-changes in sbcl-0.7.0 relative to sbcl-0.6.13:
-* major incompatible change: The default fasl file extension, i.e. the
- default extension for files produced by COMPILE-FILE, has changed
- to ".fasl", for all architectures. (No longer ".x86f" and ".axpf".)
-* compiler changes:
- ** There are many changes in the implementation of the compiler.
- SBCL is now essentially a compiler-only implementation of ANSI
- Common Lisp. EVAL still "interprets" a few special cases, but
- almost all the interesting cases are handled by creating
- a LAMBDA expression, calling COMPILE on it, then calling
- FUNCALL on the result.
- ** The EVAL-WHEN code has been rewritten to be ANSI-compliant, and
- various related bugs (IR1-1, IR1-2, IR1-3, IR1-3a) have gone away.
- Since the code is newer, there might still be some new bugs
- (though not as many as before Martin Atzmueller's fixes:-). But
- the new code is substantially simpler and clearer, and hopefully
- any remaining bugs will be simpler, less fundamental, and more
- fixable then the bugs in the old code.
- ** The revised compiler is still a little unsteady on its feet.
- In particular,
- *** The debugging information it produces (particularly the names
- of FUNCTION objects) is sometimes much less useful than what
- the old compiler produced.
- *** The support for inlining FOO when you (DECLAIM (INLINE FOO))
- then do (DEFUN FOO ..) in a non-null lexical environment (e.g.
- within a MACROLET) has been temporarily weakened.
- ** There are new compiler optimizations for various functions:
- *** the sequence functions FIND, POSITION, FIND-IF, POSITION-IF,
- FIND-IF-NOT, POSITION-IF-NOT, and FILL
- *** the math functions TRUNCATE, FLOOR, and CEILING
- *** the function-of-all-trades COERCE
- Mostly these should be transparent, but there's one
- potentially-annoying problem (bug 117): when the compiler
- inline-expands a function and does type analysis on the result,
- it can create control paths which have type mismatches, and
- when it can't prove that those control paths aren't taken,
- it will issue WARNINGs about the type mismatches. This is
- a particular problem in practice for the new sequence functions.
- It's not clear how this should be fixed, and for now, a
- workaround is given in the entry for 117 in the BUGS file.
- ** (Because of the interaction between the two previous items --
- occasional inlining problems and new inline expansions -- some
- of the new sequence function optimizations won't really kick in
- completely until debugging information, and then inlining, are
- straightened out in some future version.)
-* minor incompatible changes:
- ** As part of a bug fix by Christophe Rhodes to DIRECTORY behavior,
- DIRECTORY no longer implicitly promotes NIL slots of its
- pathname argument to :WILD. In particular, when you ask for the
- contents of a directory (which you used to be able to do without
- explicit wildcards, e.g. (DIRECTORY "/tmp/")) you now need to use
- explicit wildcards, e.g. (DIRECTORY "/tmp/*.*").
- ** changes in behavior that ANSI explicitly defines to be
- implementation dependent:
- *** The new compiler-only implementation still conforms with ANSI,
- but acts a little different than before. Besides the obvious
- changes in performance tradeoffs (that the cost per form passed
- to EVAL has gone up, and the cost per form executed by EVAL
- has gone down), the behavior of the system changes a little
- because there are no longer any interpreted function objects.
- COMPILED-FUNCTION-P is now synonymous with FUNCTIONP, and
- e.g. doing COMPILE on the output of interactive DEFUN is
- now a no-op.
- *** The value of INTERNAL-TIME-UNITS-PER-SECOND has been increased
- from 100 to 1000.
- *** The default for the USE list in MAKE-PACKAGE and DEFPACKAGE
- has changed from (:CL) to NIL.
- *** The CHAR-NAME of unprintable ASCII characters which, unlike
- e.g. #\Newline and #\Tab, don't have names specified in the
- ANSI Common Lisp standard, is now based on their ASCII symbolic
- names (#\Nul, #\Soh, #\Stx, etc.) The old CMU-CL-style names
- (#\Null, #\^a, #\^b, etc.) are still accepted by NAME-CHAR, but
- are no longer used for output.
- ** changes in internal implementation constants:
- *** The default value of *BYTES-CONSED-BETWEEN-GCS* has doubled, to
- 4 million. (If your application spends a lot of time GCing and
- you have a lot of RAM, you might want to experiment with
- increasing it even more.)
- ** The SB-C-CALL package has been merged into the SB-ALIEN package.
- However, almost all old code should still continue to work without
- immediate update, as SB-C-CALL is now a (deprecated) nickname
- for SB-ALIEN.
- ** Old operator names in the style DEF-FOO are now deprecated in
- favor of new corresponding names DEFINE-FOO, for consistency with
- the naming convention used in the ANSI standard (DEFSTRUCT, DEFVAR,
- DEFINE-CONDITION, DEFINE-MODIFY-MACRO..). This mostly affects
- internal symbols, but a few supported extensions like
- SB-ALIEN:DEF-ALIEN-FUNCTION are also affected. (So e.g.
- DEF-ALIEN-FUNCTION becomes DEFINE-ALIEN-FUNCTION.)
- ** The debugger prompt sequence now goes "5]", "5[2]", "5[3]",
- etc. as you get deeper into recursive calls to the debugger
- command loop, instead of the old "5]", "5]]", "5]]]"
- sequence. (I was motivated to do this when squabbles between
- ILISP and SBCL left me very deeply nested in the debugger. In the
- short term, this change will probably provoke more ILISP/SBCL
- squabbles, but hopefully it will be an improvement in the long run.)
- ** SB-ALIEN:DEFINE-ALIEN-FUNCTION (also known by the old deprecated
- name DEF-ALIEN-FUNCTION) now does DECLAIM FTYPE for the defined
- function, since declaiming return types involving aliens is
- (1) annoyingly messy to do by hand and (2) vital to efficient
- compilation of code which calls such functions.
- ** SB-ALIEN:LOAD-FOREIGN and SB-ALIEN:LOAD-1-FOREIGN are no
- longer reexported by the SB-EXT package. They're solely useful
- for alien code, so it seems more logical that you should get
- them from the SB-ALIEN package, not in SB-EXT.
- ** :SB-CONSTRAIN-FLOAT-TYPE, :SB-PROPAGATE-FLOAT-TYPE, and
- :SB-PROPAGATE-FUN-TYPE are no longer considered to be optional
- features. Instead, the code that they used to control is always
- built into the system.
-* many other bug fixes
- ** DEFSTRUCT and DEFCLASS have been substantially updated to take
- advantage of the new EVAL-WHEN stuff and to clean them up in
- general, and they are now more ANSI-compliant in a number of
- ways. Martin Atzmueller is responsible for a lot of this.
- ** Besides the cleanups discussed above, Martin Atzmueller fixed
- several other bugs:
- *** fixes in READ-SEQUENCE and WRITE-SEQUENCE
- *** correct ERROR type for various file operations
- *** some fixes for Lisp streams
- *** DEFMETHOD syntax checking
- *** changing old weird representation of debug information as
- strings (which, among their other deficiencies, don't transform
- correctly when you rename packages, and don't change their
- print representation when you change things like *PACKAGE*
- and *PRINT-LENGTH*) to symbols and lists of symbols
- He also made several improvements and fixed several bugs in DESCRIBE.
- ** Alexey Dejneka fixed many bugs, including classic bugs and bugs he
- discovered himself:
- *** misbehavior of WRITE-STRING/WRITE-LINE
- *** LOOP over keys of a hash table, LOOP bugs 49b and 81 and 103,
- and several other LOOP problems as well
- *** DIRECTORY when similar filenames are present
- *** DEFGENERIC with :METHOD options
- *** bug 126, in (MAKE-STRING N :INITIAL-ELEMENT #\SPACE))
- *** bug in the optimization of ARRAY-ELEMENT-TYPE
- *** argument ordering in FIND with :TEST option
- *** mishandled package designator argument in APROPOS-LIST
- *** various problems in the backquote readmacro
- *** a bug in APROPOS
- *** probably some others that I'm not describing very well here,
- since the CVS log documents them by reference to sbcl-devel
- messages, and the SourceForge archives aren't working well.:-(
- ** Dan Barlow improved the Alpha port (and is making progress on the
- PPC port, for those of you who think different).
- ** Besides the DIRECTORY fixes and changes mentioned elsewhere,
- Christophe Rhodes cleaned up the system self-test scripts (in tests/*),
- contributed the optimization of FIND-IF-NOT and POSITION-IF-NOT, and
- continues to work on the SPARC port (for those of you in a position
- to look down upon our little PC-compatible boxes from a great height).
- ** PPRINT-LOGICAL-BLOCK now copies the *PRINT-LINES* value on entry
- and uses that copy, rather than the current dynamic value, when
- it's trying to decide whether to truncate output. Thus e.g.
- (let ((*print-lines* 50))
- (pprint-logical-block (stream nil)
- (dotimes (i 10)
- (let ((*print-lines* 8))
- (print (aref possiblybigthings i) stream)))))
- should now truncate the logical block only at 50 lines, instead of
- often truncating it at 8 lines, as it did before.
-* The doc/cmucl/ directory, containing old CMU CL documentation
- from the time of the fork, is no longer part of the base system.
- SourceForge has shut down its anonymous FTP service, and with it
- my original plan for distributing the old CMU CL documentation
- there. For now, if you need these files you can download an old
- SBCL source release and extract them from it.
-* The fasl file version number changed again, for dozens of reasons,
- some of which are apparent above.
-
-changes in sbcl-0.7.1 relative to sbcl-0.7.0:
-* mostly bug fixes:
- ** SB-ALIEN:LOAD-FOREIGN and SB-ALIEN:LOAD-1-FOREIGN are set
- up properly again. (There was a packaging bug in 0.7.0 which
- left their definitions in SB-SYS::LOAD-FOREIGN and
- SB-SYS::LOAD-1-FOREIGN. LOAD-FOREIGN and LOAD-1-FOREIGN are
- vital for most things which interface to C-level interfaces,
- like extensions working with sockets or databases or
- Perl-compatible regexes or whatever, and the need to fix
- this bug is the main reason that 0.7.1 was released so
- soon after 0.7.0.)
- ** DEFGENERIC is now choosier about the methods it redefines, so that
- reLOADing a previously-LOADed file containing DEFGENERICs does
- the right thing now. Thus, the Lispy edit/reLOAD-a-little/test
- cycle now works as it should. (thanks to Alexey Dejneka)
- ** Bug 106 (types (COMPLEX FOO) where FOO is an obscure type) was
- fixed by Christophe Rhodes. (He actually submitted this patch
- months ago, and I delayed until after 0.7.0.)
- ** Bug 111 (internal compiler confusion about runtime checks on
- FUNCTION types) was fixed by Alexey Dejneka.
-* Some internal cleanups (getting rid of variables which aren't
- needed now that the byte interpreter is gone) caused the fasl
- file format number to change again.
-
-changes in sbcl-0.7.2 relative to sbcl-0.7.1:
- * incompatible change: The compiler is now less aggressive about
- tail call optimization, doing it only when (> SPACE DEBUG) or
- (> SPEED DEBUG). (This is an incompatible change because there are
- programs which relied on the old CMU-CL-style behavior to optimize
- away their unbounded recursion which will now die of stack overflow.)
- * minor incompatible change: The default BYTES-CONSED-BETWEEN-GCS
- for non-GENCGC systems has been increased to 20M (since that
- seems much closer to the likely performance optimum for modern
- systems than the old 4M value was)
- * minor incompatible change: new larger values for *DEBUG-PRINT-LENGTH*
- and *DEBUG-PRINT-LEVEL*
- * SBCL runs on SPARC systems now. (thanks to Christophe Rhodes' port
- of CMU CL's support for SPARC, and various endianness and other
- SBCL portability fixes due to Christophe Rhodes and Dan Barlow)
- * new syntactic sugar for the Unix command line: --load foo.bar is now
- an alternate notation for --eval '(load "foo.bar")'.
- * bug fixes:
- ** The system now detects stack overflow and handles it gracefully,
- at least for (OR (> SAFETY (MAX SPEED SPACE)) (= SAFETY 3))
- optimization settings. (This is a good thing in general, and
- its introduction in this version should be particularly timely
- for anyone whose code fails because of suppression of tail
- recursion!)
- ** The system now hunts for the C variable "environ" in a more
- devious way, to avoid segfaults when the C library version differs
- between compile time and run time. (thanks to Christophe Rhodes)
- ** INTEGER-valued CATCH tags now work. (thanks to Alexey Dejneka,
- and also to Christophe Rhodes for porting the fix to non-X86 CPUs)
- ** The compiler no longer issues bogus style warnings for undefined
- classes in the same source file as the DEFCLASSes which defined
- them. (thanks to Stig E Sandoe for reporting and Martin Atzmueller
- for fixing this)
- ** fixes in CONDITION class precedence list for undefined function
- errors (thanks to Alexei Dejneka)
- ** *DEFAULT-PATHNAME-DEFAULTS* is used more consistently and
- correctly. (thanks to Dan Barlow)
- ** portability fixes aiming at bootstrapping under CLISP (thanks
- to Dave McDonald and Christophe Rhodes)
- ** FORMAT fixes (thanks to Robert Strandh and Dan Barlow)
- ** fixes in type translation and and type inference (thanks to
- Christophe Rhodes)
- ** fixes to optimizer internal errors (thanks to Alexei Dejneka)
- ** various fixes in the new ports (thanks to Dan Barlow)
- * several changes related to debugging:
- ** suppression of tail recursion, as noted above
- ** stack overflow detection, as noted above
- ** The default implementation of TRACE has changed. :ENCAPSULATE T
- is now the default. (For some time encapsulation has been more
- reliable than the breakpoint-based :ENCAPSULATE NIL
- implementation, at least on X86 systems; and I just noticed that
- encapsulation also seems closer to the spirit of the ANSI
- specification.)
+changes in sbcl-0.8.4 relative to sbcl-0.8.3:
+ * incompatible change: The --disable-debugger command line
+ option now clobbers the debugger at a more fundamental
+ level, by redefining #'INVOKE-DEBUGGER instead of by
+ rebinding *DEBUGGER-HOOK*. The main difference is that BREAK
+ is specified by ANSI to ignore *DEBUGGER-HOOK* and
+ INVOKE-DEBUGGER regardless. Under the old system, BREAK would
+ enter the debugger REPL and then suffer recursive errors
+ because *DEBUG-IO* is also messed up in --disable-debugger mode;
+ while under the new system, BREAK in --disable-debugger mode
+ terminates the system just as an unhandled error would.
+ * fixed compiler performance when processing loops with a step >1;
+ * bug fix: DOCUMENTATION now retrieves generic function
+ documentation. Also, DOCUMENTATION and (SETF DOCUMENTATION)
+ support has been systematized, and now supports the methods
+ specified by ANSI, along with a default method and a method for
+ slot documentation. (reported by Nathan Froyd)
+ * bug fix: effective methods associated with a generic function are
+ no longer cached over a change of that generic function's method
+ combination. (reported by Andreas Fuchs)
+ * bug fix: RUN-PROGRAM now does not fail if some element in $PATH
+ names a non-existent directory. (thanks to Andreas Fuchs)
+ * bug fix: ROUND and TRUNCATE could, under certain circumstances on
+ the PPC platform, lead to stack corruption; this has been fixed.
+ (reported by Rainer Joswig)
+ * bug fix: ASH on an (UNSIGNED-BYTE 32) with a shift of -32 or lower
+ no longer ever returns 1 instead of 0. (thanks to Lars Brinkhoff)
+ * fixed bug 285: TRUNCATE on bignum arguments, and indeed bignum
+ arithmetic in general, is now much more reliable on the PPC
+ platform.
+ * bug fix: LOGCOUNT on (UNSIGNED-BYTE 64) objects on the Alpha
+ platform now returns the right answer.
+ * optimization: restored some effective method precomputation in
+ CLOS (turned off by an ANSI fix in sbcl-0.8.3); the amount of
+ precomputation is now tunable.
+ * optimization: compiler-internal data structure use has been
+ reviewed, and changes have been made that should improve the
+ performance of the compiler by about 20%.
+ * optimization: performance of FILL (and :INITIAL-ELEMENT) on
+ simple-base-strings and simple-bit-vectors is improved.
+ * optimization: the optimization of 32-bit logical and arithmetic
+ functions introduced in version 0.8.3 on the x86 has been
+ implemented on the mips, ppc and sparc platforms; an
+ implementation of the same facility, but for 64-bit arithmetic,
+ has been added for the alpha.
+ * microoptimization: the compiler is better able to make use of the
+ x86 LEA instruction for multiplication by constants.
+ * bug fix: in some situations compiler did not report usage of
+ generic arithmetic in (SPEED 3) policy.
+ * bug 145b fix: compiler used wrong type specifier while converting
+ MEMBER-types to numeric.
+ * bug fix: COMPILE-FILE must bind *READTABLE*. (reported by Doug
+ McNaught)
+ * bug fix: (SETF AREF) on byte-sized-element arrays with constant
+ index argument now works properly on the Alpha platform.
+ * bug fix: floating point exception treatment on the Alpha platform
+ is improved.
+ * bug fix: FILE-POSITION works much better on string input and
+ output streams. (thanks to Nikodemus Siivola)
+ * bug fix: many threading/garbage collection symptoms sorted.
+ SB-THREAD:INTERRUPT-THREAD now safe to call on a thread that
+ might be pseudo-atomic.
+ * internal change: Stopping for GC is now done with signals not
+ ptrace. GC is now done in whichever thread wanted it, instead of
+ in the parent.
+ * bug fix: GC hooks (missing since 0.8) reinstated, so finalizers
+ work again.
+ * bug fix: result form in DO is not contained in the implicit
+ TAGBODY.
+ * incompatible change: ICR structure is changed; the value part of
+ CONTINUATION is now called LVAR; corresponding functions are
+ renamed (e.g. SB-C::CONTINUATION-TYPE has become SB-C::LVAR-TYPE).
+ * added type deriver for ISQRT (thanks to Robert E. Brown).
+ * bug fix: better support for loading from the command line when an
+ initialization file sets (READTABLE-CASE *READTABLE*). (thanks
+ to Adam Warner)
+ * fixed some bugs revealed by Paul Dietz' test suite:
+ ** the RETURN clause in LOOP is now equivalent to DO (RETURN ...).
+ ** ROUND and FROUND now give the right answer when given very
+ small float arguments.
+ ** (FLOAT X) for X of type DOUBLE-FLOAT now returns X in all
+ circumstances.
+ ** optimizer for (EXPT X 0) did not work for X not of type FLOAT.
+ ** (GCD 0 <negative-integer>) returned <negative-integer>.
+ ** LCM should return a non-negative integer.
+ ** PARSE-INTEGER returned the index of a terminator instead of the
+ upper bounding index of a substring in case :JUNK-ALLOWED NIL.
+ ** PARSE-INTEGER returned an incorrect index being applied to a
+ displaced string.
+ ** LCM with two arguments of 0 returns 0 rather than signalling
+ DIVISION-BY-ZERO.
+ ** unsigned addition of a 32-bit constant with the high bit set no
+ longer causes an internal compiler error on the x86.
+ ** LOGBITP accepts a non-negative bignum as its INDEX argument.
+ ** compiler incorrectly derived types of DPB and DEPOSIT-FIELD
+ with negative last argument.
+ ** byte specifiers with zero size and position no longer cause
+ an error during type derivation.
+ ** bignum multiplication on the Alpha platform now returns the
+ right answer.
+ * porting: The system now builds on SuSE AMD64, although it still
+ generates a 32-bit binary.
+ * .fasl file incompatibility: The fasl file version number has
+ been incremented (because of the changes to internal compiler
+ data structures referred to above).