posix_memalign(3)

💖 posix_memalign(3) — Library Functions Manual 💖

okay bestie, welcome to the posix_memalign page, a certified classic from the Library Functions Manual. yes this is the alignment girlie you've heard about. let's get into it ✨

✨ NAME ✨

posix_memalign — allocate aligned memory

🫶 LIBRARY 🫶

Standard C library (libc, -lc)

🌸 SYNOPSIS 🌸

#include <stdlib.h>

int posix_memalign(void **memptr, size_t alignment, size_t size);

🎀 Feature Test Macro Requirements for glibc (peep feature_test_macros(7)) 🎀

• posix_memalign():
  • _POSIX_C_SOURCE >= 200112L

💅 DESCRIPTION 💅

so posix_memalign() allocates size bytes and drops the address of that freshly-allocated memory into *memptr. and the address it gives you? it's gonna be a multiple of alignment, which — and this is important, no cap — must be a power of two and a multiple of sizeof(void *). that same address can later be successfully passed to free(3). iconic, honestly.

if size is 0, then the value placed in *memptr is either NULL or a unique pointer value. either way it's valid, so don't stress 💖

heads up bestie: the memory is not zeroed. it's giving mystery box energy. if you wanted clean memory you're gonna have to do that yourself.

also, the way that aligned_alloc(3) is right there being a standard function with the same functionality and a more ergonomic prototype… yeah. use that instead. the manpage literally said it, not me 🙈

⭐ RETURN VALUE ⭐

posix_memalign() returns zero on success, or one of the error values listed in the next section on failure. and the plot twist: the value of errno is not set. you read that right. do not go checking errno for this one, she doesn't play that game.

🍓 ERRORS 🍓

• EINVAL — the alignment argument was not a power of two, or was not a multiple of sizeof(void *). basically you gave her a number she cannot work with 💀
• ENOMEM — out of memory. the machine said nope.

👑 ATTRIBUTES 👑

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

• posix_memalign() — Thread safety: MT-Safe

MT-Safe!! queen behavior, she's safe to call from multiple threads 🫶

🌈 STANDARDS 🌈

POSIX.1-2024.

🦋 HISTORY 🦋

glibc 2.1.91. Also POSIX.1d, POSIX.1-2001. she's been around, she's got lore 💜

⭐ CAVEATS ⭐

on Linux (and other systems), posix_memalign() does not modify memptr on failure. so like, if it flops, your pointer is untouched — don't go assuming it got set to garbage or NULL. a requirement standardizing this behavior was added in POSIX.1-2008 TC2. iykyk 😌

💖 SEE ALSO 💖

• aligned_alloc(3)

---

and that's the whole page, bestie!! 🎀 this one came to you straight from Linux man-pages 6.20, dated 2026-02-08, and it's officially wrapped. now go use aligned_alloc(3) like the manpage told you to. mwah 💋✨