Hacker News Reader

Comment by throwaway2037 | Hacker News Reader

throwaway2037

The article starts with:

    > This criticism is aimed at RAII — the language-defining feature of C++, which was wholesale imported to Rust as well.

    > because allocating resources becomes easy, RAII encourages a sloppy attitude to resources

Then lists 4x bullet points about why it is bad.

I never once heard this criticism for RAII in C++. Am I missing something? Coming from C to C++, RAII was a godsend for me regarding resource clean-up.

View on HN

It's quite common knowledge in the gamedev world (at least the part that writes C++ code, the other parts that uses C# or JS has roughly the same problem, just caused by unpredictable GC behaviour).

Towards the end of the 90s, game development transitioned from C to C++, and was also infected by the almighty OOP brain virus. The result was often code which allocated each tiny object on the heap and managed object lifetime through smart pointers (often shared pointers for everything). Then deep into development after a million lines of code there's 'suddenly' thousands of tiny alloc/frees per frame and all over the codebase but hidden from view because the free happens implicitly via RAII - and I have to admit in shame that I was actively contributing to this code style in my youth before the quite obvious realization (supported by hard profiling data) that automatic memory management isn't free ;)

Also the infamous '25k allocs per keystroke' in Chrome because of sloppy std::string usage:

https://groups.google.com/a/chromium.org/g/chromium-dev/c/EU...

Apart from performance, the other problem is debuggability. If you have a lot of 'memory allocation noise' to sift though in the memory debugger, it's hard to find the one allocation that causes problems.

> It's quite common knowledge in the gamedev world (at least the part that writes C++ code, the other parts that uses C# or JS has roughly the same problem, just caused by unpredictable GC behaviour).

Isn't it pretty common in gamedev to use bump allocators in a pool purely because of this. Actually isn't that pretty common in a lot of performance critical code because it is significantly more efficient?

I feel like RAII doesn't cause this, RAII solves resource leaks but if you turn off you brain you are still going to have the same issue in C. I mean how common is it in C to have a END: or FREE: or CLEANUP: label with a goto in a code path. That also is an allocation and a free in a scope just like you would have in C++...

> Isn't it pretty common in gamedev to use bump allocators

Yes, but such allocators mostly only make sense when you can simply reset the entire allocator without having to call destruction or housekeeping code for each item, e.g RAII cleanup wouldn't help much for individual allocated items, at most to discard the entire allocator. But once you only have a few allocators instead of thousands of individual items to track, RAII isn't all that useful either since keeping track of a handful things is also trivial with manual memory management.

I think it's mostly about RAII being so convenient that you stop thinking about memory management cost, garbage collectors have that exact same problem, they give you the illusion of a perfect memory system which you don't need to worry about. And then the cost slowly gets bigger and bigger until it can't be ignored anymore (e.g. I bet nobody on the Chrome team explicitly wanted a keystroke to make 25000 memory allocations, it just slowly grew under the hood unnoticed until somebody cared to look).

Many codebases might never get to the point were automatic memory management becomes a problem, but when it becomes a problem then it's often too late to fix because that problem is smeared over the entire codebade.

Nobody says that an object with memory allocated using a bump allocator can't use RAII for other uses. The destructor could even call delete on the pointer to the memory in the bump allocator and the bump allocator can happily implement delete as a noop or a tombstone operation of some sort. Now this does mean when you design your objects (see some talks on objects vs values for nuance there, oversimplified TLDR object exist in a place in memory but values are things that can be passed around) you need to make them aware of the allocator that allocated them. Values are generally stored in containers which the STL ones are allocator aware and you should be making your custom ones allocator aware too.

Generally the expensive part of memory allocation isn't calling malloc, it's the underlying operations, when the underlying operation is +sizeof(T) and free is a noop you can happily keep using RAII and not care.

CLARIFICATION: I'm saying having a T* member is an antipattern, make it std::unique_ptr with a custom deleter or it should be in a container, struct of arrays...

I don't think that self-destructing individual objects via 'smart-references' are a good idea, especially when you need to carry an allocator around (either in the object or in the smart-reference).

Destruction should be explicitly performed at a specific time and place, and not happen decentralized in some random location of the code when a reference goes out of scope (I realize that this dismisses the whole idea of RAII and destructors and garbage collection in general, but ¯\_(ツ)_/¯).

I needed to Google about "bump allocators". I found this: https://stackoverflow.com/questions/62076569/is-this-impleme...

    > It's a really naive allocator that can be very fast due to the tiny amount of housekeeping involved, but you have to live with a pretty heavy constraint: there's no "free" operation for individual request - you just destroy the whole thing.

To me, this sounds like an arena allocator. I remember seeing them in Apache SVN source code a squillion years ago. What's the big deal? For RAII, we can just do placement new in the ctor, and "arena free/delete" in dtor -- which might do nothing in practice.

Also, was the original article focused on gamedev? I don't see any mention of it. For us normie programmers working on CRUD apps, we don't need bump/arena allocators. Ctor->new, RAII/Dtor->delete is just fine for us.

> Also, was the original article focused on gamedev? I don't see any mention of it. For us normie programmers working on CRUD apps, we don't need bump/arena allocators. Ctor->new, RAII/Dtor->delete is just fine for us.

Yes arena allocators is the word I couldn't remember, a bump allocator is a way to implement an arena allocator and probably the most popular.

Also I 100%, in general you shouldn't need an arena allocator, but if you are building a library you should be aware of allocators and make however you are allocating memory configurable so that if you do need to some day change it it doesn't involve a complete rewrite.

And yet CryEngine, Unreal, and many others do just fine.

Not sure if you're joking or not, actually. Both of those are pretty well known for performance problems - especially in games from the mentioned era. When Crysis came out, it was notoriously laggy and the lag never really went away. Unreal lag is also super typical, I can sometimes guess if the game is using Unreal just based on that. This might have different causes, or game-specific ones, but the variability of it seems to match the non-deterministic nature of this kind of allocation scheme.

And naturally every game influencer complaining about those engines has been able to prove that memory management was the issue of their complaints.

As much as you were able to prove that CryEngine, Unreal, and many others do just fine.

I have sales numbers, and market share adoption across the industry, proving that they do just fine, on my side.

Whereas I am expecting profiler data showing memory issues caused by RAII patterns, from your side, or any of those influencers on whatever platform they complain about.

Yea, because sales numbers are a measure of software performance and therefore Jira is the fastest software in the observable universe.

Then don't complain about what you cannot prove.

Your comments here are vague and weirdly snarky and somewhat irrelevant to the topic and you'll require me to prove.. what exactly? Should I prove that Crysis was laggy when it came out? Or that I experience lags when I play Unreal games?

I guess you would like me to prove that the lag is caused by RAII but notice that I didn't actually wrote that - that's your reading of it. I just noted that the symptoms seem to point to it.

> I have sales numbers, and market share adoption across the industry, proving that they do just fine, on my side.

This proves what about their resource usage?

It proves they stand above the competition, across all levels, selected by AAA studios that keep delivering across the industry.

They should know a couple of things about how to deliver something into production.

Additionally influencers complaining about them, usually never even wrote a basic hello world, let alone understand how memory management works.

I'm sorry, how is that related to the topic of performance?

It is something AAA game studios tend to care about when picking their tools of trade.

So's popularity. I'm sure they consider a lot when picking tools. You're claiming this proves something that requires hard evidence to disprove, but I don't think the connection is as strong as you imply.

So, in the context of a thread about resource usage, what does the commercial success of the tools prove?

That shows how little you know about AAA studios culture regarding performance.

My credentials are having been a IGDA member for several years, already been in places like SCEE SOHO studio, regular GDCE attendee during its time in UK, and first GCDE in continental.

And to turn you question around, it proves that arguing about RAII impact on commercial success of the tools is completly irrelevant without solid proofs, they are printing money.

This thread talks about performance, you're the one bringing commercial stuff into it.

> Poor performance. Usually, it is significantly more efficient to allocate and free resources in batches.

This one is big. It is a lot of accounting work to individually allocate and deallocate a lot of objects.

The Zig approach is to force you to decide on and write all of your allocation and deallocation code, which I found leads to more performant code almost by default – Rust is explicitly leaving that on the table. C obviously works the same way, but doesn't have an arena allocator in the standard library.

Re: C++ vs Rust, it might be more of a pain in Rust because of this: https://news.ycombinator.com/item?id=33637092

I am happy that Rust defaults to the easier, saner, safer approach by default, but lets you bypass RAII if you want to do so.

Yeah, I agree. It makes sense and I'm glad that Drop is in no way guaranteed. I am excited for allocator-api to eventually stabilize, too!

In language performance benchmarks, you often see C++ being slower than C. Well, that's a big part of the reason. There is no real reason for C++ to be slower since you can just write C code in C++, with maybe a few details like "restrict" that are not even used that much.

The big reason is that RAII encourage fine grained resource allocation and deallocaion, while explicit malloc() and free() will encourage batch allocation and deallocation, in-place modification, and reuse of previously allocated buffers, which are more efficient.

The part about "out of memory" situations is usually not of concern on a PC, memory is plentiful and the OS will manage the situation for you, which may include killing the process in a way you can't do much about. But on embedded systems, it matters.

Often, you won't hear these criticisms. C programmers are not the most vocal about programming languages, the opposite of Rust programmers I would say. I guess that's a cultural thing, with C being simple and stable. C++ programmers are more vocal, but most are happy about RAII, so they won't complain, I can see game developers as an exception though.

    > In language performance benchmarks, you often see C++ being slower than C.

This is the first that I have ever seen such a statement. And, I am sure we could legitimately write exactly the opposite: <<you often see C being slower than C++>>. Not much being said here. As I understand, C++ template code can be much faster than equivalent code in C because the compiler has more metadata for optimisation. The classic case is C's qsort() vs C++'s sort(), where the C++ version is much faster due to compiler optimisation.

You're most likely to hear it from kernel programmers and Go/Zig/Odin advocates, but it rarely comes up as a criticism of C++ in particular. Perhaps that's because RAII is merely "distasteful" in that cohort, whereas there are many other qualities of C++ that might be considered total showstoppers for adoption before matters of taste are ever on the table.

There was an HN thread a few months ago[0] debating whether RAII is reason enough to disqualify Rust as a systems language, with some strong opinions in both directions.

[0]: https://news.ycombinator.com/item?id=42291417

Thx for linked thread. It was an interesting read to me.

I have got a feeling that the division comes from two different system programming folks. Hard realtime folks see unbounded time and uh-oh its fatal. On other side soft realtime folks ask about probabilities and profiling data and its relatively fine for them if probability is low enough. Where both of the camps would agree if number of effects is bound somehow.

The "hard mode" described in this article covers an issue I do keep running into with Golang funnily enough: I really hate the global imports, and I really want to just use dependency injection for things like OS-level syscalls...which is really what's happening here largely.

So it's interesting to see something very similar crop up here in a different language and domain: throw a true barrier down on when and how your code can request resources, as an explicit acknowledgement that the usage of them is a significant thing.

Reply on news.ycombinator.com