Kevin Rosenberg [Wed, 3 Dec 2003 16:18:57 +0000 (16:18 +0000)]
0.8.6.25:
* doc/DOCDOC: Provide more complete documentation on documentation
generation prerequisits.
Kevin Rosenberg [Tue, 2 Dec 2003 03:12:05 +0000 (03:12 +0000)]
Add fink operating system detection
Kevin Rosenberg [Tue, 2 Dec 2003 01:47:55 +0000 (01:47 +0000)]
0.8.6.24:
* doc/*: Revert to Docbook 4.1 DOCTYPE for compatibility with
older systems such as Debian Woody. Add catalog for Darwin's
fink and for partially for Redhat.
Need to add OS detection for fink in Makefile and stylesheet
location for RedHat in catalog-redhat.xml
Andreas Fuchs [Mon, 1 Dec 2003 19:58:54 +0000 (19:58 +0000)]
0.8.6.23:
add a missing \ to the SBCL banner, to de-confuse GCCs that don't
allow multi-line strings.
William Harold Newman [Mon, 1 Dec 2003 15:19:52 +0000 (15:19 +0000)]
0.8.6.22:
LESS TEXT: quieter startup banner and debug prompt mostly
as suggested by Nikodemus Siivola and Dan Barlow
on sbcl-devel
really updating FASL-FILE-VERSION (as attempted in my last
checkin) this time for sure
William Harold Newman [Mon, 1 Dec 2003 14:53:51 +0000 (14:53 +0000)]
0.8.6.22:
LESS TEXT: quieter startup banner and debug prompt mostly
as suggested by Nikodemus Siivola and Dan Barlow
on sbcl-devel
really updating FASL-FILE-VERSION (as attempted in my last
checkin) this time for sure
William Harold Newman [Sun, 30 Nov 2003 21:21:33 +0000 (21:21 +0000)]
0.8.6.21:
bumped fasl file version so I don't forget at release time
Christophe Rhodes [Sun, 30 Nov 2003 17:11:05 +0000 (17:11 +0000)]
0.8.6.20:
Merge simple-streams patch (David Licteblau sbcl-devel 2003-11-09)
... including frobs to core code to unspecial-case Gray streams
Christophe Rhodes [Sun, 30 Nov 2003 16:08:48 +0000 (16:08 +0000)]
0.8.6.19:
Fix for CONS.SUBTYPEP.xx from PFD's test suite
... more work in the CONS union method.
ALERT ALERT ALERT
this change reportedly makes PFD's random tester crawl, slowing it
by a factor of 15 on input terms (type specifiers) of size 50.
Options include
* ignoring this;
* having a cutoff in time or complexity beyond which we return
NIL, NIL;
* optimizing (NOT <cons>) some more;
* ...
but at least it's slightly more correct now.
Christophe Rhodes [Sun, 30 Nov 2003 15:49:04 +0000 (15:49 +0000)]
0.8.6.18:
ASDF-INSTALL fix (parse *proxy* properly, as per Sean Ross sbcl-devel
2003-10-23)
... start parse-integer at the right place
SHOW patchery from Nikodemus Siivola (sbcl-devel 2003-11-28)
Christophe Rhodes [Sun, 30 Nov 2003 15:19:34 +0000 (15:19 +0000)]
0.8.6.17:
Fix for bogus type warning (Rudi sbcl-devel 2003-09-21)
... I don't really know what I'm doing here. Presumably
SLOTS could never be NIL at this point? Bah.
Daniel Barlow [Sat, 29 Nov 2003 23:54:20 +0000 (23:54 +0000)]
0.8.6.16
Tidy up a bit
GET-FOREGROUND now prints 'Resuming thread n' messages on
thread swith, to make it slightly harder to get lost
THREAD-REPL-PROMPT-FUN goes away: it was a kludge anyway and
most of what it did is done in the standard prompt function
gc_thread_pid no more. Don't need sys/ptrace.h. Delete call
to mach_init() which doesn't exist anyway. Ignore
uninteresting signals in parent thread, and lose
parent_sighandler
Christophe Rhodes [Sat, 29 Nov 2003 22:20:37 +0000 (22:20 +0000)]
0.8.6.15:
Yay PFD real tests!
... REHASH-THRESHOLD may be specified to be of type (REAL 0 1)
[ aside: (REAL (0) 1) would be much preferred for
aesthetic reasons ] but we'd better not divide by it
unconditionally, because otherwise confusion ensues
... define a constant below which we shall not go
Christophe Rhodes [Sat, 29 Nov 2003 20:41:47 +0000 (20:41 +0000)]
0.8.6.14:
Ouch. That was hard work.
... contrib fix primarily for SunOS but also I think for Darwin;
at least, if it was working it was only by accident.
... uname -m never returns either "solaris" or "darwin", so that
code was wrong to start with;
... change the protocol so that we do EXTRA_CFLAGS rather than
a complete override; now we don't need the solaris or
darwin clauses in asdf-module.mk that weren't being used
anyway;
... delete more stuff in make-target-contrib.sh so that we can
actually see the effect of these changes rather than
blindly reusing an old alien.so (which explains why I
hadn't noticed this before, *sigh*; I ran a completely
clean build on Solaris for the first time in months
yesterday)
... now sb-bsd-sockets and dependents work again. I hope I haven't
broken x86-64 in the process.
Christophe Rhodes [Sat, 29 Nov 2003 11:25:31 +0000 (11:25 +0000)]
0.8.6.13:
Merge fix for bidirectional (that is IO, not anything Unicodely)
streams (Adam Warner/Gerd Moellmann sbcl-devel/cmucl-imp 2003-11-xx)
... added a test for it; be careful with that standard io syntax :)
Kevin Rosenberg [Sat, 29 Nov 2003 05:38:43 +0000 (05:38 +0000)]
0.8.6.12:
* contrib/sb-aclrepl/repl.lisp: Support recent sb-thread changes.
repl.lisp now compiles, but I have not tested sb-aclrepl with
multiple listeners.
Daniel Barlow [Sat, 29 Nov 2003 00:35:40 +0000 (00:35 +0000)]
0.8.6.11
Some clean up with unix signals possible now that we denote
them with numbers instead of keywords
Juggled the order in target-thread.lisp to make it compile
without warning
Threads now signal SIG_THREAD_EXIT in the parent, not
SIGALRM. CLONE_PARENT is no longer used, so the creating
Lisp thread gets this signal instead of the original C process
thread-exit_handler is the SIG_THREAD_EXIT handler. It calls
the new static function HANDLE-THREAD-EXIT to manipulate
*SESSION*
SB!THREAD::*FOREGROUND-THREAD-STACK* ius dead, remove from
static variables list
Andreas Fuchs [Fri, 28 Nov 2003 23:07:26 +0000 (23:07 +0000)]
0.8.6.10:
Fix up the fix for FreeBSD and sb-posix:
* revert the checks for sb-posix::eisdir to the saner
sb-posix::eexists and sb-posix::ebusy.
* instead, don't fail when any ".ERROR" tests fail, but warn.
Daniel Barlow [Fri, 28 Nov 2003 04:41:05 +0000 (04:41 +0000)]
0.8.6.10
The diff is kind of noisy, but it's really pretty simple
The goal here is to deal with multiple 'session's in a
slightly more sendible way, and for QUIT to kill off
background threads if any are running
We create a SESSION struct with various locks and queues,
so replacing about 4 special variables with one *SESSION*
structure object.
MAKE-THREAD adds the new thread id to the current session
WITH-NEW-SESSION removes the current thread from the session,
then creates a new session containing only the current thread
and rebinds *SESSION* to it. It's used by MAKE-LISTENER-THREAD
and is also exported albeit in a "we reserve the right to
change this but are unlikely to" fashion
TERMINATE-SESSION terminates all threads in the session other
than the caller. %END-OF-THE-WORLD-HANDLER calls it
Also exported *INVOKE-DEBUGGER-HOOK* from SB-EXT for SLIME and
similar alternate debugging interfaces to use.
Christophe Rhodes [Thu, 27 Nov 2003 22:44:36 +0000 (22:44 +0000)]
0.8.6.8:
Fix for GET-SETF-EXPANSION problem (Nikodemus Siivola sbcl-devel
2003-11-27)
Andreas Fuchs [Thu, 27 Nov 2003 21:50:02 +0000 (21:50 +0000)]
0.8.6.7:
Fix sb-posix tests for BSDs:
* Replace #xffffffff for directory modes with a more portable
constant (BSD's mode_t is 16 bits, so this makes the type
checker happy)
* Conditionalize the (buggy) return value for mkdir/rmdir on
*BSD.
Kevin Rosenberg [Thu, 27 Nov 2003 08:46:22 +0000 (08:46 +0000)]
0.8.6.6:
* Add two missing functions from 0.8.6.5
Daniel Barlow [Thu, 27 Nov 2003 06:21:04 +0000 (06:21 +0000)]
0.8.6.5
"Well, the hours are pretty good"
Merged the resistance-is-futex branch: see commit messages on
branch for scary details
"... but now I come to think about it, most of the actual minutes
are pretty lousy"
Kevin Rosenberg [Wed, 26 Nov 2003 20:22:00 +0000 (20:22 +0000)]
0.8.6.4:
* install.sh: install doc/sbcl-asdf-install.1
* doc/DOCDOC: add note about catalog file
* doc/sbcl-html.dsl: remove obsolete file
Daniel Barlow [Wed, 26 Nov 2003 16:31:57 +0000 (16:31 +0000)]
0.8.6.3
Fix finalization bug in sb-bsd-sockets contrib that was causing
strange SLIME errors
Add SUPPORT file with advice on mailing list etiquette and
suggestions for where to go if you need more help
(note for consultants: add yourself to this list)
William Harold Newman [Wed, 26 Nov 2003 15:16:06 +0000 (15:16 +0000)]
0.8.6.2:
trivial tidying in new XSL version of doc/
Kevin Rosenberg [Wed, 26 Nov 2003 03:35:22 +0000 (03:35 +0000)]
Remove PDF file on clean, have clean.sh invoke "make clean"
Kevin Rosenberg [Wed, 26 Nov 2003 02:12:56 +0000 (02:12 +0000)]
Only build HTML [not PDF] on default build
Kevin Rosenberg [Wed, 26 Nov 2003 02:11:09 +0000 (02:11 +0000)]
0.8.6.1:
* clean.sh: Allow recursive includes in src/runtime/Config
* doc/*: Convert from SGML format and processing to XML
Currently, catalogs are in place for Debian and RedHat which
allow automated use of local files instead of the tools
downloading DTDs and XSLs from their canonical URLs.
I'm glad to help setup catalogs for other operating systems.
William Harold Newman [Tue, 25 Nov 2003 19:34:45 +0000 (19:34 +0000)]
0.8.6:
release, tagged as sbcl_0_8_6
William Harold Newman [Sat, 22 Nov 2003 02:40:12 +0000 (02:40 +0000)]
0.8.5.50:
nikodemus siivola patch from sbcl-devel 2003-11-18: tweaked
lookup order for alien variables to cope with loading
foreign libraries
Kevin Rosenberg [Wed, 19 Nov 2003 22:46:22 +0000 (22:46 +0000)]
0.8.5.49:
* Revert previous tools-for-build/modify-ldt-struct-name commit
(which never actually included the necessary C file) for the
simpler test of LINUX_VERSION_CODE suggested by Dan Barlow.
Kevin Rosenberg [Wed, 19 Nov 2003 13:48:07 +0000 (13:48 +0000)]
0.8.4.48:
* Add tools-for-build program to determine if asm/ldt.h has the
new name for the modify_ldt structure.
* Fixes to properly compile contribs on x86_64 systems
Alexey Dejneka [Wed, 19 Nov 2003 08:41:33 +0000 (08:41 +0000)]
0.8.5.47:
* Describe PFD bug MISC.186.
Alexey Dejneka [Mon, 17 Nov 2003 15:41:26 +0000 (15:41 +0000)]
0.8.5.45:
* Describe modular arithmetic optimization.
Alexey Dejneka [Mon, 17 Nov 2003 11:47:09 +0000 (11:47 +0000)]
0.8.5.45:
* Fix PFD bugs MISC.177, 182: in copy propagation a local
lambda argument has a hidden write.
Christophe Rhodes [Sun, 16 Nov 2003 23:52:04 +0000 (23:52 +0000)]
0.8.5.44:
SB-ACLREPL threading
... delete erroneous previous attempt (binding specials outside
DEFUN forms)
... define a repl fun maker, which (a) has state in closure
variables and (b) saves its state to said variables on
non-local exit.
... rearrange sb-aclrepl package manipulation a little
... RESET-CMD now searches for the sb-impl::toplevel restart by name
All this requires some core support
... *REPL-FUN* is ok for stateless repls, but not for stateful ones;
change protocol to make a repl-fun by calling
*REPL-FUN-GENERATOR* instead (I feel a bit bad about this,
but since this is all in the SB-IMPL:: package I don't feel
/too/ bad about breaking the interface. I doubt anything but
SB-ACLREPL used it anyway)
... rebind cl specials per TOPLEVEL so that IN-PACKAGEs don't
collide (don't use WITH-STANDARD-IO-SYNTAX, duh)
Alexey Dejneka [Sun, 16 Nov 2003 17:40:18 +0000 (17:40 +0000)]
0.8.5.43:
* Fix miscellaneous PFD bugs
** REPLACE-TOPLEVEL-XEPS: update LEAF-REFS;
** SUBSTITUTE-LVAR-USES: if new LVAR is NIL, flush dests.
Alexey Dejneka [Sun, 16 Nov 2003 10:19:13 +0000 (10:19 +0000)]
0.8.5.42:
* IR1-OPTIMIZE: whenever possible, delete all marked blocks.
Kevin Rosenberg [Sat, 15 Nov 2003 22:13:27 +0000 (22:13 +0000)]
0.8.5.41:
* contrib/sb-aclrepl/: Fix error with cl:inspect
double evaluating arguments. Bug reported by Christophe Rhodes
on sbcl-devel 2003-11-15.
Alexey Dejneka [Sat, 15 Nov 2003 18:34:34 +0000 (18:34 +0000)]
0.8.5.40:
* Fix PFD bug MISC.172: restart IR1-OPTIMIZE-RETURN after
assignment-convertion;
* fix PFD bug MISC.173: in FIND-DFO-AUX skip blocks to be
deleted.
Christophe Rhodes [Sat, 15 Nov 2003 16:12:54 +0000 (16:12 +0000)]
0.8.5.39:
Fix modular ASH bug on PPC
Christophe Rhodes [Sat, 15 Nov 2003 15:21:00 +0000 (15:21 +0000)]
0.8.5.38:
Fix PFDtest HANDLER-CASE.28
... wrap in LOCALLY
Alexey Dejneka [Fri, 14 Nov 2003 12:44:42 +0000 (12:44 +0000)]
0.8.5.37:
* Fix PFD bug MISC.165: unify treating of TAIL-CALLS in
LET-convertion.
Alexey Dejneka [Fri, 14 Nov 2003 08:45:50 +0000 (08:45 +0000)]
0.8.5.36:
* Fix PFD bugs MISC.166, 167, 168: do not perform tail- and
let-convertion, if block of lambda's BIND node is to be
deleted.
Alexey Dejneka [Fri, 14 Nov 2003 04:49:06 +0000 (04:49 +0000)]
0.8.5.35:
* Fix PFD bugs MISC.152, 153: be more accurate with deleted
code in MAYBE-INFER-ITERATION-VAR-TYPE.
Daniel Barlow [Thu, 13 Nov 2003 15:37:52 +0000 (15:37 +0000)]
0.8.5.34
Add *INVOKE-DEBUGGER-HOOK*, which works basically like the
ANSI *DEBUGGER-HOOK* but isn't bound to NIL in calls to
BREAK, so can be used to trap _all_ debugger entry, not just
some of it. Useful for anyone who wants to completely remove
the debugger, or supplant it with something else.
Based on a patch from David Lichteblau, but this one only has
a single function instead of a list in the hook.
William Harold Newman [Tue, 11 Nov 2003 13:30:45 +0000 (13:30 +0000)]
0.8.5.33:
bumped .fasl file version number as per Tim Daly Jr.
incompatibility report on sbcl-devel 2003-11-09
Alexey Dejneka [Tue, 11 Nov 2003 09:09:39 +0000 (09:09 +0000)]
0.8.5.32:
* Fix PFD bug MISC.158: X86 (xor x x) instruction does not
work for X in memory.
Kevin Rosenberg [Tue, 11 Nov 2003 05:49:06 +0000 (05:49 +0000)]
0.8.5.31:
* contrib/sb-aclrepl/toplevel.lisp: Revert from lexically binding
certain state variables.
Daniel Barlow [Mon, 10 Nov 2003 23:44:22 +0000 (23:44 +0000)]
0.8.5.30
"Documentation" update for asdf-install
Add README file for asdf itself
Christophe Rhodes [Mon, 10 Nov 2003 23:26:37 +0000 (23:26 +0000)]
0.8.5.29:
Completely boring housekeeping commit
... rename grovel_headers.c to grovel-headers.c, as that underscore
was the only non-logical-pathname-compliant character in
the whole of the source and objects filenames.
... delete all redundant sb!vm: prefixes throughout
src/compiler/$arch/*.lisp
... nothing interesting at all, in fact. Builds and passes tests
on at least x86 and sparc.
Christophe Rhodes [Sun, 9 Nov 2003 13:35:30 +0000 (13:35 +0000)]
0.8.5.28:
Fix some loop badness (as reported by John Klein sbcl-devel
2003-11-09)
... step variable guessed initializers should be 1, not 0,
because the step type explicitly excludes 0.
... log some more loop badness in BUGS
Christophe Rhodes [Sat, 8 Nov 2003 21:08:13 +0000 (21:08 +0000)]
0.8.5.27:
SB-{POSIX/BSD-SOCKETS} cleanups
... make sb-bsd-sockets-tests depend on sb-posix so we can check
to see if /dev/log is actually a socket;
... test that /, not /root, is not writeable by others;
... (taking advantage of our 0.x version number) alter
SB-POSIX:S-ISFOO to use the BOOLEAN alien type, rather
than MODE-T -- the point being that the macros have no
function other than returning a boolean. Adjust
the sb-posix manifesto too, to document this.
Christophe Rhodes [Sat, 8 Nov 2003 15:04:45 +0000 (15:04 +0000)]
0.8.5.26:
Record filesystem information during the build in debug information
relative to SYS, not to /home/kevin/sbcl or equivalent.
... change to core code: tiny if ugly;
... change to warm load: only to refer to files as lpns rather
than physical namestrings;
... change to vanilla-modules: tiny;
... change to asdf modules: well, erm, less tiny. I'm not pleased
by the need to duplicate similar logic in every .asd, but
I think it's a similar symptom to the duplication of the
TEST-OP logic in many of them: the coupling between asdf
and the rest of the system is slightly wrong.
(At startup, SYS by default has translations that refer back to
the build-time values; distributors may wish to alter this for
their binaries, for instance by setting translations to
distribution-known values when dumping any final image)
William Harold Newman [Sat, 8 Nov 2003 01:32:02 +0000 (01:32 +0000)]
0.8.5.25:
merged Miles Egan's general SGML cleanup manual patch (from
sbcl-devel today)
deleted "!ENTITY mdash" hack (since ME's patch reminded me that
at some point I learned that DocBook itself seems to
define mdash)
William Harold Newman [Thu, 6 Nov 2003 23:33:11 +0000 (23:33 +0000)]
0.8.5.24:
merged CSR's well-aged patch to make X86 port stop
complaining about register class strangeness
in array checking VOPs
added #!+ conditionalization on the hack to suppress
warnings about analogous strangeness on other
ports, hoping I didn't screw up anything I can't test
Patrik Nordebo [Thu, 6 Nov 2003 21:50:26 +0000 (21:50 +0000)]
0.8.5.22:
gencgc cleanup:
... Replaced all (but one) literal 4096 (and 0xfff) with
PAGE_BYTES constant. I think the ones I changed were all about
page size, but I added a FIXME where I was unsure. There is
one bare 4096 left, but that's not a page size, I think.
... Removed gencgc.h, because it wasn't actually needed and
mostly duplicated stuff from gencgc-internal.h
William Harold Newman [Tue, 4 Nov 2003 00:40:20 +0000 (00:40 +0000)]
0.8.5.22:
archived some arcane knowledge of porting Hell^W^W^W^W^W wisdom
Rudi Schlatte [Mon, 3 Nov 2003 19:58:08 +0000 (19:58 +0000)]
0.8.5.21:
* Merge David Lichteblau "simple-streams bug with non-SAP
buffers" patch (sbcl-devel 2003-11-02)
Kevin Rosenberg [Mon, 3 Nov 2003 19:05:19 +0000 (19:05 +0000)]
0.8.5.20:
* contrib/sb-aclrepl/toplevel.lisp: Commit patch from David
Lictheblau to better support multiple listener processes.
Alexey Dejneka [Sun, 2 Nov 2003 09:17:53 +0000 (09:17 +0000)]
0.8.5.19:
* Fix PFD bugs MISC.100, 102, 105, 107, 112;
... IR1-MERGE-CASTS: do not put merge asserted type
contradicting the derived one;
... IR1-OPTIMIZE-COMBINATION: try terminate block after
PROPAGATE-FUN-CHANGE;
* FLUSH-DEAD-CODE: if the block is split under us, restart.
Daniel Barlow [Sun, 2 Nov 2003 00:36:07 +0000 (00:36 +0000)]
0.8.5.18
Minor change to sb-bsd-sockets tests: instead of
conditionalising the AF_LOCAL client test on (not (or sunos
darwin)), just skip it unless /dev/log exists. This make life
simpler if we're e.g. on a platform that usually has /dev/log
Christophe Rhodes [Thu, 30 Oct 2003 17:27:36 +0000 (17:27 +0000)]
0.8.5.17:
Possibly the least useful fix I shall ever make while working on
SBCL:
Implement ldso_stub__foo for MIPS/Linux, so that the system can
be built to link dynamically, allowing linking of foreign code
in
... many thanks to Ralf Baechle for education about the MIPS ABI
Now foreign.test.sh passes, and we build and pass tests for all
13 contribs! Woohoo!
Is there a user in the house?
Christophe Rhodes [Thu, 30 Oct 2003 09:41:37 +0000 (09:41 +0000)]
0.8.5.16:
Aren't we glad we have solaris around to keep us honest?
... adjust make.sh message to let it print under real sh(1)
... adjust sb-posix tests to let them pass (one true error in
the tests, and one weird EINVAL from Solaris)
Christophe Rhodes [Wed, 29 Oct 2003 19:14:59 +0000 (19:14 +0000)]
0.8.5.15:
More SB-POSIX stuff
... merge patch from Zach Beane (sbcl-devel 2003-10-28) to wrap
the stat S_ISFOO macros into callable functions
... merge patch CSR sbcl-devel to detect presence of various
symbols at compile-time for conditional function
definition
... slight frobs to make these two coexist: "interface" must
depend on "alien", and these calls are NEVER-FAILS
... tests from Xach via #lisp IRC
... as yet unresolved: duplication of UNIX-DSO stuff
Summarize contrib successes at the end of make.sh
... hope no-one's trisdekaphobic :-)
Christophe Rhodes [Wed, 29 Oct 2003 12:54:50 +0000 (12:54 +0000)]
0.8.5.14:
Be less assertive about LRAs, since perfectly valid instructions
can have LRA widetags. (Brian Downing sbcl-devel 2003-10-29)
... minimally-intrusive and minimally-DWIM patch
Define and use NATIVE-FILENAME for sb-posix
... also adjust the test not to run RUN-PROGRAM, since we
have problems with that :-/
Patrik Nordebo [Wed, 29 Oct 2003 06:38:40 +0000 (06:38 +0000)]
0.8.5.13:
G5/PPC970 fix:
... Replace all uses of the mcrxr instruction with the mtxer
instruction, because mcrxr is reallly slooow on the PPC970 and
we didn't actually want it for anything but clearing XER.
Daniel Barlow [Wed, 29 Oct 2003 03:37:22 +0000 (03:37 +0000)]
0.8.5.12
SB-POSIX now converts pathnames to Unix filenames using
SB-INT:UNIX-NAMESTRING, instead of to CL namestrings (which
are not the same thing, especially when wildchard characters
are involved) with NAMESTRING
Bug report and fix both courtesy of Zach Beane. Thanks.
Christophe Rhodes [Mon, 27 Oct 2003 22:16:58 +0000 (22:16 +0000)]
0.8.5.11:
Modifications for which I could make a plausible case on the
basis of being 64-bit SAP implementations for HPPA and MIPS
... testing? What's that? It compiles, ship it!
Christophe Rhodes [Mon, 27 Oct 2003 11:32:12 +0000 (11:32 +0000)]
0.8.5.10:
PPC support for 64-bit SAP references
Christophe Rhodes [Mon, 27 Oct 2003 09:35:17 +0000 (09:35 +0000)]
0.8.5.9:
SPARC support for 64-bit SAPs, needed for sb-posix
Alexey Dejneka [Sun, 26 Oct 2003 15:34:38 +0000 (15:34 +0000)]
0.8.5.8:
* Fix bugs MISC.101 and MISC.103 found by Paul Dietz:
FLUSH-DEST: set BLOCK-TYPE-CHECK attribute for USEs' blocks.
Alexey Dejneka [Sun, 26 Oct 2003 14:15:06 +0000 (14:15 +0000)]
0.8.5.7:
* Fix bug MISC.99 found by Paul Dietz: on X86 in
FAST-ASH-C/UNSIGNED=>UNSIGNED VOP add case (>= shift 32)
(the code is shared with -MOD32 version, where "result type
constraint" does not guarantee overflow absence).
Christophe Rhodes [Sun, 26 Oct 2003 13:14:09 +0000 (13:14 +0000)]
0.8.5.6:
Fix for LOGAND OPTIMIZER and 0s
... use UNSIGNED-BYTE*
Alexey Dejneka [Sun, 26 Oct 2003 11:36:24 +0000 (11:36 +0000)]
0.8.5.5:
* Fix bug reported by Brian Downing: do not perform
MV-LET-convertion, if the last optional entry has
references.
... new consistency condition: function in a local
mv-combination must be of kind MV-LET.
Daniel Barlow [Sun, 26 Oct 2003 04:04:28 +0000 (04:04 +0000)]
0.8.5.5
BUILD_ROOT
Christophe Rhodes [Sat, 25 Oct 2003 21:40:48 +0000 (21:40 +0000)]
0.8.5.3:
CLISP minor issues:
... fix float-related bug: since CLISP doesn't support
denormalized single-floats, we need to construct
LEAST-POSITIVE-SHORT-FLOAT explicitly
... we need to perform a similar trick for BOOLE as we already
do for BYTE. Rewrite the horrible package hackery to be
slightly more robust, in mitigation for the perpetuation
of this horror.
Christophe Rhodes [Sat, 25 Oct 2003 21:34:35 +0000 (21:34 +0000)]
0.8.5.2:
Large increase in entropy, only mitigated by the fact that we
cover more of POSIX and I've added a test suite...
... make SB-GROVEL accessors not cons quite so much while
looking for the address of the thing they're referencing
... add whole bundle of new functions in SB-POSIX...
... which necessitates SAP-REF-64 and friends, which probably
work everywhere except MIPS
... declaim *GC-INHIBIT* as a fixnum, because if more than 2^29
threads are inhibiting gc we probably have more problems
... beginnings of a test suite for SB-POSIX, which incidentally
shows up various problems with using it (see FIXME
comments dotted around)
... please feel free to fix any and all of the issues marked;
I'm by no means claiming ownership of this :-)
Daniel Barlow [Sat, 25 Oct 2003 20:56:01 +0000 (20:56 +0000)]
0.8.5.1
$PREFIX/share/man/man1, not $PREFIX/man/man1
William Harold Newman [Sat, 25 Oct 2003 17:14:09 +0000 (17:14 +0000)]
0.8.5:
release, mostly tagged as sbcl_0_8_5
(I thoughtlessly did "cvs tag" before "cvs commit", so I
think the old NEWS and version.lisp-expr files were
probably tagged, not the new ones.)
Daniel Barlow [Fri, 24 Oct 2003 10:23:55 +0000 (10:23 +0000)]
0.8.4.40
Removed debugging message about unexpectedly enabled signals.
These days we're older and wiser and expect that stuff to
happen
contrib/experimental-thread.patch, as the name suggests, is
a patch that may help thread stability problems, but hasn't
been sufficiently well tested to get merged into SBCL proper
this late int he release cycle
Christophe Rhodes [Wed, 22 Oct 2003 16:27:18 +0000 (16:27 +0000)]
0.8.4.39:
CLISP build "fixes"
... don't emit structure-sharing code from macros
(fixes "SB-IMPL::|,| does not name a function" clisp
error)
... rewrite bad LOOP as per TFBurdick sbcl-devel 2003-10
I say `"fixes"' because new sbcl code has exposed some
clisp bugs, so the resulting codebase doesn't build from
clisp-2.31; however, I have patched my clisp and it works to the
extent that the resultant sbcl binary only fails one test.
Christophe Rhodes [Wed, 22 Oct 2003 11:30:19 +0000 (11:30 +0000)]
0.8.4.38:
Solaris fixes (from Nikodemus Siivola sbcl-devel 2003-10-22)
... sh(1) apparently doesn't like $(foo). Make it `foo`.
... allow building from cmucl/sparc, by suppressing underflow
traps
... /dev/random doesn't exist. /dev/zero better had, though
Christophe Rhodes [Mon, 20 Oct 2003 15:29:20 +0000 (15:29 +0000)]
0.8.4.37:
Fix bug 46k: READ-BYTE on STRING-INPUT-STREAMs
... add a test
... some other BUGS filtrage (BIT-VECTOR is required to be
disjoint from SIMPLE-VECTOR; other fixed bugs)
Christophe Rhodes [Mon, 20 Oct 2003 13:31:06 +0000 (13:31 +0000)]
0.8.4.36:
Fix bug 213a
... CONS-TYPE-LENGTH-INFO to walk CONS-TYPE lists
... delete the neat but ultimately flawed (CONS NIL T) test
and use a proper test instead
... test suite additions.
Add idea from Michael Hudson (sbcl-devel 2003-08-26) to exit
early from Darwin compilations when the stack size
limit is too small.
Christophe Rhodes [Sun, 19 Oct 2003 19:10:14 +0000 (19:10 +0000)]
(logically part of 0.8.4.35 commit, but forgot to cvs add)
Christophe Rhodes [Sun, 19 Oct 2003 19:09:11 +0000 (19:09 +0000)]
0.8.4.35:
One last little PCL frob
... when declaring defgenerics, don't foolishly assume that all
extended function names are of the form (SETF ..)
... now internal PCL generic functions can have lambda lists
... and that's all of the ones at startup, so let's write an
internalsish test to ensure that it stays that way
Christophe Rhodes [Sun, 19 Oct 2003 18:08:07 +0000 (18:08 +0000)]
0.8.4.34:
Make MODULE-PROVIDE-CONTRIB obey the protocol
... hacky solution as per CSR sbcl-devel 2003-10-18
Andreas Fuchs [Sun, 19 Oct 2003 17:52:13 +0000 (17:52 +0000)]
0.8.4.33:
Add a check for working CONSTANTLY in OpenMCL (some CVS versions
throw an error)
Daniel Barlow [Sat, 18 Oct 2003 17:17:43 +0000 (17:17 +0000)]
0.8.4.32
More for SB-INTROSPECT, shamelessly inspired by Helmut
Eller's SLIME code for CMUCL.
FIND-FUNCTION-CALLERS, FIND-FUNCTION-CALLEES search in CODE
objects for FDEFN references and return functions. Users of
the former interface are probably advised to cache the
answers, or to accept that it's presently rather slow.
Christophe Rhodes [Sat, 18 Oct 2003 12:55:41 +0000 (12:55 +0000)]
0.8.4.31:
Minor PCL fixette
... when defining early generic functions in the bootstrap
process, clobber the lambda list if the gf has already
been created (e.g. NO-APPLICABLE-METHOD) by an early
DEFMETHOD.
Christophe Rhodes [Sat, 18 Oct 2003 10:14:52 +0000 (10:14 +0000)]
0.8.4.30:
Be more careful over automatically generated generic function
lambda lists
... when generating PCL-internal GFs, pass :LAMBDA-LIST to
ENSURE-GENERIC-FUNCTION
... when generating accessor GFs, pass :LAMBDA-LIST if the
function is not already created (where you want to
preserve the user's lambda list instead)
... tests for required behaviour
Adjust INVALID-FASL patch slightly
... comment in package-data-list.lisp-expr
... remove unneccessary sb!ext:: prefixes
Andreas Fuchs [Fri, 17 Oct 2003 16:16:02 +0000 (16:16 +0000)]
0.8.4.29:
LOOP fixups - whee, I love digging around in code from 1986
* make SB-LOOP::LOOP-SEQUENCER no longer choke on NIL
as a name for for-as-arithmetic counters
* also make it throw a PROGRAM-ERROR when it encounters
a list as a counter variable.
Andreas Fuchs [Fri, 17 Oct 2003 15:44:22 +0000 (15:44 +0000)]
0.8.4.28:
ASDF users, rejoice!
* Add a condition SB-EXT:INVALID-FASL (subtype ERROR)
* Make SB-FASL::CHECK-FASL-HEADER raise conditions that are subtype
SB-EXT:INVALID-FASL on the appropriate errors (or the other way
around, I forget)
* Make SB-FASL::INTERNAL-LOAD also raise a condition if the FASL
header is broken.
Christophe Rhodes [Fri, 17 Oct 2003 14:27:44 +0000 (14:27 +0000)]
0.8.4.27:
OK, here's the deal.
... *MODULE-PROVIDER-FUNCTIONS* no longer call PROVIDE. It's
now the job of user code to do that (as it probably
should always have been.
... OK, so as a result, our one-file contribs now need to
PROVIDE themselves. Make it so.
... fix asdf-install/asdf autorequiration; an installed sbcl
can now require asdf-install (as well as all the other
asdfized contribs) as the first command in a clean core
... update contrib/{README,STANDARDS} to match what we currently
do.
Daniel Barlow [Thu, 16 Oct 2003 12:14:33 +0000 (12:14 +0000)]
0.8.4.26
Innie? Outie?
Take the first debug-fun from the fun-map vector in debug-info
instead of looking for one with a matching name, which may not
exist. We only want it for the top-level form number anyway
compiler/debug-dump.lisp; write the tlf number at all debug
quality levels. 0 would take up just as much space as n does
anyway
Daniel Barlow [Thu, 16 Oct 2003 12:03:32 +0000 (12:03 +0000)]
0.8.4.25
Update text files in contrib/ directory
Christophe Rhodes [Thu, 16 Oct 2003 09:34:55 +0000 (09:34 +0000)]
0.8.4.24:
Since it's rapidly becoming an FAQ
... additional magic in asdf-module.mk to make synthetic fasls
allowing REQUIRE on an asdf module to work without
a loaded ASDF.
... does not work for the ASDF-INSTALL contrib, because REQUIRE
goes for the file named "asdf-install" rather than
"asdf-install.fasl" in preference. Probably want to fix
REQUIRE for that, but that's pending the other REQUIRE
issues
Christophe Rhodes [Wed, 15 Oct 2003 16:28:14 +0000 (16:28 +0000)]
0.8.4.23:
Fix for bug 191c (and some of PFD's tests)
... do proper keyword argument checking in the effective method
... not the cleanest fix in the world (note especially
the use of PROGN as an optimization inhibitor)
... I'm not telling you how long it took me to find the
NCONC -> APPEND bug in fngen.lisp