;;;; garbage collection and allocation-related code ;;;; This software is part of the SBCL system. See the README file for ;;;; more information. ;;;; ;;;; This software is derived from the CMU CL system, which was ;;;; written at Carnegie Mellon University and released into the ;;;; public domain. The software is in the public domain and is ;;;; provided with absolutely no warranty. See the COPYING and CREDITS ;;;; files for more information. (in-package "SB!KERNEL") ;;;; DYNAMIC-USAGE and friends (declaim (special sb!vm:*read-only-space-free-pointer* sb!vm:*static-space-free-pointer*)) (eval-when (:compile-toplevel :execute) (sb!xc:defmacro def-c-var-frob (lisp-fun c-var-name) `(progn #!-sb-fluid (declaim (inline ,lisp-fun)) (defun ,lisp-fun () (sb!alien:extern-alien ,c-var-name (sb!alien:unsigned 32)))))) #!-gencgc (progn ;; This is called once per PROFILEd function call, so it's worth a ;; little possible space cost to reduce its time cost. #!-sb-fluid (declaim (inline current-dynamic-space-start)) (def-c-var-frob current-dynamic-space-start "current_dynamic_space")) #!-sb-fluid (declaim (inline dynamic-usage)) ; to reduce PROFILEd call overhead #!+gencgc (def-c-var-frob dynamic-usage "bytes_allocated") #!-gencgc (defun dynamic-usage () (the (unsigned-byte 32) (- (sb!sys:sap-int (sb!c::dynamic-space-free-pointer)) (current-dynamic-space-start)))) (defun static-space-usage () (- (* sb!vm:*static-space-free-pointer* sb!vm:n-word-bytes) sb!vm:static-space-start)) (defun read-only-space-usage () (- (* sb!vm::*read-only-space-free-pointer* sb!vm:n-word-bytes) sb!vm:read-only-space-start)) (defun control-stack-usage () #!-stack-grows-downward-not-upward (- (sb!sys:sap-int (sb!c::control-stack-pointer-sap)) sb!vm:control-stack-start) #!+stack-grows-downward-not-upward (- sb!vm:control-stack-end (sb!sys:sap-int (sb!c::control-stack-pointer-sap)))) (defun binding-stack-usage () (- (sb!sys:sap-int (sb!c::binding-stack-pointer-sap)) sb!vm:binding-stack-start)) ;;;; ROOM (defun room-minimal-info () (format t "Dynamic space usage is: ~10:D bytes.~%" (dynamic-usage)) (format t "Read-only space usage is: ~10:D bytes.~%" (read-only-space-usage)) (format t "Static space usage is: ~10:D bytes.~%" (static-space-usage)) (format t "Control stack usage is: ~10:D bytes.~%" (control-stack-usage)) (format t "Binding stack usage is: ~10:D bytes.~%" (binding-stack-usage)) (format t "Garbage collection is currently ~:[enabled~;DISABLED~].~%" *gc-inhibit*)) (defun room-intermediate-info () (room-minimal-info) (sb!vm:memory-usage :count-spaces '(:dynamic) :print-spaces t :cutoff 0.05f0 :print-summary nil)) (defun room-maximal-info () (room-minimal-info) (sb!vm:memory-usage :count-spaces '(:static :dynamic)) (sb!vm:instance-usage :dynamic :top-n 10) (sb!vm:instance-usage :static :top-n 10)) (defun room (&optional (verbosity :default)) #!+sb-doc "Print to *STANDARD-OUTPUT* information about the state of internal storage and its management. The optional argument controls the verbosity of output. If it is T, ROOM prints out a maximal amount of information. If it is NIL, ROOM prints out a minimal amount of information. If it is :DEFAULT or it is not supplied, ROOM prints out an intermediate amount of information." (fresh-line) (ecase verbosity ((t) (room-maximal-info)) ((nil) (room-minimal-info)) (:default (room-intermediate-info))) (values)) ;;;; GET-BYTES-CONSED ;;; the total number of bytes freed so far (including any freeing ;;; which goes on in PURIFY) ;;; ;;; (We save this so that we can calculate the total number of bytes ;;; ever allocated by adding this to the number of bytes currently ;;; allocated and never freed.) (declaim (type unsigned-byte *n-bytes-freed-or-purified*)) (defvar *n-bytes-freed-or-purified* 0) (push (lambda () (setf *n-bytes-freed-or-purified* 0)) ;; KLUDGE: It's probably not quite safely right either to do ;; this in *BEFORE-SAVE-INITIALIZATIONS* (since consing, or even ;; worse, something which depended on (GET-BYTES-CONSED), might ;; happen after that) or in *AFTER-SAVE-INITIALIZATIONS*. But ;; it's probably not a big problem, and there seems to be no ;; other obvious time to do it. -- WHN 2001-07-30 *after-save-initializations*) (declaim (ftype (function () unsigned-byte) get-bytes-consed)) (defun get-bytes-consed () #!+sb-doc "Return the number of bytes consed since the program began. Typically this result will be a consed bignum, so if you have an application (e.g. profiling) which can't tolerate the overhead of consing bignums, you'll probably want either to hack in at a lower level (as the code in the SB-PROFILE package does), or to design a more microefficient interface and submit it as a patch." (+ (dynamic-usage) *n-bytes-freed-or-purified*)) ;;;; variables and constants ;;; the minimum amount of dynamic space which must be consed before a ;;; GC will be triggered ;;; ;;; Unlike CMU CL, we don't export this variable. (There's no need to, ;;; since our BYTES-CONSED-BETWEEN-GCS function is SETFable.) (defvar *bytes-consed-between-gcs* #!+gencgc (* 4 (expt 10 6)) ;; Stop-and-copy GC is really really slow when used too often. CSR ;; reported that even on his old 64 Mb SPARC, 20 Mb is much faster ;; than 4 Mb when rebuilding SBCL ca. 0.7.1. For modern machines ;; with >> 128 Mb memory, the optimum could be significantly more ;; than this, but at least 20 Mb should be better than 4 Mb. #!-gencgc (* 20 (expt 10 6))) (declaim (type index *bytes-consed-between-gcs*)) ;;;; GC hooks (defvar *before-gc-hooks* nil ; actually initialized in cold init #!+sb-doc "A list of functions that are called before garbage collection occurs. The functions should take no arguments.") (defvar *after-gc-hooks* nil ; actually initialized in cold init #!+sb-doc "A list of functions that are called after garbage collection occurs. The functions should take no arguments.") (defvar *gc-notify-stream* nil ; (actually initialized in cold init) #!+sb-doc "When non-NIL, this must be a STREAM; and the functions bound to *GC-NOTIFY-BEFORE* and *GC-NOTIFY-AFTER* are called with the STREAM value before and after a garbage collection occurs respectively.") (defvar *gc-run-time* 0 #!+sb-doc "the total CPU time spent doing garbage collection (as reported by GET-INTERNAL-RUN-TIME)") (declaim (type index *gc-run-time*)) ;;; a limit to help catch programs which allocate too much memory, ;;; since a hard heap overflow is so hard to recover from ;;; ;;; FIXME: Like *GC-TRIGGER*, this variable (1) should probably be ;;; denominated in a larger unit than bytes and (2) should probably be ;;; renamed so that it's clear from the name what unit it's ;;; denominated in. (declaim (type (or unsigned-byte null) *soft-heap-limit*)) (defvar *soft-heap-limit* ;; As long as *GC-TRIGGER* is DECLAIMed as INDEX, we know that ;; MOST-POSITIVE-FIXNUM is a hard limit on how much memory can be ;; allocated. (Not necessarily *the* hard limit, which is fairly ;; likely something like a Unix per-process limit that we don't know ;; about, but a hard limit anyway.) And this gives us a reasonable ;; conservative default for the soft limit... (- most-positive-fixnum *bytes-consed-between-gcs*)) ;;;; The following specials are used to control when garbage ;;;; collection occurs. ;;; When the dynamic usage increases beyond this amount, the system ;;; notes that a garbage collection needs to occur by setting ;;; *NEED-TO-COLLECT-GARBAGE* to T. It starts out as NIL meaning ;;; nobody has figured out what it should be yet. ;;; ;;; FIXME: *GC-TRIGGER* seems to be denominated in bytes, not words. ;;; And limiting it to INDEX is fairly reasonable in order to avoid ;;; bignum arithmetic on every allocation, and to minimize the need ;;; for thought about weird gotchas of the GC-control mechanism itself ;;; consing as it operates. But as of sbcl-0.7.5, 512Mbytes of memory ;;; costs $54.95 at Fry's in Dallas but cheap consumer 64-bit machines ;;; are still over the horizon, so gratuitously limiting our heap size ;;; to FIXNUM bytes seems fairly stupid. It'd be reasonable to ;;; (1) allow arbitrary UNSIGNED-BYTE values of *GC-TRIGGER*, or ;;; (2) redenominate this variable in words instead of bytes, postponing ;;; the problem to heaps which exceed 50% of the machine's address ;;; space, or even ;;; (3) redemoninate this variable in CONS-sized two-word units, ;;; allowing it to cover the entire memory space at the price of ;;; possible loss of clarity. ;;; (And whatever is done, it'd also be good to rename the variable so ;;; that it's clear what unit it's denominated in.) (declaim (type (or index null) *gc-trigger*)) (defvar *gc-trigger* nil) ;;; When non-NIL, inhibits garbage collection. (defvar *gc-inhibit*) ; initialized in cold init ;;; This flag is used to prevent recursive entry into the garbage ;;; collector. (defvar *already-maybe-gcing*) ; initialized in cold init ;;; When T, indicates that the dynamic usage has exceeded the value ;;; *GC-TRIGGER*. (defvar *need-to-collect-garbage* nil) ; initialized in cold init (defun default-gc-notify-before (notify-stream bytes-in-use) (declare (type stream notify-stream)) (format notify-stream "~&; GC is beginning with ~:D bytes in use at internal runtime ~:D.~%" bytes-in-use (get-internal-run-time)) (finish-output notify-stream)) (defparameter *gc-notify-before* #'default-gc-notify-before #!+sb-doc "The function bound to this variable is invoked before GC'ing (unless *GC-NOTIFY-STREAM* is NIL) with the value of *GC-NOTIFY-STREAM* and current amount of dynamic usage (in bytes). It should notify the user that the system is going to GC.") (defun default-gc-notify-after (notify-stream bytes-retained bytes-freed new-trigger) (declare (type stream notify-stream)) (format notify-stream "~&; GC has finished with ~:D bytes in use (~:D bytes freed)~@ ; at internal runtime ~:D. The new GC trigger is ~:D bytes.~%" bytes-retained bytes-freed (get-internal-run-time) new-trigger) (finish-output notify-stream)) (defparameter *gc-notify-after* #'default-gc-notify-after #!+sb-doc "The function bound to this variable is invoked after GC'ing with the value of *GC-NOTIFY-STREAM*, the amount of dynamic usage (in bytes) now free, the number of bytes freed by the GC, and the new GC trigger threshold; or if *GC-NOTIFY-STREAM* is NIL, it's not invoked. The function should notify the user that the system has finished GC'ing.") ;;;; internal GC (sb!alien:define-alien-routine collect-garbage sb!alien:int #!+gencgc (last-gen sb!alien:int)) (sb!alien:define-alien-routine set-auto-gc-trigger sb!alien:void (dynamic-usage sb!alien:unsigned-long)) (sb!alien:define-alien-routine clear-auto-gc-trigger sb!alien:void) ;;; This variable contains the function that does the real GC. This is ;;; for low-level GC experimentation. Do not touch it if you do not ;;; know what you are doing. (defvar *internal-gc* #'collect-garbage) ;;;; SUB-GC ;;; This is used to carefully invoke hooks. (eval-when (:compile-toplevel :execute) (sb!xc:defmacro carefully-funcall (function &rest args) `(handler-case (funcall ,function ,@args) (error (cond) (warn "(FUNCALL ~S~{ ~S~}) lost:~%~A" ',function ',args cond) nil)))) ;;; SUB-GC decides when and if to do a garbage collection. The FORCE-P ;;; flags controls whether a GC should occur even if the dynamic usage ;;; is not greater than *GC-TRIGGER*. ;;; ;;; For GENCGC all generations < GEN will be GC'ed. (defun sub-gc (&key force-p (gen 0)) (/show0 "entering SUB-GC") (unless *already-maybe-gcing* (let* ((*already-maybe-gcing* t) (start-time (get-internal-run-time)) (pre-gc-dynamic-usage (dynamic-usage)) ;; Currently we only check *SOFT-HEAP-LIMIT* at GC time, ;; not for every allocation. That makes it cheap to do, ;; even if it is a little ugly. (soft-heap-limit-exceeded? (and *soft-heap-limit* (> pre-gc-dynamic-usage *soft-heap-limit*))) (*soft-heap-limit* (if soft-heap-limit-exceeded? (+ pre-gc-dynamic-usage *bytes-consed-between-gcs*) *soft-heap-limit*))) (when soft-heap-limit-exceeded? (cerror "Continue with GC." "soft heap limit exceeded (temporary new limit=~W)" *soft-heap-limit*)) (when (and *gc-trigger* (> pre-gc-dynamic-usage *gc-trigger*)) (setf *need-to-collect-garbage* t)) (when (or force-p (and *need-to-collect-garbage* (not *gc-inhibit*))) ;; KLUDGE: Wow, we really mask interrupts all the time we're ;; collecting garbage? That seems like a long time.. -- WHN 19991129 (without-interrupts ;; FIXME: We probably shouldn't do this evil thing to ;; *STANDARD-OUTPUT* in a binding which is wrapped around ;; calls to user-settable GC hook functions. (let ((*standard-output* *terminal-io*)) (when *gc-notify-stream* (if (streamp *gc-notify-stream*) (carefully-funcall *gc-notify-before* *gc-notify-stream* pre-gc-dynamic-usage) (warn "*GC-NOTIFY-STREAM* is set, but not a STREAM -- ignored."))) (dolist (hook *before-gc-hooks*) (carefully-funcall hook)) (when *gc-trigger* (clear-auto-gc-trigger)) (let* (;; We do DYNAMIC-USAGE once more here in order to ;; get a more accurate measurement of the space ;; actually freed, since the messing around, e.g. ;; GC-notify stuff, since the DYNAMIC-USAGE which ;; triggered GC could've done a fair amount of ;; consing.) (pre-internal-gc-dynamic-usage (dynamic-usage)) (ignore-me (funcall *internal-gc* gen)) (post-gc-dynamic-usage (dynamic-usage)) (n-bytes-freed (- pre-internal-gc-dynamic-usage post-gc-dynamic-usage)) ;; In sbcl-0.6.12.39, the raw N-BYTES-FREED from ;; GENCGC could sometimes be substantially negative ;; (e.g. -5872). I haven't looked into what causes ;; that, but I suspect it has to do with ;; fluctuating inefficiency in the way that the ;; GENCGC packs things into page boundaries. ;; Bumping the raw result up to 0 is a little ugly, ;; but shouldn't be a problem, and it's even ;; possible to sort of justify it: the packing ;; inefficiency which has caused (DYNAMIC-USAGE) to ;; grow is effectively consing, or at least ;; overhead of consing, so it's sort of correct to ;; add it to the running total of consing. ("Man ;; isn't a rational animal, he's a rationalizing ;; animal.":-) -- WHN 2001-06-23 (eff-n-bytes-freed (max 0 n-bytes-freed))) (declare (ignore ignore-me)) (/show0 "got (DYNAMIC-USAGE) and EFF-N-BYTES-FREED") (incf *n-bytes-freed-or-purified* eff-n-bytes-freed) (/show0 "clearing *NEED-TO-COLLECT-GARBAGE*") (setf *need-to-collect-garbage* nil) (/show0 "calculating NEW-GC-TRIGGER") (let ((new-gc-trigger (+ post-gc-dynamic-usage *bytes-consed-between-gcs*))) (/show0 "setting *GC-TRIGGER*") (setf *gc-trigger* new-gc-trigger)) (/show0 "calling SET-AUTO-GC-TRIGGER") (set-auto-gc-trigger *gc-trigger*) (dolist (hook *after-gc-hooks*) (/show0 "doing a hook from *AFTER-GC--HOOKS*") ;; FIXME: This hook should be called with the same ;; kind of information as *GC-NOTIFY-AFTER*. In ;; particular, it would be nice for the hook function ;; to be able to adjust *GC-TRIGGER* intelligently to ;; e.g. 108% of total memory usage. (carefully-funcall hook)) (when *gc-notify-stream* (if (streamp *gc-notify-stream*) (carefully-funcall *gc-notify-after* *gc-notify-stream* post-gc-dynamic-usage eff-n-bytes-freed *gc-trigger*) (warn "*GC-NOTIFY-STREAM* is set, but not a stream -- ignored."))))) (scrub-control-stack))) ;XXX again? we did this from C ... (incf *gc-run-time* (- (get-internal-run-time) start-time)))) ;; FIXME: should probably return (VALUES), here and in RETURN-FROM nil) ;;; This routine is called by the allocation miscops to decide whether ;;; a GC should occur. The argument, OBJECT, is the newly allocated ;;; object which must be returned to the caller. (defun maybe-gc (&optional object) (sub-gc) object) ;;; This is the user-advertised garbage collection function. (defun gc (&key (gen 0) (full nil) &allow-other-keys) #!+(and sb-doc gencgc) "Initiate a garbage collection. GEN controls the number of generations to garbage collect." #!+(and sb-doc (not gencgc)) "Initiate a garbage collection. GEN may be provided for compatibility with generational garbage collectors, but is ignored in this implementation." (sub-gc :force-p t :gen (if full 6 gen))) ;;;; auxiliary functions (defun bytes-consed-between-gcs () #!+sb-doc "Return the amount of memory that will be allocated before the next garbage collection is initiated. This can be set with SETF." *bytes-consed-between-gcs*) (defun (setf bytes-consed-between-gcs) (val) ;; FIXME: Shouldn't this (and the DECLAIM for the underlying variable) ;; be for a strictly positive number type, e.g. ;; (AND (INTEGER 1) FIXNUM)? (declare (type index val)) (let ((old *bytes-consed-between-gcs*)) (setf *bytes-consed-between-gcs* val) (when *gc-trigger* (setf *gc-trigger* (+ *gc-trigger* (- val old))) (cond ((<= (dynamic-usage) *gc-trigger*) (clear-auto-gc-trigger) (set-auto-gc-trigger *gc-trigger*)) (t ;; FIXME: If SCRUB-CONTROL-STACK is required here, why ;; isn't it built into SUB-GC? And *is* it required here? (sb!sys:scrub-control-stack) (sub-gc))))) val) (defun gc-on () #!+sb-doc "Enable the garbage collector." (setq *gc-inhibit* nil) (when *need-to-collect-garbage* (sub-gc)) nil) (defun gc-off () #!+sb-doc "Disable the garbage collector." (setq *gc-inhibit* t) nil) ;;;; initialization stuff (defun gc-reinit () (when *gc-trigger* (if (< *gc-trigger* (dynamic-usage)) (sub-gc) (set-auto-gc-trigger *gc-trigger*))))