X-Git-Url: http://repo.macrolet.net/gitweb/?a=blobdiff_plain;f=NEWS;h=7bdf0e52842b164be9d7256af2b04798a133e3a1;hb=7217c0c2f1f8dc7934f8b0a25b9d3c4c4c6fe950;hp=68c904240a0f54a7705dbaa36fc77c35f158a98c;hpb=237ec432952f0e7d4a4bcd5f683942a253cac56a;p=sbcl.git

diff --git a/NEWS b/NEWS
index 68c9042..7bdf0e5 100644
--- a/NEWS
+++ b/NEWS
@@ -1,4 +1,67 @@
 ;;;; -*- coding: utf-8; fill-column: 78 -*-
+changes relative to sbcl-1.0.30:
+  * improvement: stack allocation is should now be possible in all nested
+    inlining cases: failure to stack allocate when equivalent code is manually
+    open coded is now considered a bug.
+  * improvement: the Unicode character database has been upgraded to the
+    Unicode 5.0.1 standard, giving names and properties to a number of new
+    characters, and providing a few extra characters with case
+    transformations.
+  * new feature: experimental :EMIT-CFASL parameter to COMPILE-FILE can
+    be used to output toplevel compile-time effects into a separate .CFASL
+    file.
+  * optimization: COERCE to VECTOR, STRING, SIMPLE-STRING and recognizable
+    one-dimenstional subtypes of ARRAY is upto 70% faster when the coercion is
+    actually needed.
+  * optimization: division of floating point numbers by constants uses
+    multiplication by reciprocal when an exact reciprocal exists.
+  * optimization: multiplication of single- and double-floats floats by
+    constant two has been optimized.
+  * optimization: ARRAY-IN-BOUNDS-P is resolved at compile-time when
+    sufficient type information is available. (thanks to Leslie Polzer)
+  * improvement: a STYLE-WARNING is signalled when a generic function
+    clobbers an earlier FTYPE proclamation.
+  * improvement: the compiler is able to track the effective type of
+    generic function across method addition and removal even in the
+    absence of an explicit DEFGENERIC.
+  * improvement: DESCRIBE now reports on symbols naming undefined
+    but assumed or declared function as well.
+  * improvement: recompilation of systems using SB-GROVEL now works
+    (thanks to Leslie Polzer)
+  * improvements to SB-CLTL2 (thanks to Larry D'Anna):
+    ** functions DECLARATION-INFORMATION, PARSE-MACRO, and ENCLOSE have been
+       documented.
+    ** AUGMENT-ENVIRONMENT and DEFINE-DECLARATION have been implemented.
+    ** DECLARATION-INFORMATION now supports declaration name DECLARATION as
+       well as user defined declaration names.
+    ** VARIABLE-INFORMATION is now aware of alien variables.
+  * improvement: improved address space layout on OpenBSD (thanks to Josh
+    Elsasser)
+  * improvement: pretty-printing of various Lisp forms has been improved
+    (thanks to Tobias Rittweiler)
+  * bug fix: a failing AVER in CONVERT-MV-CALL has been fixed. (thanks to
+    Larry D'Anna)
+  * bug fix: a failing AVER in %ALLOCATE-CLOSURES conversion has been fixed
+    (thanks to Larry D'Anna)
+  * bug fix: SLEEP supports times over 100 million seconds on long on OpenBSD
+    as well. (reported by Josh Elsasser)
+  * bug fix: DELETE-FILE on streams no longer closes the stream with :ABORT T,
+    leading to possible attempts to delete the same file twice. See docstring
+    on DELETE-FILE for details. (reported by John Fremlin)
+  * bug fix: the low-level debugger had 32-bit assumptions and was missing
+    information about some array types. (thanks to Luis Oliveira)
+  * bug fix: moderately complex combinations of inline expansions could
+    be miscompiled if the result was declared to be dynamic extent.
+  * bug fix: in some cases no compiler note about failure to stack allocate
+    was emitted, even if the objects were in fact heap allocated.
+  * bug fix: minor violation of "otherwise inaccessible" rule for stack
+    allocation could cause objects users might reasonably expect to
+    be heap allocated to be stack allocated.
+  * bug fix: DESCRIBE signalled an error for generic functions under
+    certain circumstances. (thanks to Leslie Polzer)
+  * bug fix: Fixed spelling of an error message.
+
+changes in sbcl-1.0.30 relative to sbcl-1.0.29:
   * minor incompatible change: SB-THREAD:JOIN-THREAD-ERROR-THREAD and
     SB-THREAD:INTERRUPT-THREAD-ERROR-THREAD have been deprecated in favor
     of SB-THREAD:THREAD-ERROR-THREAD.
@@ -9,17 +72,43 @@
     values in other threads.
   * new feature: SB-INTROSPECT:ALLOCATION-INFORMATION provides information
     about object allocation.
+  * optimization: division of a real float by a complex float is implemented
+    with a specialised code sequence.
+  * optimization: MAKE-INSTANCE with non-constant class-argument but constant
+    keywords is an order of magnitude faster.
+  * optimization: MAKE-INSTANCE with constant keyword arguments is x2-4 faster
+    in the presence of :AROUND or non-standard primary INITIALIZE-INSTANCE
+    methods, and similarly for non-standard metaclass classes as long as there
+    are no methods additional on MAKE-INSTANCE.
   * optimization: more efficient type-checks for FIXNUMs when the value
     is known to be a signed word on x86 and x86-64.
-  * optimization: (EXPT -1 INTEGER) is compiled into (IF (EVENP INTEGER) 1 -1).
-    (thanks to Stas Boukarev)
+  * optimization: compiler now optimizes (EXPT -1 INTEGER), (EXPT -1.0 INTEGER),
+    and (EXPT -1.0d0 INTEGER) into an ODDP test. (thanks to Stas Boukarev and
+    Paul Khuong)
+  * optimization: compiler is smarter about delegating argument type checks to
+    callees.
+  * optimization: several character functions are now compiled somewhat more
+    efficiently. (reported by Lynn Quam)
+  * optimization: the compiler now derives simple types for LOAD-VALUE-FORMs.
+  * improvement: less unsafe constant folding in floating point arithmetic,
+    especially for mixed complex/real -float operations.
+  * optimization: constant double and single floats are stored in native
+    unboxed format on x86[-64].
+  * optimization: smarter code for arithmetic operations with constant floats,
+    complex floats, or integers on x86[-64].
+  * optimization: smarter code for conjugate/multiplication of float complexes
+    and abs/negate of floats on x86-64.
+  * optimization: more efficient complex float and real float operations on
+    x86-64.
+  * improvement: complex float division is slightly more stable.
   * improvement: DESCRIBE output has been reworked to be easier to read and
     contains more pertinent information.
   * improvement: failure to provide requested stack allocation compiler notes
     provided in all cases (requested stack allocation not happening without a
     note being issued is now considered a bug.)
-  * optimization: compiler is smarter about delegating argument type checks to
-    callees.
+  * bug fix: SB-POSIX exports the documented types and functions
+    FILE-DESCRIPTOR and FILENAME, and also the corresponding -DESCRIPTOR
+    types. (reported by "abhi")
   * bug fix: on 64 bit platforms FILL worked incorrectly on arrays with
     upgraded element type (COMPLEX SINGLE-FLOAT), regression from 1.0.28.55.
     (thanks to Paul Khuong)
@@ -36,7 +125,7 @@
   * bug fix: bogus undefined variable warnings from fopcompiled references to
     global variables. (thanks to Lars Rune Nøstdal)
   * bug fix: foreign function names should now appear in backtraces on
-    FC6 as well. (reported by Tomasz Skutnik and obias Rautenkranz)
+    FC6 as well. (reported by Tomasz Skutnik and Tobias Rautenkranz)
   * bug fix: SETF compiler macro documentation strings are not discarded
     anymore.
   * bug fix: GENTEMP is now unaffected by pretty printer dispatch table.
@@ -44,6 +133,8 @@
   * bug fix: SLEEP accepts large integer arguments, truncating them to
     SIGNED-WORD on the assumption that sleeping for 68 years is sufficient
     for anyone. (reported by Leslie Polzer, thanks to Stas Boukarev)
+  * bug fix: compiler notes for expensive slot type checks could be emitted
+    at runtime MAKE-INSTANCE calls. (reported by Samium Gromoff)
 
 changes in sbcl-1.0.29 relative to 1.0.28:
   * IMPORTANT: bug database has moved from the BUGS file to Launchpad