SBCL 1.0.24 released

Source is available in SourceForge CVS and on the SourceForge File
Releases page.

Some readers of this list might not have heard the sad news of the
death of Thiemo Seufer in a car accident on December the 26th; the
SBCL community, along with many others (as described in an obituary at
<http://lwn.net/Articles/313092/>) mourns the loss of a colleague and
friend.

Best wishes to all,

Christophe

changes in sbcl-1.0.24 relative to 1.0.23:
  * new feature: ARRAY-STORAGE-VECTOR provides access to the underlying data
    vector of a multidimensional SIMPLE-ARRAY.
  * new feature: the system now signals a continuable error if standard
    readtable modification is attempted. (thanks to Tobias Rittweiler)
  * new feature: DIRECTORY has been extended with a non-standard keyword
    argument :RESOLVE-SYMLINKS. (thanks to TC-Rucho)
  * enhancement: SB-BSD-SOCKETS:MAKE-SOCKET-STREAM keyword argument handling
    has been robustified and documented better. (thanks to Robert Goldman)
  * optimization: faster generic arithmetic dispatch on x86 and x86-64.
  * optimization: FORMAT ~D and ~F are now approximately 5% faster.
  * tradeoff: constant FORMAT control strings are now compiled unless
    SPACE > SPEED (previously only when SPEED > SPACE.)
  * bug fix: Red Hat Enterprise 3 mmap randomization workaround. (thanks
    to Thomas Burdick)
  * bug fix: DEFCLASS and ENSURE-CLASS-USING-CLASS are now expected to
    be thread safe.
  * bug fix: lexical type declarations are now correctly reported by
    SB-CLTL2. (reported by Larry D'Anna)
  * bug fix: STRING-TO-OCTETS did not handle :START properly when
    using UTF-8 as external format. (thanks to Luis Oliveira)
  * bug fix: errors from invalid fill-pointer values to (SETF FILL-POINTER)
    are signalled correctly. (thanks to Stas Boukarev)
  * bug fix: SET-MACRO-CHARACTER accepts NIL as the readtable
    designator. (thanks to Tobias Rittweiler)
  * bug fix: SET-DISPATCH-MACRO-CHARACTER accepts NIL as the readtable
    designator, and returns T instead of the function. (thanks to
    Tobias Rittweiler)
  * bug fix: direct superclasses of STANDARD-CLASS and
    FUNCALLABLE-STANDARD-CLASS now default to STANDARD-OBJECT and
    FUNCALLABLE-STANDARD-OBJECT as required by AMOP.
  * bug fix: compiling a call to SLOT-VALUE with a constant slot-name
    when no class with the named slot yet exists no longer causes a
    compile-time style-warning.
  * bug fix: :ALLOCATION :CLASS slots are type-checked properly
    in safe code. (reported by Didier Verna)
  * bug fix: #430; nested structure constructors can stack allocate.
  * bug fix: on some 64-bit platforms dynamic space size was truncated
    to #xffff0000 bytes. (reported by Benjamin Lambert)
  * bug fix: setting *READ-SUPPRESS* to T no longer renders the default
    REPL unusable. (reported by Daniel Herring)
  * bug fix: return values of READ-SEQUENCE did not take :START into
    account on file streams, regressions since 1.0.12.22. (reported by
    Thomas Russ, patch by Paul Khuong)
  * bug fix: using SET or (SETF SYMBOL-VALUE) to change the value of a
    method specializer used to confuse permuation vector optimization.
  * bug fix: system inserted bogus implicit type declarations for local
    special variables in DEFMETHOD bodies.
  * bug fix: #354; duplicated frames in backtraces due to
    non-tail-call-optimized XEPs to functions with return type NIL
    have been elimited.
  * bug fix: #357; MAKE-INSTANCE/SHARED-INITIALIZE now
    initializes structure object slots according to DEFSTRUCT initforms,
    and DEFSTRUCT forms :INCLUDEind structure classes defined using
    DEFCLASS :METACLASS STRUCTURE-CLASS now inherit their initforms.
    (reported by Bruno Haible and Stephen Wilson)
  * bug fix: #395; fill-pointer output streams used now support
    element-type BASE-CHAR as well.
  * bug fix: compiler error when attempting to derive return value of
    ARRAY-ELEMENT-TYPE when the array type was a union of intersection
    types.
  * bug fix: address-spaces overlapped on OpenBSD. (thanks to Josh
    Elsasser)
  * bug fix: Mac OS X binaries should now be portable between Leopard
    and Tiger.

------------------------------------------------------------------------------