William Harold Newman [Thu, 7 Jun 2001 17:15:54 +0000 (17:15 +0000)]
0.6.12.24:
added a few tests as examples of what I'd like to have before
merging the Mai/Atzmueller inline type test patches
merged DB Alpha-dynamic-loading patches (sbcl-devel 2001-05-11)
William Harold Newman [Thu, 7 Jun 2001 13:01:24 +0000 (13:01 +0000)]
0.6.12.23:
merged flaky2_branch back into main branch
William Harold Newman [Wed, 6 Jun 2001 22:59:07 +0000 (22:59 +0000)]
0.6.12.21.flaky2.2:
(This version seems not to be flaky any more. I'm just
checking it in as a halfway step to merging it
back into the main branch.
tweaked scavenge() argument handling so that it's easier to see
what was going on when we look at it with gdb
deleted unused undocumented DIRECT_SCAV stuff
deleted unused undocumented SC_NS_GEN_CK stuff
stopped tests/interface.pure.lisp from generating un-GC-able
nonsense values when a SYMBOL-FUNCTION is a closure;
and tried to make it more accurately check for
undocumented external functions as well
William Harold Newman [Wed, 6 Jun 2001 21:45:48 +0000 (21:45 +0000)]
(no version):
updating/committing again to get CVS to recognize deletion
of early-load and addition of early-fasl
William Harold Newman [Wed, 6 Jun 2001 21:43:50 +0000 (21:43 +0000)]
0.6.12.22:
DB logical pathname and COMPILE-FILE-PATHNAME patch from
sbcl-devel 2001-06-01
made FASL file version number independent of CPU type; also
made byte FASL file version number the same
many many changes more or less related to changing the FASL
file version number..
..created SB!FASL package to hold stuff like +FASL-FILE-VERSION+
which is common to loading and dumping
..renamed early-load.lisp to early-fasl.lisp
..renamed FASL-FILE stuff to FASL-OUTPUT
..*LAP-OUTPUT-FILE* doesn't need to be special (and so is
renamed to LAP-FASL-OUTPUT).
..exported FAST-READ stuff from SB-INT, so that it's visible
in SB-FASL
..deleted old stale load-related variables
*LOAD-BYTE-COMPILED-CODE-TO-DYNAMIC-SPACE*
*LOAD-X86-TLF-TO-DYNAMIC-SPACE*
*ENABLE-DYNAMIC-SPACE-CODE*
(hardwiring code which referred to them to use their
effectively-constant values T, NIL, T instead). (Now
we no longer need to worry about what package they're
in..)
..made some symbols external to SB-KERNEL so that
now-in-SB-FASL code could find them (and so that
some of the old :: prefixes could go away):
*FREE-INTERRUPT-CONTEXT-INDEX*,
*!INITIAL-FOO* for all FOO,
*CURRENT-CATCH-BLOCK*,
*CURRENT-UNWIND-PROTECT-BLOCK*,
*PSEUDO-ATOMIC-ATOMIC*, *PSEUDO-ATOMIC-INTERRUPTED*
..deleted unneeded "SB!IMPL::" prefixes for various
other *STATIC-SYMBOLS*-related symbols which're
exported (mostly from SB-KERNEL) already
..deleted unused %INITIAL-FUNCTION (smashing Alpha
*STATIC-SYMBOLS* indices)
suppressed some stuff in side-effectful-pathnames.test.sh
until we merge the DB cleanups from flaky2_branch
William Harold Newman [Sun, 3 Jun 2001 14:58:59 +0000 (14:58 +0000)]
0.6.12.21.flaky2.1:
Since enabling QSHOW and QSHOW_SIGNALS showed that a relatively
early symptom of the memory-corruption problem is
dereferencing a 0 pointer, I added special handling
for not-our-responsibility SIGSEGV/SIGBUS in
gencgc.c.
William Harold Newman [Thu, 31 May 2001 21:12:04 +0000 (21:12 +0000)]
0.6.12.21.flaky2:
(This version has a flaky GC/memory-corruption problem. It's
perfectly repeatable both on my OpenBSD box and on
my Linux box -- just "sh run-tests.sh" -- which is
why I thought it was worth saving this before I
messed with it too much. But it's flaky in that all
sorts of little harmless-looking program or test
changes made while trying to isolate the problem can
cause it to vanish.)
Dan Barlow's logical pathname tests sbcl-devel 2001-05-31
added DEFUN SANE-DEFAULT-PATHNAME-DEFAULTS
made all relative-to-absolute conversions go through
*DEFAULT-PATHNAME-DEFAULTS*, so Unix current directory
is only used to initialize *DEFAULT-PATHNAME-DEFAULTS*
(and thus UNIX-MAYBE-PREPEND-CURRENT-DIRECTORY goes
away)
William Harold Newman [Wed, 30 May 2001 23:02:35 +0000 (23:02 +0000)]
0.6.12.21:
(realized I was confused about DEFAULT-DIRECTORY ('cause it's
in SB-INT, not SB-INT) so it doesn't need to be
marked deprecated after all. As I get older, perhaps I
should work on smaller systems..)
deleted unused %SET-DEFAULT-DIRECTORY stuff
deleted related Unix chdir() stuff
rewrote UNIX-CURRENT-DIRECTORY to use getcwd() and do its
own error-handling; got rid of DEFAULT-DIRECTORY
deleted related Unix getwd() stuff
restored SIZE-T (but this time through grovel_headers.c)
in order to interface to getcwd() more cleanly
deleted commented-out clutter in ldso-stubs.S
renamed UNIX-CURRENT-DIRECTORY to POSIX-GETCWD, and made
POSIX-GETCWD/ variant to do the old CMU-CL-style
trailing #\/ thing
William Harold Newman [Wed, 30 May 2001 01:19:51 +0000 (01:19 +0000)]
0.6.12.20:
made monitor.c quit() command read from the same stream
as the other ldb commands
renamed monitor.c devtty stuff to ldb_in
miscellaneous other tidying up in monitor.c
William Harold Newman [Wed, 30 May 2001 00:01:23 +0000 (00:01 +0000)]
0.6.12.19:
deprecated #'SB-EXT:DEFAULT-DIRECTORY, since
*DEFAULT-PATHNAME-DEFAULTS* is the ANSI way to do it
failed to get rid of DEFVAR *DEFAULT-PATHNAME-DEFAULTS*
initialization, since someone is using this var
too early in cold init. Maybe next time..
removed stale (SEARCH-LIST "default:") stuff
deleted unused CADDR-T, DADDR-T, QADDR-T, QUAD-T, UQUAD-T,
SIZE-T, SSIZE-T, +MAX-S-LONG+, +MAX-U-LONG+, SWBLK-T
William Harold Newman [Mon, 28 May 2001 23:44:10 +0000 (23:44 +0000)]
0.6.12.18:
added code to initialize *DEFAULT-PATHNAME-DEFAULTS*,
to start making it a little more ANSI
started documenting visible-in-the-interface variables
whose old values are discarded when a core is saved
and loaded (e.g. *DEFAULT-PATHNAME-DEFAULTS*)
tweaked side-effectful-pathnames.test.sh so it
accepts the new more-ANSI behavior of
*DEFAULT-PATHNAME-DEFAULTS*
hacked LOAD to physicalize pathnames before merging them,
to work around apparent MERGE-PATHNAMES problems
and let tests/filesys.pure.lisp work with new
nontrivial *DEFAULT-PATHNAME-DEFAULTS*; also
hacked COMPILE-FILE-PATHNAME similarly
factored out DEFUN PHYSICALIZE-PATHNAME
made FILE-LENGTH "work" (up to 32 bits) on OpenBSD again
William Harold Newman [Sun, 27 May 2001 15:20:15 +0000 (15:20 +0000)]
0.6.12.17:
revised address map for OpenBSD (fixing memory corruption
problem)
William Harold Newman [Sun, 27 May 2001 13:02:55 +0000 (13:02 +0000)]
0.6.12.16:
made LOAD handle EXTERNAL-FORMAT just the same way as
COMPILE-FILE already did, refusing any value other
than :DEFAULT
removed unused (and deprecated on OpenBSD) setregid/setreuid
stuff
converted "i386" to "__i386__" at various
"egrep 'def.*[^_]i386'" locations
added src/runtime/trymap.c (and noted that trymap fails for
the mapped area which the OpenBSD port uses for its
control stack, so now the mystery is no longer why
the OpenBSD port now complains about memory faults,
but why the OpenBSD port ever worked in the past)
William Harold Newman [Sat, 26 May 2001 20:33:48 +0000 (20:33 +0000)]
0.6.12.15:
turned interrupt_install_low_level_handler() into
undoably_install_low_level_interrupt_handler(), which
uses atexit() to arrange for the handler to be removed
at exit, for tidiness in general and specifically
because without the the interaction between the GC
and the exit()-time segfaults caused by memory
corruption is making it tricky to debug the
OpenBSD port
William Harold Newman [Fri, 25 May 2001 22:52:31 +0000 (22:52 +0000)]
0.6.12.14:
fixes from Arthur Lemmens' 2001-05-05 sbcl-devel message I..
..wrapped an EVAL-WHEN around the definition of +EMPTY-HT-SLOT+
..removed irrelevant COUNT term in in LOOP form in
compiler/x86/call.lisp
..moved DEFCONDITION SIMPLE-TYPE-ERROR from error.lisp to
late-target-error.lisp
..moved DEFUN DO-PENDING-INTERRUPT from signal.lisp to
target-signal.lisp, since its definition ("magically
converted to a BREAK instruction") only makes sense
to Python (and causes a warning in Lispworks)
William Harold Newman [Wed, 23 May 2001 14:00:15 +0000 (14:00 +0000)]
0.6.12.13:
set up :SB-AFTER-XC-CORE feature so slam.sh is more convenient
The various core address mismatch states should be fatal
errors, not just fprintf()s.
William Harold Newman [Tue, 22 May 2001 16:10:46 +0000 (16:10 +0000)]
0.6.12.12:
(I went hunting unsuccessfully OpenBSD GC problems, didn't
find them, and made small cleanups instead.)
Dan Barlow's way of dealing with 64-bit slots in wrapped_stat
is better than mine, and it's possible that mine is
even responsible for my OpenBSD memory corruption
problems, so convert the wrapped_st_size slot to
32 bits.
also renamed DB's my_dev_t to more-explicit ffi_dev_t
William Harold Newman [Sun, 20 May 2001 19:34:35 +0000 (19:34 +0000)]
0.6.12.11:
added #\% prefixes for names of helper functions in hash table
stuff, so they won't look so friendly and supportedish
bumped fasl file version number for the new hash table forms
restored support for compiler trace output (inspired by
MNA's patch, but controlled by the CMU-CL-style
&KEY arg interface, not a special variable)
tweaked other &KEY options of COMPILE-FILE: no more
:ENTRY-POINTS, and anticipating deprecating other stuff
William Harold Newman [Sun, 20 May 2001 12:53:51 +0000 (12:53 +0000)]
0.6.12.10:
MNA sbcl-devel 2001-05-12 "logical pathname patch & readable
hashtables" patch
William Harold Newman [Sat, 19 May 2001 19:59:08 +0000 (19:59 +0000)]
0.6.12.9:
merged MNA port sbcl-devel 2001-05-11 of Tim Moore CMU CL
improved disassembly patch
miscellaneous other SBCLification and modernization of
target-disassem.lisp
William Harold Newman [Sat, 19 May 2001 01:03:02 +0000 (01:03 +0000)]
0.6.12.8:
merged flaky1 branch back into main branch, now that it's
stable again.
(Dan Barlow's email on sbcl-devel warned that one problem
with branches is that useful checkin comments
end up being hard to find when they're stuck on
branches. I'll check the cvs2cl.pl output
after this checkin, and if it doesn't include
the notes for the various checkins on
the flaky1 branch, I'll make another checkin
below just as an excuse to add log notes which
copy or summarize them.)
The flaky1 branch was rather a lot of work for what turned
out to be not very many fixes. ("The discipline of
refactoring is hard to learn and easy to
lose sight of, even if only for a moment. I
still lose sight more often than I care to admit.
I'll do two or three refactorings in a row without
rerunning the test cases. Of course I can get away
with it. I'm confident. I've practiced. Boom! A
test fails, and I can't see which of my changes
caused the problem. At this moment, you will be
mightily tempted to just debug your way out of
trouble. After all, you got those tests to run
in the first place. How hard could it be to get
them running again? Stop. You are out of control,
and you have no idea what it will take to get back
in control by going forward. Go back to your
last known good configuration after each one."
-- Kent Beck, on pp. 410-411 in Martin Fowler's
_Refactoring_)
William Harold Newman [Sat, 19 May 2001 00:13:14 +0000 (00:13 +0000)]
0.6.12.7.flaky1.2:
(This version seems not to be flaky any more, and so should
be the end of the flaky1 branch: I'm just checking
it in in preparation for "cvs update -j flaky1" or some
such thing to merge the changes back into the main
branch.)
tweaked run-tests.sh machinery to use an absolute path for
SBCL, on general principles and specifically so that
Dan Barlow's new ENSURE-DIRECTORIES-EXIST tests can run
made (UNIX-NAMESTRING "") equivalent to (UNIX-NAMESTRING "."),
as the simplest way to make ENSURE-DIRECTORIES-EXIST
work for relative pathnames. (As per Dan's 2001-05-18
sbcl-devel mail, it can't be ANSI conforming until
*DEFAULT-PATHNAME-DEFAULTS* is made ANSI, and that's
too messy for now, so just do the simple thing.)
dropped search list support from UNIX-NAMESTRING while I was
at it, since search lists ain't supported no more
removed undocumented hardly-used EXECUTABLE-ONLY arg from
UNIX-NAMESTRING, moving the "is it executable?" test
into run-program.lisp
removed REMOVEME stuff
William Harold Newman [Fri, 18 May 2001 13:22:50 +0000 (13:22 +0000)]
0.6.12.7.flaky1.1:
(As per Daniel Barlow sbcl-devel 2001-05-17, the removal of
the mysterious "" special case in UNIX-STAT will
probably keep this version from building itself,
so I didn't even try. It does, however, at least
do "sh run-tests.sh" successfully.)
fixed declaration in %EXTRACT-STAT-RESULTS so that the
system won't have to interpret the alien reference
at runtime on every call to the function
I found how to fix the boot/gencgc/purify problem which was
leaving initial-function unmapped -- I could copy
current_region_free_pointer into
boxed_region.free_pointer just before
gc_alloc_update_page_pages() in save.c. However,
that left me with other flaky GC problems. So I gave
up my GC hacking in this version as a bad job, and
copied original 0.6.12.7 GC-related files over
the modified versions: gencgc.c, save.c, purify.c,
gc.lisp, save.lisp
made %EXTRACT-STAT-RESULTS inline to suppress an apparent
memory corruption bug
William Harold Newman [Tue, 15 May 2001 14:08:31 +0000 (14:08 +0000)]
0.6.12.7.flaky1:
(This system state is seriously screwed up. It did build,
both on Linux and on OpenBSD, but signalled an
error after writing out sbcl.core on OpenBSD, and
although it runs a little both on Linux and on
OpenBSD, it hangs in the regression tests. It
also has some temporary hacks marked REMOVEME,
including one which suppresses PURIFY when
building the system (!).)
made the system build on OpenBSD again..
..stubbed stuff out as a quick fix to the problem of FFI to
64-bit stat.st_size
tried to make the system run on OpenBSD again..
..initialized current_dynamic_space, since it's now used
instead of DYNAMIC_SPACE_START in PURIFY
..added new assertions about GENCGC alloc_region stuff being
reset when it should be
renamed Lisp-level struct stat stuff to struct wrapped_stat
tried to tidy up Lisp-level stat stuff; removed mysterious
(STRING= NAME "") behavior from UNIX-STAT
added slam.sh to help in low-level compile-and-try cycle
pulled alloc_region-is-reset logic out into separate
functions, and added more assertions on it (hunting
for a bug which broke the old assertions)
renamed gc_alloc_large to gc_alloc_possibly_large, and
gc_quick_alloc_large_unboxed to
gc_quick_alloc_unboxed_possibly_large
enabled various GC checks, e.g. gencgc_zero_check=1
deleted unused (and bizarre..) gencgc_verify_zero_fill()
turned off PURIFY in order to get the system to build, since
PURIFY seems to be misbehaving (leaving INIT-FUNCTION
above the cutoff address..)
William Harold Newman [Thu, 10 May 2001 15:18:02 +0000 (15:18 +0000)]
0.6.12.7:
Make sure that we "mkdir output/" before we use it in the build
process. (MNA pointed out that we used it before we
made it.)
Use "uname -m" to figure out sbcl_arch default.
merged MNA PCL fixes from sbcl-devel 2001-05-09 (including
port of Pierre Mai's method combination fixes from
cmucl-imp 2001-04-26)
added regression test for method combination fix
Bug 14 was mostly fixed already: the problem with
INVALID-METHOD-ERROR complaining about being outside
a method combination function went away some time ago.
The MNA/Mai patch above improves method combination
error reporting further, so it's definitely time
to retire 14 from BUGS.
William Harold Newman [Wed, 9 May 2001 19:20:18 +0000 (19:20 +0000)]
0.6.12.6:
merged Dan Barlow's patches sbcl-devel 2001-05-08 and
2001-05-09
William Harold Newman [Wed, 9 May 2001 00:02:00 +0000 (00:02 +0000)]
0.6.12.5:
made tests/foreign.test.sh more OpenBSD-friendly
more Alpha patch tweaking..
..removed extra "#define QSHOW 1" so that C-level
describe_internal_error() becomes a debugging-only
thing again
..copied boilerplate text into the head of the new files
(and moved some information from the head of the new
files to CREDITS)
..lotso cosmetic formatting changes
..deleted src/compiler/alpha/print.lisp, since it's
redundant with src/compiler/alpha/show.lisp
..deleted "dead file" src/runtime/alpha-validate.h
..removed in-the-flow-of-control EXPORTs and USE-PACKAGEs
(in favor of centralized control in
package-data.lisp-expr)
William Harold Newman [Tue, 8 May 2001 00:13:54 +0000 (00:13 +0000)]
0.6.12.4:
made grovel-headers.c run automatically as part of the build
process (so that the automatically-generated file
output/stuff-groveled-from-headers.lisp replaced the
old symlink src/code/target-os-types.lisp)
redid grovel_headers.c DEFCONSTANT macro as function
removed 'ln -sf $sbcl_arch-linux-types.lisp' stuff (from
src/runtime/make-config.sh??)
redid the definition of a stem in stems-and-flags.lisp,
as a relative path from SBCL base directory instead
of from the src/ subdirectory, so that it can now
refer to "output/stuff-groveled-from-headers.lisp
William Harold Newman [Mon, 7 May 2001 20:16:21 +0000 (20:16 +0000)]
0.6.12.3:
NOTE: This is a slightly screwed-up checkin, which won't
quite build as is, because the checked-in changes from
the Alpha port expect various src/code/*types.lisp
files (e.g. src/code/x86-linux-types.lisp) which
Daniel Barlow created by hand once and for all. Those
are in my working directory but I didn't check them in
because almost immediately, probably 0.6.12.4, I
expect to rearrange things so that the build
process generates a types file on every run of
make.sh. So since CVS add/delete is a little mixed up,
I thought I'd just skip it, even though it means that
this version isn't quite buildable unless you grab
those files from DB's patch.
merged Daniel Barlow's port to the Alpha CPU sbcl-devel
2001-05-04
moved CIRCULAR-LIST-P to early-extensions.lisp (since it
logically belongs there, and DB's new definition,
unlike the old one, doesn't have a physical dependence
on IGNORE-ERRORS which prevents it from being there)
merged boot-extensions.lisp, early-extensions.lisp, and
late-extensions.lisp, since there's no longer any
distinction between the files
deleted CVS files from sbcl-alpha-extra-files.tar, since
they were probably an oversight, I think
also src/compiler/alpha/z050.pdf:-|
The patch change
- (/hexstr arguments)
+ #!+sb-show (/hexstr arguments)
shouldn't be necessary because /SHOW, like all the
/FOO macros, is already conditional on SB-SHOW. So I
undid this change (and added some comments in
src/code/show.lisp explaining the intended behavior
of /FOO macros).
merged stat_wrapper.c into the relatively new, related
wrap.c file; deleted stat_wrapper.h since stat wrappers
seem never intended to be used by C code anyway
renamed build-tools/ to tools-for-build/ in the interest of
unambiguous (even painfully unambiguous)
globally-visible names
tweaked tests/alien.impure.lisp so that the test uses a
non-GENCGC-dependent variable (instead of having to
have the test conditional on #+GENCGC).
William Harold Newman [Mon, 7 May 2001 01:37:00 +0000 (01:37 +0000)]
0.6.12.2:
MNA port of DTC VALUES declarations patch
sbcl-devel/2001-05-04
William Harold Newman [Sun, 6 May 2001 18:37:58 +0000 (18:37 +0000)]
0.6.12.1:
MNA dumping-a-logical-host patch sbcl-devel/2001-05-02
MNA catch-init-file-errors patch sbcl-devel/2001-05-03
MNA port of DTC UNREAD-CHAR/CLEAR-INPUT patch for
encapsulated streams sbcl-devel/2001-05-03
William Harold Newman [Sat, 5 May 2001 02:44:44 +0000 (02:44 +0000)]
0.6.12:
removed a little REMOVEME stuff, otherwise like 0.6.11.45
William Harold Newman [Thu, 3 May 2001 22:17:59 +0000 (22:17 +0000)]
0.6.11.45:
more MNA byte-compilation patches (sbcl-devel 2001-04-30 and
2001-05-02)
William Harold Newman [Fri, 27 Apr 2001 19:11:34 +0000 (19:11 +0000)]
0.6.11.43:
MNA warm init byte-compilation patch (sbcl-devel 2001-04-14);
except that I didn't clobber the DEFKNOWNs for
funcallable instance accessors
fixed the byte-compile-time warning problem in describe.lisp
by redefining SB-PCL::GET-SLOTS-OR-NIL as a function
merged pcl/fin.lisp into pcl/low.lisp in order to unscrew
some assumed-to-be-function-but-then-defined-as-macro
problems, and because now that our PCL is SBCL-specific
pcl/fin.lisp stuff can mostly wither away anyway
inlined CLOS-SLOTS-REF and (SETF CLOS-SLOTS-REF)
deleted DOCTOR-DFUN-FOR-THE-DEBUGGER and
RECORD-DEFINITION
William Harold Newman [Thu, 26 Apr 2001 15:47:52 +0000 (15:47 +0000)]
0.6.11.42:
MNA port (sbcl-devel 2001-04-14) of DTC cmucl-imp patch: Use
VALUE-CELL objects rather than conses for the
interpreter's value cells.
more ANSI compliance: BREAK binds *DEBUGGER-HOOK* to NIL
before calling INVOKE-DEBUGGER.
SIGINT handling no longer uses CL:BREAK (which rebinds
*DEBUGGER-HOOK* before doing INVOKE-DEBUGGER, screwing
up --noprogrammer handling) but instead calls
INVOKE-DEBUGGER more directly.
NOPROGRAMMER-DEBUG-HOOK-FUN now terminates "normally"
(calling Unix exit(3), not falling into infinite
recursion in Lisp-level error trapping) even
when shell/pipeline/SIGINT or other mischief hoses
its low-level output streams.
renamed FASLOAD and SLOLOAD to LOAD-AS-FASL and LOAD-AS-SOURCE,
so that users have a better chance of understanding
them when they appears in BACKTRACE
William Harold Newman [Tue, 17 Apr 2001 20:50:30 +0000 (20:50 +0000)]
0.6.11.41:
moved Unix opendir/readdir/closedir iteration down to C level
so that cruft can be read directly from #include files
replaced old *UNIX-ERROR* cruft with STRERROR and
SIMPLE-PERROR and a few DEFCONSTANTs
increased ANSIness of error reporting in fd-stream.lisp
William Harold Newman [Tue, 17 Apr 2001 14:43:42 +0000 (14:43 +0000)]
0.6.11.40:
added code to support coming changes in (probably) 0.6.11.41..
..moving Unix opendir/readdir/closedir iteration down to the C
level, so that all structure layouts and whatnot can be
read directly from #include files, so that directory
operations become inherently portable (or at least as
much as C/Unix ever is:-) and work on OpenBSD
..replacing old *UNIX-ERROR* cruft with STRERROR and
SIMPLE-PERROR
GET-ERRNO belongs in SB-C-CALL, it's not Unix-specific.
William Harold Newman [Mon, 16 Apr 2001 16:33:07 +0000 (16:33 +0000)]
0.6.11.39:
trivial changes syncing my OpenBSD and Linux checkouts
William Harold Newman [Mon, 16 Apr 2001 14:05:41 +0000 (14:05 +0000)]
0.6.11.38:
got rid of CONDITION-FUNCTION-NAME logic, since it was messy
and it didn't work very well and, given BACKTRACE, it
seemed mostly redundant
bumped fasl file version since CONDITION layout changed
renamed FIND-CALLER-NAME to FIND-CALLER-NAME-AND-FRAME
used BREAK's *STACK-TOP-HINT* idiom in ERROR and CERROR too
removed PRINT-SIMPLE-ERROR stuff, so that SIMPLE-ERROR just
prints as SIMPLE-CONDITION
WHITESPACE-CHAR-P belongs in target-char.lisp (and in SB!INT).
William Harold Newman [Sun, 15 Apr 2001 16:42:05 +0000 (16:42 +0000)]
0.6.11.37:
MNA 2001-04-13 CHECK-TYPE/STORE-VALUE patch
replaced CHECK-TYPE calls with lighter-weight stuff
rewrote CHECK-TYPE-VAR with lighter-weight stuff
redid STREAM-ASSOCIATED-WITH-FILE-P
renamed PARSE-OPERANDS to !PARSE-VOP-OPERANDS, and
GROVEL-OPERANDS to !GROVEL-VOP-OPERANDS, and
PARSE-OPERAND-TYPES to !PARSE-VOP-OPERAND-TYPES
(and queued up various FOO -> !FOO renamings for
after Alpha port is merged)
made INVALID-METHOD-ERROR and METHOD-COMBINATION-ERROR
stop screwing around with DEFVARs
added *DEBUG-BEGINNER-HELP-P*
William Harold Newman [Sun, 15 Apr 2001 00:24:44 +0000 (00:24 +0000)]
0.6.11.36:
added support for --noprogrammer command line option
tweaked tests/*.test.sh stuff to use the most recently built
SBCL in the current tree (as other tests/* stuff does)
renamed some FD-STREAM stuff to be more explicit
PICK-BACKUP-NAME and ENSURE-ONE-OF needn't be so general.
reduced generality of BEEP too
ENSURE-ONE-OF should signal a TYPE-ERROR.
converted some other CERRORs to ERRORs; redid OPEN errors
rearranged error output to use pretty printer more
William Harold Newman [Thu, 12 Apr 2001 22:50:34 +0000 (22:50 +0000)]
0.6.11.35:
fixes for problems reported by Nathan Froyd on sbcl-imp
2001-04-11..
..changed "EXT:" to "SB-EXT:" in OUTPUT-FLOAT-INFINITY
..fixed screwed-up indentation in iterated DESCRIBE
fixed --notty/--noprogrammer confusion reported by Christopher
Rhodes sbcl-devel 2001-04-08, and generally rewrote
sbcl.1 --noprogrammer documentation in anticipation of
the way it's likely to actually work
various other sbcl.1 revisions
NO-APPLICABLE-METHOD doesn't need to do CERROR. (And its
message can be printed more prettily, too.:-)
William Harold Newman [Sat, 7 Apr 2001 17:23:13 +0000 (17:23 +0000)]
0.6.11.34:
eliminated distinction between ordinary optimization policy
and *INTERFACE-POLICY*
eliminated *INTERFACE-POLICY* stuff, including INTERFACE flags
to IR1-CONVERT-AUX-BINDINGS, IR1-CONVERT-LAMBDA-BODY,
IR1-CONVERT-SPECIAL-BINDINGS
converted &OPTIONAL args in IR1-CONVERT-LAMBDA-BODY to &KEY
*POLICY-BASIC-QUALITIES* becomes *POLICY-QUALITIES*
merged redundant OPTIMIZATION-QUALITY-P and
POLICY-QUALITY-NAME-P
(Bug 87 was fixed by getting rid of interface policy stuff.)
merged nearly-irrelevant SB-INSPECT package into SB-IMPL
rearranged new function definition vs. declaration type
checking, so that the compiler won't die when
(DECLAIM (FTYPE FUNCTION F)) (DEFUN F ..)
removed src/compiler/c.log (>10-year-old log file)
added *.impure-cload.lisp category of tests, and moved
compiler-1.impure.lisp to that category so that
it's compiled as it should be
William Harold Newman [Fri, 6 Apr 2001 18:08:11 +0000 (18:08 +0000)]
0.6.11.33:
bug fixes from cmucl-imp..
..duplicate keys in macro arg lists, Pierre Mai, 2001-03-30
..SXHASH of strings with fill pointers, Tim Moore, 2001-03-29
(was actually already fixed in SBCL, but I added a
test case or two to make sure that it stays fixed)
added/enabled regression tests for SXHASH and PSXHASH
Screwed-up lambda list syntax isn't a continuable error.
William Harold Newman [Fri, 6 Apr 2001 13:02:09 +0000 (13:02 +0000)]
0.6.11.32:
MNA patches from sbcl-devel..
..package cleanup (2001-03-29)
..bug 94 compiler type mismatch nondetection fix (2001-03-30)
(started trying to fix bug 87, following MNA 2001-04-04 patch,
but fix doesn't work yet)
renamed TYPES-INTERSECT and VALUES-TYPES-INTERSECT to
TYPES-EQUAL-OR-INTERSECT and
VALUES-TYPES-EQUAL-OR-INTERSECT
removed redundant quotes from self-evaluating constants
build-under-CMU-CL fixes from Christophe sbcl-devel 2001-04-05
added CMU CL 18c workaround in cross-float.lisp
William Harold Newman [Wed, 4 Apr 2001 19:47:19 +0000 (19:47 +0000)]
0.6.11.31:
Pierre Mai READ-SEQUENCE/CONCATENATED-STREAM fix (cmucl-imp
2001-03-26)
removed broken ".~D" file versioning
added PRINT-OBJECT for LOGICAL-HOST so that LOGICAL-PATHNAMEs
would print more nicely
added ~_ in PRINT-OBJECT (LOGICAL-PATHNAME T)
removed unused &OPTIONAL INDENT from DEFPRINTER-FOO functions
removed redundant/ugly indentation in DEFPRINTER expansion
William Harold Newman [Thu, 29 Mar 2001 01:50:35 +0000 (01:50 +0000)]
0.6.11.30:
MNA source cleanups (sbcl-devel 2001-03-29), including Raymond
Toy's float trap patch (cmucl-imp 2001-03-28)
commented out unused stuff in dfun.lisp
William Harold Newman [Thu, 29 Mar 2001 00:17:20 +0000 (00:17 +0000)]
0.6.11.29:
fixed bug 92: (< SB-EXT:SINGLE-FLOAT-POSITIVE-INFINITY 100)
now works, thanks to Raymond Toy's patch to CMU CL.
added more infinity-related tests in tests/
deleted bad DECLAIM FTYPE for PARSE-DEFTRANSFORM, to fix bug
reported by Christophe Rhodes sbcl-devel 2001-03-28
converted some INTERNs to SYMBOLICATE or KEYWORDICATE
converted DEF-DEBUG-COMMAND to !DEF-DEBUG-COMMAND as per FIXME
William Harold Newman [Tue, 27 Mar 2001 21:27:50 +0000 (21:27 +0000)]
0.6.11.28:
removed FIXME "MAKE-LOAD-FORM is supposed to be handled
here, too", since it's of date; in SBCL DEF!STRUCT
handles MAKE-LOAD-FORM by a different mechanism, and
vanilla DEFSTRUCT doesn't mess with it at all. (thanks
to MNA question)
tweaked FIND-FREE-REALLY-FUNCTION so that it doesn't complain
about an undefined function when the function is
defined in the running Lisp (to fix undefined function
warnings reported by MNA sbcl-devel 2001-03-26)
moved CONSTANTLY-FOO functions to SB!INT (as per old FIXME)
William Harold Newman [Tue, 27 Mar 2001 19:50:40 +0000 (19:50 +0000)]
0.6.11.27:
redid BYTE-BLT to use memmove(3) (so GENESIS will work again!)
deleted unused %SP-BYTE-BLT, and some unused stuff in SB!SYS
added error checking in DEF!STRUCT to catch any programmer
errors of (DEFSTRUCT FOO ..) followed by
(DEF!STRUCT (.. (:INCLUDE FOO)) ..)
MNA MAKE-LOAD-FORM UNIX-HOST bug and patch (fixing a
particular instance of the DEF!STRUCT programmer error)
William Harold Newman [Mon, 26 Mar 2001 20:55:55 +0000 (20:55 +0000)]
0.6.11.26:
restored CMU CL's :PROPAGATE-FLOAT-TYPE and :PROPAGATE-FUN-TYPE
functionality
changed design of CROSS-FLOAT-INFINITY-KLUDGE
renamed :PROPAGATE-FLOAT-TYPE and :PROPAGATE-FUN-TYPE
features to :SB-PROPAGATE-FLOAT-TYPE and
:SB-PROPAGATE-FUN-TYPE
renamed :CONSTRAIN-FLOAT-TYPE to :SB-CONSTRAIN-FLOAT-TYPE too
clarified TYPE-ERROR :TYPE in DEFUN SANE-PACKAGE
tweaked DEFCONSTANT handling so that CMU CL can handle the
DEFCONSTANT/DEFTYPE interaction in bit-bash.lisp
created SB-XC versions of various CL constants (e.g.
LEAST-POSITIVE-SINGLE-FLOAT) so that cross-compilation
of their DEFCONSTANT forms works more cleanly
reviewed uses of various CL constants and added SB!XC:
prefixes in compiled-on-host code: CALL-ARGUMENTS-LIMIT,
CHAR-CODE-LIMIT
split char.lisp into char.lisp and target-char.lisp so that
there'd be a suitable place to define CHAR-CODE-LIMIT
rewrote SOURCE-TRANSFORM-CXR stuff to avoid FORMAT call at
cold init time
suppressed bogus default DEFSTRUCT-generated COPY-READTABLE
hunting PROPAGATE-FOO-TYPE bug in compilation of OUTPUT-GSPACE..
..made MAKE-NUMERIC-TYPE enforce closed-bound invariant for
NUMERIC-TYPE-CLASS='INTEGER
..made MODIFIED-NUMERIC-TYPE to enforce NUMERIC-TYPE-CLASS
invariants; made slots read-only; rewrote
modify-NUMERIC-TYPE code to use MODIFIED-NUMERIC-TYPE
..moved FLET VALIDATE logic from CONSTRAIN-INTEGER-TYPE
and CONSTRAIN-FLOAT-TYPE to MAKE-NUMERIC-TYPE.
..Now COPY-NUMERIC-TYPE can go away.
..renamed BOUND-VALUE to SB!INT:TYPE-BOUND-NUMBER
factored out CTYPE-OF-NUMBER code
cross-type.lisp is type system code, so do it in SB!KERNEL.
William Harold Newman [Thu, 22 Mar 2001 18:04:23 +0000 (18:04 +0000)]
0.6.11.25:
fixed bug 45a: restored COMPLEX-ACOS and friends from
sbcl-2.4.8's #-OLD-SPECFUN code, so that e.g.
(ACOS (COMPLEX 1 1)) works now
William Harold Newman [Thu, 22 Mar 2001 13:51:36 +0000 (13:51 +0000)]
0.6.11.24:
restored floating point infinity support..
..Grep for SB-INFINITIES matches in SBCL and fix them.
..Remove DEFMACRO INFINITE and its calls.
..added a few floating point infinity test cases
William Harold Newman [Thu, 22 Mar 2001 01:03:34 +0000 (01:03 +0000)]
0.6.11.23:
converted UNION :SIMPLE-SUBTYPEP method to use EVERY/TYPE
made tests/type.*-xc.lisp dependent on :SB-TEST
replaced ASSERT calls with AVER
William Harold Newman [Wed, 21 Mar 2001 21:47:36 +0000 (21:47 +0000)]
0.6.11.22:
made REAL and FLOAT types be represented by UNION-TYPEs, to
fix the bug discussed on cmucl-imp as "bug in type
handling" ca. 2001-02-12
made COMPLEX type translator able to deal with new REAL and
FLOAT representations
enabled old commented-out type tests now that they can work
William Harold Newman [Wed, 21 Mar 2001 12:29:12 +0000 (12:29 +0000)]
0.6.11.21:
fixed T-to-AT typo in DEFUN LISTEN
made CROSS-TYPEP understand SATISFIES types
rewrote the TYPE-UNION method :SIMPLE-INTERSECTION2 (and
:COMPLEX-INTERSECTION2) to return UNION-TYPE when
that's simpler than the result of just punting; now
bug 89 is fixed, (AND KEYWORD (OR KEYWORD NULL)=KEYWORD
William Harold Newman [Wed, 21 Mar 2001 01:12:20 +0000 (01:12 +0000)]
0.6.11.20:
preparing to attack bug 89..
..allowed TYPE-INTERSECTION2, and its type methods, to return
INTERSECTION-TYPE values
..rewrote SIMPLIFIED-COMPOUND-TYPES so that it handles
COMPOUND-TYPE values returned from its SIMPLIFY2
William Harold Newman [Tue, 20 Mar 2001 16:49:07 +0000 (16:49 +0000)]
0.6.11.19:
fixed bug 88: made CROSS-TYPEP support KEYWORD, so that
(SUBTYPEP '(MEMBER :FOO) 'KEYWORD)=>T,T
made CROSS-TYPEP use EVERY/TYPE and ANY/TYPE for AND and
OR types
William Harold Newman [Tue, 20 Mar 2001 14:59:26 +0000 (14:59 +0000)]
0.6.11.18:
miscellaneous cleanups..
..made COPY-TYPE-CLASS-COLDLY use *TYPE-CLASS-FUNCTION-SLOTS*
to reduce duplication of information
..renamed all the DEFCONSTANT FOO definitions in reader.lisp
to DEFCONSTANT +CHAR-ATTR-FOO+ style
..renamed IN-BUFFER-LENGTH to +IN-BUFFER-LENGTH+ too
..and IN-BUFFER-EXTRA to +IN-BUFFER-EXTRA+
..moved byte-interp adjacent to other byte-foo in build order
..removed EVAL-WHEN around DEFTYPEs in bit-bash.lisp
..fixed DECLAIM of *FASL-FILE*, as per kon@iki.fi cmucl-help
2001-03-19 bug report
William Harold Newman [Mon, 19 Mar 2001 23:10:58 +0000 (23:10 +0000)]
0.6.11.17:
(like the changes in 0.6.11.13, except for -UNION this time
instead of -INTERSECTION)
converted SIMPLE-/COMPLEX- -UNION to -UNION2
The old UNION :COMPLEX-UNION and UNION :SIMPLE-UNION methods
go away in favor of TYPE-UNION and TYPE-UNION2 logic
renamed old 2-arg TYPE-UNION to TYPE-UNION2, and revised
it to be more like TYPE-INTERSECTION2
defined new &REST-arg TYPE-UNION similar to the &REST-arg
TYPE-INTERSECTION defined in 0.6.11.13
made some old TYPE-UNION calls use &REST-arg generality
MAKE-UNION-TYPE-OR-SOMETHING goes away in favor of new
TYPE-UNION.
VANILLA-UNION becomes HIERARCHICAL-UNION2.
removed support for pre-ANSI SATISFIES types in CTYPEP
William Harold Newman [Fri, 16 Mar 2001 22:40:29 +0000 (22:40 +0000)]
0.6.11.16:
Bug 86 isn't a bug. INTERSECTION is an unknown type. And
(specifier-type '(and (or number vector) real)) is ok.
moved FIXNUMP, BIGNUMP, and RATIOP to SB!INT
deleted TARGET-FIXNUMP in favor of SB-INT:FIXNUMP
added a xc definition of SB-INT:FIXNUMP in cross-type.lisp
fixed doc string for SB-INT:FIXNUMP
tweaked type predicate wrappers to facilitate lexical search
William Harold Newman [Fri, 16 Mar 2001 20:52:40 +0000 (20:52 +0000)]
0.6.11.15:
some cleanups related to the type hackathon in 0.6.11.13..
..restored :TYPE declaration for FORMAT slot in NUMERIC-TYPE
..restored :TYPE declaration for TYPES slot in COMPOUND-TYPE
..moved LIST, CONS, and NULL to a more logical point in
*BUILT-IN-CLASSES*
..rearranged CTYPE, ANY/TYPE, and EVERY/TYPE to share code
..added tests related to CTYPE of COMPOUND-TYPE
..redid INTERSECTION :SIMPLE-SUBTYPEP to share EVERY/TYPE too
added tests for ANY/TYPE and EVERY/TYPE, fixed EVERY/TYPE
moved SWAPPED-ARGS-FUN earlier to facilitate inlining, putting
it in SB!INT so it can go in early-extensions.lisp
deleted unused LETF and LETF*
William Harold Newman [Wed, 14 Mar 2001 13:46:52 +0000 (13:46 +0000)]
0.6.11.14:
response to MNA message sbcl-devel 2001-03-07..
..merged test case to exercise byte compiler bug
..merged patch to fix the bug
also merged Dan Barlow's explanation (sbcl-devel 2001-03-12)
into a comment in the source
William Harold Newman [Mon, 12 Mar 2001 13:47:43 +0000 (13:47 +0000)]
0.6.11.13:
converted SIMPLE-/COMPLEX- -INTERSECTION to -INTERSECTION2
reviewed old SIMPLE-/COMPLEX- -INTERSECTION methods to make
them OK in the new world
made TYPE-INTERSECTION2 and TYPE-APPROX-INTERSECTION2 for
some things which used to call TYPE-INTERSECTION
made new with-&REST-list TYPE-INTERSECTION to replace
MAKE-INTERSECTION-TYPE-OR-SOMETHING
HIERARCHICAL-INTERSECTION is more descriptive than
VANILLA-INTERSECTION.
deleted unused SIMPLIFY2 stuff
deleted unused TYPE-INTERSECT and VALUES-TYPE-ALLOWP
Pure BOOLEAN TYPE-ENUMERABLE is too much trouble; relax to
generalized boolean instead.
tweaked DEFUN-CACHED so that it will work early in cold init
(if some clueless bozo happens to mess up types so
badly that it gets called early in cold init:-)
Now the AND type translator can just use TYPE-INTERSECTION.
added *SOFT-HEAP-LIMIT*
removed some unused symbols in SB!UNIX
tweaked /SHOW0 to automatically stringify and concatenate its
arguments, to abbreviate an idiom that I use a lot
wrote /HEXSTR0 to abbreviate another idiom
defined /PRIMITIVE-PRINT to abbreviate another idiom
DO-LOAD-TIME-CODE-FIXUP and DO-LOAD-TIME-VALUE-FIXUP are
only needed at cold init.
added tests/type.after-xc.lisp and tests/type.before-xc.lisp
upgraded SB-XC:TYPEP to handle (TYPEP #(1 2 3) 'VECTOR) etc.
Intersection with *EMPTY-TYPE* and *UNIVERSAL-TYPE* should
behave (and be commutative).
Guard against some type method operations on *. (easier than
worrying about them, or trying to make them right..)
moved STYLE-WARN and friends to SB-INT to eliminate thought
about whether they're visible e.g. in ir1tran.lisp
tripped over bug 84
made SB!C-CALL use SB!INT and SB!EXT so I can use /SHOW0 et al.
DEF-PRIMITIVE-TYPE stuff can become !DEF-PRIMITIVE-TYPE.
DEF-BOUNDED-TYPE and DEFINE-FLOAT-FORMAT can become
!DEF-BOUNDED-TYPE and !DEFINE-FLOAT-FORMAT.
moved DEFTYPE FLOAT-FORMAT earlier so that it's visible in
early-type.lisp (for NUMERIC-TYPE FORMAT slot :TYPE)
moved LIST, CONS, and NULL classes earlier in *BUILT-IN-CLASSES*
to try to help with RATIO cold init problem
removed pre-ANSI keyword-only stuff in ARG-INFO-KEYWORD and
in various &KEY-related error messages and symbol names
bug 12 fixed: (SUBTYPEP 'KEYWORD 'SYMBOL)=>T,T
added INTERSECTION-TYPE support to CTYPEP
William Harold Newman [Mon, 5 Mar 2001 01:30:08 +0000 (01:30 +0000)]
0.6.11.12:
moved !PACKAGE-COLD-INIT earlier in cold init so that KEYWORDP
will work earlier
tidied up UNIX-SIGNAL machinery: no more DEFCONSTANTs,
no more re-interning, no more descriptions
tweaked RUN-PROGRAM to use UNIX-SIGNAL-NUMBER function instead
of old bare SIGFOO constants
It's easier and nicer just to coerce all RUN-PROGRAM args
than to require that they be SIMPLE-STRINGs.
made test/*.test.sh use local sbcl, not installed sbcl, just
as other tests do
William Harold Newman [Sat, 3 Mar 2001 18:25:29 +0000 (18:25 +0000)]
0.6.11.11:
reduced code duplication code in FDEFINITION-OBJECT
defined *XTYPE?* to help support ongoing type experiments
started conditionally fully enabling INTERSECTION-TYPE
allowed HAIRY-TYPE elements in INTERSECTION-TYPEs after all,
since otherwise INTERSECTION-TYPE can't fix bug 12
redefined KEYWORD type as (AND SYMBOL (SATISFIES KEYWORDP))
added tests for bug 12 fixedness
William Harold Newman [Fri, 2 Mar 2001 23:48:33 +0000 (23:48 +0000)]
0.6.11.10:
deleted trivial PRINT-OBJECT (FIXUP T) method
tweaked %%TYPEP to start on bug 10, but failed to fix up
type translators (so bug still exists)
Reject non-SYMBOL-valued predicates in SATISFIES types.
Reject bogus NOT type specifiers too. (modulo bug 10 anyway)
ANSI says TYPEP can return generalized boolean, not just
pure BOOLEAN, and having DEFKNOWN TYPEP declare it
narrower than that was unnecessarily fragile.
made %INCF-PCOUNTER-OR-FIXNUM non-inline, at least until it's
less buggy
INCF-PCOUNTER shouldn't be inline either.
PCOUNTER-FIXNUM slot should be unsigned.
added :COPIER NIL to lots of DEFSTRUCTs (just to make target
system a little smaller)
William Harold Newman [Fri, 2 Mar 2001 17:36:19 +0000 (17:36 +0000)]
0.6.11.9:
enabled some INTERSECTION-TYPE stuff
made INTERSECTION-TYPE and UNION-TYPE share a parent
This changes the layout of UNION-TYPE: new fasl version again..
defined placeholder SOURCE-TRANSFORM-INTERSECTION-TYPEP
William Harold Newman [Thu, 1 Mar 2001 17:29:17 +0000 (17:29 +0000)]
0.6.11.8:
deleted obsolete CONS-UNIQUE-TAG
That changes byte codes, so increment fasl file version.
William Harold Newman [Thu, 1 Mar 2001 16:00:12 +0000 (16:00 +0000)]
0.6.11.7:
hacking on 0.6.11.5 patches, continued..
..MAKE-HASH-TABLE :TEST #'EQUAL uses 'EQUAL instead.
..tested that PRINT-PRETTY-ON-STREAM? stuff doesn't get
confused on Gray streams
..added test case for new byte compiler bug
..removed extra DUP in patched GENERATE-BYTE-CODE-FOR-SET
new DEFCONSTANT cleanups, as per AL, sbcl-devel, 2001-02-27
Wrapping DEFCONSTANT +EMPTY-HT-SLOT+ in EVAL-WHEN seems to
cause problems, so instead I moved the +EMPTY-HT-SLOT+
definition 'way early in stems-and-flags.lisp-expr.
William Harold Newman [Wed, 28 Feb 2001 14:04:20 +0000 (14:04 +0000)]
0.6.11.6:
hacking on 0.6.11.5 patches..
..Various FIXNUM things in ASH DERIVE-TYPE need to be
TARGET-FIXNUM things.
..tweaked ASH DERIVE-TYPE to make it fit in 80 columns
..defined BECOME-DEFINED-FUNCTION-NAME and
PRINT-PRETTY-ON-STREAM? to reduce cut/paste
..deleted unused %FUNCTION-HEADER-ARGLIST-SLOT and
%FUNCTION-HEADER-NAME-SLOT
William Harold Newman [Tue, 27 Feb 2001 22:09:13 +0000 (22:09 +0000)]
0.6.11.5:
MNA hack to DEFCONSTANT-EQX to support cmucl-2.4.19 as xc host
patches ported by MNA from CMU CL..
..no need to export the unused symbols %FUNCTION-HEADER-TYPE,
%FUNCTION-HEADER-ARGLIST, and %FUNCTION-HEADER-NAME
..Upon a stack trace ambiguity in X86-CALL-CONTEXT, choose the
Lisp frame in preference to the C frame as this is the
frame of interest.
..Limit the stack trace failure warning in X86-CALL-CONTEXT to
fails for the immediate frame rather then failures
deeper within the search.
..Catch missing slots in CONDITION-READER-FUNCTION, and signal
an error.
..PPRINT-NEWLINE, PPRINT-INDENT, and PPRINT-TAB should do
nothing when *PRINT-PRETTY* is false.
..Correct the pretty printing by PRINT-UNREADABLE-OBJECT. Only
attempt to print pretty when the stream is a
PRETTY-STREAM (and when *PRINT-PRETTY*) to ensure that
all output goes to the same stream.
..Have PRINT-UNREADABLE-OBJECT respect *PRINT-PRETTY*.
..Fix OUTPUT-CHARACTER to escape the CHAR-NAME. Rework
QUOTE-STRING to not write the delimiting quotes so
that it can be used by OUTPUT-CHARACTER.
..Within GENERATE-BYTE-CODE-FOR-SET, avoid trying to set a
lexical variable with no refs since the compiler
deletes such variables.
..Within SET-LEAF-VALUE-LAMBDA-VAR, avoid trying to set a
lexical variable with no refs since the compiler
deletes such variables.
..Rework the ASH DERIVE-TYPE optimizer to better handle large
negative bounds. Based on suggestions from Raymond Toy.
..Add a defensive declaration to PARSE-SPECIALIZERS.
..Add a defensive declaration to PARSE-DEFMETHOD, as well.
..Silence compiler warnings about undefined function
<hairy-make-instance-name> when compiling a method
containing a MAKE-INSTANCE call.
..Correct the printing of instance objects for which the
length was being incorrectly calculated.
William Harold Newman [Tue, 27 Feb 2001 18:03:28 +0000 (18:03 +0000)]
(belated 0.6.11.2 checkin notes):
Add missing EVAL-WHEN wrappers for DEFCONSTANT, as per bug
report from Arthur Lemmens sbcl-devel 2001-02-23..
..the use of #.MAX-VOP-TN-REFS in vmdef.lisp
..(not in the sequence of DEFENUMs in early-objdef.lisp, at
least not yet, since it caused problems when I did)
Wrap DEFUN SYMBOLICATE in EVAL-WHEN so that we don't need
to worry about using it in DEFENUM.
0.6.11.4:
some foreshadowing for reenabling :PROPAGATE-FLOAT-TYPE
and :PROPAGATE-FUN-TYPE features
CONCAT-PNAMES goes away in favor of SYMBOLICATE.
William Harold Newman [Tue, 27 Feb 2001 13:40:55 +0000 (13:40 +0000)]
0.6.11.3:
Wrap the DEFENUMs in early-objdef.lisp in EVAL-WHEN, to
finish 0.6.11.2/Lemmens bug fix. (This way works,
but EVAL-WHEN inside the DEFENUM macro didn't.)
William Harold Newman [Mon, 26 Feb 2001 19:33:09 +0000 (19:33 +0000)]
0.6.11.1:
Address the out-of-order use of SAP-INT-TYPE pointed out by
Arthur Lemmens on sbcl-devel 2001-02-21..
..Split target-allocate.lisp out of target-sap.lisp.
..target-sap.lisp moves late in stems-and-flags.lisp-expr, while
target-allocate.lisp stays in the same place.
William Harold Newman [Sun, 25 Feb 2001 18:20:45 +0000 (18:20 +0000)]
0.6.11:
tag sbcl_0_6_11
William Harold Newman [Fri, 23 Feb 2001 17:54:34 +0000 (17:54 +0000)]
0.6.10.23:
hacking MNA "pcl cleanups" megapatch, phase IV..
The SB-PCL package now USE-PACKAGEs SB-INT and SB-EXT.
SB-INT no longer exports unused E.
SB-INT:ITERATE is now called SB-INT:NAMED-LET, to avoid
collision with SB-PCL::ITERATE.
The SB-ITERATE and SB-WALKER packages now use SB-INT
and SB-EXT too.
Now lotso SB-INT: prefixes in src/pcl/*.lisp can go away.
William Harold Newman [Fri, 23 Feb 2001 12:40:08 +0000 (12:40 +0000)]
0.6.10.22:
Sometimes anonymous function names are STRINGs. Make
%DESCRIBE-DOC and %DESCRIBE-FUNCTION-NAME handle that.
hacking MNA "pcl cleanups" megapatch, phase III..
(EVAL-WHEN (:LOAD-TOPLEVEL :EXECUTE) ..) becomes (PROGN ..)
or is elided completely.
William Harold Newman [Thu, 22 Feb 2001 20:48:03 +0000 (20:48 +0000)]
0.6.10.21:
turned 'intern.*concatenate' grep matches to SYMBOLICATE
T is a self-evaluating constant and doesn't need to be quoted.
So is NIL.
hacking MNA "pcl cleanups" megapatch, phase II..
NAME-GET-FDEFINITION and NAME-SET-FDEFINITION become
FDEFINITION and (SETF FDEFINITION).
William Harold Newman [Thu, 22 Feb 2001 17:27:20 +0000 (17:27 +0000)]
0.6.10.20:
hacking MNA "pcl cleanups" megapatch, phase I..
SB-PCL::%INSTANCE-REF and SB-PCL::INSTANCE-REF become
SB-PCL::CLOS-SLOTS-REF, an inline function.
DEF-CONSTANTLY-FUN doesn't want FDEFINITION after all.
William Harold Newman [Thu, 22 Feb 2001 13:44:56 +0000 (13:44 +0000)]
0.6.10.19:
MNA pointed out that bug #25 is gone.
applied MNA "pcl cleanups" megapatch from sbcl-devel 2001-02-19
(will be hacked on some more soon, as per my reply and
ensuing discussion)
William Harold Newman [Sun, 18 Feb 2001 17:20:21 +0000 (17:20 +0000)]
0.6.10.18:
MNA bugfixes from sbcl-devel 2001-02-14..
fixed BUG 34 (concerning byte-comp.lisp)
fixed error reporting accessing a missing PCL slot
increased *COMPILER-ERROR-FOO* defaults (by even more than MNA
proposed)
William Harold Newman [Wed, 14 Feb 2001 14:20:31 +0000 (14:20 +0000)]
0.6.10.17:
removed LIST-FIRST arg from ANY/TYPE and EVERY/TYPE
replaced MAKE-UNION-TYPE with MAKE-UNION-TYPE-OR-SOMETHING
increased *COMPILER-ERROR-PRINT-FOO* values at MNA's suggestion
MNA pointed out that bug 80 was fixed by his earlier patch.
William Harold Newman [Tue, 13 Feb 2001 18:12:30 +0000 (18:12 +0000)]
0.6.10.16:
simplified ANY-TYPE-OP and EVERY-TYPE-OP, and renamed them to
ANY/TYPE and EVERY/TYPE since the interface changed
PUNT-TYPE-METHOD now local and used only once, renamed to PUNT
William Harold Newman [Sun, 11 Feb 2001 23:11:36 +0000 (23:11 +0000)]
0.6.10.15:
MNA patches from sbcl-devel e-mail 2001-02-10: Make %DEFUN
update INFO better, and (originally due to Paolo
Amoroso on cmucl-imp) clean up debugger restarts.
also degraded FTYPE mismatch to a STYLE-WARNING instead of
a full WARNING
William Harold Newman [Sat, 10 Feb 2001 22:17:17 +0000 (22:17 +0000)]
0.6.10.14:
fixing bug 40..
UPGRADED-ARRAY-ELEMENT-TYPE now signals an error when the type
is undefined.
TYPEP and SUBTYPEP now catch type arguments which are arrays
with undefined element types.
also cleaned up UPGRADED-COMPLEX-PART-TYPE a little..
UPGRADED-COMPLEX-PART-TYPE signals an error for undefined type.
UPGRADED-COMPLEX-PART-TYPE always returns a value which is a
subtype of REAL.
William Harold Newman [Thu, 8 Feb 2001 15:08:37 +0000 (15:08 +0000)]
0.6.10.13:
fixed problems reported by Douglas Brebner 2000-01-01..
make.sh and friends now use foo/bar instead of foo/bar/ to
specify directories, since foo/bar/ fails on NetBSD.
(SETF ALIEN:EXTERN-ALIEN) works (instead of dying with
undefined function SB-KERNEL:%SET-SAP-REF-32).
William Harold Newman [Wed, 7 Feb 2001 22:11:47 +0000 (22:11 +0000)]
0.6.10.12:
defined OBJECT-NOT-COMPLEX-VECTOR-ERROR
merged MNA patch for #'APPLY error-reporting
added MNA regression test for bug #39
William Harold Newman [Wed, 7 Feb 2001 20:04:34 +0000 (20:04 +0000)]
0.6.10.11:
fixed problem pointed out by Raymond Wiker: DEFCONSTANT-EQX
now EVALs EXPR only once per situation
tweaked some compiler output for better line breaking
William Harold Newman [Wed, 7 Feb 2001 03:35:08 +0000 (03:35 +0000)]
0.6.10.10:
defined methods for INTERSECTION-TYPE by analogy with
UNION-TYPE methods
added a few tests for the type system
William Harold Newman [Thu, 1 Feb 2001 18:42:08 +0000 (18:42 +0000)]
0.6.10.9:
moved BLOCK PUNT-TYPE-METHOD into !DEFINE-TYPE-METHOD
macroexpansion template
made AND types expand into INTERSECTION-TYPE unless they're
too hairy
William Harold Newman [Fri, 26 Jan 2001 23:27:52 +0000 (23:27 +0000)]
0.6.10.8:
made Dan Barlow's proposed changes #1 and #4 (from his
sbcl-devel message 2000-01-25)..
#1: x86-validate.h stuff is no longer machine-dependent, so it
is merged into validate.h. x86-validate.h goes away.
#4: runtime.c calls arch_init() again (as in original CMU CL)
William Harold Newman [Thu, 25 Jan 2001 18:31:50 +0000 (18:31 +0000)]
0.6.10.7:
merged MNA "minor patches" collection (sbcl-devel 2000-01-25):
logical pathname support for LOAD-1-FOREIGN, and
changes in regression tests (reflect 0.6.10.5 change,
new tests for Gray streams, and fix reader test)
fixed run-tests.sh to check all *.pure.lisp (not just one!)
William Harold Newman [Thu, 25 Jan 2001 14:42:00 +0000 (14:42 +0000)]
0.6.10.6:
defined INTERSECTION-TYPE by analogy with UNION-TYPE (aiming
to fix bug #12)
renamed PUNT blocknames in type methods to PUNT-TYPE-METHOD
renamed PUNT-IF-TOO-LONG to PUNT-PRINT-IF-TOO-LONG
William Harold Newman [Thu, 25 Jan 2001 03:08:56 +0000 (03:08 +0000)]
0.6.10.5:
MNA patch: Empty namestrings are now "" instead of NIL.
(sbcl-devel 2001-01-24)
William Harold Newman [Mon, 22 Jan 2001 03:39:16 +0000 (03:39 +0000)]
0.6.10.4:
merged MNA "Gray streams" (+ packaging tweaks too) patch from
sbcl-devel 2001-01-20
deleted old gray-stream-stubs stuff
boosted fasl file version (new stream class layout)
William Harold Newman [Sat, 20 Jan 2001 03:30:51 +0000 (03:30 +0000)]
0.6.10.3:
DESCRIBE now works on STRUCTURE-OBJECTs again.
William Harold Newman [Fri, 19 Jan 2001 18:02:07 +0000 (18:02 +0000)]
0.6.10.2:
merged MNA LOAD-FOREIGN patch (sbcl-devel 2001-01-10)
tweaked LOAD-FOREIGN and RUN-PROGRAM to use :ENVIRONMENT
instead of lossy :ENV by preference
renamed GET-UNIX-ENVIRONMENT to SB-EXT:POSIX-ENVIRON, by
analogy with SB-EXT:POSIX-GETENV
removed (MERGE-PATHNAMES PROGRAM "path:") in RUN-PROGRAM,
since it refers to a search list which isn't
defined in SBCL (and since I'd like to remove all
support for search lists anyway).
SB-EXT is a public interface package not really intended as a
home for implementations; switched IN-PACKAGEs
to use SB-IMPL instead
To support this, SB-IMPL should USE SB-C-CALL and SB-ALIEN
as SB-EXT does.
made RUN-PROGRAM default to copying Unix environment
added LOAD-FOREIGN and RUN-PROGRAM test cases
William Harold Newman [Wed, 17 Jan 2001 14:45:42 +0000 (14:45 +0000)]
0.6.10.1:
made revised STRING-FOO functions accept string designators
instead of just strings (thanks to MNA bug report)
William Harold Newman [Mon, 15 Jan 2001 02:28:27 +0000 (02:28 +0000)]
0.6.10:
got confused and tagged -- I think -- 0.6.9.23 as sbcl_0_6_10
redid distribution scripts so that .tar files expand into
sbcl-x.y.z/ instead of ./