Home My Page Projects Code Snippets Project Openings SML/NJ
Summary Activity Forums Tracker Lists Tasks Docs Surveys News SCM Files

SCM Repository

[smlnj] Diff of /sml/trunk/HISTORY
ViewVC logotype

Diff of /sml/trunk/HISTORY

Parent Directory Parent Directory | Revision Log Revision Log | View Patch Patch

revision 632, Sat Apr 29 15:50:42 2000 UTC revision 733, Fri Nov 17 05:13:45 2000 UTC
# Line 11  Line 11 
11  Date:  Date:
12  Tag: <post-commit CVS tag>  Tag: <post-commit CVS tag>
13  Description:  Description:
14    
15    ----------------------------------------------------------------------
16    Name: Matthias Blume
17    Date: 2000/11/17 14:10:00 JST
18    Tag: blume-20001117-pickle-lib
19    Description:
20    
21    1. Eliminated comp-lib.cm
22    2. Made pickle-lib.cm
23    3. Eliminated all uses of intset.sml (from comp-lib.cm)
24    4. Replaced all uses of intmap.{sig,sml} (from comp-lib.cm) with
25       equivalent constructs from smlnj-lib.cm (INtHashTable).
26    5. Point 4. also goes for those uses of intmap.* in MLRISC.
27       Duplicated intmap modules thrown out.
28    6. Hunted down all duplicated SCC code and replaced it with
29       equivalent stuff (GraphSCCFn from smlnj-lib.cm).
30    7. Rewrote Feedback module.
31    8. Moved sortedlist.sml into viscomp-lib.cm.  Eventually it
32       should be thrown out and equivalent modules from smlnj-lib.cm
33       should be used (IntRedBlackSet, IntListSet, ...).
34    
35    Confirmed that compiler compiles to fixpoint.
36    
37    ----------------------------------------------------------------------
38    Name: Allen Leung
39    Date: 2000/11/10 18:00:00
40    Tag: leunga-20001110-new-x86-fp
41    
42    A new x86 floating point code generator has been added.
43    By default this is turned off.  To turn this on, do:
44    
45        CM.autoload "$smlnj/compiler.cm";
46        Compiler.Control.MLRISC.getFlag "x86-fast-fp" := true;
47    
48    Changes:
49    
50    1.  Changed FTAN to FPTAN so that the assembly output is correct.
51    2.  Changed the extension callback for FTANGENT to generate:
52    
53              fptan
54              fstp  %st(0)
55        instead of
56              fptan
57              fstpl ftempmem
58    
59    3.  Numerous assembly fixes for x86.
60    
61    5.  Cleaned up the machine code output module x86/x86MC.sml and added
62        support for a whole bunch of instructions and addressing modes:
63    
64          fadd/fsub/fsubr/fmul/fdiv/fdivr  %st, %st(n)
65          faddp/fsubp/fsubrp/fmulp/fdivp/fdivrp  %st, %st(n)
66          fadd/fsub/fsubr/fmul/fdiv/fdivr  %st(n), %st
67          fiadd/fisub/fisubr/fimul/fidiv/fidivr mem
68          fxch %st(n)
69          fld %st(n)
70          fst %st(n)
71          fst mem
72          fstp %st(n)
73          fucom %st(n)
74          fucomp %st(n)
75    
76        All these are now generated when the fast fp mode is turned on.
77    
78    6.  Removed the dedicated registers %st(0), ..., %st(7) from X86CpsRegs
79    
80    ----------------------------------------------------------------------
81    Name: Matthias Blume
82    Date: 2000/11/09 11:20:00 JST
83    Tag: blume-20001109-scc
84    Description:
85    
86    Eliminated some code duplication:
87    
88    1. Added "where" clause to GraphSCCFn in SML/NJ Library.
89       (Otherwise the functor is useless.)
90    2. Used GraphSCCFn where SCCUtilFun was used previously.
91    3. Got rid of SCCUtilFun (in comp-lib.cm).
92    
93    ----------------------------------------------------------------------
94    Name: Lal George
95    Date: 2000/11/06 09:02:21 EST 2000
96    Tag: Release_110_30
97    Description:
98    
99    - Version 110.30
100    ----------------------------------------------------------------------
101    Name: Matthias Blume
102    Date: 2000/11/04 14:45:00
103    Tag: blume-20001104-mlbuild
104    Description:
105    
106    - Made ml-build faster on startup.
107    - Documentation fixes.
108    
109    ----------------------------------------------------------------------
110    Name: Matthias Blume
111    Date: 2000/11/02 17:00:00 JST
112    Tag: blume-20001102-condcomp
113    Description:
114    
115    - Small tweaks to pickler -- new BOOTFILES!
116    - Version bumped to 110.29.2.
117    - Added conditional compilation facility to init.cmi (see comment there).
118    ----------------------------------------------------------------------
119    Name: Allen Leung
120    Date: 2000/10/23 19:31:00
121    Tag: leunga-20001023-demo-ra
122    
123    1. Minor RA changes that improves spilling on x86 (affects Moby and C-- only)
124    2. Test programs for the graph library updated
125    3. Some new MLRISC demo programs added
126    
127    ----------------------------------------------------------------------
128    Name: Matthias Blume
129    Date: 2000/08/31 22:15:00 JST
130    Tag: blume-20001017-errmsg
131    Description:
132    
133    More error message grief: Where there used to be no messages, there
134    now were some that had bogus error regions.  Fixed.
135    
136    ----------------------------------------------------------------------
137    Name: Matthias Blume
138    Date: 2000/08/31 17:30:00 JST
139    Tag: blume-20001017-v110p29p1
140    Description:
141    
142    I made a version 110.29.1 with new bootfiles.
143    
144    Changes:  Modified pickler/unpickler for faster and leaner unpickling.
145              CM documentation changes and a small bugfix in CM's error reporting.
146    
147    ----------------------------------------------------------------------
148    Name: Lal George
149    Date: 2000/09/27 14:42:35 EDT
150    Tag: george-20000927-nodestatus
151    Description:
152    
153    Changed the type of the nodestatus, so that:
154    
155            SPILLED(~1)             is now SPILLED
156            SPILLED(m) where m>=0   is now MEMREG(m)
157            SPILLED(s) where s<~1   is now SPILL_LOC(~s)
158    
159    ----------------------------------------------------------------------
160    Name: Matthias Blume
161    Date: 2000/09/07 14:45:00 JST
162    Tag: blume-20000907-cmerrmsg
163    Description:
164    
165    Small tweak to CM to avoid getting ML syntax error messages twice.
166    
167    ----------------------------------------------------------------------
168    Name: Matthias Blume
169    Date: 2000/08/31 18:00:00 JST
170    Tag: blume-20000831-cvsbootfiles
171    Description:
172    
173    New URL for boot files (because the 110.29 files on the BL server do
174    now work correctly with my updated install scripts for yacc and lex).
175    
176    ----------------------------------------------------------------------
177    Name: Matthias Blume
178    Date: 2000/08/08 12:33:00 JST
179    Tag: blume-20000808-manual
180    Description:
181    
182    Tiny update to CM manual.
183    
184    ----------------------------------------------------------------------
185    Name: Allen Leung
186    Date: 2000/08/7 19:31:00
187    Tag: leunga-20000807-a-whole-bunch-of-stuff
188    
189      Moby, C--, SSA, x86, machine descriptions etc.  Should only affect C--
190    and Mobdy.
191    
192    1.  x86
193    
194       a.  Fixes to peephole module by John and Dan.
195       b.  Assembly fix to SETcc by Allen.
196       c.  Fix to c-call by John.
197       d.  Fix to spilling by John.  (This one deals with the missing FSTPT case)
198       e.  Instruction selection optimization to SETcc as suggested by John.
199    
200           For example,
201    
202            MV(32, x, COND(32, CMP(32, LT, a, b), LI 1, LI 0))
203    
204           should generate:
205    
206            MOVL a, x
207            SUBL b, x
208            SHRL 31, x
209    
210    2.  IR stuff
211    
212         A bunch of new DJ-graph related algorithms added.  These
213         speed up SSA construction.
214    
215    3.  SSA + Scheduling
216    
217         Added code for SSA and scheduling to the repository
218    
219    ----------------------------------------------------------------------
220    Name: Lal George
221    Date: 2000/07/27 11:53:14 EDT
222    
223    Tag: lal-20000727-linux-ppc
224    Description:
225    
226     Made changes to support Linux PPC.
227     p.s. I have confirmation that the 110.29 boot files work fine.
228    
229    ----------------------------------------------------------------------
230    Name: Matthias Blume
231    Date: 2000/07/27 17:40:00 JST
232    Tag: blume-20000727-scripts
233    Description:
234    
235    !!!! WARNING !!!!
236    You must recompile the runtime system!
237    !!!! WARNING !!!!
238    
239    This is basically another round of script-enhancements:
240    
241    1. sml, ml-build, and ml-makedepend accept options -D and -U to define
242       and undefine CM preprocessor symbols.
243    
244    2. ml-build avoids generating a new heap image if it finds that the
245       existing one is still ok.  (The condition is that no ML file had to
246       be recompiled and all ML files are found to be older that the heap
247       file.)
248    
249       To make this work smoothly, I also hacked the runtime system as
250       well as SMLofNJ.SysInfo to get access to the heap image suffix
251       (.sparc-solaris, ...) that is currently being used.
252    
253       Moreover, the signature of CM.mk_standalone has changed.  See the
254       CM manual.
255    
256    3. ml-makedepend accepts additional options -n, -a, and -o.  (See the
257       CM manual for details.)
258    
259    4. More CM manual updates:
260        - all of the above has been documented.
261        - there is now a section describing the (CM-related) command line
262          arguments that are accepted by the "sml" command
263    
264    ----------------------------------------------------------------------
265    Name: Matthias Blume
266    Date: 2000/07/25 16:20:00 JST
267    Tag: blume-20000725-makedepend
268    Description:
269    
270    Added a script called ml-makedepend.  This can be used in makefiles
271    for Unix' make in a way very similar to the "makedepend" command for
272    C.
273    
274    The script internally uses function CM.sources.
275    
276    Synopsis:
277    
278        ml-makedepend [-f makefile] cmfile targetname
279    
280    The default for the makefile is "makefile" (or "Makefile" should
281    "makefile" not exist).
282    
283    ml-makedepend adds a cmfile/targetname-specific section to this
284    makefile (after removing the previous version of this section).  The
285    section contains a single dependency specification with targetname on
286    the LHS (targetname is an arbitrary name), and a list of files derived
287    from the cmfile on the RHS.  Some of the files on the RHS are
288    ARCH/OPSYS-specific.  Therefore, ml-makedepend inserts references to
289    "make" variables $(ARCH) and $(OPSYS) in place of the corresponding
290    path names.  The makefile writer is responsible for making sure that
291    these variables have correct at the time "make" is invoked.
292    
293    ----------------------------------------------------------------------
294    Name: Matthias Blume
295    Date: 2000/07/22 23:30:00 JST
296    Tag: blume-20000722-urlupdate
297    Description:
298    
299    Changed BOOT and config/srcarchiveurl to point to BL server:
300    
301        ftp://ftp.research.bell-labs.com/dist/smlnj/working/110.29/
302    
303    ----------------------------------------------------------------------
304    Name: Matthias Blume
305    Date: 2000/07/18 18:00:00 JST
306    Tag: blume-20000718-Version_110_29
307    Description:
308    
309    1. Updated src/compiler/TopLevel/main/version.sml to version 110.29
310    
311    2. Updated config/version to 110.29
312    
313    3. Updated config/srcarchiveurl
314    
315    3. New boot files!
316       ftp://ftp.cs.princeton.edu/pub/people/blume/sml/110.29-autofetch
317    
318    ----------------------------------------------------------------------
319    Name: Matthias Blume
320    Date: 2000/07/11 13:58:00 JST
321    Tag: blume-20000711-doctypo
322    Description:
323    
324    Fixed a few typos in CM manual.
325    
326    ----------------------------------------------------------------------
327    Name: Allen Leung
328    Date: 2000/06/15 00:38:00
329    Tag: leunga-20000704-sparc-x86
330    
331    1. x86 peephole improvement sp += k; sp -= k => nop  [from John]
332    2. fix to x86 RET bug [found by Dan Grossman]
333    3. sparc assembly bug fix for ticc instructions [found by Fermin]
334    
335       Affects c-- and moby only
336    
337    ----------------------------------------------------------------------
338    Name: Matthias Blume
339    Date: 2000/07/04 15:26:00
340    Tag: blume-20000704-trigger
341    Description:
342    
343    1. Improvements to CM manual.
344    2. SMLofNJ.Internals.BTrace.trigger reinstated as an alternative way
345       of getting a back-trace.  The function, when called, raises an
346       internal exception which explicitly carries the full back-trace history,
347       so it is unaffected by any intervening handle-raise pairs ("trivial"
348       or not).  The interactive loop will print that history once it arrives
349       at top level.
350       Short of having all exceptions implicitly carry the full history, the
351       recommended way of using this facility is:
352         - compile your program with instrumentation "on"
353         - run it, when it raises an exception, look at the history
354         - if the history is "cut off" because of some handler, go and modify
355           your program so that it explicitly calls BTrace.trigger
356         - recompile (still instrumented), and rerun; look at the full history
357    
358    ----------------------------------------------------------------------
359    Name: Matthias Blume
360    Date: 2000/07/03 15:36:00 JST
361    Tag: blume-20000702-manual
362    Description:
363    
364    Small corrections and updates to CM manual.
365    
366    ----------------------------------------------------------------------
367    Name: Matthias Blume
368    Date: 2000/06/29 16:04:00 JST
369    Tag: blume-20000629-yacctool
370    Description:
371    
372    Changes:
373    
374    1. Class "mlyacc" now takes separate arguments to pass options to
375       generated .sml- and .sig-files independently.
376    2. Corresponding CM manual updates.
377    3. BTrace module now also reports call sites.  (However, for loop clusters
378       it only shows from where the cluster was entered.)  There are associated
379       modifications to core.sml, internals.{sig,sml}, btrace.sml, and btimp.sml.
380    
381    ----------------------------------------------------------------------
382    Name: Matthias Blume
383    Date: 2000/06/27 16:51:00 JST
384    Tag: blume-20000627-noweb
385    Description:
386    
387    Changes:
388    
389     1. Implemented "subdir" and "witness" options for noweb tool.
390        This caused some slight internal changes in CM's tool implementation.
391     2. Fixed bug in "tool plugin" mechanism.  This is essentially cleaning
392        some remaining issues from earlier path anchor changes.
393     3. Updated CM manual accordingly.
394    
395     4. Changed implementation of back-tracing so that I now consider it
396        ready for prime-time.
397    
398        In particular, you don't have to explicitly trigger the back-trace
399        anymore.  Instead, if you are running BTrace-instrumented code and
400        there is an uncaught exception (regardless of whether or not it was
401        raised in instrumented code), the top-level evalloop will print
402        the back-trace.
403    
404        Features:
405    
406          - Instrumented and uninstrumented code work together seemlessly.
407            (Of course, uninstrumented code is never mentioned in actual
408             back-traces.)
409    
410          - Asymptotic time- and space-complexity of instrumented code is
411            equal to that of uninstrumented code.  (This means that
412            tail-recursion is preserved by the instrumentation phase.)
413    
414          - Modules whose code has been instrumented in different sessions
415            work together without problem.
416    
417          - There is no penalty whatsoever on uninstrumented code.
418    
419          - There is no penalty on "raise" expressions, even in
420            instrumented code.
421    
422        A potential bug (or perhaps it is a feature, too):
423    
424          A back-trace reaches no further than the outermost instrumented
425          non-trivial "raise".  Here, a "trivial" raise is one that is the
426          sole RHS of a "handle" rule.  Thus, back-traces reach trough
427    
428               <exp> handle e => raise e
429    
430          and even
431    
432               <exp> handle Foo => raise Bar
433    
434          and, of course, through
435    
436               <exp> handle Foo => ...
437    
438         if the exception was not Foo.
439    
440         Back-traces always reach right through any un-instrumented code
441         including any of its "handle" expressions, trivial or not.
442    
443       To try this out, do the following:
444    
445         - Erase all existing binfiles for your program.
446           (You may keep binfiles for those modules where you think you
447            definitely don't need back-tracing.)
448         - Turn on back-trace instrumentation:
449              SMLofNJ.Internals.BTrace.mode (SOME true);
450         - Recompile your program.  (I.e., run "CM.make" or "use".)
451         - You may now turn instrumentation off again (if you want):
452              SMLofNJ.Internals.BTrace.mode (SOME false);
453         - Run your program as usual.  If it raises an exception that
454           reaches the interactive toplevel, then a back-trace will
455           automatically be printed.  After that, the toplevel loop
456           will print the exception history as usual.
457    
458    ----------------------------------------------------------------------
459    Name: Matthias Blume
460    Date: 2000/06/26 09:56:46 JST
461    Tag: blume-20000626-setup
462    Description:
463    
464    CM: - setup-parameter to "sml" added; this can be used to run arbitrary
465          ML code before and after compiling a file (e.g., to set compiler
466          flags)
467    
468    Compiler: - improved btrace API (in core.sml, internals.{sig,sml})
469              - associated changes to btrace.sml (BTrace instrumentation pass)
470              - cleaner implementation of btimp.sml (BTrace tracing and report
471                module)
472    
473    CM manual: * new path encoding documented
474               * description of setup-parameter to "sml" added
475    
476    The biggest user-visible change to back-tracing is that it is no
477    longer necessary to compile all traced modules within the same
478    session.  (This was a real limitation.)
479    
480    ----------------------------------------------------------------------
481    Name: Matthias Blume
482    Date: 2000/06/24 12:40:00 JST
483    Tag: blume-20000624-startup
484    Description:
485    
486    Fixes startup slowdown problem.  (I was calling SrcPath.sync a _tad_
487    bit too often -- to put it mildly. :)
488    
489    ----------------------------------------------------------------------
490    Name: Matthias Blume
491    Date: 2000/06/23 18:20:00 JST
492    Tag: blume-20000623-btrace
493    Description:
494    
495    This updates adds a backtrace facility to aid programmers in debugging
496    their programs.  This involves the following changes:
497    
498    1. Module system/smlnj/init/core.sml (structure _Core) now has hooks for
499       keeping track of the current call stack.  When programs are compiled
500       in a special mode, the compiler will insert calls to these hooks
501       into the user program.
502       "Hook" means that it is possible for different implementations of
503       back-tracing to register themselves (at different times).
504    
505    2. compiler/MiscUtil/profile/btrace.sml implements the annotation phase
506       as an Absyn.dec->Absyn.dec rewrite.  Normally this phase is turned off.
507       It can be turned on using this call:
508         SMLofNJ.Internals.BTrace.mode (SOME true);
509       Turning it off again:
510         SMLofNJ.Internals.BTrace.mode (SOME false);
511       Querying the current status:
512         SMLofNJ.Internals.BTrace.mode NONE;
513       Annotated programs are about twice as big as normal ones, and they
514       run a factor of 2 to 4 slower with a dummy back-trace plugin (one
515       where all hooks do nothing).  The slowdown with a plugin that is
516       actually useful (such as the one supplied by default) is even greater,
517       but in the case of the default plugin it is still only an constant
518       factor (amortized).
519    
520    3. system/Basis/Implementation/NJ/internals.{sig,sml} have been augmented
521       with a sub-structure BTrace for controlling back-tracing.  In particular,
522       the above-mentioned function "mode" controls whether the annotation
523       phase is invoked by the compiler.  Another important function is
524       "trigger": when called it aborts the current execution and causes
525       the top-level loop to print a full back-trace.
526    
527    4. compiler/MiscUtil/profile/btimp.sml is the current default plugin
528       for back-tracing.  It keeps track of the dynamic call stack and in
529       addition to that it keeps a partial history at each "level" of that
530       stack.  For example, if a tail-calls b, b tail-calls c, and c tail-calls
531       d and b (at separate times, dynamically), then the report will show:
532    
533       GOTO   d
534             /c
535       GOTO  \b
536       CALL   a
537    
538       This shows that there was an initial non-tail call of a, then a
539       tail-call to b or c, looping behavior in a cluster of functions that
540       consist of b and c, and then a goto from that cluster (i.e., either from
541       b or from c) to d.
542    
543       Note that (depending on the user program) the amount of information
544       that the back-trace module has to keep track of at each level is bounded
545       by a constant.  Thus, the whole implementation has the same asymptotical
546       complexity as the original program (both in space and in time).
547    
548    5. compiler/TopLevel/interact/evalloop.sml has been modified to
549       handle the special exception SMLofNJ.Internals.BTrace.BTrace
550       which is raised by the "trigger" function mentioned above.
551    
552    Notes on usage:
553    
554    - Annotated code works well together with unannotated code:
555    Unannotated calls simply do not show up at all in the backtrace.
556    
557    - It is not a good idea to let modules that were annotated during
558    different sessions run at the same time.  This is because the compiler
559    chooses small integers to identify individual functions, and there
560    will be clashes if different modules were compiled in separate sessions.
561    (Nothing will crash, and you will even be told about the clashes, but
562    back-trace information will in general not be useful.)
563    
564    - Back-tracing can be confused by callcc and capture.
565    
566    - The only way of getting a back-trace right now is to explicitly
567    invoke the "trigger" function from your user program.  Eventually, we
568    should make every exception carry back-trace information (if
569    available).  But since this creates more overhead at "raise"-time
570    (similar to the current exnHistory overhead), I have not yet
571    implemented this.  (The implementation will be rather easy.)  With
572    exceptions carrying back-trace information, this facility will be even
573    more useful because users don't need to modify their programs...
574    
575    - While it is possible to compile the compiler with back-trace
576    annotations turned on (I did it to get some confidence in
577    correctness), you must make absolutely sure that core.sml and
578    btimp.sml are compiled WITHOUT annotation!  (core.sml cannot actually
579    be compiled with annotation because there is no core access yet, but
580    if you compile btimp.sml with annotation, then the system will go into
581    an infinite recursion and crash.)
582    Since CM currently does not know about BTrace, the only way to turn
583    annotations on and off for different modules of the compiler is to
584    interrupt CMB.make, change the settings, and re-invoke it.  Of course,
585    this is awkward and clumsy.
586    
587    Sample sessions:
588    
589    Standard ML of New Jersey v110.28.1 [FLINT v1.5], June 5, 2000
590    - SMLofNJ.Internals.BTrace.mode (SOME true);
591    [autoloading]
592    [autoloading done]
593    val it = false : bool
594    - structure X = struct
595    -     fun main n = let
596    -         fun a (x, 0) = d x
597    -           | a (x, n) = b (x, n - 1)
598    -         and b (x, n) = c (x, n)
599    -         and c (x, n) = a (x, n)
600    -         and d x = e (x, 3)
601    -         and e (x, 0) = f x
602    -           | e (x, n) = e (x, n - 1)
603    -         and f 0 = SMLofNJ.Internals.BTrace.trigger ()
604    -           | f n = n * g (n - 1)
605    -         and g n = a (n, 3)
606    -     in
607    -         f n
608    -     end
609    - end;
610    structure X : sig val main : int -> int end
611    - X.main 3;
612    *** BACK-TRACE ***
613    GOTO   stdIn:4.2-13.20: X.main[2].f
614    GOTO-( stdIn:4.2-13.20: X.main[2].e
615    GOTO   stdIn:4.2-13.20: X.main[2].d
616         / stdIn:4.2-13.20: X.main[2].a
617         | stdIn:4.2-13.20: X.main[2].b
618    GOTO-\ stdIn:4.2-13.20: X.main[2].c
619    CALL   stdIn:4.2-13.20: X.main[2].g
620    GOTO   stdIn:4.2-13.20: X.main[2].f
621    GOTO-( stdIn:4.2-13.20: X.main[2].e
622    GOTO   stdIn:4.2-13.20: X.main[2].d
623         / stdIn:4.2-13.20: X.main[2].a
624         | stdIn:4.2-13.20: X.main[2].b
625    GOTO-\ stdIn:4.2-13.20: X.main[2].c
626    CALL   stdIn:4.2-13.20: X.main[2].g
627    GOTO   stdIn:4.2-13.20: X.main[2].f
628    GOTO-( stdIn:4.2-13.20: X.main[2].e
629    GOTO   stdIn:4.2-13.20: X.main[2].d
630         / stdIn:4.2-13.20: X.main[2].a
631         | stdIn:4.2-13.20: X.main[2].b
632    GOTO-\ stdIn:4.2-13.20: X.main[2].c
633    CALL   stdIn:4.2-13.20: X.main[2].g
634    GOTO   stdIn:4.2-13.20: X.main[2].f
635    CALL   stdIn:2.15-17.4: X.main[2]
636    -
637    
638    (Note that because of a FLINt bug the above code currently does not
639    compile without BTrace turned on.)
640    
641    Here is another example, using my modified Tiger compiler:
642    
643    Standard ML of New Jersey v110.28.1 [FLINT v1.5], June 5, 2000
644    - SMLofNJ.Internals.BTrace.mode (SOME true);
645    [autoloading]
646    [autoloading done]
647    val it = false : bool
648    - CM.make "sources.cm";
649    [autoloading]
650    ...
651    [autoloading done]
652    [scanning sources.cm]
653    [parsing (sources.cm):parse.sml]
654    [creating directory CM/SKEL ...]
655    [parsing (sources.cm):tiger.lex.sml]
656    ...
657    [wrote CM/sparc-unix/semant.sml]
658    [compiling (sources.cm):main.sml]
659    [wrote CM/sparc-unix/main.sml]
660    [New bindings added.]
661    val it = true : bool
662    - Main.compile ("../testcases/merge.tig", "foo.out");
663    *** BACK-TRACE ***
664    CALL   lib/semant.sml:99.2-396.21: SemantFun[2].transExp.trvar
665    CALL   lib/semant.sml:99.2-396.21: SemantFun[2].transExp.trexp
666    CALL   lib/semant.sml:289.3-295.22: SemantFun[2].transExp.trexp.check[2]
667    GOTO   lib/semant.sml:289.3-295.22: SemantFun[2].transExp.trexp.check[2]
668    CALL   lib/semant.sml:99.2-396.21: SemantFun[2].transExp.trexp
669    CALL   lib/semant.sml:99.2-396.21: SemantFun[2].transExp.trexp
670    CALL   lib/semant.sml:488.3-505.6: SemantFun[2].transDec.trdec[2].transBody[2]
671         / lib/semant.sml:411.65-543.8: SemantFun[2].transDec
672    CALL-\ lib/semant.sml:413.2-540.9: SemantFun[2].transDec.trdec[2]
673    CALL   lib/semant.sml:99.2-396.21: SemantFun[2].transExp.trexp
674    CALL   lib/semant.sml:8.52-558.4: SemantFun[2].transProg[2]
675    CALL   main.sml:1.18-118.4: Main.compile[2]
676    -
677    
678    ----------------------------------------------------------------------
679    Name: Matthias Blumen
680    Date: 2000/06/21 18:00:00 JST
681    Tag: blume-20000621-manual
682    Description:
683    
684    CM manual update: Path environments documented.
685    
686    ----------------------------------------------------------------------
687    Name: Matthias Blume
688    Date: 2000/06/19 13:40:00
689    Tag: blume-20000619-manual
690    Description:
691    
692    CM manual and system/README update.  This only covers the fact that
693    there are no more implicit anchors.  (Path environments and the "bind"
694    option to "cm" have yet to be documented.)
695    
696    ----------------------------------------------------------------------
697    Name: Matthias Blume
698    Date: 2000/06/19 11:05:00 JST
699    Tag: blume-20000619-chdir-bugfix
700    Description:
701    
702    Fixed a bug in new SrcPath module that sometimes led to a bad chDir call.
703    
704    ----------------------------------------------------------------------
705    Name: Matthias Blume
706    Date: 2000/06/18 22:00:10 JST
707    Tag: blume-20000618-implicit-anchors-really-gone
708    Description:
709    
710    I updates the previous HISTORY entry where I forgot to mention that
711    implicit anchors are no longer with us.
712    
713    The current update also gets rid of the (now useless) controller
714    CM.Control.implicit_anchors.
715    
716    ----------------------------------------------------------------------
717    Name: Matthias Blume
718    Date: 2000/06/16 17:30:00 JST
719    Tag: blume-20000616-anchorenv
720    Description:
721    
722    This patch implements the long anticipated (just kidding :) "anchor
723    environment" mechanism.  In the course of doing this, I also
724    re-implemented CM's internal "SrcPath" module from scratch.  The new
725    one should be more robust in certain boundary cases.  In any case, it
726    is a lot cleaner than its predecessor (IMHO).
727    
728    This time, although there is yet another boot file format change, I
729    kept the unpickler backward-compatible.  As a result, no new bootfiles
730    are necessary and bootstrapping is straightforward.  (You cannot read
731    new bootfiles into an old system, but the other way around is no
732    problem.)
733    
734    Visible changes:
735    
736    ** 0. Implicit path anchors (without the leading $-symbol) are no
737    longer recognized at all. This means that such path names are not
738    illegal either.  For example, the name basis.cm simply refers to a
739    local file called "basis.cm" (i.e, the name is an ordinary path
740    relative to .cm-files directory).  Or, to put it differently, only
741    names that start with $ are anchored paths.
742    
743    ** 1. The $<singlearc> abbreviation for $/<singlearc> has finally
744    vanished.
745    
746    John (Reppy) had critizised this as soon as I originally proposed and
747    implemented it, but at that time I did not really deeply believe
748    him. :) Now I came full-circle because I need the $<singlearc> syntax
749    in another place where it cannot be seen as an abbreviation for
750    $/<singlearc>.  To avoid the confusion, $<singlearc> now means what it
751    seems to mean (i.e., it "expands" into the corresponding anchor
752    value).
753    
754    However, when paths are used as members in CM description files, it
755    continues to be true that there must be at least another arc after the
756    anchor.  This is now enforced separately during semantic analysis
757    (i.e., from a lexical/syntactical point of view, the notation is ok.)
758    
759    ** 2. The "cm" class now accepts an option "bind".  The option's value
760    is a sub-option list of precisely two items -- one labeled "anchor"
761    and the other one labeled "value".  As you might expect, "anchor" is
762    used to specify an anchor name to be bound, and "value" specifies what
763    the anchor is being bound to.
764    
765    The value must be a directory name and can be given in either standard
766    syntax (including the possibility that it is itself an anchored path)
767    or native syntax.
768    
769    Examples:
770    
771       foo.cm (bind:(anchor:bar value:$mystuff/bar))
772       lib.cm (bind:(anchor:a value:"H:\\x\\y\\z"))  (* only works under windows *)
773    
774    and so on.
775    
776    The meaning of this is that the .cm-file will be processed with an
777    augmented anchor environment where the given anchor(s) is/are bound to
778    the given values(s).
779    
780    The rationale for having this feature is this: Suppose you are trying
781    to use two different (already stable) libraries a.cm and b.cm (that
782    you perhaps didn't write yourself).  Further, suppose each of these
783    two libraries internally uses its own auxiliary library $aux/lib.cm.
784    Normally you would now have a problem because the anchor "lib" can not
785    be bound to more than one value globally.  Therefore, the project that
786    uses both a.cm and b.cm must locally redirect the anchor to some other
787    place:
788    
789       a.cm (bind:(anchor:lib value:/usr/lib/smlnj/a-stuff))
790       b.cm (bind:(anchor:lib value:/usr/lib/smlnj/b-stuff))
791    
792    This hard-wires $lib/aux.cm to /usr/lib/smlnj/a-stuff/aux.cm or
793    /usr/lib/smlnj/b-stuff/aux.cm, respectively.
794    
795    Hard-wiring path names is a bit inflexible (and CM will verbosely warn
796    you when you do so at the time of CM.stabilize).  Therefore, you can
797    also use an anchored path as the value:
798    
799      a.cm (bind:(anchor:lib value:$a-lib))
800      b.cm (bind:(anchor:lib value:$b-lib))
801    
802    Now you can globally configure (using the usual CM.Anchor.anchor or
803    pathconfig machinery) bindings for "a-lib" and "b-lib".  Since "lib"
804    itself is always locally bound, setting it globally is no longer
805    meaningful or necessary (but it does not hurt either).  In fact, "lib"
806    can still be used as a global anchor for separate purposes.  As a
807    matter of fact, one can locally define "lib" in terms of a global
808    "lib":
809    
810      a.cm (bind:(anchor:lib value:$lib/a))
811      b.cm (bind:(anchor:lib value:$lib/b))
812    
813    ** 3: The encoding of path names has changed.  This affects the way
814    path names are shown in CM's progress report and also the internal
815    protocol encoding used for parallel make.
816    
817    The encoding now uses one or more ':'-separated segments.  Each
818    segments corresponds to a file that has been specified relative to the
819    file given by its preceding segment.  The first segment is either
820    relative to the CWD, absolute, or anchored.  Each segment itself is
821    basically a Unix pathname; all segments but the first are relative.
822    
823    Example:
824    
825       $foo/bar/baz.cm:a/b/c.sml
826    
827    This path denotes the file bar/a/b/c.sml relative to the directory
828    denoted by anchor "foo".  Notice that the encoding also includes
829    baz.cm which is the .cm-file that listed a/b/c.sml.  As usual, such
830    paths are resolved relative to the .cm-files directory, so baz.cm must
831    be ignored to get the "real" pathname.
832    
833    To make this fact more obvious, CM puts the names of such "virtual
834    arcs" into parentheses when they appear in progress reports. (No
835    parentheses will appear in the internal protocol encoding.)  Thus,
836    what you really see is:
837    
838      $foo/bar/(baz.cm):a/b/c.sml
839    
840    I find this notation to be much more informative than before.
841    
842    Another new feature of the encoding is that special characters
843    including parentheses, colons, (back)slashes, and white space are
844    written as \ddd (where ddd is the decimal encoding of the character).
845    
846    *** The CM manual still needs to be updated.
847    
848    ----------------------------------------------------------------------
849    Name: Allen Leung
850    Date: 2000/06/15 00:38:00
851    Tag: leunga-20000615-x86-peephole
852    
853    x86 Peephole fix by Fermin.  Affects c-- and moby only.
854    
855    ----------------------------------------------------------------------
856    Name: Matthias Blume
857    Date: 2000/06/12 11:40:00
858    Tag: blume-20000612-parmakefix
859    Description:
860    
861    More cleanup after changing the file naming scheme: This time I
862    repaired the parallel make mechanism for CMB.make which I broke earlier.
863    
864    ----------------------------------------------------------------------
865    Name: Allen Leung
866    Date: 2000/06/09 01:25:00
867    Tag: leunga-20000609-various
868    
869    None of these things should affect normal SML/NJ operations
870    
871    1. Peephole improvements provided by Fermin (c--)
872    2. New annotation DEFUSE for adding extra dependence (moby)
873    3. New X86 LOCK instructions (moby)
874    4. New machine description language for reservation tables (scheduling)
875    5. Fixes to various optimization/analysis modules (branch chaining, dominator
876       trees etc.)
877    6. I've changed the CM files so that they can work with versions
878       110.0.6, 110.25 and 110.28
879    
880    ----------------------------------------------------------------------
881    Name: Matthias Blume
882    Date: 2000/06/09 12:40:00
883    Tag: blume-20000609-log
884    Description:
885    
886    - Removed all(?) remaining RCS Log entries from sources.
887    
888    - Fixed bug in ml-yacc and ml-lex sources (use explicit anchors for
889      anchored paths).
890    
891    ----------------------------------------------------------------------
892    Name: Matthias Blume
893    Date: 2000/06/07 17:00:00 JST
894    Tag: blume-20000607-no-implicit-anchors
895    Description:
896    
897    1. This update changes the default setting for
898    CM.Control.implicit_anchors from true to false.  This means that
899    implicit anchors are no longer permitted by default.  I also tried to
900    make sure that nothing else still relies on implicit anchors.
901    (This is the next step on the schedule towards a CM that does not even
902    have the notion of implicit anchors anymore.)
903    
904    2. More CM manual updates.
905    
906    3. I managed to track down and fix the pickling bug I mentioned last
907    time.  Because of the previously existing workaround, this entails no
908    immediate practical changes.
909    
910    ----------------------------------------------------------------------
911    Name: Matthias Blume
912    Date: 2000/06/06 11:15:00 JST
913    Tag: blume-20000606-lazierpickle
914    Description:
915    
916    !!!! NEW BOOT FILES !!!!
917    
918    * The main purpose of this update is to make library pickles lazier in
919    order to reduce the initial space penalty for autoloading a library.
920    As a result, it is now possible to have $smlnj/compiler.cm
921    pre-registered.  This should take care of the many complaints or
922    inquiries about missing structure Compiler.  This required changes to
923    CM's internal data structures and small tweaks to some algorithms.
924    
925    As a neat additional effect, it is no longer necessary (for the sake
926    of lean heap image files) to distinguish between a "minimal" CM and a
927    "full" CM.  Now, there is only one CM (i.e., the "full" version:
928    $smlnj/cm.cm aka $smlnj/cm/full.cm), and it is always available at the
929    interactive top level. ($smlnj/cm/minimal.cm is gone.)
930    
931    To make the life of compiler-hackers easier, "makeml" now also
932    pre-registers $smlnj/cmb.cm (aka $smlnj/cmb/current.cm).  In other
933    words, after you bootstrap a new sml for the first time, you will not
934    have to autoload $smlnj/cmb.cm again afterwards.  (The first time
935    around you will still have to do it, though.)
936    
937    * A second change consists of major updates to the CM manual.  There
938    are now several appendices with summary information and also a full
939    specification of the CM description file syntax.
940    
941    * In directory src/system I added the script "allcross".  This script
942    invokes sml and cross-compiles the compiler for all supported
943    architectures.  (Useful when providing a new set of boot files.)
944    
945    * There seems to be a latent bug in my "lazy pickles" mechanism.  I
946    added a small tweak to pickle-util.sml to work around this problem,
947    but it is not a proper fix yet.  I will investigate further.  (The
948    effect of the bug was an inflation of library pickle size.)
949    
950    * Version number increased to 110.28.1 (to avoid compatibility problems).
951    
952    ----------------------------------------------------------------------
953    Name: Allen Leung
954    Date: 2000/05/25 17:28 EDT
955    Tag: leunga-20000525-ra
956    Description:
957    
958      Fixed a bug in freezing phase of the register allocator.
959    
960    ----------------------------------------------------------------------
961    Name: Allen Leung
962    Date: 2000/05/15 22:53 EDT
963    Tag: leunga-20000515-alpha-x86-ra
964    Description:
965    
966      1. Alpha
967    
968          Slight cleanup.  Removed the instruction SGNXL
969    
970      2. X86
971    
972          Added the following instructions to the instruction set:
973    
974            ROLx, RORx,
975            BTx, BTSx, BTLx, BTRx,
976            XCHGx, and variants with the LOCK prefix
977    
978      3. Register Allocation
979    
980          The module ra-rewrite-with-renaming has been improved.
981    
982      These have no effect on SML/NJ.
983    
984    ----------------------------------------------------------------------
985    Name: Matthias Blume
986    Date: 2000/05/15 16:20:00 JST
987    Tag: blume-20000515-lightrebuild
988    Description:
989    
990    1. I added an alternative to "-rebuild" to "makeml".  The difference is
991       that prior to calling CMB.make' the CM-variable "LIGHT" will be
992       defined.  In effect, the command will not build any cross-compiler
993       backends and therefore finish more quickly.
994    
995       The "fixpt" script also takes a "-light" switch to be able to use
996       this new facility while compiling for a fixpoint.
997    
998    2. I replaced all mentions of anchored paths in group owner specifications
999       with simple relative paths (usually starting with "..").
1000       The rationale is that a library's internal workings should not be
1001       compromised by the lack of some anchor.  (An anchor is necessary
1002       for someone who wants to refer to the library by an anchored path,
1003       but it should not be necessary to build the same library in the first
1004       place.)
1005    
1006    3. I changed the way CM's tool mechanism determines the shell command
1007       string used for things like ml-yacc etc. so that it does not break
1008       when CM.Control.implicit_anchors is turned off.
1009    
1010    ----------------------------------------------------------------------
1011    Name: Matthias Blume
1012    Date: 2000/05/12 18:20:00 JST
1013    Tag: blume-20000512-ml-build
1014    Description:
1015    
1016    Fixed a bug in config/_ml-build that prevented ml-yacc and ml-lex from
1017    getting installed properly (by config/install.sh).
1018    
1019    ----------------------------------------------------------------------
1020    Name: Matthias Blume
1021    Date: 2000/05/12 17:30:00 JST
1022    Tag: blume-20000512-anchors
1023    Description:
1024    
1025    !!! NEW BOOT FILES !!!
1026    
1027    This change is in preparation of fading out support for "implicitly
1028    anchored path names".  I went through all sources and used the
1029    explicit (and relatively new) $-notation.  See system/README and the
1030    CM manual for more info on this.
1031    
1032    I also modified the anchoring scheme for some things such as "smlnj",
1033    "MLRISC", "cm", etc. to take advantage of the fact that explicit
1034    anchors are more expressive: anchor name and first arc do not have to
1035    coincide.  This entails the following user-visible change:
1036    
1037    You have to write $smlnj/foo/bar instead of smlnj/foo/bar.  In
1038    particular, when you fire up sml with a command-line argument, say,
1039    e.g.:
1040    
1041       sml '$smlnj/cmb.cm'
1042    
1043    At the ML toplevel prompt:
1044    
1045       CM.autoload "$smlnj/cmb.cm";
1046    
1047    There is also a new controller in CM.Control that can be used to turn
1048    off all remaining support for implicit anchors by saying:
1049    
1050        CM.autoload "$smlnj/
1051        #set CM.Control.implicit_anchors false;
1052    
1053    This causes CM to reject implicitly anchored paths.  This is (for the
1054    time being) less permissive than the "final" version where there will
1055    be no more such implicit anchors and relative paths will be just that:
1056    relative.
1057    
1058    The next step (version after next version?) will be to make the
1059    default for CM.Control.implicit_anchors false.  After the dust has
1060    settled, I can then produce the "final" version of this...
1061    
1062    Note: Since bootstrapping is a bit tricky, I provided new boot files.
1063    
1064    ----------------------------------------------------------------------
1065    Name: Matthias Blume
1066    Date: 2000/05/11 16:30:00 JST
1067    Tag: blume-20000511-sources
1068    Description:
1069    
1070    The main change is that I added function CM.sources as a generalized
1071    version of the earlier CM.makedepend.  This entails the following
1072    additional changes:
1073    
1074      - CM.makedepend has been dropped.
1075    
1076      - CM manual has been updated.
1077    
1078      - TOOLS signature and API have been changed.
1079    
1080    ----------------------------------------------------------------------
1081    Name: Allen Leung
1082    Date: 2000/05/10 21:17 EDT
1083    Tag: leunga-20000510-moby-c--ssa
1084    Description:
1085    
1086      Various bug fixes and new features for C--, Moby and MLRISC optimizations.
1087    None of these affect SML/NJ.
1088    
1089    1. Register Allocation
1090    
1091        a. A new ra spilling module (ra/ra-spill-with-renaming) is implemented.
1092           This module tries to remove local (i.e. basic block level) redundancies
1093           during spilling.
1094    
1095        b. A new framework for performing region based register allocation.
1096           Not yet entirely functional.
1097    
1098    2. X86
1099    
1100       a. DefUse for POP was missing the stack pointer [found by Lal]
1101       b. Reload for CALL was incorrect in X86Spill [found by John]
1102       c. Various fixes in X86Spill so that it can be used correctly for
1103          the new spilling module.
1104    
1105    3. SSA/IR
1106    
1107       a. New module ir/dj-dataflow.sml implements elimination based
1108          data flow analysis.
1109    
1110    4. MLRiscGen
1111    
1112       a. Fix for gc type annotation
1113    
1114    5. MDGen
1115    
1116       Various fixes for machine description -> ml code translation.  For ssa
1117       only.
1118    
1119    ----------------------------------------------------------------------
1120    Name: Allen Leung
1121    Date: 2000/05/08 22:17 EDT
1122    Tag: leunga-20000508-labexp
1123    Description:
1124    
1125      Fermin has found a few assembly problems with constant expressions
1126      generated in LabelExp.  Mostly, the problems involve extra parentheses,
1127      which choke on dumb assemblers.  This is his fix.
1128    
1129    ----------------------------------------------------------------------
1130    Name: Dave MacQueen
1131    Date: 2000/04/09 14:00 EDT
1132    Tag: dbm-20000502-Version_110_28
1133    Description:
1134    
1135    1. Updated src/compiler/TopLevel/main/version.sml to version 110.28
1136    
1137    2. Updated config/version to 110.28
1138    
1139    3. Updated config/srcarchiveurl
1140    
1141    3. New boot files!
1142       ftp://ftp.research.bell-labs.com/dist/smlnj/working/110.28/
1143    
1144    ----------------------------------------------------------------------
1145    Name: Matthias Blume
1146    Date: 2000/05/01 19:05:00 JST
1147    Tag: blume-20000501-noweb
1148    Description:
1149    
1150    A new noweb tool has been added.  The existing system is entirely
1151    unaffected by this, but some CM users have asked for renewed noweb
1152    support.  Everything is documented in the CM manual.
1153    
1154    New (plugin) libraries:
1155    
1156       noweb-tool.cm
1157       nw-ext.cm
1158    
1159    ----------------------------------------------------------------------
1160    Name: Dave MacQueen
1161    Date: 2000/04/30 12:40PM EDT
1162    Tag: dbm-20000430-bug_fixes
1163    Description:
1164    
1165    1. Fix for bug 1498
1166       smlnj/src/system/Basis/Implementation/Unsafe/object.sig
1167       smlnj/src/system/Basis/Implementation/Unsafe/object.sml
1168         added toRealArray function
1169       smlnj/src/compiler/MiscUtil/print/ppobj.sml
1170         added check for tag Obj.RealArray to array printing case in ppObj
1171    
1172    2. Fix for bug 1510
1173       smlnj/src/compiler/Semant/types/typesutil.sml
1174         fixed definition of dummyargs (used by equalTycon) so that
1175         dummy args are distinct types
1176    
1177  ----------------------------------------------------------------------  ----------------------------------------------------------------------
1178  Name: Matthias Blume  Name: Matthias Blume
1179  Date: 2000/04/30 01:00:00 JST  Date: 2000/04/30 01:00:00 JST
# Line 277  Line 1440 
1440    
1441  3.  Assembly  3.  Assembly
1442    
1443      When generating assemby, resolve the value of client defined constants,      When generating assembly, resolve the value of client defined constants,
1444      instead of generating symbolic values.  This is controlled by the      instead of generating symbolic values.  This is controlled by the
1445      new flag "asm-resolve-constants", which is default to true.      new flag "asm-resolve-constants", which is default to true.
1446    
# Line 300  Line 1463 
1463    
1464  To this end, I arranged that instead of "structure Core" as "structure  To this end, I arranged that instead of "structure Core" as "structure
1465  _Core" is bound in the pervasive environment.  Core access is done via  _Core" is bound in the pervasive environment.  Core access is done via
1466  _Core (which can never be accidentially rebound because _Core is not a  _Core (which can never be accidentally rebound because _Core is not a
1467  legal surface-syntax symbol).  legal surface-syntax symbol).
1468    
1469  The current solution is much cleaner because the core environment is  The current solution is much cleaner because the core environment is
# Line 310  Line 1473 
1473  with dynamic and symbolic parts of the core environment.  with dynamic and symbolic parts of the core environment.
1474    
1475  Remaining hackery (to bind the "magic" symbol _Core) is localized in the  Remaining hackery (to bind the "magic" symbol _Core) is localized in the
1476  compilation mananger's bootstrap compiler (actually: in the "init group"  compilation manager's bootstrap compiler (actually: in the "init group"
1477  handling).  See the comments in src/system/smlnj/init/init.cmi for  handling).  See the comments in src/system/smlnj/init/init.cmi for
1478  more details.  more details.
1479    
# Line 425  Line 1588 
1588     (specified in the .cm file at each instance where the tool's class is     (specified in the .cm file at each instance where the tool's class is
1589     used).     used).
1590    
1591     This was done to accomodate the new "make" and "shell" tools which     This was done to accommodate the new "make" and "shell" tools which
1592     facilitate fairly seemless hookup to portions of code managed using     facilitate fairly seamless hookup to portions of code managed using
1593     Makefiles or Shell scripts.     Makefiles or Shell scripts.
1594    
1595     There are no classes "shared" or "private" anymore.  Instead, the     There are no classes "shared" or "private" anymore.  Instead, the
# Line 438  Line 1601 
1601    
1602     All existing tools are described in the CM manual.     All existing tools are described in the CM manual.
1603    
1604     - Slightly better error handling.  (CM now surpresses many followup     - Slightly better error handling.  (CM now suppresses many followup
1605     error messages that tended to be more annoying than helpful.)     error messages that tended to be more annoying than helpful.)
1606    
1607  2. Major changes to the compiler's static environment data structures.  2. Major changes to the compiler's static environment data structures.
# Line 572  Line 1735 
1735    
1736       I've changed andl to testl in the floating point test sequence       I've changed andl to testl in the floating point test sequence
1737       whenever appropriate.  The Intel optimization guide states that       whenever appropriate.  The Intel optimization guide states that
1738       testl is perferable to andl.       testl is preferable to andl.
1739    
1740  7. RA (x86 only)  7. RA (x86 only)
1741    
# Line 754  Line 1917 
1917    
1918  1. Tools.registerStdShellCmdTool (from smlnj/cm/tool.cm) takes an  1. Tools.registerStdShellCmdTool (from smlnj/cm/tool.cm) takes an
1919  additional argument called "template" which is an optional string that  additional argument called "template" which is an optional string that
1920  specifiel the layout of the tool command line.  See the CM manual for  specifies the layout of the tool command line.  See the CM manual for
1921  explanation.  explanation.
1922    
1923  2. A special-purpose tool can be "regisitered" by simply dropping the  2. A special-purpose tool can be "registered" by simply dropping the
1924  corresponding <...>-tool.cm (and/or <...>-ext.cm) into the same  corresponding <...>-tool.cm (and/or <...>-ext.cm) into the same
1925  directory where the .cm file lives that uses this tool.  (The  directory where the .cm file lives that uses this tool.  (The
1926  behavior/misfeature until now was to look for the tool description  behavior/misfeature until now was to look for the tool description
# Line 801  Line 1964 
1964  rebuild.  Having sets of the form <base><k>.{bin,boot}.<arch>-unix for  rebuild.  Having sets of the form <base><k>.{bin,boot}.<arch>-unix for
1965  <k>=1,2,... is normally not a good idea when invoking fixpt.  However,  <k>=1,2,... is normally not a good idea when invoking fixpt.  However,
1966  they might be the result of an earlier partial run of fixpt (which  they might be the result of an earlier partial run of fixpt (which
1967  perhaps got accidentially killed).  In this case, fixpt will quickly  perhaps got accidentally killed).  In this case, fixpt will quickly
1968  move through what exists before continuing where it left off earlier,  move through what exists before continuing where it left off earlier,
1969  and, thus, saves a lot of time.  and, thus, saves a lot of time.
1970    
# Line 851  Line 2014 
2014    it from that remote directory.    it from that remote directory.
2015    This should simplify installation further:  For machines that have    This should simplify installation further:  For machines that have
2016    access to the internet, just fetch <version>-config.tgz, unpack it,    access to the internet, just fetch <version>-config.tgz, unpack it,
2017    edit config/targets, and go (run config/install.sh).  The scipt will    edit config/targets, and go (run config/install.sh).  The script will
2018    fetch everything else that it might need all by itself.    fetch everything else that it might need all by itself.
2019    
2020    For CVS users, this mechanism is not relevant for source archives, but    For CVS users, this mechanism is not relevant for source archives, but

Legend:
Removed from v.632  
changed lines
  Added in v.733

root@smlnj-gforge.cs.uchicago.edu
ViewVC Help
Powered by ViewVC 1.0.0