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

SCM Repository

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

Annotation of /sml/trunk/NOTES/HISTORY

Parent Directory Parent Directory | Revision Log Revision Log


Revision 2406 - (view) (download)

1 : blume 2237 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 : mrainey 2406 Name: Mike Rainey (http://www.cs.uchicago.edu/~mrainey)
18 :     Date: 2007/04/12
19 :     Description:
20 :    
21 :     Substantially changed the signature and implementation of AMD64
22 :     SVID. It now looks similar to the ia32 SVID, yet uses staged
23 :     allocation.
24 :    
25 :     ----------------------------------------------------------------------
26 : jhr 2387 Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
27 :     Date: 2007/04/07
28 :     Description:
29 :    
30 :     More Basis fixes: the WORD signature was missing {to,from}Large.
31 :     Thanks to Andreas Rossberg.
32 :    
33 :     ----------------------------------------------------------------------
34 : blume 2368 Name: Matthias Blume (http://www.tti-c.org/blume/)
35 :     Date: 2007/03/22 16:07:00 CDT
36 :     Tag: release-110.63
37 :     Description:
38 :    
39 :     Bumped version number to 110.63.
40 :    
41 :     !!!NEW BOOTFILES!!!
42 :    
43 :     ----------------------------------------------------------------------
44 : mrainey 2360 Name: Mike Rainey (http://www.cs.uchicago.edu/~mrainey)
45 :     Date: 2007/03/19
46 :     Description:
47 :    
48 :     Fixed bogus operand sizes in AMD64 instruction spilling.
49 :    
50 :     ----------------------------------------------------------------------
51 : jhr 2358 Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
52 :     Date: 2007/03/19
53 :     Description:
54 :    
55 :     Fixed a number of inconsistencies between the Posix.TTY structure
56 :     and the Basis specification. Thanks to Adam Chilpala.
57 :    
58 :     ----------------------------------------------------------------------
59 : mrainey 2345 Name: Mike Rainey (http://www.cs.uchicago.edu/~mrainey)
60 :     Date: 2007/02/26
61 :     Description:
62 :    
63 :     Added preliminary support in MLRISC for Staged Allocation, a technique
64 :     for specifying calling conventions (see
65 :     http://www.eecs.harvard.edu/~nr/pubs/staged-abstract.html).
66 :     Initially, we plan to use this code to generate C calls for the AMD64.
67 :    
68 :     The staged allocation code base resides in MLRISC/staged-allocation,
69 :     and specialized calling conventions go in MLRISC/ARCH/staged-allocation.
70 :    
71 :     ----------------------------------------------------------------------
72 : jhr 2342 Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
73 :     Date: 2007/02/20
74 :     Description:
75 :    
76 :     Bug fix: when SaveCState was called with two values to save, a subsequent
77 :     GC could cause the RestoreCState to fail because the saved state had been
78 :     promoted to tagless pair.
79 :    
80 :     ----------------------------------------------------------------------
81 : leunga 2340 Name: Allen Leung (leunga (at) reservoir (dot) com)
82 :     Date: 2007/02/20 19:08:00 EST
83 :     Description:
84 :    
85 :     Bug triggered by:
86 :    
87 :     val a1 = Word8Array.array(a1, 0w0);
88 :     val _ = Word8Array.update(a1, 0, 0w128);
89 :    
90 :     x86MCEmitter crashed when the immediate operand to MOVB was outside
91 :     of the range -128 ... 127. I've changed the code so that the range
92 :     check is disabled. Only the low order 8 bits of the immediate operand
93 : jhr 2342 are now significant.
94 : leunga 2340
95 :    
96 :     ----------------------------------------------------------------------
97 : blume 2304 Name: Matthias Blume (http://www.tti-c.org/blume/)
98 : blume 2336 Date: 2007/02/15 09:20:00 CST
99 :     Description:
100 :    
101 :     Eliminated any mention of lexgen. (Lexgen was an early precursor
102 :     to ml-ulex.)
103 :    
104 :     You should update your admin directory, so the shell scripts for
105 :     maintaining your local copy of the repository reflect this change.
106 :    
107 :     ----------------------------------------------------------------------
108 :     Name: Matthias Blume (http://www.tti-c.org/blume/)
109 : blume 2333 Date: 2007/02/14 15:45:00 CST
110 :     Description:
111 :    
112 :     Fixed bug in CM's parallel make facility that failed to have
113 :     the master re-link modules after letting slaves compile them.
114 :    
115 :     ----------------------------------------------------------------------
116 :     Name: Matthias Blume (http://www.tti-c.org/blume/)
117 : blume 2331 Date: 2007/02/12 15:10:00 CST
118 :     Description:
119 :    
120 :     Fixed typo in ml-build script that prevented library anchors
121 :     from being registered.
122 :    
123 :     ----------------------------------------------------------------------
124 :     Name: Matthias Blume (http://www.tti-c.org/blume/)
125 : blume 2304 Date: 2007/02/02 15:20:00 CST
126 :     Tag: release-110.62
127 :     Description:
128 :    
129 :     New working release (110.62). NEW BOOTFILES!
130 :    
131 :     ----------------------------------------------------------------------
132 : adrassi 2296 Name: Aaron Turon (http://www.cs.uchicago.edu/~adrassi)
133 :     Date: 2007/01/31
134 :     Description:
135 :    
136 :     Brought ml-lpt manual up-to-date with the code. Many minor
137 :     improvements to the tools, and a few minor bugfixes. Prepared
138 :     ml-lpt-lib for integration with new UTF8 structure. Changed
139 :     the interface for creating streams in ml-ulex (we now support
140 :     stream creation from several kinds of sources). Overall, the
141 :     tools are now quite stable, and their interfaces are unlikely to
142 :     change in a way that would break compatibility. More work,
143 :     however, is needed in the documentation.
144 :    
145 :     ----------------------------------------------------------------------
146 : blume 2277 Name: Matthias Blume (http://www.tti-c.org/blume/)
147 : blume 2298 Date: 2007/02/01 10:05:00 CST
148 :     Description:
149 :    
150 :     Implemented library installer. Moved CM plugin code for
151 :     ml-burg, ml-lex, and ml-yacc out of CM source tree and into
152 :     their respective trees. Implemented CM plugin for ml-ulex
153 :     and ml-antlr. Used library installer for ml-burg, ml-ulex,
154 :     and ml-antlr.
155 :    
156 :     For ml-yacc and ml-lex we continue to have permanently "plugged-in"
157 :     CM tools. (It turns out to be too messy to do otherwise because
158 :     there is too much code that during installation relies on the presence
159 :     of these tools -- resulting in a tricky ordering problem.)
160 :    
161 :     ----------------------------------------------------------------------
162 :     Name: Matthias Blume (http://www.tti-c.org/blume/)
163 : blume 2277 Date: 2007/01/30 11:05:00 CST
164 :     Description:
165 :    
166 :     Added SMLofNJ.shiftArgs which is like a shell's "shift" command.
167 :     Modified CM's startup code to use shiftArgs as it processes
168 :     command line arguments. This way, the init code in each .sml-file
169 :     or library that is mentioned at the top level will see only
170 :     those arguments that have not yet been processed at this point.
171 :     In other words, the init code can "seize control" and process the
172 :     remaining command line.
173 :    
174 :     ----------------------------------------------------------------------
175 : jhr 2269 Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
176 : jhr 2358 Date: 2007/01/30
177 :     Description:
178 :    
179 :     Added fromList function to the ORD_SET interface and lookup to the
180 :     ORD_MAP interface. See the SML/NJ Library CHANGES file for details.
181 :    
182 :     ----------------------------------------------------------------------
183 :     Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
184 : jhr 2269 Date: 2007/01/28
185 :     Description:
186 :    
187 :     Added the UTF8 structure and signature from the Moby compiler to
188 :     the SML/NJ library (and the CML library). These modules will replace
189 :     the version in the ml-lpt-lib.
190 :    
191 :     ----------------------------------------------------------------------
192 : blume 2237 Name: Matthias Blume (http://www.tti-c.org/blume/)
193 : blume 2267 Date: 2007/01/26 15:10:00 CST
194 :     Description:
195 :    
196 :     Added entries to handle ml-lpt-lib.cm in installer.
197 :    
198 :     ----------------------------------------------------------------------
199 :     Name: Matthias Blume (http://www.tti-c.org/blume/)
200 : blume 2259 Date: 2006/12/15 11:10:00 CST
201 : blume 2237 Tag: release-110.61
202 :     Description:
203 :    
204 :     New working release (110.61). NEW BOOTFILES!
205 :    
206 : blume 2259 This supersedes the pre-brown-paper-bug release (see below).
207 :    
208 : blume 2237 ----------------------------------------------------------------------
209 : blume 2259 Name: Matthias Blume (http://www.tti-c.org/blume/)
210 :     Date: 2006/12/15 11:00:00 CST
211 :     Description:
212 :    
213 :     Fixed brown-paper-bag bug with CM's pathname handling, which made
214 :     installation fail under Win32.
215 :    
216 :     ----------------------------------------------------------------------
217 :     Name: Matthias Blume (http://www.tti-c.org/blume/)
218 :     Date: 2006/12/14 12:00:00 CST
219 :     Description:
220 :    
221 :     Obsolete!
222 :    
223 :     [New working release (110.61). NEW BOOTFILES!]
224 :    
225 :     ----------------------------------------------------------------------
226 : jhr 2234 Name: John Reppy (http://cs.uchicago.edu/~jhr)
227 :     Date: 2006/14/06 09:50:00 CST
228 :     Description:
229 :    
230 :     Fixed the code in runtime/c-libs/posix-tty/{tcgetattr,tcsetattr}.c to
231 :     get the c_cc termios data copied correctly. Also moved the allocation
232 :     of the string to avoid problems if it caused a GC.
233 :    
234 :     Thanks to Timothy Bourke for the bug report and fix.
235 :    
236 :     ----------------------------------------------------------------------
237 : blume 2232 Name: Matthias Blume (http://www.tti-c.org/blume)
238 :     Date: 2006/12/08 09:50:00 CST
239 :     Description:
240 :    
241 :     Added code to CM's "standard shell tool" implementation which
242 :     causes it to tolerate (with a warning) the situation where the shell
243 :     command fails (e.g., due to the shell command's non-existence)
244 :     as long as all target files exist.
245 :    
246 :     This makes it possible to, e.g., build ml-yacc from sources even if
247 :     svn checkout messed up the time stamps on files in such a way that
248 :     yacc.grm is younger than yacc.grm.sml or yacc.grm.sig. (Ml-yacc would
249 :     be needed to re-process yacc.grm, but obviously it might not yet be
250 :     available at that time.)
251 :    
252 :     ----------------------------------------------------------------------
253 : jhr 2228 Name: John Reppy (http://cs.uchicago.edu/~jhr)
254 :     Date: 2006/12/06 12:06:00 CST
255 :     Description:
256 :    
257 :     Fixed the types of recvVecFrom, recvVecFrom', recvVecFromNB, and
258 :     recvVecFromNB' in the SOCKET signature. This error is actually
259 :     in the SML Basis specification too.
260 :    
261 :     ----------------------------------------------------------------------
262 : mblume 1923 Name: Matthias Blume (http://www.tti-c.org/blume/)
263 : blume 2226 Date: 2006/12/05 22:40:00 CST
264 :     Description:
265 :    
266 :     CM now reports undefined anchors as errors and aborts execution
267 :     rather than silently pressing on using bogus values.
268 :    
269 :     ----------------------------------------------------------------------
270 :     Name: Matthias Blume (http://www.tti-c.org/blume/)
271 : blume 2224 Date: 2006/11/29 10:00:00 CST
272 :     Description:
273 :    
274 :     Use Say.vsay for printing the "[autoloading]" message, so
275 :     #set CM.Control.verbose false (or -Ccm.verbose=false) can be used
276 :     to suppress them.
277 :    
278 :     ----------------------------------------------------------------------
279 :     Name: Matthias Blume (http://www.tti-c.org/blume/)
280 : blume 2219 Date: 2006/11/10 11:00:00 CST
281 :     Description:
282 :    
283 :     Fixed bug in CM where "with:" specifications that affect compilation
284 :     (as opposed to parsing) were ignored.
285 :    
286 :     ----------------------------------------------------------------------
287 :     Name: Matthias Blume (http://www.tti-c.org/blume/)
288 : blume 2193 Date: 2006/11/09 14:30:00 CST
289 :     Tag: release-110.60
290 :     Description:
291 :    
292 :     New working release (110.60). NEW BOOTFILES!
293 :    
294 :     This is the first subversion-hosted release. There are also changes
295 :     to the directory layout. Here is a table:
296 :    
297 :     (All pathnames are relative to the SML/NJ "root" directory.)
298 :    
299 :     config -> config
300 :    
301 :     src/cm -> base/cm
302 :     src/compiler -> base/compiler
303 :     src/runtime -> base/runtime
304 :     src/system -> base/system
305 :    
306 :     src/cm/pgraph -> pgraph
307 :    
308 :     src/READMES -> base/READMES
309 :     src/$notefile -> base/NOTES/$notefile
310 :     for <notefile> in BOOT CVSNOTES CYGWININSTALL HISTORY INSTALL
311 :     MACOSXINSTALL WININSTALL
312 :    
313 :     src/smlnj-lib -> smlnj-lib
314 :     src/MLRISC -> MLRISC
315 :     ckit -> ckit
316 :     src/cml -> cml
317 :     src/eXene -> eXene
318 :     src/heap2asm -> heap2asm
319 :     src/lexgen -> lexgen
320 :     src/ml-burg -> ml-burg
321 :     src/ml-lex -> ml-lex
322 :     src/ml-yacc -> ml-yacc
323 :     src/ml-nlffi-lib -> nlffi/lib
324 :     src/ml-nlffigen -> nlffi/gen
325 :     src/smlnj-c -> smlnj-c
326 :     src/tools/TraceDebugProf -> trace-debug-profile
327 :    
328 :     In addition, there is also a new
329 :    
330 :     ml-lpt
331 :    
332 :     directory containing two new program generator tools: ml-ulex and
333 :     ml-antlr (a lexer generator that handles unicode and an ANTRL-inspired
334 :     LL(k) parser generator). These tools are currently "beta-quality"
335 :    
336 :     The latest versios of the sources can now be obtained anonymously
337 :     via subversion. For this, it is useful to first check out
338 :    
339 :     svn://smlnj-gforge.cs.uchicago.edu/smlnj/admin
340 :    
341 :     and put the resulting directory on your shell's PATH. This provides
342 :     access to three shell scripts: checkout-all.sh, stat-all.sh,
343 :     and refresh-all.sh.
344 :    
345 :     To create a freshly checked-out copy of the sources, do
346 :    
347 :     checkout-all.sh [dir]
348 :    
349 :     where dir is the optional SML/NJ root directory (default is ".").
350 :    
351 :     This creates the above directory layout. Each subdirectory of
352 :     the root is under individual subversion control. The stat-all.sh
353 :     and refresh-al.sh scripts apply "svn stat" or "svn update" to
354 :     each of these subtrees.
355 :    
356 :     ----------------------------------------------------------------------
357 :     Name: Matthias Blume (http://www.tti-c.org/blume/)
358 : blume 2155 Date: 2006/11/02 12:20:00 CST
359 :     Description:
360 :    
361 :     Reorganized directory layout.
362 :    
363 :     This is a temporary solution, more reorganization is to come.
364 :    
365 :     The basic idea is to have a number of toplevel trees, each corresponding
366 :     to a well-defined part of the overall system. Each part can be
367 :     maintained individually, even in separate source repositories,
368 :     although currently we still serve everything out of the main
369 :     smlnj-gforge tree.
370 :    
371 :     Installer and and scripts have been updated to reflect the new
372 :     layout. The installer (base/src/system/smlnj/installer) is now
373 :     "scriptable" to avoid burning too much knowledge about the
374 :     layout into SML source code. The main script used by the installer
375 :     is in config/actions.
376 :    
377 :     The main change is that many of the subdirectories of what used to
378 :     be known as "src" have moved to the toplevel. The "src" directory
379 :     itself has moved down into a subtree called "base". (We may eventually
380 :     get rid of the extra level of indirection represented by "src".)
381 :    
382 :     The layout is now as follows:
383 :    
384 :     toplevel tree name | default repository
385 :     | (using gf=svn://smlnj-gforge.cs.uchicago.edu/smlnj)
386 :     -------------------+----------------------------------------------------
387 :     config | $gf/config/trunk
388 :     |
389 :     base | $gf/sml/trunk
390 :     |
391 :     smlnj-lib | $gf/smlnj-lib/trunk
392 :     MLRISC | $gf/MLRISC/trunk
393 :     |
394 :     ml-yacc | $gf/ml-yacc/trunk
395 :     ml-lex | $gf/ml-lex/trunk
396 :     ml-burg | $gf/ml-burg/trunk
397 :     lexgen | $gf/lexgen/trunk
398 :     heap2asm | $gf/heap2asm/trunk
399 :     |
400 :     cml | $gf/cml/trunk
401 :     eXene | $gf/eXene/trunk
402 :     ckit | $gf/ckit/trunk
403 :     nlffi | $gf/nlffi/trunk
404 :     |
405 :     smlnj-c | $gf/smlnj-c/trunk
406 :    
407 :     In $gf/admin there are a few useful shell scripts for checking out
408 :     and maintaining the entire collection of trees:
409 :    
410 :     admin/checkout-all.sh [dir]
411 :     -- optionally creates dir and checks out all trees from their
412 :     default repositories; if dir is missing, checkout into .
413 :     admin/refresh-all.sh [dir]
414 :     -- looks at all trees (from the above list) in dir (default: .)
415 :     and runs "svn update" if the tree exists and is under
416 :     subversion control;
417 :     non-existing or non-subversion trees are skipped
418 :     admin/stat-all.sh [dir]
419 :     -- like refresh-all.sh, but runs "svn stat" instead
420 :    
421 :     ----------------------------------------------------------------------
422 :     Name: Matthias Blume (http://www.tti-c.org/blume/)
423 :     Date: 2006/10/05 11:30:00 CDT
424 : mblume 2066 Tag: blume-20061005-merge-amd64
425 :     Description:
426 :    
427 :     Merged code for AMD64 backend (Mike Rainey's work).
428 :     Everything is hooked up but untested.
429 :    
430 :     ----------------------------------------------------------------------
431 :     Name: Matthias Blume (http://www.tti-c.org/blume/)
432 : mblume 1931 Date: 2006/05/17 14:00:00 CST
433 :     Tag: Release_110_59
434 :     Description:
435 :    
436 : blume 2193 New working version (110.59). NEW BOOTFILES!
437 : mblume 1931
438 :     I am freezing 110.59. Changes other than the version-number increase:
439 :    
440 :     eXene:
441 :    
442 :     * committed changes to eXene from Alley Stoughton:
443 :     "fixed bugs in X authorization and resource handling, as well
444 :     as in the pile and viewport widgets"
445 :    
446 :     Runtime:
447 :     * fixed linking problem with NetBSD 3.x.
448 :    
449 :     Lexgen:
450 :     * lexgen tool handles non-ascii characters in 7-bit mode the same
451 :     way that ml-lex does
452 :    
453 :     * lexgen propagates exceptions the same way that ml-lex does
454 :    
455 :     CML:
456 :     * Fixed a bug in the SyncVar polling functions (iGetPoll,
457 :     mTakePoll, and mGetPoll) that could lead to livelock.
458 :    
459 :     ----------------------------------------------------------------------
460 :     Name: Matthias Blume (http://www.tti-c.org/blume/)
461 : mblume 1926 Date: 2006/05/12 15:20:00 CST
462 :     Tag: blume-20060515-makedepend
463 :     Description:
464 :    
465 :     Implemented ml-makedepend (i.e., CM.sources) in a better (more robust)
466 :     way. This should hopefully fix the ml-makedepend problem permanently.
467 :    
468 :     ----------------------------------------------------------------------
469 :     Name: Matthias Blume (http://www.tti-c.org/blume/)
470 : mblume 1925 Date: 2006/05/12 15:15:00 CST
471 :     Tag: blume-20060512-makedepend
472 :     Description:
473 :    
474 :     Fixed long-standing bug with ml-makedepend where it would output a
475 :     spurious dependency to a non-existing file. (This is a simple fix.
476 :     It might need further looking into.)
477 :    
478 :     ----------------------------------------------------------------------
479 :     Name: Matthias Blume (http://www.tti-c.org/blume/)
480 : mblume 1923 Date: 2006/04/20 10:30:00 CDT
481 :     Tag: blume-20060420-heap2exec
482 :     Description:
483 :    
484 :     Committed patches received from Johannes 5 Joemann (joemann@befree.free.de)
485 :     that enable heap2exec under Linux and FreeBSD.
486 :    
487 :     ----------------------------------------------------------------------
488 : jhr 1921 Name: John Reppy
489 :     Date: 2006/04/14
490 : blume 2155 Tag:
491 : jhr 1921 Description:
492 :    
493 :     MLRISC changes:
494 :     1) renamed GAS_PSEUDO_OPS to AS_PSEUDO_OPS and put it in its own file.
495 :     2) added support for NOTB and XORB operators in pseudo-op expressions
496 :     3) added DarwinPseudoOp functor that supports Darwin's assembler syntax.
497 :     4) added support for 64-bit integer literals
498 :    
499 :     ----------------------------------------------------------------------
500 : jhr 1922 Name: Matthias Blume (http://www.tti-c.org/blume/)
501 : mblume 1907 Date: yyyy/mm/dd
502 :     Tag: Release_110_58
503 :     Description:
504 :    
505 :     New working version (110.58). NEW BOOTFILES!
506 :    
507 :     ----------------------------------------------------------------------
508 : mblume 1896 Name: Matthias Blume (http://www.tti-c.org/blume/)
509 : mblume 1905 Date: 2006/03/01 22:15:00 CST
510 :     Tag: blume-20060301-lexgen
511 :     Description:
512 :    
513 :     Incorporated several bugfixes to lexgen. Compiler now compiles to
514 :     fixpoint when using lexgen instead of ml-lex.
515 :    
516 :     ----------------------------------------------------------------------
517 :     Name: Matthias Blume (http://www.tti-c.org/blume/)
518 : mblume 1902 Date: 2006/02/28 22:45:00 CST
519 :     Tag: blume-20060228-lexgen
520 :     Description:
521 :    
522 :     Removed ml-flex and added lexgen instead, using Aaron Turon's newly
523 :     provided tarball. The generated lexers still have problems.
524 :    
525 :     ----------------------------------------------------------------------
526 :     Name: Matthias Blume (http://www.tti-c.org/blume/)
527 : mblume 1898 Date: 2006/02/26 23:30:00 CST
528 :     Tag: blume-20060226-ml-flex
529 :     Description:
530 :    
531 :     Removed ml-flex' dependency on regexp-lib.cm. Turned installation
532 :     of ml-flex on by default.
533 :    
534 :     ----------------------------------------------------------------------
535 :     Name: Matthias Blume (http://www.tti-c.org/blume/)
536 : mblume 1896 Date: 2006/02/24 16:30:00 CST
537 :     Tag: blume-20060224-ml-flex
538 :     Description:
539 :    
540 :     Added ml-flex sources. Partially integrated, but should not be
541 :     turned on yet! (Read: leave it commented-out in config/targets!)
542 :    
543 :     ----------------------------------------------------------------------
544 : jhr 1893 Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
545 :     Date: 2006/02/23
546 :     Tag: jhr-20060223-x86-64
547 :     Description:
548 :     Changes to support building on x86-64 systems (using the 32-bit mode).
549 :     Also cleaned up signal handling on Linux. Support for pre-2.2 Linux
550 :     kernels dropped.
551 :    
552 :     ----------------------------------------------------------------------
553 : allenleung 1890 Name: Allen Leung (leunga (at) reservoir (dot) com)
554 :     Date: 2006/02/22 15:07:00 EST
555 :     Tag: leunga-20060222-mdl-gen
556 :     Description:
557 :    
558 :     CM has changed. Updated the script for rebuilding the MLRISC generated
559 :     files.
560 :    
561 :     ----------------------------------------------------------------------
562 : mblume 1886 Name: Matthias Blume (http://www.tti-c.org/blume/)
563 :     Date: 2006/02/14 17:30:00 CST
564 :     Tag: blume-20060214-intel-mac
565 :     Description:
566 :    
567 :     Hooked code for Darwin-specific Intel ABI into main compiler.
568 :     (This is a temporary solution which relies on the fact that the
569 :     compiler itself does not use NLFFI. Eventually we need to divorce
570 :     intel mac from generic x86 unix code and make separate sets of binaries.)
571 :    
572 :     ----------------------------------------------------------------------
573 : jhr 1882 Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
574 : jhr 1885 Date: 2006/02/13
575 :     Tag: jhr-20060213-intel-mac
576 :     Description:
577 :     Changed MLRISC x86 CCalls for partial support of MacOS X ABI.
578 :    
579 :     ----------------------------------------------------------------------
580 :     Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
581 : jhr 1882 Date: 2006/02/06
582 :     Tag: jhr-20060206-intel-mac
583 :     Description:
584 :     Changes to support MacOS X on Intel hardware. The C-calls support in
585 :     MLRISC must also be updated to support the MacOS X ABI.
586 :    
587 :     ----------------------------------------------------------------------
588 : mblume 1863 Name: Matthias Blume (http://www.tti-c.org/blume/)
589 : mblume 1879 Date: 2005/12/16 00:15:00 CST
590 :     Tag: blume-20051216-cm
591 :     Description:
592 :    
593 :     Improved error reporting and handling in CM.
594 :    
595 :     ----------------------------------------------------------------------
596 :     Name: Matthias Blume (http://www.tti-c.org/blume/)
597 : mblume 1878 Date: 2005/11/21 14:00:00 CST
598 :     Tag: blume-20051121-html-lib
599 :     Description:
600 :    
601 :     Enabling $/html-lib.cm in config/preloads did not work. This is fixed
602 :     now. (Since the anchor mapping for html-lib.cm is not yet in effect
603 : blume 2155 at the time when config/preloads is processed, the library has to
604 : mblume 1878 be referred to by another name. In 110.57 this would be
605 :     $SMLNJ-LIB/HTML/html-lib.cm. I arranged for $smlnj/smlnj-lib/html-lib.cm
606 :     to be valid as well.)
607 :    
608 :     Thanks to Todd Wilson (Fresno) for alerting me to this issue.
609 :    
610 :     ----------------------------------------------------------------------
611 :     Name: Matthias Blume (http://www.tti-c.org/blume/)
612 : mblume 1874 Date: 2005/11/19 13:45:00 CST
613 :     Tag: Release_110_57
614 :     Description:
615 :    
616 :     Moved release tag for 110.57.
617 :    
618 :     ----------------------------------------------------------------------
619 :     Name: Matthias Blume (http://www.tti-c.org/blume/)
620 : mblume 1873 Date: 2005/11/19 23:35:00 CST
621 :     Tag: blume-20051119-fish
622 :     Description:
623 :    
624 :     Fixed a problem in config/install.sh where it tries to "fish" the
625 :     name of the CM metadata directory from the wrong place (because
626 :     the physical location of basis.cm has changed). Also, corrected the
627 :     path anchor for $/html-lib.cm. (Thanks to M. Fluet for pointing out
628 :     these problems.)
629 :    
630 :     ----------------------------------------------------------------------
631 :     Name: Matthias Blume (http://www.tti-c.org/blume/)
632 : mblume 1867 Date: 2005/11/16 14:30:00 CST
633 : mblume 1874 Tag: blume-orig-110_57
634 : mblume 1867 Description:
635 :    
636 :     New working version (110.57). NEW BOOTFILES!
637 :    
638 : mblume 1874 UPDATE: The release tag has been moved to include a fix for a brown-paper-
639 :     bag glitch (see above).
640 :    
641 : mblume 1867 ----------------------------------------------------------------------
642 :     Name: Matthias Blume (http://www.tti-c.org/blume/)
643 : mblume 1866 Date: 2005/11/16 10:25:00 CST
644 :     Tag: blume-22051116-btrace
645 :     Description:
646 :    
647 :     Fixed problem with bogus exception message when using back-trace
648 :     facility.
649 :    
650 :     ----------------------------------------------------------------------
651 :     Name: Matthias Blume (http://www.tti-c.org/blume/)
652 : mblume 1865 Date: 2005/11/15 22:35:00 CST
653 :     Tag: blume-22051115-array2copy
654 :     Description:
655 :    
656 :     Added simple implementation of Array2.copy. (Warning: mostly untested.)
657 :    
658 :     ----------------------------------------------------------------------
659 :     Name: Matthias Blume (http://www.tti-c.org/blume/)
660 : mblume 1864 Date: 2005/11/15 21:35:00 CST
661 :     Tag: blume-22051115-versiontool
662 :     Description:
663 :    
664 :     Reversed change to src/system/smlnj/internals/versiontool.cm. This
665 :     file gets loaded as a tool -- by the equivalent of CM.make during the
666 :     run of CMB.make. Thus, CMB's path configuration is meaningless for
667 :     it. Instead, it has the status of "user code", so it should use
668 :     $/basis.cm to refer to the Basis library. (At least that's true for
669 :     the purpose of bootstrapping the previous change. In the future it
670 :     might make sense to have versiontool.cm refer to
671 :     $smlnj/basis/basis.cm, i.e., the version of the Basis that the
672 :     compiler itself uses.)
673 :    
674 :     Also patched src/system/testml to have it activate those extra anchor
675 :     bindings in config/extrapathconfig.
676 :    
677 :     ----------------------------------------------------------------------
678 :     Name: Matthias Blume (http://www.tti-c.org/blume/)
679 : mblume 1863 Date: 2005/11/15 16:35:00 CST
680 :     Tag: blume-22051115-libraries
681 :     Description:
682 :    
683 :     This change affects the way the following libraries are tied into
684 :     the system:
685 :    
686 :     $/basis.cm
687 :     $/smlnj-lib.cm
688 :     $/pp-lib.cm
689 :     $/controls-lib.cm
690 :     $/html-lib.cm
691 :     $/ml-yacc-lib.cm
692 :    
693 :     These libraries are now internally (as seen from the source code
694 :     of the implementation itself) known by the following names:
695 :    
696 :     $smlnj/basis/basis.cm
697 :     $smlnj/smlnj-lib/smlnj-lib.cm
698 :     $smlnj/smlnj-lib/pp-lib.cm
699 :     $smlnj/smlnj-lib/controls-lib.cm
700 :     $smlnj/smlnj-lib/html-lib.cm
701 :     $smlnj/ml-yacc/ml-yacc-lib.cm
702 :    
703 :     This makes it possible to work with code that requires different
704 :     versions of these libraries, and which refers to these libraries using
705 :     their "default" names (i.e., the first set of names above). In other
706 :     words, one can un-define or re-define those default names without
707 :     compromising the proper functioning of the compiler itself.
708 :    
709 :     A similar procedure had already been performed for several of the
710 :     MLRISC libraries that are linked into the compiler. I did some
711 :     cleanup on this code.
712 :    
713 :     A new file in the config directory (named extrapathconfig) is
714 :     responsible for setting up path anchors that the compiler itself does
715 :     not need, but that are typically required by user code.
716 :    
717 :     ----------------------------------------------------------------------
718 : mblume 1862 Name: Matthias Blume
719 :     Date: 2005/11/07 17:05:00 CST
720 :     Tag: blume-20051107-slice
721 :     Description:
722 :    
723 :     Fixed erroneous out-of-bounds test in the "update" function of
724 :     various *ArraySlice modules. (Thanks to Vesa A. Norrman for pointing
725 :     out the problem.)
726 :    
727 :     Pushed some Basis changes through ML-Lex, CML, and eXene.
728 :    
729 :     ----------------------------------------------------------------------
730 : jhr 1858 Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
731 : jhr 1861 Date: 2005/11/07
732 :     Description:
733 :     Fixed a Basis incompatibility: The depreciated function Substring.all
734 :     was removed (use Substring.full instead).
735 :    
736 :     ----------------------------------------------------------------------
737 :     Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
738 : jhr 1858 Date: 2005/11/05
739 :     Description:
740 :     Tweaked interval set API in SML/NJ library; see the CHANGES file for details.
741 :    
742 :     ----------------------------------------------------------------------
743 : mblume 1856 Name: Matthias Blume (http://www.tti-c.org/blume/)
744 :     Date: 2005/11/02 16:25:00 CST
745 :     Tag: blume-20051102-hex
746 :     Description:
747 :    
748 :     * Runtime system bootstrap code now accepts hex digits in BOOTLIST
749 :     in either upper case or lower case format.
750 :     * Pushed changes to names of Pack<N>{Big,Little} structures through
751 :     CML and eXene.
752 :    
753 :     ----------------------------------------------------------------------
754 : jhr 1852 Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
755 : jhr 1855 Date: 2005/11/02
756 :     Description:
757 :     Fixed a Basis incompatibility: Pack<N>{Big,Little} structures should
758 :     be called PackWord<N>{Big,Little}.
759 :    
760 :     ----------------------------------------------------------------------
761 :     Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
762 : jhr 1852 Date: 2005/10/28
763 :     Description:
764 :     Fixed a minor Basis incompatibility: hex digits should be upper case.
765 :    
766 :     ----------------------------------------------------------------------
767 : mblume 1846 Name: Matthias Blume (blume (at) tti - c (dot) org)
768 :     Date: 2005/10/25 16:00:00 CDT
769 :     Tag: Release_110_56
770 :     Description:
771 :    
772 :     New working version (110.56). NEW BOOTFILES!
773 :    
774 :     ----------------------------------------------------------------------
775 : jhr 1838 Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
776 : jhr 1840 Date: 2005/10/25
777 :     Description:
778 :     Added interval sets to utility library (signatures INTERVAL_DOMAIN
779 :     and INTERVAL_SET, and functor IntervalSetFn).
780 :    
781 :     ----------------------------------------------------------------------
782 :     Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
783 : jhr 1838 Date: 2005/10/14
784 :     Description:
785 :     Add Zhong Shao's fix for datatype equality functions.
786 :    
787 :     ----------------------------------------------------------------------
788 : allenleung 1836 Name: Allen Leung (leunga (at) reservoir (dot) com)
789 : allenleung 1837 Date: 2005/10/14 12:30:00 EST
790 :     Tag: leunga-20051014-x86-spill
791 :     Description:
792 :    
793 :     Bug fix a bug found by Carl Hauser.
794 :     There was a typo in the reload code for FCMP in x86SpillInstr.sml.
795 :    
796 :     ----------------------------------------------------------------------
797 :     Name: Allen Leung (leunga (at) reservoir (dot) com)
798 : allenleung 1836 Date: 2005/10/14 11:15:00 EST
799 :     Tag: leunga-20051014-x86-asm
800 :     Description:
801 :    
802 :     Removed some debugging code in file x86Asm.sml.
803 :     The function emit_operand was printing out debugging output.
804 :    
805 :     ----------------------------------------------------------------------
806 : jhr 1813 Name: John Reppy
807 : jhr 1831 Date: 2005/07/27
808 :     Tag:
809 :     Description:
810 :    
811 : jhr 1832 Fixed ml-lex to recognize "\r" as representing carriage return.
812 :    
813 :     ----------------------------------------------------------------------
814 :     Name: John Reppy
815 :     Date: 2005/07/27
816 :     Tag:
817 :     Description:
818 :    
819 : jhr 1831 Fixed ml-yacc to work on files with non-native end-of-line
820 :     encodings (e.g., Windows text file on a Unix system).
821 :    
822 :     ----------------------------------------------------------------------
823 :     Name: John Reppy
824 : jhr 1824 Date: 2005/07/20
825 :     Tag:
826 :     Description:
827 :    
828 :     Added changes from Dominic Evans (oldmanuk (at) gmail (dot) com)
829 :     to support HPUX 11.
830 :    
831 :     ----------------------------------------------------------------------
832 :     Name: John Reppy
833 : jhr 1813 Date: 2005/07/06
834 :     Tag:
835 :     Description:
836 :    
837 :     Changes to the SML/NJ library. See smlnj-lib/CHANGES for details.
838 :    
839 :     ----------------------------------------------------------------------
840 : mblume 1808 Name: Matthias Blume (blume (at) tti - c (dot) org)
841 :     Date: 2005/07/06 00:45:00 CDT
842 :     Tag: blume-20050706-slice-copy
843 :     Description:
844 :    
845 :     Fixed reversed logic for deciding whether to "copy up" or "copy down"
846 :     in *-array-slice.sml.
847 :    
848 :     ----------------------------------------------------------------------
849 : allenleung 1806 Name: Allen Leung (leunga (at) reservoir (dot) com)
850 : allenleung 1807 Date: 2005/05/31 17:00:00 EST
851 :     Tag: leunga-20050531-cygwin-fault-2
852 :     Description:
853 :    
854 :     A typo in the cygwin code fixed.
855 :    
856 :     ----------------------------------------------------------------------
857 :     Name: Allen Leung (leunga (at) reservoir (dot) com)
858 : allenleung 1806 Date: 2005/05/31 16:47:00 EST
859 :     Tag: leunga-20050531-cygwin-fault
860 :     Description:
861 :    
862 : blume 2155 Updated Cygwin's fault/signal handling to match the Windows version.
863 : allenleung 1806 Updated the export list.
864 :    
865 :     ----------------------------------------------------------------------
866 : mblume 1791 Name: Matthias Blume (blume (at) tti - c (dot) org)
867 : mblume 1795 Date: 2005/05/18 13:35:00 CDT
868 :     Tag: Release_110_54
869 :     Description:
870 :    
871 :     New working version (110.54). NEW BOOTFILES!
872 :    
873 :     ----------------------------------------------------------------------
874 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
875 : mblume 1794 Date: 2005/05/18 11:58:00 CDT
876 :     Tag: blume-20050518-installer
877 :     Description:
878 :    
879 :     Added support scripts for Mac OS X PackageMaker and modified
880 :     config/install.sh so that it supports re-dumping a heap image
881 :     after customization.
882 :    
883 :     ----------------------------------------------------------------------
884 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
885 : mblume 1793 Date: 2005/05/18 10:55:00 CDT
886 :     Tag: blume-20050518-realdiv-noovld
887 :     Description:
888 :    
889 :     Un-overloaded / to work around bug in overloading resolution code.
890 :    
891 :     ----------------------------------------------------------------------
892 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
893 : mblume 1792 Date: 2005/05/16 23:50:00 CDT
894 :     Tag: blume-20050516-redump-heap
895 :     Description:
896 :    
897 :     Added mechanism for re-creating a heap file for the interactive system
898 :     after configuration variables have been changed.
899 :    
900 :     CM.redump_heap : string -> unit
901 :    
902 :     This is much like SMLofNJ.exportML, but starting from the resulting
903 : blume 2155 heap does not return to the caller of CM.redump_heap but
904 : mblume 1792 restarts the interactive system from scratch. The original call of
905 :     CM.redump_heap does not return but ends the interactive session.
906 :     Thus, CM.redump_heap is a lot like SMLofNJ.exportFn.
907 :    
908 :     Internally, redump_heap winds the dynamic execution context back to
909 :     the point where the original heap image was created and re-executes
910 :     the heap image generation code in the boot code.
911 :    
912 :     ----------------------------------------------------------------------
913 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
914 : mblume 1791 Date: 2005/05/09 21:55:00 CDT
915 :     Tag: blume-20050509-word64
916 :     Description:
917 :    
918 :     Added a hack to the existing hack known as Word64 to make fromString
919 :     behave correctly. I am still not sure whether Word64.scan will work
920 :     as specified with respect to the interaction of radix and prefix.
921 :    
922 :     ----------------------------------------------------------------------
923 : allenleung 1785 Name: Allen Leung (leunga (at) reservoir (dot) com)
924 :     Date: 2005/05/04 11:50:00 EST
925 :     Tag: leunga-20050504-checkgc
926 :     Description:
927 :    
928 : blume 2155 Added a gc protocol checking phase. This phase is enabled with
929 : allenleung 1785 the flag "check-gc". "debug-check-gc" turns on the verbose mode.
930 :    
931 :     ----------------------------------------------------------------------
932 : mblume 1741 Name: Matthias Blume (blume (at) tti - c (dot) org)
933 : mblume 1784 Date: 2005/05/04 10:45:00 CDT
934 :     Tag: blume-20050504-intinf
935 :     Description:
936 :    
937 :     Fixed a bug in the implementation of div and mod for IntInf.
938 :     Thanks to Neophytos Michael for reporting the problem.
939 :    
940 :     ----------------------------------------------------------------------
941 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
942 : mblume 1783 Date: 2005/05/04 10:35:00 CDT
943 :     Tag: blume-20050504-join
944 :     Description:
945 :    
946 :     Added a "join" combinator to the ParserComb module in smlnj-lib.cm.
947 :    
948 :     ----------------------------------------------------------------------
949 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
950 : mblume 1770 Date: 2005/02/28 23:40:00 CST
951 :     Tag: blume-20050228-mVar
952 :     Description:
953 :    
954 :     Fixed serious bug (brown paper bag variety) in new implementation of
955 :     structure Atom in CML. (I had accidentally used a mailbox instead of
956 :     an mvar, leaving the door open for races.)
957 :    
958 :     ----------------------------------------------------------------------
959 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
960 : mblume 1761 Date: 2005/02/25 15:00:00 CST
961 :     Tag: Release_110_53
962 :     Description:
963 :    
964 :     New working version (110.53). NEW BOOTFILES!
965 :    
966 :     ----------------------------------------------------------------------
967 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
968 : mblume 1760 Date: 2005/02/25 14:50:00 CST
969 :     Tag: blume-20050225-susp
970 :     Description:
971 :    
972 :     Brought back SMLofNJ.Susp. The underlying suspension type is the one
973 :     implemented in Core -- which means that it is the same as the one used
974 :     by the lazy extension.
975 :    
976 :     ----------------------------------------------------------------------
977 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
978 : mblume 1759 Date: 2005/02/24 16:50:00 CST
979 :     Tag: blume-20050224-cml-atom
980 :     Description:
981 :    
982 :     Simpler and at the same time more general implementation of structure
983 :     Atom in CML.
984 :    
985 :     ----------------------------------------------------------------------
986 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
987 : mblume 1758 Date: 2005/02/15 17:35:00 CST
988 :     Tag: blume-20050215-tools
989 :     Description:
990 :    
991 :     Created new "tools" directory under "src" and moved "TraceDebugProf"
992 :     there.
993 :    
994 :     ----------------------------------------------------------------------
995 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
996 : mblume 1755 Date: 2005/02/10 17:55:00 CST
997 :     Tag: blume-20050210-longlong
998 :     Description:
999 :    
1000 :     Implemented "long long" arguments and results for NLFFI. (Only the
1001 :     PPC/MacOS implementation is complete, the other backends still need to
1002 :     be updated.)
1003 :    
1004 :     ----------------------------------------------------------------------
1005 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1006 : mblume 1751 Date: 2005/01/24 17:40:00 CST
1007 :     Tag: blume-20050124-mlyacc
1008 :     Description:
1009 :    
1010 :     Minor cleanup in ML-Yacc rule printing mechanism. This should fix a
1011 :     problem with certain "as" patterns which previously got rendered
1012 :     using incorrect syntax.
1013 :    
1014 :     ----------------------------------------------------------------------
1015 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1016 : mblume 1750 Date: 2005/01/18 12:00:00 CST
1017 :     Tag: blume-20050118-profile
1018 :     Description:
1019 :    
1020 :     Made time profiling code (interrupt handler) in runtime system aware
1021 :     of new array representation.
1022 :    
1023 :     ----------------------------------------------------------------------
1024 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1025 : mblume 1747 Date: 2005/01/14 18:00:00 CST
1026 :     Tag: blume-20050114-heap2exec
1027 :     Description:
1028 :    
1029 :     Implemented new (but still experimental) heap2exec facility. This is
1030 :     tested under Mac OS X and should work under Linux (will test shortly).
1031 :     It will probably also work on the Sparc (will test some time later).
1032 :    
1033 :     - removed old "HACKED_STANDALONE" hack from runtime
1034 :    
1035 :     To be able to test this, uncomment the request for "heap2asm" in
1036 :     config/targets prior to installation. (Notice that this is different
1037 :     from "heap2exec" mentioned below. Not a typo.)
1038 :    
1039 :     To perform an actual test, run the command
1040 :    
1041 :     $ bin/heap2exec heapfile execfile
1042 :    
1043 :     (You can put heap2exec on your shell's path.)
1044 :    
1045 :     For example, run
1046 :    
1047 :     $ bin/heap2exec bin/.heap/ml-yacc.ppc-darwin mly
1048 :    
1049 :     This will create a standalone executable called "mly" which you can
1050 :     then invoke directly as a command.
1051 :    
1052 :     ----------------------------------------------------------------------
1053 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1054 : mblume 1746 Date: 2005/01/07 11:44:00 CST
1055 :     Tag: blume-20050107-mlstring
1056 :     Description:
1057 :    
1058 :     fixed off-by-one error in ML_STRING macro (globals.c)
1059 :    
1060 :     ----------------------------------------------------------------------
1061 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1062 : mblume 1744 Date: 2004/12/23 18:00:00 CST
1063 :     Tag: blume-20041223-santa
1064 :     Description:
1065 :    
1066 :     Made ml-build script "smarter" (but only very little).
1067 :    
1068 :     ----------------------------------------------------------------------
1069 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1070 : mblume 1742 Date: 2004/12/21 15:05:00 CST
1071 :     Tag: blume-20041221-longlong
1072 :     Description:
1073 :    
1074 :     * Implemented access to signed and unsigned long long data in NLFFI.
1075 :     (The parameter-passing part of the picture has not complete. But
1076 :     data structure access seems to work.)
1077 :    
1078 :     * Fixed CM's incorrect assumption that the PPC is little-endian.
1079 :     (On the Mac, it is big-endian. And that's currently our only
1080 :     PPC platform.)
1081 :    
1082 :     ----------------------------------------------------------------------
1083 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1084 : mblume 1741 Date: 2004/12/21 12:50:00 CST
1085 :     Tag: blume-20041221-memory
1086 :     Description:
1087 :    
1088 :     Some cleanup in the $c/memory.cm library: separated some concerns by
1089 :     moving allocation code and memory access code each into their own
1090 :     files.
1091 :    
1092 :     ----------------------------------------------------------------------
1093 : allenleung 1740 Name: Allen Leung (leunga (at) reservoir (dot) com)
1094 :     Date: 2004/12/17 16:12:00 EST
1095 :     Tag: leunga-20041217-cygwin-smlnj-home
1096 :     Description:
1097 :    
1098 : blume 2155 The Unix I/O library of SML/NJ on cygwin does not understand
1099 :     Windows style pathname, so problems arise when SMLNJ_HOME is set
1100 :     to a Windows style pathname. _run-sml now converts SMLNJ_HOME
1101 :     to a POSIX pathname on cygwin.
1102 : allenleung 1740
1103 :     ----------------------------------------------------------------------
1104 : mblume 1737 Name: Matthias Blume (blume (at) tti - c (dot) org)
1105 : mblume 1736 Date: 2004/12/16 13:35:00 CST
1106 :     Tag: Release_110_52
1107 :     Description:
1108 :    
1109 :     Last-minute changes incorporated into 110.52. Release tag moved.
1110 :    
1111 :     The changes:
1112 :     - HashString.hashString' -> HashString.hashSubstring
1113 :     - bug fix in UnivariateStats
1114 :    
1115 :     ----------------------------------------------------------------------
1116 : mblume 1702 Name: Matthias Blume (blume (at) tti - c (dot) org)
1117 : mblume 1733 Date: 2004/12/15 23:40:00 CST
1118 :     Tag: blume-20041215-hashSubstring
1119 :     Description:
1120 :    
1121 :     - HashString.hashString' -> HashString.hashSubstring
1122 :     - corresponding changes in atom.sml
1123 :     - "de-compressed" (aka. un-obfuscated) code for UnivariateStats and
1124 :     added some comments
1125 :    
1126 :     ----------------------------------------------------------------------
1127 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1128 : mblume 1728 Date: 2004/12/15 15:30:00 CST
1129 : mblume 1736 Tag: (Release_110_52)
1130 : mblume 1728 Description:
1131 :    
1132 :     New working version (110.52). NEW BOOTFILES!
1133 :    
1134 :     ----------------------------------------------------------------------
1135 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1136 : mblume 1727 Date: 2004/12/15 12:45:00 CST
1137 :     Tag: blume-20041215-spaces
1138 :     Description:
1139 :    
1140 :     More on the space problem (this time for Win32).
1141 :    
1142 :     ----------------------------------------------------------------------
1143 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1144 : mblume 1725 Date: 2004/12/14 17:30:00 CST
1145 :     Tag: blume-20041214-spaces
1146 :     Description:
1147 :    
1148 :     Hacked some of the scripts (in particular: the installer) to cope with
1149 :     spaces in filenames a bit better. But beware: the current "solution"
1150 :     is likely still full of bugs and inherently incomplete. (We need to
1151 :     do away with those shell scripts for a comprehensive solution.)
1152 :    
1153 :     ----------------------------------------------------------------------
1154 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1155 : mblume 1724 Date: 2004/12/13 14:45:00 CST
1156 :     Tag: blume-20041213-ml-makedepend
1157 :     Description:
1158 :    
1159 :     Fixed bug in code for ml-makedepend.
1160 :    
1161 :     ----------------------------------------------------------------------
1162 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1163 : mblume 1720 Date: 2004/12/09 16:30:00 CST
1164 :     Tag: blume-20041209-statistics
1165 :     Description:
1166 :    
1167 :     Added two simple but potentially useful statistics modules to SML/NJ Library.
1168 :     (See CHANGES file there.)
1169 :    
1170 :     ----------------------------------------------------------------------
1171 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1172 : mblume 1718 Date: 2004/12/01 16:50:00 CST
1173 :     Tag: blume-20041201-atom
1174 :     Description:
1175 :    
1176 :     smlnj-lib:
1177 :    
1178 :     Added function HashString.hashString' for substrings.
1179 :     Hand-inlined CharVector.foldl into HashString (for speed).
1180 :     Modified implementation of structure Atom to avoid extracting
1181 :     strings from substrings unless necessary.
1182 :     (Also see CHANGES file for smlnj-lib.)
1183 :    
1184 :     ----------------------------------------------------------------------
1185 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1186 : mblume 1717 Date: 2004/11/24 22:15:00 CST
1187 :     Tag: blume-20041124-cml
1188 :     Description:
1189 :    
1190 :     Made sure CML compiles when Position = Int64.
1191 :    
1192 :     ----------------------------------------------------------------------
1193 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1194 : mblume 1715 Date: 2004/11/24 14:30:00 CST
1195 :     Tag: blume-20041124-position
1196 :     Description:
1197 :    
1198 :     The compiler can now be compiled in a mode that makes structure Position
1199 :     equal to Int64. The default, however, is unchanged (Position = Int31)
1200 :     for the time being.
1201 :    
1202 :     To enable 64-bit positions, use the following procedure:
1203 :    
1204 :     1. Start sml
1205 :     2. Autoload $smlnj/cmb.cm (if not already autoloaded)
1206 :     3. Type
1207 :     #set (CMB.symval "USE_64_BIT_POSITIONS") (SOME 1);
1208 :     4. Run CMB.make() as usual.
1209 :    
1210 :     This is barely tested. The only test so far was a little SML program
1211 :     counting the number of characters in an 8-gigabyte file by
1212 :     reading it character-by-character. That test was successful.
1213 :    
1214 :     In support of 64-bit positions, a number of new functions have been
1215 :     added to the runtime system.
1216 :    
1217 :     ----------------------------------------------------------------------
1218 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1219 : mblume 1716 Date: 2004/11/23 14:45:00 CST
1220 :     Tag: blume-20041123-useFile
1221 : mblume 1712 Description:
1222 :    
1223 :     Fixed a problem with unhelpful error messages related to problems with
1224 :     .cm- or .sml files that appear as part of the sml command line.
1225 :    
1226 :     ----------------------------------------------------------------------
1227 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1228 : mblume 1703 Date: 2004/11/18 15:40:00 CST
1229 :     Tag: Release_110_51
1230 :     Description:
1231 :    
1232 :     New working version (110.51). NEW BOOTFILES!
1233 :    
1234 :     ----------------------------------------------------------------------
1235 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1236 : mblume 1702 Date: 2004/11/18 15:35:00 CST
1237 :     Tag: <none>
1238 :     Description:
1239 :    
1240 :     Enabled dlopen and friends for FreeBSD (as recommended by Johannes 5
1241 :     Joemann).
1242 :    
1243 :     ----------------------------------------------------------------------
1244 : allenleung 1701 Name: Allen Leung (leunga (at) reservoir (dot) com)
1245 :     Date: 2004/11/17 16:05:21 EST 2004
1246 :     Tag: leunga-20041117-mlrisc-live-kill
1247 :     Description:
1248 :    
1249 : blume 2155 Added support for MLTree constructs LIVE and KILL
1250 : allenleung 1701 to all the architectures.
1251 :    
1252 :     ----------------------------------------------------------------------
1253 : mblume 1687 Name: Matthias Blume (blume (at) tti - c (dot) org)
1254 : mblume 1689 Date: 2004/11/13 00:20:00 CST
1255 :     Tag: blume-20041113-versiontool
1256 :     Description:
1257 :    
1258 :     - Stripped down the versiontool: It now only handles the version number.
1259 :     The date string is generated at bootstrap time (during makeml).
1260 :    
1261 :     - In a previous commit, fixed a minor issue with how polyequal is being
1262 :     translated. In particular, the code now "looks through" abstractions.
1263 :     This results in slightly fewer polyEqual warnings and hopefully slightly
1264 :     more efficient code. Important examples for where this matters are
1265 :     the new int64 and word64 types.
1266 :    
1267 :     ----------------------------------------------------------------------
1268 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1269 : mblume 1687 Date: 2004/11/12 00:30:00 CST
1270 :     Tag: blume-20041112-int64
1271 :     Description:
1272 :    
1273 :     Structure Int64 fully hooked in. (The implementation is not very
1274 :     efficient, though.)
1275 :    
1276 :     ----------------------------------------------------------------------
1277 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1278 : mblume 1686 Date: 2004/11/11 17:30:00 CST
1279 :     Tag: blume-20041111-more64
1280 :     Description:
1281 :    
1282 :     All the pieces of Word64 are now there, with the exception of the
1283 :     conversions from and to LargeWord. (Eventually these need to be identities,
1284 :     but for the time being they don't even make sense because LargeWord is
1285 :     32-bit wide.)
1286 :    
1287 :     Also started to add similar support for Int64, but major pieces of that
1288 :     are still missing.
1289 :    
1290 :     ----------------------------------------------------------------------
1291 : mblume 1668 Name: Matthias Blume (blume (at) tti - c (dot) org)
1292 : mblume 1684 Date: 2004/11/11 00:15:00 CST
1293 :     Tag: blume-20041111-word64
1294 :     Description:
1295 :    
1296 :     Structure Word64 is now (almost) complete, word literals and patterns
1297 :     seem to work. There are a few odd pieces missing. In particular,
1298 :     I didn't do the {from,to}LargeWord parts because LargeWord is still Word32
1299 :     at the moment.
1300 :    
1301 :     Making Word64 official would mean that LargeWord becomes Word64. But
1302 :     this requires extreme care because most word-word conversions have to
1303 :     go through LargeWord, so making a mistake means loss of efficiency or
1304 :     worse. Eventually there will be a solution similar to (but actually
1305 :     simpler than) what I did with IntInf.
1306 :    
1307 :     ----------------------------------------------------------------------
1308 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1309 : mblume 1683 Date: 2004/11/10 18:12:00 CST
1310 :     Tag: blume-20041110-64bit
1311 :     Description:
1312 :    
1313 :     More 64-bit hacking (but still not even half-way there yet).
1314 :     Also, some assorted improvements to the handling of 8-bit words.
1315 :    
1316 :     ----------------------------------------------------------------------
1317 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1318 : mblume 1682 Date: 2004/11/09 17:50:00 CST
1319 : mblume 1683 Tag: <none>
1320 : mblume 1682 Description:
1321 :    
1322 :     Started adding some infrastructure for supporting 64-bit int- and
1323 :     word-types. (Still in its very early stages.)
1324 :    
1325 :     ----------------------------------------------------------------------
1326 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1327 : mblume 1672 Date: 2004/10/28 10:45:00 CDT
1328 :     Tag: Release_110_50
1329 :     Description:
1330 :    
1331 :     New working version (110.50). NEW BOOTFILES!
1332 :    
1333 :     =====================
1334 :    
1335 :     Also:
1336 :    
1337 :     - Changend config/srcarchiveurl from a file just
1338 :     containing the URL string into a file containing
1339 :     shell script code. The code has access to the $VERSION variable.
1340 :     - Made corresponding changes to config/install.sh and config/unpack.
1341 :     - Default contents of config/srcarchiveurl uses $VERSION and
1342 :     normally does not have to be edited to reflect a version change.
1343 :    
1344 :     (As a result, a version change can be done by just editing
1345 :     config/version, the rest is now automatic.)
1346 :    
1347 :     ----------------------------------------------------------------------
1348 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1349 : mblume 1669 Date: 2004/10/27 17:50:00 CDT
1350 :     Tag: blume-20041027-btrace-msg
1351 :     Description:
1352 :    
1353 :     BackTrace.monitor now also reports the source of the exception that
1354 :     triggered the trace.
1355 :    
1356 :     ----------------------------------------------------------------------
1357 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1358 : mblume 1668 Date: 2004/10/27 17:20:00 CDT
1359 :     Tag: blume-20041027-x86-c-calls
1360 :     Description:
1361 :    
1362 :     This is the HISTORY entry for two earlier commits, both concerning
1363 :     the x86 c-calls code in MLRISC:
1364 :    
1365 :     - added a missing LOAD in the code that deals with struct arguments
1366 :     - made sure the caller does not add the wrong number of bytes to the
1367 :     stack pointer after a call of a function returning a struct
1368 :     (the callee already pops the implicit argument which points to
1369 :     the space reserved for the result)
1370 :    
1371 :     ----------------------------------------------------------------------
1372 : allenleung 1662 Name: Allen Leung (leunga (at) reservoir (dot) com)
1373 :     Date: 2004/10/24 14:00:00 EST
1374 :     Tag: leunga-20041024-x86-gas-fucomip
1375 :     Description:
1376 :    
1377 :     John discovered a bug in the syntax of fucomip.
1378 :     The opcodes FU?COMIP? have been changed to
1379 :    
1380 :     fu?comip? %st(i), %st
1381 :    
1382 :     ----------------------------------------------------------------------
1383 : mblume 1622 Name: Matthias Blume (blume (at) tti - c (dot) org)
1384 : mblume 1656 Date: 2004/10/20 15:06:00 CDT
1385 :     Tag: blume-20041020-standalone-backtrace
1386 :     Description:
1387 :    
1388 :     Added a mechanism for getting back-trace information from standalone
1389 :     programs. Here is how it works:
1390 :    
1391 :     1. The part of the program from which you want to get backtrace
1392 :     information (usually the whole program) should be wrapped with
1393 :     BackTrace.monitor. This is a (unit->'a)->'a function, and your
1394 :     main program could be modified from something like
1395 :    
1396 :     fun main (pgm, args) = ...
1397 :    
1398 :     to
1399 :    
1400 :     fun main (pgm, args) = BackTrace.monitor (fn () => ...)
1401 :    
1402 :     2. To be able to access BackTrace.monitor, you have to add
1403 :     library $smlnj-tdp/plugins.cm to the .cm file that contains your
1404 :     main function.
1405 :    
1406 :     3. Remove all compiled code (i.e., all the .cm/ subdirectories that
1407 :     CM might have created in the past for your project).
1408 :    
1409 :     4. Build the system using this command line:
1410 :    
1411 :     ml-build -Ctdp.instrument=true \$smlnj-tdp/back-trace.cm \
1412 :     myprog.cm MyProg.main myprog
1413 :    
1414 :     instead of the usual
1415 :    
1416 :     ml-build myprog.cm MyProg.main myprog
1417 :    
1418 :     I changed a library name:
1419 :    
1420 :     $/trace-debug-profile.cm --> $smlnj-tdp/plugins.cm
1421 :    
1422 :     New libraries:
1423 :    
1424 :     $smlnj-tdp/back-trace.cm
1425 :     -- when loaded causes the back-trace plugin to be installed
1426 :     $smlnj-tdp/coverage.cm
1427 :     -- when loaded causes the coverage plugin to be installed
1428 :    
1429 :     ----------------------------------------------------------------------
1430 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1431 : mblume 1655 Date: 2004/10/18 16:45:00 CDT
1432 :     Tag: blume-20041018-groupowner
1433 :     Description:
1434 :    
1435 :     Added an "obsolete" warning for the "group owner" syntax to CM's parser.
1436 :    
1437 :     Eliminated group owner specs from .cm files throughout the source tree.
1438 :    
1439 :     ----------------------------------------------------------------------
1440 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1441 : mblume 1653 Date: 2004/10/15 15:45:00 CDT
1442 :     Tag: blume-20041015-coverage
1443 :     Description:
1444 :    
1445 :     * Test coverage tool added!
1446 :     * Further reorganization of tracing-, debugging-, and profiling support:
1447 :    
1448 :     - moved original BTImp -- now called BackTrace -- into a separate
1449 :     library called $/trace-debug-profile.cm
1450 :     - eliminated all mentions of BTrace from SMLofNJ.Internals
1451 :     - only the instrumentation mechanism is now left in the compiler proper
1452 :     - BackTrace module is a plugin which is NOT plugged in by default
1453 :     - Coverage module is another such plugin
1454 :    
1455 :     To get the benefits of any of these plugin modules, the code in
1456 :     question must be compiled with tdp instrumentation turned on. This
1457 :     can be done by setting SMLofNJ.Internals.TDP.mode to true.
1458 :     (The ref cell is also controlled via the -Ctdp.instrument=... switch.)
1459 :    
1460 :     Plugins are selected at link time. (Pre-compiled instrumented code
1461 :     can be re-loaded with different plugins in effect.) When an
1462 :     instrumented module is linked, whatever plugins are at that time
1463 :     enabled will come into effect for that module.
1464 :    
1465 :     To enable the back-trace plugin, load library $/trace-debug-profile.cm
1466 :     and invoke BackTrace.install() (e.g., from the interactive prompt).
1467 :     To enable the coverage plugin, load the same library and invoke
1468 :     Coverage.install().
1469 :    
1470 :     Back-traces are generated automatically on uncaught exceptions and
1471 :     when the code in question explicitly invokes BackTrace.trigger().
1472 :    
1473 :     Coverage (and execution frequency-) information must be queried
1474 :     explicitly by calling Coverage.not_covered and Coverage.hot_spots.
1475 :    
1476 :     ----------------------------------------------------------------------
1477 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1478 : mblume 1652 Date: 2004/10/14 17:40:00 CDT
1479 :     Tag: blume-20041014-tdp-core
1480 :     Description:
1481 :    
1482 :     Snapshot of a significant overhaul of how the trace/debug/profile support
1483 :     is hooked into the system (specifically: Core and SMLofNJ.Internals).
1484 :    
1485 :     ----------------------------------------------------------------------
1486 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1487 : mblume 1650 Date: 2004/10/13 16:34:00 CDT
1488 :     Tag: blume-20041013-tdp
1489 :     Description:
1490 :    
1491 :     Some rationalization of names:
1492 :    
1493 :     structure BTrace -> structure TDPInstrument
1494 :     etc.
1495 :    
1496 :     This is is preparation of using the original back-trace
1497 :     instrumentation for other purposes. "TDP" stands for
1498 :     Trace/Debug/Profile.
1499 :    
1500 :     The control flag controlling whether instrumentation is on or off is now
1501 :     registered under a different name, so instead of running sml as
1502 :    
1503 :     sml -Cinstrument.btrace-mode=true
1504 :    
1505 :     one has to say
1506 :    
1507 :     sml -Ctdp.instrument=true
1508 :    
1509 :     ----------------------------------------------------------------------
1510 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1511 : mblume 1645 Date: 2004/10/11 16:37:00 CDT
1512 :     Tag: blume-20041011-regions
1513 :     Description:
1514 :    
1515 :     Made some minor modifications to elabcore.sml to have source regions
1516 :     be propagated more tightly -- resulting in better (i.e., smaller)
1517 :     regions being reported in error- and debug messages.
1518 :    
1519 :     ----------------------------------------------------------------------
1520 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1521 : mblume 1644 Date: 2004/10/08 22:50:00 CDT
1522 :     Tag: blume-20041008-cmkw
1523 :     Description:
1524 :    
1525 :     Fixed handling of keywords in .cm files: After seeing "is" the lexer
1526 :     treats subsequent occurrences of "group", "library", "source", "is",
1527 :     "*", and "-" as ordinary identifiers rather than keywords.
1528 :    
1529 :     Most seriously, this fixes a problem with CM's "shell" tool. The tool
1530 :     is supposed to accept a tool argument called "source", but this did
1531 :     not work because of the clash with the keyword.
1532 :    
1533 :     ----------------------------------------------------------------------
1534 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1535 : mblume 1642 Date: 2004/10/07 16:00:00 CDT
1536 :     Tag: blume-20041007-cleanup
1537 :     Description:
1538 :    
1539 :     Assorted cleanup work:
1540 :    
1541 :     - got rid of intstrmap in favor of using the library's
1542 :     hash table implementation
1543 :     - threw out most of the pathnames stuff, as it was not used anyway
1544 :     - simplified tokentable implementation
1545 :     - fixed some minor spelling errors
1546 :    
1547 :     ----------------------------------------------------------------------
1548 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1549 : mblume 1641 Date: 2004/10/06 15:15:15 CDT
1550 :     Tag: blume-20041006-handler
1551 :     Description:
1552 :    
1553 :     Cleaned up the absyn to reflect the invariant that HANDLE always
1554 :     carries a FNexp as part of the type definition. This eliminates some
1555 :     superfluous sanity checks at runtime down the road.
1556 :    
1557 :     Some minor cleanup of the btrace code.
1558 :    
1559 :     ----------------------------------------------------------------------
1560 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1561 : mblume 1640 Date: 2004/10/01 10:20:30 CDT
1562 :     Tag: blume-20041001-slave
1563 :     Description:
1564 :    
1565 :     Added hack to make slave mode work in the presence of the version
1566 :     tool. (Still, since the master does two passes over the code for
1567 :     CMB.make, the release number gets bumped twice when slaves are
1568 :     attached. I don't know if this is worth fixing...)
1569 :    
1570 :     ----------------------------------------------------------------------
1571 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1572 : mblume 1639 Date: 2004/09/30 10:55:00 CDT
1573 :     Tag: blume-20040930-version
1574 :     Description:
1575 :    
1576 :     * Moved the "version" magic into its own little library under
1577 : mblume 1672 src/system/smlnj/internal. This avoids expensive reconstruction of
1578 :     a stable src/compiler/core.cm.
1579 : mblume 1639
1580 : mblume 1672 * At the same time, structure CompilerVersion is now known as
1581 :     structure SMLNJVersion.
1582 : mblume 1639
1583 :     * Arranged for the version tool to NOT kick in when rebuilding the system
1584 :     (makeml -rebuild, fixpt). Otherwise one would never reach a fixpoint.
1585 :     Also, loading the versiontool does not work when rebuilding the system
1586 :     because CM is not properly initialized at that time.
1587 :    
1588 :     ----------------------------------------------------------------------
1589 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1590 : mblume 1638 Date: 2004/09/29 14:00:00 CDT
1591 :     Tag: blume-20040929-autoversion
1592 :     Description:
1593 :    
1594 :     Implemented some CM magic to have
1595 :     file src/compiler/TopLevel/main/version.sml
1596 :     generated automagically.
1597 :     The version is taken from two files: config/version and config/release.
1598 :     The first is expected to contain a two-part version number such as 110.49.
1599 :     The second should contain a single number, but it may be missing.
1600 :    
1601 :     If the environment variable VERSIONTOOL_BUMP_RELEASE is defined at the
1602 :     time the version tool is loaded (which is the first time you say CMB.make),
1603 :     then the tool will increment the value stored in config/release every
1604 :     time CMB.make is invoked.
1605 :    
1606 :     The binfile format is now insensitive to anything beyond the first
1607 :     two components of a version number, so bumping the release does not render
1608 :     binfiles incompatible. Auto-bumping can be used to keep track of versions
1609 :     during development without invalidating existing binfiles.
1610 :    
1611 :     In any case, every CMB.make updates the date information in version.sml.
1612 :     (This is the date that is printed in the banner.)
1613 :    
1614 :     ----------------------------------------------------------------------
1615 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1616 : mblume 1634 Date: 2004/09/28 10:53:00 CDT
1617 :     Tag: blume-20040928-controls
1618 :     Description:
1619 :    
1620 :     Some cleanup of the controls code.
1621 :    
1622 :     ----------------------------------------------------------------------
1623 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1624 : mblume 1633 Date: 2004/09/27 22:08:00 CDT
1625 :     Tag: blume-20040927-controls
1626 :     Description:
1627 :    
1628 :     Added two pieces of functionality to the Controls interface:
1629 :    
1630 :     1. val save'restore: 'a control -> unit -> unit
1631 :    
1632 :     grabs the current value of the control in stage 1 and restores it
1633 :     in stage 2.
1634 :    
1635 :     2. val set' : 'a control * 'a -> unit -> unit
1636 :    
1637 :     stores the given value into the control in stage 2 (i.e., delayed)
1638 :     but does all error checking in stage 1.
1639 :     (This is for string controls that need to do parse their argument --
1640 :     something that might fail. In some cases, notably in CM, one
1641 :     already knows the intended argument but wants to delay the actual
1642 :     assignment until a time when error recovery would be more difficult.)
1643 :    
1644 :     Changed the handling of controls in tool arguments to classes "sml" and
1645 :     "lazysml":
1646 :     - use Controls.save'restore as a more robust way of restoring the
1647 :     old value (in particular: without having to re-parse the string)
1648 :     - use controls to handle the "overload" keyword in the init group
1649 :     (I believe this change actually fixes a long-standing obscure bug.)
1650 :    
1651 :     ----------------------------------------------------------------------
1652 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1653 : mblume 1632 Date: 2004/09/27 17:00:00 CDT
1654 :     Tag: blume-20040927-lazysml
1655 :     Description:
1656 :    
1657 :     Added a new tool class called "lazysml" to CM's tool chest. The only
1658 :     difference to "sml" is that compilation is done with Control.lazysml
1659 :     set to true. A source of class "lazysml" is automatically recognized
1660 :     by a file name suffix of ".lml".
1661 :    
1662 :     In addition to the above feature, the original class "sml" now also
1663 :     supports a tool argument "lazy" which has the same effect. As a
1664 :     result, the following three lines are equivalent:
1665 :    
1666 :     foo.sml : lazysml
1667 :     foo.sml : sml (lazy)
1668 :     foo.sml (lazy)
1669 :    
1670 :     The setting goes into effect both during parsing and during
1671 :     compilation. The original setting is restored right after parsing and
1672 :     after compilation, respectively.
1673 :    
1674 :     In addition to all the above, there is also a general mechanism to set
1675 :     ANY of the "controls" that are available at the command line via
1676 :     "-C..." on a per-sml-file basis. The same rules that apply for "lazy"
1677 :     apply as well. (In fact, "lazy" is implemented as a special case of
1678 :     the general mechanism.)
1679 :    
1680 :     The .cm file syntax uses a new keyword tool argument called "with".
1681 :     There are several ways of indicating the desired settings:
1682 :    
1683 :     foo.sml (with:parser.quotations=true)
1684 :     foo.sml (with:(name:parser.quotations value:true))
1685 :     foo.sml (with:(name:name1 value:value1 name:name2 value:value2 ...))
1686 :     foo.sml (with:(name1=value1 name2=value2 ...))
1687 :     foo.sml (with:(name1=value1 name:name2 value:value2 name3=value3 ...))
1688 :    
1689 :     etc.
1690 :    
1691 :     Another possible abbreviation is to leave out the =v or value:v part
1692 :     if the name refers to a boolean control (in which case the value is
1693 :     taken to be true). Thus, one could get lazy sml also by saying:
1694 :    
1695 :     foo.sml (with:parser.lazy-keyword=true)
1696 :     foo.sml (with:parser.lazy-keyword)
1697 :     foo.sml (with:(name:parser.lazy-keyword value:true))
1698 :     foo.sml (with:(name:parser.lazy-keyword))
1699 :    
1700 :     ----------------------------------------------------------------------
1701 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1702 : mblume 1631 Date: 2004/09/24 16:22:00 CDT
1703 :     Tag: blume-20040924-ppc-long-branch
1704 :     Description:
1705 :    
1706 :     Turned message about "emiting long form of branch" off by default.
1707 :     Added a control flag to turn it back on when desired.
1708 :    
1709 :     ----------------------------------------------------------------------
1710 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1711 : mblume 1630 Date: 2004/09/24 16:05:00 CDT
1712 :     Tag: blume-20040924-rounding
1713 :     Description:
1714 :    
1715 :     Applied patch for setting rounding modes under Mac OS X. Thanks to
1716 :     Melissa O'Neill for providing the code!
1717 :    
1718 :     ----------------------------------------------------------------------
1719 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1720 : mblume 1629 Date: 2004/09/23 17:30:00 CDT
1721 :     Tag: blume-20040923-envvars
1722 :     Description:
1723 :    
1724 :     1. Changed definition of type ControlRegistry.registry_tree to
1725 :     include control_info (i.e., the name of the controlling environment
1726 :     variable).
1727 :    
1728 :     2. Added command-line flags -e and -E to print the names of environment
1729 :     variables that can be used to control internal settings.
1730 :     (This uses the new API mentioned in 1.)
1731 :    
1732 :     ----------------------------------------------------------------------
1733 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1734 : mblume 1622 Date: 2004/09/13 16:50:00 CDT
1735 :     Tag: Release_110_49
1736 :     Description:
1737 :    
1738 :     New working version (110.49). NEW BOOTFILES!
1739 :    
1740 :     ----------------------------------------------------------------------
1741 : mblume 1621 Name: Matthias Blume
1742 :     Date: 2004/09/13 16:20:00 CDT
1743 :     Tag: blume-20040913-config-mlrisc
1744 :     Description:
1745 :    
1746 :     Put target "mlrisc" back into the default list.
1747 :     (There is no harm in having it, and some users have expressed their
1748 :     wish to have "mlrisc" included by default.)
1749 :    
1750 :     ----------------------------------------------------------------------
1751 : jhr 1620 Name: John Reppy
1752 :     Date: 2004/09/13
1753 :     Tag: jhr-20040913-signals
1754 :     Description:
1755 :    
1756 :     Fixed the signal masking code to properly nest mask/unmask operations
1757 :     on a per-signal basis.
1758 :    
1759 :     ----------------------------------------------------------------------
1760 : mblume 1619 Name: Matthias Blume (blume (at) tti - c (dot) org)
1761 :     Date: 2004/09/08 13:20:00 CDT
1762 :     Tag: blume-20040908-heap-magic
1763 :     Description:
1764 :    
1765 :     Bumped the heap macig number to 0x09082004 to account for the changed
1766 :     layout of the ML frame under MacOS X.
1767 :    
1768 :     ----------------------------------------------------------------------
1769 : allenleung 1618 Name: Allen Leung (leunga (at) reservoir (dot) com)
1770 :     Date: 2004/09/03 11:26:00 EST
1771 :     Tag: leunga-20040903-cygwin-install
1772 :     Description:
1773 :    
1774 :     Added a patch to _arch-n-opsys to enable the Cygwin runtime.
1775 :     The Cygwin runtime is turned on by setting the environment
1776 :     variable SMLNJ_CYGWIN_RUNTIME to 1.
1777 :    
1778 :     ----------------------------------------------------------------------
1779 : mblume 1615 Name: Matthias Blume (blume (at) tti - c (dot) org)
1780 : mblume 1617 Date: 2004/08/31 17:15:00 CDT
1781 :     Tag: blume-20040831-core
1782 :     Description:
1783 :    
1784 :     Added some exports to src/compiler/core.cm upon request by J. Joemann.
1785 :    
1786 :     ----------------------------------------------------------------------
1787 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1788 : mblume 1616 Date: 2004/08/30 17:55:00 CDT
1789 :     Tag: blume-20040830-installer
1790 :     Description:
1791 :    
1792 :     Upon request by Johannes Joemann:
1793 :    
1794 :     - improved ML code of installer to fall back to coping when renaming
1795 :     fails (i.e., when source and target are on different file systems);
1796 :     the code compiles but has yet to be tested in anger
1797 :     - removed mlrisc from list of default targets (config/targets)
1798 :    
1799 :     ----------------------------------------------------------------------
1800 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1801 : mblume 1615 Date: 2004/08/27 17:20:00 CDT
1802 :     Tag: blume-20040827-ptreql
1803 :     Description:
1804 :    
1805 :     Added ptreql primop to structure InlineT (upon request from Larry
1806 :     Paulson).
1807 :    
1808 :     ----------------------------------------------------------------------
1809 : allenleung 1613 Name: Allen Leung (leunga (at) reservoir (dot) com)
1810 : allenleung 1614 Date: 2004/08/15 21:21:00 EST
1811 :     Tag: leunga-110_48-udgraph
1812 :     Description:
1813 :    
1814 :     Another bug fix from Carl Hauser:
1815 :    
1816 :     diff /net/niflab/smlnj48/src/MLRISC/graphs/udgraph.sml udgraph.sml
1817 :     > 48c48
1818 :     > < | rmv((e as (k,_))::es,L) = rmv(es,if k = i then es else
1819 :     > e::L)
1820 :     > ---
1821 :     > > | rmv((e as (k,_))::es,L) = rmv(es,if k = i then L else e::L)
1822 :     > Without this, any deletion of an edge in an undirected graph does severe
1823 :     > violence to the graph.
1824 :    
1825 :     ----------------------------------------------------------------------
1826 :     Name: Allen Leung (leunga (at) reservoir (dot) com)
1827 : allenleung 1613 Date: 2004/08/10 23:35:00 EST
1828 :     Tag: leunga-110_48-ppc
1829 :     Description:
1830 :    
1831 :     The IBM/MacOS syntax switch on PPC was incorrectly swapped.
1832 :    
1833 :     ----------------------------------------------------------------------
1834 : mblume 1602 Name: Matthias Blume (blume (at) tti - c (dot) org)
1835 : mblume 1615 Date: 2004/08/10 12:00:00 CDT
1836 : mblume 1602 Tag: Release_110_48
1837 :     Description:
1838 :    
1839 :     New working version (110.48). NEW BOOTFILES!
1840 :    
1841 :     ----------------------------------------------------------------------
1842 : allenleung 1599 Name: Allen Leung (leunga (at) reservoir (dot) com)
1843 :     Date: 2004/08/09 12:21:00 EST
1844 :     Tag: leunga-110_47-dijsktra
1845 :     Description:
1846 :    
1847 :     Bug fix from Carl Hauser:
1848 :    
1849 : blume 2155 single_source_shortest_paths in dijkstra.sml was observed to get wrong
1850 : allenleung 1599 answers (by comparing to single_source_shortest_paths in bellman-ford.sml).
1851 :    
1852 : blume 2155 The problem is that following the expression A.update(dist,s,Num.zero)
1853 : allenleung 1599 it is necessary to update the priority queue using Q.decreaseWeight(Q,s).
1854 :    
1855 :     ----------------------------------------------------------------------
1856 : mblume 1595 Name: Matthias Blume (blume (at) tti - c (dot) org)
1857 :     Date: 2004/08/06 18:10:00 CDT
1858 :     Tag: blume-20040806-cmdline
1859 :     Description:
1860 :    
1861 :     Fiddled with handling of command-line options:
1862 :    
1863 :     * sml now quits after processing the command line
1864 :     if -H, -S, -h<n>, or -s<n> appears as the last
1865 :     command-line argument
1866 :     * a new option -q terminates the session when encountered on
1867 :     the command line; subsequent arguments will be ignored
1868 :     * bug fixes: short (erroneous) arguments are no longer ignored
1869 :     completely
1870 :    
1871 :     ----------------------------------------------------------------------
1872 : allenleung 1593 Name: Allen Leung (leunga (at) reservoir (dot) com)
1873 :     Date: 2004/08/04 18:17:00 EST
1874 :     Tag: leunga-110_47-ppc-ibm-asm
1875 :     Description:
1876 :    
1877 :     - Added minimal IBM assembly syntax support for PowerPC.
1878 :    
1879 : blume 2155 - Cygwin: manually changed the file cygwin.def. Some exported symbols have
1880 : allenleung 1593 been altered in the runtime. We need an automatic way to keep the file
1881 :     in sync.
1882 :    
1883 :     ----------------------------------------------------------------------
1884 : mblume 1548 Name: Matthias Blume (blume (at) tti - c (dot) org)
1885 : mblume 1581 Date: 2004/08/04 14:00:00 CDT
1886 :     Tag: Release_110_47
1887 :     Description:
1888 :    
1889 :     New working version (110.47). NEW BOOTFILES!
1890 :    
1891 :     ----------------------------------------------------------------------
1892 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1893 : mblume 1580 Date: 2004/08/03 14:25:00 CDT
1894 :     Tag: blume-20040803-callingconv
1895 :     Description:
1896 :    
1897 :     Added low-level support for choosing C calling conventions by
1898 :     twiddling the type of rawccall. (See
1899 :     src/compiler/Semant/types/cproto.sml for details.)
1900 :    
1901 :     ----------------------------------------------------------------------
1902 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1903 : mblume 1579 Date: 2004/08/02 15:55:00 CDT
1904 :     Tag: blume-20040802-backout
1905 :     Description:
1906 :    
1907 :     Backed out of change to win32-filesys.c. The earlier patch to
1908 :     get_file_time caused CM to produce files with the wrong time stamp.
1909 :    
1910 :     ----------------------------------------------------------------------
1911 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1912 : mblume 1573 Date: 2004/08/02 14:45:00 CDT
1913 :     Tag: blume-20040802-nlffi-win32
1914 :     Description:
1915 :    
1916 :     Added NLFFI support for Win32, adapted from a patch provided by David
1917 :     Hansel. This is currently completely untested. Also, the issue
1918 :     concerning stdcall vs. ccall is still unresolved.
1919 :    
1920 :     ----------------------------------------------------------------------
1921 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1922 : mblume 1572 Date: 2004/07/30 17:55:00 CDT
1923 :     Tag: blume-20040730-various
1924 :     Description:
1925 :    
1926 :     Gearing up towards 110.47...
1927 :    
1928 :     - various minor bugfixes to ml-nlffigen
1929 :     - a beginning of a manual for nlffi
1930 :    
1931 :     - eliminated 'export name=value' in config/install.sh as this does
1932 :     not work with certain versions of /bin/sh
1933 :     (Thanks to David King at Motorola for catching this.)
1934 :    
1935 :     - several bugfixes provided or suggested by David Hansel at Reactive Systems:
1936 :     - added a test for tm==NULL to gmtime.c and localtime.c
1937 :     - applied patch for incorrect GetFileTime under win32
1938 :     - toSeconds -> toMilliseconds in Win32/win32-process.sml
1939 :    
1940 :     ----------------------------------------------------------------------
1941 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1942 : mblume 1557 Date: 2004/07/21 18:20:00 CDT
1943 :     Tag: blume-20040721-nlffigen
1944 :     Description:
1945 :    
1946 :     - Fixed minor issue in ml-nlffigen: Now generate structure T_foo
1947 :     for a typedef to an incomplete type, but leave out the "typ" member.
1948 :     (This is just for consistency.)
1949 :     - Started to produce what is supposed to become better (i.e., comprehensive)
1950 :     documentation of what ml-nlffigen does and produces.
1951 :    
1952 :     ----------------------------------------------------------------------
1953 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1954 : mblume 1548 Date: 2004/07/14 16:25:00 CDT
1955 :     Tag: blume-20040714-union
1956 :     Description:
1957 :    
1958 :     Added C_UNION to c-calls/c-types.sml and updated the machinery
1959 :     (ml-nlffigen, cproto.sml) that conveys C function interface
1960 :     information to the code generator.
1961 :    
1962 :     However, the actual architecture-specific implementation of function
1963 :     arguments and results that are C unions is still not implemented.
1964 :    
1965 :     ----------------------------------------------------------------------
1966 : allenleung 1539 Name: Allen Leung (leunga (at) reservoir (dot) com)
1967 : allenleung 1546 Date: 2004/07/14 14:38:00 EST
1968 :     Tag: leunga-110_46_1-ppc-lwzu
1969 :     Description:
1970 :    
1971 :     Added these instructions to the PowerPC architecture:
1972 :     LBZU(X), LHZU(X), LWZU(X),
1973 : blume 2155 STWU(X), STFDU, STFSU
1974 :    
1975 : allenleung 1546 etc...
1976 :    
1977 :     Note: I haven't added their instruction encoding into the description.
1978 :    
1979 :     ----------------------------------------------------------------------
1980 :     Name: Allen Leung (leunga (at) reservoir (dot) com)
1981 : allenleung 1539 Date: 2004/07/13 15:04:00 EST
1982 :     Tag: leunga-110_46_1-ppc-lwarx
1983 :     Description:
1984 :    
1985 :     Added the two instructions LWARX and STWCX to the PowerPC
1986 : blume 2155 instruction set.
1987 : allenleung 1539
1988 :     A (untested) rewrite of loop-structure.sml. The old version
1989 :     is completely broken.
1990 :    
1991 :     ----------------------------------------------------------------------
1992 : mblume 1479 Name: Matthias Blume (blume (at) tti - c (dot) org)
1993 : mblume 1537 Date: 2004/07/13 13:50:00 CDT
1994 :     Tag: blume-20040713-nlffi
1995 :     Description:
1996 :    
1997 :     - use paramAlloc to report c-calls with too many arguments
1998 :     (for PPC version where parameter area is pre-allocated)
1999 :     - added ccall_maxargspace to machspec (to implement the above)
2000 :     - made "make" commend in CM's "make" tool configurable
2001 :     - added option (default: on) for passing the name of the SML/NJ's "bin"
2002 :     directory to "make"; the call looks like this:
2003 :    
2004 :     make <options> SMLNJ_BINDIR=<dir> <target>
2005 :    
2006 :     This can be used by the Makefile to, e.g., pick the "right" version
2007 :     of ml-nlffigen.
2008 :     - minor code tweaks
2009 :    
2010 :     ----------------------------------------------------------------------
2011 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2012 : mblume 1532 Date: 2004/07/12 22:50:00 CDT
2013 :     Tag: blume-110_46_1-macosx-nlffi
2014 :     Description:
2015 :    
2016 :     NLFFI under Mac OS X now working (sort of). This is largely untested,
2017 :     though.
2018 :    
2019 :     Note: 1. You have to make a new, clean build of the runtime system.
2020 :     2. There are new BOOTFILES, you have to use them!
2021 :     (Doing the bootstrap process yourself would be *very* painful!
2022 :     If you absolutely have to do it, build the system under
2023 :     a different architecture and then cross-compile.)
2024 :    
2025 :     Version bumped to 110.46.1 to account for runtime data format changes.
2026 :    
2027 :     ----------------------------------------------------------------------
2028 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2029 : mblume 1515 Date: 2004/06/18 14:30:00 CDT
2030 :     Tag: blume-20040618-unix
2031 :     Description:
2032 :    
2033 :     Changed the implementation of structure Unix so that the same stream
2034 :     is returned every time one of the {text,bin}{In,Out}streamOf functions
2035 :     is invoked on the same proc. This is not what the spec currently
2036 :     says -- although IMO it arguably should. (See discussion below.)
2037 :    
2038 :     ----------------------------------------------------------------------
2039 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2040 : mblume 1507 Date: 2004/06/17 18:15:00 CDT
2041 :     Tag: Release_110_46
2042 :     Description:
2043 :    
2044 :     New working version (110.46). NEW BOOTFILES!
2045 :    
2046 :     ----------------------------------------------------------------------
2047 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2048 : mblume 1506 Date: 2004/06/17 17:20:00 CDT
2049 :     Tag: blume-20040617-timer-unix
2050 :     Description:
2051 :    
2052 :     Changed the interface of structures Timer and Unix to match the most
2053 :     recent Basis spec.
2054 :    
2055 :     In the case of Unix there still seems to be an open/weird issue:
2056 :    
2057 :     The {text,bin}{In,Out}streamOf functions are supposed to create
2058 :     fresh streams whenever they are called -- as opposed to have them
2059 :     return the same stream every time. This design is supposed to
2060 :     prevent space leaks caused by proc values hanging on to streams.
2061 :    
2062 :     The reap function, on the other hand, is supposed to close the
2063 :     streams. This cannot be done without having a handle on the
2064 :     stream in proc after all...
2065 :    
2066 :     I took the liberty to implement the following stopgap solution:
2067 :    
2068 :     The proc value hangs on to the most recently created stream(s).
2069 :     Reap closes those. If either or both of the two streams hadn't
2070 :     been created at all yet, then reap will close the corresponding
2071 :     file descriptors directly.
2072 :    
2073 :     PS: I don't understand the original space leak argument anymore. If
2074 :     a proc hangs on to the imperative stream, then I/O operations on those
2075 :     will advance the state of the cached stream and avoid the space leak.
2076 :    
2077 :     ----------------------------------------------------------------------
2078 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2079 : mblume 1500 Date: 2004/05/28 16:45:00 CDT
2080 :     Tag: blume-20040528-basis
2081 :     Description:
2082 :    
2083 :     Added signature PACK_REAL and exported functor PrimIO.
2084 :    
2085 :     ----------------------------------------------------------------------
2086 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2087 : mblume 1498 Date: 2004/05/25 16:00:00 CDT
2088 :     Tag: blume-20040525-group-owner
2089 :     Description:
2090 :    
2091 :     CM now ignores (but still accepts) the "owner" information in group
2092 :     descriptions. The owner of a group is its next enclosing
2093 :     library. Each group must have a unique owner. (There is a virtual
2094 : mblume 1505 "toplevel" library that owns groups which are not nested within a real
2095 : mblume 1498 library.) Previously, each group had to explicitly declare its owner,
2096 :     and CM would check that such a declaration is correct. The new scheme
2097 :     is to have CM check that for each group there is precisely one owning
2098 :     library.
2099 :    
2100 :     The advantage of the new scheme is that the programmer no longer needs
2101 :     to maintain the somewhat annoying owner information. The downside is
2102 :     that CM cannot enforce the ownership rule across multiple runs of
2103 :     CM.make. Fortunately, enclosing the same group in two different
2104 :     libraries A and B which are not part of the same program does not
2105 :     cause real problems.
2106 :    
2107 :     ----------------------------------------------------------------------
2108 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2109 : mblume 1496 Date: 2004/05/20 16:00:00 CDT
2110 :     Tag: blume-20040520-win32
2111 :     Description:
2112 :    
2113 :     Made the win32 version work again. (Strangely, a misplaced comma had
2114 :     slipped into win32-process.c which prevented the runtime from being
2115 :     compiled correctly.)
2116 :    
2117 :     Also, included a minor addition to ml-build.bat analogous to what was
2118 :     done in blume-20040519-ml-build.
2119 :    
2120 :     ----------------------------------------------------------------------
2121 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2122 : mblume 1494 Date: 2004/05/19 22:10:00 CDT
2123 :     Tag: blume-20040519-ml-build
2124 :     Description:
2125 :    
2126 :     Arranged for ml-build to clean up after itself a little bit better.
2127 :     The script generates a temporary SML source file and compiles it using
2128 :     CM, so CM generates metadata (GUID, SKEL, objectfile) for it. It now
2129 :     gets rid of those at the end, so they don't accumulate under .cm.
2130 :    
2131 :     This required a minor change to install.sh because the name of the
2132 :     metadata directory (default: .cm) is actually configurable at
2133 :     installation time.
2134 :    
2135 :     ----------------------------------------------------------------------
2136 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2137 : mblume 1491 Date: 2004/05/18 15:50:00 CDT
2138 :     Tag: blume-20040518-mkreader
2139 :     Description:
2140 :    
2141 :     Added Posix.IO.mk{Bin,Text}{Reader,Writer} by lifting their respective
2142 :     implementations from internal modules PosixBinPrimIO and PosixTextPrimIO.
2143 :    
2144 :     ----------------------------------------------------------------------
2145 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2146 : mblume 1489 Date: 2004/05/11 14:35:00 CDT
2147 :     Tag: blume-20040511-win32sock
2148 :     Description:
2149 :    
2150 :     Added previously missing support for many socket-related functions
2151 :     under win32. Thanks to David Hansel <hansel@reactive-systems.com>
2152 :     for the voluminous patch!
2153 :    
2154 :     (I have not tested this patch under win32 yet.)
2155 :    
2156 : mblume 1490 Here is David's e-mail:
2157 :    
2158 :     Hi,
2159 :    
2160 :     Attached to this email you find a diff against sml/nj 110.45
2161 :     that will enable socket support under Windows.
2162 :    
2163 :     To apply the patch (using unix or cygwin)
2164 :     1) gunzip runtime.diff.gz
2165 :     2) "cd" into "src/runtime" in the source tree of a fresh
2166 :     110.45 installation.
2167 :     3) patch -p 1 < [your/path/to]runtime.diff
2168 :    
2169 : blume 2155 The code compiles fine but has NOT yet been extensively tested.
2170 :     I only ran a few tests for basic socket client functionality
2171 :     (which worked fine). Especially the functions that use ioctl
2172 : mblume 1490 are not tested at all and might not work (see below).
2173 :    
2174 :     I implemented this since we want to move to a newer version of sml/nj
2175 :     but need socket support in order to use it. This is the first time I
2176 :     even had a look at the sml/nj source, so please review my changes
2177 :     before making this part of the distribution! Here are a few issues
2178 : blume 2155 that I think might be better for someone to solve who is more
2179 : mblume 1490 familiar with the sml/nj source (and socket programming):
2180 :    
2181 :     - getnetbyaddr.c and getnetbyname.c will raise a "not implemented"
2182 :     exception since I could not figure out what the windows equivalent
2183 :     of these functions is
2184 :    
2185 :     - In sockets-osdep.h there are a some #include statements that are
2186 :     only used in a few files that include sockets-osdep.h
2187 :    
2188 :     - In smlnj-sock-lib.c, function init_fn() calls WSAStartup() but
2189 :     does not process its return value since I don't know how to
2190 :     report an error upwards.
2191 :    
2192 :     - It would probably be good to have a call to WSACleanup() when
2193 : blume 2155 the library is unloaded (if there is such a possibility).
2194 : mblume 1490 Otherwise I think Windows will take care of this automatically
2195 :     when the process finishes.
2196 :    
2197 :     - I used ioctlsocket() as a replacement for ioctl() but I have
2198 :     no idea if that is actually the proper replacement on Windows.
2199 :    
2200 :     - All these issues are marked in the code by "FIXME" comments.
2201 :    
2202 : blume 2155 We use sml/nj extensively in our products and are quite happy
2203 :     with it. I hope this contribution will help you.
2204 : mblume 1490
2205 :     Keep up the good work!
2206 :    
2207 :     David
2208 :    
2209 : mblume 1489 ----------------------------------------------------------------------
2210 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2211 : mblume 1488 Date: 2004/05/11 14:20:00 CDT
2212 :     Tag: blume-20040511-installml
2213 :     Description:
2214 :    
2215 :     Fixed two bugs in installml script. (Thanks to Vesa A. Norrman for
2216 :     the patch.)
2217 :    
2218 :     ----------------------------------------------------------------------
2219 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2220 : mblume 1487 Date: 2004/05/11 14:05:00 CDT
2221 :     Tag: blume-20040511-nlffi-netbsd
2222 :     Description:
2223 :    
2224 :     Added support for nlffi under netbsd. (Thanks to Vesa A. Norrman for
2225 :     the patch.)
2226 :    
2227 :     ----------------------------------------------------------------------
2228 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2229 : mblume 1486 Date: 2004/05/11 12:05:00 CDT
2230 :     Tag: blume-20040511-exports
2231 :     Description:
2232 :    
2233 :     As per request by Adam Chlipala <adam@hcoop.net>, extended various
2234 :     export lists in compiler-related .cm-files.
2235 :    
2236 :     ----------------------------------------------------------------------
2237 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2238 : mblume 1485 Date: 2004/05/11 11:35:00 CDT
2239 :     Tag: blume-20040511-allsource
2240 :     Description:
2241 :    
2242 :     The installer now honors the "src-smlnj" target again, although its meaning
2243 :     has changed from "all sources required for the compiler" to "all sources
2244 :     the installer knows about". In other words, if you enable "src-smlnj"
2245 :     in the "targets" file, then the installer will pull in sources for
2246 :     everything. (Notice that this refers to source code only. Compiled
2247 :     code is still only installed for modules that were requested explicitly
2248 :     or which are required for other modules that were requested explicitly.)
2249 :    
2250 :     ----------------------------------------------------------------------
2251 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2252 : mblume 1482 Date: 2004/04/23 17:40:00 CDT
2253 :     Tag: blume-20040423-ieee-scan
2254 :     Description:
2255 :    
2256 :     Fixed IEEEReal.scan (and .fromString) so that if there is an overflow
2257 :     in the exponent calculation we get INF or ZERO (depending on the mantissa
2258 :     and the sign of the exponent).
2259 :    
2260 :     ----------------------------------------------------------------------
2261 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2262 : mblume 1480 Date: 2004/04/23 10:40:00 CDT
2263 :     Tag: blume-20040423-ml-build
2264 :     Description:
2265 :    
2266 :     The ml-build script now terminates with a non-0 status when something
2267 :     goes wrong.
2268 :    
2269 :     ----------------------------------------------------------------------
2270 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2271 : mblume 1479 Date: 2004/04/22 16:35:00 CDT
2272 :     Tag: blume-20040422-Option
2273 :     Description:
2274 :    
2275 :     Made exception Option to be the same as exception Option.Option
2276 :     (as it should be).
2277 :    
2278 :     ----------------------------------------------------------------------
2279 : allenleung 1468 Name: Allen Leung (leunga (at) reservoir (dot) com)
2280 :     Date: 2004/03/19 14:40:00 EST
2281 :     Tag: leunga-20040319-cygwin-nlffi
2282 :     Description:
2283 :    
2284 :     Fixed the runtime so that ml-nlffi-lib runs on the cygwin version
2285 :     of SML/NJ. The problem is that
2286 :    
2287 :     lib = dlopen(NULL, ...)
2288 :     f = dlsym(lib, "malloc");
2289 :    
2290 : blume 2155 does not work on Windows unless we explicitly export symbols
2291 :     such as 'malloc' during linking. We fixed this by explicitly
2292 : allenleung 1468 exporting the required symbols with the magic gcc incantation:
2293 :    
2294 :     -Wl,--export-all cygwin.def
2295 :    
2296 :     where cygwin.def is a file containing all the symbols that
2297 : blume 2155 we wish to export.
2298 : allenleung 1468
2299 : blume 2155 I suspect this is a Windows problem and we'll have to
2300 :     do the same (somehow with windows compilers) when
2301 : allenleung 1468 we build the native win32 version with the system
2302 : blume 2155 calls LoadLibrary/GetProcAddress.
2303 : allenleung 1468
2304 :     ----------------------------------------------------------------------
2305 : mblume 1384 Name: Matthias Blume (blume (at) tti - c (dot) org)
2306 : mblume 1465 Date: 2004/03/04 16:35:00 CST
2307 :     Tag: blume-20040304-intinf-fmt
2308 :     Description:
2309 :    
2310 :     Fixed problem with IntInf.fmt (sign would show up on the right instead
2311 :     of on the left for BIN, OCT, and HEX).
2312 :    
2313 :     ----------------------------------------------------------------------
2314 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2315 : mblume 1464 Date: 2004/03/04 11:25:00 CST
2316 :     Tag: blume-20040304-symlinks
2317 :     Description:
2318 :    
2319 :     Fixed problem with installer script (unix only) where bin/ml-yacc and
2320 :     friends pointed (via symlinks) to absolute locations instead of just
2321 :     .run-sml. This was reported by Vesa A Norrman.
2322 :    
2323 :     ----------------------------------------------------------------------
2324 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2325 : mblume 1455 Date: 2004/02/13 14:50:00 CST
2326 :     Tag: Release_110_45
2327 :     Description:
2328 :    
2329 :     New working version (110.45). New bootfiles.
2330 :    
2331 :     ----------------------------------------------------------------------
2332 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2333 : mblume 1448 Date: 2004/01/26 15:15:15 CST
2334 :     Tag: blume-20040126-toplevel
2335 :     Description:
2336 :    
2337 :     Improved handling of exceptions at the interactive toplevel.
2338 :    
2339 :     ----------------------------------------------------------------------
2340 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2341 :     Date: 2004/01/26 11:25:00
2342 : mblume 1446 Tag: blume-20040126-app
2343 :     Description:
2344 :    
2345 :     Type of top-level "app" corrected.
2346 : mblume 1447 Added code for setting vp_limitPtrMask to Win32-specific runtime.
2347 : mblume 1446
2348 :     ----------------------------------------------------------------------
2349 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2350 : mblume 1427 Date: 2003/11/18 17:10 CST
2351 :     Tag: blume-20031118-basis-fiddle
2352 :     Description:
2353 :    
2354 :     - changed Timer interface to what might become the spec
2355 :     - POSIX_FLAGS -> BIT_FLAGS according to spec
2356 :     - some other minor discrepancies wrt. spec eliminated
2357 :    
2358 :     ----------------------------------------------------------------------
2359 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2360 : mblume 1414 Date: 2003/11/06 12:00:00 CST
2361 :     Tag: Release_110_44
2362 :     Description:
2363 :    
2364 :     New working version (110.44). New bootfiles.
2365 :    
2366 :     ----------------------------------------------------------------------
2367 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2368 : mblume 1411 Date: 2003/11/04 11:50:00 CST
2369 :     Tag: blume-20031104-move-libraries
2370 :     Description:
2371 :    
2372 :     Eliminated the "dont_move_libraries" directive in config/targets.
2373 :     (The mechanism was broken and could not be fixed easily. Moreover,
2374 :     there does not seem to be any reason not to move all libraries into
2375 :     lib during installation. I originally implemented this directive as a
2376 :     backward-compatibility feature when I first introduced the new CM.
2377 :     Now that things have been stable for a long time and going back to the
2378 :     old CM is not an option, there is no reason to keep it around.)
2379 :    
2380 :     ----------------------------------------------------------------------
2381 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2382 : mblume 1409 Date: 2003/11/03 16:00:00 CST
2383 :     Tag: blume-20031103-installdir
2384 :     Description:
2385 :    
2386 :     Made installer honor INSTALLDIR variable again. (Thanks to Chris
2387 :     Richards for pointing out the problem and providing the solution.)
2388 :    
2389 :     ----------------------------------------------------------------------
2390 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2391 : mblume 1400 Date: 2003/10/01 17:05:00 CDT
2392 :     Tag: blume-20031001-lal-mlrisc
2393 :     Description:
2394 :    
2395 :     MLRISC bug fix from Lal.
2396 :    
2397 :     ----------------------------------------------------------------------
2398 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2399 : mblume 1399 Date: 2003/09/30 16:10:00 CDT
2400 :     Tag: blume-20030930-primio-bat
2401 : mblume 1398 Description:
2402 :    
2403 :     1. Added openVector, nullRd, and nullWr to PRIM_IO.
2404 :     2. Improved .bat files (for Win32 port) to make things work under Win95.
2405 :     (thanks to Aaron S. Hawley for this one)
2406 :    
2407 :     ----------------------------------------------------------------------
2408 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2409 : mblume 1395 Date: 2003/09/26 16:05:00 CDT
2410 :     Tag: blume-20030926-wrappriv
2411 :     Description:
2412 :    
2413 :     Added missing wrapper for privilege "primitive" in $smlnj/viscomp/core.cm.
2414 :    
2415 :     ----------------------------------------------------------------------
2416 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2417 : mblume 1394 Date: 2003/09/26 15:00:00 CDT
2418 :     Tag: blume-20030926-110_43_3
2419 :     Description:
2420 :    
2421 :     - additional cleanup
2422 :     - version number bump, NEW BOOTFILES
2423 :    
2424 :     ----------------------------------------------------------------------
2425 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2426 : mblume 1393 Date: 2003/09/26 12:00:00 CDT
2427 :     Tag: blume-20030926-ppautoload
2428 :     Description:
2429 :    
2430 :     I modified the read-eval-print loop so that the autoloader gets
2431 :     invoked whenever the prettyprinter tries to look up a symbol that
2432 :     is not currently defined in the toplevel environment but which
2433 :     appears in CM's autoload registry. As a result, we see far fewer of
2434 :     those ?.Foo.Bar.xxx names in the prettyprinter's output.
2435 :    
2436 :     In addition to this I tried to clean up some pieces of the Basis
2437 :     implementation (e.g., Socket, Word8Array) in order to prevent other
2438 :     instances of these ?.Foo.Bar.xxx names from being printed.
2439 :    
2440 :     The mechanism that picks names for types still needs some work, though.
2441 :     (Right now it seems that if there is a type A.t which is defined to
2442 : mblume 1394 be B.u, but B is unavailable at toplevel, then A.t gets printed as
2443 :     "?.B.u" although the perhaps more sensible solution would be to use
2444 : mblume 1393 "A.t" in this case. In other words, the prettyprinter should follow
2445 :     a chain of DEFtycs not farther than there are corresponding toplevel
2446 :     names in the current environment.)
2447 :    
2448 :     ----------------------------------------------------------------------
2449 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2450 : mblume 1391 Date: 2003/09/24 16:31:00 CDT
2451 :     Tag: blume-20030924-installer
2452 :     Description:
2453 :    
2454 :     Another installer tweak: All the ML code for the installer is now
2455 :     compiled during CMB.make and put into a little library called
2456 :     $smlnj/installer.cm. The installation then simply invokes
2457 :    
2458 :     sml -m $smlnj/installer.cm
2459 :    
2460 :     and everything happens automagically.
2461 :    
2462 :     Win32: ML code senses value of environment variable SMLNJ_HOME.
2463 :     Unix: ML code senses values of environment variables ROOT, CONFIGDIR,
2464 :     and BINDIR.
2465 :    
2466 :     The new scheme guarantees that the ML code responsible for the installation
2467 :     is in sync with the APIs of the main system. Also, the installer is
2468 :     somewhat faster because the installer script is precompiled.
2469 :    
2470 :     ----------------------------------------------------------------------
2471 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2472 : mblume 1390 Date: 2003/09/24 15:35:00 CDT
2473 :     Tag: blume-20030924-synsock
2474 :     Description:
2475 :    
2476 :     Added a signature SYNCHRONOUS_SOCKET to basis.cm. This is like SOCKET
2477 :     but excludes all non-blocking operations. Defined SOCKET (in Basis)
2478 :     and CML_SOCKET in terms of SYNCHRONOUS_SOCKET. Removed superfluous
2479 :     implementations of non-blocking operations from CML's Socket
2480 :     structure.
2481 :    
2482 :     ----------------------------------------------------------------------
2483 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2484 : mblume 1389 Date: 2003/09/24 15:10:05 CDT
2485 :     Tag: blume-20030924-sockets
2486 :     Description:
2487 :    
2488 :     1. Fixed SOCKET API and implementation to match Basis spec.
2489 :     This required changing the internal representation of sockets to one
2490 :     that remembers (for each socket file descriptor) whether it is currently
2491 :     blocking or non-blocking. This state is maintained lazily (i.e., a system
2492 :     call is made only if the state actually needs to change).
2493 :    
2494 :     2. OS-specific details of sockets were moved into separate files, thus
2495 :     making it possible to unify the bulk of the socket implementations
2496 :     between Unix and Win32.
2497 :    
2498 :     3. CML's socket API changed accordingly.
2499 :     (Note that we need to remove non-blocking functions from this API
2500 :     since they are redundant in the case of CML!)
2501 :    
2502 :     4. CML's socket implementation now makes use of non-blocking functions
2503 :     provided by Basis, thus removing all OS-dependent code from this part
2504 :     of CML.
2505 :    
2506 :     5. Changed Real64.precision from 52 to 53. Minor cleanup in Real64 code.
2507 :    
2508 :     ----------------------------------------------------------------------
2509 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2510 : mblume 1387 Date: 2003/09/22 12:10:00 CDT
2511 : mblume 1388 Tag: blume-20030922-110_43_2
2512 : mblume 1387 Description:
2513 :    
2514 :     Made a new interim version and bootfiles for developer's bootstrapping
2515 :     convenience.
2516 :    
2517 :     110.43.2 -- NEW BOOTFILES
2518 :    
2519 :     ----------------------------------------------------------------------
2520 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2521 : mblume 1385 Date: 2003/09/19 15:55:00 CDT
2522 :     Tag: blume-20030919-cmdir
2523 :     Description:
2524 :    
2525 :     1. new-install.sh -> install.sh
2526 :     2. changed default CM "metadata" directory name to ".cm" (instead of "CM")
2527 :     3. tweaked installer so that another name instead of .cm can be chosen
2528 :     at install time (by setting the CM_DIR_ARC environment variable
2529 :     during installation); once installation is complete, the name is
2530 :     fixed
2531 :    
2532 :     ----------------------------------------------------------------------
2533 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2534 : mblume 1384 Date: 2003/09/18 16:00:00 CDT
2535 :     Tag: blume-20030918-110_43_1
2536 :     Description:
2537 :    
2538 :     Made a new interim version and bootfiles for developer's bootstrapping
2539 :     convenience.
2540 :    
2541 :     110.43.1 -- NEW BOOTFILES
2542 :    
2543 :     ----------------------------------------------------------------------
2544 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2545 : mblume 1383 Date: 2003/09/18 15:20:00 CDT
2546 :     Tag: blume-20030918-misc
2547 :     Description:
2548 :    
2549 :     1. Exported fractionsPerSecond etc. from TimeImp (but not from Time as
2550 :     this seems to be controversial at the moment) and used those in
2551 :     Posix.ProcEnv.times.
2552 :    
2553 :     2. Added Time.{from,to}Nanoseconds to Time.
2554 :    
2555 :     3. Improved Real.{from,to}LargeInt by avoiding needless calculations.
2556 :     For example, fromLargeInt never needs to look at more than 3 "big
2557 :     digits" to get its 53 bits of precision.
2558 :    
2559 :     ----------------------------------------------------------------------
2560 : mblume 1384 Name: Matthias Blume (blume (at) tti - c (dot) org)
2561 : mblume 1381 Date: 2003/09/17 16:30:00 CDT
2562 :     Tag: blume-20030917-real32-slices
2563 :     Description:
2564 :    
2565 :     Added an entry to the primitive environment
2566 :     (compiler/Semant/statenv/prim.sml) for int32->real64 conversion and
2567 :     added code to compiler/CodeGen/main/mlriscGen.sml to implement it.
2568 :    
2569 :     Removed some of the "magic" constants in real64.sml and replaced them
2570 :     with code that generates these values from their corresponding
2571 :     integer counterparts.
2572 :    
2573 :     Made all(?) the slice-related changes to the Basis and made everything
2574 :     compile again...
2575 :    
2576 :     ----------------------------------------------------------------------
2577 : mblume 1368 Name: Matthias Blume (blume (at) tti - c (dot) org)
2578 : mblume 1375 Date: 2003/09/15 17:45:00 CDT
2579 :     Tag: blume-20030915-rbase
2580 :     Description:
2581 :    
2582 :     Fixed bug in Real.fromLargeInt.
2583 :    
2584 :     ----------------------------------------------------------------------
2585 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2586 : mblume 1369 Date: 2003/09/13 18:11:00 CDT
2587 :     Tag: blume-20030913-libinstall
2588 :     Description:
2589 :    
2590 :     Minor bugfix in config/libinstall (set anchor with path to
2591 :     standalone tool after installing it, otherwise libraries that
2592 :     need ml-lex or ml-yacc won't compile the first time the installer
2593 :     runs).
2594 :    
2595 :     ----------------------------------------------------------------------
2596 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2597 : mblume 1368 Date: 2003/09/12 11:45:00 CDT
2598 :     Tag: blume-20030912-various
2599 :     Description:
2600 :    
2601 :     - fixed bug in Real.toLargeInt
2602 :     - fixed bug in Posix.ProcEnv.times
2603 :     - changed inputLine functions to return an option
2604 :     - minor installer improvements / bugfixes
2605 :     - changed default @SMLalloc parameter for x86/celeron to 64k
2606 :    
2607 :     ----------------------------------------------------------------------
2608 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2609 : mblume 1356 Date: 2003/09/09 22:00:00 CDT
2610 :     Tag: Release_110_43
2611 :     Description:
2612 :    
2613 :     New working release 110.43. New bootfiles.
2614 :    
2615 :     ----------------------------------------------------------------------
2616 : mblume 1368 Name: Matthias Blume (blume (at) tti - c (dot) org)
2617 : mblume 1354 Date: 2003/09/09 19:20:00 CDT
2618 :     Tag: blume-20030909-installer
2619 :     Description:
2620 :    
2621 :     Rewrote large parts of config/install.sh in SML (config/libinstall.sml).
2622 :     Modified config/install.bat to take advantage of it. Also modified
2623 :     config/install.sh (and called it config/new-install.sh) to take advantage
2624 :     of it on Unix systems. (The SML code is (supposed to be) platform-
2625 :     independent.)
2626 :    
2627 :     The installer can now install everything under Win32
2628 :     as well as under *nix as long as it compiles.
2629 :    
2630 :     Other changes:
2631 :    
2632 :     - made CML compile again under Win32
2633 :     - made eXene compile under Win32 (by providing a fake structure UnixSock
2634 :     and by using OS.Process.getEnv instead of Posix.ProcEnv.getenv)
2635 :     - fixed a bug in nowhere: it assumed that type OS.Process.status is the
2636 :     same as type int; under Win32 it isn't
2637 :     - fixed some slice-related problems in the win32-specific parts of CML
2638 :     - added a functor argument "sameVol" to os-path-fn.sml in the Basis
2639 :     (under Win32, the volume name is case-insensitive, and the
2640 :     OS.Path code compares volume names for equality)
2641 :    
2642 :     ----------------------------------------------------------------------
2643 : mblume 1368 Name: Matthias Blume (blume (at) tti - c (dot) org)
2644 : mblume 1353 Date: 2003/09/08 11:55:00 CDT
2645 :     Tag: blume-20030908-fullpath
2646 :     Description:
2647 :    
2648 :     Made Win32 version of OS.FileSys.fullPath return current directory
2649 :     when given an empty string. This is what the spec says, and incidentally,
2650 :     CM depends on it. (CM otherwise goes into an infinite loop in certain
2651 :     cases when presented with the name of a non-existing .cm file.)
2652 :    
2653 :     ----------------------------------------------------------------------
2654 : mblume 1368 Name: Matthias Blume (blume (at) tti - c (dot) org)
2655 : mblume 1350 Date: 2003/09/04 16:30:00 CDT
2656 : mblume 1352 Tag: blume-20030905-slices-etc
2657 : mblume 1350 Description:
2658 :    
2659 :     1. Changed interface to vectors and arrays in Basis to match
2660 :     (draft) Basis spec.
2661 :     2. Added signatures and implementations of slices according to
2662 :     Basis spec.
2663 :     3. Edited source code throughout the system to make it compile again
2664 :     under 1. and 2. (In some cases code had to be added to have it
2665 :     match the new signatures.)
2666 :     4. MLRISC should be backward-compatible: the copies of the originals
2667 :     of files that needed to change under 3. were retained, the .cm files
2668 :     check the compiler version number and use old versions when
2669 :     appropriate.
2670 :     5. Changed type of OS.FileSys.readDir and Posix.FileSys.readdir to
2671 :     dirstream -> string option (in accordance with Basis spec).
2672 :     6. When generating code that counts lines, ml-lex used function
2673 :     CharVector.foldli, taking advantage of its old interface.
2674 :     This has been replaced with the corresponding code from
2675 :     CharVectorSlice. (html-lex must be re-lexed!)
2676 :     7. BitArray in smlnj-lib/Util has been extended/modified to match the
2677 :     new MONO_ARRAY signature. (Do we need BitArraySlice?)
2678 :     8. Removed temporary additions (fromInternal, toInternal) from the
2679 :     (now obsolete) IntInf in smlnj-lib/Util.
2680 :     9. Cleaned up structure Byte.
2681 :     10. Added localOffset, scan, and fromString to Date (according to spec).
2682 :     Cleaned/corrected implementation of Date.
2683 :     (Still need to check for correctness; implement better canonicalizeDate.)
2684 :     11. Added "scan" to signature IEEE_REAL.
2685 :     12. Some improvements to IntInf [in particular: efficiency-hack for
2686 :     mod and rem when second operand is 2 (for parity checks).]
2687 :     13. Changed representation of type Time.time, using a single IntInf.int
2688 :     value counting microseconds. This considerably simplified the
2689 :     implementation of structure Time. We now support negative time
2690 :     values; scan and fromString handle signs.
2691 :     14. Functor PrimIO now takes two additional arguments (VectorSlice and
2692 :     ArraySlice).
2693 :    
2694 :     ----------------------------------------------------------------------
2695 : mblume 1368 Name: Matthias Blume (blume (at) tti - c (dot) org)
2696 : mblume 1347 Date: 2003/08/28 17:00:00 CDT
2697 :     Tag: blume-20030828-intinf
2698 :     Description:
2699 :    
2700 :     This is a major update which comes with a version number bump
2701 :     (110.42.99 -- yes, we are really close to 110.43 :-), NEW BOOTFILES,
2702 :     and an implementation of IntInf in the Basis.
2703 :    
2704 :     There are a fairly large number of related changes and updates throughout
2705 :     the system:
2706 :    
2707 :     Basis:
2708 :     - Implemented IntInf.
2709 :     - Made LargeInt a projection of IntInf (by filtering through INTEGER).
2710 : mblume 1348 - Added some missing Real64 operations, most notably Real.toLargeInt.
2711 : mblume 1347 - Added FixedInt as a synonym for Int32.
2712 :    
2713 :     compiler:
2714 :     * Added support for a built-in intinf type.
2715 :     - literals
2716 :     - pattern matching
2717 :     - conversion shortcuts (Int32.fromLarge o Int.toLarge etc.)
2718 :     - overloading on literals and operations
2719 :    
2720 :     This required adding a primitive type intinf, some additional
2721 :     primops, and implementations for several non-trivial intinf
2722 :     operations in Core. (The intinf type is completely abstract
2723 :     to the compiler; all operations get delegated back to the Core.)
2724 :    
2725 :     * Intinf equality is handled by polyequal. However, the compiler
2726 :     does not print its usual warning in this case (since polyequal
2727 :     is the right thing to do there).
2728 :    
2729 :     * Improved the organization of structure InlineT.
2730 :    
2731 :     * A word about conversion primops:
2732 :     If conversions involving intinf do not cancel out during
2733 : mblume 1348 CPS contract, then the compiler must insert calls to Core functions.
2734 : mblume 1347 Since all core access must be resolved already during the FLINT
2735 :     translate phase, it would be too late a the time of CPS contract
2736 : mblume 1348 to add new Core calls. For this reason, conversion primops
2737 : mblume 1347 for intinf carry two arguments: 1. the numeric argument that
2738 :     they are supposed to convert, and 2. the Core function that
2739 :     can help with this conversion if necessary. If CPS contract
2740 : mblume 1348 eliminates a primop, then the associated Core function becomes
2741 :     dead and goes away. Intinf conversion primops that do not get
2742 :     eliminated by CPS contract get rewritten into calls of their
2743 :     core functions by a separate, new phase.
2744 : mblume 1347
2745 :     interactive system:
2746 :     - Control.Print.intinfDepth controls max length of intinf constants
2747 :     being printed. (Analogous to Control.Print.stringDepth.)
2748 :     - Cleanup in printutil and pputil: got rid of unused stuff and
2749 :     duplicates; replaced some of the code with code that makes better
2750 :     use of library functionality.
2751 :    
2752 :     CM:
2753 :     Bugfix: parse-errors in init group (system/smlnj/init/init.cmi)
2754 :     are no longer silent.
2755 :    
2756 :     CKIT:
2757 :     Fixed mismatched uses of Int32 and LargeInt. I always decided
2758 :     in favor of LargeInt -- which is now the same as IntInf.
2759 :     CKIT-knowledgable people should check whether this is what's
2760 :     intended and otherwise change things back to using Int32 or
2761 :     FixedInt.
2762 :    
2763 :     Throughout the code:
2764 : mblume 1348 Started using IntInf.int literals and built-in operations
2765 : mblume 1347 (e.g., comparison with 0) where this seems appropriate.
2766 :    
2767 :    
2768 :     ----------------------------------------------------------------------
2769 : macqueen 1343 Name: Dave MacQueen (dbm@cs.uchicago.edu)
2770 :     Date: 2003/08/13 11:36:00 CDT
2771 :     Tag: dbm-20030813-mcz-merge1
2772 :     Description:
2773 :    
2774 :     Merging changes from the mcz-branch development branch into trunk.
2775 :     These changes involve replacement of the emulated old prettyprinter
2776 :     interface with direct use of the SML/NJ Lib PP library, and fixing
2777 :     of a couple of bugs (895, 1186) relating to error messages. A new
2778 :     prettyprinter for ast datatypes (Elaborator/print/ppast.{sig,sml})
2779 :     has been added.
2780 :    
2781 :     ----------------------------------------------------------------------
2782 : mblume 1368 Name: Matthias Blume (blume (at) tti - c (dot) org)
2783 : mblume 1342 Date: 2003/08/11 15:45:00 CDT
2784 :     Tag: blume-20030811-windows
2785 :     Description:
2786 :    
2787 :     Version number bumped to 110.42.9. NEW BOOTFILES!!!
2788 :    
2789 :     http://smlnj.cs.uchicago.edu/dist/working/110.42.9/
2790 :    
2791 :     This patch restores SML/NJ's ability to run under win32. There are a
2792 :     number of changes, including fixes for several bugs that had gone
2793 :     unnoticed until now:
2794 :    
2795 :     - uname "CYGWIN_NT*" is recognized as win32 (This is relevant only when
2796 :     trying to run the win32 version from within cygwin.)
2797 :    
2798 :     - There are a number of simple .bat scripts that substitute for their
2799 :     corresponding Unix shell-scripts. (See below.)
2800 :    
2801 :     - The internals of ml-build have been modified slightly. The main
2802 :     difference is that instead of calling ".link-sml" (or link-sml.bat)
2803 :     using OS.Process.system, the ML process delegates this task back
2804 :     to the script. Otherwise problems arise in mixed environments such
2805 :     as Cygwin where scripts look and work like Unix scripts, but
2806 :     where OS.Process.system cannot run them.
2807 :    
2808 :     - In CM, the srcpath pickler used native pathname syntax -- which
2809 :     is incorrect in the case of cross-compilation. The new pickle format
2810 :     is independent of platform-specific naming conventions.
2811 :    
2812 :     - Path configuration files (such as lib/pathconfig) can now choose
2813 :     between native and standard syntax. Placing a line of the form
2814 :    
2815 :     standard!
2816 :    
2817 :     into the file causes all subsequent paths to be interpreted using
2818 :     CM standard pathname syntax (= Unix conventions); a line
2819 :    
2820 :     native!
2821 :    
2822 :     switches back to native style. This was needed so that
2823 :     path config files can be written portably, see src/system/pathconfig.
2824 :    
2825 :     - Runtime system:
2826 :    
2827 :     - win32-filesys.c: get_file_time and set_file_time now
2828 :     access modification time, not creation time.
2829 :    
2830 :     - I/O code made aware of new array representation.
2831 :    
2832 :     - Bug fixes in X86.prim.masm.
2833 :    
2834 :     - src/system/makeml made aware of win32. (For use under cygwin
2835 :     and other Unix-environments for windows.)
2836 :    
2837 :     - In Basis, fixed off-by-one error in win32-io.sml (function vecF)
2838 :     which caused BinIO.inputAll to fail consistently.
2839 :    
2840 :     .bat scripts:
2841 :    
2842 :     Windows .bat scripts assume that SMLNJ_HOME is defined.
2843 :    
2844 :     - sml.bat, ml-yacc.bat, ml-lex.bat: Driver scripts for standalone
2845 :     applications (sml, ml-yacc, ml-lex).
2846 :     - ml-build.bat: analogous to ml-build.
2847 :     - config\install.bat: Analogous to config/install.sh. This requires
2848 :     that SMLNJ_HOME is set and that Microsoft Visual C is ready to use.
2849 :     (nmake etc. must be on the path, and vcvars32 must have been run.)
2850 :     Moreover, sources for ml-lex and ml-yacc need to exist under src,
2851 :     and the bootfile hierarchy must have been unpacked under
2852 :     sml.boot.x86-win32.
2853 :     The script is very primitive and does a poor job at error checking.
2854 :     It only installs the base system, ml-lex, and ml-yacc. No other
2855 :     libraries are being installed (i.e., you get only those that
2856 :     are part of the compiler.)
2857 :     - link-sml.bat: analogous to .link-sml, but not currently used
2858 :    
2859 :     Unrelated bug fixes:
2860 :    
2861 :     - ml-nlffigen now exports structures ST_* corresponding to incomplete
2862 :     types.
2863 :     - Added getDevice to PP/src/pp-debug-fn.sml. (Would not compile
2864 :     otherwise.)
2865 :    
2866 :     ----------------------------------------------------------------------
2867 : macqueen 1337 Name: Dave MacQueen (macqueen@cs.uchicago.edu)
2868 :     Date: 2003/06/17
2869 :     Tag: macqueen-20030617-bug895
2870 :     Description:
2871 :    
2872 :     Modified compiler/Elaborator/print/pptype.sml to fix bug 895.
2873 :     Tag will be used for new development branch (mcz-branch) for
2874 :     use by MacQueen, (Lucasz) Zairek, and (George) Cao at uchicago.
2875 :    
2876 :     ----------------------------------------------------------------------
2877 : mblume 1368 Name: Matthias Blume (blume (at) tti - c (dot) org)
2878 : mblume 1335 Date: 2003/05/27 16:55:00 CDT
2879 :     Tag: blume-20030527-polyeq
2880 :     Description:
2881 :    
2882 :     Tried to eliminated most cases of polymorphic equality.
2883 :    
2884 :     ----------------------------------------------------------------------
2885 : mblume 1368 Name: Matthias Blume (blume (at) tti - c (dot) org)
2886 : mblume 1334 Date: 2003/05/21 17:45:00 CDT
2887 :     Tag: blume-20030517-complete
2888 :     Description:
2889 :    
2890 :     Two changes:
2891 :    
2892 :     1. Added a flag for controlling whether non-exhaustive bindings will
2893 :     be treated as errors (default is false).
2894 :     2. Cleaned up the *entire* source tree so that CMB.make goes through
2895 :     without a single non-exhaustive match- or bind warning.
2896 :    
2897 :     ----------------------------------------------------------------------
2898 : mblume 1368 Name: Matthias Blume (blume (at) tti - c (dot) org)
2899 : mblume 1332 Date: 2003/05/17 10:20:00 CDT
2900 :     Tag: blume-20030517-absyn
2901 :     Description:
2902 :    
2903 :     1. Added cases for IF, WHILE, ANDALSO, and ORELSE to Absyn.
2904 :    
2905 :     This mainly affects the quality of error messages. However, some
2906 :     of the code is now more straightforward than before. (Treatment of
2907 :     the above four constructs in translate.sml is much simpler than
2908 :     the "macro-expansion" that was going on before. Plus, the mach-
2909 :     compiler no longer gets invoked just to be able to compile an
2910 :     if-expression.)
2911 :    
2912 :     2. The ErrorMsg.Error exception is now caught and absorbed by the
2913 :     interactive loop.
2914 :    
2915 :     ----------------------------------------------------------------------
2916 : allenleung 1327 Name: Allen Leung
2917 :     Date: 2003/05/16 13:05:00 CDT
2918 :     Tag: leunga-20030516-cygwin-runtime
2919 :     Description:
2920 :    
2921 : blume 2155 Ported the runtime system to cygwin, which uses the unix
2922 : allenleung 1327 x86-unix bin files. Missing/buggy features:
2923 :    
2924 : blume 2155 o getnetbyname, getnetbyaddr: these functions seem to be missing in
2925 : allenleung 1327 the Cygwin library.
2926 :     o Ctrl-C handling may be flaky.
2927 :     o Windows system calls and Windows I/O are not supported.
2928 :    
2929 :     A new set of binfiles is located at:
2930 :    
2931 :     http://www.dorsai.org/~leunga/boot.x86-unix.tgz
2932 :    
2933 :     This is only needed for bootstrapping the cygwin version of smlnj.
2934 :     Other x86 versions can use the existing binfiles.
2935 :    
2936 :     ----------------------------------------------------------------------
2937 : mblume 1317 Name: Matthias Blume
2938 :     Date: 2003/04/08 15:42:00 CDT
2939 :     Tag: blume-20030408-listpair
2940 :     Description:
2941 :    
2942 :     1. Added a target 'mlrisc' to installer.
2943 :    
2944 :     2. Added missing elements to structure ListPair.
2945 :    
2946 :     ----------------------------------------------------------------------
2947 : leunga 1296 Name: Allen Leung
2948 :     Date: 2003/01/07 10:40:00 EST
2949 :     Tag: leunga-20030107-int-rem
2950 :     Description:
2951 :    
2952 :     Fixed a bug in Int.rem(x,y) where y is a power of 2 on x86.
2953 :     The arguments to the SUBL instruction were swapped.
2954 :    
2955 :     ----------------------------------------------------------------------
2956 : blume 1229 Name: Matthias Blume
2957 : blume 1294 Date: 2002/12/12 16:25:00 EST
2958 :     Tag: blume-20021212-risc-ra
2959 :     Description:
2960 :    
2961 :     Fixed a serious bug in the rewrite code for FP spilling/reloading that
2962 :     sent the RA into an infinite loop when floating point registers get
2963 :     spilled. (Because of this bug, e.g., nucleic stopped compiling between
2964 :     110.37 and 110.38.)
2965 :     There was another set of potential problems related to the handling of
2966 :     MLRISC annotations (but those did not yet cause real problems, apparently).
2967 :    
2968 :     ----------------------------------------------------------------------
2969 :     Name: Matthias Blume
2970 : blume 1291 Date: 2002/12/06 22:40:00 EST
2971 :     Tag: blume-20021206-cm-fileid
2972 :     Description:
2973 :    
2974 :     Added a call of SrcPath.sync at the beginning of Parse.parse (in CM).
2975 :     This fixes the problem of CM getting confused by files that suddenly
2976 :     change their identity (e.g., by getting unlinked and recreated by some
2977 :     text editor such as vi). There might be a better/cheaper/cleaner way
2978 :     of doing this, but for now this will have to do.
2979 :    
2980 :     ----------------------------------------------------------------------
2981 :     Name: Matthias Blume
2982 : blume 1288 Date: 2002/10/28 09:50:00 EST
2983 : blume 2155 Tag: blume-20021028-typecheck
2984 : blume 1288 Description:
2985 :    
2986 :     Exported structure Typecheck from $smlnj/viscomp/core.cm.
2987 :    
2988 :     ----------------------------------------------------------------------
2989 :     Name: Matthias Blume
2990 : blume 1282 Date: 2002/10/17 09:10:00 EDT
2991 : blume 1279 Tag: Release_110_42
2992 :     Description:
2993 :    
2994 : blume 1282 In good old tradition, there has been a slight hiccup so that we have
2995 :     to patch 110.42 after the fact. The old release tag has been replaced
2996 :     (see below).
2997 :    
2998 :     The change solves a problem with two competing approaches the
2999 :     configuration problem regarding MacOS 10.1 vs. MacOS 10.2 which got in
3000 :     each other's way.
3001 :    
3002 :     This change only affects the runtime system code and the installer script.
3003 :     (No new bootfiles.)
3004 :    
3005 :     ----------------------------------------------------------------------
3006 :     Name: Matthias Blume
3007 :     Date: 2002/10/16 12:00:00 EDT
3008 :     Tag: Release_110_42_removed
3009 :     Description:
3010 :    
3011 : blume 1279 New working release. New bootfiles.
3012 :    
3013 :     ----------------------------------------------------------------------
3014 :     Name: Matthias Blume
3015 : blume 1273 Date: 2002/10/10 13:10:00 EDT
3016 :     Tag: blume-20021010-ppc-divs
3017 :     Description:
3018 :    
3019 :     The mltree operator DIVS must be implemented with an overflow check on
3020 :     the PPC because the hardware indicates divide-by-zero using "overflow" as
3021 :     well.
3022 :    
3023 :     ----------------------------------------------------------------------
3024 :     Name: Matthias Blume
3025 : blume 1261 Date: 2002/07/23 11:45:00
3026 :     Tag: blume-20020723-smlnj-home
3027 :     Description:
3028 :    
3029 :     Sml now senses the SMLNJ_HOME environment variable. If this is set,
3030 :     then the bin dir is assumed to be in $SMLNJ_HOME/bin and (unless
3031 :     CM_PATHCONFIG is also set), the path configuration file is assumed
3032 :     to be in $SMLNJ_HOME/lib/pathconfig. This way one can easily move
3033 :     the entire tree to some other place and everything will "just work".
3034 :    
3035 :     (Companion commands such as ml-build and ml-makedepend also sense this
3036 :     variable.)
3037 :    
3038 :     ----------------------------------------------------------------------
3039 :     Name: Matthias Blume
3040 : blume 1259 Date: 2002/07/12 21:19:00 EDT
3041 :     Tag: blume-20020712-liveness
3042 :     Description:
3043 :    
3044 :     Exported two useful "step" functions from liveness module (MLRISC).
3045 :    
3046 :     ----------------------------------------------------------------------
3047 :     Name: Matthias Blume
3048 : blume 1253 Date: 2002/07/05 16:00 EDT
3049 :     Tag: Release_110_41
3050 :     Description:
3051 :    
3052 :     New working release. New bootfiles.
3053 :    
3054 :     ----------------------------------------------------------------------
3055 :     Name: Matthias Blume
3056 : blume 1250 Date: 2002/07/05 10:25:00 EDT
3057 :     Tag: blume-20020705-btimp
3058 :     Description:
3059 :    
3060 :     Exported structure BTImp from $smlnj/viscomp/debugprof.cm so that
3061 :     other clients can set up backtracing support.
3062 :    
3063 :     ----------------------------------------------------------------------
3064 :     Name: Matthias Blume
3065 : blume 1249 Date: 2002/06/25 17:23:00 EDT
3066 :     Tag: blume-20020625-fpmax
3067 :     Description:
3068 :    
3069 :     Fixed a bug in translation of INLMAX (and INLMIN) for the floating-point
3070 :     case. (The sense of the isNaN test was reversed -- which made min and
3071 :     max always return their first argument.)
3072 :    
3073 :     ----------------------------------------------------------------------
3074 :     Name: Matthias Blume
3075 : blume 1240 Date: 2002/06/11
3076 :     Tag: blume-20020611-unixpath
3077 :     Description:
3078 :    
3079 :     Back-ported OS.Path.{from,to}UnixPath from idlbasis-devel branch.
3080 :    
3081 :     ----------------------------------------------------------------------
3082 :     Name: Matthias Blume
3083 : blume 1238 Date: 2002/06/10 16:35:00 EDT
3084 :     Tag: blume-20020610-ieeereal
3085 :     Description:
3086 :    
3087 :     I back-ported my implementation of IEEEReal.fromString from the
3088 :     idlbasis-devel branch so that we can test it.
3089 :    
3090 :     Another small change is that ppDec tries to give more information
3091 :     than just "<sig>" in the case of functors. However, this code is
3092 :     broken in some mysterious way if the functor's body's signature
3093 :     has not been declared by ascription but gets inferred from the
3094 :     implementation. This needs fixing...
3095 :    
3096 :     ----------------------------------------------------------------------
3097 :     Name: Matthias Blume
3098 : blume 1229 Date: 2002/05/31
3099 :     Tag: blume-20020531-btrace-mode
3100 :     Description:
3101 :    
3102 :     Resurrected SMLofNJ.Internals.BTrace.mode. (It accidentally fell by
3103 :     the wayside when I switched over to using Controls everywhere.)
3104 :    
3105 :     ----------------------------------------------------------------------
3106 : george 1222 Name: Lal George
3107 :     Date: 2002/05/23 12:21:40 EDT
3108 :     Tag: george-20020523-visual-labels
3109 :     Description:
3110 :    
3111 :     Labels are now displayed in the graphical output to make
3112 :     the fall-through and target blocks obvious.
3113 :    
3114 :     ----------------------------------------------------------------------
3115 : blume 1204 Name: Matthias Blume
3116 : blume 1220 Date: 2002/05/22 11:03:00 EDT
3117 :     Tag: blume-20020522-shrink
3118 :     Description:
3119 :    
3120 :     John tweaked yesterday's fix for 1131 to handle an out-of-memory
3121 :     situation that comes up when allocating huge arrays.
3122 :    
3123 :     ----------------------------------------------------------------------
3124 :     Name: Matthias Blume
3125 : blume 1215 Date: 2002/05/21 16:00:00 EDT
3126 :     Tag: Release_110_40
3127 :     Description:
3128 :    
3129 :     New working release (110.40). New bootfiles.
3130 :    
3131 :     [Also: John Reppy fixed GC bug 1131.]
3132 :    
3133 :     ----------------------------------------------------------------------
3134 :     Name: Matthias Blume
3135 : blume 1212 Date: 2002/05/21 12:35:00 EDT
3136 :     Tag: blume-20020521-cmdoc
3137 :     Description:
3138 :    
3139 :     CM documentation update.
3140 :    
3141 :     ----------------------------------------------------------------------
3142 :     Name: Matthias Blume
3143 : blume 1211 Date: 2002/05/21 10:55:00 EDT
3144 :     Tag: blume-20020521-misc
3145 :     Description:
3146 :    
3147 :     - John tweaked runtime to be silent on heap export (except when
3148 :     GC messages are on).
3149 :     - I added a few more things (cross-compiling versions of CMB) to
3150 :     config/preloads (as suggestions).
3151 :    
3152 :     ----------------------------------------------------------------------
3153 :     Name: Matthias Blume
3154 : blume 1208 Date: 2002/05/20 22:25:00 EDT
3155 :     Tag: blume-20020520-controls
3156 :     Description:
3157 :    
3158 :     - Added ControlUtil structure to control-lib.cm.
3159 :     - Use it throughout.
3160 :     - Used Controls facility to define MLRISC controls (as opposed to
3161 :     registering MLRISC control ref cells with Controls after the fact)
3162 :     - Fixed messed-up controls priorities.
3163 :    
3164 :     * Removed again all the stuff from config/preloads that one wouldn't
3165 :     be able to preload at the time the initial heap image is built.
3166 :     (Many libraries, e.g., CML, do not exist yet at this time. The
3167 :     only libraries that can be preloaded via config/preloads are those
3168 :     that come bundled with the bootfiles.)
3169 :    
3170 :     ----------------------------------------------------------------------
3171 :     Name: Matthias Blume
3172 : blume 1204 Date: 2002/05/20 10:59:00 EDT
3173 :     Tag: blume-20020520-preloads
3174 :     Description:
3175 :    
3176 :     Added a lot of commented-out suggestions for things to be included
3177 :     in config/preloads.
3178 :    
3179 :     ----------------------------------------------------------------------
3180 : leunga 1203 Name: Allen Leung
3181 :     Date: 2002/05/18 14:20:00 EDT
3182 :     Tag: leunga-20020518-mdl
3183 :     Description:
3184 :    
3185 :     o Made the mdl tool stuff compile and run again.
3186 :     o I've disabled all the stuff that depends on RTL specifications; they
3187 :     are all badly broken anyway.
3188 :    
3189 :     ----------------------------------------------------------------------
3190 : blume 1201 Name: Matthias Blume
3191 :     Date: 2002/05/17 16:49:00 EDT
3192 :     Tag: blume-20020517-controls
3193 :     Description:
3194 :    
3195 :     0. John Reppy made several modifications to the SML/NJ library.
3196 :     In particular, there is a shiny new controls-lib.cm.
3197 :    
3198 :     1. Pushed new controls interface through compiler so that everything
3199 :     compiles again.
3200 :    
3201 :     2. Added FormatComb and FORMAT_COMB to the CML version of the
3202 :     SML/NJ library (so that CML compiles again).
3203 :    
3204 :     3. Modified init scripts because XXX_DEFAULT environment variables
3205 :     are no longer with us. (Boot-time initialization is now done
3206 :     using the same environment variables that are also used for
3207 :     startup-time initialization of controls.)
3208 :    
3209 :     ----------------------------------------------------------------------
3210 : george 1192 Name: Lal George
3211 :     Date: 2002/05/15 09:20:10 EDT
3212 :     Tag: george-20020515-pseudo-op-decls
3213 :     Description:
3214 :    
3215 :     All pseudo-ops emitted before the first segment declaration
3216 :     such as TEXT, DATA, and BSS directives are assumed to be global
3217 :     declarations and are emitted first in the assembly file. This is
3218 :     useful in a number of situations where one has pseudo-ops that are not
3219 :     specific to any segment, and also works around the constraint that one
3220 :     cannot have client pseudo-ops in the TEXT segment.
3221 :    
3222 :     Because no segment is associated with these declarations it is
3223 :     an error to allocate any space or objects before the first segment
3224 :     directive and an exception will be raised. However, we cannot make
3225 :     this check for client pseudo-ops.
3226 :    
3227 :     These top level declarations are a field in the CFG graph_info.
3228 :     In theory you can continue to add to this field after the CFG has been
3229 :     built -- provided you know what you are doing;-)
3230 :    
3231 :     ----------------------------------------------------------------------
3232 : blume 1175 Name: Matthias Blume
3233 : blume 1190 Date: 2002/05/13 16:40:00 EDT
3234 :     Tag: blume-20020513-pp-etc
3235 :     Description:
3236 :    
3237 :     A few minor bugfixes:
3238 :    
3239 :     - Stopgap measure for bug recently reported by Elsa Gunter (ppDec).
3240 :     (Bogus printouts for redefined bindings still occur. Compiler
3241 :     bug should no longer occur now. We need to redo the prettyprinter
3242 :     from scratch.)
3243 :    
3244 :     - CM pathname printer now also adds escape sequences for ( and )
3245 :    
3246 :     - commend and docu fixes for ml-nlffi
3247 :    
3248 :     ----------------------------------------------------------------------
3249 :     Name: Matthias Blume
3250 : blume 1189 Date: 2002/05/10 16:40:00 EDT
3251 :     Tag: blume-20020510-erg-textio
3252 :     Description:
3253 :    
3254 :     Applied the following bugfix provided by Emden Gansner:
3255 :    
3256 :     Output is corrupted when outputSubstr is used rather than output.
3257 :    
3258 : blume 2155 The problem occurs when a substring
3259 : blume 1189
3260 :     ss = (s, dataStart, dataLen)
3261 :    
3262 : blume 2155 where dataStart > 0, fills a stream buffer with avail bytes left.
3263 :     avail bytes of s, starting at index dataStart, are copied into the
3264 : blume 1189 buffer, the buffer is flushed, and then the remaining dataLen-avail
3265 : blume 2155 bytes of ss are copied into the beginning of the buffer. Instead of
3266 :     starting this copy at index dataStart+avail in s, the current code
3267 : blume 1189 starts the copy at index avail.
3268 :    
3269 :     Fix:
3270 :     In text-io-fn.sml, change line 695 from
3271 :     val needsFlush = copyVec(v, avail, dataLen-avail, buf, 0)
3272 : blume 2155 to
3273 : blume 1189 val needsFlush = copyVec(v, dataStart+avail, dataLen-avail, buf, 0)
3274 :    
3275 :     ----------------------------------------------------------------------
3276 :     Name: Matthias Blume
3277 : blume 1186 Date: 2002/04/12 13:55:00 EDT
3278 :     Tag: blume-20020412-assyntax
3279 :     Description:
3280 :    
3281 :     1. Grabbed newer assyntax.h from the XFree86 project.
3282 :     2. Fiddled with how to compile X86.prim.asm without warnings.
3283 :     3. (Very) Minor cleanup in CM.
3284 :    
3285 :     ----------------------------------------------------------------------
3286 :     Name: Matthias Blume
3287 : blume 1185 Date: 2002/04/01 (no joke!) 17:07:00 EST
3288 :     Tag: blume-20020401-x86div
3289 :     Description:
3290 :    
3291 :     Added full support for div/mod/rem/quot on the x86, using the machine
3292 :     instruction's two results (without clumsily recomputing the remainder)
3293 :     directly where appropriate.
3294 :    
3295 :     Some more extensive power-of-two support was added to the x86 instruction
3296 :     selector (avoiding expensive divs, mods, and muls where they can be
3297 :     replaced with cheaper shifts and masks). However, this sort of thing
3298 :     ought to be done earlier, e.g., within the CPS optimizer so that
3299 :     all architectures benefit from it.
3300 :    
3301 :     The compiler compiles to a fixed point, but changes might be somewhat
3302 :     fragile nevertheless. Please, report any strange things that you might
3303 :     see wrt. div/mod/quot/rem...
3304 :    
3305 :     ----------------------------------------------------------------------
3306 :     Name: Matthias Blume
3307 : blume 1184 Date: 2002/03/29 17:22:00
3308 :     Tag: blume-20020329-div
3309 :     Description:
3310 :    
3311 :     Fixed my broken div/mod logic. Unfortunately, this means that the
3312 :     inline code for div/mod now has one more comparison than before.
3313 :     Fast paths (quotient > 0 or remainder = 0) are not affected, though.
3314 :     The problem was with quotient = 0, because that alone does not tell
3315 :     us which way the rounding went. One then has to look at whether
3316 :     remainder and divisor have the same sign... :(
3317 :    
3318 :     Anyway, I replaced the bootfiles with fresh ones...
3319 :    
3320 :     ----------------------------------------------------------------------
3321 :     Name: Matthias Blume
3322 : blume 1183 Date: 2002/03/29 14:10:00 EST
3323 :     Tag: blume-20020329-inlprims
3324 :     Description:
3325 :    
3326 :     NEW BOOTFILES!!! Version number bumped to 110.39.3.
3327 :    
3328 :     Primops have changed. This means that the bin/boot-file formats have
3329 :     changed as well.
3330 :    
3331 :     To make sure that there is no confusion, I made a new version.
3332 :    
3333 :    
3334 :     CHANGES:
3335 :    
3336 :     * removed REMT from mltree (remainder should never overflow).
3337 :    
3338 :     * added primops to deal with divisions of all flavors to the frontend
3339 :    
3340 :     * handled these primops all the way through so they map to their respective
3341 :     MLRISC support
3342 :    
3343 :     * used these primops in the implementation of Int, Int32, Word, Word32
3344 :    
3345 :     * removed INLDIV, INLMOD, and INLREM as they are no longer necessary
3346 :    
3347 :     * parameterized INLMIN, INLMAX, and INLABS by a numkind
3348 :    
3349 :     * translate.sml now deals with all flavors of INL{MIN,MAX,ABS}, including
3350 :     floating point
3351 :    
3352 :     * used INL{MIN,MAX,ABS} in the implementation of Int, Int32, Word, Word32,
3353 :     and Real (but Real.abs maps to a separate floating-point-only primop)
3354 :    
3355 :    
3356 :     TODO items:
3357 :    
3358 :     * Hacked Alpha32 instruction selection, disabling the selection of REMx
3359 :     instructions because the machine instruction encoder cannot handle
3360 :     them. (Hppa, PPC, and Sparc instruction selection did not handle
3361 :     REM in the first place, and REM is supported by the x86 machine coder.)
3362 :    
3363 :     * Handle DIV and MOD with DIV_TO_NEGINF directly in the x86 instruction
3364 :     selection phase. (The two can be streamlined because the hardware
3365 :     delivers both quotient and remainder at the same time anyway.)
3366 :    
3367 :     * Think about what to do with "valOf(Int32.minInt) div ~1" and friends.
3368 :     (Currently the behavior is inconsistent both across architectures and
3369 :     wrt. the draft Basis spec.)
3370 :    
3371 :     * Word8 should eventually be handled natively, too.
3372 :    
3373 :     * There seems to be one serious bug in mltree-gen.sml. It appears, though,
3374 :     as if there currently is no execution path that could trigger it in
3375 :     SML/NJ. (The assumptions underlying functions arith and promotable do not
3376 :     hold for things like multiplication and division.)
3377 :    
3378 :     ----------------------------------------------------------------------
3379 :     Name: Matthias Blume
3380 : blume 1181 Date: 2002/03/27 16:27:00 EST
3381 :     Tag: blume-20020327-mlrisc-divisions
3382 :     Description:
3383 :    
3384 :     Added support for all four division operations (ML's div, mod, quot,
3385 :     and rem) to MLRISC. In the course of doing so, I also rationalized
3386 :     the naming (no more annoying switch-around of DIV and QUOT), by
3387 :     parameterizing the operation by div_rounding_mode (which can be either
3388 :     DIV_TO_ZERO or DIV_TO_NEGINF).
3389 :    
3390 :     The generic MLTreeGen functor takes care of compiling all four
3391 :     operations down to only round-to-zero div.
3392 :    
3393 :     Missing pieces:
3394 :    
3395 :     * Doing something smarter than relying on MLTreeGen on architectures
3396 :     like, e.g., the x86 where hardware division delivers both quotient and
3397 :     remainder at the same time. With this, the implementation of the
3398 :     round-to-neginf operations could be further streamlined.
3399 :    
3400 :     * Remove inlining support for div/mod/rem from the frontend and replace it
3401 :     with primops that get carried through to the backend. Do this for all
3402 :     int and word types.
3403 :    
3404 :     ----------------------------------------------------------------------
3405 :     Name: Matthias Blume
3406 : blume 1180 Date: 2002/03/25 17:25:00 EST
3407 :     Tag: blume-20020325-divmod
3408 :     Description:
3409 :    
3410 :     I improved (hopefully without breaking them) the implementation of Int.div,
3411 :     Int.mod, and Int.rem. For this, the code in translate.sml now takes
3412 :     advantage of the following observations:
3413 :    
3414 :     Let q = x quot y r = x rem y
3415 :     d = x div y m = x mod y
3416 :    
3417 :     where "quot" is the round-to-zero version of integer division that
3418 :     hardware usually provides. Then we have:
3419 :    
3420 :     r = x - q * y where neither the * nor the - will overflow
3421 :     d = if q >= 0 orelse x = q * y then q else q - 1
3422 :     where neither the * nor the - will overflow
3423 :     m = if q >= 0 orelse r = 0 then r else r + y
3424 :     where the + will not overflow
3425 :    
3426 :     This results in substantial simplification of the generated code.
3427 :     The following table shows the number of CFG nodes and edges generated
3428 :     for
3429 :     fun f (x, y) = x OPER y
3430 :     (* with OPER \in div, mod, quot, rem *)
3431 :    
3432 :    
3433 :     OPER | nodes(old) | edges(old) | nodes(new) | edges(new)
3434 :     --------------------------------------------------------
3435 :     div | 24 | 39 | 12 | 16
3436 :     mod | 41 | 71 | 12 | 16
3437 :     quot | 8 | 10 | 8 | 10
3438 :     rem | 10 | 14 | 8 | 10
3439 :    
3440 :    
3441 :     ----------------------------------------------------------------------
3442 :     Name: Matthias Blume
3443 : blume 1179 Date: 2002/03/25 22:06:00 EST
3444 :     Tag: blume-20020325-cprotobug
3445 :     Description:
3446 :    
3447 :     Fixed a bug in cproto (c prototype decoder).
3448 :    
3449 :     ----------------------------------------------------------------------
3450 :     Name: Matthias Blume
3451 : blume 1178 Date: 2002/03/25 16:00:00 EST
3452 :     Tag: blume-20020325-raw-primops
3453 :     Description:
3454 :    
3455 :     I did some cleanup to Allen's new primop code and
3456 :     replaced yesterday's bootfiles with new ones.
3457 :     (But they are stored in the same place.)
3458 :    
3459 :     ----------------------------------------------------------------------
3460 :     Name: Matthias Blume
3461 : blume 1175 Date: 2002/03/24 22:40:00 EST
3462 :     Tag: blume-20020324-bootfiles
3463 :     Description:
3464 :    
3465 :     Made the bootfiles that Allen asked for.
3466 :    
3467 :     ----------------------------------------------------------------------
3468 : leunga 1174 Name: Allen Leung
3469 :     Date: 2002/03/23 15:50:00 EST
3470 :     Tag: leunga-20020323-flint-cps-rcc-primops
3471 :     Description:
3472 :    
3473 :     1. Changes to FLINT primops:
3474 :    
3475 :     (* make a call to a C-function;
3476 :     * The primop carries C function prototype information and specifies
3477 :     * which of its (ML-) arguments are floating point. C prototype
3478 :     * information is for use by the backend, ML information is for
3479 :     * use by the CPS converter. *)
3480 :     | RAW_CCALL of { c_proto: CTypes.c_proto,
3481 :     ml_args: ccall_type list,
3482 :     ml_res_opt: ccall_type option,
3483 :     reentrant : bool
3484 :     } option
3485 :     (* Allocate uninitialized storage on the heap.
3486 :     * The record is meant to hold short-lived C objects, i.e., they
3487 : blume 2155 * are not ML pointers. With the tag, the representation is
3488 : leunga 1174 * the same as RECORD with tag tag_raw32 (sz=4), or tag_fblock (sz=8)
3489 :     *)
3490 : blume 2155 | RAW_RECORD of {tag:bool,sz:int}
3491 : leunga 1174 and ccall_type = CCALL_INT32 | CCALL_REAL64 | CCALL_ML_PTR
3492 :    
3493 :     2. These CPS primops are now overloaded:
3494 :    
3495 :     rawload of {kind:numkind}
3496 :     rawstore of {kind:numkind}
3497 :    
3498 :     The one argument form is:
3499 :    
3500 :     rawload {kind} address
3501 :    
3502 :     The two argument form is:
3503 : blume 2155
3504 : leunga 1174 rawload {kind} [ml object, byte-offset]
3505 :    
3506 :     3. RAW_CCALL/RCC now takes two extra arguments:
3507 :    
3508 :     a. The first is whether the C call is reentrant, i.e., whether
3509 :     ML state should be saved and restored.
3510 :     b. The second argument is a string argument specifying the name of
3511 :     library and the C function.
3512 :    
3513 :     These things are currently not handled in the code generator, yet.
3514 :    
3515 :     4. In CProto,
3516 :    
3517 : blume 2155 An encoding type of "bool" means "ml object" and is mapped into
3518 :     C prototype of PTR. Note that "bool" is different than "string",
3519 : leunga 1174 even though "string" is also mapped into PTR, because "bool"
3520 :     is assigned an CPS type of BOGt, while "string" is assigned INT32t.
3521 :    
3522 :     5. Pickler/unpicker
3523 :    
3524 :     Changed to handle RAW_RECORD and newest RAW_CCALL
3525 :    
3526 :     6. MLRiscGen,
3527 :    
3528 :     1. Changed to handle the new rawload/rawstore/rawrecord operators.
3529 :     2. Code for handling C Calls has been moved to a new module CPSCCalls,
3530 :     in the file CodeGen/cpscompile/cps-c-calls.sml
3531 :    
3532 :     7. Added the conditional move operator
3533 :    
3534 :     condmove of branch
3535 :    
3536 : blume 2155 to cps. Generation of this is still buggy so it is currently
3537 : leunga 1174 disabled.
3538 : blume 2155
3539 : leunga 1174 ----------------------------------------------------------------------
3540 : george 1168 Name: Lal George
3541 : blume 2155 Date: 2002/03/22 14:18:25 EST
3542 : george 1171 Tag: george-20020322-cps-branch-prob
3543 : george 1168 Description:
3544 :    
3545 : blume 2155 Implemented the Ball-Larus branch prediction-heuristics, and
3546 : george 1168 incorporated graphical viewers for control flow graphs.
3547 :    
3548 : george 1171 Ball-Larus Heuristics:
3549 : george 1168 ---------------------
3550 :     See the file compiler/CodeGen/cpscompile/cpsBranchProb.sml.
3551 :    
3552 :     By design it uses the Dempster-Shafer theory for combining
3553 :     probabilities. For example, in the function:
3554 :    
3555 :     fun f(n,acc) = if n = 0 then acc else f(n-1, n*acc)
3556 :    
3557 :     the ball-larus heuristics predicts that the n=0 is unlikely
3558 :     (OH-heuristic), and the 'then' branch is unlikely because of the
3559 :     RH-heuristic -- giving the 'then' branch an even lower combined
3560 : george 1171 probability using the Dempster-Shafer theory.
3561 : george 1168
3562 :     Finally, John Reppy's loop analysis in MLRISC, further lowers the
3563 :     probability of the 'then' branch because of the loop in the else
3564 :     branch.
3565 :    
3566 :    
3567 :     Graphical Viewing:
3568 : blume 2155 ------------------
3569 : george 1168 I merely plugged in Allen's graphical viewers into the compiler. The
3570 :     additional code is not much. At the top level, saying:
3571 :    
3572 :     Control.MLRISC.getFlag "cfg-graphical-view" := true;
3573 :    
3574 :     will display the graphical view of the control flow graph just before
3575 :     back-patching. daVinci must be in your path for this to work. If
3576 :     daVinci is not available, then the default viewer can be changed
3577 :     using:
3578 :    
3579 :     Control.MLRISC.getString "viewer"
3580 :    
3581 :     which can be set to "dot" or "vcg" for the corresponding viewers. Of
3582 :     course, these viewers must be in your path.
3583 :    
3584 :     The above will display the compilation unit at the level of clusters,
3585 :     many of which are small, boring, and un-interesting. Also setting:
3586 :    
3587 :     Control.MLRISC.getInt "cfg-graphical-view_size"
3588 :    
3589 :     will display clusters that are larger than the value set by the above.
3590 :    
3591 : blume 2155
3592 : george 1168 ----------------------------------------------------------------------
3593 : blume 1159 Name: Matthias Blume
3594 :     Date: 2002/03/21 22:20:00 EST
3595 :     Tag: blume-20020321-kmp-bugfix
3596 :     Description:
3597 :    
3598 :     Changed the interface to the KMP routine in PreString and fixed
3599 :     a minor bug in one place where it was used.
3600 :    
3601 :     ----------------------------------------------------------------------
3602 : leunga 1156 Name: Allen Leung
3603 : leunga 1158 Date: 2002/03/21 20:30:00 EST
3604 :     Tag: leunga-20020321-cfg
3605 :     Description:
3606 :    
3607 :     Fixed a potential problem in cfg edge splitting.
3608 :    
3609 :     ----------------------------------------------------------------------
3610 :     Name: Allen Leung
3611 : leunga 1156 Date: 2002/03/21 17:15:00 EST
3612 :     Tag: leunga-20020321-x86-fp-cfg
3613 :     Description:
3614 :    
3615 : blume 2155 1. Recoded the buggy parts of x86-fp.
3616 : leunga 1156
3617 :     a. All the block reordering code has been removed.
3618 :     We now depend on the block placement phases to do this work.
3619 :    
3620 :     b. Critical edge splitting code has been simplified and moved into the
3621 :     CFG modules, as where they belong.
3622 :    
3623 :     Both of these were quite buggy and complex. The code is now much, much
3624 :     simpler.
3625 :    
3626 :     2. X86 backend.
3627 :    
3628 :     a. Added instructions for 64-bit support. Instruction selection for
3629 :     64-bit has not been committed, however, since that
3630 : blume 2155 requires changes to MLTREE which haven't been approved by
3631 : leunga 1156 Lal and John.
3632 :    
3633 : blume 2155 b. Added support for FUCOMI and FUCOMIP when generating code for
3634 : leunga 1156 PentiumPro and above. We only generate these instructions in
3635 :     the fast-fp mode.
3636 :    
3637 : blume 2155 c. Added cases for JP and JNP in X86FreqProps.
3638 :    
3639 : leunga 1156 3. CFG
3640 :    
3641 :     CFG now has a bunch of methods for edge splitting and merging.
3642 :    
3643 :     4. Machine description.
3644 :    
3645 :     John's simplification of MLTREE_BASIS.fcond broke a few machine
3646 :     description things:
3647 :    
3648 :     rtl-build.{sig,sml} and hppa.mdl fixed.
3649 :    
3650 : blume 2155 NOTE: the machine description stuff in the repository is still broken.
3651 :     Again, I can't put my fixes in because that involves
3652 :     changes to MLTREE.
3653 : leunga 1156
3654 :     ----------------------------------------------------------------------
3655 : blume 1145 Name: Matthias Blume
3656 : blume 1155 Date: 2002/03/20 15:55:00 EST
3657 :     Tag: blume-20020320-kmp
3658 :     Description:
3659 :    
3660 :     Implemented Knuth-Morris-Pratt string matching in PreString and used
3661 :     it for String.isSubstring, Substring.isSubstring, and
3662 :     Substring.position.
3663 :    
3664 :     (Might need some stress-testing. Simple examples worked fine.)
3665 :    
3666 :     ----------------------------------------------------------------------
3667 :     Name: Matthias Blume
3668 : blume 1152 Date: 2002/03/19 16:37:00 EST
3669 :     Tag: blume-20020319-witnesses
3670 :     Description:
3671 :    
3672 :     Added a structure C.W and functions convert/Ptr.convert to ml-nlffi-lib.
3673 :    
3674 :     This implements a generic mechanism for changing constness qualifiers
3675 :     anywhere within big C types without resorting to outright "casts".
3676 :     (So far, functions such as C.rw/C.ro or C.Ptr.rw/C.Ptr.ro only let you
3677 :     modify the constness at the outermost level.)
3678 :     The implementation of "convert" is based on the idea of "witness"
3679 :     values -- values that are not used by the operation but whose types
3680 :     "testify" to their applicability. On the implementation side, "convert"
3681 :     is simply a projection (returning its second curried argument). With
3682 :     cross-module inlining, it should not result in any machine code being
3683 :     generated.
3684 :    
3685 :     ----------------------------------------------------------------------
3686 :     Name: Matthias Blume
3687 : blume 1148 Date: 2002/03/15 16:40:00 EST
3688 :     Tag: blume-20020315-basis
3689 :     Description:
3690 :    
3691 :     Provided (preliminary?) implementations for
3692 :    
3693 :     {String,Substring}.{concatWith,isSuffix,isSubstring}
3694 :    
3695 :     and
3696 :    
3697 :     Substring.full
3698 :    
3699 :     Those are in the Basis spec but they were missing in SML/NJ.
3700 :    
3701 :     ----------------------------------------------------------------------
3702 :     Name: Matthias Blume
3703 : blume 1145 Date: 2002/03/14 21:30:00 EST
3704 :     Tag: blume-20020314-controls
3705 :     Description:
3706 :    
3707 :     Controls:
3708 :     ---------
3709 :    
3710 :     1. Factored out the recently-added Controls : CONTROLS stuff and put
3711 :     it into its own library $/controls-lib.cm. The source tree for
3712 :     this is under src/smlnj-lib/Controls.
3713 :    
3714 :     2. Changed the names of types and functions in this interface, so they
3715 :     make a bit more "sense":
3716 :    
3717 :     module -> registry
3718 :     'a registry -> 'a group
3719 :    
3720 :     3. The interface now deals in ref cells only. The getter/setter interface
3721 :     is (mostly) gone.
3722 :    
3723 :     4. Added a function that lets one register an already-existing ref cell.
3724 :    
3725 :     5. Made the corresponding modifications to the rest of the code so that
3726 :     everything compiles again.
3727 :    
3728 :     6. Changed the implementation of Controls.MLRISC back to something closer
3729 :     to the original. In particular, this module (and therefore MLRISC)
3730 :     does not depend on Controls. There now is some link-time code in
3731 :     int-sys.sml that registers the MLRISC controls with the Controls
3732 :     module.
3733 :    
3734 :     CM:
3735 :     ---
3736 :    
3737 :     * One can now specify the lambda-split aggressiveness in init.cmi.
3738 :    
3739 :     ----------------------------------------------------------------------
3740 : leunga 1142 Name: Allen Leung
3741 :     Date: 2002/03/13 17:30:00 EST
3742 :     Tag: leunga-20020313-x86-fp-unary
3743 :     Description:
3744 :    
3745 :     Bug fix for:
3746 :    
3747 :     > leunga@weaselbane:~/Yale/tmp/sml-dist{21} bin/sml
3748 :     > Standard ML of New Jersey v110.39.1 [FLINT v1.5], March 08, 2002
3749 :     > - fun f(x,(y,z)) = Real.~ y;
3750 :     > [autoloading]
3751 :     > [autoloading done]
3752 :     > fchsl (%eax), 184(%esp)
3753 :     > Error: MLRisc bug: X86MCEmitter.emitInstr
3754 : blume 2155 >
3755 : leunga 1142 > uncaught exception Error
3756 :     > raised at: ../MLRISC/control/mlriscErrormsg.sml:16.14-16.19
3757 :    
3758 :     The problem was that the code generator did not generate any fp registers
3759 :     in this case, and the ra didn't know that it needed to run the X86FP phase to
3760 :     translate the pseudo fp instruction. This only happened with unary fp
3761 :     operators in certain situations.
3762 :    
3763 :     ----------------------------------------------------------------------
3764 : blume 1137 Name: Matthias Blume
3765 : blume 1140 Date: 2002/03/13 14:00:00 EST
3766 :     Tag: blume-20020313-overload-etc
3767 :     Description:
3768 :    
3769 :     1. Added _overload as a synonym for overload for backward compatibility.
3770 :     (Control.overloadKW must be true for either version to be accepted.)
3771 :    
3772 :     2. Fixed bug in install script that caused more things to be installed
3773 :     than what was requested in config/targets.
3774 :    
3775 :     3. Made CM aware of the (_)overload construct so that autoloading
3776 :     works.
3777 :    
3778 :     ----------------------------------------------------------------------
3779 :     Name: Matthias Blume
3780 : blume 1138 Date: 2002/03/12 22:03:00 EST
3781 :     Tag: blume-20020312-url
3782 :     Description:
3783 :    
3784 :     Forgot to update BOOT and srcarchiveurl.
3785 :    
3786 :     ----------------------------------------------------------------------
3787 :     Name: Matthias Blume
3788 : blume 1137 Date: 2002/03/12 17:30:00 EST
3789 :     Tag: blume-20020312-version110392
3790 :     Description:
3791 :    
3792 :     Yet another version number bump (because of small changes to the
3793 :     binfile format). Version number is now 110.39.2. NEW BOOTFILES!
3794 :    
3795 :     Changes:
3796 :    
3797 :     The new pid generation scheme described a few weeks ago was overly
3798 :     complicated. I implemented a new mechanism that is simpler and
3799 :     provides a bit more "stability": Once CM has seen a compilation
3800 :     unit, it keeps its identity constant (as long as you do not delete
3801 :     those crucial CM/GUID/* files). This means that when you change
3802 : blume 1145 an interface, compile, then go back to the old interface, and
3803 : blume 1137 compile again, you arrive at the original pid.
3804 :    
3805 :     There now also is a mechanism that instructs CM to use the plain
3806 :     environment hash as a module's pid (effectively making its GUID
3807 :     the empty string). For this, "noguid" must be specified as an
3808 :     option to the .sml file in question within its .cm file.
3809 :     This is most useful for code that is being generated by tools such
3810 :     as ml-nlffigen (because during development programmers tend to
3811 :     erase the tool's entire output directory tree including CM's cached
3812 :     GUIDs). "noguid" is somewhat dangerous (since it can be used to locally
3813 :     revert to the old, broken behavior of SML/NJ, but in specific cases
3814 :     where there is no danger of interface confusion, its use is ok
3815 :     (I think).
3816 :    
3817 :     ml-nlffigen by default generates "noguid" annotations. They can be
3818 :     turned off by specifying -guid in its command line.
3819 :    
3820 :     ----------------------------------------------------------------------
3821 : george 1136 Name: Lal George
3822 :     Date: 2002/03/12 12 14:42:36 EST
3823 :     Tag: george-20020312-frequency-computation
3824 :     Description:
3825 : blume 2155
3826 : george 1136 Integrated jump chaining and static block frequency into the
3827 :     compiler. More details and numbers later.
3828 : leunga 1025
3829 : george 1003 ----------------------------------------------------------------------
3830 : george 1133 Name: Lal George
3831 :     Date: 2002/03/11 11 22:38:53 EST
3832 :     Tag: george-20020311-jump-chain-elim
3833 :     Description:
3834 :    
3835 :     Tested the jump chain elimination on all architectures (except the
3836 :     hppa). This is on by default right now and is profitable for the
3837 :     alpha and x86, however, it may not be profitable for the sparc and ppc
3838 :     when compiling the compiler.
3839 :    
3840 : blume 2155 The gc test will typically jump to a label at the end of the cluster,
3841 :     where there is another jump to an external cluster containing the actual
3842 : george 1133 code to invoke gc. This is to allow factoring of common gc invocation
3843 :     sequences. That is to say, we generate:
3844 :    
3845 :     f:
3846 :     testgc
3847 :     ja L1 % jump if above to L1
3848 :    
3849 :     L1:
3850 :     jmp L2
3851 :    
3852 :    
3853 :     After jump chain elimination the 'ja L1' instructions is converted to
3854 :     'ja L2'. On the sparc and ppc, many of the 'ja L2' instructions may end
3855 :     up being implemented in their long form (if L2 is far away) using:
3856 :    
3857 :     jbe L3 % jump if below or equal to L3
3858 :     jmp L2
3859 :     L3:
3860 :     ...
3861 :    
3862 :    
3863 :     For large compilation units L2 may be far away.
3864 :    
3865 :    
3866 :     ----------------------------------------------------------------------
3867 : blume 1132 Name: Matthias Blume
3868 :     Date: 2002/03/11 13:30:00 EST
3869 :     Tag: blume-20020311-mltreeeval
3870 :     Description:
3871 :    
3872 :     A functor parameter was missing.
3873 :    
3874 :     ----------------------------------------------------------------------
3875 : leunga 1130 Name: Allen Leung
3876 : leunga 1131 Date: 2002/03/11 10:30:00 EST
3877 : leunga 1142 Tag: leunga-20020311-runtime-string0
3878 : leunga 1131 Description:
3879 :    
3880 : blume 2155 The representation of the empty string now points to a
3881 :     legal null terminated C string instead of unit. It is now possible
3882 :     to convert an ML string into C string with InlineT.CharVector.getData.
3883 : leunga 1131 This compiles into one single machine instruction.
3884 :    
3885 :     ----------------------------------------------------------------------
3886 :     Name: Allen Leung
3887 : leunga 1130 Date: 2002/03/10 23:55:00 EST
3888 :     Tag: leunga-20020310-x86-call
3889 :     Description:
3890 :    
3891 :     Added machine generation for CALL instruction (relative displacement mode)
3892 :    
3893 :     ----------------------------------------------------------------------
3894 : blume 1128 Name: Matthias Blume
3895 :     Date: 2002/03/08 16:05:00
3896 :     Tag: blume-20020308-entrypoints
3897 :     Description:
3898 :    
3899 :     Version number bumped to 110.39.1. NEW BOOTFILES!
3900 :    
3901 :     Entrypoints: non-zero offset into a code object where execution should begin.
3902 :    
3903 :     - Added the notion of an entrypoint to CodeObj.
3904 :     - Added reading/writing of entrypoint info to Binfile.
3905 :     - Made runtime system bootloader aware of entrypoints.
3906 :     - Use the address of the label of the first function given to mlriscGen
3907 :     as the entrypoint. This address is currently always 0, but it will
3908 :     not be 0 once we turn on block placement.
3909 :     - Removed the linkage cluster code (which was The Other Way(tm) of dealing
3910 :     with entry points) from mlriscGen.
3911 :    
3912 :     ----------------------------------------------------------------------
3913 : leunga 1127 Name: Allen Leung
3914 :     Date: 2002/03/07 20:45:00 EST
3915 :     Tag: leunga-20020307-x86-cmov
3916 :     Description:
3917 :    
3918 :     Bug fixes for CMOVcc on x86.
3919 :    
3920 :     1. Added machine code generation for CMOVcc
3921 :     2. CMOVcc is now generated in preference over SETcc on PentiumPro or above.
3922 :     3. CMOVcc cannot have an immediate operand as argument.
3923 :    
3924 :     ----------------------------------------------------------------------
3925 : blume 1126 Name: Matthias Blume
3926 :     Date: 2002/03/07 16:15:00 EST
3927 :     Tag: blume-20020307-controls
3928 :     Description:
3929 :    
3930 :     This is a very large but mostly boring patch which makes (almost)
3931 :     every tuneable compiler knob (i.e., pretty much everything under
3932 :     Control.* plus a few other things) configurable via both the command
3933 :     line and environment variables in the style CM did its configuration
3934 :     until now.
3935 :    
3936 :     Try starting sml with '-h' (or, if you are brave, '-H')
3937 :    
3938 :     To this end, I added a structure Controls : CONTROLS to smlnj-lib.cm which
3939 :     implements the underlying generic mechanism.
3940 :    
3941 :     The interface to some of the existing such facilities has changed somewhat.
3942 :     For example, the MLRiscControl module now provides mkFoo instead of getFoo.
3943 :     (The getFoo interface is still there for backward-compatibility, but its
3944 :     use is deprecated.)
3945 :    
3946 :     The ml-build script passes -Cxxx=yyy command-line arguments through so
3947 :     that one can now twiddle the compiler settings when using this "batch"
3948 :     compiler.
3949 :    
3950 :     TODO items:
3951 :    
3952 :     We should go through and throw out all controls that are no longer
3953 :     connected to anything. Moreover, we should go through and provide
3954 :     meaningful (and correct!) documentation strings for those controls
3955 :     that still are connected.
3956 :    
3957 :     Currently, multiple calls to Controls.new are accepted (only the first
3958 :     has any effect). Eventually we should make sure that every control
3959 :     is being made (via Controls.new) exactly once. Future access can then
3960 :     be done using Controls.acc.
3961 :    
3962 :     Finally, it would probably be a good idea to use the getter-setter
3963 :     interface to controls rather than ref cells. For the time being, both
3964 :     styles are provided by the Controls module, but getter-setter pairs are
3965 :     better if thread-safety is of any concern because they can be wrapped.
3966 :    
3967 :     *****************************************
3968 :    
3969 :     One bug fix: The function blockPlacement in three of the MLRISC
3970 :     backpatch files used to be hard-wired to one of two possibilities at
3971 :     link time (according to the value of the placementFlag). But (I
3972 :     think) it should rather sense the flag every time.
3973 :    
3974 :     *****************************************
3975 :    
3976 :     Other assorted changes (by other people who did not supply a HISTORY entry):
3977 :    
3978 :     1. the cross-module inliner now works much better (Monnier)
3979 :     2. representation of weights, frequencies, and probabilities in MLRISC
3980 :     changed in preparation of using those for weighted block placement
3981 :     (Reppy, George)
3982 :    
3983 :     ----------------------------------------------------------------------
3984 : george 1116 Name: Lal George
3985 : george 1124 Date: 2002/03/07 14:44:24 EST 2002
3986 :     Tag: george-20020307-weighted-block-placement
3987 :    
3988 :     Tested the weighted block placement optimization on all architectures
3989 :     (except the hppa) using AMPL to generate the block and edge frequencies.
3990 :     Changes were required in the machine properties to correctly
3991 : blume 2155 categorize trap instructions. There is an MLRISC flag
3992 :     "weighted-block-placement" that can be used to enable weighted block
3993 : george 1124 placement, but this will be ineffective without block/edge
3994 :     frequencies (coming soon).
3995 :    
3996 :    
3997 :     ----------------------------------------------------------------------
3998 :     Name: Lal George
3999 : blume 2155 Date: 2002/03/05 17:24:48 EST
4000 : george 1116 Tag: george-20020305-linkage-cluster
4001 :    
4002 : george 1124 In order to support the block placement optimization, a new cluster
4003 :     is generated as the very first cluster (called the linkage cluster).
4004 : blume 2155 It contains a single jump to the 'real' entry point for the compilation
4005 : george 1124 unit. Block placement has no effect on the linkage cluster itself, but
4006 :     all the other clusters have full freedom in the manner in which they
4007 : blume 2155 reorder blocks or functions.
4008 : george 1116
4009 :     On the x86 the typical linkage code that is generated is:
4010 :     ----------------------
4011 :     .align 2
4012 :     L0:
4013 :     addl $L1-L0, 72(%esp)
4014 : george 1124 jmp L1
4015 : george 1116
4016 :    
4017 :     .align 2
4018 :     L1:
4019 :     ----------------------
4020 :    
4021 : blume 2155 72(%esp) is the memory location for the stdlink register. This
4022 : george 1116 must contain the address of the CPS function being called. In the
4023 : george 1124 above example, it contains the address of L0; before
4024 : george 1116 calling L1 (the real entry point for the compilation unit), it
4025 :     must contain the address for L1, and hence
4026 :    
4027 :     addl $L1-L0, 72(%esp)
4028 :    
4029 : george 1124 I have tested this on all architectures except the hppa.The increase
4030 :     in code size is of course negligible
4031 : george 1116
4032 :     ----------------------------------------------------------------------
4033 : leunga 1115 Name: Allen Leung
4034 :     Date: 2002/03/03 13:20:00 EST
4035 :     Tag: leunga-20020303-mlrisc-tools
4036 :    
4037 :     Added #[ ... ] expressions to mlrisc tools
4038 :    
4039 :     ----------------------------------------------------------------------
4040 : blume 1096 Name: Matthias Blume
4041 : blume 1098 Date: 2002/02/27 12:29:00 EST
4042 :     Tag: blume-20020227-cdebug
4043 :     Description:
4044 :    
4045 :     - made types in structure C and C_Debug to be equal
4046 :     - got rid of code duplication (c-int.sml vs. c-int-debug.sml)
4047 :     - there no longer is a C_Int_Debug (C_Debug is directly derived from C)
4048 :    
4049 :     ----------------------------------------------------------------------
4050 :     Name: Matthias Blume
4051 : blume 1096 Date: 2002/02/26 12:00:00 EST
4052 :     Tag: blume-20020226-ffi
4053 :     Description:
4054 :    
4055 :     1. Fixed a minor bug in CM's "noweb" tool:
4056 :     If numbering is turned off, then truly don't number (i.e., do not
4057 :     supply the -L option to noweb). The previous behavior was to supply
4058 :     -L'' -- which caused noweb to use the "default" line numbering scheme.
4059 :     Thanks to Chris Richards for pointing this out (and supplying the fix).
4060 :    
4061 :     2. Once again, I reworked some aspects of the FFI:
4062 :    
4063 :     A. The incomplete/complete type business:
4064 :    
4065 :     - Signatures POINTER_TO_INCOMPLETE_TYPE and accompanying functors are
4066 :     gone!
4067 :     - ML types representing an incomplete type are now *equal* to
4068 :     ML types representing their corresponding complete types (just like
4069 :     in C). This is still safe because ml-nlffigen will not generate
4070 :     RTTI for incomplete types, nor will it generate functions that
4071 :     require access to such RTTI. But when ML code generated from both
4072 :     incomplete and complete versions of the C type meet, the ML types
4073 :     are trivially interoperable.
4074 :    
4075 :     NOTE: These changes restore the full generality of the translation
4076 :     (which was previously lost when I eliminated functorization)!
4077 :    
4078 :     B. Enum types:
4079 :    
4080 :     - Structure C now has a type constructor "enum" that is similar to
4081 :     how the "su" constructor works. However, "enum" is not a phantom
4082 :     type because each "T enum" has values (and is isomorphic to
4083 :     MLRep.Signed.int).
4084 :     - There are generic access operations for enum objects (using
4085 :     MLRep.Signed.int).
4086 :     - ml-nlffigen will generate a structure E_foo for each "enum foo".
4087 :     * The structure contains the definition of type "mlrep" (the ML-side
4088 :     representation type of the enum). Normally, mlrep is the same
4089 :     as "MLRep.Signed.int", but if ml-nlffigen was invoked with "-ec",
4090 :     then mlrep will be defined as a datatype -- thus facilitating
4091 :     pattern matching on mlrep values.
4092 :     ("-ec" will be suppressed if there are duplicate values in an
4093 :     enumeration.)
4094 :     * Constructors ("-ec") or values (no "-ec") e_xxx of type mlrep
4095 :     will be generated for each C enum constant xxx.
4096 :     * Conversion functions m2i and i2m convert between mlrep and
4097 :     MLRep.Signed.int. (Without "-ec", these functions are identities.)
4098 :     * Coversion functions c and ml convert between mlrep and "tag enum".
4099 :     * Access functions (get/set) fetch and store mlrep values.
4100 :     - By default (unless ml-nlffigen was invoked with "-nocollect"), unnamed
4101 :     enumerations are merged into one single enumeration represented by
4102 :     structure E_'.
4103 :    
4104 :     ----------------------------------------------------------------------
4105 : leunga 1080 Name: Allen Leung
4106 : leunga 1094 Date: 2002/02/25 04:45:00 EST
4107 :     Tag: leunga-20020225-cps-spill
4108 :    
4109 :     This is a new implementation of the CPS spill phase.
4110 :     The new phase is in the new file compiler/CodeGen/cpscompile/spill-new.sml
4111 :     In case of problems, replace it with the old file spill.sml
4112 :    
4113 : blume 2155 The current compiler runs into some serious performance problems when
4114 : leunga 1094 constructing a large record. This can happen when we try to compile a
4115 :     structure with many items. Even a very simple structure like the following
4116 :     makes the compiler slow down.
4117 :    
4118 :     structure Foo = struct
4119 :     val x_1 = 0w1 : Word32.int
4120 :     val x_2 = 0w2 : Word32.int
4121 :     val x_3 = 0w3 : Word32.int
4122 :     ...
4123 :     val x_N = 0wN : Word32.int
4124 :     end
4125 :    
4126 : blume 2155 The following table shows the compile time, from N=1000 to N=4000,
4127 : leunga 1094 with the old compiler:
4128 :    
4129 :     N
4130 : blume 2155 1000 CPS 100 spill 0.04u 0.00s 0.00g
4131 :     MLRISC ra 0.06u 0.00s 0.05g
4132 : leunga 1094 (spills = 0 reloads = 0)
4133 : blume 2155 TOTAL 0.63u 0.07s 0.21g
4134 : leunga 1094
4135 : blume 2155 1100 CPS 100 spill 8.25u 0.32s 0.64g
4136 :     MLRISC ra 5.68u 0.59s 3.93g
4137 : leunga 1094 (spills = 0 reloads = 0)
4138 : blume 2155 TOTAL 14.71u 0.99s 4.81g
4139 : leunga 1094
4140 : blume 2155 1500 CPS 100 spill 58.55u 2.34s 1.74g
4141 :     MLRISC ra 5.54u 0.65s 3.91g
4142 : leunga 1094 (spills = 543 reloads = 1082)
4143 : blume 2155 TOTAL 65.40u 3.13s 6.00g
4144 : leunga 1094
4145 : blume 2155 2000 CPS 100 spill 126.69u 4.84s 3.08g
4146 :     MLRISC ra 0.80u 0.10s 0.55g
4147 : leunga 1094 (spills = 42 reloads = 84)
4148 : blume 2155 TOTAL 129.42u 5.10s 4.13g
4149 : leunga 1094
4150 : blume 2155 3000 CPS 100 spill 675.59u 19.03s 11.64g
4151 :     MLRISC ra 2.69u 0.27s 1.38g
4152 : leunga 1094 (spills = 62 reloads = 124)
4153 : blume 2155 TOTAL 682.48u 19.61s 13.99g
4154 : leunga 1094
4155 : blume 2155 4000 CPS 100 spill 2362.82u 56.28s 43.60g
4156 :     MLRISC ra 4.96u 0.27s 2.72g
4157 : leunga 1094 (spills = 85 reloads = 170)
4158 : blume 2155 TOTAL 2375.26u 57.21s 48.00g
4159 : leunga 1094
4160 : blume 2155 As you can see the old cps spill module suffers from some serious
4161 :     performance problem. But since I cannot decipher the old code fully,
4162 :     instead of patching the problems up, I'm reimplementing it
4163 :     with a different algorithm. The new code is more modular,
4164 :     smaller when compiled, and substantially faster
4165 :     (O(n log n) time and O(n) space). Timing of the new spill module:
4166 : leunga 1094
4167 : blume 2155 4000 CPS 100 spill 0.02u 0.00s 0.00g
4168 :     MLRISC ra 0.25u 0.02s 0.15g
4169 :     (spills=1 reloads=3)
4170 :     TOTAL 7.74u 0.34s 1.62g
4171 : leunga 1094
4172 :     Implementation details:
4173 :    
4174 : blume 2155 As far as I can tell, the purpose of the CPS spill module is to make sure the
4175 :     number of live variables at any program point (the bandwidth)
4176 :     does not exceed a certain limit, which is determined by the
4177 :     size of the spill area.
4178 : leunga 1094
4179 : blume 2155 When the bandwidth is too large, we decrease the register pressure by
4180 : leunga 1094 packing live variables into spill records. How we achieve this is
4181 :     completely different than what we did in the old code.
4182 : &nb