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

SCM Repository

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

Diff of /sml/trunk/HISTORY

Parent Directory Parent Directory | Revision Log Revision Log | View Patch Patch

revision 588, Fri Mar 31 09:00:02 2000 UTC revision 1029, Fri Jan 18 21:42:35 2002 UTC
# Line 8  Line 8 
8  The form of an entry should be:  The form of an entry should be:
9    
10  Name:  Name:
11  Date:  Date: yyyy/mm/dd
12  Tag: <post-commit CVS tag>  Tag: <post-commit CVS tag>
13  Description:  Description:
14    
15  ----------------------------------------------------------------------  ----------------------------------------------------------------------
16  Name: Matthias Blume  Name: Matthias Blume
17    Date: 2002/01/18
18    Tag: blume-20020118-more-ffifiddle
19    Description:
20    
21    Today's FFI fiddling:
22    
23      - Provided a structure CGetSet with "convenient" versions of C.Get.* and
24        C.Set.* that use concrete (MLRep.*) arguments and results instead
25        of abstract ones.
26    
27      - Provided word-style bit operations etc. for "int" representation
28        types in MLRep.S<Foo>Bitops where <Foo> ranges over Char, Int, Short,
29        and Long.
30    
31    ----------------------------------------------------------------------
32    Name: Matthias Blume
33    Date: 2002/01/18
34    Tag: blume-20020118-use-x86-fp
35    Description:
36    
37    Now that x86-fast-fp seems to be working, I turned it back on again
38    by default.  (Seems to work fine now, even with the FFI.)
39    
40    Other than that, I added some documentation about the FFI to
41    src/ml-nlffigen/README and updated the FFI test examples in
42    src/ml-nlffi-lib/Tests/*.
43    
44    ----------------------------------------------------------------------
45    Name: Allen Leung
46    Date: 2002/01/17
47    Tag: leunga-20020117-x86-fast-fp-call
48    Description:
49    
50       1. Fixed a problem with handling return fp values when x86's fast fp
51          mode is turned on.
52    
53       2. Minor pretty printing fix for cellset.  Print %st(0) as %st(0) instead
54          of %f32.
55    
56       3. Added a constructor INT32lit to the ast of MLRISC tools.
57    
58    ----------------------------------------------------------------------
59    Name: Matthias Blume
60    Date: 2002/01/16
61    Tag: blume-20020116-ffifiddle
62    Description:
63    
64    More fiddling with the FFI interface:
65    
66     - Make constness 'c instead of rw wherever possible.  This eliminates
67       the need for certain explicit coercions.  (However, due to ML's
68       value polymorphism, there will still be many cases where explicit
69       coercions are necessary.  Phantom types are not the whole answer
70       to modeling a subtyping relationship in ML.)
71    
72     - ro/rw coersions for pointers added.  (Avoids the detour through */&.)
73    
74     - "printf" test example added to src/ml-nlffi-lib/Tests.  (Demonstrates
75       clumsy workaround for varargs problem.)
76    
77    ----------------------------------------------------------------------
78    Name: Lal George
79    Date: 2002/01/15
80    Tag: <none>
81    Description:
82    
83    1. Since COPY instructions are no longer native to the architecture,
84       a generic functor can be used to implement the expandCopies function.
85    
86    2. Allowed EXPORT and IMPORT pseudo-op declarations to appear inside a
87       TEXT segment.
88    
89    ----------------------------------------------------------------------
90    Name: Matthias Blume
91    Date: 2002/01/15
92    Tag: blume-20020115-ffiupdates
93    Description:
94    
95    1. Fix for bug resulting in single-precision float values being returned
96       incorrectly from FFI calls.
97    
98    2. Small modifications to C FFI API:
99    
100        - memory-allocation routines return straight objects (no options)
101          and raise an exception in out-of-memory situations
102        - unsafe extensions to cast between function pointers and pointers
103          from/to ints
104        - added structure C_Debug as an alternative to structure C where
105          pointer-dereferencing (|*| and |*!) always check for null-pointers
106        - added open_lib' to DynLinkage;  open_lib' works like open_lib
107          but also takes a (possibly empty) list of existing library handles
108          that the current library depends on
109    
110    ----------------------------------------------------------------------
111    Name: Matthias Blume
112    Date: 2002/01/10
113    Tag: blume-20020110-newffigen
114    Description:
115    
116    1. Updates to portable graph code.
117    
118    2. Major update to ml-nlffigen and ml-nlffi-lib.  Things are much
119       more scalable now so that even huge interfaces such as the one
120       for GTK compile in finite time and space. :-)
121       See src/ml-nlffigen/README for details on what's new.
122    
123    ----------------------------------------------------------------------
124    Name: Lal George
125    Date: 2001/01/09 14:31:35 EST 2002
126    Tag: george-20011206-rm-native-copy
127    Description:
128    
129            Removed the native COPY and FCOPY instructions
130            from all the architectures and replaced it with the
131            explicit COPY instruction from the previous commit.
132    
133            It is now possible to simplify many of the optimizations
134            modules that manipulate copies. This has not been
135            done in this change.
136    
137    ----------------------------------------------------------------------
138    Name: Lal George
139    Date: 2001/12/06 16:50:13 EST 2001
140    Tag: george-20011206-mlrisc-instruction
141    Description:
142    
143    Changed the representation of instructions from being fully abstract
144    to being partially concrete. That is to say:
145    
146      from
147            type instruction
148    
149      to
150            type instr                              (* machine instruction *)
151    
152            datatype instruction =
153                LIVE of {regs: C.cellset, spilled: C.cellset}
154              | KILL of {regs: C.cellset, spilled: C.cellset}
155              | COPYXXX of {k: CB.cellkind, dst: CB.cell list, src: CB.cell list}
156              | ANNOTATION of {i: instruction, a: Annotations.annotation}
157              | INSTR of instr
158    
159    This makes the handling of certain special instructions that appear on
160    all architectures easier and uniform.
161    
162    LIVE and KILL say that a list of registers are live or killed at the
163    program point where they appear. No spill code is generated when an
164    element of the 'regs' field is spilled, but the register is moved to
165    the 'spilled' (which is present, more for debugging than anything else).
166    
167    LIVE replaces the (now deprecated) DEFFREG instruction on the alpha.
168    We used to generate:
169    
170            DEFFREG f1
171            f1 := f2 + f3
172            trapb
173    
174    but now generate:
175    
176            f1 := f2 + f3
177            trapb
178            LIVE {regs=[f1,f2,f3], spilled=[]}
179    
180    Furthermore, the DEFFREG (hack) required that all floating point instruction
181    use all registers mentioned in the instruction. Therefore f1 := f2 + f3,
182    defines f1 and uses [f1,f2,f3]! This hack is no longer required resulting
183    in a cleaner alpha implementation. (Hopefully, intel will not get rid of
184    this architecture).
185    
186    COPYXXX is intended to replace the parallel COPY and FCOPY  available on
187    all the architectures. This will result in further simplification of the
188    register allocator that must be aware of them for coalescing purposes, and
189    will also simplify certain aspects of the machine description that provides
190    callbacks related to parallel copies.
191    
192    ANNOTATION should be obvious, and now INSTR represents the honest to God
193    machine instruction set!
194    
195    The <arch>/instructions/<arch>Instr.sml files define certain utility
196    functions for making porting easier -- essentially converting upper case
197    to lower case. All machine instructions (of type instr) are in upper case,
198    and the lower case form generates an MLRISC instruction. For example on
199    the alpha we have:
200    
201      datatype instr =
202         LDA of {r:cell, b:cell, d:operand}
203       | ...
204    
205      val lda : {r:cell, b:cell, d:operand} -> instruction
206        ...
207    
208    where lda is just (INSTR o LDA), etc.
209    
210    ----------------------------------------------------------------------
211    Name: Matthias Blume
212    Date: 2001/11/22 21:40:00 EST
213    Tag: Release_110_37
214    Description:
215    
216    Release 110.37.  This time for real.
217    
218    ----------------------------------------------------------------------
219    Name: Matthias Blume
220    Date: 2001/11/21 16:35:00 EST
221    Tag: blume-20011121-foot-in-mouth
222    Description:
223    
224    Removed the "Release_110_37" tag because of a serious bug.
225    This will be re-tagged once the bug is fixed.
226    
227    ----------------------------------------------------------------------
228    Name: Matthias Blume
229    Date: 2001/11/21 16:14:00 EST
230    Tag: blume-20011121-forgottenfile
231    Description:
232    
233    Forgot to add a file.  (Just a .tex-file -- part of
234    the CM manual source.)
235    
236    ----------------------------------------------------------------------
237    Name: Matthias Blume
238    Date: 2001/11/21 16:10:00 EST
239    Tag: blume-20011121-invalid_110_37
240    Description:
241    
242    Note: I removed the original tag "Release_110_37" from this commit
243          because we found a serious bug in all non-x86 backends.
244          - Matthias
245    
246    1. Modifications to the SML/NJ code generator and to the runtime system
247       so that code object name strings are directly inserted into code
248       objects at code generation time.  The only business the runtime system
249       has with this is now to read the name strings on occasions.
250       (The encoding of the name string has also changed somewhat.)
251    
252    2. CM now implements a simple "set calculus" for specifying export lists.
253       In particular, it is now possible to refer to the export lists of
254       other libraries/groups/sources and form unions as well as differences.
255       See the latest CM manual for details.
256    
257    3. An separate notion of "proxy" libraries has again be eliminated from
258       CM's model.  (Proxy libraries are now simply a special case of using
259       the export list calculus.)
260    
261    4. Some of the existing libraries now take advantage of the new set
262       calculus.
263       (Notice that not all libraries have been converted because some
264       of the existing .cm-files are supposed to be backward compatible
265       with 110.0.x.)
266    
267    5. Some cleanup in stand-alone programs.  (Don't use "exnMessage" -- use
268       "General.exnMessage"!  The former relies on a certain hook to be
269       initialized, and that often does not happen in the stand-alone case.)
270    
271    ----------------------------------------------------------------------
272    Name: Lal George
273    Date: 2001/11/21  13:56:18 EST
274    Tag: george-2001121-pseudo-ops
275    Description:
276    
277      Implemented a complete redesign of MLRISC pseudo-ops. Now there
278      ought to never be any question of incompatabilities with
279      pseudo-op syntax expected by host assemblers.
280    
281      For now, only modules supporting GAS syntax are implemented
282      but more should follow, such as MASM, and vendor assembler
283      syntax, e.g. IBM as, Sun as, etc.
284    
285    ----------------------------------------------------------------------
286    Name: Matthias Blume
287    Date: 2001/11/14 11:52:00 EST
288    Tag: blume-20011114-srcname
289    Description:
290    
291    1. Routed the name of the current source file to mlriscgen where it
292       should be directly emitted into the code object.  (This last part
293       is yet to be done.)
294    
295    2. Some cleanup of the pgraph code to make it match the proposal that
296       I put out the other day.  (The proposal notwithstanding, things are
297       still in flux here.)
298    
299    ----------------------------------------------------------------------
300    Name: Lal George
301    Date: 2001/11/14 09:44:04 EST
302    Tag:
303    Description:
304    
305      Fix for a backpatching bug reported by Allen.
306    
307      Because the boundary between short and long span-dependent
308      instructions is +/- 128, there are an astounding number of
309      span-dependent instructions whose size is over estimated.
310    
311      Allen came up with the idea of letting the size of span
312      dependent instructions be non-monotonic, for a maxIter
313      number of times, after which the size must be monotonically
314      increasing.
315    
316      This table shows the number of span-dependent instructions
317      whose size was over-estimated as a function of maxIter, for the
318      file Parse/parse/ml.grm.sml:
319    
320         maxIter            # of instructions:
321            10                      687
322            20                      438
323            30                      198
324            40                        0
325    
326      In compiling the compiler, there is no significant difference in
327      compilation speed between maxIter=10 and maxIter=40. Actually,
328      my measurements showed that maxIter=40 was a tad faster than
329      maxIter=10! Also 96% of the  files in the compiler reach a fix
330      point within 13 iterations, so fixing maxIter at 40, while high,
331      is okay.
332    
333    ----------------------------------------------------------------------
334    Name: Matthias Blume
335    Date: 2001/10/31 15:25:00 EST
336    Tag: blume-20011031-pgraph
337    Description:
338    
339    CKIT:
340    * Changed the "Function" constructor of type Ast.ctype to carry optional
341      argument identifiers.
342    * Changed the return type of TypeUtil.getFunction accordingly.
343    * Type equality ignores the argument names.
344    * TypeUtil.composite tries to preserve argument names but gives up quickly
345      if there is a mismatch.
346    
347    installation script:
348    * attempts to use "curl" if available (unless "wget" is available as well)
349    
350    CM:
351    * has an experimental implementation of "portable graphs" which I will
352      soon propose as an implementation-independent library format
353    * there are also new libraries $/pgraph.cm and $/pgraph-util.cm
354    
355    NLFFI-LIB:
356    * some cleanup (all cosmetic)
357    
358    NLFFIGEN:
359    * temporarily disabled the mechanism that suppresses ML output for
360      C definitions whose identifiers start with an underscore character
361    * generate val bindings for enum constants
362    * user can request that only one style (light or heavy) is being used;
363      default is to use both (command-line arguments: -heavy and -light)
364    * fixed bug in handling of function types involving incomplete pointers
365    * generate ML entry points that take record arguments (i.e., using
366      named arguments) for C functions that have a prototype with named
367      arguments
368      (see changes to CKIT)
369    
370    ----------------------------------------------------------------------
371    Name: Allen Leung
372    Date: 2001/10/27 20:34:00 EDT
373    Tag: leunga-20011027-x86-fast-fp-call
374    Description:
375    
376       Fixed the bug described in blume-20010920-slowfp.
377    
378       The fix involves
379          1. generating FCOPYs in FSTP in ia32-svid
380          2. marking a CALL with the appropriate annotation
381    
382    ----------------------------------------------------------------------
383    Name: Matthias Blume
384    Date: 2001/10/16 11:32:00 EDT
385    Tag: blume-20011016-netbsd
386    Description:
387    
388    Underscore patch from Chris Richards (fixing problem with compiling
389    runtime system under recent NetBSD).
390    
391    ----------------------------------------------------------------------
392    Name: Allen Leung
393    Date: 2001/10/12 17:18:32 EDT 2001
394    Tag: leung-20011012-x86-printflowgraph
395    Description:
396    
397    X86RA now uses a valid (instead of dummy) PrintFlowgraph module.
398    
399    ----------------------------------------------------------------------
400    Name: Lal George
401    Date: 2001/10/11 23:51:34 EDT
402    Tag: george-20011011-too-many-instrs
403    Description:
404    
405    The representation of a program point never expected to see more
406    than 65536 instructions in a basic block!
407    
408    ----------------------------------------------------------------------
409    Name: Lal George
410    Date: 2001/10/09 09:41:37 EDT
411    Tag: george-20011008-mlrisc-labels
412    Description:
413    
414    Changed the machine description files to support printing of
415    local and global labels in assembly code, based on host assembler
416    conventions.
417    
418    ----------------------------------------------------------------------
419    Name: Matthias Blume
420    Date: 2001/09/25 15:25:00 EDT
421    Tag: blume-20010925-exninfo
422    Description:
423    
424    I provided a non-hook implementation of exnName (at the toplevel) and
425    made the "dummy" implementation of exnMessage (at the toplevel) more
426    useful: if nothing gets "hooked in", then at least you are going to
427    see the exception name and a message indicating why you don't see more.
428    
429    [For the time being, programs that need exnMessage and want to use
430    ml-build should either use General.exnMessage (strongly recommended) or
431    refer to structure General at some other point so that CM sees a
432    static dependency.]
433    
434    [Similar remarks go for "print" and "use":  If you want to use their
435    functionality in stand-alone programs generated by ml-build, then use
436    TextIO.output and Backend.Interact.useFile (from $smlnj/compiler.cm).]
437    
438    ----------------------------------------------------------------------
439    Name: Matthias Blume
440    Date: 2001/09/20 17:28:00 EDT
441    Tag: blume-20010920-slowfp
442    Description:
443    
444    Allen says that x86-fast-fp is not safe yet, so I turned it off again...
445    
446    ----------------------------------------------------------------------
447    Name: Matthias Blume
448    Date: 2001/09/20 17:20:00 EDT
449    Tag: blume-20010920-canonicalpaths
450    Description:
451    
452    0. Updated the BOOT file (something that I forgot to do earlier).
453    
454    1. Small internal change to CM so that it avoids "/../" in filenames
455       as much as possible (but only where it is safe).
456    
457    2. Changed config/_run-sml (resulting in a changed bin/.run-sml) so
458       that arguments that contain delimiters are passed through correctly.
459       This change also means that all "special" arguments of the form
460       @SMLxxx... must come first.
461    
462    3. Changed install script to put relative anchor names for tool commands
463       into pathconfig.
464    
465    ----------------------------------------------------------------------
466    Name: Matthias Blume
467    >>>>>>> 1.169
468    Date: 2001/09/18 15:35:00 EDT
469    Tag: blume-20010918-readme11036
470    Description:
471    
472    Added README files.
473    
474    ----------------------------------------------------------------------
475    Name: Matthias Blume
476    Date: 2001/09/18 11:45:00 EDT
477    Tag: Release_110_36 (retag)
478    Description:
479    
480    Fixed mistake in config/preloads. Retagged as 110.36.
481    
482    ----------------------------------------------------------------------
483    Name: Matthias Blume
484    Date: 2001/09/18 09:40:00 EDT
485    Tag: Release_110_36_orig (tag changed)
486    Description:
487    
488    New version (110.36).  New bootfiles.
489    
490    ----------------------------------------------------------------------
491    Name: Matthias Blume
492    Date: 2001/09/14 16:15:00 EDT
493    Tag: blume-20010914-x86fastfp
494    Description:
495    
496    John committed some changes that Allen made, in particular a (hopefully)
497    correctly working version of the x86-fp module.
498    
499    I changed the default setting of the Control.MLRISC.getFlag "x86-fast-fp"
500    flag to "true".  Everything seems to compile to a fixpoint ok, and
501    "mandelbrot" speeds up by about 15%.
502    
503    ----------------------------------------------------------------------
504    Name: Matthias Blume
505    Date: 2001/09/13 11:20:00 EDT
506    Tag: blume-20010913-minimal
507    Description:
508    
509    1. Stefan Monnier's patch to fix a miscompilation problem that
510       was brought to light by John Reppy's work on Moby.
511    
512    2. Implemented a minimal "structure Compiler" that contains just
513       "version" and "architecture".  The minimal version will be
514       available when the full version is not.  This is for backward-
515       compatibility with code that wants to test Compiler.version.
516    
517    ----------------------------------------------------------------------
518    Name: Matthias Blume
519    Date: 2001/08/28 14:03:00 EDT
520    Tag: blume-20010828-ml-lex
521    Description:
522    
523    Fix for bug 1581, received from Neophytos Michael.
524    
525    ----------------------------------------------------------------------
526    Name: Matthias Blume
527    Date: 2001/08/27 11:20:00 EDT
528    Tag: blume-20010827-readme11035
529    Description:
530    
531    Fleshed out the README file for 110.35.
532    
533    ----------------------------------------------------------------------
534    Name: Matthias Blume
535    Date: 2001/08/24 17:10:00 EDT
536    Tag: Release_110_35
537    Description:
538    
539    New version number (110.35).  New bootfiles.
540    
541    ----------------------------------------------------------------------
542    Name: Lal George
543    Date: 2001/08/24 13:47:18 EDT 2001
544    Tag: george-20010824-MLRISC-graphs
545    Description:
546    
547     removed clusters from MLRISC completely and replaced with graphs.
548    
549    ----------------------------------------------------------------------
550    Name: Matthias Blume
551    Date: 2001/08/23 17:50:00 EDT
552    Tag: blume-20010823-toplevel
553    Description:
554    
555    - some reorganization of the code that implements various kinds of
556      environments in the compiler (static, dynamic, symbolic, combined)
557    - re-implemented the EnvRef module so that evalStream works properly
558      (if the stream contains references to "use", "CM.make", etc.)
559    - cleaned up evalloop.sml and interact.sml (but they need more cleaning)
560    
561    ----------------------------------------------------------------------
562    Name: Matthias Blume
563    Date: 2001/08/20 15:50 EDT
564    Tag: blume20010820-slipup
565    Description:
566    
567    I forgot to commit a few files.  Here they are...
568    
569    ----------------------------------------------------------------------
570    Name: Matthias Blume
571    Date: 2001/08/20 15:35:00 EDT
572    Tag: blume-20010820-debugprof
573    Description:
574    
575    !!!! NEW BOOTFILES !!!!
576    
577    This is another round of reorganizing the compiler sources.  This
578    time the main goal was to factor out all the "instrumentation"
579    passes (for profiling and backtracing) into their own library.
580    The difficulty was to do it in such a way that it does not depend
581    on elaborate.cm but only on elabdata.cm.
582    
583    Therefore there have been further changes to both elaborate.cm and
584    elabdata.cm -- more "generic" things have been moved from the former
585    to the latter.  As a result, I was forced to split the assignment
586    of numbers indicating "primtyc"s into two portions: SML-generic and
587    SML/NJ-specific.  Since it would have been awkward to maintain,
588    I bit the bullet and actually _changed_ the mapping between these
589    numbers and primtycs.  The bottom line of this is that you need
590    a new set of bin- and bootfiles.
591    
592    I have built new bootfiles for all architectures, so doing a fresh
593    checkout and config/install.sh should be all you need.
594    
595    The newly created library's name is
596    
597        $smlnj/viscomp/debugprof.cm
598    
599    and its sources live under
600    
601        src/compiler/DebugProf
602    
603    ----------------------------------------------------------------------
604    Name: Matthias Blume
605    Date: 2001/08/15 17:15:00 EDT
606    Tag: blume-20010815-compreorg
607    Description:
608    
609    This is a first cut at reorganizing the CM libraries that make up the
610    core of the compiler.  The idea is to separate out pieces that could
611    be used independently by tools, e.g., the parser, the typechecker, etc.
612    
613    The current status is a step in this direction, but it is not quite
614    satisfactory yet.  Expect more changes in the future.
615    
616    Here is the current (new) organization...
617    
618        What used to be $smlnj/viscomp/core.cm is now divided into
619        six CM libraries:
620    
621             $smlnj/viscomp/basics.cm
622                           /parser.cm
623                           /elabdata.cm
624                           /elaborate.cm
625                           /execute.cm
626                           /core.cm
627    
628        The CM files for these libraries live under src/system/smlnj/viscomp.
629        All these libraries are proxy libraries that contain precisely
630        one CM library component.  Here are the locations of the components
631        (all within the src/compiler tree):
632    
633             Basics/basics.cm
634             Parse/parser.cm
635             ElabData/elabdata.cm
636             Elaborator/elaborate.cm
637             Execution/execute.cm
638             core.cm
639    
640         [This organization is the same that has been used already
641         for a while for the architecture-specific parts of the visible
642         compiler and for the old version of core.cm.]
643    
644         As you will notice, many source files have been moved from their
645         respective original locations to a new home in one of the above
646         subtrees.
647    
648         The division of labor between the new libraries is the following:
649    
650             basics.cm:
651                - Simple, basic definitions that pertain to many (or all) of
652                  the other libraries.
653             parser.cm:
654                - The SML parser, producing output of type Ast.dec.
655                - The type family for Ast is also defined and exported here.
656             elabdata.cm:
657                - The datatypes that describe input and output of the elaborator.
658                  This includes types, absyn, and static environments.
659             elaborator.cm:
660                - The SML/NJ type checker and elaborator.
661                  This maps an Ast.dec (with a given static environment) to
662                  an Absyn.dec (with a new static environment).
663                - This libraries implements certain modules that used to be
664                  structures as functors (to remove dependencies on FLINT).
665             execute.cm:
666                - Everything having to do with executing binary code objects.
667                - Dynamic environments.
668             core.cm:
669                - SML/NJ-specific instantiations of the elaborator and MLRISC.
670                - Top-level modules.
671                - FLINT (this should eventually become its own library)
672    
673    Notes:
674    
675    I am not 100% happy with the way I separated the elaborator (and its
676    data structures) from FLINT.  Two instances of the same problem:
677    
678        1. Data structures contain certain fields that carry FLINT-specific
679           information.  I hacked around this using exn and the property list
680           module from smlnj-lib.  But the fact that there are middle-end
681           specific fields around at all is a bit annoying.
682    
683        2. The elaborator calculates certain FLINT-related information.  I tried
684           to make this as abstract as I could using functorization, but, again,
685           the fact that the elaborator has to perform calculations on behalf
686           of the middle-end at all is not nice.
687    
688        3. Having to used exn and property lists is unfortunate because it
689           weakens type checking.  The other alternative (parameterizing
690           nearly *everything*) is not appealing, though.
691    
692    I removed the "rebinding =" warning hack because due to the new organization
693    it was awkward to maintain it.  As a result, the compiler now issues some of
694    these warnings when compiling init.cmi during bootstrap compilation. On
695    the plus side, you also get a warning when you do, for example:
696       val op = = Int32.+
697    which was not the case up to now.
698    
699    I placed "assign" and "deref" into the _Core structure so that the
700    code that deals with the "lazy" keyword can find them there.  This
701    removes the need for having access to the primitive environment
702    during elaboration.
703    
704    ----------------------------------------------------------------------
705    Name: Matthias Blume
706    Date: 2001/08/13
707    Tag: blume-20010813-closures
708    Description:
709    
710    This fix was sent to us by Zhong Shao.  It is supposed to improve the
711    performance of certain loops by avoiding needless closure allocation.
712    
713    ----------------------------------------------------------------------
714    Name: Lal George
715    Date: 2001/07/31 10:03:23 EDT 2001
716    Tag: george-20010731-x86-fmalloc
717    Description: Fixed bug in x86 calls
718    
719        There was a bug where call instructions would mysteriously
720        vanish. The call instruction had to be one that returned
721        a floating point value.
722    
723    ----------------------------------------------------------------------
724    Name: Lal George
725    Date: 2001/07/19 16:36:29 EDT 2001
726    Tag: george-20010719-simple-cells
727    Description:
728    
729    I have dramatically simplified the interface for CELLS in MLRISC.
730    
731    In summary, the cells interface is broken up into three parts:
732    
733      1. CellsBasis : CELLS_BASIS
734    
735            CellsBasis is a top level structure and common for all
736            architectures.  it contains the definitions of basic datatypes
737            and utility  functions over these types.
738    
739      2. functor Cells() : CELLS
740    
741            Cells generates an interface for CELLS that incorporates the
742            specific resources on the target architecture, such as the
743            presence of special register classes, their number and size,
744            and various useful substructures.
745    
746      3. <ARCH>CELLS
747    
748            e.g. SparcCells: SPARCCELLS
749    
750            <ARCH>CELLS usually contains additional bindings for special
751            registers  on the architecture, such as:
752    
753                    val r0 : cell           (* register zero *)
754                    val y : cell            (* Y register *)
755                    val psr : cell          (* processor status register *)
756                    ...
757    
758            The structure returned by applying the Cells functor is opened
759            in this interface.
760    
761    The main implication of all this is that the datatypes for cells is
762    split between CellsBasis and CELLS -- a fairly simple change for user
763    code.
764    
765    In the old scheme the CELLS interface had a definitional binding of
766    the form:
767    
768            signature CELLS = sig
769    
770               structure CellsBasis = CellsBasis
771    
772               ...
773    
774            end
775    
776    With all the sharing constraints that goes on in MLRISC, this old
777    design  quickly leads to errors such as:
778    
779            "structure definition spec inside of sharing ... "
780    
781    
782    and appears to require an unacceptable amount of sharing and where
783    constraint hackery.
784    
785    I think this error message (the interaction of definitional specs and
786    sharing) requires more explanation on our web page.
787    
788    ----------------------------------------------------------------------
789    Name: Matthias Blume
790    Date: 2001/07/19 15:00:00 EDT
791    Tag: blume-20010719-libreorg
792    Description:
793    
794    This update puts together a fairly extensive but straightforward change
795    to the way the libraries that implement the interactive system are
796    organized:
797    
798       The biggest change is the elimination of structure Compiler.  As a
799       replacement for this structure, there is now a CM library
800       (known as $smlnj/compiler.cm or $smlnj/compiler/current.cm)
801       that exports all the substructures of the original structure Compiler
802       directly.  So instead of saying Compiler.Foo.bar one now simply
803       says Foo.bar.  (The CM libraries actually export a collection of
804       structures that is richer than the collection of substructures of
805       structure Compiler.)
806    
807       To make the transition smooth, there is a separate library called
808       $smlnj/compiler/compiler.cm which puts together and exports the
809       original structure Compiler (or at least something very close to it).
810    
811       There are five members of the original structure Compiler
812       that are not exported directly but which instead became members
813       of a new structure Backend (described by signature BACKEND).  These are:
814       structure Profile (: PROFILE), structure Compile (: COMPILE), structure
815       Interact (: INTERACT), structure Machine (: MACHINE), and val
816       architecture (: string).
817    
818       Structure Compiler.Version has become structure CompilerVersion.
819    
820       Cross-compilers for alpha32, hppa, ppc, sparc, and x86 are provided
821       by $smlnj/compiler/<arch>.cm where <arch> is alpha32, hppa, ppc, sparc,
822       or x86, respectively.
823       Each of these exports the same frontend structures that
824       $smlnj/compiler.cm exports.  But they do not have a structure Backend
825       and instead export some structure <Arch>Backend where <Arch> is Alpha32,
826       Hppa, PPC, Sparc, or X86, respectively.
827    
828       Library $smlnj/compiler/all.cm exports the union of the exports of
829       $smlnj/compiler/<arch>.cm
830    
831       There are no structures <Arch>Compiler anymore, use
832       $smlnj/compiler/<arch>.cm instead.
833    
834       Library host-compiler-0.cm is gone.  Instead, the internal library
835       that instantiates CM is now called cm0.cm.  Selection of the host
836       compiler (backend) is no longer done here but. (Responsibility for it
837       now lies with $smlnj/compiler/current.cm.  This seems to be more
838       logical.)
839    
840       Many individual files have been moved or renamed.  Some files have
841       been split into multiple files, and some "dead" files have been deleted.
842    
843    Aside from these changes to library organization, there are also changes
844    to the way the code itself is organized:
845    
846       Structure Binfile has been re-implemented in such a way that it no
847       longer needs any knowledge of the compiler.  It exclusively deals
848       with the details of binfile layout.  It no longer invokes the
849       compiler (for the purpose of creating new prospective binfile
850       content), and it no longer has any knowledge of how to interpret
851       pickles.
852    
853       Structure Compile (: COMPILE) has been stripped down to the bare
854       essentials of compilation.  It no longer deals with linking/execution.
855       The interface has been cleaned up considerably.
856    
857       Utility routines for dealing with linking and execution have been
858       moved into their own substructures.
859    
860       (The ultimate goal of these changes is to provide a light-weight
861       binfile loader/linker (at least for, e.g., stable libraries) that
862       does not require CM or the compiler to be present.)
863    
864    CM documentation has been updated to reflect the changes to library
865    organization.
866    
867    ----------------------------------------------------------------------
868    Name: Matthias Blume
869    Date: 2001/07/10 17:30:00 EDT
870    Tag: Release_110_34
871    Description:
872    
873    Minor tweak to 110.34 (re-tagged):
874    
875      - README.html file added to CVS repository
876      - runtime compiles properly under FreeBSD 3.X and 4.X
877    
878    ----------------------------------------------------------------------
879    Name: Matthias Blume
880    Date: 2001/07/10 17:30:00 EDT
881    Tag: Release_110_34
882    Description:
883    
884    New version number (110.34). New bootfiles.
885    
886    ----------------------------------------------------------------------
887    Name: Matthias Blume
888    Date: 2001/07/09 16:00:00 EDT
889    Tag: blume-20010709-more-varargs
890    Description:
891    
892    I changed the handling of varargs in ml-nlffigen again:
893    The ellipsis ... will now simply be ignored (with an accompanying warning).
894    
895    The immediate effect is that you can actually call a varargs function
896    from ML -- but you can't actually supply any arguments beyond the ones
897    specified explicitly.  (For example, you can call printf with its format
898    string, but you cannot pass additional arguments.)
899    
900    This behavior is only marginally more useful than the one before, but
901    it has the advantage that a function or, more importantly, a function
902    type never gets dropped on the floor, thus avoiding follow-up problems with
903    other types that refer to the offending one.
904    
905    ----------------------------------------------------------------------
906    Name: Matthias Blume
907    Date: 2001/07/09 11:25:00 EDT
908    Tag: blume-20010709-varargs
909    Description:
910    
911    1. ckit-lib.cm now exports structure Error
912    2. ml-nlffigen reports occurences of "..." (i.e., varargs function types)
913       with a warning accompanied by a source location.  Moreover, it
914       merely skips the offending function or type and proceeds with the
915       rest of its work.u  As a result, one can safely feed C code containing
916       "..." to ml-nlffigen.
917    3. There are some internal improvements to CM, providing slightly
918       more general string substitutions in the tools subsystem.
919    
920    ----------------------------------------------------------------------
921    Name: Matthias Blume
922    Date: 2001/06/27 15:10:00 EDT
923    Tag: blume-20010627-concur
924    Description:
925    
926    Fixed a small bug in CM's handling of parallel compilation.
927    (You could observe the bug by Control-C-interrupting an ordinary
928    CMB.make or CM.stabilize and then attaching some compile servers.
929    The result was that all of a sudden the previously interrupted
930    compilation would continue on its own.  This was because of
931    an over-optimization: CM did not bother to clean out certain queues
932    when no servers were attached "anyway", resulting in the contents
933    of these queues to grab control when new servers did get attached.)
934    
935    There is also another minor update to the CM manual.
936    
937    ----------------------------------------------------------------------
938    Name: Matthias Blume
939    Date: 2001/06/26 16:15:00 EDT
940    Tag: blume-20010626-cmdoc
941    Description:
942    
943    Minor typo fixed in CM manual (syntax diagram for libraries).
944    
945    ----------------------------------------------------------------------
946    Name: Matthias Blume
947    Date: 2001/06/25 22:55:00 EDT
948    Tag: blume-20010625-x86pc
949    Description:
950    
951    Fixed a nasty bug in the X86 assembly code that caused signal
952    handlers to fail (crash) randomly.
953    
954    ----------------------------------------------------------------------
955    Name: Matthias Blume
956    Date: 2001/06/25 12:05:00 EDT
957    Tag: blume-20010625-nlffigen
958    Description:
959    
960    This update fixes a number of minor bugs in ml-nlffigen as reported by
961    Nick Carter <nbc@andrew.cmu.edu>.
962    
963      1. Silly but ok typedefs of the form "typedef void myvoid;" are now accepted.
964      2. Default names for generated files are now derived from the name of
965         the C file *without its directory*.  In particular, this causes generated
966         files to be placed locally even if the C file is in some system directory.
967      3. Default names for generated signatures and structures are also derived
968         from the C file name without its directory.  This avoids silly things
969         like "structure GL/GL".
970         (Other silly names are still possible because ml-nlffigen does not do
971          a thorough check of whether generated names are legal ML identifiers.
972          When in doubt, use command line arguments to force particular names.)
973    
974    ----------------------------------------------------------------------
975    Name: Matthias Blume
976    Date: 2001/06/21 12:25:00 EDT
977    Tag: blume-20010621-eXene
978    Description:
979    
980    eXene now compiles and (sort of) works again.
981    
982    The library name (for version > 110.33) is $/eXene.cm.
983    
984    I also added an new example in src/eXene/examples/nbody.  See the
985    README file there for details.
986    
987    ----------------------------------------------------------------------
988    Name: Matthias Blume
989    Date: 2001/06/20 16:40:00 EDT
990    Tag: blume-20010620-cml
991    Description:
992    
993    CML now compiles and works again.
994    
995    Libraries (for version > 110.33):
996    
997      $cml/cml.cm            Main CML library.
998      $cml/basis.cm          CML's version of $/basis.cm.
999      $cml/cml-internal.cm   Internal helper library.
1000      $cml/core-cml.cm       Internal helper library.
1001      $cml-lib/trace-cml.cm  Tracing facility.
1002      $cml-lib/smlnj-lib.cm  CML's version of $/smlnj-lib.cm
1003    
1004    The installer (config/install.sh) has been taught how to properly
1005    install this stuff.
1006    
1007    ----------------------------------------------------------------------
1008    Name: Matthias Blume
1009    Date: 2001/06/19 17:55:00 EDT
1010    Tag: blume-20010619-instantiate
1011    Description:
1012    
1013    This un-breaks the fix for bug 1432.
1014    (The bug was originally fixed in 110.9 but I broke it again some
1015    time after that.)
1016    
1017    ----------------------------------------------------------------------
1018    Name: Matthias Blume
1019    Date: 2001/06/19 17:25:00 EDT
1020    Tag: blume-20010619-signals
1021    Description:
1022    
1023    This should (hopefully) fix the long-standing signal handling bug.
1024    (The runtime system was constructing a continuation record with an
1025    incorrect descriptor which would cause the GC to drop data on the floor...)
1026    
1027    ----------------------------------------------------------------------
1028    Name: Matthias Blume
1029    Date: 2001/06/15 15:05:00 EDT
1030    Tag: blume-20010615-moresparc
1031    Description:
1032    
1033    Here is a short late-hour update related to Sparc c-calls:
1034    
1035     -- made handling of double-word arguments a bit smarter
1036    
1037     -- instruction selection phase tries to collapse certain clumsily
1038        constructed ML-Trees; typical example:
1039    
1040            ADD(ty,ADD(_,e,LI d1),LI d2)  ->  ADD(ty,e,LI(d1+d2))
1041    
1042        This currently has no further impact on SML/NJ since mlriscGen does
1043        not seem to generate such patterns in the first place, and c-calls
1044        (which did generate them in the beginning) has meanwhile been fixed
1045        so as to avoid them as well.
1046    
1047    ----------------------------------------------------------------------
1048    Name: Matthias Blume
1049    Date: 2001/06/15 15:05:00 EDT
1050    Tag: blume-20010615-sparc
1051    Description:
1052    
1053    The purpose of this update is to provide an implementation of NLFFI
1054    on Sparc machines.
1055    
1056    Here are the changes in detail:
1057    
1058       * src/MLRISC/sparc/c-calls/sparc-c-calls.sml is a new file containing
1059       the Sparc implementation of the c-calls API.
1060       * The Sparc backend of SML/NJ has been modified to uniformely use %fp
1061       for accessing the ML frame.  Thus, we have a real frame pointer and
1062       can freely modify %sp without need for an omit-frame-ptr phase.
1063       The vfp logic in src/compiler/CodeGen/* has been changed to accomodate
1064       this case.
1065       * ml-nlffigen has been taught to produce code for different architectures
1066       and calling conventions.
1067       * In a way similar to what was done in the x86 case, the Sparc
1068       backend uses its own specific extension to mltree.  (For example,
1069       it needs to be able to generate UNIMP instructions which are part
1070       of the calling convention.)
1071       * ml-nlffi-lib was reorganized to make it more modular (in particular,
1072       to make it easier to plug in new machine- and os-dependent parts).
1073    
1074    There are some other fairly unrelated bug fixes and cleanups as well:
1075    
1076       * I further hacked the .cm files for MLRISC tools (like MDLGen) so
1077       that they properly share their libraries with existing SML/NJ libraries.
1078       * I fixed a minor cosmetic bug in CM, supressing certain spurious
1079       follow-up error messages.
1080       * Updates to CM/CMB documentation.
1081    
1082    TODO items:
1083    
1084       * MLRISC should use a different register as its asmTemp on the Sparc.
1085         (The current %o2 is a really bad choice because it is part of the
1086         calling conventions, so things might interfere in unexpected ways.)
1087    
1088    ----------------------------------------------------------------------
1089    Name: Matthias Blume
1090    Date: 2001/06/07
1091    Tag: blume-20010607-calls
1092    Description:
1093    
1094    A number of internal changes related to C calls and calling conventions:
1095    
1096    1. ML-Tree CALL statements now carry a "pops" field.  It indicates the
1097       number of bytes popped implicitly (by the callee).  In most cases
1098       this field is 0 but on x86/win32 it is some non-zero value.  This
1099       is information provided for the benefit of the "omit-frameptr" pass.
1100    2. The CALL instruction on the x86 carries a similar "pops" field.
1101       The instruction selection phase copies its value from the ML-Tree
1102       CALL statement.
1103    3. On all other architectures, the instruction selection phase checks
1104       whether "pops=0" and complains if not.
1105    4. The c-calls implementation for x86 now accepts two calling conventions:
1106       "ccall" and "stdcall".  When "ccall" is selected, the caller cleans
1107       up after the call and pops is set to 0.  For "stdcall", the caller
1108       does nothing, leaving the cleanup to the callee; pops is set to
1109       the number of bytes that were pushed onto the stack.
1110    5. The cproto decoder (compiler/Semant/types/cproto.sml) now can
1111       distinguish between "ccall" and "stdcall".
1112    6. The UNIMP instruction has been added to the supported Sparc instruction
1113       set. (This is needed for implementing the official C calling convention
1114       on this architecture.)
1115    7. I fixed some of the .cm files under src/MLRISC/Tools to make them
1116       work with the latest CM.
1117    
1118    ----------------------------------------------------------------------
1119    Name: Matthias Blume
1120    Date: 2001/06/05 15:10:00 EDT
1121    Tag: blume-20010605-cm-index
1122    Description:
1123    
1124    0. The "lambdasplit" parameter for class "sml" in CM has been documented.
1125    
1126    1. CM can now generate "index files".  These are human-readable files
1127       that list on a per-.cm-file basis each toplevel symbol defined or
1128       imported.  The location of the index file for
1129       <p>/<d>.cm is <p>/CM/INDEX/<d>.cm.
1130       To enable index-file generation, set CM.Control.generate_index to true
1131       or export an environment-symbol: export CM_GENERATE_INDEX=true.
1132    
1133       The CM manual has been updated accordingly.
1134    
1135    2. I made some slight modifications to the c-calls API in MLRISC.
1136    
1137         a) There is now a callback to support saving/restoring of
1138            dedicated but caller-save registers around the actual call
1139            instruction.
1140         b) One can optionally specify a comment-annotation for the
1141            call instruction.
1142    
1143    3. SML/NJ (mlriscGen.sml) uses this new API for the rawccall primop.
1144       (For example, the comment annotation shows the C prototype of
1145        the function being called.)
1146    
1147    ----------------------------------------------------------------------
1148    Name: Matthias Blume
1149    Date: 2001/06/01 13:30:00 EDT
1150    Tag: blume-20010601-nlffi-cleanup
1151    Description:
1152    
1153    This is mostly a cleanup of MLFFI stuff:
1154    
1155       - some signature files have been put into a more exposed place
1156       - the ugly 'f type parameter is gone (simplifies types tremendously!)
1157       - ml-nlffigen changed accordingly
1158       - tutorial updated
1159    
1160    Other changes:
1161    
1162       - author's affiliation in CM manual(s) updated
1163       - some more recognized keywords added to Allen's sml.sty
1164    
1165    ----------------------------------------------------------------------
1166    Name: Matthias Blume
1167    Date: 2001/05/25 15:30:00 EDT
1168    Tag: blume-20010525-iptr
1169    Description:
1170    
1171      - put the official 110.33-README (as it appears on the ftp server) under
1172        CVS
1173      - fixed a small bug related to incomplete pointer types in
1174        ml-nlffigen
1175      - small cosmetic change to the ml-nlffi-lib's "arr" type constructor
1176        (it does not need the 'f type parameter)
1177    
1178    ----------------------------------------------------------------------
1179    Name: Matthias Blume
1180    Date: 2001/05/23 14:30:00 EDT
1181    Tag: Release_110_33
1182    Description:
1183    
1184    New version number (110.33).  New bootfiles.
1185    
1186    ----------------------------------------------------------------------
1187    Name: Matthias Blume
1188    Date: 2001/05/22 18:06:00 EDT
1189    Tag: blume-20010522-targets
1190    Description:
1191    
1192    Made install.sh use file config/targets.customized if it exists, falling
1193    back to config/targets if it doesn't.  This way one can have a customized
1194    version of the targets file without touching the "real thing", thus
1195    eliminating the constant fear of accidentally checking something bogus
1196    back into the CVS repository...  (File config/targets.customized must
1197    not be added to the repository!)
1198    
1199    ----------------------------------------------------------------------
1200    Name: Matthias Blume
1201    Date: 2001/05/22 16:30:00 EDT
1202    Tag: blume-20010522-minitut
1203    Description:
1204    
1205    1. Bug fix in ml-nlffigen; now (hopefully) correctly handling
1206       struct returns.
1207    2. Added src/ml-nlffi-lib/Doc/mini-tutorial.txt.  This is some very
1208       incomplete, preliminary documentation for NLFFI.
1209    
1210    ----------------------------------------------------------------------
1211    Name: Matthias Blume
1212    Date: 2001/05/14 11:30:00 EDT
1213    Tag: blume-20010514-script
1214    Description:
1215    
1216    Some bugs in install script fixed.
1217    
1218    In addition to that I also made a slight change to the NLFFI API:
1219    Functors generated by ml-nlffigen now take the dynamic library as a
1220    straight functor argument, not as a suspended one.  (The original
1221    functor code used to force the suspension right away anyway, so there
1222    was nothing gained by this complication of the interface.)
1223    
1224    ----------------------------------------------------------------------
1225    Name: Matthias Blume
1226    Date: 2001/05/11 14:35:00 EDT
1227    Tag: blume-20010511-ml-nlffi
1228    Description:
1229    
1230    I finally took the plunge and added my new FFI code to the main
1231    repository.  For x86-linux it is now ready for prime-time.
1232    
1233    There are two new subdirectories of "src":
1234    
1235      - ml-nlffi-lib:
1236           The utility library for programs using the FFI interface.
1237           Here is the implementation of $/c.cm and its associated low-level
1238           partners $/c-int.cm and $/memory.cm.
1239      - ml-nlffigen:
1240           A stand-alone program for generating ML glue code from C source
1241           code.
1242    
1243    Building ml-nlffigen requires $/ckit-lib.cm.
1244    
1245    The config/install.sh script has been updates to do the Right Thing
1246    (hopefully).
1247    
1248    Notice that the source tree for the C-Kit will not be put under "src"
1249    but directly under the installation root directory.  (This is the
1250    structure that currently exists on the CVS server when you check out
1251    module "sml".)  Fortunately, config/install.sh knows about this oddity.
1252    
1253    Bugs: No documentation yet.
1254    
1255    ----------------------------------------------------------------------
1256    Name: Matthias Blume
1257    Date: 2001/05/09 16:35:00 EDT
1258    Tag: blume-20010509-cpscontract
1259    Description:
1260    
1261    Fixed a bug in the accounting code in cpsopt/contract.sml.  (The
1262    wrapper/unwrapper elimination did not decrement usage counts and some
1263    dead variables got overlooked by the dead-up logic.)
1264    
1265    ----------------------------------------------------------------------
1266    Name: Lal George
1267    Date: 2001/05/08  17:26:09 EDT
1268    Tag: george-20010508-omit-frameptr
1269    Description:
1270    
1271    Changes to implement the omit-frame-pointer optimization to support
1272    raw C calls. For now, there is only support on the Intel x86, but
1273    other architectures will follow as more experience is gained with this.
1274    
1275    
1276    ----------------------------------------------------------------------
1277    Name: Matthias Blume
1278    Date: 2001/05/07 14:40:00 EDT
1279    Tag: blume-20010507-proxies
1280    Description:
1281    
1282    I made into "proxy libraries" all libraries that qualify for such a
1283    change.  (A qualifying library is a library that has another library or
1284    groups as its sole member and repeats that member's export list
1285    verbatim.  A proxy library avoids this repetition by omitting its export
1286    list, effectively inheriting the list that its (only) member exports.
1287    See the CM manual for more explanation.)
1288    The main effect is that explicit export lists for these libraries
1289    do not have to be kepts in sync, making maintenance a bit easier.
1290    
1291    I also added copyright notices to many .cm-files.
1292    
1293    Last but not least, I made a new set of bootfiles.
1294    
1295    ----------------------------------------------------------------------
1296    Name: Matthias Blume
1297    Date: 2001/05/04 17:00:00 EDT
1298    Tag: blume-20010504-cm-lsplit
1299    Description:
1300    
1301    0. John merged pending changes to $/smlnj-lib.cm
1302    
1303    1. Allen's previous change accidentally backed out of one of Lal's
1304       earlier changes.  I undid this mistake (re-introducing Lal's change).
1305    
1306    2. I used the new topOrder' function from graph-scc.sml (from $/smlnj-lib.cm)
1307       within the compiler where applicable.  There is some code simplification
1308       because of that.
1309    
1310    3. The "split" phase (in FLINT) is now part of the default list of phases.
1311       Compiler.Control.LambdaSplitting.* can be used to globally control the
1312       lambda-splitting (cross-module-inlining) engine.  In addition to that,
1313       it can now also be controlled on a per-source basis: CM has been taught
1314       a new tool parameter applicable to ML source files.
1315    
1316       - To turn lambda-splitting off completely:
1317            local open Compiler.Control.LambdaSplitting in
1318                val _ = set Off
1319            end
1320       - To make "no lambda-splitting" the global default (but allow per-source
1321         overriding); this is the initial setting:
1322            local open Compiler.Control.LambdaSplitting in
1323                val _ = set (Default NONE)
1324            end
1325       - To make "lambda-splitting with aggressiveness a" the global default
1326         (and allow per-source overriding):
1327            local open Compiler.Control.LambdaSplitting in
1328                val _ = set (Default (SOME a))
1329            end
1330    
1331       - To turn lambda-splitting off for a given ML souce file (say: a.sml)
1332         write (in the respective .cm-file):
1333            a.sml (lambdasplitting:off)
1334       - To turn lambda-splitting for a.sml on with minimal aggressiveness:
1335            a.sml (lambdasplitting:on)
1336       - To turn lambda-splitting for a.sml on with aggressiveness <a> (where
1337         <a> is a decimal non-negative integer):
1338            a.sml (lambdasplitting:<a>)
1339       - To turn lambda-splitting for a.sml on with maximal aggressiveness:
1340            a.sml (lambdasplitting:infinity)
1341       - To use the global default for a.sml:
1342            a.sml (lambdasplitting:default)
1343         or simply
1344            a.sml
1345    
1346    ----------------------------------------------------------------------
1347    Name: Allen Leung
1348    Date: 2001/05/04 01:57:00 EDT
1349    Tag: leunga-20010504-sync
1350    Description:
1351    
1352      MLRISC features.
1353    
1354      1. Fix to CMPXCHG instructions.
1355      2. Changed RA interface to allow annotations in callbacks.
1356      3. Added a new method to the stream interface to allow annotations updates.
1357    
1358    ----------------------------------------------------------------------
1359    Name: Matthias Blume
1360    Date: 2001/05/01 11:45:00 EDT
1361    Tag: blume-20010501-pcedittmp
1362    Description:
1363    
1364    Changed install.sh to use the current working directory instead of
1365    /usr/tmp for a temporary file (pcedittmp).  The previous choice
1366    of /usr/tmp caused trouble with MacOS X because of file premission
1367    problems.
1368    
1369    ----------------------------------------------------------------------
1370    Name: Matthias Blume
1371    Date: 2001/04/20 11:10:00 EDT
1372    Tag: blume-20010420-inMLflag
1373    Description:
1374    
1375     - added vp_limitPtrMask to vproc-state.h
1376       (for use by the raw-C-calls mechanism to implement proper interrupt
1377        handling)
1378     - made the ML compiler aware of various data-structure offsets so it
1379       can generate code for accessing the vp_inML flag and vp_limitPtrMask
1380     - tweaked mlriscGen.sml to have it emit interrupt-handling code for
1381       raw C-calls
1382    
1383    ----------------------------------------------------------------------
1384    Name: Lal George
1385    Date: 2001/04/20 09:15:28 EDT
1386    Tag: george-20010420-macosX
1387    Description:
1388    
1389     - Changes to port to Mac OS X; Darwin.
1390    
1391     - In the process I found that sqrt was broken on the PPC, because the
1392       fsqrt instruction is not implemented.
1393    
1394    ----------------------------------------------------------------------
1395    Name: Matthias Blume
1396    Date: 2001/04/18 12:45:00 EDT
1397    Tag: blume-20010418-ccalls
1398    Description:
1399    
1400     - fixed two off-by-4 errors in the x86-specific c-calls implementation
1401       (this bug prevented structure arguments containing pointers from being
1402        passed correctly)
1403     - changed the raw-C-call code in mlriscGen.sml in such a way that
1404       structure arguments are represented as a pointer to the beginning
1405       of the structure (instead of having a series of synthesized arguments,
1406       one for each structure member)
1407    
1408     - made makeml script's verbosity level configurable via environment
1409       variable (MAKEML_VERBOSITY)
1410    
1411     - eliminated placeholder implementations for f32l, w16s, i16s, and f32s
1412       in rawmem-x86.sml; we are now using the real thing
1413    
1414    ----------------------------------------------------------------------
1415    Name: Matthias Blume
1416    Date: 2001/03/22 16:25:00 EST
1417    Tag: blume-20010322-bootfiles
1418    Description:
1419    
1420    Created a new set of bootfiles (for your automatic installation convenience).
1421    
1422    ----------------------------------------------------------------------
1423    Name: Matthias Blume
1424    Date: 2001/03/22 15:10:00 EST
1425    Tag: blume-20010322-rawmem-parcm
1426    Description:
1427    
1428    1. All "raw memory access" primitives for the new FFI are implemented now
1429       (at least on the x86).
1430    2. Some further cleanup of CM's parallel make mechanism.
1431    
1432    ----------------------------------------------------------------------
1433    Name: Matthias Blume
1434    Date: 2001/03/19 17:53:00 EST
1435    Tag: blume-20010319-parallel
1436    Description:
1437    
1438    Parallel make (using compile servers) now works again.
1439    
1440    To this end, CM.stabilize and CMB.make have been modified to work in
1441    two passes when compile servers are attached:
1442       1. Compile everything, do not perform stabilization; this pass
1443          uses compile servers
1444       2. Stabilize everything; this pass does not use compile servers
1445    If there are no compile servers, the two passes are combined into one
1446    (as before).  Splitting the passes increases the inherent parallelism
1447    in the dependency graph because the entire graph including all
1448    libraries is available at the same time.  This, in turn, improves
1449    server utilization.  The downside is that the master process will
1450    have to do some extra work after compilation is done (because for
1451    technical reasons it must re-read all the binfiles during stabilization).
1452    
1453    ----------------------------------------------------------------------
1454    Name: Matthias Blume
1455    Date: 2001/03/16 12:22:00 EST
1456    Tag: blume-20010316-bootfiles
1457    Description:
1458    
1459    Created a new set of bootfiles (for your automatic installation convenience).
1460    
1461    ----------------------------------------------------------------------
1462    Name: Matthias Blume
1463    Date: 2001/03/16 11:00:00 EST
1464    Tag: blume-20010316-MLTREE-fixup
1465    Description:
1466    
1467    This is a minor fixup for an (untagged) earlier commit by Allen.
1468    (A file was missing).
1469    
1470    ----------------------------------------------------------------------
1471    Name: Allen Leung
1472    Date: Mon Mar  5 18:54:57 EST 2001
1473    Tag: leunga-20010305-cut-support
1474    
1475    1. New support for alternative control-flow in MLTREE.
1476       Currently we support
1477    
1478          FLOW_TO(CALL ...., [k1,...,kn])
1479    
1480       This is needed for 'cuts to' in C-- and try/handle-like constructs
1481       in Moby
1482    
1483       New assembler flag "asm-show-cutsto" to turn on control-flow debugging.
1484    
1485    2. Register Allocator
1486    
1487       Changes in interface [from Fermin, John]
1488    
1489    3. Alpha 8-bit SLL support [Fermin]
1490    
1491    4. All architectures
1492    
1493       A new module (ClusterExpandCopies) for expanding parallel copies.
1494    
1495    ----------------------------------------------------------------------
1496    Name: Allen Leung
1497    Date: 2001/02/27 23:07:00 EST
1498    Tag: leunga-20010227-minor-stuff
1499    
1500    1. Alpha bug fix for CMOVNE
1501    2. Handle mltree COND(..,FCMP ...,...)
1502    3. Bug fix in simplifier
1503    
1504    ----------------------------------------------------------------------
1505    Name: Matthias Blume
1506    Date: 2001/01/30 17:50:00 EST
1507    Tag: blume-20010130-sync
1508    Description:
1509    
1510    This is just a minor update to sync my devel branch with the main brach.
1511    The only visible change is the addition of some README files.
1512    
1513    ----------------------------------------------------------------------
1514    Name: Matthias Blume
1515    Date: 2001/01/12 23:30:00 JST
1516    Tag: blume-20010112-bootfiles
1517    Description:
1518    
1519    Made a new set of bootfiles that goes with the current state of the
1520    repository.
1521    
1522    ----------------------------------------------------------------------
1523    Name: Matthias Blume
1524    Date: 2001/01/12 21:20:00 JST
1525    Tag: blume-20010112-sync
1526    Description:
1527    
1528    I am just flushing out some minor changes that had accumulated in
1529    my private branch in order to sync with the main tree.  (This is
1530    mainly because I had CVS trouble when trying to merge _into_ my
1531    private branch.)
1532    
1533    Most people should be completely unaffected by this.
1534    
1535    ----------------------------------------------------------------------
1536    Name: Allen Leung
1537    Date: Thu Jan 11 21:03:00 EST 2001
1538    Tag: leunga-20010111-labexp=mltree
1539    Description:
1540    
1541    1.  Removed the type LabelExp and replace it by MLTree.
1542    2.  Rewritten mltree-simplify with the pattern matcher tool.
1543    3.  There were some bugs in alpha code generator which would break
1544        64-bit code generation.
1545    4.  Redo the tools to generate code with the
1546    5.  The CM files in MLRISC (and in src/system/smlnj/MLRISC)
1547        are now generated by perl scripts.
1548    
1549    ----------------------------------------------------------------------
1550    Name: Matthias Blume
1551    Date: 2001/01/10 21:55:00 JST
1552    Tag: blume-20010110-rcc
1553    Description:
1554    
1555    The RCC stuff now seems to work (but only on the x86).
1556    This required hacking of the c-calls interface (and -implementation) in
1557    MLRISC.
1558    
1559    Normal compiler users should be unaffected.
1560    
1561    ----------------------------------------------------------------------
1562    Name: Matthias Blume
1563    Date: 2001/01/09 01:20:00 JST
1564    Tag: blume-20010109-rcc
1565    Description:
1566    
1567    This is a fairly big patch, flushing out a large number of pending
1568    changes that I made to my development copy over the last couple of days.
1569    
1570    Of practical relevance at this moment is a workaround for a pickling
1571    bug that Allen ran into the other day.  The cause of the bug itself is
1572    still unknown and it might be hard to fix it properly, but the
1573    workaround has some merits of its own (namely somewhat reducing pickling
1574    overhead for certain libraries).  Therefore, I think this solution should
1575    be satisfactory at this time.
1576    
1577    The rest of the changes (i.e., the vast majority) has to do with my
1578    ongoing efforts of providing direct support for C function calls from
1579    ML.  At the moment there is a new primop "RAW_CCALL", typing magic
1580    in types/cproto.sml (invoked from FLINT/trans/translate.sml), a new
1581    case in the FLINT CPS datatype (RCC), changes to cps/convert.sml to
1582    translate uses of RAW_CCALL into RCC, and changes to mlriscGen.sml to
1583    handle RCC.
1584    
1585    The last part (the changes to mlriscGen.sml) are still known to be
1586    wrong on the x86 and not implemented on all other architectures.  But
1587    the infrastructure is in place. I had to change a few functor
1588    signatures in the backend to be able to route the CCalls interface
1589    from MLRISC there, and I had to specialize the mltree type (on the
1590    x86) to include the necessary extensions. (The extensions themselves
1591    were already there and redy to go in MLRISC/x86).
1592    
1593    Everything should be very happy as soon as someone helps me with
1594    mlriscGen.sml...
1595    
1596    In any case, nothing of this should matter to anyone as long as the
1597    new primop is not being used (which is going to be the case unless you
1598    find it where I hid it :). The rest of the compiler is completely
1599    unaffected.
1600    
1601    ----------------------------------------------------------------------
1602    Name: Matthias Blume
1603    Date: 2001/01/05 00:30:00 JST
1604    Tag: blume-20010105-primops
1605    Description:
1606    
1607    Added some experimental support for work that I am doing right now.
1608    These changes mostly concern added primops, but there is also a new
1609    experimental C library in the runtime system (but currently not enabled
1610    anywhere except on Linux/X86).
1611    
1612    In the course of adding primops (and playing with them), I discovered that
1613    Zhong's INL_PRIM hack (no type info for certain primops) was, in fact, badly
1614    broken.  (Zhong was very right he labeled this stuff as "major gross hack".)
1615    To recover, I made type information in INL_PRIM mandatory and changed
1616    prim.sml as well as built-in.sml accordingly.  The InLine structure now
1617    has complete, correct type information (i.e., no bottom types).
1618    
1619    Since all these changes mean that we need new binfiles, I also bumped the
1620    version number to 110.32.1.
1621    
1622    ----------------------------------------------------------------------
1623    Name: Matthias Blume
1624    Date: 2000/12/30 22:10:00 JST
1625    Tag: blume-20001230-various
1626    Description:
1627    
1628    Added proxy libraries for MLRISC and let MLRISC libraries refer
1629    to each other using path anchors.  (See CM manual for explanation.)
1630    
1631    Updated CM documentation.
1632    
1633    Fixed some bugs in CM.
1634    
1635    Implemented "proxy" libraries (= syntactic sugar for CM).
1636    
1637    Added "-quiet" option to makeml and changed runtime system accordingly.
1638    
1639    Added cleanup handler for exportML to reset timers and compiler stats.
1640    
1641    ----------------------------------------------------------------------
1642    Name: Lal George
1643    Date: 2000/12/22 22:22:58 EST 2000
1644    Tag: Release_110_32
1645    Description:
1646    
1647            Infinite precision used throughout MLRISC.
1648            see MLRISC/mltree/machine-int.sig
1649    
1650    ----------------------------------------------------------------------
1651    Name: Matthias Blume
1652    Date: 2000/12/22 23:16:00 JST
1653    Tag: blume-20001222-warn
1654    Description:
1655    
1656    Corrected wording and formatting of some CM warning message which I
1657    broke in my previous patch.
1658    
1659    ----------------------------------------------------------------------
1660    Name: Matthias Blume
1661    Date: 2000/12/22 21:20:00 JST
1662    Tag: blume-20001222-anchorenv
1663    Description:
1664    
1665    Fixed CM's handling of anchor environments in connection with CMB.make.
1666    
1667    ----------------------------------------------------------------------
1668    Name: Matthias Blume
1669    Date: 2000/12/22 13:15:00 JST
1670    Tag: blume-20001222-cleanup
1671    Description:
1672    
1673    Removed src/cm/ffi which does not (and did not) belong here.
1674    
1675    ----------------------------------------------------------------------
1676    Name: Matthias Blume
1677    Date: 2000/12/21 23:55:00 JST
1678    Tag: blume-20001221-exn
1679    Description:
1680    
1681    Probably most important: CM no longer silently swallows all exceptions
1682    in the compiler.
1683    Plus: some other minor CM changes.  For example, CM now reports some
1684    sizes for generated binfiles (code, data, envpickle, lambdapickle).
1685    
1686    ----------------------------------------------------------------------
1687    Name: Matthias Blume
1688    Date: 2000/12/15 00:01:05 JST
1689    Tag: blume-20001215-dirtool
1690    Description:
1691    
1692    - "dir" tool added.
1693    - improvements and cleanup to Tools structure
1694    - documentation updates
1695    
1696    ----------------------------------------------------------------------
1697    Name: Allen Leung
1698    Date: Thu Dec 14 03:45:24 EST 2000
1699    Description:
1700    Tag:  leunga-20001214-int-inf
1701    Description:
1702    
1703       In IntInf, added these standard functions, which are missing from our
1704    implementation:
1705    
1706        andb : int * int -> int
1707        xorb : int * int -> int
1708        orb  : int * int -> int
1709        notb : int -> int
1710         <<   : int * word -> int
1711        ~>>  : int * word -> int
1712    
1713       Not tested, I hope they are correct.
1714    
1715    ----------------------------------------------------------------------
1716    Name: Allen Leung
1717    Date: Fri Dec  8 19:23:26 EST 2000
1718    Description:
1719    Tag:  leunga-20001208-nowhere
1720    Description:
1721    
1722      Slight improvements to the 'nowhere' tool to handle OR-patterns,
1723    to generate better error messages etc.  Plus a brief manual.
1724    
1725    ----------------------------------------------------------------------
1726    Name: Lal George
1727    Date: 2000/12/08 09:54:02 EST 2000
1728    Tag: Release_110_31
1729    Description:
1730    
1731    - Version 110.31
1732    ----------------------------------------------------------------------
1733    Name: Allen Leung
1734    Date: Thu Dec  7 22:01:04 EST 2000
1735    Tag:  leunga-20001207-cell-monster-hack
1736    Description:
1737    
1738    Major MLRISC internal changes.  Affect all clients.
1739    Summary:
1740    
1741    1.  Type CELLS.cell = int is now replaced by a datatype.
1742        As a result, the old regmap is now gone.  Almost all interfaces
1743        in MLRISC change as a consequence.
1744    
1745    2.  A new brand version of machine description tool (v3.0) that generates
1746        modules expecting the new interface.  The old version is removed.
1747    
1748    3.  The RA interface has been further abstracted into two new functors.
1749        RISC_RA and X86RA.  These functors have much simpler interfaces.
1750        [See also directory MLRISC/demo.]
1751    
1752    4.  Some other new source->source code generation tools are available:
1753    
1754        a. MLRISC/Tools/RewriteGen -- generate rewriters from rules.
1755        b. MLRISC/Tools/WhereGen -- expands conditional pattern matching rules.
1756           I use this tool to generate the peephole optimizers---with the new
1757           cell type changes, peephole rules are becoming difficult to write
1758           without conditional pattern matching.
1759    
1760    5.  More Intmap -> IntHashTable change.  Previous changes by Matthias didn't
1761        cover the entire MLRISC source tree so many things broke.
1762    
1763    6.  CM files have been moved to the subdirectory MLRISC/cm.
1764        They are moved because there are a lot of them and they clutter up the
1765        root dir.
1766    
1767    7.  More detailed documentation to come...
1768    
1769        NOTE: To rebuild from 110.30 (ftp distribution), you'll have to do
1770        a makeml -rebuild first.  This is because of other other
1771        changes that Matthias has made (see below).
1772    
1773    
1774    ----------------------------------------------------------------------
1775    Name: Matthias Blume
1776    Date: 2000/11/30 23:12:00 JST
1777    Tag: blume-20001130-filereorg
1778    Description:
1779    
1780    Some manual updates and some file reorganizations in CM.
1781    
1782    ----------------------------------------------------------------------
1783    Name: Matthias Blume
1784    Date: 2000/11/24 17:45:00 JST
1785    Tag: blume-20001124-link
1786    Description:
1787    
1788    Drastically improved link traversal code for the case that the dynamic
1789    value was already loaded at bootstrap time.  As a result, CM and CMB
1790    now both load blazingly fast -- even on a very slow machine.  Also,
1791    memory consumption has been further reduced by this.
1792    
1793    Warning: The format of the PIDMAP file has changed.  THerefore, to
1794    bootstrap you have to do this:
1795    
1796    1. Run CMB.make
1797    2. Make a symbolic link for the boot directory:
1798         ln -s sml.boot.ARCH-OS xxx
1799    3. "Rebuild" the boot directory:
1800         ./makeml -boot xxx -rebuild sml ; rm xxx
1801    4. Boot normally:
1802          ./makeml
1803    
1804    ----------------------------------------------------------------------
1805    Name: Matthias Blume
1806    Date: 2000/11/21 21:20:00 JST
1807    Tag: blume-20001121-tools
1808    Description:
1809    
1810    Continued hacking on autoloading problem -- with success this time.
1811    Also changed tool-plugin mechanism.  See new CM manual.
1812    
1813    ----------------------------------------------------------------------
1814    Name: Matthias Blume
1815    Date: 2000/11/19 14:30:00 JST
1816    Tag:  blume-20001119-autoload
1817    Description:
1818    
1819    Some hacking to make autoloading faster.  Success for CMB, no success
1820    so far for CM.  There is a reduced structure CM' that autoloads faster.
1821    (This is a temporary, non-documented hack to be eliminated again when
1822    the general problem is solved.)
1823    
1824    ----------------------------------------------------------------------
1825    Name: Matthias Blume
1826    Date: 2000/11/17 14:10:00 JST
1827    Tag: blume-20001117-pickle-lib
1828    Description:
1829    
1830    1. Eliminated comp-lib.cm
1831    2. Made pickle-lib.cm
1832    3. Eliminated all uses of intset.sml (from comp-lib.cm)
1833    4. Replaced all uses of intmap.{sig,sml} (from comp-lib.cm) with
1834       equivalent constructs from smlnj-lib.cm (INtHashTable).
1835    5. Point 4. also goes for those uses of intmap.* in MLRISC.
1836       Duplicated intmap modules thrown out.
1837    6. Hunted down all duplicated SCC code and replaced it with
1838       equivalent stuff (GraphSCCFn from smlnj-lib.cm).
1839    7. Rewrote Feedback module.
1840    8. Moved sortedlist.sml into viscomp-lib.cm.  Eventually it
1841       should be thrown out and equivalent modules from smlnj-lib.cm
1842       should be used (IntRedBlackSet, IntListSet, ...).
1843    
1844    Confirmed that compiler compiles to fixpoint.
1845    
1846    ----------------------------------------------------------------------
1847    Name: Allen Leung
1848    Date: 2000/11/10 18:00:00
1849    Tag: leunga-20001110-new-x86-fp
1850    
1851    A new x86 floating point code generator has been added.
1852    By default this is turned off.  To turn this on, do:
1853    
1854        CM.autoload "$smlnj/compiler.cm";
1855        Compiler.Control.MLRISC.getFlag "x86-fast-fp" := true;
1856    
1857    Changes:
1858    
1859    1.  Changed FTAN to FPTAN so that the assembly output is correct.
1860    2.  Changed the extension callback for FTANGENT to generate:
1861    
1862              fptan
1863              fstp  %st(0)
1864        instead of
1865              fptan
1866              fstpl ftempmem
1867    
1868    3.  Numerous assembly fixes for x86.
1869    
1870    5.  Cleaned up the machine code output module x86/x86MC.sml and added
1871        support for a whole bunch of instructions and addressing modes:
1872    
1873          fadd/fsub/fsubr/fmul/fdiv/fdivr  %st, %st(n)
1874          faddp/fsubp/fsubrp/fmulp/fdivp/fdivrp  %st, %st(n)
1875          fadd/fsub/fsubr/fmul/fdiv/fdivr  %st(n), %st
1876          fiadd/fisub/fisubr/fimul/fidiv/fidivr mem
1877          fxch %st(n)
1878          fld %st(n)
1879          fst %st(n)
1880          fst mem
1881          fstp %st(n)
1882          fucom %st(n)
1883          fucomp %st(n)
1884    
1885        All these are now generated when the fast fp mode is turned on.
1886    
1887    6.  Removed the dedicated registers %st(0), ..., %st(7) from X86CpsRegs
1888    
1889    ----------------------------------------------------------------------
1890    Name: Matthias Blume
1891    Date: 2000/11/09 11:20:00 JST
1892    Tag: blume-20001109-scc
1893    Description:
1894    
1895    Eliminated some code duplication:
1896    
1897    1. Added "where" clause to GraphSCCFn in SML/NJ Library.
1898       (Otherwise the functor is useless.)
1899    2. Used GraphSCCFn where SCCUtilFun was used previously.
1900    3. Got rid of SCCUtilFun (in comp-lib.cm).
1901    
1902    ----------------------------------------------------------------------
1903    Name: Lal George
1904    Date: 2000/11/06 09:02:21 EST 2000
1905    Tag: Release_110_30
1906    Description:
1907    
1908    - Version 110.30
1909    ----------------------------------------------------------------------
1910    Name: Matthias Blume
1911    Date: 2000/11/04 14:45:00
1912    Tag: blume-20001104-mlbuild
1913    Description:
1914    
1915    - Made ml-build faster on startup.
1916    - Documentation fixes.
1917    
1918    ----------------------------------------------------------------------
1919    Name: Matthias Blume
1920    Date: 2000/11/02 17:00:00 JST
1921    Tag: blume-20001102-condcomp
1922    Description:
1923    
1924    - Small tweaks to pickler -- new BOOTFILES!
1925    - Version bumped to 110.29.2.
1926    - Added conditional compilation facility to init.cmi (see comment there).
1927    ----------------------------------------------------------------------
1928    Name: Allen Leung
1929    Date: 2000/10/23 19:31:00
1930    Tag: leunga-20001023-demo-ra
1931    
1932    1. Minor RA changes that improves spilling on x86 (affects Moby and C-- only)
1933    2. Test programs for the graph library updated
1934    3. Some new MLRISC demo programs added
1935    
1936    ----------------------------------------------------------------------
1937    Name: Matthias Blume
1938    Date: 2000/08/31 22:15:00 JST
1939    Tag: blume-20001017-errmsg
1940    Description:
1941    
1942    More error message grief: Where there used to be no messages, there
1943    now were some that had bogus error regions.  Fixed.
1944    
1945    ----------------------------------------------------------------------
1946    Name: Matthias Blume
1947    Date: 2000/08/31 17:30:00 JST
1948    Tag: blume-20001017-v110p29p1
1949    Description:
1950    
1951    I made a version 110.29.1 with new bootfiles.
1952    
1953    Changes:  Modified pickler/unpickler for faster and leaner unpickling.
1954              CM documentation changes and a small bugfix in CM's error reporting.
1955    
1956    ----------------------------------------------------------------------
1957    Name: Lal George
1958    Date: 2000/09/27 14:42:35 EDT
1959    Tag: george-20000927-nodestatus
1960    Description:
1961    
1962    Changed the type of the nodestatus, so that:
1963    
1964            SPILLED(~1)             is now SPILLED
1965            SPILLED(m) where m>=0   is now MEMREG(m)
1966            SPILLED(s) where s<~1   is now SPILL_LOC(~s)
1967    
1968    ----------------------------------------------------------------------
1969    Name: Matthias Blume
1970    Date: 2000/09/07 14:45:00 JST
1971    Tag: blume-20000907-cmerrmsg
1972    Description:
1973    
1974    Small tweak to CM to avoid getting ML syntax error messages twice.
1975    
1976    ----------------------------------------------------------------------
1977    Name: Matthias Blume
1978    Date: 2000/08/31 18:00:00 JST
1979    Tag: blume-20000831-cvsbootfiles
1980    Description:
1981    
1982    New URL for boot files (because the 110.29 files on the BL server do
1983    now work correctly with my updated install scripts for yacc and lex).
1984    
1985    ----------------------------------------------------------------------
1986    Name: Matthias Blume
1987    Date: 2000/08/08 12:33:00 JST
1988    Tag: blume-20000808-manual
1989    Description:
1990    
1991    Tiny update to CM manual.
1992    
1993    ----------------------------------------------------------------------
1994    Name: Allen Leung
1995    Date: 2000/08/7 19:31:00
1996    Tag: leunga-20000807-a-whole-bunch-of-stuff
1997    
1998      Moby, C--, SSA, x86, machine descriptions etc.  Should only affect C--
1999    and Mobdy.
2000    
2001    1.  x86
2002    
2003       a.  Fixes to peephole module by John and Dan.
2004       b.  Assembly fix to SETcc by Allen.
2005       c.  Fix to c-call by John.
2006       d.  Fix to spilling by John.  (This one deals with the missing FSTPT case)
2007       e.  Instruction selection optimization to SETcc as suggested by John.
2008    
2009           For example,
2010    
2011            MV(32, x, COND(32, CMP(32, LT, a, b), LI 1, LI 0))
2012    
2013           should generate:
2014    
2015            MOVL a, x
2016            SUBL b, x
2017            SHRL 31, x
2018    
2019    2.  IR stuff
2020    
2021         A bunch of new DJ-graph related algorithms added.  These
2022         speed up SSA construction.
2023    
2024    3.  SSA + Scheduling
2025    
2026         Added code for SSA and scheduling to the repository
2027    
2028    ----------------------------------------------------------------------
2029    Name: Lal George
2030    Date: 2000/07/27 11:53:14 EDT
2031    
2032    Tag: lal-20000727-linux-ppc
2033    Description:
2034    
2035     Made changes to support Linux PPC.
2036     p.s. I have confirmation that the 110.29 boot files work fine.
2037    
2038    ----------------------------------------------------------------------
2039    Name: Matthias Blume
2040    Date: 2000/07/27 17:40:00 JST
2041    Tag: blume-20000727-scripts
2042    Description:
2043    
2044    !!!! WARNING !!!!
2045    You must recompile the runtime system!
2046    !!!! WARNING !!!!
2047    
2048    This is basically another round of script-enhancements:
2049    
2050    1. sml, ml-build, and ml-makedepend accept options -D and -U to define
2051       and undefine CM preprocessor symbols.
2052    
2053    2. ml-build avoids generating a new heap image if it finds that the
2054       existing one is still ok.  (The condition is that no ML file had to
2055       be recompiled and all ML files are found to be older that the heap
2056       file.)
2057    
2058       To make this work smoothly, I also hacked the runtime system as
2059       well as SMLofNJ.SysInfo to get access to the heap image suffix
2060       (.sparc-solaris, ...) that is currently being used.
2061    
2062       Moreover, the signature of CM.mk_standalone has changed.  See the
2063       CM manual.
2064    
2065    3. ml-makedepend accepts additional options -n, -a, and -o.  (See the
2066       CM manual for details.)
2067    
2068    4. More CM manual updates:
2069        - all of the above has been documented.
2070        - there is now a section describing the (CM-related) command line
2071          arguments that are accepted by the "sml" command
2072    
2073    ----------------------------------------------------------------------
2074    Name: Matthias Blume
2075    Date: 2000/07/25 16:20:00 JST
2076    Tag: blume-20000725-makedepend
2077    Description:
2078    
2079    Added a script called ml-makedepend.  This can be used in makefiles
2080    for Unix' make in a way very similar to the "makedepend" command for
2081    C.
2082    
2083    The script internally uses function CM.sources.
2084    
2085    Synopsis:
2086    
2087        ml-makedepend [-f makefile] cmfile targetname
2088    
2089    The default for the makefile is "makefile" (or "Makefile" should
2090    "makefile" not exist).
2091    
2092    ml-makedepend adds a cmfile/targetname-specific section to this
2093    makefile (after removing the previous version of this section).  The
2094    section contains a single dependency specification with targetname on
2095    the LHS (targetname is an arbitrary name), and a list of files derived
2096    from the cmfile on the RHS.  Some of the files on the RHS are
2097    ARCH/OPSYS-specific.  Therefore, ml-makedepend inserts references to
2098    "make" variables $(ARCH) and $(OPSYS) in place of the corresponding
2099    path names.  The makefile writer is responsible for making sure that
2100    these variables have correct at the time "make" is invoked.
2101    
2102    ----------------------------------------------------------------------
2103    Name: Matthias Blume
2104    Date: 2000/07/22 23:30:00 JST
2105    Tag: blume-20000722-urlupdate
2106    Description:
2107    
2108    Changed BOOT and config/srcarchiveurl to point to BL server:
2109    
2110        ftp://ftp.research.bell-labs.com/dist/smlnj/working/110.29/
2111    
2112    ----------------------------------------------------------------------
2113    Name: Matthias Blume
2114    Date: 2000/07/18 18:00:00 JST
2115    Tag: blume-20000718-Version_110_29
2116    Description:
2117    
2118    1. Updated src/compiler/TopLevel/main/version.sml to version 110.29
2119    
2120    2. Updated config/version to 110.29
2121    
2122    3. Updated config/srcarchiveurl
2123    
2124    3. New boot files!
2125       ftp://ftp.cs.princeton.edu/pub/people/blume/sml/110.29-autofetch
2126    
2127    ----------------------------------------------------------------------
2128    Name: Matthias Blume
2129    Date: 2000/07/11 13:58:00 JST
2130    Tag: blume-20000711-doctypo
2131    Description:
2132    
2133    Fixed a few typos in CM manual.
2134    
2135    ----------------------------------------------------------------------
2136    Name: Allen Leung
2137    Date: 2000/06/15 00:38:00
2138    Tag: leunga-20000704-sparc-x86
2139    
2140    1. x86 peephole improvement sp += k; sp -= k => nop  [from John]
2141    2. fix to x86 RET bug [found by Dan Grossman]
2142    3. sparc assembly bug fix for ticc instructions [found by Fermin]
2143    
2144       Affects c-- and moby only
2145    
2146    ----------------------------------------------------------------------
2147    Name: Matthias Blume
2148    Date: 2000/07/04 15:26:00
2149    Tag: blume-20000704-trigger
2150    Description:
2151    
2152    1. Improvements to CM manual.
2153    2. SMLofNJ.Internals.BTrace.trigger reinstated as an alternative way
2154       of getting a back-trace.  The function, when called, raises an
2155       internal exception which explicitly carries the full back-trace history,
2156       so it is unaffected by any intervening handle-raise pairs ("trivial"
2157       or not).  The interactive loop will print that history once it arrives
2158       at top level.
2159       Short of having all exceptions implicitly carry the full history, the
2160       recommended way of using this facility is:
2161         - compile your program with instrumentation "on"
2162         - run it, when it raises an exception, look at the history
2163         - if the history is "cut off" because of some handler, go and modify
2164           your program so that it explicitly calls BTrace.trigger
2165         - recompile (still instrumented), and rerun; look at the full history
2166    
2167    ----------------------------------------------------------------------
2168    Name: Matthias Blume
2169    Date: 2000/07/03 15:36:00 JST
2170    Tag: blume-20000702-manual
2171    Description:
2172    
2173    Small corrections and updates to CM manual.
2174    
2175    ----------------------------------------------------------------------
2176    Name: Matthias Blume
2177    Date: 2000/06/29 16:04:00 JST
2178    Tag: blume-20000629-yacctool
2179    Description:
2180    
2181    Changes:
2182    
2183    1. Class "mlyacc" now takes separate arguments to pass options to
2184       generated .sml- and .sig-files independently.
2185    2. Corresponding CM manual updates.
2186    3. BTrace module now also reports call sites.  (However, for loop clusters
2187       it only shows from where the cluster was entered.)  There are associated
2188       modifications to core.sml, internals.{sig,sml}, btrace.sml, and btimp.sml.
2189    
2190    ----------------------------------------------------------------------
2191    Name: Matthias Blume
2192    Date: 2000/06/27 16:51:00 JST
2193    Tag: blume-20000627-noweb
2194    Description:
2195    
2196    Changes:
2197    
2198     1. Implemented "subdir" and "witness" options for noweb tool.
2199        This caused some slight internal changes in CM's tool implementation.
2200     2. Fixed bug in "tool plugin" mechanism.  This is essentially cleaning
2201        some remaining issues from earlier path anchor changes.
2202     3. Updated CM manual accordingly.
2203    
2204     4. Changed implementation of back-tracing so that I now consider it
2205        ready for prime-time.
2206    
2207        In particular, you don't have to explicitly trigger the back-trace
2208        anymore.  Instead, if you are running BTrace-instrumented code and
2209        there is an uncaught exception (regardless of whether or not it was
2210        raised in instrumented code), the top-level evalloop will print
2211        the back-trace.
2212    
2213        Features:
2214    
2215          - Instrumented and uninstrumented code work together seemlessly.
2216            (Of course, uninstrumented code is never mentioned in actual
2217             back-traces.)
2218    
2219          - Asymptotic time- and space-complexity of instrumented code is
2220            equal to that of uninstrumented code.  (This means that
2221            tail-recursion is preserved by the instrumentation phase.)
2222    
2223          - Modules whose code has been instrumented in different sessions
2224            work together without problem.
2225    
2226          - There is no penalty whatsoever on uninstrumented code.
2227    
2228          - There is no penalty on "raise" expressions, even in
2229            instrumented code.
2230    
2231        A potential bug (or perhaps it is a feature, too):
2232    
2233          A back-trace reaches no further than the outermost instrumented
2234          non-trivial "raise".  Here, a "trivial" raise is one that is the
2235          sole RHS of a "handle" rule.  Thus, back-traces reach trough
2236    
2237               <exp> handle e => raise e
2238    
2239          and even
2240    
2241               <exp> handle Foo => raise Bar
2242    
2243          and, of course, through
2244    
2245               <exp> handle Foo => ...
2246    
2247         if the exception was not Foo.
2248    
2249         Back-traces always reach right through any un-instrumented code
2250         including any of its "handle" expressions, trivial or not.
2251    
2252       To try this out, do the following:
2253    
2254         - Erase all existing binfiles for your program.
2255           (You may keep binfiles for those modules where you think you
2256            definitely don't need back-tracing.)
2257         - Turn on back-trace instrumentation:
2258              SMLofNJ.Internals.BTrace.mode (SOME true);
2259         - Recompile your program.  (I.e., run "CM.make" or "use".)
2260         - You may now turn instrumentation off again (if you want):
2261              SMLofNJ.Internals.BTrace.mode (SOME false);
2262         - Run your program as usual.  If it raises an exception that
2263           reaches the interactive toplevel, then a back-trace will
2264           automatically be printed.  After that, the toplevel loop
2265           will print the exception history as usual.
2266    
2267    ----------------------------------------------------------------------
2268    Name: Matthias Blume
2269    Date: 2000/06/26 09:56:46 JST
2270    Tag: blume-20000626-setup
2271    Description:
2272    
2273    CM: - setup-parameter to "sml" added; this can be used to run arbitrary
2274          ML code before and after compiling a file (e.g., to set compiler
2275          flags)
2276    
2277    Compiler: - improved btrace API (in core.sml, internals.{sig,sml})
2278              - associated changes to btrace.sml (BTrace instrumentation pass)
2279              - cleaner implementation of btimp.sml (BTrace tracing and report
2280                module)
2281    
2282    CM manual: * new path encoding documented
2283               * description of setup-parameter to "sml" added
2284    
2285    The biggest user-visible change to back-tracing is that it is no
2286    longer necessary to compile all traced modules within the same
2287    session.  (This was a real limitation.)
2288    
2289    ----------------------------------------------------------------------
2290    Name: Matthias Blume
2291    Date: 2000/06/24 12:40:00 JST
2292    Tag: blume-20000624-startup
2293    Description:
2294    
2295    Fixes startup slowdown problem.  (I was calling SrcPath.sync a _tad_
2296    bit too often -- to put it mildly. :)
2297    
2298    ----------------------------------------------------------------------
2299    Name: Matthias Blume
2300    Date: 2000/06/23 18:20:00 JST
2301    Tag: blume-20000623-btrace
2302    Description:
2303    
2304    This updates adds a backtrace facility to aid programmers in debugging
2305    their programs.  This involves the following changes:
2306    
2307    1. Module system/smlnj/init/core.sml (structure _Core) now has hooks for
2308       keeping track of the current call stack.  When programs are compiled
2309       in a special mode, the compiler will insert calls to these hooks
2310       into the user program.
2311       "Hook" means that it is possible for different implementations of
2312       back-tracing to register themselves (at different times).
2313    
2314    2. compiler/MiscUtil/profile/btrace.sml implements the annotation phase
2315       as an Absyn.dec->Absyn.dec rewrite.  Normally this phase is turned off.
2316       It can be turned on using this call:
2317         SMLofNJ.Internals.BTrace.mode (SOME true);
2318       Turning it off again:
2319         SMLofNJ.Internals.BTrace.mode (SOME false);
2320       Querying the current status:
2321         SMLofNJ.Internals.BTrace.mode NONE;
2322       Annotated programs are about twice as big as normal ones, and they
2323       run a factor of 2 to 4 slower with a dummy back-trace plugin (one
2324       where all hooks do nothing).  The slowdown with a plugin that is
2325       actually useful (such as the one supplied by default) is even greater,
2326       but in the case of the default plugin it is still only an constant
2327       factor (amortized).
2328    
2329    3. system/Basis/Implementation/NJ/internals.{sig,sml} have been augmented
2330       with a sub-structure BTrace for controlling back-tracing.  In particular,
2331       the above-mentioned function "mode" controls whether the annotation
2332       phase is invoked by the compiler.  Another important function is
2333       "trigger": when called it aborts the current execution and causes
2334       the top-level loop to print a full back-trace.
2335    
2336    4. compiler/MiscUtil/profile/btimp.sml is the current default plugin
2337       for back-tracing.  It keeps track of the dynamic call stack and in
2338       addition to that it keeps a partial history at each "level" of that
2339       stack.  For example, if a tail-calls b, b tail-calls c, and c tail-calls
2340       d and b (at separate times, dynamically), then the report will show:
2341    
2342       GOTO   d
2343             /c
2344       GOTO  \b
2345       CALL   a
2346    
2347       This shows that there was an initial non-tail call of a, then a
2348       tail-call to b or c, looping behavior in a cluster of functions that
2349       consist of b and c, and then a goto from that cluster (i.e., either from
2350       b or from c) to d.
2351    
2352       Note that (depending on the user program) the amount of information
2353       that the back-trace module has to keep track of at each level is bounded
2354       by a constant.  Thus, the whole implementation has the same asymptotical
2355       complexity as the original program (both in space and in time).
2356    
2357    5. compiler/TopLevel/interact/evalloop.sml has been modified to
2358       handle the special exception SMLofNJ.Internals.BTrace.BTrace
2359       which is raised by the "trigger" function mentioned above.
2360    
2361    Notes on usage:
2362    
2363    - Annotated code works well together with unannotated code:
2364    Unannotated calls simply do not show up at all in the backtrace.
2365    
2366    - It is not a good idea to let modules that were annotated during
2367    different sessions run at the same time.  This is because the compiler
2368    chooses small integers to identify individual functions, and there
2369    will be clashes if different modules were compiled in separate sessions.
2370    (Nothing will crash, and you will even be told about the clashes, but
2371    back-trace information will in general not be useful.)
2372    
2373    - Back-tracing can be confused by callcc and capture.
2374    
2375    - The only way of getting a back-trace right now is to explicitly
2376    invoke the "trigger" function from your user program.  Eventually, we
2377    should make every exception carry back-trace information (if
2378    available).  But since this creates more overhead at "raise"-time
2379    (similar to the current exnHistory overhead), I have not yet
2380    implemented this.  (The implementation will be rather easy.)  With
2381    exceptions carrying back-trace information, this facility will be even
2382    more useful because users don't need to modify their programs...
2383    
2384    - While it is possible to compile the compiler with back-trace
2385    annotations turned on (I did it to get some confidence in
2386    correctness), you must make absolutely sure that core.sml and
2387    btimp.sml are compiled WITHOUT annotation!  (core.sml cannot actually
2388    be compiled with annotation because there is no core access yet, but
2389    if you compile btimp.sml with annotation, then the system will go into
2390    an infinite recursion and crash.)
2391    Since CM currently does not know about BTrace, the only way to turn
2392    annotations on and off for different modules of the compiler is to
2393    interrupt CMB.make, change the settings, and re-invoke it.  Of course,
2394    this is awkward and clumsy.
2395    
2396    Sample sessions:
2397    
2398    Standard ML of New Jersey v110.28.1 [FLINT v1.5], June 5, 2000
2399    - SMLofNJ.Internals.BTrace.mode (SOME true);
2400    [autoloading]
2401    [autoloading done]
2402    val it = false : bool
2403    - structure X = struct
2404    -     fun main n = let
2405    -         fun a (x, 0) = d x
2406    -           | a (x, n) = b (x, n - 1)
2407    -         and b (x, n) = c (x, n)
2408    -         and c (x, n) = a (x, n)
2409    -         and d x = e (x, 3)
2410    -         and e (x, 0) = f x
2411    -           | e (x, n) = e (x, n - 1)
2412    -         and f 0 = SMLofNJ.Internals.BTrace.trigger ()
2413    -           | f n = n * g (n - 1)
2414    -         and g n = a (n, 3)
2415    -     in
2416    -         f n
2417    -     end
2418    - end;
2419    structure X : sig val main : int -> int end
2420    - X.main 3;
2421    *** BACK-TRACE ***
2422    GOTO   stdIn:4.2-13.20: X.main[2].f
2423    GOTO-( stdIn:4.2-13.20: X.main[2].e
2424    GOTO   stdIn:4.2-13.20: X.main[2].d
2425         / stdIn:4.2-13.20: X.main[2].a
2426         | stdIn:4.2-13.20: X.main[2].b
2427    GOTO-\ stdIn:4.2-13.20: X.main[2].c
2428    CALL   stdIn:4.2-13.20: X.main[2].g
2429    GOTO   stdIn:4.2-13.20: X.main[2].f
2430    GOTO-( stdIn:4.2-13.20: X.main[2].e
2431    GOTO   stdIn:4.2-13.20: X.main[2].d
2432         / stdIn:4.2-13.20: X.main[2].a
2433         | stdIn:4.2-13.20: X.main[2].b
2434    GOTO-\ stdIn:4.2-13.20: X.main[2].c
2435    CALL   stdIn:4.2-13.20: X.main[2].g
2436    GOTO   stdIn:4.2-13.20: X.main[2].f
2437    GOTO-( stdIn:4.2-13.20: X.main[2].e
2438    GOTO   stdIn:4.2-13.20: X.main[2].d
2439         / stdIn:4.2-13.20: X.main[2].a
2440         | stdIn:4.2-13.20: X.main[2].b
2441    GOTO-\ stdIn:4.2-13.20: X.main[2].c
2442    CALL   stdIn:4.2-13.20: X.main[2].g
2443    GOTO   stdIn:4.2-13.20: X.main[2].f
2444    CALL   stdIn:2.15-17.4: X.main[2]
2445    -
2446    
2447    (Note that because of a FLINt bug the above code currently does not
2448    compile without BTrace turned on.)
2449    
2450    Here is another example, using my modified Tiger compiler:
2451    
2452    Standard ML of New Jersey v110.28.1 [FLINT v1.5], June 5, 2000
2453    - SMLofNJ.Internals.BTrace.mode (SOME true);
2454    [autoloading]
2455    [autoloading done]
2456    val it = false : bool
2457    - CM.make "sources.cm";
2458    [autoloading]
2459    ...
2460    [autoloading done]
2461    [scanning sources.cm]
2462    [parsing (sources.cm):parse.sml]
2463    [creating directory CM/SKEL ...]
2464    [parsing (sources.cm):tiger.lex.sml]
2465    ...
2466    [wrote CM/sparc-unix/semant.sml]
2467    [compiling (sources.cm):main.sml]
2468    [wrote CM/sparc-unix/main.sml]
2469    [New bindings added.]
2470    val it = true : bool
2471    - Main.compile ("../testcases/merge.tig", "foo.out");
2472    *** BACK-TRACE ***
2473    CALL   lib/semant.sml:99.2-396.21: SemantFun[2].transExp.trvar
2474    CALL   lib/semant.sml:99.2-396.21: SemantFun[2].transExp.trexp
2475    CALL   lib/semant.sml:289.3-295.22: SemantFun[2].transExp.trexp.check[2]
2476    GOTO   lib/semant.sml:289.3-295.22: SemantFun[2].transExp.trexp.check[2]
2477    CALL   lib/semant.sml:99.2-396.21: SemantFun[2].transExp.trexp
2478    CALL   lib/semant.sml:99.2-396.21: SemantFun[2].transExp.trexp
2479    CALL   lib/semant.sml:488.3-505.6: SemantFun[2].transDec.trdec[2].transBody[2]
2480         / lib/semant.sml:411.65-543.8: SemantFun[2].transDec
2481    CALL-\ lib/semant.sml:413.2-540.9: SemantFun[2].transDec.trdec[2]
2482    CALL   lib/semant.sml:99.2-396.21: SemantFun[2].transExp.trexp
2483    CALL   lib/semant.sml:8.52-558.4: SemantFun[2].transProg[2]
2484    CALL   main.sml:1.18-118.4: Main.compile[2]
2485    -
2486    
2487    ----------------------------------------------------------------------
2488    Name: Matthias Blumen
2489    Date: 2000/06/21 18:00:00 JST
2490    Tag: blume-20000621-manual
2491    Description:
2492    
2493    CM manual update: Path environments documented.
2494    
2495    ----------------------------------------------------------------------
2496    Name: Matthias Blume
2497    Date: 2000/06/19 13:40:00
2498    Tag: blume-20000619-manual
2499    Description:
2500    
2501    CM manual and system/README update.  This only covers the fact that
2502    there are no more implicit anchors.  (Path environments and the "bind"
2503    option to "cm" have yet to be documented.)
2504    
2505    ----------------------------------------------------------------------
2506    Name: Matthias Blume
2507    Date: 2000/06/19 11:05:00 JST
2508    Tag: blume-20000619-chdir-bugfix
2509    Description:
2510    
2511    Fixed a bug in new SrcPath module that sometimes led to a bad chDir call.
2512    
2513    ----------------------------------------------------------------------
2514    Name: Matthias Blume
2515    Date: 2000/06/18 22:00:10 JST
2516    Tag: blume-20000618-implicit-anchors-really-gone
2517    Description:
2518    
2519    I updates the previous HISTORY entry where I forgot to mention that
2520    implicit anchors are no longer with us.
2521    
2522    The current update also gets rid of the (now useless) controller
2523    CM.Control.implicit_anchors.
2524    
2525    ----------------------------------------------------------------------
2526    Name: Matthias Blume
2527    Date: 2000/06/16 17:30:00 JST
2528    Tag: blume-20000616-anchorenv
2529    Description:
2530    
2531    This patch implements the long anticipated (just kidding :) "anchor
2532    environment" mechanism.  In the course of doing this, I also
2533    re-implemented CM's internal "SrcPath" module from scratch.  The new
2534    one should be more robust in certain boundary cases.  In any case, it
2535    is a lot cleaner than its predecessor (IMHO).
2536    
2537    This time, although there is yet another boot file format change, I
2538    kept the unpickler backward-compatible.  As a result, no new bootfiles
2539    are necessary and bootstrapping is straightforward.  (You cannot read
2540    new bootfiles into an old system, but the other way around is no
2541    problem.)
2542    
2543    Visible changes:
2544    
2545    ** 0. Implicit path anchors (without the leading $-symbol) are no
2546    longer recognized at all. This means that such path names are not
2547    illegal either.  For example, the name basis.cm simply refers to a
2548    local file called "basis.cm" (i.e, the name is an ordinary path
2549    relative to .cm-files directory).  Or, to put it differently, only
2550    names that start with $ are anchored paths.
2551    
2552    ** 1. The $<singlearc> abbreviation for $/<singlearc> has finally
2553    vanished.
2554    
2555    John (Reppy) had critizised this as soon as I originally proposed and
2556    implemented it, but at that time I did not really deeply believe
2557    him. :) Now I came full-circle because I need the $<singlearc> syntax
2558    in another place where it cannot be seen as an abbreviation for
2559    $/<singlearc>.  To avoid the confusion, $<singlearc> now means what it
2560    seems to mean (i.e., it "expands" into the corresponding anchor
2561    value).
2562    
2563    However, when paths are used as members in CM description files, it
2564    continues to be true that there must be at least another arc after the
2565    anchor.  This is now enforced separately during semantic analysis
2566    (i.e., from a lexical/syntactical point of view, the notation is ok.)
2567    
2568    ** 2. The "cm" class now accepts an option "bind".  The option's value
2569    is a sub-option list of precisely two items -- one labeled "anchor"
2570    and the other one labeled "value".  As you might expect, "anchor" is
2571    used to specify an anchor name to be bound, and "value" specifies what
2572    the anchor is being bound to.
2573    
2574    The value must be a directory name and can be given in either standard
2575    syntax (including the possibility that it is itself an anchored path)
2576    or native syntax.
2577    
2578    Examples:
2579    
2580       foo.cm (bind:(anchor:bar value:$mystuff/bar))
2581       lib.cm (bind:(anchor:a value:"H:\\x\\y\\z"))  (* only works under windows *)
2582    
2583    and so on.
2584    
2585    The meaning of this is that the .cm-file will be processed with an
2586    augmented anchor environment where the given anchor(s) is/are bound to
2587    the given values(s).
2588    
2589    The rationale for having this feature is this: Suppose you are trying
2590    to use two different (already stable) libraries a.cm and b.cm (that
2591    you perhaps didn't write yourself).  Further, suppose each of these
2592    two libraries internally uses its own auxiliary library $aux/lib.cm.
2593    Normally you would now have a problem because the anchor "lib" can not
2594    be bound to more than one value globally.  Therefore, the project that
2595    uses both a.cm and b.cm must locally redirect the anchor to some other
2596    place:
2597    
2598       a.cm (bind:(anchor:lib value:/usr/lib/smlnj/a-stuff))
2599       b.cm (bind:(anchor:lib value:/usr/lib/smlnj/b-stuff))
2600    
2601    This hard-wires $lib/aux.cm to /usr/lib/smlnj/a-stuff/aux.cm or
2602    /usr/lib/smlnj/b-stuff/aux.cm, respectively.
2603    
2604    Hard-wiring path names is a bit inflexible (and CM will verbosely warn
2605    you when you do so at the time of CM.stabilize).  Therefore, you can
2606    also use an anchored path as the value:
2607    
2608      a.cm (bind:(anchor:lib value:$a-lib))
2609      b.cm (bind:(anchor:lib value:$b-lib))
2610    
2611    Now you can globally configure (using the usual CM.Anchor.anchor or
2612    pathconfig machinery) bindings for "a-lib" and "b-lib".  Since "lib"
2613    itself is always locally bound, setting it globally is no longer
2614    meaningful or necessary (but it does not hurt either).  In fact, "lib"
2615    can still be used as a global anchor for separate purposes.  As a
2616    matter of fact, one can locally define "lib" in terms of a global
2617    "lib":
2618    
2619      a.cm (bind:(anchor:lib value:$lib/a))
2620      b.cm (bind:(anchor:lib value:$lib/b))
2621    
2622    ** 3: The encoding of path names has changed.  This affects the way
2623    path names are shown in CM's progress report and also the internal
2624    protocol encoding used for parallel make.
2625    
2626    The encoding now uses one or more ':'-separated segments.  Each
2627    segments corresponds to a file that has been specified relative to the
2628    file given by its preceding segment.  The first segment is either
2629    relative to the CWD, absolute, or anchored.  Each segment itself is
2630    basically a Unix pathname; all segments but the first are relative.
2631    
2632    Example:
2633    
2634       $foo/bar/baz.cm:a/b/c.sml
2635    
2636    This path denotes the file bar/a/b/c.sml relative to the directory
2637    denoted by anchor "foo".  Notice that the encoding also includes
2638    baz.cm which is the .cm-file that listed a/b/c.sml.  As usual, such
2639    paths are resolved relative to the .cm-files directory, so baz.cm must
2640    be ignored to get the "real" pathname.
2641    
2642    To make this fact more obvious, CM puts the names of such "virtual
2643    arcs" into parentheses when they appear in progress reports. (No
2644    parentheses will appear in the internal protocol encoding.)  Thus,
2645    what you really see is:
2646    
2647      $foo/bar/(baz.cm):a/b/c.sml
2648    
2649    I find this notation to be much more informative than before.
2650    
2651    Another new feature of the encoding is that special characters
2652    including parentheses, colons, (back)slashes, and white space are
2653    written as \ddd (where ddd is the decimal encoding of the character).
2654    
2655    *** The CM manual still needs to be updated.
2656    
2657    ----------------------------------------------------------------------
2658    Name: Allen Leung
2659    Date: 2000/06/15 00:38:00
2660    Tag: leunga-20000615-x86-peephole
2661    
2662    x86 Peephole fix by Fermin.  Affects c-- and moby only.
2663    
2664    ----------------------------------------------------------------------
2665    Name: Matthias Blume
2666    Date: 2000/06/12 11:40:00
2667    Tag: blume-20000612-parmakefix
2668    Description:
2669    
2670    More cleanup after changing the file naming scheme: This time I
2671    repaired the parallel make mechanism for CMB.make which I broke earlier.
2672    
2673    ----------------------------------------------------------------------
2674    Name: Allen Leung
2675    Date: 2000/06/09 01:25:00
2676    Tag: leunga-20000609-various
2677    
2678    None of these things should affect normal SML/NJ operations
2679    
2680    1. Peephole improvements provided by Fermin (c--)
2681    2. New annotation DEFUSE for adding extra dependence (moby)
2682    3. New X86 LOCK instructions (moby)
2683    4. New machine description language for reservation tables (scheduling)
2684    5. Fixes to various optimization/analysis modules (branch chaining, dominator
2685       trees etc.)
2686    6. I've changed the CM files so that they can work with versions
2687       110.0.6, 110.25 and 110.28
2688    
2689    ----------------------------------------------------------------------
2690    Name: Matthias Blume
2691    Date: 2000/06/09 12:40:00
2692    Tag: blume-20000609-log
2693    Description:
2694    
2695    - Removed all(?) remaining RCS Log entries from sources.
2696    
2697    - Fixed bug in ml-yacc and ml-lex sources (use explicit anchors for
2698      anchored paths).
2699    
2700    ----------------------------------------------------------------------
2701    Name: Matthias Blume
2702    Date: 2000/06/07 17:00:00 JST
2703    Tag: blume-20000607-no-implicit-anchors
2704    Description:
2705    
2706    1. This update changes the default setting for
2707    CM.Control.implicit_anchors from true to false.  This means that
2708    implicit anchors are no longer permitted by default.  I also tried to
2709    make sure that nothing else still relies on implicit anchors.
2710    (This is the next step on the schedule towards a CM that does not even
2711    have the notion of implicit anchors anymore.)
2712    
2713    2. More CM manual updates.
2714    
2715    3. I managed to track down and fix the pickling bug I mentioned last
2716    time.  Because of the previously existing workaround, this entails no
2717    immediate practical changes.
2718    
2719    ----------------------------------------------------------------------
2720    Name: Matthias Blume
2721    Date: 2000/06/06 11:15:00 JST
2722    Tag: blume-20000606-lazierpickle
2723    Description:
2724    
2725    !!!! NEW BOOT FILES !!!!
2726    
2727    * The main purpose of this update is to make library pickles lazier in
2728    order to reduce the initial space penalty for autoloading a library.
2729    As a result, it is now possible to have $smlnj/compiler.cm
2730    pre-registered.  This should take care of the many complaints or
2731    inquiries about missing structure Compiler.  This required changes to
2732    CM's internal data structures and small tweaks to some algorithms.
2733    
2734    As a neat additional effect, it is no longer necessary (for the sake
2735    of lean heap image files) to distinguish between a "minimal" CM and a
2736    "full" CM.  Now, there is only one CM (i.e., the "full" version:
2737    $smlnj/cm.cm aka $smlnj/cm/full.cm), and it is always available at the
2738    interactive top level. ($smlnj/cm/minimal.cm is gone.)
2739    
2740    To make the life of compiler-hackers easier, "makeml" now also
2741    pre-registers $smlnj/cmb.cm (aka $smlnj/cmb/current.cm).  In other
2742    words, after you bootstrap a new sml for the first time, you will not
2743    have to autoload $smlnj/cmb.cm again afterwards.  (The first time
2744    around you will still have to do it, though.)
2745    
2746    * A second change consists of major updates to the CM manual.  There
2747    are now several appendices with summary information and also a full
2748    specification of the CM description file syntax.
2749    
2750    * In directory src/system I added the script "allcross".  This script
2751    invokes sml and cross-compiles the compiler for all supported
2752    architectures.  (Useful when providing a new set of boot files.)
2753    
2754    * There seems to be a latent bug in my "lazy pickles" mechanism.  I
2755    added a small tweak to pickle-util.sml to work around this problem,
2756    but it is not a proper fix yet.  I will investigate further.  (The
2757    effect of the bug was an inflation of library pickle size.)
2758    
2759    * Version number increased to 110.28.1 (to avoid compatibility problems).
2760    
2761    ----------------------------------------------------------------------
2762    Name: Allen Leung
2763    Date: 2000/05/25 17:28 EDT
2764    Tag: leunga-20000525-ra
2765    Description:
2766    
2767      Fixed a bug in freezing phase of the register allocator.
2768    
2769    ----------------------------------------------------------------------
2770    Name: Allen Leung
2771    Date: 2000/05/15 22:53 EDT
2772    Tag: leunga-20000515-alpha-x86-ra
2773    Description:
2774    
2775      1. Alpha
2776    
2777          Slight cleanup.  Removed the instruction SGNXL
2778    
2779      2. X86
2780    
2781          Added the following instructions to the instruction set:
2782    
2783            ROLx, RORx,
2784            BTx, BTSx, BTLx, BTRx,
2785            XCHGx, and variants with the LOCK prefix
2786    
2787      3. Register Allocation
2788    
2789          The module ra-rewrite-with-renaming has been improved.
2790    
2791      These have no effect on SML/NJ.
2792    
2793    ----------------------------------------------------------------------
2794    Name: Matthias Blume
2795    Date: 2000/05/15 16:20:00 JST
2796    Tag: blume-20000515-lightrebuild
2797    Description:
2798    
2799    1. I added an alternative to "-rebuild" to "makeml".  The difference is
2800       that prior to calling CMB.make' the CM-variable "LIGHT" will be
2801       defined.  In effect, the command will not build any cross-compiler
2802       backends and therefore finish more quickly.
2803    
2804       The "fixpt" script also takes a "-light" switch to be able to use
2805       this new facility while compiling for a fixpoint.
2806    
2807    2. I replaced all mentions of anchored paths in group owner specifications
2808       with simple relative paths (usually starting with "..").
2809       The rationale is that a library's internal workings should not be
2810       compromised by the lack of some anchor.  (An anchor is necessary
2811       for someone who wants to refer to the library by an anchored path,
2812       but it should not be necessary to build the same library in the first
2813       place.)
2814    
2815    3. I changed the way CM's tool mechanism determines the shell command
2816       string used for things like ml-yacc etc. so that it does not break
2817       when CM.Control.implicit_anchors is turned off.
2818    
2819    ----------------------------------------------------------------------
2820    Name: Matthias Blume
2821    Date: 2000/05/12 18:20:00 JST
2822    Tag: blume-20000512-ml-build
2823    Description:
2824    
2825    Fixed a bug in config/_ml-build that prevented ml-yacc and ml-lex from
2826    getting installed properly (by config/install.sh).
2827    
2828    ----------------------------------------------------------------------
2829    Name: Matthias Blume
2830    Date: 2000/05/12 17:30:00 JST
2831    Tag: blume-20000512-anchors
2832    Description:
2833    
2834    !!! NEW BOOT FILES !!!
2835    
2836    This change is in preparation of fading out support for "implicitly
2837    anchored path names".  I went through all sources and used the
2838    explicit (and relatively new) $-notation.  See system/README and the
2839    CM manual for more info on this.
2840    
2841    I also modified the anchoring scheme for some things such as "smlnj",
2842    "MLRISC", "cm", etc. to take advantage of the fact that explicit
2843    anchors are more expressive: anchor name and first arc do not have to
2844    coincide.  This entails the following user-visible change:
2845    
2846    You have to write $smlnj/foo/bar instead of smlnj/foo/bar.  In
2847    particular, when you fire up sml with a command-line argument, say,
2848    e.g.:
2849    
2850       sml '$smlnj/cmb.cm'
2851    
2852    At the ML toplevel prompt:
2853    
2854       CM.autoload "$smlnj/cmb.cm";
2855    
2856    There is also a new controller in CM.Control that can be used to turn
2857    off all remaining support for implicit anchors by saying:
2858    
2859        CM.autoload "$smlnj/
2860        #set CM.Control.implicit_anchors false;
2861    
2862    This causes CM to reject implicitly anchored paths.  This is (for the
2863    time being) less permissive than the "final" version where there will
2864    be no more such implicit anchors and relative paths will be just that:
2865    relative.
2866    
2867    The next step (version after next version?) will be to make the
2868    default for CM.Control.implicit_anchors false.  After the dust has
2869    settled, I can then produce the "final" version of this...
2870    
2871    Note: Since bootstrapping is a bit tricky, I provided new boot files.
2872    
2873    ----------------------------------------------------------------------
2874    Name: Matthias Blume
2875    Date: 2000/05/11 16:30:00 JST
2876    Tag: blume-20000511-sources
2877    Description:
2878    
2879    The main change is that I added function CM.sources as a generalized
2880    version of the earlier CM.makedepend.  This entails the following
2881    additional changes:
2882    
2883      - CM.makedepend has been dropped.
2884    
2885      - CM manual has been updated.
2886    
2887      - TOOLS signature and API have been changed.
2888    
2889    ----------------------------------------------------------------------
2890    Name: Allen Leung
2891    Date: 2000/05/10 21:17 EDT
2892    Tag: leunga-20000510-moby-c--ssa
2893    Description:
2894    
2895      Various bug fixes and new features for C--, Moby and MLRISC optimizations.
2896    None of these affect SML/NJ.
2897    
2898    1. Register Allocation
2899    
2900        a. A new ra spilling module (ra/ra-spill-with-renaming) is implemented.
2901           This module tries to remove local (i.e. basic block level) redundancies
2902           during spilling.
2903    
2904        b. A new framework for performing region based register allocation.
2905           Not yet entirely functional.
2906    
2907    2. X86
2908    
2909       a. DefUse for POP was missing the stack pointer [found by Lal]
2910       b. Reload for CALL was incorrect in X86Spill [found by John]
2911       c. Various fixes in X86Spill so that it can be used correctly for
2912          the new spilling module.
2913    
2914    3. SSA/IR
2915    
2916       a. New module ir/dj-dataflow.sml implements elimination based
2917          data flow analysis.
2918    
2919    4. MLRiscGen
2920    
2921       a. Fix for gc type annotation
2922    
2923    5. MDGen
2924    
2925       Various fixes for machine description -> ml code translation.  For ssa
2926       only.
2927    
2928    ----------------------------------------------------------------------
2929    Name: Allen Leung
2930    Date: 2000/05/08 22:17 EDT
2931    Tag: leunga-20000508-labexp
2932    Description:
2933    
2934      Fermin has found a few assembly problems with constant expressions
2935      generated in LabelExp.  Mostly, the problems involve extra parentheses,
2936      which choke on dumb assemblers.  This is his fix.
2937    
2938    ----------------------------------------------------------------------
2939    Name: Dave MacQueen
2940    Date: 2000/04/09 14:00 EDT
2941    Tag: dbm-20000502-Version_110_28
2942    Description:
2943    
2944    1. Updated src/compiler/TopLevel/main/version.sml to version 110.28
2945    
2946    2. Updated config/version to 110.28
2947    
2948    3. Updated config/srcarchiveurl
2949    
2950    3. New boot files!
2951       ftp://ftp.research.bell-labs.com/dist/smlnj/working/110.28/
2952    
2953    ----------------------------------------------------------------------
2954    Name: Matthias Blume
2955    Date: 2000/05/01 19:05:00 JST
2956    Tag: blume-20000501-noweb
2957    Description:
2958    
2959    A new noweb tool has been added.  The existing system is entirely
2960    unaffected by this, but some CM users have asked for renewed noweb
2961    support.  Everything is documented in the CM manual.
2962    
2963    New (plugin) libraries:
2964    
2965       noweb-tool.cm
2966       nw-ext.cm
2967    
2968    ----------------------------------------------------------------------
2969    Name: Dave MacQueen
2970    Date: 2000/04/30 12:40PM EDT
2971    Tag: dbm-20000430-bug_fixes
2972    Description:
2973    
2974    1. Fix for bug 1498
2975       smlnj/src/system/Basis/Implementation/Unsafe/object.sig
2976       smlnj/src/system/Basis/Implementation/Unsafe/object.sml
2977         added toRealArray function
2978       smlnj/src/compiler/MiscUtil/print/ppobj.sml
2979         added check for tag Obj.RealArray to array printing case in ppObj
2980    
2981    2. Fix for bug 1510
2982       smlnj/src/compiler/Semant/types/typesutil.sml
2983         fixed definition of dummyargs (used by equalTycon) so that
2984         dummy args are distinct types
2985    
2986    ----------------------------------------------------------------------
2987    Name: Matthias Blume
2988    Date: 2000/04/30 01:00:00 JST
2989    Tag: blume-20000430-versions
2990    Description:
2991    
2992    1. CM version numbering added.  This is an implementation of Lal's
2993       proposal for adding version numbers and version checking to .cm
2994       files.  Lal said that his proposal was just that -- a proposal.
2995       For the time being I went ahead and implemented it so that people
2996       can comment on it.  Everything is completely backward-compatible
2997       (except for the stable library format, i.e., new bootfiles!).
2998    
2999       As usual, see the CM manual for details.
3000    
3001    2. An alternative syntax for anchored paths has been implemented.
3002       Dave has recently voiced the same concerns that I had when I did
3003       this, so there should be some support.  My take is that eventually
3004       I will let support for the current syntax (where anchors are
3005       "implicit") fade out in favor of the new, explicit syntax.
3006       In order to be backward-compatible, both old and new syntax are
3007       currently supported.
3008    
3009       Again, see the CM manual for details.
3010    
3011    3. Parallel make is trying to be slightly smarter:  When the master
3012       process finds a "bottleneck", i.e., when there is only one
3013       compilation unit that can be compiled and everybody else is
3014       waiting on it, then it will simply compile it directly instead
3015       of clumsily telling one of the slaves to do it.
3016    
3017    4. Support for "unsharing" added.  This is necessary in order to be
3018       able to have two different versions of the same library running
3019       at the same time (e.g., for trying out a new MLRISC while still
3020       having the old MLRISC linked into the current compiler, etc.)
3021       See the CM manual.
3022    
3023    5. Simple "makedepend" functionality added for generating Makefile
3024       dependency information.  (This is rather crude at the moment.
3025       Expect some changes here in the future.)
3026    
3027    6. ".fun" added as a recognized suffix for ML files. Also documented
3028       explicitly in the manual that the fallback behavior (unknown suffix
3029       -> ML file) is not an official feature!
3030    
3031    7. Small changes to the pickler for stable libraries.
3032    
3033    8. Several internal changes to CM (for cleanup/improvement).
3034    
3035    
3036    !!!! NEW BINFILES !!!!
3037    
3038    ----------------------------------------------------------------------
3039    Name: Matthias Blume
3040    Date: 2000/04/28 17:30:00 JST
3041    Tag: blume-20000428-pathconfig
3042    Description:
3043    
3044    1. I changed config/install.sh to remove duplicate entries from the
3045       lib/pathconfig file at the end.  Moreover, the final version of
3046       lib/pathconfig is sorted alphabetically.  The same (sorting) is done
3047       in src/system/installml.
3048    
3049    2. The config/install.sh script now consistently uses relative
3050       pathnames in lib/pathconfig whenever the anchor is in the lib
3051       directory.  (So far this was true for the libraries that come
3052       pre-compiled and bundled as part of the bootfiles but not for
3053       libraries that are compiled by the script itself.)
3054    
3055    ----------------------------------------------------------------------
3056    Name: Matthias Blume
3057    Date: 2000/04/26 13:10:00 JST
3058    Tag: blume-20000426-fun_suffix
3059    Description:
3060    
3061    Added ".fun" as a recognized file name suffix (for ML code).
3062    
3063    ----------------------------------------------------------------------
3064    Name: Allen Leung
3065    Date: 2000/04/25 17:00:00 EST
3066    Tag: leunga-20000425-alpha-ra
3067    Description:
3068    
3069    1. Alpha
3070    
3071        PSEUDOARITH was missing in AlphaRewrite.  This causes an endless loop
3072    in C--.
3073    
3074    2. RA
3075    
3076       Added a flag "ra-dump-size" to print out the size of the flowgraph
3077       and the interference graph.
3078    
3079    ----------------------------------------------------------------------
3080    Name: Dave MacQueen
3081    Date: 2000/04/25/
3082    Tag: dbm-20000425-mlyacc_doc_examples
3083    Description:
3084      Updated mlyacc.tex sections 5 and 7 for SML '97 and CM.
3085      Updated all three examples in src/ml-yacc/examples to run
3086      under 110.* using CM.make.
3087    
3088    ----------------------------------------------------------------------
3089    Name: Allen Leung
3090    Date: 2000/04/20 23:04:00 EST
3091    Tag: leunga-20000420-ssa-c---stuff
3092    Description:
3093    
3094      This update synchronizes my repository with Yale's.  Most of these
3095    changes, however, do not affect SML/NJ at all (the RA is an exception).
3096    
3097    1. Register Allocator
3098    
3099       a. An improvement in the interference graph construction:
3100          Given a copy
3101    
3102                s <- t
3103    
3104          no interference edge between s and t is added for this definition of s.
3105    
3106       b. I've added two new spill heuristic modules that Fermin and I developed
3107          (in the new library RA.cm). These are unused in SML/NJ but maybe
3108          useful for others (Moby?)
3109    
3110    2. X86
3111    
3112       a. Various fixes in the backend provided by Fermin [C--] and Lal.
3113    
3114    3. Alpha
3115    
3116       a. Added the BSR instruction and code generation that goes with it [C--]
3117       b. Other fixes too numerous to recount provided by Fermin [C--]
3118    
3119    4. Regmaps
3120    
3121       a. The regmaps are not initialized with the identity physical bindings
3122          at creation time.  This is unneeded.
3123    
3124    5. MLRISC Optimizations
3125    
3126       a. The DJ-Graph module can now compute the iterated dominance frontiers
3127          intersects with liveness incrementally in linear time! Woohoo!
3128          This is now used in my new SSA construction algorithm.
3129    
3130       b. THe branch reorganization module is now smarter about linear chains of
3131          basic blocks.
3132    
3133    
3134    ----------------------------------------------------------------------
3135    Name: Matthias Blume
3136    Date: 2000/04/12 13:52:00 JST
3137    Tag: blume_main_v110p27_1
3138    Description:
3139    
3140    Changed install.sh script to handle archive files without version number
3141    and to use "boot.<arch>-<os>" instead of "sml.boot.<arch>-<os>" for the
3142    name of the boot file archive.
3143    
3144    ----------------------------------------------------------------------
3145    Name: Dave MacQueen
3146    Date: 2000/04/09 14:00 EDT
3147    Tag: dbm-20000410-Version_110_27
3148    Description:
3149    
3150    1. Updated src/compiler/TopLevel/main/version.sml to version 110.27
3151    
3152    2. Updated src/config/version to 110.27
3153    
3154    3. New boot files!
3155    
3156    ----------------------------------------------------------------------
3157    Name: Allen Leung
3158    Date: 2000/04/09 19:09:00 EST
3159    Tag: leunga-20000409-misc
3160    Description:
3161    
3162    1.  Yet another fix for x86 assembly for idivl, imull, mull and friends.
3163    
3164    2.  Miscellaneous improvements to MLRISC (unused in sml/nj)
3165    
3166    ----------------------------------------------------------------------
3167    Name: Stefan
3168    Date: 2000/04/07 10:00:00 EDT
3169    Tag: monnier-20000406-branch-handling
3170    Description:
3171    
3172    Improved handling of branches (mostly those generated from
3173    polymorphic equality), removed switchoff and changed the
3174    default optimization settings (more cpsopt and less flintopt).
3175    
3176    ----------------------------------------------------------------------
3177    Name: Allen Leung
3178    Date: 2000/04/06 01:30:00 EST
3179    Tag: leunga-20000406-peephole-x86-SSA-2
3180    Description:
3181    
3182       Forgot a few files.
3183    
3184    ----------------------------------------------------------------------
3185    Name: Allen Leung
3186    Date: 2000/04/06 00:36:00 EST
3187    Tag: leunga-20000406-peephole-x86-SSA
3188    Description:
3189    
3190    1.  New Peephole code
3191    
3192    2.  Minor improvement to X86 instruction selection
3193    
3194    3.  Various fixes to SSA and machine description -> code translator
3195    
3196    ----------------------------------------------------------------------
3197    Name: Matthias Blume
3198    Date: 2000/04/05 12:30:00 JST
3199    Tag: blume_main_v110p26p2_3
3200    Description:
3201    
3202    This update just merges three minor cosmetic updates to CM's sources
3203    to get ready for the 110.27 code freeze on Friday.  No functionality
3204    has changed.
3205    
3206    ----------------------------------------------------------------------
3207    Name: Allen Leung
3208    Date: 2000/04/04 19:39:00 EST
3209    Tag: leunga-20000404-x86-asm
3210    Description:
3211    
3212    1.  Fixed a problem in X86 assembly.
3213    
3214        Things like
3215    
3216           jmp %eax
3217           jmp (%eax)
3218    
3219        should be output as
3220    
3221           jmp *%eax
3222           jmp *(%eax)
3223    
3224    2.  Assembly output
3225    
3226          Added a new flag
3227    
3228              "asm-indent-copies" (default to false)
3229    
3230          When this flag is on, parallel copies will be indented an extra level.
3231    
3232    ----------------------------------------------------------------------
3233    Name: Allen Leung
3234    Date: 2000/04/04 03:18:00 EST
3235    Tag: leunga-20000404-C--Moby
3236    Description:
3237    
3238        All of these fixes are related to C--, Moby, and my own optimization
3239        stuff; so they shouldn't affect SML/NJ.
3240    
3241    1.  X86
3242    
3243        Various fixes related floating point, and extensions.
3244    
3245    2.  Alpha
3246    
3247        Some extra patterns related to loads with signed/zero extension
3248        provided by Fermin.
3249    
3250    3.  Assembly
3251    
3252        When generating assembly, resolve the value of client defined constants,
3253        instead of generating symbolic values.  This is controlled by the
3254        new flag "asm-resolve-constants", which is default to true.
3255    
3256    4.  Machine Descriptions
3257    
3258        a. The precedence parser was slightly broken when parsing infixr symbols.
3259        b. The type generalizing code had the bound variables reversed, resulting
3260           in a problem during arity raising.
3261        c. Various fixes in machine descriptions.
3262    
3263    ----------------------------------------------------------------------
3264    Name: Matthias Blume
3265    Date: 2000/04/03 16:05:00 JST
3266    Tag: blume_main_v110p26p2_2
3267    Description:
3268    
3269    I eliminated coreEnv from compInfo.  Access to the "Core" structure is
3270    now done via the ordinary static environment that is context to each
3271    compilation unit.
3272    
3273    To this end, I arranged that instead of "structure Core" as "structure
3274    _Core" is bound in the pervasive environment.  Core access is done via
3275    _Core (which can never be accidentally rebound because _Core is not a
3276    legal surface-syntax symbol).
3277    
3278    The current solution is much cleaner because the core environment is
3279    now simply part of the pervasive environment which is part of every
3280    compilation unit's context anyway.  In particular, this eliminates all
3281    special-case handling that was necessary until now in order to deal
3282    with dynamic and symbolic parts of the core environment.
3283    
3284    Remaining hackery (to bind the "magic" symbol _Core) is localized in the
3285    compilation manager's bootstrap compiler (actually: in the "init group"
3286    handling).  See the comments in src/system/smlnj/init/init.cmi for
3287    more details.
3288    
3289    I also tried to track down all mentions of "Core" (as string argument
3290    to Symbol.strSymbol) in the compiler and replaced them with a
3291    reference to the new CoreSym.coreSym.  Seems cleaner since the actual
3292    name appears in one place only.
3293    
3294    Binfile and bootfile format have not changed, but the switchover from
3295    the old "init.cmi" to the new one is a bit tricky, so I supplied new
3296    bootfiles anyway.
3297    
3298    ----------------------------------------------------------------------
3299    Name: Allen Leung
3300    Date: 2000/04/02 21:17:00 EST
3301    Tag: leunga-20000402-mltree
3302    Description:
3303    
3304       1. Renamed the constructor CALL in MLTREE by popular demand.
3305       2. Added a bunch of files from my repository.  These are currently
3306          used by other non-SMLNJ backends.
3307    
3308    ----------------------------------------------------------------------
3309    Name: Allen Leung
3310    Date: 2000/03/31 21:15:00 EST
3311    Tag: leunga-20000331-aliasing
3312    Description:
3313    
3314    This update contains a rewritten (and hopefully more correct) module
3315    for extracting aliasing information from CPS.
3316    
3317       To turn on this feature:
3318    
3319            Compiler.Control.CG.memDisambiguate := true
3320    
3321       To pretty print the region information with assembly
3322    
3323           Compiler.Control.MLRISC.getFlag "asm-show-region" := true;
3324    
3325       To control how many levels of aliasing information are printed, use:
3326    
3327           Compiler.Control.MLRISC.getInt "points-to-show-level" := n
3328    
3329       The default of n is 3.
3330    
3331    ----------------------------------------------------------------------
3332    Name: David MacQueen
3333    Date: 2000/03/31 11:15:00 EST
3334    Tag: dbm-20000331-runtime_fix
3335    Description:
3336    
3337    This update contains:
3338    
3339    1. runtime/c-lib/c-libraries.c
3340       includes added in revision 1.2 caused compilation errors on hppa-hpux
3341    
3342    2. fix for bug 1556
3343       system/Basis/Implementation/NJ/internal-signals.sml
3344    
3345    ----------------------------------------------------------------------
3346    Name: Matthias Blume
3347  Date: 2000/03/31 18:00:00 JST  Date: 2000/03/31 18:00:00 JST
3348  Tag: blume_main_v110p26p2_1  Tag: blume_main_v110p26p2_1
3349  Description:  Description:
# Line 67  Line 3397 
3397     (specified in the .cm file at each instance where the tool's class is     (specified in the .cm file at each instance where the tool's class is
3398     used).     used).
3399    
3400     This was done to accomodate the new "make" and "shell" tools which     This was done to accommodate the new "make" and "shell" tools which
3401     facilitate fairly seemless hookup to portions of code managed using     facilitate fairly seamless hookup to portions of code managed using
3402     Makefiles or Shell scripts.     Makefiles or Shell scripts.
3403    
3404     There are no classes "shared" or "private" anymore.  Instead, the     There are no classes "shared" or "private" anymore.  Instead, the
# Line 80  Line 3410 
3410    
3411     All existing tools are described in the CM manual.     All existing tools are described in the CM manual.
3412    
3413     - Slightly better error handling.  (CM now surpresses many followup     - Slightly better error handling.  (CM now suppresses many followup
3414     error messages that tended to be more annoying than helpful.)     error messages that tended to be more annoying than helpful.)
3415    
3416  2. Major changes to the compiler's static environment data structures.  2. Major changes to the compiler's static environment data structures.
# Line 111  Line 3441 
3441       elaborator). There were a lot of changes during my "linkpath" trials       elaborator). There were a lot of changes during my "linkpath" trials
3442       that could have been reverted to their original state but weren't.       that could have been reverted to their original state but weren't.
3443       Please, don't be too harsh on me for messing with this code a bit more       Please, don't be too harsh on me for messing with this code a bit more
3444       than what was strictly necessary...  (I _did_ resist the tempation       than what was strictly necessary...  (I _did_ resist the temptation
3445       of doing any "global reformatting" to avoid an untimely death at       of doing any "global reformatting" to avoid an untimely death at
3446       Dave's hands. :)       Dave's hands. :)
3447    
# Line 214  Line 3544 
3544    
3545       I've changed andl to testl in the floating point test sequence       I've changed andl to testl in the floating point test sequence
3546       whenever appropriate.  The Intel optimization guide states that       whenever appropriate.  The Intel optimization guide states that
3547       testl is perferable to andl.       testl is preferable to andl.
3548    
3549  7. RA (x86 only)  7. RA (x86 only)
3550    
# Line 396  Line 3726 
3726    
3727  1. Tools.registerStdShellCmdTool (from smlnj/cm/tool.cm) takes an  1. Tools.registerStdShellCmdTool (from smlnj/cm/tool.cm) takes an
3728  additional argument called "template" which is an optional string that  additional argument called "template" which is an optional string that
3729  specifiel the layout of the tool command line.  See the CM manual for  specifies the layout of the tool command line.  See the CM manual for
3730  explanation.  explanation.
3731    
3732  2. A special-purpose tool can be "regisitered" by simply dropping the  2. A special-purpose tool can be "registered" by simply dropping the
3733  corresponding <...>-tool.cm (and/or <...>-ext.cm) into the same  corresponding <...>-tool.cm (and/or <...>-ext.cm) into the same
3734  directory where the .cm file lives that uses this tool.  (The  directory where the .cm file lives that uses this tool.  (The
3735  behavior/misfeature until now was to look for the tool description  behavior/misfeature until now was to look for the tool description
# Line 443  Line 3773 
3773  rebuild.  Having sets of the form <base><k>.{bin,boot}.<arch>-unix for  rebuild.  Having sets of the form <base><k>.{bin,boot}.<arch>-unix for
3774  <k>=1,2,... is normally not a good idea when invoking fixpt.  However,  <k>=1,2,... is normally not a good idea when invoking fixpt.  However,
3775  they might be the result of an earlier partial run of fixpt (which  they might be the result of an earlier partial run of fixpt (which
3776  perhaps got accidentially killed).  In this case, fixpt will quickly  perhaps got accidentally killed).  In this case, fixpt will quickly
3777  move through what exists before continuing where it left off earlier,  move through what exists before continuing where it left off earlier,
3778  and, thus, saves a lot of time.  and, thus, saves a lot of time.
3779    
# Line 493  Line 3823 
3823    it from that remote directory.    it from that remote directory.
3824    This should simplify installation further:  For machines that have    This should simplify installation further:  For machines that have
3825    access to the internet, just fetch <version>-config.tgz, unpack it,    access to the internet, just fetch <version>-config.tgz, unpack it,
3826    edit config/targets, and go (run config/install.sh).  The scipt will    edit config/targets, and go (run config/install.sh).  The script will
3827    fetch everything else that it might need all by itself.    fetch everything else that it might need all by itself.
3828    
3829    For CVS users, this mechanism is not relevant for source archives, but    For CVS users, this mechanism is not relevant for source archives, but

Legend:
Removed from v.588  
changed lines
  Added in v.1029

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