| blob | c9a936ec0d87d87b2e5764e31e6fca408dab7596 |
1 ;;;; the top level interfaces to the compiler, plus some other
2 ;;;; compiler-related stuff (e.g. CL:CALL-ARGUMENTS-LIMIT) which
3 ;;;; doesn't obviously belong anywhere else
5 ;;;; This software is part of the SBCL system. See the README file for
6 ;;;; more information.
7 ;;;;
8 ;;;; This software is derived from the CMU CL system, which was
9 ;;;; written at Carnegie Mellon University and released into the
10 ;;;; public domain. The software is in the public domain and is
11 ;;;; provided with absolutely no warranty. See the COPYING and CREDITS
12 ;;;; files for more information.
16 ;;; FIXME: Doesn't this belong somewhere else, like early-c.lisp?
34 ;;; Whether call of a function which cannot be defined causes a full
35 ;;; warning.
41 ;;; Set to NIL to disable loop analysis for register allocation.
44 ;;; Bind this to a stream to capture various internal debugging output.
47 ;;; The current block compilation state. These are initialized to the
48 ;;; :BLOCK-COMPILE and :ENTRY-POINTS arguments that COMPILE-FILE was
49 ;;; called with.
50 ;;;
51 ;;; *BLOCK-COMPILE-ARG* holds the original value of the :BLOCK-COMPILE
52 ;;; argument, which overrides any internal declarations.
59 ;;; When block compiling, used by PROCESS-FORM to accumulate top level
60 ;;; lambdas resulting from compiling subforms. (In reverse order.)
65 #!+sb-doc
68 #!+sb-doc
71 #!+sb-doc
72 "When this is true, the compiler prints to *ERROR-OUTPUT* progress
73 information about the phases of compilation of each function. (This
77 #!+sb-doc
78 "The defaulted pathname of the file currently being compiled, or NIL if not
81 #!+sb-doc
82 "The TRUENAME of the file currently being compiled, or NIL if not
89 ;;; the SOURCE-INFO structure for the current compilation. This is
90 ;;; null globally to indicate that we aren't currently in any
91 ;;; identifiable compilation.
94 ;;; This is true if we are within a WITH-COMPILATION-UNIT form (which
95 ;;; normally causes nested uses to be no-ops).
98 ;;; Count of the number of compilation units dynamically enclosed by
99 ;;; the current active WITH-COMPILATION-UNIT that were unwound out of.
102 ;;; Mumble conditional on *COMPILE-PROGRESS*.
113 \f
114 ;;;; WITH-COMPILATION-UNIT and WITH-COMPILATION-VALUES
117 #!+sb-doc
118 "WITH-COMPILATION-UNIT ({Key Value}*) Form*
119 This form affects compilations that take place within its dynamic extent. It
120 is intended to be wrapped around the compilation of all files in the same
121 system. These keywords are defined:
122 :OVERRIDE Boolean-Form
123 One of the effects of this form is to delay undefined warnings
124 until the end of the form, instead of giving them at the end of each
125 compilation. If OVERRIDE is NIL (the default), then the outermost
126 WITH-COMPILATION-UNIT form grabs the undefined warnings. Specifying
127 OVERRIDE true causes that form to grab any enclosed warnings, even if
128 it is enclosed by another WITH-COMPILATION-UNIT."
135 ;; Inside another WITH-COMPILATION-UNIT, a WITH-COMPILATION-UNIT is
136 ;; ordinarily (unless OVERRIDE) basically a no-op.
160 ;;; Is FUN-NAME something that no conforming program can rely on
161 ;;; defining as a function?
166 ;;; This is to be called at the end of a compilation unit. It signals
167 ;;; any residual warnings about unknown stuff, then prints the total
168 ;;; error counts. ABORT-P should be true when the compilation unit was
169 ;;; aborted by throwing out. ABORT-COUNT is the number of dynamically
170 ;;; enclosed nested compilation units that were aborted.
191 #+sb-xc-host nil
195 "~@<There is no function named ~S. References to ~S in ~
196 some contexts (like starts of blocks) have special ~
197 meaning, but here it would have to be a function, ~
198 and that shouldn't be right.~:@>"
199 name name))
202 "~@<The ~(~A~) ~S is undefined, and its name is ~
203 reserved by ANSI CL so that even if it were ~
204 defined later, the code doing so would not be ~
205 portable.~:@>"
206 kind name)))
215 "~W more use~:P of undefined ~(~A~) ~S"
216 more kind name)
218 "~W more use~:P of undefined ~(~A~) ~S"
219 more kind name)))))))
228 "~:[This ~(~A~) is~;These ~(~A~)s are~] undefined:~
229 ~% ~{~<~% ~1:;~S~>~^ ~}"
232 "~:[This ~(~A~) is~;These ~(~A~)s are~] undefined:~
233 ~% ~{~<~% ~1:;~S~>~^ ~}"
245 ~[~:;~:*~& caught ~W fatal ERROR condition~:P~]~
246 ~[~:;~:*~& caught ~W ERROR condition~:P~]~
247 ~[~:;~:*~& caught ~W WARNING condition~:P~]~
248 ~[~:;~:*~& caught ~W STYLE-WARNING condition~:P~]~
249 ~[~:;~:*~& printed ~W note~:P~]"
250 abort-p
258 ;;; Evaluate BODY, then return (VALUES BODY-VALUE WARNINGS-P
259 ;;; FAILURE-P), where BODY-VALUE is the first value of the body, and
260 ;;; WARNINGS-P and FAILURE-P are as in CL:COMPILE or CL:COMPILE-FILE.
261 ;;; This also wraps up WITH-IR1-NAMESPACE functionality.
269 \f
270 ;;;; component compilation
273 #!+sb-doc
274 "The upper limit on the number of times that we will consecutively do IR1
275 optimization that doesn't introduce any new code. A finite limit is
281 ;;; Repeatedly optimize COMPONENT until no further optimizations can
282 ;;; be found or we hit our iteration limit. When we hit the limit, we
283 ;;; clear the component and block REOPTIMIZE flags to discourage the
284 ;;; next optimization attempt from pounding on the same code.
328 ;;; KLUDGE: This was bumped from 5 to 10 in a DTC patch ported by MNA
329 ;;; from CMU CL into sbcl-0.6.11.44, the same one which allowed IR1
330 ;;; transforms to be delayed. Either DTC or MNA or both didn't explain
331 ;;; why, and I don't know what the rationale was. -- WHN 2001-04-28
332 ;;;
333 ;;; FIXME: It would be good to document why it's important to have a
334 ;;; large value here, and what the drawbacks of an excessively large
335 ;;; value are; and it might also be good to make it depend on
336 ;;; optimization policy.
342 ;;; Iterate doing FIND-DFO until no new dead code is discovered.
355 ;;; Do all the IR1 phases for a non-top-level component.
381 ;; We delay the generation of type checks until
382 ;; the type constraints have had time to
383 ;; propagate, else the compiler can confuse itself.
413 ;; Assign BLOCK-NUMBER for any cleanup blocks introduced by
414 ;; stack analysis. There shouldn't be any unreachable code after
415 ;; control, so this won't delete anything.
463 #-sb-xc-host
475 code-length
476 trace-table
477 fixup-notes
483 code-length
484 trace-table
485 fixup-notes
489 ;; We're done, so don't bother keeping anything around.
494 ;;; Delete components with no external entry points before we try to
495 ;;; generate code. Unreachable closures can cause IR2 conversion to
496 ;;; puke on itself, since it is the reference to the closure which
497 ;;; normally causes the components to be combined.
504 (:external
512 ;; miscellaneous sanity checks
513 ;;
514 ;; FIXME: These are basically pretty wimpy compared to the checks done
515 ;; by the old CHECK-IR1-CONSISTENCY code. It would be really nice to
516 ;; make those internal consistency checks work again and use them.
521 ;; sanity check to prevent weirdness from propagating insidiously as
522 ;; far from its root cause as it did in bug 138: Make sure that
523 ;; thing-to-COMPONENT links are consistent.
538 #|
550 |#
552 ;; FIXME: What is MAYBE-MUMBLE for? Do we need it any more?
569 \f
570 ;;;; clearing global data structures
571 ;;;;
572 ;;;; FIXME: Is it possible to get rid of this stuff, getting rid of
573 ;;;; global data structures entirely when possible and consing up the
574 ;;;; others from scratch instead of clearing and reusing them?
576 ;;; Clear the INFO in constants in the *FREE-VARS*, etc. In
577 ;;; addition to allowing stuff to be reclaimed, this is required for
578 ;;; correct assignment of constant offsets, since we need to assign a
579 ;;; new offset for each component. We don't clear the FUNCTIONAL-INFO
580 ;;; slots, since they are used to keep track of functions across
581 ;;; component boundaries.
594 ;;; Blow away the REFS for all global variables, and let COMPONENT
595 ;;; be recycled.
607 x))
615 ;;; Clear global variables used by the compiler.
616 ;;;
617 ;;; FIXME: It seems kinda nasty and unmaintainable to have to do this,
618 ;;; and it adds overhead even when people aren't using the compiler.
619 ;;; Perhaps we could make these global vars unbound except when
620 ;;; actually in use, so that this function could go away.
623 ;; Clear global tables.
629 ;; Clear debug counters and tables.
645 ;; (Note: The CMU CL code used to set CL::*GENSYM-COUNTER* to zero here.
646 ;; Superficially, this seemed harmful -- the user could reasonably be
647 ;; surprised if *GENSYM-COUNTER* turned back to zero when something was
648 ;; compiled. A closer inspection showed that this actually turned out to be
649 ;; harmless in practice, because CLEAR-STUFF was only called from within
650 ;; forms which bound CL::*GENSYM-COUNTER* to zero. However, this means that
651 ;; even though zeroing CL::*GENSYM-COUNTER* here turned out to be harmless in
652 ;; practice, it was also useless in practice. So we don't do it any more.)
655 \f
656 ;;;; trace output
658 ;;; Print out some useful info about COMPONENT to STREAM.
679 \f
680 ;;;; file reading
681 ;;;;
682 ;;;; When reading from a file, we have to keep track of some source
683 ;;;; information. We also exploit our ability to back up for printing
684 ;;;; the error context and for recovering from errors.
685 ;;;;
686 ;;;; The interface we provide to this stuff is the stream-oid
687 ;;;; SOURCE-INFO structure. The bookkeeping is done as a side effect
688 ;;;; of getting the next source form.
690 ;;; A FILE-INFO structure holds all the source information for a
691 ;;; given file.
693 ;; If a file, the truename of the corresponding source file. If from
694 ;; a Lisp form, :LISP. If from a stream, :STREAM.
696 ;; the defaulted, but not necessarily absolute file name (i.e. prior
697 ;; to TRUENAME call.) Null if not a file. This is used to set
698 ;; *COMPILE-FILE-PATHNAME*, and if absolute, is dumped in the
699 ;; debug-info.
701 ;; the file's write date (if relevant)
703 ;; the source path root number of the first form in this file (i.e.
704 ;; the total number of forms converted previously in this
705 ;; compilation)
707 ;; parallel vectors containing the forms read out of the file and
708 ;; the file positions that reading of each form started at (i.e. the
709 ;; end of the previous form)
713 ;;; The SOURCE-INFO structure provides a handle on all the source
714 ;;; information for an entire compilation.
716 #-no-ansi-print-object
720 ;; the UT that compilation started at
722 ;; the FILE-INFO structure for this compilation
724 ;; the stream that we are using to read the FILE-INFO, or NIL if
725 ;; no stream has been opened yet
728 ;;; Given a pathname, return a SOURCE-INFO structure.
731 :untruename file
736 ;;; Return a SOURCE-INFO to describe the incremental compilation of FORM.
743 ;;; Return a SOURCE-INFO which will read from STREAM.
749 ;;; Return a form read from STREAM; or for EOF use the trick,
750 ;;; popularized by Kent Pitman, of returning STREAM itself. If an
751 ;;; error happens, then convert it to standard abort-the-compilation
752 ;;; error condition (possibly recording some extra location
753 ;;; information).
758 :condition condition
759 ;; We don't need to supply :POSITION here because
760 ;; READER-ERRORs already know their position in the file.
761 ))
762 ;; ANSI, in its wisdom, says that READ should return END-OF-FILE
763 ;; (and that this is not a READER-ERROR) when it encounters end of
764 ;; file in the middle of something it's trying to read.
767 :condition condition
768 ;; We need to supply :POSITION here because the END-OF-FILE
769 ;; condition doesn't carry the position that the user
770 ;; probably cares about, where the failed READ began.
773 ;;; If STREAM is present, return it, otherwise open a stream to the
774 ;;; current file. There must be a current file.
775 ;;;
776 ;;; FIXME: This is probably an unnecessarily roundabout way to do
777 ;;; things now that we process a single file in COMPILE-FILE (unlike
778 ;;; the old CMU CL code, which accepted multiple files). Also, the old
779 ;;; comment said
780 ;;; When we open a new file, we also reset *PACKAGE* and policy.
781 ;;; This gives the effect of rebinding around each file.
782 ;;; which doesn't seem to be true now. Check to make sure that if
783 ;;; such rebinding is necessary, it's still done somewhere.
793 ;;; Close the stream in INFO if it is open.
801 ;;; Read and compile the source file.
818 nil)))))))
820 ;;; Return the INDEX'th source form read from INFO and the position
821 ;;; where it was read.
827 \f
828 ;;;; processing of top level forms
830 ;;; This is called by top level form processing when we are ready to
831 ;;; actually compile something. If *BLOCK-COMPILE* is T, then we still
832 ;;; convert the form, but delay compilation, pushing the result on
833 ;;; *TOPLEVEL-LAMBDAS* instead.
843 ;;; Macroexpand FORM in the current environment with an error handler.
844 ;;; We only expand one level, so that we retain all the intervening
845 ;;; forms in the source path.
853 condition))))
855 ;;; Process a PROGN-like portion of a top level form. FORMS is a list of
856 ;;; the forms, and PATH is the source path of the FORM they came out of.
857 ;;; COMPILE-TIME-TOO is as in ANSI "3.2.3.1 Processing of Top Level Forms".
863 ;;; Process a top level use of LOCALLY, or anything else (e.g.
864 ;;; MACROLET) at top level which has declarations and ordinary forms.
865 ;;; We parse declarations and then recursively process the body.
871 ;; FIXME: VALUES declaration
872 ;;
873 ;; Binding *POLICY* is pretty much of a hack, since it
874 ;; causes LOCALLY to "capture" enclosed proclamations. It
875 ;; is necessary because CONVERT-AND-MAYBE-COMPILE uses the
876 ;; value of *POLICY* as the policy. The need for this hack
877 ;; is due to the quirk that there is no way to represent in
878 ;; a POLICY that an optimize quality came from the default.
879 ;;
880 ;; FIXME: Ideally, something should be done so that DECLAIM
881 ;; inside LOCALLY works OK. Failing that, at least we could
882 ;; issue a warning instead of silently screwing up.
884 ;; This is probably also a hack
886 ;; ditto
890 ;;; Parse an EVAL-WHEN situations list, returning three flags,
891 ;;; (VALUES COMPILE-TOPLEVEL LOAD-TOPLEVEL EXECUTE), indicating
892 ;;; the types of situations present in the list.
896 '(:compile-toplevel
897 compile
898 :load-toplevel
899 load
900 :execute
901 eval)))
906 deprecated-names)))
908 situations)
913 ;;; utilities for extracting COMPONENTs of FUNCTIONALs
924 result))))
928 result))))
931 &key
932 name
934 ;; I'd thought NIL should
935 ;; work, but it doesn't.
936 ;; -- WHN 2001-09-20
946 definition
948 name))))
958 fun)))
960 ;;; Compile LAMBDA-EXPRESSION into *COMPILE-OBJECT*, returning a
961 ;;; description of the result.
962 ;;; * If *COMPILE-OBJECT* is a CORE-OBJECT, then write the function
963 ;;; into core and return the compiled FUNCTION value.
964 ;;; * If *COMPILE-OBJECT* is a fasl file, then write the function
965 ;;; into the fasl file and return a dump handle.
966 ;;;
967 ;;; If NAME is provided, then we try to use it as the name of the
968 ;;; function for debugging/diagnostic information.
971 &key
972 name
974 ;; This magical idiom seems to be the appropriate
975 ;; path for compiling standalone LAMBDAs, judging
976 ;; from the CMU CL code and experiment, so it's a
977 ;; nice default for things where we don't have a
978 ;; real source path (as in e.g. inside CL:COMPILE).
987 :name name
990 ;; FIXME: The compile-it code from here on is sort of a
991 ;; twisted version of the code in COMPILE-TOPLEVEL. It'd be
992 ;; better to find a way to share the code there; or
993 ;; alternatively, to use this code to replace the code there.
994 ;; (The second alternative might be pretty easy if we used
995 ;; the :LOCALL-ONLY option to IR1-FOR-LAMBDA. Then maybe the
996 ;; whole FUNCTIONAL-KIND=:TOPLEVEL case could go away..)
1018 result
1019 ;; KLUDGE: This code duplicates some other code in this
1020 ;; file. In the great reorganzation, the flow of program
1021 ;; logic changed from the original CMUCL model, and that
1022 ;; path (as of sbcl-0.7.5 in SUB-COMPILE-FILE) was no
1023 ;; longer followed for CORE-OBJECTS, leading to BUG
1024 ;; 156. This place is transparently not the right one for
1025 ;; this code, but I don't have a clear enough overview of
1026 ;; the compiler to know how to rearrange it all so that
1027 ;; this operation fits in nicely, and it was blocking
1028 ;; reimplementation of (DECLAIM (INLINE FOO)) (MACROLET
1029 ;; ((..)) (DEFUN FOO ...))
1030 ;;
1031 ;; FIXME: This KLUDGE doesn't solve all the problem in an
1032 ;; ideal way, as (1) definitions typed in at the REPL
1033 ;; without an INLINE declaration will give a NULL
1034 ;; FUNCTION-LAMBDA-EXPRESSION (allowable, but not ideal)
1035 ;; and (2) INLINE declarations will yield a
1036 ;; FUNCTION-LAMBDA-EXPRESSION headed by
1037 ;; SB-C:LAMBDA-WITH-LEXENV, even for null LEXENV. -- CSR,
1038 ;; 2002-07-02
1039 ;;
1040 ;; (2) is probably fairly easy to fix -- it is, after all,
1041 ;; a matter of list manipulation (or possibly of teaching
1042 ;; CL:FUNCTION about SB-C:LAMBDA-WITH-LEXENV). (1) is
1043 ;; significantly harder, as the association between
1044 ;; function object and source is a tricky one.
1045 ;;
1046 ;; FUNCTION-LAMBDA-EXPRESSION "works" (i.e. returns a
1047 ;; non-NULL list) when the function in question has been
1048 ;; compiled by (COMPILE <x> '(LAMBDA ...)); it does not
1049 ;; work when it has been compiled as part of the top-level
1050 ;; EVAL strategy of compiling everything inside (LAMBDA ()
1051 ;; ...). -- CSR, 2002-11-02
1063 (%compile lambda-expression
1065 :name name
1070 ;;; Process a top level FORM with the specified source PATH.
1071 ;;; * If this is a magic top level form, then do stuff.
1072 ;;; * If this is a macro, then expand it.
1073 ;;; * Otherwise, just compile it.
1074 ;;;
1075 ;;; COMPILE-TIME-TOO is as defined in ANSI
1076 ;;; "3.2.3.1 Processing of Top Level Forms".
1087 path)
1091 ;; When we're cross-compiling, consider: what should we
1092 ;; do when we hit e.g.
1093 ;; (EVAL-WHEN (:COMPILE-TOPLEVEL)
1094 ;; (DEFUN FOO (X) (+ 7 X)))?
1095 ;; DEFUN has a macro definition in the cross-compiler,
1096 ;; and a different macro definition in the target
1097 ;; compiler. The only sensible thing is to use the
1098 ;; target compiler's macro definition, since the
1099 ;; cross-compiler's macro is in general into target
1100 ;; functions which can't meaningfully be executed at
1101 ;; cross-compilation time. So make sure we do the EVAL
1102 ;; here, before we macroexpand.
1103 ;;
1104 ;; Then things get even dicier with something like
1105 ;; (DEFCONSTANT-EQX SB!XC:LAMBDA-LIST-KEYWORDS ..)
1106 ;; where we have to make sure that we don't uncross
1107 ;; the SB!XC: prefix before we do EVAL, because otherwise
1108 ;; we'd be trying to redefine the cross-compilation host's
1109 ;; constants.
1110 ;;
1111 ;; (Isn't it fun to cross-compile Common Lisp?:-)
1112 #+sb-xc-host
1117 ;; like SB!XC:DEFCONSTANT and SB!XC:DEFTYPE
1118 ;; should be equivalent to their CL: counterparts
1119 ;; when being compiled as target code. We leave
1120 ;; the rest of the form uncrossed because macros
1121 ;; might yet expand into EVAL-WHEN stuff, and
1122 ;; things inside EVAL-WHEN can't be uncrossed
1123 ;; until after we've EVALed them in the
1124 ;; cross-compilation host.)
1128 slightly-uncrossed)))
1130 ;; (Now that we're no longer processing toplevel
1131 ;; forms, and hence no longer need to worry about
1132 ;; EVAL-WHEN, we can uncross everything.)
1134 ;; (We have to demote COMPILE-TIME-TOO to NIL
1135 ;; here, no matter what it was before, since
1136 ;; otherwise we'd tend to EVAL subforms more than
1137 ;; once, because of WHEN COMPILE-TIME-TOO form
1138 ;; above.)
1140 ;; When we're not cross-compiling, we only need to
1141 ;; macroexpand once, so we can follow the 1-thru-6
1142 ;; sequence of steps in ANSI's "3.2.3.1 Processing of
1143 ;; Top Level Forms".
1144 #-sb-xc-host
1152 path
1153 compile-time-too))))))
1155 #+sb-xc-host
1156 ;; (There are no xc EVAL-WHEN issues in the ATOM case until
1157 ;; (1) SBCL gets smart enough to handle global
1158 ;; DEFINE-SYMBOL-MACRO or SYMBOL-MACROLET and (2) SBCL
1159 ;; implementors start using symbol macros in a way which
1160 ;; interacts with SB-XC/CL distinction.)
1162 #-sb-xc-host
1168 form))))
1170 ;; In the cross-compiler, top level COLD-FSET arranges
1171 ;; for static linking at cold init time.
1172 #+sb-xc-host
1178 lambda-expression
1179 path)))
1185 ;; CT, LT, and E here are as in Figure 3-7 of ANSI
1186 ;; "3.2.3.1 Processing of Top Level Forms".
1191 e))))
1193 body path new-compile-time-too))
1199 magic
1204 path
1205 compile-time-too))
1209 magic
1213 path
1214 compile-time-too
1224 \f
1225 ;;;; load time value support
1226 ;;;;
1227 ;;;; (See EMIT-MAKE-LOAD-FORM.)
1229 ;;; Return T if we are currently producing a fasl file and hence
1230 ;;; constants need to be dumped carefully.
1234 ;;; Compile FORM and arrange for it to be called at load-time. Return
1235 ;;; the dumper handle and our best guess at the type of the object.
1245 ;;; Compile the FORMS and arrange for them to be called (for effect,
1246 ;;; not value) at load time.
1251 ;;; Do the actual work of COMPILE-LOAD-TIME-VALUE or
1252 ;;; COMPILE-MAKE-LOAD-FORM-INIT-FORMS.
1258 lambda)))
1260 ;;; This is called by COMPILE-TOPLEVEL when it was passed T for
1261 ;;; LOAD-TIME-VALUE-P (which happens in COMPILE-LOAD-TIME-STUFF). We
1262 ;;; don't try to combine this component with anything else and frob
1263 ;;; the name. If not in a :TOPLEVEL component, then don't bother
1264 ;;; compiling, because it was merged with a run-time component.
1273 \f
1274 ;;;; COMPILE-FILE
1284 ;;; Smash LAMBDAS into a single component, compile it, and arrange for
1285 ;;; the resulting function to be called.
1295 ;;; Compile top level code and call the top level lambdas. We pick off
1296 ;;; top level lambdas in non-top-level components here, calling
1297 ;;; SUB-c-t-l-l on each subsequence of normal top level lambdas.
1306 lambdas
1307 ;; this used to read ":start start", but
1308 ;; start can be greater than len, which
1309 ;; is an error according to ANSI - CSR,
1310 ;; 2002-04-25
1312 len)))
1321 ;;; Compile LAMBDAS (a list of CLAMBDAs for top level forms) into the
1322 ;;; object file.
1323 ;;;
1324 ;;; LOAD-TIME-VALUE-P seems to control whether it's MAKE-LOAD-FORM and
1325 ;;; COMPILE-LOAD-TIME-VALUE stuff. -- WHN 20000201
1359 ;;; Actually compile any stuff that has been queued up for block
1360 ;;; compilation.
1378 lexenv))
1382 nil
1398 lexenv))
1408 ;;; Read all forms from INFO and compile them, with output to OBJECT.
1409 ;;; Return (VALUES NIL WARNINGS-P FAILURE-P).
1437 ;; FIXME: Do we need this rebinding here? It's a literal
1438 ;; translation of the old CMU CL rebinding to
1439 ;; (OR *BACKEND-INFO-ENVIRONMENT* *INFO-ENVIRONMENT*),
1440 ;; and it's not obvious whether the rebinding to itself is
1441 ;; needed that SBCL doesn't need *BACKEND-INFO-ENVIRONMENT*.
1458 nil)))
1459 ;; Some errors are sufficiently bewildering that we just fail
1460 ;; immediately, without trying to recover and compile more of
1461 ;; the input file.
1465 "~@<compilation aborted because of fatal error: ~2I~_~A~:>"
1466 condition)
1469 ;;; Return a pathname for the named file. The file must exist.
1478 new
1491 ;;; Print some junk at the beginning and end of compilation.
1499 file-info)
1501 :print-weekday nil
1507 won
1513 ;;; Open some files and call SUB-COMPILE-FILE. If something unwinds
1514 ;;; out of the compile, then abort the writing of the output file, so
1515 ;;; that we don't overwrite it with known garbage.
1518 &key
1520 ;; ANSI options
1522 ;; FIXME: ANSI doesn't seem to say anything about
1523 ;; *COMPILE-VERBOSE* and *COMPILE-PRINT* being rebound by this
1524 ;; function..
1529 ;; extensions
1533 #!+sb-doc
1534 "Compile INPUT-FILE, producing a corresponding fasl file and returning
1535 its filename. Besides the ANSI &KEY arguments :OUTPUT-FILE, :VERBOSE,
1536 :PRINT, and :EXTERNAL-FORMAT,the following extensions are supported:
1537 :TRACE-FILE
1538 If given, internal data structures are dumped to the specified
1539 file, or if a value of T is given, to a file of *.trace type
1540 derived from the input file name.
1541 Also, as a workaround for vaguely-non-ANSI behavior, the :BLOCK-COMPILE
1542 argument is quasi-supported, to determine whether multiple
1543 functions are compiled together as a unit, resolving function
1544 references at compile time. NIL means that global function names
1545 are never resolved at compilation time. Currently NIL is the
1548 circumstances, the compiler's runtime scales badly when it
1549 tries to do this for large files. If/when this performance
1550 problem is fixed, the block compilation default behavior will
1551 probably be made dependent on the SPEED and COMPILATION-SPEED
1552 optimization values, and the :BLOCK-COMPILE argument will probably
1553 become deprecated."
1580 default-trace-file-pathname
1582 default-trace-file-pathname))))
1591 dummy)
1612 ;; Hack around filesystem race condition...
1614 nil)
1615 warnings-p
1616 failure-p)))
1617 \f
1618 ;;; a helper function for COMPILE-FILE-PATHNAME: the default for
1619 ;;; the OUTPUT-FILE argument
1620 ;;;
1621 ;;; ANSI: The defaults for the OUTPUT-FILE are taken from the pathname
1622 ;;; that results from merging the INPUT-FILE with the value of
1623 ;;; *DEFAULT-PATHNAME-DEFAULTS*, except that the type component should
1624 ;;; default to the appropriate implementation-defined default type for
1625 ;;; compiled files.
1629 retyped))
1631 ;;; KLUDGE: Part of the ANSI spec for this seems contradictory:
1632 ;;; If INPUT-FILE is a logical pathname and OUTPUT-FILE is unsupplied,
1633 ;;; the result is a logical pathname. If INPUT-FILE is a logical
1634 ;;; pathname, it is translated into a physical pathname as if by
1635 ;;; calling TRANSLATE-LOGICAL-PATHNAME.
1636 ;;; So I haven't really tried to make this precisely ANSI-compatible
1637 ;;; at the level of e.g. whether it returns logical pathname or a
1638 ;;; physical pathname. Patches to make it more correct are welcome.
1639 ;;; -- WHN 2000-12-09
1641 &key
1643 input-file))
1645 #!+sb-doc
1646 "Return a pathname describing what file COMPILE-FILE would write to given
1647 these arguments."
1649 \f
1650 ;;;; MAKE-LOAD-FORM stuff
1652 ;;; The entry point for MAKE-LOAD-FORM support. When IR1 conversion
1653 ;;; finds a constant structure, it invokes this to arrange for proper
1654 ;;; dumping. If it turns out that the constant has already been
1655 ;;; dumped, then we don't need to do anything.
1656 ;;;
1657 ;;; If the constant hasn't been dumped, then we check to see whether
1658 ;;; we are in the process of creating it. We detect this by
1659 ;;; maintaining the special *CONSTANTS-BEING-CREATED* as a list of all
1660 ;;; the constants we are in the process of creating. Actually, each
1661 ;;; entry is a list of the constant and any init forms that need to be
1662 ;;; processed on behalf of that constant.
1663 ;;;
1664 ;;; It's not necessarily an error for this to happen. If we are
1665 ;;; processing the init form for some object that showed up *after*
1666 ;;; the original reference to this constant, then we just need to
1667 ;;; defer the processing of that init form. To detect this, we
1668 ;;; maintain *CONSTANTS-CREATED-SINCE-LAST-INIT* as a list of the
1669 ;;; constants created since the last time we started processing an
1670 ;;; init form. If the constant passed to emit-make-load-form shows up
1671 ;;; in this list, then there is a circular chain through creation
1672 ;;; forms, which is an error.
1673 ;;;
1674 ;;; If there is some intervening init form, then we blow out of
1675 ;;; processing it by throwing to the tag PENDING-INIT. The value we
1676 ;;; throw is the entry from *CONSTANTS-BEING-CREATED*. This is so the
1677 ;;; offending init form can be tacked onto the init forms for the
1678 ;;; circular object.
1679 ;;;
1680 ;;; If the constant doesn't show up in *CONSTANTS-BEING-CREATED*, then
1681 ;;; we have to create it. We call MAKE-LOAD-FORM and check to see
1682 ;;; whether the creation form is the magic value
1683 ;;; :SB-JUST-DUMP-IT-NORMALLY. If it is, then we don't do anything. The
1684 ;;; dumper will eventually get its hands on the object and use the
1685 ;;; normal structure dumping noise on it.
1686 ;;;
1687 ;;; Otherwise, we bind *CONSTANTS-BEING-CREATED* and
1688 ;;; *CONSTANTS-CREATED-SINCE- LAST-INIT* and compile the creation form
1689 ;;; much the way LOAD-TIME-VALUE does. When this finishes, we tell the
1690 ;;; dumper to use that result instead whenever it sees this constant.
1691 ;;;
1692 ;;; Now we try to compile the init form. We bind
1693 ;;; *CONSTANTS-CREATED-SINCE-LAST-INIT* to NIL and compile the init
1694 ;;; form (and any init forms that were added because of circularity
1695 ;;; detection). If this works, great. If not, we add the init forms to
1696 ;;; the init forms for the object that caused the problems and let it
1697 ;;; deal with it.
1700 ;;; FIXME: Shouldn't these^ variables be unbound outside LET forms?
1704 ;; KLUDGE: This special hack is because I was too lazy
1705 ;; to rework DEF!STRUCT so that the MAKE-LOAD-FORM
1706 ;; function of LAYOUT returns nontrivial forms when
1707 ;; building the cross-compiler but :IGNORE-IT when
1708 ;; cross-compiling or running under the target Lisp. --
1709 ;; WHN 19990914
1722 (:sb-just-dump-it-normally
1724 t)
1725 (:ignore-it
1726 nil)
1743 constant
1745 creation-form)
1747 nil)
1749 constant)))
1755 collect name into names
1756 collect form into forms
1758 nil)))
1763 \f
1764 ;;;; Host compile time definitions
1765 #+sb-xc-host
1770 #+sb-xc-host