Annotation of /sml/trunk/HISTORY

Parent Directory | Revision Log

---

Revision 1034 - (view) (download)

1 :	dbm	570	This is the HISTORY file for the Yale SML/NJ CVS repository.
2 :
3 :			An entry should be made for _every_ commit to the repository.
4 :			The entries in this file will be used when creating the README
5 :			for new versions, so keep that in mind when writing the
6 :			description.
7 :
8 :			The form of an entry should be:
9 :
10 :			Name:
11 :	george	1003	Date: yyyy/mm/dd
12 :	dbm	570	Tag: <post-commit CVS tag>
13 :			Description:
14 :	leunga	1025
15 :	george	1003	----------------------------------------------------------------------
16 :	george	1034	Name: Lal George
17 :			Date: 2002/01/24
18 :			Tag: george-20020124-risc-ra-interface
19 :			Description:
20 :
21 :			There is a dramatic simplification in the interface to the
22 :			register allocator for RISC architectures as a result of making
23 :			parallel copy instructions explicit.
24 :
25 :			----------------------------------------------------------------------
26 :	blume	1028	Name: Matthias Blume
27 :	blume	1032	Date: 2002/01/22
28 :			Tag: blume-20020122-x86-ccalls
29 :			Description:
30 :
31 :			Bug fix for c-calls on x86 (having to do with how char- and
32 :			short-arguments are being handled).
33 :
34 :			----------------------------------------------------------------------
35 :			Name: Matthias Blume
36 :	blume	1031	Date: 2002/01/21
37 :			Tag: blume-20020121-ff
38 :			Description:
39 :
40 :			Another day of fiddling with the FFI...
41 :
42 :			1. Bug fix/workaround: CKIT does not complain about negative array
43 :			dimensions, so ml-nlffigen has to guard itself against this possibility.
44 :			(Otherwise a negative dimension would send it into an infinite loop.)
45 :
46 :			2. Some of the abstract types (light objects, light pointers, most "base"
47 :			types) in structure C are now eqtypes.
48 :
49 :			3. Added constructors and test functions for NULL function pointers.
50 :
51 :			----------------------------------------------------------------------
52 :			Name: Matthias Blume
53 :	blume	1028	Date: 2002/01/18
54 :	blume	1030	Tag: blume-20020118-ready-for-new-release
55 :			Description:
56 :
57 :			Made config/srcarchiveurl point to a new place. (Will provide boot
58 :			files shortly.)
59 :
60 :			Maybe we christen this to be 110.38?
61 :
62 :			----------------------------------------------------------------------
63 :			Name: Matthias Blume
64 :			Date: 2002/01/18
65 :	blume	1029	Tag: blume-20020118-more-ffifiddle
66 :			Description:
67 :
68 :			Today's FFI fiddling:
69 :
70 :			- Provided a structure CGetSet with "convenient" versions of C.Get.* and
71 :			C.Set.* that use concrete (MLRep.*) arguments and results instead
72 :			of abstract ones.
73 :
74 :			- Provided word-style bit operations etc. for "int" representation
75 :			types in MLRep.S<Foo>Bitops where <Foo> ranges over Char, Int, Short,
76 :			and Long.
77 :
78 :			----------------------------------------------------------------------
79 :			Name: Matthias Blume
80 :			Date: 2002/01/18
81 :	blume	1028	Tag: blume-20020118-use-x86-fp
82 :			Description:
83 :
84 :			Now that x86-fast-fp seems to be working, I turned it back on again
85 :			by default. (Seems to work fine now, even with the FFI.)
86 :
87 :			Other than that, I added some documentation about the FFI to
88 :			src/ml-nlffigen/README and updated the FFI test examples in
89 :			src/ml-nlffi-lib/Tests/*.
90 :
91 :			----------------------------------------------------------------------
92 :	leunga	1025	Name: Allen Leung
93 :			Date: 2002/01/17
94 :			Tag: leunga-20020117-x86-fast-fp-call
95 :			Description:
96 :
97 :			1. Fixed a problem with handling return fp values when x86's fast fp
98 :			mode is turned on.
99 :
100 :			2. Minor pretty printing fix for cellset. Print %st(0) as %st(0) instead
101 :			of %f32.
102 :
103 :			3. Added a constructor INT32lit to the ast of MLRISC tools.
104 :
105 :			----------------------------------------------------------------------
106 :	blume	1021	Name: Matthias Blume
107 :			Date: 2002/01/16
108 :			Tag: blume-20020116-ffifiddle
109 :			Description:
110 :
111 :			More fiddling with the FFI interface:
112 :
113 :			- Make constness 'c instead of rw wherever possible. This eliminates
114 :			the need for certain explicit coercions. (However, due to ML's
115 :			value polymorphism, there will still be many cases where explicit
116 :			coercions are necessary. Phantom types are not the whole answer
117 :			to modeling a subtyping relationship in ML.)
118 :
119 :			- ro/rw coersions for pointers added. (Avoids the detour through */&.)
120 :
121 :			- "printf" test example added to src/ml-nlffi-lib/Tests. (Demonstrates
122 :			clumsy workaround for varargs problem.)
123 :
124 :			----------------------------------------------------------------------
125 :	george	1016	Name: Lal George
126 :			Date: 2002/01/15
127 :			Tag: <none>
128 :			Description:
129 :
130 :			1. Since COPY instructions are no longer native to the architecture,
131 :			a generic functor can be used to implement the expandCopies function.
132 :
133 :			2. Allowed EXPORT and IMPORT pseudo-op declarations to appear inside a
134 :			TEXT segment.
135 :
136 :			----------------------------------------------------------------------
137 :	blume	1011	Name: Matthias Blume
138 :	blume	1015	Date: 2002/01/15
139 :			Tag: blume-20020115-ffiupdates
140 :			Description:
141 :
142 :			1. Fix for bug resulting in single-precision float values being returned
143 :			incorrectly from FFI calls.
144 :
145 :			2. Small modifications to C FFI API:
146 :
147 :			- memory-allocation routines return straight objects (no options)
148 :			and raise an exception in out-of-memory situations
149 :			- unsafe extensions to cast between function pointers and pointers
150 :			from/to ints
151 :			- added structure C_Debug as an alternative to structure C where
152 :			pointer-dereferencing (|*| and |*!) always check for null-pointers
153 :			- added open_lib' to DynLinkage; open_lib' works like open_lib
154 :			but also takes a (possibly empty) list of existing library handles
155 :			that the current library depends on
156 :
157 :			----------------------------------------------------------------------
158 :			Name: Matthias Blume
159 :	blume	1011	Date: 2002/01/10
160 :			Tag: blume-20020110-newffigen
161 :			Description:
162 :
163 :			1. Updates to portable graph code.
164 :
165 :			2. Major update to ml-nlffigen and ml-nlffi-lib. Things are much
166 :			more scalable now so that even huge interfaces such as the one
167 :			for GTK compile in finite time and space. :-)
168 :			See src/ml-nlffigen/README for details on what's new.
169 :
170 :			----------------------------------------------------------------------
171 :	george	1003	Name: Lal George
172 :	george	1009	Date: 2001/01/09 14:31:35 EST 2002
173 :			Tag: george-20011206-rm-native-copy
174 :			Description:
175 :
176 :			Removed the native COPY and FCOPY instructions
177 :			from all the architectures and replaced it with the
178 :			explicit COPY instruction from the previous commit.
179 :
180 :			It is now possible to simplify many of the optimizations
181 :			modules that manipulate copies. This has not been
182 :			done in this change.
183 :
184 :			----------------------------------------------------------------------
185 :			Name: Lal George
186 :	george	1003	Date: 2001/12/06 16:50:13 EST 2001
187 :			Tag: george-20011206-mlrisc-instruction
188 :			Description:
189 :	george	824
190 :	george	1003	Changed the representation of instructions from being fully abstract
191 :			to being partially concrete. That is to say:
192 :
193 :			from
194 :			type instruction
195 :
196 :			to
197 :			type instr (* machine instruction *)
198 :
199 :			datatype instruction =
200 :			LIVE of {regs: C.cellset, spilled: C.cellset}
201 :			| KILL of {regs: C.cellset, spilled: C.cellset}
202 :			| COPYXXX of {k: CB.cellkind, dst: CB.cell list, src: CB.cell list}
203 :			| ANNOTATION of {i: instruction, a: Annotations.annotation}
204 :			| INSTR of instr
205 :
206 :			This makes the handling of certain special instructions that appear on
207 :			all architectures easier and uniform.
208 :
209 :			LIVE and KILL say that a list of registers are live or killed at the
210 :			program point where they appear. No spill code is generated when an
211 :			element of the 'regs' field is spilled, but the register is moved to
212 :			the 'spilled' (which is present, more for debugging than anything else).
213 :
214 :			LIVE replaces the (now deprecated) DEFFREG instruction on the alpha.
215 :			We used to generate:
216 :
217 :			DEFFREG f1
218 :			f1 := f2 + f3
219 :			trapb
220 :
221 :			but now generate:
222 :
223 :			f1 := f2 + f3
224 :			trapb
225 :			LIVE {regs=[f1,f2,f3], spilled=[]}
226 :
227 :			Furthermore, the DEFFREG (hack) required that all floating point instruction
228 :			use all registers mentioned in the instruction. Therefore f1 := f2 + f3,
229 :			defines f1 and uses [f1,f2,f3]! This hack is no longer required resulting
230 :			in a cleaner alpha implementation. (Hopefully, intel will not get rid of
231 :			this architecture).
232 :
233 :			COPYXXX is intended to replace the parallel COPY and FCOPY available on
234 :			all the architectures. This will result in further simplification of the
235 :			register allocator that must be aware of them for coalescing purposes, and
236 :			will also simplify certain aspects of the machine description that provides
237 :			callbacks related to parallel copies.
238 :
239 :			ANNOTATION should be obvious, and now INSTR represents the honest to God
240 :			machine instruction set!
241 :
242 :			The <arch>/instructions/<arch>Instr.sml files define certain utility
243 :			functions for making porting easier -- essentially converting upper case
244 :			to lower case. All machine instructions (of type instr) are in upper case,
245 :			and the lower case form generates an MLRISC instruction. For example on
246 :			the alpha we have:
247 :
248 :			datatype instr =
249 :			LDA of {r:cell, b:cell, d:operand}
250 :			| ...
251 :
252 :			val lda : {r:cell, b:cell, d:operand} -> instruction
253 :			...
254 :
255 :			where lda is just (INSTR o LDA), etc.
256 :
257 :	leunga	815	----------------------------------------------------------------------
258 :	blume	986	Name: Matthias Blume
259 :	blume	993	Date: 2001/11/22 21:40:00 EST
260 :			Tag: Release_110_37
261 :			Description:
262 :
263 :			Release 110.37. This time for real.
264 :
265 :			----------------------------------------------------------------------
266 :			Name: Matthias Blume
267 :	blume	988	Date: 2001/11/21 16:35:00 EST
268 :			Tag: blume-20011121-foot-in-mouth
269 :			Description:
270 :
271 :			Removed the "Release_110_37" tag because of a serious bug.
272 :			This will be re-tagged once the bug is fixed.
273 :
274 :			----------------------------------------------------------------------
275 :			Name: Matthias Blume
276 :	blume	987	Date: 2001/11/21 16:14:00 EST
277 :			Tag: blume-20011121-forgottenfile
278 :			Description:
279 :
280 :			Forgot to add a file. (Just a .tex-file -- part of
281 :			the CM manual source.)
282 :
283 :			----------------------------------------------------------------------
284 :			Name: Matthias Blume
285 :	blume	986	Date: 2001/11/21 16:10:00 EST
286 :	blume	988	Tag: blume-20011121-invalid_110_37
287 :	blume	986	Description:
288 :
289 :	blume	988	Note: I removed the original tag "Release_110_37" from this commit
290 :			because we found a serious bug in all non-x86 backends.
291 :			- Matthias
292 :
293 :	blume	986	1. Modifications to the SML/NJ code generator and to the runtime system
294 :			so that code object name strings are directly inserted into code
295 :			objects at code generation time. The only business the runtime system
296 :			has with this is now to read the name strings on occasions.
297 :			(The encoding of the name string has also changed somewhat.)
298 :
299 :			2. CM now implements a simple "set calculus" for specifying export lists.
300 :			In particular, it is now possible to refer to the export lists of
301 :			other libraries/groups/sources and form unions as well as differences.
302 :			See the latest CM manual for details.
303 :
304 :			3. An separate notion of "proxy" libraries has again be eliminated from
305 :			CM's model. (Proxy libraries are now simply a special case of using
306 :			the export list calculus.)
307 :
308 :			4. Some of the existing libraries now take advantage of the new set
309 :			calculus.
310 :			(Notice that not all libraries have been converted because some
311 :			of the existing .cm-files are supposed to be backward compatible
312 :			with 110.0.x.)
313 :
314 :			5. Some cleanup in stand-alone programs. (Don't use "exnMessage" -- use
315 :			"General.exnMessage"! The former relies on a certain hook to be
316 :			initialized, and that often does not happen in the stand-alone case.)
317 :
318 :			----------------------------------------------------------------------
319 :	george	984	Name: Lal George
320 :			Date: 2001/11/21 13:56:18 EST
321 :			Tag: george-2001121-pseudo-ops
322 :			Description:
323 :
324 :			Implemented a complete redesign of MLRISC pseudo-ops. Now there
325 :			ought to never be any question of incompatabilities with
326 :			pseudo-op syntax expected by host assemblers.
327 :
328 :			For now, only modules supporting GAS syntax are implemented
329 :			but more should follow, such as MASM, and vendor assembler
330 :			syntax, e.g. IBM as, Sun as, etc.
331 :
332 :			----------------------------------------------------------------------
333 :	blume	977	Name: Matthias Blume
334 :			Date: 2001/11/14 11:52:00 EST
335 :			Tag: blume-20011114-srcname
336 :			Description:
337 :
338 :			1. Routed the name of the current source file to mlriscgen where it
339 :			should be directly emitted into the code object. (This last part
340 :			is yet to be done.)
341 :
342 :			2. Some cleanup of the pgraph code to make it match the proposal that
343 :			I put out the other day. (The proposal notwithstanding, things are
344 :			still in flux here.)
345 :
346 :			----------------------------------------------------------------------
347 :	george	976	Name: Lal George
348 :			Date: 2001/11/14 09:44:04 EST
349 :			Tag:
350 :			Description:
351 :
352 :			Fix for a backpatching bug reported by Allen.
353 :
354 :			Because the boundary between short and long span-dependent
355 :			instructions is +/- 128, there are an astounding number of
356 :			span-dependent instructions whose size is over estimated.
357 :
358 :			Allen came up with the idea of letting the size of span
359 :			dependent instructions be non-monotonic, for a maxIter
360 :			number of times, after which the size must be monotonically
361 :			increasing.
362 :
363 :			This table shows the number of span-dependent instructions
364 :			whose size was over-estimated as a function of maxIter, for the
365 :			file Parse/parse/ml.grm.sml:
366 :
367 :			maxIter # of instructions:
368 :			10 687
369 :			20 438
370 :			30 198
371 :			40 0
372 :
373 :			In compiling the compiler, there is no significant difference in
374 :			compilation speed between maxIter=10 and maxIter=40. Actually,
375 :			my measurements showed that maxIter=40 was a tad faster than
376 :			maxIter=10! Also 96% of the files in the compiler reach a fix
377 :			point within 13 iterations, so fixing maxIter at 40, while high,
378 :			is okay.
379 :
380 :			----------------------------------------------------------------------
381 :	blume	975	Name: Matthias Blume
382 :			Date: 2001/10/31 15:25:00 EST
383 :			Tag: blume-20011031-pgraph
384 :			Description:
385 :
386 :			CKIT:
387 :			* Changed the "Function" constructor of type Ast.ctype to carry optional
388 :			argument identifiers.
389 :			* Changed the return type of TypeUtil.getFunction accordingly.
390 :			* Type equality ignores the argument names.
391 :			* TypeUtil.composite tries to preserve argument names but gives up quickly
392 :			if there is a mismatch.
393 :
394 :			installation script:
395 :			* attempts to use "curl" if available (unless "wget" is available as well)
396 :
397 :			CM:
398 :			* has an experimental implementation of "portable graphs" which I will
399 :			soon propose as an implementation-independent library format
400 :			* there are also new libraries $/pgraph.cm and $/pgraph-util.cm
401 :
402 :			NLFFI-LIB:
403 :			* some cleanup (all cosmetic)
404 :
405 :			NLFFIGEN:
406 :			* temporarily disabled the mechanism that suppresses ML output for
407 :			C definitions whose identifiers start with an underscore character
408 :			* generate val bindings for enum constants
409 :			* user can request that only one style (light or heavy) is being used;
410 :			default is to use both (command-line arguments: -heavy and -light)
411 :			* fixed bug in handling of function types involving incomplete pointers
412 :			* generate ML entry points that take record arguments (i.e., using
413 :			named arguments) for C functions that have a prototype with named
414 :			arguments
415 :			(see changes to CKIT)
416 :
417 :			----------------------------------------------------------------------
418 :	leunga	972	Name: Allen Leung
419 :			Date: 2001/10/27 20:34:00 EDT
420 :			Tag: leunga-20011027-x86-fast-fp-call
421 :			Description:
422 :
423 :			Fixed the bug described in blume-20010920-slowfp.
424 :
425 :			The fix involves
426 :			1. generating FCOPYs in FSTP in ia32-svid
427 :			2. marking a CALL with the appropriate annotation
428 :
429 :			----------------------------------------------------------------------
430 :	blume	961	Name: Matthias Blume
431 :			Date: 2001/10/16 11:32:00 EDT
432 :			Tag: blume-20011016-netbsd
433 :			Description:
434 :
435 :			Underscore patch from Chris Richards (fixing problem with compiling
436 :			runtime system under recent NetBSD).
437 :
438 :			----------------------------------------------------------------------
439 :	george	959	Name: Allen Leung
440 :			Date: 2001/10/12 17:18:32 EDT 2001
441 :			Tag: leung-20011012-x86-printflowgraph
442 :			Description:
443 :
444 :			X86RA now uses a valid (instead of dummy) PrintFlowgraph module.
445 :
446 :			----------------------------------------------------------------------
447 :	george	952	Name: Lal George
448 :	george	958	Date: 2001/10/11 23:51:34 EDT
449 :			Tag: george-20011011-too-many-instrs
450 :			Description:
451 :
452 :			The representation of a program point never expected to see more
453 :			than 65536 instructions in a basic block!
454 :
455 :			----------------------------------------------------------------------
456 :			Name: Lal George
457 :	george	952	Date: 2001/10/09 09:41:37 EDT
458 :			Tag: george-20011008-mlrisc-labels
459 :			Description:
460 :
461 :			Changed the machine description files to support printing of
462 :			local and global labels in assembly code, based on host assembler
463 :			conventions.
464 :
465 :			----------------------------------------------------------------------
466 :	blume	916	Name: Matthias Blume
467 :	blume	939	Date: 2001/09/25 15:25:00 EDT
468 :			Tag: blume-20010925-exninfo
469 :			Description:
470 :
471 :			I provided a non-hook implementation of exnName (at the toplevel) and
472 :			made the "dummy" implementation of exnMessage (at the toplevel) more
473 :			useful: if nothing gets "hooked in", then at least you are going to
474 :			see the exception name and a message indicating why you don't see more.
475 :
476 :			[For the time being, programs that need exnMessage and want to use
477 :			ml-build should either use General.exnMessage (strongly recommended) or
478 :			refer to structure General at some other point so that CM sees a
479 :			static dependency.]
480 :
481 :			[Similar remarks go for "print" and "use": If you want to use their
482 :			functionality in stand-alone programs generated by ml-build, then use
483 :			TextIO.output and Backend.Interact.useFile (from $smlnj/compiler.cm).]
484 :
485 :			----------------------------------------------------------------------
486 :			Name: Matthias Blume
487 :	blume	938	Date: 2001/09/20 17:28:00 EDT
488 :			Tag: blume-20010920-slowfp
489 :			Description:
490 :
491 :			Allen says that x86-fast-fp is not safe yet, so I turned it off again...
492 :
493 :			----------------------------------------------------------------------
494 :			Name: Matthias Blume
495 :	blume	937	Date: 2001/09/20 17:20:00 EDT
496 :			Tag: blume-20010920-canonicalpaths
497 :			Description:
498 :
499 :			0. Updated the BOOT file (something that I forgot to do earlier).
500 :
501 :			1. Small internal change to CM so that it avoids "/../" in filenames
502 :			as much as possible (but only where it is safe).
503 :
504 :			2. Changed config/_run-sml (resulting in a changed bin/.run-sml) so
505 :			that arguments that contain delimiters are passed through correctly.
506 :			This change also means that all "special" arguments of the form
507 :			@SMLxxx... must come first.
508 :
509 :			3. Changed install script to put relative anchor names for tool commands
510 :			into pathconfig.
511 :
512 :			----------------------------------------------------------------------
513 :			Name: Matthias Blume
514 :	george	952	>>>>>>> 1.169
515 :	blume	932	Date: 2001/09/18 15:35:00 EDT
516 :			Tag: blume-20010918-readme11036
517 :			Description:
518 :
519 :			Added README files.
520 :
521 :			----------------------------------------------------------------------
522 :			Name: Matthias Blume
523 :	blume	929	Date: 2001/09/18 11:45:00 EDT
524 :			Tag: Release_110_36 (retag)
525 :			Description:
526 :
527 :			Fixed mistake in config/preloads. Retagged as 110.36.
528 :
529 :			----------------------------------------------------------------------
530 :			Name: Matthias Blume
531 :	blume	928	Date: 2001/09/18 09:40:00 EDT
532 :	blume	929	Tag: Release_110_36_orig (tag changed)
533 :	blume	928	Description:
534 :
535 :			New version (110.36). New bootfiles.
536 :
537 :			----------------------------------------------------------------------
538 :			Name: Matthias Blume
539 :	blume	927	Date: 2001/09/14 16:15:00 EDT
540 :			Tag: blume-20010914-x86fastfp
541 :			Description:
542 :
543 :			John committed some changes that Allen made, in particular a (hopefully)
544 :			correctly working version of the x86-fp module.
545 :
546 :			I changed the default setting of the Control.MLRISC.getFlag "x86-fast-fp"
547 :			flag to "true". Everything seems to compile to a fixpoint ok, and
548 :			"mandelbrot" speeds up by about 15%.
549 :
550 :			----------------------------------------------------------------------
551 :			Name: Matthias Blume
552 :	blume	923	Date: 2001/09/13 11:20:00 EDT
553 :			Tag: blume-20010913-minimal
554 :			Description:
555 :
556 :			1. Stefan Monnier's patch to fix a miscompilation problem that
557 :			was brought to light by John Reppy's work on Moby.
558 :
559 :			2. Implemented a minimal "structure Compiler" that contains just
560 :			"version" and "architecture". The minimal version will be
561 :			available when the full version is not. This is for backward-
562 :			compatibility with code that wants to test Compiler.version.
563 :
564 :			----------------------------------------------------------------------
565 :			Name: Matthias Blume
566 :	blume	920	Date: 2001/08/28 14:03:00 EDT
567 :			Tag: blume-20010828-ml-lex
568 :			Description:
569 :
570 :			Fix for bug 1581, received from Neophytos Michael.
571 :
572 :			----------------------------------------------------------------------
573 :			Name: Matthias Blume
574 :	blume	918	Date: 2001/08/27 11:20:00 EDT
575 :			Tag: blume-20010827-readme11035
576 :			Description:
577 :
578 :			Fleshed out the README file for 110.35.
579 :
580 :			----------------------------------------------------------------------
581 :			Name: Matthias Blume
582 :	blume	916	Date: 2001/08/24 17:10:00 EDT
583 :			Tag: Release_110_35
584 :			Description:
585 :
586 :			New version number (110.35). New bootfiles.
587 :
588 :			----------------------------------------------------------------------
589 :	george	915	Name: Lal George
590 :	george	909	Date: 2001/08/24 13:47:18 EDT 2001
591 :			Tag: george-20010824-MLRISC-graphs
592 :			Description:
593 :
594 :			removed clusters from MLRISC completely and replaced with graphs.
595 :
596 :			----------------------------------------------------------------------
597 :			Name: Matthias Blume
598 :	blume	905	Date: 2001/08/23 17:50:00 EDT
599 :			Tag: blume-20010823-toplevel
600 :			Description:
601 :
602 :			- some reorganization of the code that implements various kinds of
603 :			environments in the compiler (static, dynamic, symbolic, combined)
604 :			- re-implemented the EnvRef module so that evalStream works properly
605 :			(if the stream contains references to "use", "CM.make", etc.)
606 :			- cleaned up evalloop.sml and interact.sml (but they need more cleaning)
607 :
608 :			----------------------------------------------------------------------
609 :			Name: Matthias Blume
610 :	blume	904	Date: 2001/08/20 15:50 EDT
611 :			Tag: blume20010820-slipup
612 :			Description:
613 :
614 :			I forgot to commit a few files. Here they are...
615 :
616 :			----------------------------------------------------------------------
617 :			Name: Matthias Blume
618 :	blume	903	Date: 2001/08/20 15:35:00 EDT
619 :			Tag: blume-20010820-debugprof
620 :			Description:
621 :
622 :			!!!! NEW BOOTFILES !!!!
623 :
624 :			This is another round of reorganizing the compiler sources. This
625 :			time the main goal was to factor out all the "instrumentation"
626 :			passes (for profiling and backtracing) into their own library.
627 :			The difficulty was to do it in such a way that it does not depend
628 :			on elaborate.cm but only on elabdata.cm.
629 :
630 :			Therefore there have been further changes to both elaborate.cm and
631 :			elabdata.cm -- more "generic" things have been moved from the former
632 :			to the latter. As a result, I was forced to split the assignment
633 :			of numbers indicating "primtyc"s into two portions: SML-generic and
634 :			SML/NJ-specific. Since it would have been awkward to maintain,
635 :			I bit the bullet and actually _changed_ the mapping between these
636 :			numbers and primtycs. The bottom line of this is that you need
637 :			a new set of bin- and bootfiles.
638 :
639 :			I have built new bootfiles for all architectures, so doing a fresh
640 :			checkout and config/install.sh should be all you need.
641 :
642 :			The newly created library's name is
643 :
644 :			$smlnj/viscomp/debugprof.cm
645 :
646 :			and its sources live under
647 :
648 :			src/compiler/DebugProf
649 :
650 :			----------------------------------------------------------------------
651 :			Name: Matthias Blume
652 :	blume	902	Date: 2001/08/15 17:15:00 EDT
653 :			Tag: blume-20010815-compreorg
654 :			Description:
655 :
656 :			This is a first cut at reorganizing the CM libraries that make up the
657 :			core of the compiler. The idea is to separate out pieces that could
658 :			be used independently by tools, e.g., the parser, the typechecker, etc.
659 :
660 :			The current status is a step in this direction, but it is not quite
661 :			satisfactory yet. Expect more changes in the future.
662 :
663 :			Here is the current (new) organization...
664 :
665 :			What used to be $smlnj/viscomp/core.cm is now divided into
666 :			six CM libraries:
667 :
668 :			$smlnj/viscomp/basics.cm
669 :			/parser.cm
670 :			/elabdata.cm
671 :			/elaborate.cm
672 :			/execute.cm
673 :			/core.cm
674 :
675 :			The CM files for these libraries live under src/system/smlnj/viscomp.
676 :			All these libraries are proxy libraries that contain precisely
677 :			one CM library component. Here are the locations of the components
678 :			(all within the src/compiler tree):
679 :
680 :			Basics/basics.cm
681 :			Parse/parser.cm
682 :			ElabData/elabdata.cm
683 :			Elaborator/elaborate.cm
684 :			Execution/execute.cm
685 :			core.cm
686 :
687 :			[This organization is the same that has been used already
688 :			for a while for the architecture-specific parts of the visible
689 :			compiler and for the old version of core.cm.]
690 :
691 :			As you will notice, many source files have been moved from their
692 :			respective original locations to a new home in one of the above
693 :			subtrees.
694 :
695 :			The division of labor between the new libraries is the following:
696 :
697 :			basics.cm:
698 :			- Simple, basic definitions that pertain to many (or all) of
699 :			the other libraries.
700 :			parser.cm:
701 :			- The SML parser, producing output of type Ast.dec.
702 :			- The type family for Ast is also defined and exported here.
703 :			elabdata.cm:
704 :			- The datatypes that describe input and output of the elaborator.
705 :			This includes types, absyn, and static environments.
706 :			elaborator.cm:
707 :			- The SML/NJ type checker and elaborator.
708 :			This maps an Ast.dec (with a given static environment) to
709 :			an Absyn.dec (with a new static environment).
710 :			- This libraries implements certain modules that used to be
711 :			structures as functors (to remove dependencies on FLINT).
712 :			execute.cm:
713 :			- Everything having to do with executing binary code objects.
714 :			- Dynamic environments.
715 :			core.cm:
716 :			- SML/NJ-specific instantiations of the elaborator and MLRISC.
717 :			- Top-level modules.
718 :			- FLINT (this should eventually become its own library)
719 :
720 :			Notes:
721 :
722 :			I am not 100% happy with the way I separated the elaborator (and its
723 :			data structures) from FLINT. Two instances of the same problem:
724 :
725 :			1. Data structures contain certain fields that carry FLINT-specific
726 :			information. I hacked around this using exn and the property list
727 :			module from smlnj-lib. But the fact that there are middle-end
728 :			specific fields around at all is a bit annoying.
729 :
730 :			2. The elaborator calculates certain FLINT-related information. I tried
731 :			to make this as abstract as I could using functorization, but, again,
732 :			the fact that the elaborator has to perform calculations on behalf
733 :			of the middle-end at all is not nice.
734 :
735 :			3. Having to used exn and property lists is unfortunate because it
736 :			weakens type checking. The other alternative (parameterizing
737 :			nearly *everything*) is not appealing, though.
738 :
739 :			I removed the "rebinding =" warning hack because due to the new organization
740 :			it was awkward to maintain it. As a result, the compiler now issues some of
741 :			these warnings when compiling init.cmi during bootstrap compilation. On
742 :			the plus side, you also get a warning when you do, for example:
743 :			val op = = Int32.+
744 :			which was not the case up to now.
745 :
746 :			I placed "assign" and "deref" into the _Core structure so that the
747 :			code that deals with the "lazy" keyword can find them there. This
748 :			removes the need for having access to the primitive environment
749 :			during elaboration.
750 :
751 :			----------------------------------------------------------------------
752 :			Name: Matthias Blume
753 :	blume	898	Date: 2001/08/13
754 :			Tag: blume-20010813-closures
755 :			Description:
756 :
757 :			This fix was sent to us by Zhong Shao. It is supposed to improve the
758 :			performance of certain loops by avoiding needless closure allocation.
759 :
760 :			----------------------------------------------------------------------
761 :	george	890	Name: Lal George
762 :	george	896	Date: 2001/07/31 10:03:23 EDT 2001
763 :			Tag: george-20010731-x86-fmalloc
764 :			Description: Fixed bug in x86 calls
765 :
766 :			There was a bug where call instructions would mysteriously
767 :			vanish. The call instruction had to be one that returned
768 :			a floating point value.
769 :
770 :			----------------------------------------------------------------------
771 :			Name: Lal George
772 :	george	890	Date: 2001/07/19 16:36:29 EDT 2001
773 :			Tag: george-20010719-simple-cells
774 :			Description:
775 :
776 :			I have dramatically simplified the interface for CELLS in MLRISC.
777 :
778 :			In summary, the cells interface is broken up into three parts:
779 :
780 :			1. CellsBasis : CELLS_BASIS
781 :
782 :			CellsBasis is a top level structure and common for all
783 :			architectures. it contains the definitions of basic datatypes
784 :			and utility functions over these types.
785 :
786 :			2. functor Cells() : CELLS
787 :
788 :			Cells generates an interface for CELLS that incorporates the
789 :			specific resources on the target architecture, such as the
790 :			presence of special register classes, their number and size,
791 :			and various useful substructures.
792 :
793 :			3. <ARCH>CELLS
794 :
795 :			e.g. SparcCells: SPARCCELLS
796 :
797 :			<ARCH>CELLS usually contains additional bindings for special
798 :			registers on the architecture, such as:
799 :
800 :			val r0 : cell (* register zero *)
801 :			val y : cell (* Y register *)
802 :			val psr : cell (* processor status register *)
803 :			...
804 :
805 :			The structure returned by applying the Cells functor is opened
806 :			in this interface.
807 :
808 :			The main implication of all this is that the datatypes for cells is
809 :			split between CellsBasis and CELLS -- a fairly simple change for user
810 :			code.
811 :
812 :			In the old scheme the CELLS interface had a definitional binding of
813 :			the form:
814 :
815 :			signature CELLS = sig
816 :
817 :			structure CellsBasis = CellsBasis
818 :
819 :			...
820 :
821 :			end
822 :
823 :			With all the sharing constraints that goes on in MLRISC, this old
824 :			design quickly leads to errors such as:
825 :
826 :			"structure definition spec inside of sharing ... "
827 :
828 :
829 :			and appears to require an unacceptable amount of sharing and where
830 :			constraint hackery.
831 :
832 :			I think this error message (the interaction of definitional specs and
833 :			sharing) requires more explanation on our web page.
834 :
835 :			----------------------------------------------------------------------
836 :	blume	826	Name: Matthias Blume
837 :	blume	879	Date: 2001/07/19 15:00:00 EDT
838 :			Tag: blume-20010719-libreorg
839 :			Description:
840 :
841 :			This update puts together a fairly extensive but straightforward change
842 :			to the way the libraries that implement the interactive system are
843 :			organized:
844 :
845 :			The biggest change is the elimination of structure Compiler. As a
846 :			replacement for this structure, there is now a CM library
847 :			(known as $smlnj/compiler.cm or $smlnj/compiler/current.cm)
848 :			that exports all the substructures of the original structure Compiler
849 :			directly. So instead of saying Compiler.Foo.bar one now simply
850 :			says Foo.bar. (The CM libraries actually export a collection of
851 :			structures that is richer than the collection of substructures of
852 :			structure Compiler.)
853 :
854 :			To make the transition smooth, there is a separate library called
855 :			$smlnj/compiler/compiler.cm which puts together and exports the
856 :			original structure Compiler (or at least something very close to it).
857 :
858 :			There are five members of the original structure Compiler
859 :			that are not exported directly but which instead became members
860 :			of a new structure Backend (described by signature BACKEND). These are:
861 :			structure Profile (: PROFILE), structure Compile (: COMPILE), structure
862 :			Interact (: INTERACT), structure Machine (: MACHINE), and val
863 :			architecture (: string).
864 :
865 :			Structure Compiler.Version has become structure CompilerVersion.
866 :
867 :			Cross-compilers for alpha32, hppa, ppc, sparc, and x86 are provided
868 :			by $smlnj/compiler/<arch>.cm where <arch> is alpha32, hppa, ppc, sparc,
869 :			or x86, respectively.
870 :			Each of these exports the same frontend structures that
871 :			$smlnj/compiler.cm exports. But they do not have a structure Backend
872 :			and instead export some structure <Arch>Backend where <Arch> is Alpha32,
873 :			Hppa, PPC, Sparc, or X86, respectively.
874 :
875 :			Library $smlnj/compiler/all.cm exports the union of the exports of
876 :			$smlnj/compiler/<arch>.cm
877 :
878 :			There are no structures <Arch>Compiler anymore, use
879 :			$smlnj/compiler/<arch>.cm instead.
880 :
881 :			Library host-compiler-0.cm is gone. Instead, the internal library
882 :			that instantiates CM is now called cm0.cm. Selection of the host
883 :			compiler (backend) is no longer done here but. (Responsibility for it
884 :			now lies with $smlnj/compiler/current.cm. This seems to be more
885 :			logical.)
886 :
887 :			Many individual files have been moved or renamed. Some files have
888 :			been split into multiple files, and some "dead" files have been deleted.
889 :
890 :			Aside from these changes to library organization, there are also changes
891 :			to the way the code itself is organized:
892 :
893 :			Structure Binfile has been re-implemented in such a way that it no
894 :			longer needs any knowledge of the compiler. It exclusively deals
895 :			with the details of binfile layout. It no longer invokes the
896 :			compiler (for the purpose of creating new prospective binfile
897 :			content), and it no longer has any knowledge of how to interpret
898 :			pickles.
899 :
900 :			Structure Compile (: COMPILE) has been stripped down to the bare
901 :			essentials of compilation. It no longer deals with linking/execution.
902 :			The interface has been cleaned up considerably.
903 :
904 :			Utility routines for dealing with linking and execution have been
905 :			moved into their own substructures.
906 :
907 :			(The ultimate goal of these changes is to provide a light-weight
908 :			binfile loader/linker (at least for, e.g., stable libraries) that
909 :			does not require CM or the compiler to be present.)
910 :
911 :			CM documentation has been updated to reflect the changes to library
912 :			organization.
913 :
914 :			----------------------------------------------------------------------
915 :			Name: Matthias Blume
916 :	blume	875	Date: 2001/07/10 17:30:00 EDT
917 :			Tag: Release_110_34
918 :			Description:
919 :
920 :	blume	876	Minor tweak to 110.34 (re-tagged):
921 :
922 :			- README.html file added to CVS repository
923 :			- runtime compiles properly under FreeBSD 3.X and 4.X
924 :
925 :			----------------------------------------------------------------------
926 :			Name: Matthias Blume
927 :			Date: 2001/07/10 17:30:00 EDT
928 :			Tag: Release_110_34
929 :			Description:
930 :
931 :	blume	875	New version number (110.34). New bootfiles.
932 :
933 :			----------------------------------------------------------------------
934 :			Name: Matthias Blume
935 :	blume	874	Date: 2001/07/09 16:00:00 EDT
936 :			Tag: blume-20010709-more-varargs
937 :			Description:
938 :
939 :			I changed the handling of varargs in ml-nlffigen again:
940 :			The ellipsis ... will now simply be ignored (with an accompanying warning).
941 :
942 :			The immediate effect is that you can actually call a varargs function
943 :			from ML -- but you can't actually supply any arguments beyond the ones
944 :			specified explicitly. (For example, you can call printf with its format
945 :			string, but you cannot pass additional arguments.)
946 :
947 :			This behavior is only marginally more useful than the one before, but
948 :			it has the advantage that a function or, more importantly, a function
949 :			type never gets dropped on the floor, thus avoiding follow-up problems with
950 :			other types that refer to the offending one.
951 :
952 :			----------------------------------------------------------------------
953 :			Name: Matthias Blume
954 :	blume	873	Date: 2001/07/09 11:25:00 EDT
955 :			Tag: blume-20010709-varargs
956 :			Description:
957 :
958 :			1. ckit-lib.cm now exports structure Error
959 :			2. ml-nlffigen reports occurences of "..." (i.e., varargs function types)
960 :			with a warning accompanied by a source location. Moreover, it
961 :			merely skips the offending function or type and proceeds with the
962 :			rest of its work.u As a result, one can safely feed C code containing
963 :			"..." to ml-nlffigen.
964 :			3. There are some internal improvements to CM, providing slightly
965 :			more general string substitutions in the tools subsystem.
966 :
967 :			----------------------------------------------------------------------
968 :			Name: Matthias Blume
969 :	blume	854	Date: 2001/06/27 15:10:00 EDT
970 :			Tag: blume-20010627-concur
971 :			Description:
972 :
973 :			Fixed a small bug in CM's handling of parallel compilation.
974 :			(You could observe the bug by Control-C-interrupting an ordinary
975 :			CMB.make or CM.stabilize and then attaching some compile servers.
976 :			The result was that all of a sudden the previously interrupted
977 :			compilation would continue on its own. This was because of
978 :			an over-optimization: CM did not bother to clean out certain queues
979 :			when no servers were attached "anyway", resulting in the contents
980 :			of these queues to grab control when new servers did get attached.)
981 :
982 :			There is also another minor update to the CM manual.
983 :
984 :			----------------------------------------------------------------------
985 :			Name: Matthias Blume
986 :	blume	853	Date: 2001/06/26 16:15:00 EDT
987 :			Tag: blume-20010626-cmdoc
988 :			Description:
989 :
990 :			Minor typo fixed in CM manual (syntax diagram for libraries).
991 :
992 :			----------------------------------------------------------------------
993 :			Name: Matthias Blume
994 :	blume	852	Date: 2001/06/25 22:55:00 EDT
995 :			Tag: blume-20010625-x86pc
996 :			Description:
997 :
998 :			Fixed a nasty bug in the X86 assembly code that caused signal
999 :			handlers to fail (crash) randomly.
1000 :
1001 :			----------------------------------------------------------------------
1002 :			Name: Matthias Blume
1003 :	blume	846	Date: 2001/06/25 12:05:00 EDT
1004 :			Tag: blume-20010625-nlffigen
1005 :			Description:
1006 :
1007 :			This update fixes a number of minor bugs in ml-nlffigen as reported by
1008 :			Nick Carter <nbc@andrew.cmu.edu>.
1009 :
1010 :			1. Silly but ok typedefs of the form "typedef void myvoid;" are now accepted.
1011 :			2. Default names for generated files are now derived from the name of
1012 :			the C file *without its directory*. In particular, this causes generated
1013 :			files to be placed locally even if the C file is in some system directory.
1014 :			3. Default names for generated signatures and structures are also derived
1015 :			from the C file name without its directory. This avoids silly things
1016 :			like "structure GL/GL".
1017 :			(Other silly names are still possible because ml-nlffigen does not do
1018 :	blume	847	a thorough check of whether generated names are legal ML identifiers.
1019 :			When in doubt, use command line arguments to force particular names.)
1020 :	blume	846
1021 :			----------------------------------------------------------------------
1022 :			Name: Matthias Blume
1023 :	blume	845	Date: 2001/06/21 12:25:00 EDT
1024 :			Tag: blume-20010621-eXene
1025 :			Description:
1026 :
1027 :			eXene now compiles and (sort of) works again.
1028 :
1029 :			The library name (for version > 110.33) is $/eXene.cm.
1030 :
1031 :			I also added an new example in src/eXene/examples/nbody. See the
1032 :			README file there for details.
1033 :
1034 :			----------------------------------------------------------------------
1035 :			Name: Matthias Blume
1036 :	blume	844	Date: 2001/06/20 16:40:00 EDT
1037 :			Tag: blume-20010620-cml
1038 :			Description:
1039 :
1040 :			CML now compiles and works again.
1041 :
1042 :			Libraries (for version > 110.33):
1043 :
1044 :			$cml/cml.cm Main CML library.
1045 :			$cml/basis.cm CML's version of $/basis.cm.
1046 :			$cml/cml-internal.cm Internal helper library.
1047 :			$cml/core-cml.cm Internal helper library.
1048 :			$cml-lib/trace-cml.cm Tracing facility.
1049 :			$cml-lib/smlnj-lib.cm CML's version of $/smlnj-lib.cm
1050 :
1051 :			The installer (config/install.sh) has been taught how to properly
1052 :			install this stuff.
1053 :
1054 :			----------------------------------------------------------------------
1055 :			Name: Matthias Blume
1056 :	blume	843	Date: 2001/06/19 17:55:00 EDT
1057 :			Tag: blume-20010619-instantiate
1058 :			Description:
1059 :
1060 :			This un-breaks the fix for bug 1432.
1061 :			(The bug was originally fixed in 110.9 but I broke it again some
1062 :			time after that.)
1063 :
1064 :			----------------------------------------------------------------------
1065 :			Name: Matthias Blume
1066 :	blume	842	Date: 2001/06/19 17:25:00 EDT
1067 :			Tag: blume-20010619-signals
1068 :			Description:
1069 :
1070 :			This should (hopefully) fix the long-standing signal handling bug.
1071 :			(The runtime system was constructing a continuation record with an
1072 :			incorrect descriptor which would cause the GC to drop data on the floor...)
1073 :
1074 :			----------------------------------------------------------------------
1075 :			Name: Matthias Blume
1076 :	blume	840	Date: 2001/06/15 15:05:00 EDT
1077 :	blume	841	Tag: blume-20010615-moresparc
1078 :			Description:
1079 :
1080 :			Here is a short late-hour update related to Sparc c-calls:
1081 :
1082 :			-- made handling of double-word arguments a bit smarter
1083 :
1084 :			-- instruction selection phase tries to collapse certain clumsily
1085 :			constructed ML-Trees; typical example:
1086 :
1087 :			ADD(ty,ADD(_,e,LI d1),LI d2) -> ADD(ty,e,LI(d1+d2))
1088 :
1089 :			This currently has no further impact on SML/NJ since mlriscGen does
1090 :			not seem to generate such patterns in the first place, and c-calls
1091 :			(which did generate them in the beginning) has meanwhile been fixed
1092 :			so as to avoid them as well.
1093 :
1094 :			----------------------------------------------------------------------
1095 :			Name: Matthias Blume
1096 :			Date: 2001/06/15 15:05:00 EDT
1097 :	blume	840	Tag: blume-20010615-sparc
1098 :			Description:
1099 :
1100 :			The purpose of this update is to provide an implementation of NLFFI
1101 :			on Sparc machines.
1102 :
1103 :			Here are the changes in detail:
1104 :
1105 :			* src/MLRISC/sparc/c-calls/sparc-c-calls.sml is a new file containing
1106 :			the Sparc implementation of the c-calls API.
1107 :			* The Sparc backend of SML/NJ has been modified to uniformely use %fp
1108 :			for accessing the ML frame. Thus, we have a real frame pointer and
1109 :			can freely modify %sp without need for an omit-frame-ptr phase.
1110 :			The vfp logic in src/compiler/CodeGen/* has been changed to accomodate
1111 :			this case.
1112 :			* ml-nlffigen has been taught to produce code for different architectures
1113 :			and calling conventions.
1114 :			* In a way similar to what was done in the x86 case, the Sparc
1115 :			backend uses its own specific extension to mltree. (For example,
1116 :			it needs to be able to generate UNIMP instructions which are part
1117 :			of the calling convention.)
1118 :			* ml-nlffi-lib was reorganized to make it more modular (in particular,
1119 :			to make it easier to plug in new machine- and os-dependent parts).
1120 :
1121 :			There are some other fairly unrelated bug fixes and cleanups as well:
1122 :
1123 :			* I further hacked the .cm files for MLRISC tools (like MDLGen) so
1124 :			that they properly share their libraries with existing SML/NJ libraries.
1125 :			* I fixed a minor cosmetic bug in CM, supressing certain spurious
1126 :			follow-up error messages.
1127 :			* Updates to CM/CMB documentation.
1128 :
1129 :			TODO items:
1130 :
1131 :			* MLRISC should use a different register as its asmTemp on the Sparc.
1132 :			(The current %o2 is a really bad choice because it is part of the
1133 :			calling conventions, so things might interfere in unexpected ways.)
1134 :
1135 :			----------------------------------------------------------------------
1136 :			Name: Matthias Blume
1137 :	blume	839	Date: 2001/06/07
1138 :			Tag: blume-20010607-calls
1139 :			Description:
1140 :
1141 :			A number of internal changes related to C calls and calling conventions:
1142 :
1143 :			1. ML-Tree CALL statements now carry a "pops" field. It indicates the
1144 :			number of bytes popped implicitly (by the callee). In most cases
1145 :			this field is 0 but on x86/win32 it is some non-zero value. This
1146 :			is information provided for the benefit of the "omit-frameptr" pass.
1147 :			2. The CALL instruction on the x86 carries a similar "pops" field.
1148 :			The instruction selection phase copies its value from the ML-Tree
1149 :			CALL statement.
1150 :			3. On all other architectures, the instruction selection phase checks
1151 :			whether "pops=0" and complains if not.
1152 :			4. The c-calls implementation for x86 now accepts two calling conventions:
1153 :			"ccall" and "stdcall". When "ccall" is selected, the caller cleans
1154 :			up after the call and pops is set to 0. For "stdcall", the caller
1155 :			does nothing, leaving the cleanup to the callee; pops is set to
1156 :			the number of bytes that were pushed onto the stack.
1157 :			5. The cproto decoder (compiler/Semant/types/cproto.sml) now can
1158 :			distinguish between "ccall" and "stdcall".
1159 :			6. The UNIMP instruction has been added to the supported Sparc instruction
1160 :			set. (This is needed for implementing the official C calling convention
1161 :			on this architecture.)
1162 :			7. I fixed some of the .cm files under src/MLRISC/Tools to make them
1163 :			work with the latest CM.
1164 :
1165 :			----------------------------------------------------------------------
1166 :			Name: Matthias Blume
1167 :	blume	838	Date: 2001/06/05 15:10:00 EDT
1168 :			Tag: blume-20010605-cm-index
1169 :			Description:
1170 :
1171 :			0. The "lambdasplit" parameter for class "sml" in CM has been documented.
1172 :
1173 :			1. CM can now generate "index files". These are human-readable files
1174 :			that list on a per-.cm-file basis each toplevel symbol defined or
1175 :			imported. The location of the index file for
1176 :			<p>/<d>.cm is <p>/CM/INDEX/<d>.cm.
1177 :			To enable index-file generation, set CM.Control.generate_index to true
1178 :			or export an environment-symbol: export CM_GENERATE_INDEX=true.
1179 :
1180 :			The CM manual has been updated accordingly.
1181 :
1182 :			2. I made some slight modifications to the c-calls API in MLRISC.
1183 :
1184 :			a) There is now a callback to support saving/restoring of
1185 :			dedicated but caller-save registers around the actual call
1186 :			instruction.
1187 :			b) One can optionally specify a comment-annotation for the
1188 :			call instruction.
1189 :
1190 :			3. SML/NJ (mlriscGen.sml) uses this new API for the rawccall primop.
1191 :			(For example, the comment annotation shows the C prototype of
1192 :			the function being called.)
1193 :
1194 :			----------------------------------------------------------------------
1195 :			Name: Matthias Blume
1196 :	blume	837	Date: 2001/06/01 13:30:00 EDT
1197 :			Tag: blume-20010601-nlffi-cleanup
1198 :			Description:
1199 :
1200 :			This is mostly a cleanup of MLFFI stuff:
1201 :
1202 :			- some signature files have been put into a more exposed place
1203 :			- the ugly 'f type parameter is gone (simplifies types tremendously!)
1204 :			- ml-nlffigen changed accordingly
1205 :			- tutorial updated
1206 :
1207 :			Other changes:
1208 :
1209 :			- author's affiliation in CM manual(s) updated
1210 :			- some more recognized keywords added to Allen's sml.sty
1211 :
1212 :			----------------------------------------------------------------------
1213 :			Name: Matthias Blume
1214 :	blume	836	Date: 2001/05/25 15:30:00 EDT
1215 :			Tag: blume-20010525-iptr
1216 :			Description:
1217 :
1218 :			- put the official 110.33-README (as it appears on the ftp server) under
1219 :			CVS
1220 :			- fixed a small bug related to incomplete pointer types in
1221 :			ml-nlffigen
1222 :			- small cosmetic change to the ml-nlffi-lib's "arr" type constructor
1223 :			(it does not need the 'f type parameter)
1224 :
1225 :			----------------------------------------------------------------------
1226 :			Name: Matthias Blume
1227 :	blume	834	Date: 2001/05/23 14:30:00 EDT
1228 :			Tag: Release_110_33
1229 :			Description:
1230 :
1231 :			New version number (110.33). New bootfiles.
1232 :
1233 :			----------------------------------------------------------------------
1234 :			Name: Matthias Blume
1235 :	blume	833	Date: 2001/05/22 18:06:00 EDT
1236 :			Tag: blume-20010522-targets
1237 :			Description:
1238 :
1239 :			Made install.sh use file config/targets.customized if it exists, falling
1240 :			back to config/targets if it doesn't. This way one can have a customized
1241 :			version of the targets file without touching the "real thing", thus
1242 :			eliminating the constant fear of accidentally checking something bogus
1243 :			back into the CVS repository... (File config/targets.customized must
1244 :			not be added to the repository!)
1245 :
1246 :			----------------------------------------------------------------------
1247 :			Name: Matthias Blume
1248 :	blume	832	Date: 2001/05/22 16:30:00 EDT
1249 :			Tag: blume-20010522-minitut
1250 :			Description:
1251 :
1252 :			1. Bug fix in ml-nlffigen; now (hopefully) correctly handling
1253 :			struct returns.
1254 :			2. Added src/ml-nlffi-lib/Doc/mini-tutorial.txt. This is some very
1255 :			incomplete, preliminary documentation for NLFFI.
1256 :
1257 :			----------------------------------------------------------------------
1258 :			Name: Matthias Blume
1259 :	blume	829	Date: 2001/05/14 11:30:00 EDT
1260 :			Tag: blume-20010514-script
1261 :			Description:
1262 :
1263 :			Some bugs in install script fixed.
1264 :
1265 :			In addition to that I also made a slight change to the NLFFI API:
1266 :			Functors generated by ml-nlffigen now take the dynamic library as a
1267 :			straight functor argument, not as a suspended one. (The original
1268 :			functor code used to force the suspension right away anyway, so there
1269 :			was nothing gained by this complication of the interface.)
1270 :
1271 :			----------------------------------------------------------------------
1272 :			Name: Matthias Blume
1273 :	blume	828	Date: 2001/05/11 14:35:00 EDT
1274 :			Tag: blume-20010511-ml-nlffi
1275 :			Description:
1276 :
1277 :			I finally took the plunge and added my new FFI code to the main
1278 :			repository. For x86-linux it is now ready for prime-time.
1279 :
1280 :			There are two new subdirectories of "src":
1281 :
1282 :			- ml-nlffi-lib:
1283 :			The utility library for programs using the FFI interface.
1284 :			Here is the implementation of $/c.cm and its associated low-level
1285 :			partners $/c-int.cm and $/memory.cm.
1286 :			- ml-nlffigen:
1287 :			A stand-alone program for generating ML glue code from C source
1288 :			code.
1289 :
1290 :			Building ml-nlffigen requires $/ckit-lib.cm.
1291 :
1292 :			The config/install.sh script has been updates to do the Right Thing
1293 :			(hopefully).
1294 :
1295 :			Notice that the source tree for the C-Kit will not be put under "src"
1296 :			but directly under the installation root directory. (This is the
1297 :			structure that currently exists on the CVS server when you check out
1298 :			module "sml".) Fortunately, config/install.sh knows about this oddity.
1299 :
1300 :			Bugs: No documentation yet.
1301 :
1302 :			----------------------------------------------------------------------
1303 :			Name: Matthias Blume
1304 :	blume	826	Date: 2001/05/09 16:35:00 EDT
1305 :			Tag: blume-20010509-cpscontract
1306 :			Description:
1307 :
1308 :			Fixed a bug in the accounting code in cpsopt/contract.sml. (The
1309 :			wrapper/unwrapper elimination did not decrement usage counts and some
1310 :			dead variables got overlooked by the dead-up logic.)
1311 :
1312 :			----------------------------------------------------------------------
1313 :	george	824	Name: Lal George
1314 :			Date: 2001/05/08 17:26:09 EDT
1315 :			Tag: george-20010508-omit-frameptr
1316 :			Description:
1317 :
1318 :			Changes to implement the omit-frame-pointer optimization to support
1319 :			raw C calls. For now, there is only support on the Intel x86, but
1320 :			other architectures will follow as more experience is gained with this.
1321 :
1322 :
1323 :			----------------------------------------------------------------------
1324 :	blume	818	Name: Matthias Blume
1325 :	blume	819	Date: 2001/05/07 14:40:00 EDT
1326 :			Tag: blume-20010507-proxies
1327 :			Description:
1328 :
1329 :			I made into "proxy libraries" all libraries that qualify for such a
1330 :			change. (A qualifying library is a library that has another library or
1331 :			groups as its sole member and repeats that member's export list
1332 :			verbatim. A proxy library avoids this repetition by omitting its export
1333 :			list, effectively inheriting the list that its (only) member exports.
1334 :			See the CM manual for more explanation.)
1335 :			The main effect is that explicit export lists for these libraries
1336 :			do not have to be kepts in sync, making maintenance a bit easier.
1337 :
1338 :			I also added copyright notices to many .cm-files.
1339 :
1340 :			Last but not least, I made a new set of bootfiles.
1341 :
1342 :			----------------------------------------------------------------------
1343 :			Name: Matthias Blume
1344 :	blume	818	Date: 2001/05/04 17:00:00 EDT
1345 :			Tag: blume-20010504-cm-lsplit
1346 :			Description:
1347 :
1348 :			0. John merged pending changes to $/smlnj-lib.cm
1349 :
1350 :			1. Allen's previous change accidentally backed out of one of Lal's
1351 :			earlier changes. I undid this mistake (re-introducing Lal's change).
1352 :
1353 :			2. I used the new topOrder' function from graph-scc.sml (from $/smlnj-lib.cm)
1354 :			within the compiler where applicable. There is some code simplification
1355 :			because of that.
1356 :
1357 :			3. The "split" phase (in FLINT) is now part of the default list of phases.
1358 :			Compiler.Control.LambdaSplitting.* can be used to globally control the
1359 :			lambda-splitting (cross-module-inlining) engine. In addition to that,
1360 :			it can now also be controlled on a per-source basis: CM has been taught
1361 :			a new tool parameter applicable to ML source files.
1362 :
1363 :			- To turn lambda-splitting off completely:
1364 :			local open Compiler.Control.LambdaSplitting in
1365 :			val _ = set Off
1366 :			end
1367 :			- To make "no lambda-splitting" the global default (but allow per-source
1368 :			overriding); this is the initial setting:
1369 :			local open Compiler.Control.LambdaSplitting in
1370 :			val _ = set (Default NONE)
1371 :			end
1372 :			- To make "lambda-splitting with aggressiveness a" the global default
1373 :			(and allow per-source overriding):
1374 :			local open Compiler.Control.LambdaSplitting in
1375 :			val _ = set (Default (SOME a))
1376 :			end
1377 :
1378 :			- To turn lambda-splitting off for a given ML souce file (say: a.sml)
1379 :			write (in the respective .cm-file):
1380 :			a.sml (lambdasplitting:off)
1381 :			- To turn lambda-splitting for a.sml on with minimal aggressiveness:
1382 :			a.sml (lambdasplitting:on)
1383 :			- To turn lambda-splitting for a.sml on with aggressiveness <a> (where
1384 :			<a> is a decimal non-negative integer):
1385 :			a.sml (lambdasplitting:<a>)
1386 :			- To turn lambda-splitting for a.sml on with maximal aggressiveness:
1387 :			a.sml (lambdasplitting:infinity)
1388 :			- To use the global default for a.sml:
1389 :			a.sml (lambdasplitting:default)
1390 :			or simply
1391 :			a.sml
1392 :
1393 :			----------------------------------------------------------------------
1394 :	leunga	815	Name: Allen Leung
1395 :			Date: 2001/05/04 01:57:00 EDT
1396 :			Tag: leunga-20010504-sync
1397 :			Description:
1398 :	leunga	788
1399 :	leunga	815	MLRISC features.
1400 :
1401 :			1. Fix to CMPXCHG instructions.
1402 :			2. Changed RA interface to allow annotations in callbacks.
1403 :			3. Added a new method to the stream interface to allow annotations updates.
1404 :
1405 :	george	749	----------------------------------------------------------------------
1406 :	blume	812	Name: Matthias Blume
1407 :	blume	814	Date: 2001/05/01 11:45:00 EDT
1408 :			Tag: blume-20010501-pcedittmp
1409 :			Description:
1410 :
1411 :			Changed install.sh to use the current working directory instead of
1412 :			/usr/tmp for a temporary file (pcedittmp). The previous choice
1413 :			of /usr/tmp caused trouble with MacOS X because of file premission
1414 :			problems.
1415 :
1416 :			----------------------------------------------------------------------
1417 :			Name: Matthias Blume
1418 :	blume	812	Date: 2001/04/20 11:10:00 EDT
1419 :			Tag: blume-20010420-inMLflag
1420 :			Description:
1421 :
1422 :			- added vp_limitPtrMask to vproc-state.h
1423 :			(for use by the raw-C-calls mechanism to implement proper interrupt
1424 :			handling)
1425 :			- made the ML compiler aware of various data-structure offsets so it
1426 :			can generate code for accessing the vp_inML flag and vp_limitPtrMask
1427 :			- tweaked mlriscGen.sml to have it emit interrupt-handling code for
1428 :			raw C-calls
1429 :
1430 :			----------------------------------------------------------------------
1431 :	george	810	Name: Lal George
1432 :			Date: 2001/04/20 09:15:28 EDT
1433 :			Tag: george-20010420-macosX
1434 :			Description:
1435 :
1436 :			- Changes to port to Mac OS X; Darwin.
1437 :
1438 :			- In the process I found that sqrt was broken on the PPC, because the
1439 :			fsqrt instruction is not implemented.
1440 :
1441 :			----------------------------------------------------------------------
1442 :	blume	799	Name: Matthias Blume
1443 :	blume	808	Date: 2001/04/18 12:45:00 EDT
1444 :			Tag: blume-20010418-ccalls
1445 :			Description:
1446 :
1447 :			- fixed two off-by-4 errors in the x86-specific c-calls implementation
1448 :			(this bug prevented structure arguments containing pointers from being
1449 :			passed correctly)
1450 :			- changed the raw-C-call code in mlriscGen.sml in such a way that
1451 :			structure arguments are represented as a pointer to the beginning
1452 :			of the structure (instead of having a series of synthesized arguments,
1453 :			one for each structure member)
1454 :
1455 :			- made makeml script's verbosity level configurable via environment
1456 :			variable (MAKEML_VERBOSITY)
1457 :
1458 :	blume	813	- eliminated placeholder implementations for f32l, w16s, i16s, and f32s
1459 :	blume	808	in rawmem-x86.sml; we are now using the real thing
1460 :
1461 :			----------------------------------------------------------------------
1462 :			Name: Matthias Blume
1463 :	blume	806	Date: 2001/03/22 16:25:00 EST
1464 :			Tag: blume-20010322-bootfiles
1465 :			Description:
1466 :
1467 :			Created a new set of bootfiles (for your automatic installation convenience).
1468 :
1469 :			----------------------------------------------------------------------
1470 :			Name: Matthias Blume
1471 :	blume	805	Date: 2001/03/22 15:10:00 EST
1472 :			Tag: blume-20010322-rawmem-parcm
1473 :			Description:
1474 :
1475 :			1. All "raw memory access" primitives for the new FFI are implemented now
1476 :			(at least on the x86).
1477 :			2. Some further cleanup of CM's parallel make mechanism.
1478 :
1479 :			----------------------------------------------------------------------
1480 :			Name: Matthias Blume
1481 :	blume	801	Date: 2001/03/19 17:53:00 EST
1482 :			Tag: blume-20010319-parallel
1483 :			Description:
1484 :
1485 :			Parallel make (using compile servers) now works again.
1486 :
1487 :			To this end, CM.stabilize and CMB.make have been modified to work in
1488 :			two passes when compile servers are attached:
1489 :			1. Compile everything, do not perform stabilization; this pass
1490 :			uses compile servers
1491 :			2. Stabilize everything; this pass does not use compile servers
1492 :			If there are no compile servers, the two passes are combined into one
1493 :			(as before). Splitting the passes increases the inherent parallelism
1494 :			in the dependency graph because the entire graph including all
1495 :			libraries is available at the same time. This, in turn, improves
1496 :			server utilization. The downside is that the master process will
1497 :			have to do some extra work after compilation is done (because for
1498 :			technical reasons it must re-read all the binfiles during stabilization).
1499 :
1500 :			----------------------------------------------------------------------
1501 :			Name: Matthias Blume
1502 :	blume	800	Date: 2001/03/16 12:22:00 EST
1503 :			Tag: blume-20010316-bootfiles
1504 :			Description:
1505 :
1506 :			Created a new set of bootfiles (for your automatic installation convenience).
1507 :
1508 :			----------------------------------------------------------------------
1509 :			Name: Matthias Blume
1510 :	blume	799	Date: 2001/03/16 11:00:00 EST
1511 :			Tag: blume-20010316-MLTREE-fixup
1512 :			Description:
1513 :
1514 :			This is a minor fixup for an (untagged) earlier commit by Allen.
1515 :			(A file was missing).
1516 :
1517 :			----------------------------------------------------------------------
1518 :	leunga	788	Name: Allen Leung
1519 :	leunga	796	Date: Mon Mar 5 18:54:57 EST 2001
1520 :			Tag: leunga-20010305-cut-support
1521 :
1522 :			1. New support for alternative control-flow in MLTREE.
1523 :			Currently we support
1524 :
1525 :			FLOW_TO(CALL ...., [k1,...,kn])
1526 :
1527 :			This is needed for 'cuts to' in C-- and try/handle-like constructs
1528 :			in Moby
1529 :
1530 :			New assembler flag "asm-show-cutsto" to turn on control-flow debugging.
1531 :
1532 :			2. Register Allocator
1533 :
1534 :			Changes in interface [from Fermin, John]
1535 :
1536 :			3. Alpha 8-bit SLL support [Fermin]
1537 :
1538 :			4. All architectures
1539 :
1540 :			A new module (ClusterExpandCopies) for expanding parallel copies.
1541 :
1542 :			----------------------------------------------------------------------
1543 :			Name: Allen Leung
1544 :	leunga	788	Date: 2001/02/27 23:07:00 EST
1545 :			Tag: leunga-20010227-minor-stuff
1546 :
1547 :			1. Alpha bug fix for CMOVNE
1548 :			2. Handle mltree COND(..,FCMP ...,...)
1549 :			3. Bug fix in simplifier
1550 :
1551 :			----------------------------------------------------------------------
1552 :	blume	777	Name: Matthias Blume
1553 :	blume	781	Date: 2001/01/30 17:50:00 EST
1554 :			Tag: blume-20010130-sync
1555 :			Description:
1556 :
1557 :			This is just a minor update to sync my devel branch with the main brach.
1558 :			The only visible change is the addition of some README files.
1559 :
1560 :			----------------------------------------------------------------------
1561 :			Name: Matthias Blume
1562 :	blume	778	Date: 2001/01/12 23:30:00 JST
1563 :			Tag: blume-20010112-bootfiles
1564 :			Description:
1565 :
1566 :			Made a new set of bootfiles that goes with the current state of the
1567 :			repository.
1568 :
1569 :			----------------------------------------------------------------------
1570 :			Name: Matthias Blume
1571 :	blume	777	Date: 2001/01/12 21:20:00 JST
1572 :			Tag: blume-20010112-sync
1573 :			Description:
1574 :
1575 :			I am just flushing out some minor changes that had accumulated in
1576 :			my private branch in order to sync with the main tree. (This is
1577 :			mainly because I had CVS trouble when trying to merge _into_ my
1578 :			private branch.)
1579 :
1580 :			Most people should be completely unaffected by this.
1581 :
1582 :			----------------------------------------------------------------------
1583 :	leunga	775	Name: Allen Leung
1584 :			Date: Thu Jan 11 21:03:00 EST 2001
1585 :			Tag: leunga-20010111-labexp=mltree
1586 :			Description:
1587 :
1588 :			1. Removed the type LabelExp and replace it by MLTree.
1589 :			2. Rewritten mltree-simplify with the pattern matcher tool.
1590 :			3. There were some bugs in alpha code generator which would break
1591 :			64-bit code generation.
1592 :			4. Redo the tools to generate code with the
1593 :			5. The CM files in MLRISC (and in src/system/smlnj/MLRISC)
1594 :			are now generated by perl scripts.
1595 :
1596 :			----------------------------------------------------------------------
1597 :	blume	771	Name: Matthias Blume
1598 :	blume	774	Date: 2001/01/10 21:55:00 JST
1599 :			Tag: blume-20010110-rcc
1600 :			Description:
1601 :
1602 :			The RCC stuff now seems to work (but only on the x86).
1603 :			This required hacking of the c-calls interface (and -implementation) in
1604 :			MLRISC.
1605 :
1606 :			Normal compiler users should be unaffected.
1607 :
1608 :			----------------------------------------------------------------------
1609 :			Name: Matthias Blume
1610 :	blume	773	Date: 2001/01/09 01:20:00 JST
1611 :			Tag: blume-20010109-rcc
1612 :			Description:
1613 :
1614 :			This is a fairly big patch, flushing out a large number of pending
1615 :			changes that I made to my development copy over the last couple of days.
1616 :
1617 :			Of practical relevance at this moment is a workaround for a pickling
1618 :			bug that Allen ran into the other day. The cause of the bug itself is
1619 :			still unknown and it might be hard to fix it properly, but the
1620 :			workaround has some merits of its own (namely somewhat reducing pickling
1621 :			overhead for certain libraries). Therefore, I think this solution should
1622 :			be satisfactory at this time.
1623 :
1624 :			The rest of the changes (i.e., the vast majority) has to do with my
1625 :			ongoing efforts of providing direct support for C function calls from
1626 :			ML. At the moment there is a new primop "RAW_CCALL", typing magic
1627 :			in types/cproto.sml (invoked from FLINT/trans/translate.sml), a new
1628 :			case in the FLINT CPS datatype (RCC), changes to cps/convert.sml to
1629 :			translate uses of RAW_CCALL into RCC, and changes to mlriscGen.sml to
1630 :			handle RCC.
1631 :
1632 :			The last part (the changes to mlriscGen.sml) are still known to be
1633 :			wrong on the x86 and not implemented on all other architectures. But
1634 :			the infrastructure is in place. I had to change a few functor
1635 :			signatures in the backend to be able to route the CCalls interface
1636 :			from MLRISC there, and I had to specialize the mltree type (on the
1637 :			x86) to include the necessary extensions. (The extensions themselves
1638 :			were already there and redy to go in MLRISC/x86).
1639 :
1640 :			Everything should be very happy as soon as someone helps me with
1641 :			mlriscGen.sml...
1642 :
1643 :			In any case, nothing of this should matter to anyone as long as the
1644 :			new primop is not being used (which is going to be the case unless you
1645 :			find it where I hid it :). The rest of the compiler is completely
1646 :			unaffected.
1647 :
1648 :			----------------------------------------------------------------------
1649 :			Name: Matthias Blume
1650 :	blume	772	Date: 2001/01/05 00:30:00 JST
1651 :			Tag: blume-20010105-primops
1652 :			Description:
1653 :
1654 :			Added some experimental support for work that I am doing right now.
1655 :			These changes mostly concern added primops, but there is also a new
1656 :			experimental C library in the runtime system (but currently not enabled
1657 :			anywhere except on Linux/X86).
1658 :
1659 :			In the course of adding primops (and playing with them), I discovered that
1660 :			Zhong's INL_PRIM hack (no type info for certain primops) was, in fact, badly
1661 :			broken. (Zhong was very right he labeled this stuff as "major gross hack".)
1662 :			To recover, I made type information in INL_PRIM mandatory and changed
1663 :			prim.sml as well as built-in.sml accordingly. The InLine structure now
1664 :			has complete, correct type information (i.e., no bottom types).
1665 :
1666 :			Since all these changes mean that we need new binfiles, I also bumped the
1667 :			version number to 110.32.1.
1668 :
1669 :			----------------------------------------------------------------------
1670 :			Name: Matthias Blume
1671 :	blume	771	Date: 2000/12/30 22:10:00 JST
1672 :			Tag: blume-20001230-various
1673 :			Description:
1674 :
1675 :			Added proxy libraries for MLRISC and let MLRISC libraries refer
1676 :			to each other using path anchors. (See CM manual for explanation.)
1677 :
1678 :			Updated CM documentation.
1679 :
1680 :			Fixed some bugs in CM.
1681 :
1682 :			Implemented "proxy" libraries (= syntactic sugar for CM).
1683 :
1684 :			Added "-quiet" option to makeml and changed runtime system accordingly.
1685 :
1686 :			Added cleanup handler for exportML to reset timers and compiler stats.
1687 :
1688 :			----------------------------------------------------------------------
1689 :	george	761	Name: Lal George
1690 :			Date: 2000/12/22 22:22:58 EST 2000
1691 :			Tag: Release_110_32
1692 :			Description:
1693 :
1694 :			Infinite precision used throughout MLRISC.
1695 :			see MLRISC/mltree/machine-int.sig
1696 :
1697 :			----------------------------------------------------------------------
1698 :	blume	756	Name: Matthias Blume
1699 :	blume	760	Date: 2000/12/22 23:16:00 JST
1700 :			Tag: blume-20001222-warn
1701 :			Description:
1702 :	blume	759
1703 :	blume	760	Corrected wording and formatting of some CM warning message which I
1704 :			broke in my previous patch.
1705 :	blume	759
1706 :			----------------------------------------------------------------------
1707 :			Name: Matthias Blume
1708 :	blume	763	Date: 2000/12/22 21:20:00 JST
1709 :			Tag: blume-20001222-anchorenv
1710 :			Description:
1711 :
1712 :			Fixed CM's handling of anchor environments in connection with CMB.make.
1713 :
1714 :			----------------------------------------------------------------------
1715 :			Name: Matthias Blume
1716 :	blume	758	Date: 2000/12/22 13:15:00 JST
1717 :			Tag: blume-20001222-cleanup
1718 :			Description:
1719 :
1720 :			Removed src/cm/ffi which does not (and did not) belong here.
1721 :
1722 :			----------------------------------------------------------------------
1723 :			Name: Matthias Blume
1724 :	blume	757	Date: 2000/12/21 23:55:00 JST
1725 :			Tag: blume-20001221-exn
1726 :			Description:
1727 :
1728 :			Probably most important: CM no longer silently swallows all exceptions
1729 :			in the compiler.
1730 :			Plus: some other minor CM changes. For example, CM now reports some
1731 :			sizes for generated binfiles (code, data, envpickle, lambdapickle).
1732 :
1733 :			----------------------------------------------------------------------
1734 :			Name: Matthias Blume
1735 :	blume	756	Date: 2000/12/15 00:01:05 JST
1736 :			Tag: blume-20001215-dirtool
1737 :			Description:
1738 :
1739 :			- "dir" tool added.
1740 :			- improvements and cleanup to Tools structure
1741 :			- documentation updates
1742 :
1743 :			----------------------------------------------------------------------
1744 :	leunga	752	Name: Allen Leung
1745 :	leunga	755	Date: Thu Dec 14 03:45:24 EST 2000
1746 :			Description:
1747 :			Tag: leunga-20001214-int-inf
1748 :			Description:
1749 :
1750 :			In IntInf, added these standard functions, which are missing from our
1751 :			implementation:
1752 :
1753 :			andb : int * int -> int
1754 :			xorb : int * int -> int
1755 :			orb : int * int -> int
1756 :			notb : int -> int
1757 :			<< : int * word -> int
1758 :			~>> : int * word -> int
1759 :
1760 :			Not tested, I hope they are correct.
1761 :
1762 :			----------------------------------------------------------------------
1763 :			Name: Allen Leung
1764 :	leunga	752	Date: Fri Dec 8 19:23:26 EST 2000
1765 :			Description:
1766 :			Tag: leunga-20001208-nowhere
1767 :			Description:
1768 :
1769 :			Slight improvements to the 'nowhere' tool to handle OR-patterns,
1770 :			to generate better error messages etc. Plus a brief manual.
1771 :
1772 :			----------------------------------------------------------------------
1773 :	george	749	Name: Lal George
1774 :			Date: 2000/12/08 09:54:02 EST 2000
1775 :			Tag: Release_110_31
1776 :			Description:
1777 :	george	705
1778 :	george	749	- Version 110.31
1779 :	leunga	591	----------------------------------------------------------------------
1780 :	leunga	744	Name: Allen Leung
1781 :			Date: Thu Dec 7 22:01:04 EST 2000
1782 :			Tag: leunga-20001207-cell-monster-hack
1783 :			Description:
1784 :
1785 :			Major MLRISC internal changes. Affect all clients.
1786 :			Summary:
1787 :
1788 :			1. Type CELLS.cell = int is now replaced by a datatype.
1789 :			As a result, the old regmap is now gone. Almost all interfaces
1790 :			in MLRISC change as a consequence.
1791 :
1792 :			2. A new brand version of machine description tool (v3.0) that generates
1793 :			modules expecting the new interface. The old version is removed.
1794 :
1795 :			3. The RA interface has been further abstracted into two new functors.
1796 :			RISC_RA and X86RA. These functors have much simpler interfaces.
1797 :			[See also directory MLRISC/demo.]
1798 :
1799 :			4. Some other new source->source code generation tools are available:
1800 :
1801 :			a. MLRISC/Tools/RewriteGen -- generate rewriters from rules.
1802 :			b. MLRISC/Tools/WhereGen -- expands conditional pattern matching rules.
1803 :			I use this tool to generate the peephole optimizers---with the new
1804 :			cell type changes, peephole rules are becoming difficult to write
1805 :			without conditional pattern matching.
1806 :
1807 :			5. More Intmap -> IntHashTable change. Previous changes by Matthias didn't
1808 :			cover the entire MLRISC source tree so many things broke.
1809 :
1810 :			6. CM files have been moved to the subdirectory MLRISC/cm.
1811 :			They are moved because there are a lot of them and they clutter up the
1812 :			root dir.
1813 :
1814 :			7. More detailed documentation to come...
1815 :
1816 :			NOTE: To rebuild from 110.30 (ftp distribution), you'll have to do
1817 :			a makeml -rebuild first. This is because of other other
1818 :			changes that Matthias has made (see below).
1819 :
1820 :
1821 :			----------------------------------------------------------------------
1822 :	blume	733	Name: Matthias Blume
1823 :	blume	742	Date: 2000/11/30 23:12:00 JST
1824 :			Tag: blume-20001130-filereorg
1825 :			Description:
1826 :
1827 :			Some manual updates and some file reorganizations in CM.
1828 :
1829 :			----------------------------------------------------------------------
1830 :			Name: Matthias Blume
1831 :	blume	737	Date: 2000/11/24 17:45:00 JST
1832 :			Tag: blume-20001124-link
1833 :			Description:
1834 :
1835 :			Drastically improved link traversal code for the case that the dynamic
1836 :			value was already loaded at bootstrap time. As a result, CM and CMB
1837 :			now both load blazingly fast -- even on a very slow machine. Also,
1838 :			memory consumption has been further reduced by this.
1839 :
1840 :			Warning: The format of the PIDMAP file has changed. THerefore, to
1841 :			bootstrap you have to do this:
1842 :
1843 :			1. Run CMB.make
1844 :			2. Make a symbolic link for the boot directory:
1845 :			ln -s sml.boot.ARCH-OS xxx
1846 :			3. "Rebuild" the boot directory:
1847 :			./makeml -boot xxx -rebuild sml ; rm xxx
1848 :			4. Boot normally:
1849 :			./makeml
1850 :
1851 :			----------------------------------------------------------------------
1852 :			Name: Matthias Blume
1853 :	blume	735	Date: 2000/11/21 21:20:00 JST
1854 :			Tag: blume-20001121-tools
1855 :			Description:
1856 :
1857 :			Continued hacking on autoloading problem -- with success this time.
1858 :			Also changed tool-plugin mechanism. See new CM manual.
1859 :
1860 :			----------------------------------------------------------------------
1861 :			Name: Matthias Blume
1862 :	blume	734	Date: 2000/11/19 14:30:00 JST
1863 :			Tag: blume-20001119-autoload
1864 :			Description:
1865 :
1866 :			Some hacking to make autoloading faster. Success for CMB, no success
1867 :			so far for CM. There is a reduced structure CM' that autoloads faster.
1868 :			(This is a temporary, non-documented hack to be eliminated again when
1869 :			the general problem is solved.)
1870 :
1871 :			----------------------------------------------------------------------
1872 :			Name: Matthias Blume
1873 :	blume	733	Date: 2000/11/17 14:10:00 JST
1874 :			Tag: blume-20001117-pickle-lib
1875 :			Description:
1876 :
1877 :			1. Eliminated comp-lib.cm
1878 :			2. Made pickle-lib.cm
1879 :			3. Eliminated all uses of intset.sml (from comp-lib.cm)
1880 :			4. Replaced all uses of intmap.{sig,sml} (from comp-lib.cm) with
1881 :			equivalent constructs from smlnj-lib.cm (INtHashTable).
1882 :			5. Point 4. also goes for those uses of intmap.* in MLRISC.
1883 :			Duplicated intmap modules thrown out.
1884 :			6. Hunted down all duplicated SCC code and replaced it with
1885 :			equivalent stuff (GraphSCCFn from smlnj-lib.cm).
1886 :			7. Rewrote Feedback module.
1887 :			8. Moved sortedlist.sml into viscomp-lib.cm. Eventually it
1888 :			should be thrown out and equivalent modules from smlnj-lib.cm
1889 :			should be used (IntRedBlackSet, IntListSet, ...).
1890 :
1891 :			Confirmed that compiler compiles to fixpoint.
1892 :
1893 :			----------------------------------------------------------------------
1894 :	leunga	731	Name: Allen Leung
1895 :			Date: 2000/11/10 18:00:00
1896 :			Tag: leunga-20001110-new-x86-fp
1897 :
1898 :			A new x86 floating point code generator has been added.
1899 :			By default this is turned off. To turn this on, do:
1900 :
1901 :			CM.autoload "$smlnj/compiler.cm";
1902 :			Compiler.Control.MLRISC.getFlag "x86-fast-fp" := true;
1903 :
1904 :			Changes:
1905 :
1906 :			1. Changed FTAN to FPTAN so that the assembly output is correct.
1907 :			2. Changed the extension callback for FTANGENT to generate:
1908 :
1909 :			fptan
1910 :			fstp %st(0)
1911 :			instead of
1912 :			fptan
1913 :			fstpl ftempmem
1914 :
1915 :			3. Numerous assembly fixes for x86.
1916 :
1917 :			5. Cleaned up the machine code output module x86/x86MC.sml and added
1918 :			support for a whole bunch of instructions and addressing modes:
1919 :
1920 :			fadd/fsub/fsubr/fmul/fdiv/fdivr %st, %st(n)
1921 :			faddp/fsubp/fsubrp/fmulp/fdivp/fdivrp %st, %st(n)
1922 :			fadd/fsub/fsubr/fmul/fdiv/fdivr %st(n), %st
1923 :			fiadd/fisub/fisubr/fimul/fidiv/fidivr mem
1924 :			fxch %st(n)
1925 :			fld %st(n)
1926 :			fst %st(n)
1927 :			fst mem
1928 :			fstp %st(n)
1929 :			fucom %st(n)
1930 :			fucomp %st(n)
1931 :
1932 :			All these are now generated when the fast fp mode is turned on.
1933 :
1934 :			6. Removed the dedicated registers %st(0), ..., %st(7) from X86CpsRegs
1935 :
1936 :			----------------------------------------------------------------------
1937 :	blume	729	Name: Matthias Blume
1938 :			Date: 2000/11/09 11:20:00 JST
1939 :			Tag: blume-20001109-scc
1940 :			Description:
1941 :
1942 :			Eliminated some code duplication:
1943 :
1944 :			1. Added "where" clause to GraphSCCFn in SML/NJ Library.
1945 :			(Otherwise the functor is useless.)
1946 :			2. Used GraphSCCFn where SCCUtilFun was used previously.
1947 :			3. Got rid of SCCUtilFun (in comp-lib.cm).
1948 :
1949 :			----------------------------------------------------------------------
1950 :	george	721	Name: Lal George
1951 :			Date: 2000/11/06 09:02:21 EST 2000
1952 :			Tag: Release_110_30
1953 :			Description:
1954 :
1955 :			- Version 110.30
1956 :			----------------------------------------------------------------------
1957 :	blume	715	Name: Matthias Blume
1958 :	blume	716	Date: 2000/11/04 14:45:00
1959 :			Tag: blume-20001104-mlbuild
1960 :			Description:
1961 :
1962 :			- Made ml-build faster on startup.
1963 :			- Documentation fixes.
1964 :
1965 :			----------------------------------------------------------------------
1966 :			Name: Matthias Blume
1967 :	blume	715	Date: 2000/11/02 17:00:00 JST
1968 :			Tag: blume-20001102-condcomp
1969 :			Description:
1970 :
1971 :			- Small tweaks to pickler -- new BOOTFILES!
1972 :			- Version bumped to 110.29.2.
1973 :			- Added conditional compilation facility to init.cmi (see comment there).
1974 :			----------------------------------------------------------------------
1975 :	leunga	713	Name: Allen Leung
1976 :			Date: 2000/10/23 19:31:00
1977 :			Tag: leunga-20001023-demo-ra
1978 :
1979 :			1. Minor RA changes that improves spilling on x86 (affects Moby and C-- only)
1980 :			2. Test programs for the graph library updated
1981 :			3. Some new MLRISC demo programs added
1982 :
1983 :			----------------------------------------------------------------------
1984 :	blume	710	Name: Matthias Blume
1985 :	blume	711	Date: 2000/08/31 22:15:00 JST
1986 :			Tag: blume-20001017-errmsg
1987 :			Description:
1988 :
1989 :			More error message grief: Where there used to be no messages, there
1990 :			now were some that had bogus error regions. Fixed.
1991 :
1992 :			----------------------------------------------------------------------
1993 :			Name: Matthias Blume
1994 :	blume	710	Date: 2000/08/31 17:30:00 JST
1995 :			Tag: blume-20001017-v110p29p1
1996 :			Description:
1997 :
1998 :			I made a version 110.29.1 with new bootfiles.
1999 :
2000 :			Changes: Modified pickler/unpickler for faster and leaner unpickling.
2001 :			CM documentation changes and a small bugfix in CM's error reporting.
2002 :
2003 :			----------------------------------------------------------------------
2004 :	george	705	Name: Lal George
2005 :			Date: 2000/09/27 14:42:35 EDT
2006 :			Tag: george-20000927-nodestatus
2007 :			Description:
2008 :
2009 :			Changed the type of the nodestatus, so that:
2010 :
2011 :			SPILLED(~1) is now SPILLED
2012 :			SPILLED(m) where m>=0 is now MEMREG(m)
2013 :			SPILLED(s) where s<~1 is now SPILL_LOC(~s)
2014 :
2015 :			----------------------------------------------------------------------
2016 :	blume	697	Name: Matthias Blume
2017 :	blume	703	Date: 2000/09/07 14:45:00 JST
2018 :			Tag: blume-20000907-cmerrmsg
2019 :			Description:
2020 :
2021 :			Small tweak to CM to avoid getting ML syntax error messages twice.
2022 :
2023 :			----------------------------------------------------------------------
2024 :			Name: Matthias Blume
2025 :	blume	701	Date: 2000/08/31 18:00:00 JST
2026 :			Tag: blume-20000831-cvsbootfiles
2027 :			Description:
2028 :
2029 :			New URL for boot files (because the 110.29 files on the BL server do
2030 :			now work correctly with my updated install scripts for yacc and lex).
2031 :
2032 :			----------------------------------------------------------------------
2033 :			Name: Matthias Blume
2034 :	blume	697	Date: 2000/08/08 12:33:00 JST
2035 :			Tag: blume-20000808-manual
2036 :			Description:
2037 :
2038 :			Tiny update to CM manual.
2039 :
2040 :			----------------------------------------------------------------------
2041 :	leunga	695	Name: Allen Leung
2042 :			Date: 2000/08/7 19:31:00
2043 :			Tag: leunga-20000807-a-whole-bunch-of-stuff
2044 :
2045 :			Moby, C--, SSA, x86, machine descriptions etc. Should only affect C--
2046 :			and Mobdy.
2047 :
2048 :			1. x86
2049 :
2050 :			a. Fixes to peephole module by John and Dan.
2051 :			b. Assembly fix to SETcc by Allen.
2052 :			c. Fix to c-call by John.
2053 :			d. Fix to spilling by John. (This one deals with the missing FSTPT case)
2054 :			e. Instruction selection optimization to SETcc as suggested by John.
2055 :
2056 :			For example,
2057 :
2058 :			MV(32, x, COND(32, CMP(32, LT, a, b), LI 1, LI 0))
2059 :
2060 :			should generate:
2061 :
2062 :			MOVL a, x
2063 :			SUBL b, x
2064 :			SHRL 31, x
2065 :
2066 :			2. IR stuff
2067 :
2068 :			A bunch of new DJ-graph related algorithms added. These
2069 :			speed up SSA construction.
2070 :
2071 :			3. SSA + Scheduling
2072 :
2073 :			Added code for SSA and scheduling to the repository
2074 :
2075 :			----------------------------------------------------------------------
2076 :	george	694	Name: Lal George
2077 :			Date: 2000/07/27 11:53:14 EDT
2078 :
2079 :			Tag: lal-20000727-linux-ppc
2080 :			Description:
2081 :
2082 :			Made changes to support Linux PPC.
2083 :			p.s. I have confirmation that the 110.29 boot files work fine.
2084 :
2085 :			----------------------------------------------------------------------
2086 :	blume	684	Name: Matthias Blume
2087 :	blume	692	Date: 2000/07/27 17:40:00 JST
2088 :			Tag: blume-20000727-scripts
2089 :			Description:
2090 :
2091 :			!!!! WARNING !!!!
2092 :			You must recompile the runtime system!
2093 :			!!!! WARNING !!!!
2094 :
2095 :			This is basically another round of script-enhancements:
2096 :
2097 :			1. sml, ml-build, and ml-makedepend accept options -D and -U to define
2098 :			and undefine CM preprocessor symbols.
2099 :
2100 :			2. ml-build avoids generating a new heap image if it finds that the
2101 :			existing one is still ok. (The condition is that no ML file had to
2102 :			be recompiled and all ML files are found to be older that the heap
2103 :			file.)
2104 :
2105 :			To make this work smoothly, I also hacked the runtime system as
2106 :			well as SMLofNJ.SysInfo to get access to the heap image suffix
2107 :			(.sparc-solaris, ...) that is currently being used.
2108 :
2109 :			Moreover, the signature of CM.mk_standalone has changed. See the
2110 :			CM manual.
2111 :
2112 :			3. ml-makedepend accepts additional options -n, -a, and -o. (See the
2113 :			CM manual for details.)
2114 :
2115 :			4. More CM manual updates:
2116 :			- all of the above has been documented.
2117 :			- there is now a section describing the (CM-related) command line
2118 :			arguments that are accepted by the "sml" command
2119 :
2120 :			----------------------------------------------------------------------
2121 :			Name: Matthias Blume
2122 :	blume	691	Date: 2000/07/25 16:20:00 JST
2123 :			Tag: blume-20000725-makedepend
2124 :			Description:
2125 :
2126 :			Added a script called ml-makedepend. This can be used in makefiles
2127 :			for Unix' make in a way very similar to the "makedepend" command for
2128 :			C.
2129 :
2130 :			The script internally uses function CM.sources.
2131 :
2132 :			Synopsis:
2133 :
2134 :			ml-makedepend [-f makefile] cmfile targetname
2135 :
2136 :			The default for the makefile is "makefile" (or "Makefile" should
2137 :			"makefile" not exist).
2138 :
2139 :			ml-makedepend adds a cmfile/targetname-specific section to this
2140 :			makefile (after removing the previous version of this section). The
2141 :			section contains a single dependency specification with targetname on
2142 :			the LHS (targetname is an arbitrary name), and a list of files derived
2143 :			from the cmfile on the RHS. Some of the files on the RHS are
2144 :			ARCH/OPSYS-specific. Therefore, ml-makedepend inserts references to
2145 :			"make" variables $(ARCH) and $(OPSYS) in place of the corresponding
2146 :			path names. The makefile writer is responsible for making sure that
2147 :			these variables have correct at the time "make" is invoked.
2148 :
2149 :			----------------------------------------------------------------------
2150 :			Name: Matthias Blume
2151 :	blume	690	Date: 2000/07/22 23:30:00 JST
2152 :			Tag: blume-20000722-urlupdate
2153 :			Description:
2154 :
2155 :			Changed BOOT and config/srcarchiveurl to point to BL server:
2156 :
2157 :			ftp://ftp.research.bell-labs.com/dist/smlnj/working/110.29/
2158 :
2159 :			----------------------------------------------------------------------
2160 :			Name: Matthias Blume
2161 :	blume	686	Date: 2000/07/18 18:00:00 JST
2162 :			Tag: blume-20000718-Version_110_29
2163 :			Description:
2164 :
2165 :			1. Updated src/compiler/TopLevel/main/version.sml to version 110.29
2166 :
2167 :			2. Updated config/version to 110.29
2168 :
2169 :			3. Updated config/srcarchiveurl
2170 :
2171 :			3. New boot files!
2172 :			ftp://ftp.cs.princeton.edu/pub/people/blume/sml/110.29-autofetch
2173 :
2174 :			----------------------------------------------------------------------
2175 :			Name: Matthias Blume
2176 :	blume	684	Date: 2000/07/11 13:58:00 JST
2177 :			Tag: blume-20000711-doctypo
2178 :			Description:
2179 :
2180 :			Fixed a few typos in CM manual.
2181 :
2182 :			----------------------------------------------------------------------
2183 :	leunga	683	Name: Allen Leung
2184 :			Date: 2000/06/15 00:38:00
2185 :			Tag: leunga-20000704-sparc-x86
2186 :
2187 :			1. x86 peephole improvement sp += k; sp -= k => nop [from John]
2188 :			2. fix to x86 RET bug [found by Dan Grossman]
2189 :			3. sparc assembly bug fix for ticc instructions [found by Fermin]
2190 :
2191 :			Affects c-- and moby only
2192 :
2193 :			----------------------------------------------------------------------
2194 :	blume	680	Name: Matthias Blume
2195 :	blume	682	Date: 2000/07/04 15:26:00
2196 :			Tag: blume-20000704-trigger
2197 :			Description:
2198 :
2199 :			1. Improvements to CM manual.
2200 :			2. SMLofNJ.Internals.BTrace.trigger reinstated as an alternative way
2201 :			of getting a back-trace. The function, when called, raises an
2202 :			internal exception which explicitly carries the full back-trace history,
2203 :			so it is unaffected by any intervening handle-raise pairs ("trivial"
2204 :			or not). The interactive loop will print that history once it arrives
2205 :			at top level.
2206 :			Short of having all exceptions implicitly carry the full history, the
2207 :			recommended way of using this facility is:
2208 :			- compile your program with instrumentation "on"
2209 :			- run it, when it raises an exception, look at the history
2210 :			- if the history is "cut off" because of some handler, go and modify
2211 :			your program so that it explicitly calls BTrace.trigger
2212 :			- recompile (still instrumented), and rerun; look at the full history
2213 :
2214 :			----------------------------------------------------------------------
2215 :			Name: Matthias Blume
2216 :			Date: 2000/07/03 15:36:00 JST
2217 :	blume	680	Tag: blume-20000702-manual
2218 :			Description:
2219 :
2220 :			Small corrections and updates to CM manual.
2221 :
2222 :			----------------------------------------------------------------------
2223 :	blume	682	Name: Matthias Blume
2224 :	blume	679	Date: 2000/06/29 16:04:00 JST
2225 :			Tag: blume-20000629-yacctool
2226 :			Description:
2227 :
2228 :			Changes:
2229 :
2230 :			1. Class "mlyacc" now takes separate arguments to pass options to
2231 :			generated .sml- and .sig-files independently.
2232 :			2. Corresponding CM manual updates.
2233 :			3. BTrace module now also reports call sites. (However, for loop clusters
2234 :			it only shows from where the cluster was entered.) There are associated
2235 :			modifications to core.sml, internals.{sig,sml}, btrace.sml, and btimp.sml.
2236 :
2237 :			----------------------------------------------------------------------
2238 :	blume	675	Name: Matthias Blume
2239 :	blume	678	Date: 2000/06/27 16:51:00 JST
2240 :			Tag: blume-20000627-noweb
2241 :			Description:
2242 :
2243 :			Changes:
2244 :
2245 :			1. Implemented "subdir" and "witness" options for noweb tool.
2246 :			This caused some slight internal changes in CM's tool implementation.
2247 :			2. Fixed bug in "tool plugin" mechanism. This is essentially cleaning
2248 :			some remaining issues from earlier path anchor changes.
2249 :			3. Updated CM manual accordingly.
2250 :
2251 :			4. Changed implementation of back-tracing so that I now consider it
2252 :			ready for prime-time.
2253 :
2254 :			In particular, you don't have to explicitly trigger the back-trace
2255 :			anymore. Instead, if you are running BTrace-instrumented code and
2256 :			there is an uncaught exception (regardless of whether or not it was
2257 :			raised in instrumented code), the top-level evalloop will print
2258 :			the back-trace.
2259 :
2260 :			Features:
2261 :
2262 :			- Instrumented and uninstrumented code work together seemlessly.
2263 :			(Of course, uninstrumented code is never mentioned in actual
2264 :			back-traces.)
2265 :
2266 :			- Asymptotic time- and space-complexity of instrumented code is
2267 :			equal to that of uninstrumented code. (This means that
2268 :			tail-recursion is preserved by the instrumentation phase.)
2269 :
2270 :			- Modules whose code has been instrumented in different sessions
2271 :			work together without problem.
2272 :
2273 :			- There is no penalty whatsoever on uninstrumented code.
2274 :
2275 :			- There is no penalty on "raise" expressions, even in
2276 :			instrumented code.
2277 :
2278 :			A potential bug (or perhaps it is a feature, too):
2279 :
2280 :			A back-trace reaches no further than the outermost instrumented
2281 :			non-trivial "raise". Here, a "trivial" raise is one that is the
2282 :			sole RHS of a "handle" rule. Thus, back-traces reach trough
2283 :
2284 :			<exp> handle e => raise e
2285 :
2286 :			and even
2287 :
2288 :			<exp> handle Foo => raise Bar
2289 :
2290 :			and, of course, through
2291 :
2292 :			<exp> handle Foo => ...
2293 :
2294 :			if the exception was not Foo.
2295 :
2296 :			Back-traces always reach right through any un-instrumented code
2297 :			including any of its "handle" expressions, trivial or not.
2298 :
2299 :			To try this out, do the following:
2300 :
2301 :			- Erase all existing binfiles for your program.
2302 :			(You may keep binfiles for those modules where you think you
2303 :			definitely don't need back-tracing.)
2304 :			- Turn on back-trace instrumentation:
2305 :			SMLofNJ.Internals.BTrace.mode (SOME true);
2306 :			- Recompile your program. (I.e., run "CM.make" or "use".)
2307 :			- You may now turn instrumentation off again (if you want):
2308 :			SMLofNJ.Internals.BTrace.mode (SOME false);
2309 :			- Run your program as usual. If it raises an exception that
2310 :			reaches the interactive toplevel, then a back-trace will
2311 :			automatically be printed. After that, the toplevel loop
2312 :			will print the exception history as usual.
2313 :
2314 :			----------------------------------------------------------------------
2315 :			Name: Matthias Blume
2316 :	blume	677	Date: 2000/06/26 09:56:46 JST
2317 :			Tag: blume-20000626-setup
2318 :			Description:
2319 :
2320 :			CM: - setup-parameter to "sml" added; this can be used to run arbitrary
2321 :			ML code before and after compiling a file (e.g., to set compiler
2322 :			flags)
2323 :
2324 :			Compiler: - improved btrace API (in core.sml, internals.{sig,sml})
2325 :			- associated changes to btrace.sml (BTrace instrumentation pass)
2326 :			- cleaner implementation of btimp.sml (BTrace tracing and report
2327 :			module)
2328 :
2329 :			CM manual: * new path encoding documented
2330 :			* description of setup-parameter to "sml" added
2331 :
2332 :			The biggest user-visible change to back-tracing is that it is no
2333 :			longer necessary to compile all traced modules within the same
2334 :			session. (This was a real limitation.)
2335 :
2336 :			----------------------------------------------------------------------
2337 :			Name: Matthias Blume
2338 :	blume	676	Date: 2000/06/24 12:40:00 JST
2339 :			Tag: blume-20000624-startup
2340 :			Description:
2341 :
2342 :			Fixes startup slowdown problem. (I was calling SrcPath.sync a _tad_
2343 :			bit too often -- to put it mildly. :)
2344 :
2345 :			----------------------------------------------------------------------
2346 :			Name: Matthias Blume
2347 :	blume	675	Date: 2000/06/23 18:20:00 JST
2348 :			Tag: blume-20000623-btrace
2349 :			Description:
2350 :
2351 :			This updates adds a backtrace facility to aid programmers in debugging
2352 :			their programs. This involves the following changes:
2353 :
2354 :			1. Module system/smlnj/init/core.sml (structure _Core) now has hooks for
2355 :			keeping track of the current call stack. When programs are compiled
2356 :			in a special mode, the compiler will insert calls to these hooks
2357 :			into the user program.
2358 :			"Hook" means that it is possible for different implementations of
2359 :			back-tracing to register themselves (at different times).
2360 :
2361 :			2. compiler/MiscUtil/profile/btrace.sml implements the annotation phase
2362 :			as an Absyn.dec->Absyn.dec rewrite. Normally this phase is turned off.
2363 :			It can be turned on using this call:
2364 :			SMLofNJ.Internals.BTrace.mode (SOME true);
2365 :			Turning it off again:
2366 :			SMLofNJ.Internals.BTrace.mode (SOME false);
2367 :			Querying the current status:
2368 :			SMLofNJ.Internals.BTrace.mode NONE;
2369 :			Annotated programs are about twice as big as normal ones, and they
2370 :			run a factor of 2 to 4 slower with a dummy back-trace plugin (one
2371 :			where all hooks do nothing). The slowdown with a plugin that is
2372 :			actually useful (such as the one supplied by default) is even greater,
2373 :			but in the case of the default plugin it is still only an constant
2374 :			factor (amortized).
2375 :
2376 :			3. system/Basis/Implementation/NJ/internals.{sig,sml} have been augmented
2377 :			with a sub-structure BTrace for controlling back-tracing. In particular,
2378 :			the above-mentioned function "mode" controls whether the annotation
2379 :			phase is invoked by the compiler. Another important function is
2380 :			"trigger": when called it aborts the current execution and causes
2381 :			the top-level loop to print a full back-trace.
2382 :
2383 :			4. compiler/MiscUtil/profile/btimp.sml is the current default plugin
2384 :			for back-tracing. It keeps track of the dynamic call stack and in
2385 :			addition to that it keeps a partial history at each "level" of that
2386 :			stack. For example, if a tail-calls b, b tail-calls c, and c tail-calls
2387 :			d and b (at separate times, dynamically), then the report will show:
2388 :
2389 :			GOTO d
2390 :			/c
2391 :			GOTO \b
2392 :			CALL a
2393 :
2394 :			This shows that there was an initial non-tail call of a, then a
2395 :			tail-call to b or c, looping behavior in a cluster of functions that
2396 :			consist of b and c, and then a goto from that cluster (i.e., either from
2397 :			b or from c) to d.
2398 :
2399 :			Note that (depending on the user program) the amount of information
2400 :			that the back-trace module has to keep track of at each level is bounded
2401 :			by a constant. Thus, the whole implementation has the same asymptotical
2402 :			complexity as the original program (both in space and in time).
2403 :
2404 :			5. compiler/TopLevel/interact/evalloop.sml has been modified to
2405 :			handle the special exception SMLofNJ.Internals.BTrace.BTrace
2406 :			which is raised by the "trigger" function mentioned above.
2407 :
2408 :			Notes on usage:
2409 :
2410 :			- Annotated code works well together with unannotated code:
2411 :			Unannotated calls simply do not show up at all in the backtrace.
2412 :
2413 :			- It is not a good idea to let modules that were annotated during
2414 :			different sessions run at the same time. This is because the compiler
2415 :			chooses small integers to identify individual functions, and there
2416 :			will be clashes if different modules were compiled in separate sessions.
2417 :			(Nothing will crash, and you will even be told about the clashes, but
2418 :			back-trace information will in general not be useful.)
2419 :
2420 :			- Back-tracing can be confused by callcc and capture.
2421 :
2422 :			- The only way of getting a back-trace right now is to explicitly
2423 :			invoke the "trigger" function from your user program. Eventually, we
2424 :			should make every exception carry back-trace information (if
2425 :			available). But since this creates more overhead at "raise"-time
2426 :			(similar to the current exnHistory overhead), I have not yet
2427 :			implemented this. (The implementation will be rather easy.) With
2428 :			exceptions carrying back-trace information, this facility will be even
2429 :			more useful because users don't need to modify their programs...
2430 :
2431 :			- While it is possible to compile the compiler with back-trace
2432 :			annotations turned on (I did it to get some confidence in
2433 :			correctness), you must make absolutely sure that core.sml and
2434 :			btimp.sml are compiled WITHOUT annotation! (core.sml cannot actually
2435 :			be compiled with annotation because there is no core access yet, but
2436 :			if you compile btimp.sml with annotation, then the system will go into
2437 :			an infinite recursion and crash.)
2438 :			Since CM currently does not know about BTrace, the only way to turn
2439 :			annotations on and off for different modules of the compiler is to
2440 :			interrupt CMB.make, change the settings, and re-invoke it. Of course,
2441 :			this is awkward and clumsy.
2442 :
2443 :			Sample sessions:
2444 :
2445 :			Standard ML of New Jersey v110.28.1 [FLINT v1.5], June 5, 2000
2446 :			- SMLofNJ.Internals.BTrace.mode (SOME true);
2447 :			[autoloading]
2448 :			[autoloading done]
2449 :			val it = false : bool
2450 :			- structure X = struct
2451 :			- fun main n = let
2452 :			- fun a (x, 0) = d x
2453 :			- | a (x, n) = b (x, n - 1)
2454 :			- and b (x, n) = c (x, n)
2455 :			- and c (x, n) = a (x, n)
2456 :			- and d x = e (x, 3)
2457 :			- and e (x, 0) = f x
2458 :			- | e (x, n) = e (x, n - 1)
2459 :			- and f 0 = SMLofNJ.Internals.BTrace.trigger ()
2460 :			- | f n = n * g (n - 1)
2461 :			- and g n = a (n, 3)
2462 :			- in
2463 :			- f n
2464 :			- end
2465 :			- end;
2466 :			structure X : sig val main : int -> int end
2467 :			- X.main 3;
2468 :			*** BACK-TRACE ***
2469 :			GOTO stdIn:4.2-13.20: X.main[2].f
2470 :			GOTO-( stdIn:4.2-13.20: X.main[2].e
2471 :			GOTO stdIn:4.2-13.20: X.main[2].d
2472 :			/ stdIn:4.2-13.20: X.main[2].a
2473 :			| stdIn:4.2-13.20: X.main[2].b
2474 :			GOTO-\ stdIn:4.2-13.20: X.main[2].c
2475 :			CALL stdIn:4.2-13.20: X.main[2].g
2476 :			GOTO stdIn:4.2-13.20: X.main[2].f
2477 :			GOTO-( stdIn:4.2-13.20: X.main[2].e
2478 :			GOTO stdIn:4.2-13.20: X.main[2].d
2479 :			/ stdIn:4.2-13.20: X.main[2].a
2480 :			| stdIn:4.2-13.20: X.main[2].b
2481 :			GOTO-\ stdIn:4.2-13.20: X.main[2].c
2482 :			CALL stdIn:4.2-13.20: X.main[2].g
2483 :			GOTO stdIn:4.2-13.20: X.main[2].f
2484 :			GOTO-( stdIn:4.2-13.20: X.main[2].e
2485 :			GOTO stdIn:4.2-13.20: X.main[2].d
2486 :			/ stdIn:4.2-13.20: X.main[2].a
2487 :			| stdIn:4.2-13.20: X.main[2].b
2488 :			GOTO-\ stdIn:4.2-13.20: X.main[2].c
2489 :			CALL stdIn:4.2-13.20: X.main[2].g
2490 :			GOTO stdIn:4.2-13.20: X.main[2].f
2491 :			CALL stdIn:2.15-17.4: X.main[2]
2492 :			-
2493 :
2494 :			(Note that because of a FLINt bug the above code currently does not
2495 :			compile without BTrace turned on.)
2496 :
2497 :			Here is another example, using my modified Tiger compiler:
2498 :
2499 :			Standard ML of New Jersey v110.28.1 [FLINT v1.5], June 5, 2000
2500 :			- SMLofNJ.Internals.BTrace.mode (SOME true);
2501 :			[autoloading]
2502 :			[autoloading done]
2503 :			val it = false : bool
2504 :			- CM.make "sources.cm";
2505 :			[autoloading]
2506 :			...
2507 :			[autoloading done]
2508 :			[scanning sources.cm]
2509 :			[parsing (sources.cm):parse.sml]
2510 :			[creating directory CM/SKEL ...]
2511 :			[parsing (sources.cm):tiger.lex.sml]
2512 :			...
2513 :			[wrote CM/sparc-unix/semant.sml]
2514 :			[compiling (sources.cm):main.sml]
2515 :			[wrote CM/sparc-unix/main.sml]
2516 :			[New bindings added.]
2517 :			val it = true : bool
2518 :			- Main.compile ("../testcases/merge.tig", "foo.out");
2519 :			*** BACK-TRACE ***
2520 :			CALL lib/semant.sml:99.2-396.21: SemantFun[2].transExp.trvar
2521 :			CALL lib/semant.sml:99.2-396.21: SemantFun[2].transExp.trexp
2522 :			CALL lib/semant.sml:289.3-295.22: SemantFun[2].transExp.trexp.check[2]
2523 :			GOTO lib/semant.sml:289.3-295.22: SemantFun[2].transExp.trexp.check[2]
2524 :			CALL lib/semant.sml:99.2-396.21: SemantFun[2].transExp.trexp
2525 :			CALL lib/semant.sml:99.2-396.21: SemantFun[2].transExp.trexp
2526 :			CALL lib/semant.sml:488.3-505.6: SemantFun[2].transDec.trdec[2].transBody[2]
2527 :			/ lib/semant.sml:411.65-543.8: SemantFun[2].transDec
2528 :			CALL-\ lib/semant.sml:413.2-540.9: SemantFun[2].transDec.trdec[2]
2529 :			CALL lib/semant.sml:99.2-396.21: SemantFun[2].transExp.trexp
2530 :			CALL lib/semant.sml:8.52-558.4: SemantFun[2].transProg[2]
2531 :			CALL main.sml:1.18-118.4: Main.compile[2]
2532 :			-
2533 :
2534 :			----------------------------------------------------------------------
2535 :	blume	673	Name: Matthias Blumen
2536 :			Date: 2000/06/21 18:00:00 JST
2537 :			Tag: blume-20000621-manual
2538 :			Description:
2539 :
2540 :			CM manual update: Path environments documented.
2541 :
2542 :			----------------------------------------------------------------------
2543 :	blume	666	Name: Matthias Blume
2544 :	blume	672	Date: 2000/06/19 13:40:00
2545 :			Tag: blume-20000619-manual
2546 :			Description:
2547 :
2548 :			CM manual and system/README update. This only covers the fact that
2549 :			there are no more implicit anchors. (Path environments and the "bind"
2550 :			option to "cm" have yet to be documented.)
2551 :
2552 :			----------------------------------------------------------------------
2553 :			Name: Matthias Blume
2554 :	blume	671	Date: 2000/06/19 11:05:00 JST
2555 :			Tag: blume-20000619-chdir-bugfix
2556 :			Description:
2557 :
2558 :			Fixed a bug in new SrcPath module that sometimes led to a bad chDir call.
2559 :
2560 :			----------------------------------------------------------------------
2561 :			Name: Matthias Blume
2562 :	blume	670	Date: 2000/06/18 22:00:10 JST
2563 :			Tag: blume-20000618-implicit-anchors-really-gone
2564 :			Description:
2565 :
2566 :			I updates the previous HISTORY entry where I forgot to mention that
2567 :			implicit anchors are no longer with us.
2568 :
2569 :			The current update also gets rid of the (now useless) controller
2570 :			CM.Control.implicit_anchors.
2571 :
2572 :			----------------------------------------------------------------------
2573 :			Name: Matthias Blume
2574 :	blume	666	Date: 2000/06/16 17:30:00 JST
2575 :			Tag: blume-20000616-anchorenv
2576 :			Description:
2577 :
2578 :			This patch implements the long anticipated (just kidding :) "anchor
2579 :			environment" mechanism. In the course of doing this, I also
2580 :			re-implemented CM's internal "SrcPath" module from scratch. The new
2581 :			one should be more robust in certain boundary cases. In any case, it
2582 :			is a lot cleaner than its predecessor (IMHO).
2583 :
2584 :			This time, although there is yet another boot file format change, I
2585 :			kept the unpickler backward-compatible. As a result, no new bootfiles
2586 :			are necessary and bootstrapping is straightforward. (You cannot read
2587 :			new bootfiles into an old system, but the other way around is no
2588 :			problem.)
2589 :
2590 :			Visible changes:
2591 :
2592 :	blume	670	** 0. Implicit path anchors (without the leading $-symbol) are no
2593 :			longer recognized at all. This means that such path names are not
2594 :			illegal either. For example, the name basis.cm simply refers to a
2595 :			local file called "basis.cm" (i.e, the name is an ordinary path
2596 :			relative to .cm-files directory). Or, to put it differently, only
2597 :			names that start with $ are anchored paths.
2598 :	blume	666
2599 :	blume	670	** 1. The $<singlearc> abbreviation for $/<singlearc> has finally
2600 :			vanished.
2601 :
2602 :			John (Reppy) had critizised this as soon as I originally proposed and
2603 :	blume	666	implemented it, but at that time I did not really deeply believe
2604 :			him. :) Now I came full-circle because I need the $<singlearc> syntax
2605 :			in another place where it cannot be seen as an abbreviation for
2606 :			$/<singlearc>. To avoid the confusion, $<singlearc> now means what it
2607 :			seems to mean (i.e., it "expands" into the corresponding anchor
2608 :			value).
2609 :
2610 :			However, when paths are used as members in CM description files, it
2611 :			continues to be true that there must be at least another arc after the
2612 :			anchor. This is now enforced separately during semantic analysis
2613 :	blume	670	(i.e., from a lexical/syntactical point of view, the notation is ok.)
2614 :	blume	666
2615 :			** 2. The "cm" class now accepts an option "bind". The option's value
2616 :			is a sub-option list of precisely two items -- one labeled "anchor"
2617 :			and the other one labeled "value". As you might expect, "anchor" is
2618 :			used to specify an anchor name to be bound, and "value" specifies what
2619 :			the anchor is being bound to.
2620 :
2621 :			The value must be a directory name and can be given in either standard
2622 :			syntax (including the possibility that it is itself an anchored path)
2623 :			or native syntax.
2624 :
2625 :			Examples:
2626 :
2627 :			foo.cm (bind:(anchor:bar value:$mystuff/bar))
2628 :			lib.cm (bind:(anchor:a value:"H:\\x\\y\\z")) (* only works under windows *)
2629 :
2630 :			and so on.
2631 :
2632 :			The meaning of this is that the .cm-file will be processed with an
2633 :			augmented anchor environment where the given anchor(s) is/are bound to
2634 :			the given values(s).
2635 :
2636 :			The rationale for having this feature is this: Suppose you are trying
2637 :			to use two different (already stable) libraries a.cm and b.cm (that
2638 :			you perhaps didn't write yourself). Further, suppose each of these
2639 :			two libraries internally uses its own auxiliary library $aux/lib.cm.
2640 :			Normally you would now have a problem because the anchor "lib" can not
2641 :			be bound to more than one value globally. Therefore, the project that
2642 :			uses both a.cm and b.cm must locally redirect the anchor to some other
2643 :			place:
2644 :
2645 :			a.cm (bind:(anchor:lib value:/usr/lib/smlnj/a-stuff))
2646 :			b.cm (bind:(anchor:lib value:/usr/lib/smlnj/b-stuff))
2647 :
2648 :			This hard-wires $lib/aux.cm to /usr/lib/smlnj/a-stuff/aux.cm or
2649 :			/usr/lib/smlnj/b-stuff/aux.cm, respectively.
2650 :
2651 :			Hard-wiring path names is a bit inflexible (and CM will verbosely warn
2652 :			you when you do so at the time of CM.stabilize). Therefore, you can
2653 :			also use an anchored path as the value:
2654 :
2655 :			a.cm (bind:(anchor:lib value:$a-lib))
2656 :			b.cm (bind:(anchor:lib value:$b-lib))
2657 :
2658 :			Now you can globally configure (using the usual CM.Anchor.anchor or
2659 :			pathconfig machinery) bindings for "a-lib" and "b-lib". Since "lib"
2660 :			itself is always locally bound, setting it globally is no longer
2661 :			meaningful or necessary (but it does not hurt either). In fact, "lib"
2662 :			can still be used as a global anchor for separate purposes. As a
2663 :			matter of fact, one can locally define "lib" in terms of a global
2664 :			"lib":
2665 :
2666 :			a.cm (bind:(anchor:lib value:$lib/a))
2667 :			b.cm (bind:(anchor:lib value:$lib/b))
2668 :
2669 :			** 3: The encoding of path names has changed. This affects the way
2670 :			path names are shown in CM's progress report and also the internal
2671 :			protocol encoding used for parallel make.
2672 :
2673 :			The encoding now uses one or more ':'-separated segments. Each
2674 :			segments corresponds to a file that has been specified relative to the
2675 :			file given by its preceding segment. The first segment is either
2676 :			relative to the CWD, absolute, or anchored. Each segment itself is
2677 :			basically a Unix pathname; all segments but the first are relative.
2678 :
2679 :			Example:
2680 :
2681 :			$foo/bar/baz.cm:a/b/c.sml
2682 :
2683 :			This path denotes the file bar/a/b/c.sml relative to the directory
2684 :			denoted by anchor "foo". Notice that the encoding also includes
2685 :			baz.cm which is the .cm-file that listed a/b/c.sml. As usual, such
2686 :			paths are resolved relative to the .cm-files directory, so baz.cm must
2687 :			be ignored to get the "real" pathname.
2688 :
2689 :			To make this fact more obvious, CM puts the names of such "virtual
2690 :			arcs" into parentheses when they appear in progress reports. (No
2691 :			parentheses will appear in the internal protocol encoding.) Thus,
2692 :			what you really see is:
2693 :
2694 :			$foo/bar/(baz.cm):a/b/c.sml
2695 :
2696 :			I find this notation to be much more informative than before.
2697 :
2698 :			Another new feature of the encoding is that special characters
2699 :			including parentheses, colons, (back)slashes, and white space are
2700 :			written as \ddd (where ddd is the decimal encoding of the character).
2701 :
2702 :	blume	670	*** The CM manual still needs to be updated.
2703 :
2704 :	blume	666	----------------------------------------------------------------------
2705 :	leunga	660	Name: Allen Leung
2706 :			Date: 2000/06/15 00:38:00
2707 :			Tag: leunga-20000615-x86-peephole
2708 :
2709 :			x86 Peephole fix by Fermin. Affects c-- and moby only.
2710 :
2711 :			----------------------------------------------------------------------
2712 :	blume	658	Name: Matthias Blume
2713 :			Date: 2000/06/12 11:40:00
2714 :			Tag: blume-20000612-parmakefix
2715 :			Description:
2716 :
2717 :			More cleanup after changing the file naming scheme: This time I
2718 :			repaired the parallel make mechanism for CMB.make which I broke earlier.
2719 :
2720 :			----------------------------------------------------------------------
2721 :	leunga	657	Name: Allen Leung
2722 :			Date: 2000/06/09 01:25:00
2723 :			Tag: leunga-20000609-various
2724 :
2725 :			None of these things should affect normal SML/NJ operations
2726 :
2727 :			1. Peephole improvements provided by Fermin (c--)
2728 :			2. New annotation DEFUSE for adding extra dependence (moby)
2729 :			3. New X86 LOCK instructions (moby)
2730 :			4. New machine description language for reservation tables (scheduling)
2731 :			5. Fixes to various optimization/analysis modules (branch chaining, dominator
2732 :			trees etc.)
2733 :			6. I've changed the CM files so that they can work with versions
2734 :			110.0.6, 110.25 and 110.28
2735 :
2736 :			----------------------------------------------------------------------
2737 :	blume	652	Name: Matthias Blume
2738 :	blume	656	Date: 2000/06/09 12:40:00
2739 :			Tag: blume-20000609-log
2740 :			Description:
2741 :
2742 :			- Removed all(?) remaining RCS Log entries from sources.
2743 :
2744 :			- Fixed bug in ml-yacc and ml-lex sources (use explicit anchors for
2745 :			anchored paths).
2746 :
2747 :			----------------------------------------------------------------------
2748 :			Name: Matthias Blume
2749 :	blume	653	Date: 2000/06/07 17:00:00 JST
2750 :			Tag: blume-20000607-no-implicit-anchors
2751 :			Description:
2752 :
2753 :			1. This update changes the default setting for
2754 :			CM.Control.implicit_anchors from true to false. This means that
2755 :			implicit anchors are no longer permitted by default. I also tried to
2756 :			make sure that nothing else still relies on implicit anchors.
2757 :			(This is the next step on the schedule towards a CM that does not even
2758 :			have the notion of implicit anchors anymore.)
2759 :
2760 :			2. More CM manual updates.
2761 :
2762 :			3. I managed to track down and fix the pickling bug I mentioned last
2763 :			time. Because of the previously existing workaround, this entails no
2764 :			immediate practical changes.
2765 :
2766 :			----------------------------------------------------------------------
2767 :			Name: Matthias Blume
2768 :	blume	652	Date: 2000/06/06 11:15:00 JST
2769 :			Tag: blume-20000606-lazierpickle
2770 :			Description:
2771 :
2772 :			!!!! NEW BOOT FILES !!!!
2773 :
2774 :			* The main purpose of this update is to make library pickles lazier in
2775 :			order to reduce the initial space penalty for autoloading a library.
2776 :			As a result, it is now possible to have $smlnj/compiler.cm
2777 :			pre-registered. This should take care of the many complaints or
2778 :			inquiries about missing structure Compiler. This required changes to
2779 :			CM's internal data structures and small tweaks to some algorithms.
2780 :
2781 :			As a neat additional effect, it is no longer necessary (for the sake
2782 :			of lean heap image files) to distinguish between a "minimal" CM and a
2783 :			"full" CM. Now, there is only one CM (i.e., the "full" version:
2784 :			$smlnj/cm.cm aka $smlnj/cm/full.cm), and it is always available at the
2785 :			interactive top level. ($smlnj/cm/minimal.cm is gone.)
2786 :
2787 :			To make the life of compiler-hackers easier, "makeml" now also
2788 :			pre-registers $smlnj/cmb.cm (aka $smlnj/cmb/current.cm). In other
2789 :			words, after you bootstrap a new sml for the first time, you will not
2790 :			have to autoload $smlnj/cmb.cm again afterwards. (The first time
2791 :			around you will still have to do it, though.)
2792 :
2793 :			* A second change consists of major updates to the CM manual. There
2794 :			are now several appendices with summary information and also a full
2795 :			specification of the CM description file syntax.
2796 :
2797 :			* In directory src/system I added the script "allcross". This script
2798 :			invokes sml and cross-compiles the compiler for all supported
2799 :			architectures. (Useful when providing a new set of boot files.)
2800 :
2801 :			* There seems to be a latent bug in my "lazy pickles" mechanism. I
2802 :			added a small tweak to pickle-util.sml to work around this problem,
2803 :			but it is not a proper fix yet. I will investigate further. (The
2804 :			effect of the bug was an inflation of library pickle size.)
2805 :
2806 :			* Version number increased to 110.28.1 (to avoid compatibility problems).
2807 :
2808 :			----------------------------------------------------------------------
2809 :	leunga	646	Name: Allen Leung
2810 :	leunga	648	Date: 2000/05/25 17:28 EDT
2811 :			Tag: leunga-20000525-ra
2812 :			Description:
2813 :
2814 :			Fixed a bug in freezing phase of the register allocator.
2815 :
2816 :			----------------------------------------------------------------------
2817 :			Name: Allen Leung
2818 :	leunga	647	Date: 2000/05/15 22:53 EDT
2819 :	leunga	646	Tag: leunga-20000515-alpha-x86-ra
2820 :			Description:
2821 :
2822 :			1. Alpha
2823 :
2824 :			Slight cleanup. Removed the instruction SGNXL
2825 :
2826 :			2. X86
2827 :
2828 :			Added the following instructions to the instruction set:
2829 :
2830 :			ROLx, RORx,
2831 :			BTx, BTSx, BTLx, BTRx,
2832 :			XCHGx, and variants with the LOCK prefix
2833 :
2834 :			3. Register Allocation
2835 :
2836 :			The module ra-rewrite-with-renaming has been improved.
2837 :
2838 :	leunga	647	These have no effect on SML/NJ.
2839 :
2840 :	leunga	646	----------------------------------------------------------------------
2841 :	blume	642	Name: Matthias Blume
2842 :	blume	645	Date: 2000/05/15 16:20:00 JST
2843 :			Tag: blume-20000515-lightrebuild
2844 :			Description:
2845 :
2846 :			1. I added an alternative to "-rebuild" to "makeml". The difference is
2847 :			that prior to calling CMB.make' the CM-variable "LIGHT" will be
2848 :			defined. In effect, the command will not build any cross-compiler
2849 :			backends and therefore finish more quickly.
2850 :
2851 :			The "fixpt" script also takes a "-light" switch to be able to use
2852 :			this new facility while compiling for a fixpoint.
2853 :
2854 :			2. I replaced all mentions of anchored paths in group owner specifications
2855 :			with simple relative paths (usually starting with "..").
2856 :			The rationale is that a library's internal workings should not be
2857 :			compromised by the lack of some anchor. (An anchor is necessary
2858 :			for someone who wants to refer to the library by an anchored path,
2859 :			but it should not be necessary to build the same library in the first
2860 :			place.)
2861 :
2862 :			3. I changed the way CM's tool mechanism determines the shell command
2863 :			string used for things like ml-yacc etc. so that it does not break
2864 :			when CM.Control.implicit_anchors is turned off.
2865 :
2866 :			----------------------------------------------------------------------
2867 :			Name: Matthias Blume
2868 :	blume	644	Date: 2000/05/12 18:20:00 JST
2869 :			Tag: blume-20000512-ml-build
2870 :			Description:
2871 :
2872 :			Fixed a bug in config/_ml-build that prevented ml-yacc and ml-lex from
2873 :			getting installed properly (by config/install.sh).
2874 :
2875 :			----------------------------------------------------------------------
2876 :			Name: Matthias Blume
2877 :	blume	643	Date: 2000/05/12 17:30:00 JST
2878 :			Tag: blume-20000512-anchors
2879 :			Description:
2880 :
2881 :			!!! NEW BOOT FILES !!!
2882 :
2883 :			This change is in preparation of fading out support for "implicitly
2884 :			anchored path names". I went through all sources and used the
2885 :			explicit (and relatively new) $-notation. See system/README and the
2886 :			CM manual for more info on this.
2887 :
2888 :			I also modified the anchoring scheme for some things such as "smlnj",
2889 :			"MLRISC", "cm", etc. to take advantage of the fact that explicit
2890 :			anchors are more expressive: anchor name and first arc do not have to
2891 :			coincide. This entails the following user-visible change:
2892 :
2893 :			You have to write $smlnj/foo/bar instead of smlnj/foo/bar. In
2894 :			particular, when you fire up sml with a command-line argument, say,
2895 :			e.g.:
2896 :
2897 :			sml '$smlnj/cmb.cm'
2898 :
2899 :			At the ML toplevel prompt:
2900 :
2901 :			CM.autoload "$smlnj/cmb.cm";
2902 :
2903 :			There is also a new controller in CM.Control that can be used to turn
2904 :			off all remaining support for implicit anchors by saying:
2905 :
2906 :			CM.autoload "$smlnj/
2907 :			#set CM.Control.implicit_anchors false;
2908 :
2909 :			This causes CM to reject implicitly anchored paths. This is (for the
2910 :			time being) less permissive than the "final" version where there will
2911 :			be no more such implicit anchors and relative paths will be just that:
2912 :			relative.
2913 :
2914 :			The next step (version after next version?) will be to make the
2915 :			default for CM.Control.implicit_anchors false. After the dust has
2916 :			settled, I can then produce the "final" version of this...
2917 :
2918 :			Note: Since bootstrapping is a bit tricky, I provided new boot files.
2919 :
2920 :			----------------------------------------------------------------------
2921 :			Name: Matthias Blume
2922 :	blume	642	Date: 2000/05/11 16:30:00 JST
2923 :			Tag: blume-20000511-sources
2924 :			Description:
2925 :
2926 :			The main change is that I added function CM.sources as a generalized
2927 :			version of the earlier CM.makedepend. This entails the following
2928 :			additional changes:
2929 :
2930 :			- CM.makedepend has been dropped.
2931 :
2932 :			- CM manual has been updated.
2933 :
2934 :			- TOOLS signature and API have been changed.
2935 :
2936 :			----------------------------------------------------------------------
2937 :	leunga	640	Name: Allen Leung
2938 :	leunga	641	Date: 2000/05/10 21:17 EDT
2939 :			Tag: leunga-20000510-moby-c--ssa
2940 :			Description:
2941 :
2942 :			Various bug fixes and new features for C--, Moby and MLRISC optimizations.
2943 :			None of these affect SML/NJ.
2944 :
2945 :			1. Register Allocation
2946 :
2947 :			a. A new ra spilling module (ra/ra-spill-with-renaming) is implemented.
2948 :			This module tries to remove local (i.e. basic block level) redundancies
2949 :			during spilling.
2950 :
2951 :			b. A new framework for performing region based register allocation.
2952 :			Not yet entirely functional.
2953 :
2954 :			2. X86
2955 :
2956 :			a. DefUse for POP was missing the stack pointer [found by Lal]
2957 :			b. Reload for CALL was incorrect in X86Spill [found by John]
2958 :			c. Various fixes in X86Spill so that it can be used correctly for
2959 :			the new spilling module.
2960 :
2961 :			3. SSA/IR
2962 :
2963 :			a. New module ir/dj-dataflow.sml implements elimination based
2964 :			data flow analysis.
2965 :
2966 :			4. MLRiscGen
2967 :
2968 :			a. Fix for gc type annotation
2969 :
2970 :			5. MDGen
2971 :
2972 :			Various fixes for machine description -> ml code translation. For ssa
2973 :			only.
2974 :
2975 :			----------------------------------------------------------------------
2976 :			Name: Allen Leung
2977 :	leunga	640	Date: 2000/05/08 22:17 EDT
2978 :			Tag: leunga-20000508-labexp
2979 :			Description:
2980 :
2981 :			Fermin has found a few assembly problems with constant expressions
2982 :			generated in LabelExp. Mostly, the problems involve extra parentheses,
2983 :			which choke on dumb assemblers. This is his fix.
2984 :
2985 :			----------------------------------------------------------------------
2986 :	dbm	635	Name: Dave MacQueen
2987 :			Date: 2000/04/09 14:00 EDT
2988 :			Tag: dbm-20000502-Version_110_28
2989 :			Description:
2990 :
2991 :			1. Updated src/compiler/TopLevel/main/version.sml to version 110.28
2992 :
2993 :			2. Updated config/version to 110.28
2994 :
2995 :			3. Updated config/srcarchiveurl
2996 :
2997 :			3. New boot files!
2998 :			ftp://ftp.research.bell-labs.com/dist/smlnj/working/110.28/
2999 :
3000 :			----------------------------------------------------------------------
3001 :	blume	634	Name: Matthias Blume
3002 :			Date: 2000/05/01 19:05:00 JST
3003 :			Tag: blume-20000501-noweb
3004 :			Description:
3005 :
3006 :			A new noweb tool has been added. The existing system is entirely
3007 :			unaffected by this, but some CM users have asked for renewed noweb
3008 :			support. Everything is documented in the CM manual.
3009 :
3010 :			New (plugin) libraries:
3011 :
3012 :			noweb-tool.cm
3013 :			nw-ext.cm
3014 :
3015 :			----------------------------------------------------------------------
3016 :	dbm	633	Name: Dave MacQueen
3017 :			Date: 2000/04/30 12:40PM EDT
3018 :			Tag: dbm-20000430-bug_fixes
3019 :			Description:
3020 :
3021 :			1. Fix for bug 1498
3022 :			smlnj/src/system/Basis/Implementation/Unsafe/object.sig
3023 :			smlnj/src/system/Basis/Implementation/Unsafe/object.sml
3024 :			added toRealArray function
3025 :			smlnj/src/compiler/MiscUtil/print/ppobj.sml
3026 :			added check for tag Obj.RealArray to array printing case in ppObj
3027 :
3028 :			2. Fix for bug 1510
3029 :			smlnj/src/compiler/Semant/types/typesutil.sml
3030 :			fixed definition of dummyargs (used by equalTycon) so that
3031 :			dummy args are distinct types
3032 :
3033 :			----------------------------------------------------------------------
3034 :	blume	629	Name: Matthias Blume
3035 :	blume	632	Date: 2000/04/30 01:00:00 JST
3036 :			Tag: blume-20000430-versions
3037 :			Description:
3038 :
3039 :			1. CM version numbering added. This is an implementation of Lal's
3040 :			proposal for adding version numbers and version checking to .cm
3041 :			files. Lal said that his proposal was just that -- a proposal.
3042 :			For the time being I went ahead and implemented it so that people
3043 :			can comment on it. Everything is completely backward-compatible
3044 :			(except for the stable library format, i.e., new bootfiles!).
3045 :
3046 :			As usual, see the CM manual for details.
3047 :
3048 :			2. An alternative syntax for anchored paths has been implemented.
3049 :			Dave has recently voiced the same concerns that I had when I did
3050 :			this, so there should be some support. My take is that eventually
3051 :			I will let support for the current syntax (where anchors are
3052 :			"implicit") fade out in favor of the new, explicit syntax.
3053 :			In order to be backward-compatible, both old and new syntax are
3054 :			currently supported.
3055 :
3056 :			Again, see the CM manual for details.
3057 :
3058 :			3. Parallel make is trying to be slightly smarter: When the master
3059 :			process finds a "bottleneck", i.e., when there is only one
3060 :			compilation unit that can be compiled and everybody else is
3061 :			waiting on it, then it will simply compile it directly instead
3062 :			of clumsily telling one of the slaves to do it.
3063 :
3064 :			4. Support for "unsharing" added. This is necessary in order to be
3065 :			able to have two different versions of the same library running
3066 :			at the same time (e.g., for trying out a new MLRISC while still
3067 :			having the old MLRISC linked into the current compiler, etc.)
3068 :			See the CM manual.
3069 :
3070 :			5. Simple "makedepend" functionality added for generating Makefile
3071 :			dependency information. (This is rather crude at the moment.
3072 :			Expect some changes here in the future.)
3073 :
3074 :			6. ".fun" added as a recognized suffix for ML files. Also documented
3075 :			explicitly in the manual that the fallback behavior (unknown suffix
3076 :			-> ML file) is not an official feature!
3077 :
3078 :			7. Small changes to the pickler for stable libraries.
3079 :
3080 :			8. Several internal changes to CM (for cleanup/improvement).
3081 :
3082 :
3083 :			!!!! NEW BINFILES !!!!
3084 :
3085 :			----------------------------------------------------------------------
3086 :			Name: Matthias Blume
3087 :	blume	631	Date: 2000/04/28 17:30:00 JST
3088 :			Tag: blume-20000428-pathconfig
3089 :			Description:
3090 :
3091 :			1. I changed config/install.sh to remove duplicate entries from the
3092 :			lib/pathconfig file at the end. Moreover, the final version of
3093 :			lib/pathconfig is sorted alphabetically. The same (sorting) is done
3094 :			in src/system/installml.
3095 :
3096 :			2. The config/install.sh script now consistently uses relative
3097 :			pathnames in lib/pathconfig whenever the anchor is in the lib
3098 :			directory. (So far this was true for the libraries that come
3099 :			pre-compiled and bundled as part of the bootfiles but not for
3100 :			libraries that are compiled by the script itself.)
3101 :
3102 :			----------------------------------------------------------------------
3103 :			Name: Matthias Blume
3104 :	blume	629	Date: 2000/04/26 13:10:00 JST
3105 :			Tag: blume-20000426-fun_suffix
3106 :			Description:
3107 :
3108 :			Added ".fun" as a recognized file name suffix (for ML code).
3109 :
3110 :			----------------------------------------------------------------------
3111 :	leunga	628	Name: Allen Leung
3112 :			Date: 2000/04/25 17:00:00 EST
3113 :			Tag: leunga-20000425-alpha-ra
3114 :			Description:
3115 :
3116 :			1. Alpha
3117 :
3118 :			PSEUDOARITH was missing in AlphaRewrite. This causes an endless loop
3119 :			in C--.
3120 :
3121 :			2. RA
3122 :
3123 :			Added a flag "ra-dump-size" to print out the size of the flowgraph
3124 :			and the interference graph.
3125 :
3126 :			----------------------------------------------------------------------
3127 :	dbm	627	Name: Dave MacQueen
3128 :			Date: 2000/04/25/
3129 :			Tag: dbm-20000425-mlyacc_doc_examples
3130 :			Description:
3131 :			Updated mlyacc.tex sections 5 and 7 for SML '97 and CM.
3132 :			Updated all three examples in src/ml-yacc/examples to run
3133 :			under 110.* using CM.make.
3134 :
3135 :			----------------------------------------------------------------------
3136 :	leunga	624	Name: Allen Leung
3137 :			Date: 2000/04/20 23:04:00 EST
3138 :			Tag: leunga-20000420-ssa-c---stuff
3139 :			Description:
3140 :
3141 :			This update synchronizes my repository with Yale's. Most of these
3142 :			changes, however, do not affect SML/NJ at all (the RA is an exception).
3143 :
3144 :			1. Register Allocator
3145 :
3146 :			a. An improvement in the interference graph construction:
3147 :			Given a copy
3148 :
3149 :			s <- t
3150 :
3151 :	leunga	628	no interference edge between s and t is added for this definition of s.
3152 :	leunga	624
3153 :			b. I've added two new spill heuristic modules that Fermin and I developed
3154 :			(in the new library RA.cm). These are unused in SML/NJ but maybe
3155 :			useful for others (Moby?)
3156 :
3157 :			2. X86
3158 :
3159 :			a. Various fixes in the backend provided by Fermin [C--] and Lal.
3160 :
3161 :			3. Alpha
3162 :
3163 :			a. Added the BSR instruction and code generation that goes with it [C--]
3164 :			b. Other fixes too numerous to recount provided by Fermin [C--]
3165 :
3166 :			4. Regmaps
3167 :
3168 :			a. The regmaps are not initialized with the identity physical bindings
3169 :			at creation time. This is unneeded.
3170 :
3171 :			5. MLRISC Optimizations
3172 :
3173 :			a. The DJ-Graph module can now compute the iterated dominance frontiers
3174 :			intersects with liveness incrementally in linear time! Woohoo!
3175 :			This is now used in my new SSA construction algorithm.
3176 :
3177 :			b. THe branch reorganization module is now smarter about linear chains of
3178 :			basic blocks.
3179 :
3180 :
3181 :			----------------------------------------------------------------------
3182 :	blume	615	Name: Matthias Blume
3183 :			Date: 2000/04/12 13:52:00 JST
3184 :			Tag: blume_main_v110p27_1
3185 :			Description:
3186 :
3187 :			Changed install.sh script to handle archive files without version number
3188 :			and to use "boot.<arch>-<os>" instead of "sml.boot.<arch>-<os>" for the
3189 :			name of the boot file archive.
3190 :
3191 :			----------------------------------------------------------------------
3192 :	dbm	607	Name: Dave MacQueen
3193 :			Date: 2000/04/09 14:00 EDT
3194 :			Tag: dbm-20000410-Version_110_27
3195 :			Description:
3196 :
3197 :			1. Updated src/compiler/TopLevel/main/version.sml to version 110.27
3198 :
3199 :			2. Updated src/config/version to 110.27
3200 :
3201 :			3. New boot files!
3202 :
3203 :			----------------------------------------------------------------------
3204 :	leunga	606	Name: Allen Leung
3205 :			Date: 2000/04/09 19:09:00 EST
3206 :			Tag: leunga-20000409-misc
3207 :			Description:
3208 :
3209 :			1. Yet another fix for x86 assembly for idivl, imull, mull and friends.
3210 :
3211 :			2. Miscellaneous improvements to MLRISC (unused in sml/nj)
3212 :
3213 :			----------------------------------------------------------------------
3214 :	monnier	605	Name: Stefan
3215 :			Date: 2000/04/07 10:00:00 EDT
3216 :			Tag: monnier-20000406-branch-handling
3217 :			Description:
3218 :
3219 :			Improved handling of branches (mostly those generated from
3220 :			polymorphic equality), removed switchoff and changed the
3221 :			default optimization settings (more cpsopt and less flintopt).
3222 :
3223 :			----------------------------------------------------------------------
3224 :	leunga	601	Name: Allen Leung
3225 :	leunga	602	Date: 2000/04/06 01:30:00 EST
3226 :			Tag: leunga-20000406-peephole-x86-SSA-2
3227 :			Description:
3228 :
3229 :			Forgot a few files.
3230 :
3231 :			----------------------------------------------------------------------
3232 :			Name: Allen Leung
3233 :	leunga	601	Date: 2000/04/06 00:36:00 EST
3234 :			Tag: leunga-20000406-peephole-x86-SSA
3235 :			Description:
3236 :
3237 :			1. New Peephole code
3238 :
3239 :			2. Minor improvement to X86 instruction selection
3240 :
3241 :			3. Various fixes to SSA and machine description -> code translator
3242 :
3243 :			----------------------------------------------------------------------
3244 :	blume	595	Name: Matthias Blume
3245 :			Date: 2000/04/05 12:30:00 JST
3246 :			Tag: blume_main_v110p26p2_3
3247 :			Description:
3248 :
3249 :			This update just merges three minor cosmetic updates to CM's sources
3250 :			to get ready for the 110.27 code freeze on Friday. No functionality
3251 :			has changed.
3252 :
3253 :			----------------------------------------------------------------------
3254 :	leunga	593	Name: Allen Leung
3255 :	leunga	594	Date: 2000/04/04 19:39:00 EST
3256 :			Tag: leunga-20000404-x86-asm
3257 :			Description:
3258 :
3259 :			1. Fixed a problem in X86 assembly.
3260 :
3261 :			Things like
3262 :
3263 :			jmp %eax
3264 :			jmp (%eax)
3265 :
3266 :			should be output as
3267 :
3268 :			jmp *%eax
3269 :			jmp *(%eax)
3270 :
3271 :			2. Assembly output
3272 :
3273 :			Added a new flag
3274 :
3275 :			"asm-indent-copies" (default to false)
3276 :
3277 :			When this flag is on, parallel copies will be indented an extra level.
3278 :
3279 :			----------------------------------------------------------------------
3280 :			Name: Allen Leung
3281 :	leunga	593	Date: 2000/04/04 03:18:00 EST
3282 :			Tag: leunga-20000404-C--Moby
3283 :			Description:
3284 :
3285 :			All of these fixes are related to C--, Moby, and my own optimization
3286 :			stuff; so they shouldn't affect SML/NJ.
3287 :
3288 :			1. X86
3289 :
3290 :			Various fixes related floating point, and extensions.
3291 :
3292 :			2. Alpha
3293 :
3294 :			Some extra patterns related to loads with signed/zero extension
3295 :			provided by Fermin.
3296 :
3297 :			3. Assembly
3298 :
3299 :	monnier	659	When generating assembly, resolve the value of client defined constants,
3300 :	leunga	593	instead of generating symbolic values. This is controlled by the
3301 :			new flag "asm-resolve-constants", which is default to true.
3302 :
3303 :			4. Machine Descriptions
3304 :
3305 :			a. The precedence parser was slightly broken when parsing infixr symbols.
3306 :			b. The type generalizing code had the bound variables reversed, resulting
3307 :			in a problem during arity raising.
3308 :			c. Various fixes in machine descriptions.
3309 :
3310 :			----------------------------------------------------------------------
3311 :	blume	592	Name: Matthias Blume
3312 :			Date: 2000/04/03 16:05:00 JST
3313 :			Tag: blume_main_v110p26p2_2
3314 :			Description:
3315 :
3316 :			I eliminated coreEnv from compInfo. Access to the "Core" structure is
3317 :			now done via the ordinary static environment that is context to each
3318 :			compilation unit.
3319 :
3320 :			To this end, I arranged that instead of "structure Core" as "structure
3321 :			_Core" is bound in the pervasive environment. Core access is done via
3322 :	monnier	659	_Core (which can never be accidentally rebound because _Core is not a
3323 :	blume	592	legal surface-syntax symbol).
3324 :
3325 :			The current solution is much cleaner because the core environment is
3326 :			now simply part of the pervasive environment which is part of every
3327 :			compilation unit's context anyway. In particular, this eliminates all
3328 :			special-case handling that was necessary until now in order to deal
3329 :			with dynamic and symbolic parts of the core environment.
3330 :
3331 :			Remaining hackery (to bind the "magic" symbol _Core) is localized in the
3332 :	monnier	659	compilation manager's bootstrap compiler (actually: in the "init group"
3333 :	blume	592	handling). See the comments in src/system/smlnj/init/init.cmi for
3334 :			more details.
3335 :
3336 :			I also tried to track down all mentions of "Core" (as string argument
3337 :			to Symbol.strSymbol) in the compiler and replaced them with a
3338 :			reference to the new CoreSym.coreSym. Seems cleaner since the actual
3339 :			name appears in one place only.
3340 :
3341 :			Binfile and bootfile format have not changed, but the switchover from
3342 :			the old "init.cmi" to the new one is a bit tricky, so I supplied new
3343 :			bootfiles anyway.
3344 :
3345 :			----------------------------------------------------------------------
3346 :	leunga	591	Name: Allen Leung
3347 :			Date: 2000/04/02 21:17:00 EST
3348 :			Tag: leunga-20000402-mltree
3349 :			Description:
3350 :	leunga	585
3351 :	leunga	591	1. Renamed the constructor CALL in MLTREE by popular demand.
3352 :			2. Added a bunch of files from my repository. These are currently
3353 :			used by other non-SMLNJ backends.
3354 :
3355 :	leunga	576	----------------------------------------------------------------------
3356 :	leunga	590	Name: Allen Leung
3357 :			Date: 2000/03/31 21:15:00 EST
3358 :			Tag: leunga-20000331-aliasing
3359 :			Description:
3360 :
3361 :			This update contains a rewritten (and hopefully more correct) module
3362 :			for extracting aliasing information from CPS.
3363 :
3364 :			To turn on this feature:
3365 :
3366 :			Compiler.Control.CG.memDisambiguate := true
3367 :
3368 :			To pretty print the region information with assembly
3369 :
3370 :			Compiler.Control.MLRISC.getFlag "asm-show-region" := true;
3371 :
3372 :			To control how many levels of aliasing information are printed, use:
3373 :
3374 :			Compiler.Control.MLRISC.getInt "points-to-show-level" := n
3375 :
3376 :			The default of n is 3.
3377 :
3378 :			----------------------------------------------------------------------
3379 :	dbm	589	Name: David MacQueen
3380 :			Date: 2000/03/31 11:15:00 EST
3381 :			Tag: dbm-20000331-runtime_fix
3382 :			Description:
3383 :
3384 :			This update contains:
3385 :
3386 :			1. runtime/c-lib/c-libraries.c
3387 :			includes added in revision 1.2 caused compilation errors on hppa-hpux
3388 :
3389 :			2. fix for bug 1556
3390 :			system/Basis/Implementation/NJ/internal-signals.sml
3391 :
3392 :			----------------------------------------------------------------------
3393 :	blume	587	Name: Matthias Blume
3394 :	blume	588	Date: 2000/03/31 18:00:00 JST
3395 :			Tag: blume_main_v110p26p2_1
3396 :			Description:
3397 :
3398 :			This update contains:
3399 :
3400 :			1. A small change to CM's handling of stable libraries:
3401 :			CM now maintains one "global" modmap that is used for all stable
3402 :			libraries. The use of such a global modmap maximizes sharing and
3403 :			minimizes the need for re-traversing parts of environments during
3404 :			modmap construction. (However, this has minor impact since modmap
3405 :			construction seems to account for just one percent or less of total
3406 :			compile time.)
3407 :
3408 :			2. I added a "genmap" phase to the statistics. This is where I got the
3409 :			"one percent" number (see above).
3410 :
3411 :			3. CM's new tool parameter mechanism just became _even_ better. :)
3412 :			- The parser understands named parameters and recursive options.
3413 :			- The "make" and "shell" tools use these new features.
3414 :			(This makes it a lot easier to cascade these tools.)
3415 :			- There is a small syntax change: named parameters use a
3416 :
3417 :			<name> : ( <option> ... ) or
3418 :			<name> : <string>
3419 :
3420 :			syntax. Previously, named parameters were implemented in an
3421 :			ad-hoc fashion by each tool individually (by parsing strings)
3422 :			and had the form
3423 :
3424 :			<name>=<string>
3425 :
3426 :			See the CM manual for a full description of these issues.
3427 :
3428 :			----------------------------------------------------------------------
3429 :			Name: Matthias Blume
3430 :	blume	587	Date: 2000/03/30 18:00:00 JST
3431 :			Tag: blume_main_v110p26p2_0
3432 :			Description:
3433 :
3434 :			!!!!! WARNING !!!!!!
3435 :			!! New binfiles !!
3436 :			!!!!!!!!!!!!!!!!!!!!
3437 :
3438 :			This update contains:
3439 :
3440 :			1. Moderate changes to CM:
3441 :
3442 :			- Changes to CM's tools mechanism. In particular, it is now possible
3443 :			to have tools that accept additional "command line" parameters
3444 :			(specified in the .cm file at each instance where the tool's class is
3445 :			used).
3446 :
3447 :	monnier	659	This was done to accommodate the new "make" and "shell" tools which
3448 :			facilitate fairly seamless hookup to portions of code managed using
3449 :	blume	587	Makefiles or Shell scripts.
3450 :
3451 :			There are no classes "shared" or "private" anymore. Instead, the
3452 :			sharing annotation is now a parameter to the "sml" class.
3453 :
3454 :			There is a bit of generic machinery for implementing one's own
3455 :			tools that accept command-line parameters. However, I am not yet fully
3456 :			satisfied with that part, so expect changes here in the future.
3457 :
3458 :			All existing tools are described in the CM manual.
3459 :
3460 :	monnier	659	- Slightly better error handling. (CM now suppresses many followup
3461 :	blume	587	error messages that tended to be more annoying than helpful.)
3462 :
3463 :			2. Major changes to the compiler's static environment data structures.
3464 :
3465 :			- no CMStaticEnv anymore.
3466 :			- no CMEnv, no "BareEnvironment" (actually, _only_ BareEnvironment,
3467 :			but it is called Environment), no conversions between different
3468 :			kinds of static environments
3469 :
3470 :			- There is still a notion of a "modmap", but such modmaps are generated
3471 :			on demand at the time when they are needed. This sounds slow, but I
3472 :			sped up the code that generates modmaps enough for this not to lead to
3473 :			a slowdown of the compiler (at least I didn't detect any).
3474 :
3475 :			- To facilitate rapid modmap generation, static environments now
3476 :			contain an (optional) "modtree" structure. Modtree annotations are
3477 :			constructed by the unpickler during unpickling. (This means that
3478 :			the elaborator does not have to worry about modtrees at all.)
3479 :			Modtrees have the advantage that they are compositional in the same
3480 :			way as the environment data structure itself is compositional.
3481 :			As a result, modtrees never hang on to parts of an environment that
3482 :			has already been rendered "stale" by filtering or rebinding.
3483 :
3484 :			- I went through many, many trials and errors before arriving at the
3485 :			current solution. (The initial idea of "linkpaths" did not work.)
3486 :			But the result of all this is that I have touched a lot of files that
3487 :			depend on the "modules" and "types" data structures (most of the
3488 :			elaborator). There were a lot of changes during my "linkpath" trials
3489 :			that could have been reverted to their original state but weren't.
3490 :			Please, don't be too harsh on me for messing with this code a bit more
3491 :	monnier	921	than what was strictly necessary... (I _did_ resist the temptation
3492 :	blume	587	of doing any "global reformatting" to avoid an untimely death at
3493 :			Dave's hands. :)
3494 :
3495 :			- One positive aspect of the previous point: At least I made sure that
3496 :			all files that I touched now compile without warnings (other than
3497 :			"polyEqual").
3498 :
3499 :			- compiler now tends to run "leaner" (i.e., ties up less memory in
3500 :			redundant modmaps)
3501 :
3502 :			----------------------------------------------------------------------
3503 :	leunga	580	Name: Allen Leung
3504 :	leunga	585	Date: 2000/03/29 18:00:00
3505 :			Tag: leunga-20000327-mlriscGen_hppa_alpha_x86
3506 :	leunga	586	Boot files (optional): ftp://react-ilp.cs.nyu.edu/leunga/110.26.1-sml.boot.x86-unix-20000330.tar.gz
3507 :	leunga	585	Description:
3508 :
3509 :			This update contains *MAJOR* changes to the way code is generated from CPS
3510 :			in the module mlriscGen, and in various backend modules.
3511 :
3512 :			CHANGES
3513 :			=======
3514 :
3515 :			1. MLRiscGen: forward propagation fix.
3516 :
3517 :			There was a bug in forward propagation introduced at about the same time
3518 :			as the MLRISC x86 backend, which prohibits coalescing to be
3519 :			performed effectively in loops.
3520 :
3521 :			Effect: speed up of loops in RISC architectures.
3522 :			By itself, this actually slowed down certain benchmarks on the x86.
3523 :
3524 :			2. MLRiscGen: forward propagating addresses from consing.
3525 :
3526 :			I've changed the way consing code is generated. Basically I separated
3527 :			out the initialization part:
3528 :
3529 :			store tag, offset(allocptr)
3530 :			store elem1, offset+4(allocptr)
3531 :			store elem2, offset+8(allocptr)
3532 :			...
3533 :			store elemn, offset+4n(allocptr)
3534 :
3535 :			and the address computation part:
3536 :
3537 :			celladdr <- offset+4+alloctpr
3538 :
3539 :			and move the address computation part
3540 :
3541 :			Effect: register pressure is generally lower as a result. This
3542 :			makes compilation of certain expressions much faster, such as
3543 :			long lists with non-trivial elements.
3544 :
3545 :			[(0,0), (0,0), .... (0,0)]
3546 :
3547 :			3. MLRiscGen: base pointer elimination.
3548 :
3549 :			As part of the linkage mechanism, we generate the sequence:
3550 :
3551 :			L: ... <- start of the code fragment
3552 :
3553 :			L1:
3554 :			base pointer <- linkreg - L1 + L
3555 :
3556 :			The base pointer was then used for computing relocatable addresses
3557 :			in the code fragment. Frequently (such as in lots of continuations)
3558 :			this is not needed. We now eliminate this sequence whenever possible.
3559 :
3560 :			For compile time efficiency, I'm using a very stupid local heuristic.
3561 :			But in general, this should be done as a control flow analysis.
3562 :
3563 :			Effect: Smaller code size. Speed up of most programs.
3564 :
3565 :			4. Hppa back end
3566 :
3567 :			Long jumps in span dependence resolution used to depend on the existence
3568 :			of the base pointer.
3569 :
3570 :			A jump to a long label L was expanded into the following sequence:
3571 :
3572 :			LDIL %hi(L-8192), %r29
3573 :			LDO %lo(L-8192)(%r29), %r29
3574 :			ADD %r29, baseptr, %r29
3575 :			BV,n %r0(%r29)
3576 :
3577 :			In the presence of change (3) above, this will not work. I've changed
3578 :			it so that the following sequence of instructions are generated, which
3579 :			doesn't mention the base pointer at all:
3580 :
3581 :			BL,n L', %r29 /* branch and link, L' + 4 -> %r29 */
3582 :			L': ADDIL L-(L'+4), %r29 /* Compute address of L */
3583 :			BV,n %r0(%r29) /* Jump */
3584 :
3585 :			5. Alpha back end
3586 :
3587 :			New alpha instructions LDB/LDW have been added, as per Fermin's
3588 :			suggestions. This is unrelated to all other changes.
3589 :
3590 :			6. X86 back end
3591 :
3592 :			I've changed andl to testl in the floating point test sequence
3593 :			whenever appropriate. The Intel optimization guide states that
3594 :	monnier	659	testl is preferable to andl.
3595 :	leunga	585
3596 :			7. RA (x86 only)
3597 :
3598 :			I've improved the spill propagation algorithm, using an approximation
3599 :			of maximal weighted independent sets. This seems to be necessary to
3600 :			alleviate the negative effect in light of the slow down in (1).
3601 :
3602 :			I'll write down the algorithm one of these days.
3603 :
3604 :			8. MLRiscGen: frequencies
3605 :
3606 :			I've added an annotation that states that all call gc blocks have zero
3607 :			execution frequencies. This improves register allocation on the x86.
3608 :
3609 :			BENCHMARKS
3610 :			==========
3611 :
3612 :			I've only perform the comparison on 110.25.
3613 :
3614 :			The platforms are:
3615 :
3616 :			HPPA A four processor HP machine (E9000) with 5G of memory.
3617 :			X86 A 300Hhz Pentium II with 128M of memory, and
3618 :			SPARC An Ultra sparc 2 with 512M of memory.
3619 :
3620 :			I used the following parameters for the SML benchmarks:
3621 :
3622 :			@SMLalloc
3623 :			HPPA 256k
3624 :			SPARC 512k
3625 :			X86 256k
3626 :
3627 :			COMPILATION TIME
3628 :			----------------
3629 :			Here are the numbers comparing the compilation times of the compilers.
3630 :			I've only compared 110.25 compiling the new sources versus
3631 :			a fixpoint version of the new compiler compiling the same.
3632 :
3633 :			110.25 New
3634 :			Total Time in RA Spill+Reload Total Time In RA Spill+Reload
3635 :			HPPA 627s 116s 2684+3584 599s 95s 1003+1879
3636 :			SPARC 892s 173s 2891+3870 708s 116s 1004+1880
3637 :			X86 999s 315s 94006+130691 987s 296s 108877+141957
3638 :
3639 :			110.25 New
3640 :			Code Size Code Size
3641 :			HPPA 8596736 8561421
3642 :			SPARC 8974299 8785143
3643 :			X86 9029180 8716783
3644 :
3645 :			So in summary, things are at least as good as before. Dramatic
3646 :			reduction in compilation is obtained on the Sparc; I can't explain it,
3647 :			but it is reproducible. Perhaps someone should try to reproduce this
3648 :			on their own machines.
3649 :
3650 :			SML BENCHMARKS
3651 :			--------------
3652 :
3653 :			On the average, all benchmarks perform at least as well as before.
3654 :
3655 :			HPPA Compilation Time Spill+Reload Run Time
3656 :			110.25 New 110.25 New 110.25 New
3657 :
3658 :			barnesHut 3.158 3.015 4.75% 1+1 0+0 2.980 2.922 2.00%
3659 :			boyer 6.152 5.708 7.77% 0+0 0+0 0.218 0.213 2.34%
3660 :			count-graphs 1.168 1.120 4.32% 0+0 0+0 22.705 23.073 -1.60%
3661 :			fft 0.877 0.792 10.74% 1+3 1+3 0.602 0.587 2.56%
3662 :			knuthBendix 3.180 2.857 11.32% 0+0 0+0 0.675 0.662 2.02%
3663 :			lexgen 6.190 5.290 17.01% 0+0 0+0 0.913 0.788 15.86%
3664 :			life 0.803 0.703 14.22% 25+25 0+0 0.153 0.140 9.52%
3665 :			logic 2.048 2.007 2.08% 6+6 1+1 4.133 4.008 3.12%
3666 :			mandelbrot 0.077 0.080 -4.17% 0+0 0+0 0.765 0.712 7.49%
3667 :			mlyacc 22.932 20.937 9.53% 154+181 32+57 0.468 0.430 8.91%
3668 :			nucleic 5.183 5.060 2.44% 2+2 0+0 0.125 0.120 4.17%
3669 :			ratio-regions 3.357 3.142 6.84% 0+0 0+0 116.225 113.173 2.70%
3670 :			ray 1.283 1.290 -0.52% 0+0 0+0 2.887 2.855 1.11%
3671 :			simple 6.307 6.032 4.56% 28+30 5+7 3.705 3.658 1.28%
3672 :			tsp 0.888 0.862 3.09% 0+0 0+0 7.040 6.893 2.13%
3673 :			vliw 24.378 23.455 3.94% 106+127 25+45 2.758 2.707 1.91%
3674 :			--------------------------------------------------------------------------
3675 :			Average 6.12% 4.09%
3676 :
3677 :			SPARC Compilation Time Spill+Reload Run Time
3678 :			110.25 New 110.25 New 110.25 New
3679 :
3680 :			barnesHut 3.778 3.592 5.20% 2+2 0+0 3.648 3.453 5.65%
3681 :			boyer 6.632 6.110 8.54% 0+0 0+0 0.258 0.242 6.90%
3682 :			count-graphs 1.435 1.325 8.30% 0+0 0+0 33.672 34.737 -3.07%
3683 :			fft 0.980 0.940 4.26% 3+9 2+6 0.838 0.827 1.41%
3684 :			knuthBendix 3.590 3.138 14.39% 0+0 0+0 0.962 0.967 -0.52%
3685 :			lexgen 6.593 6.072 8.59% 1+1 0+0 1.077 1.078 -0.15%
3686 :			life 0.972 0.868 11.90% 26+26 0+0 0.143 0.140 2.38%
3687 :			logic 2.525 2.387 5.80% 7+7 1+1 5.625 5.158 9.05%
3688 :			mandelbrot 0.090 0.093 -3.57% 0+0 0+0 0.855 0.728 17.39%
3689 :			mlyacc 26.732 23.827 12.19% 162+189 32+57 0.550 0.560 -1.79%
3690 :			nucleic 6.233 6.197 0.59% 3+3 0+0 0.163 0.173 -5.77%
3691 :			ratio-regions 3.780 3.507 7.79% 0+0 0+0 133.993 131.035 2.26%
3692 :			ray 1.595 1.550 2.90% 1+1 0+0 3.440 3.418 0.63%
3693 :			simple 6.972 6.487 7.48% 29+32 5+7 3.523 3.525 -0.05%
3694 :			tsp 1.115 1.063 4.86% 0+0 0+0 7.393 7.265 1.77%
3695 :			vliw 27.765 24.818 11.87% 110+135 25+45 2.265 2.135 6.09%
3696 :			----------------------------------------------------------------------------
3697 :			Average 6.94% 2.64%
3698 :
3699 :			X86 Compilation Time Spill+Reload Run Time
3700 :			110.25 New 110.25 New 110.25 New
3701 :
3702 :			barnesHut 5.530 5.420 2.03% 593+893 597+915 3.532 3.440 2.66%
3703 :			boyer 8.768 7.747 13.19% 493+199 301+289 0.327 0.297 10.11%
3704 :			count-graphs 2.040 2.010 1.49% 298+394 315+457 26.578 28.660 -7.26%
3705 :			fft 1.327 1.302 1.92% 112+209 115+210 1.055 0.962 9.71%
3706 :			knuthBendix 5.218 5.475 -4.69% 451+598 510+650 0.928 0.932 -0.36%
3707 :			lexgen 9.970 9.623 3.60% 1014+841 1157+885 0.947 0.928 1.97%
3708 :			life 1.183 1.183 0.00% 162+182 145+148 0.127 0.103 22.58%
3709 :			logic 3.285 3.512 -6.45% 514+684 591+836 5.682 5.577 1.88%
3710 :			mandelbrot 0.147 0.143 2.33% 38+41 33+54 0.703 0.690 1.93%
3711 :			mlyacc 35.457 32.763 8.22% 3496+4564 3611+4860 0.552 0.550 0.30%
3712 :			nucleic 7.100 6.888 3.07% 239+168 201+158 0.175 0.173 0.96%
3713 :			ratio-regions 6.388 6.843 -6.65% 1182+257 981+300 120.142 120.345 -0.17%
3714 :			ray 2.332 2.338 -0.29% 346+398 402+494 3.593 3.540 1.51%
3715 :			simple 9.912 9.903 0.08% 1475+941 1579+1168 3.057 3.178 -3.83%
3716 :			tsp 1.623 1.532 5.98% 266+200 250+211 8.045 7.878 2.12%
3717 :			vliw 33.947 35.470 -4.29% 2629+2774 2877+3171 2.072 1.890 9.61%
3718 :			----------------------------------------------------------------------------
3719 :			Average 1.22% 3.36%
3720 :
3721 :			----------------------------------------------------------------------
3722 :			Name: Allen Leung
3723 :	leunga	583	Date: 2000/03/23 16:25:00
3724 :			Tag: leunga-20000323-fix_x86_alpha
3725 :			Description:
3726 :
3727 :			1. X86 fixes/changes
3728 :
3729 :			a. The old code generated for SETcc was completely wrong.
3730 :			The Intel optimization guide is VERY misleading.
3731 :
3732 :			2. ALPHA fixes/changes
3733 :
3734 :			a. Added the instructions LDBU, LDWU, STB, STW as per Fermin's suggestion.
3735 :			b. Added a new mode byteWordLoadStores to the functor parameter to Alpha()
3736 :			c. Added reassociation code for address computation.
3737 :
3738 :			----------------------------------------------------------------------
3739 :			Name: Allen Leung
3740 :	leunga	580	Date: 2000/03/22 01:23:00
3741 :			Tag: leunga-20000322-fix_x86_hppa_ra
3742 :			Description:
3743 :
3744 :			1. X86 fixes/changes
3745 :
3746 :			a. x86Rewrite bug with MUL3 (found by Lal)
3747 :			b. Added the instructions FSTS, FSTL
3748 :
3749 :			2. PA-RISC fixes/changes
3750 :
3751 :			a. B label should not be a delay slot candidate! Why did this work?
3752 :			b. ADDT(32, REG(32, r), LI n) now generates one instruction instead of two,
3753 :			as it should be.
3754 :			c. The assembly syntax for fstds and fstdd was wrong.
3755 :			d. Added the composite instruction COMICLR/LDO, which is the immediate
3756 :			operand variant of COMCLR/LDO.
3757 :
3758 :			3. Generic MLRISC
3759 :
3760 :			a. shuffle.sml rewritten to be slightly more efficient
3761 :			b. DIV bug in mltree-simplify fixed (found by Fermin)
3762 :
3763 :			4. Register Allocator
3764 :
3765 :			a. I now release the interference graph earlier during spilling.
3766 :			May improve memory usage.
3767 :
3768 :			----------------------------------------------------------------------
3769 :	blume	577	Name: Matthias Blume
3770 :	blume	578	Date: 2000/03/14 14:15:32
3771 :			Tag: blume_main_v110p26p1_2
3772 :			Description:
3773 :
3774 :			1. Tools.registerStdShellCmdTool (from smlnj/cm/tool.cm) takes an
3775 :			additional argument called "template" which is an optional string that
3776 :	monnier	659	specifies the layout of the tool command line. See the CM manual for
3777 :	blume	578	explanation.
3778 :
3779 :	monnier	659	2. A special-purpose tool can be "registered" by simply dropping the
3780 :	blume	578	corresponding <...>-tool.cm (and/or <...>-ext.cm) into the same
3781 :			directory where the .cm file lives that uses this tool. (The
3782 :			behavior/misfeature until now was to look for the tool description
3783 :			files in the current working directory.) As before, tool description
3784 :			files could also be anchored -- in which case they can live anywhere
3785 :			they like. Following the recent e-mail discussion, this change should
3786 :			make it easier to have special-purpose tools that are shipped together
3787 :			with the sources of the program that uses them.
3788 :
3789 :			----------------------------------------------------------------------
3790 :			Name: Matthias Blume
3791 :	blume	577	Date: 2000/03/10 07:48:34
3792 :			Tag: blume_main_v110p26p1_1
3793 :			Description:
3794 :
3795 :			I added a re-written version of Dave's fixpt script to src/system.
3796 :			Changes relative to the original version:
3797 :			- sh-ified (not everybody has ksh)
3798 :			- automatically figures out which architecture it runs on
3799 :			- uses ./makeml a bit more cleverly
3800 :			- never invokes ./installml (and, thus, does not clobber your
3801 :			good and working installation of sml in case something goes wrong)
3802 :			- accepts max iteration count using option "-iter <n>"
3803 :			- accepts a "base" name using option "-base <base>"
3804 :
3805 :			It does not build any extraneous heap images but directly rebuilds
3806 :			bin- and boot-hierarchies using makeml's "-rebuild" switch. Finally,
3807 :			it can incorporate existing bin- and boot- hierarchies. For example,
3808 :			suppose the base is set to "sml" (which is the default). Then it
3809 :			successively builds
3810 :
3811 :			sml.bin.<arch>-unix and sml.boot.<arch>-unix
3812 :			then sml1.bin.<arch>-unix and sml1.boot.<arch>-unix
3813 :			then sml2.bin.<arch>-unix and sml2.boot.<arch>-unix
3814 :			...
3815 :			then sml<n>.bin.<arch>-unix and sml<n>.boot.<arch>-unix
3816 :
3817 :			and so on. If any of these already exist, it will just use what's
3818 :			there. In particular, many people will have the initial set of bin
3819 :			and boot files around, so this saves time for at least one full
3820 :			rebuild. Having sets of the form <base><k>.{bin,boot}.<arch>-unix for
3821 :			<k>=1,2,... is normally not a good idea when invoking fixpt. However,
3822 :			they might be the result of an earlier partial run of fixpt (which
3823 :	monnier	659	perhaps got accidentally killed). In this case, fixpt will quickly
3824 :	blume	577	move through what exists before continuing where it left off earlier,
3825 :			and, thus, saves a lot of time.
3826 :
3827 :			----------------------------------------------------------------------
3828 :	leunga	576	Name: Allen Leung
3829 :			Date: 00/03/10 02:20:00
3830 :			Tag: leunga-20000310-fix_x86_asm_ra
3831 :			Description:
3832 :	dbm	570
3833 :	leunga	576	More assembly output problems involving the indexed addressing mode
3834 :			on the x86 have been found and corrected. Thanks to Fermin Reig for the
3835 :			fix.
3836 :
3837 :			The interface and implementation of the register allocator have been changed
3838 :			slightly to accommodate the possibility to skip the register allocation
3839 :			phases completely and go directly to memory allocation. This is needed
3840 :			for C-- use.
3841 :
3842 :	dbm	570	----------------------------------------------------------------------
3843 :	blume	572	Name: Matthias Blume
3844 :	blume	575	Date: 00/03/09 10:23:53
3845 :			Tag: blume_main_v110p26p1_0
3846 :			Description:
3847 :
3848 :			* Complete re-organization of library names. Many libraries have been
3849 :			consolidated so that they share the same path anchor. For example,
3850 :			all MLRISC-related libraries are anchored at MLRISC, most libraries that
3851 :			are SML/NJ-specific are under "smlnj". Notice that names like
3852 :			host-cmb.cm or host-compiler.cm no longer exist. See system/README
3853 :			for a complete description of the new naming scheme. Quick reference:
3854 :
3855 :			host-cmb.cm -> smlnj/cmb.cm
3856 :			host-compiler.cm -> smlnj/compiler.cm
3857 :			full-cm.cm -> smlnj/cm.cm
3858 :			<arch>-<os>.cm -> smlnj/cmb/<arch>-<os>.cm
3859 :			<arch>-compiler.cm -> smlnj/compiler/<arch>.cm
3860 :
3861 :			* Bug fixes in CM.
3862 :			- exceptions in user code are being passed through (i.e., reach top level)
3863 :			- more bugs in paranoia mode fixed
3864 :			- bug related to checking group owners fixed
3865 :
3866 :			* New install.sh script that automagically fetches archive files:
3867 :			The new file config/srcarchiveurl must contain the URL of the
3868 :			(remote) directory that contains bin files (or other source archives).
3869 :			If install.sh does not find the archive locally, it tries to get
3870 :			it from that remote directory.
3871 :			This should simplify installation further: For machines that have
3872 :			access to the internet, just fetch <version>-config.tgz, unpack it,
3873 :	monnier	659	edit config/targets, and go (run config/install.sh). The script will
3874 :	blume	575	fetch everything else that it might need all by itself.
3875 :
3876 :			For CVS users, this mechanism is not relevant for source archives, but
3877 :			it is convenient for getting new sets of binfiles.
3878 :
3879 :			Archives should be tar files compressed with either gzip, compress, or
3880 :			bzip2. The script recognizes .tgz, .tar, tar.gz, tz, .tar.Z, and .tar.bz2.
3881 :
3882 :			----------------------------------------------------------------------
3883 :			Name: Matthias Blume
3884 :	blume	572	Date: 2000/03/07 04:01:04
3885 :			Tag: blume_main_v110_26_2
3886 :	dbm	570	Description:
3887 :	blume	572	- size info in BOOTLIST
3888 :			* no fixed upper limits for number of bootfiles or length of
3889 :			bootfile names in runtime
3890 :			* falling back to old behavior if no BOOTLIST size info found
3891 :			- allocation size heuristics in .run-sml
3892 :			* tries to read cache size from /proc/cpuinfo (this is important for
3893 :			small-cache Celeron systems!)
3894 :			- install.sh robustified
3895 :			- CM manual updates
3896 :			- paranoid mode
3897 :			* no more CMB.deliver() (i.e., all done by CMB.make())
3898 :			* can re-use existing sml.boot.* files
3899 :			* init.cmi now treated as library
3900 :			* library stamps for consistency checks
3901 :			- sml.boot.<arch>-<os>/PIDMAP file
3902 :			* This file is read by the CM startup code. This is used to minimize
3903 :			the amount of dynamic state that needs to be stowed away for the
3904 :			purpose of sharing between interactive system and user code.
3905 :			- CM.Anchor.anchor instead of CM.Anchor.{set,cancel}
3906 :			* Upon request by Elsa. Anchors now controlled by get-set-pair
3907 :			like most other CM state variables.
3908 :			- Compiler.CMSA eliminated
3909 :			* No longer supported by CM anyway.
3910 :			- fixed bugs in pickler that kept biting Stefan
3911 :			* past refs to past refs (was caused by the possibility that
3912 :			ad-hoc sharing is more discriminating than hash-cons sharing)
3913 :			* integer overflow on LargeInt.minInt
3914 :			- ml-{lex,yacc} build scripts now use new mechanism
3915 :			for building standalone programs
3916 :			- fixed several gcc -Wall warnings that were caused by missing header
3917 :			files, missing initializations, etc., in runtime (not all warnings
3918 :			eliminated, though)

---

root@smlnj-gforge.cs.uchicago.edu	ViewVC Help
Powered by ViewVC 1.0.0