dmem_cgroup_try_charge(9)

📦 version 7.3.1, in nixpkgs unstable

💖 dmem_cgroup_try_charge(9) 💖

okay so bestie, welcome to the Kernel API Manual — this one's about the kernel's dmem cgroup stuff, and honestly it's giving "let me try to charge this allocation and see if the region says yes" energy. sit down, grab a matcha 🍓, we're doing this.

✨ NAME ✨

dmem_cgroup_try_charge — try charging a new allocation to a region. that's it, that's the vibe 🎀

🌸 SYNOPSIS 🌸

int dmem_cgroup_try_charge (struct dmem_cgroup_region *region , u64 size
, struct dmem_cgroup_pool_state **ret_pool , struct
dmem_cgroup_pool_state **ret_limit_pool );

like, copy this exactly pls, no cap, the compiler is not forgiving 🙈

🦋 ARGUMENTS 🦋

  • region — the dmem region you're charging. your main character 🫶
  • size — how many bytes you're charging. in bytes, bestie, bytes. 🔥
  • ret_pool — on a successful allocation, this gets set to the pool that actually got charged. so you know who took the tab 💅
  • ret_limit_pool — on a failed allocation, this is the limiting pool. the one that said "not today" 😭

💜 DESCRIPTION 💜

so this function charges the region region for a size of size bytes. that's the whole bit. ⭐

if the function succeeds, ret_pool is set — and here's the important part, so listen up: that pool must be passed to dmem_cgroup_uncharge when you're undoing the allocation. don't just vibe off into the sunset and forget it, fr fr 🧚

when this function fails with -EAGAIN and ret_limit_pool is non-null, it gets set to the pool for which the limit is hit. like, it literally tells you which pool is the problem 😤 super useful — you can use it for eviction as the argument to dmem_cgroup_evict_valuable. this reference must be freed with dmem_cgroup_pool_state_put. yes, must. not "should". we said what we said. 🎯

okay quick recap because I know you're multitasking: success = ret_pool, and please actually uncharge it later. -EAGAIN + non-null ret_limit_pool = the limiting pool, and please actually put it later. two different refs, two different cleanups. iykyk 💀

🌈 RETURN 🌈

  • 0 on success — iconic, we love a zero ✨
  • -EAGAIN on hitting a limit — the limit said nah 💅
  • or a negative errno on failure — some other little gremlin went wrong 🙈

👑 SEE ALSO 👑

aaaand that's the page, bestie! this one lives over in ./kernel/cgroup and the version we're looking at is dated October 2026 — so like, cosy little kernel-api-manual moment from the future. slay 💖🎀✨

📦 where this page comes from