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

sml/trunk/HISTORY revision 594, Tue Apr 4 23:41:47 2000 UTC sml/trunk/NOTES/HISTORY revision 2226, Wed Dec 6 04:45:33 2006 UTC
# Line 7  Line 7 
7    
8  The form of an entry should be:  The form of an entry should be:
9    
10  Name:  
11  Date:  Name: Your Name (web page or e-mail)
12  Tag: <post-commit CVS tag>  Date: yyyy/mm/dd hh:mm:ss
13    Tag: cvs-tag-for-this-change
14    Description:
15    
16    ----------------------------------------------------------------------
17    Name: Matthias Blume (http://www.tti-c.org/blume/)
18    Date: 2006/12/05 22:40:00 CST
19    Description:
20    
21    CM now reports undefined anchors as errors and aborts execution
22    rather than silently pressing on using bogus values.
23    
24    ----------------------------------------------------------------------
25    Name: Matthias Blume (http://www.tti-c.org/blume/)
26    Date: 2006/11/29 10:00:00 CST
27    Description:
28    
29    Use Say.vsay for printing the "[autoloading]" message, so
30    #set CM.Control.verbose false (or -Ccm.verbose=false) can be used
31    to suppress them.
32    
33    ----------------------------------------------------------------------
34    Name: Matthias Blume (http://www.tti-c.org/blume/)
35    Date: 2006/11/10 11:00:00 CST
36    Description:
37    
38    Fixed bug in CM where "with:" specifications that affect compilation
39    (as opposed to parsing) were ignored.
40    
41    ----------------------------------------------------------------------
42    Name: Matthias Blume (http://www.tti-c.org/blume/)
43    Date: 2006/11/09 14:30:00 CST
44    Tag: release-110.60
45    Description:
46    
47    New working release (110.60).  NEW BOOTFILES!
48    
49    This is the first subversion-hosted release.  There are also changes
50    to the directory layout.  Here is a table:
51    
52    (All pathnames are relative to the SML/NJ "root" directory.)
53    
54      config -> config
55    
56      src/cm -> base/cm
57      src/compiler -> base/compiler
58      src/runtime -> base/runtime
59      src/system -> base/system
60    
61      src/cm/pgraph -> pgraph
62    
63      src/READMES -> base/READMES
64      src/$notefile -> base/NOTES/$notefile
65         for <notefile> in BOOT CVSNOTES CYGWININSTALL HISTORY INSTALL
66                           MACOSXINSTALL WININSTALL
67    
68      src/smlnj-lib -> smlnj-lib
69      src/MLRISC -> MLRISC
70      ckit -> ckit
71      src/cml -> cml
72      src/eXene -> eXene
73      src/heap2asm -> heap2asm
74      src/lexgen -> lexgen
75      src/ml-burg -> ml-burg
76      src/ml-lex -> ml-lex
77      src/ml-yacc -> ml-yacc
78      src/ml-nlffi-lib -> nlffi/lib
79      src/ml-nlffigen -> nlffi/gen
80      src/smlnj-c -> smlnj-c
81      src/tools/TraceDebugProf -> trace-debug-profile
82    
83    In addition, there is also a new
84    
85      ml-lpt
86    
87    directory containing two new program generator tools: ml-ulex and
88    ml-antlr (a lexer generator that handles unicode and an ANTRL-inspired
89    LL(k) parser generator). These tools are currently "beta-quality"
90    
91    The latest versios of the sources can now be obtained anonymously
92    via subversion.  For this, it is useful to first check out
93    
94       svn://smlnj-gforge.cs.uchicago.edu/smlnj/admin
95    
96    and put the resulting directory on your shell's PATH.  This provides
97    access to three shell scripts: checkout-all.sh, stat-all.sh,
98    and refresh-all.sh.
99    
100    To create a freshly checked-out copy of the sources, do
101    
102       checkout-all.sh [dir]
103    
104    where dir is the optional SML/NJ root directory (default is ".").
105    
106    This creates the above directory layout.  Each subdirectory of
107    the root is under individual subversion control.  The stat-all.sh
108    and refresh-al.sh scripts apply "svn stat" or "svn update" to
109    each of these subtrees.
110    
111    ----------------------------------------------------------------------
112    Name: Matthias Blume (http://www.tti-c.org/blume/)
113    Date: 2006/11/02 12:20:00 CST
114    Description:
115    
116    Reorganized directory layout.
117    
118    This is a temporary solution, more reorganization is to come.
119    
120    The basic idea is to have a number of toplevel trees, each corresponding
121    to a well-defined part of the overall system.  Each part can be
122    maintained individually, even in separate source repositories,
123    although currently we still serve everything out of the main
124    smlnj-gforge tree.
125    
126    Installer and and scripts have been updated to reflect the new
127    layout.  The installer (base/src/system/smlnj/installer) is now
128    "scriptable" to avoid burning too much knowledge about the
129    layout into SML source code.  The main script used by the installer
130    is in config/actions.
131    
132    The main change is that many of the subdirectories of what used to
133    be known as "src" have moved to the toplevel.  The "src" directory
134    itself has moved down into a subtree called "base".  (We may eventually
135    get rid of the extra level of indirection represented by "src".)
136    
137    The layout is now as follows:
138    
139      toplevel tree name | default repository
140                         | (using gf=svn://smlnj-gforge.cs.uchicago.edu/smlnj)
141      -------------------+----------------------------------------------------
142      config             | $gf/config/trunk
143                         |
144      base               | $gf/sml/trunk
145                         |
146      smlnj-lib          | $gf/smlnj-lib/trunk
147      MLRISC             | $gf/MLRISC/trunk
148                         |
149      ml-yacc            | $gf/ml-yacc/trunk
150      ml-lex             | $gf/ml-lex/trunk
151      ml-burg            | $gf/ml-burg/trunk
152      lexgen             | $gf/lexgen/trunk
153      heap2asm           | $gf/heap2asm/trunk
154                         |
155      cml                | $gf/cml/trunk
156      eXene              | $gf/eXene/trunk
157      ckit               | $gf/ckit/trunk
158      nlffi              | $gf/nlffi/trunk
159                         |
160      smlnj-c            | $gf/smlnj-c/trunk
161    
162    In $gf/admin there are a few useful shell scripts for checking out
163    and maintaining the entire collection of trees:
164    
165       admin/checkout-all.sh [dir]
166          -- optionally creates dir and checks out all trees from their
167             default repositories; if dir is missing, checkout into .
168       admin/refresh-all.sh [dir]
169          -- looks at all trees (from the above list) in dir (default: .)
170             and runs "svn update" if the tree exists and is under
171             subversion control;
172             non-existing or non-subversion trees are skipped
173       admin/stat-all.sh [dir]
174          -- like refresh-all.sh, but runs "svn stat" instead
175    
176    ----------------------------------------------------------------------
177    Name: Matthias Blume (http://www.tti-c.org/blume/)
178    Date: 2006/10/05 11:30:00 CDT
179    Tag: blume-20061005-merge-amd64
180    Description:
181    
182    Merged code for AMD64 backend (Mike Rainey's work).
183    Everything is hooked up but untested.
184    
185    ----------------------------------------------------------------------
186    Name: Matthias Blume (http://www.tti-c.org/blume/)
187    Date: 2006/05/17 14:00:00 CST
188    Tag: Release_110_59
189    Description:
190    
191    New working version (110.59).  NEW BOOTFILES!
192    
193    I am freezing 110.59.  Changes other than the version-number increase:
194    
195      eXene:
196    
197         * committed changes to eXene from Alley Stoughton:
198             "fixed bugs in X authorization and resource handling, as well
199              as in the pile and viewport widgets"
200    
201      Runtime:
202        * fixed linking problem with NetBSD 3.x.
203    
204      Lexgen:
205        * lexgen tool handles non-ascii characters in 7-bit mode the same
206          way that ml-lex does
207    
208        * lexgen propagates exceptions the same way that ml-lex does
209    
210      CML:
211        * Fixed a bug in the SyncVar polling functions (iGetPoll,
212          mTakePoll, and mGetPoll) that could lead to livelock.
213    
214    ----------------------------------------------------------------------
215    Name: Matthias Blume (http://www.tti-c.org/blume/)
216    Date: 2006/05/12 15:20:00 CST
217    Tag: blume-20060515-makedepend
218    Description:
219    
220    Implemented ml-makedepend (i.e., CM.sources) in a better (more robust)
221    way.  This should hopefully fix the ml-makedepend problem permanently.
222    
223    ----------------------------------------------------------------------
224    Name: Matthias Blume (http://www.tti-c.org/blume/)
225    Date: 2006/05/12 15:15:00 CST
226    Tag: blume-20060512-makedepend
227    Description:
228    
229    Fixed long-standing bug with ml-makedepend where it would output a
230    spurious dependency to a non-existing file.  (This is a simple fix.
231    It might need further looking into.)
232    
233    ----------------------------------------------------------------------
234    Name: Matthias Blume (http://www.tti-c.org/blume/)
235    Date: 2006/04/20 10:30:00 CDT
236    Tag: blume-20060420-heap2exec
237    Description:
238    
239    Committed patches received from Johannes 5 Joemann (joemann@befree.free.de)
240    that enable heap2exec under Linux and FreeBSD.
241    
242    ----------------------------------------------------------------------
243    Name: John Reppy
244    Date: 2006/04/14
245    Tag:
246    Description:
247    
248    MLRISC changes:
249      1) renamed GAS_PSEUDO_OPS to AS_PSEUDO_OPS and put it in its own file.
250      2) added support for NOTB and XORB operators in pseudo-op expressions
251      3) added DarwinPseudoOp functor that supports Darwin's assembler syntax.
252      4) added support for 64-bit integer literals
253    
254    ----------------------------------------------------------------------
255    Name: Matthias Blume (http://www.tti-c.org/blume/)
256    Date: yyyy/mm/dd
257    Tag: Release_110_58
258    Description:
259    
260    New working version (110.58).  NEW BOOTFILES!
261    
262    ----------------------------------------------------------------------
263    Name: Matthias Blume (http://www.tti-c.org/blume/)
264    Date: 2006/03/01 22:15:00 CST
265    Tag: blume-20060301-lexgen
266    Description:
267    
268    Incorporated several bugfixes to lexgen.  Compiler now compiles to
269    fixpoint when using lexgen instead of ml-lex.
270    
271    ----------------------------------------------------------------------
272    Name: Matthias Blume (http://www.tti-c.org/blume/)
273    Date: 2006/02/28 22:45:00 CST
274    Tag: blume-20060228-lexgen
275    Description:
276    
277    Removed ml-flex and added lexgen instead, using Aaron Turon's newly
278    provided tarball.  The generated lexers still have problems.
279    
280    ----------------------------------------------------------------------
281    Name: Matthias Blume (http://www.tti-c.org/blume/)
282    Date: 2006/02/26 23:30:00 CST
283    Tag: blume-20060226-ml-flex
284    Description:
285    
286    Removed ml-flex' dependency on regexp-lib.cm.  Turned installation
287    of ml-flex on by default.
288    
289    ----------------------------------------------------------------------
290    Name: Matthias Blume (http://www.tti-c.org/blume/)
291    Date: 2006/02/24 16:30:00 CST
292    Tag: blume-20060224-ml-flex
293    Description:
294    
295    Added ml-flex sources.  Partially integrated, but should not be
296    turned on yet!  (Read: leave it commented-out in config/targets!)
297    
298    ----------------------------------------------------------------------
299    Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
300    Date: 2006/02/23
301    Tag: jhr-20060223-x86-64
302    Description:
303    Changes to support building on x86-64 systems (using the 32-bit mode).
304    Also cleaned up signal handling on Linux.  Support for pre-2.2 Linux
305    kernels dropped.
306    
307    ----------------------------------------------------------------------
308    Name: Allen Leung (leunga (at) reservoir (dot) com)
309    Date: 2006/02/22 15:07:00 EST
310    Tag: leunga-20060222-mdl-gen
311    Description:
312    
313      CM has changed.  Updated the script for rebuilding the MLRISC generated
314    files.
315    
316    ----------------------------------------------------------------------
317    Name: Matthias Blume (http://www.tti-c.org/blume/)
318    Date: 2006/02/14 17:30:00 CST
319    Tag: blume-20060214-intel-mac
320    Description:
321    
322    Hooked code for Darwin-specific Intel ABI into main compiler.
323    (This is a temporary solution which relies on the fact that the
324    compiler itself does not use NLFFI.  Eventually we need to divorce
325    intel mac from generic x86 unix code and make separate sets of binaries.)
326    
327    ----------------------------------------------------------------------
328    Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
329    Date: 2006/02/13
330    Tag: jhr-20060213-intel-mac
331    Description:
332    Changed MLRISC x86 CCalls for partial support of MacOS X ABI.
333    
334    ----------------------------------------------------------------------
335    Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
336    Date: 2006/02/06
337    Tag: jhr-20060206-intel-mac
338    Description:
339    Changes to support MacOS X on Intel hardware.  The C-calls support in
340    MLRISC must also be updated to support the MacOS X ABI.
341    
342    ----------------------------------------------------------------------
343    Name: Matthias Blume (http://www.tti-c.org/blume/)
344    Date: 2005/12/16 00:15:00 CST
345    Tag: blume-20051216-cm
346    Description:
347    
348    Improved error reporting and handling in CM.
349    
350    ----------------------------------------------------------------------
351    Name: Matthias Blume (http://www.tti-c.org/blume/)
352    Date: 2005/11/21 14:00:00 CST
353    Tag: blume-20051121-html-lib
354    Description:
355    
356    Enabling $/html-lib.cm in config/preloads did not work.  This is fixed
357    now.  (Since the anchor mapping for html-lib.cm is not yet in effect
358    at the time when config/preloads is processed, the library has to
359    be referred to by another name.  In 110.57 this would be
360    $SMLNJ-LIB/HTML/html-lib.cm.  I arranged for $smlnj/smlnj-lib/html-lib.cm
361    to be valid as well.)
362    
363    Thanks to Todd Wilson (Fresno) for alerting me to this issue.
364    
365    ----------------------------------------------------------------------
366    Name: Matthias Blume (http://www.tti-c.org/blume/)
367    Date: 2005/11/19 13:45:00 CST
368    Tag: Release_110_57
369    Description:
370    
371    Moved release tag for 110.57.
372    
373    ----------------------------------------------------------------------
374    Name: Matthias Blume (http://www.tti-c.org/blume/)
375    Date: 2005/11/19 23:35:00 CST
376    Tag: blume-20051119-fish
377    Description:
378    
379    Fixed a problem in config/install.sh where it tries to "fish" the
380    name of the CM metadata directory from the wrong place (because
381    the physical location of basis.cm has changed).  Also, corrected the
382    path anchor for $/html-lib.cm.  (Thanks to M. Fluet for pointing out
383    these problems.)
384    
385    ----------------------------------------------------------------------
386    Name: Matthias Blume (http://www.tti-c.org/blume/)
387    Date: 2005/11/16 14:30:00 CST
388    Tag: blume-orig-110_57
389    Description:
390    
391    New working version (110.57).  NEW BOOTFILES!
392    
393    UPDATE: The release tag has been moved to include a fix for a brown-paper-
394            bag glitch (see above).
395    
396    ----------------------------------------------------------------------
397    Name: Matthias Blume (http://www.tti-c.org/blume/)
398    Date: 2005/11/16 10:25:00 CST
399    Tag: blume-22051116-btrace
400    Description:
401    
402    Fixed problem with bogus exception message when using back-trace
403    facility.
404    
405    ----------------------------------------------------------------------
406    Name: Matthias Blume (http://www.tti-c.org/blume/)
407    Date: 2005/11/15 22:35:00 CST
408    Tag: blume-22051115-array2copy
409    Description:
410    
411    Added simple implementation of Array2.copy.  (Warning: mostly untested.)
412    
413    ----------------------------------------------------------------------
414    Name: Matthias Blume (http://www.tti-c.org/blume/)
415    Date: 2005/11/15 21:35:00 CST
416    Tag: blume-22051115-versiontool
417    Description:
418    
419    Reversed change to src/system/smlnj/internals/versiontool.cm.  This
420    file gets loaded as a tool -- by the equivalent of CM.make during the
421    run of CMB.make.  Thus, CMB's path configuration is meaningless for
422    it.  Instead, it has the status of "user code", so it should use
423    $/basis.cm to refer to the Basis library.  (At least that's true for
424    the purpose of bootstrapping the previous change.  In the future it
425    might make sense to have versiontool.cm refer to
426    $smlnj/basis/basis.cm, i.e., the version of the Basis that the
427    compiler itself uses.)
428    
429    Also patched src/system/testml to have it activate those extra anchor
430    bindings in config/extrapathconfig.
431    
432    ----------------------------------------------------------------------
433    Name: Matthias Blume (http://www.tti-c.org/blume/)
434    Date: 2005/11/15 16:35:00 CST
435    Tag: blume-22051115-libraries
436    Description:
437    
438    This change affects the way the following libraries are tied into
439    the system:
440    
441       $/basis.cm
442       $/smlnj-lib.cm
443       $/pp-lib.cm
444       $/controls-lib.cm
445       $/html-lib.cm
446       $/ml-yacc-lib.cm
447    
448    These libraries are now internally (as seen from the source code
449    of the implementation itself) known by the following names:
450    
451       $smlnj/basis/basis.cm
452       $smlnj/smlnj-lib/smlnj-lib.cm
453       $smlnj/smlnj-lib/pp-lib.cm
454       $smlnj/smlnj-lib/controls-lib.cm
455       $smlnj/smlnj-lib/html-lib.cm
456       $smlnj/ml-yacc/ml-yacc-lib.cm
457    
458    This makes it possible to work with code that requires different
459    versions of these libraries, and which refers to these libraries using
460    their "default" names (i.e., the first set of names above).  In other
461    words, one can un-define or re-define those default names without
462    compromising the proper functioning of the compiler itself.
463    
464    A similar procedure had already been performed for several of the
465    MLRISC libraries that are linked into the compiler.  I did some
466    cleanup on this code.
467    
468    A new file in the config directory (named extrapathconfig) is
469    responsible for setting up path anchors that the compiler itself does
470    not need, but that are typically required by user code.
471    
472    ----------------------------------------------------------------------
473    Name: Matthias Blume
474    Date: 2005/11/07 17:05:00 CST
475    Tag: blume-20051107-slice
476    Description:
477    
478    Fixed erroneous out-of-bounds test in the "update" function of
479    various *ArraySlice modules.  (Thanks to Vesa A. Norrman for pointing
480    out the problem.)
481    
482    Pushed some Basis changes through ML-Lex, CML, and eXene.
483    
484    ----------------------------------------------------------------------
485    Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
486    Date: 2005/11/07
487    Description:
488    Fixed a Basis incompatibility: The depreciated function Substring.all
489    was removed (use Substring.full instead).
490    
491    ----------------------------------------------------------------------
492    Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
493    Date: 2005/11/05
494    Description:
495    Tweaked interval set API in SML/NJ library; see the CHANGES file for details.
496    
497    ----------------------------------------------------------------------
498    Name: Matthias Blume (http://www.tti-c.org/blume/)
499    Date: 2005/11/02 16:25:00 CST
500    Tag: blume-20051102-hex
501    Description:
502    
503    * Runtime system bootstrap code now accepts hex digits in BOOTLIST
504      in either upper case or lower case format.
505    * Pushed changes to names of Pack<N>{Big,Little} structures through
506      CML and eXene.
507    
508    ----------------------------------------------------------------------
509    Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
510    Date: 2005/11/02
511    Description:
512    Fixed a Basis incompatibility: Pack<N>{Big,Little} structures should
513    be called PackWord<N>{Big,Little}.
514    
515    ----------------------------------------------------------------------
516    Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
517    Date: 2005/10/28
518    Description:
519    Fixed a minor Basis incompatibility: hex digits should be upper case.
520    
521    ----------------------------------------------------------------------
522    Name: Matthias Blume (blume (at) tti - c (dot) org)
523    Date: 2005/10/25 16:00:00 CDT
524    Tag: Release_110_56
525    Description:
526    
527    New working version (110.56).  NEW BOOTFILES!
528    
529    ----------------------------------------------------------------------
530    Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
531    Date: 2005/10/25
532    Description:
533    Added interval sets to utility library (signatures INTERVAL_DOMAIN
534    and INTERVAL_SET, and functor IntervalSetFn).
535    
536    ----------------------------------------------------------------------
537    Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
538    Date: 2005/10/14
539    Description:
540    Add Zhong Shao's fix for datatype equality functions.
541    
542    ----------------------------------------------------------------------
543    Name: Allen Leung (leunga (at) reservoir (dot) com)
544    Date: 2005/10/14 12:30:00 EST
545    Tag: leunga-20051014-x86-spill
546    Description:
547    
548    Bug fix a bug found by Carl Hauser.
549    There was a typo in the reload code for FCMP in x86SpillInstr.sml.
550    
551    ----------------------------------------------------------------------
552    Name: Allen Leung (leunga (at) reservoir (dot) com)
553    Date: 2005/10/14 11:15:00 EST
554    Tag: leunga-20051014-x86-asm
555    Description:
556    
557    Removed some debugging code in file x86Asm.sml.
558    The function emit_operand was printing out debugging output.
559    
560    ----------------------------------------------------------------------
561    Name: John Reppy
562    Date: 2005/07/27
563    Tag:
564    Description:
565    
566    Fixed ml-lex to recognize "\r" as representing carriage return.
567    
568    ----------------------------------------------------------------------
569    Name: John Reppy
570    Date: 2005/07/27
571    Tag:
572    Description:
573    
574    Fixed ml-yacc to work on files with non-native end-of-line
575    encodings (e.g., Windows text file on a Unix system).
576    
577    ----------------------------------------------------------------------
578    Name: John Reppy
579    Date: 2005/07/20
580    Tag:
581    Description:
582    
583    Added changes from Dominic Evans (oldmanuk (at) gmail (dot) com)
584    to support HPUX 11.
585    
586    ----------------------------------------------------------------------
587    Name: John Reppy
588    Date: 2005/07/06
589    Tag:
590    Description:
591    
592    Changes to the SML/NJ library.  See smlnj-lib/CHANGES for details.
593    
594    ----------------------------------------------------------------------
595    Name: Matthias Blume (blume (at) tti - c (dot) org)
596    Date: 2005/07/06 00:45:00 CDT
597    Tag: blume-20050706-slice-copy
598    Description:
599    
600    Fixed reversed logic for deciding whether to "copy up" or "copy down"
601    in *-array-slice.sml.
602    
603    ----------------------------------------------------------------------
604    Name: Allen Leung (leunga (at) reservoir (dot) com)
605    Date: 2005/05/31 17:00:00 EST
606    Tag: leunga-20050531-cygwin-fault-2
607    Description:
608    
609    A typo in the cygwin code fixed.
610    
611    ----------------------------------------------------------------------
612    Name: Allen Leung (leunga (at) reservoir (dot) com)
613    Date: 2005/05/31 16:47:00 EST
614    Tag: leunga-20050531-cygwin-fault
615    Description:
616    
617    Updated Cygwin's fault/signal handling to match the Windows version.
618    Updated the export list.
619    
620    ----------------------------------------------------------------------
621    Name: Matthias Blume (blume (at) tti - c (dot) org)
622    Date: 2005/05/18 13:35:00 CDT
623    Tag: Release_110_54
624    Description:
625    
626    New working version (110.54).  NEW BOOTFILES!
627    
628    ----------------------------------------------------------------------
629    Name: Matthias Blume (blume (at) tti - c (dot) org)
630    Date: 2005/05/18 11:58:00 CDT
631    Tag: blume-20050518-installer
632    Description:
633    
634    Added support scripts for Mac OS X PackageMaker and modified
635    config/install.sh so that it supports re-dumping a heap image
636    after customization.
637    
638    ----------------------------------------------------------------------
639    Name: Matthias Blume (blume (at) tti - c (dot) org)
640    Date: 2005/05/18 10:55:00 CDT
641    Tag: blume-20050518-realdiv-noovld
642    Description:
643    
644    Un-overloaded / to work around bug in overloading resolution code.
645    
646    ----------------------------------------------------------------------
647    Name: Matthias Blume (blume (at) tti - c (dot) org)
648    Date: 2005/05/16 23:50:00 CDT
649    Tag: blume-20050516-redump-heap
650    Description:
651    
652    Added mechanism for re-creating a heap file for the interactive system
653    after configuration variables have been changed.
654    
655       CM.redump_heap : string -> unit
656    
657    This is much like SMLofNJ.exportML, but starting from the resulting
658    heap does not return to the caller of CM.redump_heap but
659    restarts the interactive system from scratch.  The original call of
660    CM.redump_heap does not return but ends the interactive session.
661    Thus, CM.redump_heap is a lot like SMLofNJ.exportFn.
662    
663    Internally, redump_heap winds the dynamic execution context back to
664    the point where the original heap image was created and re-executes
665    the heap image generation code in the boot code.
666    
667    ----------------------------------------------------------------------
668    Name: Matthias Blume (blume (at) tti - c (dot) org)
669    Date: 2005/05/09 21:55:00 CDT
670    Tag: blume-20050509-word64
671    Description:
672    
673    Added a hack to the existing hack known as Word64 to make fromString
674    behave correctly.  I am still not sure whether Word64.scan will work
675    as specified with respect to the interaction of radix and prefix.
676    
677    ----------------------------------------------------------------------
678    Name: Allen Leung (leunga (at) reservoir (dot) com)
679    Date: 2005/05/04 11:50:00 EST
680    Tag: leunga-20050504-checkgc
681    Description:
682    
683    Added a gc protocol checking phase.  This phase is enabled with
684    the flag "check-gc".  "debug-check-gc" turns on the verbose mode.
685    
686    ----------------------------------------------------------------------
687    Name: Matthias Blume (blume (at) tti - c (dot) org)
688    Date: 2005/05/04 10:45:00 CDT
689    Tag: blume-20050504-intinf
690    Description:
691    
692    Fixed a bug in the implementation of div and mod for IntInf.
693    Thanks to Neophytos Michael for reporting the problem.
694    
695    ----------------------------------------------------------------------
696    Name: Matthias Blume (blume (at) tti - c (dot) org)
697    Date: 2005/05/04 10:35:00 CDT
698    Tag: blume-20050504-join
699    Description:
700    
701    Added a "join" combinator to the ParserComb module in smlnj-lib.cm.
702    
703    ----------------------------------------------------------------------
704    Name: Matthias Blume (blume (at) tti - c (dot) org)
705    Date: 2005/02/28 23:40:00 CST
706    Tag: blume-20050228-mVar
707    Description:
708    
709    Fixed serious bug (brown paper bag variety) in new implementation of
710    structure Atom in CML.  (I had accidentally used a mailbox instead of
711    an mvar, leaving the door open for races.)
712    
713    ----------------------------------------------------------------------
714    Name: Matthias Blume (blume (at) tti - c (dot) org)
715    Date: 2005/02/25 15:00:00 CST
716    Tag: Release_110_53
717    Description:
718    
719    New working version (110.53).  NEW BOOTFILES!
720    
721    ----------------------------------------------------------------------
722    Name: Matthias Blume (blume (at) tti - c (dot) org)
723    Date: 2005/02/25 14:50:00 CST
724    Tag: blume-20050225-susp
725    Description:
726    
727    Brought back SMLofNJ.Susp.  The underlying suspension type is the one
728    implemented in Core -- which means that it is the same as the one used
729    by the lazy extension.
730    
731    ----------------------------------------------------------------------
732    Name: Matthias Blume (blume (at) tti - c (dot) org)
733    Date: 2005/02/24 16:50:00 CST
734    Tag: blume-20050224-cml-atom
735    Description:
736    
737    Simpler and at the same time more general implementation of structure
738    Atom in CML.
739    
740    ----------------------------------------------------------------------
741    Name: Matthias Blume (blume (at) tti - c (dot) org)
742    Date: 2005/02/15 17:35:00 CST
743    Tag: blume-20050215-tools
744    Description:
745    
746    Created new "tools" directory under "src" and moved "TraceDebugProf"
747    there.
748    
749    ----------------------------------------------------------------------
750    Name: Matthias Blume (blume (at) tti - c (dot) org)
751    Date: 2005/02/10 17:55:00 CST
752    Tag: blume-20050210-longlong
753    Description:
754    
755    Implemented "long long" arguments and results for NLFFI.  (Only the
756    PPC/MacOS implementation is complete, the other backends still need to
757    be updated.)
758    
759    ----------------------------------------------------------------------
760    Name: Matthias Blume (blume (at) tti - c (dot) org)
761    Date: 2005/01/24 17:40:00 CST
762    Tag: blume-20050124-mlyacc
763    Description:
764    
765    Minor cleanup in ML-Yacc rule printing mechanism.  This should fix a
766    problem with certain "as" patterns which previously got rendered
767    using incorrect syntax.
768    
769    ----------------------------------------------------------------------
770    Name: Matthias Blume (blume (at) tti - c (dot) org)
771    Date: 2005/01/18 12:00:00 CST
772    Tag: blume-20050118-profile
773    Description:
774    
775    Made time profiling code (interrupt handler) in runtime system aware
776    of new array representation.
777    
778    ----------------------------------------------------------------------
779    Name: Matthias Blume (blume (at) tti - c (dot) org)
780    Date: 2005/01/14 18:00:00 CST
781    Tag: blume-20050114-heap2exec
782    Description:
783    
784    Implemented new (but still experimental) heap2exec facility.  This is
785    tested under Mac OS X and should work under Linux (will test shortly).
786    It will probably also work on the Sparc (will test some time later).
787    
788      - removed old "HACKED_STANDALONE" hack from runtime
789    
790    To be able to test this, uncomment the request for "heap2asm" in
791    config/targets prior to installation.  (Notice that this is different
792    from "heap2exec" mentioned below.  Not a typo.)
793    
794    To perform an actual test, run the command
795    
796       $ bin/heap2exec heapfile execfile
797    
798    (You can put heap2exec on your shell's path.)
799    
800    For example, run
801    
802      $ bin/heap2exec bin/.heap/ml-yacc.ppc-darwin mly
803    
804    This will create a standalone executable called "mly" which you can
805    then invoke directly as a command.
806    
807    ----------------------------------------------------------------------
808    Name: Matthias Blume (blume (at) tti - c (dot) org)
809    Date: 2005/01/07 11:44:00 CST
810    Tag: blume-20050107-mlstring
811    Description:
812    
813    fixed off-by-one error in ML_STRING macro (globals.c)
814    
815    ----------------------------------------------------------------------
816    Name: Matthias Blume (blume (at) tti - c (dot) org)
817    Date: 2004/12/23 18:00:00 CST
818    Tag: blume-20041223-santa
819    Description:
820    
821    Made ml-build script "smarter" (but only very little).
822    
823    ----------------------------------------------------------------------
824    Name: Matthias Blume (blume (at) tti - c (dot) org)
825    Date: 2004/12/21 15:05:00 CST
826    Tag: blume-20041221-longlong
827    Description:
828    
829    * Implemented access to signed and unsigned long long data in NLFFI.
830    (The parameter-passing part of the picture has not complete. But
831    data structure access seems to work.)
832    
833    * Fixed CM's incorrect assumption that the PPC is little-endian.
834      (On the Mac, it is big-endian.  And that's currently our only
835      PPC platform.)
836    
837    ----------------------------------------------------------------------
838    Name: Matthias Blume (blume (at) tti - c (dot) org)
839    Date: 2004/12/21 12:50:00 CST
840    Tag: blume-20041221-memory
841    Description:
842    
843    Some cleanup in the $c/memory.cm library: separated some concerns by
844    moving allocation code and memory access code each into their own
845    files.
846    
847    ----------------------------------------------------------------------
848    Name: Allen Leung (leunga (at) reservoir (dot) com)
849    Date: 2004/12/17 16:12:00 EST
850    Tag: leunga-20041217-cygwin-smlnj-home
851    Description:
852    
853        The Unix I/O library of SML/NJ on cygwin does not understand
854    Windows style pathname, so problems arise when SMLNJ_HOME is set
855    to a Windows style pathname.  _run-sml now converts SMLNJ_HOME
856    to a POSIX pathname on cygwin.
857    
858    ----------------------------------------------------------------------
859    Name: Matthias Blume (blume (at) tti - c (dot) org)
860    Date: 2004/12/16 13:35:00 CST
861    Tag: Release_110_52
862    Description:
863    
864    Last-minute changes incorporated into 110.52.  Release tag moved.
865    
866    The changes:
867       - HashString.hashString' -> HashString.hashSubstring
868       - bug fix in UnivariateStats
869    
870    ----------------------------------------------------------------------
871    Name: Matthias Blume (blume (at) tti - c (dot) org)
872    Date: 2004/12/15 23:40:00 CST
873    Tag: blume-20041215-hashSubstring
874    Description:
875    
876    - HashString.hashString' -> HashString.hashSubstring
877    - corresponding changes in atom.sml
878    - "de-compressed" (aka. un-obfuscated) code for UnivariateStats and
879      added some comments
880    
881    ----------------------------------------------------------------------
882    Name: Matthias Blume (blume (at) tti - c (dot) org)
883    Date: 2004/12/15 15:30:00 CST
884    Tag: (Release_110_52)
885    Description:
886    
887    New working version (110.52).  NEW BOOTFILES!
888    
889    ----------------------------------------------------------------------
890    Name: Matthias Blume (blume (at) tti - c (dot) org)
891    Date: 2004/12/15 12:45:00 CST
892    Tag: blume-20041215-spaces
893    Description:
894    
895    More on the space problem (this time for Win32).
896    
897    ----------------------------------------------------------------------
898    Name: Matthias Blume (blume (at) tti - c (dot) org)
899    Date: 2004/12/14 17:30:00 CST
900    Tag: blume-20041214-spaces
901    Description:
902    
903    Hacked some of the scripts (in particular: the installer) to cope with
904    spaces in filenames a bit better.  But beware: the current "solution"
905    is likely still full of bugs and inherently incomplete.  (We need to
906    do away with those shell scripts for a comprehensive solution.)
907    
908    ----------------------------------------------------------------------
909    Name: Matthias Blume (blume (at) tti - c (dot) org)
910    Date: 2004/12/13 14:45:00 CST
911    Tag: blume-20041213-ml-makedepend
912    Description:
913    
914    Fixed bug in code for ml-makedepend.
915    
916    ----------------------------------------------------------------------
917    Name: Matthias Blume (blume (at) tti - c (dot) org)
918    Date: 2004/12/09 16:30:00 CST
919    Tag: blume-20041209-statistics
920    Description:
921    
922    Added two simple but potentially useful statistics modules to SML/NJ Library.
923    (See CHANGES file there.)
924    
925    ----------------------------------------------------------------------
926    Name: Matthias Blume (blume (at) tti - c (dot) org)
927    Date: 2004/12/01 16:50:00 CST
928    Tag: blume-20041201-atom
929    Description:
930    
931    smlnj-lib:
932    
933      Added function HashString.hashString' for substrings.
934      Hand-inlined CharVector.foldl into HashString (for speed).
935      Modified implementation of structure Atom to avoid extracting
936      strings from substrings unless necessary.
937      (Also see CHANGES file for smlnj-lib.)
938    
939    ----------------------------------------------------------------------
940    Name: Matthias Blume (blume (at) tti - c (dot) org)
941    Date: 2004/11/24 22:15:00 CST
942    Tag: blume-20041124-cml
943    Description:
944    
945    Made sure CML compiles when Position = Int64.
946    
947    ----------------------------------------------------------------------
948    Name: Matthias Blume (blume (at) tti - c (dot) org)
949    Date: 2004/11/24 14:30:00 CST
950    Tag: blume-20041124-position
951    Description:
952    
953    The compiler can now be compiled in a mode that makes structure Position
954    equal to Int64.  The default, however, is unchanged (Position = Int31)
955    for the time being.
956    
957    To enable 64-bit positions, use the following procedure:
958    
959    1. Start sml
960    2. Autoload $smlnj/cmb.cm (if not already autoloaded)
961    3. Type
962          #set (CMB.symval "USE_64_BIT_POSITIONS") (SOME 1);
963    4. Run CMB.make() as usual.
964    
965    This is barely tested.  The only test so far was a little SML program
966    counting the number of characters in an 8-gigabyte file by
967    reading it character-by-character.  That test was successful.
968    
969    In support of 64-bit positions, a number of new functions have been
970    added to the runtime system.
971    
972    ----------------------------------------------------------------------
973    Name: Matthias Blume (blume (at) tti - c (dot) org)
974    Date: 2004/11/23 14:45:00 CST
975    Tag: blume-20041123-useFile
976    Description:
977    
978    Fixed a problem with unhelpful error messages related to problems with
979    .cm- or .sml files that appear as part of the sml command line.
980    
981    ----------------------------------------------------------------------
982    Name: Matthias Blume (blume (at) tti - c (dot) org)
983    Date: 2004/11/18 15:40:00 CST
984    Tag: Release_110_51
985    Description:
986    
987    New working version (110.51).  NEW BOOTFILES!
988    
989    ----------------------------------------------------------------------
990    Name: Matthias Blume (blume (at) tti - c (dot) org)
991    Date: 2004/11/18 15:35:00 CST
992    Tag: <none>
993    Description:
994    
995    Enabled dlopen and friends for FreeBSD (as recommended by Johannes 5
996    Joemann).
997    
998    ----------------------------------------------------------------------
999    Name: Allen Leung (leunga (at) reservoir (dot) com)
1000    Date: 2004/11/17 16:05:21 EST 2004
1001    Tag: leunga-20041117-mlrisc-live-kill
1002    Description:
1003    
1004       Added support for MLTree constructs LIVE and KILL
1005    to all the architectures.
1006    
1007    ----------------------------------------------------------------------
1008    Name: Matthias Blume (blume (at) tti - c (dot) org)
1009    Date: 2004/11/13 00:20:00 CST
1010    Tag: blume-20041113-versiontool
1011    Description:
1012    
1013    - Stripped down the versiontool: It now only handles the version number.
1014      The date string is generated at bootstrap time (during makeml).
1015    
1016    - In a previous commit, fixed a minor issue with how polyequal is being
1017      translated.  In particular, the code now "looks through" abstractions.
1018      This results in slightly fewer polyEqual warnings and hopefully slightly
1019      more efficient code.  Important examples for where this matters are
1020      the new int64 and word64 types.
1021    
1022    ----------------------------------------------------------------------
1023    Name: Matthias Blume (blume (at) tti - c (dot) org)
1024    Date: 2004/11/12 00:30:00 CST
1025    Tag: blume-20041112-int64
1026    Description:
1027    
1028    Structure Int64 fully hooked in.  (The implementation is not very
1029    efficient, though.)
1030    
1031    ----------------------------------------------------------------------
1032    Name: Matthias Blume (blume (at) tti - c (dot) org)
1033    Date: 2004/11/11 17:30:00 CST
1034    Tag: blume-20041111-more64
1035    Description:
1036    
1037    All the pieces of Word64 are now there, with the exception of the
1038    conversions from and to LargeWord. (Eventually these need to be identities,
1039    but for the time being they don't even make sense because LargeWord is
1040    32-bit wide.)
1041    
1042    Also started to add similar support for Int64, but major pieces of that
1043    are still missing.
1044    
1045    ----------------------------------------------------------------------
1046    Name: Matthias Blume (blume (at) tti - c (dot) org)
1047    Date: 2004/11/11 00:15:00 CST
1048    Tag: blume-20041111-word64
1049    Description:
1050    
1051    Structure Word64 is now (almost) complete, word literals and patterns
1052    seem to work.  There are a few odd pieces missing.  In particular,
1053    I didn't do the {from,to}LargeWord parts because LargeWord is still Word32
1054    at the moment.
1055    
1056    Making Word64 official would mean that LargeWord becomes Word64.  But
1057    this requires extreme care because most word-word conversions have to
1058    go through LargeWord, so making a mistake means loss of efficiency or
1059    worse.  Eventually there will be a solution similar to (but actually
1060    simpler than) what I did with IntInf.
1061    
1062    ----------------------------------------------------------------------
1063    Name: Matthias Blume (blume (at) tti - c (dot) org)
1064    Date: 2004/11/10 18:12:00 CST
1065    Tag: blume-20041110-64bit
1066    Description:
1067    
1068    More 64-bit hacking (but still not even half-way there yet).
1069    Also, some assorted improvements to the handling of 8-bit words.
1070    
1071    ----------------------------------------------------------------------
1072    Name: Matthias Blume (blume (at) tti - c (dot) org)
1073    Date: 2004/11/09 17:50:00 CST
1074    Tag: <none>
1075    Description:
1076    
1077    Started adding some infrastructure for supporting 64-bit int- and
1078    word-types.  (Still in its very early stages.)
1079    
1080    ----------------------------------------------------------------------
1081    Name: Matthias Blume (blume (at) tti - c (dot) org)
1082    Date: 2004/10/28 10:45:00 CDT
1083    Tag: Release_110_50
1084    Description:
1085    
1086    New working version (110.50).  NEW BOOTFILES!
1087    
1088                 =====================
1089    
1090    Also:
1091    
1092      - Changend config/srcarchiveurl from a file just
1093        containing the URL string into a file containing
1094        shell script code.  The code has access to the $VERSION variable.
1095      - Made corresponding changes to config/install.sh and config/unpack.
1096      - Default contents of config/srcarchiveurl uses $VERSION and
1097        normally does not have to be edited to reflect a version change.
1098    
1099      (As a result, a version change can be done by just editing
1100       config/version, the rest is now automatic.)
1101    
1102    ----------------------------------------------------------------------
1103    Name: Matthias Blume (blume (at) tti - c (dot) org)
1104    Date: 2004/10/27 17:50:00 CDT
1105    Tag: blume-20041027-btrace-msg
1106    Description:
1107    
1108    BackTrace.monitor now also reports the source of the exception that
1109    triggered the trace.
1110    
1111    ----------------------------------------------------------------------
1112    Name: Matthias Blume (blume (at) tti - c (dot) org)
1113    Date: 2004/10/27 17:20:00 CDT
1114    Tag: blume-20041027-x86-c-calls
1115    Description:
1116    
1117    This is the HISTORY entry for two earlier commits, both concerning
1118    the x86 c-calls code in MLRISC:
1119    
1120      - added a missing LOAD in the code that deals with struct arguments
1121      - made sure the caller does not add the wrong number of bytes to the
1122        stack pointer after a call of a function returning a struct
1123        (the callee already pops the implicit argument which points to
1124         the space reserved for the result)
1125    
1126    ----------------------------------------------------------------------
1127    Name: Allen Leung (leunga (at) reservoir (dot) com)
1128    Date: 2004/10/24 14:00:00 EST
1129    Tag: leunga-20041024-x86-gas-fucomip
1130    Description:
1131    
1132       John discovered a bug in the syntax of fucomip.
1133       The opcodes FU?COMIP? have been changed to
1134    
1135           fu?comip? %st(i), %st
1136    
1137    ----------------------------------------------------------------------
1138    Name: Matthias Blume (blume (at) tti - c (dot) org)
1139    Date: 2004/10/20 15:06:00 CDT
1140    Tag: blume-20041020-standalone-backtrace
1141    Description:
1142    
1143    Added a mechanism for getting back-trace information from standalone
1144    programs.  Here is how it works:
1145    
1146       1. The part of the program from which you want to get backtrace
1147          information (usually the whole program) should be wrapped with
1148          BackTrace.monitor.  This is a (unit->'a)->'a function, and your
1149          main program could be modified from something like
1150    
1151             fun main (pgm, args) = ...
1152    
1153         to
1154    
1155            fun main (pgm, args) = BackTrace.monitor (fn () => ...)
1156    
1157       2. To be able to access BackTrace.monitor, you have to add
1158          library $smlnj-tdp/plugins.cm to the .cm file that contains your
1159          main function.
1160    
1161       3. Remove all compiled code (i.e., all the .cm/ subdirectories that
1162          CM might have created in the past for your project).
1163    
1164       4. Build the system using this command line:
1165    
1166             ml-build -Ctdp.instrument=true \$smlnj-tdp/back-trace.cm \
1167                      myprog.cm MyProg.main myprog
1168    
1169          instead of the usual
1170    
1171             ml-build myprog.cm MyProg.main myprog
1172    
1173    I changed a library name:
1174    
1175       $/trace-debug-profile.cm --> $smlnj-tdp/plugins.cm
1176    
1177    New libraries:
1178    
1179       $smlnj-tdp/back-trace.cm
1180          -- when loaded causes the back-trace plugin to be installed
1181       $smlnj-tdp/coverage.cm
1182          -- when loaded causes the coverage plugin to be installed
1183    
1184    ----------------------------------------------------------------------
1185    Name: Matthias Blume (blume (at) tti - c (dot) org)
1186    Date: 2004/10/18 16:45:00 CDT
1187    Tag: blume-20041018-groupowner
1188    Description:
1189    
1190    Added an "obsolete" warning for the "group owner" syntax to CM's parser.
1191    
1192    Eliminated group owner specs from .cm files throughout the source tree.
1193    
1194    ----------------------------------------------------------------------
1195    Name: Matthias Blume (blume (at) tti - c (dot) org)
1196    Date: 2004/10/15 15:45:00 CDT
1197    Tag: blume-20041015-coverage
1198    Description:
1199    
1200    * Test coverage tool added!
1201    * Further reorganization of tracing-, debugging-, and profiling support:
1202    
1203        - moved original BTImp -- now called BackTrace -- into a separate
1204          library called $/trace-debug-profile.cm
1205        - eliminated all mentions of BTrace from SMLofNJ.Internals
1206        - only the instrumentation mechanism is now left in the compiler proper
1207        - BackTrace module is a plugin which is NOT plugged in by default
1208        - Coverage module is another such plugin
1209    
1210    To get the benefits of any of these plugin modules, the code in
1211    question must be compiled with tdp instrumentation turned on.  This
1212    can be done by setting SMLofNJ.Internals.TDP.mode to true.
1213    (The ref cell is also controlled via the -Ctdp.instrument=... switch.)
1214    
1215    Plugins are selected at link time.  (Pre-compiled instrumented code
1216    can be re-loaded with different plugins in effect.)  When an
1217    instrumented module is linked, whatever plugins are at that time
1218    enabled will come into effect for that module.
1219    
1220    To enable the back-trace plugin, load library $/trace-debug-profile.cm
1221    and invoke BackTrace.install() (e.g., from the interactive prompt).
1222    To enable the coverage plugin, load the same library and invoke
1223    Coverage.install().
1224    
1225    Back-traces are generated automatically on uncaught exceptions and
1226    when the code in question explicitly invokes BackTrace.trigger().
1227    
1228    Coverage (and execution frequency-) information must be queried
1229    explicitly by calling Coverage.not_covered and Coverage.hot_spots.
1230    
1231    ----------------------------------------------------------------------
1232    Name: Matthias Blume (blume (at) tti - c (dot) org)
1233    Date: 2004/10/14 17:40:00 CDT
1234    Tag: blume-20041014-tdp-core
1235    Description:
1236    
1237    Snapshot of a significant overhaul of how the trace/debug/profile support
1238    is hooked into the system (specifically: Core and SMLofNJ.Internals).
1239    
1240    ----------------------------------------------------------------------
1241    Name: Matthias Blume (blume (at) tti - c (dot) org)
1242    Date: 2004/10/13 16:34:00 CDT
1243    Tag: blume-20041013-tdp
1244    Description:
1245    
1246    Some rationalization of names:
1247    
1248       structure BTrace -> structure TDPInstrument
1249       etc.
1250    
1251    This is is preparation of using the original back-trace
1252    instrumentation for other purposes.  "TDP" stands for
1253    Trace/Debug/Profile.
1254    
1255    The control flag controlling whether instrumentation is on or off is now
1256    registered under a different name, so instead of running sml as
1257    
1258     sml -Cinstrument.btrace-mode=true
1259    
1260    one has to say
1261    
1262     sml -Ctdp.instrument=true
1263    
1264    ----------------------------------------------------------------------
1265    Name: Matthias Blume (blume (at) tti - c (dot) org)
1266    Date: 2004/10/11 16:37:00 CDT
1267    Tag: blume-20041011-regions
1268    Description:
1269    
1270    Made some minor modifications to elabcore.sml to have source regions
1271    be propagated more tightly -- resulting in better (i.e., smaller)
1272    regions being reported in error- and debug messages.
1273    
1274    ----------------------------------------------------------------------
1275    Name: Matthias Blume (blume (at) tti - c (dot) org)
1276    Date: 2004/10/08 22:50:00 CDT
1277    Tag: blume-20041008-cmkw
1278    Description:
1279    
1280    Fixed handling of keywords in .cm files: After seeing "is" the lexer
1281    treats subsequent occurrences of "group", "library", "source", "is",
1282    "*", and "-" as ordinary identifiers rather than keywords.
1283    
1284    Most seriously, this fixes a problem with CM's "shell" tool.  The tool
1285    is supposed to accept a tool argument called "source", but this did
1286    not work because of the clash with the keyword.
1287    
1288    ----------------------------------------------------------------------
1289    Name: Matthias Blume (blume (at) tti - c (dot) org)
1290    Date: 2004/10/07 16:00:00 CDT
1291    Tag: blume-20041007-cleanup
1292    Description:
1293    
1294    Assorted cleanup work:
1295    
1296      - got rid of intstrmap in favor of using the library's
1297        hash table implementation
1298      - threw out most of the pathnames stuff, as it was not used anyway
1299      - simplified tokentable implementation
1300      - fixed some minor spelling errors
1301    
1302    ----------------------------------------------------------------------
1303    Name: Matthias Blume (blume (at) tti - c (dot) org)
1304    Date: 2004/10/06 15:15:15 CDT
1305    Tag: blume-20041006-handler
1306    Description:
1307    
1308    Cleaned up the absyn to reflect the invariant that HANDLE always
1309    carries a FNexp as part of the type definition.  This eliminates some
1310    superfluous sanity checks at runtime down the road.
1311    
1312    Some minor cleanup of the btrace code.
1313    
1314    ----------------------------------------------------------------------
1315    Name: Matthias Blume (blume (at) tti - c (dot) org)
1316    Date: 2004/10/01 10:20:30 CDT
1317    Tag: blume-20041001-slave
1318    Description:
1319    
1320    Added hack to make slave mode work in the presence of the version
1321    tool.  (Still, since the master does two passes over the code for
1322    CMB.make, the release number gets bumped twice when slaves are
1323    attached.  I don't know if this is worth fixing...)
1324    
1325    ----------------------------------------------------------------------
1326    Name: Matthias Blume (blume (at) tti - c (dot) org)
1327    Date: 2004/09/30 10:55:00 CDT
1328    Tag: blume-20040930-version
1329    Description:
1330    
1331    * Moved the "version" magic into its own little library under
1332      src/system/smlnj/internal.  This avoids expensive reconstruction of
1333      a stable src/compiler/core.cm.
1334    
1335    * At the same time, structure CompilerVersion is now known as
1336      structure SMLNJVersion.
1337    
1338    * Arranged for the version tool to NOT kick in when rebuilding the system
1339      (makeml -rebuild, fixpt).  Otherwise one would never reach a fixpoint.
1340      Also, loading the versiontool does not work when rebuilding the system
1341      because CM is not properly initialized at that time.
1342    
1343    ----------------------------------------------------------------------
1344    Name: Matthias Blume (blume (at) tti - c (dot) org)
1345    Date: 2004/09/29 14:00:00 CDT
1346    Tag: blume-20040929-autoversion
1347    Description:
1348    
1349    Implemented some CM magic to have
1350        file src/compiler/TopLevel/main/version.sml
1351    generated automagically.
1352    The version is taken from two files: config/version and config/release.
1353    The first is expected to contain a two-part version number such as 110.49.
1354    The second should contain a single number, but it may be missing.
1355    
1356    If the environment variable VERSIONTOOL_BUMP_RELEASE is defined at the
1357    time the version tool is loaded (which is the first time you say CMB.make),
1358    then the tool will increment the value stored in config/release every
1359    time CMB.make is invoked.
1360    
1361    The binfile format is now insensitive to anything beyond the first
1362    two components of a version number, so bumping the release does not render
1363    binfiles incompatible.  Auto-bumping can be used to keep track of versions
1364    during development without invalidating existing binfiles.
1365    
1366    In any case, every CMB.make updates the date information in version.sml.
1367    (This is the date that is printed in the banner.)
1368    
1369    ----------------------------------------------------------------------
1370    Name: Matthias Blume (blume (at) tti - c (dot) org)
1371    Date: 2004/09/28 10:53:00 CDT
1372    Tag: blume-20040928-controls
1373    Description:
1374    
1375    Some cleanup of the controls code.
1376    
1377    ----------------------------------------------------------------------
1378    Name: Matthias Blume (blume (at) tti - c (dot) org)
1379    Date: 2004/09/27 22:08:00 CDT
1380    Tag: blume-20040927-controls
1381    Description:
1382    
1383    Added two pieces of functionality to the Controls interface:
1384    
1385    1.   val save'restore: 'a control -> unit -> unit
1386    
1387       grabs the current value of the control in stage 1 and restores it
1388       in stage 2.
1389    
1390    2.   val set' : 'a control * 'a -> unit -> unit
1391    
1392       stores the given value into the control in stage 2 (i.e., delayed)
1393       but does all error checking in stage 1.
1394       (This is for string controls that need to do parse their argument --
1395        something that might fail.  In some cases, notably in CM, one
1396        already knows the intended argument but wants to delay the actual
1397        assignment until a time when error recovery would be more difficult.)
1398    
1399    Changed the handling of controls in tool arguments to classes "sml" and
1400    "lazysml":
1401       - use Controls.save'restore as a more robust way of restoring the
1402         old value (in particular: without having to re-parse the string)
1403       - use controls to handle the "overload" keyword in the init group
1404         (I believe this change actually fixes a long-standing obscure bug.)
1405    
1406    ----------------------------------------------------------------------
1407    Name: Matthias Blume (blume (at) tti - c (dot) org)
1408    Date: 2004/09/27 17:00:00 CDT
1409    Tag: blume-20040927-lazysml
1410    Description:
1411    
1412    Added a new tool class called "lazysml" to CM's tool chest.  The only
1413    difference to "sml" is that compilation is done with Control.lazysml
1414    set to true.  A source of class "lazysml" is automatically recognized
1415    by a file name suffix of ".lml".
1416    
1417    In addition to the above feature, the original class "sml" now also
1418    supports a tool argument "lazy" which has the same effect.  As a
1419    result, the following three lines are equivalent:
1420    
1421        foo.sml : lazysml
1422        foo.sml : sml (lazy)
1423        foo.sml (lazy)
1424    
1425    The setting goes into effect both during parsing and during
1426    compilation.  The original setting is restored right after parsing and
1427    after compilation, respectively.
1428    
1429    In addition to all the above, there is also a general mechanism to set
1430    ANY of the "controls" that are available at the command line via
1431    "-C..." on a per-sml-file basis.  The same rules that apply for "lazy"
1432    apply as well. (In fact, "lazy" is implemented as a special case of
1433    the general mechanism.)
1434    
1435    The .cm file syntax uses a new keyword tool argument called "with".
1436    There are several ways of indicating the desired settings:
1437    
1438        foo.sml (with:parser.quotations=true)
1439        foo.sml (with:(name:parser.quotations value:true))
1440        foo.sml (with:(name:name1 value:value1 name:name2 value:value2 ...))
1441        foo.sml (with:(name1=value1 name2=value2 ...))
1442        foo.sml (with:(name1=value1 name:name2 value:value2 name3=value3 ...))
1443    
1444        etc.
1445    
1446    Another possible abbreviation is to leave out the =v or value:v part
1447    if the name refers to a boolean control (in which case the value is
1448    taken to be true).  Thus, one could get lazy sml also by saying:
1449    
1450        foo.sml (with:parser.lazy-keyword=true)
1451        foo.sml (with:parser.lazy-keyword)
1452        foo.sml (with:(name:parser.lazy-keyword value:true))
1453        foo.sml (with:(name:parser.lazy-keyword))
1454    
1455    ----------------------------------------------------------------------
1456    Name: Matthias Blume (blume (at) tti - c (dot) org)
1457    Date: 2004/09/24 16:22:00 CDT
1458    Tag: blume-20040924-ppc-long-branch
1459    Description:
1460    
1461    Turned message about "emiting long form of branch" off by default.
1462    Added a control flag to turn it back on when desired.
1463    
1464    ----------------------------------------------------------------------
1465    Name: Matthias Blume (blume (at) tti - c (dot) org)
1466    Date: 2004/09/24 16:05:00 CDT
1467    Tag: blume-20040924-rounding
1468    Description:
1469    
1470    Applied patch for setting rounding modes under Mac OS X.  Thanks to
1471    Melissa O'Neill for providing the code!
1472    
1473    ----------------------------------------------------------------------
1474    Name: Matthias Blume (blume (at) tti - c (dot) org)
1475    Date: 2004/09/23 17:30:00 CDT
1476    Tag: blume-20040923-envvars
1477    Description:
1478    
1479    1. Changed definition of type ControlRegistry.registry_tree to
1480       include control_info (i.e., the name of the controlling environment
1481       variable).
1482    
1483    2. Added command-line flags -e and -E to print the names of environment
1484       variables that can be used to control internal settings.
1485       (This uses the new API mentioned in 1.)
1486    
1487    ----------------------------------------------------------------------
1488    Name: Matthias Blume (blume (at) tti - c (dot) org)
1489    Date: 2004/09/13 16:50:00 CDT
1490    Tag: Release_110_49
1491    Description:
1492    
1493    New working version (110.49).  NEW BOOTFILES!
1494    
1495    ----------------------------------------------------------------------
1496    Name: Matthias Blume
1497    Date: 2004/09/13 16:20:00 CDT
1498    Tag: blume-20040913-config-mlrisc
1499    Description:
1500    
1501    Put target "mlrisc" back into the default list.
1502    (There is no harm in having it, and some users have expressed their
1503    wish to have "mlrisc" included by default.)
1504    
1505    ----------------------------------------------------------------------
1506    Name: John Reppy
1507    Date: 2004/09/13
1508    Tag: jhr-20040913-signals
1509    Description:
1510    
1511    Fixed the signal masking code to properly nest mask/unmask operations
1512    on a per-signal basis.
1513    
1514    ----------------------------------------------------------------------
1515    Name: Matthias Blume (blume (at) tti - c (dot) org)
1516    Date: 2004/09/08 13:20:00 CDT
1517    Tag: blume-20040908-heap-magic
1518    Description:
1519    
1520    Bumped the heap macig number to 0x09082004 to account for the changed
1521    layout of the ML frame under MacOS X.
1522    
1523    ----------------------------------------------------------------------
1524    Name: Allen Leung (leunga (at) reservoir (dot) com)
1525    Date: 2004/09/03 11:26:00 EST
1526    Tag: leunga-20040903-cygwin-install
1527    Description:
1528    
1529       Added a patch to _arch-n-opsys to enable the Cygwin runtime.
1530    The Cygwin runtime is turned on by setting the environment
1531    variable SMLNJ_CYGWIN_RUNTIME to 1.
1532    
1533    ----------------------------------------------------------------------
1534    Name: Matthias Blume (blume (at) tti - c (dot) org)
1535    Date: 2004/08/31 17:15:00 CDT
1536    Tag: blume-20040831-core
1537    Description:
1538    
1539    Added some exports to src/compiler/core.cm upon request by J. Joemann.
1540    
1541    ----------------------------------------------------------------------
1542    Name: Matthias Blume (blume (at) tti - c (dot) org)
1543    Date: 2004/08/30 17:55:00 CDT
1544    Tag: blume-20040830-installer
1545    Description:
1546    
1547    Upon request by Johannes Joemann:
1548    
1549    - improved ML code of installer to fall back to coping when renaming
1550      fails (i.e., when source and target are on different file systems);
1551      the code compiles but has yet to be tested in anger
1552    - removed mlrisc from list of default targets (config/targets)
1553    
1554    ----------------------------------------------------------------------
1555    Name: Matthias Blume (blume (at) tti - c (dot) org)
1556    Date: 2004/08/27 17:20:00 CDT
1557    Tag: blume-20040827-ptreql
1558    Description:
1559    
1560    Added ptreql primop to structure InlineT (upon request from Larry
1561    Paulson).
1562    
1563    ----------------------------------------------------------------------
1564    Name: Allen Leung (leunga (at) reservoir (dot) com)
1565    Date: 2004/08/15 21:21:00 EST
1566    Tag: leunga-110_48-udgraph
1567    Description:
1568    
1569    Another bug fix from Carl Hauser:
1570    
1571    diff /net/niflab/smlnj48/src/MLRISC/graphs/udgraph.sml udgraph.sml
1572    > 48c48
1573    > <              | rmv((e as (k,_))::es,L) = rmv(es,if k = i then es else
1574    > e::L)
1575    > ---
1576    > >              | rmv((e as (k,_))::es,L) = rmv(es,if k = i then L else e::L)
1577    > Without this, any deletion of an edge in an undirected graph does severe
1578    > violence to the graph.
1579    
1580    ----------------------------------------------------------------------
1581    Name: Allen Leung (leunga (at) reservoir (dot) com)
1582    Date: 2004/08/10 23:35:00 EST
1583    Tag: leunga-110_48-ppc
1584    Description:
1585    
1586       The IBM/MacOS syntax switch on PPC was incorrectly swapped.
1587    
1588    ----------------------------------------------------------------------
1589    Name: Matthias Blume (blume (at) tti - c (dot) org)
1590    Date: 2004/08/10 12:00:00 CDT
1591    Tag: Release_110_48
1592    Description:
1593    
1594    New working version (110.48).  NEW BOOTFILES!
1595    
1596    ----------------------------------------------------------------------
1597    Name: Allen Leung (leunga (at) reservoir (dot) com)
1598    Date: 2004/08/09 12:21:00 EST
1599    Tag: leunga-110_47-dijsktra
1600    Description:
1601    
1602    Bug fix from Carl Hauser:
1603    
1604    single_source_shortest_paths in dijkstra.sml was observed to get wrong
1605    answers (by comparing to single_source_shortest_paths in bellman-ford.sml).
1606    
1607    The problem is that following the expression A.update(dist,s,Num.zero)
1608    it is necessary to update the priority queue using Q.decreaseWeight(Q,s).
1609    
1610    ----------------------------------------------------------------------
1611    Name: Matthias Blume (blume (at) tti - c (dot) org)
1612    Date: 2004/08/06 18:10:00 CDT
1613    Tag: blume-20040806-cmdline
1614    Description:
1615    
1616    Fiddled with handling of command-line options:
1617    
1618      * sml now quits after processing the command line
1619        if -H, -S, -h<n>, or -s<n> appears as the last
1620        command-line argument
1621      * a new option -q terminates the session when encountered on
1622        the command line; subsequent arguments will be ignored
1623      * bug fixes: short (erroneous) arguments are no longer ignored
1624        completely
1625    
1626    ----------------------------------------------------------------------
1627    Name: Allen Leung (leunga (at) reservoir (dot) com)
1628    Date: 2004/08/04 18:17:00 EST
1629    Tag: leunga-110_47-ppc-ibm-asm
1630    Description:
1631    
1632      - Added minimal IBM assembly syntax support for PowerPC.
1633    
1634      - Cygwin: manually changed the file cygwin.def.  Some exported symbols have
1635        been altered in the runtime.  We need an automatic way to keep the file
1636        in sync.
1637    
1638    ----------------------------------------------------------------------
1639    Name: Matthias Blume (blume (at) tti - c (dot) org)
1640    Date: 2004/08/04 14:00:00 CDT
1641    Tag: Release_110_47
1642    Description:
1643    
1644    New working version (110.47).  NEW BOOTFILES!
1645    
1646    ----------------------------------------------------------------------
1647    Name: Matthias Blume (blume (at) tti - c (dot) org)
1648    Date: 2004/08/03 14:25:00 CDT
1649    Tag: blume-20040803-callingconv
1650    Description:
1651    
1652    Added low-level support for choosing C calling conventions by
1653    twiddling the type of rawccall.  (See
1654    src/compiler/Semant/types/cproto.sml for details.)
1655    
1656    ----------------------------------------------------------------------
1657    Name: Matthias Blume (blume (at) tti - c (dot) org)
1658    Date: 2004/08/02 15:55:00 CDT
1659    Tag: blume-20040802-backout
1660    Description:
1661    
1662    Backed out of change to win32-filesys.c.  The earlier patch to
1663    get_file_time caused CM to produce files with the wrong time stamp.
1664    
1665    ----------------------------------------------------------------------
1666    Name: Matthias Blume (blume (at) tti - c (dot) org)
1667    Date: 2004/08/02 14:45:00 CDT
1668    Tag: blume-20040802-nlffi-win32
1669    Description:
1670    
1671    Added NLFFI support for Win32, adapted from a patch provided by David
1672    Hansel.  This is currently completely untested.  Also, the issue
1673    concerning stdcall vs. ccall is still unresolved.
1674    
1675    ----------------------------------------------------------------------
1676    Name: Matthias Blume (blume (at) tti - c (dot) org)
1677    Date: 2004/07/30 17:55:00 CDT
1678    Tag: blume-20040730-various
1679    Description:
1680    
1681    Gearing up towards 110.47...
1682    
1683    - various minor bugfixes to ml-nlffigen
1684    - a beginning of a manual for nlffi
1685    
1686    - eliminated 'export name=value' in config/install.sh as this does
1687      not work with certain versions of /bin/sh
1688      (Thanks to David King at Motorola for catching this.)
1689    
1690    - several bugfixes provided or suggested by David Hansel at Reactive Systems:
1691      - added a test for tm==NULL to gmtime.c and localtime.c
1692      - applied patch for incorrect GetFileTime under win32
1693      - toSeconds -> toMilliseconds in Win32/win32-process.sml
1694    
1695    ----------------------------------------------------------------------
1696    Name: Matthias Blume (blume (at) tti - c (dot) org)
1697    Date: 2004/07/21 18:20:00 CDT
1698    Tag: blume-20040721-nlffigen
1699    Description:
1700    
1701    - Fixed minor issue in ml-nlffigen: Now generate structure T_foo
1702      for a typedef to an incomplete type, but leave out the "typ" member.
1703      (This is just for consistency.)
1704    - Started to produce what is supposed to become better (i.e., comprehensive)
1705      documentation of what ml-nlffigen does and produces.
1706    
1707    ----------------------------------------------------------------------
1708    Name: Matthias Blume (blume (at) tti - c (dot) org)
1709    Date: 2004/07/14 16:25:00 CDT
1710    Tag: blume-20040714-union
1711    Description:
1712    
1713    Added C_UNION to c-calls/c-types.sml and updated the machinery
1714    (ml-nlffigen, cproto.sml) that conveys C function interface
1715    information to the code generator.
1716    
1717    However, the actual architecture-specific implementation of function
1718    arguments and results that are C unions is still not implemented.
1719    
1720    ----------------------------------------------------------------------
1721    Name: Allen Leung (leunga (at) reservoir (dot) com)
1722    Date: 2004/07/14 14:38:00 EST
1723    Tag: leunga-110_46_1-ppc-lwzu
1724    Description:
1725    
1726        Added these instructions to the PowerPC architecture:
1727        LBZU(X), LHZU(X), LWZU(X),
1728        STWU(X), STFDU, STFSU
1729    
1730        etc...
1731    
1732        Note: I haven't added their instruction encoding into the description.
1733    
1734    ----------------------------------------------------------------------
1735    Name: Allen Leung (leunga (at) reservoir (dot) com)
1736    Date: 2004/07/13 15:04:00 EST
1737    Tag: leunga-110_46_1-ppc-lwarx
1738    Description:
1739    
1740       Added the two instructions LWARX and STWCX to the PowerPC
1741    instruction set.
1742    
1743       A (untested) rewrite of loop-structure.sml.  The old version
1744    is completely broken.
1745    
1746    ----------------------------------------------------------------------
1747    Name: Matthias Blume (blume (at) tti - c (dot) org)
1748    Date: 2004/07/13 13:50:00 CDT
1749    Tag: blume-20040713-nlffi
1750    Description:
1751    
1752    - use paramAlloc to report c-calls with too many arguments
1753      (for PPC version where parameter area is pre-allocated)
1754    - added ccall_maxargspace to machspec (to implement the above)
1755    - made "make" commend in CM's "make" tool configurable
1756    - added option (default: on) for passing the name of the SML/NJ's "bin"
1757      directory to "make"; the call looks like this:
1758    
1759         make <options> SMLNJ_BINDIR=<dir> <target>
1760    
1761      This can be used by the Makefile to, e.g., pick the "right" version
1762      of ml-nlffigen.
1763    - minor code tweaks
1764    
1765    ----------------------------------------------------------------------
1766    Name: Matthias Blume (blume (at) tti - c (dot) org)
1767    Date: 2004/07/12 22:50:00 CDT
1768    Tag: blume-110_46_1-macosx-nlffi
1769    Description:
1770    
1771    NLFFI under Mac OS X now working (sort of).  This is largely untested,
1772    though.
1773    
1774    Note:  1. You have to make a new, clean build of the runtime system.
1775           2. There are new BOOTFILES, you have to use them!
1776              (Doing the bootstrap process yourself would be *very* painful!
1777               If you absolutely have to do it, build the system under
1778               a different architecture and then cross-compile.)
1779    
1780    Version bumped to 110.46.1 to account for runtime data format changes.
1781    
1782    ----------------------------------------------------------------------
1783    Name: Matthias Blume (blume (at) tti - c (dot) org)
1784    Date: 2004/06/18 14:30:00 CDT
1785    Tag: blume-20040618-unix
1786    Description:
1787    
1788    Changed the implementation of structure Unix so that the same stream
1789    is returned every time one of the {text,bin}{In,Out}streamOf functions
1790    is invoked on the same proc.  This is not what the spec currently
1791    says -- although IMO it arguably should.  (See discussion below.)
1792    
1793    ----------------------------------------------------------------------
1794    Name: Matthias Blume (blume (at) tti - c (dot) org)
1795    Date: 2004/06/17 18:15:00 CDT
1796    Tag: Release_110_46
1797    Description:
1798    
1799    New working version (110.46).  NEW BOOTFILES!
1800    
1801    ----------------------------------------------------------------------
1802    Name: Matthias Blume (blume (at) tti - c (dot) org)
1803    Date: 2004/06/17 17:20:00 CDT
1804    Tag: blume-20040617-timer-unix
1805    Description:
1806    
1807    Changed the interface of structures Timer and Unix to match the most
1808    recent Basis spec.
1809    
1810    In the case of Unix there still seems to be an open/weird issue:
1811    
1812       The {text,bin}{In,Out}streamOf functions are supposed to create
1813       fresh streams whenever they are called -- as opposed to have them
1814       return the same stream every time.  This design is supposed to
1815       prevent space leaks caused by proc values hanging on to streams.
1816    
1817       The reap function, on the other hand, is supposed to close the
1818       streams.  This cannot be done without having a handle on the
1819       stream in proc after all...
1820    
1821    I took the liberty to implement the following stopgap solution:
1822    
1823      The proc value hangs on to the most recently created stream(s).
1824      Reap closes those.  If either or both of the two streams hadn't
1825      been created at all yet, then reap will close the corresponding
1826      file descriptors directly.
1827    
1828    PS: I don't understand the original space leak argument anymore.  If
1829    a proc hangs on to the imperative stream, then I/O operations on those
1830    will advance the state of the cached stream and avoid the space leak.
1831    
1832    ----------------------------------------------------------------------
1833    Name: Matthias Blume (blume (at) tti - c (dot) org)
1834    Date: 2004/05/28 16:45:00 CDT
1835    Tag: blume-20040528-basis
1836    Description:
1837    
1838    Added signature PACK_REAL and exported functor PrimIO.
1839    
1840    ----------------------------------------------------------------------
1841    Name: Matthias Blume (blume (at) tti - c (dot) org)
1842    Date: 2004/05/25 16:00:00 CDT
1843    Tag: blume-20040525-group-owner
1844    Description:
1845    
1846    CM now ignores (but still accepts) the "owner" information in group
1847    descriptions.  The owner of a group is its next enclosing
1848    library. Each group must have a unique owner.  (There is a virtual
1849    "toplevel" library that owns groups which are not nested within a real
1850    library.)  Previously, each group had to explicitly declare its owner,
1851    and CM would check that such a declaration is correct.  The new scheme
1852    is to have CM check that for each group there is precisely one owning
1853    library.
1854    
1855    The advantage of the new scheme is that the programmer no longer needs
1856    to maintain the somewhat annoying owner information.  The downside is
1857    that CM cannot enforce the ownership rule across multiple runs of
1858    CM.make.  Fortunately, enclosing the same group in two different
1859    libraries A and B which are not part of the same program does not
1860    cause real problems.
1861    
1862    ----------------------------------------------------------------------
1863    Name: Matthias Blume (blume (at) tti - c (dot) org)
1864    Date: 2004/05/20 16:00:00 CDT
1865    Tag: blume-20040520-win32
1866    Description:
1867    
1868    Made the win32 version work again.  (Strangely, a misplaced comma had
1869    slipped into win32-process.c which prevented the runtime from being
1870    compiled correctly.)
1871    
1872    Also, included a minor addition to ml-build.bat analogous to what was
1873    done in blume-20040519-ml-build.
1874    
1875    ----------------------------------------------------------------------
1876    Name: Matthias Blume (blume (at) tti - c (dot) org)
1877    Date: 2004/05/19 22:10:00 CDT
1878    Tag: blume-20040519-ml-build
1879    Description:
1880    
1881    Arranged for ml-build to clean up after itself a little bit better.
1882    The script generates a temporary SML source file and compiles it using
1883    CM, so CM generates metadata (GUID, SKEL, objectfile) for it.  It now
1884    gets rid of those at the end, so they don't accumulate under .cm.
1885    
1886    This required a minor change to install.sh because the name of the
1887    metadata directory (default: .cm) is actually configurable at
1888    installation time.
1889    
1890    ----------------------------------------------------------------------
1891    Name: Matthias Blume (blume (at) tti - c (dot) org)
1892    Date: 2004/05/18 15:50:00 CDT
1893    Tag: blume-20040518-mkreader
1894    Description:
1895    
1896    Added Posix.IO.mk{Bin,Text}{Reader,Writer} by lifting their respective
1897    implementations from internal modules PosixBinPrimIO and PosixTextPrimIO.
1898    
1899    ----------------------------------------------------------------------
1900    Name: Matthias Blume (blume (at) tti - c (dot) org)
1901    Date: 2004/05/11 14:35:00 CDT
1902    Tag: blume-20040511-win32sock
1903    Description:
1904    
1905    Added previously missing support for many socket-related functions
1906    under win32.  Thanks to David Hansel <hansel@reactive-systems.com>
1907    for the voluminous patch!
1908    
1909    (I have not tested this patch under win32 yet.)
1910    
1911    Here is David's e-mail:
1912    
1913    Hi,
1914    
1915    Attached to this email you find a diff against sml/nj 110.45
1916    that will enable socket support under Windows.
1917    
1918    To apply the patch (using unix or cygwin)
1919    1) gunzip runtime.diff.gz
1920    2) "cd" into "src/runtime" in the source tree of a fresh
1921       110.45 installation.
1922    3) patch -p 1 < [your/path/to]runtime.diff
1923    
1924    The code compiles fine but has NOT yet been extensively tested.
1925    I only ran a few tests for basic socket client functionality
1926    (which worked fine).  Especially the functions that use ioctl
1927    are not tested at all and might not work (see below).
1928    
1929    I implemented this since we want to move to a newer version of sml/nj
1930    but need socket support in order to use it.  This is the first time I
1931    even had a look at the sml/nj source,  so please review my changes
1932    before making this part of the distribution!  Here are a few issues
1933    that I think might be better for someone to solve who is more
1934    familiar with the sml/nj source (and socket programming):
1935    
1936    - getnetbyaddr.c and getnetbyname.c will raise a "not implemented"
1937      exception since I could not figure out what the windows equivalent
1938      of these functions is
1939    
1940    - In sockets-osdep.h there are a some #include statements that are
1941      only used in a few files that include sockets-osdep.h
1942    
1943    - In smlnj-sock-lib.c, function init_fn() calls WSAStartup() but
1944      does not process its return value since I don't know how to
1945      report an error upwards.
1946    
1947    - It would probably be good to have a call to WSACleanup() when
1948      the library is unloaded (if there is such a possibility).
1949      Otherwise I think Windows will take care of this automatically
1950      when the process finishes.
1951    
1952    - I used ioctlsocket() as a replacement for ioctl() but I have
1953      no idea if that is actually the proper replacement on Windows.
1954    
1955    - All these issues are marked in the code by "FIXME" comments.
1956    
1957    We use sml/nj extensively in our products and are quite happy
1958    with it.  I hope this contribution will help you.
1959    
1960    Keep up the good work!
1961    
1962    David
1963    
1964    ----------------------------------------------------------------------
1965    Name: Matthias Blume (blume (at) tti - c (dot) org)
1966    Date: 2004/05/11 14:20:00 CDT
1967    Tag: blume-20040511-installml
1968    Description:
1969    
1970    Fixed two bugs in installml script.  (Thanks to Vesa A. Norrman for
1971    the patch.)
1972    
1973    ----------------------------------------------------------------------
1974    Name: Matthias Blume (blume (at) tti - c (dot) org)
1975    Date: 2004/05/11 14:05:00 CDT
1976    Tag: blume-20040511-nlffi-netbsd
1977    Description:
1978    
1979    Added support for nlffi under netbsd.  (Thanks to Vesa A. Norrman for
1980    the patch.)
1981    
1982    ----------------------------------------------------------------------
1983    Name: Matthias Blume (blume (at) tti - c (dot) org)
1984    Date: 2004/05/11 12:05:00 CDT
1985    Tag: blume-20040511-exports
1986    Description:
1987    
1988    As per request by Adam Chlipala <adam@hcoop.net>, extended various
1989    export lists in compiler-related .cm-files.
1990    
1991    ----------------------------------------------------------------------
1992    Name: Matthias Blume (blume (at) tti - c (dot) org)
1993    Date: 2004/05/11 11:35:00 CDT
1994    Tag: blume-20040511-allsource
1995    Description:
1996    
1997    The installer now honors the "src-smlnj" target again, although its meaning
1998    has changed from "all sources required for the compiler" to "all sources
1999    the installer knows about".  In other words, if you enable "src-smlnj"
2000    in the "targets" file, then the installer will pull in sources for
2001    everything.  (Notice that this refers to source code only.  Compiled
2002    code is still only installed for modules that were requested explicitly
2003    or which are required for other modules that were requested explicitly.)
2004    
2005    ----------------------------------------------------------------------
2006    Name: Matthias Blume (blume (at) tti - c (dot) org)
2007    Date: 2004/04/23 17:40:00 CDT
2008    Tag: blume-20040423-ieee-scan
2009    Description:
2010    
2011    Fixed IEEEReal.scan (and .fromString) so that if there is an overflow
2012    in the exponent calculation we get INF or ZERO (depending on the mantissa
2013    and the sign of the exponent).
2014    
2015    ----------------------------------------------------------------------
2016    Name: Matthias Blume (blume (at) tti - c (dot) org)
2017    Date: 2004/04/23 10:40:00 CDT
2018    Tag: blume-20040423-ml-build
2019    Description:
2020    
2021    The ml-build script now terminates with a non-0 status when something
2022    goes wrong.
2023    
2024    ----------------------------------------------------------------------
2025    Name: Matthias Blume (blume (at) tti - c (dot) org)
2026    Date: 2004/04/22 16:35:00 CDT
2027    Tag: blume-20040422-Option
2028    Description:
2029    
2030    Made exception Option to be the same as exception Option.Option
2031    (as it should be).
2032    
2033    ----------------------------------------------------------------------
2034    Name: Allen Leung (leunga (at) reservoir (dot) com)
2035    Date: 2004/03/19 14:40:00 EST
2036    Tag: leunga-20040319-cygwin-nlffi
2037    Description:
2038    
2039    Fixed the runtime so that ml-nlffi-lib runs on the cygwin version
2040    of SML/NJ.  The problem is that
2041    
2042       lib = dlopen(NULL, ...)
2043       f   = dlsym(lib, "malloc");
2044    
2045    does not work on Windows unless we explicitly export symbols
2046    such as 'malloc' during linking.  We fixed this by explicitly
2047    exporting the required symbols with the magic gcc incantation:
2048    
2049        -Wl,--export-all cygwin.def
2050    
2051    where cygwin.def is a file containing all the symbols that
2052    we wish to export.
2053    
2054    I suspect this is a Windows problem and we'll have to
2055    do the same (somehow with windows compilers) when
2056    we build the native win32 version with the system
2057    calls LoadLibrary/GetProcAddress.
2058    
2059    ----------------------------------------------------------------------
2060    Name: Matthias Blume (blume (at) tti - c (dot) org)
2061    Date: 2004/03/04 16:35:00 CST
2062    Tag: blume-20040304-intinf-fmt
2063    Description:
2064    
2065    Fixed problem with IntInf.fmt (sign would show up on the right instead
2066    of on the left for BIN, OCT, and HEX).
2067    
2068    ----------------------------------------------------------------------
2069    Name: Matthias Blume (blume (at) tti - c (dot) org)
2070    Date: 2004/03/04 11:25:00 CST
2071    Tag: blume-20040304-symlinks
2072    Description:
2073    
2074    Fixed problem with installer script (unix only) where bin/ml-yacc and
2075    friends pointed (via symlinks) to absolute locations instead of just
2076    .run-sml.  This was reported by Vesa A Norrman.
2077    
2078    ----------------------------------------------------------------------
2079    Name: Matthias Blume (blume (at) tti - c (dot) org)
2080    Date: 2004/02/13 14:50:00 CST
2081    Tag: Release_110_45
2082    Description:
2083    
2084    New working version (110.45). New bootfiles.
2085    
2086    ----------------------------------------------------------------------
2087    Name: Matthias Blume (blume (at) tti - c (dot) org)
2088    Date: 2004/01/26 15:15:15 CST
2089    Tag: blume-20040126-toplevel
2090    Description:
2091    
2092    Improved handling of exceptions at the interactive toplevel.
2093    
2094    ----------------------------------------------------------------------
2095    Name: Matthias Blume (blume (at) tti - c (dot) org)
2096    Date: 2004/01/26 11:25:00
2097    Tag: blume-20040126-app
2098    Description:
2099    
2100    Type of top-level "app" corrected.
2101    Added code for setting vp_limitPtrMask to Win32-specific runtime.
2102    
2103    ----------------------------------------------------------------------
2104    Name: Matthias Blume (blume (at) tti - c (dot) org)
2105    Date: 2003/11/18 17:10 CST
2106    Tag: blume-20031118-basis-fiddle
2107    Description:
2108    
2109    - changed Timer interface to what might become the spec
2110    - POSIX_FLAGS -> BIT_FLAGS according to spec
2111    - some other minor discrepancies wrt. spec eliminated
2112    
2113    ----------------------------------------------------------------------
2114    Name: Matthias Blume (blume (at) tti - c (dot) org)
2115    Date: 2003/11/06 12:00:00 CST
2116    Tag: Release_110_44
2117    Description:
2118    
2119    New working version (110.44). New bootfiles.
2120    
2121    ----------------------------------------------------------------------
2122    Name: Matthias Blume (blume (at) tti - c (dot) org)
2123    Date: 2003/11/04 11:50:00 CST
2124    Tag: blume-20031104-move-libraries
2125    Description:
2126    
2127    Eliminated the "dont_move_libraries" directive in config/targets.
2128    (The mechanism was broken and could not be fixed easily.  Moreover,
2129    there does not seem to be any reason not to move all libraries into
2130    lib during installation.  I originally implemented this directive as a
2131    backward-compatibility feature when I first introduced the new CM.
2132    Now that things have been stable for a long time and going back to the
2133    old CM is not an option, there is no reason to keep it around.)
2134    
2135    ----------------------------------------------------------------------
2136    Name: Matthias Blume (blume (at) tti - c (dot) org)
2137    Date: 2003/11/03 16:00:00 CST
2138    Tag: blume-20031103-installdir
2139    Description:
2140    
2141    Made installer honor INSTALLDIR variable again.  (Thanks to Chris
2142    Richards for pointing out the problem and providing the solution.)
2143    
2144    ----------------------------------------------------------------------
2145    Name: Matthias Blume (blume (at) tti - c (dot) org)
2146    Date: 2003/10/01 17:05:00 CDT
2147    Tag: blume-20031001-lal-mlrisc
2148    Description:
2149    
2150    MLRISC bug fix from Lal.
2151    
2152    ----------------------------------------------------------------------
2153    Name: Matthias Blume (blume (at) tti - c (dot) org)
2154    Date: 2003/09/30 16:10:00 CDT
2155    Tag: blume-20030930-primio-bat
2156    Description:
2157    
2158    1. Added openVector, nullRd, and nullWr to PRIM_IO.
2159    2. Improved .bat files (for Win32 port) to make things work under Win95.
2160       (thanks to Aaron S. Hawley for this one)
2161    
2162    ----------------------------------------------------------------------
2163    Name: Matthias Blume (blume (at) tti - c (dot) org)
2164    Date: 2003/09/26 16:05:00 CDT
2165    Tag: blume-20030926-wrappriv
2166    Description:
2167    
2168    Added missing wrapper for privilege "primitive" in $smlnj/viscomp/core.cm.
2169    
2170    ----------------------------------------------------------------------
2171    Name: Matthias Blume (blume (at) tti - c (dot) org)
2172    Date: 2003/09/26 15:00:00 CDT
2173    Tag: blume-20030926-110_43_3
2174    Description:
2175    
2176    - additional cleanup
2177    - version number bump, NEW BOOTFILES
2178    
2179    ----------------------------------------------------------------------
2180    Name: Matthias Blume (blume (at) tti - c (dot) org)
2181    Date: 2003/09/26 12:00:00 CDT
2182    Tag: blume-20030926-ppautoload
2183    Description:
2184    
2185    I modified the read-eval-print loop so that the autoloader gets
2186    invoked whenever the prettyprinter tries to look up a symbol that
2187    is not currently defined in the toplevel environment but which
2188    appears in CM's autoload registry.  As a result, we see far fewer of
2189    those ?.Foo.Bar.xxx names in the prettyprinter's output.
2190    
2191    In addition to this I tried to clean up some pieces of the Basis
2192    implementation (e.g., Socket, Word8Array) in order to prevent other
2193    instances of these ?.Foo.Bar.xxx names from being printed.
2194    
2195    The mechanism that picks names for types still needs some work, though.
2196    (Right now it seems that if there is a type A.t which is defined to
2197    be B.u, but B is unavailable at toplevel, then A.t gets printed as
2198    "?.B.u" although the perhaps more sensible solution would be to use
2199    "A.t" in this case.  In other words, the prettyprinter should follow
2200    a chain of DEFtycs not farther than there are corresponding toplevel
2201    names in the current environment.)
2202    
2203    ----------------------------------------------------------------------
2204    Name: Matthias Blume (blume (at) tti - c (dot) org)
2205    Date: 2003/09/24 16:31:00 CDT
2206    Tag: blume-20030924-installer
2207    Description:
2208    
2209    Another installer tweak:  All the ML code for the installer is now
2210    compiled during CMB.make and put into a little library called
2211    $smlnj/installer.cm.  The installation then simply invokes
2212    
2213       sml -m $smlnj/installer.cm
2214    
2215    and everything happens automagically.
2216    
2217    Win32:  ML code senses value of environment variable SMLNJ_HOME.
2218    Unix:   ML code senses values of environment variables ROOT, CONFIGDIR,
2219            and BINDIR.
2220    
2221    The new scheme guarantees that the ML code responsible for the installation
2222    is in sync with the APIs of the main system.  Also, the installer is
2223    somewhat faster because the installer script is precompiled.
2224    
2225    ----------------------------------------------------------------------
2226    Name: Matthias Blume (blume (at) tti - c (dot) org)
2227    Date: 2003/09/24 15:35:00 CDT
2228    Tag: blume-20030924-synsock
2229    Description:
2230    
2231    Added a signature SYNCHRONOUS_SOCKET to basis.cm.  This is like SOCKET
2232    but excludes all non-blocking operations.  Defined SOCKET (in Basis)
2233    and CML_SOCKET in terms of SYNCHRONOUS_SOCKET.  Removed superfluous
2234    implementations of non-blocking operations from CML's Socket
2235    structure.
2236    
2237    ----------------------------------------------------------------------
2238    Name: Matthias Blume (blume (at) tti - c (dot) org)
2239    Date: 2003/09/24 15:10:05 CDT
2240    Tag: blume-20030924-sockets
2241    Description:
2242    
2243    1. Fixed SOCKET API and implementation to match Basis spec.
2244       This required changing the internal representation of sockets to one
2245       that remembers (for each socket file descriptor) whether it is currently
2246       blocking or non-blocking.  This state is maintained lazily (i.e., a system
2247       call is made only if the state actually needs to change).
2248    
2249    2. OS-specific details of sockets were moved into separate files, thus
2250       making it possible to unify the bulk of the socket implementations
2251       between Unix and Win32.
2252    
2253    3. CML's socket API changed accordingly.
2254       (Note that we need to remove non-blocking functions from this API
2255       since they are redundant in the case of CML!)
2256    
2257    4. CML's socket implementation now makes use of non-blocking functions
2258       provided by Basis, thus removing all OS-dependent code from this part
2259       of CML.
2260    
2261    5. Changed Real64.precision from 52 to 53.  Minor cleanup in Real64 code.
2262    
2263    ----------------------------------------------------------------------
2264    Name: Matthias Blume (blume (at) tti - c (dot) org)
2265    Date: 2003/09/22 12:10:00 CDT
2266    Tag: blume-20030922-110_43_2
2267    Description:
2268    
2269    Made a new interim version and bootfiles for developer's bootstrapping
2270    convenience.
2271    
2272       110.43.2 -- NEW BOOTFILES
2273    
2274    ----------------------------------------------------------------------
2275    Name: Matthias Blume (blume (at) tti - c (dot) org)
2276    Date: 2003/09/19 15:55:00 CDT
2277    Tag: blume-20030919-cmdir
2278    Description:
2279    
2280    1. new-install.sh -> install.sh
2281    2. changed default CM "metadata" directory name to ".cm" (instead of "CM")
2282    3. tweaked installer so that another name instead of .cm can be chosen
2283       at install time (by setting the CM_DIR_ARC environment variable
2284       during installation); once installation is complete, the name is
2285       fixed
2286    
2287    ----------------------------------------------------------------------
2288    Name: Matthias Blume (blume (at) tti - c (dot) org)
2289    Date: 2003/09/18 16:00:00 CDT
2290    Tag: blume-20030918-110_43_1
2291    Description:
2292    
2293    Made a new interim version and bootfiles for developer's bootstrapping
2294    convenience.
2295    
2296       110.43.1 -- NEW BOOTFILES
2297    
2298    ----------------------------------------------------------------------
2299    Name: Matthias Blume (blume (at) tti - c (dot) org)
2300    Date: 2003/09/18 15:20:00 CDT
2301    Tag: blume-20030918-misc
2302    Description:
2303    
2304    1. Exported fractionsPerSecond etc. from TimeImp (but not from Time as
2305       this seems to be controversial at the moment) and used those in
2306       Posix.ProcEnv.times.
2307    
2308    2. Added Time.{from,to}Nanoseconds to Time.
2309    
2310    3. Improved Real.{from,to}LargeInt by avoiding needless calculations.
2311       For example, fromLargeInt never needs to look at more than 3 "big
2312       digits" to get its 53 bits of precision.
2313    
2314    ----------------------------------------------------------------------
2315    Name: Matthias Blume (blume (at) tti - c (dot) org)
2316    Date: 2003/09/17 16:30:00 CDT
2317    Tag: blume-20030917-real32-slices
2318    Description:
2319    
2320    Added an entry to the primitive environment
2321    (compiler/Semant/statenv/prim.sml) for int32->real64 conversion and
2322    added code to compiler/CodeGen/main/mlriscGen.sml to implement it.
2323    
2324    Removed some of the "magic" constants in real64.sml and replaced them
2325    with code that generates these values from their corresponding
2326    integer counterparts.
2327    
2328    Made all(?) the slice-related changes to the Basis and made everything
2329    compile again...
2330    
2331    ----------------------------------------------------------------------
2332    Name: Matthias Blume (blume (at) tti - c (dot) org)
2333    Date: 2003/09/15 17:45:00 CDT
2334    Tag: blume-20030915-rbase
2335    Description:
2336    
2337    Fixed bug in Real.fromLargeInt.
2338    
2339    ----------------------------------------------------------------------
2340    Name: Matthias Blume (blume (at) tti - c (dot) org)
2341    Date: 2003/09/13 18:11:00 CDT
2342    Tag: blume-20030913-libinstall
2343    Description:
2344    
2345    Minor bugfix in config/libinstall (set anchor with path to
2346    standalone tool after installing it, otherwise libraries that
2347    need ml-lex or ml-yacc won't compile the first time the installer
2348    runs).
2349    
2350    ----------------------------------------------------------------------
2351    Name: Matthias Blume (blume (at) tti - c (dot) org)
2352    Date: 2003/09/12 11:45:00 CDT
2353    Tag: blume-20030912-various
2354    Description:
2355    
2356     - fixed bug in Real.toLargeInt
2357     - fixed bug in Posix.ProcEnv.times
2358     - changed inputLine functions to return an option
2359     - minor installer improvements / bugfixes
2360     - changed default @SMLalloc parameter for x86/celeron to 64k
2361    
2362    ----------------------------------------------------------------------
2363    Name: Matthias Blume (blume (at) tti - c (dot) org)
2364    Date: 2003/09/09 22:00:00 CDT
2365    Tag: Release_110_43
2366    Description:
2367    
2368        New working release 110.43.  New bootfiles.
2369    
2370    ----------------------------------------------------------------------
2371    Name: Matthias Blume (blume (at) tti - c (dot) org)
2372    Date: 2003/09/09 19:20:00 CDT
2373    Tag: blume-20030909-installer
2374    Description:
2375    
2376    Rewrote large parts of config/install.sh in SML (config/libinstall.sml).
2377    Modified config/install.bat to take advantage of it.  Also modified
2378    config/install.sh (and called it config/new-install.sh) to take advantage
2379    of it on Unix systems.  (The SML code is (supposed to be) platform-
2380    independent.)
2381    
2382    The installer can now install everything under Win32
2383    as well as under *nix as long as it compiles.
2384    
2385    Other changes:
2386    
2387      - made CML compile again under Win32
2388      - made eXene compile under Win32 (by providing a fake structure UnixSock
2389        and by using OS.Process.getEnv instead of Posix.ProcEnv.getenv)
2390      - fixed a bug in nowhere: it assumed that type OS.Process.status is the
2391        same as type int;  under Win32 it isn't
2392      - fixed some slice-related problems in the win32-specific parts of CML
2393      - added a functor argument "sameVol" to os-path-fn.sml in the Basis
2394        (under Win32, the volume name is case-insensitive, and the
2395        OS.Path code compares volume names for equality)
2396    
2397    ----------------------------------------------------------------------
2398    Name: Matthias Blume (blume (at) tti - c (dot) org)
2399    Date: 2003/09/08 11:55:00 CDT
2400    Tag: blume-20030908-fullpath
2401    Description:
2402    
2403    Made Win32 version of OS.FileSys.fullPath return current directory
2404    when given an empty string.  This is what the spec says, and incidentally,
2405    CM depends on it.  (CM otherwise goes into an infinite loop in certain
2406    cases when presented with the name of a non-existing .cm file.)
2407    
2408    ----------------------------------------------------------------------
2409    Name: Matthias Blume (blume (at) tti - c (dot) org)
2410    Date: 2003/09/04 16:30:00 CDT
2411    Tag: blume-20030905-slices-etc
2412    Description:
2413    
2414      1. Changed interface to vectors and arrays in Basis to match
2415         (draft) Basis spec.
2416      2. Added signatures and implementations of slices according to
2417         Basis spec.
2418      3. Edited source code throughout the system to make it compile again
2419         under 1. and 2.  (In some cases code had to be added to have it
2420         match the new signatures.)
2421      4. MLRISC should be backward-compatible: the copies of the originals
2422         of files that needed to change under 3. were retained, the .cm files
2423         check the compiler version number and use old versions when
2424         appropriate.
2425      5. Changed type of OS.FileSys.readDir and Posix.FileSys.readdir to
2426         dirstream -> string option (in accordance with Basis spec).
2427      6. When generating code that counts lines, ml-lex used function
2428         CharVector.foldli, taking advantage of its old interface.
2429         This has been replaced with the corresponding code from
2430         CharVectorSlice.  (html-lex must be re-lexed!)
2431      7. BitArray in smlnj-lib/Util has been extended/modified to match the
2432         new MONO_ARRAY signature.  (Do we need BitArraySlice?)
2433      8. Removed temporary additions (fromInternal, toInternal) from the
2434         (now obsolete) IntInf in smlnj-lib/Util.
2435      9. Cleaned up structure Byte.
2436     10. Added localOffset, scan, and fromString to Date (according to spec).
2437         Cleaned/corrected implementation of Date.
2438         (Still need to check for correctness; implement better canonicalizeDate.)
2439     11. Added "scan" to signature IEEE_REAL.
2440     12. Some improvements to IntInf [in particular: efficiency-hack for
2441         mod and rem when second operand is 2 (for parity checks).]
2442     13. Changed representation of type Time.time, using a single IntInf.int
2443         value counting microseconds.  This considerably simplified the
2444         implementation of structure Time.  We now support negative time
2445         values; scan and fromString handle signs.
2446     14. Functor PrimIO now takes two additional arguments (VectorSlice and
2447         ArraySlice).
2448    
2449    ----------------------------------------------------------------------
2450    Name: Matthias Blume (blume (at) tti - c (dot) org)
2451    Date: 2003/08/28 17:00:00 CDT
2452    Tag: blume-20030828-intinf
2453    Description:
2454    
2455    This is a major update which comes with a version number bump
2456    (110.42.99 -- yes, we are really close to 110.43 :-), NEW BOOTFILES,
2457    and an implementation of IntInf in the Basis.
2458    
2459    There are a fairly large number of related changes and updates throughout
2460    the system:
2461    
2462    Basis:
2463            - Implemented IntInf.
2464            - Made LargeInt a projection of IntInf (by filtering through INTEGER).
2465            - Added some missing Real64 operations, most notably Real.toLargeInt.
2466            - Added FixedInt as a synonym for Int32.
2467    
2468    compiler:
2469            * Added support for a built-in intinf type.
2470              - literals
2471              - pattern matching
2472              - conversion shortcuts (Int32.fromLarge o Int.toLarge etc.)
2473              - overloading on literals and operations
2474    
2475            This required adding a primitive type intinf, some additional
2476            primops, and implementations for several non-trivial intinf
2477            operations in Core.  (The intinf type is completely abstract
2478            to the compiler; all operations get delegated back to the Core.)
2479    
2480            * Intinf equality is handled by polyequal.  However, the compiler
2481            does not print its usual warning in this case (since polyequal
2482            is the right thing to do there).
2483    
2484            * Improved the organization of structure InlineT.
2485    
2486            * A word about conversion primops:
2487            If conversions involving intinf do not cancel out during
2488            CPS contract, then the compiler must insert calls to Core functions.
2489            Since all core access must be resolved already during the FLINT
2490            translate phase, it would be too late a the time of CPS contract
2491            to add new Core calls.  For this reason, conversion primops
2492            for intinf carry two arguments:  1. the numeric argument that
2493            they are supposed to convert, and 2. the Core function that
2494            can help with this conversion if necessary.  If CPS contract
2495            eliminates a primop, then the associated Core function becomes
2496            dead and goes away.  Intinf conversion primops that do not get
2497            eliminated by CPS contract get rewritten into calls of their
2498            core functions by a separate, new phase.
2499    
2500    interactive system:
2501            - Control.Print.intinfDepth controls max length of intinf constants
2502            being printed. (Analogous to Control.Print.stringDepth.)
2503            - Cleanup in printutil and pputil: got rid of unused stuff and
2504            duplicates; replaced some of the code with code that makes better
2505            use of library functionality.
2506    
2507    CM:
2508            Bugfix: parse-errors in init group (system/smlnj/init/init.cmi)
2509            are no longer silent.
2510    
2511    CKIT:
2512            Fixed mismatched uses of Int32 and LargeInt.  I always decided
2513            in favor of LargeInt -- which is now the same as IntInf.
2514            CKIT-knowledgable people should check whether this is what's
2515            intended and otherwise change things back to using Int32 or
2516            FixedInt.
2517    
2518    Throughout the code:
2519            Started using IntInf.int literals and built-in operations
2520            (e.g., comparison with 0) where this seems appropriate.
2521    
2522    
2523    ----------------------------------------------------------------------
2524    Name: Dave MacQueen (dbm@cs.uchicago.edu)
2525    Date: 2003/08/13 11:36:00 CDT
2526    Tag: dbm-20030813-mcz-merge1
2527    Description:
2528    
2529    Merging changes from the mcz-branch development branch into trunk.
2530    These changes involve replacement of the emulated old prettyprinter
2531    interface with direct use of the SML/NJ Lib PP library, and fixing
2532    of a couple of bugs (895, 1186) relating to error messages.  A new
2533    prettyprinter for ast datatypes (Elaborator/print/ppast.{sig,sml})
2534    has been added.
2535    
2536    ----------------------------------------------------------------------
2537    Name: Matthias Blume (blume (at) tti - c (dot) org)
2538    Date: 2003/08/11 15:45:00 CDT
2539    Tag: blume-20030811-windows
2540    Description:
2541    
2542    Version number bumped to 110.42.9.  NEW BOOTFILES!!!
2543    
2544        http://smlnj.cs.uchicago.edu/dist/working/110.42.9/
2545    
2546    This patch restores SML/NJ's ability to run under win32.  There are a
2547    number of changes, including fixes for several bugs that had gone
2548    unnoticed until now:
2549    
2550      - uname "CYGWIN_NT*" is recognized as win32 (This is relevant only when
2551        trying to run the win32 version from within cygwin.)
2552    
2553      - There are a number of simple .bat scripts that substitute for their
2554        corresponding Unix shell-scripts.  (See below.)
2555    
2556      - The internals of ml-build have been modified slightly.  The main
2557        difference is that instead of calling ".link-sml" (or link-sml.bat)
2558        using OS.Process.system, the ML process delegates this task back
2559        to the script.  Otherwise problems arise in mixed environments such
2560        as Cygwin where scripts look and work like Unix scripts, but
2561        where OS.Process.system cannot run them.
2562    
2563      - In CM, the srcpath pickler used native pathname syntax -- which
2564        is incorrect in the case of cross-compilation.  The new pickle format
2565        is independent of platform-specific naming conventions.
2566    
2567      - Path configuration files (such as lib/pathconfig) can now choose
2568        between native and standard syntax.  Placing a line of the form
2569    
2570            standard!
2571    
2572        into the file causes all subsequent paths to be interpreted using
2573        CM standard pathname syntax (= Unix conventions); a line
2574    
2575            native!
2576    
2577        switches back to native style.  This was needed so that
2578        path config files can be written portably, see src/system/pathconfig.
2579    
2580      - Runtime system:
2581    
2582          - win32-filesys.c:  get_file_time and set_file_time now
2583            access modification time, not creation time.
2584    
2585          - I/O code made aware of new array representation.
2586    
2587          - Bug fixes in X86.prim.masm.
2588    
2589       - src/system/makeml made aware of win32. (For use under cygwin
2590            and other Unix-environments for windows.)
2591    
2592       - In Basis, fixed off-by-one error in win32-io.sml (function vecF)
2593         which caused BinIO.inputAll to fail consistently.
2594    
2595    .bat scripts:
2596    
2597       Windows .bat scripts assume that SMLNJ_HOME is defined.
2598    
2599       - sml.bat, ml-yacc.bat, ml-lex.bat:  Driver scripts for standalone
2600         applications (sml, ml-yacc, ml-lex).
2601       - ml-build.bat: analogous to ml-build.
2602       - config\install.bat:  Analogous to config/install.sh.  This requires
2603         that SMLNJ_HOME is set and that Microsoft Visual C is ready to use.
2604         (nmake etc. must be on the path, and vcvars32 must have been run.)
2605         Moreover, sources for ml-lex and ml-yacc need to exist under src,
2606         and the bootfile hierarchy must have been unpacked under
2607         sml.boot.x86-win32.
2608         The script is very primitive and does a poor job at error checking.
2609         It only installs the base system, ml-lex, and ml-yacc.  No other
2610         libraries are being installed (i.e., you get only those that
2611         are part of the compiler.)
2612       - link-sml.bat:  analogous to .link-sml, but not currently used
2613    
2614    Unrelated bug fixes:
2615    
2616      - ml-nlffigen now exports structures ST_* corresponding to incomplete
2617        types.
2618      - Added getDevice to PP/src/pp-debug-fn.sml.  (Would not compile
2619        otherwise.)
2620    
2621    ----------------------------------------------------------------------
2622    Name: Dave MacQueen (macqueen@cs.uchicago.edu)
2623    Date: 2003/06/17
2624    Tag: macqueen-20030617-bug895
2625    Description:
2626    
2627    Modified compiler/Elaborator/print/pptype.sml to fix bug 895.
2628    Tag will be used for new development branch (mcz-branch) for
2629    use by MacQueen, (Lucasz) Zairek, and (George) Cao at uchicago.
2630    
2631    ----------------------------------------------------------------------
2632    Name: Matthias Blume (blume (at) tti - c (dot) org)
2633    Date: 2003/05/27 16:55:00 CDT
2634    Tag: blume-20030527-polyeq
2635    Description:
2636    
2637    Tried to eliminated most cases of polymorphic equality.
2638    
2639    ----------------------------------------------------------------------
2640    Name: Matthias Blume (blume (at) tti - c (dot) org)
2641    Date: 2003/05/21 17:45:00 CDT
2642    Tag: blume-20030517-complete
2643    Description:
2644    
2645    Two changes:
2646    
2647      1. Added a flag for controlling whether non-exhaustive bindings will
2648         be treated as errors (default is false).
2649      2. Cleaned up the *entire* source tree so that CMB.make goes through
2650         without a single non-exhaustive match- or bind warning.
2651    
2652    ----------------------------------------------------------------------
2653    Name: Matthias Blume (blume (at) tti - c (dot) org)
2654    Date: 2003/05/17 10:20:00 CDT
2655    Tag: blume-20030517-absyn
2656    Description:
2657    
2658    1. Added cases for IF, WHILE, ANDALSO, and ORELSE to Absyn.
2659    
2660       This mainly affects the quality of error messages.  However, some
2661       of the code is now more straightforward than before. (Treatment of
2662       the above four constructs in translate.sml is much simpler than
2663       the "macro-expansion" that was going on before.  Plus, the mach-
2664       compiler no longer gets invoked just to be able to compile an
2665       if-expression.)
2666    
2667    2. The ErrorMsg.Error exception is now caught and absorbed by the
2668       interactive loop.
2669    
2670    ----------------------------------------------------------------------
2671    Name: Allen Leung
2672    Date: 2003/05/16 13:05:00 CDT
2673    Tag: leunga-20030516-cygwin-runtime
2674    Description:
2675    
2676        Ported the runtime system to cygwin, which uses the unix
2677    x86-unix bin files.  Missing/buggy features:
2678    
2679       o getnetbyname, getnetbyaddr: these functions seem to be missing in
2680         the Cygwin library.
2681       o Ctrl-C handling may be flaky.
2682       o Windows system calls and Windows I/O are not supported.
2683    
2684       A new set of binfiles is located at:
2685    
2686         http://www.dorsai.org/~leunga/boot.x86-unix.tgz
2687    
2688       This is only needed for bootstrapping the cygwin version of smlnj.
2689       Other x86 versions can use the existing binfiles.
2690    
2691    ----------------------------------------------------------------------
2692    Name: Matthias Blume
2693    Date: 2003/04/08 15:42:00 CDT
2694    Tag: blume-20030408-listpair
2695    Description:
2696    
2697    1. Added a target 'mlrisc' to installer.
2698    
2699    2. Added missing elements to structure ListPair.
2700    
2701    ----------------------------------------------------------------------
2702    Name: Allen Leung
2703    Date: 2003/01/07 10:40:00 EST
2704    Tag: leunga-20030107-int-rem
2705    Description:
2706    
2707    Fixed a bug in Int.rem(x,y) where y is a power of 2 on x86.
2708    The arguments to the SUBL instruction were swapped.
2709    
2710    ----------------------------------------------------------------------
2711    Name: Matthias Blume
2712    Date: 2002/12/12 16:25:00 EST
2713    Tag: blume-20021212-risc-ra
2714    Description:
2715    
2716    Fixed a serious bug in the rewrite code for FP spilling/reloading that
2717    sent the RA into an infinite loop when floating point registers get
2718    spilled.  (Because of this bug, e.g., nucleic stopped compiling between
2719    110.37 and 110.38.)
2720    There was another set of potential problems related to the handling of
2721    MLRISC annotations (but those did not yet cause real problems, apparently).
2722    
2723    ----------------------------------------------------------------------
2724    Name: Matthias Blume
2725    Date: 2002/12/06 22:40:00 EST
2726    Tag: blume-20021206-cm-fileid
2727    Description:
2728    
2729    Added a call of SrcPath.sync at the beginning of Parse.parse (in CM).
2730    This fixes the problem of CM getting confused by files that suddenly
2731    change their identity (e.g., by getting unlinked and recreated by some
2732    text editor such as vi).  There might be a better/cheaper/cleaner way
2733    of doing this, but for now this will have to do.
2734    
2735    ----------------------------------------------------------------------
2736    Name: Matthias Blume
2737    Date: 2002/10/28 09:50:00 EST
2738    Tag: blume-20021028-typecheck
2739    Description:
2740    
2741    Exported structure Typecheck from $smlnj/viscomp/core.cm.
2742    
2743    ----------------------------------------------------------------------
2744    Name: Matthias Blume
2745    Date: 2002/10/17 09:10:00 EDT
2746    Tag: Release_110_42
2747    Description:
2748    
2749    In good old tradition, there has been a slight hiccup so that we have
2750    to patch 110.42 after the fact.   The old release tag has been replaced
2751    (see below).
2752    
2753    The change solves a problem with two competing approaches the
2754    configuration problem regarding MacOS 10.1 vs. MacOS 10.2 which got in
2755    each other's way.
2756    
2757    This change only affects the runtime system code and the installer script.
2758    (No new bootfiles.)
2759    
2760    ----------------------------------------------------------------------
2761    Name: Matthias Blume
2762    Date: 2002/10/16 12:00:00 EDT
2763    Tag: Release_110_42_removed
2764    Description:
2765    
2766    New working release.  New bootfiles.
2767    
2768    ----------------------------------------------------------------------
2769    Name: Matthias Blume
2770    Date: 2002/10/10 13:10:00 EDT
2771    Tag: blume-20021010-ppc-divs
2772    Description:
2773    
2774    The mltree operator DIVS must be implemented with an overflow check on
2775    the PPC because the hardware indicates divide-by-zero using "overflow" as
2776    well.
2777    
2778    ----------------------------------------------------------------------
2779    Name: Matthias Blume
2780    Date: 2002/07/23 11:45:00
2781    Tag: blume-20020723-smlnj-home
2782    Description:
2783    
2784    Sml now senses the SMLNJ_HOME environment variable.  If this is set,
2785    then the bin dir is assumed to be in $SMLNJ_HOME/bin and (unless
2786    CM_PATHCONFIG is also set), the path configuration file is assumed
2787    to be in $SMLNJ_HOME/lib/pathconfig.  This way one can easily move
2788    the entire tree to some other place and everything will "just work".
2789    
2790    (Companion commands such as ml-build and ml-makedepend also sense this
2791    variable.)
2792    
2793    ----------------------------------------------------------------------
2794    Name: Matthias Blume
2795    Date: 2002/07/12 21:19:00 EDT
2796    Tag: blume-20020712-liveness
2797    Description:
2798    
2799    Exported two useful "step" functions from liveness module (MLRISC).
2800    
2801    ----------------------------------------------------------------------
2802    Name: Matthias Blume
2803    Date: 2002/07/05 16:00 EDT
2804    Tag: Release_110_41
2805    Description:
2806    
2807    New working release.  New bootfiles.
2808    
2809    ----------------------------------------------------------------------
2810    Name: Matthias Blume
2811    Date: 2002/07/05 10:25:00 EDT
2812    Tag: blume-20020705-btimp
2813    Description:
2814    
2815    Exported structure BTImp from $smlnj/viscomp/debugprof.cm so that
2816    other clients can set up backtracing support.
2817    
2818    ----------------------------------------------------------------------
2819    Name: Matthias Blume
2820    Date: 2002/06/25 17:23:00 EDT
2821    Tag: blume-20020625-fpmax
2822    Description:
2823    
2824    Fixed a bug in translation of INLMAX (and INLMIN) for the floating-point
2825    case.  (The sense of the isNaN test was reversed -- which made min and
2826    max always return their first argument.)
2827    
2828    ----------------------------------------------------------------------
2829    Name: Matthias Blume
2830    Date: 2002/06/11
2831    Tag: blume-20020611-unixpath
2832    Description:
2833    
2834    Back-ported OS.Path.{from,to}UnixPath from idlbasis-devel branch.
2835    
2836    ----------------------------------------------------------------------
2837    Name: Matthias Blume
2838    Date: 2002/06/10 16:35:00 EDT
2839    Tag: blume-20020610-ieeereal
2840    Description:
2841    
2842    I back-ported my implementation of IEEEReal.fromString from the
2843    idlbasis-devel branch so that we can test it.
2844    
2845    Another small change is that ppDec tries to give more information
2846    than just "<sig>" in the case of functors.  However, this code is
2847    broken in some mysterious way if the functor's body's signature
2848    has not been declared by ascription but gets inferred from the
2849    implementation.  This needs fixing...
2850    
2851    ----------------------------------------------------------------------
2852    Name: Matthias Blume
2853    Date: 2002/05/31
2854    Tag: blume-20020531-btrace-mode
2855    Description:
2856    
2857    Resurrected SMLofNJ.Internals.BTrace.mode.  (It accidentally fell by
2858    the wayside when I switched over to using Controls everywhere.)
2859    
2860    ----------------------------------------------------------------------
2861    Name: Lal George
2862    Date: 2002/05/23 12:21:40 EDT
2863    Tag: george-20020523-visual-labels
2864    Description:
2865    
2866    Labels are now displayed in the graphical output to make
2867    the fall-through and target blocks obvious.
2868    
2869    ----------------------------------------------------------------------
2870    Name: Matthias Blume
2871    Date: 2002/05/22 11:03:00 EDT
2872    Tag: blume-20020522-shrink
2873    Description:
2874    
2875    John tweaked yesterday's fix for 1131 to handle an out-of-memory
2876    situation that comes up when allocating huge arrays.
2877    
2878    ----------------------------------------------------------------------
2879    Name: Matthias Blume
2880    Date: 2002/05/21 16:00:00 EDT
2881    Tag: Release_110_40
2882    Description:
2883    
2884    New working release (110.40).  New bootfiles.
2885    
2886    [Also: John Reppy fixed GC bug 1131.]
2887    
2888    ----------------------------------------------------------------------
2889    Name: Matthias Blume
2890    Date: 2002/05/21 12:35:00 EDT
2891    Tag: blume-20020521-cmdoc
2892    Description:
2893    
2894    CM documentation update.
2895    
2896    ----------------------------------------------------------------------
2897    Name: Matthias Blume
2898    Date: 2002/05/21 10:55:00 EDT
2899    Tag: blume-20020521-misc
2900    Description:
2901    
2902    - John tweaked runtime to be silent on heap export (except when
2903      GC messages are on).
2904    - I added a few more things (cross-compiling versions of CMB) to
2905      config/preloads (as suggestions).
2906    
2907    ----------------------------------------------------------------------
2908    Name: Matthias Blume
2909    Date: 2002/05/20 22:25:00 EDT
2910    Tag: blume-20020520-controls
2911    Description:
2912    
2913    - Added ControlUtil structure to control-lib.cm.
2914    - Use it throughout.
2915    - Used Controls facility to define MLRISC controls (as opposed to
2916      registering MLRISC control ref cells with Controls after the fact)
2917    - Fixed messed-up controls priorities.
2918    
2919    * Removed again all the stuff from config/preloads that one wouldn't
2920      be able to preload at the time the initial heap image is built.
2921      (Many libraries, e.g., CML, do not exist yet at this time.  The
2922       only libraries that can be preloaded via config/preloads are those
2923       that come bundled with the bootfiles.)
2924    
2925    ----------------------------------------------------------------------
2926    Name: Matthias Blume
2927    Date: 2002/05/20 10:59:00 EDT
2928    Tag: blume-20020520-preloads
2929    Description:
2930    
2931    Added a lot of commented-out suggestions for things to be included
2932    in config/preloads.
2933    
2934    ----------------------------------------------------------------------
2935    Name: Allen Leung
2936    Date: 2002/05/18 14:20:00 EDT
2937    Tag: leunga-20020518-mdl
2938    Description:
2939    
2940    o Made the mdl tool stuff compile and run again.
2941    o I've disabled all the stuff that depends on RTL specifications; they
2942      are all badly broken anyway.
2943    
2944    ----------------------------------------------------------------------
2945    Name: Matthias Blume
2946    Date: 2002/05/17 16:49:00 EDT
2947    Tag: blume-20020517-controls
2948    Description:
2949    
2950    0. John Reppy made several modifications to the SML/NJ library.
2951       In particular, there is a shiny new controls-lib.cm.
2952    
2953    1. Pushed new controls interface through compiler so that everything
2954       compiles again.
2955    
2956    2. Added FormatComb and FORMAT_COMB to the CML version of the
2957       SML/NJ library (so that CML compiles again).
2958    
2959    3. Modified init scripts because XXX_DEFAULT environment variables
2960       are no longer with us.  (Boot-time initialization is now done
2961       using the same environment variables that are also used for
2962       startup-time initialization of controls.)
2963    
2964    ----------------------------------------------------------------------
2965    Name: Lal George
2966    Date: 2002/05/15 09:20:10 EDT
2967    Tag: george-20020515-pseudo-op-decls
2968    Description:
2969    
2970            All pseudo-ops emitted before the first segment declaration
2971    such as TEXT, DATA, and BSS directives are assumed to be global
2972    declarations and are emitted first in the assembly file. This is
2973    useful in a number of situations where one has pseudo-ops that are not
2974    specific to any segment, and also works around the constraint that one
2975    cannot have client pseudo-ops in the TEXT segment.
2976    
2977            Because no segment is associated with these declarations it is
2978    an error to allocate any space or objects before the first segment
2979    directive and an exception will be raised. However, we cannot make
2980    this check for client pseudo-ops.
2981    
2982            These top level declarations are a field in the CFG graph_info.
2983    In theory you can continue to add to this field after the CFG has been
2984    built -- provided you know what you are doing;-)
2985    
2986    ----------------------------------------------------------------------
2987    Name: Matthias Blume
2988    Date: 2002/05/13 16:40:00 EDT
2989    Tag: blume-20020513-pp-etc
2990    Description:
2991    
2992    A few minor bugfixes:
2993    
2994      - Stopgap measure for bug recently reported by Elsa Gunter (ppDec).
2995        (Bogus printouts for redefined bindings still occur.  Compiler
2996        bug should no longer occur now.  We need to redo the prettyprinter
2997        from scratch.)
2998    
2999      - CM pathname printer now also adds escape sequences for ( and )
3000    
3001      - commend and docu fixes for ml-nlffi
3002    
3003    ----------------------------------------------------------------------
3004    Name: Matthias Blume
3005    Date: 2002/05/10 16:40:00 EDT
3006    Tag: blume-20020510-erg-textio
3007    Description:
3008    
3009    Applied the following bugfix provided by Emden Gansner:
3010    
3011        Output is corrupted when outputSubstr is used rather than output.
3012    
3013        The problem occurs when a substring
3014    
3015            ss = (s, dataStart, dataLen)
3016    
3017        where dataStart > 0, fills a stream buffer with avail bytes left.
3018        avail bytes of s, starting at index dataStart, are copied into the
3019        buffer, the buffer is flushed, and then the remaining dataLen-avail
3020        bytes of ss are copied into the beginning of the buffer. Instead of
3021        starting this copy at index dataStart+avail in s, the current code
3022        starts the copy at index avail.
3023    
3024        Fix:
3025        In text-io-fn.sml, change line 695 from
3026             val needsFlush = copyVec(v, avail, dataLen-avail, buf, 0)
3027        to
3028             val needsFlush = copyVec(v, dataStart+avail, dataLen-avail, buf, 0)
3029    
3030    ----------------------------------------------------------------------
3031    Name: Matthias Blume
3032    Date: 2002/04/12 13:55:00 EDT
3033    Tag: blume-20020412-assyntax
3034    Description:
3035    
3036    1. Grabbed newer assyntax.h from the XFree86 project.
3037    2. Fiddled with how to compile X86.prim.asm without warnings.
3038    3. (Very) Minor cleanup in CM.
3039    
3040    ----------------------------------------------------------------------
3041    Name: Matthias Blume
3042    Date: 2002/04/01 (no joke!) 17:07:00 EST
3043    Tag: blume-20020401-x86div
3044    Description:
3045    
3046    Added full support for div/mod/rem/quot on the x86, using the machine
3047    instruction's two results (without clumsily recomputing the remainder)
3048    directly where appropriate.
3049    
3050    Some more extensive power-of-two support was added to the x86 instruction
3051    selector (avoiding expensive divs, mods, and muls where they can be
3052    replaced with cheaper shifts and masks).  However, this sort of thing
3053    ought to be done earlier, e.g., within the CPS optimizer so that
3054    all architectures benefit from it.
3055    
3056    The compiler compiles to a fixed point, but changes might be somewhat
3057    fragile nevertheless.  Please, report any strange things that you might
3058    see wrt. div/mod/quot/rem...
3059    
3060    ----------------------------------------------------------------------
3061    Name: Matthias Blume
3062    Date: 2002/03/29 17:22:00
3063    Tag: blume-20020329-div
3064    Description:
3065    
3066    Fixed my broken div/mod logic.  Unfortunately, this means that the
3067    inline code for div/mod now has one more comparison than before.
3068    Fast paths (quotient > 0 or remainder = 0) are not affected, though.
3069    The problem was with quotient = 0, because that alone does not tell
3070    us which way the rounding went.  One then has to look at whether
3071    remainder and divisor have the same sign...  :(
3072    
3073    Anyway, I replaced the bootfiles with fresh ones...
3074    
3075    ----------------------------------------------------------------------
3076    Name: Matthias Blume
3077    Date: 2002/03/29 14:10:00 EST
3078    Tag: blume-20020329-inlprims
3079    Description:
3080    
3081    NEW BOOTFILES!!!    Version number bumped to 110.39.3.
3082    
3083    Primops have changed. This means that the bin/boot-file formats have
3084    changed as well.
3085    
3086    To make sure that there is no confusion, I made a new version.
3087    
3088    
3089    CHANGES:
3090    
3091    * removed REMT from mltree (remainder should never overflow).
3092    
3093    * added primops to deal with divisions of all flavors to the frontend
3094    
3095    * handled these primops all the way through so they map to their respective
3096      MLRISC support
3097    
3098    * used these primops in the implementation of Int, Int32, Word, Word32
3099    
3100    * removed INLDIV, INLMOD, and INLREM as they are no longer necessary
3101    
3102    * parameterized INLMIN, INLMAX, and INLABS by a numkind
3103    
3104    * translate.sml now deals with all flavors of INL{MIN,MAX,ABS}, including
3105      floating point
3106    
3107    * used INL{MIN,MAX,ABS} in the implementation of Int, Int32, Word, Word32,
3108      and Real (but Real.abs maps to a separate floating-point-only primop)
3109    
3110    
3111    TODO items:
3112    
3113    * Hacked Alpha32 instruction selection, disabling the selection of REMx
3114      instructions because the machine instruction encoder cannot handle
3115      them.  (Hppa, PPC, and Sparc instruction selection did not handle
3116      REM in the first place, and REM is supported by the x86 machine coder.)
3117    
3118    * Handle DIV and MOD with DIV_TO_NEGINF directly in the x86 instruction
3119      selection phase.  (The two can be streamlined because the hardware
3120      delivers both quotient and remainder at the same time anyway.)
3121    
3122    * Think about what to do with "valOf(Int32.minInt) div ~1" and friends.
3123      (Currently the behavior is inconsistent both across architectures and
3124      wrt. the draft Basis spec.)
3125    
3126    * Word8 should eventually be handled natively, too.
3127    
3128    * There seems to be one serious bug in mltree-gen.sml.  It appears, though,
3129      as if there currently is no execution path that could trigger it in
3130      SML/NJ.  (The assumptions underlying functions arith and promotable do not
3131      hold for things like multiplication and division.)
3132    
3133    ----------------------------------------------------------------------
3134    Name: Matthias Blume
3135    Date: 2002/03/27 16:27:00 EST
3136    Tag: blume-20020327-mlrisc-divisions
3137    Description:
3138    
3139    Added support for all four division operations (ML's div, mod, quot,
3140    and rem) to MLRISC.  In the course of doing so, I also rationalized
3141    the naming (no more annoying switch-around of DIV and QUOT), by
3142    parameterizing the operation by div_rounding_mode (which can be either
3143    DIV_TO_ZERO or DIV_TO_NEGINF).
3144    
3145    The generic MLTreeGen functor takes care of compiling all four
3146    operations down to only round-to-zero div.
3147    
3148    Missing pieces:
3149    
3150      * Doing something smarter than relying on MLTreeGen on architectures
3151        like, e.g., the x86 where hardware division delivers both quotient and
3152        remainder at the same time.  With this, the implementation of the
3153        round-to-neginf operations could be further streamlined.
3154    
3155      * Remove inlining support for div/mod/rem from the frontend and replace it
3156        with primops that get carried through to the backend.  Do this for all
3157        int and word types.
3158    
3159    ----------------------------------------------------------------------
3160    Name: Matthias Blume
3161    Date: 2002/03/25 17:25:00 EST
3162    Tag: blume-20020325-divmod
3163    Description:
3164    
3165    I improved (hopefully without breaking them) the implementation of Int.div,
3166    Int.mod, and Int.rem.   For this, the code in translate.sml now takes
3167    advantage of the following observations:
3168    
3169      Let  q = x quot y      r = x rem y
3170           d = x div  y      m = x mod y
3171    
3172    where "quot" is the round-to-zero version of integer division that
3173    hardware usually provides.  Then we have:
3174    
3175         r = x - q * y        where neither the * nor the - will overflow
3176         d = if q >= 0 orelse x = q * y then q else q - 1
3177                              where neither the * nor the - will overflow
3178         m = if q >= 0 orelse r = 0 then r else r + y
3179                              where the + will not overflow
3180    
3181    This results in substantial simplification of the generated code.
3182    The following table shows the number of CFG nodes and edges generated
3183    for
3184            fun f (x, y) = x OPER y
3185            (* with OPER \in div, mod, quot, rem *)
3186    
3187    
3188        OPER | nodes(old) | edges(old) | nodes(new) | edges(new)
3189        --------------------------------------------------------
3190         div |         24 |         39 |         12 |         16
3191         mod |         41 |         71 |         12 |         16
3192        quot |          8 |         10 |          8 |         10
3193         rem |         10 |         14 |          8 |         10
3194    
3195    
3196    ----------------------------------------------------------------------
3197    Name: Matthias Blume
3198    Date: 2002/03/25 22:06:00 EST
3199    Tag: blume-20020325-cprotobug
3200    Description:
3201    
3202    Fixed a bug in cproto (c prototype decoder).
3203    
3204    ----------------------------------------------------------------------
3205    Name: Matthias Blume
3206    Date: 2002/03/25 16:00:00 EST
3207    Tag: blume-20020325-raw-primops
3208    Description:
3209    
3210    I did some cleanup to Allen's new primop code and
3211    replaced yesterday's bootfiles with new ones.
3212    (But they are stored in the same place.)
3213    
3214    ----------------------------------------------------------------------
3215    Name: Matthias Blume
3216    Date: 2002/03/24 22:40:00 EST
3217    Tag: blume-20020324-bootfiles
3218    Description:
3219    
3220    Made the bootfiles that Allen asked for.
3221    
3222    ----------------------------------------------------------------------
3223    Name: Allen Leung
3224    Date: 2002/03/23 15:50:00 EST
3225    Tag: leunga-20020323-flint-cps-rcc-primops
3226    Description:
3227    
3228      1. Changes to FLINT primops:
3229    
3230        (* make a call to a C-function;
3231         * The primop carries C function prototype information and specifies
3232         * which of its (ML-) arguments are floating point. C prototype
3233         * information is for use by the backend, ML information is for
3234         * use by the CPS converter. *)
3235      | RAW_CCALL of { c_proto: CTypes.c_proto,
3236                       ml_args: ccall_type list,
3237                       ml_res_opt: ccall_type option,
3238                       reentrant : bool
3239                     } option
3240       (* Allocate uninitialized storage on the heap.
3241        * The record is meant to hold short-lived C objects, i.e., they
3242        * are not ML pointers.  With the tag, the representation is
3243        * the same as RECORD with tag tag_raw32 (sz=4), or tag_fblock (sz=8)
3244        *)
3245      | RAW_RECORD of {tag:bool,sz:int}
3246      and ccall_type = CCALL_INT32 | CCALL_REAL64 | CCALL_ML_PTR
3247    
3248      2.  These CPS primops are now overloaded:
3249    
3250           rawload of {kind:numkind}
3251           rawstore of {kind:numkind}
3252    
3253          The one argument form is:
3254    
3255             rawload {kind} address
3256    
3257          The two argument form is:
3258    
3259             rawload {kind} [ml object, byte-offset]
3260    
3261      3. RAW_CCALL/RCC now takes two extra arguments:
3262    
3263         a. The first is whether the C call is reentrant, i.e., whether
3264            ML state should be saved and restored.
3265         b. The second argument is a string argument specifying the name of
3266            library and the C function.
3267    
3268         These things are currently not handled in the code generator, yet.
3269    
3270      4. In CProto,
3271    
3272         An encoding type of "bool" means "ml object" and is mapped into
3273         C prototype of PTR.  Note that "bool" is different than "string",
3274         even though "string" is also mapped into PTR, because "bool"
3275         is assigned an CPS type of BOGt, while "string" is assigned INT32t.
3276    
3277      5. Pickler/unpicker
3278    
3279         Changed to handle RAW_RECORD and newest RAW_CCALL
3280    
3281      6. MLRiscGen,
3282    
3283         1. Changed to handle the new rawload/rawstore/rawrecord operators.
3284         2. Code for handling C Calls has been moved to a new module CPSCCalls,
3285            in the file CodeGen/cpscompile/cps-c-calls.sml
3286    
3287      7. Added the conditional move operator
3288    
3289             condmove of branch
3290    
3291         to cps.  Generation of this is still buggy so it is currently
3292         disabled.
3293    
3294    ----------------------------------------------------------------------
3295    Name: Lal George
3296    Date: 2002/03/22 14:18:25 EST
3297    Tag: george-20020322-cps-branch-prob
3298    Description:
3299    
3300    Implemented the Ball-Larus branch prediction-heuristics, and
3301    incorporated graphical viewers for control flow graphs.
3302    
3303    Ball-Larus Heuristics:
3304    ---------------------
3305    See the file compiler/CodeGen/cpscompile/cpsBranchProb.sml.
3306    
3307    By design it uses the Dempster-Shafer theory for combining
3308    probabilities.  For example, in the function:
3309    
3310        fun f(n,acc) = if n = 0 then acc else f(n-1, n*acc)
3311    
3312    the ball-larus heuristics predicts that the n=0 is unlikely
3313    (OH-heuristic), and the 'then' branch is unlikely because of the
3314    RH-heuristic -- giving the 'then' branch an even lower combined
3315    probability using the Dempster-Shafer theory.
3316    
3317    Finally, John Reppy's loop analysis in MLRISC, further lowers the
3318    probability of the 'then' branch because of the loop in the else
3319    branch.
3320    
3321    
3322    Graphical Viewing:
3323    ------------------
3324    I merely plugged in Allen's graphical viewers into the compiler. The
3325    additional code is not much. At the top level, saying:
3326    
3327            Control.MLRISC.getFlag "cfg-graphical-view" := true;
3328    
3329    will display the graphical view of the control flow graph just before
3330    back-patching.  daVinci must be in your path for this to work. If
3331    daVinci is not available, then the default viewer can be changed
3332    using:
3333    
3334            Control.MLRISC.getString "viewer"