SCM Repository
Annotation of /sml/trunk/NOTES/HISTORY
Parent Directory
|
Revision Log
Revision 3426 - (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 : | jhr | 3426 | Here is a test entry |
17 : | |||
18 : | mrainey | 2777 | ---------------------------------------------------------------------- |
19 : | jhr | 3417 | Name John Reppy (http://cs.uchicago.edu/~jhr) |
20 : | Date 2009/07/09 | ||
21 : | Description: | ||
22 : | |||
23 : | Removed redundant implementations of various top-level operations by | ||
24 : | consolidating them in base/system/smlnj/init/pervasive.sml. This | ||
25 : | change also fixes a bug in that the top-level version of round was | ||
26 : | incorrect. | ||
27 : | |||
28 : | ---------------------------------------------------------------------- | ||
29 : | blume | 3402 | Name: Matthias Blume (http://www.tti-c.org/blume/) |
30 : | blume | 3407 | Date: 2009/06/15 11:17:00 CDT |
31 : | Description: | ||
32 : | |||
33 : | Bumped version number to 110.70. New bootfiles. | ||
34 : | |||
35 : | ---------------------------------------------------------------------- | ||
36 : | Name: Matthias Blume (http://www.tti-c.org/blume/) | ||
37 : | blume | 3402 | Date: 2009/06/12 11:02:00 CDT |
38 : | Description: | ||
39 : | |||
40 : | 1. Corrected problem in config/actions that led to the so-called | ||
41 : | "unpickling bug" which appeared in version 110.68. | ||
42 : | |||
43 : | 2. Provided fix for the 64-bit pattern match bug. | ||
44 : | |||
45 : | ---------------------------------------------------------------------- | ||
46 : | jhr | 3283 | Name John Reppy (http://cs.uchicago.edu/~jhr) |
47 : | Date 2009/03/21 | ||
48 : | Description: | ||
49 : | |||
50 : | Fixed bug in Int32.fmt when the argument was the minimum int and the | ||
51 : | radix was something other than DEC. | ||
52 : | |||
53 : | ---------------------------------------------------------------------- | ||
54 : | jhr | 3274 | Name Aaron Turon (turon@ccs.neu.edu) |
55 : | Date 2009/02/21 | ||
56 : | Description: | ||
57 : | |||
58 : | Fixed bugs in how ml-antlr parsed ML types in %tokens specifications. | ||
59 : | |||
60 : | ---------------------------------------------------------------------- | ||
61 : | jriehl | 3245 | Name: Jon Riehl (http://www.jonriehl.com/) |
62 : | jriehl | 3267 | Date: 2009/01/13 |
63 : | Description: | ||
64 : | |||
65 : | Picking up some additional fixes for 110.69, including a fix for | ||
66 : | spaces in CM file paths. | ||
67 : | |||
68 : | ---------------------------------------------------------------------- | ||
69 : | Name: Jon Riehl (http://www.jonriehl.com/) | ||
70 : | jriehl | 3245 | Date: 2008/12/22 |
71 : | Description: | ||
72 : | |||
73 : | Version bumped to 110.69. New bootfiles. | ||
74 : | |||
75 : | ---------------------------------------------------------------------- | ||
76 : | jhr | 3244 | Name: John Reppy (http://www.cs.uchicago.edu/~jhr) |
77 : | Date: 2008/12/06 | ||
78 : | Description: | ||
79 : | |||
80 : | Re-enabled some CPS optimizations (first_contract and eta). The most | ||
81 : | important effect of this change is to make uses of SMLofNJ.Cont.capture | ||
82 : | be properly tail recursive. | ||
83 : | |||
84 : | ---------------------------------------------------------------------- | ||
85 : | mrainey | 3242 | Name: Mike Rainey (http://www.cs.uchicago.edu/~mrainey) |
86 : | Date: 2008/12/03 | ||
87 : | Description: | ||
88 : | |||
89 : | New concurrency-related instructions for x86 and amd64 code | ||
90 : | generators. | ||
91 : | - PAUSE: Notify the CPU that the program is spin waiting. | ||
92 : | - MFENCE: memory fence for reads and writes. | ||
93 : | - SFENCE: memory fence for writes. | ||
94 : | - LFENCE: memory fence for reads. | ||
95 : | |||
96 : | ---------------------------------------------------------------------- | ||
97 : | jhr | 3191 | Name: John Reppy (http://www.cs.uchicago.edu/~jhr) |
98 : | jhr | 3240 | Date: 2008/12/02 |
99 : | Description: | ||
100 : | |||
101 : | jhr | 3244 | Added makefile and other support for building runtime on OpenBSD. |
102 : | jhr | 3240 | |
103 : | ---------------------------------------------------------------------- | ||
104 : | Name: John Reppy (http://www.cs.uchicago.edu/~jhr) | ||
105 : | jhr | 3191 | Date: 2008/08/13 |
106 : | Description: | ||
107 : | |||
108 : | Version bumped to 110.68. New bootfiles. | ||
109 : | |||
110 : | ---------------------------------------------------------------------- | ||
111 : | Name Aaron Turon (turon@ccs.neu.edu) | ||
112 : | adrassi | 3187 | Date 2008/08/11 |
113 : | Description: | ||
114 : | |||
115 : | Minor fix to ml-ulex backend for regexps that match any character | ||
116 : | and perform a single action. Previously the emitted code would | ||
117 : | not allow the regexps to match any character at all. | ||
118 : | |||
119 : | ---------------------------------------------------------------------- | ||
120 : | jhr | 3024 | Name: John Reppy (http://www.cs.uchicago.edu/~jhr) |
121 : | jhr | 3182 | Date: 2008/08/05 |
122 : | Description: | ||
123 : | |||
124 : | Added build support for OpenBSD (thanks to Brian O'Hanlon). | ||
125 : | |||
126 : | ---------------------------------------------------------------------- | ||
127 : | Name: John Reppy (http://www.cs.uchicago.edu/~jhr) | ||
128 : | jhr | 3106 | Date: 2008/07/12 |
129 : | Description: | ||
130 : | |||
131 : | Fixed Int64.fromString to use base-10. | ||
132 : | |||
133 : | ---------------------------------------------------------------------- | ||
134 : | Name: John Reppy (http://www.cs.uchicago.edu/~jhr) | ||
135 : | jhr | 3024 | Date: 2008/04/12 |
136 : | Description: | ||
137 : | |||
138 : | Various updates to the ml-lpt tree. The documentation has been updated; | ||
139 : | bugs in the parsing of negation and character classes in ml-ulex have | ||
140 : | been fixed; and changes have been made to make the ml-lpt tools more | ||
141 : | compatible with MLton (and other SML implementations). Thanks to | ||
142 : | Matthew Fluet and Aaron Turon for their patches. | ||
143 : | |||
144 : | ---------------------------------------------------------------------- | ||
145 : | larsberg | 2973 | Name: Lars Bergstrom (larsberg@cs.uchicago.edu) |
146 : | larsberg | 2999 | Date: 2008/20/04 |
147 : | Description: | ||
148 : | |||
149 : | Implemented timer-based profiling on Windows, with behavior as close | ||
150 : | to the *nix ITIMER-based profiling as possible. | ||
151 : | ---------------------------------------------------------------------- | ||
152 : | Name: Lars Bergstrom (larsberg@cs.uchicago.edu) | ||
153 : | larsberg | 2973 | Date: 2008/07/04 |
154 : | Description: | ||
155 : | |||
156 : | Finished off the Windows subset of the basis library. Added process | ||
157 : | support and various configuration and system identification utils. | ||
158 : | ---------------------------------------------------------------------- | ||
159 : | jhr | 2959 | Name: John Reppy (http://www.cs.uchicago.edu/~jhr) |
160 : | Date: 2008/03/18 | ||
161 : | Description: | ||
162 : | |||
163 : | Major changes to the RegExp library: see smlnj-lib/CHANGES for details. | ||
164 : | ---------------------------------------------------------------------- | ||
165 : | larsberg | 2933 | Name: Lars Bergstrom (larsberg@cs.uchicago.edu) |
166 : | larsberg | 2941 | Date: 2008/02/14 |
167 : | Description: | ||
168 : | |||
169 : | larsberg | 2943 | Added the Windows Status structure |
170 : | ---------------------------------------------------------------------- | ||
171 : | Name: Lars Bergstrom (larsberg@cs.uchicago.edu) | ||
172 : | Date: 2008/02/14 | ||
173 : | Description: | ||
174 : | |||
175 : | larsberg | 2941 | Added the Windows DDE structure |
176 : | ---------------------------------------------------------------------- | ||
177 : | Name: Lars Bergstrom (larsberg@cs.uchicago.edu) | ||
178 : | larsberg | 2936 | Date: 2008/02/05 |
179 : | Description: | ||
180 : | |||
181 : | Added the Windows Config structure | ||
182 : | ---------------------------------------------------------------------- | ||
183 : | Name: Lars Bergstrom (larsberg@cs.uchicago.edu) | ||
184 : | larsberg | 2933 | Date: 2008/01/31 |
185 : | Description: | ||
186 : | |||
187 : | Added outline of the Windows basis library and the basic registry | ||
188 : | functionality. | ||
189 : | ---------------------------------------------------------------------- | ||
190 : | mrainey | 2930 | Name: Mike Rainey (http://www.cs.uchicago.edu/~mrainey) |
191 : | Date: 2008/01/23 | ||
192 : | Description: | ||
193 : | |||
194 : | Fixed the amd64 code generator to compile with the current MLRISC. | ||
195 : | ---------------------------------------------------------------------- | ||
196 : | larsberg | 2874 | Name: Lars Bergstrom (larsberg@cs.uchicago.edu) |
197 : | larsberg | 2917 | Date: 2007/11/26 |
198 : | Description: | ||
199 : | |||
200 : | nlffi was updated to work on Windows. It needed to pass in the correct | ||
201 : | value for the name of the kernel32 DLL to obtain 'base' bindings. | ||
202 : | Additionally updated the README for the most basic nlffi sample with | ||
203 : | what you need to do on MacOSX and on Windows to make it work. | ||
204 : | ---------------------------------------------------------------------- | ||
205 : | Name: Lars Bergstrom (larsberg@cs.uchicago.edu) | ||
206 : | larsberg | 2908 | Date: 2007/11/21 |
207 : | Description: | ||
208 : | |||
209 : | Overwrite the SMLNJ_HOME environment variable on installation | ||
210 : | Properly change the package code so that subsequent version | ||
211 : | installations prompt for uninstall (instead of 'repair/remove') | ||
212 : | ---------------------------------------------------------------------- | ||
213 : | Name: Lars Bergstrom (larsberg@cs.uchicago.edu) | ||
214 : | larsberg | 2909 | Date: 2007/11/14 |
215 : | larsberg | 2874 | Description: |
216 : | |||
217 : | Cleaned up WININSTALL file for new MSI-based setup. | ||
218 : | ---------------------------------------------------------------------- | ||
219 : | blume | 2871 | Name: Matthias Blume (http://www.tti-c.org/blume/) |
220 : | Date: 2007/11/13 14:05:00 CST | ||
221 : | Description: | ||
222 : | |||
223 : | Version bumped to 110.67. New bootfiles. | ||
224 : | |||
225 : | ---------------------------------------------------------------------- | ||
226 : | jhr | 2842 | Name: John Reppy (http://www.cs.uchicago.edu/~jhr) |
227 : | jhr | 2853 | Date: 2007/11/12 |
228 : | Description: | ||
229 : | |||
230 : | Ensure that the size of the allocation space is at least 128K. | ||
231 : | ---------------------------------------------------------------------- | ||
232 : | Name: John Reppy (http://www.cs.uchicago.edu/~jhr) | ||
233 : | jhr | 2842 | Date: 2007/11/05 |
234 : | Description: | ||
235 : | |||
236 : | Fixed type error in ml-lpt library that occurs when compiling | ||
237 : | against a basis that was compiled with the USE_64_BIT_POSITIONS | ||
238 : | symbol set. | ||
239 : | (Thanks to Johannes Joemann) | ||
240 : | ---------------------------------------------------------------------- | ||
241 : | mrainey | 2840 | Name: Mike Rainey (http://www.cs.uchicago.edu/~mrainey) |
242 : | Date: 2007/11/03 | ||
243 : | Description: | ||
244 : | |||
245 : | Fixed the amd64 code generator to compile with the current MLRISC. | ||
246 : | ---------------------------------------------------------------------- | ||
247 : | larsberg | 2835 | Name: Lars Bergstrom (web page or e-mail) |
248 : | Date: 2007/11/02 | ||
249 : | Description: | ||
250 : | |||
251 : | Made sml.bat more resilient to either not having run the installer | ||
252 : | and having no SMLNJ_HOME set or having just shuffled the directory | ||
253 : | around. | ||
254 : | ---------------------------------------------------------------------- | ||
255 : | jhr | 2821 | Name: John Reppy (http://www.cs.uchicago.edu/~jhr) |
256 : | jhr | 2831 | Date: 2007/11/01 |
257 : | Description: | ||
258 : | |||
259 : | A collection of bug fixes for machine.sml in the Reactive library. | ||
260 : | (Thanks to Timothy Bourke) | ||
261 : | ---------------------------------------------------------------------- | ||
262 : | Name: John Reppy (http://www.cs.uchicago.edu/~jhr) | ||
263 : | jhr | 2821 | Date: 2007/10/28 |
264 : | Description: | ||
265 : | |||
266 : | Patches for Mac OS X 10.5 (Leopard). | ||
267 : | ---------------------------------------------------------------------- | ||
268 : | mrainey | 2809 | Name: Mike Rainey (http://www.cs.uchicago.edu/~mrainey) |
269 : | mrainey | 2813 | Date: 2007/10/28 |
270 : | Description: | ||
271 : | |||
272 : | mrainey | 2815 | Fixed some bugs in the AMD64 floating-point spilling code. |
273 : | mrainey | 2813 | ---------------------------------------------------------------------- |
274 : | Name: Mike Rainey (http://www.cs.uchicago.edu/~mrainey) | ||
275 : | mrainey | 2812 | Date: 2007/10/25 |
276 : | Description: | ||
277 : | |||
278 : | Added support for the atomic XCHG instruction. | ||
279 : | ---------------------------------------------------------------------- | ||
280 : | Name: Mike Rainey (http://www.cs.uchicago.edu/~mrainey) | ||
281 : | mrainey | 2809 | Date: 2007/10/22 |
282 : | Description: | ||
283 : | |||
284 : | Added AMD64 support for floating-point negation. | ||
285 : | ---------------------------------------------------------------------- | ||
286 : | larsberg | 2807 | Name: Lars Bergstrom (web page or e-mail) |
287 : | Date: 2007/10/22 | ||
288 : | Description: | ||
289 : | |||
290 : | Fixed ^C handling in Windows | ||
291 : | Added a Windows installer | ||
292 : | Made it possible to build for Windows on a mapped drive from Parallels | ||
293 : | ---------------------------------------------------------------------- | ||
294 : | mrainey | 2777 | Name: Mike Rainey (http://www.cs.uchicago.edu/~mrainey) |
295 : | mrainey | 2806 | Date: 2007/10/22 |
296 : | Description: | ||
297 : | |||
298 : | Added AMD64 support for the atomic fetch and add instruction. | ||
299 : | ---------------------------------------------------------------------- | ||
300 : | Name: Mike Rainey (http://www.cs.uchicago.edu/~mrainey) | ||
301 : | mrainey | 2794 | Date: 2007/10/18 |
302 : | Description: | ||
303 : | |||
304 : | mrainey | 2804 | The GAS output now favors p2align over align, since the |
305 : | former is guaranteed to be consistent over multiple architectures | ||
306 : | and the latter is not. | ||
307 : | ---------------------------------------------------------------------- | ||
308 : | Name: Mike Rainey (http://www.cs.uchicago.edu/~mrainey) | ||
309 : | Date: 2007/10/18 | ||
310 : | Description: | ||
311 : | |||
312 : | mrainey | 2794 | Fixed a bug in register spilling. |
313 : | ---------------------------------------------------------------------- | ||
314 : | Name: Mike Rainey (http://www.cs.uchicago.edu/~mrainey) | ||
315 : | mrainey | 2790 | Date: 2007/10/2 |
316 : | Description: | ||
317 : | |||
318 : | Added demo support for AMD64 for quick testing and fixed support for | ||
319 : | 64-bit label constants. | ||
320 : | ---------------------------------------------------------------------- | ||
321 : | Name: Mike Rainey (http://www.cs.uchicago.edu/~mrainey) | ||
322 : | mrainey | 2782 | Date: 2007/09/20 |
323 : | Description: | ||
324 : | |||
325 : | Added support for the MLRISC COND instruction and the x86-64 CMOVCC | ||
326 : | instruction. | ||
327 : | ---------------------------------------------------------------------- | ||
328 : | Name: Mike Rainey (http://www.cs.uchicago.edu/~mrainey) | ||
329 : | mrainey | 2777 | Date: 2007/09/17 |
330 : | Description: | ||
331 : | |||
332 : | Fixed an instruction-selection bug when loading 64-bit labels. | ||
333 : | |||
334 : | ---------------------------------------------------------------------- | ||
335 : | gkuan | 2774 | Name: George Kuan (http://www.cs.uchicago.edu/~gkuan) |
336 : | Date: 2007/09/14 | ||
337 : | Description: | ||
338 : | |||
339 : | Fixed Elaborator and Translate performance bugs | ||
340 : | |||
341 : | ---------------------------------------------------------------------- | ||
342 : | mrainey | 2770 | Name: Mike Rainey (http://www.cs.uchicago.edu/~mrainey) |
343 : | Date: 2007/09/12 | ||
344 : | Description: | ||
345 : | |||
346 : | Fixed xorl memory argument bug. | ||
347 : | |||
348 : | mblume | 1480 | ---------------------------------------------------------------------- |
349 : | gkuan | 2732 | Name: George Kuan (http://www.cs.uchicago.edu/~gkuan) |
350 : | Date: 2007/07/27 | ||
351 : | Description: | ||
352 : | |||
353 : | Added f64sgn (for Real64.signBit) as a primop defined | ||
354 : | in MLRiscGen. signBit(~0.0) not handled correctly. | ||
355 : | |||
356 : | ---------------------------------------------------------------------- | ||
357 : | jhr | 2727 | Name: John Reppy (http://www.cs.uchicago.edu/~jhr) |
358 : | Date: 2007/06/21 | ||
359 : | Description: | ||
360 : | |||
361 : | Fixed bug in {TextIO,BinIO}.StreamIO.endOfStream that would incorrectly | ||
362 : | signal end of stream. | ||
363 : | |||
364 : | ---------------------------------------------------------------------- | ||
365 : | blume | 2676 | Name: Matthias Blume (http://www.tti-c.org/blume/) |
366 : | blume | 2720 | Date: 2007/06/12 11:22:00 CDT |
367 : | Description: | ||
368 : | |||
369 : | Eliminated config/allsources. The information is now drawn directly | ||
370 : | from config/actions. | ||
371 : | |||
372 : | ---------------------------------------------------------------------- | ||
373 : | Name: Matthias Blume (http://www.tti-c.org/blume/) | ||
374 : | blume | 2700 | Date: 2007/06/07 23:30:00 CDT |
375 : | Tag: release-110.65 | ||
376 : | Description: | ||
377 : | |||
378 : | Version 110.65. NEW BOOTFILES! | ||
379 : | |||
380 : | ---------------------------------------------------------------------- | ||
381 : | Name: Matthias Blume (http://www.tti-c.org/blume/) | ||
382 : | blume | 2692 | Date: 2007/06/06 21:45:00 CDT |
383 : | Description: | ||
384 : | |||
385 : | - Aaron: fixed a number of bugs in ml-ulex. | ||
386 : | - Matthias: | ||
387 : | - added CM control "cm.force-tools"; this is false by default; | ||
388 : | when set to true, then tools like ml-yacc, ml-lex, ml-ulex, etc. | ||
389 : | will be forced to run regardless of whether or not their targets | ||
390 : | are up-to-date | ||
391 : | - changed installer code so that config/install.sh will re-build | ||
392 : | heap images for all tools even if those heap images already existed | ||
393 : | |||
394 : | ---------------------------------------------------------------------- | ||
395 : | Name: Matthias Blume (http://www.tti-c.org/blume/) | ||
396 : | blume | 2688 | Date: 2007/06/04 21:21:00 CDT |
397 : | Description: | ||
398 : | |||
399 : | After Aaron Turon's bug fix for ml-ulex (handling the ^ character | ||
400 : | in legacy mode), re-ran the lexer generator on all lex input files | ||
401 : | and committed the results. | ||
402 : | |||
403 : | This should fix the problem with ckit and nlffi that was reported | ||
404 : | by Vesa A. Norrman. | ||
405 : | |||
406 : | ---------------------------------------------------------------------- | ||
407 : | Name: Matthias Blume (http://www.tti-c.org/blume/) | ||
408 : | blume | 2676 | Date: 2007/05/31 11:40:00 CDT |
409 : | Tag: release-110.64 | ||
410 : | Description: | ||
411 : | |||
412 : | Version 110.64. NEW BOOTFILES! | ||
413 : | |||
414 : | ---------------------------------------------------------------------- | ||
415 : | dbm | 2650 | Name: Dave MacQueen (dbm@cs.uchicago.edu) |
416 : | Date: 2007/05/31 14:55:00 CDT | ||
417 : | Revision: 2650 | ||
418 : | Description: | ||
419 : | |||
420 : | 3rd merge of base from primop-branch-3 into the trunk. | ||
421 : | Additional bug fixes included: | ||
422 : | |||
423 : | * Timer.cpu_timer, etc. type printing corrected (by making | ||
424 : | Timer have opaque sig constraint in basis/Implementation/timer.sml) | ||
425 : | * Infinite loop in FLINT (tests/typing/tests/25.sml) (fix by | ||
426 : | Stefan Monnier) | ||
427 : | |||
428 : | ---------------------------------------------------------------------- | ||
429 : | blume | 2645 | Name: Matthias Blume (http://www.tti-c.org/blume/) |
430 : | Date: 2007/05/31 11:40:00 CDT | ||
431 : | Description: | ||
432 : | |||
433 : | Fixed some bugs in new Div code in FLINT/trans/translate.sml. | ||
434 : | |||
435 : | Also, changed the handling of "no core access": When translate.sml | ||
436 : | needs access to a core exception at a time when the core has not | ||
437 : | been set up yet (this only happens when compiling system/smlnj/init/*), | ||
438 : | then don't bother generating the corresponding tests. | ||
439 : | |||
440 : | dbm | 2650 | The old scheme was to generate a bogus value to be used in place of |
441 : | the exception. Unfortunately, that confuses the plambda type checker. | ||
442 : | Moreover, it does not do any good, because at runtime we don't expect | ||
443 : | such an exception to be ever raised. (The code in system/smlnj/init/* | ||
444 : | has to be written very carefully with this in mind!) | ||
445 : | blume | 2645 | |
446 : | ---------------------------------------------------------------------- | ||
447 : | mrainey | 2639 | Name: Mike Rainey (http://www.cs.uchicago.edu/~mrainey) |
448 : | Date: 2007/05/29 | ||
449 : | Description: | ||
450 : | |||
451 : | Added FSQRT instructions for the AMD64 code generator. | ||
452 : | |||
453 : | ---------------------------------------------------------------------- | ||
454 : | blume | 2626 | Name: Matthias Blume (http://www.tti-c.org/blume/) |
455 : | Date: 2007/05/29 19:00:00 CDT | ||
456 : | Description: | ||
457 : | |||
458 : | FLINT/trans/translate.sml now wraps all DIV/MOD/QUOT/REM operations | ||
459 : | with an explicit test for zero division. This should fixes several | ||
460 : | regressions and makes it possible for downstream optimization phases | ||
461 : | to treat these operations as "pure" when they are applied to unsigned | ||
462 : | operands. | ||
463 : | |||
464 : | ---------------------------------------------------------------------- | ||
465 : | mrainey | 2622 | Name: Mike Rainey (http://www.cs.uchicago.edu/~mrainey) |
466 : | Date: 2007/05/29 | ||
467 : | Description: | ||
468 : | |||
469 : | Added the new MLRISC code generator for the AMD64. This version, | ||
470 : | in contrast to the previous one, uses SSE registers and instructions | ||
471 : | for all floating-point computations. | ||
472 : | |||
473 : | ---------------------------------------------------------------------- | ||
474 : | jhr | 2613 | Name: John Reppy (http://www.cs.uchicago.edu/~jhr) |
475 : | jhr | 2617 | Date: 2007/05/29 |
476 : | jhr | 2613 | Description: |
477 : | |||
478 : | jhr | 2617 | A number of fixes related to the formatting of dates. These include fixes |
479 : | for bugs #1415 and #1416. We also now correctly handle format characters that | ||
480 : | lie outside the specified set. | ||
481 : | jhr | 2613 | |
482 : | ---------------------------------------------------------------------- | ||
483 : | blume | 2610 | Name: Matthias Blume (http://www.tti-c.org/blume/) |
484 : | Date: 2007/05/23 22:30:00 CDT | ||
485 : | Description: | ||
486 : | |||
487 : | CMB (and CM) now automatically defines the CM "preprocessor" symbol | ||
488 : | NO_PLUGINS during "makeml -rebuild" or when CM operates in "slave" mode. | ||
489 : | |||
490 : | (In addition, CMB_REBUILD_MODE is defined for makeml -rebuild, | ||
491 : | CM_SLAVE_MODE is defined in attached slaves while running CM.make | ||
492 : | or CM.rebuild, and CMB_SLAVE_MODE is defined in attached slaves | ||
493 : | while running CMB.make. The point is that the single symbol | ||
494 : | NO_PLUGINS is definde in all three cases.) | ||
495 : | |||
496 : | I changed the three locations within the sources that get compiled | ||
497 : | during CMB.make where ml-yacc or ml-lex input is processed: When | ||
498 : | NO_PLUGINS defined, then the use of the mlyacc and mllex tools | ||
499 : | is bypassed. | ||
500 : | |||
501 : | When bootstrapping new versions of the compiler, there can be | ||
502 : | situations where the plugin tools for ml-yacc and ml-lex (or ml-ulex) | ||
503 : | are not available or otherwise not operational. In this case one can | ||
504 : | manually define the NO_PLUGINS symbol prior to running CMB.make(). To | ||
505 : | do so, the following command should be issued at the interactive | ||
506 : | prompt: | ||
507 : | |||
508 : | #set (CMB.symval "NO_PLUGINS") (SOME 1); | ||
509 : | |||
510 : | A CMB.make with NO_PLUGINS defined relies on the existence of the | ||
511 : | files that normally would be generated by ml-yacc and ml-lex. | ||
512 : | (Copies of these files are in the repository.) | ||
513 : | |||
514 : | ---------------------------------------------------------------------- | ||
515 : | dbm | 2603 | Name: David MacQueen (dbm@cs.uchicago.edu) |
516 : | Date: 2007/5/23 17:23:00 CDT | ||
517 : | Revision: 2603 | ||
518 : | Description: | ||
519 : | |||
520 : | 2nd merge of base from primop-branch-3 into the trunk. | ||
521 : | Additional bug fixes included: | ||
522 : | Date.scan and Date.fromString fixed; | ||
523 : | Overloading resolution fixed and some type printing problems corrected. | ||
524 : | |||
525 : | ---------------------------------------------------------------------- | ||
526 : | blume | 2578 | Name: Matthias Blume (http://www.tti-c.org/blume/) |
527 : | blume | 2589 | Date: 2007/05/23 09:50:00 CDT |
528 : | Description: | ||
529 : | |||
530 : | Changed the installation mechanism for CM tool plugins. These are | ||
531 : | just libaries and now get installed like ordinary libaries. | ||
532 : | |||
533 : | There are now a number of new installation targets that give some | ||
534 : | fine-grain control over what classes and suffixes are known, and what | ||
535 : | they will map to. | ||
536 : | See config/targets for details. | ||
537 : | |||
538 : | The code that caused plugin installation as part of running a | ||
539 : | tool's "build" script has been removed. (The build script is for | ||
540 : | building, not for installing.) | ||
541 : | |||
542 : | ---------------------------------------------------------------------- | ||
543 : | Name: Matthias Blume (http://www.tti-c.org/blume/) | ||
544 : | blume | 2578 | Date: 2007/05/22 11:50:00 CDT |
545 : | Description: | ||
546 : | |||
547 : | Added a boolean control named cm.tolerate-tool-failures (env. variable | ||
548 : | name: CM_TOLERATE_TOOL_FAILURES). The default is false and makes | ||
549 : | CM fail if a shell tool reports a non-success exit status. If | ||
550 : | the control is set to true, then CM will press on after tool failures | ||
551 : | in the event that the target files exist (even though they are | ||
552 : | considered outdated). Turning the control to true can be useful for | ||
553 : | bootstrapping. | ||
554 : | |||
555 : | ---------------------------------------------------------------------- | ||
556 : | adrassi | 2570 | Name: Aaron Turon (http://www.cs.uchicago.edu/~adrassi) |
557 : | Date: 2007/05/19 | ||
558 : | Description: | ||
559 : | |||
560 : | Merge of ml-lpt revisions for 110.64. | ||
561 : | |||
562 : | The name of several ml-lpt-lib modules has changed: | ||
563 : | Repair => AntlrRepair | ||
564 : | StreamPos => AntlrStreamPos | ||
565 : | ErrHandlerFn => AntlrErrHandler | ||
566 : | EBNF => AntlrEBNF | ||
567 : | |||
568 : | The ml-antlr specification format has changed: declarations such as | ||
569 : | %tokens and nonterminal definitions can occur *multiple* times in the | ||
570 : | same specification. The semantics are such that each new declaration | ||
571 : | extends the previous ones. This does not apply to %start or %name, | ||
572 : | of course. | ||
573 : | |||
574 : | Importing a grammar via %import now includes *all* declarations in from | ||
575 : | the imported grammar, except for %name, %entry, and %start. Tokens | ||
576 : | and nonterminals can be dropped using the new %dropping clause of the | ||
577 : | %import directive; the separate %drop and %extend have been removed. | ||
578 : | |||
579 : | We now allow optional type annotations on nonterminals, using the %nonterms | ||
580 : | directive as in ml-yacc. | ||
581 : | |||
582 : | The refcell construct is now implemented using SML's regular reference | ||
583 : | cells, so the :== and !! notation has been deprecated. | ||
584 : | |||
585 : | The ml-antlr tool now does much more checking of specifications, and | ||
586 : | its error messages have been greatly improved. Error repair for | ||
587 : | generated parsers has been completely rewritten, and is now both | ||
588 : | much faster and more accurate. | ||
589 : | |||
590 : | ml-ulex is now more lenient with escape codes (non-SML-standard | ||
591 : | escape codes are now interpreted literally, so e.g. \| denotes "|"). | ||
592 : | Also, character classes may now include a "-" character at the beginning | ||
593 : | as is standard in most other regexp tools. | ||
594 : | |||
595 : | All of these changes are documented in the user guide, which has | ||
596 : | been updated and improved with this merge. | ||
597 : | |||
598 : | ---------------------------------------------------------------------- | ||
599 : | dbm | 2492 | Name: David MacQueen (http://www.cs.uchicago.edu/~jhr) |
600 : | Date: 2007/05/03 | ||
601 : | Revision: r2492 | ||
602 : | Description: | ||
603 : | |||
604 : | Merge of the primop3 branch (base) into the trunk to create 110.63.1. | ||
605 : | Significant changes in FLINT and the front end, mostly having to do | ||
606 : | with a reorganized system for handling primops. Various bug fixes | ||
607 : | and improvements in printing signatures. | ||
608 : | |||
609 : | ---------------------------------------------------------------------- | ||
610 : | adrassi | 2551 | Name: Aaron Turon (http://www.cs.uchicago.edu/~adrassi) |
611 : | Date: 2007/05/02 | ||
612 : | Description: | ||
613 : | |||
614 : | Preliminary commit of large ml-lpt revisions (more to come for 110.64). | ||
615 : | The ml-antlr specification format has changed: declarations such as | ||
616 : | %tokens and nonterminal definitions can occur *multiple* times in the | ||
617 : | same specification. The semantics are such that each new declaration | ||
618 : | extends the previous ones. Grammar extension constructs have also | ||
619 : | changed. We now allow type annotations on nonterminals. Finally, | ||
620 : | the refcell construct is now implemented using SML's regular reference | ||
621 : | cells, so the :== and !! notation has been deprecated. All of these | ||
622 : | changes will appear in the 110.64 user guide for ml-lpt. | ||
623 : | |||
624 : | The ml-antlr tool now does much more checking of specifications, and | ||
625 : | its error messages have been greatly improved. There has also been | ||
626 : | some work on the error repair process for generated parsers, but this | ||
627 : | will be further improved in 110.64. | ||
628 : | |||
629 : | ---------------------------------------------------------------------- | ||
630 : | dbm | 2603 | Name: David MacQueen (http://www.cs.uchicago.edu/~jhr) |
631 : | Date: 2007/05/03 | ||
632 : | Revision: r2492 | ||
633 : | Description: | ||
634 : | |||
635 : | Merge of the primop3 branch (base) into the trunk to create 110.63.1. | ||
636 : | Significant changes in FLINT and the front end, mostly having to do | ||
637 : | with a reorganized system for handling primops. Various bug fixes | ||
638 : | and improvements in printing signatures. | ||
639 : | |||
640 : | ---------------------------------------------------------------------- | ||
641 : | Name: Aaron Turon (http://www.cs.uchicago.edu/~adrassi) | ||
642 : | Date: 2007/05/02 | ||
643 : | Description: | ||
644 : | |||
645 : | Preliminary commit of large ml-lpt revisions (more to come for 110.64). | ||
646 : | The ml-antlr specification format has changed: declarations such as | ||
647 : | %tokens and nonterminal definitions can occur *multiple* times in the | ||
648 : | same specification. The semantics are such that each new declaration | ||
649 : | extends the previous ones. Grammar extension constructs have also | ||
650 : | changed. We now allow type annotations on nonterminals. Finally, | ||
651 : | the refcell construct is now implemented using SML's regular reference | ||
652 : | cells, so the :== and !! notation has been deprecated. All of these | ||
653 : | changes will appear in the 110.64 user guide for ml-lpt. | ||
654 : | |||
655 : | The ml-antlr tool now does much more checking of specifications, and | ||
656 : | its error messages have been greatly improved. There has also been | ||
657 : | some work on the error repair process for generated parsers, but this | ||
658 : | will be further improved in 110.64. | ||
659 : | |||
660 : | ---------------------------------------------------------------------- | ||
661 : | jhr | 2444 | Name: John Reppy (http://www.cs.uchicago.edu/~jhr) |
662 : | jhr | 2450 | Date: 2007/04/24 |
663 : | Description: | ||
664 : | |||
665 : | More Basis fixes: The Char.fromString (etc.) functions did not handle | ||
666 : | the "\uxxxx" escape sequence. There is still an outstanding bug with | ||
667 : | String.fromString the tail is a format escape. I added a comment to | ||
668 : | this effect in Basis/Implementation/string.sml. | ||
669 : | Thanks to Andreas Rossberg. | ||
670 : | |||
671 : | ---------------------------------------------------------------------- | ||
672 : | Name: John Reppy (http://www.cs.uchicago.edu/~jhr) | ||
673 : | Date: 2007/04/24 | ||
674 : | Description: | ||
675 : | |||
676 : | Added next function to Fifo and Queue modules in the SML/NJ Library. | ||
677 : | |||
678 : | ---------------------------------------------------------------------- | ||
679 : | Name: John Reppy (http://www.cs.uchicago.edu/~jhr) | ||
680 : | jhr | 2444 | Date: 2007/04/23 |
681 : | Description: | ||
682 : | |||
683 : | jhr | 2445 | More Basis fixes: Time.fmt dropped the leading "~" for negative time |
684 : | values that had no whole part. | ||
685 : | |||
686 : | ---------------------------------------------------------------------- | ||
687 : | Name: John Reppy (http://www.cs.uchicago.edu/~jhr) | ||
688 : | Date: 2007/04/23 | ||
689 : | Description: | ||
690 : | |||
691 : | jhr | 2444 | More Basis fixes: the OS.Path module did not include the InvalidArc |
692 : | exception and did not do sufficient argument checking. | ||
693 : | jhr | 2450 | Thanks to Stephen Weeks and Adam Chilpala. |
694 : | jhr | 2444 | |
695 : | ---------------------------------------------------------------------- | ||
696 : | mrainey | 2406 | Name: Mike Rainey (http://www.cs.uchicago.edu/~mrainey) |
697 : | mrainey | 2410 | Date: 2007/04/16 |
698 : | Description: | ||
699 : | |||
700 : | Fixed an unbound functor bug in the AMD64 CM file. | ||
701 : | |||
702 : | ---------------------------------------------------------------------- | ||
703 : | Name: Mike Rainey (http://www.cs.uchicago.edu/~mrainey) | ||
704 : | mrainey | 2406 | Date: 2007/04/12 |
705 : | Description: | ||
706 : | |||
707 : | Substantially changed the signature and implementation of AMD64 | ||
708 : | SVID. It now looks similar to the ia32 SVID, yet uses staged | ||
709 : | allocation. | ||
710 : | |||
711 : | ---------------------------------------------------------------------- | ||
712 : | jhr | 2387 | Name: John Reppy (http://www.cs.uchicago.edu/~jhr) |
713 : | Date: 2007/04/07 | ||
714 : | Description: | ||
715 : | |||
716 : | More Basis fixes: the WORD signature was missing {to,from}Large. | ||
717 : | Thanks to Andreas Rossberg. | ||
718 : | |||
719 : | ---------------------------------------------------------------------- | ||
720 : | blume | 2368 | Name: Matthias Blume (http://www.tti-c.org/blume/) |
721 : | Date: 2007/03/22 16:07:00 CDT | ||
722 : | Tag: release-110.63 | ||
723 : | Description: | ||
724 : | |||
725 : | Bumped version number to 110.63. | ||
726 : | |||
727 : | !!!NEW BOOTFILES!!! | ||
728 : | |||
729 : | ---------------------------------------------------------------------- | ||
730 : | mrainey | 2360 | Name: Mike Rainey (http://www.cs.uchicago.edu/~mrainey) |
731 : | Date: 2007/03/19 | ||
732 : | Description: | ||
733 : | |||
734 : | Fixed bogus operand sizes in AMD64 instruction spilling. | ||
735 : | |||
736 : | ---------------------------------------------------------------------- | ||
737 : | jhr | 2358 | Name: John Reppy (http://www.cs.uchicago.edu/~jhr) |
738 : | Date: 2007/03/19 | ||
739 : | Description: | ||
740 : | |||
741 : | Fixed a number of inconsistencies between the Posix.TTY structure | ||
742 : | and the Basis specification. Thanks to Adam Chilpala. | ||
743 : | |||
744 : | ---------------------------------------------------------------------- | ||
745 : | mrainey | 2345 | Name: Mike Rainey (http://www.cs.uchicago.edu/~mrainey) |
746 : | Date: 2007/02/26 | ||
747 : | Description: | ||
748 : | |||
749 : | Added preliminary support in MLRISC for Staged Allocation, a technique | ||
750 : | for specifying calling conventions (see | ||
751 : | http://www.eecs.harvard.edu/~nr/pubs/staged-abstract.html). | ||
752 : | Initially, we plan to use this code to generate C calls for the AMD64. | ||
753 : | |||
754 : | The staged allocation code base resides in MLRISC/staged-allocation, | ||
755 : | and specialized calling conventions go in MLRISC/ARCH/staged-allocation. | ||
756 : | |||
757 : | ---------------------------------------------------------------------- | ||
758 : | jhr | 2342 | Name: John Reppy (http://www.cs.uchicago.edu/~jhr) |
759 : | Date: 2007/02/20 | ||
760 : | Description: | ||
761 : | |||
762 : | Bug fix: when SaveCState was called with two values to save, a subsequent | ||
763 : | GC could cause the RestoreCState to fail because the saved state had been | ||
764 : | promoted to tagless pair. | ||
765 : | |||
766 : | ---------------------------------------------------------------------- | ||
767 : | leunga | 2340 | Name: Allen Leung (leunga (at) reservoir (dot) com) |
768 : | Date: 2007/02/20 19:08:00 EST | ||
769 : | Description: | ||
770 : | |||
771 : | Bug triggered by: | ||
772 : | |||
773 : | val a1 = Word8Array.array(a1, 0w0); | ||
774 : | val _ = Word8Array.update(a1, 0, 0w128); | ||
775 : | |||
776 : | x86MCEmitter crashed when the immediate operand to MOVB was outside | ||
777 : | of the range -128 ... 127. I've changed the code so that the range | ||
778 : | check is disabled. Only the low order 8 bits of the immediate operand | ||
779 : | jhr | 2342 | are now significant. |
780 : | leunga | 2340 | |
781 : | |||
782 : | ---------------------------------------------------------------------- | ||
783 : | blume | 2304 | Name: Matthias Blume (http://www.tti-c.org/blume/) |
784 : | blume | 2336 | Date: 2007/02/15 09:20:00 CST |
785 : | Description: | ||
786 : | |||
787 : | Eliminated any mention of lexgen. (Lexgen was an early precursor | ||
788 : | to ml-ulex.) | ||
789 : | |||
790 : | You should update your admin directory, so the shell scripts for | ||
791 : | maintaining your local copy of the repository reflect this change. | ||
792 : | |||
793 : | ---------------------------------------------------------------------- | ||
794 : | Name: Matthias Blume (http://www.tti-c.org/blume/) | ||
795 : | blume | 2333 | Date: 2007/02/14 15:45:00 CST |
796 : | Description: | ||
797 : | |||
798 : | Fixed bug in CM's parallel make facility that failed to have | ||
799 : | the master re-link modules after letting slaves compile them. | ||
800 : | |||
801 : | ---------------------------------------------------------------------- | ||
802 : | Name: Matthias Blume (http://www.tti-c.org/blume/) | ||
803 : | blume | 2331 | Date: 2007/02/12 15:10:00 CST |
804 : | Description: | ||
805 : | |||
806 : | Fixed typo in ml-build script that prevented library anchors | ||
807 : | from being registered. | ||
808 : | |||
809 : | ---------------------------------------------------------------------- | ||
810 : | Name: Matthias Blume (http://www.tti-c.org/blume/) | ||
811 : | blume | 2304 | Date: 2007/02/02 15:20:00 CST |
812 : | Tag: release-110.62 | ||
813 : | Description: | ||
814 : | |||
815 : | New working release (110.62). NEW BOOTFILES! | ||
816 : | |||
817 : | ---------------------------------------------------------------------- | ||
818 : | adrassi | 2296 | Name: Aaron Turon (http://www.cs.uchicago.edu/~adrassi) |
819 : | Date: 2007/01/31 | ||
820 : | Description: | ||
821 : | |||
822 : | Brought ml-lpt manual up-to-date with the code. Many minor | ||
823 : | improvements to the tools, and a few minor bugfixes. Prepared | ||
824 : | ml-lpt-lib for integration with new UTF8 structure. Changed | ||
825 : | the interface for creating streams in ml-ulex (we now support | ||
826 : | stream creation from several kinds of sources). Overall, the | ||
827 : | tools are now quite stable, and their interfaces are unlikely to | ||
828 : | change in a way that would break compatibility. More work, | ||
829 : | however, is needed in the documentation. | ||
830 : | |||
831 : | ---------------------------------------------------------------------- | ||
832 : | blume | 2277 | Name: Matthias Blume (http://www.tti-c.org/blume/) |
833 : | blume | 2298 | Date: 2007/02/01 10:05:00 CST |
834 : | Description: | ||
835 : | |||
836 : | Implemented library installer. Moved CM plugin code for | ||
837 : | ml-burg, ml-lex, and ml-yacc out of CM source tree and into | ||
838 : | their respective trees. Implemented CM plugin for ml-ulex | ||
839 : | and ml-antlr. Used library installer for ml-burg, ml-ulex, | ||
840 : | and ml-antlr. | ||
841 : | |||
842 : | For ml-yacc and ml-lex we continue to have permanently "plugged-in" | ||
843 : | CM tools. (It turns out to be too messy to do otherwise because | ||
844 : | there is too much code that during installation relies on the presence | ||
845 : | of these tools -- resulting in a tricky ordering problem.) | ||
846 : | |||
847 : | ---------------------------------------------------------------------- | ||
848 : | Name: Matthias Blume (http://www.tti-c.org/blume/) | ||
849 : | blume | 2277 | Date: 2007/01/30 11:05:00 CST |
850 : | Description: | ||
851 : | |||
852 : | Added SMLofNJ.shiftArgs which is like a shell's "shift" command. | ||
853 : | Modified CM's startup code to use shiftArgs as it processes | ||
854 : | command line arguments. This way, the init code in each .sml-file | ||
855 : | or library that is mentioned at the top level will see only | ||
856 : | those arguments that have not yet been processed at this point. | ||
857 : | In other words, the init code can "seize control" and process the | ||
858 : | remaining command line. | ||
859 : | |||
860 : | ---------------------------------------------------------------------- | ||
861 : | jhr | 2269 | Name: John Reppy (http://www.cs.uchicago.edu/~jhr) |
862 : | jhr | 2358 | Date: 2007/01/30 |
863 : | Description: | ||
864 : | |||
865 : | Added fromList function to the ORD_SET interface and lookup to the | ||
866 : | ORD_MAP interface. See the SML/NJ Library CHANGES file for details. | ||
867 : | |||
868 : | ---------------------------------------------------------------------- | ||
869 : | Name: John Reppy (http://www.cs.uchicago.edu/~jhr) | ||
870 : | jhr | 2269 | Date: 2007/01/28 |
871 : | Description: | ||
872 : | |||
873 : | Added the UTF8 structure and signature from the Moby compiler to | ||
874 : | the SML/NJ library (and the CML library). These modules will replace | ||
875 : | the version in the ml-lpt-lib. | ||
876 : | |||
877 : | ---------------------------------------------------------------------- | ||
878 : | blume | 2237 | Name: Matthias Blume (http://www.tti-c.org/blume/) |
879 : | blume | 2267 | Date: 2007/01/26 15:10:00 CST |
880 : | Description: | ||
881 : | |||
882 : | Added entries to handle ml-lpt-lib.cm in installer. | ||
883 : | |||
884 : | ---------------------------------------------------------------------- | ||
885 : | Name: Matthias Blume (http://www.tti-c.org/blume/) | ||
886 : | blume | 2259 | Date: 2006/12/15 11:10:00 CST |
887 : | blume | 2237 | Tag: release-110.61 |
888 : | Description: | ||
889 : | |||
890 : | New working release (110.61). NEW BOOTFILES! | ||
891 : | |||
892 : | blume | 2259 | This supersedes the pre-brown-paper-bug release (see below). |
893 : | |||
894 : | blume | 2237 | ---------------------------------------------------------------------- |
895 : | blume | 2259 | Name: Matthias Blume (http://www.tti-c.org/blume/) |
896 : | Date: 2006/12/15 11:00:00 CST | ||
897 : | Description: | ||
898 : | |||
899 : | Fixed brown-paper-bag bug with CM's pathname handling, which made | ||
900 : | installation fail under Win32. | ||
901 : | |||
902 : | ---------------------------------------------------------------------- | ||
903 : | Name: Matthias Blume (http://www.tti-c.org/blume/) | ||
904 : | Date: 2006/12/14 12:00:00 CST | ||
905 : | Description: | ||
906 : | |||
907 : | Obsolete! | ||
908 : | |||
909 : | [New working release (110.61). NEW BOOTFILES!] | ||
910 : | |||
911 : | ---------------------------------------------------------------------- | ||
912 : | jhr | 2234 | Name: John Reppy (http://cs.uchicago.edu/~jhr) |
913 : | Date: 2006/14/06 09:50:00 CST | ||
914 : | Description: | ||
915 : | |||
916 : | Fixed the code in runtime/c-libs/posix-tty/{tcgetattr,tcsetattr}.c to | ||
917 : | get the c_cc termios data copied correctly. Also moved the allocation | ||
918 : | of the string to avoid problems if it caused a GC. | ||
919 : | |||
920 : | Thanks to Timothy Bourke for the bug report and fix. | ||
921 : | |||
922 : | ---------------------------------------------------------------------- | ||
923 : | blume | 2232 | Name: Matthias Blume (http://www.tti-c.org/blume) |
924 : | Date: 2006/12/08 09:50:00 CST | ||
925 : | Description: | ||
926 : | |||
927 : | Added code to CM's "standard shell tool" implementation which | ||
928 : | causes it to tolerate (with a warning) the situation where the shell | ||
929 : | command fails (e.g., due to the shell command's non-existence) | ||
930 : | as long as all target files exist. | ||
931 : | |||
932 : | This makes it possible to, e.g., build ml-yacc from sources even if | ||
933 : | svn checkout messed up the time stamps on files in such a way that | ||
934 : | yacc.grm is younger than yacc.grm.sml or yacc.grm.sig. (Ml-yacc would | ||
935 : | be needed to re-process yacc.grm, but obviously it might not yet be | ||
936 : | available at that time.) | ||
937 : | |||
938 : | ---------------------------------------------------------------------- | ||
939 : | jhr | 2228 | Name: John Reppy (http://cs.uchicago.edu/~jhr) |
940 : | Date: 2006/12/06 12:06:00 CST | ||
941 : | Description: | ||
942 : | |||
943 : | Fixed the types of recvVecFrom, recvVecFrom', recvVecFromNB, and | ||
944 : | recvVecFromNB' in the SOCKET signature. This error is actually | ||
945 : | in the SML Basis specification too. | ||
946 : | |||
947 : | ---------------------------------------------------------------------- | ||
948 : | mblume | 1923 | Name: Matthias Blume (http://www.tti-c.org/blume/) |
949 : | blume | 2226 | Date: 2006/12/05 22:40:00 CST |
950 : | Description: | ||
951 : | |||
952 : | CM now reports undefined anchors as errors and aborts execution | ||
953 : | rather than silently pressing on using bogus values. | ||
954 : | |||
955 : | ---------------------------------------------------------------------- | ||
956 : | Name: Matthias Blume (http://www.tti-c.org/blume/) | ||
957 : | blume | 2224 | Date: 2006/11/29 10:00:00 CST |
958 : | Description: | ||
959 : | |||
960 : | Use Say.vsay for printing the "[autoloading]" message, so | ||
961 : | #set CM.Control.verbose false (or -Ccm.verbose=false) can be used | ||
962 : | to suppress them. | ||
963 : | |||
964 : | ---------------------------------------------------------------------- | ||
965 : | Name: Matthias Blume (http://www.tti-c.org/blume/) | ||
966 : | blume | 2219 | Date: 2006/11/10 11:00:00 CST |
967 : | Description: | ||
968 : | |||
969 : | Fixed bug in CM where "with:" specifications that affect compilation | ||
970 : | (as opposed to parsing) were ignored. | ||
971 : | |||
972 : | ---------------------------------------------------------------------- | ||
973 : | Name: Matthias Blume (http://www.tti-c.org/blume/) | ||
974 : | blume | 2193 | Date: 2006/11/09 14:30:00 CST |
975 : | Tag: release-110.60 | ||
976 : | Description: | ||
977 : | |||
978 : | New working release (110.60). NEW BOOTFILES! | ||
979 : | |||
980 : | This is the first subversion-hosted release. There are also changes | ||
981 : | to the directory layout. Here is a table: | ||
982 : | |||
983 : | (All pathnames are relative to the SML/NJ "root" directory.) | ||
984 : | |||
985 : | config -> config | ||
986 : | |||
987 : | src/cm -> base/cm | ||
988 : | src/compiler -> base/compiler | ||
989 : | src/runtime -> base/runtime | ||
990 : | src/system -> base/system | ||
991 : | |||
992 : | src/cm/pgraph -> pgraph | ||
993 : | |||
994 : | src/READMES -> base/READMES | ||
995 : | src/$notefile -> base/NOTES/$notefile | ||
996 : | for <notefile> in BOOT CVSNOTES CYGWININSTALL HISTORY INSTALL | ||
997 : | MACOSXINSTALL WININSTALL | ||
998 : | |||
999 : | src/smlnj-lib -> smlnj-lib | ||
1000 : | src/MLRISC -> MLRISC | ||
1001 : | ckit -> ckit | ||
1002 : | src/cml -> cml | ||
1003 : | src/eXene -> eXene | ||
1004 : | src/heap2asm -> heap2asm | ||
1005 : | src/lexgen -> lexgen | ||
1006 : | src/ml-burg -> ml-burg | ||
1007 : | src/ml-lex -> ml-lex | ||
1008 : | src/ml-yacc -> ml-yacc | ||
1009 : | src/ml-nlffi-lib -> nlffi/lib | ||
1010 : | src/ml-nlffigen -> nlffi/gen | ||
1011 : | src/smlnj-c -> smlnj-c | ||
1012 : | src/tools/TraceDebugProf -> trace-debug-profile | ||
1013 : | |||
1014 : | In addition, there is also a new | ||
1015 : | |||
1016 : | ml-lpt | ||
1017 : | |||
1018 : | directory containing two new program generator tools: ml-ulex and | ||
1019 : | ml-antlr (a lexer generator that handles unicode and an ANTRL-inspired | ||
1020 : | LL(k) parser generator). These tools are currently "beta-quality" | ||
1021 : | |||
1022 : | The latest versios of the sources can now be obtained anonymously | ||
1023 : | via subversion. For this, it is useful to first check out | ||
1024 : | |||
1025 : | svn://smlnj-gforge.cs.uchicago.edu/smlnj/admin | ||
1026 : | |||
1027 : | and put the resulting directory on your shell's PATH. This provides | ||
1028 : | access to three shell scripts: checkout-all.sh, stat-all.sh, | ||
1029 : | and refresh-all.sh. | ||
1030 : | |||
1031 : | To create a freshly checked-out copy of the sources, do | ||
1032 : | |||
1033 : | checkout-all.sh [dir] | ||
1034 : | |||
1035 : | where dir is the optional SML/NJ root directory (default is "."). | ||
1036 : | |||
1037 : | This creates the above directory layout. Each subdirectory of | ||
1038 : | the root is under individual subversion control. The stat-all.sh | ||
1039 : | and refresh-al.sh scripts apply "svn stat" or "svn update" to | ||
1040 : | each of these subtrees. | ||
1041 : | |||
1042 : | ---------------------------------------------------------------------- | ||
1043 : | Name: Matthias Blume (http://www.tti-c.org/blume/) | ||
1044 : | blume | 2155 | Date: 2006/11/02 12:20:00 CST |
1045 : | Description: | ||
1046 : | |||
1047 : | Reorganized directory layout. | ||
1048 : | |||
1049 : | This is a temporary solution, more reorganization is to come. | ||
1050 : | |||
1051 : | The basic idea is to have a number of toplevel trees, each corresponding | ||
1052 : | to a well-defined part of the overall system. Each part can be | ||
1053 : | maintained individually, even in separate source repositories, | ||
1054 : | although currently we still serve everything out of the main | ||
1055 : | smlnj-gforge tree. | ||
1056 : | |||
1057 : | Installer and and scripts have been updated to reflect the new | ||
1058 : | layout. The installer (base/src/system/smlnj/installer) is now | ||
1059 : | "scriptable" to avoid burning too much knowledge about the | ||
1060 : | layout into SML source code. The main script used by the installer | ||
1061 : | is in config/actions. | ||
1062 : | |||
1063 : | The main change is that many of the subdirectories of what used to | ||
1064 : | be known as "src" have moved to the toplevel. The "src" directory | ||
1065 : | itself has moved down into a subtree called "base". (We may eventually | ||
1066 : | get rid of the extra level of indirection represented by "src".) | ||
1067 : | |||
1068 : | The layout is now as follows: | ||
1069 : | |||
1070 : | toplevel tree name | default repository | ||
1071 : | | (using gf=svn://smlnj-gforge.cs.uchicago.edu/smlnj) | ||
1072 : | -------------------+---------------------------------------------------- | ||
1073 : | config | $gf/config/trunk | ||
1074 : | | | ||
1075 : | base | $gf/sml/trunk | ||
1076 : | | | ||
1077 : | smlnj-lib | $gf/smlnj-lib/trunk | ||
1078 : | MLRISC | $gf/MLRISC/trunk | ||
1079 : | | | ||
1080 : | ml-yacc | $gf/ml-yacc/trunk | ||
1081 : | ml-lex | $gf/ml-lex/trunk | ||
1082 : | ml-burg | $gf/ml-burg/trunk | ||
1083 : | lexgen | $gf/lexgen/trunk | ||
1084 : | heap2asm | $gf/heap2asm/trunk | ||
1085 : | | | ||
1086 : | cml | $gf/cml/trunk | ||
1087 : | eXene | $gf/eXene/trunk | ||
1088 : | ckit | $gf/ckit/trunk | ||
1089 : | nlffi | $gf/nlffi/trunk | ||
1090 : | | | ||
1091 : | smlnj-c | $gf/smlnj-c/trunk | ||
1092 : | |||
1093 : | In $gf/admin there are a few useful shell scripts for checking out | ||
1094 : | and maintaining the entire collection of trees: | ||
1095 : | |||
1096 : | admin/checkout-all.sh [dir] | ||
1097 : | -- optionally creates dir and checks out all trees from their | ||
1098 : | default repositories; if dir is missing, checkout into . | ||
1099 : | admin/refresh-all.sh [dir] | ||
1100 : | -- looks at all trees (from the above list) in dir (default: .) | ||
1101 : | and runs "svn update" if the tree exists and is under | ||
1102 : | subversion control; | ||
1103 : | non-existing or non-subversion trees are skipped | ||
1104 : | admin/stat-all.sh [dir] | ||
1105 : | -- like refresh-all.sh, but runs "svn stat" instead | ||
1106 : | |||
1107 : | ---------------------------------------------------------------------- | ||
1108 : | Name: Matthias Blume (http://www.tti-c.org/blume/) | ||
1109 : | Date: 2006/10/05 11:30:00 CDT | ||
1110 : | mblume | 2066 | Tag: blume-20061005-merge-amd64 |
1111 : | Description: | ||
1112 : | |||
1113 : | Merged code for AMD64 backend (Mike Rainey's work). | ||
1114 : | Everything is hooked up but untested. | ||
1115 : | |||
1116 : | ---------------------------------------------------------------------- | ||
1117 : | Name: Matthias Blume (http://www.tti-c.org/blume/) | ||
1118 : | mblume | 1931 | Date: 2006/05/17 14:00:00 CST |
1119 : | Tag: Release_110_59 | ||
1120 : | Description: | ||
1121 : | |||
1122 : | blume | 2193 | New working version (110.59). NEW BOOTFILES! |
1123 : | mblume | 1931 | |
1124 : | I am freezing 110.59. Changes other than the version-number increase: | ||
1125 : | |||
1126 : | eXene: | ||
1127 : | |||
1128 : | * committed changes to eXene from Alley Stoughton: | ||
1129 : | "fixed bugs in X authorization and resource handling, as well | ||
1130 : | as in the pile and viewport widgets" | ||
1131 : | |||
1132 : | Runtime: | ||
1133 : | * fixed linking problem with NetBSD 3.x. | ||
1134 : | |||
1135 : | Lexgen: | ||
1136 : | * lexgen tool handles non-ascii characters in 7-bit mode the same | ||
1137 : | way that ml-lex does | ||
1138 : | |||
1139 : | * lexgen propagates exceptions the same way that ml-lex does | ||
1140 : | |||
1141 : | CML: | ||
1142 : | * Fixed a bug in the SyncVar polling functions (iGetPoll, | ||
1143 : | mTakePoll, and mGetPoll) that could lead to livelock. | ||
1144 : | |||
1145 : | ---------------------------------------------------------------------- | ||
1146 : | Name: Matthias Blume (http://www.tti-c.org/blume/) | ||
1147 : | mblume | 1926 | Date: 2006/05/12 15:20:00 CST |
1148 : | Tag: blume-20060515-makedepend | ||
1149 : | Description: | ||
1150 : | |||
1151 : | Implemented ml-makedepend (i.e., CM.sources) in a better (more robust) | ||
1152 : | way. This should hopefully fix the ml-makedepend problem permanently. | ||
1153 : | |||
1154 : | ---------------------------------------------------------------------- | ||
1155 : | Name: Matthias Blume (http://www.tti-c.org/blume/) | ||
1156 : | mblume | 1925 | Date: 2006/05/12 15:15:00 CST |
1157 : | Tag: blume-20060512-makedepend | ||
1158 : | Description: | ||
1159 : | |||
1160 : | Fixed long-standing bug with ml-makedepend where it would output a | ||
1161 : | spurious dependency to a non-existing file. (This is a simple fix. | ||
1162 : | It might need further looking into.) | ||
1163 : | |||
1164 : | ---------------------------------------------------------------------- | ||
1165 : | Name: Matthias Blume (http://www.tti-c.org/blume/) | ||
1166 : | mblume | 1923 | Date: 2006/04/20 10:30:00 CDT |
1167 : | Tag: blume-20060420-heap2exec | ||
1168 : | Description: | ||
1169 : | |||
1170 : | Committed patches received from Johannes 5 Joemann (joemann@befree.free.de) | ||
1171 : | that enable heap2exec under Linux and FreeBSD. | ||
1172 : | |||
1173 : | ---------------------------------------------------------------------- | ||
1174 : | jhr | 1921 | Name: John Reppy |
1175 : | Date: 2006/04/14 | ||
1176 : | blume | 2155 | Tag: |
1177 : | jhr | 1921 | Description: |
1178 : | |||
1179 : | MLRISC changes: | ||
1180 : | 1) renamed GAS_PSEUDO_OPS to AS_PSEUDO_OPS and put it in its own file. | ||
1181 : | 2) added support for NOTB and XORB operators in pseudo-op expressions | ||
1182 : | 3) added DarwinPseudoOp functor that supports Darwin's assembler syntax. | ||
1183 : | 4) added support for 64-bit integer literals | ||
1184 : | |||
1185 : | ---------------------------------------------------------------------- | ||
1186 : | jhr | 1922 | Name: Matthias Blume (http://www.tti-c.org/blume/) |
1187 : | mblume | 1907 | Date: yyyy/mm/dd |
1188 : | Tag: Release_110_58 | ||
1189 : | Description: | ||
1190 : | |||
1191 : | New working version (110.58). NEW BOOTFILES! | ||
1192 : | |||
1193 : | ---------------------------------------------------------------------- | ||
1194 : | mblume | 1896 | Name: Matthias Blume (http://www.tti-c.org/blume/) |
1195 : | mblume | 1905 | Date: 2006/03/01 22:15:00 CST |
1196 : | Tag: blume-20060301-lexgen | ||
1197 : | Description: | ||
1198 : | |||
1199 : | Incorporated several bugfixes to lexgen. Compiler now compiles to | ||
1200 : | fixpoint when using lexgen instead of ml-lex. | ||
1201 : | |||
1202 : | ---------------------------------------------------------------------- | ||
1203 : | Name: Matthias Blume (http://www.tti-c.org/blume/) | ||
1204 : | mblume | 1902 | Date: 2006/02/28 22:45:00 CST |
1205 : | Tag: blume-20060228-lexgen | ||
1206 : | Description: | ||
1207 : | |||
1208 : | Removed ml-flex and added lexgen instead, using Aaron Turon's newly | ||
1209 : | provided tarball. The generated lexers still have problems. | ||
1210 : | |||
1211 : | ---------------------------------------------------------------------- | ||
1212 : | Name: Matthias Blume (http://www.tti-c.org/blume/) | ||
1213 : | mblume | 1898 | Date: 2006/02/26 23:30:00 CST |
1214 : | Tag: blume-20060226-ml-flex | ||
1215 : | Description: | ||
1216 : | |||
1217 : | Removed ml-flex' dependency on regexp-lib.cm. Turned installation | ||
1218 : | of ml-flex on by default. | ||
1219 : | |||
1220 : | ---------------------------------------------------------------------- | ||
1221 : | Name: Matthias Blume (http://www.tti-c.org/blume/) | ||
1222 : | mblume | 1896 | Date: 2006/02/24 16:30:00 CST |
1223 : | Tag: blume-20060224-ml-flex | ||
1224 : | Description: | ||
1225 : | |||
1226 : | Added ml-flex sources. Partially integrated, but should not be | ||
1227 : | turned on yet! (Read: leave it commented-out in config/targets!) | ||
1228 : | |||
1229 : | ---------------------------------------------------------------------- | ||
1230 : | jhr | 1893 | Name: John Reppy (http://www.cs.uchicago.edu/~jhr) |
1231 : | Date: 2006/02/23 | ||
1232 : | Tag: jhr-20060223-x86-64 | ||
1233 : | Description: | ||
1234 : | Changes to support building on x86-64 systems (using the 32-bit mode). | ||
1235 : | Also cleaned up signal handling on Linux. Support for pre-2.2 Linux | ||
1236 : | kernels dropped. | ||
1237 : | |||
1238 : | ---------------------------------------------------------------------- | ||
1239 : | allenleung | 1890 | Name: Allen Leung (leunga (at) reservoir (dot) com) |
1240 : | Date: 2006/02/22 15:07:00 EST | ||
1241 : | Tag: leunga-20060222-mdl-gen | ||
1242 : | Description: | ||
1243 : | |||
1244 : | CM has changed. Updated the script for rebuilding the MLRISC generated | ||
1245 : | files. | ||
1246 : | |||
1247 : | ---------------------------------------------------------------------- | ||
1248 : | mblume | 1886 | Name: Matthias Blume (http://www.tti-c.org/blume/) |
1249 : | Date: 2006/02/14 17:30:00 CST | ||
1250 : | Tag: blume-20060214-intel-mac | ||
1251 : | Description: | ||
1252 : | |||
1253 : | Hooked code for Darwin-specific Intel ABI into main compiler. | ||
1254 : | (This is a temporary solution which relies on the fact that the | ||
1255 : | compiler itself does not use NLFFI. Eventually we need to divorce | ||
1256 : | intel mac from generic x86 unix code and make separate sets of binaries.) | ||
1257 : | |||
1258 : | ---------------------------------------------------------------------- | ||
1259 : | jhr | 1882 | Name: John Reppy (http://www.cs.uchicago.edu/~jhr) |
1260 : | jhr | 1885 | Date: 2006/02/13 |
1261 : | Tag: jhr-20060213-intel-mac | ||
1262 : | Description: | ||
1263 : | Changed MLRISC x86 CCalls for partial support of MacOS X ABI. | ||
1264 : | |||
1265 : | ---------------------------------------------------------------------- | ||
1266 : | Name: John Reppy (http://www.cs.uchicago.edu/~jhr) | ||
1267 : | jhr | 1882 | Date: 2006/02/06 |
1268 : | Tag: jhr-20060206-intel-mac | ||
1269 : | Description: | ||
1270 : | Changes to support MacOS X on Intel hardware. The C-calls support in | ||
1271 : | MLRISC must also be updated to support the MacOS X ABI. | ||
1272 : | |||
1273 : | ---------------------------------------------------------------------- | ||
1274 : | mblume | 1863 | Name: Matthias Blume (http://www.tti-c.org/blume/) |
1275 : | mblume | 1879 | Date: 2005/12/16 00:15:00 CST |
1276 : | Tag: blume-20051216-cm | ||
1277 : | Description: | ||
1278 : | |||
1279 : | Improved error reporting and handling in CM. | ||
1280 : | |||
1281 : | ---------------------------------------------------------------------- | ||
1282 : | Name: Matthias Blume (http://www.tti-c.org/blume/) | ||
1283 : | mblume | 1878 | Date: 2005/11/21 14:00:00 CST |
1284 : | Tag: blume-20051121-html-lib | ||
1285 : | Description: | ||
1286 : | |||
1287 : | Enabling $/html-lib.cm in config/preloads did not work. This is fixed | ||
1288 : | now. (Since the anchor mapping for html-lib.cm is not yet in effect | ||
1289 : | blume | 2155 | at the time when config/preloads is processed, the library has to |
1290 : | mblume | 1878 | be referred to by another name. In 110.57 this would be |
1291 : | $SMLNJ-LIB/HTML/html-lib.cm. I arranged for $smlnj/smlnj-lib/html-lib.cm | ||
1292 : | to be valid as well.) | ||
1293 : | |||
1294 : | Thanks to Todd Wilson (Fresno) for alerting me to this issue. | ||
1295 : | |||
1296 : | ---------------------------------------------------------------------- | ||
1297 : | Name: Matthias Blume (http://www.tti-c.org/blume/) | ||
1298 : | mblume | 1874 | Date: 2005/11/19 13:45:00 CST |
1299 : | Tag: Release_110_57 | ||
1300 : | Description: | ||
1301 : | |||
1302 : | Moved release tag for 110.57. | ||
1303 : | |||
1304 : | ---------------------------------------------------------------------- | ||
1305 : | Name: Matthias Blume (http://www.tti-c.org/blume/) | ||
1306 : | mblume | 1873 | Date: 2005/11/19 23:35:00 CST |
1307 : | Tag: blume-20051119-fish | ||
1308 : | Description: | ||
1309 : | |||
1310 : | Fixed a problem in config/install.sh where it tries to "fish" the | ||
1311 : | name of the CM metadata directory from the wrong place (because | ||
1312 : | the physical location of basis.cm has changed). Also, corrected the | ||
1313 : | path anchor for $/html-lib.cm. (Thanks to M. Fluet for pointing out | ||
1314 : | these problems.) | ||
1315 : | |||
1316 : | ---------------------------------------------------------------------- | ||
1317 : | Name: Matthias Blume (http://www.tti-c.org/blume/) | ||
1318 : | mblume | 1867 | Date: 2005/11/16 14:30:00 CST |
1319 : | mblume | 1874 | Tag: blume-orig-110_57 |
1320 : | mblume | 1867 | Description: |
1321 : | |||
1322 : | New working version (110.57). NEW BOOTFILES! | ||
1323 : | |||
1324 : | mblume | 1874 | UPDATE: The release tag has been moved to include a fix for a brown-paper- |
1325 : | bag glitch (see above). | ||
1326 : | |||
1327 : | mblume | 1867 | ---------------------------------------------------------------------- |
1328 : | Name: Matthias Blume (http://www.tti-c.org/blume/) | ||
1329 : | mblume | 1866 | Date: 2005/11/16 10:25:00 CST |
1330 : | Tag: blume-22051116-btrace | ||
1331 : | Description: | ||
1332 : | |||
1333 : | Fixed problem with bogus exception message when using back-trace | ||
1334 : | facility. | ||
1335 : | |||
1336 : | ---------------------------------------------------------------------- | ||
1337 : | Name: Matthias Blume (http://www.tti-c.org/blume/) | ||
1338 : | mblume | 1865 | Date: 2005/11/15 22:35:00 CST |
1339 : | Tag: blume-22051115-array2copy | ||
1340 : | Description: | ||
1341 : | |||
1342 : | Added simple implementation of Array2.copy. (Warning: mostly untested.) | ||
1343 : | |||
1344 : | ---------------------------------------------------------------------- | ||
1345 : | Name: Matthias Blume (http://www.tti-c.org/blume/) | ||
1346 : | mblume | 1864 | Date: 2005/11/15 21:35:00 CST |
1347 : | Tag: blume-22051115-versiontool | ||
1348 : | Description: | ||
1349 : | |||
1350 : | Reversed change to src/system/smlnj/internals/versiontool.cm. This | ||
1351 : | file gets loaded as a tool -- by the equivalent of CM.make during the | ||
1352 : | run of CMB.make. Thus, CMB's path configuration is meaningless for | ||
1353 : | it. Instead, it has the status of "user code", so it should use | ||
1354 : | $/basis.cm to refer to the Basis library. (At least that's true for | ||
1355 : | the purpose of bootstrapping the previous change. In the future it | ||
1356 : | might make sense to have versiontool.cm refer to | ||
1357 : | $smlnj/basis/basis.cm, i.e., the version of the Basis that the | ||
1358 : | compiler itself uses.) | ||
1359 : | |||
1360 : | Also patched src/system/testml to have it activate those extra anchor | ||
1361 : | bindings in config/extrapathconfig. | ||
1362 : | |||
1363 : | ---------------------------------------------------------------------- | ||
1364 : | Name: Matthias Blume (http://www.tti-c.org/blume/) | ||
1365 : | mblume | 1863 | Date: 2005/11/15 16:35:00 CST |
1366 : | Tag: blume-22051115-libraries | ||
1367 : | Description: | ||
1368 : | |||
1369 : | This change affects the way the following libraries are tied into | ||
1370 : | the system: | ||
1371 : | |||
1372 : | $/basis.cm | ||
1373 : | $/smlnj-lib.cm | ||
1374 : | $/pp-lib.cm | ||
1375 : | $/controls-lib.cm | ||
1376 : | $/html-lib.cm | ||
1377 : | $/ml-yacc-lib.cm | ||
1378 : | |||
1379 : | These libraries are now internally (as seen from the source code | ||
1380 : | of the implementation itself) known by the following names: | ||
1381 : | |||
1382 : | $smlnj/basis/basis.cm | ||
1383 : | $smlnj/smlnj-lib/smlnj-lib.cm | ||
1384 : | $smlnj/smlnj-lib/pp-lib.cm | ||
1385 : | $smlnj/smlnj-lib/controls-lib.cm | ||
1386 : | $smlnj/smlnj-lib/html-lib.cm | ||
1387 : | $smlnj/ml-yacc/ml-yacc-lib.cm | ||
1388 : | |||
1389 : | This makes it possible to work with code that requires different | ||
1390 : | versions of these libraries, and which refers to these libraries using | ||
1391 : | their "default" names (i.e., the first set of names above). In other | ||
1392 : | words, one can un-define or re-define those default names without | ||
1393 : | compromising the proper functioning of the compiler itself. | ||
1394 : | |||
1395 : | A similar procedure had already been performed for several of the | ||
1396 : | MLRISC libraries that are linked into the compiler. I did some | ||
1397 : | cleanup on this code. | ||
1398 : | |||
1399 : | A new file in the config directory (named extrapathconfig) is | ||
1400 : | responsible for setting up path anchors that the compiler itself does | ||
1401 : | not need, but that are typically required by user code. | ||
1402 : | |||
1403 : | ---------------------------------------------------------------------- | ||
1404 : | mblume | 1862 | Name: Matthias Blume |
1405 : | Date: 2005/11/07 17:05:00 CST | ||
1406 : | Tag: blume-20051107-slice | ||
1407 : | Description: | ||
1408 : | |||
1409 : | Fixed erroneous out-of-bounds test in the "update" function of | ||
1410 : | various *ArraySlice modules. (Thanks to Vesa A. Norrman for pointing | ||
1411 : | out the problem.) | ||
1412 : | |||
1413 : | Pushed some Basis changes through ML-Lex, CML, and eXene. | ||
1414 : | |||
1415 : | ---------------------------------------------------------------------- | ||
1416 : | jhr | 1858 | Name: John Reppy (http://www.cs.uchicago.edu/~jhr) |
1417 : | jhr | 1861 | Date: 2005/11/07 |
1418 : | Description: | ||
1419 : | Fixed a Basis incompatibility: The depreciated function Substring.all | ||
1420 : | was removed (use Substring.full instead). | ||
1421 : | |||
1422 : | ---------------------------------------------------------------------- | ||
1423 : | Name: John Reppy (http://www.cs.uchicago.edu/~jhr) | ||
1424 : | jhr | 1858 | Date: 2005/11/05 |
1425 : | Description: | ||
1426 : | Tweaked interval set API in SML/NJ library; see the CHANGES file for details. | ||
1427 : | |||
1428 : | ---------------------------------------------------------------------- | ||
1429 : | mblume | 1856 | Name: Matthias Blume (http://www.tti-c.org/blume/) |
1430 : | Date: 2005/11/02 16:25:00 CST | ||
1431 : | Tag: blume-20051102-hex | ||
1432 : | Description: | ||
1433 : | |||
1434 : | * Runtime system bootstrap code now accepts hex digits in BOOTLIST | ||
1435 : | in either upper case or lower case format. | ||
1436 : | * Pushed changes to names of Pack<N>{Big,Little} structures through | ||
1437 : | CML and eXene. | ||
1438 : | |||
1439 : | ---------------------------------------------------------------------- | ||
1440 : | jhr | 1852 | Name: John Reppy (http://www.cs.uchicago.edu/~jhr) |
1441 : | jhr | 1855 | Date: 2005/11/02 |
1442 : | Description: | ||
1443 : | Fixed a Basis incompatibility: Pack<N>{Big,Little} structures should | ||
1444 : | be called PackWord<N>{Big,Little}. | ||
1445 : | |||
1446 : | ---------------------------------------------------------------------- | ||
1447 : | Name: John Reppy (http://www.cs.uchicago.edu/~jhr) | ||
1448 : | jhr | 1852 | Date: 2005/10/28 |
1449 : | Description: | ||
1450 : | Fixed a minor Basis incompatibility: hex digits should be upper case. | ||
1451 : | |||
1452 : | ---------------------------------------------------------------------- | ||
1453 : | mblume | 1846 | Name: Matthias Blume (blume (at) tti - c (dot) org) |
1454 : | Date: 2005/10/25 16:00:00 CDT | ||
1455 : | Tag: Release_110_56 | ||
1456 : | Description: | ||
1457 : | |||
1458 : | New working version (110.56). NEW BOOTFILES! | ||
1459 : | |||
1460 : | ---------------------------------------------------------------------- | ||
1461 : | jhr | 1838 | Name: John Reppy (http://www.cs.uchicago.edu/~jhr) |
1462 : | jhr | 1840 | Date: 2005/10/25 |
1463 : | Description: | ||
1464 : | Added interval sets to utility library (signatures INTERVAL_DOMAIN | ||
1465 : | and INTERVAL_SET, and functor IntervalSetFn). | ||
1466 : | |||
1467 : | ---------------------------------------------------------------------- | ||
1468 : | Name: John Reppy (http://www.cs.uchicago.edu/~jhr) | ||
1469 : | jhr | 1838 | Date: 2005/10/14 |
1470 : | Description: | ||
1471 : | Add Zhong Shao's fix for datatype equality functions. | ||
1472 : | |||
1473 : | ---------------------------------------------------------------------- | ||
1474 : | allenleung | 1836 | Name: Allen Leung (leunga (at) reservoir (dot) com) |
1475 : | allenleung | 1837 | Date: 2005/10/14 12:30:00 EST |
1476 : | Tag: leunga-20051014-x86-spill | ||
1477 : | Description: | ||
1478 : | |||
1479 : | Bug fix a bug found by Carl Hauser. | ||
1480 : | There was a typo in the reload code for FCMP in x86SpillInstr.sml. | ||
1481 : | |||
1482 : | ---------------------------------------------------------------------- | ||
1483 : | Name: Allen Leung (leunga (at) reservoir (dot) com) | ||
1484 : | allenleung | 1836 | Date: 2005/10/14 11:15:00 EST |
1485 : | Tag: leunga-20051014-x86-asm | ||
1486 : | Description: | ||
1487 : | |||
1488 : | Removed some debugging code in file x86Asm.sml. | ||
1489 : | The function emit_operand was printing out debugging output. | ||
1490 : | |||
1491 : | ---------------------------------------------------------------------- | ||
1492 : | jhr | 1813 | Name: John Reppy |
1493 : | jhr | 1831 | Date: 2005/07/27 |
1494 : | Tag: | ||
1495 : | Description: | ||
1496 : | |||
1497 : | jhr | 1832 | Fixed ml-lex to recognize "\r" as representing carriage return. |
1498 : | |||
1499 : | ---------------------------------------------------------------------- | ||
1500 : | Name: John Reppy | ||
1501 : | Date: 2005/07/27 | ||
1502 : | Tag: | ||
1503 : | Description: | ||
1504 : | |||
1505 : | jhr | 1831 | Fixed ml-yacc to work on files with non-native end-of-line |
1506 : | encodings (e.g., Windows text file on a Unix system). | ||
1507 : | |||
1508 : | ---------------------------------------------------------------------- | ||
1509 : | Name: John Reppy | ||
1510 : | jhr | 1824 | Date: 2005/07/20 |
1511 : | Tag: | ||
1512 : | Description: | ||
1513 : | |||
1514 : | Added changes from Dominic Evans (oldmanuk (at) gmail (dot) com) | ||
1515 : | to support HPUX 11. | ||
1516 : | |||
1517 : | ---------------------------------------------------------------------- | ||
1518 : | Name: John Reppy | ||
1519 : | jhr | 1813 | Date: 2005/07/06 |
1520 : | Tag: | ||
1521 : | Description: | ||
1522 : | |||
1523 : | Changes to the SML/NJ library. See smlnj-lib/CHANGES for details. | ||
1524 : | |||
1525 : | ---------------------------------------------------------------------- | ||
1526 : | mblume | 1808 | Name: Matthias Blume (blume (at) tti - c (dot) org) |
1527 : | Date: 2005/07/06 00:45:00 CDT | ||
1528 : | Tag: blume-20050706-slice-copy | ||
1529 : | Description: | ||
1530 : | |||
1531 : | Fixed reversed logic for deciding whether to "copy up" or "copy down" | ||
1532 : | in *-array-slice.sml. | ||
1533 : | |||
1534 : | ---------------------------------------------------------------------- | ||
1535 : | allenleung | 1806 | Name: Allen Leung (leunga (at) reservoir (dot) com) |
1536 : | allenleung | 1807 | Date: 2005/05/31 17:00:00 EST |
1537 : | Tag: leunga-20050531-cygwin-fault-2 | ||
1538 : | Description: | ||
1539 : | |||
1540 : | A typo in the cygwin code fixed. | ||
1541 : | |||
1542 : | ---------------------------------------------------------------------- | ||
1543 : | Name: Allen Leung (leunga (at) reservoir (dot) com) | ||
1544 : | allenleung | 1806 | Date: 2005/05/31 16:47:00 EST |
1545 : | Tag: leunga-20050531-cygwin-fault | ||
1546 : | Description: | ||
1547 : | |||
1548 : | blume | 2155 | Updated Cygwin's fault/signal handling to match the Windows version. |
1549 : | allenleung | 1806 | Updated the export list. |
1550 : | |||
1551 : | ---------------------------------------------------------------------- | ||
1552 : | mblume | 1791 | Name: Matthias Blume (blume (at) tti - c (dot) org) |
1553 : | mblume | 1795 | Date: 2005/05/18 13:35:00 CDT |
1554 : | Tag: Release_110_54 | ||
1555 : | Description: | ||
1556 : | |||
1557 : | New working version (110.54). NEW BOOTFILES! | ||
1558 : | |||
1559 : | ---------------------------------------------------------------------- | ||
1560 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
1561 : | mblume | 1794 | Date: 2005/05/18 11:58:00 CDT |
1562 : | Tag: blume-20050518-installer | ||
1563 : | Description: | ||
1564 : | |||
1565 : | Added support scripts for Mac OS X PackageMaker and modified | ||
1566 : | config/install.sh so that it supports re-dumping a heap image | ||
1567 : | after customization. | ||
1568 : | |||
1569 : | ---------------------------------------------------------------------- | ||
1570 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
1571 : | mblume | 1793 | Date: 2005/05/18 10:55:00 CDT |
1572 : | Tag: blume-20050518-realdiv-noovld | ||
1573 : | Description: | ||
1574 : | |||
1575 : | Un-overloaded / to work around bug in overloading resolution code. | ||
1576 : | |||
1577 : | ---------------------------------------------------------------------- | ||
1578 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
1579 : | mblume | 1792 | Date: 2005/05/16 23:50:00 CDT |
1580 : | Tag: blume-20050516-redump-heap | ||
1581 : | Description: | ||
1582 : | |||
1583 : | Added mechanism for re-creating a heap file for the interactive system | ||
1584 : | after configuration variables have been changed. | ||
1585 : | |||
1586 : | CM.redump_heap : string -> unit | ||
1587 : | |||
1588 : | This is much like SMLofNJ.exportML, but starting from the resulting | ||
1589 : | blume | 2155 | heap does not return to the caller of CM.redump_heap but |
1590 : | mblume | 1792 | restarts the interactive system from scratch. The original call of |
1591 : | CM.redump_heap does not return but ends the interactive session. | ||
1592 : | Thus, CM.redump_heap is a lot like SMLofNJ.exportFn. | ||
1593 : | |||
1594 : | Internally, redump_heap winds the dynamic execution context back to | ||
1595 : | the point where the original heap image was created and re-executes | ||
1596 : | the heap image generation code in the boot code. | ||
1597 : | |||
1598 : | ---------------------------------------------------------------------- | ||
1599 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
1600 : | mblume | 1791 | Date: 2005/05/09 21:55:00 CDT |
1601 : | Tag: blume-20050509-word64 | ||
1602 : | Description: | ||
1603 : | |||
1604 : | Added a hack to the existing hack known as Word64 to make fromString | ||
1605 : | behave correctly. I am still not sure whether Word64.scan will work | ||
1606 : | as specified with respect to the interaction of radix and prefix. | ||
1607 : | |||
1608 : | ---------------------------------------------------------------------- | ||
1609 : | allenleung | 1785 | Name: Allen Leung (leunga (at) reservoir (dot) com) |
1610 : | Date: 2005/05/04 11:50:00 EST | ||
1611 : | Tag: leunga-20050504-checkgc | ||
1612 : | Description: | ||
1613 : | |||
1614 : | blume | 2155 | Added a gc protocol checking phase. This phase is enabled with |
1615 : | allenleung | 1785 | the flag "check-gc". "debug-check-gc" turns on the verbose mode. |
1616 : | |||
1617 : | ---------------------------------------------------------------------- | ||
1618 : | mblume | 1741 | Name: Matthias Blume (blume (at) tti - c (dot) org) |
1619 : | mblume | 1784 | Date: 2005/05/04 10:45:00 CDT |
1620 : | Tag: blume-20050504-intinf | ||
1621 : | Description: | ||
1622 : | |||
1623 : | Fixed a bug in the implementation of div and mod for IntInf. | ||
1624 : | Thanks to Neophytos Michael for reporting the problem. | ||
1625 : | |||
1626 : | ---------------------------------------------------------------------- | ||
1627 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
1628 : | mblume | 1783 | Date: 2005/05/04 10:35:00 CDT |
1629 : | Tag: blume-20050504-join | ||
1630 : | Description: | ||
1631 : | |||
1632 : | Added a "join" combinator to the ParserComb module in smlnj-lib.cm. | ||
1633 : | |||
1634 : | ---------------------------------------------------------------------- | ||
1635 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
1636 : | mblume | 1770 | Date: 2005/02/28 23:40:00 CST |
1637 : | Tag: blume-20050228-mVar | ||
1638 : | Description: | ||
1639 : | |||
1640 : | Fixed serious bug (brown paper bag variety) in new implementation of | ||
1641 : | structure Atom in CML. (I had accidentally used a mailbox instead of | ||
1642 : | an mvar, leaving the door open for races.) | ||
1643 : | |||
1644 : | ---------------------------------------------------------------------- | ||
1645 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
1646 : | mblume | 1761 | Date: 2005/02/25 15:00:00 CST |
1647 : | Tag: Release_110_53 | ||
1648 : | Description: | ||
1649 : | |||
1650 : | New working version (110.53). NEW BOOTFILES! | ||
1651 : | |||
1652 : | ---------------------------------------------------------------------- | ||
1653 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
1654 : | mblume | 1760 | Date: 2005/02/25 14:50:00 CST |
1655 : | Tag: blume-20050225-susp | ||
1656 : | Description: | ||
1657 : | |||
1658 : | Brought back SMLofNJ.Susp. The underlying suspension type is the one | ||
1659 : | implemented in Core -- which means that it is the same as the one used | ||
1660 : | by the lazy extension. | ||
1661 : | |||
1662 : | ---------------------------------------------------------------------- | ||
1663 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
1664 : | mblume | 1759 | Date: 2005/02/24 16:50:00 CST |
1665 : | Tag: blume-20050224-cml-atom | ||
1666 : | Description: | ||
1667 : | |||
1668 : | Simpler and at the same time more general implementation of structure | ||
1669 : | Atom in CML. | ||
1670 : | |||
1671 : | ---------------------------------------------------------------------- | ||
1672 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
1673 : | mblume | 1758 | Date: 2005/02/15 17:35:00 CST |
1674 : | Tag: blume-20050215-tools | ||
1675 : | Description: | ||
1676 : | |||
1677 : | Created new "tools" directory under "src" and moved "TraceDebugProf" | ||
1678 : | there. | ||
1679 : | |||
1680 : | ---------------------------------------------------------------------- | ||
1681 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
1682 : | mblume | 1755 | Date: 2005/02/10 17:55:00 CST |
1683 : | Tag: blume-20050210-longlong | ||
1684 : | Description: | ||
1685 : | |||
1686 : | Implemented "long long" arguments and results for NLFFI. (Only the | ||
1687 : | PPC/MacOS implementation is complete, the other backends still need to | ||
1688 : | be updated.) | ||
1689 : | |||
1690 : | ---------------------------------------------------------------------- | ||
1691 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
1692 : | mblume | 1751 | Date: 2005/01/24 17:40:00 CST |
1693 : | Tag: blume-20050124-mlyacc | ||
1694 : | Description: | ||
1695 : | |||
1696 : | Minor cleanup in ML-Yacc rule printing mechanism. This should fix a | ||
1697 : | problem with certain "as" patterns which previously got rendered | ||
1698 : | using incorrect syntax. | ||
1699 : | |||
1700 : | ---------------------------------------------------------------------- | ||
1701 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
1702 : | mblume | 1750 | Date: 2005/01/18 12:00:00 CST |
1703 : | Tag: blume-20050118-profile | ||
1704 : | Description: | ||
1705 : | |||
1706 : | Made time profiling code (interrupt handler) in runtime system aware | ||
1707 : | of new array representation. | ||
1708 : | |||
1709 : | ---------------------------------------------------------------------- | ||
1710 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
1711 : | mblume | 1747 | Date: 2005/01/14 18:00:00 CST |
1712 : | Tag: blume-20050114-heap2exec | ||
1713 : | Description: | ||
1714 : | |||
1715 : | Implemented new (but still experimental) heap2exec facility. This is | ||
1716 : | tested under Mac OS X and should work under Linux (will test shortly). | ||
1717 : | It will probably also work on the Sparc (will test some time later). | ||
1718 : | |||
1719 : | - removed old "HACKED_STANDALONE" hack from runtime | ||
1720 : | |||
1721 : | To be able to test this, uncomment the request for "heap2asm" in | ||
1722 : | config/targets prior to installation. (Notice that this is different | ||
1723 : | from "heap2exec" mentioned below. Not a typo.) | ||
1724 : | |||
1725 : | To perform an actual test, run the command | ||
1726 : | |||
1727 : | $ bin/heap2exec heapfile execfile | ||
1728 : | |||
1729 : | (You can put heap2exec on your shell's path.) | ||
1730 : | |||
1731 : | For example, run | ||
1732 : | |||
1733 : | $ bin/heap2exec bin/.heap/ml-yacc.ppc-darwin mly | ||
1734 : | |||
1735 : | This will create a standalone executable called "mly" which you can | ||
1736 : | then invoke directly as a command. | ||
1737 : | |||
1738 : | ---------------------------------------------------------------------- | ||
1739 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
1740 : | mblume | 1746 | Date: 2005/01/07 11:44:00 CST |
1741 : | Tag: blume-20050107-mlstring | ||
1742 : | Description: | ||
1743 : | |||
1744 : | fixed off-by-one error in ML_STRING macro (globals.c) | ||
1745 : | |||
1746 : | ---------------------------------------------------------------------- | ||
1747 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
1748 : | mblume | 1744 | Date: 2004/12/23 18:00:00 CST |
1749 : | Tag: blume-20041223-santa | ||
1750 : | Description: | ||
1751 : | |||
1752 : | Made ml-build script "smarter" (but only very little). | ||
1753 : | |||
1754 : | ---------------------------------------------------------------------- | ||
1755 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
1756 : | mblume | 1742 | Date: 2004/12/21 15:05:00 CST |
1757 : | Tag: blume-20041221-longlong | ||
1758 : | Description: | ||
1759 : | |||
1760 : | * Implemented access to signed and unsigned long long data in NLFFI. | ||
1761 : | (The parameter-passing part of the picture has not complete. But | ||
1762 : | data structure access seems to work.) | ||
1763 : | |||
1764 : | * Fixed CM's incorrect assumption that the PPC is little-endian. | ||
1765 : | (On the Mac, it is big-endian. And that's currently our only | ||
1766 : | PPC platform.) | ||
1767 : | |||
1768 : | ---------------------------------------------------------------------- | ||
1769 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
1770 : | mblume | 1741 | Date: 2004/12/21 12:50:00 CST |
1771 : | Tag: blume-20041221-memory | ||
1772 : | Description: | ||
1773 : | |||
1774 : | Some cleanup in the $c/memory.cm library: separated some concerns by | ||
1775 : | moving allocation code and memory access code each into their own | ||
1776 : | files. | ||
1777 : | |||
1778 : | ---------------------------------------------------------------------- | ||
1779 : | allenleung | 1740 | Name: Allen Leung (leunga (at) reservoir (dot) com) |
1780 : | Date: 2004/12/17 16:12:00 EST | ||
1781 : | Tag: leunga-20041217-cygwin-smlnj-home | ||
1782 : | Description: | ||
1783 : | |||
1784 : | blume | 2155 | The Unix I/O library of SML/NJ on cygwin does not understand |
1785 : | Windows style pathname, so problems arise when SMLNJ_HOME is set | ||
1786 : | to a Windows style pathname. _run-sml now converts SMLNJ_HOME | ||
1787 : | to a POSIX pathname on cygwin. | ||
1788 : | allenleung | 1740 | |
1789 : | ---------------------------------------------------------------------- | ||
1790 : | mblume | 1737 | Name: Matthias Blume (blume (at) tti - c (dot) org) |
1791 : | mblume | 1736 | Date: 2004/12/16 13:35:00 CST |
1792 : | Tag: Release_110_52 | ||
1793 : | Description: | ||
1794 : | |||
1795 : | Last-minute changes incorporated into 110.52. Release tag moved. | ||
1796 : | |||
1797 : | The changes: | ||
1798 : | - HashString.hashString' -> HashString.hashSubstring | ||
1799 : | - bug fix in UnivariateStats | ||
1800 : | |||
1801 : | ---------------------------------------------------------------------- | ||
1802 : | mblume | 1702 | Name: Matthias Blume (blume (at) tti - c (dot) org) |
1803 : | mblume | 1733 | Date: 2004/12/15 23:40:00 CST |
1804 : | Tag: blume-20041215-hashSubstring | ||
1805 : | Description: | ||
1806 : | |||
1807 : | - HashString.hashString' -> HashString.hashSubstring | ||
1808 : | - corresponding changes in atom.sml | ||
1809 : | - "de-compressed" (aka. un-obfuscated) code for UnivariateStats and | ||
1810 : | added some comments | ||
1811 : | |||
1812 : | ---------------------------------------------------------------------- | ||
1813 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
1814 : | mblume | 1728 | Date: 2004/12/15 15:30:00 CST |
1815 : | mblume | 1736 | Tag: (Release_110_52) |
1816 : | mblume | 1728 | Description: |
1817 : | |||
1818 : | New working version (110.52). NEW BOOTFILES! | ||
1819 : | |||
1820 : | ---------------------------------------------------------------------- | ||
1821 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
1822 : | mblume | 1727 | Date: 2004/12/15 12:45:00 CST |
1823 : | Tag: blume-20041215-spaces | ||
1824 : | Description: | ||
1825 : | |||
1826 : | More on the space problem (this time for Win32). | ||
1827 : | |||
1828 : | ---------------------------------------------------------------------- | ||
1829 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
1830 : | mblume | 1725 | Date: 2004/12/14 17:30:00 CST |
1831 : | Tag: blume-20041214-spaces | ||
1832 : | Description: | ||
1833 : | |||
1834 : | Hacked some of the scripts (in particular: the installer) to cope with | ||
1835 : | spaces in filenames a bit better. But beware: the current "solution" | ||
1836 : | is likely still full of bugs and inherently incomplete. (We need to | ||
1837 : | do away with those shell scripts for a comprehensive solution.) | ||
1838 : | |||
1839 : | ---------------------------------------------------------------------- | ||
1840 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
1841 : | mblume | 1724 | Date: 2004/12/13 14:45:00 CST |
1842 : | Tag: blume-20041213-ml-makedepend | ||
1843 : | Description: | ||
1844 : | |||
1845 : | Fixed bug in code for ml-makedepend. | ||
1846 : | |||
1847 : | ---------------------------------------------------------------------- | ||
1848 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
1849 : | mblume | 1720 | Date: 2004/12/09 16:30:00 CST |
1850 : | Tag: blume-20041209-statistics | ||
1851 : | Description: | ||
1852 : | |||
1853 : | Added two simple but potentially useful statistics modules to SML/NJ Library. | ||
1854 : | (See CHANGES file there.) | ||
1855 : | |||
1856 : | ---------------------------------------------------------------------- | ||
1857 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
1858 : | mblume | 1718 | Date: 2004/12/01 16:50:00 CST |
1859 : | Tag: blume-20041201-atom | ||
1860 : | Description: | ||
1861 : | |||
1862 : | smlnj-lib: | ||
1863 : | |||
1864 : | Added function HashString.hashString' for substrings. | ||
1865 : | Hand-inlined CharVector.foldl into HashString (for speed). | ||
1866 : | Modified implementation of structure Atom to avoid extracting | ||
1867 : | strings from substrings unless necessary. | ||
1868 : | (Also see CHANGES file for smlnj-lib.) | ||
1869 : | |||
1870 : | ---------------------------------------------------------------------- | ||
1871 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
1872 : | mblume | 1717 | Date: 2004/11/24 22:15:00 CST |
1873 : | Tag: blume-20041124-cml | ||
1874 : | Description: | ||
1875 : | |||
1876 : | Made sure CML compiles when Position = Int64. | ||
1877 : | |||
1878 : | ---------------------------------------------------------------------- | ||
1879 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
1880 : | mblume | 1715 | Date: 2004/11/24 14:30:00 CST |
1881 : | Tag: blume-20041124-position | ||
1882 : | Description: | ||
1883 : | |||
1884 : | The compiler can now be compiled in a mode that makes structure Position | ||
1885 : | equal to Int64. The default, however, is unchanged (Position = Int31) | ||
1886 : | for the time being. | ||
1887 : | |||
1888 : | To enable 64-bit positions, use the following procedure: | ||
1889 : | |||
1890 : | 1. Start sml | ||
1891 : | 2. Autoload $smlnj/cmb.cm (if not already autoloaded) | ||
1892 : | 3. Type | ||
1893 : | #set (CMB.symval "USE_64_BIT_POSITIONS") (SOME 1); | ||
1894 : | 4. Run CMB.make() as usual. | ||
1895 : | |||
1896 : | This is barely tested. The only test so far was a little SML program | ||
1897 : | counting the number of characters in an 8-gigabyte file by | ||
1898 : | reading it character-by-character. That test was successful. | ||
1899 : | |||
1900 : | In support of 64-bit positions, a number of new functions have been | ||
1901 : | added to the runtime system. | ||
1902 : | |||
1903 : | ---------------------------------------------------------------------- | ||
1904 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
1905 : | mblume | 1716 | Date: 2004/11/23 14:45:00 CST |
1906 : | Tag: blume-20041123-useFile | ||
1907 : | mblume | 1712 | Description: |
1908 : | |||
1909 : | Fixed a problem with unhelpful error messages related to problems with | ||
1910 : | .cm- or .sml files that appear as part of the sml command line. | ||
1911 : | |||
1912 : | ---------------------------------------------------------------------- | ||
1913 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
1914 : | mblume | 1703 | Date: 2004/11/18 15:40:00 CST |
1915 : | Tag: Release_110_51 | ||
1916 : | Description: | ||
1917 : | |||
1918 : | New working version (110.51). NEW BOOTFILES! | ||
1919 : | |||
1920 : | ---------------------------------------------------------------------- | ||
1921 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
1922 : | mblume | 1702 | Date: 2004/11/18 15:35:00 CST |
1923 : | Tag: <none> | ||
1924 : | Description: | ||
1925 : | |||
1926 : | Enabled dlopen and friends for FreeBSD (as recommended by Johannes 5 | ||
1927 : | Joemann). | ||
1928 : | |||
1929 : | ---------------------------------------------------------------------- | ||
1930 : | allenleung | 1701 | Name: Allen Leung (leunga (at) reservoir (dot) com) |
1931 : | Date: 2004/11/17 16:05:21 EST 2004 | ||
1932 : | Tag: leunga-20041117-mlrisc-live-kill | ||
1933 : | Description: | ||
1934 : | |||
1935 : | blume | 2155 | Added support for MLTree constructs LIVE and KILL |
1936 : | allenleung | 1701 | to all the architectures. |
1937 : | |||
1938 : | ---------------------------------------------------------------------- | ||
1939 : | mblume | 1687 | Name: Matthias Blume (blume (at) tti - c (dot) org) |
1940 : | mblume | 1689 | Date: 2004/11/13 00:20:00 CST |
1941 : | Tag: blume-20041113-versiontool | ||
1942 : | Description: | ||
1943 : | |||
1944 : | - Stripped down the versiontool: It now only handles the version number. | ||
1945 : | The date string is generated at bootstrap time (during makeml). | ||
1946 : | |||
1947 : | - In a previous commit, fixed a minor issue with how polyequal is being | ||
1948 : | translated. In particular, the code now "looks through" abstractions. | ||
1949 : | This results in slightly fewer polyEqual warnings and hopefully slightly | ||
1950 : | more efficient code. Important examples for where this matters are | ||
1951 : | the new int64 and word64 types. | ||
1952 : | |||
1953 : | ---------------------------------------------------------------------- | ||
1954 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
1955 : | mblume | 1687 | Date: 2004/11/12 00:30:00 CST |
1956 : | Tag: blume-20041112-int64 | ||
1957 : | Description: | ||
1958 : | |||
1959 : | Structure Int64 fully hooked in. (The implementation is not very | ||
1960 : | efficient, though.) | ||
1961 : | |||
1962 : | ---------------------------------------------------------------------- | ||
1963 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
1964 : | mblume | 1686 | Date: 2004/11/11 17:30:00 CST |
1965 : | Tag: blume-20041111-more64 | ||
1966 : | Description: | ||
1967 : | |||
1968 : | All the pieces of Word64 are now there, with the exception of the | ||
1969 : | conversions from and to LargeWord. (Eventually these need to be identities, | ||
1970 : | but for the time being they don't even make sense because LargeWord is | ||
1971 : | 32-bit wide.) | ||
1972 : | |||
1973 : | Also started to add similar support for Int64, but major pieces of that | ||
1974 : | are still missing. | ||
1975 : | |||
1976 : | ---------------------------------------------------------------------- | ||
1977 : | mblume | 1668 | Name: Matthias Blume (blume (at) tti - c (dot) org) |
1978 : | mblume | 1684 | Date: 2004/11/11 00:15:00 CST |
1979 : | Tag: blume-20041111-word64 | ||
1980 : | Description: | ||
1981 : | |||
1982 : | Structure Word64 is now (almost) complete, word literals and patterns | ||
1983 : | seem to work. There are a few odd pieces missing. In particular, | ||
1984 : | I didn't do the {from,to}LargeWord parts because LargeWord is still Word32 | ||
1985 : | at the moment. | ||
1986 : | |||
1987 : | Making Word64 official would mean that LargeWord becomes Word64. But | ||
1988 : | this requires extreme care because most word-word conversions have to | ||
1989 : | go through LargeWord, so making a mistake means loss of efficiency or | ||
1990 : | worse. Eventually there will be a solution similar to (but actually | ||
1991 : | simpler than) what I did with IntInf. | ||
1992 : | |||
1993 : | ---------------------------------------------------------------------- | ||
1994 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
1995 : | mblume | 1683 | Date: 2004/11/10 18:12:00 CST |
1996 : | Tag: blume-20041110-64bit | ||
1997 : | Description: | ||
1998 : | |||
1999 : | More 64-bit hacking (but still not even half-way there yet). | ||
2000 : | Also, some assorted improvements to the handling of 8-bit words. | ||
2001 : | |||
2002 : | ---------------------------------------------------------------------- | ||
2003 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
2004 : | mblume | 1682 | Date: 2004/11/09 17:50:00 CST |
2005 : | mblume | 1683 | Tag: <none> |
2006 : | mblume | 1682 | Description: |
2007 : | |||
2008 : | Started adding some infrastructure for supporting 64-bit int- and | ||
2009 : | word-types. (Still in its very early stages.) | ||
2010 : | |||
2011 : | ---------------------------------------------------------------------- | ||
2012 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
2013 : | mblume | 1672 | Date: 2004/10/28 10:45:00 CDT |
2014 : | Tag: Release_110_50 | ||
2015 : | Description: | ||
2016 : | |||
2017 : | New working version (110.50). NEW BOOTFILES! | ||
2018 : | |||
2019 : | ===================== | ||
2020 : | |||
2021 : | Also: | ||
2022 : | |||
2023 : | - Changend config/srcarchiveurl from a file just | ||
2024 : | containing the URL string into a file containing | ||
2025 : | shell script code. The code has access to the $VERSION variable. | ||
2026 : | - Made corresponding changes to config/install.sh and config/unpack. | ||
2027 : | - Default contents of config/srcarchiveurl uses $VERSION and | ||
2028 : | normally does not have to be edited to reflect a version change. | ||
2029 : | |||
2030 : | (As a result, a version change can be done by just editing | ||
2031 : | config/version, the rest is now automatic.) | ||
2032 : | |||
2033 : | ---------------------------------------------------------------------- | ||
2034 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
2035 : | mblume | 1669 | Date: 2004/10/27 17:50:00 CDT |
2036 : | Tag: blume-20041027-btrace-msg | ||
2037 : | Description: | ||
2038 : | |||
2039 : | BackTrace.monitor now also reports the source of the exception that | ||
2040 : | triggered the trace. | ||
2041 : | |||
2042 : | ---------------------------------------------------------------------- | ||
2043 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
2044 : | mblume | 1668 | Date: 2004/10/27 17:20:00 CDT |
2045 : | Tag: blume-20041027-x86-c-calls | ||
2046 : | Description: | ||
2047 : | |||
2048 : | This is the HISTORY entry for two earlier commits, both concerning | ||
2049 : | the x86 c-calls code in MLRISC: | ||
2050 : | |||
2051 : | - added a missing LOAD in the code that deals with struct arguments | ||
2052 : | - made sure the caller does not add the wrong number of bytes to the | ||
2053 : | stack pointer after a call of a function returning a struct | ||
2054 : | (the callee already pops the implicit argument which points to | ||
2055 : | the space reserved for the result) | ||
2056 : | |||
2057 : | ---------------------------------------------------------------------- | ||
2058 : | allenleung | 1662 | Name: Allen Leung (leunga (at) reservoir (dot) com) |
2059 : | Date: 2004/10/24 14:00:00 EST | ||
2060 : | Tag: leunga-20041024-x86-gas-fucomip | ||
2061 : | Description: | ||
2062 : | |||
2063 : | John discovered a bug in the syntax of fucomip. | ||
2064 : | The opcodes FU?COMIP? have been changed to | ||
2065 : | |||
2066 : | fu?comip? %st(i), %st | ||
2067 : | |||
2068 : | ---------------------------------------------------------------------- | ||
2069 : | mblume | 1622 | Name: Matthias Blume (blume (at) tti - c (dot) org) |
2070 : | mblume | 1656 | Date: 2004/10/20 15:06:00 CDT |
2071 : | Tag: blume-20041020-standalone-backtrace | ||
2072 : | Description: | ||
2073 : | |||
2074 : | Added a mechanism for getting back-trace information from standalone | ||
2075 : | programs. Here is how it works: | ||
2076 : | |||
2077 : | 1. The part of the program from which you want to get backtrace | ||
2078 : | information (usually the whole program) should be wrapped with | ||
2079 : | BackTrace.monitor. This is a (unit->'a)->'a function, and your | ||
2080 : | main program could be modified from something like | ||
2081 : | |||
2082 : | fun main (pgm, args) = ... | ||
2083 : | |||
2084 : | to | ||
2085 : | |||
2086 : | fun main (pgm, args) = BackTrace.monitor (fn () => ...) | ||
2087 : | |||
2088 : | 2. To be able to access BackTrace.monitor, you have to add | ||
2089 : | library $smlnj-tdp/plugins.cm to the .cm file that contains your | ||
2090 : | main function. | ||
2091 : | |||
2092 : | 3. Remove all compiled code (i.e., all the .cm/ subdirectories that | ||
2093 : | CM might have created in the past for your project). | ||
2094 : | |||
2095 : | 4. Build the system using this command line: | ||
2096 : | |||
2097 : | ml-build -Ctdp.instrument=true \$smlnj-tdp/back-trace.cm \ | ||
2098 : | myprog.cm MyProg.main myprog | ||
2099 : | |||
2100 : | instead of the usual | ||
2101 : | |||
2102 : | ml-build myprog.cm MyProg.main myprog | ||
2103 : | |||
2104 : | I changed a library name: | ||
2105 : | |||
2106 : | $/trace-debug-profile.cm --> $smlnj-tdp/plugins.cm | ||
2107 : | |||
2108 : | New libraries: | ||
2109 : | |||
2110 : | $smlnj-tdp/back-trace.cm | ||
2111 : | -- when loaded causes the back-trace plugin to be installed | ||
2112 : | $smlnj-tdp/coverage.cm | ||
2113 : | -- when loaded causes the coverage plugin to be installed | ||
2114 : | |||
2115 : | ---------------------------------------------------------------------- | ||
2116 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
2117 : | mblume | 1655 | Date: 2004/10/18 16:45:00 CDT |
2118 : | Tag: blume-20041018-groupowner | ||
2119 : | Description: | ||
2120 : | |||
2121 : | Added an "obsolete" warning for the "group owner" syntax to CM's parser. | ||
2122 : | |||
2123 : | Eliminated group owner specs from .cm files throughout the source tree. | ||
2124 : | |||
2125 : | ---------------------------------------------------------------------- | ||
2126 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
2127 : | mblume | 1653 | Date: 2004/10/15 15:45:00 CDT |
2128 : | Tag: blume-20041015-coverage | ||
2129 : | Description: | ||
2130 : | |||
2131 : | * Test coverage tool added! | ||
2132 : | * Further reorganization of tracing-, debugging-, and profiling support: | ||
2133 : | |||
2134 : | - moved original BTImp -- now called BackTrace -- into a separate | ||
2135 : | library called $/trace-debug-profile.cm | ||
2136 : | - eliminated all mentions of BTrace from SMLofNJ.Internals | ||
2137 : | - only the instrumentation mechanism is now left in the compiler proper | ||
2138 : | - BackTrace module is a plugin which is NOT plugged in by default | ||
2139 : | - Coverage module is another such plugin | ||
2140 : | |||
2141 : | To get the benefits of any of these plugin modules, the code in | ||
2142 : | question must be compiled with tdp instrumentation turned on. This | ||
2143 : | can be done by setting SMLofNJ.Internals.TDP.mode to true. | ||
2144 : | (The ref cell is also controlled via the -Ctdp.instrument=... switch.) | ||
2145 : | |||
2146 : | Plugins are selected at link time. (Pre-compiled instrumented code | ||
2147 : | can be re-loaded with different plugins in effect.) When an | ||
2148 : | instrumented module is linked, whatever plugins are at that time | ||
2149 : | enabled will come into effect for that module. | ||
2150 : | |||
2151 : | To enable the back-trace plugin, load library $/trace-debug-profile.cm | ||
2152 : | and invoke BackTrace.install() (e.g., from the interactive prompt). | ||
2153 : | To enable the coverage plugin, load the same library and invoke | ||
2154 : | Coverage.install(). | ||
2155 : | |||
2156 : | Back-traces are generated automatically on uncaught exceptions and | ||
2157 : | when the code in question explicitly invokes BackTrace.trigger(). | ||
2158 : | |||
2159 : | Coverage (and execution frequency-) information must be queried | ||
2160 : | explicitly by calling Coverage.not_covered and Coverage.hot_spots. | ||
2161 : | |||
2162 : | ---------------------------------------------------------------------- | ||
2163 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
2164 : | mblume | 1652 | Date: 2004/10/14 17:40:00 CDT |
2165 : | Tag: blume-20041014-tdp-core | ||
2166 : | Description: | ||
2167 : | |||
2168 : | Snapshot of a significant overhaul of how the trace/debug/profile support | ||
2169 : | is hooked into the system (specifically: Core and SMLofNJ.Internals). | ||
2170 : | |||
2171 : | ---------------------------------------------------------------------- | ||
2172 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
2173 : | mblume | 1650 | Date: 2004/10/13 16:34:00 CDT |
2174 : | Tag: blume-20041013-tdp | ||
2175 : | Description: | ||
2176 : | |||
2177 : | Some rationalization of names: | ||
2178 : | |||
2179 : | structure BTrace -> structure TDPInstrument | ||
2180 : | etc. | ||
2181 : | |||
2182 : | This is is preparation of using the original back-trace | ||
2183 : | instrumentation for other purposes. "TDP" stands for | ||
2184 : | Trace/Debug/Profile. | ||
2185 : | |||
2186 : | The control flag controlling whether instrumentation is on or off is now | ||
2187 : | registered under a different name, so instead of running sml as | ||
2188 : | |||
2189 : | sml -Cinstrument.btrace-mode=true | ||
2190 : | |||
2191 : | one has to say | ||
2192 : | |||
2193 : | sml -Ctdp.instrument=true | ||
2194 : | |||
2195 : | ---------------------------------------------------------------------- | ||
2196 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
2197 : | mblume | 1645 | Date: 2004/10/11 16:37:00 CDT |
2198 : | Tag: blume-20041011-regions | ||
2199 : | Description: | ||
2200 : | |||
2201 : | Made some minor modifications to elabcore.sml to have source regions | ||
2202 : | be propagated more tightly -- resulting in better (i.e., smaller) | ||
2203 : | regions being reported in error- and debug messages. | ||
2204 : | |||
2205 : | ---------------------------------------------------------------------- | ||
2206 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
2207 : | mblume | 1644 | Date: 2004/10/08 22:50:00 CDT |
2208 : | Tag: blume-20041008-cmkw | ||
2209 : | Description: | ||
2210 : | |||
2211 : | Fixed handling of keywords in .cm files: After seeing "is" the lexer | ||
2212 : | treats subsequent occurrences of "group", "library", "source", "is", | ||
2213 : | "*", and "-" as ordinary identifiers rather than keywords. | ||
2214 : | |||
2215 : | Most seriously, this fixes a problem with CM's "shell" tool. The tool | ||
2216 : | is supposed to accept a tool argument called "source", but this did | ||
2217 : | not work because of the clash with the keyword. | ||
2218 : | |||
2219 : | ---------------------------------------------------------------------- | ||
2220 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
2221 : | mblume | 1642 | Date: 2004/10/07 16:00:00 CDT |
2222 : | Tag: blume-20041007-cleanup | ||
2223 : | Description: | ||
2224 : | |||
2225 : | Assorted cleanup work: | ||
2226 : | |||
2227 : | - got rid of intstrmap in favor of using the library's | ||
2228 : | hash table implementation | ||
2229 : | - threw out most of the pathnames stuff, as it was not used anyway | ||
2230 : | - simplified tokentable implementation | ||
2231 : | - fixed some minor spelling errors | ||
2232 : | |||
2233 : | ---------------------------------------------------------------------- | ||
2234 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
2235 : | mblume | 1641 | Date: 2004/10/06 15:15:15 CDT |
2236 : | Tag: blume-20041006-handler | ||
2237 : | Description: | ||
2238 : | |||
2239 : | Cleaned up the absyn to reflect the invariant that HANDLE always | ||
2240 : | carries a FNexp as part of the type definition. This eliminates some | ||
2241 : | superfluous sanity checks at runtime down the road. | ||
2242 : | |||
2243 : | Some minor cleanup of the btrace code. | ||
2244 : | |||
2245 : | ---------------------------------------------------------------------- | ||
2246 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
2247 : | mblume | 1640 | Date: 2004/10/01 10:20:30 CDT |
2248 : | Tag: blume-20041001-slave | ||
2249 : | Description: | ||
2250 : | |||
2251 : | Added hack to make slave mode work in the presence of the version | ||
2252 : | tool. (Still, since the master does two passes over the code for | ||
2253 : | CMB.make, the release number gets bumped twice when slaves are | ||
2254 : | attached. I don't know if this is worth fixing...) | ||
2255 : | |||
2256 : | ---------------------------------------------------------------------- | ||
2257 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
2258 : | mblume | 1639 | Date: 2004/09/30 10:55:00 CDT |
2259 : | Tag: blume-20040930-version | ||
2260 : | Description: | ||
2261 : | |||
2262 : | * Moved the "version" magic into its own little library under | ||
2263 : | mblume | 1672 | src/system/smlnj/internal. This avoids expensive reconstruction of |
2264 : | a stable src/compiler/core.cm. | ||
2265 : | mblume | 1639 | |
2266 : | mblume | 1672 | * At the same time, structure CompilerVersion is now known as |
2267 : | structure SMLNJVersion. | ||
2268 : | mblume | 1639 | |
2269 : | * Arranged for the version tool to NOT kick in when rebuilding the system | ||
2270 : | (makeml -rebuild, fixpt). Otherwise one would never reach a fixpoint. | ||
2271 : | Also, loading the versiontool does not work when rebuilding the system | ||
2272 : | because CM is not properly initialized at that time. | ||
2273 : | |||
2274 : | ---------------------------------------------------------------------- | ||
2275 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
2276 : | mblume | 1638 | Date: 2004/09/29 14:00:00 CDT |
2277 : | Tag: blume-20040929-autoversion | ||
2278 : | Description: | ||
2279 : | |||
2280 : | Implemented some CM magic to have | ||
2281 : | file src/compiler/TopLevel/main/version.sml | ||
2282 : | generated automagically. | ||
2283 : | The version is taken from two files: config/version and config/release. | ||
2284 : | The first is expected to contain a two-part version number such as 110.49. | ||
2285 : | The second should contain a single number, but it may be missing. | ||
2286 : | |||
2287 : | If the environment variable VERSIONTOOL_BUMP_RELEASE is defined at the | ||
2288 : | time the version tool is loaded (which is the first time you say CMB.make), | ||
2289 : | then the tool will increment the value stored in config/release every | ||
2290 : | time CMB.make is invoked. | ||
2291 : | |||
2292 : | The binfile format is now insensitive to anything beyond the first | ||
2293 : | two components of a version number, so bumping the release does not render | ||
2294 : | binfiles incompatible. Auto-bumping can be used to keep track of versions | ||
2295 : | during development without invalidating existing binfiles. | ||
2296 : | |||
2297 : | In any case, every CMB.make updates the date information in version.sml. | ||
2298 : | (This is the date that is printed in the banner.) | ||
2299 : | |||
2300 : | ---------------------------------------------------------------------- | ||
2301 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
2302 : | mblume | 1634 | Date: 2004/09/28 10:53:00 CDT |
2303 : | Tag: blume-20040928-controls | ||
2304 : | Description: | ||
2305 : | |||
2306 : | Some cleanup of the controls code. | ||
2307 : | |||
2308 : | ---------------------------------------------------------------------- | ||
2309 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
2310 : | mblume | 1633 | Date: 2004/09/27 22:08:00 CDT |
2311 : | Tag: blume-20040927-controls | ||
2312 : | Description: | ||
2313 : | |||
2314 : | Added two pieces of functionality to the Controls interface: | ||
2315 : | |||
2316 : | 1. val save'restore: 'a control -> unit -> unit | ||
2317 : | |||
2318 : | grabs the current value of the control in stage 1 and restores it | ||
2319 : | in stage 2. | ||
2320 : | |||
2321 : | 2. val set' : 'a control * 'a -> unit -> unit | ||
2322 : | |||
2323 : | stores the given value into the control in stage 2 (i.e., delayed) | ||
2324 : | but does all error checking in stage 1. | ||
2325 : | (This is for string controls that need to do parse their argument -- | ||
2326 : | something that might fail. In some cases, notably in CM, one | ||
2327 : | already knows the intended argument but wants to delay the actual | ||
2328 : | assignment until a time when error recovery would be more difficult.) | ||
2329 : | |||
2330 : | Changed the handling of controls in tool arguments to classes "sml" and | ||
2331 : | "lazysml": | ||
2332 : | - use Controls.save'restore as a more robust way of restoring the | ||
2333 : | old value (in particular: without having to re-parse the string) | ||
2334 : | - use controls to handle the "overload" keyword in the init group | ||
2335 : | (I believe this change actually fixes a long-standing obscure bug.) | ||
2336 : | |||
2337 : | ---------------------------------------------------------------------- | ||
2338 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
2339 : | mblume | 1632 | Date: 2004/09/27 17:00:00 CDT |
2340 : | Tag: blume-20040927-lazysml | ||
2341 : | Description: | ||
2342 : | |||
2343 : | Added a new tool class called "lazysml" to CM's tool chest. The only | ||
2344 : | difference to "sml" is that compilation is done with Control.lazysml | ||
2345 : | set to true. A source of class "lazysml" is automatically recognized | ||
2346 : | by a file name suffix of ".lml". | ||
2347 : | |||
2348 : | In addition to the above feature, the original class "sml" now also | ||
2349 : | supports a tool argument "lazy" which has the same effect. As a | ||
2350 : | result, the following three lines are equivalent: | ||
2351 : | |||
2352 : | foo.sml : lazysml | ||
2353 : | foo.sml : sml (lazy) | ||
2354 : | foo.sml (lazy) | ||
2355 : | |||
2356 : | The setting goes into effect both during parsing and during | ||
2357 : | compilation. The original setting is restored right after parsing and | ||
2358 : | after compilation, respectively. | ||
2359 : | |||
2360 : | In addition to all the above, there is also a general mechanism to set | ||
2361 : | ANY of the "controls" that are available at the command line via | ||
2362 : | "-C..." on a per-sml-file basis. The same rules that apply for "lazy" | ||
2363 : | apply as well. (In fact, "lazy" is implemented as a special case of | ||
2364 : | the general mechanism.) | ||
2365 : | |||
2366 : | The .cm file syntax uses a new keyword tool argument called "with". | ||
2367 : | There are several ways of indicating the desired settings: | ||
2368 : | |||
2369 : | foo.sml (with:parser.quotations=true) | ||
2370 : | foo.sml (with:(name:parser.quotations value:true)) | ||
2371 : | foo.sml (with:(name:name1 value:value1 name:name2 value:value2 ...)) | ||
2372 : | foo.sml (with:(name1=value1 name2=value2 ...)) | ||
2373 : | foo.sml (with:(name1=value1 name:name2 value:value2 name3=value3 ...)) | ||
2374 : | |||
2375 : | etc. | ||
2376 : | |||
2377 : | Another possible abbreviation is to leave out the =v or value:v part | ||
2378 : | if the name refers to a boolean control (in which case the value is | ||
2379 : | taken to be true). Thus, one could get lazy sml also by saying: | ||
2380 : | |||
2381 : | foo.sml (with:parser.lazy-keyword=true) | ||
2382 : | foo.sml (with:parser.lazy-keyword) | ||
2383 : | foo.sml (with:(name:parser.lazy-keyword value:true)) | ||
2384 : | foo.sml (with:(name:parser.lazy-keyword)) | ||
2385 : | |||
2386 : | ---------------------------------------------------------------------- | ||
2387 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
2388 : | mblume | 1631 | Date: 2004/09/24 16:22:00 CDT |
2389 : | Tag: blume-20040924-ppc-long-branch | ||
2390 : | Description: | ||
2391 : | |||
2392 : | Turned message about "emiting long form of branch" off by default. | ||
2393 : | Added a control flag to turn it back on when desired. | ||
2394 : | |||
2395 : | ---------------------------------------------------------------------- | ||
2396 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
2397 : | mblume | 1630 | Date: 2004/09/24 16:05:00 CDT |
2398 : | Tag: blume-20040924-rounding | ||
2399 : | Description: | ||
2400 : | |||
2401 : | Applied patch for setting rounding modes under Mac OS X. Thanks to | ||
2402 : | Melissa O'Neill for providing the code! | ||
2403 : | |||
2404 : | ---------------------------------------------------------------------- | ||
2405 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
2406 : | mblume | 1629 | Date: 2004/09/23 17:30:00 CDT |
2407 : | Tag: blume-20040923-envvars | ||
2408 : | Description: | ||
2409 : | |||
2410 : | 1. Changed definition of type ControlRegistry.registry_tree to | ||
2411 : | include control_info (i.e., the name of the controlling environment | ||
2412 : | variable). | ||
2413 : | |||
2414 : | 2. Added command-line flags -e and -E to print the names of environment | ||
2415 : | variables that can be used to control internal settings. | ||
2416 : | (This uses the new API mentioned in 1.) | ||
2417 : | |||
2418 : | ---------------------------------------------------------------------- | ||
2419 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
2420 : | mblume | 1622 | Date: 2004/09/13 16:50:00 CDT |
2421 : | Tag: Release_110_49 | ||
2422 : | Description: | ||
2423 : | |||
2424 : | New working version (110.49). NEW BOOTFILES! | ||
2425 : | |||
2426 : | ---------------------------------------------------------------------- | ||
2427 : | mblume | 1621 | Name: Matthias Blume |
2428 : | Date: 2004/09/13 16:20:00 CDT | ||
2429 : | Tag: blume-20040913-config-mlrisc | ||
2430 : | Description: | ||
2431 : | |||
2432 : | Put target "mlrisc" back into the default list. | ||
2433 : | (There is no harm in having it, and some users have expressed their | ||
2434 : | wish to have "mlrisc" included by default.) | ||
2435 : | |||
2436 : | ---------------------------------------------------------------------- | ||
2437 : | jhr | 1620 | Name: John Reppy |
2438 : | Date: 2004/09/13 | ||
2439 : | Tag: jhr-20040913-signals | ||
2440 : | Description: | ||
2441 : | |||
2442 : | Fixed the signal masking code to properly nest mask/unmask operations | ||
2443 : | on a per-signal basis. | ||
2444 : | |||
2445 : | ---------------------------------------------------------------------- | ||
2446 : | mblume | 1619 | Name: Matthias Blume (blume (at) tti - c (dot) org) |
2447 : | Date: 2004/09/08 13:20:00 CDT | ||
2448 : | Tag: blume-20040908-heap-magic | ||
2449 : | Description: | ||
2450 : | |||
2451 : | Bumped the heap macig number to 0x09082004 to account for the changed | ||
2452 : | layout of the ML frame under MacOS X. | ||
2453 : | |||
2454 : | ---------------------------------------------------------------------- | ||
2455 : | allenleung | 1618 | Name: Allen Leung (leunga (at) reservoir (dot) com) |
2456 : | Date: 2004/09/03 11:26:00 EST | ||
2457 : | Tag: leunga-20040903-cygwin-install | ||
2458 : | Description: | ||
2459 : | |||
2460 : | Added a patch to _arch-n-opsys to enable the Cygwin runtime. | ||
2461 : | The Cygwin runtime is turned on by setting the environment | ||
2462 : | variable SMLNJ_CYGWIN_RUNTIME to 1. | ||
2463 : | |||
2464 : | ---------------------------------------------------------------------- | ||
2465 : | mblume | 1615 | Name: Matthias Blume (blume (at) tti - c (dot) org) |
2466 : | mblume | 1617 | Date: 2004/08/31 17:15:00 CDT |
2467 : | Tag: blume-20040831-core | ||
2468 : | Description: | ||
2469 : | |||
2470 : | Added some exports to src/compiler/core.cm upon request by J. Joemann. | ||
2471 : | |||
2472 : | ---------------------------------------------------------------------- | ||
2473 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
2474 : | mblume | 1616 | Date: 2004/08/30 17:55:00 CDT |
2475 : | Tag: blume-20040830-installer | ||
2476 : | Description: | ||
2477 : | |||
2478 : | Upon request by Johannes Joemann: | ||
2479 : | |||
2480 : | - improved ML code of installer to fall back to coping when renaming | ||
2481 : | fails (i.e., when source and target are on different file systems); | ||
2482 : | the code compiles but has yet to be tested in anger | ||
2483 : | - removed mlrisc from list of default targets (config/targets) | ||
2484 : | |||
2485 : | ---------------------------------------------------------------------- | ||
2486 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
2487 : | mblume | 1615 | Date: 2004/08/27 17:20:00 CDT |
2488 : | Tag: blume-20040827-ptreql | ||
2489 : | Description: | ||
2490 : | |||
2491 : | Added ptreql primop to structure InlineT (upon request from Larry | ||
2492 : | Paulson). | ||
2493 : | |||
2494 : | ---------------------------------------------------------------------- | ||
2495 : | allenleung | 1613 | Name: Allen Leung (leunga (at) reservoir (dot) com) |
2496 : | allenleung | 1614 | Date: 2004/08/15 21:21:00 EST |
2497 : | Tag: leunga-110_48-udgraph | ||
2498 : | Description: | ||
2499 : | |||
2500 : | Another bug fix from Carl Hauser: | ||
2501 : | |||
2502 : | diff /net/niflab/smlnj48/src/MLRISC/graphs/udgraph.sml udgraph.sml | ||
2503 : | > 48c48 | ||
2504 : | > < | rmv((e as (k,_))::es,L) = rmv(es,if k = i then es else | ||
2505 : | > e::L) | ||
2506 : | > --- | ||
2507 : | > > | rmv((e as (k,_))::es,L) = rmv(es,if k = i then L else e::L) | ||
2508 : | > Without this, any deletion of an edge in an undirected graph does severe | ||
2509 : | > violence to the graph. | ||
2510 : | |||
2511 : | ---------------------------------------------------------------------- | ||
2512 : | Name: Allen Leung (leunga (at) reservoir (dot) com) | ||
2513 : | allenleung | 1613 | Date: 2004/08/10 23:35:00 EST |
2514 : | Tag: leunga-110_48-ppc | ||
2515 : | Description: | ||
2516 : | |||
2517 : | The IBM/MacOS syntax switch on PPC was incorrectly swapped. | ||
2518 : | |||
2519 : | ---------------------------------------------------------------------- | ||
2520 : | mblume | 1602 | Name: Matthias Blume (blume (at) tti - c (dot) org) |
2521 : | mblume | 1615 | Date: 2004/08/10 12:00:00 CDT |
2522 : | mblume | 1602 | Tag: Release_110_48 |
2523 : | Description: | ||
2524 : | |||
2525 : | New working version (110.48). NEW BOOTFILES! | ||
2526 : | |||
2527 : | ---------------------------------------------------------------------- | ||
2528 : | allenleung | 1599 | Name: Allen Leung (leunga (at) reservoir (dot) com) |
2529 : | Date: 2004/08/09 12:21:00 EST | ||
2530 : | Tag: leunga-110_47-dijsktra | ||
2531 : | Description: | ||
2532 : | |||
2533 : | Bug fix from Carl Hauser: | ||
2534 : | |||
2535 : | blume | 2155 | single_source_shortest_paths in dijkstra.sml was observed to get wrong |
2536 : | allenleung | 1599 | answers (by comparing to single_source_shortest_paths in bellman-ford.sml). |
2537 : | |||
2538 : | blume | 2155 | The problem is that following the expression A.update(dist,s,Num.zero) |
2539 : | allenleung | 1599 | it is necessary to update the priority queue using Q.decreaseWeight(Q,s). |
2540 : | |||
2541 : | ---------------------------------------------------------------------- | ||
2542 : | mblume | 1595 | Name: Matthias Blume (blume (at) tti - c (dot) org) |
2543 : | Date: 2004/08/06 18:10:00 CDT | ||
2544 : | Tag: blume-20040806-cmdline | ||
2545 : | Description: | ||
2546 : | |||
2547 : | Fiddled with handling of command-line options: | ||
2548 : | |||
2549 : | * sml now quits after processing the command line | ||
2550 : | if -H, -S, -h<n>, or -s<n> appears as the last | ||
2551 : | command-line argument | ||
2552 : | * a new option -q terminates the session when encountered on | ||
2553 : | the command line; subsequent arguments will be ignored | ||
2554 : | * bug fixes: short (erroneous) arguments are no longer ignored | ||
2555 : | completely | ||
2556 : | |||
2557 : | ---------------------------------------------------------------------- | ||
2558 : | allenleung | 1593 | Name: Allen Leung (leunga (at) reservoir (dot) com) |
2559 : | Date: 2004/08/04 18:17:00 EST | ||
2560 : | Tag: leunga-110_47-ppc-ibm-asm | ||
2561 : | Description: | ||
2562 : | |||
2563 : | - Added minimal IBM assembly syntax support for PowerPC. | ||
2564 : | |||
2565 : | blume | 2155 | - Cygwin: manually changed the file cygwin.def. Some exported symbols have |
2566 : | allenleung | 1593 | been altered in the runtime. We need an automatic way to keep the file |
2567 : | in sync. | ||
2568 : | |||
2569 : | ---------------------------------------------------------------------- | ||
2570 : | mblume | 1548 | Name: Matthias Blume (blume (at) tti - c (dot) org) |
2571 : | mblume | 1581 | Date: 2004/08/04 14:00:00 CDT |
2572 : | Tag: Release_110_47 | ||
2573 : | Description: | ||
2574 : | |||
2575 : | New working version (110.47). NEW BOOTFILES! | ||
2576 : | |||
2577 : | ---------------------------------------------------------------------- | ||
2578 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
2579 : | mblume | 1580 | Date: 2004/08/03 14:25:00 CDT |
2580 : | Tag: blume-20040803-callingconv | ||
2581 : | Description: | ||
2582 : | |||
2583 : | Added low-level support for choosing C calling conventions by | ||
2584 : | twiddling the type of rawccall. (See | ||
2585 : | src/compiler/Semant/types/cproto.sml for details.) | ||
2586 : | |||
2587 : | ---------------------------------------------------------------------- | ||
2588 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
2589 : | mblume | 1579 | Date: 2004/08/02 15:55:00 CDT |
2590 : | Tag: blume-20040802-backout | ||
2591 : | Description: | ||
2592 : | |||
2593 : | Backed out of change to win32-filesys.c. The earlier patch to | ||
2594 : | get_file_time caused CM to produce files with the wrong time stamp. | ||
2595 : | |||
2596 : | ---------------------------------------------------------------------- | ||
2597 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
2598 : | mblume | 1573 | Date: 2004/08/02 14:45:00 CDT |
2599 : | Tag: blume-20040802-nlffi-win32 | ||
2600 : | Description: | ||
2601 : | |||
2602 : | Added NLFFI support for Win32, adapted from a patch provided by David | ||
2603 : | Hansel. This is currently completely untested. Also, the issue | ||
2604 : | concerning stdcall vs. ccall is still unresolved. | ||
2605 : | |||
2606 : | ---------------------------------------------------------------------- | ||
2607 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
2608 : | mblume | 1572 | Date: 2004/07/30 17:55:00 CDT |
2609 : | Tag: blume-20040730-various | ||
2610 : | Description: | ||
2611 : | |||
2612 : | Gearing up towards 110.47... | ||
2613 : | |||
2614 : | - various minor bugfixes to ml-nlffigen | ||
2615 : | - a beginning of a manual for nlffi | ||
2616 : | |||
2617 : | - eliminated 'export name=value' in config/install.sh as this does | ||
2618 : | not work with certain versions of /bin/sh | ||
2619 : | (Thanks to David King at Motorola for catching this.) | ||
2620 : | |||
2621 : | - several bugfixes provided or suggested by David Hansel at Reactive Systems: | ||
2622 : | - added a test for tm==NULL to gmtime.c and localtime.c | ||
2623 : | - applied patch for incorrect GetFileTime under win32 | ||
2624 : | - toSeconds -> toMilliseconds in Win32/win32-process.sml | ||
2625 : | |||
2626 : | ---------------------------------------------------------------------- | ||
2627 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
2628 : | mblume | 1557 | Date: 2004/07/21 18:20:00 CDT |
2629 : | Tag: blume-20040721-nlffigen | ||
2630 : | Description: | ||
2631 : | |||
2632 : | - Fixed minor issue in ml-nlffigen: Now generate structure T_foo | ||
2633 : | for a typedef to an incomplete type, but leave out the "typ" member. | ||
2634 : | (This is just for consistency.) | ||
2635 : | - Started to produce what is supposed to become better (i.e., comprehensive) | ||
2636 : | documentation of what ml-nlffigen does and produces. | ||
2637 : | |||
2638 : | ---------------------------------------------------------------------- | ||
2639 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
2640 : | mblume | 1548 | Date: 2004/07/14 16:25:00 CDT |
2641 : | Tag: blume-20040714-union | ||
2642 : | Description: | ||
2643 : | |||
2644 : | Added C_UNION to c-calls/c-types.sml and updated the machinery | ||
2645 : | (ml-nlffigen, cproto.sml) that conveys C function interface | ||
2646 : | information to the code generator. | ||
2647 : | |||
2648 : | However, the actual architecture-specific implementation of function | ||
2649 : | arguments and results that are C unions is still not implemented. | ||
2650 : | |||
2651 : | ---------------------------------------------------------------------- | ||
2652 : | allenleung | 1539 | Name: Allen Leung (leunga (at) reservoir (dot) com) |
2653 : | allenleung | 1546 | Date: 2004/07/14 14:38:00 EST |
2654 : | Tag: leunga-110_46_1-ppc-lwzu | ||
2655 : | Description: | ||
2656 : | |||
2657 : | Added these instructions to the PowerPC architecture: | ||
2658 : | LBZU(X), LHZU(X), LWZU(X), | ||
2659 : | blume | 2155 | STWU(X), STFDU, STFSU |
2660 : | |||
2661 : | allenleung | 1546 | etc... |
2662 : | |||
2663 : | Note: I haven't added their instruction encoding into the description. | ||
2664 : | |||
2665 : | ---------------------------------------------------------------------- | ||
2666 : | Name: Allen Leung (leunga (at) reservoir (dot) com) | ||
2667 : | allenleung | 1539 | Date: 2004/07/13 15:04:00 EST |
2668 : | Tag: leunga-110_46_1-ppc-lwarx | ||
2669 : | Description: | ||
2670 : | |||
2671 : | Added the two instructions LWARX and STWCX to the PowerPC | ||
2672 : | blume | 2155 | instruction set. |
2673 : | allenleung | 1539 | |
2674 : | A (untested) rewrite of loop-structure.sml. The old version | ||
2675 : | is completely broken. | ||
2676 : | |||
2677 : | ---------------------------------------------------------------------- | ||
2678 : | mblume | 1479 | Name: Matthias Blume (blume (at) tti - c (dot) org) |
2679 : | mblume | 1537 | Date: 2004/07/13 13:50:00 CDT |
2680 : | Tag: blume-20040713-nlffi | ||
2681 : | Description: | ||
2682 : | |||
2683 : | - use paramAlloc to report c-calls with too many arguments | ||
2684 : | (for PPC version where parameter area is pre-allocated) | ||
2685 : | - added ccall_maxargspace to machspec (to implement the above) | ||
2686 : | - made "make" commend in CM's "make" tool configurable | ||
2687 : | - added option (default: on) for passing the name of the SML/NJ's "bin" | ||
2688 : | directory to "make"; the call looks like this: | ||
2689 : | |||
2690 : | make <options> SMLNJ_BINDIR=<dir> <target> | ||
2691 : | |||
2692 : | This can be used by the Makefile to, e.g., pick the "right" version | ||
2693 : | of ml-nlffigen. | ||
2694 : | - minor code tweaks | ||
2695 : | |||
2696 : | ---------------------------------------------------------------------- | ||
2697 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
2698 : | mblume | 1532 | Date: 2004/07/12 22:50:00 CDT |
2699 : | Tag: blume-110_46_1-macosx-nlffi | ||
2700 : | Description: | ||
2701 : | |||
2702 : | NLFFI under Mac OS X now working (sort of). This is largely untested, | ||
2703 : | though. | ||
2704 : | |||
2705 : | Note: 1. You have to make a new, clean build of the runtime system. | ||
2706 : | 2. There are new BOOTFILES, you have to use them! | ||
2707 : | (Doing the bootstrap process yourself would be *very* painful! | ||
2708 : | If you absolutely have to do it, build the system under | ||
2709 : | a different architecture and then cross-compile.) | ||
2710 : | |||
2711 : | Version bumped to 110.46.1 to account for runtime data format changes. | ||
2712 : | |||
2713 : | ---------------------------------------------------------------------- | ||
2714 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
2715 : | mblume | 1515 | Date: 2004/06/18 14:30:00 CDT |
2716 : | Tag: blume-20040618-unix | ||
2717 : | Description: | ||
2718 : | |||
2719 : | Changed the implementation of structure Unix so that the same stream | ||
2720 : | is returned every time one of the {text,bin}{In,Out}streamOf functions | ||
2721 : | is invoked on the same proc. This is not what the spec currently | ||
2722 : | says -- although IMO it arguably should. (See discussion below.) | ||
2723 : | |||
2724 : | ---------------------------------------------------------------------- | ||
2725 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
2726 : | mblume | 1507 | Date: 2004/06/17 18:15:00 CDT |
2727 : | Tag: Release_110_46 | ||
2728 : | Description: | ||
2729 : | |||
2730 : | New working version (110.46). NEW BOOTFILES! | ||
2731 : | |||
2732 : | ---------------------------------------------------------------------- | ||
2733 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
2734 : | mblume | 1506 | Date: 2004/06/17 17:20:00 CDT |
2735 : | Tag: blume-20040617-timer-unix | ||
2736 : | Description: | ||
2737 : | |||
2738 : | Changed the interface of structures Timer and Unix to match the most | ||
2739 : | recent Basis spec. | ||
2740 : | |||
2741 : | In the case of Unix there still seems to be an open/weird issue: | ||
2742 : | |||
2743 : | The {text,bin}{In,Out}streamOf functions are supposed to create | ||
2744 : | fresh streams whenever they are called -- as opposed to have them | ||
2745 : | return the same stream every time. This design is supposed to | ||
2746 : | prevent space leaks caused by proc values hanging on to streams. | ||
2747 : | |||
2748 : | The reap function, on the other hand, is supposed to close the | ||
2749 : | streams. This cannot be done without having a handle on the | ||
2750 : | stream in proc after all... | ||
2751 : | |||
2752 : | I took the liberty to implement the following stopgap solution: | ||
2753 : | |||
2754 : | The proc value hangs on to the most recently created stream(s). | ||
2755 : | Reap closes those. If either or both of the two streams hadn't | ||
2756 : | been created at all yet, then reap will close the corresponding | ||
2757 : | file descriptors directly. | ||
2758 : | |||
2759 : | PS: I don't understand the original space leak argument anymore. If | ||
2760 : | a proc hangs on to the imperative stream, then I/O operations on those | ||
2761 : | will advance the state of the cached stream and avoid the space leak. | ||
2762 : | |||
2763 : | ---------------------------------------------------------------------- | ||
2764 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
2765 : | mblume | 1500 | Date: 2004/05/28 16:45:00 CDT |
2766 : | Tag: blume-20040528-basis | ||
2767 : | Description: | ||
2768 : | |||
2769 : | Added signature PACK_REAL and exported functor PrimIO. | ||
2770 : | |||
2771 : | ---------------------------------------------------------------------- | ||
2772 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
2773 : | mblume | 1498 | Date: 2004/05/25 16:00:00 CDT |
2774 : | Tag: blume-20040525-group-owner | ||
2775 : | Description: | ||
2776 : | |||
2777 : | CM now ignores (but still accepts) the "owner" information in group | ||
2778 : | descriptions. The owner of a group is its next enclosing | ||
2779 : | library. Each group must have a unique owner. (There is a virtual | ||
2780 : | mblume | 1505 | "toplevel" library that owns groups which are not nested within a real |
2781 : | mblume | 1498 | library.) Previously, each group had to explicitly declare its owner, |
2782 : | and CM would check that such a declaration is correct. The new scheme | ||
2783 : | is to have CM check that for each group there is precisely one owning | ||
2784 : | library. | ||
2785 : | |||
2786 : | The advantage of the new scheme is that the programmer no longer needs | ||
2787 : | to maintain the somewhat annoying owner information. The downside is | ||
2788 : | that CM cannot enforce the ownership rule across multiple runs of | ||
2789 : | CM.make. Fortunately, enclosing the same group in two different | ||
2790 : | libraries A and B which are not part of the same program does not | ||
2791 : | cause real problems. | ||
2792 : | |||
2793 : | ---------------------------------------------------------------------- | ||
2794 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
2795 : | mblume | 1496 | Date: 2004/05/20 16:00:00 CDT |
2796 : | Tag: blume-20040520-win32 | ||
2797 : | Description: | ||
2798 : | |||
2799 : | Made the win32 version work again. (Strangely, a misplaced comma had | ||
2800 : | slipped into win32-process.c which prevented the runtime from being | ||
2801 : | compiled correctly.) | ||
2802 : | |||
2803 : | Also, included a minor addition to ml-build.bat analogous to what was | ||
2804 : | done in blume-20040519-ml-build. | ||
2805 : | |||
2806 : | ---------------------------------------------------------------------- | ||
2807 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
2808 : | mblume | 1494 | Date: 2004/05/19 22:10:00 CDT |
2809 : | Tag: blume-20040519-ml-build | ||
2810 : | Description: | ||
2811 : | |||
2812 : | Arranged for ml-build to clean up after itself a little bit better. | ||
2813 : | The script generates a temporary SML source file and compiles it using | ||
2814 : | CM, so CM generates metadata (GUID, SKEL, objectfile) for it. It now | ||
2815 : | gets rid of those at the end, so they don't accumulate under .cm. | ||
2816 : | |||
2817 : | This required a minor change to install.sh because the name of the | ||
2818 : | metadata directory (default: .cm) is actually configurable at | ||
2819 : | installation time. | ||
2820 : | |||
2821 : | ---------------------------------------------------------------------- | ||
2822 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
2823 : | mblume | 1491 | Date: 2004/05/18 15:50:00 CDT |
2824 : | Tag: blume-20040518-mkreader | ||
2825 : | Description: | ||
2826 : | |||
2827 : | Added Posix.IO.mk{Bin,Text}{Reader,Writer} by lifting their respective | ||
2828 : | implementations from internal modules PosixBinPrimIO and PosixTextPrimIO. | ||
2829 : | |||
2830 : | ---------------------------------------------------------------------- | ||
2831 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
2832 : | mblume | 1489 | Date: 2004/05/11 14:35:00 CDT |
2833 : | Tag: blume-20040511-win32sock | ||
2834 : | Description: | ||
2835 : | |||
2836 : | Added previously missing support for many socket-related functions | ||
2837 : | under win32. Thanks to David Hansel <hansel@reactive-systems.com> | ||
2838 : | for the voluminous patch! | ||
2839 : | |||
2840 : | (I have not tested this patch under win32 yet.) | ||
2841 : | |||
2842 : | mblume | 1490 | Here is David's e-mail: |
2843 : | |||
2844 : | Hi, | ||
2845 : | |||
2846 : | Attached to this email you find a diff against sml/nj 110.45 | ||
2847 : | that will enable socket support under Windows. | ||
2848 : | |||
2849 : | To apply the patch (using unix or cygwin) | ||
2850 : | 1) gunzip runtime.diff.gz | ||
2851 : | 2) "cd" into "src/runtime" in the source tree of a fresh | ||
2852 : | 110.45 installation. | ||
2853 : | 3) patch -p 1 < [your/path/to]runtime.diff | ||
2854 : | |||
2855 : | blume | 2155 | The code compiles fine but has NOT yet been extensively tested. |
2856 : | I only ran a few tests for basic socket client functionality | ||
2857 : | (which worked fine). Especially the functions that use ioctl | ||
2858 : | mblume | 1490 | are not tested at all and might not work (see below). |
2859 : | |||
2860 : | I implemented this since we want to move to a newer version of sml/nj | ||
2861 : | but need socket support in order to use it. This is the first time I | ||
2862 : | even had a look at the sml/nj source, so please review my changes | ||
2863 : | before making this part of the distribution! Here are a few issues | ||
2864 : | blume | 2155 | that I think might be better for someone to solve who is more |
2865 : | mblume | 1490 | familiar with the sml/nj source (and socket programming): |
2866 : | |||
2867 : | - getnetbyaddr.c and getnetbyname.c will raise a "not implemented" | ||
2868 : | exception since I could not figure out what the windows equivalent | ||
2869 : | of these functions is | ||
2870 : | |||
2871 : | - In sockets-osdep.h there are a some #include statements that are | ||
2872 : | only used in a few files that include sockets-osdep.h | ||
2873 : | |||
2874 : | - In smlnj-sock-lib.c, function init_fn() calls WSAStartup() but | ||
2875 : | does not process its return value since I don't know how to | ||
2876 : | report an error upwards. | ||
2877 : | |||
2878 : | - It would probably be good to have a call to WSACleanup() when | ||
2879 : | blume | 2155 | the library is unloaded (if there is such a possibility). |
2880 : | mblume | 1490 | Otherwise I think Windows will take care of this automatically |
2881 : | when the process finishes. | ||
2882 : | |||
2883 : | - I used ioctlsocket() as a replacement for ioctl() but I have | ||
2884 : | no idea if that is actually the proper replacement on Windows. | ||
2885 : | |||
2886 : | - All these issues are marked in the code by "FIXME" comments. | ||
2887 : | |||
2888 : | blume | 2155 | We use sml/nj extensively in our products and are quite happy |
2889 : | with it. I hope this contribution will help you. | ||
2890 : | mblume | 1490 | |
2891 : | Keep up the good work! | ||
2892 : | |||
2893 : | David | ||
2894 : | |||
2895 : | mblume | 1489 | ---------------------------------------------------------------------- |
2896 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
2897 : | mblume | 1488 | Date: 2004/05/11 14:20:00 CDT |
2898 : | Tag: blume-20040511-installml | ||
2899 : | Description: | ||
2900 : | |||
2901 : | Fixed two bugs in installml script. (Thanks to Vesa A. Norrman for | ||
2902 : | the patch.) | ||
2903 : | |||
2904 : | ---------------------------------------------------------------------- | ||
2905 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
2906 : | mblume | 1487 | Date: 2004/05/11 14:05:00 CDT |
2907 : | Tag: blume-20040511-nlffi-netbsd | ||
2908 : | Description: | ||
2909 : | |||
2910 : | Added support for nlffi under netbsd. (Thanks to Vesa A. Norrman for | ||
2911 : | the patch.) | ||
2912 : | |||
2913 : | ---------------------------------------------------------------------- | ||
2914 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
2915 : | mblume | 1486 | Date: 2004/05/11 12:05:00 CDT |
2916 : | Tag: blume-20040511-exports | ||
2917 : | Description: | ||
2918 : | |||
2919 : | As per request by Adam Chlipala <adam@hcoop.net>, extended various | ||
2920 : | export lists in compiler-related .cm-files. | ||
2921 : | |||
2922 : | ---------------------------------------------------------------------- | ||
2923 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
2924 : | mblume | 1485 | Date: 2004/05/11 11:35:00 CDT |
2925 : | Tag: blume-20040511-allsource | ||
2926 : | Description: | ||
2927 : | |||
2928 : | The installer now honors the "src-smlnj" target again, although its meaning | ||
2929 : | has changed from "all sources required for the compiler" to "all sources | ||
2930 : | the installer knows about". In other words, if you enable "src-smlnj" | ||
2931 : | in the "targets" file, then the installer will pull in sources for | ||
2932 : | everything. (Notice that this refers to source code only. Compiled | ||
2933 : | code is still only installed for modules that were requested explicitly | ||
2934 : | or which are required for other modules that were requested explicitly.) | ||
2935 : | |||
2936 : | ---------------------------------------------------------------------- | ||
2937 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
2938 : | mblume | 1482 | Date: 2004/04/23 17:40:00 CDT |
2939 : | Tag: blume-20040423-ieee-scan | ||
2940 : | Description: | ||
2941 : | |||
2942 : | Fixed IEEEReal.scan (and .fromString) so that if there is an overflow | ||
2943 : | in the exponent calculation we get INF or ZERO (depending on the mantissa | ||
2944 : | and the sign of the exponent). | ||
2945 : | |||
2946 : | ---------------------------------------------------------------------- | ||
2947 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
2948 : | mblume | 1480 | Date: 2004/04/23 10:40:00 CDT |
2949 : | Tag: blume-20040423-ml-build | ||
2950 : | Description: | ||
2951 : | |||
2952 : | The ml-build script now terminates with a non-0 status when something | ||
2953 : | goes wrong. | ||
2954 : | |||
2955 : | ---------------------------------------------------------------------- | ||
2956 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
2957 : | mblume | 1479 | Date: 2004/04/22 16:35:00 CDT |
2958 : | Tag: blume-20040422-Option | ||
2959 : | Description: | ||
2960 : | |||
2961 : | Made exception Option to be the same as exception Option.Option | ||
2962 : | (as it should be). | ||
2963 : | |||
2964 : | ---------------------------------------------------------------------- | ||
2965 : | allenleung | 1468 | Name: Allen Leung (leunga (at) reservoir (dot) com) |
2966 : | Date: 2004/03/19 14:40:00 EST | ||
2967 : | Tag: leunga-20040319-cygwin-nlffi | ||
2968 : | Description: | ||
2969 : | |||
2970 : | Fixed the runtime so that ml-nlffi-lib runs on the cygwin version | ||
2971 : | of SML/NJ. The problem is that | ||
2972 : | |||
2973 : | lib = dlopen(NULL, ...) | ||
2974 : | f = dlsym(lib, "malloc"); | ||
2975 : | |||
2976 : | blume | 2155 | does not work on Windows unless we explicitly export symbols |
2977 : | such as 'malloc' during linking. We fixed this by explicitly | ||
2978 : | allenleung | 1468 | exporting the required symbols with the magic gcc incantation: |
2979 : | |||
2980 : | -Wl,--export-all cygwin.def | ||
2981 : | |||
2982 : | where cygwin.def is a file containing all the symbols that | ||
2983 : | blume | 2155 | we wish to export. |
2984 : | allenleung | 1468 | |
2985 : | blume | 2155 | I suspect this is a Windows problem and we'll have to |
2986 : | do the same (somehow with windows compilers) when | ||
2987 : | allenleung | 1468 | we build the native win32 version with the system |
2988 : | blume | 2155 | calls LoadLibrary/GetProcAddress. |
2989 : | allenleung | 1468 | |
2990 : | ---------------------------------------------------------------------- | ||
2991 : | mblume | 1384 | Name: Matthias Blume (blume (at) tti - c (dot) org) |
2992 : | mblume | 1465 | Date: 2004/03/04 16:35:00 CST |
2993 : | Tag: blume-20040304-intinf-fmt | ||
2994 : | Description: | ||
2995 : | |||
2996 : | Fixed problem with IntInf.fmt (sign would show up on the right instead | ||
2997 : | of on the left for BIN, OCT, and HEX). | ||
2998 : | |||
2999 : | ---------------------------------------------------------------------- | ||
3000 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
3001 : | mblume | 1464 | Date: 2004/03/04 11:25:00 CST |
3002 : | Tag: blume-20040304-symlinks | ||
3003 : | Description: | ||
3004 : | |||
3005 : | Fixed problem with installer script (unix only) where bin/ml-yacc and | ||
3006 : | friends pointed (via symlinks) to absolute locations instead of just | ||
3007 : | .run-sml. This was reported by Vesa A Norrman. | ||
3008 : | |||
3009 : | ---------------------------------------------------------------------- | ||
3010 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
3011 : | mblume | 1455 | Date: 2004/02/13 14:50:00 CST |
3012 : | Tag: Release_110_45 | ||
3013 : | Description: | ||
3014 : | |||
3015 : | New working version (110.45). New bootfiles. | ||
3016 : | |||
3017 : | ---------------------------------------------------------------------- | ||
3018 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
3019 : | mblume | 1448 | Date: 2004/01/26 15:15:15 CST |
3020 : | Tag: blume-20040126-toplevel | ||
3021 : | Description: | ||
3022 : | |||
3023 : | Improved handling of exceptions at the interactive toplevel. | ||
3024 : | |||
3025 : | ---------------------------------------------------------------------- | ||
3026 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
3027 : | Date: 2004/01/26 11:25:00 | ||
3028 : | mblume | 1446 | Tag: blume-20040126-app |
3029 : | Description: | ||
3030 : | |||
3031 : | Type of top-level "app" corrected. | ||
3032 : | mblume | 1447 | Added code for setting vp_limitPtrMask to Win32-specific runtime. |
3033 : | mblume | 1446 | |
3034 : | ---------------------------------------------------------------------- | ||
3035 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
3036 : | mblume | 1427 | Date: 2003/11/18 17:10 CST |
3037 : | Tag: blume-20031118-basis-fiddle | ||
3038 : | Description: | ||
3039 : | |||
3040 : | - changed Timer interface to what might become the spec | ||
3041 : | - POSIX_FLAGS -> BIT_FLAGS according to spec | ||
3042 : | - some other minor discrepancies wrt. spec eliminated | ||
3043 : | |||
3044 : | ---------------------------------------------------------------------- | ||
3045 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
3046 : | mblume | 1414 | Date: 2003/11/06 12:00:00 CST |
3047 : | Tag: Release_110_44 | ||
3048 : | Description: | ||
3049 : | |||
3050 : | New working version (110.44). New bootfiles. | ||
3051 : | |||
3052 : | ---------------------------------------------------------------------- | ||
3053 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
3054 : | mblume | 1411 | Date: 2003/11/04 11:50:00 CST |
3055 : | Tag: blume-20031104-move-libraries | ||
3056 : | Description: | ||
3057 : | |||
3058 : | Eliminated the "dont_move_libraries" directive in config/targets. | ||
3059 : | (The mechanism was broken and could not be fixed easily. Moreover, | ||
3060 : | there does not seem to be any reason not to move all libraries into | ||
3061 : | lib during installation. I originally implemented this directive as a | ||
3062 : | backward-compatibility feature when I first introduced the new CM. | ||
3063 : | Now that things have been stable for a long time and going back to the | ||
3064 : | old CM is not an option, there is no reason to keep it around.) | ||
3065 : | |||
3066 : | ---------------------------------------------------------------------- | ||
3067 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
3068 : | mblume | 1409 | Date: 2003/11/03 16:00:00 CST |
3069 : | Tag: blume-20031103-installdir | ||
3070 : | Description: | ||
3071 : | |||
3072 : | Made installer honor INSTALLDIR variable again. (Thanks to Chris | ||
3073 : | Richards for pointing out the problem and providing the solution.) | ||
3074 : | |||
3075 : | ---------------------------------------------------------------------- | ||
3076 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
3077 : | mblume | 1400 | Date: 2003/10/01 17:05:00 CDT |
3078 : | Tag: blume-20031001-lal-mlrisc | ||
3079 : | Description: | ||
3080 : | |||
3081 : | MLRISC bug fix from Lal. | ||
3082 : | |||
3083 : | ---------------------------------------------------------------------- | ||
3084 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
3085 : | mblume | 1399 | Date: 2003/09/30 16:10:00 CDT |
3086 : | Tag: blume-20030930-primio-bat | ||
3087 : | mblume | 1398 | Description: |
3088 : | |||
3089 : | 1. Added openVector, nullRd, and nullWr to PRIM_IO. | ||
3090 : | 2. Improved .bat files (for Win32 port) to make things work under Win95. | ||
3091 : | (thanks to Aaron S. Hawley for this one) | ||
3092 : | |||
3093 : | ---------------------------------------------------------------------- | ||
3094 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
3095 : | mblume | 1395 | Date: 2003/09/26 16:05:00 CDT |
3096 : | Tag: blume-20030926-wrappriv | ||
3097 : | Description: | ||
3098 : | |||
3099 : | Added missing wrapper for privilege "primitive" in $smlnj/viscomp/core.cm. | ||
3100 : | |||
3101 : | ---------------------------------------------------------------------- | ||
3102 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
3103 : | mblume | 1394 | Date: 2003/09/26 15:00:00 CDT |
3104 : | Tag: blume-20030926-110_43_3 | ||
3105 : | Description: | ||
3106 : | |||
3107 : | - additional cleanup | ||
3108 : | - version number bump, NEW BOOTFILES | ||
3109 : | |||
3110 : | ---------------------------------------------------------------------- | ||
3111 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
3112 : | mblume | 1393 | Date: 2003/09/26 12:00:00 CDT |
3113 : | Tag: blume-20030926-ppautoload | ||
3114 : | Description: | ||
3115 : | |||
3116 : | I modified the read-eval-print loop so that the autoloader gets | ||
3117 : | invoked whenever the prettyprinter tries to look up a symbol that | ||
3118 : | is not currently defined in the toplevel environment but which | ||
3119 : | appears in CM's autoload registry. As a result, we see far fewer of | ||
3120 : | those ?.Foo.Bar.xxx names in the prettyprinter's output. | ||
3121 : | |||
3122 : | In addition to this I tried to clean up some pieces of the Basis | ||
3123 : | implementation (e.g., Socket, Word8Array) in order to prevent other | ||
3124 : | instances of these ?.Foo.Bar.xxx names from being printed. | ||
3125 : | |||
3126 : | The mechanism that picks names for types still needs some work, though. | ||
3127 : | (Right now it seems that if there is a type A.t which is defined to | ||
3128 : | mblume | 1394 | be B.u, but B is unavailable at toplevel, then A.t gets printed as |
3129 : | "?.B.u" although the perhaps more sensible solution would be to use | ||
3130 : | mblume | 1393 | "A.t" in this case. In other words, the prettyprinter should follow |
3131 : | a chain of DEFtycs not farther than there are corresponding toplevel | ||
3132 : | names in the current environment.) | ||
3133 : | |||
3134 : | ---------------------------------------------------------------------- | ||
3135 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
3136 : | mblume | 1391 | Date: 2003/09/24 16:31:00 CDT |
3137 : | Tag: blume-20030924-installer | ||
3138 : | Description: | ||
3139 : | |||
3140 : | Another installer tweak: All the ML code for the installer is now | ||
3141 : | compiled during CMB.make and put into a little library called | ||
3142 : | $smlnj/installer.cm. The installation then simply invokes | ||
3143 : | |||
3144 : | sml -m $smlnj/installer.cm | ||
3145 : | |||
3146 : | and everything happens automagically. | ||
3147 : | |||
3148 : | Win32: ML code senses value of environment variable SMLNJ_HOME. | ||
3149 : | Unix: ML code senses values of environment variables ROOT, CONFIGDIR, | ||
3150 : | and BINDIR. | ||
3151 : | |||
3152 : | The new scheme guarantees that the ML code responsible for the installation | ||
3153 : | is in sync with the APIs of the main system. Also, the installer is | ||
3154 : | somewhat faster because the installer script is precompiled. | ||
3155 : | |||
3156 : | ---------------------------------------------------------------------- | ||
3157 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
3158 : | mblume | 1390 | Date: 2003/09/24 15:35:00 CDT |
3159 : | Tag: blume-20030924-synsock | ||
3160 : | Description: | ||
3161 : | |||
3162 : | Added a signature SYNCHRONOUS_SOCKET to basis.cm. This is like SOCKET | ||
3163 : | but excludes all non-blocking operations. Defined SOCKET (in Basis) | ||
3164 : | and CML_SOCKET in terms of SYNCHRONOUS_SOCKET. Removed superfluous | ||
3165 : | implementations of non-blocking operations from CML's Socket | ||
3166 : | structure. | ||
3167 : | |||
3168 : | ---------------------------------------------------------------------- | ||
3169 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
3170 : | mblume | 1389 | Date: 2003/09/24 15:10:05 CDT |
3171 : | Tag: blume-20030924-sockets | ||
3172 : | Description: | ||
3173 : | |||
3174 : | 1. Fixed SOCKET API and implementation to match Basis spec. | ||
3175 : | This required changing the internal representation of sockets to one | ||
3176 : | that remembers (for each socket file descriptor) whether it is currently | ||
3177 : | blocking or non-blocking. This state is maintained lazily (i.e., a system | ||
3178 : | call is made only if the state actually needs to change). | ||
3179 : | |||
3180 : | 2. OS-specific details of sockets were moved into separate files, thus | ||
3181 : | making it possible to unify the bulk of the socket implementations | ||
3182 : | between Unix and Win32. | ||
3183 : | |||
3184 : | 3. CML's socket API changed accordingly. | ||
3185 : | (Note that we need to remove non-blocking functions from this API | ||
3186 : | since they are redundant in the case of CML!) | ||
3187 : | |||
3188 : | 4. CML's socket implementation now makes use of non-blocking functions | ||
3189 : | provided by Basis, thus removing all OS-dependent code from this part | ||
3190 : | of CML. | ||
3191 : | |||
3192 : | 5. Changed Real64.precision from 52 to 53. Minor cleanup in Real64 code. | ||
3193 : | |||
3194 : | ---------------------------------------------------------------------- | ||
3195 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
3196 : | mblume | 1387 | Date: 2003/09/22 12:10:00 CDT |
3197 : | mblume | 1388 | Tag: blume-20030922-110_43_2 |
3198 : | mblume | 1387 | Description: |
3199 : | |||
3200 : | Made a new interim version and bootfiles for developer's bootstrapping | ||
3201 : | convenience. | ||
3202 : | |||
3203 : | 110.43.2 -- NEW BOOTFILES | ||
3204 : | |||
3205 : | ---------------------------------------------------------------------- | ||
3206 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
3207 : | mblume | 1385 | Date: 2003/09/19 15:55:00 CDT |
3208 : | Tag: blume-20030919-cmdir | ||
3209 : | Description: | ||
3210 : | |||
3211 : | 1. new-install.sh -> install.sh | ||
3212 : | 2. changed default CM "metadata" directory name to ".cm" (instead of "CM") | ||
3213 : | 3. tweaked installer so that another name instead of .cm can be chosen | ||
3214 : | at install time (by setting the CM_DIR_ARC environment variable | ||
3215 : | during installation); once installation is complete, the name is | ||
3216 : | fixed | ||
3217 : | |||
3218 : | ---------------------------------------------------------------------- | ||
3219 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
3220 : | mblume | 1384 | Date: 2003/09/18 16:00:00 CDT |
3221 : | Tag: blume-20030918-110_43_1 | ||
3222 : | Description: | ||
3223 : | |||
3224 : | Made a new interim version and bootfiles for developer's bootstrapping | ||
3225 : | convenience. | ||
3226 : | |||
3227 : | 110.43.1 -- NEW BOOTFILES | ||
3228 : | |||
3229 : | ---------------------------------------------------------------------- | ||
3230 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
3231 : | mblume | 1383 | Date: 2003/09/18 15:20:00 CDT |
3232 : | Tag: blume-20030918-misc | ||
3233 : | Description: | ||
3234 : | |||
3235 : | 1. Exported fractionsPerSecond etc. from TimeImp (but not from Time as | ||
3236 : | this seems to be controversial at the moment) and used those in | ||
3237 : | Posix.ProcEnv.times. | ||
3238 : | |||
3239 : | 2. Added Time.{from,to}Nanoseconds to Time. | ||
3240 : | |||
3241 : | 3. Improved Real.{from,to}LargeInt by avoiding needless calculations. | ||
3242 : | For example, fromLargeInt never needs to look at more than 3 "big | ||
3243 : | digits" to get its 53 bits of precision. | ||
3244 : | |||
3245 : | ---------------------------------------------------------------------- | ||
3246 : | mblume | 1384 | Name: Matthias Blume (blume (at) tti - c (dot) org) |
3247 : | mblume | 1381 | Date: 2003/09/17 16:30:00 CDT |
3248 : | Tag: blume-20030917-real32-slices | ||
3249 : | Description: | ||
3250 : | |||
3251 : | Added an entry to the primitive environment | ||
3252 : | (compiler/Semant/statenv/prim.sml) for int32->real64 conversion and | ||
3253 : | added code to compiler/CodeGen/main/mlriscGen.sml to implement it. | ||
3254 : | |||
3255 : | Removed some of the "magic" constants in real64.sml and replaced them | ||
3256 : | with code that generates these values from their corresponding | ||
3257 : | integer counterparts. | ||
3258 : | |||
3259 : | Made all(?) the slice-related changes to the Basis and made everything | ||
3260 : | compile again... | ||
3261 : | |||
3262 : | ---------------------------------------------------------------------- | ||
3263 : | mblume | 1368 | Name: Matthias Blume (blume (at) tti - c (dot) org) |
3264 : | mblume | 1375 | Date: 2003/09/15 17:45:00 CDT |
3265 : | Tag: blume-20030915-rbase | ||
3266 : | Description: | ||
3267 : | |||
3268 : | Fixed bug in Real.fromLargeInt. | ||
3269 : | |||
3270 : | ---------------------------------------------------------------------- | ||
3271 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
3272 : | mblume | 1369 | Date: 2003/09/13 18:11:00 CDT |
3273 : | Tag: blume-20030913-libinstall | ||
3274 : | Description: | ||
3275 : | |||
3276 : | Minor bugfix in config/libinstall (set anchor with path to | ||
3277 : | standalone tool after installing it, otherwise libraries that | ||
3278 : | need ml-lex or ml-yacc won't compile the first time the installer | ||
3279 : | runs). | ||
3280 : | |||
3281 : | ---------------------------------------------------------------------- | ||
3282 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
3283 : | mblume | 1368 | Date: 2003/09/12 11:45:00 CDT |
3284 : | Tag: blume-20030912-various | ||
3285 : | Description: | ||
3286 : | |||
3287 : | - fixed bug in Real.toLargeInt | ||
3288 : | - fixed bug in Posix.ProcEnv.times | ||
3289 : | - changed inputLine functions to return an option | ||
3290 : | - minor installer improvements / bugfixes | ||
3291 : | - changed default @SMLalloc parameter for x86/celeron to 64k | ||
3292 : | |||
3293 : | ---------------------------------------------------------------------- | ||
3294 : | Name: Matthias Blume (blume (at) tti - c (dot) org) | ||
3295 : | mblume | 1356 | Date: 2003/09/09 22:00:00 CDT |
3296 : | Tag: Release_110_43 | ||
3297 : | Description: | ||
3298 : | |||
3299 : | New working release 110.43. New bootfiles. | ||
3300 : | |||
3301 : | ---------------------------------------------------------------------- | ||
3302 : | mblume | 1368 | Name: Matthias Blume (blume (at) tti - c (dot) org) |
3303 : | mblume | 1354 | Date: 2003/09/09 19:20:00 CDT |
3304 : | Tag: blume-20030909-installer | ||
3305 : | Description: | ||
3306 : | |||
3307 : | Rewrote large parts of config/install.sh in SML (config/libinstall.sml). | ||
3308 : | Modified config/install.bat to take advantage of it. Also modified | ||
3309 : | config/install.sh (and called it config/new-install.sh) to take advantage | ||
3310 : | of it on Unix systems. (The SML code is (supposed to be) platform- | ||
3311 : | independent.) | ||
3312 : | |||
3313 : | The installer can now install everything under Win32 | ||
3314 : | as well as under *nix as long as it compiles. | ||
3315 : | |||
3316 : | Other changes: | ||
3317 : | |||
3318 : | - made CML compile again under Win32 | ||
3319 : | - made eXene compile under Win32 (by providing a fake structure UnixSock | ||
3320 : | and by using OS.Process.getEnv instead of Posix.ProcEnv.getenv) | ||
3321 : | - fixed a bug in nowhere: it assumed that type OS.Process.status is the | ||
3322 : | same as type int; under Win32 it isn't | ||
3323 : | - fixed some slice-related problems in the win32-specific parts of CML | ||
3324 : | - added a functor argument "sameVol" to os-path-fn.sml in the Basis | ||
3325 : | (under Win32, the volume name is case-insensitive, and the | ||
3326 : | OS.Path code compares volume names for equality) | ||
3327 : | |||
3328 : | ---------------------------------------------------------------------- | ||
3329 : | mblume | 1368 | Name: Matthias Blume (blume (at) tti - c (dot) org) |
3330 : | mblume | 1353 | Date: 2003/09/08 11:55:00 CDT |
3331 : | Tag: blume-20030908-fullpath | ||
3332 : | Description: | ||
3333 : | |||
3334 : | Made Win32 version of OS.FileSys.fullPath return current directory | ||
3335 : | when given an empty string. This is what the spec says, and incidentally, | ||
3336 : | CM depends on it. (CM otherwise goes into an infinite loop in certain | ||
3337 : | cases when presented with the name of a non-existing .cm file.) | ||
3338 : | |||
3339 : | ---------------------------------------------------------------------- | ||
3340 : | mblume | 1368 | Name: Matthias Blume (blume (at) tti - c (dot) org) |
3341 : | mblume | 1350 | Date: 2003/09/04 16:30:00 CDT |
3342 : | mblume | 1352 | Tag: blume-20030905-slices-etc |
3343 : | mblume | 1350 | Description: |
3344 : | |||
3345 : | 1. Changed interface to vectors and arrays in Basis to match | ||
3346 : | (draft) Basis spec. | ||
3347 : | 2. Added signatures and implementations of slices according to | ||
3348 : | Basis spec. | ||
3349 : | 3. Edited source code throughout the system to make it compile again | ||
3350 : | under 1. and 2. (In some cases code had to be added to have it | ||
3351 : | match the new signatures.) | ||
3352 : | 4. MLRISC should be backward-compatible: the copies of the originals | ||
3353 : | of files that needed to change under 3. were retained, the .cm files | ||
3354 : | check the compiler version number and use old versions when | ||
3355 : | appropriate. | ||
3356 : | 5. Changed type of OS.FileSys.readDir and Posix.FileSys.readdir to | ||
3357 : | dirstream -> string option (in accordance with Basis spec). | ||
3358 : | 6. When generating code that counts lines, ml-lex used function | ||
3359 : | CharVector.foldli, taking advantage of its old interface. | ||
3360 : | This has been replaced with the corresponding code from | ||
3361 : | CharVectorSlice. (html-lex must be re-lexed!) | ||
3362 : | 7. BitArray in smlnj-lib/Util has been extended/modified to match the | ||
3363 : | new MONO_ARRAY signature. (Do we need BitArraySlice?) | ||
3364 : | 8. Removed temporary additions (fromInternal, toInternal) from the | ||
3365 : | (now obsolete) IntInf in smlnj-lib/Util. | ||
3366 : | 9. Cleaned up structure Byte. | ||
3367 : | 10. Added localOffset, scan, and fromString to Date (according to spec). | ||
3368 : | Cleaned/corrected implementation of Date. | ||
3369 : | (Still need to check for correctness; implement better canonicalizeDate.) | ||
3370 : | 11. Added "scan" to signature IEEE_REAL. | ||
3371 : | 12. Some improvements to IntInf [in particular: efficiency-hack for | ||
3372 : | mod and rem when second operand is 2 (for parity checks).] | ||
3373 : | 13. Changed representation of type Time.time, using a single IntInf.int | ||
3374 : | value counting microseconds. This considerably simplified the | ||
3375 : | implementation of structure Time. We now support negative time | ||
3376 : | values; scan and fromString handle signs. | ||
3377 : | 14. Functor PrimIO now takes two additional arguments (VectorSlice and | ||
3378 : | ArraySlice). | ||
3379 : | |||
3380 : | ---------------------------------------------------------------------- | ||
3381 : | mblume | 1368 | Name: Matthias Blume (blume (at) tti - c (dot) org) |
3382 : | mblume | 1347 | Date: 2003/08/28 17:00:00 CDT |
3383 : | Tag: blume-20030828-intinf | ||
3384 : | Description: | ||
3385 : | |||
3386 : | This is a major update which comes with a version number bump | ||
3387 : | (110.42.99 -- yes, we are really close to 110.43 :-), NEW BOOTFILES, | ||
3388 : | and an implementation of IntInf in the Basis. | ||
3389 : | |||
3390 : | There are a fairly large number of related changes and updates throughout | ||
3391 : | the system: | ||
3392 : | |||
3393 : | Basis: | ||
3394 : | - Implemented IntInf. | ||
3395 : | - Made LargeInt a projection of IntInf (by filtering through INTEGER). | ||
3396 : | mblume | 1348 | - Added some missing Real64 operations, most notably Real.toLargeInt. |
3397 : | mblume | 1347 | - Added FixedInt as a synonym for Int32. |
3398 : | |||
3399 : | compiler: | ||
3400 : | * Added support for a built-in intinf type. | ||
3401 : | - literals | ||
3402 : | - pattern matching | ||
3403 : | - conversion shortcuts (Int32.fromLarge o Int.toLarge etc.) | ||
3404 : | - overloading on literals and operations | ||
3405 : | |||
3406 : | This required adding a primitive type intinf, some additional | ||
3407 : | primops, and implementations for several non-trivial intinf | ||
3408 : | operations in Core. (The intinf type is completely abstract | ||
3409 : | to the compiler; all operations get delegated back to the Core.) | ||
3410 : | |||
3411 : | * Intinf equality is handled by polyequal. However, the compiler | ||
3412 : | does not print its usual warning in this case (since polyequal | ||
3413 : | is the right thing to do there). | ||
3414 : | |||
3415 : | * Improved the organization of structure InlineT. | ||
3416 : | |||
3417 : | * A word about conversion primops: | ||
3418 : | If conversions involving intinf do not cancel out during | ||
3419 : | mblume | 1348 | CPS contract, then the compiler must insert calls to Core functions. |
3420 : | mblume | 1347 | Since all core access must be resolved already during the FLINT |
3421 : | translate phase, it would be too late a the time of CPS contract | ||
3422 : | mblume | 1348 | to add new Core calls. For this reason, conversion primops |
3423 : | mblume | 1347 | for intinf carry two arguments: 1. the numeric argument that |
3424 : | they are supposed to convert, and 2. the Core function that | ||
3425 : | can help with this conversion if necessary. If CPS contract | ||
3426 : | mblume | 1348 | eliminates a primop, then the associated Core function becomes |
3427 : | dead and goes away. Intinf conversion primops that do not get | ||
3428 : | eliminated by CPS contract get rewritten into calls of their | ||
3429 : | core functions by a separate, new phase. | ||
3430 : | mblume | 1347 | |
3431 : | interactive system: | ||
3432 : | - Control.Print.intinfDepth controls max length of intinf constants | ||
3433 : | being printed. (Analogous to Control.Print.stringDepth.) | ||
3434 : | - Cleanup in printutil and pputil: got rid of unused stuff and | ||
3435 : | duplicates; replaced some of the code with code that makes better | ||
3436 : | use of library functionality. | ||
3437 : | |||
3438 : | CM: | ||
3439 : | Bugfix: parse-errors in init group (system/smlnj/init/init.cmi) | ||
3440 : | are no longer silent. | ||
3441 : | |||
3442 : | CKIT: | ||
3443 : | Fixed mismatched uses of Int32 and LargeInt. I always decided | ||
3444 : | in favor of LargeInt -- which is now the same as IntInf. | ||
3445 : | CKIT-knowledgable people should check whether this is what's | ||
3446 : | intended and otherwise change things back to using Int32 or | ||
3447 : | FixedInt. | ||
3448 : | |||
3449 : | Throughout the code: | ||
3450 : | mblume | 1348 | Started using IntInf.int literals and built-in operations |
3451 : | mblume | 1347 | (e.g., comparison with 0) where this seems appropriate. |
3452 : | |||
3453 : | |||
3454 : | ---------------------------------------------------------------------- | ||
3455 : | macqueen | 1343 | Name: Dave MacQueen (dbm@cs.uchicago.edu) |
3456 : | Date: 2003/08/13 11:36:00 CDT | ||
3457 : | Tag: dbm-20030813-mcz-merge1 | ||
3458 : | Description: | ||
3459 : | |||
3460 : | Merging changes from the mcz-branch development branch into trunk. | ||
3461 : | These changes involve replacement of the emulated old prettyprinter | ||
3462 : | interface with direct use of the SML/NJ Lib PP library, and fixing | ||
3463 : | of a couple of bugs (895, 1186) relating to error messages. A new | ||
3464 : | prettyprinter for ast datatypes (Elaborator/print/ppast.{sig,sml}) | ||
3465 : | has been added. | ||
3466 : | |||
3467 : | ---------------------------------------------------------------------- | ||
3468 : | mblume | 1368 | Name: Matthias Blume (blume (at) tti - c (dot) org) |
3469 : | mblume | 1342 | Date: 2003/08/11 15:45:00 CDT |
3470 : | Tag: blume-20030811-windows | ||
3471 : | Description: | ||
3472 : | |||
3473 : | Version number bumped to 110.42.9. NEW BOOTFILES!!! | ||
3474 : | |||
3475 : | http://smlnj.cs.uchicago.edu/dist/working/110.42.9/ | ||
3476 : | |||
3477 : | This patch restores SML/NJ's ability to run under win32. There are a | ||
3478 : | number of changes, including fixes for several bugs that had gone | ||
3479 : | unnoticed until now: | ||
3480 : | |||
3481 : | - uname "CYGWIN_NT*" is recognized as win32 (This is relevant only when | ||
3482 : | trying to run the win32 version from within cygwin.) | ||
3483 : | |||
3484 : | - There are a number of simple .bat scripts that substitute for their | ||
3485 : | corresponding Unix shell-scripts. (See below.) | ||
3486 : | |||
3487 : | - The internals of ml-build have been modified slightly. The main | ||
3488 : | difference is that instead of calling ".link-sml" (or link-sml.bat) | ||
3489 : | using OS.Process.system, the ML process delegates this task back | ||
3490 : | to the script. Otherwise problems arise in mixed environments such | ||
3491 : | as Cygwin where scripts look and work like Unix scripts, but | ||
3492 : | where OS.Process.system cannot run them. | ||
3493 : | |||
3494 : | - In CM, the srcpath pickler used native pathname syntax -- which | ||
3495 : | is incorrect in the case of cross-compilation. The new pickle format | ||
3496 : | is independent of platform-specific naming conventions. | ||
3497 : | |||
3498 : | - Path configuration files (such as lib/pathconfig) can now choose | ||
3499 : | between native and standard syntax. Placing a line of the form | ||
3500 : | |||
3501 : | standard! | ||
3502 : | |||
3503 : | into the file causes all subsequent paths to be interpreted using | ||
3504 : | CM standard pathname syntax (= Unix conventions); a line | ||
3505 : | |||
3506 : | native! | ||
3507 : | |||
3508 : | switches back to native style. This was needed so that | ||
3509 : | path config files can be written portably, see src/system/pathconfig. | ||
3510 : | |||
3511 : | - Runtime system: | ||
3512 : | |||
3513 : | - win32-filesys.c: get_file_time and set_file_time now | ||
3514 : | access modification time, not creation time. | ||
3515 : | |||
3516 : | - I/O code made aware of new array representation. | ||
3517 : | |||
3518 : | - Bug fixes in X86.prim.masm. | ||
3519 : | |||
3520 : | - src/system/makeml made aware of win32. (For use under cygwin | ||
3521 : | and other Unix-environments for windows.) | ||
3522 : | |||
3523 : | - In Basis, fixed off-by-one error in win32-io.sml (function vecF) | ||
3524 : | which caused BinIO.inputAll to fail consistently. | ||
3525 : | |||
3526 : | .bat scripts: | ||
3527 : | |||
3528 : | Windows .bat scripts assume that SMLNJ_HOME is defined. | ||
3529 : | |||
3530 : | - sml.bat, ml-yacc.bat, ml-lex.bat: Driver scripts for standalone | ||
3531 : | applications (sml, ml-yacc, ml-lex). | ||
3532 : | - ml-build.bat: analogous to ml-build. | ||
3533 : | - config\install.bat: Analogous to config/install.sh. This requires | ||
3534 : | that SMLNJ_HOME is set and that Microsoft Visual C is ready to use. | ||
3535 : | (nmake etc. must be on the path, and vcvars32 must have been run.) | ||
3536 : | Moreover, sources for ml-lex and ml-yacc need to exist under src, | ||
3537 : | and the bootfile hierarchy must have been unpacked under | ||
3538 : | sml.boot.x86-win32. | ||
3539 : | The script is very primitive and does a poor job at error checking. | ||
3540 : | It only installs the base system, ml-lex, and ml-yacc. No other | ||
3541 : | libraries are being installed (i.e., you get only those that | ||
3542 : | are part of the compiler.) | ||
3543 : | - link-sml.bat: analogous to .link-sml, but not currently used | ||
3544 : | |||
3545 : | Unrelated bug fixes: | ||
3546 : | |||
3547 : | - ml-nlffigen now exports structures ST_* corresponding to incomplete | ||
3548 : | types. | ||
3549 : | - Added getDevice to PP/src/pp-debug-fn.sml. (Would not compile | ||
3550 : | otherwise.) | ||
3551 : | |||
3552 : | ---------------------------------------------------------------------- | ||
3553 : | macqueen | 1337 | Name: Dave MacQueen (macqueen@cs.uchicago.edu) |
3554 : | Date: 2003/06/17 | ||
3555 : | Tag: macqueen-20030617-bug895 | ||
3556 : | Description: | ||
3557 : | |||
3558 : | Modified compiler/Elaborator/print/pptype.sml to fix bug 895. | ||
3559 : | Tag will be used for new development branch (mcz-branch) for | ||
3560 : | use by MacQueen, (Lucasz) Zairek, and (George) Cao at uchicago. | ||
3561 : | |||
3562 : | ---------------------------------------------------------------------- | ||
3563 : | mblume | 1368 | Name: Matthias Blume (blume (at) tti - c (dot) org) |
3564 : | mblume | 1335 | Date: 2003/05/27 16:55:00 CDT |
3565 : | Tag: blume-20030527-polyeq | ||
3566 : | Description: | ||
3567 : | |||
3568 : | Tried to eliminated most cases of polymorphic equality. | ||
3569 : | |||
3570 : | ---------------------------------------------------------------------- | ||
3571 : | mblume | 1368 | Name: Matthias Blume (blume (at) tti - c (dot) org) |
3572 : | mblume | 1334 | Date: 2003/05/21 17:45:00 CDT |
3573 : | Tag: blume-20030517-complete | ||
3574 : | Description: | ||
3575 : | |||
3576 : | Two changes: | ||
3577 : | |||
3578 : | 1. Added a flag for controlling whether non-exhaustive bindings will | ||
3579 : | be treated as errors (default is false). | ||
3580 : | 2. Cleaned up the *entire* source tree so that CMB.make goes through | ||
3581 : | without a single non-exhaustive match- or bind warning. | ||
3582 : | |||
3583 : | ---------------------------------------------------------------------- | ||
3584 : | mblume | 1368 | Name: Matthias Blume (blume (at) tti - c (dot) org) |
3585 : | mblume | 1332 | Date: 2003/05/17 10:20:00 CDT |
3586 : | Tag: blume-20030517-absyn | ||
3587 : | Description: | ||
3588 : | |||
3589 : | 1. Added cases for IF, WHILE, ANDALSO, and ORELSE to Absyn. | ||
3590 : | |||
3591 : | This mainly affects the quality of error messages. However, some | ||
3592 : | of the code is now more straightforward than before. (Treatment of | ||
3593 : | the above four constructs in translate.sml is much simpler than | ||
3594 : | the "macro-expansion" that was going on before. Plus, the mach- | ||
3595 : | compiler no longer gets invoked just to be able to compile an | ||
3596 : | if-expression.) | ||
3597 : | |||
3598 : | 2. The ErrorMsg.Error exception is now caught and absorbed by the | ||
3599 : | interactive loop. | ||
3600 : | |||
3601 : | ---------------------------------------------------------------------- | ||
3602 : | allenleung | 1327 | Name: Allen Leung |
3603 : | Date: 2003/05/16 13:05:00 CDT | ||
3604 : | Tag: leunga-20030516-cygwin-runtime | ||
3605 : | Description: | ||
3606 : | |||
3607 : | blume | 2155 | Ported the runtime system to cygwin, which uses the unix |
3608 : | allenleung | 1327 | x86-unix bin files. Missing/buggy features: |
3609 : | |||
3610 : | blume | 2155 | o getnetbyname, getnetbyaddr: these functions seem to be missing in |
3611 : | allenleung | 1327 | the Cygwin library. |
3612 : | o Ctrl-C handling may be flaky. | ||
3613 : | o Windows system calls and Windows I/O are not supported. | ||
3614 : | |||
3615 : | A new set of binfiles is located at: | ||
3616 : | |||
3617 : | http://www.dorsai.org/~leunga/boot.x86-unix.tgz | ||
3618 : | |||
3619 : | This is only needed for bootstrapping the cygwin version of smlnj. | ||
3620 : | Other x86 versions can use the existing binfiles. | ||
3621 : | |||
3622 : | ---------------------------------------------------------------------- | ||
3623 : | mblume | 1317 | Name: Matthias Blume |
3624 : | Date: 2003/04/08 15:42:00 CDT | ||
3625 : | Tag: blume-20030408-listpair | ||
3626 : | Description: | ||
3627 : | |||
3628 : | 1. Added a target 'mlrisc' to installer. | ||
3629 : | |||
3630 : | 2. Added missing elements to structure ListPair. | ||
3631 : | |||
3632 : | ---------------------------------------------------------------------- | ||
3633 : | leunga | 1296 | Name: Allen Leung |
3634 : | Date: 2003/01/07 10:40:00 EST | ||
3635 : | Tag: leunga-20030107-int-rem | ||
3636 : | Description: | ||
3637 : | |||
3638 : | Fixed a bug in Int.rem(x,y) where y is a power of 2 on x86. | ||
3639 : | The arguments to the SUBL instruction were swapped. | ||
3640 : | |||
3641 : | ---------------------------------------------------------------------- | ||
3642 : | blume | 1229 | Name: Matthias Blume |
3643 : | blume | 1294 | Date: 2002/12/12 16:25:00 EST |
3644 : | Tag: blume-20021212-risc-ra | ||
3645 : | Description: | ||
3646 : | |||
3647 : | Fixed a serious bug in the rewrite code for FP spilling/reloading that | ||
3648 : | sent the RA into an infinite loop when floating point registers get | ||
3649 : | spilled. (Because of this bug, e.g., nucleic stopped compiling between | ||
3650 : | 110.37 and 110.38.) | ||
3651 : | There was another set of potential problems related to the handling of | ||
3652 : | MLRISC annotations (but those did not yet cause real problems, apparently). | ||
3653 : | |||
3654 : | ---------------------------------------------------------------------- | ||
3655 : | Name: Matthias Blume | ||
3656 : | blume | 1291 | Date: 2002/12/06 22:40:00 EST |
3657 : | Tag: blume-20021206-cm-fileid | ||
3658 : | Description: | ||
3659 : | |||
3660 : | Added a call of SrcPath.sync at the beginning of Parse.parse (in CM). | ||
3661 : | This fixes the problem of CM getting confused by files that suddenly | ||
3662 : | change their identity (e.g., by getting unlinked and recreated by some | ||
3663 : | text editor such as vi). There might be a better/cheaper/cleaner way | ||
3664 : | of doing this, but for now this will have to do. | ||
3665 : | |||
3666 : | ---------------------------------------------------------------------- | ||
3667 : | Name: Matthias Blume | ||
3668 : | blume | 1288 | Date: 2002/10/28 09:50:00 EST |
3669 : | blume | 2155 | Tag: blume-20021028-typecheck |
3670 : | blume | 1288 | Description: |
3671 : | |||
3672 : | Exported structure Typecheck from $smlnj/viscomp/core.cm. | ||
3673 : | |||
3674 : | ---------------------------------------------------------------------- | ||
3675 : | Name: Matthias Blume | ||
3676 : | blume | 1282 | Date: 2002/10/17 09:10:00 EDT |
3677 : | blume | 1279 | Tag: Release_110_42 |
3678 : | Description: | ||
3679 : | |||
3680 : | blume | 1282 | In good old tradition, there has been a slight hiccup so that we have |
3681 : | to patch 110.42 after the fact. The old release tag has been replaced | ||
3682 : | (see below). | ||
3683 : | |||
3684 : | The change solves a problem with two competing approaches the | ||
3685 : | configuration problem regarding MacOS 10.1 vs. MacOS 10.2 which got in | ||
3686 : | each other's way. | ||
3687 : | |||
3688 : | This change only affects the runtime system code and the installer script. | ||
3689 : | (No new bootfiles.) | ||
3690 : | |||
3691 : | ---------------------------------------------------------------------- | ||
3692 : | Name: Matthias Blume | ||
3693 : | Date: 2002/10/16 12:00:00 EDT | ||
3694 : | Tag: Release_110_42_removed | ||
3695 : | Description: | ||
3696 : | |||
3697 : | blume | 1279 | New working release. New bootfiles. |
3698 : | |||
3699 : | ---------------------------------------------------------------------- | ||
3700 : | Name: Matthias Blume | ||
3701 : | blume | 1273 | Date: 2002/10/10 13:10:00 EDT |
3702 : | Tag: blume-20021010-ppc-divs | ||
3703 : | Description: | ||
3704 : | |||
3705 : | The mltree operator DIVS must be implemented with an overflow check on | ||
3706 : | the PPC because the hardware indicates divide-by-zero using "overflow" as | ||
3707 : | well. | ||
3708 : | |||
3709 : | ---------------------------------------------------------------------- | ||
3710 : | Name: Matthias Blume | ||
3711 : | blume | 1261 | Date: 2002/07/23 11:45:00 |
3712 : | Tag: blume-20020723-smlnj-home | ||
3713 : | Description: | ||
3714 : | |||
3715 : | Sml now senses the SMLNJ_HOME environment variable. If this is set, | ||
3716 : | then the bin dir is assumed to be in $SMLNJ_HOME/bin and (unless | ||
3717 : | CM_PATHCONFIG is also set), the path configuration file is assumed | ||
3718 : | to be in $SMLNJ_HOME/lib/pathconfig. This way one can easily move | ||
3719 : | the entire tree to some other place and everything will "just work". | ||
3720 : | |||
3721 : | (Companion commands such as ml-build and ml-makedepend also sense this | ||
3722 : | variable.) | ||
3723 : | |||
3724 : | ---------------------------------------------------------------------- | ||
3725 : | Name: Matthias Blume | ||
3726 : | blume | 1259 | Date: 2002/07/12 21:19:00 EDT |
3727 : | Tag: blume-20020712-liveness | ||
3728 : | Description: | ||
3729 : | |||
3730 : | Exported two useful "step" functions from liveness module (MLRISC). | ||
3731 : | |||
3732 : | ---------------------------------------------------------------------- | ||
3733 : | Name: Matthias Blume | ||
3734 : | blume | 1253 | Date: 2002/07/05 16:00 EDT |
3735 : | Tag: Release_110_41 | ||
3736 : | Description: | ||
3737 : | |||
3738 : | New working release. New bootfiles. | ||
3739 : | |||
3740 : | ---------------------------------------------------------------------- | ||
3741 : | Name: Matthias Blume | ||
3742 : | blume | 1250 | Date: 2002/07/05 10:25:00 EDT |
3743 : | Tag: blume-20020705-btimp | ||
3744 : | Description: | ||
3745 : | |||
3746 : | Exported structure BTImp from $smlnj/viscomp/debugprof.cm so that | ||
3747 : | other clients can set up backtracing support. | ||
3748 : | |||
3749 : | ---------------------------------------------------------------------- | ||
3750 : | Name: Matthias Blume | ||
3751 : | blume | 1249 | Date: 2002/06/25 17:23:00 EDT |
3752 : | Tag: blume-20020625-fpmax | ||
3753 : | Description: | ||
3754 : | |||
3755 : | Fixed a bug in translation of INLMAX (and INLMIN) for the floating-point | ||
3756 : | case. (The sense of the isNaN test was reversed -- which made min and | ||
3757 : | max always return their first argument.) | ||
3758 : | |||
3759 : | ---------------------------------------------------------------------- | ||
3760 : | Name: Matthias Blume | ||
3761 : | blume | 1240 | Date: 2002/06/11 |
3762 : | Tag: blume-20020611-unixpath | ||
3763 : | Description: | ||
3764 : | |||
3765 : | Back-ported OS.Path.{from,to}UnixPath from idlbasis-devel branch. | ||
3766 : | |||
3767 : | ---------------------------------------------------------------------- | ||
3768 : | Name: Matthias Blume | ||
3769 : | blume | 1238 | Date: 2002/06/10 16:35:00 EDT |
3770 : | Tag: blume-20020610-ieeereal | ||
3771 : | Description: | ||
3772 : | |||
3773 : | I back-ported my implementation of IEEEReal.fromString from the | ||
3774 : | idlbasis-devel branch so that we can test it. | ||
3775 : | |||
3776 : | Another small change is that ppDec tries to give more information | ||
3777 : | than just "<sig>" in the case of functors. However, this code is | ||
3778 : | broken in some mysterious way if the functor's body's signature | ||
3779 : | has not been declared by ascription but gets inferred from the | ||
3780 : | implementation. This needs fixing... | ||
3781 : | |||
3782 : | ---------------------------------------------------------------------- | ||
3783 : | Name: Matthias Blume | ||
3784 : | blume | 1229 | Date: 2002/05/31 |
3785 : | Tag: blume-20020531-btrace-mode | ||
3786 : | Description: | ||
3787 : | |||
3788 : | Resurrected SMLofNJ.Internals.BTrace.mode. (It accidentally fell by | ||
3789 : | the wayside when I switched over to using Controls everywhere.) | ||
3790 : | |||
3791 : | ---------------------------------------------------------------------- | ||
3792 : | george | 1222 | Name: Lal George |
3793 : | Date: 2002/05/23 12:21:40 EDT | ||
3794 : | Tag: george-20020523-visual-labels | ||
3795 : | Description: | ||
3796 : | |||
3797 : | Labels are now displayed in the graphical output to make | ||
3798 : | the fall-through and target blocks obvious. | ||
3799 : | |||
3800 : | ---------------------------------------------------------------------- | ||
3801 : | blume | 1204 | Name: Matthias Blume |
3802 : | blume | 1220 | Date: 2002/05/22 11:03:00 EDT |
3803 : | Tag: blume-20020522-shrink | ||
3804 : | Description: | ||
3805 : | |||
3806 : | John tweaked yesterday's fix for 1131 to handle an out-of-memory | ||
3807 : | situation that comes up when allocating huge arrays. | ||
3808 : | |||
3809 : | ---------------------------------------------------------------------- | ||
3810 : | Name: Matthias Blume | ||
3811 : | blume | 1215 | Date: 2002/05/21 16:00:00 EDT |
3812 : | Tag: Release_110_40 | ||
3813 : | Description: | ||
3814 : | |||
3815 : | New working release (110.40). New bootfiles. | ||
3816 : | |||
3817 : | [Also: John Reppy fixed GC bug 1131.] | ||
3818 : | |||
3819 : | ---------------------------------------------------------------------- | ||
3820 : | Name: Matthias Blume | ||
3821 : | blume | 1212 | Date: 2002/05/21 12:35:00 EDT |
3822 : | Tag: blume-20020521-cmdoc | ||
3823 : | Description: | ||
3824 : | |||
3825 : | CM documentation update. | ||
3826 : | |||
3827 : | ---------------------------------------------------------------------- | ||
3828 : | Name: Matthias Blume | ||
3829 : | blume | 1211 | Date: 2002/05/21 10:55:00 EDT |
3830 : | Tag: blume-20020521-misc | ||
3831 : | Description: | ||
3832 : | |||
3833 : | - John tweaked runtime to be silent on heap export (except when | ||
3834 : | GC messages are on). | ||
3835 : | - I added a few more things (cross-compiling versions of CMB) to | ||
3836 : | config/preloads (as suggestions). | ||
3837 : | |||
3838 : | ---------------------------------------------------------------------- | ||
3839 : | Name: Matthias Blume | ||
3840 : | blume | 1208 | Date: 2002/05/20 22:25:00 EDT |
3841 : | Tag: blume-20020520-controls | ||
3842 : | Description: | ||
3843 : | |||
3844 : | - Added ControlUtil structure to control-lib.cm. | ||
3845 : | - Use it throughout. | ||
3846 : | - Used Controls facility to define MLRISC controls (as opposed to | ||
3847 : | registering MLRISC control ref cells with Controls after the fact) | ||
3848 : | - Fixed messed-up controls priorities. | ||
3849 : | |||
3850 : | * Removed again all the stuff from config/preloads that one wouldn't | ||
3851 : | be able to preload at the time the initial heap image is built. | ||
3852 : | (Many libraries, e.g., CML, do not exist yet at this time. The | ||
3853 : | only libraries that can be preloaded via config/preloads are those | ||
3854 : | that come bundled with the bootfiles.) | ||
3855 : | |||
3856 : | ---------------------------------------------------------------------- | ||
3857 : | Name: Matthias Blume | ||
3858 : | blume | 1204 | Date: 2002/05/20 10:59:00 EDT |
3859 : | Tag: blume-20020520-preloads | ||
3860 : | Description: | ||
3861 : | |||
3862 : | Added a lot of commented-out suggestions for things to be included | ||
3863 : | in config/preloads. | ||
3864 : | |||
3865 : | ---------------------------------------------------------------------- | ||
3866 : | leunga | 1203 | Name: Allen Leung |
3867 : | Date: 2002/05/18 14:20:00 EDT | ||
3868 : | Tag: leunga-20020518-mdl | ||
3869 : | Description: | ||
3870 : | |||
3871 : | o Made the mdl tool stuff compile and run again. | ||
3872 : | o I've disabled all the stuff that depends on RTL specifications; they | ||
3873 : | are all badly broken anyway. | ||
3874 : | |||
3875 : | ---------------------------------------------------------------------- | ||
3876 : | blume | 1201 | Name: Matthias Blume |
3877 : | Date: 2002/05/17 16:49:00 EDT | ||
3878 : | Tag: blume-20020517-controls | ||
3879 : | Description: | ||
3880 : | |||
3881 : | 0. John Reppy made several modifications to the SML/NJ library. | ||
3882 : | In particular, there is a shiny new controls-lib.cm. | ||
3883 : | |||
3884 : | 1. Pushed new controls interface through compiler so that everything | ||
3885 : | compiles again. | ||
3886 : | |||
3887 : | 2. Added FormatComb and FORMAT_COMB to the CML version of the | ||
3888 : | SML/NJ library (so that CML compiles again). | ||
3889 : | |||
3890 : | 3. Modified init scripts because XXX_DEFAULT environment variables | ||
3891 : | are no longer with us. (Boot-time initialization is now done | ||
3892 : | using the same environment variables that are also used for | ||
3893 : | startup-time initialization of controls.) | ||
3894 : | |||
3895 : | ---------------------------------------------------------------------- | ||
3896 : | george | 1192 | Name: Lal George |
3897 : | Date: 2002/05/15 09:20:10 EDT | ||
3898 : | Tag: george-20020515-pseudo-op-decls | ||
3899 : | Description: | ||
3900 : | |||
3901 : | All pseudo-ops emitted before the first segment declaration | ||
3902 : | such as TEXT, DATA, and BSS directives are assumed to be global | ||
3903 : | declarations and are emitted first in the assembly file. This is | ||
3904 : | useful in a number of situations where one has pseudo-ops that are not | ||
3905 : | specific to any segment, and also works around the constraint that one | ||
3906 : | cannot have client pseudo-ops in the TEXT segment. | ||
3907 : | |||
3908 : | Because no segment is associated with these declarations it is | ||
3909 : | an error to allocate any space or objects before the first segment | ||
3910 : | directive and an exception will be raised. However, we cannot make | ||
3911 : | this check for client pseudo-ops. | ||
3912 : | |||
3913 : | These top level declarations are a field in the CFG graph_info. | ||
3914 : | In theory you can continue to add to this field after the CFG has been | ||
3915 : | built -- provided you know what you are doing;-) | ||
3916 : | |||
3917 : | ---------------------------------------------------------------------- | ||
3918 : | blume | 1175 | Name: Matthias Blume |
3919 : | blume | 1190 | Date: 2002/05/13 16:40:00 EDT |
3920 : | Tag: blume-20020513-pp-etc | ||
3921 : | Description: | ||
3922 : | |||
3923 : | A few minor bugfixes: | ||
3924 : | |||
3925 : | - Stopgap measure for bug recently reported by Elsa Gunter (ppDec). | ||
3926 : | (Bogus printouts for redefined bindings still occur. Compiler | ||
3927 : | bug should no longer occur now. We need to redo the prettyprinter | ||
3928 : | from scratch.) | ||
3929 : | |||
3930 : | - CM pathname printer now also adds escape sequences for ( and ) | ||
3931 : | |||
3932 : | - commend and docu fixes for ml-nlffi | ||
3933 : | |||
3934 : | ---------------------------------------------------------------------- | ||
3935 : | Name: Matthias Blume | ||
3936 : | blume | 1189 | Date: 2002/05/10 16:40:00 EDT |
3937 : | Tag: blume-20020510-erg-textio | ||
3938 : | Description: | ||
3939 : | |||
3940 : | Applied the following bugfix provided by Emden Gansner: | ||
3941 : | |||
3942 : | Output is corrupted when outputSubstr is used rather than output. | ||
3943 : | |||
3944 : | blume | 2155 | The problem occurs when a substring |
3945 : | blume | 1189 | |
3946 : | ss = (s, dataStart, dataLen) | ||
3947 : | |||
3948 : | blume | 2155 | where dataStart > 0, fills a stream buffer with avail bytes left. |
3949 : | avail bytes of s, starting at index dataStart, are copied into the | ||
3950 : | blume | 1189 | buffer, the buffer is flushed, and then the remaining dataLen-avail |
3951 : | blume | 2155 | bytes of ss are copied into the beginning of the buffer. Instead of |
3952 : | starting this copy at index dataStart+avail in s, the current code | ||
3953 : | blume | 1189 | starts the copy at index avail. |
3954 : | |||
3955 : | Fix: | ||
3956 : | In text-io-fn.sml, change line 695 from | ||
3957 : | val needsFlush = copyVec(v, avail, dataLen-avail, buf, 0) | ||
3958 : | blume | 2155 | to |
3959 : | blume | 1189 | val needsFlush = copyVec(v, dataStart+avail, dataLen-avail, buf, 0) |
3960 : | |||
3961 : | ---------------------------------------------------------------------- | ||
3962 : | Name: Matthias Blume | ||
3963 : | blume | 1186 | Date: 2002/04/12 13:55:00 EDT |
3964 : | Tag: blume-20020412-assyntax | ||
3965 : | Description: | ||
3966 : | |||
3967 : | 1. Grabbed newer assyntax.h from the XFree86 project. | ||
3968 : | 2. Fiddled with how to compile X86.prim.asm without warnings. | ||
3969 : | 3. (Very) Minor cleanup in CM. | ||
3970 : | |||
3971 : | ---------------------------------------------------------------------- | ||
3972 : | Name: Matthias Blume | ||
3973 : | blume | 1185 | Date: 2002/04/01 (no joke!) 17:07:00 EST |
3974 : | Tag: blume-20020401-x86div | ||
3975 : | Description: | ||
3976 : | |||
3977 : | Added full support for div/mod/rem/quot on the x86, using the machine | ||
3978 : | instruction's two results (without clumsily recomputing the remainder) | ||
3979 : | directly where appropriate. | ||
3980 : | |||
3981 : | Some more extensive power-of-two support was added to the x86 instruction | ||
3982 : | selector (avoiding expensive divs, mods, and muls where they can be | ||
3983 : | replaced with cheaper shifts and masks). However, this sort of thing | ||
3984 : | ought to be done earlier, e.g., within the CPS optimizer so that | ||
3985 : | all architectures benefit from it. | ||
3986 : | |||
3987 : | The compiler compiles to a fixed point, but changes might be somewhat | ||
3988 : | fragile nevertheless. Please, report any strange things that you might | ||
3989 : | see wrt. div/mod/quot/rem... | ||
3990 : | |||
3991 : | ---------------------------------------------------------------------- | ||
3992 : | Name: Matthias Blume | ||
3993 : | blume | 1184 | Date: 2002/03/29 17:22:00 |
3994 : | Tag: blume-20020329-div | ||
3995 : | Description: | ||
3996 : | |||
3997 : | Fixed my broken div/mod logic. Unfortunately, this means that the | ||
3998 : | inline code for div/mod now has one more comparison than before. | ||
3999 : | Fast paths (quotient > 0 or remainder = 0) are not affected, though. | ||
4000 : | The problem was with quotient = 0, because that alone does not tell | ||
4001 : | us which way the rounding went. One then has to look at whether | ||
4002 : | remainder and divisor have the same sign... :( | ||
4003 : | |||
4004 : | Anyway, I replaced the bootfiles with fresh ones... | ||
4005 : | |||
4006 : | ---------------------------------------------------------------------- | ||
4007 : | Name: Matthias Blume | ||
4008 : | blume | 1183 | Date: 2002/03/29 14:10:00 EST |
4009 : | Tag: blume-20020329-inlprims | ||
4010 : | Description: | ||
4011 : | |||
4012 : | NEW BOOTFILES!!! Version number bumped to 110.39.3. | ||
4013 : | |||
4014 : | Primops have changed. This means that the bin/boot-file formats have | ||
4015 : | changed as well. | ||
4016 : | |||
4017 : | To make sure that there is no confusion, I made a new version. | ||
4018 : | |||
4019 : | |||
4020 : | CHANGES: | ||
4021 : | |||
4022 : | * removed REMT from mltree (remainder should never overflow). | ||
4023 : | |||
4024 : | * added primops to deal with divisions of all flavors to the frontend | ||
4025 : | |||
4026 : | * handled these primops all the way through so they map to their respective | ||
4027 : | MLRISC support | ||
4028 : | |||
4029 : | * used these primops in the implementation of Int, Int32, Word, Word32 | ||
4030 : | |||
4031 : | * removed INLDIV, INLMOD, and INLREM as they are no longer necessary | ||
4032 : | |||
4033 : | * parameterized INLMIN, INLMAX, and INLABS by a numkind | ||
4034 : | |||
4035 : | * translate.sml now deals with all flavors of INL{MIN,MAX,ABS}, including | ||
4036 : | floating point | ||
4037 : | |||
4038 : | * used INL{MIN,MAX,ABS} in the implementation of Int, Int32, Word, Word32, | ||
4039 : | and Real (but Real.abs maps to a separate floating-point-only primop) | ||
4040 : | |||
4041 : | |||
4042 : | TODO items: | ||
4043 : | |||
4044 : | * Hacked Alpha32 instruction selection, disabling the selection of REMx | ||
4045 : | instructions because the machine instruction encoder cannot handle | ||
4046 : | them. (Hppa, PPC, and Sparc instruction selection did not handle | ||
4047 : | REM in the first place, and REM is supported by the x86 machine coder.) | ||
4048 : | |||
4049 : | * Handle DIV and MOD with DIV_TO_NEGINF directly in the x86 instruction | ||
4050 : | selection phase. (The two can be streamlined because the hardware | ||
4051 : | delivers both quotient and remainder at the same time anyway.) | ||
4052 : | |||
4053 : | * Think about what to do with "valOf(Int32.minInt) div ~1" and friends. | ||
4054 : | (Currently the behavior is inconsistent both across architectures and | ||
4055 : | wrt. the draft Basis spec.) | ||
4056 : | |||
4057 : | * Word8 should eventually be handled natively, too. | ||
4058 : | |||
4059 : | * There seems to be one serious bug in mltree-gen.sml. It appears, though, | ||
4060 : | as if there currently is no execution path that could trigger it in | ||
4061 : | SML/NJ. (The assumptions underlying functions arith and promotable do not | ||
4062 : | hold for things like multiplication and division.) | ||
4063 : | |||
4064 : | ---------------------------------------------------------------------- | ||
4065 : | Name: Matthias Blume | ||
4066 : | blume | 1181 | Date: 2002/03/27 16:27:00 EST |
4067 : | Tag: blume-20020327-mlrisc-divisions | ||
4068 : | Description: | ||
4069 : | |||
4070 : | Added support for all four division operations (ML's div, mod, quot, | ||
4071 : | and rem) to MLRISC. In the course of doing so, I also rationalized | ||
4072 : | the naming (no more annoying switch-around of DIV and QUOT), by | ||
4073 : | parameterizing the operation by div_rounding_mode (which can be either | ||
4074 : | DIV_TO_ZERO or DIV_TO_NEGINF). | ||
4075 : | |||
4076 : | The generic MLTreeGen functor takes care of compiling all four | ||
4077 : | operations down to only round-to-zero div. | ||
4078 : | |||
4079 : | Missing pieces: | ||
4080 : | |||
4081 : | * Doing something smarter than relying on MLTreeGen on architectures | ||
4082 : | like, e.g., the x86 where hardware division delivers both quotient and | ||
4083 : | remainder at the same time. With this, the implementation of the | ||
4084 : | round-to-neginf operations could be further streamlined. | ||
4085 : | |||
4086 : | * Remove inlining support for div/mod/rem from the frontend and replace it | ||
4087 : | with primops that get carried through to the backend. Do this for all | ||
4088 : | int and word types. | ||
4089 : | |||
4090 : | ---------------------------------------------------------------------- | ||
4091 : | Name: Matthias Blume | ||
4092 : | blume | 1180 | Date: 2002/03/25 17:25:00 EST |
4093 : | Tag: blume-20020325-divmod | ||
4094 : | Description: | ||
4095 : | |||
4096 : | I improved (hopefully without breaking them) the implementation of Int.div, | ||
4097 : | Int.mod, and Int.rem. For this, the code in translate.sml now takes | ||
4098 : | advantage of the following observations: | ||
4099 : | |||
4100 : | Let q = x quot y r = x rem y | ||
4101 : | d = x div y m = x mod y | ||
4102 : | |||
4103 : | where "quot" is the round-to-zero version of integer division that | ||
4104 : | hardware usually provides. Then we have: | ||
4105 : | |||
4106 : | r = x - q * y where neither the * nor the - will overflow | ||
4107 : | d = if q >= 0 orelse x = q * y then q else q - 1 | ||
4108 : | where neither the * nor the - will overflow | ||
4109 : | m = if q >= 0 orelse r = 0 then r else r + y | ||
4110 : | where the + will not overflow | ||
4111 : | |||
4112 : | This results in substantial simplification of the generated code. | ||
4113 : | The following table shows the number of CFG nodes and edges generated | ||
4114 : | for | ||
4115 : | fun f (x, y) = x OPER y | ||
4116 : | (* with OPER \in div, mod, quot, rem *) | ||
4117 : | |||
4118 : | |||
4119 : | OPER | nodes(old) | edges(old) | nodes(new) | edges(new) | ||
4120 : | -------------------------------------------------------- | ||
4121 : | div | 24 | 39 | 12 | 16 | ||
4122 : | mod | 41 | 71 | 12 | 16 | ||
4123 : | quot | 8 | 10 | 8 | 10 | ||
4124 : | rem | 10 | 14 | 8 | 10 | ||
4125 : | |||
4126 : | |||
4127 : | ---------------------------------------------------------------------- | ||
4128 : | Name: Matthias Blume | ||
4129 : | blume | 1179 | Date: 2002/03/25 22:06:00 EST |
4130 : | Tag: blume-20020325-cprotobug | ||
4131 : | Description: | ||
4132 : | |||
4133 : | Fixed a bug in cproto (c prototype decoder). | ||
4134 : | |||
4135 : | ---------------------------------------------------------------------- | ||
4136 : | Name: Matthias Blume | ||
4137 : | blume | 1178 | Date: 2002/03/25 16:00:00 EST |
4138 : | Tag: blume-20020325-raw-primops | ||
4139 : | Description: | ||
4140 : | |||
4141 : | I did some cleanup to Allen's new primop code and | ||
4142 : | replaced yesterday's bootfiles with new ones. | ||
4143 : | (But they are stored in the same place.) | ||
4144 : | |||
4145 : | ---------------------------------------------------------------------- | ||
4146 : | Name: Matthias Blume | ||
4147 : | blume | 1175 | Date: 2002/03/24 22:40:00 EST |
4148 : | Tag: blume-20020324-bootfiles | ||
4149 : | Description: | ||
4150 : | |||
4151 : | Made the bootfiles that Allen asked for. | ||
4152 : | |||
4153 : | ---------------------------------------------------------------------- | ||
4154 : | leunga | 1174 | Name: Allen Leung |
4155 : | Date: 2002/03/23 15:50:00 EST | ||
4156 : | Tag: leunga-20020323-flint-cps-rcc-primops | ||
4157 : | Description: | ||
4158 : | |||
4159 : | 1. Changes to FLINT primops: | ||
4160 : | |||
4161 : | (* make a call to a C-function; | ||
4162 : | * The primop carries C function prototype information and specifies | ||
4163 : | * which of its (ML-) arguments are floating point. C prototype | ||
4164 : | * information is for use by the backend, ML information is for | ||
4165 : | * use by the CPS converter. *) | ||
4166 : | | RAW_CCALL of { c_proto: CTypes.c_proto, | ||
4167 : | ml_args: ccall_type list, | ||
4168 : | ml_res_opt: ccall_type option, | ||
4169 : | reentrant : bool | ||
4170 : | } option | ||
4171 : | (* Allocate uninitialized storage on the heap. | ||
4172 : | * The record is meant to hold short-lived C objects, i.e., they | ||
4173 : | blume | 2155 | * are not ML pointers. With the tag, the representation is |
4174 : | leunga | 1174 | * the same as RECORD with tag tag_raw32 (sz=4), or tag_fblock (sz=8) |
4175 : | *) | ||
4176 : | blume | 2155 | | RAW_RECORD of {tag:bool,sz:int} |
4177 : | leunga | 1174 | and ccall_type = CCALL_INT32 | CCALL_REAL64 | CCALL_ML_PTR |
4178 : | |||
4179 : | 2. These CPS primops are now overloaded: | ||
4180 : | |||
4181 : | rawload of {kind:numkind} | ||
4182 : | rawstore of {kind:numkind} | ||
4183 : | |||
4184 : | The one argument form is: | ||
4185 : | |||
4186 : | rawload {kind} address | ||
4187 : | |||
4188 : | The two argument form is: | ||
4189 : | blume | 2155 | |
4190 : | leunga | 1174 | rawload {kind} [ml object, byte-offset] |
4191 : | |||
4192 : | 3. RAW_CCALL/RCC now takes two extra arguments: | ||
4193 : | |||
4194 : | a. The first is whether the C call is reentrant, i.e., whether | ||
4195 : | ML state should be saved and restored. | ||
4196 : | b. The second argument is a string argument specifying the name of | ||
4197 : |