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 2444 - (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 : jhr 2444 Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
18 :     Date: 2007/04/23
19 :     Description:
20 :    
21 :     More Basis fixes: the OS.Path module did not include the InvalidArc
22 :     exception and did not do sufficient argument checking.
23 :    
24 :     ----------------------------------------------------------------------
25 : mrainey 2406 Name: Mike Rainey (http://www.cs.uchicago.edu/~mrainey)
26 : mrainey 2410 Date: 2007/04/16
27 :     Description:
28 :    
29 :     Fixed an unbound functor bug in the AMD64 CM file.
30 :    
31 :     ----------------------------------------------------------------------
32 :     Name: Mike Rainey (http://www.cs.uchicago.edu/~mrainey)
33 : mrainey 2406 Date: 2007/04/12
34 :     Description:
35 :    
36 :     Substantially changed the signature and implementation of AMD64
37 :     SVID. It now looks similar to the ia32 SVID, yet uses staged
38 :     allocation.
39 :    
40 :     ----------------------------------------------------------------------
41 : jhr 2387 Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
42 :     Date: 2007/04/07
43 :     Description:
44 :    
45 :     More Basis fixes: the WORD signature was missing {to,from}Large.
46 :     Thanks to Andreas Rossberg.
47 :    
48 :     ----------------------------------------------------------------------
49 : blume 2368 Name: Matthias Blume (http://www.tti-c.org/blume/)
50 :     Date: 2007/03/22 16:07:00 CDT
51 :     Tag: release-110.63
52 :     Description:
53 :    
54 :     Bumped version number to 110.63.
55 :    
56 :     !!!NEW BOOTFILES!!!
57 :    
58 :     ----------------------------------------------------------------------
59 : mrainey 2360 Name: Mike Rainey (http://www.cs.uchicago.edu/~mrainey)
60 :     Date: 2007/03/19
61 :     Description:
62 :    
63 :     Fixed bogus operand sizes in AMD64 instruction spilling.
64 :    
65 :     ----------------------------------------------------------------------
66 : jhr 2358 Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
67 :     Date: 2007/03/19
68 :     Description:
69 :    
70 :     Fixed a number of inconsistencies between the Posix.TTY structure
71 :     and the Basis specification. Thanks to Adam Chilpala.
72 :    
73 :     ----------------------------------------------------------------------
74 : mrainey 2345 Name: Mike Rainey (http://www.cs.uchicago.edu/~mrainey)
75 :     Date: 2007/02/26
76 :     Description:
77 :    
78 :     Added preliminary support in MLRISC for Staged Allocation, a technique
79 :     for specifying calling conventions (see
80 :     http://www.eecs.harvard.edu/~nr/pubs/staged-abstract.html).
81 :     Initially, we plan to use this code to generate C calls for the AMD64.
82 :    
83 :     The staged allocation code base resides in MLRISC/staged-allocation,
84 :     and specialized calling conventions go in MLRISC/ARCH/staged-allocation.
85 :    
86 :     ----------------------------------------------------------------------
87 : jhr 2342 Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
88 :     Date: 2007/02/20
89 :     Description:
90 :    
91 :     Bug fix: when SaveCState was called with two values to save, a subsequent
92 :     GC could cause the RestoreCState to fail because the saved state had been
93 :     promoted to tagless pair.
94 :    
95 :     ----------------------------------------------------------------------
96 : leunga 2340 Name: Allen Leung (leunga (at) reservoir (dot) com)
97 :     Date: 2007/02/20 19:08:00 EST
98 :     Description:
99 :    
100 :     Bug triggered by:
101 :    
102 :     val a1 = Word8Array.array(a1, 0w0);
103 :     val _ = Word8Array.update(a1, 0, 0w128);
104 :    
105 :     x86MCEmitter crashed when the immediate operand to MOVB was outside
106 :     of the range -128 ... 127. I've changed the code so that the range
107 :     check is disabled. Only the low order 8 bits of the immediate operand
108 : jhr 2342 are now significant.
109 : leunga 2340
110 :    
111 :     ----------------------------------------------------------------------
112 : blume 2304 Name: Matthias Blume (http://www.tti-c.org/blume/)
113 : blume 2336 Date: 2007/02/15 09:20:00 CST
114 :     Description:
115 :    
116 :     Eliminated any mention of lexgen. (Lexgen was an early precursor
117 :     to ml-ulex.)
118 :    
119 :     You should update your admin directory, so the shell scripts for
120 :     maintaining your local copy of the repository reflect this change.
121 :    
122 :     ----------------------------------------------------------------------
123 :     Name: Matthias Blume (http://www.tti-c.org/blume/)
124 : blume 2333 Date: 2007/02/14 15:45:00 CST
125 :     Description:
126 :    
127 :     Fixed bug in CM's parallel make facility that failed to have
128 :     the master re-link modules after letting slaves compile them.
129 :    
130 :     ----------------------------------------------------------------------
131 :     Name: Matthias Blume (http://www.tti-c.org/blume/)
132 : blume 2331 Date: 2007/02/12 15:10:00 CST
133 :     Description:
134 :    
135 :     Fixed typo in ml-build script that prevented library anchors
136 :     from being registered.
137 :    
138 :     ----------------------------------------------------------------------
139 :     Name: Matthias Blume (http://www.tti-c.org/blume/)
140 : blume 2304 Date: 2007/02/02 15:20:00 CST
141 :     Tag: release-110.62
142 :     Description:
143 :    
144 :     New working release (110.62). NEW BOOTFILES!
145 :    
146 :     ----------------------------------------------------------------------
147 : adrassi 2296 Name: Aaron Turon (http://www.cs.uchicago.edu/~adrassi)
148 :     Date: 2007/01/31
149 :     Description:
150 :    
151 :     Brought ml-lpt manual up-to-date with the code. Many minor
152 :     improvements to the tools, and a few minor bugfixes. Prepared
153 :     ml-lpt-lib for integration with new UTF8 structure. Changed
154 :     the interface for creating streams in ml-ulex (we now support
155 :     stream creation from several kinds of sources). Overall, the
156 :     tools are now quite stable, and their interfaces are unlikely to
157 :     change in a way that would break compatibility. More work,
158 :     however, is needed in the documentation.
159 :    
160 :     ----------------------------------------------------------------------
161 : blume 2277 Name: Matthias Blume (http://www.tti-c.org/blume/)
162 : blume 2298 Date: 2007/02/01 10:05:00 CST
163 :     Description:
164 :    
165 :     Implemented library installer. Moved CM plugin code for
166 :     ml-burg, ml-lex, and ml-yacc out of CM source tree and into
167 :     their respective trees. Implemented CM plugin for ml-ulex
168 :     and ml-antlr. Used library installer for ml-burg, ml-ulex,
169 :     and ml-antlr.
170 :    
171 :     For ml-yacc and ml-lex we continue to have permanently "plugged-in"
172 :     CM tools. (It turns out to be too messy to do otherwise because
173 :     there is too much code that during installation relies on the presence
174 :     of these tools -- resulting in a tricky ordering problem.)
175 :    
176 :     ----------------------------------------------------------------------
177 :     Name: Matthias Blume (http://www.tti-c.org/blume/)
178 : blume 2277 Date: 2007/01/30 11:05:00 CST
179 :     Description:
180 :    
181 :     Added SMLofNJ.shiftArgs which is like a shell's "shift" command.
182 :     Modified CM's startup code to use shiftArgs as it processes
183 :     command line arguments. This way, the init code in each .sml-file
184 :     or library that is mentioned at the top level will see only
185 :     those arguments that have not yet been processed at this point.
186 :     In other words, the init code can "seize control" and process the
187 :     remaining command line.
188 :    
189 :     ----------------------------------------------------------------------
190 : jhr 2269 Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
191 : jhr 2358 Date: 2007/01/30
192 :     Description:
193 :    
194 :     Added fromList function to the ORD_SET interface and lookup to the
195 :     ORD_MAP interface. See the SML/NJ Library CHANGES file for details.
196 :    
197 :     ----------------------------------------------------------------------
198 :     Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
199 : jhr 2269 Date: 2007/01/28
200 :     Description:
201 :    
202 :     Added the UTF8 structure and signature from the Moby compiler to
203 :     the SML/NJ library (and the CML library). These modules will replace
204 :     the version in the ml-lpt-lib.
205 :    
206 :     ----------------------------------------------------------------------
207 : blume 2237 Name: Matthias Blume (http://www.tti-c.org/blume/)
208 : blume 2267 Date: 2007/01/26 15:10:00 CST
209 :     Description:
210 :    
211 :     Added entries to handle ml-lpt-lib.cm in installer.
212 :    
213 :     ----------------------------------------------------------------------
214 :     Name: Matthias Blume (http://www.tti-c.org/blume/)
215 : blume 2259 Date: 2006/12/15 11:10:00 CST
216 : blume 2237 Tag: release-110.61
217 :     Description:
218 :    
219 :     New working release (110.61). NEW BOOTFILES!
220 :    
221 : blume 2259 This supersedes the pre-brown-paper-bug release (see below).
222 :    
223 : blume 2237 ----------------------------------------------------------------------
224 : blume 2259 Name: Matthias Blume (http://www.tti-c.org/blume/)
225 :     Date: 2006/12/15 11:00:00 CST
226 :     Description:
227 :    
228 :     Fixed brown-paper-bag bug with CM's pathname handling, which made
229 :     installation fail under Win32.
230 :    
231 :     ----------------------------------------------------------------------
232 :     Name: Matthias Blume (http://www.tti-c.org/blume/)
233 :     Date: 2006/12/14 12:00:00 CST
234 :     Description:
235 :    
236 :     Obsolete!
237 :    
238 :     [New working release (110.61). NEW BOOTFILES!]
239 :    
240 :     ----------------------------------------------------------------------
241 : jhr 2234 Name: John Reppy (http://cs.uchicago.edu/~jhr)
242 :     Date: 2006/14/06 09:50:00 CST
243 :     Description:
244 :    
245 :     Fixed the code in runtime/c-libs/posix-tty/{tcgetattr,tcsetattr}.c to
246 :     get the c_cc termios data copied correctly. Also moved the allocation
247 :     of the string to avoid problems if it caused a GC.
248 :    
249 :     Thanks to Timothy Bourke for the bug report and fix.
250 :    
251 :     ----------------------------------------------------------------------
252 : blume 2232 Name: Matthias Blume (http://www.tti-c.org/blume)
253 :     Date: 2006/12/08 09:50:00 CST
254 :     Description:
255 :    
256 :     Added code to CM's "standard shell tool" implementation which
257 :     causes it to tolerate (with a warning) the situation where the shell
258 :     command fails (e.g., due to the shell command's non-existence)
259 :     as long as all target files exist.
260 :    
261 :     This makes it possible to, e.g., build ml-yacc from sources even if
262 :     svn checkout messed up the time stamps on files in such a way that
263 :     yacc.grm is younger than yacc.grm.sml or yacc.grm.sig. (Ml-yacc would
264 :     be needed to re-process yacc.grm, but obviously it might not yet be
265 :     available at that time.)
266 :    
267 :     ----------------------------------------------------------------------
268 : jhr 2228 Name: John Reppy (http://cs.uchicago.edu/~jhr)
269 :     Date: 2006/12/06 12:06:00 CST
270 :     Description:
271 :    
272 :     Fixed the types of recvVecFrom, recvVecFrom', recvVecFromNB, and
273 :     recvVecFromNB' in the SOCKET signature. This error is actually
274 :     in the SML Basis specification too.
275 :    
276 :     ----------------------------------------------------------------------
277 : mblume 1923 Name: Matthias Blume (http://www.tti-c.org/blume/)
278 : blume 2226 Date: 2006/12/05 22:40:00 CST
279 :     Description:
280 :    
281 :     CM now reports undefined anchors as errors and aborts execution
282 :     rather than silently pressing on using bogus values.
283 :    
284 :     ----------------------------------------------------------------------
285 :     Name: Matthias Blume (http://www.tti-c.org/blume/)
286 : blume 2224 Date: 2006/11/29 10:00:00 CST
287 :     Description:
288 :    
289 :     Use Say.vsay for printing the "[autoloading]" message, so
290 :     #set CM.Control.verbose false (or -Ccm.verbose=false) can be used
291 :     to suppress them.
292 :    
293 :     ----------------------------------------------------------------------
294 :     Name: Matthias Blume (http://www.tti-c.org/blume/)
295 : blume 2219 Date: 2006/11/10 11:00:00 CST
296 :     Description:
297 :    
298 :     Fixed bug in CM where "with:" specifications that affect compilation
299 :     (as opposed to parsing) were ignored.
300 :    
301 :     ----------------------------------------------------------------------
302 :     Name: Matthias Blume (http://www.tti-c.org/blume/)
303 : blume 2193 Date: 2006/11/09 14:30:00 CST
304 :     Tag: release-110.60
305 :     Description:
306 :    
307 :     New working release (110.60). NEW BOOTFILES!
308 :    
309 :     This is the first subversion-hosted release. There are also changes
310 :     to the directory layout. Here is a table:
311 :    
312 :     (All pathnames are relative to the SML/NJ "root" directory.)
313 :    
314 :     config -> config
315 :    
316 :     src/cm -> base/cm
317 :     src/compiler -> base/compiler
318 :     src/runtime -> base/runtime
319 :     src/system -> base/system
320 :    
321 :     src/cm/pgraph -> pgraph
322 :    
323 :     src/READMES -> base/READMES
324 :     src/$notefile -> base/NOTES/$notefile
325 :     for <notefile> in BOOT CVSNOTES CYGWININSTALL HISTORY INSTALL
326 :     MACOSXINSTALL WININSTALL
327 :    
328 :     src/smlnj-lib -> smlnj-lib
329 :     src/MLRISC -> MLRISC
330 :     ckit -> ckit
331 :     src/cml -> cml
332 :     src/eXene -> eXene
333 :     src/heap2asm -> heap2asm
334 :     src/lexgen -> lexgen
335 :     src/ml-burg -> ml-burg
336 :     src/ml-lex -> ml-lex
337 :     src/ml-yacc -> ml-yacc
338 :     src/ml-nlffi-lib -> nlffi/lib
339 :     src/ml-nlffigen -> nlffi/gen
340 :     src/smlnj-c -> smlnj-c
341 :     src/tools/TraceDebugProf -> trace-debug-profile
342 :    
343 :     In addition, there is also a new
344 :    
345 :     ml-lpt
346 :    
347 :     directory containing two new program generator tools: ml-ulex and
348 :     ml-antlr (a lexer generator that handles unicode and an ANTRL-inspired
349 :     LL(k) parser generator). These tools are currently "beta-quality"
350 :    
351 :     The latest versios of the sources can now be obtained anonymously
352 :     via subversion. For this, it is useful to first check out
353 :    
354 :     svn://smlnj-gforge.cs.uchicago.edu/smlnj/admin
355 :    
356 :     and put the resulting directory on your shell's PATH. This provides
357 :     access to three shell scripts: checkout-all.sh, stat-all.sh,
358 :     and refresh-all.sh.
359 :    
360 :     To create a freshly checked-out copy of the sources, do
361 :    
362 :     checkout-all.sh [dir]
363 :    
364 :     where dir is the optional SML/NJ root directory (default is ".").
365 :    
366 :     This creates the above directory layout. Each subdirectory of
367 :     the root is under individual subversion control. The stat-all.sh
368 :     and refresh-al.sh scripts apply "svn stat" or "svn update" to
369 :     each of these subtrees.
370 :    
371 :     ----------------------------------------------------------------------
372 :     Name: Matthias Blume (http://www.tti-c.org/blume/)
373 : blume 2155 Date: 2006/11/02 12:20:00 CST
374 :     Description:
375 :    
376 :     Reorganized directory layout.
377 :    
378 :     This is a temporary solution, more reorganization is to come.
379 :    
380 :     The basic idea is to have a number of toplevel trees, each corresponding
381 :     to a well-defined part of the overall system. Each part can be
382 :     maintained individually, even in separate source repositories,
383 :     although currently we still serve everything out of the main
384 :     smlnj-gforge tree.
385 :    
386 :     Installer and and scripts have been updated to reflect the new
387 :     layout. The installer (base/src/system/smlnj/installer) is now
388 :     "scriptable" to avoid burning too much knowledge about the
389 :     layout into SML source code. The main script used by the installer
390 :     is in config/actions.
391 :    
392 :     The main change is that many of the subdirectories of what used to
393 :     be known as "src" have moved to the toplevel. The "src" directory
394 :     itself has moved down into a subtree called "base". (We may eventually
395 :     get rid of the extra level of indirection represented by "src".)
396 :    
397 :     The layout is now as follows:
398 :    
399 :     toplevel tree name | default repository
400 :     | (using gf=svn://smlnj-gforge.cs.uchicago.edu/smlnj)
401 :     -------------------+----------------------------------------------------
402 :     config | $gf/config/trunk
403 :     |
404 :     base | $gf/sml/trunk
405 :     |
406 :     smlnj-lib | $gf/smlnj-lib/trunk
407 :     MLRISC | $gf/MLRISC/trunk
408 :     |
409 :     ml-yacc | $gf/ml-yacc/trunk
410 :     ml-lex | $gf/ml-lex/trunk
411 :     ml-burg | $gf/ml-burg/trunk
412 :     lexgen | $gf/lexgen/trunk
413 :     heap2asm | $gf/heap2asm/trunk
414 :     |
415 :     cml | $gf/cml/trunk
416 :     eXene | $gf/eXene/trunk
417 :     ckit | $gf/ckit/trunk
418 :     nlffi | $gf/nlffi/trunk
419 :     |
420 :     smlnj-c | $gf/smlnj-c/trunk
421 :    
422 :     In $gf/admin there are a few useful shell scripts for checking out
423 :     and maintaining the entire collection of trees:
424 :    
425 :     admin/checkout-all.sh [dir]
426 :     -- optionally creates dir and checks out all trees from their
427 :     default repositories; if dir is missing, checkout into .
428 :     admin/refresh-all.sh [dir]
429 :     -- looks at all trees (from the above list) in dir (default: .)
430 :     and runs "svn update" if the tree exists and is under
431 :     subversion control;
432 :     non-existing or non-subversion trees are skipped
433 :     admin/stat-all.sh [dir]
434 :     -- like refresh-all.sh, but runs "svn stat" instead
435 :    
436 :     ----------------------------------------------------------------------
437 :     Name: Matthias Blume (http://www.tti-c.org/blume/)
438 :     Date: 2006/10/05 11:30:00 CDT
439 : mblume 2066 Tag: blume-20061005-merge-amd64
440 :     Description:
441 :    
442 :     Merged code for AMD64 backend (Mike Rainey's work).
443 :     Everything is hooked up but untested.
444 :    
445 :     ----------------------------------------------------------------------
446 :     Name: Matthias Blume (http://www.tti-c.org/blume/)
447 : mblume 1931 Date: 2006/05/17 14:00:00 CST
448 :     Tag: Release_110_59
449 :     Description:
450 :    
451 : blume 2193 New working version (110.59). NEW BOOTFILES!
452 : mblume 1931
453 :     I am freezing 110.59. Changes other than the version-number increase:
454 :    
455 :     eXene:
456 :    
457 :     * committed changes to eXene from Alley Stoughton:
458 :     "fixed bugs in X authorization and resource handling, as well
459 :     as in the pile and viewport widgets"
460 :    
461 :     Runtime:
462 :     * fixed linking problem with NetBSD 3.x.
463 :    
464 :     Lexgen:
465 :     * lexgen tool handles non-ascii characters in 7-bit mode the same
466 :     way that ml-lex does
467 :    
468 :     * lexgen propagates exceptions the same way that ml-lex does
469 :    
470 :     CML:
471 :     * Fixed a bug in the SyncVar polling functions (iGetPoll,
472 :     mTakePoll, and mGetPoll) that could lead to livelock.
473 :    
474 :     ----------------------------------------------------------------------
475 :     Name: Matthias Blume (http://www.tti-c.org/blume/)
476 : mblume 1926 Date: 2006/05/12 15:20:00 CST
477 :     Tag: blume-20060515-makedepend
478 :     Description:
479 :    
480 :     Implemented ml-makedepend (i.e., CM.sources) in a better (more robust)
481 :     way. This should hopefully fix the ml-makedepend problem permanently.
482 :    
483 :     ----------------------------------------------------------------------
484 :     Name: Matthias Blume (http://www.tti-c.org/blume/)
485 : mblume 1925 Date: 2006/05/12 15:15:00 CST
486 :     Tag: blume-20060512-makedepend
487 :     Description:
488 :    
489 :     Fixed long-standing bug with ml-makedepend where it would output a
490 :     spurious dependency to a non-existing file. (This is a simple fix.
491 :     It might need further looking into.)
492 :    
493 :     ----------------------------------------------------------------------
494 :     Name: Matthias Blume (http://www.tti-c.org/blume/)
495 : mblume 1923 Date: 2006/04/20 10:30:00 CDT
496 :     Tag: blume-20060420-heap2exec
497 :     Description:
498 :    
499 :     Committed patches received from Johannes 5 Joemann (joemann@befree.free.de)
500 :     that enable heap2exec under Linux and FreeBSD.
501 :    
502 :     ----------------------------------------------------------------------
503 : jhr 1921 Name: John Reppy
504 :     Date: 2006/04/14
505 : blume 2155 Tag:
506 : jhr 1921 Description:
507 :    
508 :     MLRISC changes:
509 :     1) renamed GAS_PSEUDO_OPS to AS_PSEUDO_OPS and put it in its own file.
510 :     2) added support for NOTB and XORB operators in pseudo-op expressions
511 :     3) added DarwinPseudoOp functor that supports Darwin's assembler syntax.
512 :     4) added support for 64-bit integer literals
513 :    
514 :     ----------------------------------------------------------------------
515 : jhr 1922 Name: Matthias Blume (http://www.tti-c.org/blume/)
516 : mblume 1907 Date: yyyy/mm/dd
517 :     Tag: Release_110_58
518 :     Description:
519 :    
520 :     New working version (110.58). NEW BOOTFILES!
521 :    
522 :     ----------------------------------------------------------------------
523 : mblume 1896 Name: Matthias Blume (http://www.tti-c.org/blume/)
524 : mblume 1905 Date: 2006/03/01 22:15:00 CST
525 :     Tag: blume-20060301-lexgen
526 :     Description:
527 :    
528 :     Incorporated several bugfixes to lexgen. Compiler now compiles to
529 :     fixpoint when using lexgen instead of ml-lex.
530 :    
531 :     ----------------------------------------------------------------------
532 :     Name: Matthias Blume (http://www.tti-c.org/blume/)
533 : mblume 1902 Date: 2006/02/28 22:45:00 CST
534 :     Tag: blume-20060228-lexgen
535 :     Description:
536 :    
537 :     Removed ml-flex and added lexgen instead, using Aaron Turon's newly
538 :     provided tarball. The generated lexers still have problems.
539 :    
540 :     ----------------------------------------------------------------------
541 :     Name: Matthias Blume (http://www.tti-c.org/blume/)
542 : mblume 1898 Date: 2006/02/26 23:30:00 CST
543 :     Tag: blume-20060226-ml-flex
544 :     Description:
545 :    
546 :     Removed ml-flex' dependency on regexp-lib.cm. Turned installation
547 :     of ml-flex on by default.
548 :    
549 :     ----------------------------------------------------------------------
550 :     Name: Matthias Blume (http://www.tti-c.org/blume/)
551 : mblume 1896 Date: 2006/02/24 16:30:00 CST
552 :     Tag: blume-20060224-ml-flex
553 :     Description:
554 :    
555 :     Added ml-flex sources. Partially integrated, but should not be
556 :     turned on yet! (Read: leave it commented-out in config/targets!)
557 :    
558 :     ----------------------------------------------------------------------
559 : jhr 1893 Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
560 :     Date: 2006/02/23
561 :     Tag: jhr-20060223-x86-64
562 :     Description:
563 :     Changes to support building on x86-64 systems (using the 32-bit mode).
564 :     Also cleaned up signal handling on Linux. Support for pre-2.2 Linux
565 :     kernels dropped.
566 :    
567 :     ----------------------------------------------------------------------
568 : allenleung 1890 Name: Allen Leung (leunga (at) reservoir (dot) com)
569 :     Date: 2006/02/22 15:07:00 EST
570 :     Tag: leunga-20060222-mdl-gen
571 :     Description:
572 :    
573 :     CM has changed. Updated the script for rebuilding the MLRISC generated
574 :     files.
575 :    
576 :     ----------------------------------------------------------------------
577 : mblume 1886 Name: Matthias Blume (http://www.tti-c.org/blume/)
578 :     Date: 2006/02/14 17:30:00 CST
579 :     Tag: blume-20060214-intel-mac
580 :     Description:
581 :    
582 :     Hooked code for Darwin-specific Intel ABI into main compiler.
583 :     (This is a temporary solution which relies on the fact that the
584 :     compiler itself does not use NLFFI. Eventually we need to divorce
585 :     intel mac from generic x86 unix code and make separate sets of binaries.)
586 :    
587 :     ----------------------------------------------------------------------
588 : jhr 1882 Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
589 : jhr 1885 Date: 2006/02/13
590 :     Tag: jhr-20060213-intel-mac
591 :     Description:
592 :     Changed MLRISC x86 CCalls for partial support of MacOS X ABI.
593 :    
594 :     ----------------------------------------------------------------------
595 :     Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
596 : jhr 1882 Date: 2006/02/06
597 :     Tag: jhr-20060206-intel-mac
598 :     Description:
599 :     Changes to support MacOS X on Intel hardware. The C-calls support in
600 :     MLRISC must also be updated to support the MacOS X ABI.
601 :    
602 :     ----------------------------------------------------------------------
603 : mblume 1863 Name: Matthias Blume (http://www.tti-c.org/blume/)
604 : mblume 1879 Date: 2005/12/16 00:15:00 CST
605 :     Tag: blume-20051216-cm
606 :     Description:
607 :    
608 :     Improved error reporting and handling in CM.
609 :    
610 :     ----------------------------------------------------------------------
611 :     Name: Matthias Blume (http://www.tti-c.org/blume/)
612 : mblume 1878 Date: 2005/11/21 14:00:00 CST
613 :     Tag: blume-20051121-html-lib
614 :     Description:
615 :    
616 :     Enabling $/html-lib.cm in config/preloads did not work. This is fixed
617 :     now. (Since the anchor mapping for html-lib.cm is not yet in effect
618 : blume 2155 at the time when config/preloads is processed, the library has to
619 : mblume 1878 be referred to by another name. In 110.57 this would be
620 :     $SMLNJ-LIB/HTML/html-lib.cm. I arranged for $smlnj/smlnj-lib/html-lib.cm
621 :     to be valid as well.)
622 :    
623 :     Thanks to Todd Wilson (Fresno) for alerting me to this issue.
624 :    
625 :     ----------------------------------------------------------------------
626 :     Name: Matthias Blume (http://www.tti-c.org/blume/)
627 : mblume 1874 Date: 2005/11/19 13:45:00 CST
628 :     Tag: Release_110_57
629 :     Description:
630 :    
631 :     Moved release tag for 110.57.
632 :    
633 :     ----------------------------------------------------------------------
634 :     Name: Matthias Blume (http://www.tti-c.org/blume/)
635 : mblume 1873 Date: 2005/11/19 23:35:00 CST
636 :     Tag: blume-20051119-fish
637 :     Description:
638 :    
639 :     Fixed a problem in config/install.sh where it tries to "fish" the
640 :     name of the CM metadata directory from the wrong place (because
641 :     the physical location of basis.cm has changed). Also, corrected the
642 :     path anchor for $/html-lib.cm. (Thanks to M. Fluet for pointing out
643 :     these problems.)
644 :    
645 :     ----------------------------------------------------------------------
646 :     Name: Matthias Blume (http://www.tti-c.org/blume/)
647 : mblume 1867 Date: 2005/11/16 14:30:00 CST
648 : mblume 1874 Tag: blume-orig-110_57
649 : mblume 1867 Description:
650 :    
651 :     New working version (110.57). NEW BOOTFILES!
652 :    
653 : mblume 1874 UPDATE: The release tag has been moved to include a fix for a brown-paper-
654 :     bag glitch (see above).
655 :    
656 : mblume 1867 ----------------------------------------------------------------------
657 :     Name: Matthias Blume (http://www.tti-c.org/blume/)
658 : mblume 1866 Date: 2005/11/16 10:25:00 CST
659 :     Tag: blume-22051116-btrace
660 :     Description:
661 :    
662 :     Fixed problem with bogus exception message when using back-trace
663 :     facility.
664 :    
665 :     ----------------------------------------------------------------------
666 :     Name: Matthias Blume (http://www.tti-c.org/blume/)
667 : mblume 1865 Date: 2005/11/15 22:35:00 CST
668 :     Tag: blume-22051115-array2copy
669 :     Description:
670 :    
671 :     Added simple implementation of Array2.copy. (Warning: mostly untested.)
672 :    
673 :     ----------------------------------------------------------------------
674 :     Name: Matthias Blume (http://www.tti-c.org/blume/)
675 : mblume 1864 Date: 2005/11/15 21:35:00 CST
676 :     Tag: blume-22051115-versiontool
677 :     Description:
678 :    
679 :     Reversed change to src/system/smlnj/internals/versiontool.cm. This
680 :     file gets loaded as a tool -- by the equivalent of CM.make during the
681 :     run of CMB.make. Thus, CMB's path configuration is meaningless for
682 :     it. Instead, it has the status of "user code", so it should use
683 :     $/basis.cm to refer to the Basis library. (At least that's true for
684 :     the purpose of bootstrapping the previous change. In the future it
685 :     might make sense to have versiontool.cm refer to
686 :     $smlnj/basis/basis.cm, i.e., the version of the Basis that the
687 :     compiler itself uses.)
688 :    
689 :     Also patched src/system/testml to have it activate those extra anchor
690 :     bindings in config/extrapathconfig.
691 :    
692 :     ----------------------------------------------------------------------
693 :     Name: Matthias Blume (http://www.tti-c.org/blume/)
694 : mblume 1863 Date: 2005/11/15 16:35:00 CST
695 :     Tag: blume-22051115-libraries
696 :     Description:
697 :    
698 :     This change affects the way the following libraries are tied into
699 :     the system:
700 :    
701 :     $/basis.cm
702 :     $/smlnj-lib.cm
703 :     $/pp-lib.cm
704 :     $/controls-lib.cm
705 :     $/html-lib.cm
706 :     $/ml-yacc-lib.cm
707 :    
708 :     These libraries are now internally (as seen from the source code
709 :     of the implementation itself) known by the following names:
710 :    
711 :     $smlnj/basis/basis.cm
712 :     $smlnj/smlnj-lib/smlnj-lib.cm
713 :     $smlnj/smlnj-lib/pp-lib.cm
714 :     $smlnj/smlnj-lib/controls-lib.cm
715 :     $smlnj/smlnj-lib/html-lib.cm
716 :     $smlnj/ml-yacc/ml-yacc-lib.cm
717 :    
718 :     This makes it possible to work with code that requires different
719 :     versions of these libraries, and which refers to these libraries using
720 :     their "default" names (i.e., the first set of names above). In other
721 :     words, one can un-define or re-define those default names without
722 :     compromising the proper functioning of the compiler itself.
723 :    
724 :     A similar procedure had already been performed for several of the
725 :     MLRISC libraries that are linked into the compiler. I did some
726 :     cleanup on this code.
727 :    
728 :     A new file in the config directory (named extrapathconfig) is
729 :     responsible for setting up path anchors that the compiler itself does
730 :     not need, but that are typically required by user code.
731 :    
732 :     ----------------------------------------------------------------------
733 : mblume 1862 Name: Matthias Blume
734 :     Date: 2005/11/07 17:05:00 CST
735 :     Tag: blume-20051107-slice
736 :     Description:
737 :    
738 :     Fixed erroneous out-of-bounds test in the "update" function of
739 :     various *ArraySlice modules. (Thanks to Vesa A. Norrman for pointing
740 :     out the problem.)
741 :    
742 :     Pushed some Basis changes through ML-Lex, CML, and eXene.
743 :    
744 :     ----------------------------------------------------------------------
745 : jhr 1858 Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
746 : jhr 1861 Date: 2005/11/07
747 :     Description:
748 :     Fixed a Basis incompatibility: The depreciated function Substring.all
749 :     was removed (use Substring.full instead).
750 :    
751 :     ----------------------------------------------------------------------
752 :     Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
753 : jhr 1858 Date: 2005/11/05
754 :     Description:
755 :     Tweaked interval set API in SML/NJ library; see the CHANGES file for details.
756 :    
757 :     ----------------------------------------------------------------------
758 : mblume 1856 Name: Matthias Blume (http://www.tti-c.org/blume/)
759 :     Date: 2005/11/02 16:25:00 CST
760 :     Tag: blume-20051102-hex
761 :     Description:
762 :    
763 :     * Runtime system bootstrap code now accepts hex digits in BOOTLIST
764 :     in either upper case or lower case format.
765 :     * Pushed changes to names of Pack<N>{Big,Little} structures through
766 :     CML and eXene.
767 :    
768 :     ----------------------------------------------------------------------
769 : jhr 1852 Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
770 : jhr 1855 Date: 2005/11/02
771 :     Description:
772 :     Fixed a Basis incompatibility: Pack<N>{Big,Little} structures should
773 :     be called PackWord<N>{Big,Little}.
774 :    
775 :     ----------------------------------------------------------------------
776 :     Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
777 : jhr 1852 Date: 2005/10/28
778 :     Description:
779 :     Fixed a minor Basis incompatibility: hex digits should be upper case.
780 :    
781 :     ----------------------------------------------------------------------
782 : mblume 1846 Name: Matthias Blume (blume (at) tti - c (dot) org)
783 :     Date: 2005/10/25 16:00:00 CDT
784 :     Tag: Release_110_56
785 :     Description:
786 :    
787 :     New working version (110.56). NEW BOOTFILES!
788 :    
789 :     ----------------------------------------------------------------------
790 : jhr 1838 Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
791 : jhr 1840 Date: 2005/10/25
792 :     Description:
793 :     Added interval sets to utility library (signatures INTERVAL_DOMAIN
794 :     and INTERVAL_SET, and functor IntervalSetFn).
795 :    
796 :     ----------------------------------------------------------------------
797 :     Name: John Reppy (http://www.cs.uchicago.edu/~jhr)
798 : jhr 1838 Date: 2005/10/14
799 :     Description:
800 :     Add Zhong Shao's fix for datatype equality functions.
801 :    
802 :     ----------------------------------------------------------------------
803 : allenleung 1836 Name: Allen Leung (leunga (at) reservoir (dot) com)
804 : allenleung 1837 Date: 2005/10/14 12:30:00 EST
805 :     Tag: leunga-20051014-x86-spill
806 :     Description:
807 :    
808 :     Bug fix a bug found by Carl Hauser.
809 :     There was a typo in the reload code for FCMP in x86SpillInstr.sml.
810 :    
811 :     ----------------------------------------------------------------------
812 :     Name: Allen Leung (leunga (at) reservoir (dot) com)
813 : allenleung 1836 Date: 2005/10/14 11:15:00 EST
814 :     Tag: leunga-20051014-x86-asm
815 :     Description:
816 :    
817 :     Removed some debugging code in file x86Asm.sml.
818 :     The function emit_operand was printing out debugging output.
819 :    
820 :     ----------------------------------------------------------------------
821 : jhr 1813 Name: John Reppy
822 : jhr 1831 Date: 2005/07/27
823 :     Tag:
824 :     Description:
825 :    
826 : jhr 1832 Fixed ml-lex to recognize "\r" as representing carriage return.
827 :    
828 :     ----------------------------------------------------------------------
829 :     Name: John Reppy
830 :     Date: 2005/07/27
831 :     Tag:
832 :     Description:
833 :    
834 : jhr 1831 Fixed ml-yacc to work on files with non-native end-of-line
835 :     encodings (e.g., Windows text file on a Unix system).
836 :    
837 :     ----------------------------------------------------------------------
838 :     Name: John Reppy
839 : jhr 1824 Date: 2005/07/20
840 :     Tag:
841 :     Description:
842 :    
843 :     Added changes from Dominic Evans (oldmanuk (at) gmail (dot) com)
844 :     to support HPUX 11.
845 :    
846 :     ----------------------------------------------------------------------
847 :     Name: John Reppy
848 : jhr 1813 Date: 2005/07/06
849 :     Tag:
850 :     Description:
851 :    
852 :     Changes to the SML/NJ library. See smlnj-lib/CHANGES for details.
853 :    
854 :     ----------------------------------------------------------------------
855 : mblume 1808 Name: Matthias Blume (blume (at) tti - c (dot) org)
856 :     Date: 2005/07/06 00:45:00 CDT
857 :     Tag: blume-20050706-slice-copy
858 :     Description:
859 :    
860 :     Fixed reversed logic for deciding whether to "copy up" or "copy down"
861 :     in *-array-slice.sml.
862 :    
863 :     ----------------------------------------------------------------------
864 : allenleung 1806 Name: Allen Leung (leunga (at) reservoir (dot) com)
865 : allenleung 1807 Date: 2005/05/31 17:00:00 EST
866 :     Tag: leunga-20050531-cygwin-fault-2
867 :     Description:
868 :    
869 :     A typo in the cygwin code fixed.
870 :    
871 :     ----------------------------------------------------------------------
872 :     Name: Allen Leung (leunga (at) reservoir (dot) com)
873 : allenleung 1806 Date: 2005/05/31 16:47:00 EST
874 :     Tag: leunga-20050531-cygwin-fault
875 :     Description:
876 :    
877 : blume 2155 Updated Cygwin's fault/signal handling to match the Windows version.
878 : allenleung 1806 Updated the export list.
879 :    
880 :     ----------------------------------------------------------------------
881 : mblume 1791 Name: Matthias Blume (blume (at) tti - c (dot) org)
882 : mblume 1795 Date: 2005/05/18 13:35:00 CDT
883 :     Tag: Release_110_54
884 :     Description:
885 :    
886 :     New working version (110.54). NEW BOOTFILES!
887 :    
888 :     ----------------------------------------------------------------------
889 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
890 : mblume 1794 Date: 2005/05/18 11:58:00 CDT
891 :     Tag: blume-20050518-installer
892 :     Description:
893 :    
894 :     Added support scripts for Mac OS X PackageMaker and modified
895 :     config/install.sh so that it supports re-dumping a heap image
896 :     after customization.
897 :    
898 :     ----------------------------------------------------------------------
899 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
900 : mblume 1793 Date: 2005/05/18 10:55:00 CDT
901 :     Tag: blume-20050518-realdiv-noovld
902 :     Description:
903 :    
904 :     Un-overloaded / to work around bug in overloading resolution code.
905 :    
906 :     ----------------------------------------------------------------------
907 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
908 : mblume 1792 Date: 2005/05/16 23:50:00 CDT
909 :     Tag: blume-20050516-redump-heap
910 :     Description:
911 :    
912 :     Added mechanism for re-creating a heap file for the interactive system
913 :     after configuration variables have been changed.
914 :    
915 :     CM.redump_heap : string -> unit
916 :    
917 :     This is much like SMLofNJ.exportML, but starting from the resulting
918 : blume 2155 heap does not return to the caller of CM.redump_heap but
919 : mblume 1792 restarts the interactive system from scratch. The original call of
920 :     CM.redump_heap does not return but ends the interactive session.
921 :     Thus, CM.redump_heap is a lot like SMLofNJ.exportFn.
922 :    
923 :     Internally, redump_heap winds the dynamic execution context back to
924 :     the point where the original heap image was created and re-executes
925 :     the heap image generation code in the boot code.
926 :    
927 :     ----------------------------------------------------------------------
928 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
929 : mblume 1791 Date: 2005/05/09 21:55:00 CDT
930 :     Tag: blume-20050509-word64
931 :     Description:
932 :    
933 :     Added a hack to the existing hack known as Word64 to make fromString
934 :     behave correctly. I am still not sure whether Word64.scan will work
935 :     as specified with respect to the interaction of radix and prefix.
936 :    
937 :     ----------------------------------------------------------------------
938 : allenleung 1785 Name: Allen Leung (leunga (at) reservoir (dot) com)
939 :     Date: 2005/05/04 11:50:00 EST
940 :     Tag: leunga-20050504-checkgc
941 :     Description:
942 :    
943 : blume 2155 Added a gc protocol checking phase. This phase is enabled with
944 : allenleung 1785 the flag "check-gc". "debug-check-gc" turns on the verbose mode.
945 :    
946 :     ----------------------------------------------------------------------
947 : mblume 1741 Name: Matthias Blume (blume (at) tti - c (dot) org)
948 : mblume 1784 Date: 2005/05/04 10:45:00 CDT
949 :     Tag: blume-20050504-intinf
950 :     Description:
951 :    
952 :     Fixed a bug in the implementation of div and mod for IntInf.
953 :     Thanks to Neophytos Michael for reporting the problem.
954 :    
955 :     ----------------------------------------------------------------------
956 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
957 : mblume 1783 Date: 2005/05/04 10:35:00 CDT
958 :     Tag: blume-20050504-join
959 :     Description:
960 :    
961 :     Added a "join" combinator to the ParserComb module in smlnj-lib.cm.
962 :    
963 :     ----------------------------------------------------------------------
964 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
965 : mblume 1770 Date: 2005/02/28 23:40:00 CST
966 :     Tag: blume-20050228-mVar
967 :     Description:
968 :    
969 :     Fixed serious bug (brown paper bag variety) in new implementation of
970 :     structure Atom in CML. (I had accidentally used a mailbox instead of
971 :     an mvar, leaving the door open for races.)
972 :    
973 :     ----------------------------------------------------------------------
974 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
975 : mblume 1761 Date: 2005/02/25 15:00:00 CST
976 :     Tag: Release_110_53
977 :     Description:
978 :    
979 :     New working version (110.53). NEW BOOTFILES!
980 :    
981 :     ----------------------------------------------------------------------
982 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
983 : mblume 1760 Date: 2005/02/25 14:50:00 CST
984 :     Tag: blume-20050225-susp
985 :     Description:
986 :    
987 :     Brought back SMLofNJ.Susp. The underlying suspension type is the one
988 :     implemented in Core -- which means that it is the same as the one used
989 :     by the lazy extension.
990 :    
991 :     ----------------------------------------------------------------------
992 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
993 : mblume 1759 Date: 2005/02/24 16:50:00 CST
994 :     Tag: blume-20050224-cml-atom
995 :     Description:
996 :    
997 :     Simpler and at the same time more general implementation of structure
998 :     Atom in CML.
999 :    
1000 :     ----------------------------------------------------------------------
1001 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1002 : mblume 1758 Date: 2005/02/15 17:35:00 CST
1003 :     Tag: blume-20050215-tools
1004 :     Description:
1005 :    
1006 :     Created new "tools" directory under "src" and moved "TraceDebugProf"
1007 :     there.
1008 :    
1009 :     ----------------------------------------------------------------------
1010 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1011 : mblume 1755 Date: 2005/02/10 17:55:00 CST
1012 :     Tag: blume-20050210-longlong
1013 :     Description:
1014 :    
1015 :     Implemented "long long" arguments and results for NLFFI. (Only the
1016 :     PPC/MacOS implementation is complete, the other backends still need to
1017 :     be updated.)
1018 :    
1019 :     ----------------------------------------------------------------------
1020 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1021 : mblume 1751 Date: 2005/01/24 17:40:00 CST
1022 :     Tag: blume-20050124-mlyacc
1023 :     Description:
1024 :    
1025 :     Minor cleanup in ML-Yacc rule printing mechanism. This should fix a
1026 :     problem with certain "as" patterns which previously got rendered
1027 :     using incorrect syntax.
1028 :    
1029 :     ----------------------------------------------------------------------
1030 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1031 : mblume 1750 Date: 2005/01/18 12:00:00 CST
1032 :     Tag: blume-20050118-profile
1033 :     Description:
1034 :    
1035 :     Made time profiling code (interrupt handler) in runtime system aware
1036 :     of new array representation.
1037 :    
1038 :     ----------------------------------------------------------------------
1039 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1040 : mblume 1747 Date: 2005/01/14 18:00:00 CST
1041 :     Tag: blume-20050114-heap2exec
1042 :     Description:
1043 :    
1044 :     Implemented new (but still experimental) heap2exec facility. This is
1045 :     tested under Mac OS X and should work under Linux (will test shortly).
1046 :     It will probably also work on the Sparc (will test some time later).
1047 :    
1048 :     - removed old "HACKED_STANDALONE" hack from runtime
1049 :    
1050 :     To be able to test this, uncomment the request for "heap2asm" in
1051 :     config/targets prior to installation. (Notice that this is different
1052 :     from "heap2exec" mentioned below. Not a typo.)
1053 :    
1054 :     To perform an actual test, run the command
1055 :    
1056 :     $ bin/heap2exec heapfile execfile
1057 :    
1058 :     (You can put heap2exec on your shell's path.)
1059 :    
1060 :     For example, run
1061 :    
1062 :     $ bin/heap2exec bin/.heap/ml-yacc.ppc-darwin mly
1063 :    
1064 :     This will create a standalone executable called "mly" which you can
1065 :     then invoke directly as a command.
1066 :    
1067 :     ----------------------------------------------------------------------
1068 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1069 : mblume 1746 Date: 2005/01/07 11:44:00 CST
1070 :     Tag: blume-20050107-mlstring
1071 :     Description:
1072 :    
1073 :     fixed off-by-one error in ML_STRING macro (globals.c)
1074 :    
1075 :     ----------------------------------------------------------------------
1076 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1077 : mblume 1744 Date: 2004/12/23 18:00:00 CST
1078 :     Tag: blume-20041223-santa
1079 :     Description:
1080 :    
1081 :     Made ml-build script "smarter" (but only very little).
1082 :    
1083 :     ----------------------------------------------------------------------
1084 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1085 : mblume 1742 Date: 2004/12/21 15:05:00 CST
1086 :     Tag: blume-20041221-longlong
1087 :     Description:
1088 :    
1089 :     * Implemented access to signed and unsigned long long data in NLFFI.
1090 :     (The parameter-passing part of the picture has not complete. But
1091 :     data structure access seems to work.)
1092 :    
1093 :     * Fixed CM's incorrect assumption that the PPC is little-endian.
1094 :     (On the Mac, it is big-endian. And that's currently our only
1095 :     PPC platform.)
1096 :    
1097 :     ----------------------------------------------------------------------
1098 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1099 : mblume 1741 Date: 2004/12/21 12:50:00 CST
1100 :     Tag: blume-20041221-memory
1101 :     Description:
1102 :    
1103 :     Some cleanup in the $c/memory.cm library: separated some concerns by
1104 :     moving allocation code and memory access code each into their own
1105 :     files.
1106 :    
1107 :     ----------------------------------------------------------------------
1108 : allenleung 1740 Name: Allen Leung (leunga (at) reservoir (dot) com)
1109 :     Date: 2004/12/17 16:12:00 EST
1110 :     Tag: leunga-20041217-cygwin-smlnj-home
1111 :     Description:
1112 :    
1113 : blume 2155 The Unix I/O library of SML/NJ on cygwin does not understand
1114 :     Windows style pathname, so problems arise when SMLNJ_HOME is set
1115 :     to a Windows style pathname. _run-sml now converts SMLNJ_HOME
1116 :     to a POSIX pathname on cygwin.
1117 : allenleung 1740
1118 :     ----------------------------------------------------------------------
1119 : mblume 1737 Name: Matthias Blume (blume (at) tti - c (dot) org)
1120 : mblume 1736 Date: 2004/12/16 13:35:00 CST
1121 :     Tag: Release_110_52
1122 :     Description:
1123 :    
1124 :     Last-minute changes incorporated into 110.52. Release tag moved.
1125 :    
1126 :     The changes:
1127 :     - HashString.hashString' -> HashString.hashSubstring
1128 :     - bug fix in UnivariateStats
1129 :    
1130 :     ----------------------------------------------------------------------
1131 : mblume 1702 Name: Matthias Blume (blume (at) tti - c (dot) org)
1132 : mblume 1733 Date: 2004/12/15 23:40:00 CST
1133 :     Tag: blume-20041215-hashSubstring
1134 :     Description:
1135 :    
1136 :     - HashString.hashString' -> HashString.hashSubstring
1137 :     - corresponding changes in atom.sml
1138 :     - "de-compressed" (aka. un-obfuscated) code for UnivariateStats and
1139 :     added some comments
1140 :    
1141 :     ----------------------------------------------------------------------
1142 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1143 : mblume 1728 Date: 2004/12/15 15:30:00 CST
1144 : mblume 1736 Tag: (Release_110_52)
1145 : mblume 1728 Description:
1146 :    
1147 :     New working version (110.52). NEW BOOTFILES!
1148 :    
1149 :     ----------------------------------------------------------------------
1150 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1151 : mblume 1727 Date: 2004/12/15 12:45:00 CST
1152 :     Tag: blume-20041215-spaces
1153 :     Description:
1154 :    
1155 :     More on the space problem (this time for Win32).
1156 :    
1157 :     ----------------------------------------------------------------------
1158 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1159 : mblume 1725 Date: 2004/12/14 17:30:00 CST
1160 :     Tag: blume-20041214-spaces
1161 :     Description:
1162 :    
1163 :     Hacked some of the scripts (in particular: the installer) to cope with
1164 :     spaces in filenames a bit better. But beware: the current "solution"
1165 :     is likely still full of bugs and inherently incomplete. (We need to
1166 :     do away with those shell scripts for a comprehensive solution.)
1167 :    
1168 :     ----------------------------------------------------------------------
1169 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1170 : mblume 1724 Date: 2004/12/13 14:45:00 CST
1171 :     Tag: blume-20041213-ml-makedepend
1172 :     Description:
1173 :    
1174 :     Fixed bug in code for ml-makedepend.
1175 :    
1176 :     ----------------------------------------------------------------------
1177 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1178 : mblume 1720 Date: 2004/12/09 16:30:00 CST
1179 :     Tag: blume-20041209-statistics
1180 :     Description:
1181 :    
1182 :     Added two simple but potentially useful statistics modules to SML/NJ Library.
1183 :     (See CHANGES file there.)
1184 :    
1185 :     ----------------------------------------------------------------------
1186 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1187 : mblume 1718 Date: 2004/12/01 16:50:00 CST
1188 :     Tag: blume-20041201-atom
1189 :     Description:
1190 :    
1191 :     smlnj-lib:
1192 :    
1193 :     Added function HashString.hashString' for substrings.
1194 :     Hand-inlined CharVector.foldl into HashString (for speed).
1195 :     Modified implementation of structure Atom to avoid extracting
1196 :     strings from substrings unless necessary.
1197 :     (Also see CHANGES file for smlnj-lib.)
1198 :    
1199 :     ----------------------------------------------------------------------
1200 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1201 : mblume 1717 Date: 2004/11/24 22:15:00 CST
1202 :     Tag: blume-20041124-cml
1203 :     Description:
1204 :    
1205 :     Made sure CML compiles when Position = Int64.
1206 :    
1207 :     ----------------------------------------------------------------------
1208 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1209 : mblume 1715 Date: 2004/11/24 14:30:00 CST
1210 :     Tag: blume-20041124-position
1211 :     Description:
1212 :    
1213 :     The compiler can now be compiled in a mode that makes structure Position
1214 :     equal to Int64. The default, however, is unchanged (Position = Int31)
1215 :     for the time being.
1216 :    
1217 :     To enable 64-bit positions, use the following procedure:
1218 :    
1219 :     1. Start sml
1220 :     2. Autoload $smlnj/cmb.cm (if not already autoloaded)
1221 :     3. Type
1222 :     #set (CMB.symval "USE_64_BIT_POSITIONS") (SOME 1);
1223 :     4. Run CMB.make() as usual.
1224 :    
1225 :     This is barely tested. The only test so far was a little SML program
1226 :     counting the number of characters in an 8-gigabyte file by
1227 :     reading it character-by-character. That test was successful.
1228 :    
1229 :     In support of 64-bit positions, a number of new functions have been
1230 :     added to the runtime system.
1231 :    
1232 :     ----------------------------------------------------------------------
1233 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1234 : mblume 1716 Date: 2004/11/23 14:45:00 CST
1235 :     Tag: blume-20041123-useFile
1236 : mblume 1712 Description:
1237 :    
1238 :     Fixed a problem with unhelpful error messages related to problems with
1239 :     .cm- or .sml files that appear as part of the sml command line.
1240 :    
1241 :     ----------------------------------------------------------------------
1242 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1243 : mblume 1703 Date: 2004/11/18 15:40:00 CST
1244 :     Tag: Release_110_51
1245 :     Description:
1246 :    
1247 :     New working version (110.51). NEW BOOTFILES!
1248 :    
1249 :     ----------------------------------------------------------------------
1250 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1251 : mblume 1702 Date: 2004/11/18 15:35:00 CST
1252 :     Tag: <none>
1253 :     Description:
1254 :    
1255 :     Enabled dlopen and friends for FreeBSD (as recommended by Johannes 5
1256 :     Joemann).
1257 :    
1258 :     ----------------------------------------------------------------------
1259 : allenleung 1701 Name: Allen Leung (leunga (at) reservoir (dot) com)
1260 :     Date: 2004/11/17 16:05:21 EST 2004
1261 :     Tag: leunga-20041117-mlrisc-live-kill
1262 :     Description:
1263 :    
1264 : blume 2155 Added support for MLTree constructs LIVE and KILL
1265 : allenleung 1701 to all the architectures.
1266 :    
1267 :     ----------------------------------------------------------------------
1268 : mblume 1687 Name: Matthias Blume (blume (at) tti - c (dot) org)
1269 : mblume 1689 Date: 2004/11/13 00:20:00 CST
1270 :     Tag: blume-20041113-versiontool
1271 :     Description:
1272 :    
1273 :     - Stripped down the versiontool: It now only handles the version number.
1274 :     The date string is generated at bootstrap time (during makeml).
1275 :    
1276 :     - In a previous commit, fixed a minor issue with how polyequal is being
1277 :     translated. In particular, the code now "looks through" abstractions.
1278 :     This results in slightly fewer polyEqual warnings and hopefully slightly
1279 :     more efficient code. Important examples for where this matters are
1280 :     the new int64 and word64 types.
1281 :    
1282 :     ----------------------------------------------------------------------
1283 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1284 : mblume 1687 Date: 2004/11/12 00:30:00 CST
1285 :     Tag: blume-20041112-int64
1286 :     Description:
1287 :    
1288 :     Structure Int64 fully hooked in. (The implementation is not very
1289 :     efficient, though.)
1290 :    
1291 :     ----------------------------------------------------------------------
1292 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1293 : mblume 1686 Date: 2004/11/11 17:30:00 CST
1294 :     Tag: blume-20041111-more64
1295 :     Description:
1296 :    
1297 :     All the pieces of Word64 are now there, with the exception of the
1298 :     conversions from and to LargeWord. (Eventually these need to be identities,
1299 :     but for the time being they don't even make sense because LargeWord is
1300 :     32-bit wide.)
1301 :    
1302 :     Also started to add similar support for Int64, but major pieces of that
1303 :     are still missing.
1304 :    
1305 :     ----------------------------------------------------------------------
1306 : mblume 1668 Name: Matthias Blume (blume (at) tti - c (dot) org)
1307 : mblume 1684 Date: 2004/11/11 00:15:00 CST
1308 :     Tag: blume-20041111-word64
1309 :     Description:
1310 :    
1311 :     Structure Word64 is now (almost) complete, word literals and patterns
1312 :     seem to work. There are a few odd pieces missing. In particular,
1313 :     I didn't do the {from,to}LargeWord parts because LargeWord is still Word32
1314 :     at the moment.
1315 :    
1316 :     Making Word64 official would mean that LargeWord becomes Word64. But
1317 :     this requires extreme care because most word-word conversions have to
1318 :     go through LargeWord, so making a mistake means loss of efficiency or
1319 :     worse. Eventually there will be a solution similar to (but actually
1320 :     simpler than) what I did with IntInf.
1321 :    
1322 :     ----------------------------------------------------------------------
1323 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1324 : mblume 1683 Date: 2004/11/10 18:12:00 CST
1325 :     Tag: blume-20041110-64bit
1326 :     Description:
1327 :    
1328 :     More 64-bit hacking (but still not even half-way there yet).
1329 :     Also, some assorted improvements to the handling of 8-bit words.
1330 :    
1331 :     ----------------------------------------------------------------------
1332 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1333 : mblume 1682 Date: 2004/11/09 17:50:00 CST
1334 : mblume 1683 Tag: <none>
1335 : mblume 1682 Description:
1336 :    
1337 :     Started adding some infrastructure for supporting 64-bit int- and
1338 :     word-types. (Still in its very early stages.)
1339 :    
1340 :     ----------------------------------------------------------------------
1341 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1342 : mblume 1672 Date: 2004/10/28 10:45:00 CDT
1343 :     Tag: Release_110_50
1344 :     Description:
1345 :    
1346 :     New working version (110.50). NEW BOOTFILES!
1347 :    
1348 :     =====================
1349 :    
1350 :     Also:
1351 :    
1352 :     - Changend config/srcarchiveurl from a file just
1353 :     containing the URL string into a file containing
1354 :     shell script code. The code has access to the $VERSION variable.
1355 :     - Made corresponding changes to config/install.sh and config/unpack.
1356 :     - Default contents of config/srcarchiveurl uses $VERSION and
1357 :     normally does not have to be edited to reflect a version change.
1358 :    
1359 :     (As a result, a version change can be done by just editing
1360 :     config/version, the rest is now automatic.)
1361 :    
1362 :     ----------------------------------------------------------------------
1363 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1364 : mblume 1669 Date: 2004/10/27 17:50:00 CDT
1365 :     Tag: blume-20041027-btrace-msg
1366 :     Description:
1367 :    
1368 :     BackTrace.monitor now also reports the source of the exception that
1369 :     triggered the trace.
1370 :    
1371 :     ----------------------------------------------------------------------
1372 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1373 : mblume 1668 Date: 2004/10/27 17:20:00 CDT
1374 :     Tag: blume-20041027-x86-c-calls
1375 :     Description:
1376 :    
1377 :     This is the HISTORY entry for two earlier commits, both concerning
1378 :     the x86 c-calls code in MLRISC:
1379 :    
1380 :     - added a missing LOAD in the code that deals with struct arguments
1381 :     - made sure the caller does not add the wrong number of bytes to the
1382 :     stack pointer after a call of a function returning a struct
1383 :     (the callee already pops the implicit argument which points to
1384 :     the space reserved for the result)
1385 :    
1386 :     ----------------------------------------------------------------------
1387 : allenleung 1662 Name: Allen Leung (leunga (at) reservoir (dot) com)
1388 :     Date: 2004/10/24 14:00:00 EST
1389 :     Tag: leunga-20041024-x86-gas-fucomip
1390 :     Description:
1391 :    
1392 :     John discovered a bug in the syntax of fucomip.
1393 :     The opcodes FU?COMIP? have been changed to
1394 :    
1395 :     fu?comip? %st(i), %st
1396 :    
1397 :     ----------------------------------------------------------------------
1398 : mblume 1622 Name: Matthias Blume (blume (at) tti - c (dot) org)
1399 : mblume 1656 Date: 2004/10/20 15:06:00 CDT
1400 :     Tag: blume-20041020-standalone-backtrace
1401 :     Description:
1402 :    
1403 :     Added a mechanism for getting back-trace information from standalone
1404 :     programs. Here is how it works:
1405 :    
1406 :     1. The part of the program from which you want to get backtrace
1407 :     information (usually the whole program) should be wrapped with
1408 :     BackTrace.monitor. This is a (unit->'a)->'a function, and your
1409 :     main program could be modified from something like
1410 :    
1411 :     fun main (pgm, args) = ...
1412 :    
1413 :     to
1414 :    
1415 :     fun main (pgm, args) = BackTrace.monitor (fn () => ...)
1416 :    
1417 :     2. To be able to access BackTrace.monitor, you have to add
1418 :     library $smlnj-tdp/plugins.cm to the .cm file that contains your
1419 :     main function.
1420 :    
1421 :     3. Remove all compiled code (i.e., all the .cm/ subdirectories that
1422 :     CM might have created in the past for your project).
1423 :    
1424 :     4. Build the system using this command line:
1425 :    
1426 :     ml-build -Ctdp.instrument=true \$smlnj-tdp/back-trace.cm \
1427 :     myprog.cm MyProg.main myprog
1428 :    
1429 :     instead of the usual
1430 :    
1431 :     ml-build myprog.cm MyProg.main myprog
1432 :    
1433 :     I changed a library name:
1434 :    
1435 :     $/trace-debug-profile.cm --> $smlnj-tdp/plugins.cm
1436 :    
1437 :     New libraries:
1438 :    
1439 :     $smlnj-tdp/back-trace.cm
1440 :     -- when loaded causes the back-trace plugin to be installed
1441 :     $smlnj-tdp/coverage.cm
1442 :     -- when loaded causes the coverage plugin to be installed
1443 :    
1444 :     ----------------------------------------------------------------------
1445 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1446 : mblume 1655 Date: 2004/10/18 16:45:00 CDT
1447 :     Tag: blume-20041018-groupowner
1448 :     Description:
1449 :    
1450 :     Added an "obsolete" warning for the "group owner" syntax to CM's parser.
1451 :    
1452 :     Eliminated group owner specs from .cm files throughout the source tree.
1453 :    
1454 :     ----------------------------------------------------------------------
1455 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1456 : mblume 1653 Date: 2004/10/15 15:45:00 CDT
1457 :     Tag: blume-20041015-coverage
1458 :     Description:
1459 :    
1460 :     * Test coverage tool added!
1461 :     * Further reorganization of tracing-, debugging-, and profiling support:
1462 :    
1463 :     - moved original BTImp -- now called BackTrace -- into a separate
1464 :     library called $/trace-debug-profile.cm
1465 :     - eliminated all mentions of BTrace from SMLofNJ.Internals
1466 :     - only the instrumentation mechanism is now left in the compiler proper
1467 :     - BackTrace module is a plugin which is NOT plugged in by default
1468 :     - Coverage module is another such plugin
1469 :    
1470 :     To get the benefits of any of these plugin modules, the code in
1471 :     question must be compiled with tdp instrumentation turned on. This
1472 :     can be done by setting SMLofNJ.Internals.TDP.mode to true.
1473 :     (The ref cell is also controlled via the -Ctdp.instrument=... switch.)
1474 :    
1475 :     Plugins are selected at link time. (Pre-compiled instrumented code
1476 :     can be re-loaded with different plugins in effect.) When an
1477 :     instrumented module is linked, whatever plugins are at that time
1478 :     enabled will come into effect for that module.
1479 :    
1480 :     To enable the back-trace plugin, load library $/trace-debug-profile.cm
1481 :     and invoke BackTrace.install() (e.g., from the interactive prompt).
1482 :     To enable the coverage plugin, load the same library and invoke
1483 :     Coverage.install().
1484 :    
1485 :     Back-traces are generated automatically on uncaught exceptions and
1486 :     when the code in question explicitly invokes BackTrace.trigger().
1487 :    
1488 :     Coverage (and execution frequency-) information must be queried
1489 :     explicitly by calling Coverage.not_covered and Coverage.hot_spots.
1490 :    
1491 :     ----------------------------------------------------------------------
1492 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1493 : mblume 1652 Date: 2004/10/14 17:40:00 CDT
1494 :     Tag: blume-20041014-tdp-core
1495 :     Description:
1496 :    
1497 :     Snapshot of a significant overhaul of how the trace/debug/profile support
1498 :     is hooked into the system (specifically: Core and SMLofNJ.Internals).
1499 :    
1500 :     ----------------------------------------------------------------------
1501 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1502 : mblume 1650 Date: 2004/10/13 16:34:00 CDT
1503 :     Tag: blume-20041013-tdp
1504 :     Description:
1505 :    
1506 :     Some rationalization of names:
1507 :    
1508 :     structure BTrace -> structure TDPInstrument
1509 :     etc.
1510 :    
1511 :     This is is preparation of using the original back-trace
1512 :     instrumentation for other purposes. "TDP" stands for
1513 :     Trace/Debug/Profile.
1514 :    
1515 :     The control flag controlling whether instrumentation is on or off is now
1516 :     registered under a different name, so instead of running sml as
1517 :    
1518 :     sml -Cinstrument.btrace-mode=true
1519 :    
1520 :     one has to say
1521 :    
1522 :     sml -Ctdp.instrument=true
1523 :    
1524 :     ----------------------------------------------------------------------
1525 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1526 : mblume 1645 Date: 2004/10/11 16:37:00 CDT
1527 :     Tag: blume-20041011-regions
1528 :     Description:
1529 :    
1530 :     Made some minor modifications to elabcore.sml to have source regions
1531 :     be propagated more tightly -- resulting in better (i.e., smaller)
1532 :     regions being reported in error- and debug messages.
1533 :    
1534 :     ----------------------------------------------------------------------
1535 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1536 : mblume 1644 Date: 2004/10/08 22:50:00 CDT
1537 :     Tag: blume-20041008-cmkw
1538 :     Description:
1539 :    
1540 :     Fixed handling of keywords in .cm files: After seeing "is" the lexer
1541 :     treats subsequent occurrences of "group", "library", "source", "is",
1542 :     "*", and "-" as ordinary identifiers rather than keywords.
1543 :    
1544 :     Most seriously, this fixes a problem with CM's "shell" tool. The tool
1545 :     is supposed to accept a tool argument called "source", but this did
1546 :     not work because of the clash with the keyword.
1547 :    
1548 :     ----------------------------------------------------------------------
1549 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1550 : mblume 1642 Date: 2004/10/07 16:00:00 CDT
1551 :     Tag: blume-20041007-cleanup
1552 :     Description:
1553 :    
1554 :     Assorted cleanup work:
1555 :    
1556 :     - got rid of intstrmap in favor of using the library's
1557 :     hash table implementation
1558 :     - threw out most of the pathnames stuff, as it was not used anyway
1559 :     - simplified tokentable implementation
1560 :     - fixed some minor spelling errors
1561 :    
1562 :     ----------------------------------------------------------------------
1563 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1564 : mblume 1641 Date: 2004/10/06 15:15:15 CDT
1565 :     Tag: blume-20041006-handler
1566 :     Description:
1567 :    
1568 :     Cleaned up the absyn to reflect the invariant that HANDLE always
1569 :     carries a FNexp as part of the type definition. This eliminates some
1570 :     superfluous sanity checks at runtime down the road.
1571 :    
1572 :     Some minor cleanup of the btrace code.
1573 :    
1574 :     ----------------------------------------------------------------------
1575 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1576 : mblume 1640 Date: 2004/10/01 10:20:30 CDT
1577 :     Tag: blume-20041001-slave
1578 :     Description:
1579 :    
1580 :     Added hack to make slave mode work in the presence of the version
1581 :     tool. (Still, since the master does two passes over the code for
1582 :     CMB.make, the release number gets bumped twice when slaves are
1583 :     attached. I don't know if this is worth fixing...)
1584 :    
1585 :     ----------------------------------------------------------------------
1586 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1587 : mblume 1639 Date: 2004/09/30 10:55:00 CDT
1588 :     Tag: blume-20040930-version
1589 :     Description:
1590 :    
1591 :     * Moved the "version" magic into its own little library under
1592 : mblume 1672 src/system/smlnj/internal. This avoids expensive reconstruction of
1593 :     a stable src/compiler/core.cm.
1594 : mblume 1639
1595 : mblume 1672 * At the same time, structure CompilerVersion is now known as
1596 :     structure SMLNJVersion.
1597 : mblume 1639
1598 :     * Arranged for the version tool to NOT kick in when rebuilding the system
1599 :     (makeml -rebuild, fixpt). Otherwise one would never reach a fixpoint.
1600 :     Also, loading the versiontool does not work when rebuilding the system
1601 :     because CM is not properly initialized at that time.
1602 :    
1603 :     ----------------------------------------------------------------------
1604 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1605 : mblume 1638 Date: 2004/09/29 14:00:00 CDT
1606 :     Tag: blume-20040929-autoversion
1607 :     Description:
1608 :    
1609 :     Implemented some CM magic to have
1610 :     file src/compiler/TopLevel/main/version.sml
1611 :     generated automagically.
1612 :     The version is taken from two files: config/version and config/release.
1613 :     The first is expected to contain a two-part version number such as 110.49.
1614 :     The second should contain a single number, but it may be missing.
1615 :    
1616 :     If the environment variable VERSIONTOOL_BUMP_RELEASE is defined at the
1617 :     time the version tool is loaded (which is the first time you say CMB.make),
1618 :     then the tool will increment the value stored in config/release every
1619 :     time CMB.make is invoked.
1620 :    
1621 :     The binfile format is now insensitive to anything beyond the first
1622 :     two components of a version number, so bumping the release does not render
1623 :     binfiles incompatible. Auto-bumping can be used to keep track of versions
1624 :     during development without invalidating existing binfiles.
1625 :    
1626 :     In any case, every CMB.make updates the date information in version.sml.
1627 :     (This is the date that is printed in the banner.)
1628 :    
1629 :     ----------------------------------------------------------------------
1630 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1631 : mblume 1634 Date: 2004/09/28 10:53:00 CDT
1632 :     Tag: blume-20040928-controls
1633 :     Description:
1634 :    
1635 :     Some cleanup of the controls code.
1636 :    
1637 :     ----------------------------------------------------------------------
1638 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1639 : mblume 1633 Date: 2004/09/27 22:08:00 CDT
1640 :     Tag: blume-20040927-controls
1641 :     Description:
1642 :    
1643 :     Added two pieces of functionality to the Controls interface:
1644 :    
1645 :     1. val save'restore: 'a control -> unit -> unit
1646 :    
1647 :     grabs the current value of the control in stage 1 and restores it
1648 :     in stage 2.
1649 :    
1650 :     2. val set' : 'a control * 'a -> unit -> unit
1651 :    
1652 :     stores the given value into the control in stage 2 (i.e., delayed)
1653 :     but does all error checking in stage 1.
1654 :     (This is for string controls that need to do parse their argument --
1655 :     something that might fail. In some cases, notably in CM, one
1656 :     already knows the intended argument but wants to delay the actual
1657 :     assignment until a time when error recovery would be more difficult.)
1658 :    
1659 :     Changed the handling of controls in tool arguments to classes "sml" and
1660 :     "lazysml":
1661 :     - use Controls.save'restore as a more robust way of restoring the
1662 :     old value (in particular: without having to re-parse the string)
1663 :     - use controls to handle the "overload" keyword in the init group
1664 :     (I believe this change actually fixes a long-standing obscure bug.)
1665 :    
1666 :     ----------------------------------------------------------------------
1667 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1668 : mblume 1632 Date: 2004/09/27 17:00:00 CDT
1669 :     Tag: blume-20040927-lazysml
1670 :     Description:
1671 :    
1672 :     Added a new tool class called "lazysml" to CM's tool chest. The only
1673 :     difference to "sml" is that compilation is done with Control.lazysml
1674 :     set to true. A source of class "lazysml" is automatically recognized
1675 :     by a file name suffix of ".lml".
1676 :    
1677 :     In addition to the above feature, the original class "sml" now also
1678 :     supports a tool argument "lazy" which has the same effect. As a
1679 :     result, the following three lines are equivalent:
1680 :    
1681 :     foo.sml : lazysml
1682 :     foo.sml : sml (lazy)
1683 :     foo.sml (lazy)
1684 :    
1685 :     The setting goes into effect both during parsing and during
1686 :     compilation. The original setting is restored right after parsing and
1687 :     after compilation, respectively.
1688 :    
1689 :     In addition to all the above, there is also a general mechanism to set
1690 :     ANY of the "controls" that are available at the command line via
1691 :     "-C..." on a per-sml-file basis. The same rules that apply for "lazy"
1692 :     apply as well. (In fact, "lazy" is implemented as a special case of
1693 :     the general mechanism.)
1694 :    
1695 :     The .cm file syntax uses a new keyword tool argument called "with".
1696 :     There are several ways of indicating the desired settings:
1697 :    
1698 :     foo.sml (with:parser.quotations=true)
1699 :     foo.sml (with:(name:parser.quotations value:true))
1700 :     foo.sml (with:(name:name1 value:value1 name:name2 value:value2 ...))
1701 :     foo.sml (with:(name1=value1 name2=value2 ...))
1702 :     foo.sml (with:(name1=value1 name:name2 value:value2 name3=value3 ...))
1703 :    
1704 :     etc.
1705 :    
1706 :     Another possible abbreviation is to leave out the =v or value:v part
1707 :     if the name refers to a boolean control (in which case the value is
1708 :     taken to be true). Thus, one could get lazy sml also by saying:
1709 :    
1710 :     foo.sml (with:parser.lazy-keyword=true)
1711 :     foo.sml (with:parser.lazy-keyword)
1712 :     foo.sml (with:(name:parser.lazy-keyword value:true))
1713 :     foo.sml (with:(name:parser.lazy-keyword))
1714 :    
1715 :     ----------------------------------------------------------------------
1716 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1717 : mblume 1631 Date: 2004/09/24 16:22:00 CDT
1718 :     Tag: blume-20040924-ppc-long-branch
1719 :     Description:
1720 :    
1721 :     Turned message about "emiting long form of branch" off by default.
1722 :     Added a control flag to turn it back on when desired.
1723 :    
1724 :     ----------------------------------------------------------------------
1725 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1726 : mblume 1630 Date: 2004/09/24 16:05:00 CDT
1727 :     Tag: blume-20040924-rounding
1728 :     Description:
1729 :    
1730 :     Applied patch for setting rounding modes under Mac OS X. Thanks to
1731 :     Melissa O'Neill for providing the code!
1732 :    
1733 :     ----------------------------------------------------------------------
1734 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1735 : mblume 1629 Date: 2004/09/23 17:30:00 CDT
1736 :     Tag: blume-20040923-envvars
1737 :     Description:
1738 :    
1739 :     1. Changed definition of type ControlRegistry.registry_tree to
1740 :     include control_info (i.e., the name of the controlling environment
1741 :     variable).
1742 :    
1743 :     2. Added command-line flags -e and -E to print the names of environment
1744 :     variables that can be used to control internal settings.
1745 :     (This uses the new API mentioned in 1.)
1746 :    
1747 :     ----------------------------------------------------------------------
1748 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1749 : mblume 1622 Date: 2004/09/13 16:50:00 CDT
1750 :     Tag: Release_110_49
1751 :     Description:
1752 :    
1753 :     New working version (110.49). NEW BOOTFILES!
1754 :    
1755 :     ----------------------------------------------------------------------
1756 : mblume 1621 Name: Matthias Blume
1757 :     Date: 2004/09/13 16:20:00 CDT
1758 :     Tag: blume-20040913-config-mlrisc
1759 :     Description:
1760 :    
1761 :     Put target "mlrisc" back into the default list.
1762 :     (There is no harm in having it, and some users have expressed their
1763 :     wish to have "mlrisc" included by default.)
1764 :    
1765 :     ----------------------------------------------------------------------
1766 : jhr 1620 Name: John Reppy
1767 :     Date: 2004/09/13
1768 :     Tag: jhr-20040913-signals
1769 :     Description:
1770 :    
1771 :     Fixed the signal masking code to properly nest mask/unmask operations
1772 :     on a per-signal basis.
1773 :    
1774 :     ----------------------------------------------------------------------
1775 : mblume 1619 Name: Matthias Blume (blume (at) tti - c (dot) org)
1776 :     Date: 2004/09/08 13:20:00 CDT
1777 :     Tag: blume-20040908-heap-magic
1778 :     Description:
1779 :    
1780 :     Bumped the heap macig number to 0x09082004 to account for the changed
1781 :     layout of the ML frame under MacOS X.
1782 :    
1783 :     ----------------------------------------------------------------------
1784 : allenleung 1618 Name: Allen Leung (leunga (at) reservoir (dot) com)
1785 :     Date: 2004/09/03 11:26:00 EST
1786 :     Tag: leunga-20040903-cygwin-install
1787 :     Description:
1788 :    
1789 :     Added a patch to _arch-n-opsys to enable the Cygwin runtime.
1790 :     The Cygwin runtime is turned on by setting the environment
1791 :     variable SMLNJ_CYGWIN_RUNTIME to 1.
1792 :    
1793 :     ----------------------------------------------------------------------
1794 : mblume 1615 Name: Matthias Blume (blume (at) tti - c (dot) org)
1795 : mblume 1617 Date: 2004/08/31 17:15:00 CDT
1796 :     Tag: blume-20040831-core
1797 :     Description:
1798 :    
1799 :     Added some exports to src/compiler/core.cm upon request by J. Joemann.
1800 :    
1801 :     ----------------------------------------------------------------------
1802 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1803 : mblume 1616 Date: 2004/08/30 17:55:00 CDT
1804 :     Tag: blume-20040830-installer
1805 :     Description:
1806 :    
1807 :     Upon request by Johannes Joemann:
1808 :    
1809 :     - improved ML code of installer to fall back to coping when renaming
1810 :     fails (i.e., when source and target are on different file systems);
1811 :     the code compiles but has yet to be tested in anger
1812 :     - removed mlrisc from list of default targets (config/targets)
1813 :    
1814 :     ----------------------------------------------------------------------
1815 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1816 : mblume 1615 Date: 2004/08/27 17:20:00 CDT
1817 :     Tag: blume-20040827-ptreql
1818 :     Description:
1819 :    
1820 :     Added ptreql primop to structure InlineT (upon request from Larry
1821 :     Paulson).
1822 :    
1823 :     ----------------------------------------------------------------------
1824 : allenleung 1613 Name: Allen Leung (leunga (at) reservoir (dot) com)
1825 : allenleung 1614 Date: 2004/08/15 21:21:00 EST
1826 :     Tag: leunga-110_48-udgraph
1827 :     Description:
1828 :    
1829 :     Another bug fix from Carl Hauser:
1830 :    
1831 :     diff /net/niflab/smlnj48/src/MLRISC/graphs/udgraph.sml udgraph.sml
1832 :     > 48c48
1833 :     > < | rmv((e as (k,_))::es,L) = rmv(es,if k = i then es else
1834 :     > e::L)
1835 :     > ---
1836 :     > > | rmv((e as (k,_))::es,L) = rmv(es,if k = i then L else e::L)
1837 :     > Without this, any deletion of an edge in an undirected graph does severe
1838 :     > violence to the graph.
1839 :    
1840 :     ----------------------------------------------------------------------
1841 :     Name: Allen Leung (leunga (at) reservoir (dot) com)
1842 : allenleung 1613 Date: 2004/08/10 23:35:00 EST
1843 :     Tag: leunga-110_48-ppc
1844 :     Description:
1845 :    
1846 :     The IBM/MacOS syntax switch on PPC was incorrectly swapped.
1847 :    
1848 :     ----------------------------------------------------------------------
1849 : mblume 1602 Name: Matthias Blume (blume (at) tti - c (dot) org)
1850 : mblume 1615 Date: 2004/08/10 12:00:00 CDT
1851 : mblume 1602 Tag: Release_110_48
1852 :     Description:
1853 :    
1854 :     New working version (110.48). NEW BOOTFILES!
1855 :    
1856 :     ----------------------------------------------------------------------
1857 : allenleung 1599 Name: Allen Leung (leunga (at) reservoir (dot) com)
1858 :     Date: 2004/08/09 12:21:00 EST
1859 :     Tag: leunga-110_47-dijsktra
1860 :     Description:
1861 :    
1862 :     Bug fix from Carl Hauser:
1863 :    
1864 : blume 2155 single_source_shortest_paths in dijkstra.sml was observed to get wrong
1865 : allenleung 1599 answers (by comparing to single_source_shortest_paths in bellman-ford.sml).
1866 :    
1867 : blume 2155 The problem is that following the expression A.update(dist,s,Num.zero)
1868 : allenleung 1599 it is necessary to update the priority queue using Q.decreaseWeight(Q,s).
1869 :    
1870 :     ----------------------------------------------------------------------
1871 : mblume 1595 Name: Matthias Blume (blume (at) tti - c (dot) org)
1872 :     Date: 2004/08/06 18:10:00 CDT
1873 :     Tag: blume-20040806-cmdline
1874 :     Description:
1875 :    
1876 :     Fiddled with handling of command-line options:
1877 :    
1878 :     * sml now quits after processing the command line
1879 :     if -H, -S, -h<n>, or -s<n> appears as the last
1880 :     command-line argument
1881 :     * a new option -q terminates the session when encountered on
1882 :     the command line; subsequent arguments will be ignored
1883 :     * bug fixes: short (erroneous) arguments are no longer ignored
1884 :     completely
1885 :    
1886 :     ----------------------------------------------------------------------
1887 : allenleung 1593 Name: Allen Leung (leunga (at) reservoir (dot) com)
1888 :     Date: 2004/08/04 18:17:00 EST
1889 :     Tag: leunga-110_47-ppc-ibm-asm
1890 :     Description:
1891 :    
1892 :     - Added minimal IBM assembly syntax support for PowerPC.
1893 :    
1894 : blume 2155 - Cygwin: manually changed the file cygwin.def. Some exported symbols have
1895 : allenleung 1593 been altered in the runtime. We need an automatic way to keep the file
1896 :     in sync.
1897 :    
1898 :     ----------------------------------------------------------------------
1899 : mblume 1548 Name: Matthias Blume (blume (at) tti - c (dot) org)
1900 : mblume 1581 Date: 2004/08/04 14:00:00 CDT
1901 :     Tag: Release_110_47
1902 :     Description:
1903 :    
1904 :     New working version (110.47). NEW BOOTFILES!
1905 :    
1906 :     ----------------------------------------------------------------------
1907 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1908 : mblume 1580 Date: 2004/08/03 14:25:00 CDT
1909 :     Tag: blume-20040803-callingconv
1910 :     Description:
1911 :    
1912 :     Added low-level support for choosing C calling conventions by
1913 :     twiddling the type of rawccall. (See
1914 :     src/compiler/Semant/types/cproto.sml for details.)
1915 :    
1916 :     ----------------------------------------------------------------------
1917 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1918 : mblume 1579 Date: 2004/08/02 15:55:00 CDT
1919 :     Tag: blume-20040802-backout
1920 :     Description:
1921 :    
1922 :     Backed out of change to win32-filesys.c. The earlier patch to
1923 :     get_file_time caused CM to produce files with the wrong time stamp.
1924 :    
1925 :     ----------------------------------------------------------------------
1926 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1927 : mblume 1573 Date: 2004/08/02 14:45:00 CDT
1928 :     Tag: blume-20040802-nlffi-win32
1929 :     Description:
1930 :    
1931 :     Added NLFFI support for Win32, adapted from a patch provided by David
1932 :     Hansel. This is currently completely untested. Also, the issue
1933 :     concerning stdcall vs. ccall is still unresolved.
1934 :    
1935 :     ----------------------------------------------------------------------
1936 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1937 : mblume 1572 Date: 2004/07/30 17:55:00 CDT
1938 :     Tag: blume-20040730-various
1939 :     Description:
1940 :    
1941 :     Gearing up towards 110.47...
1942 :    
1943 :     - various minor bugfixes to ml-nlffigen
1944 :     - a beginning of a manual for nlffi
1945 :    
1946 :     - eliminated 'export name=value' in config/install.sh as this does
1947 :     not work with certain versions of /bin/sh
1948 :     (Thanks to David King at Motorola for catching this.)
1949 :    
1950 :     - several bugfixes provided or suggested by David Hansel at Reactive Systems:
1951 :     - added a test for tm==NULL to gmtime.c and localtime.c
1952 :     - applied patch for incorrect GetFileTime under win32
1953 :     - toSeconds -> toMilliseconds in Win32/win32-process.sml
1954 :    
1955 :     ----------------------------------------------------------------------
1956 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1957 : mblume 1557 Date: 2004/07/21 18:20:00 CDT
1958 :     Tag: blume-20040721-nlffigen
1959 :     Description:
1960 :    
1961 :     - Fixed minor issue in ml-nlffigen: Now generate structure T_foo
1962 :     for a typedef to an incomplete type, but leave out the "typ" member.
1963 :     (This is just for consistency.)
1964 :     - Started to produce what is supposed to become better (i.e., comprehensive)
1965 :     documentation of what ml-nlffigen does and produces.
1966 :    
1967 :     ----------------------------------------------------------------------
1968 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
1969 : mblume 1548 Date: 2004/07/14 16:25:00 CDT
1970 :     Tag: blume-20040714-union
1971 :     Description:
1972 :    
1973 :     Added C_UNION to c-calls/c-types.sml and updated the machinery
1974 :     (ml-nlffigen, cproto.sml) that conveys C function interface
1975 :     information to the code generator.
1976 :    
1977 :     However, the actual architecture-specific implementation of function
1978 :     arguments and results that are C unions is still not implemented.
1979 :    
1980 :     ----------------------------------------------------------------------
1981 : allenleung 1539 Name: Allen Leung (leunga (at) reservoir (dot) com)
1982 : allenleung 1546 Date: 2004/07/14 14:38:00 EST
1983 :     Tag: leunga-110_46_1-ppc-lwzu
1984 :     Description:
1985 :    
1986 :     Added these instructions to the PowerPC architecture:
1987 :     LBZU(X), LHZU(X), LWZU(X),
1988 : blume 2155 STWU(X), STFDU, STFSU
1989 :    
1990 : allenleung 1546 etc...
1991 :    
1992 :     Note: I haven't added their instruction encoding into the description.
1993 :    
1994 :     ----------------------------------------------------------------------
1995 :     Name: Allen Leung (leunga (at) reservoir (dot) com)
1996 : allenleung 1539 Date: 2004/07/13 15:04:00 EST
1997 :     Tag: leunga-110_46_1-ppc-lwarx
1998 :     Description:
1999 :    
2000 :     Added the two instructions LWARX and STWCX to the PowerPC
2001 : blume 2155 instruction set.
2002 : allenleung 1539
2003 :     A (untested) rewrite of loop-structure.sml. The old version
2004 :     is completely broken.
2005 :    
2006 :     ----------------------------------------------------------------------
2007 : mblume 1479 Name: Matthias Blume (blume (at) tti - c (dot) org)
2008 : mblume 1537 Date: 2004/07/13 13:50:00 CDT
2009 :     Tag: blume-20040713-nlffi
2010 :     Description:
2011 :    
2012 :     - use paramAlloc to report c-calls with too many arguments
2013 :     (for PPC version where parameter area is pre-allocated)
2014 :     - added ccall_maxargspace to machspec (to implement the above)
2015 :     - made "make" commend in CM's "make" tool configurable
2016 :     - added option (default: on) for passing the name of the SML/NJ's "bin"
2017 :     directory to "make"; the call looks like this:
2018 :    
2019 :     make <options> SMLNJ_BINDIR=<dir> <target>
2020 :    
2021 :     This can be used by the Makefile to, e.g., pick the "right" version
2022 :     of ml-nlffigen.
2023 :     - minor code tweaks
2024 :    
2025 :     ----------------------------------------------------------------------
2026 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2027 : mblume 1532 Date: 2004/07/12 22:50:00 CDT
2028 :     Tag: blume-110_46_1-macosx-nlffi
2029 :     Description:
2030 :    
2031 :     NLFFI under Mac OS X now working (sort of). This is largely untested,
2032 :     though.
2033 :    
2034 :     Note: 1. You have to make a new, clean build of the runtime system.
2035 :     2. There are new BOOTFILES, you have to use them!
2036 :     (Doing the bootstrap process yourself would be *very* painful!
2037 :     If you absolutely have to do it, build the system under
2038 :     a different architecture and then cross-compile.)
2039 :    
2040 :     Version bumped to 110.46.1 to account for runtime data format changes.
2041 :    
2042 :     ----------------------------------------------------------------------
2043 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2044 : mblume 1515 Date: 2004/06/18 14:30:00 CDT
2045 :     Tag: blume-20040618-unix
2046 :     Description:
2047 :    
2048 :     Changed the implementation of structure Unix so that the same stream
2049 :     is returned every time one of the {text,bin}{In,Out}streamOf functions
2050 :     is invoked on the same proc. This is not what the spec currently
2051 :     says -- although IMO it arguably should. (See discussion below.)
2052 :    
2053 :     ----------------------------------------------------------------------
2054 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2055 : mblume 1507 Date: 2004/06/17 18:15:00 CDT
2056 :     Tag: Release_110_46
2057 :     Description:
2058 :    
2059 :     New working version (110.46). NEW BOOTFILES!
2060 :    
2061 :     ----------------------------------------------------------------------
2062 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2063 : mblume 1506 Date: 2004/06/17 17:20:00 CDT
2064 :     Tag: blume-20040617-timer-unix
2065 :     Description:
2066 :    
2067 :     Changed the interface of structures Timer and Unix to match the most
2068 :     recent Basis spec.
2069 :    
2070 :     In the case of Unix there still seems to be an open/weird issue:
2071 :    
2072 :     The {text,bin}{In,Out}streamOf functions are supposed to create
2073 :     fresh streams whenever they are called -- as opposed to have them
2074 :     return the same stream every time. This design is supposed to
2075 :     prevent space leaks caused by proc values hanging on to streams.
2076 :    
2077 :     The reap function, on the other hand, is supposed to close the
2078 :     streams. This cannot be done without having a handle on the
2079 :     stream in proc after all...
2080 :    
2081 :     I took the liberty to implement the following stopgap solution:
2082 :    
2083 :     The proc value hangs on to the most recently created stream(s).
2084 :     Reap closes those. If either or both of the two streams hadn't
2085 :     been created at all yet, then reap will close the corresponding
2086 :     file descriptors directly.
2087 :    
2088 :     PS: I don't understand the original space leak argument anymore. If
2089 :     a proc hangs on to the imperative stream, then I/O operations on those
2090 :     will advance the state of the cached stream and avoid the space leak.
2091 :    
2092 :     ----------------------------------------------------------------------
2093 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2094 : mblume 1500 Date: 2004/05/28 16:45:00 CDT
2095 :     Tag: blume-20040528-basis
2096 :     Description:
2097 :    
2098 :     Added signature PACK_REAL and exported functor PrimIO.
2099 :    
2100 :     ----------------------------------------------------------------------
2101 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2102 : mblume 1498 Date: 2004/05/25 16:00:00 CDT
2103 :     Tag: blume-20040525-group-owner
2104 :     Description:
2105 :    
2106 :     CM now ignores (but still accepts) the "owner" information in group
2107 :     descriptions. The owner of a group is its next enclosing
2108 :     library. Each group must have a unique owner. (There is a virtual
2109 : mblume 1505 "toplevel" library that owns groups which are not nested within a real
2110 : mblume 1498 library.) Previously, each group had to explicitly declare its owner,
2111 :     and CM would check that such a declaration is correct. The new scheme
2112 :     is to have CM check that for each group there is precisely one owning
2113 :     library.
2114 :    
2115 :     The advantage of the new scheme is that the programmer no longer needs
2116 :     to maintain the somewhat annoying owner information. The downside is
2117 :     that CM cannot enforce the ownership rule across multiple runs of
2118 :     CM.make. Fortunately, enclosing the same group in two different
2119 :     libraries A and B which are not part of the same program does not
2120 :     cause real problems.
2121 :    
2122 :     ----------------------------------------------------------------------
2123 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2124 : mblume 1496 Date: 2004/05/20 16:00:00 CDT
2125 :     Tag: blume-20040520-win32
2126 :     Description:
2127 :    
2128 :     Made the win32 version work again. (Strangely, a misplaced comma had
2129 :     slipped into win32-process.c which prevented the runtime from being
2130 :     compiled correctly.)
2131 :    
2132 :     Also, included a minor addition to ml-build.bat analogous to what was
2133 :     done in blume-20040519-ml-build.
2134 :    
2135 :     ----------------------------------------------------------------------
2136 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2137 : mblume 1494 Date: 2004/05/19 22:10:00 CDT
2138 :     Tag: blume-20040519-ml-build
2139 :     Description:
2140 :    
2141 :     Arranged for ml-build to clean up after itself a little bit better.
2142 :     The script generates a temporary SML source file and compiles it using
2143 :     CM, so CM generates metadata (GUID, SKEL, objectfile) for it. It now
2144 :     gets rid of those at the end, so they don't accumulate under .cm.
2145 :    
2146 :     This required a minor change to install.sh because the name of the
2147 :     metadata directory (default: .cm) is actually configurable at
2148 :     installation time.
2149 :    
2150 :     ----------------------------------------------------------------------
2151 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2152 : mblume 1491 Date: 2004/05/18 15:50:00 CDT
2153 :     Tag: blume-20040518-mkreader
2154 :     Description:
2155 :    
2156 :     Added Posix.IO.mk{Bin,Text}{Reader,Writer} by lifting their respective
2157 :     implementations from internal modules PosixBinPrimIO and PosixTextPrimIO.
2158 :    
2159 :     ----------------------------------------------------------------------
2160 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2161 : mblume 1489 Date: 2004/05/11 14:35:00 CDT
2162 :     Tag: blume-20040511-win32sock
2163 :     Description:
2164 :    
2165 :     Added previously missing support for many socket-related functions
2166 :     under win32. Thanks to David Hansel <hansel@reactive-systems.com>
2167 :     for the voluminous patch!
2168 :    
2169 :     (I have not tested this patch under win32 yet.)
2170 :    
2171 : mblume 1490 Here is David's e-mail:
2172 :    
2173 :     Hi,
2174 :    
2175 :     Attached to this email you find a diff against sml/nj 110.45
2176 :     that will enable socket support under Windows.
2177 :    
2178 :     To apply the patch (using unix or cygwin)
2179 :     1) gunzip runtime.diff.gz
2180 :     2) "cd" into "src/runtime" in the source tree of a fresh
2181 :     110.45 installation.
2182 :     3) patch -p 1 < [your/path/to]runtime.diff
2183 :    
2184 : blume 2155 The code compiles fine but has NOT yet been extensively tested.
2185 :     I only ran a few tests for basic socket client functionality
2186 :     (which worked fine). Especially the functions that use ioctl
2187 : mblume 1490 are not tested at all and might not work (see below).
2188 :    
2189 :     I implemented this since we want to move to a newer version of sml/nj
2190 :     but need socket support in order to use it. This is the first time I
2191 :     even had a look at the sml/nj source, so please review my changes
2192 :     before making this part of the distribution! Here are a few issues
2193 : blume 2155 that I think might be better for someone to solve who is more
2194 : mblume 1490 familiar with the sml/nj source (and socket programming):
2195 :    
2196 :     - getnetbyaddr.c and getnetbyname.c will raise a "not implemented"
2197 :     exception since I could not figure out what the windows equivalent
2198 :     of these functions is
2199 :    
2200 :     - In sockets-osdep.h there are a some #include statements that are
2201 :     only used in a few files that include sockets-osdep.h
2202 :    
2203 :     - In smlnj-sock-lib.c, function init_fn() calls WSAStartup() but
2204 :     does not process its return value since I don't know how to
2205 :     report an error upwards.
2206 :    
2207 :     - It would probably be good to have a call to WSACleanup() when
2208 : blume 2155 the library is unloaded (if there is such a possibility).
2209 : mblume 1490 Otherwise I think Windows will take care of this automatically
2210 :     when the process finishes.
2211 :    
2212 :     - I used ioctlsocket() as a replacement for ioctl() but I have
2213 :     no idea if that is actually the proper replacement on Windows.
2214 :    
2215 :     - All these issues are marked in the code by "FIXME" comments.
2216 :    
2217 : blume 2155 We use sml/nj extensively in our products and are quite happy
2218 :     with it. I hope this contribution will help you.
2219 : mblume 1490
2220 :     Keep up the good work!
2221 :    
2222 :     David
2223 :    
2224 : mblume 1489 ----------------------------------------------------------------------
2225 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2226 : mblume 1488 Date: 2004/05/11 14:20:00 CDT
2227 :     Tag: blume-20040511-installml
2228 :     Description:
2229 :    
2230 :     Fixed two bugs in installml script. (Thanks to Vesa A. Norrman for
2231 :     the patch.)
2232 :    
2233 :     ----------------------------------------------------------------------
2234 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2235 : mblume 1487 Date: 2004/05/11 14:05:00 CDT
2236 :     Tag: blume-20040511-nlffi-netbsd
2237 :     Description:
2238 :    
2239 :     Added support for nlffi under netbsd. (Thanks to Vesa A. Norrman for
2240 :     the patch.)
2241 :    
2242 :     ----------------------------------------------------------------------
2243 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2244 : mblume 1486 Date: 2004/05/11 12:05:00 CDT
2245 :     Tag: blume-20040511-exports
2246 :     Description:
2247 :    
2248 :     As per request by Adam Chlipala <adam@hcoop.net>, extended various
2249 :     export lists in compiler-related .cm-files.
2250 :    
2251 :     ----------------------------------------------------------------------
2252 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2253 : mblume 1485 Date: 2004/05/11 11:35:00 CDT
2254 :     Tag: blume-20040511-allsource
2255 :     Description:
2256 :    
2257 :     The installer now honors the "src-smlnj" target again, although its meaning
2258 :     has changed from "all sources required for the compiler" to "all sources
2259 :     the installer knows about". In other words, if you enable "src-smlnj"
2260 :     in the "targets" file, then the installer will pull in sources for
2261 :     everything. (Notice that this refers to source code only. Compiled
2262 :     code is still only installed for modules that were requested explicitly
2263 :     or which are required for other modules that were requested explicitly.)
2264 :    
2265 :     ----------------------------------------------------------------------
2266 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2267 : mblume 1482 Date: 2004/04/23 17:40:00 CDT
2268 :     Tag: blume-20040423-ieee-scan
2269 :     Description:
2270 :    
2271 :     Fixed IEEEReal.scan (and .fromString) so that if there is an overflow
2272 :     in the exponent calculation we get INF or ZERO (depending on the mantissa
2273 :     and the sign of the exponent).
2274 :    
2275 :     ----------------------------------------------------------------------
2276 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2277 : mblume 1480 Date: 2004/04/23 10:40:00 CDT
2278 :     Tag: blume-20040423-ml-build
2279 :     Description:
2280 :    
2281 :     The ml-build script now terminates with a non-0 status when something
2282 :     goes wrong.
2283 :    
2284 :     ----------------------------------------------------------------------
2285 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2286 : mblume 1479 Date: 2004/04/22 16:35:00 CDT
2287 :     Tag: blume-20040422-Option
2288 :     Description:
2289 :    
2290 :     Made exception Option to be the same as exception Option.Option
2291 :     (as it should be).
2292 :    
2293 :     ----------------------------------------------------------------------
2294 : allenleung 1468 Name: Allen Leung (leunga (at) reservoir (dot) com)
2295 :     Date: 2004/03/19 14:40:00 EST
2296 :     Tag: leunga-20040319-cygwin-nlffi
2297 :     Description:
2298 :    
2299 :     Fixed the runtime so that ml-nlffi-lib runs on the cygwin version
2300 :     of SML/NJ. The problem is that
2301 :    
2302 :     lib = dlopen(NULL, ...)
2303 :     f = dlsym(lib, "malloc");
2304 :    
2305 : blume 2155 does not work on Windows unless we explicitly export symbols
2306 :     such as 'malloc' during linking. We fixed this by explicitly
2307 : allenleung 1468 exporting the required symbols with the magic gcc incantation:
2308 :    
2309 :     -Wl,--export-all cygwin.def
2310 :    
2311 :     where cygwin.def is a file containing all the symbols that
2312 : blume 2155 we wish to export.
2313 : allenleung 1468
2314 : blume 2155 I suspect this is a Windows problem and we'll have to
2315 :     do the same (somehow with windows compilers) when
2316 : allenleung 1468 we build the native win32 version with the system
2317 : blume 2155 calls LoadLibrary/GetProcAddress.
2318 : allenleung 1468
2319 :     ----------------------------------------------------------------------
2320 : mblume 1384 Name: Matthias Blume (blume (at) tti - c (dot) org)
2321 : mblume 1465 Date: 2004/03/04 16:35:00 CST
2322 :     Tag: blume-20040304-intinf-fmt
2323 :     Description:
2324 :    
2325 :     Fixed problem with IntInf.fmt (sign would show up on the right instead
2326 :     of on the left for BIN, OCT, and HEX).
2327 :    
2328 :     ----------------------------------------------------------------------
2329 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2330 : mblume 1464 Date: 2004/03/04 11:25:00 CST
2331 :     Tag: blume-20040304-symlinks
2332 :     Description:
2333 :    
2334 :     Fixed problem with installer script (unix only) where bin/ml-yacc and
2335 :     friends pointed (via symlinks) to absolute locations instead of just
2336 :     .run-sml. This was reported by Vesa A Norrman.
2337 :    
2338 :     ----------------------------------------------------------------------
2339 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2340 : mblume 1455 Date: 2004/02/13 14:50:00 CST
2341 :     Tag: Release_110_45
2342 :     Description:
2343 :    
2344 :     New working version (110.45). New bootfiles.
2345 :    
2346 :     ----------------------------------------------------------------------
2347 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2348 : mblume 1448 Date: 2004/01/26 15:15:15 CST
2349 :     Tag: blume-20040126-toplevel
2350 :     Description:
2351 :    
2352 :     Improved handling of exceptions at the interactive toplevel.
2353 :    
2354 :     ----------------------------------------------------------------------
2355 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2356 :     Date: 2004/01/26 11:25:00
2357 : mblume 1446 Tag: blume-20040126-app
2358 :     Description:
2359 :    
2360 :     Type of top-level "app" corrected.
2361 : mblume 1447 Added code for setting vp_limitPtrMask to Win32-specific runtime.
2362 : mblume 1446
2363 :     ----------------------------------------------------------------------
2364 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2365 : mblume 1427 Date: 2003/11/18 17:10 CST
2366 :     Tag: blume-20031118-basis-fiddle
2367 :     Description:
2368 :    
2369 :     - changed Timer interface to what might become the spec
2370 :     - POSIX_FLAGS -> BIT_FLAGS according to spec
2371 :     - some other minor discrepancies wrt. spec eliminated
2372 :    
2373 :     ----------------------------------------------------------------------
2374 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2375 : mblume 1414 Date: 2003/11/06 12:00:00 CST
2376 :     Tag: Release_110_44
2377 :     Description:
2378 :    
2379 :     New working version (110.44). New bootfiles.
2380 :    
2381 :     ----------------------------------------------------------------------
2382 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2383 : mblume 1411 Date: 2003/11/04 11:50:00 CST
2384 :     Tag: blume-20031104-move-libraries
2385 :     Description:
2386 :    
2387 :     Eliminated the "dont_move_libraries" directive in config/targets.
2388 :     (The mechanism was broken and could not be fixed easily. Moreover,
2389 :     there does not seem to be any reason not to move all libraries into
2390 :     lib during installation. I originally implemented this directive as a
2391 :     backward-compatibility feature when I first introduced the new CM.
2392 :     Now that things have been stable for a long time and going back to the
2393 :     old CM is not an option, there is no reason to keep it around.)
2394 :    
2395 :     ----------------------------------------------------------------------
2396 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2397 : mblume 1409 Date: 2003/11/03 16:00:00 CST
2398 :     Tag: blume-20031103-installdir
2399 :     Description:
2400 :    
2401 :     Made installer honor INSTALLDIR variable again. (Thanks to Chris
2402 :     Richards for pointing out the problem and providing the solution.)
2403 :    
2404 :     ----------------------------------------------------------------------
2405 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2406 : mblume 1400 Date: 2003/10/01 17:05:00 CDT
2407 :     Tag: blume-20031001-lal-mlrisc
2408 :     Description:
2409 :    
2410 :     MLRISC bug fix from Lal.
2411 :    
2412 :     ----------------------------------------------------------------------
2413 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2414 : mblume 1399 Date: 2003/09/30 16:10:00 CDT
2415 :     Tag: blume-20030930-primio-bat
2416 : mblume 1398 Description:
2417 :    
2418 :     1. Added openVector, nullRd, and nullWr to PRIM_IO.
2419 :     2. Improved .bat files (for Win32 port) to make things work under Win95.
2420 :     (thanks to Aaron S. Hawley for this one)
2421 :    
2422 :     ----------------------------------------------------------------------
2423 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2424 : mblume 1395 Date: 2003/09/26 16:05:00 CDT
2425 :     Tag: blume-20030926-wrappriv
2426 :     Description:
2427 :    
2428 :     Added missing wrapper for privilege "primitive" in $smlnj/viscomp/core.cm.
2429 :    
2430 :     ----------------------------------------------------------------------
2431 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2432 : mblume 1394 Date: 2003/09/26 15:00:00 CDT
2433 :     Tag: blume-20030926-110_43_3
2434 :     Description:
2435 :    
2436 :     - additional cleanup
2437 :     - version number bump, NEW BOOTFILES
2438 :    
2439 :     ----------------------------------------------------------------------
2440 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2441 : mblume 1393 Date: 2003/09/26 12:00:00 CDT
2442 :     Tag: blume-20030926-ppautoload
2443 :     Description:
2444 :    
2445 :     I modified the read-eval-print loop so that the autoloader gets
2446 :     invoked whenever the prettyprinter tries to look up a symbol that
2447 :     is not currently defined in the toplevel environment but which
2448 :     appears in CM's autoload registry. As a result, we see far fewer of
2449 :     those ?.Foo.Bar.xxx names in the prettyprinter's output.
2450 :    
2451 :     In addition to this I tried to clean up some pieces of the Basis
2452 :     implementation (e.g., Socket, Word8Array) in order to prevent other
2453 :     instances of these ?.Foo.Bar.xxx names from being printed.
2454 :    
2455 :     The mechanism that picks names for types still needs some work, though.
2456 :     (Right now it seems that if there is a type A.t which is defined to
2457 : mblume 1394 be B.u, but B is unavailable at toplevel, then A.t gets printed as
2458 :     "?.B.u" although the perhaps more sensible solution would be to use
2459 : mblume 1393 "A.t" in this case. In other words, the prettyprinter should follow
2460 :     a chain of DEFtycs not farther than there are corresponding toplevel
2461 :     names in the current environment.)
2462 :    
2463 :     ----------------------------------------------------------------------
2464 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2465 : mblume 1391 Date: 2003/09/24 16:31:00 CDT
2466 :     Tag: blume-20030924-installer
2467 :     Description:
2468 :    
2469 :     Another installer tweak: All the ML code for the installer is now
2470 :     compiled during CMB.make and put into a little library called
2471 :     $smlnj/installer.cm. The installation then simply invokes
2472 :    
2473 :     sml -m $smlnj/installer.cm
2474 :    
2475 :     and everything happens automagically.
2476 :    
2477 :     Win32: ML code senses value of environment variable SMLNJ_HOME.
2478 :     Unix: ML code senses values of environment variables ROOT, CONFIGDIR,
2479 :     and BINDIR.
2480 :    
2481 :     The new scheme guarantees that the ML code responsible for the installation
2482 :     is in sync with the APIs of the main system. Also, the installer is
2483 :     somewhat faster because the installer script is precompiled.
2484 :    
2485 :     ----------------------------------------------------------------------
2486 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2487 : mblume 1390 Date: 2003/09/24 15:35:00 CDT
2488 :     Tag: blume-20030924-synsock
2489 :     Description:
2490 :    
2491 :     Added a signature SYNCHRONOUS_SOCKET to basis.cm. This is like SOCKET
2492 :     but excludes all non-blocking operations. Defined SOCKET (in Basis)
2493 :     and CML_SOCKET in terms of SYNCHRONOUS_SOCKET. Removed superfluous
2494 :     implementations of non-blocking operations from CML's Socket
2495 :     structure.
2496 :    
2497 :     ----------------------------------------------------------------------
2498 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2499 : mblume 1389 Date: 2003/09/24 15:10:05 CDT
2500 :     Tag: blume-20030924-sockets
2501 :     Description:
2502 :    
2503 :     1. Fixed SOCKET API and implementation to match Basis spec.
2504 :     This required changing the internal representation of sockets to one
2505 :     that remembers (for each socket file descriptor) whether it is currently
2506 :     blocking or non-blocking. This state is maintained lazily (i.e., a system
2507 :     call is made only if the state actually needs to change).
2508 :    
2509 :     2. OS-specific details of sockets were moved into separate files, thus
2510 :     making it possible to unify the bulk of the socket implementations
2511 :     between Unix and Win32.
2512 :    
2513 :     3. CML's socket API changed accordingly.
2514 :     (Note that we need to remove non-blocking functions from this API
2515 :     since they are redundant in the case of CML!)
2516 :    
2517 :     4. CML's socket implementation now makes use of non-blocking functions
2518 :     provided by Basis, thus removing all OS-dependent code from this part
2519 :     of CML.
2520 :    
2521 :     5. Changed Real64.precision from 52 to 53. Minor cleanup in Real64 code.
2522 :    
2523 :     ----------------------------------------------------------------------
2524 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2525 : mblume 1387 Date: 2003/09/22 12:10:00 CDT
2526 : mblume 1388 Tag: blume-20030922-110_43_2
2527 : mblume 1387 Description:
2528 :    
2529 :     Made a new interim version and bootfiles for developer's bootstrapping
2530 :     convenience.
2531 :    
2532 :     110.43.2 -- NEW BOOTFILES
2533 :    
2534 :     ----------------------------------------------------------------------
2535 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2536 : mblume 1385 Date: 2003/09/19 15:55:00 CDT
2537 :     Tag: blume-20030919-cmdir
2538 :     Description:
2539 :    
2540 :     1. new-install.sh -> install.sh
2541 :     2. changed default CM "metadata" directory name to ".cm" (instead of "CM")
2542 :     3. tweaked installer so that another name instead of .cm can be chosen
2543 :     at install time (by setting the CM_DIR_ARC environment variable
2544 :     during installation); once installation is complete, the name is
2545 :     fixed
2546 :    
2547 :     ----------------------------------------------------------------------
2548 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2549 : mblume 1384 Date: 2003/09/18 16:00:00 CDT
2550 :     Tag: blume-20030918-110_43_1
2551 :     Description:
2552 :    
2553 :     Made a new interim version and bootfiles for developer's bootstrapping
2554 :     convenience.
2555 :    
2556 :     110.43.1 -- NEW BOOTFILES
2557 :    
2558 :     ----------------------------------------------------------------------
2559 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2560 : mblume 1383 Date: 2003/09/18 15:20:00 CDT
2561 :     Tag: blume-20030918-misc
2562 :     Description:
2563 :    
2564 :     1. Exported fractionsPerSecond etc. from TimeImp (but not from Time as
2565 :     this seems to be controversial at the moment) and used those in
2566 :     Posix.ProcEnv.times.
2567 :    
2568 :     2. Added Time.{from,to}Nanoseconds to Time.
2569 :    
2570 :     3. Improved Real.{from,to}LargeInt by avoiding needless calculations.
2571 :     For example, fromLargeInt never needs to look at more than 3 "big
2572 :     digits" to get its 53 bits of precision.
2573 :    
2574 :     ----------------------------------------------------------------------
2575 : mblume 1384 Name: Matthias Blume (blume (at) tti - c (dot) org)
2576 : mblume 1381 Date: 2003/09/17 16:30:00 CDT
2577 :     Tag: blume-20030917-real32-slices
2578 :     Description:
2579 :    
2580 :     Added an entry to the primitive environment
2581 :     (compiler/Semant/statenv/prim.sml) for int32->real64 conversion and
2582 :     added code to compiler/CodeGen/main/mlriscGen.sml to implement it.
2583 :    
2584 :     Removed some of the "magic" constants in real64.sml and replaced them
2585 :     with code that generates these values from their corresponding
2586 :     integer counterparts.
2587 :    
2588 :     Made all(?) the slice-related changes to the Basis and made everything
2589 :     compile again...
2590 :    
2591 :     ----------------------------------------------------------------------
2592 : mblume 1368 Name: Matthias Blume (blume (at) tti - c (dot) org)
2593 : mblume 1375 Date: 2003/09/15 17:45:00 CDT
2594 :     Tag: blume-20030915-rbase
2595 :     Description:
2596 :    
2597 :     Fixed bug in Real.fromLargeInt.
2598 :    
2599 :     ----------------------------------------------------------------------
2600 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2601 : mblume 1369 Date: 2003/09/13 18:11:00 CDT
2602 :     Tag: blume-20030913-libinstall
2603 :     Description:
2604 :    
2605 :     Minor bugfix in config/libinstall (set anchor with path to
2606 :     standalone tool after installing it, otherwise libraries that
2607 :     need ml-lex or ml-yacc won't compile the first time the installer
2608 :     runs).
2609 :    
2610 :     ----------------------------------------------------------------------
2611 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2612 : mblume 1368 Date: 2003/09/12 11:45:00 CDT
2613 :     Tag: blume-20030912-various
2614 :     Description:
2615 :    
2616 :     - fixed bug in Real.toLargeInt
2617 :     - fixed bug in Posix.ProcEnv.times
2618 :     - changed inputLine functions to return an option
2619 :     - minor installer improvements / bugfixes
2620 :     - changed default @SMLalloc parameter for x86/celeron to 64k
2621 :    
2622 :     ----------------------------------------------------------------------
2623 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
2624 : mblume 1356 Date: 2003/09/09 22:00:00 CDT
2625 :     Tag: Release_110_43
2626 :     Description:
2627 :    
2628 :     New working release 110.43. New bootfiles.
2629 :    
2630 :     ----------------------------------------------------------------------
2631 : mblume 1368 Name: Matthias Blume (blume (at) tti - c (dot) org)
2632 : mblume 1354 Date: 2003/09/09 19:20:00 CDT
2633 :     Tag: blume-20030909-installer
2634 :     Description:
2635 :    
2636 :     Rewrote large parts of config/install.sh in SML (config/libinstall.sml).
2637 :     Modified config/install.bat to take advantage of it. Also modified
2638 :     config/install.sh (and called it config/new-install.sh) to take advantage
2639 :     of it on Unix systems. (The SML code is (supposed to be) platform-
2640 :     independent.)
2641 :    
2642 :     The installer can now install everything under Win32
2643 :     as well as under *nix as long as it compiles.
2644 :    
2645 :     Other changes:
2646 :    
2647 :     - made CML compile again under Win32
2648 :     - made eXene compile under Win32 (by providing a fake structure UnixSock
2649 :     and by using OS.Process.getEnv instead of Posix.ProcEnv.getenv)
2650 :     - fixed a bug in nowhere: it assumed that type OS.Process.status is the
2651 :     same as type int; under Win32 it isn't
2652 :     - fixed some slice-related problems in the win32-specific parts of CML
2653 :     - added a functor argument "sameVol" to os-path-fn.sml in the Basis
2654 :     (under Win32, the volume name is case-insensitive, and the
2655 :     OS.Path code compares volume names for equality)
2656 :    
2657 :     ----------------------------------------------------------------------
2658 : mblume 1368 Name: Matthias Blume (blume (at) tti - c (dot) org)
2659 : mblume 1353 Date: 2003/09/08 11:55:00 CDT
2660 :     Tag: blume-20030908-fullpath
2661 :     Description:
2662 :    
2663 :     Made Win32 version of OS.FileSys.fullPath return current directory
2664 :     when given an empty string. This is what the spec says, and incidentally,
2665 :     CM depends on it. (CM otherwise goes into an infinite loop in certain
2666 :     cases when presented with the name of a non-existing .cm file.)
2667 :    
2668 :     ----------------------------------------------------------------------
2669 : mblume 1368 Name: Matthias Blume (blume (at) tti - c (dot) org)
2670 : mblume 1350 Date: 2003/09/04 16:30:00 CDT
2671 : mblume 1352 Tag: blume-20030905-slices-etc
2672 : mblume 1350 Description:
2673 :    
2674 :     1. Changed interface to vectors and arrays in Basis to match
2675 :     (draft) Basis spec.
2676 :     2. Added signatures and implementations of slices according to
2677 :     Basis spec.
2678 :     3. Edited source code throughout the system to make it compile again
2679 :     under 1. and 2. (In some cases code had to be added to have it
2680 :     match the new signatures.)
2681 :     4. MLRISC should be backward-compatible: the copies of the originals
2682 :     of files that needed to change under 3. were retained, the .cm files
2683 :     check the compiler version number and use old versions when
2684 :     appropriate.
2685 :     5. Changed type of OS.FileSys.readDir and Posix.FileSys.readdir to
2686 :     dirstream -> string option (in accordance with Basis spec).
2687 :     6. When generating code that counts lines, ml-lex used function
2688 :     CharVector.foldli, taking advantage of its old interface.
2689 :     This has been replaced with the corresponding code from
2690 :     CharVectorSlice. (html-lex must be re-lexed!)
2691 :     7. BitArray in smlnj-lib/Util has been extended/modified to match the
2692 :     new MONO_ARRAY signature. (Do we need BitArraySlice?)
2693 :     8. Removed temporary additions (fromInternal, toInternal) from the
2694 :     (now obsolete) IntInf in smlnj-lib/Util.
2695 :     9. Cleaned up structure Byte.
2696 :     10. Added localOffset, scan, and fromString to Date (according to spec).
2697 :     Cleaned/corrected implementation of Date.
2698 :     (Still need to check for correctness; implement better canonicalizeDate.)
2699 :     11. Added "scan" to signature IEEE_REAL.
2700 :     12. Some improvements to IntInf [in particular: efficiency-hack for
2701 :     mod and rem when second operand is 2 (for parity checks).]
2702 :     13. Changed representation of type Time.time, using a single IntInf.int
2703 :     value counting microseconds. This considerably simplified the
2704 :     implementation of structure Time. We now support negative time
2705 :     values; scan and fromString handle signs.
2706 :     14. Functor PrimIO now takes two additional arguments (VectorSlice and
2707 :     ArraySlice).
2708 :    
2709 :     ----------------------------------------------------------------------
2710 : mblume 1368 Name: Matthias Blume (blume (at) tti - c (dot) org)
2711 : mblume 1347 Date: 2003/08/28 17:00:00 CDT
2712 :     Tag: blume-20030828-intinf
2713 :     Description:
2714 :    
2715 :     This is a major update which comes with a version number bump
2716 :     (110.42.99 -- yes, we are really close to 110.43 :-), NEW BOOTFILES,
2717 :     and an implementation of IntInf in the Basis.
2718 :    
2719 :     There are a fairly large number of related changes and updates throughout
2720 :     the system:
2721 :    
2722 :     Basis:
2723 :     - Implemented IntInf.
2724 :     - Made LargeInt a projection of IntInf (by filtering through INTEGER).
2725 : mblume 1348 - Added some missing Real64 operations, most notably Real.toLargeInt.
2726 : mblume 1347 - Added FixedInt as a synonym for Int32.
2727 :    
2728 :     compiler:
2729 :     * Added support for a built-in intinf type.
2730 :     - literals
2731 :     - pattern matching
2732 :     - conversion shortcuts (Int32.fromLarge o Int.toLarge etc.)
2733 :     - overloading on literals and operations
2734 :    
2735 :     This required adding a primitive type intinf, some additional
2736 :     primops, and implementations for several non-trivial intinf
2737 :     operations in Core. (The intinf type is completely abstract
2738 :     to the compiler; all operations get delegated back to the Core.)
2739 :    
2740 :     * Intinf equality is handled by polyequal. However, the compiler
2741 :     does not print its usual warning in this case (since polyequal
2742 :     is the right thing to do there).
2743 :    
2744 :     * Improved the organization of structure InlineT.
2745 :    
2746 :     * A word about conversion primops:
2747 :     If conversions involving intinf do not cancel out during
2748 : mblume 1348 CPS contract, then the compiler must insert calls to Core functions.
2749 : mblume 1347 Since all core access must be resolved already during the FLINT
2750 :     translate phase, it would be too late a the time of CPS contract
2751 : mblume 1348 to add new Core calls. For this reason, conversion primops
2752 : mblume 1347 for intinf carry two arguments: 1. the numeric argument that
2753 :     they are supposed to convert, and 2. the Core function that
2754 :     can help with this conversion if necessary. If CPS contract
2755 : mblume 1348 eliminates a primop, then the associated Core function becomes
2756 :     dead and goes away. Intinf conversion primops that do not get
2757 :     eliminated by CPS contract get rewritten into calls of their
2758 :     core functions by a separate, new phase.
2759 : mblume 1347
2760 :     interactive system:
2761 :     - Control.Print.intinfDepth controls max length of intinf constants
2762 :     being printed. (Analogous to Control.Print.stringDepth.)
2763 :     - Cleanup in printutil and pputil: got rid of unused stuff and
2764 :     duplicates; replaced some of the code with code that makes better
2765 :     use of library functionality.
2766 :    
2767 :     CM:
2768 :     Bugfix: parse-errors in init group (system/smlnj/init/init.cmi)
2769 :     are no longer silent.
2770 :    
2771 :     CKIT:
2772 :     Fixed mismatched uses of Int32 and LargeInt. I always decided
2773 :     in favor of LargeInt -- which is now the same as IntInf.
2774 :     CKIT-knowledgable people should check whether this is what's
2775 :     intended and otherwise change things back to using Int32 or
2776 :     FixedInt.
2777 :    
2778 :     Throughout the code:
2779 : mblume 1348 Started using IntInf.int literals and built-in operations
2780 : mblume 1347 (e.g., comparison with 0) where this seems appropriate.
2781 :    
2782 :    
2783 :     ----------------------------------------------------------------------
2784 : macqueen 1343 Name: Dave MacQueen (dbm@cs.uchicago.edu)
2785 :     Date: 2003/08/13 11:36:00 CDT
2786 :     Tag: dbm-20030813-mcz-merge1
2787 :     Description:
2788 :    
2789 :     Merging changes from the mcz-branch development branch into trunk.
2790 :     These changes involve replacement of the emulated old prettyprinter
2791 :     interface with direct use of the SML/NJ Lib PP library, and fixing
2792 :     of a couple of bugs (895, 1186) relating to error messages. A new
2793 :     prettyprinter for ast datatypes (Elaborator/print/ppast.{sig,sml})
2794 :     has been added.
2795 :    
2796 :     ----------------------------------------------------------------------
2797 : mblume 1368 Name: Matthias Blume (blume (at) tti - c (dot) org)
2798 : mblume 1342 Date: 2003/08/11 15:45:00 CDT
2799 :     Tag: blume-20030811-windows
2800 :     Description:
2801 :    
2802 :     Version number bumped to 110.42.9. NEW BOOTFILES!!!
2803 :    
2804 :     http://smlnj.cs.uchicago.edu/dist/working/110.42.9/
2805 :    
2806 :     This patch restores SML/NJ's ability to run under win32. There are a
2807 :     number of changes, including fixes for several bugs that had gone
2808 :     unnoticed until now:
2809 :    
2810 :     - uname "CYGWIN_NT*" is recognized as win32 (This is relevant only when
2811 :     trying to run the win32 version from within cygwin.)
2812 :    
2813 :     - There are a number of simple .bat scripts that substitute for their
2814 :     corresponding Unix shell-scripts. (See below.)
2815 :    
2816 :     - The internals of ml-build have been modified slightly. The main
2817 :     difference is that instead of calling ".link-sml" (or link-sml.bat)
2818 :     using OS.Process.system, the ML process delegates this task back
2819 :     to the script. Otherwise problems arise in mixed environments such
2820 :     as Cygwin where scripts look and work like Unix scripts, but
2821 :     where OS.Process.system cannot run them.
2822 :    
2823 :     - In CM, the srcpath pickler used native pathname syntax -- which
2824 :     is incorrect in the case of cross-compilation. The new pickle format
2825 :     is independent of platform-specific naming conventions.
2826 :    
2827 :     - Path configuration files (such as lib/pathconfig) can now choose
2828 :     between native and standard syntax. Placing a line of the form
2829 :    
2830 :     standard!
2831 :    
2832 :     into the file causes all subsequent paths to be interpreted using
2833 :     CM standard pathname syntax (= Unix conventions); a line
2834 :    
2835 :     native!
2836 :    
2837 :     switches back to native style. This was needed so that
2838 :     path config files can be written portably, see src/system/pathconfig.
2839 :    
2840 :     - Runtime system:
2841 :    
2842 :     - win32-filesys.c: get_file_time and set_file_time now
2843 :     access modification time, not creation time.
2844 :    
2845 :     - I/O code made aware of new array representation.
2846 :    
2847 :     - Bug fixes in X86.prim.masm.
2848 :    
2849 :     - src/system/makeml made aware of win32. (For use under cygwin
2850 :     and other Unix-environments for windows.)
2851 :    
2852 :     - In Basis, fixed off-by-one error in win32-io.sml (function vecF)
2853 :     which caused BinIO.inputAll to fail consistently.
2854 :    
2855 :     .bat scripts:
2856 :    
2857 :     Windows .bat scripts assume that SMLNJ_HOME is defined.
2858 :    
2859 :     - sml.bat, ml-yacc.bat, ml-lex.bat: Driver scripts for standalone
2860 :     applications (sml, ml-yacc, ml-lex).
2861 :     - ml-build.bat: analogous to ml-build.
2862 :     - config\install.bat: Analogous to config/install.sh. This requires
2863 :     that SMLNJ_HOME is set and that Microsoft Visual C is ready to use.
2864 :     (nmake etc. must be on the path, and vcvars32 must have been run.)
2865 :     Moreover, sources for ml-lex and ml-yacc need to exist under src,
2866 :     and the bootfile hierarchy must have been unpacked under
2867 :     sml.boot.x86-win32.
2868 :     The script is very primitive and does a poor job at error checking.
2869 :     It only installs the base system, ml-lex, and ml-yacc. No other
2870 :     libraries are being installed (i.e., you get only those that
2871 :     are part of the compiler.)
2872 :     - link-sml.bat: analogous to .link-sml, but not currently used
2873 :    
2874 :     Unrelated bug fixes:
2875 :    
2876 :     - ml-nlffigen now exports structures ST_* corresponding to incomplete
2877 :     types.
2878 :     - Added getDevice to PP/src/pp-debug-fn.sml. (Would not compile
2879 :     otherwise.)
2880 :    
2881 :     ----------------------------------------------------------------------
2882 : macqueen 1337 Name: Dave MacQueen (macqueen@cs.uchicago.edu)
2883 :     Date: 2003/06/17
2884 :     Tag: macqueen-20030617-bug895
2885 :     Description:
2886 :    
2887 :     Modified compiler/Elaborator/print/pptype.sml to fix bug 895.
2888 :     Tag will be used for new development branch (mcz-branch) for
2889 :     use by MacQueen, (Lucasz) Zairek, and (George) Cao at uchicago.
2890 :    
2891 :     ----------------------------------------------------------------------
2892 : mblume 1368 Name: Matthias Blume (blume (at) tti - c (dot) org)
2893 : mblume 1335 Date: 2003/05/27 16:55:00 CDT
2894 :     Tag: blume-20030527-polyeq
2895 :     Description:
2896 :    
2897 :     Tried to eliminated most cases of polymorphic equality.
2898 :    
2899 :     ----------------------------------------------------------------------
2900 : mblume 1368 Name: Matthias Blume (blume (at) tti - c (dot) org)
2901 : mblume 1334 Date: 2003/05/21 17:45:00 CDT
2902 :     Tag: blume-20030517-complete
2903 :     Description:
2904 :    
2905 :     Two changes:
2906 :    
2907 :     1. Added a flag for controlling whether non-exhaustive bindings will
2908 :     be treated as errors (default is false).
2909 :     2. Cleaned up the *entire* source tree so that CMB.make goes through
2910 :     without a single non-exhaustive match- or bind warning.
2911 :    
2912 :     ----------------------------------------------------------------------
2913 : mblume 1368 Name: Matthias Blume (blume (at) tti - c (dot) org)
2914 : mblume 1332 Date: 2003/05/17 10:20:00 CDT
2915 :     Tag: blume-20030517-absyn
2916 :     Description:
2917 :    
2918 :     1. Added cases for IF, WHILE, ANDALSO, and ORELSE to Absyn.
2919 :    
2920 :     This mainly affects the quality of error messages. However, some
2921 :     of the code is now more straightforward than before. (Treatment of
2922 :     the above four constructs in translate.sml is much simpler than
2923 :     the "macro-expansion" that was going on before. Plus, the mach-
2924 :     compiler no longer gets invoked just to be able to compile an
2925 :     if-expression.)
2926 :    
2927 :     2. The ErrorMsg.Error exception is now caught and absorbed by the
2928 :     interactive loop.
2929 :    
2930 :     ----------------------------------------------------------------------
2931 : allenleung 1327 Name: Allen Leung
2932 :     Date: 2003/05/16 13:05:00 CDT
2933 :     Tag: leunga-20030516-cygwin-runtime
2934 :     Description:
2935 :    
2936 : blume 2155 Ported the runtime system to cygwin, which uses the unix
2937 : allenleung 1327 x86-unix bin files. Missing/buggy features:
2938 :    
2939 : blume 2155 o getnetbyname, getnetbyaddr: these functions seem to be missing in
2940 : allenleung 1327 the Cygwin library.
2941 :     o Ctrl-C handling may be flaky.
2942 :     o Windows system calls and Windows I/O are not supported.
2943 :    
2944 :     A new set of binfiles is located at:
2945 :    
2946 :     http://www.dorsai.org/~leunga/boot.x86-unix.tgz
2947 :    
2948 :     This is only needed for bootstrapping the cygwin version of smlnj.
2949 :     Other x86 versions can use the existing binfiles.
2950 :    
2951 :     ----------------------------------------------------------------------
2952 : mblume 1317 Name: Matthias Blume
2953 :     Date: 2003/04/08 15:42:00 CDT
2954 :     Tag: blume-20030408-listpair
2955 :     Description:
2956 :    
2957 :     1. Added a target 'mlrisc' to installer.
2958 :    
2959 :     2. Added missing elements to structure ListPair.
2960 :    
2961 :     ----------------------------------------------------------------------
2962 : leunga 1296 Name: Allen Leung
2963 :     Date: 2003/01/07 10:40:00 EST
2964 :     Tag: leunga-20030107-int-rem
2965 :     Description:
2966 :    
2967 :     Fixed a bug in Int.rem(x,y) where y is a power of 2 on x86.
2968 :     The arguments to the SUBL instruction were swapped.
2969 :    
2970 :     ----------------------------------------------------------------------
2971 : blume 1229 Name: Matthias Blume
2972 : blume 1294 Date: 2002/12/12 16:25:00 EST
2973 :     Tag: blume-20021212-risc-ra
2974 :     Description:
2975 :    
2976 :     Fixed a serious bug in the rewrite code for FP spilling/reloading that
2977 :     sent the RA into an infinite loop when floating point registers get
2978 :     spilled. (Because of this bug, e.g., nucleic stopped compiling between
2979 :     110.37 and 110.38.)
2980 :     There was another set of potential problems related to the handling of
2981 :     MLRISC annotations (but those did not yet cause real problems, apparently).
2982 :    
2983 :     ----------------------------------------------------------------------
2984 :     Name: Matthias Blume
2985 : blume 1291 Date: 2002/12/06 22:40:00 EST
2986 :     Tag: blume-20021206-cm-fileid
2987 :     Description:
2988 :    
2989 :     Added a call of SrcPath.sync at the beginning of Parse.parse (in CM).
2990 :     This fixes the problem of CM getting confused by files that suddenly
2991 :     change their identity (e.g., by getting unlinked and recreated by some
2992 :     text editor such as vi). There might be a better/cheaper/cleaner way
2993 :     of doing this, but for now this will have to do.
2994 :    
2995 :     ----------------------------------------------------------------------
2996 :     Name: Matthias Blume
2997 : blume 1288 Date: 2002/10/28 09:50:00 EST
2998 : blume 2155 Tag: blume-20021028-typecheck
2999 : blume 1288 Description:
3000 :    
3001 :     Exported structure Typecheck from $smlnj/viscomp/core.cm.
3002 :    
3003 :     ----------------------------------------------------------------------
3004 :     Name: Matthias Blume
3005 : blume 1282 Date: 2002/10/17 09:10:00 EDT
3006 : blume 1279 Tag: Release_110_42
3007 :     Description:
3008 :    
3009 : blume 1282 In good old tradition, there has been a slight hiccup so that we have
3010 :     to patch 110.42 after the fact. The old release tag has been replaced
3011 :     (see below).
3012 :    
3013 :     The change solves a problem with two competing approaches the
3014 :     configuration problem regarding MacOS 10.1 vs. MacOS 10.2 which got in
3015 :     each other's way.
3016 :    
3017 :     This change only affects the runtime system code and the installer script.
3018 :     (No new bootfiles.)
3019 :    
3020 :     ----------------------------------------------------------------------
3021 :     Name: Matthias Blume
3022 :     Date: 2002/10/16 12:00:00 EDT
3023 :     Tag: Release_110_42_removed
3024 :     Description:
3025 :    
3026 : blume 1279 New working release. New bootfiles.
3027 :    
3028 :     ----------------------------------------------------------------------
3029 :     Name: Matthias Blume
3030 : blume 1273 Date: 2002/10/10 13:10:00 EDT
3031 :     Tag: blume-20021010-ppc-divs
3032 :     Description:
3033 :    
3034 :     The mltree operator DIVS must be implemented with an overflow check on
3035 :     the PPC because the hardware indicates divide-by-zero using "overflow" as
3036 :     well.
3037 :    
3038 :     ----------------------------------------------------------------------
3039 :     Name: Matthias Blume
3040 : blume 1261 Date: 2002/07/23 11:45:00
3041 :     Tag: blume-20020723-smlnj-home
3042 :     Description:
3043 :    
3044 :     Sml now senses the SMLNJ_HOME environment variable. If this is set,
3045 :     then the bin dir is assumed to be in $SMLNJ_HOME/bin and (unless
3046 :     CM_PATHCONFIG is also set), the path configuration file is assumed
3047 :     to be in $SMLNJ_HOME/lib/pathconfig. This way one can easily move
3048 :     the entire tree to some other place and everything will "just work".
3049 :    
3050 :     (Companion commands such as ml-build and ml-makedepend also sense this
3051 :     variable.)
3052 :    
3053 :     ----------------------------------------------------------------------
3054 :     Name: Matthias Blume
3055 : blume 1259 Date: 2002/07/12 21:19:00 EDT
3056 :     Tag: blume-20020712-liveness
3057 :     Description:
3058 :    
3059 :     Exported two useful "step" functions from liveness module (MLRISC).
3060 :    
3061 :     ----------------------------------------------------------------------
3062 :     Name: Matthias Blume
3063 : blume 1253 Date: 2002/07/05 16:00 EDT
3064 :     Tag: Release_110_41
3065 :     Description:
3066 :    
3067 :     New working release. New bootfiles.
3068 :    
3069 :     ----------------------------------------------------------------------
3070 :     Name: Matthias Blume
3071 : blume 1250 Date: 2002/07/05 10:25:00 EDT
3072 :     Tag: blume-20020705-btimp
3073 :     Description:
3074 :    
3075 :     Exported structure BTImp from $smlnj/viscomp/debugprof.cm so that
3076 :     other clients can set up backtracing support.
3077 :    
3078 :     ----------------------------------------------------------------------
3079 :     Name: Matthias Blume
3080 : blume 1249 Date: 2002/06/25 17:23:00 EDT
3081 :     Tag: blume-20020625-fpmax
3082 :     Description:
3083 :    
3084 :     Fixed a bug in translation of INLMAX (and INLMIN) for the floating-point
3085 :     case. (The sense of the isNaN test was reversed -- which made min and
3086 :     max always return their first argument.)
3087 :    
3088 :     ----------------------------------------------------------------------
3089 :     Name: Matthias Blume
3090 : blume 1240 Date: 2002/06/11
3091 :     Tag: blume-20020611-unixpath
3092 :     Description:
3093 :    
3094 :     Back-ported OS.Path.{from,to}UnixPath from idlbasis-devel branch.
3095 :    
3096 :     ----------------------------------------------------------------------
3097 :     Name: Matthias Blume
3098 : blume 1238 Date: 2002/06/10 16:35:00 EDT
3099 :     Tag: blume-20020610-ieeereal
3100 :     Description:
3101 :    
3102 :     I back-ported my implementation of IEEEReal.fromString from the
3103 :     idlbasis-devel branch so that we can test it.
3104 :    
3105 :     Another small change is that ppDec tries to give more information
3106 :     than just "<sig>" in the case of functors. However, this code is
3107 :     broken in some mysterious way if the functor's body's signature
3108 :     has not been declared by ascription but gets inferred from the
3109 :     implementation. This needs fixing...
3110 :    
3111 :     ----------------------------------------------------------------------
3112 :     Name: Matthias Blume
3113 : blume 1229 Date: 2002/05/31
3114 :     Tag: blume-20020531-btrace-mode
3115 :     Description:
3116 :    
3117 :     Resurrected SMLofNJ.Internals.BTrace.mode. (It accidentally fell by
3118 :     the wayside when I switched over to using Controls everywhere.)
3119 :    
3120 :     ----------------------------------------------------------------------
3121 : george 1222 Name: Lal George
3122 :     Date: 2002/05/23 12:21:40 EDT
3123 :     Tag: george-20020523-visual-labels
3124 :     Description:
3125 :    
3126 :     Labels are now displayed in the graphical output to make
3127 :     the fall-through and target blocks obvious.
3128 :    
3129 :     ----------------------------------------------------------------------
3130 : blume 1204 Name: Matthias Blume
3131 : blume 1220 Date: 2002/05/22 11:03:00 EDT
3132 :     Tag: blume-20020522-shrink
3133 :     Description:
3134 :    
3135 :     John tweaked yesterday's fix for 1131 to handle an out-of-memory
3136 :     situation that comes up when allocating huge arrays.
3137 :    
3138 :     ----------------------------------------------------------------------
3139 :     Name: Matthias Blume
3140 : blume 1215 Date: 2002/05/21 16:00:00 EDT
3141 :     Tag: Release_110_40
3142 :     Description:
3143 :    
3144 :     New working release (110.40). New bootfiles.
3145 :    
3146 :     [Also: John Reppy fixed GC bug 1131.]
3147 :    
3148 :     ----------------------------------------------------------------------
3149 :     Name: Matthias Blume
3150 : blume 1212 Date: 2002/05/21 12:35:00 EDT
3151 :     Tag: blume-20020521-cmdoc
3152 :     Description:
3153 :    
3154 :     CM documentation update.
3155 :    
3156 :     ----------------------------------------------------------------------
3157 :     Name: Matthias Blume
3158 : blume 1211 Date: 2002/05/21 10:55:00 EDT
3159 :     Tag: blume-20020521-misc
3160 :     Description:
3161 :    
3162 :     - John tweaked runtime to be silent on heap export (except when
3163 :     GC messages are on).
3164 :     - I added a few more things (cross-compiling versions of CMB) to
3165 :     config/preloads (as suggestions).
3166 :    
3167 :     ----------------------------------------------------------------------
3168 :     Name: Matthias Blume
3169 : blume 1208 Date: 2002/05/20 22:25:00 EDT
3170 :     Tag: blume-20020520-controls
3171 :     Description:
3172 :    
3173 :     - Added ControlUtil structure to control-lib.cm.
3174 :     - Use it throughout.
3175 :     - Used Controls facility to define MLRISC controls (as opposed to
3176 :     registering MLRISC control ref cells with Controls after the fact)
3177 :     - Fixed messed-up controls priorities.
3178 :    
3179 :     * Removed again all the stuff from config/preloads that one wouldn't
3180 :     be able to preload at the time the initial heap image is built.
3181 :     (Many libraries, e.g., CML, do not exist yet at this time. The
3182 :     only libraries that can be preloaded via config/preloads are those
3183 :     that come bundled with the bootfiles.)
3184 :    
3185 :     ----------------------------------------------------------------------
3186 :     Name: Matthias Blume
3187 : blume 1204 Date: 2002/05/20 10:59:00 EDT
3188 :     Tag: blume-20020520-preloads
3189 :     Description:
3190 :    
3191 :     Added a lot of commented-out suggestions for things to be included
3192 :     in config/preloads.
3193 :    
3194 :     ----------------------------------------------------------------------
3195 : leunga 1203 Name: Allen Leung
3196 :     Date: 2002/05/18 14:20:00 EDT
3197 :     Tag: leunga-20020518-mdl
3198 :     Description:
3199 :    
3200 :     o Made the mdl tool stuff compile and run again.
3201 :     o I've disabled all the stuff that depends on RTL specifications; they
3202 :     are all badly broken anyway.
3203 :    
3204 :     ----------------------------------------------------------------------
3205 : blume 1201 Name: Matthias Blume
3206 :     Date: 2002/05/17 16:49:00 EDT
3207 :     Tag: blume-20020517-controls
3208 :     Description:
3209 :    
3210 :     0. John Reppy made several modifications to the SML/NJ library.
3211 :     In particular, there is a shiny new controls-lib.cm.
3212 :    
3213 :     1. Pushed new controls interface through compiler so that everything
3214 :     compiles again.
3215 :    
3216 :     2. Added FormatComb and FORMAT_COMB to the CML version of the
3217 :     SML/NJ library (so that CML compiles again).
3218 :    
3219 :     3. Modified init scripts because XXX_DEFAULT environment variables
3220 :     are no longer with us. (Boot-time initialization is now done
3221 :     using the same environment variables that are also used for
3222 :     startup-time initialization of controls.)
3223 :    
3224 :     ----------------------------------------------------------------------
3225 : george 1192 Name: Lal George
3226 :     Date: 2002/05/15 09:20:10 EDT
3227 :     Tag: george-20020515-pseudo-op-decls
3228 :     Description:
3229 :    
3230 :     All pseudo-ops emitted before the first segment declaration
3231 :     such as TEXT, DATA, and BSS directives are assumed to be global
3232 :     declarations and are emitted first in the assembly file. This is
3233 :     useful in a number of situations where one has pseudo-ops that are not
3234 :     specific to any segment, and also works around the constraint that one
3235 :     cannot have client pseudo-ops in the TEXT segment.
3236 :    
3237 :     Because no segment is associated with these declarations it is
3238 :     an error to allocate any space or objects before the first segment
3239 :     directive and an exception will be raised. However, we cannot make
3240 :     this check for client pseudo-ops.
3241 :    
3242 :     These top level declarations are a field in the CFG graph_info.
3243 :     In theory you can continue to add to this field after the CFG has been
3244 :     built -- provided you know what you are doing;-)
3245 :    
3246 :     ----------------------------------------------------------------------
3247 : blume 1175 Name: Matthias Blume
3248 : blume 1190 Date: 2002/05/13 16:40:00 EDT
3249 :     Tag: blume-20020513-pp-etc
3250 :     Description:
3251 :    
3252 :     A few minor bugfixes:
3253 :    
3254 :     - Stopgap measure for bug recently reported by Elsa Gunter (ppDec).
3255 :     (Bogus printouts for redefined bindings still occur. Compiler
3256 :     bug should no longer occur now. We need to redo the prettyprinter
3257 :     from scratch.)
3258 :    
3259 :     - CM pathname printer now also adds escape sequences for ( and )
3260 :    
3261 :     - commend and docu fixes for ml-nlffi
3262 :    
3263 :     ----------------------------------------------------------------------
3264 :     Name: Matthias Blume
3265 : blume 1189 Date: 2002/05/10 16:40:00 EDT
3266 :     Tag: blume-20020510-erg-textio
3267 :     Description:
3268 :    
3269 :     Applied the following bugfix provided by Emden Gansner:
3270 :    
3271 :     Output is corrupted when outputSubstr is used rather than output.
3272 :    
3273 : blume 2155 The problem occurs when a substring
3274 : blume 1189
3275 :     ss = (s, dataStart, dataLen)
3276 :    
3277 : blume 2155 where dataStart > 0, fills a stream buffer with avail bytes left.
3278 :     avail bytes of s, starting at index dataStart, are copied into the
3279 : blume 1189 buffer, the buffer is flushed, and then the remaining dataLen-avail
3280 : blume 2155 bytes of ss are copied into the beginning of the buffer. Instead of
3281 :     starting this copy at index dataStart+avail in s, the current code
3282 : blume 1189 starts the copy at index avail.
3283 :    
3284 :     Fix:
3285 :     In text-io-fn.sml, change line 695 from
3286 :     val needsFlush = copyVec(v, avail, dataLen-avail, buf, 0)
3287 : blume 2155 to
3288 : blume 1189 val needsFlush = copyVec(v, dataStart+avail, dataLen-avail, buf, 0)
3289 :    
3290 :     ----------------------------------------------------------------------
3291 :     Name: Matthias Blume
3292 : blume 1186 Date: 2002/04/12 13:55:00 EDT
3293 :     Tag: blume-20020412-assyntax
3294 :     Description:
3295 :    
3296 :     1. Grabbed newer assyntax.h from the XFree86 project.
3297 :     2. Fiddled with how to compile X86.prim.asm without warnings.
3298 :     3. (Very) Minor cleanup in CM.
3299 :    
3300 :     ----------------------------------------------------------------------
3301 :     Name: Matthias Blume
3302 : blume 1185 Date: 2002/04/01 (no joke!) 17:07:00 EST
3303 :     Tag: blume-20020401-x86div
3304 :     Description:
3305 :    
3306 :     Added full support for div/mod/rem/quot on the x86, using the machine
3307 :     instruction's two results (without clumsily recomputing the remainder)
3308 :     directly where appropriate.
3309 :    
3310 :     Some more extensive power-of-two support was added to the x86 instruction
3311 :     selector (avoiding expensive divs, mods, and muls where they can be
3312 :     replaced with cheaper shifts and masks). However, this sort of thing
3313 :     ought to be done earlier, e.g., within the CPS optimizer so that
3314 :     all architectures benefit from it.
3315 :    
3316 :     The compiler compiles to a fixed point, but changes might be somewhat
3317 :     fragile nevertheless. Please, report any strange things that you might
3318 :     see wrt. div/mod/quot/rem...
3319 :    
3320 :     ----------------------------------------------------------------------
3321 :     Name: Matthias Blume
3322 : blume 1184 Date: 2002/03/29 17:22:00
3323 :     Tag: blume-20020329-div
3324 :     Description:
3325 :    
3326 :     Fixed my broken div/mod logic. Unfortunately, this means that the
3327 :     inline code for div/mod now has one more comparison than before.
3328 :     Fast paths (quotient > 0 or remainder = 0) are not affected, though.
3329 :     The problem was with quotient = 0, because that alone does not tell
3330 :     us which way the rounding went. One then has to look at whether
3331 :     remainder and divisor have the same sign... :(
3332 :    
3333 :     Anyway, I replaced the bootfiles with fresh ones...
3334 :    
3335 :     ----------------------------------------------------------------------
3336 :     Name: Matthias Blume
3337 : blume 1183 Date: 2002/03/29 14:10:00 EST
3338 :     Tag: blume-20020329-inlprims
3339 :     Description:
3340 :    
3341 :     NEW BOOTFILES!!! Version number bumped to 110.39.3.
3342 :    
3343 :     Primops have changed. This means that the bin/boot-file formats have
3344 :     changed as well.
3345 :    
3346 :     To make sure that there is no confusion, I made a new version.
3347 :    
3348 :    
3349 :     CHANGES:
3350 :    
3351 :     * removed REMT from mltree (remainder should never overflow).
3352 :    
3353 :     * added primops to deal with divisions of all flavors to the frontend
3354 :    
3355 :     * handled these primops all the way through so they map to their respective
3356 :     MLRISC support
3357 :    
3358 :     * used these primops in the implementation of Int, Int32, Word, Word32
3359 :    
3360 :     * removed INLDIV, INLMOD, and INLREM as they are no longer necessary
3361 :    
3362 :     * parameterized INLMIN, INLMAX, and INLABS by a numkind
3363 :    
3364 :     * translate.sml now deals with all flavors of INL{MIN,MAX,ABS}, including
3365 :     floating point
3366 :    
3367 :     * used INL{MIN,MAX,ABS} in the implementation of Int, Int32, Word, Word32,
3368 :     and Real (but Real.abs maps to a separate floating-point-only primop)
3369 :    
3370 :    
3371 :     TODO items:
3372 :    
3373 :     * Hacked Alpha32 instruction selection, disabling the selection of REMx
3374 :     instructions because the machine instruction encoder cannot handle
3375 :     them. (Hppa, PPC, and Sparc instruction selection did not handle
3376 :     REM in the first place, and REM is supported by the x86 machine coder.)
3377 :    
3378 :     * Handle DIV and MOD with DIV_TO_NEGINF directly in the x86 instruction
3379 :     selection phase. (The two can be streamlined because the hardware
3380 :     delivers both quotient and remainder at the same time anyway.)
3381 :    
3382 :     * Think about what to do with "valOf(Int32.minInt) div ~1" and friends.
3383 :     (Currently the behavior is inconsistent both across architectures and
3384 :     wrt. the draft Basis spec.)
3385 :    
3386 :     * Word8 should eventually be handled natively, too.
3387 :    
3388 :     * There seems to be one serious bug in mltree-gen.sml. It appears, though,
3389 :     as if there currently is no execution path that could trigger it in
3390 :     SML/NJ. (The assumptions underlying functions arith and promotable do not
3391 :     hold for things like multiplication and division.)
3392 :    
3393 :     ----------------------------------------------------------------------
3394 :     Name: Matthias Blume
3395 : blume 1181 Date: 2002/03/27 16:27:00 EST
3396 :     Tag: blume-20020327-mlrisc-divisions
3397 :     Description:
3398 :    
3399 :     Added support for all four division operations (ML's div, mod, quot,
3400 :     and rem) to MLRISC. In the course of doing so, I also rationalized
3401 :     the naming (no more annoying switch-around of DIV and QUOT), by
3402 :     parameterizing the operation by div_rounding_mode (which can be either
3403 :     DIV_TO_ZERO or DIV_TO_NEGINF).
3404 :    
3405 :     The generic MLTreeGen functor takes care of compiling all four
3406 :     operations down to only round-to-zero div.
3407 :    
3408 :     Missing pieces:
3409 :    
3410 :     * Doing something smarter than relying on MLTreeGen on architectures
3411 :     like, e.g., the x86 where hardware division delivers both quotient and
3412 :     remainder at the same time. With this, the implementation of the
3413 :     round-to-neginf operations could be further streamlined.
3414 :    
3415 :     * Remove inlining support for div/mod/rem from the frontend and replace it
3416 :     with primops that get carried through to the backend. Do this for all
3417 :     int and word types.
3418 :    
3419 :     ----------------------------------------------------------------------
3420 :     Name: Matthias Blume
3421 : blume 1180 Date: 2002/03/25 17:25:00 EST
3422 :     Tag: blume-20020325-divmod
3423 :     Description:
3424 :    
3425 :     I improved (hopefully without breaking them) the implementation of Int.div,
3426 :     Int.mod, and Int.rem. For this, the code in translate.sml now takes
3427 :     advantage of the following observations:
3428 :    
3429 :     Let q = x quot y r = x rem y
3430 :     d = x div y m = x mod y
3431 :    
3432 :     where "quot" is the round-to-zero version of integer division that
3433 :     hardware usually provides. Then we have:
3434 :    
3435 :     r = x - q * y where neither the * nor the - will overflow
3436 :     d = if q >= 0 orelse x = q * y then q else q - 1
3437 :     where neither the * nor the - will overflow
3438 :     m = if q >= 0 orelse r = 0 then r else r + y
3439 :     where the + will not overflow
3440 :    
3441 :     This results in substantial simplification of the generated code.
3442 :     The following table shows the number of CFG nodes and edges generated
3443 :     for
3444 :     fun f (x, y) = x OPER y
3445 :     (* with OPER \in div, mod, quot, rem *)
3446 :    
3447 :    
3448 :     OPER | nodes(old) | edges(old) | nodes(new) | edges(new)
3449 :     --------------------------------------------------------
3450 :     div | 24 | 39 | 12 | 16
3451 :     mod | 41 | 71 | 12 | 16
3452 :     quot | 8 | 10 | 8 | 10
3453 :     rem | 10 | 14 | 8 | 10
3454 :    
3455 :    
3456 :     ----------------------------------------------------------------------
3457 :     Name: Matthias Blume
3458 : blume 1179 Date: 2002/03/25 22:06:00 EST
3459 :     Tag: blume-20020325-cprotobug
3460 :     Description:
3461 :    
3462 :     Fixed a bug in cproto (c prototype decoder).
3463 :    
3464 :     ----------------------------------------------------------------------
3465 :     Name: Matthias Blume
3466 : blume 1178 Date: 2002/03/25 16:00:00 EST
3467 :     Tag: blume-20020325-raw-primops
3468 :     Description:
3469 :    
3470 :     I did some cleanup to Allen's new primop code and
3471 :     replaced yesterday's bootfiles with new ones.
3472 :     (But they are stored in the same place.)
3473 :    
3474 :     ----------------------------------------------------------------------
3475 :     Name: Matthias Blume
3476 : blume 1175 Date: 2002/03/24 22:40:00 EST
3477 :     Tag: blume-20020324-bootfiles
3478 :     Description:
3479 :    
3480 :     Made the bootfiles that Allen asked for.
3481 :    
3482 :     ----------------------------------------------------------------------
3483 : leunga 1174 Name: Allen Leung
3484 :     Date: 2002/03/23 15:50:00 EST
3485 :     Tag: leunga-20020323-flint-cps-rcc-primops
3486 :     Description:
3487 :    
3488 :     1. Changes to FLINT primops:
3489 :    
3490 :     (* make a call to a C-function;
3491 :     * The primop carries C function prototype information and specifies
3492 :     * which of its (ML-) arguments are floating point. C prototype
3493 :     * information is for use by the backend, ML information is for
3494 :     * use by the CPS converter. *)
3495 :     | RAW_CCALL of { c_proto: CTypes.c_proto,
3496 :     ml_args: ccall_type list,
3497 :     ml_res_opt: ccall_type option,
3498 :     reentrant : bool
3499 :     } option
3500 :     (* Allocate uninitialized storage on the heap.
3501 :     * The record is meant to hold short-lived C objects, i.e., they
3502 : blume 2155 * are not ML pointers. With the tag, the representation is
3503 : leunga 1174 * the same as RECORD with tag tag_raw32 (sz=4), or tag_fblock (sz=8)
3504 :     *)
3505 : blume 2155 | RAW_RECORD of {tag:bool,sz:int}
3506 : leunga 1174 and ccall_type = CCALL_INT32 | CCALL_REAL64 | CCALL_ML_PTR
3507 :    
3508 :     2. These CPS primops are now overloaded:
3509 :    
3510 :     rawload of {kind:numkind}
3511 :     rawstore of {kind:numkind}
3512 :    
3513 :     The one argument form is:
3514 :    
3515 :     rawload {kind} address
3516 :    
3517 :     The two argument form is:
3518 : blume 2155
3519 : leunga 1174 rawload {kind} [ml object, byte-offset]
3520 :    
3521 :     3. RAW_CCALL/RCC now takes two extra arguments:
3522 :    
3523 :     a. The first is whether the C call is reentrant, i.e., whether
3524 :     ML state should be saved and restored.
3525 :     b. The second argument is a string argument specifying the name of
3526 :     library and the C function.
3527 :    
3528 :     These things are currently not handled in the code generator, yet.
3529 :    
3530 :     4. In CProto,
3531 :    
3532 : blume 2155 An encoding type of "bool" means "ml object" and is mapped into
3533 :     C prototype of PTR. Note that "bool" is different than "string",
3534 : leunga 1174 even though "string" is also mapped into PTR, because "bool"
3535 :     is assigned an CPS type of BOGt, while "string" is assigned INT32t.
3536 :    
3537 :     5. Pickler/unpicker
3538 :    
3539 :     Changed to handle RAW_RECORD and newest RAW_CCALL
3540 :    
3541 :     6. MLRiscGen,
3542 :    
3543 :     1. Changed to handle the new rawload/rawstore/rawrecord operators.
3544 :     2. Code for handling C Calls has been moved to a new module CPSCCalls,
3545 :     in the file CodeGen/cpscompile/cps-c-calls.sml
3546 :    
3547 :     7. Added the conditional move operator
3548 :    
3549 :     condmove of branch
3550 :    
3551 : blume 2155 to cps. Generation of this is still buggy so it is currently
3552 : leunga 1174 disabled.
3553 : blume 2155
3554 : leunga 1174 ----------------------------------------------------------------------
3555 : george 1168 Name: Lal George
3556 : blume 2155 Date: 2002/03/22 14:18:25 EST
3557 : george 1171 Tag: george-20020322-cps-branch-prob
3558 : george 1168 Description:
3559 :    
3560 : blume 2155 Implemented the Ball-Larus branch prediction-heuristics, and
3561 : george 1168 incorporated graphical viewers for control flow graphs.
3562 :    
3563 : george 1171 Ball-Larus Heuristics:
3564 : george 1168 ---------------------
3565 :     See the file compiler/CodeGen/cpscompile/cpsBranchProb.sml.
3566 :    
3567 :     By design it uses the Dempster-Shafer theory for combining
3568 :     probabilities. For example, in the function:
3569 :    
3570 :     fun f(n,acc) = if n = 0 then acc else f(n-1, n*acc)
3571 :    
3572 :     the ball-larus heuristics predicts that the n=0 is unlikely
3573 :     (OH-heuristic), and the 'then' branch is unlikely because of the
3574 :     RH-heuristic -- giving the 'then' branch an even lower combined
3575 : george 1171 probability using the Dempster-Shafer theory.
3576 : george 1168
3577 :     Finally, John Reppy's loop analysis in MLRISC, further lowers the
3578 :     probability of the 'then' branch because of the loop in the else
3579 :     branch.
3580 :    
3581 :    
3582 :     Graphical Viewing:
3583 : blume 2155 ------------------
3584 : george 1168 I merely plugged in Allen's graphical viewers into the compiler. The
3585 :     additional code is not much. At the top level, saying:
3586 :    
3587 :     Control.MLRISC.getFlag "cfg-graphical-view" := true;
3588 :    
3589 :     will display the graphical view of the control flow graph just before
3590 :     back-patching. daVinci must be in your path for this to work. If
3591 :     daVinci is not available, then the default viewer can be changed
3592 :     using:
3593 :    
3594 :     Control.MLRISC.getString "viewer"
3595 :    
3596 :     which can be set to "dot" or "vcg" for the corresponding viewers. Of
3597 :     course, these viewers must be in your path.
3598 :    
3599 :     The above will display the compilation unit at the level of clusters,
3600 :     many of which are small, boring, and un-interesting. Also setting:
3601 :    
3602 :     Control.MLRISC.getInt "cfg-graphical-view_size"
3603 :    
3604 :     will display clusters that are larger than the value set by the above.
3605 :    
3606 : blume 2155
3607 : george 1168 ----------------------------------------------------------------------
3608 : blume 1159 Name: Matthias Blume
3609 :     Date: 2002/03/21 22:20:00 EST
3610 :     Tag: blume-20020321-kmp-bugfix
3611 :     Description:
3612 :    
3613 :     Changed the interface to the KMP routine in PreString and fixed
3614 :     a minor bug in one place where it was used.
3615 :    
3616 :     ----------------------------------------------------------------------
3617 : leunga 1156 Name: Allen Leung
3618 : leunga 1158 Date: 2002/03/21 20:30:00 EST
3619 :     Tag: leunga-20020321-cfg
3620 :     Description:
3621 :    
3622 :     Fixed a potential problem in cfg edge splitting.
3623 :    
3624 :     ----------------------------------------------------------------------
3625 :     Name: Allen Leung
3626 : leunga 1156 Date: 2002/03/21 17:15:00 EST
3627 :     Tag: leunga-20020321-x86-fp-cfg
3628 :     Description:
3629 :    
3630 : blume 2155 1. Recoded the buggy parts of x86-fp.
3631 : leunga 1156
3632 :     a. All the block reordering code has been removed.
3633 :     We now depend on the block placement phases to do this work.
3634 :    
3635 :     b. Critical edge splitting code has been simplified and moved into the
3636 :     CFG modules, as where they belong.
3637 :    
3638 :     Both of these were quite buggy and complex. The code is now much, much
3639 :     simpler.
3640 :    
3641 :     2. X86 backend.
3642 :    
3643 :     a. Added instructions for 64-bit support. Instruction selection for
3644 :     64-bit has not been committed, however, since that
3645 : blume 2155 requires changes to MLTREE which haven't been approved by
3646 : leunga 1156 Lal and John.
3647 :    
3648 : blume 2155 b. Added support for FUCOMI and FUCOMIP when generating code for
3649 : leunga 1156 PentiumPro and above. We only generate these instructions in
3650 :     the fast-fp mode.
3651 :    
3652 : blume 2155 c. Added cases for JP and JNP in X86FreqProps.
3653 :    
3654 : leunga 1156 3. CFG
3655 :    
3656 :     CFG now has a bunch of methods for edge splitting and merging.
3657 :    
3658 :     4. Machine description.
3659 :    
3660 :     John's simplification of MLTREE_BASIS.fcond broke a few machine
3661 :     description things:
3662 :    
3663 :     rtl-build.{sig,sml} and hppa.mdl fixed.
3664 :    
3665 : blume 2155 NOTE: the machine description stuff in the repository is still broken.
3666 :     Again, I can't put my fixes in because that involves
3667 :     changes to MLTREE.
3668 : leunga 1156
3669 :     ----------------------------------------------------------------------
3670 : blume 1145 Name: Matthias Blume
3671 : blume 1155 Date: 2002/03/20 15:55:00 EST
3672 :     Tag: blume-20020320-kmp
3673 :     Description:
3674 :    
3675 :     Implemented Knuth-Morris-Pratt string matching in PreString and used
3676 :     it for String.isSubstring, Substring.isSubstring, and
3677 :     Substring.position.
3678 :    
3679 :     (Might need some stress-testing. Simple examples worked fine.)
3680 :    
3681 :     ----------------------------------------------------------------------
3682 :     Name: Matthias Blume
3683 : blume 1152 Date: 2002/03/19 16:37:00 EST
3684 :     Tag: blume-20020319-witnesses
3685 :     Description:
3686 :    
3687 :     Added a structure C.W and functions convert/Ptr.convert to ml-nlffi-lib.
3688 :    
3689 :     This implements a generic mechanism for changing constness qualifiers
3690 :     anywhere within big C types without resorting to outright "casts".
3691 :     (So far, functions such as C.rw/C.ro or C.Ptr.rw/C.Ptr.ro only let you
3692 :     modify the constness at the outermost level.)
3693 :     The implementation of "convert" is based on the idea of "witness"
3694 :     values -- values that are not used by the operation but whose types
3695 :     "testify" to their applicability. On the implementation side, "convert"
3696 :     is simply a projection (returning its second curried argument). With
3697 :     cross-module inlining, it should not result in any machine code being
3698 :     generated.
3699 :    
3700 :     ----------------------------------------------------------------------
3701 :     Name: Matthias Blume
3702 : blume 1148 Date: 2002/03/15 16:40:00 EST
3703 :     Tag: blume-20020315-basis
3704 :     Description:
3705 :    
3706 :     Provided (preliminary?) implementations for
3707 :    
3708 :     {String,Substring}.{concatWith,isSuffix,isSubstring}
3709 :    
3710 :     and
3711 :    
3712 :     Substring.full
3713 :    
3714 :     Those are in the Basis spec but they were missing in SML/NJ.
3715 :    
3716 :     ----------------------------------------------------------------------
3717 :     Name: Matthias Blume
3718 : blume 1145 Date: 2002/03/14 21:30:00 EST
3719 :     Tag: blume-20020314-controls
3720 :     Description:
3721 :    
3722 :     Controls:
3723 :     ---------
3724 :    
3725 :     1. Factored out the recently-added Controls : CONTROLS stuff and put
3726 :     it into its own library $/controls-lib.cm. The source tree for
3727 :     this is under src/smlnj-lib/Controls.
3728 :    
3729 :     2. Changed the names of types and functions in this interface, so they
3730 :     make a bit more "sense":
3731 :    
3732 :     module -> registry
3733 :     'a registry -> 'a group
3734 :    
3735 :     3. The interface now deals in ref cells only. The getter/setter interface
3736 :     is (mostly) gone.
3737 :    
3738 :     4. Added a function that lets one register an already-existing ref cell.
3739 :    
3740 :     5. Made the corresponding modifications to the rest of the code so that
3741 :     everything compiles again.
3742 :    
3743 :     6. Changed the implementation of Controls.MLRISC back to something closer
3744 :     to the original. In particular, this module (and therefore MLRISC)
3745 :     does not depend on Controls. There now is some link-time code in
3746 :     int-sys.sml that registers the MLRISC controls with the Controls
3747 :     module.
3748 :    
3749 :     CM:
3750 :     ---
3751 :    
3752 :     * One can now specify the lambda-split aggressiveness in init.cmi.
3753 :    
3754 :     ----------------------------------------------------------------------
3755 : leunga 1142 Name: Allen Leung
3756 :     Date: 2002/03/13 17:30:00 EST
3757 :     Tag: leunga-20020313-x86-fp-unary
3758 :     Description:
3759 :    
3760 :     Bug fix for:
3761 :    
3762 :     > leunga@weaselbane:~/Yale/tmp/sml-dist{21} bin/sml
3763 :     > Standard ML of New Jersey v110.39.1 [FLINT v1.5], March 08, 2002
3764 :     > - fun f(x,(y,z)) = Real.~ y;
3765 :     > [autoloading]
3766 :     > [autoloading done]
3767 :     > fchsl (%eax), 184(%esp)
3768 :     > Error: MLRisc bug: X86MCEmitter.emitInstr
3769 : blume 2155 >
3770 : leunga 1142 > uncaught exception Error
3771 :     > raised at: ../MLRISC/control/mlriscErrormsg.sml:16.14-16.19
3772 :    
3773 :     The problem was that the code generator did not generate any fp registers
3774 :     in this case, and the ra didn't know that it needed to run the X86FP phase to
3775 :     translate the pseudo fp instruction. This only happened with unary fp
3776 :     operators in certain situations.
3777 :    
3778 :     ----------------------------------------------------------------------
3779 : blume 1137 Name: Matthias Blume
3780 : blume 1140 Date: 2002/03/13 14:00:00 EST
3781 :     Tag: blume-20020313-overload-etc
3782 :     Description:
3783 :    
3784 :     1. Added _overload as a synonym for overload for backward compatibility.
3785 :     (Control.overloadKW must be true for either version to be accepted.)
3786 :    
3787 :     2. Fixed bug in install script that caused more things to be installed
3788 :     than what was requested in config/targets.
3789 :    
3790 :     3. Made CM aware of the (_)overload construct so that autoloading
3791 :     works.
3792 :    
3793 :     ----------------------------------------------------------------------
3794 :     Name: Matthias Blume
3795 : blume 1138 Date: 2002/03/12 22:03:00 EST
3796 :     Tag: blume-20020312-url
3797 :     Description:
3798 :    
3799 :     Forgot to update BOOT and srcarchiveurl.
3800 :    
3801 :     ----------------------------------------------------------------------
3802 :     Name: Matthias Blume
3803 : blume 1137 Date: 2002/03/12 17:30:00 EST
3804 :     Tag: blume-20020312-version110392
3805 :     Description:
3806 :    
3807 :     Yet another version number bump (because of small changes to the
3808 :     binfile format). Version number is now 110.39.2. NEW BOOTFILES!
3809 :    
3810 :     Changes:
3811 :    
3812 :     The new pid generation scheme described a few weeks ago was overly
3813 :     complicated. I implemented a new mechanism that is simpler and
3814 :     provides a bit more "stability": Once CM has seen a compilation
3815 :     unit, it keeps its identity constant (as long as you do not delete
3816 :     those crucial CM/GUID/* files). This means that when you change
3817 : blume 1145 an interface, compile, then go back to the old interface, and
3818 : blume 1137 compile again, you arrive at the original pid.
3819 :    
3820 :     There now also is a mechanism that instructs CM to use the plain
3821 :     environment hash as a module's pid (effectively making its GUID
3822 :     the empty string). For this, "noguid" must be specified as an
3823 :     option to the .sml file in question within its .cm file.
3824 :     This is most useful for code that is being generated by tools such
3825 :     as ml-nlffigen (because during development programmers tend to
3826 :     erase the tool's entire output directory tree including CM's cached
3827 :     GUIDs). "noguid" is somewhat dangerous (since it can be used to locally
3828 :     revert to the old, broken behavior of SML/NJ, but in specific cases
3829 :     where there is no danger of interface confusion, its use is ok
3830 :     (I think).
3831 :    
3832 :     ml-nlffigen by default generates "noguid" annotations. They can be
3833 :     turned off by specifying -guid in its command line.
3834 :    
3835 :     ----------------------------------------------------------------------
3836 : george 1136 Name: Lal George
3837 :     Date: 2002/03/12 12 14:42:36 EST
3838 :     Tag: george-20020312-frequency-computation
3839 :     Description:
3840 : blume 2155
3841 : george 1136 Integrated jump chaining and static block frequency into the
3842 :     compiler. More details and numbers later.
3843 : leunga 1025
3844 : george 1003 ----------------------------------------------------------------------
3845 : george 1133 Name: Lal George
3846 :     Date: 2002/03/11 11 22:38:53 EST
3847 :     Tag: george-20020311-jump-chain-elim
3848 :     Description:
3849 :    
3850 :     Tested the jump chain elimination on all architectures (except the
3851 :     hppa). This is on by default right now and is profitable for the
3852 :     alpha and x86, however, it may not be profitable for the sparc and ppc
3853 :     when compiling the compiler.
3854 :    
3855 : blume 2155 The gc test will typically jump to a label at the end of the cluster,
3856 :     where there is another jump to an external cluster containing the actual
3857 : george 1133 code to invoke gc. This is to allow factoring of common gc invocation
3858 :     sequences. That is to say, we generate:
3859 :    
3860 :     f:
3861 :     testgc
3862 :     ja L1 % jump if above to L1
3863 :    
3864 :     L1:
3865 :     jmp L2
3866 :    
3867 :    
3868 :     After jump chain elimination the 'ja L1' instructions is converted to
3869 :     'ja L2'. On the sparc and ppc, many of the 'ja L2' instructions may end
3870 :     up being implemented in their long form (if L2 is far away) using:
3871 :    
3872 :     jbe L3 % jump if below or equal to L3
3873 :     jmp L2
3874 :     L3:
3875 :     ...
3876 :    
3877 :    
3878 :     For large compilation units L2 may be far away.
3879 :    
3880 :    
3881 :     ----------------------------------------------------------------------
3882 : blume 1132 Name: Matthias Blume
3883 :     Date: 2002/03/11 13:30:00 EST
3884 :     Tag: blume-20020311-mltreeeval
3885 :     Description:
3886 :    
3887 :     A functor parameter was missing.
3888 :    
3889 :     ----------------------------------------------------------------------
3890 : leunga 1130 Name: Allen Leung
3891 : leunga 1131 Date: 2002/03/11 10:30:00 EST
3892 : leunga 1142 Tag: leunga-20020311-runtime-string0
3893 : leunga 1131 Description:
3894 :    
3895 : blume 2155 The representation of the empty string now points to a
3896 :     legal null terminated C string instead of unit. It is now possible
3897 :     to convert an ML string into C string with InlineT.CharVector.getData.
3898 : leunga 1131 This compiles into one single machine instruction.
3899 :    
3900 :     ----------------------------------------------------------------------
3901 :     Name: Allen Leung
3902 : leunga 1130 Date: 2002/03/10 23:55:00 EST
3903 :     Tag: leunga-20020310-x86-call
3904 :     Description:
3905 :    
3906 :     Added machine generation for CALL instruction (relative displacement mode)
3907 :    
3908 :     ----------------------------------------------------------------------
3909 : blume 1128 Name: Matthias Blume
3910 :     Date: 2002/03/08 16:05:00
3911 :     Tag: blume-20020308-entrypoints
3912 :     Description:
3913 :    
3914 :     Version number bumped to 110.39.1. NEW BOOTFILES!
3915 :    
3916 :     Entrypoints: non-zero offset into a code object where execution should begin.
3917 :    
3918 :     - Added the notion of an entrypoint to CodeObj.
3919 :     - Added reading/writing of entrypoint info to Binfile.
3920 :     - Made runtime system bootloader aware of entrypoints.
3921 :     - Use the address of the label of the first function given to mlriscGen
3922 :     as the entrypoint. This address is currently always 0, but it will
3923 :     not be 0 once we turn on block placement.
3924 :     - Removed the linkage cluster code (which was The Other Way(tm) of dealing
3925 :     with entry points) from mlriscGen.
3926 :    
3927 :     ----------------------------------------------------------------------
3928 : leunga 1127 Name: Allen Leung
3929 :     Date: 2002/03/07 20:45:00 EST
3930 :     Tag: leunga-20020307-x86-cmov
3931 :     Description:
3932 :    
3933 :     Bug fixes for CMOVcc on x86.
3934 :    
3935 :     1. Added machine code generation for CMOVcc
3936 :     2. CMOVcc is now generated in preference over SETcc on PentiumPro or above.
3937 :     3. CMOVcc cannot have an immediate operand as argument.
3938 :    
3939 :     ----------------------------------------------------------------------
3940 : blume 1126 Name: Matthias Blume
3941 :     Date: 2002/03/07 16:15:00 EST
3942 :     Tag: blume-20020307-controls
3943 :     Description:
3944 :    
3945 :     This is a very large but mostly boring patch which makes (almost)
3946 :     every tuneable compiler knob (i.e., pretty much everything under
3947 :     Control.* plus a few other things) configurable via both the command
3948 :     line and environment variables in the style CM did its configuration
3949 :     until now.
3950 :    
3951 :     Try starting sml with '-h' (or, if you are brave, '-H')
3952 :    
3953 :     To this end, I added a structure Controls : CONTROLS to smlnj-lib.cm which
3954 :     implements the underlying generic mechanism.
3955 :    
3956 :     The interface to some of the existing such facilities has changed somewhat.
3957 :     For example, the MLRiscControl module now provides mkFoo instead of getFoo.
3958 :     (The getFoo interface is still there for backward-compatibility, but its
3959 :     use is deprecated.)
3960 :    
3961 :     The ml-build script passes -Cxxx=yyy command-line arguments through so
3962 :     that one can now twiddle the compiler settings when using this "batch"
3963 :     compiler.
3964 :    
3965 :     TODO items:
3966 :    
3967 :     We should go through and throw out all controls that are no longer
3968 :     connected to anything. Moreover, we should go through and provide
3969 :     meaningful (and correct!) documentation strings for those controls
3970 :     that still are connected.
3971 :    
3972 :     Currently, multiple calls to Controls.new are accepted (only the first
3973 :     has any effect). Eventually we should make sure that every control
3974 :     is being made (via Controls.new) exactly once. Future access can then
3975 :     be done using Controls.acc.
3976 :    
3977 :     Finally, it would probably be a good idea to use the getter-setter
3978 :     interface to controls rather than ref cells. For the time being, both
3979 :     styles are provided by the Controls module, but getter-setter pairs are
3980 :     better if thread-safety is of any concern because they can be wrapped.
3981 :    
3982 :     *****************************************
3983 :    
3984 :     One bug fix: The function blockPlacement in three of the MLRISC
3985 :     backpatch files used to be hard-wired to one of two possibilities at
3986 :     link time (according to the value of the placementFlag). But (I
3987 :     think) it should rather sense the flag every time.
3988 :    
3989 :     *****************************************
3990 :    
3991 :     Other assorted changes (by other people who did not supply a HISTORY entry):
3992 :    
3993 :     1. the cross-module inliner now works much better (Monnier)
3994 :     2. representation of weights, frequencies, and probabilities in MLRISC
3995 :     changed in preparation of using those for weighted block placement
3996 :     (Reppy, George)
3997 :    
3998 :     ----------------------------------------------------------------------
3999 : george 1116 Name: Lal George
4000 : george 1124 Date: 2002/03/07 14:44:24 EST 2002
4001 :     Tag: george-20020307-weighted-block-placement
4002 :    
4003 :     Tested the weighted block placement optimization on all architectures
4004 :     (except the hppa) using AMPL to generate the block and edge frequencies.
4005 :     Changes were required in the machine properties to correctly
4006 : blume 2155 categorize trap instructions. There is an MLRISC flag
4007 :     "weighted-block-placement" that can be used to enable weighted block
4008 : george 1124 placement, but this will be ineffective without block/edge
4009 :     frequencies (coming soon).
4010 :    
4011 :    
4012 :     ----------------------------------------------------------------------
4013 :     Name: Lal George
4014 : blume 2155 Date: 2002/03/05 17:24:48 EST
4015 : george 1116 Tag: george-20020305-linkage-cluster
4016 :    
4017 : george 1124 In order to support the block placement optimization, a new cluster
4018 :     is generated as the very first cluster (called the linkage cluster).
4019 : blume 2155 It contains a single jump to the 'real' entry point for the compilation
4020 : george 1124 unit. Block placement has no effect on the linkage cluster itself, but
4021 :     all the other clusters have full freedom in the manner in which they
4022 : blume 2155 reorder blocks or functions.
4023 : george 1116
4024 :     On the x86 the typical linkage code that is generated is:
4025 :     ----------------------
4026 :     .align 2
4027 :     L0:
4028 :     addl $L1-L0, 72(%esp)
4029 : george 1124 jmp L1
4030 : george 1116
4031 :    
4032 :     .align 2
4033 :     L1:
4034 :     ----------------------
4035 :    
4036 : blume 2155 72(%esp) is the memory location for the stdlink register. This
4037 : george 1116 must contain the address of the CPS function being called. In the
4038 : george 1124 above example, it contains the address of L0; before
4039 : george 1116 calling L1 (the real entry point for the compilation unit), it
4040 :     must contain the address for L1, and hence
4041 :    
4042 :     addl $L1-L0, 72(%esp)
4043 :    
4044 : george 1124 I have tested this on all architectures except the hppa.The increase
4045 :     in code size is of course negligible
4046 : george 1116
4047 :     ----------------------------------------------------------------------
4048 : leunga 1115 Name: Allen Leung
4049 :     Date: 2002/03/03 13:20:00 EST
4050 :     Tag: leunga-20020303-mlrisc-tools
4051 :    
4052 :     Added #[ ... ] expressions to mlrisc tools
4053 :    
4054 :     ----------------------------------------------------------------------
4055 : blume 1096 Name: Matthias Blume
4056 : blume 1098 Date: 2002/02/27 12:29:00 EST
4057 :     Tag: blume-20020227-cdebug
4058 :     Description:
4059 :    
4060 :     - made types in structure C and C_Debug to be equal
4061 :     - got rid of code duplication (c-int.sml vs. c-int-debug.sml)
4062 :     - there no longer is a C_Int_Debug (C_Debug is directly derived from C)
4063 :    
4064 :     ----------------------------------------------------------------------
4065 :     Name: Matthias Blume
4066 : blume 1096 Date: 2002/02/26 12:00:00 EST
4067 :     Tag: blume-20020226-ffi
4068 :     Description:
4069 :    
4070 :     1. Fixed a minor bug in CM's "noweb" tool:
4071 :     If numbering is turned off, then truly don't number (i.e., do not
4072 :     supply the -L option to noweb). The previous behavior was to supply
4073 :     -L'' -- which caused noweb to use the "default" line numbering scheme.
4074 :     Thanks to Chris Richards for pointing this out (and supplying the fix).
4075 :    
4076 :     2. Once again, I reworked some aspects of the FFI:
4077 :    
4078 :     A. The incomplete/complete type business:
4079 :    
4080 :     - Signatures POINTER_TO_INCOMPLETE_TYPE and accompanying functors are
4081 :     gone!
4082 :     - ML types representing an incomplete type are now *equal* to
4083 :     ML types representing their corresponding complete types (just like
4084 :     in C). This is still safe because ml-nlffigen will not generate
4085 :     RTTI for incomplete types, nor will it generate functions that
4086 :     require access to such RTTI. But when ML code generated from both
4087 :     incomplete and complete versions of the C type meet, the ML types
4088 :     are trivially interoperable.
4089 :    
4090 :     NOTE: These changes restore the full generality of the translation
4091 :     (which was previously lost when I eliminated functorization)!
4092 :    
4093 :     B. Enum types:
4094 :    
4095 :     - Structure C now has a type constructor "enum" that is similar to
4096 :     how the "su" constructor works. However, "enum" is not a phantom
4097 :     type because each "T enum" has values (and is isomorphic to
4098 :     MLRep.Signed.int).
4099 :     - There are generic access operations for enum objects (using
4100 :     MLRep.Signed.int).
4101 :     - ml-nlffigen will generate a structure E_foo for each "enum foo".
4102 :     * The structure contains the definition of type "mlrep" (the ML-side
4103 :     representation type of the enum). Normally, mlrep is the same
4104 :     as "MLRep.Signed.int", but if ml-nlffigen was invoked with "-ec",
4105 :     then mlrep will be defined as a datatype -- thus facilitating
4106 :     pattern matching on mlrep values.
4107 :     ("-ec" will be suppressed if there are duplicate values in an
4108 :     enumeration.)
4109 :     * Constructors ("-ec") or values (no "-ec") e_xxx of type mlrep
4110 :     will be generated for each C enum constant xxx.
4111 :     * Conversion functions m2i and i2m convert between mlrep and
4112 :     MLRep.Signed.int. (Without "-ec", these functions are identities.)
4113 :     * Coversion functions c and ml convert between mlrep and "tag enum".
4114 :     * Access functions (get/set) fetch and store mlrep values.
4115 :     - By default (unless ml-nlffigen was invoked with "-nocollect"), unnamed
4116 :     enumerations are merged into one single enumeration represented by
4117 :     structure E_'.
4118 :    
4119 :     ----------------------------------------------------------------------
4120 : leunga 1080 Name: Allen Leung
4121 : leunga 1094 Date: 2002/02/25 04:45:00 EST
4122 :     Tag: leunga-20020225-cps-spill
4123 :    
4124 :     This is a new implementation of the CPS spill phase.
4125 :     The new phase is in the new file compiler/CodeGen/cpscompile/spill-new.sml
4126 :     In case of problems, replace it with the old file spill.sml
4127 :    
4128 : blume 2155 The current compiler runs into some serious performance problems when
4129 : leunga 1094 constructing a large record. This can happen when we try to compile a
4130 :     structure with many items. Even a very simple structure like the following
4131 :     makes the compiler slow down.
4132 :    
4133 :     structure Foo = struct
4134 :     val x_1 = 0w1 : Word32.int
4135 :     val x_2 = 0w2 : Word32.int
4136 :     val x_3 = 0w3 : Word32.int
4137 :     ...
4138 :     val x_N = 0wN : Word32.int
4139 :     end
4140 :    
4141 : blume 2155 The following table shows the compile time, from N=1000 to N=4000,
4142 : leunga 1094 with the old compiler:
4143 :    
4144 :     N
4145 : blume 2155 1000 CPS 100 spill 0.04u 0.00s 0.00g
4146 :     MLRISC ra 0.06u 0.00s 0.05g
4147 : leunga 1094 (spills = 0 reloads = 0)
4148 : blume 2155 TOTAL 0.63u 0.07s 0.21g
4149 : leunga 1094
4150 : blume 2155 1100 CPS 100 spill 8.25u 0.32s 0.64g
4151 :     MLRISC ra 5.68u 0.59s 3.93g
4152 : leunga 1094 (spills = 0 reloads = 0)
4153 : blume 2155 TOTAL 14.71u 0.99s 4.81g
4154 : leunga 1094
4155 : blume 2155 1500 CPS 100 spill 58.55u 2.34s 1.74g
4156 :     MLRISC ra 5.54u 0.65s 3.91g
4157 : leunga 1094 (spills = 543 reloads = 1082)
4158 : blume 2155 TOTAL 65.40u 3.13s 6.00g
4159 : leunga 1094
4160 : blume 2155 2000 CPS 100 spill 126.69u 4.84s 3.08g
4161 :     MLRISC ra 0.80u 0.10s 0.55g
4162 : leunga 1094 (spills = 42 reloads = 84)
4163 : blume 2155 TOTAL 129.42u 5.10s 4.13g
4164 : leunga 1094
4165 : blume 2155 3000 CPS 100 spill 675.59u 19.03s 11.64g
4166 :     MLRISC ra 2.69u 0.27s 1.38g
4167 : leunga 1094 (spills = 62 reloads = 124)
4168 : blume 2155 TOTAL 682.48u 19.61s 13.99g
4169 : leunga 1094
4170 : blume 2155 4000 CPS 100 spill 2362.82u 56.28s 43.60g
4171 :     MLRISC ra 4.96u 0.27s 2.72g
4172 : leunga 1094 (spills = 85 reloads = 170)
4173 : blume 2155 TOTAL 2375.26u 57.21s 48.00g
4174 : leunga 1094
4175 : blume 2155 As you can see the old cps spill module suffers from some serious
4176 :     performance problem. But since I cannot decipher the old code fully,
4177 :     instead of patching the problems up, I'm reimplementing it
4178 :     with a different algorithm. The new code is more modular,
4179 :     smaller when compiled, and substantially faster
4180 :     (O(n log n) time and O(n) space). Timing of the new spill module:
4181 : leunga 1094
4182 : blume 2155 4000 CPS 100 spill 0.02u 0.00s 0.00g
4183 :