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/NOTES/HISTORY
ViewVC logotype

Diff of /sml/trunk/NOTES/HISTORY

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

revision 773, Mon Jan 8 16:18:37 2001 UTC revision 1926, Mon May 15 20:15:32 2006 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: Matthias Blume (http://www.tti-c.org/blume/)
17    Date: 2006/05/12 15:20:00 CST
18    Tag: blume-20060515-makedepend
19    Description:
20    
21    Implemented ml-makedepend (i.e., CM.sources) in a better (more robust)
22    way.  This should hopefully fix the ml-makedepend problem permanently.
23    
24    ----------------------------------------------------------------------
25    Name: Matthias Blume (http://www.tti-c.org/blume/)
26    Date: 2006/05/12 15:15:00 CST
27    Tag: blume-20060512-makedepend
28    Description:
29    
30    Fixed long-standing bug with ml-makedepend where it would output a
31    spurious dependency to a non-existing file.  (This is a simple fix.
32    It might need further looking into.)
33    
34    ----------------------------------------------------------------------
35    Name: Matthias Blume (http://www.tti-c.org/blume/)
36    Date: 2006/04/20 10:30:00 CDT
37    Tag: blume-20060420-heap2exec
38    Description:
39    
40    Committed patches received from Johannes 5 Joemann (joemann@befree.free.de)
41    that enable heap2exec under Linux and FreeBSD.
42    
43    ----------------------------------------------------------------------
44    Name: John Reppy
45    Date: 2006/04/14
46    Tag:
47    Description:
48    
49    MLRISC changes:
50      1) renamed GAS_PSEUDO_OPS to AS_PSEUDO_OPS and put it in its own file.
51      2) added support for NOTB and XORB operators in pseudo-op expressions
52      3) added DarwinPseudoOp functor that supports Darwin's assembler syntax.
53      4) added support for 64-bit integer literals
54    
55    ----------------------------------------------------------------------
56    Name: Matthias Blume (http://www.tti-c.org/blume/)
57    Date: yyyy/mm/dd
58    Tag: Release_110_58
59    Description:
60    
61    New working version (110.58).  NEW BOOTFILES!
62    
63    ----------------------------------------------------------------------
64    Name: Matthias Blume (http://www.tti-c.org/blume/)
65    Date: 2006/03/01 22:15:00 CST
66    Tag: blume-20060301-lexgen
67    Description:
68    
69    Incorporated several bugfixes to lexgen.  Compiler now compiles to
70    fixpoint when using lexgen instead of ml-lex.
71    
72    ----------------------------------------------------------------------
73    Name: Matthias Blume (http://www.tti-c.org/blume/)
74    Date: 2006/02/28 22:45:00 CST
75    Tag: blume-20060228-lexgen
76    Description:
77    
78    Removed ml-flex and added lexgen instead, using Aaron Turon's newly
79    provided tarball.  The generated lexers still have problems.
80    
81    ----------------------------------------------------------------------
82    Name: Matthias Blume (http://www.tti-c.org/blume/)
83    Date: 2006/02/26 23:30:00 CST
84    Tag: blume-20060226-ml-flex
85    Description:
86    
87    Removed ml-flex' dependency on regexp-lib.cm.  Turned installation
88    of ml-flex on by default.
89    
90    ----------------------------------------------------------------------
91    Name: Matthias Blume (http://www.tti-c.org/blume/)
92    Date: 2006/02/24 16:30:00 CST
93    Tag: blume-20060224-ml-flex
94    Description:
95    
96    Added ml-flex sources.  Partially integrated, but should not be
97    turned on yet!  (Read: leave it commented-out in config/targets!)
98    
99    ----------------------------------------------------------------------
100    Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
101    Date: 2006/02/23
102    Tag: jhr-20060223-x86-64
103    Description:
104    Changes to support building on x86-64 systems (using the 32-bit mode).
105    Also cleaned up signal handling on Linux.  Support for pre-2.2 Linux
106    kernels dropped.
107    
108    ----------------------------------------------------------------------
109    Name: Allen Leung (leunga (at) reservoir (dot) com)
110    Date: 2006/02/22 15:07:00 EST
111    Tag: leunga-20060222-mdl-gen
112    Description:
113    
114      CM has changed.  Updated the script for rebuilding the MLRISC generated
115    files.
116    
117    ----------------------------------------------------------------------
118    Name: Matthias Blume (http://www.tti-c.org/blume/)
119    Date: 2006/02/14 17:30:00 CST
120    Tag: blume-20060214-intel-mac
121    Description:
122    
123    Hooked code for Darwin-specific Intel ABI into main compiler.
124    (This is a temporary solution which relies on the fact that the
125    compiler itself does not use NLFFI.  Eventually we need to divorce
126    intel mac from generic x86 unix code and make separate sets of binaries.)
127    
128    ----------------------------------------------------------------------
129    Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
130    Date: 2006/02/13
131    Tag: jhr-20060213-intel-mac
132    Description:
133    Changed MLRISC x86 CCalls for partial support of MacOS X ABI.
134    
135    ----------------------------------------------------------------------
136    Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
137    Date: 2006/02/06
138    Tag: jhr-20060206-intel-mac
139    Description:
140    Changes to support MacOS X on Intel hardware.  The C-calls support in
141    MLRISC must also be updated to support the MacOS X ABI.
142    
143    ----------------------------------------------------------------------
144    Name: Matthias Blume (http://www.tti-c.org/blume/)
145    Date: 2005/12/16 00:15:00 CST
146    Tag: blume-20051216-cm
147    Description:
148    
149    Improved error reporting and handling in CM.
150    
151    ----------------------------------------------------------------------
152    Name: Matthias Blume (http://www.tti-c.org/blume/)
153    Date: 2005/11/21 14:00:00 CST
154    Tag: blume-20051121-html-lib
155    Description:
156    
157    Enabling $/html-lib.cm in config/preloads did not work.  This is fixed
158    now.  (Since the anchor mapping for html-lib.cm is not yet in effect
159    at the time when config/preloads is processed, the library has to
160    be referred to by another name.  In 110.57 this would be
161    $SMLNJ-LIB/HTML/html-lib.cm.  I arranged for $smlnj/smlnj-lib/html-lib.cm
162    to be valid as well.)
163    
164    Thanks to Todd Wilson (Fresno) for alerting me to this issue.
165    
166    ----------------------------------------------------------------------
167    Name: Matthias Blume (http://www.tti-c.org/blume/)
168    Date: 2005/11/19 13:45:00 CST
169    Tag: Release_110_57
170    Description:
171    
172    Moved release tag for 110.57.
173    
174    ----------------------------------------------------------------------
175    Name: Matthias Blume (http://www.tti-c.org/blume/)
176    Date: 2005/11/19 23:35:00 CST
177    Tag: blume-20051119-fish
178    Description:
179    
180    Fixed a problem in config/install.sh where it tries to "fish" the
181    name of the CM metadata directory from the wrong place (because
182    the physical location of basis.cm has changed).  Also, corrected the
183    path anchor for $/html-lib.cm.  (Thanks to M. Fluet for pointing out
184    these problems.)
185    
186    ----------------------------------------------------------------------
187    Name: Matthias Blume (http://www.tti-c.org/blume/)
188    Date: 2005/11/16 14:30:00 CST
189    Tag: blume-orig-110_57
190    Description:
191    
192    New working version (110.57).  NEW BOOTFILES!
193    
194    UPDATE: The release tag has been moved to include a fix for a brown-paper-
195            bag glitch (see above).
196    
197    ----------------------------------------------------------------------
198    Name: Matthias Blume (http://www.tti-c.org/blume/)
199    Date: 2005/11/16 10:25:00 CST
200    Tag: blume-22051116-btrace
201    Description:
202    
203    Fixed problem with bogus exception message when using back-trace
204    facility.
205    
206    ----------------------------------------------------------------------
207    Name: Matthias Blume (http://www.tti-c.org/blume/)
208    Date: 2005/11/15 22:35:00 CST
209    Tag: blume-22051115-array2copy
210    Description:
211    
212    Added simple implementation of Array2.copy.  (Warning: mostly untested.)
213    
214    ----------------------------------------------------------------------
215    Name: Matthias Blume (http://www.tti-c.org/blume/)
216    Date: 2005/11/15 21:35:00 CST
217    Tag: blume-22051115-versiontool
218    Description:
219    
220    Reversed change to src/system/smlnj/internals/versiontool.cm.  This
221    file gets loaded as a tool -- by the equivalent of CM.make during the
222    run of CMB.make.  Thus, CMB's path configuration is meaningless for
223    it.  Instead, it has the status of "user code", so it should use
224    $/basis.cm to refer to the Basis library.  (At least that's true for
225    the purpose of bootstrapping the previous change.  In the future it
226    might make sense to have versiontool.cm refer to
227    $smlnj/basis/basis.cm, i.e., the version of the Basis that the
228    compiler itself uses.)
229    
230    Also patched src/system/testml to have it activate those extra anchor
231    bindings in config/extrapathconfig.
232    
233    ----------------------------------------------------------------------
234    Name: Matthias Blume (http://www.tti-c.org/blume/)
235    Date: 2005/11/15 16:35:00 CST
236    Tag: blume-22051115-libraries
237    Description:
238    
239    This change affects the way the following libraries are tied into
240    the system:
241    
242       $/basis.cm
243       $/smlnj-lib.cm
244       $/pp-lib.cm
245       $/controls-lib.cm
246       $/html-lib.cm
247       $/ml-yacc-lib.cm
248    
249    These libraries are now internally (as seen from the source code
250    of the implementation itself) known by the following names:
251    
252       $smlnj/basis/basis.cm
253       $smlnj/smlnj-lib/smlnj-lib.cm
254       $smlnj/smlnj-lib/pp-lib.cm
255       $smlnj/smlnj-lib/controls-lib.cm
256       $smlnj/smlnj-lib/html-lib.cm
257       $smlnj/ml-yacc/ml-yacc-lib.cm
258    
259    This makes it possible to work with code that requires different
260    versions of these libraries, and which refers to these libraries using
261    their "default" names (i.e., the first set of names above).  In other
262    words, one can un-define or re-define those default names without
263    compromising the proper functioning of the compiler itself.
264    
265    A similar procedure had already been performed for several of the
266    MLRISC libraries that are linked into the compiler.  I did some
267    cleanup on this code.
268    
269    A new file in the config directory (named extrapathconfig) is
270    responsible for setting up path anchors that the compiler itself does
271    not need, but that are typically required by user code.
272    
273    ----------------------------------------------------------------------
274    Name: Matthias Blume
275    Date: 2005/11/07 17:05:00 CST
276    Tag: blume-20051107-slice
277    Description:
278    
279    Fixed erroneous out-of-bounds test in the "update" function of
280    various *ArraySlice modules.  (Thanks to Vesa A. Norrman for pointing
281    out the problem.)
282    
283    Pushed some Basis changes through ML-Lex, CML, and eXene.
284    
285    ----------------------------------------------------------------------
286    Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
287    Date: 2005/11/07
288    Description:
289    Fixed a Basis incompatibility: The depreciated function Substring.all
290    was removed (use Substring.full instead).
291    
292    ----------------------------------------------------------------------
293    Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
294    Date: 2005/11/05
295    Description:
296    Tweaked interval set API in SML/NJ library; see the CHANGES file for details.
297    
298    ----------------------------------------------------------------------
299    Name: Matthias Blume (http://www.tti-c.org/blume/)
300    Date: 2005/11/02 16:25:00 CST
301    Tag: blume-20051102-hex
302    Description:
303    
304    * Runtime system bootstrap code now accepts hex digits in BOOTLIST
305      in either upper case or lower case format.
306    * Pushed changes to names of Pack<N>{Big,Little} structures through
307      CML and eXene.
308    
309    ----------------------------------------------------------------------
310    Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
311    Date: 2005/11/02
312    Description:
313    Fixed a Basis incompatibility: Pack<N>{Big,Little} structures should
314    be called PackWord<N>{Big,Little}.
315    
316    ----------------------------------------------------------------------
317    Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
318    Date: 2005/10/28
319    Description:
320    Fixed a minor Basis incompatibility: hex digits should be upper case.
321    
322    ----------------------------------------------------------------------
323    Name: Matthias Blume (blume (at) tti - c (dot) org)
324    Date: 2005/10/25 16:00:00 CDT
325    Tag: Release_110_56
326    Description:
327    
328    New working version (110.56).  NEW BOOTFILES!
329    
330    ----------------------------------------------------------------------
331    Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
332    Date: 2005/10/25
333    Description:
334    Added interval sets to utility library (signatures INTERVAL_DOMAIN
335    and INTERVAL_SET, and functor IntervalSetFn).
336    
337    ----------------------------------------------------------------------
338    Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
339    Date: 2005/10/14
340    Description:
341    Add Zhong Shao's fix for datatype equality functions.
342    
343    ----------------------------------------------------------------------
344    Name: Allen Leung (leunga (at) reservoir (dot) com)
345    Date: 2005/10/14 12:30:00 EST
346    Tag: leunga-20051014-x86-spill
347    Description:
348    
349    Bug fix a bug found by Carl Hauser.
350    There was a typo in the reload code for FCMP in x86SpillInstr.sml.
351    
352    ----------------------------------------------------------------------
353    Name: Allen Leung (leunga (at) reservoir (dot) com)
354    Date: 2005/10/14 11:15:00 EST
355    Tag: leunga-20051014-x86-asm
356    Description:
357    
358    Removed some debugging code in file x86Asm.sml.
359    The function emit_operand was printing out debugging output.
360    
361    ----------------------------------------------------------------------
362    Name: John Reppy
363    Date: 2005/07/27
364    Tag:
365    Description:
366    
367    Fixed ml-lex to recognize "\r" as representing carriage return.
368    
369    ----------------------------------------------------------------------
370    Name: John Reppy
371    Date: 2005/07/27
372    Tag:
373    Description:
374    
375    Fixed ml-yacc to work on files with non-native end-of-line
376    encodings (e.g., Windows text file on a Unix system).
377    
378    ----------------------------------------------------------------------
379    Name: John Reppy
380    Date: 2005/07/20
381    Tag:
382    Description:
383    
384    Added changes from Dominic Evans (oldmanuk (at) gmail (dot) com)
385    to support HPUX 11.
386    
387    ----------------------------------------------------------------------
388    Name: John Reppy
389    Date: 2005/07/06
390    Tag:
391    Description:
392    
393    Changes to the SML/NJ library.  See smlnj-lib/CHANGES for details.
394    
395    ----------------------------------------------------------------------
396    Name: Matthias Blume (blume (at) tti - c (dot) org)
397    Date: 2005/07/06 00:45:00 CDT
398    Tag: blume-20050706-slice-copy
399    Description:
400    
401    Fixed reversed logic for deciding whether to "copy up" or "copy down"
402    in *-array-slice.sml.
403    
404    ----------------------------------------------------------------------
405    Name: Allen Leung (leunga (at) reservoir (dot) com)
406    Date: 2005/05/31 17:00:00 EST
407    Tag: leunga-20050531-cygwin-fault-2
408    Description:
409    
410    A typo in the cygwin code fixed.
411    
412    ----------------------------------------------------------------------
413    Name: Allen Leung (leunga (at) reservoir (dot) com)
414    Date: 2005/05/31 16:47:00 EST
415    Tag: leunga-20050531-cygwin-fault
416    Description:
417    
418    Updated Cygwin's fault/signal handling to match the Windows version.
419    Updated the export list.
420    
421    ----------------------------------------------------------------------
422    Name: Matthias Blume (blume (at) tti - c (dot) org)
423    Date: 2005/05/18 13:35:00 CDT
424    Tag: Release_110_54
425    Description:
426    
427    New working version (110.54).  NEW BOOTFILES!
428    
429    ----------------------------------------------------------------------
430    Name: Matthias Blume (blume (at) tti - c (dot) org)
431    Date: 2005/05/18 11:58:00 CDT
432    Tag: blume-20050518-installer
433    Description:
434    
435    Added support scripts for Mac OS X PackageMaker and modified
436    config/install.sh so that it supports re-dumping a heap image
437    after customization.
438    
439    ----------------------------------------------------------------------
440    Name: Matthias Blume (blume (at) tti - c (dot) org)
441    Date: 2005/05/18 10:55:00 CDT
442    Tag: blume-20050518-realdiv-noovld
443    Description:
444    
445    Un-overloaded / to work around bug in overloading resolution code.
446    
447    ----------------------------------------------------------------------
448    Name: Matthias Blume (blume (at) tti - c (dot) org)
449    Date: 2005/05/16 23:50:00 CDT
450    Tag: blume-20050516-redump-heap
451    Description:
452    
453    Added mechanism for re-creating a heap file for the interactive system
454    after configuration variables have been changed.
455    
456       CM.redump_heap : string -> unit
457    
458    This is much like SMLofNJ.exportML, but starting from the resulting
459    heap does not return to the caller of CM.redump_heap but
460    restarts the interactive system from scratch.  The original call of
461    CM.redump_heap does not return but ends the interactive session.
462    Thus, CM.redump_heap is a lot like SMLofNJ.exportFn.
463    
464    Internally, redump_heap winds the dynamic execution context back to
465    the point where the original heap image was created and re-executes
466    the heap image generation code in the boot code.
467    
468    ----------------------------------------------------------------------
469    Name: Matthias Blume (blume (at) tti - c (dot) org)
470    Date: 2005/05/09 21:55:00 CDT
471    Tag: blume-20050509-word64
472    Description:
473    
474    Added a hack to the existing hack known as Word64 to make fromString
475    behave correctly.  I am still not sure whether Word64.scan will work
476    as specified with respect to the interaction of radix and prefix.
477    
478    ----------------------------------------------------------------------
479    Name: Allen Leung (leunga (at) reservoir (dot) com)
480    Date: 2005/05/04 11:50:00 EST
481    Tag: leunga-20050504-checkgc
482    Description:
483    
484    Added a gc protocol checking phase.  This phase is enabled with
485    the flag "check-gc".  "debug-check-gc" turns on the verbose mode.
486    
487    ----------------------------------------------------------------------
488    Name: Matthias Blume (blume (at) tti - c (dot) org)
489    Date: 2005/05/04 10:45:00 CDT
490    Tag: blume-20050504-intinf
491    Description:
492    
493    Fixed a bug in the implementation of div and mod for IntInf.
494    Thanks to Neophytos Michael for reporting the problem.
495    
496    ----------------------------------------------------------------------
497    Name: Matthias Blume (blume (at) tti - c (dot) org)
498    Date: 2005/05/04 10:35:00 CDT
499    Tag: blume-20050504-join
500    Description:
501    
502    Added a "join" combinator to the ParserComb module in smlnj-lib.cm.
503    
504    ----------------------------------------------------------------------
505    Name: Matthias Blume (blume (at) tti - c (dot) org)
506    Date: 2005/02/28 23:40:00 CST
507    Tag: blume-20050228-mVar
508    Description:
509    
510    Fixed serious bug (brown paper bag variety) in new implementation of
511    structure Atom in CML.  (I had accidentally used a mailbox instead of
512    an mvar, leaving the door open for races.)
513    
514    ----------------------------------------------------------------------
515    Name: Matthias Blume (blume (at) tti - c (dot) org)
516    Date: 2005/02/25 15:00:00 CST
517    Tag: Release_110_53
518    Description:
519    
520    New working version (110.53).  NEW BOOTFILES!
521    
522    ----------------------------------------------------------------------
523    Name: Matthias Blume (blume (at) tti - c (dot) org)
524    Date: 2005/02/25 14:50:00 CST
525    Tag: blume-20050225-susp
526    Description:
527    
528    Brought back SMLofNJ.Susp.  The underlying suspension type is the one
529    implemented in Core -- which means that it is the same as the one used
530    by the lazy extension.
531    
532    ----------------------------------------------------------------------
533    Name: Matthias Blume (blume (at) tti - c (dot) org)
534    Date: 2005/02/24 16:50:00 CST
535    Tag: blume-20050224-cml-atom
536    Description:
537    
538    Simpler and at the same time more general implementation of structure
539    Atom in CML.
540    
541    ----------------------------------------------------------------------
542    Name: Matthias Blume (blume (at) tti - c (dot) org)
543    Date: 2005/02/15 17:35:00 CST
544    Tag: blume-20050215-tools
545    Description:
546    
547    Created new "tools" directory under "src" and moved "TraceDebugProf"
548    there.
549    
550    ----------------------------------------------------------------------
551    Name: Matthias Blume (blume (at) tti - c (dot) org)
552    Date: 2005/02/10 17:55:00 CST
553    Tag: blume-20050210-longlong
554    Description:
555    
556    Implemented "long long" arguments and results for NLFFI.  (Only the
557    PPC/MacOS implementation is complete, the other backends still need to
558    be updated.)
559    
560    ----------------------------------------------------------------------
561    Name: Matthias Blume (blume (at) tti - c (dot) org)
562    Date: 2005/01/24 17:40:00 CST
563    Tag: blume-20050124-mlyacc
564    Description:
565    
566    Minor cleanup in ML-Yacc rule printing mechanism.  This should fix a
567    problem with certain "as" patterns which previously got rendered
568    using incorrect syntax.
569    
570    ----------------------------------------------------------------------
571    Name: Matthias Blume (blume (at) tti - c (dot) org)
572    Date: 2005/01/18 12:00:00 CST
573    Tag: blume-20050118-profile
574    Description:
575    
576    Made time profiling code (interrupt handler) in runtime system aware
577    of new array representation.
578    
579    ----------------------------------------------------------------------
580    Name: Matthias Blume (blume (at) tti - c (dot) org)
581    Date: 2005/01/14 18:00:00 CST
582    Tag: blume-20050114-heap2exec
583    Description:
584    
585    Implemented new (but still experimental) heap2exec facility.  This is
586    tested under Mac OS X and should work under Linux (will test shortly).
587    It will probably also work on the Sparc (will test some time later).
588    
589      - removed old "HACKED_STANDALONE" hack from runtime
590    
591    To be able to test this, uncomment the request for "heap2asm" in
592    config/targets prior to installation.  (Notice that this is different
593    from "heap2exec" mentioned below.  Not a typo.)
594    
595    To perform an actual test, run the command
596    
597       $ bin/heap2exec heapfile execfile
598    
599    (You can put heap2exec on your shell's path.)
600    
601    For example, run
602    
603      $ bin/heap2exec bin/.heap/ml-yacc.ppc-darwin mly
604    
605    This will create a standalone executable called "mly" which you can
606    then invoke directly as a command.
607    
608    ----------------------------------------------------------------------
609    Name: Matthias Blume (blume (at) tti - c (dot) org)
610    Date: 2005/01/07 11:44:00 CST
611    Tag: blume-20050107-mlstring
612    Description:
613    
614    fixed off-by-one error in ML_STRING macro (globals.c)
615    
616    ----------------------------------------------------------------------
617    Name: Matthias Blume (blume (at) tti - c (dot) org)
618    Date: 2004/12/23 18:00:00 CST
619    Tag: blume-20041223-santa
620    Description:
621    
622    Made ml-build script "smarter" (but only very little).
623    
624    ----------------------------------------------------------------------
625    Name: Matthias Blume (blume (at) tti - c (dot) org)
626    Date: 2004/12/21 15:05:00 CST
627    Tag: blume-20041221-longlong
628    Description:
629    
630    * Implemented access to signed and unsigned long long data in NLFFI.
631    (The parameter-passing part of the picture has not complete. But
632    data structure access seems to work.)
633    
634    * Fixed CM's incorrect assumption that the PPC is little-endian.
635      (On the Mac, it is big-endian.  And that's currently our only
636      PPC platform.)
637    
638    ----------------------------------------------------------------------
639    Name: Matthias Blume (blume (at) tti - c (dot) org)
640    Date: 2004/12/21 12:50:00 CST
641    Tag: blume-20041221-memory
642    Description:
643    
644    Some cleanup in the $c/memory.cm library: separated some concerns by
645    moving allocation code and memory access code each into their own
646    files.
647    
648    ----------------------------------------------------------------------
649    Name: Allen Leung (leunga (at) reservoir (dot) com)
650    Date: 2004/12/17 16:12:00 EST
651    Tag: leunga-20041217-cygwin-smlnj-home
652    Description:
653    
654        The Unix I/O library of SML/NJ on cygwin does not understand
655    Windows style pathname, so problems arise when SMLNJ_HOME is set
656    to a Windows style pathname.  _run-sml now converts SMLNJ_HOME
657    to a POSIX pathname on cygwin.
658    
659    ----------------------------------------------------------------------
660    Name: Matthias Blume (blume (at) tti - c (dot) org)
661    Date: 2004/12/16 13:35:00 CST
662    Tag: Release_110_52
663    Description:
664    
665    Last-minute changes incorporated into 110.52.  Release tag moved.
666    
667    The changes:
668       - HashString.hashString' -> HashString.hashSubstring
669       - bug fix in UnivariateStats
670    
671    ----------------------------------------------------------------------
672    Name: Matthias Blume (blume (at) tti - c (dot) org)
673    Date: 2004/12/15 23:40:00 CST
674    Tag: blume-20041215-hashSubstring
675    Description:
676    
677    - HashString.hashString' -> HashString.hashSubstring
678    - corresponding changes in atom.sml
679    - "de-compressed" (aka. un-obfuscated) code for UnivariateStats and
680      added some comments
681    
682    ----------------------------------------------------------------------
683    Name: Matthias Blume (blume (at) tti - c (dot) org)
684    Date: 2004/12/15 15:30:00 CST
685    Tag: (Release_110_52)
686    Description:
687    
688    New working version (110.52).  NEW BOOTFILES!
689    
690    ----------------------------------------------------------------------
691    Name: Matthias Blume (blume (at) tti - c (dot) org)
692    Date: 2004/12/15 12:45:00 CST
693    Tag: blume-20041215-spaces
694    Description:
695    
696    More on the space problem (this time for Win32).
697    
698    ----------------------------------------------------------------------
699    Name: Matthias Blume (blume (at) tti - c (dot) org)
700    Date: 2004/12/14 17:30:00 CST
701    Tag: blume-20041214-spaces
702    Description:
703    
704    Hacked some of the scripts (in particular: the installer) to cope with
705    spaces in filenames a bit better.  But beware: the current "solution"
706    is likely still full of bugs and inherently incomplete.  (We need to
707    do away with those shell scripts for a comprehensive solution.)
708    
709    ----------------------------------------------------------------------
710    Name: Matthias Blume (blume (at) tti - c (dot) org)
711    Date: 2004/12/13 14:45:00 CST
712    Tag: blume-20041213-ml-makedepend
713    Description:
714    
715    Fixed bug in code for ml-makedepend.
716    
717    ----------------------------------------------------------------------
718    Name: Matthias Blume (blume (at) tti - c (dot) org)
719    Date: 2004/12/09 16:30:00 CST
720    Tag: blume-20041209-statistics
721    Description:
722    
723    Added two simple but potentially useful statistics modules to SML/NJ Library.
724    (See CHANGES file there.)
725    
726    ----------------------------------------------------------------------
727    Name: Matthias Blume (blume (at) tti - c (dot) org)
728    Date: 2004/12/01 16:50:00 CST
729    Tag: blume-20041201-atom
730    Description:
731    
732    smlnj-lib:
733    
734      Added function HashString.hashString' for substrings.
735      Hand-inlined CharVector.foldl into HashString (for speed).
736      Modified implementation of structure Atom to avoid extracting
737      strings from substrings unless necessary.
738      (Also see CHANGES file for smlnj-lib.)
739    
740    ----------------------------------------------------------------------
741    Name: Matthias Blume (blume (at) tti - c (dot) org)
742    Date: 2004/11/24 22:15:00 CST
743    Tag: blume-20041124-cml
744    Description:
745    
746    Made sure CML compiles when Position = Int64.
747    
748    ----------------------------------------------------------------------
749    Name: Matthias Blume (blume (at) tti - c (dot) org)
750    Date: 2004/11/24 14:30:00 CST
751    Tag: blume-20041124-position
752    Description:
753    
754    The compiler can now be compiled in a mode that makes structure Position
755    equal to Int64.  The default, however, is unchanged (Position = Int31)
756    for the time being.
757    
758    To enable 64-bit positions, use the following procedure:
759    
760    1. Start sml
761    2. Autoload $smlnj/cmb.cm (if not already autoloaded)
762    3. Type
763          #set (CMB.symval "USE_64_BIT_POSITIONS") (SOME 1);
764    4. Run CMB.make() as usual.
765    
766    This is barely tested.  The only test so far was a little SML program
767    counting the number of characters in an 8-gigabyte file by
768    reading it character-by-character.  That test was successful.
769    
770    In support of 64-bit positions, a number of new functions have been
771    added to the runtime system.
772    
773    ----------------------------------------------------------------------
774    Name: Matthias Blume (blume (at) tti - c (dot) org)
775    Date: 2004/11/23 14:45:00 CST
776    Tag: blume-20041123-useFile
777    Description:
778    
779    Fixed a problem with unhelpful error messages related to problems with
780    .cm- or .sml files that appear as part of the sml command line.
781    
782    ----------------------------------------------------------------------
783    Name: Matthias Blume (blume (at) tti - c (dot) org)
784    Date: 2004/11/18 15:40:00 CST
785    Tag: Release_110_51
786    Description:
787    
788    New working version (110.51).  NEW BOOTFILES!
789    
790    ----------------------------------------------------------------------
791    Name: Matthias Blume (blume (at) tti - c (dot) org)
792    Date: 2004/11/18 15:35:00 CST
793    Tag: <none>
794    Description:
795    
796    Enabled dlopen and friends for FreeBSD (as recommended by Johannes 5
797    Joemann).
798    
799    ----------------------------------------------------------------------
800    Name: Allen Leung (leunga (at) reservoir (dot) com)
801    Date: 2004/11/17 16:05:21 EST 2004
802    Tag: leunga-20041117-mlrisc-live-kill
803    Description:
804    
805       Added support for MLTree constructs LIVE and KILL
806    to all the architectures.
807    
808    ----------------------------------------------------------------------
809    Name: Matthias Blume (blume (at) tti - c (dot) org)
810    Date: 2004/11/13 00:20:00 CST
811    Tag: blume-20041113-versiontool
812    Description:
813    
814    - Stripped down the versiontool: It now only handles the version number.
815      The date string is generated at bootstrap time (during makeml).
816    
817    - In a previous commit, fixed a minor issue with how polyequal is being
818      translated.  In particular, the code now "looks through" abstractions.
819      This results in slightly fewer polyEqual warnings and hopefully slightly
820      more efficient code.  Important examples for where this matters are
821      the new int64 and word64 types.
822    
823    ----------------------------------------------------------------------
824    Name: Matthias Blume (blume (at) tti - c (dot) org)
825    Date: 2004/11/12 00:30:00 CST
826    Tag: blume-20041112-int64
827    Description:
828    
829    Structure Int64 fully hooked in.  (The implementation is not very
830    efficient, though.)
831    
832    ----------------------------------------------------------------------
833    Name: Matthias Blume (blume (at) tti - c (dot) org)
834    Date: 2004/11/11 17:30:00 CST
835    Tag: blume-20041111-more64
836    Description:
837    
838    All the pieces of Word64 are now there, with the exception of the
839    conversions from and to LargeWord. (Eventually these need to be identities,
840    but for the time being they don't even make sense because LargeWord is
841    32-bit wide.)
842    
843    Also started to add similar support for Int64, but major pieces of that
844    are still missing.
845    
846    ----------------------------------------------------------------------
847    Name: Matthias Blume (blume (at) tti - c (dot) org)
848    Date: 2004/11/11 00:15:00 CST
849    Tag: blume-20041111-word64
850    Description:
851    
852    Structure Word64 is now (almost) complete, word literals and patterns
853    seem to work.  There are a few odd pieces missing.  In particular,
854    I didn't do the {from,to}LargeWord parts because LargeWord is still Word32
855    at the moment.
856    
857    Making Word64 official would mean that LargeWord becomes Word64.  But
858    this requires extreme care because most word-word conversions have to
859    go through LargeWord, so making a mistake means loss of efficiency or
860    worse.  Eventually there will be a solution similar to (but actually
861    simpler than) what I did with IntInf.
862    
863    ----------------------------------------------------------------------
864    Name: Matthias Blume (blume (at) tti - c (dot) org)
865    Date: 2004/11/10 18:12:00 CST
866    Tag: blume-20041110-64bit
867    Description:
868    
869    More 64-bit hacking (but still not even half-way there yet).
870    Also, some assorted improvements to the handling of 8-bit words.
871    
872    ----------------------------------------------------------------------
873    Name: Matthias Blume (blume (at) tti - c (dot) org)
874    Date: 2004/11/09 17:50:00 CST
875    Tag: <none>
876    Description:
877    
878    Started adding some infrastructure for supporting 64-bit int- and
879    word-types.  (Still in its very early stages.)
880    
881    ----------------------------------------------------------------------
882    Name: Matthias Blume (blume (at) tti - c (dot) org)
883    Date: 2004/10/28 10:45:00 CDT
884    Tag: Release_110_50
885    Description:
886    
887    New working version (110.50).  NEW BOOTFILES!
888    
889                 =====================
890    
891    Also:
892    
893      - Changend config/srcarchiveurl from a file just
894        containing the URL string into a file containing
895        shell script code.  The code has access to the $VERSION variable.
896      - Made corresponding changes to config/install.sh and config/unpack.
897      - Default contents of config/srcarchiveurl uses $VERSION and
898        normally does not have to be edited to reflect a version change.
899    
900      (As a result, a version change can be done by just editing
901       config/version, the rest is now automatic.)
902    
903    ----------------------------------------------------------------------
904    Name: Matthias Blume (blume (at) tti - c (dot) org)
905    Date: 2004/10/27 17:50:00 CDT
906    Tag: blume-20041027-btrace-msg
907    Description:
908    
909    BackTrace.monitor now also reports the source of the exception that
910    triggered the trace.
911    
912    ----------------------------------------------------------------------
913    Name: Matthias Blume (blume (at) tti - c (dot) org)
914    Date: 2004/10/27 17:20:00 CDT
915    Tag: blume-20041027-x86-c-calls
916    Description:
917    
918    This is the HISTORY entry for two earlier commits, both concerning
919    the x86 c-calls code in MLRISC:
920    
921      - added a missing LOAD in the code that deals with struct arguments
922      - made sure the caller does not add the wrong number of bytes to the
923        stack pointer after a call of a function returning a struct
924        (the callee already pops the implicit argument which points to
925         the space reserved for the result)
926    
927    ----------------------------------------------------------------------
928    Name: Allen Leung (leunga (at) reservoir (dot) com)
929    Date: 2004/10/24 14:00:00 EST
930    Tag: leunga-20041024-x86-gas-fucomip
931    Description:
932    
933       John discovered a bug in the syntax of fucomip.
934       The opcodes FU?COMIP? have been changed to
935    
936           fu?comip? %st(i), %st
937    
938    ----------------------------------------------------------------------
939    Name: Matthias Blume (blume (at) tti - c (dot) org)
940    Date: 2004/10/20 15:06:00 CDT
941    Tag: blume-20041020-standalone-backtrace
942    Description:
943    
944    Added a mechanism for getting back-trace information from standalone
945    programs.  Here is how it works:
946    
947       1. The part of the program from which you want to get backtrace
948          information (usually the whole program) should be wrapped with
949          BackTrace.monitor.  This is a (unit->'a)->'a function, and your
950          main program could be modified from something like
951    
952             fun main (pgm, args) = ...
953    
954         to
955    
956            fun main (pgm, args) = BackTrace.monitor (fn () => ...)
957    
958       2. To be able to access BackTrace.monitor, you have to add
959          library $smlnj-tdp/plugins.cm to the .cm file that contains your
960          main function.
961    
962       3. Remove all compiled code (i.e., all the .cm/ subdirectories that
963          CM might have created in the past for your project).
964    
965       4. Build the system using this command line:
966    
967             ml-build -Ctdp.instrument=true \$smlnj-tdp/back-trace.cm \
968                      myprog.cm MyProg.main myprog
969    
970          instead of the usual
971    
972             ml-build myprog.cm MyProg.main myprog
973    
974    I changed a library name:
975    
976       $/trace-debug-profile.cm --> $smlnj-tdp/plugins.cm
977    
978    New libraries:
979    
980       $smlnj-tdp/back-trace.cm
981          -- when loaded causes the back-trace plugin to be installed
982       $smlnj-tdp/coverage.cm
983          -- when loaded causes the coverage plugin to be installed
984    
985    ----------------------------------------------------------------------
986    Name: Matthias Blume (blume (at) tti - c (dot) org)
987    Date: 2004/10/18 16:45:00 CDT
988    Tag: blume-20041018-groupowner
989    Description:
990    
991    Added an "obsolete" warning for the "group owner" syntax to CM's parser.
992    
993    Eliminated group owner specs from .cm files throughout the source tree.
994    
995    ----------------------------------------------------------------------
996    Name: Matthias Blume (blume (at) tti - c (dot) org)
997    Date: 2004/10/15 15:45:00 CDT
998    Tag: blume-20041015-coverage
999    Description:
1000    
1001    * Test coverage tool added!
1002    * Further reorganization of tracing-, debugging-, and profiling support:
1003    
1004        - moved original BTImp -- now called BackTrace -- into a separate
1005          library called $/trace-debug-profile.cm
1006        - eliminated all mentions of BTrace from SMLofNJ.Internals
1007        - only the instrumentation mechanism is now left in the compiler proper
1008        - BackTrace module is a plugin which is NOT plugged in by default
1009        - Coverage module is another such plugin
1010    
1011    To get the benefits of any of these plugin modules, the code in
1012    question must be compiled with tdp instrumentation turned on.  This
1013    can be done by setting SMLofNJ.Internals.TDP.mode to true.
1014    (The ref cell is also controlled via the -Ctdp.instrument=... switch.)
1015    
1016    Plugins are selected at link time.  (Pre-compiled instrumented code
1017    can be re-loaded with different plugins in effect.)  When an
1018    instrumented module is linked, whatever plugins are at that time
1019    enabled will come into effect for that module.
1020    
1021    To enable the back-trace plugin, load library $/trace-debug-profile.cm
1022    and invoke BackTrace.install() (e.g., from the interactive prompt).
1023    To enable the coverage plugin, load the same library and invoke
1024    Coverage.install().
1025    
1026    Back-traces are generated automatically on uncaught exceptions and
1027    when the code in question explicitly invokes BackTrace.trigger().
1028    
1029    Coverage (and execution frequency-) information must be queried
1030    explicitly by calling Coverage.not_covered and Coverage.hot_spots.
1031    
1032    ----------------------------------------------------------------------
1033    Name: Matthias Blume (blume (at) tti - c (dot) org)
1034    Date: 2004/10/14 17:40:00 CDT
1035    Tag: blume-20041014-tdp-core
1036    Description:
1037    
1038    Snapshot of a significant overhaul of how the trace/debug/profile support
1039    is hooked into the system (specifically: Core and SMLofNJ.Internals).
1040    
1041    ----------------------------------------------------------------------
1042    Name: Matthias Blume (blume (at) tti - c (dot) org)
1043    Date: 2004/10/13 16:34:00 CDT
1044    Tag: blume-20041013-tdp
1045    Description:
1046    
1047    Some rationalization of names:
1048    
1049       structure BTrace -> structure TDPInstrument
1050       etc.
1051    
1052    This is is preparation of using the original back-trace
1053    instrumentation for other purposes.  "TDP" stands for
1054    Trace/Debug/Profile.
1055    
1056    The control flag controlling whether instrumentation is on or off is now
1057    registered under a different name, so instead of running sml as
1058    
1059     sml -Cinstrument.btrace-mode=true
1060    
1061    one has to say
1062    
1063     sml -Ctdp.instrument=true
1064    
1065    ----------------------------------------------------------------------
1066    Name: Matthias Blume (blume (at) tti - c (dot) org)
1067    Date: 2004/10/11 16:37:00 CDT
1068    Tag: blume-20041011-regions
1069    Description:
1070    
1071    Made some minor modifications to elabcore.sml to have source regions
1072    be propagated more tightly -- resulting in better (i.e., smaller)
1073    regions being reported in error- and debug messages.
1074    
1075    ----------------------------------------------------------------------
1076    Name: Matthias Blume (blume (at) tti - c (dot) org)
1077    Date: 2004/10/08 22:50:00 CDT
1078    Tag: blume-20041008-cmkw
1079    Description:
1080    
1081    Fixed handling of keywords in .cm files: After seeing "is" the lexer
1082    treats subsequent occurrences of "group", "library", "source", "is",
1083    "*", and "-" as ordinary identifiers rather than keywords.
1084    
1085    Most seriously, this fixes a problem with CM's "shell" tool.  The tool
1086    is supposed to accept a tool argument called "source", but this did
1087    not work because of the clash with the keyword.
1088    
1089    ----------------------------------------------------------------------
1090    Name: Matthias Blume (blume (at) tti - c (dot) org)
1091    Date: 2004/10/07 16:00:00 CDT
1092    Tag: blume-20041007-cleanup
1093    Description:
1094    
1095    Assorted cleanup work:
1096    
1097      - got rid of intstrmap in favor of using the library's
1098        hash table implementation
1099      - threw out most of the pathnames stuff, as it was not used anyway
1100      - simplified tokentable implementation
1101      - fixed some minor spelling errors
1102    
1103    ----------------------------------------------------------------------
1104    Name: Matthias Blume (blume (at) tti - c (dot) org)
1105    Date: 2004/10/06 15:15:15 CDT
1106    Tag: blume-20041006-handler
1107    Description:
1108    
1109    Cleaned up the absyn to reflect the invariant that HANDLE always
1110    carries a FNexp as part of the type definition.  This eliminates some
1111    superfluous sanity checks at runtime down the road.
1112    
1113    Some minor cleanup of the btrace code.
1114    
1115    ----------------------------------------------------------------------
1116    Name: Matthias Blume (blume (at) tti - c (dot) org)
1117    Date: 2004/10/01 10:20:30 CDT
1118    Tag: blume-20041001-slave
1119    Description:
1120    
1121    Added hack to make slave mode work in the presence of the version
1122    tool.  (Still, since the master does two passes over the code for
1123    CMB.make, the release number gets bumped twice when slaves are
1124    attached.  I don't know if this is worth fixing...)
1125    
1126    ----------------------------------------------------------------------
1127    Name: Matthias Blume (blume (at) tti - c (dot) org)
1128    Date: 2004/09/30 10:55:00 CDT
1129    Tag: blume-20040930-version
1130    Description:
1131    
1132    * Moved the "version" magic into its own little library under
1133      src/system/smlnj/internal.  This avoids expensive reconstruction of
1134      a stable src/compiler/core.cm.
1135    
1136    * At the same time, structure CompilerVersion is now known as
1137      structure SMLNJVersion.
1138    
1139    * Arranged for the version tool to NOT kick in when rebuilding the system
1140      (makeml -rebuild, fixpt).  Otherwise one would never reach a fixpoint.
1141      Also, loading the versiontool does not work when rebuilding the system
1142      because CM is not properly initialized at that time.
1143    
1144    ----------------------------------------------------------------------
1145    Name: Matthias Blume (blume (at) tti - c (dot) org)
1146    Date: 2004/09/29 14:00:00 CDT
1147    Tag: blume-20040929-autoversion
1148    Description:
1149    
1150    Implemented some CM magic to have
1151        file src/compiler/TopLevel/main/version.sml
1152    generated automagically.
1153    The version is taken from two files: config/version and config/release.
1154    The first is expected to contain a two-part version number such as 110.49.
1155    The second should contain a single number, but it may be missing.
1156    
1157    If the environment variable VERSIONTOOL_BUMP_RELEASE is defined at the
1158    time the version tool is loaded (which is the first time you say CMB.make),
1159    then the tool will increment the value stored in config/release every
1160    time CMB.make is invoked.
1161    
1162    The binfile format is now insensitive to anything beyond the first
1163    two components of a version number, so bumping the release does not render
1164    binfiles incompatible.  Auto-bumping can be used to keep track of versions
1165    during development without invalidating existing binfiles.
1166    
1167    In any case, every CMB.make updates the date information in version.sml.
1168    (This is the date that is printed in the banner.)
1169    
1170    ----------------------------------------------------------------------
1171    Name: Matthias Blume (blume (at) tti - c (dot) org)
1172    Date: 2004/09/28 10:53:00 CDT
1173    Tag: blume-20040928-controls
1174    Description:
1175    
1176    Some cleanup of the controls code.
1177    
1178    ----------------------------------------------------------------------
1179    Name: Matthias Blume (blume (at) tti - c (dot) org)
1180    Date: 2004/09/27 22:08:00 CDT
1181    Tag: blume-20040927-controls
1182    Description:
1183    
1184    Added two pieces of functionality to the Controls interface:
1185    
1186    1.   val save'restore: 'a control -> unit -> unit
1187    
1188       grabs the current value of the control in stage 1 and restores it
1189       in stage 2.
1190    
1191    2.   val set' : 'a control * 'a -> unit -> unit
1192    
1193       stores the given value into the control in stage 2 (i.e., delayed)
1194       but does all error checking in stage 1.
1195       (This is for string controls that need to do parse their argument --
1196        something that might fail.  In some cases, notably in CM, one
1197        already knows the intended argument but wants to delay the actual
1198        assignment until a time when error recovery would be more difficult.)
1199    
1200    Changed the handling of controls in tool arguments to classes "sml" and
1201    "lazysml":
1202       - use Controls.save'restore as a more robust way of restoring the
1203         old value (in particular: without having to re-parse the string)
1204       - use controls to handle the "overload" keyword in the init group
1205         (I believe this change actually fixes a long-standing obscure bug.)
1206    
1207    ----------------------------------------------------------------------
1208    Name: Matthias Blume (blume (at) tti - c (dot) org)
1209    Date: 2004/09/27 17:00:00 CDT
1210    Tag: blume-20040927-lazysml
1211    Description:
1212    
1213    Added a new tool class called "lazysml" to CM's tool chest.  The only
1214    difference to "sml" is that compilation is done with Control.lazysml
1215    set to true.  A source of class "lazysml" is automatically recognized
1216    by a file name suffix of ".lml".
1217    
1218    In addition to the above feature, the original class "sml" now also
1219    supports a tool argument "lazy" which has the same effect.  As a
1220    result, the following three lines are equivalent:
1221    
1222        foo.sml : lazysml
1223        foo.sml : sml (lazy)
1224        foo.sml (lazy)
1225    
1226    The setting goes into effect both during parsing and during
1227    compilation.  The original setting is restored right after parsing and
1228    after compilation, respectively.
1229    
1230    In addition to all the above, there is also a general mechanism to set
1231    ANY of the "controls" that are available at the command line via
1232    "-C..." on a per-sml-file basis.  The same rules that apply for "lazy"
1233    apply as well. (In fact, "lazy" is implemented as a special case of
1234    the general mechanism.)
1235    
1236    The .cm file syntax uses a new keyword tool argument called "with".
1237    There are several ways of indicating the desired settings:
1238    
1239        foo.sml (with:parser.quotations=true)
1240        foo.sml (with:(name:parser.quotations value:true))
1241        foo.sml (with:(name:name1 value:value1 name:name2 value:value2 ...))
1242        foo.sml (with:(name1=value1 name2=value2 ...))
1243        foo.sml (with:(name1=value1 name:name2 value:value2 name3=value3 ...))
1244    
1245        etc.
1246    
1247    Another possible abbreviation is to leave out the =v or value:v part
1248    if the name refers to a boolean control (in which case the value is
1249    taken to be true).  Thus, one could get lazy sml also by saying:
1250    
1251        foo.sml (with:parser.lazy-keyword=true)
1252        foo.sml (with:parser.lazy-keyword)
1253        foo.sml (with:(name:parser.lazy-keyword value:true))
1254        foo.sml (with:(name:parser.lazy-keyword))
1255    
1256    ----------------------------------------------------------------------
1257    Name: Matthias Blume (blume (at) tti - c (dot) org)
1258    Date: 2004/09/24 16:22:00 CDT
1259    Tag: blume-20040924-ppc-long-branch
1260    Description:
1261    
1262    Turned message about "emiting long form of branch" off by default.
1263    Added a control flag to turn it back on when desired.
1264    
1265    ----------------------------------------------------------------------
1266    Name: Matthias Blume (blume (at) tti - c (dot) org)
1267    Date: 2004/09/24 16:05:00 CDT
1268    Tag: blume-20040924-rounding
1269    Description:
1270    
1271    Applied patch for setting rounding modes under Mac OS X.  Thanks to
1272    Melissa O'Neill for providing the code!
1273    
1274    ----------------------------------------------------------------------
1275    Name: Matthias Blume (blume (at) tti - c (dot) org)
1276    Date: 2004/09/23 17:30:00 CDT
1277    Tag: blume-20040923-envvars
1278    Description:
1279    
1280    1. Changed definition of type ControlRegistry.registry_tree to
1281       include control_info (i.e., the name of the controlling environment
1282       variable).
1283    
1284    2. Added command-line flags -e and -E to print the names of environment
1285       variables that can be used to control internal settings.
1286       (This uses the new API mentioned in 1.)
1287    
1288    ----------------------------------------------------------------------
1289    Name: Matthias Blume (blume (at) tti - c (dot) org)
1290    Date: 2004/09/13 16:50:00 CDT
1291    Tag: Release_110_49
1292    Description:
1293    
1294    New working version (110.49).  NEW BOOTFILES!
1295    
1296    ----------------------------------------------------------------------
1297    Name: Matthias Blume
1298    Date: 2004/09/13 16:20:00 CDT
1299    Tag: blume-20040913-config-mlrisc
1300    Description:
1301    
1302    Put target "mlrisc" back into the default list.
1303    (There is no harm in having it, and some users have expressed their
1304    wish to have "mlrisc" included by default.)
1305    
1306    ----------------------------------------------------------------------
1307    Name: John Reppy
1308    Date: 2004/09/13
1309    Tag: jhr-20040913-signals
1310    Description:
1311    
1312    Fixed the signal masking code to properly nest mask/unmask operations
1313    on a per-signal basis.
1314    
1315    ----------------------------------------------------------------------
1316    Name: Matthias Blume (blume (at) tti - c (dot) org)
1317    Date: 2004/09/08 13:20:00 CDT
1318    Tag: blume-20040908-heap-magic
1319    Description:
1320    
1321    Bumped the heap macig number to 0x09082004 to account for the changed
1322    layout of the ML frame under MacOS X.
1323    
1324    ----------------------------------------------------------------------
1325    Name: Allen Leung (leunga (at) reservoir (dot) com)
1326    Date: 2004/09/03 11:26:00 EST
1327    Tag: leunga-20040903-cygwin-install
1328    Description:
1329    
1330       Added a patch to _arch-n-opsys to enable the Cygwin runtime.
1331    The Cygwin runtime is turned on by setting the environment
1332    variable SMLNJ_CYGWIN_RUNTIME to 1.
1333    
1334    ----------------------------------------------------------------------
1335    Name: Matthias Blume (blume (at) tti - c (dot) org)
1336    Date: 2004/08/31 17:15:00 CDT
1337    Tag: blume-20040831-core
1338    Description:
1339    
1340    Added some exports to src/compiler/core.cm upon request by J. Joemann.
1341    
1342    ----------------------------------------------------------------------
1343    Name: Matthias Blume (blume (at) tti - c (dot) org)
1344    Date: 2004/08/30 17:55:00 CDT
1345    Tag: blume-20040830-installer
1346    Description:
1347    
1348    Upon request by Johannes Joemann:
1349    
1350    - improved ML code of installer to fall back to coping when renaming
1351      fails (i.e., when source and target are on different file systems);
1352      the code compiles but has yet to be tested in anger
1353    - removed mlrisc from list of default targets (config/targets)
1354    
1355    ----------------------------------------------------------------------
1356    Name: Matthias Blume (blume (at) tti - c (dot) org)
1357    Date: 2004/08/27 17:20:00 CDT
1358    Tag: blume-20040827-ptreql
1359    Description:
1360    
1361    Added ptreql primop to structure InlineT (upon request from Larry
1362    Paulson).
1363    
1364    ----------------------------------------------------------------------
1365    Name: Allen Leung (leunga (at) reservoir (dot) com)
1366    Date: 2004/08/15 21:21:00 EST
1367    Tag: leunga-110_48-udgraph
1368    Description:
1369    
1370    Another bug fix from Carl Hauser:
1371    
1372    diff /net/niflab/smlnj48/src/MLRISC/graphs/udgraph.sml udgraph.sml
1373    > 48c48
1374    > <              | rmv((e as (k,_))::es,L) = rmv(es,if k = i then es else
1375    > e::L)
1376    > ---
1377    > >              | rmv((e as (k,_))::es,L) = rmv(es,if k = i then L else e::L)
1378    > Without this, any deletion of an edge in an undirected graph does severe
1379    > violence to the graph.
1380    
1381    ----------------------------------------------------------------------
1382    Name: Allen Leung (leunga (at) reservoir (dot) com)
1383    Date: 2004/08/10 23:35:00 EST
1384    Tag: leunga-110_48-ppc
1385    Description:
1386    
1387       The IBM/MacOS syntax switch on PPC was incorrectly swapped.
1388    
1389    ----------------------------------------------------------------------
1390    Name: Matthias Blume (blume (at) tti - c (dot) org)
1391    Date: 2004/08/10 12:00:00 CDT
1392    Tag: Release_110_48
1393    Description:
1394    
1395    New working version (110.48).  NEW BOOTFILES!
1396    
1397    ----------------------------------------------------------------------
1398    Name: Allen Leung (leunga (at) reservoir (dot) com)
1399    Date: 2004/08/09 12:21:00 EST
1400    Tag: leunga-110_47-dijsktra
1401    Description:
1402    
1403    Bug fix from Carl Hauser:
1404    
1405    single_source_shortest_paths in dijkstra.sml was observed to get wrong
1406    answers (by comparing to single_source_shortest_paths in bellman-ford.sml).
1407    
1408    The problem is that following the expression A.update(dist,s,Num.zero)
1409    it is necessary to update the priority queue using Q.decreaseWeight(Q,s).
1410    
1411    ----------------------------------------------------------------------
1412    Name: Matthias Blume (blume (at) tti - c (dot) org)
1413    Date: 2004/08/06 18:10:00 CDT
1414    Tag: blume-20040806-cmdline
1415    Description:
1416    
1417    Fiddled with handling of command-line options:
1418    
1419      * sml now quits after processing the command line
1420        if -H, -S, -h<n>, or -s<n> appears as the last
1421        command-line argument
1422      * a new option -q terminates the session when encountered on
1423        the command line; subsequent arguments will be ignored
1424      * bug fixes: short (erroneous) arguments are no longer ignored
1425        completely
1426    
1427    ----------------------------------------------------------------------
1428    Name: Allen Leung (leunga (at) reservoir (dot) com)
1429    Date: 2004/08/04 18:17:00 EST
1430    Tag: leunga-110_47-ppc-ibm-asm
1431    Description:
1432    
1433      - Added minimal IBM assembly syntax support for PowerPC.
1434    
1435      - Cygwin: manually changed the file cygwin.def.  Some exported symbols have
1436        been altered in the runtime.  We need an automatic way to keep the file
1437        in sync.
1438    
1439    ----------------------------------------------------------------------
1440    Name: Matthias Blume (blume (at) tti - c (dot) org)
1441    Date: 2004/08/04 14:00:00 CDT
1442    Tag: Release_110_47
1443    Description:
1444    
1445    New working version (110.47).  NEW BOOTFILES!
1446    
1447    ----------------------------------------------------------------------
1448    Name: Matthias Blume (blume (at) tti - c (dot) org)
1449    Date: 2004/08/03 14:25:00 CDT
1450    Tag: blume-20040803-callingconv
1451    Description:
1452    
1453    Added low-level support for choosing C calling conventions by
1454    twiddling the type of rawccall.  (See
1455    src/compiler/Semant/types/cproto.sml for details.)
1456    
1457    ----------------------------------------------------------------------
1458    Name: Matthias Blume (blume (at) tti - c (dot) org)
1459    Date: 2004/08/02 15:55:00 CDT
1460    Tag: blume-20040802-backout
1461    Description:
1462    
1463    Backed out of change to win32-filesys.c.  The earlier patch to
1464    get_file_time caused CM to produce files with the wrong time stamp.
1465    
1466    ----------------------------------------------------------------------
1467    Name: Matthias Blume (blume (at) tti - c (dot) org)
1468    Date: 2004/08/02 14:45:00 CDT
1469    Tag: blume-20040802-nlffi-win32
1470    Description:
1471    
1472    Added NLFFI support for Win32, adapted from a patch provided by David
1473    Hansel.  This is currently completely untested.  Also, the issue
1474    concerning stdcall vs. ccall is still unresolved.
1475    
1476    ----------------------------------------------------------------------
1477    Name: Matthias Blume (blume (at) tti - c (dot) org)
1478    Date: 2004/07/30 17:55:00 CDT
1479    Tag: blume-20040730-various
1480    Description:
1481    
1482    Gearing up towards 110.47...
1483    
1484    - various minor bugfixes to ml-nlffigen
1485    - a beginning of a manual for nlffi
1486    
1487    - eliminated 'export name=value' in config/install.sh as this does
1488      not work with certain versions of /bin/sh
1489      (Thanks to David King at Motorola for catching this.)
1490    
1491    - several bugfixes provided or suggested by David Hansel at Reactive Systems:
1492      - added a test for tm==NULL to gmtime.c and localtime.c
1493      - applied patch for incorrect GetFileTime under win32
1494      - toSeconds -> toMilliseconds in Win32/win32-process.sml
1495    
1496    ----------------------------------------------------------------------
1497    Name: Matthias Blume (blume (at) tti - c (dot) org)
1498    Date: 2004/07/21 18:20:00 CDT
1499    Tag: blume-20040721-nlffigen
1500    Description:
1501    
1502    - Fixed minor issue in ml-nlffigen: Now generate structure T_foo
1503      for a typedef to an incomplete type, but leave out the "typ" member.
1504      (This is just for consistency.)
1505    - Started to produce what is supposed to become better (i.e., comprehensive)
1506      documentation of what ml-nlffigen does and produces.
1507    
1508    ----------------------------------------------------------------------
1509    Name: Matthias Blume (blume (at) tti - c (dot) org)
1510    Date: 2004/07/14 16:25:00 CDT
1511    Tag: blume-20040714-union
1512    Description:
1513    
1514    Added C_UNION to c-calls/c-types.sml and updated the machinery
1515    (ml-nlffigen, cproto.sml) that conveys C function interface
1516    information to the code generator.
1517    
1518    However, the actual architecture-specific implementation of function
1519    arguments and results that are C unions is still not implemented.
1520    
1521    ----------------------------------------------------------------------
1522    Name: Allen Leung (leunga (at) reservoir (dot) com)
1523    Date: 2004/07/14 14:38:00 EST
1524    Tag: leunga-110_46_1-ppc-lwzu
1525    Description:
1526    
1527        Added these instructions to the PowerPC architecture:
1528        LBZU(X), LHZU(X), LWZU(X),
1529        STWU(X), STFDU, STFSU
1530    
1531        etc...
1532    
1533        Note: I haven't added their instruction encoding into the description.
1534    
1535    ----------------------------------------------------------------------
1536    Name: Allen Leung (leunga (at) reservoir (dot) com)
1537    Date: 2004/07/13 15:04:00 EST
1538    Tag: leunga-110_46_1-ppc-lwarx
1539    Description:
1540    
1541       Added the two instructions LWARX and STWCX to the PowerPC
1542    instruction set.
1543    
1544       A (untested) rewrite of loop-structure.sml.  The old version
1545    is completely broken.
1546    
1547    ----------------------------------------------------------------------
1548    Name: Matthias Blume (blume (at) tti - c (dot) org)
1549    Date: 2004/07/13 13:50:00 CDT
1550    Tag: blume-20040713-nlffi
1551    Description:
1552    
1553    - use paramAlloc to report c-calls with too many arguments
1554      (for PPC version where parameter area is pre-allocated)
1555    - added ccall_maxargspace to machspec (to implement the above)
1556    - made "make" commend in CM's "make" tool configurable
1557    - added option (default: on) for passing the name of the SML/NJ's "bin"
1558      directory to "make"; the call looks like this:
1559    
1560         make <options> SMLNJ_BINDIR=<dir> <target>
1561    
1562      This can be used by the Makefile to, e.g., pick the "right" version
1563      of ml-nlffigen.
1564    - minor code tweaks
1565    
1566    ----------------------------------------------------------------------
1567    Name: Matthias Blume (blume (at) tti - c (dot) org)
1568    Date: 2004/07/12 22:50:00 CDT
1569    Tag: blume-110_46_1-macosx-nlffi
1570    Description:
1571    
1572    NLFFI under Mac OS X now working (sort of).  This is largely untested,
1573    though.
1574    
1575    Note:  1. You have to make a new, clean build of the runtime system.
1576           2. There are new BOOTFILES, you have to use them!
1577              (Doing the bootstrap process yourself would be *very* painful!
1578               If you absolutely have to do it, build the system under
1579               a different architecture and then cross-compile.)
1580    
1581    Version bumped to 110.46.1 to account for runtime data format changes.
1582    
1583    ----------------------------------------------------------------------
1584    Name: Matthias Blume (blume (at) tti - c (dot) org)
1585    Date: 2004/06/18 14:30:00 CDT
1586    Tag: blume-20040618-unix
1587    Description:
1588    
1589    Changed the implementation of structure Unix so that the same stream
1590    is returned every time one of the {text,bin}{In,Out}streamOf functions
1591    is invoked on the same proc.  This is not what the spec currently
1592    says -- although IMO it arguably should.  (See discussion below.)
1593    
1594    ----------------------------------------------------------------------
1595    Name: Matthias Blume (blume (at) tti - c (dot) org)
1596    Date: 2004/06/17 18:15:00 CDT
1597    Tag: Release_110_46
1598    Description:
1599    
1600    New working version (110.46).  NEW BOOTFILES!
1601    
1602    ----------------------------------------------------------------------
1603    Name: Matthias Blume (blume (at) tti - c (dot) org)
1604    Date: 2004/06/17 17:20:00 CDT
1605    Tag: blume-20040617-timer-unix
1606    Description:
1607    
1608    Changed the interface of structures Timer and Unix to match the most
1609    recent Basis spec.
1610    
1611    In the case of Unix there still seems to be an open/weird issue:
1612    
1613       The {text,bin}{In,Out}streamOf functions are supposed to create
1614       fresh streams whenever they are called -- as opposed to have them
1615       return the same stream every time.  This design is supposed to
1616       prevent space leaks caused by proc values hanging on to streams.
1617    
1618       The reap function, on the other hand, is supposed to close the
1619       streams.  This cannot be done without having a handle on the
1620       stream in proc after all...
1621    
1622    I took the liberty to implement the following stopgap solution:
1623    
1624      The proc value hangs on to the most recently created stream(s).
1625      Reap closes those.  If either or both of the two streams hadn't
1626      been created at all yet, then reap will close the corresponding
1627      file descriptors directly.
1628    
1629    PS: I don't understand the original space leak argument anymore.  If
1630    a proc hangs on to the imperative stream, then I/O operations on those
1631    will advance the state of the cached stream and avoid the space leak.
1632    
1633    ----------------------------------------------------------------------
1634    Name: Matthias Blume (blume (at) tti - c (dot) org)
1635    Date: 2004/05/28 16:45:00 CDT
1636    Tag: blume-20040528-basis
1637    Description:
1638    
1639    Added signature PACK_REAL and exported functor PrimIO.
1640    
1641    ----------------------------------------------------------------------
1642    Name: Matthias Blume (blume (at) tti - c (dot) org)
1643    Date: 2004/05/25 16:00:00 CDT
1644    Tag: blume-20040525-group-owner
1645    Description:
1646    
1647    CM now ignores (but still accepts) the "owner" information in group
1648    descriptions.  The owner of a group is its next enclosing
1649    library. Each group must have a unique owner.  (There is a virtual
1650    "toplevel" library that owns groups which are not nested within a real
1651    library.)  Previously, each group had to explicitly declare its owner,
1652    and CM would check that such a declaration is correct.  The new scheme
1653    is to have CM check that for each group there is precisely one owning
1654    library.
1655    
1656    The advantage of the new scheme is that the programmer no longer needs
1657    to maintain the somewhat annoying owner information.  The downside is
1658    that CM cannot enforce the ownership rule across multiple runs of
1659    CM.make.  Fortunately, enclosing the same group in two different
1660    libraries A and B which are not part of the same program does not
1661    cause real problems.
1662    
1663    ----------------------------------------------------------------------
1664    Name: Matthias Blume (blume (at) tti - c (dot) org)
1665    Date: 2004/05/20 16:00:00 CDT
1666    Tag: blume-20040520-win32
1667    Description:
1668    
1669    Made the win32 version work again.  (Strangely, a misplaced comma had
1670    slipped into win32-process.c which prevented the runtime from being
1671    compiled correctly.)
1672    
1673    Also, included a minor addition to ml-build.bat analogous to what was
1674    done in blume-20040519-ml-build.
1675    
1676    ----------------------------------------------------------------------
1677    Name: Matthias Blume (blume (at) tti - c (dot) org)
1678    Date: 2004/05/19 22:10:00 CDT
1679    Tag: blume-20040519-ml-build
1680    Description:
1681    
1682    Arranged for ml-build to clean up after itself a little bit better.
1683    The script generates a temporary SML source file and compiles it using
1684    CM, so CM generates metadata (GUID, SKEL, objectfile) for it.  It now
1685    gets rid of those at the end, so they don't accumulate under .cm.
1686    
1687    This required a minor change to install.sh because the name of the
1688    metadata directory (default: .cm) is actually configurable at
1689    installation time.
1690    
1691    ----------------------------------------------------------------------
1692    Name: Matthias Blume (blume (at) tti - c (dot) org)
1693    Date: 2004/05/18 15:50:00 CDT
1694    Tag: blume-20040518-mkreader
1695    Description:
1696    
1697    Added Posix.IO.mk{Bin,Text}{Reader,Writer} by lifting their respective
1698    implementations from internal modules PosixBinPrimIO and PosixTextPrimIO.
1699    
1700    ----------------------------------------------------------------------
1701    Name: Matthias Blume (blume (at) tti - c (dot) org)
1702    Date: 2004/05/11 14:35:00 CDT
1703    Tag: blume-20040511-win32sock
1704    Description:
1705    
1706    Added previously missing support for many socket-related functions
1707    under win32.  Thanks to David Hansel <hansel@reactive-systems.com>
1708    for the voluminous patch!
1709    
1710    (I have not tested this patch under win32 yet.)
1711    
1712    Here is David's e-mail:
1713    
1714    Hi,
1715    
1716    Attached to this email you find a diff against sml/nj 110.45
1717    that will enable socket support under Windows.
1718    
1719    To apply the patch (using unix or cygwin)
1720    1) gunzip runtime.diff.gz
1721    2) "cd" into "src/runtime" in the source tree of a fresh
1722       110.45 installation.
1723    3) patch -p 1 < [your/path/to]runtime.diff
1724    
1725    The code compiles fine but has NOT yet been extensively tested.
1726    I only ran a few tests for basic socket client functionality
1727    (which worked fine).  Especially the functions that use ioctl
1728    are not tested at all and might not work (see below).
1729    
1730    I implemented this since we want to move to a newer version of sml/nj
1731    but need socket support in order to use it.  This is the first time I
1732    even had a look at the sml/nj source,  so please review my changes
1733    before making this part of the distribution!  Here are a few issues
1734    that I think might be better for someone to solve who is more
1735    familiar with the sml/nj source (and socket programming):
1736    
1737    - getnetbyaddr.c and getnetbyname.c will raise a "not implemented"
1738      exception since I could not figure out what the windows equivalent
1739      of these functions is
1740    
1741    - In sockets-osdep.h there are a some #include statements that are
1742      only used in a few files that include sockets-osdep.h
1743    
1744    - In smlnj-sock-lib.c, function init_fn() calls WSAStartup() but
1745      does not process its return value since I don't know how to
1746      report an error upwards.
1747    
1748    - It would probably be good to have a call to WSACleanup() when
1749      the library is unloaded (if there is such a possibility).
1750      Otherwise I think Windows will take care of this automatically
1751      when the process finishes.
1752    
1753    - I used ioctlsocket() as a replacement for ioctl() but I have
1754      no idea if that is actually the proper replacement on Windows.
1755    
1756    - All these issues are marked in the code by "FIXME" comments.
1757    
1758    We use sml/nj extensively in our products and are quite happy
1759    with it.  I hope this contribution will help you.
1760    
1761    Keep up the good work!
1762    
1763    David
1764    
1765    ----------------------------------------------------------------------
1766    Name: Matthias Blume (blume (at) tti - c (dot) org)
1767    Date: 2004/05/11 14:20:00 CDT
1768    Tag: blume-20040511-installml
1769    Description:
1770    
1771    Fixed two bugs in installml script.  (Thanks to Vesa A. Norrman for
1772    the patch.)
1773    
1774    ----------------------------------------------------------------------
1775    Name: Matthias Blume (blume (at) tti - c (dot) org)
1776    Date: 2004/05/11 14:05:00 CDT
1777    Tag: blume-20040511-nlffi-netbsd
1778    Description:
1779    
1780    Added support for nlffi under netbsd.  (Thanks to Vesa A. Norrman for
1781    the patch.)
1782    
1783    ----------------------------------------------------------------------
1784    Name: Matthias Blume (blume (at) tti - c (dot) org)
1785    Date: 2004/05/11 12:05:00 CDT
1786    Tag: blume-20040511-exports
1787    Description:
1788    
1789    As per request by Adam Chlipala <adam@hcoop.net>, extended various
1790    export lists in compiler-related .cm-files.
1791    
1792    ----------------------------------------------------------------------
1793    Name: Matthias Blume (blume (at) tti - c (dot) org)
1794    Date: 2004/05/11 11:35:00 CDT
1795    Tag: blume-20040511-allsource
1796    Description:
1797    
1798    The installer now honors the "src-smlnj" target again, although its meaning
1799    has changed from "all sources required for the compiler" to "all sources
1800    the installer knows about".  In other words, if you enable "src-smlnj"
1801    in the "targets" file, then the installer will pull in sources for
1802    everything.  (Notice that this refers to source code only.  Compiled
1803    code is still only installed for modules that were requested explicitly
1804    or which are required for other modules that were requested explicitly.)
1805    
1806    ----------------------------------------------------------------------
1807    Name: Matthias Blume (blume (at) tti - c (dot) org)
1808    Date: 2004/04/23 17:40:00 CDT
1809    Tag: blume-20040423-ieee-scan
1810    Description:
1811    
1812    Fixed IEEEReal.scan (and .fromString) so that if there is an overflow
1813    in the exponent calculation we get INF or ZERO (depending on the mantissa
1814    and the sign of the exponent).
1815    
1816    ----------------------------------------------------------------------
1817    Name: Matthias Blume (blume (at) tti - c (dot) org)
1818    Date: 2004/04/23 10:40:00 CDT
1819    Tag: blume-20040423-ml-build
1820    Description:
1821    
1822    The ml-build script now terminates with a non-0 status when something
1823    goes wrong.
1824    
1825    ----------------------------------------------------------------------
1826    Name: Matthias Blume (blume (at) tti - c (dot) org)
1827    Date: 2004/04/22 16:35:00 CDT
1828    Tag: blume-20040422-Option
1829    Description:
1830    
1831    Made exception Option to be the same as exception Option.Option
1832    (as it should be).
1833    
1834    ----------------------------------------------------------------------
1835    Name: Allen Leung (leunga (at) reservoir (dot) com)
1836    Date: 2004/03/19 14:40:00 EST
1837    Tag: leunga-20040319-cygwin-nlffi
1838    Description:
1839    
1840    Fixed the runtime so that ml-nlffi-lib runs on the cygwin version
1841    of SML/NJ.  The problem is that
1842    
1843       lib = dlopen(NULL, ...)
1844       f   = dlsym(lib, "malloc");
1845    
1846    does not work on Windows unless we explicitly export symbols
1847    such as 'malloc' during linking.  We fixed this by explicitly
1848    exporting the required symbols with the magic gcc incantation:
1849    
1850        -Wl,--export-all cygwin.def
1851    
1852    where cygwin.def is a file containing all the symbols that
1853    we wish to export.
1854    
1855    I suspect this is a Windows problem and we'll have to
1856    do the same (somehow with windows compilers) when
1857    we build the native win32 version with the system
1858    calls LoadLibrary/GetProcAddress.
1859    
1860    ----------------------------------------------------------------------
1861    Name: Matthias Blume (blume (at) tti - c (dot) org)
1862    Date: 2004/03/04 16:35:00 CST
1863    Tag: blume-20040304-intinf-fmt
1864    Description:
1865    
1866    Fixed problem with IntInf.fmt (sign would show up on the right instead
1867    of on the left for BIN, OCT, and HEX).
1868    
1869    ----------------------------------------------------------------------
1870    Name: Matthias Blume (blume (at) tti - c (dot) org)
1871    Date: 2004/03/04 11:25:00 CST
1872    Tag: blume-20040304-symlinks
1873    Description:
1874    
1875    Fixed problem with installer script (unix only) where bin/ml-yacc and
1876    friends pointed (via symlinks) to absolute locations instead of just
1877    .run-sml.  This was reported by Vesa A Norrman.
1878    
1879    ----------------------------------------------------------------------
1880    Name: Matthias Blume (blume (at) tti - c (dot) org)
1881    Date: 2004/02/13 14:50:00 CST
1882    Tag: Release_110_45
1883    Description:
1884    
1885    New working version (110.45). New bootfiles.
1886    
1887    ----------------------------------------------------------------------
1888    Name: Matthias Blume (blume (at) tti - c (dot) org)
1889    Date: 2004/01/26 15:15:15 CST
1890    Tag: blume-20040126-toplevel
1891    Description:
1892    
1893    Improved handling of exceptions at the interactive toplevel.
1894    
1895    ----------------------------------------------------------------------
1896    Name: Matthias Blume (blume (at) tti - c (dot) org)
1897    Date: 2004/01/26 11:25:00
1898    Tag: blume-20040126-app
1899    Description:
1900    
1901    Type of top-level "app" corrected.
1902    Added code for setting vp_limitPtrMask to Win32-specific runtime.
1903    
1904    ----------------------------------------------------------------------
1905    Name: Matthias Blume (blume (at) tti - c (dot) org)
1906    Date: 2003/11/18 17:10 CST
1907    Tag: blume-20031118-basis-fiddle
1908    Description:
1909    
1910    - changed Timer interface to what might become the spec
1911    - POSIX_FLAGS -> BIT_FLAGS according to spec
1912    - some other minor discrepancies wrt. spec eliminated
1913    
1914    ----------------------------------------------------------------------
1915    Name: Matthias Blume (blume (at) tti - c (dot) org)
1916    Date: 2003/11/06 12:00:00 CST
1917    Tag: Release_110_44
1918    Description:
1919    
1920    New working version (110.44). New bootfiles.
1921    
1922    ----------------------------------------------------------------------
1923    Name: Matthias Blume (blume (at) tti - c (dot) org)
1924    Date: 2003/11/04 11:50:00 CST
1925    Tag: blume-20031104-move-libraries
1926    Description:
1927    
1928    Eliminated the "dont_move_libraries" directive in config/targets.
1929    (The mechanism was broken and could not be fixed easily.  Moreover,
1930    there does not seem to be any reason not to move all libraries into
1931    lib during installation.  I originally implemented this directive as a
1932    backward-compatibility feature when I first introduced the new CM.
1933    Now that things have been stable for a long time and going back to the
1934    old CM is not an option, there is no reason to keep it around.)
1935    
1936    ----------------------------------------------------------------------
1937    Name: Matthias Blume (blume (at) tti - c (dot) org)
1938    Date: 2003/11/03 16:00:00 CST
1939    Tag: blume-20031103-installdir
1940    Description:
1941    
1942    Made installer honor INSTALLDIR variable again.  (Thanks to Chris
1943    Richards for pointing out the problem and providing the solution.)
1944    
1945    ----------------------------------------------------------------------
1946    Name: Matthias Blume (blume (at) tti - c (dot) org)
1947    Date: 2003/10/01 17:05:00 CDT
1948    Tag: blume-20031001-lal-mlrisc
1949    Description:
1950    
1951    MLRISC bug fix from Lal.
1952    
1953    ----------------------------------------------------------------------
1954    Name: Matthias Blume (blume (at) tti - c (dot) org)
1955    Date: 2003/09/30 16:10:00 CDT
1956    Tag: blume-20030930-primio-bat
1957    Description:
1958    
1959    1. Added openVector, nullRd, and nullWr to PRIM_IO.
1960    2. Improved .bat files (for Win32 port) to make things work under Win95.
1961       (thanks to Aaron S. Hawley for this one)
1962    
1963    ----------------------------------------------------------------------
1964    Name: Matthias Blume (blume (at) tti - c (dot) org)
1965    Date: 2003/09/26 16:05:00 CDT
1966    Tag: blume-20030926-wrappriv
1967    Description:
1968    
1969    Added missing wrapper for privilege "primitive" in $smlnj/viscomp/core.cm.
1970    
1971    ----------------------------------------------------------------------
1972    Name: Matthias Blume (blume (at) tti - c (dot) org)
1973    Date: 2003/09/26 15:00:00 CDT
1974    Tag: blume-20030926-110_43_3
1975    Description:
1976    
1977    - additional cleanup
1978    - version number bump, NEW BOOTFILES
1979    
1980    ----------------------------------------------------------------------
1981    Name: Matthias Blume (blume (at) tti - c (dot) org)
1982    Date: 2003/09/26 12:00:00 CDT
1983    Tag: blume-20030926-ppautoload
1984    Description:
1985    
1986    I modified the read-eval-print loop so that the autoloader gets
1987    invoked whenever the prettyprinter tries to look up a symbol that
1988    is not currently defined in the toplevel environment but which
1989    appears in CM's autoload registry.  As a result, we see far fewer of
1990    those ?.Foo.Bar.xxx names in the prettyprinter's output.
1991    
1992    In addition to this I tried to clean up some pieces of the Basis
1993    implementation (e.g., Socket, Word8Array) in order to prevent other
1994    instances of these ?.Foo.Bar.xxx names from being printed.
1995    
1996    The mechanism that picks names for types still needs some work, though.
1997    (Right now it seems that if there is a type A.t which is defined to
1998    be B.u, but B is unavailable at toplevel, then A.t gets printed as
1999    "?.B.u" although the perhaps more sensible solution would be to use
2000    "A.t" in this case.  In other words, the prettyprinter should follow
2001    a chain of DEFtycs not farther than there are corresponding toplevel
2002    names in the current environment.)
2003    
2004    ----------------------------------------------------------------------
2005    Name: Matthias Blume (blume (at) tti - c (dot) org)
2006    Date: 2003/09/24 16:31:00 CDT
2007    Tag: blume-20030924-installer
2008    Description:
2009    
2010    Another installer tweak:  All the ML code for the installer is now
2011    compiled during CMB.make and put into a little library called
2012    $smlnj/installer.cm.  The installation then simply invokes
2013    
2014       sml -m $smlnj/installer.cm
2015    
2016    and everything happens automagically.
2017    
2018    Win32:  ML code senses value of environment variable SMLNJ_HOME.
2019    Unix:   ML code senses values of environment variables ROOT, CONFIGDIR,
2020            and BINDIR.
2021    
2022    The new scheme guarantees that the ML code responsible for the installation
2023    is in sync with the APIs of the main system.  Also, the installer is
2024    somewhat faster because the installer script is precompiled.
2025    
2026    ----------------------------------------------------------------------
2027    Name: Matthias Blume (blume (at) tti - c (dot) org)
2028    Date: 2003/09/24 15:35:00 CDT
2029    Tag: blume-20030924-synsock
2030    Description:
2031    
2032    Added a signature SYNCHRONOUS_SOCKET to basis.cm.  This is like SOCKET
2033    but excludes all non-blocking operations.  Defined SOCKET (in Basis)
2034    and CML_SOCKET in terms of SYNCHRONOUS_SOCKET.  Removed superfluous
2035    implementations of non-blocking operations from CML's Socket
2036    structure.
2037    
2038    ----------------------------------------------------------------------
2039    Name: Matthias Blume (blume (at) tti - c (dot) org)
2040    Date: 2003/09/24 15:10:05 CDT
2041    Tag: blume-20030924-sockets
2042    Description:
2043    
2044    1. Fixed SOCKET API and implementation to match Basis spec.
2045       This required changing the internal representation of sockets to one
2046       that remembers (for each socket file descriptor) whether it is currently
2047       blocking or non-blocking.  This state is maintained lazily (i.e., a system
2048       call is made only if the state actually needs to change).
2049    
2050    2. OS-specific details of sockets were moved into separate files, thus
2051       making it possible to unify the bulk of the socket implementations
2052       between Unix and Win32.
2053    
2054    3. CML's socket API changed accordingly.
2055       (Note that we need to remove non-blocking functions from this API
2056       since they are redundant in the case of CML!)
2057    
2058    4. CML's socket implementation now makes use of non-blocking functions
2059       provided by Basis, thus removing all OS-dependent code from this part
2060       of CML.
2061    
2062    5. Changed Real64.precision from 52 to 53.  Minor cleanup in Real64 code.
2063    
2064    ----------------------------------------------------------------------
2065    Name: Matthias Blume (blume (at) tti - c (dot) org)
2066    Date: 2003/09/22 12:10:00 CDT
2067    Tag: blume-20030922-110_43_2
2068    Description:
2069    
2070    Made a new interim version and bootfiles for developer's bootstrapping
2071    convenience.
2072    
2073       110.43.2 -- NEW BOOTFILES
2074    
2075    ----------------------------------------------------------------------
2076    Name: Matthias Blume (blume (at) tti - c (dot) org)
2077    Date: 2003/09/19 15:55:00 CDT
2078    Tag: blume-20030919-cmdir
2079    Description:
2080    
2081    1. new-install.sh -> install.sh
2082    2. changed default CM "metadata" directory name to ".cm" (instead of "CM")
2083    3. tweaked installer so that another name instead of .cm can be chosen
2084       at install time (by setting the CM_DIR_ARC environment variable
2085       during installation); once installation is complete, the name is
2086       fixed
2087    
2088    ----------------------------------------------------------------------
2089    Name: Matthias Blume (blume (at) tti - c (dot) org)
2090    Date: 2003/09/18 16:00:00 CDT
2091    Tag: blume-20030918-110_43_1
2092    Description:
2093    
2094    Made a new interim version and bootfiles for developer's bootstrapping
2095    convenience.
2096    
2097       110.43.1 -- NEW BOOTFILES
2098    
2099    ----------------------------------------------------------------------
2100    Name: Matthias Blume (blume (at) tti - c (dot) org)
2101    Date: 2003/09/18 15:20:00 CDT
2102    Tag: blume-20030918-misc
2103    Description:
2104    
2105    1. Exported fractionsPerSecond etc. from TimeImp (but not from Time as
2106       this seems to be controversial at the moment) and used those in
2107       Posix.ProcEnv.times.
2108    
2109    2. Added Time.{from,to}Nanoseconds to Time.
2110    
2111    3. Improved Real.{from,to}LargeInt by avoiding needless calculations.
2112       For example, fromLargeInt never needs to look at more than 3 "big
2113       digits" to get its 53 bits of precision.
2114    
2115    ----------------------------------------------------------------------
2116    Name: Matthias Blume (blume (at) tti - c (dot) org)
2117    Date: 2003/09/17 16:30:00 CDT
2118    Tag: blume-20030917-real32-slices
2119    Description:
2120    
2121    Added an entry to the primitive environment
2122    (compiler/Semant/statenv/prim.sml) for int32->real64 conversion and
2123    added code to compiler/CodeGen/main/mlriscGen.sml to implement it.
2124    
2125    Removed some of the "magic" constants in real64.sml and replaced them
2126    with code that generates these values from their corresponding
2127    integer counterparts.
2128    
2129    Made all(?) the slice-related changes to the Basis and made everything
2130    compile again...
2131    
2132    ----------------------------------------------------------------------
2133    Name: Matthias Blume (blume (at) tti - c (dot) org)
2134    Date: 2003/09/15 17:45:00 CDT
2135    Tag: blume-20030915-rbase
2136    Description:
2137    
2138    Fixed bug in Real.fromLargeInt.
2139    
2140    ----------------------------------------------------------------------
2141    Name: Matthias Blume (blume (at) tti - c (dot) org)
2142    Date: 2003/09/13 18:11:00 CDT
2143    Tag: blume-20030913-libinstall
2144    Description:
2145    
2146    Minor bugfix in config/libinstall (set anchor with path to
2147    standalone tool after installing it, otherwise libraries that
2148    need ml-lex or ml-yacc won't compile the first time the installer
2149    runs).
2150    
2151    ----------------------------------------------------------------------
2152    Name: Matthias Blume (blume (at) tti - c (dot) org)
2153    Date: 2003/09/12 11:45:00 CDT
2154    Tag: blume-20030912-various
2155    Description:
2156    
2157     - fixed bug in Real.toLargeInt
2158     - fixed bug in Posix.ProcEnv.times
2159     - changed inputLine functions to return an option
2160     - minor installer improvements / bugfixes
2161     - changed default @SMLalloc parameter for x86/celeron to 64k
2162    
2163    ----------------------------------------------------------------------
2164    Name: Matthias Blume (blume (at) tti - c (dot) org)
2165    Date: 2003/09/09 22:00:00 CDT
2166    Tag: Release_110_43
2167    Description:
2168    
2169        New working release 110.43.  New bootfiles.
2170    
2171    ----------------------------------------------------------------------
2172    Name: Matthias Blume (blume (at) tti - c (dot) org)
2173    Date: 2003/09/09 19:20:00 CDT
2174    Tag: blume-20030909-installer
2175    Description:
2176    
2177    Rewrote large parts of config/install.sh in SML (config/libinstall.sml).
2178    Modified config/install.bat to take advantage of it.  Also modified
2179    config/install.sh (and called it config/new-install.sh) to take advantage
2180    of it on Unix systems.  (The SML code is (supposed to be) platform-
2181    independent.)
2182    
2183    The installer can now install everything under Win32
2184    as well as under *nix as long as it compiles.
2185    
2186    Other changes:
2187    
2188      - made CML compile again under Win32
2189      - made eXene compile under Win32 (by providing a fake structure UnixSock
2190        and by using OS.Process.getEnv instead of Posix.ProcEnv.getenv)
2191      - fixed a bug in nowhere: it assumed that type OS.Process.status is the
2192        same as type int;  under Win32 it isn't
2193      - fixed some slice-related problems in the win32-specific parts of CML
2194      - added a functor argument "sameVol" to os-path-fn.sml in the Basis
2195        (under Win32, the volume name is case-insensitive, and the
2196        OS.Path code compares volume names for equality)
2197    
2198    ----------------------------------------------------------------------
2199    Name: Matthias Blume (blume (at) tti - c (dot) org)
2200    Date: 2003/09/08 11:55:00 CDT
2201    Tag: blume-20030908-fullpath
2202    Description:
2203    
2204    Made Win32 version of OS.FileSys.fullPath return current directory
2205    when given an empty string.  This is what the spec says, and incidentally,
2206    CM depends on it.  (CM otherwise goes into an infinite loop in certain
2207    cases when presented with the name of a non-existing .cm file.)
2208    
2209    ----------------------------------------------------------------------
2210    Name: Matthias Blume (blume (at) tti - c (dot) org)
2211    Date: 2003/09/04 16:30:00 CDT
2212    Tag: blume-20030905-slices-etc
2213    Description:
2214    
2215      1. Changed interface to vectors and arrays in Basis to match
2216         (draft) Basis spec.
2217      2. Added signatures and implementations of slices according to
2218         Basis spec.
2219      3. Edited source code throughout the system to make it compile again
2220         under 1. and 2.  (In some cases code had to be added to have it
2221         match the new signatures.)
2222      4. MLRISC should be backward-compatible: the copies of the originals
2223         of files that needed to change under 3. were retained, the .cm files
2224         check the compiler version number and use old versions when
2225         appropriate.
2226      5. Changed type of OS.FileSys.readDir and Posix.FileSys.readdir to
2227         dirstream -> string option (in accordance with Basis spec).
2228      6. When generating code that counts lines, ml-lex used function
2229         CharVector.foldli, taking advantage of its old interface.
2230         This has been replaced with the corresponding code from
2231         CharVectorSlice.  (html-lex must be re-lexed!)
2232      7. BitArray in smlnj-lib/Util has been extended/modified to match the
2233         new MONO_ARRAY signature.  (Do we need BitArraySlice?)
2234      8. Removed temporary additions (fromInternal, toInternal) from the
2235         (now obsolete) IntInf in smlnj-lib/Util.
2236      9. Cleaned up structure Byte.
2237     10. Added localOffset, scan, and fromString to Date (according to spec).
2238         Cleaned/corrected implementation of Date.
2239         (Still need to check for correctness; implement better canonicalizeDate.)
2240     11. Added "scan" to signature IEEE_REAL.
2241     12. Some improvements to IntInf [in particular: efficiency-hack for
2242         mod and rem when second operand is 2 (for parity checks).]
2243     13. Changed representation of type Time.time, using a single IntInf.int
2244         value counting microseconds.  This considerably simplified the
2245         implementation of structure Time.  We now support negative time
2246         values; scan and fromString handle signs.
2247     14. Functor PrimIO now takes two additional arguments (VectorSlice and
2248         ArraySlice).
2249    
2250    ----------------------------------------------------------------------
2251    Name: Matthias Blume (blume (at) tti - c (dot) org)
2252    Date: 2003/08/28 17:00:00 CDT
2253    Tag: blume-20030828-intinf
2254    Description:
2255    
2256    This is a major update which comes with a version number bump
2257    (110.42.99 -- yes, we are really close to 110.43 :-), NEW BOOTFILES,
2258    and an implementation of IntInf in the Basis.
2259    
2260    There are a fairly large number of related changes and updates throughout
2261    the system:
2262    
2263    Basis:
2264            - Implemented IntInf.
2265            - Made LargeInt a projection of IntInf (by filtering through INTEGER).
2266            - Added some missing Real64 operations, most notably Real.toLargeInt.
2267            - Added FixedInt as a synonym for Int32.
2268    
2269    compiler:
2270            * Added support for a built-in intinf type.
2271              - literals
2272              - pattern matching
2273              - conversion shortcuts (Int32.fromLarge o Int.toLarge etc.)
2274              - overloading on literals and operations
2275    
2276            This required adding a primitive type intinf, some additional
2277            primops, and implementations for several non-trivial intinf
2278            operations in Core.  (The intinf type is completely abstract
2279            to the compiler; all operations get delegated back to the Core.)
2280    
2281            * Intinf equality is handled by polyequal.  However, the compiler
2282            does not print its usual warning in this case (since polyequal
2283            is the right thing to do there).
2284    
2285            * Improved the organization of structure InlineT.
2286    
2287            * A word about conversion primops:
2288            If conversions involving intinf do not cancel out during
2289            CPS contract, then the compiler must insert calls to Core functions.
2290            Since all core access must be resolved already during the FLINT
2291            translate phase, it would be too late a the time of CPS contract
2292            to add new Core calls.  For this reason, conversion primops
2293            for intinf carry two arguments:  1. the numeric argument that
2294            they are supposed to convert, and 2. the Core function that
2295            can help with this conversion if necessary.  If CPS contract
2296            eliminates a primop, then the associated Core function becomes
2297            dead and goes away.  Intinf conversion primops that do not get
2298            eliminated by CPS contract get rewritten into calls of their
2299            core functions by a separate, new phase.
2300    
2301    interactive system:
2302            - Control.Print.intinfDepth controls max length of intinf constants
2303            being printed. (Analogous to Control.Print.stringDepth.)
2304            - Cleanup in printutil and pputil: got rid of unused stuff and
2305            duplicates; replaced some of the code with code that makes better
2306            use of library functionality.
2307    
2308    CM:
2309            Bugfix: parse-errors in init group (system/smlnj/init/init.cmi)
2310            are no longer silent.
2311    
2312    CKIT:
2313            Fixed mismatched uses of Int32 and LargeInt.  I always decided
2314            in favor of LargeInt -- which is now the same as IntInf.
2315            CKIT-knowledgable people should check whether this is what's
2316            intended and otherwise change things back to using Int32 or
2317            FixedInt.
2318    
2319    Throughout the code:
2320            Started using IntInf.int literals and built-in operations
2321            (e.g., comparison with 0) where this seems appropriate.
2322    
2323    
2324    ----------------------------------------------------------------------
2325    Name: Dave MacQueen (dbm@cs.uchicago.edu)
2326    Date: 2003/08/13 11:36:00 CDT
2327    Tag: dbm-20030813-mcz-merge1
2328    Description:
2329    
2330    Merging changes from the mcz-branch development branch into trunk.
2331    These changes involve replacement of the emulated old prettyprinter
2332    interface with direct use of the SML/NJ Lib PP library, and fixing
2333    of a couple of bugs (895, 1186) relating to error messages.  A new
2334    prettyprinter for ast datatypes (Elaborator/print/ppast.{sig,sml})
2335    has been added.
2336    
2337    ----------------------------------------------------------------------
2338    Name: Matthias Blume (blume (at) tti - c (dot) org)
2339    Date: 2003/08/11 15:45:00 CDT
2340    Tag: blume-20030811-windows
2341    Description:
2342    
2343    Version number bumped to 110.42.9.  NEW BOOTFILES!!!
2344    
2345        http://smlnj.cs.uchicago.edu/dist/working/110.42.9/
2346    
2347    This patch restores SML/NJ's ability to run under win32.  There are a
2348    number of changes, including fixes for several bugs that had gone
2349    unnoticed until now:
2350    
2351      - uname "CYGWIN_NT*" is recognized as win32 (This is relevant only when
2352        trying to run the win32 version from within cygwin.)
2353    
2354      - There are a number of simple .bat scripts that substitute for their
2355        corresponding Unix shell-scripts.  (See below.)
2356    
2357      - The internals of ml-build have been modified slightly.  The main
2358        difference is that instead of calling ".link-sml" (or link-sml.bat)
2359        using OS.Process.system, the ML process delegates this task back
2360        to the script.  Otherwise problems arise in mixed environments such
2361        as Cygwin where scripts look and work like Unix scripts, but
2362        where OS.Process.system cannot run them.
2363    
2364      - In CM, the srcpath pickler used native pathname syntax -- which
2365        is incorrect in the case of cross-compilation.  The new pickle format
2366        is independent of platform-specific naming conventions.
2367    
2368      - Path configuration files (such as lib/pathconfig) can now choose
2369        between native and standard syntax.  Placing a line of the form
2370    
2371            standard!
2372    
2373        into the file causes all subsequent paths to be interpreted using
2374        CM standard pathname syntax (= Unix conventions); a line
2375    
2376            native!
2377    
2378        switches back to native style.  This was needed so that
2379        path config files can be written portably, see src/system/pathconfig.
2380    
2381      - Runtime system:
2382    
2383          - win32-filesys.c:  get_file_time and set_file_time now
2384            access modification time, not creation time.
2385    
2386          - I/O code made aware of new array representation.
2387    
2388          - Bug fixes in X86.prim.masm.
2389    
2390       - src/system/makeml made aware of win32. (For use under cygwin
2391            and other Unix-environments for windows.)
2392    
2393       - In Basis, fixed off-by-one error in win32-io.sml (function vecF)
2394         which caused BinIO.inputAll to fail consistently.
2395    
2396    .bat scripts:
2397    
2398       Windows .bat scripts assume that SMLNJ_HOME is defined.
2399    
2400       - sml.bat, ml-yacc.bat, ml-lex.bat:  Driver scripts for standalone
2401         applications (sml, ml-yacc, ml-lex).
2402       - ml-build.bat: analogous to ml-build.
2403       - config\install.bat:  Analogous to config/install.sh.  This requires
2404         that SMLNJ_HOME is set and that Microsoft Visual C is ready to use.
2405         (nmake etc. must be on the path, and vcvars32 must have been run.)
2406         Moreover, sources for ml-lex and ml-yacc need to exist under src,
2407         and the bootfile hierarchy must have been unpacked under
2408         sml.boot.x86-win32.
2409         The script is very primitive and does a poor job at error checking.
2410         It only installs the base system, ml-lex, and ml-yacc.  No other
2411         libraries are being installed (i.e., you get only those that
2412         are part of the compiler.)
2413       - link-sml.bat:  analogous to .link-sml, but not currently used
2414    
2415    Unrelated bug fixes:
2416    
2417      - ml-nlffigen now exports structures ST_* corresponding to incomplete
2418        types.
2419      - Added getDevice to PP/src/pp-debug-fn.sml.  (Would not compile
2420        otherwise.)
2421    
2422    ----------------------------------------------------------------------
2423    Name: Dave MacQueen (macqueen@cs.uchicago.edu)
2424    Date: 2003/06/17
2425    Tag: macqueen-20030617-bug895
2426    Description:
2427    
2428    Modified compiler/Elaborator/print/pptype.sml to fix bug 895.
2429    Tag will be used for new development branch (mcz-branch) for
2430    use by MacQueen, (Lucasz) Zairek, and (George) Cao at uchicago.
2431    
2432    ----------------------------------------------------------------------
2433    Name: Matthias Blume (blume (at) tti - c (dot) org)
2434    Date: 2003/05/27 16:55:00 CDT
2435    Tag: blume-20030527-polyeq
2436    Description:
2437    
2438    Tried to eliminated most cases of polymorphic equality.
2439    
2440    ----------------------------------------------------------------------
2441    Name: Matthias Blume (blume (at) tti - c (dot) org)
2442    Date: 2003/05/21 17:45:00 CDT
2443    Tag: blume-20030517-complete
2444    Description:
2445    
2446    Two changes:
2447    
2448      1. Added a flag for controlling whether non-exhaustive bindings will
2449         be treated as errors (default is false).
2450      2. Cleaned up the *entire* source tree so that CMB.make goes through
2451         without a single non-exhaustive match- or bind warning.
2452    
2453    ----------------------------------------------------------------------
2454    Name: Matthias Blume (blume (at) tti - c (dot) org)
2455    Date: 2003/05/17 10:20:00 CDT
2456    Tag: blume-20030517-absyn
2457    Description:
2458    
2459    1. Added cases for IF, WHILE, ANDALSO, and ORELSE to Absyn.
2460    
2461       This mainly affects the quality of error messages.  However, some
2462       of the code is now more straightforward than before. (Treatment of
2463       the above four constructs in translate.sml is much simpler than
2464       the "macro-expansion" that was going on before.  Plus, the mach-
2465       compiler no longer gets invoked just to be able to compile an
2466       if-expression.)
2467    
2468    2. The ErrorMsg.Error exception is now caught and absorbed by the
2469       interactive loop.
2470    
2471    ----------------------------------------------------------------------
2472    Name: Allen Leung
2473    Date: 2003/05/16 13:05:00 CDT
2474    Tag: leunga-20030516-cygwin-runtime
2475    Description:
2476    
2477        Ported the runtime system to cygwin, which uses the unix
2478    x86-unix bin files.  Missing/buggy features:
2479    
2480       o getnetbyname, getnetbyaddr: these functions seem to be missing in
2481         the Cygwin library.
2482       o Ctrl-C handling may be flaky.
2483       o Windows system calls and Windows I/O are not supported.
2484    
2485       A new set of binfiles is located at:
2486    
2487         http://www.dorsai.org/~leunga/boot.x86-unix.tgz
2488    
2489       This is only needed for bootstrapping the cygwin version of smlnj.
2490       Other x86 versions can use the existing binfiles.
2491    
2492    ----------------------------------------------------------------------
2493    Name: Matthias Blume
2494    Date: 2003/04/08 15:42:00 CDT
2495    Tag: blume-20030408-listpair
2496    Description:
2497    
2498    1. Added a target 'mlrisc' to installer.
2499    
2500    2. Added missing elements to structure ListPair.
2501    
2502    ----------------------------------------------------------------------
2503    Name: Allen Leung
2504    Date: 2003/01/07 10:40:00 EST
2505    Tag: leunga-20030107-int-rem
2506    Description:
2507    
2508    Fixed a bug in Int.rem(x,y) where y is a power of 2 on x86.
2509    The arguments to the SUBL instruction were swapped.
2510    
2511    ----------------------------------------------------------------------
2512    Name: Matthias Blume
2513    Date: 2002/12/12 16:25:00 EST
2514    Tag: blume-20021212-risc-ra
2515    Description:
2516    
2517    Fixed a serious bug in the rewrite code for FP spilling/reloading that
2518    sent the RA into an infinite loop when floating point registers get
2519    spilled.  (Because of this bug, e.g., nucleic stopped compiling between
2520    110.37 and 110.38.)
2521    There was another set of potential problems related to the handling of
2522    MLRISC annotations (but those did not yet cause real problems, apparently).
2523    
2524    ----------------------------------------------------------------------
2525    Name: Matthias Blume
2526    Date: 2002/12/06 22:40:00 EST
2527    Tag: blume-20021206-cm-fileid
2528    Description:
2529    
2530    Added a call of SrcPath.sync at the beginning of Parse.parse (in CM).
2531    This fixes the problem of CM getting confused by files that suddenly
2532    change their identity (e.g., by getting unlinked and recreated by some
2533    text editor such as vi).  There might be a better/cheaper/cleaner way
2534    of doing this, but for now this will have to do.
2535    
2536    ----------------------------------------------------------------------
2537    Name: Matthias Blume
2538    Date: 2002/10/28 09:50:00 EST
2539    Tag: blume-20021028-typecheck
2540    Description:
2541    
2542    Exported structure Typecheck from $smlnj/viscomp/core.cm.
2543    
2544    ----------------------------------------------------------------------
2545    Name: Matthias Blume
2546    Date: 2002/10/17 09:10:00 EDT
2547    Tag: Release_110_42
2548    Description:
2549    
2550    In good old tradition, there has been a slight hiccup so that we have
2551    to patch 110.42 after the fact.   The old release tag has been replaced
2552    (see below).
2553    
2554    The change solves a problem with two competing approaches the
2555    configuration problem regarding MacOS 10.1 vs. MacOS 10.2 which got in
2556    each other's way.
2557    
2558    This change only affects the runtime system code and the installer script.
2559    (No new bootfiles.)
2560    
2561    ----------------------------------------------------------------------
2562    Name: Matthias Blume
2563    Date: 2002/10/16 12:00:00 EDT
2564    Tag: Release_110_42_removed
2565    Description:
2566    
2567    New working release.  New bootfiles.
2568    
2569    ----------------------------------------------------------------------
2570    Name: Matthias Blume
2571    Date: 2002/10/10 13:10:00 EDT
2572    Tag: blume-20021010-ppc-divs
2573    Description:
2574    
2575    The mltree operator DIVS must be implemented with an overflow check on
2576    the PPC because the hardware indicates divide-by-zero using "overflow" as
2577    well.
2578    
2579    ----------------------------------------------------------------------
2580    Name: Matthias Blume
2581    Date: 2002/07/23 11:45:00
2582    Tag: blume-20020723-smlnj-home
2583    Description:
2584    
2585    Sml now senses the SMLNJ_HOME environment variable.  If this is set,
2586    then the bin dir is assumed to be in $SMLNJ_HOME/bin and (unless
2587    CM_PATHCONFIG is also set), the path configuration file is assumed
2588    to be in $SMLNJ_HOME/lib/pathconfig.  This way one can easily move
2589    the entire tree to some other place and everything will "just work".
2590    
2591    (Companion commands such as ml-build and ml-makedepend also sense this
2592    variable.)
2593    
2594    ----------------------------------------------------------------------
2595    Name: Matthias Blume
2596    Date: 2002/07/12 21:19:00 EDT
2597    Tag: blume-20020712-liveness
2598    Description:
2599    
2600    Exported two useful "step" functions from liveness module (MLRISC).
2601    
2602    ----------------------------------------------------------------------
2603    Name: Matthias Blume
2604    Date: 2002/07/05 16:00 EDT
2605    Tag: Release_110_41
2606    Description:
2607    
2608    New working release.  New bootfiles.
2609    
2610    ----------------------------------------------------------------------
2611    Name: Matthias Blume
2612    Date: 2002/07/05 10:25:00 EDT
2613    Tag: blume-20020705-btimp
2614    Description:
2615    
2616    Exported structure BTImp from $smlnj/viscomp/debugprof.cm so that
2617    other clients can set up backtracing support.
2618    
2619    ----------------------------------------------------------------------
2620    Name: Matthias Blume
2621    Date: 2002/06/25 17:23:00 EDT
2622    Tag: blume-20020625-fpmax
2623    Description:
2624    
2625    Fixed a bug in translation of INLMAX (and INLMIN) for the floating-point
2626    case.  (The sense of the isNaN test was reversed -- which made min and
2627    max always return their first argument.)
2628    
2629    ----------------------------------------------------------------------
2630    Name: Matthias Blume
2631    Date: 2002/06/11
2632    Tag: blume-20020611-unixpath
2633    Description:
2634    
2635    Back-ported OS.Path.{from,to}UnixPath from idlbasis-devel branch.
2636    
2637    ----------------------------------------------------------------------
2638    Name: Matthias Blume
2639    Date: 2002/06/10 16:35:00 EDT
2640    Tag: blume-20020610-ieeereal
2641    Description:
2642    
2643    I back-ported my implementation of IEEEReal.fromString from the
2644    idlbasis-devel branch so that we can test it.
2645    
2646    Another small change is that ppDec tries to give more information
2647    than just "<sig>" in the case of functors.  However, this code is
2648    broken in some mysterious way if the functor's body's signature
2649    has not been declared by ascription but gets inferred from the
2650    implementation.  This needs fixing...
2651    
2652    ----------------------------------------------------------------------
2653    Name: Matthias Blume
2654    Date: 2002/05/31
2655    Tag: blume-20020531-btrace-mode
2656    Description:
2657    
2658    Resurrected SMLofNJ.Internals.BTrace.mode.  (It accidentally fell by
2659    the wayside when I switched over to using Controls everywhere.)
2660    
2661    ----------------------------------------------------------------------
2662    Name: Lal George
2663    Date: 2002/05/23 12:21:40 EDT
2664    Tag: george-20020523-visual-labels
2665    Description:
2666    
2667    Labels are now displayed in the graphical output to make
2668    the fall-through and target blocks obvious.
2669    
2670    ----------------------------------------------------------------------
2671    Name: Matthias Blume
2672    Date: 2002/05/22 11:03:00 EDT
2673    Tag: blume-20020522-shrink
2674    Description:
2675    
2676    John tweaked yesterday's fix for 1131 to handle an out-of-memory
2677    situation that comes up when allocating huge arrays.
2678    
2679    ----------------------------------------------------------------------
2680    Name: Matthias Blume
2681    Date: 2002/05/21 16:00:00 EDT
2682    Tag: Release_110_40
2683    Description:
2684    
2685    New working release (110.40).  New bootfiles.
2686    
2687    [Also: John Reppy fixed GC bug 1131.]
2688    
2689    ----------------------------------------------------------------------
2690    Name: Matthias Blume
2691    Date: 2002/05/21 12:35:00 EDT
2692    Tag: blume-20020521-cmdoc
2693    Description:
2694    
2695    CM documentation update.
2696    
2697    ----------------------------------------------------------------------
2698    Name: Matthias Blume
2699    Date: 2002/05/21 10:55:00 EDT
2700    Tag: blume-20020521-misc
2701    Description:
2702    
2703    - John tweaked runtime to be silent on heap export (except when
2704      GC messages are on).
2705    - I added a few more things (cross-compiling versions of CMB) to
2706      config/preloads (as suggestions).
2707    
2708    ----------------------------------------------------------------------
2709    Name: Matthias Blume
2710    Date: 2002/05/20 22:25:00 EDT
2711    Tag: blume-20020520-controls
2712    Description:
2713    
2714    - Added ControlUtil structure to control-lib.cm.
2715    - Use it throughout.
2716    - Used Controls facility to define MLRISC controls (as opposed to
2717      registering MLRISC control ref cells with Controls after the fact)
2718    - Fixed messed-up controls priorities.
2719    
2720    * Removed again all the stuff from config/preloads that one wouldn't
2721      be able to preload at the time the initial heap image is built.
2722      (Many libraries, e.g., CML, do not exist yet at this time.  The
2723       only libraries that can be preloaded via config/preloads are those
2724       that come bundled with the bootfiles.)
2725    
2726    ----------------------------------------------------------------------
2727    Name: Matthias Blume
2728    Date: 2002/05/20 10:59:00 EDT
2729    Tag: blume-20020520-preloads
2730    Description:
2731    
2732    Added a lot of commented-out suggestions for things to be included
2733    in config/preloads.
2734    
2735    ----------------------------------------------------------------------
2736    Name: Allen Leung
2737    Date: 2002/05/18 14:20:00 EDT
2738    Tag: leunga-20020518-mdl
2739    Description:
2740    
2741    o Made the mdl tool stuff compile and run again.
2742    o I've disabled all the stuff that depends on RTL specifications; they
2743      are all badly broken anyway.
2744    
2745    ----------------------------------------------------------------------
2746    Name: Matthias Blume
2747    Date: 2002/05/17 16:49:00 EDT
2748    Tag: blume-20020517-controls
2749    Description:
2750    
2751    0. John Reppy made several modifications to the SML/NJ library.
2752       In particular, there is a shiny new controls-lib.cm.
2753    
2754    1. Pushed new controls interface through compiler so that everything
2755       compiles again.
2756    
2757    2. Added FormatComb and FORMAT_COMB to the CML version of the
2758       SML/NJ library (so that CML compiles again).
2759    
2760    3. Modified init scripts because XXX_DEFAULT environment variables
2761       are no longer with us.  (Boot-time initialization is now done
2762       using the same environment variables that are also used for
2763       startup-time initialization of controls.)
2764    
2765    ----------------------------------------------------------------------
2766    Name: Lal George
2767    Date: 2002/05/15 09:20:10 EDT
2768    Tag: george-20020515-pseudo-op-decls
2769    Description:
2770    
2771            All pseudo-ops emitted before the first segment declaration
2772    such as TEXT, DATA, and BSS directives are assumed to be global
2773    declarations and are emitted first in the assembly file. This is
2774    useful in a number of situations where one has pseudo-ops that are not
2775    specific to any segment, and also works around the constraint that one
2776    cannot have client pseudo-ops in the TEXT segment.
2777    
2778            Because no segment is associated with these declarations it is
2779    an error to allocate any space or objects before the first segment
2780    directive and an exception will be raised. However, we cannot make
2781    this check for client pseudo-ops.
2782    
2783            These top level declarations are a field in the CFG graph_info.
2784    In theory you can continue to add to this field after the CFG has been
2785    built -- provided you know what you are doing;-)
2786    
2787    ----------------------------------------------------------------------
2788    Name: Matthias Blume
2789    Date: 2002/05/13 16:40:00 EDT
2790    Tag: blume-20020513-pp-etc
2791    Description:
2792    
2793    A few minor bugfixes:
2794    
2795      - Stopgap measure for bug recently reported by Elsa Gunter (ppDec).
2796        (Bogus printouts for redefined bindings still occur.  Compiler
2797        bug should no longer occur now.  We need to redo the prettyprinter
2798        from scratch.)
2799    
2800      - CM pathname printer now also adds escape sequences for ( and )
2801    
2802      - commend and docu fixes for ml-nlffi
2803    
2804    ----------------------------------------------------------------------
2805    Name: Matthias Blume
2806    Date: 2002/05/10 16:40:00 EDT
2807    Tag: blume-20020510-erg-textio
2808    Description:
2809    
2810    Applied the following bugfix provided by Emden Gansner:
2811    
2812        Output is corrupted when outputSubstr is used rather than output.
2813    
2814        The problem occurs when a substring
2815    
2816            ss = (s, dataStart, dataLen)
2817    
2818        where dataStart > 0, fills a stream buffer with avail bytes left.
2819        avail bytes of s, starting at index dataStart, are copied into the
2820        buffer, the buffer is flushed, and then the remaining dataLen-avail
2821        bytes of ss are copied into the beginning of the buffer. Instead of
2822        starting this copy at index dataStart+avail in s, the current code
2823        starts the copy at index avail.
2824    
2825        Fix:
2826        In text-io-fn.sml, change line 695 from
2827             val needsFlush = copyVec(v, avail, dataLen-avail, buf, 0)
2828        to
2829             val needsFlush = copyVec(v, dataStart+avail, dataLen-avail, buf, 0)
2830    
2831    ----------------------------------------------------------------------
2832    Name: Matthias Blume
2833    Date: 2002/04/12 13:55:00 EDT
2834    Tag: blume-20020412-assyntax
2835    Description:
2836    
2837    1. Grabbed newer assyntax.h from the XFree86 project.
2838    2. Fiddled with how to compile X86.prim.asm without warnings.
2839    3. (Very) Minor cleanup in CM.
2840    
2841    ----------------------------------------------------------------------
2842    Name: Matthias Blume
2843    Date: 2002/04/01 (no joke!) 17:07:00 EST
2844    Tag: blume-20020401-x86div
2845    Description:
2846    
2847    Added full support for div/mod/rem/quot on the x86, using the machine
2848    instruction's two results (without clumsily recomputing the remainder)
2849    directly where appropriate.
2850    
2851    Some more extensive power-of-two support was added to the x86 instruction
2852    selector (avoiding expensive divs, mods, and muls where they can be
2853    replaced with cheaper shifts and masks).  However, this sort of thing
2854    ought to be done earlier, e.g., within the CPS optimizer so that
2855    all architectures benefit from it.
2856    
2857    The compiler compiles to a fixed point, but changes might be somewhat
2858    fragile nevertheless.  Please, report any strange things that you might
2859    see wrt. div/mod/quot/rem...
2860    
2861    ----------------------------------------------------------------------
2862    Name: Matthias Blume
2863    Date: 2002/03/29 17:22:00
2864    Tag: blume-20020329-div
2865    Description:
2866    
2867    Fixed my broken div/mod logic.  Unfortunately, this means that the
2868    inline code for div/mod now has one more comparison than before.
2869    Fast paths (quotient > 0 or remainder = 0) are not affected, though.
2870    The problem was with quotient = 0, because that alone does not tell
2871    us which way the rounding went.  One then has to look at whether
2872    remainder and divisor have the same sign...  :(
2873    
2874    Anyway, I replaced the bootfiles with fresh ones...
2875    
2876    ----------------------------------------------------------------------
2877    Name: Matthias Blume
2878    Date: 2002/03/29 14:10:00 EST
2879    Tag: blume-20020329-inlprims
2880    Description:
2881    
2882    NEW BOOTFILES!!!    Version number bumped to 110.39.3.
2883    
2884    Primops have changed. This means that the bin/boot-file formats have
2885    changed as well.
2886    
2887    To make sure that there is no confusion, I made a new version.
2888    
2889    
2890    CHANGES:
2891    
2892    * removed REMT from mltree (remainder should never overflow).
2893    
2894    * added primops to deal with divisions of all flavors to the frontend
2895    
2896    * handled these primops all the way through so they map to their respective
2897      MLRISC support
2898    
2899    * used these primops in the implementation of Int, Int32, Word, Word32
2900    
2901    * removed INLDIV, INLMOD, and INLREM as they are no longer necessary
2902    
2903    * parameterized INLMIN, INLMAX, and INLABS by a numkind
2904    
2905    * translate.sml now deals with all flavors of INL{MIN,MAX,ABS}, including
2906      floating point
2907    
2908    * used INL{MIN,MAX,ABS} in the implementation of Int, Int32, Word, Word32,
2909      and Real (but Real.abs maps to a separate floating-point-only primop)
2910    
2911    
2912    TODO items:
2913    
2914    * Hacked Alpha32 instruction selection, disabling the selection of REMx
2915      instructions because the machine instruction encoder cannot handle
2916      them.  (Hppa, PPC, and Sparc instruction selection did not handle
2917      REM in the first place, and REM is supported by the x86 machine coder.)
2918    
2919    * Handle DIV and MOD with DIV_TO_NEGINF directly in the x86 instruction
2920      selection phase.  (The two can be streamlined because the hardware
2921      delivers both quotient and remainder at the same time anyway.)
2922    
2923    * Think about what to do with "valOf(Int32.minInt) div ~1" and friends.
2924      (Currently the behavior is inconsistent both across architectures and
2925      wrt. the draft Basis spec.)
2926    
2927    * Word8 should eventually be handled natively, too.
2928    
2929    * There seems to be one serious bug in mltree-gen.sml.  It appears, though,
2930      as if there currently is no execution path that could trigger it in
2931      SML/NJ.  (The assumptions underlying functions arith and promotable do not
2932      hold for things like multiplication and division.)
2933    
2934    ----------------------------------------------------------------------
2935    Name: Matthias Blume
2936    Date: 2002/03/27 16:27:00 EST
2937    Tag: blume-20020327-mlrisc-divisions
2938    Description:
2939    
2940    Added support for all four division operations (ML's div, mod, quot,
2941    and rem) to MLRISC.  In the course of doing so, I also rationalized
2942    the naming (no more annoying switch-around of DIV and QUOT), by
2943    parameterizing the operation by div_rounding_mode (which can be either
2944    DIV_TO_ZERO or DIV_TO_NEGINF).
2945    
2946    The generic MLTreeGen functor takes care of compiling all four
2947    operations down to only round-to-zero div.
2948    
2949    Missing pieces:
2950    
2951      * Doing something smarter than relying on MLTreeGen on architectures
2952        like, e.g., the x86 where hardware division delivers both quotient and
2953        remainder at the same time.  With this, the implementation of the
2954        round-to-neginf operations could be further streamlined.
2955    
2956      * Remove inlining support for div/mod/rem from the frontend and replace it
2957        with primops that get carried through to the backend.  Do this for all
2958        int and word types.
2959    
2960    ----------------------------------------------------------------------
2961    Name: Matthias Blume
2962    Date: 2002/03/25 17:25:00 EST
2963    Tag: blume-20020325-divmod
2964    Description:
2965    
2966    I improved (hopefully without breaking them) the implementation of Int.div,
2967    Int.mod, and Int.rem.   For this, the code in translate.sml now takes
2968    advantage of the following observations:
2969    
2970      Let  q = x quot y      r = x rem y
2971           d = x div  y      m = x mod y
2972    
2973    where "quot" is the round-to-zero version of integer division that
2974    hardware usually provides.  Then we have:
2975    
2976         r = x - q * y        where neither the * nor the - will overflow
2977         d = if q >= 0 orelse x = q * y then q else q - 1
2978                              where neither the * nor the - will overflow
2979         m = if q >= 0 orelse r = 0 then r else r + y
2980                              where the + will not overflow
2981    
2982    This results in substantial simplification of the generated code.
2983    The following table shows the number of CFG nodes and edges generated
2984    for
2985            fun f (x, y) = x OPER y
2986            (* with OPER \in div, mod, quot, rem *)
2987    
2988    
2989        OPER | nodes(old) | edges(old) | nodes(new) | edges(new)
2990        --------------------------------------------------------
2991         div |         24 |         39 |         12 |         16
2992         mod |         41 |         71 |         12 |         16
2993        quot |          8 |         10 |          8 |         10
2994         rem |         10 |         14 |          8 |         10
2995    
2996    
2997    ----------------------------------------------------------------------
2998    Name: Matthias Blume
2999    Date: 2002/03/25 22:06:00 EST
3000    Tag: blume-20020325-cprotobug
3001    Description:
3002    
3003    Fixed a bug in cproto (c prototype decoder).
3004    
3005    ----------------------------------------------------------------------
3006    Name: Matthias Blume
3007    Date: 2002/03/25 16:00:00 EST
3008    Tag: blume-20020325-raw-primops
3009    Description:
3010    
3011    I did some cleanup to Allen's new primop code and
3012    replaced yesterday's bootfiles with new ones.
3013    (But they are stored in the same place.)
3014    
3015    ----------------------------------------------------------------------
3016    Name: Matthias Blume
3017    Date: 2002/03/24 22:40:00 EST
3018    Tag: blume-20020324-bootfiles
3019    Description:
3020    
3021    Made the bootfiles that Allen asked for.
3022    
3023    ----------------------------------------------------------------------
3024    Name: Allen Leung
3025    Date: 2002/03/23 15:50:00 EST
3026    Tag: leunga-20020323-flint-cps-rcc-primops
3027    Description:
3028    
3029      1. Changes to FLINT primops:
3030    
3031        (* make a call to a C-function;
3032         * The primop carries C function prototype information and specifies
3033         * which of its (ML-) arguments are floating point. C prototype
3034         * information is for use by the backend, ML information is for
3035         * use by the CPS converter. *)
3036      | RAW_CCALL of { c_proto: CTypes.c_proto,
3037                       ml_args: ccall_type list,
3038                       ml_res_opt: ccall_type option,
3039                       reentrant : bool
3040                     } option
3041       (* Allocate uninitialized storage on the heap.
3042        * The record is meant to hold short-lived C objects, i.e., they
3043        * are not ML pointers.  With the tag, the representation is
3044        * the same as RECORD with tag tag_raw32 (sz=4), or tag_fblock (sz=8)
3045        *)
3046      | RAW_RECORD of {tag:bool,sz:int}
3047      and ccall_type = CCALL_INT32 | CCALL_REAL64 | CCALL_ML_PTR
3048    
3049      2.  These CPS primops are now overloaded:
3050    
3051           rawload of {kind:numkind}
3052           rawstore of {kind:numkind}
3053    
3054          The one argument form is:
3055    
3056             rawload {kind} address
3057    
3058          The two argument form is:
3059    
3060             rawload {kind} [ml object, byte-offset]
3061    
3062      3. RAW_CCALL/RCC now takes two extra arguments:
3063    
3064         a. The first is whether the C call is reentrant, i.e., whether
3065            ML state should be saved and restored.
3066         b. The second argument is a string argument specifying the name of
3067            library and the C function.
3068    
3069         These things are currently not handled in the code generator, yet.
3070    
3071      4. In CProto,
3072    
3073         An encoding type of "bool" means "ml object" and is mapped into
3074         C prototype of PTR.  Note that "bool" is different than "string",
3075         even though "string" is also mapped into PTR, because "bool"
3076         is assigned an CPS type of BOGt, while "string" is assigned INT32t.
3077    
3078      5. Pickler/unpicker
3079    
3080         Changed to handle RAW_RECORD and newest RAW_CCALL
3081    
3082      6. MLRiscGen,
3083    
3084         1. Changed to handle the new rawload/rawstore/rawrecord operators.
3085         2. Code for handling C Calls has been moved to a new module CPSCCalls,
3086            in the file CodeGen/cpscompile/cps-c-calls.sml
3087    
3088      7. Added the conditional move operator
3089    
3090             condmove of branch
3091    
3092         to cps.  Generation of this is still buggy so it is currently
3093         disabled.
3094    
3095    ----------------------------------------------------------------------
3096    Name: Lal George
3097    Date: 2002/03/22 14:18:25 EST
3098    Tag: george-20020322-cps-branch-prob
3099    Description:
3100    
3101    Implemented the Ball-Larus branch prediction-heuristics, and
3102    incorporated graphical viewers for control flow graphs.
3103    
3104    Ball-Larus Heuristics:
3105    ---------------------
3106    See the file compiler/CodeGen/cpscompile/cpsBranchProb.sml.
3107    
3108    By design it uses the Dempster-Shafer theory for combining
3109    probabilities.  For example, in the function:
3110    
3111        fun f(n,acc) = if n = 0 then acc else f(n-1, n*acc)
3112    
3113    the ball-larus heuristics predicts that the n=0 is unlikely
3114    (OH-heuristic), and the 'then' branch is unlikely because of the
3115    RH-heuristic -- giving the 'then' branch an even lower combined
3116    probability using the Dempster-Shafer theory.
3117    
3118    Finally, John Reppy's loop analysis in MLRISC, further lowers the
3119    probability of the 'then' branch because of the loop in the else
3120    branch.
3121    
3122    
3123    Graphical Viewing:
3124    ------------------
3125    I merely plugged in Allen's graphical viewers into the compiler. The
3126    additional code is not much. At the top level, saying:
3127    
3128            Control.MLRISC.getFlag "cfg-graphical-view" := true;
3129    
3130    will display the graphical view of the control flow graph just before
3131    back-patching.  daVinci must be in your path for this to work. If
3132    daVinci is not available, then the default viewer can be changed
3133    using:
3134    
3135            Control.MLRISC.getString "viewer"
3136    
3137    which can be set to "dot" or "vcg" for the corresponding viewers. Of
3138    course, these viewers must be in your path.
3139    
3140    The above will display the compilation unit at the level of clusters,
3141    many of which are small, boring, and un-interesting. Also setting:
3142    
3143            Control.MLRISC.getInt "cfg-graphical-view_size"
3144    
3145    will display clusters that are larger than the value set by the above.
3146    
3147    
3148    ----------------------------------------------------------------------
3149    Name: Matthias Blume
3150    Date: 2002/03/21 22:20:00 EST
3151    Tag: blume-20020321-kmp-bugfix
3152    Description:
3153    
3154    Changed the interface to the KMP routine in PreString and fixed
3155    a minor bug in one place where it was used.
3156    
3157    ----------------------------------------------------------------------
3158    Name: Allen Leung
3159    Date: 2002/03/21 20:30:00 EST
3160    Tag: leunga-20020321-cfg
3161    Description:
3162    
3163      Fixed a potential problem in cfg edge splitting.
3164    
3165    ----------------------------------------------------------------------
3166    Name: Allen Leung
3167    Date: 2002/03/21 17:15:00 EST
3168    Tag: leunga-20020321-x86-fp-cfg
3169    Description:
3170    
3171      1. Recoded the buggy parts of x86-fp.
3172    
3173         a. All the block reordering code has been removed.
3174            We now depend on the block placement phases to do this work.
3175    
3176         b. Critical edge splitting code has been simplified and moved into the
3177            CFG modules, as where they belong.
3178    
3179         Both of these were quite buggy and complex.  The code is now much, much
3180         simpler.
3181    
3182      2. X86 backend.
3183    
3184         a. Added instructions for 64-bit support.  Instruction selection for
3185            64-bit has not been committed, however, since that
3186            requires changes to MLTREE which haven't been approved by
3187            Lal and John.
3188    
3189         b. Added support for FUCOMI and FUCOMIP when generating code for
3190            PentiumPro and above.  We only generate these instructions in
3191            the fast-fp mode.
3192    
3193         c. Added cases for JP and JNP in X86FreqProps.
3194    
3195      3. CFG
3196    
3197         CFG now has a bunch of methods for edge splitting and merging.
3198    
3199      4. Machine description.
3200    
3201         John's simplification of MLTREE_BASIS.fcond broke a few machine
3202         description things:
3203    
3204         rtl-build.{sig,sml} and hppa.mdl fixed.
3205    
3206         NOTE: the machine description stuff in the repository is still broken.
3207               Again, I can't put my fixes in because that involves
3208               changes to MLTREE.
3209    
3210    ----------------------------------------------------------------------
3211    Name: Matthias Blume
3212    Date: 2002/03/20 15:55:00 EST
3213    Tag: blume-20020320-kmp
3214    Description:
3215    
3216    Implemented Knuth-Morris-Pratt string matching in PreString and used
3217    it for String.isSubstring, Substring.isSubstring, and
3218    Substring.position.
3219    
3220    (Might need some stress-testing.  Simple examples worked fine.)
3221    
3222    ----------------------------------------------------------------------
3223    Name: Matthias Blume
3224    Date: 2002/03/19 16:37:00 EST
3225    Tag: blume-20020319-witnesses
3226    Description:
3227    
3228    Added a structure C.W and functions convert/Ptr.convert to ml-nlffi-lib.
3229    
3230    This implements a generic mechanism for changing constness qualifiers
3231    anywhere within big C types without resorting to outright "casts".
3232    (So far, functions such as C.rw/C.ro or C.Ptr.rw/C.Ptr.ro only let you
3233    modify the constness at the outermost level.)
3234    The implementation of "convert" is based on the idea of "witness"
3235    values -- values that are not used by the operation but whose types
3236    "testify" to their applicability.  On the implementation side, "convert"
3237    is simply a projection (returning its second curried argument).  With
3238    cross-module inlining, it should not result in any machine code being
3239    generated.
3240    
3241    ----------------------------------------------------------------------
3242    Name: Matthias Blume
3243    Date: 2002/03/15 16:40:00 EST
3244    Tag: blume-20020315-basis
3245    Description:
3246    
3247    Provided (preliminary?) implementations for
3248    
3249      {String,Substring}.{concatWith,isSuffix,isSubstring}
3250    
3251    and
3252    
3253      Substring.full
3254    
3255    Those are in the Basis spec but they were missing in SML/NJ.
3256    
3257    ----------------------------------------------------------------------
3258    Name: Matthias Blume
3259    Date: 2002/03/14 21:30:00 EST
3260    Tag: blume-20020314-controls
3261    Description:
3262    
3263    Controls:
3264    ---------
3265    
3266    1. Factored out the recently-added Controls : CONTROLS stuff and put
3267       it into its own library $/controls-lib.cm.  The source tree for
3268       this is under src/smlnj-lib/Controls.
3269    
3270    2. Changed the names of types and functions in this interface, so they
3271       make a bit more "sense":
3272    
3273          module -> registry
3274          'a registry -> 'a group
3275    
3276    3. The interface now deals in ref cells only.  The getter/setter interface
3277       is (mostly) gone.
3278    
3279    4. Added a function that lets one register an already-existing ref cell.
3280    
3281    5. Made the corresponding modifications to the rest of the code so that
3282       everything compiles again.
3283    
3284    6. Changed the implementation of Controls.MLRISC back to something closer
3285       to the original.  In particular, this module (and therefore MLRISC)
3286       does not depend on Controls.  There now is some link-time code in
3287       int-sys.sml that registers the MLRISC controls with the Controls
3288       module.
3289    
3290    CM:
3291    ---
3292    
3293      * One can now specify the lambda-split aggressiveness in init.cmi.
3294    
3295    ----------------------------------------------------------------------
3296    Name: Allen Leung
3297    Date: 2002/03/13 17:30:00 EST
3298    Tag: leunga-20020313-x86-fp-unary
3299    Description:
3300    
3301    Bug fix for:
3302    
3303    > leunga@weaselbane:~/Yale/tmp/sml-dist{21} bin/sml
3304    > Standard ML of New Jersey v110.39.1 [FLINT v1.5], March 08, 2002
3305    > - fun f(x,(y,z)) = Real.~ y;
3306    > [autoloading]
3307    > [autoloading done]
3308    >       fchsl   (%eax), 184(%esp)
3309    > Error: MLRisc bug: X86MCEmitter.emitInstr
3310    >
3311    > uncaught exception Error
3312    >   raised at: ../MLRISC/control/mlriscErrormsg.sml:16.14-16.19
3313    
3314    The problem was that the code generator did not generate any fp registers
3315    in this case, and the ra didn't know that it needed to run the X86FP phase to
3316    translate the pseudo fp instruction.   This only happened with unary fp
3317    operators in certain situations.
3318    
3319    ----------------------------------------------------------------------
3320    Name: Matthias Blume
3321    Date: 2002/03/13 14:00:00 EST
3322    Tag: blume-20020313-overload-etc
3323    Description:
3324    
3325    1. Added _overload as a synonym for overload for backward compatibility.
3326       (Control.overloadKW must be true for either version to be accepted.)
3327    
3328    2. Fixed bug in install script that caused more things to be installed
3329       than what was requested in config/targets.
3330    
3331    3. Made CM aware of the (_)overload construct so that autoloading
3332       works.
3333    
3334    ----------------------------------------------------------------------
3335    Name: Matthias Blume
3336    Date: 2002/03/12 22:03:00 EST
3337    Tag: blume-20020312-url
3338    Description:
3339    
3340    Forgot to update BOOT and srcarchiveurl.
3341    
3342    ----------------------------------------------------------------------
3343    Name: Matthias Blume
3344    Date: 2002/03/12 17:30:00 EST
3345    Tag: blume-20020312-version110392
3346    Description:
3347    
3348    Yet another version number bump (because of small changes to the
3349    binfile format).  Version number is now 110.39.2.  NEW BOOTFILES!
3350    
3351    Changes:
3352    
3353      The new pid generation scheme described a few weeks ago was overly
3354      complicated.  I implemented a new mechanism that is simpler and
3355      provides a bit more "stability":  Once CM has seen a compilation
3356      unit, it keeps its identity constant (as long as you do not delete
3357      those crucial CM/GUID/* files).  This means that when you change
3358      an interface, compile, then go back to the old interface, and
3359      compile again, you arrive at the original pid.
3360    
3361      There now also is a mechanism that instructs CM to use the plain
3362      environment hash as a module's pid (effectively making its GUID
3363      the empty string).  For this, "noguid" must be specified as an
3364      option to the .sml file in question within its .cm file.
3365      This is most useful for code that is being generated by tools such
3366      as ml-nlffigen (because during development programmers tend to
3367      erase the tool's entire output directory tree including CM's cached
3368      GUIDs).  "noguid" is somewhat dangerous (since it can be used to locally
3369      revert to the old, broken behavior of SML/NJ, but in specific cases
3370      where there is no danger of interface confusion, its use is ok
3371      (I think).
3372    
3373      ml-nlffigen by default generates "noguid" annotations.  They can be
3374      turned off by specifying -guid in its command line.
3375    
3376    ----------------------------------------------------------------------
3377    Name: Lal George
3378    Date: 2002/03/12 12 14:42:36 EST
3379    Tag: george-20020312-frequency-computation
3380    Description:
3381    
3382    Integrated jump chaining and static block frequency into the
3383    compiler. More details and numbers later.
3384    
3385    ----------------------------------------------------------------------
3386    Name: Lal George
3387    Date: 2002/03/11 11 22:38:53 EST
3388    Tag: george-20020311-jump-chain-elim
3389    Description:
3390    
3391    Tested the jump chain elimination on all architectures (except the
3392    hppa).  This is on by default right now and is profitable for the
3393    alpha and x86, however, it may not be profitable for the sparc and ppc
3394    when compiling the compiler.
3395    
3396    The gc test will typically jump to a label at the end of the cluster,
3397    where there is another jump to an external cluster containing the actual
3398    code to invoke gc. This is to allow factoring of common gc invocation
3399    sequences. That is to say, we generate:
3400    
3401            f:
3402               testgc
3403               ja   L1      % jump if above to L1
3404    
3405            L1:
3406               jmp L2
3407    
3408    
3409    After jump chain elimination the 'ja L1' instructions is converted to
3410    'ja L2'. On the sparc and ppc, many of the 'ja L2' instructions may end
3411    up being implemented in their long form (if L2 is far away) using:
3412    
3413            jbe     L3      % jump if below or equal to L3
3414            jmp     L2
3415         L3:
3416            ...
3417    
3418    
3419    For large compilation units L2  may be far away.
3420    
3421    
3422    ----------------------------------------------------------------------
3423    Name: Matthias Blume
3424    Date: 2002/03/11 13:30:00 EST
3425    Tag: blume-20020311-mltreeeval
3426    Description:
3427    
3428    A functor parameter was missing.
3429    
3430    ----------------------------------------------------------------------
3431    Name: Allen Leung
3432    Date: 2002/03/11 10:30:00 EST
3433    Tag: leunga-20020311-runtime-string0
3434    Description:
3435    
3436       The representation of the empty string now points to a
3437    legal null terminated C string instead of unit.  It is now possible
3438    to convert an ML string into C string with InlineT.CharVector.getData.
3439    This compiles into one single machine instruction.
3440    
3441    ----------------------------------------------------------------------
3442    Name: Allen Leung
3443    Date: 2002/03/10 23:55:00 EST
3444    Tag: leunga-20020310-x86-call
3445    Description:
3446    
3447       Added machine generation for CALL instruction (relative displacement mode)
3448    
3449    ----------------------------------------------------------------------
3450    Name: Matthias Blume
3451    Date: 2002/03/08 16:05:00
3452    Tag: blume-20020308-entrypoints
3453    Description:
3454    
3455    Version number bumped to 110.39.1.  NEW BOOTFILES!
3456    
3457    Entrypoints: non-zero offset into a code object where execution should begin.
3458    
3459    - Added the notion of an entrypoint to CodeObj.
3460    - Added reading/writing of entrypoint info to Binfile.
3461    - Made runtime system bootloader aware of entrypoints.
3462    - Use the address of the label of the first function given to mlriscGen
3463      as the entrypoint.  This address is currently always 0, but it will
3464      not be 0 once we turn on block placement.
3465    - Removed the linkage cluster code (which was The Other Way(tm) of dealing
3466      with entry points) from mlriscGen.
3467    
3468    ----------------------------------------------------------------------
3469    Name: Allen Leung
3470    Date: 2002/03/07 20:45:00 EST
3471    Tag: leunga-20020307-x86-cmov
3472    Description:
3473    
3474       Bug fixes for CMOVcc on x86.
3475    
3476       1. Added machine code generation for CMOVcc
3477       2. CMOVcc is now generated in preference over SETcc on PentiumPro or above.
3478       3. CMOVcc cannot have an immediate operand as argument.
3479    
3480    ----------------------------------------------------------------------
3481    Name: Matthias Blume
3482    Date: 2002/03/07 16:15:00 EST
3483    Tag: blume-20020307-controls
3484    Description:
3485    
3486    This is a very large but mostly boring patch which makes (almost)
3487    every tuneable compiler knob (i.e., pretty much everything under
3488    Control.* plus a few other things) configurable via both the command
3489    line and environment variables in the style CM did its configuration
3490    until now.
3491    
3492    Try starting sml with '-h' (or, if you are brave, '-H')
3493    
3494    To this end, I added a structure Controls : CONTROLS to smlnj-lib.cm which
3495    implements the underlying generic mechanism.
3496    
3497    The interface to some of the existing such facilities has changed somewhat.
3498    For example, the MLRiscControl module now provides mkFoo instead of getFoo.
3499    (The getFoo interface is still there for backward-compatibility, but its
3500    use is deprecated.)
3501    
3502    The ml-build script passes -Cxxx=yyy command-line arguments through so
3503    that one can now twiddle the compiler settings when using this "batch"
3504    compiler.
3505    
3506    TODO items:
3507    
3508    We should go through and throw out all controls that are no longer
3509    connected to anything.  Moreover, we should go through and provide
3510    meaningful (and correct!) documentation strings for those controls
3511    that still are connected.
3512    
3513    Currently, multiple calls to Controls.new are accepted (only the first
3514    has any effect).  Eventually we should make sure that every control
3515    is being made (via Controls.new) exactly once.  Future access can then
3516    be done using Controls.acc.
3517    
3518    Finally, it would probably be a good idea to use the getter-setter
3519    interface to controls rather than ref cells.  For the time being, both
3520    styles are provided by the Controls module, but getter-setter pairs are
3521    better if thread-safety is of any concern because they can be wrapped.
3522    
3523    *****************************************
3524    
3525    One bug fix: The function blockPlacement in three of the MLRISC
3526    backpatch files used to be hard-wired to one of two possibilities at
3527    link time (according to the value of the placementFlag).  But (I
3528    think) it should rather sense the flag every time.
3529    
3530    *****************************************
3531    
3532    Other assorted changes (by other people who did not supply a HISTORY entry):
3533    
3534    1. the cross-module inliner now works much better (Monnier)
3535    2. representation of weights, frequencies, and probabilities in MLRISC
3536       changed in preparation of using those for weighted block placement
3537       (Reppy, George)
3538    
3539    ----------------------------------------------------------------------
3540    Name: Lal George
3541    Date: 2002/03/07 14:44:24 EST 2002
3542    Tag: george-20020307-weighted-block-placement
3543    
3544    Tested the weighted block placement optimization on all architectures
3545    (except the hppa) using AMPL to generate the block and edge frequencies.
3546    Changes were required in the machine properties to correctly
3547    categorize trap instructions. There is an MLRISC flag
3548    "weighted-block-placement" that can be used to enable weighted block
3549    placement, but this will be ineffective without block/edge
3550    frequencies (coming soon).
3551    
3552    
3553    ----------------------------------------------------------------------
3554    Name: Lal George
3555    Date: 2002/03/05 17:24:48 EST
3556    Tag: george-20020305-linkage-cluster
3557    
3558    In order to support the block placement optimization, a new cluster
3559    is generated as the very first cluster (called the linkage cluster).
3560    It contains a single jump to the 'real' entry point for the compilation
3561    unit. Block placement has no effect on the linkage cluster itself, but
3562    all the other clusters  have full freedom in the manner in which they
3563    reorder blocks or functions.
3564    
3565    On the x86 the typical linkage code that is generated is:
3566       ----------------------
3567            .align 2
3568       L0:
3569            addl    $L1-L0, 72(%esp)
3570            jmp     L1
3571    
3572    
3573            .align  2
3574       L1:
3575       ----------------------
3576    
3577    72(%esp) is the memory location for the stdlink register. This
3578    must contain the address of the CPS function being called. In the
3579    above example, it contains the address of  L0; before
3580    calling L1 (the real entry point for the compilation unit), it
3581    must contain the address for L1, and hence
3582    
3583            addl $L1-L0, 72(%esp)
3584    
3585    I have tested this on all architectures except the hppa.The increase
3586    in code size is of course negligible
3587    
3588    ----------------------------------------------------------------------
3589    Name: Allen Leung
3590    Date: 2002/03/03 13:20:00 EST
3591    Tag: leunga-20020303-mlrisc-tools
3592    
3593      Added #[ ... ] expressions to mlrisc tools
3594    
3595    ----------------------------------------------------------------------
3596    Name: Matthias Blume
3597    Date: 2002/02/27 12:29:00 EST
3598    Tag: blume-20020227-cdebug
3599    Description:
3600    
3601    - made types in structure C and C_Debug to be equal
3602    - got rid of code duplication (c-int.sml vs. c-int-debug.sml)
3603    - there no longer is a C_Int_Debug (C_Debug is directly derived from C)
3604    
3605    ----------------------------------------------------------------------
3606    Name: Matthias Blume
3607    Date: 2002/02/26 12:00:00 EST
3608    Tag: blume-20020226-ffi
3609    Description:
3610    
3611    1. Fixed a minor bug in CM's "noweb" tool:
3612       If numbering is turned off, then truly don't number (i.e., do not
3613       supply the -L option to noweb).  The previous behavior was to supply
3614       -L'' -- which caused noweb to use the "default" line numbering scheme.
3615       Thanks to Chris Richards for pointing this out (and supplying the fix).
3616    
3617    2. Once again, I reworked some aspects of the FFI:
3618    
3619       A. The incomplete/complete type business:
3620    
3621       - Signatures POINTER_TO_INCOMPLETE_TYPE and accompanying functors are
3622         gone!
3623       - ML types representing an incomplete type are now *equal* to
3624         ML types representing their corresponding complete types (just like
3625         in C).  This is still safe because ml-nlffigen will not generate
3626         RTTI for incomplete types, nor will it generate functions that
3627         require access to such RTTI.   But when ML code generated from both
3628         incomplete and complete versions of the C type meet, the ML types
3629         are trivially interoperable.
3630    
3631         NOTE:  These changes restore the full generality of the translation
3632         (which was previously lost when I eliminated functorization)!
3633    
3634       B. Enum types:
3635    
3636       - Structure C now has a type constructor "enum" that is similar to
3637         how the "su" constructor works.  However, "enum" is not a phantom
3638         type because each "T enum" has values (and is isomorphic to
3639         MLRep.Signed.int).
3640       - There are generic access operations for enum objects (using
3641         MLRep.Signed.int).