aligned_alloc(3)

💖 aligned_alloc(3) 💖

okay so bestie, this is a page from the Library Functions Manual — yes we're deep in the Linux man-pages 6.20 zone — and today's main character is aligned_alloc, a libc girlie whose whole thing is giving you memory that's lined up just right. ✨

---

🌸 NAME 🌸

aligned_alloc — allocate aligned memory 🎀

📚 LIBRARY 📚

Standard C library — that's libc, -lc when you're linking. iykyk 💅

✨ SYNOPSIS ✨

#include <stdlib.h>

void *aligned_alloc(size_t alignment, size_t size);

🧚 Feature Test Macro Requirements for glibc (see feature_test_macros(7)):

• aligned_alloc(): _ISOC11_SOURCE

🦋 DESCRIPTION 🦋

aligned_alloc() grabs size bytes and hands you back a pointer to that allocated memory. The tea? ☕ The memory address is literally going to be a multiple of alignment, and alignment has to be a power of two. no cap. And then later you can pass that pointer to free(3) and it will work just fine, bestie. 🫶

⚠️ heads up: the memory is not zeroed. not even a little. don't assume it's clean. 💀

👑 RETURN VALUE 👑

Slay — on success, aligned_alloc() returns a pointer to the allocated memory. On error it returns NULL, and errno gets set so you know what went wrong. 🌈

💅 ERRORS 💅

• EINVAL — bestie. the alignment argument was not a power of two. girl. 😭
• ENOMEM — out of memory. it's giving exhausted. 💀

⭐ ATTRIBUTES ⭐

for an explanation of the terms used in this section, see attributes(7).

• Interface: aligned_alloc()
• Attribute: Thread safety
• Value: MT-Safe

so yes, she's thread safe. iconic behavior honestly. 👑

🎀 STANDARDS 🎀

C23, POSIX.1-2024.

🌸 HISTORY 🌸

glibc 2.16. C11, POSIX.1-2024.

💜 C11

okay so in C11, the spec for this function had several issues. like, actual problems. 🙈

• size had to be a multiple of alignment. otherwise the behavior was undefined.
• If alignment was not a power of two, the behavior was undefined.

DR460 reported both of those UB cases as unnecessarily dangerous, and fixed them with a Technical Corrigendum that turned them into errors. periodt. ✨

N2072 reported that the requirement that size is a multiple of alignment is superfluous, and removed it with a Technical Corrigendum.

C17 incorporates both technical corrigenda. The API has been stable ever since C17. 💖

glibc initially implemented it as silently aligning as stdc_bit_ceil(alignment) instead of alignment. Since glibc 2.38, it implements the C17 specification.

Some implementations, such as FreeBSD/jemalloc, implement the C17 specification, even though their documentation claims having undefined behavior. messy but ok. 😭

Some implementations, such as OpenBSD, implement C11 amended with DR460, even though their documentation claims having undefined behavior. same vibe honestly.

No known implementations have exploited the undefined behavior in a more dangerous way. This function should be safe to use. 🫶✨

🌈 NOTES 🌈

On many systems there are alignment restrictions — for example, on buffers used for direct block device I/O. POSIX specifies the pathconf(path,_PC_REC_XFER_ALIGN) call that tells you what alignment is needed. And now? you can use aligned_alloc() to satisfy that requirement. ate. 💅

The glibc malloc(3) always returns 8-byte aligned memory addresses, so this function is needed only if you require larger alignment values. 🔥

🦋 SEE ALSO 🦋

brk(2), getpagesize(2), free(3), malloc(3)

---

and that's the whole page bestie!! 💖 go forth and align ur memory responsibly. this was Linux man-pages 6.20, dated 2025-12-25 — mwah, happy reading, ily, bye!! 🎀✨🫶🌟