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