sbcl.git
21 years ago0.8.0.5:
Alexey Dejneka [Mon, 26 May 2003 14:42:23 +0000 (14:42 +0000)]
0.8.0.5:
        * Deleted obsolete bug 244;
        * optimizer for ARRAY-HEADER-P knows about 0-dimensional
          arrays;
          ... fixed bug 250.

21 years ago0.8.0.4:
Alexey Dejneka [Mon, 26 May 2003 08:17:13 +0000 (08:17 +0000)]
0.8.0.4:
        Fixed bug 249: local functions did not check type of unused
        arguments.

21 years ago0.8.0.3:
Alexey Dejneka [Mon, 26 May 2003 04:25:52 +0000 (04:25 +0000)]
0.8.0.3:
        Merged CAST branch.

        Changes since -cast.8:
        * separated usage of object and values types;
        * fixed warning and error reports for compile-time type
          errors;
        * inline structure slot accessors are implemented with source
          transforms;
        * enabled warning emitting for type errors in some paths to
          CAST;
        * removed check for type errors in arguments of a call of a
          flushable function;
        * source transforms are made nameless.

21 years ago0.8.0.2:
Christophe Rhodes [Sun, 25 May 2003 22:34:23 +0000 (22:34 +0000)]
0.8.0.2:
Fix stack exhaustion stack exhaustion death
... define DEFINE-FUNCTION-NAME-SYNTAX function-name-defining macro;
... use it for SETF functions, and define LEGAL-FUNCTION-NAME-P
and FUN-NAME-BLOCK-NAME in terms of VALID-FUNCTION-NAME-P;
... also define internal PCL generalized function name syntax as
such, and test for internalness in SET-ARG-INFO1;
... OAOO bonus: delete bits of SB!PCL::CLASS-PREDICATE that were
decorating the compiler;

(note: this API is interface-compatible with CMUCL's for defining
generalized function name syntax.  However, it's not currently exported
from SB-EXT because I happen to think that calling something
VALID-FUNCTION-NAME-P when it returns two values, the second of which
is syntactically significant, is a bit lame, and maybe we'll be able
to agree a better name between the two projects)

21 years ago0.8.0.1:
Christophe Rhodes [Sun, 25 May 2003 22:26:13 +0000 (22:26 +0000)]
0.8.0.1:
Fix build on FreeBSD
... don't #include <proc.h>, which defines a struct thread; (thanks
to Henrik Motakef)
... do sigaltstack() if #+c-stack-is-control-stack (which is, in
practice, all the time).

(this /may/ break building on OpenBSD, which uses the same OS-dependent
routines; it is possible that OpenBSD needs something from <proc.h>;
if so, we need to find some compromise)

21 years ago0.8.0:
William Harold Newman [Sun, 25 May 2003 15:11:35 +0000 (15:11 +0000)]
0.8.0:
release, tagged as sbcl_0_8_0

21 years ago0.8alpha.0.45:
Christophe Rhodes [Thu, 22 May 2003 16:46:58 +0000 (16:46 +0000)]
0.8alpha.0.45:
Fix profiler bug reported by APD sbcl-devel 2003-05-21:
... update *N-BYTES-FREED-OR-PURIFIED* in SUB-GC;
... add a smoke test.
Increment FASL file version number due to change in
DEFSTRUCT-SLOT-DESCRIPTION structure.

21 years ago0.8alpha.0.44:
Kevin Rosenberg [Thu, 22 May 2003 16:00:20 +0000 (16:00 +0000)]
0.8alpha.0.44:
    sb-aclrepl/inspect.lisp: Change usage of dsd-%name to dsd-name

21 years ago0.8alpha.0.43:
Kevin Rosenberg [Tue, 20 May 2003 16:15:18 +0000 (16:15 +0000)]
0.8alpha.0.43:
    contrib/sb-aclrepl:
       Rework newline handling to ensure a fresh line
       Correct README file

21 years ago0.8alpha.0.42:
Christophe Rhodes [Tue, 20 May 2003 14:42:25 +0000 (14:42 +0000)]
0.8alpha.0.42:
improvements to sb-simple-streams contrib (from Rudi Schlatte)
... don't spam *features* anymore
... writes of large chunks of data work now, instead of
failing silently

21 years ago0.8alpha.0.41:
Christophe Rhodes [Tue, 20 May 2003 10:49:26 +0000 (10:49 +0000)]
0.8alpha.0.41:
Firefighting the build, part II
... remove DSD-%NAME optimization, in the interest of making
SLOT-VALUE (and hence MAKE-LOAD-FORM-SAVING-SLOTS)
a more reliable operation.
... now the name of the slot is the symbol in the DEFSTRUCT
form, as expected; also, now the CL package is pristine,
containing only the 978 exported symbols.
Essentially this version has built from CMUCL and built itself
successfully.

21 years ago0.8alpha.0.40:
Christophe Rhodes [Tue, 20 May 2003 10:36:02 +0000 (10:36 +0000)]
0.8alpha.0.40:
Firefighting the build, part I
... make SLOT-VALUE work on :READ-ONLY T structure slots
Notes:

* This version may have a subtle breakage that may or may not
bite.  The forthcoming 0.8alpha.0.41 commit will fix that
subtle breakage, but it was fixed chronologically before in my
tree, so I haven't taken out that fix to test this one in
isolation.  I hope that makes sense...)

* The new implementation of MAKE-LOAD-FORM-SAVING-SLOTS seems
very slow; this may be a perceptual problem.  What is
incontrovertible is that it is very noisy; it chatters about
compiling many top-level forms, caused by PCL generating
LOAD-TIME-VALUE forms for ENSURE-ACCESSOR to optimize SLOT-VALUE.
A fix for this, at least for structure objects, will probably be
forthcoming in 0.8alpha.0.4x.

21 years ago0.8alpha.0.39:
Christophe Rhodes [Mon, 19 May 2003 16:25:10 +0000 (16:25 +0000)]
0.8alpha.0.39:
A couple more CLOS fixes:
... make &OPTIONAL argument count checking less lax in methods
(caught by pfdietz' MAKE-LOAD-FORM.ERROR.2)
... make :ARGUMENT-PRECEDENCE-ORDER and :METHOD-COMBINATION
DEFGENERIC options do sanity checking on their arguments
(:A-P-O caught by pfdietz' suite; :M-C
checking defensively installed :-)

21 years ago0.8alpha.0.38:
Christophe Rhodes [Mon, 19 May 2003 14:47:14 +0000 (14:47 +0000)]
0.8alpha.0.38:
Slight change to REQUIRE/PROVIDE protocol
... as observed by Tony Martinez sbcl-devel 2003-05-13, REQUIRE
takes a string designator, so allow this
... update to latest ASDF, which changes the asdf hook slightly
such that individual modules are required (ha!) to
PROVIDE themselves; make it so.

21 years ago0.8alpha.0.37:
Christophe Rhodes [Mon, 19 May 2003 14:05:17 +0000 (14:05 +0000)]
0.8alpha.0.37:
Make MAKE-LOAD-FORM(-SAVING-SLOTS) vaguely conform
... and in the process, remind myself of just how horrible the
:JUST-DUMP-IT-NORMALLY hack was.
... more methods on MAKE-LOAD-FORM;
... real, CLOS-based introspective definition of
MAKE-LOAD-FORM-SAVING-SLOTS...
... which means that we have to hold off from using MLFSS until
it's around, so make JUST-DUMP-IT-NORMALLY use
:SB-JUST-DUMP-IT-NORMALLY rather than MLFSS in its
definition for the target.
Fix the type.impure.lisp test for the new definition of
condition classes (oops).

21 years ago0.8alpha.0.36:
Christophe Rhodes [Mon, 19 May 2003 10:51:32 +0000 (10:51 +0000)]
0.8alpha.0.36:
A little tender loving care, applied to conditions:
... make the implementation of DEFINE-CONDITION agree with the
documentation string: allow :DOCUMENTATION slot options
to work.
... ANSI (and pfdietz :-) wants SLOT-EXISTS-P to work on
conditions; hook condition objects into CLOS enough to
talk about existence of slots: (new classes
CONDITION-{EFFECTIVE,DIRECT}-SLOT-DEFINITION,
CONDITION-CLASS, etc)
... it's a bit ridiculous to have SLOT-EXISTS-P working on
conditions, and then not be able to do SLOT-VALUE, so
do the work necessary to make CONDITION objects
more-or-less fully understood by PCL: (new methods on
COMPUTE-SLOTS, ALLOCATE-INSTANCE, SLOT-VALUE-USING-CLASS
and friends; new clauses in internal functions such as
GET-OPTIMIZED-STD-SLOT-VALUE-USING-CLASS-METHOD-FUNCTION;
adjustment of the braid to set up CLOS knowledge of the
new class hierarchy).

21 years ago0.8alpha.0.35:
Christophe Rhodes [Sun, 18 May 2003 15:49:11 +0000 (15:49 +0000)]
0.8alpha.0.35:
Commit patch from Rudi Schlatte for sb-simple-streams contrib
... add TODO
... implement socket streams (using sb-bsd-sockets)
... (frob by CSR: don't fail if we get a connection refused)

21 years ago0.8alpha.0.34:
Christophe Rhodes [Sun, 18 May 2003 14:42:33 +0000 (14:42 +0000)]
0.8alpha.0.34:
Make ALLOCATE-INSTANCE work on all structure classes
... if we're defined by a DEFSTRUCT, then make a closure to
allocate an instance.

21 years ago0.8alpha.0.33:
Christophe Rhodes [Fri, 16 May 2003 14:42:35 +0000 (14:42 +0000)]
0.8alpha.0.33:
Add one more contrib
... SB-MD5, based on Pierre Mai's highly-frobbed implementation of
RFC1321;
... a couple of cosmetic changes to rotate-byte to accommodate the
demands of code that wants to use it.

21 years ago0.8alpha.0.32:
Christophe Rhodes [Fri, 16 May 2003 11:37:09 +0000 (11:37 +0000)]
0.8alpha.0.32:
Improvements in build technology:
... detect and choose GNUMAKE earlier, once and only once.
... provide recompile and use-anyway restarts for warm load.

21 years ago0.8alpha.0.31:
Kevin Rosenberg [Thu, 15 May 2003 16:21:41 +0000 (16:21 +0000)]
0.8alpha.0.31:
    * sb-aclrepl/sb-aclrepl.asd: Remove compilation note workaround
    * sb-aclrepl/repl.lisp: add a few needed newlines to output

21 years ago0.8alpha.0.30:
Christophe Rhodes [Thu, 15 May 2003 16:03:33 +0000 (16:03 +0000)]
0.8alpha.0.30:
Bandage over compiler overenthusiasm for EQL-specialized methods
... reduction in annoyance for contrib/ users probably more
important than informing developers that they should
be using EQL on SINGLE-FLOAT, not T, for best
performance.

21 years ago0.8alpha.0.29:
Christophe Rhodes [Wed, 14 May 2003 14:16:41 +0000 (14:16 +0000)]
0.8alpha.0.29:
A couple of extremely minor fixes
... "oops" in type declaration in %DEFUN
... fix KLUDGE in IMMEDIATE-CONSTANT-SC for x86, and allow (in
principle) building from lisps with larger
MOST-POSITIVE-FIXNUM as well as smaller.

21 years ago0.8alpha.0.28:
Christophe Rhodes [Tue, 13 May 2003 13:55:29 +0000 (13:55 +0000)]
0.8alpha.0.28:
Fix bug 47d (DEFGENERIC must signal PROGRAM-ERROR when
attempting to create a generic function with the same name as a
special operator).
... sounds easy, huh?  No.
... make COMPILER-ERROR not inherit from ERROR any more, so that
user handlers don't (wrongly) claim to handle it;
... establish a handler for COMPILER-ERROR around the evaluator
that delegates to the compiler handlers if present, but
handles them itself if not...
... by signalling an error from a new internal restart, to allow
user handlers for ERROR and friends a chance to run.

21 years ago0.8alpha.0.27:
Christophe Rhodes [Tue, 13 May 2003 12:21:15 +0000 (12:21 +0000)]
0.8alpha.0.27:
Add skeletal simple-streams contrib from Rudi Schlatte, based on Paul
Foley's implementation for CMUCL.

21 years ago0.8alpha.0.26:
Kevin Rosenberg [Mon, 12 May 2003 18:12:35 +0000 (18:12 +0000)]
0.8alpha.0.26:
     * sb-contrib/sb-aclrepl.asd:
         - Remove work-around for optimization notes
     * sb-contrib/{toplevel,repl}.lisp:
         - Rework fresh-line handling to accomodate that *repl-read-fun* causes
        a newline of which the output-stream is unaware.

21 years ago0.8alpha.0.25:
Christophe Rhodes [Mon, 12 May 2003 14:10:30 +0000 (14:10 +0000)]
0.8alpha.0.25:
A couple of PCL fixes:
... REMOVE-METHOD should always return its generic function
argument.  Make it so.
... SHARED-INITIALIZE should initialize :CLASS slots too.

21 years ago0.8alpha.0.24:
Kevin Rosenberg [Mon, 12 May 2003 04:40:30 +0000 (04:40 +0000)]
0.8alpha.0.24:
   * sb-aclrepl/sb-aclrepl.asd:
        - Work around 'eql method specialization optimization notes
   * sb-aclrepl:repl.lisp:
        - Refactor read-cmd into small functions
        - Add relative history numbers, eg, `:-2'
        - Add history pattern match search, eg, `::foo'
        - Add optional redo query to history command, eg, `:24 ?'

21 years ago0.8alpha.0.23:
Christophe Rhodes [Fri, 9 May 2003 10:22:48 +0000 (10:22 +0000)]
0.8alpha.0.23:
Fix a couple of the CLOS bugs that have been accumulating:
... make :ALLOCATION :CLASS slots behave as they should in slot
inheritance and class redefinition.
Slightly unKLUDGEify the %SET-SYMBOL-VALUE implementation
... do it the same way whether building with #!+SB-THREAD or
not, so at least it's ugly once and only once.

21 years ago0.8alpha.0.22
Daniel Barlow [Thu, 8 May 2003 20:37:54 +0000 (20:37 +0000)]
0.8alpha.0.22
SB-GROVEL-related changes
... make sb-bsd-sockets use it
... which involved restoring size-of-{int,char,long} constants

When building contrib, reset the central-registry to avoid
picking up any code from e.g. ~/.sbcl/systems/

21 years ago0.8alpha.0.21:
Christophe Rhodes [Thu, 8 May 2003 16:17:57 +0000 (16:17 +0000)]
0.8alpha.0.21:
The "uh, I thought we had users beta-testing for us" release:
Make SETQ on globals return a value (and IN-PACKAGE, and ...)
... the SET VOP and the %SET-SYMBOL-VALUE IR2 thingy are
different, so one shouldn't translate the other;
... instead, create an IR2 translator for %SET-SYMBOL-VALUE by
hand, mimicking the effect of :SET-TRANS in
DEFINE-PRIMITIVE-OBJECT;
... this removes the need for late-symbol.lisp, so delete it.
While we're in a deleting mood...
... delete src/pcl/fast-init.lisp, which hasn't been part of the
build for the last 5 months or so.

21 years ago0.8alpha.0.20:
Christophe Rhodes [Thu, 8 May 2003 15:43:21 +0000 (15:43 +0000)]
0.8alpha.0.20:
Make OpenMCL a suitable slamming host:
... fix slam.sh
... :if-exists :supersede for *output-files-for-genesis*
Since OpenMCL builds still work, announce clisp as a suitable host
for the cross-compiler.

21 years ago0.8alpha.0.19
Daniel Barlow [Thu, 8 May 2003 12:44:18 +0000 (12:44 +0000)]
0.8alpha.0.19
We're not using ALTERNATE-SIGNAL-STACK-START, and in fact
haven't been for some time.  Remove it.

21 years ago0.8alpha.0.18:
Christophe Rhodes [Thu, 8 May 2003 11:17:24 +0000 (11:17 +0000)]
0.8alpha.0.18:
Some minor package alterations:
... export CONTROL-STACK-POINTER-VALID-P from SB!VM;
... along with new *{BINDING,CONTROL}-STACK-FOO* to replace old
non-asterisked versions;
make CONTROL-STACK-POINTER-VALID-P not style-warn on each use.

21 years ago0.8alpha.0.17
Daniel Barlow [Wed, 7 May 2003 11:38:35 +0000 (11:38 +0000)]
0.8alpha.0.17
Package frobbing fix from Matthew Danish to make sb-thread
build again

Save errno return from waitpid long enough to print the error,
as the real errno is clobbered by parent_do_garbage_collect()

21 years ago0.8alpha.0.16:
Christophe Rhodes [Wed, 7 May 2003 11:18:59 +0000 (11:18 +0000)]
0.8alpha.0.16:
Fix for SXHASH on condition objects
... was causing compilation failures when referencing explicit
constant conditions

21 years ago0.8alpha.0.15:
William Harold Newman [Wed, 7 May 2003 02:42:04 +0000 (02:42 +0000)]
0.8alpha.0.15:
broke some long lines
merged Matthew Danish's slam-on-clisp patch from sbcl-devel

21 years ago0.8alpha.0.14
Daniel Barlow [Mon, 5 May 2003 23:27:07 +0000 (23:27 +0000)]
0.8alpha.0.14
Merge thread-gc-branch.

Summary: move time-to-gc-p logic entirely into C.  Delete a
lot of Lisp stuff no longer necessary.  Make SUB-GC
thread-safe or at least thread-tolerant.  Some hooks and
variables that were previously available but not apparently
used for much are now no longer present.

21 years ago0.8alpha.0.13:
Christophe Rhodes [Mon, 5 May 2003 14:42:08 +0000 (14:42 +0000)]
0.8alpha.0.13:
(oops: add file needed from last commit)

21 years ago0.8alpha.0.13:
Christophe Rhodes [Mon, 5 May 2003 14:09:03 +0000 (14:09 +0000)]
0.8alpha.0.13:
CLISP build megapatch
... mostly putting #-SB-XC in front of :COMPILE-TOPLEVEL,
because clisp gives a full warning for function and
macro redefinition;
... workaround clisp's buggy pretty printer by not exercising it
as much: use (INHIBIT-WARNINGS 3);
... explicit :INITIAL-ELEMENT 0 when we're using 0 to mean
"uninitialized" in MAKE-ARRAY;
... SPECIAL-OPERATOR-P isn't a good test on the host for what
can become a target macro;
... slightly more portable floating point logic:
Explicitly set *READ-DEFAULT-FLOAT-FORMAT* so that we
don't create host LONG-FLOATs by accident;
LOAD-TIME-VALUE magic for negative floating point zeros;
Minor associated text file frobbage
... braindump some unrelated TODO items
Obligatory runtime code improvement
... fix one warning in gencgc.h

21 years ago0.8aplha.0.12:
Alexey Dejneka [Mon, 5 May 2003 06:31:53 +0000 (06:31 +0000)]
0.8aplha.0.12:
        Don't emit a note about inefficient return convention for a
        function, which does not return at all.

21 years ago0.8alpha.0.11:
Christophe Rhodes [Sun, 4 May 2003 19:20:33 +0000 (19:20 +0000)]
0.8alpha.0.11:
Commit laggardly test for 0.8alpha.0.4 breakage

21 years ago0.8alpha.0.10:
Alexey Dejneka [Sun, 4 May 2003 16:52:32 +0000 (16:52 +0000)]
0.8alpha.0.10:
        * &ENVIRONMENT argument in macro lambda list is bound first
          (found by Paul Dietz);
        * Added checking for duplicate variables in macro lambda lists.

21 years ago0.8alpha.0.9:
William Harold Newman [Sat, 3 May 2003 18:19:43 +0000 (18:19 +0000)]
0.8alpha.0.9:
defined WITH-UNIQUE-NAMES
grepped for 'gensym "', and used WITH-UNIQUE-NAMES instead
where it seemed more convenient
tweaked miscellaneous text I noticed in my greppage
added test case for just-fixed compiler bug
bugfix: one last s/layout-class/layout-classoid/

21 years ago0.8alpha.0.8:
Christophe Rhodes [Sat, 3 May 2003 15:32:27 +0000 (15:32 +0000)]
0.8alpha.0.8:
Delete NEGATIVE-ZERO-IS-NOT-ZERO feature conditional, and all
code compiled when it is active, as (following discussions with
Raymond Toy) it has been superseded by accurate MEMBER type
methods.
... mention its loss in NEWS, just in case anyone has actually
been using it (highly unlikely).

21 years ago0.8alpha.0.7:
Christophe Rhodes [Sat, 3 May 2003 13:02:58 +0000 (13:02 +0000)]
0.8alpha.0.7:
Fix for (VALUES) from FUN-TYPE reorganization
... treat a bare FUNCTION as (FUNCTION * *) in VALID-FUN-USE
... add a comment for the future to consider whether it's
possible that an intersection type could get in there.

21 years ago0.8aplha.0.6:
Alexey Dejneka [Sat, 3 May 2003 12:14:09 +0000 (12:14 +0000)]
0.8aplha.0.6:
        SB!KERNEL::COMPUTE-EFFECTIVE-SLOTS: compare names to names,
        not to slots (thanks to Gerd Moellmann for review).

21 years ago0.8aplha.0.5:
Alexey Dejneka [Sat, 3 May 2003 07:05:24 +0000 (07:05 +0000)]
0.8aplha.0.5:
        Fixed bug reported by WHN 2003-05-02:
        * CHANGE-REF-LEAF incompatibly changes REF derived type in
          BASIC-COMBINATION-FUN.

21 years ago0.8alpha.0.4:
Christophe Rhodes [Fri, 2 May 2003 14:56:38 +0000 (14:56 +0000)]
0.8alpha.0.4:
Canonicalize FUNCTION and VALUES types
... MAKE-FUN-TYPE and MAKE-VALUES-TYPE wrappers
... remove MAKE-VALUES-TYPE cache on grounds of it confusing me
... FUN-TYPE-NARGS needs to check to see if WILD-ARGS is set
[e.g. for LIST, defined as (FUNCTION (&REST T) LIST),
canonicalized to (FUNCTION * LIST)]
Ensure that FUN-TYPES and subtypes of FUNCTION aren't treated as
disjoint
... new type methods for FUNCTION :COMPLEX-{INTERSECT,UN}ION
Make the MAP/%MAP transform arglist look the same as their
DEFKNOWN
... not actually sure if this is necessary any more, but at one
point I got very strange errors without it.

Handle this one with slight care.  I'm not 100% convined that
it's bulletproof, though it passes tests and builds itself
without complaint.

21 years ago0.8alpha.0.3:
Kevin Rosenberg [Wed, 30 Apr 2003 19:55:03 +0000 (19:55 +0000)]
0.8alpha.0.3:
      * src/pcl/std-class.lisp, src/pcl/generic-functions.lisp:
        Change signature of {direct,effective}-slot-definiton-class
        to match AMOP

21 years ago0.8alpha.0.2:
Christophe Rhodes [Wed, 30 Apr 2003 19:37:09 +0000 (19:37 +0000)]
0.8alpha.0.2:
Fix for methods on classes with incomplete subclasses (detected
by KMR/McCLIM compilation)
... distilled test for it.

21 years ago0.8alpha.0.1:
Alexey Dejneka [Wed, 30 Apr 2003 16:29:45 +0000 (16:29 +0000)]
0.8alpha.0.1:
SB-C::REINIT-LAMBDA-PHYSENV: Always clean INDIRECT flag on
lambda vars.

21 years ago0.8alpha.0:
William Harold Newman [Wed, 30 Apr 2003 15:07:32 +0000 (15:07 +0000)]
0.8alpha.0:
release, tagged as sbcl_0_8alpha_0

21 years ago0.pre8.119:
Christophe Rhodes [Tue, 29 Apr 2003 13:19:10 +0000 (13:19 +0000)]
0.pre8.119:
Sneaking in despite the freeze: fix the long form of
DEFINE-METHOD-COMBINATION, which is allowed a documentation
string.

21 years ago0.pre8.118
Daniel Barlow [Tue, 29 Apr 2003 00:28:42 +0000 (00:28 +0000)]
0.pre8.118
Fix WITH-MUTEX bug: don't release the mutex if we didn't
manage to acquire it in the first place

21 years ago0.pre8.117:
Christophe Rhodes [Mon, 28 Apr 2003 12:57:02 +0000 (12:57 +0000)]
0.pre8.117:
:ALLOW-OTHER-KEYS NIL is perfectly valid in an initarg list, so
don't flame out when it's present (detected by Paul Dietz'
suite: test CLASS-07.10)

21 years ago0.pre8.116:
Christophe Rhodes [Mon, 28 Apr 2003 12:23:44 +0000 (12:23 +0000)]
0.pre8.116:
Fix for :DEFAULT-INITARGS with side-effects (detected by Paul
Dietz' tests, fix from Gerd Moellmann)
... ctor needed to be smarter about the separation between
locations and initarg equality
Also really add the hyperobject tests (logically part of
sbcl-0.pre8.115)

21 years ago0.pre8.115:
Christophe Rhodes [Mon, 28 Apr 2003 11:00:22 +0000 (11:00 +0000)]
0.pre8.115:
Fix for SLOT-VALUE inside (SETF SLOT-VALUE-USING-CLASS), from
Gerd Moellmann (test case a simplified version of KMR's
hyperobject)
... don't pass a NIL wrapper to the relevant PCL functions
... one or two code cleanups and comments

21 years ago0.pre8.114:
Christophe Rhodes [Mon, 28 Apr 2003 09:53:01 +0000 (09:53 +0000)]
0.pre8.114:
CHAR, SCHAR and SVREF act on vectors, not multidimensional
arrays
... so make the source-transform aware of that, for early error
reporting (observation from Gerd Moellmann)

21 years ago0.pre8.113:
Kevin Rosenberg [Sun, 27 Apr 2003 20:56:01 +0000 (20:56 +0000)]
0.pre8.113:
      * src/pcl/slots.lisp: change return value of slot-makunbound-using-class
      to OBJECT rather than NIL to conform to AMOP

21 years ago0.pre8.112:
Kevin Rosenberg [Sun, 27 Apr 2003 17:02:13 +0000 (17:02 +0000)]
0.pre8.112:
      - toplevel.lisp: implements toplevel of recursive repl.

21 years ago0.pre8.112:
Kevin Rosenberg [Sun, 27 Apr 2003 17:00:24 +0000 (17:00 +0000)]
0.pre8.112:
      - src/code/toplevel.lisp: Remove changes to REPL and rename
        REPL to REPL-FUN and add hook.
      - sb-aclrepl/tests.lisp: Add display tests.
      - sb-aclrepl/toplevel.lisp: New file. Toplevel REPL with support
        for catching signals
      - sb-aclrepl/README: state that sb-aclrepl must be loaded in
        ~/.sbclrc.

21 years ago0.pre8.111:
Alexey Dejneka [Sun, 27 Apr 2003 05:50:40 +0000 (05:50 +0000)]
0.pre8.111:
        * fast version of CONSTANT-FOLD-CALL flushes dead combination.

21 years ago0.pre8.110
Daniel Barlow [Sat, 26 Apr 2003 22:45:37 +0000 (22:45 +0000)]
0.pre8.110
Syntax error in sparc/linux C runtime

21 years ago0.pre8.109
Daniel Barlow [Sat, 26 Apr 2003 22:32:27 +0000 (22:32 +0000)]
0.pre8.109
#+sb-threads check for Linux 2.4, because it won't work in 2.2
In the process, rename early_kernel to linux_sparc_siginfo_bug,
just to make its purpose a little more obvious

Export WITH-TIMEOUT and TIMEOUT from SB-EXT.  TIMEOUT is now a
SERIOUS-CONDITION not an ERROR

21 years ago0.pre8.108:
William Harold Newman [Sat, 26 Apr 2003 14:19:52 +0000 (14:19 +0000)]
0.pre8.108:
Check *FEATURES* compatibility in fasl load.

21 years ago0.pre8.107:
Alexey Dejneka [Sat, 26 Apr 2003 03:04:52 +0000 (03:04 +0000)]
0.pre8.107:
        * Factored out code shared between MAP and MAP-INTO
          optimizers;
        * use one iteration variable for all arrays in the MAP-INTO
          optimizer.

21 years ago0.pre8.106:
William Harold Newman [Sat, 26 Apr 2003 02:43:01 +0000 (02:43 +0000)]
0.pre8.106:
doc tweaking
refactoring in preparation for checking *FEATURES* compatibility
in fasl load

21 years ago0.pre8.105
Daniel Barlow [Fri, 25 Apr 2003 23:43:07 +0000 (23:43 +0000)]
0.pre8.105
Add #include <errno.h>, for +sb-threads builds on some Linux
distros in which the existing include files don't seem to have
dragged it in already

Change the PSEUDO_ATOMIC_ATOMIC assert in alloc to print more
detail.  If you run into this bug, sbcl-devel wants to know.

21 years ago0.pre8.104:
Kevin Rosenberg [Fri, 25 Apr 2003 16:31:17 +0000 (16:31 +0000)]
0.pre8.104:
    * src/code/toplevel.lisp: Add special variables to convert SB-IMPL::REPL
      into a recursively invokable funcion
    * src/code/debug.lisp: Add hook for SB-DEBUG::DEBUG-LOOP
    * contrib/sb-aclrepl/tests.lisp: add tests for bignum inspection
    * contrib/sb-aclrepl/repl.lisp: convert to use new SB-IMPL::REPL function,
      add some debugger commands
    * contrib/sb-aclrepl/debug.lisp: use SB-DEBUG::*DEBUG-LOOP-FUN* hook.
      however, hook is not yet enabled by default  while debugger function
      continues development.

21 years ago0.pre8.103:
Alexey Dejneka [Fri, 25 Apr 2003 14:04:47 +0000 (14:04 +0000)]
0.pre8.103:
        * Added open coding of MAP-INTO for a vector destination
          (reported by Brian Downing on c.l.l)

21 years ago0.pre8.102:
Alexey Dejneka [Fri, 25 Apr 2003 04:06:31 +0000 (04:06 +0000)]
0.pre8.102:
        Compiler issues a full WARNING on calling of an undefined
        function with a name from the CL package.

21 years ago0.pre8.100:
Kevin Rosenberg [Fri, 25 Apr 2003 02:54:06 +0000 (02:54 +0000)]
0.pre8.100:
    * sb-aclrepl changes
       - Update README with examples and contact information
       - Strip out break-stack concept from repl.lisp
       - Fix bug in trimming *history* when *max-history* is reached
       - Add display of single-float, double-float, bignum hexidecimal contents
       - Fix reporting of object addresses (mask lowtag bits)
       - Add ":i slot <id>" command
       - Some non-active experimental code is in toplevel.lisp and debug.lisp,
         but this is #+ignore'd while in development.

21 years ago0.pre8.100:
Christophe Rhodes [Thu, 24 Apr 2003 15:22:49 +0000 (15:22 +0000)]
0.pre8.100:
As suggested by Fufie (Stig Sandoe) on #lisp IRC, emit a
STYLE-WARNING for slots with names that are STRING= but not EQ.
... test that we can nonetheless still make them, given the
insecurity I feel about my format string wizardry.

21 years ago0.pre8.99:
Christophe Rhodes [Thu, 24 Apr 2003 08:55:34 +0000 (08:55 +0000)]
0.pre8.99:
One more -0.0/0.0 fix:
... numeric type ranges should be compared with EQUALP, not
EQUAL, to make (FLOAT 0.0) be the same as (FLOAT -0.0)

21 years ago0.pre8.98:
Christophe Rhodes [Wed, 23 Apr 2003 17:04:52 +0000 (17:04 +0000)]
0.pre8.98:
Fix for MEMBER type (as per cmucl-imp 2003-04-23)
... (MEMBER 0.0) is not the same as (SINGLE-FLOAT 0.0 0.0);
... (MEMBER 0.0 -0.0) is the same as (SINGLE-FLOAT 0.0 0.0)
... (NOT (MEMBER 0.0)) needs to be
(OR (NOT SINGLE-FLOAT)
    (SINGLE-FLOAT * (0.0))
    (MEMBER -0.0)
    (SINGLE-FLOAT (0.0)));
... add some tests for this one.
In the process of this fix, make -0.0 and -0.0d0 dumpable by the
cross-compiler:
... more special cases in src/code/cross-float.lisp.
And also let an :SB-SHOW build proceed to the end
... don't try to print the slots of PCL objects.

21 years ago0.pre8.97:
Kevin Rosenberg [Wed, 23 Apr 2003 16:36:21 +0000 (16:36 +0000)]
0.pre8.97:
      * contrib/sb-rt/Makefile: fixed to install sb-rt as an ASDF system
      * contrib/sb-rt/rt.lisp: renamed from sb-rt.lisp
      * contrib/sb-aclrepl/sb-aclrepl.asd: Reworked to create a separate
        sb-aclrepl-tests ASDF system

21 years ago0.pre8.96:
Christophe Rhodes [Wed, 23 Apr 2003 09:38:04 +0000 (09:38 +0000)]
0.pre8.96:
Make ROOM work on cheneygc systems more than half of the time
... ensure that we map over allocated objects from the start
of the current dynamic (semi-)space, not from the start
of the whole dynamic space region.

21 years ago0.pre8.95:
Kevin Rosenberg [Wed, 23 Apr 2003 03:26:50 +0000 (03:26 +0000)]
0.pre8.95:
       - Rework sb-aclrepl.asd file to for sb-rt package
       - Rename aclrepl-tests.lisp to tests.lisp

21 years ago0.pre8.94:
Kevin Rosenberg [Tue, 22 Apr 2003 23:55:18 +0000 (23:55 +0000)]
0.pre8.94:
       - Add missing Makefile so that sb-rt installs

21 years ago0.pre8.93:
William Harold Newman [Tue, 22 Apr 2003 21:38:38 +0000 (21:38 +0000)]
0.pre8.93:
Antonio Martinez pointed out that an EQUAL test is needed for
generalized function name lookup.
(FORMAT NIL "~S ~S" '1W '1WW) => "|1W| 1WW" seems to be
correct after all.

21 years ago0.pre8.92:
Christophe Rhodes [Tue, 22 Apr 2003 15:23:09 +0000 (15:23 +0000)]
0.pre8.92:
Add SB-RT (a regression tester framework) as a contrib
... not with a great fanfare, though, as philosophically it's for
contrib use, not general public use;
... make sb-aclrepl and sb-bsd-sockets use sb-rt, not their own
local copies of rt.

21 years ago0.pre8.91:
Christophe Rhodes [Tue, 22 Apr 2003 15:10:57 +0000 (15:10 +0000)]
0.pre8.91:
Fix for ctor MAKE-INSTANCE optimization
... don't overzealously assert that a given location is unset;
users can supply duplicate initargs.

21 years ago0.pre8.90:
Christophe Rhodes [Tue, 22 Apr 2003 14:58:32 +0000 (14:58 +0000)]
0.pre8.90:
Make COPY-SYMBOL work on threaded builds
... FAST-SYMBOL-VALUE need not be fast, but must not signal an error
on unbound-symbol
... add a smoke test

21 years ago0.pre8.89:
Alexey Dejneka [Tue, 22 Apr 2003 04:42:18 +0000 (04:42 +0000)]
0.pre8.89:
        * CAST branch seems to take too long time to complete, so:
        ** fixed ENDP.* tests from Paul Dietz' test suit by disabling
          "constant folding" and flushing of IF;
        * fixed type declaration for ENDP as recommended by CSR;
        * ... and ENDP is FLUSHABLE.

21 years ago0.pre8.88:
Kevin Rosenberg [Tue, 22 Apr 2003 00:42:38 +0000 (00:42 +0000)]
0.pre8.88:
      - contrib/sb-aclrepl/rt.lisp: Add pfdietz's eq test to equalp-with-case
      - contrib/sb-aclrepl/aclrepl-tests.lisp: Take advantage of new
        equalp-with-case patch

21 years ago0.pre8.87
Daniel Barlow [Mon, 21 Apr 2003 23:57:34 +0000 (23:57 +0000)]
0.pre8.87
Bug 247 was fixed in 0.pre8.71: update tests and BUGS file to
match

21 years ago0.pre8.86:
Kevin Rosenberg [Mon, 21 Apr 2003 23:02:47 +0000 (23:02 +0000)]
0.pre8.86:
       - contrib/sb-aclrepl/repl.lisp: improve ld-cmd to allow loads
         such as ":ld ~/foo/bar".
       - contrib/sb-aclrepl/inspect.lisp: Handle cyclic lists.
       - contrib/sb-aclrepl/aclrepl-tests.lisp: add cyclic lists tests.
         Use macros to handle common test patterns greatly reducing
         code duplication.

21 years ago0.pre8.85
Daniel Barlow [Mon, 21 Apr 2003 21:06:24 +0000 (21:06 +0000)]
0.pre8.85
Merge the definitely-maybe-gc branch:

Remove all the lispland logic that checks bytes-consed
thresholds to see if it's worth GCing: C does these checks
anyway.  SUB-GC now always GCs unless *GC-INHIBIT* says not to

Delete the notify before/after calls and notify-stream.
(Planned future change: pass the same information to
*{BEFORE,AFTER}-GC-HOOKS* instead)

*SOFT-HEAP-LIMIT* and MAYBE-GC are gone (unused).
*ALREADY-MAYBE-GCING* is also dead, replaced with a proper
mutex that will also prevent simultaneous GCs from multiple
threads.  This entails moving gc.lisp and purify.lisp after
target-thread in the build-order

(Unithread WITH-MUTEX now honours its WAIT-P argument, so
actually works in this situation)

21 years ago0.pre8.84:
Alexey Dejneka [Mon, 21 Apr 2003 04:37:20 +0000 (04:37 +0000)]
0.pre8.84:
        Fixes in RESTART-CASE, inspired by Paul Dietz' test suit:
        * MACROEXPAND requires two arguments;
        * DWIM module in RESTART-CASE uses search-by-identity rather
          than search-by-name.

21 years ago0.pre8.83:
Alexey Dejneka [Sun, 20 Apr 2003 16:29:19 +0000 (16:29 +0000)]
0.pre8.83:
        NSET-EXCLUSIVE-OR does not return extra elements when its
        arguments contain duplicated elements. (reported by Paul
        Dietz)

21 years ago0.pre8.82:
Alexey Dejneka [Sun, 20 Apr 2003 10:53:42 +0000 (10:53 +0000)]
0.pre8.82:
        Fixed bugs caught by Paul Dietz' test suite:
        * CONVERT-MORE-CALL failed on ((LAMBDA (&KEY) 1)
          :ALLOW-OTHER-KEYS T) (fixed by Gerd Moellmann);
        * &WHOLE and &REST arguments in a macro lambda list may be
          patterns.

21 years ago0.pre8.81:
Kevin Rosenberg [Sun, 20 Apr 2003 08:48:46 +0000 (08:48 +0000)]
0.pre8.81:
     - sb-aclrepl: more bug fixes for the inspector, added "istep" tests
       to test interactive object traversal.

21 years ago0.pre8.80
Kevin Rosenberg [Sun, 20 Apr 2003 06:12:51 +0000 (06:12 +0000)]
0.pre8.80
     - fix inspector bug when display with positive skip value and no elements

21 years ago0.pre8.79
Kevin Rosenberg [Sun, 20 Apr 2003 05:15:10 +0000 (05:15 +0000)]
0.pre8.79
    - inspector bug fixes, refactoring, more tests

21 years ago0.pre8.78
Daniel Barlow [Sun, 20 Apr 2003 03:39:47 +0000 (03:39 +0000)]
0.pre8.78
More locking fixes exposed by using a real SMP system
... gencgc: gc_alloc_update_page_tables touches global data
    so needs wrapping in free_pages_lock
... gencgc_handle_wp_violation: we can get two CPUs in this
    routine at once, so it would be nice if the second one
      didn't barf if it found the first had been here already

Found an eliminated another THREAD_CONTROL_STACK_SIZE use

21 years ago0.pre8.77
Kevin Rosenberg [Sun, 20 Apr 2003 00:07:50 +0000 (00:07 +0000)]
0.pre8.77
     - More refactoring in inspect.lisp
     - Add ":reset" command for repl
     - Add large regression test for aclrepl, primarily for the inspector

21 years ago0.pre8.76
Daniel Barlow [Sat, 19 Apr 2003 23:19:16 +0000 (23:19 +0000)]
0.pre8.76
SMP fix fix: whatever new-fangled C-like language the compiler
on SuSE accepts is clearly a superset of the C wot I learnt at
school.  That or I mangled the patch while hand-applying it.

21 years ago0.pre8.75
Daniel Barlow [Sat, 19 Apr 2003 20:11:31 +0000 (20:11 +0000)]
0.pre8.75
SMP fix: mutex around all access to modify_ldt() function

(necessary but not, I regret, sufficient to make it work on SMP)

21 years ago0.pre8.74:
Christophe Rhodes [Sat, 19 Apr 2003 13:14:45 +0000 (13:14 +0000)]
0.pre8.74:
Couple the classes and type systems some more
... forward-referenced-classes are now valid types.

Note: this fix follows the cmucl fix perhaps slightly too closely.  It
creates CLASSOIDs for forward-referenced-classes slightly eagerly, where
previously no such CLASSOID was generated.  This may have some as-yet
unnoticed effect.