14 |
|
|
15 |
---------------------------------------------------------------------- |
---------------------------------------------------------------------- |
16 |
Name: Matthias Blume |
Name: Matthias Blume |
17 |
|
Date: 2002/02/15 17:05:00 |
18 |
|
Tag: Release_110_39 |
19 |
|
Description: |
20 |
|
|
21 |
|
Last-minute retagging is becoming a tradition... :-( |
22 |
|
|
23 |
|
This is the working release 110.39. |
24 |
|
|
25 |
|
---------------------------------------------------------------------- |
26 |
|
Name: Matthias Blume |
27 |
|
Date: 2002/02/15 16:00:00 EST |
28 |
|
Tag: Release_110_39-orig |
29 |
|
Description: |
30 |
|
|
31 |
|
Working release 110.39. New bootfiles. |
32 |
|
|
33 |
|
(Update: There was a small bug in the installer so it wouldn't work |
34 |
|
with all shells. So I retagged. -Matthias) |
35 |
|
|
36 |
|
---------------------------------------------------------------------- |
37 |
|
Name: Matthias Blume |
38 |
|
Date: 2002/02/15 14:17:00 EST |
39 |
|
Tag: blume-20020215-showbindings |
40 |
|
Description: |
41 |
|
|
42 |
|
Added EnvRef.listBoundSymbols and CM.State.showBindings. Especially |
43 |
|
the latter can be useful for exploring what bindings are available at |
44 |
|
the interactive prompt. (The first function returns only the list |
45 |
|
of symbols that are really bound, the second prints those but also the |
46 |
|
ones that CM's autoloading mechanism knows about.) |
47 |
|
|
48 |
|
---------------------------------------------------------------------- |
49 |
|
Name: Matthias Blume |
50 |
|
Date: 2002/02/15 12:08:00 EST |
51 |
|
Tag: blume-20020215-iptrs |
52 |
|
Description: |
53 |
|
|
54 |
|
Two improvements to ml-nlffigen: |
55 |
|
|
56 |
|
1. Write files only if they do not exist or if their current contents |
57 |
|
do not coincide with what's being written. (That is, avoid messing |
58 |
|
with the time stamps unless absolutely necessary.) |
59 |
|
|
60 |
|
2. Implement a "repository" mechanism for generated files related |
61 |
|
to "incomplete pointer types". See the README file for details. |
62 |
|
|
63 |
|
---------------------------------------------------------------------- |
64 |
|
Name: Matthias Blume |
65 |
|
Date: 2002/02/14 11:50:00 EST |
66 |
|
Tag: blume-20020214-quote |
67 |
|
Description: |
68 |
|
|
69 |
|
Added a type 't t_' to tag.sml (in ml-nlffi-lib.cm). This is required |
70 |
|
because of the new and improved tag generation scheme. (Thanks to Allen |
71 |
|
Leung for pointing it out.) |
72 |
|
|
73 |
|
---------------------------------------------------------------------- |
74 |
|
Name: Lal George |
75 |
|
Date: 2002/02/14 09:55:27 EST 2002 |
76 |
|
Tag: george-20020214-isabelle-bug |
77 |
|
Description: |
78 |
|
|
79 |
|
Fixed the MLRISC bug sent by Markus Wenzel regarding the compilation |
80 |
|
of Isabelle on the x86. |
81 |
|
|
82 |
|
From Allen: |
83 |
|
----------- |
84 |
|
I've found the problem: |
85 |
|
|
86 |
|
in ra-core.sml, I use the counter "blocked" to keep track of the |
87 |
|
true number of elements in the freeze queue. When the counter goes |
88 |
|
to zero, I skip examining the queue. But I've messed up the |
89 |
|
bookkeeping in combine(): |
90 |
|
|
91 |
|
else (); |
92 |
|
case !ucol of |
93 |
|
PSEUDO => (if !cntv > 0 then |
94 |
|
(if !cntu > 0 then blocked := !blocked - 1 else (); |
95 |
|
^^^^^^^^^^^^^^^^^^^^^^^ |
96 |
|
moveu := mergeMoveList(!movev, !moveu) |
97 |
|
) |
98 |
|
else (); |
99 |
|
|
100 |
|
combine() is called to coalesce two nodes u and v. |
101 |
|
I think I was thinking that if the move counts of u and v are both |
102 |
|
greater than zero then after they are coalesced then one node is |
103 |
|
removed from the freeze queue. Apparently I was thinking that |
104 |
|
both u and v are of low degree, but that's clearly not necessarily true. |
105 |
|
|
106 |
|
|
107 |
|
02/12/2002: |
108 |
|
Here's the patch. HOL now compiles. |
109 |
|
|
110 |
|
I don't know how this impact on performance (compile |
111 |
|
time or runtime). This bug caused the RA (especially on the x86) |
112 |
|
to go thru the potential spill phase when there are still nodes on the |
113 |
|
freeze queue. |
114 |
|
|
115 |
|
|
116 |
|
|
117 |
|
|
118 |
|
---------------------------------------------------------------------- |
119 |
|
Name: Matthias Blume |
120 |
|
Date: 2002/02/13 22:40:00 EST |
121 |
|
Tag: blume-20020213-fptr-rtti |
122 |
|
Description: |
123 |
|
|
124 |
|
Fixed a bug in ml-nlffigen that was introduced with one of the previous |
125 |
|
updates. |
126 |
|
|
127 |
|
---------------------------------------------------------------------- |
128 |
|
Name: Matthias Blume |
129 |
|
Date: 2002/02/13 16:41:00 EST |
130 |
|
Tag: blume-20020213-cmlpq |
131 |
|
Description: |
132 |
|
|
133 |
|
Added new priority queue export symbols (which have just been added to |
134 |
|
smlnj-lib.cm) to CML's version of smlnj-lib.cm. (Otherwise CML would |
135 |
|
not compile and the installer would choke.) |
136 |
|
|
137 |
|
---------------------------------------------------------------------- |
138 |
|
Name: Matthias Blume |
139 |
|
Date: 2002/02/13 16:15:00 EST |
140 |
|
Tag: blume-20020213-various |
141 |
|
Description: |
142 |
|
|
143 |
|
1. More tweaks to ml-nlffigen: |
144 |
|
|
145 |
|
- better internal datastructures (resulting in slight speedup) |
146 |
|
- "-match" option requires exact match |
147 |
|
- "localized" gensym counters (untagged structs/unions nested within |
148 |
|
other structs/unions or within typedefs get a fresh counter; their |
149 |
|
tag will be prefixed by a concatenation of their parents' tags) |
150 |
|
- bug fixes (related to calculation of transitive closure of types |
151 |
|
to be included in the output) |
152 |
|
|
153 |
|
2. Minor Basis updates: |
154 |
|
|
155 |
|
- added implementations for List.collate and Option.app |
156 |
|
|
157 |
|
---------------------------------------------------------------------- |
158 |
|
Name: Matthias Blume |
159 |
|
Date: 2002/02/11 15:55:00 EST |
160 |
|
Tag: blume-20020211-gensym |
161 |
|
Description: |
162 |
|
|
163 |
|
Added a "-gensym" option to command line of ml-nlffigen. This can be |
164 |
|
used to specify a "stem" -- a string that is inserted in all "gensym'd" |
165 |
|
names (ML structure names that correspond to unnamed C structs, unions, |
166 |
|
and enums), so that separate runs of ml-nlffigen do not clash. |
167 |
|
|
168 |
|
---------------------------------------------------------------------- |
169 |
|
Name: Matthias Blume |
170 |
|
Date: 2002/02/11 12:05:00 EST |
171 |
|
Tag: blume-20020211-gensml |
172 |
|
Description: |
173 |
|
|
174 |
|
A quick fix for a problem with GenSML (in the pgraph-util library): |
175 |
|
Make generation of toplevel "local" optional. (Strictly speaking, |
176 |
|
signature definitions within "local" are not legal SML.) |
177 |
|
|
178 |
|
Other than that: updates to INSTALL and cm/TODO. |
179 |
|
|
180 |
|
---------------------------------------------------------------------- |
181 |
|
Name: Matthias Blume |
182 |
|
Date: 2002/02/08 15:00:00 EST |
183 |
|
Tag: blume-20020208-uniquepid |
184 |
|
Description: |
185 |
|
|
186 |
|
0. Version number has been bumped to 110.38.1. NEW BOOTFILES!!! |
187 |
|
|
188 |
|
1. The installer (config/install.sh) has gotten smarter: |
189 |
|
|
190 |
|
- Configuration options are a bit easier to specify now |
191 |
|
(in config/targets). |
192 |
|
- Bug in recognizing .tar.bz2 files fixed. |
193 |
|
- Installer automatically resolves dependencies between |
194 |
|
configuration options (e.g., if you ask for eXene, you will |
195 |
|
also get cml -- regardless whether you asked for it or not). |
196 |
|
- Installer can run in "quieter mode" by setting the environment |
197 |
|
variable INSTALL_QUIETLY to "true". "Quieter" does not mean |
198 |
|
"completely silent", though. |
199 |
|
- Build HashCons library as part of smlnj-lib. |
200 |
|
|
201 |
|
2. A new scheme for assigning persistent identifiers to compilation |
202 |
|
units (and, by extension, to types etc.) has been put into place. |
203 |
|
This fixes a long-standing bug where types and even dynamic values |
204 |
|
can get internally confused, thereby compromising type safety |
205 |
|
(abstraction) and dynamic correctness. See |
206 |
|
|
207 |
|
http://cm.bell-labs.com/cm/cs/who/blume/pid-confusion.tgz |
208 |
|
|
209 |
|
for an example of how things could go wrong until now. |
210 |
|
|
211 |
|
The downside of the new scheme is that pids are not quite as |
212 |
|
persistent as they used to be: CM will generate a fresh pid |
213 |
|
for every compilation unit that it thinks it sees for the first |
214 |
|
time. That means that if you compile starting from a clean, fresh |
215 |
|
source tree at two different times, you end up with different |
216 |
|
binaries. |
217 |
|
|
218 |
|
Cutoff recompilation, however, has not been compromised because |
219 |
|
CM keeps pid information in special caches between runs. |
220 |
|
|
221 |
|
---------------------------------------------------------------------- |
222 |
|
Name: Lal George |
223 |
|
Date: 2002/02/07 15:34:13 EST 2002 |
224 |
|
Tag: <none> |
225 |
|
Description: |
226 |
|
|
227 |
|
Compilers that generate assembly code may produce global labels |
228 |
|
whose value is resolved at link time. The various peephole optimization |
229 |
|
modules did not take this in account. |
230 |
|
|
231 |
|
TODO. The Labels.addrOf function should really return an option |
232 |
|
type so that clients are forced to deal with this issue, rather |
233 |
|
than an exception being raised. |
234 |
|
|
235 |
|
---------------------------------------------------------------------- |
236 |
|
Name: Lal George |
237 |
|
Date: 2002/02/06 13:55:02 EST |
238 |
|
Tag: george-20020206-ra-breakup |
239 |
|
Description: |
240 |
|
|
241 |
|
1. A bug fix from Allen. |
242 |
|
|
243 |
|
A typo causes extra fstp %st(0)'s to be generated at compensation |
244 |
|
edges, which might cause stack underflow traps at runtime. This |
245 |
|
occurs in fft where there are extraneous fstps right before the 'into' |
246 |
|
trap instruction (in this case they are harmless since none of the |
247 |
|
integers overflow.) |
248 |
|
|
249 |
|
2. Pulled out various utility modules that were embedded in the modules |
250 |
|
of the register allocator. I need these modules for other purposes, but |
251 |
|
they are not complete enough to put into a library (just yet). |
252 |
|
---------------------------------------------------------------------- |
253 |
|
Name: Matthias Blume |
254 |
|
Date: 2002/01/31 16:05:00 EST |
255 |
|
Tag: blume-20020131-sparc-ccalls |
256 |
|
Description: |
257 |
|
|
258 |
|
1. C-calls on Sparc needlessly allocated a huge chunk (96 bytes) |
259 |
|
of extra stack space by mistake. Fixed. |
260 |
|
|
261 |
|
2. Bug in logic of handling of command-line options in ml-nlffigen fixed. |
262 |
|
|
263 |
|
---------------------------------------------------------------------- |
264 |
|
Name: Allen Leung |
265 |
|
Date: 2002/01/30 |
266 |
|
Tag: leunga-20020130-nowhere-bug-fix |
267 |
|
Description: |
268 |
|
|
269 |
|
MLRISC bug fixes: |
270 |
|
1. Fixed a bindings computation bug in the 'nowhere' program generator tool. |
271 |
|
2. MachineInt.fromString was negating its value. |
272 |
|
|
273 |
|
---------------------------------------------------------------------- |
274 |
|
Name: Matthias Blume |
275 |
|
Date: 2002/01/29 |
276 |
|
Tag: blume-20020129-INSTALL |
277 |
|
Description: |
278 |
|
|
279 |
|
- Added somewhat detailed installation instructions (file INSTALL). |
280 |
|
- Fixed curl-detection bug in config/install.sh. |
281 |
|
- It is now possible to select the URL getter using the URLGETTER |
282 |
|
environment variable: |
283 |
|
|
284 |
|
not set / "unknown" --> automatic detection (script tries wget, |
285 |
|
curl, and lynx) |
286 |
|
"wget" / "curl" / "lynx" --> use the specified program (script "knows" |
287 |
|
how to properly invoke them) |
288 |
|
other --> use $URLGETTER directly, it must take |
289 |
|
precisely two command-line arguments |
290 |
|
(source URL and destination file name) |
291 |
|
|
292 |
|
---------------------------------------------------------------------- |
293 |
|
Name: Matthias Blume |
294 |
|
Date: 2002/01/28 |
295 |
|
Tag: blume-20020128-sparc-ccalls |
296 |
|
Description: |
297 |
|
|
298 |
|
- Fixed problem with calculation of "used" registers in sparc-c-calls. |
299 |
|
- Make use of the allocParam argument in sparc-c-calls. |
300 |
|
|
301 |
|
---------------------------------------------------------------------- |
302 |
|
Name: Matthias Blume |
303 |
|
Date: 2002/01/28 |
304 |
|
Tag: blume-20020128-allocParam |
305 |
|
Description: |
306 |
|
|
307 |
|
John Reppy: Changes c-calls API to accept client-callback for |
308 |
|
allocating extra stack space. |
309 |
|
me: Corresponding changes to mlriscGen (using a dummy argument that |
310 |
|
does not change the current behavior). |
311 |
|
|
312 |
|
---------------------------------------------------------------------- |
313 |
|
Name: Matthias Blume |
314 |
|
Date: 2002/01/28 12:00:00 |
315 |
|
Tag: Release_110_38 |
316 |
|
Description: |
317 |
|
|
318 |
|
This time for real!!! |
319 |
|
|
320 |
|
---------------------------------------------------------------------- |
321 |
|
Name: Matthias Blume |
322 |
|
Date: 2002/01/28 10:56:00 EST |
323 |
|
Tag: blume-20020128-retraction |
324 |
|
Description: |
325 |
|
|
326 |
|
0. Retracted earlier 110.38. (The Release_110_38 tag has been replaced |
327 |
|
with blume-Release_110_38-retracted.) |
328 |
|
|
329 |
|
1. Fixed a problem with incorrect rounding modes in real64.sml. |
330 |
|
(Thanks to Andrew Mccreight <andrew.mccreight@yale.edu>.) |
331 |
|
|
332 |
|
2. A bug in ml-nlffigen related to the handling of unnamed structs, unions, |
333 |
|
and enums fixed. The naming of corresponding ML identifiers should |
334 |
|
now be consistent again. |
335 |
|
|
336 |
|
---------------------------------------------------------------------- |
337 |
|
Name: Allen Leung |
338 |
|
Date: 2002/01/27 |
339 |
|
Tag: leunga-20020127-nowhere |
340 |
|
Description: |
341 |
|
|
342 |
|
Added a target called nowhere in the configuration scripts. |
343 |
|
Enabling this will build the MLRISC 'nowhere' tool (for translating |
344 |
|
programs with where-clauses into legal SML code) during installation. |
345 |
|
|
346 |
|
---------------------------------------------------------------------- |
347 |
|
Name: Matthias Blume |
348 |
|
Date: 2002/01/25 21:27:00 EST |
349 |
|
Tag: blume-Release_110_38-retracted |
350 |
|
Description: |
351 |
|
|
352 |
|
Call it a (working) release! Version is 110.38. Bootfiles are ready. |
353 |
|
|
354 |
|
README will be added later. |
355 |
|
|
356 |
|
!!! NOTE: Re-tagged as blume-Release_110_38-retracted. Original tag |
357 |
|
(Release_110_38) removed. Reason: Last-minute bug fixes. |
358 |
|
|
359 |
|
---------------------------------------------------------------------- |
360 |
|
Name: Matthias Blume |
361 |
|
Date: 2002/01/25 |
362 |
|
Tag: blume-20020125-ffi |
363 |
|
Description: |
364 |
|
|
365 |
|
A large number of tweaks and improvements to ml-nlffi-lib and |
366 |
|
ml-nlffigen: |
367 |
|
|
368 |
|
- ML represenation types have been streamlined |
369 |
|
- getter and setter functions work with concrete values, not abstract |
370 |
|
ones where possible |
371 |
|
- ml-nlffigen command line more flexible (see README file there) |
372 |
|
- some bugs have been fixed (hopefully) |
373 |
|
|
374 |
|
---------------------------------------------------------------------- |
375 |
|
Name: Lal George |
376 |
|
Date: 2002/01/24 |
377 |
|
Tag: george-20020124-risc-ra-interface |
378 |
|
Description: |
379 |
|
|
380 |
|
There is a dramatic simplification in the interface to the |
381 |
|
register allocator for RISC architectures as a result of making |
382 |
|
parallel copy instructions explicit. |
383 |
|
|
384 |
|
---------------------------------------------------------------------- |
385 |
|
Name: Matthias Blume |
386 |
|
Date: 2002/01/22 |
387 |
|
Tag: blume-20020122-x86-ccalls |
388 |
|
Description: |
389 |
|
|
390 |
|
Bug fix for c-calls on x86 (having to do with how char- and |
391 |
|
short-arguments are being handled). |
392 |
|
|
393 |
|
---------------------------------------------------------------------- |
394 |
|
Name: Matthias Blume |
395 |
|
Date: 2002/01/21 |
396 |
|
Tag: blume-20020121-ff |
397 |
|
Description: |
398 |
|
|
399 |
|
Another day of fiddling with the FFI... |
400 |
|
|
401 |
|
1. Bug fix/workaround: CKIT does not complain about negative array |
402 |
|
dimensions, so ml-nlffigen has to guard itself against this possibility. |
403 |
|
(Otherwise a negative dimension would send it into an infinite loop.) |
404 |
|
|
405 |
|
2. Some of the abstract types (light objects, light pointers, most "base" |
406 |
|
types) in structure C are now eqtypes. |
407 |
|
|
408 |
|
3. Added constructors and test functions for NULL function pointers. |
409 |
|
|
410 |
|
---------------------------------------------------------------------- |
411 |
|
Name: Matthias Blume |
412 |
|
Date: 2002/01/18 |
413 |
|
Tag: blume-20020118-ready-for-new-release |
414 |
|
Description: |
415 |
|
|
416 |
|
Made config/srcarchiveurl point to a new place. (Will provide boot |
417 |
|
files shortly.) |
418 |
|
|
419 |
|
Maybe we christen this to be 110.38? |
420 |
|
|
421 |
|
---------------------------------------------------------------------- |
422 |
|
Name: Matthias Blume |
423 |
|
Date: 2002/01/18 |
424 |
|
Tag: blume-20020118-more-ffifiddle |
425 |
|
Description: |
426 |
|
|
427 |
|
Today's FFI fiddling: |
428 |
|
|
429 |
|
- Provided a structure CGetSet with "convenient" versions of C.Get.* and |
430 |
|
C.Set.* that use concrete (MLRep.*) arguments and results instead |
431 |
|
of abstract ones. |
432 |
|
|
433 |
|
- Provided word-style bit operations etc. for "int" representation |
434 |
|
types in MLRep.S<Foo>Bitops where <Foo> ranges over Char, Int, Short, |
435 |
|
and Long. |
436 |
|
|
437 |
|
---------------------------------------------------------------------- |
438 |
|
Name: Matthias Blume |
439 |
Date: 2002/01/18 |
Date: 2002/01/18 |
440 |
Tag: blume-20020118-use-x86-fp |
Tag: blume-20020118-use-x86-fp |
441 |
Description: |
Description: |
870 |
|
|
871 |
---------------------------------------------------------------------- |
---------------------------------------------------------------------- |
872 |
Name: Matthias Blume |
Name: Matthias Blume |
|
>>>>>>> 1.169 |
|
873 |
Date: 2001/09/18 15:35:00 EDT |
Date: 2001/09/18 15:35:00 EDT |
874 |
Tag: blume-20010918-readme11036 |
Tag: blume-20010918-readme11036 |
875 |
Description: |
Description: |