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/NOTES/HISTORY
ViewVC logotype

Diff of /sml/trunk/NOTES/HISTORY

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

revision 701, Thu Aug 31 08:46:43 2000 UTC revision 1183, Fri Mar 29 19:09:48 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
17    Date: 2002/03/29 14:10:00 EST
18    Tag: blume-20020329-inlprims
19    Description:
20    
21    NEW BOOTFILES!!!    Version number bumped to 110.39.3.
22    
23    Primops have changed. This means that the bin/boot-file formats have
24    changed as well.
25    
26    To make sure that there is no confusion, I made a new version.
27    
28    
29    CHANGES:
30    
31    * removed REMT from mltree (remainder should never overflow).
32    
33    * added primops to deal with divisions of all flavors to the frontend
34    
35    * handled these primops all the way through so they map to their respective
36      MLRISC support
37    
38    * used these primops in the implementation of Int, Int32, Word, Word32
39    
40    * removed INLDIV, INLMOD, and INLREM as they are no longer necessary
41    
42    * parameterized INLMIN, INLMAX, and INLABS by a numkind
43    
44    * translate.sml now deals with all flavors of INL{MIN,MAX,ABS}, including
45      floating point
46    
47    * used INL{MIN,MAX,ABS} in the implementation of Int, Int32, Word, Word32,
48      and Real (but Real.abs maps to a separate floating-point-only primop)
49    
50    
51    TODO items:
52    
53    * Hacked Alpha32 instruction selection, disabling the selection of REMx
54      instructions because the machine instruction encoder cannot handle
55      them.  (Hppa, PPC, and Sparc instruction selection did not handle
56      REM in the first place, and REM is supported by the x86 machine coder.)
57    
58    * Handle DIV and MOD with DIV_TO_NEGINF directly in the x86 instruction
59      selection phase.  (The two can be streamlined because the hardware
60      delivers both quotient and remainder at the same time anyway.)
61    
62    * Think about what to do with "valOf(Int32.minInt) div ~1" and friends.
63      (Currently the behavior is inconsistent both across architectures and
64      wrt. the draft Basis spec.)
65    
66    * Word8 should eventually be handled natively, too.
67    
68    * There seems to be one serious bug in mltree-gen.sml.  It appears, though,
69      as if there currently is no execution path that could trigger it in
70      SML/NJ.  (The assumptions underlying functions arith and promotable do not
71      hold for things like multiplication and division.)
72    
73    ----------------------------------------------------------------------
74    Name: Matthias Blume
75    Date: 2002/03/27 16:27:00 EST
76    Tag: blume-20020327-mlrisc-divisions
77    Description:
78    
79    Added support for all four division operations (ML's div, mod, quot,
80    and rem) to MLRISC.  In the course of doing so, I also rationalized
81    the naming (no more annoying switch-around of DIV and QUOT), by
82    parameterizing the operation by div_rounding_mode (which can be either
83    DIV_TO_ZERO or DIV_TO_NEGINF).
84    
85    The generic MLTreeGen functor takes care of compiling all four
86    operations down to only round-to-zero div.
87    
88    Missing pieces:
89    
90      * Doing something smarter than relying on MLTreeGen on architectures
91        like, e.g., the x86 where hardware division delivers both quotient and
92        remainder at the same time.  With this, the implementation of the
93        round-to-neginf operations could be further streamlined.
94    
95      * Remove inlining support for div/mod/rem from the frontend and replace it
96        with primops that get carried through to the backend.  Do this for all
97        int and word types.
98    
99    ----------------------------------------------------------------------
100    Name: Matthias Blume
101    Date: 2002/03/25 17:25:00 EST
102    Tag: blume-20020325-divmod
103    Description:
104    
105    I improved (hopefully without breaking them) the implementation of Int.div,
106    Int.mod, and Int.rem.   For this, the code in translate.sml now takes
107    advantage of the following observations:
108    
109      Let  q = x quot y      r = x rem y
110           d = x div  y      m = x mod y
111    
112    where "quot" is the round-to-zero version of integer division that
113    hardware usually provides.  Then we have:
114    
115         r = x - q * y        where neither the * nor the - will overflow
116         d = if q >= 0 orelse x = q * y then q else q - 1
117                              where neither the * nor the - will overflow
118         m = if q >= 0 orelse r = 0 then r else r + y
119                              where the + will not overflow
120    
121    This results in substantial simplification of the generated code.
122    The following table shows the number of CFG nodes and edges generated
123    for
124            fun f (x, y) = x OPER y
125            (* with OPER \in div, mod, quot, rem *)
126    
127    
128        OPER | nodes(old) | edges(old) | nodes(new) | edges(new)
129        --------------------------------------------------------
130         div |         24 |         39 |         12 |         16
131         mod |         41 |         71 |         12 |         16
132        quot |          8 |         10 |          8 |         10
133         rem |         10 |         14 |          8 |         10
134    
135    
136    ----------------------------------------------------------------------
137    Name: Matthias Blume
138    Date: 2002/03/25 22:06:00 EST
139    Tag: blume-20020325-cprotobug
140    Description:
141    
142    Fixed a bug in cproto (c prototype decoder).
143    
144    ----------------------------------------------------------------------
145    Name: Matthias Blume
146    Date: 2002/03/25 16:00:00 EST
147    Tag: blume-20020325-raw-primops
148    Description:
149    
150    I did some cleanup to Allen's new primop code and
151    replaced yesterday's bootfiles with new ones.
152    (But they are stored in the same place.)
153    
154    ----------------------------------------------------------------------
155    Name: Matthias Blume
156    Date: 2002/03/24 22:40:00 EST
157    Tag: blume-20020324-bootfiles
158    Description:
159    
160    Made the bootfiles that Allen asked for.
161    
162    ----------------------------------------------------------------------
163    Name: Allen Leung
164    Date: 2002/03/23 15:50:00 EST
165    Tag: leunga-20020323-flint-cps-rcc-primops
166    Description:
167    
168      1. Changes to FLINT primops:
169    
170        (* make a call to a C-function;
171         * The primop carries C function prototype information and specifies
172         * which of its (ML-) arguments are floating point. C prototype
173         * information is for use by the backend, ML information is for
174         * use by the CPS converter. *)
175      | RAW_CCALL of { c_proto: CTypes.c_proto,
176                       ml_args: ccall_type list,
177                       ml_res_opt: ccall_type option,
178                       reentrant : bool
179                     } option
180       (* Allocate uninitialized storage on the heap.
181        * The record is meant to hold short-lived C objects, i.e., they
182        * are not ML pointers.  With the tag, the representation is
183        * the same as RECORD with tag tag_raw32 (sz=4), or tag_fblock (sz=8)
184        *)
185      | RAW_RECORD of {tag:bool,sz:int}
186      and ccall_type = CCALL_INT32 | CCALL_REAL64 | CCALL_ML_PTR
187    
188      2.  These CPS primops are now overloaded:
189    
190           rawload of {kind:numkind}
191           rawstore of {kind:numkind}
192    
193          The one argument form is:
194    
195             rawload {kind} address
196    
197          The two argument form is:
198    
199             rawload {kind} [ml object, byte-offset]
200    
201      3. RAW_CCALL/RCC now takes two extra arguments:
202    
203         a. The first is whether the C call is reentrant, i.e., whether
204            ML state should be saved and restored.
205         b. The second argument is a string argument specifying the name of
206            library and the C function.
207    
208         These things are currently not handled in the code generator, yet.
209    
210      4. In CProto,
211    
212         An encoding type of "bool" means "ml object" and is mapped into
213         C prototype of PTR.  Note that "bool" is different than "string",
214         even though "string" is also mapped into PTR, because "bool"
215         is assigned an CPS type of BOGt, while "string" is assigned INT32t.
216    
217      5. Pickler/unpicker
218    
219         Changed to handle RAW_RECORD and newest RAW_CCALL
220    
221      6. MLRiscGen,
222    
223         1. Changed to handle the new rawload/rawstore/rawrecord operators.
224         2. Code for handling C Calls has been moved to a new module CPSCCalls,
225            in the file CodeGen/cpscompile/cps-c-calls.sml
226    
227      7. Added the conditional move operator
228    
229             condmove of branch
230    
231         to cps.  Generation of this is still buggy so it is currently
232         disabled.
233    
234    ----------------------------------------------------------------------
235    Name: Lal George
236    Date: 2002/03/22 14:18:25 EST
237    Tag: george-20020322-cps-branch-prob
238    Description:
239    
240    Implemented the Ball-Larus branch prediction-heuristics, and
241    incorporated graphical viewers for control flow graphs.
242    
243    Ball-Larus Heuristics:
244    ---------------------
245    See the file compiler/CodeGen/cpscompile/cpsBranchProb.sml.
246    
247    By design it uses the Dempster-Shafer theory for combining
248    probabilities.  For example, in the function:
249    
250        fun f(n,acc) = if n = 0 then acc else f(n-1, n*acc)
251    
252    the ball-larus heuristics predicts that the n=0 is unlikely
253    (OH-heuristic), and the 'then' branch is unlikely because of the
254    RH-heuristic -- giving the 'then' branch an even lower combined
255    probability using the Dempster-Shafer theory.
256    
257    Finally, John Reppy's loop analysis in MLRISC, further lowers the
258    probability of the 'then' branch because of the loop in the else
259    branch.
260    
261    
262    Graphical Viewing:
263    ------------------
264    I merely plugged in Allen's graphical viewers into the compiler. The
265    additional code is not much. At the top level, saying:
266    
267            Control.MLRISC.getFlag "cfg-graphical-view" := true;
268    
269    will display the graphical view of the control flow graph just before
270    back-patching.  daVinci must be in your path for this to work. If
271    daVinci is not available, then the default viewer can be changed
272    using:
273    
274            Control.MLRISC.getString "viewer"
275    
276    which can be set to "dot" or "vcg" for the corresponding viewers. Of
277    course, these viewers must be in your path.
278    
279    The above will display the compilation unit at the level of clusters,
280    many of which are small, boring, and un-interesting. Also setting:
281    
282            Control.MLRISC.getInt "cfg-graphical-view_size"
283    
284    will display clusters that are larger than the value set by the above.
285    
286    
287    ----------------------------------------------------------------------
288    Name: Matthias Blume
289    Date: 2002/03/21 22:20:00 EST
290    Tag: blume-20020321-kmp-bugfix
291    Description:
292    
293    Changed the interface to the KMP routine in PreString and fixed
294    a minor bug in one place where it was used.
295    
296    ----------------------------------------------------------------------
297    Name: Allen Leung
298    Date: 2002/03/21 20:30:00 EST
299    Tag: leunga-20020321-cfg
300    Description:
301    
302      Fixed a potential problem in cfg edge splitting.
303    
304    ----------------------------------------------------------------------
305    Name: Allen Leung
306    Date: 2002/03/21 17:15:00 EST
307    Tag: leunga-20020321-x86-fp-cfg
308    Description:
309    
310      1. Recoded the buggy parts of x86-fp.
311    
312         a. All the block reordering code has been removed.
313            We now depend on the block placement phases to do this work.
314    
315         b. Critical edge splitting code has been simplified and moved into the
316            CFG modules, as where they belong.
317    
318         Both of these were quite buggy and complex.  The code is now much, much
319         simpler.
320    
321      2. X86 backend.
322    
323         a. Added instructions for 64-bit support.  Instruction selection for
324            64-bit has not been committed, however, since that
325            requires changes to MLTREE which haven't been approved by
326            Lal and John.
327    
328         b. Added support for FUCOMI and FUCOMIP when generating code for
329            PentiumPro and above.  We only generate these instructions in
330            the fast-fp mode.
331    
332         c. Added cases for JP and JNP in X86FreqProps.
333    
334      3. CFG
335    
336         CFG now has a bunch of methods for edge splitting and merging.
337    
338      4. Machine description.
339    
340         John's simplification of MLTREE_BASIS.fcond broke a few machine
341         description things:
342    
343         rtl-build.{sig,sml} and hppa.mdl fixed.
344    
345         NOTE: the machine description stuff in the repository is still broken.
346               Again, I can't put my fixes in because that involves
347               changes to MLTREE.
348    
349    ----------------------------------------------------------------------
350    Name: Matthias Blume
351    Date: 2002/03/20 15:55:00 EST
352    Tag: blume-20020320-kmp
353    Description:
354    
355    Implemented Knuth-Morris-Pratt string matching in PreString and used
356    it for String.isSubstring, Substring.isSubstring, and
357    Substring.position.
358    
359    (Might need some stress-testing.  Simple examples worked fine.)
360    
361    ----------------------------------------------------------------------
362    Name: Matthias Blume
363    Date: 2002/03/19 16:37:00 EST
364    Tag: blume-20020319-witnesses
365    Description:
366    
367    Added a structure C.W and functions convert/Ptr.convert to ml-nlffi-lib.
368    
369    This implements a generic mechanism for changing constness qualifiers
370    anywhere within big C types without resorting to outright "casts".
371    (So far, functions such as C.rw/C.ro or C.Ptr.rw/C.Ptr.ro only let you
372    modify the constness at the outermost level.)
373    The implementation of "convert" is based on the idea of "witness"
374    values -- values that are not used by the operation but whose types
375    "testify" to their applicability.  On the implementation side, "convert"
376    is simply a projection (returning its second curried argument).  With
377    cross-module inlining, it should not result in any machine code being
378    generated.
379    
380    ----------------------------------------------------------------------
381    Name: Matthias Blume
382    Date: 2002/03/15 16:40:00 EST
383    Tag: blume-20020315-basis
384    Description:
385    
386    Provided (preliminary?) implementations for
387    
388      {String,Substring}.{concatWith,isSuffix,isSubstring}
389    
390    and
391    
392      Substring.full
393    
394    Those are in the Basis spec but they were missing in SML/NJ.
395    
396    ----------------------------------------------------------------------
397    Name: Matthias Blume
398    Date: 2002/03/14 21:30:00 EST
399    Tag: blume-20020314-controls
400    Description:
401    
402    Controls:
403    ---------
404    
405    1. Factored out the recently-added Controls : CONTROLS stuff and put
406       it into its own library $/controls-lib.cm.  The source tree for
407       this is under src/smlnj-lib/Controls.
408    
409    2. Changed the names of types and functions in this interface, so they
410       make a bit more "sense":
411    
412          module -> registry
413          'a registry -> 'a group
414    
415    3. The interface now deals in ref cells only.  The getter/setter interface
416       is (mostly) gone.
417    
418    4. Added a function that lets one register an already-existing ref cell.
419    
420    5. Made the corresponding modifications to the rest of the code so that
421       everything compiles again.
422    
423    6. Changed the implementation of Controls.MLRISC back to something closer
424       to the original.  In particular, this module (and therefore MLRISC)
425       does not depend on Controls.  There now is some link-time code in
426       int-sys.sml that registers the MLRISC controls with the Controls
427       module.
428    
429    CM:
430    ---
431    
432      * One can now specify the lambda-split aggressiveness in init.cmi.
433    
434    ----------------------------------------------------------------------
435    Name: Allen Leung
436    Date: 2002/03/13 17:30:00 EST
437    Tag: leunga-20020313-x86-fp-unary
438    Description:
439    
440    Bug fix for:
441    
442    > leunga@weaselbane:~/Yale/tmp/sml-dist{21} bin/sml
443    > Standard ML of New Jersey v110.39.1 [FLINT v1.5], March 08, 2002
444    > - fun f(x,(y,z)) = Real.~ y;
445    > [autoloading]
446    > [autoloading done]
447    >       fchsl   (%eax), 184(%esp)
448    > Error: MLRisc bug: X86MCEmitter.emitInstr
449    >
450    > uncaught exception Error
451    >   raised at: ../MLRISC/control/mlriscErrormsg.sml:16.14-16.19
452    
453    The problem was that the code generator did not generate any fp registers
454    in this case, and the ra didn't know that it needed to run the X86FP phase to
455    translate the pseudo fp instruction.   This only happened with unary fp
456    operators in certain situations.
457    
458    ----------------------------------------------------------------------
459    Name: Matthias Blume
460    Date: 2002/03/13 14:00:00 EST
461    Tag: blume-20020313-overload-etc
462    Description:
463    
464    1. Added _overload as a synonym for overload for backward compatibility.
465       (Control.overloadKW must be true for either version to be accepted.)
466    
467    2. Fixed bug in install script that caused more things to be installed
468       than what was requested in config/targets.
469    
470    3. Made CM aware of the (_)overload construct so that autoloading
471       works.
472    
473    ----------------------------------------------------------------------
474    Name: Matthias Blume
475    Date: 2002/03/12 22:03:00 EST
476    Tag: blume-20020312-url
477    Description:
478    
479    Forgot to update BOOT and srcarchiveurl.
480    
481    ----------------------------------------------------------------------
482    Name: Matthias Blume
483    Date: 2002/03/12 17:30:00 EST
484    Tag: blume-20020312-version110392
485    Description:
486    
487    Yet another version number bump (because of small changes to the
488    binfile format).  Version number is now 110.39.2.  NEW BOOTFILES!
489    
490    Changes:
491    
492      The new pid generation scheme described a few weeks ago was overly
493      complicated.  I implemented a new mechanism that is simpler and
494      provides a bit more "stability":  Once CM has seen a compilation
495      unit, it keeps its identity constant (as long as you do not delete
496      those crucial CM/GUID/* files).  This means that when you change
497      an interface, compile, then go back to the old interface, and
498      compile again, you arrive at the original pid.
499    
500      There now also is a mechanism that instructs CM to use the plain
501      environment hash as a module's pid (effectively making its GUID
502      the empty string).  For this, "noguid" must be specified as an
503      option to the .sml file in question within its .cm file.
504      This is most useful for code that is being generated by tools such
505      as ml-nlffigen (because during development programmers tend to
506      erase the tool's entire output directory tree including CM's cached
507      GUIDs).  "noguid" is somewhat dangerous (since it can be used to locally
508      revert to the old, broken behavior of SML/NJ, but in specific cases
509      where there is no danger of interface confusion, its use is ok
510      (I think).
511    
512      ml-nlffigen by default generates "noguid" annotations.  They can be
513      turned off by specifying -guid in its command line.
514    
515    ----------------------------------------------------------------------
516    Name: Lal George
517    Date: 2002/03/12 12 14:42:36 EST
518    Tag: george-20020312-frequency-computation
519    Description:
520    
521    Integrated jump chaining and static block frequency into the
522    compiler. More details and numbers later.
523    
524    ----------------------------------------------------------------------
525    Name: Lal George
526    Date: 2002/03/11 11 22:38:53 EST
527    Tag: george-20020311-jump-chain-elim
528    Description:
529    
530    Tested the jump chain elimination on all architectures (except the
531    hppa).  This is on by default right now and is profitable for the
532    alpha and x86, however, it may not be profitable for the sparc and ppc
533    when compiling the compiler.
534    
535    The gc test will typically jump to a label at the end of the cluster,
536    where there is another jump to an external cluster containing the actual
537    code to invoke gc. This is to allow factoring of common gc invocation
538    sequences. That is to say, we generate:
539    
540            f:
541               testgc
542               ja   L1      % jump if above to L1
543    
544            L1:
545               jmp L2
546    
547    
548    After jump chain elimination the 'ja L1' instructions is converted to
549    'ja L2'. On the sparc and ppc, many of the 'ja L2' instructions may end
550    up being implemented in their long form (if L2 is far away) using:
551    
552            jbe     L3      % jump if below or equal to L3
553            jmp     L2
554         L3:
555            ...
556    
557    
558    For large compilation units L2  may be far away.
559    
560    
561    ----------------------------------------------------------------------
562    Name: Matthias Blume
563    Date: 2002/03/11 13:30:00 EST
564    Tag: blume-20020311-mltreeeval
565    Description:
566    
567    A functor parameter was missing.
568    
569    ----------------------------------------------------------------------
570    Name: Allen Leung
571    Date: 2002/03/11 10:30:00 EST
572    Tag: leunga-20020311-runtime-string0
573    Description:
574    
575       The representation of the empty string now points to a
576    legal null terminated C string instead of unit.  It is now possible
577    to convert an ML string into C string with InlineT.CharVector.getData.
578    This compiles into one single machine instruction.
579    
580    ----------------------------------------------------------------------
581    Name: Allen Leung
582    Date: 2002/03/10 23:55:00 EST
583    Tag: leunga-20020310-x86-call
584    Description:
585    
586       Added machine generation for CALL instruction (relative displacement mode)
587    
588    ----------------------------------------------------------------------
589    Name: Matthias Blume
590    Date: 2002/03/08 16:05:00
591    Tag: blume-20020308-entrypoints
592    Description:
593    
594    Version number bumped to 110.39.1.  NEW BOOTFILES!
595    
596    Entrypoints: non-zero offset into a code object where execution should begin.
597    
598    - Added the notion of an entrypoint to CodeObj.
599    - Added reading/writing of entrypoint info to Binfile.
600    - Made runtime system bootloader aware of entrypoints.
601    - Use the address of the label of the first function given to mlriscGen
602      as the entrypoint.  This address is currently always 0, but it will
603      not be 0 once we turn on block placement.
604    - Removed the linkage cluster code (which was The Other Way(tm) of dealing
605      with entry points) from mlriscGen.
606    
607    ----------------------------------------------------------------------
608    Name: Allen Leung
609    Date: 2002/03/07 20:45:00 EST
610    Tag: leunga-20020307-x86-cmov
611    Description:
612    
613       Bug fixes for CMOVcc on x86.
614    
615       1. Added machine code generation for CMOVcc
616       2. CMOVcc is now generated in preference over SETcc on PentiumPro or above.
617       3. CMOVcc cannot have an immediate operand as argument.
618    
619    ----------------------------------------------------------------------
620    Name: Matthias Blume
621    Date: 2002/03/07 16:15:00 EST
622    Tag: blume-20020307-controls
623    Description:
624    
625    This is a very large but mostly boring patch which makes (almost)
626    every tuneable compiler knob (i.e., pretty much everything under
627    Control.* plus a few other things) configurable via both the command
628    line and environment variables in the style CM did its configuration
629    until now.
630    
631    Try starting sml with '-h' (or, if you are brave, '-H')
632    
633    To this end, I added a structure Controls : CONTROLS to smlnj-lib.cm which
634    implements the underlying generic mechanism.
635    
636    The interface to some of the existing such facilities has changed somewhat.
637    For example, the MLRiscControl module now provides mkFoo instead of getFoo.
638    (The getFoo interface is still there for backward-compatibility, but its
639    use is deprecated.)
640    
641    The ml-build script passes -Cxxx=yyy command-line arguments through so
642    that one can now twiddle the compiler settings when using this "batch"
643    compiler.
644    
645    TODO items:
646    
647    We should go through and throw out all controls that are no longer
648    connected to anything.  Moreover, we should go through and provide
649    meaningful (and correct!) documentation strings for those controls
650    that still are connected.
651    
652    Currently, multiple calls to Controls.new are accepted (only the first
653    has any effect).  Eventually we should make sure that every control
654    is being made (via Controls.new) exactly once.  Future access can then
655    be done using Controls.acc.
656    
657    Finally, it would probably be a good idea to use the getter-setter
658    interface to controls rather than ref cells.  For the time being, both
659    styles are provided by the Controls module, but getter-setter pairs are
660    better if thread-safety is of any concern because they can be wrapped.
661    
662    *****************************************
663    
664    One bug fix: The function blockPlacement in three of the MLRISC
665    backpatch files used to be hard-wired to one of two possibilities at
666    link time (according to the value of the placementFlag).  But (I
667    think) it should rather sense the flag every time.
668    
669    *****************************************
670    
671    Other assorted changes (by other people who did not supply a HISTORY entry):
672    
673    1. the cross-module inliner now works much better (Monnier)
674    2. representation of weights, frequencies, and probabilities in MLRISC
675       changed in preparation of using those for weighted block placement
676       (Reppy, George)
677    
678    ----------------------------------------------------------------------
679    Name: Lal George
680    Date: 2002/03/07 14:44:24 EST 2002
681    Tag: george-20020307-weighted-block-placement
682    
683    Tested the weighted block placement optimization on all architectures
684    (except the hppa) using AMPL to generate the block and edge frequencies.
685    Changes were required in the machine properties to correctly
686    categorize trap instructions. There is an MLRISC flag
687    "weighted-block-placement" that can be used to enable weighted block
688    placement, but this will be ineffective without block/edge
689    frequencies (coming soon).
690    
691    
692    ----------------------------------------------------------------------
693    Name: Lal George
694    Date: 2002/03/05 17:24:48 EST
695    Tag: george-20020305-linkage-cluster
696    
697    In order to support the block placement optimization, a new cluster
698    is generated as the very first cluster (called the linkage cluster).
699    It contains a single jump to the 'real' entry point for the compilation
700    unit. Block placement has no effect on the linkage cluster itself, but
701    all the other clusters  have full freedom in the manner in which they
702    reorder blocks or functions.
703    
704    On the x86 the typical linkage code that is generated is:
705       ----------------------
706            .align 2
707       L0:
708            addl    $L1-L0, 72(%esp)
709            jmp     L1
710    
711    
712            .align  2
713       L1:
714       ----------------------
715    
716    72(%esp) is the memory location for the stdlink register. This
717    must contain the address of the CPS function being called. In the
718    above example, it contains the address of  L0; before
719    calling L1 (the real entry point for the compilation unit), it
720    must contain the address for L1, and hence
721    
722            addl $L1-L0, 72(%esp)
723    
724    I have tested this on all architectures except the hppa.The increase
725    in code size is of course negligible
726    
727    ----------------------------------------------------------------------
728    Name: Allen Leung
729    Date: 2002/03/03 13:20:00 EST
730    Tag: leunga-20020303-mlrisc-tools
731    
732      Added #[ ... ] expressions to mlrisc tools
733    
734    ----------------------------------------------------------------------
735    Name: Matthias Blume
736    Date: 2002/02/27 12:29:00 EST
737    Tag: blume-20020227-cdebug
738    Description:
739    
740    - made types in structure C and C_Debug to be equal
741    - got rid of code duplication (c-int.sml vs. c-int-debug.sml)
742    - there no longer is a C_Int_Debug (C_Debug is directly derived from C)
743    
744    ----------------------------------------------------------------------
745    Name: Matthias Blume
746    Date: 2002/02/26 12:00:00 EST
747    Tag: blume-20020226-ffi
748    Description:
749    
750    1. Fixed a minor bug in CM's "noweb" tool:
751       If numbering is turned off, then truly don't number (i.e., do not
752       supply the -L option to noweb).  The previous behavior was to supply
753       -L'' -- which caused noweb to use the "default" line numbering scheme.
754       Thanks to Chris Richards for pointing this out (and supplying the fix).
755    
756    2. Once again, I reworked some aspects of the FFI:
757    
758       A. The incomplete/complete type business:
759    
760       - Signatures POINTER_TO_INCOMPLETE_TYPE and accompanying functors are
761         gone!
762       - ML types representing an incomplete type are now *equal* to
763         ML types representing their corresponding complete types (just like
764         in C).  This is still safe because ml-nlffigen will not generate
765         RTTI for incomplete types, nor will it generate functions that
766         require access to such RTTI.   But when ML code generated from both
767         incomplete and complete versions of the C type meet, the ML types
768         are trivially interoperable.
769    
770         NOTE:  These changes restore the full generality of the translation
771         (which was previously lost when I eliminated functorization)!
772    
773       B. Enum types:
774    
775       - Structure C now has a type constructor "enum" that is similar to
776         how the "su" constructor works.  However, "enum" is not a phantom
777         type because each "T enum" has values (and is isomorphic to
778         MLRep.Signed.int).
779       - There are generic access operations for enum objects (using
780         MLRep.Signed.int).
781       - ml-nlffigen will generate a structure E_foo for each "enum foo".
782         * The structure contains the definition of type "mlrep" (the ML-side
783         representation type of the enum).  Normally, mlrep is the same
784         as "MLRep.Signed.int", but if ml-nlffigen was invoked with "-ec",
785         then mlrep will be defined as a datatype -- thus facilitating
786         pattern matching on mlrep values.
787         ("-ec" will be suppressed if there are duplicate values in an
788          enumeration.)
789         * Constructors ("-ec") or values (no "-ec") e_xxx of type mlrep
790         will be generated for each C enum constant xxx.
791         * Conversion functions m2i and i2m convert between mlrep and
792         MLRep.Signed.int.  (Without "-ec", these functions are identities.)
793         * Coversion functions c and ml convert between mlrep and "tag enum".
794         * Access functions (get/set) fetch and store mlrep values.
795       - By default (unless ml-nlffigen was invoked with "-nocollect"), unnamed
796         enumerations are merged into one single enumeration represented by
797         structure E_'.
798    
799    ----------------------------------------------------------------------
800    Name: Allen Leung
801    Date: 2002/02/25 04:45:00 EST
802    Tag: leunga-20020225-cps-spill
803    
804    This is a new implementation of the CPS spill phase.
805    The new phase is in the new file compiler/CodeGen/cpscompile/spill-new.sml
806    In case of problems, replace it with the old file spill.sml
807    
808    The current compiler runs into some serious performance problems when
809    constructing a large record.  This can happen when we try to compile a
810    structure with many items.  Even a very simple structure like the following
811    makes the compiler slow down.
812    
813        structure Foo = struct
814           val x_1 = 0w1 : Word32.int
815           val x_2 = 0w2 : Word32.int
816           val x_3 = 0w3 : Word32.int
817           ...
818           val x_N = 0wN : Word32.int
819        end
820    
821    The following table shows the compile time, from N=1000 to N=4000,
822    with the old compiler:
823    
824    N
825    1000   CPS 100 spill                           0.04u  0.00s  0.00g
826           MLRISC ra                               0.06u  0.00s  0.05g
827              (spills = 0 reloads = 0)
828           TOTAL                                   0.63u  0.07s  0.21g
829    
830    1100   CPS 100 spill                           8.25u  0.32s  0.64g
831           MLRISC ra                               5.68u  0.59s  3.93g
832              (spills = 0 reloads = 0)
833           TOTAL                                   14.71u  0.99s  4.81g
834    
835    1500   CPS 100 spill                           58.55u  2.34s  1.74g
836           MLRISC ra                               5.54u  0.65s  3.91g
837              (spills = 543 reloads = 1082)
838           TOTAL                                   65.40u  3.13s  6.00g
839    
840    2000   CPS 100 spill                           126.69u  4.84s  3.08g
841           MLRISC ra                               0.80u  0.10s  0.55g
842              (spills = 42 reloads = 84)
843           TOTAL                                   129.42u  5.10s  4.13g
844    
845    3000   CPS 100 spill                           675.59u  19.03s  11.64g
846           MLRISC ra                               2.69u  0.27s  1.38g
847              (spills = 62 reloads = 124)
848           TOTAL                                   682.48u  19.61s  13.99g
849    
850    4000   CPS 100 spill                           2362.82u  56.28s  43.60g
851           MLRISC ra                               4.96u  0.27s  2.72g
852              (spills = 85 reloads = 170)
853           TOTAL                                   2375.26u  57.21s  48.00g
854    
855    As you can see the old cps spill module suffers from some serious
856    performance problem.  But since I cannot decipher the old code fully,
857    instead of patching the problems up, I'm reimplementing it
858    with a different algorithm.  The new code is more modular,
859    smaller when compiled, and substantially faster
860    (O(n log n) time and O(n) space).  Timing of the new spill module:
861    
862    4000  CPS 100 spill                           0.02u  0.00s  0.00g
863          MLRISC ra                               0.25u  0.02s  0.15g
864             (spills=1 reloads=3)
865          TOTAL                                   7.74u  0.34s  1.62g
866    
867    Implementation details:
868    
869    As far as I can tell, the purpose of the CPS spill module is to make sure the
870    number of live variables at any program point (the bandwidth)
871    does not exceed a certain limit, which is determined by the
872    size of the spill area.
873    
874    When the bandwidth is too large, we decrease the register pressure by
875    packing live variables into spill records.  How we achieve this is
876    completely different than what we did in the old code.
877    
878    First, there is something about the MLRiscGen code generator
879    that we should be aware of:
880    
881    o MLRiscGen performs code motion!
882    
883       In particular, it will move floating point computations and
884       address computations involving only the heap pointer to
885       their use sites (if there is only a single use).
886       What this means is that if we have a CPS record construction
887       statement
888    
889           RECORD(k,vl,w,e)
890    
891       we should never count the new record address w as live if w
892       has only one use (which is often the case).
893    
894       We should do something similar to floating point, but the transformation
895       there is much more complex, so I won't deal with that.
896    
897    Secondly, there are now two new cps primops at our disposal:
898    
899     1. rawrecord of record_kind option
900        This pure operator allocates some uninitialized storage from the heap.
901        There are two forms:
902    
903         rawrecord NONE [INT n]  allocates a tagless record of length n
904         rawrecord (SOME rk) [INT n] allocates a tagged record of length n
905                                     and initializes the tag.
906    
907     2. rawupdate of cty
908          rawupdate cty (v,i,x)
909          Assigns to x to the ith component of record v.
910          The storelist is not updated.
911    
912    We use these new primops for both spilling and increment record construction.
913    
914     1. Spilling.
915    
916        This is implemented with a linear scan algorithm (but generalized
917        to trees).  The algorithm will create a single spill record at the
918        beginning of the cps function and use rawupdate to spill to it,
919        and SELECT or SELp to reload from it.  So both spills and reloads
920        are fine-grain operations.  In contrast, in the old algorithm
921        "spills" have to be bundled together in records.
922    
923        Ideally, we should sink the spill record construction to where
924        it is needed.  We can even split the spill record into multiple ones
925        at the places where they are needed.  But CPS is not a good
926        representation for global code motion, so I'll keep it simple and
927        am not attempting this.
928    
929     2. Incremental record construction (aka record splitting).
930    
931        Long records with many component values which are simulatenously live
932        (recall that single use record addresses are not considered to
933         be live) are constructed with rawrecord and rawupdate.
934        We allocate space on the heap with rawrecord first, then gradually
935        fill it in with rawupdate.  This is the technique suggested to me
936        by Matthias.
937    
938        Some restrictions on when this is applicable:
939        1. It is not a VECTOR record.  The code generator currently does not handle
940           this case. VECTOR record uses double indirection like arrays.
941        2. All the record component values are defined in the same "basic block"
942           as the record constructor.  This is to prevent speculative
943           record construction.
944    
945    ----------------------------------------------------------------------
946    Name: Allen Leung
947    Date: 2002/02/22 01:02:00 EST
948    Tag: leunga-20020222-mlrisc-tools
949    
950    Minor bug fixes in the parser and rewriter
951    
952    ----------------------------------------------------------------------
953    Name: Allen Leung
954    Date: 2002/02/21 20:20:00 EST
955    Tag: leunga-20020221-peephole
956    
957    Regenerated the peephole files.  Some contained typos in the specification
958    and some didn't compile because of pretty printing bugs in the old version
959    of 'nowhere'.
960    
961    ----------------------------------------------------------------------
962    Name: Allen Leung
963    Date: 2002/02/19 20:20:00 EST
964    Tag: leunga-20020219-mlrisc-tools
965    Description:
966    
967       Minor bug fixes to the mlrisc-tools library:
968    
969       1.  Fixed up parsing colon suffixed keywords
970       2.  Added the ability to shut the error messages up
971       3.  Reimplemented the pretty printer and fixed up/improved
972           the pretty printing of handle and -> types.
973       4.  Fixed up generation of literal symbols in the nowhere tool.
974       5.  Added some SML keywords to to sml.sty
975    
976    ----------------------------------------------------------------------
977    Name: Matthias Blume
978    Date: 2002/02/19 16:20:00 EST
979    Tag: blume-20020219-cmffi
980    Description:
981    
982    A wild mix of changes, some minor, some major:
983    
984    * All C FFI-related libraries are now anchored under $c:
985        $/c.cm      --> $c/c.cm
986        $/c-int.cm  --> $c/internals/c-int.cm
987        $/memory.cm --> $c/memory/memory.cm
988    
989    * "make" tool (in CM) now treats its argument pathname slightly
990      differently:
991        1. If the native expansion is an absolute name, then before invoking
992           the "make" command on it, CM will apply OS.Path.mkRelative
993           (with relativeTo = OS.FileSys.getDir()) to it.
994        2. The argument will be passed through to subsequent phases of CM
995           processing without "going native".  In particular, if the argument
996           was an anchored path, then "make" will not lose track of that anchor.
997    
998    * Compiler backends now "know" their respective C calling conventions
999      instead of having to be told about it by ml-nlffigen.  This relieves
1000      ml-nlffigen from one of its burdens.
1001    
1002    * The X86Backend has been split into X86CCallBackend and X86StdCallBackend.
1003    
1004    * Export C_DEBUG and C_Debug from $c/c.cm.
1005    
1006    * C type encoding in ml-nlffi-lib has been improved to model the conceptual
1007      subtyping relationship between incomplete pointers and their complete
1008      counterparts.  For this, ('t, 'c) ptr has been changed to 'o ptr --
1009      with the convention of instantiating 'o with ('t, 'c) obj whenever
1010      the pointer target type is complete.  In the incomplete case, 'o
1011      will be instantiated with some "'c iobj" -- a type obtained by
1012      using one of the functors PointerToIncompleteType or PointerToCompleteType.
1013    
1014      Operations that work on both incomplete and complete pointer types are
1015      typed as taking an 'o ptr while operations that require the target to
1016      be known are typed as taking some ('t, 'c) obj ptr.
1017    
1018      voidptr is now a bit "more concrete", namely "type voidptr = void ptr'"
1019      where void is an eqtype without any values.  This makes it possible
1020      to work on voidptr values using functions meant to operate on light
1021      incomplete pointers.
1022    
1023    * As a result of the above, signature POINTER_TO_INCOMPLETE_TYPE has
1024      been vastly simplified.
1025    
1026    ----------------------------------------------------------------------
1027    Name: Matthias Blume
1028    Date: 2002/02/19 10:48:00 EST
1029    Tag: blume-20020219-pqfix
1030    Description:
1031    
1032    Applied Chris Okasaki's bug fix for priority queues.
1033    
1034    ----------------------------------------------------------------------
1035    Name: Matthias Blume
1036    Date: 2002/02/15 17:05:00
1037    Tag: Release_110_39
1038    Description:
1039    
1040    Last-minute retagging is becoming a tradition... :-(
1041    
1042    This is the working release 110.39.
1043    
1044    ----------------------------------------------------------------------
1045    Name: Matthias Blume
1046    Date: 2002/02/15 16:00:00 EST
1047    Tag: Release_110_39-orig
1048    Description:
1049    
1050    Working release 110.39.  New bootfiles.
1051    
1052    (Update: There was a small bug in the installer so it wouldn't work
1053    with all shells.  So I retagged. -Matthias)
1054    
1055    ----------------------------------------------------------------------
1056    Name: Matthias Blume
1057    Date: 2002/02/15 14:17:00 EST
1058    Tag: blume-20020215-showbindings
1059    Description:
1060    
1061    Added EnvRef.listBoundSymbols and CM.State.showBindings.  Especially
1062    the latter can be useful for exploring what bindings are available at
1063    the interactive prompt.  (The first function returns only the list
1064    of symbols that are really bound, the second prints those but also the
1065    ones that CM's autoloading mechanism knows about.)
1066    
1067    ----------------------------------------------------------------------
1068    Name: Matthias Blume
1069    Date: 2002/02/15 12:08:00 EST
1070    Tag: blume-20020215-iptrs
1071    Description:
1072    
1073    Two improvements to ml-nlffigen:
1074    
1075      1. Write files only if they do not exist or if their current contents
1076         do not coincide with what's being written.  (That is, avoid messing
1077         with the time stamps unless absolutely necessary.)
1078    
1079      2. Implement a "repository" mechanism for generated files related
1080         to "incomplete pointer types".   See the README file for details.
1081    
1082    ----------------------------------------------------------------------
1083    Name: Matthias Blume
1084    Date: 2002/02/14 11:50:00 EST
1085    Tag: blume-20020214-quote
1086    Description:
1087    
1088    Added a type 't t_' to tag.sml (in ml-nlffi-lib.cm).  This is required
1089    because of the new and improved tag generation scheme.  (Thanks to Allen
1090    Leung for pointing it out.)
1091    
1092    ----------------------------------------------------------------------
1093    Name: Lal George
1094    Date: 2002/02/14 09:55:27 EST 2002
1095    Tag: george-20020214-isabelle-bug
1096    Description:
1097    
1098    Fixed the MLRISC bug sent by Markus Wenzel regarding the compilation
1099    of Isabelle on the x86.
1100    
1101    From Allen:
1102    -----------
1103     I've found the problem:
1104    
1105         in ra-core.sml, I use the counter "blocked" to keep track of the
1106         true number of elements in the freeze queue.  When the counter goes
1107         to zero, I skip examining the queue.  But I've messed up the
1108         bookkeeping in combine():
1109    
1110             else ();
1111             case !ucol of
1112               PSEUDO => (if !cntv > 0 then
1113                     (if !cntu > 0 then blocked := !blocked - 1 else ();
1114                                        ^^^^^^^^^^^^^^^^^^^^^^^
1115                      moveu := mergeMoveList(!movev, !moveu)
1116                     )
1117                  else ();
1118    
1119         combine() is called to coalesce two nodes u and v.
1120         I think I was thinking that if the move counts of u and v are both
1121         greater than zero then after they are coalesced then one node is
1122         removed from the freeze queue.  Apparently I was thinking that
1123         both u and v are of low degree, but that's clearly not necessarily true.
1124    
1125    
1126    02/12/2002:
1127        Here's the patch.  HOL now compiles.
1128    
1129        I don't know how this impact on performance (compile
1130        time or runtime).  This bug caused the RA (especially on the x86)
1131        to go thru the potential spill phase when there are still nodes on the
1132        freeze queue.
1133    
1134    
1135    
1136    
1137    ----------------------------------------------------------------------
1138    Name: Matthias Blume
1139    Date: 2002/02/13 22:40:00 EST
1140    Tag: blume-20020213-fptr-rtti
1141    Description:
1142    
1143    Fixed a bug in ml-nlffigen that was introduced with one of the previous
1144    updates.
1145    
1146    ----------------------------------------------------------------------
1147    Name: Matthias Blume
1148    Date: 2002/02/13 16:41:00 EST
1149    Tag: blume-20020213-cmlpq
1150    Description:
1151    
1152    Added new priority queue export symbols (which have just been added to
1153    smlnj-lib.cm) to CML's version of smlnj-lib.cm.  (Otherwise CML would
1154    not compile and the installer would choke.)
1155    
1156    ----------------------------------------------------------------------
1157    Name: Matthias Blume
1158    Date: 2002/02/13 16:15:00 EST
1159    Tag: blume-20020213-various
1160    Description:
1161    
1162    1. More tweaks to ml-nlffigen:
1163    
1164       - better internal datastructures (resulting in slight speedup)
1165       - "-match" option requires exact match
1166       - "localized" gensym counters (untagged structs/unions nested within
1167         other structs/unions or within typedefs get a fresh counter; their
1168         tag will be prefixed by a concatenation of their parents' tags)
1169       - bug fixes (related to calculation of transitive closure of types
1170         to be included in the output)
1171    
1172    2. Minor Basis updates:
1173    
1174       - added implementations for List.collate and Option.app
1175    
1176    ----------------------------------------------------------------------
1177    Name: Matthias Blume
1178    Date: 2002/02/11 15:55:00 EST
1179    Tag: blume-20020211-gensym
1180    Description:
1181    
1182    Added a "-gensym" option to command line of ml-nlffigen.  This can be
1183    used to specify a "stem" -- a string that is inserted in all "gensym'd"
1184    names (ML structure names that correspond to unnamed C structs, unions,
1185    and enums), so that separate runs of ml-nlffigen do not clash.
1186    
1187    ----------------------------------------------------------------------
1188    Name: Matthias Blume
1189    Date: 2002/02/11 12:05:00 EST
1190    Tag: blume-20020211-gensml
1191    Description:
1192    
1193    A quick fix for a problem with GenSML (in the pgraph-util library):
1194    Make generation of toplevel "local" optional.  (Strictly speaking,
1195    signature definitions within "local" are not legal SML.)
1196    
1197    Other than that: updates to INSTALL and cm/TODO.
1198    
1199    ----------------------------------------------------------------------
1200    Name: Matthias Blume
1201    Date: 2002/02/08 15:00:00 EST
1202    Tag: blume-20020208-uniquepid
1203    Description:
1204    
1205    0. Version number has been bumped to 110.38.1.  NEW BOOTFILES!!!
1206    
1207    1. The installer (config/install.sh) has gotten smarter:
1208    
1209         - Configuration options are a bit easier to specify now
1210           (in config/targets).
1211         - Bug in recognizing .tar.bz2 files fixed.
1212         - Installer automatically resolves dependencies between
1213           configuration options (e.g., if you ask for eXene, you will
1214           also get cml -- regardless whether you asked for it or not).
1215         - Installer can run in "quieter mode" by setting the environment
1216           variable INSTALL_QUIETLY to "true".  "Quieter" does not mean
1217           "completely silent", though.
1218         - Build HashCons library as part of smlnj-lib.
1219    
1220    2. A new scheme for assigning persistent identifiers to compilation
1221       units (and, by extension, to types etc.) has been put into place.
1222       This fixes a long-standing bug where types and even dynamic values
1223       can get internally confused, thereby compromising type safety
1224       (abstraction) and dynamic correctness.  See
1225    
1226         http://cm.bell-labs.com/cm/cs/who/blume/pid-confusion.tgz
1227    
1228       for an example of how things could go wrong until now.
1229    
1230       The downside of the new scheme is that pids are not quite as
1231       persistent as they used to be: CM will generate a fresh pid
1232       for every compilation unit that it thinks it sees for the first
1233       time.  That means that if you compile starting from a clean, fresh
1234       source tree at two different times, you end up with different
1235       binaries.
1236    
1237       Cutoff recompilation, however, has not been compromised because
1238       CM keeps pid information in special caches between runs.
1239    
1240    ----------------------------------------------------------------------
1241    Name: Lal George
1242    Date: 2002/02/07 15:34:13 EST 2002
1243    Tag: <none>
1244    Description:
1245    
1246    Compilers that generate assembly code may produce  global labels
1247    whose value is resolved at link time. The various peephole optimization
1248    modules did not take this in account.
1249    
1250    TODO. The Labels.addrOf function should really return an option
1251    type so that clients are forced to deal with this issue, rather
1252    than an exception being raised.
1253    
1254    ----------------------------------------------------------------------
1255    Name: Lal George
1256    Date: 2002/02/06 13:55:02 EST
1257    Tag: george-20020206-ra-breakup
1258    Description:
1259    
1260    1. A bug fix from Allen.
1261    
1262        A typo causes extra fstp %st(0)'s to be generated at compensation
1263        edges, which might cause stack underflow traps at runtime.  This
1264        occurs in fft where there are extraneous fstps right before the 'into'
1265        trap instruction (in this case they are harmless since none of the
1266        integers overflow.)
1267    
1268    2. Pulled out various utility modules that were embedded in the modules
1269       of the register allocator. I need these modules for other purposes, but
1270       they are not complete enough to put into a library (just yet).
1271    ----------------------------------------------------------------------
1272    Name: Matthias Blume
1273    Date: 2002/01/31 16:05:00 EST
1274    Tag: blume-20020131-sparc-ccalls
1275    Description:
1276    
1277    1. C-calls on Sparc needlessly allocated a huge chunk (96 bytes)
1278       of extra stack space by mistake.  Fixed.
1279    
1280    2. Bug in logic of handling of command-line options in ml-nlffigen fixed.
1281    
1282    ----------------------------------------------------------------------
1283    Name: Allen Leung
1284    Date: 2002/01/30
1285    Tag: leunga-20020130-nowhere-bug-fix
1286    Description:
1287    
1288       MLRISC bug fixes:
1289       1. Fixed a bindings computation bug in the 'nowhere' program generator tool.
1290       2. MachineInt.fromString was negating its value.
1291    
1292    ----------------------------------------------------------------------
1293    Name: Matthias Blume
1294    Date: 2002/01/29
1295    Tag: blume-20020129-INSTALL
1296    Description:
1297    
1298    - Added somewhat detailed installation instructions (file INSTALL).
1299    - Fixed curl-detection bug in config/install.sh.
1300    - It is now possible to select the URL getter using the URLGETTER
1301      environment variable:
1302    
1303          not set / "unknown"      --> automatic detection (script tries wget,
1304                                       curl, and lynx)
1305          "wget" / "curl" / "lynx" --> use the specified program (script "knows"
1306                                       how to properly invoke them)
1307          other                    --> use $URLGETTER directly, it must take
1308                                       precisely two command-line arguments
1309                                       (source URL and destination file name)
1310    
1311    ----------------------------------------------------------------------
1312    Name: Matthias Blume
1313    Date: 2002/01/28
1314    Tag: blume-20020128-sparc-ccalls
1315    Description:
1316    
1317    - Fixed problem with calculation of "used" registers in sparc-c-calls.
1318    - Make use of the allocParam argument in sparc-c-calls.
1319    
1320    ----------------------------------------------------------------------
1321    Name: Matthias Blume
1322    Date: 2002/01/28
1323    Tag: blume-20020128-allocParam
1324    Description:
1325    
1326    John Reppy:  Changes c-calls API to accept client-callback for
1327    allocating extra stack space.
1328    me: Corresponding changes to mlriscGen (using a dummy argument that
1329        does not change the current behavior).
1330    
1331    ----------------------------------------------------------------------
1332    Name: Matthias Blume
1333    Date: 2002/01/28 12:00:00
1334    Tag: Release_110_38
1335    Description:
1336    
1337    This time for real!!!
1338    
1339    ----------------------------------------------------------------------
1340    Name: Matthias Blume
1341    Date: 2002/01/28 10:56:00 EST
1342    Tag: blume-20020128-retraction
1343    Description:
1344    
1345    0. Retracted earlier 110.38.  (The Release_110_38 tag has been replaced
1346       with blume-Release_110_38-retracted.)
1347    
1348    1. Fixed a problem with incorrect rounding modes in real64.sml.
1349       (Thanks to Andrew Mccreight <andrew.mccreight@yale.edu>.)
1350    
1351    2. A bug in ml-nlffigen related to the handling of unnamed structs, unions,
1352       and enums fixed.  The naming of corresponding ML identifiers should
1353       now be consistent again.
1354    
1355    ----------------------------------------------------------------------
1356    Name: Allen Leung
1357    Date: 2002/01/27
1358    Tag: leunga-20020127-nowhere
1359    Description:
1360    
1361       Added a target called nowhere in the configuration scripts.
1362       Enabling this will build the MLRISC 'nowhere' tool (for translating
1363       programs with where-clauses into legal SML code) during installation.
1364    
1365    ----------------------------------------------------------------------
1366    Name: Matthias Blume
1367    Date: 2002/01/25 21:27:00 EST
1368    Tag: blume-Release_110_38-retracted
1369    Description:
1370    
1371    Call it a (working) release!  Version is 110.38. Bootfiles are ready.
1372    
1373    README will be added later.
1374    
1375    !!! NOTE:  Re-tagged as blume-Release_110_38-retracted. Original tag
1376    (Release_110_38) removed.  Reason: Last-minute bug fixes.
1377    
1378    ----------------------------------------------------------------------
1379    Name: Matthias Blume
1380    Date: 2002/01/25
1381    Tag: blume-20020125-ffi
1382    Description:
1383    
1384    A large number of tweaks and improvements to ml-nlffi-lib and
1385    ml-nlffigen:
1386    
1387       - ML represenation types have been streamlined
1388       - getter and setter functions work with concrete values, not abstract
1389         ones where possible
1390       - ml-nlffigen command line more flexible (see README file there)
1391       - some bugs have been fixed (hopefully)
1392    
1393    ----------------------------------------------------------------------
1394    Name: Lal George
1395    Date: 2002/01/24
1396    Tag: george-20020124-risc-ra-interface
1397    Description:
1398    
1399       There is a dramatic simplification in the interface to the
1400       register allocator for RISC architectures as a result of making
1401       parallel copy instructions explicit.
1402    
1403    ----------------------------------------------------------------------
1404    Name: Matthias Blume
1405    Date: 2002/01/22
1406    Tag: blume-20020122-x86-ccalls
1407    Description:
1408    
1409    Bug fix for c-calls on x86 (having to do with how char- and
1410    short-arguments are being handled).
1411    
1412    ----------------------------------------------------------------------
1413    Name: Matthias Blume
1414    Date: 2002/01/21
1415    Tag: blume-20020121-ff
1416    Description:
1417    
1418    Another day of fiddling with the FFI...
1419    
1420    1. Bug fix/workaround:  CKIT does not complain about negative array
1421       dimensions, so ml-nlffigen has to guard itself against this possibility.
1422       (Otherwise a negative dimension would send it into an infinite loop.)
1423    
1424    2. Some of the abstract types (light objects, light pointers, most "base"
1425       types) in structure C are now eqtypes.
1426    
1427    3. Added constructors and test functions for NULL function pointers.
1428    
1429    ----------------------------------------------------------------------
1430    Name: Matthias Blume
1431    Date: 2002/01/18
1432    Tag: blume-20020118-ready-for-new-release
1433    Description:
1434    
1435    Made config/srcarchiveurl point to a new place.  (Will provide boot
1436    files shortly.)
1437    
1438    Maybe we christen this to be 110.38?
1439    
1440    ----------------------------------------------------------------------
1441    Name: Matthias Blume
1442    Date: 2002/01/18
1443    Tag: blume-20020118-more-ffifiddle
1444    Description:
1445    
1446    Today's FFI fiddling:
1447    
1448      - Provided a structure CGetSet with "convenient" versions of C.Get.* and
1449        C.Set.* that use concrete (MLRep.*) arguments and results instead
1450        of abstract ones.
1451    
1452      - Provided word-style bit operations etc. for "int" representation
1453        types in MLRep.S<Foo>Bitops where <Foo> ranges over Char, Int, Short,
1454        and Long.
1455    
1456    ----------------------------------------------------------------------
1457    Name: Matthias Blume
1458    Date: 2002/01/18
1459    Tag: blume-20020118-use-x86-fp
1460    Description:
1461    
1462    Now that x86-fast-fp seems to be working, I turned it back on again
1463    by default.  (Seems to work fine now, even with the FFI.)
1464    
1465    Other than that, I added some documentation about the FFI to
1466    src/ml-nlffigen/README and updated the FFI test examples in
1467    src/ml-nlffi-lib/Tests/*.
1468    
1469    ----------------------------------------------------------------------
1470    Name: Allen Leung
1471    Date: 2002/01/17
1472    Tag: leunga-20020117-x86-fast-fp-call
1473    Description:
1474    
1475       1. Fixed a problem with handling return fp values when x86's fast fp
1476          mode is turned on.
1477    
1478       2. Minor pretty printing fix for cellset.  Print %st(0) as %st(0) instead
1479          of %f32.
1480    
1481       3. Added a constructor INT32lit to the ast of MLRISC tools.
1482    
1483    ----------------------------------------------------------------------
1484    Name: Matthias Blume
1485    Date: 2002/01/16
1486    Tag: blume-20020116-ffifiddle
1487    Description:
1488    
1489    More fiddling with the FFI interface:
1490    
1491     - Make constness 'c instead of rw wherever possible.  This eliminates
1492       the need for certain explicit coercions.  (However, due to ML's
1493       value polymorphism, there will still be many cases where explicit
1494       coercions are necessary.  Phantom types are not the whole answer
1495       to modeling a subtyping relationship in ML.)
1496    
1497     - ro/rw coersions for pointers added.  (Avoids the detour through */&.)
1498    
1499     - "printf" test example added to src/ml-nlffi-lib/Tests.  (Demonstrates
1500       clumsy workaround for varargs problem.)
1501    
1502    ----------------------------------------------------------------------
1503    Name: Lal George
1504    Date: 2002/01/15
1505    Tag: <none>
1506    Description:
1507    
1508    1. Since COPY instructions are no longer native to the architecture,
1509       a generic functor can be used to implement the expandCopies function.
1510    
1511    2. Allowed EXPORT and IMPORT pseudo-op declarations to appear inside a
1512       TEXT segment.
1513    
1514    ----------------------------------------------------------------------
1515    Name: Matthias Blume
1516    Date: 2002/01/15
1517    Tag: blume-20020115-ffiupdates
1518    Description:
1519    
1520    1. Fix for bug resulting in single-precision float values being returned
1521       incorrectly from FFI calls.
1522    
1523    2. Small modifications to C FFI API:
1524    
1525        - memory-allocation routines return straight objects (no options)
1526          and raise an exception in out-of-memory situations
1527        - unsafe extensions to cast between function pointers and pointers
1528          from/to ints
1529        - added structure C_Debug as an alternative to structure C where
1530          pointer-dereferencing (|*| and |*!) always check for null-pointers
1531        - added open_lib' to DynLinkage;  open_lib' works like open_lib
1532          but also takes a (possibly empty) list of existing library handles
1533          that the current library depends on
1534    
1535    ----------------------------------------------------------------------
1536    Name: Matthias Blume
1537    Date: 2002/01/10
1538    Tag: blume-20020110-newffigen
1539    Description:
1540    
1541    1. Updates to portable graph code.
1542    
1543    2. Major update to ml-nlffigen and ml-nlffi-lib.  Things are much
1544       more scalable now so that even huge interfaces such as the one
1545       for GTK compile in finite time and space. :-)
1546       See src/ml-nlffigen/README for details on what's new.
1547    
1548    ----------------------------------------------------------------------
1549    Name: Lal George
1550    Date: 2001/01/09 14:31:35 EST 2002
1551    Tag: george-20011206-rm-native-copy
1552    Description:
1553    
1554            Removed the native COPY and FCOPY instructions
1555            from all the architectures and replaced it with the
1556            explicit COPY instruction from the previous commit.
1557    
1558            It is now possible to simplify many of the optimizations
1559            modules that manipulate copies. This has not been
1560            done in this change.
1561    
1562    ----------------------------------------------------------------------
1563    Name: Lal George
1564    Date: 2001/12/06 16:50:13 EST 2001
1565    Tag: george-20011206-mlrisc-instruction
1566    Description:
1567    
1568    Changed the representation of instructions from being fully abstract
1569    to being partially concrete. That is to say:
1570    
1571      from
1572            type instruction
1573    
1574      to
1575            type instr                              (* machine instruction *)
1576    
1577            datatype instruction =
1578                LIVE of {regs: C.cellset, spilled: C.cellset}
1579              | KILL of {regs: C.cellset, spilled: C.cellset}
1580              | COPYXXX of {k: CB.cellkind, dst: CB.cell list, src: CB.cell list}
1581              | ANNOTATION of {i: instruction, a: Annotations.annotation}
1582              | INSTR of instr
1583    
1584    This makes the handling of certain special instructions that appear on
1585    all architectures easier and uniform.
1586    
1587    LIVE and KILL say that a list of registers are live or killed at the
1588    program point where they appear. No spill code is generated when an
1589    element of the 'regs' field is spilled, but the register is moved to
1590    the 'spilled' (which is present, more for debugging than anything else).
1591    
1592    LIVE replaces the (now deprecated) DEFFREG instruction on the alpha.
1593    We used to generate:
1594    
1595            DEFFREG f1
1596            f1 := f2 + f3
1597            trapb
1598    
1599    but now generate:
1600    
1601            f1 := f2 + f3
1602            trapb
1603            LIVE {regs=[f1,f2,f3], spilled=[]}
1604    
1605    Furthermore, the DEFFREG (hack) required that all floating point instruction
1606    use all registers mentioned in the instruction. Therefore f1 := f2 + f3,
1607    defines f1 and uses [f1,f2,f3]! This hack is no longer required resulting
1608    in a cleaner alpha implementation. (Hopefully, intel will not get rid of
1609    this architecture).
1610    
1611    COPYXXX is intended to replace the parallel COPY and FCOPY  available on
1612    all the architectures. This will result in further simplification of the
1613    register allocator that must be aware of them for coalescing purposes, and
1614    will also simplify certain aspects of the machine description that provides
1615    callbacks related to parallel copies.
1616    
1617    ANNOTATION should be obvious, and now INSTR represents the honest to God
1618    machine instruction set!
1619    
1620    The <arch>/instructions/<arch>Instr.sml files define certain utility
1621    functions for making porting easier -- essentially converting upper case
1622    to lower case. All machine instructions (of type instr) are in upper case,
1623    and the lower case form generates an MLRISC instruction. For example on
1624    the alpha we have:
1625    
1626      datatype instr =
1627         LDA of {r:cell, b:cell, d:operand}
1628       | ...
1629    
1630      val lda : {r:cell, b:cell, d:operand} -> instruction
1631        ...
1632    
1633    where lda is just (INSTR o LDA), etc.
1634    
1635    ----------------------------------------------------------------------
1636    Name: Matthias Blume
1637    Date: 2001/11/22 21:40:00 EST
1638    Tag: Release_110_37
1639    Description:
1640    
1641    Release 110.37.  This time for real.
1642    
1643    ----------------------------------------------------------------------
1644    Name: Matthias Blume
1645    Date: 2001/11/21 16:35:00 EST
1646    Tag: blume-20011121-foot-in-mouth
1647    Description:
1648    
1649    Removed the "Release_110_37" tag because of a serious bug.
1650    This will be re-tagged once the bug is fixed.
1651    
1652    ----------------------------------------------------------------------
1653    Name: Matthias Blume
1654    Date: 2001/11/21 16:14:00 EST
1655    Tag: blume-20011121-forgottenfile
1656    Description:
1657    
1658    Forgot to add a file.  (Just a .tex-file -- part of
1659    the CM manual source.)
1660    
1661    ----------------------------------------------------------------------
1662    Name: Matthias Blume
1663    Date: 2001/11/21 16:10:00 EST
1664    Tag: blume-20011121-invalid_110_37
1665    Description:
1666    
1667    Note: I removed the original tag "Release_110_37" from this commit
1668          because we found a serious bug in all non-x86 backends.
1669          - Matthias
1670    
1671    1. Modifications to the SML/NJ code generator and to the runtime system
1672       so that code object name strings are directly inserted into code
1673       objects at code generation time.  The only business the runtime system
1674       has with this is now to read the name strings on occasions.
1675       (The encoding of the name string has also changed somewhat.)
1676    
1677    2. CM now implements a simple "set calculus" for specifying export lists.
1678       In particular, it is now possible to refer to the export lists of
1679       other libraries/groups/sources and form unions as well as differences.
1680       See the latest CM manual for details.
1681    
1682    3. An separate notion of "proxy" libraries has again be eliminated from
1683       CM's model.  (Proxy libraries are now simply a special case of using
1684       the export list calculus.)
1685    
1686    4. Some of the existing libraries now take advantage of the new set
1687       calculus.
1688       (Notice that not all libraries have been converted because some
1689       of the existing .cm-files are supposed to be backward compatible
1690       with 110.0.x.)
1691    
1692    5. Some cleanup in stand-alone programs.  (Don't use "exnMessage" -- use
1693       "General.exnMessage"!  The former relies on a certain hook to be
1694       initialized, and that often does not happen in the stand-alone case.)
1695    
1696    ----------------------------------------------------------------------
1697    Name: Lal George
1698    Date: 2001/11/21  13:56:18 EST
1699    Tag: george-2001121-pseudo-ops
1700    Description:
1701    
1702      Implemented a complete redesign of MLRISC pseudo-ops. Now there
1703      ought to never be any question of incompatabilities with
1704      pseudo-op syntax expected by host assemblers.
1705    
1706      For now, only modules supporting GAS syntax are implemented
1707      but more should follow, such as MASM, and vendor assembler
1708      syntax, e.g. IBM as, Sun as, etc.
1709    
1710    ----------------------------------------------------------------------
1711    Name: Matthias Blume
1712    Date: 2001/11/14 11:52:00 EST
1713    Tag: blume-20011114-srcname
1714    Description:
1715    
1716    1. Routed the name of the current source file to mlriscgen where it
1717       should be directly emitted into the code object.  (This last part
1718       is yet to be done.)
1719    
1720    2. Some cleanup of the pgraph code to make it match the proposal that
1721       I put out the other day.  (The proposal notwithstanding, things are
1722       still in flux here.)
1723    
1724    ----------------------------------------------------------------------
1725    Name: Lal George
1726    Date: 2001/11/14 09:44:04 EST
1727    Tag:
1728    Description:
1729    
1730      Fix for a backpatching bug reported by Allen.
1731    
1732      Because the boundary between short and long span-dependent
1733      instructions is +/- 128, there are an astounding number of
1734      span-dependent instructions whose size is over estimated.
1735    
1736      Allen came up with the idea of letting the size of span
1737      dependent instructions be non-monotonic, for a maxIter
1738      number of times, after which the size must be monotonically
1739      increasing.
1740    
1741      This table shows the number of span-dependent instructions
1742      whose size was over-estimated as a function of maxIter, for the
1743      file Parse/parse/ml.grm.sml:
1744    
1745         maxIter            # of instructions:
1746            10                      687
1747            20                      438
1748            30                      198
1749            40                        0
1750    
1751      In compiling the compiler, there is no significant difference in
1752      compilation speed between maxIter=10 and maxIter=40. Actually,
1753      my measurements showed that maxIter=40 was a tad faster than
1754      maxIter=10! Also 96% of the  files in the compiler reach a fix
1755      point within 13 iterations, so fixing maxIter at 40, while high,
1756      is okay.
1757    
1758    ----------------------------------------------------------------------
1759    Name: Matthias Blume
1760    Date: 2001/10/31 15:25:00 EST
1761    Tag: blume-20011031-pgraph
1762    Description:
1763    
1764    CKIT:
1765    * Changed the "Function" constructor of type Ast.ctype to carry optional
1766      argument identifiers.
1767    * Changed the return type of TypeUtil.getFunction accordingly.
1768    * Type equality ignores the argument names.
1769    * TypeUtil.composite tries to preserve argument names but gives up quickly
1770      if there is a mismatch.
1771    
1772    installation script:
1773    * attempts to use "curl" if available (unless "wget" is available as well)
1774    
1775    CM:
1776    * has an experimental implementation of "portable graphs" which I will
1777      soon propose as an implementation-independent library format
1778    * there are also new libraries $/pgraph.cm and $/pgraph-util.cm
1779    
1780    NLFFI-LIB:
1781    * some cleanup (all cosmetic)
1782    
1783    NLFFIGEN:
1784    * temporarily disabled the mechanism that suppresses ML output for
1785      C definitions whose identifiers start with an underscore character
1786    * generate val bindings for enum constants
1787    * user can request that only one style (light or heavy) is being used;
1788      default is to use both (command-line arguments: -heavy and -light)
1789    * fixed bug in handling of function types involving incomplete pointers
1790    * generate ML entry points that take record arguments (i.e., using
1791      named arguments) for C functions that have a prototype with named
1792      arguments
1793      (see changes to CKIT)
1794    
1795    ----------------------------------------------------------------------
1796    Name: Allen Leung
1797    Date: 2001/10/27 20:34:00 EDT
1798    Tag: leunga-20011027-x86-fast-fp-call
1799    Description:
1800    
1801       Fixed the bug described in blume-20010920-slowfp.
1802    
1803       The fix involves
1804          1. generating FCOPYs in FSTP in ia32-svid
1805          2. marking a CALL with the appropriate annotation
1806    
1807    ----------------------------------------------------------------------
1808    Name: Matthias Blume
1809    Date: 2001/10/16 11:32:00 EDT
1810    Tag: blume-20011016-netbsd
1811    Description:
1812    
1813    Underscore patch from Chris Richards (fixing problem with compiling
1814    runtime system under recent NetBSD).
1815    
1816    ----------------------------------------------------------------------
1817    Name: Allen Leung
1818    Date: 2001/10/12 17:18:32 EDT 2001
1819    Tag: leung-20011012-x86-printflowgraph
1820    Description:
1821    
1822    X86RA now uses a valid (instead of dummy) PrintFlowgraph module.
1823    
1824    ----------------------------------------------------------------------
1825    Name: Lal George
1826    Date: 2001/10/11 23:51:34 EDT
1827    Tag: george-20011011-too-many-instrs
1828    Description:
1829    
1830    The representation of a program point never expected to see more
1831    than 65536 instructions in a basic block!
1832    
1833    ----------------------------------------------------------------------
1834    Name: Lal George
1835    Date: 2001/10/09 09:41:37 EDT
1836    Tag: george-20011008-mlrisc-labels
1837    Description:
1838    
1839    Changed the machine description files to support printing of
1840    local and global labels in assembly code, based on host assembler
1841    conventions.
1842    
1843    ----------------------------------------------------------------------
1844    Name: Matthias Blume
1845    Date: 2001/09/25 15:25:00 EDT
1846    Tag: blume-20010925-exninfo
1847    Description:
1848    
1849    I provided a non-hook implementation of exnName (at the toplevel) and
1850    made the "dummy" implementation of exnMessage (at the toplevel) more
1851    useful: if nothing gets "hooked in", then at least you are going to
1852    see the exception name and a message indicating why you don't see more.
1853    
1854    [For the time being, programs that need exnMessage and want to use
1855    ml-build should either use General.exnMessage (strongly recommended) or
1856    refer to structure General at some other point so that CM sees a
1857    static dependency.]
1858    
1859    [Similar remarks go for "print" and "use":  If you want to use their
1860    functionality in stand-alone programs generated by ml-build, then use
1861    TextIO.output and Backend.Interact.useFile (from $smlnj/compiler.cm).]
1862    
1863    ----------------------------------------------------------------------
1864    Name: Matthias Blume
1865    Date: 2001/09/20 17:28:00 EDT
1866    Tag: blume-20010920-slowfp
1867    Description:
1868    
1869    Allen says that x86-fast-fp is not safe yet, so I turned it off again...
1870    
1871    ----------------------------------------------------------------------
1872    Name: Matthias Blume
1873    Date: 2001/09/20 17:20:00 EDT
1874    Tag: blume-20010920-canonicalpaths
1875    Description:
1876    
1877    0. Updated the BOOT file (something that I forgot to do earlier).
1878    
1879    1. Small internal change to CM so that it avoids "/../" in filenames
1880       as much as possible (but only where it is safe).
1881    
1882    2. Changed config/_run-sml (resulting in a changed bin/.run-sml) so
1883       that arguments that contain delimiters are passed through correctly.
1884       This change also means that all "special" arguments of the form
1885       @SMLxxx... must come first.
1886    
1887    3. Changed install script to put relative anchor names for tool commands
1888       into pathconfig.
1889    
1890    ----------------------------------------------------------------------
1891    Name: Matthias Blume
1892    Date: 2001/09/18 15:35:00 EDT
1893    Tag: blume-20010918-readme11036
1894    Description:
1895    
1896    Added README files.
1897    
1898    ----------------------------------------------------------------------
1899    Name: Matthias Blume
1900    Date: 2001/09/18 11:45:00 EDT
1901    Tag: Release_110_36 (retag)
1902    Description:
1903    
1904    Fixed mistake in config/preloads. Retagged as 110.36.
1905    
1906    ----------------------------------------------------------------------
1907    Name: Matthias Blume
1908    Date: 2001/09/18 09:40:00 EDT
1909    Tag: Release_110_36_orig (tag changed)
1910    Description:
1911    
1912    New version (110.36).  New bootfiles.
1913    
1914    ----------------------------------------------------------------------
1915    Name: Matthias Blume
1916    Date: 2001/09/14 16:15:00 EDT
1917    Tag: blume-20010914-x86fastfp
1918    Description:
1919    
1920    John committed some changes that Allen made, in particular a (hopefully)
1921    correctly working version of the x86-fp module.
1922    
1923    I changed the default setting of the Control.MLRISC.getFlag "x86-fast-fp"
1924    flag to "true".  Everything seems to compile to a fixpoint ok, and
1925    "mandelbrot" speeds up by about 15%.
1926    
1927    ----------------------------------------------------------------------
1928    Name: Matthias Blume
1929    Date: 2001/09/13 11:20:00 EDT
1930    Tag: blume-20010913-minimal
1931    Description:
1932    
1933    1. Stefan Monnier's patch to fix a miscompilation problem that
1934       was brought to light by John Reppy's work on Moby.
1935    
1936    2. Implemented a minimal "structure Compiler" that contains just
1937       "version" and "architecture".  The minimal version will be
1938       available when the full version is not.  This is for backward-
1939       compatibility with code that wants to test Compiler.version.
1940    
1941    ----------------------------------------------------------------------
1942    Name: Matthias Blume
1943    Date: 2001/08/28 14:03:00 EDT
1944    Tag: blume-20010828-ml-lex
1945    Description:
1946    
1947    Fix for bug 1581, received from Neophytos Michael.
1948    
1949    ----------------------------------------------------------------------
1950    Name: Matthias Blume
1951    Date: 2001/08/27 11:20:00 EDT
1952    Tag: blume-20010827-readme11035
1953    Description:
1954    
1955    Fleshed out the README file for 110.35.
1956    
1957    ----------------------------------------------------------------------
1958    Name: Matthias Blume
1959    Date: 2001/08/24 17:10:00 EDT
1960    Tag: Release_110_35
1961    Description:
1962    
1963    New version number (110.35).  New bootfiles.
1964    
1965    ----------------------------------------------------------------------
1966    Name: Lal George
1967    Date: 2001/08/24 13:47:18 EDT 2001
1968    Tag: george-20010824-MLRISC-graphs
1969    Description:
1970    
1971     removed clusters from MLRISC completely and replaced with graphs.
1972    
1973    ----------------------------------------------------------------------
1974    Name: Matthias Blume
1975    Date: 2001/08/23 17:50:00 EDT
1976    Tag: blume-20010823-toplevel
1977    Description:
1978    
1979    - some reorganization of the code that implements various kinds of
1980      environments in the compiler (static, dynamic, symbolic, combined)
1981    - re-implemented the EnvRef module so that evalStream works properly
1982      (if the stream contains references to "use", "CM.make", etc.)
1983    - cleaned up evalloop.sml and interact.sml (but they need more cleaning)
1984    
1985    ----------------------------------------------------------------------
1986    Name: Matthias Blume
1987    Date: 2001/08/20 15:50 EDT
1988    Tag: blume20010820-slipup
1989    Description:
1990    
1991    I forgot to commit a few files.  Here they are...
1992    
1993    ----------------------------------------------------------------------
1994    Name: Matthias Blume
1995    Date: 2001/08/20 15:35:00 EDT
1996    Tag: blume-20010820-debugprof
1997    Description:
1998    
1999    !!!! NEW BOOTFILES !!!!
2000    
2001    This is another round of reorganizing the compiler sources.  This
2002    time the main goal was to factor out all the "instrumentation"
2003    passes (for profiling and backtracing) into their own library.
2004    The difficulty was to do it in such a way that it does not depend
2005    on elaborate.cm but only on elabdata.cm.
2006    
2007    Therefore there have been further changes to both elaborate.cm and
2008    elabdata.cm -- more "generic" things have been moved from the former
2009    to the latter.  As a result, I was forced to split the assignment
2010    of numbers indicating "primtyc"s into two portions: SML-generic and
2011    SML/NJ-specific.  Since it would have been awkward to maintain,
2012    I bit the bullet and actually _changed_ the mapping between these
2013    numbers and primtycs.  The bottom line of this is that you need
2014    a new set of bin- and bootfiles.
2015    
2016    I have built new bootfiles for all architectures, so doing a fresh
2017    checkout and config/install.sh should be all you need.
2018    
2019    The newly created library's name is
2020    
2021        $smlnj/viscomp/debugprof.cm
2022    
2023    and its sources live under
2024    
2025        src/compiler/DebugProf
2026    
2027    ----------------------------------------------------------------------
2028    Name: Matthias Blume
2029    Date: 2001/08/15 17:15:00 EDT
2030    Tag: blume-20010815-compreorg
2031    Description:
2032    
2033    This is a first cut at reorganizing the CM libraries that make up the
2034    core of the compiler.  The idea is to separate out pieces that could
2035    be used independently by tools, e.g., the parser, the typechecker, etc.
2036    
2037    The current status is a step in this direction, but it is not quite
2038    satisfactory yet.  Expect more changes in the future.
2039    
2040    Here is the current (new) organization...
2041    
2042        What used to be $smlnj/viscomp/core.cm is now divided into
2043        six CM libraries:
2044    
2045             $smlnj/viscomp/basics.cm
2046                           /parser.cm
2047                           /elabdata.cm
2048                           /elaborate.cm
2049                           /execute.cm
2050                           /core.cm
2051    
2052        The CM files for these libraries live under src/system/smlnj/viscomp.
2053        All these libraries are proxy libraries that contain precisely
2054        one CM library component.  Here are the locations of the components
2055        (all within the src/compiler tree):
2056    
2057             Basics/basics.cm
2058             Parse/parser.cm
2059             ElabData/elabdata.cm
2060             Elaborator/elaborate.cm
2061             Execution/execute.cm
2062             core.cm
2063    
2064         [This organization is the same that has been used already
2065         for a while for the architecture-specific parts of the visible
2066         compiler and for the old version of core.cm.]
2067    
2068         As you will notice, many source files have been moved from their
2069         respective original locations to a new home in one of the above
2070         subtrees.
2071    
2072         The division of labor between the new libraries is the following:
2073    
2074             basics.cm:
2075                - Simple, basic definitions that pertain to many (or all) of
2076                  the other libraries.
2077             parser.cm:
2078                - The SML parser, producing output of type Ast.dec.
2079                - The type family for Ast is also defined and exported here.
2080             elabdata.cm:
2081                - The datatypes that describe input and output of the elaborator.
2082                  This includes types, absyn, and static environments.
2083             elaborator.cm:
2084                - The SML/NJ type checker and elaborator.
2085                  This maps an Ast.dec (with a given static environment) to
2086                  an Absyn.dec (with a new static environment).
2087                - This libraries implements certain modules that used to be
2088                  structures as functors (to remove dependencies on FLINT).
2089             execute.cm:
2090                - Everything having to do with executing binary code objects.
2091                - Dynamic environments.
2092             core.cm:
2093                - SML/NJ-specific instantiations of the elaborator and MLRISC.
2094                - Top-level modules.
2095                - FLINT (this should eventually become its own library)
2096    
2097    Notes:
2098    
2099    I am not 100% happy with the way I separated the elaborator (and its
2100    data structures) from FLINT.  Two instances of the same problem:
2101    
2102        1. Data structures contain certain fields that carry FLINT-specific
2103           information.  I hacked around this using exn and the property list
2104           module from smlnj-lib.  But the fact that there are middle-end
2105           specific fields around at all is a bit annoying.
2106    
2107        2. The elaborator calculates certain FLINT-related information.  I tried
2108           to make this as abstract as I could using functorization, but, again,
2109           the fact that the elaborator has to perform calculations on behalf
2110           of the middle-end at all is not nice.
2111    
2112        3. Having to used exn and property lists is unfortunate because it
2113           weakens type checking.  The other alternative (parameterizing
2114           nearly *everything*) is not appealing, though.
2115    
2116    I removed the "rebinding =" warning hack because due to the new organization
2117    it was awkward to maintain it.  As a result, the compiler now issues some of
2118    these warnings when compiling init.cmi during bootstrap compilation. On
2119    the plus side, you also get a warning when you do, for example:
2120       val op = = Int32.+
2121    which was not the case up to now.
2122    
2123    I placed "assign" and "deref" into the _Core structure so that the
2124    code that deals with the "lazy" keyword can find them there.  This
2125    removes the need for having access to the primitive environment
2126    during elaboration.
2127    
2128    ----------------------------------------------------------------------
2129    Name: Matthias Blume
2130    Date: 2001/08/13
2131    Tag: blume-20010813-closures
2132    Description:
2133    
2134    This fix was sent to us by Zhong Shao.  It is supposed to improve the
2135    performance of certain loops by avoiding needless closure allocation.
2136    
2137    ----------------------------------------------------------------------
2138    Name: Lal George
2139    Date: 2001/07/31 10:03:23 EDT 2001
2140    Tag: george-20010731-x86-fmalloc
2141    Description: Fixed bug in x86 calls
2142    
2143        There was a bug where call instructions would mysteriously
2144        vanish. The call instruction had to be one that returned
2145        a floating point value.
2146    
2147    ----------------------------------------------------------------------
2148    Name: Lal George
2149    Date: 2001/07/19 16:36:29 EDT 2001
2150    Tag: george-20010719-simple-cells
2151    Description:
2152    
2153    I have dramatically simplified the interface for CELLS in MLRISC.
2154    
2155    In summary, the cells interface is broken up into three parts:
2156    
2157      1. CellsBasis : CELLS_BASIS
2158    
2159            CellsBasis is a top level structure and common for all
2160            architectures.  it contains the definitions of basic datatypes
2161            and utility  functions over these types.
2162    
2163      2. functor Cells() : CELLS
2164    
2165            Cells generates an interface for CELLS that incorporates the
2166            specific resources on the target architecture, such as the
2167            presence of special register classes, their number and size,
2168            and various useful substructures.
2169    
2170      3. <ARCH>CELLS
2171    
2172            e.g. SparcCells: SPARCCELLS
2173    
2174            <ARCH>CELLS usually contains additional bindings for special
2175            registers  on the architecture, such as:
2176    
2177                    val r0 : cell           (* register zero *)
2178                    val y : cell            (* Y register *)
2179                    val psr : cell          (* processor status register *)
2180                    ...
2181    
2182            The structure returned by applying the Cells functor is opened
2183            in this interface.
2184    
2185    The main implication of all this is that the datatypes for cells is
2186    split between CellsBasis and CELLS -- a fairly simple change for user
2187    code.
2188    
2189    In the old scheme the CELLS interface had a definitional binding of
2190    the form:
2191    
2192            signature CELLS = sig
2193    
2194               structure CellsBasis = CellsBasis
2195    
2196               ...
2197    
2198            end
2199    
2200    With all the sharing constraints that goes on in MLRISC, this old
2201    design  quickly leads to errors such as:
2202    
2203            "structure definition spec inside of sharing ... "
2204    
2205    
2206    and appears to require an unacceptable amount of sharing and where
2207    constraint hackery.
2208    
2209    I think this error message (the interaction of definitional specs and
2210    sharing) requires more explanation on our web page.
2211    
2212    ----------------------------------------------------------------------
2213    Name: Matthias Blume
2214    Date: 2001/07/19 15:00:00 EDT
2215    Tag: blume-20010719-libreorg
2216    Description:
2217    
2218    This update puts together a fairly extensive but straightforward change
2219    to the way the libraries that implement the interactive system are
2220    organized:
2221    
2222       The biggest change is the elimination of structure Compiler.  As a
2223       replacement for this structure, there is now a CM library
2224       (known as $smlnj/compiler.cm or $smlnj/compiler/current.cm)
2225       that exports all the substructures of the original structure Compiler
2226       directly.  So instead of saying Compiler.Foo.bar one now simply
2227       says Foo.bar.  (The CM libraries actually export a collection of
2228       structures that is richer than the collection of substructures of
2229       structure Compiler.)
2230    
2231       To make the transition smooth, there is a separate library called
2232       $smlnj/compiler/compiler.cm which puts together and exports the
2233       original structure Compiler (or at least something very close to it).
2234    
2235       There are five members of the original structure Compiler
2236       that are not exported directly but which instead became members
2237       of a new structure Backend (described by signature BACKEND).  These are:
2238       structure Profile (: PROFILE), structure Compile (: COMPILE), structure
2239       Interact (: INTERACT), structure Machine (: MACHINE), and val
2240       architecture (: string).
2241    
2242       Structure Compiler.Version has become structure CompilerVersion.
2243    
2244       Cross-compilers for alpha32, hppa, ppc, sparc, and x86 are provided
2245       by $smlnj/compiler/<arch>.cm where <arch> is alpha32, hppa, ppc, sparc,
2246       or x86, respectively.
2247       Each of these exports the same frontend structures that
2248       $smlnj/compiler.cm exports.  But they do not have a structure Backend
2249       and instead export some structure <Arch>Backend where <Arch> is Alpha32,
2250       Hppa, PPC, Sparc, or X86, respectively.
2251    
2252       Library $smlnj/compiler/all.cm exports the union of the exports of
2253       $smlnj/compiler/<arch>.cm
2254    
2255       There are no structures <Arch>Compiler anymore, use
2256       $smlnj/compiler/<arch>.cm instead.
2257    
2258       Library host-compiler-0.cm is gone.  Instead, the internal library
2259       that instantiates CM is now called cm0.cm.  Selection of the host
2260       compiler (backend) is no longer done here but. (Responsibility for it
2261       now lies with $smlnj/compiler/current.cm.  This seems to be more
2262       logical.)
2263    
2264       Many individual files have been moved or renamed.  Some files have
2265       been split into multiple files, and some "dead" files have been deleted.
2266    
2267    Aside from these changes to library organization, there are also changes
2268    to the way the code itself is organized:
2269    
2270       Structure Binfile has been re-implemented in such a way that it no
2271       longer needs any knowledge of the compiler.  It exclusively deals
2272       with the details of binfile layout.  It no longer invokes the
2273       compiler (for the purpose of creating new prospective binfile
2274       content), and it no longer has any knowledge of how to interpret
2275       pickles.
2276    
2277       Structure Compile (: COMPILE) has been stripped down to the bare
2278       essentials of compilation.  It no longer deals with linking/execution.
2279       The interface has been cleaned up considerably.
2280    
2281       Utility routines for dealing with linking and execution have been
2282       moved into their own substructures.
2283    
2284       (The ultimate goal of these changes is to provide a light-weight
2285       binfile loader/linker (at least for, e.g., stable libraries) that
2286       does not require CM or the compiler to be present.)
2287    
2288    CM documentation has been updated to reflect the changes to library
2289    organization.
2290    
2291    ----------------------------------------------------------------------
2292    Name: Matthias Blume
2293    Date: 2001/07/10 17:30:00 EDT
2294    Tag: Release_110_34
2295    Description:
2296    
2297    Minor tweak to 110.34 (re-tagged):
2298    
2299      - README.html file added to CVS repository
2300      - runtime compiles properly under FreeBSD 3.X and 4.X
2301    
2302    ----------------------------------------------------------------------
2303    Name: Matthias Blume
2304    Date: 2001/07/10 17:30:00 EDT
2305    Tag: Release_110_34
2306    Description:
2307    
2308    New version number (110.34). New bootfiles.
2309    
2310    ----------------------------------------------------------------------
2311    Name: Matthias Blume
2312    Date: 2001/07/09 16:00:00 EDT
2313    Tag: blume-20010709-more-varargs
2314    Description:
2315    
2316    I changed the handling of varargs in ml-nlffigen again:
2317    The ellipsis ... will now simply be ignored (with an accompanying warning).
2318    
2319    The immediate effect is that you can actually call a varargs function
2320    from ML -- but you can't actually supply any arguments beyond the ones
2321    specified explicitly.  (For example, you can call printf with its format
2322    string, but you cannot pass additional arguments.)
2323    
2324    This behavior is only marginally more useful than the one before, but
2325    it has the advantage that a function or, more importantly, a function
2326    type never gets dropped on the floor, thus avoiding follow-up problems with
2327    other types that refer to the offending one.
2328    
2329    ----------------------------------------------------------------------
2330    Name: Matthias Blume
2331    Date: 2001/07/09 11:25:00 EDT
2332    Tag: blume-20010709-varargs
2333    Description:
2334    
2335    1. ckit-lib.cm now exports structure Error
2336    2. ml-nlffigen reports occurences of "..." (i.e., varargs function types)
2337       with a warning accompanied by a source location.  Moreover, it
2338       merely skips the offending function or type and proceeds with the
2339       rest of its work.u  As a result, one can safely feed C code containing
2340       "..." to ml-nlffigen.
2341    3. There are some internal improvements to CM, providing slightly
2342       more general string substitutions in the tools subsystem.
2343    
2344    ----------------------------------------------------------------------
2345    Name: Matthias Blume
2346    Date: 2001/06/27 15:10:00 EDT
2347    Tag: blume-20010627-concur
2348    Description:
2349    
2350    Fixed a small bug in CM's handling of parallel compilation.
2351    (You could observe the bug by Control-C-interrupting an ordinary
2352    CMB.make or CM.stabilize and then attaching some compile servers.
2353    The result was that all of a sudden the previously interrupted
2354    compilation would continue on its own.  This was because of
2355    an over-optimization: CM did not bother to clean out certain queues
2356    when no servers were attached "anyway", resulting in the contents
2357    of these queues to grab control when new servers did get attached.)
2358    
2359    There is also another minor update to the CM manual.
2360    
2361    ----------------------------------------------------------------------
2362    Name: Matthias Blume
2363    Date: 2001/06/26 16:15:00 EDT
2364    Tag: blume-20010626-cmdoc
2365    Description:
2366    
2367    Minor typo fixed in CM manual (syntax diagram for libraries).
2368    
2369    ----------------------------------------------------------------------
2370    Name: Matthias Blume
2371    Date: 2001/06/25 22:55:00 EDT
2372    Tag: blume-20010625-x86pc
2373    Description:
2374    
2375    Fixed a nasty bug in the X86 assembly code that caused signal
2376    handlers to fail (crash) randomly.
2377    
2378    ----------------------------------------------------------------------
2379    Name: Matthias Blume
2380    Date: 2001/06/25 12:05:00 EDT
2381    Tag: blume-20010625-nlffigen
2382    Description:
2383    
2384    This update fixes a number of minor bugs in ml-nlffigen as reported by
2385    Nick Carter <nbc@andrew.cmu.edu>.
2386    
2387      1. Silly but ok typedefs of the form "typedef void myvoid;" are now accepted.
2388      2. Default names for generated files are now derived from the name of
2389         the C file *without its directory*.  In particular, this causes generated
2390         files to be placed locally even if the C file is in some system directory.
2391      3. Default names for generated signatures and structures are also derived
2392         from the C file name without its directory.  This avoids silly things
2393         like "structure GL/GL".
2394         (Other silly names are still possible because ml-nlffigen does not do
2395          a thorough check of whether generated names are legal ML identifiers.
2396          When in doubt, use command line arguments to force particular names.)
2397    
2398    ----------------------------------------------------------------------
2399    Name: Matthias Blume
2400    Date: 2001/06/21 12:25:00 EDT
2401    Tag: blume-20010621-eXene
2402    Description:
2403    
2404    eXene now compiles and (sort of) works again.
2405    
2406    The library name (for version > 110.33) is $/eXene.cm.
2407    
2408    I also added an new example in src/eXene/examples/nbody.  See the
2409    README file there for details.
2410    
2411    ----------------------------------------------------------------------
2412    Name: Matthias Blume
2413    Date: 2001/06/20 16:40:00 EDT
2414    Tag: blume-20010620-cml
2415    Description:
2416    
2417    CML now compiles and works again.
2418    
2419    Libraries (for version > 110.33):
2420    
2421      $cml/cml.cm            Main CML library.
2422      $cml/basis.cm          CML's version of $/basis.cm.
2423      $cml/cml-internal.cm   Internal helper library.
2424      $cml/core-cml.cm       Internal helper library.
2425      $cml-lib/trace-cml.cm  Tracing facility.
2426      $cml-lib/smlnj-lib.cm  CML's version of $/smlnj-lib.cm
2427    
2428    The installer (config/install.sh) has been taught how to properly
2429    install this stuff.
2430    
2431    ----------------------------------------------------------------------
2432    Name: Matthias Blume
2433    Date: 2001/06/19 17:55:00 EDT
2434    Tag: blume-20010619-instantiate
2435    Description:
2436    
2437    This un-breaks the fix for bug 1432.
2438    (The bug was originally fixed in 110.9 but I broke it again some
2439    time after that.)
2440    
2441    ----------------------------------------------------------------------
2442    Name: Matthias Blume
2443    Date: 2001/06/19 17:25:00 EDT
2444    Tag: blume-20010619-signals
2445    Description:
2446    
2447    This should (hopefully) fix the long-standing signal handling bug.
2448    (The runtime system was constructing a continuation record with an
2449    incorrect descriptor which would cause the GC to drop data on the floor...)
2450    
2451    ----------------------------------------------------------------------
2452    Name: Matthias Blume
2453    Date: 2001/06/15 15:05:00 EDT
2454    Tag: blume-20010615-moresparc
2455    Description:
2456    
2457    Here is a short late-hour update related to Sparc c-calls:
2458    
2459     -- made handling of double-word arguments a bit smarter
2460    
2461     -- instruction selection phase tries to collapse certain clumsily
2462        constructed ML-Trees; typical example:
2463    
2464            ADD(ty,ADD(_,e,LI d1),LI d2)  ->  ADD(ty,e,LI(d1+d2))
2465    
2466        This currently has no further impact on SML/NJ since mlriscGen does
2467        not seem to generate such patterns in the first place, and c-calls
2468        (which did generate them in the beginning) has meanwhile been fixed
2469        so as to avoid them as well.
2470    
2471    ----------------------------------------------------------------------
2472    Name: Matthias Blume
2473    Date: 2001/06/15 15:05:00 EDT
2474    Tag: blume-20010615-sparc
2475    Description:
2476    
2477    The purpose of this update is to provide an implementation of NLFFI
2478    on Sparc machines.
2479    
2480    Here are the changes in detail:
2481    
2482       * src/MLRISC/sparc/c-calls/sparc-c-calls.sml is a new file containing
2483       the Sparc implementation of the c-calls API.
2484       * The Sparc backend of SML/NJ has been modified to uniformely use %fp
2485       for accessing the ML frame.  Thus, we have a real frame pointer and
2486       can freely modify %sp without need for an omit-frame-ptr phase.
2487       The vfp logic in src/compiler/CodeGen/* has been changed to accomodate
2488       this case.
2489       * ml-nlffigen has been taught to produce code for different architectures
2490       and calling conventions.
2491       * In a way similar to what was done in the x86 case, the Sparc
2492       backend uses its own specific extension to mltree.  (For example,
2493       it needs to be able to generate UNIMP instructions which are part
2494       of the calling convention.)
2495       * ml-nlffi-lib was reorganized to make it more modular (in particular,
2496       to make it easier to plug in new machine- and os-dependent parts).
2497    
2498    There are some other fairly unrelated bug fixes and cleanups as well:
2499    
2500       * I further hacked the .cm files for MLRISC tools (like MDLGen) so
2501       that they properly share their libraries with existing SML/NJ libraries.
2502       * I fixed a minor cosmetic bug in CM, supressing certain spurious
2503       follow-up error messages.
2504       * Updates to CM/CMB documentation.
2505    
2506    TODO items:
2507    
2508       * MLRISC should use a different register as its asmTemp on the Sparc.
2509         (The current %o2 is a really bad choice because it is part of the
2510         calling conventions, so things might interfere in unexpected ways.)
2511    
2512    ----------------------------------------------------------------------
2513    Name: Matthias Blume
2514    Date: 2001/06/07
2515    Tag: blume-20010607-calls
2516    Description:
2517    
2518    A number of internal changes related to C calls and calling conventions:
2519    
2520    1. ML-Tree CALL statements now carry a "pops" field.  It indicates the
2521       number of bytes popped implicitly (by the callee).  In most cases
2522       this field is 0 but on x86/win32 it is some non-zero value.  This
2523       is information provided for the benefit of the "omit-frameptr" pass.
2524    2. The CALL instruction on the x86 carries a similar "pops" field.
2525       The instruction selection phase copies its value from the ML-Tree
2526       CALL statement.
2527    3. On all other architectures, the instruction selection phase checks
2528       whether "pops=0" and complains if not.
2529    4. The c-calls implementation for x86 now accepts two calling conventions:
2530       "ccall" and "stdcall".  When "ccall" is selected, the caller cleans
2531       up after the call and pops is set to 0.  For "stdcall", the caller
2532       does nothing, leaving the cleanup to the callee; pops is set to
2533       the number of bytes that were pushed onto the stack.
2534    5. The cproto decoder (compiler/Semant/types/cproto.sml) now can
2535       distinguish between "ccall" and "stdcall".
2536    6. The UNIMP instruction has been added to the supported Sparc instruction
2537       set. (This is needed for implementing the official C calling convention
2538       on this architecture.)
2539    7. I fixed some of the .cm files under src/MLRISC/Tools to make them
2540       work with the latest CM.
2541    
2542    ----------------------------------------------------------------------
2543    Name: Matthias Blume
2544    Date: 2001/06/05 15:10:00 EDT
2545    Tag: blume-20010605-cm-index
2546    Description:
2547    
2548    0. The "lambdasplit" parameter for class "sml" in CM has been documented.
2549    
2550    1. CM can now generate "index files".  These are human-readable files
2551       that list on a per-.cm-file basis each toplevel symbol defined or
2552       imported.  The location of the index file for
2553       <p>/<d>.cm is <p>/CM/INDEX/<d>.cm.
2554       To enable index-file generation, set CM.Control.generate_index to true
2555       or export an environment-symbol: export CM_GENERATE_INDEX=true.
2556    
2557       The CM manual has been updated accordingly.
2558    
2559    2. I made some slight modifications to the c-calls API in MLRISC.
2560    
2561         a) There is now a callback to support saving/restoring of
2562            dedicated but caller-save registers around the actual call
2563            instruction.
2564         b) One can optionally specify a comment-annotation for the
2565            call instruction.
2566    
2567    3. SML/NJ (mlriscGen.sml) uses this new API for the rawccall primop.
2568       (For example, the comment annotation shows the C prototype of
2569        the function being called.)
2570    
2571    ----------------------------------------------------------------------
2572    Name: Matthias Blume
2573    Date: 2001/06/01 13:30:00 EDT
2574    Tag: blume-20010601-nlffi-cleanup
2575    Description:
2576    
2577    This is mostly a cleanup of MLFFI stuff:
2578    
2579       - some signature files have been put into a more exposed place
2580       - the ugly 'f type parameter is gone (simplifies types tremendously!)
2581       - ml-nlffigen changed accordingly
2582       - tutorial updated
2583    
2584    Other changes:
2585    
2586       - author's affiliation in CM manual(s) updated
2587       - some more recognized keywords added to Allen's sml.sty
2588    
2589    ----------------------------------------------------------------------
2590    Name: Matthias Blume
2591    Date: 2001/05/25 15:30:00 EDT
2592    Tag: blume-20010525-iptr
2593    Description:
2594    
2595      - put the official 110.33-README (as it appears on the ftp server) under
2596        CVS
2597      - fixed a small bug related to incomplete pointer types in
2598        ml-nlffigen
2599      - small cosmetic change to the ml-nlffi-lib's "arr" type constructor
2600        (it does not need the 'f type parameter)
2601    
2602    ----------------------------------------------------------------------
2603    Name: Matthias Blume
2604    Date: 2001/05/23 14:30:00 EDT
2605    Tag: Release_110_33
2606    Description:
2607    
2608    New version number (110.33).  New bootfiles.
2609    
2610    ----------------------------------------------------------------------
2611    Name: Matthias Blume
2612    Date: 2001/05/22 18:06:00 EDT
2613    Tag: blume-20010522-targets
2614    Description:
2615    
2616    Made install.sh use file config/targets.customized if it exists, falling
2617    back to config/targets if it doesn't.  This way one can have a customized
2618    version of the targets file without touching the "real thing", thus
2619    eliminating the constant fear of accidentally checking something bogus
2620    back into the CVS repository...  (File config/targets.customized must
2621    not be added to the repository!)
2622    
2623    ----------------------------------------------------------------------
2624    Name: Matthias Blume
2625    Date: 2001/05/22 16:30:00 EDT
2626    Tag: blume-20010522-minitut
2627    Description:
2628    
2629    1. Bug fix in ml-nlffigen; now (hopefully) correctly handling
2630       struct returns.
2631    2. Added src/ml-nlffi-lib/Doc/mini-tutorial.txt.  This is some very
2632       incomplete, preliminary documentation for NLFFI.
2633    
2634    ----------------------------------------------------------------------
2635    Name: Matthias Blume
2636    Date: 2001/05/14 11:30:00 EDT
2637    Tag: blume-20010514-script
2638    Description:
2639    
2640    Some bugs in install script fixed.
2641    
2642    In addition to that I also made a slight change to the NLFFI API:
2643    Functors generated by ml-nlffigen now take the dynamic library as a
2644    straight functor argument, not as a suspended one.  (The original
2645    functor code used to force the suspension right away anyway, so there
2646    was nothing gained by this complication of the interface.)
2647    
2648    ----------------------------------------------------------------------
2649    Name: Matthias Blume
2650    Date: 2001/05/11 14:35:00 EDT
2651    Tag: blume-20010511-ml-nlffi
2652    Description:
2653    
2654    I finally took the plunge and added my new FFI code to the main
2655    repository.  For x86-linux it is now ready for prime-time.
2656    
2657    There are two new subdirectories of "src":
2658    
2659      - ml-nlffi-lib:
2660           The utility library for programs using the FFI interface.
2661           Here is the implementation of $/c.cm and its associated low-level
2662           partners $/c-int.cm and $/memory.cm.
2663      - ml-nlffigen:
2664           A stand-alone program for generating ML glue code from C source
2665           code.
2666    
2667    Building ml-nlffigen requires $/ckit-lib.cm.
2668    
2669    The config/install.sh script has been updates to do the Right Thing
2670    (hopefully).
2671    
2672    Notice that the source tree for the C-Kit will not be put under "src"
2673    but directly under the installation root directory.  (This is the
2674    structure that currently exists on the CVS server when you check out
2675    module "sml".)  Fortunately, config/install.sh knows about this oddity.
2676    
2677    Bugs: No documentation yet.
2678    
2679    ----------------------------------------------------------------------
2680    Name: Matthias Blume
2681    Date: 2001/05/09 16:35:00 EDT
2682    Tag: blume-20010509-cpscontract
2683    Description:
2684    
2685    Fixed a bug in the accounting code in cpsopt/contract.sml.  (The
2686    wrapper/unwrapper elimination did not decrement usage counts and some
2687    dead variables got overlooked by the dead-up logic.)
2688    
2689    ----------------------------------------------------------------------
2690    Name: Lal George
2691    Date: 2001/05/08  17:26:09 EDT
2692    Tag: george-20010508-omit-frameptr
2693    Description:
2694    
2695    Changes to implement the omit-frame-pointer optimization to support
2696    raw C calls. For now, there is only support on the Intel x86, but
2697    other architectures will follow as more experience is gained with this.
2698    
2699    
2700    ----------------------------------------------------------------------
2701    Name: Matthias Blume
2702    Date: 2001/05/07 14:40:00 EDT
2703    Tag: blume-20010507-proxies
2704    Description:
2705    
2706    I made into "proxy libraries" all libraries that qualify for such a
2707    change.  (A qualifying library is a library that has another library or
2708    groups as its sole member and repeats that member's export list
2709    verbatim.  A proxy library avoids this repetition by omitting its export
2710    list, effectively inheriting the list that its (only) member exports.
2711    See the CM manual for more explanation.)
2712    The main effect is that explicit export lists for these libraries
2713    do not have to be kepts in sync, making maintenance a bit easier.
2714    
2715    I also added copyright notices to many .cm-files.
2716    
2717    Last but not least, I made a new set of bootfiles.
2718    
2719    ----------------------------------------------------------------------
2720    Name: Matthias Blume
2721    Date: 2001/05/04 17:00:00 EDT
2722    Tag: blume-20010504-cm-lsplit
2723    Description:
2724    
2725    0. John merged pending changes to $/smlnj-lib.cm
2726    
2727    1. Allen's previous change accidentally backed out of one of Lal's
2728       earlier changes.  I undid this mistake (re-introducing Lal's change).
2729    
2730    2. I used the new topOrder' function from graph-scc.sml (from $/smlnj-lib.cm)
2731       within the compiler where applicable.  There is some code simplification
2732       because of that.
2733    
2734    3. The "split" phase (in FLINT) is now part of the default list of phases.
2735       Compiler.Control.LambdaSplitting.* can be used to globally control the
2736       lambda-splitting (cross-module-inlining) engine.  In addition to that,
2737       it can now also be controlled on a per-source basis: CM has been taught
2738       a new tool parameter applicable to ML source files.
2739    
2740       - To turn lambda-splitting off completely:
2741            local open Compiler.Control.LambdaSplitting in
2742                val _ = set Off
2743            end
2744       - To make "no lambda-splitting" the global default (but allow per-source
2745         overriding); this is the initial setting:
2746            local open Compiler.Control.LambdaSplitting in
2747                val _ = set (Default NONE)
2748            end
2749       - To make "lambda-splitting with aggressiveness a" the global default
2750         (and allow per-source overriding):
2751            local open Compiler.Control.LambdaSplitting in
2752                val _ = set (Default (SOME a))
2753            end
2754    
2755       - To turn lambda-splitting off for a given ML souce file (say: a.sml)
2756         write (in the respective .cm-file):
2757            a.sml (lambdasplitting:off)
2758       - To turn lambda-splitting for a.sml on with minimal aggressiveness:
2759            a.sml (lambdasplitting:on)
2760       - To turn lambda-splitting for a.sml on with aggressiveness <a> (where
2761         <a> is a decimal non-negative integer):
2762            a.sml (lambdasplitting:<a>)
2763       - To turn lambda-splitting for a.sml on with maximal aggressiveness:
2764            a.sml (lambdasplitting:infinity)
2765       - To use the global default for a.sml:
2766            a.sml (lambdasplitting:default)
2767         or simply
2768            a.sml
2769    
2770    ----------------------------------------------------------------------
2771    Name: Allen Leung
2772    Date: 2001/05/04 01:57:00 EDT
2773    Tag: leunga-20010504-sync
2774    Description:
2775    
2776      MLRISC features.
2777    
2778      1. Fix to CMPXCHG instructions.
2779      2. Changed RA interface to allow annotations in callbacks.
2780      3. Added a new method to the stream interface to allow annotations updates.
2781    
2782    ----------------------------------------------------------------------
2783    Name: Matthias Blume
2784    Date: 2001/05/01 11:45:00 EDT
2785    Tag: blume-20010501-pcedittmp
2786    Description:
2787    
2788    Changed install.sh to use the current working directory instead of
2789    /usr/tmp for a temporary file (pcedittmp).  The previous choice
2790    of /usr/tmp caused trouble with MacOS X because of file premission
2791    problems.
2792    
2793    ----------------------------------------------------------------------
2794    Name: Matthias Blume
2795    Date: 2001/04/20 11:10:00 EDT
2796    Tag: blume-20010420-inMLflag
2797    Description:
2798    
2799     - added vp_limitPtrMask to vproc-state.h
2800       (for use by the raw-C-calls mechanism to implement proper interrupt
2801        handling)
2802     - made the ML compiler aware of various data-structure offsets so it
2803       can generate code for accessing the vp_inML flag and vp_limitPtrMask
2804     - tweaked mlriscGen.sml to have it emit interrupt-handling code for
2805       raw C-calls
2806    
2807    ----------------------------------------------------------------------
2808    Name: Lal George
2809    Date: 2001/04/20 09:15:28 EDT
2810    Tag: george-20010420-macosX
2811    Description:
2812    
2813     - Changes to port to Mac OS X; Darwin.
2814    
2815     - In the process I found that sqrt was broken on the PPC, because the
2816       fsqrt instruction is not implemented.
2817    
2818    ----------------------------------------------------------------------
2819    Name: Matthias Blume
2820    Date: 2001/04/18 12:45:00 EDT
2821    Tag: blume-20010418-ccalls
2822    Description:
2823    
2824     - fixed two off-by-4 errors in the x86-specific c-calls implementation
2825       (this bug prevented structure arguments containing pointers from being
2826        passed correctly)
2827     - changed the raw-C-call code in mlriscGen.sml in such a way that
2828       structure arguments are represented as a pointer to the beginning
2829       of the structure (instead of having a series of synthesized arguments,
2830       one for each structure member)
2831    
2832     - made makeml script's verbosity level configurable via environment
2833       variable (MAKEML_VERBOSITY)
2834    
2835     - eliminated placeholder implementations for f32l, w16s, i16s, and f32s
2836       in rawmem-x86.sml; we are now using the real thing
2837    
2838    ----------------------------------------------------------------------
2839    Name: Matthias Blume
2840    Date: 2001/03/22 16:25:00 EST
2841    Tag: blume-20010322-bootfiles
2842    Description:
2843    
2844    Created a new set of bootfiles (for your automatic installation convenience).
2845    
2846    ----------------------------------------------------------------------
2847    Name: Matthias Blume
2848    Date: 2001/03/22 15:10:00 EST
2849    Tag: blume-20010322-rawmem-parcm
2850    Description:
2851    
2852    1. All "raw memory access" primitives for the new FFI are implemented now
2853       (at least on the x86).
2854    2. Some further cleanup of CM's parallel make mechanism.
2855    
2856    ----------------------------------------------------------------------
2857    Name: Matthias Blume
2858    Date: 2001/03/19 17:53:00 EST
2859    Tag: blume-20010319-parallel
2860    Description:
2861    
2862    Parallel make (using compile servers) now works again.
2863    
2864    To this end, CM.stabilize and CMB.make have been modified to work in
2865    two passes when compile servers are attached:
2866       1. Compile everything, do not perform stabilization; this pass
2867          uses compile servers
2868       2. Stabilize everything; this pass does not use compile servers
2869    If there are no compile servers, the two passes are combined into one
2870    (as before).  Splitting the passes increases the inherent parallelism
2871    in the dependency graph because the entire graph including all
2872    libraries is available at the same time.  This, in turn, improves
2873    server utilization.  The downside is that the master process will
2874    have to do some extra work after compilation is done (because for
2875    technical reasons it must re-read all the binfiles during stabilization).
2876    
2877    ----------------------------------------------------------------------
2878    Name: Matthias Blume
2879    Date: 2001/03/16 12:22:00 EST
2880    Tag: blume-20010316-bootfiles
2881    Description:
2882    
2883    Created a new set of bootfiles (for your automatic installation convenience).
2884    
2885    ----------------------------------------------------------------------
2886    Name: Matthias Blume
2887    Date: 2001/03/16 11:00:00 EST
2888    Tag: blume-20010316-MLTREE-fixup
2889    Description:
2890    
2891    This is a minor fixup for an (untagged) earlier commit by Allen.
2892    (A file was missing).
2893    
2894    ----------------------------------------------------------------------
2895    Name: Allen Leung
2896    Date: Mon Mar  5 18:54:57 EST 2001
2897    Tag: leunga-20010305-cut-support
2898    
2899    1. New support for alternative control-flow in MLTREE.
2900       Currently we support
2901    
2902          FLOW_TO(CALL ...., [k1,...,kn])
2903    
2904       This is needed for 'cuts to' in C-- and try/handle-like constructs
2905       in Moby
2906    
2907       New assembler flag "asm-show-cutsto" to turn on control-flow debugging.
2908    
2909    2. Register Allocator
2910    
2911       Changes in interface [from Fermin, John]
2912    
2913    3. Alpha 8-bit SLL support [Fermin]
2914    
2915    4. All architectures
2916    
2917       A new module (ClusterExpandCopies) for expanding parallel copies.
2918    
2919    ----------------------------------------------------------------------
2920    Name: Allen Leung
2921    Date: 2001/02/27 23:07:00 EST
2922    Tag: leunga-20010227-minor-stuff
2923    
2924    1. Alpha bug fix for CMOVNE
2925    2. Handle mltree COND(..,FCMP ...,...)
2926    3. Bug fix in simplifier
2927    
2928    ----------------------------------------------------------------------
2929    Name: Matthias Blume
2930    Date: 2001/01/30 17:50:00 EST
2931    Tag: blume-20010130-sync
2932    Description:
2933    
2934    This is just a minor update to sync my devel branch with the main brach.
2935    The only visible change is the addition of some README files.
2936    
2937    ----------------------------------------------------------------------
2938    Name: Matthias Blume
2939    Date: 2001/01/12 23:30:00 JST
2940    Tag: blume-20010112-bootfiles
2941    Description:
2942    
2943    Made a new set of bootfiles that goes with the current state of the
2944    repository.
2945    
2946    ----------------------------------------------------------------------
2947    Name: Matthias Blume
2948    Date: 2001/01/12 21:20:00 JST
2949    Tag: blume-20010112-sync
2950    Description:
2951    
2952    I am just flushing out some minor changes that had accumulated in
2953    my private branch in order to sync with the main tree.  (This is
2954    mainly because I had CVS trouble when trying to merge _into_ my
2955    private branch.)
2956    
2957    Most people should be completely unaffected by this.
2958    
2959    ----------------------------------------------------------------------
2960    Name: Allen Leung
2961    Date: Thu Jan 11 21:03:00 EST 2001
2962    Tag: leunga-20010111-labexp=mltree
2963    Description:
2964    
2965    1.  Removed the type LabelExp and replace it by MLTree.
2966    2.  Rewritten mltree-simplify with the pattern matcher tool.
2967    3.  There were some bugs in alpha code generator which would break
2968        64-bit code generation.
2969    4.  Redo the tools to generate code with the
2970    5.  The CM files in MLRISC (and in src/system/smlnj/MLRISC)
2971        are now generated by perl scripts.
2972    
2973    ----------------------------------------------------------------------
2974    Name: Matthias Blume
2975    Date: 2001/01/10 21:55:00 JST
2976    Tag: blume-20010110-rcc
2977    Description:
2978    
2979    The RCC stuff now seems to work (but only on the x86).
2980    This required hacking of the c-calls interface (and -implementation) in
2981    MLRISC.
2982    
2983    Normal compiler users should be unaffected.
2984    
2985    ----------------------------------------------------------------------
2986    Name: Matthias Blume
2987    Date: 2001/01/09 01:20:00 JST
2988    Tag: blume-20010109-rcc
2989    Description:
2990    
2991    This is a fairly big patch, flushing out a large number of pending
2992    changes that I made to my development copy over the last couple of days.
2993    
2994    Of practical relevance at this moment is a workaround for a pickling
2995    bug that Allen ran into the other day.  The cause of the bug itself is
2996    still unknown and it might be hard to fix it properly, but the
2997    workaround has some merits of its own (namely somewhat reducing pickling
2998    overhead for certain libraries).  Therefore, I think this solution should
2999    be satisfactory at this time.
3000    
3001    The rest of the changes (i.e., the vast majority) has to do with my
3002    ongoing efforts of providing direct support for C function calls from
3003    ML.  At the moment there is a new primop "RAW_CCALL", typing magic
3004    in types/cproto.sml (invoked from FLINT/trans/translate.sml), a new
3005    case in the FLINT CPS datatype (RCC), changes to cps/convert.sml to
3006    translate uses of RAW_CCALL into RCC, and changes to mlriscGen.sml to
3007    handle RCC.
3008    
3009    The last part (the changes to mlriscGen.sml) are still known to be
3010    wrong on the x86 and not implemented on all other architectures.  But
3011    the infrastructure is in place. I had to change a few functor
3012    signatures in the backend to be able to route the CCalls interface
3013    from MLRISC there, and I had to specialize the mltree type (on the
3014    x86) to include the necessary extensions. (The extensions themselves
3015    were already there and redy to go in MLRISC/x86).
3016    
3017    Everything should be very happy as soon as someone helps me with
3018    mlriscGen.sml...
3019    
3020    In any case, nothing of this should matter to anyone as long as the
3021    new primop is not being used (which is going to be the case unless you
3022    find it where I hid it :). The rest of the compiler is completely
3023    unaffected.
3024    
3025    ----------------------------------------------------------------------
3026    Name: Matthias Blume
3027    Date: 2001/01/05 00:30:00 JST
3028    Tag: blume-20010105-primops
3029    Description:
3030    
3031    Added some experimental support for work that I am doing right now.
3032    These changes mostly concern added primops, but there is also a new
3033    experimental C library in the runtime system (but currently not enabled
3034    anywhere except on Linux/X86).
3035    
3036    In the course of adding primops (and playing with them), I discovered that
3037    Zhong's INL_PRIM hack (no type info for certain primops) was, in fact, badly
3038    broken.  (Zhong was very right he labeled this stuff as "major gross hack".)
3039    To recover, I made type information in INL_PRIM mandatory and changed
3040    prim.sml as well as built-in.sml accordingly.  The InLine structure now
3041    has complete, correct type information (i.e., no bottom types).
3042    
3043    Since all these changes mean that we need new binfiles, I also bumped the
3044    version number to 110.32.1.
3045    
3046    ----------------------------------------------------------------------
3047    Name: Matthias Blume
3048    Date: 2000/12/30 22:10:00 JST
3049    Tag: blume-20001230-various
3050    Description:
3051    
3052    Added proxy libraries for MLRISC and let MLRISC libraries refer
3053    to each other using path anchors.  (See CM manual for explanation.)
3054    
3055    Updated CM documentation.
3056    
3057    Fixed some bugs in CM.
3058    
3059    Implemented "proxy" libraries (= syntactic sugar for CM).
3060    
3061    Added "-quiet" option to makeml and changed runtime system accordingly.
3062    
3063    Added cleanup handler for exportML to reset timers and compiler stats.
3064    
3065    ----------------------------------------------------------------------
3066    Name: Lal George
3067    Date: 2000/12/22 22:22:58 EST 2000
3068    Tag: Release_110_32
3069    Description:
3070    
3071            Infinite precision used throughout MLRISC.
3072            see MLRISC/mltree/machine-int.sig
3073    
3074    ----------------------------------------------------------------------
3075    Name: Matthias Blume
3076    Date: 2000/12/22 23:16:00 JST
3077    Tag: blume-20001222-warn
3078    Description:
3079    
3080    Corrected wording and formatting of some CM warning message which I
3081    broke in my previous patch.
3082    
3083    ----------------------------------------------------------------------
3084    Name: Matthias Blume
3085    Date: 2000/12/22 21:20:00 JST
3086    Tag: blume-20001222-anchorenv
3087    Description:
3088    
3089    Fixed CM's handling of anchor environments in connection with CMB.make.
3090    
3091    ----------------------------------------------------------------------
3092    Name: Matthias Blume
3093    Date: 2000/12/22 13:15:00 JST
3094    Tag: blume-20001222-cleanup
3095    Description:
3096    
3097    Removed src/cm/ffi which does not (and did not) belong here.
3098    
3099    ----------------------------------------------------------------------
3100    Name: Matthias Blume
3101    Date: 2000/12/21 23:55:00 JST
3102    Tag: blume-20001221-exn
3103    Description:
3104    
3105    Probably most important: CM no longer silently swallows all exceptions
3106    in the compiler.
3107    Plus: some other minor CM changes.  For example, CM now reports some
3108    sizes for generated binfiles (code, data, envpickle, lambdapickle).
3109    
3110    ----------------------------------------------------------------------
3111    Name: Matthias Blume
3112    Date: 2000/12/15 00:01:05 JST
3113    Tag: blume-20001215-dirtool
3114    Description:
3115    
3116    - "dir" tool added.
3117    - improvements and cleanup to Tools structure
3118    - documentation updates
3119    
3120    ----------------------------------------------------------------------
3121    Name: Allen Leung
3122    Date: Thu Dec 14 03:45:24 EST 2000
3123    Description:
3124    Tag:  leunga-20001214-int-inf
3125    Description:
3126    
3127       In IntInf, added these standard functions, which are missing from our
3128    implementation:
3129    
3130        andb : int * int -> int
3131        xorb : int * int -> int
3132        orb  : int * int -> int
3133        notb : int -> int
3134         <<   : int * word -> int
3135        ~>>  : int * word -> int
3136    
3137       Not tested, I hope they are correct.
3138    
3139    ----------------------------------------------------------------------
3140    Name: Allen Leung
3141    Date: Fri Dec  8 19:23:26 EST 2000
3142    Description:
3143    Tag:  leunga-20001208-nowhere
3144    Description:
3145    
3146      Slight improvements to the 'nowhere' tool to handle OR-patterns,
3147    to generate better error messages etc.  Plus a brief manual.
3148    
3149    ----------------------------------------------------------------------
3150    Name: Lal George
3151    Date: 2000/12/08 09:54:02 EST 2000
3152    Tag: Release_110_31
3153    Description:
3154    
3155    - Version 110.31
3156    ----------------------------------------------------------------------
3157    Name: Allen Leung
3158    Date: Thu Dec  7 22:01:04 EST 2000
3159    Tag:  leunga-20001207-cell-monster-hack
3160    Description:
3161    
3162    Major MLRISC internal changes.  Affect all clients.
3163    Summary:
3164    
3165    1.  Type CELLS.cell = int is now replaced by a datatype.
3166        As a result, the old regmap is now gone.  Almost all interfaces
3167        in MLRISC change as a consequence.
3168    
3169    2.  A new brand version of machine description tool (v3.0) that generates
3170        modules expecting the new interface.  The old version is removed.
3171    
3172    3.  The RA interface has been further abstracted into two new functors.
3173        RISC_RA and X86RA.  These functors have much simpler interfaces.
3174        [See also directory MLRISC/demo.]
3175    
3176    4.  Some other new source->source code generation tools are available:
3177    
3178        a. MLRISC/Tools/RewriteGen -- generate rewriters from rules.
3179        b. MLRISC/Tools/WhereGen -- expands conditional pattern matching rules.
3180           I use this tool to generate the peephole optimizers---with the new
3181           cell type changes, peephole rules are becoming difficult to write
3182           without conditional pattern matching.
3183    
3184    5.  More Intmap -> IntHashTable change.  Previous changes by Matthias didn't
3185        cover the entire MLRISC source tree so many things broke.
3186    
3187    6.  CM files have been moved to the subdirectory MLRISC/cm.
3188        They are moved because there are a lot of them and they clutter up the
3189        root dir.
3190    
3191    7.  More detailed documentation to come...
3192    
3193        NOTE: To rebuild from 110.30 (ftp distribution), you'll have to do
3194        a makeml -rebuild first.  This is because of other other
3195        changes that Matthias has made (see below).
3196    
3197    
3198    ----------------------------------------------------------------------
3199    Name: Matthias Blume
3200    Date: 2000/11/30 23:12:00 JST
3201    Tag: blume-20001130-filereorg
3202    Description:
3203    
3204    Some manual updates and some file reorganizations in CM.
3205    
3206    ----------------------------------------------------------------------
3207    Name: Matthias Blume
3208    Date: 2000/11/24 17:45:00 JST
3209    Tag: blume-20001124-link
3210    Description:
3211    
3212    Drastically improved link traversal code for the case that the dynamic
3213    value was already loaded at bootstrap time.  As a result, CM and CMB
3214    now both load blazingly fast -- even on a very slow machine.  Also,
3215    memory consumption has been further reduced by this.
3216    
3217    Warning: The format of the PIDMAP file has changed.  THerefore, to
3218    bootstrap you have to do this:
3219    
3220    1. Run CMB.make
3221    2. Make a symbolic link for the boot directory:
3222         ln -s sml.boot.ARCH-OS xxx
3223    3. "Rebuild" the boot directory:
3224         ./makeml -boot xxx -rebuild sml ; rm xxx
3225    4. Boot normally:
3226          ./makeml
3227    
3228    ----------------------------------------------------------------------
3229    Name: Matthias Blume
3230    Date: 2000/11/21 21:20:00 JST
3231    Tag: blume-20001121-tools
3232    Description:
3233    
3234    Continued hacking on autoloading problem -- with success this time.
3235    Also changed tool-plugin mechanism.  See new CM manual.
3236    
3237    ----------------------------------------------------------------------
3238    Name: Matthias Blume
3239    Date: 2000/11/19 14:30:00 JST
3240    Tag:  blume-20001119-autoload
3241    Description:
3242    
3243    Some hacking to make autoloading faster.  Success for CMB, no success
3244    so far for CM.  There is a reduced structure CM' that autoloads faster.
3245    (This is a temporary, non-documented hack to be eliminated again when
3246    the general problem is solved.)
3247    
3248    ----------------------------------------------------------------------
3249    Name: Matthias Blume
3250    Date: 2000/11/17 14:10:00 JST
3251    Tag: blume-20001117-pickle-lib
3252    Description:
3253    
3254    1. Eliminated comp-lib.cm
3255    2. Made pickle-lib.cm
3256    3. Eliminated all uses of intset.sml (from comp-lib.cm)
3257    4. Replaced all uses of intmap.{sig,sml} (from comp-lib.cm) with
3258       equivalent constructs from smlnj-lib.cm (INtHashTable).
3259    5. Point 4. also goes for those uses of intmap.* in MLRISC.
3260       Duplicated intmap modules thrown out.
3261    6. Hunted down all duplicated SCC code and replaced it with
3262       equivalent stuff (GraphSCCFn from smlnj-lib.cm).
3263    7. Rewrote Feedback module.
3264    8. Moved sortedlist.sml into viscomp-lib.cm.  Eventually it
3265       should be thrown out and equivalent modules from smlnj-lib.cm
3266       should be used (IntRedBlackSet, IntListSet, ...).
3267    
3268    Confirmed that compiler compiles to fixpoint.
3269    
3270    ----------------------------------------------------------------------
3271    Name: Allen Leung
3272    Date: 2000/11/10 18:00:00
3273    Tag: leunga-20001110-new-x86-fp
3274    
3275    A new x86 floating point code generator has been added.
3276    By default this is turned off.  To turn this on, do:
3277    
3278        CM.autoload "$smlnj/compiler.cm";
3279        Compiler.Control.MLRISC.getFlag "x86-fast-fp" := true;
3280    
3281    Changes:
3282    
3283    1.  Changed FTAN to FPTAN so that the assembly output is correct.
3284    2.  Changed the extension callback for FTANGENT to generate:
3285    
3286              fptan
3287              fstp  %st(0)
3288        instead of
3289              fptan
3290              fstpl ftempmem
3291    
3292    3.  Numerous assembly fixes for x86.
3293    
3294    5.  Cleaned up the machine code output module x86/x86MC.sml and added
3295        support for a whole bunch of instructions and addressing modes:
3296    
3297          fadd/fsub/fsubr/fmul/fdiv/fdivr  %st, %st(n)
3298          faddp/fsubp/fsubrp/fmulp/fdivp/fdivrp  %st, %st(n)
3299          fadd/fsub/fsubr/fmul/fdiv/fdivr  %st(n), %st
3300          fiadd/fisub/fisubr/fimul/fidiv/fidivr mem
3301          fxch %st(n)
3302          fld %st(n)
3303          fst %st(n)
3304          fst mem
3305          fstp %st(n)
3306          fucom %st(n)
3307          fucomp %st(n)
3308    
3309        All these are now generated when the fast fp mode is turned on.
3310    
3311    6.  Removed the dedicated registers %st(0), ..., %st(7) from X86CpsRegs
3312    
3313    ----------------------------------------------------------------------
3314    Name: Matthias Blume
3315    Date: 2000/11/09 11:20:00 JST
3316    Tag: blume-20001109-scc
3317    Description:
3318    
3319    Eliminated some code duplication:
3320    
3321    1. Added "where" clause to GraphSCCFn in SML/NJ Library.
3322       (Otherwise the functor is useless.)
3323    2. Used GraphSCCFn where SCCUtilFun was used previously.
3324    3. Got rid of SCCUtilFun (in comp-lib.cm).
3325    
3326    ----------------------------------------------------------------------
3327    Name: Lal George
3328    Date: 2000/11/06 09:02:21 EST 2000
3329    Tag: Release_110_30
3330    Description:
3331    
3332    - Version 110.30
3333    ----------------------------------------------------------------------
3334    Name: Matthias Blume
3335    Date: 2000/11/04 14:45:00
3336    Tag: blume-20001104-mlbuild
3337    Description:
3338    
3339    - Made ml-build faster on startup.
3340    - Documentation fixes.
3341    
3342    ----------------------------------------------------------------------
3343    Name: Matthias Blume
3344    Date: 2000/11/02 17:00:00 JST
3345    Tag: blume-20001102-condcomp
3346    Description:
3347    
3348    - Small tweaks to pickler -- new BOOTFILES!
3349    - Version bumped to 110.29.2.
3350    - Added conditional compilation facility to init.cmi (see comment there).
3351    ----------------------------------------------------------------------
3352    Name: Allen Leung
3353    Date: 2000/10/23 19:31:00
3354    Tag: leunga-20001023-demo-ra
3355    
3356    1. Minor RA changes that improves spilling on x86 (affects Moby and C-- only)
3357    2. Test programs for the graph library updated
3358    3. Some new MLRISC demo programs added
3359    
3360    ----------------------------------------------------------------------
3361    Name: Matthias Blume
3362    Date: 2000/08/31 22:15:00 JST
3363    Tag: blume-20001017-errmsg
3364    Description:
3365    
3366    More error message grief: Where there used to be no messages, there
3367    now were some that had bogus error regions.  Fixed.
3368    
3369    ----------------------------------------------------------------------
3370    Name: Matthias Blume
3371    Date: 2000/08/31 17:30:00 JST
3372    Tag: blume-20001017-v110p29p1
3373    Description:
3374    
3375    I made a version 110.29.1 with new bootfiles.
3376    
3377    Changes:  Modified pickler/unpickler for faster and leaner unpickling.
3378              CM documentation changes and a small bugfix in CM's error reporting.
3379    
3380    ----------------------------------------------------------------------
3381    Name: Lal George
3382    Date: 2000/09/27 14:42:35 EDT
3383    Tag: george-20000927-nodestatus
3384    Description:
3385    
3386    Changed the type of the nodestatus, so that:
3387    
3388            SPILLED(~1)             is now SPILLED
3389            SPILLED(m) where m>=0   is now MEMREG(m)
3390            SPILLED(s) where s<~1   is now SPILL_LOC(~s)
3391    
3392    ----------------------------------------------------------------------
3393    Name: Matthias Blume
3394    Date: 2000/09/07 14:45:00 JST
3395    Tag: blume-20000907-cmerrmsg
3396    Description:
3397    
3398    Small tweak to CM to avoid getting ML syntax error messages twice.
3399    
3400  ----------------------------------------------------------------------  ----------------------------------------------------------------------
3401  Name: Matthias Blume  Name: Matthias Blume
3402  Date: 2000/08/31 18:00:00 JST  Date: 2000/08/31 18:00:00 JST
# Line 1479  Line 4865 
4865       elaborator). There were a lot of changes during my "linkpath" trials       elaborator). There were a lot of changes during my "linkpath" trials
4866       that could have been reverted to their original state but weren't.       that could have been reverted to their original state but weren't.
4867       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
4868       than what was strictly necessary...  (I _did_ resist the tempation       than what was strictly necessary...  (I _did_ resist the temptation
4869       of doing any "global reformatting" to avoid an untimely death at       of doing any "global reformatting" to avoid an untimely death at
4870       Dave's hands. :)       Dave's hands. :)
4871    

Legend:
Removed from v.701  
changed lines
  Added in v.1183

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