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/branches/primop-branch-3/NOTES/HISTORY
ViewVC logotype

Annotation of /sml/branches/primop-branch-3/NOTES/HISTORY

Parent Directory Parent Directory | Revision Log Revision Log


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

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