4 ** *.pure.lisp tests run with assertoid.lisp loaded; assertoid
5 is moved to its own package, for use in *.impure.lisp.
6 ** non-x86 ports now pass irrat.pure.lisp
7 ** sparc and ppc now pass bit-vector.impure-cload.lisp
8 * faster bootstrapping (both make.sh and slam.sh)
9 ** added mechanisms for automatically finding dead code, and
10 used them to remove dead code
11 ** moved stuff from warm init into cold init where possible
12 (so that slam.sh will run faster and also just because
13 ideally everything would be in cold init)
14 ** profiled and tweaked
15 * fixed (TRACE :REPORT PROFILE ...) interface to profiling
16 * more EVAL/EVAL-WHEN/%COMPILE/DEFUN/DEFSTRUCT cleanup:
17 ** made %COMPILE understand magicality of DEFUN FOO
18 w.r.t. e.g. preexisting inlineness of FOO
19 ** used %COMPILE where COMPILE-TOP-LEVEL used to be used
20 ** removed now-redundant COMPILE-TOP-LEVEL and
21 FUNCTIONAL-KIND=:TOP-LEVEL stuff from the compiler
22 ** (ideally, but perhaps too hard, given what I've discovered
23 about the godawful internals of function debug names):
24 made FUNCTION-NAME logic work on closures, so that
25 various public functions like CL:PACKAGEP which
26 are now implemented as closures (because
27 they're structure slot accessors) won't be so
29 * outstanding embarrassments
30 ** :IGNORE-ERRORS-P cruft in stems-and-flags.lisp-expr. (It's
31 reasonable to support this as a crutch when initially
32 bootstrapping from balky xc hosts with their own
33 idiosyncratic ideas of what merits FAILURE-P, but it's
34 embarrassing to have to use it when bootstrapping
36 ** weird double-loading (first in GENESIS, then in warm init)
37 of src/assembly/target/*.lisp stuff, and the associated
38 weirdness of the half-baked state (compiler almost but
39 not quite ready for prime time..) of the system after
41 * fixups now feasible because of pre7 changes
42 ** ANSIfied DECLAIM INLINE stuff (deprecating MAYBE-INLINE,
43 including e.g. on the man page)
44 * miscellaneous simple refactoring
46 ** renamed %PRIMITIVE to %VOP
47 * These days ANSI C has inline functions, so..
48 ** redid many cpp macros as inline functions:
49 HeaderValue, Pointerp, CEILING, ALIGNED_SIZE,
50 GET_FREE_POINTER, SET_FREE_POINTER,
51 GET_GC_TRIGGER, SET_GC_TRIGGER, GetBSP, SetBSP,
52 os_trunc_foo(), os_round_up_foo()
53 ** removed various avoid-evaluating-C-macro-arg-twice
55 * Either get rid of or at least rework the fdefinition/encapsulation
56 system so that (SYMBOL-FUNCTION 'FOO) is identically equal to
58 =======================================================================
61 * refactored in preparation for moving CLOS into cold init and merging
62 SB-PCL:FOO with CL:FOO (for FOO=CLASS, FOO=CLASS-OF, etc.)
63 ** systematized support for MOP (new regression tests, maybe
64 new SB-MOP package..) to try to make sure things don't
65 get mislaid in the upcoming CLOS restructuring
66 ** extracted type system from SB-KERNEL into new SB-TYPE
68 ** reimplemented GENERIC-FUNCTION as a primitive object (or
69 maybe made SB-MOP:FUNCALLABLE-STANDARD-OBJECT the
70 primitive object, and then let GENERIC-FUNCTIONs
71 inherit from that) instead of structures with
72 :ALTERNATE-METACLASS and funcallableness. Now
73 FUNCALLABLE-INSTANCE can go away. (And now the new
74 funcallable primitive objects need to go into
75 collections like *FUN-HEADER-WIDETAGS* where
76 FUNCALLABLE-INSTANCE objects used to be.)
77 ** reimplemented CONDITIONs as primitive objects instead of
78 structures with :ALTERNATE-METACLASS. Now (between
79 this and the change to GENERIC-FUNCTIONs)
80 DEFSTRUCT :ALTERNATE-METACLASS can go away.
81 ** (maybe) Now INSTANCE_POINTER_LOWTAG can become just
82 STRUCTURE_POINTER_LOWTAG, and the concept of
83 SB-KERNEL:INSTANCE (including INSTANCEP,
84 (SPECIFIER-TYPE 'INSTANCE), etc.) can go away.
85 * moved CLOS into cold init, in order to allow CLOS to be used in the
86 implementation of the core system (e.g. the type system and the
87 compiler) and in order to support merger of CL:CLASS with
89 * (maybe) eliminated warm init altogether in favor of cold init
90 * (maybe, especially if warm init can be eliminated) rationalized
91 the build process, fixing miscellaneous pre-0.5.0 stuff that's
92 transparently not the right thing
93 ** removed separate build directories, now just building in
94 place with .sbclcoldfasl extensions
95 * (maybe) more refactoring in preparation for merging SB-PCL:FOO
96 into CL:FOO: reimplemented type system OO dispatch
97 (!DEFINE-TYPE-METHOD, etc.) in terms of CLOS OO dispatch
98 * merged SB-PCL:FOO into CL:FOO (and similarly CLASS-OF, etc.)
99 * added some automatic tests for basic binary compatibility, in hopes
100 that it might be practical to maintain binary compatibility
101 between minor maintenance releases on the stable branch (but no
102 promises, sorry, since I've never tried to do this before, and
103 have no idea how much of a pain this'll be)
104 ========================================================================
105 for 1.0 (fixes of lower priority which I'd nonetheless be embarrassed
106 to leave unfixed in 1.0):
107 * all too many BUGS entries and FIXMEs
108 =======================================================================
109 other priorities, no particular time:
111 * bug fixes, especially really annoying bugs (ANSI or not) and any
112 ANSI bugs (i.e. not just bugs in extras like the debugger or
113 "declarations are assertions", but violations of the standard)
114 * better communication with the outside world (scratching WHN's
115 personal itch): I don't want socket-level stuff so much as I
116 want RPC-level or higher (CORBA?) interfaces and (possibly
117 through RPC or CORBA) GUI support
118 =======================================================================
119 important but out of scope (for WHN, anyway: Patches from other people
120 are still welcome!) until after 1.0:
122 * sadly deteriorated support for ANSI-style block compilation
123 (static linking of DEFUNs within a single file or
124 WITH-COMPILATION-UNIT)
125 * various GC issues (exuberant cut-and-paste coding,
126 possibly dangerously over-conservative handling
127 of neighbors of function objects, general GC efficiency)
128 * package issues other than SB!TYPE, SB!MOP, and dead exported
130 * Any systematic effort to fix compiler consistency checks is
131 out of scope. (However, it still might be possible to
132 determine that some or all of them are hopelessly stale
134 =======================================================================
135 other known issues with no particular target date:
137 bugs listed on the man page
139 hundreds of FIXME notes in the sources from WHN
141 various other unfinished business from CMU CL and before, marked with
142 "XX" or "XXX" or "###" or "***" or "???" or "pfw" or "@@@@" or "zzzzz"
143 or probably also other codes that I haven't noticed or have forgotten.
145 (Things marked as KLUDGE are in general things which are ugly or
146 confusing, but that, for whatever reason, may stay that way
148 =======================================================================
149 "There's nothing an agnostic can't do as long as he doesn't know
150 whether he believes in anything or not."
153 "God grant me serenity to accept the code I cannot change, courage to
154 change the code I can, and wisdom to know the difference."
157 "Accumulation of half-understood design decisions eventually chokes a
158 program as a water weed chokes a canal. By refactoring you can ensure
159 that your full understanding of how the program should be designed is
160 always reflected in the program. As a water weed quickly spreads its
161 tendrils, partially understood design decisions quickly spread their
162 effects throughout your program. No one or two or even ten individual
163 actions will be enough to eradicate the problem."
164 -- Martin Fowler, in _Refactoring: Improving the Design of Existing
167 "I wish I didn't know now what I didn't know then."