Annotation of /sml/branches/primop-branch-3/NOTES/HISTORY

Parent Directory | Revision Log

---

Revision 2931 - (view) (download)

<

1 :	dbm	2431	This is the HISTORY file for the SML/NJ repository.
2 :	dbm	570
3 :			An entry should be made for _every_ commit to the repository.
4 :			The entries in this file will be used when creating the README
5 :			for new versions, so keep that in mind when writing the
6 :			description.
7 :
8 :			The form of an entry should be:
9 :
10 :	mblume	1931
11 :			Name: Your Name (web page or e-mail)
12 :	blume	2155	Date: yyyy/mm/dd hh:mm:ss
13 :	mblume	1931	Tag: cvs-tag-for-this-change
14 :	mblume	1480	Description:
15 :
16 :			----------------------------------------------------------------------
17 :	gkuan	2931	Name: Mike Rainey (http://www.cs.uchicago.edu/~mrainey)
18 :			Date: 2008/01/23
19 :			Description:
20 :
21 :			Fixed the amd64 code generator to compile with the current MLRISC.
22 :			----------------------------------------------------------------------
23 :	gkuan	2883	Name: Lars Bergstrom (larsberg@cs.uchicago.edu)
24 :	gkuan	2931	Date: 2007/11/26
25 :	gkuan	2883	Description:
26 :
27 :	gkuan	2931	nlffi was updated to work on Windows. It needed to pass in the correct
28 :			value for the name of the kernel32 DLL to obtain 'base' bindings.
29 :			Additionally updated the README for the most basic nlffi sample with
30 :			what you need to do on MacOSX and on Windows to make it work.
31 :			----------------------------------------------------------------------
32 :			Name: Lars Bergstrom (larsberg@cs.uchicago.edu)
33 :			Date: 2007/11/21
34 :			Description:
35 :
36 :			Overwrite the SMLNJ_HOME environment variable on installation
37 :			Properly change the package code so that subsequent version
38 :			installations prompt for uninstall (instead of 'repair/remove')
39 :			----------------------------------------------------------------------
40 :			Name: Lars Bergstrom (larsberg@cs.uchicago.edu)
41 :			Date: 2007/11/14
42 :			Description:
43 :
44 :	gkuan	2883	Cleaned up WININSTALL file for new MSI-based setup.
45 :			----------------------------------------------------------------------
46 :			Name: Matthias Blume (http://www.tti-c.org/blume/)
47 :			Date: 2007/11/13 14:05:00 CST
48 :			Description:
49 :
50 :			Version bumped to 110.67. New bootfiles.
51 :
52 :			----------------------------------------------------------------------
53 :			Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
54 :			Date: 2007/11/12
55 :			Description:
56 :
57 :			Ensure that the size of the allocation space is at least 128K.
58 :			----------------------------------------------------------------------
59 :			Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
60 :			Date: 2007/11/05
61 :			Description:
62 :
63 :			Fixed type error in ml-lpt library that occurs when compiling
64 :			against a basis that was compiled with the USE_64_BIT_POSITIONS
65 :			symbol set.
66 :			(Thanks to Johannes Joemann)
67 :			----------------------------------------------------------------------
68 :			Name: Mike Rainey (http://www.cs.uchicago.edu/~mrainey)
69 :			Date: 2007/11/03
70 :			Description:
71 :
72 :			Fixed the amd64 code generator to compile with the current MLRISC.
73 :			----------------------------------------------------------------------
74 :			Name: Lars Bergstrom (web page or e-mail)
75 :			Date: 2007/11/02
76 :			Description:
77 :
78 :			Made sml.bat more resilient to either not having run the installer
79 :			and having no SMLNJ_HOME set or having just shuffled the directory
80 :			around.
81 :			----------------------------------------------------------------------
82 :			Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
83 :			Date: 2007/11/01
84 :			Description:
85 :
86 :			A collection of bug fixes for machine.sml in the Reactive library.
87 :			(Thanks to Timothy Bourke)
88 :			----------------------------------------------------------------------
89 :			Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
90 :			Date: 2007/10/28
91 :			Description:
92 :
93 :			Patches for Mac OS X 10.5 (Leopard).
94 :			----------------------------------------------------------------------
95 :			Name: Mike Rainey (http://www.cs.uchicago.edu/~mrainey)
96 :			Date: 2007/10/28
97 :			Description:
98 :
99 :			Fixed some bugs in the AMD64 floating-point spilling code.
100 :			----------------------------------------------------------------------
101 :			Name: Mike Rainey (http://www.cs.uchicago.edu/~mrainey)
102 :			Date: 2007/10/25
103 :			Description:
104 :
105 :			Added support for the atomic XCHG instruction.
106 :			----------------------------------------------------------------------
107 :			Name: Mike Rainey (http://www.cs.uchicago.edu/~mrainey)
108 :			Date: 2007/10/22
109 :			Description:
110 :
111 :			Added AMD64 support for floating-point negation.
112 :			----------------------------------------------------------------------
113 :			Name: Lars Bergstrom (web page or e-mail)
114 :			Date: 2007/10/22
115 :			Description:
116 :
117 :			Fixed ^C handling in Windows
118 :			Added a Windows installer
119 :			Made it possible to build for Windows on a mapped drive from Parallels
120 :			----------------------------------------------------------------------
121 :			Name: Mike Rainey (http://www.cs.uchicago.edu/~mrainey)
122 :			Date: 2007/10/22
123 :			Description:
124 :
125 :			Added AMD64 support for the atomic fetch and add instruction.
126 :			----------------------------------------------------------------------
127 :			Name: Mike Rainey (http://www.cs.uchicago.edu/~mrainey)
128 :			Date: 2007/10/18
129 :			Description:
130 :
131 :			The GAS output now favors p2align over align, since the
132 :			former is guaranteed to be consistent over multiple architectures
133 :			and the latter is not.
134 :			----------------------------------------------------------------------
135 :			Name: Mike Rainey (http://www.cs.uchicago.edu/~mrainey)
136 :			Date: 2007/10/18
137 :			Description:
138 :
139 :			Fixed a bug in register spilling.
140 :			----------------------------------------------------------------------
141 :			Name: Mike Rainey (http://www.cs.uchicago.edu/~mrainey)
142 :			Date: 2007/10/2
143 :			Description:
144 :
145 :			Added demo support for AMD64 for quick testing and fixed support for
146 :			64-bit label constants.
147 :			----------------------------------------------------------------------
148 :			Name: Mike Rainey (http://www.cs.uchicago.edu/~mrainey)
149 :			Date: 2007/09/20
150 :			Description:
151 :
152 :			Added support for the MLRISC COND instruction and the x86-64 CMOVCC
153 :			instruction.
154 :			----------------------------------------------------------------------
155 :			Name: Mike Rainey (http://www.cs.uchicago.edu/~mrainey)
156 :			Date: 2007/09/17
157 :			Description:
158 :
159 :			Fixed an instruction-selection bug when loading 64-bit labels.
160 :
161 :			----------------------------------------------------------------------
162 :	gkuan	2742	Name: George Kuan (http://www.cs.uchicago.edu/~gkuan)
163 :	gkuan	2883	Date: 2007/09/14
164 :			Description:
165 :
166 :			Fixed Elaborator and Translate performance bugs
167 :
168 :			----------------------------------------------------------------------
169 :			Name: Mike Rainey (http://www.cs.uchicago.edu/~mrainey)
170 :			Date: 2007/09/12
171 :			Description:
172 :
173 :			Fixed xorl memory argument bug.
174 :
175 :			----------------------------------------------------------------------
176 :			Name: George Kuan (http://www.cs.uchicago.edu/~gkuan)
177 :	gkuan	2742	Date: 2007/07/27
178 :			Description:
179 :
180 :			Added f64sgn (for Real64.signBit) as a primop defined
181 :			in MLRiscGen. signBit(~0.0) not handled correctly.
182 :
183 :			----------------------------------------------------------------------
184 :			Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
185 :			Date: 2007/06/21
186 :			Description:
187 :
188 :			Fixed bug in {TextIO,BinIO}.StreamIO.endOfStream that would incorrectly
189 :			signal end of stream.
190 :
191 :			----------------------------------------------------------------------
192 :			Name: Matthias Blume (http://www.tti-c.org/blume/)
193 :			Date: 2007/06/12 11:22:00 CDT
194 :			Description:
195 :
196 :			Eliminated config/allsources. The information is now drawn directly
197 :			from config/actions.
198 :
199 :			----------------------------------------------------------------------
200 :			Name: Matthias Blume (http://www.tti-c.org/blume/)
201 :			Date: 2007/06/07 23:30:00 CDT
202 :			Tag: release-110.65
203 :			Description:
204 :
205 :			Version 110.65. NEW BOOTFILES!
206 :
207 :			----------------------------------------------------------------------
208 :			Name: Matthias Blume (http://www.tti-c.org/blume/)
209 :			Date: 2007/06/06 21:45:00 CDT
210 :			Description:
211 :
212 :			- Aaron: fixed a number of bugs in ml-ulex.
213 :			- Matthias:
214 :			- added CM control "cm.force-tools"; this is false by default;
215 :			when set to true, then tools like ml-yacc, ml-lex, ml-ulex, etc.
216 :			will be forced to run regardless of whether or not their targets
217 :			are up-to-date
218 :			- changed installer code so that config/install.sh will re-build
219 :			heap images for all tools even if those heap images already existed
220 :
221 :			----------------------------------------------------------------------
222 :			Name: Matthias Blume (http://www.tti-c.org/blume/)
223 :			Date: 2007/06/04 21:21:00 CDT
224 :			Description:
225 :
226 :			After Aaron Turon's bug fix for ml-ulex (handling the ^ character
227 :			in legacy mode), re-ran the lexer generator on all lex input files
228 :			and committed the results.
229 :
230 :			This should fix the problem with ckit and nlffi that was reported
231 :			by Vesa A. Norrman.
232 :
233 :			----------------------------------------------------------------------
234 :			Name: Matthias Blume (http://www.tti-c.org/blume/)
235 :			Date: 2007/05/31 11:40:00 CDT
236 :			Tag: release-110.64
237 :			Description:
238 :
239 :			Version 110.64. NEW BOOTFILES!
240 :
241 :			----------------------------------------------------------------------
242 :	dbm	2649	Name: Dave MacQueen (dbm@cs.uchicago.edu)
243 :	dbm	2737	Date: 2007/05/31 14:55:00 CDT
244 :			Revision: 2650
245 :	dbm	2649	Description:
246 :
247 :			3rd merge of base from primop-branch-3 into the trunk.
248 :			Additional bug fixes included:
249 :
250 :			* Timer.cpu_timer, etc. type printing corrected (by making
251 :			Timer have opaque sig constraint in basis/Implementation/timer.sml)
252 :			* Infinite loop in FLINT (tests/typing/tests/25.sml) (fix by
253 :			Stefan Monnier)
254 :
255 :			----------------------------------------------------------------------
256 :	dbm	2611	Name: Matthias Blume (http://www.tti-c.org/blume/)
257 :	dbm	2648	Date: 2007/05/31 11:40:00 CDT
258 :			Description:
259 :
260 :			Fixed some bugs in new Div code in FLINT/trans/translate.sml.
261 :
262 :			Also, changed the handling of "no core access": When translate.sml
263 :			needs access to a core exception at a time when the core has not
264 :			been set up yet (this only happens when compiling system/smlnj/init/*),
265 :			then don't bother generating the corresponding tests.
266 :
267 :	dbm	2649	The old scheme was to generate a bogus value to be used in place of
268 :			the exception. Unfortunately, that confuses the plambda type checker.
269 :			Moreover, it does not do any good, because at runtime we don't expect
270 :			such an exception to be ever raised. (The code in system/smlnj/init/*
271 :			has to be written very carefully with this in mind!)
272 :	dbm	2648
273 :			----------------------------------------------------------------------
274 :			Name: Mike Rainey (http://www.cs.uchicago.edu/~mrainey)
275 :			Date: 2007/05/29
276 :			Description:
277 :
278 :			Added FSQRT instructions for the AMD64 code generator.
279 :
280 :			----------------------------------------------------------------------
281 :			Name: Matthias Blume (http://www.tti-c.org/blume/)
282 :	dbm	2636	Date: 2007/05/29 19:00:00 CDT
283 :			Description:
284 :
285 :			FLINT/trans/translate.sml now wraps all DIV/MOD/QUOT/REM operations
286 :			with an explicit test for zero division. This should fixes several
287 :			regressions and makes it possible for downstream optimization phases
288 :			to treat these operations as "pure" when they are applied to unsigned
289 :			operands.
290 :
291 :			----------------------------------------------------------------------
292 :			Name: Mike Rainey (http://www.cs.uchicago.edu/~mrainey)
293 :			Date: 2007/05/29
294 :			Description:
295 :
296 :			Added the new MLRISC code generator for the AMD64. This version,
297 :			in contrast to the previous one, uses SSE registers and instructions
298 :			for all floating-point computations.
299 :
300 :			----------------------------------------------------------------------
301 :			Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
302 :			Date: 2007/05/29
303 :			Description:
304 :
305 :			A number of fixes related to the formatting of dates. These include fixes
306 :			for bugs #1415 and #1416. We also now correctly handle format characters that
307 :			lie outside the specified set.
308 :
309 :			----------------------------------------------------------------------
310 :			Name: Matthias Blume (http://www.tti-c.org/blume/)
311 :	dbm	2611	Date: 2007/05/23 22:30:00 CDT
312 :			Description:
313 :
314 :			CMB (and CM) now automatically defines the CM "preprocessor" symbol
315 :			NO_PLUGINS during "makeml -rebuild" or when CM operates in "slave" mode.
316 :
317 :			(In addition, CMB_REBUILD_MODE is defined for makeml -rebuild,
318 :			CM_SLAVE_MODE is defined in attached slaves while running CM.make
319 :			or CM.rebuild, and CMB_SLAVE_MODE is defined in attached slaves
320 :			while running CMB.make. The point is that the single symbol
321 :			NO_PLUGINS is definde in all three cases.)
322 :
323 :			I changed the three locations within the sources that get compiled
324 :			during CMB.make where ml-yacc or ml-lex input is processed: When
325 :			NO_PLUGINS defined, then the use of the mlyacc and mllex tools
326 :			is bypassed.
327 :
328 :			When bootstrapping new versions of the compiler, there can be
329 :			situations where the plugin tools for ml-yacc and ml-lex (or ml-ulex)
330 :			are not available or otherwise not operational. In this case one can
331 :			manually define the NO_PLUGINS symbol prior to running CMB.make(). To
332 :			do so, the following command should be issued at the interactive
333 :			prompt:
334 :
335 :			#set (CMB.symval "NO_PLUGINS") (SOME 1);
336 :
337 :			A CMB.make with NO_PLUGINS defined relies on the existence of the
338 :			files that normally would be generated by ml-yacc and ml-lex.
339 :			(Copies of these files are in the repository.)
340 :
341 :			----------------------------------------------------------------------
342 :	dbm	2600	Name: David MacQueen (dbm@cs.uchicago.edu)
343 :			Date: 2007/5/23 17:23:00 CDT
344 :	dbm	2636	Revision: 2603
345 :	dbm	2600	Description:
346 :
347 :	dbm	2636	2nd merge of base from primop-branch-3 into the trunk.
348 :			Additional bug fixes included:
349 :	dbm	2600	Date.scan and Date.fromString fixed;
350 :			Overloading resolution fixed and some type printing problems corrected.
351 :
352 :			----------------------------------------------------------------------
353 :	dbm	2579	Name: Matthias Blume (http://www.tti-c.org/blume/)
354 :	dbm	2595	Date: 2007/05/23 09:50:00 CDT
355 :			Description:
356 :
357 :			Changed the installation mechanism for CM tool plugins. These are
358 :			just libaries and now get installed like ordinary libaries.
359 :
360 :			There are now a number of new installation targets that give some
361 :			fine-grain control over what classes and suffixes are known, and what
362 :			they will map to.
363 :			See config/targets for details.
364 :
365 :			The code that caused plugin installation as part of running a
366 :			tool's "build" script has been removed. (The build script is for
367 :			building, not for installing.)
368 :
369 :			----------------------------------------------------------------------
370 :			Name: Matthias Blume (http://www.tti-c.org/blume/)
371 :	dbm	2579	Date: 2007/05/22 11:50:00 CDT
372 :			Description:
373 :
374 :			Added a boolean control named cm.tolerate-tool-failures (env. variable
375 :			name: CM_TOLERATE_TOOL_FAILURES). The default is false and makes
376 :			CM fail if a shell tool reports a non-success exit status. If
377 :			the control is set to true, then CM will press on after tool failures
378 :			in the event that the target files exist (even though they are
379 :			considered outdated). Turning the control to true can be useful for
380 :			bootstrapping.
381 :
382 :			----------------------------------------------------------------------
383 :			Name: Aaron Turon (http://www.cs.uchicago.edu/~adrassi)
384 :			Date: 2007/05/19
385 :			Description:
386 :
387 :			Merge of ml-lpt revisions for 110.64.
388 :
389 :			The name of several ml-lpt-lib modules has changed:
390 :			Repair => AntlrRepair
391 :			StreamPos => AntlrStreamPos
392 :			ErrHandlerFn => AntlrErrHandler
393 :			EBNF => AntlrEBNF
394 :
395 :			The ml-antlr specification format has changed: declarations such as
396 :			%tokens and nonterminal definitions can occur *multiple* times in the
397 :			same specification. The semantics are such that each new declaration
398 :			extends the previous ones. This does not apply to %start or %name,
399 :			of course.
400 :
401 :			Importing a grammar via %import now includes *all* declarations in from
402 :			the imported grammar, except for %name, %entry, and %start. Tokens
403 :			and nonterminals can be dropped using the new %dropping clause of the
404 :			%import directive; the separate %drop and %extend have been removed.
405 :
406 :			We now allow optional type annotations on nonterminals, using the %nonterms
407 :			directive as in ml-yacc.
408 :
409 :			The refcell construct is now implemented using SML's regular reference
410 :			cells, so the :== and !! notation has been deprecated.
411 :
412 :			The ml-antlr tool now does much more checking of specifications, and
413 :			its error messages have been greatly improved. Error repair for
414 :			generated parsers has been completely rewritten, and is now both
415 :			much faster and more accurate.
416 :
417 :			ml-ulex is now more lenient with escape codes (non-SML-standard
418 :			escape codes are now interpreted literally, so e.g. \| denotes "|").
419 :			Also, character classes may now include a "-" character at the beginning
420 :			as is standard in most other regexp tools.
421 :
422 :			All of these changes are documented in the user guide, which has
423 :			been updated and improved with this merge.
424 :
425 :			----------------------------------------------------------------------
426 :	dbm	2524	Name: David MacQueen (http://www.cs.uchicago.edu/~jhr)
427 :			Date: 2007/05/03
428 :			Revision: r2492
429 :			Description:
430 :
431 :			Merge of the primop3 branch (base) into the trunk to create 110.63.1.
432 :			Significant changes in FLINT and the front end, mostly having to do
433 :			with a reorganized system for handling primops. Various bug fixes
434 :			and improvements in printing signatures.
435 :
436 :			----------------------------------------------------------------------
437 :	dbm	2579	Name: Aaron Turon (http://www.cs.uchicago.edu/~adrassi)
438 :			Date: 2007/05/02
439 :			Description:
440 :
441 :			Preliminary commit of large ml-lpt revisions (more to come for 110.64).
442 :			The ml-antlr specification format has changed: declarations such as
443 :			%tokens and nonterminal definitions can occur *multiple* times in the
444 :			same specification. The semantics are such that each new declaration
445 :			extends the previous ones. Grammar extension constructs have also
446 :			changed. We now allow type annotations on nonterminals. Finally,
447 :			the refcell construct is now implemented using SML's regular reference
448 :			cells, so the :== and !! notation has been deprecated. All of these
449 :			changes will appear in the 110.64 user guide for ml-lpt.
450 :
451 :			The ml-antlr tool now does much more checking of specifications, and
452 :			its error messages have been greatly improved. There has also been
453 :			some work on the error repair process for generated parsers, but this
454 :			will be further improved in 110.64.
455 :
456 :			----------------------------------------------------------------------
457 :	dbm	2636	Name: David MacQueen (http://www.cs.uchicago.edu/~jhr)
458 :			Date: 2007/05/03
459 :			Revision: r2492
460 :			Description:
461 :
462 :			Merge of the primop3 branch (base) into the trunk to create 110.63.1.
463 :			Significant changes in FLINT and the front end, mostly having to do
464 :			with a reorganized system for handling primops. Various bug fixes
465 :			and improvements in printing signatures.
466 :
467 :			----------------------------------------------------------------------
468 :			Name: Aaron Turon (http://www.cs.uchicago.edu/~adrassi)
469 :			Date: 2007/05/02
470 :			Description:
471 :
472 :			Preliminary commit of large ml-lpt revisions (more to come for 110.64).
473 :			The ml-antlr specification format has changed: declarations such as
474 :			%tokens and nonterminal definitions can occur *multiple* times in the
475 :			same specification. The semantics are such that each new declaration
476 :			extends the previous ones. Grammar extension constructs have also
477 :			changed. We now allow type annotations on nonterminals. Finally,
478 :			the refcell construct is now implemented using SML's regular reference
479 :			cells, so the :== and !! notation has been deprecated. All of these
480 :			changes will appear in the 110.64 user guide for ml-lpt.
481 :
482 :			The ml-antlr tool now does much more checking of specifications, and
483 :			its error messages have been greatly improved. There has also been
484 :			some work on the error repair process for generated parsers, but this
485 :			will be further improved in 110.64.
486 :
487 :			----------------------------------------------------------------------
488 :	dbm	2464	Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
489 :			Date: 2007/04/24
490 :			Description:
491 :
492 :			More Basis fixes: The Char.fromString (etc.) functions did not handle
493 :			the "\uxxxx" escape sequence. There is still an outstanding bug with
494 :			String.fromString the tail is a format escape. I added a comment to
495 :			this effect in Basis/Implementation/string.sml.
496 :			Thanks to Andreas Rossberg.
497 :
498 :			----------------------------------------------------------------------
499 :			Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
500 :			Date: 2007/04/24
501 :			Description:
502 :
503 :			Added next function to Fifo and Queue modules in the SML/NJ Library.
504 :
505 :			----------------------------------------------------------------------
506 :			Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
507 :			Date: 2007/04/23
508 :			Description:
509 :
510 :			More Basis fixes: Time.fmt dropped the leading "~" for negative time
511 :			values that had no whole part.
512 :
513 :			----------------------------------------------------------------------
514 :			Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
515 :			Date: 2007/04/23
516 :			Description:
517 :
518 :			More Basis fixes: the OS.Path module did not include the InvalidArc
519 :			exception and did not do sufficient argument checking.
520 :			Thanks to Stephen Weeks and Adam Chilpala.
521 :
522 :			----------------------------------------------------------------------
523 :	dbm	2431	Name: Mike Rainey (http://www.cs.uchicago.edu/~mrainey)
524 :			Date: 2007/04/16
525 :			Description:
526 :
527 :			Fixed an unbound functor bug in the AMD64 CM file.
528 :
529 :			----------------------------------------------------------------------
530 :			Name: Mike Rainey (http://www.cs.uchicago.edu/~mrainey)
531 :			Date: 2007/04/12
532 :			Description:
533 :
534 :			Substantially changed the signature and implementation of AMD64
535 :			SVID. It now looks similar to the ia32 SVID, yet uses staged
536 :			allocation.
537 :
538 :			----------------------------------------------------------------------
539 :			Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
540 :			Date: 2007/04/07
541 :			Description:
542 :
543 :			More Basis fixes: the WORD signature was missing {to,from}Large.
544 :			Thanks to Andreas Rossberg.
545 :
546 :			----------------------------------------------------------------------
547 :	mblume	1923	Name: Matthias Blume (http://www.tti-c.org/blume/)
548 :	dbm	2431	Date: 2007/03/22 16:07:00 CDT
549 :			Tag: release-110.63
550 :			Description:
551 :
552 :			Bumped version number to 110.63.
553 :
554 :			!!!NEW BOOTFILES!!!
555 :
556 :			----------------------------------------------------------------------
557 :			Name: Mike Rainey (http://www.cs.uchicago.edu/~mrainey)
558 :			Date: 2007/03/19
559 :			Description:
560 :
561 :			Fixed bogus operand sizes in AMD64 instruction spilling.
562 :
563 :			----------------------------------------------------------------------
564 :			Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
565 :			Date: 2007/03/19
566 :			Description:
567 :
568 :			Fixed a number of inconsistencies between the Posix.TTY structure
569 :			and the Basis specification. Thanks to Adam Chilpala.
570 :
571 :			----------------------------------------------------------------------
572 :			Name: Mike Rainey (http://www.cs.uchicago.edu/~mrainey)
573 :			Date: 2007/02/26
574 :			Description:
575 :
576 :			Added preliminary support in MLRISC for Staged Allocation, a technique
577 :			for specifying calling conventions (see
578 :			http://www.eecs.harvard.edu/~nr/pubs/staged-abstract.html).
579 :			Initially, we plan to use this code to generate C calls for the AMD64.
580 :
581 :			The staged allocation code base resides in MLRISC/staged-allocation,
582 :			and specialized calling conventions go in MLRISC/ARCH/staged-allocation.
583 :
584 :			----------------------------------------------------------------------
585 :			Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
586 :			Date: 2007/02/20
587 :			Description:
588 :
589 :			Bug fix: when SaveCState was called with two values to save, a subsequent
590 :			GC could cause the RestoreCState to fail because the saved state had been
591 :			promoted to tagless pair.
592 :
593 :			----------------------------------------------------------------------
594 :			Name: Allen Leung (leunga (at) reservoir (dot) com)
595 :			Date: 2007/02/20 19:08:00 EST
596 :			Description:
597 :
598 :			Bug triggered by:
599 :
600 :			val a1 = Word8Array.array(a1, 0w0);
601 :			val _ = Word8Array.update(a1, 0, 0w128);
602 :
603 :			x86MCEmitter crashed when the immediate operand to MOVB was outside
604 :			of the range -128 ... 127. I've changed the code so that the range
605 :			check is disabled. Only the low order 8 bits of the immediate operand
606 :			are now significant.
607 :
608 :
609 :			----------------------------------------------------------------------
610 :			Name: Matthias Blume (http://www.tti-c.org/blume/)
611 :			Date: 2007/02/15 09:20:00 CST
612 :			Description:
613 :
614 :			Eliminated any mention of lexgen. (Lexgen was an early precursor
615 :			to ml-ulex.)
616 :
617 :			You should update your admin directory, so the shell scripts for
618 :			maintaining your local copy of the repository reflect this change.
619 :
620 :			----------------------------------------------------------------------
621 :			Name: Matthias Blume (http://www.tti-c.org/blume/)
622 :			Date: 2007/02/14 15:45:00 CST
623 :			Description:
624 :
625 :			Fixed bug in CM's parallel make facility that failed to have
626 :			the master re-link modules after letting slaves compile them.
627 :
628 :			----------------------------------------------------------------------
629 :			Name: Matthias Blume (http://www.tti-c.org/blume/)
630 :			Date: 2007/02/12 15:10:00 CST
631 :			Description:
632 :
633 :			Fixed typo in ml-build script that prevented library anchors
634 :			from being registered.
635 :
636 :			----------------------------------------------------------------------
637 :			Name: Matthias Blume (http://www.tti-c.org/blume/)
638 :			Date: 2007/02/02 15:20:00 CST
639 :			Tag: release-110.62
640 :			Description:
641 :
642 :			New working release (110.62). NEW BOOTFILES!
643 :
644 :			----------------------------------------------------------------------
645 :			Name: Aaron Turon (http://www.cs.uchicago.edu/~adrassi)
646 :			Date: 2007/01/31
647 :			Description:
648 :
649 :			Brought ml-lpt manual up-to-date with the code. Many minor
650 :			improvements to the tools, and a few minor bugfixes. Prepared
651 :			ml-lpt-lib for integration with new UTF8 structure. Changed
652 :			the interface for creating streams in ml-ulex (we now support
653 :			stream creation from several kinds of sources). Overall, the
654 :			tools are now quite stable, and their interfaces are unlikely to
655 :			change in a way that would break compatibility. More work,
656 :			however, is needed in the documentation.
657 :
658 :			----------------------------------------------------------------------
659 :			Name: Matthias Blume (http://www.tti-c.org/blume/)
660 :			Date: 2007/02/01 10:05:00 CST
661 :			Description:
662 :
663 :			Implemented library installer. Moved CM plugin code for
664 :			ml-burg, ml-lex, and ml-yacc out of CM source tree and into
665 :			their respective trees. Implemented CM plugin for ml-ulex
666 :			and ml-antlr. Used library installer for ml-burg, ml-ulex,
667 :			and ml-antlr.
668 :
669 :			For ml-yacc and ml-lex we continue to have permanently "plugged-in"
670 :			CM tools. (It turns out to be too messy to do otherwise because
671 :			there is too much code that during installation relies on the presence
672 :			of these tools -- resulting in a tricky ordering problem.)
673 :
674 :			----------------------------------------------------------------------
675 :			Name: Matthias Blume (http://www.tti-c.org/blume/)
676 :			Date: 2007/01/30 11:05:00 CST
677 :			Description:
678 :
679 :			Added SMLofNJ.shiftArgs which is like a shell's "shift" command.
680 :			Modified CM's startup code to use shiftArgs as it processes
681 :			command line arguments. This way, the init code in each .sml-file
682 :			or library that is mentioned at the top level will see only
683 :			those arguments that have not yet been processed at this point.
684 :			In other words, the init code can "seize control" and process the
685 :			remaining command line.
686 :
687 :			----------------------------------------------------------------------
688 :			Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
689 :			Date: 2007/01/30
690 :			Description:
691 :
692 :			Added fromList function to the ORD_SET interface and lookup to the
693 :			ORD_MAP interface. See the SML/NJ Library CHANGES file for details.
694 :
695 :			----------------------------------------------------------------------
696 :			Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
697 :			Date: 2007/01/28
698 :			Description:
699 :
700 :			Added the UTF8 structure and signature from the Moby compiler to
701 :			the SML/NJ library (and the CML library). These modules will replace
702 :			the version in the ml-lpt-lib.
703 :
704 :			----------------------------------------------------------------------
705 :			Name: Matthias Blume (http://www.tti-c.org/blume/)
706 :			Date: 2007/01/26 15:10:00 CST
707 :			Description:
708 :
709 :			Added entries to handle ml-lpt-lib.cm in installer.
710 :
711 :			----------------------------------------------------------------------
712 :			Name: Matthias Blume (http://www.tti-c.org/blume/)
713 :			Date: 2006/12/15 11:10:00 CST
714 :			Tag: release-110.61
715 :			Description:
716 :
717 :			New working release (110.61). NEW BOOTFILES!
718 :
719 :			This supersedes the pre-brown-paper-bug release (see below).
720 :
721 :			----------------------------------------------------------------------
722 :			Name: Matthias Blume (http://www.tti-c.org/blume/)
723 :			Date: 2006/12/15 11:00:00 CST
724 :			Description:
725 :
726 :			Fixed brown-paper-bag bug with CM's pathname handling, which made
727 :			installation fail under Win32.
728 :
729 :			----------------------------------------------------------------------
730 :			Name: Matthias Blume (http://www.tti-c.org/blume/)
731 :			Date: 2006/12/14 12:00:00 CST
732 :			Description:
733 :
734 :			Obsolete!
735 :
736 :			[New working release (110.61). NEW BOOTFILES!]
737 :
738 :			----------------------------------------------------------------------
739 :			Name: John Reppy (http://cs.uchicago.edu/~jhr)
740 :			Date: 2006/14/06 09:50:00 CST
741 :			Description:
742 :
743 :			Fixed the code in runtime/c-libs/posix-tty/{tcgetattr,tcsetattr}.c to
744 :			get the c_cc termios data copied correctly. Also moved the allocation
745 :			of the string to avoid problems if it caused a GC.
746 :
747 :			Thanks to Timothy Bourke for the bug report and fix.
748 :
749 :			----------------------------------------------------------------------
750 :			Name: Matthias Blume (http://www.tti-c.org/blume)
751 :			Date: 2006/12/08 09:50:00 CST
752 :			Description:
753 :
754 :			Added code to CM's "standard shell tool" implementation which
755 :			causes it to tolerate (with a warning) the situation where the shell
756 :			command fails (e.g., due to the shell command's non-existence)
757 :			as long as all target files exist.
758 :
759 :			This makes it possible to, e.g., build ml-yacc from sources even if
760 :			svn checkout messed up the time stamps on files in such a way that
761 :			yacc.grm is younger than yacc.grm.sml or yacc.grm.sig. (Ml-yacc would
762 :			be needed to re-process yacc.grm, but obviously it might not yet be
763 :			available at that time.)
764 :
765 :			----------------------------------------------------------------------
766 :			Name: John Reppy (http://cs.uchicago.edu/~jhr)
767 :			Date: 2006/12/06 12:06:00 CST
768 :			Description:
769 :
770 :			Fixed the types of recvVecFrom, recvVecFrom', recvVecFromNB, and
771 :			recvVecFromNB' in the SOCKET signature. This error is actually
772 :			in the SML Basis specification too.
773 :
774 :			----------------------------------------------------------------------
775 :			Name: Matthias Blume (http://www.tti-c.org/blume/)
776 :			Date: 2006/12/05 22:40:00 CST
777 :			Description:
778 :
779 :			CM now reports undefined anchors as errors and aborts execution
780 :			rather than silently pressing on using bogus values.
781 :
782 :			----------------------------------------------------------------------
783 :			Name: Matthias Blume (http://www.tti-c.org/blume/)
784 :			Date: 2006/11/29 10:00:00 CST
785 :			Description:
786 :
787 :			Use Say.vsay for printing the "[autoloading]" message, so
788 :			#set CM.Control.verbose false (or -Ccm.verbose=false) can be used
789 :			to suppress them.
790 :
791 :			----------------------------------------------------------------------
792 :			Name: Matthias Blume (http://www.tti-c.org/blume/)
793 :	blume	2219	Date: 2006/11/10 11:00:00 CST
794 :			Description:
795 :
796 :			Fixed bug in CM where "with:" specifications that affect compilation
797 :			(as opposed to parsing) were ignored.
798 :
799 :			----------------------------------------------------------------------
800 :			Name: Matthias Blume (http://www.tti-c.org/blume/)
801 :	blume	2193	Date: 2006/11/09 14:30:00 CST
802 :			Tag: release-110.60
803 :			Description:
804 :
805 :			New working release (110.60). NEW BOOTFILES!
806 :
807 :			This is the first subversion-hosted release. There are also changes
808 :			to the directory layout. Here is a table:
809 :
810 :			(All pathnames are relative to the SML/NJ "root" directory.)
811 :
812 :			config -> config
813 :
814 :			src/cm -> base/cm
815 :			src/compiler -> base/compiler
816 :			src/runtime -> base/runtime
817 :			src/system -> base/system
818 :
819 :			src/cm/pgraph -> pgraph
820 :
821 :			src/READMES -> base/READMES
822 :			src/$notefile -> base/NOTES/$notefile
823 :			for <notefile> in BOOT CVSNOTES CYGWININSTALL HISTORY INSTALL
824 :			MACOSXINSTALL WININSTALL
825 :
826 :			src/smlnj-lib -> smlnj-lib
827 :			src/MLRISC -> MLRISC
828 :			ckit -> ckit
829 :			src/cml -> cml
830 :			src/eXene -> eXene
831 :			src/heap2asm -> heap2asm
832 :			src/lexgen -> lexgen
833 :			src/ml-burg -> ml-burg
834 :			src/ml-lex -> ml-lex
835 :			src/ml-yacc -> ml-yacc
836 :			src/ml-nlffi-lib -> nlffi/lib
837 :			src/ml-nlffigen -> nlffi/gen
838 :			src/smlnj-c -> smlnj-c
839 :			src/tools/TraceDebugProf -> trace-debug-profile
840 :
841 :			In addition, there is also a new
842 :
843 :			ml-lpt
844 :
845 :			directory containing two new program generator tools: ml-ulex and
846 :			ml-antlr (a lexer generator that handles unicode and an ANTRL-inspired
847 :			LL(k) parser generator). These tools are currently "beta-quality"
848 :
849 :			The latest versios of the sources can now be obtained anonymously
850 :			via subversion. For this, it is useful to first check out
851 :
852 :			svn://smlnj-gforge.cs.uchicago.edu/smlnj/admin
853 :
854 :			and put the resulting directory on your shell's PATH. This provides
855 :			access to three shell scripts: checkout-all.sh, stat-all.sh,
856 :			and refresh-all.sh.
857 :
858 :			To create a freshly checked-out copy of the sources, do
859 :
860 :			checkout-all.sh [dir]
861 :
862 :			where dir is the optional SML/NJ root directory (default is ".").
863 :
864 :			This creates the above directory layout. Each subdirectory of
865 :			the root is under individual subversion control. The stat-all.sh
866 :			and refresh-al.sh scripts apply "svn stat" or "svn update" to
867 :			each of these subtrees.
868 :
869 :			----------------------------------------------------------------------
870 :			Name: Matthias Blume (http://www.tti-c.org/blume/)
871 :	blume	2155	Date: 2006/11/02 12:20:00 CST
872 :			Description:
873 :
874 :			Reorganized directory layout.
875 :
876 :			This is a temporary solution, more reorganization is to come.
877 :
878 :			The basic idea is to have a number of toplevel trees, each corresponding
879 :			to a well-defined part of the overall system. Each part can be
880 :			maintained individually, even in separate source repositories,
881 :			although currently we still serve everything out of the main
882 :			smlnj-gforge tree.
883 :
884 :			Installer and and scripts have been updated to reflect the new
885 :			layout. The installer (base/src/system/smlnj/installer) is now
886 :			"scriptable" to avoid burning too much knowledge about the
887 :			layout into SML source code. The main script used by the installer
888 :			is in config/actions.
889 :
890 :			The main change is that many of the subdirectories of what used to
891 :			be known as "src" have moved to the toplevel. The "src" directory
892 :			itself has moved down into a subtree called "base". (We may eventually
893 :			get rid of the extra level of indirection represented by "src".)
894 :
895 :			The layout is now as follows:
896 :
897 :			toplevel tree name | default repository
898 :			| (using gf=svn://smlnj-gforge.cs.uchicago.edu/smlnj)
899 :			-------------------+----------------------------------------------------
900 :			config | $gf/config/trunk
901 :			|
902 :			base | $gf/sml/trunk
903 :			|
904 :			smlnj-lib | $gf/smlnj-lib/trunk
905 :			MLRISC | $gf/MLRISC/trunk
906 :			|
907 :			ml-yacc | $gf/ml-yacc/trunk
908 :			ml-lex | $gf/ml-lex/trunk
909 :			ml-burg | $gf/ml-burg/trunk
910 :			lexgen | $gf/lexgen/trunk
911 :			heap2asm | $gf/heap2asm/trunk
912 :			|
913 :			cml | $gf/cml/trunk
914 :			eXene | $gf/eXene/trunk
915 :			ckit | $gf/ckit/trunk
916 :			nlffi | $gf/nlffi/trunk
917 :			|
918 :			smlnj-c | $gf/smlnj-c/trunk
919 :
920 :			In $gf/admin there are a few useful shell scripts for checking out
921 :			and maintaining the entire collection of trees:
922 :
923 :			admin/checkout-all.sh [dir]
924 :			-- optionally creates dir and checks out all trees from their
925 :			default repositories; if dir is missing, checkout into .
926 :			admin/refresh-all.sh [dir]
927 :			-- looks at all trees (from the above list) in dir (default: .)
928 :			and runs "svn update" if the tree exists and is under
929 :			subversion control;
930 :			non-existing or non-subversion trees are skipped
931 :			admin/stat-all.sh [dir]
932 :			-- like refresh-all.sh, but runs "svn stat" instead
933 :
934 :			----------------------------------------------------------------------
935 :			Name: Matthias Blume (http://www.tti-c.org/blume/)
936 :			Date: 2006/10/05 11:30:00 CDT
937 :	mblume	2066	Tag: blume-20061005-merge-amd64
938 :			Description:
939 :
940 :			Merged code for AMD64 backend (Mike Rainey's work).
941 :			Everything is hooked up but untested.
942 :
943 :			----------------------------------------------------------------------
944 :			Name: Matthias Blume (http://www.tti-c.org/blume/)
945 :	mblume	1931	Date: 2006/05/17 14:00:00 CST
946 :			Tag: Release_110_59
947 :			Description:
948 :
949 :	blume	2193	New working version (110.59). NEW BOOTFILES!
950 :	mblume	1931
951 :			I am freezing 110.59. Changes other than the version-number increase:
952 :
953 :			eXene:
954 :
955 :			* committed changes to eXene from Alley Stoughton:
956 :			"fixed bugs in X authorization and resource handling, as well
957 :			as in the pile and viewport widgets"
958 :
959 :			Runtime:
960 :			* fixed linking problem with NetBSD 3.x.
961 :
962 :			Lexgen:
963 :			* lexgen tool handles non-ascii characters in 7-bit mode the same
964 :			way that ml-lex does
965 :
966 :			* lexgen propagates exceptions the same way that ml-lex does
967 :
968 :			CML:
969 :			* Fixed a bug in the SyncVar polling functions (iGetPoll,
970 :			mTakePoll, and mGetPoll) that could lead to livelock.
971 :
972 :			----------------------------------------------------------------------
973 :			Name: Matthias Blume (http://www.tti-c.org/blume/)
974 :	mblume	1926	Date: 2006/05/12 15:20:00 CST
975 :			Tag: blume-20060515-makedepend
976 :			Description:
977 :
978 :			Implemented ml-makedepend (i.e., CM.sources) in a better (more robust)
979 :			way. This should hopefully fix the ml-makedepend problem permanently.
980 :
981 :			----------------------------------------------------------------------
982 :			Name: Matthias Blume (http://www.tti-c.org/blume/)
983 :	mblume	1925	Date: 2006/05/12 15:15:00 CST
984 :			Tag: blume-20060512-makedepend
985 :			Description:
986 :
987 :			Fixed long-standing bug with ml-makedepend where it would output a
988 :			spurious dependency to a non-existing file. (This is a simple fix.
989 :			It might need further looking into.)
990 :
991 :			----------------------------------------------------------------------
992 :			Name: Matthias Blume (http://www.tti-c.org/blume/)
993 :	mblume	1923	Date: 2006/04/20 10:30:00 CDT
994 :			Tag: blume-20060420-heap2exec
995 :			Description:
996 :
997 :			Committed patches received from Johannes 5 Joemann (joemann@befree.free.de)
998 :			that enable heap2exec under Linux and FreeBSD.
999 :
1000 :			----------------------------------------------------------------------
1001 :	jhr	1921	Name: John Reppy
1002 :			Date: 2006/04/14
1003 :	blume	2155	Tag:
1004 :	jhr	1921	Description:
1005 :
1006 :			MLRISC changes:
1007 :			1) renamed GAS_PSEUDO_OPS to AS_PSEUDO_OPS and put it in its own file.
1008 :			2) added support for NOTB and XORB operators in pseudo-op expressions
1009 :			3) added DarwinPseudoOp functor that supports Darwin's assembler syntax.
1010 :			4) added support for 64-bit integer literals
1011 :
1012 :			----------------------------------------------------------------------
1013 :	jhr	1922	Name: Matthias Blume (http://www.tti-c.org/blume/)
1014 :	mblume	1907	Date: yyyy/mm/dd
1015 :			Tag: Release_110_58
1016 :			Description:
1017 :
1018 :			New working version (110.58). NEW BOOTFILES!
1019 :
1020 :			----------------------------------------------------------------------
1021 :	mblume	1896	Name: Matthias Blume (http://www.tti-c.org/blume/)
1022 :	mblume	1905	Date: 2006/03/01 22:15:00 CST
1023 :			Tag: blume-20060301-lexgen
1024 :			Description:
1025 :
1026 :			Incorporated several bugfixes to lexgen. Compiler now compiles to
1027 :			fixpoint when using lexgen instead of ml-lex.
1028 :
1029 :			----------------------------------------------------------------------
1030 :			Name: Matthias Blume (http://www.tti-c.org/blume/)
1031 :	mblume	1902	Date: 2006/02/28 22:45:00 CST
1032 :			Tag: blume-20060228-lexgen
1033 :			Description:
1034 :
1035 :			Removed ml-flex and added lexgen instead, using Aaron Turon's newly
1036 :			provided tarball. The generated lexers still have problems.
1037 :
1038 :			----------------------------------------------------------------------
1039 :			Name: Matthias Blume (http://www.tti-c.org/blume/)
1040 :	mblume	1898	Date: 2006/02/26 23:30:00 CST
1041 :			Tag: blume-20060226-ml-flex
1042 :			Description:
1043 :
1044 :			Removed ml-flex' dependency on regexp-lib.cm. Turned installation
1045 :			of ml-flex on by default.
1046 :
1047 :			----------------------------------------------------------------------
1048 :			Name: Matthias Blume (http://www.tti-c.org/blume/)
1049 :	mblume	1896	Date: 2006/02/24 16:30:00 CST
1050 :			Tag: blume-20060224-ml-flex
1051 :			Description:
1052 :
1053 :			Added ml-flex sources. Partially integrated, but should not be
1054 :			turned on yet! (Read: leave it commented-out in config/targets!)
1055 :
1056 :			----------------------------------------------------------------------
1057 :	jhr	1893	Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
1058 :			Date: 2006/02/23
1059 :			Tag: jhr-20060223-x86-64
1060 :			Description:
1061 :			Changes to support building on x86-64 systems (using the 32-bit mode).
1062 :			Also cleaned up signal handling on Linux. Support for pre-2.2 Linux
1063 :			kernels dropped.
1064 :
1065 :			----------------------------------------------------------------------
1066 :	allenleung	1890	Name: Allen Leung (leunga (at) reservoir (dot) com)
1067 :			Date: 2006/02/22 15:07:00 EST
1068 :			Tag: leunga-20060222-mdl-gen
1069 :			Description:
1070 :
1071 :			CM has changed. Updated the script for rebuilding the MLRISC generated
1072 :			files.
1073 :
1074 :			----------------------------------------------------------------------
1075 :	mblume	1886	Name: Matthias Blume (http://www.tti-c.org/blume/)
1076 :			Date: 2006/02/14 17:30:00 CST
1077 :			Tag: blume-20060214-intel-mac
1078 :			Description:
1079 :
1080 :			Hooked code for Darwin-specific Intel ABI into main compiler.
1081 :			(This is a temporary solution which relies on the fact that the
1082 :			compiler itself does not use NLFFI. Eventually we need to divorce
1083 :			intel mac from generic x86 unix code and make separate sets of binaries.)
1084 :
1085 :			----------------------------------------------------------------------
1086 :	jhr	1882	Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
1087 :	jhr	1885	Date: 2006/02/13
1088 :			Tag: jhr-20060213-intel-mac
1089 :			Description:
1090 :			Changed MLRISC x86 CCalls for partial support of MacOS X ABI.
1091 :
1092 :			----------------------------------------------------------------------
1093 :			Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
1094 :	jhr	1882	Date: 2006/02/06
1095 :			Tag: jhr-20060206-intel-mac
1096 :			Description:
1097 :			Changes to support MacOS X on Intel hardware. The C-calls support in
1098 :			MLRISC must also be updated to support the MacOS X ABI.
1099 :
1100 :			----------------------------------------------------------------------
1101 :	mblume	1863	Name: Matthias Blume (http://www.tti-c.org/blume/)
1102 :	mblume	1879	Date: 2005/12/16 00:15:00 CST
1103 :			Tag: blume-20051216-cm
1104 :			Description:
1105 :
1106 :			Improved error reporting and handling in CM.
1107 :
1108 :			----------------------------------------------------------------------
1109 :			Name: Matthias Blume (http://www.tti-c.org/blume/)
1110 :	mblume	1878	Date: 2005/11/21 14:00:00 CST
1111 :			Tag: blume-20051121-html-lib
1112 :			Description:
1113 :
1114 :			Enabling $/html-lib.cm in config/preloads did not work. This is fixed
1115 :			now. (Since the anchor mapping for html-lib.cm is not yet in effect
1116 :	blume	2155	at the time when config/preloads is processed, the library has to
1117 :	mblume	1878	be referred to by another name. In 110.57 this would be
1118 :			$SMLNJ-LIB/HTML/html-lib.cm. I arranged for $smlnj/smlnj-lib/html-lib.cm
1119 :			to be valid as well.)
1120 :
1121 :			Thanks to Todd Wilson (Fresno) for alerting me to this issue.
1122 :
1123 :			----------------------------------------------------------------------
1124 :			Name: Matthias Blume (http://www.tti-c.org/blume/)
1125 :	mblume	1874	Date: 2005/11/19 13:45:00 CST
1126 :			Tag: Release_110_57
1127 :			Description:
1128 :
1129 :			Moved release tag for 110.57.
1130 :
1131 :			----------------------------------------------------------------------
1132 :			Name: Matthias Blume (http://www.tti-c.org/blume/)
1133 :	mblume	1873	Date: 2005/11/19 23:35:00 CST
1134 :			Tag: blume-20051119-fish
1135 :			Description:
1136 :
1137 :			Fixed a problem in config/install.sh where it tries to "fish" the
1138 :			name of the CM metadata directory from the wrong place (because
1139 :			the physical location of basis.cm has changed). Also, corrected the
1140 :			path anchor for $/html-lib.cm. (Thanks to M. Fluet for pointing out
1141 :			these problems.)
1142 :
1143 :			----------------------------------------------------------------------
1144 :			Name: Matthias Blume (http://www.tti-c.org/blume/)
1145 :	mblume	1867	Date: 2005/11/16 14:30:00 CST
1146 :	mblume	1874	Tag: blume-orig-110_57
1147 :	mblume	1867	Description:
1148 :
1149 :			New working version (110.57). NEW BOOTFILES!
1150 :
1151 :	mblume	1874	UPDATE: The release tag has been moved to include a fix for a brown-paper-
1152 :			bag glitch (see above).
1153 :
1154 :	mblume	1867	----------------------------------------------------------------------
1155 :			Name: Matthias Blume (http://www.tti-c.org/blume/)
1156 :	mblume	1866	Date: 2005/11/16 10:25:00 CST
1157 :			Tag: blume-22051116-btrace
1158 :			Description:
1159 :
1160 :			Fixed problem with bogus exception message when using back-trace
1161 :			facility.
1162 :
1163 :			----------------------------------------------------------------------
1164 :			Name: Matthias Blume (http://www.tti-c.org/blume/)
1165 :	mblume	1865	Date: 2005/11/15 22:35:00 CST
1166 :			Tag: blume-22051115-array2copy
1167 :			Description:
1168 :
1169 :			Added simple implementation of Array2.copy. (Warning: mostly untested.)
1170 :
1171 :			----------------------------------------------------------------------
1172 :			Name: Matthias Blume (http://www.tti-c.org/blume/)
1173 :	mblume	1864	Date: 2005/11/15 21:35:00 CST
1174 :			Tag: blume-22051115-versiontool
1175 :			Description:
1176 :
1177 :			Reversed change to src/system/smlnj/internals/versiontool.cm. This
1178 :			file gets loaded as a tool -- by the equivalent of CM.make during the
1179 :			run of CMB.make. Thus, CMB's path configuration is meaningless for
1180 :			it. Instead, it has the status of "user code", so it should use
1181 :			$/basis.cm to refer to the Basis library. (At least that's true for
1182 :			the purpose of bootstrapping the previous change. In the future it
1183 :			might make sense to have versiontool.cm refer to
1184 :			$smlnj/basis/basis.cm, i.e., the version of the Basis that the
1185 :			compiler itself uses.)
1186 :
1187 :			Also patched src/system/testml to have it activate those extra anchor
1188 :			bindings in config/extrapathconfig.
1189 :
1190 :			----------------------------------------------------------------------
1191 :			Name: Matthias Blume (http://www.tti-c.org/blume/)
1192 :	mblume	1863	Date: 2005/11/15 16:35:00 CST
1193 :			Tag: blume-22051115-libraries
1194 :			Description:
1195 :
1196 :			This change affects the way the following libraries are tied into
1197 :			the system:
1198 :
1199 :			$/basis.cm
1200 :			$/smlnj-lib.cm
1201 :			$/pp-lib.cm
1202 :			$/controls-lib.cm
1203 :			$/html-lib.cm
1204 :			$/ml-yacc-lib.cm
1205 :
1206 :			These libraries are now internally (as seen from the source code
1207 :			of the implementation itself) known by the following names:
1208 :
1209 :			$smlnj/basis/basis.cm
1210 :			$smlnj/smlnj-lib/smlnj-lib.cm
1211 :			$smlnj/smlnj-lib/pp-lib.cm
1212 :			$smlnj/smlnj-lib/controls-lib.cm
1213 :			$smlnj/smlnj-lib/html-lib.cm
1214 :			$smlnj/ml-yacc/ml-yacc-lib.cm
1215 :
1216 :			This makes it possible to work with code that requires different
1217 :			versions of these libraries, and which refers to these libraries using
1218 :			their "default" names (i.e., the first set of names above). In other
1219 :			words, one can un-define or re-define those default names without
1220 :			compromising the proper functioning of the compiler itself.
1221 :
1222 :			A similar procedure had already been performed for several of the
1223 :			MLRISC libraries that are linked into the compiler. I did some
1224 :			cleanup on this code.
1225 :
1226 :			A new file in the config directory (named extrapathconfig) is
1227 :			responsible for setting up path anchors that the compiler itself does
1228 :			not need, but that are typically required by user code.
1229 :
1230 :			----------------------------------------------------------------------
1231 :	mblume	1862	Name: Matthias Blume
1232 :			Date: 2005/11/07 17:05:00 CST
1233 :			Tag: blume-20051107-slice
1234 :			Description:
1235 :
1236 :			Fixed erroneous out-of-bounds test in the "update" function of
1237 :			various *ArraySlice modules. (Thanks to Vesa A. Norrman for pointing
1238 :			out the problem.)
1239 :
1240 :			Pushed some Basis changes through ML-Lex, CML, and eXene.
1241 :
1242 :			----------------------------------------------------------------------
1243 :	jhr	1858	Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
1244 :	jhr	1861	Date: 2005/11/07
1245 :			Description:
1246 :			Fixed a Basis incompatibility: The depreciated function Substring.all
1247 :			was removed (use Substring.full instead).
1248 :
1249 :			----------------------------------------------------------------------
1250 :			Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
1251 :	jhr	1858	Date: 2005/11/05
1252 :			Description:
1253 :			Tweaked interval set API in SML/NJ library; see the CHANGES file for details.
1254 :
1255 :			----------------------------------------------------------------------
1256 :	mblume	1856	Name: Matthias Blume (http://www.tti-c.org/blume/)
1257 :			Date: 2005/11/02 16:25:00 CST
1258 :			Tag: blume-20051102-hex
1259 :			Description:
1260 :
1261 :			* Runtime system bootstrap code now accepts hex digits in BOOTLIST
1262 :			in either upper case or lower case format.
1263 :			* Pushed changes to names of Pack<N>{Big,Little} structures through
1264 :			CML and eXene.
1265 :
1266 :			----------------------------------------------------------------------
1267 :	jhr	1852	Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
1268 :	jhr	1855	Date: 2005/11/02
1269 :			Description:
1270 :			Fixed a Basis incompatibility: Pack<N>{Big,Little} structures should
1271 :			be called PackWord<N>{Big,Little}.
1272 :
1273 :			----------------------------------------------------------------------
1274 :			Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
1275 :	jhr	1852	Date: 2005/10/28
1276 :			Description:
1277 :			Fixed a minor Basis incompatibility: hex digits should be upper case.
1278 :
1279 :			----------------------------------------------------------------------
1280 :	mblume	1846	Name: Matthias Blume (blume (at) tti - c (dot) org)
1281 :			Date: 2005/10/25 16:00:00 CDT
1282 :			Tag: Release_110_56
1283 :			Description:
1284 :
1285 :			New working version (110.56). NEW BOOTFILES!
1286 :
1287 :			----------------------------------------------------------------------
1288 :	jhr	1838	Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
1289 :	jhr	1840	Date: 2005/10/25
1290 :			Description:
1291 :			Added interval sets to utility library (signatures INTERVAL_DOMAIN
1292 :			and INTERVAL_SET, and functor IntervalSetFn).
1293 :
1294 :			----------------------------------------------------------------------
1295 :			Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
1296 :	jhr	1838	Date: 2005/10/14
1297 :			Description:
1298 :			Add Zhong Shao's fix for datatype equality functions.
1299 :
1300 :			----------------------------------------------------------------------
1301 :	allenleung	1836	Name: Allen Leung (leunga (at) reservoir (dot) com)
1302 :	allenleung	1837	Date: 2005/10/14 12:30:00 EST
1303 :			Tag: leunga-20051014-x86-spill
1304 :			Description:
1305 :
1306 :			Bug fix a bug found by Carl Hauser.
1307 :			There was a typo in the reload code for FCMP in x86SpillInstr.sml.
1308 :
1309 :			----------------------------------------------------------------------
1310 :			Name: Allen Leung (leunga (at) reservoir (dot) com)
1311 :	allenleung	1836	Date: 2005/10/14 11:15:00 EST
1312 :			Tag: leunga-20051014-x86-asm
1313 :			Description:
1314 :
1315 :			Removed some debugging code in file x86Asm.sml.
1316 :			The function emit_operand was printing out debugging output.
1317 :
1318 :			----------------------------------------------------------------------
1319 :	jhr	1813	Name: John Reppy
1320 :	jhr	1831	Date: 2005/07/27
1321 :			Tag:
1322 :			Description:
1323 :
1324 :	jhr	1832	Fixed ml-lex to recognize "\r" as representing carriage return.
1325 :
1326 :			----------------------------------------------------------------------
1327 :			Name: John Reppy
1328 :			Date: 2005/07/27
1329 :			Tag:
1330 :			Description:
1331 :
1332 :	jhr	1831	Fixed ml-yacc to work on files with non-native end-of-line
1333 :			encodings (e.g., Windows text file on a Unix system).
1334 :
1335 :			----------------------------------------------------------------------
1336 :			Name: John Reppy
1337 :	jhr	1824	Date: 2005/07/20
1338 :			Tag:
1339 :			Description:
1340 :
1341 :			Added changes from Dominic Evans (oldmanuk (at) gmail (dot) com)
1342 :			to support HPUX 11.
1343 :
1344 :			----------------------------------------------------------------------
1345 :			Name: John Reppy
1346 :	jhr	1813	Date: 2005/07/06
1347 :			Tag:
1348 :			Description:
1349 :
1350 :			Changes to the SML/NJ library. See smlnj-lib/CHANGES for details.
1351 :
1352 :			----------------------------------------------------------------------
1353 :	mblume	1808	Name: Matthias Blume (blume (at) tti - c (dot) org)
1354 :			Date: 2005/07/06 00:45:00 CDT
1355 :			Tag: blume-20050706-slice-copy
1356 :			Description:
1357 :
1358 :			Fixed reversed logic for deciding whether to "copy up" or "copy down"
1359 :			in *-array-slice.sml.
1360 :
1361 :			----------------------------------------------------------------------
1362 :	allenleung	1806	Name: Allen Leung (leunga (at) reservoir (dot) com)
1363 :	allenleung	1807	Date: 2005/05/31 17:00:00 EST
1364 :			Tag: leunga-20050531-cygwin-fault-2
1365 :			Description:
1366 :
1367 :			A typo in the cygwin code fixed.
1368 :
1369 :			----------------------------------------------------------------------
1370 :			Name: Allen Leung (leunga (at) reservoir (dot) com)
1371 :	allenleung	1806	Date: 2005/05/31 16:47:00 EST
1372 :			Tag: leunga-20050531-cygwin-fault
1373 :			Description:
1374 :
1375 :	blume	2155	Updated Cygwin's fault/signal handling to match the Windows version.
1376 :	allenleung	1806	Updated the export list.
1377 :
1378 :			----------------------------------------------------------------------
1379 :	mblume	1791	Name: Matthias Blume (blume (at) tti - c (dot) org)
1380 :	mblume	1795	Date: 2005/05/18 13:35:00 CDT
1381 :			Tag: Release_110_54
1382 :			Description:
1383 :
1384 :			New working version (110.54). NEW BOOTFILES!
1385 :
1386 :			----------------------------------------------------------------------
1387 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
1388 :	mblume	1794	Date: 2005/05/18 11:58:00 CDT
1389 :			Tag: blume-20050518-installer
1390 :			Description:
1391 :
1392 :			Added support scripts for Mac OS X PackageMaker and modified
1393 :			config/install.sh so that it supports re-dumping a heap image
1394 :			after customization.
1395 :
1396 :			----------------------------------------------------------------------
1397 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
1398 :	mblume	1793	Date: 2005/05/18 10:55:00 CDT
1399 :			Tag: blume-20050518-realdiv-noovld
1400 :			Description:
1401 :
1402 :			Un-overloaded / to work around bug in overloading resolution code.
1403 :
1404 :			----------------------------------------------------------------------
1405 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
1406 :	mblume	1792	Date: 2005/05/16 23:50:00 CDT
1407 :			Tag: blume-20050516-redump-heap
1408 :			Description:
1409 :
1410 :			Added mechanism for re-creating a heap file for the interactive system
1411 :			after configuration variables have been changed.
1412 :
1413 :			CM.redump_heap : string -> unit
1414 :
1415 :			This is much like SMLofNJ.exportML, but starting from the resulting
1416 :	blume	2155	heap does not return to the caller of CM.redump_heap but
1417 :	mblume	1792	restarts the interactive system from scratch. The original call of
1418 :			CM.redump_heap does not return but ends the interactive session.
1419 :			Thus, CM.redump_heap is a lot like SMLofNJ.exportFn.
1420 :
1421 :			Internally, redump_heap winds the dynamic execution context back to
1422 :			the point where the original heap image was created and re-executes
1423 :			the heap image generation code in the boot code.
1424 :
1425 :			----------------------------------------------------------------------
1426 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
1427 :	mblume	1791	Date: 2005/05/09 21:55:00 CDT
1428 :			Tag: blume-20050509-word64
1429 :			Description:
1430 :
1431 :			Added a hack to the existing hack known as Word64 to make fromString
1432 :			behave correctly. I am still not sure whether Word64.scan will work
1433 :			as specified with respect to the interaction of radix and prefix.
1434 :
1435 :			----------------------------------------------------------------------
1436 :	allenleung	1785	Name: Allen Leung (leunga (at) reservoir (dot) com)
1437 :			Date: 2005/05/04 11:50:00 EST
1438 :			Tag: leunga-20050504-checkgc
1439 :			Description:
1440 :
1441 :	blume	2155	Added a gc protocol checking phase. This phase is enabled with
1442 :	allenleung	1785	the flag "check-gc". "debug-check-gc" turns on the verbose mode.
1443 :
1444 :			----------------------------------------------------------------------
1445 :	mblume	1741	Name: Matthias Blume (blume (at) tti - c (dot) org)
1446 :	mblume	1784	Date: 2005/05/04 10:45:00 CDT
1447 :			Tag: blume-20050504-intinf
1448 :			Description:
1449 :
1450 :			Fixed a bug in the implementation of div and mod for IntInf.
1451 :			Thanks to Neophytos Michael for reporting the problem.
1452 :
1453 :			----------------------------------------------------------------------
1454 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
1455 :	mblume	1783	Date: 2005/05/04 10:35:00 CDT
1456 :			Tag: blume-20050504-join
1457 :			Description:
1458 :
1459 :			Added a "join" combinator to the ParserComb module in smlnj-lib.cm.
1460 :
1461 :			----------------------------------------------------------------------
1462 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
1463 :	mblume	1770	Date: 2005/02/28 23:40:00 CST
1464 :			Tag: blume-20050228-mVar
1465 :			Description:
1466 :
1467 :			Fixed serious bug (brown paper bag variety) in new implementation of
1468 :			structure Atom in CML. (I had accidentally used a mailbox instead of
1469 :			an mvar, leaving the door open for races.)
1470 :
1471 :			----------------------------------------------------------------------
1472 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
1473 :	mblume	1761	Date: 2005/02/25 15:00:00 CST
1474 :			Tag: Release_110_53
1475 :			Description:
1476 :
1477 :			New working version (110.53). NEW BOOTFILES!
1478 :
1479 :			----------------------------------------------------------------------
1480 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
1481 :	mblume	1760	Date: 2005/02/25 14:50:00 CST
1482 :			Tag: blume-20050225-susp
1483 :			Description:
1484 :
1485 :			Brought back SMLofNJ.Susp. The underlying suspension type is the one
1486 :			implemented in Core -- which means that it is the same as the one used
1487 :			by the lazy extension.
1488 :
1489 :			----------------------------------------------------------------------
1490 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
1491 :	mblume	1759	Date: 2005/02/24 16:50:00 CST
1492 :			Tag: blume-20050224-cml-atom
1493 :			Description:
1494 :
1495 :			Simpler and at the same time more general implementation of structure
1496 :			Atom in CML.
1497 :
1498 :			----------------------------------------------------------------------
1499 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
1500 :	mblume	1758	Date: 2005/02/15 17:35:00 CST
1501 :			Tag: blume-20050215-tools
1502 :			Description:
1503 :
1504 :			Created new "tools" directory under "src" and moved "TraceDebugProf"
1505 :			there.
1506 :
1507 :			----------------------------------------------------------------------
1508 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
1509 :	mblume	1755	Date: 2005/02/10 17:55:00 CST
1510 :			Tag: blume-20050210-longlong
1511 :			Description:
1512 :
1513 :			Implemented "long long" arguments and results for NLFFI. (Only the
1514 :			PPC/MacOS implementation is complete, the other backends still need to
1515 :			be updated.)
1516 :
1517 :			----------------------------------------------------------------------
1518 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
1519 :	mblume	1751	Date: 2005/01/24 17:40:00 CST
1520 :			Tag: blume-20050124-mlyacc
1521 :			Description:
1522 :
1523 :			Minor cleanup in ML-Yacc rule printing mechanism. This should fix a
1524 :			problem with certain "as" patterns which previously got rendered
1525 :			using incorrect syntax.
1526 :
1527 :			----------------------------------------------------------------------
1528 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
1529 :	mblume	1750	Date: 2005/01/18 12:00:00 CST
1530 :			Tag: blume-20050118-profile
1531 :			Description:
1532 :
1533 :			Made time profiling code (interrupt handler) in runtime system aware
1534 :			of new array representation.
1535 :
1536 :			----------------------------------------------------------------------
1537 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
1538 :	mblume	1747	Date: 2005/01/14 18:00:00 CST
1539 :			Tag: blume-20050114-heap2exec
1540 :			Description:
1541 :
1542 :			Implemented new (but still experimental) heap2exec facility. This is
1543 :			tested under Mac OS X and should work under Linux (will test shortly).
1544 :			It will probably also work on the Sparc (will test some time later).
1545 :
1546 :			- removed old "HACKED_STANDALONE" hack from runtime
1547 :
1548 :			To be able to test this, uncomment the request for "heap2asm" in
1549 :			config/targets prior to installation. (Notice that this is different
1550 :			from "heap2exec" mentioned below. Not a typo.)
1551 :
1552 :			To perform an actual test, run the command
1553 :
1554 :			$ bin/heap2exec heapfile execfile
1555 :
1556 :			(You can put heap2exec on your shell's path.)
1557 :
1558 :			For example, run
1559 :
1560 :			$ bin/heap2exec bin/.heap/ml-yacc.ppc-darwin mly
1561 :
1562 :			This will create a standalone executable called "mly" which you can
1563 :			then invoke directly as a command.
1564 :
1565 :			----------------------------------------------------------------------
1566 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
1567 :	mblume	1746	Date: 2005/01/07 11:44:00 CST
1568 :			Tag: blume-20050107-mlstring
1569 :			Description:
1570 :
1571 :			fixed off-by-one error in ML_STRING macro (globals.c)
1572 :
1573 :			----------------------------------------------------------------------
1574 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
1575 :	mblume	1744	Date: 2004/12/23 18:00:00 CST
1576 :			Tag: blume-20041223-santa
1577 :			Description:
1578 :
1579 :			Made ml-build script "smarter" (but only very little).
1580 :
1581 :			----------------------------------------------------------------------
1582 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
1583 :	mblume	1742	Date: 2004/12/21 15:05:00 CST
1584 :			Tag: blume-20041221-longlong
1585 :			Description:
1586 :
1587 :			* Implemented access to signed and unsigned long long data in NLFFI.
1588 :			(The parameter-passing part of the picture has not complete. But
1589 :			data structure access seems to work.)
1590 :
1591 :			* Fixed CM's incorrect assumption that the PPC is little-endian.
1592 :			(On the Mac, it is big-endian. And that's currently our only
1593 :			PPC platform.)
1594 :
1595 :			----------------------------------------------------------------------
1596 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
1597 :	mblume	1741	Date: 2004/12/21 12:50:00 CST
1598 :			Tag: blume-20041221-memory
1599 :			Description:
1600 :
1601 :			Some cleanup in the $c/memory.cm library: separated some concerns by
1602 :			moving allocation code and memory access code each into their own
1603 :			files.
1604 :
1605 :			----------------------------------------------------------------------
1606 :	allenleung	1740	Name: Allen Leung (leunga (at) reservoir (dot) com)
1607 :			Date: 2004/12/17 16:12:00 EST
1608 :			Tag: leunga-20041217-cygwin-smlnj-home
1609 :			Description:
1610 :
1611 :	blume	2155	The Unix I/O library of SML/NJ on cygwin does not understand
1612 :			Windows style pathname, so problems arise when SMLNJ_HOME is set
1613 :			to a Windows style pathname. _run-sml now converts SMLNJ_HOME
1614 :			to a POSIX pathname on cygwin.
1615 :	allenleung	1740
1616 :			----------------------------------------------------------------------
1617 :	mblume	1737	Name: Matthias Blume (blume (at) tti - c (dot) org)
1618 :	mblume	1736	Date: 2004/12/16 13:35:00 CST
1619 :			Tag: Release_110_52
1620 :			Description:
1621 :
1622 :			Last-minute changes incorporated into 110.52. Release tag moved.
1623 :
1624 :			The changes:
1625 :			- HashString.hashString' -> HashString.hashSubstring
1626 :			- bug fix in UnivariateStats
1627 :
1628 :			----------------------------------------------------------------------
1629 :	mblume	1702	Name: Matthias Blume (blume (at) tti - c (dot) org)
1630 :	mblume	1733	Date: 2004/12/15 23:40:00 CST
1631 :			Tag: blume-20041215-hashSubstring
1632 :			Description:
1633 :
1634 :			- HashString.hashString' -> HashString.hashSubstring
1635 :			- corresponding changes in atom.sml
1636 :			- "de-compressed" (aka. un-obfuscated) code for UnivariateStats and
1637 :			added some comments
1638 :
1639 :			----------------------------------------------------------------------
1640 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
1641 :	mblume	1728	Date: 2004/12/15 15:30:00 CST
1642 :	mblume	1736	Tag: (Release_110_52)
1643 :	mblume	1728	Description:
1644 :
1645 :			New working version (110.52). NEW BOOTFILES!
1646 :
1647 :			----------------------------------------------------------------------
1648 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
1649 :	mblume	1727	Date: 2004/12/15 12:45:00 CST
1650 :			Tag: blume-20041215-spaces
1651 :			Description:
1652 :
1653 :			More on the space problem (this time for Win32).
1654 :
1655 :			----------------------------------------------------------------------
1656 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
1657 :	mblume	1725	Date: 2004/12/14 17:30:00 CST
1658 :			Tag: blume-20041214-spaces
1659 :			Description:
1660 :
1661 :			Hacked some of the scripts (in particular: the installer) to cope with
1662 :			spaces in filenames a bit better. But beware: the current "solution"
1663 :			is likely still full of bugs and inherently incomplete. (We need to
1664 :			do away with those shell scripts for a comprehensive solution.)
1665 :
1666 :			----------------------------------------------------------------------
1667 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
1668 :	mblume	1724	Date: 2004/12/13 14:45:00 CST
1669 :			Tag: blume-20041213-ml-makedepend
1670 :			Description:
1671 :
1672 :			Fixed bug in code for ml-makedepend.
1673 :
1674 :			----------------------------------------------------------------------
1675 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
1676 :	mblume	1720	Date: 2004/12/09 16:30:00 CST
1677 :			Tag: blume-20041209-statistics
1678 :			Description:
1679 :
1680 :			Added two simple but potentially useful statistics modules to SML/NJ Library.
1681 :			(See CHANGES file there.)
1682 :
1683 :			----------------------------------------------------------------------
1684 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
1685 :	mblume	1718	Date: 2004/12/01 16:50:00 CST
1686 :			Tag: blume-20041201-atom
1687 :			Description:
1688 :
1689 :			smlnj-lib:
1690 :
1691 :			Added function HashString.hashString' for substrings.
1692 :			Hand-inlined CharVector.foldl into HashString (for speed).
1693 :			Modified implementation of structure Atom to avoid extracting
1694 :			strings from substrings unless necessary.
1695 :			(Also see CHANGES file for smlnj-lib.)
1696 :
1697 :			----------------------------------------------------------------------
1698 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
1699 :	mblume	1717	Date: 2004/11/24 22:15:00 CST
1700 :			Tag: blume-20041124-cml
1701 :			Description:
1702 :
1703 :			Made sure CML compiles when Position = Int64.
1704 :
1705 :			----------------------------------------------------------------------
1706 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
1707 :	mblume	1715	Date: 2004/11/24 14:30:00 CST
1708 :			Tag: blume-20041124-position
1709 :			Description:
1710 :
1711 :			The compiler can now be compiled in a mode that makes structure Position
1712 :			equal to Int64. The default, however, is unchanged (Position = Int31)
1713 :			for the time being.
1714 :
1715 :			To enable 64-bit positions, use the following procedure:
1716 :
1717 :			1. Start sml
1718 :			2. Autoload $smlnj/cmb.cm (if not already autoloaded)
1719 :			3. Type
1720 :			#set (CMB.symval "USE_64_BIT_POSITIONS") (SOME 1);
1721 :			4. Run CMB.make() as usual.
1722 :
1723 :			This is barely tested. The only test so far was a little SML program
1724 :			counting the number of characters in an 8-gigabyte file by
1725 :			reading it character-by-character. That test was successful.
1726 :
1727 :			In support of 64-bit positions, a number of new functions have been
1728 :			added to the runtime system.
1729 :
1730 :			----------------------------------------------------------------------
1731 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
1732 :	mblume	1716	Date: 2004/11/23 14:45:00 CST
1733 :			Tag: blume-20041123-useFile
1734 :	mblume	1712	Description:
1735 :
1736 :			Fixed a problem with unhelpful error messages related to problems with
1737 :			.cm- or .sml files that appear as part of the sml command line.
1738 :
1739 :			----------------------------------------------------------------------
1740 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
1741 :	mblume	1703	Date: 2004/11/18 15:40:00 CST
1742 :			Tag: Release_110_51
1743 :			Description:
1744 :
1745 :			New working version (110.51). NEW BOOTFILES!
1746 :
1747 :			----------------------------------------------------------------------
1748 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
1749 :	mblume	1702	Date: 2004/11/18 15:35:00 CST
1750 :			Tag: <none>
1751 :			Description:
1752 :
1753 :			Enabled dlopen and friends for FreeBSD (as recommended by Johannes 5
1754 :			Joemann).
1755 :
1756 :			----------------------------------------------------------------------
1757 :	allenleung	1701	Name: Allen Leung (leunga (at) reservoir (dot) com)
1758 :			Date: 2004/11/17 16:05:21 EST 2004
1759 :			Tag: leunga-20041117-mlrisc-live-kill
1760 :			Description:
1761 :
1762 :	blume	2155	Added support for MLTree constructs LIVE and KILL
1763 :	allenleung	1701	to all the architectures.
1764 :
1765 :			----------------------------------------------------------------------
1766 :	mblume	1687	Name: Matthias Blume (blume (at) tti - c (dot) org)
1767 :	mblume	1689	Date: 2004/11/13 00:20:00 CST
1768 :			Tag: blume-20041113-versiontool
1769 :			Description:
1770 :
1771 :			- Stripped down the versiontool: It now only handles the version number.
1772 :			The date string is generated at bootstrap time (during makeml).
1773 :
1774 :			- In a previous commit, fixed a minor issue with how polyequal is being
1775 :			translated. In particular, the code now "looks through" abstractions.
1776 :			This results in slightly fewer polyEqual warnings and hopefully slightly
1777 :			more efficient code. Important examples for where this matters are
1778 :			the new int64 and word64 types.
1779 :
1780 :			----------------------------------------------------------------------
1781 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
1782 :	mblume	1687	Date: 2004/11/12 00:30:00 CST
1783 :			Tag: blume-20041112-int64
1784 :			Description:
1785 :
1786 :			Structure Int64 fully hooked in. (The implementation is not very
1787 :			efficient, though.)
1788 :
1789 :			----------------------------------------------------------------------
1790 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
1791 :	mblume	1686	Date: 2004/11/11 17:30:00 CST
1792 :			Tag: blume-20041111-more64
1793 :			Description:
1794 :
1795 :			All the pieces of Word64 are now there, with the exception of the
1796 :			conversions from and to LargeWord. (Eventually these need to be identities,
1797 :			but for the time being they don't even make sense because LargeWord is
1798 :			32-bit wide.)
1799 :
1800 :			Also started to add similar support for Int64, but major pieces of that
1801 :			are still missing.
1802 :
1803 :			----------------------------------------------------------------------
1804 :	mblume	1668	Name: Matthias Blume (blume (at) tti - c (dot) org)
1805 :	mblume	1684	Date: 2004/11/11 00:15:00 CST
1806 :			Tag: blume-20041111-word64
1807 :			Description:
1808 :
1809 :			Structure Word64 is now (almost) complete, word literals and patterns
1810 :			seem to work. There are a few odd pieces missing. In particular,
1811 :			I didn't do the {from,to}LargeWord parts because LargeWord is still Word32
1812 :			at the moment.
1813 :
1814 :			Making Word64 official would mean that LargeWord becomes Word64. But
1815 :			this requires extreme care because most word-word conversions have to
1816 :			go through LargeWord, so making a mistake means loss of efficiency or
1817 :			worse. Eventually there will be a solution similar to (but actually
1818 :			simpler than) what I did with IntInf.
1819 :
1820 :			----------------------------------------------------------------------
1821 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
1822 :	mblume	1683	Date: 2004/11/10 18:12:00 CST
1823 :			Tag: blume-20041110-64bit
1824 :			Description:
1825 :
1826 :			More 64-bit hacking (but still not even half-way there yet).
1827 :			Also, some assorted improvements to the handling of 8-bit words.
1828 :
1829 :			----------------------------------------------------------------------
1830 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
1831 :	mblume	1682	Date: 2004/11/09 17:50:00 CST
1832 :	mblume	1683	Tag: <none>
1833 :	mblume	1682	Description:
1834 :
1835 :			Started adding some infrastructure for supporting 64-bit int- and
1836 :			word-types. (Still in its very early stages.)
1837 :
1838 :			----------------------------------------------------------------------
1839 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
1840 :	mblume	1672	Date: 2004/10/28 10:45:00 CDT
1841 :			Tag: Release_110_50
1842 :			Description:
1843 :
1844 :			New working version (110.50). NEW BOOTFILES!
1845 :
1846 :			=====================
1847 :
1848 :			Also:
1849 :
1850 :			- Changend config/srcarchiveurl from a file just
1851 :			containing the URL string into a file containing
1852 :			shell script code. The code has access to the $VERSION variable.
1853 :			- Made corresponding changes to config/install.sh and config/unpack.
1854 :			- Default contents of config/srcarchiveurl uses $VERSION and
1855 :			normally does not have to be edited to reflect a version change.
1856 :
1857 :			(As a result, a version change can be done by just editing
1858 :			config/version, the rest is now automatic.)
1859 :
1860 :			----------------------------------------------------------------------
1861 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
1862 :	mblume	1669	Date: 2004/10/27 17:50:00 CDT
1863 :			Tag: blume-20041027-btrace-msg
1864 :			Description:
1865 :
1866 :			BackTrace.monitor now also reports the source of the exception that
1867 :			triggered the trace.
1868 :
1869 :			----------------------------------------------------------------------
1870 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
1871 :	mblume	1668	Date: 2004/10/27 17:20:00 CDT
1872 :			Tag: blume-20041027-x86-c-calls
1873 :			Description:
1874 :
1875 :			This is the HISTORY entry for two earlier commits, both concerning
1876 :			the x86 c-calls code in MLRISC:
1877 :
1878 :			- added a missing LOAD in the code that deals with struct arguments
1879 :			- made sure the caller does not add the wrong number of bytes to the
1880 :			stack pointer after a call of a function returning a struct
1881 :			(the callee already pops the implicit argument which points to
1882 :			the space reserved for the result)
1883 :
1884 :			----------------------------------------------------------------------
1885 :	allenleung	1662	Name: Allen Leung (leunga (at) reservoir (dot) com)
1886 :			Date: 2004/10/24 14:00:00 EST
1887 :			Tag: leunga-20041024-x86-gas-fucomip
1888 :			Description:
1889 :
1890 :			John discovered a bug in the syntax of fucomip.
1891 :			The opcodes FU?COMIP? have been changed to
1892 :
1893 :			fu?comip? %st(i), %st
1894 :
1895 :			----------------------------------------------------------------------
1896 :	mblume	1622	Name: Matthias Blume (blume (at) tti - c (dot) org)
1897 :	mblume	1656	Date: 2004/10/20 15:06:00 CDT
1898 :			Tag: blume-20041020-standalone-backtrace
1899 :			Description:
1900 :
1901 :			Added a mechanism for getting back-trace information from standalone
1902 :			programs. Here is how it works:
1903 :
1904 :			1. The part of the program from which you want to get backtrace
1905 :			information (usually the whole program) should be wrapped with
1906 :			BackTrace.monitor. This is a (unit->'a)->'a function, and your
1907 :			main program could be modified from something like
1908 :
1909 :			fun main (pgm, args) = ...
1910 :
1911 :			to
1912 :
1913 :			fun main (pgm, args) = BackTrace.monitor (fn () => ...)
1914 :
1915 :			2. To be able to access BackTrace.monitor, you have to add
1916 :			library $smlnj-tdp/plugins.cm to the .cm file that contains your
1917 :			main function.
1918 :
1919 :			3. Remove all compiled code (i.e., all the .cm/ subdirectories that
1920 :			CM might have created in the past for your project).
1921 :
1922 :			4. Build the system using this command line:
1923 :
1924 :			ml-build -Ctdp.instrument=true \$smlnj-tdp/back-trace.cm \
1925 :			myprog.cm MyProg.main myprog
1926 :
1927 :			instead of the usual
1928 :
1929 :			ml-build myprog.cm MyProg.main myprog
1930 :
1931 :			I changed a library name:
1932 :
1933 :			$/trace-debug-profile.cm --> $smlnj-tdp/plugins.cm
1934 :
1935 :			New libraries:
1936 :
1937 :			$smlnj-tdp/back-trace.cm
1938 :			-- when loaded causes the back-trace plugin to be installed
1939 :			$smlnj-tdp/coverage.cm
1940 :			-- when loaded causes the coverage plugin to be installed
1941 :
1942 :			----------------------------------------------------------------------
1943 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
1944 :	mblume	1655	Date: 2004/10/18 16:45:00 CDT
1945 :			Tag: blume-20041018-groupowner
1946 :			Description:
1947 :
1948 :			Added an "obsolete" warning for the "group owner" syntax to CM's parser.
1949 :
1950 :			Eliminated group owner specs from .cm files throughout the source tree.
1951 :
1952 :			----------------------------------------------------------------------
1953 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
1954 :	mblume	1653	Date: 2004/10/15 15:45:00 CDT
1955 :			Tag: blume-20041015-coverage
1956 :			Description:
1957 :
1958 :			* Test coverage tool added!
1959 :			* Further reorganization of tracing-, debugging-, and profiling support:
1960 :
1961 :			- moved original BTImp -- now called BackTrace -- into a separate
1962 :			library called $/trace-debug-profile.cm
1963 :			- eliminated all mentions of BTrace from SMLofNJ.Internals
1964 :			- only the instrumentation mechanism is now left in the compiler proper
1965 :			- BackTrace module is a plugin which is NOT plugged in by default
1966 :			- Coverage module is another such plugin
1967 :
1968 :			To get the benefits of any of these plugin modules, the code in
1969 :			question must be compiled with tdp instrumentation turned on. This
1970 :			can be done by setting SMLofNJ.Internals.TDP.mode to true.
1971 :			(The ref cell is also controlled via the -Ctdp.instrument=... switch.)
1972 :
1973 :			Plugins are selected at link time. (Pre-compiled instrumented code
1974 :			can be re-loaded with different plugins in effect.) When an
1975 :			instrumented module is linked, whatever plugins are at that time
1976 :			enabled will come into effect for that module.
1977 :
1978 :			To enable the back-trace plugin, load library $/trace-debug-profile.cm
1979 :			and invoke BackTrace.install() (e.g., from the interactive prompt).
1980 :			To enable the coverage plugin, load the same library and invoke
1981 :			Coverage.install().
1982 :
1983 :			Back-traces are generated automatically on uncaught exceptions and
1984 :			when the code in question explicitly invokes BackTrace.trigger().
1985 :
1986 :			Coverage (and execution frequency-) information must be queried
1987 :			explicitly by calling Coverage.not_covered and Coverage.hot_spots.
1988 :
1989 :			----------------------------------------------------------------------
1990 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
1991 :	mblume	1652	Date: 2004/10/14 17:40:00 CDT
1992 :			Tag: blume-20041014-tdp-core
1993 :			Description:
1994 :
1995 :			Snapshot of a significant overhaul of how the trace/debug/profile support
1996 :			is hooked into the system (specifically: Core and SMLofNJ.Internals).
1997 :
1998 :			----------------------------------------------------------------------
1999 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2000 :	mblume	1650	Date: 2004/10/13 16:34:00 CDT
2001 :			Tag: blume-20041013-tdp
2002 :			Description:
2003 :
2004 :			Some rationalization of names:
2005 :
2006 :			structure BTrace -> structure TDPInstrument
2007 :			etc.
2008 :
2009 :			This is is preparation of using the original back-trace
2010 :			instrumentation for other purposes. "TDP" stands for
2011 :			Trace/Debug/Profile.
2012 :
2013 :			The control flag controlling whether instrumentation is on or off is now
2014 :			registered under a different name, so instead of running sml as
2015 :
2016 :			sml -Cinstrument.btrace-mode=true
2017 :
2018 :			one has to say
2019 :
2020 :			sml -Ctdp.instrument=true
2021 :
2022 :			----------------------------------------------------------------------
2023 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2024 :	mblume	1645	Date: 2004/10/11 16:37:00 CDT
2025 :			Tag: blume-20041011-regions
2026 :			Description:
2027 :
2028 :			Made some minor modifications to elabcore.sml to have source regions
2029 :			be propagated more tightly -- resulting in better (i.e., smaller)
2030 :			regions being reported in error- and debug messages.
2031 :
2032 :			----------------------------------------------------------------------
2033 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2034 :	mblume	1644	Date: 2004/10/08 22:50:00 CDT
2035 :			Tag: blume-20041008-cmkw
2036 :			Description:
2037 :
2038 :			Fixed handling of keywords in .cm files: After seeing "is" the lexer
2039 :			treats subsequent occurrences of "group", "library", "source", "is",
2040 :			"*", and "-" as ordinary identifiers rather than keywords.
2041 :
2042 :			Most seriously, this fixes a problem with CM's "shell" tool. The tool
2043 :			is supposed to accept a tool argument called "source", but this did
2044 :			not work because of the clash with the keyword.
2045 :
2046 :			----------------------------------------------------------------------
2047 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2048 :	mblume	1642	Date: 2004/10/07 16:00:00 CDT
2049 :			Tag: blume-20041007-cleanup
2050 :			Description:
2051 :
2052 :			Assorted cleanup work:
2053 :
2054 :			- got rid of intstrmap in favor of using the library's
2055 :			hash table implementation
2056 :			- threw out most of the pathnames stuff, as it was not used anyway
2057 :			- simplified tokentable implementation
2058 :			- fixed some minor spelling errors
2059 :
2060 :			----------------------------------------------------------------------
2061 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2062 :	mblume	1641	Date: 2004/10/06 15:15:15 CDT
2063 :			Tag: blume-20041006-handler
2064 :			Description:
2065 :
2066 :			Cleaned up the absyn to reflect the invariant that HANDLE always
2067 :			carries a FNexp as part of the type definition. This eliminates some
2068 :			superfluous sanity checks at runtime down the road.
2069 :
2070 :			Some minor cleanup of the btrace code.
2071 :
2072 :			----------------------------------------------------------------------
2073 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2074 :	mblume	1640	Date: 2004/10/01 10:20:30 CDT
2075 :			Tag: blume-20041001-slave
2076 :			Description:
2077 :
2078 :			Added hack to make slave mode work in the presence of the version
2079 :			tool. (Still, since the master does two passes over the code for
2080 :			CMB.make, the release number gets bumped twice when slaves are
2081 :			attached. I don't know if this is worth fixing...)
2082 :
2083 :			----------------------------------------------------------------------
2084 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2085 :	mblume	1639	Date: 2004/09/30 10:55:00 CDT
2086 :			Tag: blume-20040930-version
2087 :			Description:
2088 :
2089 :			* Moved the "version" magic into its own little library under
2090 :	mblume	1672	src/system/smlnj/internal. This avoids expensive reconstruction of
2091 :			a stable src/compiler/core.cm.
2092 :	mblume	1639
2093 :	mblume	1672	* At the same time, structure CompilerVersion is now known as
2094 :			structure SMLNJVersion.
2095 :	mblume	1639
2096 :			* Arranged for the version tool to NOT kick in when rebuilding the system
2097 :			(makeml -rebuild, fixpt). Otherwise one would never reach a fixpoint.
2098 :			Also, loading the versiontool does not work when rebuilding the system
2099 :			because CM is not properly initialized at that time.
2100 :
2101 :			----------------------------------------------------------------------
2102 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2103 :	mblume	1638	Date: 2004/09/29 14:00:00 CDT
2104 :			Tag: blume-20040929-autoversion
2105 :			Description:
2106 :
2107 :			Implemented some CM magic to have
2108 :			file src/compiler/TopLevel/main/version.sml
2109 :			generated automagically.
2110 :			The version is taken from two files: config/version and config/release.
2111 :			The first is expected to contain a two-part version number such as 110.49.
2112 :			The second should contain a single number, but it may be missing.
2113 :
2114 :			If the environment variable VERSIONTOOL_BUMP_RELEASE is defined at the
2115 :			time the version tool is loaded (which is the first time you say CMB.make),
2116 :			then the tool will increment the value stored in config/release every
2117 :			time CMB.make is invoked.
2118 :
2119 :			The binfile format is now insensitive to anything beyond the first
2120 :			two components of a version number, so bumping the release does not render
2121 :			binfiles incompatible. Auto-bumping can be used to keep track of versions
2122 :			during development without invalidating existing binfiles.
2123 :
2124 :			In any case, every CMB.make updates the date information in version.sml.
2125 :			(This is the date that is printed in the banner.)
2126 :
2127 :			----------------------------------------------------------------------
2128 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2129 :	mblume	1634	Date: 2004/09/28 10:53:00 CDT
2130 :			Tag: blume-20040928-controls
2131 :			Description:
2132 :
2133 :			Some cleanup of the controls code.
2134 :
2135 :			----------------------------------------------------------------------
2136 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2137 :	mblume	1633	Date: 2004/09/27 22:08:00 CDT
2138 :			Tag: blume-20040927-controls
2139 :			Description:
2140 :
2141 :			Added two pieces of functionality to the Controls interface:
2142 :
2143 :			1. val save'restore: 'a control -> unit -> unit
2144 :
2145 :			grabs the current value of the control in stage 1 and restores it
2146 :			in stage 2.
2147 :
2148 :			2. val set' : 'a control * 'a -> unit -> unit
2149 :
2150 :			stores the given value into the control in stage 2 (i.e., delayed)
2151 :			but does all error checking in stage 1.
2152 :			(This is for string controls that need to do parse their argument --
2153 :			something that might fail. In some cases, notably in CM, one
2154 :			already knows the intended argument but wants to delay the actual
2155 :			assignment until a time when error recovery would be more difficult.)
2156 :
2157 :			Changed the handling of controls in tool arguments to classes "sml" and
2158 :			"lazysml":
2159 :			- use Controls.save'restore as a more robust way of restoring the
2160 :			old value (in particular: without having to re-parse the string)
2161 :			- use controls to handle the "overload" keyword in the init group
2162 :			(I believe this change actually fixes a long-standing obscure bug.)
2163 :
2164 :			----------------------------------------------------------------------
2165 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2166 :	mblume	1632	Date: 2004/09/27 17:00:00 CDT
2167 :			Tag: blume-20040927-lazysml
2168 :			Description:
2169 :
2170 :			Added a new tool class called "lazysml" to CM's tool chest. The only
2171 :			difference to "sml" is that compilation is done with Control.lazysml
2172 :			set to true. A source of class "lazysml" is automatically recognized
2173 :			by a file name suffix of ".lml".
2174 :
2175 :			In addition to the above feature, the original class "sml" now also
2176 :			supports a tool argument "lazy" which has the same effect. As a
2177 :			result, the following three lines are equivalent:
2178 :
2179 :			foo.sml : lazysml
2180 :			foo.sml : sml (lazy)
2181 :			foo.sml (lazy)
2182 :
2183 :			The setting goes into effect both during parsing and during
2184 :			compilation. The original setting is restored right after parsing and
2185 :			after compilation, respectively.
2186 :
2187 :			In addition to all the above, there is also a general mechanism to set
2188 :			ANY of the "controls" that are available at the command line via
2189 :			"-C..." on a per-sml-file basis. The same rules that apply for "lazy"
2190 :			apply as well. (In fact, "lazy" is implemented as a special case of
2191 :			the general mechanism.)
2192 :
2193 :			The .cm file syntax uses a new keyword tool argument called "with".
2194 :			There are several ways of indicating the desired settings:
2195 :
2196 :			foo.sml (with:parser.quotations=true)
2197 :			foo.sml (with:(name:parser.quotations value:true))
2198 :			foo.sml (with:(name:name1 value:value1 name:name2 value:value2 ...))
2199 :			foo.sml (with:(name1=value1 name2=value2 ...))
2200 :			foo.sml (with:(name1=value1 name:name2 value:value2 name3=value3 ...))
2201 :
2202 :			etc.
2203 :
2204 :			Another possible abbreviation is to leave out the =v or value:v part
2205 :			if the name refers to a boolean control (in which case the value is
2206 :			taken to be true). Thus, one could get lazy sml also by saying:
2207 :
2208 :			foo.sml (with:parser.lazy-keyword=true)
2209 :			foo.sml (with:parser.lazy-keyword)
2210 :			foo.sml (with:(name:parser.lazy-keyword value:true))
2211 :			foo.sml (with:(name:parser.lazy-keyword))
2212 :
2213 :			----------------------------------------------------------------------
2214 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2215 :	mblume	1631	Date: 2004/09/24 16:22:00 CDT
2216 :			Tag: blume-20040924-ppc-long-branch
2217 :			Description:
2218 :
2219 :			Turned message about "emiting long form of branch" off by default.
2220 :			Added a control flag to turn it back on when desired.
2221 :
2222 :			----------------------------------------------------------------------
2223 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2224 :	mblume	1630	Date: 2004/09/24 16:05:00 CDT
2225 :			Tag: blume-20040924-rounding
2226 :			Description:
2227 :
2228 :			Applied patch for setting rounding modes under Mac OS X. Thanks to
2229 :			Melissa O'Neill for providing the code!
2230 :
2231 :			----------------------------------------------------------------------
2232 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2233 :	mblume	1629	Date: 2004/09/23 17:30:00 CDT
2234 :			Tag: blume-20040923-envvars
2235 :			Description:
2236 :
2237 :			1. Changed definition of type ControlRegistry.registry_tree to
2238 :			include control_info (i.e., the name of the controlling environment
2239 :			variable).
2240 :
2241 :			2. Added command-line flags -e and -E to print the names of environment
2242 :			variables that can be used to control internal settings.
2243 :			(This uses the new API mentioned in 1.)
2244 :
2245 :			----------------------------------------------------------------------
2246 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2247 :	mblume	1622	Date: 2004/09/13 16:50:00 CDT
2248 :			Tag: Release_110_49
2249 :			Description:
2250 :
2251 :			New working version (110.49). NEW BOOTFILES!
2252 :
2253 :			----------------------------------------------------------------------
2254 :	mblume	1621	Name: Matthias Blume
2255 :			Date: 2004/09/13 16:20:00 CDT
2256 :			Tag: blume-20040913-config-mlrisc
2257 :			Description:
2258 :
2259 :			Put target "mlrisc" back into the default list.
2260 :			(There is no harm in having it, and some users have expressed their
2261 :			wish to have "mlrisc" included by default.)
2262 :
2263 :			----------------------------------------------------------------------
2264 :	jhr	1620	Name: John Reppy
2265 :			Date: 2004/09/13
2266 :			Tag: jhr-20040913-signals
2267 :			Description:
2268 :
2269 :			Fixed the signal masking code to properly nest mask/unmask operations
2270 :			on a per-signal basis.
2271 :
2272 :			----------------------------------------------------------------------
2273 :	mblume	1619	Name: Matthias Blume (blume (at) tti - c (dot) org)
2274 :			Date: 2004/09/08 13:20:00 CDT
2275 :			Tag: blume-20040908-heap-magic
2276 :			Description:
2277 :
2278 :			Bumped the heap macig number to 0x09082004 to account for the changed
2279 :			layout of the ML frame under MacOS X.
2280 :
2281 :			----------------------------------------------------------------------
2282 :	allenleung	1618	Name: Allen Leung (leunga (at) reservoir (dot) com)
2283 :			Date: 2004/09/03 11:26:00 EST
2284 :			Tag: leunga-20040903-cygwin-install
2285 :			Description:
2286 :
2287 :			Added a patch to _arch-n-opsys to enable the Cygwin runtime.
2288 :			The Cygwin runtime is turned on by setting the environment
2289 :			variable SMLNJ_CYGWIN_RUNTIME to 1.
2290 :
2291 :			----------------------------------------------------------------------
2292 :	mblume	1615	Name: Matthias Blume (blume (at) tti - c (dot) org)
2293 :	mblume	1617	Date: 2004/08/31 17:15:00 CDT
2294 :			Tag: blume-20040831-core
2295 :			Description:
2296 :
2297 :			Added some exports to src/compiler/core.cm upon request by J. Joemann.
2298 :
2299 :			----------------------------------------------------------------------
2300 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2301 :	mblume	1616	Date: 2004/08/30 17:55:00 CDT
2302 :			Tag: blume-20040830-installer
2303 :			Description:
2304 :
2305 :			Upon request by Johannes Joemann:
2306 :
2307 :			- improved ML code of installer to fall back to coping when renaming
2308 :			fails (i.e., when source and target are on different file systems);
2309 :			the code compiles but has yet to be tested in anger
2310 :			- removed mlrisc from list of default targets (config/targets)
2311 :
2312 :			----------------------------------------------------------------------
2313 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2314 :	mblume	1615	Date: 2004/08/27 17:20:00 CDT
2315 :			Tag: blume-20040827-ptreql
2316 :			Description:
2317 :
2318 :			Added ptreql primop to structure InlineT (upon request from Larry
2319 :			Paulson).
2320 :
2321 :			----------------------------------------------------------------------
2322 :	allenleung	1613	Name: Allen Leung (leunga (at) reservoir (dot) com)
2323 :	allenleung	1614	Date: 2004/08/15 21:21:00 EST
2324 :			Tag: leunga-110_48-udgraph
2325 :			Description:
2326 :
2327 :			Another bug fix from Carl Hauser:
2328 :
2329 :			diff /net/niflab/smlnj48/src/MLRISC/graphs/udgraph.sml udgraph.sml
2330 :			> 48c48
2331 :			> < | rmv((e as (k,_))::es,L) = rmv(es,if k = i then es else
2332 :			> e::L)
2333 :			> ---
2334 :			> > | rmv((e as (k,_))::es,L) = rmv(es,if k = i then L else e::L)
2335 :			> Without this, any deletion of an edge in an undirected graph does severe
2336 :			> violence to the graph.
2337 :
2338 :			----------------------------------------------------------------------
2339 :			Name: Allen Leung (leunga (at) reservoir (dot) com)
2340 :	allenleung	1613	Date: 2004/08/10 23:35:00 EST
2341 :			Tag: leunga-110_48-ppc
2342 :			Description:
2343 :
2344 :			The IBM/MacOS syntax switch on PPC was incorrectly swapped.
2345 :
2346 :			----------------------------------------------------------------------
2347 :	mblume	1602	Name: Matthias Blume (blume (at) tti - c (dot) org)
2348 :	mblume	1615	Date: 2004/08/10 12:00:00 CDT
2349 :	mblume	1602	Tag: Release_110_48
2350 :			Description:
2351 :
2352 :			New working version (110.48). NEW BOOTFILES!
2353 :
2354 :			----------------------------------------------------------------------
2355 :	allenleung	1599	Name: Allen Leung (leunga (at) reservoir (dot) com)
2356 :			Date: 2004/08/09 12:21:00 EST
2357 :			Tag: leunga-110_47-dijsktra
2358 :			Description:
2359 :
2360 :			Bug fix from Carl Hauser:
2361 :
2362 :	blume	2155	single_source_shortest_paths in dijkstra.sml was observed to get wrong
2363 :	allenleung	1599	answers (by comparing to single_source_shortest_paths in bellman-ford.sml).
2364 :
2365 :	blume	2155	The problem is that following the expression A.update(dist,s,Num.zero)
2366 :	allenleung	1599	it is necessary to update the priority queue using Q.decreaseWeight(Q,s).
2367 :
2368 :			----------------------------------------------------------------------
2369 :	mblume	1595	Name: Matthias Blume (blume (at) tti - c (dot) org)
2370 :			Date: 2004/08/06 18:10:00 CDT
2371 :			Tag: blume-20040806-cmdline
2372 :			Description:
2373 :
2374 :			Fiddled with handling of command-line options:
2375 :
2376 :			* sml now quits after processing the command line
2377 :			if -H, -S, -h<n>, or -s<n> appears as the last
2378 :			command-line argument
2379 :			* a new option -q terminates the session when encountered on
2380 :			the command line; subsequent arguments will be ignored
2381 :			* bug fixes: short (erroneous) arguments are no longer ignored
2382 :			completely
2383 :
2384 :			----------------------------------------------------------------------
2385 :	allenleung	1593	Name: Allen Leung (leunga (at) reservoir (dot) com)
2386 :			Date: 2004/08/04 18:17:00 EST
2387 :			Tag: leunga-110_47-ppc-ibm-asm
2388 :			Description:
2389 :
2390 :			- Added minimal IBM assembly syntax support for PowerPC.
2391 :
2392 :	blume	2155	- Cygwin: manually changed the file cygwin.def. Some exported symbols have
2393 :	allenleung	1593	been altered in the runtime. We need an automatic way to keep the file
2394 :			in sync.
2395 :
2396 :			----------------------------------------------------------------------
2397 :	mblume	1548	Name: Matthias Blume (blume (at) tti - c (dot) org)
2398 :	mblume	1581	Date: 2004/08/04 14:00:00 CDT
2399 :			Tag: Release_110_47
2400 :			Description:
2401 :
2402 :			New working version (110.47). NEW BOOTFILES!
2403 :
2404 :			----------------------------------------------------------------------
2405 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2406 :	mblume	1580	Date: 2004/08/03 14:25:00 CDT
2407 :			Tag: blume-20040803-callingconv
2408 :			Description:
2409 :
2410 :			Added low-level support for choosing C calling conventions by
2411 :			twiddling the type of rawccall. (See
2412 :			src/compiler/Semant/types/cproto.sml for details.)
2413 :
2414 :			----------------------------------------------------------------------
2415 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2416 :	mblume	1579	Date: 2004/08/02 15:55:00 CDT
2417 :			Tag: blume-20040802-backout
2418 :			Description:
2419 :
2420 :			Backed out of change to win32-filesys.c. The earlier patch to
2421 :			get_file_time caused CM to produce files with the wrong time stamp.
2422 :
2423 :			----------------------------------------------------------------------
2424 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2425 :	mblume	1573	Date: 2004/08/02 14:45:00 CDT
2426 :			Tag: blume-20040802-nlffi-win32
2427 :			Description:
2428 :
2429 :			Added NLFFI support for Win32, adapted from a patch provided by David
2430 :			Hansel. This is currently completely untested. Also, the issue
2431 :			concerning stdcall vs. ccall is still unresolved.
2432 :
2433 :			----------------------------------------------------------------------
2434 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2435 :	mblume	1572	Date: 2004/07/30 17:55:00 CDT
2436 :			Tag: blume-20040730-various
2437 :			Description:
2438 :
2439 :			Gearing up towards 110.47...
2440 :
2441 :			- various minor bugfixes to ml-nlffigen
2442 :			- a beginning of a manual for nlffi
2443 :
2444 :			- eliminated 'export name=value' in config/install.sh as this does
2445 :			not work with certain versions of /bin/sh
2446 :			(Thanks to David King at Motorola for catching this.)
2447 :
2448 :			- several bugfixes provided or suggested by David Hansel at Reactive Systems:
2449 :			- added a test for tm==NULL to gmtime.c and localtime.c
2450 :			- applied patch for incorrect GetFileTime under win32
2451 :			- toSeconds -> toMilliseconds in Win32/win32-process.sml
2452 :
2453 :			----------------------------------------------------------------------
2454 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2455 :	mblume	1557	Date: 2004/07/21 18:20:00 CDT
2456 :			Tag: blume-20040721-nlffigen
2457 :			Description:
2458 :
2459 :			- Fixed minor issue in ml-nlffigen: Now generate structure T_foo
2460 :			for a typedef to an incomplete type, but leave out the "typ" member.
2461 :			(This is just for consistency.)
2462 :			- Started to produce what is supposed to become better (i.e., comprehensive)
2463 :			documentation of what ml-nlffigen does and produces.
2464 :
2465 :			----------------------------------------------------------------------
2466 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2467 :	mblume	1548	Date: 2004/07/14 16:25:00 CDT
2468 :			Tag: blume-20040714-union
2469 :			Description:
2470 :
2471 :			Added C_UNION to c-calls/c-types.sml and updated the machinery
2472 :			(ml-nlffigen, cproto.sml) that conveys C function interface
2473 :			information to the code generator.
2474 :
2475 :			However, the actual architecture-specific implementation of function
2476 :			arguments and results that are C unions is still not implemented.
2477 :
2478 :			----------------------------------------------------------------------
2479 :	allenleung	1539	Name: Allen Leung (leunga (at) reservoir (dot) com)
2480 :	allenleung	1546	Date: 2004/07/14 14:38:00 EST
2481 :			Tag: leunga-110_46_1-ppc-lwzu
2482 :			Description:
2483 :
2484 :			Added these instructions to the PowerPC architecture:
2485 :			LBZU(X), LHZU(X), LWZU(X),
2486 :	blume	2155	STWU(X), STFDU, STFSU
2487 :
2488 :	allenleung	1546	etc...
2489 :
2490 :			Note: I haven't added their instruction encoding into the description.
2491 :
2492 :			----------------------------------------------------------------------
2493 :			Name: Allen Leung (leunga (at) reservoir (dot) com)
2494 :	allenleung	1539	Date: 2004/07/13 15:04:00 EST
2495 :			Tag: leunga-110_46_1-ppc-lwarx
2496 :			Description:
2497 :
2498 :			Added the two instructions LWARX and STWCX to the PowerPC
2499 :	blume	2155	instruction set.
2500 :	allenleung	1539
2501 :			A (untested) rewrite of loop-structure.sml. The old version
2502 :			is completely broken.
2503 :
2504 :			----------------------------------------------------------------------
2505 :	mblume	1479	Name: Matthias Blume (blume (at) tti - c (dot) org)
2506 :	mblume	1537	Date: 2004/07/13 13:50:00 CDT
2507 :			Tag: blume-20040713-nlffi
2508 :			Description:
2509 :
2510 :			- use paramAlloc to report c-calls with too many arguments
2511 :			(for PPC version where parameter area is pre-allocated)
2512 :			- added ccall_maxargspace to machspec (to implement the above)
2513 :			- made "make" commend in CM's "make" tool configurable
2514 :			- added option (default: on) for passing the name of the SML/NJ's "bin"
2515 :			directory to "make"; the call looks like this:
2516 :
2517 :			make <options> SMLNJ_BINDIR=<dir> <target>
2518 :
2519 :			This can be used by the Makefile to, e.g., pick the "right" version
2520 :			of ml-nlffigen.
2521 :			- minor code tweaks
2522 :
2523 :			----------------------------------------------------------------------
2524 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2525 :	mblume	1532	Date: 2004/07/12 22:50:00 CDT
2526 :			Tag: blume-110_46_1-macosx-nlffi
2527 :			Description:
2528 :
2529 :			NLFFI under Mac OS X now working (sort of). This is largely untested,
2530 :			though.
2531 :
2532 :			Note: 1. You have to make a new, clean build of the runtime system.
2533 :			2. There are new BOOTFILES, you have to use them!
2534 :			(Doing the bootstrap process yourself would be *very* painful!
2535 :			If you absolutely have to do it, build the system under
2536 :			a different architecture and then cross-compile.)
2537 :
2538 :			Version bumped to 110.46.1 to account for runtime data format changes.
2539 :
2540 :			----------------------------------------------------------------------
2541 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2542 :	mblume	1515	Date: 2004/06/18 14:30:00 CDT
2543 :			Tag: blume-20040618-unix
2544 :			Description:
2545 :
2546 :			Changed the implementation of structure Unix so that the same stream
2547 :			is returned every time one of the {text,bin}{In,Out}streamOf functions
2548 :			is invoked on the same proc. This is not what the spec currently
2549 :			says -- although IMO it arguably should. (See discussion below.)
2550 :
2551 :			----------------------------------------------------------------------
2552 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2553 :	mblume	1507	Date: 2004/06/17 18:15:00 CDT
2554 :			Tag: Release_110_46
2555 :			Description:
2556 :
2557 :			New working version (110.46). NEW BOOTFILES!
2558 :
2559 :			----------------------------------------------------------------------
2560 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2561 :	mblume	1506	Date: 2004/06/17 17:20:00 CDT
2562 :			Tag: blume-20040617-timer-unix
2563 :			Description:
2564 :
2565 :			Changed the interface of structures Timer and Unix to match the most
2566 :			recent Basis spec.
2567 :
2568 :			In the case of Unix there still seems to be an open/weird issue:
2569 :
2570 :			The {text,bin}{In,Out}streamOf functions are supposed to create
2571 :			fresh streams whenever they are called -- as opposed to have them
2572 :			return the same stream every time. This design is supposed to
2573 :			prevent space leaks caused by proc values hanging on to streams.
2574 :
2575 :			The reap function, on the other hand, is supposed to close the
2576 :			streams. This cannot be done without having a handle on the
2577 :			stream in proc after all...
2578 :
2579 :			I took the liberty to implement the following stopgap solution:
2580 :
2581 :			The proc value hangs on to the most recently created stream(s).
2582 :			Reap closes those. If either or both of the two streams hadn't
2583 :			been created at all yet, then reap will close the corresponding
2584 :			file descriptors directly.
2585 :
2586 :			PS: I don't understand the original space leak argument anymore. If
2587 :			a proc hangs on to the imperative stream, then I/O operations on those
2588 :			will advance the state of the cached stream and avoid the space leak.
2589 :
2590 :			----------------------------------------------------------------------
2591 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2592 :	mblume	1500	Date: 2004/05/28 16:45:00 CDT
2593 :			Tag: blume-20040528-basis
2594 :			Description:
2595 :
2596 :			Added signature PACK_REAL and exported functor PrimIO.
2597 :
2598 :			----------------------------------------------------------------------
2599 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2600 :	mblume	1498	Date: 2004/05/25 16:00:00 CDT
2601 :			Tag: blume-20040525-group-owner
2602 :			Description:
2603 :
2604 :			CM now ignores (but still accepts) the "owner" information in group
2605 :			descriptions. The owner of a group is its next enclosing
2606 :			library. Each group must have a unique owner. (There is a virtual
2607 :	mblume	1505	"toplevel" library that owns groups which are not nested within a real
2608 :	mblume	1498	library.) Previously, each group had to explicitly declare its owner,
2609 :			and CM would check that such a declaration is correct. The new scheme
2610 :			is to have CM check that for each group there is precisely one owning
2611 :			library.
2612 :
2613 :			The advantage of the new scheme is that the programmer no longer needs
2614 :			to maintain the somewhat annoying owner information. The downside is
2615 :			that CM cannot enforce the ownership rule across multiple runs of
2616 :			CM.make. Fortunately, enclosing the same group in two different
2617 :			libraries A and B which are not part of the same program does not
2618 :			cause real problems.
2619 :
2620 :			----------------------------------------------------------------------
2621 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2622 :	mblume	1496	Date: 2004/05/20 16:00:00 CDT
2623 :			Tag: blume-20040520-win32
2624 :			Description:
2625 :
2626 :			Made the win32 version work again. (Strangely, a misplaced comma had
2627 :			slipped into win32-process.c which prevented the runtime from being
2628 :			compiled correctly.)
2629 :
2630 :			Also, included a minor addition to ml-build.bat analogous to what was
2631 :			done in blume-20040519-ml-build.
2632 :
2633 :			----------------------------------------------------------------------
2634 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2635 :	mblume	1494	Date: 2004/05/19 22:10:00 CDT
2636 :			Tag: blume-20040519-ml-build
2637 :			Description:
2638 :
2639 :			Arranged for ml-build to clean up after itself a little bit better.
2640 :			The script generates a temporary SML source file and compiles it using
2641 :			CM, so CM generates metadata (GUID, SKEL, objectfile) for it. It now
2642 :			gets rid of those at the end, so they don't accumulate under .cm.
2643 :
2644 :			This required a minor change to install.sh because the name of the
2645 :			metadata directory (default: .cm) is actually configurable at
2646 :			installation time.
2647 :
2648 :			----------------------------------------------------------------------
2649 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2650 :	mblume	1491	Date: 2004/05/18 15:50:00 CDT
2651 :			Tag: blume-20040518-mkreader
2652 :			Description:
2653 :
2654 :			Added Posix.IO.mk{Bin,Text}{Reader,Writer} by lifting their respective
2655 :			implementations from internal modules PosixBinPrimIO and PosixTextPrimIO.
2656 :
2657 :			----------------------------------------------------------------------
2658 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2659 :	mblume	1489	Date: 2004/05/11 14:35:00 CDT
2660 :			Tag: blume-20040511-win32sock
2661 :			Description:
2662 :
2663 :			Added previously missing support for many socket-related functions
2664 :			under win32. Thanks to David Hansel <hansel@reactive-systems.com>
2665 :			for the voluminous patch!
2666 :
2667 :			(I have not tested this patch under win32 yet.)
2668 :
2669 :	mblume	1490	Here is David's e-mail:
2670 :
2671 :			Hi,
2672 :
2673 :			Attached to this email you find a diff against sml/nj 110.45
2674 :			that will enable socket support under Windows.
2675 :
2676 :			To apply the patch (using unix or cygwin)
2677 :			1) gunzip runtime.diff.gz
2678 :			2) "cd" into "src/runtime" in the source tree of a fresh
2679 :			110.45 installation.
2680 :			3) patch -p 1 < [your/path/to]runtime.diff
2681 :
2682 :	blume	2155	The code compiles fine but has NOT yet been extensively tested.
2683 :			I only ran a few tests for basic socket client functionality
2684 :			(which worked fine). Especially the functions that use ioctl
2685 :	mblume	1490	are not tested at all and might not work (see below).
2686 :
2687 :			I implemented this since we want to move to a newer version of sml/nj
2688 :			but need socket support in order to use it. This is the first time I
2689 :			even had a look at the sml/nj source, so please review my changes
2690 :			before making this part of the distribution! Here are a few issues
2691 :	blume	2155	that I think might be better for someone to solve who is more
2692 :	mblume	1490	familiar with the sml/nj source (and socket programming):
2693 :
2694 :			- getnetbyaddr.c and getnetbyname.c will raise a "not implemented"
2695 :			exception since I could not figure out what the windows equivalent
2696 :			of these functions is
2697 :
2698 :			- In sockets-osdep.h there are a some #include statements that are
2699 :			only used in a few files that include sockets-osdep.h
2700 :
2701 :			- In smlnj-sock-lib.c, function init_fn() calls WSAStartup() but
2702 :			does not process its return value since I don't know how to
2703 :			report an error upwards.
2704 :
2705 :			- It would probably be good to have a call to WSACleanup() when
2706 :	blume	2155	the library is unloaded (if there is such a possibility).
2707 :	mblume	1490	Otherwise I think Windows will take care of this automatically
2708 :			when the process finishes.
2709 :
2710 :			- I used ioctlsocket() as a replacement for ioctl() but I have
2711 :			no idea if that is actually the proper replacement on Windows.
2712 :
2713 :			- All these issues are marked in the code by "FIXME" comments.
2714 :
2715 :	blume	2155	We use sml/nj extensively in our products and are quite happy
2716 :			with it. I hope this contribution will help you.
2717 :	mblume	1490
2718 :			Keep up the good work!
2719 :
2720 :			David
2721 :
2722 :	mblume	1489	----------------------------------------------------------------------
2723 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2724 :	mblume	1488	Date: 2004/05/11 14:20:00 CDT
2725 :			Tag: blume-20040511-installml
2726 :			Description:
2727 :
2728 :			Fixed two bugs in installml script. (Thanks to Vesa A. Norrman for
2729 :			the patch.)
2730 :
2731 :			----------------------------------------------------------------------
2732 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2733 :	mblume	1487	Date: 2004/05/11 14:05:00 CDT
2734 :			Tag: blume-20040511-nlffi-netbsd
2735 :			Description:
2736 :
2737 :			Added support for nlffi under netbsd. (Thanks to Vesa A. Norrman for
2738 :			the patch.)
2739 :
2740 :			----------------------------------------------------------------------
2741 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2742 :	mblume	1486	Date: 2004/05/11 12:05:00 CDT
2743 :			Tag: blume-20040511-exports
2744 :			Description:
2745 :
2746 :			As per request by Adam Chlipala <adam@hcoop.net>, extended various
2747 :			export lists in compiler-related .cm-files.
2748 :
2749 :			----------------------------------------------------------------------
2750 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2751 :	mblume	1485	Date: 2004/05/11 11:35:00 CDT
2752 :			Tag: blume-20040511-allsource
2753 :			Description:
2754 :
2755 :			The installer now honors the "src-smlnj" target again, although its meaning
2756 :			has changed from "all sources required for the compiler" to "all sources
2757 :			the installer knows about". In other words, if you enable "src-smlnj"
2758 :			in the "targets" file, then the installer will pull in sources for
2759 :			everything. (Notice that this refers to source code only. Compiled
2760 :			code is still only installed for modules that were requested explicitly
2761 :			or which are required for other modules that were requested explicitly.)
2762 :
2763 :			----------------------------------------------------------------------
2764 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2765 :	mblume	1482	Date: 2004/04/23 17:40:00 CDT
2766 :			Tag: blume-20040423-ieee-scan
2767 :			Description:
2768 :
2769 :			Fixed IEEEReal.scan (and .fromString) so that if there is an overflow
2770 :			in the exponent calculation we get INF or ZERO (depending on the mantissa
2771 :			and the sign of the exponent).
2772 :
2773 :			----------------------------------------------------------------------
2774 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2775 :	mblume	1480	Date: 2004/04/23 10:40:00 CDT
2776 :			Tag: blume-20040423-ml-build
2777 :			Description:
2778 :
2779 :			The ml-build script now terminates with a non-0 status when something
2780 :			goes wrong.
2781 :
2782 :			----------------------------------------------------------------------
2783 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2784 :	mblume	1479	Date: 2004/04/22 16:35:00 CDT
2785 :			Tag: blume-20040422-Option
2786 :			Description:
2787 :
2788 :			Made exception Option to be the same as exception Option.Option
2789 :			(as it should be).
2790 :
2791 :			----------------------------------------------------------------------
2792 :	allenleung	1468	Name: Allen Leung (leunga (at) reservoir (dot) com)
2793 :			Date: 2004/03/19 14:40:00 EST
2794 :			Tag: leunga-20040319-cygwin-nlffi
2795 :			Description:
2796 :
2797 :			Fixed the runtime so that ml-nlffi-lib runs on the cygwin version
2798 :			of SML/NJ. The problem is that
2799 :
2800 :			lib = dlopen(NULL, ...)
2801 :			f = dlsym(lib, "malloc");
2802 :
2803 :	blume	2155	does not work on Windows unless we explicitly export symbols
2804 :			such as 'malloc' during linking. We fixed this by explicitly
2805 :	allenleung	1468	exporting the required symbols with the magic gcc incantation:
2806 :
2807 :			-Wl,--export-all cygwin.def
2808 :
2809 :			where cygwin.def is a file containing all the symbols that
2810 :	blume	2155	we wish to export.
2811 :	allenleung	1468
2812 :	blume	2155	I suspect this is a Windows problem and we'll have to
2813 :			do the same (somehow with windows compilers) when
2814 :	allenleung	1468	we build the native win32 version with the system
2815 :	blume	2155	calls LoadLibrary/GetProcAddress.
2816 :	allenleung	1468
2817 :			----------------------------------------------------------------------
2818 :	mblume	1384	Name: Matthias Blume (blume (at) tti - c (dot) org)
2819 :	mblume	1465	Date: 2004/03/04 16:35:00 CST
2820 :			Tag: blume-20040304-intinf-fmt
2821 :			Description:
2822 :
2823 :			Fixed problem with IntInf.fmt (sign would show up on the right instead
2824 :			of on the left for BIN, OCT, and HEX).
2825 :
2826 :			----------------------------------------------------------------------
2827 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2828 :	mblume	1464	Date: 2004/03/04 11:25:00 CST
2829 :			Tag: blume-20040304-symlinks
2830 :			Description:
2831 :
2832 :			Fixed problem with installer script (unix only) where bin/ml-yacc and
2833 :			friends pointed (via symlinks) to absolute locations instead of just
2834 :			.run-sml. This was reported by Vesa A Norrman.
2835 :
2836 :			----------------------------------------------------------------------
2837 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2838 :	mblume	1455	Date: 2004/02/13 14:50:00 CST
2839 :			Tag: Release_110_45
2840 :			Description:
2841 :
2842 :			New working version (110.45). New bootfiles.
2843 :
2844 :			----------------------------------------------------------------------
2845 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2846 :	mblume	1448	Date: 2004/01/26 15:15:15 CST
2847 :			Tag: blume-20040126-toplevel
2848 :			Description:
2849 :
2850 :			Improved handling of exceptions at the interactive toplevel.
2851 :
2852 :			----------------------------------------------------------------------
2853 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2854 :			Date: 2004/01/26 11:25:00
2855 :	mblume	1446	Tag: blume-20040126-app
2856 :			Description:
2857 :
2858 :			Type of top-level "app" corrected.
2859 :	mblume	1447	Added code for setting vp_limitPtrMask to Win32-specific runtime.
2860 :	mblume	1446
2861 :			----------------------------------------------------------------------
2862 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2863 :	mblume	1427	Date: 2003/11/18 17:10 CST
2864 :			Tag: blume-20031118-basis-fiddle
2865 :			Description:
2866 :
2867 :			- changed Timer interface to what might become the spec
2868 :			- POSIX_FLAGS -> BIT_FLAGS according to spec
2869 :			- some other minor discrepancies wrt. spec eliminated
2870 :
2871 :			----------------------------------------------------------------------
2872 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2873 :	mblume	1414	Date: 2003/11/06 12:00:00 CST
2874 :			Tag: Release_110_44
2875 :			Description:
2876 :
2877 :			New working version (110.44). New bootfiles.
2878 :
2879 :			----------------------------------------------------------------------
2880 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2881 :	mblume	1411	Date: 2003/11/04 11:50:00 CST
2882 :			Tag: blume-20031104-move-libraries
2883 :			Description:
2884 :
2885 :			Eliminated the "dont_move_libraries" directive in config/targets.
2886 :			(The mechanism was broken and could not be fixed easily. Moreover,
2887 :			there does not seem to be any reason not to move all libraries into
2888 :			lib during installation. I originally implemented this directive as a
2889 :			backward-compatibility feature when I first introduced the new CM.
2890 :			Now that things have been stable for a long time and going back to the
2891 :			old CM is not an option, there is no reason to keep it around.)
2892 :
2893 :			----------------------------------------------------------------------
2894 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2895 :	mblume	1409	Date: 2003/11/03 16:00:00 CST
2896 :			Tag: blume-20031103-installdir
2897 :			Description:
2898 :
2899 :			Made installer honor INSTALLDIR variable again. (Thanks to Chris
2900 :			Richards for pointing out the problem and providing the solution.)
2901 :
2902 :			----------------------------------------------------------------------
2903 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2904 :	mblume	1400	Date: 2003/10/01 17:05:00 CDT
2905 :			Tag: blume-20031001-lal-mlrisc
2906 :			Description:
2907 :
2908 :			MLRISC bug fix from Lal.
2909 :
2910 :			----------------------------------------------------------------------
2911 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2912 :	mblume	1399	Date: 2003/09/30 16:10:00 CDT
2913 :			Tag: blume-20030930-primio-bat
2914 :	mblume	1398	Description:
2915 :
2916 :			1. Added openVector, nullRd, and nullWr to PRIM_IO.
2917 :			2. Improved .bat files (for Win32 port) to make things work under Win95.
2918 :			(thanks to Aaron S. Hawley for this one)
2919 :
2920 :			----------------------------------------------------------------------
2921 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2922 :	mblume	1395	Date: 2003/09/26 16:05:00 CDT
2923 :			Tag: blume-20030926-wrappriv
2924 :			Description:
2925 :
2926 :			Added missing wrapper for privilege "primitive" in $smlnj/viscomp/core.cm.
2927 :
2928 :			----------------------------------------------------------------------
2929 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2930 :	mblume	1394	Date: 2003/09/26 15:00:00 CDT
2931 :			Tag: blume-20030926-110_43_3
2932 :			Description:
2933 :
2934 :			- additional cleanup
2935 :			- version number bump, NEW BOOTFILES
2936 :
2937 :			----------------------------------------------------------------------
2938 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2939 :	mblume	1393	Date: 2003/09/26 12:00:00 CDT
2940 :			Tag: blume-20030926-ppautoload
2941 :			Description:
2942 :
2943 :			I modified the read-eval-print loop so that the autoloader gets
2944 :			invoked whenever the prettyprinter tries to look up a symbol that
2945 :			is not currently defined in the toplevel environment but which
2946 :			appears in CM's autoload registry. As a result, we see far fewer of
2947 :			those ?.Foo.Bar.xxx names in the prettyprinter's output.
2948 :
2949 :			In addition to this I tried to clean up some pieces of the Basis
2950 :			implementation (e.g., Socket, Word8Array) in order to prevent other
2951 :			instances of these ?.Foo.Bar.xxx names from being printed.
2952 :
2953 :			The mechanism that picks names for types still needs some work, though.
2954 :			(Right now it seems that if there is a type A.t which is defined to
2955 :	mblume	1394	be B.u, but B is unavailable at toplevel, then A.t gets printed as
2956 :			"?.B.u" although the perhaps more sensible solution would be to use
2957 :	mblume	1393	"A.t" in this case. In other words, the prettyprinter should follow
2958 :			a chain of DEFtycs not farther than there are corresponding toplevel
2959 :			names in the current environment.)
2960 :
2961 :			----------------------------------------------------------------------
2962 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2963 :	mblume	1391	Date: 2003/09/24 16:31:00 CDT
2964 :			Tag: blume-20030924-installer
2965 :			Description:
2966 :
2967 :			Another installer tweak: All the ML code for the installer is now
2968 :			compiled during CMB.make and put into a little library called
2969 :			$smlnj/installer.cm. The installation then simply invokes
2970 :
2971 :			sml -m $smlnj/installer.cm
2972 :
2973 :			and everything happens automagically.
2974 :
2975 :			Win32: ML code senses value of environment variable SMLNJ_HOME.
2976 :			Unix: ML code senses values of environment variables ROOT, CONFIGDIR,
2977 :			and BINDIR.
2978 :
2979 :			The new scheme guarantees that the ML code responsible for the installation
2980 :			is in sync with the APIs of the main system. Also, the installer is
2981 :			somewhat faster because the installer script is precompiled.
2982 :
2983 :			----------------------------------------------------------------------
2984 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2985 :	mblume	1390	Date: 2003/09/24 15:35:00 CDT
2986 :			Tag: blume-20030924-synsock
2987 :			Description:
2988 :
2989 :			Added a signature SYNCHRONOUS_SOCKET to basis.cm. This is like SOCKET
2990 :			but excludes all non-blocking operations. Defined SOCKET (in Basis)
2991 :			and CML_SOCKET in terms of SYNCHRONOUS_SOCKET. Removed superfluous
2992 :			implementations of non-blocking operations from CML's Socket
2993 :			structure.
2994 :
2995 :			----------------------------------------------------------------------
2996 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
2997 :	mblume	1389	Date: 2003/09/24 15:10:05 CDT
2998 :			Tag: blume-20030924-sockets
2999 :			Description:
3000 :
3001 :			1. Fixed SOCKET API and implementation to match Basis spec.
3002 :			This required changing the internal representation of sockets to one
3003 :			that remembers (for each socket file descriptor) whether it is currently
3004 :			blocking or non-blocking. This state is maintained lazily (i.e., a system
3005 :			call is made only if the state actually needs to change).
3006 :
3007 :			2. OS-specific details of sockets were moved into separate files, thus
3008 :			making it possible to unify the bulk of the socket implementations
3009 :			between Unix and Win32.
3010 :
3011 :			3. CML's socket API changed accordingly.
3012 :			(Note that we need to remove non-blocking functions from this API
3013 :			since they are redundant in the case of CML!)
3014 :
3015 :			4. CML's socket implementation now makes use of non-blocking functions
3016 :			provided by Basis, thus removing all OS-dependent code from this part
3017 :			of CML.
3018 :
3019 :			5. Changed Real64.precision from 52 to 53. Minor cleanup in Real64 code.
3020 :
3021 :			----------------------------------------------------------------------
3022 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
3023 :	mblume	1387	Date: 2003/09/22 12:10:00 CDT
3024 :	mblume	1388	Tag: blume-20030922-110_43_2
3025 :	mblume	1387	Description:
3026 :
3027 :			Made a new interim version and bootfiles for developer's bootstrapping
3028 :			convenience.
3029 :
3030 :			110.43.2 -- NEW BOOTFILES
3031 :
3032 :			----------------------------------------------------------------------
3033 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
3034 :	mblume	1385	Date: 2003/09/19 15:55:00 CDT
3035 :			Tag: blume-20030919-cmdir
3036 :			Description:
3037 :
3038 :			1. new-install.sh -> install.sh
3039 :			2. changed default CM "metadata" directory name to ".cm" (instead of "CM")
3040 :			3. tweaked installer so that another name instead of .cm can be chosen
3041 :			at install time (by setting the CM_DIR_ARC environment variable
3042 :			during installation); once installation is complete, the name is
3043 :			fixed
3044 :
3045 :			----------------------------------------------------------------------
3046 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
3047 :	mblume	1384	Date: 2003/09/18 16:00:00 CDT
3048 :			Tag: blume-20030918-110_43_1
3049 :			Description:
3050 :
3051 :			Made a new interim version and bootfiles for developer's bootstrapping
3052 :			convenience.
3053 :
3054 :			110.43.1 -- NEW BOOTFILES
3055 :
3056 :			----------------------------------------------------------------------
3057 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
3058 :	mblume	1383	Date: 2003/09/18 15:20:00 CDT
3059 :			Tag: blume-20030918-misc
3060 :			Description:
3061 :
3062 :			1. Exported fractionsPerSecond etc. from TimeImp (but not from Time as
3063 :			this seems to be controversial at the moment) and used those in
3064 :			Posix.ProcEnv.times.
3065 :
3066 :			2. Added Time.{from,to}Nanoseconds to Time.
3067 :
3068 :			3. Improved Real.{from,to}LargeInt by avoiding needless calculations.
3069 :			For example, fromLargeInt never needs to look at more than 3 "big
3070 :			digits" to get its 53 bits of precision.
3071 :
3072 :			----------------------------------------------------------------------
3073 :	mblume	1384	Name: Matthias Blume (blume (at) tti - c (dot) org)
3074 :	mblume	1381	Date: 2003/09/17 16:30:00 CDT
3075 :			Tag: blume-20030917-real32-slices
3076 :			Description:
3077 :
3078 :			Added an entry to the primitive environment
3079 :			(compiler/Semant/statenv/prim.sml) for int32->real64 conversion and
3080 :			added code to compiler/CodeGen/main/mlriscGen.sml to implement it.
3081 :
3082 :			Removed some of the "magic" constants in real64.sml and replaced them
3083 :			with code that generates these values from their corresponding
3084 :			integer counterparts.
3085 :
3086 :			Made all(?) the slice-related changes to the Basis and made everything
3087 :			compile again...
3088 :
3089 :			----------------------------------------------------------------------
3090 :	mblume	1368	Name: Matthias Blume (blume (at) tti - c (dot) org)
3091 :	mblume	1375	Date: 2003/09/15 17:45:00 CDT
3092 :			Tag: blume-20030915-rbase
3093 :			Description:
3094 :
3095 :			Fixed bug in Real.fromLargeInt.
3096 :
3097 :			----------------------------------------------------------------------
3098 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
3099 :	mblume	1369	Date: 2003/09/13 18:11:00 CDT
3100 :			Tag: blume-20030913-libinstall
3101 :			Description:
3102 :
3103 :			Minor bugfix in config/libinstall (set anchor with path to
3104 :			standalone tool after installing it, otherwise libraries that
3105 :			need ml-lex or ml-yacc won't compile the first time the installer
3106 :			runs).
3107 :
3108 :			----------------------------------------------------------------------
3109 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
3110 :	mblume	1368	Date: 2003/09/12 11:45:00 CDT
3111 :			Tag: blume-20030912-various
3112 :			Description:
3113 :
3114 :			- fixed bug in Real.toLargeInt
3115 :			- fixed bug in Posix.ProcEnv.times
3116 :			- changed inputLine functions to return an option
3117 :			- minor installer improvements / bugfixes
3118 :			- changed default @SMLalloc parameter for x86/celeron to 64k
3119 :
3120 :			----------------------------------------------------------------------
3121 :			Name: Matthias Blume (blume (at) tti - c (dot) org)
3122 :	mblume	1356	Date: 2003/09/09 22:00:00 CDT
3123 :			Tag: Release_110_43
3124 :			Description:
3125 :
3126 :			New working release 110.43. New bootfiles.
3127 :
3128 :			----------------------------------------------------------------------
3129 :	mblume	1368	Name: Matthias Blume (blume (at) tti - c (dot) org)
3130 :	mblume	1354	Date: 2003/09/09 19:20:00 CDT
3131 :			Tag: blume-20030909-installer
3132 :			Description:
3133 :
3134 :			Rewrote large parts of config/install.sh in SML (config/libinstall.sml).
3135 :			Modified config/install.bat to take advantage of it. Also modified
3136 :			config/install.sh (and called it config/new-install.sh) to take advantage
3137 :			of it on Unix systems. (The SML code is (supposed to be) platform-
3138 :			independent.)
3139 :
3140 :			The installer can now install everything under Win32
3141 :			as well as under *nix as long as it compiles.
3142 :
3143 :			Other changes:
3144 :
3145 :			- made CML compile again under Win32
3146 :			- made eXene compile under Win32 (by providing a fake structure UnixSock
3147 :			and by using OS.Process.getEnv instead of Posix.ProcEnv.getenv)
3148 :			- fixed a bug in nowhere: it assumed that type OS.Process.status is the
3149 :			same as type int; under Win32 it isn't
3150 :			- fixed some slice-related problems in the win32-specific parts of CML
3151 :			- added a functor argument "sameVol" to os-path-fn.sml in the Basis
3152 :			(under Win32, the volume name is case-insensitive, and the
3153 :			OS.Path code compares volume names for equality)
3154 :
3155 :			----------------------------------------------------------------------
3156 :	mblume	1368	Name: Matthias Blume (blume (at) tti - c (dot) org)
3157 :	mblume	1353	Date: 2003/09/08 11:55:00 CDT
3158 :			Tag: blume-20030908-fullpath
3159 :			Description:
3160 :
3161 :			Made Win32 version of OS.FileSys.fullPath return current directory
3162 :			when given an empty string. This is what the spec says, and incidentally,
3163 :			CM depends on it. (CM otherwise goes into an infinite loop in certain
3164 :			cases when presented with the name of a non-existing .cm file.)
3165 :
3166 :			----------------------------------------------------------------------
3167 :	mblume	1368	Name: Matthias Blume (blume (at) tti - c (dot) org)
3168 :	mblume	1350	Date: 2003/09/04 16:30:00 CDT
3169 :	mblume	1352	Tag: blume-20030905-slices-etc
3170 :	mblume	1350	Description:
3171 :
3172 :			1. Changed interface to vectors and arrays in Basis to match
3173 :			(draft) Basis spec.
3174 :			2. Added signatures and implementations of slices according to
3175 :			Basis spec.
3176 :			3. Edited source code throughout the system to make it compile again
3177 :			under 1. and 2. (In some cases code had to be added to have it
3178 :			match the new signatures.)
3179 :			4. MLRISC should be backward-compatible: the copies of the originals
3180 :			of files that needed to change under 3. were retained, the .cm files
3181 :			check the compiler version number and use old versions when
3182 :			appropriate.
3183 :			5. Changed type of OS.FileSys.readDir and Posix.FileSys.readdir to
3184 :			dirstream -> string option (in accordance with Basis spec).
3185 :			6. When generating code that counts lines, ml-lex used function
3186 :			CharVector.foldli, taking advantage of its old interface.
3187 :			This has been replaced with the corresponding code from
3188 :			CharVectorSlice. (html-lex must be re-lexed!)
3189 :			7. BitArray in smlnj-lib/Util has been extended/modified to match the
3190 :			new MONO_ARRAY signature. (Do we need BitArraySlice?)
3191 :			8. Removed temporary additions (fromInternal, toInternal) from the
3192 :			(now obsolete) IntInf in smlnj-lib/Util.
3193 :			9. Cleaned up structure Byte.
3194 :			10. Added localOffset, scan, and fromString to Date (according to spec).
3195 :			Cleaned/corrected implementation of Date.
3196 :			(Still need to check for correctness; implement better canonicalizeDate.)
3197 :			11. Added "scan" to signature IEEE_REAL.
3198 :			12. Some improvements to IntInf [in particular: efficiency-hack for
3199 :			mod and rem when second operand is 2 (for parity checks).]
3200 :			13. Changed representation of type Time.time, using a single IntInf.int
3201 :			value counting microseconds. This considerably simplified the
3202 :			implementation of structure Time. We now support negative time
3203 :			values; scan and fromString handle signs.
3204 :			14. Functor PrimIO now takes two additional arguments (VectorSlice and
3205 :			ArraySlice).
3206 :
3207 :			----------------------------------------------------------------------
3208 :	mblume	1368	Name: Matthias Blume (blume (at) tti - c (dot) org)
3209 :	mblume	1347	Date: 2003/08/28 17:00:00 CDT
3210 :			Tag: blume-20030828-intinf
3211 :			Description:
3212 :
3213 :			This is a major update which comes with a version number bump
3214 :			(110.42.99 -- yes, we are really close to 110.43 :-), NEW BOOTFILES,
3215 :			and an implementation of IntInf in the Basis.
3216 :
3217 :			There are a fairly large number of related changes and updates throughout
3218 :			the system:
3219 :
3220 :			Basis:
3221 :			- Implemented IntInf.
3222 :			- Made LargeInt a projection of IntInf (by filtering through INTEGER).
3223 :	mblume	1348	- Added some missing Real64 operations, most notably Real.toLargeInt.
3224 :	mblume	1347	- Added FixedInt as a synonym for Int32.
3225 :
3226 :			compiler:
3227 :			* Added support for a built-in intinf type.
3228 :			- literals
3229 :			- pattern matching
3230 :			- conversion shortcuts (Int32.fromLarge o Int.toLarge etc.)
3231 :			- overloading on literals and operations
3232 :
3233 :			This required adding a primitive type intinf, some additional
3234 :			primops, and implementations for several non-trivial intinf
3235 :			operations in Core. (The intinf type is completely abstract
3236 :			to the compiler; all operations get delegated back to the Core.)
3237 :
3238 :			* Intinf equality is handled by polyequal. However, the compiler
3239 :			does not print its usual warning in this case (since polyequal
3240 :			is the right thing to do there).
3241 :
3242 :			* Improved the organization of structure InlineT.
3243 :
3244 :			* A word about conversion primops:
3245 :			If conversions involving intinf do not cancel out during
3246 :	mblume	1348	CPS contract, then the compiler must insert calls to Core functions.
3247 :	mblume	1347	Since all core access must be resolved already during the FLINT
3248 :			translate phase, it would be too late a the time of CPS contract
3249 :	mblume	1348	to add new Core calls. For this reason, conversion primops
3250 :	mblume	1347	for intinf carry two arguments: 1. the numeric argument that
3251 :			they are supposed to convert, and 2. the Core function that
3252 :			can help with this conversion if necessary. If CPS contract
3253 :	mblume	1348	eliminates a primop, then the associated Core function becomes
3254 :			dead and goes away. Intinf conversion primops that do not get
3255 :			eliminated by CPS contract get rewritten into calls of their
3256 :			core functions by a separate, new phase.
3257 :	mblume	1347
3258 :			interactive system:
3259 :			- Control.Print.intinfDepth controls max length of intinf constants
3260 :			being printed. (Analogous to Control.Print.stringDepth.)
3261 :			- Cleanup in printutil and pputil: got rid of unused stuff and
3262 :			duplicates; replaced some of the code with code that makes better
3263 :			use of library functionality.
3264 :
3265 :			CM:
3266 :			Bugfix: parse-errors in init group (system/smlnj/init/init.cmi)
3267 :			are no longer silent.
3268 :
3269 :			CKIT:
3270 :			Fixed mismatched uses of Int32 and LargeInt. I always decided
3271 :			in favor of LargeInt -- which is now the same as IntInf.
3272 :			CKIT-knowledgable people should check whether this is what's
3273 :			intended and otherwise change things back to using Int32 or
3274 :			FixedInt.
3275 :
3276 :			Throughout the code:
3277 :	mblume	1348	Started using IntInf.int literals and built-in operations
3278 :	mblume	1347	(e.g., comparison with 0) where this seems appropriate.
3279 :
3280 :
3281 :			----------------------------------------------------------------------
3282 :	macqueen	1343	Name: Dave MacQueen (dbm@cs.uchicago.edu)
3283 :			Date: 2003/08/13 11:36:00 CDT
3284 :			Tag: dbm-20030813-mcz-merge1
3285 :			Description:
3286 :
3287 :			Merging changes from the mcz-branch development branch into trunk.
3288 :			These changes involve replacement of the emulated old prettyprinter
3289 :			interface with direct use of the SML/NJ Lib PP library, and fixing
3290 :			of a couple of bugs (895, 1186) relating to error messages. A new
3291 :			prettyprinter for ast datatypes (Elaborator/print/ppast.{sig,sml})
3292 :			has been added.
3293 :
3294 :			----------------------------------------------------------------------
3295 :	mblume	1368	Name: Matthias Blume (blume (at) tti - c (dot) org)
3296 :	mblume	1342	Date: 2003/08/11 15:45:00 CDT
3297 :			Tag: blume-20030811-windows
3298 :			Description:
3299 :
3300 :			Version number bumped to 110.42.9. NEW BOOTFILES!!!
3301 :
3302 :			http://smlnj.cs.uchicago.edu/dist/working/110.42.9/
3303 :
3304 :			This patch restores SML/NJ's ability to run under win32. There are a
3305 :			number of changes, including fixes for several bugs that had gone
3306 :			unnoticed until now:
3307 :
3308 :			- uname "CYGWIN_NT*" is recognized as win32 (This is relevant only when
3309 :			trying to run the win32 version from within cygwin.)
3310 :
3311 :			- There are a number of simple .bat scripts that substitute for their
3312 :			corresponding Unix shell-scripts. (See below.)
3313 :
3314 :			- The internals of ml-build have been modified slightly. The main
3315 :			difference is that instead of calling ".link-sml" (or link-sml.bat)
3316 :			using OS.Process.system, the ML process delegates this task back
3317 :			to the script. Otherwise problems arise in mixed environments such
3318 :			as Cygwin where scripts look and work like Unix scripts, but
3319 :			where OS.Process.system cannot run them.
3320 :
3321 :			- In CM, the srcpath pickler used native pathname syntax -- which
3322 :			is incorrect in the case of cross-compilation. The new pickle format
3323 :			is independent of platform-specific naming conventions.
3324 :
3325 :			- Path configuration files (such as lib/pathconfig) can now choose
3326 :			between native and standard syntax. Placing a line of the form
3327 :
3328 :			standard!
3329 :
3330 :			into the file causes all subsequent paths to be interpreted using
3331 :			CM standard pathname syntax (= Unix conventions); a line
3332 :
3333 :			native!
3334 :
3335 :			switches back to native style. This was needed so that
3336 :			path config files can be written portably, see src/system/pathconfig.
3337 :
3338 :			- Runtime system:
3339 :
3340 :			- win32-filesys.c: get_file_time and set_file_time now
3341 :			access modification time, not creation time.
3342 :
3343 :			- I/O code made aware of new array representation.
3344 :
3345 :			- Bug fixes in X86.prim.masm.
3346 :
3347 :			- src/system/makeml made aware of win32. (For use under cygwin
3348 :			and other Unix-environments for windows.)
3349 :
3350 :			- In Basis, fixed off-by-one error in win32-io.sml (function vecF)
3351 :			which caused BinIO.inputAll to fail consistently.
3352 :
3353 :			.bat scripts:
3354 :
3355 :			Windows .bat scripts assume that SMLNJ_HOME is defined.
3356 :
3357 :			- sml.bat, ml-yacc.bat, ml-lex.bat: Driver scripts for standalone
3358 :			applications (sml, ml-yacc, ml-lex).
3359 :			- ml-build.bat: analogous to ml-build.
3360 :			- config\install.bat: Analogous to config/install.sh. This requires
3361 :			that SMLNJ_HOME is set and that Microsoft Visual C is ready to use.
3362 :			(nmake etc. must be on the path, and vcvars32 must have been run.)
3363 :			Moreover, sources for ml-lex and ml-yacc need to exist under src,
3364 :			and the bootfile hierarchy must have been unpacked under
3365 :			sml.boot.x86-win32.
3366 :			The script is very primitive and does a poor job at error checking.
3367 :			It only installs the base system, ml-lex, and ml-yacc. No other
3368 :			libraries are being installed (i.e., you get only those that
3369 :			are part of the compiler.)
3370 :			- link-sml.bat: analogous to .link-sml, but not currently used
3371 :
3372 :			Unrelated bug fixes:
3373 :
3374 :			- ml-nlffigen now exports structures ST_* corresponding to incomplete
3375 :			types.
3376 :			- Added getDevice to PP/src/pp-debug-fn.sml. (Would not compile
3377 :			otherwise.)
3378 :
3379 :			----------------------------------------------------------------------
3380 :	macqueen	1337	Name: Dave MacQueen (macqueen@cs.uchicago.edu)
3381 :			Date: 2003/06/17
3382 :			Tag: macqueen-20030617-bug895
3383 :			Description:
3384 :
3385 :			Modified compiler/Elaborator/print/pptype.sml to fix bug 895.
3386 :			Tag will be used for new development branch (mcz-branch) for
3387 :			use by MacQueen, (Lucasz) Zairek, and (George) Cao at uchicago.
3388 :
3389 :			----------------------------------------------------------------------
3390 :	mblume	1368	Name: Matthias Blume (blume (at) tti - c (dot) org)
3391 :	mblume	1335	Date: 2003/05/27 16:55:00 CDT
3392 :			Tag: blume-20030527-polyeq
3393 :			Description:
3394 :
3395 :			Tried to eliminated most cases of polymorphic equality.
3396 :
3397 :			----------------------------------------------------------------------
3398 :	mblume	1368	Name: Matthias Blume (blume (at) tti - c (dot) org)
3399 :	mblume	1334	Date: 2003/05/21 17:45:00 CDT
3400 :			Tag: blume-20030517-complete
3401 :			Description:
3402 :
3403 :			Two changes:
3404 :
3405 :			1. Added a flag for controlling whether non-exhaustive bindings will
3406 :			be treated as errors (default is false).
3407 :			2. Cleaned up the *entire* source tree so that CMB.make goes through
3408 :			without a single non-exhaustive match- or bind warning.
3409 :
3410 :			----------------------------------------------------------------------
3411 :	mblume	1368	Name: Matthias Blume (blume (at) tti - c (dot) org)
3412 :	mblume	1332	Date: 2003/05/17 10:20:00 CDT
3413 :			Tag: blume-20030517-absyn
3414 :			Description:
3415 :
3416 :			1. Added cases for IF, WHILE, ANDALSO, and ORELSE to Absyn.
3417 :
3418 :			This mainly affects the quality of error messages. However, some
3419 :			of the code is now more straightforward than before. (Treatment of
3420 :			the above four constructs in translate.sml is much simpler than
3421 :			the "macro-expansion" that was going on before. Plus, the mach-
3422 :			compiler no longer gets invoked just to be able to compile an
3423 :			if-expression.)
3424 :
3425 :			2. The ErrorMsg.Error exception is now caught and absorbed by the
3426 :			interactive loop.
3427 :
3428 :			----------------------------------------------------------------------
3429 :	allenleung	1327	Name: Allen Leung
3430 :			Date: 2003/05/16 13:05:00 CDT
3431 :			Tag: leunga-20030516-cygwin-runtime
3432 :			Description:
3433 :
3434 :	blume	2155	Ported the runtime system to cygwin, which uses the unix
3435 :	allenleung	1327	x86-unix bin files. Missing/buggy features:
3436 :
3437 :	blume	2155	o getnetbyname, getnetbyaddr: these functions seem to be missing in
3438 :	allenleung	1327	the Cygwin library.
3439 :			o Ctrl-C handling may be flaky.
3440 :			o Windows system calls and Windows I/O are not supported.
3441 :
3442 :			A new set of binfiles is located at:
3443 :
3444 :			http://www.dorsai.org/~leunga/boot.x86-unix.tgz
3445 :
3446 :			This is only needed for bootstrapping the cygwin version of smlnj.
3447 :			Other x86 versions can use the existing binfiles.
3448 :
3449 :			----------------------------------------------------------------------
3450 :	mblume	1317	Name: Matthias Blume
3451 :			Date: 2003/04/08 15:42:00 CDT
3452 :			Tag: blume-20030408-listpair
3453 :			Description:
3454 :
3455 :			1. Added a target 'mlrisc' to installer.
3456 :
3457 :			2. Added missing elements to structure ListPair.
3458 :
3459 :			----------------------------------------------------------------------
3460 :	leunga	1296	Name: Allen Leung
3461 :			Date: 2003/01/07 10:40:00 EST
3462 :			Tag: leunga-20030107-int-rem
3463 :			Description:
3464 :
3465 :			Fixed a bug in Int.rem(x,y) where y is a power of 2 on x86.
3466 :			The arguments to the SUBL instruction were swapped.
3467 :
3468 :			----------------------------------------------------------------------
3469 :	blume	1229	Name: Matthias Blume
3470 :	blume	1294	Date: 2002/12/12 16:25:00 EST
3471 :			Tag: blume-20021212-risc-ra
3472 :			Description:
3473 :
3474 :			Fixed a serious bug in the rewrite code for FP spilling/reloading that
3475 :			sent the RA into an infinite loop when floating point registers get
3476 :			spilled. (Because of this bug, e.g., nucleic stopped compiling between
3477 :			110.37 and 110.38.)
3478 :			There was another set of potential problems related to the handling of
3479 :			MLRISC annotations (but those did not yet cause real problems, apparently).
3480 :
3481 :			----------------------------------------------------------------------
3482 :			Name: Matthias Blume
3483 :	blume	1291	Date: 2002/12/06 22:40:00 EST
3484 :			Tag: blume-20021206-cm-fileid
3485 :			Description:
3486 :
3487 :			Added a call of SrcPath.sync at the beginning of Parse.parse (in CM).
3488 :			This fixes the problem of CM getting confused by files that suddenly
3489 :			change their identity (e.g., by getting unlinked and recreated by some
3490 :			text editor such as vi). There might be a better/cheaper/cleaner way
3491 :			of doing this, but for now this will have to do.
3492 :
3493 :			----------------------------------------------------------------------
3494 :			Name: Matthias Blume
3495 :	blume	1288	Date: 2002/10/28 09:50:00 EST
3496 :	blume	2155	Tag: blume-20021028-typecheck
3497 :	blume	1288	Description:
3498 :
3499 :			Exported structure Typecheck from $smlnj/viscomp/core.cm.
3500 :
3501 :			----------------------------------------------------------------------
3502 :			Name: Matthias Blume
3503 :	blume	1282	Date: 2002/10/17 09:10:00 EDT
3504 :	blume	1279	Tag: Release_110_42
3505 :			Description:
3506 :
3507 :	blume	1282	In good old tradition, there has been a slight hiccup so that we have
3508 :			to patch 110.42 after the fact. The old release tag has been replaced
3509 :			(see below).
3510 :
3511 :			The change solves a problem with two competing approaches the
3512 :			configuration problem regarding MacOS 10.1 vs. MacOS 10.2 which got in
3513 :			each other's way.
3514 :
3515 :			This change only affects the runtime system code and the installer script.
3516 :			(No new bootfiles.)
3517 :
3518 :			----------------------------------------------------------------------
3519 :			Name: Matthias Blume
3520 :			Date: 2002/10/16 12:00:00 EDT
3521 :			Tag: Release_110_42_removed
3522 :			Description:
3523 :
3524 :	blume	1279	New working release. New bootfiles.
3525 :
3526 :			----------------------------------------------------------------------
3527 :			Name: Matthias Blume
3528 :	blume	1273	Date: 2002/10/10 13:10:00 EDT
3529 :			Tag: blume-20021010-ppc-divs
3530 :			Description:
3531 :
3532 :			The mltree operator DIVS must be implemented with an overflow check on
3533 :			the PPC because the hardware indicates divide-by-zero using "overflow" as
3534 :			well.
3535 :
3536 :			----------------------------------------------------------------------
3537 :			Name: Matthias Blume
3538 :	blume	1261	Date: 2002/07/23 11:45:00
3539 :			Tag: blume-20020723-smlnj-home
3540 :			Description:
3541 :
3542 :			Sml now senses the SMLNJ_HOME environment variable. If this is set,
3543 :			then the bin dir is assumed to be in $SMLNJ_HOME/bin and (unless
3544 :			CM_PATHCONFIG is also set), the path configuration file is assumed
3545 :			to be in $SMLNJ_HOME/lib/pathconfig. This way one can easily move
3546 :			the entire tree to some other place and everything will "just work".
3547 :
3548 :			(Companion commands such as ml-build and ml-makedepend also sense this
3549 :			variable.)
3550 :
3551 :			----------------------------------------------------------------------
3552 :			Name: Matthias Blume
3553 :	blume	1259	Date: 2002/07/12 21:19:00 EDT
3554 :			Tag: blume-20020712-liveness
3555 :			Description:
3556 :
3557 :			Exported two useful "step" functions from liveness module (MLRISC).
3558 :
3559 :			----------------------------------------------------------------------
3560 :			Name: Matthias Blume
3561 :	blume	1253	Date: 2002/07/05 16:00 EDT
3562 :			Tag: Release_110_41
3563 :			Description:
3564 :
3565 :			New working release. New bootfiles.
3566 :
3567 :			----------------------------------------------------------------------
3568 :			Name: Matthias Blume
3569 :	blume	1250	Date: 2002/07/05 10:25:00 EDT
3570 :			Tag: blume-20020705-btimp
3571 :			Description:
3572 :
3573 :			Exported structure BTImp from $smlnj/viscomp/debugprof.cm so that
3574 :			other clients can set up backtracing support.
3575 :
3576 :			----------------------------------------------------------------------
3577 :			Name: Matthias Blume
3578 :	blume	1249	Date: 2002/06/25 17:23:00 EDT
3579 :			Tag: blume-20020625-fpmax
3580 :			Description:
3581 :
3582 :			Fixed a bug in translation of INLMAX (and INLMIN) for the floating-point
3583 :			case. (The sense of the isNaN test was reversed -- which made min and
3584 :			max always return their first argument.)
3585 :
3586 :			----------------------------------------------------------------------
3587 :			Name: Matthias Blume
3588 :	blume	1240	Date: 2002/06/11
3589 :			Tag: blume-20020611-unixpath
3590 :			Description:
3591 :
3592 :			Back-ported OS.Path.{from,to}UnixPath from idlbasis-devel branch.
3593 :
3594 :			----------------------------------------------------------------------
3595 :			Name: Matthias Blume
3596 :	blume	1238	Date: 2002/06/10 16:35:00 EDT
3597 :			Tag: blume-20020610-ieeereal
3598 :			Description:
3599 :
3600 :			I back-ported my implementation of IEEEReal.fromString from the
3601 :			idlbasis-devel branch so that we can test it.
3602 :
3603 :			Another small change is that ppDec tries to give more information
3604 :			than just "<sig>" in the case of functors. However, this code is
3605 :			broken in some mysterious way if the functor's body's signature
3606 :			has not been declared by ascription but gets inferred from the
3607 :			implementation. This needs fixing...
3608 :
3609 :			----------------------------------------------------------------------
3610 :			Name: Matthias Blume
3611 :	blume	1229	Date: 2002/05/31
3612 :			Tag: blume-20020531-btrace-mode
3613 :			Description:
3614 :
3615 :			Resurrected SMLofNJ.Internals.BTrace.mode. (It accidentally fell by
3616 :			the wayside when I switched over to using Controls everywhere.)
3617 :
3618 :			----------------------------------------------------------------------
3619 :	george	1222	Name: Lal George
3620 :			Date: 2002/05/23 12:21:40 EDT
3621 :			Tag: george-20020523-visual-labels
3622 :			Description:
3623 :
3624 :			Labels are now displayed in the graphical output to make
3625 :			the fall-through and target blocks obvious.
3626 :
3627 :			----------------------------------------------------------------------
3628 :	blume	1204	Name: Matthias Blume
3629 :	blume	1220	Date: 2002/05/22 11:03:00 EDT
3630 :			Tag: blume-20020522-shrink
3631 :			Description:
3632 :
3633 :			John tweaked yesterday's fix for 1131 to handle an out-of-memory
3634 :			situation that comes up when allocating huge arrays.
3635 :
3636 :			----------------------------------------------------------------------
3637 :			Name: Matthias Blume
3638 :	blume	1215	Date: 2002/05/21 16:00:00 EDT
3639 :			Tag: Release_110_40
3640 :			Description:
3641 :
3642 :			New working release (110.40). New bootfiles.
3643 :
3644 :			[Also: John Reppy fixed GC bug 1131.]
3645 :
3646 :			----------------------------------------------------------------------
3647 :			Name: Matthias Blume
3648 :	blume	1212	Date: 2002/05/21 12:35:00 EDT
3649 :			Tag: blume-20020521-cmdoc
3650 :			Description:
3651 :
3652 :			CM documentation update.
3653 :
3654 :			----------------------------------------------------------------------
3655 :			Name: Matthias Blume
3656 :	blume	1211	Date: 2002/05/21 10:55:00 EDT
3657 :			Tag: blume-20020521-misc
3658 :			Description:
3659 :
3660 :			- John tweaked runtime to be silent on heap export (except when
3661 :			GC messages are on).
3662 :			- I added a few more things (cross-compiling versions of CMB) to
3663 :			config/preloads (as suggestions).
3664 :
3665 :			----------------------------------------------------------------------
3666 :			Name: Matthias Blume
3667 :	blume	1208	Date: 2002/05/20 22:25:00 EDT
3668 :			Tag: blume-20020520-controls
3669 :			Description:
3670 :
3671 :			- Added ControlUtil structure to control-lib.cm.
3672 :			- Use it throughout.
3673 :			- Used Controls facility to define MLRISC controls (as opposed to
3674 :			registering MLRISC control ref cells with Controls after the fact)
3675 :			- Fixed messed-up controls priorities.
3676 :
3677 :			* Removed again all the stuff from config/preloads that one wouldn't
3678 :			be able to preload at the time the initial heap image is built.
3679 :			(Many libraries, e.g., CML, do not exist yet at this time. The
3680 :			only libraries that can be preloaded via config/preloads are those
3681 :			that come bundled with the bootfiles.)
3682 :
3683 :			----------------------------------------------------------------------
3684 :			Name: Matthias Blume
3685 :	blume	1204	Date: 2002/05/20 10:59:00 EDT
3686 :			Tag: blume-20020520-preloads
3687 :			Description:
3688 :
3689 :			Added a lot of commented-out suggestions for things to be included
3690 :			in config/preloads.
3691 :
3692 :			----------------------------------------------------------------------
3693 :	leunga	1203	Name: Allen Leung
3694 :			Date: 2002/05/18 14:20:00 EDT
3695 :			Tag: leunga-20020518-mdl
3696 :			Description:
3697 :
3698 :			o Made the mdl tool stuff compile and run again.
3699 :			o I've disabled all the stuff that depends on RTL specifications; they
3700 :			are all badly broken anyway.
3701 :
3702 :			----------------------------------------------------------------------
3703 :	blume	1201	Name: Matthias Blume
3704 :			Date: 2002/05/17 16:49:00 EDT
3705 :			Tag: blume-20020517-controls
3706 :			Description:
3707 :
3708 :			0. John Reppy made several modifications to the SML/NJ library.
3709 :			In particular, there is a shiny new controls-lib.cm.
3710 :
3711 :			1. Pushed new controls interface through compiler so that everything
3712 :			compiles again.
3713 :
3714 :			2. Added FormatComb and FORMAT_COMB to the CML version of the
3715 :			SML/NJ library (so that CML compiles again).
3716 :
3717 :			3. Modified init scripts because XXX_DEFAULT environment variables
3718 :			are no longer with us. (Boot-time initialization is now done
3719 :			using the same environment variables that are also used for
3720 :			startup-time initialization of controls.)
3721 :
3722 :			----------------------------------------------------------------------
3723 :	george	1192	Name: Lal George
3724 :			Date: 2002/05/15 09:20:10 EDT
3725 :			Tag: george-20020515-pseudo-op-decls
3726 :			Description:
3727 :
3728 :			All pseudo-ops emitted before the first segment declaration
3729 :			such as TEXT, DATA, and BSS directives are assumed to be global
3730 :			declarations and are emitted first in the assembly file. This is
3731 :			useful in a number of situations where one has pseudo-ops that are not
3732 :			specific to any segment, and also works around the constraint that one
3733 :			cannot have client pseudo-ops in the TEXT segment.
3734 :
3735 :			Because no segment is associated with these declarations it is
3736 :			an error to allocate any space or objects before the first segment
3737 :			directive and an exception will be raised. However, we cannot make
3738 :			this check for client pseudo-ops.
3739 :
3740 :			These top level declarations are a field in the CFG graph_info.
3741 :			In theory you can continue to add to this field after the CFG has been
3742 :			built -- provided you know what you are doing;-)
3743 :
3744 :			----------------------------------------------------------------------
3745 :	blume	1175	Name: Matthias Blume
3746 :	blume	1190	Date: 2002/05/13 16:40:00 EDT
3747 :			Tag: blume-20020513-pp-etc
3748 :			Description:
3749 :
3750 :			A few minor bugfixes:
3751 :
3752 :			- Stopgap measure for bug recently reported by Elsa Gunter (ppDec).
3753 :			(Bogus printouts for redefined bindings still occur. Compiler
3754 :			bug should no longer occur now. We need to redo the prettyprinter
3755 :			from scratch.)
3756 :
3757 :			- CM pathname printer now also adds escape sequences for ( and )
3758 :
3759 :			- commend and docu fixes for ml-nlffi
3760 :
3761 :			----------------------------------------------------------------------
3762 :			Name: Matthias Blume
3763 :	blume	1189	Date: 2002/05/10 16:40:00 EDT
3764 :			Tag: blume-20020510-erg-textio
3765 :			Description:
3766 :
3767 :			Applied the following bugfix provided by Emden Gansner:
3768 :
3769 :			Output is corrupted when outputSubstr is used rather than output.
3770 :
3771 :	blume	2155	The problem occurs when a substring
3772 :	blume	1189
3773 :			ss = (s, dataStart, dataLen)
3774 :
3775 :	blume	2155	where dataStart > 0, fills a stream buffer with avail bytes left.
3776 :			avail bytes of s, starting at index dataStart, are copied into the
3777 :	blume	1189	buffer, the buffer is flushed, and then the remaining dataLen-avail
3778 :	blume	2155	bytes of ss are copied into the beginning of the buffer. Instead of
3779 :			starting this copy at index dataStart+avail in s, the current code
3780 :	blume	1189	starts the copy at index avail.
3781 :
3782 :			Fix:
3783 :			In text-io-fn.sml, change line 695 from
3784 :			val needsFlush = copyVec(v, avail, dataLen-avail, buf, 0)
3785 :	blume	2155	to
3786 :	blume	1189	val needsFlush = copyVec(v, dataStart+avail, dataLen-avail, buf, 0)
3787 :
3788 :			----------------------------------------------------------------------
3789 :			Name: Matthias Blume
3790 :	blume	1186	Date: 2002/04/12 13:55:00 EDT
3791 :			Tag: blume-20020412-assyntax
3792 :			Description:
3793 :
3794 :			1. Grabbed newer assyntax.h from the XFree86 project.
3795 :			2. Fiddled with how to compile X86.prim.asm without warnings.
3796 :			3. (Very) Minor cleanup in CM.
3797 :
3798 :			----------------------------------------------------------------------
3799 :			Name: Matthias Blume
3800 :	blume	1185	Date: 2002/04/01 (no joke!) 17:07:00 EST
3801 :			Tag: blume-20020401-x86div
3802 :			Description:
3803 :
3804 :			Added full support for div/mod/rem/quot on the x86, using the machine
3805 :			instruction's two results (without clumsily recomputing the remainder)
3806 :			directly where appropriate.
3807 :
3808 :			Some more extensive power-of-two support was added to the x86 instruction
3809 :			selector (avoiding expensive divs, mods, and muls where they can be
3810 :			replaced with cheaper shifts and masks). However, this sort of thing
3811 :			ought to be done earlier, e.g., within the CPS optimizer so that
3812 :			all architectures benefit from it.
3813 :
3814 :			The compiler compiles to a fixed point, but changes might be somewhat
3815 :			fragile nevertheless. Please, report any strange things that you might
3816 :			see wrt. div/mod/quot/rem...
3817 :
3818 :			----------------------------------------------------------------------
3819 :			Name: Matthias Blume
3820 :	blume	1184	Date: 2002/03/29 17:22:00
3821 :			Tag: blume-20020329-div
3822 :			Description:
3823 :
3824 :			Fixed my broken div/mod logic. Unfortunately, this means that the
3825 :			inline code for div/mod now has one more comparison than before.
3826 :			Fast paths (quotient > 0 or remainder = 0) are not affected, though.
3827 :			The problem was with quotient = 0, because that alone does not tell
3828 :			us which way the rounding went. One then has to look at whether
3829 :			remainder and divisor have the same sign... :(
3830 :
3831 :			Anyway, I replaced the bootfiles with fresh ones...
3832 :
3833 :			----------------------------------------------------------------------
3834 :			Name: Matthias Blume
3835 :	blume	1183	Date: 2002/03/29 14:10:00 EST
3836 :			Tag: blume-20020329-inlprims
3837 :			Description:
3838 :
3839 :			NEW BOOTFILES!!! Version number bumped to 110.39.3.
3840 :
3841 :			Primops have changed. This means that the bin/boot-file formats have
3842 :			changed as well.
3843 :
3844 :			To make sure that there is no confusion, I made a new version.
3845 :
3846 :
3847 :			CHANGES:
3848 :
3849 :			* removed REMT from mltree (remainder should never overflow).
3850 :
3851 :			* added primops to deal with divisions of all flavors to the frontend
3852 :
3853 :			* handled these primops all the way through so they map to their respective
3854 :			MLRISC support
3855 :
3856 :			* used these primops in the implementation of Int, Int32, Word, Word32
3857 :
3858 :			* removed INLDIV, INLMOD, and INLREM as they are no longer necessary
3859 :
3860 :			* parameterized INLMIN, INLMAX, and INLABS by a numkind
3861 :
3862 :			* translate.sml now deals with all flavors of INL{MIN,MAX,ABS}, including
3863 :			floating point
3864 :
3865 :			* used INL{MIN,MAX,ABS} in the implementation of Int, Int32, Word, Word32,
3866 :			and Real (but Real.abs maps to a separate floating-point-only primop)
3867 :
3868 :
3869 :			TODO items:
3870 :
3871 :			* Hacked Alpha32 instruction selection, disabling the selection of REMx
3872 :			instructions because the machine instruction encoder cannot handle
3873 :			them. (Hppa, PPC, and Sparc instruction selection did not handle
3874 :			REM in the first place, and REM is supported by the x86 machine coder.)
3875 :
3876 :			* Handle DIV and MOD with DIV_TO_NEGINF directly in the x86 instruction
3877 :			selection phase. (The two can be streamlined because the hardware
3878 :			delivers both quotient and remainder at the same time anyway.)
3879 :
3880 :			* Think about what to do with "valOf(Int32.minInt) div ~1" and friends.
3881 :			(Currently the behavior is inconsistent both across architectures and
3882 :			wrt. the draft Basis spec.)
3883 :
3884 :			* Word8 should eventually be handled natively, too.
3885 :
3886 :			* There seems to be one serious bug in mltree-gen.sml. It appears, though,
3887 :			as if there currently is no execution path that could trigger it in
3888 :			SML/NJ. (The assumptions underlying functions arith and promotable do not
3889 :			hold for things like multiplication and division.)
3890 :
3891 :			----------------------------------------------------------------------
3892 :			Name: Matthias Blume
3893 :	blume	1181	Date: 2002/03/27 16:27:00 EST
3894 :			Tag: blume-20020327-mlrisc-divisions
3895 :			Description:
3896 :
3897 :			Added support for all four division operations (ML's div, mod, quot,
3898 :			and rem) to MLRISC. In the course of doing so, I also rationalized
3899 :			the naming (no more annoying switch-around of DIV and QUOT), by
3900 :			parameterizing the operation by div_rounding_mode (which can be either
3901 :			DIV_TO_ZERO or DIV_TO_NEGINF).
3902 :
3903 :			The generic MLTreeGen functor takes care of compiling all four
3904 :			operations down to only round-to-zero div.
3905 :
3906 :			Missing pieces:
3907 :
3908 :			* Doing something smarter than relying on MLTreeGen on architectures
3909 :			like, e.g., the x86 where hardware division delivers both quotient and
3910 :			remainder at the same time. With this, the implementation of the
3911 :			round-to-neginf operations could be further streamlined.
3912 :
3913 :			* Remove inlining support for div/mod/rem from the frontend and replace it
3914 :			with primops that get carried through to the backend. Do this for all
3915 :			int and word types.
3916 :
3917 :			----------------------------------------------------------------------
3918 :			Name: Matthias Blume
3919 :	blume	1180	Date: 2002/03/25 17:25:00 EST
3920 :			Tag: blume-20020325-divmod
3921 :			Description:
3922 :
3923 :			I improved (hopefully without breaking them) the implementation of Int.div,
3924 :			Int.mod, and Int.rem. For this, the code in translate.sml now takes
3925 :			advantage of the following observations:
3926 :
3927 :			Let q = x quot y r = x rem y
3928 :			d = x div y m = x mod y
3929 :
3930 :			where "quot" is the round-to-zero version of integer division that
3931 :			hardware usually provides. Then we have:
3932 :
3933 :			r = x - q * y where neither the * nor the - will overflow
3934 :			d = if q >= 0 orelse x = q * y then q else q - 1
3935 :			where neither the * nor the - will overflow
3936 :			m = if q >= 0 orelse r = 0 then r else r + y
3937 :			where the + will not overflow
3938 :
3939 :			This results in substantial simplification of the generated code.
3940 :			The following table shows the number of CFG nodes and edges generated
3941 :			for
3942 :			fun f (x, y) = x OPER y
3943 :			(* with OPER \in div, mod, quot, rem *)
3944 :
3945 :
3946 :			OPER | nodes(old) | edges(old) | nodes(new) | edges(new)
3947 :			--------------------------------------------------------
3948 :			div | 24 | 39 | 12 | 16
3949 :			mod | 41 | 71 | 12 | 16
3950 :			quot | 8 | 10 | 8 | 10
3951 :			rem | 10 | 14 | 8 | 10
3952 :
3953 :
3954 :			----------------------------------------------------------------------
3955 :			Name: Matthias Blume
3956 :	blume	1179	Date: 2002/03/25 22:06:00 EST
3957 :			Tag: blume-20020325-cprotobug
3958 :			Description:
3959 :
3960 :			Fixed a bug in cproto (c prototype decoder).
3961 :
3962 :			----------------------------------------------------------------------
3963 :			Name: Matthias Blume
3964 :	blume	1178	Date: 2002/03/25 16:00:00 EST
3965 :			Tag: blume-20020325-raw-primops
3966 :			Description:
3967 :
3968 :			I did some cleanup to Allen's new primop code and
3969 :			replaced yesterday's bootfiles with new ones.
3970 :			(But they are stored in the same place.)
3971 :
3972 :			----------------------------------------------------------------------
3973 :			Name: Matthias Blume
3974 :	blume	1175	Date: 2002/03/24 22:40:00 EST
3975 :			Tag: blume-20020324-bootfiles
3976 :			Description:
3977 :
3978 :			Made the bootfiles that Allen asked for.
3979 :
3980 :			----------------------------------------------------------------------
3981 :	leunga	1174	Name: Allen Leung
3982 :			Date: 2002/03/23 15:50:00 EST
3983 :			Tag: leunga-20020323-flint-cps-rcc-primops
3984 :			Description:
3985 :
3986 :			1. Changes to FLINT primops:
3987 :
3988 :			(* make a call to a C-function;
3989 :			* The primop carries C function prototype information and specifies
3990 :			* which of its (ML-) arguments are floating point. C prototype
3991 :			* information is for use by the backend, ML information is for
3992 :			* use by the CPS converter. *)
3993 :			| RAW_CCALL of { c_proto: CTypes.c_proto,
3994 :			ml_args: ccall_type list,
3995 :			ml_res_opt: ccall_type option,
3996 :			reentrant : bool
3997 :			} option
3998 :			(* Allocate uninitialized storage on the heap.
3999 :			* The record is meant to hold short-lived C objects, i.e., they
4000 :	blume	2155	* are not ML pointers. With the tag, the representation is
4001 :	leunga	1174	* the same as RECORD with tag tag_raw32 (sz=4), or tag_fblock (sz=8)
4002 :			*)
4003 :	blume	2155	| RAW_RECORD of {tag:bool,sz:int}
4004 :	leunga	1174	and ccall_type = CCALL_INT32 | CCALL_REAL64 | CCALL_ML_PTR
4005 :
4006 :			2. These CPS primops are now overloaded:
4007 :
4008 :			rawload of {kind:numkind}
4009 :			rawstore of {kind:numkind}
4010 :
4011 :			The one argument form is:
4012 :
4013 :			rawload {kind} address
4014 :
4015 :			The two argument form is:
4016 :	blume	2155
4017 :	leunga	1174	rawload {kind} [ml object, byte-offset]
4018 :
4019 :			3. RAW_CCALL/RCC now takes two extra arguments:
4020 :
4021 :			a. The first is whether the C call is reentrant, i.e., whether
4022 :			ML state should be saved and restored.
4023 :			b. The second argument is a string argument specifying the name of
4024 :			library and the C function.
4025 :
4026 :			These things are currently not handled in the code generator, yet.
4027 :
4028 :			4. In CProto,
4029 :
4030 :	blume	2155	An encoding type of "bool" means "ml object" and is mapped into
4031 :			C prototype of PTR. Note that "bool" is different than "string",
4032 :	leunga	1174	even though "string" is also mapped into PTR, because "bool"
4033 :			is assigned an CPS type of BOGt, while "string" is assigned INT32t.
4034 :
4035 :			5. Pickler/unpicker
4036 :
4037 :			Changed to handle RAW_RECORD and newest RAW_CCALL
4038 :
4039 :			6. MLRiscGen,
4040 :
4041 :			1. Changed to handle the new rawload/rawstore/rawrecord operators.
4042 :			2. Code for handling C Calls has been moved to a new module CPSCCalls,
4043 :			in the file CodeGen/cpscompile/cps-c-calls.sml
4044 :
4045 :			7. Added the conditional move operator
4046 :
4047 :			condmove of branch
4048 :
4049 :	blume	2155	to cps. Generation of this is still buggy so it is currently
4050 :	leunga	1174	disabled.
4051 :	blume	2155
4052 :	leunga	1174	----------------------------------------------------------------------
4053 :	george	1168	Name: Lal George
4054 :	blume	2155	Date: 2002/03/22 14:18:25 EST
4055 :	george	1171	Tag: george-20020322-cps-branch-prob
4056 :	george	1168	Description:
4057 :
4058 :	blume	2155	Implemented the Ball-Larus branch prediction-heuristics, and
4059 :	george	1168	incorporated graphical viewers for control flow graphs.
4060 :
4061 :	george	1171	Ball-Larus Heuristics:
4062 :	george	1168	---------------------
4063 :			See the file compiler/CodeGen/cpscompile/cpsBranchProb.sml.
4064 :
4065 :			By design it uses the Dempster-Shafer theory for combining
4066 :			probabilities. For example, in the function:
4067 :
4068 :			fun f(n,acc) = if n = 0 then acc else f(n-1, n*acc)
4069 :
4070 :			the ball-larus heuristics predicts that the n=0 is unlikely
4071 :			(OH-heuristic), and the 'then' branch is unlikely because of the
4072 :			RH-heuristic -- giving the 'then' branch an even lower combined
4073 :	george	1171	probability using the Dempster-Shafer theory.
4074 :	george	1168
4075 :			Finally, John Reppy's loop analysis in MLRISC, further lowers the
4076 :			probability of the 'then' branch because of the loop in the else
4077 :			branch.
4078 :
4079 :
4080 :			Graphical Viewing:
4081 :	blume	2155	------------------
4082 :	george	1168	I merely plugged in Allen's graphical viewers into the compiler. The
4083 :			additional code is not much. At the top level, saying:
4084 :
4085 :			Control.MLRISC.getFlag "cfg-graphical-view" := true;
4086 :
4087 :			will display the graphical view of the control flow graph just before
4088 :			back-patching. daVinci must be in your path for this to work. If
4089 :			daVinci is not available, then the default viewer can be changed
4090 :			using:
4091 :
4092 :			Control.MLRISC.getString "viewer"
4093 :
4094 :			which can be set to "dot" or "vcg" for the corresponding viewers. Of
4095 :			course, these viewers must be in your path.
4096 :
4097 :			The above will display the compilation unit at the level of clusters,
4098 :			many of which are small, boring, and un-interesting. Also setting:
4099 :
4100 :			Control.MLRISC.getInt "cfg-graphical-view_size"
4101 :
4102 :			will display clusters that are larger than the value set by the above.
4103 :
4104 :	blume	2155
4105 :	george	1168	----------------------------------------------------------------------
4106 :	blume	1159	Name: Matthias Blume
4107 :			Date: 2002/03/21 22:20:00 EST
4108 :			Tag: blume-20020321-kmp-bugfix
4109 :			Description:
4110 :
4111 :			Changed the interface to the KMP routine in PreString and fixed
4112 :			a minor bug in one place where it was used.
4113 :
4114 :			----------------------------------------------------------------------
4115 :	leunga	1156	Name: Allen Leung
4116 :	leunga	1158	Date: 2002/03/21 20:30:00 EST
4117 :			Tag: leunga-20020321-cfg
4118 :			Description:
4119 :
4120 :			Fixed a potential problem in cfg edge splitting.
4121 :
4122 :			----------------------------------------------------------------------
4123 :			Name: Allen Leung
4124 :	leunga	1156	Date: 2002/03/21 17:15:00 EST
4125 :			Tag: leunga-20020321-x86-fp-cfg
4126 :			Description:
4127 :
4128 :	blume	2155	1. Recoded the buggy parts of x86-fp.
4129 :	leunga	1156
4130 :			a. All the block reordering code has been removed.
4131 :			We now depend on the block placement phases to do this work.
4132 :
4133 :			b. Critical edge splitting code has been simplified and moved into the
4134 :			CFG modules, as where they belong.
4135 :
4136 :			Both of these were quite buggy and complex. The code is now much, much
4137 :			simpler.
4138 :
4139 :			2. X86 backend.
4140 :
4141 :			a. Added instructions for 64-bit support. Instruction selection for
4142 :			64-bit has not been committed, however, since that
4143 :	blume	2155	requires changes to MLTREE which haven't been approved by
4144 :	leunga	1156	Lal and John.
4145 :
4146 :	blume	2155	b. Added support for FUCOMI and FUCOMIP when generating code for
4147 :	leunga	1156	PentiumPro and above. We only generate these instructions in
4148 :			the fast-fp mode.
4149 :
4150 :	blume	2155	c. Added cases for JP and JNP in X86FreqProps.
4151 :
4152 :	leunga	1156	3. CFG
4153 :
4154 :			CFG now has a bunch of methods for edge splitting and merging.
4155 :
4156 :			4. Machine description.
4157 :
4158 :			John's simplification of MLTREE_BASIS.fcond broke a few machine
4159 :			description things:
4160 :
4161 :			rtl-build.{sig,sml} and hppa.mdl fixed.
4162 :
4163 :	blume	2155	NOTE: the machine description stuff in the repository is still broken.
4164 :			Again, I can't put my fixes in because that involves
4165 :			changes to MLTREE.
4166 :	leunga	1156
4167 :			----------------------------------------------------------------------
4168 :	blume	1145	Name: Matthias Blume
4169 :	blume	1155	Date: 2002/03/20 15:55:00 EST
4170 :			Tag: blume-20020320-kmp
4171 :			Description:
4172 :
4173 :			Implemented Knuth-Morris-Pratt string matching in PreString and used
4174 :			it for String.isSubstring, Substring.isSubstring, and
4175 :			Substring.position.
4176 :
4177 :			(Might need some stress-testing. Simple examples worked fine.)
4178 :
4179 :			----------------------------------------------------------------------
4180 :			Name: Matthias Blume
4181 :	blume	1152	Date: 2002/03/19 16:37:00 EST
4182 :			Tag: blume-20020319-witnesses
4183 :			Description:
4184 :
4185 :			Added a structure C.W and functions convert/Ptr.convert to ml-nlffi-lib.
4186 :
4187 :			This implements a generic mechanism for changing constness qualifiers
4188 :			anywhere within big C types without resorting to outright "casts".
4189 :			(So far, functions such as C.rw/C.ro or C.Ptr.rw/C.Ptr.ro only let you
4190 :			modify the constness at the outermost level.)
4191 :			The implementation of "convert" is based on the idea of "witness"
4192 :			values -- values that are not used by the operation but whose types
4193 :			"testify" to their applicability. On the implementation side, "convert"
4194 :			is simply a projection (returning its second curried argument). With
4195 :			cross-module inlining, it should not result in any machine code being
4196 :			generated.
4197 :
4198 :			----------------------------------------------------------------------
4199 :			Name: Matthias Blume
4200 :	blume	1148	Date: 2002/03/15 16:40:00 EST
4201 :			Tag: blume-20020315-basis
4202 :			Description:
4203 :
4204 :			Provided (preliminary?) implementations for
4205 :
4206 :			{String,Substring}.{concatWith,isSuffix,isSubstring}
4207 :
4208 :			and
4209 :
4210 :			Substring.full
4211 :
4212 :			Those are in the Basis spec but they were missing in SML/NJ.
4213 :
4214 :			----------------------------------------------------------------------
4215 :			Name: Matthias Blume
4216 :	blume	1145	Date: 2002/03/14 21:30:00 EST
4217 :			Tag: blume-20020314-controls
4218 :			Description:
4219 :
4220 :			Controls:
4221 :			---------
4222 :
4223 :			1. Factored out the recently-added Controls : CONTROLS stuff and put
4224 :			it into its own library $/controls-lib.cm. The source tree for
4225 :			this is under src/smlnj-lib/Controls.
4226 :
4227 :			2. Changed the names of types and functions in this interface, so they
4228 :			make a bit more "sense":
4229 :
4230 :			module -> registry
4231 :			'a registry -> 'a group
4232 :
4233 :			3. The interface now deals in ref cells only. The getter/setter interface
4234 :			is (mostly) gone.
4235 :
4236 :			4. Added a function that lets one register an already-existing ref cell.
4237 :
4238 :			5. Made the corresponding modifications to the rest of the code so that
4239 :			everything compiles again.