Home My Page Projects Code Snippets Project Openings SML/NJ
Summary Activity Forums Tracker Lists Tasks Docs Surveys News SCM Files

SCM Repository

[smlnj] Annotation of /sml/trunk/HISTORY
ViewVC logotype

Annotation of /sml/trunk/HISTORY

Parent Directory Parent Directory | Revision Log Revision Log


Revision 675 - (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 :     Date:
12 :     Tag: <post-commit CVS tag>
13 :     Description:
14 : leunga 591 ----------------------------------------------------------------------
15 : blume 675 Name: Matthias Blume
16 :     Date: 2000/06/23 18:20:00 JST
17 :     Tag: blume-20000623-btrace
18 :     Description:
19 :    
20 :     This updates adds a backtrace facility to aid programmers in debugging
21 :     their programs. This involves the following changes:
22 :    
23 :     1. Module system/smlnj/init/core.sml (structure _Core) now has hooks for
24 :     keeping track of the current call stack. When programs are compiled
25 :     in a special mode, the compiler will insert calls to these hooks
26 :     into the user program.
27 :     "Hook" means that it is possible for different implementations of
28 :     back-tracing to register themselves (at different times).
29 :    
30 :     2. compiler/MiscUtil/profile/btrace.sml implements the annotation phase
31 :     as an Absyn.dec->Absyn.dec rewrite. Normally this phase is turned off.
32 :     It can be turned on using this call:
33 :     SMLofNJ.Internals.BTrace.mode (SOME true);
34 :     Turning it off again:
35 :     SMLofNJ.Internals.BTrace.mode (SOME false);
36 :     Querying the current status:
37 :     SMLofNJ.Internals.BTrace.mode NONE;
38 :     Annotated programs are about twice as big as normal ones, and they
39 :     run a factor of 2 to 4 slower with a dummy back-trace plugin (one
40 :     where all hooks do nothing). The slowdown with a plugin that is
41 :     actually useful (such as the one supplied by default) is even greater,
42 :     but in the case of the default plugin it is still only an constant
43 :     factor (amortized).
44 :    
45 :     3. system/Basis/Implementation/NJ/internals.{sig,sml} have been augmented
46 :     with a sub-structure BTrace for controlling back-tracing. In particular,
47 :     the above-mentioned function "mode" controls whether the annotation
48 :     phase is invoked by the compiler. Another important function is
49 :     "trigger": when called it aborts the current execution and causes
50 :     the top-level loop to print a full back-trace.
51 :    
52 :     4. compiler/MiscUtil/profile/btimp.sml is the current default plugin
53 :     for back-tracing. It keeps track of the dynamic call stack and in
54 :     addition to that it keeps a partial history at each "level" of that
55 :     stack. For example, if a tail-calls b, b tail-calls c, and c tail-calls
56 :     d and b (at separate times, dynamically), then the report will show:
57 :    
58 :     GOTO d
59 :     /c
60 :     GOTO \b
61 :     CALL a
62 :    
63 :     This shows that there was an initial non-tail call of a, then a
64 :     tail-call to b or c, looping behavior in a cluster of functions that
65 :     consist of b and c, and then a goto from that cluster (i.e., either from
66 :     b or from c) to d.
67 :    
68 :     Note that (depending on the user program) the amount of information
69 :     that the back-trace module has to keep track of at each level is bounded
70 :     by a constant. Thus, the whole implementation has the same asymptotical
71 :     complexity as the original program (both in space and in time).
72 :    
73 :     5. compiler/TopLevel/interact/evalloop.sml has been modified to
74 :     handle the special exception SMLofNJ.Internals.BTrace.BTrace
75 :     which is raised by the "trigger" function mentioned above.
76 :    
77 :     Notes on usage:
78 :    
79 :     - Annotated code works well together with unannotated code:
80 :     Unannotated calls simply do not show up at all in the backtrace.
81 :    
82 :     - It is not a good idea to let modules that were annotated during
83 :     different sessions run at the same time. This is because the compiler
84 :     chooses small integers to identify individual functions, and there
85 :     will be clashes if different modules were compiled in separate sessions.
86 :     (Nothing will crash, and you will even be told about the clashes, but
87 :     back-trace information will in general not be useful.)
88 :    
89 :     - Back-tracing can be confused by callcc and capture.
90 :    
91 :     - The only way of getting a back-trace right now is to explicitly
92 :     invoke the "trigger" function from your user program. Eventually, we
93 :     should make every exception carry back-trace information (if
94 :     available). But since this creates more overhead at "raise"-time
95 :     (similar to the current exnHistory overhead), I have not yet
96 :     implemented this. (The implementation will be rather easy.) With
97 :     exceptions carrying back-trace information, this facility will be even
98 :     more useful because users don't need to modify their programs...
99 :    
100 :     - While it is possible to compile the compiler with back-trace
101 :     annotations turned on (I did it to get some confidence in
102 :     correctness), you must make absolutely sure that core.sml and
103 :     btimp.sml are compiled WITHOUT annotation! (core.sml cannot actually
104 :     be compiled with annotation because there is no core access yet, but
105 :     if you compile btimp.sml with annotation, then the system will go into
106 :     an infinite recursion and crash.)
107 :     Since CM currently does not know about BTrace, the only way to turn
108 :     annotations on and off for different modules of the compiler is to
109 :     interrupt CMB.make, change the settings, and re-invoke it. Of course,
110 :     this is awkward and clumsy.
111 :    
112 :     Sample sessions:
113 :    
114 :     Standard ML of New Jersey v110.28.1 [FLINT v1.5], June 5, 2000
115 :     - SMLofNJ.Internals.BTrace.mode (SOME true);
116 :     [autoloading]
117 :     [autoloading done]
118 :     val it = false : bool
119 :     - structure X = struct
120 :     - fun main n = let
121 :     - fun a (x, 0) = d x
122 :     - | a (x, n) = b (x, n - 1)
123 :     - and b (x, n) = c (x, n)
124 :     - and c (x, n) = a (x, n)
125 :     - and d x = e (x, 3)
126 :     - and e (x, 0) = f x
127 :     - | e (x, n) = e (x, n - 1)
128 :     - and f 0 = SMLofNJ.Internals.BTrace.trigger ()
129 :     - | f n = n * g (n - 1)
130 :     - and g n = a (n, 3)
131 :     - in
132 :     - f n
133 :     - end
134 :     - end;
135 :     structure X : sig val main : int -> int end
136 :     - X.main 3;
137 :     *** BACK-TRACE ***
138 :     GOTO stdIn:4.2-13.20: X.main[2].f
139 :     GOTO-( stdIn:4.2-13.20: X.main[2].e
140 :     GOTO stdIn:4.2-13.20: X.main[2].d
141 :     / stdIn:4.2-13.20: X.main[2].a
142 :     | stdIn:4.2-13.20: X.main[2].b
143 :     GOTO-\ stdIn:4.2-13.20: X.main[2].c
144 :     CALL stdIn:4.2-13.20: X.main[2].g
145 :     GOTO stdIn:4.2-13.20: X.main[2].f
146 :     GOTO-( stdIn:4.2-13.20: X.main[2].e
147 :     GOTO stdIn:4.2-13.20: X.main[2].d
148 :     / stdIn:4.2-13.20: X.main[2].a
149 :     | stdIn:4.2-13.20: X.main[2].b
150 :     GOTO-\ stdIn:4.2-13.20: X.main[2].c
151 :     CALL stdIn:4.2-13.20: X.main[2].g
152 :     GOTO stdIn:4.2-13.20: X.main[2].f
153 :     GOTO-( stdIn:4.2-13.20: X.main[2].e
154 :     GOTO stdIn:4.2-13.20: X.main[2].d
155 :     / stdIn:4.2-13.20: X.main[2].a
156 :     | stdIn:4.2-13.20: X.main[2].b
157 :     GOTO-\ stdIn:4.2-13.20: X.main[2].c
158 :     CALL stdIn:4.2-13.20: X.main[2].g
159 :     GOTO stdIn:4.2-13.20: X.main[2].f
160 :     CALL stdIn:2.15-17.4: X.main[2]
161 :     -
162 :    
163 :     (Note that because of a FLINt bug the above code currently does not
164 :     compile without BTrace turned on.)
165 :    
166 :     Here is another example, using my modified Tiger compiler:
167 :    
168 :     Standard ML of New Jersey v110.28.1 [FLINT v1.5], June 5, 2000
169 :     - SMLofNJ.Internals.BTrace.mode (SOME true);
170 :     [autoloading]
171 :     [autoloading done]
172 :     val it = false : bool
173 :     - CM.make "sources.cm";
174 :     [autoloading]
175 :     ...
176 :     [autoloading done]
177 :     [scanning sources.cm]
178 :     [parsing (sources.cm):parse.sml]
179 :     [creating directory CM/SKEL ...]
180 :     [parsing (sources.cm):tiger.lex.sml]
181 :     ...
182 :     [wrote CM/sparc-unix/semant.sml]
183 :     [compiling (sources.cm):main.sml]
184 :     [wrote CM/sparc-unix/main.sml]
185 :     [New bindings added.]
186 :     val it = true : bool
187 :     - Main.compile ("../testcases/merge.tig", "foo.out");
188 :     *** BACK-TRACE ***
189 :     CALL lib/semant.sml:99.2-396.21: SemantFun[2].transExp.trvar
190 :     CALL lib/semant.sml:99.2-396.21: SemantFun[2].transExp.trexp
191 :     CALL lib/semant.sml:289.3-295.22: SemantFun[2].transExp.trexp.check[2]
192 :     GOTO lib/semant.sml:289.3-295.22: SemantFun[2].transExp.trexp.check[2]
193 :     CALL lib/semant.sml:99.2-396.21: SemantFun[2].transExp.trexp
194 :     CALL lib/semant.sml:99.2-396.21: SemantFun[2].transExp.trexp
195 :     CALL lib/semant.sml:488.3-505.6: SemantFun[2].transDec.trdec[2].transBody[2]
196 :     / lib/semant.sml:411.65-543.8: SemantFun[2].transDec
197 :     CALL-\ lib/semant.sml:413.2-540.9: SemantFun[2].transDec.trdec[2]
198 :     CALL lib/semant.sml:99.2-396.21: SemantFun[2].transExp.trexp
199 :     CALL lib/semant.sml:8.52-558.4: SemantFun[2].transProg[2]
200 :     CALL main.sml:1.18-118.4: Main.compile[2]
201 :     -
202 :    
203 :     ----------------------------------------------------------------------
204 : blume 673 Name: Matthias Blumen
205 :     Date: 2000/06/21 18:00:00 JST
206 :     Tag: blume-20000621-manual
207 :     Description:
208 :    
209 :     CM manual update: Path environments documented.
210 :    
211 :     ----------------------------------------------------------------------
212 : blume 666 Name: Matthias Blume
213 : blume 672 Date: 2000/06/19 13:40:00
214 :     Tag: blume-20000619-manual
215 :     Description:
216 :    
217 :     CM manual and system/README update. This only covers the fact that
218 :     there are no more implicit anchors. (Path environments and the "bind"
219 :     option to "cm" have yet to be documented.)
220 :    
221 :     ----------------------------------------------------------------------
222 :     Name: Matthias Blume
223 : blume 671 Date: 2000/06/19 11:05:00 JST
224 :     Tag: blume-20000619-chdir-bugfix
225 :     Description:
226 :    
227 :     Fixed a bug in new SrcPath module that sometimes led to a bad chDir call.
228 :    
229 :     ----------------------------------------------------------------------
230 :     Name: Matthias Blume
231 : blume 670 Date: 2000/06/18 22:00:10 JST
232 :     Tag: blume-20000618-implicit-anchors-really-gone
233 :     Description:
234 :    
235 :     I updates the previous HISTORY entry where I forgot to mention that
236 :     implicit anchors are no longer with us.
237 :    
238 :     The current update also gets rid of the (now useless) controller
239 :     CM.Control.implicit_anchors.
240 :    
241 :     ----------------------------------------------------------------------
242 :     Name: Matthias Blume
243 : blume 666 Date: 2000/06/16 17:30:00 JST
244 :     Tag: blume-20000616-anchorenv
245 :     Description:
246 :    
247 :     This patch implements the long anticipated (just kidding :) "anchor
248 :     environment" mechanism. In the course of doing this, I also
249 :     re-implemented CM's internal "SrcPath" module from scratch. The new
250 :     one should be more robust in certain boundary cases. In any case, it
251 :     is a lot cleaner than its predecessor (IMHO).
252 :    
253 :     This time, although there is yet another boot file format change, I
254 :     kept the unpickler backward-compatible. As a result, no new bootfiles
255 :     are necessary and bootstrapping is straightforward. (You cannot read
256 :     new bootfiles into an old system, but the other way around is no
257 :     problem.)
258 :    
259 :     Visible changes:
260 :    
261 : blume 670 ** 0. Implicit path anchors (without the leading $-symbol) are no
262 :     longer recognized at all. This means that such path names are not
263 :     illegal either. For example, the name basis.cm simply refers to a
264 :     local file called "basis.cm" (i.e, the name is an ordinary path
265 :     relative to .cm-files directory). Or, to put it differently, only
266 :     names that start with $ are anchored paths.
267 : blume 666
268 : blume 670 ** 1. The $<singlearc> abbreviation for $/<singlearc> has finally
269 :     vanished.
270 :    
271 :     John (Reppy) had critizised this as soon as I originally proposed and
272 : blume 666 implemented it, but at that time I did not really deeply believe
273 :     him. :) Now I came full-circle because I need the $<singlearc> syntax
274 :     in another place where it cannot be seen as an abbreviation for
275 :     $/<singlearc>. To avoid the confusion, $<singlearc> now means what it
276 :     seems to mean (i.e., it "expands" into the corresponding anchor
277 :     value).
278 :    
279 :     However, when paths are used as members in CM description files, it
280 :     continues to be true that there must be at least another arc after the
281 :     anchor. This is now enforced separately during semantic analysis
282 : blume 670 (i.e., from a lexical/syntactical point of view, the notation is ok.)
283 : blume 666
284 :     ** 2. The "cm" class now accepts an option "bind". The option's value
285 :     is a sub-option list of precisely two items -- one labeled "anchor"
286 :     and the other one labeled "value". As you might expect, "anchor" is
287 :     used to specify an anchor name to be bound, and "value" specifies what
288 :     the anchor is being bound to.
289 :    
290 :     The value must be a directory name and can be given in either standard
291 :     syntax (including the possibility that it is itself an anchored path)
292 :     or native syntax.
293 :    
294 :     Examples:
295 :    
296 :     foo.cm (bind:(anchor:bar value:$mystuff/bar))
297 :     lib.cm (bind:(anchor:a value:"H:\\x\\y\\z")) (* only works under windows *)
298 :    
299 :     and so on.
300 :    
301 :     The meaning of this is that the .cm-file will be processed with an
302 :     augmented anchor environment where the given anchor(s) is/are bound to
303 :     the given values(s).
304 :    
305 :     The rationale for having this feature is this: Suppose you are trying
306 :     to use two different (already stable) libraries a.cm and b.cm (that
307 :     you perhaps didn't write yourself). Further, suppose each of these
308 :     two libraries internally uses its own auxiliary library $aux/lib.cm.
309 :     Normally you would now have a problem because the anchor "lib" can not
310 :     be bound to more than one value globally. Therefore, the project that
311 :     uses both a.cm and b.cm must locally redirect the anchor to some other
312 :     place:
313 :    
314 :     a.cm (bind:(anchor:lib value:/usr/lib/smlnj/a-stuff))
315 :     b.cm (bind:(anchor:lib value:/usr/lib/smlnj/b-stuff))
316 :    
317 :     This hard-wires $lib/aux.cm to /usr/lib/smlnj/a-stuff/aux.cm or
318 :     /usr/lib/smlnj/b-stuff/aux.cm, respectively.
319 :    
320 :     Hard-wiring path names is a bit inflexible (and CM will verbosely warn
321 :     you when you do so at the time of CM.stabilize). Therefore, you can
322 :     also use an anchored path as the value:
323 :    
324 :     a.cm (bind:(anchor:lib value:$a-lib))
325 :     b.cm (bind:(anchor:lib value:$b-lib))
326 :    
327 :     Now you can globally configure (using the usual CM.Anchor.anchor or
328 :     pathconfig machinery) bindings for "a-lib" and "b-lib". Since "lib"
329 :     itself is always locally bound, setting it globally is no longer
330 :     meaningful or necessary (but it does not hurt either). In fact, "lib"
331 :     can still be used as a global anchor for separate purposes. As a
332 :     matter of fact, one can locally define "lib" in terms of a global
333 :     "lib":
334 :    
335 :     a.cm (bind:(anchor:lib value:$lib/a))
336 :     b.cm (bind:(anchor:lib value:$lib/b))
337 :    
338 :     ** 3: The encoding of path names has changed. This affects the way
339 :     path names are shown in CM's progress report and also the internal
340 :     protocol encoding used for parallel make.
341 :    
342 :     The encoding now uses one or more ':'-separated segments. Each
343 :     segments corresponds to a file that has been specified relative to the
344 :     file given by its preceding segment. The first segment is either
345 :     relative to the CWD, absolute, or anchored. Each segment itself is
346 :     basically a Unix pathname; all segments but the first are relative.
347 :    
348 :     Example:
349 :    
350 :     $foo/bar/baz.cm:a/b/c.sml
351 :    
352 :     This path denotes the file bar/a/b/c.sml relative to the directory
353 :     denoted by anchor "foo". Notice that the encoding also includes
354 :     baz.cm which is the .cm-file that listed a/b/c.sml. As usual, such
355 :     paths are resolved relative to the .cm-files directory, so baz.cm must
356 :     be ignored to get the "real" pathname.
357 :    
358 :     To make this fact more obvious, CM puts the names of such "virtual
359 :     arcs" into parentheses when they appear in progress reports. (No
360 :     parentheses will appear in the internal protocol encoding.) Thus,
361 :     what you really see is:
362 :    
363 :     $foo/bar/(baz.cm):a/b/c.sml
364 :    
365 :     I find this notation to be much more informative than before.
366 :    
367 :     Another new feature of the encoding is that special characters
368 :     including parentheses, colons, (back)slashes, and white space are
369 :     written as \ddd (where ddd is the decimal encoding of the character).
370 :    
371 : blume 670 *** The CM manual still needs to be updated.
372 :    
373 : blume 666 ----------------------------------------------------------------------
374 : leunga 660 Name: Allen Leung
375 :     Date: 2000/06/15 00:38:00
376 :     Tag: leunga-20000615-x86-peephole
377 :    
378 :     x86 Peephole fix by Fermin. Affects c-- and moby only.
379 :    
380 :     ----------------------------------------------------------------------
381 : blume 658 Name: Matthias Blume
382 :     Date: 2000/06/12 11:40:00
383 :     Tag: blume-20000612-parmakefix
384 :     Description:
385 :    
386 :     More cleanup after changing the file naming scheme: This time I
387 :     repaired the parallel make mechanism for CMB.make which I broke earlier.
388 :    
389 :     ----------------------------------------------------------------------
390 : leunga 657 Name: Allen Leung
391 :     Date: 2000/06/09 01:25:00
392 :     Tag: leunga-20000609-various
393 :    
394 :     None of these things should affect normal SML/NJ operations
395 :    
396 :     1. Peephole improvements provided by Fermin (c--)
397 :     2. New annotation DEFUSE for adding extra dependence (moby)
398 :     3. New X86 LOCK instructions (moby)
399 :     4. New machine description language for reservation tables (scheduling)
400 :     5. Fixes to various optimization/analysis modules (branch chaining, dominator
401 :     trees etc.)
402 :     6. I've changed the CM files so that they can work with versions
403 :     110.0.6, 110.25 and 110.28
404 :    
405 :     ----------------------------------------------------------------------
406 : blume 652 Name: Matthias Blume
407 : blume 656 Date: 2000/06/09 12:40:00
408 :     Tag: blume-20000609-log
409 :     Description:
410 :    
411 :     - Removed all(?) remaining RCS Log entries from sources.
412 :    
413 :     - Fixed bug in ml-yacc and ml-lex sources (use explicit anchors for
414 :     anchored paths).
415 :    
416 :     ----------------------------------------------------------------------
417 :     Name: Matthias Blume
418 : blume 653 Date: 2000/06/07 17:00:00 JST
419 :     Tag: blume-20000607-no-implicit-anchors
420 :     Description:
421 :    
422 :     1. This update changes the default setting for
423 :     CM.Control.implicit_anchors from true to false. This means that
424 :     implicit anchors are no longer permitted by default. I also tried to
425 :     make sure that nothing else still relies on implicit anchors.
426 :     (This is the next step on the schedule towards a CM that does not even
427 :     have the notion of implicit anchors anymore.)
428 :    
429 :     2. More CM manual updates.
430 :    
431 :     3. I managed to track down and fix the pickling bug I mentioned last
432 :     time. Because of the previously existing workaround, this entails no
433 :     immediate practical changes.
434 :    
435 :     ----------------------------------------------------------------------
436 :     Name: Matthias Blume
437 : blume 652 Date: 2000/06/06 11:15:00 JST
438 :     Tag: blume-20000606-lazierpickle
439 :     Description:
440 :    
441 :     !!!! NEW BOOT FILES !!!!
442 :    
443 :     * The main purpose of this update is to make library pickles lazier in
444 :     order to reduce the initial space penalty for autoloading a library.
445 :     As a result, it is now possible to have $smlnj/compiler.cm
446 :     pre-registered. This should take care of the many complaints or
447 :     inquiries about missing structure Compiler. This required changes to
448 :     CM's internal data structures and small tweaks to some algorithms.
449 :    
450 :     As a neat additional effect, it is no longer necessary (for the sake
451 :     of lean heap image files) to distinguish between a "minimal" CM and a
452 :     "full" CM. Now, there is only one CM (i.e., the "full" version:
453 :     $smlnj/cm.cm aka $smlnj/cm/full.cm), and it is always available at the
454 :     interactive top level. ($smlnj/cm/minimal.cm is gone.)
455 :    
456 :     To make the life of compiler-hackers easier, "makeml" now also
457 :     pre-registers $smlnj/cmb.cm (aka $smlnj/cmb/current.cm). In other
458 :     words, after you bootstrap a new sml for the first time, you will not
459 :     have to autoload $smlnj/cmb.cm again afterwards. (The first time
460 :     around you will still have to do it, though.)
461 :    
462 :     * A second change consists of major updates to the CM manual. There
463 :     are now several appendices with summary information and also a full
464 :     specification of the CM description file syntax.
465 :    
466 :     * In directory src/system I added the script "allcross". This script
467 :     invokes sml and cross-compiles the compiler for all supported
468 :     architectures. (Useful when providing a new set of boot files.)
469 :    
470 :     * There seems to be a latent bug in my "lazy pickles" mechanism. I
471 :     added a small tweak to pickle-util.sml to work around this problem,
472 :     but it is not a proper fix yet. I will investigate further. (The
473 :     effect of the bug was an inflation of library pickle size.)
474 :    
475 :     * Version number increased to 110.28.1 (to avoid compatibility problems).
476 :    
477 :     ----------------------------------------------------------------------
478 : leunga 646 Name: Allen Leung
479 : leunga 648 Date: 2000/05/25 17:28 EDT
480 :     Tag: leunga-20000525-ra
481 :     Description:
482 :    
483 :     Fixed a bug in freezing phase of the register allocator.
484 :    
485 :     ----------------------------------------------------------------------
486 :     Name: Allen Leung
487 : leunga 647 Date: 2000/05/15 22:53 EDT
488 : leunga 646 Tag: leunga-20000515-alpha-x86-ra
489 :     Description:
490 :    
491 :     1. Alpha
492 :    
493 :     Slight cleanup. Removed the instruction SGNXL
494 :    
495 :     2. X86
496 :    
497 :     Added the following instructions to the instruction set:
498 :    
499 :     ROLx, RORx,
500 :     BTx, BTSx, BTLx, BTRx,
501 :     XCHGx, and variants with the LOCK prefix
502 :    
503 :     3. Register Allocation
504 :    
505 :     The module ra-rewrite-with-renaming has been improved.
506 :    
507 : leunga 647 These have no effect on SML/NJ.
508 :    
509 : leunga 646 ----------------------------------------------------------------------
510 : blume 642 Name: Matthias Blume
511 : blume 645 Date: 2000/05/15 16:20:00 JST
512 :     Tag: blume-20000515-lightrebuild
513 :     Description:
514 :    
515 :     1. I added an alternative to "-rebuild" to "makeml". The difference is
516 :     that prior to calling CMB.make' the CM-variable "LIGHT" will be
517 :     defined. In effect, the command will not build any cross-compiler
518 :     backends and therefore finish more quickly.
519 :    
520 :     The "fixpt" script also takes a "-light" switch to be able to use
521 :     this new facility while compiling for a fixpoint.
522 :    
523 :     2. I replaced all mentions of anchored paths in group owner specifications
524 :     with simple relative paths (usually starting with "..").
525 :     The rationale is that a library's internal workings should not be
526 :     compromised by the lack of some anchor. (An anchor is necessary
527 :     for someone who wants to refer to the library by an anchored path,
528 :     but it should not be necessary to build the same library in the first
529 :     place.)
530 :    
531 :     3. I changed the way CM's tool mechanism determines the shell command
532 :     string used for things like ml-yacc etc. so that it does not break
533 :     when CM.Control.implicit_anchors is turned off.
534 :    
535 :     ----------------------------------------------------------------------
536 :     Name: Matthias Blume
537 : blume 644 Date: 2000/05/12 18:20:00 JST
538 :     Tag: blume-20000512-ml-build
539 :     Description:
540 :    
541 :     Fixed a bug in config/_ml-build that prevented ml-yacc and ml-lex from
542 :     getting installed properly (by config/install.sh).
543 :    
544 :     ----------------------------------------------------------------------
545 :     Name: Matthias Blume
546 : blume 643 Date: 2000/05/12 17:30:00 JST
547 :     Tag: blume-20000512-anchors
548 :     Description:
549 :    
550 :     !!! NEW BOOT FILES !!!
551 :    
552 :     This change is in preparation of fading out support for "implicitly
553 :     anchored path names". I went through all sources and used the
554 :     explicit (and relatively new) $-notation. See system/README and the
555 :     CM manual for more info on this.
556 :    
557 :     I also modified the anchoring scheme for some things such as "smlnj",
558 :     "MLRISC", "cm", etc. to take advantage of the fact that explicit
559 :     anchors are more expressive: anchor name and first arc do not have to
560 :     coincide. This entails the following user-visible change:
561 :    
562 :     You have to write $smlnj/foo/bar instead of smlnj/foo/bar. In
563 :     particular, when you fire up sml with a command-line argument, say,
564 :     e.g.:
565 :    
566 :     sml '$smlnj/cmb.cm'
567 :    
568 :     At the ML toplevel prompt:
569 :    
570 :     CM.autoload "$smlnj/cmb.cm";
571 :    
572 :     There is also a new controller in CM.Control that can be used to turn
573 :     off all remaining support for implicit anchors by saying:
574 :    
575 :     CM.autoload "$smlnj/
576 :     #set CM.Control.implicit_anchors false;
577 :    
578 :     This causes CM to reject implicitly anchored paths. This is (for the
579 :     time being) less permissive than the "final" version where there will
580 :     be no more such implicit anchors and relative paths will be just that:
581 :     relative.
582 :    
583 :     The next step (version after next version?) will be to make the
584 :     default for CM.Control.implicit_anchors false. After the dust has
585 :     settled, I can then produce the "final" version of this...
586 :    
587 :     Note: Since bootstrapping is a bit tricky, I provided new boot files.
588 :    
589 :     ----------------------------------------------------------------------
590 :     Name: Matthias Blume
591 : blume 642 Date: 2000/05/11 16:30:00 JST
592 :     Tag: blume-20000511-sources
593 :     Description:
594 :    
595 :     The main change is that I added function CM.sources as a generalized
596 :     version of the earlier CM.makedepend. This entails the following
597 :     additional changes:
598 :    
599 :     - CM.makedepend has been dropped.
600 :    
601 :     - CM manual has been updated.
602 :    
603 :     - TOOLS signature and API have been changed.
604 :    
605 :     ----------------------------------------------------------------------
606 : leunga 640 Name: Allen Leung
607 : leunga 641 Date: 2000/05/10 21:17 EDT
608 :     Tag: leunga-20000510-moby-c--ssa
609 :     Description:
610 :    
611 :     Various bug fixes and new features for C--, Moby and MLRISC optimizations.
612 :     None of these affect SML/NJ.
613 :    
614 :     1. Register Allocation
615 :    
616 :     a. A new ra spilling module (ra/ra-spill-with-renaming) is implemented.
617 :     This module tries to remove local (i.e. basic block level) redundancies
618 :     during spilling.
619 :    
620 :     b. A new framework for performing region based register allocation.
621 :     Not yet entirely functional.
622 :    
623 :     2. X86
624 :    
625 :     a. DefUse for POP was missing the stack pointer [found by Lal]
626 :     b. Reload for CALL was incorrect in X86Spill [found by John]
627 :     c. Various fixes in X86Spill so that it can be used correctly for
628 :     the new spilling module.
629 :    
630 :     3. SSA/IR
631 :    
632 :     a. New module ir/dj-dataflow.sml implements elimination based
633 :     data flow analysis.
634 :    
635 :     4. MLRiscGen
636 :    
637 :     a. Fix for gc type annotation
638 :    
639 :     5. MDGen
640 :    
641 :     Various fixes for machine description -> ml code translation. For ssa
642 :     only.
643 :    
644 :     ----------------------------------------------------------------------
645 :     Name: Allen Leung
646 : leunga 640 Date: 2000/05/08 22:17 EDT
647 :     Tag: leunga-20000508-labexp
648 :     Description:
649 :    
650 :     Fermin has found a few assembly problems with constant expressions
651 :     generated in LabelExp. Mostly, the problems involve extra parentheses,
652 :     which choke on dumb assemblers. This is his fix.
653 :    
654 :     ----------------------------------------------------------------------
655 : dbm 635 Name: Dave MacQueen
656 :     Date: 2000/04/09 14:00 EDT
657 :     Tag: dbm-20000502-Version_110_28
658 :     Description:
659 :    
660 :     1. Updated src/compiler/TopLevel/main/version.sml to version 110.28
661 :    
662 :     2. Updated config/version to 110.28
663 :    
664 :     3. Updated config/srcarchiveurl
665 :    
666 :     3. New boot files!
667 :     ftp://ftp.research.bell-labs.com/dist/smlnj/working/110.28/
668 :    
669 :     ----------------------------------------------------------------------
670 : blume 634 Name: Matthias Blume
671 :     Date: 2000/05/01 19:05:00 JST
672 :     Tag: blume-20000501-noweb
673 :     Description:
674 :    
675 :     A new noweb tool has been added. The existing system is entirely
676 :     unaffected by this, but some CM users have asked for renewed noweb
677 :     support. Everything is documented in the CM manual.
678 :    
679 :     New (plugin) libraries:
680 :    
681 :     noweb-tool.cm
682 :     nw-ext.cm
683 :    
684 :     ----------------------------------------------------------------------
685 : dbm 633 Name: Dave MacQueen
686 :     Date: 2000/04/30 12:40PM EDT
687 :     Tag: dbm-20000430-bug_fixes
688 :     Description:
689 :    
690 :     1. Fix for bug 1498
691 :     smlnj/src/system/Basis/Implementation/Unsafe/object.sig
692 :     smlnj/src/system/Basis/Implementation/Unsafe/object.sml
693 :     added toRealArray function
694 :     smlnj/src/compiler/MiscUtil/print/ppobj.sml
695 :     added check for tag Obj.RealArray to array printing case in ppObj
696 :    
697 :     2. Fix for bug 1510
698 :     smlnj/src/compiler/Semant/types/typesutil.sml
699 :     fixed definition of dummyargs (used by equalTycon) so that
700 :     dummy args are distinct types
701 :    
702 :     ----------------------------------------------------------------------
703 : blume 629 Name: Matthias Blume
704 : blume 632 Date: 2000/04/30 01:00:00 JST
705 :     Tag: blume-20000430-versions
706 :     Description:
707 :    
708 :     1. CM version numbering added. This is an implementation of Lal's
709 :     proposal for adding version numbers and version checking to .cm
710 :     files. Lal said that his proposal was just that -- a proposal.
711 :     For the time being I went ahead and implemented it so that people
712 :     can comment on it. Everything is completely backward-compatible
713 :     (except for the stable library format, i.e., new bootfiles!).
714 :    
715 :     As usual, see the CM manual for details.
716 :    
717 :     2. An alternative syntax for anchored paths has been implemented.
718 :     Dave has recently voiced the same concerns that I had when I did
719 :     this, so there should be some support. My take is that eventually
720 :     I will let support for the current syntax (where anchors are
721 :     "implicit") fade out in favor of the new, explicit syntax.
722 :     In order to be backward-compatible, both old and new syntax are
723 :     currently supported.
724 :    
725 :     Again, see the CM manual for details.
726 :    
727 :     3. Parallel make is trying to be slightly smarter: When the master
728 :     process finds a "bottleneck", i.e., when there is only one
729 :     compilation unit that can be compiled and everybody else is
730 :     waiting on it, then it will simply compile it directly instead
731 :     of clumsily telling one of the slaves to do it.
732 :    
733 :     4. Support for "unsharing" added. This is necessary in order to be
734 :     able to have two different versions of the same library running
735 :     at the same time (e.g., for trying out a new MLRISC while still
736 :     having the old MLRISC linked into the current compiler, etc.)
737 :     See the CM manual.
738 :    
739 :     5. Simple "makedepend" functionality added for generating Makefile
740 :     dependency information. (This is rather crude at the moment.
741 :     Expect some changes here in the future.)
742 :    
743 :     6. ".fun" added as a recognized suffix for ML files. Also documented
744 :     explicitly in the manual that the fallback behavior (unknown suffix
745 :     -> ML file) is not an official feature!
746 :    
747 :     7. Small changes to the pickler for stable libraries.
748 :    
749 :     8. Several internal changes to CM (for cleanup/improvement).
750 :    
751 :    
752 :     !!!! NEW BINFILES !!!!
753 :    
754 :     ----------------------------------------------------------------------
755 :     Name: Matthias Blume
756 : blume 631 Date: 2000/04/28 17:30:00 JST
757 :     Tag: blume-20000428-pathconfig
758 :     Description:
759 :    
760 :     1. I changed config/install.sh to remove duplicate entries from the
761 :     lib/pathconfig file at the end. Moreover, the final version of
762 :     lib/pathconfig is sorted alphabetically. The same (sorting) is done
763 :     in src/system/installml.
764 :    
765 :     2. The config/install.sh script now consistently uses relative
766 :     pathnames in lib/pathconfig whenever the anchor is in the lib
767 :     directory. (So far this was true for the libraries that come
768 :     pre-compiled and bundled as part of the bootfiles but not for
769 :     libraries that are compiled by the script itself.)
770 :    
771 :     ----------------------------------------------------------------------
772 :     Name: Matthias Blume
773 : blume 629 Date: 2000/04/26 13:10:00 JST
774 :     Tag: blume-20000426-fun_suffix
775 :     Description:
776 :    
777 :     Added ".fun" as a recognized file name suffix (for ML code).
778 :    
779 :     ----------------------------------------------------------------------
780 : leunga 628 Name: Allen Leung
781 :     Date: 2000/04/25 17:00:00 EST
782 :     Tag: leunga-20000425-alpha-ra
783 :     Description:
784 :    
785 :     1. Alpha
786 :    
787 :     PSEUDOARITH was missing in AlphaRewrite. This causes an endless loop
788 :     in C--.
789 :    
790 :     2. RA
791 :    
792 :     Added a flag "ra-dump-size" to print out the size of the flowgraph
793 :     and the interference graph.
794 :    
795 :     ----------------------------------------------------------------------
796 : dbm 627 Name: Dave MacQueen
797 :     Date: 2000/04/25/
798 :     Tag: dbm-20000425-mlyacc_doc_examples
799 :     Description:
800 :     Updated mlyacc.tex sections 5 and 7 for SML '97 and CM.
801 :     Updated all three examples in src/ml-yacc/examples to run
802 :     under 110.* using CM.make.
803 :    
804 :     ----------------------------------------------------------------------
805 : leunga 624 Name: Allen Leung
806 :     Date: 2000/04/20 23:04:00 EST
807 :     Tag: leunga-20000420-ssa-c---stuff
808 :     Description:
809 :    
810 :     This update synchronizes my repository with Yale's. Most of these
811 :     changes, however, do not affect SML/NJ at all (the RA is an exception).
812 :    
813 :     1. Register Allocator
814 :    
815 :     a. An improvement in the interference graph construction:
816 :     Given a copy
817 :    
818 :     s <- t
819 :    
820 : leunga 628 no interference edge between s and t is added for this definition of s.
821 : leunga 624
822 :     b. I've added two new spill heuristic modules that Fermin and I developed
823 :     (in the new library RA.cm). These are unused in SML/NJ but maybe
824 :     useful for others (Moby?)
825 :    
826 :     2. X86
827 :    
828 :     a. Various fixes in the backend provided by Fermin [C--] and Lal.
829 :    
830 :     3. Alpha
831 :    
832 :     a. Added the BSR instruction and code generation that goes with it [C--]
833 :     b. Other fixes too numerous to recount provided by Fermin [C--]
834 :    
835 :     4. Regmaps
836 :    
837 :     a. The regmaps are not initialized with the identity physical bindings
838 :     at creation time. This is unneeded.
839 :    
840 :     5. MLRISC Optimizations
841 :    
842 :     a. The DJ-Graph module can now compute the iterated dominance frontiers
843 :     intersects with liveness incrementally in linear time! Woohoo!
844 :     This is now used in my new SSA construction algorithm.
845 :    
846 :     b. THe branch reorganization module is now smarter about linear chains of
847 :     basic blocks.
848 :    
849 :    
850 :     ----------------------------------------------------------------------
851 : blume 615 Name: Matthias Blume
852 :     Date: 2000/04/12 13:52:00 JST
853 :     Tag: blume_main_v110p27_1
854 :     Description:
855 :    
856 :     Changed install.sh script to handle archive files without version number
857 :     and to use "boot.<arch>-<os>" instead of "sml.boot.<arch>-<os>" for the
858 :     name of the boot file archive.
859 :    
860 :     ----------------------------------------------------------------------
861 : dbm 607 Name: Dave MacQueen
862 :     Date: 2000/04/09 14:00 EDT
863 :     Tag: dbm-20000410-Version_110_27
864 :     Description:
865 :    
866 :     1. Updated src/compiler/TopLevel/main/version.sml to version 110.27
867 :    
868 :     2. Updated src/config/version to 110.27
869 :    
870 :     3. New boot files!
871 :    
872 :     ----------------------------------------------------------------------
873 : leunga 606 Name: Allen Leung
874 :     Date: 2000/04/09 19:09:00 EST
875 :     Tag: leunga-20000409-misc
876 :     Description:
877 :    
878 :     1. Yet another fix for x86 assembly for idivl, imull, mull and friends.
879 :    
880 :     2. Miscellaneous improvements to MLRISC (unused in sml/nj)
881 :    
882 :     ----------------------------------------------------------------------
883 : monnier 605 Name: Stefan
884 :     Date: 2000/04/07 10:00:00 EDT
885 :     Tag: monnier-20000406-branch-handling
886 :     Description:
887 :    
888 :     Improved handling of branches (mostly those generated from
889 :     polymorphic equality), removed switchoff and changed the
890 :     default optimization settings (more cpsopt and less flintopt).
891 :    
892 :     ----------------------------------------------------------------------
893 : leunga 601 Name: Allen Leung
894 : leunga 602 Date: 2000/04/06 01:30:00 EST
895 :     Tag: leunga-20000406-peephole-x86-SSA-2
896 :     Description:
897 :    
898 :     Forgot a few files.
899 :    
900 :     ----------------------------------------------------------------------
901 :     Name: Allen Leung
902 : leunga 601 Date: 2000/04/06 00:36:00 EST
903 :     Tag: leunga-20000406-peephole-x86-SSA
904 :     Description:
905 :    
906 :     1. New Peephole code
907 :    
908 :     2. Minor improvement to X86 instruction selection
909 :    
910 :     3. Various fixes to SSA and machine description -> code translator
911 :    
912 :     ----------------------------------------------------------------------
913 : blume 595 Name: Matthias Blume
914 :     Date: 2000/04/05 12:30:00 JST
915 :     Tag: blume_main_v110p26p2_3
916 :     Description:
917 :    
918 :     This update just merges three minor cosmetic updates to CM's sources
919 :     to get ready for the 110.27 code freeze on Friday. No functionality
920 :     has changed.
921 :    
922 :     ----------------------------------------------------------------------
923 : leunga 593 Name: Allen Leung
924 : leunga 594 Date: 2000/04/04 19:39:00 EST
925 :     Tag: leunga-20000404-x86-asm
926 :     Description:
927 :    
928 :     1. Fixed a problem in X86 assembly.
929 :    
930 :     Things like
931 :    
932 :     jmp %eax
933 :     jmp (%eax)
934 :    
935 :     should be output as
936 :    
937 :     jmp *%eax
938 :     jmp *(%eax)
939 :    
940 :     2. Assembly output
941 :    
942 :     Added a new flag
943 :    
944 :     "asm-indent-copies" (default to false)
945 :    
946 :     When this flag is on, parallel copies will be indented an extra level.
947 :    
948 :     ----------------------------------------------------------------------
949 :     Name: Allen Leung
950 : leunga 593 Date: 2000/04/04 03:18:00 EST
951 :     Tag: leunga-20000404-C--Moby
952 :     Description:
953 :    
954 :     All of these fixes are related to C--, Moby, and my own optimization
955 :     stuff; so they shouldn't affect SML/NJ.
956 :    
957 :     1. X86
958 :    
959 :     Various fixes related floating point, and extensions.
960 :    
961 :     2. Alpha
962 :    
963 :     Some extra patterns related to loads with signed/zero extension
964 :     provided by Fermin.
965 :    
966 :     3. Assembly
967 :    
968 : monnier 659 When generating assembly, resolve the value of client defined constants,
969 : leunga 593 instead of generating symbolic values. This is controlled by the
970 :     new flag "asm-resolve-constants", which is default to true.
971 :    
972 :     4. Machine Descriptions
973 :    
974 :     a. The precedence parser was slightly broken when parsing infixr symbols.
975 :     b. The type generalizing code had the bound variables reversed, resulting
976 :     in a problem during arity raising.
977 :     c. Various fixes in machine descriptions.
978 :    
979 :     ----------------------------------------------------------------------
980 : blume 592 Name: Matthias Blume
981 :     Date: 2000/04/03 16:05:00 JST
982 :     Tag: blume_main_v110p26p2_2
983 :     Description:
984 :    
985 :     I eliminated coreEnv from compInfo. Access to the "Core" structure is
986 :     now done via the ordinary static environment that is context to each
987 :     compilation unit.
988 :    
989 :     To this end, I arranged that instead of "structure Core" as "structure
990 :     _Core" is bound in the pervasive environment. Core access is done via
991 : monnier 659 _Core (which can never be accidentally rebound because _Core is not a
992 : blume 592 legal surface-syntax symbol).
993 :    
994 :     The current solution is much cleaner because the core environment is
995 :     now simply part of the pervasive environment which is part of every
996 :     compilation unit's context anyway. In particular, this eliminates all
997 :     special-case handling that was necessary until now in order to deal
998 :     with dynamic and symbolic parts of the core environment.
999 :    
1000 :     Remaining hackery (to bind the "magic" symbol _Core) is localized in the
1001 : monnier 659 compilation manager's bootstrap compiler (actually: in the "init group"
1002 : blume 592 handling). See the comments in src/system/smlnj/init/init.cmi for
1003 :     more details.
1004 :    
1005 :     I also tried to track down all mentions of "Core" (as string argument
1006 :     to Symbol.strSymbol) in the compiler and replaced them with a
1007 :     reference to the new CoreSym.coreSym. Seems cleaner since the actual
1008 :     name appears in one place only.
1009 :    
1010 :     Binfile and bootfile format have not changed, but the switchover from
1011 :     the old "init.cmi" to the new one is a bit tricky, so I supplied new
1012 :     bootfiles anyway.
1013 :    
1014 :     ----------------------------------------------------------------------
1015 : leunga 591 Name: Allen Leung
1016 :     Date: 2000/04/02 21:17:00 EST
1017 :     Tag: leunga-20000402-mltree
1018 :     Description:
1019 : leunga 585
1020 : leunga 591 1. Renamed the constructor CALL in MLTREE by popular demand.
1021 :     2. Added a bunch of files from my repository. These are currently
1022 :     used by other non-SMLNJ backends.
1023 :    
1024 : leunga 576 ----------------------------------------------------------------------
1025 : leunga 590 Name: Allen Leung
1026 :     Date: 2000/03/31 21:15:00 EST
1027 :     Tag: leunga-20000331-aliasing
1028 :     Description:
1029 :    
1030 :     This update contains a rewritten (and hopefully more correct) module
1031 :     for extracting aliasing information from CPS.
1032 :    
1033 :     To turn on this feature:
1034 :    
1035 :     Compiler.Control.CG.memDisambiguate := true
1036 :    
1037 :     To pretty print the region information with assembly
1038 :    
1039 :     Compiler.Control.MLRISC.getFlag "asm-show-region" := true;
1040 :    
1041 :     To control how many levels of aliasing information are printed, use:
1042 :    
1043 :     Compiler.Control.MLRISC.getInt "points-to-show-level" := n
1044 :    
1045 :     The default of n is 3.
1046 :    
1047 :     ----------------------------------------------------------------------
1048 : dbm 589 Name: David MacQueen
1049 :     Date: 2000/03/31 11:15:00 EST
1050 :     Tag: dbm-20000331-runtime_fix
1051 :     Description:
1052 :    
1053 :     This update contains:
1054 :    
1055 :     1. runtime/c-lib/c-libraries.c
1056 :     includes added in revision 1.2 caused compilation errors on hppa-hpux
1057 :    
1058 :     2. fix for bug 1556
1059 :     system/Basis/Implementation/NJ/internal-signals.sml
1060 :    
1061 :     ----------------------------------------------------------------------
1062 : blume 587 Name: Matthias Blume
1063 : blume 588 Date: 2000/03/31 18:00:00 JST
1064 :     Tag: blume_main_v110p26p2_1
1065 :     Description:
1066 :    
1067 :     This update contains:
1068 :    
1069 :     1. A small change to CM's handling of stable libraries:
1070 :     CM now maintains one "global" modmap that is used for all stable
1071 :     libraries. The use of such a global modmap maximizes sharing and
1072 :     minimizes the need for re-traversing parts of environments during
1073 :     modmap construction. (However, this has minor impact since modmap
1074 :     construction seems to account for just one percent or less of total
1075 :     compile time.)
1076 :    
1077 :     2. I added a "genmap" phase to the statistics. This is where I got the
1078 :     "one percent" number (see above).
1079 :    
1080 :     3. CM's new tool parameter mechanism just became _even_ better. :)
1081 :     - The parser understands named parameters and recursive options.
1082 :     - The "make" and "shell" tools use these new features.
1083 :     (This makes it a lot easier to cascade these tools.)
1084 :     - There is a small syntax change: named parameters use a
1085 :    
1086 :     <name> : ( <option> ... ) or
1087 :     <name> : <string>
1088 :    
1089 :     syntax. Previously, named parameters were implemented in an
1090 :     ad-hoc fashion by each tool individually (by parsing strings)
1091 :     and had the form
1092 :    
1093 :     <name>=<string>
1094 :    
1095 :     See the CM manual for a full description of these issues.
1096 :    
1097 :     ----------------------------------------------------------------------
1098 :     Name: Matthias Blume
1099 : blume 587 Date: 2000/03/30 18:00:00 JST
1100 :     Tag: blume_main_v110p26p2_0
1101 :     Description:
1102 :    
1103 :     !!!!! WARNING !!!!!!
1104 :     !! New binfiles !!
1105 :     !!!!!!!!!!!!!!!!!!!!
1106 :    
1107 :     This update contains:
1108 :    
1109 :     1. Moderate changes to CM:
1110 :    
1111 :     - Changes to CM's tools mechanism. In particular, it is now possible
1112 :     to have tools that accept additional "command line" parameters
1113 :     (specified in the .cm file at each instance where the tool's class is
1114 :     used).
1115 :    
1116 : monnier 659 This was done to accommodate the new "make" and "shell" tools which
1117 :     facilitate fairly seamless hookup to portions of code managed using
1118 : blume 587 Makefiles or Shell scripts.
1119 :    
1120 :     There are no classes "shared" or "private" anymore. Instead, the
1121 :     sharing annotation is now a parameter to the "sml" class.
1122 :    
1123 :     There is a bit of generic machinery for implementing one's own
1124 :     tools that accept command-line parameters. However, I am not yet fully
1125 :     satisfied with that part, so expect changes here in the future.
1126 :    
1127 :     All existing tools are described in the CM manual.
1128 :    
1129 : monnier 659 - Slightly better error handling. (CM now suppresses many followup
1130 : blume 587 error messages that tended to be more annoying than helpful.)
1131 :    
1132 :     2. Major changes to the compiler's static environment data structures.
1133 :    
1134 :     - no CMStaticEnv anymore.
1135 :     - no CMEnv, no "BareEnvironment" (actually, _only_ BareEnvironment,
1136 :     but it is called Environment), no conversions between different
1137 :     kinds of static environments
1138 :    
1139 :     - There is still a notion of a "modmap", but such modmaps are generated
1140 :     on demand at the time when they are needed. This sounds slow, but I
1141 :     sped up the code that generates modmaps enough for this not to lead to
1142 :     a slowdown of the compiler (at least I didn't detect any).
1143 :    
1144 :     - To facilitate rapid modmap generation, static environments now
1145 :     contain an (optional) "modtree" structure. Modtree annotations are
1146 :     constructed by the unpickler during unpickling. (This means that
1147 :     the elaborator does not have to worry about modtrees at all.)
1148 :     Modtrees have the advantage that they are compositional in the same
1149 :     way as the environment data structure itself is compositional.
1150 :     As a result, modtrees never hang on to parts of an environment that
1151 :     has already been rendered "stale" by filtering or rebinding.
1152 :    
1153 :     - I went through many, many trials and errors before arriving at the
1154 :     current solution. (The initial idea of "linkpaths" did not work.)
1155 :     But the result of all this is that I have touched a lot of files that
1156 :     depend on the "modules" and "types" data structures (most of the
1157 :     elaborator). There were a lot of changes during my "linkpath" trials
1158 :     that could have been reverted to their original state but weren't.
1159 :     Please, don't be too harsh on me for messing with this code a bit more
1160 :     than what was strictly necessary... (I _did_ resist the tempation
1161 :     of doing any "global reformatting" to avoid an untimely death at
1162 :     Dave's hands. :)
1163 :    
1164 :     - One positive aspect of the previous point: At least I made sure that
1165 :     all files that I touched now compile without warnings (other than
1166 :     "polyEqual").
1167 :    
1168 :     - compiler now tends to run "leaner" (i.e., ties up less memory in
1169 :     redundant modmaps)
1170 :    
1171 :     ----------------------------------------------------------------------
1172 : leunga 580 Name: Allen Leung
1173 : leunga 585 Date: 2000/03/29 18:00:00
1174 :     Tag: leunga-20000327-mlriscGen_hppa_alpha_x86
1175 : leunga 586 Boot files (optional): ftp://react-ilp.cs.nyu.edu/leunga/110.26.1-sml.boot.x86-unix-20000330.tar.gz
1176 : leunga 585 Description:
1177 :    
1178 :     This update contains *MAJOR* changes to the way code is generated from CPS
1179 :     in the module mlriscGen, and in various backend modules.
1180 :    
1181 :     CHANGES
1182 :     =======
1183 :    
1184 :     1. MLRiscGen: forward propagation fix.
1185 :    
1186 :     There was a bug in forward propagation introduced at about the same time
1187 :     as the MLRISC x86 backend, which prohibits coalescing to be
1188 :     performed effectively in loops.
1189 :    
1190 :     Effect: speed up of loops in RISC architectures.
1191 :     By itself, this actually slowed down certain benchmarks on the x86.
1192 :    
1193 :     2. MLRiscGen: forward propagating addresses from consing.
1194 :    
1195 :     I've changed the way consing code is generated. Basically I separated
1196 :     out the initialization part:
1197 :    
1198 :     store tag, offset(allocptr)
1199 :     store elem1, offset+4(allocptr)
1200 :     store elem2, offset+8(allocptr)
1201 :     ...
1202 :     store elemn, offset+4n(allocptr)
1203 :    
1204 :     and the address computation part:
1205 :    
1206 :     celladdr <- offset+4+alloctpr
1207 :    
1208 :     and move the address computation part
1209 :    
1210 :     Effect: register pressure is generally lower as a result. This
1211 :     makes compilation of certain expressions much faster, such as
1212 :     long lists with non-trivial elements.
1213 :    
1214 :     [(0,0), (0,0), .... (0,0)]
1215 :    
1216 :     3. MLRiscGen: base pointer elimination.
1217 :    
1218 :     As part of the linkage mechanism, we generate the sequence:
1219 :    
1220 :     L: ... <- start of the code fragment
1221 :    
1222 :     L1:
1223 :     base pointer <- linkreg - L1 + L
1224 :    
1225 :     The base pointer was then used for computing relocatable addresses
1226 :     in the code fragment. Frequently (such as in lots of continuations)
1227 :     this is not needed. We now eliminate this sequence whenever possible.
1228 :    
1229 :     For compile time efficiency, I'm using a very stupid local heuristic.
1230 :     But in general, this should be done as a control flow analysis.
1231 :    
1232 :     Effect: Smaller code size. Speed up of most programs.
1233 :    
1234 :     4. Hppa back end
1235 :    
1236 :     Long jumps in span dependence resolution used to depend on the existence
1237 :     of the base pointer.
1238 :    
1239 :     A jump to a long label L was expanded into the following sequence:
1240 :    
1241 :     LDIL %hi(L-8192), %r29
1242 :     LDO %lo(L-8192)(%r29), %r29
1243 :     ADD %r29, baseptr, %r29
1244 :     BV,n %r0(%r29)
1245 :    
1246 :     In the presence of change (3) above, this will not work. I've changed
1247 :     it so that the following sequence of instructions are generated, which
1248 :     doesn't mention the base pointer at all:
1249 :    
1250 :     BL,n L', %r29 /* branch and link, L' + 4 -> %r29 */
1251 :     L': ADDIL L-(L'+4), %r29 /* Compute address of L */
1252 :     BV,n %r0(%r29) /* Jump */
1253 :    
1254 :     5. Alpha back end
1255 :    
1256 :     New alpha instructions LDB/LDW have been added, as per Fermin's
1257 :     suggestions. This is unrelated to all other changes.
1258 :    
1259 :     6. X86 back end
1260 :    
1261 :     I've changed andl to testl in the floating point test sequence
1262 :     whenever appropriate. The Intel optimization guide states that
1263 : monnier 659 testl is preferable to andl.
1264 : leunga 585
1265 :     7. RA (x86 only)
1266 :    
1267 :     I've improved the spill propagation algorithm, using an approximation
1268 :     of maximal weighted independent sets. This seems to be necessary to
1269 :     alleviate the negative effect in light of the slow down in (1).
1270 :    
1271 :     I'll write down the algorithm one of these days.
1272 :    
1273 :     8. MLRiscGen: frequencies
1274 :    
1275 :     I've added an annotation that states that all call gc blocks have zero
1276 :     execution frequencies. This improves register allocation on the x86.
1277 :    
1278 :     BENCHMARKS
1279 :     ==========
1280 :    
1281 :     I've only perform the comparison on 110.25.
1282 :    
1283 :     The platforms are:
1284 :    
1285 :     HPPA A four processor HP machine (E9000) with 5G of memory.
1286 :     X86 A 300Hhz Pentium II with 128M of memory, and
1287 :     SPARC An Ultra sparc 2 with 512M of memory.
1288 :    
1289 :     I used the following parameters for the SML benchmarks:
1290 :    
1291 :     @SMLalloc
1292 :     HPPA 256k
1293 :     SPARC 512k
1294 :     X86 256k
1295 :    
1296 :     COMPILATION TIME
1297 :     ----------------
1298 :     Here are the numbers comparing the compilation times of the compilers.
1299 :     I've only compared 110.25 compiling the new sources versus
1300 :     a fixpoint version of the new compiler compiling the same.
1301 :    
1302 :     110.25 New
1303 :     Total Time in RA Spill+Reload Total Time In RA Spill+Reload
1304 :     HPPA 627s 116s 2684+3584 599s 95s 1003+1879
1305 :     SPARC 892s 173s 2891+3870 708s 116s 1004+1880
1306 :     X86 999s 315s 94006+130691 987s 296s 108877+141957
1307 :    
1308 :     110.25 New
1309 :     Code Size Code Size
1310 :     HPPA 8596736 8561421
1311 :     SPARC 8974299 8785143
1312 :     X86 9029180 8716783
1313 :    
1314 :     So in summary, things are at least as good as before. Dramatic
1315 :     reduction in compilation is obtained on the Sparc; I can't explain it,
1316 :     but it is reproducible. Perhaps someone should try to reproduce this
1317 :     on their own machines.
1318 :    
1319 :     SML BENCHMARKS
1320 :     --------------
1321 :    
1322 :     On the average, all benchmarks perform at least as well as before.
1323 :    
1324 :     HPPA Compilation Time Spill+Reload Run Time
1325 :     110.25 New 110.25 New 110.25 New
1326 :    
1327 :     barnesHut 3.158 3.015 4.75% 1+1 0+0 2.980 2.922 2.00%
1328 :     boyer 6.152 5.708 7.77% 0+0 0+0 0.218 0.213 2.34%
1329 :     count-graphs 1.168 1.120 4.32% 0+0 0+0 22.705 23.073 -1.60%
1330 :     fft 0.877 0.792 10.74% 1+3 1+3 0.602 0.587 2.56%
1331 :     knuthBendix 3.180 2.857 11.32% 0+0 0+0 0.675 0.662 2.02%
1332 :     lexgen 6.190 5.290 17.01% 0+0 0+0 0.913 0.788 15.86%
1333 :     life 0.803 0.703 14.22% 25+25 0+0 0.153 0.140 9.52%
1334 :     logic 2.048 2.007 2.08% 6+6 1+1 4.133 4.008 3.12%
1335 :     mandelbrot 0.077 0.080 -4.17% 0+0 0+0 0.765 0.712 7.49%
1336 :     mlyacc 22.932 20.937 9.53% 154+181 32+57 0.468 0.430 8.91%
1337 :     nucleic 5.183 5.060 2.44% 2+2 0+0 0.125 0.120 4.17%
1338 :     ratio-regions 3.357 3.142 6.84% 0+0 0+0 116.225 113.173 2.70%
1339 :     ray 1.283 1.290 -0.52% 0+0 0+0 2.887 2.855 1.11%
1340 :     simple 6.307 6.032 4.56% 28+30 5+7 3.705 3.658 1.28%
1341 :     tsp 0.888 0.862 3.09% 0+0 0+0 7.040 6.893 2.13%
1342 :     vliw 24.378 23.455 3.94% 106+127 25+45 2.758 2.707 1.91%
1343 :     --------------------------------------------------------------------------
1344 :     Average 6.12% 4.09%
1345 :    
1346 :     SPARC Compilation Time Spill+Reload Run Time
1347 :     110.25 New 110.25 New 110.25 New
1348 :    
1349 :     barnesHut 3.778 3.592 5.20% 2+2 0+0 3.648 3.453 5.65%
1350 :     boyer 6.632 6.110 8.54% 0+0 0+0 0.258 0.242 6.90%
1351 :     count-graphs 1.435 1.325 8.30% 0+0 0+0 33.672 34.737 -3.07%
1352 :     fft 0.980 0.940 4.26% 3+9 2+6 0.838 0.827 1.41%
1353 :     knuthBendix 3.590 3.138 14.39% 0+0 0+0 0.962 0.967 -0.52%
1354 :     lexgen 6.593 6.072 8.59% 1+1 0+0 1.077 1.078 -0.15%
1355 :     life 0.972 0.868 11.90% 26+26 0+0 0.143 0.140 2.38%
1356 :     logic 2.525 2.387 5.80% 7+7 1+1 5.625 5.158 9.05%
1357 :     mandelbrot 0.090 0.093 -3.57% 0+0 0+0 0.855 0.728 17.39%
1358 :     mlyacc 26.732 23.827 12.19% 162+189 32+57 0.550 0.560 -1.79%
1359 :     nucleic 6.233 6.197 0.59% 3+3 0+0 0.163 0.173 -5.77%
1360 :     ratio-regions 3.780 3.507 7.79% 0+0 0+0 133.993 131.035 2.26%
1361 :     ray 1.595 1.550 2.90% 1+1 0+0 3.440 3.418 0.63%
1362 :     simple 6.972 6.487 7.48% 29+32 5+7 3.523 3.525 -0.05%
1363 :     tsp 1.115 1.063 4.86% 0+0 0+0 7.393 7.265 1.77%
1364 :     vliw 27.765 24.818 11.87% 110+135 25+45 2.265 2.135 6.09%
1365 :     ----------------------------------------------------------------------------
1366 :     Average 6.94% 2.64%
1367 :    
1368 :     X86 Compilation Time Spill+Reload Run Time
1369 :     110.25 New 110.25 New 110.25 New
1370 :    
1371 :     barnesHut 5.530 5.420 2.03% 593+893 597+915 3.532 3.440 2.66%
1372 :     boyer 8.768 7.747 13.19% 493+199 301+289 0.327 0.297 10.11%
1373 :     count-graphs 2.040 2.010 1.49% 298+394 315+457 26.578 28.660 -7.26%
1374 :     fft 1.327 1.302 1.92% 112+209 115+210 1.055 0.962 9.71%
1375 :     knuthBendix 5.218 5.475 -4.69% 451+598 510+650 0.928 0.932 -0.36%
1376 :     lexgen 9.970 9.623 3.60% 1014+841 1157+885 0.947 0.928 1.97%
1377 :     life 1.183 1.183 0.00% 162+182 145+148 0.127 0.103 22.58%
1378 :     logic 3.285 3.512 -6.45% 514+684 591+836 5.682 5.577 1.88%
1379 :     mandelbrot 0.147 0.143 2.33% 38+41 33+54 0.703 0.690 1.93%
1380 :     mlyacc 35.457 32.763 8.22% 3496+4564 3611+4860 0.552 0.550 0.30%
1381 :     nucleic 7.100 6.888 3.07% 239+168 201+158 0.175 0.173 0.96%
1382 :     ratio-regions 6.388 6.843 -6.65% 1182+257 981+300 120.142 120.345 -0.17%
1383 :     ray 2.332 2.338 -0.29% 346+398 402+494 3.593 3.540 1.51%
1384 :     simple 9.912 9.903 0.08% 1475+941 1579+1168 3.057 3.178 -3.83%
1385 :     tsp 1.623 1.532 5.98% 266+200 250+211 8.045 7.878 2.12%
1386 :     vliw 33.947 35.470 -4.29% 2629+2774 2877+3171 2.072 1.890 9.61%
1387 :     ----------------------------------------------------------------------------
1388 :     Average 1.22% 3.36%
1389 :    
1390 :     ----------------------------------------------------------------------
1391 :     Name: Allen Leung
1392 : leunga 583 Date: 2000/03/23 16:25:00
1393 :     Tag: leunga-20000323-fix_x86_alpha
1394 :     Description:
1395 :    
1396 :     1. X86 fixes/changes
1397 :    
1398 :     a. The old code generated for SETcc was completely wrong.
1399 :     The Intel optimization guide is VERY misleading.
1400 :    
1401 :     2. ALPHA fixes/changes
1402 :    
1403 :     a. Added the instructions LDBU, LDWU, STB, STW as per Fermin's suggestion.
1404 :     b. Added a new mode byteWordLoadStores to the functor parameter to Alpha()
1405 :     c. Added reassociation code for address computation.
1406 :    
1407 :     ----------------------------------------------------------------------
1408 :     Name: Allen Leung
1409 : leunga 580 Date: 2000/03/22 01:23:00
1410 :     Tag: leunga-20000322-fix_x86_hppa_ra
1411 :     Description:
1412 :    
1413 :     1. X86 fixes/changes
1414 :    
1415 :     a. x86Rewrite bug with MUL3 (found by Lal)
1416 :     b. Added the instructions FSTS, FSTL
1417 :    
1418 :     2. PA-RISC fixes/changes
1419 :    
1420 :     a. B label should not be a delay slot candidate! Why did this work?
1421 :     b. ADDT(32, REG(32, r), LI n) now generates one instruction instead of two,
1422 :     as it should be.
1423 :     c. The assembly syntax for fstds and fstdd was wrong.
1424 :     d. Added the composite instruction COMICLR/LDO, which is the immediate
1425 :     operand variant of COMCLR/LDO.
1426 :    
1427 :     3. Generic MLRISC
1428 :    
1429 :     a. shuffle.sml rewritten to be slightly more efficient
1430 :     b. DIV bug in mltree-simplify fixed (found by Fermin)
1431 :    
1432 :     4. Register Allocator
1433 :    
1434 :     a. I now release the interference graph earlier during spilling.
1435 :     May improve memory usage.
1436 :    
1437 :     ----------------------------------------------------------------------
1438 : blume 577 Name: Matthias Blume
1439 : blume 578 Date: 2000/03/14 14:15:32
1440 :     Tag: blume_main_v110p26p1_2
1441 :     Description:
1442 :    
1443 :     1. Tools.registerStdShellCmdTool (from smlnj/cm/tool.cm) takes an
1444 :     additional argument called "template" which is an optional string that
1445 : monnier 659 specifies the layout of the tool command line. See the CM manual for
1446 : blume 578 explanation.
1447 :    
1448 : monnier 659 2. A special-purpose tool can be "registered" by simply dropping the
1449 : blume 578 corresponding <...>-tool.cm (and/or <...>-ext.cm) into the same
1450 :     directory where the .cm file lives that uses this tool. (The
1451 :     behavior/misfeature until now was to look for the tool description
1452 :     files in the current working directory.) As before, tool description
1453 :     files could also be anchored -- in which case they can live anywhere
1454 :     they like. Following the recent e-mail discussion, this change should
1455 :     make it easier to have special-purpose tools that are shipped together
1456 :     with the sources of the program that uses them.
1457 :    
1458 :     ----------------------------------------------------------------------
1459 :     Name: Matthias Blume
1460 : blume 577 Date: 2000/03/10 07:48:34
1461 :     Tag: blume_main_v110p26p1_1
1462 :     Description:
1463 :    
1464 :     I added a re-written version of Dave's fixpt script to src/system.
1465 :     Changes relative to the original version:
1466 :     - sh-ified (not everybody has ksh)
1467 :     - automatically figures out which architecture it runs on
1468 :     - uses ./makeml a bit more cleverly
1469 :     - never invokes ./installml (and, thus, does not clobber your
1470 :     good and working installation of sml in case something goes wrong)
1471 :     - accepts max iteration count using option "-iter <n>"
1472 :     - accepts a "base" name using option "-base <base>"
1473 :    
1474 :     It does not build any extraneous heap images but directly rebuilds
1475 :     bin- and boot-hierarchies using makeml's "-rebuild" switch. Finally,
1476 :     it can incorporate existing bin- and boot- hierarchies. For example,
1477 :     suppose the base is set to "sml" (which is the default). Then it
1478 :     successively builds
1479 :    
1480 :     sml.bin.<arch>-unix and sml.boot.<arch>-unix
1481 :     then sml1.bin.<arch>-unix and sml1.boot.<arch>-unix
1482 :     then sml2.bin.<arch>-unix and sml2.boot.<arch>-unix
1483 :     ...
1484 :     then sml<n>.bin.<arch>-unix and sml<n>.boot.<arch>-unix
1485 :    
1486 :     and so on. If any of these already exist, it will just use what's
1487 :     there. In particular, many people will have the initial set of bin
1488 :     and boot files around, so this saves time for at least one full
1489 :     rebuild. Having sets of the form <base><k>.{bin,boot}.<arch>-unix for
1490 :     <k>=1,2,... is normally not a good idea when invoking fixpt. However,
1491 :     they might be the result of an earlier partial run of fixpt (which
1492 : monnier 659 perhaps got accidentally killed). In this case, fixpt will quickly
1493 : blume 577 move through what exists before continuing where it left off earlier,
1494 :     and, thus, saves a lot of time.
1495 :    
1496 :     ----------------------------------------------------------------------
1497 : leunga 576 Name: Allen Leung
1498 :     Date: 00/03/10 02:20:00
1499 :     Tag: leunga-20000310-fix_x86_asm_ra
1500 :     Description:
1501 : dbm 570
1502 : leunga 576 More assembly output problems involving the indexed addressing mode
1503 :     on the x86 have been found and corrected. Thanks to Fermin Reig for the
1504 :     fix.
1505 :    
1506 :     The interface and implementation of the register allocator have been changed
1507 :     slightly to accommodate the possibility to skip the register allocation
1508 :     phases completely and go directly to memory allocation. This is needed
1509 :     for C-- use.
1510 :    
1511 : dbm 570 ----------------------------------------------------------------------
1512 : blume 572 Name: Matthias Blume
1513 : blume 575 Date: 00/03/09 10:23:53
1514 :     Tag: blume_main_v110p26p1_0
1515 :     Description:
1516 :    
1517 :     * Complete re-organization of library names. Many libraries have been
1518 :     consolidated so that they share the same path anchor. For example,
1519 :     all MLRISC-related libraries are anchored at MLRISC, most libraries that
1520 :     are SML/NJ-specific are under "smlnj". Notice that names like
1521 :     host-cmb.cm or host-compiler.cm no longer exist. See system/README
1522 :     for a complete description of the new naming scheme. Quick reference:
1523 :    
1524 :     host-cmb.cm -> smlnj/cmb.cm
1525 :     host-compiler.cm -> smlnj/compiler.cm
1526 :     full-cm.cm -> smlnj/cm.cm
1527 :     <arch>-<os>.cm -> smlnj/cmb/<arch>-<os>.cm
1528 :     <arch>-compiler.cm -> smlnj/compiler/<arch>.cm
1529 :    
1530 :     * Bug fixes in CM.
1531 :     - exceptions in user code are being passed through (i.e., reach top level)
1532 :     - more bugs in paranoia mode fixed
1533 :     - bug related to checking group owners fixed
1534 :    
1535 :     * New install.sh script that automagically fetches archive files:
1536 :     The new file config/srcarchiveurl must contain the URL of the
1537 :     (remote) directory that contains bin files (or other source archives).
1538 :     If install.sh does not find the archive locally, it tries to get
1539 :     it from that remote directory.
1540 :     This should simplify installation further: For machines that have
1541 :     access to the internet, just fetch <version>-config.tgz, unpack it,
1542 : monnier 659 edit config/targets, and go (run config/install.sh). The script will
1543 : blume 575 fetch everything else that it might need all by itself.
1544 :    
1545 :     For CVS users, this mechanism is not relevant for source archives, but
1546 :     it is convenient for getting new sets of binfiles.
1547 :    
1548 :     Archives should be tar files compressed with either gzip, compress, or
1549 :     bzip2. The script recognizes .tgz, .tar, tar.gz, tz, .tar.Z, and .tar.bz2.
1550 :    
1551 :     ----------------------------------------------------------------------
1552 :     Name: Matthias Blume
1553 : blume 572 Date: 2000/03/07 04:01:04
1554 :     Tag: blume_main_v110_26_2
1555 : dbm 570 Description:
1556 : blume 572 - size info in BOOTLIST
1557 :     * no fixed upper limits for number of bootfiles or length of
1558 :     bootfile names in runtime
1559 :     * falling back to old behavior if no BOOTLIST size info found
1560 :     - allocation size heuristics in .run-sml
1561 :     * tries to read cache size from /proc/cpuinfo (this is important for
1562 :     small-cache Celeron systems!)
1563 :     - install.sh robustified
1564 :     - CM manual updates
1565 :     - paranoid mode
1566 :     * no more CMB.deliver() (i.e., all done by CMB.make())
1567 :     * can re-use existing sml.boot.* files
1568 :     * init.cmi now treated as library
1569 :     * library stamps for consistency checks
1570 :     - sml.boot.<arch>-<os>/PIDMAP file
1571 :     * This file is read by the CM startup code. This is used to minimize
1572 :     the amount of dynamic state that needs to be stowed away for the
1573 :     purpose of sharing between interactive system and user code.
1574 :     - CM.Anchor.anchor instead of CM.Anchor.{set,cancel}
1575 :     * Upon request by Elsa. Anchors now controlled by get-set-pair
1576 :     like most other CM state variables.
1577 :     - Compiler.CMSA eliminated
1578 :     * No longer supported by CM anyway.
1579 :     - fixed bugs in pickler that kept biting Stefan
1580 :     * past refs to past refs (was caused by the possibility that
1581 :     ad-hoc sharing is more discriminating than hash-cons sharing)
1582 :     * integer overflow on LargeInt.minInt
1583 :     - ml-{lex,yacc} build scripts now use new mechanism
1584 :     for building standalone programs
1585 :     - fixed several gcc -Wall warnings that were caused by missing header
1586 :     files, missing initializations, etc., in runtime (not all warnings
1587 :     eliminated, though)

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