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 578, Tue Mar 14 05:16:29 2000 UTC revision 701, Thu Aug 31 08:46:43 2000 UTC
# Line 13  Line 13 
13  Description:  Description:
14  ----------------------------------------------------------------------  ----------------------------------------------------------------------
15  Name: Matthias Blume  Name: Matthias Blume
16    Date: 2000/08/31 18:00:00 JST
17    Tag: blume-20000831-cvsbootfiles
18    Description:
19    
20    New URL for boot files (because the 110.29 files on the BL server do
21    now work correctly with my updated install scripts for yacc and lex).
22    
23    ----------------------------------------------------------------------
24    Name: Matthias Blume
25    Date: 2000/08/08 12:33:00 JST
26    Tag: blume-20000808-manual
27    Description:
28    
29    Tiny update to CM manual.
30    
31    ----------------------------------------------------------------------
32    Name: Allen Leung
33    Date: 2000/08/7 19:31:00
34    Tag: leunga-20000807-a-whole-bunch-of-stuff
35    
36      Moby, C--, SSA, x86, machine descriptions etc.  Should only affect C--
37    and Mobdy.
38    
39    1.  x86
40    
41       a.  Fixes to peephole module by John and Dan.
42       b.  Assembly fix to SETcc by Allen.
43       c.  Fix to c-call by John.
44       d.  Fix to spilling by John.  (This one deals with the missing FSTPT case)
45       e.  Instruction selection optimization to SETcc as suggested by John.
46    
47           For example,
48    
49            MV(32, x, COND(32, CMP(32, LT, a, b), LI 1, LI 0))
50    
51           should generate:
52    
53            MOVL a, x
54            SUBL b, x
55            SHRL 31, x
56    
57    2.  IR stuff
58    
59         A bunch of new DJ-graph related algorithms added.  These
60         speed up SSA construction.
61    
62    3.  SSA + Scheduling
63    
64         Added code for SSA and scheduling to the repository
65    
66    ----------------------------------------------------------------------
67    Name: Lal George
68    Date: 2000/07/27 11:53:14 EDT
69    
70    Tag: lal-20000727-linux-ppc
71    Description:
72    
73     Made changes to support Linux PPC.
74     p.s. I have confirmation that the 110.29 boot files work fine.
75    
76    ----------------------------------------------------------------------
77    Name: Matthias Blume
78    Date: 2000/07/27 17:40:00 JST
79    Tag: blume-20000727-scripts
80    Description:
81    
82    !!!! WARNING !!!!
83    You must recompile the runtime system!
84    !!!! WARNING !!!!
85    
86    This is basically another round of script-enhancements:
87    
88    1. sml, ml-build, and ml-makedepend accept options -D and -U to define
89       and undefine CM preprocessor symbols.
90    
91    2. ml-build avoids generating a new heap image if it finds that the
92       existing one is still ok.  (The condition is that no ML file had to
93       be recompiled and all ML files are found to be older that the heap
94       file.)
95    
96       To make this work smoothly, I also hacked the runtime system as
97       well as SMLofNJ.SysInfo to get access to the heap image suffix
98       (.sparc-solaris, ...) that is currently being used.
99    
100       Moreover, the signature of CM.mk_standalone has changed.  See the
101       CM manual.
102    
103    3. ml-makedepend accepts additional options -n, -a, and -o.  (See the
104       CM manual for details.)
105    
106    4. More CM manual updates:
107        - all of the above has been documented.
108        - there is now a section describing the (CM-related) command line
109          arguments that are accepted by the "sml" command
110    
111    ----------------------------------------------------------------------
112    Name: Matthias Blume
113    Date: 2000/07/25 16:20:00 JST
114    Tag: blume-20000725-makedepend
115    Description:
116    
117    Added a script called ml-makedepend.  This can be used in makefiles
118    for Unix' make in a way very similar to the "makedepend" command for
119    C.
120    
121    The script internally uses function CM.sources.
122    
123    Synopsis:
124    
125        ml-makedepend [-f makefile] cmfile targetname
126    
127    The default for the makefile is "makefile" (or "Makefile" should
128    "makefile" not exist).
129    
130    ml-makedepend adds a cmfile/targetname-specific section to this
131    makefile (after removing the previous version of this section).  The
132    section contains a single dependency specification with targetname on
133    the LHS (targetname is an arbitrary name), and a list of files derived
134    from the cmfile on the RHS.  Some of the files on the RHS are
135    ARCH/OPSYS-specific.  Therefore, ml-makedepend inserts references to
136    "make" variables $(ARCH) and $(OPSYS) in place of the corresponding
137    path names.  The makefile writer is responsible for making sure that
138    these variables have correct at the time "make" is invoked.
139    
140    ----------------------------------------------------------------------
141    Name: Matthias Blume
142    Date: 2000/07/22 23:30:00 JST
143    Tag: blume-20000722-urlupdate
144    Description:
145    
146    Changed BOOT and config/srcarchiveurl to point to BL server:
147    
148        ftp://ftp.research.bell-labs.com/dist/smlnj/working/110.29/
149    
150    ----------------------------------------------------------------------
151    Name: Matthias Blume
152    Date: 2000/07/18 18:00:00 JST
153    Tag: blume-20000718-Version_110_29
154    Description:
155    
156    1. Updated src/compiler/TopLevel/main/version.sml to version 110.29
157    
158    2. Updated config/version to 110.29
159    
160    3. Updated config/srcarchiveurl
161    
162    3. New boot files!
163       ftp://ftp.cs.princeton.edu/pub/people/blume/sml/110.29-autofetch
164    
165    ----------------------------------------------------------------------
166    Name: Matthias Blume
167    Date: 2000/07/11 13:58:00 JST
168    Tag: blume-20000711-doctypo
169    Description:
170    
171    Fixed a few typos in CM manual.
172    
173    ----------------------------------------------------------------------
174    Name: Allen Leung
175    Date: 2000/06/15 00:38:00
176    Tag: leunga-20000704-sparc-x86
177    
178    1. x86 peephole improvement sp += k; sp -= k => nop  [from John]
179    2. fix to x86 RET bug [found by Dan Grossman]
180    3. sparc assembly bug fix for ticc instructions [found by Fermin]
181    
182       Affects c-- and moby only
183    
184    ----------------------------------------------------------------------
185    Name: Matthias Blume
186    Date: 2000/07/04 15:26:00
187    Tag: blume-20000704-trigger
188    Description:
189    
190    1. Improvements to CM manual.
191    2. SMLofNJ.Internals.BTrace.trigger reinstated as an alternative way
192       of getting a back-trace.  The function, when called, raises an
193       internal exception which explicitly carries the full back-trace history,
194       so it is unaffected by any intervening handle-raise pairs ("trivial"
195       or not).  The interactive loop will print that history once it arrives
196       at top level.
197       Short of having all exceptions implicitly carry the full history, the
198       recommended way of using this facility is:
199         - compile your program with instrumentation "on"
200         - run it, when it raises an exception, look at the history
201         - if the history is "cut off" because of some handler, go and modify
202           your program so that it explicitly calls BTrace.trigger
203         - recompile (still instrumented), and rerun; look at the full history
204    
205    ----------------------------------------------------------------------
206    Name: Matthias Blume
207    Date: 2000/07/03 15:36:00 JST
208    Tag: blume-20000702-manual
209    Description:
210    
211    Small corrections and updates to CM manual.
212    
213    ----------------------------------------------------------------------
214    Name: Matthias Blume
215    Date: 2000/06/29 16:04:00 JST
216    Tag: blume-20000629-yacctool
217    Description:
218    
219    Changes:
220    
221    1. Class "mlyacc" now takes separate arguments to pass options to
222       generated .sml- and .sig-files independently.
223    2. Corresponding CM manual updates.
224    3. BTrace module now also reports call sites.  (However, for loop clusters
225       it only shows from where the cluster was entered.)  There are associated
226       modifications to core.sml, internals.{sig,sml}, btrace.sml, and btimp.sml.
227    
228    ----------------------------------------------------------------------
229    Name: Matthias Blume
230    Date: 2000/06/27 16:51:00 JST
231    Tag: blume-20000627-noweb
232    Description:
233    
234    Changes:
235    
236     1. Implemented "subdir" and "witness" options for noweb tool.
237        This caused some slight internal changes in CM's tool implementation.
238     2. Fixed bug in "tool plugin" mechanism.  This is essentially cleaning
239        some remaining issues from earlier path anchor changes.
240     3. Updated CM manual accordingly.
241    
242     4. Changed implementation of back-tracing so that I now consider it
243        ready for prime-time.
244    
245        In particular, you don't have to explicitly trigger the back-trace
246        anymore.  Instead, if you are running BTrace-instrumented code and
247        there is an uncaught exception (regardless of whether or not it was
248        raised in instrumented code), the top-level evalloop will print
249        the back-trace.
250    
251        Features:
252    
253          - Instrumented and uninstrumented code work together seemlessly.
254            (Of course, uninstrumented code is never mentioned in actual
255             back-traces.)
256    
257          - Asymptotic time- and space-complexity of instrumented code is
258            equal to that of uninstrumented code.  (This means that
259            tail-recursion is preserved by the instrumentation phase.)
260    
261          - Modules whose code has been instrumented in different sessions
262            work together without problem.
263    
264          - There is no penalty whatsoever on uninstrumented code.
265    
266          - There is no penalty on "raise" expressions, even in
267            instrumented code.
268    
269        A potential bug (or perhaps it is a feature, too):
270    
271          A back-trace reaches no further than the outermost instrumented
272          non-trivial "raise".  Here, a "trivial" raise is one that is the
273          sole RHS of a "handle" rule.  Thus, back-traces reach trough
274    
275               <exp> handle e => raise e
276    
277          and even
278    
279               <exp> handle Foo => raise Bar
280    
281          and, of course, through
282    
283               <exp> handle Foo => ...
284    
285         if the exception was not Foo.
286    
287         Back-traces always reach right through any un-instrumented code
288         including any of its "handle" expressions, trivial or not.
289    
290       To try this out, do the following:
291    
292         - Erase all existing binfiles for your program.
293           (You may keep binfiles for those modules where you think you
294            definitely don't need back-tracing.)
295         - Turn on back-trace instrumentation:
296              SMLofNJ.Internals.BTrace.mode (SOME true);
297         - Recompile your program.  (I.e., run "CM.make" or "use".)
298         - You may now turn instrumentation off again (if you want):
299              SMLofNJ.Internals.BTrace.mode (SOME false);
300         - Run your program as usual.  If it raises an exception that
301           reaches the interactive toplevel, then a back-trace will
302           automatically be printed.  After that, the toplevel loop
303           will print the exception history as usual.
304    
305    ----------------------------------------------------------------------
306    Name: Matthias Blume
307    Date: 2000/06/26 09:56:46 JST
308    Tag: blume-20000626-setup
309    Description:
310    
311    CM: - setup-parameter to "sml" added; this can be used to run arbitrary
312          ML code before and after compiling a file (e.g., to set compiler
313          flags)
314    
315    Compiler: - improved btrace API (in core.sml, internals.{sig,sml})
316              - associated changes to btrace.sml (BTrace instrumentation pass)
317              - cleaner implementation of btimp.sml (BTrace tracing and report
318                module)
319    
320    CM manual: * new path encoding documented
321               * description of setup-parameter to "sml" added
322    
323    The biggest user-visible change to back-tracing is that it is no
324    longer necessary to compile all traced modules within the same
325    session.  (This was a real limitation.)
326    
327    ----------------------------------------------------------------------
328    Name: Matthias Blume
329    Date: 2000/06/24 12:40:00 JST
330    Tag: blume-20000624-startup
331    Description:
332    
333    Fixes startup slowdown problem.  (I was calling SrcPath.sync a _tad_
334    bit too often -- to put it mildly. :)
335    
336    ----------------------------------------------------------------------
337    Name: Matthias Blume
338    Date: 2000/06/23 18:20:00 JST
339    Tag: blume-20000623-btrace
340    Description:
341    
342    This updates adds a backtrace facility to aid programmers in debugging
343    their programs.  This involves the following changes:
344    
345    1. Module system/smlnj/init/core.sml (structure _Core) now has hooks for
346       keeping track of the current call stack.  When programs are compiled
347       in a special mode, the compiler will insert calls to these hooks
348       into the user program.
349       "Hook" means that it is possible for different implementations of
350       back-tracing to register themselves (at different times).
351    
352    2. compiler/MiscUtil/profile/btrace.sml implements the annotation phase
353       as an Absyn.dec->Absyn.dec rewrite.  Normally this phase is turned off.
354       It can be turned on using this call:
355         SMLofNJ.Internals.BTrace.mode (SOME true);
356       Turning it off again:
357         SMLofNJ.Internals.BTrace.mode (SOME false);
358       Querying the current status:
359         SMLofNJ.Internals.BTrace.mode NONE;
360       Annotated programs are about twice as big as normal ones, and they
361       run a factor of 2 to 4 slower with a dummy back-trace plugin (one
362       where all hooks do nothing).  The slowdown with a plugin that is
363       actually useful (such as the one supplied by default) is even greater,
364       but in the case of the default plugin it is still only an constant
365       factor (amortized).
366    
367    3. system/Basis/Implementation/NJ/internals.{sig,sml} have been augmented
368       with a sub-structure BTrace for controlling back-tracing.  In particular,
369       the above-mentioned function "mode" controls whether the annotation
370       phase is invoked by the compiler.  Another important function is
371       "trigger": when called it aborts the current execution and causes
372       the top-level loop to print a full back-trace.
373    
374    4. compiler/MiscUtil/profile/btimp.sml is the current default plugin
375       for back-tracing.  It keeps track of the dynamic call stack and in
376       addition to that it keeps a partial history at each "level" of that
377       stack.  For example, if a tail-calls b, b tail-calls c, and c tail-calls
378       d and b (at separate times, dynamically), then the report will show:
379    
380       GOTO   d
381             /c
382       GOTO  \b
383       CALL   a
384    
385       This shows that there was an initial non-tail call of a, then a
386       tail-call to b or c, looping behavior in a cluster of functions that
387       consist of b and c, and then a goto from that cluster (i.e., either from
388       b or from c) to d.
389    
390       Note that (depending on the user program) the amount of information
391       that the back-trace module has to keep track of at each level is bounded
392       by a constant.  Thus, the whole implementation has the same asymptotical
393       complexity as the original program (both in space and in time).
394    
395    5. compiler/TopLevel/interact/evalloop.sml has been modified to
396       handle the special exception SMLofNJ.Internals.BTrace.BTrace
397       which is raised by the "trigger" function mentioned above.
398    
399    Notes on usage:
400    
401    - Annotated code works well together with unannotated code:
402    Unannotated calls simply do not show up at all in the backtrace.
403    
404    - It is not a good idea to let modules that were annotated during
405    different sessions run at the same time.  This is because the compiler
406    chooses small integers to identify individual functions, and there
407    will be clashes if different modules were compiled in separate sessions.
408    (Nothing will crash, and you will even be told about the clashes, but
409    back-trace information will in general not be useful.)
410    
411    - Back-tracing can be confused by callcc and capture.
412    
413    - The only way of getting a back-trace right now is to explicitly
414    invoke the "trigger" function from your user program.  Eventually, we
415    should make every exception carry back-trace information (if
416    available).  But since this creates more overhead at "raise"-time
417    (similar to the current exnHistory overhead), I have not yet
418    implemented this.  (The implementation will be rather easy.)  With
419    exceptions carrying back-trace information, this facility will be even
420    more useful because users don't need to modify their programs...
421    
422    - While it is possible to compile the compiler with back-trace
423    annotations turned on (I did it to get some confidence in
424    correctness), you must make absolutely sure that core.sml and
425    btimp.sml are compiled WITHOUT annotation!  (core.sml cannot actually
426    be compiled with annotation because there is no core access yet, but
427    if you compile btimp.sml with annotation, then the system will go into
428    an infinite recursion and crash.)
429    Since CM currently does not know about BTrace, the only way to turn
430    annotations on and off for different modules of the compiler is to
431    interrupt CMB.make, change the settings, and re-invoke it.  Of course,
432    this is awkward and clumsy.
433    
434    Sample sessions:
435    
436    Standard ML of New Jersey v110.28.1 [FLINT v1.5], June 5, 2000
437    - SMLofNJ.Internals.BTrace.mode (SOME true);
438    [autoloading]
439    [autoloading done]
440    val it = false : bool
441    - structure X = struct
442    -     fun main n = let
443    -         fun a (x, 0) = d x
444    -           | a (x, n) = b (x, n - 1)
445    -         and b (x, n) = c (x, n)
446    -         and c (x, n) = a (x, n)
447    -         and d x = e (x, 3)
448    -         and e (x, 0) = f x
449    -           | e (x, n) = e (x, n - 1)
450    -         and f 0 = SMLofNJ.Internals.BTrace.trigger ()
451    -           | f n = n * g (n - 1)
452    -         and g n = a (n, 3)
453    -     in
454    -         f n
455    -     end
456    - end;
457    structure X : sig val main : int -> int end
458    - X.main 3;
459    *** BACK-TRACE ***
460    GOTO   stdIn:4.2-13.20: X.main[2].f
461    GOTO-( stdIn:4.2-13.20: X.main[2].e
462    GOTO   stdIn:4.2-13.20: X.main[2].d
463         / stdIn:4.2-13.20: X.main[2].a
464         | stdIn:4.2-13.20: X.main[2].b
465    GOTO-\ stdIn:4.2-13.20: X.main[2].c
466    CALL   stdIn:4.2-13.20: X.main[2].g
467    GOTO   stdIn:4.2-13.20: X.main[2].f
468    GOTO-( stdIn:4.2-13.20: X.main[2].e
469    GOTO   stdIn:4.2-13.20: X.main[2].d
470         / stdIn:4.2-13.20: X.main[2].a
471         | stdIn:4.2-13.20: X.main[2].b
472    GOTO-\ stdIn:4.2-13.20: X.main[2].c
473    CALL   stdIn:4.2-13.20: X.main[2].g
474    GOTO   stdIn:4.2-13.20: X.main[2].f
475    GOTO-( stdIn:4.2-13.20: X.main[2].e
476    GOTO   stdIn:4.2-13.20: X.main[2].d
477         / stdIn:4.2-13.20: X.main[2].a
478         | stdIn:4.2-13.20: X.main[2].b
479    GOTO-\ stdIn:4.2-13.20: X.main[2].c
480    CALL   stdIn:4.2-13.20: X.main[2].g
481    GOTO   stdIn:4.2-13.20: X.main[2].f
482    CALL   stdIn:2.15-17.4: X.main[2]
483    -
484    
485    (Note that because of a FLINt bug the above code currently does not
486    compile without BTrace turned on.)
487    
488    Here is another example, using my modified Tiger compiler:
489    
490    Standard ML of New Jersey v110.28.1 [FLINT v1.5], June 5, 2000
491    - SMLofNJ.Internals.BTrace.mode (SOME true);
492    [autoloading]
493    [autoloading done]
494    val it = false : bool
495    - CM.make "sources.cm";
496    [autoloading]
497    ...
498    [autoloading done]
499    [scanning sources.cm]
500    [parsing (sources.cm):parse.sml]
501    [creating directory CM/SKEL ...]
502    [parsing (sources.cm):tiger.lex.sml]
503    ...
504    [wrote CM/sparc-unix/semant.sml]
505    [compiling (sources.cm):main.sml]
506    [wrote CM/sparc-unix/main.sml]
507    [New bindings added.]
508    val it = true : bool
509    - Main.compile ("../testcases/merge.tig", "foo.out");
510    *** BACK-TRACE ***
511    CALL   lib/semant.sml:99.2-396.21: SemantFun[2].transExp.trvar
512    CALL   lib/semant.sml:99.2-396.21: SemantFun[2].transExp.trexp
513    CALL   lib/semant.sml:289.3-295.22: SemantFun[2].transExp.trexp.check[2]
514    GOTO   lib/semant.sml:289.3-295.22: SemantFun[2].transExp.trexp.check[2]
515    CALL   lib/semant.sml:99.2-396.21: SemantFun[2].transExp.trexp
516    CALL   lib/semant.sml:99.2-396.21: SemantFun[2].transExp.trexp
517    CALL   lib/semant.sml:488.3-505.6: SemantFun[2].transDec.trdec[2].transBody[2]
518         / lib/semant.sml:411.65-543.8: SemantFun[2].transDec
519    CALL-\ lib/semant.sml:413.2-540.9: SemantFun[2].transDec.trdec[2]
520    CALL   lib/semant.sml:99.2-396.21: SemantFun[2].transExp.trexp
521    CALL   lib/semant.sml:8.52-558.4: SemantFun[2].transProg[2]
522    CALL   main.sml:1.18-118.4: Main.compile[2]
523    -
524    
525    ----------------------------------------------------------------------
526    Name: Matthias Blumen
527    Date: 2000/06/21 18:00:00 JST
528    Tag: blume-20000621-manual
529    Description:
530    
531    CM manual update: Path environments documented.
532    
533    ----------------------------------------------------------------------
534    Name: Matthias Blume
535    Date: 2000/06/19 13:40:00
536    Tag: blume-20000619-manual
537    Description:
538    
539    CM manual and system/README update.  This only covers the fact that
540    there are no more implicit anchors.  (Path environments and the "bind"
541    option to "cm" have yet to be documented.)
542    
543    ----------------------------------------------------------------------
544    Name: Matthias Blume
545    Date: 2000/06/19 11:05:00 JST
546    Tag: blume-20000619-chdir-bugfix
547    Description:
548    
549    Fixed a bug in new SrcPath module that sometimes led to a bad chDir call.
550    
551    ----------------------------------------------------------------------
552    Name: Matthias Blume
553    Date: 2000/06/18 22:00:10 JST
554    Tag: blume-20000618-implicit-anchors-really-gone
555    Description:
556    
557    I updates the previous HISTORY entry where I forgot to mention that
558    implicit anchors are no longer with us.
559    
560    The current update also gets rid of the (now useless) controller
561    CM.Control.implicit_anchors.
562    
563    ----------------------------------------------------------------------
564    Name: Matthias Blume
565    Date: 2000/06/16 17:30:00 JST
566    Tag: blume-20000616-anchorenv
567    Description:
568    
569    This patch implements the long anticipated (just kidding :) "anchor
570    environment" mechanism.  In the course of doing this, I also
571    re-implemented CM's internal "SrcPath" module from scratch.  The new
572    one should be more robust in certain boundary cases.  In any case, it
573    is a lot cleaner than its predecessor (IMHO).
574    
575    This time, although there is yet another boot file format change, I
576    kept the unpickler backward-compatible.  As a result, no new bootfiles
577    are necessary and bootstrapping is straightforward.  (You cannot read
578    new bootfiles into an old system, but the other way around is no
579    problem.)
580    
581    Visible changes:
582    
583    ** 0. Implicit path anchors (without the leading $-symbol) are no
584    longer recognized at all. This means that such path names are not
585    illegal either.  For example, the name basis.cm simply refers to a
586    local file called "basis.cm" (i.e, the name is an ordinary path
587    relative to .cm-files directory).  Or, to put it differently, only
588    names that start with $ are anchored paths.
589    
590    ** 1. The $<singlearc> abbreviation for $/<singlearc> has finally
591    vanished.
592    
593    John (Reppy) had critizised this as soon as I originally proposed and
594    implemented it, but at that time I did not really deeply believe
595    him. :) Now I came full-circle because I need the $<singlearc> syntax
596    in another place where it cannot be seen as an abbreviation for
597    $/<singlearc>.  To avoid the confusion, $<singlearc> now means what it
598    seems to mean (i.e., it "expands" into the corresponding anchor
599    value).
600    
601    However, when paths are used as members in CM description files, it
602    continues to be true that there must be at least another arc after the
603    anchor.  This is now enforced separately during semantic analysis
604    (i.e., from a lexical/syntactical point of view, the notation is ok.)
605    
606    ** 2. The "cm" class now accepts an option "bind".  The option's value
607    is a sub-option list of precisely two items -- one labeled "anchor"
608    and the other one labeled "value".  As you might expect, "anchor" is
609    used to specify an anchor name to be bound, and "value" specifies what
610    the anchor is being bound to.
611    
612    The value must be a directory name and can be given in either standard
613    syntax (including the possibility that it is itself an anchored path)
614    or native syntax.
615    
616    Examples:
617    
618       foo.cm (bind:(anchor:bar value:$mystuff/bar))
619       lib.cm (bind:(anchor:a value:"H:\\x\\y\\z"))  (* only works under windows *)
620    
621    and so on.
622    
623    The meaning of this is that the .cm-file will be processed with an
624    augmented anchor environment where the given anchor(s) is/are bound to
625    the given values(s).
626    
627    The rationale for having this feature is this: Suppose you are trying
628    to use two different (already stable) libraries a.cm and b.cm (that
629    you perhaps didn't write yourself).  Further, suppose each of these
630    two libraries internally uses its own auxiliary library $aux/lib.cm.
631    Normally you would now have a problem because the anchor "lib" can not
632    be bound to more than one value globally.  Therefore, the project that
633    uses both a.cm and b.cm must locally redirect the anchor to some other
634    place:
635    
636       a.cm (bind:(anchor:lib value:/usr/lib/smlnj/a-stuff))
637       b.cm (bind:(anchor:lib value:/usr/lib/smlnj/b-stuff))
638    
639    This hard-wires $lib/aux.cm to /usr/lib/smlnj/a-stuff/aux.cm or
640    /usr/lib/smlnj/b-stuff/aux.cm, respectively.
641    
642    Hard-wiring path names is a bit inflexible (and CM will verbosely warn
643    you when you do so at the time of CM.stabilize).  Therefore, you can
644    also use an anchored path as the value:
645    
646      a.cm (bind:(anchor:lib value:$a-lib))
647      b.cm (bind:(anchor:lib value:$b-lib))
648    
649    Now you can globally configure (using the usual CM.Anchor.anchor or
650    pathconfig machinery) bindings for "a-lib" and "b-lib".  Since "lib"
651    itself is always locally bound, setting it globally is no longer
652    meaningful or necessary (but it does not hurt either).  In fact, "lib"
653    can still be used as a global anchor for separate purposes.  As a
654    matter of fact, one can locally define "lib" in terms of a global
655    "lib":
656    
657      a.cm (bind:(anchor:lib value:$lib/a))
658      b.cm (bind:(anchor:lib value:$lib/b))
659    
660    ** 3: The encoding of path names has changed.  This affects the way
661    path names are shown in CM's progress report and also the internal
662    protocol encoding used for parallel make.
663    
664    The encoding now uses one or more ':'-separated segments.  Each
665    segments corresponds to a file that has been specified relative to the
666    file given by its preceding segment.  The first segment is either
667    relative to the CWD, absolute, or anchored.  Each segment itself is
668    basically a Unix pathname; all segments but the first are relative.
669    
670    Example:
671    
672       $foo/bar/baz.cm:a/b/c.sml
673    
674    This path denotes the file bar/a/b/c.sml relative to the directory
675    denoted by anchor "foo".  Notice that the encoding also includes
676    baz.cm which is the .cm-file that listed a/b/c.sml.  As usual, such
677    paths are resolved relative to the .cm-files directory, so baz.cm must
678    be ignored to get the "real" pathname.
679    
680    To make this fact more obvious, CM puts the names of such "virtual
681    arcs" into parentheses when they appear in progress reports. (No
682    parentheses will appear in the internal protocol encoding.)  Thus,
683    what you really see is:
684    
685      $foo/bar/(baz.cm):a/b/c.sml
686    
687    I find this notation to be much more informative than before.
688    
689    Another new feature of the encoding is that special characters
690    including parentheses, colons, (back)slashes, and white space are
691    written as \ddd (where ddd is the decimal encoding of the character).
692    
693    *** The CM manual still needs to be updated.
694    
695    ----------------------------------------------------------------------
696    Name: Allen Leung
697    Date: 2000/06/15 00:38:00
698    Tag: leunga-20000615-x86-peephole
699    
700    x86 Peephole fix by Fermin.  Affects c-- and moby only.
701    
702    ----------------------------------------------------------------------
703    Name: Matthias Blume
704    Date: 2000/06/12 11:40:00
705    Tag: blume-20000612-parmakefix
706    Description:
707    
708    More cleanup after changing the file naming scheme: This time I
709    repaired the parallel make mechanism for CMB.make which I broke earlier.
710    
711    ----------------------------------------------------------------------
712    Name: Allen Leung
713    Date: 2000/06/09 01:25:00
714    Tag: leunga-20000609-various
715    
716    None of these things should affect normal SML/NJ operations
717    
718    1. Peephole improvements provided by Fermin (c--)
719    2. New annotation DEFUSE for adding extra dependence (moby)
720    3. New X86 LOCK instructions (moby)
721    4. New machine description language for reservation tables (scheduling)
722    5. Fixes to various optimization/analysis modules (branch chaining, dominator
723       trees etc.)
724    6. I've changed the CM files so that they can work with versions
725       110.0.6, 110.25 and 110.28
726    
727    ----------------------------------------------------------------------
728    Name: Matthias Blume
729    Date: 2000/06/09 12:40:00
730    Tag: blume-20000609-log
731    Description:
732    
733    - Removed all(?) remaining RCS Log entries from sources.
734    
735    - Fixed bug in ml-yacc and ml-lex sources (use explicit anchors for
736      anchored paths).
737    
738    ----------------------------------------------------------------------
739    Name: Matthias Blume
740    Date: 2000/06/07 17:00:00 JST
741    Tag: blume-20000607-no-implicit-anchors
742    Description:
743    
744    1. This update changes the default setting for
745    CM.Control.implicit_anchors from true to false.  This means that
746    implicit anchors are no longer permitted by default.  I also tried to
747    make sure that nothing else still relies on implicit anchors.
748    (This is the next step on the schedule towards a CM that does not even
749    have the notion of implicit anchors anymore.)
750    
751    2. More CM manual updates.
752    
753    3. I managed to track down and fix the pickling bug I mentioned last
754    time.  Because of the previously existing workaround, this entails no
755    immediate practical changes.
756    
757    ----------------------------------------------------------------------
758    Name: Matthias Blume
759    Date: 2000/06/06 11:15:00 JST
760    Tag: blume-20000606-lazierpickle
761    Description:
762    
763    !!!! NEW BOOT FILES !!!!
764    
765    * The main purpose of this update is to make library pickles lazier in
766    order to reduce the initial space penalty for autoloading a library.
767    As a result, it is now possible to have $smlnj/compiler.cm
768    pre-registered.  This should take care of the many complaints or
769    inquiries about missing structure Compiler.  This required changes to
770    CM's internal data structures and small tweaks to some algorithms.
771    
772    As a neat additional effect, it is no longer necessary (for the sake
773    of lean heap image files) to distinguish between a "minimal" CM and a
774    "full" CM.  Now, there is only one CM (i.e., the "full" version:
775    $smlnj/cm.cm aka $smlnj/cm/full.cm), and it is always available at the
776    interactive top level. ($smlnj/cm/minimal.cm is gone.)
777    
778    To make the life of compiler-hackers easier, "makeml" now also
779    pre-registers $smlnj/cmb.cm (aka $smlnj/cmb/current.cm).  In other
780    words, after you bootstrap a new sml for the first time, you will not
781    have to autoload $smlnj/cmb.cm again afterwards.  (The first time
782    around you will still have to do it, though.)
783    
784    * A second change consists of major updates to the CM manual.  There
785    are now several appendices with summary information and also a full
786    specification of the CM description file syntax.
787    
788    * In directory src/system I added the script "allcross".  This script
789    invokes sml and cross-compiles the compiler for all supported
790    architectures.  (Useful when providing a new set of boot files.)
791    
792    * There seems to be a latent bug in my "lazy pickles" mechanism.  I
793    added a small tweak to pickle-util.sml to work around this problem,
794    but it is not a proper fix yet.  I will investigate further.  (The
795    effect of the bug was an inflation of library pickle size.)
796    
797    * Version number increased to 110.28.1 (to avoid compatibility problems).
798    
799    ----------------------------------------------------------------------
800    Name: Allen Leung
801    Date: 2000/05/25 17:28 EDT
802    Tag: leunga-20000525-ra
803    Description:
804    
805      Fixed a bug in freezing phase of the register allocator.
806    
807    ----------------------------------------------------------------------
808    Name: Allen Leung
809    Date: 2000/05/15 22:53 EDT
810    Tag: leunga-20000515-alpha-x86-ra
811    Description:
812    
813      1. Alpha
814    
815          Slight cleanup.  Removed the instruction SGNXL
816    
817      2. X86
818    
819          Added the following instructions to the instruction set:
820    
821            ROLx, RORx,
822            BTx, BTSx, BTLx, BTRx,
823            XCHGx, and variants with the LOCK prefix
824    
825      3. Register Allocation
826    
827          The module ra-rewrite-with-renaming has been improved.
828    
829      These have no effect on SML/NJ.
830    
831    ----------------------------------------------------------------------
832    Name: Matthias Blume
833    Date: 2000/05/15 16:20:00 JST
834    Tag: blume-20000515-lightrebuild
835    Description:
836    
837    1. I added an alternative to "-rebuild" to "makeml".  The difference is
838       that prior to calling CMB.make' the CM-variable "LIGHT" will be
839       defined.  In effect, the command will not build any cross-compiler
840       backends and therefore finish more quickly.
841    
842       The "fixpt" script also takes a "-light" switch to be able to use
843       this new facility while compiling for a fixpoint.
844    
845    2. I replaced all mentions of anchored paths in group owner specifications
846       with simple relative paths (usually starting with "..").
847       The rationale is that a library's internal workings should not be
848       compromised by the lack of some anchor.  (An anchor is necessary
849       for someone who wants to refer to the library by an anchored path,
850       but it should not be necessary to build the same library in the first
851       place.)
852    
853    3. I changed the way CM's tool mechanism determines the shell command
854       string used for things like ml-yacc etc. so that it does not break
855       when CM.Control.implicit_anchors is turned off.
856    
857    ----------------------------------------------------------------------
858    Name: Matthias Blume
859    Date: 2000/05/12 18:20:00 JST
860    Tag: blume-20000512-ml-build
861    Description:
862    
863    Fixed a bug in config/_ml-build that prevented ml-yacc and ml-lex from
864    getting installed properly (by config/install.sh).
865    
866    ----------------------------------------------------------------------
867    Name: Matthias Blume
868    Date: 2000/05/12 17:30:00 JST
869    Tag: blume-20000512-anchors
870    Description:
871    
872    !!! NEW BOOT FILES !!!
873    
874    This change is in preparation of fading out support for "implicitly
875    anchored path names".  I went through all sources and used the
876    explicit (and relatively new) $-notation.  See system/README and the
877    CM manual for more info on this.
878    
879    I also modified the anchoring scheme for some things such as "smlnj",
880    "MLRISC", "cm", etc. to take advantage of the fact that explicit
881    anchors are more expressive: anchor name and first arc do not have to
882    coincide.  This entails the following user-visible change:
883    
884    You have to write $smlnj/foo/bar instead of smlnj/foo/bar.  In
885    particular, when you fire up sml with a command-line argument, say,
886    e.g.:
887    
888       sml '$smlnj/cmb.cm'
889    
890    At the ML toplevel prompt:
891    
892       CM.autoload "$smlnj/cmb.cm";
893    
894    There is also a new controller in CM.Control that can be used to turn
895    off all remaining support for implicit anchors by saying:
896    
897        CM.autoload "$smlnj/
898        #set CM.Control.implicit_anchors false;
899    
900    This causes CM to reject implicitly anchored paths.  This is (for the
901    time being) less permissive than the "final" version where there will
902    be no more such implicit anchors and relative paths will be just that:
903    relative.
904    
905    The next step (version after next version?) will be to make the
906    default for CM.Control.implicit_anchors false.  After the dust has
907    settled, I can then produce the "final" version of this...
908    
909    Note: Since bootstrapping is a bit tricky, I provided new boot files.
910    
911    ----------------------------------------------------------------------
912    Name: Matthias Blume
913    Date: 2000/05/11 16:30:00 JST
914    Tag: blume-20000511-sources
915    Description:
916    
917    The main change is that I added function CM.sources as a generalized
918    version of the earlier CM.makedepend.  This entails the following
919    additional changes:
920    
921      - CM.makedepend has been dropped.
922    
923      - CM manual has been updated.
924    
925      - TOOLS signature and API have been changed.
926    
927    ----------------------------------------------------------------------
928    Name: Allen Leung
929    Date: 2000/05/10 21:17 EDT
930    Tag: leunga-20000510-moby-c--ssa
931    Description:
932    
933      Various bug fixes and new features for C--, Moby and MLRISC optimizations.
934    None of these affect SML/NJ.
935    
936    1. Register Allocation
937    
938        a. A new ra spilling module (ra/ra-spill-with-renaming) is implemented.
939           This module tries to remove local (i.e. basic block level) redundancies
940           during spilling.
941    
942        b. A new framework for performing region based register allocation.
943           Not yet entirely functional.
944    
945    2. X86
946    
947       a. DefUse for POP was missing the stack pointer [found by Lal]
948       b. Reload for CALL was incorrect in X86Spill [found by John]
949       c. Various fixes in X86Spill so that it can be used correctly for
950          the new spilling module.
951    
952    3. SSA/IR
953    
954       a. New module ir/dj-dataflow.sml implements elimination based
955          data flow analysis.
956    
957    4. MLRiscGen
958    
959       a. Fix for gc type annotation
960    
961    5. MDGen
962    
963       Various fixes for machine description -> ml code translation.  For ssa
964       only.
965    
966    ----------------------------------------------------------------------
967    Name: Allen Leung
968    Date: 2000/05/08 22:17 EDT
969    Tag: leunga-20000508-labexp
970    Description:
971    
972      Fermin has found a few assembly problems with constant expressions
973      generated in LabelExp.  Mostly, the problems involve extra parentheses,
974      which choke on dumb assemblers.  This is his fix.
975    
976    ----------------------------------------------------------------------
977    Name: Dave MacQueen
978    Date: 2000/04/09 14:00 EDT
979    Tag: dbm-20000502-Version_110_28
980    Description:
981    
982    1. Updated src/compiler/TopLevel/main/version.sml to version 110.28
983    
984    2. Updated config/version to 110.28
985    
986    3. Updated config/srcarchiveurl
987    
988    3. New boot files!
989       ftp://ftp.research.bell-labs.com/dist/smlnj/working/110.28/
990    
991    ----------------------------------------------------------------------
992    Name: Matthias Blume
993    Date: 2000/05/01 19:05:00 JST
994    Tag: blume-20000501-noweb
995    Description:
996    
997    A new noweb tool has been added.  The existing system is entirely
998    unaffected by this, but some CM users have asked for renewed noweb
999    support.  Everything is documented in the CM manual.
1000    
1001    New (plugin) libraries:
1002    
1003       noweb-tool.cm
1004       nw-ext.cm
1005    
1006    ----------------------------------------------------------------------
1007    Name: Dave MacQueen
1008    Date: 2000/04/30 12:40PM EDT
1009    Tag: dbm-20000430-bug_fixes
1010    Description:
1011    
1012    1. Fix for bug 1498
1013       smlnj/src/system/Basis/Implementation/Unsafe/object.sig
1014       smlnj/src/system/Basis/Implementation/Unsafe/object.sml
1015         added toRealArray function
1016       smlnj/src/compiler/MiscUtil/print/ppobj.sml
1017         added check for tag Obj.RealArray to array printing case in ppObj
1018    
1019    2. Fix for bug 1510
1020       smlnj/src/compiler/Semant/types/typesutil.sml
1021         fixed definition of dummyargs (used by equalTycon) so that
1022         dummy args are distinct types
1023    
1024    ----------------------------------------------------------------------
1025    Name: Matthias Blume
1026    Date: 2000/04/30 01:00:00 JST
1027    Tag: blume-20000430-versions
1028    Description:
1029    
1030    1. CM version numbering added.  This is an implementation of Lal's
1031       proposal for adding version numbers and version checking to .cm
1032       files.  Lal said that his proposal was just that -- a proposal.
1033       For the time being I went ahead and implemented it so that people
1034       can comment on it.  Everything is completely backward-compatible
1035       (except for the stable library format, i.e., new bootfiles!).
1036    
1037       As usual, see the CM manual for details.
1038    
1039    2. An alternative syntax for anchored paths has been implemented.
1040       Dave has recently voiced the same concerns that I had when I did
1041       this, so there should be some support.  My take is that eventually
1042       I will let support for the current syntax (where anchors are
1043       "implicit") fade out in favor of the new, explicit syntax.
1044       In order to be backward-compatible, both old and new syntax are
1045       currently supported.
1046    
1047       Again, see the CM manual for details.
1048    
1049    3. Parallel make is trying to be slightly smarter:  When the master
1050       process finds a "bottleneck", i.e., when there is only one
1051       compilation unit that can be compiled and everybody else is
1052       waiting on it, then it will simply compile it directly instead
1053       of clumsily telling one of the slaves to do it.
1054    
1055    4. Support for "unsharing" added.  This is necessary in order to be
1056       able to have two different versions of the same library running
1057       at the same time (e.g., for trying out a new MLRISC while still
1058       having the old MLRISC linked into the current compiler, etc.)
1059       See the CM manual.
1060    
1061    5. Simple "makedepend" functionality added for generating Makefile
1062       dependency information.  (This is rather crude at the moment.
1063       Expect some changes here in the future.)
1064    
1065    6. ".fun" added as a recognized suffix for ML files. Also documented
1066       explicitly in the manual that the fallback behavior (unknown suffix
1067       -> ML file) is not an official feature!
1068    
1069    7. Small changes to the pickler for stable libraries.
1070    
1071    8. Several internal changes to CM (for cleanup/improvement).
1072    
1073    
1074    !!!! NEW BINFILES !!!!
1075    
1076    ----------------------------------------------------------------------
1077    Name: Matthias Blume
1078    Date: 2000/04/28 17:30:00 JST
1079    Tag: blume-20000428-pathconfig
1080    Description:
1081    
1082    1. I changed config/install.sh to remove duplicate entries from the
1083       lib/pathconfig file at the end.  Moreover, the final version of
1084       lib/pathconfig is sorted alphabetically.  The same (sorting) is done
1085       in src/system/installml.
1086    
1087    2. The config/install.sh script now consistently uses relative
1088       pathnames in lib/pathconfig whenever the anchor is in the lib
1089       directory.  (So far this was true for the libraries that come
1090       pre-compiled and bundled as part of the bootfiles but not for
1091       libraries that are compiled by the script itself.)
1092    
1093    ----------------------------------------------------------------------
1094    Name: Matthias Blume
1095    Date: 2000/04/26 13:10:00 JST
1096    Tag: blume-20000426-fun_suffix
1097    Description:
1098    
1099    Added ".fun" as a recognized file name suffix (for ML code).
1100    
1101    ----------------------------------------------------------------------
1102    Name: Allen Leung
1103    Date: 2000/04/25 17:00:00 EST
1104    Tag: leunga-20000425-alpha-ra
1105    Description:
1106    
1107    1. Alpha
1108    
1109        PSEUDOARITH was missing in AlphaRewrite.  This causes an endless loop
1110    in C--.
1111    
1112    2. RA
1113    
1114       Added a flag "ra-dump-size" to print out the size of the flowgraph
1115       and the interference graph.
1116    
1117    ----------------------------------------------------------------------
1118    Name: Dave MacQueen
1119    Date: 2000/04/25/
1120    Tag: dbm-20000425-mlyacc_doc_examples
1121    Description:
1122      Updated mlyacc.tex sections 5 and 7 for SML '97 and CM.
1123      Updated all three examples in src/ml-yacc/examples to run
1124      under 110.* using CM.make.
1125    
1126    ----------------------------------------------------------------------
1127    Name: Allen Leung
1128    Date: 2000/04/20 23:04:00 EST
1129    Tag: leunga-20000420-ssa-c---stuff
1130    Description:
1131    
1132      This update synchronizes my repository with Yale's.  Most of these
1133    changes, however, do not affect SML/NJ at all (the RA is an exception).
1134    
1135    1. Register Allocator
1136    
1137       a. An improvement in the interference graph construction:
1138          Given a copy
1139    
1140                s <- t
1141    
1142          no interference edge between s and t is added for this definition of s.
1143    
1144       b. I've added two new spill heuristic modules that Fermin and I developed
1145          (in the new library RA.cm). These are unused in SML/NJ but maybe
1146          useful for others (Moby?)
1147    
1148    2. X86
1149    
1150       a. Various fixes in the backend provided by Fermin [C--] and Lal.
1151    
1152    3. Alpha
1153    
1154       a. Added the BSR instruction and code generation that goes with it [C--]
1155       b. Other fixes too numerous to recount provided by Fermin [C--]
1156    
1157    4. Regmaps
1158    
1159       a. The regmaps are not initialized with the identity physical bindings
1160          at creation time.  This is unneeded.
1161    
1162    5. MLRISC Optimizations
1163    
1164       a. The DJ-Graph module can now compute the iterated dominance frontiers
1165          intersects with liveness incrementally in linear time! Woohoo!
1166          This is now used in my new SSA construction algorithm.
1167    
1168       b. THe branch reorganization module is now smarter about linear chains of
1169          basic blocks.
1170    
1171    
1172    ----------------------------------------------------------------------
1173    Name: Matthias Blume
1174    Date: 2000/04/12 13:52:00 JST
1175    Tag: blume_main_v110p27_1
1176    Description:
1177    
1178    Changed install.sh script to handle archive files without version number
1179    and to use "boot.<arch>-<os>" instead of "sml.boot.<arch>-<os>" for the
1180    name of the boot file archive.
1181    
1182    ----------------------------------------------------------------------
1183    Name: Dave MacQueen
1184    Date: 2000/04/09 14:00 EDT
1185    Tag: dbm-20000410-Version_110_27
1186    Description:
1187    
1188    1. Updated src/compiler/TopLevel/main/version.sml to version 110.27
1189    
1190    2. Updated src/config/version to 110.27
1191    
1192    3. New boot files!
1193    
1194    ----------------------------------------------------------------------
1195    Name: Allen Leung
1196    Date: 2000/04/09 19:09:00 EST
1197    Tag: leunga-20000409-misc
1198    Description:
1199    
1200    1.  Yet another fix for x86 assembly for idivl, imull, mull and friends.
1201    
1202    2.  Miscellaneous improvements to MLRISC (unused in sml/nj)
1203    
1204    ----------------------------------------------------------------------
1205    Name: Stefan
1206    Date: 2000/04/07 10:00:00 EDT
1207    Tag: monnier-20000406-branch-handling
1208    Description:
1209    
1210    Improved handling of branches (mostly those generated from
1211    polymorphic equality), removed switchoff and changed the
1212    default optimization settings (more cpsopt and less flintopt).
1213    
1214    ----------------------------------------------------------------------
1215    Name: Allen Leung
1216    Date: 2000/04/06 01:30:00 EST
1217    Tag: leunga-20000406-peephole-x86-SSA-2
1218    Description:
1219    
1220       Forgot a few files.
1221    
1222    ----------------------------------------------------------------------
1223    Name: Allen Leung
1224    Date: 2000/04/06 00:36:00 EST
1225    Tag: leunga-20000406-peephole-x86-SSA
1226    Description:
1227    
1228    1.  New Peephole code
1229    
1230    2.  Minor improvement to X86 instruction selection
1231    
1232    3.  Various fixes to SSA and machine description -> code translator
1233    
1234    ----------------------------------------------------------------------
1235    Name: Matthias Blume
1236    Date: 2000/04/05 12:30:00 JST
1237    Tag: blume_main_v110p26p2_3
1238    Description:
1239    
1240    This update just merges three minor cosmetic updates to CM's sources
1241    to get ready for the 110.27 code freeze on Friday.  No functionality
1242    has changed.
1243    
1244    ----------------------------------------------------------------------
1245    Name: Allen Leung
1246    Date: 2000/04/04 19:39:00 EST
1247    Tag: leunga-20000404-x86-asm
1248    Description:
1249    
1250    1.  Fixed a problem in X86 assembly.
1251    
1252        Things like
1253    
1254           jmp %eax
1255           jmp (%eax)
1256    
1257        should be output as
1258    
1259           jmp *%eax
1260           jmp *(%eax)
1261    
1262    2.  Assembly output
1263    
1264          Added a new flag
1265    
1266              "asm-indent-copies" (default to false)
1267    
1268          When this flag is on, parallel copies will be indented an extra level.
1269    
1270    ----------------------------------------------------------------------
1271    Name: Allen Leung
1272    Date: 2000/04/04 03:18:00 EST
1273    Tag: leunga-20000404-C--Moby
1274    Description:
1275    
1276        All of these fixes are related to C--, Moby, and my own optimization
1277        stuff; so they shouldn't affect SML/NJ.
1278    
1279    1.  X86
1280    
1281        Various fixes related floating point, and extensions.
1282    
1283    2.  Alpha
1284    
1285        Some extra patterns related to loads with signed/zero extension
1286        provided by Fermin.
1287    
1288    3.  Assembly
1289    
1290        When generating assembly, resolve the value of client defined constants,
1291        instead of generating symbolic values.  This is controlled by the
1292        new flag "asm-resolve-constants", which is default to true.
1293    
1294    4.  Machine Descriptions
1295    
1296        a. The precedence parser was slightly broken when parsing infixr symbols.
1297        b. The type generalizing code had the bound variables reversed, resulting
1298           in a problem during arity raising.
1299        c. Various fixes in machine descriptions.
1300    
1301    ----------------------------------------------------------------------
1302    Name: Matthias Blume
1303    Date: 2000/04/03 16:05:00 JST
1304    Tag: blume_main_v110p26p2_2
1305    Description:
1306    
1307    I eliminated coreEnv from compInfo.  Access to the "Core" structure is
1308    now done via the ordinary static environment that is context to each
1309    compilation unit.
1310    
1311    To this end, I arranged that instead of "structure Core" as "structure
1312    _Core" is bound in the pervasive environment.  Core access is done via
1313    _Core (which can never be accidentally rebound because _Core is not a
1314    legal surface-syntax symbol).
1315    
1316    The current solution is much cleaner because the core environment is
1317    now simply part of the pervasive environment which is part of every
1318    compilation unit's context anyway.  In particular, this eliminates all
1319    special-case handling that was necessary until now in order to deal
1320    with dynamic and symbolic parts of the core environment.
1321    
1322    Remaining hackery (to bind the "magic" symbol _Core) is localized in the
1323    compilation manager's bootstrap compiler (actually: in the "init group"
1324    handling).  See the comments in src/system/smlnj/init/init.cmi for
1325    more details.
1326    
1327    I also tried to track down all mentions of "Core" (as string argument
1328    to Symbol.strSymbol) in the compiler and replaced them with a
1329    reference to the new CoreSym.coreSym.  Seems cleaner since the actual
1330    name appears in one place only.
1331    
1332    Binfile and bootfile format have not changed, but the switchover from
1333    the old "init.cmi" to the new one is a bit tricky, so I supplied new
1334    bootfiles anyway.
1335    
1336    ----------------------------------------------------------------------
1337    Name: Allen Leung
1338    Date: 2000/04/02 21:17:00 EST
1339    Tag: leunga-20000402-mltree
1340    Description:
1341    
1342       1. Renamed the constructor CALL in MLTREE by popular demand.
1343       2. Added a bunch of files from my repository.  These are currently
1344          used by other non-SMLNJ backends.
1345    
1346    ----------------------------------------------------------------------
1347    Name: Allen Leung
1348    Date: 2000/03/31 21:15:00 EST
1349    Tag: leunga-20000331-aliasing
1350    Description:
1351    
1352    This update contains a rewritten (and hopefully more correct) module
1353    for extracting aliasing information from CPS.
1354    
1355       To turn on this feature:
1356    
1357            Compiler.Control.CG.memDisambiguate := true
1358    
1359       To pretty print the region information with assembly
1360    
1361           Compiler.Control.MLRISC.getFlag "asm-show-region" := true;
1362    
1363       To control how many levels of aliasing information are printed, use:
1364    
1365           Compiler.Control.MLRISC.getInt "points-to-show-level" := n
1366    
1367       The default of n is 3.
1368    
1369    ----------------------------------------------------------------------
1370    Name: David MacQueen
1371    Date: 2000/03/31 11:15:00 EST
1372    Tag: dbm-20000331-runtime_fix
1373    Description:
1374    
1375    This update contains:
1376    
1377    1. runtime/c-lib/c-libraries.c
1378       includes added in revision 1.2 caused compilation errors on hppa-hpux
1379    
1380    2. fix for bug 1556
1381       system/Basis/Implementation/NJ/internal-signals.sml
1382    
1383    ----------------------------------------------------------------------
1384    Name: Matthias Blume
1385    Date: 2000/03/31 18:00:00 JST
1386    Tag: blume_main_v110p26p2_1
1387    Description:
1388    
1389    This update contains:
1390    
1391    1. A small change to CM's handling of stable libraries:
1392       CM now maintains one "global" modmap that is used for all stable
1393       libraries.  The use of such a global modmap maximizes sharing and
1394       minimizes the need for re-traversing parts of environments during
1395       modmap construction.  (However, this has minor impact since modmap
1396       construction seems to account for just one percent or less of total
1397       compile time.)
1398    
1399    2. I added a "genmap" phase to the statistics.  This is where I got the
1400       "one percent" number (see above).
1401    
1402    3. CM's new tool parameter mechanism just became _even_ better. :)
1403       - The parser understands named parameters and recursive options.
1404       - The "make" and "shell" tools use these new features.
1405         (This makes it a lot easier to cascade these tools.)
1406       - There is a small syntax change: named parameters use a
1407    
1408           <name> : ( <option> ... )            or
1409           <name> : <string>
1410    
1411         syntax.  Previously, named parameters were implemented in an
1412         ad-hoc fashion by each tool individually (by parsing strings)
1413         and had the form
1414    
1415           <name>=<string>
1416    
1417       See the CM manual for a full description of these issues.
1418    
1419    ----------------------------------------------------------------------
1420    Name: Matthias Blume
1421    Date: 2000/03/30 18:00:00 JST
1422    Tag: blume_main_v110p26p2_0
1423    Description:
1424    
1425    !!!!! WARNING !!!!!!
1426    !!  New binfiles  !!
1427    !!!!!!!!!!!!!!!!!!!!
1428    
1429    This update contains:
1430    
1431    1. Moderate changes to CM:
1432    
1433       - Changes to CM's tools mechanism.  In particular, it is now possible
1434       to have tools that accept additional "command line" parameters
1435       (specified in the .cm file at each instance where the tool's class is
1436       used).
1437    
1438       This was done to accommodate the new "make" and "shell" tools which
1439       facilitate fairly seamless hookup to portions of code managed using
1440       Makefiles or Shell scripts.
1441    
1442       There are no classes "shared" or "private" anymore.  Instead, the
1443       sharing annotation is now a parameter to the "sml" class.
1444    
1445       There is a bit of generic machinery for implementing one's own
1446       tools that accept command-line parameters.  However, I am not yet fully
1447       satisfied with that part, so expect changes here in the future.
1448    
1449       All existing tools are described in the CM manual.
1450    
1451       - Slightly better error handling.  (CM now suppresses many followup
1452       error messages that tended to be more annoying than helpful.)
1453    
1454    2. Major changes to the compiler's static environment data structures.
1455    
1456       - no CMStaticEnv anymore.
1457            - no CMEnv, no "BareEnvironment" (actually, _only_ BareEnvironment,
1458              but it is called Environment), no conversions between different
1459              kinds of static environments
1460    
1461       - There is still a notion of a "modmap", but such modmaps are generated
1462         on demand at the time when they are needed.  This sounds slow, but I
1463         sped up the code that generates modmaps enough for this not to lead to
1464         a slowdown of the compiler (at least I didn't detect any).
1465    
1466       - To facilitate rapid modmap generation, static environments now
1467         contain an (optional) "modtree" structure.  Modtree annotations are
1468         constructed by the unpickler during unpickling.  (This means that
1469         the elaborator does not have to worry about modtrees at all.)
1470         Modtrees have the advantage that they are compositional in the same
1471         way as the environment data structure itself is compositional.
1472         As a result, modtrees never hang on to parts of an environment that
1473         has already been rendered "stale" by filtering or rebinding.
1474    
1475       - I went through many, many trials and errors before arriving at the
1476         current solution.  (The initial idea of "linkpaths" did not work.)
1477         But the result of all this is that I have touched a lot of files that
1478         depend on the "modules" and "types" data structures (most of the
1479         elaborator). There were a lot of changes during my "linkpath" trials
1480         that could have been reverted to their original state but weren't.
1481         Please, don't be too harsh on me for messing with this code a bit more
1482         than what was strictly necessary...  (I _did_ resist the tempation
1483         of doing any "global reformatting" to avoid an untimely death at
1484         Dave's hands. :)
1485    
1486       - One positive aspect of the previous point:  At least I made sure that
1487         all files that I touched now compile without warnings (other than
1488         "polyEqual").
1489    
1490       - compiler now tends to run "leaner" (i.e., ties up less memory in
1491         redundant modmaps)
1492    
1493    ----------------------------------------------------------------------
1494    Name: Allen Leung
1495    Date: 2000/03/29 18:00:00
1496    Tag: leunga-20000327-mlriscGen_hppa_alpha_x86
1497    Boot files (optional): ftp://react-ilp.cs.nyu.edu/leunga/110.26.1-sml.boot.x86-unix-20000330.tar.gz
1498    Description:
1499    
1500       This update contains *MAJOR* changes to the way code is generated from CPS
1501    in the module mlriscGen, and in various backend modules.
1502    
1503    CHANGES
1504    =======
1505    
1506    1. MLRiscGen: forward propagation fix.
1507    
1508       There was a bug in forward propagation introduced at about the same time
1509       as the MLRISC x86 backend, which prohibits coalescing to be
1510       performed effectively in loops.
1511    
1512       Effect: speed up of loops in RISC architectures.
1513               By itself, this actually slowed down certain benchmarks on the x86.
1514    
1515    2. MLRiscGen:  forward propagating addresses from consing.
1516    
1517       I've changed the way consing code is generated.  Basically I separated
1518       out the initialization part:
1519    
1520            store tag,   offset(allocptr)
1521            store elem1, offset+4(allocptr)
1522            store elem2, offset+8(allocptr)
1523            ...
1524            store elemn, offset+4n(allocptr)
1525    
1526       and the address computation part:
1527    
1528            celladdr <- offset+4+alloctpr
1529    
1530       and move the address computation part
1531    
1532       Effect:  register pressure is generally lower as a result.  This
1533                makes compilation of certain expressions much faster, such as
1534                long lists with non-trivial elements.
1535    
1536                 [(0,0), (0,0), .... (0,0)]
1537    
1538    3. MLRiscGen: base pointer elimination.
1539    
1540        As part of the linkage mechanism, we generate the sequence:
1541    
1542         L:  ...  <- start of the code fragment
1543    
1544         L1:
1545             base pointer <- linkreg - L1 + L
1546    
1547         The base pointer was then used for computing relocatable addresses
1548       in the code fragment.  Frequently (such as in lots of continuations)
1549       this is not needed.  We now eliminate this sequence whenever possible.
1550    
1551         For compile time efficiency, I'm using a very stupid local heuristic.
1552       But in general, this should be done as a control flow analysis.
1553    
1554       Effect:  Smaller code size.  Speed up of most programs.
1555    
1556    4. Hppa back end
1557    
1558         Long jumps in span dependence resolution used to depend on the existence
1559      of the base pointer.
1560    
1561         A jump to a long label L was expanded into the following sequence:
1562    
1563          LDIL %hi(L-8192), %r29
1564          LDO  %lo(L-8192)(%r29), %r29
1565          ADD  %r29, baseptr, %r29
1566          BV,n %r0(%r29)
1567    
1568         In the presence of change (3) above, this will not work.  I've changed
1569       it so that the following sequence of instructions are generated, which
1570       doesn't mention the base pointer at all:
1571    
1572             BL,n  L', %r29           /* branch and link, L' + 4 -> %r29 */
1573        L':  ADDIL L-(L'+4), %r29     /* Compute address of L */
1574             BV,n  %r0(%r29)          /* Jump */
1575    
1576    5. Alpha back end
1577    
1578          New alpha instructions LDB/LDW have been added, as per Fermin's
1579       suggestions.   This is unrelated to all other changes.
1580    
1581    6. X86 back end
1582    
1583         I've changed andl to testl in the floating point test sequence
1584         whenever appropriate.  The Intel optimization guide states that
1585         testl is preferable to andl.
1586    
1587    7. RA (x86 only)
1588    
1589         I've improved the spill propagation algorithm, using an approximation
1590       of maximal weighted independent sets.   This seems to be necessary to
1591       alleviate the negative effect in light of the slow down in (1).
1592    
1593         I'll write down the algorithm one of these days.
1594    
1595    8. MLRiscGen: frequencies
1596    
1597         I've added an annotation that states that all call gc blocks have zero
1598       execution frequencies.  This improves register allocation on the x86.
1599    
1600    BENCHMARKS
1601    ==========
1602    
1603       I've only perform the comparison on 110.25.
1604    
1605       The platforms are:
1606    
1607        HPPA  A four processor HP machine (E9000) with 5G of memory.
1608        X86   A 300Hhz Pentium II with 128M of memory, and
1609        SPARC An Ultra sparc 2 with 512M of memory.
1610    
1611       I used the following parameters for the SML benchmarks:
1612    
1613                 @SMLalloc
1614         HPPA    256k
1615         SPARC   512k
1616         X86     256k
1617    
1618    COMPILATION TIME
1619    ----------------
1620       Here are the numbers comparing the compilation times of the compilers.
1621       I've only compared 110.25 compiling the new sources versus
1622       a fixpoint version of the new compiler compiling the same.
1623    
1624                     110.25                                  New
1625               Total  Time in RA  Spill+Reload   Total  Time In RA Spill+Reload
1626         HPPA   627s    116s        2684+3584     599s    95s       1003+1879
1627         SPARC  892s    173s        2891+3870     708s    116s      1004+1880
1628         X86    999s    315s       94006+130691   987s    296s    108877+141957
1629    
1630                   110.25         New
1631                Code Size      Code Size
1632         HPPA   8596736         8561421
1633         SPARC  8974299         8785143
1634         X86    9029180         8716783
1635    
1636       So in summary, things are at least as good as before.   Dramatic
1637       reduction in compilation is obtained on the Sparc; I can't explain it,
1638       but it is reproducible.  Perhaps someone should try to reproduce this
1639       on their own machines.
1640    
1641    SML BENCHMARKS
1642    --------------
1643    
1644        On the average, all benchmarks perform at least as well as before.
1645    
1646          HPPA         Compilation Time     Spill+Reload      Run Time
1647                     110.25  New            110.25    New   110.25  New
1648    
1649          barnesHut  3.158  3.015  4.75%    1+1       0+0   2.980  2.922   2.00%
1650              boyer  6.152  5.708  7.77%    0+0       0+0   0.218  0.213   2.34%
1651       count-graphs  1.168  1.120  4.32%    0+0       0+0  22.705 23.073  -1.60%
1652                fft  0.877  0.792 10.74%    1+3       1+3   0.602  0.587   2.56%
1653        knuthBendix  3.180  2.857 11.32%    0+0       0+0   0.675  0.662   2.02%
1654             lexgen  6.190  5.290 17.01%    0+0       0+0   0.913  0.788  15.86%
1655               life  0.803  0.703 14.22%   25+25      0+0   0.153  0.140   9.52%
1656              logic  2.048  2.007  2.08%    6+6       1+1   4.133  4.008   3.12%
1657         mandelbrot  0.077  0.080 -4.17%    0+0       0+0   0.765  0.712   7.49%
1658             mlyacc 22.932 20.937  9.53%  154+181    32+57  0.468  0.430   8.91%
1659            nucleic  5.183  5.060  2.44%    2+2       0+0   0.125  0.120   4.17%
1660      ratio-regions  3.357  3.142  6.84%    0+0       0+0  116.225 113.173 2.70%
1661                ray  1.283  1.290 -0.52%    0+0       0+0   2.887  2.855   1.11%
1662             simple  6.307  6.032  4.56%   28+30      5+7   3.705  3.658   1.28%
1663                tsp  0.888  0.862  3.09%    0+0       0+0   7.040  6.893   2.13%
1664               vliw 24.378 23.455  3.94%  106+127    25+45  2.758  2.707   1.91%
1665      --------------------------------------------------------------------------
1666       Average                     6.12%                                   4.09%
1667    
1668          SPARC        Compilation Time     Spill+Reload      Run Time
1669                     110.25  New            110.25    New   110.25  New
1670    
1671          barnesHut  3.778  3.592  5.20%    2+2       0+0   3.648  3.453    5.65%
1672              boyer  6.632  6.110  8.54%    0+0       0+0   0.258  0.242    6.90%
1673       count-graphs  1.435  1.325  8.30%    0+0       0+0  33.672 34.737   -3.07%
1674                fft  0.980  0.940  4.26%    3+9       2+6   0.838  0.827    1.41%
1675        knuthBendix  3.590  3.138 14.39%    0+0       0+0   0.962  0.967   -0.52%
1676             lexgen  6.593  6.072  8.59%    1+1       0+0   1.077  1.078   -0.15%
1677               life  0.972  0.868 11.90%   26+26      0+0   0.143  0.140    2.38%
1678              logic  2.525  2.387  5.80%    7+7       1+1   5.625  5.158    9.05%
1679         mandelbrot  0.090  0.093 -3.57%    0+0       0+0   0.855  0.728   17.39%
1680             mlyacc 26.732 23.827 12.19%  162+189    32+57  0.550  0.560   -1.79%
1681            nucleic  6.233  6.197  0.59%    3+3       0+0   0.163  0.173   -5.77%
1682      ratio-regions  3.780  3.507  7.79%    0+0       0+0 133.993 131.035   2.26%
1683                ray  1.595  1.550  2.90%    1+1       0+0   3.440  3.418    0.63%
1684             simple  6.972  6.487  7.48%   29+32      5+7   3.523  3.525   -0.05%
1685                tsp  1.115  1.063  4.86%    0+0       0+0   7.393  7.265    1.77%
1686               vliw 27.765 24.818 11.87%  110+135    25+45  2.265  2.135    6.09%
1687      ----------------------------------------------------------------------------
1688       Average                     6.94%                                    2.64%
1689    
1690          X86          Compilation Time     Spill+Reload      Run Time
1691                     110.25  New            110.25    New   110.25  New
1692    
1693          barnesHut  5.530  5.420  2.03%  593+893   597+915   3.532  3.440   2.66%
1694              boyer  8.768  7.747 13.19%  493+199   301+289   0.327  0.297  10.11%
1695       count-graphs  2.040  2.010  1.49%  298+394   315+457  26.578 28.660  -7.26%
1696                fft  1.327  1.302  1.92%  112+209   115+210   1.055  0.962   9.71%
1697        knuthBendix  5.218  5.475 -4.69%  451+598   510+650   0.928  0.932  -0.36%
1698             lexgen  9.970  9.623  3.60% 1014+841  1157+885   0.947  0.928   1.97%
1699               life  1.183  1.183  0.00%  162+182   145+148   0.127  0.103  22.58%
1700              logic  3.285  3.512 -6.45%  514+684   591+836   5.682  5.577   1.88%
1701         mandelbrot  0.147  0.143  2.33%   38+41     33+54    0.703  0.690   1.93%
1702             mlyacc 35.457 32.763  8.22% 3496+4564 3611+4860  0.552  0.550   0.30%
1703            nucleic  7.100  6.888  3.07%  239+168   201+158   0.175  0.173   0.96%
1704      ratio-regions  6.388  6.843 -6.65% 1182+257   981+300  120.142 120.345 -0.17%
1705                ray  2.332  2.338 -0.29%  346+398   402+494   3.593  3.540   1.51%
1706             simple  9.912  9.903  0.08% 1475+941  1579+1168  3.057  3.178  -3.83%
1707                tsp  1.623  1.532  5.98%  266+200   250+211   8.045  7.878   2.12%
1708               vliw 33.947 35.470 -4.29% 2629+2774 2877+3171  2.072  1.890   9.61%
1709      ----------------------------------------------------------------------------
1710       Average                     1.22%                                     3.36%
1711    
1712    ----------------------------------------------------------------------
1713    Name: Allen Leung
1714    Date: 2000/03/23 16:25:00
1715    Tag: leunga-20000323-fix_x86_alpha
1716    Description:
1717    
1718    1. X86 fixes/changes
1719    
1720       a.  The old code generated for SETcc was completely wrong.
1721           The Intel optimization guide is VERY misleading.
1722    
1723    2. ALPHA fixes/changes
1724    
1725       a.  Added the instructions LDBU, LDWU, STB, STW as per Fermin's suggestion.
1726       b.  Added a new mode byteWordLoadStores to the functor parameter to Alpha()
1727       c.  Added reassociation code for address computation.
1728    
1729    ----------------------------------------------------------------------
1730    Name: Allen Leung
1731    Date: 2000/03/22 01:23:00
1732    Tag: leunga-20000322-fix_x86_hppa_ra
1733    Description:
1734    
1735    1. X86 fixes/changes
1736    
1737       a.  x86Rewrite bug with MUL3 (found by Lal)
1738       b.  Added the instructions FSTS, FSTL
1739    
1740    2. PA-RISC fixes/changes
1741    
1742       a.  B label should not be a delay slot candidate!  Why did this work?
1743       b.  ADDT(32, REG(32, r), LI n) now generates one instruction instead of two,
1744           as it should be.
1745       c.  The assembly syntax for fstds and fstdd was wrong.
1746       d.  Added the composite instruction COMICLR/LDO, which is the immediate
1747           operand variant of COMCLR/LDO.
1748    
1749    3. Generic MLRISC
1750    
1751       a.  shuffle.sml rewritten to be slightly more efficient
1752       b.  DIV bug in mltree-simplify fixed (found by Fermin)
1753    
1754    4. Register Allocator
1755    
1756       a.  I now release the interference graph earlier during spilling.
1757           May improve memory usage.
1758    
1759    ----------------------------------------------------------------------
1760    Name: Matthias Blume
1761  Date: 2000/03/14 14:15:32  Date: 2000/03/14 14:15:32
1762  Tag: blume_main_v110p26p1_2  Tag: blume_main_v110p26p1_2
1763  Description:  Description:
1764    
1765  1. Tools.registerStdShellCmdTool (from smlnj/cm/tool.cm) takes an  1. Tools.registerStdShellCmdTool (from smlnj/cm/tool.cm) takes an
1766  additional argument called "template" which is an optional string that  additional argument called "template" which is an optional string that
1767  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
1768  explanation.  explanation.
1769    
1770  2. A special-purpose tool can be "regisitered" by simply dropping the  2. A special-purpose tool can be "registered" by simply dropping the
1771  corresponding <...>-tool.cm (and/or <...>-ext.cm) into the same  corresponding <...>-tool.cm (and/or <...>-ext.cm) into the same
1772  directory where the .cm file lives that uses this tool.  (The  directory where the .cm file lives that uses this tool.  (The
1773  behavior/misfeature until now was to look for the tool description  behavior/misfeature until now was to look for the tool description
# Line 66  Line 1811 
1811  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
1812  <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,
1813  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
1814  perhaps got accidentially killed).  In this case, fixpt will quickly  perhaps got accidentally killed).  In this case, fixpt will quickly
1815  move through what exists before continuing where it left off earlier,  move through what exists before continuing where it left off earlier,
1816  and, thus, saves a lot of time.  and, thus, saves a lot of time.
1817    
# Line 116  Line 1861 
1861    it from that remote directory.    it from that remote directory.
1862    This should simplify installation further:  For machines that have    This should simplify installation further:  For machines that have
1863    access to the internet, just fetch <version>-config.tgz, unpack it,    access to the internet, just fetch <version>-config.tgz, unpack it,
1864    edit config/targets, and go (run config/install.sh).  The scipt will    edit config/targets, and go (run config/install.sh).  The script will
1865    fetch everything else that it might need all by itself.    fetch everything else that it might need all by itself.
1866    
1867    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.578  
changed lines
  Added in v.701

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