3 Bugs can be reported on the help mailing list
4 sbcl-help@lists.sourceforge.net
5 or on the development mailing list
6 sbcl-devel@lists.sourceforge.net
8 Please include enough information in a bug report that someone reading
9 it can reproduce the problem, i.e. don't write
10 Subject: apparent bug in PRINT-OBJECT (or *PRINT-LENGTH*?)
11 PRINT-OBJECT doesn't seem to work with *PRINT-LENGTH*. Is this a bug?
13 Subject: apparent bug in PRINT-OBJECT (or *PRINT-LENGTH*?)
14 In sbcl-1.2.3 running under OpenBSD 4.5 on my Alpha box, when
15 I compile and load the file
16 (DEFSTRUCT (FOO (:PRINT-OBJECT (LAMBDA (X Y)
17 (LET ((*PRINT-LENGTH* 4))
20 then at the command line type
22 the program loops endlessly instead of printing the object.
27 There is also some information on bugs in the manual page and
28 in the TODO file. Eventually more such information may move here.
30 The gaps in the number sequence belong to old bug descriptions which
31 have gone away (typically because they were fixed, but sometimes for
32 other reasons, e.g. because they were moved elsewhere).
36 DEFSTRUCT almost certainly should overwrite the old LAYOUT information
37 instead of just punting when a contradictory structure definition
38 is loaded. As it is, if you redefine DEFSTRUCTs in a way which
39 changes their layout, you probably have to rebuild your entire
40 program, even if you know or guess enough about the internals of
41 SBCL to wager that this (undefined in ANSI) operation would be safe.
43 3: "type checking of structure slots"
45 ANSI specifies that a type mismatch in a structure slot
46 initialization value should not cause a warning.
48 This one might not be fixed for a while because while we're big
49 believers in ANSI compatibility and all, (1) there's no obvious
50 simple way to do it (short of disabling all warnings for type
51 mismatches everywhere), and (2) there's a good portable
52 workaround, and (3) by their own reasoning, it looks as though
53 ANSI may have gotten it wrong. ANSI justifies this specification
55 The restriction against issuing a warning for type mismatches
56 between a slot-initform and the corresponding slot's :TYPE
57 option is necessary because a slot-initform must be specified
58 in order to specify slot options; in some cases, no suitable
60 However, in SBCL (as in CMU CL or, for that matter, any compiler
61 which really understands Common Lisp types) a suitable default
62 does exist, in all cases, because the compiler understands the
63 concept of functions which never return (i.e. has return type NIL).
64 Thus, as a portable workaround, you can use a call to some
65 known-never-to-return function as the default. E.g.
67 (BAR (ERROR "missing :BAR argument")
68 :TYPE SOME-TYPE-TOO-HAIRY-TO-CONSTRUCT-AN-INSTANCE-OF))
70 (DECLAIM (FTYPE (FUNCTION () NIL) MISSING-ARG))
71 (DEFUN REQUIRED-ARG () ; workaround for SBCL non-ANSI slot init typing
72 (ERROR "missing required argument"))
74 (BAR (REQUIRED-ARG) :TYPE TRICKY-TYPE-OF-SOME-SORT)
75 (BLETCH (REQUIRED-ARG) :TYPE TRICKY-TYPE-OF-SOME-SORT)
76 (N-REFS-SO-FAR 0 :TYPE (INTEGER 0)))
77 Such code should compile without complaint and work correctly either
78 on SBCL or on any other completely compliant Common Lisp system.
80 b: &AUX argument in a boa-constructor without a default value means
81 "do not initilize this slot" and does not cause type error. But
82 an error may be signalled at read time and it would be good if
88 The "compiling top-level form:" output ought to be condensed.
89 Perhaps any number of such consecutive lines ought to turn into a
90 single "compiling top-level forms:" line.
93 It would be nice if the
95 (during macroexpansion)
96 said what macroexpansion was at fault, e.g.
98 (during macroexpansion of IN-PACKAGE,
99 during macroexpansion of DEFFOO)
102 (I *think* this is a bug. It certainly seems like strange behavior. But
103 the ANSI spec is scary, dark, and deep.. -- WHN)
104 (FORMAT NIL "~,1G" 1.4) => "1. "
105 (FORMAT NIL "~3,1G" 1.4) => "1. "
108 Sometimes (SB-EXT:QUIT) fails with
109 Argh! maximum interrupt nesting depth (4096) exceeded, exiting
110 Process inferior-lisp exited abnormally with code 1
111 I haven't noticed a repeatable case of this yet.
114 The printer doesn't report closures very well. This is true in
118 #<Closure Over Function "DEFUN STRUCTURE-SLOT-ACCESSOR" {134D1A1}>
119 It would be nice to make closures have a settable name slot,
120 and make things like DEFSTRUCT and FLET, which create closures,
121 set helpful values into this slot.
124 And as long as we're wishing, it would be awfully nice if INSPECT could
125 also report on closures, telling about the values of the bound variables.
128 The compiler assumes that any time a function of declared FTYPE
129 doesn't signal an error, its arguments were of the declared type.
130 E.g. compiling and loading
131 (DECLAIM (OPTIMIZE (SAFETY 3)))
132 (DEFUN FACTORIAL (X) (GAMMA (1+ X)))
134 (DECLAIM (FTYPE (FUNCTION (UNSIGNED-BYTE)) FACTORIAL))
136 (COND ((> (FACTORIAL X) 1.0E6)
137 (FORMAT T "too big~%"))
139 (FORMAT T "exactly ~S~%" (FACTORIAL X)))
141 (FORMAT T "approximately ~S~%" (FACTORIAL X)))))
144 will cause the INTEGERP case to be selected, giving bogus output a la
146 This violates the "declarations are assertions" principle.
147 According to the ANSI spec, in the section "System Class FUNCTION",
148 this is a case of "lying to the compiler", but the lying is done
149 by the code which calls FACTORIAL with non-UNSIGNED-BYTE arguments,
150 not by the unexpectedly general definition of FACTORIAL. In any case,
151 "declarations are assertions" means that lying to the compiler should
152 cause an error to be signalled, and should not cause a bogus
153 result to be returned. Thus, the compiler should not assume
154 that arbitrary functions check their argument types. (It might
155 make sense to add another flag (CHECKED?) to DEFKNOWN to
156 identify functions which *do* check their argument types.)
157 (Also, verify that the compiler handles declared function
158 return types as assertions.)
161 The definitions of SIGCONTEXT-FLOAT-REGISTER and
162 %SET-SIGCONTEXT-FLOAT-REGISTER in x86-vm.lisp say they're not
163 supported on FreeBSD because the floating point state is not saved,
164 but at least as of FreeBSD 4.0, the floating point state *is* saved,
165 so they could be supported after all. Very likely
166 SIGCONTEXT-FLOATING-POINT-MODES could now be supported, too.
169 a slew of floating-point-related errors reported by Peter Van Eynde
171 c: Many expressions generate floating infinity on x86/Linux:
176 PVE's regression tests want them to raise errors. sbcl-0.7.0.5
177 on x86/Linux generates the infinities instead. That might or
178 might not be conforming behavior, but it's also inconsistent,
179 which is almost certainly wrong. (Inconsistency: (/ 1 0.0)
180 should give the same result as (/ 1.0 0.0), but instead (/ 1 0.0)
181 generates SINGLE-FLOAT-POSITIVE-INFINITY and (/ 1.0 0.0)
183 d: (in section12.erg) various forms a la
184 (FLOAT 1 DOUBLE-FLOAT-EPSILON)
185 don't give the right behavior.
188 The debugger LIST-LOCATIONS command doesn't work properly.
189 (How should it work properly?)
192 Compiling and loading
193 (DEFUN FAIL (X) (THROW 'FAIL-TAG X))
195 then requesting a BACKTRACE at the debugger prompt gives no information
196 about where in the user program the problem occurred.
199 Using the pretty-printer from the command prompt gives funny
200 results, apparently because the pretty-printer doesn't know
201 about user's command input, including the user's carriage return
202 that the user, and therefore the pretty-printer thinks that
203 the new output block should start indented 2 or more characters
204 rightward of the correct location.
207 As reported by Winton Davies on a CMU CL mailing list 2000-01-10,
208 and reported for SBCL by Martin Atzmueller 2000-10-20: (TRACE GETHASH)
209 crashes SBCL. In general tracing anything which is used in the
210 implementation of TRACE is likely to have the same problem.
213 ANSI says in one place that type declarations can be abbreviated even
214 when the type name is not a symbol, e.g.
215 (DECLAIM ((VECTOR T) *FOOVECTOR*))
216 SBCL doesn't support this. But ANSI says in another place that this
217 isn't allowed. So it's not clear this is a bug after all. (See the
218 e-mail on cmucl-help@cons.org on 2001-01-16 and 2001-01-17 from WHN
222 as pointed out by Dan Barlow on sbcl-devel 2000-07-02:
223 The PICK-TEMPORARY-FILE-NAME utility used by LOAD-FOREIGN uses
224 an easily guessable temporary filename in a way which might open
225 applications using LOAD-FOREIGN to hijacking by malicious users
226 on the same machine. Incantations for doing this safely are
227 floating around the net in various "how to write secure programs
228 despite Unix" documents, and it would be good to (1) fix this in
229 LOAD-FOREIGN, and (2) hunt for any other code which uses temporary
230 files and make it share the same new safe logic.
232 (partially alleviated in sbcl-0.7.9.32 by a fix by Matthew Danish to
233 make the temporary filename less easily guessable)
236 RANDOM-INTEGER-EXTRA-BITS=10 may not be large enough for the RANDOM
237 RNG to be high quality near RANDOM-FIXNUM-MAX; it looks as though
238 the mean of the distribution can be systematically O(0.1%) wrong.
239 Just increasing R-I-E-B is probably not a good solution, since
240 it would decrease efficiency more than is probably necessary. Perhaps
241 using some sort of accept/reject method would be better.
244 Internally the compiler sometimes evaluates
245 (sb-kernel:type/= (specifier-type '*) (specifier-type t))
246 (I stumbled across this when I added an
247 (assert (not (eq type1 *wild-type*)))
248 in the NAMED :SIMPLE-= type method.) '* isn't really a type, and
249 in a type context should probably be translated to T, and so it's
250 probably wrong to ask whether it's equal to the T type and then (using
251 the EQ type comparison in the NAMED :SIMPLE-= type method) return NIL.
252 (I haven't tried to investigate this bug enough to guess whether
253 there might be any user-level symptoms.)
255 In fact, the type system is likely to depend on this inequality not
256 holding... * is not equivalent to T in many cases, such as
257 (VECTOR *) /= (VECTOR T).
260 The facility for dumping a running Lisp image to disk gets confused
261 when run without the PURIFY option, and creates an unnecessarily large
262 core file (apparently representing memory usage up to the previous
263 high-water mark). Moreover, when the file is loaded, it confuses the
264 GC, so that thereafter memory usage can never be reduced below that
267 (As of 0.8.7.3 it's likely that the latter half of this bug is fixed.
268 The interaction between gencgc and the variables used by
269 save-lisp-and-die is still nonoptimal, though, so no respite from
273 In sbcl-0.6.11.41 (and in all earlier SBCL, and in CMU
274 CL), out-of-line structure slot setters are horribly inefficient
275 whenever the type of the slot is declared, because out-of-line
276 structure slot setters are implemented as closures to save space,
277 so the compiler doesn't compile the type test into code, but
278 instead just saves the type in a lexical closure and interprets it
280 To exercise the problem, compile and load
281 (cl:in-package :cl-user)
283 (bar (error "missing") :type bar))
286 (loop (setf (foo-bar *foo*) x)))
288 (defvar *bar* (make-bar))
289 (defvar *foo* (make-foo :bar *bar*))
290 (defvar *setf-foo-bar* #'(setf foo-bar))
292 (loop (funcall *setf-foo-bar* x *foo*)))
293 then run (WASTREL1 *BAR*) or (WASTREL2 *BAR*), hit Ctrl-C, and
294 use BACKTRACE, to see it's spending all essentially all its time
295 in %TYPEP and VALUES-SPECIFIER-TYPE and so forth.
296 One possible solution would be simply to give up on
297 representing structure slot accessors as functions, and represent
298 them as macroexpansions instead. This can be inconvenient for users,
299 but it's not clear that it's worse than trying to help by expanding
300 into a horribly inefficient implementation.
301 As a workaround for the problem, #'(SETF FOO) expressions
302 can be replaced with (EFFICIENT-SETF-FUNCTION FOO), where
303 (defmacro efficient-setf-function (place-function-name)
304 (or #+sbcl (and (sb-int:info :function :accessor-for place-function-name)
305 ;; a workaround for the problem, encouraging the
306 ;; inline expansion of the structure accessor, so
307 ;; that the compiler can optimize its type test
308 (let ((new-value (gensym "NEW-VALUE-"))
309 (structure-value (gensym "STRUCTURE-VALUE-")))
310 `(lambda (,new-value ,structure-value)
311 (setf (,place-function-name ,structure-value)
313 ;; no problem, can just use the ordinary expansion
314 `(function (setf ,place-function-name))))
317 There's apparently a bug in CEILING optimization which caused
318 Douglas Crosher to patch the CMU CL version. Martin Atzmueller
319 applied the patches to SBCL and they didn't seem to cause problems
320 (as reported sbcl-devel 2001-05-04). However, since the patches
321 modify nontrivial code which was apparently written incorrectly
322 the first time around, until regression tests are written I'm not
323 comfortable merging the patches in the CVS version of SBCL.
326 (TIME (ROOM T)) reports more than 200 Mbytes consed even for
327 a clean, just-started SBCL system. And it seems to be right:
328 (ROOM T) can bring a small computer to its knees for a *long*
329 time trying to GC afterwards. Surely there's some more economical
330 way to implement (ROOM T).
332 Daniel Barlow doesn't know what fixed this, but observes that it
333 doesn't seem to be the case in 0.8.7.3 any more. Instead, (ROOM T)
334 in a fresh SBCL causes
336 debugger invoked on a SB-INT:BUG in thread 5911:
337 failed AVER: "(SAP= CURRENT END)"
339 unless a GC has happened beforehand.
342 When the compiler inline expands functions, it may be that different
343 kinds of return values are generated from different code branches.
344 E.g. an inline expansion of POSITION generates integer results
345 from one branch, and NIL results from another. When that inline
346 expansion is used in a context where only one of those results
349 (aref *a1* (position x *a2*)))
350 and the compiler can't prove that the unacceptable branch is
351 never taken, then bogus type mismatch warnings can be generated.
352 If you need to suppress the type mismatch warnings, you can
353 suppress the inline expansion,
355 #+sbcl (declare (notinline position)) ; to suppress bug 117 bogowarnings
356 (aref *a1* (position x *a2*)))
357 or, sometimes, suppress them by declaring the result to be of an
360 (aref *a1* (the integer (position x *a2*))))
362 This is not a new compiler problem in 0.7.0, but the new compiler
363 transforms for FIND, POSITION, FIND-IF, and POSITION-IF make it
364 more conspicuous. If you don't need performance from these functions,
365 and the bogus warnings are a nuisance for you, you can return to
366 your pre-0.7.0 state of grace with
367 #+sbcl (declaim (notinline find position find-if position-if)) ; bug 117..
372 as reported by Eric Marsden on cmucl-imp@cons.org 2001-08-14:
373 (= (FLOAT 1 DOUBLE-FLOAT-EPSILON)
374 (+ (FLOAT 1 DOUBLE-FLOAT-EPSILON) DOUBLE-FLOAT-EPSILON)) => T
375 when of course it should be NIL. (He says it only fails for X86,
376 not SPARC; dunno about Alpha.)
378 Also, "the same problem exists for LONG-FLOAT-EPSILON,
379 DOUBLE-FLOAT-NEGATIVE-EPSILON, LONG-FLOAT-NEGATIVE-EPSILON (though
380 for the -negative- the + is replaced by a - in the test)."
382 Raymond Toy comments that this is tricky on the X86 since its FPU
383 uses 80-bit precision internally.
385 Bruno Haible comments:
386 The values are those that are expected for an IEEE double-float
387 arithmetic. The problem appears to be that the rounding is not
388 IEEE on x86 compliant: namely, values are first rounded to 64
389 bits mantissa precision, then only to 53 bits mantissa
390 precision. This gives different results than rounding to 53 bits
391 mantissa precision in a single step.
393 The quick "fix", to permanently change the FPU control word from
394 0x037f to 0x027f, will give problems with the fdlibm code that is
395 used for computing transcendental functions like sinh() etc.
396 so maybe we need to change the FPU control word to that for Lisp
397 code, and adjust it to the safe 0x037f for calls to C?
400 As of version 0.pre7.14, SBCL's implementation of MACROLET makes
401 the entire lexical environment at the point of MACROLET available
402 in the bodies of the macroexpander functions. In particular, it
403 allows the function bodies (which run at compile time) to try to
404 access lexical variables (which are only defined at runtime).
405 It doesn't even issue a warning, which is bad.
407 The SBCL behavior arguably conforms to the ANSI spec (since the
408 spec says that the behavior is undefined, ergo anything conforms).
409 However, it would be better to issue a compile-time error.
410 Unfortunately I (WHN) don't see any simple way to detect this
411 condition in order to issue such an error, so for the meantime
412 SBCL just does this weird broken "conforming" thing.
414 The ANSI standard says, in the definition of the special operator
416 The macro-expansion functions defined by MACROLET are defined
417 in the lexical environment in which the MACROLET form appears.
418 Declarations and MACROLET and SYMBOL-MACROLET definitions affect
419 the local macro definitions in a MACROLET, but the consequences
420 are undefined if the local macro definitions reference any
421 local variable or function bindings that are visible in that
423 Then it seems to contradict itself by giving the example
425 (macrolet ((fudge (z)
426 ;The parameters x and flag are not accessible
427 ; at this point; a reference to flag would be to
428 ; the global variable of that name.
429 ` (if flag (* ,z ,z) ,z)))
430 ;The parameters x and flag are accessible here.
434 The comment "a reference to flag would be to the global variable
435 of the same name" sounds like good behavior for the system to have.
436 but actual specification quoted above says that the actual behavior
439 (Since 0.7.8.23 macroexpanders are defined in a restricted version
440 of the lexical environment, containing no lexical variables and
441 functions, which seems to conform to ANSI and CLtL2, but signalling
442 a STYLE-WARNING for references to variables similar to locals might
446 (as reported by Gabe Garza on cmucl-help 2001-09-21)
448 (defun test-pred (x y)
452 (func (lambda () x)))
453 (print (eq func func))
454 (print (test-pred func func))
455 (delete func (list func))))
456 Now calling (TEST-CASE) gives output
459 (#<FUNCTION {500A9EF9}>)
460 Evidently Python thinks of the lambda as a code transformation so
461 much that it forgets that it's also an object.
464 Ideally, uninterning a symbol would allow it, and its associated
465 FDEFINITION and PROCLAIM data, to be reclaimed by the GC. However,
466 at least as of sbcl-0.7.0, this isn't the case. Information about
467 FDEFINITIONs and PROCLAIMed properties is stored in globaldb.lisp
468 essentially in ordinary (non-weak) hash tables keyed by symbols.
469 Thus, once a system has an entry in this system, it tends to live
470 forever, even when it is uninterned and all other references to it
473 141: "pretty printing and backquote"
476 ``(FOO SB-IMPL::BACKQ-COMMA-AT S)
478 c. (reported by Paul F. Dietz)
480 `(LAMBDA (SB-IMPL::BACKQ-COMMA X))
483 (reported by Jesse Bouwman 2001-10-24 through the unfortunately
484 prominent SourceForge web/db bug tracking system, which is
485 unfortunately not a reliable way to get a timely response from
486 the SBCL maintainers)
487 In the course of trying to build a test case for an
488 application error, I encountered this behavior:
489 If you start up sbcl, and then lay on CTRL-C for a
490 minute or two, the lisp process will eventually say:
491 %PRIMITIVE HALT called; the party is over.
492 and throw you into the monitor. If I start up lisp,
493 attach to the process with strace, and then do the same
494 (abusive) thing, I get instead:
495 access failure in heap page not marked as write-protected
496 and the monitor again. I don't know enough to have the
497 faintest idea of what is going on here.
498 This is with sbcl 6.12, uname -a reports:
499 Linux prep 2.2.19 #4 SMP Tue Apr 24 13:59:52 CDT 2001 i686 unknown
500 I (WHN) have verified that the same thing occurs on sbcl-0.pre7.141
501 under OpenBSD 2.9 on my X86 laptop. Do be patient when you try it:
502 it took more than two minutes (but less than five) for me.
506 ANSI allows types `(COMPLEX ,FOO) to use very hairy values for
507 FOO, e.g. (COMPLEX (AND REAL (SATISFIES ODDP))). The old CMU CL
508 COMPLEX implementation didn't deal with this, and hasn't been
509 upgraded to do so. (This doesn't seem to be a high priority
510 conformance problem, since seems hard to construct useful code
513 b. (fixed in 0.8.3.43)
516 Floating point errors are reported poorly. E.g. on x86 OpenBSD
519 debugger invoked on condition of type SB-KERNEL:FLOATING-POINT-EXCEPTION:
520 An arithmetic error SB-KERNEL:FLOATING-POINT-EXCEPTION was signalled.
521 No traps are enabled? How can this be?
522 It should be possible to be much more specific (overflow, division
523 by zero, etc.) and of course the "How can this be?" should be fixable.
525 See also bugs #45.c and #183
528 (reported by Robert E. Brown 2002-04-16)
529 When a function is called with too few arguments, causing the
530 debugger to be entered, the uninitialized slots in the bad call frame
531 seem to cause GCish problems, being interpreted as tagged data even
532 though they're not. In particular, executing ROOM in the
533 debugger at that point causes AVER failures:
536 * (lisp-implementation-version)
542 failed AVER: "(SAP= CURRENT END)"
543 (Christophe Rhodes reports that this doesn't occur on the SPARC, which
544 isn't too surprising since there are many differences in stack
545 implementation and GC conservatism between the X86 and other ports.)
547 This is probably the same bug as 216
550 In sbcl-0.7.3.11, compiling the (illegal) code
551 (in-package :cl-user)
552 (defmethod prove ((uustk uustk))
555 gives the (not terribly clear) error message
557 ; (during macroexpansion of (DEFMETHOD PROVE ...))
558 ; can't get template for (FROB NIL NIL)
559 The problem seems to be that the code walker used by the DEFMETHOD
560 macro is unhappy with the illegal syntax in the method body, and
561 is giving an unclear error message.
564 The compiler sometimes tries to constant-fold expressions before
565 it checks to see whether they can be reached. This can lead to
566 bogus warnings about errors in the constant folding, e.g. in code
569 (WRITE-STRING (> X 0) "+" "0"))
570 compiled in a context where the compiler can prove that X is NIL,
571 and the compiler complains that (> X 0) causes a type error because
572 NIL isn't a valid argument to #'>. Until sbcl-0.7.4.10 or so this
573 caused a full WARNING, which made the bug really annoying because then
574 COMPILE and COMPILE-FILE returned FAILURE-P=T for perfectly legal
575 code. Since then the warning has been downgraded to STYLE-WARNING,
576 so it's still a bug but at least it's a little less annoying.
578 183: "IEEE floating point issues"
579 Even where floating point handling is being dealt with relatively
580 well (as of sbcl-0.7.5, on sparc/sunos and alpha; see bug #146), the
581 accrued-exceptions and current-exceptions part of the fp control
582 word don't seem to bear much relation to reality. E.g. on
586 debugger invoked on condition of type DIVISION-BY-ZERO:
587 arithmetic error DIVISION-BY-ZERO signalled
588 0] (sb-vm::get-floating-point-modes)
590 (:TRAPS (:OVERFLOW :INVALID :DIVIDE-BY-ZERO)
591 :ROUNDING-MODE :NEAREST
592 :CURRENT-EXCEPTIONS NIL
593 :ACCRUED-EXCEPTIONS (:INEXACT)
596 * (sb-vm::get-floating-point-modes)
597 (:TRAPS (:OVERFLOW :INVALID :DIVIDE-BY-ZERO)
598 :ROUNDING-MODE :NEAREST
599 :CURRENT-EXCEPTIONS (:INEXACT)
600 :ACCRUED-EXCEPTIONS (:INEXACT)
603 188: "compiler performance fiasco involving type inference and UNION-TYPE"
607 (declare (optimize (safety 3)))
608 (declare (optimize (compilation-speed 2)))
609 (declare (optimize (speed 1) (debug 1) (space 1)))
611 (declare (type (integer 0) start))
612 (print (incf start 22))
613 (print (incf start 26))
614 (print (incf start 28)))
616 (declare (type (integer 0) start))
617 (print (incf start 22))
618 (print (incf start 26)))
620 (declare (type (integer 0) start))
621 (print (incf start 22))
622 (print (incf start 26))))))
624 This example could be solved with clever enough constraint
625 propagation or with SSA, but consider
630 The careful type of X is {2k} :-(. Is it really important to be
631 able to work with unions of many intervals?
633 191: "Miscellaneous PCL deficiencies"
634 (reported by Alexey Dejneka sbcl-devel 2002-08-04)
635 a. DEFCLASS does not inform the compiler about generated
636 functions. Compiling a file with
640 (WITH-SLOTS (A-CLASS-X) A
642 results in a STYLE-WARNING:
644 SB-SLOT-ACCESSOR-NAME::|COMMON-LISP-USER A-CLASS-X slot READER|
646 APD's fix for this was checked in to sbcl-0.7.6.20, but Pierre
647 Mai points out that the declamation of functions is in fact
648 incorrect in some cases (most notably for structure
649 classes). This means that at present erroneous attempts to use
650 WITH-SLOTS and the like on classes with metaclass STRUCTURE-CLASS
651 won't get the corresponding STYLE-WARNING.
652 c. (fixed in 0.8.4.23)
654 201: "Incautious type inference from compound types"
655 a. (reported by APD sbcl-devel 2002-09-17)
657 (LET ((Y (CAR (THE (CONS INTEGER *) X))))
659 (FORMAT NIL "~S IS ~S, Y = ~S"
666 (FOO ' (1 . 2)) => "NIL IS INTEGER, Y = 1"
670 (declare (type (array * (4 4)) x))
672 (setq x (make-array '(4 4)))
673 (adjust-array y '(3 5))
674 (= (array-dimension y 0) (eval `(array-dimension ,y 0)))))
676 * (foo (make-array '(4 4) :adjustable t))
679 205: "environment issues in cross compiler"
680 (These bugs have no impact on user code, but should be fixed or
682 a. Macroexpanders introduced with MACROLET are defined in the null
684 b. The body of (EVAL-WHEN (:COMPILE-TOPLEVEL) ...) is evaluated in
685 the null lexical environment.
686 c. The cross-compiler cannot inline functions defined in a non-null
689 206: ":SB-FLUID feature broken"
690 (reported by Antonio Martinez-Shotton sbcl-devel 2002-10-07)
691 Enabling :SB-FLUID in the target-features list in sbcl-0.7.8 breaks
694 207: "poorly distributed SXHASH results for compound data"
695 SBCL's SXHASH could probably try a little harder. ANSI: "the
696 intent is that an implementation should make a good-faith
697 effort to produce hash-codes that are well distributed
698 within the range of non-negative fixnums". But
699 (let ((hits (make-hash-table)))
702 (let* ((ij (cons i j))
703 (newlist (push ij (gethash (sxhash ij) hits))))
705 (format t "~&collision: ~S~%" newlist))))))
706 reports lots of collisions in sbcl-0.7.8. A stronger MIX function
707 would be an obvious way of fix. Maybe it would be acceptably efficient
708 to redo MIX using a lookup into a 256-entry s-box containing
709 29-bit pseudorandom numbers?
711 211: "keywords processing"
712 a. :ALLOW-OTHER-KEYS T should allow a function to receive an odd
713 number of keyword arguments.
716 (flet ((foo (&key y) (list y)))
717 (list (foo :y 1 :y 2)))
719 issues confusing message
724 ; caught STYLE-WARNING:
725 ; The variable #:G15 is defined but never used.
727 212: "Sequence functions and circular arguments"
728 COERCE, MERGE and CONCATENATE go into an infinite loop when given
729 circular arguments; it would be good for the user if they could be
730 given an error instead (ANSI 17.1.1 allows this behaviour on the part
731 of the implementation, as conforming code cannot give non-proper
732 sequences to these functions. MAP also has this problem (and
733 solution), though arguably the convenience of being able to do
734 (MAP 'LIST '+ FOO '#1=(1 . #1#))
735 might be classed as more important (though signalling an error when
736 all of the arguments are circular is probably desireable).
738 213: "Sequence functions and type checking"
739 b. MAP, when given a type argument that is SUBTYPEP LIST, does not
740 check that it will return a sequence of the given type. Fixing
741 it along the same lines as the others (cf. work done around
742 sbcl-0.7.8.45) is possible, but doing so efficiently didn't look
743 entirely straightforward.
744 c. All of these functions will silently accept a type of the form
746 whether or not the return value is of this type. This is
747 probably permitted by ANSI (see "Exceptional Situations" under
748 ANSI MAKE-SEQUENCE), but the DERIVE-TYPE mechanism does not
749 know about this escape clause, so code of the form
750 (INTEGERP (CAR (MAKE-SEQUENCE '(CONS INTEGER *) 2)))
751 can erroneously return T.
753 215: ":TEST-NOT handling by functions"
754 a. FIND and POSITION currently signal errors when given non-NIL for
755 both their :TEST and (deprecated) :TEST-NOT arguments, but by
756 ANSI 17.2 "the consequences are unspecified", which by ANSI 1.4.2
757 means that the effect is "unpredictable but harmless". It's not
758 clear what that actually means; it may preclude conforming
759 implementations from signalling errors.
760 b. COUNT, REMOVE and the like give priority to a :TEST-NOT argument
761 when conflict occurs. As a quality of implementation issue, it
762 might be preferable to treat :TEST and :TEST-NOT as being in some
763 sense the same &KEY, and effectively take the first test function in
765 c. Again, a quality of implementation issue: it would be good to issue a
766 STYLE-WARNING at compile-time for calls with :TEST-NOT, and a
767 WARNING for calls with both :TEST and :TEST-NOT; possibly this
768 latter should be WARNed about at execute-time too.
770 216: "debugger confused by frames with invalid number of arguments"
771 In sbcl-0.7.8.51, executing e.g. (VECTOR-PUSH-EXTEND T), BACKTRACE, Q
772 leaves the system confused, enough so that (QUIT) no longer works.
773 It's as though the process of working with the uninitialized slot in
774 the bad VECTOR-PUSH-EXTEND frame causes GC problems, though that may
775 not be the actual problem. (CMU CL 18c doesn't have problems with this.)
777 This is probably the same bug as 162
779 217: "Bad type operations with FUNCTION types"
782 * (values-type-union (specifier-type '(function (base-char)))
783 (specifier-type '(function (integer))))
785 #<FUN-TYPE (FUNCTION (BASE-CHAR) *)>
787 It causes insertion of wrong type assertions into generated
791 (let ((f (etypecase x
792 (character #'write-char)
793 (integer #'write-byte))))
796 (character (write-char x s))
797 (integer (write-byte x s)))))
799 Then (FOO #\1 *STANDARD-OUTPUT*) signals type error.
801 (In 0.7.9.1 the result type is (FUNCTION * *), so Python does not
802 produce invalid code, but type checking is not accurate.)
804 233: bugs in constraint propagation
806 (declaim (optimize (speed 2) (safety 3)))
808 (if (typep (prog1 x (setq x y)) 'double-float)
811 (foo 1d0 5) => segmentation violation
813 235: "type system and inline expansion"
815 (declaim (ftype (function (cons) number) acc))
816 (declaim (inline acc))
818 (the number (car c)))
821 (values (locally (declare (optimize (safety 0)))
823 (locally (declare (optimize (safety 3)))
826 (foo '(nil) '(t)) => NIL, T.
828 237: "Environment arguments to type functions"
829 a. Functions SUBTYPEP, TYPEP, UPGRADED-ARRAY-ELEMENT-TYPE, and
830 UPGRADED-COMPLEX-PART-TYPE now have an optional environment
831 argument, but they ignore it completely. This is almost
832 certainly not correct.
833 b. Also, the compiler's optimizers for TYPEP have not been informed
834 about the new argument; consequently, they will not transform
835 calls of the form (TYPEP 1 'INTEGER NIL), even though this is
836 just as optimizeable as (TYPEP 1 'INTEGER).
838 238: "REPL compiler overenthusiasm for CLOS code"
840 * (defclass foo () ())
841 * (defmethod bar ((x foo) (foo foo)) (call-next-method))
842 causes approximately 100 lines of code deletion notes. Some
843 discussion on this issue happened under the title 'Three "interesting"
844 bugs in PCL', resulting in a fix for this oververbosity from the
845 compiler proper; however, the problem persists in the interactor
846 because the notion of original source is not preserved: for the
847 compiler, the original source of the above expression is (DEFMETHOD
848 BAR ((X FOO) (FOO FOO)) (CALL-NEXT-METHOD)), while by the time the
849 compiler gets its hands on the code needing compilation from the REPL,
850 it has been macroexpanded several times.
852 A symptom of the same underlying problem, reported by Tony Martinez:
854 (with-input-from-string (*query-io* " no")
856 (simple-type-error () 'error))
858 ; (SB-KERNEL:FLOAT-WAIT)
860 ; note: deleting unreachable code
861 ; compilation unit finished
864 241: "DEFCLASS mysteriously remembers uninterned accessor names."
865 (from tonyms on #lisp IRC 2003-02-25)
866 In sbcl-0.7.12.55, typing
867 (defclass foo () ((bar :accessor foo-bar)))
870 (defclass foo () ((bar :accessor foo-bar)))
871 gives the error message
872 "#:FOO-BAR already names an ordinary function or a macro."
873 So it's somehow checking the uninterned old accessor name instead
874 of the new requested accessor name, which seems broken to me (WHN).
876 242: "WRITE-SEQUENCE suboptimality"
877 (observed from clx performance)
878 In sbcl-0.7.13, WRITE-SEQUENCE of a sequence of type
879 (SIMPLE-ARRAY (UNSIGNED-BYTE 8) (*)) on a stream with element-type
880 (UNSIGNED-BYTE 8) will write to the stream one byte at a time,
881 rather than writing the sequence in one go, leading to severe
882 performance degradation.
884 243: "STYLE-WARNING overenthusiasm for unused variables"
885 (observed from clx compilation)
886 In sbcl-0.7.14, in the presence of the macros
887 (DEFMACRO FOO (X) `(BAR ,X))
888 (DEFMACRO BAR (X) (DECLARE (IGNORABLE X)) 'NIL)
889 somewhat surprising style warnings are emitted for
890 (COMPILE NIL '(LAMBDA (Y) (FOO Y))):
892 ; (LAMBDA (Y) (FOO Y))
894 ; caught STYLE-WARNING:
895 ; The variable Y is defined but never used.
897 245: bugs in disassembler
898 b. On X86 operand size prefix is not recognized.
901 (defun foo (&key (a :x))
905 does not cause a warning. (BTW: old SBCL issued a warning, but for a
906 function, which was never called!)
909 Compiler does not emit warnings for
911 a. (lambda () (svref (make-array 8 :adjustable t) 1))
914 (list (let ((y (the real x)))
915 (unless (floatp y) (error ""))
920 (declare (optimize (debug 0)))
921 (declare (type vector x))
922 (list (fill-pointer x)
926 Complex array type does not have corresponding type specifier.
928 This is a problem because the compiler emits optimization notes when
929 you use a non-simple array, and without a type specifier for hairy
930 array types, there's no good way to tell it you're doing it
931 intentionally so that it should shut up and just compile the code.
933 Another problem is confusing error message "asserted type ARRAY
934 conflicts with derived type (VALUES SIMPLE-VECTOR &OPTIONAL)" during
935 compiling (LAMBDA (V) (VALUES (SVREF V 0) (VECTOR-POP V))).
937 The last problem is that when type assertions are converted to type
938 checks, types are represented with type specifiers, so we could lose
939 complex attribute. (Now this is probably not important, because
940 currently checks for complex arrays seem to be performed by
944 (compile nil '(lambda () (aref (make-array 0) 0))) compiles without
945 warning. Analogous cases with the index and length being equal and
946 greater than 0 are warned for; the problem here seems to be that the
947 type required for an array reference of this type is (INTEGER 0 (0))
948 which is canonicalized to NIL.
953 (t1 (specifier-type s)))
954 (eval `(defstruct ,s))
955 (type= t1 (specifier-type s)))
960 b. The same for CSUBTYPEP.
962 262: "yet another bug in inline expansion of local functions"
966 (declare (integer x y))
969 (declare (integer u))
970 (if (> (1+ (the unsigned-byte u)) 0)
972 (return (+ 38 (cos (/ u 78)))))))
973 (declare (inline xyz))
975 (* (funcall (eval #'xyz) x)
977 (funcall (if (> x 5) #'xyz #'identity)
982 Urgh... It's time to write IR1-copier.
985 David Lichteblau provided (sbcl-devel 2003-06-01) a patch to fix
986 behaviour of streams with element-type (SIGNED-BYTE 8). The patch
987 looks reasonable, if not obviously correct; however, it caused the
988 PPC/Linux port to segfault during warm-init while loading
989 src/pcl/std-class.fasl. A workaround patch was made, but it would
990 be nice to understand why the first patch caused problems, and to
991 fix the cause if possible.
993 268: "wrong free declaration scope"
994 The following code must signal type error:
996 (locally (declare (optimize (safety 3)))
997 (flet ((foo (x &optional (y (car x)))
998 (declare (optimize (safety 0)))
1000 (funcall (eval #'foo) 1)))
1003 SCALE-FLOAT should accept any integer for its second argument.
1006 In the following function constraint propagator optimizes nothing:
1009 (declare (integer x))
1010 (declare (optimize speed))
1018 Compilation of the following two forms causes "X is unbound" error:
1020 (symbol-macrolet ((x pi))
1021 (macrolet ((foo (y) (+ x y)))
1022 (declaim (inline bar))
1028 (See (COERCE (CDR X) 'FUNCTION) in IR1-CONVERT-INLINE-LAMBDA.)
1031 CLHS says that type declaration of a symbol macro should not affect
1032 its expansion, but in SBCL it does. (If you like magic and want to
1033 fix it, don't forget to change all uses of MACROEXPAND to
1037 The following code (taken from CLOCC) takes a lot of time to compile:
1040 (declare (type (integer 0 #.large-constant) n))
1043 (fixed in 0.8.2.51, but a test case would be good)
1046 (defmethod fee ((x fixnum))
1049 (fee 1) => type error
1056 (declare (optimize speed))
1057 (loop for i of-type (integer 0) from 0 by 2 below 10
1060 uses generic arithmetic.
1062 b. (fixed in 0.8.3.6)
1064 279: type propagation error -- correctly inferred type goes astray?
1065 In sbcl-0.8.3 and sbcl-0.8.1.47, the warning
1066 The binding of ABS-FOO is a (VALUES (INTEGER 0 0)
1067 &OPTIONAL), not a (INTEGER 1 536870911)
1068 is emitted when compiling this file:
1069 (declaim (ftype (function ((integer 0 #.most-positive-fixnum))
1070 (integer #.most-negative-fixnum 0))
1075 (let* (;; Uncomment this for a type mismatch warning indicating
1076 ;; that the type of (FOO X) is correctly understood.
1077 #+nil (fs-foo (float-sign (foo x)))
1078 ;; Uncomment this for a type mismatch warning
1079 ;; indicating that the type of (ABS (FOO X)) is
1080 ;; correctly understood.
1081 #+nil (fs-abs-foo (float-sign (abs (foo x))))
1082 ;; something wrong with this one though
1083 (abs-foo (abs (foo x))))
1084 (declare (type (integer 1 100) abs-foo))
1089 280: bogus WARNING about duplicate function definition
1090 In sbcl-0.8.3 and sbcl-0.8.1.47, if BS.MIN is defined inline,
1092 (declaim (inline bs.min))
1093 (defun bs.min (bases) nil)
1094 before compiling the file below, the compiler warns
1095 Duplicate definition for BS.MIN found in one static
1096 unit (usually a file).
1098 (declaim (special *minus* *plus* *stagnant*))
1099 (defun b.*.min (&optional (x () xp) (y () yp) &rest rest)
1101 (define-compiler-macro b.*.min (&rest rest)
1103 (defun afish-d-rbd (pd)
1105 (b.*.min (foo-d-rbd *stagnant*))
1106 (multiple-value-bind (reduce-fn initial-value)
1108 (list (values #'bs.min 0))
1109 (vector (values #'bs.min *plus*)))
1110 (let ((cv-ks (cv (kpd.ks pd))))
1111 (funcall reduce-fn d-rbds)))))
1112 (defun bfish-d-rbd (pd)
1114 (b.*.min (foo-d-rbd *stagnant*))
1115 (multiple-value-bind (reduce-fn initial-value)
1117 (list (values #'bs.min *minus*))
1118 (vector (values #'bs.min 0)))
1119 (let ((cv-ks (cv (kpd.ks pd))))
1120 (funcall reduce-fn d-rbds)))))
1122 281: COMPUTE-EFFECTIVE-METHOD error signalling.
1123 (slightly obscured by a non-0 default value for
1124 SB-PCL::*MAX-EMF-PRECOMPUTE-METHODS*)
1125 It would be natural for COMPUTE-EFFECTIVE-METHOD to signal errors
1126 when it finds a method with invalid qualifiers. However, it
1127 shouldn't signal errors when any such methods are not applicable to
1128 the particular call being evaluated, and certainly it shouldn't when
1129 simply precomputing effective methods that may never be called.
1130 (setf sb-pcl::*max-emf-precompute-methods* 0)
1132 (:method-combination +)
1133 (:method ((x symbol)) 1)
1134 (:method + ((x number)) x))
1135 (foo 1) -> ERROR, but should simply return 1
1137 The issue seems to be that construction of a discriminating function
1138 calls COMPUTE-EFFECTIVE-METHOD with methods that are not all applicable.
1140 283: Thread safety: libc functions
1141 There are places that we call unsafe-for-threading libc functions
1142 that we should find alternatives for, or put locks around. Known or
1143 strongly suspected problems, as of 0.8.3.10: please update this
1144 bug instead of creating new ones
1146 localtime() - called for timezone calculations in code/time.lisp
1148 284: Thread safety: special variables
1149 There are lots of special variables in SBCL, and I feel sure that at
1150 least some of them are indicative of potentially thread-unsafe
1151 parts of the system. See doc/internals/notes/threading-specials
1153 286: "recursive known functions"
1154 Self-call recognition conflicts with known function
1155 recognition. Currently cross compiler and target COMPILE do not
1156 recognize recursion, and in target compiler it can be disabled. We
1157 can always disable it for known functions with RECURSIVE attribute,
1158 but there remains a possibility of a function with a
1159 (tail)-recursive simplification pass and transforms/VOPs for base
1162 287: PPC/Linux miscompilation or corruption in first GC
1163 When the runtime is compiled with -O3 on certain PPC/Linux machines, a
1164 segmentation fault is reported at the point of first triggered GC,
1165 during the compilation of DEFSTRUCT WRAPPER. As a temporary workaround,
1166 the runtime is no longer compiled with -O3 on PPC/Linux, but it is likely
1167 that this merely obscures, not solves, the underlying problem; as and when
1168 underlying problems are fixed, it would be worth trying again to provoke
1171 288: fundamental cross-compilation issues (from old UGLINESS file)
1172 Using host floating point numbers to represent target floating point
1173 numbers, or host characters to represent target characters, is
1174 theoretically shaky. (The characters are OK as long as the characters
1175 are in the ANSI-guaranteed character set, though, so they aren't a
1176 real problem as long as the sources don't need anything but that;
1177 the floats are a real problem.)
1179 289: "type checking and source-transforms"
1181 (block nil (let () (funcall #'+ (eval 'nil) (eval '1) (return :good))))
1184 Our policy is to check argument types at the moment of a call. It
1185 disagrees with ANSI, which says that type assertions are put
1186 immediately onto argument expressions, but is easier to implement in
1187 IR1 and is more compatible to type inference, inline expansion,
1188 etc. IR1-transforms automatically keep this policy, but source
1189 transforms for associative functions (such as +), being applied
1190 during IR1-convertion, do not. It may be tolerable for direct calls
1191 (+ x y z), but for (FUNCALL #'+ x y z) it is non-conformant.
1193 b. Another aspect of this problem is efficiency. [x y + z +]
1194 requires less registers than [x y z + +]. This transformation is
1195 currently performed with source transforms, but it would be good to
1196 also perform it in IR1 optimization phase.
1198 290: Alpha floating point and denormalized traps
1199 In SBCL 0.8.3.6x on the alpha, we work around what appears to be a
1200 hardware or kernel deficiency: the status of the enable/disable
1201 denormalized-float traps bit seems to be ambiguous; by the time we
1202 get to os_restore_fp_control after a trap, denormalized traps seem
1203 to be enabled. Since we don't want a trap every time someone uses a
1204 denormalized float, in general, we mask out that bit when we restore
1205 the control word; however, this clobbers any change the user might
1209 (reported by Adam Warner, sbcl-devel 2003-09-23)
1211 The --load toplevel argument does not perform any sanitization of its
1212 argument. As a result, files with Lisp pathname pattern characters
1213 (#\* or #\?, for instance) or quotation marks can cause the system
1214 to perform arbitrary behaviour.
1217 LOOP with non-constant arithmetic step clauses suffers from overzealous
1218 type constraint: code of the form
1219 (loop for d of-type double-float from 0d0 to 10d0 by x collect d)
1220 compiles to a type restriction on X of (AND DOUBLE-FLOAT (REAL
1221 (0))). However, an integral value of X should be legal, because
1222 successive adds of integers to double-floats produces double-floats,
1223 so none of the type restrictions in the code is violated.
1225 300: (reported by Peter Graves) Function PEEK-CHAR checks PEEK-TYPE
1226 argument type only after having read a character. This is caused
1227 with EXPLICIT-CHECK attribute in DEFKNOWN. The similar problem
1228 exists with =, /=, <, >, <=, >=. They were fixed, but it is probably
1229 less error prone to have EXPLICIT-CHECK be a local declaration,
1230 being put into the definition, instead of an attribute being kept in
1231 a separate file; maybe also put it into SB-EXT?
1233 301: ARRAY-SIMPLE-=-TYPE-METHOD breaks on corner cases which can arise
1234 in NOTE-ASSUMED-TYPES
1235 In sbcl-0.8.7.32, compiling the file
1237 (declare (type integer x))
1238 (declare (type (vector (or hash-table bit)) y))
1241 (declare (type integer x))
1242 (declare (type (simple-array base (2)) y))
1245 failed AVER: "(NOT (AND (NOT EQUALP) CERTAINP))"
1247 302: Undefined type messes up DATA-VECTOR-REF expansion.
1249 (defun dis (s ei x y)
1250 (declare (type (simple-array function (2)) s) (type ei ei))
1251 (funcall (aref s ei) x y))
1252 on sbcl-0.8.7.36/X86/Linux causes a BUG to be signalled:
1253 full call to SB-KERNEL:DATA-VECTOR-REF
1255 303: "nonlinear LVARs" (aka MISC.293)
1257 (multiple-value-call #'list
1259 (multiple-value-prog1
1260 (eval '(values :a :b :c))
1266 (throw 'bar (values 3 4)))))))))))
1268 (BUU 1) returns garbage.
1270 The problem is that both EVALs sequentially write to the same LVAR.
1273 (Reported by Dave Roberts.)
1274 Local INLINE/NOTINLINE declaration removes local FTYPE declaration:
1277 (declare (ftype (function () (integer 0 10)) fee)
1281 uses generic arithmetic with INLINE and fixnum without.
1283 306: "Imprecise unions of array types"
1285 (declare (optimize speed)
1286 (type (or (array cons) (array vector)) x))
1288 (foo #((0))) => TYPE-ERROR
1295 ,@(loop for x across sb-vm:*specialized-array-element-type-properties*
1296 collect `(array ,(sb-vm:saetp-specifier x)))))
1297 => NIL, T (when it should be T, T)
1299 308: "Characters without names"
1300 (reported by Bruno Haible sbcl-devel "character names are missing"
1302 (graphic-char-p (code-char 255))
1304 (char-name (code-char 255))
1307 SBCL is unsure of what to do about characters with codes in the
1308 range 128-255. Currently they are treated as non-graphic, but don't
1309 have names, which is not compliant with the standard. Various fixes
1310 are possible, such as
1311 * giving them names such as NON-ASCII-128;
1312 * reducing CHAR-CODE-LIMIT to 127 (almost certainly unpopular);
1313 * making the characters graphic (makes a certain amount of sense);
1314 * biting the bullet and implementing Unicode (probably quite hard).
1316 309: "Dubious values for implementation limits"
1317 (reported by Bruno Haible sbcl-devel "Incorrect value of
1318 multiple-values-limit" 2004-04-19)
1319 (values-list (make-list 1000000)), on x86/linux, signals a stack
1320 exhaustion condition, despite MULTIPLE-VALUES-LIMIT being
1321 significantly larger than 1000000. There are probably similar
1322 dubious values for CALL-ARGUMENTS-LIMIT (see cmucl-help/cmucl-imp
1323 around the same time regarding a call to LIST on sparc with 1000
1324 arguments) and other implementation limit constants.
1326 311: "Tokeniser not thread-safe"
1327 (see also Robert Marlow sbcl-help "Multi threaded read chucking a
1329 The tokenizer's use of *read-buffer* and *read-buffer-length* causes
1330 spurious errors should two threads attempt to tokenise at the same
1334 (reported by Jon Dyte)
1335 SBCL issues a warning "Duplicate definition of FOO" compiling
1337 (declaim (inline foo))
1341 (list (foo y) (if (> y 1) (funcall (if (> y 0) #'foo #'identity) y))))
1343 (probably related to the bug 280.)
1345 314: "LOOP :INITIALLY clauses and scope of initializers"
1346 reported by Bruno Haible sbcl-devel "various SBCL bugs" from CLISP
1347 test suite, originally by Thomas F. Burdick.
1348 ;; <http://www.lisp.org/HyperSpec/Body/sec_6-1-7-2.html>
1349 ;; According to the HyperSpec 6.1.2.1.4, in for-as-equals-then, var is
1350 ;; initialized to the result of evaluating form1. 6.1.7.2 says that
1351 ;; initially clauses are evaluated in the loop prologue, which precedes all
1352 ;; loop code except for the initial settings provided by with, for, or as.
1353 (loop :for x = 0 :then (1+ x)
1354 :for y = (1+ x) :then (ash y 1)
1355 :for z :across #(1 3 9 27 81 243)
1357 :initially (assert (zerop x)) :initially (assert (= 2 w))
1358 :until (>= w 100) :collect w)
1359 Expected: (2 6 15 38)
1362 315: "no bounds check for access to displaced array"
1363 reported by Bruno Haible sbcl-devel "various SBCL bugs" from CLISP
1365 (locally (declare (optimize (safety 3) (speed 0)))
1366 (let* ((x (make-array 10 :fill-pointer 4 :element-type 'character
1367 :initial-element #\space :adjustable t))
1368 (y (make-array 10 :fill-pointer 4 :element-type 'character
1370 (adjust-array x '(5))
1373 SBCL 0.8.10 elides the bounds check somewhere along the line, and
1374 returns #\Nul (where an error would be much preferable, since a test
1375 of that form but with (setf (char y 5) #\Space) potentially corrupts
1376 the heap and certainly confuses the world if that string is used by
1379 317: "FORMAT of floating point numbers"
1380 reported by Bruno Haible sbcl-devel "various SBCL bugs" from CLISP
1382 (format nil "~1F" 10) => "0." ; "10." expected
1383 (format nil "~0F" 10) => "0." ; "10." expected
1384 (format nil "~2F" 1234567.1) => "1000000." ; "1234567." expected
1385 it would be nice if whatever fixed this also untangled the two
1386 competing implementations of floating point printing (Steele and
1387 White, and Burger and Dybvig) present in src/code/print.lisp
1389 318: "stack overflow in compiler warning with redefined class"
1390 reported by Bruno Haible sbcl-devel "various SBCL bugs" from CLISP
1392 (setq *print-pretty* nil)
1394 (setf (find-class 'foo) nil)
1395 (defstruct foo slot-1)
1397 ...#<SB-KERNEL:STRUCTURE-CLASSOID #<SB-KERNEL:STRUCTURE-CLASSOID #<SB-KERNEL:STRUCTURE-CLASSOID #<SB-KERNEL:STRUCTUREControl stack guard page temporarily disabled: proceed with caution
1398 (it's not really clear what it should give: is (SETF FIND-CLASS)
1399 meant to be enough to delete structure classes from the system?
1400 Giving a stack overflow is definitely suboptimal, though.)
1402 319: "backquote with comma inside array"
1403 reported by Bruno Haible sbcl-devel "various SBCL bugs" from CLISP
1405 (read-from-string "`#1A(1 2 ,(+ 2 2) 4)")
1407 #(1 2 ((SB-IMPL::|,|) + 2 2) 4)
1408 which probably isn't intentional.
1410 321: "DEFINE-METHOD-COMBINATION lambda list parsing"
1411 reported by Bruno Haible sbcl-devel "various SBCL bugs" from CLISP
1413 (define-method-combination w-args ()
1415 (:arguments arg1 arg2 &aux (extra :extra))
1416 `(progn ,@(mapcar (lambda (method) `(call-method ,method)) method-list)))
1417 gives a (caught) compile-time error, which can be exposed by
1418 (defgeneric mc-test-w-args (p1 p2 s)
1419 (:method-combination w-args)
1420 (:method ((p1 number) (p2 t) s)
1421 (vector-push-extend (list 'number p1 p2) s))
1422 (:method ((p1 string) (p2 t) s)
1423 (vector-push-extend (list 'string p1 p2) s))
1424 (:method ((p1 t) (p2 t) s) (vector-push-extend (list t p1 p2) s)))
1426 323: "REPLACE, BIT-BASH and large strings"
1427 The transform for REPLACE on simple-base-strings uses BIT-BASH, which
1428 at present has an upper limit in size. Consequently, in sbcl-0.8.10
1430 (declare (optimize speed (safety 1)))
1431 (let ((x (make-string 140000000))
1432 (y (make-string 140000000)))
1433 (length (replace x y))))
1436 debugger invoked on a TYPE-ERROR in thread 2412:
1437 The value 1120000000 is not of type (MOD 536870911).
1438 (see also "more and better sequence transforms" sbcl-devel 2004-05-10)
1440 324: "STREAMs and :ELEMENT-TYPE with large bytesize"
1441 In theory, (open foo :element-type '(unsigned-byte <x>)) should work
1442 for all positive integral <x>. At present, it only works for <x> up
1443 to about 1024 (and similarly for signed-byte), so
1444 (open "/dev/zero" :element-type '(unsigned-byte 1025))
1445 gives an error in sbcl-0.8.10.
1447 325: "CLOSE :ABORT T on supeseding streams"
1448 Closing a stream opened with :IF-EXISTS :SUPERSEDE with :ABORT T leaves no
1449 file on disk, even if one existed before opening.
1451 The illegality of this is not crystal clear, as the ANSI dictionary
1452 entry for CLOSE says that when :ABORT is T superseded files are not
1453 superseded (ie. the original should be restored), whereas the OPEN
1454 entry says about :IF-EXISTS :SUPERSEDE "If possible, the
1455 implementation should not destroy the old file until the new stream
1456 is closed." -- implying that even though undesirable, early deletion
1457 is legal. Restoring the original would none the less be the polite