11 |
Date: |
Date: |
12 |
Tag: <post-commit CVS tag> |
Tag: <post-commit CVS tag> |
13 |
Description: |
Description: |
14 |
|
|
15 |
|
---------------------------------------------------------------------- |
16 |
|
Name: Matthias Blume |
17 |
|
Date: 2001/03/19 17:53:00 EST |
18 |
|
Tag: blume-20010319-parallel |
19 |
|
Description: |
20 |
|
|
21 |
|
Parallel make (using compile servers) now works again. |
22 |
|
|
23 |
|
To this end, CM.stabilize and CMB.make have been modified to work in |
24 |
|
two passes when compile servers are attached: |
25 |
|
1. Compile everything, do not perform stabilization; this pass |
26 |
|
uses compile servers |
27 |
|
2. Stabilize everything; this pass does not use compile servers |
28 |
|
If there are no compile servers, the two passes are combined into one |
29 |
|
(as before). Splitting the passes increases the inherent parallelism |
30 |
|
in the dependency graph because the entire graph including all |
31 |
|
libraries is available at the same time. This, in turn, improves |
32 |
|
server utilization. The downside is that the master process will |
33 |
|
have to do some extra work after compilation is done (because for |
34 |
|
technical reasons it must re-read all the binfiles during stabilization). |
35 |
|
|
36 |
|
---------------------------------------------------------------------- |
37 |
|
Name: Matthias Blume |
38 |
|
Date: 2001/03/16 12:22:00 EST |
39 |
|
Tag: blume-20010316-bootfiles |
40 |
|
Description: |
41 |
|
|
42 |
|
Created a new set of bootfiles (for your automatic installation convenience). |
43 |
|
|
44 |
|
---------------------------------------------------------------------- |
45 |
|
Name: Matthias Blume |
46 |
|
Date: 2001/03/16 11:00:00 EST |
47 |
|
Tag: blume-20010316-MLTREE-fixup |
48 |
|
Description: |
49 |
|
|
50 |
|
This is a minor fixup for an (untagged) earlier commit by Allen. |
51 |
|
(A file was missing). |
52 |
|
|
53 |
|
---------------------------------------------------------------------- |
54 |
|
Name: Allen Leung |
55 |
|
Date: Mon Mar 5 18:54:57 EST 2001 |
56 |
|
Tag: leunga-20010305-cut-support |
57 |
|
|
58 |
|
1. New support for alternative control-flow in MLTREE. |
59 |
|
Currently we support |
60 |
|
|
61 |
|
FLOW_TO(CALL ...., [k1,...,kn]) |
62 |
|
|
63 |
|
This is needed for 'cuts to' in C-- and try/handle-like constructs |
64 |
|
in Moby |
65 |
|
|
66 |
|
New assembler flag "asm-show-cutsto" to turn on control-flow debugging. |
67 |
|
|
68 |
|
2. Register Allocator |
69 |
|
|
70 |
|
Changes in interface [from Fermin, John] |
71 |
|
|
72 |
|
3. Alpha 8-bit SLL support [Fermin] |
73 |
|
|
74 |
|
4. All architectures |
75 |
|
|
76 |
|
A new module (ClusterExpandCopies) for expanding parallel copies. |
77 |
|
|
78 |
|
---------------------------------------------------------------------- |
79 |
|
Name: Allen Leung |
80 |
|
Date: 2001/02/27 23:07:00 EST |
81 |
|
Tag: leunga-20010227-minor-stuff |
82 |
|
|
83 |
|
1. Alpha bug fix for CMOVNE |
84 |
|
2. Handle mltree COND(..,FCMP ...,...) |
85 |
|
3. Bug fix in simplifier |
86 |
|
|
87 |
|
---------------------------------------------------------------------- |
88 |
|
Name: Matthias Blume |
89 |
|
Date: 2001/01/30 17:50:00 EST |
90 |
|
Tag: blume-20010130-sync |
91 |
|
Description: |
92 |
|
|
93 |
|
This is just a minor update to sync my devel branch with the main brach. |
94 |
|
The only visible change is the addition of some README files. |
95 |
|
|
96 |
|
---------------------------------------------------------------------- |
97 |
|
Name: Matthias Blume |
98 |
|
Date: 2001/01/12 23:30:00 JST |
99 |
|
Tag: blume-20010112-bootfiles |
100 |
|
Description: |
101 |
|
|
102 |
|
Made a new set of bootfiles that goes with the current state of the |
103 |
|
repository. |
104 |
|
|
105 |
|
---------------------------------------------------------------------- |
106 |
|
Name: Matthias Blume |
107 |
|
Date: 2001/01/12 21:20:00 JST |
108 |
|
Tag: blume-20010112-sync |
109 |
|
Description: |
110 |
|
|
111 |
|
I am just flushing out some minor changes that had accumulated in |
112 |
|
my private branch in order to sync with the main tree. (This is |
113 |
|
mainly because I had CVS trouble when trying to merge _into_ my |
114 |
|
private branch.) |
115 |
|
|
116 |
|
Most people should be completely unaffected by this. |
117 |
|
|
118 |
|
---------------------------------------------------------------------- |
119 |
|
Name: Allen Leung |
120 |
|
Date: Thu Jan 11 21:03:00 EST 2001 |
121 |
|
Tag: leunga-20010111-labexp=mltree |
122 |
|
Description: |
123 |
|
|
124 |
|
1. Removed the type LabelExp and replace it by MLTree. |
125 |
|
2. Rewritten mltree-simplify with the pattern matcher tool. |
126 |
|
3. There were some bugs in alpha code generator which would break |
127 |
|
64-bit code generation. |
128 |
|
4. Redo the tools to generate code with the |
129 |
|
5. The CM files in MLRISC (and in src/system/smlnj/MLRISC) |
130 |
|
are now generated by perl scripts. |
131 |
|
|
132 |
|
---------------------------------------------------------------------- |
133 |
|
Name: Matthias Blume |
134 |
|
Date: 2001/01/10 21:55:00 JST |
135 |
|
Tag: blume-20010110-rcc |
136 |
|
Description: |
137 |
|
|
138 |
|
The RCC stuff now seems to work (but only on the x86). |
139 |
|
This required hacking of the c-calls interface (and -implementation) in |
140 |
|
MLRISC. |
141 |
|
|
142 |
|
Normal compiler users should be unaffected. |
143 |
|
|
144 |
|
---------------------------------------------------------------------- |
145 |
|
Name: Matthias Blume |
146 |
|
Date: 2001/01/09 01:20:00 JST |
147 |
|
Tag: blume-20010109-rcc |
148 |
|
Description: |
149 |
|
|
150 |
|
This is a fairly big patch, flushing out a large number of pending |
151 |
|
changes that I made to my development copy over the last couple of days. |
152 |
|
|
153 |
|
Of practical relevance at this moment is a workaround for a pickling |
154 |
|
bug that Allen ran into the other day. The cause of the bug itself is |
155 |
|
still unknown and it might be hard to fix it properly, but the |
156 |
|
workaround has some merits of its own (namely somewhat reducing pickling |
157 |
|
overhead for certain libraries). Therefore, I think this solution should |
158 |
|
be satisfactory at this time. |
159 |
|
|
160 |
|
The rest of the changes (i.e., the vast majority) has to do with my |
161 |
|
ongoing efforts of providing direct support for C function calls from |
162 |
|
ML. At the moment there is a new primop "RAW_CCALL", typing magic |
163 |
|
in types/cproto.sml (invoked from FLINT/trans/translate.sml), a new |
164 |
|
case in the FLINT CPS datatype (RCC), changes to cps/convert.sml to |
165 |
|
translate uses of RAW_CCALL into RCC, and changes to mlriscGen.sml to |
166 |
|
handle RCC. |
167 |
|
|
168 |
|
The last part (the changes to mlriscGen.sml) are still known to be |
169 |
|
wrong on the x86 and not implemented on all other architectures. But |
170 |
|
the infrastructure is in place. I had to change a few functor |
171 |
|
signatures in the backend to be able to route the CCalls interface |
172 |
|
from MLRISC there, and I had to specialize the mltree type (on the |
173 |
|
x86) to include the necessary extensions. (The extensions themselves |
174 |
|
were already there and redy to go in MLRISC/x86). |
175 |
|
|
176 |
|
Everything should be very happy as soon as someone helps me with |
177 |
|
mlriscGen.sml... |
178 |
|
|
179 |
|
In any case, nothing of this should matter to anyone as long as the |
180 |
|
new primop is not being used (which is going to be the case unless you |
181 |
|
find it where I hid it :). The rest of the compiler is completely |
182 |
|
unaffected. |
183 |
|
|
184 |
|
---------------------------------------------------------------------- |
185 |
|
Name: Matthias Blume |
186 |
|
Date: 2001/01/05 00:30:00 JST |
187 |
|
Tag: blume-20010105-primops |
188 |
|
Description: |
189 |
|
|
190 |
|
Added some experimental support for work that I am doing right now. |
191 |
|
These changes mostly concern added primops, but there is also a new |
192 |
|
experimental C library in the runtime system (but currently not enabled |
193 |
|
anywhere except on Linux/X86). |
194 |
|
|
195 |
|
In the course of adding primops (and playing with them), I discovered that |
196 |
|
Zhong's INL_PRIM hack (no type info for certain primops) was, in fact, badly |
197 |
|
broken. (Zhong was very right he labeled this stuff as "major gross hack".) |
198 |
|
To recover, I made type information in INL_PRIM mandatory and changed |
199 |
|
prim.sml as well as built-in.sml accordingly. The InLine structure now |
200 |
|
has complete, correct type information (i.e., no bottom types). |
201 |
|
|
202 |
|
Since all these changes mean that we need new binfiles, I also bumped the |
203 |
|
version number to 110.32.1. |
204 |
|
|
205 |
|
---------------------------------------------------------------------- |
206 |
|
Name: Matthias Blume |
207 |
|
Date: 2000/12/30 22:10:00 JST |
208 |
|
Tag: blume-20001230-various |
209 |
|
Description: |
210 |
|
|
211 |
|
Added proxy libraries for MLRISC and let MLRISC libraries refer |
212 |
|
to each other using path anchors. (See CM manual for explanation.) |
213 |
|
|
214 |
|
Updated CM documentation. |
215 |
|
|
216 |
|
Fixed some bugs in CM. |
217 |
|
|
218 |
|
Implemented "proxy" libraries (= syntactic sugar for CM). |
219 |
|
|
220 |
|
Added "-quiet" option to makeml and changed runtime system accordingly. |
221 |
|
|
222 |
|
Added cleanup handler for exportML to reset timers and compiler stats. |
223 |
|
|
224 |
|
---------------------------------------------------------------------- |
225 |
|
Name: Lal George |
226 |
|
Date: 2000/12/22 22:22:58 EST 2000 |
227 |
|
Tag: Release_110_32 |
228 |
|
Description: |
229 |
|
|
230 |
|
Infinite precision used throughout MLRISC. |
231 |
|
see MLRISC/mltree/machine-int.sig |
232 |
|
|
233 |
|
---------------------------------------------------------------------- |
234 |
|
Name: Matthias Blume |
235 |
|
Date: 2000/12/22 23:16:00 JST |
236 |
|
Tag: blume-20001222-warn |
237 |
|
Description: |
238 |
|
|
239 |
|
Corrected wording and formatting of some CM warning message which I |
240 |
|
broke in my previous patch. |
241 |
|
|
242 |
|
---------------------------------------------------------------------- |
243 |
|
Name: Matthias Blume |
244 |
|
Date: 2000/12/22 21:20:00 JST |
245 |
|
Tag: blume-20001222-anchorenv |
246 |
|
Description: |
247 |
|
|
248 |
|
Fixed CM's handling of anchor environments in connection with CMB.make. |
249 |
|
|
250 |
|
---------------------------------------------------------------------- |
251 |
|
Name: Matthias Blume |
252 |
|
Date: 2000/12/22 13:15:00 JST |
253 |
|
Tag: blume-20001222-cleanup |
254 |
|
Description: |
255 |
|
|
256 |
|
Removed src/cm/ffi which does not (and did not) belong here. |
257 |
|
|
258 |
|
---------------------------------------------------------------------- |
259 |
|
Name: Matthias Blume |
260 |
|
Date: 2000/12/21 23:55:00 JST |
261 |
|
Tag: blume-20001221-exn |
262 |
|
Description: |
263 |
|
|
264 |
|
Probably most important: CM no longer silently swallows all exceptions |
265 |
|
in the compiler. |
266 |
|
Plus: some other minor CM changes. For example, CM now reports some |
267 |
|
sizes for generated binfiles (code, data, envpickle, lambdapickle). |
268 |
|
|
269 |
|
---------------------------------------------------------------------- |
270 |
|
Name: Matthias Blume |
271 |
|
Date: 2000/12/15 00:01:05 JST |
272 |
|
Tag: blume-20001215-dirtool |
273 |
|
Description: |
274 |
|
|
275 |
|
- "dir" tool added. |
276 |
|
- improvements and cleanup to Tools structure |
277 |
|
- documentation updates |
278 |
|
|
279 |
|
---------------------------------------------------------------------- |
280 |
|
Name: Allen Leung |
281 |
|
Date: Thu Dec 14 03:45:24 EST 2000 |
282 |
|
Description: |
283 |
|
Tag: leunga-20001214-int-inf |
284 |
|
Description: |
285 |
|
|
286 |
|
In IntInf, added these standard functions, which are missing from our |
287 |
|
implementation: |
288 |
|
|
289 |
|
andb : int * int -> int |
290 |
|
xorb : int * int -> int |
291 |
|
orb : int * int -> int |
292 |
|
notb : int -> int |
293 |
|
<< : int * word -> int |
294 |
|
~>> : int * word -> int |
295 |
|
|
296 |
|
Not tested, I hope they are correct. |
297 |
|
|
298 |
|
---------------------------------------------------------------------- |
299 |
|
Name: Allen Leung |
300 |
|
Date: Fri Dec 8 19:23:26 EST 2000 |
301 |
|
Description: |
302 |
|
Tag: leunga-20001208-nowhere |
303 |
|
Description: |
304 |
|
|
305 |
|
Slight improvements to the 'nowhere' tool to handle OR-patterns, |
306 |
|
to generate better error messages etc. Plus a brief manual. |
307 |
|
|
308 |
|
---------------------------------------------------------------------- |
309 |
|
Name: Lal George |
310 |
|
Date: 2000/12/08 09:54:02 EST 2000 |
311 |
|
Tag: Release_110_31 |
312 |
|
Description: |
313 |
|
|
314 |
|
- Version 110.31 |
315 |
|
---------------------------------------------------------------------- |
316 |
|
Name: Allen Leung |
317 |
|
Date: Thu Dec 7 22:01:04 EST 2000 |
318 |
|
Tag: leunga-20001207-cell-monster-hack |
319 |
|
Description: |
320 |
|
|
321 |
|
Major MLRISC internal changes. Affect all clients. |
322 |
|
Summary: |
323 |
|
|
324 |
|
1. Type CELLS.cell = int is now replaced by a datatype. |
325 |
|
As a result, the old regmap is now gone. Almost all interfaces |
326 |
|
in MLRISC change as a consequence. |
327 |
|
|
328 |
|
2. A new brand version of machine description tool (v3.0) that generates |
329 |
|
modules expecting the new interface. The old version is removed. |
330 |
|
|
331 |
|
3. The RA interface has been further abstracted into two new functors. |
332 |
|
RISC_RA and X86RA. These functors have much simpler interfaces. |
333 |
|
[See also directory MLRISC/demo.] |
334 |
|
|
335 |
|
4. Some other new source->source code generation tools are available: |
336 |
|
|
337 |
|
a. MLRISC/Tools/RewriteGen -- generate rewriters from rules. |
338 |
|
b. MLRISC/Tools/WhereGen -- expands conditional pattern matching rules. |
339 |
|
I use this tool to generate the peephole optimizers---with the new |
340 |
|
cell type changes, peephole rules are becoming difficult to write |
341 |
|
without conditional pattern matching. |
342 |
|
|
343 |
|
5. More Intmap -> IntHashTable change. Previous changes by Matthias didn't |
344 |
|
cover the entire MLRISC source tree so many things broke. |
345 |
|
|
346 |
|
6. CM files have been moved to the subdirectory MLRISC/cm. |
347 |
|
They are moved because there are a lot of them and they clutter up the |
348 |
|
root dir. |
349 |
|
|
350 |
|
7. More detailed documentation to come... |
351 |
|
|
352 |
|
NOTE: To rebuild from 110.30 (ftp distribution), you'll have to do |
353 |
|
a makeml -rebuild first. This is because of other other |
354 |
|
changes that Matthias has made (see below). |
355 |
|
|
356 |
|
|
357 |
|
---------------------------------------------------------------------- |
358 |
|
Name: Matthias Blume |
359 |
|
Date: 2000/11/30 23:12:00 JST |
360 |
|
Tag: blume-20001130-filereorg |
361 |
|
Description: |
362 |
|
|
363 |
|
Some manual updates and some file reorganizations in CM. |
364 |
|
|
365 |
|
---------------------------------------------------------------------- |
366 |
|
Name: Matthias Blume |
367 |
|
Date: 2000/11/24 17:45:00 JST |
368 |
|
Tag: blume-20001124-link |
369 |
|
Description: |
370 |
|
|
371 |
|
Drastically improved link traversal code for the case that the dynamic |
372 |
|
value was already loaded at bootstrap time. As a result, CM and CMB |
373 |
|
now both load blazingly fast -- even on a very slow machine. Also, |
374 |
|
memory consumption has been further reduced by this. |
375 |
|
|
376 |
|
Warning: The format of the PIDMAP file has changed. THerefore, to |
377 |
|
bootstrap you have to do this: |
378 |
|
|
379 |
|
1. Run CMB.make |
380 |
|
2. Make a symbolic link for the boot directory: |
381 |
|
ln -s sml.boot.ARCH-OS xxx |
382 |
|
3. "Rebuild" the boot directory: |
383 |
|
./makeml -boot xxx -rebuild sml ; rm xxx |
384 |
|
4. Boot normally: |
385 |
|
./makeml |
386 |
|
|
387 |
|
---------------------------------------------------------------------- |
388 |
|
Name: Matthias Blume |
389 |
|
Date: 2000/11/21 21:20:00 JST |
390 |
|
Tag: blume-20001121-tools |
391 |
|
Description: |
392 |
|
|
393 |
|
Continued hacking on autoloading problem -- with success this time. |
394 |
|
Also changed tool-plugin mechanism. See new CM manual. |
395 |
|
|
396 |
|
---------------------------------------------------------------------- |
397 |
|
Name: Matthias Blume |
398 |
|
Date: 2000/11/19 14:30:00 JST |
399 |
|
Tag: blume-20001119-autoload |
400 |
|
Description: |
401 |
|
|
402 |
|
Some hacking to make autoloading faster. Success for CMB, no success |
403 |
|
so far for CM. There is a reduced structure CM' that autoloads faster. |
404 |
|
(This is a temporary, non-documented hack to be eliminated again when |
405 |
|
the general problem is solved.) |
406 |
|
|
407 |
|
---------------------------------------------------------------------- |
408 |
|
Name: Matthias Blume |
409 |
|
Date: 2000/11/17 14:10:00 JST |
410 |
|
Tag: blume-20001117-pickle-lib |
411 |
|
Description: |
412 |
|
|
413 |
|
1. Eliminated comp-lib.cm |
414 |
|
2. Made pickle-lib.cm |
415 |
|
3. Eliminated all uses of intset.sml (from comp-lib.cm) |
416 |
|
4. Replaced all uses of intmap.{sig,sml} (from comp-lib.cm) with |
417 |
|
equivalent constructs from smlnj-lib.cm (INtHashTable). |
418 |
|
5. Point 4. also goes for those uses of intmap.* in MLRISC. |
419 |
|
Duplicated intmap modules thrown out. |
420 |
|
6. Hunted down all duplicated SCC code and replaced it with |
421 |
|
equivalent stuff (GraphSCCFn from smlnj-lib.cm). |
422 |
|
7. Rewrote Feedback module. |
423 |
|
8. Moved sortedlist.sml into viscomp-lib.cm. Eventually it |
424 |
|
should be thrown out and equivalent modules from smlnj-lib.cm |
425 |
|
should be used (IntRedBlackSet, IntListSet, ...). |
426 |
|
|
427 |
|
Confirmed that compiler compiles to fixpoint. |
428 |
|
|
429 |
|
---------------------------------------------------------------------- |
430 |
|
Name: Allen Leung |
431 |
|
Date: 2000/11/10 18:00:00 |
432 |
|
Tag: leunga-20001110-new-x86-fp |
433 |
|
|
434 |
|
A new x86 floating point code generator has been added. |
435 |
|
By default this is turned off. To turn this on, do: |
436 |
|
|
437 |
|
CM.autoload "$smlnj/compiler.cm"; |
438 |
|
Compiler.Control.MLRISC.getFlag "x86-fast-fp" := true; |
439 |
|
|
440 |
|
Changes: |
441 |
|
|
442 |
|
1. Changed FTAN to FPTAN so that the assembly output is correct. |
443 |
|
2. Changed the extension callback for FTANGENT to generate: |
444 |
|
|
445 |
|
fptan |
446 |
|
fstp %st(0) |
447 |
|
instead of |
448 |
|
fptan |
449 |
|
fstpl ftempmem |
450 |
|
|
451 |
|
3. Numerous assembly fixes for x86. |
452 |
|
|
453 |
|
5. Cleaned up the machine code output module x86/x86MC.sml and added |
454 |
|
support for a whole bunch of instructions and addressing modes: |
455 |
|
|
456 |
|
fadd/fsub/fsubr/fmul/fdiv/fdivr %st, %st(n) |
457 |
|
faddp/fsubp/fsubrp/fmulp/fdivp/fdivrp %st, %st(n) |
458 |
|
fadd/fsub/fsubr/fmul/fdiv/fdivr %st(n), %st |
459 |
|
fiadd/fisub/fisubr/fimul/fidiv/fidivr mem |
460 |
|
fxch %st(n) |
461 |
|
fld %st(n) |
462 |
|
fst %st(n) |
463 |
|
fst mem |
464 |
|
fstp %st(n) |
465 |
|
fucom %st(n) |
466 |
|
fucomp %st(n) |
467 |
|
|
468 |
|
All these are now generated when the fast fp mode is turned on. |
469 |
|
|
470 |
|
6. Removed the dedicated registers %st(0), ..., %st(7) from X86CpsRegs |
471 |
|
|
472 |
|
---------------------------------------------------------------------- |
473 |
|
Name: Matthias Blume |
474 |
|
Date: 2000/11/09 11:20:00 JST |
475 |
|
Tag: blume-20001109-scc |
476 |
|
Description: |
477 |
|
|
478 |
|
Eliminated some code duplication: |
479 |
|
|
480 |
|
1. Added "where" clause to GraphSCCFn in SML/NJ Library. |
481 |
|
(Otherwise the functor is useless.) |
482 |
|
2. Used GraphSCCFn where SCCUtilFun was used previously. |
483 |
|
3. Got rid of SCCUtilFun (in comp-lib.cm). |
484 |
|
|
485 |
|
---------------------------------------------------------------------- |
486 |
|
Name: Lal George |
487 |
|
Date: 2000/11/06 09:02:21 EST 2000 |
488 |
|
Tag: Release_110_30 |
489 |
|
Description: |
490 |
|
|
491 |
|
- Version 110.30 |
492 |
|
---------------------------------------------------------------------- |
493 |
|
Name: Matthias Blume |
494 |
|
Date: 2000/11/04 14:45:00 |
495 |
|
Tag: blume-20001104-mlbuild |
496 |
|
Description: |
497 |
|
|
498 |
|
- Made ml-build faster on startup. |
499 |
|
- Documentation fixes. |
500 |
|
|
501 |
|
---------------------------------------------------------------------- |
502 |
|
Name: Matthias Blume |
503 |
|
Date: 2000/11/02 17:00:00 JST |
504 |
|
Tag: blume-20001102-condcomp |
505 |
|
Description: |
506 |
|
|
507 |
|
- Small tweaks to pickler -- new BOOTFILES! |
508 |
|
- Version bumped to 110.29.2. |
509 |
|
- Added conditional compilation facility to init.cmi (see comment there). |
510 |
|
---------------------------------------------------------------------- |
511 |
|
Name: Allen Leung |
512 |
|
Date: 2000/10/23 19:31:00 |
513 |
|
Tag: leunga-20001023-demo-ra |
514 |
|
|
515 |
|
1. Minor RA changes that improves spilling on x86 (affects Moby and C-- only) |
516 |
|
2. Test programs for the graph library updated |
517 |
|
3. Some new MLRISC demo programs added |
518 |
|
|
519 |
|
---------------------------------------------------------------------- |
520 |
|
Name: Matthias Blume |
521 |
|
Date: 2000/08/31 22:15:00 JST |
522 |
|
Tag: blume-20001017-errmsg |
523 |
|
Description: |
524 |
|
|
525 |
|
More error message grief: Where there used to be no messages, there |
526 |
|
now were some that had bogus error regions. Fixed. |
527 |
|
|
528 |
|
---------------------------------------------------------------------- |
529 |
|
Name: Matthias Blume |
530 |
|
Date: 2000/08/31 17:30:00 JST |
531 |
|
Tag: blume-20001017-v110p29p1 |
532 |
|
Description: |
533 |
|
|
534 |
|
I made a version 110.29.1 with new bootfiles. |
535 |
|
|
536 |
|
Changes: Modified pickler/unpickler for faster and leaner unpickling. |
537 |
|
CM documentation changes and a small bugfix in CM's error reporting. |
538 |
|
|
539 |
|
---------------------------------------------------------------------- |
540 |
|
Name: Lal George |
541 |
|
Date: 2000/09/27 14:42:35 EDT |
542 |
|
Tag: george-20000927-nodestatus |
543 |
|
Description: |
544 |
|
|
545 |
|
Changed the type of the nodestatus, so that: |
546 |
|
|
547 |
|
SPILLED(~1) is now SPILLED |
548 |
|
SPILLED(m) where m>=0 is now MEMREG(m) |
549 |
|
SPILLED(s) where s<~1 is now SPILL_LOC(~s) |
550 |
|
|
551 |
|
---------------------------------------------------------------------- |
552 |
|
Name: Matthias Blume |
553 |
|
Date: 2000/09/07 14:45:00 JST |
554 |
|
Tag: blume-20000907-cmerrmsg |
555 |
|
Description: |
556 |
|
|
557 |
|
Small tweak to CM to avoid getting ML syntax error messages twice. |
558 |
|
|
559 |
|
---------------------------------------------------------------------- |
560 |
|
Name: Matthias Blume |
561 |
|
Date: 2000/08/31 18:00:00 JST |
562 |
|
Tag: blume-20000831-cvsbootfiles |
563 |
|
Description: |
564 |
|
|
565 |
|
New URL for boot files (because the 110.29 files on the BL server do |
566 |
|
now work correctly with my updated install scripts for yacc and lex). |
567 |
|
|
568 |
|
---------------------------------------------------------------------- |
569 |
|
Name: Matthias Blume |
570 |
|
Date: 2000/08/08 12:33:00 JST |
571 |
|
Tag: blume-20000808-manual |
572 |
|
Description: |
573 |
|
|
574 |
|
Tiny update to CM manual. |
575 |
|
|
576 |
|
---------------------------------------------------------------------- |
577 |
|
Name: Allen Leung |
578 |
|
Date: 2000/08/7 19:31:00 |
579 |
|
Tag: leunga-20000807-a-whole-bunch-of-stuff |
580 |
|
|
581 |
|
Moby, C--, SSA, x86, machine descriptions etc. Should only affect C-- |
582 |
|
and Mobdy. |
583 |
|
|
584 |
|
1. x86 |
585 |
|
|
586 |
|
a. Fixes to peephole module by John and Dan. |
587 |
|
b. Assembly fix to SETcc by Allen. |
588 |
|
c. Fix to c-call by John. |
589 |
|
d. Fix to spilling by John. (This one deals with the missing FSTPT case) |
590 |
|
e. Instruction selection optimization to SETcc as suggested by John. |
591 |
|
|
592 |
|
For example, |
593 |
|
|
594 |
|
MV(32, x, COND(32, CMP(32, LT, a, b), LI 1, LI 0)) |
595 |
|
|
596 |
|
should generate: |
597 |
|
|
598 |
|
MOVL a, x |
599 |
|
SUBL b, x |
600 |
|
SHRL 31, x |
601 |
|
|
602 |
|
2. IR stuff |
603 |
|
|
604 |
|
A bunch of new DJ-graph related algorithms added. These |
605 |
|
speed up SSA construction. |
606 |
|
|
607 |
|
3. SSA + Scheduling |
608 |
|
|
609 |
|
Added code for SSA and scheduling to the repository |
610 |
|
|
611 |
|
---------------------------------------------------------------------- |
612 |
|
Name: Lal George |
613 |
|
Date: 2000/07/27 11:53:14 EDT |
614 |
|
|
615 |
|
Tag: lal-20000727-linux-ppc |
616 |
|
Description: |
617 |
|
|
618 |
|
Made changes to support Linux PPC. |
619 |
|
p.s. I have confirmation that the 110.29 boot files work fine. |
620 |
|
|
621 |
|
---------------------------------------------------------------------- |
622 |
|
Name: Matthias Blume |
623 |
|
Date: 2000/07/27 17:40:00 JST |
624 |
|
Tag: blume-20000727-scripts |
625 |
|
Description: |
626 |
|
|
627 |
|
!!!! WARNING !!!! |
628 |
|
You must recompile the runtime system! |
629 |
|
!!!! WARNING !!!! |
630 |
|
|
631 |
|
This is basically another round of script-enhancements: |
632 |
|
|
633 |
|
1. sml, ml-build, and ml-makedepend accept options -D and -U to define |
634 |
|
and undefine CM preprocessor symbols. |
635 |
|
|
636 |
|
2. ml-build avoids generating a new heap image if it finds that the |
637 |
|
existing one is still ok. (The condition is that no ML file had to |
638 |
|
be recompiled and all ML files are found to be older that the heap |
639 |
|
file.) |
640 |
|
|
641 |
|
To make this work smoothly, I also hacked the runtime system as |
642 |
|
well as SMLofNJ.SysInfo to get access to the heap image suffix |
643 |
|
(.sparc-solaris, ...) that is currently being used. |
644 |
|
|
645 |
|
Moreover, the signature of CM.mk_standalone has changed. See the |
646 |
|
CM manual. |
647 |
|
|
648 |
|
3. ml-makedepend accepts additional options -n, -a, and -o. (See the |
649 |
|
CM manual for details.) |
650 |
|
|
651 |
|
4. More CM manual updates: |
652 |
|
- all of the above has been documented. |
653 |
|
- there is now a section describing the (CM-related) command line |
654 |
|
arguments that are accepted by the "sml" command |
655 |
|
|
656 |
|
---------------------------------------------------------------------- |
657 |
|
Name: Matthias Blume |
658 |
|
Date: 2000/07/25 16:20:00 JST |
659 |
|
Tag: blume-20000725-makedepend |
660 |
|
Description: |
661 |
|
|
662 |
|
Added a script called ml-makedepend. This can be used in makefiles |
663 |
|
for Unix' make in a way very similar to the "makedepend" command for |
664 |
|
C. |
665 |
|
|
666 |
|
The script internally uses function CM.sources. |
667 |
|
|
668 |
|
Synopsis: |
669 |
|
|
670 |
|
ml-makedepend [-f makefile] cmfile targetname |
671 |
|
|
672 |
|
The default for the makefile is "makefile" (or "Makefile" should |
673 |
|
"makefile" not exist). |
674 |
|
|
675 |
|
ml-makedepend adds a cmfile/targetname-specific section to this |
676 |
|
makefile (after removing the previous version of this section). The |
677 |
|
section contains a single dependency specification with targetname on |
678 |
|
the LHS (targetname is an arbitrary name), and a list of files derived |
679 |
|
from the cmfile on the RHS. Some of the files on the RHS are |
680 |
|
ARCH/OPSYS-specific. Therefore, ml-makedepend inserts references to |
681 |
|
"make" variables $(ARCH) and $(OPSYS) in place of the corresponding |
682 |
|
path names. The makefile writer is responsible for making sure that |
683 |
|
these variables have correct at the time "make" is invoked. |
684 |
|
|
685 |
|
---------------------------------------------------------------------- |
686 |
|
Name: Matthias Blume |
687 |
|
Date: 2000/07/22 23:30:00 JST |
688 |
|
Tag: blume-20000722-urlupdate |
689 |
|
Description: |
690 |
|
|
691 |
|
Changed BOOT and config/srcarchiveurl to point to BL server: |
692 |
|
|
693 |
|
ftp://ftp.research.bell-labs.com/dist/smlnj/working/110.29/ |
694 |
|
|
695 |
|
---------------------------------------------------------------------- |
696 |
|
Name: Matthias Blume |
697 |
|
Date: 2000/07/18 18:00:00 JST |
698 |
|
Tag: blume-20000718-Version_110_29 |
699 |
|
Description: |
700 |
|
|
701 |
|
1. Updated src/compiler/TopLevel/main/version.sml to version 110.29 |
702 |
|
|
703 |
|
2. Updated config/version to 110.29 |
704 |
|
|
705 |
|
3. Updated config/srcarchiveurl |
706 |
|
|
707 |
|
3. New boot files! |
708 |
|
ftp://ftp.cs.princeton.edu/pub/people/blume/sml/110.29-autofetch |
709 |
|
|
710 |
|
---------------------------------------------------------------------- |
711 |
|
Name: Matthias Blume |
712 |
|
Date: 2000/07/11 13:58:00 JST |
713 |
|
Tag: blume-20000711-doctypo |
714 |
|
Description: |
715 |
|
|
716 |
|
Fixed a few typos in CM manual. |
717 |
|
|
718 |
---------------------------------------------------------------------- |
---------------------------------------------------------------------- |
719 |
Name: Allen Leung |
Name: Allen Leung |
720 |
Date: 2000/06/15 00:38:00 |
Date: 2000/06/15 00:38:00 |