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/NOTES/HISTORY
ViewVC logotype

Annotation of /sml/trunk/NOTES/HISTORY

Parent Directory Parent Directory | Revision Log Revision Log


Revision 615 - (view) (download)
Original Path: sml/trunk/HISTORY

1 : dbm 570 This is the HISTORY file for the Yale SML/NJ CVS repository.
2 :    
3 :     An entry should be made for _every_ commit to the repository.
4 :     The entries in this file will be used when creating the README
5 :     for new versions, so keep that in mind when writing the
6 :     description.
7 :    
8 :     The form of an entry should be:
9 :    
10 :     Name:
11 :     Date:
12 :     Tag: <post-commit CVS tag>
13 :     Description:
14 : leunga 591 ----------------------------------------------------------------------
15 : blume 615 Name: Matthias Blume
16 :     Date: 2000/04/12 13:52:00 JST
17 :     Tag: blume_main_v110p27_1
18 :     Description:
19 :    
20 :     Changed install.sh script to handle archive files without version number
21 :     and to use "boot.<arch>-<os>" instead of "sml.boot.<arch>-<os>" for the
22 :     name of the boot file archive.
23 :    
24 :     ----------------------------------------------------------------------
25 : dbm 607 Name: Dave MacQueen
26 :     Date: 2000/04/09 14:00 EDT
27 :     Tag: dbm-20000410-Version_110_27
28 :     Description:
29 :    
30 :     1. Updated src/compiler/TopLevel/main/version.sml to version 110.27
31 :    
32 :     2. Updated src/config/version to 110.27
33 :    
34 :     3. New boot files!
35 :    
36 :     ----------------------------------------------------------------------
37 : leunga 606 Name: Allen Leung
38 :     Date: 2000/04/09 19:09:00 EST
39 :     Tag: leunga-20000409-misc
40 :     Description:
41 :    
42 :     1. Yet another fix for x86 assembly for idivl, imull, mull and friends.
43 :    
44 :     2. Miscellaneous improvements to MLRISC (unused in sml/nj)
45 :    
46 :     ----------------------------------------------------------------------
47 : monnier 605 Name: Stefan
48 :     Date: 2000/04/07 10:00:00 EDT
49 :     Tag: monnier-20000406-branch-handling
50 :     Description:
51 :    
52 :     Improved handling of branches (mostly those generated from
53 :     polymorphic equality), removed switchoff and changed the
54 :     default optimization settings (more cpsopt and less flintopt).
55 :    
56 :     ----------------------------------------------------------------------
57 : leunga 601 Name: Allen Leung
58 : leunga 602 Date: 2000/04/06 01:30:00 EST
59 :     Tag: leunga-20000406-peephole-x86-SSA-2
60 :     Description:
61 :    
62 :     Forgot a few files.
63 :    
64 :     ----------------------------------------------------------------------
65 :     Name: Allen Leung
66 : leunga 601 Date: 2000/04/06 00:36:00 EST
67 :     Tag: leunga-20000406-peephole-x86-SSA
68 :     Description:
69 :    
70 :     1. New Peephole code
71 :    
72 :     2. Minor improvement to X86 instruction selection
73 :    
74 :     3. Various fixes to SSA and machine description -> code translator
75 :    
76 :     ----------------------------------------------------------------------
77 : blume 595 Name: Matthias Blume
78 :     Date: 2000/04/05 12:30:00 JST
79 :     Tag: blume_main_v110p26p2_3
80 :     Description:
81 :    
82 :     This update just merges three minor cosmetic updates to CM's sources
83 :     to get ready for the 110.27 code freeze on Friday. No functionality
84 :     has changed.
85 :    
86 :     ----------------------------------------------------------------------
87 : leunga 593 Name: Allen Leung
88 : leunga 594 Date: 2000/04/04 19:39:00 EST
89 :     Tag: leunga-20000404-x86-asm
90 :     Description:
91 :    
92 :     1. Fixed a problem in X86 assembly.
93 :    
94 :     Things like
95 :    
96 :     jmp %eax
97 :     jmp (%eax)
98 :    
99 :     should be output as
100 :    
101 :     jmp *%eax
102 :     jmp *(%eax)
103 :    
104 :     2. Assembly output
105 :    
106 :     Added a new flag
107 :    
108 :     "asm-indent-copies" (default to false)
109 :    
110 :     When this flag is on, parallel copies will be indented an extra level.
111 :    
112 :     ----------------------------------------------------------------------
113 :     Name: Allen Leung
114 : leunga 593 Date: 2000/04/04 03:18:00 EST
115 :     Tag: leunga-20000404-C--Moby
116 :     Description:
117 :    
118 :     All of these fixes are related to C--, Moby, and my own optimization
119 :     stuff; so they shouldn't affect SML/NJ.
120 :    
121 :     1. X86
122 :    
123 :     Various fixes related floating point, and extensions.
124 :    
125 :     2. Alpha
126 :    
127 :     Some extra patterns related to loads with signed/zero extension
128 :     provided by Fermin.
129 :    
130 :     3. Assembly
131 :    
132 :     When generating assemby, resolve the value of client defined constants,
133 :     instead of generating symbolic values. This is controlled by the
134 :     new flag "asm-resolve-constants", which is default to true.
135 :    
136 :     4. Machine Descriptions
137 :    
138 :     a. The precedence parser was slightly broken when parsing infixr symbols.
139 :     b. The type generalizing code had the bound variables reversed, resulting
140 :     in a problem during arity raising.
141 :     c. Various fixes in machine descriptions.
142 :    
143 :     ----------------------------------------------------------------------
144 : blume 592 Name: Matthias Blume
145 :     Date: 2000/04/03 16:05:00 JST
146 :     Tag: blume_main_v110p26p2_2
147 :     Description:
148 :    
149 :     I eliminated coreEnv from compInfo. Access to the "Core" structure is
150 :     now done via the ordinary static environment that is context to each
151 :     compilation unit.
152 :    
153 :     To this end, I arranged that instead of "structure Core" as "structure
154 :     _Core" is bound in the pervasive environment. Core access is done via
155 :     _Core (which can never be accidentially rebound because _Core is not a
156 :     legal surface-syntax symbol).
157 :    
158 :     The current solution is much cleaner because the core environment is
159 :     now simply part of the pervasive environment which is part of every
160 :     compilation unit's context anyway. In particular, this eliminates all
161 :     special-case handling that was necessary until now in order to deal
162 :     with dynamic and symbolic parts of the core environment.
163 :    
164 :     Remaining hackery (to bind the "magic" symbol _Core) is localized in the
165 :     compilation mananger's bootstrap compiler (actually: in the "init group"
166 :     handling). See the comments in src/system/smlnj/init/init.cmi for
167 :     more details.
168 :    
169 :     I also tried to track down all mentions of "Core" (as string argument
170 :     to Symbol.strSymbol) in the compiler and replaced them with a
171 :     reference to the new CoreSym.coreSym. Seems cleaner since the actual
172 :     name appears in one place only.
173 :    
174 :     Binfile and bootfile format have not changed, but the switchover from
175 :     the old "init.cmi" to the new one is a bit tricky, so I supplied new
176 :     bootfiles anyway.
177 :    
178 :     ----------------------------------------------------------------------
179 : leunga 591 Name: Allen Leung
180 :     Date: 2000/04/02 21:17:00 EST
181 :     Tag: leunga-20000402-mltree
182 :     Description:
183 : leunga 585
184 : leunga 591 1. Renamed the constructor CALL in MLTREE by popular demand.
185 :     2. Added a bunch of files from my repository. These are currently
186 :     used by other non-SMLNJ backends.
187 :    
188 : leunga 576 ----------------------------------------------------------------------
189 : leunga 590 Name: Allen Leung
190 :     Date: 2000/03/31 21:15:00 EST
191 :     Tag: leunga-20000331-aliasing
192 :     Description:
193 :    
194 :     This update contains a rewritten (and hopefully more correct) module
195 :     for extracting aliasing information from CPS.
196 :    
197 :     To turn on this feature:
198 :    
199 :     Compiler.Control.CG.memDisambiguate := true
200 :    
201 :     To pretty print the region information with assembly
202 :    
203 :     Compiler.Control.MLRISC.getFlag "asm-show-region" := true;
204 :    
205 :     To control how many levels of aliasing information are printed, use:
206 :    
207 :     Compiler.Control.MLRISC.getInt "points-to-show-level" := n
208 :    
209 :     The default of n is 3.
210 :    
211 :     ----------------------------------------------------------------------
212 : dbm 589 Name: David MacQueen
213 :     Date: 2000/03/31 11:15:00 EST
214 :     Tag: dbm-20000331-runtime_fix
215 :     Description:
216 :    
217 :     This update contains:
218 :    
219 :     1. runtime/c-lib/c-libraries.c
220 :     includes added in revision 1.2 caused compilation errors on hppa-hpux
221 :    
222 :     2. fix for bug 1556
223 :     system/Basis/Implementation/NJ/internal-signals.sml
224 :    
225 :     ----------------------------------------------------------------------
226 : blume 587 Name: Matthias Blume
227 : blume 588 Date: 2000/03/31 18:00:00 JST
228 :     Tag: blume_main_v110p26p2_1
229 :     Description:
230 :    
231 :     This update contains:
232 :    
233 :     1. A small change to CM's handling of stable libraries:
234 :     CM now maintains one "global" modmap that is used for all stable
235 :     libraries. The use of such a global modmap maximizes sharing and
236 :     minimizes the need for re-traversing parts of environments during
237 :     modmap construction. (However, this has minor impact since modmap
238 :     construction seems to account for just one percent or less of total
239 :     compile time.)
240 :    
241 :     2. I added a "genmap" phase to the statistics. This is where I got the
242 :     "one percent" number (see above).
243 :    
244 :     3. CM's new tool parameter mechanism just became _even_ better. :)
245 :     - The parser understands named parameters and recursive options.
246 :     - The "make" and "shell" tools use these new features.
247 :     (This makes it a lot easier to cascade these tools.)
248 :     - There is a small syntax change: named parameters use a
249 :    
250 :     <name> : ( <option> ... ) or
251 :     <name> : <string>
252 :    
253 :     syntax. Previously, named parameters were implemented in an
254 :     ad-hoc fashion by each tool individually (by parsing strings)
255 :     and had the form
256 :    
257 :     <name>=<string>
258 :    
259 :     See the CM manual for a full description of these issues.
260 :    
261 :     ----------------------------------------------------------------------
262 :     Name: Matthias Blume
263 : blume 587 Date: 2000/03/30 18:00:00 JST
264 :     Tag: blume_main_v110p26p2_0
265 :     Description:
266 :    
267 :     !!!!! WARNING !!!!!!
268 :     !! New binfiles !!
269 :     !!!!!!!!!!!!!!!!!!!!
270 :    
271 :     This update contains:
272 :    
273 :     1. Moderate changes to CM:
274 :    
275 :     - Changes to CM's tools mechanism. In particular, it is now possible
276 :     to have tools that accept additional "command line" parameters
277 :     (specified in the .cm file at each instance where the tool's class is
278 :     used).
279 :    
280 :     This was done to accomodate the new "make" and "shell" tools which
281 :     facilitate fairly seemless hookup to portions of code managed using
282 :     Makefiles or Shell scripts.
283 :    
284 :     There are no classes "shared" or "private" anymore. Instead, the
285 :     sharing annotation is now a parameter to the "sml" class.
286 :    
287 :     There is a bit of generic machinery for implementing one's own
288 :     tools that accept command-line parameters. However, I am not yet fully
289 :     satisfied with that part, so expect changes here in the future.
290 :    
291 :     All existing tools are described in the CM manual.
292 :    
293 :     - Slightly better error handling. (CM now surpresses many followup
294 :     error messages that tended to be more annoying than helpful.)
295 :    
296 :     2. Major changes to the compiler's static environment data structures.
297 :    
298 :     - no CMStaticEnv anymore.
299 :     - no CMEnv, no "BareEnvironment" (actually, _only_ BareEnvironment,
300 :     but it is called Environment), no conversions between different
301 :     kinds of static environments
302 :    
303 :     - There is still a notion of a "modmap", but such modmaps are generated
304 :     on demand at the time when they are needed. This sounds slow, but I
305 :     sped up the code that generates modmaps enough for this not to lead to
306 :     a slowdown of the compiler (at least I didn't detect any).
307 :    
308 :     - To facilitate rapid modmap generation, static environments now
309 :     contain an (optional) "modtree" structure. Modtree annotations are
310 :     constructed by the unpickler during unpickling. (This means that
311 :     the elaborator does not have to worry about modtrees at all.)
312 :     Modtrees have the advantage that they are compositional in the same
313 :     way as the environment data structure itself is compositional.
314 :     As a result, modtrees never hang on to parts of an environment that
315 :     has already been rendered "stale" by filtering or rebinding.
316 :    
317 :     - I went through many, many trials and errors before arriving at the
318 :     current solution. (The initial idea of "linkpaths" did not work.)
319 :     But the result of all this is that I have touched a lot of files that
320 :     depend on the "modules" and "types" data structures (most of the
321 :     elaborator). There were a lot of changes during my "linkpath" trials
322 :     that could have been reverted to their original state but weren't.
323 :     Please, don't be too harsh on me for messing with this code a bit more
324 :     than what was strictly necessary... (I _did_ resist the tempation
325 :     of doing any "global reformatting" to avoid an untimely death at
326 :     Dave's hands. :)
327 :    
328 :     - One positive aspect of the previous point: At least I made sure that
329 :     all files that I touched now compile without warnings (other than
330 :     "polyEqual").
331 :    
332 :     - compiler now tends to run "leaner" (i.e., ties up less memory in
333 :     redundant modmaps)
334 :    
335 :     ----------------------------------------------------------------------
336 : leunga 580 Name: Allen Leung
337 : leunga 585 Date: 2000/03/29 18:00:00
338 :     Tag: leunga-20000327-mlriscGen_hppa_alpha_x86
339 : leunga 586 Boot files (optional): ftp://react-ilp.cs.nyu.edu/leunga/110.26.1-sml.boot.x86-unix-20000330.tar.gz
340 : leunga 585 Description:
341 :    
342 :     This update contains *MAJOR* changes to the way code is generated from CPS
343 :     in the module mlriscGen, and in various backend modules.
344 :    
345 :     CHANGES
346 :     =======
347 :    
348 :     1. MLRiscGen: forward propagation fix.
349 :    
350 :     There was a bug in forward propagation introduced at about the same time
351 :     as the MLRISC x86 backend, which prohibits coalescing to be
352 :     performed effectively in loops.
353 :    
354 :     Effect: speed up of loops in RISC architectures.
355 :     By itself, this actually slowed down certain benchmarks on the x86.
356 :    
357 :     2. MLRiscGen: forward propagating addresses from consing.
358 :    
359 :     I've changed the way consing code is generated. Basically I separated
360 :     out the initialization part:
361 :    
362 :     store tag, offset(allocptr)
363 :     store elem1, offset+4(allocptr)
364 :     store elem2, offset+8(allocptr)
365 :     ...
366 :     store elemn, offset+4n(allocptr)
367 :    
368 :     and the address computation part:
369 :    
370 :     celladdr <- offset+4+alloctpr
371 :    
372 :     and move the address computation part
373 :    
374 :     Effect: register pressure is generally lower as a result. This
375 :     makes compilation of certain expressions much faster, such as
376 :     long lists with non-trivial elements.
377 :    
378 :     [(0,0), (0,0), .... (0,0)]
379 :    
380 :     3. MLRiscGen: base pointer elimination.
381 :    
382 :     As part of the linkage mechanism, we generate the sequence:
383 :    
384 :     L: ... <- start of the code fragment
385 :    
386 :     L1:
387 :     base pointer <- linkreg - L1 + L
388 :    
389 :     The base pointer was then used for computing relocatable addresses
390 :     in the code fragment. Frequently (such as in lots of continuations)
391 :     this is not needed. We now eliminate this sequence whenever possible.
392 :    
393 :     For compile time efficiency, I'm using a very stupid local heuristic.
394 :     But in general, this should be done as a control flow analysis.
395 :    
396 :     Effect: Smaller code size. Speed up of most programs.
397 :    
398 :     4. Hppa back end
399 :    
400 :     Long jumps in span dependence resolution used to depend on the existence
401 :     of the base pointer.
402 :    
403 :     A jump to a long label L was expanded into the following sequence:
404 :    
405 :     LDIL %hi(L-8192), %r29
406 :     LDO %lo(L-8192)(%r29), %r29
407 :     ADD %r29, baseptr, %r29
408 :     BV,n %r0(%r29)
409 :    
410 :     In the presence of change (3) above, this will not work. I've changed
411 :     it so that the following sequence of instructions are generated, which
412 :     doesn't mention the base pointer at all:
413 :    
414 :     BL,n L', %r29 /* branch and link, L' + 4 -> %r29 */
415 :     L': ADDIL L-(L'+4), %r29 /* Compute address of L */
416 :     BV,n %r0(%r29) /* Jump */
417 :    
418 :     5. Alpha back end
419 :    
420 :     New alpha instructions LDB/LDW have been added, as per Fermin's
421 :     suggestions. This is unrelated to all other changes.
422 :    
423 :     6. X86 back end
424 :    
425 :     I've changed andl to testl in the floating point test sequence
426 :     whenever appropriate. The Intel optimization guide states that
427 :     testl is perferable to andl.
428 :    
429 :     7. RA (x86 only)
430 :    
431 :     I've improved the spill propagation algorithm, using an approximation
432 :     of maximal weighted independent sets. This seems to be necessary to
433 :     alleviate the negative effect in light of the slow down in (1).
434 :    
435 :     I'll write down the algorithm one of these days.
436 :    
437 :     8. MLRiscGen: frequencies
438 :    
439 :     I've added an annotation that states that all call gc blocks have zero
440 :     execution frequencies. This improves register allocation on the x86.
441 :    
442 :     BENCHMARKS
443 :     ==========
444 :    
445 :     I've only perform the comparison on 110.25.
446 :    
447 :     The platforms are:
448 :    
449 :     HPPA A four processor HP machine (E9000) with 5G of memory.
450 :     X86 A 300Hhz Pentium II with 128M of memory, and
451 :     SPARC An Ultra sparc 2 with 512M of memory.
452 :    
453 :     I used the following parameters for the SML benchmarks:
454 :    
455 :     @SMLalloc
456 :     HPPA 256k
457 :     SPARC 512k
458 :     X86 256k
459 :    
460 :     COMPILATION TIME
461 :     ----------------
462 :     Here are the numbers comparing the compilation times of the compilers.
463 :     I've only compared 110.25 compiling the new sources versus
464 :     a fixpoint version of the new compiler compiling the same.
465 :    
466 :     110.25 New
467 :     Total Time in RA Spill+Reload Total Time In RA Spill+Reload
468 :     HPPA 627s 116s 2684+3584 599s 95s 1003+1879
469 :     SPARC 892s 173s 2891+3870 708s 116s 1004+1880
470 :     X86 999s 315s 94006+130691 987s 296s 108877+141957
471 :    
472 :     110.25 New
473 :     Code Size Code Size
474 :     HPPA 8596736 8561421
475 :     SPARC 8974299 8785143
476 :     X86 9029180 8716783
477 :    
478 :     So in summary, things are at least as good as before. Dramatic
479 :     reduction in compilation is obtained on the Sparc; I can't explain it,
480 :     but it is reproducible. Perhaps someone should try to reproduce this
481 :     on their own machines.
482 :    
483 :     SML BENCHMARKS
484 :     --------------
485 :    
486 :     On the average, all benchmarks perform at least as well as before.
487 :    
488 :     HPPA Compilation Time Spill+Reload Run Time
489 :     110.25 New 110.25 New 110.25 New
490 :    
491 :     barnesHut 3.158 3.015 4.75% 1+1 0+0 2.980 2.922 2.00%
492 :     boyer 6.152 5.708 7.77% 0+0 0+0 0.218 0.213 2.34%
493 :     count-graphs 1.168 1.120 4.32% 0+0 0+0 22.705 23.073 -1.60%
494 :     fft 0.877 0.792 10.74% 1+3 1+3 0.602 0.587 2.56%
495 :     knuthBendix 3.180 2.857 11.32% 0+0 0+0 0.675 0.662 2.02%
496 :     lexgen 6.190 5.290 17.01% 0+0 0+0 0.913 0.788 15.86%
497 :     life 0.803 0.703 14.22% 25+25 0+0 0.153 0.140 9.52%
498 :     logic 2.048 2.007 2.08% 6+6 1+1 4.133 4.008 3.12%
499 :     mandelbrot 0.077 0.080 -4.17% 0+0 0+0 0.765 0.712 7.49%
500 :     mlyacc 22.932 20.937 9.53% 154+181 32+57 0.468 0.430 8.91%
501 :     nucleic 5.183 5.060 2.44% 2+2 0+0 0.125 0.120 4.17%
502 :     ratio-regions 3.357 3.142 6.84% 0+0 0+0 116.225 113.173 2.70%
503 :     ray 1.283 1.290 -0.52% 0+0 0+0 2.887 2.855 1.11%
504 :     simple 6.307 6.032 4.56% 28+30 5+7 3.705 3.658 1.28%
505 :     tsp 0.888 0.862 3.09% 0+0 0+0 7.040 6.893 2.13%
506 :     vliw 24.378 23.455 3.94% 106+127 25+45 2.758 2.707 1.91%
507 :     --------------------------------------------------------------------------
508 :     Average 6.12% 4.09%
509 :    
510 :     SPARC Compilation Time Spill+Reload Run Time
511 :     110.25 New 110.25 New 110.25 New
512 :    
513 :     barnesHut 3.778 3.592 5.20% 2+2 0+0 3.648 3.453 5.65%
514 :     boyer 6.632 6.110 8.54% 0+0 0+0 0.258 0.242 6.90%
515 :     count-graphs 1.435 1.325 8.30% 0+0 0+0 33.672 34.737 -3.07%
516 :     fft 0.980 0.940 4.26% 3+9 2+6 0.838 0.827 1.41%
517 :     knuthBendix 3.590 3.138 14.39% 0+0 0+0 0.962 0.967 -0.52%
518 :     lexgen 6.593 6.072 8.59% 1+1 0+0 1.077 1.078 -0.15%
519 :     life 0.972 0.868 11.90% 26+26 0+0 0.143 0.140 2.38%
520 :     logic 2.525 2.387 5.80% 7+7 1+1 5.625 5.158 9.05%
521 :     mandelbrot 0.090 0.093 -3.57% 0+0 0+0 0.855 0.728 17.39%
522 :     mlyacc 26.732 23.827 12.19% 162+189 32+57 0.550 0.560 -1.79%
523 :     nucleic 6.233 6.197 0.59% 3+3 0+0 0.163 0.173 -5.77%
524 :     ratio-regions 3.780 3.507 7.79% 0+0 0+0 133.993 131.035 2.26%
525 :     ray 1.595 1.550 2.90% 1+1 0+0 3.440 3.418 0.63%
526 :     simple 6.972 6.487 7.48% 29+32 5+7 3.523 3.525 -0.05%
527 :     tsp 1.115 1.063 4.86% 0+0 0+0 7.393 7.265 1.77%
528 :     vliw 27.765 24.818 11.87% 110+135 25+45 2.265 2.135 6.09%
529 :     ----------------------------------------------------------------------------
530 :     Average 6.94% 2.64%
531 :    
532 :     X86 Compilation Time Spill+Reload Run Time
533 :     110.25 New 110.25 New 110.25 New
534 :    
535 :     barnesHut 5.530 5.420 2.03% 593+893 597+915 3.532 3.440 2.66%
536 :     boyer 8.768 7.747 13.19% 493+199 301+289 0.327 0.297 10.11%
537 :     count-graphs 2.040 2.010 1.49% 298+394 315+457 26.578 28.660 -7.26%
538 :     fft 1.327 1.302 1.92% 112+209 115+210 1.055 0.962 9.71%
539 :     knuthBendix 5.218 5.475 -4.69% 451+598 510+650 0.928 0.932 -0.36%
540 :     lexgen 9.970 9.623 3.60% 1014+841 1157+885 0.947 0.928 1.97%
541 :     life 1.183 1.183 0.00% 162+182 145+148 0.127 0.103 22.58%
542 :     logic 3.285 3.512 -6.45% 514+684 591+836 5.682 5.577 1.88%
543 :     mandelbrot 0.147 0.143 2.33% 38+41 33+54 0.703 0.690 1.93%
544 :     mlyacc 35.457 32.763 8.22% 3496+4564 3611+4860 0.552 0.550 0.30%
545 :     nucleic 7.100 6.888 3.07% 239+168 201+158 0.175 0.173 0.96%
546 :     ratio-regions 6.388 6.843 -6.65% 1182+257 981+300 120.142 120.345 -0.17%
547 :     ray 2.332 2.338 -0.29% 346+398 402+494 3.593 3.540 1.51%
548 :     simple 9.912 9.903 0.08% 1475+941 1579+1168 3.057 3.178 -3.83%
549 :     tsp 1.623 1.532 5.98% 266+200 250+211 8.045 7.878 2.12%
550 :     vliw 33.947 35.470 -4.29% 2629+2774 2877+3171 2.072 1.890 9.61%
551 :     ----------------------------------------------------------------------------
552 :     Average 1.22% 3.36%
553 :    
554 :     ----------------------------------------------------------------------
555 :     Name: Allen Leung
556 : leunga 583 Date: 2000/03/23 16:25:00
557 :     Tag: leunga-20000323-fix_x86_alpha
558 :     Description:
559 :    
560 :     1. X86 fixes/changes
561 :    
562 :     a. The old code generated for SETcc was completely wrong.
563 :     The Intel optimization guide is VERY misleading.
564 :    
565 :     2. ALPHA fixes/changes
566 :    
567 :     a. Added the instructions LDBU, LDWU, STB, STW as per Fermin's suggestion.
568 :     b. Added a new mode byteWordLoadStores to the functor parameter to Alpha()
569 :     c. Added reassociation code for address computation.
570 :    
571 :     ----------------------------------------------------------------------
572 :     Name: Allen Leung
573 : leunga 580 Date: 2000/03/22 01:23:00
574 :     Tag: leunga-20000322-fix_x86_hppa_ra
575 :     Description:
576 :    
577 :     1. X86 fixes/changes
578 :    
579 :     a. x86Rewrite bug with MUL3 (found by Lal)
580 :     b. Added the instructions FSTS, FSTL
581 :    
582 :     2. PA-RISC fixes/changes
583 :    
584 :     a. B label should not be a delay slot candidate! Why did this work?
585 :     b. ADDT(32, REG(32, r), LI n) now generates one instruction instead of two,
586 :     as it should be.
587 :     c. The assembly syntax for fstds and fstdd was wrong.
588 :     d. Added the composite instruction COMICLR/LDO, which is the immediate
589 :     operand variant of COMCLR/LDO.
590 :    
591 :     3. Generic MLRISC
592 :    
593 :     a. shuffle.sml rewritten to be slightly more efficient
594 :     b. DIV bug in mltree-simplify fixed (found by Fermin)
595 :    
596 :     4. Register Allocator
597 :    
598 :     a. I now release the interference graph earlier during spilling.
599 :     May improve memory usage.
600 :    
601 :     ----------------------------------------------------------------------
602 : blume 577 Name: Matthias Blume
603 : blume 578 Date: 2000/03/14 14:15:32
604 :     Tag: blume_main_v110p26p1_2
605 :     Description:
606 :    
607 :     1. Tools.registerStdShellCmdTool (from smlnj/cm/tool.cm) takes an
608 :     additional argument called "template" which is an optional string that
609 :     specifiel the layout of the tool command line. See the CM manual for
610 :     explanation.
611 :    
612 :     2. A special-purpose tool can be "regisitered" by simply dropping the
613 :     corresponding <...>-tool.cm (and/or <...>-ext.cm) into the same
614 :     directory where the .cm file lives that uses this tool. (The
615 :     behavior/misfeature until now was to look for the tool description
616 :     files in the current working directory.) As before, tool description
617 :     files could also be anchored -- in which case they can live anywhere
618 :     they like. Following the recent e-mail discussion, this change should
619 :     make it easier to have special-purpose tools that are shipped together
620 :     with the sources of the program that uses them.
621 :    
622 :     ----------------------------------------------------------------------
623 :     Name: Matthias Blume
624 : blume 577 Date: 2000/03/10 07:48:34
625 :     Tag: blume_main_v110p26p1_1
626 :     Description:
627 :    
628 :     I added a re-written version of Dave's fixpt script to src/system.
629 :     Changes relative to the original version:
630 :     - sh-ified (not everybody has ksh)
631 :     - automatically figures out which architecture it runs on
632 :     - uses ./makeml a bit more cleverly
633 :     - never invokes ./installml (and, thus, does not clobber your
634 :     good and working installation of sml in case something goes wrong)
635 :     - accepts max iteration count using option "-iter <n>"
636 :     - accepts a "base" name using option "-base <base>"
637 :    
638 :     It does not build any extraneous heap images but directly rebuilds
639 :     bin- and boot-hierarchies using makeml's "-rebuild" switch. Finally,
640 :     it can incorporate existing bin- and boot- hierarchies. For example,
641 :     suppose the base is set to "sml" (which is the default). Then it
642 :     successively builds
643 :    
644 :     sml.bin.<arch>-unix and sml.boot.<arch>-unix
645 :     then sml1.bin.<arch>-unix and sml1.boot.<arch>-unix
646 :     then sml2.bin.<arch>-unix and sml2.boot.<arch>-unix
647 :     ...
648 :     then sml<n>.bin.<arch>-unix and sml<n>.boot.<arch>-unix
649 :    
650 :     and so on. If any of these already exist, it will just use what's
651 :     there. In particular, many people will have the initial set of bin
652 :     and boot files around, so this saves time for at least one full
653 :     rebuild. Having sets of the form <base><k>.{bin,boot}.<arch>-unix for
654 :     <k>=1,2,... is normally not a good idea when invoking fixpt. However,
655 :     they might be the result of an earlier partial run of fixpt (which
656 :     perhaps got accidentially killed). In this case, fixpt will quickly
657 :     move through what exists before continuing where it left off earlier,
658 :     and, thus, saves a lot of time.
659 :    
660 :     ----------------------------------------------------------------------
661 : leunga 576 Name: Allen Leung
662 :     Date: 00/03/10 02:20:00
663 :     Tag: leunga-20000310-fix_x86_asm_ra
664 :     Description:
665 : dbm 570
666 : leunga 576 More assembly output problems involving the indexed addressing mode
667 :     on the x86 have been found and corrected. Thanks to Fermin Reig for the
668 :     fix.
669 :    
670 :     The interface and implementation of the register allocator have been changed
671 :     slightly to accommodate the possibility to skip the register allocation
672 :     phases completely and go directly to memory allocation. This is needed
673 :     for C-- use.
674 :    
675 : dbm 570 ----------------------------------------------------------------------
676 : blume 572 Name: Matthias Blume
677 : blume 575 Date: 00/03/09 10:23:53
678 :     Tag: blume_main_v110p26p1_0
679 :     Description:
680 :    
681 :     * Complete re-organization of library names. Many libraries have been
682 :     consolidated so that they share the same path anchor. For example,
683 :     all MLRISC-related libraries are anchored at MLRISC, most libraries that
684 :     are SML/NJ-specific are under "smlnj". Notice that names like
685 :     host-cmb.cm or host-compiler.cm no longer exist. See system/README
686 :     for a complete description of the new naming scheme. Quick reference:
687 :    
688 :     host-cmb.cm -> smlnj/cmb.cm
689 :     host-compiler.cm -> smlnj/compiler.cm
690 :     full-cm.cm -> smlnj/cm.cm
691 :     <arch>-<os>.cm -> smlnj/cmb/<arch>-<os>.cm
692 :     <arch>-compiler.cm -> smlnj/compiler/<arch>.cm
693 :    
694 :     * Bug fixes in CM.
695 :     - exceptions in user code are being passed through (i.e., reach top level)
696 :     - more bugs in paranoia mode fixed
697 :     - bug related to checking group owners fixed
698 :    
699 :     * New install.sh script that automagically fetches archive files:
700 :     The new file config/srcarchiveurl must contain the URL of the
701 :     (remote) directory that contains bin files (or other source archives).
702 :     If install.sh does not find the archive locally, it tries to get
703 :     it from that remote directory.
704 :     This should simplify installation further: For machines that have
705 :     access to the internet, just fetch <version>-config.tgz, unpack it,
706 :     edit config/targets, and go (run config/install.sh). The scipt will
707 :     fetch everything else that it might need all by itself.
708 :    
709 :     For CVS users, this mechanism is not relevant for source archives, but
710 :     it is convenient for getting new sets of binfiles.
711 :    
712 :     Archives should be tar files compressed with either gzip, compress, or
713 :     bzip2. The script recognizes .tgz, .tar, tar.gz, tz, .tar.Z, and .tar.bz2.
714 :    
715 :     ----------------------------------------------------------------------
716 :     Name: Matthias Blume
717 : blume 572 Date: 2000/03/07 04:01:04
718 :     Tag: blume_main_v110_26_2
719 : dbm 570 Description:
720 : blume 572 - size info in BOOTLIST
721 :     * no fixed upper limits for number of bootfiles or length of
722 :     bootfile names in runtime
723 :     * falling back to old behavior if no BOOTLIST size info found
724 :     - allocation size heuristics in .run-sml
725 :     * tries to read cache size from /proc/cpuinfo (this is important for
726 :     small-cache Celeron systems!)
727 :     - install.sh robustified
728 :     - CM manual updates
729 :     - paranoid mode
730 :     * no more CMB.deliver() (i.e., all done by CMB.make())
731 :     * can re-use existing sml.boot.* files
732 :     * init.cmi now treated as library
733 :     * library stamps for consistency checks
734 :     - sml.boot.<arch>-<os>/PIDMAP file
735 :     * This file is read by the CM startup code. This is used to minimize
736 :     the amount of dynamic state that needs to be stowed away for the
737 :     purpose of sharing between interactive system and user code.
738 :     - CM.Anchor.anchor instead of CM.Anchor.{set,cancel}
739 :     * Upon request by Elsa. Anchors now controlled by get-set-pair
740 :     like most other CM state variables.
741 :     - Compiler.CMSA eliminated
742 :     * No longer supported by CM anyway.
743 :     - fixed bugs in pickler that kept biting Stefan
744 :     * past refs to past refs (was caused by the possibility that
745 :     ad-hoc sharing is more discriminating than hash-cons sharing)
746 :     * integer overflow on LargeInt.minInt
747 :     - ml-{lex,yacc} build scripts now use new mechanism
748 :     for building standalone programs
749 :     - fixed several gcc -Wall warnings that were caused by missing header
750 :     files, missing initializations, etc., in runtime (not all warnings
751 :     eliminated, though)

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