sbcl-2.0.2 released

Christophe Rhodes <[email protected]> Sat, 29 Feb 2020 21:16:32 +0000
Newsgroups gmane.lisp.steel-bank.announce
Message-ID <87sgitytzj.fsf@shin>
Dear all,

I have just released sbcl-2.0.2, and uploaded sources and a binary to
SourceForge.

There might be cause for a little more caution than usual when
considering whether to upgrade to this release.  The fix to REMHASH (and
indeed the report of the bug) came very late in the release cycle, and
has not have seen as much testing as I might have liked.  We also have a
report of a regression in dynamic-extent allocation, where under some
circumstances (involving fairly complicated compiled code) objects which
were allocated on the stack in sbcl-2.0.1 are not stack-allocated in
this release.  If this affects you, a test case on sbcl-devel would be
much appreciated!

Other notes on the changes in this release are appended below.

Best wishes,

Christophe

changes in sbcl-2.0.2 relative to sbcl-2.0.1:
  * new feature: CMUCL-style block compilation support has been added,
    allowing whole program optimization. It has been documented in a new
    section of the manual entitled "Advanced Compiler Use and Efficiency
    Hints". In particular, users of block compilation will find a large
    speedup for numerical code, as functions which call or return floating
    point values will keep everything unboxed.
  * bug fix: in CHANGE-CLASS, handle non-standard slot-value-using-class for
    :INSTANCE and :CLASS slots better.  (reported by Shinmera on #sbcl)
  * bug fix: REMHASH in an EQUAL or EQUALP hash-table works correctly on a
    key for which EQUAL (respectively EQUALP) is reduced to EQ.  (lp#1865094)
  * bug fix: do not emit a warning for DEFCLASS slots with mismatched :TYPE
    and :INITFORM when the class has a non-standard metaclass.
  * bug fix: DISASSEMBLE on SPARC is less likely to spuriously annotate code
    sequences as error traps.