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 1391 - (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 : mblume 1384 Name:
11 : george 1003 Date: yyyy/mm/dd
12 : dbm 570 Tag: <post-commit CVS tag>
13 :     Description:
14 : leunga 1142
15 : george 1136 ----------------------------------------------------------------------
16 : mblume 1384 Name: Matthias Blume (blume (at) tti - c (dot) org)
17 : mblume 1391 Date: 2003/09/24 16:31:00 CDT
18 :     Tag: blume-20030924-installer
19 :     Description:
20 :    
21 :     Another installer tweak: All the ML code for the installer is now
22 :     compiled during CMB.make and put into a little library called
23 :     $smlnj/installer.cm. The installation then simply invokes
24 :    
25 :     sml -m $smlnj/installer.cm
26 :    
27 :     and everything happens automagically.
28 :    
29 :     Win32: ML code senses value of environment variable SMLNJ_HOME.
30 :     Unix: ML code senses values of environment variables ROOT, CONFIGDIR,
31 :     and BINDIR.
32 :    
33 :     The new scheme guarantees that the ML code responsible for the installation
34 :     is in sync with the APIs of the main system. Also, the installer is
35 :     somewhat faster because the installer script is precompiled.
36 :    
37 :     ----------------------------------------------------------------------
38 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
39 : mblume 1390 Date: 2003/09/24 15:35:00 CDT
40 :     Tag: blume-20030924-synsock
41 :     Description:
42 :    
43 :     Added a signature SYNCHRONOUS_SOCKET to basis.cm. This is like SOCKET
44 :     but excludes all non-blocking operations. Defined SOCKET (in Basis)
45 :     and CML_SOCKET in terms of SYNCHRONOUS_SOCKET. Removed superfluous
46 :     implementations of non-blocking operations from CML's Socket
47 :     structure.
48 :    
49 :     ----------------------------------------------------------------------
50 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
51 : mblume 1389 Date: 2003/09/24 15:10:05 CDT
52 :     Tag: blume-20030924-sockets
53 :     Description:
54 :    
55 :     1. Fixed SOCKET API and implementation to match Basis spec.
56 :     This required changing the internal representation of sockets to one
57 :     that remembers (for each socket file descriptor) whether it is currently
58 :     blocking or non-blocking. This state is maintained lazily (i.e., a system
59 :     call is made only if the state actually needs to change).
60 :    
61 :     2. OS-specific details of sockets were moved into separate files, thus
62 :     making it possible to unify the bulk of the socket implementations
63 :     between Unix and Win32.
64 :    
65 :     3. CML's socket API changed accordingly.
66 :     (Note that we need to remove non-blocking functions from this API
67 :     since they are redundant in the case of CML!)
68 :    
69 :     4. CML's socket implementation now makes use of non-blocking functions
70 :     provided by Basis, thus removing all OS-dependent code from this part
71 :     of CML.
72 :    
73 :     5. Changed Real64.precision from 52 to 53. Minor cleanup in Real64 code.
74 :    
75 :     ----------------------------------------------------------------------
76 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
77 : mblume 1387 Date: 2003/09/22 12:10:00 CDT
78 : mblume 1388 Tag: blume-20030922-110_43_2
79 : mblume 1387 Description:
80 :    
81 :     Made a new interim version and bootfiles for developer's bootstrapping
82 :     convenience.
83 :    
84 :     110.43.2 -- NEW BOOTFILES
85 :    
86 :     ----------------------------------------------------------------------
87 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
88 : mblume 1385 Date: 2003/09/19 15:55:00 CDT
89 :     Tag: blume-20030919-cmdir
90 :     Description:
91 :    
92 :     1. new-install.sh -> install.sh
93 :     2. changed default CM "metadata" directory name to ".cm" (instead of "CM")
94 :     3. tweaked installer so that another name instead of .cm can be chosen
95 :     at install time (by setting the CM_DIR_ARC environment variable
96 :     during installation); once installation is complete, the name is
97 :     fixed
98 :    
99 :     ----------------------------------------------------------------------
100 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
101 : mblume 1384 Date: 2003/09/18 16:00:00 CDT
102 :     Tag: blume-20030918-110_43_1
103 :     Description:
104 :    
105 :     Made a new interim version and bootfiles for developer's bootstrapping
106 :     convenience.
107 :    
108 :     110.43.1 -- NEW BOOTFILES
109 :    
110 :     ----------------------------------------------------------------------
111 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
112 : mblume 1383 Date: 2003/09/18 15:20:00 CDT
113 :     Tag: blume-20030918-misc
114 :     Description:
115 :    
116 :     1. Exported fractionsPerSecond etc. from TimeImp (but not from Time as
117 :     this seems to be controversial at the moment) and used those in
118 :     Posix.ProcEnv.times.
119 :    
120 :     2. Added Time.{from,to}Nanoseconds to Time.
121 :    
122 :     3. Improved Real.{from,to}LargeInt by avoiding needless calculations.
123 :     For example, fromLargeInt never needs to look at more than 3 "big
124 :     digits" to get its 53 bits of precision.
125 :    
126 :     ----------------------------------------------------------------------
127 : mblume 1384 Name: Matthias Blume (blume (at) tti - c (dot) org)
128 : mblume 1381 Date: 2003/09/17 16:30:00 CDT
129 :     Tag: blume-20030917-real32-slices
130 :     Description:
131 :    
132 :     Added an entry to the primitive environment
133 :     (compiler/Semant/statenv/prim.sml) for int32->real64 conversion and
134 :     added code to compiler/CodeGen/main/mlriscGen.sml to implement it.
135 :    
136 :     Removed some of the "magic" constants in real64.sml and replaced them
137 :     with code that generates these values from their corresponding
138 :     integer counterparts.
139 :    
140 :     Made all(?) the slice-related changes to the Basis and made everything
141 :     compile again...
142 :    
143 :     ----------------------------------------------------------------------
144 : mblume 1368 Name: Matthias Blume (blume (at) tti - c (dot) org)
145 : mblume 1375 Date: 2003/09/15 17:45:00 CDT
146 :     Tag: blume-20030915-rbase
147 :     Description:
148 :    
149 :     Fixed bug in Real.fromLargeInt.
150 :    
151 :     ----------------------------------------------------------------------
152 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
153 : mblume 1369 Date: 2003/09/13 18:11:00 CDT
154 :     Tag: blume-20030913-libinstall
155 :     Description:
156 :    
157 :     Minor bugfix in config/libinstall (set anchor with path to
158 :     standalone tool after installing it, otherwise libraries that
159 :     need ml-lex or ml-yacc won't compile the first time the installer
160 :     runs).
161 :    
162 :     ----------------------------------------------------------------------
163 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
164 : mblume 1368 Date: 2003/09/12 11:45:00 CDT
165 :     Tag: blume-20030912-various
166 :     Description:
167 :    
168 :     - fixed bug in Real.toLargeInt
169 :     - fixed bug in Posix.ProcEnv.times
170 :     - changed inputLine functions to return an option
171 :     - minor installer improvements / bugfixes
172 :     - changed default @SMLalloc parameter for x86/celeron to 64k
173 :    
174 :     ----------------------------------------------------------------------
175 :     Name: Matthias Blume (blume (at) tti - c (dot) org)
176 : mblume 1356 Date: 2003/09/09 22:00:00 CDT
177 :     Tag: Release_110_43
178 :     Description:
179 :    
180 :     New working release 110.43. New bootfiles.
181 :    
182 :     ----------------------------------------------------------------------
183 : mblume 1368 Name: Matthias Blume (blume (at) tti - c (dot) org)
184 : mblume 1354 Date: 2003/09/09 19:20:00 CDT
185 :     Tag: blume-20030909-installer
186 :     Description:
187 :    
188 :     Rewrote large parts of config/install.sh in SML (config/libinstall.sml).
189 :     Modified config/install.bat to take advantage of it. Also modified
190 :     config/install.sh (and called it config/new-install.sh) to take advantage
191 :     of it on Unix systems. (The SML code is (supposed to be) platform-
192 :     independent.)
193 :    
194 :     The installer can now install everything under Win32
195 :     as well as under *nix as long as it compiles.
196 :    
197 :     Other changes:
198 :    
199 :     - made CML compile again under Win32
200 :     - made eXene compile under Win32 (by providing a fake structure UnixSock
201 :     and by using OS.Process.getEnv instead of Posix.ProcEnv.getenv)
202 :     - fixed a bug in nowhere: it assumed that type OS.Process.status is the
203 :     same as type int; under Win32 it isn't
204 :     - fixed some slice-related problems in the win32-specific parts of CML
205 :     - added a functor argument "sameVol" to os-path-fn.sml in the Basis
206 :     (under Win32, the volume name is case-insensitive, and the
207 :     OS.Path code compares volume names for equality)
208 :    
209 :     ----------------------------------------------------------------------
210 : mblume 1368 Name: Matthias Blume (blume (at) tti - c (dot) org)
211 : mblume 1353 Date: 2003/09/08 11:55:00 CDT
212 :     Tag: blume-20030908-fullpath
213 :     Description:
214 :    
215 :     Made Win32 version of OS.FileSys.fullPath return current directory
216 :     when given an empty string. This is what the spec says, and incidentally,
217 :     CM depends on it. (CM otherwise goes into an infinite loop in certain
218 :     cases when presented with the name of a non-existing .cm file.)
219 :    
220 :     ----------------------------------------------------------------------
221 : mblume 1368 Name: Matthias Blume (blume (at) tti - c (dot) org)
222 : mblume 1350 Date: 2003/09/04 16:30:00 CDT
223 : mblume 1352 Tag: blume-20030905-slices-etc
224 : mblume 1350 Description:
225 :    
226 :     1. Changed interface to vectors and arrays in Basis to match
227 :     (draft) Basis spec.
228 :     2. Added signatures and implementations of slices according to
229 :     Basis spec.
230 :     3. Edited source code throughout the system to make it compile again
231 :     under 1. and 2. (In some cases code had to be added to have it
232 :     match the new signatures.)
233 :     4. MLRISC should be backward-compatible: the copies of the originals
234 :     of files that needed to change under 3. were retained, the .cm files
235 :     check the compiler version number and use old versions when
236 :     appropriate.
237 :     5. Changed type of OS.FileSys.readDir and Posix.FileSys.readdir to
238 :     dirstream -> string option (in accordance with Basis spec).
239 :     6. When generating code that counts lines, ml-lex used function
240 :     CharVector.foldli, taking advantage of its old interface.
241 :     This has been replaced with the corresponding code from
242 :     CharVectorSlice. (html-lex must be re-lexed!)
243 :     7. BitArray in smlnj-lib/Util has been extended/modified to match the
244 :     new MONO_ARRAY signature. (Do we need BitArraySlice?)
245 :     8. Removed temporary additions (fromInternal, toInternal) from the
246 :     (now obsolete) IntInf in smlnj-lib/Util.
247 :     9. Cleaned up structure Byte.
248 :     10. Added localOffset, scan, and fromString to Date (according to spec).
249 :     Cleaned/corrected implementation of Date.
250 :     (Still need to check for correctness; implement better canonicalizeDate.)
251 :     11. Added "scan" to signature IEEE_REAL.
252 :     12. Some improvements to IntInf [in particular: efficiency-hack for
253 :     mod and rem when second operand is 2 (for parity checks).]
254 :     13. Changed representation of type Time.time, using a single IntInf.int
255 :     value counting microseconds. This considerably simplified the
256 :     implementation of structure Time. We now support negative time
257 :     values; scan and fromString handle signs.
258 :     14. Functor PrimIO now takes two additional arguments (VectorSlice and
259 :     ArraySlice).
260 :    
261 :     ----------------------------------------------------------------------
262 : mblume 1368 Name: Matthias Blume (blume (at) tti - c (dot) org)
263 : mblume 1347 Date: 2003/08/28 17:00:00 CDT
264 :     Tag: blume-20030828-intinf
265 :     Description:
266 :    
267 :     This is a major update which comes with a version number bump
268 :     (110.42.99 -- yes, we are really close to 110.43 :-), NEW BOOTFILES,
269 :     and an implementation of IntInf in the Basis.
270 :    
271 :     There are a fairly large number of related changes and updates throughout
272 :     the system:
273 :    
274 :     Basis:
275 :     - Implemented IntInf.
276 :     - Made LargeInt a projection of IntInf (by filtering through INTEGER).
277 : mblume 1348 - Added some missing Real64 operations, most notably Real.toLargeInt.
278 : mblume 1347 - Added FixedInt as a synonym for Int32.
279 :    
280 :     compiler:
281 :     * Added support for a built-in intinf type.
282 :     - literals
283 :     - pattern matching
284 :     - conversion shortcuts (Int32.fromLarge o Int.toLarge etc.)
285 :     - overloading on literals and operations
286 :    
287 :     This required adding a primitive type intinf, some additional
288 :     primops, and implementations for several non-trivial intinf
289 :     operations in Core. (The intinf type is completely abstract
290 :     to the compiler; all operations get delegated back to the Core.)
291 :    
292 :     * Intinf equality is handled by polyequal. However, the compiler
293 :     does not print its usual warning in this case (since polyequal
294 :     is the right thing to do there).
295 :    
296 :     * Improved the organization of structure InlineT.
297 :    
298 :     * A word about conversion primops:
299 :     If conversions involving intinf do not cancel out during
300 : mblume 1348 CPS contract, then the compiler must insert calls to Core functions.
301 : mblume 1347 Since all core access must be resolved already during the FLINT
302 :     translate phase, it would be too late a the time of CPS contract
303 : mblume 1348 to add new Core calls. For this reason, conversion primops
304 : mblume 1347 for intinf carry two arguments: 1. the numeric argument that
305 :     they are supposed to convert, and 2. the Core function that
306 :     can help with this conversion if necessary. If CPS contract
307 : mblume 1348 eliminates a primop, then the associated Core function becomes
308 :     dead and goes away. Intinf conversion primops that do not get
309 :     eliminated by CPS contract get rewritten into calls of their
310 :     core functions by a separate, new phase.
311 : mblume 1347
312 :     interactive system:
313 :     - Control.Print.intinfDepth controls max length of intinf constants
314 :     being printed. (Analogous to Control.Print.stringDepth.)
315 :     - Cleanup in printutil and pputil: got rid of unused stuff and
316 :     duplicates; replaced some of the code with code that makes better
317 :     use of library functionality.
318 :    
319 :     CM:
320 :     Bugfix: parse-errors in init group (system/smlnj/init/init.cmi)
321 :     are no longer silent.
322 :    
323 :     CKIT:
324 :     Fixed mismatched uses of Int32 and LargeInt. I always decided
325 :     in favor of LargeInt -- which is now the same as IntInf.
326 :     CKIT-knowledgable people should check whether this is what's
327 :     intended and otherwise change things back to using Int32 or
328 :     FixedInt.
329 :    
330 :     Throughout the code:
331 : mblume 1348 Started using IntInf.int literals and built-in operations
332 : mblume 1347 (e.g., comparison with 0) where this seems appropriate.
333 :    
334 :    
335 :     ----------------------------------------------------------------------
336 : macqueen 1343 Name: Dave MacQueen (dbm@cs.uchicago.edu)
337 :     Date: 2003/08/13 11:36:00 CDT
338 :     Tag: dbm-20030813-mcz-merge1
339 :     Description:
340 :    
341 :     Merging changes from the mcz-branch development branch into trunk.
342 :     These changes involve replacement of the emulated old prettyprinter
343 :     interface with direct use of the SML/NJ Lib PP library, and fixing
344 :     of a couple of bugs (895, 1186) relating to error messages. A new
345 :     prettyprinter for ast datatypes (Elaborator/print/ppast.{sig,sml})
346 :     has been added.
347 :    
348 :     ----------------------------------------------------------------------
349 : mblume 1368 Name: Matthias Blume (blume (at) tti - c (dot) org)
350 : mblume 1342 Date: 2003/08/11 15:45:00 CDT
351 :     Tag: blume-20030811-windows
352 :     Description:
353 :    
354 :     Version number bumped to 110.42.9. NEW BOOTFILES!!!
355 :    
356 :     http://smlnj.cs.uchicago.edu/dist/working/110.42.9/
357 :    
358 :     This patch restores SML/NJ's ability to run under win32. There are a
359 :     number of changes, including fixes for several bugs that had gone
360 :     unnoticed until now:
361 :    
362 :     - uname "CYGWIN_NT*" is recognized as win32 (This is relevant only when
363 :     trying to run the win32 version from within cygwin.)
364 :    
365 :     - There are a number of simple .bat scripts that substitute for their
366 :     corresponding Unix shell-scripts. (See below.)
367 :    
368 :     - The internals of ml-build have been modified slightly. The main
369 :     difference is that instead of calling ".link-sml" (or link-sml.bat)
370 :     using OS.Process.system, the ML process delegates this task back
371 :     to the script. Otherwise problems arise in mixed environments such
372 :     as Cygwin where scripts look and work like Unix scripts, but
373 :     where OS.Process.system cannot run them.
374 :    
375 :     - In CM, the srcpath pickler used native pathname syntax -- which
376 :     is incorrect in the case of cross-compilation. The new pickle format
377 :     is independent of platform-specific naming conventions.
378 :    
379 :     - Path configuration files (such as lib/pathconfig) can now choose
380 :     between native and standard syntax. Placing a line of the form
381 :    
382 :     standard!
383 :    
384 :     into the file causes all subsequent paths to be interpreted using
385 :     CM standard pathname syntax (= Unix conventions); a line
386 :    
387 :     native!
388 :    
389 :     switches back to native style. This was needed so that
390 :     path config files can be written portably, see src/system/pathconfig.
391 :    
392 :     - Runtime system:
393 :    
394 :     - win32-filesys.c: get_file_time and set_file_time now
395 :     access modification time, not creation time.
396 :    
397 :     - I/O code made aware of new array representation.
398 :    
399 :     - Bug fixes in X86.prim.masm.
400 :    
401 :     - src/system/makeml made aware of win32. (For use under cygwin
402 :     and other Unix-environments for windows.)
403 :    
404 :     - In Basis, fixed off-by-one error in win32-io.sml (function vecF)
405 :     which caused BinIO.inputAll to fail consistently.
406 :    
407 :     .bat scripts:
408 :    
409 :     Windows .bat scripts assume that SMLNJ_HOME is defined.
410 :    
411 :     - sml.bat, ml-yacc.bat, ml-lex.bat: Driver scripts for standalone
412 :     applications (sml, ml-yacc, ml-lex).
413 :     - ml-build.bat: analogous to ml-build.
414 :     - config\install.bat: Analogous to config/install.sh. This requires
415 :     that SMLNJ_HOME is set and that Microsoft Visual C is ready to use.
416 :     (nmake etc. must be on the path, and vcvars32 must have been run.)
417 :     Moreover, sources for ml-lex and ml-yacc need to exist under src,
418 :     and the bootfile hierarchy must have been unpacked under
419 :     sml.boot.x86-win32.
420 :     The script is very primitive and does a poor job at error checking.
421 :     It only installs the base system, ml-lex, and ml-yacc. No other
422 :     libraries are being installed (i.e., you get only those that
423 :     are part of the compiler.)
424 :     - link-sml.bat: analogous to .link-sml, but not currently used
425 :    
426 :     Unrelated bug fixes:
427 :    
428 :     - ml-nlffigen now exports structures ST_* corresponding to incomplete
429 :     types.
430 :     - Added getDevice to PP/src/pp-debug-fn.sml. (Would not compile
431 :     otherwise.)
432 :    
433 :     ----------------------------------------------------------------------
434 : macqueen 1337 Name: Dave MacQueen (macqueen@cs.uchicago.edu)
435 :     Date: 2003/06/17
436 :     Tag: macqueen-20030617-bug895
437 :     Description:
438 :    
439 :     Modified compiler/Elaborator/print/pptype.sml to fix bug 895.
440 :     Tag will be used for new development branch (mcz-branch) for
441 :     use by MacQueen, (Lucasz) Zairek, and (George) Cao at uchicago.
442 :    
443 :     ----------------------------------------------------------------------
444 : mblume 1368 Name: Matthias Blume (blume (at) tti - c (dot) org)
445 : mblume 1335 Date: 2003/05/27 16:55:00 CDT
446 :     Tag: blume-20030527-polyeq
447 :     Description:
448 :    
449 :     Tried to eliminated most cases of polymorphic equality.
450 :    
451 :     ----------------------------------------------------------------------
452 : mblume 1368 Name: Matthias Blume (blume (at) tti - c (dot) org)
453 : mblume 1334 Date: 2003/05/21 17:45:00 CDT
454 :     Tag: blume-20030517-complete
455 :     Description:
456 :    
457 :     Two changes:
458 :    
459 :     1. Added a flag for controlling whether non-exhaustive bindings will
460 :     be treated as errors (default is false).
461 :     2. Cleaned up the *entire* source tree so that CMB.make goes through
462 :     without a single non-exhaustive match- or bind warning.
463 :    
464 :     ----------------------------------------------------------------------
465 : mblume 1368 Name: Matthias Blume (blume (at) tti - c (dot) org)
466 : mblume 1332 Date: 2003/05/17 10:20:00 CDT
467 :     Tag: blume-20030517-absyn
468 :     Description:
469 :    
470 :     1. Added cases for IF, WHILE, ANDALSO, and ORELSE to Absyn.
471 :    
472 :     This mainly affects the quality of error messages. However, some
473 :     of the code is now more straightforward than before. (Treatment of
474 :     the above four constructs in translate.sml is much simpler than
475 :     the "macro-expansion" that was going on before. Plus, the mach-
476 :     compiler no longer gets invoked just to be able to compile an
477 :     if-expression.)
478 :    
479 :     2. The ErrorMsg.Error exception is now caught and absorbed by the
480 :     interactive loop.
481 :    
482 :     ----------------------------------------------------------------------
483 : allenleung 1327 Name: Allen Leung
484 :     Date: 2003/05/16 13:05:00 CDT
485 :     Tag: leunga-20030516-cygwin-runtime
486 :     Description:
487 :    
488 :     Ported the runtime system to cygwin, which uses the unix
489 :     x86-unix bin files. Missing/buggy features:
490 :    
491 :     o getnetbyname, getnetbyaddr: these functions seem to be missing in
492 :     the Cygwin library.
493 :     o Ctrl-C handling may be flaky.
494 :     o Windows system calls and Windows I/O are not supported.
495 :    
496 :     A new set of binfiles is located at:
497 :    
498 :     http://www.dorsai.org/~leunga/boot.x86-unix.tgz
499 :    
500 :     This is only needed for bootstrapping the cygwin version of smlnj.
501 :     Other x86 versions can use the existing binfiles.
502 :    
503 :     ----------------------------------------------------------------------
504 : mblume 1317 Name: Matthias Blume
505 :     Date: 2003/04/08 15:42:00 CDT
506 :     Tag: blume-20030408-listpair
507 :     Description:
508 :    
509 :     1. Added a target 'mlrisc' to installer.
510 :    
511 :     2. Added missing elements to structure ListPair.
512 :    
513 :     ----------------------------------------------------------------------
514 : leunga 1296 Name: Allen Leung
515 :     Date: 2003/01/07 10:40:00 EST
516 :     Tag: leunga-20030107-int-rem
517 :     Description:
518 :    
519 :     Fixed a bug in Int.rem(x,y) where y is a power of 2 on x86.
520 :     The arguments to the SUBL instruction were swapped.
521 :    
522 :     ----------------------------------------------------------------------
523 : blume 1229 Name: Matthias Blume
524 : blume 1294 Date: 2002/12/12 16:25:00 EST
525 :     Tag: blume-20021212-risc-ra
526 :     Description:
527 :    
528 :     Fixed a serious bug in the rewrite code for FP spilling/reloading that
529 :     sent the RA into an infinite loop when floating point registers get
530 :     spilled. (Because of this bug, e.g., nucleic stopped compiling between
531 :     110.37 and 110.38.)
532 :     There was another set of potential problems related to the handling of
533 :     MLRISC annotations (but those did not yet cause real problems, apparently).
534 :    
535 :     ----------------------------------------------------------------------
536 :     Name: Matthias Blume
537 : blume 1291 Date: 2002/12/06 22:40:00 EST
538 :     Tag: blume-20021206-cm-fileid
539 :     Description:
540 :    
541 :     Added a call of SrcPath.sync at the beginning of Parse.parse (in CM).
542 :     This fixes the problem of CM getting confused by files that suddenly
543 :     change their identity (e.g., by getting unlinked and recreated by some
544 :     text editor such as vi). There might be a better/cheaper/cleaner way
545 :     of doing this, but for now this will have to do.
546 :    
547 :     ----------------------------------------------------------------------
548 :     Name: Matthias Blume
549 : blume 1288 Date: 2002/10/28 09:50:00 EST
550 :     Tag: blume-20021028-typecheck
551 :     Description:
552 :    
553 :     Exported structure Typecheck from $smlnj/viscomp/core.cm.
554 :    
555 :     ----------------------------------------------------------------------
556 :     Name: Matthias Blume
557 : blume 1282 Date: 2002/10/17 09:10:00 EDT
558 : blume 1279 Tag: Release_110_42
559 :     Description:
560 :    
561 : blume 1282 In good old tradition, there has been a slight hiccup so that we have
562 :     to patch 110.42 after the fact. The old release tag has been replaced
563 :     (see below).
564 :    
565 :     The change solves a problem with two competing approaches the
566 :     configuration problem regarding MacOS 10.1 vs. MacOS 10.2 which got in
567 :     each other's way.
568 :    
569 :     This change only affects the runtime system code and the installer script.
570 :     (No new bootfiles.)
571 :    
572 :     ----------------------------------------------------------------------
573 :     Name: Matthias Blume
574 :     Date: 2002/10/16 12:00:00 EDT
575 :     Tag: Release_110_42_removed
576 :     Description:
577 :    
578 : blume 1279 New working release. New bootfiles.
579 :    
580 :     ----------------------------------------------------------------------
581 :     Name: Matthias Blume
582 : blume 1273 Date: 2002/10/10 13:10:00 EDT
583 :     Tag: blume-20021010-ppc-divs
584 :     Description:
585 :    
586 :     The mltree operator DIVS must be implemented with an overflow check on
587 :     the PPC because the hardware indicates divide-by-zero using "overflow" as
588 :     well.
589 :    
590 :     ----------------------------------------------------------------------
591 :     Name: Matthias Blume
592 : blume 1261 Date: 2002/07/23 11:45:00
593 :     Tag: blume-20020723-smlnj-home
594 :     Description:
595 :    
596 :     Sml now senses the SMLNJ_HOME environment variable. If this is set,
597 :     then the bin dir is assumed to be in $SMLNJ_HOME/bin and (unless
598 :     CM_PATHCONFIG is also set), the path configuration file is assumed
599 :     to be in $SMLNJ_HOME/lib/pathconfig. This way one can easily move
600 :     the entire tree to some other place and everything will "just work".
601 :    
602 :     (Companion commands such as ml-build and ml-makedepend also sense this
603 :     variable.)
604 :    
605 :     ----------------------------------------------------------------------
606 :     Name: Matthias Blume
607 : blume 1259 Date: 2002/07/12 21:19:00 EDT
608 :     Tag: blume-20020712-liveness
609 :     Description:
610 :    
611 :     Exported two useful "step" functions from liveness module (MLRISC).
612 :    
613 :     ----------------------------------------------------------------------
614 :     Name: Matthias Blume
615 : blume 1253 Date: 2002/07/05 16:00 EDT
616 :     Tag: Release_110_41
617 :     Description:
618 :    
619 :     New working release. New bootfiles.
620 :    
621 :     ----------------------------------------------------------------------
622 :     Name: Matthias Blume
623 : blume 1250 Date: 2002/07/05 10:25:00 EDT
624 :     Tag: blume-20020705-btimp
625 :     Description:
626 :    
627 :     Exported structure BTImp from $smlnj/viscomp/debugprof.cm so that
628 :     other clients can set up backtracing support.
629 :    
630 :     ----------------------------------------------------------------------
631 :     Name: Matthias Blume
632 : blume 1249 Date: 2002/06/25 17:23:00 EDT
633 :     Tag: blume-20020625-fpmax
634 :     Description:
635 :    
636 :     Fixed a bug in translation of INLMAX (and INLMIN) for the floating-point
637 :     case. (The sense of the isNaN test was reversed -- which made min and
638 :     max always return their first argument.)
639 :    
640 :     ----------------------------------------------------------------------
641 :     Name: Matthias Blume
642 : blume 1240 Date: 2002/06/11
643 :     Tag: blume-20020611-unixpath
644 :     Description:
645 :    
646 :     Back-ported OS.Path.{from,to}UnixPath from idlbasis-devel branch.
647 :    
648 :     ----------------------------------------------------------------------
649 :     Name: Matthias Blume
650 : blume 1238 Date: 2002/06/10 16:35:00 EDT
651 :     Tag: blume-20020610-ieeereal
652 :     Description:
653 :    
654 :     I back-ported my implementation of IEEEReal.fromString from the
655 :     idlbasis-devel branch so that we can test it.
656 :    
657 :     Another small change is that ppDec tries to give more information
658 :     than just "<sig>" in the case of functors. However, this code is
659 :     broken in some mysterious way if the functor's body's signature
660 :     has not been declared by ascription but gets inferred from the
661 :     implementation. This needs fixing...
662 :    
663 :     ----------------------------------------------------------------------
664 :     Name: Matthias Blume
665 : blume 1229 Date: 2002/05/31
666 :     Tag: blume-20020531-btrace-mode
667 :     Description:
668 :    
669 :     Resurrected SMLofNJ.Internals.BTrace.mode. (It accidentally fell by
670 :     the wayside when I switched over to using Controls everywhere.)
671 :    
672 :     ----------------------------------------------------------------------
673 : george 1222 Name: Lal George
674 :     Date: 2002/05/23 12:21:40 EDT
675 :     Tag: george-20020523-visual-labels
676 :     Description:
677 :    
678 :     Labels are now displayed in the graphical output to make
679 :     the fall-through and target blocks obvious.
680 :    
681 :     ----------------------------------------------------------------------
682 : blume 1204 Name: Matthias Blume
683 : blume 1220 Date: 2002/05/22 11:03:00 EDT
684 :     Tag: blume-20020522-shrink
685 :     Description:
686 :    
687 :     John tweaked yesterday's fix for 1131 to handle an out-of-memory
688 :     situation that comes up when allocating huge arrays.
689 :    
690 :     ----------------------------------------------------------------------
691 :     Name: Matthias Blume
692 : blume 1215 Date: 2002/05/21 16:00:00 EDT
693 :     Tag: Release_110_40
694 :     Description:
695 :    
696 :     New working release (110.40). New bootfiles.
697 :    
698 :     [Also: John Reppy fixed GC bug 1131.]
699 :    
700 :     ----------------------------------------------------------------------
701 :     Name: Matthias Blume
702 : blume 1212 Date: 2002/05/21 12:35:00 EDT
703 :     Tag: blume-20020521-cmdoc
704 :     Description:
705 :    
706 :     CM documentation update.
707 :    
708 :     ----------------------------------------------------------------------
709 :     Name: Matthias Blume
710 : blume 1211 Date: 2002/05/21 10:55:00 EDT
711 :     Tag: blume-20020521-misc
712 :     Description:
713 :    
714 :     - John tweaked runtime to be silent on heap export (except when
715 :     GC messages are on).
716 :     - I added a few more things (cross-compiling versions of CMB) to
717 :     config/preloads (as suggestions).
718 :    
719 :     ----------------------------------------------------------------------
720 :     Name: Matthias Blume
721 : blume 1208 Date: 2002/05/20 22:25:00 EDT
722 :     Tag: blume-20020520-controls
723 :     Description:
724 :    
725 :     - Added ControlUtil structure to control-lib.cm.
726 :     - Use it throughout.
727 :     - Used Controls facility to define MLRISC controls (as opposed to
728 :     registering MLRISC control ref cells with Controls after the fact)
729 :     - Fixed messed-up controls priorities.
730 :    
731 :     * Removed again all the stuff from config/preloads that one wouldn't
732 :     be able to preload at the time the initial heap image is built.
733 :     (Many libraries, e.g., CML, do not exist yet at this time. The
734 :     only libraries that can be preloaded via config/preloads are those
735 :     that come bundled with the bootfiles.)
736 :    
737 :     ----------------------------------------------------------------------
738 :     Name: Matthias Blume
739 : blume 1204 Date: 2002/05/20 10:59:00 EDT
740 :     Tag: blume-20020520-preloads
741 :     Description:
742 :    
743 :     Added a lot of commented-out suggestions for things to be included
744 :     in config/preloads.
745 :    
746 :     ----------------------------------------------------------------------
747 : leunga 1203 Name: Allen Leung
748 :     Date: 2002/05/18 14:20:00 EDT
749 :     Tag: leunga-20020518-mdl
750 :     Description:
751 :    
752 :     o Made the mdl tool stuff compile and run again.
753 :     o I've disabled all the stuff that depends on RTL specifications; they
754 :     are all badly broken anyway.
755 :    
756 :     ----------------------------------------------------------------------
757 : blume 1201 Name: Matthias Blume
758 :     Date: 2002/05/17 16:49:00 EDT
759 :     Tag: blume-20020517-controls
760 :     Description:
761 :    
762 :     0. John Reppy made several modifications to the SML/NJ library.
763 :     In particular, there is a shiny new controls-lib.cm.
764 :    
765 :     1. Pushed new controls interface through compiler so that everything
766 :     compiles again.
767 :    
768 :     2. Added FormatComb and FORMAT_COMB to the CML version of the
769 :     SML/NJ library (so that CML compiles again).
770 :    
771 :     3. Modified init scripts because XXX_DEFAULT environment variables
772 :     are no longer with us. (Boot-time initialization is now done
773 :     using the same environment variables that are also used for
774 :     startup-time initialization of controls.)
775 :    
776 :     ----------------------------------------------------------------------
777 : george 1192 Name: Lal George
778 :     Date: 2002/05/15 09:20:10 EDT
779 :     Tag: george-20020515-pseudo-op-decls
780 :     Description:
781 :    
782 :     All pseudo-ops emitted before the first segment declaration
783 :     such as TEXT, DATA, and BSS directives are assumed to be global
784 :     declarations and are emitted first in the assembly file. This is
785 :     useful in a number of situations where one has pseudo-ops that are not
786 :     specific to any segment, and also works around the constraint that one
787 :     cannot have client pseudo-ops in the TEXT segment.
788 :    
789 :     Because no segment is associated with these declarations it is
790 :     an error to allocate any space or objects before the first segment
791 :     directive and an exception will be raised. However, we cannot make
792 :     this check for client pseudo-ops.
793 :    
794 :     These top level declarations are a field in the CFG graph_info.
795 :     In theory you can continue to add to this field after the CFG has been
796 :     built -- provided you know what you are doing;-)
797 :    
798 :     ----------------------------------------------------------------------
799 : blume 1175 Name: Matthias Blume
800 : blume 1190 Date: 2002/05/13 16:40:00 EDT
801 :     Tag: blume-20020513-pp-etc
802 :     Description:
803 :    
804 :     A few minor bugfixes:
805 :    
806 :     - Stopgap measure for bug recently reported by Elsa Gunter (ppDec).
807 :     (Bogus printouts for redefined bindings still occur. Compiler
808 :     bug should no longer occur now. We need to redo the prettyprinter
809 :     from scratch.)
810 :    
811 :     - CM pathname printer now also adds escape sequences for ( and )
812 :    
813 :     - commend and docu fixes for ml-nlffi
814 :    
815 :     ----------------------------------------------------------------------
816 :     Name: Matthias Blume
817 : blume 1189 Date: 2002/05/10 16:40:00 EDT
818 :     Tag: blume-20020510-erg-textio
819 :     Description:
820 :    
821 :     Applied the following bugfix provided by Emden Gansner:
822 :    
823 :     Output is corrupted when outputSubstr is used rather than output.
824 :    
825 :     The problem occurs when a substring
826 :    
827 :     ss = (s, dataStart, dataLen)
828 :    
829 :     where dataStart > 0, fills a stream buffer with avail bytes left.
830 :     avail bytes of s, starting at index dataStart, are copied into the
831 :     buffer, the buffer is flushed, and then the remaining dataLen-avail
832 :     bytes of ss are copied into the beginning of the buffer. Instead of
833 :     starting this copy at index dataStart+avail in s, the current code
834 :     starts the copy at index avail.
835 :    
836 :     Fix:
837 :     In text-io-fn.sml, change line 695 from
838 :     val needsFlush = copyVec(v, avail, dataLen-avail, buf, 0)
839 :     to
840 :     val needsFlush = copyVec(v, dataStart+avail, dataLen-avail, buf, 0)
841 :    
842 :     ----------------------------------------------------------------------
843 :     Name: Matthias Blume
844 : blume 1186 Date: 2002/04/12 13:55:00 EDT
845 :     Tag: blume-20020412-assyntax
846 :     Description:
847 :    
848 :     1. Grabbed newer assyntax.h from the XFree86 project.
849 :     2. Fiddled with how to compile X86.prim.asm without warnings.
850 :     3. (Very) Minor cleanup in CM.
851 :    
852 :     ----------------------------------------------------------------------
853 :     Name: Matthias Blume
854 : blume 1185 Date: 2002/04/01 (no joke!) 17:07:00 EST
855 :     Tag: blume-20020401-x86div
856 :     Description:
857 :    
858 :     Added full support for div/mod/rem/quot on the x86, using the machine
859 :     instruction's two results (without clumsily recomputing the remainder)
860 :     directly where appropriate.
861 :    
862 :     Some more extensive power-of-two support was added to the x86 instruction
863 :     selector (avoiding expensive divs, mods, and muls where they can be
864 :     replaced with cheaper shifts and masks). However, this sort of thing
865 :     ought to be done earlier, e.g., within the CPS optimizer so that
866 :     all architectures benefit from it.
867 :    
868 :     The compiler compiles to a fixed point, but changes might be somewhat
869 :     fragile nevertheless. Please, report any strange things that you might
870 :     see wrt. div/mod/quot/rem...
871 :    
872 :     ----------------------------------------------------------------------
873 :     Name: Matthias Blume
874 : blume 1184 Date: 2002/03/29 17:22:00
875 :     Tag: blume-20020329-div
876 :     Description:
877 :    
878 :     Fixed my broken div/mod logic. Unfortunately, this means that the
879 :     inline code for div/mod now has one more comparison than before.
880 :     Fast paths (quotient > 0 or remainder = 0) are not affected, though.
881 :     The problem was with quotient = 0, because that alone does not tell
882 :     us which way the rounding went. One then has to look at whether
883 :     remainder and divisor have the same sign... :(
884 :    
885 :     Anyway, I replaced the bootfiles with fresh ones...
886 :    
887 :     ----------------------------------------------------------------------
888 :     Name: Matthias Blume
889 : blume 1183 Date: 2002/03/29 14:10:00 EST
890 :     Tag: blume-20020329-inlprims
891 :     Description:
892 :    
893 :     NEW BOOTFILES!!! Version number bumped to 110.39.3.
894 :    
895 :     Primops have changed. This means that the bin/boot-file formats have
896 :     changed as well.
897 :    
898 :     To make sure that there is no confusion, I made a new version.
899 :    
900 :    
901 :     CHANGES:
902 :    
903 :     * removed REMT from mltree (remainder should never overflow).
904 :    
905 :     * added primops to deal with divisions of all flavors to the frontend
906 :    
907 :     * handled these primops all the way through so they map to their respective
908 :     MLRISC support
909 :    
910 :     * used these primops in the implementation of Int, Int32, Word, Word32
911 :    
912 :     * removed INLDIV, INLMOD, and INLREM as they are no longer necessary
913 :    
914 :     * parameterized INLMIN, INLMAX, and INLABS by a numkind
915 :    
916 :     * translate.sml now deals with all flavors of INL{MIN,MAX,ABS}, including
917 :     floating point
918 :    
919 :     * used INL{MIN,MAX,ABS} in the implementation of Int, Int32, Word, Word32,
920 :     and Real (but Real.abs maps to a separate floating-point-only primop)
921 :    
922 :    
923 :     TODO items:
924 :    
925 :     * Hacked Alpha32 instruction selection, disabling the selection of REMx
926 :     instructions because the machine instruction encoder cannot handle
927 :     them. (Hppa, PPC, and Sparc instruction selection did not handle
928 :     REM in the first place, and REM is supported by the x86 machine coder.)
929 :    
930 :     * Handle DIV and MOD with DIV_TO_NEGINF directly in the x86 instruction
931 :     selection phase. (The two can be streamlined because the hardware
932 :     delivers both quotient and remainder at the same time anyway.)
933 :    
934 :     * Think about what to do with "valOf(Int32.minInt) div ~1" and friends.
935 :     (Currently the behavior is inconsistent both across architectures and
936 :     wrt. the draft Basis spec.)
937 :    
938 :     * Word8 should eventually be handled natively, too.
939 :    
940 :     * There seems to be one serious bug in mltree-gen.sml. It appears, though,
941 :     as if there currently is no execution path that could trigger it in
942 :     SML/NJ. (The assumptions underlying functions arith and promotable do not
943 :     hold for things like multiplication and division.)
944 :    
945 :     ----------------------------------------------------------------------
946 :     Name: Matthias Blume
947 : blume 1181 Date: 2002/03/27 16:27:00 EST
948 :     Tag: blume-20020327-mlrisc-divisions
949 :     Description:
950 :    
951 :     Added support for all four division operations (ML's div, mod, quot,
952 :     and rem) to MLRISC. In the course of doing so, I also rationalized
953 :     the naming (no more annoying switch-around of DIV and QUOT), by
954 :     parameterizing the operation by div_rounding_mode (which can be either
955 :     DIV_TO_ZERO or DIV_TO_NEGINF).
956 :    
957 :     The generic MLTreeGen functor takes care of compiling all four
958 :     operations down to only round-to-zero div.
959 :    
960 :     Missing pieces:
961 :    
962 :     * Doing something smarter than relying on MLTreeGen on architectures
963 :     like, e.g., the x86 where hardware division delivers both quotient and
964 :     remainder at the same time. With this, the implementation of the
965 :     round-to-neginf operations could be further streamlined.
966 :    
967 :     * Remove inlining support for div/mod/rem from the frontend and replace it
968 :     with primops that get carried through to the backend. Do this for all
969 :     int and word types.
970 :    
971 :     ----------------------------------------------------------------------
972 :     Name: Matthias Blume
973 : blume 1180 Date: 2002/03/25 17:25:00 EST
974 :     Tag: blume-20020325-divmod
975 :     Description:
976 :    
977 :     I improved (hopefully without breaking them) the implementation of Int.div,
978 :     Int.mod, and Int.rem. For this, the code in translate.sml now takes
979 :     advantage of the following observations:
980 :    
981 :     Let q = x quot y r = x rem y
982 :     d = x div y m = x mod y
983 :    
984 :     where "quot" is the round-to-zero version of integer division that
985 :     hardware usually provides. Then we have:
986 :    
987 :     r = x - q * y where neither the * nor the - will overflow
988 :     d = if q >= 0 orelse x = q * y then q else q - 1
989 :     where neither the * nor the - will overflow
990 :     m = if q >= 0 orelse r = 0 then r else r + y
991 :     where the + will not overflow
992 :    
993 :     This results in substantial simplification of the generated code.
994 :     The following table shows the number of CFG nodes and edges generated
995 :     for
996 :     fun f (x, y) = x OPER y
997 :     (* with OPER \in div, mod, quot, rem *)
998 :    
999 :    
1000 :     OPER | nodes(old) | edges(old) | nodes(new) | edges(new)
1001 :     --------------------------------------------------------
1002 :     div | 24 | 39 | 12 | 16
1003 :     mod | 41 | 71 | 12 | 16
1004 :     quot | 8 | 10 | 8 | 10
1005 :     rem | 10 | 14 | 8 | 10
1006 :    
1007 :    
1008 :     ----------------------------------------------------------------------
1009 :     Name: Matthias Blume
1010 : blume 1179 Date: 2002/03/25 22:06:00 EST
1011 :     Tag: blume-20020325-cprotobug
1012 :     Description:
1013 :    
1014 :     Fixed a bug in cproto (c prototype decoder).
1015 :    
1016 :     ----------------------------------------------------------------------
1017 :     Name: Matthias Blume
1018 : blume 1178 Date: 2002/03/25 16:00:00 EST
1019 :     Tag: blume-20020325-raw-primops
1020 :     Description:
1021 :    
1022 :     I did some cleanup to Allen's new primop code and
1023 :     replaced yesterday's bootfiles with new ones.
1024 :     (But they are stored in the same place.)
1025 :    
1026 :     ----------------------------------------------------------------------
1027 :     Name: Matthias Blume
1028 : blume 1175 Date: 2002/03/24 22:40:00 EST
1029 :     Tag: blume-20020324-bootfiles
1030 :     Description:
1031 :    
1032 :     Made the bootfiles that Allen asked for.
1033 :    
1034 :     ----------------------------------------------------------------------
1035 : leunga 1174 Name: Allen Leung
1036 :     Date: 2002/03/23 15:50:00 EST
1037 :     Tag: leunga-20020323-flint-cps-rcc-primops
1038 :     Description:
1039 :    
1040 :     1. Changes to FLINT primops:
1041 :    
1042 :     (* make a call to a C-function;
1043 :     * The primop carries C function prototype information and specifies
1044 :     * which of its (ML-) arguments are floating point. C prototype
1045 :     * information is for use by the backend, ML information is for
1046 :     * use by the CPS converter. *)
1047 :     | RAW_CCALL of { c_proto: CTypes.c_proto,
1048 :     ml_args: ccall_type list,
1049 :     ml_res_opt: ccall_type option,
1050 :     reentrant : bool
1051 :     } option
1052 :     (* Allocate uninitialized storage on the heap.
1053 :     * The record is meant to hold short-lived C objects, i.e., they
1054 :     * are not ML pointers. With the tag, the representation is
1055 :     * the same as RECORD with tag tag_raw32 (sz=4), or tag_fblock (sz=8)
1056 :     *)
1057 :     | RAW_RECORD of {tag:bool,sz:int}
1058 :     and ccall_type = CCALL_INT32 | CCALL_REAL64 | CCALL_ML_PTR
1059 :    
1060 :     2. These CPS primops are now overloaded:
1061 :    
1062 :     rawload of {kind:numkind}
1063 :     rawstore of {kind:numkind}
1064 :    
1065 :     The one argument form is:
1066 :    
1067 :     rawload {kind} address
1068 :    
1069 :     The two argument form is:
1070 :    
1071 :     rawload {kind} [ml object, byte-offset]
1072 :    
1073 :     3. RAW_CCALL/RCC now takes two extra arguments:
1074 :    
1075 :     a. The first is whether the C call is reentrant, i.e., whether
1076 :     ML state should be saved and restored.
1077 :     b. The second argument is a string argument specifying the name of
1078 :     library and the C function.
1079 :    
1080 :     These things are currently not handled in the code generator, yet.
1081 :    
1082 :     4. In CProto,
1083 :    
1084 :     An encoding type of "bool" means "ml object" and is mapped into
1085 :     C prototype of PTR. Note that "bool" is different than "string",
1086 :     even though "string" is also mapped into PTR, because "bool"
1087 :     is assigned an CPS type of BOGt, while "string" is assigned INT32t.
1088 :    
1089 :     5. Pickler/unpicker
1090 :    
1091 :     Changed to handle RAW_RECORD and newest RAW_CCALL
1092 :    
1093 :     6. MLRiscGen,
1094 :    
1095 :     1. Changed to handle the new rawload/rawstore/rawrecord operators.
1096 :     2. Code for handling C Calls has been moved to a new module CPSCCalls,
1097 :     in the file CodeGen/cpscompile/cps-c-calls.sml
1098 :    
1099 :     7. Added the conditional move operator
1100 :    
1101 :     condmove of branch
1102 :    
1103 :     to cps. Generation of this is still buggy so it is currently
1104 :     disabled.
1105 :    
1106 :     ----------------------------------------------------------------------
1107 : george 1168 Name: Lal George
1108 :     Date: 2002/03/22 14:18:25 EST
1109 : george 1171 Tag: george-20020322-cps-branch-prob
1110 : george 1168 Description:
1111 :    
1112 : george 1171 Implemented the Ball-Larus branch prediction-heuristics, and
1113 : george 1168 incorporated graphical viewers for control flow graphs.
1114 :    
1115 : george 1171 Ball-Larus Heuristics:
1116 : george 1168 ---------------------
1117 :     See the file compiler/CodeGen/cpscompile/cpsBranchProb.sml.
1118 :    
1119 :     By design it uses the Dempster-Shafer theory for combining
1120 :     probabilities. For example, in the function:
1121 :    
1122 :     fun f(n,acc) = if n = 0 then acc else f(n-1, n*acc)
1123 :    
1124 :     the ball-larus heuristics predicts that the n=0 is unlikely
1125 :     (OH-heuristic), and the 'then' branch is unlikely because of the
1126 :     RH-heuristic -- giving the 'then' branch an even lower combined
1127 : george 1171 probability using the Dempster-Shafer theory.
1128 : george 1168
1129 :     Finally, John Reppy's loop analysis in MLRISC, further lowers the
1130 :     probability of the 'then' branch because of the loop in the else
1131 :     branch.
1132 :    
1133 :    
1134 :     Graphical Viewing:
1135 :     ------------------
1136 :     I merely plugged in Allen's graphical viewers into the compiler. The
1137 :     additional code is not much. At the top level, saying:
1138 :    
1139 :     Control.MLRISC.getFlag "cfg-graphical-view" := true;
1140 :    
1141 :     will display the graphical view of the control flow graph just before
1142 :     back-patching. daVinci must be in your path for this to work. If
1143 :     daVinci is not available, then the default viewer can be changed
1144 :     using:
1145 :    
1146 :     Control.MLRISC.getString "viewer"
1147 :    
1148 :     which can be set to "dot" or "vcg" for the corresponding viewers. Of
1149 :     course, these viewers must be in your path.
1150 :    
1151 :     The above will display the compilation unit at the level of clusters,
1152 :     many of which are small, boring, and un-interesting. Also setting:
1153 :    
1154 :     Control.MLRISC.getInt "cfg-graphical-view_size"
1155 :    
1156 :     will display clusters that are larger than the value set by the above.
1157 :    
1158 :    
1159 :     ----------------------------------------------------------------------
1160 : blume 1159 Name: Matthias Blume
1161 :     Date: 2002/03/21 22:20:00 EST
1162 :     Tag: blume-20020321-kmp-bugfix
1163 :     Description:
1164 :    
1165 :     Changed the interface to the KMP routine in PreString and fixed
1166 :     a minor bug in one place where it was used.
1167 :    
1168 :     ----------------------------------------------------------------------
1169 : leunga 1156 Name: Allen Leung
1170 : leunga 1158 Date: 2002/03/21 20:30:00 EST
1171 :     Tag: leunga-20020321-cfg
1172 :     Description:
1173 :    
1174 :     Fixed a potential problem in cfg edge splitting.
1175 :    
1176 :     ----------------------------------------------------------------------
1177 :     Name: Allen Leung
1178 : leunga 1156 Date: 2002/03/21 17:15:00 EST
1179 :     Tag: leunga-20020321-x86-fp-cfg
1180 :     Description:
1181 :    
1182 :     1. Recoded the buggy parts of x86-fp.
1183 :    
1184 :     a. All the block reordering code has been removed.
1185 :     We now depend on the block placement phases to do this work.
1186 :    
1187 :     b. Critical edge splitting code has been simplified and moved into the
1188 :     CFG modules, as where they belong.
1189 :    
1190 :     Both of these were quite buggy and complex. The code is now much, much
1191 :     simpler.
1192 :    
1193 :     2. X86 backend.
1194 :    
1195 :     a. Added instructions for 64-bit support. Instruction selection for
1196 :     64-bit has not been committed, however, since that
1197 :     requires changes to MLTREE which haven't been approved by
1198 :     Lal and John.
1199 :    
1200 :     b. Added support for FUCOMI and FUCOMIP when generating code for
1201 :     PentiumPro and above. We only generate these instructions in
1202 :     the fast-fp mode.
1203 :    
1204 :     c. Added cases for JP and JNP in X86FreqProps.
1205 :    
1206 :     3. CFG
1207 :    
1208 :     CFG now has a bunch of methods for edge splitting and merging.
1209 :    
1210 :     4. Machine description.
1211 :    
1212 :     John's simplification of MLTREE_BASIS.fcond broke a few machine
1213 :     description things:
1214 :    
1215 :     rtl-build.{sig,sml} and hppa.mdl fixed.
1216 :    
1217 :     NOTE: the machine description stuff in the repository is still broken.
1218 :     Again, I can't put my fixes in because that involves
1219 :     changes to MLTREE.
1220 :    
1221 :     ----------------------------------------------------------------------
1222 : blume 1145 Name: Matthias Blume
1223 : blume 1155 Date: 2002/03/20 15:55:00 EST
1224 :     Tag: blume-20020320-kmp
1225 :     Description:
1226 :    
1227 :     Implemented Knuth-Morris-Pratt string matching in PreString and used
1228 :     it for String.isSubstring, Substring.isSubstring, and
1229 :     Substring.position.
1230 :    
1231 :     (Might need some stress-testing. Simple examples worked fine.)
1232 :    
1233 :     ----------------------------------------------------------------------
1234 :     Name: Matthias Blume
1235 : blume 1152 Date: 2002/03/19 16:37:00 EST
1236 :     Tag: blume-20020319-witnesses
1237 :     Description:
1238 :    
1239 :     Added a structure C.W and functions convert/Ptr.convert to ml-nlffi-lib.
1240 :    
1241 :     This implements a generic mechanism for changing constness qualifiers
1242 :     anywhere within big C types without resorting to outright "casts".
1243 :     (So far, functions such as C.rw/C.ro or C.Ptr.rw/C.Ptr.ro only let you
1244 :     modify the constness at the outermost level.)
1245 :     The implementation of "convert" is based on the idea of "witness"
1246 :     values -- values that are not used by the operation but whose types
1247 :     "testify" to their applicability. On the implementation side, "convert"
1248 :     is simply a projection (returning its second curried argument). With
1249 :     cross-module inlining, it should not result in any machine code being
1250 :     generated.
1251 :    
1252 :     ----------------------------------------------------------------------
1253 :     Name: Matthias Blume
1254 : blume 1148 Date: 2002/03/15 16:40:00 EST
1255 :     Tag: blume-20020315-basis
1256 :     Description:
1257 :    
1258 :     Provided (preliminary?) implementations for
1259 :    
1260 :     {String,Substring}.{concatWith,isSuffix,isSubstring}
1261 :    
1262 :     and
1263 :    
1264 :     Substring.full
1265 :    
1266 :     Those are in the Basis spec but they were missing in SML/NJ.
1267 :    
1268 :     ----------------------------------------------------------------------
1269 :     Name: Matthias Blume
1270 : blume 1145 Date: 2002/03/14 21:30:00 EST
1271 :     Tag: blume-20020314-controls
1272 :     Description:
1273 :    
1274 :     Controls:
1275 :     ---------
1276 :    
1277 :     1. Factored out the recently-added Controls : CONTROLS stuff and put
1278 :     it into its own library $/controls-lib.cm. The source tree for
1279 :     this is under src/smlnj-lib/Controls.
1280 :    
1281 :     2. Changed the names of types and functions in this interface, so they
1282 :     make a bit more "sense":
1283 :    
1284 :     module -> registry
1285 :     'a registry -> 'a group
1286 :    
1287 :     3. The interface now deals in ref cells only. The getter/setter interface
1288 :     is (mostly) gone.
1289 :    
1290 :     4. Added a function that lets one register an already-existing ref cell.
1291 :    
1292 :     5. Made the corresponding modifications to the rest of the code so that
1293 :     everything compiles again.
1294 :    
1295 :     6. Changed the implementation of Controls.MLRISC back to something closer
1296 :     to the original. In particular, this module (and therefore MLRISC)
1297 :     does not depend on Controls. There now is some link-time code in
1298 :     int-sys.sml that registers the MLRISC controls with the Controls
1299 :     module.
1300 :    
1301 :     CM:
1302 :     ---
1303 :    
1304 :     * One can now specify the lambda-split aggressiveness in init.cmi.
1305 :    
1306 :     ----------------------------------------------------------------------
1307 : leunga 1142 Name: Allen Leung
1308 :     Date: 2002/03/13 17:30:00 EST
1309 :     Tag: leunga-20020313-x86-fp-unary
1310 :     Description:
1311 :    
1312 :     Bug fix for:
1313 :    
1314 :     > leunga@weaselbane:~/Yale/tmp/sml-dist{21} bin/sml
1315 :     > Standard ML of New Jersey v110.39.1 [FLINT v1.5], March 08, 2002
1316 :     > - fun f(x,(y,z)) = Real.~ y;
1317 :     > [autoloading]
1318 :     > [autoloading done]
1319 :     > fchsl (%eax), 184(%esp)
1320 :     > Error: MLRisc bug: X86MCEmitter.emitInstr
1321 :     >
1322 :     > uncaught exception Error
1323 :     > raised at: ../MLRISC/control/mlriscErrormsg.sml:16.14-16.19
1324 :    
1325 :     The problem was that the code generator did not generate any fp registers
1326 :     in this case, and the ra didn't know that it needed to run the X86FP phase to
1327 :     translate the pseudo fp instruction. This only happened with unary fp
1328 :     operators in certain situations.
1329 :    
1330 :     ----------------------------------------------------------------------
1331 : blume 1137 Name: Matthias Blume
1332 : blume 1140 Date: 2002/03/13 14:00:00 EST
1333 :     Tag: blume-20020313-overload-etc
1334 :     Description:
1335 :    
1336 :     1. Added _overload as a synonym for overload for backward compatibility.
1337 :     (Control.overloadKW must be true for either version to be accepted.)
1338 :    
1339 :     2. Fixed bug in install script that caused more things to be installed
1340 :     than what was requested in config/targets.
1341 :    
1342 :     3. Made CM aware of the (_)overload construct so that autoloading
1343 :     works.
1344 :    
1345 :     ----------------------------------------------------------------------
1346 :     Name: Matthias Blume
1347 : blume 1138 Date: 2002/03/12 22:03:00 EST
1348 :     Tag: blume-20020312-url
1349 :     Description:
1350 :    
1351 :     Forgot to update BOOT and srcarchiveurl.
1352 :    
1353 :     ----------------------------------------------------------------------
1354 :     Name: Matthias Blume
1355 : blume 1137 Date: 2002/03/12 17:30:00 EST
1356 :     Tag: blume-20020312-version110392
1357 :     Description:
1358 :    
1359 :     Yet another version number bump (because of small changes to the
1360 :     binfile format). Version number is now 110.39.2. NEW BOOTFILES!
1361 :    
1362 :     Changes:
1363 :    
1364 :     The new pid generation scheme described a few weeks ago was overly
1365 :     complicated. I implemented a new mechanism that is simpler and
1366 :     provides a bit more "stability": Once CM has seen a compilation
1367 :     unit, it keeps its identity constant (as long as you do not delete
1368 :     those crucial CM/GUID/* files). This means that when you change
1369 : blume 1145 an interface, compile, then go back to the old interface, and
1370 : blume 1137 compile again, you arrive at the original pid.
1371 :    
1372 :     There now also is a mechanism that instructs CM to use the plain
1373 :     environment hash as a module's pid (effectively making its GUID
1374 :     the empty string). For this, "noguid" must be specified as an
1375 :     option to the .sml file in question within its .cm file.
1376 :     This is most useful for code that is being generated by tools such
1377 :     as ml-nlffigen (because during development programmers tend to
1378 :     erase the tool's entire output directory tree including CM's cached
1379 :     GUIDs). "noguid" is somewhat dangerous (since it can be used to locally
1380 :     revert to the old, broken behavior of SML/NJ, but in specific cases
1381 :     where there is no danger of interface confusion, its use is ok
1382 :     (I think).
1383 :    
1384 :     ml-nlffigen by default generates "noguid" annotations. They can be
1385 :     turned off by specifying -guid in its command line.
1386 :    
1387 :     ----------------------------------------------------------------------
1388 : george 1136 Name: Lal George
1389 :     Date: 2002/03/12 12 14:42:36 EST
1390 :     Tag: george-20020312-frequency-computation
1391 :     Description:
1392 :    
1393 :     Integrated jump chaining and static block frequency into the
1394 :     compiler. More details and numbers later.
1395 : leunga 1025
1396 : george 1003 ----------------------------------------------------------------------
1397 : george 1133 Name: Lal George
1398 :     Date: 2002/03/11 11 22:38:53 EST
1399 :     Tag: george-20020311-jump-chain-elim
1400 :     Description:
1401 :    
1402 :     Tested the jump chain elimination on all architectures (except the
1403 :     hppa). This is on by default right now and is profitable for the
1404 :     alpha and x86, however, it may not be profitable for the sparc and ppc
1405 :     when compiling the compiler.
1406 :    
1407 :     The gc test will typically jump to a label at the end of the cluster,
1408 :     where there is another jump to an external cluster containing the actual
1409 :     code to invoke gc. This is to allow factoring of common gc invocation
1410 :     sequences. That is to say, we generate:
1411 :    
1412 :     f:
1413 :     testgc
1414 :     ja L1 % jump if above to L1
1415 :    
1416 :     L1:
1417 :     jmp L2
1418 :    
1419 :    
1420 :     After jump chain elimination the 'ja L1' instructions is converted to
1421 :     'ja L2'. On the sparc and ppc, many of the 'ja L2' instructions may end
1422 :     up being implemented in their long form (if L2 is far away) using:
1423 :    
1424 :     jbe L3 % jump if below or equal to L3
1425 :     jmp L2
1426 :     L3:
1427 :     ...
1428 :    
1429 :    
1430 :     For large compilation units L2 may be far away.
1431 :    
1432 :    
1433 :     ----------------------------------------------------------------------
1434 : blume 1132 Name: Matthias Blume
1435 :     Date: 2002/03/11 13:30:00 EST
1436 :     Tag: blume-20020311-mltreeeval
1437 :     Description:
1438 :    
1439 :     A functor parameter was missing.
1440 :    
1441 :     ----------------------------------------------------------------------
1442 : leunga 1130 Name: Allen Leung
1443 : leunga 1131 Date: 2002/03/11 10:30:00 EST
1444 : leunga 1142 Tag: leunga-20020311-runtime-string0
1445 : leunga 1131 Description:
1446 :    
1447 :     The representation of the empty string now points to a
1448 :     legal null terminated C string instead of unit. It is now possible
1449 :     to convert an ML string into C string with InlineT.CharVector.getData.
1450 :     This compiles into one single machine instruction.
1451 :    
1452 :     ----------------------------------------------------------------------
1453 :     Name: Allen Leung
1454 : leunga 1130 Date: 2002/03/10 23:55:00 EST
1455 :     Tag: leunga-20020310-x86-call
1456 :     Description:
1457 :    
1458 :     Added machine generation for CALL instruction (relative displacement mode)
1459 :    
1460 :     ----------------------------------------------------------------------
1461 : blume 1128 Name: Matthias Blume
1462 :     Date: 2002/03/08 16:05:00
1463 :     Tag: blume-20020308-entrypoints
1464 :     Description:
1465 :    
1466 :     Version number bumped to 110.39.1. NEW BOOTFILES!
1467 :    
1468 :     Entrypoints: non-zero offset into a code object where execution should begin.
1469 :    
1470 :     - Added the notion of an entrypoint to CodeObj.
1471 :     - Added reading/writing of entrypoint info to Binfile.
1472 :     - Made runtime system bootloader aware of entrypoints.
1473 :     - Use the address of the label of the first function given to mlriscGen
1474 :     as the entrypoint. This address is currently always 0, but it will
1475 :     not be 0 once we turn on block placement.
1476 :     - Removed the linkage cluster code (which was The Other Way(tm) of dealing
1477 :     with entry points) from mlriscGen.
1478 :    
1479 :     ----------------------------------------------------------------------
1480 : leunga 1127 Name: Allen Leung
1481 :     Date: 2002/03/07 20:45:00 EST
1482 :     Tag: leunga-20020307-x86-cmov
1483 :     Description:
1484 :    
1485 :     Bug fixes for CMOVcc on x86.
1486 :    
1487 :     1. Added machine code generation for CMOVcc
1488 :     2. CMOVcc is now generated in preference over SETcc on PentiumPro or above.
1489 :     3. CMOVcc cannot have an immediate operand as argument.
1490 :    
1491 :     ----------------------------------------------------------------------
1492 : blume 1126 Name: Matthias Blume
1493 :     Date: 2002/03/07 16:15:00 EST
1494 :     Tag: blume-20020307-controls
1495 :     Description:
1496 :    
1497 :     This is a very large but mostly boring patch which makes (almost)
1498 :     every tuneable compiler knob (i.e., pretty much everything under
1499 :     Control.* plus a few other things) configurable via both the command
1500 :     line and environment variables in the style CM did its configuration
1501 :     until now.
1502 :    
1503 :     Try starting sml with '-h' (or, if you are brave, '-H')
1504 :    
1505 :     To this end, I added a structure Controls : CONTROLS to smlnj-lib.cm which
1506 :     implements the underlying generic mechanism.
1507 :    
1508 :     The interface to some of the existing such facilities has changed somewhat.
1509 :     For example, the MLRiscControl module now provides mkFoo instead of getFoo.
1510 :     (The getFoo interface is still there for backward-compatibility, but its
1511 :     use is deprecated.)
1512 :    
1513 :     The ml-build script passes -Cxxx=yyy command-line arguments through so
1514 :     that one can now twiddle the compiler settings when using this "batch"
1515 :     compiler.
1516 :    
1517 :     TODO items:
1518 :    
1519 :     We should go through and throw out all controls that are no longer
1520 :     connected to anything. Moreover, we should go through and provide
1521 :     meaningful (and correct!) documentation strings for those controls
1522 :     that still are connected.
1523 :    
1524 :     Currently, multiple calls to Controls.new are accepted (only the first
1525 :     has any effect). Eventually we should make sure that every control
1526 :     is being made (via Controls.new) exactly once. Future access can then
1527 :     be done using Controls.acc.
1528 :    
1529 :     Finally, it would probably be a good idea to use the getter-setter
1530 :     interface to controls rather than ref cells. For the time being, both
1531 :     styles are provided by the Controls module, but getter-setter pairs are
1532 :     better if thread-safety is of any concern because they can be wrapped.
1533 :    
1534 :     *****************************************
1535 :    
1536 :     One bug fix: The function blockPlacement in three of the MLRISC
1537 :     backpatch files used to be hard-wired to one of two possibilities at
1538 :     link time (according to the value of the placementFlag). But (I
1539 :     think) it should rather sense the flag every time.
1540 :    
1541 :     *****************************************
1542 :    
1543 :     Other assorted changes (by other people who did not supply a HISTORY entry):
1544 :    
1545 :     1. the cross-module inliner now works much better (Monnier)
1546 :     2. representation of weights, frequencies, and probabilities in MLRISC
1547 :     changed in preparation of using those for weighted block placement
1548 :     (Reppy, George)
1549 :    
1550 :     ----------------------------------------------------------------------
1551 : george 1116 Name: Lal George
1552 : george 1124 Date: 2002/03/07 14:44:24 EST 2002
1553 :     Tag: george-20020307-weighted-block-placement
1554 :    
1555 :     Tested the weighted block placement optimization on all architectures
1556 :     (except the hppa) using AMPL to generate the block and edge frequencies.
1557 :     Changes were required in the machine properties to correctly
1558 :     categorize trap instructions. There is an MLRISC flag
1559 :     "weighted-block-placement" that can be used to enable weighted block
1560 :     placement, but this will be ineffective without block/edge
1561 :     frequencies (coming soon).
1562 :    
1563 :    
1564 :     ----------------------------------------------------------------------
1565 :     Name: Lal George
1566 : george 1116 Date: 2002/03/05 17:24:48 EST
1567 :     Tag: george-20020305-linkage-cluster
1568 :    
1569 : george 1124 In order to support the block placement optimization, a new cluster
1570 :     is generated as the very first cluster (called the linkage cluster).
1571 :     It contains a single jump to the 'real' entry point for the compilation
1572 :     unit. Block placement has no effect on the linkage cluster itself, but
1573 :     all the other clusters have full freedom in the manner in which they
1574 :     reorder blocks or functions.
1575 : george 1116
1576 :     On the x86 the typical linkage code that is generated is:
1577 :     ----------------------
1578 :     .align 2
1579 :     L0:
1580 :     addl $L1-L0, 72(%esp)
1581 : george 1124 jmp L1
1582 : george 1116
1583 :    
1584 :     .align 2
1585 :     L1:
1586 :     ----------------------
1587 :    
1588 :     72(%esp) is the memory location for the stdlink register. This
1589 :     must contain the address of the CPS function being called. In the
1590 : george 1124 above example, it contains the address of L0; before
1591 : george 1116 calling L1 (the real entry point for the compilation unit), it
1592 :     must contain the address for L1, and hence
1593 :    
1594 :     addl $L1-L0, 72(%esp)
1595 :    
1596 : george 1124 I have tested this on all architectures except the hppa.The increase
1597 :     in code size is of course negligible
1598 : george 1116
1599 :     ----------------------------------------------------------------------
1600 : leunga 1115 Name: Allen Leung
1601 :     Date: 2002/03/03 13:20:00 EST
1602 :     Tag: leunga-20020303-mlrisc-tools
1603 :    
1604 :     Added #[ ... ] expressions to mlrisc tools
1605 :    
1606 :     ----------------------------------------------------------------------
1607 : blume 1096 Name: Matthias Blume
1608 : blume 1098 Date: 2002/02/27 12:29:00 EST
1609 :     Tag: blume-20020227-cdebug
1610 :     Description:
1611 :    
1612 :     - made types in structure C and C_Debug to be equal
1613 :     - got rid of code duplication (c-int.sml vs. c-int-debug.sml)
1614 :     - there no longer is a C_Int_Debug (C_Debug is directly derived from C)
1615 :    
1616 :     ----------------------------------------------------------------------
1617 :     Name: Matthias Blume
1618 : blume 1096 Date: 2002/02/26 12:00:00 EST
1619 :     Tag: blume-20020226-ffi
1620 :     Description:
1621 :    
1622 :     1. Fixed a minor bug in CM's "noweb" tool:
1623 :     If numbering is turned off, then truly don't number (i.e., do not
1624 :     supply the -L option to noweb). The previous behavior was to supply
1625 :     -L'' -- which caused noweb to use the "default" line numbering scheme.
1626 :     Thanks to Chris Richards for pointing this out (and supplying the fix).
1627 :    
1628 :     2. Once again, I reworked some aspects of the FFI:
1629 :    
1630 :     A. The incomplete/complete type business:
1631 :    
1632 :     - Signatures POINTER_TO_INCOMPLETE_TYPE and accompanying functors are
1633 :     gone!
1634 :     - ML types representing an incomplete type are now *equal* to
1635 :     ML types representing their corresponding complete types (just like
1636 :     in C). This is still safe because ml-nlffigen will not generate
1637 :     RTTI for incomplete types, nor will it generate functions that
1638 :     require access to such RTTI. But when ML code generated from both
1639 :     incomplete and complete versions of the C type meet, the ML types
1640 :     are trivially interoperable.
1641 :    
1642 :     NOTE: These changes restore the full generality of the translation
1643 :     (which was previously lost when I eliminated functorization)!
1644 :    
1645 :     B. Enum types:
1646 :    
1647 :     - Structure C now has a type constructor "enum" that is similar to
1648 :     how the "su" constructor works. However, "enum" is not a phantom
1649 :     type because each "T enum" has values (and is isomorphic to
1650 :     MLRep.Signed.int).
1651 :     - There are generic access operations for enum objects (using
1652 :     MLRep.Signed.int).
1653 :     - ml-nlffigen will generate a structure E_foo for each "enum foo".
1654 :     * The structure contains the definition of type "mlrep" (the ML-side
1655 :     representation type of the enum). Normally, mlrep is the same
1656 :     as "MLRep.Signed.int", but if ml-nlffigen was invoked with "-ec",
1657 :     then mlrep will be defined as a datatype -- thus facilitating
1658 :     pattern matching on mlrep values.
1659 :     ("-ec" will be suppressed if there are duplicate values in an
1660 :     enumeration.)
1661 :     * Constructors ("-ec") or values (no "-ec") e_xxx of type mlrep
1662 :     will be generated for each C enum constant xxx.
1663 :     * Conversion functions m2i and i2m convert between mlrep and
1664 :     MLRep.Signed.int. (Without "-ec", these functions are identities.)
1665 :     * Coversion functions c and ml convert between mlrep and "tag enum".
1666 :     * Access functions (get/set) fetch and store mlrep values.
1667 :     - By default (unless ml-nlffigen was invoked with "-nocollect"), unnamed
1668 :     enumerations are merged into one single enumeration represented by
1669 :     structure E_'.
1670 :    
1671 :     ----------------------------------------------------------------------
1672 : leunga 1080 Name: Allen Leung
1673 : leunga 1094 Date: 2002/02/25 04:45:00 EST
1674 :     Tag: leunga-20020225-cps-spill
1675 :    
1676 :     This is a new implementation of the CPS spill phase.
1677 :     The new phase is in the new file compiler/CodeGen/cpscompile/spill-new.sml
1678 :     In case of problems, replace it with the old file spill.sml
1679 :    
1680 :     The current compiler runs into some serious performance problems when
1681 :     constructing a large record. This can happen when we try to compile a
1682 :     structure with many items. Even a very simple structure like the following
1683 :     makes the compiler slow down.
1684 :    
1685 :     structure Foo = struct
1686 :     val x_1 = 0w1 : Word32.int
1687 :     val x_2 = 0w2 : Word32.int
1688 :     val x_3 = 0w3 : Word32.int
1689 :     ...
1690 :     val x_N = 0wN : Word32.int
1691 :     end
1692 :    
1693 :     The following table shows the compile time, from N=1000 to N=4000,
1694 :     with the old compiler:
1695 :    
1696 :     N
1697 :     1000 CPS 100 spill 0.04u 0.00s 0.00g
1698 :     MLRISC ra 0.06u 0.00s 0.05g
1699 :     (spills = 0 reloads = 0)
1700 :     TOTAL 0.63u 0.07s 0.21g
1701 :    
1702 :     1100 CPS 100 spill 8.25u 0.32s 0.64g
1703 :     MLRISC ra 5.68u 0.59s 3.93g
1704 :     (spills = 0 reloads = 0)
1705 :     TOTAL 14.71u 0.99s 4.81g
1706 :    
1707 :     1500 CPS 100 spill 58.55u 2.34s 1.74g
1708 :     MLRISC ra 5.54u 0.65s 3.91g
1709 :     (spills = 543 reloads = 1082)
1710 :     TOTAL 65.40u 3.13s 6.00g
1711 :    
1712 :     2000 CPS 100 spill 126.69u 4.84s 3.08g
1713 :     MLRISC ra 0.80u 0.10s 0.55g
1714 :     (spills = 42 reloads = 84)
1715 :     TOTAL 129.42u 5.10s 4.13g
1716 :    
1717 :     3000 CPS 100 spill 675.59u 19.03s 11.64g
1718 :     MLRISC ra 2.69u 0.27s 1.38g
1719 :     (spills = 62 reloads = 124)
1720 :     TOTAL 682.48u 19.61s 13.99g
1721 :    
1722 :     4000 CPS 100 spill 2362.82u 56.28s 43.60g
1723 :     MLRISC ra 4.96u 0.27s 2.72g
1724 :     (spills = 85 reloads = 170)
1725 :     TOTAL 2375.26u 57.21s 48.00g
1726 :    
1727 :     As you can see the old cps spill module suffers from some serious
1728 : leunga 1115 performance problem. But since I cannot decipher the old code fully,
1729 :     instead of patching the problems up, I'm reimplementing it
1730 : leunga 1094 with a different algorithm. The new code is more modular,
1731 :     smaller when compiled, and substantially faster
1732 :     (O(n log n) time and O(n) space). Timing of the new spill module:
1733 :    
1734 :     4000 CPS 100 spill 0.02u 0.00s 0.00g
1735 :     MLRISC ra 0.25u 0.02s 0.15g
1736 :     (spills=1 reloads=3)
1737 :     TOTAL 7.74u 0.34s 1.62g
1738 :    
1739 :     Implementation details:
1740 :    
1741 :     As far as I can tell, the purpose of the CPS spill module is to make sure the
1742 :     number of live variables at any program point (the bandwidth)
1743 :     does not exceed a certain limit, which is determined by the
1744 :     size of the spill area.
1745 :    
1746 :     When the bandwidth is too large, we decrease the register pressure by
1747 :     packing live variables into spill records. How we achieve this is
1748 :     completely different than what we did in the old code.
1749 :    
1750 :     First, there is something about the MLRiscGen code generator
1751 :     that we should be aware of:
1752 :    
1753 :     o MLRiscGen performs code motion!
1754 :    
1755 :     In particular, it will move floating point computations and
1756 :     address computations involving only the heap pointer to
1757 :     their use sites (if there is only a single use).
1758 :     What this means is that if we have a CPS record construction
1759 :     statement
1760 :    
1761 :     RECORD(k,vl,w,e)
1762 :    
1763 :     we should never count the new record address w as live if w
1764 :     has only one use (which is often the case).
1765 :    
1766 :     We should do something similar to floating point, but the transformation
1767 :     there is much more complex, so I won't deal with that.
1768 :    
1769 :     Secondly, there are now two new cps primops at our disposal:
1770 :    
1771 :     1. rawrecord of record_kind option
1772 :     This pure operator allocates some uninitialized storage from the heap.
1773 :     There are two forms:
1774 :    
1775 :     rawrecord NONE [INT n] allocates a tagless record of length n
1776 :     rawrecord (SOME rk) [INT n] allocates a tagged record of length n
1777 :     and initializes the tag.
1778 :    
1779 :     2. rawupdate of cty
1780 :     rawupdate cty (v,i,x)
1781 :     Assigns to x to the ith component of record v.
1782 :     The storelist is not updated.
1783 :    
1784 :     We use these new primops for both spilling and increment record construction.
1785 :    
1786 :     1. Spilling.
1787 :    
1788 :     This is implemented with a linear scan algorithm (but generalized
1789 :     to trees). The algorithm will create a single spill record at the
1790 :     beginning of the cps function and use rawupdate to spill to it,
1791 :     and SELECT or SELp to reload from it. So both spills and reloads
1792 :     are fine-grain operations. In contrast, in the old algorithm
1793 :     "spills" have to be bundled together in records.
1794 :    
1795 :     Ideally, we should sink the spill record construction to where
1796 :     it is needed. We can even split the spill record into multiple ones
1797 :     at the places where they are needed. But CPS is not a good
1798 :     representation for global code motion, so I'll keep it simple and
1799 :     am not attempting this.
1800 :    
1801 :     2. Incremental record construction (aka record splitting).
1802 :    
1803 :     Long records with many component values which are simulatenously live
1804 :     (recall that single use record addresses are not considered to
1805 :     be live) are constructed with rawrecord and rawupdate.
1806 :     We allocate space on the heap with rawrecord first, then gradually
1807 :     fill it in with rawupdate. This is the technique suggested to me
1808 :     by Matthias.
1809 :    
1810 :     Some restrictions on when this is applicable:
1811 :     1. It is not a VECTOR record. The code generator currently does not handle
1812 :     this case. VECTOR record uses double indirection like arrays.
1813 :     2. All the record component values are defined in the same "basic block"
1814 :     as the record constructor. This is to prevent speculative
1815 :     record construction.
1816 :    
1817 :     ----------------------------------------------------------------------
1818 :     Name: Allen Leung
1819 : leunga 1086 Date: 2002/02/22 01:02:00 EST
1820 :     Tag: leunga-20020222-mlrisc-tools
1821 :    
1822 :     Minor bug fixes in the parser and rewriter
1823 :    
1824 :     ----------------------------------------------------------------------
1825 :     Name: Allen Leung
1826 : leunga 1085 Date: 2002/02/21 20:20:00 EST
1827 :     Tag: leunga-20020221-peephole
1828 :    
1829 :     Regenerated the peephole files. Some contained typos in the specification
1830 :     and some didn't compile because of pretty printing bugs in the old version
1831 :     of 'nowhere'.
1832 :    
1833 :     ----------------------------------------------------------------------
1834 :     Name: Allen Leung
1835 : leunga 1079 Date: 2002/02/19 20:20:00 EST
1836 :     Tag: leunga-20020219-mlrisc-tools
1837 :     Description:
1838 :    
1839 :     Minor bug fixes to the mlrisc-tools library:
1840 :    
1841 :     1. Fixed up parsing colon suffixed keywords
1842 :     2. Added the ability to shut the error messages up
1843 :     3. Reimplemented the pretty printer and fixed up/improved
1844 :     the pretty printing of handle and -> types.
1845 :     4. Fixed up generation of literal symbols in the nowhere tool.
1846 :     5. Added some SML keywords to to sml.sty
1847 :    
1848 :     ----------------------------------------------------------------------
1849 :     Name: Matthias Blume
1850 : blume 1078 Date: 2002/02/19 16:20:00 EST
1851 :     Tag: blume-20020219-cmffi
1852 :     Description:
1853 :    
1854 :     A wild mix of changes, some minor, some major:
1855 :    
1856 :     * All C FFI-related libraries are now anchored under $c:
1857 :     $/c.cm --> $c/c.cm
1858 :     $/c-int.cm --> $c/internals/c-int.cm
1859 :     $/memory.cm --> $c/memory/memory.cm
1860 :    
1861 :     * "make" tool (in CM) now treats its argument pathname slightly
1862 :     differently:
1863 :     1. If the native expansion is an absolute name, then before invoking
1864 :     the "make" command on it, CM will apply OS.Path.mkRelative
1865 :     (with relativeTo = OS.FileSys.getDir()) to it.
1866 :     2. The argument will be passed through to subsequent phases of CM
1867 :     processing without "going native". In particular, if the argument
1868 :     was an anchored path, then "make" will not lose track of that anchor.
1869 :    
1870 :     * Compiler backends now "know" their respective C calling conventions
1871 :     instead of having to be told about it by ml-nlffigen. This relieves
1872 :     ml-nlffigen from one of its burdens.
1873 :    
1874 :     * The X86Backend has been split into X86CCallBackend and X86StdCallBackend.
1875 :    
1876 :     * Export C_DEBUG and C_Debug from $c/c.cm.
1877 :    
1878 :     * C type encoding in ml-nlffi-lib has been improved to model the conceptual
1879 :     subtyping relationship between incomplete pointers and their complete
1880 :     counterparts. For this, ('t, 'c) ptr has been changed to 'o ptr --
1881 :     with the convention of instantiating 'o with ('t, 'c) obj whenever
1882 :     the pointer target type is complete. In the incomplete case, 'o
1883 :     will be instantiated with some "'c iobj" -- a type obtained by
1884 :     using one of the functors PointerToIncompleteType or PointerToCompleteType.
1885 :    
1886 :     Operations that work on both incomplete and complete pointer types are
1887 :     typed as taking an 'o ptr while operations that require the target to
1888 :     be known are typed as taking some ('t, 'c) obj ptr.
1889 :    
1890 :     voidptr is now a bit "more concrete", namely "type voidptr = void ptr'"
1891 :     where void is an eqtype without any values. This makes it possible
1892 :     to work on voidptr values using functions meant to operate on light
1893 :     incomplete pointers.
1894 :    
1895 :     * As a result of the above, signature POINTER_TO_INCOMPLETE_TYPE has
1896 :     been vastly simplified.
1897 :    
1898 :     ----------------------------------------------------------------------
1899 :     Name: Matthias Blume
1900 : blume 1076 Date: 2002/02/19 10:48:00 EST
1901 :     Tag: blume-20020219-pqfix
1902 :     Description:
1903 :    
1904 :     Applied Chris Okasaki's bug fix for priority queues.
1905 :    
1906 :     ----------------------------------------------------------------------
1907 :     Name: Matthias Blume
1908 : blume 1073 Date: 2002/02/15 17:05:00
1909 : blume 1069 Tag: Release_110_39
1910 :     Description:
1911 :    
1912 : blume 1073 Last-minute retagging is becoming a tradition... :-(
1913 :    
1914 :     This is the working release 110.39.
1915 :    
1916 :     ----------------------------------------------------------------------
1917 :     Name: Matthias Blume
1918 :     Date: 2002/02/15 16:00:00 EST
1919 :     Tag: Release_110_39-orig
1920 :     Description:
1921 :    
1922 : blume 1069 Working release 110.39. New bootfiles.
1923 :    
1924 : blume 1073 (Update: There was a small bug in the installer so it wouldn't work
1925 :     with all shells. So I retagged. -Matthias)
1926 :    
1927 : blume 1069 ----------------------------------------------------------------------
1928 :     Name: Matthias Blume
1929 : blume 1068 Date: 2002/02/15 14:17:00 EST
1930 :     Tag: blume-20020215-showbindings
1931 :     Description:
1932 :    
1933 :     Added EnvRef.listBoundSymbols and CM.State.showBindings. Especially
1934 :     the latter can be useful for exploring what bindings are available at
1935 :     the interactive prompt. (The first function returns only the list
1936 :     of symbols that are really bound, the second prints those but also the
1937 :     ones that CM's autoloading mechanism knows about.)
1938 :    
1939 :     ----------------------------------------------------------------------
1940 :     Name: Matthias Blume
1941 : blume 1067 Date: 2002/02/15 12:08:00 EST
1942 :     Tag: blume-20020215-iptrs
1943 :     Description:
1944 :    
1945 :     Two improvements to ml-nlffigen:
1946 :    
1947 :     1. Write files only if they do not exist or if their current contents
1948 :     do not coincide with what's being written. (That is, avoid messing
1949 :     with the time stamps unless absolutely necessary.)
1950 :    
1951 :     2. Implement a "repository" mechanism for generated files related
1952 :     to "incomplete pointer types". See the README file for details.
1953 :    
1954 :     ----------------------------------------------------------------------
1955 :     Name: Matthias Blume
1956 : blume 1066 Date: 2002/02/14 11:50:00 EST
1957 :     Tag: blume-20020214-quote
1958 :     Description:
1959 :    
1960 :     Added a type 't t_' to tag.sml (in ml-nlffi-lib.cm). This is required
1961 :     because of the new and improved tag generation scheme. (Thanks to Allen
1962 :     Leung for pointing it out.)
1963 :    
1964 :     ----------------------------------------------------------------------
1965 : george 1065 Name: Lal George
1966 :     Date: 2002/02/14 09:55:27 EST 2002
1967 :     Tag: george-20020214-isabelle-bug
1968 :     Description:
1969 :    
1970 :     Fixed the MLRISC bug sent by Markus Wenzel regarding the compilation
1971 :     of Isabelle on the x86.
1972 :    
1973 :     From Allen:
1974 :     -----------
1975 :     I've found the problem:
1976 :    
1977 :     in ra-core.sml, I use the counter "blocked" to keep track of the
1978 :     true number of elements in the freeze queue. When the counter goes
1979 :     to zero, I skip examining the queue. But I've messed up the
1980 :     bookkeeping in combine():
1981 :    
1982 :     else ();
1983 :     case !ucol of
1984 :     PSEUDO => (if !cntv > 0 then
1985 :     (if !cntu > 0 then blocked := !blocked - 1 else ();
1986 :     ^^^^^^^^^^^^^^^^^^^^^^^
1987 :     moveu := mergeMoveList(!movev, !moveu)
1988 :     )
1989 :     else ();
1990 :    
1991 :     combine() is called to coalesce two nodes u and v.
1992 :     I think I was thinking that if the move counts of u and v are both
1993 :     greater than zero then after they are coalesced then one node is
1994 :     removed from the freeze queue. Apparently I was thinking that
1995 :     both u and v are of low degree, but that's clearly not necessarily true.
1996 :    
1997 :    
1998 :     02/12/2002:
1999 :     Here's the patch. HOL now compiles.
2000 :    
2001 :     I don't know how this impact on performance (compile
2002 :     time or runtime). This bug caused the RA (especially on the x86)
2003 :     to go thru the potential spill phase when there are still nodes on the
2004 :     freeze queue.
2005 :    
2006 :    
2007 :    
2008 :    
2009 :     ----------------------------------------------------------------------
2010 : blume 1058 Name: Matthias Blume
2011 : blume 1064 Date: 2002/02/13 22:40:00 EST
2012 :     Tag: blume-20020213-fptr-rtti
2013 :     Description:
2014 :    
2015 :     Fixed a bug in ml-nlffigen that was introduced with one of the previous
2016 :     updates.
2017 :    
2018 :     ----------------------------------------------------------------------
2019 :     Name: Matthias Blume
2020 : blume 1063 Date: 2002/02/13 16:41:00 EST
2021 :     Tag: blume-20020213-cmlpq
2022 :     Description:
2023 :    
2024 :     Added new priority queue export symbols (which have just been added to
2025 :     smlnj-lib.cm) to CML's version of smlnj-lib.cm. (Otherwise CML would
2026 :     not compile and the installer would choke.)
2027 :    
2028 :     ----------------------------------------------------------------------
2029 :     Name: Matthias Blume
2030 : blume 1062 Date: 2002/02/13 16:15:00 EST
2031 :     Tag: blume-20020213-various
2032 :     Description:
2033 :    
2034 :     1. More tweaks to ml-nlffigen:
2035 :    
2036 :     - better internal datastructures (resulting in slight speedup)
2037 :     - "-match" option requires exact match
2038 :     - "localized" gensym counters (untagged structs/unions nested within
2039 :     other structs/unions or within typedefs get a fresh counter; their
2040 :     tag will be prefixed by a concatenation of their parents' tags)
2041 :     - bug fixes (related to calculation of transitive closure of types
2042 :     to be included in the output)
2043 :    
2044 :     2. Minor Basis updates:
2045 :    
2046 :     - added implementations for List.collate and Option.app
2047 :    
2048 :     ----------------------------------------------------------------------
2049 :     Name: Matthias Blume
2050 : blume 1060 Date: 2002/02/11 15:55:00 EST
2051 :     Tag: blume-20020211-gensym
2052 :     Description:
2053 :    
2054 :     Added a "-gensym" option to command line of ml-nlffigen. This can be
2055 :     used to specify a "stem" -- a string that is inserted in all "gensym'd"
2056 :     names (ML structure names that correspond to unnamed C structs, unions,
2057 :     and enums), so that separate runs of ml-nlffigen do not clash.
2058 :    
2059 :     ----------------------------------------------------------------------
2060 :     Name: Matthias Blume
2061 : blume 1059 Date: 2002/02/11 12:05:00 EST
2062 :     Tag: blume-20020211-gensml
2063 :     Description:
2064 :    
2065 :     A quick fix for a problem with GenSML (in the pgraph-util library):
2066 :     Make generation of toplevel "local" optional. (Strictly speaking,
2067 :     signature definitions within "local" are not legal SML.)
2068 :    
2069 :     Other than that: updates to INSTALL and cm/TODO.
2070 :    
2071 :     ----------------------------------------------------------------------
2072 :     Name: Matthias Blume
2073 : blume 1058 Date: 2002/02/08 15:00:00 EST
2074 :     Tag: blume-20020208-uniquepid
2075 :     Description:
2076 :    
2077 :     0. Version number has been bumped to 110.38.1. NEW BOOTFILES!!!
2078 :    
2079 :     1. The installer (config/install.sh) has gotten smarter:
2080 :    
2081 :     - Configuration options are a bit easier to specify now
2082 :     (in config/targets).
2083 :     - Bug in recognizing .tar.bz2 files fixed.
2084 :     - Installer automatically resolves dependencies between
2085 :     configuration options (e.g., if you ask for eXene, you will
2086 :     also get cml -- regardless whether you asked for it or not).
2087 :     - Installer can run in "quieter mode" by setting the environment
2088 :     variable INSTALL_QUIETLY to "true". "Quieter" does not mean
2089 :     "completely silent", though.
2090 :     - Build HashCons library as part of smlnj-lib.
2091 :    
2092 :     2. A new scheme for assigning persistent identifiers to compilation
2093 :     units (and, by extension, to types etc.) has been put into place.
2094 :     This fixes a long-standing bug where types and even dynamic values
2095 :     can get internally confused, thereby compromising type safety
2096 :     (abstraction) and dynamic correctness. See
2097 :    
2098 :     http://cm.bell-labs.com/cm/cs/who/blume/pid-confusion.tgz
2099 :    
2100 :     for an example of how things could go wrong until now.
2101 :    
2102 :     The downside of the new scheme is that pids are not quite as
2103 :     persistent as they used to be: CM will generate a fresh pid
2104 :     for every compilation unit that it thinks it sees for the first
2105 :     time. That means that if you compile starting from a clean, fresh
2106 :     source tree at two different times, you end up with different
2107 :     binaries.
2108 :    
2109 :     Cutoff recompilation, however, has not been compromised because
2110 :     CM keeps pid information in special caches between runs.
2111 :    
2112 :     ----------------------------------------------------------------------
2113 : george 1053 Name: Lal George
2114 : george 1055 Date: 2002/02/07 15:34:13 EST 2002
2115 :     Tag: <none>
2116 :     Description:
2117 :    
2118 :     Compilers that generate assembly code may produce global labels
2119 :     whose value is resolved at link time. The various peephole optimization
2120 :     modules did not take this in account.
2121 :    
2122 :     TODO. The Labels.addrOf function should really return an option
2123 :     type so that clients are forced to deal with this issue, rather
2124 :     than an exception being raised.
2125 :    
2126 :     ----------------------------------------------------------------------
2127 :     Name: Lal George
2128 : george 1053 Date: 2002/02/06 13:55:02 EST
2129 :     Tag: george-20020206-ra-breakup
2130 :     Description:
2131 :    
2132 :     1. A bug fix from Allen.
2133 :    
2134 :     A typo causes extra fstp %st(0)'s to be generated at compensation
2135 :     edges, which might cause stack underflow traps at runtime. This
2136 :     occurs in fft where there are extraneous fstps right before the 'into'
2137 :     trap instruction (in this case they are harmless since none of the
2138 :     integers overflow.)
2139 :    
2140 : george 1054 2. Pulled out various utility modules that were embedded in the modules
2141 : george 1053 of the register allocator. I need these modules for other purposes, but
2142 : george 1054 they are not complete enough to put into a library (just yet).
2143 : george 1053 ----------------------------------------------------------------------
2144 : blume 1049 Name: Matthias Blume
2145 :     Date: 2002/01/31 16:05:00 EST
2146 :     Tag: blume-20020131-sparc-ccalls
2147 :     Description:
2148 :    
2149 :     1. C-calls on Sparc needlessly allocated a huge chunk (96 bytes)
2150 :     of extra stack space by mistake. Fixed.
2151 :    
2152 :     2. Bug in logic of handling of command-line options in ml-nlffigen fixed.
2153 :    
2154 :     ----------------------------------------------------------------------
2155 : leunga 1048 Name: Allen Leung
2156 :     Date: 2002/01/30
2157 :     Tag: leunga-20020130-nowhere-bug-fix
2158 :     Description:
2159 :    
2160 :     MLRISC bug fixes:
2161 :     1. Fixed a bindings computation bug in the 'nowhere' program generator tool.
2162 :     2. MachineInt.fromString was negating its value.
2163 :    
2164 :     ----------------------------------------------------------------------
2165 : blume 1039 Name: Matthias Blume
2166 : blume 1047 Date: 2002/01/29
2167 :     Tag: blume-20020129-INSTALL
2168 :     Description:
2169 :    
2170 :     - Added somewhat detailed installation instructions (file INSTALL).
2171 :     - Fixed curl-detection bug in config/install.sh.
2172 :     - It is now possible to select the URL getter using the URLGETTER
2173 :     environment variable:
2174 :    
2175 :     not set / "unknown" --> automatic detection (script tries wget,
2176 :     curl, and lynx)
2177 :     "wget" / "curl" / "lynx" --> use the specified program (script "knows"
2178 :     how to properly invoke them)
2179 :     other --> use $URLGETTER directly, it must take
2180 :     precisely two command-line arguments
2181 :     (source URL and destination file name)
2182 :    
2183 :     ----------------------------------------------------------------------
2184 :     Name: Matthias Blume
2185 : blume 1044 Date: 2002/01/28
2186 : blume 1045 Tag: blume-20020128-sparc-ccalls
2187 :     Description:
2188 :    
2189 :     - Fixed problem with calculation of "used" registers in sparc-c-calls.
2190 :     - Make use of the allocParam argument in sparc-c-calls.
2191 :    
2192 :     ----------------------------------------------------------------------
2193 :     Name: Matthias Blume
2194 :     Date: 2002/01/28
2195 : blume 1044 Tag: blume-20020128-allocParam
2196 :     Description:
2197 :    
2198 :     John Reppy: Changes c-calls API to accept client-callback for
2199 :     allocating extra stack space.
2200 :     me: Corresponding changes to mlriscGen (using a dummy argument that
2201 :     does not change the current behavior).
2202 :    
2203 :     ----------------------------------------------------------------------
2204 :     Name: Matthias Blume
2205 : blume 1040 Date: 2002/01/28 12:00:00
2206 :     Tag: Release_110_38
2207 :     Description:
2208 :    
2209 :     This time for real!!!
2210 :    
2211 :     ----------------------------------------------------------------------
2212 :     Name: Matthias Blume
2213 : blume 1039 Date: 2002/01/28 10:56:00 EST
2214 :     Tag: blume-20020128-retraction
2215 :     Description:
2216 :    
2217 :     0. Retracted earlier 110.38. (The Release_110_38 tag has been replaced
2218 :     with blume-Release_110_38-retracted.)
2219 :    
2220 :     1. Fixed a problem with incorrect rounding modes in real64.sml.
2221 :     (Thanks to Andrew Mccreight <andrew.mccreight@yale.edu>.)
2222 :    
2223 :     2. A bug in ml-nlffigen related to the handling of unnamed structs, unions,
2224 :     and enums fixed. The naming of corresponding ML identifiers should
2225 :     now be consistent again.
2226 :    
2227 :     ----------------------------------------------------------------------
2228 : leunga 1038 Name: Allen Leung
2229 :     Date: 2002/01/27
2230 :     Tag: leunga-20020127-nowhere
2231 :     Description:
2232 :    
2233 :     Added a target called nowhere in the configuration scripts.
2234 :     Enabling this will build the MLRISC 'nowhere' tool (for translating
2235 :     programs with where-clauses into legal SML code) during installation.
2236 :    
2237 :     ----------------------------------------------------------------------
2238 : blume 1036 Name: Matthias Blume
2239 : leunga 1038 Date: 2002/01/25 21:27:00 EST
2240 : blume 1039 Tag: blume-Release_110_38-retracted
2241 : blume 1037 Description:
2242 :    
2243 :     Call it a (working) release! Version is 110.38. Bootfiles are ready.
2244 :    
2245 :     README will be added later.
2246 :    
2247 : blume 1039 !!! NOTE: Re-tagged as blume-Release_110_38-retracted. Original tag
2248 :     (Release_110_38) removed. Reason: Last-minute bug fixes.
2249 :    
2250 : blume 1037 ----------------------------------------------------------------------
2251 :     Name: Matthias Blume
2252 : blume 1036 Date: 2002/01/25
2253 :     Tag: blume-20020125-ffi
2254 :     Description:
2255 :    
2256 :     A large number of tweaks and improvements to ml-nlffi-lib and
2257 :     ml-nlffigen:
2258 :    
2259 :     - ML represenation types have been streamlined
2260 :     - getter and setter functions work with concrete values, not abstract
2261 :     ones where possible
2262 :     - ml-nlffigen command line more flexible (see README file there)
2263 :     - some bugs have been fixed (hopefully)
2264 :    
2265 :     ----------------------------------------------------------------------
2266 : george 1034 Name: Lal George
2267 :     Date: 2002/01/24
2268 :     Tag: george-20020124-risc-ra-interface
2269 :     Description:
2270 :    
2271 :     There is a dramatic simplification in the interface to the
2272 :     register allocator for RISC architectures as a result of making
2273 :     parallel copy instructions explicit.
2274 :    
2275 :     ----------------------------------------------------------------------
2276 : blume 1028 Name: Matthias Blume
2277 : blume 1032 Date: 2002/01/22
2278 :     Tag: blume-20020122-x86-ccalls
2279 :     Description:
2280 :    
2281 :     Bug fix for c-calls on x86 (having to do with how char- and
2282 :     short-arguments are being handled).
2283 :    
2284 :     ----------------------------------------------------------------------
2285 :     Name: Matthias Blume
2286 : blume 1031 Date: 2002/01/21
2287 :     Tag: blume-20020121-ff
2288 :     Description:
2289 :    
2290 :     Another day of fiddling with the FFI...
2291 :    
2292 :     1. Bug fix/workaround: CKIT does not complain about negative array
2293 :     dimensions, so ml-nlffigen has to guard itself against this possibility.
2294 :     (Otherwise a negative dimension would send it into an infinite loop.)
2295 :    
2296 :     2. Some of the abstract types (light objects, light pointers, most "base"
2297 :     types) in structure C are now eqtypes.
2298 :    
2299 :     3. Added constructors and test functions for NULL function pointers.
2300 :    
2301 :     ----------------------------------------------------------------------
2302 :     Name: Matthias Blume
2303 : blume 1028 Date: 2002/01/18
2304 : blume 1030 Tag: blume-20020118-ready-for-new-release
2305 :     Description:
2306 :    
2307 :     Made config/srcarchiveurl point to a new place. (Will provide boot
2308 :     files shortly.)
2309 :    
2310 :     Maybe we christen this to be 110.38?
2311 :    
2312 :     ----------------------------------------------------------------------
2313 :     Name: Matthias Blume
2314 :     Date: 2002/01/18
2315 : blume 1029 Tag: blume-20020118-more-ffifiddle
2316 :     Description:
2317 :    
2318 :     Today's FFI fiddling:
2319 :    
2320 :     - Provided a structure CGetSet with "convenient" versions of C.Get.* and
2321 :     C.Set.* that use concrete (MLRep.*) arguments and results instead
2322 :     of abstract ones.
2323 :    
2324 :     - Provided word-style bit operations etc. for "int" representation
2325 :     types in MLRep.S<Foo>Bitops where <Foo> ranges over Char, Int, Short,
2326 :     and Long.
2327 :    
2328 :     ----------------------------------------------------------------------
2329 :     Name: Matthias Blume
2330 :     Date: 2002/01/18
2331 : blume 1028 Tag: blume-20020118-use-x86-fp
2332 :     Description:
2333 :    
2334 :     Now that x86-fast-fp seems to be working, I turned it back on again
2335 :     by default. (Seems to work fine now, even with the FFI.)
2336 :    
2337 :     Other than that, I added some documentation about the FFI to
2338 :     src/ml-nlffigen/README and updated the FFI test examples in
2339 :     src/ml-nlffi-lib/Tests/*.
2340 :    
2341 :     ----------------------------------------------------------------------
2342 : leunga 1025 Name: Allen Leung
2343 :     Date: 2002/01/17
2344 :     Tag: leunga-20020117-x86-fast-fp-call
2345 :     Description:
2346 :    
2347 :     1. Fixed a problem with handling return fp values when x86's fast fp
2348 :     mode is turned on.
2349 :    
2350 :     2. Minor pretty printing fix for cellset. Print %st(0) as %st(0) instead
2351 :     of %f32.
2352 :    
2353 :     3. Added a constructor INT32lit to the ast of MLRISC tools.
2354 :    
2355 :     ----------------------------------------------------------------------
2356 : blume 1021 Name: Matthias Blume
2357 :     Date: 2002/01/16
2358 :     Tag: blume-20020116-ffifiddle
2359 :     Description:
2360 :    
2361 :     More fiddling with the FFI interface:
2362 :    
2363 :     - Make constness 'c instead of rw wherever possible. This eliminates
2364 :     the need for certain explicit coercions. (However, due to ML's
2365 :     value polymorphism, there will still be many cases where explicit
2366 :     coercions are necessary. Phantom types are not the whole answer
2367 :     to modeling a subtyping relationship in ML.)
2368 :    
2369 :     - ro/rw coersions for pointers added. (Avoids the detour through */&.)
2370 :    
2371 :     - "printf" test example added to src/ml-nlffi-lib/Tests. (Demonstrates
2372 :     clumsy workaround for varargs problem.)
2373 :    
2374 :     ----------------------------------------------------------------------
2375 : george 1016 Name: Lal George
2376 :     Date: 2002/01/15
2377 :     Tag: <none>
2378 :     Description:
2379 :    
2380 :     1. Since COPY instructions are no longer native to the architecture,
2381 :     a generic functor can be used to implement the expandCopies function.
2382 :    
2383 :     2. Allowed EXPORT and IMPORT pseudo-op declarations to appear inside a
2384 :     TEXT segment.
2385 :    
2386 :     ----------------------------------------------------------------------
2387 : blume 1011 Name: Matthias Blume
2388 : blume 1015 Date: 2002/01/15
2389 :     Tag: blume-20020115-ffiupdates
2390 :     Description:
2391 :    
2392 :     1. Fix for bug resulting in single-precision float values being returned
2393 :     incorrectly from FFI calls.
2394 :    
2395 :     2. Small modifications to C FFI API:
2396 :    
2397 :     - memory-allocation routines return straight objects (no options)
2398 :     and raise an exception in out-of-memory situations
2399 :     - unsafe extensions to cast between function pointers and pointers
2400 :     from/to ints
2401 :     - added structure C_Debug as an alternative to structure C where
2402 :     pointer-dereferencing (|*| and |*!) always check for null-pointers
2403 :     - added open_lib' to DynLinkage; open_lib' works like open_lib
2404 :     but also takes a (possibly empty) list of existing library handles
2405 :     that the current library depends on
2406 :    
2407 :     ----------------------------------------------------------------------
2408 :     Name: Matthias Blume
2409 : blume 1011 Date: 2002/01/10
2410 :     Tag: blume-20020110-newffigen
2411 :     Description:
2412 :    
2413 :     1. Updates to portable graph code.
2414 :    
2415 :     2. Major update to ml-nlffigen and ml-nlffi-lib. Things are much
2416 :     more scalable now so that even huge interfaces such as the one
2417 :     for GTK compile in finite time and space. :-)
2418 :     See src/ml-nlffigen/README for details on what's new.
2419 :    
2420 :     ----------------------------------------------------------------------
2421 : george 1003 Name: Lal George
2422 : george 1009 Date: 2001/01/09 14:31:35 EST 2002
2423 :     Tag: george-20011206-rm-native-copy
2424 :     Description:
2425 :    
2426 :     Removed the native COPY and FCOPY instructions
2427 :     from all the architectures and replaced it with the
2428 :     explicit COPY instruction from the previous commit.
2429 :    
2430 :     It is now possible to simplify many of the optimizations
2431 :     modules that manipulate copies. This has not been
2432 :     done in this change.
2433 :    
2434 :     ----------------------------------------------------------------------
2435 :     Name: Lal George
2436 : george 1003 Date: 2001/12/06 16:50:13 EST 2001
2437 :     Tag: george-20011206-mlrisc-instruction
2438 :     Description:
2439 : george 824
2440 : george 1003 Changed the representation of instructions from being fully abstract
2441 :     to being partially concrete. That is to say:
2442 :    
2443 :     from
2444 :     type instruction
2445 :    
2446 :     to
2447 :     type instr (* machine instruction *)
2448 :    
2449 :     datatype instruction =
2450 :     LIVE of {regs: C.cellset, spilled: C.cellset}
2451 :     | KILL of {regs: C.cellset, spilled: C.cellset}
2452 :     | COPYXXX of {k: CB.cellkind, dst: CB.cell list, src: CB.cell list}
2453 :     | ANNOTATION of {i: instruction, a: Annotations.annotation}
2454 :     | INSTR of instr
2455 :    
2456 :     This makes the handling of certain special instructions that appear on
2457 :     all architectures easier and uniform.
2458 :    
2459 :     LIVE and KILL say that a list of registers are live or killed at the
2460 :     program point where they appear. No spill code is generated when an
2461 :     element of the 'regs' field is spilled, but the register is moved to
2462 :     the 'spilled' (which is present, more for debugging than anything else).
2463 :    
2464 :     LIVE replaces the (now deprecated) DEFFREG instruction on the alpha.
2465 :     We used to generate:
2466 :    
2467 :     DEFFREG f1
2468 :     f1 := f2 + f3
2469 :     trapb
2470 :    
2471 :     but now generate:
2472 :    
2473 :     f1 := f2 + f3
2474 :     trapb
2475 :     LIVE {regs=[f1,f2,f3], spilled=[]}
2476 :    
2477 :     Furthermore, the DEFFREG (hack) required that all floating point instruction
2478 :     use all registers mentioned in the instruction. Therefore f1 := f2 + f3,
2479 :     defines f1 and uses [f1,f2,f3]! This hack is no longer required resulting
2480 :     in a cleaner alpha implementation. (Hopefully, intel will not get rid of
2481 :     this architecture).
2482 :    
2483 :     COPYXXX is intended to replace the parallel COPY and FCOPY available on
2484 :     all the architectures. This will result in further simplification of the
2485 :     register allocator that must be aware of them for coalescing purposes, and
2486 :     will also simplify certain aspects of the machine description that provides
2487 :     callbacks related to parallel copies.
2488 :    
2489 :     ANNOTATION should be obvious, and now INSTR represents the honest to God
2490 :     machine instruction set!
2491 :    
2492 :     The <arch>/instructions/<arch>Instr.sml files define certain utility
2493 :     functions for making porting easier -- essentially converting upper case
2494 :     to lower case. All machine instructions (of type instr) are in upper case,
2495 :     and the lower case form generates an MLRISC instruction. For example on
2496 :     the alpha we have:
2497 :    
2498 :     datatype instr =
2499 :     LDA of {r:cell, b:cell, d:operand}
2500 :     | ...
2501 :    
2502 :     val lda : {r:cell, b:cell, d:operand} -> instruction
2503 :     ...
2504 :    
2505 :     where lda is just (INSTR o LDA), etc.
2506 :    
2507 : leunga 815 ----------------------------------------------------------------------
2508 : blume 986 Name: Matthias Blume
2509 : blume 993 Date: 2001/11/22 21:40:00 EST
2510 :     Tag: Release_110_37
2511 :     Description:
2512 :    
2513 :     Release 110.37. This time for real.
2514 :    
2515 :     ----------------------------------------------------------------------
2516 :     Name: Matthias Blume
2517 : blume 988 Date: 2001/11/21 16:35:00 EST
2518 :     Tag: blume-20011121-foot-in-mouth
2519 :     Description:
2520 :    
2521 :     Removed the "Release_110_37" tag because of a serious bug.
2522 :     This will be re-tagged once the bug is fixed.
2523 :    
2524 :     ----------------------------------------------------------------------
2525 :     Name: Matthias Blume
2526 : blume 987 Date: 2001/11/21 16:14:00 EST
2527 :     Tag: blume-20011121-forgottenfile
2528 :     Description:
2529 :    
2530 :     Forgot to add a file. (Just a .tex-file -- part of
2531 :     the CM manual source.)
2532 :    
2533 :     ----------------------------------------------------------------------
2534 :     Name: Matthias Blume
2535 : blume 986 Date: 2001/11/21 16:10:00 EST
2536 : blume 988 Tag: blume-20011121-invalid_110_37
2537 : blume 986 Description:
2538 :    
2539 : blume 988 Note: I removed the original tag "Release_110_37" from this commit
2540 :     because we found a serious bug in all non-x86 backends.
2541 :     - Matthias
2542 :    
2543 : blume 986 1. Modifications to the SML/NJ code generator and to the runtime system
2544 :     so that code object name strings are directly inserted into code
2545 :     objects at code generation time. The only business the runtime system
2546 :     has with this is now to read the name strings on occasions.
2547 :     (The encoding of the name string has also changed somewhat.)
2548 :    
2549 :     2. CM now implements a simple "set calculus" for specifying export lists.
2550 :     In particular, it is now possible to refer to the export lists of
2551 :     other libraries/groups/sources and form unions as well as differences.
2552 :     See the latest CM manual for details.
2553 :    
2554 :     3. An separate notion of "proxy" libraries has again be eliminated from
2555 :     CM's model. (Proxy libraries are now simply a special case of using
2556 :     the export list calculus.)
2557 :    
2558 :     4. Some of the existing libraries now take advantage of the new set
2559 :     calculus.
2560 :     (Notice that not all libraries have been converted because some
2561 :     of the existing .cm-files are supposed to be backward compatible
2562 :     with 110.0.x.)
2563 :    
2564 :     5. Some cleanup in stand-alone programs. (Don't use "exnMessage" -- use
2565 :     "General.exnMessage"! The former relies on a certain hook to be
2566 :     initialized, and that often does not happen in the stand-alone case.)
2567 :    
2568 :     ----------------------------------------------------------------------
2569 : george 984 Name: Lal George
2570 :     Date: 2001/11/21 13:56:18 EST
2571 :     Tag: george-2001121-pseudo-ops
2572 :     Description:
2573 :    
2574 :     Implemented a complete redesign of MLRISC pseudo-ops. Now there
2575 :     ought to never be any question of incompatabilities with
2576 :     pseudo-op syntax expected by host assemblers.
2577 :    
2578 :     For now, only modules supporting GAS syntax are implemented
2579 :     but more should follow, such as MASM, and vendor assembler
2580 :     syntax, e.g. IBM as, Sun as, etc.
2581 :    
2582 :     ----------------------------------------------------------------------
2583 : blume 977 Name: Matthias Blume
2584 :     Date: 2001/11/14 11:52:00 EST
2585 :     Tag: blume-20011114-srcname
2586 :     Description:
2587 :    
2588 :     1. Routed the name of the current source file to mlriscgen where it
2589 :     should be directly emitted into the code object. (This last part
2590 :     is yet to be done.)
2591 :    
2592 :     2. Some cleanup of the pgraph code to make it match the proposal that
2593 :     I put out the other day. (The proposal notwithstanding, things are
2594 :     still in flux here.)
2595 :    
2596 :     ----------------------------------------------------------------------
2597 : george 976 Name: Lal George
2598 :     Date: 2001/11/14 09:44:04 EST
2599 :     Tag:
2600 :     Description:
2601 :    
2602 :     Fix for a backpatching bug reported by Allen.
2603 :    
2604 :     Because the boundary between short and long span-dependent
2605 :     instructions is +/- 128, there are an astounding number of
2606 :     span-dependent instructions whose size is over estimated.
2607 :    
2608 :     Allen came up with the idea of letting the size of span
2609 :     dependent instructions be non-monotonic, for a maxIter
2610 :     number of times, after which the size must be monotonically
2611 :     increasing.
2612 :    
2613 :     This table shows the number of span-dependent instructions
2614 :     whose size was over-estimated as a function of maxIter, for the
2615 :     file Parse/parse/ml.grm.sml:
2616 :    
2617 :     maxIter # of instructions:
2618 :     10 687
2619 :     20 438
2620 :     30 198
2621 :     40 0
2622 :    
2623 :     In compiling the compiler, there is no significant difference in
2624 :     compilation speed between maxIter=10 and maxIter=40. Actually,
2625 :     my measurements showed that maxIter=40 was a tad faster than
2626 :     maxIter=10! Also 96% of the files in the compiler reach a fix
2627 :     point within 13 iterations, so fixing maxIter at 40, while high,
2628 :     is okay.
2629 :    
2630 :     ----------------------------------------------------------------------
2631 : blume 975 Name: Matthias Blume
2632 :     Date: 2001/10/31 15:25:00 EST
2633 :     Tag: blume-20011031-pgraph
2634 :     Description:
2635 :    
2636 :     CKIT:
2637 :     * Changed the "Function" constructor of type Ast.ctype to carry optional
2638 :     argument identifiers.
2639 :     * Changed the return type of TypeUtil.getFunction accordingly.
2640 :     * Type equality ignores the argument names.
2641 :     * TypeUtil.composite tries to preserve argument names but gives up quickly
2642 :     if there is a mismatch.
2643 :    
2644 :     installation script:
2645 :     * attempts to use "curl" if available (unless "wget" is available as well)
2646 :    
2647 :     CM:
2648 :     * has an experimental implementation of "portable graphs" which I will
2649 :     soon propose as an implementation-independent library format
2650 :     * there are also new libraries $/pgraph.cm and $/pgraph-util.cm
2651 :    
2652 :     NLFFI-LIB:
2653 :     * some cleanup (all cosmetic)
2654 :    
2655 :     NLFFIGEN:
2656 :     * temporarily disabled the mechanism that suppresses ML output for
2657 :     C definitions whose identifiers start with an underscore character
2658 :     * generate val bindings for enum constants
2659 :     * user can request that only one style (light or heavy) is being used;
2660 :     default is to use both (command-line arguments: -heavy and -light)
2661 :     * fixed bug in handling of function types involving incomplete pointers
2662 :     * generate ML entry points that take record arguments (i.e., using
2663 :     named arguments) for C functions that have a prototype with named
2664 :     arguments
2665 :     (see changes to CKIT)
2666 :    
2667 :     ----------------------------------------------------------------------
2668 : leunga 972 Name: Allen Leung
2669 :     Date: 2001/10/27 20:34:00 EDT
2670 :     Tag: leunga-20011027-x86-fast-fp-call
2671 :     Description:
2672 :    
2673 :     Fixed the bug described in blume-20010920-slowfp.
2674 :    
2675 :     The fix involves
2676 :     1. generating FCOPYs in FSTP in ia32-svid
2677 :     2. marking a CALL with the appropriate annotation
2678 :    
2679 :     ----------------------------------------------------------------------
2680 : blume 961 Name: Matthias Blume
2681 :     Date: 2001/10/16 11:32:00 EDT
2682 :     Tag: blume-20011016-netbsd
2683 :     Description:
2684 :    
2685 :     Underscore patch from Chris Richards (fixing problem with compiling
2686 :     runtime system under recent NetBSD).
2687 :    
2688 :     ----------------------------------------------------------------------
2689 : george 959 Name: Allen Leung
2690 :     Date: 2001/10/12 17:18:32 EDT 2001
2691 :     Tag: leung-20011012-x86-printflowgraph
2692 :     Description:
2693 :    
2694 :     X86RA now uses a valid (instead of dummy) PrintFlowgraph module.
2695 :    
2696 :     ----------------------------------------------------------------------
2697 : george 952 Name: Lal George
2698 : george 958 Date: 2001/10/11 23:51:34 EDT
2699 :     Tag: george-20011011-too-many-instrs
2700 :     Description:
2701 :    
2702 :     The representation of a program point never expected to see more
2703 :     than 65536 instructions in a basic block!
2704 :    
2705 :     ----------------------------------------------------------------------
2706 :     Name: Lal George
2707 : george 952 Date: 2001/10/09 09:41:37 EDT
2708 :     Tag: george-20011008-mlrisc-labels
2709 :     Description:
2710 :    
2711 :     Changed the machine description files to support printing of
2712 :     local and global labels in assembly code, based on host assembler
2713 :     conventions.
2714 :    
2715 :     ----------------------------------------------------------------------
2716 : blume 916 Name: Matthias Blume
2717 : blume 939 Date: 2001/09/25 15:25:00 EDT
2718 :     Tag: blume-20010925-exninfo
2719 :     Description:
2720 :    
2721 :     I provided a non-hook implementation of exnName (at the toplevel) and
2722 :     made the "dummy" implementation of exnMessage (at the toplevel) more
2723 :     useful: if nothing gets "hooked in", then at least you are going to
2724 :     see the exception name and a message indicating why you don't see more.
2725 :    
2726 :     [For the time being, programs that need exnMessage and want to use
2727 :     ml-build should either use General.exnMessage (strongly recommended) or
2728 :     refer to structure General at some other point so that CM sees a
2729 :     static dependency.]
2730 :    
2731 :     [Similar remarks go for "print" and "use": If you want to use their
2732 :     functionality in stand-alone programs generated by ml-build, then use
2733 :     TextIO.output and Backend.Interact.useFile (from $smlnj/compiler.cm).]
2734 :    
2735 :     ----------------------------------------------------------------------
2736 :     Name: Matthias Blume
2737 : blume 938 Date: 2001/09/20 17:28:00 EDT
2738 :     Tag: blume-20010920-slowfp
2739 :     Description:
2740 :    
2741 :     Allen says that x86-fast-fp is not safe yet, so I turned it off again...
2742 :    
2743 :     ----------------------------------------------------------------------
2744 :     Name: Matthias Blume
2745 : blume 937 Date: 2001/09/20 17:20:00 EDT
2746 :     Tag: blume-20010920-canonicalpaths
2747 :     Description:
2748 :    
2749 :     0. Updated the BOOT file (something that I forgot to do earlier).
2750 :    
2751 :     1. Small internal change to CM so that it avoids "/../" in filenames
2752 :     as much as possible (but only where it is safe).
2753 :    
2754 :     2. Changed config/_run-sml (resulting in a changed bin/.run-sml) so
2755 :     that arguments that contain delimiters are passed through correctly.
2756 :     This change also means that all "special" arguments of the form
2757 :     @SMLxxx... must come first.
2758 :    
2759 :     3. Changed install script to put relative anchor names for tool commands
2760 :     into pathconfig.
2761 :    
2762 :     ----------------------------------------------------------------------
2763 :     Name: Matthias Blume
2764 : blume 932 Date: 2001/09/18 15:35:00 EDT
2765 :     Tag: blume-20010918-readme11036
2766 :     Description:
2767 :    
2768 :     Added README files.
2769 :    
2770 :     ----------------------------------------------------------------------
2771 :     Name: Matthias Blume
2772 : blume 929 Date: 2001/09/18 11:45:00 EDT
2773 :     Tag: Release_110_36 (retag)
2774 :     Description:
2775 :    
2776 :     Fixed mistake in config/preloads. Retagged as 110.36.
2777 :    
2778 :     ----------------------------------------------------------------------
2779 :     Name: Matthias Blume
2780 : blume 928 Date: 2001/09/18 09:40:00 EDT
2781 : blume 929 Tag: Release_110_36_orig (tag changed)
2782 : blume 928 Description:
2783 :    
2784 :     New version (110.36). New bootfiles.
2785 :    
2786 :     ----------------------------------------------------------------------
2787 :     Name: Matthias Blume
2788 : blume 927 Date: 2001/09/14 16:15:00 EDT
2789 :     Tag: blume-20010914-x86fastfp
2790 :     Description:
2791 :    
2792 :     John committed some changes that Allen made, in particular a (hopefully)
2793 :     correctly working version of the x86-fp module.
2794 :    
2795 :     I changed the default setting of the Control.MLRISC.getFlag "x86-fast-fp"
2796 :     flag to "true". Everything seems to compile to a fixpoint ok, and
2797 :     "mandelbrot" speeds up by about 15%.
2798 :    
2799 :     ----------------------------------------------------------------------
2800 :     Name: Matthias Blume
2801 : blume 923 Date: 2001/09/13 11:20:00 EDT
2802 :     Tag: blume-20010913-minimal
2803 :     Description:
2804 :    
2805 :     1. Stefan Monnier's patch to fix a miscompilation problem that
2806 :     was brought to light by John Reppy's work on Moby.
2807 :    
2808 :     2. Implemented a minimal "structure Compiler" that contains just
2809 :     "version" and "architecture". The minimal version will be
2810 :     available when the full version is not. This is for backward-
2811 :     compatibility with code that wants to test Compiler.version.
2812 :    
2813 :     ----------------------------------------------------------------------
2814 :     Name: Matthias Blume
2815 : blume 920 Date: 2001/08/28 14:03:00 EDT
2816 :     Tag: blume-20010828-ml-lex
2817 :     Description:
2818 :    
2819 :     Fix for bug 1581, received from Neophytos Michael.
2820 :    
2821 :     ----------------------------------------------------------------------
2822 :     Name: Matthias Blume
2823 : blume 918 Date: 2001/08/27 11:20:00 EDT
2824 :     Tag: blume-20010827-readme11035
2825 :     Description:
2826 :    
2827 :     Fleshed out the README file for 110.35.
2828 :    
2829 :     ----------------------------------------------------------------------
2830 :     Name: Matthias Blume
2831 : blume 916 Date: 2001/08/24 17:10:00 EDT
2832 :     Tag: Release_110_35
2833 :     Description:
2834 :    
2835 :     New version number (110.35). New bootfiles.
2836 :    
2837 :     ----------------------------------------------------------------------
2838 : george 915 Name: Lal George
2839 : george 909 Date: 2001/08/24 13:47:18 EDT 2001
2840 :     Tag: george-20010824-MLRISC-graphs
2841 :     Description:
2842 :    
2843 :     removed clusters from MLRISC completely and replaced with graphs.
2844 :    
2845 :     ----------------------------------------------------------------------
2846 :     Name: Matthias Blume
2847 : blume 905 Date: 2001/08/23 17:50:00 EDT
2848 :     Tag: blume-20010823-toplevel
2849 :     Description:
2850 :    
2851 :     - some reorganization of the code that implements various kinds of
2852 :     environments in the compiler (static, dynamic, symbolic, combined)
2853 :     - re-implemented the EnvRef module so that evalStream works properly
2854 :     (if the stream contains references to "use", "CM.make", etc.)
2855 :     - cleaned up evalloop.sml and interact.sml (but they need more cleaning)
2856 :    
2857 :     ----------------------------------------------------------------------
2858 :     Name: Matthias Blume
2859 : blume 904 Date: 2001/08/20 15:50 EDT
2860 :     Tag: blume20010820-slipup
2861 :     Description:
2862 :    
2863 :     I forgot to commit a few files. Here they are...
2864 :    
2865 :     ----------------------------------------------------------------------
2866 :     Name: Matthias Blume
2867 : blume 903 Date: 2001/08/20 15:35:00 EDT
2868 :     Tag: blume-20010820-debugprof
2869 :     Description:
2870 :    
2871 :     !!!! NEW BOOTFILES !!!!
2872 :    
2873 :     This is another round of reorganizing the compiler sources. This
2874 :     time the main goal was to factor out all the "instrumentation"
2875 :     passes (for profiling and backtracing) into their own library.
2876 :     The difficulty was to do it in such a way that it does not depend
2877 :     on elaborate.cm but only on elabdata.cm.
2878 :    
2879 :     Therefore there have been further changes to both elaborate.cm and
2880 :     elabdata.cm -- more "generic" things have been moved from the former
2881 :     to the latter. As a result, I was forced to split the assignment
2882 :     of numbers indicating "primtyc"s into two portions: SML-generic and
2883 :     SML/NJ-specific. Since it would have been awkward to maintain,
2884 :     I bit the bullet and actually _changed_ the mapping between these
2885 :     numbers and primtycs. The bottom line of this is that you need
2886 :     a new set of bin- and bootfiles.
2887 :    
2888 :     I have built new bootfiles for all architectures, so doing a fresh
2889 :     checkout and config/install.sh should be all you need.
2890 :    
2891 :     The newly created library's name is
2892 :    
2893 :     $smlnj/viscomp/debugprof.cm
2894 :    
2895 :     and its sources live under
2896 :    
2897 :     src/compiler/DebugProf
2898 :    
2899 :     ----------------------------------------------------------------------
2900 :     Name: Matthias Blume
2901 : blume 902 Date: 2001/08/15 17:15:00 EDT
2902 :     Tag: blume-20010815-compreorg
2903 :     Description:
2904 :    
2905 :     This is a first cut at reorganizing the CM libraries that make up the
2906 :     core of the compiler. The idea is to separate out pieces that could
2907 :     be used independently by tools, e.g., the parser, the typechecker, etc.
2908 :    
2909 :     The current status is a step in this direction, but it is not quite
2910 :     satisfactory yet. Expect more changes in the future.
2911 :    
2912 :     Here is the current (new) organization...
2913 :    
2914 :     What used to be $smlnj/viscomp/core.cm is now divided into
2915 :     six CM libraries:
2916 :    
2917 :     $smlnj/viscomp/basics.cm
2918 :     /parser.cm
2919 :     /elabdata.cm
2920 :     /elaborate.cm
2921 :     /execute.cm
2922 :     /core.cm
2923 :    
2924 :     The CM files for these libraries live under src/system/smlnj/viscomp.
2925 :     All these libraries are proxy libraries that contain precisely
2926 :     one CM library component. Here are the locations of the components
2927 :     (all within the src/compiler tree):
2928 :    
2929 :     Basics/basics.cm
2930 :     Parse/parser.cm
2931 :     ElabData/elabdata.cm
2932 :     Elaborator/elaborate.cm
2933 :     Execution/execute.cm
2934 :     core.cm
2935 :    
2936 :     [This organization is the same that has been used already
2937 :     for a while for the architecture-specific parts of the visible
2938 :     compiler and for the old version of core.cm.]
2939 :    
2940 :     As you will notice, many source files have been moved from their
2941 :     respective original locations to a new home in one of the above
2942 :     subtrees.
2943 :    
2944 :     The division of labor between the new libraries is the following:
2945 :    
2946 :     basics.cm:
2947 :     - Simple, basic definitions that pertain to many (or all) of
2948 :     the other libraries.
2949 :     parser.cm:
2950 :     - The SML parser, producing output of type Ast.dec.
2951 :     - The type family for Ast is also defined and exported here.
2952 :     elabdata.cm:
2953 :     - The datatypes that describe input and output of the elaborator.
2954 :     This includes types, absyn, and static environments.
2955 :     elaborator.cm:
2956 :     - The SML/NJ type checker and elaborator.
2957 :     This maps an Ast.dec (with a given static environment) to
2958 :     an Absyn.dec (with a new static environment).
2959 :     - This libraries implements certain modules that used to be
2960 :     structures as functors (to remove dependencies on FLINT).
2961 :     execute.cm:
2962 :     - Everything having to do with executing binary code objects.
2963 :     - Dynamic environments.
2964 :     core.cm:
2965 :     - SML/NJ-specific instantiations of the elaborator and MLRISC.
2966 :     - Top-level modules.
2967 :     - FLINT (this should eventually become its own library)
2968 :    
2969 :     Notes:
2970 :    
2971 :     I am not 100% happy with the way I separated the elaborator (and its
2972 :     data structures) from FLINT. Two instances of the same problem:
2973 :    
2974 :     1. Data structures contain certain fields that carry FLINT-specific
2975 :     information. I hacked around this using exn and the property list
2976 :     module from smlnj-lib. But the fact that there are middle-end
2977 :     specific fields around at all is a bit annoying.
2978 :    
2979 :     2. The elaborator calculates certain FLINT-related information. I tried
2980 :     to make this as abstract as I could using functorization, but, again,
2981 :     the fact that the elaborator has to perform calculations on behalf
2982 :     of the middle-end at all is not nice.
2983 :    
2984 :     3. Having to used exn and property lists is unfortunate because it
2985 :     weakens type checking. The other alternative (parameterizing
2986 :     nearly *everything*) is not appealing, though.
2987 :    
2988 :     I removed the "rebinding =" warning hack because due to the new organization
2989 :     it was awkward to maintain it. As a result, the compiler now issues some of
2990 :     these warnings when compiling init.cmi during bootstrap compilation. On
2991 :     the plus side, you also get a warning when you do, for example:
2992 :     val op = = Int32.+
2993 :     which was not the case up to now.
2994 :    
2995 :     I placed "assign" and "deref" into the _Core structure so that the
2996 :     code that deals with the "lazy" keyword can find them there. This
2997 :     removes the need for having access to the primitive environment
2998 :     during elaboration.
2999 :    
3000 :     ----------------------------------------------------------------------
3001 :     Name: Matthias Blume
3002 : blume 898 Date: 2001/08/13
3003 :     Tag: blume-20010813-closures
3004 :     Description:
3005 :    
3006 :     This fix was sent to us by Zhong Shao. It is supposed to improve the
3007 :     performance of certain loops by avoiding needless closure allocation.
3008 :    
3009 :     ----------------------------------------------------------------------
3010 : george 890 Name: Lal George
3011 : george 896 Date: 2001/07/31 10:03:23 EDT 2001
3012 :     Tag: george-20010731-x86-fmalloc
3013 :     Description: Fixed bug in x86 calls
3014 :    
3015 :     There was a bug where call instructions would mysteriously
3016 :     vanish. The call instruction had to be one that returned
3017 :     a floating point value.
3018 :    
3019 :     ----------------------------------------------------------------------
3020 :     Name: Lal George
3021 : george 890 Date: 2001/07/19 16:36:29 EDT 2001
3022 :     Tag: george-20010719-simple-cells
3023 :     Description:
3024 :    
3025 :     I have dramatically simplified the interface for CELLS in MLRISC.
3026 :    
3027 :     In summary, the cells interface is broken up into three parts:
3028 :    
3029 :     1. CellsBasis : CELLS_BASIS
3030 :    
3031 :     CellsBasis is a top level structure and common for all
3032 :     architectures. it contains the definitions of basic datatypes
3033 :     and utility functions over these types.
3034 :    
3035 :     2. functor Cells() : CELLS
3036 :    
3037 :     Cells generates an interface for CELLS that incorporates the
3038 :     specific resources on the target architecture, such as the
3039 :     presence of special register classes, their number and size,
3040 :     and various useful substructures.
3041 :    
3042 :     3. <ARCH>CELLS
3043 :    
3044 :     e.g. SparcCells: SPARCCELLS
3045 :    
3046 :     <ARCH>CELLS usually contains additional bindings for special
3047 :     registers on the architecture, such as:
3048 :    
3049 :     val r0 : cell (* register zero *)
3050 :     val y : cell (* Y register *)
3051 :     val psr : cell (* processor status register *)
3052 :     ...
3053 :    
3054 :     The structure returned by applying the Cells functor is opened
3055 :     in this interface.
3056 :    
3057 :     The main implication of all this is that the datatypes for cells is
3058 :     split between CellsBasis and CELLS -- a fairly simple change for user
3059 :     code.
3060 :    
3061 :     In the old scheme the CELLS interface had a definitional binding of
3062 :     the form:
3063 :    
3064 :     signature CELLS = sig
3065 :    
3066 :     structure CellsBasis = CellsBasis
3067 :    
3068 :     ...
3069 :    
3070 :     end
3071 :    
3072 :     With all the sharing constraints that goes on in MLRISC, this old
3073 :     design quickly leads to errors such as:
3074 :    
3075 :     "structure definition spec inside of sharing ... "
3076 :    
3077 :    
3078 :     and appears to require an unacceptable amount of sharing and where
3079 :     constraint hackery.
3080 :    
3081 :     I think this error message (the interaction of definitional specs and
3082 :     sharing) requires more explanation on our web page.
3083 :    
3084 :     ----------------------------------------------------------------------
3085 : blume 826 Name: Matthias Blume
3086 : blume 879 Date: 2001/07/19 15:00:00 EDT
3087 :     Tag: blume-20010719-libreorg
3088 :     Description:
3089 :    
3090 :     This update puts together a fairly extensive but straightforward change
3091 :     to the way the libraries that implement the interactive system are
3092 :     organized:
3093 :    
3094 :     The biggest change is the elimination of structure Compiler. As a
3095 :     replacement for this structure, there is now a CM library
3096 :     (known as $smlnj/compiler.cm or $smlnj/compiler/current.cm)
3097 :     that exports all the substructures of the original structure Compiler
3098 :     directly. So instead of saying Compiler.Foo.bar one now simply
3099 :     says Foo.bar. (The CM libraries actually export a collection of
3100 :     structures that is richer than the collection of substructures of
3101 :     structure Compiler.)
3102 :    
3103 :     To make the transition smooth, there is a separate library called
3104 :     $smlnj/compiler/compiler.cm which puts together and exports the
3105 :     original structure Compiler (or at least something very close to it).
3106 :    
3107 :     There are five members of the original structure Compiler
3108 :     that are not exported directly but which instead became members
3109 :     of a new structure Backend (described by signature BACKEND). These are:
3110 :     structure Profile (: PROFILE), structure Compile (: COMPILE), structure
3111 :     Interact (: INTERACT), structure Machine (: MACHINE), and val
3112 :     architecture (: string).
3113 :    
3114 :     Structure Compiler.Version has become structure CompilerVersion.
3115 :    
3116 :     Cross-compilers for alpha32, hppa, ppc, sparc, and x86 are provided
3117 :     by $smlnj/compiler/<arch>.cm where <arch> is alpha32, hppa, ppc, sparc,
3118 :     or x86, respectively.
3119 :     Each of these exports the same frontend structures that
3120 :     $smlnj/compiler.cm exports. But they do not have a structure Backend
3121 :     and instead export some structure <Arch>Backend where <Arch> is Alpha32,
3122 :     Hppa, PPC, Sparc, or X86, respectively.
3123 :    
3124 :     Library $smlnj/compiler/all.cm exports the union of the exports of
3125 :     $smlnj/compiler/<arch>.cm
3126 :    
3127 :     There are no structures <Arch>Compiler anymore, use
3128 :     $smlnj/compiler/<arch>.cm instead.
3129 :    
3130 :     Library host-compiler-0.cm is gone. Instead, the internal library
3131 :     that instantiates CM is now called cm0.cm. Selection of the host
3132 :     compiler (backend) is no longer done here but. (Responsibility for it
3133 :     now lies with $smlnj/compiler/current.cm. This seems to be more
3134 :     logical.)
3135 :    
3136 :     Many individual files have been moved or renamed. Some files have
3137 :     been split into multiple files, and some "dead" files have been deleted.
3138 :    
3139 :     Aside from these changes to library organization, there are also changes
3140 :     to the way the code itself is organized:
3141 :    
3142 :     Structure Binfile has been re-implemented in such a way that it no
3143 :     longer needs any knowledge of the compiler. It exclusively deals
3144 :     with the details of binfile layout. It no longer invokes the
3145 :     compiler (for the purpose of creating new prospective binfile
3146 :     content), and it no longer has any knowledge of how to interpret
3147 :     pickles.
3148 :    
3149 :     Structure Compile (: COMPILE) has been stripped down to the bare
3150 :     essentials of compilation. It no longer deals with linking/execution.
3151 :     The interface has been cleaned up considerably.
3152 :    
3153 :     Utility routines for dealing with linking and execution have been
3154 :     moved into their own substructures.
3155 :    
3156 :     (The ultimate goal of these changes is to provide a light-weight
3157 :     binfile loader/linker (at least for, e.g., stable libraries) that
3158 :     does not require CM or the compiler to be present.)
3159 :    
3160 :     CM documentation has been updated to reflect the changes to library
3161 :     organization.
3162 :    
3163 :     ----------------------------------------------------------------------
3164 :     Name: Matthias Blume
3165 : blume 875 Date: 2001/07/10 17:30:00 EDT
3166 :     Tag: Release_110_34
3167 :     Description:
3168 :    
3169 : blume 876 Minor tweak to 110.34 (re-tagged):
3170 :    
3171 :     - README.html file added to CVS repository
3172 :     - runtime compiles properly under FreeBSD 3.X and 4.X
3173 :    
3174 :     ----------------------------------------------------------------------
3175 :     Name: Matthias Blume
3176 :     Date: 2001/07/10 17:30:00 EDT
3177 :     Tag: Release_110_34
3178 :     Description:
3179 :    
3180 : blume 875 New version number (110.34). New bootfiles.
3181 :    
3182 :     ----------------------------------------------------------------------
3183 :     Name: Matthias Blume
3184 : blume 874 Date: 2001/07/09 16:00:00 EDT
3185 :     Tag: blume-20010709-more-varargs
3186 :     Description:
3187 :    
3188 :     I changed the handling of varargs in ml-nlffigen again:
3189 :     The ellipsis ... will now simply be ignored (with an accompanying warning).
3190 :    
3191 :     The immediate effect is that you can actually call a varargs function
3192 :     from ML -- but you can't actually supply any arguments beyond the ones
3193 :     specified explicitly. (For example, you can call printf with its format
3194 :     string, but you cannot pass additional arguments.)
3195 :    
3196 :     This behavior is only marginally more useful than the one before, but
3197 :     it has the advantage that a function or, more importantly, a function
3198 :     type never gets dropped on the floor, thus avoiding follow-up problems with
3199 :     other types that refer to the offending one.
3200 :    
3201 :     ----------------------------------------------------------------------
3202 :     Name: Matthias Blume
3203 : blume 873 Date: 2001/07/09 11:25:00 EDT
3204 :     Tag: blume-20010709-varargs
3205 :     Description:
3206 :    
3207 :     1. ckit-lib.cm now exports structure Error
3208 :     2. ml-nlffigen reports occurences of "..." (i.e., varargs function types)
3209 :     with a warning accompanied by a source location. Moreover, it
3210 :     merely skips the offending function or type and proceeds with the
3211 :     rest of its work.u As a result, one can safely feed C code containing
3212 :     "..." to ml-nlffigen.
3213 :     3. There are some internal improvements to CM, providing slightly
3214 :     more general string substitutions in the tools subsystem.
3215 :    
3216 :     ----------------------------------------------------------------------
3217 :     Name: Matthias Blume
3218 : blume 854 Date: 2001/06/27 15:10:00 EDT
3219 :     Tag: blume-20010627-concur
3220 :     Description:
3221 :    
3222 :     Fixed a small bug in CM's handling of parallel compilation.
3223 :     (You could observe the bug by Control-C-interrupting an ordinary
3224 :     CMB.make or CM.stabilize and then attaching some compile servers.
3225 :     The result was that all of a sudden the previously interrupted
3226 :     compilation would continue on its own. This was because of
3227 :     an over-optimization: CM did not bother to clean out certain queues
3228 :     when no servers were attached "anyway", resulting in the contents
3229 :     of these queues to grab control when new servers did get attached.)
3230 :    
3231 :     There is also another minor update to the CM manual.
3232 :    
3233 :     ----------------------------------------------------------------------
3234 :     Name: Matthias Blume
3235 : blume 853 Date: 2001/06/26 16:15:00 EDT
3236 :     Tag: blume-20010626-cmdoc
3237 :     Description:
3238 :    
3239 :     Minor typo fixed in CM manual (syntax diagram for libraries).
3240 :    
3241 :     ----------------------------------------------------------------------
3242 :     Name: Matthias Blume
3243 : blume 852 Date: 2001/06/25 22:55:00 EDT
3244 :     Tag: blume-20010625-x86pc
3245 :     Description:
3246 :    
3247 :     Fixed a nasty bug in the X86 assembly code that caused signal
3248 :     handlers to fail (crash) randomly.
3249 :    
3250 :     ----------------------------------------------------------------------
3251 :     Name: Matthias Blume
3252 : blume 846 Date: 2001/06/25 12:05:00 EDT
3253 :     Tag: blume-20010625-nlffigen
3254 :     Description:
3255 :    
3256 :     This update fixes a number of minor bugs in ml-nlffigen as reported by
3257 :     Nick Carter <nbc@andrew.cmu.edu>.
3258 :    
3259 :     1. Silly but ok typedefs of the form "typedef void myvoid;" are now accepted.
3260 :     2. Default names for generated files are now derived from the name of
3261 :     the C file *without its directory*. In particular, this causes generated
3262 :     files to be placed locally even if the C file is in some system directory.
3263 :     3. Default names for generated signatures and structures are also derived
3264 :     from the C file name without its directory. This avoids silly things
3265 :     like "structure GL/GL".
3266 :     (Other silly names are still possible because ml-nlffigen does not do
3267 : blume 847 a thorough check of whether generated names are legal ML identifiers.
3268 :     When in doubt, use command line arguments to force particular names.)
3269 : blume 846
3270 :     ----------------------------------------------------------------------
3271 :     Name: Matthias Blume
3272 : blume 845 Date: 2001/06/21 12:25:00 EDT
3273 :     Tag: blume-20010621-eXene
3274 :     Description:
3275 :    
3276 :     eXene now compiles and (sort of) works again.
3277 :    
3278 :     The library name (for version > 110.33) is $/eXene.cm.
3279 :    
3280 :     I also added an new example in src/eXene/examples/nbody. See the
3281 :     README file there for details.
3282 :    
3283 :     ----------------------------------------------------------------------
3284 :     Name: Matthias Blume
3285 : blume 844 Date: 2001/06/20 16:40:00 EDT
3286 :     Tag: blume-20010620-cml
3287 :     Description:
3288 :    
3289 :     CML now compiles and works again.
3290 :    
3291 :     Libraries (for version > 110.33):
3292 :    
3293 :     $cml/cml.cm Main CML library.
3294 :     $cml/basis.cm CML's version of $/basis.cm.
3295 :     $cml/cml-internal.cm Internal helper library.
3296 :     $cml/core-cml.cm Internal helper library.
3297 :     $cml-lib/trace-cml.cm Tracing facility.
3298 :     $cml-lib/smlnj-lib.cm CML's version of $/smlnj-lib.cm
3299 :    
3300 :     The installer (config/install.sh) has been taught how to properly
3301 :     install this stuff.
3302 :    
3303 :     ----------------------------------------------------------------------
3304 :     Name: Matthias Blume
3305 : blume 843 Date: 2001/06/19 17:55:00 EDT
3306 :     Tag: blume-20010619-instantiate
3307 :     Description:
3308 :    
3309 :     This un-breaks the fix for bug 1432.
3310 :     (The bug was originally fixed in 110.9 but I broke it again some
3311 :     time after that.)
3312 :    
3313 :     ----------------------------------------------------------------------
3314 :     Name: Matthias Blume
3315 : blume 842 Date: 2001/06/19 17:25:00 EDT
3316 :     Tag: blume-20010619-signals
3317 :     Description:
3318 :    
3319 :     This should (hopefully) fix the long-standing signal handling bug.
3320 :     (The runtime system was constructing a continuation record with an
3321 :     incorrect descriptor which would cause the GC to drop data on the floor...)
3322 :    
3323 :     ----------------------------------------------------------------------
3324 :     Name: Matthias Blume
3325 : blume 840 Date: 2001/06/15 15:05:00 EDT
3326 : blume 841 Tag: blume-20010615-moresparc
3327 :     Description:
3328 :    
3329 :     Here is a short late-hour update related to Sparc c-calls:
3330 :    
3331 :     -- made handling of double-word arguments a bit smarter
3332 :    
3333 :     -- instruction selection phase tries to collapse certain clumsily
3334 :     constructed ML-Trees; typical example:
3335 :    
3336 :     ADD(ty,ADD(_,e,LI d1),LI d2) -> ADD(ty,e,LI(d1+d2))
3337 :    
3338 :     This currently has no further impact on SML/NJ since mlriscGen does
3339 :     not seem to generate such patterns in the first place, and c-calls
3340 :     (which did generate them in the beginning) has meanwhile been fixed
3341 :     so as to avoid them as well.
3342 :    
3343 :     ----------------------------------------------------------------------
3344 :     Name: Matthias Blume
3345 :     Date: 2001/06/15 15:05:00 EDT
3346 : blume 840 Tag: blume-20010615-sparc
3347 :     Description:
3348 :    
3349 :     The purpose of this update is to provide an implementation of NLFFI
3350 :     on Sparc machines.
3351 :    
3352 :     Here are the changes in detail:
3353 :    
3354 :     * src/MLRISC/sparc/c-calls/sparc-c-calls.sml is a new file containing
3355 :     the Sparc implementation of the c-calls API.
3356 :     * The Sparc backend of SML/NJ has been modified to uniformely use %fp
3357 :     for accessing the ML frame. Thus, we have a real frame pointer and
3358 :     can freely modify %sp without need for an omit-frame-ptr phase.
3359 :     The vfp logic in src/compiler/CodeGen/* has been changed to accomodate
3360 :     this case.
3361 :     * ml-nlffigen has been taught to produce code for different architectures
3362 :     and calling conventions.
3363 :     * In a way similar to what was done in the x86 case, the Sparc
3364 :     backend uses its own specific extension to mltree. (For example,
3365 :     it needs to be able to generate UNIMP instructions which are part
3366 :     of the calling convention.)
3367 :     * ml-nlffi-lib was reorganized to make it more modular (in particular,
3368 :     to make it easier to plug in new machine- and os-dependent parts).
3369 :    
3370 :     There are some other fairly unrelated bug fixes and cleanups as well:
3371 :    
3372 :     * I further hacked the .cm files for MLRISC tools (like MDLGen) so
3373 :     that they properly share their libraries with existing SML/NJ libraries.
3374 :     * I fixed a minor cosmetic bug in CM, supressing certain spurious
3375 :     follow-up error messages.
3376 :     * Updates to CM/CMB documentation.
3377 :    
3378 :     TODO items:
3379 :    
3380 :     * MLRISC should use a different register as its asmTemp on the Sparc.
3381 :     (The current %o2 is a really bad choice because it is part of the
3382 :     calling conventions, so things might interfere in unexpected ways.)
3383 :    
3384 :     ----------------------------------------------------------------------
3385 :     Name: Matthias Blume
3386 : blume 839 Date: 2001/06/07
3387 :     Tag: blume-20010607-calls
3388 :     Description:
3389 :    
3390 :     A number of internal changes related to C calls and calling conventions:
3391 :    
3392 :     1. ML-Tree CALL statements now carry a "pops" field. It indicates the
3393 :     number of bytes popped implicitly (by the callee). In most cases
3394 :     this field is 0 but on x86/win32 it is some non-zero value. This
3395 :     is information provided for the benefit of the "omit-frameptr" pass.
3396 :     2. The CALL instruction on the x86 carries a similar "pops" field.
3397 :     The instruction selection phase copies its value from the ML-Tree
3398 :     CALL statement.
3399 :     3. On all other architectures, the instruction selection phase checks
3400 :     whether "pops=0" and complains if not.
3401 :     4. The c-calls implementation for x86 now accepts two calling conventions:
3402 :     "ccall" and "stdcall". When "ccall" is selected, the caller cleans
3403 :     up after the call and pops is set to 0. For "stdcall", the caller
3404 :     does nothing, leaving the cleanup to the callee; pops is set to
3405 :     the number of bytes that were pushed onto the stack.
3406 :     5. The cproto decoder (compiler/Semant/types/cproto.sml) now can
3407 :     distinguish between "ccall" and "stdcall".
3408 :     6. The UNIMP instruction has been added to the supported Sparc instruction
3409 :     set. (This is needed for implementing the official C calling convention
3410 :     on this architecture.)
3411 :     7. I fixed some of the .cm files under src/MLRISC/Tools to make them
3412 :     work with the latest CM.
3413 :    
3414 :     ----------------------------------------------------------------------
3415 :     Name: Matthias Blume
3416 : blume 838 Date: 2001/06/05 15:10:00 EDT
3417 :     Tag: blume-20010605-cm-index
3418 :     Description:
3419 :    
3420 :     0. The "lambdasplit" parameter for class "sml" in CM has been documented.
3421 :    
3422 :     1. CM can now generate "index files". These are human-readable files
3423 :     that list on a per-.cm-file basis each toplevel symbol defined or
3424 :     imported. The location of the index file for
3425 :     <p>/<d>.cm is <p>/CM/INDEX/<d>.cm.
3426 :     To enable index-file generation, set CM.Control.generate_index to true
3427 :     or export an environment-symbol: export CM_GENERATE_INDEX=true.
3428 :    
3429 :     The CM manual has been updated accordingly.
3430 :    
3431 :     2. I made some slight modifications to the c-calls API in MLRISC.
3432 :    
3433 :     a) There is now a callback to support saving/restoring of
3434 :     dedicated but caller-save registers around the actual call
3435 :     instruction.
3436 :     b) One can optionally specify a comment-annotation for the
3437 :     call instruction.
3438 :    
3439 :     3. SML/NJ (mlriscGen.sml) uses this new API for the rawccall primop.
3440 :     (For example, the comment annotation shows the C prototype of
3441 :     the function being called.)
3442 :    
3443 :     ----------------------------------------------------------------------
3444 :     Name: Matthias Blume
3445 : blume 837 Date: 2001/06/01 13:30:00 EDT
3446 :     Tag: blume-20010601-nlffi-cleanup
3447 :     Description:
3448 :    
3449 :     This is mostly a cleanup of MLFFI stuff:
3450 :    
3451 :     - some signature files have been put into a more exposed place
3452 :     - the ugly 'f type parameter is gone (simplifies types tremendously!)
3453 :     - ml-nlffigen changed accordingly
3454 :     - tutorial updated
3455 :    
3456 :     Other changes:
3457 :    
3458 :     - author's affiliation in CM manual(s) updated
3459 :     - some more recognized keywords added to Allen's sml.sty
3460 :    
3461 :     ----------------------------------------------------------------------
3462 :     Name: Matthias Blume
3463 : blume 836 Date: 2001/05/25 15:30:00 EDT
3464 :     Tag: blume-20010525-iptr
3465 :     Description:
3466 :    
3467 :     - put the official 110.33-README (as it appears on the ftp server) under
3468 :     CVS
3469 :     - fixed a small bug related to incomplete pointer types in
3470 :     ml-nlffigen
3471 :     - small cosmetic change to the ml-nlffi-lib's "arr" type constructor
3472 :     (it does not need the 'f type parameter)
3473 :    
3474 :     ----------------------------------------------------------------------
3475 :     Name: Matthias Blume
3476 : blume 834 Date: 2001/05/23 14:30:00 EDT
3477 :     Tag: Release_110_33
3478 :     Description:
3479 :    
3480 :     New version number (110.33). New bootfiles.
3481 :    
3482 :     ----------------------------------------------------------------------
3483 :     Name: Matthias Blume
3484 : blume 833 Date: 2001/05/22 18:06:00 EDT
3485 :     Tag: blume-20010522-targets
3486 :     Description:
3487 :    
3488 :     Made install.sh use file config/targets.customized if it exists, falling
3489 :     back to config/targets if it doesn't. This way one can have a customized
3490 :     version of the targets file without touching the "real thing", thus
3491 :     eliminating the constant fear of accidentally checking something bogus
3492 :     back into the CVS repository... (File config/targets.customized must
3493 :     not be added to the repository!)
3494 :    
3495 :     ----------------------------------------------------------------------
3496 :     Name: Matthias Blume
3497 : blume 832 Date: 2001/05/22 16:30:00 EDT
3498 :     Tag: blume-20010522-minitut
3499 :     Description:
3500 :    
3501 :     1. Bug fix in ml-nlffigen; now (hopefully) correctly handling
3502 :     struct returns.
3503 :     2. Added src/ml-nlffi-lib/Doc/mini-tutorial.txt. This is some very
3504 :     incomplete, preliminary documentation for NLFFI.
3505 :    
3506 :     ----------------------------------------------------------------------
3507 :     Name: Matthias Blume
3508 : blume 829 Date: 2001/05/14 11:30:00 EDT
3509 :     Tag: blume-20010514-script
3510 :     Description:
3511 :    
3512 :     Some bugs in install script fixed.
3513 :    
3514 :     In addition to that I also made a slight change to the NLFFI API:
3515 :     Functors generated by ml-nlffigen now take the dynamic library as a
3516 :     straight functor argument, not as a suspended one. (The original
3517 :     functor code used to force the suspension right away anyway, so there
3518 :     was nothing gained by this complication of the interface.)
3519 :    
3520 :     ----------------------------------------------------------------------
3521 :     Name: Matthias Blume
3522 : blume 828 Date: 2001/05/11 14:35:00 EDT
3523 :     Tag: blume-20010511-ml-nlffi
3524 :     Description:
3525 :    
3526 :     I finally took the plunge and added my new FFI code to the main
3527 :     repository. For x86-linux it is now ready for prime-time.
3528 :    
3529 :     There are two new subdirectories of "src":
3530 :    
3531 :     - ml-nlffi-lib:
3532 :     The utility library for programs using the FFI interface.
3533 :     Here is the implementation of $/c.cm and its associated low-level
3534 :     partners $/c-int.cm and $/memory.cm.
3535 :     - ml-nlffigen:
3536 :     A stand-alone program for generating ML glue code from C source
3537 :     code.
3538 :    
3539 :     Building ml-nlffigen requires $/ckit-lib.cm.
3540 :    
3541 :     The config/install.sh script has been updates to do the Right Thing
3542 :     (hopefully).
3543 :    
3544 :     Notice that the source tree for the C-Kit will not be put under "src"
3545 :     but directly under the installation root directory. (This is the
3546 :     structure that currently exists on the CVS server when you check out
3547 :     module "sml".) Fortunately, config/install.sh knows about this oddity.
3548 :    
3549 :     Bugs: No documentation yet.
3550 :    
3551 :     ----------------------------------------------------------------------
3552 :     Name: Matthias Blume
3553 : blume 826 Date: 2001/05/09 16:35:00 EDT
3554 :     Tag: blume-20010509-cpscontract
3555 :     Description:
3556 :    
3557 :     Fixed a bug in the accounting code in cpsopt/contract.sml. (The
3558 :     wrapper/unwrapper elimination did not decrement usage counts and some
3559 :     dead variables got overlooked by the dead-up logic.)
3560 :    
3561 :     ----------------------------------------------------------------------
3562 : george 824 Name: Lal George
3563 :     Date: 2001/05/08 17:26:09 EDT
3564 :     Tag: george-20010508-omit-frameptr
3565 :     Description:
3566 :    
3567 :     Changes to implement the omit-frame-pointer optimization to support
3568 :     raw C calls. For now, there is only support on the Intel x86, but
3569 :     other architectures will follow as more experience is gained with this.
3570 :    
3571 :    
3572 :     ----------------------------------------------------------------------
3573 : blume 818 Name: Matthias Blume
3574 : blume 819 Date: 2001/05/07 14:40:00 EDT
3575 :     Tag: blume-20010507-proxies
3576 :     Description:
3577 :    
3578 :     I made into "proxy libraries" all libraries that qualify for such a
3579 :     change. (A qualifying library is a library that has another library or
3580 :     groups as its sole member and repeats that member's export list
3581 :     verbatim. A proxy library avoids this repetition by omitting its export
3582 :     list, effectively inheriting the list that its (only) member exports.
3583 :     See the CM manual for more explanation.)
3584 :     The main effect is that explicit export lists for these libraries
3585 :     do not have to be kepts in sync, making maintenance a bit easier.
3586 :    
3587 :     I also added copyright notices to many .cm-files.
3588 :    
3589 :     Last but not least, I made a new set of bootfiles.
3590 :    
3591 :     ----------------------------------------------------------------------
3592 :     Name: Matthias Blume
3593 : blume 818 Date: 2001/05/04 17:00:00 EDT
3594 :     Tag: blume-20010504-cm-lsplit
3595 :     Description:
3596 :    
3597 :     0. John merged pending changes to $/smlnj-lib.cm
3598 :    
3599 :     1. Allen's previous change accidentally backed out of one of Lal's
3600 :     earlier changes. I undid this mistake (re-introducing Lal's change).
3601 :    
3602 :     2. I used the new topOrder' function from graph-scc.sml (from $/smlnj-lib.cm)
3603 :     within the compiler where applicable. There is some code simplification
3604 :     because of that.
3605 :    
3606 :     3. The "split" phase (in FLINT) is now part of the default list of phases.
3607 :     Compiler.Control.LambdaSplitting.* can be used to globally control the
3608 :     lambda-splitting (cross-module-inlining) engine. In addition to that,
3609 :     it can now also be controlled on a per-source basis: CM has been taught
3610 :     a new tool parameter applicable to ML source files.
3611 :    
3612 :     - To turn lambda-splitting off completely:
3613 :     local open Compiler.Control.LambdaSplitting in
3614 :     val _ = set Off
3615 :     end
3616 :     - To make "no lambda-splitting" the global default (but allow per-source
3617 :     overriding); this is the initial setting:
3618 :     local open Compiler.Control.LambdaSplitting in
3619 :     val _ = set (Default NONE)
3620 :     end
3621 :     - To make "lambda-splitting with aggressiveness a" the global default
3622 :     (and allow per-source overriding):
3623 :     local open Compiler.Control.LambdaSplitting in
3624 :     val _ = set (Default (SOME a))
3625 :     end
3626 :    
3627 :     - To turn lambda-splitting off for a given ML souce file (say: a.sml)
3628 :     write (in the respective .cm-file):
3629 :     a.sml (lambdasplitting:off)
3630 :     - To turn lambda-splitting for a.sml on with minimal aggressiveness:
3631 :     a.sml (lambdasplitting:on)
3632 :     - To turn lambda-splitting for a.sml on with aggressiveness <a> (where
3633 :     <a> is a decimal non-negative integer):
3634 :     a.sml (lambdasplitting:<a>)
3635 :     - To turn lambda-splitting for a.sml on with maximal aggressiveness:
3636 :     a.sml (lambdasplitting:infinity)
3637 :     - To use the global default for a.sml:
3638 :     a.sml (lambdasplitting:default)
3639 :     or simply
3640 :     a.sml
3641 :    
3642 :     ----------------------------------------------------------------------
3643 : leunga 815 Name: Allen Leung
3644 :     Date: 2001/05/04 01:57:00 EDT
3645 :     Tag: leunga-20010504-sync
3646 :     Description:
3647 : leunga 788
3648 : leunga 815 MLRISC features.
3649 :    
3650 :     1. Fix to CMPXCHG instructions.
3651 :     2. Changed RA interface to allow annotations in callbacks.
3652 :     3. Added a new method to the stream interface to allow annotations updates.
3653 :    
3654 : george 749 ----------------------------------------------------------------------
3655 : blume 812 Name: Matthias Blume
3656 : blume 814 Date: 2001/05/01 11:45:00 EDT
3657 :     Tag: blume-20010501-pcedittmp
3658 :     Description:
3659 :    
3660 :     Changed install.sh to use the current working directory instead of
3661 :     /usr/tmp for a temporary file (pcedittmp). The previous choice
3662 :     of /usr/tmp caused trouble with MacOS X because of file premission
3663 :     problems.
3664 :    
3665 :     ----------------------------------------------------------------------
3666 :     Name: Matthias Blume
3667 : blume 812 Date: 2001/04/20 11:10:00 EDT
3668 :     Tag: blume-20010420-inMLflag
3669 :     Description:
3670 :    
3671 :     - added vp_limitPtrMask to vproc-state.h
3672 :     (for use by the raw-C-calls mechanism to implement proper interrupt
3673 :     handling)
3674 :     - made the ML compiler aware of various data-structure offsets so it
3675 :     can generate code for accessing the vp_inML flag and vp_limitPtrMask
3676 :     - tweaked mlriscGen.sml to have it emit interrupt-handling code for
3677 :     raw C-calls
3678 :    
3679 :     ----------------------------------------------------------------------
3680 : george 810 Name: Lal George
3681 :     Date: 2001/04/20 09:15:28 EDT
3682 :     Tag: george-20010420-macosX
3683 :     Description:
3684 :    
3685 :     - Changes to port to Mac OS X; Darwin.
3686 :    
3687 :     - In the process I found that sqrt was broken on the PPC, because the
3688 :     fsqrt instruction is not implemented.
3689 :    
3690 :     ----------------------------------------------------------------------
3691 : blume 799 Name: Matthias Blume
3692 : blume 808 Date: 2001/04/18 12:45:00 EDT
3693 :     Tag: blume-20010418-ccalls
3694 :     Description:
3695 :    
3696 :     - fixed two off-by-4 errors in the x86-specific c-calls implementation
3697 :     (this bug prevented structure arguments containing pointers from being
3698 :     passed correctly)
3699 :     - changed the raw-C-call code in mlriscGen.sml in such a way that
3700 :     structure arguments are represented as a pointer to the beginning
3701 :     of the structure (instead of having a series of synthesized arguments,
3702 :     one for each structure member)
3703 :    
3704 :     - made makeml script's verbosity level configurable via environment
3705 :     variable (MAKEML_VERBOSITY)
3706 :    
3707 : blume 813 - eliminated placeholder implementations for f32l, w16s, i16s, and f32s
3708 : blume 808 in rawmem-x86.sml; we are now using the real thing
3709 :    
3710 :     ----------------------------------------------------------------------
3711 :     Name: Matthias Blume
3712 : blume 806 Date: 2001/03/22 16:25:00 EST
3713 :     Tag: blume-20010322-bootfiles
3714 :     Description:
3715 :    
3716 :     Created a new set of bootfiles (for your automatic installation convenience).
3717 :    
3718 :     ----------------------------------------------------------------------
3719 :     Name: Matthias Blume
3720 : blume 805 Date: 2001/03/22 15:10:00 EST
3721 :     Tag: blume-20010322-rawmem-parcm
3722 :     Description:
3723 :    
3724 :     1. All "raw memory access" primitives for the new FFI are implemented now
3725 :     (at least on the x86).
3726 :     2. Some further cleanup of CM's parallel make mechanism.
3727 :    
3728 :     ----------------------------------------------------------------------
3729 :     Name: Matthias Blume
3730 : blume 801 Date: 2001/03/19 17:53:00 EST
3731 :     Tag: blume-20010319-parallel
3732 :     Description:
3733 :    
3734 :     Parallel make (using compile servers) now works again.
3735 :    
3736 :     To this end, CM.stabilize and CMB.make have been modified to work in
3737 :     two passes when compile servers are attached:
3738 :     1. Compile everything, do not perform stabilization; this pass
3739 :     uses compile servers
3740 :     2. Stabilize everything; this pass does not use compile servers
3741 :     If there are no compile servers, the two passes are combined into one
3742 :     (as before). Splitting the passes increases the inherent parallelism
3743 :     in the dependency graph because the entire graph including all
3744 :     libraries is available at the same time. This, in turn, improves
3745 :     server utilization. The downside is that the master process will
3746 :     have to do some extra work after compilation is done (because for
3747 :     technical reasons it must re-read all the binfiles during stabilization).
3748 :    
3749 :     ----------------------------------------------------------------------
3750 :     Name: Matthias Blume
3751 : blume 800 Date: 2001/03/16 12:22:00 EST
3752 :     Tag: blume-20010316-bootfiles
3753 :     Description:
3754 :    
3755 :     Created a new set of bootfiles (for your automatic installation convenience).
3756 :    
3757 :     ----------------------------------------------------------------------
3758 :     Name: Matthias Blume
3759 : blume 799 Date: 2001/03/16 11:00:00 EST
3760 :     Tag: blume-20010316-MLTREE-fixup
3761 :     Description:
3762 :    
3763 :     This is a minor fixup for an (untagged) earlier commit by Allen.
3764 :     (A file was missing).
3765 :    
3766 :     ----------------------------------------------------------------------
3767 : leunga 788 Name: Allen Leung
3768 : leunga 796 Date: Mon Mar 5 18:54:57 EST 2001
3769 :     Tag: leunga-20010305-cut-support
3770 :    
3771 :     1. New support for alternative control-flow in MLTREE.
3772 :     Currently we support
3773 :    
3774 :     FLOW_TO(CALL ...., [k1,...,kn])
3775 :    
3776 :     This is needed for 'cuts to' in C-- and try/handle-like constructs
3777 :     in Moby
3778 :    
3779 :     New assembler flag "asm-show-cutsto" to turn on control-flow debugging.
3780 :    
3781 :     2. Register Allocator
3782 :    
3783 :     Changes in interface [from Fermin, John]
3784 :    
3785 :     3. Alpha 8-bit SLL support [Fermin]
3786 :    
3787 :     4. All architectures
3788 :    
3789 :     A new module (ClusterExpandCopies) for expanding parallel copies.
3790 :    
3791 :     ----------------------------------------------------------------------
3792 :     Name: Allen Leung
3793 : leunga 788 Date: 2001/02/27 23:07:00 EST
3794 :     Tag: leunga-20010227-minor-stuff
3795 :    
3796 :     1. Alpha bug fix for CMOVNE
3797 :     2. Handle mltree COND(..,FCMP ...,...)
3798 :     3. Bug fix in simplifier
3799 :    
3800 :     ----------------------------------------------------------------------
3801 : blume 777 Name: Matthias Blume
3802 : blume 781 Date: 2001/01/30 17:50:00 EST
3803 :     Tag: blume-20010130-sync
3804 :     Description:
3805 :    
3806 :     This is just a minor update to sync my devel branch with the main brach.
3807 :     The only visible change is the addition of some README files.
3808 :    
3809 :     ----------------------------------------------------------------------
3810 :     Name: Matthias Blume
3811 : blume 778 Date: 2001/01/12 23:30:00 JST
3812 :     Tag: blume-20010112-bootfiles
3813 :     Description:
3814 :    
3815 :     Made a new set of bootfiles that goes with the current state of the
3816 :     repository.
3817 :    
3818 :     ----------------------------------------------------------------------
3819 :     Name: Matthias Blume
3820 : blume 777 Date: 2001/01/12 21:20:00 JST
3821 :     Tag: blume-20010112-sync
3822 :     Description:
3823 :    
3824 :     I am just flushing out some minor changes that had accumulated in
3825 :     my private branch in order to sync with the main tree. (This is
3826 :     mainly because I had CVS trouble when trying to merge _into_ my
3827 :     private branch.)
3828 :    
3829 :     Most people should be completely unaffected by this.
3830 :    
3831 :     ----------------------------------------------------------------------
3832 : leunga 775 Name: Allen Leung
3833 :     Date: Thu Jan 11 21:03:00 EST 2001
3834 :     Tag: leunga-20010111-labexp=mltree
3835 :     Description:
3836 :    
3837 :     1. Removed the type LabelExp and replace it by MLTree.
3838 :     2. Rewritten mltree-simplify with the pattern matcher tool.
3839 :     3. There were some bugs in alpha code generator which would break
3840 :     64-bit code generation.
3841 :     4. Redo the tools to generate code with the
3842 :     5. The CM files in MLRISC (and in src/system/smlnj/MLRISC)
3843 :     are now generated by perl scripts.
3844 :    
3845 :     ----------------------------------------------------------------------
3846 : blume 771 Name: Matthias Blume
3847 : blume 774 Date: 2001/01/10 21:55:00 JST
3848 :     Tag: blume-20010110-rcc
3849 :     Description:
3850 :    
3851 :     The RCC stuff now seems to work (but only on the x86).
3852 :     This required hacking of the c-calls interface (and -implementation) in
3853 :     MLRISC.
3854 :    
3855 :     Normal compiler users should be unaffected.
3856 :    
3857 :     ----------------------------------------------------------------------
3858 :     Name: Matthias Blume
3859 : blume 773 Date: 2001/01/09 01:20:00 JST
3860 :     Tag: blume-20010109-rcc
3861 :     Description:
3862 :    
3863 :     This is a fairly big patch, flushing out a large number of pending
3864 :     changes that I made to my development copy over the last couple of days.
3865 :    
3866 :     Of practical relevance at this moment is a workaround for a pickling
3867 :     bug that Allen ran into the other day. The cause of the bug itself is
3868 :     still unknown and it might be hard to fix it properly, but the
3869 :     workaround has some merits of its own (namely somewhat reducing pickling
3870 :     overhead for certain libraries). Therefore, I think this solution should
3871 :     be satisfactory at this time.
3872 :    
3873 :     The rest of the changes (i.e., the vast majority) has to do with my
3874 :     ongoing efforts of providing direct support for C function calls from
3875 :     ML. At the moment there is a new primop "RAW_CCALL", typing magic
3876 :     in types/cproto.sml (invoked from FLINT/trans/translate.sml), a new
3877 :     case in the FLINT CPS datatype (RCC), changes to cps/convert.sml to
3878 :     translate uses of RAW_CCALL into RCC, and changes to mlriscGen.sml to
3879 :     handle RCC.
3880 :    
3881 :     The last part (the changes to mlriscGen.sml) are still known to be
3882 :     wrong on the x86 and not implemented on all other architectures. But
3883 :     the infrastructure is in place. I had to change a few functor
3884 :     signatures in the backend to be able to route the CCalls interface
3885 :     from MLRISC there, and I had to specialize the mltree type (on the
3886 :     x86) to include the necessary extensions. (The extensions themselves
3887 :     were already there and redy to go in MLRISC/x86).
3888 :    
3889 :     Everything should be very happy as soon as someone helps me with
3890 :     mlriscGen.sml...
3891 :    
3892 :     In any case, nothing of this should matter to anyone as long as the
3893 :     new primop is not being used (which is going to be the case unless you
3894 :     find it where I hid it :). The rest of the compiler is completely
3895 :     unaffected.
3896 :    
3897 :     ----------------------------------------------------------------------
3898 :     Name: Matthias Blume
3899 : blume 772 Date: 2001/01/05 00:30:00 JST
3900 :     Tag: blume-20010105-primops
3901 :     Description:
3902 :    
3903 :     Added some experimental support for work that I am doing right now.
3904 :     These changes mostly concern added primops, but there is also a new
3905 :     experimental C library in the runtime system (but currently not enabled
3906 :     anywhere except on Linux/X86).
3907 :    
3908 :     In the course of adding primops (and playing with them), I discovered that
3909 :     Zhong's INL_PRIM hack (no type info for certain primops) was, in fact, badly
3910 :     broken. (Zhong was very right he labeled this stuff as "major gross hack".)
3911 :     To recover, I made type information in INL_PRIM mandatory and changed
3912 :     prim.sml as well as built-in.sml accordingly. The InLine structure now
3913 :     has complete, correct type information (i.e., no bottom types).
3914 :    
3915 :     Since all these changes mean that we need new binfiles, I also bumped the
3916 :     version number to 110.32.1.
3917 :    
3918 :     ----------------------------------------------------------------------
3919 :     Name: Matthias Blume
3920 : blume 771 Date: 2000/12/30 22:10:00 JST
3921 :     Tag: blume-20001230-various
3922 :     Description:
3923 :    
3924 :     Added proxy libraries for MLRISC and let MLRISC libraries refer
3925 :     to each other using path anchors. (See CM manual for explanation.)
3926 :    
3927 :     Updated CM documentation.
3928 :    
3929 :     Fixed some bugs in CM.
3930 :    
3931 :     Implemented "proxy" libraries (= syntactic sugar for CM).
3932 :    
3933 :     Added "-quiet" option to makeml and changed runtime system accordingly.
3934 :    
3935 :     Added cleanup handler for exportML to reset timers and compiler stats.
3936 :    
3937 :     ----------------------------------------------------------------------
3938 : george 761 Name: Lal George
3939 :     Date: 2000/12/22 22:22:58 EST 2000
3940 :     Tag: Release_110_32
3941 :     Description:
3942 :    
3943 :     Infinite precision used throughout MLRISC.
3944 :     see MLRISC/mltree/machine-int.sig
3945 :    
3946 :     ----------------------------------------------------------------------
3947 : blume 756 Name: Matthias Blume
3948 : blume 760 Date: 2000/12/22 23:16:00 JST
3949 :     Tag: blume-20001222-warn
3950 :     Description:
3951 : blume 759
3952 : blume 760 Corrected wording and formatting of some CM warning message which I
3953 :     broke in my previous patch.
3954 : blume 759
3955 :     ----------------------------------------------------------------------
3956 :     Name: Matthias Blume
3957 : blume 763 Date: 2000/12/22 21:20:00 JST
3958 :     Tag: blume-20001222-anchorenv
3959 :     Description:
3960 :    
3961 :     Fixed CM's handling of anchor environments in connection with CMB.make.
3962 :    
3963 :     ----------------------------------------------------------------------
3964 :     Name: Matthias Blume
3965 : blume 758 Date: 2000/12/22 13:15:00 JST
3966 :     Tag: blume-20001222-cleanup
3967 :     Description:
3968 :    
3969 :     Removed src/cm/ffi which does not (and did not) belong here.
3970 :    
3971 :     ----------------------------------------------------------------------
3972 :     Name: Matthias Blume
3973 : blume 757 Date: 2000/12/21 23:55:00 JST
3974 :     Tag: blume-20001221-exn
3975 :     Description:
3976 :    
3977 :     Probably most important: CM no longer silently swallows all exceptions
3978 :     in the compiler.
3979 :     Plus: some other minor CM changes. For example, CM now reports some
3980 :     sizes for generated binfiles (code, data, envpickle, lambdapickle).
3981 :    
3982 :     ----------------------------------------------------------------------
3983 :     Name: Matthias Blume
3984 : blume 756 Date: 2000/12/15 00:01:05 JST
3985 :     Tag: blume-20001215-dirtool
3986 :     Description:
3987 :    
3988 :     - "dir" tool added.
3989 :     - improvements and cleanup to Tools structure
3990 :     - documentation updates
3991 :    
3992 :     ----------------------------------------------------------------------
3993 : leunga 752 Name: Allen Leung
3994 : leunga 755 Date: Thu Dec 14 03:45:24 EST 2000
3995 :     Description:
3996 :     Tag: leunga-20001214-int-inf
3997 :     Description:
3998 :    
3999 :     In IntInf, added these standard functions, which are missing from our
4000 :     implementation:
4001 :    
4002 :     andb : int * int -> int
4003 :     xorb : int * int -> int
4004 :     orb : int * int -> int
4005 :     notb : int -> int
4006 :     << : int * word -> int
4007 :     ~>> : int * word -> int
4008 :    
4009 :     Not tested, I hope they are correct.
4010 :    
4011 :     ----------------------------------------------------------------------
4012 :     Name: Allen Leung
4013 : leunga 752 Date: Fri Dec 8 19:23:26 EST 2000
4014 :     Description:
4015 :     Tag: leunga-20001208-nowhere
4016 :     Description:
4017 :    
4018 :     Slight improvements to the 'nowhere' tool to handle OR-patterns,
4019 :     to generate better error messages etc. Plus a brief manual.
4020 :    
4021 :     ----------------------------------------------------------------------
4022 : george 749 Name: Lal George
4023 :     Date: 2000/12/08 09:54:02 EST 2000
4024 :     Tag: Release_110_31
4025 :     Description:
4026 : george 705
4027 : george 749 - Version 110.31
4028 : leunga 591 ----------------------------------------------------------------------
4029 : leunga 744 Name: Allen Leung
4030 :     Date: Thu Dec 7 22:01:04 EST 2000
4031 :     Tag: leunga-20001207-cell-monster-hack
4032 :     Description:
4033 :    
4034 :     Major MLRISC internal changes. Affect all clients.
4035 :     Summary:
4036 :    
4037 :     1. Type CELLS.cell = int is now replaced by a datatype.
4038 :     As a result, the old regmap is now gone. Almost all interfaces
4039 :     in MLRISC change as a consequence.
4040 :    
4041 :     2. A new brand version of machine description tool (v3.0) that generates
4042 :     modules expecting the new interface. The old version is removed.
4043 :    
4044 :     3. The RA interface has been further abstracted into two new functors.
4045 :     RISC_RA and X86RA. These functors have much simpler interfaces.
4046 :     [See also directory MLRISC/demo.]
4047 :    
4048 :     4. Some other new source->source code generation tools are available:
4049 :    
4050 :     a. MLRISC/Tools/RewriteGen -- generate rewriters from rules.
4051 :     b. MLRISC/Tools/WhereGen -- expands conditional pattern matching rules.
4052 :     I use this tool to generate the peephole optimizers---with the new
4053 :     cell type changes, peephole rules are becoming difficult to write
4054 :     without conditional pattern matching.
4055 :    
4056 :     5. More Intmap -> IntHashTable change. Previous changes by Matthias didn't
4057 :     cover the entire MLRISC source tree so many things broke.
4058 :    
4059 :     6. CM files have been moved to the subdirectory MLRISC/cm.
4060 :     They are moved because there are a lot of them and they clutter up the
4061 :     root dir.
4062 :    
4063 :     7. More detailed documentation to come...
4064 :    
4065 :     NOTE: To rebuild from 110.30 (ftp distribution), you'll have to do
4066 :     a makeml -rebuild first. This is because of other other
4067 :     changes that Matthias has made (see below).
4068 :    
4069 :    
4070 :     ----------------------------------------------------------------------
4071 : blume 733 Name: Matthias Blume
4072 : blume 742 Date: 2000/11/30 23:12:00 JST
4073 :     Tag: blume-20001130-filereorg
4074 :     Description:
4075 :    
4076 :     Some manual updates and some file reorganizations in CM.
4077 :    
4078 :     ----------------------------------------------------------------------
4079 :     Name: Matthias Blume
4080 : blume 737 Date: 2000/11/24 17:45:00 JST
4081 :     Tag: blume-20001124-link
4082 :     Description:
4083 :    
4084 :     Drastically improved link traversal code for the case that the dynamic
4085 :     value was already loaded at bootstrap time. As a result, CM and CMB
4086 :     now both load blazingly fast -- even on a very slow machine. Also,
4087 :     memory consumption has been further reduced by this.
4088 :    
4089 :     Warning: The format of the PIDMAP file has changed. THerefore, to
4090 :     bootstrap you have to do this:
4091 :    
4092 :     1. Run CMB.make
4093 :     2. Make a symbolic link for the boot directory:
4094 :     ln -s sml.boot.ARCH-OS xxx
4095 :     3. "Rebuild" the boot directory:
4096 :     ./makeml -boot xxx -rebuild sml ; rm xxx
4097 :     4. Boot normally:
4098 :     ./makeml
4099 :    
4100 :     ----------------------------------------------------------------------
4101 :     Name: Matthias Blume
4102 : blume 735 Date: 2000/11/21 21:20:00 JST
4103 :     Tag: blume-20001121-tools
4104 :     Description:
4105 :    
4106 :     Continued hacking on autoloading problem -- with success this time.
4107 :     Also changed tool-plugin mechanism. See new CM manual.
4108 :    
4109 :     ----------------------------------------------------------------------
4110 :     Name: Matthias Blume
4111 : blume 734 Date: 2000/11/19 14:30:00 JST
4112 :     Tag: blume-20001119-autoload
4113 :     Description:
4114 :    
4115 :     Some hacking to make autoloading faster. Success for CMB, no success
4116 :     so far for CM. There is a reduced structure CM' that autoloads faster.
4117 :     (This is a temporary, non-documented hack to be eliminated again when
4118 :     the general problem is solved.)
4119 :    
4120 :     ----------------------------------------------------------------------
4121 :     Name: Matthias Blume
4122 : blume 733 Date: 2000/11/17 14:10:00 JST
4123 :     Tag: blume-20001117-pickle-lib
4124 :     Description:
4125 :    
4126 :     1. Eliminated comp-lib.cm
4127 :     2. Made pickle-lib.cm
4128 :     3. Eliminated all uses of intset.sml (from comp-lib.cm)
4129 :     4. Replaced all uses of intmap.{sig,sml} (from comp-lib.cm) with
4130 :     equivalent constructs from smlnj-lib.cm (INtHashTable).
4131 :     5. Point 4. also goes for those uses of intmap.* in MLRISC.
4132 :     Duplicated intmap modules thrown out.
4133 :     6. Hunted down all duplicated SCC code and replaced it with
4134 :     equivalent stuff (GraphSCCFn from smlnj-lib.cm).
4135 :     7. Rewrote Feedback module.
4136 :     8. Moved sortedlist.sml into viscomp-lib.cm. Eventually it
4137 :     should be thrown out and equivalent modules from smlnj-lib.cm
4138 :     should be used (IntRedBlackSet, IntListSet, ...).
4139 :    
4140 :     Confirmed that compiler compiles to fixpoint.
4141 :    
4142 :     ----------------------------------------------------------------------
4143 : leunga 731 Name: Allen Leung
4144 :     Date: 2000/11/10 18:00:00
4145 :     Tag: leunga-20001110-new-x86-fp
4146 :    
4147 :     A new x86 floating point code generator has been added.
4148 :     By default this is turned off. To turn this on, do:
4149 :    
4150 :     CM.autoload "$smlnj/compiler.cm";
4151 :     Compiler.Control.MLRISC.getFlag "x86-fast-fp" := true;
4152 :    
4153 :     Changes:
4154 :    
4155 :     1. Changed FTAN to FPTAN so that the assembly output is correct.
4156 :     2. Changed the extension callback for FTANGENT to generate:
4157 :    
4158 :     fptan
4159 :     fstp %st(0)
4160 :     instead of
4161 :     fptan
4162 :     fstpl ftempmem
4163 :    
4164 :     3. Numerous assembly fixes for x86.
4165 :    
4166 :     5. Cleaned up the machine code output module x86/x86MC.sml and added
4167 :     support for a whole bunch of instructions and addressing modes:
4168 :    
4169 :     fadd/fsub/fsubr/fmul/fdiv/fdivr %st, %st(n)
4170 :     faddp/fsubp/fsubrp/fmulp/fdivp/fdivrp %st, %st(n)
4171 :     fadd/fsub/fsubr/fmul/fdiv/fdivr %st(n), %st
4172 :     fiadd/fisub/fisubr/fimul/fidiv/fidivr mem
4173 :     fxch %st(n)
4174 :     fld %st(n)
4175 :     fst %st(n)
4176 :     fst mem
4177 :     fstp %st(n)
4178 :     fucom %st(n)
4179 :     fucomp %st(n)
4180 :    
4181 :     All these are now generated when the fast fp mode is turned on.
4182 :    
4183 :     6. Removed the dedicated registers %st(0), ..., %st(7) from X86CpsRegs
4184 :    
4185 :     ----------------------------------------------------------------------
4186 : blume 729 Name: Matthias Blume
4187 :     Date: 2000/11/09 11:20:00 JST
4188 :     Tag: blume-20001109-scc
4189 :     Description:
4190 :    
4191 :     Eliminated some code duplication:
4192 :    
4193 :     1. Added "where" clause to GraphSCCFn in SML/NJ Library.
4194 :     (Otherwise the functor is useless.)
4195 :     2. Used GraphSCCFn where SCCUtilFun was used previously.
4196 :     3. Got rid of SCCUtilFun (in comp-lib.cm).
4197 :    
4198 :     ----------------------------------------------------------------------
4199 : george 721 Name: Lal George
4200 :     Date: 2000/11/06 09:02:21 EST 2000
4201 :     Tag: Release_110_30
4202 :     Description:
4203 :    
4204 :     - Version 110.30
4205 :     ----------------------------------------------------------------------
4206 : blume 715 Name: Matthias Blume
4207 : blume 716 Date: 2000/11/04 14:45:00
4208 :     Tag: blume-20001104-mlbuild
4209 :     Description:
4210 :    
4211 :     - Made ml-build faster on startup.
4212 :     - Documentation fixes.
4213 :    
4214 :     ----------------------------------------------------------------------
4215 :     Name: Matthias Blume
4216 : blume 715 Date: 2000/11/02 17:00:00 JST
4217 :     Tag: blume-20001102-condcomp
4218 :     Description:
4219 :    
4220 :     - Small tweaks to pickler -- new BOOTFILES!
4221 :     - Version bumped to 110.29.2.
4222 :     - Added conditional compilation facility to init.cmi (see comment there).
4223 :     ----------------------------------------------------------------------
4224 : leunga 713 Name: Allen Leung
4225 :     Date: 2000/10/23 19:31:00
4226 :     Tag: leunga-20001023-demo-ra
4227 :    
4228 :     1. Minor RA changes that improves spilling on x86 (affects Moby and C-- only)
4229 :     2. Test programs for the graph library updated
4230 :     3. Some new MLRISC demo programs added
4231 :    
4232 :     ----------------------------------------------------------------------
4233 : blume 710 Name: Matthias Blume
4234 : blume 711 Date: 2000/08/31 22:15:00 JST
4235 :     Tag: blume-20001017-errmsg
4236 :     Description:
4237 :    
4238 :     More error message grief: Where there used to be no messages, there
4239 :     now were some that had bogus error regions. Fixed.
4240 :    
4241 :     ----------------------------------------------------------------------
4242 :     Name: Matthias Blume
4243 : blume 710 Date: 2000/08/31 17:30:00 JST
4244 :     Tag: blume-20001017-v110p29p1
4245 :     Description:
4246 :    
4247 :     I made a version 110.29.1 with new bootfiles.
4248 :    
4249 :     Changes: Modified pickler/unpickler for faster and leaner unpickling.
4250 :     CM documentation changes and a small bugfix in CM's error reporting.
4251 :    
4252 :     ----------------------------------------------------------------------
4253 : george 705 Name: Lal George
4254 :     Date: 2000/09/27 14:42:35 EDT
4255 :     Tag: george-20000927-nodestatus
4256 :     Description:
4257 :    
4258 :     Changed the type of the nodestatus, so that:
4259 :    
4260 :     SPILLED(~1) is now SPILLED
4261 :     SPILLED(m) where m>=0 is now MEMREG(m)
4262 :     SPILLED(s) where s<~1 is now SPILL_LOC(~s)
4263 :    
4264 :     ----------------------------------------------------------------------
4265 : blume 697 Name: Matthias Blume
4266 : blume 703 Date: 2000/09/07 14:45:00 JST
4267 :     Tag: blume-20000907-cmerrmsg
4268 :     Description:
4269 :    
4270 :     Small tweak to CM to avoid getting ML syntax error messages twice.
4271 :    
4272 :     ----------------------------------------------------------------------
4273 :     Name: Matthias Blume
4274 : blume 701 Date: 2000/08/31 18:00:00 JST
4275 :     Tag: blume-20000831-cvsbootfiles
4276 :     Description:
4277 :    
4278 :     New URL for boot files (because the 110.29 files on the BL server do
4279 :     now work correctly with my updated install scripts for yacc and lex).
4280 :    
4281 :     ----------------------------------------------------------------------
4282 :     Name: Matthias Blume
4283 : blume 697 Date: 2000/08/08 12:33:00 JST
4284 :     Tag: blume-20000808-manual
4285 :     Description:
4286 :    
4287 :     Tiny update to CM manual.
4288 :    
4289 :     ----------------------------------------------------------------------
4290 : leunga 695 Name: Allen Leung
4291 :     Date: 2000/08/7 19:31:00
4292 :     Tag: leunga-20000807-a-whole-bunch-of-stuff
4293 :    
4294 :     Moby, C--, SSA, x86, machine descriptions etc. Should only affect C--
4295 :     and Mobdy.
4296 :    
4297 :     1. x86
4298 :    
4299 :     a. Fixes to peephole module by John and Dan.
4300 :     b. Assembly fix to SETcc by Allen.
4301 :     c. Fix to c-call by John.
4302 :     d. Fix to spilling by John. (This one deals with the missing FSTPT case)
4303 :     e. Instruction selection optimization to SETcc as suggested by John.
4304 :    
4305 :     For example,
4306 :    
4307 :     MV(32, x, COND(32, CMP(32, LT, a, b), LI 1, LI 0))
4308 :    
4309 :     should generate:
4310 :    
4311 :     MOVL a, x
4312 :     SUBL b, x
4313 :     SHRL 31, x
4314 :    
4315 :     2. IR stuff
4316 :    
4317 :     A bunch of new DJ-graph related algorithms added. These
4318 :     speed up SSA construction.
4319 :    
4320 :     3. SSA + Scheduling
4321 :    
4322 :     Added code for SSA and scheduling to the repository
4323 :    
4324 :     ----------------------------------------------------------------------
4325 : george 694 Name: Lal George
4326 :     Date: 2000/07/27 11:53:14 EDT
4327 :    
4328 :     Tag: lal-20000727-linux-ppc
4329 :     Description:
4330 :    
4331 :     Made changes to support Linux PPC.
4332 :     p.s. I have confirmation that the 110.29 boot files work fine.
4333 :    
4334 :     ----------------------------------------------------------------------
4335 : blume 684 Name: Matthias Blume
4336 : blume 692 Date: 2000/07/27 17:40:00 JST
4337 :     Tag: blume-20000727-scripts
4338 :     Description:
4339 :    
4340 :     !!!! WARNING !!!!
4341 :     You must recompile the runtime system!
4342 :     !!!! WARNING !!!!
4343 :    
4344 :     This is basically another round of script-enhancements:
4345 :    
4346 :     1. sml, ml-build, and ml-makedepend accept options -D and -U to define
4347 :     and undefine CM preprocessor symbols.
4348 :    
4349 :     2. ml-build avoids generating a new heap image if it finds that the
4350 :     existing one is still ok. (The condition is that no ML file had to
4351 :     be recompiled and all ML files are found to be older that the heap
4352 :     file.)
4353 :    
4354 :     To make this work smoothly, I also hacked the runtime system as
4355 :     well as SMLofNJ.SysInfo to get access to the heap image suffix
4356 :     (.sparc-solaris, ...) that is currently being used.
4357 :    
4358 :     Moreover, the signature of CM.mk_standalone has changed. See the
4359 :     CM manual.
4360 :    
43