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

SCM Repository

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

Diff of /sml/trunk/NOTES/HISTORY

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

revision 939, Tue Sep 25 19:25:48 2001 UTC revision 1905, Thu Mar 2 04:14:04 2006 UTC
# Line 8  Line 8 
8  The form of an entry should be:  The form of an entry should be:
9    
10  Name:  Name:
11  Date:  Date: yyyy/mm/dd
12  Tag: <post-commit CVS tag>  Tag: <post-commit CVS tag>
13  Description:  Description:
14    
15  ----------------------------------------------------------------------  ----------------------------------------------------------------------
16    Name: Matthias Blume (http://www.tti-c.org/blume/)
17    Date: 2006/03/01 22:15:00 CST
18    Tag: blume-20060301-lexgen
19    Description:
20    
21    Incorporated several bugfixes to lexgen.  Compiler now compiles to
22    fixpoint when using lexgen instead of ml-lex.
23    
24    ----------------------------------------------------------------------
25    Name: Matthias Blume (http://www.tti-c.org/blume/)
26    Date: 2006/02/28 22:45:00 CST
27    Tag: blume-20060228-lexgen
28    Description:
29    
30    Removed ml-flex and added lexgen instead, using Aaron Turon's newly
31    provided tarball.  The generated lexers still have problems.
32    
33    ----------------------------------------------------------------------
34    Name: Matthias Blume (http://www.tti-c.org/blume/)
35    Date: 2006/02/26 23:30:00 CST
36    Tag: blume-20060226-ml-flex
37    Description:
38    
39    Removed ml-flex' dependency on regexp-lib.cm.  Turned installation
40    of ml-flex on by default.
41    
42    ----------------------------------------------------------------------
43    Name: Matthias Blume (http://www.tti-c.org/blume/)
44    Date: 2006/02/24 16:30:00 CST
45    Tag: blume-20060224-ml-flex
46    Description:
47    
48    Added ml-flex sources.  Partially integrated, but should not be
49    turned on yet!  (Read: leave it commented-out in config/targets!)
50    
51    ----------------------------------------------------------------------
52    Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
53    Date: 2006/02/23
54    Tag: jhr-20060223-x86-64
55    Description:
56    Changes to support building on x86-64 systems (using the 32-bit mode).
57    Also cleaned up signal handling on Linux.  Support for pre-2.2 Linux
58    kernels dropped.
59    
60    ----------------------------------------------------------------------
61    Name: Allen Leung (leunga (at) reservoir (dot) com)
62    Date: 2006/02/22 15:07:00 EST
63    Tag: leunga-20060222-mdl-gen
64    Description:
65    
66      CM has changed.  Updated the script for rebuilding the MLRISC generated
67    files.
68    
69    ----------------------------------------------------------------------
70    Name: Matthias Blume (http://www.tti-c.org/blume/)
71    Date: 2006/02/14 17:30:00 CST
72    Tag: blume-20060214-intel-mac
73    Description:
74    
75    Hooked code for Darwin-specific Intel ABI into main compiler.
76    (This is a temporary solution which relies on the fact that the
77    compiler itself does not use NLFFI.  Eventually we need to divorce
78    intel mac from generic x86 unix code and make separate sets of binaries.)
79    
80    ----------------------------------------------------------------------
81    Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
82    Date: 2006/02/13
83    Tag: jhr-20060213-intel-mac
84    Description:
85    Changed MLRISC x86 CCalls for partial support of MacOS X ABI.
86    
87    ----------------------------------------------------------------------
88    Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
89    Date: 2006/02/06
90    Tag: jhr-20060206-intel-mac
91    Description:
92    Changes to support MacOS X on Intel hardware.  The C-calls support in
93    MLRISC must also be updated to support the MacOS X ABI.
94    
95    ----------------------------------------------------------------------
96    Name: Matthias Blume (http://www.tti-c.org/blume/)
97    Date: 2005/12/16 00:15:00 CST
98    Tag: blume-20051216-cm
99    Description:
100    
101    Improved error reporting and handling in CM.
102    
103    ----------------------------------------------------------------------
104    Name: Matthias Blume (http://www.tti-c.org/blume/)
105    Date: 2005/11/21 14:00:00 CST
106    Tag: blume-20051121-html-lib
107    Description:
108    
109    Enabling $/html-lib.cm in config/preloads did not work.  This is fixed
110    now.  (Since the anchor mapping for html-lib.cm is not yet in effect
111    at the time when config/preloads is processed, the library has to
112    be referred to by another name.  In 110.57 this would be
113    $SMLNJ-LIB/HTML/html-lib.cm.  I arranged for $smlnj/smlnj-lib/html-lib.cm
114    to be valid as well.)
115    
116    Thanks to Todd Wilson (Fresno) for alerting me to this issue.
117    
118    ----------------------------------------------------------------------
119    Name: Matthias Blume (http://www.tti-c.org/blume/)
120    Date: 2005/11/19 13:45:00 CST
121    Tag: Release_110_57
122    Description:
123    
124    Moved release tag for 110.57.
125    
126    ----------------------------------------------------------------------
127    Name: Matthias Blume (http://www.tti-c.org/blume/)
128    Date: 2005/11/19 23:35:00 CST
129    Tag: blume-20051119-fish
130    Description:
131    
132    Fixed a problem in config/install.sh where it tries to "fish" the
133    name of the CM metadata directory from the wrong place (because
134    the physical location of basis.cm has changed).  Also, corrected the
135    path anchor for $/html-lib.cm.  (Thanks to M. Fluet for pointing out
136    these problems.)
137    
138    ----------------------------------------------------------------------
139    Name: Matthias Blume (http://www.tti-c.org/blume/)
140    Date: 2005/11/16 14:30:00 CST
141    Tag: blume-orig-110_57
142    Description:
143    
144    New working version (110.57).  NEW BOOTFILES!
145    
146    UPDATE: The release tag has been moved to include a fix for a brown-paper-
147            bag glitch (see above).
148    
149    ----------------------------------------------------------------------
150    Name: Matthias Blume (http://www.tti-c.org/blume/)
151    Date: 2005/11/16 10:25:00 CST
152    Tag: blume-22051116-btrace
153    Description:
154    
155    Fixed problem with bogus exception message when using back-trace
156    facility.
157    
158    ----------------------------------------------------------------------
159    Name: Matthias Blume (http://www.tti-c.org/blume/)
160    Date: 2005/11/15 22:35:00 CST
161    Tag: blume-22051115-array2copy
162    Description:
163    
164    Added simple implementation of Array2.copy.  (Warning: mostly untested.)
165    
166    ----------------------------------------------------------------------
167    Name: Matthias Blume (http://www.tti-c.org/blume/)
168    Date: 2005/11/15 21:35:00 CST
169    Tag: blume-22051115-versiontool
170    Description:
171    
172    Reversed change to src/system/smlnj/internals/versiontool.cm.  This
173    file gets loaded as a tool -- by the equivalent of CM.make during the
174    run of CMB.make.  Thus, CMB's path configuration is meaningless for
175    it.  Instead, it has the status of "user code", so it should use
176    $/basis.cm to refer to the Basis library.  (At least that's true for
177    the purpose of bootstrapping the previous change.  In the future it
178    might make sense to have versiontool.cm refer to
179    $smlnj/basis/basis.cm, i.e., the version of the Basis that the
180    compiler itself uses.)
181    
182    Also patched src/system/testml to have it activate those extra anchor
183    bindings in config/extrapathconfig.
184    
185    ----------------------------------------------------------------------
186    Name: Matthias Blume (http://www.tti-c.org/blume/)
187    Date: 2005/11/15 16:35:00 CST
188    Tag: blume-22051115-libraries
189    Description:
190    
191    This change affects the way the following libraries are tied into
192    the system:
193    
194       $/basis.cm
195       $/smlnj-lib.cm
196       $/pp-lib.cm
197       $/controls-lib.cm
198       $/html-lib.cm
199       $/ml-yacc-lib.cm
200    
201    These libraries are now internally (as seen from the source code
202    of the implementation itself) known by the following names:
203    
204       $smlnj/basis/basis.cm
205       $smlnj/smlnj-lib/smlnj-lib.cm
206       $smlnj/smlnj-lib/pp-lib.cm
207       $smlnj/smlnj-lib/controls-lib.cm
208       $smlnj/smlnj-lib/html-lib.cm
209       $smlnj/ml-yacc/ml-yacc-lib.cm
210    
211    This makes it possible to work with code that requires different
212    versions of these libraries, and which refers to these libraries using
213    their "default" names (i.e., the first set of names above).  In other
214    words, one can un-define or re-define those default names without
215    compromising the proper functioning of the compiler itself.
216    
217    A similar procedure had already been performed for several of the
218    MLRISC libraries that are linked into the compiler.  I did some
219    cleanup on this code.
220    
221    A new file in the config directory (named extrapathconfig) is
222    responsible for setting up path anchors that the compiler itself does
223    not need, but that are typically required by user code.
224    
225    ----------------------------------------------------------------------
226    Name: Matthias Blume
227    Date: 2005/11/07 17:05:00 CST
228    Tag: blume-20051107-slice
229    Description:
230    
231    Fixed erroneous out-of-bounds test in the "update" function of
232    various *ArraySlice modules.  (Thanks to Vesa A. Norrman for pointing
233    out the problem.)
234    
235    Pushed some Basis changes through ML-Lex, CML, and eXene.
236    
237    ----------------------------------------------------------------------
238    Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
239    Date: 2005/11/07
240    Description:
241    Fixed a Basis incompatibility: The depreciated function Substring.all
242    was removed (use Substring.full instead).
243    
244    ----------------------------------------------------------------------
245    Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
246    Date: 2005/11/05
247    Description:
248    Tweaked interval set API in SML/NJ library; see the CHANGES file for details.
249    
250    ----------------------------------------------------------------------
251    Name: Matthias Blume (http://www.tti-c.org/blume/)
252    Date: 2005/11/02 16:25:00 CST
253    Tag: blume-20051102-hex
254    Description:
255    
256    * Runtime system bootstrap code now accepts hex digits in BOOTLIST
257      in either upper case or lower case format.
258    * Pushed changes to names of Pack<N>{Big,Little} structures through
259      CML and eXene.
260    
261    ----------------------------------------------------------------------
262    Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
263    Date: 2005/11/02
264    Description:
265    Fixed a Basis incompatibility: Pack<N>{Big,Little} structures should
266    be called PackWord<N>{Big,Little}.
267    
268    ----------------------------------------------------------------------
269    Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
270    Date: 2005/10/28
271    Description:
272    Fixed a minor Basis incompatibility: hex digits should be upper case.
273    
274    ----------------------------------------------------------------------
275    Name: Matthias Blume (blume (at) tti - c (dot) org)
276    Date: 2005/10/25 16:00:00 CDT
277    Tag: Release_110_56
278    Description:
279    
280    New working version (110.56).  NEW BOOTFILES!
281    
282    ----------------------------------------------------------------------
283    Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
284    Date: 2005/10/25
285    Description:
286    Added interval sets to utility library (signatures INTERVAL_DOMAIN
287    and INTERVAL_SET, and functor IntervalSetFn).
288    
289    ----------------------------------------------------------------------
290    Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
291    Date: 2005/10/14
292    Description:
293    Add Zhong Shao's fix for datatype equality functions.
294    
295    ----------------------------------------------------------------------
296    Name: Allen Leung (leunga (at) reservoir (dot) com)
297    Date: 2005/10/14 12:30:00 EST
298    Tag: leunga-20051014-x86-spill
299    Description:
300    
301    Bug fix a bug found by Carl Hauser.
302    There was a typo in the reload code for FCMP in x86SpillInstr.sml.
303    
304    ----------------------------------------------------------------------
305    Name: Allen Leung (leunga (at) reservoir (dot) com)
306    Date: 2005/10/14 11:15:00 EST
307    Tag: leunga-20051014-x86-asm
308    Description:
309    
310    Removed some debugging code in file x86Asm.sml.
311    The function emit_operand was printing out debugging output.
312    
313    ----------------------------------------------------------------------
314    Name: John Reppy
315    Date: 2005/07/27
316    Tag:
317    Description:
318    
319    Fixed ml-lex to recognize "\r" as representing carriage return.
320    
321    ----------------------------------------------------------------------
322    Name: John Reppy
323    Date: 2005/07/27
324    Tag:
325    Description:
326    
327    Fixed ml-yacc to work on files with non-native end-of-line
328    encodings (e.g., Windows text file on a Unix system).
329    
330    ----------------------------------------------------------------------
331    Name: John Reppy
332    Date: 2005/07/20
333    Tag:
334    Description:
335    
336    Added changes from Dominic Evans (oldmanuk (at) gmail (dot) com)
337    to support HPUX 11.
338    
339    ----------------------------------------------------------------------
340    Name: John Reppy
341    Date: 2005/07/06
342    Tag:
343    Description:
344    
345    Changes to the SML/NJ library.  See smlnj-lib/CHANGES for details.
346    
347    ----------------------------------------------------------------------
348    Name: Matthias Blume (blume (at) tti - c (dot) org)
349    Date: 2005/07/06 00:45:00 CDT
350    Tag: blume-20050706-slice-copy
351    Description:
352    
353    Fixed reversed logic for deciding whether to "copy up" or "copy down"
354    in *-array-slice.sml.
355    
356    ----------------------------------------------------------------------
357    Name: Allen Leung (leunga (at) reservoir (dot) com)
358    Date: 2005/05/31 17:00:00 EST
359    Tag: leunga-20050531-cygwin-fault-2
360    Description:
361    
362    A typo in the cygwin code fixed.
363    
364    ----------------------------------------------------------------------
365    Name: Allen Leung (leunga (at) reservoir (dot) com)
366    Date: 2005/05/31 16:47:00 EST
367    Tag: leunga-20050531-cygwin-fault
368    Description:
369    
370    Updated Cygwin's fault/signal handling to match the Windows version.
371    Updated the export list.
372    
373    ----------------------------------------------------------------------
374    Name: Matthias Blume (blume (at) tti - c (dot) org)
375    Date: 2005/05/18 13:35:00 CDT
376    Tag: Release_110_54
377    Description:
378    
379    New working version (110.54).  NEW BOOTFILES!
380    
381    ----------------------------------------------------------------------
382    Name: Matthias Blume (blume (at) tti - c (dot) org)
383    Date: 2005/05/18 11:58:00 CDT
384    Tag: blume-20050518-installer
385    Description:
386    
387    Added support scripts for Mac OS X PackageMaker and modified
388    config/install.sh so that it supports re-dumping a heap image
389    after customization.
390    
391    ----------------------------------------------------------------------
392    Name: Matthias Blume (blume (at) tti - c (dot) org)
393    Date: 2005/05/18 10:55:00 CDT
394    Tag: blume-20050518-realdiv-noovld
395    Description:
396    
397    Un-overloaded / to work around bug in overloading resolution code.
398    
399    ----------------------------------------------------------------------
400    Name: Matthias Blume (blume (at) tti - c (dot) org)
401    Date: 2005/05/16 23:50:00 CDT
402    Tag: blume-20050516-redump-heap
403    Description:
404    
405    Added mechanism for re-creating a heap file for the interactive system
406    after configuration variables have been changed.
407    
408       CM.redump_heap : string -> unit
409    
410    This is much like SMLofNJ.exportML, but starting from the resulting
411    heap does not return to the caller of CM.redump_heap but
412    restarts the interactive system from scratch.  The original call of
413    CM.redump_heap does not return but ends the interactive session.
414    Thus, CM.redump_heap is a lot like SMLofNJ.exportFn.
415    
416    Internally, redump_heap winds the dynamic execution context back to
417    the point where the original heap image was created and re-executes
418    the heap image generation code in the boot code.
419    
420    ----------------------------------------------------------------------
421    Name: Matthias Blume (blume (at) tti - c (dot) org)
422    Date: 2005/05/09 21:55:00 CDT
423    Tag: blume-20050509-word64
424    Description:
425    
426    Added a hack to the existing hack known as Word64 to make fromString
427    behave correctly.  I am still not sure whether Word64.scan will work
428    as specified with respect to the interaction of radix and prefix.
429    
430    ----------------------------------------------------------------------
431    Name: Allen Leung (leunga (at) reservoir (dot) com)
432    Date: 2005/05/04 11:50:00 EST
433    Tag: leunga-20050504-checkgc
434    Description:
435    
436    Added a gc protocol checking phase.  This phase is enabled with
437    the flag "check-gc".  "debug-check-gc" turns on the verbose mode.
438    
439    ----------------------------------------------------------------------
440    Name: Matthias Blume (blume (at) tti - c (dot) org)
441    Date: 2005/05/04 10:45:00 CDT
442    Tag: blume-20050504-intinf
443    Description:
444    
445    Fixed a bug in the implementation of div and mod for IntInf.
446    Thanks to Neophytos Michael for reporting the problem.
447    
448    ----------------------------------------------------------------------
449    Name: Matthias Blume (blume (at) tti - c (dot) org)
450    Date: 2005/05/04 10:35:00 CDT
451    Tag: blume-20050504-join
452    Description:
453    
454    Added a "join" combinator to the ParserComb module in smlnj-lib.cm.
455    
456    ----------------------------------------------------------------------
457    Name: Matthias Blume (blume (at) tti - c (dot) org)
458    Date: 2005/02/28 23:40:00 CST
459    Tag: blume-20050228-mVar
460    Description:
461    
462    Fixed serious bug (brown paper bag variety) in new implementation of
463    structure Atom in CML.  (I had accidentally used a mailbox instead of
464    an mvar, leaving the door open for races.)
465    
466    ----------------------------------------------------------------------
467    Name: Matthias Blume (blume (at) tti - c (dot) org)
468    Date: 2005/02/25 15:00:00 CST
469    Tag: Release_110_53
470    Description:
471    
472    New working version (110.53).  NEW BOOTFILES!
473    
474    ----------------------------------------------------------------------
475    Name: Matthias Blume (blume (at) tti - c (dot) org)
476    Date: 2005/02/25 14:50:00 CST
477    Tag: blume-20050225-susp
478    Description:
479    
480    Brought back SMLofNJ.Susp.  The underlying suspension type is the one
481    implemented in Core -- which means that it is the same as the one used
482    by the lazy extension.
483    
484    ----------------------------------------------------------------------
485    Name: Matthias Blume (blume (at) tti - c (dot) org)
486    Date: 2005/02/24 16:50:00 CST
487    Tag: blume-20050224-cml-atom
488    Description:
489    
490    Simpler and at the same time more general implementation of structure
491    Atom in CML.
492    
493    ----------------------------------------------------------------------
494    Name: Matthias Blume (blume (at) tti - c (dot) org)
495    Date: 2005/02/15 17:35:00 CST
496    Tag: blume-20050215-tools
497    Description:
498    
499    Created new "tools" directory under "src" and moved "TraceDebugProf"
500    there.
501    
502    ----------------------------------------------------------------------
503    Name: Matthias Blume (blume (at) tti - c (dot) org)
504    Date: 2005/02/10 17:55:00 CST
505    Tag: blume-20050210-longlong
506    Description:
507    
508    Implemented "long long" arguments and results for NLFFI.  (Only the
509    PPC/MacOS implementation is complete, the other backends still need to
510    be updated.)
511    
512    ----------------------------------------------------------------------
513    Name: Matthias Blume (blume (at) tti - c (dot) org)
514    Date: 2005/01/24 17:40:00 CST
515    Tag: blume-20050124-mlyacc
516    Description:
517    
518    Minor cleanup in ML-Yacc rule printing mechanism.  This should fix a
519    problem with certain "as" patterns which previously got rendered
520    using incorrect syntax.
521    
522    ----------------------------------------------------------------------
523    Name: Matthias Blume (blume (at) tti - c (dot) org)
524    Date: 2005/01/18 12:00:00 CST
525    Tag: blume-20050118-profile
526    Description:
527    
528    Made time profiling code (interrupt handler) in runtime system aware
529    of new array representation.
530    
531    ----------------------------------------------------------------------
532    Name: Matthias Blume (blume (at) tti - c (dot) org)
533    Date: 2005/01/14 18:00:00 CST
534    Tag: blume-20050114-heap2exec
535    Description:
536    
537    Implemented new (but still experimental) heap2exec facility.  This is
538    tested under Mac OS X and should work under Linux (will test shortly).
539    It will probably also work on the Sparc (will test some time later).
540    
541      - removed old "HACKED_STANDALONE" hack from runtime
542    
543    To be able to test this, uncomment the request for "heap2asm" in
544    config/targets prior to installation.  (Notice that this is different
545    from "heap2exec" mentioned below.  Not a typo.)
546    
547    To perform an actual test, run the command
548    
549       $ bin/heap2exec heapfile execfile
550    
551    (You can put heap2exec on your shell's path.)
552    
553    For example, run
554    
555      $ bin/heap2exec bin/.heap/ml-yacc.ppc-darwin mly
556    
557    This will create a standalone executable called "mly" which you can
558    then invoke directly as a command.
559    
560    ----------------------------------------------------------------------
561    Name: Matthias Blume (blume (at) tti - c (dot) org)
562    Date: 2005/01/07 11:44:00 CST
563    Tag: blume-20050107-mlstring
564    Description:
565    
566    fixed off-by-one error in ML_STRING macro (globals.c)
567    
568    ----------------------------------------------------------------------
569    Name: Matthias Blume (blume (at) tti - c (dot) org)
570    Date: 2004/12/23 18:00:00 CST
571    Tag: blume-20041223-santa
572    Description:
573    
574    Made ml-build script "smarter" (but only very little).
575    
576    ----------------------------------------------------------------------
577    Name: Matthias Blume (blume (at) tti - c (dot) org)
578    Date: 2004/12/21 15:05:00 CST
579    Tag: blume-20041221-longlong
580    Description:
581    
582    * Implemented access to signed and unsigned long long data in NLFFI.
583    (The parameter-passing part of the picture has not complete. But
584    data structure access seems to work.)
585    
586    * Fixed CM's incorrect assumption that the PPC is little-endian.
587      (On the Mac, it is big-endian.  And that's currently our only
588      PPC platform.)
589    
590    ----------------------------------------------------------------------
591    Name: Matthias Blume (blume (at) tti - c (dot) org)
592    Date: 2004/12/21 12:50:00 CST
593    Tag: blume-20041221-memory
594    Description:
595    
596    Some cleanup in the $c/memory.cm library: separated some concerns by
597    moving allocation code and memory access code each into their own
598    files.
599    
600    ----------------------------------------------------------------------
601    Name: Allen Leung (leunga (at) reservoir (dot) com)
602    Date: 2004/12/17 16:12:00 EST
603    Tag: leunga-20041217-cygwin-smlnj-home
604    Description:
605    
606        The Unix I/O library of SML/NJ on cygwin does not understand
607    Windows style pathname, so problems arise when SMLNJ_HOME is set
608    to a Windows style pathname.  _run-sml now converts SMLNJ_HOME
609    to a POSIX pathname on cygwin.
610    
611    ----------------------------------------------------------------------
612    Name: Matthias Blume (blume (at) tti - c (dot) org)
613    Date: 2004/12/16 13:35:00 CST
614    Tag: Release_110_52
615    Description:
616    
617    Last-minute changes incorporated into 110.52.  Release tag moved.
618    
619    The changes:
620       - HashString.hashString' -> HashString.hashSubstring
621       - bug fix in UnivariateStats
622    
623    ----------------------------------------------------------------------
624    Name: Matthias Blume (blume (at) tti - c (dot) org)
625    Date: 2004/12/15 23:40:00 CST
626    Tag: blume-20041215-hashSubstring
627    Description:
628    
629    - HashString.hashString' -> HashString.hashSubstring
630    - corresponding changes in atom.sml
631    - "de-compressed" (aka. un-obfuscated) code for UnivariateStats and
632      added some comments
633    
634    ----------------------------------------------------------------------
635    Name: Matthias Blume (blume (at) tti - c (dot) org)
636    Date: 2004/12/15 15:30:00 CST
637    Tag: (Release_110_52)
638    Description:
639    
640    New working version (110.52).  NEW BOOTFILES!
641    
642    ----------------------------------------------------------------------
643    Name: Matthias Blume (blume (at) tti - c (dot) org)
644    Date: 2004/12/15 12:45:00 CST
645    Tag: blume-20041215-spaces
646    Description:
647    
648    More on the space problem (this time for Win32).
649    
650    ----------------------------------------------------------------------
651    Name: Matthias Blume (blume (at) tti - c (dot) org)
652    Date: 2004/12/14 17:30:00 CST
653    Tag: blume-20041214-spaces
654    Description:
655    
656    Hacked some of the scripts (in particular: the installer) to cope with
657    spaces in filenames a bit better.  But beware: the current "solution"
658    is likely still full of bugs and inherently incomplete.  (We need to
659    do away with those shell scripts for a comprehensive solution.)
660    
661    ----------------------------------------------------------------------
662    Name: Matthias Blume (blume (at) tti - c (dot) org)
663    Date: 2004/12/13 14:45:00 CST
664    Tag: blume-20041213-ml-makedepend
665    Description:
666    
667    Fixed bug in code for ml-makedepend.
668    
669    ----------------------------------------------------------------------
670    Name: Matthias Blume (blume (at) tti - c (dot) org)
671    Date: 2004/12/09 16:30:00 CST
672    Tag: blume-20041209-statistics
673    Description:
674    
675    Added two simple but potentially useful statistics modules to SML/NJ Library.
676    (See CHANGES file there.)
677    
678    ----------------------------------------------------------------------
679    Name: Matthias Blume (blume (at) tti - c (dot) org)
680    Date: 2004/12/01 16:50:00 CST
681    Tag: blume-20041201-atom
682    Description:
683    
684    smlnj-lib:
685    
686      Added function HashString.hashString' for substrings.
687      Hand-inlined CharVector.foldl into HashString (for speed).
688      Modified implementation of structure Atom to avoid extracting
689      strings from substrings unless necessary.
690      (Also see CHANGES file for smlnj-lib.)
691    
692    ----------------------------------------------------------------------
693    Name: Matthias Blume (blume (at) tti - c (dot) org)
694    Date: 2004/11/24 22:15:00 CST
695    Tag: blume-20041124-cml
696    Description:
697    
698    Made sure CML compiles when Position = Int64.
699    
700    ----------------------------------------------------------------------
701    Name: Matthias Blume (blume (at) tti - c (dot) org)
702    Date: 2004/11/24 14:30:00 CST
703    Tag: blume-20041124-position
704    Description:
705    
706    The compiler can now be compiled in a mode that makes structure Position
707    equal to Int64.  The default, however, is unchanged (Position = Int31)
708    for the time being.
709    
710    To enable 64-bit positions, use the following procedure:
711    
712    1. Start sml
713    2. Autoload $smlnj/cmb.cm (if not already autoloaded)
714    3. Type
715          #set (CMB.symval "USE_64_BIT_POSITIONS") (SOME 1);
716    4. Run CMB.make() as usual.
717    
718    This is barely tested.  The only test so far was a little SML program
719    counting the number of characters in an 8-gigabyte file by
720    reading it character-by-character.  That test was successful.
721    
722    In support of 64-bit positions, a number of new functions have been
723    added to the runtime system.
724    
725    ----------------------------------------------------------------------
726    Name: Matthias Blume (blume (at) tti - c (dot) org)
727    Date: 2004/11/23 14:45:00 CST
728    Tag: blume-20041123-useFile
729    Description:
730    
731    Fixed a problem with unhelpful error messages related to problems with
732    .cm- or .sml files that appear as part of the sml command line.
733    
734    ----------------------------------------------------------------------
735    Name: Matthias Blume (blume (at) tti - c (dot) org)
736    Date: 2004/11/18 15:40:00 CST
737    Tag: Release_110_51
738    Description:
739    
740    New working version (110.51).  NEW BOOTFILES!
741    
742    ----------------------------------------------------------------------
743    Name: Matthias Blume (blume (at) tti - c (dot) org)
744    Date: 2004/11/18 15:35:00 CST
745    Tag: <none>
746    Description:
747    
748    Enabled dlopen and friends for FreeBSD (as recommended by Johannes 5
749    Joemann).
750    
751    ----------------------------------------------------------------------
752    Name: Allen Leung (leunga (at) reservoir (dot) com)
753    Date: 2004/11/17 16:05:21 EST 2004
754    Tag: leunga-20041117-mlrisc-live-kill
755    Description:
756    
757       Added support for MLTree constructs LIVE and KILL
758    to all the architectures.
759    
760    ----------------------------------------------------------------------
761    Name: Matthias Blume (blume (at) tti - c (dot) org)
762    Date: 2004/11/13 00:20:00 CST
763    Tag: blume-20041113-versiontool
764    Description:
765    
766    - Stripped down the versiontool: It now only handles the version number.
767      The date string is generated at bootstrap time (during makeml).
768    
769    - In a previous commit, fixed a minor issue with how polyequal is being
770      translated.  In particular, the code now "looks through" abstractions.
771      This results in slightly fewer polyEqual warnings and hopefully slightly
772      more efficient code.  Important examples for where this matters are
773      the new int64 and word64 types.
774    
775    ----------------------------------------------------------------------
776    Name: Matthias Blume (blume (at) tti - c (dot) org)
777    Date: 2004/11/12 00:30:00 CST
778    Tag: blume-20041112-int64
779    Description:
780    
781    Structure Int64 fully hooked in.  (The implementation is not very
782    efficient, though.)
783    
784    ----------------------------------------------------------------------
785    Name: Matthias Blume (blume (at) tti - c (dot) org)
786    Date: 2004/11/11 17:30:00 CST
787    Tag: blume-20041111-more64
788    Description:
789    
790    All the pieces of Word64 are now there, with the exception of the
791    conversions from and to LargeWord. (Eventually these need to be identities,
792    but for the time being they don't even make sense because LargeWord is
793    32-bit wide.)
794    
795    Also started to add similar support for Int64, but major pieces of that
796    are still missing.
797    
798    ----------------------------------------------------------------------
799    Name: Matthias Blume (blume (at) tti - c (dot) org)
800    Date: 2004/11/11 00:15:00 CST
801    Tag: blume-20041111-word64
802    Description:
803    
804    Structure Word64 is now (almost) complete, word literals and patterns
805    seem to work.  There are a few odd pieces missing.  In particular,
806    I didn't do the {from,to}LargeWord parts because LargeWord is still Word32
807    at the moment.
808    
809    Making Word64 official would mean that LargeWord becomes Word64.  But
810    this requires extreme care because most word-word conversions have to
811    go through LargeWord, so making a mistake means loss of efficiency or
812    worse.  Eventually there will be a solution similar to (but actually
813    simpler than) what I did with IntInf.
814    
815    ----------------------------------------------------------------------
816    Name: Matthias Blume (blume (at) tti - c (dot) org)
817    Date: 2004/11/10 18:12:00 CST
818    Tag: blume-20041110-64bit
819    Description:
820    
821    More 64-bit hacking (but still not even half-way there yet).
822    Also, some assorted improvements to the handling of 8-bit words.
823    
824    ----------------------------------------------------------------------
825    Name: Matthias Blume (blume (at) tti - c (dot) org)
826    Date: 2004/11/09 17:50:00 CST
827    Tag: <none>
828    Description:
829    
830    Started adding some infrastructure for supporting 64-bit int- and
831    word-types.  (Still in its very early stages.)
832    
833    ----------------------------------------------------------------------
834    Name: Matthias Blume (blume (at) tti - c (dot) org)
835    Date: 2004/10/28 10:45:00 CDT
836    Tag: Release_110_50
837    Description:
838    
839    New working version (110.50).  NEW BOOTFILES!
840    
841                 =====================
842    
843    Also:
844    
845      - Changend config/srcarchiveurl from a file just
846        containing the URL string into a file containing
847        shell script code.  The code has access to the $VERSION variable.
848      - Made corresponding changes to config/install.sh and config/unpack.
849      - Default contents of config/srcarchiveurl uses $VERSION and
850        normally does not have to be edited to reflect a version change.
851    
852      (As a result, a version change can be done by just editing
853       config/version, the rest is now automatic.)
854    
855    ----------------------------------------------------------------------
856    Name: Matthias Blume (blume (at) tti - c (dot) org)
857    Date: 2004/10/27 17:50:00 CDT
858    Tag: blume-20041027-btrace-msg
859    Description:
860    
861    BackTrace.monitor now also reports the source of the exception that
862    triggered the trace.
863    
864    ----------------------------------------------------------------------
865    Name: Matthias Blume (blume (at) tti - c (dot) org)
866    Date: 2004/10/27 17:20:00 CDT
867    Tag: blume-20041027-x86-c-calls
868    Description:
869    
870    This is the HISTORY entry for two earlier commits, both concerning
871    the x86 c-calls code in MLRISC:
872    
873      - added a missing LOAD in the code that deals with struct arguments
874      - made sure the caller does not add the wrong number of bytes to the
875        stack pointer after a call of a function returning a struct
876        (the callee already pops the implicit argument which points to
877         the space reserved for the result)
878    
879    ----------------------------------------------------------------------
880    Name: Allen Leung (leunga (at) reservoir (dot) com)
881    Date: 2004/10/24 14:00:00 EST
882    Tag: leunga-20041024-x86-gas-fucomip
883    Description:
884    
885       John discovered a bug in the syntax of fucomip.
886       The opcodes FU?COMIP? have been changed to
887    
888           fu?comip? %st(i), %st
889    
890    ----------------------------------------------------------------------
891    Name: Matthias Blume (blume (at) tti - c (dot) org)
892    Date: 2004/10/20 15:06:00 CDT
893    Tag: blume-20041020-standalone-backtrace
894    Description:
895    
896    Added a mechanism for getting back-trace information from standalone
897    programs.  Here is how it works:
898    
899       1. The part of the program from which you want to get backtrace
900          information (usually the whole program) should be wrapped with
901          BackTrace.monitor.  This is a (unit->'a)->'a function, and your
902          main program could be modified from something like
903    
904             fun main (pgm, args) = ...
905    
906         to
907    
908            fun main (pgm, args) = BackTrace.monitor (fn () => ...)
909    
910       2. To be able to access BackTrace.monitor, you have to add
911          library $smlnj-tdp/plugins.cm to the .cm file that contains your
912          main function.
913    
914       3. Remove all compiled code (i.e., all the .cm/ subdirectories that
915          CM might have created in the past for your project).
916    
917       4. Build the system using this command line:
918    
919             ml-build -Ctdp.instrument=true \$smlnj-tdp/back-trace.cm \
920                      myprog.cm MyProg.main myprog
921    
922          instead of the usual
923    
924             ml-build myprog.cm MyProg.main myprog
925    
926    I changed a library name:
927    
928       $/trace-debug-profile.cm --> $smlnj-tdp/plugins.cm
929    
930    New libraries:
931    
932       $smlnj-tdp/back-trace.cm
933          -- when loaded causes the back-trace plugin to be installed
934       $smlnj-tdp/coverage.cm
935          -- when loaded causes the coverage plugin to be installed
936    
937    ----------------------------------------------------------------------
938    Name: Matthias Blume (blume (at) tti - c (dot) org)
939    Date: 2004/10/18 16:45:00 CDT
940    Tag: blume-20041018-groupowner
941    Description:
942    
943    Added an "obsolete" warning for the "group owner" syntax to CM's parser.
944    
945    Eliminated group owner specs from .cm files throughout the source tree.
946    
947    ----------------------------------------------------------------------
948    Name: Matthias Blume (blume (at) tti - c (dot) org)
949    Date: 2004/10/15 15:45:00 CDT
950    Tag: blume-20041015-coverage
951    Description:
952    
953    * Test coverage tool added!
954    * Further reorganization of tracing-, debugging-, and profiling support:
955    
956        - moved original BTImp -- now called BackTrace -- into a separate
957          library called $/trace-debug-profile.cm
958        - eliminated all mentions of BTrace from SMLofNJ.Internals
959        - only the instrumentation mechanism is now left in the compiler proper
960        - BackTrace module is a plugin which is NOT plugged in by default
961        - Coverage module is another such plugin
962    
963    To get the benefits of any of these plugin modules, the code in
964    question must be compiled with tdp instrumentation turned on.  This
965    can be done by setting SMLofNJ.Internals.TDP.mode to true.
966    (The ref cell is also controlled via the -Ctdp.instrument=... switch.)
967    
968    Plugins are selected at link time.  (Pre-compiled instrumented code
969    can be re-loaded with different plugins in effect.)  When an
970    instrumented module is linked, whatever plugins are at that time
971    enabled will come into effect for that module.
972    
973    To enable the back-trace plugin, load library $/trace-debug-profile.cm
974    and invoke BackTrace.install() (e.g., from the interactive prompt).
975    To enable the coverage plugin, load the same library and invoke
976    Coverage.install().
977    
978    Back-traces are generated automatically on uncaught exceptions and
979    when the code in question explicitly invokes BackTrace.trigger().
980    
981    Coverage (and execution frequency-) information must be queried
982    explicitly by calling Coverage.not_covered and Coverage.hot_spots.
983    
984    ----------------------------------------------------------------------
985    Name: Matthias Blume (blume (at) tti - c (dot) org)
986    Date: 2004/10/14 17:40:00 CDT
987    Tag: blume-20041014-tdp-core
988    Description:
989    
990    Snapshot of a significant overhaul of how the trace/debug/profile support
991    is hooked into the system (specifically: Core and SMLofNJ.Internals).
992    
993    ----------------------------------------------------------------------
994    Name: Matthias Blume (blume (at) tti - c (dot) org)
995    Date: 2004/10/13 16:34:00 CDT
996    Tag: blume-20041013-tdp
997    Description:
998    
999    Some rationalization of names:
1000    
1001       structure BTrace -> structure TDPInstrument
1002       etc.
1003    
1004    This is is preparation of using the original back-trace
1005    instrumentation for other purposes.  "TDP" stands for
1006    Trace/Debug/Profile.
1007    
1008    The control flag controlling whether instrumentation is on or off is now
1009    registered under a different name, so instead of running sml as
1010    
1011     sml -Cinstrument.btrace-mode=true
1012    
1013    one has to say
1014    
1015     sml -Ctdp.instrument=true
1016    
1017    ----------------------------------------------------------------------
1018    Name: Matthias Blume (blume (at) tti - c (dot) org)
1019    Date: 2004/10/11 16:37:00 CDT
1020    Tag: blume-20041011-regions
1021    Description:
1022    
1023    Made some minor modifications to elabcore.sml to have source regions
1024    be propagated more tightly -- resulting in better (i.e., smaller)
1025    regions being reported in error- and debug messages.
1026    
1027    ----------------------------------------------------------------------
1028    Name: Matthias Blume (blume (at) tti - c (dot) org)
1029    Date: 2004/10/08 22:50:00 CDT
1030    Tag: blume-20041008-cmkw
1031    Description:
1032    
1033    Fixed handling of keywords in .cm files: After seeing "is" the lexer
1034    treats subsequent occurrences of "group", "library", "source", "is",
1035    "*", and "-" as ordinary identifiers rather than keywords.
1036    
1037    Most seriously, this fixes a problem with CM's "shell" tool.  The tool
1038    is supposed to accept a tool argument called "source", but this did
1039    not work because of the clash with the keyword.
1040    
1041    ----------------------------------------------------------------------
1042    Name: Matthias Blume (blume (at) tti - c (dot) org)
1043    Date: 2004/10/07 16:00:00 CDT
1044    Tag: blume-20041007-cleanup
1045    Description:
1046    
1047    Assorted cleanup work:
1048    
1049      - got rid of intstrmap in favor of using the library's
1050        hash table implementation
1051      - threw out most of the pathnames stuff, as it was not used anyway
1052      - simplified tokentable implementation
1053      - fixed some minor spelling errors
1054    
1055    ----------------------------------------------------------------------
1056    Name: Matthias Blume (blume (at) tti - c (dot) org)
1057    Date: 2004/10/06 15:15:15 CDT
1058    Tag: blume-20041006-handler
1059    Description:
1060    
1061    Cleaned up the absyn to reflect the invariant that HANDLE always
1062    carries a FNexp as part of the type definition.  This eliminates some
1063    superfluous sanity checks at runtime down the road.
1064    
1065    Some minor cleanup of the btrace code.
1066    
1067    ----------------------------------------------------------------------
1068    Name: Matthias Blume (blume (at) tti - c (dot) org)
1069    Date: 2004/10/01 10:20:30 CDT
1070    Tag: blume-20041001-slave
1071    Description:
1072    
1073    Added hack to make slave mode work in the presence of the version
1074    tool.  (Still, since the master does two passes over the code for
1075    CMB.make, the release number gets bumped twice when slaves are
1076    attached.  I don't know if this is worth fixing...)
1077    
1078    ----------------------------------------------------------------------
1079    Name: Matthias Blume (blume (at) tti - c (dot) org)
1080    Date: 2004/09/30 10:55:00 CDT
1081    Tag: blume-20040930-version
1082    Description:
1083    
1084    * Moved the "version" magic into its own little library under
1085      src/system/smlnj/internal.  This avoids expensive reconstruction of
1086      a stable src/compiler/core.cm.
1087    
1088    * At the same time, structure CompilerVersion is now known as
1089      structure SMLNJVersion.
1090    
1091    * Arranged for the version tool to NOT kick in when rebuilding the system
1092      (makeml -rebuild, fixpt).  Otherwise one would never reach a fixpoint.
1093      Also, loading the versiontool does not work when rebuilding the system
1094      because CM is not properly initialized at that time.
1095    
1096    ----------------------------------------------------------------------
1097    Name: Matthias Blume (blume (at) tti - c (dot) org)
1098    Date: 2004/09/29 14:00:00 CDT
1099    Tag: blume-20040929-autoversion
1100    Description:
1101    
1102    Implemented some CM magic to have
1103        file src/compiler/TopLevel/main/version.sml
1104    generated automagically.
1105    The version is taken from two files: config/version and config/release.
1106    The first is expected to contain a two-part version number such as 110.49.
1107    The second should contain a single number, but it may be missing.
1108    
1109    If the environment variable VERSIONTOOL_BUMP_RELEASE is defined at the
1110    time the version tool is loaded (which is the first time you say CMB.make),
1111    then the tool will increment the value stored in config/release every
1112    time CMB.make is invoked.
1113    
1114    The binfile format is now insensitive to anything beyond the first
1115    two components of a version number, so bumping the release does not render
1116    binfiles incompatible.  Auto-bumping can be used to keep track of versions
1117    during development without invalidating existing binfiles.
1118    
1119    In any case, every CMB.make updates the date information in version.sml.
1120    (This is the date that is printed in the banner.)
1121    
1122    ----------------------------------------------------------------------
1123    Name: Matthias Blume (blume (at) tti - c (dot) org)
1124    Date: 2004/09/28 10:53:00 CDT
1125    Tag: blume-20040928-controls
1126    Description:
1127    
1128    Some cleanup of the controls code.
1129    
1130    ----------------------------------------------------------------------
1131    Name: Matthias Blume (blume (at) tti - c (dot) org)
1132    Date: 2004/09/27 22:08:00 CDT
1133    Tag: blume-20040927-controls
1134    Description:
1135    
1136    Added two pieces of functionality to the Controls interface:
1137    
1138    1.   val save'restore: 'a control -> unit -> unit
1139    
1140       grabs the current value of the control in stage 1 and restores it
1141       in stage 2.
1142    
1143    2.   val set' : 'a control * 'a -> unit -> unit
1144    
1145       stores the given value into the control in stage 2 (i.e., delayed)
1146       but does all error checking in stage 1.
1147       (This is for string controls that need to do parse their argument --
1148        something that might fail.  In some cases, notably in CM, one
1149        already knows the intended argument but wants to delay the actual
1150        assignment until a time when error recovery would be more difficult.)
1151    
1152    Changed the handling of controls in tool arguments to classes "sml" and
1153    "lazysml":
1154       - use Controls.save'restore as a more robust way of restoring the
1155         old value (in particular: without having to re-parse the string)
1156       - use controls to handle the "overload" keyword in the init group
1157         (I believe this change actually fixes a long-standing obscure bug.)
1158    
1159    ----------------------------------------------------------------------
1160    Name: Matthias Blume (blume (at) tti - c (dot) org)
1161    Date: 2004/09/27 17:00:00 CDT
1162    Tag: blume-20040927-lazysml
1163    Description:
1164    
1165    Added a new tool class called "lazysml" to CM's tool chest.  The only
1166    difference to "sml" is that compilation is done with Control.lazysml
1167    set to true.  A source of class "lazysml" is automatically recognized
1168    by a file name suffix of ".lml".
1169    
1170    In addition to the above feature, the original class "sml" now also
1171    supports a tool argument "lazy" which has the same effect.  As a
1172    result, the following three lines are equivalent:
1173    
1174        foo.sml : lazysml
1175        foo.sml : sml (lazy)
1176        foo.sml (lazy)
1177    
1178    The setting goes into effect both during parsing and during
1179    compilation.  The original setting is restored right after parsing and
1180    after compilation, respectively.
1181    
1182    In addition to all the above, there is also a general mechanism to set
1183    ANY of the "controls" that are available at the command line via
1184    "-C..." on a per-sml-file basis.  The same rules that apply for "lazy"
1185    apply as well. (In fact, "lazy" is implemented as a special case of
1186    the general mechanism.)
1187    
1188    The .cm file syntax uses a new keyword tool argument called "with".
1189    There are several ways of indicating the desired settings:
1190    
1191        foo.sml (with:parser.quotations=true)
1192        foo.sml (with:(name:parser.quotations value:true))
1193        foo.sml (with:(name:name1 value:value1 name:name2 value:value2 ...))
1194        foo.sml (with:(name1=value1 name2=value2 ...))
1195        foo.sml (with:(name1=value1 name:name2 value:value2 name3=value3 ...))
1196    
1197        etc.
1198    
1199    Another possible abbreviation is to leave out the =v or value:v part
1200    if the name refers to a boolean control (in which case the value is
1201    taken to be true).  Thus, one could get lazy sml also by saying:
1202    
1203        foo.sml (with:parser.lazy-keyword=true)
1204        foo.sml (with:parser.lazy-keyword)
1205        foo.sml (with:(name:parser.lazy-keyword value:true))
1206        foo.sml (with:(name:parser.lazy-keyword))
1207    
1208    ----------------------------------------------------------------------
1209    Name: Matthias Blume (blume (at) tti - c (dot) org)
1210    Date: 2004/09/24 16:22:00 CDT
1211    Tag: blume-20040924-ppc-long-branch
1212    Description:
1213    
1214    Turned message about "emiting long form of branch" off by default.
1215    Added a control flag to turn it back on when desired.
1216    
1217    ----------------------------------------------------------------------
1218    Name: Matthias Blume (blume (at) tti - c (dot) org)
1219    Date: 2004/09/24 16:05:00 CDT
1220    Tag: blume-20040924-rounding
1221    Description:
1222    
1223    Applied patch for setting rounding modes under Mac OS X.  Thanks to
1224    Melissa O'Neill for providing the code!
1225    
1226    ----------------------------------------------------------------------
1227    Name: Matthias Blume (blume (at) tti - c (dot) org)
1228    Date: 2004/09/23 17:30:00 CDT
1229    Tag: blume-20040923-envvars
1230    Description:
1231    
1232    1. Changed definition of type ControlRegistry.registry_tree to
1233       include control_info (i.e., the name of the controlling environment
1234       variable).
1235    
1236    2. Added command-line flags -e and -E to print the names of environment
1237       variables that can be used to control internal settings.
1238       (This uses the new API mentioned in 1.)
1239    
1240    ----------------------------------------------------------------------
1241    Name: Matthias Blume (blume (at) tti - c (dot) org)
1242    Date: 2004/09/13 16:50:00 CDT
1243    Tag: Release_110_49
1244    Description:
1245    
1246    New working version (110.49).  NEW BOOTFILES!
1247    
1248    ----------------------------------------------------------------------
1249    Name: Matthias Blume
1250    Date: 2004/09/13 16:20:00 CDT
1251    Tag: blume-20040913-config-mlrisc
1252    Description:
1253    
1254    Put target "mlrisc" back into the default list.
1255    (There is no harm in having it, and some users have expressed their
1256    wish to have "mlrisc" included by default.)
1257    
1258    ----------------------------------------------------------------------
1259    Name: John Reppy
1260    Date: 2004/09/13
1261    Tag: jhr-20040913-signals
1262    Description:
1263    
1264    Fixed the signal masking code to properly nest mask/unmask operations
1265    on a per-signal basis.
1266    
1267    ----------------------------------------------------------------------
1268    Name: Matthias Blume (blume (at) tti - c (dot) org)
1269    Date: 2004/09/08 13:20:00 CDT
1270    Tag: blume-20040908-heap-magic
1271    Description:
1272    
1273    Bumped the heap macig number to 0x09082004 to account for the changed
1274    layout of the ML frame under MacOS X.
1275    
1276    ----------------------------------------------------------------------
1277    Name: Allen Leung (leunga (at) reservoir (dot) com)
1278    Date: 2004/09/03 11:26:00 EST
1279    Tag: leunga-20040903-cygwin-install
1280    Description:
1281    
1282       Added a patch to _arch-n-opsys to enable the Cygwin runtime.
1283    The Cygwin runtime is turned on by setting the environment
1284    variable SMLNJ_CYGWIN_RUNTIME to 1.
1285    
1286    ----------------------------------------------------------------------
1287    Name: Matthias Blume (blume (at) tti - c (dot) org)
1288    Date: 2004/08/31 17:15:00 CDT
1289    Tag: blume-20040831-core
1290    Description:
1291    
1292    Added some exports to src/compiler/core.cm upon request by J. Joemann.
1293    
1294    ----------------------------------------------------------------------
1295    Name: Matthias Blume (blume (at) tti - c (dot) org)
1296    Date: 2004/08/30 17:55:00 CDT
1297    Tag: blume-20040830-installer
1298    Description:
1299    
1300    Upon request by Johannes Joemann:
1301    
1302    - improved ML code of installer to fall back to coping when renaming
1303      fails (i.e., when source and target are on different file systems);
1304      the code compiles but has yet to be tested in anger
1305    - removed mlrisc from list of default targets (config/targets)
1306    
1307    ----------------------------------------------------------------------
1308    Name: Matthias Blume (blume (at) tti - c (dot) org)
1309    Date: 2004/08/27 17:20:00 CDT
1310    Tag: blume-20040827-ptreql
1311    Description:
1312    
1313    Added ptreql primop to structure InlineT (upon request from Larry
1314    Paulson).
1315    
1316    ----------------------------------------------------------------------
1317    Name: Allen Leung (leunga (at) reservoir (dot) com)
1318    Date: 2004/08/15 21:21:00 EST
1319    Tag: leunga-110_48-udgraph
1320    Description:
1321    
1322    Another bug fix from Carl Hauser:
1323    
1324    diff /net/niflab/smlnj48/src/MLRISC/graphs/udgraph.sml udgraph.sml
1325    > 48c48
1326    > <              | rmv((e as (k,_))::es,L) = rmv(es,if k = i then es else
1327    > e::L)
1328    > ---
1329    > >              | rmv((e as (k,_))::es,L) = rmv(es,if k = i then L else e::L)
1330    > Without this, any deletion of an edge in an undirected graph does severe
1331    > violence to the graph.
1332    
1333    ----------------------------------------------------------------------
1334    Name: Allen Leung (leunga (at) reservoir (dot) com)
1335    Date: 2004/08/10 23:35:00 EST
1336    Tag: leunga-110_48-ppc
1337    Description:
1338    
1339       The IBM/MacOS syntax switch on PPC was incorrectly swapped.
1340    
1341    ----------------------------------------------------------------------
1342    Name: Matthias Blume (blume (at) tti - c (dot) org)
1343    Date: 2004/08/10 12:00:00 CDT
1344    Tag: Release_110_48
1345    Description:
1346    
1347    New working version (110.48).  NEW BOOTFILES!
1348    
1349    ----------------------------------------------------------------------
1350    Name: Allen Leung (leunga (at) reservoir (dot) com)
1351    Date: 2004/08/09 12:21:00 EST
1352    Tag: leunga-110_47-dijsktra
1353    Description:
1354    
1355    Bug fix from Carl Hauser:
1356    
1357    single_source_shortest_paths in dijkstra.sml was observed to get wrong
1358    answers (by comparing to single_source_shortest_paths in bellman-ford.sml).
1359    
1360    The problem is that following the expression A.update(dist,s,Num.zero)
1361    it is necessary to update the priority queue using Q.decreaseWeight(Q,s).
1362    
1363    ----------------------------------------------------------------------
1364    Name: Matthias Blume (blume (at) tti - c (dot) org)
1365    Date: 2004/08/06 18:10:00 CDT
1366    Tag: blume-20040806-cmdline
1367    Description:
1368    
1369    Fiddled with handling of command-line options:
1370    
1371      * sml now quits after processing the command line
1372        if -H, -S, -h<n>, or -s<n> appears as the last
1373        command-line argument
1374      * a new option -q terminates the session when encountered on
1375        the command line; subsequent arguments will be ignored
1376      * bug fixes: short (erroneous) arguments are no longer ignored
1377        completely
1378    
1379    ----------------------------------------------------------------------
1380    Name: Allen Leung (leunga (at) reservoir (dot) com)
1381    Date: 2004/08/04 18:17:00 EST
1382    Tag: leunga-110_47-ppc-ibm-asm
1383    Description:
1384    
1385      - Added minimal IBM assembly syntax support for PowerPC.
1386    
1387      - Cygwin: manually changed the file cygwin.def.  Some exported symbols have
1388        been altered in the runtime.  We need an automatic way to keep the file
1389        in sync.
1390    
1391    ----------------------------------------------------------------------
1392    Name: Matthias Blume (blume (at) tti - c (dot) org)
1393    Date: 2004/08/04 14:00:00 CDT
1394    Tag: Release_110_47
1395    Description:
1396    
1397    New working version (110.47).  NEW BOOTFILES!
1398    
1399    ----------------------------------------------------------------------
1400    Name: Matthias Blume (blume (at) tti - c (dot) org)
1401    Date: 2004/08/03 14:25:00 CDT
1402    Tag: blume-20040803-callingconv
1403    Description:
1404    
1405    Added low-level support for choosing C calling conventions by
1406    twiddling the type of rawccall.  (See
1407    src/compiler/Semant/types/cproto.sml for details.)
1408    
1409    ----------------------------------------------------------------------
1410    Name: Matthias Blume (blume (at) tti - c (dot) org)
1411    Date: 2004/08/02 15:55:00 CDT
1412    Tag: blume-20040802-backout
1413    Description:
1414    
1415    Backed out of change to win32-filesys.c.  The earlier patch to
1416    get_file_time caused CM to produce files with the wrong time stamp.
1417    
1418    ----------------------------------------------------------------------
1419    Name: Matthias Blume (blume (at) tti - c (dot) org)
1420    Date: 2004/08/02 14:45:00 CDT
1421    Tag: blume-20040802-nlffi-win32
1422    Description:
1423    
1424    Added NLFFI support for Win32, adapted from a patch provided by David
1425    Hansel.  This is currently completely untested.  Also, the issue
1426    concerning stdcall vs. ccall is still unresolved.
1427    
1428    ----------------------------------------------------------------------
1429    Name: Matthias Blume (blume (at) tti - c (dot) org)
1430    Date: 2004/07/30 17:55:00 CDT
1431    Tag: blume-20040730-various
1432    Description:
1433    
1434    Gearing up towards 110.47...
1435    
1436    - various minor bugfixes to ml-nlffigen
1437    - a beginning of a manual for nlffi
1438    
1439    - eliminated 'export name=value' in config/install.sh as this does
1440      not work with certain versions of /bin/sh
1441      (Thanks to David King at Motorola for catching this.)
1442    
1443    - several bugfixes provided or suggested by David Hansel at Reactive Systems:
1444      - added a test for tm==NULL to gmtime.c and localtime.c
1445      - applied patch for incorrect GetFileTime under win32
1446      - toSeconds -> toMilliseconds in Win32/win32-process.sml
1447    
1448    ----------------------------------------------------------------------
1449    Name: Matthias Blume (blume (at) tti - c (dot) org)
1450    Date: 2004/07/21 18:20:00 CDT
1451    Tag: blume-20040721-nlffigen
1452    Description:
1453    
1454    - Fixed minor issue in ml-nlffigen: Now generate structure T_foo
1455      for a typedef to an incomplete type, but leave out the "typ" member.
1456      (This is just for consistency.)
1457    - Started to produce what is supposed to become better (i.e., comprehensive)
1458      documentation of what ml-nlffigen does and produces.
1459    
1460    ----------------------------------------------------------------------
1461    Name: Matthias Blume (blume (at) tti - c (dot) org)
1462    Date: 2004/07/14 16:25:00 CDT
1463    Tag: blume-20040714-union
1464    Description:
1465    
1466    Added C_UNION to c-calls/c-types.sml and updated the machinery
1467    (ml-nlffigen, cproto.sml) that conveys C function interface
1468    information to the code generator.
1469    
1470    However, the actual architecture-specific implementation of function
1471    arguments and results that are C unions is still not implemented.
1472    
1473    ----------------------------------------------------------------------
1474    Name: Allen Leung (leunga (at) reservoir (dot) com)
1475    Date: 2004/07/14 14:38:00 EST
1476    Tag: leunga-110_46_1-ppc-lwzu
1477    Description:
1478    
1479        Added these instructions to the PowerPC architecture:
1480        LBZU(X), LHZU(X), LWZU(X),
1481        STWU(X), STFDU, STFSU
1482    
1483        etc...
1484    
1485        Note: I haven't added their instruction encoding into the description.
1486    
1487    ----------------------------------------------------------------------
1488    Name: Allen Leung (leunga (at) reservoir (dot) com)
1489    Date: 2004/07/13 15:04:00 EST
1490    Tag: leunga-110_46_1-ppc-lwarx
1491    Description:
1492    
1493       Added the two instructions LWARX and STWCX to the PowerPC
1494    instruction set.
1495    
1496       A (untested) rewrite of loop-structure.sml.  The old version
1497    is completely broken.
1498    
1499    ----------------------------------------------------------------------
1500    Name: Matthias Blume (blume (at) tti - c (dot) org)
1501    Date: 2004/07/13 13:50:00 CDT
1502    Tag: blume-20040713-nlffi
1503    Description:
1504    
1505    - use paramAlloc to report c-calls with too many arguments
1506      (for PPC version where parameter area is pre-allocated)
1507    - added ccall_maxargspace to machspec (to implement the above)
1508    - made "make" commend in CM's "make" tool configurable
1509    - added option (default: on) for passing the name of the SML/NJ's "bin"
1510      directory to "make"; the call looks like this:
1511    
1512         make <options> SMLNJ_BINDIR=<dir> <target>
1513    
1514      This can be used by the Makefile to, e.g., pick the "right" version
1515      of ml-nlffigen.
1516    - minor code tweaks
1517    
1518    ----------------------------------------------------------------------
1519    Name: Matthias Blume (blume (at) tti - c (dot) org)
1520    Date: 2004/07/12 22:50:00 CDT
1521    Tag: blume-110_46_1-macosx-nlffi
1522    Description:
1523    
1524    NLFFI under Mac OS X now working (sort of).  This is largely untested,
1525    though.
1526    
1527    Note:  1. You have to make a new, clean build of the runtime system.
1528           2. There are new BOOTFILES, you have to use them!
1529              (Doing the bootstrap process yourself would be *very* painful!
1530               If you absolutely have to do it, build the system under
1531               a different architecture and then cross-compile.)
1532    
1533    Version bumped to 110.46.1 to account for runtime data format changes.
1534    
1535    ----------------------------------------------------------------------
1536    Name: Matthias Blume (blume (at) tti - c (dot) org)
1537    Date: 2004/06/18 14:30:00 CDT
1538    Tag: blume-20040618-unix
1539    Description:
1540    
1541    Changed the implementation of structure Unix so that the same stream
1542    is returned every time one of the {text,bin}{In,Out}streamOf functions
1543    is invoked on the same proc.  This is not what the spec currently
1544    says -- although IMO it arguably should.  (See discussion below.)
1545    
1546    ----------------------------------------------------------------------
1547    Name: Matthias Blume (blume (at) tti - c (dot) org)
1548    Date: 2004/06/17 18:15:00 CDT
1549    Tag: Release_110_46
1550    Description:
1551    
1552    New working version (110.46).  NEW BOOTFILES!
1553    
1554    ----------------------------------------------------------------------
1555    Name: Matthias Blume (blume (at) tti - c (dot) org)
1556    Date: 2004/06/17 17:20:00 CDT
1557    Tag: blume-20040617-timer-unix
1558    Description:
1559    
1560    Changed the interface of structures Timer and Unix to match the most
1561    recent Basis spec.
1562    
1563    In the case of Unix there still seems to be an open/weird issue:
1564    
1565       The {text,bin}{In,Out}streamOf functions are supposed to create
1566       fresh streams whenever they are called -- as opposed to have them
1567       return the same stream every time.  This design is supposed to
1568       prevent space leaks caused by proc values hanging on to streams.
1569    
1570       The reap function, on the other hand, is supposed to close the
1571       streams.  This cannot be done without having a handle on the
1572       stream in proc after all...
1573    
1574    I took the liberty to implement the following stopgap solution:
1575    
1576      The proc value hangs on to the most recently created stream(s).
1577      Reap closes those.  If either or both of the two streams hadn't
1578      been created at all yet, then reap will close the corresponding
1579      file descriptors directly.
1580    
1581    PS: I don't understand the original space leak argument anymore.  If
1582    a proc hangs on to the imperative stream, then I/O operations on those
1583    will advance the state of the cached stream and avoid the space leak.
1584    
1585    ----------------------------------------------------------------------
1586    Name: Matthias Blume (blume (at) tti - c (dot) org)
1587    Date: 2004/05/28 16:45:00 CDT
1588    Tag: blume-20040528-basis
1589    Description:
1590    
1591    Added signature PACK_REAL and exported functor PrimIO.
1592    
1593    ----------------------------------------------------------------------
1594    Name: Matthias Blume (blume (at) tti - c (dot) org)
1595    Date: 2004/05/25 16:00:00 CDT
1596    Tag: blume-20040525-group-owner
1597    Description:
1598    
1599    CM now ignores (but still accepts) the "owner" information in group
1600    descriptions.  The owner of a group is its next enclosing
1601    library. Each group must have a unique owner.  (There is a virtual
1602    "toplevel" library that owns groups which are not nested within a real
1603    library.)  Previously, each group had to explicitly declare its owner,
1604    and CM would check that such a declaration is correct.  The new scheme
1605    is to have CM check that for each group there is precisely one owning
1606    library.
1607    
1608    The advantage of the new scheme is that the programmer no longer needs
1609    to maintain the somewhat annoying owner information.  The downside is
1610    that CM cannot enforce the ownership rule across multiple runs of
1611    CM.make.  Fortunately, enclosing the same group in two different
1612    libraries A and B which are not part of the same program does not
1613    cause real problems.
1614    
1615    ----------------------------------------------------------------------
1616    Name: Matthias Blume (blume (at) tti - c (dot) org)
1617    Date: 2004/05/20 16:00:00 CDT
1618    Tag: blume-20040520-win32
1619    Description:
1620    
1621    Made the win32 version work again.  (Strangely, a misplaced comma had
1622    slipped into win32-process.c which prevented the runtime from being
1623    compiled correctly.)
1624    
1625    Also, included a minor addition to ml-build.bat analogous to what was
1626    done in blume-20040519-ml-build.
1627    
1628    ----------------------------------------------------------------------
1629    Name: Matthias Blume (blume (at) tti - c (dot) org)
1630    Date: 2004/05/19 22:10:00 CDT
1631    Tag: blume-20040519-ml-build
1632    Description:
1633    
1634    Arranged for ml-build to clean up after itself a little bit better.
1635    The script generates a temporary SML source file and compiles it using
1636    CM, so CM generates metadata (GUID, SKEL, objectfile) for it.  It now
1637    gets rid of those at the end, so they don't accumulate under .cm.
1638    
1639    This required a minor change to install.sh because the name of the
1640    metadata directory (default: .cm) is actually configurable at
1641    installation time.
1642    
1643    ----------------------------------------------------------------------
1644    Name: Matthias Blume (blume (at) tti - c (dot) org)
1645    Date: 2004/05/18 15:50:00 CDT
1646    Tag: blume-20040518-mkreader
1647    Description:
1648    
1649    Added Posix.IO.mk{Bin,Text}{Reader,Writer} by lifting their respective
1650    implementations from internal modules PosixBinPrimIO and PosixTextPrimIO.
1651    
1652    ----------------------------------------------------------------------
1653    Name: Matthias Blume (blume (at) tti - c (dot) org)
1654    Date: 2004/05/11 14:35:00 CDT
1655    Tag: blume-20040511-win32sock
1656    Description:
1657    
1658    Added previously missing support for many socket-related functions
1659    under win32.  Thanks to David Hansel <hansel@reactive-systems.com>
1660    for the voluminous patch!
1661    
1662    (I have not tested this patch under win32 yet.)
1663    
1664    Here is David's e-mail:
1665    
1666    Hi,
1667    
1668    Attached to this email you find a diff against sml/nj 110.45
1669    that will enable socket support under Windows.
1670    
1671    To apply the patch (using unix or cygwin)
1672    1) gunzip runtime.diff.gz
1673    2) "cd" into "src/runtime" in the source tree of a fresh
1674       110.45 installation.
1675    3) patch -p 1 < [your/path/to]runtime.diff
1676    
1677    The code compiles fine but has NOT yet been extensively tested.
1678    I only ran a few tests for basic socket client functionality
1679    (which worked fine).  Especially the functions that use ioctl
1680    are not tested at all and might not work (see below).
1681    
1682    I implemented this since we want to move to a newer version of sml/nj
1683    but need socket support in order to use it.  This is the first time I
1684    even had a look at the sml/nj source,  so please review my changes
1685    before making this part of the distribution!  Here are a few issues
1686    that I think might be better for someone to solve who is more
1687    familiar with the sml/nj source (and socket programming):
1688    
1689    - getnetbyaddr.c and getnetbyname.c will raise a "not implemented"
1690      exception since I could not figure out what the windows equivalent
1691      of these functions is
1692    
1693    - In sockets-osdep.h there are a some #include statements that are
1694      only used in a few files that include sockets-osdep.h
1695    
1696    - In smlnj-sock-lib.c, function init_fn() calls WSAStartup() but
1697      does not process its return value since I don't know how to
1698      report an error upwards.
1699    
1700    - It would probably be good to have a call to WSACleanup() when
1701      the library is unloaded (if there is such a possibility).
1702      Otherwise I think Windows will take care of this automatically
1703      when the process finishes.
1704    
1705    - I used ioctlsocket() as a replacement for ioctl() but I have
1706      no idea if that is actually the proper replacement on Windows.
1707    
1708    - All these issues are marked in the code by "FIXME" comments.
1709    
1710    We use sml/nj extensively in our products and are quite happy
1711    with it.  I hope this contribution will help you.
1712    
1713    Keep up the good work!
1714    
1715    David
1716    
1717    ----------------------------------------------------------------------
1718    Name: Matthias Blume (blume (at) tti - c (dot) org)
1719    Date: 2004/05/11 14:20:00 CDT
1720    Tag: blume-20040511-installml
1721    Description:
1722    
1723    Fixed two bugs in installml script.  (Thanks to Vesa A. Norrman for
1724    the patch.)
1725    
1726    ----------------------------------------------------------------------
1727    Name: Matthias Blume (blume (at) tti - c (dot) org)
1728    Date: 2004/05/11 14:05:00 CDT
1729    Tag: blume-20040511-nlffi-netbsd
1730    Description:
1731    
1732    Added support for nlffi under netbsd.  (Thanks to Vesa A. Norrman for
1733    the patch.)
1734    
1735    ----------------------------------------------------------------------
1736    Name: Matthias Blume (blume (at) tti - c (dot) org)
1737    Date: 2004/05/11 12:05:00 CDT
1738    Tag: blume-20040511-exports
1739    Description:
1740    
1741    As per request by Adam Chlipala <adam@hcoop.net>, extended various
1742    export lists in compiler-related .cm-files.
1743    
1744    ----------------------------------------------------------------------
1745    Name: Matthias Blume (blume (at) tti - c (dot) org)
1746    Date: 2004/05/11 11:35:00 CDT
1747    Tag: blume-20040511-allsource
1748    Description:
1749    
1750    The installer now honors the "src-smlnj" target again, although its meaning
1751    has changed from "all sources required for the compiler" to "all sources
1752    the installer knows about".  In other words, if you enable "src-smlnj"
1753    in the "targets" file, then the installer will pull in sources for
1754    everything.  (Notice that this refers to source code only.  Compiled
1755    code is still only installed for modules that were requested explicitly
1756    or which are required for other modules that were requested explicitly.)
1757    
1758    ----------------------------------------------------------------------
1759    Name: Matthias Blume (blume (at) tti - c (dot) org)
1760    Date: 2004/04/23 17:40:00 CDT
1761    Tag: blume-20040423-ieee-scan
1762    Description:
1763    
1764    Fixed IEEEReal.scan (and .fromString) so that if there is an overflow
1765    in the exponent calculation we get INF or ZERO (depending on the mantissa
1766    and the sign of the exponent).
1767    
1768    ----------------------------------------------------------------------
1769    Name: Matthias Blume (blume (at) tti - c (dot) org)
1770    Date: 2004/04/23 10:40:00 CDT
1771    Tag: blume-20040423-ml-build
1772    Description:
1773    
1774    The ml-build script now terminates with a non-0 status when something
1775    goes wrong.
1776    
1777    ----------------------------------------------------------------------
1778    Name: Matthias Blume (blume (at) tti - c (dot) org)
1779    Date: 2004/04/22 16:35:00 CDT
1780    Tag: blume-20040422-Option
1781    Description:
1782    
1783    Made exception Option to be the same as exception Option.Option
1784    (as it should be).
1785    
1786    ----------------------------------------------------------------------
1787    Name: Allen Leung (leunga (at) reservoir (dot) com)
1788    Date: 2004/03/19 14:40:00 EST
1789    Tag: leunga-20040319-cygwin-nlffi
1790    Description:
1791    
1792    Fixed the runtime so that ml-nlffi-lib runs on the cygwin version
1793    of SML/NJ.  The problem is that
1794    
1795       lib = dlopen(NULL, ...)
1796       f   = dlsym(lib, "malloc");
1797    
1798    does not work on Windows unless we explicitly export symbols
1799    such as 'malloc' during linking.  We fixed this by explicitly
1800    exporting the required symbols with the magic gcc incantation:
1801    
1802        -Wl,--export-all cygwin.def
1803    
1804    where cygwin.def is a file containing all the symbols that
1805    we wish to export.
1806    
1807    I suspect this is a Windows problem and we'll have to
1808    do the same (somehow with windows compilers) when
1809    we build the native win32 version with the system
1810    calls LoadLibrary/GetProcAddress.
1811    
1812    ----------------------------------------------------------------------
1813    Name: Matthias Blume (blume (at) tti - c (dot) org)
1814    Date: 2004/03/04 16:35:00 CST
1815    Tag: blume-20040304-intinf-fmt
1816    Description:
1817    
1818    Fixed problem with IntInf.fmt (sign would show up on the right instead
1819    of on the left for BIN, OCT, and HEX).
1820    
1821    ----------------------------------------------------------------------
1822    Name: Matthias Blume (blume (at) tti - c (dot) org)
1823    Date: 2004/03/04 11:25:00 CST
1824    Tag: blume-20040304-symlinks
1825    Description:
1826    
1827    Fixed problem with installer script (unix only) where bin/ml-yacc and
1828    friends pointed (via symlinks) to absolute locations instead of just
1829    .run-sml.  This was reported by Vesa A Norrman.
1830    
1831    ----------------------------------------------------------------------
1832    Name: Matthias Blume (blume (at) tti - c (dot) org)
1833    Date: 2004/02/13 14:50:00 CST
1834    Tag: Release_110_45
1835    Description:
1836    
1837    New working version (110.45). New bootfiles.
1838    
1839    ----------------------------------------------------------------------
1840    Name: Matthias Blume (blume (at) tti - c (dot) org)
1841    Date: 2004/01/26 15:15:15 CST
1842    Tag: blume-20040126-toplevel
1843    Description:
1844    
1845    Improved handling of exceptions at the interactive toplevel.
1846    
1847    ----------------------------------------------------------------------
1848    Name: Matthias Blume (blume (at) tti - c (dot) org)
1849    Date: 2004/01/26 11:25:00
1850    Tag: blume-20040126-app
1851    Description:
1852    
1853    Type of top-level "app" corrected.
1854    Added code for setting vp_limitPtrMask to Win32-specific runtime.
1855    
1856    ----------------------------------------------------------------------
1857    Name: Matthias Blume (blume (at) tti - c (dot) org)
1858    Date: 2003/11/18 17:10 CST
1859    Tag: blume-20031118-basis-fiddle
1860    Description:
1861    
1862    - changed Timer interface to what might become the spec
1863    - POSIX_FLAGS -> BIT_FLAGS according to spec
1864    - some other minor discrepancies wrt. spec eliminated
1865    
1866    ----------------------------------------------------------------------
1867    Name: Matthias Blume (blume (at) tti - c (dot) org)
1868    Date: 2003/11/06 12:00:00 CST
1869    Tag: Release_110_44
1870    Description:
1871    
1872    New working version (110.44). New bootfiles.
1873    
1874    ----------------------------------------------------------------------
1875    Name: Matthias Blume (blume (at) tti - c (dot) org)
1876    Date: 2003/11/04 11:50:00 CST
1877    Tag: blume-20031104-move-libraries
1878    Description:
1879    
1880    Eliminated the "dont_move_libraries" directive in config/targets.
1881    (The mechanism was broken and could not be fixed easily.  Moreover,
1882    there does not seem to be any reason not to move all libraries into
1883    lib during installation.  I originally implemented this directive as a
1884    backward-compatibility feature when I first introduced the new CM.
1885    Now that things have been stable for a long time and going back to the
1886    old CM is not an option, there is no reason to keep it around.)
1887    
1888    ----------------------------------------------------------------------
1889    Name: Matthias Blume (blume (at) tti - c (dot) org)
1890    Date: 2003/11/03 16:00:00 CST
1891    Tag: blume-20031103-installdir
1892    Description:
1893    
1894    Made installer honor INSTALLDIR variable again.  (Thanks to Chris
1895    Richards for pointing out the problem and providing the solution.)
1896    
1897    ----------------------------------------------------------------------
1898    Name: Matthias Blume (blume (at) tti - c (dot) org)
1899    Date: 2003/10/01 17:05:00 CDT
1900    Tag: blume-20031001-lal-mlrisc
1901    Description:
1902    
1903    MLRISC bug fix from Lal.
1904    
1905    ----------------------------------------------------------------------
1906    Name: Matthias Blume (blume (at) tti - c (dot) org)
1907    Date: 2003/09/30 16:10:00 CDT
1908    Tag: blume-20030930-primio-bat
1909    Description:
1910    
1911    1. Added openVector, nullRd, and nullWr to PRIM_IO.
1912    2. Improved .bat files (for Win32 port) to make things work under Win95.
1913       (thanks to Aaron S. Hawley for this one)
1914    
1915    ----------------------------------------------------------------------
1916    Name: Matthias Blume (blume (at) tti - c (dot) org)
1917    Date: 2003/09/26 16:05:00 CDT
1918    Tag: blume-20030926-wrappriv
1919    Description:
1920    
1921    Added missing wrapper for privilege "primitive" in $smlnj/viscomp/core.cm.
1922    
1923    ----------------------------------------------------------------------
1924    Name: Matthias Blume (blume (at) tti - c (dot) org)
1925    Date: 2003/09/26 15:00:00 CDT
1926    Tag: blume-20030926-110_43_3
1927    Description:
1928    
1929    - additional cleanup
1930    - version number bump, NEW BOOTFILES
1931    
1932    ----------------------------------------------------------------------
1933    Name: Matthias Blume (blume (at) tti - c (dot) org)
1934    Date: 2003/09/26 12:00:00 CDT
1935    Tag: blume-20030926-ppautoload
1936    Description:
1937    
1938    I modified the read-eval-print loop so that the autoloader gets
1939    invoked whenever the prettyprinter tries to look up a symbol that
1940    is not currently defined in the toplevel environment but which
1941    appears in CM's autoload registry.  As a result, we see far fewer of
1942    those ?.Foo.Bar.xxx names in the prettyprinter's output.
1943    
1944    In addition to this I tried to clean up some pieces of the Basis
1945    implementation (e.g., Socket, Word8Array) in order to prevent other
1946    instances of these ?.Foo.Bar.xxx names from being printed.
1947    
1948    The mechanism that picks names for types still needs some work, though.
1949    (Right now it seems that if there is a type A.t which is defined to
1950    be B.u, but B is unavailable at toplevel, then A.t gets printed as
1951    "?.B.u" although the perhaps more sensible solution would be to use
1952    "A.t" in this case.  In other words, the prettyprinter should follow
1953    a chain of DEFtycs not farther than there are corresponding toplevel
1954    names in the current environment.)
1955    
1956    ----------------------------------------------------------------------
1957    Name: Matthias Blume (blume (at) tti - c (dot) org)
1958    Date: 2003/09/24 16:31:00 CDT
1959    Tag: blume-20030924-installer
1960    Description:
1961    
1962    Another installer tweak:  All the ML code for the installer is now
1963    compiled during CMB.make and put into a little library called
1964    $smlnj/installer.cm.  The installation then simply invokes
1965    
1966       sml -m $smlnj/installer.cm
1967    
1968    and everything happens automagically.
1969    
1970    Win32:  ML code senses value of environment variable SMLNJ_HOME.
1971    Unix:   ML code senses values of environment variables ROOT, CONFIGDIR,
1972            and BINDIR.
1973    
1974    The new scheme guarantees that the ML code responsible for the installation
1975    is in sync with the APIs of the main system.  Also, the installer is
1976    somewhat faster because the installer script is precompiled.
1977    
1978    ----------------------------------------------------------------------
1979    Name: Matthias Blume (blume (at) tti - c (dot) org)
1980    Date: 2003/09/24 15:35:00 CDT
1981    Tag: blume-20030924-synsock
1982    Description:
1983    
1984    Added a signature SYNCHRONOUS_SOCKET to basis.cm.  This is like SOCKET
1985    but excludes all non-blocking operations.  Defined SOCKET (in Basis)
1986    and CML_SOCKET in terms of SYNCHRONOUS_SOCKET.  Removed superfluous
1987    implementations of non-blocking operations from CML's Socket
1988    structure.
1989    
1990    ----------------------------------------------------------------------
1991    Name: Matthias Blume (blume (at) tti - c (dot) org)
1992    Date: 2003/09/24 15:10:05 CDT
1993    Tag: blume-20030924-sockets
1994    Description:
1995    
1996    1. Fixed SOCKET API and implementation to match Basis spec.
1997       This required changing the internal representation of sockets to one
1998       that remembers (for each socket file descriptor) whether it is currently
1999       blocking or non-blocking.  This state is maintained lazily (i.e., a system
2000       call is made only if the state actually needs to change).
2001    
2002    2. OS-specific details of sockets were moved into separate files, thus
2003       making it possible to unify the bulk of the socket implementations
2004       between Unix and Win32.
2005    
2006    3. CML's socket API changed accordingly.
2007       (Note that we need to remove non-blocking functions from this API
2008       since they are redundant in the case of CML!)
2009    
2010    4. CML's socket implementation now makes use of non-blocking functions
2011       provided by Basis, thus removing all OS-dependent code from this part
2012       of CML.
2013    
2014    5. Changed Real64.precision from 52 to 53.  Minor cleanup in Real64 code.
2015    
2016    ----------------------------------------------------------------------
2017    Name: Matthias Blume (blume (at) tti - c (dot) org)
2018    Date: 2003/09/22 12:10:00 CDT
2019    Tag: blume-20030922-110_43_2
2020    Description:
2021    
2022    Made a new interim version and bootfiles for developer's bootstrapping
2023    convenience.
2024    
2025       110.43.2 -- NEW BOOTFILES
2026    
2027    ----------------------------------------------------------------------
2028    Name: Matthias Blume (blume (at) tti - c (dot) org)
2029    Date: 2003/09/19 15:55:00 CDT
2030    Tag: blume-20030919-cmdir
2031    Description:
2032    
2033    1. new-install.sh -> install.sh
2034    2. changed default CM "metadata" directory name to ".cm" (instead of "CM")
2035    3. tweaked installer so that another name instead of .cm can be chosen
2036       at install time (by setting the CM_DIR_ARC environment variable
2037       during installation); once installation is complete, the name is
2038       fixed
2039    
2040    ----------------------------------------------------------------------
2041    Name: Matthias Blume (blume (at) tti - c (dot) org)
2042    Date: 2003/09/18 16:00:00 CDT
2043    Tag: blume-20030918-110_43_1
2044    Description:
2045    
2046    Made a new interim version and bootfiles for developer's bootstrapping
2047    convenience.
2048    
2049       110.43.1 -- NEW BOOTFILES
2050    
2051    ----------------------------------------------------------------------
2052    Name: Matthias Blume (blume (at) tti - c (dot) org)
2053    Date: 2003/09/18 15:20:00 CDT
2054    Tag: blume-20030918-misc
2055    Description:
2056    
2057    1. Exported fractionsPerSecond etc. from TimeImp (but not from Time as
2058       this seems to be controversial at the moment) and used those in
2059       Posix.ProcEnv.times.
2060    
2061    2. Added Time.{from,to}Nanoseconds to Time.
2062    
2063    3. Improved Real.{from,to}LargeInt by avoiding needless calculations.
2064       For example, fromLargeInt never needs to look at more than 3 "big
2065       digits" to get its 53 bits of precision.
2066    
2067    ----------------------------------------------------------------------
2068    Name: Matthias Blume (blume (at) tti - c (dot) org)
2069    Date: 2003/09/17 16:30:00 CDT
2070    Tag: blume-20030917-real32-slices
2071    Description:
2072    
2073    Added an entry to the primitive environment
2074    (compiler/Semant/statenv/prim.sml) for int32->real64 conversion and
2075    added code to compiler/CodeGen/main/mlriscGen.sml to implement it.
2076    
2077    Removed some of the "magic" constants in real64.sml and replaced them
2078    with code that generates these values from their corresponding
2079    integer counterparts.
2080    
2081    Made all(?) the slice-related changes to the Basis and made everything
2082    compile again...
2083    
2084    ----------------------------------------------------------------------
2085    Name: Matthias Blume (blume (at) tti - c (dot) org)
2086    Date: 2003/09/15 17:45:00 CDT
2087    Tag: blume-20030915-rbase
2088    Description:
2089    
2090    Fixed bug in Real.fromLargeInt.
2091    
2092    ----------------------------------------------------------------------
2093    Name: Matthias Blume (blume (at) tti - c (dot) org)
2094    Date: 2003/09/13 18:11:00 CDT
2095    Tag: blume-20030913-libinstall
2096    Description:
2097    
2098    Minor bugfix in config/libinstall (set anchor with path to
2099    standalone tool after installing it, otherwise libraries that
2100    need ml-lex or ml-yacc won't compile the first time the installer
2101    runs).
2102    
2103    ----------------------------------------------------------------------
2104    Name: Matthias Blume (blume (at) tti - c (dot) org)
2105    Date: 2003/09/12 11:45:00 CDT
2106    Tag: blume-20030912-various
2107    Description:
2108    
2109     - fixed bug in Real.toLargeInt
2110     - fixed bug in Posix.ProcEnv.times
2111     - changed inputLine functions to return an option
2112     - minor installer improvements / bugfixes
2113     - changed default @SMLalloc parameter for x86/celeron to 64k
2114    
2115    ----------------------------------------------------------------------
2116    Name: Matthias Blume (blume (at) tti - c (dot) org)
2117    Date: 2003/09/09 22:00:00 CDT
2118    Tag: Release_110_43
2119    Description:
2120    
2121        New working release 110.43.  New bootfiles.
2122    
2123    ----------------------------------------------------------------------
2124    Name: Matthias Blume (blume (at) tti - c (dot) org)
2125    Date: 2003/09/09 19:20:00 CDT
2126    Tag: blume-20030909-installer
2127    Description:
2128    
2129    Rewrote large parts of config/install.sh in SML (config/libinstall.sml).
2130    Modified config/install.bat to take advantage of it.  Also modified
2131    config/install.sh (and called it config/new-install.sh) to take advantage
2132    of it on Unix systems.  (The SML code is (supposed to be) platform-
2133    independent.)
2134    
2135    The installer can now install everything under Win32
2136    as well as under *nix as long as it compiles.
2137    
2138    Other changes:
2139    
2140      - made CML compile again under Win32
2141      - made eXene compile under Win32 (by providing a fake structure UnixSock
2142        and by using OS.Process.getEnv instead of Posix.ProcEnv.getenv)
2143      - fixed a bug in nowhere: it assumed that type OS.Process.status is the
2144        same as type int;  under Win32 it isn't
2145      - fixed some slice-related problems in the win32-specific parts of CML
2146      - added a functor argument "sameVol" to os-path-fn.sml in the Basis
2147        (under Win32, the volume name is case-insensitive, and the
2148        OS.Path code compares volume names for equality)
2149    
2150    ----------------------------------------------------------------------
2151    Name: Matthias Blume (blume (at) tti - c (dot) org)
2152    Date: 2003/09/08 11:55:00 CDT
2153    Tag: blume-20030908-fullpath
2154    Description:
2155    
2156    Made Win32 version of OS.FileSys.fullPath return current directory
2157    when given an empty string.  This is what the spec says, and incidentally,
2158    CM depends on it.  (CM otherwise goes into an infinite loop in certain
2159    cases when presented with the name of a non-existing .cm file.)
2160    
2161    ----------------------------------------------------------------------
2162    Name: Matthias Blume (blume (at) tti - c (dot) org)
2163    Date: 2003/09/04 16:30:00 CDT
2164    Tag: blume-20030905-slices-etc
2165    Description:
2166    
2167      1. Changed interface to vectors and arrays in Basis to match
2168         (draft) Basis spec.
2169      2. Added signatures and implementations of slices according to
2170         Basis spec.
2171      3. Edited source code throughout the system to make it compile again
2172         under 1. and 2.  (In some cases code had to be added to have it
2173         match the new signatures.)
2174      4. MLRISC should be backward-compatible: the copies of the originals
2175         of files that needed to change under 3. were retained, the .cm files
2176         check the compiler version number and use old versions when
2177         appropriate.
2178      5. Changed type of OS.FileSys.readDir and Posix.FileSys.readdir to
2179         dirstream -> string option (in accordance with Basis spec).
2180      6. When generating code that counts lines, ml-lex used function
2181         CharVector.foldli, taking advantage of its old interface.
2182         This has been replaced with the corresponding code from
2183         CharVectorSlice.  (html-lex must be re-lexed!)
2184      7. BitArray in smlnj-lib/Util has been extended/modified to match the
2185         new MONO_ARRAY signature.  (Do we need BitArraySlice?)
2186      8. Removed temporary additions (fromInternal, toInternal) from the
2187         (now obsolete) IntInf in smlnj-lib/Util.
2188      9. Cleaned up structure Byte.
2189     10. Added localOffset, scan, and fromString to Date (according to spec).
2190         Cleaned/corrected implementation of Date.
2191         (Still need to check for correctness; implement better canonicalizeDate.)
2192     11. Added "scan" to signature IEEE_REAL.
2193     12. Some improvements to IntInf [in particular: efficiency-hack for
2194         mod and rem when second operand is 2 (for parity checks).]
2195     13. Changed representation of type Time.time, using a single IntInf.int
2196         value counting microseconds.  This considerably simplified the
2197         implementation of structure Time.  We now support negative time
2198         values; scan and fromString handle signs.
2199     14. Functor PrimIO now takes two additional arguments (VectorSlice and
2200         ArraySlice).
2201    
2202    ----------------------------------------------------------------------
2203    Name: Matthias Blume (blume (at) tti - c (dot) org)
2204    Date: 2003/08/28 17:00:00 CDT
2205    Tag: blume-20030828-intinf
2206    Description:
2207    
2208    This is a major update which comes with a version number bump
2209    (110.42.99 -- yes, we are really close to 110.43 :-), NEW BOOTFILES,
2210    and an implementation of IntInf in the Basis.
2211    
2212    There are a fairly large number of related changes and updates throughout
2213    the system:
2214    
2215    Basis:
2216            - Implemented IntInf.
2217            - Made LargeInt a projection of IntInf (by filtering through INTEGER).
2218            - Added some missing Real64 operations, most notably Real.toLargeInt.
2219            - Added FixedInt as a synonym for Int32.
2220    
2221    compiler:
2222            * Added support for a built-in intinf type.
2223              - literals
2224              - pattern matching
2225              - conversion shortcuts (Int32.fromLarge o Int.toLarge etc.)
2226              - overloading on literals and operations
2227    
2228            This required adding a primitive type intinf, some additional
2229            primops, and implementations for several non-trivial intinf
2230            operations in Core.  (The intinf type is completely abstract
2231            to the compiler; all operations get delegated back to the Core.)
2232    
2233            * Intinf equality is handled by polyequal.  However, the compiler
2234            does not print its usual warning in this case (since polyequal
2235            is the right thing to do there).
2236    
2237            * Improved the organization of structure InlineT.
2238    
2239            * A word about conversion primops:
2240            If conversions involving intinf do not cancel out during
2241            CPS contract, then the compiler must insert calls to Core functions.
2242            Since all core access must be resolved already during the FLINT
2243            translate phase, it would be too late a the time of CPS contract
2244            to add new Core calls.  For this reason, conversion primops
2245            for intinf carry two arguments:  1. the numeric argument that
2246            they are supposed to convert, and 2. the Core function that
2247            can help with this conversion if necessary.  If CPS contract
2248            eliminates a primop, then the associated Core function becomes
2249            dead and goes away.  Intinf conversion primops that do not get
2250            eliminated by CPS contract get rewritten into calls of their
2251            core functions by a separate, new phase.
2252    
2253    interactive system:
2254            - Control.Print.intinfDepth controls max length of intinf constants
2255            being printed. (Analogous to Control.Print.stringDepth.)
2256            - Cleanup in printutil and pputil: got rid of unused stuff and
2257            duplicates; replaced some of the code with code that makes better
2258            use of library functionality.
2259    
2260    CM:
2261            Bugfix: parse-errors in init group (system/smlnj/init/init.cmi)
2262            are no longer silent.
2263    
2264    CKIT:
2265            Fixed mismatched uses of Int32 and LargeInt.  I always decided
2266            in favor of LargeInt -- which is now the same as IntInf.
2267            CKIT-knowledgable people should check whether this is what's
2268            intended and otherwise change things back to using Int32 or
2269            FixedInt.
2270    
2271    Throughout the code:
2272            Started using IntInf.int literals and built-in operations
2273            (e.g., comparison with 0) where this seems appropriate.
2274    
2275    
2276    ----------------------------------------------------------------------
2277    Name: Dave MacQueen (dbm@cs.uchicago.edu)
2278    Date: 2003/08/13 11:36:00 CDT
2279    Tag: dbm-20030813-mcz-merge1
2280    Description:
2281    
2282    Merging changes from the mcz-branch development branch into trunk.
2283    These changes involve replacement of the emulated old prettyprinter
2284    interface with direct use of the SML/NJ Lib PP library, and fixing
2285    of a couple of bugs (895, 1186) relating to error messages.  A new
2286    prettyprinter for ast datatypes (Elaborator/print/ppast.{sig,sml})
2287    has been added.
2288    
2289    ----------------------------------------------------------------------
2290    Name: Matthias Blume (blume (at) tti - c (dot) org)
2291    Date: 2003/08/11 15:45:00 CDT
2292    Tag: blume-20030811-windows
2293    Description:
2294    
2295    Version number bumped to 110.42.9.  NEW BOOTFILES!!!
2296    
2297        http://smlnj.cs.uchicago.edu/dist/working/110.42.9/
2298    
2299    This patch restores SML/NJ's ability to run under win32.  There are a
2300    number of changes, including fixes for several bugs that had gone
2301    unnoticed until now:
2302    
2303      - uname "CYGWIN_NT*" is recognized as win32 (This is relevant only when
2304        trying to run the win32 version from within cygwin.)
2305    
2306      - There are a number of simple .bat scripts that substitute for their
2307        corresponding Unix shell-scripts.  (See below.)
2308    
2309      - The internals of ml-build have been modified slightly.  The main
2310        difference is that instead of calling ".link-sml" (or link-sml.bat)
2311        using OS.Process.system, the ML process delegates this task back
2312        to the script.  Otherwise problems arise in mixed environments such
2313        as Cygwin where scripts look and work like Unix scripts, but
2314        where OS.Process.system cannot run them.
2315    
2316      - In CM, the srcpath pickler used native pathname syntax -- which
2317        is incorrect in the case of cross-compilation.  The new pickle format
2318        is independent of platform-specific naming conventions.
2319    
2320      - Path configuration files (such as lib/pathconfig) can now choose
2321        between native and standard syntax.  Placing a line of the form
2322    
2323            standard!
2324    
2325        into the file causes all subsequent paths to be interpreted using
2326        CM standard pathname syntax (= Unix conventions); a line
2327    
2328            native!
2329    
2330        switches back to native style.  This was needed so that
2331        path config files can be written portably, see src/system/pathconfig.
2332    
2333      - Runtime system:
2334    
2335          - win32-filesys.c:  get_file_time and set_file_time now
2336            access modification time, not creation time.
2337    
2338          - I/O code made aware of new array representation.
2339    
2340          - Bug fixes in X86.prim.masm.
2341    
2342       - src/system/makeml made aware of win32. (For use under cygwin
2343            and other Unix-environments for windows.)
2344    
2345       - In Basis, fixed off-by-one error in win32-io.sml (function vecF)
2346         which caused BinIO.inputAll to fail consistently.
2347    
2348    .bat scripts:
2349    
2350       Windows .bat scripts assume that SMLNJ_HOME is defined.
2351    
2352       - sml.bat, ml-yacc.bat, ml-lex.bat:  Driver scripts for standalone
2353         applications (sml, ml-yacc, ml-lex).
2354       - ml-build.bat: analogous to ml-build.
2355       - config\install.bat:  Analogous to config/install.sh.  This requires
2356         that SMLNJ_HOME is set and that Microsoft Visual C is ready to use.
2357         (nmake etc. must be on the path, and vcvars32 must have been run.)
2358         Moreover, sources for ml-lex and ml-yacc need to exist under src,
2359         and the bootfile hierarchy must have been unpacked under
2360         sml.boot.x86-win32.
2361         The script is very primitive and does a poor job at error checking.
2362         It only installs the base system, ml-lex, and ml-yacc.  No other
2363         libraries are being installed (i.e., you get only those that
2364         are part of the compiler.)
2365       - link-sml.bat:  analogous to .link-sml, but not currently used
2366    
2367    Unrelated bug fixes:
2368    
2369      - ml-nlffigen now exports structures ST_* corresponding to incomplete
2370        types.
2371      - Added getDevice to PP/src/pp-debug-fn.sml.  (Would not compile
2372        otherwise.)
2373    
2374    ----------------------------------------------------------------------
2375    Name: Dave MacQueen (macqueen@cs.uchicago.edu)
2376    Date: 2003/06/17
2377    Tag: macqueen-20030617-bug895
2378    Description:
2379    
2380    Modified compiler/Elaborator/print/pptype.sml to fix bug 895.
2381    Tag will be used for new development branch (mcz-branch) for
2382    use by MacQueen, (Lucasz) Zairek, and (George) Cao at uchicago.
2383    
2384    ----------------------------------------------------------------------
2385    Name: Matthias Blume (blume (at) tti - c (dot) org)
2386    Date: 2003/05/27 16:55:00 CDT
2387    Tag: blume-20030527-polyeq
2388    Description:
2389    
2390    Tried to eliminated most cases of polymorphic equality.
2391    
2392    ----------------------------------------------------------------------
2393    Name: Matthias Blume (blume (at) tti - c (dot) org)
2394    Date: 2003/05/21 17:45:00 CDT
2395    Tag: blume-20030517-complete
2396    Description:
2397    
2398    Two changes:
2399    
2400      1. Added a flag for controlling whether non-exhaustive bindings will
2401         be treated as errors (default is false).
2402      2. Cleaned up the *entire* source tree so that CMB.make goes through
2403         without a single non-exhaustive match- or bind warning.
2404    
2405    ----------------------------------------------------------------------
2406    Name: Matthias Blume (blume (at) tti - c (dot) org)
2407    Date: 2003/05/17 10:20:00 CDT
2408    Tag: blume-20030517-absyn
2409    Description:
2410    
2411    1. Added cases for IF, WHILE, ANDALSO, and ORELSE to Absyn.
2412    
2413       This mainly affects the quality of error messages.  However, some
2414       of the code is now more straightforward than before. (Treatment of
2415       the above four constructs in translate.sml is much simpler than
2416       the "macro-expansion" that was going on before.  Plus, the mach-
2417       compiler no longer gets invoked just to be able to compile an
2418       if-expression.)
2419    
2420    2. The ErrorMsg.Error exception is now caught and absorbed by the
2421       interactive loop.
2422    
2423    ----------------------------------------------------------------------
2424    Name: Allen Leung
2425    Date: 2003/05/16 13:05:00 CDT
2426    Tag: leunga-20030516-cygwin-runtime
2427    Description:
2428    
2429        Ported the runtime system to cygwin, which uses the unix
2430    x86-unix bin files.  Missing/buggy features:
2431    
2432       o getnetbyname, getnetbyaddr: these functions seem to be missing in
2433         the Cygwin library.
2434       o Ctrl-C handling may be flaky.
2435       o Windows system calls and Windows I/O are not supported.
2436    
2437       A new set of binfiles is located at:
2438    
2439         http://www.dorsai.org/~leunga/boot.x86-unix.tgz
2440    
2441       This is only needed for bootstrapping the cygwin version of smlnj.
2442       Other x86 versions can use the existing binfiles.
2443    
2444    ----------------------------------------------------------------------
2445    Name: Matthias Blume
2446    Date: 2003/04/08 15:42:00 CDT
2447    Tag: blume-20030408-listpair
2448    Description:
2449    
2450    1. Added a target 'mlrisc' to installer.
2451    
2452    2. Added missing elements to structure ListPair.
2453    
2454    ----------------------------------------------------------------------
2455    Name: Allen Leung
2456    Date: 2003/01/07 10:40:00 EST
2457    Tag: leunga-20030107-int-rem
2458    Description:
2459    
2460    Fixed a bug in Int.rem(x,y) where y is a power of 2 on x86.
2461    The arguments to the SUBL instruction were swapped.
2462    
2463    ----------------------------------------------------------------------
2464    Name: Matthias Blume
2465    Date: 2002/12/12 16:25:00 EST
2466    Tag: blume-20021212-risc-ra
2467    Description:
2468    
2469    Fixed a serious bug in the rewrite code for FP spilling/reloading that
2470    sent the RA into an infinite loop when floating point registers get
2471    spilled.  (Because of this bug, e.g., nucleic stopped compiling between
2472    110.37 and 110.38.)
2473    There was another set of potential problems related to the handling of
2474    MLRISC annotations (but those did not yet cause real problems, apparently).
2475    
2476    ----------------------------------------------------------------------
2477    Name: Matthias Blume
2478    Date: 2002/12/06 22:40:00 EST
2479    Tag: blume-20021206-cm-fileid
2480    Description:
2481    
2482    Added a call of SrcPath.sync at the beginning of Parse.parse (in CM).
2483    This fixes the problem of CM getting confused by files that suddenly
2484    change their identity (e.g., by getting unlinked and recreated by some
2485    text editor such as vi).  There might be a better/cheaper/cleaner way
2486    of doing this, but for now this will have to do.
2487    
2488    ----------------------------------------------------------------------
2489    Name: Matthias Blume
2490    Date: 2002/10/28 09:50:00 EST
2491    Tag: blume-20021028-typecheck
2492    Description:
2493    
2494    Exported structure Typecheck from $smlnj/viscomp/core.cm.
2495    
2496    ----------------------------------------------------------------------
2497    Name: Matthias Blume
2498    Date: 2002/10/17 09:10:00 EDT
2499    Tag: Release_110_42
2500    Description:
2501    
2502    In good old tradition, there has been a slight hiccup so that we have
2503    to patch 110.42 after the fact.   The old release tag has been replaced
2504    (see below).
2505    
2506    The change solves a problem with two competing approaches the
2507    configuration problem regarding MacOS 10.1 vs. MacOS 10.2 which got in
2508    each other's way.
2509    
2510    This change only affects the runtime system code and the installer script.
2511    (No new bootfiles.)
2512    
2513    ----------------------------------------------------------------------
2514    Name: Matthias Blume
2515    Date: 2002/10/16 12:00:00 EDT
2516    Tag: Release_110_42_removed
2517    Description:
2518    
2519    New working release.  New bootfiles.
2520    
2521    ----------------------------------------------------------------------
2522    Name: Matthias Blume
2523    Date: 2002/10/10 13:10:00 EDT
2524    Tag: blume-20021010-ppc-divs
2525    Description:
2526    
2527    The mltree operator DIVS must be implemented with an overflow check on
2528    the PPC because the hardware indicates divide-by-zero using "overflow" as
2529    well.
2530    
2531    ----------------------------------------------------------------------
2532    Name: Matthias Blume
2533    Date: 2002/07/23 11:45:00
2534    Tag: blume-20020723-smlnj-home
2535    Description:
2536    
2537    Sml now senses the SMLNJ_HOME environment variable.  If this is set,
2538    then the bin dir is assumed to be in $SMLNJ_HOME/bin and (unless
2539    CM_PATHCONFIG is also set), the path configuration file is assumed
2540    to be in $SMLNJ_HOME/lib/pathconfig.  This way one can easily move
2541    the entire tree to some other place and everything will "just work".
2542    
2543    (Companion commands such as ml-build and ml-makedepend also sense this
2544    variable.)
2545    
2546    ----------------------------------------------------------------------
2547    Name: Matthias Blume
2548    Date: 2002/07/12 21:19:00 EDT
2549    Tag: blume-20020712-liveness
2550    Description:
2551    
2552    Exported two useful "step" functions from liveness module (MLRISC).
2553    
2554    ----------------------------------------------------------------------
2555    Name: Matthias Blume
2556    Date: 2002/07/05 16:00 EDT
2557    Tag: Release_110_41
2558    Description:
2559    
2560    New working release.  New bootfiles.
2561    
2562    ----------------------------------------------------------------------
2563    Name: Matthias Blume
2564    Date: 2002/07/05 10:25:00 EDT
2565    Tag: blume-20020705-btimp
2566    Description:
2567    
2568    Exported structure BTImp from $smlnj/viscomp/debugprof.cm so that
2569    other clients can set up backtracing support.
2570    
2571    ----------------------------------------------------------------------
2572    Name: Matthias Blume
2573    Date: 2002/06/25 17:23:00 EDT
2574    Tag: blume-20020625-fpmax
2575    Description:
2576    
2577    Fixed a bug in translation of INLMAX (and INLMIN) for the floating-point
2578    case.  (The sense of the isNaN test was reversed -- which made min and
2579    max always return their first argument.)
2580    
2581    ----------------------------------------------------------------------
2582    Name: Matthias Blume
2583    Date: 2002/06/11
2584    Tag: blume-20020611-unixpath
2585    Description:
2586    
2587    Back-ported OS.Path.{from,to}UnixPath from idlbasis-devel branch.
2588    
2589    ----------------------------------------------------------------------
2590    Name: Matthias Blume
2591    Date: 2002/06/10 16:35:00 EDT
2592    Tag: blume-20020610-ieeereal
2593    Description:
2594    
2595    I back-ported my implementation of IEEEReal.fromString from the
2596    idlbasis-devel branch so that we can test it.
2597    
2598    Another small change is that ppDec tries to give more information
2599    than just "<sig>" in the case of functors.  However, this code is
2600    broken in some mysterious way if the functor's body's signature
2601    has not been declared by ascription but gets inferred from the
2602    implementation.  This needs fixing...
2603    
2604    ----------------------------------------------------------------------
2605    Name: Matthias Blume
2606    Date: 2002/05/31
2607    Tag: blume-20020531-btrace-mode
2608    Description:
2609    
2610    Resurrected SMLofNJ.Internals.BTrace.mode.  (It accidentally fell by
2611    the wayside when I switched over to using Controls everywhere.)
2612    
2613    ----------------------------------------------------------------------
2614    Name: Lal George
2615    Date: 2002/05/23 12:21:40 EDT
2616    Tag: george-20020523-visual-labels
2617    Description:
2618    
2619    Labels are now displayed in the graphical output to make
2620    the fall-through and target blocks obvious.
2621    
2622    ----------------------------------------------------------------------
2623    Name: Matthias Blume
2624    Date: 2002/05/22 11:03:00 EDT
2625    Tag: blume-20020522-shrink
2626    Description:
2627    
2628    John tweaked yesterday's fix for 1131 to handle an out-of-memory
2629    situation that comes up when allocating huge arrays.
2630    
2631    ----------------------------------------------------------------------
2632    Name: Matthias Blume
2633    Date: 2002/05/21 16:00:00 EDT
2634    Tag: Release_110_40
2635    Description:
2636    
2637    New working release (110.40).  New bootfiles.
2638    
2639    [Also: John Reppy fixed GC bug 1131.]
2640    
2641    ----------------------------------------------------------------------
2642    Name: Matthias Blume
2643    Date: 2002/05/21 12:35:00 EDT
2644    Tag: blume-20020521-cmdoc
2645    Description:
2646    
2647    CM documentation update.
2648    
2649    ----------------------------------------------------------------------
2650    Name: Matthias Blume
2651    Date: 2002/05/21 10:55:00 EDT
2652    Tag: blume-20020521-misc
2653    Description:
2654    
2655    - John tweaked runtime to be silent on heap export (except when
2656      GC messages are on).
2657    - I added a few more things (cross-compiling versions of CMB) to
2658      config/preloads (as suggestions).
2659    
2660    ----------------------------------------------------------------------
2661    Name: Matthias Blume
2662    Date: 2002/05/20 22:25:00 EDT
2663    Tag: blume-20020520-controls
2664    Description:
2665    
2666    - Added ControlUtil structure to control-lib.cm.
2667    - Use it throughout.
2668    - Used Controls facility to define MLRISC controls (as opposed to
2669      registering MLRISC control ref cells with Controls after the fact)
2670    - Fixed messed-up controls priorities.
2671    
2672    * Removed again all the stuff from config/preloads that one wouldn't
2673      be able to preload at the time the initial heap image is built.
2674      (Many libraries, e.g., CML, do not exist yet at this time.  The
2675       only libraries that can be preloaded via config/preloads are those
2676       that come bundled with the bootfiles.)
2677    
2678    ----------------------------------------------------------------------
2679    Name: Matthias Blume
2680    Date: 2002/05/20 10:59:00 EDT
2681    Tag: blume-20020520-preloads
2682    Description:
2683    
2684    Added a lot of commented-out suggestions for things to be included
2685    in config/preloads.
2686    
2687    ----------------------------------------------------------------------
2688    Name: Allen Leung
2689    Date: 2002/05/18 14:20:00 EDT
2690    Tag: leunga-20020518-mdl
2691    Description:
2692    
2693    o Made the mdl tool stuff compile and run again.
2694    o I've disabled all the stuff that depends on RTL specifications; they
2695      are all badly broken anyway.
2696    
2697    ----------------------------------------------------------------------
2698    Name: Matthias Blume
2699    Date: 2002/05/17 16:49:00 EDT
2700    Tag: blume-20020517-controls
2701    Description:
2702    
2703    0. John Reppy made several modifications to the SML/NJ library.
2704       In particular, there is a shiny new controls-lib.cm.
2705    
2706    1. Pushed new controls interface through compiler so that everything
2707       compiles again.
2708    
2709    2. Added FormatComb and FORMAT_COMB to the CML version of the
2710       SML/NJ library (so that CML compiles again).
2711    
2712    3. Modified init scripts because XXX_DEFAULT environment variables
2713       are no longer with us.  (Boot-time initialization is now done
2714       using the same environment variables that are also used for
2715       startup-time initialization of controls.)
2716    
2717    ----------------------------------------------------------------------
2718    Name: Lal George
2719    Date: 2002/05/15 09:20:10 EDT
2720    Tag: george-20020515-pseudo-op-decls
2721    Description:
2722    
2723            All pseudo-ops emitted before the first segment declaration
2724    such as TEXT, DATA, and BSS directives are assumed to be global
2725    declarations and are emitted first in the assembly file. This is
2726    useful in a number of situations where one has pseudo-ops that are not
2727    specific to any segment, and also works around the constraint that one
2728    cannot have client pseudo-ops in the TEXT segment.
2729    
2730            Because no segment is associated with these declarations it is
2731    an error to allocate any space or objects before the first segment
2732    directive and an exception will be raised. However, we cannot make
2733    this check for client pseudo-ops.
2734    
2735            These top level declarations are a field in the CFG graph_info.
2736    In theory you can continue to add to this field after the CFG has been
2737    built -- provided you know what you are doing;-)
2738    
2739    ----------------------------------------------------------------------
2740    Name: Matthias Blume
2741    Date: 2002/05/13 16:40:00 EDT
2742    Tag: blume-20020513-pp-etc
2743    Description:
2744    
2745    A few minor bugfixes:
2746    
2747      - Stopgap measure for bug recently reported by Elsa Gunter (ppDec).
2748        (Bogus printouts for redefined bindings still occur.  Compiler
2749        bug should no longer occur now.  We need to redo the prettyprinter
2750        from scratch.)
2751    
2752      - CM pathname printer now also adds escape sequences for ( and )
2753    
2754      - commend and docu fixes for ml-nlffi
2755    
2756    ----------------------------------------------------------------------
2757    Name: Matthias Blume
2758    Date: 2002/05/10 16:40:00 EDT
2759    Tag: blume-20020510-erg-textio
2760    Description:
2761    
2762    Applied the following bugfix provided by Emden Gansner:
2763    
2764        Output is corrupted when outputSubstr is used rather than output.
2765    
2766        The problem occurs when a substring
2767    
2768            ss = (s, dataStart, dataLen)
2769    
2770        where dataStart > 0, fills a stream buffer with avail bytes left.
2771        avail bytes of s, starting at index dataStart, are copied into the
2772        buffer, the buffer is flushed, and then the remaining dataLen-avail
2773        bytes of ss are copied into the beginning of the buffer. Instead of
2774        starting this copy at index dataStart+avail in s, the current code
2775        starts the copy at index avail.
2776    
2777        Fix:
2778        In text-io-fn.sml, change line 695 from
2779             val needsFlush = copyVec(v, avail, dataLen-avail, buf, 0)
2780        to
2781             val needsFlush = copyVec(v, dataStart+avail, dataLen-avail, buf, 0)
2782    
2783    ----------------------------------------------------------------------
2784    Name: Matthias Blume
2785    Date: 2002/04/12 13:55:00 EDT
2786    Tag: blume-20020412-assyntax
2787    Description:
2788    
2789    1. Grabbed newer assyntax.h from the XFree86 project.
2790    2. Fiddled with how to compile X86.prim.asm without warnings.
2791    3. (Very) Minor cleanup in CM.
2792    
2793    ----------------------------------------------------------------------
2794    Name: Matthias Blume
2795    Date: 2002/04/01 (no joke!) 17:07:00 EST
2796    Tag: blume-20020401-x86div
2797    Description:
2798    
2799    Added full support for div/mod/rem/quot on the x86, using the machine
2800    instruction's two results (without clumsily recomputing the remainder)
2801    directly where appropriate.
2802    
2803    Some more extensive power-of-two support was added to the x86 instruction
2804    selector (avoiding expensive divs, mods, and muls where they can be
2805    replaced with cheaper shifts and masks).  However, this sort of thing
2806    ought to be done earlier, e.g., within the CPS optimizer so that
2807    all architectures benefit from it.
2808    
2809    The compiler compiles to a fixed point, but changes might be somewhat
2810    fragile nevertheless.  Please, report any strange things that you might
2811    see wrt. div/mod/quot/rem...
2812    
2813    ----------------------------------------------------------------------
2814    Name: Matthias Blume
2815    Date: 2002/03/29 17:22:00
2816    Tag: blume-20020329-div
2817    Description:
2818    
2819    Fixed my broken div/mod logic.  Unfortunately, this means that the
2820    inline code for div/mod now has one more comparison than before.
2821    Fast paths (quotient > 0 or remainder = 0) are not affected, though.
2822    The problem was with quotient = 0, because that alone does not tell
2823    us which way the rounding went.  One then has to look at whether
2824    remainder and divisor have the same sign...  :(
2825    
2826    Anyway, I replaced the bootfiles with fresh ones...
2827    
2828    ----------------------------------------------------------------------
2829    Name: Matthias Blume
2830    Date: 2002/03/29 14:10:00 EST
2831    Tag: blume-20020329-inlprims
2832    Description:
2833    
2834    NEW BOOTFILES!!!    Version number bumped to 110.39.3.
2835    
2836    Primops have changed. This means that the bin/boot-file formats have
2837    changed as well.
2838    
2839    To make sure that there is no confusion, I made a new version.
2840    
2841    
2842    CHANGES:
2843    
2844    * removed REMT from mltree (remainder should never overflow).
2845    
2846    * added primops to deal with divisions of all flavors to the frontend
2847    
2848    * handled these primops all the way through so they map to their respective
2849      MLRISC support
2850    
2851    * used these primops in the implementation of Int, Int32, Word, Word32
2852    
2853    * removed INLDIV, INLMOD, and INLREM as they are no longer necessary
2854    
2855    * parameterized INLMIN, INLMAX, and INLABS by a numkind
2856    
2857    * translate.sml now deals with all flavors of INL{MIN,MAX,ABS}, including
2858      floating point
2859    
2860    * used INL{MIN,MAX,ABS} in the implementation of Int, Int32, Word, Word32,
2861      and Real (but Real.abs maps to a separate floating-point-only primop)
2862    
2863    
2864    TODO items:
2865    
2866    * Hacked Alpha32 instruction selection, disabling the selection of REMx
2867      instructions because the machine instruction encoder cannot handle
2868      them.  (Hppa, PPC, and Sparc instruction selection did not handle
2869      REM in the first place, and REM is supported by the x86 machine coder.)
2870    
2871    * Handle DIV and MOD with DIV_TO_NEGINF directly in the x86 instruction
2872      selection phase.  (The two can be streamlined because the hardware
2873      delivers both quotient and remainder at the same time anyway.)
2874    
2875    * Think about what to do with "valOf(Int32.minInt) div ~1" and friends.
2876      (Currently the behavior is inconsistent both across architectures and
2877      wrt. the draft Basis spec.)
2878    
2879    * Word8 should eventually be handled natively, too.
2880    
2881    * There seems to be one serious bug in mltree-gen.sml.  It appears, though,
2882      as if there currently is no execution path that could trigger it in
2883      SML/NJ.  (The assumptions underlying functions arith and promotable do not
2884      hold for things like multiplication and division.)
2885    
2886    ----------------------------------------------------------------------
2887    Name: Matthias Blume
2888    Date: 2002/03/27 16:27:00 EST
2889    Tag: blume-20020327-mlrisc-divisions
2890    Description:
2891    
2892    Added support for all four division operations (ML's div, mod, quot,
2893    and rem) to MLRISC.  In the course of doing so, I also rationalized
2894    the naming (no more annoying switch-around of DIV and QUOT), by
2895    parameterizing the operation by div_rounding_mode (which can be either
2896    DIV_TO_ZERO or DIV_TO_NEGINF).
2897    
2898    The generic MLTreeGen functor takes care of compiling all four
2899    operations down to only round-to-zero div.
2900    
2901    Missing pieces:
2902    
2903      * Doing something smarter than relying on MLTreeGen on architectures
2904        like, e.g., the x86 where hardware division delivers both quotient and
2905        remainder at the same time.  With this, the implementation of the
2906        round-to-neginf operations could be further streamlined.
2907    
2908      * Remove inlining support for div/mod/rem from the frontend and replace it
2909        with primops that get carried through to the backend.  Do this for all
2910        int and word types.
2911    
2912    ----------------------------------------------------------------------
2913    Name: Matthias Blume
2914    Date: 2002/03/25 17:25:00 EST
2915    Tag: blume-20020325-divmod
2916    Description:
2917    
2918    I improved (hopefully without breaking them) the implementation of Int.div,
2919    Int.mod, and Int.rem.   For this, the code in translate.sml now takes
2920    advantage of the following observations:
2921    
2922      Let  q = x quot y      r = x rem y
2923           d = x div  y      m = x mod y
2924    
2925    where "quot" is the round-to-zero version of integer division that
2926    hardware usually provides.  Then we have:
2927    
2928         r = x - q * y        where neither the * nor the - will overflow
2929         d = if q >= 0 orelse x = q * y then q else q - 1
2930                              where neither the * nor the - will overflow
2931         m = if q >= 0 orelse r = 0 then r else r + y
2932                              where the + will not overflow
2933    
2934    This results in substantial simplification of the generated code.
2935    The following table shows the number of CFG nodes and edges generated
2936    for
2937            fun f (x, y) = x OPER y
2938            (* with OPER \in div, mod, quot, rem *)
2939    
2940    
2941        OPER | nodes(old) | edges(old) | nodes(new) | edges(new)
2942        --------------------------------------------------------
2943         div |         24 |         39 |         12 |         16
2944         mod |         41 |         71 |         12 |         16
2945        quot |          8 |         10 |          8 |         10
2946         rem |         10 |         14 |          8 |         10
2947    
2948    
2949    ----------------------------------------------------------------------
2950    Name: Matthias Blume
2951    Date: 2002/03/25 22:06:00 EST
2952    Tag: blume-20020325-cprotobug
2953    Description:
2954    
2955    Fixed a bug in cproto (c prototype decoder).
2956    
2957    ----------------------------------------------------------------------
2958    Name: Matthias Blume
2959    Date: 2002/03/25 16:00:00 EST
2960    Tag: blume-20020325-raw-primops
2961    Description:
2962    
2963    I did some cleanup to Allen's new primop code and
2964    replaced yesterday's bootfiles with new ones.
2965    (But they are stored in the same place.)
2966    
2967    ----------------------------------------------------------------------
2968    Name: Matthias Blume
2969    Date: 2002/03/24 22:40:00 EST
2970    Tag: blume-20020324-bootfiles
2971    Description:
2972    
2973    Made the bootfiles that Allen asked for.
2974    
2975    ----------------------------------------------------------------------
2976    Name: Allen Leung
2977    Date: 2002/03/23 15:50:00 EST
2978    Tag: leunga-20020323-flint-cps-rcc-primops
2979    Description:
2980    
2981      1. Changes to FLINT primops:
2982    
2983        (* make a call to a C-function;
2984         * The primop carries C function prototype information and specifies
2985         * which of its (ML-) arguments are floating point. C prototype
2986         * information is for use by the backend, ML information is for
2987         * use by the CPS converter. *)
2988      | RAW_CCALL of { c_proto: CTypes.c_proto,
2989                       ml_args: ccall_type list,
2990                       ml_res_opt: ccall_type option,
2991                       reentrant : bool
2992                     } option
2993       (* Allocate uninitialized storage on the heap.
2994        * The record is meant to hold short-lived C objects, i.e., they
2995        * are not ML pointers.  With the tag, the representation is
2996        * the same as RECORD with tag tag_raw32 (sz=4), or tag_fblock (sz=8)
2997        *)
2998      | RAW_RECORD of {tag:bool,sz:int}
2999      and ccall_type = CCALL_INT32 | CCALL_REAL64 | CCALL_ML_PTR
3000    
3001      2.  These CPS primops are now overloaded:
3002    
3003           rawload of {kind:numkind}
3004           rawstore of {kind:numkind}
3005    
3006          The one argument form is:
3007    
3008             rawload {kind} address
3009    
3010          The two argument form is:
3011    
3012             rawload {kind} [ml object, byte-offset]
3013    
3014      3. RAW_CCALL/RCC now takes two extra arguments:
3015    
3016         a. The first is whether the C call is reentrant, i.e., whether
3017            ML state should be saved and restored.
3018         b. The second argument is a string argument specifying the name of
3019            library and the C function.
3020    
3021         These things are currently not handled in the code generator, yet.
3022    
3023      4. In CProto,
3024    
3025         An encoding type of "bool" means "ml object" and is mapped into
3026         C prototype of PTR.  Note that "bool" is different than "string",
3027         even though "string" is also mapped into PTR, because "bool"
3028         is assigned an CPS type of BOGt, while "string" is assigned INT32t.
3029    
3030      5. Pickler/unpicker
3031    
3032         Changed to handle RAW_RECORD and newest RAW_CCALL
3033    
3034      6. MLRiscGen,
3035    
3036         1. Changed to handle the new rawload/rawstore/rawrecord operators.
3037         2. Code for handling C Calls has been moved to a new module CPSCCalls,
3038            in the file CodeGen/cpscompile/cps-c-calls.sml
3039    
3040      7. Added the conditional move operator
3041    
3042             condmove of branch
3043    
3044         to cps.  Generation of this is still buggy so it is currently
3045         disabled.
3046    
3047    ----------------------------------------------------------------------
3048    Name: Lal George
3049    Date: 2002/03/22 14:18:25 EST
3050    Tag: george-20020322-cps-branch-prob
3051    Description:
3052    
3053    Implemented the Ball-Larus branch prediction-heuristics, and
3054    incorporated graphical viewers for control flow graphs.
3055    
3056    Ball-Larus Heuristics:
3057    ---------------------
3058    See the file compiler/CodeGen/cpscompile/cpsBranchProb.sml.
3059    
3060    By design it uses the Dempster-Shafer theory for combining
3061    probabilities.  For example, in the function:
3062    
3063        fun f(n,acc) = if n = 0 then acc else f(n-1, n*acc)
3064    
3065    the ball-larus heuristics predicts that the n=0 is unlikely
3066    (OH-heuristic), and the 'then' branch is unlikely because of the
3067    RH-heuristic -- giving the 'then' branch an even lower combined
3068    probability using the Dempster-Shafer theory.
3069    
3070    Finally, John Reppy's loop analysis in MLRISC, further lowers the
3071    probability of the 'then' branch because of the loop in the else
3072    branch.
3073    
3074    
3075    Graphical Viewing:
3076    ------------------
3077    I merely plugged in Allen's graphical viewers into the compiler. The
3078    additional code is not much. At the top level, saying:
3079    
3080            Control.MLRISC.getFlag "cfg-graphical-view" := true;
3081    
3082    will display the graphical view of the control flow graph just before
3083    back-patching.  daVinci must be in your path for this to work. If
3084    daVinci is not available, then the default viewer can be changed
3085    using:
3086    
3087            Control.MLRISC.getString "viewer"
3088    
3089    which can be set to "dot" or "vcg" for the corresponding viewers. Of
3090    course, these viewers must be in your path.
3091    
3092    The above will display the compilation unit at the level of clusters,
3093    many of which are small, boring, and un-interesting. Also setting:
3094    
3095            Control.MLRISC.getInt "cfg-graphical-view_size"
3096    
3097    will display clusters that are larger than the value set by the above.
3098    
3099    
3100    ----------------------------------------------------------------------
3101    Name: Matthias Blume
3102    Date: 2002/03/21 22:20:00 EST
3103    Tag: blume-20020321-kmp-bugfix
3104    Description:
3105    
3106    Changed the interface to the KMP routine in PreString and fixed
3107    a minor bug in one place where it was used.
3108    
3109    ----------------------------------------------------------------------
3110    Name: Allen Leung
3111    Date: 2002/03/21 20:30:00 EST
3112    Tag: leunga-20020321-cfg
3113    Description:
3114    
3115      Fixed a potential problem in cfg edge splitting.
3116    
3117    ----------------------------------------------------------------------
3118    Name: Allen Leung
3119    Date: 2002/03/21 17:15:00 EST
3120    Tag: leunga-20020321-x86-fp-cfg
3121    Description:
3122    
3123      1. Recoded the buggy parts of x86-fp.
3124    
3125         a. All the block reordering code has been removed.
3126            We now depend on the block placement phases to do this work.
3127    
3128         b. Critical edge splitting code has been simplified and moved into the
3129            CFG modules, as where they belong.
3130    
3131         Both of these were quite buggy and complex.  The code is now much, much
3132         simpler.
3133    
3134      2. X86 backend.
3135    
3136         a. Added instructions for 64-bit support.  Instruction selection for
3137            64-bit has not been committed, however, since that
3138            requires changes to MLTREE which haven't been approved by
3139            Lal and John.
3140    
3141         b. Added support for FUCOMI and FUCOMIP when generating code for
3142            PentiumPro and above.  We only generate these instructions in
3143            the fast-fp mode.
3144    
3145         c. Added cases for JP and JNP in X86FreqProps.
3146    
3147      3. CFG
3148    
3149         CFG now has a bunch of methods for edge splitting and merging.
3150    
3151      4. Machine description.
3152    
3153         John's simplification of MLTREE_BASIS.fcond broke a few machine
3154         description things:
3155    
3156         rtl-build.{sig,sml} and hppa.mdl fixed.
3157    
3158         NOTE: the machine description stuff in the repository is still broken.
3159               Again, I can't put my fixes in because that involves
3160               changes to MLTREE.
3161    
3162    ----------------------------------------------------------------------
3163    Name: Matthias Blume
3164    Date: 2002/03/20 15:55:00 EST
3165    Tag: blume-20020320-kmp
3166    Description:
3167    
3168    Implemented Knuth-Morris-Pratt string matching in PreString and used
3169    it for String.isSubstring, Substring.isSubstring, and
3170    Substring.position.
3171    
3172    (Might need some stress-testing.  Simple examples worked fine.)
3173    
3174    ----------------------------------------------------------------------
3175    Name: Matthias Blume
3176    Date: 2002/03/19 16:37:00 EST
3177    Tag: blume-20020319-witnesses
3178    Description:
3179    
3180    Added a structure C.W and functions convert/Ptr.convert to ml-nlffi-lib.
3181    
3182    This implements a generic mechanism for changing constness qualifiers
3183    anywhere within big C types without resorting to outright "casts".
3184    (So far, functions such as C.rw/C.ro or C.Ptr.rw/C.Ptr.ro only let you
3185    modify the constness at the outermost level.)
3186    The implementation of "convert" is based on the idea of "witness"
3187    values -- values that are not used by the operation but whose types
3188    "testify" to their applicability.  On the implementation side, "convert"
3189    is simply a projection (returning its second curried argument).  With
3190    cross-module inlining, it should not result in any machine code being
3191    generated.
3192    
3193    ----------------------------------------------------------------------
3194    Name: Matthias Blume
3195    Date: 2002/03/15 16:40:00 EST
3196    Tag: blume-20020315-basis
3197    Description:
3198    
3199    Provided (preliminary?) implementations for
3200    
3201      {String,Substring}.{concatWith,isSuffix,isSubstring}
3202    
3203    and
3204    
3205      Substring.full
3206    
3207    Those are in the Basis spec but they were missing in SML/NJ.
3208    
3209    ----------------------------------------------------------------------
3210    Name: Matthias Blume
3211    Date: 2002/03/14 21:30:00 EST
3212    Tag: blume-20020314-controls
3213    Description:
3214    
3215    Controls:
3216    ---------
3217    
3218    1. Factored out the recently-added Controls : CONTROLS stuff and put
3219       it into its own library $/controls-lib.cm.  The source tree for
3220       this is under src/smlnj-lib/Controls.
3221    
3222    2. Changed the names of types and functions in this interface, so they
3223       make a bit more "sense":
3224    
3225          module -> registry
3226          'a registry -> 'a group
3227    
3228    3. The interface now deals in ref cells only.  The getter/setter interface
3229       is (mostly) gone.
3230    
3231    4. Added a function that lets one register an already-existing ref cell.
3232    
3233    5. Made the corresponding modifications to the rest of the code so that
3234       everything compiles again.
3235    
3236    6. Changed the implementation of Controls.MLRISC back to something closer
3237       to the original.  In particular, this module (and therefore MLRISC)
3238       does not depend on Controls.  There now is some link-time code in
3239       int-sys.sml that registers the MLRISC controls with the Controls
3240       module.
3241    
3242    CM:
3243    ---
3244    
3245      * One can now specify the lambda-split aggressiveness in init.cmi.
3246    
3247    ----------------------------------------------------------------------
3248    Name: Allen Leung
3249    Date: 2002/03/13 17:30:00 EST
3250    Tag: leunga-20020313-x86-fp-unary
3251    Description:
3252    
3253    Bug fix for:
3254    
3255    > leunga@weaselbane:~/Yale/tmp/sml-dist{21} bin/sml
3256    > Standard ML of New Jersey v110.39.1 [FLINT v1.5], March 08, 2002
3257    > - fun f(x,(y,z)) = Real.~ y;
3258    > [autoloading]
3259    > [autoloading done]
3260    >       fchsl   (%eax), 184(%esp)
3261    > Error: MLRisc bug: X86MCEmitter.emitInstr
3262    >
3263    > uncaught exception Error
3264    >   raised at: ../MLRISC/control/mlriscErrormsg.sml:16.14-16.19
3265    
3266    The problem was that the code generator did not generate any fp registers
3267    in this case, and the ra didn't know that it needed to run the X86FP phase to
3268    translate the pseudo fp instruction.   This only happened with unary fp
3269    operators in certain situations.
3270    
3271    ----------------------------------------------------------------------
3272    Name: Matthias Blume
3273    Date: 2002/03/13 14:00:00 EST
3274    Tag: blume-20020313-overload-etc
3275    Description:
3276    
3277    1. Added _overload as a synonym for overload for backward compatibility.
3278       (Control.overloadKW must be true for either version to be accepted.)
3279    
3280    2. Fixed bug in install script that caused more things to be installed
3281       than what was requested in config/targets.
3282    
3283    3. Made CM aware of the (_)overload construct so that autoloading
3284       works.
3285    
3286    ----------------------------------------------------------------------
3287    Name: Matthias Blume
3288    Date: 2002/03/12 22:03:00 EST
3289    Tag: blume-20020312-url
3290    Description:
3291    
3292    Forgot to update BOOT and srcarchiveurl.
3293    
3294    ----------------------------------------------------------------------
3295    Name: Matthias Blume
3296    Date: 2002/03/12 17:30:00 EST
3297    Tag: blume-20020312-version110392
3298    Description:
3299    
3300    Yet another version number bump (because of small changes to the
3301    binfile format).  Version number is now 110.39.2.  NEW BOOTFILES!
3302    
3303    Changes:
3304    
3305      The new pid generation scheme described a few weeks ago was overly
3306      complicated.  I implemented a new mechanism that is simpler and
3307      provides a bit more "stability":  Once CM has seen a compilation
3308      unit, it keeps its identity constant (as long as you do not delete
3309      those crucial CM/GUID/* files).  This means that when you change
3310      an interface, compile, then go back to the old interface, and
3311      compile again, you arrive at the original pid.
3312    
3313      There now also is a mechanism that instructs CM to use the plain
3314      environment hash as a module's pid (effectively making its GUID
3315      the empty string).  For this, "noguid" must be specified as an
3316      option to the .sml file in question within its .cm file.
3317      This is most useful for code that is being generated by tools such
3318      as ml-nlffigen (because during development programmers tend to
3319      erase the tool's entire output directory tree including CM's cached
3320      GUIDs).  "noguid" is somewhat dangerous (since it can be used to locally
3321      revert to the old, broken behavior of SML/NJ, but in specific cases
3322      where there is no danger of interface confusion, its use is ok
3323      (I think).
3324    
3325      ml-nlffigen by default generates "noguid" annotations.  They can be
3326      turned off by specifying -guid in its command line.
3327    
3328    ----------------------------------------------------------------------
3329    Name: Lal George
3330    Date: 2002/03/12 12 14:42:36 EST
3331    Tag: george-20020312-frequency-computation
3332    Description:
3333    
3334    Integrated jump chaining and static block frequency into the
3335    compiler. More details and numbers later.
3336    
3337    ----------------------------------------------------------------------
3338    Name: Lal George
3339    Date: 2002/03/11 11 22:38:53 EST
3340    Tag: george-20020311-jump-chain-elim
3341    Description:
3342    
3343    Tested the jump chain elimination on all architectures (except the
3344    hppa).  This is on by default right now and is profitable for the
3345    alpha and x86, however, it may not be profitable for the sparc and ppc
3346    when compiling the compiler.
3347    
3348    The gc test will typically jump to a label at the end of the cluster,
3349    where there is another jump to an external cluster containing the actual
3350    code to invoke gc. This is to allow factoring of common gc invocation
3351    sequences. That is to say, we generate:
3352    
3353            f:
3354               testgc
3355               ja   L1      % jump if above to L1
3356    
3357            L1:
3358               jmp L2
3359    
3360    
3361    After jump chain elimination the 'ja L1' instructions is converted to
3362    'ja L2'. On the sparc and ppc, many of the 'ja L2' instructions may end
3363    up being implemented in their long form (if L2 is far away) using:
3364    
3365            jbe     L3      % jump if below or equal to L3
3366            jmp     L2
3367         L3:
3368            ...
3369    
3370    
3371    For large compilation units L2  may be far away.
3372    
3373    
3374    ----------------------------------------------------------------------
3375    Name: Matthias Blume
3376    Date: 2002/03/11 13:30:00 EST
3377    Tag: blume-20020311-mltreeeval
3378    Description:
3379    
3380    A functor parameter was missing.
3381    
3382    ----------------------------------------------------------------------
3383    Name: Allen Leung
3384    Date: 2002/03/11 10:30:00 EST
3385    Tag: leunga-20020311-runtime-string0
3386    Description:
3387    
3388       The representation of the empty string now points to a
3389    legal null terminated C string instead of unit.  It is now possible
3390    to convert an ML string into C string with InlineT.CharVector.getData.
3391    This compiles into one single machine instruction.
3392    
3393    ----------------------------------------------------------------------
3394    Name: Allen Leung
3395    Date: 2002/03/10 23:55:00 EST
3396    Tag: leunga-20020310-x86-call
3397    Description:
3398    
3399       Added machine generation for CALL instruction (relative displacement mode)
3400    
3401    ----------------------------------------------------------------------
3402    Name: Matthias Blume
3403    Date: 2002/03/08 16:05:00
3404    Tag: blume-20020308-entrypoints
3405    Description:
3406    
3407    Version number bumped to 110.39.1.  NEW BOOTFILES!
3408    
3409    Entrypoints: non-zero offset into a code object where execution should begin.
3410    
3411    - Added the notion of an entrypoint to CodeObj.
3412    - Added reading/writing of entrypoint info to Binfile.
3413    - Made runtime system bootloader aware of entrypoints.
3414    - Use the address of the label of the first function given to mlriscGen
3415      as the entrypoint.  This address is currently always 0, but it will
3416      not be 0 once we turn on block placement.
3417    - Removed the linkage cluster code (which was The Other Way(tm) of dealing
3418      with entry points) from mlriscGen.
3419    
3420    ----------------------------------------------------------------------
3421    Name: Allen Leung
3422    Date: 2002/03/07 20:45:00 EST
3423    Tag: leunga-20020307-x86-cmov
3424    Description:
3425    
3426       Bug fixes for CMOVcc on x86.
3427    
3428       1. Added machine code generation for CMOVcc
3429       2. CMOVcc is now generated in preference over SETcc on PentiumPro or above.
3430       3. CMOVcc cannot have an immediate operand as argument.
3431    
3432    ----------------------------------------------------------------------
3433    Name: Matthias Blume
3434    Date: 2002/03/07 16:15:00 EST
3435    Tag: blume-20020307-controls
3436    Description:
3437    
3438    This is a very large but mostly boring patch which makes (almost)
3439    every tuneable compiler knob (i.e., pretty much everything under
3440    Control.* plus a few other things) configurable via both the command
3441    line and environment variables in the style CM did its configuration
3442    until now.
3443    
3444    Try starting sml with '-h' (or, if you are brave, '-H')
3445    
3446    To this end, I added a structure Controls : CONTROLS to smlnj-lib.cm which
3447    implements the underlying generic mechanism.
3448    
3449    The interface to some of the existing such facilities has changed somewhat.
3450    For example, the MLRiscControl module now provides mkFoo instead of getFoo.
3451    (The getFoo interface is still there for backward-compatibility, but its
3452    use is deprecated.)
3453    
3454    The ml-build script passes -Cxxx=yyy command-line arguments through so
3455    that one can now twiddle the compiler settings when using this "batch"
3456    compiler.
3457    
3458    TODO items:
3459    
3460    We should go through and throw out all controls that are no longer
3461    connected to anything.  Moreover, we should go through and provide
3462    meaningful (and correct!) documentation strings for those controls
3463    that still are connected.
3464    
3465    Currently, multiple calls to Controls.new are accepted (only the first
3466    has any effect).  Eventually we should make sure that every control
3467    is being made (via Controls.new) exactly once.  Future access can then
3468    be done using Controls.acc.
3469    
3470    Finally, it would probably be a good idea to use the getter-setter
3471    interface to controls rather than ref cells.  For the time being, both
3472    styles are provided by the Controls module, but getter-setter pairs are
3473    better if thread-safety is of any concern because they can be wrapped.
3474    
3475    *****************************************
3476    
3477    One bug fix: The function blockPlacement in three of the MLRISC
3478    backpatch files used to be hard-wired to one of two possibilities at
3479    link time (according to the value of the placementFlag).  But (I
3480    think) it should rather sense the flag every time.
3481    
3482    *****************************************
3483    
3484    Other assorted changes (by other people who did not supply a HISTORY entry):
3485    
3486    1. the cross-module inliner now works much better (Monnier)
3487    2. representation of weights, frequencies, and probabilities in MLRISC
3488       changed in preparation of using those for weighted block placement
3489       (Reppy, George)
3490    
3491    ----------------------------------------------------------------------
3492    Name: Lal George
3493    Date: 2002/03/07 14:44:24 EST 2002
3494    Tag: george-20020307-weighted-block-placement
3495    
3496    Tested the weighted block placement optimization on all architectures
3497    (except the hppa) using AMPL to generate the block and edge frequencies.
3498    Changes were required in the machine properties to correctly
3499    categorize trap instructions. There is an MLRISC flag
3500    "weighted-block-placement" that can be used to enable weighted block
3501    placement, but this will be ineffective without block/edge
3502    frequencies (coming soon).
3503    
3504    
3505    ----------------------------------------------------------------------
3506    Name: Lal George
3507    Date: 2002/03/05 17:24:48 EST
3508    Tag: george-20020305-linkage-cluster
3509    
3510    In order to support the block placement optimization, a new cluster
3511    is generated as the very first cluster (called the linkage cluster).
3512    It contains a single jump to the 'real' entry point for the compilation
3513    unit. Block placement has no effect on the linkage cluster itself, but
3514    all the other clusters  have full freedom in the manner in which they
3515    reorder blocks or functions.
3516    
3517    On the x86 the typical linkage code that is generated is:
3518       ----------------------
3519            .align 2
3520       L0:
3521            addl    $L1-L0, 72(%esp)
3522            jmp     L1
3523    
3524    
3525            .align  2
3526       L1:
3527       ----------------------
3528    
3529    72(%esp) is the memory location for the stdlink register. This
3530    must contain the address of the CPS function being called. In the
3531    above example, it contains the address of  L0; before
3532    calling L1 (the real entry point for the compilation unit), it
3533    must contain the address for L1, and hence
3534    
3535            addl $L1-L0, 72(%esp)
3536    
3537    I have tested this on all architectures except the hppa.The increase
3538    in code size is of course negligible
3539    
3540    ----------------------------------------------------------------------
3541    Name: Allen Leung
3542    Date: 2002/03/03 13:20:00 EST
3543    Tag: leunga-20020303-mlrisc-tools
3544    
3545      Added #[ ... ] expressions to mlrisc tools
3546    
3547    ----------------------------------------------------------------------
3548    Name: Matthias Blume
3549    Date: 2002/02/27 12:29:00 EST
3550    Tag: blume-20020227-cdebug
3551    Description:
3552    
3553    - made types in structure C and C_Debug to be equal
3554    - got rid of code duplication (c-int.sml vs. c-int-debug.sml)
3555    - there no longer is a C_Int_Debug (C_Debug is directly derived from C)
3556    
3557    ----------------------------------------------------------------------
3558    Name: Matthias Blume
3559    Date: 2002/02/26 12:00:00 EST
3560    Tag: blume-20020226-ffi
3561    Description:
3562    
3563    1. Fixed a minor bug in CM's "noweb" tool:
3564       If numbering is turned off, then truly don't number (i.e., do not
3565       supply the -L option to noweb).  The previous behavior was to supply
3566       -L'' -- which caused noweb to use the "default" line numbering scheme.
3567       Thanks to Chris Richards for pointing this out (and supplying the fix).
3568    
3569    2. Once again, I reworked some aspects of the FFI:
3570    
3571       A. The incomplete/complete type business:
3572    
3573       - Signatures POINTER_TO_INCOMPLETE_TYPE and accompanying functors are
3574         gone!
3575       - ML types representing an incomplete type are now *equal* to
3576         ML types representing their corresponding complete types (just like
3577         in C).  This is still safe because ml-nlffigen will not generate
3578         RTTI for incomplete types, nor will it generate functions that
3579         require access to such RTTI.   But when ML code generated from both
3580         incomplete and complete versions of the C type meet, the ML types
3581         are trivially interoperable.
3582    
3583         NOTE:  These changes restore the full generality of the translation
3584         (which was previously lost when I eliminated functorization)!
3585    
3586       B. Enum types:
3587    
3588       - Structure C now has a type constructor "enum" that is similar to
3589         how the "su" constructor works.  However, "enum" is not a phantom
3590         type because each "T enum" has values (and is isomorphic to
3591         MLRep.Signed.int).
3592       - There are generic access operations for enum objects (using
3593         MLRep.Signed.int).
3594       - ml-nlffigen will generate a structure E_foo for each "enum foo".
3595         * The structure contains the definition of type "mlrep" (the ML-side
3596         representation type of the enum).  Normally, mlrep is the same
3597         as "MLRep.Signed.int", but if ml-nlffigen was invoked with "-ec",
3598         then mlrep will be defined as a datatype -- thus facilitating
3599         pattern matching on mlrep values.
3600         ("-ec" will be suppressed if there are duplicate values in an
3601          enumeration.)
3602         * Constructors ("-ec") or values (no "-ec") e_xxx of type mlrep
3603         will be generated for each C enum constant xxx.
3604         * Conversion functions m2i and i2m convert between mlrep and
3605         MLRep.Signed.int.  (Without "-ec", these functions are identities.)
3606         * Coversion functions c and ml convert between mlrep and "tag enum".
3607         * Access functions (get/set) fetch and store mlrep values.
3608       - By default (unless ml-nlffigen was invoked with "-nocollect"), unnamed
3609         enumerations are merged into one single enumeration represented by
3610         structure E_'.
3611    
3612    ----------------------------------------------------------------------
3613    Name: Allen Leung
3614    Date: 2002/02/25 04:45:00 EST
3615    Tag: leunga-20020225-cps-spill
3616    
3617    This is a new implementation of the CPS spill phase.
3618    The new phase is in the new file compiler/CodeGen/cpscompile/spill-new.sml
3619    In case of problems, replace it with the old file spill.sml
3620    
3621    The current compiler runs into some serious performance problems when
3622    constructing a large record.  This can happen when we try to compile a
3623    structure with many items.  Even a very simple structure like the following
3624    makes the compiler slow down.
3625    
3626        structure Foo = struct
3627           val x_1 = 0w1 : Word32.int
3628           val x_2 = 0w2 : Word32.int
3629           val x_3 = 0w3 : Word32.int
3630           ...
3631           val x_N = 0wN : Word32.int
3632        end
3633    
3634    The following table shows the compile time, from N=1000 to N=4000,
3635    with the old compiler:
3636    
3637    N
3638    1000   CPS 100 spill                           0.04u  0.00s  0.00g
3639           MLRISC ra                       &nb