Work at SourceForge, help us to make it a better place! We have an immediate need for a Support Technician in our San Francisco or Denver office.

Close

sbcl Log


Commit Date  
[400988] by Nikodemus Siivola Nikodemus Siivola

0.8.12.23: Signs of mellowing out in SBCL? Reversed NEWS file
to lastest-first order, and moved planned changes bit
to TODO.

2004-07-05 16:27:47 Tree
[fdc38a] by Nikodemus Siivola Nikodemus Siivola

0.8.12.22: Minor changes to make SBCL build itself cleanly with
package locks enabled.

2004-07-05 02:30:42 Tree
[c2fd99] by Nikodemus Siivola Nikodemus Siivola

0.8.12.21: The same finnish twit who broke the manual also left
unaccounted several pathological cases in his COMPILER-ERROR
reorganization.
* Revert to a less fancy, but more robust scheme, one that
never requires MAKE-LOAD-FORM for dumping
COMPILED-PROGRAM-ERRORS. The informational benefits
of the earlier patch are unaffected.
* Tests for at least some of those pathological cases in case
someone comes up with the same brilliant idea.

2004-07-04 22:30:03 Tree
[f469ab] by Nikodemus Siivola Nikodemus Siivola

0.8.12.20: Seems that some finnish twit broke the manual building.
Fixed.

2004-07-04 15:52:15 Tree
[29f644] by Nikodemus Siivola Nikodemus Siivola

0.8.12.19: Restructure the manual chapter "Package Locks" into consepts
and a dictionary, which also incidentally fixes the broken
section numbering there.

2004-07-04 15:03:46 Tree
[fea8ea] by Nikodemus Siivola Nikodemus Siivola

0.8.12.18: Rearranging COMPILER-ERROR protocol
* Enhance the protocol to attach the original
condition and source to the PROGRAM-ERROR.
* Make compile-time package-lock-violations from
lexical constructs signal runtime PROGRAM-ERRORS,
and not drop into debugger during compilation while
still allowing users to handle them during compilation.
* While at it, make file compiler errors from undumpable
constants more informative.
* Plus a few cosmetic changes to package locking guts.

2004-07-04 13:03:27 Tree
[0332a8] by Nikodemus Siivola Nikodemus Siivola

0.8.12.17: Really. Arrgh. This is becoming a bad habit.

2004-07-04 02:22:18 Tree
[0eb427] by Nikodemus Siivola Nikodemus Siivola

0.8.12.17: Fix TRACE :BREAK T in the presence of package
locks and grap credit for package locks.

2004-07-04 02:12:44 Tree
[964e64] by Christophe Rhodes Christophe Rhodes

0.8.12.16:
Fix BUG #334
... do bookkeeping behind the user's back for
effective-slot-defitions generated by the user
... for :class slots, allocate a location and place it in the
class' class-slot-cells;
... for :class / :instance slots, set the slot-definition-class
slot to the new class;
... add minimal test for reasonable behaviour.

2004-07-02 08:14:01 Tree
[a14326] by Christophe Rhodes Christophe Rhodes

0.8.12.15:
Fix for (declare ignore) treatment in methods when there is use
of SETQ in the body
... wow, no-one noticed before me? Cool!

2004-07-01 11:41:22 Tree
[05b900] by William Harold Newman William Harold Newman

0.8.12.14:
bugfix: missing level of LISTness in :REFERENCES in
FIND-METHOD-LENGTH-MISMATCH
tidying: added contrib/sb-sprof/.cvsignore

2004-06-30 23:08:32 Tree
[e60e0a] by Christophe Rhodes Christophe Rhodes

0.8.12.13:
Fix for MAKE-LOAD-FORM-SAVING-SLOTS on conditions
... it's not required to work, which is why we hadn't noticed
that it was totally broken. However, since it can work
with minimal extra effort, might as well get it right.

2004-06-30 20:43:08 Tree
[6bb7ec] by Nikodemus Siivola Nikodemus Siivola

0.8.12.12: Restore buildability with :SB-FUTEX and :SB-PACKAGE-LOCKS.
Reported by Kevin Rosenberg on #lisp.

2004-06-29 18:54:02 Tree
[c0c27f] by Christophe Rhodes Christophe Rhodes

0.8.12.11:
Fix bug #340
... SETF of VALUES now works as per spec (CLHS 5.1.2.3)

2004-06-29 13:25:02 Tree
[76874d] by Nikodemus Siivola Nikodemus Siivola

0.8.12.10: Fix bug 338: "MOP specializers as type specifiers"
(reported by Bruno Haible sbcl-devel 2004-06-11)
... Adding a type translator in SHARED-INITIALIZE :AFTER
does the trick.
... Test case.

2004-06-29 12:42:52 Tree
[0d1030] by Christophe Rhodes Christophe Rhodes

0.8.12.9:
Indentation change to debug-dump.lisp
... resulting from complete failure to find where source info
is conditionally dumped on (debug 2).
Also log the (SETF VALUES) bug

2004-06-29 12:13:44 Tree
[dde834] by Nikodemus Siivola Nikodemus Siivola

0.8.12.8: Really this time. Note to self: remeber to save the
version.lisp-expr buffer before commiting.

2004-06-29 10:02:48 Tree
[0f861c] by Nikodemus Siivola Nikodemus Siivola

0.8.12.8: NEWS entry for package locks, plus add :sb-package-locks
to base-target-features.lisp-expr.

2004-06-29 09:53:13 Tree
[ff9259] by Nikodemus Siivola Nikodemus Siivola

0.8.12.7: Merge package locks, AKA "what can go wrong with a 3783 line patch?"
... Controlled by the presence of :sb-package-locks in target
features.
... This builds both with and without package locks on both
x86 Linux and SunOS Sparc, with both CMUCL and SBCL
as host -- so chances are it should build elsewhere as
well.
... Remaining TODO: turn package locking errors from lexical
constructs to program errors in the produced code, fix
the bits in SBCL that hit host's SBCL-tyle package locks
(relevant FIXME is in src/cold/shared.lisp).

2004-06-29 08:50:51 Tree
[8b64d5] by Christophe Rhodes Christophe Rhodes

0.8.12.6:
Give sb-sprof a chance of working on non-x86 non-gencgc.
... implement search_dynamic_space and friends in cheneygc
... share component_ptr_from_pc between the GCs, and define
an alien routine for it unconditionally
... (provide 'sb-sprof)

2004-06-28 16:27:29 Tree
[2fb9b0] by Andreas Fuchs Andreas Fuchs

0.8.12.5:
Extract the version number from version.lisp-expr for the manual

* doc/manual/extract-values.sh does that for us and fills them
into values.template.

2004-06-28 11:38:42 Tree
[048b0b] by Daniel Barlow Daniel Barlow

once more, with branching

2004-06-27 18:15:58 Tree
[4ebdc8] by Daniel Barlow Daniel Barlow

Youn are not expected to understand this. I don't

2004-06-26 17:48:22 Tree
[1ae37c] by Christophe Rhodes Christophe Rhodes

0.8.12.4:
MORE REFERENCES
... rearrange src/code/condition.lisp a little to allow slightly
more references to appear;
... add some references in various error-producing forms in PCL

2004-06-26 17:28:11 Tree
[0e43db] by Christophe Rhodes Christophe Rhodes

0.8.12.3:
Fix (get-macro-character #\Space)
... now returns NIL, not undefined-macro-char
... tests for the standard characters

2004-06-26 14:43:50 Tree
Older >