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 928, Tue Sep 18 13:38:46 2001 UTC revision 1855, Wed Nov 2 21:12:31 2005 UTC
# Line 8  Line 8 
8  The form of an entry should be:  The form of an entry should be:
9    
10  Name:  Name:
11  Date:  Date: yyyy/mm/dd
12  Tag: <post-commit CVS tag>  Tag: <post-commit CVS tag>
13  Description:  Description:
14    
15  ----------------------------------------------------------------------  ----------------------------------------------------------------------
16    Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
17    Date: 2005/11/02
18    Description:
19    Fixed a Basis incompatibility: Pack<N>{Big,Little} structures should
20    be called PackWord<N>{Big,Little}.
21    
22    ----------------------------------------------------------------------
23    Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
24    Date: 2005/10/28
25    Description:
26    Fixed a minor Basis incompatibility: hex digits should be upper case.
27    
28    ----------------------------------------------------------------------
29    Name: Matthias Blume (blume (at) tti - c (dot) org)
30    Date: 2005/10/25 16:00:00 CDT
31    Tag: Release_110_56
32    Description:
33    
34    New working version (110.56).  NEW BOOTFILES!
35    
36    ----------------------------------------------------------------------
37    Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
38    Date: 2005/10/25
39    Description:
40    Added interval sets to utility library (signatures INTERVAL_DOMAIN
41    and INTERVAL_SET, and functor IntervalSetFn).
42    
43    ----------------------------------------------------------------------
44    Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
45    Date: 2005/10/14
46    Description:
47    Add Zhong Shao's fix for datatype equality functions.
48    
49    ----------------------------------------------------------------------
50    Name: Allen Leung (leunga (at) reservoir (dot) com)
51    Date: 2005/10/14 12:30:00 EST
52    Tag: leunga-20051014-x86-spill
53    Description:
54    
55    Bug fix a bug found by Carl Hauser.
56    There was a typo in the reload code for FCMP in x86SpillInstr.sml.
57    
58    ----------------------------------------------------------------------
59    Name: Allen Leung (leunga (at) reservoir (dot) com)
60    Date: 2005/10/14 11:15:00 EST
61    Tag: leunga-20051014-x86-asm
62    Description:
63    
64    Removed some debugging code in file x86Asm.sml.
65    The function emit_operand was printing out debugging output.
66    
67    ----------------------------------------------------------------------
68    Name: John Reppy
69    Date: 2005/07/27
70    Tag:
71    Description:
72    
73    Fixed ml-lex to recognize "\r" as representing carriage return.
74    
75    ----------------------------------------------------------------------
76    Name: John Reppy
77    Date: 2005/07/27
78    Tag:
79    Description:
80    
81    Fixed ml-yacc to work on files with non-native end-of-line
82    encodings (e.g., Windows text file on a Unix system).
83    
84    ----------------------------------------------------------------------
85    Name: John Reppy
86    Date: 2005/07/20
87    Tag:
88    Description:
89    
90    Added changes from Dominic Evans (oldmanuk (at) gmail (dot) com)
91    to support HPUX 11.
92    
93    ----------------------------------------------------------------------
94    Name: John Reppy
95    Date: 2005/07/06
96    Tag:
97    Description:
98    
99    Changes to the SML/NJ library.  See smlnj-lib/CHANGES for details.
100    
101    ----------------------------------------------------------------------
102    Name: Matthias Blume (blume (at) tti - c (dot) org)
103    Date: 2005/07/06 00:45:00 CDT
104    Tag: blume-20050706-slice-copy
105    Description:
106    
107    Fixed reversed logic for deciding whether to "copy up" or "copy down"
108    in *-array-slice.sml.
109    
110    ----------------------------------------------------------------------
111    Name: Allen Leung (leunga (at) reservoir (dot) com)
112    Date: 2005/05/31 17:00:00 EST
113    Tag: leunga-20050531-cygwin-fault-2
114    Description:
115    
116    A typo in the cygwin code fixed.
117    
118    ----------------------------------------------------------------------
119    Name: Allen Leung (leunga (at) reservoir (dot) com)
120    Date: 2005/05/31 16:47:00 EST
121    Tag: leunga-20050531-cygwin-fault
122    Description:
123    
124    Updated Cygwin's fault/signal handling to match the Windows version.
125    Updated the export list.
126    
127    ----------------------------------------------------------------------
128    Name: Matthias Blume (blume (at) tti - c (dot) org)
129    Date: 2005/05/18 13:35:00 CDT
130    Tag: Release_110_54
131    Description:
132    
133    New working version (110.54).  NEW BOOTFILES!
134    
135    ----------------------------------------------------------------------
136    Name: Matthias Blume (blume (at) tti - c (dot) org)
137    Date: 2005/05/18 11:58:00 CDT
138    Tag: blume-20050518-installer
139    Description:
140    
141    Added support scripts for Mac OS X PackageMaker and modified
142    config/install.sh so that it supports re-dumping a heap image
143    after customization.
144    
145    ----------------------------------------------------------------------
146    Name: Matthias Blume (blume (at) tti - c (dot) org)
147    Date: 2005/05/18 10:55:00 CDT
148    Tag: blume-20050518-realdiv-noovld
149    Description:
150    
151    Un-overloaded / to work around bug in overloading resolution code.
152    
153    ----------------------------------------------------------------------
154    Name: Matthias Blume (blume (at) tti - c (dot) org)
155    Date: 2005/05/16 23:50:00 CDT
156    Tag: blume-20050516-redump-heap
157    Description:
158    
159    Added mechanism for re-creating a heap file for the interactive system
160    after configuration variables have been changed.
161    
162       CM.redump_heap : string -> unit
163    
164    This is much like SMLofNJ.exportML, but starting from the resulting
165    heap does not return to the caller of CM.redump_heap but
166    restarts the interactive system from scratch.  The original call of
167    CM.redump_heap does not return but ends the interactive session.
168    Thus, CM.redump_heap is a lot like SMLofNJ.exportFn.
169    
170    Internally, redump_heap winds the dynamic execution context back to
171    the point where the original heap image was created and re-executes
172    the heap image generation code in the boot code.
173    
174    ----------------------------------------------------------------------
175    Name: Matthias Blume (blume (at) tti - c (dot) org)
176    Date: 2005/05/09 21:55:00 CDT
177    Tag: blume-20050509-word64
178    Description:
179    
180    Added a hack to the existing hack known as Word64 to make fromString
181    behave correctly.  I am still not sure whether Word64.scan will work
182    as specified with respect to the interaction of radix and prefix.
183    
184    ----------------------------------------------------------------------
185    Name: Allen Leung (leunga (at) reservoir (dot) com)
186    Date: 2005/05/04 11:50:00 EST
187    Tag: leunga-20050504-checkgc
188    Description:
189    
190    Added a gc protocol checking phase.  This phase is enabled with
191    the flag "check-gc".  "debug-check-gc" turns on the verbose mode.
192    
193    ----------------------------------------------------------------------
194    Name: Matthias Blume (blume (at) tti - c (dot) org)
195    Date: 2005/05/04 10:45:00 CDT
196    Tag: blume-20050504-intinf
197    Description:
198    
199    Fixed a bug in the implementation of div and mod for IntInf.
200    Thanks to Neophytos Michael for reporting the problem.
201    
202    ----------------------------------------------------------------------
203    Name: Matthias Blume (blume (at) tti - c (dot) org)
204    Date: 2005/05/04 10:35:00 CDT
205    Tag: blume-20050504-join
206    Description:
207    
208    Added a "join" combinator to the ParserComb module in smlnj-lib.cm.
209    
210    ----------------------------------------------------------------------
211    Name: Matthias Blume (blume (at) tti - c (dot) org)
212    Date: 2005/02/28 23:40:00 CST
213    Tag: blume-20050228-mVar
214    Description:
215    
216    Fixed serious bug (brown paper bag variety) in new implementation of
217    structure Atom in CML.  (I had accidentally used a mailbox instead of
218    an mvar, leaving the door open for races.)
219    
220    ----------------------------------------------------------------------
221    Name: Matthias Blume (blume (at) tti - c (dot) org)
222    Date: 2005/02/25 15:00:00 CST
223    Tag: Release_110_53
224    Description:
225    
226    New working version (110.53).  NEW BOOTFILES!
227    
228    ----------------------------------------------------------------------
229    Name: Matthias Blume (blume (at) tti - c (dot) org)
230    Date: 2005/02/25 14:50:00 CST
231    Tag: blume-20050225-susp
232    Description:
233    
234    Brought back SMLofNJ.Susp.  The underlying suspension type is the one
235    implemented in Core -- which means that it is the same as the one used
236    by the lazy extension.
237    
238    ----------------------------------------------------------------------
239    Name: Matthias Blume (blume (at) tti - c (dot) org)
240    Date: 2005/02/24 16:50:00 CST
241    Tag: blume-20050224-cml-atom
242    Description:
243    
244    Simpler and at the same time more general implementation of structure
245    Atom in CML.
246    
247    ----------------------------------------------------------------------
248    Name: Matthias Blume (blume (at) tti - c (dot) org)
249    Date: 2005/02/15 17:35:00 CST
250    Tag: blume-20050215-tools
251    Description:
252    
253    Created new "tools" directory under "src" and moved "TraceDebugProf"
254    there.
255    
256    ----------------------------------------------------------------------
257    Name: Matthias Blume (blume (at) tti - c (dot) org)
258    Date: 2005/02/10 17:55:00 CST
259    Tag: blume-20050210-longlong
260    Description:
261    
262    Implemented "long long" arguments and results for NLFFI.  (Only the
263    PPC/MacOS implementation is complete, the other backends still need to
264    be updated.)
265    
266    ----------------------------------------------------------------------
267    Name: Matthias Blume (blume (at) tti - c (dot) org)
268    Date: 2005/01/24 17:40:00 CST
269    Tag: blume-20050124-mlyacc
270    Description:
271    
272    Minor cleanup in ML-Yacc rule printing mechanism.  This should fix a
273    problem with certain "as" patterns which previously got rendered
274    using incorrect syntax.
275    
276    ----------------------------------------------------------------------
277    Name: Matthias Blume (blume (at) tti - c (dot) org)
278    Date: 2005/01/18 12:00:00 CST
279    Tag: blume-20050118-profile
280    Description:
281    
282    Made time profiling code (interrupt handler) in runtime system aware
283    of new array representation.
284    
285    ----------------------------------------------------------------------
286    Name: Matthias Blume (blume (at) tti - c (dot) org)
287    Date: 2005/01/14 18:00:00 CST
288    Tag: blume-20050114-heap2exec
289    Description:
290    
291    Implemented new (but still experimental) heap2exec facility.  This is
292    tested under Mac OS X and should work under Linux (will test shortly).
293    It will probably also work on the Sparc (will test some time later).
294    
295      - removed old "HACKED_STANDALONE" hack from runtime
296    
297    To be able to test this, uncomment the request for "heap2asm" in
298    config/targets prior to installation.  (Notice that this is different
299    from "heap2exec" mentioned below.  Not a typo.)
300    
301    To perform an actual test, run the command
302    
303       $ bin/heap2exec heapfile execfile
304    
305    (You can put heap2exec on your shell's path.)
306    
307    For example, run
308    
309      $ bin/heap2exec bin/.heap/ml-yacc.ppc-darwin mly
310    
311    This will create a standalone executable called "mly" which you can
312    then invoke directly as a command.
313    
314    ----------------------------------------------------------------------
315    Name: Matthias Blume (blume (at) tti - c (dot) org)
316    Date: 2005/01/07 11:44:00 CST
317    Tag: blume-20050107-mlstring
318    Description:
319    
320    fixed off-by-one error in ML_STRING macro (globals.c)
321    
322    ----------------------------------------------------------------------
323    Name: Matthias Blume (blume (at) tti - c (dot) org)
324    Date: 2004/12/23 18:00:00 CST
325    Tag: blume-20041223-santa
326    Description:
327    
328    Made ml-build script "smarter" (but only very little).
329    
330    ----------------------------------------------------------------------
331    Name: Matthias Blume (blume (at) tti - c (dot) org)
332    Date: 2004/12/21 15:05:00 CST
333    Tag: blume-20041221-longlong
334    Description:
335    
336    * Implemented access to signed and unsigned long long data in NLFFI.
337    (The parameter-passing part of the picture has not complete. But
338    data structure access seems to work.)
339    
340    * Fixed CM's incorrect assumption that the PPC is little-endian.
341      (On the Mac, it is big-endian.  And that's currently our only
342      PPC platform.)
343    
344    ----------------------------------------------------------------------
345    Name: Matthias Blume (blume (at) tti - c (dot) org)
346    Date: 2004/12/21 12:50:00 CST
347    Tag: blume-20041221-memory
348    Description:
349    
350    Some cleanup in the $c/memory.cm library: separated some concerns by
351    moving allocation code and memory access code each into their own
352    files.
353    
354    ----------------------------------------------------------------------
355    Name: Allen Leung (leunga (at) reservoir (dot) com)
356    Date: 2004/12/17 16:12:00 EST
357    Tag: leunga-20041217-cygwin-smlnj-home
358    Description:
359    
360        The Unix I/O library of SML/NJ on cygwin does not understand
361    Windows style pathname, so problems arise when SMLNJ_HOME is set
362    to a Windows style pathname.  _run-sml now converts SMLNJ_HOME
363    to a POSIX pathname on cygwin.
364    
365    ----------------------------------------------------------------------
366    Name: Matthias Blume (blume (at) tti - c (dot) org)
367    Date: 2004/12/16 13:35:00 CST
368    Tag: Release_110_52
369    Description:
370    
371    Last-minute changes incorporated into 110.52.  Release tag moved.
372    
373    The changes:
374       - HashString.hashString' -> HashString.hashSubstring
375       - bug fix in UnivariateStats
376    
377    ----------------------------------------------------------------------
378    Name: Matthias Blume (blume (at) tti - c (dot) org)
379    Date: 2004/12/15 23:40:00 CST
380    Tag: blume-20041215-hashSubstring
381    Description:
382    
383    - HashString.hashString' -> HashString.hashSubstring
384    - corresponding changes in atom.sml
385    - "de-compressed" (aka. un-obfuscated) code for UnivariateStats and
386      added some comments
387    
388    ----------------------------------------------------------------------
389    Name: Matthias Blume (blume (at) tti - c (dot) org)
390    Date: 2004/12/15 15:30:00 CST
391    Tag: (Release_110_52)
392    Description:
393    
394    New working version (110.52).  NEW BOOTFILES!
395    
396    ----------------------------------------------------------------------
397    Name: Matthias Blume (blume (at) tti - c (dot) org)
398    Date: 2004/12/15 12:45:00 CST
399    Tag: blume-20041215-spaces
400    Description:
401    
402    More on the space problem (this time for Win32).
403    
404    ----------------------------------------------------------------------
405    Name: Matthias Blume (blume (at) tti - c (dot) org)
406    Date: 2004/12/14 17:30:00 CST
407    Tag: blume-20041214-spaces
408    Description:
409    
410    Hacked some of the scripts (in particular: the installer) to cope with
411    spaces in filenames a bit better.  But beware: the current "solution"
412    is likely still full of bugs and inherently incomplete.  (We need to
413    do away with those shell scripts for a comprehensive solution.)
414    
415    ----------------------------------------------------------------------
416    Name: Matthias Blume (blume (at) tti - c (dot) org)
417    Date: 2004/12/13 14:45:00 CST
418    Tag: blume-20041213-ml-makedepend
419    Description:
420    
421    Fixed bug in code for ml-makedepend.
422    
423    ----------------------------------------------------------------------
424    Name: Matthias Blume (blume (at) tti - c (dot) org)
425    Date: 2004/12/09 16:30:00 CST
426    Tag: blume-20041209-statistics
427    Description:
428    
429    Added two simple but potentially useful statistics modules to SML/NJ Library.
430    (See CHANGES file there.)
431    
432    ----------------------------------------------------------------------
433    Name: Matthias Blume (blume (at) tti - c (dot) org)
434    Date: 2004/12/01 16:50:00 CST
435    Tag: blume-20041201-atom
436    Description:
437    
438    smlnj-lib:
439    
440      Added function HashString.hashString' for substrings.
441      Hand-inlined CharVector.foldl into HashString (for speed).
442      Modified implementation of structure Atom to avoid extracting
443      strings from substrings unless necessary.
444      (Also see CHANGES file for smlnj-lib.)
445    
446    ----------------------------------------------------------------------
447    Name: Matthias Blume (blume (at) tti - c (dot) org)
448    Date: 2004/11/24 22:15:00 CST
449    Tag: blume-20041124-cml
450    Description:
451    
452    Made sure CML compiles when Position = Int64.
453    
454    ----------------------------------------------------------------------
455    Name: Matthias Blume (blume (at) tti - c (dot) org)
456    Date: 2004/11/24 14:30:00 CST
457    Tag: blume-20041124-position
458    Description:
459    
460    The compiler can now be compiled in a mode that makes structure Position
461    equal to Int64.  The default, however, is unchanged (Position = Int31)
462    for the time being.
463    
464    To enable 64-bit positions, use the following procedure:
465    
466    1. Start sml
467    2. Autoload $smlnj/cmb.cm (if not already autoloaded)
468    3. Type
469          #set (CMB.symval "USE_64_BIT_POSITIONS") (SOME 1);
470    4. Run CMB.make() as usual.
471    
472    This is barely tested.  The only test so far was a little SML program
473    counting the number of characters in an 8-gigabyte file by
474    reading it character-by-character.  That test was successful.
475    
476    In support of 64-bit positions, a number of new functions have been
477    added to the runtime system.
478    
479    ----------------------------------------------------------------------
480    Name: Matthias Blume (blume (at) tti - c (dot) org)
481    Date: 2004/11/23 14:45:00 CST
482    Tag: blume-20041123-useFile
483    Description:
484    
485    Fixed a problem with unhelpful error messages related to problems with
486    .cm- or .sml files that appear as part of the sml command line.
487    
488    ----------------------------------------------------------------------
489    Name: Matthias Blume (blume (at) tti - c (dot) org)
490    Date: 2004/11/18 15:40:00 CST
491    Tag: Release_110_51
492    Description:
493    
494    New working version (110.51).  NEW BOOTFILES!
495    
496    ----------------------------------------------------------------------
497    Name: Matthias Blume (blume (at) tti - c (dot) org)
498    Date: 2004/11/18 15:35:00 CST
499    Tag: <none>
500    Description:
501    
502    Enabled dlopen and friends for FreeBSD (as recommended by Johannes 5
503    Joemann).
504    
505    ----------------------------------------------------------------------
506    Name: Allen Leung (leunga (at) reservoir (dot) com)
507    Date: 2004/11/17 16:05:21 EST 2004
508    Tag: leunga-20041117-mlrisc-live-kill
509    Description:
510    
511       Added support for MLTree constructs LIVE and KILL
512    to all the architectures.
513    
514    ----------------------------------------------------------------------
515    Name: Matthias Blume (blume (at) tti - c (dot) org)
516    Date: 2004/11/13 00:20:00 CST
517    Tag: blume-20041113-versiontool
518    Description:
519    
520    - Stripped down the versiontool: It now only handles the version number.
521      The date string is generated at bootstrap time (during makeml).
522    
523    - In a previous commit, fixed a minor issue with how polyequal is being
524      translated.  In particular, the code now "looks through" abstractions.
525      This results in slightly fewer polyEqual warnings and hopefully slightly
526      more efficient code.  Important examples for where this matters are
527      the new int64 and word64 types.
528    
529    ----------------------------------------------------------------------
530    Name: Matthias Blume (blume (at) tti - c (dot) org)
531    Date: 2004/11/12 00:30:00 CST
532    Tag: blume-20041112-int64
533    Description:
534    
535    Structure Int64 fully hooked in.  (The implementation is not very
536    efficient, though.)
537    
538    ----------------------------------------------------------------------
539    Name: Matthias Blume (blume (at) tti - c (dot) org)
540    Date: 2004/11/11 17:30:00 CST
541    Tag: blume-20041111-more64
542    Description:
543    
544    All the pieces of Word64 are now there, with the exception of the
545    conversions from and to LargeWord. (Eventually these need to be identities,
546    but for the time being they don't even make sense because LargeWord is
547    32-bit wide.)
548    
549    Also started to add similar support for Int64, but major pieces of that
550    are still missing.
551    
552    ----------------------------------------------------------------------
553    Name: Matthias Blume (blume (at) tti - c (dot) org)
554    Date: 2004/11/11 00:15:00 CST
555    Tag: blume-20041111-word64
556    Description:
557    
558    Structure Word64 is now (almost) complete, word literals and patterns
559    seem to work.  There are a few odd pieces missing.  In particular,
560    I didn't do the {from,to}LargeWord parts because LargeWord is still Word32
561    at the moment.
562    
563    Making Word64 official would mean that LargeWord becomes Word64.  But
564    this requires extreme care because most word-word conversions have to
565    go through LargeWord, so making a mistake means loss of efficiency or
566    worse.  Eventually there will be a solution similar to (but actually
567    simpler than) what I did with IntInf.
568    
569    ----------------------------------------------------------------------
570    Name: Matthias Blume (blume (at) tti - c (dot) org)
571    Date: 2004/11/10 18:12:00 CST
572    Tag: blume-20041110-64bit
573    Description:
574    
575    More 64-bit hacking (but still not even half-way there yet).
576    Also, some assorted improvements to the handling of 8-bit words.
577    
578    ----------------------------------------------------------------------
579    Name: Matthias Blume (blume (at) tti - c (dot) org)
580    Date: 2004/11/09 17:50:00 CST
581    Tag: <none>
582    Description:
583    
584    Started adding some infrastructure for supporting 64-bit int- and
585    word-types.  (Still in its very early stages.)
586    
587    ----------------------------------------------------------------------
588    Name: Matthias Blume (blume (at) tti - c (dot) org)
589    Date: 2004/10/28 10:45:00 CDT
590    Tag: Release_110_50
591    Description:
592    
593    New working version (110.50).  NEW BOOTFILES!
594    
595                 =====================
596    
597    Also:
598    
599      - Changend config/srcarchiveurl from a file just
600        containing the URL string into a file containing
601        shell script code.  The code has access to the $VERSION variable.
602      - Made corresponding changes to config/install.sh and config/unpack.
603      - Default contents of config/srcarchiveurl uses $VERSION and
604        normally does not have to be edited to reflect a version change.
605    
606      (As a result, a version change can be done by just editing
607       config/version, the rest is now automatic.)
608    
609    ----------------------------------------------------------------------
610    Name: Matthias Blume (blume (at) tti - c (dot) org)
611    Date: 2004/10/27 17:50:00 CDT
612    Tag: blume-20041027-btrace-msg
613    Description:
614    
615    BackTrace.monitor now also reports the source of the exception that
616    triggered the trace.
617    
618    ----------------------------------------------------------------------
619    Name: Matthias Blume (blume (at) tti - c (dot) org)
620    Date: 2004/10/27 17:20:00 CDT
621    Tag: blume-20041027-x86-c-calls
622    Description:
623    
624    This is the HISTORY entry for two earlier commits, both concerning
625    the x86 c-calls code in MLRISC:
626    
627      - added a missing LOAD in the code that deals with struct arguments
628      - made sure the caller does not add the wrong number of bytes to the
629        stack pointer after a call of a function returning a struct
630        (the callee already pops the implicit argument which points to
631         the space reserved for the result)
632    
633    ----------------------------------------------------------------------
634    Name: Allen Leung (leunga (at) reservoir (dot) com)
635    Date: 2004/10/24 14:00:00 EST
636    Tag: leunga-20041024-x86-gas-fucomip
637    Description:
638    
639       John discovered a bug in the syntax of fucomip.
640       The opcodes FU?COMIP? have been changed to
641    
642           fu?comip? %st(i), %st
643    
644    ----------------------------------------------------------------------
645    Name: Matthias Blume (blume (at) tti - c (dot) org)
646    Date: 2004/10/20 15:06:00 CDT
647    Tag: blume-20041020-standalone-backtrace
648    Description:
649    
650    Added a mechanism for getting back-trace information from standalone
651    programs.  Here is how it works:
652    
653       1. The part of the program from which you want to get backtrace
654          information (usually the whole program) should be wrapped with
655          BackTrace.monitor.  This is a (unit->'a)->'a function, and your
656          main program could be modified from something like
657    
658             fun main (pgm, args) = ...
659    
660         to
661    
662            fun main (pgm, args) = BackTrace.monitor (fn () => ...)
663    
664       2. To be able to access BackTrace.monitor, you have to add
665          library $smlnj-tdp/plugins.cm to the .cm file that contains your
666          main function.
667    
668       3. Remove all compiled code (i.e., all the .cm/ subdirectories that
669          CM might have created in the past for your project).
670    
671       4. Build the system using this command line:
672    
673             ml-build -Ctdp.instrument=true \$smlnj-tdp/back-trace.cm \
674                      myprog.cm MyProg.main myprog
675    
676          instead of the usual
677    
678             ml-build myprog.cm MyProg.main myprog
679    
680    I changed a library name:
681    
682       $/trace-debug-profile.cm --> $smlnj-tdp/plugins.cm
683    
684    New libraries:
685    
686       $smlnj-tdp/back-trace.cm
687          -- when loaded causes the back-trace plugin to be installed
688       $smlnj-tdp/coverage.cm
689          -- when loaded causes the coverage plugin to be installed
690    
691    ----------------------------------------------------------------------
692    Name: Matthias Blume (blume (at) tti - c (dot) org)
693    Date: 2004/10/18 16:45:00 CDT
694    Tag: blume-20041018-groupowner
695    Description:
696    
697    Added an "obsolete" warning for the "group owner" syntax to CM's parser.
698    
699    Eliminated group owner specs from .cm files throughout the source tree.
700    
701    ----------------------------------------------------------------------
702    Name: Matthias Blume (blume (at) tti - c (dot) org)
703    Date: 2004/10/15 15:45:00 CDT
704    Tag: blume-20041015-coverage
705    Description:
706    
707    * Test coverage tool added!
708    * Further reorganization of tracing-, debugging-, and profiling support:
709    
710        - moved original BTImp -- now called BackTrace -- into a separate
711          library called $/trace-debug-profile.cm
712        - eliminated all mentions of BTrace from SMLofNJ.Internals
713        - only the instrumentation mechanism is now left in the compiler proper
714        - BackTrace module is a plugin which is NOT plugged in by default
715        - Coverage module is another such plugin
716    
717    To get the benefits of any of these plugin modules, the code in
718    question must be compiled with tdp instrumentation turned on.  This
719    can be done by setting SMLofNJ.Internals.TDP.mode to true.
720    (The ref cell is also controlled via the -Ctdp.instrument=... switch.)
721    
722    Plugins are selected at link time.  (Pre-compiled instrumented code
723    can be re-loaded with different plugins in effect.)  When an
724    instrumented module is linked, whatever plugins are at that time
725    enabled will come into effect for that module.
726    
727    To enable the back-trace plugin, load library $/trace-debug-profile.cm
728    and invoke BackTrace.install() (e.g., from the interactive prompt).
729    To enable the coverage plugin, load the same library and invoke
730    Coverage.install().
731    
732    Back-traces are generated automatically on uncaught exceptions and
733    when the code in question explicitly invokes BackTrace.trigger().
734    
735    Coverage (and execution frequency-) information must be queried
736    explicitly by calling Coverage.not_covered and Coverage.hot_spots.
737    
738    ----------------------------------------------------------------------
739    Name: Matthias Blume (blume (at) tti - c (dot) org)
740    Date: 2004/10/14 17:40:00 CDT
741    Tag: blume-20041014-tdp-core
742    Description:
743    
744    Snapshot of a significant overhaul of how the trace/debug/profile support
745    is hooked into the system (specifically: Core and SMLofNJ.Internals).
746    
747    ----------------------------------------------------------------------
748    Name: Matthias Blume (blume (at) tti - c (dot) org)
749    Date: 2004/10/13 16:34:00 CDT
750    Tag: blume-20041013-tdp
751    Description:
752    
753    Some rationalization of names:
754    
755       structure BTrace -> structure TDPInstrument
756       etc.
757    
758    This is is preparation of using the original back-trace
759    instrumentation for other purposes.  "TDP" stands for
760    Trace/Debug/Profile.
761    
762    The control flag controlling whether instrumentation is on or off is now
763    registered under a different name, so instead of running sml as
764    
765     sml -Cinstrument.btrace-mode=true
766    
767    one has to say
768    
769     sml -Ctdp.instrument=true
770    
771    ----------------------------------------------------------------------
772    Name: Matthias Blume (blume (at) tti - c (dot) org)
773    Date: 2004/10/11 16:37:00 CDT
774    Tag: blume-20041011-regions
775    Description:
776    
777    Made some minor modifications to elabcore.sml to have source regions
778    be propagated more tightly -- resulting in better (i.e., smaller)
779    regions being reported in error- and debug messages.
780    
781    ----------------------------------------------------------------------
782    Name: Matthias Blume (blume (at) tti - c (dot) org)
783    Date: 2004/10/08 22:50:00 CDT
784    Tag: blume-20041008-cmkw
785    Description:
786    
787    Fixed handling of keywords in .cm files: After seeing "is" the lexer
788    treats subsequent occurrences of "group", "library", "source", "is",
789    "*", and "-" as ordinary identifiers rather than keywords.
790    
791    Most seriously, this fixes a problem with CM's "shell" tool.  The tool
792    is supposed to accept a tool argument called "source", but this did
793    not work because of the clash with the keyword.
794    
795    ----------------------------------------------------------------------
796    Name: Matthias Blume (blume (at) tti - c (dot) org)
797    Date: 2004/10/07 16:00:00 CDT
798    Tag: blume-20041007-cleanup
799    Description:
800    
801    Assorted cleanup work:
802    
803      - got rid of intstrmap in favor of using the library's
804        hash table implementation
805      - threw out most of the pathnames stuff, as it was not used anyway
806      - simplified tokentable implementation
807      - fixed some minor spelling errors
808    
809    ----------------------------------------------------------------------
810    Name: Matthias Blume (blume (at) tti - c (dot) org)
811    Date: 2004/10/06 15:15:15 CDT
812    Tag: blume-20041006-handler
813    Description:
814    
815    Cleaned up the absyn to reflect the invariant that HANDLE always
816    carries a FNexp as part of the type definition.  This eliminates some
817    superfluous sanity checks at runtime down the road.
818    
819    Some minor cleanup of the btrace code.
820    
821    ----------------------------------------------------------------------
822    Name: Matthias Blume (blume (at) tti - c (dot) org)
823    Date: 2004/10/01 10:20:30 CDT
824    Tag: blume-20041001-slave
825    Description:
826    
827    Added hack to make slave mode work in the presence of the version
828    tool.  (Still, since the master does two passes over the code for
829    CMB.make, the release number gets bumped twice when slaves are
830    attached.  I don't know if this is worth fixing...)
831    
832    ----------------------------------------------------------------------
833    Name: Matthias Blume (blume (at) tti - c (dot) org)
834    Date: 2004/09/30 10:55:00 CDT
835    Tag: blume-20040930-version
836    Description:
837    
838    * Moved the "version" magic into its own little library under
839      src/system/smlnj/internal.  This avoids expensive reconstruction of
840      a stable src/compiler/core.cm.
841    
842    * At the same time, structure CompilerVersion is now known as
843      structure SMLNJVersion.
844    
845    * Arranged for the version tool to NOT kick in when rebuilding the system
846      (makeml -rebuild, fixpt).  Otherwise one would never reach a fixpoint.
847      Also, loading the versiontool does not work when rebuilding the system
848      because CM is not properly initialized at that time.
849    
850    ----------------------------------------------------------------------
851    Name: Matthias Blume (blume (at) tti - c (dot) org)
852    Date: 2004/09/29 14:00:00 CDT
853    Tag: blume-20040929-autoversion
854    Description:
855    
856    Implemented some CM magic to have
857        file src/compiler/TopLevel/main/version.sml
858    generated automagically.
859    The version is taken from two files: config/version and config/release.
860    The first is expected to contain a two-part version number such as 110.49.
861    The second should contain a single number, but it may be missing.
862    
863    If the environment variable VERSIONTOOL_BUMP_RELEASE is defined at the
864    time the version tool is loaded (which is the first time you say CMB.make),
865    then the tool will increment the value stored in config/release every
866    time CMB.make is invoked.
867    
868    The binfile format is now insensitive to anything beyond the first
869    two components of a version number, so bumping the release does not render
870    binfiles incompatible.  Auto-bumping can be used to keep track of versions
871    during development without invalidating existing binfiles.
872    
873    In any case, every CMB.make updates the date information in version.sml.
874    (This is the date that is printed in the banner.)
875    
876    ----------------------------------------------------------------------
877    Name: Matthias Blume (blume (at) tti - c (dot) org)
878    Date: 2004/09/28 10:53:00 CDT
879    Tag: blume-20040928-controls
880    Description:
881    
882    Some cleanup of the controls code.
883    
884    ----------------------------------------------------------------------
885    Name: Matthias Blume (blume (at) tti - c (dot) org)
886    Date: 2004/09/27 22:08:00 CDT
887    Tag: blume-20040927-controls
888    Description:
889    
890    Added two pieces of functionality to the Controls interface:
891    
892    1.   val save'restore: 'a control -> unit -> unit
893    
894       grabs the current value of the control in stage 1 and restores it
895       in stage 2.
896    
897    2.   val set' : 'a control * 'a -> unit -> unit
898    
899       stores the given value into the control in stage 2 (i.e., delayed)
900       but does all error checking in stage 1.
901       (This is for string controls that need to do parse their argument --
902        something that might fail.  In some cases, notably in CM, one
903        already knows the intended argument but wants to delay the actual
904        assignment until a time when error recovery would be more difficult.)
905    
906    Changed the handling of controls in tool arguments to classes "sml" and
907    "lazysml":
908       - use Controls.save'restore as a more robust way of restoring the
909         old value (in particular: without having to re-parse the string)
910       - use controls to handle the "overload" keyword in the init group
911         (I believe this change actually fixes a long-standing obscure bug.)
912    
913    ----------------------------------------------------------------------
914    Name: Matthias Blume (blume (at) tti - c (dot) org)
915    Date: 2004/09/27 17:00:00 CDT
916    Tag: blume-20040927-lazysml
917    Description:
918    
919    Added a new tool class called "lazysml" to CM's tool chest.  The only
920    difference to "sml" is that compilation is done with Control.lazysml
921    set to true.  A source of class "lazysml" is automatically recognized
922    by a file name suffix of ".lml".
923    
924    In addition to the above feature, the original class "sml" now also
925    supports a tool argument "lazy" which has the same effect.  As a
926    result, the following three lines are equivalent:
927    
928        foo.sml : lazysml
929        foo.sml : sml (lazy)
930        foo.sml (lazy)
931    
932    The setting goes into effect both during parsing and during
933    compilation.  The original setting is restored right after parsing and
934    after compilation, respectively.
935    
936    In addition to all the above, there is also a general mechanism to set
937    ANY of the "controls" that are available at the command line via
938    "-C..." on a per-sml-file basis.  The same rules that apply for "lazy"
939    apply as well. (In fact, "lazy" is implemented as a special case of
940    the general mechanism.)
941    
942    The .cm file syntax uses a new keyword tool argument called "with".
943    There are several ways of indicating the desired settings:
944    
945        foo.sml (with:parser.quotations=true)
946        foo.sml (with:(name:parser.quotations value:true))
947        foo.sml (with:(name:name1 value:value1 name:name2 value:value2 ...))
948        foo.sml (with:(name1=value1 name2=value2 ...))
949        foo.sml (with:(name1=value1 name:name2 value:value2 name3=value3 ...))
950    
951        etc.
952    
953    Another possible abbreviation is to leave out the =v or value:v part
954    if the name refers to a boolean control (in which case the value is
955    taken to be true).  Thus, one could get lazy sml also by saying:
956    
957        foo.sml (with:parser.lazy-keyword=true)
958        foo.sml (with:parser.lazy-keyword)
959        foo.sml (with:(name:parser.lazy-keyword value:true))
960        foo.sml (with:(name:parser.lazy-keyword))
961    
962    ----------------------------------------------------------------------
963    Name: Matthias Blume (blume (at) tti - c (dot) org)
964    Date: 2004/09/24 16:22:00 CDT
965    Tag: blume-20040924-ppc-long-branch
966    Description:
967    
968    Turned message about "emiting long form of branch" off by default.
969    Added a control flag to turn it back on when desired.
970    
971    ----------------------------------------------------------------------
972    Name: Matthias Blume (blume (at) tti - c (dot) org)
973    Date: 2004/09/24 16:05:00 CDT
974    Tag: blume-20040924-rounding
975    Description:
976    
977    Applied patch for setting rounding modes under Mac OS X.  Thanks to
978    Melissa O'Neill for providing the code!
979    
980    ----------------------------------------------------------------------
981    Name: Matthias Blume (blume (at) tti - c (dot) org)
982    Date: 2004/09/23 17:30:00 CDT
983    Tag: blume-20040923-envvars
984    Description:
985    
986    1. Changed definition of type ControlRegistry.registry_tree to
987       include control_info (i.e., the name of the controlling environment
988       variable).
989    
990    2. Added command-line flags -e and -E to print the names of environment
991       variables that can be used to control internal settings.
992       (This uses the new API mentioned in 1.)
993    
994    ----------------------------------------------------------------------
995    Name: Matthias Blume (blume (at) tti - c (dot) org)
996    Date: 2004/09/13 16:50:00 CDT
997    Tag: Release_110_49
998    Description:
999    
1000    New working version (110.49).  NEW BOOTFILES!
1001    
1002    ----------------------------------------------------------------------
1003    Name: Matthias Blume
1004    Date: 2004/09/13 16:20:00 CDT
1005    Tag: blume-20040913-config-mlrisc
1006    Description:
1007    
1008    Put target "mlrisc" back into the default list.
1009    (There is no harm in having it, and some users have expressed their
1010    wish to have "mlrisc" included by default.)
1011    
1012    ----------------------------------------------------------------------
1013    Name: John Reppy
1014    Date: 2004/09/13
1015    Tag: jhr-20040913-signals
1016    Description:
1017    
1018    Fixed the signal masking code to properly nest mask/unmask operations
1019    on a per-signal basis.
1020    
1021    ----------------------------------------------------------------------
1022    Name: Matthias Blume (blume (at) tti - c (dot) org)
1023    Date: 2004/09/08 13:20:00 CDT
1024    Tag: blume-20040908-heap-magic
1025    Description:
1026    
1027    Bumped the heap macig number to 0x09082004 to account for the changed
1028    layout of the ML frame under MacOS X.
1029    
1030    ----------------------------------------------------------------------
1031    Name: Allen Leung (leunga (at) reservoir (dot) com)
1032    Date: 2004/09/03 11:26:00 EST
1033    Tag: leunga-20040903-cygwin-install
1034    Description:
1035    
1036       Added a patch to _arch-n-opsys to enable the Cygwin runtime.
1037    The Cygwin runtime is turned on by setting the environment
1038    variable SMLNJ_CYGWIN_RUNTIME to 1.
1039    
1040    ----------------------------------------------------------------------
1041    Name: Matthias Blume (blume (at) tti - c (dot) org)
1042    Date: 2004/08/31 17:15:00 CDT
1043    Tag: blume-20040831-core
1044    Description:
1045    
1046    Added some exports to src/compiler/core.cm upon request by J. Joemann.
1047    
1048    ----------------------------------------------------------------------
1049    Name: Matthias Blume (blume (at) tti - c (dot) org)
1050    Date: 2004/08/30 17:55:00 CDT
1051    Tag: blume-20040830-installer
1052    Description:
1053    
1054    Upon request by Johannes Joemann:
1055    
1056    - improved ML code of installer to fall back to coping when renaming
1057      fails (i.e., when source and target are on different file systems);
1058      the code compiles but has yet to be tested in anger
1059    - removed mlrisc from list of default targets (config/targets)
1060    
1061    ----------------------------------------------------------------------
1062    Name: Matthias Blume (blume (at) tti - c (dot) org)
1063    Date: 2004/08/27 17:20:00 CDT
1064    Tag: blume-20040827-ptreql
1065    Description:
1066    
1067    Added ptreql primop to structure InlineT (upon request from Larry
1068    Paulson).
1069    
1070    ----------------------------------------------------------------------
1071    Name: Allen Leung (leunga (at) reservoir (dot) com)
1072    Date: 2004/08/15 21:21:00 EST
1073    Tag: leunga-110_48-udgraph
1074    Description:
1075    
1076    Another bug fix from Carl Hauser:
1077    
1078    diff /net/niflab/smlnj48/src/MLRISC/graphs/udgraph.sml udgraph.sml
1079    > 48c48
1080    > <              | rmv((e as (k,_))::es,L) = rmv(es,if k = i then es else
1081    > e::L)
1082    > ---
1083    > >              | rmv((e as (k,_))::es,L) = rmv(es,if k = i then L else e::L)
1084    > Without this, any deletion of an edge in an undirected graph does severe
1085    > violence to the graph.
1086    
1087    ----------------------------------------------------------------------
1088    Name: Allen Leung (leunga (at) reservoir (dot) com)
1089    Date: 2004/08/10 23:35:00 EST
1090    Tag: leunga-110_48-ppc
1091    Description:
1092    
1093       The IBM/MacOS syntax switch on PPC was incorrectly swapped.
1094    
1095    ----------------------------------------------------------------------
1096    Name: Matthias Blume (blume (at) tti - c (dot) org)
1097    Date: 2004/08/10 12:00:00 CDT
1098    Tag: Release_110_48
1099    Description:
1100    
1101    New working version (110.48).  NEW BOOTFILES!
1102    
1103    ----------------------------------------------------------------------
1104    Name: Allen Leung (leunga (at) reservoir (dot) com)
1105    Date: 2004/08/09 12:21:00 EST
1106    Tag: leunga-110_47-dijsktra
1107    Description:
1108    
1109    Bug fix from Carl Hauser:
1110    
1111    single_source_shortest_paths in dijkstra.sml was observed to get wrong
1112    answers (by comparing to single_source_shortest_paths in bellman-ford.sml).
1113    
1114    The problem is that following the expression A.update(dist,s,Num.zero)
1115    it is necessary to update the priority queue using Q.decreaseWeight(Q,s).
1116    
1117    ----------------------------------------------------------------------
1118    Name: Matthias Blume (blume (at) tti - c (dot) org)
1119    Date: 2004/08/06 18:10:00 CDT
1120    Tag: blume-20040806-cmdline
1121    Description:
1122    
1123    Fiddled with handling of command-line options:
1124    
1125      * sml now quits after processing the command line
1126        if -H, -S, -h<n>, or -s<n> appears as the last
1127        command-line argument
1128      * a new option -q terminates the session when encountered on
1129        the command line; subsequent arguments will be ignored
1130      * bug fixes: short (erroneous) arguments are no longer ignored
1131        completely
1132    
1133    ----------------------------------------------------------------------
1134    Name: Allen Leung (leunga (at) reservoir (dot) com)
1135    Date: 2004/08/04 18:17:00 EST
1136    Tag: leunga-110_47-ppc-ibm-asm
1137    Description:
1138    
1139      - Added minimal IBM assembly syntax support for PowerPC.
1140    
1141      - Cygwin: manually changed the file cygwin.def.  Some exported symbols have
1142        been altered in the runtime.  We need an automatic way to keep the file
1143        in sync.
1144    
1145    ----------------------------------------------------------------------
1146    Name: Matthias Blume (blume (at) tti - c (dot) org)
1147    Date: 2004/08/04 14:00:00 CDT
1148    Tag: Release_110_47
1149    Description:
1150    
1151    New working version (110.47).  NEW BOOTFILES!
1152    
1153    ----------------------------------------------------------------------
1154    Name: Matthias Blume (blume (at) tti - c (dot) org)
1155    Date: 2004/08/03 14:25:00 CDT
1156    Tag: blume-20040803-callingconv
1157    Description:
1158    
1159    Added low-level support for choosing C calling conventions by
1160    twiddling the type of rawccall.  (See
1161    src/compiler/Semant/types/cproto.sml for details.)
1162    
1163    ----------------------------------------------------------------------
1164    Name: Matthias Blume (blume (at) tti - c (dot) org)
1165    Date: 2004/08/02 15:55:00 CDT
1166    Tag: blume-20040802-backout
1167    Description:
1168    
1169    Backed out of change to win32-filesys.c.  The earlier patch to
1170    get_file_time caused CM to produce files with the wrong time stamp.
1171    
1172    ----------------------------------------------------------------------
1173    Name: Matthias Blume (blume (at) tti - c (dot) org)
1174    Date: 2004/08/02 14:45:00 CDT
1175    Tag: blume-20040802-nlffi-win32
1176    Description:
1177    
1178    Added NLFFI support for Win32, adapted from a patch provided by David
1179    Hansel.  This is currently completely untested.  Also, the issue
1180    concerning stdcall vs. ccall is still unresolved.
1181    
1182    ----------------------------------------------------------------------
1183    Name: Matthias Blume (blume (at) tti - c (dot) org)
1184    Date: 2004/07/30 17:55:00 CDT
1185    Tag: blume-20040730-various
1186    Description:
1187    
1188    Gearing up towards 110.47...
1189    
1190    - various minor bugfixes to ml-nlffigen
1191    - a beginning of a manual for nlffi
1192    
1193    - eliminated 'export name=value' in config/install.sh as this does
1194      not work with certain versions of /bin/sh
1195      (Thanks to David King at Motorola for catching this.)
1196    
1197    - several bugfixes provided or suggested by David Hansel at Reactive Systems:
1198      - added a test for tm==NULL to gmtime.c and localtime.c
1199      - applied patch for incorrect GetFileTime under win32
1200      - toSeconds -> toMilliseconds in Win32/win32-process.sml
1201    
1202    ----------------------------------------------------------------------
1203    Name: Matthias Blume (blume (at) tti - c (dot) org)
1204    Date: 2004/07/21 18:20:00 CDT
1205    Tag: blume-20040721-nlffigen
1206    Description:
1207    
1208    - Fixed minor issue in ml-nlffigen: Now generate structure T_foo
1209      for a typedef to an incomplete type, but leave out the "typ" member.
1210      (This is just for consistency.)
1211    - Started to produce what is supposed to become better (i.e., comprehensive)
1212      documentation of what ml-nlffigen does and produces.
1213    
1214    ----------------------------------------------------------------------
1215    Name: Matthias Blume (blume (at) tti - c (dot) org)
1216    Date: 2004/07/14 16:25:00 CDT
1217    Tag: blume-20040714-union
1218    Description:
1219    
1220    Added C_UNION to c-calls/c-types.sml and updated the machinery
1221    (ml-nlffigen, cproto.sml) that conveys C function interface
1222    information to the code generator.
1223    
1224    However, the actual architecture-specific implementation of function
1225    arguments and results that are C unions is still not implemented.
1226    
1227    ----------------------------------------------------------------------
1228    Name: Allen Leung (leunga (at) reservoir (dot) com)
1229    Date: 2004/07/14 14:38:00 EST
1230    Tag: leunga-110_46_1-ppc-lwzu
1231    Description:
1232    
1233        Added these instructions to the PowerPC architecture:
1234        LBZU(X), LHZU(X), LWZU(X),
1235        STWU(X), STFDU, STFSU
1236    
1237        etc...
1238    
1239        Note: I haven't added their instruction encoding into the description.
1240    
1241    ----------------------------------------------------------------------
1242    Name: Allen Leung (leunga (at) reservoir (dot) com)
1243    Date: 2004/07/13 15:04:00 EST
1244    Tag: leunga-110_46_1-ppc-lwarx
1245    Description:
1246    
1247       Added the two instructions LWARX and STWCX to the PowerPC
1248    instruction set.
1249    
1250       A (untested) rewrite of loop-structure.sml.  The old version
1251    is completely broken.
1252    
1253    ----------------------------------------------------------------------
1254    Name: Matthias Blume (blume (at) tti - c (dot) org)
1255    Date: 2004/07/13 13:50:00 CDT
1256    Tag: blume-20040713-nlffi
1257    Description:
1258    
1259    - use paramAlloc to report c-calls with too many arguments
1260      (for PPC version where parameter area is pre-allocated)
1261    - added ccall_maxargspace to machspec (to implement the above)
1262    - made "make" commend in CM's "make" tool configurable
1263    - added option (default: on) for passing the name of the SML/NJ's "bin"
1264      directory to "make"; the call looks like this:
1265    
1266         make <options> SMLNJ_BINDIR=<dir> <target>
1267    
1268      This can be used by the Makefile to, e.g., pick the "right" version
1269      of ml-nlffigen.
1270    - minor code tweaks
1271    
1272    ----------------------------------------------------------------------
1273    Name: Matthias Blume (blume (at) tti - c (dot) org)
1274    Date: 2004/07/12 22:50:00 CDT
1275    Tag: blume-110_46_1-macosx-nlffi
1276    Description:
1277    
1278    NLFFI under Mac OS X now working (sort of).  This is largely untested,
1279    though.
1280    
1281    Note:  1. You have to make a new, clean build of the runtime system.
1282           2. There are new BOOTFILES, you have to use them!
1283              (Doing the bootstrap process yourself would be *very* painful!
1284               If you absolutely have to do it, build the system under
1285               a different architecture and then cross-compile.)
1286    
1287    Version bumped to 110.46.1 to account for runtime data format changes.
1288    
1289    ----------------------------------------------------------------------
1290    Name: Matthias Blume (blume (at) tti - c (dot) org)
1291    Date: 2004/06/18 14:30:00 CDT
1292    Tag: blume-20040618-unix
1293    Description:
1294    
1295    Changed the implementation of structure Unix so that the same stream
1296    is returned every time one of the {text,bin}{In,Out}streamOf functions
1297    is invoked on the same proc.  This is not what the spec currently
1298    says -- although IMO it arguably should.  (See discussion below.)
1299    
1300    ----------------------------------------------------------------------
1301    Name: Matthias Blume (blume (at) tti - c (dot) org)
1302    Date: 2004/06/17 18:15:00 CDT
1303    Tag: Release_110_46
1304    Description:
1305    
1306    New working version (110.46).  NEW BOOTFILES!
1307    
1308    ----------------------------------------------------------------------
1309    Name: Matthias Blume (blume (at) tti - c (dot) org)
1310    Date: 2004/06/17 17:20:00 CDT
1311    Tag: blume-20040617-timer-unix
1312    Description:
1313    
1314    Changed the interface of structures Timer and Unix to match the most
1315    recent Basis spec.
1316    
1317    In the case of Unix there still seems to be an open/weird issue:
1318    
1319       The {text,bin}{In,Out}streamOf functions are supposed to create
1320       fresh streams whenever they are called -- as opposed to have them
1321       return the same stream every time.  This design is supposed to
1322       prevent space leaks caused by proc values hanging on to streams.
1323    
1324       The reap function, on the other hand, is supposed to close the
1325       streams.  This cannot be done without having a handle on the
1326       stream in proc after all...
1327    
1328    I took the liberty to implement the following stopgap solution:
1329    
1330      The proc value hangs on to the most recently created stream(s).
1331      Reap closes those.  If either or both of the two streams hadn't
1332      been created at all yet, then reap will close the corresponding
1333      file descriptors directly.
1334    
1335    PS: I don't understand the original space leak argument anymore.  If
1336    a proc hangs on to the imperative stream, then I/O operations on those
1337    will advance the state of the cached stream and avoid the space leak.
1338    
1339    ----------------------------------------------------------------------
1340    Name: Matthias Blume (blume (at) tti - c (dot) org)
1341    Date: 2004/05/28 16:45:00 CDT
1342    Tag: blume-20040528-basis
1343    Description:
1344    
1345    Added signature PACK_REAL and exported functor PrimIO.
1346    
1347    ----------------------------------------------------------------------
1348    Name: Matthias Blume (blume (at) tti - c (dot) org)
1349    Date: 2004/05/25 16:00:00 CDT
1350    Tag: blume-20040525-group-owner
1351    Description:
1352    
1353    CM now ignores (but still accepts) the "owner" information in group
1354    descriptions.  The owner of a group is its next enclosing
1355    library. Each group must have a unique owner.  (There is a virtual
1356    "toplevel" library that owns groups which are not nested within a real
1357    library.)  Previously, each group had to explicitly declare its owner,
1358    and CM would check that such a declaration is correct.  The new scheme
1359    is to have CM check that for each group there is precisely one owning
1360    library.
1361    
1362    The advantage of the new scheme is that the programmer no longer needs
1363    to maintain the somewhat annoying owner information.  The downside is
1364    that CM cannot enforce the ownership rule across multiple runs of
1365    CM.make.  Fortunately, enclosing the same group in two different
1366    libraries A and B which are not part of the same program does not
1367    cause real problems.
1368    
1369    ----------------------------------------------------------------------
1370    Name: Matthias Blume (blume (at) tti - c (dot) org)
1371    Date: 2004/05/20 16:00:00 CDT
1372    Tag: blume-20040520-win32
1373    Description:
1374    
1375    Made the win32 version work again.  (Strangely, a misplaced comma had
1376    slipped into win32-process.c which prevented the runtime from being
1377    compiled correctly.)
1378    
1379    Also, included a minor addition to ml-build.bat analogous to what was
1380    done in blume-20040519-ml-build.
1381    
1382    ----------------------------------------------------------------------
1383    Name: Matthias Blume (blume (at) tti - c (dot) org)
1384    Date: 2004/05/19 22:10:00 CDT
1385    Tag: blume-20040519-ml-build
1386    Description:
1387    
1388    Arranged for ml-build to clean up after itself a little bit better.
1389    The script generates a temporary SML source file and compiles it using
1390    CM, so CM generates metadata (GUID, SKEL, objectfile) for it.  It now
1391    gets rid of those at the end, so they don't accumulate under .cm.
1392    
1393    This required a minor change to install.sh because the name of the
1394    metadata directory (default: .cm) is actually configurable at
1395    installation time.
1396    
1397    ----------------------------------------------------------------------
1398    Name: Matthias Blume (blume (at) tti - c (dot) org)
1399    Date: 2004/05/18 15:50:00 CDT
1400    Tag: blume-20040518-mkreader
1401    Description:
1402    
1403    Added Posix.IO.mk{Bin,Text}{Reader,Writer} by lifting their respective
1404    implementations from internal modules PosixBinPrimIO and PosixTextPrimIO.
1405    
1406    ----------------------------------------------------------------------
1407    Name: Matthias Blume (blume (at) tti - c (dot) org)
1408    Date: 2004/05/11 14:35:00 CDT
1409    Tag: blume-20040511-win32sock
1410    Description:
1411    
1412    Added previously missing support for many socket-related functions
1413    under win32.  Thanks to David Hansel <hansel@reactive-systems.com>
1414    for the voluminous patch!
1415    
1416    (I have not tested this patch under win32 yet.)
1417    
1418    Here is David's e-mail:
1419    
1420    Hi,
1421    
1422    Attached to this email you find a diff against sml/nj 110.45
1423    that will enable socket support under Windows.
1424    
1425    To apply the patch (using unix or cygwin)
1426    1) gunzip runtime.diff.gz
1427    2) "cd" into "src/runtime" in the source tree of a fresh
1428       110.45 installation.
1429    3) patch -p 1 < [your/path/to]runtime.diff
1430    
1431    The code compiles fine but has NOT yet been extensively tested.
1432    I only ran a few tests for basic socket client functionality
1433    (which worked fine).  Especially the functions that use ioctl
1434    are not tested at all and might not work (see below).
1435    
1436    I implemented this since we want to move to a newer version of sml/nj
1437    but need socket support in order to use it.  This is the first time I
1438    even had a look at the sml/nj source,  so please review my changes
1439    before making this part of the distribution!  Here are a few issues
1440    that I think might be better for someone to solve who is more
1441    familiar with the sml/nj source (and socket programming):
1442    
1443    - getnetbyaddr.c and getnetbyname.c will raise a "not implemented"
1444      exception since I could not figure out what the windows equivalent
1445      of these functions is
1446    
1447    - In sockets-osdep.h there are a some #include statements that are
1448      only used in a few files that include sockets-osdep.h
1449    
1450    - In smlnj-sock-lib.c, function init_fn() calls WSAStartup() but
1451      does not process its return value since I don't know how to
1452      report an error upwards.
1453    
1454    - It would probably be good to have a call to WSACleanup() when
1455      the library is unloaded (if there is such a possibility).
1456      Otherwise I think Windows will take care of this automatically
1457      when the process finishes.
1458    
1459    - I used ioctlsocket() as a replacement for ioctl() but I have
1460      no idea if that is actually the proper replacement on Windows.
1461    
1462    - All these issues are marked in the code by "FIXME" comments.
1463    
1464    We use sml/nj extensively in our products and are quite happy
1465    with it.  I hope this contribution will help you.
1466    
1467    Keep up the good work!
1468    
1469    David
1470    
1471    ----------------------------------------------------------------------
1472    Name: Matthias Blume (blume (at) tti - c (dot) org)
1473    Date: 2004/05/11 14:20:00 CDT
1474    Tag: blume-20040511-installml
1475    Description:
1476    
1477    Fixed two bugs in installml script.  (Thanks to Vesa A. Norrman for
1478    the patch.)
1479    
1480    ----------------------------------------------------------------------
1481    Name: Matthias Blume (blume (at) tti - c (dot) org)
1482    Date: 2004/05/11 14:05:00 CDT
1483    Tag: blume-20040511-nlffi-netbsd
1484    Description:
1485    
1486    Added support for nlffi under netbsd.  (Thanks to Vesa A. Norrman for
1487    the patch.)
1488    
1489    ----------------------------------------------------------------------
1490    Name: Matthias Blume (blume (at) tti - c (dot) org)
1491    Date: 2004/05/11 12:05:00 CDT
1492    Tag: blume-20040511-exports
1493    Description:
1494    
1495    As per request by Adam Chlipala <adam@hcoop.net>, extended various
1496    export lists in compiler-related .cm-files.
1497    
1498    ----------------------------------------------------------------------
1499    Name: Matthias Blume (blume (at) tti - c (dot) org)
1500    Date: 2004/05/11 11:35:00 CDT
1501    Tag: blume-20040511-allsource
1502    Description:
1503    
1504    The installer now honors the "src-smlnj" target again, although its meaning
1505    has changed from "all sources required for the compiler" to "all sources
1506    the installer knows about".  In other words, if you enable "src-smlnj"
1507    in the "targets" file, then the installer will pull in sources for
1508    everything.  (Notice that this refers to source code only.  Compiled
1509    code is still only installed for modules that were requested explicitly
1510    or which are required for other modules that were requested explicitly.)
1511    
1512    ----------------------------------------------------------------------
1513    Name: Matthias Blume (blume (at) tti - c (dot) org)
1514    Date: 2004/04/23 17:40:00 CDT
1515    Tag: blume-20040423-ieee-scan
1516    Description:
1517    
1518    Fixed IEEEReal.scan (and .fromString) so that if there is an overflow
1519    in the exponent calculation we get INF or ZERO (depending on the mantissa
1520    and the sign of the exponent).
1521    
1522    ----------------------------------------------------------------------
1523    Name: Matthias Blume (blume (at) tti - c (dot) org)
1524    Date: 2004/04/23 10:40:00 CDT
1525    Tag: blume-20040423-ml-build
1526    Description:
1527    
1528    The ml-build script now terminates with a non-0 status when something
1529    goes wrong.
1530    
1531    ----------------------------------------------------------------------
1532    Name: Matthias Blume (blume (at) tti - c (dot) org)
1533    Date: 2004/04/22 16:35:00 CDT
1534    Tag: blume-20040422-Option
1535    Description:
1536    
1537    Made exception Option to be the same as exception Option.Option
1538    (as it should be).
1539    
1540    ----------------------------------------------------------------------
1541    Name: Allen Leung (leunga (at) reservoir (dot) com)
1542    Date: 2004/03/19 14:40:00 EST
1543    Tag: leunga-20040319-cygwin-nlffi
1544    Description:
1545    
1546    Fixed the runtime so that ml-nlffi-lib runs on the cygwin version
1547    of SML/NJ.  The problem is that
1548    
1549       lib = dlopen(NULL, ...)
1550       f   = dlsym(lib, "malloc");
1551    
1552    does not work on Windows unless we explicitly export symbols
1553    such as 'malloc' during linking.  We fixed this by explicitly
1554    exporting the required symbols with the magic gcc incantation:
1555    
1556        -Wl,--export-all cygwin.def
1557    
1558    where cygwin.def is a file containing all the symbols that
1559    we wish to export.
1560    
1561    I suspect this is a Windows problem and we'll have to
1562    do the same (somehow with windows compilers) when
1563    we build the native win32 version with the system
1564    calls LoadLibrary/GetProcAddress.
1565    
1566    ----------------------------------------------------------------------
1567    Name: Matthias Blume (blume (at) tti - c (dot) org)
1568    Date: 2004/03/04 16:35:00 CST
1569    Tag: blume-20040304-intinf-fmt
1570    Description:
1571    
1572    Fixed problem with IntInf.fmt (sign would show up on the right instead
1573    of on the left for BIN, OCT, and HEX).
1574    
1575    ----------------------------------------------------------------------
1576    Name: Matthias Blume (blume (at) tti - c (dot) org)
1577    Date: 2004/03/04 11:25:00 CST
1578    Tag: blume-20040304-symlinks
1579    Description:
1580    
1581    Fixed problem with installer script (unix only) where bin/ml-yacc and
1582    friends pointed (via symlinks) to absolute locations instead of just
1583    .run-sml.  This was reported by Vesa A Norrman.
1584    
1585    ----------------------------------------------------------------------
1586    Name: Matthias Blume (blume (at) tti - c (dot) org)
1587    Date: 2004/02/13 14:50:00 CST
1588    Tag: Release_110_45
1589    Description:
1590    
1591    New working version (110.45). New bootfiles.
1592    
1593    ----------------------------------------------------------------------
1594    Name: Matthias Blume (blume (at) tti - c (dot) org)
1595    Date: 2004/01/26 15:15:15 CST
1596    Tag: blume-20040126-toplevel
1597    Description:
1598    
1599    Improved handling of exceptions at the interactive toplevel.
1600    
1601    ----------------------------------------------------------------------
1602    Name: Matthias Blume (blume (at) tti - c (dot) org)
1603    Date: 2004/01/26 11:25:00
1604    Tag: blume-20040126-app
1605    Description:
1606    
1607    Type of top-level "app" corrected.
1608    Added code for setting vp_limitPtrMask to Win32-specific runtime.
1609    
1610    ----------------------------------------------------------------------
1611    Name: Matthias Blume (blume (at) tti - c (dot) org)
1612    Date: 2003/11/18 17:10 CST
1613    Tag: blume-20031118-basis-fiddle
1614    Description:
1615    
1616    - changed Timer interface to what might become the spec
1617    - POSIX_FLAGS -> BIT_FLAGS according to spec
1618    - some other minor discrepancies wrt. spec eliminated
1619    
1620    ----------------------------------------------------------------------
1621    Name: Matthias Blume (blume (at) tti - c (dot) org)
1622    Date: 2003/11/06 12:00:00 CST
1623    Tag: Release_110_44
1624    Description:
1625    
1626    New working version (110.44). New bootfiles.
1627    
1628    ----------------------------------------------------------------------
1629    Name: Matthias Blume (blume (at) tti - c (dot) org)
1630    Date: 2003/11/04 11:50:00 CST
1631    Tag: blume-20031104-move-libraries
1632    Description:
1633    
1634    Eliminated the "dont_move_libraries" directive in config/targets.
1635    (The mechanism was broken and could not be fixed easily.  Moreover,
1636    there does not seem to be any reason not to move all libraries into
1637    lib during installation.  I originally implemented this directive as a
1638    backward-compatibility feature when I first introduced the new CM.
1639    Now that things have been stable for a long time and going back to the
1640    old CM is not an option, there is no reason to keep it around.)
1641    
1642    ----------------------------------------------------------------------
1643    Name: Matthias Blume (blume (at) tti - c (dot) org)
1644    Date: 2003/11/03 16:00:00 CST
1645    Tag: blume-20031103-installdir
1646    Description:
1647    
1648    Made installer honor INSTALLDIR variable again.  (Thanks to Chris
1649    Richards for pointing out the problem and providing the solution.)
1650    
1651    ----------------------------------------------------------------------
1652    Name: Matthias Blume (blume (at) tti - c (dot) org)
1653    Date: 2003/10/01 17:05:00 CDT
1654    Tag: blume-20031001-lal-mlrisc
1655    Description:
1656    
1657    MLRISC bug fix from Lal.
1658    
1659    ----------------------------------------------------------------------
1660    Name: Matthias Blume (blume (at) tti - c (dot) org)
1661    Date: 2003/09/30 16:10:00 CDT
1662    Tag: blume-20030930-primio-bat
1663    Description:
1664    
1665    1. Added openVector, nullRd, and nullWr to PRIM_IO.
1666    2. Improved .bat files (for Win32 port) to make things work under Win95.
1667       (thanks to Aaron S. Hawley for this one)
1668    
1669    ----------------------------------------------------------------------
1670    Name: Matthias Blume (blume (at) tti - c (dot) org)
1671    Date: 2003/09/26 16:05:00 CDT
1672    Tag: blume-20030926-wrappriv
1673    Description:
1674    
1675    Added missing wrapper for privilege "primitive" in $smlnj/viscomp/core.cm.
1676    
1677    ----------------------------------------------------------------------
1678    Name: Matthias Blume (blume (at) tti - c (dot) org)
1679    Date: 2003/09/26 15:00:00 CDT
1680    Tag: blume-20030926-110_43_3
1681    Description:
1682    
1683    - additional cleanup
1684    - version number bump, NEW BOOTFILES
1685    
1686    ----------------------------------------------------------------------
1687    Name: Matthias Blume (blume (at) tti - c (dot) org)
1688    Date: 2003/09/26 12:00:00 CDT
1689    Tag: blume-20030926-ppautoload
1690    Description:
1691    
1692    I modified the read-eval-print loop so that the autoloader gets
1693    invoked whenever the prettyprinter tries to look up a symbol that
1694    is not currently defined in the toplevel environment but which
1695    appears in CM's autoload registry.  As a result, we see far fewer of
1696    those ?.Foo.Bar.xxx names in the prettyprinter's output.
1697    
1698    In addition to this I tried to clean up some pieces of the Basis
1699    implementation (e.g., Socket, Word8Array) in order to prevent other
1700    instances of these ?.Foo.Bar.xxx names from being printed.
1701    
1702    The mechanism that picks names for types still needs some work, though.
1703    (Right now it seems that if there is a type A.t which is defined to
1704    be B.u, but B is unavailable at toplevel, then A.t gets printed as
1705    "?.B.u" although the perhaps more sensible solution would be to use
1706    "A.t" in this case.  In other words, the prettyprinter should follow
1707    a chain of DEFtycs not farther than there are corresponding toplevel
1708    names in the current environment.)
1709    
1710    ----------------------------------------------------------------------
1711    Name: Matthias Blume (blume (at) tti - c (dot) org)
1712    Date: 2003/09/24 16:31:00 CDT
1713    Tag: blume-20030924-installer
1714    Description:
1715    
1716    Another installer tweak:  All the ML code for the installer is now
1717    compiled during CMB.make and put into a little library called
1718    $smlnj/installer.cm.  The installation then simply invokes
1719    
1720       sml -m $smlnj/installer.cm
1721    
1722    and everything happens automagically.
1723    
1724    Win32:  ML code senses value of environment variable SMLNJ_HOME.
1725    Unix:   ML code senses values of environment variables ROOT, CONFIGDIR,
1726            and BINDIR.
1727    
1728    The new scheme guarantees that the ML code responsible for the installation
1729    is in sync with the APIs of the main system.  Also, the installer is
1730    somewhat faster because the installer script is precompiled.
1731    
1732    ----------------------------------------------------------------------
1733    Name: Matthias Blume (blume (at) tti - c (dot) org)
1734    Date: 2003/09/24 15:35:00 CDT
1735    Tag: blume-20030924-synsock
1736    Description:
1737    
1738    Added a signature SYNCHRONOUS_SOCKET to basis.cm.  This is like SOCKET
1739    but excludes all non-blocking operations.  Defined SOCKET (in Basis)
1740    and CML_SOCKET in terms of SYNCHRONOUS_SOCKET.  Removed superfluous
1741    implementations of non-blocking operations from CML's Socket
1742    structure.
1743    
1744    ----------------------------------------------------------------------
1745    Name: Matthias Blume (blume (at) tti - c (dot) org)
1746    Date: 2003/09/24 15:10:05 CDT
1747    Tag: blume-20030924-sockets
1748    Description:
1749    
1750    1. Fixed SOCKET API and implementation to match Basis spec.
1751       This required changing the internal representation of sockets to one
1752       that remembers (for each socket file descriptor) whether it is currently
1753       blocking or non-blocking.  This state is maintained lazily (i.e., a system
1754       call is made only if the state actually needs to change).
1755    
1756    2. OS-specific details of sockets were moved into separate files, thus
1757       making it possible to unify the bulk of the socket implementations
1758       between Unix and Win32.
1759    
1760    3. CML's socket API changed accordingly.
1761       (Note that we need to remove non-blocking functions from this API
1762       since they are redundant in the case of CML!)
1763    
1764    4. CML's socket implementation now makes use of non-blocking functions
1765       provided by Basis, thus removing all OS-dependent code from this part
1766       of CML.
1767    
1768    5. Changed Real64.precision from 52 to 53.  Minor cleanup in Real64 code.
1769    
1770    ----------------------------------------------------------------------
1771    Name: Matthias Blume (blume (at) tti - c (dot) org)
1772    Date: 2003/09/22 12:10:00 CDT
1773    Tag: blume-20030922-110_43_2
1774    Description:
1775    
1776    Made a new interim version and bootfiles for developer's bootstrapping
1777    convenience.
1778    
1779       110.43.2 -- NEW BOOTFILES
1780    
1781    ----------------------------------------------------------------------
1782    Name: Matthias Blume (blume (at) tti - c (dot) org)
1783    Date: 2003/09/19 15:55:00 CDT
1784    Tag: blume-20030919-cmdir
1785    Description:
1786    
1787    1. new-install.sh -> install.sh
1788    2. changed default CM "metadata" directory name to ".cm" (instead of "CM")
1789    3. tweaked installer so that another name instead of .cm can be chosen
1790       at install time (by setting the CM_DIR_ARC environment variable
1791       during installation); once installation is complete, the name is
1792       fixed
1793    
1794    ----------------------------------------------------------------------
1795    Name: Matthias Blume (blume (at) tti - c (dot) org)
1796    Date: 2003/09/18 16:00:00 CDT
1797    Tag: blume-20030918-110_43_1
1798    Description:
1799    
1800    Made a new interim version and bootfiles for developer's bootstrapping
1801    convenience.
1802    
1803       110.43.1 -- NEW BOOTFILES
1804    
1805    ----------------------------------------------------------------------
1806    Name: Matthias Blume (blume (at) tti - c (dot) org)
1807    Date: 2003/09/18 15:20:00 CDT
1808    Tag: blume-20030918-misc
1809    Description:
1810    
1811    1. Exported fractionsPerSecond etc. from TimeImp (but not from Time as
1812       this seems to be controversial at the moment) and used those in
1813       Posix.ProcEnv.times.
1814    
1815    2. Added Time.{from,to}Nanoseconds to Time.
1816    
1817    3. Improved Real.{from,to}LargeInt by avoiding needless calculations.
1818       For example, fromLargeInt never needs to look at more than 3 "big
1819       digits" to get its 53 bits of precision.
1820    
1821    ----------------------------------------------------------------------
1822    Name: Matthias Blume (blume (at) tti - c (dot) org)
1823    Date: 2003/09/17 16:30:00 CDT
1824    Tag: blume-20030917-real32-slices
1825    Description:
1826    
1827    Added an entry to the primitive environment
1828    (compiler/Semant/statenv/prim.sml) for int32->real64 conversion and
1829    added code to compiler/CodeGen/main/mlriscGen.sml to implement it.
1830    
1831    Removed some of the "magic" constants in real64.sml and replaced them
1832    with code that generates these values from their corresponding
1833    integer counterparts.
1834    
1835    Made all(?) the slice-related changes to the Basis and made everything
1836    compile again...
1837    
1838    ----------------------------------------------------------------------
1839    Name: Matthias Blume (blume (at) tti - c (dot) org)
1840    Date: 2003/09/15 17:45:00 CDT
1841    Tag: blume-20030915-rbase
1842    Description:
1843    
1844    Fixed bug in Real.fromLargeInt.
1845    
1846    ----------------------------------------------------------------------
1847    Name: Matthias Blume (blume (at) tti - c (dot) org)
1848    Date: 2003/09/13 18:11:00 CDT
1849    Tag: blume-20030913-libinstall
1850    Description:
1851    
1852    Minor bugfix in config/libinstall (set anchor with path to
1853    standalone tool after installing it, otherwise libraries that
1854    need ml-lex or ml-yacc won't compile the first time the installer
1855    runs).
1856    
1857    ----------------------------------------------------------------------
1858    Name: Matthias Blume (blume (at) tti - c (dot) org)
1859    Date: 2003/09/12 11:45:00 CDT
1860    Tag: blume-20030912-various
1861    Description:
1862    
1863     - fixed bug in Real.toLargeInt
1864     - fixed bug in Posix.ProcEnv.times
1865     - changed inputLine functions to return an option
1866     - minor installer improvements / bugfixes
1867     - changed default @SMLalloc parameter for x86/celeron to 64k
1868    
1869    ----------------------------------------------------------------------
1870    Name: Matthias Blume (blume (at) tti - c (dot) org)
1871    Date: 2003/09/09 22:00:00 CDT
1872    Tag: Release_110_43
1873    Description:
1874    
1875        New working release 110.43.  New bootfiles.
1876    
1877    ----------------------------------------------------------------------
1878    Name: Matthias Blume (blume (at) tti - c (dot) org)
1879    Date: 2003/09/09 19:20:00 CDT
1880    Tag: blume-20030909-installer
1881    Description:
1882    
1883    Rewrote large parts of config/install.sh in SML (config/libinstall.sml).
1884    Modified config/install.bat to take advantage of it.  Also modified
1885    config/install.sh (and called it config/new-install.sh) to take advantage
1886    of it on Unix systems.  (The SML code is (supposed to be) platform-
1887    independent.)
1888    
1889    The installer can now install everything under Win32
1890    as well as under *nix as long as it compiles.
1891    
1892    Other changes:
1893    
1894      - made CML compile again under Win32
1895      - made eXene compile under Win32 (by providing a fake structure UnixSock
1896        and by using OS.Process.getEnv instead of Posix.ProcEnv.getenv)
1897      - fixed a bug in nowhere: it assumed that type OS.Process.status is the
1898        same as type int;  under Win32 it isn't
1899      - fixed some slice-related problems in the win32-specific parts of CML
1900      - added a functor argument "sameVol" to os-path-fn.sml in the Basis
1901        (under Win32, the volume name is case-insensitive, and the
1902        OS.Path code compares volume names for equality)
1903    
1904    ----------------------------------------------------------------------
1905    Name: Matthias Blume (blume (at) tti - c (dot) org)
1906    Date: 2003/09/08 11:55:00 CDT
1907    Tag: blume-20030908-fullpath
1908    Description:
1909    
1910    Made Win32 version of OS.FileSys.fullPath return current directory
1911    when given an empty string.  This is what the spec says, and incidentally,
1912    CM depends on it.  (CM otherwise goes into an infinite loop in certain
1913    cases when presented with the name of a non-existing .cm file.)
1914    
1915    ----------------------------------------------------------------------
1916    Name: Matthias Blume (blume (at) tti - c (dot) org)
1917    Date: 2003/09/04 16:30:00 CDT
1918    Tag: blume-20030905-slices-etc
1919    Description:
1920    
1921      1. Changed interface to vectors and arrays in Basis to match
1922         (draft) Basis spec.
1923      2. Added signatures and implementations of slices according to
1924         Basis spec.
1925      3. Edited source code throughout the system to make it compile again
1926         under 1. and 2.  (In some cases code had to be added to have it
1927         match the new signatures.)
1928      4. MLRISC should be backward-compatible: the copies of the originals
1929         of files that needed to change under 3. were retained, the .cm files
1930         check the compiler version number and use old versions when
1931         appropriate.
1932      5. Changed type of OS.FileSys.readDir and Posix.FileSys.readdir to
1933         dirstream -> string option (in accordance with Basis spec).
1934      6. When generating code that counts lines, ml-lex used function
1935         CharVector.foldli, taking advantage of its old interface.
1936         This has been replaced with the corresponding code from
1937         CharVectorSlice.  (html-lex must be re-lexed!)
1938      7. BitArray in smlnj-lib/Util has been extended/modified to match the
1939         new MONO_ARRAY signature.  (Do we need BitArraySlice?)
1940      8. Removed temporary additions (fromInternal, toInternal) from the
1941         (now obsolete) IntInf in smlnj-lib/Util.
1942      9. Cleaned up structure Byte.
1943     10. Added localOffset, scan, and fromString to Date (according to spec).
1944         Cleaned/corrected implementation of Date.
1945         (Still need to check for correctness; implement better canonicalizeDate.)
1946     11. Added "scan" to signature IEEE_REAL.
1947     12. Some improvements to IntInf [in particular: efficiency-hack for
1948         mod and rem when second operand is 2 (for parity checks).]
1949     13. Changed representation of type Time.time, using a single IntInf.int
1950         value counting microseconds.  This considerably simplified the
1951         implementation of structure Time.  We now support negative time
1952         values; scan and fromString handle signs.
1953     14. Functor PrimIO now takes two additional arguments (VectorSlice and
1954         ArraySlice).
1955    
1956    ----------------------------------------------------------------------
1957    Name: Matthias Blume (blume (at) tti - c (dot) org)
1958    Date: 2003/08/28 17:00:00 CDT
1959    Tag: blume-20030828-intinf
1960    Description:
1961    
1962    This is a major update which comes with a version number bump
1963    (110.42.99 -- yes, we are really close to 110.43 :-), NEW BOOTFILES,
1964    and an implementation of IntInf in the Basis.
1965    
1966    There are a fairly large number of related changes and updates throughout
1967    the system:
1968    
1969    Basis:
1970            - Implemented IntInf.
1971            - Made LargeInt a projection of IntInf (by filtering through INTEGER).
1972            - Added some missing Real64 operations, most notably Real.toLargeInt.
1973            - Added FixedInt as a synonym for Int32.
1974    
1975    compiler:
1976            * Added support for a built-in intinf type.
1977              - literals
1978              - pattern matching
1979              - conversion shortcuts (Int32.fromLarge o Int.toLarge etc.)
1980              - overloading on literals and operations
1981    
1982            This required adding a primitive type intinf, some additional
1983            primops, and implementations for several non-trivial intinf
1984            operations in Core.  (The intinf type is completely abstract
1985            to the compiler; all operations get delegated back to the Core.)
1986    
1987            * Intinf equality is handled by polyequal.  However, the compiler
1988            does not print its usual warning in this case (since polyequal
1989            is the right thing to do there).
1990    
1991            * Improved the organization of structure InlineT.
1992    
1993            * A word about conversion primops:
1994            If conversions involving intinf do not cancel out during
1995            CPS contract, then the compiler must insert calls to Core functions.
1996            Since all core access must be resolved already during the FLINT
1997            translate phase, it would be too late a the time of CPS contract
1998            to add new Core calls.  For this reason, conversion primops
1999            for intinf carry two arguments:  1. the numeric argument that
2000            they are supposed to convert, and 2. the Core function that
2001            can help with this conversion if necessary.  If CPS contract
2002            eliminates a primop, then the associated Core function becomes
2003            dead and goes away.  Intinf conversion primops that do not get
2004            eliminated by CPS contract get rewritten into calls of their
2005            core functions by a separate, new phase.
2006    
2007    interactive system:
2008            - Control.Print.intinfDepth controls max length of intinf constants
2009            being printed. (Analogous to Control.Print.stringDepth.)
2010            - Cleanup in printutil and pputil: got rid of unused stuff and
2011            duplicates; replaced some of the code with code that makes better
2012            use of library functionality.
2013    
2014    CM:
2015            Bugfix: parse-errors in init group (system/smlnj/init/init.cmi)
2016            are no longer silent.
2017    
2018    CKIT:
2019            Fixed mismatched uses of Int32 and LargeInt.  I always decided
2020            in favor of LargeInt -- which is now the same as IntInf.
2021            CKIT-knowledgable people should check whether this is what's
2022            intended and otherwise change things back to using Int32 or
2023            FixedInt.
2024    
2025    Throughout the code:
2026            Started using IntInf.int literals and built-in operations
2027            (e.g., comparison with 0) where this seems appropriate.
2028    
2029    
2030    ----------------------------------------------------------------------
2031    Name: Dave MacQueen (dbm@cs.uchicago.edu)
2032    Date: 2003/08/13 11:36:00 CDT
2033    Tag: dbm-20030813-mcz-merge1
2034    Description:
2035    
2036    Merging changes from the mcz-branch development branch into trunk.
2037    These changes involve replacement of the emulated old prettyprinter
2038    interface with direct use of the SML/NJ Lib PP library, and fixing
2039    of a couple of bugs (895, 1186) relating to error messages.  A new
2040    prettyprinter for ast datatypes (Elaborator/print/ppast.{sig,sml})
2041    has been added.
2042    
2043    ----------------------------------------------------------------------
2044    Name: Matthias Blume (blume (at) tti - c (dot) org)
2045    Date: 2003/08/11 15:45:00 CDT
2046    Tag: blume-20030811-windows
2047    Description:
2048    
2049    Version number bumped to 110.42.9.  NEW BOOTFILES!!!
2050    
2051        http://smlnj.cs.uchicago.edu/dist/working/110.42.9/
2052    
2053    This patch restores SML/NJ's ability to run under win32.  There are a
2054    number of changes, including fixes for several bugs that had gone
2055    unnoticed until now:
2056    
2057      - uname "CYGWIN_NT*" is recognized as win32 (This is relevant only when
2058        trying to run the win32 version from within cygwin.)
2059    
2060      - There are a number of simple .bat scripts that substitute for their
2061        corresponding Unix shell-scripts.  (See below.)
2062    
2063      - The internals of ml-build have been modified slightly.  The main
2064        difference is that instead of calling ".link-sml" (or link-sml.bat)
2065        using OS.Process.system, the ML process delegates this task back
2066        to the script.  Otherwise problems arise in mixed environments such
2067        as Cygwin where scripts look and work like Unix scripts, but
2068        where OS.Process.system cannot run them.
2069    
2070      - In CM, the srcpath pickler used native pathname syntax -- which
2071        is incorrect in the case of cross-compilation.  The new pickle format
2072        is independent of platform-specific naming conventions.
2073    
2074      - Path configuration files (such as lib/pathconfig) can now choose
2075        between native and standard syntax.  Placing a line of the form
2076    
2077            standard!
2078    
2079        into the file causes all subsequent paths to be interpreted using
2080        CM standard pathname syntax (= Unix conventions); a line
2081    
2082            native!
2083    
2084        switches back to native style.  This was needed so that
2085        path config files can be written portably, see src/system/pathconfig.
2086    
2087      - Runtime system:
2088    
2089          - win32-filesys.c:  get_file_time and set_file_time now
2090            access modification time, not creation time.
2091    
2092          - I/O code made aware of new array representation.
2093    
2094          - Bug fixes in X86.prim.masm.
2095    
2096       - src/system/makeml made aware of win32. (For use under cygwin
2097            and other Unix-environments for windows.)
2098    
2099       - In Basis, fixed off-by-one error in win32-io.sml (function vecF)
2100         which caused BinIO.inputAll to fail consistently.
2101    
2102    .bat scripts:
2103    
2104       Windows .bat scripts assume that SMLNJ_HOME is defined.
2105    
2106       - sml.bat, ml-yacc.bat, ml-lex.bat:  Driver scripts for standalone
2107         applications (sml, ml-yacc, ml-lex).
2108       - ml-build.bat: analogous to ml-build.
2109       - config\install.bat:  Analogous to config/install.sh.  This requires
2110         that SMLNJ_HOME is set and that Microsoft Visual C is ready to use.
2111         (nmake etc. must be on the path, and vcvars32 must have been run.)
2112         Moreover, sources for ml-lex and ml-yacc need to exist under src,
2113         and the bootfile hierarchy must have been unpacked under
2114         sml.boot.x86-win32.
2115         The script is very primitive and does a poor job at error checking.
2116         It only installs the base system, ml-lex, and ml-yacc.  No other
2117         libraries are being installed (i.e., you get only those that
2118         are part of the compiler.)
2119       - link-sml.bat:  analogous to .link-sml, but not currently used
2120    
2121    Unrelated bug fixes:
2122    
2123      - ml-nlffigen now exports structures ST_* corresponding to incomplete
2124        types.
2125      - Added getDevice to PP/src/pp-debug-fn.sml.  (Would not compile
2126        otherwise.)
2127    
2128    ----------------------------------------------------------------------
2129    Name: Dave MacQueen (macqueen@cs.uchicago.edu)
2130    Date: 2003/06/17
2131    Tag: macqueen-20030617-bug895
2132    Description:
2133    
2134    Modified compiler/Elaborator/print/pptype.sml to fix bug 895.
2135    Tag will be used for new development branch (mcz-branch) for
2136    use by MacQueen, (Lucasz) Zairek, and (George) Cao at uchicago.
2137    
2138    ----------------------------------------------------------------------
2139    Name: Matthias Blume (blume (at) tti - c (dot) org)
2140    Date: 2003/05/27 16:55:00 CDT
2141    Tag: blume-20030527-polyeq
2142    Description:
2143    
2144    Tried to eliminated most cases of polymorphic equality.
2145    
2146    ----------------------------------------------------------------------
2147    Name: Matthias Blume (blume (at) tti - c (dot) org)
2148    Date: 2003/05/21 17:45:00 CDT
2149    Tag: blume-20030517-complete
2150    Description:
2151    
2152    Two changes:
2153    
2154      1. Added a flag for controlling whether non-exhaustive bindings will
2155         be treated as errors (default is false).
2156      2. Cleaned up the *entire* source tree so that CMB.make goes through
2157         without a single non-exhaustive match- or bind warning.
2158    
2159    ----------------------------------------------------------------------
2160    Name: Matthias Blume (blume (at) tti - c (dot) org)
2161    Date: 2003/05/17 10:20:00 CDT
2162    Tag: blume-20030517-absyn
2163    Description:
2164    
2165    1. Added cases for IF, WHILE, ANDALSO, and ORELSE to Absyn.
2166    
2167       This mainly affects the quality of error messages.  However, some
2168       of the code is now more straightforward than before. (Treatment of
2169       the above four constructs in translate.sml is much simpler than
2170       the "macro-expansion" that was going on before.  Plus, the mach-
2171       compiler no longer gets invoked just to be able to compile an
2172       if-expression.)
2173    
2174    2. The ErrorMsg.Error exception is now caught and absorbed by the
2175       interactive loop.
2176    
2177    ----------------------------------------------------------------------
2178    Name: Allen Leung
2179    Date: 2003/05/16 13:05:00 CDT
2180    Tag: leunga-20030516-cygwin-runtime
2181    Description:
2182    
2183        Ported the runtime system to cygwin, which uses the unix
2184    x86-unix bin files.  Missing/buggy features:
2185    
2186       o getnetbyname, getnetbyaddr: these functions seem to be missing in
2187         the Cygwin library.
2188       o Ctrl-C handling may be flaky.
2189       o Windows system calls and Windows I/O are not supported.
2190    
2191       A new set of binfiles is located at:
2192    
2193         http://www.dorsai.org/~leunga/boot.x86-unix.tgz
2194    
2195       This is only needed for bootstrapping the cygwin version of smlnj.
2196       Other x86 versions can use the existing binfiles.
2197    
2198    ----------------------------------------------------------------------
2199    Name: Matthias Blume
2200    Date: 2003/04/08 15:42:00 CDT
2201    Tag: blume-20030408-listpair
2202    Description:
2203    
2204    1. Added a target 'mlrisc' to installer.
2205    
2206    2. Added missing elements to structure ListPair.
2207    
2208    ----------------------------------------------------------------------
2209    Name: Allen Leung
2210    Date: 2003/01/07 10:40:00 EST
2211    Tag: leunga-20030107-int-rem
2212    Description:
2213    
2214    Fixed a bug in Int.rem(x,y) where y is a power of 2 on x86.
2215    The arguments to the SUBL instruction were swapped.
2216    
2217    ----------------------------------------------------------------------
2218    Name: Matthias Blume
2219    Date: 2002/12/12 16:25:00 EST
2220    Tag: blume-20021212-risc-ra
2221    Description:
2222    
2223    Fixed a serious bug in the rewrite code for FP spilling/reloading that
2224    sent the RA into an infinite loop when floating point registers get
2225    spilled.  (Because of this bug, e.g., nucleic stopped compiling between
2226    110.37 and 110.38.)
2227    There was another set of potential problems related to the handling of
2228    MLRISC annotations (but those did not yet cause real problems, apparently).
2229    
2230    ----------------------------------------------------------------------
2231    Name: Matthias Blume
2232    Date: 2002/12/06 22:40:00 EST
2233    Tag: blume-20021206-cm-fileid
2234    Description:
2235    
2236    Added a call of SrcPath.sync at the beginning of Parse.parse (in CM).
2237    This fixes the problem of CM getting confused by files that suddenly
2238    change their identity (e.g., by getting unlinked and recreated by some
2239    text editor such as vi).  There might be a better/cheaper/cleaner way
2240    of doing this, but for now this will have to do.
2241    
2242    ----------------------------------------------------------------------
2243    Name: Matthias Blume
2244    Date: 2002/10/28 09:50:00 EST
2245    Tag: blume-20021028-typecheck
2246    Description:
2247    
2248    Exported structure Typecheck from $smlnj/viscomp/core.cm.
2249    
2250    ----------------------------------------------------------------------
2251    Name: Matthias Blume
2252    Date: 2002/10/17 09:10:00 EDT
2253    Tag: Release_110_42
2254    Description:
2255    
2256    In good old tradition, there has been a slight hiccup so that we have
2257    to patch 110.42 after the fact.   The old release tag has been replaced
2258    (see below).
2259    
2260    The change solves a problem with two competing approaches the
2261    configuration problem regarding MacOS 10.1 vs. MacOS 10.2 which got in
2262    each other's way.
2263    
2264    This change only affects the runtime system code and the installer script.
2265    (No new bootfiles.)
2266    
2267    ----------------------------------------------------------------------
2268    Name: Matthias Blume
2269    Date: 2002/10/16 12:00:00 EDT
2270    Tag: Release_110_42_removed
2271    Description:
2272    
2273    New working release.  New bootfiles.
2274    
2275    ----------------------------------------------------------------------
2276    Name: Matthias Blume
2277    Date: 2002/10/10 13:10:00 EDT
2278    Tag: blume-20021010-ppc-divs
2279    Description:
2280    
2281    The mltree operator DIVS must be implemented with an overflow check on
2282    the PPC because the hardware indicates divide-by-zero using "overflow" as
2283    well.
2284    
2285    ----------------------------------------------------------------------
2286    Name: Matthias Blume
2287    Date: 2002/07/23 11:45:00
2288    Tag: blume-20020723-smlnj-home
2289    Description:
2290    
2291    Sml now senses the SMLNJ_HOME environment variable.  If this is set,
2292    then the bin dir is assumed to be in $SMLNJ_HOME/bin and (unless
2293    CM_PATHCONFIG is also set), the path configuration file is assumed
2294    to be in $SMLNJ_HOME/lib/pathconfig.  This way one can easily move
2295    the entire tree to some other place and everything will "just work".
2296    
2297    (Companion commands such as ml-build and ml-makedepend also sense this
2298    variable.)
2299    
2300    ----------------------------------------------------------------------
2301    Name: Matthias Blume
2302    Date: 2002/07/12 21:19:00 EDT
2303    Tag: blume-20020712-liveness
2304    Description:
2305    
2306    Exported two useful "step" functions from liveness module (MLRISC).
2307    
2308    ----------------------------------------------------------------------
2309    Name: Matthias Blume
2310    Date: 2002/07/05 16:00 EDT
2311    Tag: Release_110_41
2312    Description:
2313    
2314    New working release.  New bootfiles.
2315    
2316    ----------------------------------------------------------------------
2317    Name: Matthias Blume
2318    Date: 2002/07/05 10:25:00 EDT
2319    Tag: blume-20020705-btimp
2320    Description:
2321    
2322    Exported structure BTImp from $smlnj/viscomp/debugprof.cm so that
2323    other clients can set up backtracing support.
2324    
2325    ----------------------------------------------------------------------
2326    Name: Matthias Blume
2327    Date: 2002/06/25 17:23:00 EDT
2328    Tag: blume-20020625-fpmax
2329    Description:
2330    
2331    Fixed a bug in translation of INLMAX (and INLMIN) for the floating-point
2332    case.  (The sense of the isNaN test was reversed -- which made min and
2333    max always return their first argument.)
2334    
2335    ----------------------------------------------------------------------
2336    Name: Matthias Blume
2337    Date: 2002/06/11
2338    Tag: blume-20020611-unixpath
2339    Description:
2340    
2341    Back-ported OS.Path.{from,to}UnixPath from idlbasis-devel branch.
2342    
2343    ----------------------------------------------------------------------
2344    Name: Matthias Blume
2345    Date: 2002/06/10 16:35:00 EDT
2346    Tag: blume-20020610-ieeereal
2347    Description:
2348    
2349    I back-ported my implementation of IEEEReal.fromString from the
2350    idlbasis-devel branch so that we can test it.
2351    
2352    Another small change is that ppDec tries to give more information
2353    than just "<sig>" in the case of functors.  However, this code is
2354    broken in some mysterious way if the functor's body's signature
2355    has not been declared by ascription but gets inferred from the
2356    implementation.  This needs fixing...
2357    
2358    ----------------------------------------------------------------------
2359    Name: Matthias Blume
2360    Date: 2002/05/31
2361    Tag: blume-20020531-btrace-mode
2362    Description:
2363    
2364    Resurrected SMLofNJ.Internals.BTrace.mode.  (It accidentally fell by
2365    the wayside when I switched over to using Controls everywhere.)
2366    
2367    ----------------------------------------------------------------------
2368    Name: Lal George
2369    Date: 2002/05/23 12:21:40 EDT
2370    Tag: george-20020523-visual-labels
2371    Description:
2372    
2373    Labels are now displayed in the graphical output to make
2374    the fall-through and target blocks obvious.
2375    
2376    ----------------------------------------------------------------------
2377    Name: Matthias Blume
2378    Date: 2002/05/22 11:03:00 EDT
2379    Tag: blume-20020522-shrink
2380    Description:
2381    
2382    John tweaked yesterday's fix for 1131 to handle an out-of-memory
2383    situation that comes up when allocating huge arrays.
2384    
2385    ----------------------------------------------------------------------
2386    Name: Matthias Blume
2387    Date: 2002/05/21 16:00:00 EDT
2388    Tag: Release_110_40
2389    Description:
2390    
2391    New working release (110.40).  New bootfiles.
2392    
2393    [Also: John Reppy fixed GC bug 1131.]
2394    
2395    ----------------------------------------------------------------------
2396    Name: Matthias Blume
2397    Date: 2002/05/21 12:35:00 EDT
2398    Tag: blume-20020521-cmdoc
2399    Description:
2400    
2401    CM documentation update.
2402    
2403    ----------------------------------------------------------------------
2404    Name: Matthias Blume
2405    Date: 2002/05/21 10:55:00 EDT
2406    Tag: blume-20020521-misc
2407    Description:
2408    
2409    - John tweaked runtime to be silent on heap export (except when
2410      GC messages are on).
2411    - I added a few more things (cross-compiling versions of CMB) to
2412      config/preloads (as suggestions).
2413    
2414    ----------------------------------------------------------------------
2415    Name: Matthias Blume
2416    Date: 2002/05/20 22:25:00 EDT
2417    Tag: blume-20020520-controls
2418    Description:
2419    
2420    - Added ControlUtil structure to control-lib.cm.
2421    - Use it throughout.
2422    - Used Controls facility to define MLRISC controls (as opposed to
2423      registering MLRISC control ref cells with Controls after the fact)
2424    - Fixed messed-up controls priorities.
2425    
2426    * Removed again all the stuff from config/preloads that one wouldn't
2427      be able to preload at the time the initial heap image is built.
2428      (Many libraries, e.g., CML, do not exist yet at this time.  The
2429       only libraries that can be preloaded via config/preloads are those
2430       that come bundled with the bootfiles.)
2431    
2432    ----------------------------------------------------------------------
2433    Name: Matthias Blume
2434    Date: 2002/05/20 10:59:00 EDT
2435    Tag: blume-20020520-preloads
2436    Description:
2437    
2438    Added a lot of commented-out suggestions for things to be included
2439    in config/preloads.
2440    
2441    ----------------------------------------------------------------------
2442    Name: Allen Leung
2443    Date: 2002/05/18 14:20:00 EDT
2444    Tag: leunga-20020518-mdl
2445    Description:
2446    
2447    o Made the mdl tool stuff compile and run again.
2448    o I've disabled all the stuff that depends on RTL specifications; they
2449      are all badly broken anyway.
2450    
2451    ----------------------------------------------------------------------
2452    Name: Matthias Blume
2453    Date: 2002/05/17 16:49:00 EDT
2454    Tag: blume-20020517-controls
2455    Description:
2456    
2457    0. John Reppy made several modifications to the SML/NJ library.
2458       In particular, there is a shiny new controls-lib.cm.
2459    
2460    1. Pushed new controls interface through compiler so that everything
2461       compiles again.
2462    
2463    2. Added FormatComb and FORMAT_COMB to the CML version of the
2464       SML/NJ library (so that CML compiles again).
2465    
2466    3. Modified init scripts because XXX_DEFAULT environment variables
2467       are no longer with us.  (Boot-time initialization is now done
2468       using the same environment variables that are also used for
2469       startup-time initialization of controls.)
2470    
2471    ----------------------------------------------------------------------
2472    Name: Lal George
2473    Date: 2002/05/15 09:20:10 EDT
2474    Tag: george-20020515-pseudo-op-decls
2475    Description:
2476    
2477            All pseudo-ops emitted before the first segment declaration
2478    such as TEXT, DATA, and BSS directives are assumed to be global
2479    declarations and are emitted first in the assembly file. This is
2480    useful in a number of situations where one has pseudo-ops that are not
2481    specific to any segment, and also works around the constraint that one
2482    cannot have client pseudo-ops in the TEXT segment.
2483    
2484            Because no segment is associated with these declarations it is
2485    an error to allocate any space or objects before the first segment
2486    directive and an exception will be raised. However, we cannot make
2487    this check for client pseudo-ops.
2488    
2489            These top level declarations are a field in the CFG graph_info.
2490    In theory you can continue to add to this field after the CFG has been
2491    built -- provided you know what you are doing;-)
2492    
2493    ----------------------------------------------------------------------
2494    Name: Matthias Blume
2495    Date: 2002/05/13 16:40:00 EDT
2496    Tag: blume-20020513-pp-etc
2497    Description:
2498    
2499    A few minor bugfixes:
2500    
2501      - Stopgap measure for bug recently reported by Elsa Gunter (ppDec).
2502        (Bogus printouts for redefined bindings still occur.  Compiler
2503        bug should no longer occur now.  We need to redo the prettyprinter
2504        from scratch.)
2505    
2506      - CM pathname printer now also adds escape sequences for ( and )
2507    
2508      - commend and docu fixes for ml-nlffi
2509    
2510    ----------------------------------------------------------------------
2511    Name: Matthias Blume
2512    Date: 2002/05/10 16:40:00 EDT
2513    Tag: blume-20020510-erg-textio
2514    Description:
2515    
2516    Applied the following bugfix provided by Emden Gansner:
2517    
2518        Output is corrupted when outputSubstr is used rather than output.
2519    
2520        The problem occurs when a substring
2521    
2522            ss = (s, dataStart, dataLen)
2523    
2524        where dataStart > 0, fills a stream buffer with avail bytes left.
2525        avail bytes of s, starting at index dataStart, are copied into the
2526        buffer, the buffer is flushed, and then the remaining dataLen-avail
2527        bytes of ss are copied into the beginning of the buffer. Instead of
2528        starting this copy at index dataStart+avail in s, the current code
2529        starts the copy at index avail.
2530    
2531        Fix:
2532        In text-io-fn.sml, change line 695 from
2533             val needsFlush = copyVec(v, avail, dataLen-avail, buf, 0)
2534        to
2535             val needsFlush = copyVec(v, dataStart+avail, dataLen-avail, buf, 0)
2536    
2537    ----------------------------------------------------------------------
2538    Name: Matthias Blume
2539    Date: 2002/04/12 13:55:00 EDT
2540    Tag: blume-20020412-assyntax
2541    Description:
2542    
2543    1. Grabbed newer assyntax.h from the XFree86 project.
2544    2. Fiddled with how to compile X86.prim.asm without warnings.
2545    3. (Very) Minor cleanup in CM.
2546    
2547    ----------------------------------------------------------------------
2548    Name: Matthias Blume
2549    Date: 2002/04/01 (no joke!) 17:07:00 EST
2550    Tag: blume-20020401-x86div
2551    Description:
2552    
2553    Added full support for div/mod/rem/quot on the x86, using the machine
2554    instruction's two results (without clumsily recomputing the remainder)
2555    directly where appropriate.
2556    
2557    Some more extensive power-of-two support was added to the x86 instruction
2558    selector (avoiding expensive divs, mods, and muls where they can be
2559    replaced with cheaper shifts and masks).  However, this sort of thing
2560    ought to be done earlier, e.g., within the CPS optimizer so that
2561    all architectures benefit from it.
2562    
2563    The compiler compiles to a fixed point, but changes might be somewhat
2564    fragile nevertheless.  Please, report any strange things that you might
2565    see wrt. div/mod/quot/rem...
2566    
2567    ----------------------------------------------------------------------
2568    Name: Matthias Blume
2569    Date: 2002/03/29 17:22:00
2570    Tag: blume-20020329-div
2571    Description:
2572    
2573    Fixed my broken div/mod logic.  Unfortunately, this means that the
2574    inline code for div/mod now has one more comparison than before.
2575    Fast paths (quotient > 0 or remainder = 0) are not affected, though.
2576    The problem was with quotient = 0, because that alone does not tell
2577    us which way the rounding went.  One then has to look at whether
2578    remainder and divisor have the same sign...  :(
2579    
2580    Anyway, I replaced the bootfiles with fresh ones...
2581    
2582    ----------------------------------------------------------------------
2583    Name: Matthias Blume
2584    Date: 2002/03/29 14:10:00 EST
2585    Tag: blume-20020329-inlprims
2586    Description:
2587    
2588    NEW BOOTFILES!!!    Version number bumped to 110.39.3.
2589    
2590    Primops have changed. This means that the bin/boot-file formats have
2591    changed as well.
2592    
2593    To make sure that there is no confusion, I made a new version.
2594    
2595    
2596    CHANGES:
2597    
2598    * removed REMT from mltree (remainder should never overflow).
2599    
2600    * added primops to deal with divisions of all flavors to the frontend
2601    
2602    * handled these primops all the way through so they map to their respective
2603      MLRISC support
2604    
2605    * used these primops in the implementation of Int, Int32, Word, Word32
2606    
2607    * removed INLDIV, INLMOD, and INLREM as they are no longer necessary
2608    
2609    * parameterized INLMIN, INLMAX, and INLABS by a numkind
2610    
2611    * translate.sml now deals with all flavors of INL{MIN,MAX,ABS}, including
2612      floating point
2613    
2614    * used INL{MIN,MAX,ABS} in the implementation of Int, Int32, Word, Word32,
2615      and Real (but Real.abs maps to a separate floating-point-only primop)
2616    
2617    
2618    TODO items:
2619    
2620    * Hacked Alpha32 instruction selection, disabling the selection of REMx
2621      instructions because the machine instruction encoder cannot handle
2622      them.  (Hppa, PPC, and Sparc instruction selection did not handle
2623      REM in the first place, and REM is supported by the x86 machine coder.)
2624    
2625    * Handle DIV and MOD with DIV_TO_NEGINF directly in the x86 instruction
2626      selection phase.  (The two can be streamlined because the hardware
2627      delivers both quotient and remainder at the same time anyway.)
2628    
2629    * Think about what to do with "valOf(Int32.minInt) div ~1" and friends.
2630      (Currently the behavior is inconsistent both across architectures and
2631      wrt. the draft Basis spec.)
2632    
2633    * Word8 should eventually be handled natively, too.
2634    
2635    * There seems to be one serious bug in mltree-gen.sml.  It appears, though,
2636      as if there currently is no execution path that could trigger it in
2637      SML/NJ.  (The assumptions underlying functions arith and promotable do not
2638      hold for things like multiplication and division.)
2639    
2640    ----------------------------------------------------------------------
2641    Name: Matthias Blume
2642    Date: 2002/03/27 16:27:00 EST
2643    Tag: blume-20020327-mlrisc-divisions
2644    Description:
2645    
2646    Added support for all four division operations (ML's div, mod, quot,
2647    and rem) to MLRISC.  In the course of doing so, I also rationalized
2648    the naming (no more annoying switch-around of DIV and QUOT), by
2649    parameterizing the operation by div_rounding_mode (which can be either
2650    DIV_TO_ZERO or DIV_TO_NEGINF).
2651    
2652    The generic MLTreeGen functor takes care of compiling all four
2653    operations down to only round-to-zero div.
2654    
2655    Missing pieces:
2656    
2657      * Doing something smarter than relying on MLTreeGen on architectures
2658        like, e.g., the x86 where hardware division delivers both quotient and
2659        remainder at the same time.  With this, the implementation of the
2660        round-to-neginf operations could be further streamlined.
2661    
2662      * Remove inlining support for div/mod/rem from the frontend and replace it
2663        with primops that get carried through to the backend.  Do this for all
2664        int and word types.
2665    
2666    ----------------------------------------------------------------------
2667    Name: Matthias Blume
2668    Date: 2002/03/25 17:25:00 EST
2669    Tag: blume-20020325-divmod
2670    Description:
2671    
2672    I improved (hopefully without breaking them) the implementation of Int.div,
2673    Int.mod, and Int.rem.   For this, the code in translate.sml now takes
2674    advantage of the following observations:
2675    
2676      Let  q = x quot y      r = x rem y
2677           d = x div  y      m = x mod y
2678    
2679    where "quot" is the round-to-zero version of integer division that
2680    hardware usually provides.  Then we have:
2681    
2682         r = x - q * y        where neither the * nor the - will overflow
2683         d = if q >= 0 orelse x = q * y then q else q - 1
2684                              where neither the * nor the - will overflow
2685         m = if q >= 0 orelse r = 0 then r else r + y
2686                              where the + will not overflow
2687    
2688    This results in substantial simplification of the generated code.
2689    The following table shows the number of CFG nodes and edges generated
2690    for
2691            fun f (x, y) = x OPER y
2692            (* with OPER \in div, mod, quot, rem *)
2693    
2694    
2695        OPER | nodes(old) | edges(old) | nodes(new) | edges(new)
2696        --------------------------------------------------------
2697         div |         24 |         39 |         12 |         16
2698         mod |         41 |         71 |         12 |         16
2699        quot |          8 |         10 |          8 |         10
2700         rem |         10 |         14 |          8 |         10
2701    
2702    
2703    ----------------------------------------------------------------------
2704    Name: Matthias Blume
2705    Date: 2002/03/25 22:06:00 EST
2706    Tag: blume-20020325-cprotobug
2707    Description:
2708    
2709    Fixed a bug in cproto (c prototype decoder).
2710    
2711    ----------------------------------------------------------------------
2712    Name: Matthias Blume
2713    Date: 2002/03/25 16:00:00 EST
2714    Tag: blume-20020325-raw-primops
2715    Description:
2716    
2717    I did some cleanup to Allen's new primop code and
2718    replaced yesterday's bootfiles with new ones.
2719    (But they are stored in the same place.)
2720    
2721    ----------------------------------------------------------------------
2722    Name: Matthias Blume
2723    Date: 2002/03/24 22:40:00 EST
2724    Tag: blume-20020324-bootfiles
2725    Description:
2726    
2727    Made the bootfiles that Allen asked for.
2728    
2729    ----------------------------------------------------------------------
2730    Name: Allen Leung
2731    Date: 2002/03/23 15:50:00 EST
2732    Tag: leunga-20020323-flint-cps-rcc-primops
2733    Description:
2734    
2735      1. Changes to FLINT primops:
2736    
2737        (* make a call to a C-function;
2738         * The primop carries C function prototype information and specifies
2739         * which of its (ML-) arguments are floating point. C prototype
2740         * information is for use by the backend, ML information is for
2741         * use by the CPS converter. *)
2742      | RAW_CCALL of { c_proto: CTypes.c_proto,
2743                       ml_args: ccall_type list,
2744                       ml_res_opt: ccall_type option,
2745                       reentrant : bool
2746                     } option
2747       (* Allocate uninitialized storage on the heap.
2748        * The record is meant to hold short-lived C objects, i.e., they
2749        * are not ML pointers.  With the tag, the representation is
2750        * the same as RECORD with tag tag_raw32 (sz=4), or tag_fblock (sz=8)
2751        *)
2752      | RAW_RECORD of {tag:bool,sz:int}
2753      and ccall_type = CCALL_INT32 | CCALL_REAL64 | CCALL_ML_PTR
2754    
2755      2.  These CPS primops are now overloaded:
2756    
2757           rawload of {kind:numkind}
2758           rawstore of {kind:numkind}
2759    
2760          The one argument form is:
2761    
2762             rawload {kind} address
2763    
2764          The two argument form is:
2765    
2766             rawload {kind} [ml object, byte-offset]
2767    
2768      3. RAW_CCALL/RCC now takes two extra arguments:
2769    
2770         a. The first is whether the C call is reentrant, i.e., whether
2771            ML state should be saved and restored.
2772         b. The second argument is a string argument specifying the name of
2773            library and the C function.
2774    
2775         These things are currently not handled in the code generator, yet.
2776    
2777      4. In CProto,
2778    
2779         An encoding type of "bool" means "ml object" and is mapped into
2780         C prototype of PTR.  Note that "bool" is different than "string",
2781         even though "string" is also mapped into PTR, because "bool"
2782         is assigned an CPS type of BOGt, while "string" is assigned INT32t.
2783    
2784      5. Pickler/unpicker
2785    
2786         Changed to handle RAW_RECORD and newest RAW_CCALL
2787    
2788      6. MLRiscGen,
2789    
2790         1. Changed to handle the new rawload/rawstore/rawrecord operators.
2791         2. Code for handling C Calls has been moved to a new module CPSCCalls,
2792            in the file CodeGen/cpscompile/cps-c-calls.sml
2793    
2794      7. Added the conditional move operator
2795    
2796             condmove of branch
2797    
2798         to cps.  Generation of this is still buggy so it is currently
2799         disabled.
2800    
2801    ----------------------------------------------------------------------
2802    Name: Lal George
2803    Date: 2002/03/22 14:18:25 EST
2804    Tag: george-20020322-cps-branch-prob
2805    Description:
2806    
2807    Implemented the Ball-Larus branch prediction-heuristics, and
2808    incorporated graphical viewers for control flow graphs.
2809    
2810    Ball-Larus Heuristics:
2811    ---------------------
2812    See the file compiler/CodeGen/cpscompile/cpsBranchProb.sml.
2813    
2814    By design it uses the Dempster-Shafer theory for combining
2815    probabilities.  For example, in the function:
2816    
2817        fun f(n,acc) = if n = 0 then acc else f(n-1, n*acc)
2818    
2819    the ball-larus heuristics predicts that the n=0 is unlikely
2820    (OH-heuristic), and the 'then' branch is unlikely because of the
2821    RH-heuristic -- giving the 'then' branch an even lower combined
2822    probability using the Dempster-Shafer theory.
2823    
2824    Finally, John Reppy's loop analysis in MLRISC, further lowers the
2825    probability of the 'then' branch because of the loop in the else
2826    branch.
2827    
2828    
2829    Graphical Viewing:
2830    ------------------
2831    I merely plugged in Allen's graphical viewers into the compiler. The
2832    additional code is not much. At the top level, saying:
2833    
2834            Control.MLRISC.getFlag "cfg-graphical-view" := true;
2835    
2836    will display the graphical view of the control flow graph just before
2837    back-patching.  daVinci must be in your path for this to work. If
2838    daVinci is not available, then the default viewer can be changed
2839    using:
2840    
2841            Control.MLRISC.getString "viewer"
2842    
2843    which can be set to "dot" or "vcg" for the corresponding viewers. Of
2844    course, these viewers must be in your path.
2845    
2846    The above will display the compilation unit at the level of clusters,
2847    many of which are small, boring, and un-interesting. Also setting:
2848    
2849            Control.MLRISC.getInt "cfg-graphical-view_size"
2850    
2851    will display clusters that are larger than the value set by the above.
2852    
2853    
2854    ----------------------------------------------------------------------
2855    Name: Matthias Blume
2856    Date: 2002/03/21 22:20:00 EST
2857    Tag: blume-20020321-kmp-bugfix
2858    Description:
2859    
2860    Changed the interface to the KMP routine in PreString and fixed
2861    a minor bug in one place where it was used.
2862    
2863    ----------------------------------------------------------------------
2864    Name: Allen Leung
2865    Date: 2002/03/21 20:30:00 EST
2866    Tag: leunga-20020321-cfg
2867    Description:
2868    
2869      Fixed a potential problem in cfg edge splitting.
2870    
2871    ----------------------------------------------------------------------
2872    Name: Allen Leung
2873    Date: 2002/03/21 17:15:00 EST
2874    Tag: leunga-20020321-x86-fp-cfg
2875    Description:
2876    
2877      1. Recoded the buggy parts of x86-fp.
2878    
2879         a. All the block reordering code has been removed.
2880            We now depend on the block placement phases to do this work.
2881    
2882         b. Critical edge splitting code has been simplified and moved into the
2883            CFG modules, as where they belong.
2884    
2885         Both of these were quite buggy and complex.  The code is now much, much
2886         simpler.
2887    
2888      2. X86 backend.
2889    
2890         a. Added instructions for 64-bit support.  Instruction selection for
2891            64-bit has not been committed, however, since that
2892            requires changes to MLTREE which haven't been approved by
2893            Lal and John.
2894    
2895         b. Added support for FUCOMI and FUCOMIP when generating code for
2896            PentiumPro and above.  We only generate these instructions in
2897            the fast-fp mode.
2898    
2899         c. Added cases for JP and JNP in X86FreqProps.
2900    
2901      3. CFG
2902    
2903         CFG now has a bunch of methods for edge splitting and merging.
2904    
2905      4. Machine description.
2906    
2907         John's simplification of MLTREE_BASIS.fcond broke a few machine
2908         description things:
2909    
2910         rtl-build.{sig,sml} and hppa.mdl fixed.
2911    
2912         NOTE: the machine description stuff in the repository is still broken.
2913               Again, I can't put my fixes in because that involves
2914               changes to MLTREE.
2915    
2916    ----------------------------------------------------------------------
2917    Name: Matthias Blume
2918    Date: 2002/03/20 15:55:00 EST
2919    Tag: blume-20020320-kmp
2920    Description:
2921    
2922    Implemented Knuth-Morris-Pratt string matching in PreString and used
2923    it for String.isSubstring, Substring.isSubstring, and
2924    Substring.position.
2925    
2926    (Might need some stress-testing.  Simple examples worked fine.)
2927    
2928    ----------------------------------------------------------------------
2929    Name: Matthias Blume
2930    Date: 2002/03/19 16:37:00 EST
2931    Tag: blume-20020319-witnesses
2932    Description:
2933    
2934    Added a structure C.W and functions convert/Ptr.convert to ml-nlffi-lib.
2935    
2936    This implements a generic mechanism for changing constness qualifiers
2937    anywhere within big C types without resorting to outright "casts".
2938    (So far, functions such as C.rw/C.ro or C.Ptr.rw/C.Ptr.ro only let you
2939    modify the constness at the outermost level.)
2940    The implementation of "convert" is based on the idea of "witness"
2941    values -- values that are not used by the operation but whose types
2942    "testify" to their applicability.  On the implementation side, "convert"
2943    is simply a projection (returning its second curried argument).  With
2944    cross-module inlining, it should not result in any machine code being
2945    generated.
2946    
2947    ----------------------------------------------------------------------
2948    Name: Matthias Blume
2949    Date: 2002/03/15 16:40:00 EST
2950    Tag: blume-20020315-basis
2951    Description:
2952    
2953    Provided (preliminary?) implementations for
2954    
2955      {String,Substring}.{concatWith,isSuffix,isSubstring}
2956    
2957    and
2958    
2959      Substring.full
2960    
2961    Those are in the Basis spec but they were missing in SML/NJ.
2962    
2963    ----------------------------------------------------------------------
2964    Name: Matthias Blume
2965    Date: 2002/03/14 21:30:00 EST
2966    Tag: blume-20020314-controls
2967    Description:
2968    
2969    Controls:
2970    ---------
2971    
2972    1. Factored out the recently-added Controls : CONTROLS stuff and put
2973       it into its own library $/controls-lib.cm.  The source tree for
2974       this is under src/smlnj-lib/Controls.
2975    
2976    2. Changed the names of types and functions in this interface, so they
2977       make a bit more "sense":
2978    
2979          module -> registry
2980          'a registry -> 'a group
2981    
2982    3. The interface now deals in ref cells only.  The getter/setter interface
2983       is (mostly) gone.
2984    
2985    4. Added a function that lets one register an already-existing ref cell.
2986    
2987    5. Made the corresponding modifications to the rest of the code so that
2988       everything compiles again.
2989    
2990    6. Changed the implementation of Controls.MLRISC back to something closer
2991       to the original.  In particular, this module (and therefore MLRISC)
2992       does not depend on Controls.  There now is some link-time code in
2993       int-sys.sml that registers the MLRISC controls with the Controls
2994       module.
2995    
2996    CM:
2997    ---
2998    
2999      * One can now specify the lambda-split aggressiveness in init.cmi.
3000    
3001    ----------------------------------------------------------------------
3002    Name: Allen Leung
3003    Date: 2002/03/13 17:30:00 EST
3004    Tag: leunga-20020313-x86-fp-unary
3005    Description:
3006    
3007    Bug fix for:
3008    
3009    > leunga@weaselbane:~/Yale/tmp/sml-dist{21} bin/sml
3010    > Standard ML of New Jersey v110.39.1 [FLINT v1.5], March 08, 2002
3011    > - fun f(x,(y,z)) = Real.~ y;
3012    > [autoloading]
3013    > [autoloading done]
3014    >       fchsl   (%eax), 184(%esp)
3015    > Error: MLRisc bug: X86MCEmitter.emitInstr
3016    >
3017    > uncaught exception Error
3018    >   raised at: ../MLRISC/control/mlriscErrormsg.sml:16.14-16.19
3019    
3020    The problem was that the code generator did not generate any fp registers
3021    in this case, and the ra didn't know that it needed to run the X86FP phase to
3022    translate the pseudo fp instruction.   This only happened with unary fp
3023    operators in certain situations.
3024    
3025    ----------------------------------------------------------------------
3026    Name: Matthias Blume
3027    Date: 2002/03/13 14:00:00 EST
3028    Tag: blume-20020313-overload-etc
3029    Description:
3030    
3031    1. Added _overload as a synonym for overload for backward compatibility.
3032       (Control.overloadKW must be true for either version to be accepted.)
3033    
3034    2. Fixed bug in install script that caused more things to be installed
3035       than what was requested in config/targets.
3036    
3037    3. Made CM aware of the (_)overload construct so that autoloading
3038       works.
3039    
3040    ----------------------------------------------------------------------
3041    Name: Matthias Blume
3042    Date: 2002/03/12 22:03:00 EST
3043    Tag: blume-20020312-url
3044    Description:
3045    
3046    Forgot to update BOOT and srcarchiveurl.
3047    
3048    ----------------------------------------------------------------------
3049    Name: Matthias Blume
3050    Date: 2002/03/12 17:30:00 EST
3051    Tag: blume-20020312-version110392
3052    Description:
3053    
3054    Yet another version number bump (because of small changes to the
3055    binfile format).  Version number is now 110.39.2.  NEW BOOTFILES!
3056    
3057    Changes:
3058    
3059      The new pid generation scheme described a few weeks ago was overly
3060      complicated.  I implemented a new mechanism that is simpler and
3061      provides a bit more "stability":  Once CM has seen a compilation
3062      unit, it keeps its identity constant (as long as you do not delete
3063      those crucial CM/GUID/* files).  This means that when you change
3064      an interface, compile, then go back to the old interface, and
3065      compile again, you arrive at the original pid.
3066    
3067      There now also is a mechanism that instructs CM to use the plain
3068      environment hash as a module's pid (effectively making its GUID
3069      the empty string).  For this, "noguid" must be specified as an
3070      option to the .sml file in question within its .cm file.
3071      This is most useful for code that is being generated by tools such
3072      as ml-nlffigen (because during development programmers tend to
3073      erase the tool's entire output directory tree including CM's cached
3074      GUIDs).  "noguid" is somewhat dangerous (since it can be used to locally
3075      revert to the old, broken behavior of SML/NJ, but in specific cases
3076      where there is no danger of interface confusion, its use is ok
3077      (I think).
3078    
3079      ml-nlffigen by default generates "noguid" annotations.  They can be
3080      turned off by specifying -guid in its command line.
3081    
3082    ----------------------------------------------------------------------
3083    Name: Lal George
3084    Date: 2002/03/12 12 14:42:36 EST
3085    Tag: george-20020312-frequency-computation
3086    Description:
3087    
3088    Integrated jump chaining and static block frequency into the
3089    compiler. More details and numbers later.
3090    
3091    ----------------------------------------------------------------------
3092    Name: Lal George
3093    Date: 2002/03/11 11 22:38:53 EST
3094    Tag: george-20020311-jump-chain-elim
3095    Description:
3096    
3097    Tested the jump chain elimination on all architectures (except the
3098    hppa).  This is on by default right now and is profitable for the
3099    alpha and x86, however, it may not be profitable for the sparc and ppc
3100    when compiling the compiler.
3101    
3102    The gc test will typically jump to a label at the end of the cluster,
3103    where there is another jump to an external cluster containing the actual
3104    code to invoke gc. This is to allow factoring of common gc invocation
3105    sequences. That is to say, we generate:
3106    
3107            f:
3108               testgc
3109               ja   L1      % jump if above to L1
3110    
3111            L1:
3112               jmp L2
3113    
3114    
3115    After jump chain elimination the 'ja L1' instructions is converted to
3116    'ja L2'. On the sparc and ppc, many of the 'ja L2' instructions may end
3117    up being implemented in their long form (if L2 is far away) using:
3118    
3119            jbe     L3      % jump if below or equal to L3
3120            jmp     L2
3121         L3:
3122            ...
3123    
3124    
3125    For large compilation units L2  may be far away.
3126    
3127    
3128    ----------------------------------------------------------------------
3129    Name: Matthias Blume
3130    Date: 2002/03/11 13:30:00 EST
3131    Tag: blume-20020311-mltreeeval
3132    Description:
3133    
3134    A functor parameter was missing.
3135    
3136    ----------------------------------------------------------------------
3137    Name: Allen Leung
3138    Date: 2002/03/11 10:30:00 EST
3139    Tag: leunga-20020311-runtime-string0
3140    Description:
3141    
3142       The representation of the empty string now points to a
3143    legal null terminated C string instead of unit.  It is now possible
3144    to convert an ML string into C string with InlineT.CharVector.getData.
3145    This compiles into one single machine instruction.
3146    
3147    ----------------------------------------------------------------------
3148    Name: Allen Leung
3149    Date: 2002/03/10 23:55:00 EST
3150    Tag: leunga-20020310-x86-call
3151    Description:
3152    
3153       Added machine generation for CALL instruction (relative displacement mode)
3154    
3155    ----------------------------------------------------------------------
3156    Name: Matthias Blume
3157    Date: 2002/03/08 16:05:00
3158    Tag: blume-20020308-entrypoints
3159    Description:
3160    
3161    Version number bumped to 110.39.1.  NEW BOOTFILES!
3162    
3163    Entrypoints: non-zero offset into a code object where execution should begin.
3164    
3165    - Added the notion of an entrypoint to CodeObj.
3166    - Added reading/writing of entrypoint info to Binfile.
3167    - Made runtime system bootloader aware of entrypoints.
3168    - Use the address of the label of the first function given to mlriscGen
3169      as the entrypoint.  This address is currently always 0, but it will
3170      not be 0 once we turn on block placement.
3171    - Removed the linkage cluster code (which was The Other Way(tm) of dealing
3172      with entry points) from mlriscGen.
3173    
3174    ----------------------------------------------------------------------
3175    Name: Allen Leung
3176    Date: 2002/03/07 20:45:00 EST
3177    Tag: leunga-20020307-x86-cmov
3178    Description:
3179    
3180       Bug fixes for CMOVcc on x86.
3181    
3182       1. Added machine code generation for CMOVcc
3183       2. CMOVcc is now generated in preference over SETcc on PentiumPro or above.
3184       3. CMOVcc cannot have an immediate operand as argument.
3185    
3186    ----------------------------------------------------------------------
3187    Name: Matthias Blume
3188    Date: 2002/03/07 16:15:00 EST
3189    Tag: blume-20020307-controls
3190    Description:
3191    
3192    This is a very large but mostly boring patch which makes (almost)
3193    every tuneable compiler knob (i.e., pretty much everything under
3194    Control.* plus a few other things) configurable via both the command
3195    line and environment variables in the style CM did its configuration
3196    until now.
3197    
3198    Try starting sml with '-h' (or, if you are brave, '-H')
3199    
3200    To this end, I added a structure Controls : CONTROLS to smlnj-lib.cm which
3201    implements the underlying generic mechanism.
3202    
3203    The interface to some of the existing such facilities has changed somewhat.
3204    For example, the MLRiscControl module now provides mkFoo instead of getFoo.
3205    (The getFoo interface is still there for backward-compatibility, but its
3206    use is deprecated.)
3207    
3208    The ml-build script passes -Cxxx=yyy command-line arguments through so
3209    that one can now twiddle the compiler settings when using this "batch"
3210    compiler.
3211    
3212    TODO items:
3213    
3214    We should go through and throw out all controls that are no longer
3215    connected to anything.  Moreover, we should go through and provide
3216    meaningful (and correct!) documentation strings for those controls
3217    that still are connected.
3218    
3219    Currently, multiple calls to Controls.new are accepted (only the first
3220    has any effect).  Eventually we should make sure that every control
3221    is being made (via Controls.new) exactly once.  Future access can then
3222    be done using Controls.acc.
3223    
3224    Finally, it would probably be a good idea to use the getter-setter
3225    interface to controls rather than ref cells.  For the time being, both
3226    styles are provided by the Controls module, but getter-setter pairs are
3227    better if thread-safety is of any concern because they can be wrapped.
3228    
3229    *****************************************
3230    
3231    One bug fix: The function blockPlacement in three of the MLRISC
3232    backpatch files used to be hard-wired to one of two possibilities at
3233    link time (according to the value of the placementFlag).  But (I
3234    think) it should rather sense the flag every time.
3235    
3236    *****************************************
3237    
3238    Other assorted changes (by other people who did not supply a HISTORY entry):
3239    
3240    1. the cross-module inliner now works much better (Monnier)
3241    2. representation of weights, frequencies, and probabilities in MLRISC
3242       changed in preparation of using those for weighted block placement
3243       (Reppy, George)
3244    
3245    ----------------------------------------------------------------------
3246    Name: Lal George
3247    Date: 2002/03/07 14:44:24 EST 2002
3248    Tag: george-20020307-weighted-block-placement
3249    
3250    Tested the weighted block placement optimization on all architectures
3251    (except the hppa) using AMPL to generate the block and edge frequencies.
3252    Changes were required in the machine properties to correctly
3253    categorize trap instructions. There is an MLRISC flag
3254    "weighted-block-placement" that can be used to enable weighted block
3255    placement, but this will be ineffective without block/edge
3256    frequencies (coming soon).
3257    
3258    
3259    ----------------------------------------------------------------------
3260    Name: Lal George
3261    Date: 2002/03/05 17:24:48 EST
3262    Tag: george-20020305-linkage-cluster
3263    
3264    In order to support the block placement optimization, a new cluster
3265    is generated as the very first cluster (called the linkage cluster).
3266    It contains a single jump to the 'real' entry point for the compilation
3267    unit. Block placement has no effect on the linkage cluster itself, but
3268    all the other clusters  have full freedom in the manner in which they
3269    reorder blocks or functions.
3270    
3271    On the x86 the typical linkage code that is generated is:
3272       ----------------------
3273            .align 2
3274       L0:
3275            addl    $L1-L0, 72(%esp)
3276            jmp     L1
3277    
3278    
3279            .align  2
3280       L1:
3281       ----------------------
3282    
3283    72(%esp) is the memory location for the stdlink register. This
3284    must contain the address of the CPS function being called. In the
3285    above example, it contains the address of  L0; before
3286    calling L1 (the real entry point for the compilation unit), it
3287    must contain the address for L1, and hence
3288    
3289            addl $L1-L0, 72(%esp)
3290    
3291    I have tested this on all architectures except the hppa.The increase
3292    in code size is of course negligible
3293    
3294    ----------------------------------------------------------------------
3295    Name: Allen Leung
3296    Date: 2002/03/03 13:20:00 EST
3297    Tag: leunga-20020303-mlrisc-tools
3298    
3299      Added #[ ... ] expressions to mlrisc tools
3300    
3301    ----------------------------------------------------------------------
3302    Name: Matthias Blume
3303    Date: 2002/02/27 12:29:00 EST
3304    Tag: blume-20020227-cdebug
3305    Description:
3306    
3307    - made types in structure C and C_Debug to be equal
3308    - got rid of code duplication (c-int.sml vs. c-int-debug.sml)
3309    - there no longer is a C_Int_Debug (C_Debug is directly derived from C)
3310    
3311    ----------------------------------------------------------------------
3312    Name: Matthias Blume
3313    Date: 2002/02/26 12:00:00 EST
3314    Tag: blume-20020226-ffi
3315    Description:
3316    
3317    1. Fixed a minor bug in CM's "noweb" tool:
3318       If numbering is turned off, then truly don't number (i.e., do not
3319       supply the -L option to noweb).  The previous behavior was to supply
3320       -L'' -- which caused noweb to use the "default" line numbering scheme.
3321       Thanks to Chris Richards for pointing this out (and supplying the fix).
3322    
3323    2. Once again, I reworked some aspects of the FFI:
3324    
3325       A. The incomplete/complete type business:
3326    
3327       - Signatures POINTER_TO_INCOMPLETE_TYPE and accompanying functors are
3328         gone!
3329       - ML types representing an incomplete type are now *equal* to
3330         ML types representing their corresponding complete types (just like
3331         in C).  This is still safe because ml-nlffigen will not generate
3332         RTTI for incomplete types, nor will it generate functions that
3333         require access to such RTTI.   But when ML code generated from both
3334         incomplete and complete versions of the C type meet, the ML types
3335         are trivially interoperable.
3336    
3337         NOTE:  These changes restore the full generality of the translation
3338         (which was previously lost when I eliminated functorization)!
3339    
3340       B. Enum types:
3341    
3342       - Structure C now has a type constructor "enum" that is similar to
3343         how the "su" constructor works.  However, "enum" is not a phantom
3344         type because each "T enum" has values (and is isomorphic to
3345         MLRep.Signed.int).
3346       - There are generic access operations for enum objects (using
3347         MLRep.Signed.int).
3348       - ml-nlffigen will generate a structure E_foo for each "enum foo".
3349         * The structure contains the definition of type "mlrep" (the ML-side
3350         representation type of the enum).  Normally, mlrep is the same
3351         as "MLRep.Signed.int", but if ml-nlffigen was invoked with "-ec",
3352         then mlrep will be defined as a datatype -- thus facilitating
3353         pattern matching on mlrep values.
3354         ("-ec" will be suppressed if there are duplicate values in an
3355          enumeration.)
3356         * Constructors ("-ec") or values (no "-ec") e_xxx of type mlrep
3357         will be generated for each C enum constant xxx.
3358         * Conversion functions m2i and i2m convert between mlrep and
3359         MLRep.Signed.int.  (Without "-ec", these functions are identities.)
3360         * Coversion functions c and ml convert between mlrep and "tag enum".
3361         * Access functions (get/set) fetch and store mlrep values.
3362       - By default (unless ml-nlffigen was invoked with "-nocollect"), unnamed
3363         enumerations are merged into one single enumeration represented by
3364         structure E_'.
3365    
3366    ----------------------------------------------------------------------
3367    Name: Allen Leung
3368    Date: 2002/02/25 04:45:00 EST
3369    Tag: leunga-20020225-cps-spill
3370    
3371    This is a new implementation of the CPS spill phase.
3372    The new phase is in the new file compiler/CodeGen/cpscompile/spill-new.sml
3373    In case of problems, replace it with the old file spill.sml
3374    
3375    The current compiler runs into some serious performance problems when
3376    constructing a large record.  This can happen when we try to compile a
3377    structure with many items.  Even a very simple structure like the following
3378    makes the compiler slow down.
3379    
3380        structure Foo = struct
3381           val x_1 = 0w1 : Word32.int
3382           val x_2 = 0w2 : Word32.int
3383           val x_3 = 0w3 : Word32.int
3384           ...
3385           val x_N = 0wN : Word32.int
3386        end
3387    
3388    The following table shows the compile time, from N=1000 to N=4000,
3389    with the old compiler:
3390    
3391    N
3392    1000   CPS 100 spill                           0.04u  0.00s  0.00g
3393           MLRISC ra                               0.06u  0.00s  0.05g
3394              (spills = 0 reloads = 0)
3395           TOTAL                                   0.63u  0.07s  0.21g
3396    
3397    1100   CPS 100 spill                           8.25u  0.32s  0.64g
3398           MLRISC ra                               5.68u  0.59s  3.93g
3399              (spills = 0 reloads = 0)
3400           TOTAL                                   14.71u  0.99s  4.81g
3401    
3402    1500   CPS 100 spill                           58.55u  2.34s  1.74g
3403           MLRISC ra                               5.54u  0.65s  3.91g
3404              (spills = 543 reloads = 1082)
3405           TOTAL                                   65.40u  3.13s  6.00g
3406    
3407    2000   CPS 100 spill                           126.69u  4.84s  3.08g
3408           MLRISC ra                               0.80u  0.10s  0.55g
3409              (spills = 42 reloads = 84)
3410           TOTAL                                   129.42u  5.10s  4.13g
3411    
3412    3000   CPS 100 spill                           675.59u  19.03s  11.64g
3413           MLRISC ra                               2.69u  0.27s  1.38g
3414              (spills = 62 reloads = 124)
3415           TOTAL                                   682.48u  19.61s  13.99g
3416    
3417    4000   CPS 100 spill                           2362.82u  56.28s  43.60g
3418           MLRISC ra                               4.96u  0.27s  2.72g
3419              (spills = 85 reloads = 170)
3420           TOTAL                                   2375.26u  57.21s  48.00g
3421    
3422    As you can see the old cps spill module suffers from some serious
3423    performance problem.  But since I cannot decipher the old code fully,
3424    instead of patching the problems up, I'm reimplementing it
3425    with a different algorithm.  The new code is more modular,
3426    smaller when compiled, and substantially faster
3427    (O(n log n) time and O(n) space).  Timing of the new spill module:
3428    
3429    4000  CPS 100 spill                           0.02u  0.00s  0.00g
3430          MLRISC ra                               0.25u  0.02s  0.15g
3431             (spills=1 reloads=3)
3432          TOTAL                                   7.74u  0.34s  1.62g
3433    
3434    Implementation details:
3435    
3436    As far as I can tell, the purpose of the CPS spill module is to make sure the
3437    number of live variables at any program point (the bandwidth)
3438    does not exceed a certain limit, which is determined by the
3439    size of the spill area.
3440    
3441    When the bandwidth is too large, we decrease the register pressure by
3442    packing live variables into spill records.  How we achieve this is
3443    completely different than what we did in the old code.
3444    
3445    First, there is something about the MLRiscGen code generator
3446    that we should be aware of:
3447    
3448    o MLRiscGen performs code motion!
3449    
3450       In particular, it will move floating point computations and
3451       address computations involving only the heap pointer to
3452       their use sites (if there is only a single use).
3453       What this means is that if we have a CPS record construction
3454       statement
3455    
3456           RECORD(k,vl,w,e)
3457    
3458       we should never count the new record address w as live if w
3459       has only one use (which is often the case).
3460    
3461       We should do something similar to floating point, but the transformation
3462       there is much more complex, so I won't deal with that.
3463    
3464    Secondly, there are now two new cps primops at our disposal:
3465    
3466     1. rawrecord of record_kind option
3467        This pure operator allocates some uninitialized storage from the heap.
3468        There are two forms:
3469    
3470         rawrecord NONE [INT n]  allocates a tagless record of length n
3471         rawrecord (SOME rk) [INT n] allocates a tagged record of length n
3472                                     and initializes the tag.
3473    
3474     2. rawupdate of cty
3475          rawupdate cty (v,i,x)
3476          Assigns to x to the ith component of record v.
3477          The storelist is not updated.
3478    
3479    We use these new primops for both spilling and increment record construction.
3480    
3481     1. Spilling.
3482    
3483        This is implemented with a linear scan algorithm (but generalized
3484        to trees).  The algorithm will create a single spill record at the
3485        beginning of the cps function and use rawupdate to spill to it,
3486        and SELECT or SELp to reload from it.  So both spills and reloads
3487        are fine-grain operations.  In contrast, in the old algorithm
3488        "spills" have to be bundled together in records.
3489    
3490        Ideally, we should sink the spill record construction to where
3491        it is needed.  We can even split the spill record into multiple ones
3492        at the places where they are needed.  But CPS is not a good
3493        representation for global code motion, so I'll keep it simple and
3494        am not attempting this.
3495    
3496     2. Incremental record construction (aka record splitting).
3497    
3498        Long records with many component values which are simulatenously live
3499        (recall that single use record addresses are not considered to
3500         be live) are constructed with rawrecord and rawupdate.
3501        We allocate space on the heap with rawrecord first, then gradually
3502        fill it in with rawupdate.  This is the technique suggested to me
3503        by Matthias.
3504    
3505        Some restrictions on when this is applicable:
3506        1. It is not a VECTOR record.  The code generator currently does not handle
3507           this case. VECTOR record uses double indirection like arrays.
3508        2. All the record component values are defined in the same "basic block"
3509           as the record constructor.  This is to prevent speculative
3510           record construction.
3511    
3512    ----------------------------------------------------------------------
3513    Name: Allen Leung
3514    Date: 2002/02/22 01:02:00 EST
3515    Tag: leunga-20020222-mlrisc-tools
3516    
3517    Minor bug fixes in the parser and rewriter
3518    
3519    ----------------------------------------------------------------------
3520    Name: Allen Leung
3521    Date: 2002/02/21 20:20:00 EST
3522    Tag: leunga-20020221-peephole
3523    
3524    Regenerated the peephole files.  Some contained typos in the specification
3525    and some didn't compile because of pretty printing bugs in the old version
3526    of 'nowhere'.
3527    
3528    ----------------------------------------------------------------------
3529    Name: Allen Leung
3530    Date: 2002/02/19 20:20:00 EST
3531    Tag: leunga-20020219-mlrisc-tools
3532    Description:
3533    
3534       Minor bug fixes to the mlrisc-tools library:
3535    
3536       1.  Fixed up parsing colon suffixed keywords
3537       2.  Added the ability to shut the error messages up
3538       3.  Reimplemented the pretty printer and fixed up/improved
3539           the pretty printing of handle and -> types.
3540       4.  Fixed up generation of literal symbols in the nowhere tool.
3541       5.  Added some SML keywords to to sml.sty
3542    
3543    ----------------------------------------------------------------------
3544    Name: Matthias Blume
3545    Date: 2002/02/19 16:20:00 EST
3546    Tag: blume-20020219-cmffi
3547    Description:
3548    
3549    A wild mix of changes, some minor, some major:
3550    
3551    * All C FFI-related libraries are now anchored under $c:
3552        $/c.cm      --> $c/c.cm
3553        $/c-int.cm  --> $c/internals/c-int.cm
3554        $/memory.cm --> $c/memory/memory.cm
3555    
3556    * "make" tool (in CM) now treats its argument pathname slightly
3557      differently:
3558        1. If the native expansion is an absolute name, then before invoking
3559           the "make" command on it, CM will apply OS.Path.mkRelative
3560           (with relativeTo = OS.FileSys.getDir()) to it.
3561        2. The argument will be passed through to subsequent phases of CM
3562           processing without "going native".  In particular, if the argument
3563           was an anchored path, then "make" will not lose track of that anchor.
3564    
3565    * Compiler backends now "know" their respective C calling conventions
3566      instead of having to be told about it by ml-nlffigen.  This relieves
3567      ml-nlffigen from one of its burdens.
3568    
3569    * The X86Backend has been split into X86CCallBackend and X86StdCallBackend.
3570    
3571    * Export C_DEBUG and C_Debug from $c/c.cm.
3572    
3573    * C type encoding in ml-nlffi-lib has been improved to model the conceptual
3574      subtyping relationship between incomplete pointers and their complete
3575      counterparts.  For this, ('t, 'c) ptr has been changed to 'o ptr --
3576      with the convention of instantiating 'o with ('t, 'c) obj whenever
3577      the pointer target type is complete.  In the incomplete case, 'o
3578      will be instantiated with some "'c iobj" -- a type obtained by
3579      using one of the functors PointerToIncompleteType or PointerToCompleteType.
3580    
3581      Operations that work on both incomplete and complete pointer types are
3582      typed as taking an 'o ptr while operations that require the target to
3583      be known are typed as taking some ('t, 'c) obj ptr.
3584    
3585      voidptr is now a bit "more concrete", namely "type voidptr = void ptr'"
3586      where void is an eqtype without any values.  This makes it possible
3587      to work on voidptr values using functions meant to operate on light
3588      incomplete pointers.
3589    
3590    * As a result of the above, signature POINTER_TO_INCOMPLETE_TYPE has
3591      been vastly simplified.
3592    
3593    ----------------------------------------------------------------------
3594    Name: Matthias Blume
3595    Date: 2002/02/19 10:48:00 EST
3596    Tag: blume-20020219-pqfix
3597    Description:
3598    
3599    Applied Chris Okasaki's bug fix for priority queues.
3600    
3601    ----------------------------------------------------------------------
3602    Name: Matthias Blume
3603    Date: 2002/02/15 17:05:00
3604    Tag: Release_110_39
3605    Description:
3606    
3607    Last-minute retagging is becoming a tradition... :-(
3608    
3609    This is the working release 110.39.
3610    
3611    ----------------------------------------------------------------------
3612    Name: Matthias Blume
3613    Date: 2002/02/15 16:00:00 EST
3614    Tag: Release_110_39-orig
3615    Description:
3616    
3617    Working release 110.39.  New bootfiles.
3618    
3619    (Update: There was a small bug in the installer so it wouldn't work
3620    with all shells.  So I retagged. -Matthias)
3621    
3622    ----------------------------------------------------------------------
3623    Name: Matthias Blume
3624    Date: 2002/02/15 14:17:00 EST
3625    Tag: blume-20020215-showbindings
3626    Description:
3627    
3628    Added EnvRef.listBoundSymbols and CM.State.showBindings.  Especially
3629    the latter can be useful for exploring what bindings are available at
3630    the interactive prompt.  (The first function returns only the list
3631    of symbols that are really bound, the second prints those but also the
3632    ones that CM's autoloading mechanism knows about.)
3633    
3634    ----------------------------------------------------------------------
3635    Name: Matthias Blume
3636    Date: 2002/02/15 12:08:00 EST
3637    Tag: blume-20020215-iptrs
3638    Description:
3639    
3640    Two improvements to ml-nlffigen:
3641    
3642      1. Write files only if they do not exist or if their current contents
3643         do not coincide with what's being written.  (That is, avoid messing
3644         with the time stamps unless absolutely necessary.)
3645    
3646      2. Implement a "repository" mechanism for generated files related
3647         to "incomplete pointer types".   See the README file for details.
3648    
3649    ----------------------------------------------------------------------
3650    Name: Matthias Blume
3651    Date: 2002/02/14 11:50:00 EST
3652    Tag: blume-20020214-quote
3653    Description:
3654    
3655    Added a type 't t_' to tag.sml (in ml-nlffi-lib.cm).  This is required
3656    because of the new and improved tag generation scheme.  (Thanks to Allen
3657    Leung for pointing it out.)
3658    
3659    ----------------------------------------------------------------------
3660    Name: Lal George
3661    Date: 2002/02/14 09:55:27 EST 2002
3662    Tag: george-20020214-isabelle-bug
3663    Description:
3664    
3665    Fixed the MLRISC bug sent by Markus Wenzel regarding the compilation
3666    of Isabelle on the x86.
3667    
3668    From Allen:
3669    -----------
3670     I've found the problem:
3671    
3672         in ra-core.sml, I use the counter "blocked" to keep track of the
3673         true number of elements in the freeze queue.  When the counter goes
3674         to zero, I skip examining the queue.  But I've messed up the
3675         bookkeeping in combine():
3676    
3677             else ();
3678             case !ucol of
3679               PSEUDO => (if !cntv > 0 then
3680                     (if !cntu > 0 then blocked := !blocked - 1 else ();
3681                                        ^^^^^^^^^^^^^^^^^^^^^^^
3682                      moveu := mergeMoveList(!movev, !moveu)
3683                     )
3684                  else ();
3685    
3686         combine() is called to coalesce two nodes u and v.
3687         I think I was thinking that if the move counts of u and v are both
3688         greater than zero then after they are coalesced then one node is
3689         removed from the freeze queue.  Apparently I was thinking that
3690         both u and v are of low degree, but that's clearly not necessarily true.
3691    
3692    
3693    02/12/2002:
3694        Here's the patch.  HOL now compiles.
3695    
3696        I don't know how this impact on performance (compile
3697        time or runtime).  This bug caused the RA (especially on the x86)
3698        to go thru the potential spill phase when there are still nodes on the
3699        freeze queue.
3700    
3701    
3702    
3703    
3704    ----------------------------------------------------------------------
3705    Name: Matthias Blume
3706    Date: 2002/02/13 22:40:00 EST
3707    Tag: blume-20020213-fptr-rtti
3708    Description:
3709    
3710    Fixed a bug in ml-nlffigen that was introduced with one of the previous
3711    updates.
3712    
3713    ----------------------------------------------------------------------
3714    Name: Matthias Blume
3715    Date: 2002/02/13 16:41:00 EST
3716    Tag: blume-20020213-cmlpq
3717    Description:
3718    
3719    Added new priority queue export symbols (which have just been added to
3720    smlnj-lib.cm) to CML's version of smlnj-lib.cm.  (Otherwise CML would
3721    not compile and the installer would choke.)
3722    
3723    ----------------------------------------------------------------------
3724    Name: Matthias Blume
3725    Date: 2002/02/13 16:15:00 EST
3726    Tag: blume-20020213-various
3727    Description:
3728    
3729    1. More tweaks to ml-nlffigen:
3730    
3731       - better internal datastructures (resulting in slight speedup)
3732       - "-match" option requires exact match
3733       - "localized" gensym counters (untagged structs/unions nested within
3734         other structs/unions or within typedefs get a fresh counter; their
3735         tag will be prefixed by a concatenation of their parents' tags)
3736       - bug fixes (related to calculation of transitive closure of types
3737         to be included in the output)
3738    
3739    2. Minor Basis updates:
3740    
3741       - added implementations for List.collate and Option.app
3742    
3743    ----------------------------------------------------------------------
3744    Name: Matthias Blume
3745    Date: 2002/02/11 15:55:00 EST
3746    Tag: blume-20020211-gensym
3747    Description:
3748    
3749    Added a "-gensym" option to command line of ml-nlffigen.  This can be
3750    used to specify a "stem" -- a string that is inserted in all "gensym'd"
3751    names (ML structure names that correspond to unnamed C structs, unions,
3752    and enums), so that separate runs of ml-nlffigen do not clash.
3753    
3754    ----------------------------------------------------------------------
3755    Name: Matthias Blume
3756    Date: 2002/02/11 12:05:00 EST
3757    Tag: blume-20020211-gensml
3758    Description:
3759    
3760    A quick fix for a problem with GenSML (in the pgraph-util library):
3761    Make generation of toplevel "local" optional.  (Strictly speaking,
3762    signature definitions within "local" are not legal SML.)
3763    
3764    Other than that: updates to INSTALL and cm/TODO.
3765    
3766    ----------------------------------------------------------------------
3767    Name: Matthias Blume
3768    Date: 2002/02/08 15:00:00 EST
3769    Tag: blume-20020208-uniquepid
3770    Description:
3771    
3772    0. Version number has been bumped to 110.38.1.  NEW BOOTFILES!!!
3773    
3774    1. The installer (config/install.sh) has gotten smarter:
3775    
3776         - Configuration options are a bit easier to specify now
3777           (in config/targets).
3778         - Bug in recognizing .tar.bz2 files fixed.
3779         - Installer automatically resolves dependencies between
3780           configuration options (e.g., if you ask for eXene, you will
3781           also get cml -- regardless whether you asked for it or not).
3782         - Installer can run in "quieter mode" by setting the environment
3783           variable INSTALL_QUIETLY to "true".  "Quieter" does not mean
3784           "completely silent", though.
3785         - Build HashCons library as part of smlnj-lib.
3786    
3787    2. A new scheme for assigning persistent identifiers to compilation
3788       units (and, by extension, to types etc.) has been put into place.
3789       This fixes a long-standing bug where types and even dynamic values
3790       can get internally confused, thereby compromising type safety
3791       (abstraction) and dynamic correctness.  See
3792    
3793         http://cm.bell-labs.com/cm/cs/who/blume/pid-confusion.tgz
3794    
3795       for an example of how things could go wrong until now.
3796    
3797       The downside of the new scheme is that pids are not quite as
3798       persistent as they used to be: CM will generate a fresh pid
3799       for every compilation unit that it thinks it sees for the first
3800       time.  That means that if you compile starting from a clean, fresh
3801       source tree at two different times, you end up with different
3802       binaries.
3803    
3804       Cutoff recompilation, however, has not been compromised because
3805       CM keeps pid information in special caches between runs.
3806    
3807    ----------------------------------------------------------------------
3808    Name: Lal George
3809    Date: 2002/02/07 15:34:13 EST 2002
3810    Tag: <none>
3811    Description:
3812    
3813    Compilers that generate assembly code may produce  global labels
3814    whose value is resolved at link time. The various peephole optimization
3815    modules did not take this in account.
3816    
3817    TODO. The Labels.addrOf function should really return an option
3818    type so that clients are forced to deal with this issue, rather
3819    than an exception being raised.
3820    
3821    ----------------------------------------------------------------------
3822    Name: Lal George
3823    Date: 2002/02/06 13:55:02 EST
3824    Tag: george-20020206-ra-breakup
3825    Description:
3826    
3827    1. A bug fix from Allen.
3828    
3829        A typo causes extra fstp %st(0)'s to be generated at compensation
3830        edges, which might cause stack underflow traps at runtime.  This
3831        occurs in fft where there are extraneous fstps right before the 'into'
3832        trap instruction (in this case they are harmless since none of the
3833        integers overflow.)
3834    
3835    2. Pulled out various utility modules that were embedded in the modules
3836       of the register allocator. I need these modules for other purposes, but
3837       they are not complete enough to put into a library (just yet).
3838    ----------------------------------------------------------------------
3839    Name: Matthias Blume
3840    Date: 2002/01/31 16:05:00 EST
3841    Tag: blume-20020131-sparc-ccalls
3842    Description:
3843    
3844    1. C-calls on Sparc needlessly allocated a huge chunk (96 bytes)
3845       of extra stack space by mistake.  Fixed.
3846    
3847    2. Bug in logic of handling of command-line options in ml-nlffigen fixed.
3848    
3849    ----------------------------------------------------------------------
3850    Name: Allen Leung
3851    Date: 2002/01/30
3852    Tag: leunga-20020130-nowhere-bug-fix
3853    Description:
3854    
3855       MLRISC bug fixes:
3856       1. Fixed a bindings computation bug in the 'nowhere' program generator tool.
3857       2. MachineInt.fromString was neg