X-Git-Url: http://repo.macrolet.net/gitweb/?a=blobdiff_plain;f=BUGS;h=788da3e6cdf1a77920485fbb01128d60ad52b6c7;hb=227096b878fee7afae9d3bc2cee5df01449bca2d;hp=bd1c01ebaed389b3e5e8aa03edfaae3ea955a5c2;hpb=dbb323485444a8b452c84a2e1f5c99a79047bd8c;p=sbcl.git diff --git a/BUGS b/BUGS index bd1c01e..788da3e 100644 --- a/BUGS +++ b/BUGS @@ -32,8 +32,6 @@ have gone away (typically because they were fixed, but sometimes for other reasons, e.g. because they were moved elsewhere). -KNOWN BUGS OF NO SPECIAL CLASS: - 2: DEFSTRUCT almost certainly should overwrite the old LAYOUT information instead of just punting when a contradictory structure definition @@ -190,7 +188,10 @@ WORKAROUND: type safety errors reported by Peter Van Eynde July 25, 2000: k: READ-BYTE is supposed to signal TYPE-ERROR when its argument is not a binary input stream, but instead cheerfully reads from - character streams, e.g. (MAKE-STRING-INPUT-STREAM "abc"). + string-input streams, e.g. (MAKE-STRING-INPUT-STREAM "abc"). + [ Bug was reported as "from character streams", but in 0.8.3.10 we + get correct behaviour from (WITH-OPEN-FILE (i "/dev/zero") (READ-BYTE i)) ] + 60: The debugger LIST-LOCATIONS command doesn't work properly. @@ -360,6 +361,8 @@ WORKAROUND: your pre-0.7.0 state of grace with #+sbcl (declaim (notinline find position find-if position-if)) ; bug 117.. + (see also bug 279) + 118: as reported by Eric Marsden on cmucl-imp@cons.org 2001-08-14: (= (FLOAT 1 DOUBLE-FLOAT-EPSILON) @@ -490,6 +493,7 @@ WORKAROUND: it took more than two minutes (but less than five) for me. 145: + a. ANSI allows types `(COMPLEX ,FOO) to use very hairy values for FOO, e.g. (COMPLEX (AND REAL (SATISFIES ODDP))). The old CMU CL COMPLEX implementation didn't deal with this, and hasn't been @@ -497,6 +501,8 @@ WORKAROUND: conformance problem, since seems hard to construct useful code where it matters.) + b. (fixed in 0.8.3.43) + 146: Floating point errors are reported poorly. E.g. on x86 OpenBSD with sbcl-0.7.1, @@ -529,6 +535,8 @@ WORKAROUND: isn't too surprising since there are many differences in stack implementation and GC conservatism between the X86 and other ports.) + This is probably the same bug as 216 + 167: In sbcl-0.7.3.11, compiling the (illegal) code (in-package :cl-user) @@ -783,6 +791,8 @@ WORKAROUND: the bad VECTOR-PUSH-EXTEND frame causes GC problems, though that may not be the actual problem. (CMU CL 18c doesn't have problems with this.) + This is probably the same bug as 162 + 217: "Bad type operations with FUNCTION types" In sbcl.0.7.7: @@ -809,20 +819,6 @@ WORKAROUND: produce invalid code, but type checking is not accurate.) 233: bugs in constraint propagation - a. - (defun foo (x) - (declare (optimize (speed 2) (safety 3))) - (let ((y 0d0)) - (values - (the double-float x) - (setq y (+ x 1d0)) - (setq x 3d0) - (quux y (+ y 2d0) (* y 3d0))))) - (foo 4) => segmentation violation - - (see usage of CONTINUATION-ASSERTED-TYPE in USE-RESULT-CONSTRAINTS) - (see also bug 236) - b. (declaim (optimize (speed 2) (safety 3))) (defun foo (x y) @@ -1051,10 +1047,6 @@ WORKAROUND: (bignum "hip") (t "zuz"))) -272: - All forms of GC hooks (including notifiers and finalisers) are currently - (since 0.8.0) broken for gencgc (i.e. x86) users - 273: Compilation of the following two forms causes "X is unbound" error: @@ -1070,7 +1062,9 @@ WORKAROUND: 274: CLHS says that type declaration of a symbol macro should not affect - its expansion, but in SBCL it does. + its expansion, but in SBCL it does. (If you like magic and want to + fix it, don't forget to change all uses of MACROEXPAND to + MACROEXPAND*.) 275: The following code (taken from CLOCC) takes a lot of time to compile: @@ -1089,8 +1083,178 @@ WORKAROUND: (taken from CLOCC) +278: + a. + (defun foo () + (declare (optimize speed)) + (loop for i of-type (integer 0) from 0 by 2 below 10 + collect i)) + + uses generic arithmetic. + + b. (fixed in 0.8.3.6) -DEFUNCT CATEGORIES OF BUGS - IR1-#: - These labels were used for bugs related to the old IR1 interpreter. - The # values reached 6 before the category was closed down. +279: type propagation error -- correctly inferred type goes astray? + In sbcl-0.8.3 and sbcl-0.8.1.47, the warning + The binding of ABS-FOO is a (VALUES (INTEGER 0 0) + &OPTIONAL), not a (INTEGER 1 536870911) + is emitted when compiling this file: + (declaim (ftype (function ((integer 0 #.most-positive-fixnum)) + (integer #.most-negative-fixnum 0)) + foo)) + (defun foo (x) + (- x)) + (defun bar (x) + (let* (;; Uncomment this for a type mismatch warning indicating + ;; that the type of (FOO X) is correctly understood. + #+nil (fs-foo (float-sign (foo x))) + ;; Uncomment this for a type mismatch warning + ;; indicating that the type of (ABS (FOO X)) is + ;; correctly understood. + #+nil (fs-abs-foo (float-sign (abs (foo x)))) + ;; something wrong with this one though + (abs-foo (abs (foo x)))) + (declare (type (integer 1 100) abs-foo)) + (print abs-foo))) + + (see also bug 117) + +280: bogus WARNING about duplicate function definition + In sbcl-0.8.3 and sbcl-0.8.1.47, if BS.MIN is defined inline, + e.g. by + (declaim (inline bs.min)) + (defun bs.min (bases) nil) + before compiling the file below, the compiler warns + Duplicate definition for BS.MIN found in one static + unit (usually a file). + when compiling + (declaim (special *minus* *plus* *stagnant*)) + (defun b.*.min (&optional (x () xp) (y () yp) &rest rest) + (bs.min avec)) + (define-compiler-macro b.*.min (&rest rest) + `(bs.min ,@rest)) + (defun afish-d-rbd (pd) + (if *stagnant* + (b.*.min (foo-d-rbd *stagnant*)) + (multiple-value-bind (reduce-fn initial-value) + (etypecase pd + (list (values #'bs.min 0)) + (vector (values #'bs.min *plus*))) + (let ((cv-ks (cv (kpd.ks pd)))) + (funcall reduce-fn d-rbds))))) + (defun bfish-d-rbd (pd) + (if *stagnant* + (b.*.min (foo-d-rbd *stagnant*)) + (multiple-value-bind (reduce-fn initial-value) + (etypecase pd + (list (values #'bs.min *minus*)) + (vector (values #'bs.min 0))) + (let ((cv-ks (cv (kpd.ks pd)))) + (funcall reduce-fn d-rbds))))) + +281: COMPUTE-EFFECTIVE-METHOD error signalling. + (slightly obscured by a non-0 default value for + SB-PCL::*MAX-EMF-PRECOMPUTE-METHODS*) + It would be natural for COMPUTE-EFFECTIVE-METHOD to signal errors + when it finds a method with invalid qualifiers. However, it + shouldn't signal errors when any such methods are not applicable to + the particular call being evaluated, and certainly it shouldn't when + simply precomputing effective methods that may never be called. + (setf sb-pcl::*max-emf-precompute-methods* 0) + (defgeneric foo (x) + (:method-combination +) + (:method ((x symbol)) 1) + (:method + ((x number)) x)) + (foo 1) -> ERROR, but should simply return 1 + + The issue seems to be that construction of a discriminating function + calls COMPUTE-EFFECTIVE-METHOD with methods that are not all applicable. + +282: "type checking in full calls" + In current (0.8.3.6) implementation a CAST in a full call argument + is not checked; but the continuation between the CAST and the + combination has the "checked" type and CAST performs unsafe + coercion; this may lead to errors: if FOO is declared to take a + FIXNUM, this code will produce garbage on a machine with 30-bit + fixnums: + + (foo (aref (the (array (unsigned-byte 32)) x))) + +283: Thread safety: libc functions + There are places that we call unsafe-for-threading libc functions + that we should find alternatives for, or put locks around. Known or + strongly suspected problems, as of 0.8.3.10: please update this + bug instead of creating new ones + + localtime() - called for timezone calculations in code/time.lisp + +284: Thread safety: special variables + There are lots of special variables in SBCL, and I feel sure that at + least some of them are indicative of potentially thread-unsafe + parts of the system. See doc/internals/notes/threading-specials + +286: "recursive known functions" + Self-call recognition conflicts with known function + recognition. Currently cross compiler and target COMPILE do not + recognize recursion, and in target compiler it can be disabled. We + can always disable it for known functions with RECURSIVE attribute, + but there remains a possibility of a function with a + (tail)-recursive simplification pass and transforms/VOPs for base + cases. + +287: PPC/Linux miscompilation or corruption in first GC + When the runtime is compiled with -O3 on certain PPC/Linux machines, a + segmentation fault is reported at the point of first triggered GC, + during the compilation of DEFSTRUCT WRAPPER. As a temporary workaround, + the runtime is no longer compiled with -O3 on PPC/Linux, but it is likely + that this merely obscures, not solves, the underlying problem; as and when + underlying problems are fixed, it would be worth trying again to provoke + this problem. + +288: fundamental cross-compilation issues (from old UGLINESS file) + 288a: Using host floating point numbers to represent target + floating point numbers, or host characters to represent + target characters, is theoretically shaky. (The characters + are OK as long as the characters are in the ANSI-guaranteed + character set, though, so they aren't a real problem as + long as the sources don't need anything but that.) + 288b: The compiler still makes assumptions about cross-compilation-host + implementation of ANSI CL: + 288b1: Simple bit vectors are distinct from simple vectors (in + DEFINE-STORAGE-BASE and elsewhere). (Actually, I'm not *sure* + that things would really break if this weren't so, but I + strongly suspect that they would.) + 288b2: SINGLE-FLOAT is distinct from DOUBLE-FLOAT. (This is + in a sense just one aspect of bug 288a.) + +289: "type checking and source-transforms" + a. + (block nil (let () (funcall #'+ (eval 'nil) (eval '1) (return :good)))) + signals type error. + + Our policy is to check argument types at the moment of a call. It + disagrees with ANSI, which says that type assertions are put + immediately onto argument expressions, but is easier to implement in + IR1 and is more compatible to type inference, inline expansion, + etc. IR1-transforms automatically keep this policy, but source + transforms for associative functions (such as +), being applied + during IR1-convertion, do not. It may be tolerable for direct calls + (+ x y z), but for (FUNCALL #'+ x y z) it is non-conformant. + + b. Another aspect of this problem is efficiency. [x y + z +] + requires less registers than [x y z + +]. This transformation is + currently performed with source transforms, but it would be good to + also perform it in IR1 optimization phase. + +290: Alpha floating point and denormalized traps + In SBCL 0.8.3.6x on the alpha, we work around what appears to be a + hardware or kernel deficiency: the status of the enable/disable + denormalized-float traps bit seems to be ambiguous; by the time we + get to os_restore_fp_control after a trap, denormalized traps seem + to be enabled. Since we don't want a trap every time someone uses a + denormalized float, in general, we mask out that bit when we restore + the control word; however, this clobbers any change the user might + have made. + +292: + (fixed in 0.8.3.74)