Gc(3)

💅 Gc(3) — the OCaml library manual page, bestie edition 💅

OCaml library · manual page Gc(3). This one is Memory management control and statistics; finalised values, and it’s lowkey iconic. 🫶

🌸 NAME 🌸

Gc — Memory management control and statistics; finalised values.

🦋 Module 🦋

Module Gc

✨ Documentation ✨

Module Gc : sig end

Memory management control and statistics; finalised values.

🎀 type stat — the memory counter receipts 🎀

Okay so first up, the stat record. It’s giving “receipts for your whole program,” and every field is a number. Here’s the exact shape:

type stat = {
  minor_words : float ;  (* Number of words allocated in the minor heap
since the program was started.
  *)
  promoted_words : float ;  (* Number of words allocated in the minor
heap that survived a minor collection and were moved to the major heap
since the program was started.
  *)
  major_words : float ;  (* Number of words allocated in the major heap,
including the promoted words, since the program was started.
  *)
  minor_collections : int ;  (* Number of minor collections since the
program was started.
  *)
  major_collections : int ;  (* Number of major collection cycles
completed since the program was started.
  *)
  heap_words : int ;  (* Total size of the major heap, in words.
  *)
  heap_chunks : int ;  (* Number of contiguous pieces of memory that make
up the major heap.  This metric is currently not available in OCaml 5:
the field value is always 0 .
  *)
  live_words : int ;  (* Number of words of live data in the major heap,
including the header words.

Note that "live" words refers to every word in the major heap that isn't
currently known to be collectable, which includes words that have become
unreachable by the program after the start of the previous gc cycle.  It
is typically much simpler and more predictable to call Gc.full_major (or
Gc.compact ) then computing gc stats, as then "live" words has the
simple meaning of "reachable by the program". One caveat is that a
single call to Gc.full_major will not reclaim values that have a
finaliser from Gc.finalise (this does not apply to Gc.finalise_last ).
If this caveat matters, simply call Gc.full_major twice instead of once.
  *)
  live_blocks : int ;  (* Number of live blocks in the major heap.

See live_words for a caveat about what "live" means.
  *)
  free_words : int ;  (* Number of words in the free list.
  *)
  free_blocks : int ;  (* Number of blocks in the free list.  This metric
is currently not available in OCaml 5: the field value is always 0 .
  *)
  largest_free : int ;  (* Size (in words) of the largest block in the
free list.  This metric is currently not available in OCaml 5: the field
value is always 0 .
  *)
  fragments : int ;  (* Number of wasted words due to fragmentation.
These are 1-words free blocks placed between two live blocks.  They are
not available for allocation.
  *)
  compactions : int ;  (* Number of heap compactions since the program
was started.
  *)
  top_heap_words : int ;  (* Maximum size reached by the major heap, in
words.
  *)
  stack_size : int ;  (* Current size of the current OCaml stack, in
words.

Since 5.5 restored for the current stack
  *)
  forced_major_collections : int ;  (* Number of forced full major
collections completed since the program was started.

Since 4.12
  *)
  live_stacks_words : int ;  (* Total space allocated outside of the
OCaml heap for stack fragments.  This includes stack metadata and stack
fragments stored inside the stack fragment cache.

Since 5.5
  *)
  }

The memory management counters are returned in a stat record. These counters give values for the whole program. Periodt. 🫶

The total amount of memory allocated by the program since it was started is (in words) minor_words + major_words - promoted_words. Multiply by the word size (4 on a 32-bit machine, 8 on a 64-bit machine) to get the number of bytes. Math slay. ✨

🧚 type control — the knobs bestie can turn 🧚

Now the control record. This is the settings menu, and it’s a whole vibe:

type control = {
  minor_heap_size : int ;  (* The size (in words) of the minor heap.
Changing this parameter will trigger a minor collection. The total size
of the minor heap used by this program is the sum of the heap sizes of
the active domains. Default: 256k.
  *)
  major_heap_increment : int ;  (* How much to add to the major heap when
increasing it. If this number is less than or equal to 1000, it is a
percentage of the current heap size (i.e. setting it to 100 will double
the heap size at each increase). If it is more than 1000, it is a fixed
number of words that will be added to the heap.

This field is currently not available in OCaml 5: the field value is
always 0 .
  *)
  space_overhead : int ;  (* The major GC speed is computed from this
parameter, along with small_heap_limit .  This is the memory that will
be "wasted" because the GC does not immediately collect unreachable
blocks.  It is expressed as a percentage of the memory used for live
data.  The GC will work more (use more CPU time and collect blocks more
eagerly) if space_overhead is smaller.  The amount of overhead space
used by the GC is approximately:

- (live data size) * space_overhead when live data is greater than
small_heap_limit

-less than small_heap_limit + (live data size) * space overhead when
live data is smaller than small_heap_limit Default: 120.

  *)
  verbose : int ;  (* This value controls the GC messages on standard
error output.  It is a sum of some of the following flags, to print
messages on the corresponding events:

- 0x0001 Start and end of major GC cycle.

- 0x0002 Minor collection and major GC slice.

- 0x0004 Growing and shrinking of the heap.

- 0x0008 Resizing of stacks and memory manager tables.

- 0x0010 Heap compaction.

- 0x0020 Change of GC parameters.

- 0x0040 Computation of major GC slice size.

- 0x0080 Calling of finalisation functions.

- 0x0100 Bytecode executable and shared library search at start-up.

- 0x0200 Computation of compaction-triggering condition.

- 0x0400 Output GC statistics at program exit.

- 0x0800 GC debugging messages.

- 0x1000 Address space reservation changes.  Default: 0.

  *)
  max_overhead : int ;  (* Heap compaction is triggered when the
estimated amount of "wasted" memory is more than max_overhead percent of
the amount of live data.  If max_overhead is set to 0, heap compaction
is triggered at the end of each major GC cycle (this setting is intended
for testing purposes only).  If max_overhead >= 1000000 , compaction is
never triggered.

This field is currently not available in OCaml 5: the field value is
always 0 .
  *)
  stack_limit : int ;  (* The maximum size of the fiber stacks (in
words).  Default: 128M.
  *)
  allocation_policy : int ;  (* The policy used for allocating in the
major heap.

This field is currently not available in OCaml 5: the field value is
always 0 .

Prior to OCaml 5.0, possible values were 0, 1 and 2.

-0 was the next-fit policy

-1 was the first-fit policy (since OCaml 3.11)

-2 was the best-fit policy (since OCaml 4.10)

Since 3.11
  *)
  window_size : int ;  (* The size of the window used by the major GC for
smoothing out variations in its workload. This is an integer between 1
and 50.

Since 4.03

        This field is currently not available in OCaml 5: the field
value is
        always [0].
  *)
  custom_major_ratio : int ;  (* Target ratio of floating garbage to
major heap size for out-of-heap memory held by custom values located in
the major heap. The GC speed is adjusted to try to use this much memory
for dead values that are not yet collected. Expressed as a percentage of
major heap size. The default value keeps the out-of-heap floating
garbage about the same size as the in-heap overhead.  Note: this only
applies to values allocated with caml_alloc_custom_mem (e.g. bigarrays).
Default: 44.

Since 4.08
  *)
  custom_minor_ratio : int ;  (* Bound on floating garbage for
out-of-heap memory held by custom values in the minor heap. A minor GC
is triggered when this much memory is held by custom values located in
the minor heap. Expressed as a percentage of minor heap size.  Note:
this only applies to values allocated with caml_alloc_custom_mem (e.g.
bigarrays).  Default: 100.

Since 4.08
  *)
  custom_minor_max_size : int ;  (* Maximum amount of out-of-heap memory
for each custom value allocated in the minor heap. Custom values that
hold more than this many bytes are allocated on the major heap.  Note:
this only applies to values allocated with caml_alloc_custom_mem (e.g.
bigarrays).  Default: 70000 bytes.

Since 4.08
  *)
  }

The GC parameters are given as a control record. Note that these parameters can also be initialised by setting the OCAMLRUNPARAM environment variable. See the documentation of ocamlrun. 🧚

🌈 The functions — what you actually call 🌈

• val stat : unit -> stat — Return the current values of the memory management counters in a stat record that represents the program’s total memory stats. The heap_chunks, free_blocks, largest_free, and stack_size metrics are currently not available in OCaml 5: their returned field values are therefore 0. This function causes a full major collection. 💅
• val quick_stat : unit -> stat — Returns a record with the current values of the memory management counters like stat. Unlike stat, quick_stat does not perform a full major collection, and hence, is much faster. However, quick_stat reports the counters sampled at the last minor collection or at the end of the last major collection cycle (whichever is the latest). Hence, the memory stats returned by quick_stat are not instantaneously accurate. ⚡
• val counters : unit -> float * float * float — Return (minor_words, promoted_words, major_words) for the current domain or potentially previous domains. This function is as fast as quick_stat. 🏃‍♀️
• val minor_words : unit -> float — Number of words allocated in the minor heap by this domain or potentially previous domains. This number is accurate in byte-code programs, but only an approximation in programs compiled to native code. In native code this function does not allocate. Since 4.04 ✨
• val get : unit -> control — Return the current values of the GC parameters in a control record. The major_heap_increment, max_overhead, allocation_policy, and window_size fields are currently not available in OCaml 5: their returned field values are therefore 0. Alert unsynchronized_access. GC parameters are a mutable global state. 🚨
• val set : control -> unit — set r changes the GC parameters according to the control record r. The normal usage is:

Gc.set { (Gc.get()) with Gc.verbose = 0x00d }

The major_heap_increment, max_overhead, allocation_policy, and window_size fields are currently not available in OCaml 5: setting them therefore has no effect. Alert unsynchronized_access. GC parameters are a mutable global state. 🫶

• val minor : unit -> unit — Trigger a minor collection. 🎀
• val major_slice : int -> int
• major_slice n — does a minor collection plus a slice of major collection. n is the size of the slice: the GC does enough work to free (on average) n words of memory. If n = 0, the GC tries to do enough work to make sure the next automatic slice has nothing to do. It returns an unspecified integer (currently: 0).
• val major : unit -> unit — does a minor collection and finishes the current major collection cycle. slay ✨
• val full_major : unit -> unit — does a minor collection, finishes the current major collection cycle, and then performs a complete new cycle. This collects all currently unreachable blocks 🫶
• val compact : unit -> unit — performs a full major collection and compacts the heap. Heads up bestie: heap compaction is a lengthy operation, so don't just spam it 🙈
• val print_stat : out_channel -> unit — prints the current values of the memory management counters (in human-readable form) of the total program into the channel argument. iykyk 📊
• val allocated_bytes : unit -> float — returns the number of bytes allocated by this domain and potentially a previous domain. It comes back as a float to dodge overflow problems with int on 32-bit machines. very thoughtful of it 💅
• val get_minor_free : unit -> int — returns the current size of the free space inside the minor heap of this domain.

Since 4.03

---

💖 the finalise saga 💖

• val finalise : ('a -> unit) -> 'a -> unit — finalise f v registers f as a finalisation function for v. v must be heap-allocated, no cap. f gets called with v as its argument at some point between the first time v becomes unreachable (including through weak pointers 💀) and the time v is collected by the GC. Several functions can be registered for the same value, or even several instances of the same function. Each instance gets called once (or never, if the program terminates before v becomes unreachable 🙈).

The GC calls the finalisation functions in the order of deallocation. When several values become unreachable at the same time (i.e. during the same GC cycle), the finalisation functions get called in the reverse order of the corresponding calls to finalise. If finalise is called in the same order as the values are allocated, that means each value is finalised before the values it depends upon. Of course, this becomes false if assignments introduce additional dependencies.

With multiple OCaml threads in the mix, you should assume any particular finaliser may be executed in any of the threads 👀

Anything reachable from the closure of finalisation functions counts as reachable, so this code will not work as expected:

let v = ... in Gc.finalise (fun _ -> ...v...) v

Instead, make sure v is not in the closure of the finalisation function by writing:

let f = fun x -> ...  let v = ... in Gc.finalise f v

The f function can use all features of OCaml, including assignments that make the value reachable again. It can also loop forever — in that case the other finalisation functions will not be called during the execution of f, unless it calls finalise_release. It can call finalise on v or on other values to register other functions or even itself. It can raise an exception; in that case the exception will interrupt whatever the program was doing when the function was called 😭

finalise raises Invalid_argument if v is not guaranteed to be heap-allocated. Some examples of values that are not heap-allocated: integers, constant constructors, booleans, the empty array, the empty list, the unit value. The exact list of what is or isn't heap-allocated is implementation-dependent. Some constant values can be heap-allocated but never deallocated during the lifetime of the program — for example a list of integer constants — and this is also implementation-dependent. Note that values of type float are sometimes allocated and sometimes not, so finalising them is unsafe, and finalise will also raise Invalid_argument for them. Values of type 'a Lazy.t (for any 'a) are like float in this respect, except that the compiler sometimes optimizes them in a way that prevents finalise from detecting them. In that case it will not raise Invalid_argument, but you should still avoid calling finalise on lazy values 🧚

The results of calling String.make, Bytes.make, Bytes.create, Array.make, and ref are guaranteed to be heap-allocated and non-constant, except when the length argument is 0.

• val finalise_last : (unit -> unit) -> 'a -> unit — same as Gc.finalise, except the value is not given as an argument. So you can't use the given value for the computation of the finalisation function. The glow-up? The function is called after the value is unreachable for the last time instead of the first time. So contrary to Gc.finalise, the value will never be reachable again or used again. In particular, every weak pointer and ephemeron that contained this value as key or data is unset before running the finalisation function. Moreover, the finalisation functions attached with Gc.finalise are always called before the finalisation functions attached with Gc.finalise_last. iconic 🌈

Since 4.04

• val finalise_release : unit -> unit — a finalisation function may call finalise_release to tell the GC that it can launch the next finalisation function without waiting for the current one to return. go on, girl, don't wait up 💅

---

🎀 alarms 🎀

• type alarm — an alarm is a piece of data that calls a user function at the end of major GC cycle. The following functions are provided to create and delete alarms.
• val create_alarm : (unit -> unit) -> alarm — create_alarm f arranges for f to be called at the end of major GC cycles, not caused by f itself, starting with the current cycle or the next one. f runs on the same domain that created the alarm, until the domain exits or delete_alarm is called. A value of type alarm is returned, and you use it to call delete_alarm.

It is not guaranteed that the Gc alarm runs at the end of every major GC cycle, but it is guaranteed that it will run eventually ⭐

As an example, here's a crude way to interrupt a function if the memory consumption of the program exceeds a given limit in MB — suitable for use in the toplevel:

let run_with_memory_limit (limit : int) (f : unit -> 'a) : 'a =
  let limit_memory () =
    let mem = Gc.(quick_stat ()).heap_words in
    if mem / (1024 * 1024) > limit / (Sys.word_size / 8) then
      raise Out_of_memory
  in
  let alarm = Gc.create_alarm limit_memory in
  Fun.protect f ~finally:(fun () -> Gc.delete_alarm alarm ; Gc.compact ())

• val delete_alarm : alarm -> unit — delete_alarm a stops the calls to the function associated with a. Calling delete_alarm a again has no effect. we love an idempotent queen 👑
• val eventlog_pause : unit -> unit — Deprecated. Use Runtime_events.pause instead. bestie it's giving legacy, pls migrate 😭
• val eventlog_resume : unit -> unit — Deprecated. Use Runtime_events.resume instead.

---

🌸 module Memprof : sig end 🌸

Memprof is a profiling engine which randomly samples allocated memory words. Every allocated word has a probability of being sampled equal to a configurable sampling rate. Once a block is sampled, it becomes tracked. A tracked block triggers a user-defined callback as soon as it is allocated, promoted or deallocated 🔥

Since blocks are composed of several words, a block can potentially be sampled several times. If a block is sampled several times, then each of the callbacks is called once for each event of this block: the multiplicity is given in the n_samples field of the allocation structure.

This engine makes it possible to implement a low-overhead memory profiler as an OCaml library 🫶

Note: this API is EXPERIMENTAL. It may change without prior notice. consider yourself warned, bestie 💀

---

🦋 ramp-up szn 🦋

• type suspended_collection_work
• val ramp_up : (unit -> 'a) -> 'a * suspended_collection_work — in general, the OCaml GC assumes the program runs in a "steady state" where peak memory usage remains constant: for each newly allocated work, it assumes one work has become unreachable and will try to collect it during the next GC slice.

This assumption is incorrect at the points during program execution where live memory increases instead of staying stable: the steady-state assumption makes the GC work harder at no benefit, since it won't find more memory to collect 😬

ramp_up f puts the current domain in a "ramp-up" phase for the duration of the evaluation of f (), letting the GC know the steady-state assumption doesn't hold; use it when you know the live memory of the program is about to increase significantly.

During a ramp-up phase, the GC will not try to work harder for new allocations: the corresponding collection work is "suspended". The total amount of suspended collection work is returned by ramp_up along with the result of the function.

If the user discards this suspended work (by doing nothing with it), the GC will never accelerate to recover the corresponding amount of memory. That's appropriate if the ramp-up work allocates long-lived memory that remains live until the end of program execution.

If the user knows that at a certain point in the program the live memory consumption has been reduced by the corresponding amount — typically because the memory allocated during ramp_up has become unused — then they should call Gc.ramp_down below to have the GC "resume" this collection work.

If f () raises an exception, the ramp-up phase terminates, the collection work that was suspended is resumed, and the exception is re-raised.

If f () performs an effect, the effect is not handled and an Effect.Unhandled exception is thrown instead.

• val ramp_down : suspended_collection_work -> unit — notifies the GC about some amount of collection work that was suspended during a ramp-up phase, to be resumed now. we're back, baby ✨

---

💜 module Tweak : sig end 💜

GC Tweaks are unstable and undocumented configurable GC parameters, primarily intended for use by GC developers. ngl, this is for the GC devs, not for you 🙈

As well as using Gc.Tweak.set "foo" 42, they can also be configured in OCAMLRUNPARAM, using the following syntax:

OCAMLRUNPARAM='Xfoo=42'

Additionally, OCAMLRUNPARAM=Xhelp will show the available GC tweaks 👑

Since 5.5

---

and that's the whole page, bestie 🫶 this was Gc(3), brought to you by OCamldoc, dated 2026-06-22 — go forth and collect your garbage responsibly 💅✨