process)

> allocate a buffer at startup and let the system clean it up when the program exits

This is possible with the lazy_static library or (not yet in stable Rust) OnceCell. It allows you to allocate & initialize any datastructure once during runtime and get global read-only access.

I don't know Rust but I know C++.

And C++ has the potential to be faster than C, mostly thanks to metaprogramming (templates, ...). It is horrible if you have to do it, but if you are just using the standard library, you don't have to feel the pain but still take advantage of it. That's how algorithms are implemented. Because so much is known at compile time, optimizers can do a lot.

The reason C++ is generally regarded as slower is that C++ programmers tend to create objects on the heap all the time because constructors and destructors make it easy. Modern C++ also discourages raw pointers and so you get references counters all over the place, essentially turning C++ into a garbage collected language. I am not saying it is bad, but it certainly impacts performance.

But if you manage your memory in C++ just as you do in C, keeping track of all your buffers reusing them, and not using more than necessary, I can easily see C++ beat C.

> Modern C++ also discourages raw pointers and so you get references counters all over the place, essentially turning C++ into a garbage collected language.

This doesn't match my experience. It's true that modern C++ discourages owning raw pointers, but the solution is usually unique_ptr, not shared_ptr. Truly shared ownership is actually pretty uncommon IME, usually you can have one entity which obviously "owns" the object and then any other reference to it can be non-owning.

It's also worth noting that with std::move, actually changing the refcount of a share_ptr can be pretty rare even if you do have shared ownership.

This is not my experience. Most developers are just not very good at what they do, and the go-to smart pointers for not-very-good C++ developers is std:shared_ptr<T>.

This has been my experience as well - especially when C++11 came out. I have seen codebases where it has been "use std::shared_ptr for everything, becuase it is safer if/when we use threads". I know that doesn't make sense, but it just was the attitude back then.

Tbh, Back then, I didn't see a problem with it. Once i started chasing down weird bugs where objects aren't freed properly because no one knew which objects own what, I have been very cautious.

We had a few developers like that here when C++11 was introduced, but a few people gave them the smack down and now we rarely see shared pointers.

unique_ptr is pretty bad for performance as well. It is more complicated to use compared to raw pointers and encourages an OOP object-per-object piecemal code and data architecture. I've never seen a C++ program making use of unique_ptr that didn't give a strong smell of enterprise programming.

As of C++17 not so horrible, and C++2x versions even less so, unless one has some strange fetisch for SFINAE and tag dispatch.

Since 1993, I never saw any need to keep bothering with C other than having it imposed on me, C++ had enough C89 subset on it, if I ever miss coding like C and its warts.

Nowadays that compatibility is up to C11 subset.

> Nowadays that compatibility is up to C11 subset.

Not true unfortunately, the "C subset" is still stuck at something that can at best be called a fork of "C95" which was then developed into a "bastard language" that resembled C on the surface, but isn't actually C (e.g. the incomplete designated init support in C++20 is the best example of this half-assed "looks like C, but isn't actually C" philosophy).

That is why it is called a subset, duh!

> It's not unusual to see C programs written in a "typical" C style become dramatically faster when rewritten in a more modern language.

On the other hand, empirically, it is not unusual to see straightforward C programs being dramatically faster than comparable C++ programs written in enterprise style, and to also build much faster.

> Last but not least, you would have to be a masochist to write heavily multi-threaded code in C

You have to be a masochist to write heavily multi-threaded code that uses a lot of ad-hoc synchronization with mutexes and atomics. As it turns out, for many many tasks, it's also a spectacular bad way to go about parallelization, because mutexes are the _opposite_ of parallelization.

As a rule of thumb, do coarse-grained concurrency. Install a few queues, come up with a job system, and it won't be hard to get parallization right in plain C at all. Writing in C is often a good idea because what's a bad idea to do on hardware coincedes pretty well with what is painful to write.

> On the other hand, empirically, it is not unusual to see straightforward C programs being dramatically faster than comparable C++ programs written in enterprise style, and to also build much faster.

Your only comparison cases are cases where the code in question was re-written in C. This most likely means that everyone already knew it was slow and so the re-write also fixed the fundamental problems. If the code had been rewritten in C++ it would also be faster - and since C++ allows some optimizations C doesn't it would be even faster. (it is known that if you switch from gcc to g++ your code often will run faster if it compiles)

There is a reason for enterprise style C++. Most of the time it is still fast enough, and it is a lot more maintainable.

> it is known that if you switch from gcc to g++ your code often will run faster if it compiles)

I've never heard such a claim, can you back it up? And what does it say about the language?

> and it is a lot more maintainable

If you equate "maintainable" = readable, I've never once seen maintainable enterprise code. Everything is a convoluted mess that never gets anything done. Probably I haven't worked at the best shops, but then again, where are those? And why doesn't the language help mediocre programmers to write maintainable code?

I suspect that maintainability is almost exclusively a function of experience, not the programming language used. Experienced programmers do seem to agree that C-style C++ or even plain C is the way to go.

> become dramatically faster when rewritten in a more modern language

IME that's mostly a myth though. A C compiler will stamp out a specialized version just as well if it can see all the relevant function bodies (either via inlining or LTO).

"Zero cost abstraction" isn't just a C++ thing, it happens mostly in the language agnostic optimizer passes. For instance the reason why std::sort() shows up faster in benchmarks than C's qsort() is simply because std::sort() implementation is all inline template code, not because of some magic performance-enhancing qualities of the C++ template system.

inlining only goes so far. You won't get full of qsort to be inlined, and if it's not inlined, it needs to be at least cloned to be on par with std::sort, so the comparator function could get const-propagated.

AFAIK out of the major compilers, gcc has the most aggressive cloning, but it's still nowhere near to const propagate the comparator from qsort. With std::sort with a stateless comparator function object (such as std::less, which is the default), you get this for free*.

* of course this is not entirely free, as this is more prone to code bloat. But, you can always type-erase the comparator, and use a function pointer, or std::function, if this ever becomes a problem. But you can't convince a C compiler to const propagate the comparator in qsort all the way through, if the optimizer chooses that it doesn't worth it.

glibc qsort's implementation is in libc.so, not in the header. GCC doesn't have anything to work with.

It's also an apples-to-oranges comparison, since std::sort and qsort implement different algorithms.

A lot of std::sort's performance is actually from using the version without any callbacks. If you pass a comparator function which just compares two integers the obvious way, it gets much slower. So one of std::sort's biggest advantages is actually not that it uses templates, but that it's specialized for the common case of not needing a custom callback. Theoretically the compiler should make the two cases the same, but apparently GCC is too dumb (that's not a slight on GCC; I think people expect too much from compilers):

  ------------------------------------------------------------------------
  Benchmark                              Time             CPU   Iterations
  ------------------------------------------------------------------------
  std_sort_random                 52881299 ns     52873089 ns           14
  std_sort_with_callback_random   63319633 ns     63307876 ns           11
  qsort_random                   106803314 ns    106784567 ns            7
  external_sort_random            97642851 ns     97640888 ns            7
  std_sort_sorted                  8433311 ns      8432564 ns           82
  std_sort_with_callback_sorted   13868016 ns     13865170 ns           50
  qsort_sorted                    28098439 ns     28093720 ns           26
  external_sort_sorted            33629020 ns     33628108 ns           21

external_sort is just std::sort hidden behind an extern function implemented in a separate .o file. Those benchmarks are from sorting 1MB of random and already-sorted data (as indicated in the names). I think it's important to test such cases, because often online people benchmark code which is written all in a single file, whereas real-life C++ projects are usually organized in such a way that every little class is in its own little file, which gets compiled into a separated object file, and then it all gets linked together without LTO. And then those same people go on to claim performance benefits of their language without actually using the setup which enables those benefits, which IMO is a bit dishonest.

When I drill further down into everything I want to drill into, maybe I'll publish the source for the benchmarks somewhere.

check this out: https://github.com/WebKit/WebKit/blob/main/Source/bmalloc/li...

qsort only isn't inline, because libcs don't supply an inline definition. If you write your own qsort, then you'll see it getting inlined and/or function cloned for different types.

The only real difference between qsort and std::sort in terms of code generation, is that for std::sort the default assumption is to function clone and for qsort it is to generate the full slow function. Now the compiler will in most cases detect that qsort can be cloned or inlined, but sometimes it might decide not to and the fallback is, in most cases slower then the C++ fallback.

PS.: I'm just annoyed that my generic C hashtable that is written in a qsort style doesn't get function copied/inlined when it's used for more than one type.

You can't just ignore that qsort is implemented in a different library just because it could be implemented inline in the header.

It could and it should, but function specialisation without inlining is still on the wish list for llvm last I checked.

I totally agree with you, but are we really expecting "a typical PC" to have 10+ threads?

Gonna beat a dead horse here, but >50% of PCs that are surveyed by Steam have 12 threads or more.

That’s PCs that have steam installed at all.

Intel’s bare minimum current-gen i3 processor has 12 threads. That’s the absolute cheapest desktop-level processor you can get.

Your phone probably has 6 cores (though not 12 threads).

So yes, if you’re writing code for desktop hardware, it’s safe to assume you have at least 8 threads. Maybe you don’t want to consume all of them, but it’s better to let the OS handle scheduling.

https://www.techspot.com/article/2363-multi-core-cpu/

Gaming is very much not representative. There's roughly 120M active steam users, vs. ~1.4 billion windows installs.

If I look around me, for instance in my whole family we're two with Steam installed but ever household has a desktop or a laptop (and generally a 7-8 years old cheap entry-level 350€ one, you'd be hard-pressed to find even a quad-core in there)

This article doesn't say that, what it actually says is "Over 70% of Steam users have a CPU with 4 or more cores."

Steam doesn't even measure publicize information about threads on the survey, which makes it near impossible to check because not that long ago Intel locked out hyperthreading/SMT on their low/mid-grade CPUs.

Additionally, and more importantly: the Steam hardware survey _obviously_ doesn't represent the average consumer PC.

My laptop has 8 cores and 16 threads.

I'm about to buy a PC with 16 cores and 32 threads for "normal" money.

The AMD EPYC server CPUs scale to dual sockets with 64-cores each, for a whopping 256 hardware threads in a single box. That's not some sort of esoteric hyper-scale configuration, but completely ordinary off-the-shelf stuff you can get in large quantities from mainstream vendors.

A single-threaded application on a server like that will use between 0.5% to about 50% of the total available performance, depending on where its bottleneck is. It will never reach 100%!

This matters to things like CLI tools, batch jobs, and the like, many of which are written in C, especially in the Linux world. A case-in-point that demonstrates how much performance has been left on the table is ripgrep, which is a multi-threaded Rust replacement for grep.

My Ryzen 5 3600X (a years-old mid-range CPU) has 6 cores and 12 threads, I wouldn't say that's atypical.

Today, it's debatable, but if we're talking about programming languages for the future then the future is what's relevant. I don't think it will be long before 50+ thread CPUs are common. Multithreading won't be a nice-to-have feature, it will be a necessity.

Given that many phones these days have half a dozen cores...

Why not? I just opened htop on my desktop, and I have 789 threads running over 117 processes, with 12 logical cores.

> I totally agree with you, but are we really expecting "a typical PC" to have 10+ threads?

I think you are mixing hyperthreading, or SMT with regular "software" threading

They're mixing parallelism and concurrency. (nb: I might be abusing these terms too)

Parallelism aka CPU-bound tasks are limited by the number of cores you have. Concurrency aka IO-bound tasks are not, because they're usually not all runnable at once. It can be faster to go concurrent even on a single core because you can overlap IOs, but it'll use more memory and other resources.

Also, "going faster" isn't always a good thing. If you're a low priority system task, you don't want to consume all the system resources because the user's apps might need them. Or the the user doesn't want the fans to turn on, or it's a passive cooled system that shouldn't get too hot, etc.

And for both of them, it not only makes it easier to write bugs in unsafe languages, but in safe languages you can easily accidentally make things slower instead of faster just because it's complicated.

If not today, definitely tomorrow.

> it's just that C developers avoid them.

I'm not sure it's fair to say C developers avoid hash tables - I've worked on several projects with hash-table implementations in them.

The 'problem' if there is one, is that such things are rarely picked up from any sort of standard library, and are instead implemented in each project.

I'm also not really sure what the problem is with 'resorting' to void*, it's part of the language. It's not 'safe' in that the compiler won't catch your errors if you stuff any old thing in there, but that's C.

> you would have to be a masochist to write heavily multi-threaded code in C

pthreads makes it relatively straightforward. I've seen (and written) fairly sophisticated thread-pool implementations around them.

C noob here. Why isn't a hash table implementation merged into the c standard library? Is it because the stdlib has to be as thin as possible for some performance reason or something?

I did C for about 15 years....

Yeah C doesn't really go in for that sort of thing. The standard library tends to be much more about some minimal support for strings and interfaces to OS features like files, signals, memory allocation etc. It doesn't really provide much in the way of building blocks to be reused by application developers.

The recommendation out there on the net seems to be to look at Glib, which is used by gtk, for that sort of thing.

Another good alternative might be the NSPR - https://firefox-source-docs.mozilla.org/nspr/reference/index...

I used this way back in 2001-3 for a multi-platform project because it provides some good platform abstractions, and it looks like it has a hash-table implementation in amongst its other features.

"void* and function pointers" behaves essentially the same as templates, assuming the compiler inlines or function clones the function called with constant expression arguments.

All equivalent code has the same performance with a sufficiently smart compiler. In practice, compilers are often not sufficiently smart.

Without the related typechecking.

What did you mean when you said C hashtables can’t be generic? Is the (void*) not an adequate solution?

It depends on what you're trying to do. In general, marshaling everything through void pointers is possible, but it'll cost you in terms of both bug surface (it's much easier to make mistakes when you've opted out of C's weak types) and performance (you now have at least one mandatory pointer indirection, which is especially egregious when your underlying value type is something simple like an integer).

Anything you can do in C++, you can do in C. But C++ compilers will generally optimize semantically equivalent code better than C compilers will, because C++ gives the compiler more freedom.

Another perfectly good solution is to treat a C hashtable as a sort of acceleration index. That C hashtable then, rather than holding pointers simply holds `hashkey -> i` where i is the position into a typed array.

I.e. your data structure is like this:

  generic_hashtable Hash; // (hash->int)
  Foo *values; // where my typed data is.

Using void* means the compiler (almost certainly?) can’t see through it to optimize. More importantly, it looses you type safety and the self-documentation that flat_hash_map<K, V> gives you.

There is no obvious way to get a hash from (void*).

And yes, C developers also tend to avoid ordered containers.

My favorite example of a safety feature in Rust which also improves runtime performance is string slices, which are implemented as a pair of a pointer to the first character and the length (used for bounds checking). Not only does this avoid having to scan for the NUL terminator to find the string length (that is, O(1) instead of O(N), which can make the difference between an O(N) and O(N^2) algorithm), but also it allows taking substrings without copying or modifying the original string (also helped by the borrow checker allowing the programmer to safely omit defensive copies of strings).

Made this C lib for exactly the reasons you mention. I love C but got mad at the constant bounds errors and copying.

https://github.com/alcover/buffet

C has a huge performance burden from 0 terminated strings. Programs are constantly running strlen() or equivalent to get the length. Length-delineated strings, like what D has, are an order of magnitude faster.

Exactly! And because you can't create a substring without mutating the original value, you see that C code often needs to resort to unnecessary copying as well.

No. You can't create a zero-terminated substring other than a proper suffix or a buffer copy. But that's not really a surprise right?

Well then, don't use zero-terminated strings for proper string processing. You don't have to use zero-termination, even when some programmers in the 70s and 80s were convinced enough of it that abominations like strtok() landed in the standard.

>You don't have to use zero-termination

You can choose between zero-termination and having to convert strings back and forth when using idiomatic libraries.

_There are_ programs that are constantly running strlen(). C strings are the default builtin string representation that has an acceptable tradeoff for performance vs space and simplicity for where they are used: Mostly in string literals, which are expected to be small strings. Mostly for printf() and friends. Zero-terminated strings are space efficient and don't allow bike shedding like length-prefixed strings do. And don't get us started about allocation strategies.

"A magnitude faster" for doing what? Typical usages of zero-terminated strings are performance-uncritical. And note that zero-terminated doesn't preclude using separate length fields.

Sane programs use store length of strings explicitly where strings get longer and/or performance is a concern, just as it is the case with other types of arrays.

> that has an acceptable tradeoff for performance vs space and simplicity for where they are used

Is it? I've been programming strings for 45 years now. Including on 8 and 10 bit machines. All that space efficiency goes out the window when one wants a subset of a string that isn't a common tail.

The simplicity goes out the window as soon as you want a substring that isn't a common tail. Now you have memory allocation to deal with.

The performance goes out the window because now the entire string contents has to be loaded into the cache to determine its length.

> length-prefixed

Are worse. Which is why I didn't mention them.

> Sane programs use store length

Meaning they become length-delineated programs, except it's done manually, tediously, and error-prone.

Whenever I review C code, the first thing I look at are the strlen/strncpy/str** sequences. It's almost always got a bug in it, an off-by-one error.

Again, I'm not saying you should represent substrings, or strings in general for that matter, as zero terminated strings, and I'm not saying use zero terminated strings for anything longer than a couple bytes.

No, I recommend everyone to use whatever fits the situation best. It might be a 2 byte start index and a 1 byte length fields that expresses the length as a multiple of 12 bytes. It might be rope data structure. Or it might be whatever. "String" is not a super well defined thing, and I don't understand why everybody is so super concerned about a canonical string data type. String data types are for scripting languages. 99% of my usage of string (literals) is just printf and opening files, and C does these just fine.

Zero terminated strings are only a default thing for string literals that does indeed bring a little bit of simplicity and convenience (no need for a builtin string type and the associated bike shedding, and only need to pass a single pointer to functions like printf).

> Meaning they become length-delineated programs, except it's done manually, tediously, and error-prone.

Not sure when is the last time I found it "manually, tediously, and error-prone". There are very rare cases where I have to construct zero-terminated strings from code, or need to strlen() something because of an API. And even when these cases occur they don't bother me at all. Stuff just works for me generally and I'm moving on. I have probably 500 stupid bugs unrelated to string handling before I once forget a zero terminator, and when that one time happens I just fix it and move on. On the plus side, given that we're in C where there are no slice types, zero-terminated strings spare me to pass extra length values for format strings or filepaths.

Sometimes I envision being able to use slices but I have some concerns if that would be an actual improvement. Importantly it should be about arrays and not just about strings. Strings are arrays, they aren't special.

I think a good design for slices could be one whose length can never be accessed by the programmer, but which can be used for automated bounds checks. Keeping size/capacity/offset and 43 cursors into whatever buffers separate is actually correct in my view from a modularization standpoint, because "String <-> Index/Size/Offset etc." isn't a 1:1 relationship.

> Whenever I review C code, the first thing I look at are the strlen/strncpy/str* sequences. It's almost always got a bug in it, an off-by-one error.

You will have to look quite a bit to find strlen() or strncpy() in my code. I'm not advocating for them, and not advocating to build serious string processing on top of zero-terminated strings.

Zero-terminated strings was a bad decision even back then, let alone now. They make vectorization very painful, and you just needlessly have to iterate over strings at every use-site.

Except nobody cares about vectorization of your printf("Hello, World\n") or other 12-character strings. Vectorization here would in fact be a waste of build time as well as object output size, and the runtime performance would be not measureably different, possibly even slower in some cases. It's a total waste.

When you're processing actual buffers full of text or binary data, and performance matters, of course you are not advised to use an in-band signaled sentinel like zero-terminator is. Use an explicit length for those cases.

Additionally, back in the 8 and 16 bit home computer days, C programms were full of inline Assembly, because C compilers generated garbage machine code.

It was 40 years of throwing UB optimizations at it, that made its current performance come true.

This is nothing special, any language can eventually reach that reality with similar investment.

Rust does runtime bounds checks on array access, which the compiler can't elide in non-trivial cases (e.g. binary search), so if you want to write such algorithms as fast as in C, you need to use a lot of unsafe array indexing.

I would be interested in how big of a difference does it make — these branches will always go one way, so they are trivial to branch predict. Other than very slightly increasing the code size (which can have a big effect if the i cache is no longer sufficient), I fail to see it doing too badly as is.

You might be referring to slices. Arrays in rust have to have a known length at compile time.

Rustc may still need to do runtime bounds check if it can’t conclude that a dynamically computed index is in-bounds.

The length is known statically, but whether the index is in-bounds may not be.

The binary search GP talks about is exactly one such case, go on Godbolt, write up a simple binary search (with a static size so you get code), and you’ll see that the compiler checks against the literal array size on every iteration.

I have no idea how I totally missed that from the GP's comment. Thanks for correcting me.

this pretty much summarises my opinion - one nitpick - i assume you meant "omit bounds and other checks", not "emit bounds and other checks" which seems to mean the opposite of what you're intending

Rust does emit bounds and other checks, though. Optimization passes can usually clear some of them away, but you'd need to check the assembly output to be sure.

Yes, that's "omit".

"Emit" means "to send out", eg "emit a strange noise", "emit radiation".

It is specifically both:

- trying to access an arbitrary element in a slice, the compiler will emit bounds checks (`if index > len: panic()`) to avoid an uncontrolled out-of-bounds memory access — https://godbolt.org/z/cbY5ebzvK (note how if you comment out the assert, the code barely changes, because the compiler is adding an invisible assert of its own)

- if the compiler can infer that `index` will always be less than `len`, then it will omit the bounds check — https://godbolt.org/z/TTashYnjd

Yes, thank you! That's an embarrassing typo!

(And thanks to the other person as well, who presumably deleted the same comment after seeing yours.)

But not always. Some algorithms cannot be written in a way that would remove run time bound checking from rust e.g. KMP.

There are in fact ways that idiomatic Rust can be even faster than idiomatic C.

> any safety checks put into the competing language will have a runtime cost, which often is unacceptable.

And what is the runtime cost of all the mitigations put in place because we don't use a memory safe language? Stack canaries, safe stacks, ASLR, control flow integrity, code pointer integrity, runtime attestation, library re-linking and randomization. Not to mention sandboxing techniques and other system level mitigations.

I suppose I should thank the C language for job security as a security engineer.

And who says that faster runtime trumps all other considerations? I would much rather have my computer run slower than have to continuously deal with security vulnerabilities. My time is much more valuable than CPU time.

> I would much rather have my computer run slower than have to continuously deal with security vulnerabilities. My time is much more valuable than CPU time.

To be devil’s advocate and take the other side of this argument: I would rather the CPU I paid for spend its cycles performing actual application logic. Every cycle spent executing all this “safety code” overhead feels like me having to pay for a developer’s sloppiness.

I feel end users pay a high cost for all this runtime checking and all these frameworks and levels of abstraction.

First, not all language safety features have to manifest themselves as run-time checks. A properly designed language will allow many safety checks to be done at compile-time.

And second, would you really rather deal with security holes on an on-going basis?

The problem with C is not that you can write unsafe code in it, it is that the design of the language -- and pointer aliasing in particular -- makes it impossible to perform even basic sanity checks at compile time [EDIT: or even at run-time for that matter]. For example:

int x[10];

...

int y = x[11];

In any sane language, that would be a trivially-checkable compile-time error. But in C it is not because there are all kinds of things that could legally happen where the ellipses are that would make this code not violate any of C's safety constraints. That is the problem with C, and it is a fundamental problem with C's design, that it conflates pointers and arrays. C was designed for a totally different world, and it is broken beyond repair.

I'm not here to defend C, but to point out a problem that many advocates make when they cherry pick examples of how their language is both safer and more performant than C. Simply put, that example is irrelevant when the size of the array is not known at compile time. The moment that you have a dynamically allocated array, you are either dropping safety (e.g. expecting the developer to perform bounds checks when necessary) or performance (e.g. the compiler inserts bounds checks at runtime).

It is also worth considering that there is nothing preventing C compilers from inserting compile time checks to address examples such as yours. I just tried to compile a similar example in gcc and it does catch the unsafe code with warnings and the optimizer turned on.

Bad example. There's nothing[0] you could put in the ellipsis to make that code valid, and both gcc and clang will warn about it (clang on defaults, gcc with -Wall).

[0] Ok, I guess you could #define x something, but that's not interesting from a static analysis perspective.

What do you suppose you could put in the ellipsis that would make your second statement defined behavior other than preprocesor stuff or creating a new scope with a different variable named x? (Neither of which would frustrate a compiler wanting to give you a warning)

The real overhead is not safety code but inefficient design and unnecessary (and often detrimental to the user) features. The added 10% of safety checks is nothing compared to orders of magnitude of bloat modern software has. Moreover, a lot of C codebases have the same safety checks manually implemented into them, sacrificing performance in order to have less bugs. And when it comes to designing a modern C alternative, it's not just a performance/safety tradeoff, the are parts of C that are simply bad, like null-terminated strings, and are long due to be replaced with something better.

I'm suspicious that people arguing that safety checks making things slow are basing that on the very stale now idea that CPU's work sequentially. Which isn't true for super scalar machines at all. And also forgetting compilers will optimize away a lot of them.

There is an asymmetrical risk here. On one hand the compile might emit checks that aren't needed. On the other that the programmer will fail to insert a check that is needed.

>I would rather the CPU I paid for spend its cycles performing actual application logic.

The problem is that for CPU programming errors are application logic too and they are run, cf. protected memory. You don't run everything in ring0 do you?

Is Rust slower then C in general though? And how much are you paying? Probably in true terms it costs less if there are fewer bugs and vulns.

That's not the point. The point the parent is making is that yes, languages that include safety features are slower, but you will actually need those safety features in C as well, in the code. So your program will also be slower in C.

There is something backwards about how safety checks are done in C.

First it's up to the programmer to put them in. Then hopefully the compiler will optimize away the unneeded ones. If the programmer though biffs and forgets then oops. And of course classically we blame the programmer not the system he's been forced to work under.

That seems like a reasonable thing in 1982. Which is 40 years ago.

It would be better if the compiler implemented the checks automagically and removed ones it knows it doesn't need. And bonus, if the programmer puts one in, leave it alone.

Yes, I think we're actually all in agreement here. The common point is that TFA's argument that we need to keep C because it's faster is invalid. The GP's point is that it is not in fact faster because of all the things you need to do to produce code that can actually be deployed in today's world. My point is that even if C were faster (which it actually isn't) that would not matter because C is unsafe even with all the extra stuff that people stick onto it to try to make it safe. So the original claim is really a judgement call that speed matters more than safety, and this is not something about which there is any kind of consensus.

What does slower actually mean? Slower in the face of safety, while computers continue to get faster and or cheaper (depending on how you spend your transistors/cost) is a fools paradise. Don't be baited into their flawed logic.

If your transistor/cost curve has a doubling time of 3 years, 156 weeks. A 5% difference is approximately 11 weeks.

If your transistor/cost curve has a doubling time of 2 years, a 5% difference is approximately 8 weeks.

How fast do we have to get before safety is table stakes? Focusing on raw unsafe speed wouldn't be a normalized metric in any other industry. I'll spend 8-11 weeks of performance gains on correctness.

C was an inside job.

You're assuming that safety features will prevent security bugs. I think that is...optimistic at best.

Yes, they can defeat a few classes of exploits, but generally not the ones that lead to really bad outcomes.

> The Chromium project finds that around 70% of our serious security bugs are memory safety problems.

> ~70% of the vulnerabilities Microsoft assigns a CVE each year continue to be memory safety issues

Memory safety is a leading source of serious bugs in a big group of operating systems, browsers, image and file parsers, and more.

It does seem likely we can stop this, and it doesn’t seem likely to me we can fix it in C. We have tried and failed for years at that.

Solaris SPARC is the first that has suceeded at it, and now others follow it, how?

Hardware memory tagging, the language itself is beyond hope.

If you remove the mechanically preventable bugs from being a consideration, by definition the only one you now need to focus on are the ones NOT prevented by mechanism.

How is this not a win? We only have so many decisions we can make per day.

A lot of people make perfect the enemy of good.

So, let's take the most extreme example of pursuing safety, WUFFS.

Unlike general purpose languages WUFFS has a very specific purpose (Wrangling Untrusted File Formats Safely) so it doesn't need to worry that it can't solve some of your problems, which frees it to completely refuse to do stuff that's unsafe, while going real, real fast.

WUFFS gets to completely omit bounds checks, which you probably wouldn't have dared try in C because it's so obviously unsafe, but WUFFS already proved at compile time that it can't ever have bounds misses so it needn't do these checks. WUFFS gets to also omit integer overflow checks, because again it proved at compile time that your code is correct and cannot overflow for any input. And since WUFFS knows how big the data is at all times it gets to automatically generate loop unrolling and suchlike accelerations.

WUFFS isn't a general purpose language, it doesn't have strings, it doesn't have growable arrays (what C++ calls "vectors"), it doesn't have any dynamic memory allocation, but then we weren't talking about how much you love general purpose features, we were talking about safety preventing security bugs. Which is exactly what WUFFS does.

I can't respond to that unless you are more specific about what are "the ones that lead to really bad outcomes". There are a lot of arbitrary-code-execution attacks that are enabled by buffer overflows, and IMHO that's as bad as it gets. There is absolutely no legitimate excuse for a buffer overflow in today's world. Switching to safe(r) language won't prevent all attacks, but it will make a big dent.

It defeats some extremely important classes of exploits. And I'm not sure how they're not ones that lead to really bad outcomes since they lead to fun ones such as arbitrary code execution all the time. I can create a C program with hideous vulnerabilities in about five minutes without doing anything that isn't totally standard and normal (albeit obviously vulnerable so technically buggy). I'd have to actually look up how to make my code vulnerable in languages with more safety features.

And the latest trend, producing C Machines with hardware memory tagging, because none of those mitigations are actually working preventing all those CVE to take place.

You are the security engineer, so you certainly know better than me, but aren't those runtime mitigations aimed at malicious programs? Which is to say, even if a better-C was written that didn't allow people to write a program that would bump into those mitigations, the bad guys could still write their programs in assembly or C or whatever, right?

They're aimed at protecting innocent programs from exploitation. If we knew which programs were actually malicious, our lives would be much easier!

But the presence of a better language to write innocent programs in wouldn't protect the innocent programs from malicious programs written in C and assembly...

Think about web browsers. We're not really that worried about people running malicious web browsers. It's potentially a problem, but as long as people know not to run software that some random spammer sends them in an email then it's easy to avoid.

On the other hand what computer security people worry about a lot is that the web browsers made by reputable organizations and teams of competent programmers nevertheless contain security flaws that can be exploited by a maliciously-written website to cause those browsers to do unexpected and dangerous things.

Many of those security flaws in otherwise well-regarded software are due to memory management errors that just aren't present in safer languages, or they're due to type errors that wouldn't be present in more type-safe languages.

There are some implementation bugs that could be present in any language no matter how many safety features it has, but many security bugs aren't due to, say, an incorrectly specified algorithm, they're due to the programmer asking the computer to do something that's literally nonsense, like asking for the fourth element of a list that only has three elements, or recording that someone's age is apricot. Programming languages with powerful nonsense filters can remove a lot of those kinds of security bugs. (And powerful type systems often give programmers mechanisms to tell the compiler more about the program so that it can filter out more kinds of nonsense than it would otherwise.)

I don't understand this response. Nothing really stops that, regardless of implementation language. It's why we have an entire bodged and mostly ineffective AV industry, as well as a slightly less bodged and partially effective endpoint monitoring/detection industry.

Runtime mitigations exist to mitigate some of the latent risk associated with programming in unsafe programming languages. We use them because they're our best known approach to continuing to use those languages without letting script kiddies own us like it's 1993.

You don't seem to understand that those malicious programs have no way of running on someone's computer if they can't exploit some other program to get installed on the machine in the first place. If the system software on the target machine is written in a better language and has no exploits, then it doesn't matter what language the attacker uses for their software.

It would prevent or at least mitigate some classes of exploitation. Buffer overflows are very common attack vectors: https://cve.mitre.org/cgi-bin/cvekey.cgi?keyword=buffer+over... (386 results, 11 from prior years, so 375 from this year). 23 of those are in kernels, many leading to privilege escalation.

Yes, a better language would protect innocent programs. Take stack canaries. They are to protect stack corruption due to an application bug, e.g. unsafe input handling. Input handling is perfectly safe in many other languages, but C has a lot of footguns.

The language the innocent programs are written in can remove attack vectors that the malicious programs use; I don’t think it matters what the malicious programs are written in?

Sandboxing techniques can also be used for dealing with malicious programs, but most of those things listed are for programs to use to protect themselves, not to stop them from intentionally doing bad things.

This is why I hope unikernels & Rust will eventually replace docker & linux for high-performance, high-security production deployments.

I agree. I think of address space context switching overhead as the performance price we pay for not being able to run all our programs in a single address space, which we could safely do if we knew all the programs were emitted by a trusted compiler that disallows unsafe memory access. Imagine if system calls were just ordinary functions that can be called with no more than the normal function call overhead? What if they could even be inlined?

Obviously there's a lot of little details you'd have to work out. Like how to make such a trusted compiler in the first place, and how to sandbox unsafe code and legacy applications that were compiled by an untrusted compiler.

If this seems like it's far-fetched or too much work, consider what lengths high-performance hardware devices like HPC network interfaces go to avoid system calls at all costs, to the point where applications talk to the hardware directly. Is that really a sustainable practice long term? And how can anyone audit the security of such hardware devices?

This is what Microsoft Research's Singularity OS did, with the language being C#. Their argument was that the MMU's address space isolation was a 30% Unsafe Code Tax so even if C# was slower than C if they could get the slowdown to less than that it was still an overall performance win.

I'm pretty sure the discovery of Meltdown/Spectre and similar speculative execution attacks would completely wreck this model. The fix for those exploits has been to make the isolation barriers even stronger but if you don't have them at all you're wide open. If you had such an OS but then had to split it back into separate address spaces you've now lost the performance gains and just have a slower OS that is harder to develop drivers for.

Yeah, speculation attacks are a big problem. It seems like maybe if you're running a specific compiler you might be able to avoid speculation attacks by not emitting dangerous sequences of instructions, but I don't know what the state of the art is when it comes to Spectre mitigations and whether it's possible to have a compiler that can formally verify that a program is immune to any (known) speculation attacks.

> I think of address space context switching overhead as the performance price we pay for not being able to run all our programs in a single address space, which we could safely do if we knew all the programs were emitted by a trusted compiler that disallows unsafe memory access. Imagine if system calls were just ordinary functions that can be called with no more than the normal function call overhead?

That's pretty much how the Amiga worked, and that level of technology achievement is still unsurpassed today.

You already see this in PaaS and serverless for managed runtimes, I don't care if my Java and .NET code runs on bare metal, micro-kernel, unikernel, or whatever.

Rust will not replace anything because it’s impossible to write code in it. Everything you write is a syntax error that requires an exobrain to figure out.

What's the point of this hyperbole? We have ample evidence that thousands of engineers (and large companies) are successfully applying Rust.

GP has been posting the same thing in many threads when Rust comes up. Apparently they think it's funny, or they're really bad at writing Rust code and feel like venting.

...in a lot of the cases I ran into early on, the rust compiler literally told me what to change.

> C language toolchain

I'm sorry, but the C language toolchain is not great. The only part of the C language toolchain that is good is it's platform support. Every platform has a C compiler.

However, that's hardly something most devs will care about. At this point, we are pretty much all targeting Arm or x86 (Sorry PIC and MIPS devs). And every new language that's cropped up at a minimum supports both those platforms and the major OSes for those platforms.

So once you dismiss that point, what are you left with? Tools for catching memory leaks? Well, sorry, but pretty much every language has those tools and better. Particularly GCed languages. Profiling tools? Debug tools? To say the C debug tools are "good" is to say you've not used any other toolchain.

But let's also point out C's horror story that is "dependency management and building". How do you grab dependencies for and build javascript? npm install, npm build. How about C? Well... are you using auto tools? Cmake? Ninja? Scion? makefiles? etc... Oh, and how do you get dependencies? Well, hope you like reading readme files and decipher cryptic "symbol not found xyz" compiler errors.

Suffice it to say C's toolchain isn't anything to be proud of. It's a relic of the 70s that to this day gets in the way because the community has practically given up on the idea of fixing the glaring issues (Too hard, everything already depends on X working the way X does).

So why risk switching languages? Well, because the C toolchain is a dumpster fire and you likely aren't trying to write new code for IBM RPG servers.

> trying to write new code for IBM RPG servers

Hey, I know that! Challenge accep... Eeeeh, nevermind.

All the modern AS/400 servers can run unmodified AIX programs these days anyway.

You keep enumerating words that I will never put in my CV lest employers know that I have been paid to suffer at their contact.

> I'm sorry, but the C language toolchain is not great. The only part of the C language toolchain that is good is it's platform support. Every platform has a C compiler.

Please take a few days to review John Regehr’s excellent blog. I’m certain that you’ll find dispelling your ignorance rewarding.

> How do you grab dependencies for and build javascript? npm install, npm build.

Npm is the poster child for supply chain attacks.

> Please take a few days to review John Regehr’s excellent blog.

Haven't heard of him but will certainly give a read. Thanks for the heads up.

> Npm is the poster child for supply chain attacks.

The only reason C doesn't (often) have similar supply chain attacks is because pulling in dependencies is so hard that you aren't likely to end up with a 10k dependency project.

Hard to say that's really a plus.

Other ecosystems have their own problems and certainly aren't perfect. However, the toolchains are generally leaps and bounds ahead of what C currently has.

> you aren't likely to end up with a 10k dependency project.

> Hard to say that's really a plus.

I'll definitely call that a plus. All dependencies introduce risk. Sure, we can't avoid all dependencies, but carefully evaluating them and keeping the list small is a big win for maintainability and security.

If a random library has a 0.1% chance of having malicious code (I'd say it's more like 1%, but let's be generous), a 10K-dependency program is guaranteed to be pulling in at least something malicious.

>The only reason C doesn't (often) have similar supply chain attacks is because pulling in dependencies is so hard that you aren't likely to end up with a 10k dependency project.

Or the lack of canonical package manager means the developer has to actually spend time to inspect the quality of each dependency instead of `npm install is-odd`ing like there is no tomorrow.

It also acts as a "filter" to prevent adding complexity to the dependency tree.

They will do `apt-get install is-odd` instead, or yum, pacman, or...

> Please take a few days to review John Regehr’s excellent blog. I’m certain that you’ll find dispelling your ignorance rewarding.

Just a heads up, this reads to me as condescending.

> Please take a few days to review John Regehr’s excellent blog. I’m certain that you’ll find dispelling your ignorance rewarding.

I regularly read his blog and I found nothing to back you up. He mostly work on a generic compiler toolchain (LLVM) other than C-specific tools and his work easily translates to many other languages using LLVM. And even a rare C-specific tool like C-Reduce is not that hard to adapt for other languages; C-Reduce is actually language-agnostic and works well for many other curly-brace languages as well.

Good eye. I might add: the blind praise of garbage collection, and the "symbol not found xyz" bit raised an eyebrow for me.

You can't trust HN comments. Most of them sound good to laymen. But to a trained eye, it's a comment tree of cards.

> i might add the blind praise of garbage collection raised an eyebrow for me.

Like it or not, a problem C has is that it's pretty much impossible to determine why a piece of memory is currently being retained. Tools like valgrind and jemalloc can give you good starting points but really can't point to issues where "This memory is being retained because of this linked list over here"

Languages with GCs can, at any point, dump the heap and provide tools for analyzing that heap to point to the exact reason why a bit of memory is currently being retained.

That's not really "blind" praise of GCed languages, simply a fact of how they operate.

You’re not wrong that aliasing presents serious practical problems for building an efficient object graph. The thing is the defining characteristic of a C-like is that it has an ultra lightweight runtime[1]. The way I look at it, every C program has a bespoke garbage collector of variable quality. It’s fair to say the median is pretty dismal sure, but tooling does exist to tame that beast.

I think there’s a wide open research space to work on taming the aliasing problem. There are PhDs to be had, and maybe even Turing awards.

[1] C gets to cheat a little bit here because Unix and friends are its runtime.

> But aside from Jai, is anyone C alternative really looking to pursue having killer features? And if it doesn't have one, how does it prove the switch from C is worth it? It can't.

Zig's `zig cc` is a killer feature that doesn't even require using Zig-the-language at all. `zig cc` is an LLVM-based C compiler that gives you trivial cross-compilation for existing C codebases, adds effective caching, and can be easily installed on most operating systems with `tar -xzf`.

https://andrewkelley.me/post/zig-cc-powerful-drop-in-replace...

zig cc has enough bugs in it that I just went back to using clang on each platform.

What you're seeing is most likely not bugs (it is clang after all), but the result of the much stricter default compilation settings in "zig cc" which tends to break a lot of C code that hasn't been checked against a static analyzer or ASAN before.

No, they were bugs in zig CC. They were things like linking against the wrong frameworks on macOS.

Ah alright, some of those macOS specific problems had been fixed a little while ago when work on the macOS parts of the zld linker happened. I still get an occasional "framework headers not found" on macOS in the current Zig head version, but after the first failure I never seem to be able to reproduce it. It also never happens when running in CI, so I'm not sure if it isn't actually a problem with my Xcode setup.

Possibly true, but bugs can be fixed. The potential is still there.

If you want to compete with Rust on cross compilation usability, you will have to do better than that.

    pip install ziglang
    python -m ziglang

> First of all, pretty much all languages ever will make vacuous claims of "higher programmer productivity". The problem is that for a business this usually doesn't matter. Why? Because the actual programming is not the main time sink. In a business, what takes time is to actually figure out what the task really is. So something like a 10% or 20% "productivity boost" won't even register. A 100% increase in productivity might show, but even that isn't guaranteed.

I stopped reading here. This is like saying typing speed doesn't matter because most of your time is spent thinking about what to type. If your argument basically evaluates to saying that it doesn't matter how productive you are while doing the actual activity of your job itself... at some point it's like, I dunno, do you actually believe what you're saying?

Also, not to go off on too unrelated/unhinged a tangent, but, I've also started noticing in the last five years this widespread fear of actually sitting down and programming. There's a weird meta-culture where people want to do anything but sit down and write code, because it's hard and scary. I hear arguments made all the time for how important everything is — aligning stakeholders, scoping tasks, communication, mentoring juniors, soft skills, whatever. Everything except writing the actual code, which is somehow a mere technical triviality that anyone can do, or something.

Right? Productivity makes thinking easier. It’s part of a virtuous cycle.

I agree especially with that last point. There is something about programming that is very difficult. It gets _more_ difficult if you write complex things, but that's not the essence of it. I think it has something to do with raw concentration.

We have all of these tools that help us to write code while making few mistakes. From constrained DSLs/configuration languages, schemas, static typing to automated tests etc. But in the end we still have to exert all that energy to sit down, focus and do the coding.

Meetings, design, coordination etc. All of those things are important, and they can be exhausting in their own way. But programming is by far the most challenging, hour per hour (pound for pound). Making all of these micro decisions, think about the impact of your code, the readability, the correctness. It is _at least_ two times harder and more draining than anything else I do, even when I'm in a flow state and the actual work is fun, more fun than all of the other tasks.

And I agree: It is the most important thing. I mean all of the other stuff is basically there to support the core task that solves your problems: producing working software (except if it isn't but then you have political problems). There are seemingly infinite dimensions for improvement as well, tradeoffs such as performance, efficiency, robustness, usability, leverage, creativity...

The point is "marginal costs from switching languages may outweigh any benefits in plain programming productivity".

The article misses a really big item.

I’ve programmed in C for 20+ years and then recently switched to rust. When fellow C programmers ask me how Rust is going, I say “it’s modern” and “it’s consistent”. What I mean by that unlike C, Rust doesn’t have 50 years of baggage: string functions that handle null termination all differently, complicated implicit promotion rules, null-terminated strings that haven’t been a good idea for decades, apis that we’re completely busted by posix threads but still exist, functions that cannot be used safely that still exist, code snippet examples with horribly broken unsafe code that beginners copy-paste, dubious and bad type-based aliasing, rampant use of UB, etc etc etc

Sadly Rust is not a better C. Maybe it’ll grow into one some day. Today it’s a better C++. Sadly I’d still chose C to do C’s job.

All I want from “better C” is to fix the obviously bad problems from C’s history. Or rather, maybe that’s the committee’s job.

At any rate: it’s embarrassing that we’re still so dependent on a language that’s so broken. And it’s sad that there doesn’t seem to be a path forward.

When I discovered strsep was a thing I was genuinely a bit mad that I'd been "tricked" into using strtok all this time. Since then I never just look up a function in the standard library reference. It's a fair chance its API is far from ideal, especially in string.h

What is C’s job, when would you use it for a green field project? Not trying to start a flame war, genuinely curious as I see less and less need for C, other than maintaining all those existing code bases.

When working with the huge amounts of embedded silicon that is supported by C libraries and toolchains only.

Wow, this blog entry's reasoning is a leaky ship.

1. C language toolchain -- No specifics (save static analyzers, ... because C needs static analyzers to catch C specific bugs...). What do you actually feel you're missing? Most of these "new C" languages use the same backend as a C compiler. What can't you use?

2., 3., and 4. -- Just chicken and egg FUD.

Not to mention it's "'Better X' doesn't matter" fundamentally doesn't understand the competition.

A sample quote: "That's not just one but eight(!) killer features (re: Java). How many of those unique selling points do the C alternatives have? Less than Java did at least!" WTF?! "But Java had 8!" is laughable.

Another: "[A]ny safety checks put into the competing language will have a runtime cost, which often is unacceptable". We can also produce a car without windows, doors or a chassis that will survive a crash, and it may be faster too. We may even get lucky and live to tell the tale.

Again: "Aside from Jai, is anyone C alternative really looking to pursue having killer features?" Aside from Jai?! Jai. Really? Oh, I remember you just discounted safety and correctness as killer features. ... But Jai. And nothing against Jai. But a language which hasn't been, which may never be, released to the general public. Just, eyes wide open, wow.

I hope this is the peak of anti-Rust/pro-C silliness, but I know it's not.

I would reread the first paragraph. In context, this is clearly not an "anti-Rust/pro-C" essay. It's not a championing of C nor a resignation to its perpetual dominance.

It's a position paper meant to sketch out the negative space that a viable C replacement can't meaningfully compete in, so that its designers can focus attention on things that actually add value. If you don't acknowledge your enemy's strengths, you'll never really be able to fight them.

> It's a position paper meant to sketch out the negative space that a viable C replacement can't meaningfully compete in

Yeah, I get it and I'm not sure it does that job well enough to be taken seriously.

> What do you actually feel you're missing?

It explicitly says in the post... for example static analyzers. What is the "Frama-C" equivalent of any language that bills itself as a replacement / competitor of C?

> Most of these "new C" languages use the same backend as a C compiler.

Most of them use LLVM as far as I know, which does not target plenty of obscure platforms.

So the argument is we need the bug finders we've developed for C (because our C is full of bugs)?

> Most of them use LLVM as far as I know, which does not target plenty of obscure platforms.

Yeah, writing software in the old language is more convenient on obscure, rare, niche platforms, because it's older than dirt. More chicken and egg nonsense. It's not impossible to have LLVM add additional targets, right?

> Yeah, writing software in the old language is more convenient because it's older than dirt. More chicken and egg nonsense. It's not impossible to have LLVM add additional targets, right?

It seems like you're being needlessly hostile. There is nothing personal about this discussion, and it's not nonsense. I write embedded software that needs to run on PIC18 microcontrollers. Support for that in LLVM was dropped about 8 years ago. Do you think it's reasonable to say to someone "just add a new LLVM target, it's not literally impossible"?

> It seems like you're being needlessly hostile.

I'm profoundly disappointed. This article's reasoning is not good.

> Do you think it's reasonable to say to someone "just add a new LLVM target, it's not literally impossible"?

No, but that C is already pervasive because it's been around 50 years isn't a case "against an alternative to C" as much as it is a headwind. That's fair. C has a massive head start and no one should discount that. But I'm not exactly certain that was the argument he was making.

This conversation is about general use of C vs alternatives such as Rust.

If you can't use Rust because of your specific circumstances that's ok - use C! But don't use "my current circumstances prevent me from using Rust" as an argument as an argument against Rust in general - which is what you're doing.

> But don't use "my current circumstances prevent me from using Rust" as an argument as an argument against Rust in general - which is what you're doing.

First of all I'm just summarizing the article, I'm in no way arguing against people using Rust. The article itself argues against C alternatives, and explicitly describes Rust as a C++ alternative. So the article is not even arguing against Rust! If we can't agree on such a basic reading of the article, it just seems like we're going to pointlessly talk past each other in the comments.

GCC 13 will support Rust.

I think what he's saying for (1) is that the toolchain is good compared to other languages (lots of existing tools.)

I don't know why you say 2-4 is FUD.

This article is just apologism for the status quo. Nothing new here.

> Memory allocation, array and string handling are often tricky, but with the right libraries and a sound memory strategy, it can be minimized.

Such a handwavy deflection. Parsing strings in C is a joke and also a minefield bursting with vulnerabilities. And if you use null terminated strings (which you are, let’s face it) it’s slow.

Depends on the functions used. memcpy is better than strcpy

I’m sure you meant something more along the lines of “strncpy is better than strcpy”.

Nah, strncpy is pretty bad and doesn't do something anyone really expects to want.

(strncpy doesn't 0-terminate when it hit max length, nor stop when it sees its first 0 in the source buffer; it acts almost entirely like memcpy except that it writes zeroes to the full length of the destination upon seeing a zero in the source. This is allegedly useful for populating fixed-sized buffers in historical Unix, and can plausibly be useful for writing out tar headers, but in practice very little code in the wild does anything observably different if you #define strncpy memcpy.)

memcpy is better than strncpy and strcpy ;)

Ok, but there exists a sulotion for this. Using C++ with STL. Why we exactly need a new language?

There is A solution. Is it the best solution? Are there other solutions which may offer better performance or safety.

I don't we should stop trying to improve just because an existing solution exists, especially in the space of systems programming which has barely moved in decades

The only way to real achieve progress in systems programming is through generation change, one person at a time, the problem isn't technical.

Typical statement coming from people lacking experience in this field (no offense meant).

The truth is, parsing strings in C is as easy as in any other language. You have a string array and a length field, and you scrub through it from left to right with a cursor. Done.

What you do not get in C is creating lots of string objects willy-nilly, and concatenating them with a plus sign, like you do in scripting languages.

Now Object Pascal and Modula-2 are scripting languages....

Well no, and if we're being pedantic I didn't imply that.

> ...concatenating them with a plus sign, like you do in scripting languages.

If anything, "replacing C" is a good excuse to stimulate creativity. The scope is known, the land well trodden. I like seeing new things in programming languages and OS design. I like nerds having fun and giving me interesting stuff to wrap my three neurons around in the process. That's why I love computing to begin with. Not increasing productivity, not disrupting who knows what. Just cool, smart shit.

(5) is pretty irrelevant, ABI isn't language.

You can strap basically any language on top of the C ABI. Many contender languages either use it natively or offer low-cost/no-cost C ABI bindings, specifically as its the lingua franca ABI.

You can check (5) off your list if your choice of replacement language has easy C ABI interop.

[edit]

(1) isn't particularly relevant either. These days the tooling for detecting memory leaks, data races, bugs, etc, operates at the debug info (DAWRF/dSYM) level. It doesn't matter what compiler or input language generated the debugging information, the tools should work just fine. You can run valgrind on your Rust binary for instance. [1] Is there any such tooling that specifically depends on the input source being C? I guess some static analysis tools?

[1] https://nnethercote.github.io/2022/01/05/rust-and-valgrind.h...

Yeah, I'm also confused by this.

A platform will define an ABI which, as I understand it, explains how parameters are passed to and returned from a "function", what types are supported and how they're encoded, at the processor register and stack level.

C, being the "portable assembler" [1] that it is, maps its types and functions pretty easily to that platform ABI, but I've been seeing a fair amount of confusion about that ABI being about C rather than any language that can link through that platform ABI. Couldn't it just as easily be called, I dunno, the Ada ABI?

[1] not really, except for the sake of this argument

Why is this down-voted? C does not have a standard ABI, this guy here is right?

I just want:

C with:

- bounds checking

- use after free checks

- modules

- sane metaprograming/template

- tagged union built in

without:

- macro

- pre declaration

- split header/source

That's it, no borrow checker, no weird syntax, no nothing

So far D is the answser for me, but i'm worried about its future, will they keep improve the language in that direction? or will they continue with their high level stuff

Zig/Jai/Odin so far are looking interesting, the problem is they all depart from the C syntax, wich is annoying, the only syntax i'd like to change is the primitive type, i32/f32/size/usize, and that's it, no need more

I have a similar list, but I think part the difficulty with C replacements is that the sort of people who want to go out and write a new language generally do so because they have something more radical in mind.

I have pondered forking tinycc and using that as a startpoint for a more conservative set of changes like you describe. It would limit the amount of work to be done and lead to something useable quite quickly.

Have you looked at C3 (the language OP is making)? I think it hits a lot of your boxes.

I looked at it; and i'm not really a fan of the syntax, `fn` feels out of place

Also no language server is a hard pass, tooling is important

> No experienced developers

Whenever a CVE is discussed, even if the code was from Google, Microsoft, or Apple, commenters immediately point out that these companies must have hired some "bad programmers", because a "good programmer" wouldn't have made such mistake. Given that even top software companies continue producing CVEs on a regular basis, there is a shortage of people capable of writing C.

A lot of programming skills are transferable (algorithms, architecture, system APIs, debugging), so you don't really need to hire "language X programmer".

The problem with making a new language is that it seems like people do it because they want to make a new language. Don't make a new language just to make a new language. Do it because you are solving a problem by doing it, and if you are solving a problem, you already win regardless of how many people will use the language. Have the problem before you make the solution.

> It turns out that people don't always agree on what the pain points with C is.

People have different problems so they should probably use different languages. Don't even try to make a language that will solve everyone's problems.

> Also, while the company has experience recruiting for C developers, it doesn't know how to recruit for this new language.

If the language is so hard that people have to have special knowledge to use it and can't learn it quickly then it's probably not a good language.

> For a business it's whether despite the downsides the language can help the bottom line: "Is this valuable enough to outweigh the downsides?"

Exactly, it's solving real problems that matters.

> The "build it and they will come" idea is tempting to believe in

A better idea is "build it and we will come because we need it. Then maybe other people will come if they need it too."

> any safety checks put into the competing language will have a runtime cost, which often is unacceptable

This is completely wrong. The best counterexample is probably ATS http://www.ats-lang.org which is compatible with C, yet also features dependent types (allowing us to prove arbitrary statements about our programs, and check them at compile time) and linear type (allowing us to precisely track resource usage; similar to Rust)

A good example is http://ats-lang.sourceforge.net/DOCUMENT/ATS2CAIRO/HTML/c36.... which uses the Cairo graphics library, and ends with the following:

> It may seem that using cairo functions in ATS is nearly identical to using them in C (modulo syntatical difference). However, what happens at the level of typechecking in ATS is far more sophisticated than in C. In particular, linear types are assigned to cairo objects (such as contexts, surfaces, patterns, font faces, etc.) in ATS to allow them to be tracked statically, that is, at compile-time, preventing potential mismanagement of such objects. For instance, if the following line:

      val () = cairo_surface_destroy (sf) // a type error if omitted

> is removed from the program in tutprog_hello.dats, then a type-error message is issued at compile-time to indicate that the resource sf is not properly freed. A message as such can be of great value in practice for correcting potential memory leaks that may otherwise readily go unnoticed. ATS is a programming language that distinguishes itself in its practical and effective support for precise resource management.

I still occasionally write C (and C++), but have come to appreciate the sparseness of Go (although not the intricate abstractions of Rust) and keep wondering if Zig (which does cross-compilation almost as effortlessly as Go) shouldn’t be more popular.

In the end, though, you can accomplish the same (from an outcome perspective, not necessarily from a satisfactory one) in all of them, so I try not to be overly biased, although convenience vs correctness seem to be the main issues at play here.

(Edit: why the downvotes? Is Zig out of favor? Is it because this may read as a slight on Rust?)

I like Zig a lot and think it hits precisely the right niche for a C replacement. Rust is a great language, but it’s definitely not C like when it comes to the overall UI and philosophy.

I’d like to use Zig a bit more, but the standard library is still going through some significant changes (it seems) which makes it slightly harder to commit to at the moment.

In general I wonder if some of these languages would actually benefit from Jai’s approach of deferring availability until the language is ready. While getting early adopters is great for getting feedback early, it can also disuade potential users that try the language before its ready and decide to leave it alone as a result.

> This leads to a strategy of only having checks in "safe" mode. Where the "fast" mode is just as "unsafe" as C.

This is hard to describe without getting deep into Rust details, but I think the ability to encapsulate unsafe code in safe APIs is just as important as what pure safe code can do. (Of course those two things end up being intimately related.)

> And worse, what if the language omits crucial features that are present in C? Features that C advanced programmers rely on?

Seems like this might be a good place to ask: which of the C replacement languages mentioned (or not mentioned) support inline assembly? Do any of them do it better than GCC/Clang? For me, I feel like the ability to drop to assembly in a pinch is a necessary feature. I'm sure I could technically get by linking an external file, but I'd be more interested in a language with inline ASM support than one that doesn't allow it. Thanks!

Pretty much all of them at this point.

Rust: https://doc.rust-lang.org/nightly/reference/inline-assembly....

D: https://dlang.org/spec/iasm.html

Ada: https://docs.adacore.com/gnat_ugn-docs/html/gnat_ugn/gnat_ug...

Cobol: https://www.ibm.com/docs/en/cobol-zos/6.2?topic=appendixes-a...

Pascal: https://wiki.freepascal.org/Asm

Python: https://pypi.org/project/il/

You can almost name a language and find a route to inline assembly :D

Do you happen to know if is also true for the C replacement languages mentioned in the article: C3, Zig, Odin, Jai and eC?

C3: Currently being looked into https://github.com/c3lang/c3c#current-status

Zig: https://ziglang.org/documentation/master/#Assembly

Odin: Looks like it's a dead language, but it was on the docket. https://github.com/odin-lang/Odin/blob/master/misc/roadmap.m...

Jai: https://github.com/Jai-Community/Jai-Community-Library/wiki/...

eC: No clue. Not a great name for googling.

Odin is pretty much alive to JangaFX customers,

https://jangafx.com/

Yup, I was just wrong on this. Their git project just got a commit yesterday. IDK what I was looking at that made me think it was defunct.

Zig: https://ziglang.org/documentation/master/#Assembly

Zig seems to as well, though I haven't looked closely enough at any of them to see if the support goes beyond x86/x64 (or even ARM).

Finally an article I can agree with.

C is great mostly because it's easy to learn, and it has many other advantages.

I dislike all those new languages because they have too many features, and they're non trivial to learn.

I've looked at some rust codebases and I don't think there will be a lot of people who will want to maintain them. Rust is difficult, even though it's an awesome language.

A language that could compete with C would:

* Be fast to compile

* Have some quality of life things you can find in the C++ STL: map, string, etc

* Look a lot like C, meaning it would not try to innovate too much.

To summarize, it would be a lighter version of C++.

> easy to learn

What exactly do you mean? Learn the basics? Become productive? Memorize the entire specification? I'd argue C isn't significantly faster at any of these than most other languages.

It can be learned quickly, it's very readable, and taught to students without too much pain, and beginners can become productive with it.

The "difficulty bar" must be low. This is why english is a popular language.

> I'd argue C isn't significantly faster at any of these than most other languages.

Yes it is. It has a lot of flaws, but it's popular for all those other reasons.

Go?

Actually, yes. Go does a lot of the things everyone seems to want in these C replacements, the problem is that it has GC which, to a lot of people, automatically disqualifies it as an option for writing certain kinds of software you can write with manual memory management. A lot of people also tend to think they need manual memory management for $reasons when GC is actually fine.

without a GC

Zig? Odin?

What a defeatist take :/

A "better C" language doesn't have to "win the popularity race" to be useful, it just needs to (a) be easy to learn coming from C, (b) integrate well into the existing C/C++/ObjC ecosystem, and (c) easy to install and across all supported platforms.

A "better C" cannot and should not replace C for all use cases, but it should at least augment C by fixing its design wart.

The actually depressing thing is the "winner takes all" mentality though that seems to be prelevant in the programming world.

This person is actively trying to make a "better C" so it's no surprise his arguments against his own project are lukewarm.

"It is difficult to get a man to understand something, when his salary depends on his not understanding it."

Why do you put that quote? I'm not sure anybody "tried to get him understand it". I suppose the post comes mostly because of what the author learned or concluded from his own project. There might be other motivations. I think the post has some genuinely good points that need to get more mindshare.

There is also a more general "case against the self-indulgent narcissism in creating a new language". There are an insane number of languages with the longevity of a mushroom mentioned on HN.

Heh. This article was posted shortly after we instituted a ban [0] on language talks at Handmade Seattle 2022.

There's an oversupply of enthusiastic engineers trying to provide C alternatives and it's creating massive bike-shedding and yak-shaving problems.

Language designers are encouraged to give a presentation if it's about building quality software using a language. Otherwise, no.

[0] https://vimeo.com/handmadeseattle/languages-are-banned

An oft-overlooked pain point is that when interacting with C interfaces, it's not good enough to implement a C ABI FFI. There's a world of software out there where the supporrted interface is a C API, at the source level. Compatibility across different versions, or across different implementations, is assured only to programs which interact with the interfaces in ways that correspond particular C source patterns.

The biggest culprit I've met on this front is the POSIX stat family of functions.

Like any other technology that wishes to supplant the status-quo, a new programming language cannot be just 'a little bit better'. It has to be substantially better in order to overcome the inertia that the entrenched technology has established.

This is true for both software and hardware. There have been plenty of leaps and bounds such as SSD speeds compared to HDD speeds; but even there, SSDs will not supplant HDDs until there is price parity between the two technologies. Until then, data will be stored on SSD when fast access is critical and on HDD when there is too much of it to justify the cost.

Getting a company to switch operating systems, database vendors, or cloud providers is likewise an uphill battle. They don't call them 'walled gardens' for nothing.

I am building a new data management system that I think is much better than file systems at storing and managing unstructured data and better than RDBMS at managing structured data; but I wouldn't have even dreamed of starting the project if I wasn't confident that it was at least 2x better at a minimum.

Jai is not real. It should not be mentioned even on the same sentence as Zig.

It's real but without public access it's largely useless. I understand why Jonathan might want to avoid dealing with a peanut gallery trying to argue with him about what his language should be like, and that's fine, his language his rules and all of that. But until this situation changes it's impossible for anyone to really care about Jai.

You can't say Jai is not real, because there are beta testers using it. However, there are other languages that have taken and implemented various concepts that Jai had to offer. By the time Jai comes out, and is open to the public, there might not be anything compelling about it anymore.

The simplest solution is a preprocessor. The original founders of C made one, probably because they needed it. But adding another lightweight one is simple. C++ has bloat. Compilation time can be slow.

I realize that this is radical, but if people could work with me on this, this could solve all your problems, no really, it could solve all your problems. And if it can't solve all your problems yet, then modify it to

https://cboard.cprogramming.com/c-programming/181160-hi-i-ha...

There is no need for another language. Just use this

Gregory

Is your thing on GitHub?

https://github.com/gregoryc/EmeraldC/

I just made a github.

You could write a 100-line python script that compiles from your custom language to C code and it's still be better than C. There's lots of ways to re-use existing tooling and the bar really isn't that high for better languages

Why do you think more of these projects don't take this approach? I would be 10x more likely to try one in a professional setting if I knew the portion I was taking chance on was source translation layer, and I could rely on the well tested and supported C infrastructure.

The Nim language compiles Nim source code into C as an intermediate step.

I was considering this for my programming language but I will most likely not use it because it makes optimisations and garbage collection harder, and the C would not be human readable (just look at the output of the chicken compiler for an example) so there's not a huge benefit. C-- could be an option but it isn't very active or well used.

In terms of relying on tested and supported infrastructure lots of these projects use llvm.

It's frankly harder and/or less fun to generate C and integrate with whatever locally support C infrastructure you have than it is to BYO.

> Looking at C++ alternatives there are languages like D, Rust, Nim, Crystal, Beef, Carbon and others.

Why do people keep lumping Rust into the category of a "C++ alternative". Rust is a C alternative and can be used places that C can but C++ cannot. i.e the linux kernal and baremetal silicon (C++ can be used in the embedded but I've never heard of it being used without modification for that environment).

> My language (C3) is fairly recent, there are others: Zig, Odin, Jai and older languages like eC.

Further he seems to claim that Jai is a C alternative when it's explicitly described by it's creator as a C++ alternative.

I came to the same conclusion last year: C (client) and Java (server) are the final languages.

To fragment the industry with more languages is a hidden cost that wont end well for anyone.

KWh is approaching $1, prepare by choosing the proper languages.

"Any C alternative will be expected to be on par with C in performance. The problem is that C have practically no checks, so any safety checks put into the competing language will have a runtime cost, which often is unacceptable. This leads to a strategy of only having checks in "safe" mode. Where the "fast" mode is just as "unsafe" as C."

This one is the main point. Highest possible speed while consuming as little resource as possible. Put in checks, and most of those checks will run redundantly compromising performance.

While true, its lack of restrictions is also a hinderance in terms of performance. As pointers can overlap by default (unless you throw a __restrict on them), a whole class of optimizations cannot be performed. That's why Fortran and Rust can in theory beat C from a performance perspective even though they do more checking as a rule. That's just one example, but it's a pretty big one.

"We have seriously regressed, since C developed. C has destroyed our ability to advance the state of the art in automatic optimization, automatic parallelization, automatic mapping of a high-level language to the machine."

-- Fran Allen, from Coders at Work

> any safety checks put into the competing language will have a runtime cost

How about all guarantees static analysis brings to data races, memory ownership, etc? No runtime cost. C tools do not bring this.

Yep. A few years ago I implemented a skip list based rope library in C[1], and after learning rust I eventually ported it over[2].

The rust implementation was much less code than the C version. It generated a bigger assembly but it ran 20% faster or so. (I don't know why it ran faster than the C version - this was before the noalias analysis was turned on in the compiler).

Its now about 3x faster than C, thanks to some use of clever layered data structures. I could implement those optimizations in C, but I find rust easier to work with.

C has advantages, but performance is a bad reason to choose C over rust. In my experience, the runtime bounds checks it adds are remarkably cheap from a performance perspective. And its more than offset by the extra optimizations the rust compiler can do thanks to the extra knowledge the compiler has about your program. If my experience is anything to go by, naively porting C programs to rust would result in faster code a lot of the time.

And I find it easier to optimize rust code compared to C code, thanks to generics and the (excellent) crates ecosystem. If I was optimizing for runtime speed, I'd pick rust over C every time.

[1] https://github.com/josephg/librope

[2] https://github.com/josephg/jumprope-rs

The valgrind suite is a very effective answer to these needs and it's used by most C programmers. Helgrind for data races. Memcheck for memory ownership. Not to mention sophisticated tools for cache profiling, call graph analysis and heap profiling. No, they're not static tools but being dynamic has a host of advantages too.

Valgrind's memcheck is great for exercising known paths in a program, and determining whether they're likely to contain memory errors. It doesn't help you discover those paths, which is what attackers are doing when they research and then exploit your program.

We've known for decades that compiler mitigations, fuzz testing, and dynamic instrumentation are excellent and necessary components of writing more secure C. But they don't secure C programs, because C itself is fundamentally unsafe.

> No, they're not static tools but being dynamic has a host of advantages too.

Advantages that any compiled binary can enjoy.

I think the main point is that a good alternative C doesn't compete directly with C's strongest points.

Here, you can of course make a more safe language without runtime penalty just by supporting richer build-time/static checks, but the modest and meaningful gains you make within that constraint will never be enough to displace C.

You need to be providing something meaningfully more in areas where C hasn't already squeezed out most gains. The OP tastefully avoided making this article directly about C3, but if you go and look at the summary of features, you can see how this philosophy connects with what they're working to build.

C is used because it has to be used, i.e. low level environments etc. not because 'performance' most of the time.

If you could pay in performance a little bit, but have a clean, portable, beautiful debugger experience, nice libraries like 'Java' with a 'stable ABI' ... nobody would use C. We'd all use 'that new thing'. And we would just buy slightly faster hardware for the IoT whatever.

We use C++ smart pointers all over the place, and they come with a 'cost'. We mostly 'happily pay the price' because it's worth it most scenarios.

Engineers obsession with performance is a bit of a curse. Rarely do we really need that much. Kernel code, sure, IoT and remotely controlled toy cars, mostly not.

Honestly, I hate includes and preprocessor directives so much that any "better C" variant which replaces them with more modern solutions would probably be enough in my book - with the caveat being that I also like having an IDE very much and wouldn't want to revert to coding in something like emacs with plugins. BTW, C++20 has a good replacement for includes (modules), but IDE support is not there yet and thus it's not usable for me.

I wholeheartedly agree, the speed of processing includes annoys me most about C.

The sad part is that the mainstream alternatives that I have on the radar are _even slower_ to build.

And usage of the preprocessor in APIs is allowing for good things, at least for a compiled systems language like C: Macros allow to get rid of boilerplate on the syntax level and if-defs allow for compatibility. Both points' significance is reduced a long way in other languages, because they have stronger facilities for abstraction, BUT this abstraction comes at a non-significant price and cannot replace all significant uses of the preprocessor.

So I'm still stuck with C. The solution is designing header files carefully, resulting in build times that are lightning fast, at least in relation to most other mainstream languages. And there is always the option to go for "unity" builds.

What bother me most is not even the speed of compilation (although it's a major pain too), but the fact that I need to create a header file for every .c file I want to have used from outside it's compilation unit, and add method declarations in it. In 2022 it's SO BACKWARDS it hurts.

I almost never found that painful. Maybe a couple things that help: 1) Improve modularization. APIs should be small (as a rule). 2) Don't write a new file for each "class". It's not a problem to have files with more than 1 KLOC. But it is problematic to have a lot of really small files, as is (I think) customary with say Java.

I come from the Java world, so I'm very used to every semantically separate bit of code sitting in its own file. I think it's superior to having multi-KLOC source files (as is common in C/C++ world, because it minimizes the header writing and maintenance busywork), because then you can browse code via a file browser, which is typically a panel in your IDE. The C alternative to that would be having the same file opened in multiple tabs, and heavy usage of (IDE) bookmarks? I don't find that as convenient.

As to improving modularization, in order to fix compilation speed issues I was trying out the "program as a single compilation unit" approach advocated by handmade people, and I found that I had to spent a good amount of time in moving code between modules, and their subsequent headers. So annoying... Perhaps it wouldn't be so infuriating if I didn't spend the past 15 years coding in languages where this is a problem solved automatically and invisibly by the compiler.

> any safety checks put into the competing language will have a runtime cost, which often is unacceptable.

Suppose you have bounds checks on array accesses, but your program is 100% correct and the panic case is never hit. Doesn't the branch predictor essentially make the bounds check free? Or very low cost at least?

If what comes after the check is a write or any other destructive operation, then the check is not so easy to mask away. I don't have actual numbers, but my gut is a strong 'no'.

That said, a lot of pedestrian code that isn't running particularly hot would probably be better off including runtime checks by default.

It is no different from QuakeC, GOAL, and a myriad of other programming languages that game studios never had as milestone to release to the wieder public, yet they were quite usefull to their owners.

The momentum of C seems so vast that replacing it is like getting humans to switch to a base-12 number system

That's a weird analogy considering humans used base-12 (and base-20 and base-60) number systems for most of history.

Really? Wikipedia says it is historically uncommon

EDIT: What language uses base-60??

Babylonians used it. That's the origin of our time counting system of hours, minutes and seconds and our angle measuring system of degrees, minutes and seconds. Anyway, if we someday change our number system away from 10, I hope the change will be towards a smaller base, like 8 or 6, because when one is a child, it sucks to learn the times tables. Base 12 would increase the number of entries by 44%. Base 8 would decrease them by 46%, and base 6 would have only 6*6=36 very easy to remember entries.

That's not a pure number system though, because it didn't have a separate character for each digit

Minutes in an hour, seconds in a minute, and something about degrees.

Rust is not only a C++ alternative, it's a C alternative too. You don't have to have methods in your structs or use generics, traits, closures, interators etc.

You won't be able to use the standard library, but it's doable.

C has already a syntax way too rich. ISO should fix and remove from the syntax. Yes, break backward compatibility starting from "c24".

Let's go in a fantasy world where we have a B+ language which should have been C: remove typedef/_generic/typeof/restrict/enum/etc, well all those horrible things. Only one loop statement, loop {}, no switch ofc. Only sized/signed core types with properly sized/signed literals: ub/sb or u8/s8 with integer literals like 123ub/123sb, sw/sd..., udw/sdw... , uqw/sqw..., fdw/1.23fdw (f32), fqw/1.23fqw (f64) (for some hardware ymm/xmm/zmm with fancy literals, sorry I am not aware of the fine details of the C/CPP lexer). Ofc, no integer promotion and no implicit cast (void* would still not require any cast). Compile-time and runtime cast (not like with that horrible c++ syntax), same thing for consts (even though nowadays it is the the compiler which detects real compile-time constants), yes you may have to "static const" your literals. Oh, and for the C preprocessor, do fix that variadic-args function-like macro for good (I wonder if gcc the way is not better than c++ ISO way).

Without all that, it should be "reasonable" for a small team of average-skilled or an individual to write a naive compiler until they don't forget they can hit a goto at any line. And they are those who say "but feature A is cheap to implement"... 1 million cheap features later... well, you get the picture.

The really hard part for ISO was to keep C actually "cheap" to implement, this is where they are failing hard.

As for the gcc extension based C dialect for kernel developement, no salvation here, C is really too alien (or even B+). Namely the little ISA abstraction is not "enough", so either the kernel goes full "slow" with plain C and assembly (_not_ inline assembly ofc), or full assembly on all of its fast-paths which would require proper binary defined specs (I am currently researching how bad this could be).

I do believe the open source communities have the strength to maintain parallel assembly _significant_ code bases (which do the same thing), big monolithic corpos... not so sure...

> 3. Programmer productivity

> [...] Why? Because the actual programming is not the main time sink. In a business, what takes time is to actually figure out what the task really is. So something like a 10% or 20% "productivity boost" won't even register.

This reminds me of Amdahl's Law [1], about how a performance optimization (parallelism) has small returns because the performance improvement only affects a small portion of the overall process.

[1] https://en.wikipedia.org/wiki/Amdahl's_law

Counterpoint is that a language that allows rapid iteration/prototyping (and thus, increases "programmer productivity") is useful precisely because it helps figuring out what the task really is.

100%.

This annoys me so much in Java debates. The argument goes something like "It doesn't matter that Java is verbose and requires hundreds of files because the IDE generates them for you". And this is true... when you write the code. But the real friction is when you must read or change the code in response to new requirements, which is exactly what we as engineers spend the bulk of our time doing.

Ive been thinking about this topic for years and OP was finally able to put it in writing. I dont think anyone would claim english is the best, but it just so happens that the most powerful empire on earth (until recently?) uses it as its primary language. Its good enough at what it does, and has a lot of history backing it up

With this knowledge I choose to align myself with languages and platforms that respect C and complement it (Lua, Objective-C, etc) instead of languages and platforms that try to supplant it (c++, java, swift, rust, etc)

I concur with you - everything I want to do in Rust or Golang, I can do with C+Lua just as easily and with far less tooling hassle. I'm yet to find a good reason to switch from just using C+Lua for everything - but the Rust guys sure are trying. Maybe Rust+Lua makes the most sense, though ..

I will never use any other language for server apps than C. C is a language for getting things done. It's finished and final. It will not change. There will be no new surprises, no new operators, and no unpublished packages. There are just source files. Write, compile, ship, repeat. You used it in 1990 to ship, you used it in 2000 to ship, you used it in 2010 to ship, you used in 2020 to ship, and you will use it in 2030 to ship. Learn once, ship forever.

Aways nice to see good reasoning. Though I'd veer away from it if the task is too large and the program needs to be updated frequently by multiple people.

Also, you got any recommendations for a c library for server apps?

Wait… really? Not even considering C++ Boost, Go, or Rust?

No. I like when nothing changes. C is perfect. All my code from 2000 still works.

This. I know I'll never get SHOUTY DEPRECATION WARNINGS with C.

Hypothetical: I need a small, fast systems language to use for writing small, portable programs.^1 Using the language the system is written in would be ideal.

1. In other words, the size of a "hello world" program matters, as does the storage space and time I need for the compiler toolchain and libraries.

What language should I use. I am not interested in writing large, complex programs.

Has anyone checked out this?

You can use C, but with better syntax.

https://cboard.cprogramming.com/c-programming/181160-hi-i-ha...

I'm looking for contributors. :)

Just some advice, I think you'll struggle to gain traction without putting the source onto Github, as that is generally how people contribute :)

I just made a github.

https://github.com/gregoryc/EmeraldC/

I agree with most things in this article really, but then again, handmade.network has some quality in it overall.

Is there a "standard C" llvm target? That would maybe chip away at the "if you are working on some obscure embedded platform, C might be the only supported high level language" advantage.

Answering my own question, but it looks like llvm-cbe is it, as far as I can tell.

(4) is extremely important. Labor pool is smaller and there won't be a wealth of soft documentation and people that can provide immediate support for common questions.

It is, on the other hand it's a lot of easier to write correct code in, well, almost every language other than C.

To give a concrete example, I was able to write production-ready code with 2 weeks of Rust experience and nobody to provide support for questions (beyond reading StackOverflow), and that code was more reliable and less buggy then the Python/JavaScript code our company was otherwise writing, even though we were collectively much more experienced in those stacks. I doubt we'd have gotten that with C, even though we had a bunch of developers who were experienced with C.

And I wrote a firmware in C for 4-quadrant torque control in about the same time. I do not have any formal proof on how buggy/or not it is but the thing runs properly for 10 years already. If that is not "production code" I do not know what is. I am not inexperienced programmer. Rather quite opposite. But this was after I did not touch C and any microcontrollers for like 10 years.

I feel like C is somewhat manageable on microcontrollers. Programs tend to be smaller (micro!), you can often do static allocation which mitigates a lot of the memory management issues, and you're also often dealing with simple types and bit manipulation which is where C shines.

In larger programs on full-fat operating systems, programs tend to be much larger (esp. if you include libraries - dealing with libraries being one of the most complex things in C), and you have to deal with sophisticated allocation patterns, and complex types (webs of pointers), abstractions and business logic which is where C basically leaves you on your own and provides very little in terms of structure or guard rails.

After many years in this field, I am convinced we don't need a better "C" at all. We need better "C education".

I tend to disagree at this notion. I've used C for almost a decade now and I've seen many developers who were far more talented than me make similar mistakes all the time.

Some problems require solutions which are unable to be solved in C in a safe way.

the only better c that will count is if c add some features that are critical like a standard library that is more powerful i know c++ 17 have some improvements but need alternatives for most things is not good for the ecosystem, and the old only if you want, is 2022 most embedded system use this tools, we need c to have a reasonable standard library, the whole world depend on it.

Check out https://cboard.cprogramming.com/c-programming/181160-hi-i-ha...

I posted this less than 24 hours ago.

This might be what you want, and there's no overhead at all.

It might actually speed up your code. 15x faster regex

I just made this. I'm looking for contributors :)

Next: the case against Cloudflare-gated websites.

you don't enjoy having half of the Internet MitM'd?

More than you can possibly imagine.

> The status of C as the lingua franca of today's computing makes it worthwhile to write tools for it, so there are many tools being written.

The vast, vast majority of programmers aren’t doing their computing by writing C anymore. It’s been the better part of 20 years since you could plausibly call it the Lingua Franca of computing.

Lingua Franca in this case typically refers to the fact that you write a library with at least C ABI support if you want it to have the largest user base. Almost all languages' FFIs support some C ABI for their libraries.

Well argued. Lots of things factored in.

Written by someone who does not know about / does not understand compilietime safety guarantees

Here comes the Rust brigade…

Reply on news.ycombinator.com