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 774 - (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 : george 749 ----------------------------------------------------------------------
15 : blume 771 Name: Matthias Blume
16 : blume 774 Date: 2001/01/10 21:55:00 JST
17 :     Tag: blume-20010110-rcc
18 :     Description:
19 :    
20 :     The RCC stuff now seems to work (but only on the x86).
21 :     This required hacking of the c-calls interface (and -implementation) in
22 :     MLRISC.
23 :    
24 :     Normal compiler users should be unaffected.
25 :    
26 :     ----------------------------------------------------------------------
27 :     Name: Matthias Blume
28 : blume 773 Date: 2001/01/09 01:20:00 JST
29 :     Tag: blume-20010109-rcc
30 :     Description:
31 :    
32 :     This is a fairly big patch, flushing out a large number of pending
33 :     changes that I made to my development copy over the last couple of days.
34 :    
35 :     Of practical relevance at this moment is a workaround for a pickling
36 :     bug that Allen ran into the other day. The cause of the bug itself is
37 :     still unknown and it might be hard to fix it properly, but the
38 :     workaround has some merits of its own (namely somewhat reducing pickling
39 :     overhead for certain libraries). Therefore, I think this solution should
40 :     be satisfactory at this time.
41 :    
42 :     The rest of the changes (i.e., the vast majority) has to do with my
43 :     ongoing efforts of providing direct support for C function calls from
44 :     ML. At the moment there is a new primop "RAW_CCALL", typing magic
45 :     in types/cproto.sml (invoked from FLINT/trans/translate.sml), a new
46 :     case in the FLINT CPS datatype (RCC), changes to cps/convert.sml to
47 :     translate uses of RAW_CCALL into RCC, and changes to mlriscGen.sml to
48 :     handle RCC.
49 :    
50 :     The last part (the changes to mlriscGen.sml) are still known to be
51 :     wrong on the x86 and not implemented on all other architectures. But
52 :     the infrastructure is in place. I had to change a few functor
53 :     signatures in the backend to be able to route the CCalls interface
54 :     from MLRISC there, and I had to specialize the mltree type (on the
55 :     x86) to include the necessary extensions. (The extensions themselves
56 :     were already there and redy to go in MLRISC/x86).
57 :    
58 :     Everything should be very happy as soon as someone helps me with
59 :     mlriscGen.sml...
60 :    
61 :     In any case, nothing of this should matter to anyone as long as the
62 :     new primop is not being used (which is going to be the case unless you
63 :     find it where I hid it :). The rest of the compiler is completely
64 :     unaffected.
65 :    
66 :     ----------------------------------------------------------------------
67 :     Name: Matthias Blume
68 : blume 772 Date: 2001/01/05 00:30:00 JST
69 :     Tag: blume-20010105-primops
70 :     Description:
71 :    
72 :     Added some experimental support for work that I am doing right now.
73 :     These changes mostly concern added primops, but there is also a new
74 :     experimental C library in the runtime system (but currently not enabled
75 :     anywhere except on Linux/X86).
76 :    
77 :     In the course of adding primops (and playing with them), I discovered that
78 :     Zhong's INL_PRIM hack (no type info for certain primops) was, in fact, badly
79 :     broken. (Zhong was very right he labeled this stuff as "major gross hack".)
80 :     To recover, I made type information in INL_PRIM mandatory and changed
81 :     prim.sml as well as built-in.sml accordingly. The InLine structure now
82 :     has complete, correct type information (i.e., no bottom types).
83 :    
84 :     Since all these changes mean that we need new binfiles, I also bumped the
85 :     version number to 110.32.1.
86 :    
87 :     ----------------------------------------------------------------------
88 :     Name: Matthias Blume
89 : blume 771 Date: 2000/12/30 22:10:00 JST
90 :     Tag: blume-20001230-various
91 :     Description:
92 :    
93 :     Added proxy libraries for MLRISC and let MLRISC libraries refer
94 :     to each other using path anchors. (See CM manual for explanation.)
95 :    
96 :     Updated CM documentation.
97 :    
98 :     Fixed some bugs in CM.
99 :    
100 :     Implemented "proxy" libraries (= syntactic sugar for CM).
101 :    
102 :     Added "-quiet" option to makeml and changed runtime system accordingly.
103 :    
104 :     Added cleanup handler for exportML to reset timers and compiler stats.
105 :    
106 :     ----------------------------------------------------------------------
107 : george 761 Name: Lal George
108 :     Date: 2000/12/22 22:22:58 EST 2000
109 :     Tag: Release_110_32
110 :     Description:
111 :    
112 :     Infinite precision used throughout MLRISC.
113 :     see MLRISC/mltree/machine-int.sig
114 :    
115 :     ----------------------------------------------------------------------
116 : blume 756 Name: Matthias Blume
117 : blume 760 Date: 2000/12/22 23:16:00 JST
118 :     Tag: blume-20001222-warn
119 :     Description:
120 : blume 759
121 : blume 760 Corrected wording and formatting of some CM warning message which I
122 :     broke in my previous patch.
123 : blume 759
124 :     ----------------------------------------------------------------------
125 :     Name: Matthias Blume
126 : blume 763 Date: 2000/12/22 21:20:00 JST
127 :     Tag: blume-20001222-anchorenv
128 :     Description:
129 :    
130 :     Fixed CM's handling of anchor environments in connection with CMB.make.
131 :    
132 :     ----------------------------------------------------------------------
133 :     Name: Matthias Blume
134 : blume 758 Date: 2000/12/22 13:15:00 JST
135 :     Tag: blume-20001222-cleanup
136 :     Description:
137 :    
138 :     Removed src/cm/ffi which does not (and did not) belong here.
139 :    
140 :     ----------------------------------------------------------------------
141 :     Name: Matthias Blume
142 : blume 757 Date: 2000/12/21 23:55:00 JST
143 :     Tag: blume-20001221-exn
144 :     Description:
145 :    
146 :     Probably most important: CM no longer silently swallows all exceptions
147 :     in the compiler.
148 :     Plus: some other minor CM changes. For example, CM now reports some
149 :     sizes for generated binfiles (code, data, envpickle, lambdapickle).
150 :    
151 :     ----------------------------------------------------------------------
152 :     Name: Matthias Blume
153 : blume 756 Date: 2000/12/15 00:01:05 JST
154 :     Tag: blume-20001215-dirtool
155 :     Description:
156 :    
157 :     - "dir" tool added.
158 :     - improvements and cleanup to Tools structure
159 :     - documentation updates
160 :    
161 :     ----------------------------------------------------------------------
162 : leunga 752 Name: Allen Leung
163 : leunga 755 Date: Thu Dec 14 03:45:24 EST 2000
164 :     Description:
165 :     Tag: leunga-20001214-int-inf
166 :     Description:
167 :    
168 :     In IntInf, added these standard functions, which are missing from our
169 :     implementation:
170 :    
171 :     andb : int * int -> int
172 :     xorb : int * int -> int
173 :     orb : int * int -> int
174 :     notb : int -> int
175 :     << : int * word -> int
176 :     ~>> : int * word -> int
177 :    
178 :     Not tested, I hope they are correct.
179 :    
180 :     ----------------------------------------------------------------------
181 :     Name: Allen Leung
182 : leunga 752 Date: Fri Dec 8 19:23:26 EST 2000
183 :     Description:
184 :     Tag: leunga-20001208-nowhere
185 :     Description:
186 :    
187 :     Slight improvements to the 'nowhere' tool to handle OR-patterns,
188 :     to generate better error messages etc. Plus a brief manual.
189 :    
190 :     ----------------------------------------------------------------------
191 : george 749 Name: Lal George
192 :     Date: 2000/12/08 09:54:02 EST 2000
193 :     Tag: Release_110_31
194 :     Description:
195 : george 705
196 : george 749 - Version 110.31
197 : leunga 591 ----------------------------------------------------------------------
198 : leunga 744 Name: Allen Leung
199 :     Date: Thu Dec 7 22:01:04 EST 2000
200 :     Tag: leunga-20001207-cell-monster-hack
201 :     Description:
202 :    
203 :     Major MLRISC internal changes. Affect all clients.
204 :     Summary:
205 :    
206 :     1. Type CELLS.cell = int is now replaced by a datatype.
207 :     As a result, the old regmap is now gone. Almost all interfaces
208 :     in MLRISC change as a consequence.
209 :    
210 :     2. A new brand version of machine description tool (v3.0) that generates
211 :     modules expecting the new interface. The old version is removed.
212 :    
213 :     3. The RA interface has been further abstracted into two new functors.
214 :     RISC_RA and X86RA. These functors have much simpler interfaces.
215 :     [See also directory MLRISC/demo.]
216 :    
217 :     4. Some other new source->source code generation tools are available:
218 :    
219 :     a. MLRISC/Tools/RewriteGen -- generate rewriters from rules.
220 :     b. MLRISC/Tools/WhereGen -- expands conditional pattern matching rules.
221 :     I use this tool to generate the peephole optimizers---with the new
222 :     cell type changes, peephole rules are becoming difficult to write
223 :     without conditional pattern matching.
224 :    
225 :     5. More Intmap -> IntHashTable change. Previous changes by Matthias didn't
226 :     cover the entire MLRISC source tree so many things broke.
227 :    
228 :     6. CM files have been moved to the subdirectory MLRISC/cm.
229 :     They are moved because there are a lot of them and they clutter up the
230 :     root dir.
231 :    
232 :     7. More detailed documentation to come...
233 :    
234 :     NOTE: To rebuild from 110.30 (ftp distribution), you'll have to do
235 :     a makeml -rebuild first. This is because of other other
236 :     changes that Matthias has made (see below).
237 :    
238 :    
239 :     ----------------------------------------------------------------------
240 : blume 733 Name: Matthias Blume
241 : blume 742 Date: 2000/11/30 23:12:00 JST
242 :     Tag: blume-20001130-filereorg
243 :     Description:
244 :    
245 :     Some manual updates and some file reorganizations in CM.
246 :    
247 :     ----------------------------------------------------------------------
248 :     Name: Matthias Blume
249 : blume 737 Date: 2000/11/24 17:45:00 JST
250 :     Tag: blume-20001124-link
251 :     Description:
252 :    
253 :     Drastically improved link traversal code for the case that the dynamic
254 :     value was already loaded at bootstrap time. As a result, CM and CMB
255 :     now both load blazingly fast -- even on a very slow machine. Also,
256 :     memory consumption has been further reduced by this.
257 :    
258 :     Warning: The format of the PIDMAP file has changed. THerefore, to
259 :     bootstrap you have to do this:
260 :    
261 :     1. Run CMB.make
262 :     2. Make a symbolic link for the boot directory:
263 :     ln -s sml.boot.ARCH-OS xxx
264 :     3. "Rebuild" the boot directory:
265 :     ./makeml -boot xxx -rebuild sml ; rm xxx
266 :     4. Boot normally:
267 :     ./makeml
268 :    
269 :     ----------------------------------------------------------------------
270 :     Name: Matthias Blume
271 : blume 735 Date: 2000/11/21 21:20:00 JST
272 :     Tag: blume-20001121-tools
273 :     Description:
274 :    
275 :     Continued hacking on autoloading problem -- with success this time.
276 :     Also changed tool-plugin mechanism. See new CM manual.
277 :    
278 :     ----------------------------------------------------------------------
279 :     Name: Matthias Blume
280 : blume 734 Date: 2000/11/19 14:30:00 JST
281 :     Tag: blume-20001119-autoload
282 :     Description:
283 :    
284 :     Some hacking to make autoloading faster. Success for CMB, no success
285 :     so far for CM. There is a reduced structure CM' that autoloads faster.
286 :     (This is a temporary, non-documented hack to be eliminated again when
287 :     the general problem is solved.)
288 :    
289 :     ----------------------------------------------------------------------
290 :     Name: Matthias Blume
291 : blume 733 Date: 2000/11/17 14:10:00 JST
292 :     Tag: blume-20001117-pickle-lib
293 :     Description:
294 :    
295 :     1. Eliminated comp-lib.cm
296 :     2. Made pickle-lib.cm
297 :     3. Eliminated all uses of intset.sml (from comp-lib.cm)
298 :     4. Replaced all uses of intmap.{sig,sml} (from comp-lib.cm) with
299 :     equivalent constructs from smlnj-lib.cm (INtHashTable).
300 :     5. Point 4. also goes for those uses of intmap.* in MLRISC.
301 :     Duplicated intmap modules thrown out.
302 :     6. Hunted down all duplicated SCC code and replaced it with
303 :     equivalent stuff (GraphSCCFn from smlnj-lib.cm).
304 :     7. Rewrote Feedback module.
305 :     8. Moved sortedlist.sml into viscomp-lib.cm. Eventually it
306 :     should be thrown out and equivalent modules from smlnj-lib.cm
307 :     should be used (IntRedBlackSet, IntListSet, ...).
308 :    
309 :     Confirmed that compiler compiles to fixpoint.
310 :    
311 :     ----------------------------------------------------------------------
312 : leunga 731 Name: Allen Leung
313 :     Date: 2000/11/10 18:00:00
314 :     Tag: leunga-20001110-new-x86-fp
315 :    
316 :     A new x86 floating point code generator has been added.
317 :     By default this is turned off. To turn this on, do:
318 :    
319 :     CM.autoload "$smlnj/compiler.cm";
320 :     Compiler.Control.MLRISC.getFlag "x86-fast-fp" := true;
321 :    
322 :     Changes:
323 :    
324 :     1. Changed FTAN to FPTAN so that the assembly output is correct.
325 :     2. Changed the extension callback for FTANGENT to generate:
326 :    
327 :     fptan
328 :     fstp %st(0)
329 :     instead of
330 :     fptan
331 :     fstpl ftempmem
332 :    
333 :     3. Numerous assembly fixes for x86.
334 :    
335 :     5. Cleaned up the machine code output module x86/x86MC.sml and added
336 :     support for a whole bunch of instructions and addressing modes:
337 :    
338 :     fadd/fsub/fsubr/fmul/fdiv/fdivr %st, %st(n)
339 :     faddp/fsubp/fsubrp/fmulp/fdivp/fdivrp %st, %st(n)
340 :     fadd/fsub/fsubr/fmul/fdiv/fdivr %st(n), %st
341 :     fiadd/fisub/fisubr/fimul/fidiv/fidivr mem
342 :     fxch %st(n)
343 :     fld %st(n)
344 :     fst %st(n)
345 :     fst mem
346 :     fstp %st(n)
347 :     fucom %st(n)
348 :     fucomp %st(n)
349 :    
350 :     All these are now generated when the fast fp mode is turned on.
351 :    
352 :     6. Removed the dedicated registers %st(0), ..., %st(7) from X86CpsRegs
353 :    
354 :     ----------------------------------------------------------------------
355 : blume 729 Name: Matthias Blume
356 :     Date: 2000/11/09 11:20:00 JST
357 :     Tag: blume-20001109-scc
358 :     Description:
359 :    
360 :     Eliminated some code duplication:
361 :    
362 :     1. Added "where" clause to GraphSCCFn in SML/NJ Library.
363 :     (Otherwise the functor is useless.)
364 :     2. Used GraphSCCFn where SCCUtilFun was used previously.
365 :     3. Got rid of SCCUtilFun (in comp-lib.cm).
366 :    
367 :     ----------------------------------------------------------------------
368 : george 721 Name: Lal George
369 :     Date: 2000/11/06 09:02:21 EST 2000
370 :     Tag: Release_110_30
371 :     Description:
372 :    
373 :     - Version 110.30
374 :     ----------------------------------------------------------------------
375 : blume 715 Name: Matthias Blume
376 : blume 716 Date: 2000/11/04 14:45:00
377 :     Tag: blume-20001104-mlbuild
378 :     Description:
379 :    
380 :     - Made ml-build faster on startup.
381 :     - Documentation fixes.
382 :    
383 :     ----------------------------------------------------------------------
384 :     Name: Matthias Blume
385 : blume 715 Date: 2000/11/02 17:00:00 JST
386 :     Tag: blume-20001102-condcomp
387 :     Description:
388 :    
389 :     - Small tweaks to pickler -- new BOOTFILES!
390 :     - Version bumped to 110.29.2.
391 :     - Added conditional compilation facility to init.cmi (see comment there).
392 :     ----------------------------------------------------------------------
393 : leunga 713 Name: Allen Leung
394 :     Date: 2000/10/23 19:31:00
395 :     Tag: leunga-20001023-demo-ra
396 :    
397 :     1. Minor RA changes that improves spilling on x86 (affects Moby and C-- only)
398 :     2. Test programs for the graph library updated
399 :     3. Some new MLRISC demo programs added
400 :    
401 :     ----------------------------------------------------------------------
402 : blume 710 Name: Matthias Blume
403 : blume 711 Date: 2000/08/31 22:15:00 JST
404 :     Tag: blume-20001017-errmsg
405 :     Description:
406 :    
407 :     More error message grief: Where there used to be no messages, there
408 :     now were some that had bogus error regions. Fixed.
409 :    
410 :     ----------------------------------------------------------------------
411 :     Name: Matthias Blume
412 : blume 710 Date: 2000/08/31 17:30:00 JST
413 :     Tag: blume-20001017-v110p29p1
414 :     Description:
415 :    
416 :     I made a version 110.29.1 with new bootfiles.
417 :    
418 :     Changes: Modified pickler/unpickler for faster and leaner unpickling.
419 :     CM documentation changes and a small bugfix in CM's error reporting.
420 :    
421 :     ----------------------------------------------------------------------
422 : george 705 Name: Lal George
423 :     Date: 2000/09/27 14:42:35 EDT
424 :     Tag: george-20000927-nodestatus
425 :     Description:
426 :    
427 :     Changed the type of the nodestatus, so that:
428 :    
429 :     SPILLED(~1) is now SPILLED
430 :     SPILLED(m) where m>=0 is now MEMREG(m)
431 :     SPILLED(s) where s<~1 is now SPILL_LOC(~s)
432 :    
433 :     ----------------------------------------------------------------------
434 : blume 697 Name: Matthias Blume
435 : blume 703 Date: 2000/09/07 14:45:00 JST
436 :     Tag: blume-20000907-cmerrmsg
437 :     Description:
438 :    
439 :     Small tweak to CM to avoid getting ML syntax error messages twice.
440 :    
441 :     ----------------------------------------------------------------------
442 :     Name: Matthias Blume
443 : blume 701 Date: 2000/08/31 18:00:00 JST
444 :     Tag: blume-20000831-cvsbootfiles
445 :     Description:
446 :    
447 :     New URL for boot files (because the 110.29 files on the BL server do
448 :     now work correctly with my updated install scripts for yacc and lex).
449 :    
450 :     ----------------------------------------------------------------------
451 :     Name: Matthias Blume
452 : blume 697 Date: 2000/08/08 12:33:00 JST
453 :     Tag: blume-20000808-manual
454 :     Description:
455 :    
456 :     Tiny update to CM manual.
457 :    
458 :     ----------------------------------------------------------------------
459 : leunga 695 Name: Allen Leung
460 :     Date: 2000/08/7 19:31:00
461 :     Tag: leunga-20000807-a-whole-bunch-of-stuff
462 :    
463 :     Moby, C--, SSA, x86, machine descriptions etc. Should only affect C--
464 :     and Mobdy.
465 :    
466 :     1. x86
467 :    
468 :     a. Fixes to peephole module by John and Dan.
469 :     b. Assembly fix to SETcc by Allen.
470 :     c. Fix to c-call by John.
471 :     d. Fix to spilling by John. (This one deals with the missing FSTPT case)
472 :     e. Instruction selection optimization to SETcc as suggested by John.
473 :    
474 :     For example,
475 :    
476 :     MV(32, x, COND(32, CMP(32, LT, a, b), LI 1, LI 0))
477 :    
478 :     should generate:
479 :    
480 :     MOVL a, x
481 :     SUBL b, x
482 :     SHRL 31, x
483 :    
484 :     2. IR stuff
485 :    
486 :     A bunch of new DJ-graph related algorithms added. These
487 :     speed up SSA construction.
488 :    
489 :     3. SSA + Scheduling
490 :    
491 :     Added code for SSA and scheduling to the repository
492 :    
493 :     ----------------------------------------------------------------------
494 : george 694 Name: Lal George
495 :     Date: 2000/07/27 11:53:14 EDT
496 :    
497 :     Tag: lal-20000727-linux-ppc
498 :     Description:
499 :    
500 :     Made changes to support Linux PPC.
501 :     p.s. I have confirmation that the 110.29 boot files work fine.
502 :    
503 :     ----------------------------------------------------------------------
504 : blume 684 Name: Matthias Blume
505 : blume 692 Date: 2000/07/27 17:40:00 JST
506 :     Tag: blume-20000727-scripts
507 :     Description:
508 :    
509 :     !!!! WARNING !!!!
510 :     You must recompile the runtime system!
511 :     !!!! WARNING !!!!
512 :    
513 :     This is basically another round of script-enhancements:
514 :    
515 :     1. sml, ml-build, and ml-makedepend accept options -D and -U to define
516 :     and undefine CM preprocessor symbols.
517 :    
518 :     2. ml-build avoids generating a new heap image if it finds that the
519 :     existing one is still ok. (The condition is that no ML file had to
520 :     be recompiled and all ML files are found to be older that the heap
521 :     file.)
522 :    
523 :     To make this work smoothly, I also hacked the runtime system as
524 :     well as SMLofNJ.SysInfo to get access to the heap image suffix
525 :     (.sparc-solaris, ...) that is currently being used.
526 :    
527 :     Moreover, the signature of CM.mk_standalone has changed. See the
528 :     CM manual.
529 :    
530 :     3. ml-makedepend accepts additional options -n, -a, and -o. (See the
531 :     CM manual for details.)
532 :    
533 :     4. More CM manual updates:
534 :     - all of the above has been documented.
535 :     - there is now a section describing the (CM-related) command line
536 :     arguments that are accepted by the "sml" command
537 :    
538 :     ----------------------------------------------------------------------
539 :     Name: Matthias Blume
540 : blume 691 Date: 2000/07/25 16:20:00 JST
541 :     Tag: blume-20000725-makedepend
542 :     Description:
543 :    
544 :     Added a script called ml-makedepend. This can be used in makefiles
545 :     for Unix' make in a way very similar to the "makedepend" command for
546 :     C.
547 :    
548 :     The script internally uses function CM.sources.
549 :    
550 :     Synopsis:
551 :    
552 :     ml-makedepend [-f makefile] cmfile targetname
553 :    
554 :     The default for the makefile is "makefile" (or "Makefile" should
555 :     "makefile" not exist).
556 :    
557 :     ml-makedepend adds a cmfile/targetname-specific section to this
558 :     makefile (after removing the previous version of this section). The
559 :     section contains a single dependency specification with targetname on
560 :     the LHS (targetname is an arbitrary name), and a list of files derived
561 :     from the cmfile on the RHS. Some of the files on the RHS are
562 :     ARCH/OPSYS-specific. Therefore, ml-makedepend inserts references to
563 :     "make" variables $(ARCH) and $(OPSYS) in place of the corresponding
564 :     path names. The makefile writer is responsible for making sure that
565 :     these variables have correct at the time "make" is invoked.
566 :    
567 :     ----------------------------------------------------------------------
568 :     Name: Matthias Blume
569 : blume 690 Date: 2000/07/22 23:30:00 JST
570 :     Tag: blume-20000722-urlupdate
571 :     Description:
572 :    
573 :     Changed BOOT and config/srcarchiveurl to point to BL server:
574 :    
575 :     ftp://ftp.research.bell-labs.com/dist/smlnj/working/110.29/
576 :    
577 :     ----------------------------------------------------------------------
578 :     Name: Matthias Blume
579 : blume 686 Date: 2000/07/18 18:00:00 JST
580 :     Tag: blume-20000718-Version_110_29
581 :     Description:
582 :    
583 :     1. Updated src/compiler/TopLevel/main/version.sml to version 110.29
584 :    
585 :     2. Updated config/version to 110.29
586 :    
587 :     3. Updated config/srcarchiveurl
588 :    
589 :     3. New boot files!
590 :     ftp://ftp.cs.princeton.edu/pub/people/blume/sml/110.29-autofetch
591 :    
592 :     ----------------------------------------------------------------------
593 :     Name: Matthias Blume
594 : blume 684 Date: 2000/07/11 13:58:00 JST
595 :     Tag: blume-20000711-doctypo
596 :     Description:
597 :    
598 :     Fixed a few typos in CM manual.
599 :    
600 :     ----------------------------------------------------------------------
601 : leunga 683 Name: Allen Leung
602 :     Date: 2000/06/15 00:38:00
603 :     Tag: leunga-20000704-sparc-x86
604 :    
605 :     1. x86 peephole improvement sp += k; sp -= k => nop [from John]
606 :     2. fix to x86 RET bug [found by Dan Grossman]
607 :     3. sparc assembly bug fix for ticc instructions [found by Fermin]
608 :    
609 :     Affects c-- and moby only
610 :    
611 :     ----------------------------------------------------------------------
612 : blume 680 Name: Matthias Blume
613 : blume 682 Date: 2000/07/04 15:26:00
614 :     Tag: blume-20000704-trigger
615 :     Description:
616 :    
617 :     1. Improvements to CM manual.
618 :     2. SMLofNJ.Internals.BTrace.trigger reinstated as an alternative way
619 :     of getting a back-trace. The function, when called, raises an
620 :     internal exception which explicitly carries the full back-trace history,
621 :     so it is unaffected by any intervening handle-raise pairs ("trivial"
622 :     or not). The interactive loop will print that history once it arrives
623 :     at top level.
624 :     Short of having all exceptions implicitly carry the full history, the
625 :     recommended way of using this facility is:
626 :     - compile your program with instrumentation "on"
627 :     - run it, when it raises an exception, look at the history
628 :     - if the history is "cut off" because of some handler, go and modify
629 :     your program so that it explicitly calls BTrace.trigger
630 :     - recompile (still instrumented), and rerun; look at the full history
631 :    
632 :     ----------------------------------------------------------------------
633 :     Name: Matthias Blume
634 :     Date: 2000/07/03 15:36:00 JST
635 : blume 680 Tag: blume-20000702-manual
636 :     Description:
637 :    
638 :     Small corrections and updates to CM manual.
639 :    
640 :     ----------------------------------------------------------------------
641 : blume 682 Name: Matthias Blume
642 : blume 679 Date: 2000/06/29 16:04:00 JST
643 :     Tag: blume-20000629-yacctool
644 :     Description:
645 :    
646 :     Changes:
647 :    
648 :     1. Class "mlyacc" now takes separate arguments to pass options to
649 :     generated .sml- and .sig-files independently.
650 :     2. Corresponding CM manual updates.
651 :     3. BTrace module now also reports call sites. (However, for loop clusters
652 :     it only shows from where the cluster was entered.) There are associated
653 :     modifications to core.sml, internals.{sig,sml}, btrace.sml, and btimp.sml.
654 :    
655 :     ----------------------------------------------------------------------
656 : blume 675 Name: Matthias Blume
657 : blume 678 Date: 2000/06/27 16:51:00 JST
658 :     Tag: blume-20000627-noweb
659 :     Description:
660 :    
661 :     Changes:
662 :    
663 :     1. Implemented "subdir" and "witness" options for noweb tool.
664 :     This caused some slight internal changes in CM's tool implementation.
665 :     2. Fixed bug in "tool plugin" mechanism. This is essentially cleaning
666 :     some remaining issues from earlier path anchor changes.
667 :     3. Updated CM manual accordingly.
668 :    
669 :     4. Changed implementation of back-tracing so that I now consider it
670 :     ready for prime-time.
671 :    
672 :     In particular, you don't have to explicitly trigger the back-trace
673 :     anymore. Instead, if you are running BTrace-instrumented code and
674 :     there is an uncaught exception (regardless of whether or not it was
675 :     raised in instrumented code), the top-level evalloop will print
676 :     the back-trace.
677 :    
678 :     Features:
679 :    
680 :     - Instrumented and uninstrumented code work together seemlessly.
681 :     (Of course, uninstrumented code is never mentioned in actual
682 :     back-traces.)
683 :    
684 :     - Asymptotic time- and space-complexity of instrumented code is
685 :     equal to that of uninstrumented code. (This means that
686 :     tail-recursion is preserved by the instrumentation phase.)
687 :    
688 :     - Modules whose code has been instrumented in different sessions
689 :     work together without problem.
690 :    
691 :     - There is no penalty whatsoever on uninstrumented code.
692 :    
693 :     - There is no penalty on "raise" expressions, even in
694 :     instrumented code.
695 :    
696 :     A potential bug (or perhaps it is a feature, too):
697 :    
698 :     A back-trace reaches no further than the outermost instrumented
699 :     non-trivial "raise". Here, a "trivial" raise is one that is the
700 :     sole RHS of a "handle" rule. Thus, back-traces reach trough
701 :    
702 :     <exp> handle e => raise e
703 :    
704 :     and even
705 :    
706 :     <exp> handle Foo => raise Bar
707 :    
708 :     and, of course, through
709 :    
710 :     <exp> handle Foo => ...
711 :    
712 :     if the exception was not Foo.
713 :    
714 :     Back-traces always reach right through any un-instrumented code
715 :     including any of its "handle" expressions, trivial or not.
716 :    
717 :     To try this out, do the following:
718 :    
719 :     - Erase all existing binfiles for your program.
720 :     (You may keep binfiles for those modules where you think you
721 :     definitely don't need back-tracing.)
722 :     - Turn on back-trace instrumentation:
723 :     SMLofNJ.Internals.BTrace.mode (SOME true);
724 :     - Recompile your program. (I.e., run "CM.make" or "use".)
725 :     - You may now turn instrumentation off again (if you want):
726 :     SMLofNJ.Internals.BTrace.mode (SOME false);
727 :     - Run your program as usual. If it raises an exception that
728 :     reaches the interactive toplevel, then a back-trace will
729 :     automatically be printed. After that, the toplevel loop
730 :     will print the exception history as usual.
731 :    
732 :     ----------------------------------------------------------------------
733 :     Name: Matthias Blume
734 : blume 677 Date: 2000/06/26 09:56:46 JST
735 :     Tag: blume-20000626-setup
736 :     Description:
737 :    
738 :     CM: - setup-parameter to "sml" added; this can be used to run arbitrary
739 :     ML code before and after compiling a file (e.g., to set compiler
740 :     flags)
741 :    
742 :     Compiler: - improved btrace API (in core.sml, internals.{sig,sml})
743 :     - associated changes to btrace.sml (BTrace instrumentation pass)
744 :     - cleaner implementation of btimp.sml (BTrace tracing and report
745 :     module)
746 :    
747 :     CM manual: * new path encoding documented
748 :     * description of setup-parameter to "sml" added
749 :    
750 :     The biggest user-visible change to back-tracing is that it is no
751 :     longer necessary to compile all traced modules within the same
752 :     session. (This was a real limitation.)
753 :    
754 :     ----------------------------------------------------------------------
755 :     Name: Matthias Blume
756 : blume 676 Date: 2000/06/24 12:40:00 JST
757 :     Tag: blume-20000624-startup
758 :     Description:
759 :    
760 :     Fixes startup slowdown problem. (I was calling SrcPath.sync a _tad_
761 :     bit too often -- to put it mildly. :)
762 :    
763 :     ----------------------------------------------------------------------
764 :     Name: Matthias Blume
765 : blume 675 Date: 2000/06/23 18:20:00 JST
766 :     Tag: blume-20000623-btrace
767 :     Description:
768 :    
769 :     This updates adds a backtrace facility to aid programmers in debugging
770 :     their programs. This involves the following changes:
771 :    
772 :     1. Module system/smlnj/init/core.sml (structure _Core) now has hooks for
773 :     keeping track of the current call stack. When programs are compiled
774 :     in a special mode, the compiler will insert calls to these hooks
775 :     into the user program.
776 :     "Hook" means that it is possible for different implementations of
777 :     back-tracing to register themselves (at different times).
778 :    
779 :     2. compiler/MiscUtil/profile/btrace.sml implements the annotation phase
780 :     as an Absyn.dec->Absyn.dec rewrite. Normally this phase is turned off.
781 :     It can be turned on using this call:
782 :     SMLofNJ.Internals.BTrace.mode (SOME true);
783 :     Turning it off again:
784 :     SMLofNJ.Internals.BTrace.mode (SOME false);
785 :     Querying the current status:
786 :     SMLofNJ.Internals.BTrace.mode NONE;
787 :     Annotated programs are about twice as big as normal ones, and they
788 :     run a factor of 2 to 4 slower with a dummy back-trace plugin (one
789 :     where all hooks do nothing). The slowdown with a plugin that is
790 :     actually useful (such as the one supplied by default) is even greater,
791 :     but in the case of the default plugin it is still only an constant
792 :     factor (amortized).
793 :    
794 :     3. system/Basis/Implementation/NJ/internals.{sig,sml} have been augmented
795 :     with a sub-structure BTrace for controlling back-tracing. In particular,
796 :     the above-mentioned function "mode" controls whether the annotation
797 :     phase is invoked by the compiler. Another important function is
798 :     "trigger": when called it aborts the current execution and causes
799 :     the top-level loop to print a full back-trace.
800 :    
801 :     4. compiler/MiscUtil/profile/btimp.sml is the current default plugin
802 :     for back-tracing. It keeps track of the dynamic call stack and in
803 :     addition to that it keeps a partial history at each "level" of that
804 :     stack. For example, if a tail-calls b, b tail-calls c, and c tail-calls
805 :     d and b (at separate times, dynamically), then the report will show:
806 :    
807 :     GOTO d
808 :     /c
809 :     GOTO \b
810 :     CALL a
811 :    
812 :     This shows that there was an initial non-tail call of a, then a
813 :     tail-call to b or c, looping behavior in a cluster of functions that
814 :     consist of b and c, and then a goto from that cluster (i.e., either from
815 :     b or from c) to d.
816 :    
817 :     Note that (depending on the user program) the amount of information
818 :     that the back-trace module has to keep track of at each level is bounded
819 :     by a constant. Thus, the whole implementation has the same asymptotical
820 :     complexity as the original program (both in space and in time).
821 :    
822 :     5. compiler/TopLevel/interact/evalloop.sml has been modified to
823 :     handle the special exception SMLofNJ.Internals.BTrace.BTrace
824 :     which is raised by the "trigger" function mentioned above.
825 :    
826 :     Notes on usage:
827 :    
828 :     - Annotated code works well together with unannotated code:
829 :     Unannotated calls simply do not show up at all in the backtrace.
830 :    
831 :     - It is not a good idea to let modules that were annotated during
832 :     different sessions run at the same time. This is because the compiler
833 :     chooses small integers to identify individual functions, and there
834 :     will be clashes if different modules were compiled in separate sessions.
835 :     (Nothing will crash, and you will even be told about the clashes, but
836 :     back-trace information will in general not be useful.)
837 :    
838 :     - Back-tracing can be confused by callcc and capture.
839 :    
840 :     - The only way of getting a back-trace right now is to explicitly
841 :     invoke the "trigger" function from your user program. Eventually, we
842 :     should make every exception carry back-trace information (if
843 :     available). But since this creates more overhead at "raise"-time
844 :     (similar to the current exnHistory overhead), I have not yet
845 :     implemented this. (The implementation will be rather easy.) With
846 :     exceptions carrying back-trace information, this facility will be even
847 :     more useful because users don't need to modify their programs...
848 :    
849 :     - While it is possible to compile the compiler with back-trace
850 :     annotations turned on (I did it to get some confidence in
851 :     correctness), you must make absolutely sure that core.sml and
852 :     btimp.sml are compiled WITHOUT annotation! (core.sml cannot actually
853 :     be compiled with annotation because there is no core access yet, but
854 :     if you compile btimp.sml with annotation, then the system will go into
855 :     an infinite recursion and crash.)
856 :     Since CM currently does not know about BTrace, the only way to turn
857 :     annotations on and off for different modules of the compiler is to
858 :     interrupt CMB.make, change the settings, and re-invoke it. Of course,
859 :     this is awkward and clumsy.
860 :    
861 :     Sample sessions:
862 :    
863 :     Standard ML of New Jersey v110.28.1 [FLINT v1.5], June 5, 2000
864 :     - SMLofNJ.Internals.BTrace.mode (SOME true);
865 :     [autoloading]
866 :     [autoloading done]
867 :     val it = false : bool
868 :     - structure X = struct
869 :     - fun main n = let
870 :     - fun a (x, 0) = d x
871 :     - | a (x, n) = b (x, n - 1)
872 :     - and b (x, n) = c (x, n)
873 :     - and c (x, n) = a (x, n)
874 :     - and d x = e (x, 3)
875 :     - and e (x, 0) = f x
876 :     - | e (x, n) = e (x, n - 1)
877 :     - and f 0 = SMLofNJ.Internals.BTrace.trigger ()
878 :     - | f n = n * g (n - 1)
879 :     - and g n = a (n, 3)
880 :     - in
881 :     - f n
882 :     - end
883 :     - end;
884 :     structure X : sig val main : int -> int end
885 :     - X.main 3;
886 :     *** BACK-TRACE ***
887 :     GOTO stdIn:4.2-13.20: X.main[2].f
888 :     GOTO-( stdIn:4.2-13.20: X.main[2].e
889 :     GOTO stdIn:4.2-13.20: X.main[2].d
890 :     / stdIn:4.2-13.20: X.main[2].a
891 :     | stdIn:4.2-13.20: X.main[2].b
892 :     GOTO-\ stdIn:4.2-13.20: X.main[2].c
893 :     CALL stdIn:4.2-13.20: X.main[2].g
894 :     GOTO stdIn:4.2-13.20: X.main[2].f
895 :     GOTO-( stdIn:4.2-13.20: X.main[2].e
896 :     GOTO stdIn:4.2-13.20: X.main[2].d
897 :     / stdIn:4.2-13.20: X.main[2].a
898 :     | stdIn:4.2-13.20: X.main[2].b
899 :     GOTO-\ stdIn:4.2-13.20: X.main[2].c
900 :     CALL stdIn:4.2-13.20: X.main[2].g
901 :     GOTO stdIn:4.2-13.20: X.main[2].f
902 :     GOTO-( stdIn:4.2-13.20: X.main[2].e
903 :     GOTO stdIn:4.2-13.20: X.main[2].d
904 :     / stdIn:4.2-13.20: X.main[2].a
905 :     | stdIn:4.2-13.20: X.main[2].b
906 :     GOTO-\ stdIn:4.2-13.20: X.main[2].c
907 :     CALL stdIn:4.2-13.20: X.main[2].g
908 :     GOTO stdIn:4.2-13.20: X.main[2].f
909 :     CALL stdIn:2.15-17.4: X.main[2]
910 :     -
911 :    
912 :     (Note that because of a FLINt bug the above code currently does not
913 :     compile without BTrace turned on.)
914 :    
915 :     Here is another example, using my modified Tiger compiler:
916 :    
917 :     Standard ML of New Jersey v110.28.1 [FLINT v1.5], June 5, 2000
918 :     - SMLofNJ.Internals.BTrace.mode (SOME true);
919 :     [autoloading]
920 :     [autoloading done]
921 :     val it = false : bool
922 :     - CM.make "sources.cm";
923 :     [autoloading]
924 :     ...
925 :     [autoloading done]
926 :     [scanning sources.cm]
927 :     [parsing (sources.cm):parse.sml]
928 :     [creating directory CM/SKEL ...]
929 :     [parsing (sources.cm):tiger.lex.sml]
930 :     ...
931 :     [wrote CM/sparc-unix/semant.sml]
932 :     [compiling (sources.cm):main.sml]
933 :     [wrote CM/sparc-unix/main.sml]
934 :     [New bindings added.]
935 :     val it = true : bool
936 :     - Main.compile ("../testcases/merge.tig", "foo.out");
937 :     *** BACK-TRACE ***
938 :     CALL lib/semant.sml:99.2-396.21: SemantFun[2].transExp.trvar
939 :     CALL lib/semant.sml:99.2-396.21: SemantFun[2].transExp.trexp
940 :     CALL lib/semant.sml:289.3-295.22: SemantFun[2].transExp.trexp.check[2]
941 :     GOTO lib/semant.sml:289.3-295.22: SemantFun[2].transExp.trexp.check[2]
942 :     CALL lib/semant.sml:99.2-396.21: SemantFun[2].transExp.trexp
943 :     CALL lib/semant.sml:99.2-396.21: SemantFun[2].transExp.trexp
944 :     CALL lib/semant.sml:488.3-505.6: SemantFun[2].transDec.trdec[2].transBody[2]
945 :     / lib/semant.sml:411.65-543.8: SemantFun[2].transDec
946 :     CALL-\ lib/semant.sml:413.2-540.9: SemantFun[2].transDec.trdec[2]
947 :     CALL lib/semant.sml:99.2-396.21: SemantFun[2].transExp.trexp
948 :     CALL lib/semant.sml:8.52-558.4: SemantFun[2].transProg[2]
949 :     CALL main.sml:1.18-118.4: Main.compile[2]
950 :     -
951 :    
952 :     ----------------------------------------------------------------------
953 : blume 673 Name: Matthias Blumen
954 :     Date: 2000/06/21 18:00:00 JST
955 :     Tag: blume-20000621-manual
956 :     Description:
957 :    
958 :     CM manual update: Path environments documented.
959 :    
960 :     ----------------------------------------------------------------------
961 : blume 666 Name: Matthias Blume
962 : blume 672 Date: 2000/06/19 13:40:00
963 :     Tag: blume-20000619-manual
964 :     Description:
965 :    
966 :     CM manual and system/README update. This only covers the fact that
967 :     there are no more implicit anchors. (Path environments and the "bind"
968 :     option to "cm" have yet to be documented.)
969 :    
970 :     ----------------------------------------------------------------------
971 :     Name: Matthias Blume
972 : blume 671 Date: 2000/06/19 11:05:00 JST
973 :     Tag: blume-20000619-chdir-bugfix
974 :     Description:
975 :    
976 :     Fixed a bug in new SrcPath module that sometimes led to a bad chDir call.
977 :    
978 :     ----------------------------------------------------------------------
979 :     Name: Matthias Blume
980 : blume 670 Date: 2000/06/18 22:00:10 JST
981 :     Tag: blume-20000618-implicit-anchors-really-gone
982 :     Description:
983 :    
984 :     I updates the previous HISTORY entry where I forgot to mention that
985 :     implicit anchors are no longer with us.
986 :    
987 :     The current update also gets rid of the (now useless) controller
988 :     CM.Control.implicit_anchors.
989 :    
990 :     ----------------------------------------------------------------------
991 :     Name: Matthias Blume
992 : blume 666 Date: 2000/06/16 17:30:00 JST
993 :     Tag: blume-20000616-anchorenv
994 :     Description:
995 :    
996 :     This patch implements the long anticipated (just kidding :) "anchor
997 :     environment" mechanism. In the course of doing this, I also
998 :     re-implemented CM's internal "SrcPath" module from scratch. The new
999 :     one should be more robust in certain boundary cases. In any case, it
1000 :     is a lot cleaner than its predecessor (IMHO).
1001 :    
1002 :     This time, although there is yet another boot file format change, I
1003 :     kept the unpickler backward-compatible. As a result, no new bootfiles
1004 :     are necessary and bootstrapping is straightforward. (You cannot read
1005 :     new bootfiles into an old system, but the other way around is no
1006 :     problem.)
1007 :    
1008 :     Visible changes:
1009 :    
1010 : blume 670 ** 0. Implicit path anchors (without the leading $-symbol) are no
1011 :     longer recognized at all. This means that such path names are not
1012 :     illegal either. For example, the name basis.cm simply refers to a
1013 :     local file called "basis.cm" (i.e, the name is an ordinary path
1014 :     relative to .cm-files directory). Or, to put it differently, only
1015 :     names that start with $ are anchored paths.
1016 : blume 666
1017 : blume 670 ** 1. The $<singlearc> abbreviation for $/<singlearc> has finally
1018 :     vanished.
1019 :    
1020 :     John (Reppy) had critizised this as soon as I originally proposed and
1021 : blume 666 implemented it, but at that time I did not really deeply believe
1022 :     him. :) Now I came full-circle because I need the $<singlearc> syntax
1023 :     in another place where it cannot be seen as an abbreviation for
1024 :     $/<singlearc>. To avoid the confusion, $<singlearc> now means what it
1025 :     seems to mean (i.e., it "expands" into the corresponding anchor
1026 :     value).
1027 :    
1028 :     However, when paths are used as members in CM description files, it
1029 :     continues to be true that there must be at least another arc after the
1030 :     anchor. This is now enforced separately during semantic analysis
1031 : blume 670 (i.e., from a lexical/syntactical point of view, the notation is ok.)
1032 : blume 666
1033 :     ** 2. The "cm" class now accepts an option "bind". The option's value
1034 :     is a sub-option list of precisely two items -- one labeled "anchor"
1035 :     and the other one labeled "value". As you might expect, "anchor" is
1036 :     used to specify an anchor name to be bound, and "value" specifies what
1037 :     the anchor is being bound to.
1038 :    
1039 :     The value must be a directory name and can be given in either standard
1040 :     syntax (including the possibility that it is itself an anchored path)
1041 :     or native syntax.
1042 :    
1043 :     Examples:
1044 :    
1045 :     foo.cm (bind:(anchor:bar value:$mystuff/bar))
1046 :     lib.cm (bind:(anchor:a value:"H:\\x\\y\\z")) (* only works under windows *)
1047 :    
1048 :     and so on.
1049 :    
1050 :     The meaning of this is that the .cm-file will be processed with an
1051 :     augmented anchor environment where the given anchor(s) is/are bound to
1052 :     the given values(s).
1053 :    
1054 :     The rationale for having this feature is this: Suppose you are trying
1055 :     to use two different (already stable) libraries a.cm and b.cm (that
1056 :     you perhaps didn't write yourself). Further, suppose each of these
1057 :     two libraries internally uses its own auxiliary library $aux/lib.cm.
1058 :     Normally you would now have a problem because the anchor "lib" can not
1059 :     be bound to more than one value globally. Therefore, the project that
1060 :     uses both a.cm and b.cm must locally redirect the anchor to some other
1061 :     place:
1062 :    
1063 :     a.cm (bind:(anchor:lib value:/usr/lib/smlnj/a-stuff))
1064 :     b.cm (bind:(anchor:lib value:/usr/lib/smlnj/b-stuff))
1065 :    
1066 :     This hard-wires $lib/aux.cm to /usr/lib/smlnj/a-stuff/aux.cm or
1067 :     /usr/lib/smlnj/b-stuff/aux.cm, respectively.
1068 :    
1069 :     Hard-wiring path names is a bit inflexible (and CM will verbosely warn
1070 :     you when you do so at the time of CM.stabilize). Therefore, you can
1071 :     also use an anchored path as the value:
1072 :    
1073 :     a.cm (bind:(anchor:lib value:$a-lib))
1074 :     b.cm (bind:(anchor:lib value:$b-lib))
1075 :    
1076 :     Now you can globally configure (using the usual CM.Anchor.anchor or
1077 :     pathconfig machinery) bindings for "a-lib" and "b-lib". Since "lib"
1078 :     itself is always locally bound, setting it globally is no longer
1079 :     meaningful or necessary (but it does not hurt either). In fact, "lib"
1080 :     can still be used as a global anchor for separate purposes. As a
1081 :     matter of fact, one can locally define "lib" in terms of a global
1082 :     "lib":
1083 :    
1084 :     a.cm (bind:(anchor:lib value:$lib/a))
1085 :     b.cm (bind:(anchor:lib value:$lib/b))
1086 :    
1087 :     ** 3: The encoding of path names has changed. This affects the way
1088 :     path names are shown in CM's progress report and also the internal
1089 :     protocol encoding used for parallel make.
1090 :    
1091 :     The encoding now uses one or more ':'-separated segments. Each
1092 :     segments corresponds to a file that has been specified relative to the
1093 :     file given by its preceding segment. The first segment is either
1094 :     relative to the CWD, absolute, or anchored. Each segment itself is
1095 :     basically a Unix pathname; all segments but the first are relative.
1096 :    
1097 :     Example:
1098 :    
1099 :     $foo/bar/baz.cm:a/b/c.sml
1100 :    
1101 :     This path denotes the file bar/a/b/c.sml relative to the directory
1102 :     denoted by anchor "foo". Notice that the encoding also includes
1103 :     baz.cm which is the .cm-file that listed a/b/c.sml. As usual, such
1104 :     paths are resolved relative to the .cm-files directory, so baz.cm must
1105 :     be ignored to get the "real" pathname.
1106 :    
1107 :     To make this fact more obvious, CM puts the names of such "virtual
1108 :     arcs" into parentheses when they appear in progress reports. (No
1109 :     parentheses will appear in the internal protocol encoding.) Thus,
1110 :     what you really see is:
1111 :    
1112 :     $foo/bar/(baz.cm):a/b/c.sml
1113 :    
1114 :     I find this notation to be much more informative than before.
1115 :    
1116 :     Another new feature of the encoding is that special characters
1117 :     including parentheses, colons, (back)slashes, and white space are
1118 :     written as \ddd (where ddd is the decimal encoding of the character).
1119 :    
1120 : blume 670 *** The CM manual still needs to be updated.
1121 :    
1122 : blume 666 ----------------------------------------------------------------------
1123 : leunga 660 Name: Allen Leung
1124 :     Date: 2000/06/15 00:38:00
1125 :     Tag: leunga-20000615-x86-peephole
1126 :    
1127 :     x86 Peephole fix by Fermin. Affects c-- and moby only.
1128 :    
1129 :     ----------------------------------------------------------------------
1130 : blume 658 Name: Matthias Blume
1131 :     Date: 2000/06/12 11:40:00
1132 :     Tag: blume-20000612-parmakefix
1133 :     Description:
1134 :    
1135 :     More cleanup after changing the file naming scheme: This time I
1136 :     repaired the parallel make mechanism for CMB.make which I broke earlier.
1137 :    
1138 :     ----------------------------------------------------------------------
1139 : leunga 657 Name: Allen Leung
1140 :     Date: 2000/06/09 01:25:00
1141 :     Tag: leunga-20000609-various
1142 :    
1143 :     None of these things should affect normal SML/NJ operations
1144 :    
1145 :     1. Peephole improvements provided by Fermin (c--)
1146 :     2. New annotation DEFUSE for adding extra dependence (moby)
1147 :     3. New X86 LOCK instructions (moby)
1148 :     4. New machine description language for reservation tables (scheduling)
1149 :     5. Fixes to various optimization/analysis modules (branch chaining, dominator
1150 :     trees etc.)
1151 :     6. I've changed the CM files so that they can work with versions
1152 :     110.0.6, 110.25 and 110.28
1153 :    
1154 :     ----------------------------------------------------------------------
1155 : blume 652 Name: Matthias Blume
1156 : blume 656 Date: 2000/06/09 12:40:00
1157 :     Tag: blume-20000609-log
1158 :     Description:
1159 :    
1160 :     - Removed all(?) remaining RCS Log entries from sources.
1161 :    
1162 :     - Fixed bug in ml-yacc and ml-lex sources (use explicit anchors for
1163 :     anchored paths).
1164 :    
1165 :     ----------------------------------------------------------------------
1166 :     Name: Matthias Blume
1167 : blume 653 Date: 2000/06/07 17:00:00 JST
1168 :     Tag: blume-20000607-no-implicit-anchors
1169 :     Description:
1170 :    
1171 :     1. This update changes the default setting for
1172 :     CM.Control.implicit_anchors from true to false. This means that
1173 :     implicit anchors are no longer permitted by default. I also tried to
1174 :     make sure that nothing else still relies on implicit anchors.
1175 :     (This is the next step on the schedule towards a CM that does not even
1176 :     have the notion of implicit anchors anymore.)
1177 :    
1178 :     2. More CM manual updates.
1179 :    
1180 :     3. I managed to track down and fix the pickling bug I mentioned last
1181 :     time. Because of the previously existing workaround, this entails no
1182 :     immediate practical changes.
1183 :    
1184 :     ----------------------------------------------------------------------
1185 :     Name: Matthias Blume
1186 : blume 652 Date: 2000/06/06 11:15:00 JST
1187 :     Tag: blume-20000606-lazierpickle
1188 :     Description:
1189 :    
1190 :     !!!! NEW BOOT FILES !!!!
1191 :    
1192 :     * The main purpose of this update is to make library pickles lazier in
1193 :     order to reduce the initial space penalty for autoloading a library.
1194 :     As a result, it is now possible to have $smlnj/compiler.cm
1195 :     pre-registered. This should take care of the many complaints or
1196 :     inquiries about missing structure Compiler. This required changes to
1197 :     CM's internal data structures and small tweaks to some algorithms.
1198 :    
1199 :     As a neat additional effect, it is no longer necessary (for the sake
1200 :     of lean heap image files) to distinguish between a "minimal" CM and a
1201 :     "full" CM. Now, there is only one CM (i.e., the "full" version:
1202 :     $smlnj/cm.cm aka $smlnj/cm/full.cm), and it is always available at the
1203 :     interactive top level. ($smlnj/cm/minimal.cm is gone.)
1204 :    
1205 :     To make the life of compiler-hackers easier, "makeml" now also
1206 :     pre-registers $smlnj/cmb.cm (aka $smlnj/cmb/current.cm). In other
1207 :     words, after you bootstrap a new sml for the first time, you will not
1208 :     have to autoload $smlnj/cmb.cm again afterwards. (The first time
1209 :     around you will still have to do it, though.)
1210 :    
1211 :     * A second change consists of major updates to the CM manual. There
1212 :     are now several appendices with summary information and also a full
1213 :     specification of the CM description file syntax.
1214 :    
1215 :     * In directory src/system I added the script "allcross". This script
1216 :     invokes sml and cross-compiles the compiler for all supported
1217 :     architectures. (Useful when providing a new set of boot files.)
1218 :    
1219 :     * There seems to be a latent bug in my "lazy pickles" mechanism. I
1220 :     added a small tweak to pickle-util.sml to work around this problem,
1221 :     but it is not a proper fix yet. I will investigate further. (The
1222 :     effect of the bug was an inflation of library pickle size.)
1223 :    
1224 :     * Version number increased to 110.28.1 (to avoid compatibility problems).
1225 :    
1226 :     ----------------------------------------------------------------------
1227 : leunga 646 Name: Allen Leung
1228 : leunga 648 Date: 2000/05/25 17:28 EDT
1229 :     Tag: leunga-20000525-ra
1230 :     Description:
1231 :    
1232 :     Fixed a bug in freezing phase of the register allocator.
1233 :    
1234 :     ----------------------------------------------------------------------
1235 :     Name: Allen Leung
1236 : leunga 647 Date: 2000/05/15 22:53 EDT
1237 : leunga 646 Tag: leunga-20000515-alpha-x86-ra
1238 :     Description:
1239 :    
1240 :     1. Alpha
1241 :    
1242 :     Slight cleanup. Removed the instruction SGNXL
1243 :    
1244 :     2. X86
1245 :    
1246 :     Added the following instructions to the instruction set:
1247 :    
1248 :     ROLx, RORx,
1249 :     BTx, BTSx, BTLx, BTRx,
1250 :     XCHGx, and variants with the LOCK prefix
1251 :    
1252 :     3. Register Allocation
1253 :    
1254 :     The module ra-rewrite-with-renaming has been improved.
1255 :    
1256 : leunga 647 These have no effect on SML/NJ.
1257 :    
1258 : leunga 646 ----------------------------------------------------------------------
1259 : blume 642 Name: Matthias Blume
1260 : blume 645 Date: 2000/05/15 16:20:00 JST
1261 :     Tag: blume-20000515-lightrebuild
1262 :     Description:
1263 :    
1264 :     1. I added an alternative to "-rebuild" to "makeml". The difference is
1265 :     that prior to calling CMB.make' the CM-variable "LIGHT" will be
1266 :     defined. In effect, the command will not build any cross-compiler
1267 :     backends and therefore finish more quickly.
1268 :    
1269 :     The "fixpt" script also takes a "-light" switch to be able to use
1270 :     this new facility while compiling for a fixpoint.
1271 :    
1272 :     2. I replaced all mentions of anchored paths in group owner specifications
1273 :     with simple relative paths (usually starting with "..").
1274 :     The rationale is that a library's internal workings should not be
1275 :     compromised by the lack of some anchor. (An anchor is necessary
1276 :     for someone who wants to refer to the library by an anchored path,
1277 :     but it should not be necessary to build the same library in the first
1278 :     place.)
1279 :    
1280 :     3. I changed the way CM's tool mechanism determines the shell command
1281 :     string used for things like ml-yacc etc. so that it does not break
1282 :     when CM.Control.implicit_anchors is turned off.
1283 :    
1284 :     ----------------------------------------------------------------------
1285 :     Name: Matthias Blume
1286 : blume 644 Date: 2000/05/12 18:20:00 JST
1287 :     Tag: blume-20000512-ml-build
1288 :     Description:
1289 :    
1290 :     Fixed a bug in config/_ml-build that prevented ml-yacc and ml-lex from
1291 :     getting installed properly (by config/install.sh).
1292 :    
1293 :     ----------------------------------------------------------------------
1294 :     Name: Matthias Blume
1295 : blume 643 Date: 2000/05/12 17:30:00 JST
1296 :     Tag: blume-20000512-anchors
1297 :     Description:
1298 :    
1299 :     !!! NEW BOOT FILES !!!
1300 :    
1301 :     This change is in preparation of fading out support for "implicitly
1302 :     anchored path names". I went through all sources and used the
1303 :     explicit (and relatively new) $-notation. See system/README and the
1304 :     CM manual for more info on this.
1305 :    
1306 :     I also modified the anchoring scheme for some things such as "smlnj",
1307 :     "MLRISC", "cm", etc. to take advantage of the fact that explicit
1308 :     anchors are more expressive: anchor name and first arc do not have to
1309 :     coincide. This entails the following user-visible change:
1310 :    
1311 :     You have to write $smlnj/foo/bar instead of smlnj/foo/bar. In
1312 :     particular, when you fire up sml with a command-line argument, say,
1313 :     e.g.:
1314 :    
1315 :     sml '$smlnj/cmb.cm'
1316 :    
1317 :     At the ML toplevel prompt:
1318 :    
1319 :     CM.autoload "$smlnj/cmb.cm";
1320 :    
1321 :     There is also a new controller in CM.Control that can be used to turn
1322 :     off all remaining support for implicit anchors by saying:
1323 :    
1324 :     CM.autoload "$smlnj/
1325 :     #set CM.Control.implicit_anchors false;
1326 :    
1327 :     This causes CM to reject implicitly anchored paths. This is (for the
1328 :     time being) less permissive than the "final" version where there will
1329 :     be no more such implicit anchors and relative paths will be just that:
1330 :     relative.
1331 :    
1332 :     The next step (version after next version?) will be to make the
1333 :     default for CM.Control.implicit_anchors false. After the dust has
1334 :     settled, I can then produce the "final" version of this...
1335 :    
1336 :     Note: Since bootstrapping is a bit tricky, I provided new boot files.
1337 :    
1338 :     ----------------------------------------------------------------------
1339 :     Name: Matthias Blume
1340 : blume 642 Date: 2000/05/11 16:30:00 JST
1341 :     Tag: blume-20000511-sources
1342 :     Description:
1343 :    
1344 :     The main change is that I added function CM.sources as a generalized
1345 :     version of the earlier CM.makedepend. This entails the following
1346 :     additional changes:
1347 :    
1348 :     - CM.makedepend has been dropped.
1349 :    
1350 :     - CM manual has been updated.
1351 :    
1352 :     - TOOLS signature and API have been changed.
1353 :    
1354 :     ----------------------------------------------------------------------
1355 : leunga 640 Name: Allen Leung
1356 : leunga 641 Date: 2000/05/10 21:17 EDT
1357 :     Tag: leunga-20000510-moby-c--ssa
1358 :     Description:
1359 :    
1360 :     Various bug fixes and new features for C--, Moby and MLRISC optimizations.
1361 :     None of these affect SML/NJ.
1362 :    
1363 :     1. Register Allocation
1364 :    
1365 :     a. A new ra spilling module (ra/ra-spill-with-renaming) is implemented.
1366 :     This module tries to remove local (i.e. basic block level) redundancies
1367 :     during spilling.
1368 :    
1369 :     b. A new framework for performing region based register allocation.
1370 :     Not yet entirely functional.
1371 :    
1372 :     2. X86
1373 :    
1374 :     a. DefUse for POP was missing the stack pointer [found by Lal]
1375 :     b. Reload for CALL was incorrect in X86Spill [found by John]
1376 :     c. Various fixes in X86Spill so that it can be used correctly for
1377 :     the new spilling module.
1378 :    
1379 :     3. SSA/IR
1380 :    
1381 :     a. New module ir/dj-dataflow.sml implements elimination based
1382 :     data flow analysis.
1383 :    
1384 :     4. MLRiscGen
1385 :    
1386 :     a. Fix for gc type annotation
1387 :    
1388 :     5. MDGen
1389 :    
1390 :     Various fixes for machine description -> ml code translation. For ssa
1391 :     only.
1392 :    
1393 :     ----------------------------------------------------------------------
1394 :     Name: Allen Leung
1395 : leunga 640 Date: 2000/05/08 22:17 EDT
1396 :     Tag: leunga-20000508-labexp
1397 :     Description:
1398 :    
1399 :     Fermin has found a few assembly problems with constant expressions
1400 :     generated in LabelExp. Mostly, the problems involve extra parentheses,
1401 :     which choke on dumb assemblers. This is his fix.
1402 :    
1403 :     ----------------------------------------------------------------------
1404 : dbm 635 Name: Dave MacQueen
1405 :     Date: 2000/04/09 14:00 EDT
1406 :     Tag: dbm-20000502-Version_110_28
1407 :     Description:
1408 :    
1409 :     1. Updated src/compiler/TopLevel/main/version.sml to version 110.28
1410 :    
1411 :     2. Updated config/version to 110.28
1412 :    
1413 :     3. Updated config/srcarchiveurl
1414 :    
1415 :     3. New boot files!
1416 :     ftp://ftp.research.bell-labs.com/dist/smlnj/working/110.28/
1417 :    
1418 :     ----------------------------------------------------------------------
1419 : blume 634 Name: Matthias Blume
1420 :     Date: 2000/05/01 19:05:00 JST
1421 :     Tag: blume-20000501-noweb
1422 :     Description:
1423 :    
1424 :     A new noweb tool has been added. The existing system is entirely
1425 :     unaffected by this, but some CM users have asked for renewed noweb
1426 :     support. Everything is documented in the CM manual.
1427 :    
1428 :     New (plugin) libraries:
1429 :    
1430 :     noweb-tool.cm
1431 :     nw-ext.cm
1432 :    
1433 :     ----------------------------------------------------------------------
1434 : dbm 633 Name: Dave MacQueen
1435 :     Date: 2000/04/30 12:40PM EDT
1436 :     Tag: dbm-20000430-bug_fixes
1437 :     Description:
1438 :    
1439 :     1. Fix for bug 1498
1440 :     smlnj/src/system/Basis/Implementation/Unsafe/object.sig
1441 :     smlnj/src/system/Basis/Implementation/Unsafe/object.sml
1442 :     added toRealArray function
1443 :     smlnj/src/compiler/MiscUtil/print/ppobj.sml
1444 :     added check for tag Obj.RealArray to array printing case in ppObj
1445 :    
1446 :     2. Fix for bug 1510
1447 :     smlnj/src/compiler/Semant/types/typesutil.sml
1448 :     fixed definition of dummyargs (used by equalTycon) so that
1449 :     dummy args are distinct types
1450 :    
1451 :     ----------------------------------------------------------------------
1452 : blume 629 Name: Matthias Blume
1453 : blume 632 Date: 2000/04/30 01:00:00 JST
1454 :     Tag: blume-20000430-versions
1455 :     Description:
1456 :    
1457 :     1. CM version numbering added. This is an implementation of Lal's
1458 :     proposal for adding version numbers and version checking to .cm
1459 :     files. Lal said that his proposal was just that -- a proposal.
1460 :     For the time being I went ahead and implemented it so that people
1461 :     can comment on it. Everything is completely backward-compatible
1462 :     (except for the stable library format, i.e., new bootfiles!).
1463 :    
1464 :     As usual, see the CM manual for details.
1465 :    
1466 :     2. An alternative syntax for anchored paths has been implemented.
1467 :     Dave has recently voiced the same concerns that I had when I did
1468 :     this, so there should be some support. My take is that eventually
1469 :     I will let support for the current syntax (where anchors are
1470 :     "implicit") fade out in favor of the new, explicit syntax.
1471 :     In order to be backward-compatible, both old and new syntax are
1472 :     currently supported.
1473 :    
1474 :     Again, see the CM manual for details.
1475 :    
1476 :     3. Parallel make is trying to be slightly smarter: When the master
1477 :     process finds a "bottleneck", i.e., when there is only one
1478 :     compilation unit that can be compiled and everybody else is
1479 :     waiting on it, then it will simply compile it directly instead
1480 :     of clumsily telling one of the slaves to do it.
1481 :    
1482 :     4. Support for "unsharing" added. This is necessary in order to be
1483 :     able to have two different versions of the same library running
1484 :     at the same time (e.g., for trying out a new MLRISC while still
1485 :     having the old MLRISC linked into the current compiler, etc.)
1486 :     See the CM manual.
1487 :    
1488 :     5. Simple "makedepend" functionality added for generating Makefile
1489 :     dependency information. (This is rather crude at the moment.
1490 :     Expect some changes here in the future.)
1491 :    
1492 :     6. ".fun" added as a recognized suffix for ML files. Also documented
1493 :     explicitly in the manual that the fallback behavior (unknown suffix
1494 :     -> ML file) is not an official feature!
1495 :    
1496 :     7. Small changes to the pickler for stable libraries.
1497 :    
1498 :     8. Several internal changes to CM (for cleanup/improvement).
1499 :    
1500 :    
1501 :     !!!! NEW BINFILES !!!!
1502 :    
1503 :     ----------------------------------------------------------------------
1504 :     Name: Matthias Blume
1505 : blume 631 Date: 2000/04/28 17:30:00 JST
1506 :     Tag: blume-20000428-pathconfig
1507 :     Description:
1508 :    
1509 :     1. I changed config/install.sh to remove duplicate entries from the
1510 :     lib/pathconfig file at the end. Moreover, the final version of
1511 :     lib/pathconfig is sorted alphabetically. The same (sorting) is done
1512 :     in src/system/installml.
1513 :    
1514 :     2. The config/install.sh script now consistently uses relative
1515 :     pathnames in lib/pathconfig whenever the anchor is in the lib
1516 :     directory. (So far this was true for the libraries that come
1517 :     pre-compiled and bundled as part of the bootfiles but not for
1518 :     libraries that are compiled by the script itself.)
1519 :    
1520 :     ----------------------------------------------------------------------
1521 :     Name: Matthias Blume
1522 : blume 629 Date: 2000/04/26 13:10:00 JST
1523 :     Tag: blume-20000426-fun_suffix
1524 :     Description:
1525 :    
1526 :     Added ".fun" as a recognized file name suffix (for ML code).
1527 :    
1528 :     ----------------------------------------------------------------------
1529 : leunga 628 Name: Allen Leung
1530 :     Date: 2000/04/25 17:00:00 EST
1531 :     Tag: leunga-20000425-alpha-ra
1532 :     Description:
1533 :    
1534 :     1. Alpha
1535 :    
1536 :     PSEUDOARITH was missing in AlphaRewrite. This causes an endless loop
1537 :     in C--.
1538 :    
1539 :     2. RA
1540 :    
1541 :     Added a flag "ra-dump-size" to print out the size of the flowgraph
1542 :     and the interference graph.
1543 :    
1544 :     ----------------------------------------------------------------------
1545 : dbm 627 Name: Dave MacQueen
1546 :     Date: 2000/04/25/
1547 :     Tag: dbm-20000425-mlyacc_doc_examples
1548 :     Description:
1549 :     Updated mlyacc.tex sections 5 and 7 for SML '97 and CM.
1550 :     Updated all three examples in src/ml-yacc/examples to run
1551 :     under 110.* using CM.make.
1552 :    
1553 :     ----------------------------------------------------------------------
1554 : leunga 624 Name: Allen Leung
1555 :     Date: 2000/04/20 23:04:00 EST
1556 :     Tag: leunga-20000420-ssa-c---stuff
1557 :     Description:
1558 :    
1559 :     This update synchronizes my repository with Yale's. Most of these
1560 :     changes, however, do not affect SML/NJ at all (the RA is an exception).
1561 :    
1562 :     1. Register Allocator
1563 :    
1564 :     a. An improvement in the interference graph construction:
1565 :     Given a copy
1566 :    
1567 :     s <- t
1568 :    
1569 : leunga 628 no interference edge between s and t is added for this definition of s.
1570 : leunga 624
1571 :     b. I've added two new spill heuristic modules that Fermin and I developed
1572 :     (in the new library RA.cm). These are unused in SML/NJ but maybe
1573 :     useful for others (Moby?)
1574 :    
1575 :     2. X86
1576 :    
1577 :     a. Various fixes in the backend provided by Fermin [C--] and Lal.
1578 :    
1579 :     3. Alpha
1580 :    
1581 :     a. Added the BSR instruction and code generation that goes with it [C--]
1582 :     b. Other fixes too numerous to recount provided by Fermin [C--]
1583 :    
1584 :     4. Regmaps
1585 :    
1586 :     a. The regmaps are not initialized with the identity physical bindings
1587 :     at creation time. This is unneeded.
1588 :    
1589 :     5. MLRISC Optimizations
1590 :    
1591 :     a. The DJ-Graph module can now compute the iterated dominance frontiers
1592 :     intersects with liveness incrementally in linear time! Woohoo!
1593 :     This is now used in my new SSA construction algorithm.
1594 :    
1595 :     b. THe branch reorganization module is now smarter about linear chains of
1596 :     basic blocks.
1597 :    
1598 :    
1599 :     ----------------------------------------------------------------------
1600 : blume 615 Name: Matthias Blume
1601 :     Date: 2000/04/12 13:52:00 JST
1602 :     Tag: blume_main_v110p27_1
1603 :     Description:
1604 :    
1605 :     Changed install.sh script to handle archive files without version number
1606 :     and to use "boot.<arch>-<os>" instead of "sml.boot.<arch>-<os>" for the
1607 :     name of the boot file archive.
1608 :    
1609 :     ----------------------------------------------------------------------
1610 : dbm 607 Name: Dave MacQueen
1611 :     Date: 2000/04/09 14:00 EDT
1612 :     Tag: dbm-20000410-Version_110_27
1613 :     Description:
1614 :    
1615 :     1. Updated src/compiler/TopLevel/main/version.sml to version 110.27
1616 :    
1617 :     2. Updated src/config/version to 110.27
1618 :    
1619 :     3. New boot files!
1620 :    
1621 :     ----------------------------------------------------------------------
1622 : leunga 606 Name: Allen Leung
1623 :     Date: 2000/04/09 19:09:00 EST
1624 :     Tag: leunga-20000409-misc
1625 :     Description:
1626 :    
1627 :     1. Yet another fix for x86 assembly for idivl, imull, mull and friends.
1628 :    
1629 :     2. Miscellaneous improvements to MLRISC (unused in sml/nj)
1630 :    
1631 :     ----------------------------------------------------------------------
1632 : monnier 605 Name: Stefan
1633 :     Date: 2000/04/07 10:00:00 EDT
1634 :     Tag: monnier-20000406-branch-handling
1635 :     Description:
1636 :    
1637 :     Improved handling of branches (mostly those generated from
1638 :     polymorphic equality), removed switchoff and changed the
1639 :     default optimization settings (more cpsopt and less flintopt).
1640 :    
1641 :     ----------------------------------------------------------------------
1642 : leunga 601 Name: Allen Leung
1643 : leunga 602 Date: 2000/04/06 01:30:00 EST
1644 :     Tag: leunga-20000406-peephole-x86-SSA-2
1645 :     Description:
1646 :    
1647 :     Forgot a few files.
1648 :    
1649 :     ----------------------------------------------------------------------
1650 :     Name: Allen Leung
1651 : leunga 601 Date: 2000/04/06 00:36:00 EST
1652 :     Tag: leunga-20000406-peephole-x86-SSA
1653 :     Description:
1654 :    
1655 :     1. New Peephole code
1656 :    
1657 :     2. Minor improvement to X86 instruction selection
1658 :    
1659 :     3. Various fixes to SSA and machine description -> code translator
1660 :    
1661 :     ----------------------------------------------------------------------
1662 : blume 595 Name: Matthias Blume
1663 :     Date: 2000/04/05 12:30:00 JST
1664 :     Tag: blume_main_v110p26p2_3
1665 :     Description:
1666 :    
1667 :     This update just merges three minor cosmetic updates to CM's sources
1668 :     to get ready for the 110.27 code freeze on Friday. No functionality
1669 :     has changed.
1670 :    
1671 :     ----------------------------------------------------------------------
1672 : leunga 593 Name: Allen Leung
1673 : leunga 594 Date: 2000/04/04 19:39:00 EST
1674 :     Tag: leunga-20000404-x86-asm
1675 :     Description:
1676 :    
1677 :     1. Fixed a problem in X86 assembly.
1678 :    
1679 :     Things like
1680 :    
1681 :     jmp %eax
1682 :     jmp (%eax)
1683 :    
1684 :     should be output as
1685 :    
1686 :     jmp *%eax
1687 :     jmp *(%eax)
1688 :    
1689 :     2. Assembly output
1690 :    
1691 :     Added a new flag
1692 :    
1693 :     "asm-indent-copies" (default to false)
1694 :    
1695 :     When this flag is on, parallel copies will be indented an extra level.
1696 :    
1697 :     ----------------------------------------------------------------------
1698 :     Name: Allen Leung
1699 : leunga 593 Date: 2000/04/04 03:18:00 EST
1700 :     Tag: leunga-20000404-C--Moby
1701 :     Description:
1702 :    
1703 :     All of these fixes are related to C--, Moby, and my own optimization
1704 :     stuff; so they shouldn't affect SML/NJ.
1705 :    
1706 :     1. X86
1707 :    
1708 :     Various fixes related floating point, and extensions.
1709 :    
1710 :     2. Alpha
1711 :    
1712 :     Some extra patterns related to loads with signed/zero extension
1713 :     provided by Fermin.
1714 :    
1715 :     3. Assembly
1716 :    
1717 : monnier 659 When generating assembly, resolve the value of client defined constants,
1718 : leunga 593 instead of generating symbolic values. This is controlled by the
1719 :     new flag "asm-resolve-constants", which is default to true.
1720 :    
1721 :     4. Machine Descriptions
1722 :    
1723 :     a. The precedence parser was slightly broken when parsing infixr symbols.
1724 :     b. The type generalizing code had the bound variables reversed, resulting
1725 :     in a problem during arity raising.
1726 :     c. Various fixes in machine descriptions.
1727 :    
1728 :     ----------------------------------------------------------------------
1729 : blume 592 Name: Matthias Blume
1730 :     Date: 2000/04/03 16:05:00 JST
1731 :     Tag: blume_main_v110p26p2_2
1732 :     Description:
1733 :    
1734 :     I eliminated coreEnv from compInfo. Access to the "Core" structure is
1735 :     now done via the ordinary static environment that is context to each
1736 :     compilation unit.
1737 :    
1738 :     To this end, I arranged that instead of "structure Core" as "structure
1739 :     _Core" is bound in the pervasive environment. Core access is done via
1740 : monnier 659 _Core (which can never be accidentally rebound because _Core is not a
1741 : blume 592 legal surface-syntax symbol).
1742 :    
1743 :     The current solution is much cleaner because the core environment is
1744 :     now simply part of the pervasive environment which is part of every
1745 :     compilation unit's context anyway. In particular, this eliminates all
1746 :     special-case handling that was necessary until now in order to deal
1747 :     with dynamic and symbolic parts of the core environment.
1748 :    
1749 :     Remaining hackery (to bind the "magic" symbol _Core) is localized in the
1750 : monnier 659 compilation manager's bootstrap compiler (actually: in the "init group"
1751 : blume 592 handling). See the comments in src/system/smlnj/init/init.cmi for
1752 :     more details.
1753 :    
1754 :     I also tried to track down all mentions of "Core" (as string argument
1755 :     to Symbol.strSymbol) in the compiler and replaced them with a
1756 :     reference to the new CoreSym.coreSym. Seems cleaner since the actual
1757 :     name appears in one place only.
1758 :    
1759 :     Binfile and bootfile format have not changed, but the switchover from
1760 :     the old "init.cmi" to the new one is a bit tricky, so I supplied new
1761 :     bootfiles anyway.
1762 :    
1763 :     ----------------------------------------------------------------------
1764 : leunga 591 Name: Allen Leung
1765 :     Date: 2000/04/02 21:17:00 EST
1766 :     Tag: leunga-20000402-mltree
1767 :     Description:
1768 : leunga 585
1769 : leunga 591 1. Renamed the constructor CALL in MLTREE by popular demand.
1770 :     2. Added a bunch of files from my repository. These are currently
1771 :     used by other non-SMLNJ backends.
1772 :    
1773 : leunga 576 ----------------------------------------------------------------------
1774 : leunga 590 Name: Allen Leung
1775 :     Date: 2000/03/31 21:15:00 EST
1776 :     Tag: leunga-20000331-aliasing
1777 :     Description:
1778 :    
1779 :     This update contains a rewritten (and hopefully more correct) module
1780 :     for extracting aliasing information from CPS.
1781 :    
1782 :     To turn on this feature:
1783 :    
1784 :     Compiler.Control.CG.memDisambiguate := true
1785 :    
1786 :     To pretty print the region information with assembly
1787 :    
1788 :     Compiler.Control.MLRISC.getFlag "asm-show-region" := true;
1789 :    
1790 :     To control how many levels of aliasing information are printed, use:
1791 :    
1792 :     Compiler.Control.MLRISC.getInt "points-to-show-level" := n
1793 :    
1794 :     The default of n is 3.
1795 :    
1796 :     ----------------------------------------------------------------------
1797 : dbm 589 Name: David MacQueen
1798 :     Date: 2000/03/31 11:15:00 EST
1799 :     Tag: dbm-20000331-runtime_fix
1800 :     Description:
1801 :    
1802 :     This update contains:
1803 :    
1804 :     1. runtime/c-lib/c-libraries.c
1805 :     includes added in revision 1.2 caused compilation errors on hppa-hpux
1806 :    
1807 :     2. fix for bug 1556
1808 :     system/Basis/Implementation/NJ/internal-signals.sml
1809 :    
1810 :     ----------------------------------------------------------------------
1811 : blume 587 Name: Matthias Blume
1812 : blume 588 Date: 2000/03/31 18:00:00 JST
1813 :     Tag: blume_main_v110p26p2_1
1814 :     Description:
1815 :    
1816 :     This update contains:
1817 :    
1818 :     1. A small change to CM's handling of stable libraries:
1819 :     CM now maintains one "global" modmap that is used for all stable
1820 :     libraries. The use of such a global modmap maximizes sharing and
1821 :     minimizes the need for re-traversing parts of environments during
1822 :     modmap construction. (However, this has minor impact since modmap
1823 :     construction seems to account for just one percent or less of total
1824 :     compile time.)
1825 :    
1826 :     2. I added a "genmap" phase to the statistics. This is where I got the
1827 :     "one percent" number (see above).
1828 :    
1829 :     3. CM's new tool parameter mechanism just became _even_ better. :)
1830 :     - The parser understands named parameters and recursive options.
1831 :     - The "make" and "shell" tools use these new features.
1832 :     (This makes it a lot easier to cascade these tools.)
1833 :     - There is a small syntax change: named parameters use a
1834 :    
1835 :     <name> : ( <option> ... ) or
1836 :     <name> : <string>
1837 :    
1838 :     syntax. Previously, named parameters were implemented in an
1839 :     ad-hoc fashion by each tool individually (by parsing strings)
1840 :     and had the form
1841 :    
1842 :     <name>=<string>
1843 :    
1844 :     See the CM manual for a full description of these issues.
1845 :    
1846 :     ----------------------------------------------------------------------
1847 :     Name: Matthias Blume
1848 : blume 587 Date: 2000/03/30 18:00:00 JST
1849 :     Tag: blume_main_v110p26p2_0
1850 :     Description:
1851 :    
1852 :     !!!!! WARNING !!!!!!
1853 :     !! New binfiles !!
1854 :     !!!!!!!!!!!!!!!!!!!!
1855 :    
1856 :     This update contains:
1857 :    
1858 :     1. Moderate changes to CM:
1859 :    
1860 :     - Changes to CM's tools mechanism. In particular, it is now possible
1861 :     to have tools that accept additional "command line" parameters
1862 :     (specified in the .cm file at each instance where the tool's class is
1863 :     used).
1864 :    
1865 : monnier 659 This was done to accommodate the new "make" and "shell" tools which
1866 :     facilitate fairly seamless hookup to portions of code managed using
1867 : blume 587 Makefiles or Shell scripts.
1868 :    
1869 :     There are no classes "shared" or "private" anymore. Instead, the
1870 :     sharing annotation is now a parameter to the "sml" class.
1871 :    
1872 :     There is a bit of generic machinery for implementing one's own
1873 :     tools that accept command-line parameters. However, I am not yet fully
1874 :     satisfied with that part, so expect changes here in the future.
1875 :    
1876 :     All existing tools are described in the CM manual.
1877 :    
1878 : monnier 659 - Slightly better error handling. (CM now suppresses many followup
1879 : blume 587 error messages that tended to be more annoying than helpful.)
1880 :    
1881 :     2. Major changes to the compiler's static environment data structures.
1882 :    
1883 :     - no CMStaticEnv anymore.
1884 :     - no CMEnv, no "BareEnvironment" (actually, _only_ BareEnvironment,
1885 :     but it is called Environment), no conversions between different
1886 :     kinds of static environments
1887 :    
1888 :     - There is still a notion of a "modmap", but such modmaps are generated
1889 :     on demand at the time when they are needed. This sounds slow, but I
1890 :     sped up the code that generates modmaps enough for this not to lead to
1891 :     a slowdown of the compiler (at least I didn't detect any).
1892 :    
1893 :     - To facilitate rapid modmap generation, static environments now
1894 :     contain an (optional) "modtree" structure. Modtree annotations are
1895 :     constructed by the unpickler during unpickling. (This means that
1896 :     the elaborator does not have to worry about modtrees at all.)
1897 :     Modtrees have the advantage that they are compositional in the same
1898 :     way as the environment data structure itself is compositional.
1899 :     As a result, modtrees never hang on to parts of an environment that
1900 :     has already been rendered "stale" by filtering or rebinding.
1901 :    
1902 :     - I went through many, many trials and errors before arriving at the
1903 :     current solution. (The initial idea of "linkpaths" did not work.)
1904 :     But the result of all this is that I have touched a lot of files that
1905 :     depend on the "modules" and "types" data structures (most of the
1906 :     elaborator). There were a lot of changes during my "linkpath" trials
1907 :     that could have been reverted to their original state but weren't.
1908 :     Please, don't be too harsh on me for messing with this code a bit more
1909 :     than what was strictly necessary... (I _did_ resist the tempation
1910 :     of doing any "global reformatting" to avoid an untimely death at
1911 :     Dave's hands. :)
1912 :    
1913 :     - One positive aspect of the previous point: At least I made sure that
1914 :     all files that I touched now compile without warnings (other than
1915 :     "polyEqual").
1916 :    
1917 :     - compiler now tends to run "leaner" (i.e., ties up less memory in
1918 :     redundant modmaps)
1919 :    
1920 :     ----------------------------------------------------------------------
1921 : leunga 580 Name: Allen Leung
1922 : leunga 585 Date: 2000/03/29 18:00:00
1923 :     Tag: leunga-20000327-mlriscGen_hppa_alpha_x86
1924 : leunga 586 Boot files (optional): ftp://react-ilp.cs.nyu.edu/leunga/110.26.1-sml.boot.x86-unix-20000330.tar.gz
1925 : leunga 585 Description:
1926 :    
1927 :     This update contains *MAJOR* changes to the way code is generated from CPS
1928 :     in the module mlriscGen, and in various backend modules.
1929 :    
1930 :     CHANGES
1931 :     =======
1932 :    
1933 :     1. MLRiscGen: forward propagation fix.
1934 :    
1935 :     There was a bug in forward propagation introduced at about the same time
1936 :     as the MLRISC x86 backend, which prohibits coalescing to be
1937 :     performed effectively in loops.
1938 :    
1939 :     Effect: speed up of loops in RISC architectures.
1940 :     By itself, this actually slowed down certain benchmarks on the x86.
1941 :    
1942 :     2. MLRiscGen: forward propagating addresses from consing.
1943 :    
1944 :     I've changed the way consing code is generated. Basically I separated
1945 :     out the initialization part:
1946 :    
1947 :     store tag, offset(allocptr)
1948 :     store elem1, offset+4(allocptr)
1949 :     store elem2, offset+8(allocptr)
1950 :     ...
1951 :     store elemn, offset+4n(allocptr)
1952 :    
1953 :     and the address computation part:
1954 :    
1955 :     celladdr <- offset+4+alloctpr
1956 :    
1957 :     and move the address computation part
1958 :    
1959 :     Effect: register pressure is generally lower as a result. This
1960 :     makes compilation of certain expressions much faster, such as
1961 :     long lists with non-trivial elements.
1962 :    
1963 :     [(0,0), (0,0), .... (0,0)]
1964 :    
1965 :     3. MLRiscGen: base pointer elimination.
1966 :    
1967 :     As part of the linkage mechanism, we generate the sequence:
1968 :    
1969 :     L: ... <- start of the code fragment
1970 :    
1971 :     L1:
1972 :     base pointer <- linkreg - L1 + L
1973 :    
1974 :     The base pointer was then used for computing relocatable addresses
1975 :     in the code fragment. Frequently (such as in lots of continuations)
1976 :     this is not needed. We now eliminate this sequence whenever possible.
1977 :    
1978 :     For compile time efficiency, I'm using a very stupid local heuristic.
1979 :     But in general, this should be done as a control flow analysis.
1980 :    
1981 :     Effect: Smaller code size. Speed up of most programs.
1982 :    
1983 :     4. Hppa back end
1984 :    
1985 :     Long jumps in span dependence resolution used to depend on the existence
1986 :     of the base pointer.
1987 :    
1988 :     A jump to a long label L was expanded into the following sequence:
1989 :    
1990 :     LDIL %hi(L-8192), %r29
1991 :     LDO %lo(L-8192)(%r29), %r29
1992 :     ADD %r29, baseptr, %r29
1993 :     BV,n %r0(%r29)
1994 :    
1995 :     In the presence of change (3) above, this will not work. I've changed
1996 :     it so that the following sequence of instructions are generated, which
1997 :     doesn't mention the base pointer at all:
1998 :    
1999 :     BL,n L', %r29 /* branch and link, L' + 4 -> %r29 */
2000 :     L': ADDIL L-(L'+4), %r29 /* Compute address of L */
2001 :     BV,n %r0(%r29) /* Jump */
2002 :    
2003 :     5. Alpha back end
2004 :    
2005 :     New alpha instructions LDB/LDW have been added, as per Fermin's
2006 :     suggestions. This is unrelated to all other changes.
2007 :    
2008 :     6. X86 back end
2009 :    
2010 :     I've changed andl to testl in the floating point test sequence
2011 :     whenever appropriate. The Intel optimization guide states that
2012 : monnier 659 testl is preferable to andl.
2013 : leunga 585
2014 :     7. RA (x86 only)
2015 :    
2016 :     I've improved the spill propagation algorithm, using an approximation
2017 :     of maximal weighted independent sets. This seems to be necessary to
2018 :     alleviate the negative effect in light of the slow down in (1).
2019 :    
2020 :     I'll write down the algorithm one of these days.
2021 :    
2022 :     8. MLRiscGen: frequencies
2023 :    
2024 :     I've added an annotation that states that all call gc blocks have zero
2025 :     execution frequencies. This improves register allocation on the x86.
2026 :    
2027 :     BENCHMARKS
2028 :     ==========
2029 :    
2030 :     I've only perform the comparison on 110.25.
2031 :    
2032 :     The platforms are:
2033 :    
2034 :     HPPA A four processor HP machine (E9000) with 5G of memory.
2035 :     X86 A 300Hhz Pentium II with 128M of memory, and
2036 :     SPARC An Ultra sparc 2 with 512M of memory.
2037 :    
2038 :     I used the following parameters for the SML benchmarks:
2039 :    
2040 :     @SMLalloc
2041 :     HPPA 256k
2042 :     SPARC 512k
2043 :     X86 256k
2044 :    
2045 :     COMPILATION TIME
2046 :     ----------------
2047 :     Here are the numbers comparing the compilation times of the compilers.
2048 :     I've only compared 110.25 compiling the new sources versus
2049 :     a fixpoint version of the new compiler compiling the same.
2050 :    
2051 :     110.25 New
2052 :     Total Time in RA Spill+Reload Total Time In RA Spill+Reload
2053 :     HPPA 627s 116s 2684+3584 599s 95s 1003+1879
2054 :     SPARC 892s 173s 2891+3870 708s 116s 1004+1880
2055 :     X86 999s 315s 94006+130691 987s 296s 108877+141957
2056 :    
2057 :     110.25 New
2058 :     Code Size Code Size
2059 :     HPPA 8596736 8561421
2060 :     SPARC 8974299 8785143
2061 :     X86 9029180 8716783
2062 :    
2063 :     So in summary, things are at least as good as before. Dramatic
2064 :     reduction in compilation is obtained on the Sparc; I can't explain it,
2065 :     but it is reproducible. Perhaps someone should try to reproduce this
2066 :     on their own machines.
2067 :    
2068 :     SML BENCHMARKS
2069 :     --------------
2070 :    
2071 :     On the average, all benchmarks perform at least as well as before.
2072 :    
2073 :     HPPA Compilation Time Spill+Reload Run Time
2074 :     110.25 New 110.25 New 110.25 New
2075 :    
2076 :     barnesHut 3.158 3.015 4.75% 1+1 0+0 2.980 2.922 2.00%
2077 :     boyer 6.152 5.708 7.77% 0+0 0+0 0.218 0.213 2.34%
2078 :     count-graphs 1.168 1.120 4.32% 0+0 0+0 22.705 23.073 -1.60%
2079 :     fft 0.877 0.792 10.74% 1+3 1+3 0.602 0.587 2.56%
2080 :     knuthBendix 3.180 2.857 11.32% 0+0 0+0 0.675 0.662 2.02%
2081 :     lexgen 6.190 5.290 17.01% 0+0 0+0 0.913 0.788 15.86%
2082 :     life 0.803 0.703 14.22% 25+25 0+0 0.153 0.140 9.52%
2083 :     logic 2.048 2.007 2.08% 6+6 1+1 4.133 4.008 3.12%
2084 :     mandelbrot 0.077 0.080 -4.17% 0+0 0+0 0.765 0.712 7.49%
2085 :     mlyacc 22.932 20.937 9.53% 154+181 32+57 0.468 0.430 8.91%
2086 :     nucleic 5.183 5.060 2.44% 2+2 0+0 0.125 0.120 4.17%
2087 :     ratio-regions 3.357 3.142 6.84% 0+0 0+0 116.225 113.173 2.70%
2088 :     ray 1.283 1.290 -0.52% 0+0 0+0 2.887 2.855 1.11%
2089 :     simple 6.307 6.032 4.56% 28+30 5+7 3.705 3.658 1.28%
2090 :     tsp 0.888 0.862 3.09% 0+0 0+0 7.040 6.893 2.13%
2091 :     vliw 24.378 23.455 3.94% 106+127 25+45 2.758 2.707 1.91%
2092 :     --------------------------------------------------------------------------
2093 :     Average 6.12% 4.09%
2094 :    
2095 :     SPARC Compilation Time Spill+Reload Run Time
2096 :     110.25 New 110.25 New 110.25 New
2097 :    
2098 :     barnesHut 3.778 3.592 5.20% 2+2 0+0 3.648 3.453 5.65%
2099 :     boyer 6.632 6.110 8.54% 0+0 0+0 0.258 0.242 6.90%
2100 :     count-graphs 1.435 1.325 8.30% 0+0 0+0 33.672 34.737 -3.07%
2101 :     fft 0.980 0.940 4.26% 3+9 2+6 0.838 0.827 1.41%
2102 :     knuthBendix 3.590 3.138 14.39% 0+0 0+0 0.962 0.967 -0.52%
2103 :     lexgen 6.593 6.072 8.59% 1+1 0+0 1.077 1.078 -0.15%
2104 :     life 0.972 0.868 11.90% 26+26 0+0 0.143 0.140 2.38%
2105 :     logic 2.525 2.387 5.80% 7+7 1+1 5.625 5.158 9.05%
2106 :     mandelbrot 0.090 0.093 -3.57% 0+0 0+0 0.855 0.728 17.39%
2107 :     mlyacc 26.732 23.827 12.19% 162+189 32+57 0.550 0.560 -1.79%
2108 :     nucleic 6.233 6.197 0.59% 3+3 0+0 0.163 0.173 -5.77%
2109 :     ratio-regions 3.780 3.507 7.79% 0+0 0+0 133.993 131.035 2.26%
2110 :     ray 1.595 1.550 2.90% 1+1 0+0 3.440 3.418 0.63%
2111 :     simple 6.972 6.487 7.48% 29+32 5+7 3.523 3.525 -0.05%
2112 :     tsp 1.115 1.063 4.86% 0+0 0+0 7.393 7.265 1.77%
2113 :     vliw 27.765 24.818 11.87% 110+135 25+45 2.265 2.135 6.09%
2114 :     ----------------------------------------------------------------------------
2115 :     Average 6.94% 2.64%
2116 :    
2117 :     X86 Compilation Time Spill+Reload Run Time
2118 :     110.25 New 110.25 New 110.25 New
2119 :    
2120 :     barnesHut 5.530 5.420 2.03% 593+893 597+915 3.532 3.440 2.66%
2121 :     boyer 8.768 7.747 13.19% 493+199 301+289 0.327 0.297 10.11%
2122 :     count-graphs 2.040 2.010 1.49% 298+394 315+457 26.578 28.660 -7.26%
2123 :     fft 1.327 1.302 1.92% 112+209 115+210 1.055 0.962 9.71%
2124 :     knuthBendix 5.218 5.475 -4.69% 451+598 510+650 0.928 0.932 -0.36%
2125 :     lexgen 9.970 9.623 3.60% 1014+841 1157+885 0.947 0.928 1.97%
2126 :     life 1.183 1.183 0.00% 162+182 145+148 0.127 0.103 22.58%
2127 :     logic 3.285 3.512 -6.45% 514+684 591+836 5.682 5.577 1.88%
2128 :     mandelbrot 0.147 0.143 2.33% 38+41 33+54 0.703 0.690 1.93%
2129 :     mlyacc 35.457 32.763 8.22% 3496+4564 3611+4860 0.552 0.550 0.30%
2130 :     nucleic 7.100 6.888 3.07% 239+168 201+158 0.175 0.173 0.96%
2131 :     ratio-regions 6.388 6.843 -6.65% 1182+257 981+300 120.142 120.345 -0.17%
2132 :     ray 2.332 2.338 -0.29% 346+398 402+494 3.593 3.540 1.51%
2133 :     simple 9.912 9.903 0.08% 1475+941 1579+1168 3.057 3.178 -3.83%
2134 :     tsp 1.623 1.532 5.98% 266+200 250+211 8.045 7.878 2.12%
2135 :     vliw 33.947 35.470 -4.29% 2629+2774 2877+3171 2.072 1.890 9.61%
2136 :     ----------------------------------------------------------------------------
2137 :     Average 1.22% 3.36%
2138 :    
2139 :     ----------------------------------------------------------------------
2140 :     Name: Allen Leung
2141 : leunga 583 Date: 2000/03/23 16:25:00
2142 :     Tag: leunga-20000323-fix_x86_alpha
2143 :     Description:
2144 :    
2145 :     1. X86 fixes/changes
2146 :    
2147 :     a. The old code generated for SETcc was completely wrong.
2148 :     The Intel optimization guide is VERY misleading.
2149 :    
2150 :     2. ALPHA fixes/changes
2151 :    
2152 :     a. Added the instructions LDBU, LDWU, STB, STW as per Fermin's suggestion.
2153 :     b. Added a new mode byteWordLoadStores to the functor parameter to Alpha()
2154 :     c. Added reassociation code for address computation.
2155 :    
2156 :     ----------------------------------------------------------------------
2157 :     Name: Allen Leung
2158 : leunga 580 Date: 2000/03/22 01:23:00
2159 :     Tag: leunga-20000322-fix_x86_hppa_ra
2160 :     Description:
2161 :    
2162 :     1. X86 fixes/changes
2163 :    
2164 :     a. x86Rewrite bug with MUL3 (found by Lal)
2165 :     b. Added the instructions FSTS, FSTL
2166 :    
2167 :     2. PA-RISC fixes/changes
2168 :    
2169 :     a. B label should not be a delay slot candidate! Why did this work?
2170 :     b. ADDT(32, REG(32, r), LI n) now generates one instruction instead of two,
2171 :     as it should be.
2172 :     c. The assembly syntax for fstds and fstdd was wrong.
2173 :     d. Added the composite instruction COMICLR/LDO, which is the immediate
2174 :     operand variant of COMCLR/LDO.
2175 :    
2176 :     3. Generic MLRISC
2177 :    
2178 :     a. shuffle.sml rewritten to be slightly more efficient
2179 :     b. DIV bug in mltree-simplify fixed (found by Fermin)
2180 :    
2181 :     4. Register Allocator
2182 :    
2183 :     a. I now release the interference graph earlier during spilling.
2184 :     May improve memory usage.
2185 :    
2186 :     ----------------------------------------------------------------------
2187 : blume 577 Name: Matthias Blume
2188 : blume 578 Date: 2000/03/14 14:15:32
2189 :     Tag: blume_main_v110p26p1_2
2190 :     Description:
2191 :    
2192 :     1. Tools.registerStdShellCmdTool (from smlnj/cm/tool.cm) takes an
2193 :     additional argument called "template" which is an optional string that
2194 : monnier 659 specifies the layout of the tool command line. See the CM manual for
2195 : blume 578 explanation.
2196 :    
2197 : monnier 659 2. A special-purpose tool can be "registered" by simply dropping the
2198 : blume 578 corresponding <...>-tool.cm (and/or <...>-ext.cm) into the same
2199 :     directory where the .cm file lives that uses this tool. (The
2200 :     behavior/misfeature until now was to look for the tool description
2201 :     files in the current working directory.) As before, tool description
2202 :     files could also be anchored -- in which case they can live anywhere
2203 :     they like. Following the recent e-mail discussion, this change should
2204 :     make it easier to have special-purpose tools that are shipped together
2205 :     with the sources of the program that uses them.
2206 :    
2207 :     ----------------------------------------------------------------------
2208 :     Name: Matthias Blume
2209 : blume 577 Date: 2000/03/10 07:48:34
2210 :     Tag: blume_main_v110p26p1_1
2211 :     Description:
2212 :    
2213 :     I added a re-written version of Dave's fixpt script to src/system.
2214 :     Changes relative to the original version:
2215 :     - sh-ified (not everybody has ksh)
2216 :     - automatically figures out which architecture it runs on
2217 :     - uses ./makeml a bit more cleverly
2218 :     - never invokes ./installml (and, thus, does not clobber your
2219 :     good and working installation of sml in case something goes wrong)
2220 :     - accepts max iteration count using option "-iter <n>"
2221 :     - accepts a "base" name using option "-base <base>"
2222 :    
2223 :     It does not build any extraneous heap images but directly rebuilds
2224 :     bin- and boot-hierarchies using makeml's "-rebuild" switch. Finally,
2225 :     it can incorporate existing bin- and boot- hierarchies. For example,
2226 :     suppose the base is set to "sml" (which is the default). Then it
2227 :     successively builds
2228 :    
2229 :     sml.bin.<arch>-unix and sml.boot.<arch>-unix
2230 :     then sml1.bin.<arch>-unix and sml1.boot.<arch>-unix
2231 :     then sml2.bin.<arch>-unix and sml2.boot.<arch>-unix
2232 :     ...
2233 :     then sml<n>.bin.<arch>-unix and sml<n>.boot.<arch>-unix
2234 :    
2235 :     and so on. If any of these already exist, it will just use what's
2236 :     there. In particular, many people will have the initial set of bin
2237 :     and boot files around, so this saves time for at least one full
2238 :     rebuild. Having sets of the form <base><k>.{bin,boot}.<arch>-unix for
2239 :     <k>=1,2,... is normally not a good idea when invoking fixpt. However,
2240 :     they might be the result of an earlier partial run of fixpt (which
2241 : monnier 659 perhaps got accidentally killed). In this case, fixpt will quickly
2242 : blume 577 move through what exists before continuing where it left off earlier,
2243 :     and, thus, saves a lot of time.
2244 :    
2245 :     ----------------------------------------------------------------------
2246 : leunga 576 Name: Allen Leung
2247 :     Date: 00/03/10 02:20:00
2248 :     Tag: leunga-20000310-fix_x86_asm_ra
2249 :     Description:
2250 : dbm 570
2251 : leunga 576 More assembly output problems involving the indexed addressing mode
2252 :     on the x86 have been found and corrected. Thanks to Fermin Reig for the
2253 :     fix.
2254 :    
2255 :     The interface and implementation of the register allocator have been changed
2256 :     slightly to accommodate the possibility to skip the register allocation
2257 :     phases completely and go directly to memory allocation. This is needed
2258 :     for C-- use.
2259 :    
2260 : dbm 570 ----------------------------------------------------------------------
2261 : blume 572 Name: Matthias Blume
2262 : blume 575 Date: 00/03/09 10:23:53
2263 :     Tag: blume_main_v110p26p1_0
2264 :     Description:
2265 :    
2266 :     * Complete re-organization of library names. Many libraries have been
2267 :     consolidated so that they share the same path anchor. For example,
2268 :     all MLRISC-related libraries are anchored at MLRISC, most libraries that
2269 :     are SML/NJ-specific are under "smlnj". Notice that names like
2270 :     host-cmb.cm or host-compiler.cm no longer exist. See system/README
2271 :     for a complete description of the new naming scheme. Quick reference:
2272 :    
2273 :     host-cmb.cm -> smlnj/cmb.cm
2274 :     host-compiler.cm -> smlnj/compiler.cm
2275 :     full-cm.cm -> smlnj/cm.cm
2276 :     <arch>-<os>.cm -> smlnj/cmb/<arch>-<os>.cm
2277 :     <arch>-compiler.cm -> smlnj/compiler/<arch>.cm
2278 :    
2279 :     * Bug fixes in CM.
2280 :     - exceptions in user code are being passed through (i.e., reach top level)
2281 :     - more bugs in paranoia mode fixed
2282 :     - bug related to checking group owners fixed
2283 :    
2284 :     * New install.sh script that automagically fetches archive files:
2285 :     The new file config/srcarchiveurl must contain the URL of the
2286 :     (remote) directory that contains bin files (or other source archives).
2287 :     If install.sh does not find the archive locally, it tries to get
2288 :     it from that remote directory.
2289 :     This should simplify installation further: For machines that have
2290 :     access to the internet, just fetch <version>-config.tgz, unpack it,
2291 : monnier 659 edit config/targets, and go (run config/install.sh). The script will
2292 : blume 575 fetch everything else that it might need all by itself.
2293 :    
2294 :     For CVS users, this mechanism is not relevant for source archives, but
2295 :     it is convenient for getting new sets of binfiles.
2296 :    
2297 :     Archives should be tar files compressed with either gzip, compress, or
2298 :     bzip2. The script recognizes .tgz, .tar, tar.gz, tz, .tar.Z, and .tar.bz2.
2299 :    
2300 :     ----------------------------------------------------------------------
2301 :     Name: Matthias Blume
2302 : blume 572 Date: 2000/03/07 04:01:04
2303 :     Tag: blume_main_v110_26_2
2304 : dbm 570 Description:
2305 : blume 572 - size info in BOOTLIST
2306 :     * no fixed upper limits for number of bootfiles or length of
2307 :     bootfile names in runtime
2308 :     * falling back to old behavior if no BOOTLIST size info found
2309 :     - allocation size heuristics in .run-sml
2310 :     * tries to read cache size from /proc/cpuinfo (this is important for
2311 :     small-cache Celeron systems!)
2312 :     - install.sh robustified
2313 :     - CM manual updates
2314 :     - paranoid mode
2315 :     * no more CMB.deliver() (i.e., all done by CMB.make())
2316 :     * can re-use existing sml.boot.* files
2317 :     * init.cmi now treated as library
2318 :     * library stamps for consistency checks
2319 :     - sml.boot.<arch>-<os>/PIDMAP file
2320 :     * This file is read by the CM startup code. This is used to minimize
2321 :     the amount of dynamic state that needs to be stowed away for the
2322 :     purpose of sharing between interactive system and user code.
2323 :     - CM.Anchor.anchor instead of CM.Anchor.{set,cancel}
2324 :     * Upon request by Elsa. Anchors now controlled by get-set-pair
2325 :     like most other CM state variables.
2326 :     - Compiler.CMSA eliminated
2327 :     * No longer supported by CM anyway.
2328 :     - fixed bugs in pickler that kept biting Stefan
2329 :     * past refs to past refs (was caused by the possibility that
2330 :     ad-hoc sharing is more discriminating than hash-cons sharing)
2331 :     * integer overflow on LargeInt.minInt
2332 :     - ml-{lex,yacc} build scripts now use new mechanism
2333 :     for building standalone programs
2334 :     - fixed several gcc -Wall warnings that were caused by missing header
2335 :     files, missing initializations, etc., in runtime (not all warnings
2336 :     eliminated, though)

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