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: 2001/05/22 16:30:00 EDT |
18 |
|
Tag: blume-20010522-minitut |
19 |
|
Description: |
20 |
|
|
21 |
|
1. Bug fix in ml-nlffigen; now (hopefully) correctly handling |
22 |
|
struct returns. |
23 |
|
2. Added src/ml-nlffi-lib/Doc/mini-tutorial.txt. This is some very |
24 |
|
incomplete, preliminary documentation for NLFFI. |
25 |
|
|
26 |
|
---------------------------------------------------------------------- |
27 |
|
Name: Matthias Blume |
28 |
|
Date: 2001/05/14 11:30:00 EDT |
29 |
|
Tag: blume-20010514-script |
30 |
|
Description: |
31 |
|
|
32 |
|
Some bugs in install script fixed. |
33 |
|
|
34 |
|
In addition to that I also made a slight change to the NLFFI API: |
35 |
|
Functors generated by ml-nlffigen now take the dynamic library as a |
36 |
|
straight functor argument, not as a suspended one. (The original |
37 |
|
functor code used to force the suspension right away anyway, so there |
38 |
|
was nothing gained by this complication of the interface.) |
39 |
|
|
40 |
|
---------------------------------------------------------------------- |
41 |
|
Name: Matthias Blume |
42 |
|
Date: 2001/05/11 14:35:00 EDT |
43 |
|
Tag: blume-20010511-ml-nlffi |
44 |
|
Description: |
45 |
|
|
46 |
|
I finally took the plunge and added my new FFI code to the main |
47 |
|
repository. For x86-linux it is now ready for prime-time. |
48 |
|
|
49 |
|
There are two new subdirectories of "src": |
50 |
|
|
51 |
|
- ml-nlffi-lib: |
52 |
|
The utility library for programs using the FFI interface. |
53 |
|
Here is the implementation of $/c.cm and its associated low-level |
54 |
|
partners $/c-int.cm and $/memory.cm. |
55 |
|
- ml-nlffigen: |
56 |
|
A stand-alone program for generating ML glue code from C source |
57 |
|
code. |
58 |
|
|
59 |
|
Building ml-nlffigen requires $/ckit-lib.cm. |
60 |
|
|
61 |
|
The config/install.sh script has been updates to do the Right Thing |
62 |
|
(hopefully). |
63 |
|
|
64 |
|
Notice that the source tree for the C-Kit will not be put under "src" |
65 |
|
but directly under the installation root directory. (This is the |
66 |
|
structure that currently exists on the CVS server when you check out |
67 |
|
module "sml".) Fortunately, config/install.sh knows about this oddity. |
68 |
|
|
69 |
|
Bugs: No documentation yet. |
70 |
|
|
71 |
|
---------------------------------------------------------------------- |
72 |
|
Name: Matthias Blume |
73 |
|
Date: 2001/05/09 16:35:00 EDT |
74 |
|
Tag: blume-20010509-cpscontract |
75 |
|
Description: |
76 |
|
|
77 |
|
Fixed a bug in the accounting code in cpsopt/contract.sml. (The |
78 |
|
wrapper/unwrapper elimination did not decrement usage counts and some |
79 |
|
dead variables got overlooked by the dead-up logic.) |
80 |
|
|
81 |
|
---------------------------------------------------------------------- |
82 |
|
Name: Lal George |
83 |
|
Date: 2001/05/08 17:26:09 EDT |
84 |
|
Tag: george-20010508-omit-frameptr |
85 |
|
Description: |
86 |
|
|
87 |
|
Changes to implement the omit-frame-pointer optimization to support |
88 |
|
raw C calls. For now, there is only support on the Intel x86, but |
89 |
|
other architectures will follow as more experience is gained with this. |
90 |
|
|
91 |
|
|
92 |
|
---------------------------------------------------------------------- |
93 |
|
Name: Matthias Blume |
94 |
|
Date: 2001/05/07 14:40:00 EDT |
95 |
|
Tag: blume-20010507-proxies |
96 |
|
Description: |
97 |
|
|
98 |
|
I made into "proxy libraries" all libraries that qualify for such a |
99 |
|
change. (A qualifying library is a library that has another library or |
100 |
|
groups as its sole member and repeats that member's export list |
101 |
|
verbatim. A proxy library avoids this repetition by omitting its export |
102 |
|
list, effectively inheriting the list that its (only) member exports. |
103 |
|
See the CM manual for more explanation.) |
104 |
|
The main effect is that explicit export lists for these libraries |
105 |
|
do not have to be kepts in sync, making maintenance a bit easier. |
106 |
|
|
107 |
|
I also added copyright notices to many .cm-files. |
108 |
|
|
109 |
|
Last but not least, I made a new set of bootfiles. |
110 |
|
|
111 |
|
---------------------------------------------------------------------- |
112 |
|
Name: Matthias Blume |
113 |
|
Date: 2001/05/04 17:00:00 EDT |
114 |
|
Tag: blume-20010504-cm-lsplit |
115 |
|
Description: |
116 |
|
|
117 |
|
0. John merged pending changes to $/smlnj-lib.cm |
118 |
|
|
119 |
|
1. Allen's previous change accidentally backed out of one of Lal's |
120 |
|
earlier changes. I undid this mistake (re-introducing Lal's change). |
121 |
|
|
122 |
|
2. I used the new topOrder' function from graph-scc.sml (from $/smlnj-lib.cm) |
123 |
|
within the compiler where applicable. There is some code simplification |
124 |
|
because of that. |
125 |
|
|
126 |
|
3. The "split" phase (in FLINT) is now part of the default list of phases. |
127 |
|
Compiler.Control.LambdaSplitting.* can be used to globally control the |
128 |
|
lambda-splitting (cross-module-inlining) engine. In addition to that, |
129 |
|
it can now also be controlled on a per-source basis: CM has been taught |
130 |
|
a new tool parameter applicable to ML source files. |
131 |
|
|
132 |
|
- To turn lambda-splitting off completely: |
133 |
|
local open Compiler.Control.LambdaSplitting in |
134 |
|
val _ = set Off |
135 |
|
end |
136 |
|
- To make "no lambda-splitting" the global default (but allow per-source |
137 |
|
overriding); this is the initial setting: |
138 |
|
local open Compiler.Control.LambdaSplitting in |
139 |
|
val _ = set (Default NONE) |
140 |
|
end |
141 |
|
- To make "lambda-splitting with aggressiveness a" the global default |
142 |
|
(and allow per-source overriding): |
143 |
|
local open Compiler.Control.LambdaSplitting in |
144 |
|
val _ = set (Default (SOME a)) |
145 |
|
end |
146 |
|
|
147 |
|
- To turn lambda-splitting off for a given ML souce file (say: a.sml) |
148 |
|
write (in the respective .cm-file): |
149 |
|
a.sml (lambdasplitting:off) |
150 |
|
- To turn lambda-splitting for a.sml on with minimal aggressiveness: |
151 |
|
a.sml (lambdasplitting:on) |
152 |
|
- To turn lambda-splitting for a.sml on with aggressiveness <a> (where |
153 |
|
<a> is a decimal non-negative integer): |
154 |
|
a.sml (lambdasplitting:<a>) |
155 |
|
- To turn lambda-splitting for a.sml on with maximal aggressiveness: |
156 |
|
a.sml (lambdasplitting:infinity) |
157 |
|
- To use the global default for a.sml: |
158 |
|
a.sml (lambdasplitting:default) |
159 |
|
or simply |
160 |
|
a.sml |
161 |
|
|
162 |
|
---------------------------------------------------------------------- |
163 |
|
Name: Allen Leung |
164 |
|
Date: 2001/05/04 01:57:00 EDT |
165 |
|
Tag: leunga-20010504-sync |
166 |
|
Description: |
167 |
|
|
168 |
|
MLRISC features. |
169 |
|
|
170 |
|
1. Fix to CMPXCHG instructions. |
171 |
|
2. Changed RA interface to allow annotations in callbacks. |
172 |
|
3. Added a new method to the stream interface to allow annotations updates. |
173 |
|
|
174 |
|
---------------------------------------------------------------------- |
175 |
|
Name: Matthias Blume |
176 |
|
Date: 2001/05/01 11:45:00 EDT |
177 |
|
Tag: blume-20010501-pcedittmp |
178 |
|
Description: |
179 |
|
|
180 |
|
Changed install.sh to use the current working directory instead of |
181 |
|
/usr/tmp for a temporary file (pcedittmp). The previous choice |
182 |
|
of /usr/tmp caused trouble with MacOS X because of file premission |
183 |
|
problems. |
184 |
|
|
185 |
|
---------------------------------------------------------------------- |
186 |
|
Name: Matthias Blume |
187 |
|
Date: 2001/04/20 11:10:00 EDT |
188 |
|
Tag: blume-20010420-inMLflag |
189 |
|
Description: |
190 |
|
|
191 |
|
- added vp_limitPtrMask to vproc-state.h |
192 |
|
(for use by the raw-C-calls mechanism to implement proper interrupt |
193 |
|
handling) |
194 |
|
- made the ML compiler aware of various data-structure offsets so it |
195 |
|
can generate code for accessing the vp_inML flag and vp_limitPtrMask |
196 |
|
- tweaked mlriscGen.sml to have it emit interrupt-handling code for |
197 |
|
raw C-calls |
198 |
|
|
199 |
|
---------------------------------------------------------------------- |
200 |
|
Name: Lal George |
201 |
|
Date: 2001/04/20 09:15:28 EDT |
202 |
|
Tag: george-20010420-macosX |
203 |
|
Description: |
204 |
|
|
205 |
|
- Changes to port to Mac OS X; Darwin. |
206 |
|
|
207 |
|
- In the process I found that sqrt was broken on the PPC, because the |
208 |
|
fsqrt instruction is not implemented. |
209 |
|
|
210 |
|
---------------------------------------------------------------------- |
211 |
|
Name: Matthias Blume |
212 |
|
Date: 2001/04/18 12:45:00 EDT |
213 |
|
Tag: blume-20010418-ccalls |
214 |
|
Description: |
215 |
|
|
216 |
|
- fixed two off-by-4 errors in the x86-specific c-calls implementation |
217 |
|
(this bug prevented structure arguments containing pointers from being |
218 |
|
passed correctly) |
219 |
|
- changed the raw-C-call code in mlriscGen.sml in such a way that |
220 |
|
structure arguments are represented as a pointer to the beginning |
221 |
|
of the structure (instead of having a series of synthesized arguments, |
222 |
|
one for each structure member) |
223 |
|
|
224 |
|
- made makeml script's verbosity level configurable via environment |
225 |
|
variable (MAKEML_VERBOSITY) |
226 |
|
|
227 |
|
- eliminated placeholder implementations for f32l, w16s, i16s, and f32s |
228 |
|
in rawmem-x86.sml; we are now using the real thing |
229 |
|
|
230 |
|
---------------------------------------------------------------------- |
231 |
|
Name: Matthias Blume |
232 |
|
Date: 2001/03/22 16:25:00 EST |
233 |
|
Tag: blume-20010322-bootfiles |
234 |
|
Description: |
235 |
|
|
236 |
|
Created a new set of bootfiles (for your automatic installation convenience). |
237 |
|
|
238 |
|
---------------------------------------------------------------------- |
239 |
|
Name: Matthias Blume |
240 |
|
Date: 2001/03/22 15:10:00 EST |
241 |
|
Tag: blume-20010322-rawmem-parcm |
242 |
|
Description: |
243 |
|
|
244 |
|
1. All "raw memory access" primitives for the new FFI are implemented now |
245 |
|
(at least on the x86). |
246 |
|
2. Some further cleanup of CM's parallel make mechanism. |
247 |
|
|
248 |
|
---------------------------------------------------------------------- |
249 |
|
Name: Matthias Blume |
250 |
|
Date: 2001/03/19 17:53:00 EST |
251 |
|
Tag: blume-20010319-parallel |
252 |
|
Description: |
253 |
|
|
254 |
|
Parallel make (using compile servers) now works again. |
255 |
|
|
256 |
|
To this end, CM.stabilize and CMB.make have been modified to work in |
257 |
|
two passes when compile servers are attached: |
258 |
|
1. Compile everything, do not perform stabilization; this pass |
259 |
|
uses compile servers |
260 |
|
2. Stabilize everything; this pass does not use compile servers |
261 |
|
If there are no compile servers, the two passes are combined into one |
262 |
|
(as before). Splitting the passes increases the inherent parallelism |
263 |
|
in the dependency graph because the entire graph including all |
264 |
|
libraries is available at the same time. This, in turn, improves |
265 |
|
server utilization. The downside is that the master process will |
266 |
|
have to do some extra work after compilation is done (because for |
267 |
|
technical reasons it must re-read all the binfiles during stabilization). |
268 |
|
|
269 |
|
---------------------------------------------------------------------- |
270 |
|
Name: Matthias Blume |
271 |
|
Date: 2001/03/16 12:22:00 EST |
272 |
|
Tag: blume-20010316-bootfiles |
273 |
|
Description: |
274 |
|
|
275 |
|
Created a new set of bootfiles (for your automatic installation convenience). |
276 |
|
|
277 |
|
---------------------------------------------------------------------- |
278 |
|
Name: Matthias Blume |
279 |
|
Date: 2001/03/16 11:00:00 EST |
280 |
|
Tag: blume-20010316-MLTREE-fixup |
281 |
|
Description: |
282 |
|
|
283 |
|
This is a minor fixup for an (untagged) earlier commit by Allen. |
284 |
|
(A file was missing). |
285 |
|
|
286 |
|
---------------------------------------------------------------------- |
287 |
|
Name: Allen Leung |
288 |
|
Date: Mon Mar 5 18:54:57 EST 2001 |
289 |
|
Tag: leunga-20010305-cut-support |
290 |
|
|
291 |
|
1. New support for alternative control-flow in MLTREE. |
292 |
|
Currently we support |
293 |
|
|
294 |
|
FLOW_TO(CALL ...., [k1,...,kn]) |
295 |
|
|
296 |
|
This is needed for 'cuts to' in C-- and try/handle-like constructs |
297 |
|
in Moby |
298 |
|
|
299 |
|
New assembler flag "asm-show-cutsto" to turn on control-flow debugging. |
300 |
|
|
301 |
|
2. Register Allocator |
302 |
|
|
303 |
|
Changes in interface [from Fermin, John] |
304 |
|
|
305 |
|
3. Alpha 8-bit SLL support [Fermin] |
306 |
|
|
307 |
|
4. All architectures |
308 |
|
|
309 |
|
A new module (ClusterExpandCopies) for expanding parallel copies. |
310 |
|
|
311 |
|
---------------------------------------------------------------------- |
312 |
|
Name: Allen Leung |
313 |
|
Date: 2001/02/27 23:07:00 EST |
314 |
|
Tag: leunga-20010227-minor-stuff |
315 |
|
|
316 |
|
1. Alpha bug fix for CMOVNE |
317 |
|
2. Handle mltree COND(..,FCMP ...,...) |
318 |
|
3. Bug fix in simplifier |
319 |
|
|
320 |
|
---------------------------------------------------------------------- |
321 |
|
Name: Matthias Blume |
322 |
|
Date: 2001/01/30 17:50:00 EST |
323 |
|
Tag: blume-20010130-sync |
324 |
|
Description: |
325 |
|
|
326 |
|
This is just a minor update to sync my devel branch with the main brach. |
327 |
|
The only visible change is the addition of some README files. |
328 |
|
|
329 |
|
---------------------------------------------------------------------- |
330 |
|
Name: Matthias Blume |
331 |
|
Date: 2001/01/12 23:30:00 JST |
332 |
|
Tag: blume-20010112-bootfiles |
333 |
|
Description: |
334 |
|
|
335 |
|
Made a new set of bootfiles that goes with the current state of the |
336 |
|
repository. |
337 |
|
|
338 |
|
---------------------------------------------------------------------- |
339 |
|
Name: Matthias Blume |
340 |
|
Date: 2001/01/12 21:20:00 JST |
341 |
|
Tag: blume-20010112-sync |
342 |
|
Description: |
343 |
|
|
344 |
|
I am just flushing out some minor changes that had accumulated in |
345 |
|
my private branch in order to sync with the main tree. (This is |
346 |
|
mainly because I had CVS trouble when trying to merge _into_ my |
347 |
|
private branch.) |
348 |
|
|
349 |
|
Most people should be completely unaffected by this. |
350 |
|
|
351 |
|
---------------------------------------------------------------------- |
352 |
|
Name: Allen Leung |
353 |
|
Date: Thu Jan 11 21:03:00 EST 2001 |
354 |
|
Tag: leunga-20010111-labexp=mltree |
355 |
|
Description: |
356 |
|
|
357 |
|
1. Removed the type LabelExp and replace it by MLTree. |
358 |
|
2. Rewritten mltree-simplify with the pattern matcher tool. |
359 |
|
3. There were some bugs in alpha code generator which would break |
360 |
|
64-bit code generation. |
361 |
|
4. Redo the tools to generate code with the |
362 |
|
5. The CM files in MLRISC (and in src/system/smlnj/MLRISC) |
363 |
|
are now generated by perl scripts. |
364 |
|
|
365 |
|
---------------------------------------------------------------------- |
366 |
|
Name: Matthias Blume |
367 |
|
Date: 2001/01/10 21:55:00 JST |
368 |
|
Tag: blume-20010110-rcc |
369 |
|
Description: |
370 |
|
|
371 |
|
The RCC stuff now seems to work (but only on the x86). |
372 |
|
This required hacking of the c-calls interface (and -implementation) in |
373 |
|
MLRISC. |
374 |
|
|
375 |
|
Normal compiler users should be unaffected. |
376 |
|
|
377 |
|
---------------------------------------------------------------------- |
378 |
|
Name: Matthias Blume |
379 |
|
Date: 2001/01/09 01:20:00 JST |
380 |
|
Tag: blume-20010109-rcc |
381 |
|
Description: |
382 |
|
|
383 |
|
This is a fairly big patch, flushing out a large number of pending |
384 |
|
changes that I made to my development copy over the last couple of days. |
385 |
|
|
386 |
|
Of practical relevance at this moment is a workaround for a pickling |
387 |
|
bug that Allen ran into the other day. The cause of the bug itself is |
388 |
|
still unknown and it might be hard to fix it properly, but the |
389 |
|
workaround has some merits of its own (namely somewhat reducing pickling |
390 |
|
overhead for certain libraries). Therefore, I think this solution should |
391 |
|
be satisfactory at this time. |
392 |
|
|
393 |
|
The rest of the changes (i.e., the vast majority) has to do with my |
394 |
|
ongoing efforts of providing direct support for C function calls from |
395 |
|
ML. At the moment there is a new primop "RAW_CCALL", typing magic |
396 |
|
in types/cproto.sml (invoked from FLINT/trans/translate.sml), a new |
397 |
|
case in the FLINT CPS datatype (RCC), changes to cps/convert.sml to |
398 |
|
translate uses of RAW_CCALL into RCC, and changes to mlriscGen.sml to |
399 |
|
handle RCC. |
400 |
|
|
401 |
|
The last part (the changes to mlriscGen.sml) are still known to be |
402 |
|
wrong on the x86 and not implemented on all other architectures. But |
403 |
|
the infrastructure is in place. I had to change a few functor |
404 |
|
signatures in the backend to be able to route the CCalls interface |
405 |
|
from MLRISC there, and I had to specialize the mltree type (on the |
406 |
|
x86) to include the necessary extensions. (The extensions themselves |
407 |
|
were already there and redy to go in MLRISC/x86). |
408 |
|
|
409 |
|
Everything should be very happy as soon as someone helps me with |
410 |
|
mlriscGen.sml... |
411 |
|
|
412 |
|
In any case, nothing of this should matter to anyone as long as the |
413 |
|
new primop is not being used (which is going to be the case unless you |
414 |
|
find it where I hid it :). The rest of the compiler is completely |
415 |
|
unaffected. |
416 |
|
|
417 |
|
---------------------------------------------------------------------- |
418 |
|
Name: Matthias Blume |
419 |
|
Date: 2001/01/05 00:30:00 JST |
420 |
|
Tag: blume-20010105-primops |
421 |
|
Description: |
422 |
|
|
423 |
|
Added some experimental support for work that I am doing right now. |
424 |
|
These changes mostly concern added primops, but there is also a new |
425 |
|
experimental C library in the runtime system (but currently not enabled |
426 |
|
anywhere except on Linux/X86). |
427 |
|
|
428 |
|
In the course of adding primops (and playing with them), I discovered that |
429 |
|
Zhong's INL_PRIM hack (no type info for certain primops) was, in fact, badly |
430 |
|
broken. (Zhong was very right he labeled this stuff as "major gross hack".) |
431 |
|
To recover, I made type information in INL_PRIM mandatory and changed |
432 |
|
prim.sml as well as built-in.sml accordingly. The InLine structure now |
433 |
|
has complete, correct type information (i.e., no bottom types). |
434 |
|
|
435 |
|
Since all these changes mean that we need new binfiles, I also bumped the |
436 |
|
version number to 110.32.1. |
437 |
|
|
438 |
|
---------------------------------------------------------------------- |
439 |
|
Name: Matthias Blume |
440 |
|
Date: 2000/12/30 22:10:00 JST |
441 |
|
Tag: blume-20001230-various |
442 |
|
Description: |
443 |
|
|
444 |
|
Added proxy libraries for MLRISC and let MLRISC libraries refer |
445 |
|
to each other using path anchors. (See CM manual for explanation.) |
446 |
|
|
447 |
|
Updated CM documentation. |
448 |
|
|
449 |
|
Fixed some bugs in CM. |
450 |
|
|
451 |
|
Implemented "proxy" libraries (= syntactic sugar for CM). |
452 |
|
|
453 |
|
Added "-quiet" option to makeml and changed runtime system accordingly. |
454 |
|
|
455 |
|
Added cleanup handler for exportML to reset timers and compiler stats. |
456 |
|
|
457 |
|
---------------------------------------------------------------------- |
458 |
|
Name: Lal George |
459 |
|
Date: 2000/12/22 22:22:58 EST 2000 |
460 |
|
Tag: Release_110_32 |
461 |
|
Description: |
462 |
|
|
463 |
|
Infinite precision used throughout MLRISC. |
464 |
|
see MLRISC/mltree/machine-int.sig |
465 |
|
|
466 |
|
---------------------------------------------------------------------- |
467 |
|
Name: Matthias Blume |
468 |
|
Date: 2000/12/22 23:16:00 JST |
469 |
|
Tag: blume-20001222-warn |
470 |
|
Description: |
471 |
|
|
472 |
|
Corrected wording and formatting of some CM warning message which I |
473 |
|
broke in my previous patch. |
474 |
|
|
475 |
|
---------------------------------------------------------------------- |
476 |
|
Name: Matthias Blume |
477 |
|
Date: 2000/12/22 21:20:00 JST |
478 |
|
Tag: blume-20001222-anchorenv |
479 |
|
Description: |
480 |
|
|
481 |
|
Fixed CM's handling of anchor environments in connection with CMB.make. |
482 |
|
|
483 |
|
---------------------------------------------------------------------- |
484 |
|
Name: Matthias Blume |
485 |
|
Date: 2000/12/22 13:15:00 JST |
486 |
|
Tag: blume-20001222-cleanup |
487 |
|
Description: |
488 |
|
|
489 |
|
Removed src/cm/ffi which does not (and did not) belong here. |
490 |
|
|
491 |
|
---------------------------------------------------------------------- |
492 |
|
Name: Matthias Blume |
493 |
|
Date: 2000/12/21 23:55:00 JST |
494 |
|
Tag: blume-20001221-exn |
495 |
|
Description: |
496 |
|
|
497 |
|
Probably most important: CM no longer silently swallows all exceptions |
498 |
|
in the compiler. |
499 |
|
Plus: some other minor CM changes. For example, CM now reports some |
500 |
|
sizes for generated binfiles (code, data, envpickle, lambdapickle). |
501 |
|
|
502 |
---------------------------------------------------------------------- |
---------------------------------------------------------------------- |
503 |
Name: Matthias Blume |
Name: Matthias Blume |
504 |
Date: 2000/12/15 00:01:05 JST |
Date: 2000/12/15 00:01:05 JST |