SCM Repository
Diff of /sml/trunk/HISTORY
Parent Directory
|
Revision Log
|
Patch
revision 659, Mon Jun 12 07:37:22 2000 UTC | revision 1186, Fri Apr 12 17:54:31 2002 UTC | |
---|---|---|
# | Line 8 | Line 8 |
8 | The form of an entry should be: | The form of an entry should be: |
9 | ||
10 | Name: | Name: |
11 | Date: | Date: yyyy/mm/dd |
12 | Tag: <post-commit CVS tag> | Tag: <post-commit CVS tag> |
13 | Description: | Description: |
14 | ||
15 | ---------------------------------------------------------------------- | |
16 | Name: Matthias Blume | |
17 | Date: 2002/04/12 13:55:00 EDT | |
18 | Tag: blume-20020412-assyntax | |
19 | Description: | |
20 | ||
21 | 1. Grabbed newer assyntax.h from the XFree86 project. | |
22 | 2. Fiddled with how to compile X86.prim.asm without warnings. | |
23 | 3. (Very) Minor cleanup in CM. | |
24 | ||
25 | ---------------------------------------------------------------------- | |
26 | Name: Matthias Blume | |
27 | Date: 2002/04/01 (no joke!) 17:07:00 EST | |
28 | Tag: blume-20020401-x86div | |
29 | Description: | |
30 | ||
31 | Added full support for div/mod/rem/quot on the x86, using the machine | |
32 | instruction's two results (without clumsily recomputing the remainder) | |
33 | directly where appropriate. | |
34 | ||
35 | Some more extensive power-of-two support was added to the x86 instruction | |
36 | selector (avoiding expensive divs, mods, and muls where they can be | |
37 | replaced with cheaper shifts and masks). However, this sort of thing | |
38 | ought to be done earlier, e.g., within the CPS optimizer so that | |
39 | all architectures benefit from it. | |
40 | ||
41 | The compiler compiles to a fixed point, but changes might be somewhat | |
42 | fragile nevertheless. Please, report any strange things that you might | |
43 | see wrt. div/mod/quot/rem... | |
44 | ||
45 | ---------------------------------------------------------------------- | |
46 | Name: Matthias Blume | |
47 | Date: 2002/03/29 17:22:00 | |
48 | Tag: blume-20020329-div | |
49 | Description: | |
50 | ||
51 | Fixed my broken div/mod logic. Unfortunately, this means that the | |
52 | inline code for div/mod now has one more comparison than before. | |
53 | Fast paths (quotient > 0 or remainder = 0) are not affected, though. | |
54 | The problem was with quotient = 0, because that alone does not tell | |
55 | us which way the rounding went. One then has to look at whether | |
56 | remainder and divisor have the same sign... :( | |
57 | ||
58 | Anyway, I replaced the bootfiles with fresh ones... | |
59 | ||
60 | ---------------------------------------------------------------------- | |
61 | Name: Matthias Blume | |
62 | Date: 2002/03/29 14:10:00 EST | |
63 | Tag: blume-20020329-inlprims | |
64 | Description: | |
65 | ||
66 | NEW BOOTFILES!!! Version number bumped to 110.39.3. | |
67 | ||
68 | Primops have changed. This means that the bin/boot-file formats have | |
69 | changed as well. | |
70 | ||
71 | To make sure that there is no confusion, I made a new version. | |
72 | ||
73 | ||
74 | CHANGES: | |
75 | ||
76 | * removed REMT from mltree (remainder should never overflow). | |
77 | ||
78 | * added primops to deal with divisions of all flavors to the frontend | |
79 | ||
80 | * handled these primops all the way through so they map to their respective | |
81 | MLRISC support | |
82 | ||
83 | * used these primops in the implementation of Int, Int32, Word, Word32 | |
84 | ||
85 | * removed INLDIV, INLMOD, and INLREM as they are no longer necessary | |
86 | ||
87 | * parameterized INLMIN, INLMAX, and INLABS by a numkind | |
88 | ||
89 | * translate.sml now deals with all flavors of INL{MIN,MAX,ABS}, including | |
90 | floating point | |
91 | ||
92 | * used INL{MIN,MAX,ABS} in the implementation of Int, Int32, Word, Word32, | |
93 | and Real (but Real.abs maps to a separate floating-point-only primop) | |
94 | ||
95 | ||
96 | TODO items: | |
97 | ||
98 | * Hacked Alpha32 instruction selection, disabling the selection of REMx | |
99 | instructions because the machine instruction encoder cannot handle | |
100 | them. (Hppa, PPC, and Sparc instruction selection did not handle | |
101 | REM in the first place, and REM is supported by the x86 machine coder.) | |
102 | ||
103 | * Handle DIV and MOD with DIV_TO_NEGINF directly in the x86 instruction | |
104 | selection phase. (The two can be streamlined because the hardware | |
105 | delivers both quotient and remainder at the same time anyway.) | |
106 | ||
107 | * Think about what to do with "valOf(Int32.minInt) div ~1" and friends. | |
108 | (Currently the behavior is inconsistent both across architectures and | |
109 | wrt. the draft Basis spec.) | |
110 | ||
111 | * Word8 should eventually be handled natively, too. | |
112 | ||
113 | * There seems to be one serious bug in mltree-gen.sml. It appears, though, | |
114 | as if there currently is no execution path that could trigger it in | |
115 | SML/NJ. (The assumptions underlying functions arith and promotable do not | |
116 | hold for things like multiplication and division.) | |
117 | ||
118 | ---------------------------------------------------------------------- | |
119 | Name: Matthias Blume | |
120 | Date: 2002/03/27 16:27:00 EST | |
121 | Tag: blume-20020327-mlrisc-divisions | |
122 | Description: | |
123 | ||
124 | Added support for all four division operations (ML's div, mod, quot, | |
125 | and rem) to MLRISC. In the course of doing so, I also rationalized | |
126 | the naming (no more annoying switch-around of DIV and QUOT), by | |
127 | parameterizing the operation by div_rounding_mode (which can be either | |
128 | DIV_TO_ZERO or DIV_TO_NEGINF). | |
129 | ||
130 | The generic MLTreeGen functor takes care of compiling all four | |
131 | operations down to only round-to-zero div. | |
132 | ||
133 | Missing pieces: | |
134 | ||
135 | * Doing something smarter than relying on MLTreeGen on architectures | |
136 | like, e.g., the x86 where hardware division delivers both quotient and | |
137 | remainder at the same time. With this, the implementation of the | |
138 | round-to-neginf operations could be further streamlined. | |
139 | ||
140 | * Remove inlining support for div/mod/rem from the frontend and replace it | |
141 | with primops that get carried through to the backend. Do this for all | |
142 | int and word types. | |
143 | ||
144 | ---------------------------------------------------------------------- | |
145 | Name: Matthias Blume | |
146 | Date: 2002/03/25 17:25:00 EST | |
147 | Tag: blume-20020325-divmod | |
148 | Description: | |
149 | ||
150 | I improved (hopefully without breaking them) the implementation of Int.div, | |
151 | Int.mod, and Int.rem. For this, the code in translate.sml now takes | |
152 | advantage of the following observations: | |
153 | ||
154 | Let q = x quot y r = x rem y | |
155 | d = x div y m = x mod y | |
156 | ||
157 | where "quot" is the round-to-zero version of integer division that | |
158 | hardware usually provides. Then we have: | |
159 | ||
160 | r = x - q * y where neither the * nor the - will overflow | |
161 | d = if q >= 0 orelse x = q * y then q else q - 1 | |
162 | where neither the * nor the - will overflow | |
163 | m = if q >= 0 orelse r = 0 then r else r + y | |
164 | where the + will not overflow | |
165 | ||
166 | This results in substantial simplification of the generated code. | |
167 | The following table shows the number of CFG nodes and edges generated | |
168 | for | |
169 | fun f (x, y) = x OPER y | |
170 | (* with OPER \in div, mod, quot, rem *) | |
171 | ||
172 | ||
173 | OPER | nodes(old) | edges(old) | nodes(new) | edges(new) | |
174 | -------------------------------------------------------- | |
175 | div | 24 | 39 | 12 | 16 | |
176 | mod | 41 | 71 | 12 | 16 | |
177 | quot | 8 | 10 | 8 | 10 | |
178 | rem | 10 | 14 | 8 | 10 | |
179 | ||
180 | ||
181 | ---------------------------------------------------------------------- | |
182 | Name: Matthias Blume | |
183 | Date: 2002/03/25 22:06:00 EST | |
184 | Tag: blume-20020325-cprotobug | |
185 | Description: | |
186 | ||
187 | Fixed a bug in cproto (c prototype decoder). | |
188 | ||
189 | ---------------------------------------------------------------------- | |
190 | Name: Matthias Blume | |
191 | Date: 2002/03/25 16:00:00 EST | |
192 | Tag: blume-20020325-raw-primops | |
193 | Description: | |
194 | ||
195 | I did some cleanup to Allen's new primop code and | |
196 | replaced yesterday's bootfiles with new ones. | |
197 | (But they are stored in the same place.) | |
198 | ||
199 | ---------------------------------------------------------------------- | |
200 | Name: Matthias Blume | |
201 | Date: 2002/03/24 22:40:00 EST | |
202 | Tag: blume-20020324-bootfiles | |
203 | Description: | |
204 | ||
205 | Made the bootfiles that Allen asked for. | |
206 | ||
207 | ---------------------------------------------------------------------- | |
208 | Name: Allen Leung | |
209 | Date: 2002/03/23 15:50:00 EST | |
210 | Tag: leunga-20020323-flint-cps-rcc-primops | |
211 | Description: | |
212 | ||
213 | 1. Changes to FLINT primops: | |
214 | ||
215 | (* make a call to a C-function; | |
216 | * The primop carries C function prototype information and specifies | |
217 | * which of its (ML-) arguments are floating point. C prototype | |
218 | * information is for use by the backend, ML information is for | |
219 | * use by the CPS converter. *) | |
220 | | RAW_CCALL of { c_proto: CTypes.c_proto, | |
221 | ml_args: ccall_type list, | |
222 | ml_res_opt: ccall_type option, | |
223 | reentrant : bool | |
224 | } option | |
225 | (* Allocate uninitialized storage on the heap. | |
226 | * The record is meant to hold short-lived C objects, i.e., they | |
227 | * are not ML pointers. With the tag, the representation is | |
228 | * the same as RECORD with tag tag_raw32 (sz=4), or tag_fblock (sz=8) | |
229 | *) | |
230 | | RAW_RECORD of {tag:bool,sz:int} | |
231 | and ccall_type = CCALL_INT32 | CCALL_REAL64 | CCALL_ML_PTR | |
232 | ||
233 | 2. These CPS primops are now overloaded: | |
234 | ||
235 | rawload of {kind:numkind} | |
236 | rawstore of {kind:numkind} | |
237 | ||
238 | The one argument form is: | |
239 | ||
240 | rawload {kind} address | |
241 | ||
242 | The two argument form is: | |
243 | ||
244 | rawload {kind} [ml object, byte-offset] | |
245 | ||
246 | 3. RAW_CCALL/RCC now takes two extra arguments: | |
247 | ||
248 | a. The first is whether the C call is reentrant, i.e., whether | |
249 | ML state should be saved and restored. | |
250 | b. The second argument is a string argument specifying the name of | |
251 | library and the C function. | |
252 | ||
253 | These things are currently not handled in the code generator, yet. | |
254 | ||
255 | 4. In CProto, | |
256 | ||
257 | An encoding type of "bool" means "ml object" and is mapped into | |
258 | C prototype of PTR. Note that "bool" is different than "string", | |
259 | even though "string" is also mapped into PTR, because "bool" | |
260 | is assigned an CPS type of BOGt, while "string" is assigned INT32t. | |
261 | ||
262 | 5. Pickler/unpicker | |
263 | ||
264 | Changed to handle RAW_RECORD and newest RAW_CCALL | |
265 | ||
266 | 6. MLRiscGen, | |
267 | ||
268 | 1. Changed to handle the new rawload/rawstore/rawrecord operators. | |
269 | 2. Code for handling C Calls has been moved to a new module CPSCCalls, | |
270 | in the file CodeGen/cpscompile/cps-c-calls.sml | |
271 | ||
272 | 7. Added the conditional move operator | |
273 | ||
274 | condmove of branch | |
275 | ||
276 | to cps. Generation of this is still buggy so it is currently | |
277 | disabled. | |
278 | ||
279 | ---------------------------------------------------------------------- | |
280 | Name: Lal George | |
281 | Date: 2002/03/22 14:18:25 EST | |
282 | Tag: george-20020322-cps-branch-prob | |
283 | Description: | |
284 | ||
285 | Implemented the Ball-Larus branch prediction-heuristics, and | |
286 | incorporated graphical viewers for control flow graphs. | |
287 | ||
288 | Ball-Larus Heuristics: | |
289 | --------------------- | |
290 | See the file compiler/CodeGen/cpscompile/cpsBranchProb.sml. | |
291 | ||
292 | By design it uses the Dempster-Shafer theory for combining | |
293 | probabilities. For example, in the function: | |
294 | ||
295 | fun f(n,acc) = if n = 0 then acc else f(n-1, n*acc) | |
296 | ||
297 | the ball-larus heuristics predicts that the n=0 is unlikely | |
298 | (OH-heuristic), and the 'then' branch is unlikely because of the | |
299 | RH-heuristic -- giving the 'then' branch an even lower combined | |
300 | probability using the Dempster-Shafer theory. | |
301 | ||
302 | Finally, John Reppy's loop analysis in MLRISC, further lowers the | |
303 | probability of the 'then' branch because of the loop in the else | |
304 | branch. | |
305 | ||
306 | ||
307 | Graphical Viewing: | |
308 | ------------------ | |
309 | I merely plugged in Allen's graphical viewers into the compiler. The | |
310 | additional code is not much. At the top level, saying: | |
311 | ||
312 | Control.MLRISC.getFlag "cfg-graphical-view" := true; | |
313 | ||
314 | will display the graphical view of the control flow graph just before | |
315 | back-patching. daVinci must be in your path for this to work. If | |
316 | daVinci is not available, then the default viewer can be changed | |
317 | using: | |
318 | ||
319 | Control.MLRISC.getString "viewer" | |
320 | ||
321 | which can be set to "dot" or "vcg" for the corresponding viewers. Of | |
322 | course, these viewers must be in your path. | |
323 | ||
324 | The above will display the compilation unit at the level of clusters, | |
325 | many of which are small, boring, and un-interesting. Also setting: | |
326 | ||
327 | Control.MLRISC.getInt "cfg-graphical-view_size" | |
328 | ||
329 | will display clusters that are larger than the value set by the above. | |
330 | ||
331 | ||
332 | ---------------------------------------------------------------------- | |
333 | Name: Matthias Blume | |
334 | Date: 2002/03/21 22:20:00 EST | |
335 | Tag: blume-20020321-kmp-bugfix | |
336 | Description: | |
337 | ||
338 | Changed the interface to the KMP routine in PreString and fixed | |
339 | a minor bug in one place where it was used. | |
340 | ||
341 | ---------------------------------------------------------------------- | |
342 | Name: Allen Leung | |
343 | Date: 2002/03/21 20:30:00 EST | |
344 | Tag: leunga-20020321-cfg | |
345 | Description: | |
346 | ||
347 | Fixed a potential problem in cfg edge splitting. | |
348 | ||
349 | ---------------------------------------------------------------------- | |
350 | Name: Allen Leung | |
351 | Date: 2002/03/21 17:15:00 EST | |
352 | Tag: leunga-20020321-x86-fp-cfg | |
353 | Description: | |
354 | ||
355 | 1. Recoded the buggy parts of x86-fp. | |
356 | ||
357 | a. All the block reordering code has been removed. | |
358 | We now depend on the block placement phases to do this work. | |
359 | ||
360 | b. Critical edge splitting code has been simplified and moved into the | |
361 | CFG modules, as where they belong. | |
362 | ||
363 | Both of these were quite buggy and complex. The code is now much, much | |
364 | simpler. | |
365 | ||
366 | 2. X86 backend. | |
367 | ||
368 | a. Added instructions for 64-bit support. Instruction selection for | |
369 | 64-bit has not been committed, however, since that | |
370 | requires changes to MLTREE which haven't been approved by | |
371 | Lal and John. | |
372 | ||
373 | b. Added support for FUCOMI and FUCOMIP when generating code for | |
374 | PentiumPro and above. We only generate these instructions in | |
375 | the fast-fp mode. | |
376 | ||
377 | c. Added cases for JP and JNP in X86FreqProps. | |
378 | ||
379 | 3. CFG | |
380 | ||
381 | CFG now has a bunch of methods for edge splitting and merging. | |
382 | ||
383 | 4. Machine description. | |
384 | ||
385 | John's simplification of MLTREE_BASIS.fcond broke a few machine | |
386 | description things: | |
387 | ||
388 | rtl-build.{sig,sml} and hppa.mdl fixed. | |
389 | ||
390 | NOTE: the machine description stuff in the repository is still broken. | |
391 | Again, I can't put my fixes in because that involves | |
392 | changes to MLTREE. | |
393 | ||
394 | ---------------------------------------------------------------------- | |
395 | Name: Matthias Blume | |
396 | Date: 2002/03/20 15:55:00 EST | |
397 | Tag: blume-20020320-kmp | |
398 | Description: | |
399 | ||
400 | Implemented Knuth-Morris-Pratt string matching in PreString and used | |
401 | it for String.isSubstring, Substring.isSubstring, and | |
402 | Substring.position. | |
403 | ||
404 | (Might need some stress-testing. Simple examples worked fine.) | |
405 | ||
406 | ---------------------------------------------------------------------- | |
407 | Name: Matthias Blume | |
408 | Date: 2002/03/19 16:37:00 EST | |
409 | Tag: blume-20020319-witnesses | |
410 | Description: | |
411 | ||
412 | Added a structure C.W and functions convert/Ptr.convert to ml-nlffi-lib. | |
413 | ||
414 | This implements a generic mechanism for changing constness qualifiers | |
415 | anywhere within big C types without resorting to outright "casts". | |
416 | (So far, functions such as C.rw/C.ro or C.Ptr.rw/C.Ptr.ro only let you | |
417 | modify the constness at the outermost level.) | |
418 | The implementation of "convert" is based on the idea of "witness" | |
419 | values -- values that are not used by the operation but whose types | |
420 | "testify" to their applicability. On the implementation side, "convert" | |
421 | is simply a projection (returning its second curried argument). With | |
422 | cross-module inlining, it should not result in any machine code being | |
423 | generated. | |
424 | ||
425 | ---------------------------------------------------------------------- | |
426 | Name: Matthias Blume | |
427 | Date: 2002/03/15 16:40:00 EST | |
428 | Tag: blume-20020315-basis | |
429 | Description: | |
430 | ||
431 | Provided (preliminary?) implementations for | |
432 | ||
433 | {String,Substring}.{concatWith,isSuffix,isSubstring} | |
434 | ||
435 | and | |
436 | ||
437 | Substring.full | |
438 | ||
439 | Those are in the Basis spec but they were missing in SML/NJ. | |
440 | ||
441 | ---------------------------------------------------------------------- | |
442 | Name: Matthias Blume | |
443 | Date: 2002/03/14 21:30:00 EST | |
444 | Tag: blume-20020314-controls | |
445 | Description: | |
446 | ||
447 | Controls: | |
448 | --------- | |
449 | ||
450 | 1. Factored out the recently-added Controls : CONTROLS stuff and put | |
451 | it into its own library $/controls-lib.cm. The source tree for | |
452 | this is under src/smlnj-lib/Controls. | |
453 | ||
454 | 2. Changed the names of types and functions in this interface, so they | |
455 | make a bit more "sense": | |
456 | ||
457 | module -> registry | |
458 | 'a registry -> 'a group | |
459 | ||
460 | 3. The interface now deals in ref cells only. The getter/setter interface | |
461 | is (mostly) gone. | |
462 | ||
463 | 4. Added a function that lets one register an already-existing ref cell. | |
464 | ||
465 | 5. Made the corresponding modifications to the rest of the code so that | |
466 | everything compiles again. | |
467 | ||
468 | 6. Changed the implementation of Controls.MLRISC back to something closer | |
469 | to the original. In particular, this module (and therefore MLRISC) | |
470 | does not depend on Controls. There now is some link-time code in | |
471 | int-sys.sml that registers the MLRISC controls with the Controls | |
472 | module. | |
473 | ||
474 | CM: | |
475 | --- | |
476 | ||
477 | * One can now specify the lambda-split aggressiveness in init.cmi. | |
478 | ||
479 | ---------------------------------------------------------------------- | |
480 | Name: Allen Leung | |
481 | Date: 2002/03/13 17:30:00 EST | |
482 | Tag: leunga-20020313-x86-fp-unary | |
483 | Description: | |
484 | ||
485 | Bug fix for: | |
486 | ||
487 | > leunga@weaselbane:~/Yale/tmp/sml-dist{21} bin/sml | |
488 | > Standard ML of New Jersey v110.39.1 [FLINT v1.5], March 08, 2002 | |
489 | > - fun f(x,(y,z)) = Real.~ y; | |
490 | > [autoloading] | |
491 | > [autoloading done] | |
492 | > fchsl (%eax), 184(%esp) | |
493 | > Error: MLRisc bug: X86MCEmitter.emitInstr | |
494 | > | |
495 | > uncaught exception Error | |
496 | > raised at: ../MLRISC/control/mlriscErrormsg.sml:16.14-16.19 | |
497 | ||
498 | The problem was that the code generator did not generate any fp registers | |
499 | in this case, and the ra didn't know that it needed to run the X86FP phase to | |
500 | translate the pseudo fp instruction. This only happened with unary fp | |
501 | operators in certain situations. | |
502 | ||
503 | ---------------------------------------------------------------------- | |
504 | Name: Matthias Blume | |
505 | Date: 2002/03/13 14:00:00 EST | |
506 | Tag: blume-20020313-overload-etc | |
507 | Description: | |
508 | ||
509 | 1. Added _overload as a synonym for overload for backward compatibility. | |
510 | (Control.overloadKW must be true for either version to be accepted.) | |
511 | ||
512 | 2. Fixed bug in install script that caused more things to be installed | |
513 | than what was requested in config/targets. | |
514 | ||
515 | 3. Made CM aware of the (_)overload construct so that autoloading | |
516 | works. | |
517 | ||
518 | ---------------------------------------------------------------------- | |
519 | Name: Matthias Blume | |
520 | Date: 2002/03/12 22:03:00 EST | |
521 | Tag: blume-20020312-url | |
522 | Description: | |
523 | ||
524 | Forgot to update BOOT and srcarchiveurl. | |
525 | ||
526 | ---------------------------------------------------------------------- | |
527 | Name: Matthias Blume | |
528 | Date: 2002/03/12 17:30:00 EST | |
529 | Tag: blume-20020312-version110392 | |
530 | Description: | |
531 | ||
532 | Yet another version number bump (because of small changes to the | |
533 | binfile format). Version number is now 110.39.2. NEW BOOTFILES! | |
534 | ||
535 | Changes: | |
536 | ||
537 | The new pid generation scheme described a few weeks ago was overly | |
538 | complicated. I implemented a new mechanism that is simpler and | |
539 | provides a bit more "stability": Once CM has seen a compilation | |
540 | unit, it keeps its identity constant (as long as you do not delete | |
541 | those crucial CM/GUID/* files). This means that when you change | |
542 | an interface, compile, then go back to the old interface, and | |
543 | compile again, you arrive at the original pid. | |
544 | ||
545 | There now also is a mechanism that instructs CM to use the plain | |
546 | environment hash as a module's pid (effectively making its GUID | |
547 | the empty string). For this, "noguid" must be specified as an | |
548 | option to the .sml file in question within its .cm file. | |
549 | This is most useful for code that is being generated by tools such | |
550 | as ml-nlffigen (because during development programmers tend to | |
551 | erase the tool's entire output directory tree including CM's cached | |
552 | GUIDs). "noguid" is somewhat dangerous (since it can be used to locally | |
553 | revert to the old, broken behavior of SML/NJ, but in specific cases | |
554 | where there is no danger of interface confusion, its use is ok | |
555 | (I think). | |
556 | ||
557 | ml-nlffigen by default generates "noguid" annotations. They can be | |
558 | turned off by specifying -guid in its command line. | |
559 | ||
560 | ---------------------------------------------------------------------- | |
561 | Name: Lal George | |
562 | Date: 2002/03/12 12 14:42:36 EST | |
563 | Tag: george-20020312-frequency-computation | |
564 | Description: | |
565 | ||
566 | Integrated jump chaining and static block frequency into the | |
567 | compiler. More details and numbers later. | |
568 | ||
569 | ---------------------------------------------------------------------- | |
570 | Name: Lal George | |
571 | Date: 2002/03/11 11 22:38:53 EST | |
572 | Tag: george-20020311-jump-chain-elim | |
573 | Description: | |
574 | ||
575 | Tested the jump chain elimination on all architectures (except the | |
576 | hppa). This is on by default right now and is profitable for the | |
577 | alpha and x86, however, it may not be profitable for the sparc and ppc | |
578 | when compiling the compiler. | |
579 | ||
580 | The gc test will typically jump to a label at the end of the cluster, | |
581 | where there is another jump to an external cluster containing the actual | |
582 | code to invoke gc. This is to allow factoring of common gc invocation | |
583 | sequences. That is to say, we generate: | |
584 | ||
585 | f: | |
586 | testgc | |
587 | ja L1 % jump if above to L1 | |
588 | ||
589 | L1: | |
590 | jmp L2 | |
591 | ||
592 | ||
593 | After jump chain elimination the 'ja L1' instructions is converted to | |
594 | 'ja L2'. On the sparc and ppc, many of the 'ja L2' instructions may end | |
595 | up being implemented in their long form (if L2 is far away) using: | |
596 | ||
597 | jbe L3 % jump if below or equal to L3 | |
598 | jmp L2 | |
599 | L3: | |
600 | ... | |
601 | ||
602 | ||
603 | For large compilation units L2 may be far away. | |
604 | ||
605 | ||
606 | ---------------------------------------------------------------------- | |
607 | Name: Matthias Blume | |
608 | Date: 2002/03/11 13:30:00 EST | |
609 | Tag: blume-20020311-mltreeeval | |
610 | Description: | |
611 | ||
612 | A functor parameter was missing. | |
613 | ||
614 | ---------------------------------------------------------------------- | |
615 | Name: Allen Leung | |
616 | Date: 2002/03/11 10:30:00 EST | |
617 | Tag: leunga-20020311-runtime-string0 | |
618 | Description: | |
619 | ||
620 | The representation of the empty string now points to a | |
621 | legal null terminated C string instead of unit. It is now possible | |
622 | to convert an ML string into C string with InlineT.CharVector.getData. | |
623 | This compiles into one single machine instruction. | |
624 | ||
625 | ---------------------------------------------------------------------- | |
626 | Name: Allen Leung | |
627 | Date: 2002/03/10 23:55:00 EST | |
628 | Tag: leunga-20020310-x86-call | |
629 | Description: | |
630 | ||
631 | Added machine generation for CALL instruction (relative displacement mode) | |
632 | ||
633 | ---------------------------------------------------------------------- | |
634 | Name: Matthias Blume | |
635 | Date: 2002/03/08 16:05:00 | |
636 | Tag: blume-20020308-entrypoints | |
637 | Description: | |
638 | ||
639 | Version number bumped to 110.39.1. NEW BOOTFILES! | |
640 | ||
641 | Entrypoints: non-zero offset into a code object where execution should begin. | |
642 | ||
643 | - Added the notion of an entrypoint to CodeObj. | |
644 | - Added reading/writing of entrypoint info to Binfile. | |
645 | - Made runtime system bootloader aware of entrypoints. | |
646 | - Use the address of the label of the first function given to mlriscGen | |
647 | as the entrypoint. This address is currently always 0, but it will | |
648 | not be 0 once we turn on block placement. | |
649 | - Removed the linkage cluster code (which was The Other Way(tm) of dealing | |
650 | with entry points) from mlriscGen. | |
651 | ||
652 | ---------------------------------------------------------------------- | |
653 | Name: Allen Leung | |
654 | Date: 2002/03/07 20:45:00 EST | |
655 | Tag: leunga-20020307-x86-cmov | |
656 | Description: | |
657 | ||
658 | Bug fixes for CMOVcc on x86. | |
659 | ||
660 | 1. Added machine code generation for CMOVcc | |
661 | 2. CMOVcc is now generated in preference over SETcc on PentiumPro or above. | |
662 | 3. CMOVcc cannot have an immediate operand as argument. | |
663 | ||
664 | ---------------------------------------------------------------------- | |
665 | Name: Matthias Blume | |
666 | Date: 2002/03/07 16:15:00 EST | |
667 | Tag: blume-20020307-controls | |
668 | Description: | |
669 | ||
670 | This is a very large but mostly boring patch which makes (almost) | |
671 | every tuneable compiler knob (i.e., pretty much everything under | |
672 | Control.* plus a few other things) configurable via both the command | |
673 | line and environment variables in the style CM did its configuration | |
674 | until now. | |
675 | ||
676 | Try starting sml with '-h' (or, if you are brave, '-H') | |
677 | ||
678 | To this end, I added a structure Controls : CONTROLS to smlnj-lib.cm which | |
679 | implements the underlying generic mechanism. | |
680 | ||
681 | The interface to some of the existing such facilities has changed somewhat. | |
682 | For example, the MLRiscControl module now provides mkFoo instead of getFoo. | |
683 | (The getFoo interface is still there for backward-compatibility, but its | |
684 | use is deprecated.) | |
685 | ||
686 | The ml-build script passes -Cxxx=yyy command-line arguments through so | |
687 | that one can now twiddle the compiler settings when using this "batch" | |
688 | compiler. | |
689 | ||
690 | TODO items: | |
691 | ||
692 | We should go through and throw out all controls that are no longer | |
693 | connected to anything. Moreover, we should go through and provide | |
694 | meaningful (and correct!) documentation strings for those controls | |
695 | that still are connected. | |
696 | ||
697 | Currently, multiple calls to Controls.new are accepted (only the first | |
698 | has any effect). Eventually we should make sure that every control | |
699 | is being made (via Controls.new) exactly once. Future access can then | |
700 | be done using Controls.acc. | |
701 | ||
702 | Finally, it would probably be a good idea to use the getter-setter | |
703 | interface to controls rather than ref cells. For the time being, both | |
704 | styles are provided by the Controls module, but getter-setter pairs are | |
705 | better if thread-safety is of any concern because they can be wrapped. | |
706 | ||
707 | ***************************************** | |
708 | ||
709 | One bug fix: The function blockPlacement in three of the MLRISC | |
710 | backpatch files used to be hard-wired to one of two possibilities at | |
711 | link time (according to the value of the placementFlag). But (I | |
712 | think) it should rather sense the flag every time. | |
713 | ||
714 | ***************************************** | |
715 | ||
716 | Other assorted changes (by other people who did not supply a HISTORY entry): | |
717 | ||
718 | 1. the cross-module inliner now works much better (Monnier) | |
719 | 2. representation of weights, frequencies, and probabilities in MLRISC | |
720 | changed in preparation of using those for weighted block placement | |
721 | (Reppy, George) | |
722 | ||
723 | ---------------------------------------------------------------------- | |
724 | Name: Lal George | |
725 | Date: 2002/03/07 14:44:24 EST 2002 | |
726 | Tag: george-20020307-weighted-block-placement | |
727 | ||
728 | Tested the weighted block placement optimization on all architectures | |
729 | (except the hppa) using AMPL to generate the block and edge frequencies. | |
730 | Changes were required in the machine properties to correctly | |
731 | categorize trap instructions. There is an MLRISC flag | |
732 | "weighted-block-placement" that can be used to enable weighted block | |
733 | placement, but this will be ineffective without block/edge | |
734 | frequencies (coming soon). | |
735 | ||
736 | ||
737 | ---------------------------------------------------------------------- | |
738 | Name: Lal George | |
739 | Date: 2002/03/05 17:24:48 EST | |
740 | Tag: george-20020305-linkage-cluster | |
741 | ||
742 | In order to support the block placement optimization, a new cluster | |
743 | is generated as the very first cluster (called the linkage cluster). | |
744 | It contains a single jump to the 'real' entry point for the compilation | |
745 | unit. Block placement has no effect on the linkage cluster itself, but | |
746 | all the other clusters have full freedom in the manner in which they | |
747 | reorder blocks or functions. | |
748 | ||
749 | On the x86 the typical linkage code that is generated is: | |
750 | ---------------------- | |
751 | .align 2 | |
752 | L0: | |
753 | addl $L1-L0, 72(%esp) | |
754 | jmp L1 | |
755 | ||
756 | ||
757 | .align 2 | |
758 | L1: | |
759 | ---------------------- | |
760 | ||
761 | 72(%esp) is the memory location for the stdlink register. This | |
762 | must contain the address of the CPS function being called. In the | |
763 | above example, it contains the address of L0; before | |
764 | calling L1 (the real entry point for the compilation unit), it | |
765 | must contain the address for L1, and hence | |
766 | ||
767 | addl $L1-L0, 72(%esp) | |
768 | ||
769 | I have tested this on all architectures except the hppa.The increase | |
770 | in code size is of course negligible | |
771 | ||
772 | ---------------------------------------------------------------------- | |
773 | Name: Allen Leung | |
774 | Date: 2002/03/03 13:20:00 EST | |
775 | Tag: leunga-20020303-mlrisc-tools | |
776 | ||
777 | Added #[ ... ] expressions to mlrisc tools | |
778 | ||
779 | ---------------------------------------------------------------------- | |
780 | Name: Matthias Blume | |
781 | Date: 2002/02/27 12:29:00 EST | |
782 | Tag: blume-20020227-cdebug | |
783 | Description: | |
784 | ||
785 | - made types in structure C and C_Debug to be equal | |
786 | - got rid of code duplication (c-int.sml vs. c-int-debug.sml) | |
787 | - there no longer is a C_Int_Debug (C_Debug is directly derived from C) | |
788 | ||
789 | ---------------------------------------------------------------------- | |
790 | Name: Matthias Blume | |
791 | Date: 2002/02/26 12:00:00 EST | |
792 | Tag: blume-20020226-ffi | |
793 | Description: | |
794 | ||
795 | 1. Fixed a minor bug in CM's "noweb" tool: | |
796 | If numbering is turned off, then truly don't number (i.e., do not | |
797 | supply the -L option to noweb). The previous behavior was to supply | |
798 | -L'' -- which caused noweb to use the "default" line numbering scheme. | |
799 | Thanks to Chris Richards for pointing this out (and supplying the fix). | |
800 | ||
801 | 2. Once again, I reworked some aspects of the FFI: | |
802 | ||
803 | A. The incomplete/complete type business: | |
804 | ||
805 | - Signatures POINTER_TO_INCOMPLETE_TYPE and accompanying functors are | |
806 | gone! | |
807 | - ML types representing an incomplete type are now *equal* to | |
808 | ML types representing their corresponding complete types (just like | |
809 | in C). This is still safe because ml-nlffigen will not generate | |
810 | RTTI for incomplete types, nor will it generate functions that | |
811 | require access to such RTTI. But when ML code generated from both | |
812 | incomplete and complete versions of the C type meet, the ML types | |
813 | are trivially interoperable. | |
814 | ||
815 | NOTE: These changes restore the full generality of the translation | |
816 | (which was previously lost when I eliminated functorization)! | |
817 | ||
818 | B. Enum types: | |
819 | ||
820 | - Structure C now has a type constructor "enum" that is similar to | |
821 | how the "su" constructor works. However, "enum" is not a phantom | |
822 | type because each "T enum" has values (and is isomorphic to | |
823 | MLRep.Signed.int). | |
824 | - There are generic access operations for enum objects (using | |
825 | MLRep.Signed.int). | |
826 | - ml-nlffigen will generate a structure E_foo for each "enum foo". | |
827 | * The structure contains the definition of type "mlrep" (the ML-side | |
828 | representation type of the enum). Normally, mlrep is the same | |
829 | as "MLRep.Signed.int", but if ml-nlffigen was invoked with "-ec", | |
830 | then mlrep will be defined as a datatype -- thus facilitating | |
831 | pattern matching on mlrep values. | |
832 | ("-ec" will be suppressed if there are duplicate values in an | |
833 | enumeration.) | |
834 | * Constructors ("-ec") or values (no "-ec") e_xxx of type mlrep | |
835 | will be generated for each C enum constant xxx. | |
836 | * Conversion functions m2i and i2m convert between mlrep and | |
837 | MLRep.Signed.int. (Without "-ec", these functions are identities.) | |
838 | * Coversion functions c and ml convert between mlrep and "tag enum". | |
839 | * Access functions (get/set) fetch and store mlrep values. | |
840 | - By default (unless ml-nlffigen was invoked with "-nocollect"), unnamed | |
841 | enumerations are merged into one single enumeration represented by | |
842 | structure E_'. | |
843 | ||
844 | ---------------------------------------------------------------------- | |
845 | Name: Allen Leung | |
846 | Date: 2002/02/25 04:45:00 EST | |
847 | Tag: leunga-20020225-cps-spill | |
848 | ||
849 | This is a new implementation of the CPS spill phase. | |
850 | The new phase is in the new file compiler/CodeGen/cpscompile/spill-new.sml | |
851 | In case of problems, replace it with the old file spill.sml | |
852 | ||
853 | The current compiler runs into some serious performance problems when | |
854 | constructing a large record. This can happen when we try to compile a | |
855 | structure with many items. Even a very simple structure like the following | |
856 | makes the compiler slow down. | |
857 | ||
858 | structure Foo = struct | |
859 | val x_1 = 0w1 : Word32.int | |
860 | val x_2 = 0w2 : Word32.int | |
861 | val x_3 = 0w3 : Word32.int | |
862 | ... | |
863 | val x_N = 0wN : Word32.int | |
864 | end | |
865 | ||
866 | The following table shows the compile time, from N=1000 to N=4000, | |
867 | with the old compiler: | |
868 | ||
869 | N | |
870 | 1000 CPS 100 spill 0.04u 0.00s 0.00g | |
871 | MLRISC ra 0.06u 0.00s 0.05g | |
872 | (spills = 0 reloads = 0) | |
873 | TOTAL 0.63u 0.07s 0.21g | |
874 | ||
875 | 1100 CPS 100 spill 8.25u 0.32s 0.64g | |
876 | MLRISC ra 5.68u 0.59s 3.93g | |
877 | (spills = 0 reloads = 0) | |
878 | TOTAL 14.71u 0.99s 4.81g | |
879 | ||
880 | 1500 CPS 100 spill 58.55u 2.34s 1.74g | |
881 | MLRISC ra 5.54u 0.65s 3.91g | |
882 | (spills = 543 reloads = 1082) | |
883 | TOTAL 65.40u 3.13s 6.00g | |
884 | ||
885 | 2000 CPS 100 spill 126.69u 4.84s 3.08g | |
886 | MLRISC ra 0.80u 0.10s 0.55g | |
887 | (spills = 42 reloads = 84) | |
888 | TOTAL 129.42u 5.10s 4.13g | |
889 | ||
890 | 3000 CPS 100 spill 675.59u 19.03s 11.64g | |
891 | MLRISC ra 2.69u 0.27s 1.38g | |
892 | (spills = 62 reloads = 124) | |
893 | TOTAL 682.48u 19.61s 13.99g | |
894 | ||
895 | 4000 CPS 100 spill 2362.82u 56.28s 43.60g | |
896 | MLRISC ra 4.96u 0.27s 2.72g | |
897 | (spills = 85 reloads = 170) | |
898 | TOTAL 2375.26u 57.21s 48.00g | |
899 | ||
900 | As you can see the old cps spill module suffers from some serious | |
901 | performance problem. But since I cannot decipher the old code fully, | |
902 | instead of patching the problems up, I'm reimplementing it | |
903 | with a different algorithm. The new code is more modular, | |
904 | smaller when compiled, and substantially faster | |
905 | (O(n log n) time and O(n) space). Timing of the new spill module: | |
906 | ||
907 | 4000 CPS 100 spill 0.02u 0.00s 0.00g | |
908 | MLRISC ra 0.25u 0.02s 0.15g | |
909 | (spills=1 reloads=3) | |
910 | TOTAL 7.74u 0.34s 1.62g | |
911 | ||
912 | Implementation details: | |
913 | ||
914 | As far as I can tell, the purpose of the CPS spill module is to make sure the | |
915 | number of live variables at any program point (the bandwidth) | |
916 | does not exceed a certain limit, which is determined by the | |
917 | size of the spill area. | |
918 | ||
919 | When the bandwidth is too large, we decrease the register pressure by | |
920 | packing live variables into spill records. How we achieve this is | |
921 | completely different than what we did in the old code. | |
922 | ||
923 | First, there is something about the MLRiscGen code generator | |
924 | that we should be aware of: | |
925 | ||
926 | o MLRiscGen performs code motion! | |
927 | ||
928 | In particular, it will move floating point computations and | |
929 | address computations involving only the heap pointer to | |
930 | their use sites (if there is only a single use). | |
931 | What this means is that if we have a CPS record construction | |
932 | statement | |
933 | ||
934 | RECORD(k,vl,w,e) | |
935 | ||
936 | we should never count the new record address w as live if w | |
937 | has only one use (which is often the case). | |
938 | ||
939 | We should do something similar to floating point, but the transformation | |
940 | there is much more complex, so I won't deal with that. | |
941 | ||
942 | Secondly, there are now two new cps primops at our disposal: | |
943 | ||
944 | 1. rawrecord of record_kind option | |
945 | This pure operator allocates some uninitialized storage from the heap. | |
946 | There are two forms: | |
947 | ||
948 | rawrecord NONE [INT n] allocates a tagless record of length n | |
949 | rawrecord (SOME rk) [INT n] allocates a tagged record of length n | |
950 | and initializes the tag. | |
951 | ||
952 | 2. rawupdate of cty | |
953 | rawupdate cty (v,i,x) | |
954 | Assigns to x to the ith component of record v. | |
955 | The storelist is not updated. | |
956 | ||
957 | We use these new primops for both spilling and increment record construction. | |
958 | ||
959 | 1. Spilling. | |
960 | ||
961 | This is implemented with a linear scan algorithm (but generalized | |
962 | to trees). The algorithm will create a single spill record at the | |
963 | beginning of the cps function and use rawupdate to spill to it, | |
964 | and SELECT or SELp to reload from it. So both spills and reloads | |
965 | are fine-grain operations. In contrast, in the old algorithm | |
966 | "spills" have to be bundled together in records. | |
967 | ||
968 | Ideally, we should sink the spill record construction to where | |
969 | it is needed. We can even split the spill record into multiple ones | |
970 | at the places where they are needed. But CPS is not a good | |
971 | representation for global code motion, so I'll keep it simple and | |
972 | am not attempting this. | |
973 | ||
974 | 2. Incremental record construction (aka record splitting). | |
975 | ||
976 | Long records with many component values which are simulatenously live | |
977 | (recall that single use record addresses are not considered to | |
978 | be live) are constructed with rawrecord and rawupdate. | |
979 | We allocate space on the heap with rawrecord first, then gradually | |
980 | fill it in with rawupdate. This is the technique suggested to me | |
981 | by Matthias. | |
982 | ||
983 | Some restrictions on when this is applicable: | |
984 | 1. It is not a VECTOR record. The code generator currently does not handle | |
985 | this case. VECTOR record uses double indirection like arrays. | |
986 | 2. All the record component values are defined in the same "basic block" | |
987 | as the record constructor. This is to prevent speculative | |
988 | record construction. | |
989 | ||
990 | ---------------------------------------------------------------------- | |
991 | Name: Allen Leung | |
992 | Date: 2002/02/22 01:02:00 EST | |
993 | Tag: leunga-20020222-mlrisc-tools | |
994 | ||
995 | Minor bug fixes in the parser and rewriter | |
996 | ||
997 | ---------------------------------------------------------------------- | |
998 | Name: Allen Leung | |
999 | Date: 2002/02/21 20:20:00 EST | |
1000 | Tag: leunga-20020221-peephole | |
1001 | ||
1002 | Regenerated the peephole files. Some contained typos in the specification | |
1003 | and some didn't compile because of pretty printing bugs in the old version | |
1004 | of 'nowhere'. | |
1005 | ||
1006 | ---------------------------------------------------------------------- | |
1007 | Name: Allen Leung | |
1008 | Date: 2002/02/19 20:20:00 EST | |
1009 | Tag: leunga-20020219-mlrisc-tools | |
1010 | Description: | |
1011 | ||
1012 | Minor bug fixes to the mlrisc-tools library: | |
1013 | ||
1014 | 1. Fixed up parsing colon suffixed keywords | |
1015 | 2. Added the ability to shut the error messages up | |
1016 | 3. Reimplemented the pretty printer and fixed up/improved | |
1017 | the pretty printing of handle and -> types. | |
1018 | 4. Fixed up generation of literal symbols in the nowhere tool. | |
1019 | 5. Added some SML keywords to to sml.sty | |
1020 | ||
1021 | ---------------------------------------------------------------------- | |
1022 | Name: Matthias Blume | |
1023 | Date: 2002/02/19 16:20:00 EST | |
1024 | Tag: blume-20020219-cmffi | |
1025 | Description: | |
1026 | ||
1027 | A wild mix of changes, some minor, some major: | |
1028 | ||
1029 | * All C FFI-related libraries are now anchored under $c: | |
1030 | $/c.cm --> $c/c.cm | |
1031 | $/c-int.cm --> $c/internals/c-int.cm | |
1032 | $/memory.cm --> $c/memory/memory.cm | |
1033 | ||
1034 | * "make" tool (in CM) now treats its argument pathname slightly | |
1035 | differently: | |
1036 | 1. If the native expansion is an absolute name, then before invoking | |
1037 | the "make" command on it, CM will apply OS.Path.mkRelative | |
1038 | (with relativeTo = OS.FileSys.getDir()) to it. | |
1039 | 2. The argument will be passed through to subsequent phases of CM | |
1040 | processing without "going native". In particular, if the argument | |
1041 | was an anchored path, then "make" will not lose track of that anchor. | |
1042 | ||
1043 | * Compiler backends now "know" their respective C calling conventions | |
1044 | instead of having to be told about it by ml-nlffigen. This relieves | |
1045 | ml-nlffigen from one of its burdens. | |
1046 | ||
1047 | * The X86Backend has been split into X86CCallBackend and X86StdCallBackend. | |
1048 | ||
1049 | * Export C_DEBUG and C_Debug from $c/c.cm. | |
1050 | ||
1051 | * C type encoding in ml-nlffi-lib has been improved to model the conceptual | |
1052 | subtyping relationship between incomplete pointers and their complete | |
1053 | counterparts. For this, ('t, 'c) ptr has been changed to 'o ptr -- | |
1054 | with the convention of instantiating 'o with ('t, 'c) obj whenever | |
1055 | the pointer target type is complete. In the incomplete case, 'o | |
1056 | will be instantiated with some "'c iobj" -- a type obtained by | |
1057 | using one of the functors PointerToIncompleteType or PointerToCompleteType. | |
1058 | ||
1059 | Operations that work on both incomplete and complete pointer types are | |
1060 | typed as taking an 'o ptr while operations that require the target to | |
1061 | be known are typed as taking some ('t, 'c) obj ptr. | |
1062 | ||
1063 | voidptr is now a bit "more concrete", namely "type voidptr = void ptr'" | |
1064 | where void is an eqtype without any values. This makes it possible | |
1065 | to work on voidptr values using functions meant to operate on light | |
1066 | incomplete pointers. | |
1067 | ||
1068 | * As a result of the above, signature POINTER_TO_INCOMPLETE_TYPE has | |
1069 | been vastly simplified. | |
1070 | ||
1071 | ---------------------------------------------------------------------- | |
1072 | Name: Matthias Blume | |
1073 | Date: 2002/02/19 10:48:00 EST | |
1074 | Tag: blume-20020219-pqfix | |
1075 | Description: | |
1076 | ||
1077 | Applied Chris Okasaki's bug fix for priority queues. | |
1078 | ||
1079 | ---------------------------------------------------------------------- | |
1080 | Name: Matthias Blume | |
1081 | Date: 2002/02/15 17:05:00 | |
1082 | Tag: Release_110_39 | |
1083 | Description: | |
1084 | ||
1085 | Last-minute retagging is becoming a tradition... :-( | |
1086 | ||
1087 | This is the working release 110.39. | |
1088 | ||
1089 | ---------------------------------------------------------------------- | |
1090 | Name: Matthias Blume | |
1091 | Date: 2002/02/15 16:00:00 EST | |
1092 | Tag: Release_110_39-orig | |
1093 | Description: | |
1094 | ||
1095 | Working release 110.39. New bootfiles. | |
1096 | ||
1097 | (Update: There was a small bug in the installer so it wouldn't work | |
1098 | with all shells. So I retagged. -Matthias) | |
1099 | ||
1100 | ---------------------------------------------------------------------- | |
1101 | Name: Matthias Blume | |
1102 | Date: 2002/02/15 14:17:00 EST | |
1103 | Tag: blume-20020215-showbindings | |
1104 | Description: | |
1105 | ||
1106 | Added EnvRef.listBoundSymbols and CM.State.showBindings. Especially | |
1107 | the latter can be useful for exploring what bindings are available at | |
1108 | the interactive prompt. (The first function returns only the list | |
1109 | of symbols that are really bound, the second prints those but also the | |
1110 | ones that CM's autoloading mechanism knows about.) | |
1111 | ||
1112 | ---------------------------------------------------------------------- | |
1113 | Name: Matthias Blume | |
1114 | Date: 2002/02/15 12:08:00 EST | |
1115 | Tag: blume-20020215-iptrs | |
1116 | Description: | |
1117 | ||
1118 | Two improvements to ml-nlffigen: | |
1119 | ||
1120 | 1. Write files only if they do not exist or if their current contents | |
1121 | do not coincide with what's being written. (That is, avoid messing | |
1122 | with the time stamps unless absolutely necessary.) | |
1123 | ||
1124 | 2. Implement a "repository" mechanism for generated files related | |
1125 | to "incomplete pointer types". See the README file for details. | |
1126 | ||
1127 | ---------------------------------------------------------------------- | |
1128 | Name: Matthias Blume | |
1129 | Date: 2002/02/14 11:50:00 EST | |
1130 | Tag: blume-20020214-quote | |
1131 | Description: | |
1132 | ||
1133 | Added a type 't t_' to tag.sml (in ml-nlffi-lib.cm). This is required | |
1134 | because of the new and improved tag generation scheme. (Thanks to Allen | |
1135 | Leung for pointing it out.) | |
1136 | ||
1137 | ---------------------------------------------------------------------- | |
1138 | Name: Lal George | |
1139 | Date: 2002/02/14 09:55:27 EST 2002 | |
1140 | Tag: george-20020214-isabelle-bug | |
1141 | Description: | |
1142 | ||
1143 | Fixed the MLRISC bug sent by Markus Wenzel regarding the compilation | |
1144 | of Isabelle on the x86. | |
1145 | ||
1146 | From Allen: | |
1147 | ----------- | |
1148 | I've found the problem: | |
1149 | ||
1150 | in ra-core.sml, I use the counter "blocked" to keep track of the | |
1151 | true number of elements in the freeze queue. When the counter goes | |
1152 | to zero, I skip examining the queue. But I've messed up the | |
1153 | bookkeeping in combine(): | |
1154 | ||
1155 | else (); | |
1156 | case !ucol of | |
1157 | PSEUDO => (if !cntv > 0 then | |
1158 | (if !cntu > 0 then blocked := !blocked - 1 else (); | |
1159 | ^^^^^^^^^^^^^^^^^^^^^^^ | |
1160 | moveu := mergeMoveList(!movev, !moveu) | |
1161 | ) | |
1162 | else (); | |
1163 | ||
1164 | combine() is called to coalesce two nodes u and v. | |
1165 | I think I was thinking that if the move counts of u and v are both | |
1166 | greater than zero then after they are coalesced then one node is | |
1167 | removed from the freeze queue. Apparently I was thinking that | |
1168 | both u and v are of low degree, but that's clearly not necessarily true. | |
1169 | ||
1170 | ||
1171 | 02/12/2002: | |
1172 | Here's the patch. HOL now compiles. | |
1173 | ||
1174 | I don't know how this impact on performance (compile | |
1175 | time or runtime). This bug caused the RA (especially on the x86) | |
1176 | to go thru the potential spill phase when there are still nodes on the | |
1177 | freeze queue. | |
1178 | ||
1179 | ||
1180 | ||
1181 | ||
1182 | ---------------------------------------------------------------------- | |
1183 | Name: Matthias Blume | |
1184 | Date: 2002/02/13 22:40:00 EST | |
1185 | Tag: blume-20020213-fptr-rtti | |
1186 | Description: | |
1187 | ||
1188 | Fixed a bug in ml-nlffigen that was introduced with one of the previous | |
1189 | updates. | |
1190 | ||
1191 | ---------------------------------------------------------------------- | |
1192 | Name: Matthias Blume | |
1193 | Date: 2002/02/13 16:41:00 EST | |
1194 | Tag: blume-20020213-cmlpq | |
1195 | Description: | |
1196 | ||
1197 | Added new priority queue export symbols (which have just been added to | |
1198 | smlnj-lib.cm) to CML's version of smlnj-lib.cm. (Otherwise CML would | |
1199 | not compile and the installer would choke.) | |
1200 | ||
1201 | ---------------------------------------------------------------------- | |
1202 | Name: Matthias Blume | |
1203 | Date: 2002/02/13 16:15:00 EST | |
1204 | Tag: blume-20020213-various | |
1205 | Description: | |
1206 | ||
1207 | 1. More tweaks to ml-nlffigen: | |
1208 | ||
1209 | - better internal datastructures (resulting in slight speedup) | |
1210 | - "-match" option requires exact match | |
1211 | - "localized" gensym counters (untagged structs/unions nested within | |
1212 | other structs/unions or within typedefs get a fresh counter; their | |
1213 | tag will be prefixed by a concatenation of their parents' tags) | |
1214 | - bug fixes (related to calculation of transitive closure of types | |
1215 | to be included in the output) | |
1216 | ||
1217 | 2. Minor Basis updates: | |
1218 | ||
1219 | - added implementations for List.collate and Option.app | |
1220 | ||
1221 | ---------------------------------------------------------------------- | |
1222 | Name: Matthias Blume | |
1223 | Date: 2002/02/11 15:55:00 EST | |
1224 | Tag: blume-20020211-gensym | |
1225 | Description: | |
1226 | ||
1227 | Added a "-gensym" option to command line of ml-nlffigen. This can be | |
1228 | used to specify a "stem" -- a string that is inserted in all "gensym'd" | |
1229 | names (ML structure names that correspond to unnamed C structs, unions, | |
1230 | and enums), so that separate runs of ml-nlffigen do not clash. | |
1231 | ||
1232 | ---------------------------------------------------------------------- | |
1233 | Name: Matthias Blume | |
1234 | Date: 2002/02/11 12:05:00 EST | |
1235 | Tag: blume-20020211-gensml | |
1236 | Description: | |
1237 | ||
1238 | A quick fix for a problem with GenSML (in the pgraph-util library): | |
1239 | Make generation of toplevel "local" optional. (Strictly speaking, | |
1240 | signature definitions within "local" are not legal SML.) | |
1241 | ||
1242 | Other than that: updates to INSTALL and cm/TODO. | |
1243 | ||
1244 | ---------------------------------------------------------------------- | |
1245 | Name: Matthias Blume | |
1246 | Date: 2002/02/08 15:00:00 EST | |
1247 | Tag: blume-20020208-uniquepid | |
1248 | Description: | |
1249 | ||
1250 | 0. Version number has been bumped to 110.38.1. NEW BOOTFILES!!! | |
1251 | ||
1252 | 1. The installer (config/install.sh) has gotten smarter: | |
1253 | ||
1254 | - Configuration options are a bit easier to specify now | |
1255 | (in config/targets). | |
1256 | - Bug in recognizing .tar.bz2 files fixed. | |
1257 | - Installer automatically resolves dependencies between | |
1258 | configuration options (e.g., if you ask for eXene, you will | |
1259 | also get cml -- regardless whether you asked for it or not). | |
1260 | - Installer can run in "quieter mode" by setting the environment | |
1261 | variable INSTALL_QUIETLY to "true". "Quieter" does not mean | |
1262 | "completely silent", though. | |
1263 | - Build HashCons library as part of smlnj-lib. | |
1264 | ||
1265 | 2. A new scheme for assigning persistent identifiers to compilation | |
1266 | units (and, by extension, to types etc.) has been put into place. | |
1267 | This fixes a long-standing bug where types and even dynamic values | |
1268 | can get internally confused, thereby compromising type safety | |
1269 | (abstraction) and dynamic correctness. See | |
1270 | ||
1271 | http://cm.bell-labs.com/cm/cs/who/blume/pid-confusion.tgz | |
1272 | ||
1273 | for an example of how things could go wrong until now. | |
1274 | ||
1275 | The downside of the new scheme is that pids are not quite as | |
1276 | persistent as they used to be: CM will generate a fresh pid | |
1277 | for every compilation unit that it thinks it sees for the first | |
1278 | time. That means that if you compile starting from a clean, fresh | |
1279 | source tree at two different times, you end up with different | |
1280 | binaries. | |
1281 | ||
1282 | Cutoff recompilation, however, has not been compromised because | |
1283 | CM keeps pid information in special caches between runs. | |
1284 | ||
1285 | ---------------------------------------------------------------------- | |
1286 | Name: Lal George | |
1287 | Date: 2002/02/07 15:34:13 EST 2002 | |
1288 | Tag: <none> | |
1289 | Description: | |
1290 | ||
1291 | Compilers that generate assembly code may produce global labels | |
1292 | whose value is resolved at link time. The various peephole optimization | |
1293 | modules did not take this in account. | |
1294 | ||
1295 | TODO. The Labels.addrOf function should really return an option | |
1296 | type so that clients are forced to deal with this issue, rather | |
1297 | than an exception being raised. | |
1298 | ||
1299 | ---------------------------------------------------------------------- | |
1300 | Name: Lal George | |
1301 | Date: 2002/02/06 13:55:02 EST | |
1302 | Tag: george-20020206-ra-breakup | |
1303 | Description: | |
1304 | ||
1305 | 1. A bug fix from Allen. | |
1306 | ||
1307 | A typo causes extra fstp %st(0)'s to be generated at compensation | |
1308 | edges, which might cause stack underflow traps at runtime. This | |
1309 | occurs in fft where there are extraneous fstps right before the 'into' | |
1310 | trap instruction (in this case they are harmless since none of the | |
1311 | integers overflow.) | |
1312 | ||
1313 | 2. Pulled out various utility modules that were embedded in the modules | |
1314 | of the register allocator. I need these modules for other purposes, but | |
1315 | they are not complete enough to put into a library (just yet). | |
1316 | ---------------------------------------------------------------------- | |
1317 | Name: Matthias Blume | |
1318 | Date: 2002/01/31 16:05:00 EST | |
1319 | Tag: blume-20020131-sparc-ccalls | |
1320 | Description: | |
1321 | ||
1322 | 1. C-calls on Sparc needlessly allocated a huge chunk (96 bytes) | |
1323 | of extra stack space by mistake. Fixed. | |
1324 | ||
1325 | 2. Bug in logic of handling of command-line options in ml-nlffigen fixed. | |
1326 | ||
1327 | ---------------------------------------------------------------------- | |
1328 | Name: Allen Leung | |
1329 | Date: 2002/01/30 | |
1330 | Tag: leunga-20020130-nowhere-bug-fix | |
1331 | Description: | |
1332 | ||
1333 | MLRISC bug fixes: | |
1334 | 1. Fixed a bindings computation bug in the 'nowhere' program generator tool. | |
1335 | 2. MachineInt.fromString was negating its value. | |
1336 | ||
1337 | ---------------------------------------------------------------------- | |
1338 | Name: Matthias Blume | |
1339 | Date: 2002/01/29 | |
1340 | Tag: blume-20020129-INSTALL | |
1341 | Description: | |
1342 | ||
1343 | - Added somewhat detailed installation instructions (file INSTALL). | |
1344 | - Fixed curl-detection bug in config/install.sh. | |
1345 | - It is now possible to select the URL getter using the URLGETTER | |
1346 | environment variable: | |
1347 | ||
1348 | not set / "unknown" --> automatic detection (script tries wget, | |
1349 | curl, and lynx) | |
1350 | "wget" / "curl" / "lynx" --> use the specified program (script "knows" | |
1351 | how to properly invoke them) | |
1352 | other --> use $URLGETTER directly, it must take | |
1353 | precisely two command-line arguments | |
1354 | (source URL and destination file name) | |
1355 | ||
1356 | ---------------------------------------------------------------------- | |
1357 | Name: Matthias Blume | |
1358 | Date: 2002/01/28 | |
1359 | Tag: blume-20020128-sparc-ccalls | |
1360 | Description: | |
1361 | ||
1362 | - Fixed problem with calculation of "used" registers in sparc-c-calls. | |
1363 | - Make use of the allocParam argument in sparc-c-calls. | |
1364 | ||
1365 | ---------------------------------------------------------------------- | |
1366 | Name: Matthias Blume | |
1367 | Date: 2002/01/28 | |
1368 | Tag: blume-20020128-allocParam | |
1369 | Description: | |
1370 | ||
1371 | John Reppy: Changes c-calls API to accept client-callback for | |
1372 | allocating extra stack space. | |
1373 | me: Corresponding changes to mlriscGen (using a dummy argument that | |
1374 | does not change the current behavior). | |
1375 | ||
1376 | ---------------------------------------------------------------------- | |
1377 | Name: Matthias Blume | |
1378 | Date: 2002/01/28 12:00:00 | |
1379 | Tag: Release_110_38 | |
1380 | Description: | |
1381 | ||
1382 | This time for real!!! | |
1383 | ||
1384 | ---------------------------------------------------------------------- | |
1385 | Name: Matthias Blume | |
1386 | Date: 2002/01/28 10:56:00 EST | |
1387 | Tag: blume-20020128-retraction | |
1388 | Description: | |
1389 | ||
1390 | 0. Retracted earlier 110.38. (The Release_110_38 tag has been replaced | |
1391 | with blume-Release_110_38-retracted.) | |
1392 | ||
1393 | 1. Fixed a problem with incorrect rounding modes in real64.sml. | |
1394 | (Thanks to Andrew Mccreight <andrew.mccreight@yale.edu>.) | |
1395 | ||
1396 | 2. A bug in ml-nlffigen related to the handling of unnamed structs, unions, | |
1397 | and enums fixed. The naming of corresponding ML identifiers should | |
1398 | now be consistent again. | |
1399 | ||
1400 | ---------------------------------------------------------------------- | |
1401 | Name: Allen Leung | |
1402 | Date: 2002/01/27 | |
1403 | Tag: leunga-20020127-nowhere | |
1404 | Description: | |
1405 | ||
1406 | Added a target called nowhere in the configuration scripts. | |
1407 | Enabling this will build the MLRISC 'nowhere' tool (for translating | |
1408 | programs with where-clauses into legal SML code) during installation. | |
1409 | ||
1410 | ---------------------------------------------------------------------- | |
1411 | Name: Matthias Blume | |
1412 | Date: 2002/01/25 21:27:00 EST | |
1413 | Tag: blume-Release_110_38-retracted | |
1414 | Description: | |
1415 | ||
1416 | Call it a (working) release! Version is 110.38. Bootfiles are ready. | |
1417 | ||
1418 | README will be added later. | |
1419 | ||
1420 | !!! NOTE: Re-tagged as blume-Release_110_38-retracted. Original tag | |
1421 | (Release_110_38) removed. Reason: Last-minute bug fixes. | |
1422 | ||
1423 | ---------------------------------------------------------------------- | |
1424 | Name: Matthias Blume | |
1425 | Date: 2002/01/25 | |
1426 | Tag: blume-20020125-ffi | |
1427 | Description: | |
1428 | ||
1429 | A large number of tweaks and improvements to ml-nlffi-lib and | |
1430 | ml-nlffigen: | |
1431 | ||
1432 | - ML represenation types have been streamlined | |
1433 | - getter and setter functions work with concrete values, not abstract | |
1434 | ones where possible | |
1435 | - ml-nlffigen command line more flexible (see README file there) | |
1436 | - some bugs have been fixed (hopefully) | |
1437 | ||
1438 | ---------------------------------------------------------------------- | |
1439 | Name: Lal George | |
1440 | Date: 2002/01/24 | |
1441 | Tag: george-20020124-risc-ra-interface | |
1442 | Description: | |
1443 | ||
1444 | There is a dramatic simplification in the interface to the | |
1445 | register allocator for RISC architectures as a result of making | |
1446 | parallel copy instructions explicit. | |
1447 | ||
1448 | ---------------------------------------------------------------------- | |
1449 | Name: Matthias Blume | |
1450 | Date: 2002/01/22 | |
1451 | Tag: blume-20020122-x86-ccalls | |
1452 | Description: | |
1453 | ||
1454 | Bug fix for c-calls on x86 (having to do with how char- and | |
1455 | short-arguments are being handled). | |
1456 | ||
1457 | ---------------------------------------------------------------------- | |
1458 | Name: Matthias Blume | |
1459 | Date: 2002/01/21 | |
1460 | Tag: blume-20020121-ff | |
1461 | Description: | |
1462 | ||
1463 | Another day of fiddling with the FFI... | |
1464 | ||
1465 | 1. Bug fix/workaround: CKIT does not complain about negative array | |
1466 | dimensions, so ml-nlffigen has to guard itself against this possibility. | |
1467 | (Otherwise a negative dimension would send it into an infinite loop.) | |
1468 | ||
1469 | 2. Some of the abstract types (light objects, light pointers, most "base" | |
1470 | types) in structure C are now eqtypes. | |
1471 | ||
1472 | 3. Added constructors and test functions for NULL function pointers. | |
1473 | ||
1474 | ---------------------------------------------------------------------- | |
1475 | Name: Matthias Blume | |
1476 | Date: 2002/01/18 | |
1477 | Tag: blume-20020118-ready-for-new-release | |
1478 | Description: | |
1479 | ||
1480 | Made config/srcarchiveurl point to a new place. (Will provide boot | |
1481 | files shortly.) | |
1482 | ||
1483 | Maybe we christen this to be 110.38? | |
1484 | ||
1485 | ---------------------------------------------------------------------- | |
1486 | Name: Matthias Blume | |
1487 | Date: 2002/01/18 | |
1488 | Tag: blume-20020118-more-ffifiddle | |
1489 | Description: | |
1490 | ||
1491 | Today's FFI fiddling: | |
1492 | ||
1493 | - Provided a structure CGetSet with "convenient" versions of C.Get.* and | |
1494 | C.Set.* that use concrete (MLRep.*) arguments and results instead | |
1495 | of abstract ones. | |
1496 | ||
1497 | - Provided word-style bit operations etc. for "int" representation | |
1498 | types in MLRep.S<Foo>Bitops where <Foo> ranges over Char, Int, Short, | |
1499 | and Long. | |
1500 | ||
1501 | ---------------------------------------------------------------------- | |
1502 | Name: Matthias Blume | |
1503 | Date: 2002/01/18 | |
1504 | Tag: blume-20020118-use-x86-fp | |
1505 | Description: | |
1506 | ||
1507 | Now that x86-fast-fp seems to be working, I turned it back on again | |
1508 | by default. (Seems to work fine now, even with the FFI.) | |
1509 | ||
1510 | Other than that, I added some documentation about the FFI to | |
1511 | src/ml-nlffigen/README and updated the FFI test examples in | |
1512 | src/ml-nlffi-lib/Tests/*. | |
1513 | ||
1514 | ---------------------------------------------------------------------- | |
1515 | Name: Allen Leung | |
1516 | Date: 2002/01/17 | |
1517 | Tag: leunga-20020117-x86-fast-fp-call | |
1518 | Description: | |
1519 | ||
1520 | 1. Fixed a problem with handling return fp values when x86's fast fp | |
1521 | mode is turned on. | |
1522 | ||
1523 | 2. Minor pretty printing fix for cellset. Print %st(0) as %st(0) instead | |
1524 | of %f32. | |
1525 | ||
1526 | 3. Added a constructor INT32lit to the ast of MLRISC tools. | |
1527 | ||
1528 | ---------------------------------------------------------------------- | |
1529 | Name: Matthias Blume | |
1530 | Date: 2002/01/16 | |
1531 | Tag: blume-20020116-ffifiddle | |
1532 | Description: | |
1533 | ||
1534 | More fiddling with the FFI interface: | |
1535 | ||
1536 | - Make constness 'c instead of rw wherever possible. This eliminates | |
1537 | the need for certain explicit coercions. (However, due to ML's | |
1538 | value polymorphism, there will still be many cases where explicit | |
1539 | coercions are necessary. Phantom types are not the whole answer | |
1540 | to modeling a subtyping relationship in ML.) | |
1541 | ||
1542 | - ro/rw coersions for pointers added. (Avoids the detour through */&.) | |
1543 | ||
1544 | - "printf" test example added to src/ml-nlffi-lib/Tests. (Demonstrates | |
1545 | clumsy workaround for varargs problem.) | |
1546 | ||
1547 | ---------------------------------------------------------------------- | |
1548 | Name: Lal George | |
1549 | Date: 2002/01/15 | |
1550 | Tag: <none> | |
1551 | Description: | |
1552 | ||
1553 | 1. Since COPY instructions are no longer native to the architecture, | |
1554 | a generic functor can be used to implement the expandCopies function. | |
1555 | ||
1556 | 2. Allowed EXPORT and IMPORT pseudo-op declarations to appear inside a | |
1557 | TEXT segment. | |
1558 | ||
1559 | ---------------------------------------------------------------------- | |
1560 | Name: Matthias Blume | |
1561 | Date: 2002/01/15 | |
1562 | Tag: blume-20020115-ffiupdates | |
1563 | Description: | |
1564 | ||
1565 | 1. Fix for bug resulting in single-precision float values being returned | |
1566 | incorrectly from FFI calls. | |
1567 | ||
1568 | 2. Small modifications to C FFI API: | |
1569 | ||
1570 | - memory-allocation routines return straight objects (no options) | |
1571 | and raise an exception in out-of-memory situations | |
1572 | - unsafe extensions to cast between function pointers and pointers | |
1573 | from/to ints | |
1574 | - added structure C_Debug as an alternative to structure C where | |
1575 | pointer-dereferencing (|*| and |*!) always check for null-pointers | |
1576 | - added open_lib' to DynLinkage; open_lib' works like open_lib | |
1577 | but also takes a (possibly empty) list of existing library handles | |
1578 | that the current library depends on | |
1579 | ||
1580 | ---------------------------------------------------------------------- | |
1581 | Name: Matthias Blume | |
1582 | Date: 2002/01/10 | |
1583 | Tag: blume-20020110-newffigen | |
1584 | Description: | |
1585 | ||
1586 | 1. Updates to portable graph code. | |
1587 | ||
1588 | 2. Major update to ml-nlffigen and ml-nlffi-lib. Things are much | |
1589 | more scalable now so that even huge interfaces such as the one | |
1590 | for GTK compile in finite time and space. :-) | |
1591 | See src/ml-nlffigen/README for details on what's new. | |
1592 | ||
1593 | ---------------------------------------------------------------------- | |
1594 | Name: Lal George | |
1595 | Date: 2001/01/09 14:31:35 EST 2002 | |
1596 | Tag: george-20011206-rm-native-copy | |
1597 | Description: | |
1598 | ||
1599 | Removed the native COPY and FCOPY instructions | |
1600 | from all the architectures and replaced it with the | |
1601 | explicit COPY instruction from the previous commit. | |
1602 | ||
1603 | It is now possible to simplify many of the optimizations | |
1604 | modules that manipulate copies. This has not been | |
1605 | done in this change. | |
1606 | ||
1607 | ---------------------------------------------------------------------- | |
1608 | Name: Lal George | |
1609 | Date: 2001/12/06 16:50:13 EST 2001 | |
1610 | Tag: george-20011206-mlrisc-instruction | |
1611 | Description: | |
1612 | ||
1613 | Changed the representation of instructions from being fully abstract | |
1614 | to being partially concrete. That is to say: | |
1615 | ||
1616 | from | |
1617 | type instruction | |
1618 | ||
1619 | to | |
1620 | type instr (* machine instruction *) | |
1621 | ||
1622 | datatype instruction = | |
1623 | LIVE of {regs: C.cellset, spilled: C.cellset} | |
1624 | | KILL of {regs: C.cellset, spilled: C.cellset} | |
1625 | | COPYXXX of {k: CB.cellkind, dst: CB.cell list, src: CB.cell list} | |
1626 | | ANNOTATION of {i: instruction, a: Annotations.annotation} | |
1627 | | INSTR of instr | |
1628 | ||
1629 | This makes the handling of certain special instructions that appear on | |
1630 | all architectures easier and uniform. | |
1631 | ||
1632 | LIVE and KILL say that a list of registers are live or killed at the | |
1633 | program point where they appear. No spill code is generated when an | |
1634 | element of the 'regs' field is spilled, but the register is moved to | |
1635 | the 'spilled' (which is present, more for debugging than anything else). | |
1636 | ||
1637 | LIVE replaces the (now deprecated) DEFFREG instruction on the alpha. | |
1638 | We used to generate: | |
1639 | ||
1640 | DEFFREG f1 | |
1641 | f1 := f2 + f3 | |
1642 | trapb | |
1643 | ||
1644 | but now generate: | |
1645 | ||
1646 | f1 := f2 + f3 | |
1647 | trapb | |
1648 | LIVE {regs=[f1,f2,f3], spilled=[]} | |
1649 | ||
1650 | Furthermore, the DEFFREG (hack) required that all floating point instruction | |
1651 | use all registers mentioned in the instruction. Therefore f1 := f2 + f3, | |
1652 | defines f1 and uses [f1,f2,f3]! This hack is no longer required resulting | |
1653 | in a cleaner alpha implementation. (Hopefully, intel will not get rid of | |
1654 | this architecture). | |
1655 | ||
1656 | COPYXXX is intended to replace the parallel COPY and FCOPY available on | |
1657 | all the architectures. This will result in further simplification of the | |
1658 | register allocator that must be aware of them for coalescing purposes, and | |
1659 | will also simplify certain aspects of the machine description that provides | |
1660 | callbacks related to parallel copies. | |
1661 | ||
1662 | ANNOTATION should be obvious, and now INSTR represents the honest to God | |
1663 | machine instruction set! | |
1664 | ||
1665 | The <arch>/instructions/<arch>Instr.sml files define certain utility | |
1666 | functions for making porting easier -- essentially converting upper case | |
1667 | to lower case. All machine instructions (of type instr) are in upper case, | |
1668 | and the lower case form generates an MLRISC instruction. For example on | |
1669 | the alpha we have: | |
1670 | ||
1671 | datatype instr = | |
1672 | LDA of {r:cell, b:cell, d:operand} | |
1673 | | ... | |
1674 | ||
1675 | val lda : {r:cell, b:cell, d:operand} -> instruction | |
1676 | ... | |
1677 | ||
1678 | where lda is just (INSTR o LDA), etc. | |
1679 | ||
1680 | ---------------------------------------------------------------------- | |
1681 | Name: Matthias Blume | |
1682 | Date: 2001/11/22 21:40:00 EST | |
1683 | Tag: Release_110_37 | |
1684 | Description: | |
1685 | ||
1686 | Release 110.37. This time for real. | |
1687 | ||
1688 | ---------------------------------------------------------------------- | |
1689 | Name: Matthias Blume | |
1690 | Date: 2001/11/21 16:35:00 EST | |
1691 | Tag: blume-20011121-foot-in-mouth | |
1692 | Description: | |
1693 | ||
1694 | Removed the "Release_110_37" tag because of a serious bug. | |
1695 | This will be re-tagged once the bug is fixed. | |
1696 | ||
1697 | ---------------------------------------------------------------------- | |
1698 | Name: Matthias Blume | |
1699 | Date: 2001/11/21 16:14:00 EST | |
1700 | Tag: blume-20011121-forgottenfile | |
1701 | Description: | |
1702 | ||
1703 | Forgot to add a file. (Just a .tex-file -- part of | |
1704 | the CM manual source.) | |
1705 | ||
1706 | ---------------------------------------------------------------------- | |
1707 | Name: Matthias Blume | |
1708 | Date: 2001/11/21 16:10:00 EST | |
1709 | Tag: blume-20011121-invalid_110_37 | |
1710 | Description: | |
1711 | ||
1712 | Note: I removed the original tag "Release_110_37" from this commit | |
1713 | because we found a serious bug in all non-x86 backends. | |
1714 | - Matthias | |
1715 | ||
1716 | 1. Modifications to the SML/NJ code generator and to the runtime system | |
1717 | so that code object name strings are directly inserted into code | |
1718 | objects at code generation time. The only business the runtime system | |
1719 | has with this is now to read the name strings on occasions. | |
1720 | (The encoding of the name string has also changed somewhat.) | |
1721 | ||
1722 | 2. CM now implements a simple "set calculus" for specifying export lists. | |
1723 | In particular, it is now possible to refer to the export lists of | |
1724 | other libraries/groups/sources and form unions as well as differences. | |
1725 | See the latest CM manual for details. | |
1726 | ||
1727 | 3. An separate notion of "proxy" libraries has again be eliminated from | |
1728 | CM's model. (Proxy libraries are now simply a special case of using | |
1729 | the export list calculus.) | |
1730 | ||
1731 | 4. Some of the existing libraries now take advantage of the new set | |
1732 | calculus. | |
1733 | (Notice that not all libraries have been converted because some | |
1734 | of the existing .cm-files are supposed to be backward compatible | |
1735 | with 110.0.x.) | |
1736 | ||
1737 | 5. Some cleanup in stand-alone programs. (Don't use "exnMessage" -- use | |
1738 | "General.exnMessage"! The former relies on a certain hook to be | |
1739 | initialized, and that often does not happen in the stand-alone case.) | |
1740 | ||
1741 | ---------------------------------------------------------------------- | |
1742 | Name: Lal George | |
1743 | Date: 2001/11/21 13:56:18 EST | |
1744 | Tag: george-2001121-pseudo-ops | |
1745 | Description: | |
1746 | ||
1747 | Implemented a complete redesign of MLRISC pseudo-ops. Now there | |
1748 | ought to never be any question of incompatabilities with | |
1749 | pseudo-op syntax expected by host assemblers. | |
1750 | ||
1751 | For now, only modules supporting GAS syntax are implemented | |
1752 | but more should follow, such as MASM, and vendor assembler | |
1753 | syntax, e.g. IBM as, Sun as, etc. | |
1754 | ||
1755 | ---------------------------------------------------------------------- | |
1756 | Name: Matthias Blume | |
1757 | Date: 2001/11/14 11:52:00 EST | |
1758 | Tag: blume-20011114-srcname | |
1759 | Description: | |
1760 | ||
1761 | 1. Routed the name of the current source file to mlriscgen where it | |
1762 | should be directly emitted into the code object. (This last part | |
1763 | is yet to be done.) | |
1764 | ||
1765 | 2. Some cleanup of the pgraph code to make it match the proposal that | |
1766 | I put out the other day. (The proposal notwithstanding, things are | |
1767 | still in flux here.) | |
1768 | ||
1769 | ---------------------------------------------------------------------- | |
1770 | Name: Lal George | |
1771 | Date: 2001/11/14 09:44:04 EST | |
1772 | Tag: | |
1773 | Description: | |
1774 | ||
1775 | Fix for a backpatching bug reported by Allen. | |
1776 | ||
1777 | Because the boundary between short and long span-dependent | |
1778 | instructions is +/- 128, there are an astounding number of | |
1779 | span-dependent instructions whose size is over estimated. | |
1780 | ||
1781 | Allen came up with the idea of letting the size of span | |
1782 | dependent instructions be non-monotonic, for a maxIter | |
1783 | number of times, after which the size must be monotonically | |
1784 | increasing. | |
1785 | ||
1786 | This table shows the number of span-dependent instructions | |
1787 | whose size was over-estimated as a function of maxIter, for the | |
1788 | file Parse/parse/ml.grm.sml: | |
1789 | ||
1790 | maxIter # of instructions: | |
1791 | 10 687 | |
1792 | 20 438 | |
1793 | 30 198 | |
1794 | 40 0 | |
1795 | ||
1796 | In compiling the compiler, there is no significant difference in | |
1797 | compilation speed between maxIter=10 and maxIter=40. Actually, | |
1798 | my measurements showed that maxIter=40 was a tad faster than | |
1799 | maxIter=10! Also 96% of the files in the compiler reach a fix | |
1800 | point within 13 iterations, so fixing maxIter at 40, while high, | |
1801 | is okay. | |
1802 | ||
1803 | ---------------------------------------------------------------------- | |
1804 | Name: Matthias Blume | |
1805 | Date: 2001/10/31 15:25:00 EST | |
1806 | Tag: blume-20011031-pgraph | |
1807 | Description: | |
1808 | ||
1809 | CKIT: | |
1810 | * Changed the "Function" constructor of type Ast.ctype to carry optional | |
1811 | argument identifiers. | |
1812 | * Changed the return type of TypeUtil.getFunction accordingly. | |
1813 | * Type equality ignores the argument names. | |
1814 | * TypeUtil.composite tries to preserve argument names but gives up quickly | |
1815 | if there is a mismatch. | |
1816 | ||
1817 | installation script: | |
1818 | * attempts to use "curl" if available (unless "wget" is available as well) | |
1819 | ||
1820 | CM: | |
1821 | * has an experimental implementation of "portable graphs" which I will | |
1822 | soon propose as an implementation-independent library format | |
1823 | * there are also new libraries $/pgraph.cm and $/pgraph-util.cm | |
1824 | ||
1825 | NLFFI-LIB: | |
1826 | * some cleanup (all cosmetic) | |
1827 | ||
1828 | NLFFIGEN: | |
1829 | * temporarily disabled the mechanism that suppresses ML output for | |
1830 | C definitions whose identifiers start with an underscore character | |
1831 | * generate val bindings for enum constants | |
1832 | * user can request that only one style (light or heavy) is being used; | |
1833 | default is to use both (command-line arguments: -heavy and -light) | |
1834 | * fixed bug in handling of function types involving incomplete pointers | |
1835 | * generate ML entry points that take record arguments (i.e., using | |
1836 | named arguments) for C functions that have a prototype with named | |
1837 | arguments | |
1838 | (see changes to CKIT) | |
1839 | ||
1840 | ---------------------------------------------------------------------- | |
1841 | Name: Allen Leung | |
1842 | Date: 2001/10/27 20:34:00 EDT | |
1843 | Tag: leunga-20011027-x86-fast-fp-call | |
1844 | Description: | |
1845 | ||
1846 | Fixed the bug described in blume-20010920-slowfp. | |
1847 | ||
1848 | The fix involves | |
1849 | 1. generating FCOPYs in FSTP in ia32-svid | |
1850 | 2. marking a CALL with the appropriate annotation | |
1851 | ||
1852 | ---------------------------------------------------------------------- | |
1853 | Name: Matthias Blume | |
1854 | Date: 2001/10/16 11:32:00 EDT | |
1855 | Tag: blume-20011016-netbsd | |
1856 | Description: | |
1857 | ||
1858 | Underscore patch from Chris Richards (fixing problem with compiling | |
1859 | runtime system under recent NetBSD). | |
1860 | ||
1861 | ---------------------------------------------------------------------- | |
1862 | Name: Allen Leung | |
1863 | Date: 2001/10/12 17:18:32 EDT 2001 | |
1864 | Tag: leung-20011012-x86-printflowgraph | |
1865 | Description: | |
1866 | ||
1867 | X86RA now uses a valid (instead of dummy) PrintFlowgraph module. | |
1868 | ||
1869 | ---------------------------------------------------------------------- | |
1870 | Name: Lal George | |
1871 | Date: 2001/10/11 23:51:34 EDT | |
1872 | Tag: george-20011011-too-many-instrs | |
1873 | Description: | |
1874 | ||
1875 | The representation of a program point never expected to see more | |
1876 | than 65536 instructions in a basic block! | |
1877 | ||
1878 | ---------------------------------------------------------------------- | |
1879 | Name: Lal George | |
1880 | Date: 2001/10/09 09:41:37 EDT | |
1881 | Tag: george-20011008-mlrisc-labels | |
1882 | Description: | |
1883 | ||
1884 | Changed the machine description files to support printing of | |
1885 | local and global labels in assembly code, based on host assembler | |
1886 | conventions. | |
1887 | ||
1888 | ---------------------------------------------------------------------- | |
1889 | Name: Matthias Blume | |
1890 | Date: 2001/09/25 15:25:00 EDT | |
1891 | Tag: blume-20010925-exninfo | |
1892 | Description: | |
1893 | ||
1894 | I provided a non-hook implementation of exnName (at the toplevel) and | |
1895 | made the "dummy" implementation of exnMessage (at the toplevel) more | |
1896 | useful: if nothing gets "hooked in", then at least you are going to | |
1897 | see the exception name and a message indicating why you don't see more. | |
1898 | ||
1899 | [For the time being, programs that need exnMessage and want to use | |
1900 | ml-build should either use General.exnMessage (strongly recommended) or | |
1901 | refer to structure General at some other point so that CM sees a | |
1902 | static dependency.] | |
1903 | ||
1904 | [Similar remarks go for "print" and "use": If you want to use their | |
1905 | functionality in stand-alone programs generated by ml-build, then use | |
1906 | TextIO.output and Backend.Interact.useFile (from $smlnj/compiler.cm).] | |
1907 | ||
1908 | ---------------------------------------------------------------------- | |
1909 | Name: Matthias Blume | |
1910 | Date: 2001/09/20 17:28:00 EDT | |
1911 | Tag: blume-20010920-slowfp | |
1912 | Description: | |
1913 | ||
1914 | Allen says that x86-fast-fp is not safe yet, so I turned it off again... | |
1915 | ||
1916 | ---------------------------------------------------------------------- | |
1917 | Name: Matthias Blume | |
1918 | Date: 2001/09/20 17:20:00 EDT | |
1919 | Tag: blume-20010920-canonicalpaths | |
1920 | Description: | |
1921 | ||
1922 | 0. Updated the BOOT file (something that I forgot to do earlier). | |
1923 | ||
1924 | 1. Small internal change to CM so that it avoids "/../" in filenames | |
1925 | as much as possible (but only where it is safe). | |
1926 | ||
1927 | 2. Changed config/_run-sml (resulting in a changed bin/.run-sml) so | |
1928 | that arguments that contain delimiters are passed through correctly. | |
1929 | This change also means that all "special" arguments of the form | |
1930 | @SMLxxx... must come first. | |
1931 | ||
1932 | 3. Changed install script to put relative anchor names for tool commands | |
1933 | into pathconfig. | |
1934 | ||
1935 | ---------------------------------------------------------------------- | |
1936 | Name: Matthias Blume | |
1937 | Date: 2001/09/18 15:35:00 EDT | |
1938 | Tag: blume-20010918-readme11036 | |
1939 | Description: | |
1940 | ||
1941 | Added README files. | |
1942 | ||
1943 | ---------------------------------------------------------------------- | |
1944 | Name: Matthias Blume | |
1945 | Date: 2001/09/18 11:45:00 EDT | |
1946 | Tag: Release_110_36 (retag) | |
1947 | Description: | |
1948 | ||
1949 | Fixed mistake in config/preloads. Retagged as 110.36. | |
1950 | ||
1951 | ---------------------------------------------------------------------- | |
1952 | Name: Matthias Blume | |
1953 | Date: 2001/09/18 09:40:00 EDT | |
1954 | Tag: Release_110_36_orig (tag changed) | |
1955 | Description: | |
1956 | ||
1957 | New version (110.36). New bootfiles. | |
1958 | ||
1959 | ---------------------------------------------------------------------- | |
1960 | Name: Matthias Blume | |
1961 | Date: 2001/09/14 16:15:00 EDT | |
1962 | Tag: blume-20010914-x86fastfp | |
1963 | Description: | |
1964 | ||
1965 | John committed some changes that Allen made, in particular a (hopefully) | |
1966 | correctly working version of the x86-fp module. | |
1967 | ||
1968 | I changed the default setting of the Control.MLRISC.getFlag "x86-fast-fp" | |
1969 | flag to "true". Everything seems to compile to a fixpoint ok, and | |
1970 | "mandelbrot" speeds up by about 15%. | |
1971 | ||
1972 | ---------------------------------------------------------------------- | |
1973 | Name: Matthias Blume | |
1974 | Date: 2001/09/13 11:20:00 EDT | |
1975 | Tag: blume-20010913-minimal | |
1976 | Description: | |
1977 | ||
1978 | 1. Stefan Monnier's patch to fix a miscompilation problem that | |
1979 | was brought to light by John Reppy's work on Moby. | |
1980 | ||
1981 | 2. Implemented a minimal "structure Compiler" that contains just | |
1982 | "version" and "architecture". The minimal version will be | |
1983 | available when the full version is not. This is for backward- | |
1984 | compatibility with code that wants to test Compiler.version. | |
1985 | ||
1986 | ---------------------------------------------------------------------- | |
1987 | Name: Matthias Blume | |
1988 | Date: 2001/08/28 14:03:00 EDT | |
1989 | Tag: blume-20010828-ml-lex | |
1990 | Description: | |
1991 | ||
1992 | Fix for bug 1581, received from Neophytos Michael. | |
1993 | ||
1994 | ---------------------------------------------------------------------- | |
1995 | Name: Matthias Blume | |
1996 | Date: 2001/08/27 11:20:00 EDT | |
1997 | Tag: blume-20010827-readme11035 | |
1998 | Description: | |
1999 | ||
2000 | Fleshed out the README file for 110.35. | |
2001 | ||
2002 | ---------------------------------------------------------------------- | |
2003 | Name: Matthias Blume | |
2004 | Date: 2001/08/24 17:10:00 EDT | |
2005 | Tag: Release_110_35 | |
2006 | Description: | |
2007 | ||
2008 | New version number (110.35). New bootfiles. | |
2009 | ||
2010 | ---------------------------------------------------------------------- | |
2011 | Name: Lal George | |
2012 | Date: 2001/08/24 13:47:18 EDT 2001 | |
2013 | Tag: george-20010824-MLRISC-graphs | |
2014 | Description: | |
2015 | ||
2016 | removed clusters from MLRISC completely and replaced with graphs. | |
2017 | ||
2018 | ---------------------------------------------------------------------- | |
2019 | Name: Matthias Blume | |
2020 | Date: 2001/08/23 17:50:00 EDT | |
2021 | Tag: blume-20010823-toplevel | |
2022 | Description: | |
2023 | ||
2024 | - some reorganization of the code that implements various kinds of | |
2025 | environments in the compiler (static, dynamic, symbolic, combined) | |
2026 | - re-implemented the EnvRef module so that evalStream works properly | |
2027 | (if the stream contains references to "use", "CM.make", etc.) | |
2028 | - cleaned up evalloop.sml and interact.sml (but they need more cleaning) | |
2029 | ||
2030 | ---------------------------------------------------------------------- | |
2031 | Name: Matthias Blume | |
2032 | Date: 2001/08/20 15:50 EDT | |
2033 | Tag: blume20010820-slipup | |
2034 | Description: | |
2035 | ||
2036 | I forgot to commit a few files. Here they are... | |
2037 | ||
2038 | ---------------------------------------------------------------------- | |
2039 | Name: Matthias Blume | |
2040 | Date: 2001/08/20 15:35:00 EDT | |
2041 | Tag: blume-20010820-debugprof | |
2042 | Description: | |
2043 | ||
2044 | !!!! NEW BOOTFILES !!!! | |
2045 | ||
2046 | This is another round of reorganizing the compiler sources. This | |
2047 | time the main goal was to factor out all the "instrumentation" | |
2048 | passes (for profiling and backtracing) into their own library. | |
2049 | The difficulty was to do it in such a way that it does not depend | |
2050 | on elaborate.cm but only on elabdata.cm. | |
2051 | ||
2052 | Therefore there have been further changes to both elaborate.cm and | |
2053 | elabdata.cm -- more "generic" things have been moved from the former | |
2054 | to the latter. As a result, I was forced to split the assignment | |
2055 | of numbers indicating "primtyc"s into two portions: SML-generic and | |
2056 | SML/NJ-specific. Since it would have been awkward to maintain, | |
2057 | I bit the bullet and actually _changed_ the mapping between these | |
2058 | numbers and primtycs. The bottom line of this is that you need | |
2059 | a new set of bin- and bootfiles. | |
2060 | ||
2061 | I have built new bootfiles for all architectures, so doing a fresh | |
2062 | checkout and config/install.sh should be all you need. | |
2063 | ||
2064 | The newly created library's name is | |
2065 | ||
2066 | $smlnj/viscomp/debugprof.cm | |
2067 | ||
2068 | and its sources live under | |
2069 | ||
2070 | src/compiler/DebugProf | |
2071 | ||
2072 | ---------------------------------------------------------------------- | |
2073 | Name: Matthias Blume | |
2074 | Date: 2001/08/15 17:15:00 EDT | |
2075 | Tag: blume-20010815-compreorg | |
2076 | Description: | |
2077 | ||
2078 | This is a first cut at reorganizing the CM libraries that make up the | |
2079 | core of the compiler. The idea is to separate out pieces that could | |
2080 | be used independently by tools, e.g., the parser, the typechecker, etc. | |
2081 | ||
2082 | The current status is a step in this direction, but it is not quite | |
2083 | satisfactory yet. Expect more changes in the future. | |
2084 | ||
2085 | Here is the current (new) organization... | |
2086 | ||
2087 | What used to be $smlnj/viscomp/core.cm is now divided into | |
2088 | six CM libraries: | |
2089 | ||
2090 | $smlnj/viscomp/basics.cm | |
2091 | /parser.cm | |
2092 | /elabdata.cm | |
2093 | /elaborate.cm | |
2094 | /execute.cm | |
2095 | /core.cm | |
2096 | ||
2097 | The CM files for these libraries live under src/system/smlnj/viscomp. | |
2098 | All these libraries are proxy libraries that contain precisely | |
2099 | one CM library component. Here are the locations of the components | |
2100 | (all within the src/compiler tree): | |
2101 | ||
2102 | Basics/basics.cm | |
2103 | Parse/parser.cm | |
2104 | ElabData/elabdata.cm | |
2105 | Elaborator/elaborate.cm | |
2106 | Execution/execute.cm | |
2107 | core.cm | |
2108 | ||
2109 | [This organization is the same that has been used already | |
2110 | for a while for the architecture-specific parts of the visible | |
2111 | compiler and for the old version of core.cm.] | |
2112 | ||
2113 | As you will notice, many source files have been moved from their | |
2114 | respective original locations to a new home in one of the above | |
2115 | subtrees. | |
2116 | ||
2117 | The division of labor between the new libraries is the following: | |
2118 | ||
2119 | basics.cm: | |
2120 | - Simple, basic definitions that pertain to many (or all) of | |
2121 | the other libraries. | |
2122 | parser.cm: | |
2123 | - The SML parser, producing output of type Ast.dec. | |
2124 | - The type family for Ast is also defined and exported here. | |
2125 | elabdata.cm: | |
2126 | - The datatypes that describe input and output of the elaborator. | |
2127 | This includes types, absyn, and static environments. | |
2128 | elaborator.cm: | |
2129 | - The SML/NJ type checker and elaborator. | |
2130 | This maps an Ast.dec (with a given static environment) to | |
2131 | an Absyn.dec (with a new static environment). | |
2132 | - This libraries implements certain modules that used to be | |
2133 | structures as functors (to remove dependencies on FLINT). | |
2134 | execute.cm: | |
2135 | - Everything having to do with executing binary code objects. | |
2136 | - Dynamic environments. | |
2137 | core.cm: | |
2138 | - SML/NJ-specific instantiations of the elaborator and MLRISC. | |
2139 | - Top-level modules. | |
2140 | - FLINT (this should eventually become its own library) | |
2141 | ||
2142 | Notes: | |
2143 | ||
2144 | I am not 100% happy with the way I separated the elaborator (and its | |
2145 | data structures) from FLINT. Two instances of the same problem: | |
2146 | ||
2147 | 1. Data structures contain certain fields that carry FLINT-specific | |
2148 | information. I hacked around this using exn and the property list | |
2149 | module from smlnj-lib. But the fact that there are middle-end | |
2150 | specific fields around at all is a bit annoying. | |
2151 | ||
2152 | 2. The elaborator calculates certain FLINT-related information. I tried | |
2153 | to make this as abstract as I could using functorization, but, again, | |
2154 | the fact that the elaborator has to perform calculations on behalf | |
2155 | of the middle-end at all is not nice. | |
2156 | ||
2157 | 3. Having to used exn and property lists is unfortunate because it | |
2158 | weakens type checking. The other alternative (parameterizing | |
2159 | nearly *everything*) is not appealing, though. | |
2160 | ||
2161 | I removed the "rebinding =" warning hack because due to the new organization | |
2162 | it was awkward to maintain it. As a result, the compiler now issues some of | |
2163 | these warnings when compiling init.cmi during bootstrap compilation. On | |
2164 | the plus side, you also get a warning when you do, for example: | |
2165 | val op = = Int32.+ | |
2166 | which was not the case up to now. | |
2167 | ||
2168 | I placed "assign" and "deref" into the _Core structure so that the | |
2169 | code that deals with the "lazy" keyword can find them there. This | |
2170 | removes the need for having access to the primitive environment | |
2171 | during elaboration. | |
2172 | ||
2173 | ---------------------------------------------------------------------- | |
2174 | Name: Matthias Blume | |
2175 | Date: 2001/08/13 | |
2176 | Tag: blume-20010813-closures | |
2177 | Description: | |
2178 | ||
2179 | This fix was sent to us by Zhong Shao. It is supposed to improve the | |
2180 | performance of certain loops by avoiding needless closure allocation. | |
2181 | ||
2182 | ---------------------------------------------------------------------- | |
2183 | Name: Lal George | |
2184 | Date: 2001/07/31 10:03:23 EDT 2001 | |
2185 | Tag: george-20010731-x86-fmalloc | |
2186 | Description: Fixed bug in x86 calls | |
2187 | ||
2188 | There was a bug where call instructions would mysteriously | |
2189 | vanish. The call instruction had to be one that returned | |
2190 | a floating point value. | |
2191 | ||
2192 | ---------------------------------------------------------------------- | |
2193 | Name: Lal George | |
2194 | Date: 2001/07/19 16:36:29 EDT 2001 | |
2195 | Tag: george-20010719-simple-cells | |
2196 | Description: | |
2197 | ||
2198 | I have dramatically simplified the interface for CELLS in MLRISC. | |
2199 | ||
2200 | In summary, the cells interface is broken up into three parts: | |
2201 | ||
2202 | 1. CellsBasis : CELLS_BASIS | |
2203 | ||
2204 | CellsBasis is a top level structure and common for all | |
2205 | architectures. it contains the definitions of basic datatypes | |
2206 | and utility functions over these types. | |
2207 | ||
2208 | 2. functor Cells() : CELLS | |
2209 | ||
2210 | Cells generates an interface for CELLS that incorporates the | |
2211 | specific resources on the target architecture, such as the | |
2212 | presence of special register classes, their number and size, | |
2213 | and various useful substructures. | |
2214 | ||
2215 | 3. <ARCH>CELLS | |
2216 | ||
2217 | e.g. SparcCells: SPARCCELLS | |
2218 | ||
2219 | <ARCH>CELLS usually contains additional bindings for special | |
2220 | registers on the architecture, such as: | |
2221 | ||
2222 | val r0 : cell (* register zero *) | |
2223 | val y : cell (* Y register *) | |
2224 | val psr : cell (* processor status register *) | |
2225 | ... | |
2226 | ||
2227 | The structure returned by applying the Cells functor is opened | |
2228 | in this interface. | |
2229 | ||
2230 | The main implication of all this is that the datatypes for cells is | |
2231 | split between CellsBasis and CELLS -- a fairly simple change for user | |
2232 | code. | |
2233 | ||
2234 | In the old scheme the CELLS interface had a definitional binding of | |
2235 | the form: | |
2236 | ||
2237 | signature CELLS = sig | |
2238 | ||
2239 | structure CellsBasis = CellsBasis | |
2240 | ||
2241 | ... | |
2242 | ||
2243 | end | |
2244 | ||
2245 | With all the sharing constraints that goes on in MLRISC, this old | |
2246 | design quickly leads to errors such as: | |
2247 | ||
2248 | "structure definition spec inside of sharing ... " | |
2249 | ||
2250 | ||
2251 | and appears to require an unacceptable amount of sharing and where | |
2252 | constraint hackery. | |
2253 | ||
2254 | I think this error message (the interaction of definitional specs and | |
2255 | sharing) requires more explanation on our web page. | |
2256 | ||
2257 | ---------------------------------------------------------------------- | |
2258 | Name: Matthias Blume | |
2259 | Date: 2001/07/19 15:00:00 EDT | |
2260 | Tag: blume-20010719-libreorg | |
2261 | Description: | |
2262 | ||
2263 | This update puts together a fairly extensive but straightforward change | |
2264 | to the way the libraries that implement the interactive system are | |
2265 | organized: | |
2266 | ||
2267 | The biggest change is the elimination of structure Compiler. As a | |
2268 | replacement for this structure, there is now a CM library | |
2269 | (known as $smlnj/compiler.cm or $smlnj/compiler/current.cm) | |
2270 | that exports all the substructures of the original structure Compiler | |
2271 | directly. So instead of saying Compiler.Foo.bar one now simply | |
2272 | says Foo.bar. (The CM libraries actually export a collection of | |
2273 | structures that is richer than the collection of substructures of | |
2274 | structure Compiler.) | |
2275 | ||
2276 | To make the transition smooth, there is a separate library called | |
2277 | $smlnj/compiler/compiler.cm which puts together and exports the | |
2278 | original structure Compiler (or at least something very close to it). | |
2279 | ||
2280 | There are five members of the original structure Compiler | |
2281 | that are not exported directly but which instead became members | |
2282 | of a new structure Backend (described by signature BACKEND). These are: | |
2283 | structure Profile (: PROFILE), structure Compile (: COMPILE), structure | |
2284 | Interact (: INTERACT), structure Machine (: MACHINE), and val | |
2285 | architecture (: string). | |
2286 | ||
2287 | Structure Compiler.Version has become structure CompilerVersion. | |
2288 | ||
2289 | Cross-compilers for alpha32, hppa, ppc, sparc, and x86 are provided | |
2290 | by $smlnj/compiler/<arch>.cm where <arch> is alpha32, hppa, ppc, sparc, | |
2291 | or x86, respectively. | |
2292 | Each of these exports the same frontend structures that | |
2293 | $smlnj/compiler.cm exports. But they do not have a structure Backend | |
2294 | and instead export some structure <Arch>Backend where <Arch> is Alpha32, | |
2295 | Hppa, PPC, Sparc, or X86, respectively. | |
2296 | ||
2297 | Library $smlnj/compiler/all.cm exports the union of the exports of | |
2298 | $smlnj/compiler/<arch>.cm | |
2299 | ||
2300 | There are no structures <Arch>Compiler anymore, use | |
2301 | $smlnj/compiler/<arch>.cm instead. | |
2302 | ||
2303 | Library host-compiler-0.cm is gone. Instead, the internal library | |
2304 | that instantiates CM is now called cm0.cm. Selection of the host | |
2305 | compiler (backend) is no longer done here but. (Responsibility for it | |
2306 | now lies with $smlnj/compiler/current.cm. This seems to be more | |
2307 | logical.) | |
2308 | ||
2309 | Many individual files have been moved or renamed. Some files have | |
2310 | been split into multiple files, and some "dead" files have been deleted. | |
2311 | ||
2312 | Aside from these changes to library organization, there are also changes | |
2313 | to the way the code itself is organized: | |
2314 | ||
2315 | Structure Binfile has been re-implemented in such a way that it no | |
2316 | longer needs any knowledge of the compiler. It exclusively deals | |
2317 | with the details of binfile layout. It no longer invokes the | |
2318 | compiler (for the purpose of creating new prospective binfile | |
2319 | content), and it no longer has any knowledge of how to interpret | |
2320 | pickles. | |
2321 | ||
2322 | Structure Compile (: COMPILE) has been stripped down to the bare | |
2323 | essentials of compilation. It no longer deals with linking/execution. | |
2324 | The interface has been cleaned up considerably. | |
2325 | ||
2326 | Utility routines for dealing with linking and execution have been | |
2327 | moved into their own substructures. | |
2328 | ||
2329 | (The ultimate goal of these changes is to provide a light-weight | |
2330 | binfile loader/linker (at least for, e.g., stable libraries) that | |
2331 | does not require CM or the compiler to be present.) | |
2332 | ||
2333 | CM documentation has been updated to reflect the changes to library | |
2334 | organization. | |
2335 | ||
2336 | ---------------------------------------------------------------------- | |
2337 | Name: Matthias Blume | |
2338 | Date: 2001/07/10 17:30:00 EDT | |
2339 | Tag: Release_110_34 | |
2340 | Description: | |
2341 | ||
2342 | Minor tweak to 110.34 (re-tagged): | |
2343 | ||
2344 | - README.html file added to CVS repository | |
2345 | - runtime compiles properly under FreeBSD 3.X and 4.X | |
2346 | ||
2347 | ---------------------------------------------------------------------- | |
2348 | Name: Matthias Blume | |
2349 | Date: 2001/07/10 17:30:00 EDT | |
2350 | Tag: Release_110_34 | |
2351 | Description: | |
2352 | ||
2353 | New version number (110.34). New bootfiles. | |
2354 | ||
2355 | ---------------------------------------------------------------------- | |
2356 | Name: Matthias Blume | |
2357 | Date: 2001/07/09 16:00:00 EDT | |
2358 | Tag: blume-20010709-more-varargs | |
2359 | Description: | |
2360 | ||
2361 | I changed the handling of varargs in ml-nlffigen again: | |
2362 | The ellipsis ... will now simply be ignored (with an accompanying warning). | |
2363 | ||
2364 | The immediate effect is that you can actually call a varargs function | |
2365 | from ML -- but you can't actually supply any arguments beyond the ones | |
2366 | specified explicitly. (For example, you can call printf with its format | |
2367 | string, but you cannot pass additional arguments.) | |
2368 | ||
2369 | This behavior is only marginally more useful than the one before, but | |
2370 | it has the advantage that a function or, more importantly, a function | |
2371 | type never gets dropped on the floor, thus avoiding follow-up problems with | |
2372 | other types that refer to the offending one. | |
2373 | ||
2374 | ---------------------------------------------------------------------- | |
2375 | Name: Matthias Blume | |
2376 | Date: 2001/07/09 11:25:00 EDT | |
2377 | Tag: blume-20010709-varargs | |
2378 | Description: | |
2379 | ||
2380 | 1. ckit-lib.cm now exports structure Error | |
2381 | 2. ml-nlffigen reports occurences of "..." (i.e., varargs function types) | |
2382 | with a warning accompanied by a source location. Moreover, it | |
2383 | merely skips the offending function or type and proceeds with the | |
2384 | rest of its work.u As a result, one can safely feed C code containing | |
2385 | "..." to ml-nlffigen. | |
2386 | 3. There are some internal improvements to CM, providing slightly | |
2387 | more general string substitutions in the tools subsystem. | |
2388 | ||
2389 | ---------------------------------------------------------------------- | |
2390 | Name: Matthias Blume | |
2391 | Date: 2001/06/27 15:10:00 EDT | |
2392 | Tag: blume-20010627-concur | |
2393 | Description: | |
2394 | ||
2395 | Fixed a small bug in CM's handling of parallel compilation. | |
2396 | (You could observe the bug by Control-C-interrupting an ordinary | |
2397 | CMB.make or CM.stabilize and then attaching some compile servers. | |
2398 | The result was that all of a sudden the previously interrupted | |
2399 | compilation would continue on its own. This was because of | |
2400 | an over-optimization: CM did not bother to clean out certain queues | |
2401 | when no servers were attached "anyway", resulting in the contents | |
2402 | of these queues to grab control when new servers did get attached.) | |
2403 | ||
2404 | There is also another minor update to the CM manual. | |
2405 | ||
2406 | ---------------------------------------------------------------------- | |
2407 | Name: Matthias Blume | |
2408 | Date: 2001/06/26 16:15:00 EDT | |
2409 | Tag: blume-20010626-cmdoc | |
2410 | Description: | |
2411 | ||
2412 | Minor typo fixed in CM manual (syntax diagram for libraries). | |
2413 | ||
2414 | ---------------------------------------------------------------------- | |
2415 | Name: Matthias Blume | |
2416 | Date: 2001/06/25 22:55:00 EDT | |
2417 | Tag: blume-20010625-x86pc | |
2418 | Description: | |
2419 | ||
2420 | Fixed a nasty bug in the X86 assembly code that caused signal | |
2421 | handlers to fail (crash) randomly. | |
2422 | ||
2423 | ---------------------------------------------------------------------- | |
2424 | Name: Matthias Blume | |
2425 | Date: 2001/06/25 12:05:00 EDT | |
2426 | Tag: blume-20010625-nlffigen | |
2427 | Description: | |
2428 | ||
2429 | This update fixes a number of minor bugs in ml-nlffigen as reported by | |
2430 | Nick Carter <nbc@andrew.cmu.edu>. | |
2431 | ||
2432 | 1. Silly but ok typedefs of the form "typedef void myvoid;" are now accepted. | |
2433 | 2. Default names for generated files are now derived from the name of | |
2434 | the C file *without its directory*. In particular, this causes generated | |
2435 | files to be placed locally even if the C file is in some system directory. | |
2436 | 3. Default names for generated signatures and structures are also derived | |
2437 | from the C file name without its directory. This avoids silly things | |
2438 | like "structure GL/GL". | |
2439 | (Other silly names are still possible because ml-nlffigen does not do | |
2440 | a thorough check of whether generated names are legal ML identifiers. | |
2441 | When in doubt, use command line arguments to force particular names.) | |
2442 | ||
2443 | ---------------------------------------------------------------------- | |
2444 | Name: Matthias Blume | |
2445 | Date: 2001/06/21 12:25:00 EDT | |
2446 | Tag: blume-20010621-eXene | |
2447 | Description: | |
2448 | ||
2449 | eXene now compiles and (sort of) works again. | |
2450 | ||
2451 | The library name (for version > 110.33) is $/eXene.cm. | |
2452 | ||
2453 | I also added an new example in src/eXene/examples/nbody. See the | |
2454 | README file there for details. | |
2455 | ||
2456 | ---------------------------------------------------------------------- | |
2457 | Name: Matthias Blume | |
2458 | Date: 2001/06/20 16:40:00 EDT | |
2459 | Tag: blume-20010620-cml | |
2460 | Description: | |
2461 | ||
2462 | CML now compiles and works again. | |
2463 | ||
2464 | Libraries (for version > 110.33): | |
2465 | ||
2466 | $cml/cml.cm Main CML library. | |
2467 | $cml/basis.cm CML's version of $/basis.cm. | |
2468 | $cml/cml-internal.cm Internal helper library. | |
2469 | $cml/core-cml.cm Internal helper library. | |
2470 | $cml-lib/trace-cml.cm Tracing facility. | |
2471 | $cml-lib/smlnj-lib.cm CML's version of $/smlnj-lib.cm | |
2472 | ||
2473 | The installer (config/install.sh) has been taught how to properly | |
2474 | install this stuff. | |
2475 | ||
2476 | ---------------------------------------------------------------------- | |
2477 | Name: Matthias Blume | |
2478 | Date: 2001/06/19 17:55:00 EDT | |
2479 | Tag: blume-20010619-instantiate | |
2480 | Description: | |
2481 | ||
2482 | This un-breaks the fix for bug 1432. | |
2483 | (The bug was originally fixed in 110.9 but I broke it again some | |
2484 | time after that.) | |
2485 | ||
2486 | ---------------------------------------------------------------------- | |
2487 | Name: Matthias Blume | |
2488 | Date: 2001/06/19 17:25:00 EDT | |
2489 | Tag: blume-20010619-signals | |
2490 | Description: | |
2491 | ||
2492 | This should (hopefully) fix the long-standing signal handling bug. | |
2493 | (The runtime system was constructing a continuation record with an | |
2494 | incorrect descriptor which would cause the GC to drop data on the floor...) | |
2495 | ||
2496 | ---------------------------------------------------------------------- | |
2497 | Name: Matthias Blume | |
2498 | Date: 2001/06/15 15:05:00 EDT | |
2499 | Tag: blume-20010615-moresparc | |
2500 | Description: | |
2501 | ||
2502 | Here is a short late-hour update related to Sparc c-calls: | |
2503 | ||
2504 | -- made handling of double-word arguments a bit smarter | |
2505 | ||
2506 | -- instruction selection phase tries to collapse certain clumsily | |
2507 | constructed ML-Trees; typical example: | |
2508 | ||
2509 | ADD(ty,ADD(_,e,LI d1),LI d2) -> ADD(ty,e,LI(d1+d2)) | |
2510 | ||
2511 | This currently has no further impact on SML/NJ since mlriscGen does | |
2512 | not seem to generate such patterns in the first place, and c-calls | |
2513 | (which did generate them in the beginning) has meanwhile been fixed | |
2514 | so as to avoid them as well. | |
2515 | ||
2516 | ---------------------------------------------------------------------- | |
2517 | Name: Matthias Blume | |
2518 | Date: 2001/06/15 15:05:00 EDT | |
2519 | Tag: blume-20010615-sparc | |
2520 | Description: | |
2521 | ||
2522 | The purpose of this update is to provide an implementation of NLFFI | |
2523 | on Sparc machines. | |
2524 | ||
2525 | Here are the changes in detail: | |
2526 | ||
2527 | * src/MLRISC/sparc/c-calls/sparc-c-calls.sml is a new file containing | |
2528 | the Sparc implementation of the c-calls API. | |
2529 | * The Sparc backend of SML/NJ has been modified to uniformely use %fp | |
2530 | for accessing the ML frame. Thus, we have a real frame pointer and | |
2531 | can freely modify %sp without need for an omit-frame-ptr phase. | |
2532 | The vfp logic in src/compiler/CodeGen/* has been changed to accomodate | |
2533 | this case. | |
2534 | * ml-nlffigen has been taught to produce code for different architectures | |
2535 | and calling conventions. | |
2536 | * In a way similar to what was done in the x86 case, the Sparc | |
2537 | backend uses its own specific extension to mltree. (For example, | |
2538 | it needs to be able to generate UNIMP instructions which are part | |
2539 | of the calling convention.) | |
2540 | * ml-nlffi-lib was reorganized to make it more modular (in particular, | |
2541 | to make it easier to plug in new machine- and os-dependent parts). | |
2542 | ||
2543 | There are some other fairly unrelated bug fixes and cleanups as well: | |
2544 | ||
2545 | * I further hacked the .cm files for MLRISC tools (like MDLGen) so | |
2546 | that they properly share their libraries with existing SML/NJ libraries. | |
2547 | * I fixed a minor cosmetic bug in CM, supressing certain spurious | |
2548 | follow-up error messages. | |
2549 | * Updates to CM/CMB documentation. | |
2550 | ||
2551 | TODO items: | |
2552 | ||
2553 | * MLRISC should use a different register as its asmTemp on the Sparc. | |
2554 | (The current %o2 is a really bad choice because it is part of the | |
2555 | calling conventions, so things might interfere in unexpected ways.) | |
2556 | ||
2557 | ---------------------------------------------------------------------- | |
2558 | Name: Matthias Blume | |
2559 | Date: 2001/06/07 | |
2560 | Tag: blume-20010607-calls | |
2561 | Description: | |
2562 | ||
2563 | A number of internal changes related to C calls and calling conventions: | |
2564 | ||
2565 | 1. ML-Tree CALL statements now carry a "pops" field. It indicates the | |
2566 | number of bytes popped implicitly (by the callee). In most cases | |
2567 | this field is 0 but on x86/win32 it is some non-zero value. This | |
2568 | is information provided for the benefit of the "omit-frameptr" pass. | |
2569 | 2. The CALL instruction on the x86 carries a similar "pops" field. | |
2570 | The instruction selection phase copies its value from the ML-Tree | |
2571 | CALL statement. | |
2572 | 3. On all other architectures, the instruction selection phase checks | |
2573 | whether "pops=0" and complains if not. | |
2574 | 4. The c-calls implementation for x86 now accepts two calling conventions: | |
2575 | "ccall" and "stdcall". When "ccall" is selected, the caller cleans | |
2576 | up after the call and pops is set to 0. For "stdcall", the caller | |
2577 | does nothing, leaving the cleanup to the callee; pops is set to | |
2578 | the number of bytes that were pushed onto the stack. | |
2579 | 5. The cproto decoder (compiler/Semant/types/cproto.sml) now can | |
2580 | distinguish between "ccall" and "stdcall". | |
2581 | 6. The UNIMP instruction has been added to the supported Sparc instruction | |
2582 | set. (This is needed for implementing the official C calling convention | |
2583 | on this architecture.) | |
2584 | 7. I fixed some of the .cm files under src/MLRISC/Tools to make them | |
2585 | work with the latest CM. | |
2586 | ||
2587 | ---------------------------------------------------------------------- | |
2588 | Name: Matthias Blume | |
2589 | Date: 2001/06/05 15:10:00 EDT | |
2590 | Tag: blume-20010605-cm-index | |
2591 | Description: | |
2592 | ||
2593 | 0. The "lambdasplit" parameter for class "sml" in CM has been documented. | |
2594 | ||
2595 | 1. CM can now generate "index files". These are human-readable files | |
2596 | that list on a per-.cm-file basis each toplevel symbol defined or | |
2597 | imported. The location of the index file for | |
2598 | <p>/<d>.cm is <p>/CM/INDEX/<d>.cm. | |
2599 | To enable index-file generation, set CM.Control.generate_index to true | |
2600 | or export an environment-symbol: export CM_GENERATE_INDEX=true. | |
2601 | ||
2602 | The CM manual has been updated accordingly. | |
2603 | ||
2604 | 2. I made some slight modifications to the c-calls API in MLRISC. | |
2605 | ||
2606 | a) There is now a callback to support saving/restoring of | |
2607 | dedicated but caller-save registers around the actual call | |
2608 | instruction. | |
2609 | b) One can optionally specify a comment-annotation for the | |
2610 | call instruction. | |
2611 | ||
2612 | 3. SML/NJ (mlriscGen.sml) uses this new API for the rawccall primop. | |
2613 | (For example, the comment annotation shows the C prototype of | |
2614 | the function being called.) | |
2615 | ||
2616 | ---------------------------------------------------------------------- | |
2617 | Name: Matthias Blume | |
2618 | Date: 2001/06/01 13:30:00 EDT | |
2619 | Tag: blume-20010601-nlffi-cleanup | |
2620 | Description: | |
2621 | ||
2622 | This is mostly a cleanup of MLFFI stuff: | |
2623 | ||
2624 | - some signature files have been put into a more exposed place | |
2625 | - the ugly 'f type parameter is gone (simplifies types tremendously!) | |
2626 | - ml-nlffigen changed accordingly | |
2627 | - tutorial updated | |
2628 | ||
2629 | Other changes: | |
2630 | ||
2631 | - author's affiliation in CM manual(s) updated | |
2632 | - some more recognized keywords added to Allen's sml.sty | |
2633 | ||
2634 | ---------------------------------------------------------------------- | |
2635 | Name: Matthias Blume | |
2636 | Date: 2001/05/25 15:30:00 EDT | |
2637 | Tag: blume-20010525-iptr | |
2638 | Description: | |
2639 | ||
2640 | - put the official 110.33-README (as it appears on the ftp server) under | |
2641 | CVS | |
2642 | - fixed a small bug related to incomplete pointer types in | |
2643 | ml-nlffigen | |
2644 | - small cosmetic change to the ml-nlffi-lib's "arr" type constructor | |
2645 | (it does not need the 'f type parameter) | |
2646 | ||
2647 | ---------------------------------------------------------------------- | |
2648 | Name: Matthias Blume | |
2649 | Date: 2001/05/23 14:30:00 EDT | |
2650 | Tag: Release_110_33 | |
2651 | Description: | |
2652 | ||
2653 | New version number (110.33). New bootfiles. | |
2654 | ||
2655 | ---------------------------------------------------------------------- | |
2656 | Name: Matthias Blume | |
2657 | Date: 2001/05/22 18:06:00 EDT | |
2658 | Tag: blume-20010522-targets | |
2659 | Description: | |
2660 | ||
2661 | Made install.sh use file config/targets.customized if it exists, falling | |
2662 | back to config/targets if it doesn't. This way one can have a customized | |
2663 | version of the targets file without touching the "real thing", thus | |
2664 | eliminating the constant fear of accidentally checking something bogus | |
2665 | back into the CVS repository... (File config/targets.customized must | |
2666 | not be added to the repository!) | |
2667 | ||
2668 | ---------------------------------------------------------------------- | |
2669 | Name: Matthias Blume | |
2670 | Date: 2001/05/22 16:30:00 EDT | |
2671 | Tag: blume-20010522-minitut | |
2672 | Description: | |
2673 | ||
2674 | 1. Bug fix in ml-nlffigen; now (hopefully) correctly handling | |
2675 | struct returns. | |
2676 | 2. Added src/ml-nlffi-lib/Doc/mini-tutorial.txt. This is some very | |
2677 | incomplete, preliminary documentation for NLFFI. | |
2678 | ||
2679 | ---------------------------------------------------------------------- | |
2680 | Name: Matthias Blume | |
2681 | Date: 2001/05/14 11:30:00 EDT | |
2682 | Tag: blume-20010514-script | |
2683 | Description: | |
2684 | ||
2685 | Some bugs in install script fixed. | |
2686 | ||
2687 | In addition to that I also made a slight change to the NLFFI API: | |
2688 | Functors generated by ml-nlffigen now take the dynamic library as a | |
2689 | straight functor argument, not as a suspended one. (The original | |
2690 | functor code used to force the suspension right away anyway, so there | |
2691 | was nothing gained by this complication of the interface.) | |
2692 | ||
2693 | ---------------------------------------------------------------------- | |
2694 | Name: Matthias Blume | |
2695 | Date: 2001/05/11 14:35:00 EDT | |
2696 | Tag: blume-20010511-ml-nlffi | |
2697 | Description: | |
2698 | ||
2699 | I finally took the plunge and added my new FFI code to the main | |
2700 | repository. For x86-linux it is now ready for prime-time. | |
2701 | ||
2702 | There are two new subdirectories of "src": | |
2703 | ||
2704 | - ml-nlffi-lib: | |
2705 | The utility library for programs using the FFI interface. | |
2706 | Here is the implementation of $/c.cm and its associated low-level | |
2707 | partners $/c-int.cm and $/memory.cm. | |
2708 | - ml-nlffigen: | |
2709 | A stand-alone program for generating ML glue code from C source | |
2710 | code. | |
2711 | ||
2712 | Building ml-nlffigen requires $/ckit-lib.cm. | |
2713 | ||
2714 | The config/install.sh script has been updates to do the Right Thing | |
2715 | (hopefully). | |
2716 | ||
2717 | Notice that the source tree for the C-Kit will not be put under "src" | |
2718 | but directly under the installation root directory. (This is the | |
2719 | structure that currently exists on the CVS server when you check out | |
2720 | module "sml".) Fortunately, config/install.sh knows about this oddity. | |
2721 | ||
2722 | Bugs: No documentation yet. | |
2723 | ||
2724 | ---------------------------------------------------------------------- | |
2725 | Name: Matthias Blume | |
2726 | Date: 2001/05/09 16:35:00 EDT | |
2727 | Tag: blume-20010509-cpscontract | |
2728 | Description: | |
2729 | ||
2730 | Fixed a bug in the accounting code in cpsopt/contract.sml. (The | |
2731 | wrapper/unwrapper elimination did not decrement usage counts and some | |
2732 | dead variables got overlooked by the dead-up logic.) | |
2733 | ||
2734 | ---------------------------------------------------------------------- | |
2735 | Name: Lal George | |
2736 | Date: 2001/05/08 17:26:09 EDT | |
2737 | Tag: george-20010508-omit-frameptr | |
2738 | Description: | |
2739 | ||
2740 | Changes to implement the omit-frame-pointer optimization to support | |
2741 | raw C calls. For now, there is only support on the Intel x86, but | |
2742 | other architectures will follow as more experience is gained with this. | |
2743 | ||
2744 | ||
2745 | ---------------------------------------------------------------------- | |
2746 | Name: Matthias Blume | |
2747 | Date: 2001/05/07 14:40:00 EDT | |
2748 | Tag: blume-20010507-proxies | |
2749 | Description: | |
2750 | ||
2751 | I made into "proxy libraries" all libraries that qualify for such a | |
2752 | change. (A qualifying library is a library that has another library or | |
2753 | groups as its sole member and repeats that member's export list | |
2754 | verbatim. A proxy library avoids this repetition by omitting its export | |
2755 | list, effectively inheriting the list that its (only) member exports. | |
2756 | See the CM manual for more explanation.) | |
2757 | The main effect is that explicit export lists for these libraries | |
2758 | do not have to be kepts in sync, making maintenance a bit easier. | |
2759 | ||
2760 | I also added copyright notices to many .cm-files. | |
2761 | ||
2762 | Last but not least, I made a new set of bootfiles. | |
2763 | ||
2764 | ---------------------------------------------------------------------- | |
2765 | Name: Matthias Blume | |
2766 | Date: 2001/05/04 17:00:00 EDT | |
2767 | Tag: blume-20010504-cm-lsplit | |
2768 | Description: | |
2769 | ||
2770 | 0. John merged pending changes to $/smlnj-lib.cm | |
2771 | ||
2772 | 1. Allen's previous change accidentally backed out of one of Lal's | |
2773 | earlier changes. I undid this mistake (re-introducing Lal's change). | |
2774 | ||
2775 | 2. I used the new topOrder' function from graph-scc.sml (from $/smlnj-lib.cm) | |
2776 | within the compiler where applicable. There is some code simplification | |
2777 | because of that. | |
2778 | ||
2779 | 3. The "split" phase (in FLINT) is now part of the default list of phases. | |
2780 | Compiler.Control.LambdaSplitting.* can be used to globally control the | |
2781 | lambda-splitting (cross-module-inlining) engine. In addition to that, | |
2782 | it can now also be controlled on a per-source basis: CM has been taught | |
2783 | a new tool parameter applicable to ML source files. | |
2784 | ||
2785 | - To turn lambda-splitting off completely: | |
2786 | local open Compiler.Control.LambdaSplitting in | |
2787 | val _ = set Off | |
2788 | end | |
2789 | - To make "no lambda-splitting" the global default (but allow per-source | |
2790 | overriding); this is the initial setting: | |
2791 | local open Compiler.Control.LambdaSplitting in | |
2792 | val _ = set (Default NONE) | |
2793 | end | |
2794 | - To make "lambda-splitting with aggressiveness a" the global default | |
2795 | (and allow per-source overriding): | |
2796 | local open Compiler.Control.LambdaSplitting in | |
2797 | val _ = set (Default (SOME a)) | |
2798 | end | |
2799 | ||
2800 | - To turn lambda-splitting off for a given ML souce file (say: a.sml) | |
2801 | write (in the respective .cm-file): | |
2802 | a.sml (lambdasplitting:off) | |
2803 | - To turn lambda-splitting for a.sml on with minimal aggressiveness: | |
2804 | a.sml (lambdasplitting:on) | |
2805 | - To turn lambda-splitting for a.sml on with aggressiveness <a> (where | |
2806 | <a> is a decimal non-negative integer): | |
2807 | a.sml (lambdasplitting:<a>) | |
2808 | - To turn lambda-splitting for a.sml on with maximal aggressiveness: | |
2809 | a.sml (lambdasplitting:infinity) | |
2810 | - To use the global default for a.sml: | |
2811 | a.sml (lambdasplitting:default) | |
2812 | or simply | |
2813 | a.sml | |
2814 | ||
2815 | ---------------------------------------------------------------------- | |
2816 | Name: Allen Leung | |
2817 | Date: 2001/05/04 01:57:00 EDT | |
2818 | Tag: leunga-20010504-sync | |
2819 | Description: | |
2820 | ||
2821 | MLRISC features. | |
2822 | ||
2823 | 1. Fix to CMPXCHG instructions. | |
2824 | 2. Changed RA interface to allow annotations in callbacks. | |
2825 | 3. Added a new method to the stream interface to allow annotations updates. | |
2826 | ||
2827 | ---------------------------------------------------------------------- | |
2828 | Name: Matthias Blume | |
2829 | Date: 2001/05/01 11:45:00 EDT | |
2830 | Tag: blume-20010501-pcedittmp | |
2831 | Description: | |
2832 | ||
2833 | Changed install.sh to use the current working directory instead of | |
2834 | /usr/tmp for a temporary file (pcedittmp). The previous choice | |
2835 | of /usr/tmp caused trouble with MacOS X because of file premission | |
2836 | problems. | |
2837 | ||
2838 | ---------------------------------------------------------------------- | |
2839 | Name: Matthias Blume | |
2840 | Date: 2001/04/20 11:10:00 EDT | |
2841 | Tag: blume-20010420-inMLflag | |
2842 | Description: | |
2843 | ||
2844 | - added vp_limitPtrMask to vproc-state.h | |
2845 | (for use by the raw-C-calls mechanism to implement proper interrupt | |
2846 | handling) | |
2847 | - made the ML compiler aware of various data-structure offsets so it | |
2848 | can generate code for accessing the vp_inML flag and vp_limitPtrMask | |
2849 | - tweaked mlriscGen.sml to have it emit interrupt-handling code for | |
2850 | raw C-calls | |
2851 | ||
2852 | ---------------------------------------------------------------------- | |
2853 | Name: Lal George | |
2854 | Date: 2001/04/20 09:15:28 EDT | |
2855 | Tag: george-20010420-macosX | |
2856 | Description: | |
2857 | ||
2858 | - Changes to port to Mac OS X; Darwin. | |
2859 | ||
2860 | - In the process I found that sqrt was broken on the PPC, because the | |
2861 | fsqrt instruction is not implemented. | |
2862 | ||
2863 | ---------------------------------------------------------------------- | |
2864 | Name: Matthias Blume | |
2865 | Date: 2001/04/18 12:45:00 EDT | |
2866 | Tag: blume-20010418-ccalls | |
2867 | Description: | |
2868 | ||
2869 | - fixed two off-by-4 errors in the x86-specific c-calls implementation | |
2870 | (this bug prevented structure arguments containing pointers from being | |
2871 | passed correctly) | |
2872 | - changed the raw-C-call code in mlriscGen.sml in such a way that | |
2873 | structure arguments are represented as a pointer to the beginning | |
2874 | of the structure (instead of having a series of synthesized arguments, | |
2875 | one for each structure member) | |
2876 | ||
2877 | - made makeml script's verbosity level configurable via environment | |
2878 | variable (MAKEML_VERBOSITY) | |
2879 | ||
2880 | - eliminated placeholder implementations for f32l, w16s, i16s, and f32s | |
2881 | in rawmem-x86.sml; we are now using the real thing | |
2882 | ||
2883 | ---------------------------------------------------------------------- | |
2884 | Name: Matthias Blume | |
2885 | Date: 2001/03/22 16:25:00 EST | |
2886 | Tag: blume-20010322-bootfiles | |
2887 | Description: | |
2888 | ||
2889 | Created a new set of bootfiles (for your automatic installation convenience). | |
2890 | ||
2891 | ---------------------------------------------------------------------- | |
2892 | Name: Matthias Blume | |
2893 | Date: 2001/03/22 15:10:00 EST | |
2894 | Tag: blume-20010322-rawmem-parcm | |
2895 | Description: | |
2896 | ||
2897 | 1. All "raw memory access" primitives for the new FFI are implemented now | |
2898 | (at least on the x86). | |
2899 | 2. Some further cleanup of CM's parallel make mechanism. | |
2900 | ||
2901 | ---------------------------------------------------------------------- | |
2902 | Name: Matthias Blume | |
2903 | Date: 2001/03/19 17:53:00 EST | |
2904 | Tag: blume-20010319-parallel | |
2905 | Description: | |
2906 | ||
2907 | Parallel make (using compile servers) now works again. | |
2908 | ||
2909 | To this end, CM.stabilize and CMB.make have been modified to work in | |
2910 | two passes when compile servers are attached: | |
2911 | 1. Compile everything, do not perform stabilization; this pass | |
2912 | uses compile servers | |
2913 | 2. Stabilize everything; this pass does not use compile servers | |
2914 | If there are no compile servers, the two passes are combined into one | |
2915 | (as before). Splitting the passes increases the inherent parallelism | |
2916 | in the dependency graph because the entire graph including all | |
2917 | libraries is available at the same time. This, in turn, improves | |
2918 | server utilization. The downside is that the master process will | |
2919 | have to do some extra work after compilation is done (because for | |
2920 | technical reasons it must re-read all the binfiles during stabilization). | |
2921 | ||
2922 | ---------------------------------------------------------------------- | |
2923 | Name: Matthias Blume | |
2924 | Date: 2001/03/16 12:22:00 EST | |
2925 | Tag: blume-20010316-bootfiles | |
2926 | Description: | |
2927 | ||
2928 | Created a new set of bootfiles (for your automatic installation convenience). | |
2929 | ||
2930 | ---------------------------------------------------------------------- | |
2931 | Name: Matthias Blume | |
2932 | Date: 2001/03/16 11:00:00 EST | |
2933 | Tag: blume-20010316-MLTREE-fixup | |
2934 | Description: | |
2935 | ||
2936 | This is a minor fixup for an (untagged) earlier commit by Allen. | |
2937 | (A file was missing). | |
2938 | ||
2939 | ---------------------------------------------------------------------- | |
2940 | Name: Allen Leung | |
2941 | Date: Mon Mar 5 18:54:57 EST 2001 | |
2942 | Tag: leunga-20010305-cut-support | |
2943 | ||
2944 | 1. New support for alternative control-flow in MLTREE. | |
2945 | Currently we support | |
2946 | ||
2947 | FLOW_TO(CALL ...., [k1,...,kn]) | |
2948 | ||
2949 | This is needed for 'cuts to' in C-- and try/handle-like constructs | |
2950 | in Moby | |
2951 | ||
2952 | New assembler flag "asm-show-cutsto" to turn on control-flow debugging. | |
2953 | ||
2954 | 2. Register Allocator | |
2955 | ||
2956 | Changes in interface [from Fermin, John] | |
2957 | ||
2958 | 3. Alpha 8-bit SLL support [Fermin] | |
2959 | ||
2960 | 4. All architectures | |
2961 | ||
2962 | A new module (ClusterExpandCopies) for expanding parallel copies. | |
2963 | ||
2964 | ---------------------------------------------------------------------- | |
2965 | Name: Allen Leung | |
2966 | Date: 2001/02/27 23:07:00 EST | |
2967 | Tag: leunga-20010227-minor-stuff | |
2968 | ||
2969 | 1. Alpha bug fix for CMOVNE | |
2970 | 2. Handle mltree COND(..,FCMP ...,...) | |
2971 | 3. Bug fix in simplifier | |
2972 | ||
2973 | ---------------------------------------------------------------------- | |
2974 | Name: Matthias Blume | |
2975 | Date: 2001/01/30 17:50:00 EST | |
2976 | Tag: blume-20010130-sync | |
2977 | Description: | |
2978 | ||
2979 | This is just a minor update to sync my devel branch with the main brach. | |
2980 | The only visible change is the addition of some README files. | |
2981 | ||
2982 | ---------------------------------------------------------------------- | |
2983 | Name: Matthias Blume | |
2984 | Date: 2001/01/12 23:30:00 JST | |
2985 | Tag: blume-20010112-bootfiles | |
2986 | Description: | |
2987 | ||
2988 | Made a new set of bootfiles that goes with the current state of the | |
2989 | repository. | |
2990 | ||
2991 | ---------------------------------------------------------------------- | |
2992 | Name: Matthias Blume | |
2993 | Date: 2001/01/12 21:20:00 JST | |
2994 | Tag: blume-20010112-sync | |
2995 | Description: | |
2996 | ||
2997 | I am just flushing out some minor changes that had accumulated in | |
2998 | my private branch in order to sync with the main tree. (This is | |
2999 | mainly because I had CVS trouble when trying to merge _into_ my | |
3000 | private branch.) | |
3001 | ||
3002 | Most people should be completely unaffected by this. | |
3003 | ||
3004 | ---------------------------------------------------------------------- | |
3005 | Name: Allen Leung | |
3006 | Date: Thu Jan 11 21:03:00 EST 2001 | |
3007 | Tag: leunga-20010111-labexp=mltree | |
3008 | Description: | |
3009 | ||
3010 | 1. Removed the type LabelExp and replace it by MLTree. | |
3011 | 2. Rewritten mltree-simplify with the pattern matcher tool. | |
3012 | 3. There were some bugs in alpha code generator which would break | |
3013 | 64-bit code generation. | |
3014 | 4. Redo the tools to generate code with the | |
3015 | 5. The CM files in MLRISC (and in src/system/smlnj/MLRISC) | |
3016 | are now generated by perl scripts. | |
3017 | ||
3018 | ---------------------------------------------------------------------- | |
3019 | Name: Matthias Blume | |
3020 | Date: 2001/01/10 21:55:00 JST | |
3021 | Tag: blume-20010110-rcc | |
3022 | Description: | |
3023 | ||
3024 | The RCC stuff now seems to work (but only on the x86). | |
3025 | This required hacking of the c-calls interface (and -implementation) in | |
3026 | MLRISC. | |
3027 | ||
3028 | Normal compiler users should be unaffected. | |
3029 | ||
3030 | ---------------------------------------------------------------------- | |
3031 | Name: Matthias Blume | |
3032 | Date: 2001/01/09 01:20:00 JST | |
3033 | Tag: blume-20010109-rcc | |
3034 | Description: | |
3035 | ||
3036 | This is a fairly big patch, flushing out a large number of pending | |
3037 | changes that I made to my development copy over the last couple of days. | |
3038 | ||
3039 | Of practical relevance at this moment is a workaround for a pickling | |
3040 | bug that Allen ran into the other day. The cause of the bug itself is | |
3041 | still unknown and it might be hard to fix it properly, but the | |
3042 | workaround has some merits of its own (namely somewhat reducing pickling | |
3043 | overhead for certain libraries). Therefore, I think this solution should | |
3044 | be satisfactory at this time. | |
3045 | ||
3046 | The rest of the changes (i.e., the vast majority) has to do with my | |
3047 | ongoing efforts of providing direct support for C function calls from | |
3048 | ML. At the moment there is a new primop "RAW_CCALL", typing magic | |
3049 | in types/cproto.sml (invoked from FLINT/trans/translate.sml), a new | |
3050 | case in the FLINT CPS datatype (RCC), changes to cps/convert.sml to | |
3051 | translate uses of RAW_CCALL into RCC, and changes to mlriscGen.sml to | |
3052 | handle RCC. | |
3053 | ||
3054 | The last part (the changes to mlriscGen.sml) are still known to be | |
3055 | wrong on the x86 and not implemented on all other architectures. But | |
3056 | the infrastructure is in place. I had to change a few functor | |
3057 | signatures in the backend to be able to route the CCalls interface | |
3058 | from MLRISC there, and I had to specialize the mltree type (on the | |
3059 | x86) to include the necessary extensions. (The extensions themselves | |
3060 | were already there and redy to go in MLRISC/x86). | |
3061 | ||
3062 | Everything should be very happy as soon as someone helps me with | |
3063 | mlriscGen.sml... | |
3064 | ||
3065 | In any case, nothing of this should matter to anyone as long as the | |
3066 | new primop is not being used (which is going to be the case unless you | |
3067 | find it where I hid it :). The rest of the compiler is completely | |
3068 | unaffected. | |
3069 | ||
3070 | ---------------------------------------------------------------------- | |
3071 | Name: Matthias Blume | |
3072 | Date: 2001/01/05 00:30:00 JST | |
3073 | Tag: blume-20010105-primops | |
3074 | Description: | |
3075 | ||
3076 | Added some experimental support for work that I am doing right now. | |
3077 | These changes mostly concern added primops, but there is also a new | |
3078 | experimental C library in the runtime system (but currently not enabled | |
3079 | anywhere except on Linux/X86). | |
3080 | ||
3081 | In the course of adding primops (and playing with them), I discovered that | |
3082 | Zhong's INL_PRIM hack (no type info for certain primops) was, in fact, badly | |
3083 | broken. (Zhong was very right he labeled this stuff as "major gross hack".) | |
3084 | To recover, I made type information in INL_PRIM mandatory and changed | |
3085 | prim.sml as well as built-in.sml accordingly. The InLine structure now | |
3086 | has complete, correct type information (i.e., no bottom types). | |
3087 | ||
3088 | Since all these changes mean that we need new binfiles, I also bumped the | |
3089 | version number to 110.32.1. | |
3090 | ||
3091 | ---------------------------------------------------------------------- | |
3092 | Name: Matthias Blume | |
3093 | Date: 2000/12/30 22:10:00 JST | |
3094 | Tag: blume-20001230-various | |
3095 | Description: | |
3096 | ||
3097 | Added proxy libraries for MLRISC and let MLRISC libraries refer | |
3098 | to each other using path anchors. (See CM manual for explanation.) | |
3099 | ||
3100 | Updated CM documentation. | |
3101 | ||
3102 | Fixed some bugs in CM. | |
3103 | ||
3104 | Implemented "proxy" libraries (= syntactic sugar for CM). | |
3105 | ||
3106 | Added "-quiet" option to makeml and changed runtime system accordingly. | |
3107 | ||
3108 | Added cleanup handler for exportML to reset timers and compiler stats. | |
3109 | ||
3110 | ---------------------------------------------------------------------- | |
3111 | Name: Lal George | |
3112 | Date: 2000/12/22 22:22:58 EST 2000 | |
3113 | Tag: Release_110_32 | |
3114 | Description: | |
3115 | ||
3116 | Infinite precision used throughout MLRISC. | |
3117 | see MLRISC/mltree/machine-int.sig | |
3118 | ||
3119 | ---------------------------------------------------------------------- | |
3120 | Name: Matthias Blume | |
3121 | Date: 2000/12/22 23:16:00 JST | |
3122 | Tag: blume-20001222-warn | |
3123 | Description: | |
3124 | ||
3125 | Corrected wording and formatting of some CM warning message which I | |
3126 | broke in my previous patch. | |
3127 | ||
3128 | ---------------------------------------------------------------------- | |
3129 | Name: Matthias Blume | |
3130 | Date: 2000/12/22 21:20:00 JST | |
3131 | Tag: blume-20001222-anchorenv | |
3132 | Description: | |
3133 | ||
3134 | Fixed CM's handling of anchor environments in connection with CMB.make. | |
3135 | ||
3136 | ---------------------------------------------------------------------- | |
3137 | Name: Matthias Blume | |
3138 | Date: 2000/12/22 13:15:00 JST | |
3139 | Tag: blume-20001222-cleanup | |
3140 | Description: | |
3141 | ||
3142 | Removed src/cm/ffi which does not (and did not) belong here. | |
3143 | ||
3144 | ---------------------------------------------------------------------- | |
3145 | Name: Matthias Blume | |
3146 | Date: 2000/12/21 23:55:00 JST | |
3147 | Tag: blume-20001221-exn | |
3148 | Description: | |
3149 | ||
3150 | Probably most important: CM no longer silently swallows all exceptions | |
3151 | in the compiler. | |
3152 | Plus: some other minor CM changes. For example, CM now reports some | |
3153 | sizes for generated binfiles (code, data, envpickle, lambdapickle). | |
3154 | ||
3155 | ---------------------------------------------------------------------- | |
3156 | Name: Matthias Blume | |
3157 | Date: 2000/12/15 00:01:05 JST | |
3158 | Tag: blume-20001215-dirtool | |
3159 | Description: | |
3160 | ||
3161 | - "dir" tool added. | |
3162 | - improvements and cleanup to Tools structure | |
3163 | - documentation updates | |
3164 | ||
3165 | ---------------------------------------------------------------------- | |
3166 | Name: Allen Leung | |
3167 | Date: Thu Dec 14 03:45:24 EST 2000 | |
3168 | Description: | |
3169 | Tag: leunga-20001214-int-inf | |
3170 | Description: | |
3171 | ||
3172 | In IntInf, added these standard functions, which are missing from our | |
3173 | implementation: | |
3174 | ||
3175 | andb : int * int -> int | |
3176 | xorb : int * int -> int | |
3177 | orb : int * int -> int | |
3178 | notb : int -> int | |
3179 | << : int * word -> int | |
3180 | ~>> : int * word -> int | |
3181 | ||
3182 | Not tested, I hope they are correct. | |
3183 | ||
3184 | ---------------------------------------------------------------------- | |
3185 | Name: Allen Leung | |
3186 | Date: Fri Dec 8 19:23:26 EST 2000 | |
3187 | Description: | |
3188 | Tag: leunga-20001208-nowhere | |
3189 | Description: | |
3190 | ||
3191 | Slight improvements to the 'nowhere' tool to handle OR-patterns, | |
3192 | to generate better error messages etc. Plus a brief manual. | |
3193 | ||
3194 | ---------------------------------------------------------------------- | |
3195 | Name: Lal George | |
3196 | Date: 2000/12/08 09:54:02 EST 2000 | |
3197 | Tag: Release_110_31 | |
3198 | Description: | |
3199 | ||
3200 | - Version 110.31 | |
3201 | ---------------------------------------------------------------------- | |
3202 | Name: Allen Leung | |
3203 | Date: Thu Dec 7 22:01:04 EST 2000 | |
3204 | Tag: leunga-20001207-cell-monster-hack | |
3205 | Description: | |
3206 | ||
3207 | Major MLRISC internal changes. Affect all clients. | |
3208 | Summary: | |
3209 | ||
3210 | 1. Type CELLS.cell = int is now replaced by a datatype. | |
3211 | As a result, the old regmap is now gone. Almost all interfaces | |
3212 | in MLRISC change as a consequence. | |
3213 | ||
3214 | 2. A new brand version of machine description tool (v3.0) that generates | |
3215 | modules expecting the new interface. The old version is removed. | |
3216 | ||
3217 | 3. The RA interface has been further abstracted into two new functors. | |
3218 | RISC_RA and X86RA. These functors have much simpler interfaces. | |
3219 | [See also directory MLRISC/demo.] | |
3220 | ||
3221 | 4. Some other new source->source code generation tools are available: | |
3222 | ||
3223 | a. MLRISC/Tools/RewriteGen -- generate rewriters from rules. | |
3224 | b. MLRISC/Tools/WhereGen -- expands conditional pattern matching rules. | |
3225 | I use this tool to generate the peephole optimizers---with the new | |
3226 | cell type changes, peephole rules are becoming difficult to write | |
3227 | without conditional pattern matching. | |
3228 | ||
3229 | 5. More Intmap -> IntHashTable change. Previous changes by Matthias didn't | |
3230 | cover the entire MLRISC source tree so many things broke. | |
3231 | ||
3232 | 6. CM files have been moved to the subdirectory MLRISC/cm. | |
3233 | They are moved because there are a lot of them and they clutter up the | |
3234 | root dir. | |
3235 | ||
3236 | 7. More detailed documentation to come... | |
3237 | ||
3238 | NOTE: To rebuild from 110.30 (ftp distribution), you'll have to do | |
3239 | a makeml -rebuild first. This is because of other other | |
3240 | changes that Matthias has made (see below). | |
3241 | ||
3242 | ||
3243 | ---------------------------------------------------------------------- | |
3244 | Name: Matthias Blume | |
3245 | Date: 2000/11/30 23:12:00 JST | |
3246 | Tag: blume-20001130-filereorg | |
3247 | Description: | |
3248 | ||
3249 | Some manual updates and some file reorganizations in CM. | |
3250 | ||
3251 | ---------------------------------------------------------------------- | |
3252 | Name: Matthias Blume | |
3253 | Date: 2000/11/24 17:45:00 JST | |
3254 | Tag: blume-20001124-link | |
3255 | Description: | |
3256 | ||
3257 | Drastically improved link traversal code for the case that the dynamic | |
3258 | value was already loaded at bootstrap time. As a result, CM and CMB | |
3259 | now both load blazingly fast -- even on a very slow machine. Also, | |
3260 | memory consumption has been further reduced by this. | |
3261 | ||
3262 | Warning: The format of the PIDMAP file has changed. THerefore, to | |
3263 | bootstrap you have to do this: | |
3264 | ||
3265 | 1. Run CMB.make | |
3266 | 2. Make a symbolic link for the boot directory: | |
3267 | ln -s sml.boot.ARCH-OS xxx | |
3268 | 3. "Rebuild" the boot directory: | |
3269 | ./makeml -boot xxx -rebuild sml ; rm xxx | |
3270 | 4. Boot normally: | |
3271 | ./makeml | |
3272 | ||
3273 | ---------------------------------------------------------------------- | |
3274 | Name: Matthias Blume | |
3275 | Date: 2000/11/21 21:20:00 JST | |
3276 | Tag: blume-20001121-tools | |
3277 | Description: | |
3278 | ||
3279 | Continued hacking on autoloading problem -- with success this time. | |
3280 | Also changed tool-plugin mechanism. See new CM manual. | |
3281 | ||
3282 | ---------------------------------------------------------------------- | |
3283 | Name: Matthias Blume | |
3284 | Date: 2000/11/19 14:30:00 JST | |
3285 | Tag: blume-20001119-autoload | |
3286 | Description: | |
3287 | ||
3288 | Some hacking to make autoloading faster. Success for CMB, no success | |
3289 | so far for CM. There is a reduced structure CM' that autoloads faster. | |
3290 | (This is a temporary, non-documented hack to be eliminated again when | |
3291 | the general problem is solved.) | |
3292 | ||
3293 | ---------------------------------------------------------------------- | |
3294 | Name: Matthias Blume | |
3295 | Date: 2000/11/17 14:10:00 JST | |
3296 | Tag: blume-20001117-pickle-lib | |
3297 | Description: | |
3298 | ||
3299 | 1. Eliminated comp-lib.cm | |
3300 | 2. Made pickle-lib.cm | |
3301 | 3. Eliminated all uses of intset.sml (from comp-lib.cm) | |
3302 | 4. Replaced all uses of intmap.{sig,sml} (from comp-lib.cm) with | |
3303 | equivalent constructs from smlnj-lib.cm (INtHashTable). | |
3304 | 5. Point 4. also goes for those uses of intmap.* in MLRISC. | |
3305 | Duplicated intmap modules thrown out. | |
3306 | 6. Hunted down all duplicated SCC code and replaced it with | |
3307 | equivalent stuff (GraphSCCFn from smlnj-lib.cm). | |
3308 | 7. Rewrote Feedback module. | |
3309 | 8. Moved sortedlist.sml into viscomp-lib.cm. Eventually it | |
3310 | should be thrown out and equivalent modules from smlnj-lib.cm | |
3311 | should be used (IntRedBlackSet, IntListSet, ...). | |
3312 | ||
3313 | Confirmed that compiler compiles to fixpoint. | |
3314 | ||
3315 | ---------------------------------------------------------------------- | |
3316 | Name: Allen Leung | |
3317 | Date: 2000/11/10 18:00:00 | |
3318 | Tag: leunga-20001110-new-x86-fp | |
3319 | ||
3320 | A new x86 floating point code generator has been added. | |
3321 | By default this is turned off. To turn this on, do: | |
3322 | ||
3323 | CM.autoload "$smlnj/compiler.cm"; | |
3324 | Compiler.Control.MLRISC.getFlag "x86-fast-fp" := true; | |
3325 | ||
3326 | Changes: | |
3327 | ||
3328 | 1. Changed FTAN to FPTAN so that the assembly output is correct. | |
3329 | 2. Changed the extension callback for FTANGENT to generate: | |
3330 | ||
3331 | fptan | |
3332 | fstp %st(0) | |
3333 | instead of | |
3334 | fptan | |
3335 | fstpl ftempmem | |
3336 | ||
3337 | 3. Numerous assembly fixes for x86. | |
3338 | ||
3339 | 5. Cleaned up the machine code output module x86/x86MC.sml and added | |
3340 | support for a whole bunch of instructions and addressing modes: | |
3341 | ||
3342 | fadd/fsub/fsubr/fmul/fdiv/fdivr %st, %st(n) | |
3343 | faddp/fsubp/fsubrp/fmulp/fdivp/fdivrp %st, %st(n) | |
3344 | fadd/fsub/fsubr/fmul/fdiv/fdivr %st(n), %st | |
3345 | fiadd/fisub/fisubr/fimul/fidiv/fidivr mem | |
3346 | fxch %st(n) | |
3347 | fld %st(n) | |
3348 | fst %st(n) | |
3349 | fst mem | |
3350 | fstp %st(n) | |
3351 | fucom %st(n) | |
3352 | fucomp %st(n) | |
3353 | ||
3354 | All these are now generated when the fast fp mode is turned on. | |
3355 | ||
3356 | 6. Removed the dedicated registers %st(0), ..., %st(7) from X86CpsRegs | |
3357 | ||
3358 | ---------------------------------------------------------------------- | |
3359 | Name: Matthias Blume | |
3360 | Date: 2000/11/09 11:20:00 JST | |
3361 | Tag: blume-20001109-scc | |
3362 | Description: | |
3363 | ||
3364 | Eliminated some code duplication: | |
3365 | ||
3366 | 1. Added "where" clause to GraphSCCFn in SML/NJ Library. | |
3367 | (Otherwise the functor is useless.) | |
3368 | 2. Used GraphSCCFn where SCCUtilFun was used previously. | |
3369 | 3. Got rid of SCCUtilFun (in comp-lib.cm). | |
3370 | ||
3371 | ---------------------------------------------------------------------- | |
3372 | Name: Lal George | |
3373 | Date: 2000/11/06 09:02:21 EST 2000 | |
3374 | Tag: Release_110_30 | |
3375 | Description: | |
3376 | ||
3377 | - Version 110.30 | |
3378 | ---------------------------------------------------------------------- | |
3379 | Name: Matthias Blume | |
3380 | Date: 2000/11/04 14:45:00 | |
3381 | Tag: blume-20001104-mlbuild | |
3382 | Description: | |
3383 | ||
3384 | - Made ml-build faster on startup. | |
3385 | - Documentation fixes. | |
3386 | ||
3387 | ---------------------------------------------------------------------- | |
3388 | Name: Matthias Blume | |
3389 | Date: 2000/11/02 17:00:00 JST | |
3390 | Tag: blume-20001102-condcomp | |
3391 | Description: | |
3392 | ||
3393 | - Small tweaks to pickler -- new BOOTFILES! | |
3394 | - Version bumped to 110.29.2. | |
3395 | - Added conditional compilation facility to init.cmi (see comment there). | |
3396 | ---------------------------------------------------------------------- | |
3397 | Name: Allen Leung | |
3398 | Date: 2000/10/23 19:31:00 | |
3399 | Tag: leunga-20001023-demo-ra | |
3400 | ||
3401 | 1. Minor RA changes that improves spilling on x86 (affects Moby and C-- only) | |
3402 | 2. Test programs for the graph library updated | |
3403 | 3. Some new MLRISC demo programs added | |
3404 | ||
3405 | ---------------------------------------------------------------------- | |
3406 | Name: Matthias Blume | |
3407 | Date: 2000/08/31 22:15:00 JST | |
3408 | Tag: blume-20001017-errmsg | |
3409 | Description: | |
3410 | ||
3411 | More error message grief: Where there used to be no messages, there | |
3412 | now were some that had bogus error regions. Fixed. | |
3413 | ||
3414 | ---------------------------------------------------------------------- | |
3415 | Name: Matthias Blume | |
3416 | Date: 2000/08/31 17:30:00 JST | |
3417 | Tag: blume-20001017-v110p29p1 | |
3418 | Description: | |
3419 | ||
3420 | I made a version 110.29.1 with new bootfiles. | |
3421 | ||
3422 | Changes: Modified pickler/unpickler for faster and leaner unpickling. | |
3423 | CM documentation changes and a small bugfix in CM's error reporting. | |
3424 | ||
3425 | ---------------------------------------------------------------------- | |
3426 | Name: Lal George | |
3427 | Date: 2000/09/27 14:42:35 EDT | |
3428 | Tag: george-20000927-nodestatus | |
3429 | Description: | |
3430 | ||
3431 | Changed the type of the nodestatus, so that: | |
3432 | ||
3433 | SPILLED(~1) is now SPILLED | |
3434 | SPILLED(m) where m>=0 is now MEMREG(m) | |
3435 | SPILLED(s) where s<~1 is now SPILL_LOC(~s) | |
3436 | ||
3437 | ---------------------------------------------------------------------- | |
3438 | Name: Matthias Blume | |
3439 | Date: 2000/09/07 14:45:00 JST | |
3440 | Tag: blume-20000907-cmerrmsg | |
3441 | Description: | |
3442 | ||
3443 | Small tweak to CM to avoid getting ML syntax error messages twice. | |
3444 | ||
3445 | ---------------------------------------------------------------------- | |
3446 | Name: Matthias Blume | |
3447 | Date: 2000/08/31 18:00:00 JST | |
3448 | Tag: blume-20000831-cvsbootfiles | |
3449 | Description: | |
3450 | ||
3451 | New URL for boot files (because the 110.29 files on the BL server do | |
3452 | now work correctly with my updated install scripts for yacc and lex). | |
3453 | ||
3454 | ---------------------------------------------------------------------- | |
3455 | Name: Matthias Blume | |
3456 | Date: 2000/08/08 12:33:00 JST | |
3457 | Tag: blume-20000808-manual | |
3458 | Description: | |
3459 | ||
3460 | Tiny update to CM manual. | |
3461 | ||
3462 | ---------------------------------------------------------------------- | |
3463 | Name: Allen Leung | |
3464 | Date: 2000/08/7 19:31:00 | |
3465 | Tag: leunga-20000807-a-whole-bunch-of-stuff | |
3466 | ||
3467 | Moby, C--, SSA, x86, machine descriptions etc. Should only affect C-- | |
3468 | and Mobdy. | |
3469 | ||
3470 | 1. x86 | |
3471 | ||
3472 | a. Fixes to peephole module by John and Dan. | |
3473 | b. Assembly fix to SETcc by Allen. | |
3474 | c. Fix to c-call by John. | |
3475 | d. Fix to spilling by John. (This one deals with the missing FSTPT case) | |
3476 | e. Instruction selection optimization to SETcc as suggested by John. | |
3477 | ||
3478 | For example, | |
3479 | ||
3480 | MV(32, x, COND(32, CMP(32, LT, a, b), LI 1, LI 0)) | |
3481 | ||
3482 | should generate: | |
3483 | ||
3484 | MOVL a, x | |
3485 | SUBL b, x | |
3486 | SHRL 31, x | |
3487 | ||
3488 | 2. IR stuff | |
3489 | ||
3490 | A bunch of new DJ-graph related algorithms added. These | |
3491 | speed up SSA construction. | |
3492 | ||
3493 | 3. SSA + Scheduling | |
3494 | ||
3495 | Added code for SSA and scheduling to the repository | |
3496 | ||
3497 | ---------------------------------------------------------------------- | |
3498 | Name: Lal George | |
3499 | Date: 2000/07/27 11:53:14 EDT | |
3500 | ||
3501 | Tag: lal-20000727-linux-ppc | |
3502 | Description: | |
3503 | ||
3504 | Made changes to support Linux PPC. | |
3505 | p.s. I have confirmation that the 110.29 boot files work fine. | |
3506 | ||
3507 | ---------------------------------------------------------------------- | |
3508 | Name: Matthias Blume | |
3509 | Date: 2000/07/27 17:40:00 JST | |
3510 | Tag: blume-20000727-scripts | |
3511 | Description: | |
3512 | ||
3513 | !!!! WARNING !!!! | |
3514 | You must recompile the runtime system! | |
3515 | !!!! WARNING !!!! | |
3516 | ||
3517 | This is basically another round of script-enhancements: | |
3518 | ||
3519 | 1. sml, ml-build, and ml-makedepend accept options -D and -U to define | |
3520 | and undefine CM preprocessor symbols. | |
3521 | ||
3522 | 2. ml-build avoids generating a new heap image if it finds that the | |
3523 | existing one is still ok. (The condition is that no ML file had to | |
3524 | be recompiled and all ML files are found to be older that the heap | |
3525 | file.) | |
3526 | ||
3527 | To make this work smoothly, I also hacked the runtime system as | |
3528 | well as SMLofNJ.SysInfo to get access to the heap image suffix | |
3529 | (.sparc-solaris, ...) that is currently being used. | |
3530 | ||
3531 | Moreover, the signature of CM.mk_standalone has changed. See the | |
3532 | CM manual. | |
3533 | ||
3534 | 3. ml-makedepend accepts additional options -n, -a, and -o. (See the | |
3535 | CM manual for details.) | |
3536 | ||
3537 | 4. More CM manual updates: | |
3538 | - all of the above has been documented. | |
3539 | - there is now a section describing the (CM-related) command line | |
3540 | arguments that are accepted by the "sml" command | |
3541 | ||
3542 | ---------------------------------------------------------------------- | |
3543 | Name: Matthias Blume | |
3544 | Date: 2000/07/25 16:20:00 JST | |
3545 | Tag: blume-20000725-makedepend | |
3546 | Description: | |
3547 | ||
3548 | Added a script called ml-makedepend. This can be used in makefiles | |
3549 | for Unix' make in a way very similar to the "makedepend" command for | |
3550 | C. | |
3551 | ||
3552 | The script internally uses function CM.sources. | |
3553 | ||
3554 | Synopsis: | |
3555 | ||
3556 | ml-makedepend [-f makefile] cmfile targetname | |
3557 | ||
3558 | The default for the makefile is "makefile" (or "Makefile" should | |
3559 | "makefile" not exist). | |
3560 | ||
3561 | ml-makedepend adds a cmfile/targetname-specific section to this | |
3562 | makefile (after removing the previous version of this section). The | |
3563 | section contains a single dependency specification with targetname on | |
3564 | the LHS (targetname is an arbitrary name), and a list of files derived | |
3565 | from the cmfile on the RHS. Some of the files on the RHS are | |
3566 | ARCH/OPSYS-specific. Therefore, ml-makedepend inserts references to | |
3567 | "make" variables $(ARCH) and $(OPSYS) in place of the corresponding | |
3568 | path names. The makefile writer is responsible for making sure that | |
3569 | these variables have correct at the time "make" is invoked. | |
3570 | ||
3571 | ---------------------------------------------------------------------- | |
3572 | Name: Matthias Blume | |
3573 | Date: 2000/07/22 23:30:00 JST | |
3574 | Tag: blume-20000722-urlupdate | |
3575 | Description: | |
3576 | ||
3577 | Changed BOOT and config/srcarchiveurl to point to BL server: | |
3578 | ||
3579 | ftp://ftp.research.bell-labs.com/dist/smlnj/working/110.29/ | |
3580 | ||
3581 | ---------------------------------------------------------------------- | |
3582 | Name: Matthias Blume | |
3583 | Date: 2000/07/18 18:00:00 JST | |
3584 | Tag: blume-20000718-Version_110_29 | |
3585 | Description: | |
3586 | ||
3587 | 1. Updated src/compiler/TopLevel/main/version.sml to version 110.29 | |
3588 | ||
3589 | 2. Updated config/version to 110.29 | |
3590 | ||
3591 | 3. Updated config/srcarchiveurl | |
3592 | ||
3593 | 3. New boot files! | |
3594 | ftp://ftp.cs.princeton.edu/pub/people/blume/sml/110.29-autofetch | |
3595 | ||
3596 | ---------------------------------------------------------------------- | |
3597 | Name: Matthias Blume | |
3598 | Date: 2000/07/11 13:58:00 JST | |
3599 | Tag: blume-20000711-doctypo | |
3600 | Description: | |
3601 | ||
3602 | Fixed a few typos in CM manual. | |
3603 | ||
3604 | ---------------------------------------------------------------------- | |
3605 | Name: Allen Leung | |
3606 | Date: 2000/06/15 00:38:00 | |
3607 | Tag: leunga-20000704-sparc-x86 | |
3608 | ||
3609 | 1. x86 peephole improvement sp += k; sp -= k => nop [from John] | |
3610 | 2. fix to x86 RET bug [found by Dan Grossman] | |
3611 | 3. sparc assembly bug fix for ticc instructions [found by Fermin] | |
3612 | ||
3613 | Affects c-- and moby only | |
3614 | ||
3615 | ---------------------------------------------------------------------- | |
3616 | Name: Matthias Blume | |
3617 | Date: 2000/07/04 15:26:00 | |
3618 | Tag: blume-20000704-trigger | |
3619 | Description: | |
3620 | ||
3621 | 1. Improvements to CM manual. | |
3622 | 2. SMLofNJ.Internals.BTrace.trigger reinstated as an alternative way | |
3623 | of getting a back-trace. The function, when called, raises an | |