Alexey Dejneka [Fri, 30 Apr 2004 10:55:12 +0000 (10:55 +0000)]
0.8.10.2:
* Fix bug 313, reported by Antonio Menezes Leitao: in
PROPAGATE-FUN-CHANGE before applying source transform, check
that LEAF corresponds to a global function, not to some
random named identifier.
Alexey Dejneka [Fri, 30 Apr 2004 03:23:44 +0000 (03:23 +0000)]
0.8.10.1:
New bug.
William Harold Newman [Sun, 25 Apr 2004 18:11:48 +0000 (18:11 +0000)]
0.8.10:
release, tagged as sbcl_0_8_10
Alexey Dejneka [Sat, 24 Apr 2004 04:40:27 +0000 (04:40 +0000)]
0.8.9.58:
New bug.
Rudi Schlatte [Fri, 23 Apr 2004 14:36:33 +0000 (14:36 +0000)]
0.8.9.57:
Prettify docstrings:
* Detect and format itemized sections and tabulated descriptions
(see the documentation for save-lisp-and-die)
... docstrings.lisp is getting large and unordered; the next tuit will
likely be spent cleaning up and moving the docstring extractor
into contrib/
Kevin Rosenberg [Thu, 22 Apr 2004 18:09:37 +0000 (18:09 +0000)]
0.8.9.56:
* doc/catalogs/catalog-debian.xml: Update with new locations
for new xml-core package.
Alexey Dejneka [Wed, 21 Apr 2004 17:50:19 +0000 (17:50 +0000)]
0.8.9.55:
* Put credits to Lutz Euler into NEWS.
Christophe Rhodes [Tue, 20 Apr 2004 13:31:55 +0000 (13:31 +0000)]
0.8.9.54:
Log various bugs from Bruno Haible (and also tokeniser
threadsafety bug from Robert Marlow)
Christophe Rhodes [Mon, 19 Apr 2004 14:15:46 +0000 (14:15 +0000)]
0.8.9.53:
Just textual fixes.
Christophe Rhodes [Mon, 19 Apr 2004 13:46:27 +0000 (13:46 +0000)]
0.8.9.52:
Fixes fixes fixes
... restore build on linux/unithread;
... workaround apparent OpenMCL bug in the reader (#1#-related)
... fix for (funcall #'cddr ...)
Christophe Rhodes [Mon, 19 Apr 2004 11:22:55 +0000 (11:22 +0000)]
0.8.9.51:
Ignore appropriate stuff on ppc (Raymond Wiker sbcl-devel 2004-04-18)
... also add boilerplate to static-fn.lisp
Christophe Rhodes [Mon, 19 Apr 2004 11:09:43 +0000 (11:09 +0000)]
0.8.9.50:
Fix the first of Bruno Haible's test failures, more-or-less as
per Nikodemus Siivola sbcl-devel 2004-04-16
... also fix behaviour of OBSOLETE-INSTANCE-TRAP as hinted by NS
William Harold Newman [Mon, 19 Apr 2004 03:41:40 +0000 (03:41 +0000)]
0.8.9.49:
merged Scott Parish's patch for OpenBSD, with a few tweaks...
...reduced space sizes to work around ulimit problems (as a
quick hack, probably not the ideal long-term solution)
...fiddling to work around collisions with NetBSD patch; ah,
the joys of working on living, squirmy code
...I think s/LISP_FEATURE_LINUX/LISP_FEATURE_SB_THREAD/ on
create_thread() is correct; it was motivated by the
way that now new_thread_trampoline() is defined only
when LISP_FEATURE_SB_THREAD.
minor issue encountered while going through xc: clean.sh
should blow away src/runtime/genesis/.
Christophe Rhodes [Fri, 16 Apr 2004 14:17:39 +0000 (14:17 +0000)]
0.8.9.48:
Just documentation
... make the compiler messages in the Compiler chapter match
reality slightly more;
... Slightly Less Capitalization All Over The Place When It's
Not Really Necessary (I realise this may be
controversial :-)
... more index entries, etc;
... annotate various bits with FIXMEs
Alexey Dejneka [Thu, 15 Apr 2004 17:47:13 +0000 (17:47 +0000)]
0.8.9.47:
* Merged patch by Lutz Euler for the bug 245a.
Christophe Rhodes [Thu, 15 Apr 2004 13:30:07 +0000 (13:30 +0000)]
0.8.9.46:
deKLUDGE the solution for PACK-BEFORE-GC-HOOK
... well, not completely. PACK remains non-reentrant and
non-threadsafe, but at least now global data structures
don't grow without bounds;
... mostly whitespace changes, but clear the PACK structures
after every call, not once per GC cycle. Marginally
less efficient, I fear :-(
... while we're at it, fix analogously VOP-TN-REFS, and while
we're at it, document that a special is a bit of an
odd way to implement a (non-reentrant non-threadsafe)
closure.
... only one BEFORE-GC-HOOK left.
Christophe Rhodes [Thu, 15 Apr 2004 13:17:18 +0000 (13:17 +0000)]
0.8.9.45:
Slight threadcode cleanup
... add boilerplate;
... make bodies of WITH-MUTEX and WITH-RECURSIVE-LOCK accept
declarations;
... implement GET-MUTEX and RELEASE-MUTEX for unithread, but
don't actually use them yet because we're still in
"no performance penalty for unithread";
... make WITH-MUTEX available in the cross-compiler, so that
if necessary we can lock sections non-recursively.
Christophe Rhodes [Thu, 15 Apr 2004 10:57:35 +0000 (10:57 +0000)]
0.8.9.44:
Delete a bunch of UNIX-FOO symbols (and UNIX-FSYNC particularly) as
per patch Azimuth sbcl-devel 2004-04-01;
Also give build-start/build-finished times as suggested by
Perry Metzger and Brian Downing sbcl-devel 2004-04-09
Christophe Rhodes [Thu, 15 Apr 2004 09:19:17 +0000 (09:19 +0000)]
0.8.9.43:
Normalize QSHOW (Perry Metzger sbcl-devel 2004-04-04)
... #ifdef, not #if
Daniel Barlow [Wed, 14 Apr 2004 22:38:45 +0000 (22:38 +0000)]
0.8.9.42
MORE IGNORANCE - add missing IGNORE declaration in code/target-thread
Also delete the standalone asdf-install executable, due to it
being basically a bit rubbish anyway
Christophe Rhodes [Wed, 14 Apr 2004 20:07:46 +0000 (20:07 +0000)]
0.8.9.41:
More ANSI-fixes, this time for RATIONALIZE
... new implementation due to Bruno Haible;
... include explanation and helpful references;
... fixes RATIONALIZE.1 and RATIONALIZE.3
Christophe Rhodes [Wed, 14 Apr 2004 16:42:25 +0000 (16:42 +0000)]
0.8.9.40:
Yay! Finally, a patch inspired by PFD's ansi-tests
... make ECHO-STREAMs understand READ-SEQUENCE;
... add more tests than are in ansi-tests, because the interaction
with UNREAD-CHAR is potentially tricky.
Rudi Schlatte [Wed, 14 Apr 2004 08:55:42 +0000 (08:55 +0000)]
0.8.9.39:
Documentation changes:
* Document run-program
* Create docstrings from contrib modules (see MODULES in Makefile)
... include some docstrings in sb-rotate-byte and sb-md5
documentation to test it.
... unloadable contrib modules break docstring generation for now ...
* Frob docstrings slightly: convert SYMBOL-LOOKALIKES to
@code{symbol-lookalikes}, or @var{symbol-lookalikes} if they're part
of the arglist.
... written in a slightly roundabout (line-based) way that will be
useful for detecting common itemization / table formatting
idioms in sbcl's docstrings.
... Bug fix: don't downcase docstrings.
Christophe Rhodes [Tue, 13 Apr 2004 20:37:49 +0000 (20:37 +0000)]
0.8.9.38:
Ignore some SPARC variables
Christophe Rhodes [Tue, 13 Apr 2004 13:31:36 +0000 (13:31 +0000)]
0.8.9.37:
NetBSD fixes
... remove one __NetBSD__ which got in the way;
... include kernel revision detection.
... also include less expensive spinlock definitions
for x86/unithread builds.
Christophe Rhodes [Tue, 13 Apr 2004 10:30:37 +0000 (10:30 +0000)]
0.8.9.36:
Commit "ignore during cross-compilation" patch (CSR sbcl-devel
2004-04-05)
... bad treatment of IGNORE now gets a full WARNING during
cross-compilation;
... fix all the badness this reveals;
... implement SAME-ARG checking in LOGFOO type derivers;
... also add one more IGNORABLE in PCL (from Marcus Pearce);
... test for bad (signed-byte <N>) bug.
Christophe Rhodes [Sun, 11 Apr 2004 21:10:53 +0000 (21:10 +0000)]
0.8.9.35:
Slightly MORE DOCUMENTATION
... minimal, unpolished sb-md5 and sb-rotate-byte documentation
... whitespace fixup in sbcl.texinfo
... add a couple of index terms to sb-aclrepl.texinfo
(It Would Be Nice to be able to include contrib docstrings;
it ought to be possible, but possibly we need to zap the
asdf-install 'binary' and require (ha ha) that REQUIRE should
work from within the build tree [with SBCL_HOME set appropriately]
on _all_ contribs)
Christophe Rhodes [Sun, 11 Apr 2004 21:02:39 +0000 (21:02 +0000)]
0.8.9.34:
Commit unmaintainable documentationoid for early-objdef.lisp
... add hand-generated binary representation of widetags
next to automatically-generating definition;
... will fail to be maintained as cut'n'paste moves the
lines around...
... but that's OK, because I will break the knuckles of anyone
who edits it before we get 64bitness;
... no, not really. But maybe it will act as a bit of a
visual deterrent.
Rudi Schlatte [Fri, 9 Apr 2004 20:54:24 +0000 (20:54 +0000)]
0.8.9.33:
* infrastructure for contrib/ documentation in the main manual:
contrib;**;*.texinfo get automagically included in the chapter
"Contributed Modules"
... rename docstrings.sh to make-tempfiles.sh, since it makes more
than docstrings now
... test it all by converting sb-aclrepl/README to texinfo format.
Rudi Schlatte [Fri, 9 Apr 2004 16:04:48 +0000 (16:04 +0000)]
0.8.9.32:
* Documentation changes:
- Document trace and require
- Add variable index
- Remove explicit links from @node lines, since makeinfo figures
this out anyway and they would make automagic contrib/
documentation inclusion harder.
- add file TEXINFO-HINTS
* Documentation infractructure changes:
- document functions etc. with package prefix; index without
package prefix too (makeinfo seems to index the name with prefix
without being asked, so we just lay back and enjoy it)
- Move list-of-packages-to-extract-docstrings-from to Makefile
- Escape Texinfo special characters in documentation bodies as well.
Christophe Rhodes [Fri, 9 Apr 2004 14:39:49 +0000 (14:39 +0000)]
0.8.9.31:
Add Config.x86-netbsd to HEAD (oops)
Daniel Barlow [Fri, 9 Apr 2004 12:49:33 +0000 (12:49 +0000)]
0.8.9.30
Rename x86 allocate-dynamic-code-object to allocate-code-object
for consistency with other ports, as we no longer have any
other kind of code object anwyay.
Christophe Rhodes [Thu, 8 Apr 2004 13:26:01 +0000 (13:26 +0000)]
0.8.9.29:
Merge netbsd_branch
Christophe Rhodes [Thu, 8 Apr 2004 13:14:23 +0000 (13:14 +0000)]
0.8.9.6.netbsd.3:
Minor commentary and indentation fixes.
Christophe Rhodes [Thu, 8 Apr 2004 12:00:30 +0000 (12:00 +0000)]
0.8.9.6.netbsd.2:
Merge Perry Metzger "netbsd patches" sbcl-devel 2004-04-06
... don't merge SB_THREAD/spinlock cleanup; the right answer
is to define an almost-null get_spinlock() version
for non-threaded builds.
Christophe Rhodes [Thu, 8 Apr 2004 09:25:08 +0000 (09:25 +0000)]
0.8.9.28:
Restore buildability goodness on PPC and SPARC
... move n-fixnum-foo-bits and so on from sparc-specific code
to early-vm.lisp
... while I'm at it, make the n-lowtag-bits calculated rather
than #+ / #-
... restore the hideous hack in ppc type-vops UGH UGH UGH
(my brain is too slow to fix this properly)
Christophe Rhodes [Wed, 7 Apr 2004 14:22:35 +0000 (14:22 +0000)]
0.8.9.27:
Make special operators know about their user-visible arglists
... change motivated by all-new all-singing all-dancing
automagic documentation facility;
... add a hacky test for it in sb-introspect
Christophe Rhodes [Wed, 7 Apr 2004 11:30:53 +0000 (11:30 +0000)]
0.8.9.26:
More docstrings fixes
... COLLECT-DOCUMENTATION should always return RESULT, even when
the package itself is undocumented;
... make the unique name reflect the package being documented,
not the home package of the symbol, so reexports
work right;
... get special-operators more right;
... also include one or two of these things in the manual so we
get a feel for what's going on.
Christophe Rhodes [Wed, 7 Apr 2004 08:56:59 +0000 (08:56 +0000)]
0.8.9.25:
Too late! Merge fixes to docstrings mechanism
... Make it work (including being able to load contribs) on
freshly-dumped uninstalled cores;
... transform #\+ to "plus"
... generate documentation for all nominally public packages
(note that because of reexports, SB-ALIEN:* gets
documented as common-lisp-var-star. Probably
fixable)
Daniel Barlow [Tue, 6 Apr 2004 22:21:19 +0000 (22:21 +0000)]
0.8.9.24
Restore the unmap/remap-to-zero-pages behaviour that was in
versions prior to 0.8.9.20; it works a lot better on machines
without Far Too Much Memory
Rudi Schlatte [Tue, 6 Apr 2004 15:17:21 +0000 (15:17 +0000)]
0.8.9.23
- Add a documentation string extractor. Docstrings of exported
symbols of the packages listed in doc/manual/Makefile can be
included in the manual like so:
@include macro-sb-ext-define-source-context.texinfo
Rudi Schlatte [Tue, 6 Apr 2004 12:31:33 +0000 (12:31 +0000)]
0.8.9.22
Some documentation tweaks:
- Added doc/manual/Makefile (no docstring extraction yet)
- compiler.texinfo: Moved "compiler limitations" section backwards:
Better describe the cool features first, only then mention where
they don't apply
- efficiency.texinfo: remove `makeinfo' style warning
Christophe Rhodes [Tue, 6 Apr 2004 09:31:14 +0000 (09:31 +0000)]
0.8.9.21:
Fixup discussion of OPTIMIZE/DYNAMIC-EXTENT declarations as
suggested by APD on sbcl-devel
Daniel Barlow [Mon, 5 Apr 2004 23:39:14 +0000 (23:39 +0000)]
0.8.9.20
Merge Perry Metzger "much ado about zero" patch from
sbcl-devel 2004/04/04: replace various interesting memory
zeroing constructs with memset for probably speed improvement
and definite maintainability enhancement
Daniel Barlow [Mon, 5 Apr 2004 23:16:18 +0000 (23:16 +0000)]
0.8.9.18
Some cosmetic cleanups to make this a better place to start a
new x86-64 branch
... remove large chunks of long-float: it didn't work anyway
... parms.lisp happens fractionally earlier so that we can
defconstant nil-value slightly more cleanly
some raw 3s and 4s get made into functions of n-lowtag-bits
Christophe Rhodes [Mon, 5 Apr 2004 12:11:56 +0000 (12:11 +0000)]
0.8.9.18:
Beginnings of documentation for DYNAMIC-EXTENT.
... working with texinfo seems easier even if the tools aren't
perfect yet.
Also log a pair of related bugs in the type system
William Harold Newman [Mon, 5 Apr 2004 11:08:03 +0000 (11:08 +0000)]
0.8.9.17:
merged Zach Beane's patch for PROFILE output formatting
(from sbcl-devel 03 Apr 2004)
Christophe Rhodes [Sun, 4 Apr 2004 14:07:24 +0000 (14:07 +0000)]
0.8.9.16:
Make NOTINLINE calls to known functions have their return types
derived.
... depun the BASIC-COMBINATION-KIND slot; add FUN-INFO slot
to hold any applicable fun-info;
... add a couple of clauses to optimizing routines;
... test case for new behaviour, and also adjust
NOTINLINE IDENTITY declarations in other tests
Christophe Rhodes [Sat, 3 Apr 2004 18:36:25 +0000 (18:36 +0000)]
0.8.9.15:
Add (struct cons *) cast needed (observation from Perry Metzger)
William Harold Newman [Thu, 1 Apr 2004 02:05:05 +0000 (02:05 +0000)]
0.8.8.14:
tidying...
...*REVERSED-TYPE-INFO-INIT-FORMS* isn't needed at run time.
...HOLES stuff is long-dead code -- seems to be related to
something on old old CMU CL SPARC port -- so delete it.
(pointed out by nyef on #lisp)
...For that matter, FIND-HOLES is no longer used either,
so delete it too.
...belated *and*-*edit* in cut-and-paste-and-edit of FTYPE
declaration of !COLD-LOSE
Rudi Schlatte [Wed, 31 Mar 2004 20:02:12 +0000 (20:02 +0000)]
0.8.9.13
* Import initial version of manual in Texinfo markup
* Add Debugger chapter of cmucl manual
- Comment out list-locations, step, breakpoint commands, since they
do not work at the moment
- add *debug-print-variable-alist* documentation, adapt debug
optimization policy description
Christophe Rhodes [Tue, 30 Mar 2004 20:21:11 +0000 (20:21 +0000)]
0.8.9.12:
Add basic tests regarding DYNAMIC-EXTENT
... mostly ensure that nothing is broken;
... in future, we'll want to verify that things are in fact being
stack-allocated where applicable.
Christophe Rhodes [Tue, 30 Mar 2004 17:08:18 +0000 (17:08 +0000)]
0.8.9.11:
Fix for alpha branch of ash-left-mod64:
... EBROKENPARENS
Christophe Rhodes [Tue, 30 Mar 2004 16:58:25 +0000 (16:58 +0000)]
0.8.9.10:
DYNAMIC-EXTENT &REST lists.
... much as per CSR sbcl-devel 2004-03-29;
... alter listify-rest-args VOPs on non-x86 to meet the new use
(don't do anything yet with the DX parameter)
... note concerns over stack manipulation in x86 DX allocation
This version compiles and passes tests on x86 and alpha (modulo
one unrelated bugfix, coming soon)
Christophe Rhodes [Tue, 30 Mar 2004 12:24:46 +0000 (12:24 +0000)]
0.8.9.9:
"more warnings fixes" (Perry E. Metzger sbcl-devel 2004-03-29)
... except with s/GENCGC/X86/ for fun_header and return_pc_header
scav functions
Christophe Rhodes [Tue, 30 Mar 2004 11:19:52 +0000 (11:19 +0000)]
0.8.9.8:
Remove some warnings (Perry Metzinger sbcl-devel 2004-03-30
"some anti-warning patches")
... also improve the comment above extern char **environ
Add required exit clause to load.impure.lisp
Christophe Rhodes [Tue, 30 Mar 2004 09:11:56 +0000 (09:11 +0000)]
0.8.9.7:
INSTALL fixes
... mention contrib/ as files you might want to keep around;
... patch (Nikodemus Siivola sbcl-devel 2004-03-27) for
more explicit description of typical form of SBCL_HOME
Christophe Rhodes [Mon, 29 Mar 2004 10:36:30 +0000 (10:36 +0000)]
0.8.9.6.netbsd.1:
Merge patch (Perry Metzger sbcl-devel 2004-03-27) for partial
NetBSD support (some work also due to Valtteri Vuorikoski)
Alexey Dejneka [Sat, 27 Mar 2004 07:58:16 +0000 (07:58 +0000)]
0.8.9.6:
* Fix bug reported by Sean Ross: flush fill pointer from a
simple array loaded from a FASL;
* update FOP tracing during loading.
Christophe Rhodes [Thu, 25 Mar 2004 21:51:53 +0000 (21:51 +0000)]
0.8.9.5:
Pointless microoptimizations 'R' us:
... in (defun foo (x)
(declare (optimize speed) (type (signed-byte 32) x))
(logand x #xffffffff))
on the x86, there is no need to perform the hardware AND, as
we are not interested in the flags. Make it so. (The system
already automatically handles the representation such that X
and the result can live in the same register -- to see this,
wrap the LOGAND above as (ASH (LOGAND ...) -3) and disassemble.
Alexey Dejneka [Thu, 25 Mar 2004 18:51:33 +0000 (18:51 +0000)]
0.8.9.5:
* Remove bug entry 304.
Alexey Dejneka [Thu, 25 Mar 2004 18:22:50 +0000 (18:22 +0000)]
0.8.9.4:
* Fix bug 304:
** combine ASH with ASH-MODx;
** declare OFFSET arguments of EXTRACT-ALIEN-VALUE and
DEPOSIT-ALIEN-VALUE to be unbounded UNSIGNED-BYTE;
** COUNT-LOW-ORDER-ZEROS looks through CASTs;
** provide modular-version => prototype translation.
Christophe Rhodes [Thu, 25 Mar 2004 08:33:11 +0000 (08:33 +0000)]
0.8.9.3:
Take advantage of the new tagging scheme on the SPARC
... yet more cut'n'paste backend programming
William Harold Newman [Wed, 24 Mar 2004 16:45:26 +0000 (16:45 +0000)]
0.8.9.2:
trivial tidying: A test file extension of .tmp lets it be
autoremoved by clean.sh and autoignored by my ~/.cvsignore.
William Harold Newman [Wed, 24 Mar 2004 16:26:26 +0000 (16:26 +0000)]
0.8.9.1:
hacked DocBook stuff to work with my old Debian setup
William Harold Newman [Wed, 24 Mar 2004 13:13:35 +0000 (13:13 +0000)]
0.8.9:
release, tagged as sbcl_0_8_9
Alexey Dejneka [Tue, 23 Mar 2004 14:07:41 +0000 (14:07 +0000)]
0.8.8.35:
* Two new bugs revealed by the post of Dave Roberts.
Alexey Dejneka [Mon, 22 Mar 2004 13:50:43 +0000 (13:50 +0000)]
0.8.8.34:
* Describe MISC.293.
Rudi Schlatte [Sun, 21 Mar 2004 13:18:43 +0000 (13:18 +0000)]
0.8.8.33:
* Implement read-sequence, write-sequence for sequences of
type (unsigned-byte 8), (signed-byte 8), add test case
* Prettier print-object method for socket simple-streams
William Harold Newman [Fri, 19 Mar 2004 19:51:26 +0000 (19:51 +0000)]
0.8.8.32:
Make distclean.sh delete CVS/ subdirectories.
Christophe Rhodes [Thu, 18 Mar 2004 10:09:36 +0000 (10:09 +0000)]
0.8.8.31:
Slightly less arbitrary default for *MAX-EMF-PRECOMPUTE-METHODS*.
... it appears that there is no need for an upper limit, so
the only natural value is no limit at all;
... retain the tunability of the parameter, even though there
probably ought to be a better MOPpish way of expressing
it;
... awaiting with interest the effect this has on benchmarks; it
has no observable effect on e.g. McCLIM window mappping
times. (what /does/ have an effect there is compiling
and loading a file with
(SB-PCL::PRECOMPUTE-RANDOM-CODE-SEGMENTS)
in it).
Christophe Rhodes [Wed, 17 Mar 2004 20:24:16 +0000 (20:24 +0000)]
0.8.8.30:
Remove unnecessary bounds checks from REPLACE, VECTOR-POP and
VECTOR-PUSH-EXTEND (Juho Snellman sbcl-devel 2004-03-17)
Christophe Rhodes [Tue, 16 Mar 2004 18:08:22 +0000 (18:08 +0000)]
0.8.8.29:
Fix for *PRINT-READABLY* printing of symbols
... *PRINT-READABLY* targets the /standard/ readtable, not the
current readtable.
Christophe Rhodes [Tue, 16 Mar 2004 18:00:04 +0000 (18:00 +0000)]
0.8.8.28:
More floating point reader fixes for non-10 *READ-BASE*
... confusion arises if the exponent marker is also a digit.
Fix the manifest confusions;
... also a long-standing floating point reader-bugfix: reading
"ae+9" with *READ-BASE* = 11 used to give a BUG;
... fix the tests, too.
Christophe Rhodes [Tue, 16 Mar 2004 12:19:04 +0000 (12:19 +0000)]
0.8.8.27:
Fix for float reading with *READ-BASE* < 10 (part of the
failures in PRINT.SINGLE-FLOAT.RANDOM et al. from PFD)
... appropriate mysteriously-missing 9 entry for
DECIMAL-DIGIT
... some extra states in the tokenizer FSM
... tests!
Christophe Rhodes [Mon, 15 Mar 2004 15:55:06 +0000 (15:55 +0000)]
0.8.8.26:
Merge patches from NJF (sbcl-devel 2003-03-11)
... more #!-SB-DOC
... fewer :TEST-NOT
... less OAOOM
Explain what's happening around MAYBE-FP-WAIT (thanks to
Nikodemus Siivola sbcl-devel)
Patch threads.impure.lisp to allow writing the C file (Nikodemus
Siivola sbcl-devel 2004-02-27)
Fix for foreign double access on Darwin (CSR sbcl-devel 2004-02-23)
Christophe Rhodes [Mon, 15 Mar 2004 15:24:52 +0000 (15:24 +0000)]
0.8.8.25:
More SB-POSIX test fixes
... allow building as root again, since it seems that's a
desired feature
... test for not having EUID of 0 in EACCES posix tests instead
(this will probably annoy antifuchs, but I forgot to commit
version.lisp-expr for the last fix. Since this is logically a
simple continuation, I haven't bothered to increment it again)
Christophe Rhodes [Mon, 15 Mar 2004 13:27:50 +0000 (13:27 +0000)]
0.8.8.25:
Don't allow the system to be built as root
... id(1) check early in the build.
Fix the SB-POSIX test suite
... rewrite mkdir.error.3 not to assume that / is unwriteable;
... continue to assume that the user is not (fake)root, but
nonetheless try to clean up a bit more if we are.
Christophe Rhodes [Mon, 15 Mar 2004 00:00:59 +0000 (00:00 +0000)]
0.8.8.24:
Fix the build on MIPS
... DEF!CONSTANT for the registers so that the offsets are
defined in a timely fashion
... also s/DO-PENDING-INTERRUPT/RECEIVE-PENDING-INTERRUPT/ for
MIPS and HPPA
Christophe Rhodes [Wed, 10 Mar 2004 16:10:17 +0000 (16:10 +0000)]
0.8.8.23:
Better %SXHASH-SUBSTRING (patch Juho Snellman sbcl-devel 2004-03-09)
... frob comments a little
... make the same FLET workaround in %SXHASH-SIMPLE-SUBSTRING
... probably fasl-incompatible with 0.8.8.22, but I've already
changed the fasl version number once this cycle. Let's
see if anyone complains :)
... 20% faster at compiling mk-defsystem on DB's iMac
(MORE SPEED!)
Christophe Rhodes [Tue, 9 Mar 2004 14:45:37 +0000 (14:45 +0000)]
0.8.8.22:
Less pprint suboptimiality
... s/simple-string/(simple-array character (*))/ and add
explicit coercions;
... this should remove most of the performance degradation in the
pretty printer introduced around 0.8.1.x when (vector nil)
was recognized as a string type.
Christophe Rhodes [Tue, 9 Mar 2004 12:08:39 +0000 (12:08 +0000)]
0.8.8.21:
Make unknown variables signal a compile-time warning.
... slightly hacky (if (eq kind :variable) ...) in the
warning logic itself;
... rearrange bits of the build to eliminate technically
undefined forward references;
... fix for multiple *CURRENT-CATCH-BLOCK* and
*CURRENT-UNWIND-PROTECT-BLOCK*: we now have one and
only one of each variable, rather than three
Christophe Rhodes [Mon, 8 Mar 2004 22:14:31 +0000 (22:14 +0000)]
0.8.8.20:
Hey look! An ANSI bug that's more annoying than (vector nil)!
... printing #\Space is defined to yield #\
... did you see the whitespace at the end of that line?
Christophe Rhodes [Mon, 8 Mar 2004 21:59:47 +0000 (21:59 +0000)]
0.8.8.19:
Fix for *PRINT-CASE* :CAPITALIZE and word determination (see
CLHS *PRINT-CASE* and STRING-CAPITALIZE)
... also note unpleasant discovery about lack of threadsafety
in the printer
Alexey Dejneka [Sun, 7 Mar 2004 07:50:51 +0000 (07:50 +0000)]
0.8.8.18:
* VALID-FUN-USE:
** when checking result type, intersect NODE-DERIVED-TYPE with
the trusted type assertion, if it exists;
** recognize NIL for RESULT-TEST as "always succeeds".
Christophe Rhodes [Sat, 6 Mar 2004 22:09:43 +0000 (22:09 +0000)]
0.8.8.17:
Make CLOSE works on STRING-STREAMs
Christophe Rhodes [Sat, 6 Mar 2004 19:54:51 +0000 (19:54 +0000)]
0.8.8.16:
Allow CLOSE to work on (MAKE-CONCATENATED-STREAM)
William Harold Newman [Sat, 6 Mar 2004 03:02:19 +0000 (03:02 +0000)]
0.8.8.15:
hacked TRACE :PRINT code so that it does TERPRI as well as
FRESH-LINE (which is perhaps less elegant than doing
only one or the other, but seems to be more robust
than the previous FRESH-LINE-only solution)
fixed overlooked merge conflict in NEWS (pointed out by
Neil Schemenauer)
William Harold Newman [Fri, 5 Mar 2004 13:02:20 +0000 (13:02 +0000)]
0.8.8.14:
introduced SB-DEBUG:*DEBUG-PRINT-VARIABLE-ALIST* mechanism
fixed trivial bug in CONCATENATED-N-BIN
Christophe Rhodes [Thu, 4 Mar 2004 11:12:45 +0000 (11:12 +0000)]
0.8.8.13:
CONCATENATED-STREAM-STREAMS fix
... yes, do discard streams. The old behaviour said "keep streams
around for closing", but that's actually the user's job.
Christophe Rhodes [Tue, 2 Mar 2004 16:23:21 +0000 (16:23 +0000)]
0.8.8.12:
PFDFixes
... ECHO-STREAM and end-of-file
... *PRINT-BASE*, *PRINT-RADIX* and ratios
Christophe Rhodes [Tue, 2 Mar 2004 09:37:48 +0000 (09:37 +0000)]
0.8.8.11:
Exploit the new widetag structure on PPC.
Christophe Rhodes [Mon, 1 Mar 2004 23:22:25 +0000 (23:22 +0000)]
0.8.8.10:
MORE MICRO-OPTIMIZATION
... renumber all widetags. (LESS BINARY-COMPATIBILITY)
... implement some smarter type tag checking on the x86.
The smarter type checking comes in several flavours.
* If we have two adjacent lowtags, in three cases out of four
we can do "and, cmp, branch", rather than
"cmp, branch, cmp, branch";
* If we have two lowtags that are unadjacent but differ by
just one bit, we can likewise do "and, cmp, branch" rather
than "cmp, branch, cmp, branch";
* If we have a contiguous range of lowtags that includes the
most positive lowtag, we do not need to test the upper bound;
* If we have four lowtags with only two bits different, we can
do "and, cmp, branch" rather than *four* "cmp, branch" pairs.
This change will conflict with ongoing 64-bit ports. The
conflict is resolvable (there are enough widetags in the "array"
range (which in practice is 130 - 254; currently there are five
in this range unused (or three with long-float arrays)
Christophe Rhodes [Mon, 1 Mar 2004 21:32:42 +0000 (21:32 +0000)]
0.8.8.9:
Reinstate some compiler transforms for STRINGoid types.
... since STRING is now a union type, some compiler transforms
would give up (unnecessarily) with arguments of such
a type, even when the conditions were otherwise
favourable;
... extend the logic in those transforms to deal with arrays
with the same 'shape' but different specialized
array element type
Christophe Rhodes [Mon, 1 Mar 2004 20:30:23 +0000 (20:30 +0000)]
0.8.8.8:
Fix for CLEAR-INPUT
... its optional argument can be NIL or T as well as a stream;
... since we're in the area,
s/STREAMLIKE/STREAM-DESIGNATOR/
s/STRINGABLE/STRING-DESIGNATOR/
s/SYMBOLS/SYMBOLS-DESIGNATOR/
(and move SYMBOLS-DESIGNATOR and PACKAGE-DESIGNATOR to
deftypes-for-target.lisp where they belong)
Note *ZAP-ARRAY-DATA-TEMP* as a source of non-threadsafety
in ADJUST-ARRAY
Christophe Rhodes [Mon, 1 Mar 2004 20:23:30 +0000 (20:23 +0000)]
0.8.8.7:
MAKE-STRING-INPUT-STREAM fixes
... :END can be NIL;
... coerce non-(simple-array character (*))s to that type.
Also fix an erroneous test and message in ADJUST-ARRAY.
Christophe Rhodes [Mon, 1 Mar 2004 16:21:14 +0000 (16:21 +0000)]
0.8.8.6:
Some fixes for ADJUST-ARRAY
... make sure we copy the element in a zero-rank array;
... don't adjust simple arrays, even if it doesn't break
anything (because there's probably lying to compilers
going on).
Christophe Rhodes [Mon, 1 Mar 2004 15:08:21 +0000 (15:08 +0000)]
0.8.8.5:
Implemented *-MOD32 on the PPC
Christophe Rhodes [Fri, 27 Feb 2004 09:41:37 +0000 (09:41 +0000)]
0.8.8.4:
Restore cross-compilability, by making ldso-stubs.S in host-1
Christophe Rhodes [Thu, 26 Feb 2004 12:15:01 +0000 (12:15 +0000)]
0.8.8.3:
MORE BACKSLASHES
... restore buildability on x86/linux, and probably other places,
by escaping backslashes as necessary