Writing memory efficient C structs

For completeness, this description of alignment is misleading:

> Well, dear reader, this padding is added because the CPU needs memory to be aligned in sets of 4 bytes because it’s optimized in that fashion.

> ...

> Remember: since structs are aligned to 4 bytes, any padding is therefore unnecessary if the size of the struct is a multiple of 4 without the padding.

Individual data types have their own alignment (e.g., `bool`/`char` may be 1, `short` may be 2, `int` may be 4, `long` may be 8, etc.), and the alignment of a compound type (like a struct) defaults to the maximum alignment of its constituent types.

In this article, `struct Monster` has an alignment of 4 because `int` and `float` have an alignment of 4 for the author's configuration. Expanding one of the `int`s to a `long` could increase the alignment to 8 on some CPUs, and removing the `int` and `float` fields would decrease the alignment to 1 for most CPUs.

The author relies on the compiler fitting the bitfield in unused padding bytes after “speed”, but that is implementation-defined behavior (almost everything about bitfields is implementation defined). MSVC will align the unsigned bitfield to alignof(unsigned), whereas GCC and clang will use the padding bytes. To portably pack the struct, use unsigned integers and use flags (monster.flags & CAN_FLY instead of monster.can_fly).

See https://c.godbolt.org/z/K7oY77KGj

  class mybits_t : bit::bit_array<17> {
    public:
      // slice a 1 bit sized field
      // returns a proxy reference to the single bit
      auto field1() {
        return (*this)(0, 1);
      }

      // slice a 3 bit sized field 
      // returns a proxy reference to the 3 bits
      auto field2() {
        return (*this)(1, 4);
      } 
  };
  mybits_t mybits(7);
  assert(mybits.field1() == 1);
  mybits.field1() = 0;
  assert(static_cast<int>(mybits) == 6); //no implicit cast to integers, but explicit supported

This is a good article, but it bugs me that it sort of only tells half the story.

Why are you making your structs smaller. Probably for "performance". If your goal is literally just reducing memory usage, then this is all fine. Maybe you're running on a small microcontroller and you just straight up don't have much RAM.

But these days, in most areas, memory is cheap and plentiful. What's expensive is CPU cache. If you're optimizing memory use for that, it's really about runtime speed and less about total memory usage. You're trying to make things small so you have fewer cache misses and the CPU doesn't get stuck waiting as much.

In that case, the advice here is only part of the story. Using bitfields and smaller integers saves memory. But in order to access those fields and do anything with them, they need to get loaded into registers. I'm not an expert here, but my understanding is that loading a bitfield or small int into a word-sized register can sometimes have some bit of overhead, so you have to pay attention to the trade-off.

If I was optimizing for overall runtime speed and considering packing structs to do that, I'd really want to have good profiling and benchmarks in place first. Otherwise it's easy to think you're making progress when you're actually going backwards.

Note that although alignment and size are platform-specific, in practice padding can be minimized simply by using a fixed order:

  (various vector types)
  int128_t
  long double
  long long, int64_t, int_fast64_t, double
  off_t
  time_t (can be 64 bits on 32-bit platforms only if off_t also is)
  register_t, int_fast16_t, int_fast32_t (64-bit on e.g. x32)
  void *, void(*)(), intptr_t
  ptrdiff_t (theoretically wrong if fully aligned and bigger than pointer)
  size_t, ssize_t
  long (32-bit on win64; generally has 16-bit alignment if bigger than pointer)
  int32_t, float
  int (can be 16-bit)
  short, int16_t
  char, bool, int8_t, int_fast8_t
  (a trailing array member must come last; note when calculating combined size you should use offsetof instead of sizeof to not waste padding)

Besides this, `int_leastN_t` goes with `intN_t`, and all unsigned types go with their signed types. Note that types like `id_t` are hard to place however.

Although there do exist many platforms where this is not in order of size, the bigger types generally have smaller alignment on such platforms. Despite this I have split out rows occasionally despite not knowing of any such platform.

Generally speaking, sort the members in descending order by alignment, if you want the tightest packing without resorting to nonportable compiler extensions. This descending order eliminates padding between members producing the smallest offset for the last byte of the last member, and that then minimizes the padding required at the end of the struct.

In C, alignment for basic types is linked to size. A type may have an alignment requirement that is as much as its size: e.g. an 8 byte data type might be aligned as strictly as addresses divisible by 8.

The alignment of an array aggregate is that of its element type.

The alignment of a struct is that of its most strictly aligned member.

In portable coding, we don't exactly know, but we can make the worst-case assumption that every basic type has the strictest possible alignment---i.e. its size. We also don't know the sizes of all the types, but we know that integers of higher ranks are at least as large as lower ranks; i.e. sizeof(char) <= sizeof(short) <= sizeof(int) <= sizeof(long), etc.

Things are tricky when you have a mixture of floating-point, pointers and integers. E.g. pointer could be the same size as a float, or bigger.

Here is a trick you use: combine together clumps of like types, and try to make combinations that are powers of two. For instance, if your structure has two or four pointers to things, try to make them adjacent. E.g.:

  struct s {
    void *p1, *p2, *p3, p4;
    short s1, s2, s3, s4;
    float f1, f2, f3, f4;
  };

Even though float is almost certainly larger than short and may have stricter alignment, its moot because I have four of everything. Even if this is on a tiny system where pointers are 16 bits wide, four of them will make 8 bytes. So when the f1 member is being placed into the struct, the offset is at least 8 byte aligned: we don't expect padding.

If you aggregate enough members of the same type in a power-of-two sized clump, you can easily create alignment which meets or exceeds the strictest alignment in the system.

So those are the main principles: group together members of the same type. Then keeping those groups together, sort by descending alignment strictness.

No mention of pahole / dwarves? https://lwn.net/Articles/335942/ It's the standard tool used by the Linux kernel developers.

Some other things you can do to improve packing are using integer indices/offsets/relative pointers instead of absolute pointers (especially on 64-bit machines), overlapping fields for unions (as an example you may have a tag that can share some bits with another field in each member of the union), using a structs position in memory to encode information (for example, if you have a binary tree of array allocated nodes, you can specify that the right child always follows the parent in memory, then you only need to store the index of the left child), and of course structure packing pragmas/attributes.

Some related thoughts (more for C++ than C).

You can eliminate the cost of a wrapping class with no members by taking advantage of the empty base class optimization (or rather: by avoiding members in a base class you will be taking advantage of EBCO).

You can use an untagged union in C++ (taking care to avoid undefined behavior) to implement small object optimization. In it's general form, the idea is you can store some smaller struct in a union (as a char[], since C++ doesn't have native unions) with a larger structure, and use a sentinel in the larger structure to determine which it is. E.g., if large struct has some field that must not be zero, then make it zero in the smaller struct and you can distinguish between the two without a tag (like would be required with std::variant).

I'm also a huge fan in general of "make invalid states unrepresentable" and following this can often reduce redundancy and eliminate extra fields.

Optimizing struct layouts is cool. but if your motivation is to pack more monsters in your game definitely consider transposing your row oriented monster struct into a column oriented entity using Entity Component System.

Example framework:

https://github.com/skypjack/entt?tab=readme-ov-file

It has benefits besides memory/cache optimizations. It organizes logic in a much more composable way that is friendly for reuse.

Nice little article.

The next thing to really explore is how the struct is being used. Smaller is always nice, but sometimes you need it to be fast. To do that, you'll want to group together fields that are commonly read. For example, x, y, and speed are likely always read together. So, it'd make sense to group those fields together to optimize for cache reads.

The next thing to question is if your Monster is being interacted with as an individual thing or as a group of things. If it's as a group of things then it might make sense to create a "Struct of arrays" of monsters. Imagine, for example, you have a system which removes all the dead monsters. It'd be faster to iterate through an `isAlive` array rather than an array of `Monster` structs.

And now you are on your way to reinventing an ECS :)

>“… Note that I'm not an expert, in anything, so take everything I write about with a grain of salt I guess :).”

>”I’m by far an expert and I could be wrong in some specifications”

You’re missing a “not” on that 2nd sentence ;)

In all seriousness, you don’t need this disclaimer, you’re selling yourself short.

There’s nothing wrong with your post to make the disclaimer.

And if there is something wrong with the post, let the comments flood in and tell you that rather than yourself.

I don't think I've ever seen struct bitfields used in real code, it's almost always manual bit-masks against a numeric type, often via macros. (I assume due to portability)

Sometimes struct-of-arrays (in-memory column store) is the right solution, and you get automatic packing that way. C makes it so much easier to store an array of structs that people don't do this often enough, but column stores prove it has its uses.

I'm somewhat surprised to see this on the front page of HN. I distinctly remember this material being covered in a single lecture during our freshman CS course 20 years ago. I doubt this would've reached the front page in HN's early days, and it makes me wonder if we've moved away from teaching these fundamentals due to the heavy reliance on abstraction and ai agents in today's industry.

This reminds of an amazing talk by Mike Acton on Data oriented design: https://www.youtube.com/watch?v=rX0ItVEVjHc

And there is a related talk by Andrew Kelley implementing these ideas on Zig toolchain: https://www.youtube.com/watch?v=IroPQ150F6c

and i raise you this: https://www.catb.org/esr/structure-packing/

This is good stuff to have "in the toolbox," however habitually organizing structs this way is premature optimization. Much better to sensibly organize the fields and ignore alignment entirely unless you have a really good reason to do this. Good reasons could be that you know you will create a lot of these objects, or it has become a performance issue due to cache misses this would improve.

I would suggest to never use unsigned int for values that will involve any sort of calculations on them (health, damage, speed). This is considered bad practice due to overflow and should be avoided, with the exception of values that are used for UIDs, indexes, bitfields, mask and flags.

Neither TFA nor any comment mention using a struct of arrays instead of an array of structs. This technique alone eliminates all padding bytes!

I would expect many C programmers would know these things, because I do like that when writing programs in C, too (although for bit fields, I will use a single field and then named bits by macros). There are some other considerations as well though, such as static vs dynamic allocation, and the requirements of how the specific program is working.

definitely interesting points for C structs. regarding CPU it might be interesting to look a little further still regarding optimal sizes for structs regarding cachcing, figuring out hot and cold areas and either including padding or putting cold values behind pointer to a substruct.

like the examples though, very clear what the changes are and the specific impact. always a good reminder to review for these things after a session hacking in new things!

Could improve it by using enum MonsterName :u8, moving it down near the other byte sized item, and specifying the exact type of the bitfield(u8). Then it would be 16 bytes.

Remind me about Andrew Kelley conf "data-oriented design" He had explained about this topic.

for another good intro to the fun with structs, Beej's guide is always a good goto:

https://beej.us/guide/bgc/html/split-wide/structs-ii-more-fu...

...goes into offsets, padding bytes, bit-field, unions, etc etc.

Minor nitpick, 32bit max value is ~4 billion, not 4 million

Adding to this you want the struct to be exactly 64 bytes, even on Apple MX processors which have 128 bytes of cache.

This is also applicable outside of C as well.

The article casually ignores the fact that more production code uses bitmasking than bit fields.

Loved this, thanks!

Unpacked structs are also very useful for parsing network data. You take a package, cast it to an unpacked struct, and can access everything nicely.

Many a spacecraft telemetry stacks work like that.

Is it possible to apply these optimizations to Arrow?

Arrow's struct is called StructArray. Fields of StructArray have a StructType.

StructType has .bit_width and .byte_width attrs in Python and probably the other implementations too: https://arrow.apache.org/docs/python/generated/pyarrow.Struc...

Arrow supports bitfields with BooleanArray, and enums with categoricals but

"BUG: Categorical columns using the PyArrow backend requires 4x more memory" (open) https://github.com/pandas-dev/pandas/issues/58062 :

> On disk Parquet appears to store the category data as logical type String which is compressed with snappy and encoded

Arrow Flight RPC handles nested structs with enums over the wire somehow too FWIU