sbcl Log


Commit Date  
[7c850c] (clos-typechecking2-branch) by Christophe Rhodes Christophe Rhodes

0.9.6.56.clos-typechecking2.2:
Propagate type information for funcallable-standard-class
instances in instance-writer too.
... fixes the fail/sv/gf case (compiled slot-value inside
a generic function)
... probably doesn't fix subclasses with hairy intersection
types...
... doesn't fix the compiled use of slot-value outside
methods, which uses the (SLOT-ACCESSOR :GLOBAL ...
WRITER) generic function.

2005-11-24 17:09:40 Tree
[bd706e] by Christophe Rhodes Christophe Rhodes

0.9.6.56.clos-typechecking2.1:
Forward-port clos-typechecking-branch to HEAD.
... no new fixes, but I think I have some ideas for how to move
forward:
* the optimized slot setters, and probably also the
non-optimized path through (setf slot-value-using-class),
should cache the type checking predicate, rather than calling
TYPEP or %%TYPEP every time
* (SAFETY 3) should probably be replaced with (SB-C::TYPE-CHECKS
3) or whatever it's called
* FUNCTION types...
* need tests of redefinition &
UPDATE-INSTANCE-FOR-REDEFINED-CLASS; subclasses with
intersection types.
* it might be possible to perform some limited inference in
method functions.

2005-11-24 14:30:31 Tree
[5771a5] by Christophe Rhodes Christophe Rhodes

file clos-typechecking.impure.lisp was added on branch clos-typechecking2-branch on 2005-11-24 14:30:31 +0000

2005-01-03 15:49:34 Tree
[3820ea] by Christophe Rhodes Christophe Rhodes

0.8.18.11:
Maybe fix the clisp build bug (reported by Pascal Bourguignon)

2005-01-02 23:02:35 Tree
[584a5c] by Christophe Rhodes Christophe Rhodes

0.8.18.10:
Fix FORMATTER.COND.7 and similar;
... ~V[ uses only one argument to determine which clause (unless
the argument is NIL, of course)

2005-01-01 21:34:34 Tree
[753dce] by Christophe Rhodes Christophe Rhodes

0.8.18.9:
Merge a couple of patches
... a bug in 'filesys.lisp' (Artem V. Andreev);
... (coerce #c(1 2) '(complex float)) => error (Vincent Arkesteijn)

2005-01-01 21:01:27 Tree
[cce467] by Christophe Rhodes Christophe Rhodes

0.8.18.8:
Make METHOD and FAST-METHOD generalized function names
... some adjustments in NAMED-LAMBDAs;
... no more INTERN-FUN-NAME, yay.

2004-12-31 15:53:50 Tree
[3d2929] by William Harold Newman William Harold Newman

0.8.18.7:
more BUGS
misc. fiddling/tidying while trying to relearn IRn guts...
...indentation systematization
...redid printers s/#<CTRAN #1 {97EC941}>/#<CTRAN 1 {97EC941}>/
so that they look less like *PRINT-CIRCLE* bugs

2004-12-31 13:39:08 Tree
[98e0c9] by Christophe Rhodes Christophe Rhodes

0.8.18.5:
Make comments in ctor.lisp reflect the two recent fixes.

2004-12-31 12:30:11 Tree
[683e2b] by Christophe Rhodes Christophe Rhodes

0.8.18.4:
Pass defaulted initargs, not just present initargs, to methods
on SHARED-INITIALIZE and INITIALIZE-INSTANCE in ctor.lisp
... test, both for constant and variable initforms;
... I wish I could remember who reported this bug where.

2004-12-31 11:50:54 Tree
[7c5138] by Christophe Rhodes Christophe Rhodes

0.8.18.3:
Fix for invalid :default-initargs not being caught by ctor
... disable ctor if any default-initarg keys are invalid

2004-12-31 08:25:04 Tree
[2f97d8] by William Harold Newman William Harold Newman

0.8.18.2:
logged bugs 367a and 367b

2004-12-30 18:55:58 Tree
[368a2e] by Brian Mastenbrook Brian Mastenbrook

0.8.18.1: obviously I wasn't thinking this morning. It's not 0.8.19 /yet/.

2004-12-30 14:18:39 Tree
[b36697] by Brian Mastenbrook Brian Mastenbrook

0.8.19.1: PowerPC linkage tables (darwin only for now)
* Linkage tables are now implemented on PowerPC.
Right now I've only enabled it on Darwin; to enable it on Linux,
the addresses used for linkage table spaces would need to be tested.
* Lazy foreign functions do not work correctly - the code expects the signal
that occurs when jumping to a write-protected page to be the same as the
signal that occurs when reading from one. On Darwin the former is SIGBUS
and the latter is SIGSEGV; this means that the fault address testing does
not work on Darwin.

2004-12-30 13:44:26 Tree
[4ae836] (sbcl_0_8_18) by William Harold Newman William Harold Newman

0.8.18:
release, tagged as sbcl_0_8_18

2004-12-28 23:38:00 Tree
[383a00] by Brian Mastenbrook Brian Mastenbrook

0.8.17.31: "dlerror and a tale of woe"
Fix two bugs relating to dlerror() in the OS X dl* shim:
* dlerror() should return NULL when there is no error
* dlerror() should return an error when dlsym() returns NULL
Also, fix a call to cerror with only one argument.

Restores support for OS X 10.2 "Jaguar":
* Use a header file with constants for the dl* shim instead of the
OS dlfcn.h, which doesn't exist under 10.2 and does under 10.3
* Include a fake nl_langinfo which returns a codeset of UTF-8 unless
neither LC_CALL or LANG are set to C. OS X uses UTF-8 everywhere, so
this is arguably the right behavior.

2004-12-20 13:10:38 Tree
[af141f] by Alexey Dejneka Alexey Dejneka

0.8.17.30:
* Merged patch for the bug 348 by Gabor Melis.

2004-12-19 09:59:18 Tree
[cb79d7] by Alexey Dejneka Alexey Dejneka

0.8.17.29:
* Merged sbcl-0-8-17-28-signed-modular branch.

2004-12-19 07:01:04 Tree
[40318b] by Nathan Froyd Nathan Froyd

0.8.17.28:
Oops. Undo modular fixnum arithmetic changes from 0.8.17.24.

2004-12-09 16:58:40 Tree
[1cae06] by Christophe Rhodes Christophe Rhodes

0.8.17.27:
Improve ENUM support a little. (VJA sbcl-devel 2004-12-09)
... SB-ALIEN enums not limited to symbols any more;
... SB-GROVEL enum grovelation;
... tests (which pass despite the current, erm, suboptimality
of arithmetic :)

2004-12-09 16:15:57 Tree
[93c815] by Christophe Rhodes Christophe Rhodes

0.8.17.26:
Fix from VJA (sbcl-devel 2004-12-09) for EXPORT
continue-format-control
... actually give it the arguments it needs.
... write a test that doesn't actually test for the bug, but makes
me feel better anyway.

2004-12-09 15:32:51 Tree
[f4450b] by Nikodemus Siivola Nikodemus Siivola

0.8.17.25: Missing export & documentation updates
* buglet reported by Raymond Wiker
* typo reported by Devon McCullough
* remove text saying LOAD-SHARED-OBJECT must be called
before loading the definitions

2004-12-09 11:40:58 Tree
[6a7ffd] by Nathan Froyd Nathan Froyd

0.8.17.24:
Reinstate fixnum arithmetic when possible by defining modular
arithmetic mechanisms for (UNSIGNED-BYTE 29)
... this feels like a big, ugly hack, since the compiler is
(presumably) smart enough to do this when modular
arithmetic was not present;
... move some EVAL-WHEN macros into a MACROLET while we're at it;
... builds and passes tests on x86/Linux; will probably build
properly on other platforms, but will fail tests in
(at least) tests/arith.pure. These test failures seem
harmless enough and will be fixed in another revision or
two.

2004-12-08 16:31:41 Tree
[f39bad] by Christophe Rhodes Christophe Rhodes

0.8.17.23:
Fix for printing 1.0d+23. Thinko in transcription, duly caught
by Raymond Toy (bug report cmucl-help 2004-12)

2004-12-07 15:16:46 Tree
[dfe613] by Nikodemus Siivola Nikodemus Siivola

0.8.17.22: minor buglets & bugreports
* Use %COERCE-NAME-TO-FUN, not FDEFINITION for
evaluation of FUNCTION.
* (SETF MACRO-DEFINITION) must accept NIL environments
(reported by Kalle Olavi Niemitalo)
* Also record a bunch of PCL/MOP bugs reported
by Bruno Haible. Note: there are still more bugs
reported by him on the mailing list.

2004-12-07 13:30:40 Tree
Older >