| blob | bc93706717157f007ad3bef0f3e34aa31c6e3b9d |
1 ;;;; This file implements local call analysis. A local call is a
2 ;;;; function call between functions being compiled at the same time.
3 ;;;; If we can tell at compile time that such a call is legal, then we
4 ;;;; change the combination to call the correct lambda, mark it as
5 ;;;; local, and add this link to our call graph. Once a call is local,
6 ;;;; it is then eligible for let conversion, which places the body of
7 ;;;; the function inline.
8 ;;;;
9 ;;;; We cannot always do a local call even when we do have the
10 ;;;; function being called. Calls that cannot be shown to have legal
11 ;;;; arg counts are not converted.
13 ;;;; This software is part of the SBCL system. See the README file for
14 ;;;; more information.
15 ;;;;
16 ;;;; This software is derived from the CMU CL system, which was
17 ;;;; written at Carnegie Mellon University and released into the
18 ;;;; public domain. The software is in the public domain and is
19 ;;;; provided with absolutely no warranty. See the COPYING and CREDITS
20 ;;;; files for more information.
24 ;;; This function propagates information from the variables in the
25 ;;; function FUN to the actual arguments in CALL. This is also called
26 ;;; by the VALUES IR1 optimizer when it sleazily converts MV-BINDs to
27 ;;; LETs.
28 ;;;
29 ;;; We flush all arguments to CALL that correspond to unreferenced
30 ;;; variables in FUN. We leave NILs in the COMBINATION-ARGS so that
31 ;;; the remaining args still match up with their vars.
32 ;;;
33 ;;; We also apply the declared variable type assertion to the argument
34 ;;; lvars.
53 collect arg into dx-lvars
55 ;; Stack analysis wants DX value generators to end
56 ;; their blocks. Uses of mupltiple used LVARs already
57 ;; end their blocks, so we just need to process
58 ;; used-once LVARs.
66 :next call
72 :mess-up entry
86 ;;; This function handles merging the tail sets if CALL is potentially
87 ;;; tail-recursive, and is a call to a function with a different
88 ;;; TAIL-SET than CALL's FUN. This must be called whenever we alter
89 ;;; IR1 so as to place a local call in what might be a tail-recursive
90 ;;; context. Note that any call which returns its value to a RETURN is
91 ;;; considered potentially tail-recursive, since any implicit MV-PROG1
92 ;;; might be optimized away.
93 ;;;
94 ;;; We destructively modify the set for the calling function to
95 ;;; represent both, and then change all the functions in callee's set
96 ;;; to reference the first. If we do merge, we reoptimize the
97 ;;; RETURN-RESULT lvar to cause IR1-OPTIMIZE-RETURN to recompute the
98 ;;; tail set type.
112 t)))))
114 ;;; Convert a combination into a local call. We PROPAGATE-TO-ARGS, set
115 ;;; the combination kind to :LOCAL, add FUN to the CALLS of the
116 ;;; function that the call is in, call MERGE-TAIL-SETS, then replace
117 ;;; the function in the REF node with the new function.
118 ;;;
119 ;;; We change the REF last, since changing the reference can trigger
120 ;;; LET conversion of the new function, but will only do so if the
121 ;;; call is local. Note that the replacement may trigger LET
122 ;;; conversion or other changes in IR1. We must call MERGE-TAIL-SETS
123 ;;; with NEW-FUN before the substitution, since after the substitution
124 ;;; (and LET conversion), the call may no longer be recognizable as
125 ;;; tail-recursive.
139 \f
140 ;;;; external entry point creation
142 ;;; Return a LAMBDA form that can be used as the definition of the XEP
143 ;;; for FUN.
144 ;;;
145 ;;; If FUN is a LAMBDA, then we check the number of arguments
146 ;;; (conditional on policy) and call FUN with all the arguments.
147 ;;;
148 ;;; If FUN is an OPTIONAL-DISPATCH, then we dispatch off of the number
149 ;;; of supplied arguments by doing do an = test for each entry-point,
150 ;;; calling the entry with the appropriate prefix of the passed
151 ;;; arguments.
152 ;;;
153 ;;; If there is a &MORE arg, then there are a couple of optimizations
154 ;;; that we make (more for space than anything else):
155 ;;; -- If MIN-ARGS is 0, then we make the more entry a T clause, since
156 ;;; no argument count error is possible.
157 ;;; -- We can omit the = clause for the last entry-point, allowing the
158 ;;; case of 0 more args to fall through to the more entry.
159 ;;;
160 ;;; We don't bother to policy conditionalize wrong arg errors in
161 ;;; optional dispatches, since the additional overhead is negligible
162 ;;; compared to the cost of everything else going on.
163 ;;;
164 ;;; Note that if policy indicates it, argument type declarations in
165 ;;; FUN will be verified. Since nothing is known about the type of the
166 ;;; XEP arg vars, type checks will be emitted when the XEP's arg vars
167 ;;; are passed to the actual function.
190 ;; Force convertion of all entries
193 and n from min
197 ;; FIXME: Make sure that INDEX type distinguishes between
198 ;; target and host. (Probably just make the SB!XC:DEFTYPE
199 ;; different from CL:DEFTYPE.)
215 ;;; Make an external entry point (XEP) for FUN and return it. We
216 ;;; convert the result of MAKE-XEP-LAMBDA in the correct environment,
217 ;;; then associate this lambda with FUN as its XEP. After the
218 ;;; conversion, we iterate over the function's associated lambdas,
219 ;;; redoing local call analysis so that the XEP calls will get
220 ;;; converted.
221 ;;;
222 ;;; We set REANALYZE and REOPTIMIZE in the component, just in case we
223 ;;; discover an XEP after the initial local call analyze pass.
245 res)))
247 ;;; Notice a REF that is not in a local-call context. If the REF is
248 ;;; already to an XEP, then do nothing, otherwise change it to the
249 ;;; XEP, making an XEP if necessary.
250 ;;;
251 ;;; If REF is to a special :CLEANUP or :ESCAPE function, then we treat
252 ;;; it as though it was not an XEP reference (i.e. leave it alone).
260 \f
261 ;;; Attempt to convert all references to FUN to local calls. The
262 ;;; reference must be the function for a call, and the function lvar
263 ;;; must be used only once, since otherwise we cannot be sure what
264 ;;; function is to be called. The call lvar would be multiply used if
265 ;;; there is hairy stuff such as conditionals in the expression that
266 ;;; computes the function.
267 ;;;
268 ;;; If we cannot convert a reference, then we mark the referenced
269 ;;; function as an entry-point, creating a new XEP if necessary. We
270 ;;; don't try to convert calls that are in error (:ERROR kind.)
271 ;;;
272 ;;; This is broken off from LOCALL-ANALYZE-COMPONENT so that people
273 ;;; can force analysis of newly introduced calls. Note that we don't
274 ;;; do LET conversion here.
299 ;;; We examine all NEW-FUNCTIONALS in COMPONENT, attempting to convert
300 ;;; calls into local calls when it is legal. We also attempt to
301 ;;; convert each LAMBDA to a LET. LET conversion is also triggered by
302 ;;; deletion of a function reference, but functions that start out
303 ;;; eligible for conversion must be noticed sometime.
304 ;;;
305 ;;; Note that there is a lot of action going on behind the scenes
306 ;;; here, triggered by reference deletion. In particular, the
307 ;;; COMPONENT-LAMBDAS are being hacked to remove newly deleted and LET
308 ;;; converted LAMBDAs, so it is important that the LAMBDA is added to
309 ;;; the COMPONENT-LAMBDAS when it is. Also, the
310 ;;; COMPONENT-NEW-FUNCTIONALS may contain all sorts of drivel, since
311 ;;; it is not updated when we delete functions, etc. Only
312 ;;; COMPONENT-LAMBDAS is updated.
313 ;;;
314 ;;; COMPONENT-REANALYZE-FUNCTIONALS is treated similarly to
315 ;;; COMPONENT-NEW-FUNCTIONALS, but we don't add lambdas to the
316 ;;; LAMBDAS.
334 ;; Fix/check FUNCTIONAL's relationship to COMPONENT-LAMDBAS.
336 ;; Since FUNCTIONAL isn't a LAMBDA, this doesn't
337 ;; apply: no-op.
340 ; NEW-FUNCTIONALS, hence is new.
341 ;; FUNCTIONAL becomes part of COMPONENT-LAMBDAS now.
346 ;; FUNCTIONAL should be in COMPONENT-LAMBDAS already.
348 component)))))
359 ;; The original CMU CL code seemed to implicitly assume that
360 ;; COMPONENT is the only one here. Let's make that explicit.
371 ;;; If policy is auspicious and CALL is not in an XEP and we don't seem
372 ;;; to be in an infinite recursive loop, then change the reference to
373 ;;; reference a fresh copy. We return whichever function we decide to
374 ;;; reference.
392 original-functional)))))))
397 calls this LET-converted local function:~
398 ~% ~S"
402 expand because expansion refers to ~
403 the optimized away object ~S."
404 losing-local-object)))
411 original-functional)
414 converted-lambda))))
415 original-functional))
417 ;;; Dispatch to the appropriate function to attempt to convert a call.
418 ;;; REF must be a reference to a FUNCTIONAL. This is called in IR1
419 ;;; optimization as well as in local call analysis. If the call is is
420 ;;; already :LOCAL, we do nothing. If the call is already scheduled
421 ;;; for deletion, also do nothing (in addition to saving time, this
422 ;;; also avoids some problems with optimizing collections of functions
423 ;;; that are partially deleted.)
424 ;;;
425 ;;; This is called both before and after FIND-INITIAL-DFO runs. When
426 ;;; called on a :INITIAL component, we don't care whether the caller
427 ;;; and callee are in the same component. Afterward, we must stick
428 ;;; with whatever component division we have chosen.
429 ;;;
430 ;;; Before attempting to convert a call, we see whether the function
431 ;;; is supposed to be inline expanded. Call conversion proceeds as
432 ;;; before after any expansion.
433 ;;;
434 ;;; We bind *COMPILER-ERROR-CONTEXT* to the node for the call so that
435 ;;; warnings will get the right context.
450 component))))
453 original-fun))
471 ;;; Attempt to convert a multiple-value call. The only interesting
472 ;;; case is a call to a function that LOOKS-LIKE-AN-MV-BIND, has
473 ;;; exactly one reference and no XEP, and is called with one values
474 ;;; lvar.
475 ;;;
476 ;;; We change the call to be to the last optional entry point and
477 ;;; change the call to be local. Due to our preconditions, the call
478 ;;; should eventually be converted to a let, but we can't do that now,
479 ;;; since there may be stray references to the e-p lambda due to
480 ;;; optional defaulting code.
481 ;;;
482 ;;; We also use variable types for the called function to construct an
483 ;;; assertion for the values lvar.
484 ;;;
485 ;;; See CONVERT-CALL for additional notes on MERGE-TAIL-SETS, etc.
508 ;;; Attempt to convert a call to a lambda. If the number of args is
509 ;;; wrong, we give a warning and mark the call as :ERROR to remove it
510 ;;; from future consideration. If the argcount is O.K. then we just
511 ;;; convert it.
520 'local-argument-mismatch
521 :format-control
522 "function called with ~R argument~:P, but wants exactly ~R"
525 \f
526 ;;;; &OPTIONAL, &MORE and &KEYWORD calls
528 ;;; This is similar to CONVERT-LAMBDA-CALL, but deals with
529 ;;; OPTIONAL-DISPATCHes. If only fixed args are supplied, then convert
530 ;;; a call to the correct entry point. If &KEY args are supplied, then
531 ;;; dispatch to a subfunction. We don't convert calls to functions
532 ;;; that have a &MORE (or &REST) arg.
541 'local-argument-mismatch
542 :format-control
543 "function called with ~R argument~:P, but wants at least ~R"
555 'local-argument-mismatch
556 :format-control
557 "function called with ~R argument~:P, but wants at most ~R"
558 :format-arguments
563 ;;; This function is used to convert a call to an entry point when
564 ;;; complex transformations need to be done on the original arguments.
565 ;;; ENTRY is the entry point function that we are calling. VARS is a
566 ;;; list of variable names which are bound to the original call
567 ;;; arguments. IGNORES is the subset of VARS which are ignored. ARGS
568 ;;; is the list of arguments to the entry point function.
569 ;;;
570 ;;; In order to avoid gruesome graph grovelling, we introduce a new
571 ;;; function that rearranges the arguments and calls the entry point.
572 ;;; We analyze the new function and the entry point immediately so
573 ;;; that everything gets converted during the single pass.
590 ;;; Use CONVERT-HAIRY-FUN-ENTRY to convert a &MORE-arg call to a known
591 ;;; function into a local call to the MAIN-ENTRY.
592 ;;;
593 ;;; First we verify that all keywords are constant and legal. If there
594 ;;; aren't, then we warn the user and don't attempt to convert the call.
595 ;;;
596 ;;; We massage the supplied &KEY arguments into the order expected
597 ;;; by the main entry. This is done by binding all the arguments to
598 ;;; the keyword call to variables in the introduced lambda, then
599 ;;; passing these values variables in the correct order when calling
600 ;;; the main entry. Unused arguments (such as the keywords themselves)
601 ;;; are discarded simply by not passing them along.
602 ;;;
603 ;;; If there is a &REST arg, then we bundle up the args and pass them
604 ;;; to LIST.
626 (:keyword
689 (:optional
693 (:rest
696 (:keyword
714 \f
715 ;;;; LET conversion
716 ;;;;
717 ;;;; Converting to a LET has differing significance to various parts
718 ;;;; of the compiler:
719 ;;;; -- The body of a LET is spliced in immediately after the
720 ;;;; corresponding combination node, making the control transfer
721 ;;;; explicit and allowing LETs to be mashed together into a single
722 ;;;; block. The value of the LET is delivered directly to the
723 ;;;; original lvar for the call, eliminating the need to
724 ;;;; propagate information from the dummy result lvar.
725 ;;;; -- As far as IR1 optimization is concerned, it is interesting in
726 ;;;; that there is only one expression that the variable can be bound
727 ;;;; to, and this is easily substituted for.
728 ;;;; -- LETs are interesting to environment analysis and to the back
729 ;;;; end because in most ways a LET can be considered to be "the
730 ;;;; same function" as its home function.
731 ;;;; -- LET conversion has dynamic scope implications, since control
732 ;;;; transfers within the same environment are local. In a local
733 ;;;; control transfer, cleanup code must be emitted to remove
734 ;;;; dynamic bindings that are no longer in effect.
736 ;;; Set up the control transfer to the called CLAMBDA. We split the
737 ;;; call block immediately after the call, and link the head of
738 ;;; CLAMBDA to the call block. The successor block after splitting
739 ;;; (where we return to) is returned.
740 ;;;
741 ;;; If the lambda is is a different component than the call, then we
742 ;;; call JOIN-COMPONENTS. This only happens in block compilation
743 ;;; before FIND-INITIAL-DFO.
760 next-block)))
762 ;;; Remove CLAMBDA from the tail set of anything it used to be in the
763 ;;; same set as; but leave CLAMBDA with a valid tail set value of
764 ;;; its own, for the benefit of code which might try to pull
765 ;;; something out of it (e.g. return type).
767 ;; Until sbcl-0.pre7.37.flaky5.2, we did
768 ;; (LET ((TAILS (LAMBDA-TAIL-SET CLAMBDA)))
769 ;; (SETF (TAIL-SET-FUNS TAILS)
770 ;; (DELETE CLAMBDA (TAIL-SET-FUNS TAILS))))
771 ;; (SETF (LAMBDA-TAIL-SET CLAMBDA) NIL)
772 ;; here. Apparently the idea behind the (SETF .. NIL) was that since
773 ;; TAIL-SET-FUNS no longer thinks we're in the tail set, it's
774 ;; inconsistent, and perhaps unsafe, for us to think we're in the
775 ;; tail set. Unfortunately..
776 ;;
777 ;; The (SETF .. NIL) caused problems in sbcl-0.pre7.37.flaky5.2 when
778 ;; I was trying to get Python to emit :EXTERNAL LAMBDAs directly
779 ;; (instead of only being able to emit funny little :TOPLEVEL stubs
780 ;; which you called in order to get the address of an external LAMBDA):
781 ;; the external function was defined in terms of internal function,
782 ;; which was LET-converted, and then things blew up downstream when
783 ;; FINALIZE-XEP-DEFINITION tried to find out its DEFINED-TYPE from
784 ;; the now-NILed-out TAIL-SET. So..
785 ;;
786 ;; To deal with this problem, we no longer NIL out
787 ;; (LAMBDA-TAIL-SET CLAMBDA) here. Instead:
788 ;; * If we're the only function in TAIL-SET-FUNS, it should
789 ;; be safe to leave ourself linked to it, and it to you.
790 ;; * If there are other functions in TAIL-SET-FUNS, then we're
791 ;; afraid of future optimizations on those functions causing
792 ;; the TAIL-SET object no longer to be valid to describe our
793 ;; return value. Thus, we delete ourselves from that object;
794 ;; but we save a newly-allocated tail-set, derived from the old
795 ;; one, for ourselves, for the use of later code (e.g.
796 ;; FINALIZE-XEP-DEFINITION) which might want to
797 ;; know about our return type.
806 ;; The documentation on TAIL-SET-INFO doesn't tell whether it could
807 ;; remain valid in this case, so we nuke it on the theory that
808 ;; missing information tends to be less dangerous than incorrect
809 ;; information.
812 ;;; Handle the PHYSENV semantics of LET conversion. We add CLAMBDA and
813 ;;; its LETs to LETs for the CALL's home function. We merge the calls
814 ;;; for CLAMBDA with the calls for the home function, removing CLAMBDA
815 ;;; in the process. We also merge the ENTRIES.
816 ;;;
817 ;;; We also unlink the function head from the component head and set
818 ;;; COMPONENT-REANALYZE to true to indicate that the DFO should be
819 ;;; recomputed.
839 ;; CLAMBDA belongs to HOME now.
849 ;; All of CLAMBDA's LETs belong to HOME now.
855 ;; CLAMBDA no longer has an independent existence as an entity
856 ;; which has LETs.
859 ;; HOME no longer calls CLAMBDA, and owns all of CLAMBDA's old
860 ;; DFO dependencies.
864 ;; CLAMBDA no longer has an independent existence as an entity
865 ;; which calls things or has DFO dependencies.
868 ;; All of CLAMBDA's ENTRIES belong to HOME now.
872 ;; CLAMBDA no longer has an independent existence as an entity
873 ;; with ENTRIES.
878 ;;; Handle the value semantics of LET conversion. Delete FUN's return
879 ;;; node, and change the control flow to transfer to NEXT-BLOCK
880 ;;; instead. Move all the uses of the result lvar to CALL's lvar.
899 ;; FIXME: Replace the call with unsafe CAST. -- APD, 2003-01-26
906 ;;; We are converting FUN to be a LET when the call is in a non-tail
907 ;;; position. Any previously tail calls in FUN are no longer tail
908 ;;; calls, and must be restored to normal calls which transfer to
909 ;;; NEXT-BLOCK (FUN's return point.) We can't do this by DO-USES on
910 ;;; the RETURN-RESULT, because the return might have been deleted (if
911 ;;; all calls were TR.)
930 ;; The called function might be an assignment in the
931 ;; case where we are currently converting that function.
932 ;; In steady-state, assignments never appear as a called
933 ;; function.
934 (:assignment
938 ;;; Deal with returning from a LET or assignment that we are
939 ;;; converting. FUN is the function we are calling, CALL is a call to
940 ;;; FUN, and NEXT-BLOCK is the return point for a non-tail call, or
941 ;;; NULL if call is a tail call.
942 ;;;
943 ;;; If the call is not a tail call, then we must do
944 ;;; UNCONVERT-TAIL-CALLS, since a tail call is a call which returns
945 ;;; its value out of the enclosing non-let function. When call is
946 ;;; non-TR, we must convert it back to an ordinary local call, since
947 ;;; the value must be delivered to the receiver of CALL's value.
948 ;;;
949 ;;; We do different things depending on whether the caller and callee
950 ;;; have returns left:
952 ;;; -- If the callee has no return we just do MOVE-LET-CALL-CONT.
953 ;;; Either the function doesn't return, or all returns are via
954 ;;; tail-recursive local calls.
955 ;;; -- If CALL is a non-tail call, or if both have returns, then
956 ;;; we delete the callee's return, move its uses to the call's
957 ;;; result lvar, and transfer control to the appropriate
958 ;;; return point.
959 ;;; -- If the callee has a return, but the caller doesn't, then we
960 ;;; move the return to the caller.
989 ;;; Actually do LET conversion. We call subfunctions to do most of the
990 ;;; work. We do REOPTIMIZE-LVAR on the args and CALL's lvar so that
991 ;;; LET-specific IR1 optimizations get a chance. We blow away any
992 ;;; entry for the function in *FREE-FUNS* so that nobody will create
993 ;;; new references to it.
998 nil
999 next-block)))
1003 ;; If CALL has a derive type NIL, it means that "its return" is
1004 ;; unreachable, but the next BIND is still reachable; in order to
1005 ;; not confuse MAYBE-TERMINATE-BLOCK...
1008 ;;; Reoptimize all of CALL's args and its result.
1017 ;;; Are there any declarations in force to say CLAMBDA shouldn't be
1018 ;;; LET converted?
1020 ;; From the user's point of view, LET-converting something that
1021 ;; has a name is inlining it. (The user can't see what we're doing
1022 ;; with anonymous things, and suppressing inlining
1023 ;; for such things can easily give Python acute indigestion, so
1024 ;; we don't.)
1026 ;; ANSI requires that explicit NOTINLINE be respected.
1028 ;; If (= LET-CONVERSION 0) we can guess that inlining
1029 ;; generally won't be appreciated, but if the user
1030 ;; specifically requests inlining, that takes precedence over
1031 ;; our general guess.
1035 ;;; We also don't convert calls to named functions which appear in the
1036 ;;; initial component, delaying this until optimization. This
1037 ;;; minimizes the likelihood that we will LET-convert a function which
1038 ;;; may have references added due to later local inline expansion.
1045 ;;; This function is called when there is some reason to believe that
1046 ;;; CLAMBDA might be converted into a LET. This is done after local
1047 ;;; call analysis, and also when a reference is deleted. We return
1048 ;;; true if we converted.
1053 ;; We only convert to a LET when the function is a normal local
1054 ;; function, has no XEP, and is referenced in exactly one local
1055 ;; call. Conversion is also inhibited if the only reference is in
1056 ;; a block about to be deleted.
1057 ;;
1058 ;; These rules limiting LET conversion may seem unnecessarily
1059 ;; restrictive, since there are some cases where we could do the
1060 ;; return with a jump that don't satisfy these requirements. The
1061 ;; reason for doing things this way is that it makes the concept
1062 ;; of a LET much more useful at the level of IR1 semantics. The
1063 ;; :ASSIGNMENT function kind provides another way to optimize
1064 ;; calls to single-return/multiple call functions.
1065 ;;
1066 ;; We don't attempt to convert calls to functions that have an
1067 ;; XEP, since we might be embarrassed later when we want to
1068 ;; convert a newly discovered local call. Also, see
1069 ;; OK-INITIAL-CONVERT-P.
1086 nil)))
1095 t))))
1096 \f
1097 ;;;; tail local calls and assignments
1099 ;;; Return T if there are no cleanups between BLOCK1 and BLOCK2, or if
1100 ;;; they definitely won't generate any cleanup code. Currently we
1101 ;;; recognize lexical entry points that are only used locally (if at
1102 ;;; all).
1116 ;;; If a potentially TR local call really is TR, then convert it to
1117 ;;; jump directly to the called function. We also call
1118 ;;; MAYBE-CONVERT-TO-ASSIGNMENT. The first value is true if we
1119 ;;; tail-convert. The second is the value of M-C-T-A.
1126 ;; this is a tail call
1130 ;; If the call is in an XEP, we might decide to make it
1131 ;; non-tail so that we can use known return inside the
1132 ;; component.
1144 ;;; This is called when we believe it might make sense to convert
1145 ;;; CLAMBDA to an assignment. All this function really does is
1146 ;;; determine when a function with more than one call can still be
1147 ;;; combined with the calling function's environment. We can convert
1148 ;;; when:
1149 ;;; -- The function is a normal, non-entry function, and
1150 ;;; -- Except for one call, all calls must be tail recursive calls
1151 ;;; in the called function (i.e. are self-recursive tail calls)
1152 ;;; -- OK-INITIAL-CONVERT-P is true.
1153 ;;;
1154 ;;; There may be one outside call, and it need not be tail-recursive.
1155 ;;; Since all tail local calls have already been converted to direct
1156 ;;; transfers, the only control semantics needed are to splice in the
1157 ;;; body at the non-tail call. If there is no non-tail call, then we
1158 ;;; need only merge the environments. Both cases are handled by
1159 ;;; LET-CONVERT.
1160 ;;;
1161 ;;; ### It would actually be possible to allow any number of outside
1162 ;;; calls as long as they all return to the same place (i.e. have the
1163 ;;; same conceptual continuation.) A special case of this would be
1164 ;;; when all of the outside calls are tail recursive.
1191 t)
1193 nil))))))