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 * outstanding embarrassments
33 ** :IGNORE-ERRORS-P cruft in stems-and-flags.lisp-expr. (It's
34 reasonable to support this as a crutch when initially
35 bootstrapping from balky xc hosts with their own
36 idiosyncratic ideas of what merits FAILURE-P, but it's
37 embarrassing to have to use it when bootstrapping
39 ** weird double-loading (first in GENESIS, then in warm init)
40 of src/assembly/target/*.lisp stuff, and the associated
41 weirdness of the half-baked state (compiler almost but
42 not quite ready for prime time..) of the system after
44 * fixups now feasible because of pre7 changes
45 ** ANSIfied DECLAIM INLINE stuff (deprecating MAYBE-INLINE)
46 * miscellaneous simple refactoring
48 ** renamed %PRIMITIVE to %VOP
49 * These days ANSI C has inline functions, so..
50 ** redid many cpp macros as inline functions:
51 HeaderValue, Pointerp, CEILING, ALIGNED_SIZE,
52 GET_FREE_POINTER, SET_FREE_POINTER,
53 GET_GC_TRIGGER, SET_GC_TRIGGER, GetBSP, SetBSP,
54 os_trunc_foo(), os_round_up_foo()
55 ** removed various avoid-evaluating-C-macro-arg-twice
57 * Either get rid of or at least rework the fdefinition/encapsulation
58 system so that (SYMBOL-FUNCTION 'FOO) is identically equal to
60 =======================================================================
63 * refactored in preparation for moving CLOS into cold init and merging
64 SB-PCL:FOO with CL:FOO (for FOO=CLASS, FOO=CLASS-OF, etc.)
65 ** systematized support for MOP (new regression tests, maybe
66 new SB-MOP package..) to try to make sure things don't
67 get mislaid in the upcoming CLOS restructuring
68 ** extracted type system from SB-KERNEL into new SB-TYPE
70 ** reimplemented GENERIC-FUNCTION as a primitive object (or
71 maybe made SB-MOP:FUNCALLABLE-STANDARD-OBJECT the
72 primitive object, and then let GENERIC-FUNCTIONs
73 inherit from that) instead of structures with
74 :ALTERNATE-METACLASS and funcallableness. Now
75 FUNCALLABLE-INSTANCE can go away. (And now the new
76 funcallable primitive objects need to go into
77 collections like *FUN-HEADER-WIDETAGS* where
78 FUNCALLABLE-INSTANCE objects used to be.)
79 ** reimplemented CONDITIONs as primitive objects instead of
80 structures with :ALTERNATE-METACLASS. Now (between
81 this and the change to GENERIC-FUNCTIONs)
82 DEFSTRUCT :ALTERNATE-METACLASS can go away.
83 ** (maybe) Now INSTANCE_POINTER_LOWTAG can become just
84 STRUCTURE_POINTER_LOWTAG, and the concept of
85 SB-KERNEL:INSTANCE (including INSTANCEP,
86 (SPECIFIER-TYPE 'INSTANCE), etc.) can go away.
87 * moved CLOS into cold init, in order to allow CLOS to be used in the
88 implementation of the core system (e.g. the type system and the
89 compiler) and in order to support merger of CL:CLASS with
91 * (maybe) eliminated warm init altogether in favor of cold init
92 * (maybe, especially if warm init can be eliminated) rationalized
93 the build process, fixing miscellaneous pre-0.5.0 stuff that's
94 transparently not the right thing
95 ** removed separate build directories, now just building in
96 place with .sbclcoldfasl extensions
97 * (maybe) more refactoring in preparation for merging SB-PCL:FOO
98 into CL:FOO: reimplemented type system OO dispatch
99 (!DEFINE-TYPE-METHOD, etc.) in terms of CLOS OO dispatch
100 * merged SB-PCL:FOO into CL:FOO (and similarly CLASS-OF, etc.)
101 * added some automatic tests for basic binary compatibility, in hopes
102 that it might be practical to maintain binary compatibility
103 between minor maintenance releases on the stable branch (but no
104 promises, sorry, since I've never tried to do this before, and
105 have no idea how much of a pain this'll be)
106 ========================================================================
107 for 1.0 (fixes of lower priority which I'd nonetheless be embarrassed
108 to leave unfixed in 1.0):
109 * all too many BUGS entries and FIXMEs
110 =======================================================================
111 other priorities, no particular time:
113 * bug fixes, especially really annoying bugs (ANSI or not) and any
114 ANSI bugs (i.e. not just bugs in extras like the debugger or
115 "declarations are assertions", but violations of the standard)
116 * better communication with the outside world (scratching WHN's
117 personal itch): I don't want socket-level stuff so much as I
118 want RPC-level or higher (CORBA?) interfaces and (possibly
119 through RPC or CORBA) GUI support
120 =======================================================================
121 important but out of scope (for WHN, anyway: Patches from other people
122 are still welcome!) until after 1.0:
124 * sadly deteriorated support for ANSI-style block compilation
125 (static linking of DEFUNs within a single file or
126 WITH-COMPILATION-UNIT)
127 * various GC issues (exuberant cut-and-paste coding,
128 possibly dangerously over-conservative handling
129 of neighbors of function objects, general GC efficiency)
130 * package issues other than SB!TYPE, SB!MOP, and dead exported
132 * Any systematic effort to fix compiler consistency checks is
133 out of scope. (However, it still might be possible to
134 determine that some or all of them are hopelessly stale
136 =======================================================================
137 other known issues with no particular target date:
139 bugs listed on the man page
141 hundreds of FIXME notes in the sources from WHN
143 various other unfinished business from CMU CL and before, marked with
144 "XX" or "XXX" or "###" or "***" or "???" or "pfw" or "@@@@" or "zzzzz"
145 or probably also other codes that I haven't noticed or have forgotten.
147 (Things marked as KLUDGE are in general things which are ugly or
148 confusing, but that, for whatever reason, may stay that way
150 =======================================================================
151 "There's nothing an agnostic can't do as long as he doesn't know
152 whether he believes in anything or not."
155 "God grant me serenity to accept the code I cannot change, courage to
156 change the code I can, and wisdom to know the difference."
159 "Accumulation of half-understood design decisions eventually chokes a
160 program as a water weed chokes a canal. By refactoring you can ensure
161 that your full understanding of how the program should be designed is
162 always reflected in the program. As a water weed quickly spreads its
163 tendrils, partially understood design decisions quickly spread their
164 effects throughout your program. No one or two or even ten individual
165 actions will be enough to eradicate the problem."
166 -- Martin Fowler, in _Refactoring: Improving the Design of Existing
169 "I wish I didn't know now what I didn't know then."