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 677, Mon Jun 26 00:56:56 2000 UTC revision 921, Tue Sep 11 23:20:45 2001 UTC
# Line 11  Line 11 
11  Date:  Date:
12  Tag: <post-commit CVS tag>  Tag: <post-commit CVS tag>
13  Description:  Description:
14    
15    ----------------------------------------------------------------------
16    Name: Matthias Blume
17    Date: 2001/08/28 14:03:00 EDT
18    Tag: blume-20010828-ml-lex
19    Description:
20    
21    Fix for bug 1581, received from Neophytos Michael.
22    
23    ----------------------------------------------------------------------
24    Name: Matthias Blume
25    Date: 2001/08/27 11:20:00 EDT
26    Tag: blume-20010827-readme11035
27    Description:
28    
29    Fleshed out the README file for 110.35.
30    
31    ----------------------------------------------------------------------
32    Name: Matthias Blume
33    Date: 2001/08/24 17:10:00 EDT
34    Tag: Release_110_35
35    Description:
36    
37    New version number (110.35).  New bootfiles.
38    
39    ----------------------------------------------------------------------
40    Name: Lal George
41    Date: 2001/08/24 13:47:18 EDT 2001
42    Tag: george-20010824-MLRISC-graphs
43    Description:
44    
45     removed clusters from MLRISC completely and replaced with graphs.
46    
47    ----------------------------------------------------------------------
48    Name: Matthias Blume
49    Date: 2001/08/23 17:50:00 EDT
50    Tag: blume-20010823-toplevel
51    Description:
52    
53    - some reorganization of the code that implements various kinds of
54      environments in the compiler (static, dynamic, symbolic, combined)
55    - re-implemented the EnvRef module so that evalStream works properly
56      (if the stream contains references to "use", "CM.make", etc.)
57    - cleaned up evalloop.sml and interact.sml (but they need more cleaning)
58    
59    ----------------------------------------------------------------------
60    Name: Matthias Blume
61    Date: 2001/08/20 15:50 EDT
62    Tag: blume20010820-slipup
63    Description:
64    
65    I forgot to commit a few files.  Here they are...
66    
67    ----------------------------------------------------------------------
68    Name: Matthias Blume
69    Date: 2001/08/20 15:35:00 EDT
70    Tag: blume-20010820-debugprof
71    Description:
72    
73    !!!! NEW BOOTFILES !!!!
74    
75    This is another round of reorganizing the compiler sources.  This
76    time the main goal was to factor out all the "instrumentation"
77    passes (for profiling and backtracing) into their own library.
78    The difficulty was to do it in such a way that it does not depend
79    on elaborate.cm but only on elabdata.cm.
80    
81    Therefore there have been further changes to both elaborate.cm and
82    elabdata.cm -- more "generic" things have been moved from the former
83    to the latter.  As a result, I was forced to split the assignment
84    of numbers indicating "primtyc"s into two portions: SML-generic and
85    SML/NJ-specific.  Since it would have been awkward to maintain,
86    I bit the bullet and actually _changed_ the mapping between these
87    numbers and primtycs.  The bottom line of this is that you need
88    a new set of bin- and bootfiles.
89    
90    I have built new bootfiles for all architectures, so doing a fresh
91    checkout and config/install.sh should be all you need.
92    
93    The newly created library's name is
94    
95        $smlnj/viscomp/debugprof.cm
96    
97    and its sources live under
98    
99        src/compiler/DebugProf
100    
101    ----------------------------------------------------------------------
102    Name: Matthias Blume
103    Date: 2001/08/15 17:15:00 EDT
104    Tag: blume-20010815-compreorg
105    Description:
106    
107    This is a first cut at reorganizing the CM libraries that make up the
108    core of the compiler.  The idea is to separate out pieces that could
109    be used independently by tools, e.g., the parser, the typechecker, etc.
110    
111    The current status is a step in this direction, but it is not quite
112    satisfactory yet.  Expect more changes in the future.
113    
114    Here is the current (new) organization...
115    
116        What used to be $smlnj/viscomp/core.cm is now divided into
117        six CM libraries:
118    
119             $smlnj/viscomp/basics.cm
120                           /parser.cm
121                           /elabdata.cm
122                           /elaborate.cm
123                           /execute.cm
124                           /core.cm
125    
126        The CM files for these libraries live under src/system/smlnj/viscomp.
127        All these libraries are proxy libraries that contain precisely
128        one CM library component.  Here are the locations of the components
129        (all within the src/compiler tree):
130    
131             Basics/basics.cm
132             Parse/parser.cm
133             ElabData/elabdata.cm
134             Elaborator/elaborate.cm
135             Execution/execute.cm
136             core.cm
137    
138         [This organization is the same that has been used already
139         for a while for the architecture-specific parts of the visible
140         compiler and for the old version of core.cm.]
141    
142         As you will notice, many source files have been moved from their
143         respective original locations to a new home in one of the above
144         subtrees.
145    
146         The division of labor between the new libraries is the following:
147    
148             basics.cm:
149                - Simple, basic definitions that pertain to many (or all) of
150                  the other libraries.
151             parser.cm:
152                - The SML parser, producing output of type Ast.dec.
153                - The type family for Ast is also defined and exported here.
154             elabdata.cm:
155                - The datatypes that describe input and output of the elaborator.
156                  This includes types, absyn, and static environments.
157             elaborator.cm:
158                - The SML/NJ type checker and elaborator.
159                  This maps an Ast.dec (with a given static environment) to
160                  an Absyn.dec (with a new static environment).
161                - This libraries implements certain modules that used to be
162                  structures as functors (to remove dependencies on FLINT).
163             execute.cm:
164                - Everything having to do with executing binary code objects.
165                - Dynamic environments.
166             core.cm:
167                - SML/NJ-specific instantiations of the elaborator and MLRISC.
168                - Top-level modules.
169                - FLINT (this should eventually become its own library)
170    
171    Notes:
172    
173    I am not 100% happy with the way I separated the elaborator (and its
174    data structures) from FLINT.  Two instances of the same problem:
175    
176        1. Data structures contain certain fields that carry FLINT-specific
177           information.  I hacked around this using exn and the property list
178           module from smlnj-lib.  But the fact that there are middle-end
179           specific fields around at all is a bit annoying.
180    
181        2. The elaborator calculates certain FLINT-related information.  I tried
182           to make this as abstract as I could using functorization, but, again,
183           the fact that the elaborator has to perform calculations on behalf
184           of the middle-end at all is not nice.
185    
186        3. Having to used exn and property lists is unfortunate because it
187           weakens type checking.  The other alternative (parameterizing
188           nearly *everything*) is not appealing, though.
189    
190    I removed the "rebinding =" warning hack because due to the new organization
191    it was awkward to maintain it.  As a result, the compiler now issues some of
192    these warnings when compiling init.cmi during bootstrap compilation. On
193    the plus side, you also get a warning when you do, for example:
194       val op = = Int32.+
195    which was not the case up to now.
196    
197    I placed "assign" and "deref" into the _Core structure so that the
198    code that deals with the "lazy" keyword can find them there.  This
199    removes the need for having access to the primitive environment
200    during elaboration.
201    
202    ----------------------------------------------------------------------
203    Name: Matthias Blume
204    Date: 2001/08/13
205    Tag: blume-20010813-closures
206    Description:
207    
208    This fix was sent to us by Zhong Shao.  It is supposed to improve the
209    performance of certain loops by avoiding needless closure allocation.
210    
211    ----------------------------------------------------------------------
212    Name: Lal George
213    Date: 2001/07/31 10:03:23 EDT 2001
214    Tag: george-20010731-x86-fmalloc
215    Description: Fixed bug in x86 calls
216    
217        There was a bug where call instructions would mysteriously
218        vanish. The call instruction had to be one that returned
219        a floating point value.
220    
221    ----------------------------------------------------------------------
222    Name: Lal George
223    Date: 2001/07/19 16:36:29 EDT 2001
224    Tag: george-20010719-simple-cells
225    Description:
226    
227    I have dramatically simplified the interface for CELLS in MLRISC.
228    
229    In summary, the cells interface is broken up into three parts:
230    
231      1. CellsBasis : CELLS_BASIS
232    
233            CellsBasis is a top level structure and common for all
234            architectures.  it contains the definitions of basic datatypes
235            and utility  functions over these types.
236    
237      2. functor Cells() : CELLS
238    
239            Cells generates an interface for CELLS that incorporates the
240            specific resources on the target architecture, such as the
241            presence of special register classes, their number and size,
242            and various useful substructures.
243    
244      3. <ARCH>CELLS
245    
246            e.g. SparcCells: SPARCCELLS
247    
248            <ARCH>CELLS usually contains additional bindings for special
249            registers  on the architecture, such as:
250    
251                    val r0 : cell           (* register zero *)
252                    val y : cell            (* Y register *)
253                    val psr : cell          (* processor status register *)
254                    ...
255    
256            The structure returned by applying the Cells functor is opened
257            in this interface.
258    
259    The main implication of all this is that the datatypes for cells is
260    split between CellsBasis and CELLS -- a fairly simple change for user
261    code.
262    
263    In the old scheme the CELLS interface had a definitional binding of
264    the form:
265    
266            signature CELLS = sig
267    
268               structure CellsBasis = CellsBasis
269    
270               ...
271    
272            end
273    
274    With all the sharing constraints that goes on in MLRISC, this old
275    design  quickly leads to errors such as:
276    
277            "structure definition spec inside of sharing ... "
278    
279    
280    and appears to require an unacceptable amount of sharing and where
281    constraint hackery.
282    
283    I think this error message (the interaction of definitional specs and
284    sharing) requires more explanation on our web page.
285    
286    ----------------------------------------------------------------------
287    Name: Matthias Blume
288    Date: 2001/07/19 15:00:00 EDT
289    Tag: blume-20010719-libreorg
290    Description:
291    
292    This update puts together a fairly extensive but straightforward change
293    to the way the libraries that implement the interactive system are
294    organized:
295    
296       The biggest change is the elimination of structure Compiler.  As a
297       replacement for this structure, there is now a CM library
298       (known as $smlnj/compiler.cm or $smlnj/compiler/current.cm)
299       that exports all the substructures of the original structure Compiler
300       directly.  So instead of saying Compiler.Foo.bar one now simply
301       says Foo.bar.  (The CM libraries actually export a collection of
302       structures that is richer than the collection of substructures of
303       structure Compiler.)
304    
305       To make the transition smooth, there is a separate library called
306       $smlnj/compiler/compiler.cm which puts together and exports the
307       original structure Compiler (or at least something very close to it).
308    
309       There are five members of the original structure Compiler
310       that are not exported directly but which instead became members
311       of a new structure Backend (described by signature BACKEND).  These are:
312       structure Profile (: PROFILE), structure Compile (: COMPILE), structure
313       Interact (: INTERACT), structure Machine (: MACHINE), and val
314       architecture (: string).
315    
316       Structure Compiler.Version has become structure CompilerVersion.
317    
318       Cross-compilers for alpha32, hppa, ppc, sparc, and x86 are provided
319       by $smlnj/compiler/<arch>.cm where <arch> is alpha32, hppa, ppc, sparc,
320       or x86, respectively.
321       Each of these exports the same frontend structures that
322       $smlnj/compiler.cm exports.  But they do not have a structure Backend
323       and instead export some structure <Arch>Backend where <Arch> is Alpha32,
324       Hppa, PPC, Sparc, or X86, respectively.
325    
326       Library $smlnj/compiler/all.cm exports the union of the exports of
327       $smlnj/compiler/<arch>.cm
328    
329       There are no structures <Arch>Compiler anymore, use
330       $smlnj/compiler/<arch>.cm instead.
331    
332       Library host-compiler-0.cm is gone.  Instead, the internal library
333       that instantiates CM is now called cm0.cm.  Selection of the host
334       compiler (backend) is no longer done here but. (Responsibility for it
335       now lies with $smlnj/compiler/current.cm.  This seems to be more
336       logical.)
337    
338       Many individual files have been moved or renamed.  Some files have
339       been split into multiple files, and some "dead" files have been deleted.
340    
341    Aside from these changes to library organization, there are also changes
342    to the way the code itself is organized:
343    
344       Structure Binfile has been re-implemented in such a way that it no
345       longer needs any knowledge of the compiler.  It exclusively deals
346       with the details of binfile layout.  It no longer invokes the
347       compiler (for the purpose of creating new prospective binfile
348       content), and it no longer has any knowledge of how to interpret
349       pickles.
350    
351       Structure Compile (: COMPILE) has been stripped down to the bare
352       essentials of compilation.  It no longer deals with linking/execution.
353       The interface has been cleaned up considerably.
354    
355       Utility routines for dealing with linking and execution have been
356       moved into their own substructures.
357    
358       (The ultimate goal of these changes is to provide a light-weight
359       binfile loader/linker (at least for, e.g., stable libraries) that
360       does not require CM or the compiler to be present.)
361    
362    CM documentation has been updated to reflect the changes to library
363    organization.
364    
365    ----------------------------------------------------------------------
366    Name: Matthias Blume
367    Date: 2001/07/10 17:30:00 EDT
368    Tag: Release_110_34
369    Description:
370    
371    Minor tweak to 110.34 (re-tagged):
372    
373      - README.html file added to CVS repository
374      - runtime compiles properly under FreeBSD 3.X and 4.X
375    
376    ----------------------------------------------------------------------
377    Name: Matthias Blume
378    Date: 2001/07/10 17:30:00 EDT
379    Tag: Release_110_34
380    Description:
381    
382    New version number (110.34). New bootfiles.
383    
384    ----------------------------------------------------------------------
385    Name: Matthias Blume
386    Date: 2001/07/09 16:00:00 EDT
387    Tag: blume-20010709-more-varargs
388    Description:
389    
390    I changed the handling of varargs in ml-nlffigen again:
391    The ellipsis ... will now simply be ignored (with an accompanying warning).
392    
393    The immediate effect is that you can actually call a varargs function
394    from ML -- but you can't actually supply any arguments beyond the ones
395    specified explicitly.  (For example, you can call printf with its format
396    string, but you cannot pass additional arguments.)
397    
398    This behavior is only marginally more useful than the one before, but
399    it has the advantage that a function or, more importantly, a function
400    type never gets dropped on the floor, thus avoiding follow-up problems with
401    other types that refer to the offending one.
402    
403    ----------------------------------------------------------------------
404    Name: Matthias Blume
405    Date: 2001/07/09 11:25:00 EDT
406    Tag: blume-20010709-varargs
407    Description:
408    
409    1. ckit-lib.cm now exports structure Error
410    2. ml-nlffigen reports occurences of "..." (i.e., varargs function types)
411       with a warning accompanied by a source location.  Moreover, it
412       merely skips the offending function or type and proceeds with the
413       rest of its work.u  As a result, one can safely feed C code containing
414       "..." to ml-nlffigen.
415    3. There are some internal improvements to CM, providing slightly
416       more general string substitutions in the tools subsystem.
417    
418    ----------------------------------------------------------------------
419    Name: Matthias Blume
420    Date: 2001/06/27 15:10:00 EDT
421    Tag: blume-20010627-concur
422    Description:
423    
424    Fixed a small bug in CM's handling of parallel compilation.
425    (You could observe the bug by Control-C-interrupting an ordinary
426    CMB.make or CM.stabilize and then attaching some compile servers.
427    The result was that all of a sudden the previously interrupted
428    compilation would continue on its own.  This was because of
429    an over-optimization: CM did not bother to clean out certain queues
430    when no servers were attached "anyway", resulting in the contents
431    of these queues to grab control when new servers did get attached.)
432    
433    There is also another minor update to the CM manual.
434    
435    ----------------------------------------------------------------------
436    Name: Matthias Blume
437    Date: 2001/06/26 16:15:00 EDT
438    Tag: blume-20010626-cmdoc
439    Description:
440    
441    Minor typo fixed in CM manual (syntax diagram for libraries).
442    
443    ----------------------------------------------------------------------
444    Name: Matthias Blume
445    Date: 2001/06/25 22:55:00 EDT
446    Tag: blume-20010625-x86pc
447    Description:
448    
449    Fixed a nasty bug in the X86 assembly code that caused signal
450    handlers to fail (crash) randomly.
451    
452    ----------------------------------------------------------------------
453    Name: Matthias Blume
454    Date: 2001/06/25 12:05:00 EDT
455    Tag: blume-20010625-nlffigen
456    Description:
457    
458    This update fixes a number of minor bugs in ml-nlffigen as reported by
459    Nick Carter <nbc@andrew.cmu.edu>.
460    
461      1. Silly but ok typedefs of the form "typedef void myvoid;" are now accepted.
462      2. Default names for generated files are now derived from the name of
463         the C file *without its directory*.  In particular, this causes generated
464         files to be placed locally even if the C file is in some system directory.
465      3. Default names for generated signatures and structures are also derived
466         from the C file name without its directory.  This avoids silly things
467         like "structure GL/GL".
468         (Other silly names are still possible because ml-nlffigen does not do
469          a thorough check of whether generated names are legal ML identifiers.
470          When in doubt, use command line arguments to force particular names.)
471    
472    ----------------------------------------------------------------------
473    Name: Matthias Blume
474    Date: 2001/06/21 12:25:00 EDT
475    Tag: blume-20010621-eXene
476    Description:
477    
478    eXene now compiles and (sort of) works again.
479    
480    The library name (for version > 110.33) is $/eXene.cm.
481    
482    I also added an new example in src/eXene/examples/nbody.  See the
483    README file there for details.
484    
485    ----------------------------------------------------------------------
486    Name: Matthias Blume
487    Date: 2001/06/20 16:40:00 EDT
488    Tag: blume-20010620-cml
489    Description:
490    
491    CML now compiles and works again.
492    
493    Libraries (for version > 110.33):
494    
495      $cml/cml.cm            Main CML library.
496      $cml/basis.cm          CML's version of $/basis.cm.
497      $cml/cml-internal.cm   Internal helper library.
498      $cml/core-cml.cm       Internal helper library.
499      $cml-lib/trace-cml.cm  Tracing facility.
500      $cml-lib/smlnj-lib.cm  CML's version of $/smlnj-lib.cm
501    
502    The installer (config/install.sh) has been taught how to properly
503    install this stuff.
504    
505    ----------------------------------------------------------------------
506    Name: Matthias Blume
507    Date: 2001/06/19 17:55:00 EDT
508    Tag: blume-20010619-instantiate
509    Description:
510    
511    This un-breaks the fix for bug 1432.
512    (The bug was originally fixed in 110.9 but I broke it again some
513    time after that.)
514    
515    ----------------------------------------------------------------------
516    Name: Matthias Blume
517    Date: 2001/06/19 17:25:00 EDT
518    Tag: blume-20010619-signals
519    Description:
520    
521    This should (hopefully) fix the long-standing signal handling bug.
522    (The runtime system was constructing a continuation record with an
523    incorrect descriptor which would cause the GC to drop data on the floor...)
524    
525    ----------------------------------------------------------------------
526    Name: Matthias Blume
527    Date: 2001/06/15 15:05:00 EDT
528    Tag: blume-20010615-moresparc
529    Description:
530    
531    Here is a short late-hour update related to Sparc c-calls:
532    
533     -- made handling of double-word arguments a bit smarter
534    
535     -- instruction selection phase tries to collapse certain clumsily
536        constructed ML-Trees; typical example:
537    
538            ADD(ty,ADD(_,e,LI d1),LI d2)  ->  ADD(ty,e,LI(d1+d2))
539    
540        This currently has no further impact on SML/NJ since mlriscGen does
541        not seem to generate such patterns in the first place, and c-calls
542        (which did generate them in the beginning) has meanwhile been fixed
543        so as to avoid them as well.
544    
545    ----------------------------------------------------------------------
546    Name: Matthias Blume
547    Date: 2001/06/15 15:05:00 EDT
548    Tag: blume-20010615-sparc
549    Description:
550    
551    The purpose of this update is to provide an implementation of NLFFI
552    on Sparc machines.
553    
554    Here are the changes in detail:
555    
556       * src/MLRISC/sparc/c-calls/sparc-c-calls.sml is a new file containing
557       the Sparc implementation of the c-calls API.
558       * The Sparc backend of SML/NJ has been modified to uniformely use %fp
559       for accessing the ML frame.  Thus, we have a real frame pointer and
560       can freely modify %sp without need for an omit-frame-ptr phase.
561       The vfp logic in src/compiler/CodeGen/* has been changed to accomodate
562       this case.
563       * ml-nlffigen has been taught to produce code for different architectures
564       and calling conventions.
565       * In a way similar to what was done in the x86 case, the Sparc
566       backend uses its own specific extension to mltree.  (For example,
567       it needs to be able to generate UNIMP instructions which are part
568       of the calling convention.)
569       * ml-nlffi-lib was reorganized to make it more modular (in particular,
570       to make it easier to plug in new machine- and os-dependent parts).
571    
572    There are some other fairly unrelated bug fixes and cleanups as well:
573    
574       * I further hacked the .cm files for MLRISC tools (like MDLGen) so
575       that they properly share their libraries with existing SML/NJ libraries.
576       * I fixed a minor cosmetic bug in CM, supressing certain spurious
577       follow-up error messages.
578       * Updates to CM/CMB documentation.
579    
580    TODO items:
581    
582       * MLRISC should use a different register as its asmTemp on the Sparc.
583         (The current %o2 is a really bad choice because it is part of the
584         calling conventions, so things might interfere in unexpected ways.)
585    
586    ----------------------------------------------------------------------
587    Name: Matthias Blume
588    Date: 2001/06/07
589    Tag: blume-20010607-calls
590    Description:
591    
592    A number of internal changes related to C calls and calling conventions:
593    
594    1. ML-Tree CALL statements now carry a "pops" field.  It indicates the
595       number of bytes popped implicitly (by the callee).  In most cases
596       this field is 0 but on x86/win32 it is some non-zero value.  This
597       is information provided for the benefit of the "omit-frameptr" pass.
598    2. The CALL instruction on the x86 carries a similar "pops" field.
599       The instruction selection phase copies its value from the ML-Tree
600       CALL statement.
601    3. On all other architectures, the instruction selection phase checks
602       whether "pops=0" and complains if not.
603    4. The c-calls implementation for x86 now accepts two calling conventions:
604       "ccall" and "stdcall".  When "ccall" is selected, the caller cleans
605       up after the call and pops is set to 0.  For "stdcall", the caller
606       does nothing, leaving the cleanup to the callee; pops is set to
607       the number of bytes that were pushed onto the stack.
608    5. The cproto decoder (compiler/Semant/types/cproto.sml) now can
609       distinguish between "ccall" and "stdcall".
610    6. The UNIMP instruction has been added to the supported Sparc instruction
611       set. (This is needed for implementing the official C calling convention
612       on this architecture.)
613    7. I fixed some of the .cm files under src/MLRISC/Tools to make them
614       work with the latest CM.
615    
616    ----------------------------------------------------------------------
617    Name: Matthias Blume
618    Date: 2001/06/05 15:10:00 EDT
619    Tag: blume-20010605-cm-index
620    Description:
621    
622    0. The "lambdasplit" parameter for class "sml" in CM has been documented.
623    
624    1. CM can now generate "index files".  These are human-readable files
625       that list on a per-.cm-file basis each toplevel symbol defined or
626       imported.  The location of the index file for
627       <p>/<d>.cm is <p>/CM/INDEX/<d>.cm.
628       To enable index-file generation, set CM.Control.generate_index to true
629       or export an environment-symbol: export CM_GENERATE_INDEX=true.
630    
631       The CM manual has been updated accordingly.
632    
633    2. I made some slight modifications to the c-calls API in MLRISC.
634    
635         a) There is now a callback to support saving/restoring of
636            dedicated but caller-save registers around the actual call
637            instruction.
638         b) One can optionally specify a comment-annotation for the
639            call instruction.
640    
641    3. SML/NJ (mlriscGen.sml) uses this new API for the rawccall primop.
642       (For example, the comment annotation shows the C prototype of
643        the function being called.)
644    
645    ----------------------------------------------------------------------
646    Name: Matthias Blume
647    Date: 2001/06/01 13:30:00 EDT
648    Tag: blume-20010601-nlffi-cleanup
649    Description:
650    
651    This is mostly a cleanup of MLFFI stuff:
652    
653       - some signature files have been put into a more exposed place
654       - the ugly 'f type parameter is gone (simplifies types tremendously!)
655       - ml-nlffigen changed accordingly
656       - tutorial updated
657    
658    Other changes:
659    
660       - author's affiliation in CM manual(s) updated
661       - some more recognized keywords added to Allen's sml.sty
662    
663    ----------------------------------------------------------------------
664    Name: Matthias Blume
665    Date: 2001/05/25 15:30:00 EDT
666    Tag: blume-20010525-iptr
667    Description:
668    
669      - put the official 110.33-README (as it appears on the ftp server) under
670        CVS
671      - fixed a small bug related to incomplete pointer types in
672        ml-nlffigen
673      - small cosmetic change to the ml-nlffi-lib's "arr" type constructor
674        (it does not need the 'f type parameter)
675    
676    ----------------------------------------------------------------------
677    Name: Matthias Blume
678    Date: 2001/05/23 14:30:00 EDT
679    Tag: Release_110_33
680    Description:
681    
682    New version number (110.33).  New bootfiles.
683    
684    ----------------------------------------------------------------------
685    Name: Matthias Blume
686    Date: 2001/05/22 18:06:00 EDT
687    Tag: blume-20010522-targets
688    Description:
689    
690    Made install.sh use file config/targets.customized if it exists, falling
691    back to config/targets if it doesn't.  This way one can have a customized
692    version of the targets file without touching the "real thing", thus
693    eliminating the constant fear of accidentally checking something bogus
694    back into the CVS repository...  (File config/targets.customized must
695    not be added to the repository!)
696    
697    ----------------------------------------------------------------------
698    Name: Matthias Blume
699    Date: 2001/05/22 16:30:00 EDT
700    Tag: blume-20010522-minitut
701    Description:
702    
703    1. Bug fix in ml-nlffigen; now (hopefully) correctly handling
704       struct returns.
705    2. Added src/ml-nlffi-lib/Doc/mini-tutorial.txt.  This is some very
706       incomplete, preliminary documentation for NLFFI.
707    
708    ----------------------------------------------------------------------
709    Name: Matthias Blume
710    Date: 2001/05/14 11:30:00 EDT
711    Tag: blume-20010514-script
712    Description:
713    
714    Some bugs in install script fixed.
715    
716    In addition to that I also made a slight change to the NLFFI API:
717    Functors generated by ml-nlffigen now take the dynamic library as a
718    straight functor argument, not as a suspended one.  (The original
719    functor code used to force the suspension right away anyway, so there
720    was nothing gained by this complication of the interface.)
721    
722    ----------------------------------------------------------------------
723    Name: Matthias Blume
724    Date: 2001/05/11 14:35:00 EDT
725    Tag: blume-20010511-ml-nlffi
726    Description:
727    
728    I finally took the plunge and added my new FFI code to the main
729    repository.  For x86-linux it is now ready for prime-time.
730    
731    There are two new subdirectories of "src":
732    
733      - ml-nlffi-lib:
734           The utility library for programs using the FFI interface.
735           Here is the implementation of $/c.cm and its associated low-level
736           partners $/c-int.cm and $/memory.cm.
737      - ml-nlffigen:
738           A stand-alone program for generating ML glue code from C source
739           code.
740    
741    Building ml-nlffigen requires $/ckit-lib.cm.
742    
743    The config/install.sh script has been updates to do the Right Thing
744    (hopefully).
745    
746    Notice that the source tree for the C-Kit will not be put under "src"
747    but directly under the installation root directory.  (This is the
748    structure that currently exists on the CVS server when you check out
749    module "sml".)  Fortunately, config/install.sh knows about this oddity.
750    
751    Bugs: No documentation yet.
752    
753    ----------------------------------------------------------------------
754    Name: Matthias Blume
755    Date: 2001/05/09 16:35:00 EDT
756    Tag: blume-20010509-cpscontract
757    Description:
758    
759    Fixed a bug in the accounting code in cpsopt/contract.sml.  (The
760    wrapper/unwrapper elimination did not decrement usage counts and some
761    dead variables got overlooked by the dead-up logic.)
762    
763    ----------------------------------------------------------------------
764    Name: Lal George
765    Date: 2001/05/08  17:26:09 EDT
766    Tag: george-20010508-omit-frameptr
767    Description:
768    
769    Changes to implement the omit-frame-pointer optimization to support
770    raw C calls. For now, there is only support on the Intel x86, but
771    other architectures will follow as more experience is gained with this.
772    
773    
774    ----------------------------------------------------------------------
775    Name: Matthias Blume
776    Date: 2001/05/07 14:40:00 EDT
777    Tag: blume-20010507-proxies
778    Description:
779    
780    I made into "proxy libraries" all libraries that qualify for such a
781    change.  (A qualifying library is a library that has another library or
782    groups as its sole member and repeats that member's export list
783    verbatim.  A proxy library avoids this repetition by omitting its export
784    list, effectively inheriting the list that its (only) member exports.
785    See the CM manual for more explanation.)
786    The main effect is that explicit export lists for these libraries
787    do not have to be kepts in sync, making maintenance a bit easier.
788    
789    I also added copyright notices to many .cm-files.
790    
791    Last but not least, I made a new set of bootfiles.
792    
793    ----------------------------------------------------------------------
794    Name: Matthias Blume
795    Date: 2001/05/04 17:00:00 EDT
796    Tag: blume-20010504-cm-lsplit
797    Description:
798    
799    0. John merged pending changes to $/smlnj-lib.cm
800    
801    1. Allen's previous change accidentally backed out of one of Lal's
802       earlier changes.  I undid this mistake (re-introducing Lal's change).
803    
804    2. I used the new topOrder' function from graph-scc.sml (from $/smlnj-lib.cm)
805       within the compiler where applicable.  There is some code simplification
806       because of that.
807    
808    3. The "split" phase (in FLINT) is now part of the default list of phases.
809       Compiler.Control.LambdaSplitting.* can be used to globally control the
810       lambda-splitting (cross-module-inlining) engine.  In addition to that,
811       it can now also be controlled on a per-source basis: CM has been taught
812       a new tool parameter applicable to ML source files.
813    
814       - To turn lambda-splitting off completely:
815            local open Compiler.Control.LambdaSplitting in
816                val _ = set Off
817            end
818       - To make "no lambda-splitting" the global default (but allow per-source
819         overriding); this is the initial setting:
820            local open Compiler.Control.LambdaSplitting in
821                val _ = set (Default NONE)
822            end
823       - To make "lambda-splitting with aggressiveness a" the global default
824         (and allow per-source overriding):
825            local open Compiler.Control.LambdaSplitting in
826                val _ = set (Default (SOME a))
827            end
828    
829       - To turn lambda-splitting off for a given ML souce file (say: a.sml)
830         write (in the respective .cm-file):
831            a.sml (lambdasplitting:off)
832       - To turn lambda-splitting for a.sml on with minimal aggressiveness:
833            a.sml (lambdasplitting:on)
834       - To turn lambda-splitting for a.sml on with aggressiveness <a> (where
835         <a> is a decimal non-negative integer):
836            a.sml (lambdasplitting:<a>)
837       - To turn lambda-splitting for a.sml on with maximal aggressiveness:
838            a.sml (lambdasplitting:infinity)
839       - To use the global default for a.sml:
840            a.sml (lambdasplitting:default)
841         or simply
842            a.sml
843    
844    ----------------------------------------------------------------------
845    Name: Allen Leung
846    Date: 2001/05/04 01:57:00 EDT
847    Tag: leunga-20010504-sync
848    Description:
849    
850      MLRISC features.
851    
852      1. Fix to CMPXCHG instructions.
853      2. Changed RA interface to allow annotations in callbacks.
854      3. Added a new method to the stream interface to allow annotations updates.
855    
856    ----------------------------------------------------------------------
857    Name: Matthias Blume
858    Date: 2001/05/01 11:45:00 EDT
859    Tag: blume-20010501-pcedittmp
860    Description:
861    
862    Changed install.sh to use the current working directory instead of
863    /usr/tmp for a temporary file (pcedittmp).  The previous choice
864    of /usr/tmp caused trouble with MacOS X because of file premission
865    problems.
866    
867    ----------------------------------------------------------------------
868    Name: Matthias Blume
869    Date: 2001/04/20 11:10:00 EDT
870    Tag: blume-20010420-inMLflag
871    Description:
872    
873     - added vp_limitPtrMask to vproc-state.h
874       (for use by the raw-C-calls mechanism to implement proper interrupt
875        handling)
876     - made the ML compiler aware of various data-structure offsets so it
877       can generate code for accessing the vp_inML flag and vp_limitPtrMask
878     - tweaked mlriscGen.sml to have it emit interrupt-handling code for
879       raw C-calls
880    
881    ----------------------------------------------------------------------
882    Name: Lal George
883    Date: 2001/04/20 09:15:28 EDT
884    Tag: george-20010420-macosX
885    Description:
886    
887     - Changes to port to Mac OS X; Darwin.
888    
889     - In the process I found that sqrt was broken on the PPC, because the
890       fsqrt instruction is not implemented.
891    
892    ----------------------------------------------------------------------
893    Name: Matthias Blume
894    Date: 2001/04/18 12:45:00 EDT
895    Tag: blume-20010418-ccalls
896    Description:
897    
898     - fixed two off-by-4 errors in the x86-specific c-calls implementation
899       (this bug prevented structure arguments containing pointers from being
900        passed correctly)
901     - changed the raw-C-call code in mlriscGen.sml in such a way that
902       structure arguments are represented as a pointer to the beginning
903       of the structure (instead of having a series of synthesized arguments,
904       one for each structure member)
905    
906     - made makeml script's verbosity level configurable via environment
907       variable (MAKEML_VERBOSITY)
908    
909     - eliminated placeholder implementations for f32l, w16s, i16s, and f32s
910       in rawmem-x86.sml; we are now using the real thing
911    
912    ----------------------------------------------------------------------
913    Name: Matthias Blume
914    Date: 2001/03/22 16:25:00 EST
915    Tag: blume-20010322-bootfiles
916    Description:
917    
918    Created a new set of bootfiles (for your automatic installation convenience).
919    
920    ----------------------------------------------------------------------
921    Name: Matthias Blume
922    Date: 2001/03/22 15:10:00 EST
923    Tag: blume-20010322-rawmem-parcm
924    Description:
925    
926    1. All "raw memory access" primitives for the new FFI are implemented now
927       (at least on the x86).
928    2. Some further cleanup of CM's parallel make mechanism.
929    
930    ----------------------------------------------------------------------
931    Name: Matthias Blume
932    Date: 2001/03/19 17:53:00 EST
933    Tag: blume-20010319-parallel
934    Description:
935    
936    Parallel make (using compile servers) now works again.
937    
938    To this end, CM.stabilize and CMB.make have been modified to work in
939    two passes when compile servers are attached:
940       1. Compile everything, do not perform stabilization; this pass
941          uses compile servers
942       2. Stabilize everything; this pass does not use compile servers
943    If there are no compile servers, the two passes are combined into one
944    (as before).  Splitting the passes increases the inherent parallelism
945    in the dependency graph because the entire graph including all
946    libraries is available at the same time.  This, in turn, improves
947    server utilization.  The downside is that the master process will
948    have to do some extra work after compilation is done (because for
949    technical reasons it must re-read all the binfiles during stabilization).
950    
951    ----------------------------------------------------------------------
952    Name: Matthias Blume
953    Date: 2001/03/16 12:22:00 EST
954    Tag: blume-20010316-bootfiles
955    Description:
956    
957    Created a new set of bootfiles (for your automatic installation convenience).
958    
959    ----------------------------------------------------------------------
960    Name: Matthias Blume
961    Date: 2001/03/16 11:00:00 EST
962    Tag: blume-20010316-MLTREE-fixup
963    Description:
964    
965    This is a minor fixup for an (untagged) earlier commit by Allen.
966    (A file was missing).
967    
968    ----------------------------------------------------------------------
969    Name: Allen Leung
970    Date: Mon Mar  5 18:54:57 EST 2001
971    Tag: leunga-20010305-cut-support
972    
973    1. New support for alternative control-flow in MLTREE.
974       Currently we support
975    
976          FLOW_TO(CALL ...., [k1,...,kn])
977    
978       This is needed for 'cuts to' in C-- and try/handle-like constructs
979       in Moby
980    
981       New assembler flag "asm-show-cutsto" to turn on control-flow debugging.
982    
983    2. Register Allocator
984    
985       Changes in interface [from Fermin, John]
986    
987    3. Alpha 8-bit SLL support [Fermin]
988    
989    4. All architectures
990    
991       A new module (ClusterExpandCopies) for expanding parallel copies.
992    
993    ----------------------------------------------------------------------
994    Name: Allen Leung
995    Date: 2001/02/27 23:07:00 EST
996    Tag: leunga-20010227-minor-stuff
997    
998    1. Alpha bug fix for CMOVNE
999    2. Handle mltree COND(..,FCMP ...,...)
1000    3. Bug fix in simplifier
1001    
1002    ----------------------------------------------------------------------
1003    Name: Matthias Blume
1004    Date: 2001/01/30 17:50:00 EST
1005    Tag: blume-20010130-sync
1006    Description:
1007    
1008    This is just a minor update to sync my devel branch with the main brach.
1009    The only visible change is the addition of some README files.
1010    
1011    ----------------------------------------------------------------------
1012    Name: Matthias Blume
1013    Date: 2001/01/12 23:30:00 JST
1014    Tag: blume-20010112-bootfiles
1015    Description:
1016    
1017    Made a new set of bootfiles that goes with the current state of the
1018    repository.
1019    
1020    ----------------------------------------------------------------------
1021    Name: Matthias Blume
1022    Date: 2001/01/12 21:20:00 JST
1023    Tag: blume-20010112-sync
1024    Description:
1025    
1026    I am just flushing out some minor changes that had accumulated in
1027    my private branch in order to sync with the main tree.  (This is
1028    mainly because I had CVS trouble when trying to merge _into_ my
1029    private branch.)
1030    
1031    Most people should be completely unaffected by this.
1032    
1033    ----------------------------------------------------------------------
1034    Name: Allen Leung
1035    Date: Thu Jan 11 21:03:00 EST 2001
1036    Tag: leunga-20010111-labexp=mltree
1037    Description:
1038    
1039    1.  Removed the type LabelExp and replace it by MLTree.
1040    2.  Rewritten mltree-simplify with the pattern matcher tool.
1041    3.  There were some bugs in alpha code generator which would break
1042        64-bit code generation.
1043    4.  Redo the tools to generate code with the
1044    5.  The CM files in MLRISC (and in src/system/smlnj/MLRISC)
1045        are now generated by perl scripts.
1046    
1047    ----------------------------------------------------------------------
1048    Name: Matthias Blume
1049    Date: 2001/01/10 21:55:00 JST
1050    Tag: blume-20010110-rcc
1051    Description:
1052    
1053    The RCC stuff now seems to work (but only on the x86).
1054    This required hacking of the c-calls interface (and -implementation) in
1055    MLRISC.
1056    
1057    Normal compiler users should be unaffected.
1058    
1059    ----------------------------------------------------------------------
1060    Name: Matthias Blume
1061    Date: 2001/01/09 01:20:00 JST
1062    Tag: blume-20010109-rcc
1063    Description:
1064    
1065    This is a fairly big patch, flushing out a large number of pending
1066    changes that I made to my development copy over the last couple of days.
1067    
1068    Of practical relevance at this moment is a workaround for a pickling
1069    bug that Allen ran into the other day.  The cause of the bug itself is
1070    still unknown and it might be hard to fix it properly, but the
1071    workaround has some merits of its own (namely somewhat reducing pickling
1072    overhead for certain libraries).  Therefore, I think this solution should
1073    be satisfactory at this time.
1074    
1075    The rest of the changes (i.e., the vast majority) has to do with my
1076    ongoing efforts of providing direct support for C function calls from
1077    ML.  At the moment there is a new primop "RAW_CCALL", typing magic
1078    in types/cproto.sml (invoked from FLINT/trans/translate.sml), a new
1079    case in the FLINT CPS datatype (RCC), changes to cps/convert.sml to
1080    translate uses of RAW_CCALL into RCC, and changes to mlriscGen.sml to
1081    handle RCC.
1082    
1083    The last part (the changes to mlriscGen.sml) are still known to be
1084    wrong on the x86 and not implemented on all other architectures.  But
1085    the infrastructure is in place. I had to change a few functor
1086    signatures in the backend to be able to route the CCalls interface
1087    from MLRISC there, and I had to specialize the mltree type (on the
1088    x86) to include the necessary extensions. (The extensions themselves
1089    were already there and redy to go in MLRISC/x86).
1090    
1091    Everything should be very happy as soon as someone helps me with
1092    mlriscGen.sml...
1093    
1094    In any case, nothing of this should matter to anyone as long as the
1095    new primop is not being used (which is going to be the case unless you
1096    find it where I hid it :). The rest of the compiler is completely
1097    unaffected.
1098    
1099    ----------------------------------------------------------------------
1100    Name: Matthias Blume
1101    Date: 2001/01/05 00:30:00 JST
1102    Tag: blume-20010105-primops
1103    Description:
1104    
1105    Added some experimental support for work that I am doing right now.
1106    These changes mostly concern added primops, but there is also a new
1107    experimental C library in the runtime system (but currently not enabled
1108    anywhere except on Linux/X86).
1109    
1110    In the course of adding primops (and playing with them), I discovered that
1111    Zhong's INL_PRIM hack (no type info for certain primops) was, in fact, badly
1112    broken.  (Zhong was very right he labeled this stuff as "major gross hack".)
1113    To recover, I made type information in INL_PRIM mandatory and changed
1114    prim.sml as well as built-in.sml accordingly.  The InLine structure now
1115    has complete, correct type information (i.e., no bottom types).
1116    
1117    Since all these changes mean that we need new binfiles, I also bumped the
1118    version number to 110.32.1.
1119    
1120    ----------------------------------------------------------------------
1121    Name: Matthias Blume
1122    Date: 2000/12/30 22:10:00 JST
1123    Tag: blume-20001230-various
1124    Description:
1125    
1126    Added proxy libraries for MLRISC and let MLRISC libraries refer
1127    to each other using path anchors.  (See CM manual for explanation.)
1128    
1129    Updated CM documentation.
1130    
1131    Fixed some bugs in CM.
1132    
1133    Implemented "proxy" libraries (= syntactic sugar for CM).
1134    
1135    Added "-quiet" option to makeml and changed runtime system accordingly.
1136    
1137    Added cleanup handler for exportML to reset timers and compiler stats.
1138    
1139    ----------------------------------------------------------------------
1140    Name: Lal George
1141    Date: 2000/12/22 22:22:58 EST 2000
1142    Tag: Release_110_32
1143    Description:
1144    
1145            Infinite precision used throughout MLRISC.
1146            see MLRISC/mltree/machine-int.sig
1147    
1148    ----------------------------------------------------------------------
1149    Name: Matthias Blume
1150    Date: 2000/12/22 23:16:00 JST
1151    Tag: blume-20001222-warn
1152    Description:
1153    
1154    Corrected wording and formatting of some CM warning message which I
1155    broke in my previous patch.
1156    
1157    ----------------------------------------------------------------------
1158    Name: Matthias Blume
1159    Date: 2000/12/22 21:20:00 JST
1160    Tag: blume-20001222-anchorenv
1161    Description:
1162    
1163    Fixed CM's handling of anchor environments in connection with CMB.make.
1164    
1165    ----------------------------------------------------------------------
1166    Name: Matthias Blume
1167    Date: 2000/12/22 13:15:00 JST
1168    Tag: blume-20001222-cleanup
1169    Description:
1170    
1171    Removed src/cm/ffi which does not (and did not) belong here.
1172    
1173    ----------------------------------------------------------------------
1174    Name: Matthias Blume
1175    Date: 2000/12/21 23:55:00 JST
1176    Tag: blume-20001221-exn
1177    Description:
1178    
1179    Probably most important: CM no longer silently swallows all exceptions
1180    in the compiler.
1181    Plus: some other minor CM changes.  For example, CM now reports some
1182    sizes for generated binfiles (code, data, envpickle, lambdapickle).
1183    
1184    ----------------------------------------------------------------------
1185    Name: Matthias Blume
1186    Date: 2000/12/15 00:01:05 JST
1187    Tag: blume-20001215-dirtool
1188    Description:
1189    
1190    - "dir" tool added.
1191    - improvements and cleanup to Tools structure
1192    - documentation updates
1193    
1194    ----------------------------------------------------------------------
1195    Name: Allen Leung
1196    Date: Thu Dec 14 03:45:24 EST 2000
1197    Description:
1198    Tag:  leunga-20001214-int-inf
1199    Description:
1200    
1201       In IntInf, added these standard functions, which are missing from our
1202    implementation:
1203    
1204        andb : int * int -> int
1205        xorb : int * int -> int
1206        orb  : int * int -> int
1207        notb : int -> int
1208         <<   : int * word -> int
1209        ~>>  : int * word -> int
1210    
1211       Not tested, I hope they are correct.
1212    
1213    ----------------------------------------------------------------------
1214    Name: Allen Leung
1215    Date: Fri Dec  8 19:23:26 EST 2000
1216    Description:
1217    Tag:  leunga-20001208-nowhere
1218    Description:
1219    
1220      Slight improvements to the 'nowhere' tool to handle OR-patterns,
1221    to generate better error messages etc.  Plus a brief manual.
1222    
1223    ----------------------------------------------------------------------
1224    Name: Lal George
1225    Date: 2000/12/08 09:54:02 EST 2000
1226    Tag: Release_110_31
1227    Description:
1228    
1229    - Version 110.31
1230    ----------------------------------------------------------------------
1231    Name: Allen Leung
1232    Date: Thu Dec  7 22:01:04 EST 2000
1233    Tag:  leunga-20001207-cell-monster-hack
1234    Description:
1235    
1236    Major MLRISC internal changes.  Affect all clients.
1237    Summary:
1238    
1239    1.  Type CELLS.cell = int is now replaced by a datatype.
1240        As a result, the old regmap is now gone.  Almost all interfaces
1241        in MLRISC change as a consequence.
1242    
1243    2.  A new brand version of machine description tool (v3.0) that generates
1244        modules expecting the new interface.  The old version is removed.
1245    
1246    3.  The RA interface has been further abstracted into two new functors.
1247        RISC_RA and X86RA.  These functors have much simpler interfaces.
1248        [See also directory MLRISC/demo.]
1249    
1250    4.  Some other new source->source code generation tools are available:
1251    
1252        a. MLRISC/Tools/RewriteGen -- generate rewriters from rules.
1253        b. MLRISC/Tools/WhereGen -- expands conditional pattern matching rules.
1254           I use this tool to generate the peephole optimizers---with the new
1255           cell type changes, peephole rules are becoming difficult to write
1256           without conditional pattern matching.
1257    
1258    5.  More Intmap -> IntHashTable change.  Previous changes by Matthias didn't
1259        cover the entire MLRISC source tree so many things broke.
1260    
1261    6.  CM files have been moved to the subdirectory MLRISC/cm.
1262        They are moved because there are a lot of them and they clutter up the
1263        root dir.
1264    
1265    7.  More detailed documentation to come...
1266    
1267        NOTE: To rebuild from 110.30 (ftp distribution), you'll have to do
1268        a makeml -rebuild first.  This is because of other other
1269        changes that Matthias has made (see below).
1270    
1271    
1272    ----------------------------------------------------------------------
1273    Name: Matthias Blume
1274    Date: 2000/11/30 23:12:00 JST
1275    Tag: blume-20001130-filereorg
1276    Description:
1277    
1278    Some manual updates and some file reorganizations in CM.
1279    
1280    ----------------------------------------------------------------------
1281    Name: Matthias Blume
1282    Date: 2000/11/24 17:45:00 JST
1283    Tag: blume-20001124-link
1284    Description:
1285    
1286    Drastically improved link traversal code for the case that the dynamic
1287    value was already loaded at bootstrap time.  As a result, CM and CMB
1288    now both load blazingly fast -- even on a very slow machine.  Also,
1289    memory consumption has been further reduced by this.
1290    
1291    Warning: The format of the PIDMAP file has changed.  THerefore, to
1292    bootstrap you have to do this:
1293    
1294    1. Run CMB.make
1295    2. Make a symbolic link for the boot directory:
1296         ln -s sml.boot.ARCH-OS xxx
1297    3. "Rebuild" the boot directory:
1298         ./makeml -boot xxx -rebuild sml ; rm xxx
1299    4. Boot normally:
1300          ./makeml
1301    
1302    ----------------------------------------------------------------------
1303    Name: Matthias Blume
1304    Date: 2000/11/21 21:20:00 JST
1305    Tag: blume-20001121-tools
1306    Description:
1307    
1308    Continued hacking on autoloading problem -- with success this time.
1309    Also changed tool-plugin mechanism.  See new CM manual.
1310    
1311    ----------------------------------------------------------------------
1312    Name: Matthias Blume
1313    Date: 2000/11/19 14:30:00 JST
1314    Tag:  blume-20001119-autoload
1315    Description:
1316    
1317    Some hacking to make autoloading faster.  Success for CMB, no success
1318    so far for CM.  There is a reduced structure CM' that autoloads faster.
1319    (This is a temporary, non-documented hack to be eliminated again when
1320    the general problem is solved.)
1321    
1322    ----------------------------------------------------------------------
1323    Name: Matthias Blume
1324    Date: 2000/11/17 14:10:00 JST
1325    Tag: blume-20001117-pickle-lib
1326    Description:
1327    
1328    1. Eliminated comp-lib.cm
1329    2. Made pickle-lib.cm
1330    3. Eliminated all uses of intset.sml (from comp-lib.cm)
1331    4. Replaced all uses of intmap.{sig,sml} (from comp-lib.cm) with
1332       equivalent constructs from smlnj-lib.cm (INtHashTable).
1333    5. Point 4. also goes for those uses of intmap.* in MLRISC.
1334       Duplicated intmap modules thrown out.
1335    6. Hunted down all duplicated SCC code and replaced it with
1336       equivalent stuff (GraphSCCFn from smlnj-lib.cm).
1337    7. Rewrote Feedback module.
1338    8. Moved sortedlist.sml into viscomp-lib.cm.  Eventually it
1339       should be thrown out and equivalent modules from smlnj-lib.cm
1340       should be used (IntRedBlackSet, IntListSet, ...).
1341    
1342    Confirmed that compiler compiles to fixpoint.
1343    
1344    ----------------------------------------------------------------------
1345    Name: Allen Leung
1346    Date: 2000/11/10 18:00:00
1347    Tag: leunga-20001110-new-x86-fp
1348    
1349    A new x86 floating point code generator has been added.
1350    By default this is turned off.  To turn this on, do:
1351    
1352        CM.autoload "$smlnj/compiler.cm";
1353        Compiler.Control.MLRISC.getFlag "x86-fast-fp" := true;
1354    
1355    Changes:
1356    
1357    1.  Changed FTAN to FPTAN so that the assembly output is correct.
1358    2.  Changed the extension callback for FTANGENT to generate:
1359    
1360              fptan
1361              fstp  %st(0)
1362        instead of
1363              fptan
1364              fstpl ftempmem
1365    
1366    3.  Numerous assembly fixes for x86.
1367    
1368    5.  Cleaned up the machine code output module x86/x86MC.sml and added
1369        support for a whole bunch of instructions and addressing modes:
1370    
1371          fadd/fsub/fsubr/fmul/fdiv/fdivr  %st, %st(n)
1372          faddp/fsubp/fsubrp/fmulp/fdivp/fdivrp  %st, %st(n)
1373          fadd/fsub/fsubr/fmul/fdiv/fdivr  %st(n), %st
1374          fiadd/fisub/fisubr/fimul/fidiv/fidivr mem
1375          fxch %st(n)
1376          fld %st(n)
1377          fst %st(n)
1378          fst mem
1379          fstp %st(n)
1380          fucom %st(n)
1381          fucomp %st(n)
1382    
1383        All these are now generated when the fast fp mode is turned on.
1384    
1385    6.  Removed the dedicated registers %st(0), ..., %st(7) from X86CpsRegs
1386    
1387    ----------------------------------------------------------------------
1388    Name: Matthias Blume
1389    Date: 2000/11/09 11:20:00 JST
1390    Tag: blume-20001109-scc
1391    Description:
1392    
1393    Eliminated some code duplication:
1394    
1395    1. Added "where" clause to GraphSCCFn in SML/NJ Library.
1396       (Otherwise the functor is useless.)
1397    2. Used GraphSCCFn where SCCUtilFun was used previously.
1398    3. Got rid of SCCUtilFun (in comp-lib.cm).
1399    
1400    ----------------------------------------------------------------------
1401    Name: Lal George
1402    Date: 2000/11/06 09:02:21 EST 2000
1403    Tag: Release_110_30
1404    Description:
1405    
1406    - Version 110.30
1407    ----------------------------------------------------------------------
1408    Name: Matthias Blume
1409    Date: 2000/11/04 14:45:00
1410    Tag: blume-20001104-mlbuild
1411    Description:
1412    
1413    - Made ml-build faster on startup.
1414    - Documentation fixes.
1415    
1416    ----------------------------------------------------------------------
1417    Name: Matthias Blume
1418    Date: 2000/11/02 17:00:00 JST
1419    Tag: blume-20001102-condcomp
1420    Description:
1421    
1422    - Small tweaks to pickler -- new BOOTFILES!
1423    - Version bumped to 110.29.2.
1424    - Added conditional compilation facility to init.cmi (see comment there).
1425    ----------------------------------------------------------------------
1426    Name: Allen Leung
1427    Date: 2000/10/23 19:31:00
1428    Tag: leunga-20001023-demo-ra
1429    
1430    1. Minor RA changes that improves spilling on x86 (affects Moby and C-- only)
1431    2. Test programs for the graph library updated
1432    3. Some new MLRISC demo programs added
1433    
1434    ----------------------------------------------------------------------
1435    Name: Matthias Blume
1436    Date: 2000/08/31 22:15:00 JST
1437    Tag: blume-20001017-errmsg
1438    Description:
1439    
1440    More error message grief: Where there used to be no messages, there
1441    now were some that had bogus error regions.  Fixed.
1442    
1443    ----------------------------------------------------------------------
1444    Name: Matthias Blume
1445    Date: 2000/08/31 17:30:00 JST
1446    Tag: blume-20001017-v110p29p1
1447    Description:
1448    
1449    I made a version 110.29.1 with new bootfiles.
1450    
1451    Changes:  Modified pickler/unpickler for faster and leaner unpickling.
1452              CM documentation changes and a small bugfix in CM's error reporting.
1453    
1454    ----------------------------------------------------------------------
1455    Name: Lal George
1456    Date: 2000/09/27 14:42:35 EDT
1457    Tag: george-20000927-nodestatus
1458    Description:
1459    
1460    Changed the type of the nodestatus, so that:
1461    
1462            SPILLED(~1)             is now SPILLED
1463            SPILLED(m) where m>=0   is now MEMREG(m)
1464            SPILLED(s) where s<~1   is now SPILL_LOC(~s)
1465    
1466    ----------------------------------------------------------------------
1467    Name: Matthias Blume
1468    Date: 2000/09/07 14:45:00 JST
1469    Tag: blume-20000907-cmerrmsg
1470    Description:
1471    
1472    Small tweak to CM to avoid getting ML syntax error messages twice.
1473    
1474    ----------------------------------------------------------------------
1475    Name: Matthias Blume
1476    Date: 2000/08/31 18:00:00 JST
1477    Tag: blume-20000831-cvsbootfiles
1478    Description:
1479    
1480    New URL for boot files (because the 110.29 files on the BL server do
1481    now work correctly with my updated install scripts for yacc and lex).
1482    
1483    ----------------------------------------------------------------------
1484    Name: Matthias Blume
1485    Date: 2000/08/08 12:33:00 JST
1486    Tag: blume-20000808-manual
1487    Description:
1488    
1489    Tiny update to CM manual.
1490    
1491    ----------------------------------------------------------------------
1492    Name: Allen Leung
1493    Date: 2000/08/7 19:31:00
1494    Tag: leunga-20000807-a-whole-bunch-of-stuff
1495    
1496      Moby, C--, SSA, x86, machine descriptions etc.  Should only affect C--
1497    and Mobdy.
1498    
1499    1.  x86
1500    
1501       a.  Fixes to peephole module by John and Dan.
1502       b.  Assembly fix to SETcc by Allen.
1503       c.  Fix to c-call by John.
1504       d.  Fix to spilling by John.  (This one deals with the missing FSTPT case)
1505       e.  Instruction selection optimization to SETcc as suggested by John.
1506    
1507           For example,
1508    
1509            MV(32, x, COND(32, CMP(32, LT, a, b), LI 1, LI 0))
1510    
1511           should generate:
1512    
1513            MOVL a, x
1514            SUBL b, x
1515            SHRL 31, x
1516    
1517    2.  IR stuff
1518    
1519         A bunch of new DJ-graph related algorithms added.  These
1520         speed up SSA construction.
1521    
1522    3.  SSA + Scheduling
1523    
1524         Added code for SSA and scheduling to the repository
1525    
1526    ----------------------------------------------------------------------
1527    Name: Lal George
1528    Date: 2000/07/27 11:53:14 EDT
1529    
1530    Tag: lal-20000727-linux-ppc
1531    Description:
1532    
1533     Made changes to support Linux PPC.
1534     p.s. I have confirmation that the 110.29 boot files work fine.
1535    
1536    ----------------------------------------------------------------------
1537    Name: Matthias Blume
1538    Date: 2000/07/27 17:40:00 JST
1539    Tag: blume-20000727-scripts
1540    Description:
1541    
1542    !!!! WARNING !!!!
1543    You must recompile the runtime system!
1544    !!!! WARNING !!!!
1545    
1546    This is basically another round of script-enhancements:
1547    
1548    1. sml, ml-build, and ml-makedepend accept options -D and -U to define
1549       and undefine CM preprocessor symbols.
1550    
1551    2. ml-build avoids generating a new heap image if it finds that the
1552       existing one is still ok.  (The condition is that no ML file had to
1553       be recompiled and all ML files are found to be older that the heap
1554       file.)
1555    
1556       To make this work smoothly, I also hacked the runtime system as
1557       well as SMLofNJ.SysInfo to get access to the heap image suffix
1558       (.sparc-solaris, ...) that is currently being used.
1559    
1560       Moreover, the signature of CM.mk_standalone has changed.  See the
1561       CM manual.
1562    
1563    3. ml-makedepend accepts additional options -n, -a, and -o.  (See the
1564       CM manual for details.)
1565    
1566    4. More CM manual updates:
1567        - all of the above has been documented.
1568        - there is now a section describing the (CM-related) command line
1569          arguments that are accepted by the "sml" command
1570    
1571    ----------------------------------------------------------------------
1572    Name: Matthias Blume
1573    Date: 2000/07/25 16:20:00 JST
1574    Tag: blume-20000725-makedepend
1575    Description:
1576    
1577    Added a script called ml-makedepend.  This can be used in makefiles
1578    for Unix' make in a way very similar to the "makedepend" command for
1579    C.
1580    
1581    The script internally uses function CM.sources.
1582    
1583    Synopsis:
1584    
1585        ml-makedepend [-f makefile] cmfile targetname
1586    
1587    The default for the makefile is "makefile" (or "Makefile" should
1588    "makefile" not exist).
1589    
1590    ml-makedepend adds a cmfile/targetname-specific section to this
1591    makefile (after removing the previous version of this section).  The
1592    section contains a single dependency specification with targetname on
1593    the LHS (targetname is an arbitrary name), and a list of files derived
1594    from the cmfile on the RHS.  Some of the files on the RHS are
1595    ARCH/OPSYS-specific.  Therefore, ml-makedepend inserts references to
1596    "make" variables $(ARCH) and $(OPSYS) in place of the corresponding
1597    path names.  The makefile writer is responsible for making sure that
1598    these variables have correct at the time "make" is invoked.
1599    
1600    ----------------------------------------------------------------------
1601    Name: Matthias Blume
1602    Date: 2000/07/22 23:30:00 JST
1603    Tag: blume-20000722-urlupdate
1604    Description:
1605    
1606    Changed BOOT and config/srcarchiveurl to point to BL server:
1607    
1608        ftp://ftp.research.bell-labs.com/dist/smlnj/working/110.29/
1609    
1610    ----------------------------------------------------------------------
1611    Name: Matthias Blume
1612    Date: 2000/07/18 18:00:00 JST
1613    Tag: blume-20000718-Version_110_29
1614    Description:
1615    
1616    1. Updated src/compiler/TopLevel/main/version.sml to version 110.29
1617    
1618    2. Updated config/version to 110.29
1619    
1620    3. Updated config/srcarchiveurl
1621    
1622    3. New boot files!
1623       ftp://ftp.cs.princeton.edu/pub/people/blume/sml/110.29-autofetch
1624    
1625    ----------------------------------------------------------------------
1626    Name: Matthias Blume
1627    Date: 2000/07/11 13:58:00 JST
1628    Tag: blume-20000711-doctypo
1629    Description:
1630    
1631    Fixed a few typos in CM manual.
1632    
1633    ----------------------------------------------------------------------
1634    Name: Allen Leung
1635    Date: 2000/06/15 00:38:00
1636    Tag: leunga-20000704-sparc-x86
1637    
1638    1. x86 peephole improvement sp += k; sp -= k => nop  [from John]
1639    2. fix to x86 RET bug [found by Dan Grossman]
1640    3. sparc assembly bug fix for ticc instructions [found by Fermin]
1641    
1642       Affects c-- and moby only
1643    
1644    ----------------------------------------------------------------------
1645    Name: Matthias Blume
1646    Date: 2000/07/04 15:26:00
1647    Tag: blume-20000704-trigger
1648    Description:
1649    
1650    1. Improvements to CM manual.
1651    2. SMLofNJ.Internals.BTrace.trigger reinstated as an alternative way
1652       of getting a back-trace.  The function, when called, raises an
1653       internal exception which explicitly carries the full back-trace history,
1654       so it is unaffected by any intervening handle-raise pairs ("trivial"
1655       or not).  The interactive loop will print that history once it arrives
1656       at top level.
1657       Short of having all exceptions implicitly carry the full history, the
1658       recommended way of using this facility is:
1659         - compile your program with instrumentation "on"
1660         - run it, when it raises an exception, look at the history
1661         - if the history is "cut off" because of some handler, go and modify
1662           your program so that it explicitly calls BTrace.trigger
1663         - recompile (still instrumented), and rerun; look at the full history
1664    
1665    ----------------------------------------------------------------------
1666    Name: Matthias Blume
1667    Date: 2000/07/03 15:36:00 JST
1668    Tag: blume-20000702-manual
1669    Description:
1670    
1671    Small corrections and updates to CM manual.
1672    
1673    ----------------------------------------------------------------------
1674    Name: Matthias Blume
1675    Date: 2000/06/29 16:04:00 JST
1676    Tag: blume-20000629-yacctool
1677    Description:
1678    
1679    Changes:
1680    
1681    1. Class "mlyacc" now takes separate arguments to pass options to
1682       generated .sml- and .sig-files independently.
1683    2. Corresponding CM manual updates.
1684    3. BTrace module now also reports call sites.  (However, for loop clusters
1685       it only shows from where the cluster was entered.)  There are associated
1686       modifications to core.sml, internals.{sig,sml}, btrace.sml, and btimp.sml.
1687    
1688    ----------------------------------------------------------------------
1689    Name: Matthias Blume
1690    Date: 2000/06/27 16:51:00 JST
1691    Tag: blume-20000627-noweb
1692    Description:
1693    
1694    Changes:
1695    
1696     1. Implemented "subdir" and "witness" options for noweb tool.
1697        This caused some slight internal changes in CM's tool implementation.
1698     2. Fixed bug in "tool plugin" mechanism.  This is essentially cleaning
1699        some remaining issues from earlier path anchor changes.
1700     3. Updated CM manual accordingly.
1701    
1702     4. Changed implementation of back-tracing so that I now consider it
1703        ready for prime-time.
1704    
1705        In particular, you don't have to explicitly trigger the back-trace
1706        anymore.  Instead, if you are running BTrace-instrumented code and
1707        there is an uncaught exception (regardless of whether or not it was
1708        raised in instrumented code), the top-level evalloop will print
1709        the back-trace.
1710    
1711        Features:
1712    
1713          - Instrumented and uninstrumented code work together seemlessly.
1714            (Of course, uninstrumented code is never mentioned in actual
1715             back-traces.)
1716    
1717          - Asymptotic time- and space-complexity of instrumented code is
1718            equal to that of uninstrumented code.  (This means that
1719            tail-recursion is preserved by the instrumentation phase.)
1720    
1721          - Modules whose code has been instrumented in different sessions
1722            work together without problem.
1723    
1724          - There is no penalty whatsoever on uninstrumented code.
1725    
1726          - There is no penalty on "raise" expressions, even in
1727            instrumented code.
1728    
1729        A potential bug (or perhaps it is a feature, too):
1730    
1731          A back-trace reaches no further than the outermost instrumented
1732          non-trivial "raise".  Here, a "trivial" raise is one that is the
1733          sole RHS of a "handle" rule.  Thus, back-traces reach trough
1734    
1735               <exp> handle e => raise e
1736    
1737          and even
1738    
1739               <exp> handle Foo => raise Bar
1740    
1741          and, of course, through
1742    
1743               <exp> handle Foo => ...
1744    
1745         if the exception was not Foo.
1746    
1747         Back-traces always reach right through any un-instrumented code
1748         including any of its "handle" expressions, trivial or not.
1749    
1750       To try this out, do the following:
1751    
1752         - Erase all existing binfiles for your program.
1753           (You may keep binfiles for those modules where you think you
1754            definitely don't need back-tracing.)
1755         - Turn on back-trace instrumentation:
1756              SMLofNJ.Internals.BTrace.mode (SOME true);
1757         - Recompile your program.  (I.e., run "CM.make" or "use".)
1758         - You may now turn instrumentation off again (if you want):
1759              SMLofNJ.Internals.BTrace.mode (SOME false);
1760         - Run your program as usual.  If it raises an exception that
1761           reaches the interactive toplevel, then a back-trace will
1762           automatically be printed.  After that, the toplevel loop
1763           will print the exception history as usual.
1764    
1765  ----------------------------------------------------------------------  ----------------------------------------------------------------------
1766  Name: Matthias Blume  Name: Matthias Blume
1767  Date: 2000/06/26 09:56:46 JST  Date: 2000/06/26 09:56:46 JST
# Line 1188  Line 2939 
2939       elaborator). There were a lot of changes during my "linkpath" trials       elaborator). There were a lot of changes during my "linkpath" trials
2940       that could have been reverted to their original state but weren't.       that could have been reverted to their original state but weren't.
2941       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
2942       than what was strictly necessary...  (I _did_ resist the tempation       than what was strictly necessary...  (I _did_ resist the temptation
2943       of doing any "global reformatting" to avoid an untimely death at       of doing any "global reformatting" to avoid an untimely death at
2944       Dave's hands. :)       Dave's hands. :)
2945    

Legend:
Removed from v.677  
changed lines
  Added in v.921

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