X-Git-Url: http://repo.macrolet.net/gitweb/?a=blobdiff_plain;f=BUGS;h=bba6c83e684dddc234652ac2f544d3ca677d68d4;hb=94ac5b7c3ff37850210b6fc9a7593cf1c5752993;hp=39508d5f5e75f386e30c3543473a81bb11598de5;hpb=372989d837526e3100b364153d58181a2a563351;p=sbcl.git

diff --git a/BUGS b/BUGS
index 39508d5..bba6c83 100644
--- a/BUGS
+++ b/BUGS
@@ -972,18 +972,36 @@ WORKAROUND:
       (let ((x (1+ x)))
         (call-next-method)))
   Now (FOO 3) should return 3, but instead it returns 4.
-    
-137:
-  (SB-DEBUG:BACKTRACE) output should start with something
-  including the name BACKTRACE, not (as in 0.pre7.88)
-  just "0: (\"hairy arg processor\" ...)". Until about
-  sbcl-0.pre7.109, the names in BACKTRACE were all screwed
-  up compared to the nice useful names in sbcl-0.6.13.
-  Around sbcl-0.pre7.109, they were mostly fixed by using
-  NAMED-LAMBDA to implement DEFUN. However, there are still
-  some screwups left, e.g. as of sbcl-0.pre7.109, there are
-  still some functions named "hairy arg processor" and
-  "SB-INT:&MORE processor".
+
+140:
+  (reported by Alexey Dejneka sbcl-devel 2002-01-03)
+
+  SUBTYPEP does not work well with redefined classes:
+  ---
+  * (defclass a () ())
+  #<STANDARD-CLASS A>
+  * (defclass b () ())
+  #<STANDARD-CLASS B>
+  * (subtypep 'b 'a)
+  NIL
+  T
+  * (defclass b (a) ())
+  #<STANDARD-CLASS B>
+  * (subtypep 'b 'a)
+  T
+  T
+  * (defclass b () ())
+  #<STANDARD-CLASS B>
+   
+  ;;; And now...
+  * (subtypep 'b 'a)
+  T
+  T
+
+  This bug was fixed in sbcl-0.7.4.1 by invalidating the PCL wrapper
+  class upon redefinition. Unfortunately, doing so causes bug #176 to
+  appear.  Pending further investication, one or other of these bugs
+  might be present at any given time.
 
 141: 
   Pretty-printing nested backquotes doesn't work right, as 
@@ -1257,8 +1275,92 @@ WORKAROUND:
   on various operating systems. (reported by Harald Hanche-Olsen on
   cmucl-help 2002-05-31)
 
-175:
-  (fixed in sbcl-0.7.4.14)
+176:
+  reported by Alexey Dejneka 08 Jun 2002 in sbcl-devel:
+    Playing with McCLIM, I've received an error "Unbound variable WRAPPER
+    in SB-PCL::CHECK-WRAPPER-VALIDITY".
+      (defun check-wrapper-validity (instance)
+        (let* ((owrapper (wrapper-of instance)))
+          (if (not (invalid-wrapper-p owrapper))
+              owrapper
+              (let* ((state (wrapper-state wrapper)) ; !!!
+        ...
+    I've tried to replace it with OWRAPPER, but now OBSOLETE-INSTANCE-TRAP
+    breaks with "NIL is not of type SB-KERNEL:LAYOUT".
+    SBCL 0.7.4.13.
+  partial fix: The undefined variable WRAPPER resulted from an error
+  in recent refactoring, as can be seen by comparing to the code in e.g. 
+  sbcl-0.7.2. Replacing WRAPPER with OWRAPPER (done by WHN in sbcl-0.7.4.22)
+  should bring the code back to its behavior as of sbcl-0.7.2, but
+  that still leaves the OBSOLETE-INSTANCE-TRAP bug. An example of 
+  input which triggers that bug is
+    (dotimes (i 20)
+      (let ((lastname (intern (format nil "C~D" (1- i))))
+            (name (intern (format nil "C~D" i))))
+      (eval `(defclass ,name
+                 (,@(if (= i 0) nil (list lastname)))
+               ()))
+      (eval `(defmethod initialize-instance :after ((x ,name) &rest any)
+               (declare (ignore any))))))
+    (defclass b () ())
+    (defclass c0 (b) ())
+    (make-instance 'c19)
+
+  See also bug #140.
+
+178: "AVER failure compiling confused THEs in FUNCALL"
+  In sbcl-0.7.4.24, compiling
+    (defun bug178 (x)
+      (funcall (the function (the standard-object x))))
+  gives 
+    failed AVER:
+      "(AND (EQ (IR2-CONTINUATION-PRIMITIVE-TYPE 2CONT) FUNCTION-PTYPE) (EQ CHECK T))"
+  This variant compiles OK, though:
+    (defun bug178alternative (x)
+      (funcall (the nil x)))
+
+181: "bad type specifier drops compiler into debugger"
+  Compiling 
+    (in-package :cl-user)
+    (defun bar (x)
+      (declare (type 0 x))
+      (cons x x))
+  signals 
+    bad thing to be a type specifier: 0
+  which seems fine, but also enters the debugger (instead of having
+  the compiler handle the error, convert it into a COMPILER-ERROR, and
+  continue compiling) which seems wrong.
+
+182: "SPARC/Linux floating point"
+  Evaluating (/ 1.0 0.0) at the prompt causes a Bus error and complete
+  death of the environment. Other floating point operations sometimes
+  return infinities when they should raise traps (e.g. the test case
+  for bug #146, (expt 2.0 12777)).
+
+183: "IEEE floating point issues"
+  Even where floating point handling is being dealt with relatively
+  well (as of sbcl-0.7.5, on sparc/sunos and alpha; see bug #146), the
+  accrued-exceptions and current-exceptions part of the fp control
+  word don't seem to bear much relation to reality. E.g. on
+  SPARC/SunOS:
+  * (/ 1.0 0.0)
+
+  debugger invoked on condition of type DIVISION-BY-ZERO:
+    arithmetic error DIVISION-BY-ZERO signalled
+  0] (sb-vm::get-floating-point-modes)
+
+  (:TRAPS (:OVERFLOW :INVALID :DIVIDE-BY-ZERO)
+          :ROUNDING-MODE :NEAREST
+          :CURRENT-EXCEPTIONS NIL
+          :ACCRUED-EXCEPTIONS (:INEXACT)
+          :FAST-MODE NIL)
+  0] abort
+  * (sb-vm::get-floating-point-modes)
+  (:TRAPS (:OVERFLOW :INVALID :DIVIDE-BY-ZERO)
+          :ROUNDING-MODE :NEAREST
+          :CURRENT-EXCEPTIONS (:INEXACT)
+          :ACCRUED-EXCEPTIONS (:INEXACT)
+          :FAST-MODE NIL)
 
 DEFUNCT CATEGORIES OF BUGS
   IR1-#: