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

SCM Repository

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

Diff of /sml/trunk/HISTORY

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

revision 788, Wed Feb 28 04:09:48 2001 UTC revision 1131, Mon Mar 11 15:20:52 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: Allen Leung  Name: Allen Leung
17    Date: 2002/03/11 10:30:00 EST
18    Tag: leunga-20020310-runtime-string0
19    Description:
20    
21       The representation of the empty string now points to a
22    legal null terminated C string instead of unit.  It is now possible
23    to convert an ML string into C string with InlineT.CharVector.getData.
24    This compiles into one single machine instruction.
25    
26    ----------------------------------------------------------------------
27    Name: Allen Leung
28    Date: 2002/03/10 23:55:00 EST
29    Tag: leunga-20020310-x86-call
30    Description:
31    
32       Added machine generation for CALL instruction (relative displacement mode)
33    
34    ----------------------------------------------------------------------
35    Name: Matthias Blume
36    Date: 2002/03/08 16:05:00
37    Tag: blume-20020308-entrypoints
38    Description:
39    
40    Version number bumped to 110.39.1.  NEW BOOTFILES!
41    
42    Entrypoints: non-zero offset into a code object where execution should begin.
43    
44    - Added the notion of an entrypoint to CodeObj.
45    - Added reading/writing of entrypoint info to Binfile.
46    - Made runtime system bootloader aware of entrypoints.
47    - Use the address of the label of the first function given to mlriscGen
48      as the entrypoint.  This address is currently always 0, but it will
49      not be 0 once we turn on block placement.
50    - Removed the linkage cluster code (which was The Other Way(tm) of dealing
51      with entry points) from mlriscGen.
52    
53    ----------------------------------------------------------------------
54    Name: Allen Leung
55    Date: 2002/03/07 20:45:00 EST
56    Tag: leunga-20020307-x86-cmov
57    Description:
58    
59       Bug fixes for CMOVcc on x86.
60    
61       1. Added machine code generation for CMOVcc
62       2. CMOVcc is now generated in preference over SETcc on PentiumPro or above.
63       3. CMOVcc cannot have an immediate operand as argument.
64    
65    ----------------------------------------------------------------------
66    Name: Matthias Blume
67    Date: 2002/03/07 16:15:00 EST
68    Tag: blume-20020307-controls
69    Description:
70    
71    This is a very large but mostly boring patch which makes (almost)
72    every tuneable compiler knob (i.e., pretty much everything under
73    Control.* plus a few other things) configurable via both the command
74    line and environment variables in the style CM did its configuration
75    until now.
76    
77    Try starting sml with '-h' (or, if you are brave, '-H')
78    
79    To this end, I added a structure Controls : CONTROLS to smlnj-lib.cm which
80    implements the underlying generic mechanism.
81    
82    The interface to some of the existing such facilities has changed somewhat.
83    For example, the MLRiscControl module now provides mkFoo instead of getFoo.
84    (The getFoo interface is still there for backward-compatibility, but its
85    use is deprecated.)
86    
87    The ml-build script passes -Cxxx=yyy command-line arguments through so
88    that one can now twiddle the compiler settings when using this "batch"
89    compiler.
90    
91    TODO items:
92    
93    We should go through and throw out all controls that are no longer
94    connected to anything.  Moreover, we should go through and provide
95    meaningful (and correct!) documentation strings for those controls
96    that still are connected.
97    
98    Currently, multiple calls to Controls.new are accepted (only the first
99    has any effect).  Eventually we should make sure that every control
100    is being made (via Controls.new) exactly once.  Future access can then
101    be done using Controls.acc.
102    
103    Finally, it would probably be a good idea to use the getter-setter
104    interface to controls rather than ref cells.  For the time being, both
105    styles are provided by the Controls module, but getter-setter pairs are
106    better if thread-safety is of any concern because they can be wrapped.
107    
108    *****************************************
109    
110    One bug fix: The function blockPlacement in three of the MLRISC
111    backpatch files used to be hard-wired to one of two possibilities at
112    link time (according to the value of the placementFlag).  But (I
113    think) it should rather sense the flag every time.
114    
115    *****************************************
116    
117    Other assorted changes (by other people who did not supply a HISTORY entry):
118    
119    1. the cross-module inliner now works much better (Monnier)
120    2. representation of weights, frequencies, and probabilities in MLRISC
121       changed in preparation of using those for weighted block placement
122       (Reppy, George)
123    
124    ----------------------------------------------------------------------
125    Name: Lal George
126    Date: 2002/03/07 14:44:24 EST 2002
127    Tag: george-20020307-weighted-block-placement
128    
129    Tested the weighted block placement optimization on all architectures
130    (except the hppa) using AMPL to generate the block and edge frequencies.
131    Changes were required in the machine properties to correctly
132    categorize trap instructions. There is an MLRISC flag
133    "weighted-block-placement" that can be used to enable weighted block
134    placement, but this will be ineffective without block/edge
135    frequencies (coming soon).
136    
137    
138    ----------------------------------------------------------------------
139    Name: Lal George
140    Date: 2002/03/05 17:24:48 EST
141    Tag: george-20020305-linkage-cluster
142    
143    In order to support the block placement optimization, a new cluster
144    is generated as the very first cluster (called the linkage cluster).
145    It contains a single jump to the 'real' entry point for the compilation
146    unit. Block placement has no effect on the linkage cluster itself, but
147    all the other clusters  have full freedom in the manner in which they
148    reorder blocks or functions.
149    
150    On the x86 the typical linkage code that is generated is:
151       ----------------------
152            .align 2
153       L0:
154            addl    $L1-L0, 72(%esp)
155            jmp     L1
156    
157    
158            .align  2
159       L1:
160       ----------------------
161    
162    72(%esp) is the memory location for the stdlink register. This
163    must contain the address of the CPS function being called. In the
164    above example, it contains the address of  L0; before
165    calling L1 (the real entry point for the compilation unit), it
166    must contain the address for L1, and hence
167    
168            addl $L1-L0, 72(%esp)
169    
170    I have tested this on all architectures except the hppa.The increase
171    in code size is of course negligible
172    
173    ----------------------------------------------------------------------
174    Name: Allen Leung
175    Date: 2002/03/03 13:20:00 EST
176    Tag: leunga-20020303-mlrisc-tools
177    
178      Added #[ ... ] expressions to mlrisc tools
179    
180    ----------------------------------------------------------------------
181    Name: Matthias Blume
182    Date: 2002/02/27 12:29:00 EST
183    Tag: blume-20020227-cdebug
184    Description:
185    
186    - made types in structure C and C_Debug to be equal
187    - got rid of code duplication (c-int.sml vs. c-int-debug.sml)
188    - there no longer is a C_Int_Debug (C_Debug is directly derived from C)
189    
190    ----------------------------------------------------------------------
191    Name: Matthias Blume
192    Date: 2002/02/26 12:00:00 EST
193    Tag: blume-20020226-ffi
194    Description:
195    
196    1. Fixed a minor bug in CM's "noweb" tool:
197       If numbering is turned off, then truly don't number (i.e., do not
198       supply the -L option to noweb).  The previous behavior was to supply
199       -L'' -- which caused noweb to use the "default" line numbering scheme.
200       Thanks to Chris Richards for pointing this out (and supplying the fix).
201    
202    2. Once again, I reworked some aspects of the FFI:
203    
204       A. The incomplete/complete type business:
205    
206       - Signatures POINTER_TO_INCOMPLETE_TYPE and accompanying functors are
207         gone!
208       - ML types representing an incomplete type are now *equal* to
209         ML types representing their corresponding complete types (just like
210         in C).  This is still safe because ml-nlffigen will not generate
211         RTTI for incomplete types, nor will it generate functions that
212         require access to such RTTI.   But when ML code generated from both
213         incomplete and complete versions of the C type meet, the ML types
214         are trivially interoperable.
215    
216         NOTE:  These changes restore the full generality of the translation
217         (which was previously lost when I eliminated functorization)!
218    
219       B. Enum types:
220    
221       - Structure C now has a type constructor "enum" that is similar to
222         how the "su" constructor works.  However, "enum" is not a phantom
223         type because each "T enum" has values (and is isomorphic to
224         MLRep.Signed.int).
225       - There are generic access operations for enum objects (using
226         MLRep.Signed.int).
227       - ml-nlffigen will generate a structure E_foo for each "enum foo".
228         * The structure contains the definition of type "mlrep" (the ML-side
229         representation type of the enum).  Normally, mlrep is the same
230         as "MLRep.Signed.int", but if ml-nlffigen was invoked with "-ec",
231         then mlrep will be defined as a datatype -- thus facilitating
232         pattern matching on mlrep values.
233         ("-ec" will be suppressed if there are duplicate values in an
234          enumeration.)
235         * Constructors ("-ec") or values (no "-ec") e_xxx of type mlrep
236         will be generated for each C enum constant xxx.
237         * Conversion functions m2i and i2m convert between mlrep and
238         MLRep.Signed.int.  (Without "-ec", these functions are identities.)
239         * Coversion functions c and ml convert between mlrep and "tag enum".
240         * Access functions (get/set) fetch and store mlrep values.
241       - By default (unless ml-nlffigen was invoked with "-nocollect"), unnamed
242         enumerations are merged into one single enumeration represented by
243         structure E_'.
244    
245    ----------------------------------------------------------------------
246    Name: Allen Leung
247    Date: 2002/02/25 04:45:00 EST
248    Tag: leunga-20020225-cps-spill
249    
250    This is a new implementation of the CPS spill phase.
251    The new phase is in the new file compiler/CodeGen/cpscompile/spill-new.sml
252    In case of problems, replace it with the old file spill.sml
253    
254    The current compiler runs into some serious performance problems when
255    constructing a large record.  This can happen when we try to compile a
256    structure with many items.  Even a very simple structure like the following
257    makes the compiler slow down.
258    
259        structure Foo = struct
260           val x_1 = 0w1 : Word32.int
261           val x_2 = 0w2 : Word32.int
262           val x_3 = 0w3 : Word32.int
263           ...
264           val x_N = 0wN : Word32.int
265        end
266    
267    The following table shows the compile time, from N=1000 to N=4000,
268    with the old compiler:
269    
270    N
271    1000   CPS 100 spill                           0.04u  0.00s  0.00g
272           MLRISC ra                               0.06u  0.00s  0.05g
273              (spills = 0 reloads = 0)
274           TOTAL                                   0.63u  0.07s  0.21g
275    
276    1100   CPS 100 spill                           8.25u  0.32s  0.64g
277           MLRISC ra                               5.68u  0.59s  3.93g
278              (spills = 0 reloads = 0)
279           TOTAL                                   14.71u  0.99s  4.81g
280    
281    1500   CPS 100 spill                           58.55u  2.34s  1.74g
282           MLRISC ra                               5.54u  0.65s  3.91g
283              (spills = 543 reloads = 1082)
284           TOTAL                                   65.40u  3.13s  6.00g
285    
286    2000   CPS 100 spill                           126.69u  4.84s  3.08g
287           MLRISC ra                               0.80u  0.10s  0.55g
288              (spills = 42 reloads = 84)
289           TOTAL                                   129.42u  5.10s  4.13g
290    
291    3000   CPS 100 spill                           675.59u  19.03s  11.64g
292           MLRISC ra                               2.69u  0.27s  1.38g
293              (spills = 62 reloads = 124)
294           TOTAL                                   682.48u  19.61s  13.99g
295    
296    4000   CPS 100 spill                           2362.82u  56.28s  43.60g
297           MLRISC ra                               4.96u  0.27s  2.72g
298              (spills = 85 reloads = 170)
299           TOTAL                                   2375.26u  57.21s  48.00g
300    
301    As you can see the old cps spill module suffers from some serious
302    performance problem.  But since I cannot decipher the old code fully,
303    instead of patching the problems up, I'm reimplementing it
304    with a different algorithm.  The new code is more modular,
305    smaller when compiled, and substantially faster
306    (O(n log n) time and O(n) space).  Timing of the new spill module:
307    
308    4000  CPS 100 spill                           0.02u  0.00s  0.00g
309          MLRISC ra                               0.25u  0.02s  0.15g
310             (spills=1 reloads=3)
311          TOTAL                                   7.74u  0.34s  1.62g
312    
313    Implementation details:
314    
315    As far as I can tell, the purpose of the CPS spill module is to make sure the
316    number of live variables at any program point (the bandwidth)
317    does not exceed a certain limit, which is determined by the
318    size of the spill area.
319    
320    When the bandwidth is too large, we decrease the register pressure by
321    packing live variables into spill records.  How we achieve this is
322    completely different than what we did in the old code.
323    
324    First, there is something about the MLRiscGen code generator
325    that we should be aware of:
326    
327    o MLRiscGen performs code motion!
328    
329       In particular, it will move floating point computations and
330       address computations involving only the heap pointer to
331       their use sites (if there is only a single use).
332       What this means is that if we have a CPS record construction
333       statement
334    
335           RECORD(k,vl,w,e)
336    
337       we should never count the new record address w as live if w
338       has only one use (which is often the case).
339    
340       We should do something similar to floating point, but the transformation
341       there is much more complex, so I won't deal with that.
342    
343    Secondly, there are now two new cps primops at our disposal:
344    
345     1. rawrecord of record_kind option
346        This pure operator allocates some uninitialized storage from the heap.
347        There are two forms:
348    
349         rawrecord NONE [INT n]  allocates a tagless record of length n
350         rawrecord (SOME rk) [INT n] allocates a tagged record of length n
351                                     and initializes the tag.
352    
353     2. rawupdate of cty
354          rawupdate cty (v,i,x)
355          Assigns to x to the ith component of record v.
356          The storelist is not updated.
357    
358    We use these new primops for both spilling and increment record construction.
359    
360     1. Spilling.
361    
362        This is implemented with a linear scan algorithm (but generalized
363        to trees).  The algorithm will create a single spill record at the
364        beginning of the cps function and use rawupdate to spill to it,
365        and SELECT or SELp to reload from it.  So both spills and reloads
366        are fine-grain operations.  In contrast, in the old algorithm
367        "spills" have to be bundled together in records.
368    
369        Ideally, we should sink the spill record construction to where
370        it is needed.  We can even split the spill record into multiple ones
371        at the places where they are needed.  But CPS is not a good
372        representation for global code motion, so I'll keep it simple and
373        am not attempting this.
374    
375     2. Incremental record construction (aka record splitting).
376    
377        Long records with many component values which are simulatenously live
378        (recall that single use record addresses are not considered to
379         be live) are constructed with rawrecord and rawupdate.
380        We allocate space on the heap with rawrecord first, then gradually
381        fill it in with rawupdate.  This is the technique suggested to me
382        by Matthias.
383    
384        Some restrictions on when this is applicable:
385        1. It is not a VECTOR record.  The code generator currently does not handle
386           this case. VECTOR record uses double indirection like arrays.
387        2. All the record component values are defined in the same "basic block"
388           as the record constructor.  This is to prevent speculative
389           record construction.
390    
391    ----------------------------------------------------------------------
392    Name: Allen Leung
393    Date: 2002/02/22 01:02:00 EST
394    Tag: leunga-20020222-mlrisc-tools
395    
396    Minor bug fixes in the parser and rewriter
397    
398    ----------------------------------------------------------------------
399    Name: Allen Leung
400    Date: 2002/02/21 20:20:00 EST
401    Tag: leunga-20020221-peephole
402    
403    Regenerated the peephole files.  Some contained typos in the specification
404    and some didn't compile because of pretty printing bugs in the old version
405    of 'nowhere'.
406    
407    ----------------------------------------------------------------------
408    Name: Allen Leung
409    Date: 2002/02/19 20:20:00 EST
410    Tag: leunga-20020219-mlrisc-tools
411    Description:
412    
413       Minor bug fixes to the mlrisc-tools library:
414    
415       1.  Fixed up parsing colon suffixed keywords
416       2.  Added the ability to shut the error messages up
417       3.  Reimplemented the pretty printer and fixed up/improved
418           the pretty printing of handle and -> types.
419       4.  Fixed up generation of literal symbols in the nowhere tool.
420       5.  Added some SML keywords to to sml.sty
421    
422    ----------------------------------------------------------------------
423    Name: Matthias Blume
424    Date: 2002/02/19 16:20:00 EST
425    Tag: blume-20020219-cmffi
426    Description:
427    
428    A wild mix of changes, some minor, some major:
429    
430    * All C FFI-related libraries are now anchored under $c:
431        $/c.cm      --> $c/c.cm
432        $/c-int.cm  --> $c/internals/c-int.cm
433        $/memory.cm --> $c/memory/memory.cm
434    
435    * "make" tool (in CM) now treats its argument pathname slightly
436      differently:
437        1. If the native expansion is an absolute name, then before invoking
438           the "make" command on it, CM will apply OS.Path.mkRelative
439           (with relativeTo = OS.FileSys.getDir()) to it.
440        2. The argument will be passed through to subsequent phases of CM
441           processing without "going native".  In particular, if the argument
442           was an anchored path, then "make" will not lose track of that anchor.
443    
444    * Compiler backends now "know" their respective C calling conventions
445      instead of having to be told about it by ml-nlffigen.  This relieves
446      ml-nlffigen from one of its burdens.
447    
448    * The X86Backend has been split into X86CCallBackend and X86StdCallBackend.
449    
450    * Export C_DEBUG and C_Debug from $c/c.cm.
451    
452    * C type encoding in ml-nlffi-lib has been improved to model the conceptual
453      subtyping relationship between incomplete pointers and their complete
454      counterparts.  For this, ('t, 'c) ptr has been changed to 'o ptr --
455      with the convention of instantiating 'o with ('t, 'c) obj whenever
456      the pointer target type is complete.  In the incomplete case, 'o
457      will be instantiated with some "'c iobj" -- a type obtained by
458      using one of the functors PointerToIncompleteType or PointerToCompleteType.
459    
460      Operations that work on both incomplete and complete pointer types are
461      typed as taking an 'o ptr while operations that require the target to
462      be known are typed as taking some ('t, 'c) obj ptr.
463    
464      voidptr is now a bit "more concrete", namely "type voidptr = void ptr'"
465      where void is an eqtype without any values.  This makes it possible
466      to work on voidptr values using functions meant to operate on light
467      incomplete pointers.
468    
469    * As a result of the above, signature POINTER_TO_INCOMPLETE_TYPE has
470      been vastly simplified.
471    
472    ----------------------------------------------------------------------
473    Name: Matthias Blume
474    Date: 2002/02/19 10:48:00 EST
475    Tag: blume-20020219-pqfix
476    Description:
477    
478    Applied Chris Okasaki's bug fix for priority queues.
479    
480    ----------------------------------------------------------------------
481    Name: Matthias Blume
482    Date: 2002/02/15 17:05:00
483    Tag: Release_110_39
484    Description:
485    
486    Last-minute retagging is becoming a tradition... :-(
487    
488    This is the working release 110.39.
489    
490    ----------------------------------------------------------------------
491    Name: Matthias Blume
492    Date: 2002/02/15 16:00:00 EST
493    Tag: Release_110_39-orig
494    Description:
495    
496    Working release 110.39.  New bootfiles.
497    
498    (Update: There was a small bug in the installer so it wouldn't work
499    with all shells.  So I retagged. -Matthias)
500    
501    ----------------------------------------------------------------------
502    Name: Matthias Blume
503    Date: 2002/02/15 14:17:00 EST
504    Tag: blume-20020215-showbindings
505    Description:
506    
507    Added EnvRef.listBoundSymbols and CM.State.showBindings.  Especially
508    the latter can be useful for exploring what bindings are available at
509    the interactive prompt.  (The first function returns only the list
510    of symbols that are really bound, the second prints those but also the
511    ones that CM's autoloading mechanism knows about.)
512    
513    ----------------------------------------------------------------------
514    Name: Matthias Blume
515    Date: 2002/02/15 12:08:00 EST
516    Tag: blume-20020215-iptrs
517    Description:
518    
519    Two improvements to ml-nlffigen:
520    
521      1. Write files only if they do not exist or if their current contents
522         do not coincide with what's being written.  (That is, avoid messing
523         with the time stamps unless absolutely necessary.)
524    
525      2. Implement a "repository" mechanism for generated files related
526         to "incomplete pointer types".   See the README file for details.
527    
528    ----------------------------------------------------------------------
529    Name: Matthias Blume
530    Date: 2002/02/14 11:50:00 EST
531    Tag: blume-20020214-quote
532    Description:
533    
534    Added a type 't t_' to tag.sml (in ml-nlffi-lib.cm).  This is required
535    because of the new and improved tag generation scheme.  (Thanks to Allen
536    Leung for pointing it out.)
537    
538    ----------------------------------------------------------------------
539    Name: Lal George
540    Date: 2002/02/14 09:55:27 EST 2002
541    Tag: george-20020214-isabelle-bug
542    Description:
543    
544    Fixed the MLRISC bug sent by Markus Wenzel regarding the compilation
545    of Isabelle on the x86.
546    
547    From Allen:
548    -----------
549     I've found the problem:
550    
551         in ra-core.sml, I use the counter "blocked" to keep track of the
552         true number of elements in the freeze queue.  When the counter goes
553         to zero, I skip examining the queue.  But I've messed up the
554         bookkeeping in combine():
555    
556             else ();
557             case !ucol of
558               PSEUDO => (if !cntv > 0 then
559                     (if !cntu > 0 then blocked := !blocked - 1 else ();
560                                        ^^^^^^^^^^^^^^^^^^^^^^^
561                      moveu := mergeMoveList(!movev, !moveu)
562                     )
563                  else ();
564    
565         combine() is called to coalesce two nodes u and v.
566         I think I was thinking that if the move counts of u and v are both
567         greater than zero then after they are coalesced then one node is
568         removed from the freeze queue.  Apparently I was thinking that
569         both u and v are of low degree, but that's clearly not necessarily true.
570    
571    
572    02/12/2002:
573        Here's the patch.  HOL now compiles.
574    
575        I don't know how this impact on performance (compile
576        time or runtime).  This bug caused the RA (especially on the x86)
577        to go thru the potential spill phase when there are still nodes on the
578        freeze queue.
579    
580    
581    
582    
583    ----------------------------------------------------------------------
584    Name: Matthias Blume
585    Date: 2002/02/13 22:40:00 EST
586    Tag: blume-20020213-fptr-rtti
587    Description:
588    
589    Fixed a bug in ml-nlffigen that was introduced with one of the previous
590    updates.
591    
592    ----------------------------------------------------------------------
593    Name: Matthias Blume
594    Date: 2002/02/13 16:41:00 EST
595    Tag: blume-20020213-cmlpq
596    Description:
597    
598    Added new priority queue export symbols (which have just been added to
599    smlnj-lib.cm) to CML's version of smlnj-lib.cm.  (Otherwise CML would
600    not compile and the installer would choke.)
601    
602    ----------------------------------------------------------------------
603    Name: Matthias Blume
604    Date: 2002/02/13 16:15:00 EST
605    Tag: blume-20020213-various
606    Description:
607    
608    1. More tweaks to ml-nlffigen:
609    
610       - better internal datastructures (resulting in slight speedup)
611       - "-match" option requires exact match
612       - "localized" gensym counters (untagged structs/unions nested within
613         other structs/unions or within typedefs get a fresh counter; their
614         tag will be prefixed by a concatenation of their parents' tags)
615       - bug fixes (related to calculation of transitive closure of types
616         to be included in the output)
617    
618    2. Minor Basis updates:
619    
620       - added implementations for List.collate and Option.app
621    
622    ----------------------------------------------------------------------
623    Name: Matthias Blume
624    Date: 2002/02/11 15:55:00 EST
625    Tag: blume-20020211-gensym
626    Description:
627    
628    Added a "-gensym" option to command line of ml-nlffigen.  This can be
629    used to specify a "stem" -- a string that is inserted in all "gensym'd"
630    names (ML structure names that correspond to unnamed C structs, unions,
631    and enums), so that separate runs of ml-nlffigen do not clash.
632    
633    ----------------------------------------------------------------------
634    Name: Matthias Blume
635    Date: 2002/02/11 12:05:00 EST
636    Tag: blume-20020211-gensml
637    Description:
638    
639    A quick fix for a problem with GenSML (in the pgraph-util library):
640    Make generation of toplevel "local" optional.  (Strictly speaking,
641    signature definitions within "local" are not legal SML.)
642    
643    Other than that: updates to INSTALL and cm/TODO.
644    
645    ----------------------------------------------------------------------
646    Name: Matthias Blume
647    Date: 2002/02/08 15:00:00 EST
648    Tag: blume-20020208-uniquepid
649    Description:
650    
651    0. Version number has been bumped to 110.38.1.  NEW BOOTFILES!!!
652    
653    1. The installer (config/install.sh) has gotten smarter:
654    
655         - Configuration options are a bit easier to specify now
656           (in config/targets).
657         - Bug in recognizing .tar.bz2 files fixed.
658         - Installer automatically resolves dependencies between
659           configuration options (e.g., if you ask for eXene, you will
660           also get cml -- regardless whether you asked for it or not).
661         - Installer can run in "quieter mode" by setting the environment
662           variable INSTALL_QUIETLY to "true".  "Quieter" does not mean
663           "completely silent", though.
664         - Build HashCons library as part of smlnj-lib.
665    
666    2. A new scheme for assigning persistent identifiers to compilation
667       units (and, by extension, to types etc.) has been put into place.
668       This fixes a long-standing bug where types and even dynamic values
669       can get internally confused, thereby compromising type safety
670       (abstraction) and dynamic correctness.  See
671    
672         http://cm.bell-labs.com/cm/cs/who/blume/pid-confusion.tgz
673    
674       for an example of how things could go wrong until now.
675    
676       The downside of the new scheme is that pids are not quite as
677       persistent as they used to be: CM will generate a fresh pid
678       for every compilation unit that it thinks it sees for the first
679       time.  That means that if you compile starting from a clean, fresh
680       source tree at two different times, you end up with different
681       binaries.
682    
683       Cutoff recompilation, however, has not been compromised because
684       CM keeps pid information in special caches between runs.
685    
686    ----------------------------------------------------------------------
687    Name: Lal George
688    Date: 2002/02/07 15:34:13 EST 2002
689    Tag: <none>
690    Description:
691    
692    Compilers that generate assembly code may produce  global labels
693    whose value is resolved at link time. The various peephole optimization
694    modules did not take this in account.
695    
696    TODO. The Labels.addrOf function should really return an option
697    type so that clients are forced to deal with this issue, rather
698    than an exception being raised.
699    
700    ----------------------------------------------------------------------
701    Name: Lal George
702    Date: 2002/02/06 13:55:02 EST
703    Tag: george-20020206-ra-breakup
704    Description:
705    
706    1. A bug fix from Allen.
707    
708        A typo causes extra fstp %st(0)'s to be generated at compensation
709        edges, which might cause stack underflow traps at runtime.  This
710        occurs in fft where there are extraneous fstps right before the 'into'
711        trap instruction (in this case they are harmless since none of the
712        integers overflow.)
713    
714    2. Pulled out various utility modules that were embedded in the modules
715       of the register allocator. I need these modules for other purposes, but
716       they are not complete enough to put into a library (just yet).
717    ----------------------------------------------------------------------
718    Name: Matthias Blume
719    Date: 2002/01/31 16:05:00 EST
720    Tag: blume-20020131-sparc-ccalls
721    Description:
722    
723    1. C-calls on Sparc needlessly allocated a huge chunk (96 bytes)
724       of extra stack space by mistake.  Fixed.
725    
726    2. Bug in logic of handling of command-line options in ml-nlffigen fixed.
727    
728    ----------------------------------------------------------------------
729    Name: Allen Leung
730    Date: 2002/01/30
731    Tag: leunga-20020130-nowhere-bug-fix
732    Description:
733    
734       MLRISC bug fixes:
735       1. Fixed a bindings computation bug in the 'nowhere' program generator tool.
736       2. MachineInt.fromString was negating its value.
737    
738    ----------------------------------------------------------------------
739    Name: Matthias Blume
740    Date: 2002/01/29
741    Tag: blume-20020129-INSTALL
742    Description:
743    
744    - Added somewhat detailed installation instructions (file INSTALL).
745    - Fixed curl-detection bug in config/install.sh.
746    - It is now possible to select the URL getter using the URLGETTER
747      environment variable:
748    
749          not set / "unknown"      --> automatic detection (script tries wget,
750                                       curl, and lynx)
751          "wget" / "curl" / "lynx" --> use the specified program (script "knows"
752                                       how to properly invoke them)
753          other                    --> use $URLGETTER directly, it must take
754                                       precisely two command-line arguments
755                                       (source URL and destination file name)
756    
757    ----------------------------------------------------------------------
758    Name: Matthias Blume
759    Date: 2002/01/28
760    Tag: blume-20020128-sparc-ccalls
761    Description:
762    
763    - Fixed problem with calculation of "used" registers in sparc-c-calls.
764    - Make use of the allocParam argument in sparc-c-calls.
765    
766    ----------------------------------------------------------------------
767    Name: Matthias Blume
768    Date: 2002/01/28
769    Tag: blume-20020128-allocParam
770    Description:
771    
772    John Reppy:  Changes c-calls API to accept client-callback for
773    allocating extra stack space.
774    me: Corresponding changes to mlriscGen (using a dummy argument that
775        does not change the current behavior).
776    
777    ----------------------------------------------------------------------
778    Name: Matthias Blume
779    Date: 2002/01/28 12:00:00
780    Tag: Release_110_38
781    Description:
782    
783    This time for real!!!
784    
785    ----------------------------------------------------------------------
786    Name: Matthias Blume
787    Date: 2002/01/28 10:56:00 EST
788    Tag: blume-20020128-retraction
789    Description:
790    
791    0. Retracted earlier 110.38.  (The Release_110_38 tag has been replaced
792       with blume-Release_110_38-retracted.)
793    
794    1. Fixed a problem with incorrect rounding modes in real64.sml.
795       (Thanks to Andrew Mccreight <andrew.mccreight@yale.edu>.)
796    
797    2. A bug in ml-nlffigen related to the handling of unnamed structs, unions,
798       and enums fixed.  The naming of corresponding ML identifiers should
799       now be consistent again.
800    
801    ----------------------------------------------------------------------
802    Name: Allen Leung
803    Date: 2002/01/27
804    Tag: leunga-20020127-nowhere
805    Description:
806    
807       Added a target called nowhere in the configuration scripts.
808       Enabling this will build the MLRISC 'nowhere' tool (for translating
809       programs with where-clauses into legal SML code) during installation.
810    
811    ----------------------------------------------------------------------
812    Name: Matthias Blume
813    Date: 2002/01/25 21:27:00 EST
814    Tag: blume-Release_110_38-retracted
815    Description:
816    
817    Call it a (working) release!  Version is 110.38. Bootfiles are ready.
818    
819    README will be added later.
820    
821    !!! NOTE:  Re-tagged as blume-Release_110_38-retracted. Original tag
822    (Release_110_38) removed.  Reason: Last-minute bug fixes.
823    
824    ----------------------------------------------------------------------
825    Name: Matthias Blume
826    Date: 2002/01/25
827    Tag: blume-20020125-ffi
828    Description:
829    
830    A large number of tweaks and improvements to ml-nlffi-lib and
831    ml-nlffigen:
832    
833       - ML represenation types have been streamlined
834       - getter and setter functions work with concrete values, not abstract
835         ones where possible
836       - ml-nlffigen command line more flexible (see README file there)
837       - some bugs have been fixed (hopefully)
838    
839    ----------------------------------------------------------------------
840    Name: Lal George
841    Date: 2002/01/24
842    Tag: george-20020124-risc-ra-interface
843    Description:
844    
845       There is a dramatic simplification in the interface to the
846       register allocator for RISC architectures as a result of making
847       parallel copy instructions explicit.
848    
849    ----------------------------------------------------------------------
850    Name: Matthias Blume
851    Date: 2002/01/22
852    Tag: blume-20020122-x86-ccalls
853    Description:
854    
855    Bug fix for c-calls on x86 (having to do with how char- and
856    short-arguments are being handled).
857    
858    ----------------------------------------------------------------------
859    Name: Matthias Blume
860    Date: 2002/01/21
861    Tag: blume-20020121-ff
862    Description:
863    
864    Another day of fiddling with the FFI...
865    
866    1. Bug fix/workaround:  CKIT does not complain about negative array
867       dimensions, so ml-nlffigen has to guard itself against this possibility.
868       (Otherwise a negative dimension would send it into an infinite loop.)
869    
870    2. Some of the abstract types (light objects, light pointers, most "base"
871       types) in structure C are now eqtypes.
872    
873    3. Added constructors and test functions for NULL function pointers.
874    
875    ----------------------------------------------------------------------
876    Name: Matthias Blume
877    Date: 2002/01/18
878    Tag: blume-20020118-ready-for-new-release
879    Description:
880    
881    Made config/srcarchiveurl point to a new place.  (Will provide boot
882    files shortly.)
883    
884    Maybe we christen this to be 110.38?
885    
886    ----------------------------------------------------------------------
887    Name: Matthias Blume
888    Date: 2002/01/18
889    Tag: blume-20020118-more-ffifiddle
890    Description:
891    
892    Today's FFI fiddling:
893    
894      - Provided a structure CGetSet with "convenient" versions of C.Get.* and
895        C.Set.* that use concrete (MLRep.*) arguments and results instead
896        of abstract ones.
897    
898      - Provided word-style bit operations etc. for "int" representation
899        types in MLRep.S<Foo>Bitops where <Foo> ranges over Char, Int, Short,
900        and Long.
901    
902    ----------------------------------------------------------------------
903    Name: Matthias Blume
904    Date: 2002/01/18
905    Tag: blume-20020118-use-x86-fp
906    Description:
907    
908    Now that x86-fast-fp seems to be working, I turned it back on again
909    by default.  (Seems to work fine now, even with the FFI.)
910    
911    Other than that, I added some documentation about the FFI to
912    src/ml-nlffigen/README and updated the FFI test examples in
913    src/ml-nlffi-lib/Tests/*.
914    
915    ----------------------------------------------------------------------
916    Name: Allen Leung
917    Date: 2002/01/17
918    Tag: leunga-20020117-x86-fast-fp-call
919    Description:
920    
921       1. Fixed a problem with handling return fp values when x86's fast fp
922          mode is turned on.
923    
924       2. Minor pretty printing fix for cellset.  Print %st(0) as %st(0) instead
925          of %f32.
926    
927       3. Added a constructor INT32lit to the ast of MLRISC tools.
928    
929    ----------------------------------------------------------------------
930    Name: Matthias Blume
931    Date: 2002/01/16
932    Tag: blume-20020116-ffifiddle
933    Description:
934    
935    More fiddling with the FFI interface:
936    
937     - Make constness 'c instead of rw wherever possible.  This eliminates
938       the need for certain explicit coercions.  (However, due to ML's
939       value polymorphism, there will still be many cases where explicit
940       coercions are necessary.  Phantom types are not the whole answer
941       to modeling a subtyping relationship in ML.)
942    
943     - ro/rw coersions for pointers added.  (Avoids the detour through */&.)
944    
945     - "printf" test example added to src/ml-nlffi-lib/Tests.  (Demonstrates
946       clumsy workaround for varargs problem.)
947    
948    ----------------------------------------------------------------------
949    Name: Lal George
950    Date: 2002/01/15
951    Tag: <none>
952    Description:
953    
954    1. Since COPY instructions are no longer native to the architecture,
955       a generic functor can be used to implement the expandCopies function.
956    
957    2. Allowed EXPORT and IMPORT pseudo-op declarations to appear inside a
958       TEXT segment.
959    
960    ----------------------------------------------------------------------
961    Name: Matthias Blume
962    Date: 2002/01/15
963    Tag: blume-20020115-ffiupdates
964    Description:
965    
966    1. Fix for bug resulting in single-precision float values being returned
967       incorrectly from FFI calls.
968    
969    2. Small modifications to C FFI API:
970    
971        - memory-allocation routines return straight objects (no options)
972          and raise an exception in out-of-memory situations
973        - unsafe extensions to cast between function pointers and pointers
974          from/to ints
975        - added structure C_Debug as an alternative to structure C where
976          pointer-dereferencing (|*| and |*!) always check for null-pointers
977        - added open_lib' to DynLinkage;  open_lib' works like open_lib
978          but also takes a (possibly empty) list of existing library handles
979          that the current library depends on
980    
981    ----------------------------------------------------------------------
982    Name: Matthias Blume
983    Date: 2002/01/10
984    Tag: blume-20020110-newffigen
985    Description:
986    
987    1. Updates to portable graph code.
988    
989    2. Major update to ml-nlffigen and ml-nlffi-lib.  Things are much
990       more scalable now so that even huge interfaces such as the one
991       for GTK compile in finite time and space. :-)
992       See src/ml-nlffigen/README for details on what's new.
993    
994    ----------------------------------------------------------------------
995    Name: Lal George
996    Date: 2001/01/09 14:31:35 EST 2002
997    Tag: george-20011206-rm-native-copy
998    Description:
999    
1000            Removed the native COPY and FCOPY instructions
1001            from all the architectures and replaced it with the
1002            explicit COPY instruction from the previous commit.
1003    
1004            It is now possible to simplify many of the optimizations
1005            modules that manipulate copies. This has not been
1006            done in this change.
1007    
1008    ----------------------------------------------------------------------
1009    Name: Lal George
1010    Date: 2001/12/06 16:50:13 EST 2001
1011    Tag: george-20011206-mlrisc-instruction
1012    Description:
1013    
1014    Changed the representation of instructions from being fully abstract
1015    to being partially concrete. That is to say:
1016    
1017      from
1018            type instruction
1019    
1020      to
1021            type instr                              (* machine instruction *)
1022    
1023            datatype instruction =
1024                LIVE of {regs: C.cellset, spilled: C.cellset}
1025              | KILL of {regs: C.cellset, spilled: C.cellset}
1026              | COPYXXX of {k: CB.cellkind, dst: CB.cell list, src: CB.cell list}
1027              | ANNOTATION of {i: instruction, a: Annotations.annotation}
1028              | INSTR of instr
1029    
1030    This makes the handling of certain special instructions that appear on
1031    all architectures easier and uniform.
1032    
1033    LIVE and KILL say that a list of registers are live or killed at the
1034    program point where they appear. No spill code is generated when an
1035    element of the 'regs' field is spilled, but the register is moved to
1036    the 'spilled' (which is present, more for debugging than anything else).
1037    
1038    LIVE replaces the (now deprecated) DEFFREG instruction on the alpha.
1039    We used to generate:
1040    
1041            DEFFREG f1
1042            f1 := f2 + f3
1043            trapb
1044    
1045    but now generate:
1046    
1047            f1 := f2 + f3
1048            trapb
1049            LIVE {regs=[f1,f2,f3], spilled=[]}
1050    
1051    Furthermore, the DEFFREG (hack) required that all floating point instruction
1052    use all registers mentioned in the instruction. Therefore f1 := f2 + f3,
1053    defines f1 and uses [f1,f2,f3]! This hack is no longer required resulting
1054    in a cleaner alpha implementation. (Hopefully, intel will not get rid of
1055    this architecture).
1056    
1057    COPYXXX is intended to replace the parallel COPY and FCOPY  available on
1058    all the architectures. This will result in further simplification of the
1059    register allocator that must be aware of them for coalescing purposes, and
1060    will also simplify certain aspects of the machine description that provides
1061    callbacks related to parallel copies.
1062    
1063    ANNOTATION should be obvious, and now INSTR represents the honest to God
1064    machine instruction set!
1065    
1066    The <arch>/instructions/<arch>Instr.sml files define certain utility
1067    functions for making porting easier -- essentially converting upper case
1068    to lower case. All machine instructions (of type instr) are in upper case,
1069    and the lower case form generates an MLRISC instruction. For example on
1070    the alpha we have:
1071    
1072      datatype instr =
1073         LDA of {r:cell, b:cell, d:operand}
1074       | ...
1075    
1076      val lda : {r:cell, b:cell, d:operand} -> instruction
1077        ...
1078    
1079    where lda is just (INSTR o LDA), etc.
1080    
1081    ----------------------------------------------------------------------
1082    Name: Matthias Blume
1083    Date: 2001/11/22 21:40:00 EST
1084    Tag: Release_110_37
1085    Description:
1086    
1087    Release 110.37.  This time for real.
1088    
1089    ----------------------------------------------------------------------
1090    Name: Matthias Blume
1091    Date: 2001/11/21 16:35:00 EST
1092    Tag: blume-20011121-foot-in-mouth
1093    Description:
1094    
1095    Removed the "Release_110_37" tag because of a serious bug.
1096    This will be re-tagged once the bug is fixed.
1097    
1098    ----------------------------------------------------------------------
1099    Name: Matthias Blume
1100    Date: 2001/11/21 16:14:00 EST
1101    Tag: blume-20011121-forgottenfile
1102    Description:
1103    
1104    Forgot to add a file.  (Just a .tex-file -- part of
1105    the CM manual source.)
1106    
1107    ----------------------------------------------------------------------
1108    Name: Matthias Blume
1109    Date: 2001/11/21 16:10:00 EST
1110    Tag: blume-20011121-invalid_110_37
1111    Description:
1112    
1113    Note: I removed the original tag "Release_110_37" from this commit
1114          because we found a serious bug in all non-x86 backends.
1115          - Matthias
1116    
1117    1. Modifications to the SML/NJ code generator and to the runtime system
1118       so that code object name strings are directly inserted into code
1119       objects at code generation time.  The only business the runtime system
1120       has with this is now to read the name strings on occasions.
1121       (The encoding of the name string has also changed somewhat.)
1122    
1123    2. CM now implements a simple "set calculus" for specifying export lists.
1124       In particular, it is now possible to refer to the export lists of
1125       other libraries/groups/sources and form unions as well as differences.
1126       See the latest CM manual for details.
1127    
1128    3. An separate notion of "proxy" libraries has again be eliminated from
1129       CM's model.  (Proxy libraries are now simply a special case of using
1130       the export list calculus.)
1131    
1132    4. Some of the existing libraries now take advantage of the new set
1133       calculus.
1134       (Notice that not all libraries have been converted because some
1135       of the existing .cm-files are supposed to be backward compatible
1136       with 110.0.x.)
1137    
1138    5. Some cleanup in stand-alone programs.  (Don't use "exnMessage" -- use
1139       "General.exnMessage"!  The former relies on a certain hook to be
1140       initialized, and that often does not happen in the stand-alone case.)
1141    
1142    ----------------------------------------------------------------------
1143    Name: Lal George
1144    Date: 2001/11/21  13:56:18 EST
1145    Tag: george-2001121-pseudo-ops
1146    Description:
1147    
1148      Implemented a complete redesign of MLRISC pseudo-ops. Now there
1149      ought to never be any question of incompatabilities with
1150      pseudo-op syntax expected by host assemblers.
1151    
1152      For now, only modules supporting GAS syntax are implemented
1153      but more should follow, such as MASM, and vendor assembler
1154      syntax, e.g. IBM as, Sun as, etc.
1155    
1156    ----------------------------------------------------------------------
1157    Name: Matthias Blume
1158    Date: 2001/11/14 11:52:00 EST
1159    Tag: blume-20011114-srcname
1160    Description:
1161    
1162    1. Routed the name of the current source file to mlriscgen where it
1163       should be directly emitted into the code object.  (This last part
1164       is yet to be done.)
1165    
1166    2. Some cleanup of the pgraph code to make it match the proposal that
1167       I put out the other day.  (The proposal notwithstanding, things are
1168       still in flux here.)
1169    
1170    ----------------------------------------------------------------------
1171    Name: Lal George
1172    Date: 2001/11/14 09:44:04 EST
1173    Tag:
1174    Description:
1175    
1176      Fix for a backpatching bug reported by Allen.
1177    
1178      Because the boundary between short and long span-dependent
1179      instructions is +/- 128, there are an astounding number of
1180      span-dependent instructions whose size is over estimated.
1181    
1182      Allen came up with the idea of letting the size of span
1183      dependent instructions be non-monotonic, for a maxIter
1184      number of times, after which the size must be monotonically
1185      increasing.
1186    
1187      This table shows the number of span-dependent instructions
1188      whose size was over-estimated as a function of maxIter, for the
1189      file Parse/parse/ml.grm.sml:
1190    
1191         maxIter            # of instructions:
1192            10                      687
1193            20                      438
1194            30                      198
1195            40                        0
1196    
1197      In compiling the compiler, there is no significant difference in
1198      compilation speed between maxIter=10 and maxIter=40. Actually,
1199      my measurements showed that maxIter=40 was a tad faster than
1200      maxIter=10! Also 96% of the  files in the compiler reach a fix
1201      point within 13 iterations, so fixing maxIter at 40, while high,
1202      is okay.
1203    
1204    ----------------------------------------------------------------------
1205    Name: Matthias Blume
1206    Date: 2001/10/31 15:25:00 EST
1207    Tag: blume-20011031-pgraph
1208    Description:
1209    
1210    CKIT:
1211    * Changed the "Function" constructor of type Ast.ctype to carry optional
1212      argument identifiers.
1213    * Changed the return type of TypeUtil.getFunction accordingly.
1214    * Type equality ignores the argument names.
1215    * TypeUtil.composite tries to preserve argument names but gives up quickly
1216      if there is a mismatch.
1217    
1218    installation script:
1219    * attempts to use "curl" if available (unless "wget" is available as well)
1220    
1221    CM:
1222    * has an experimental implementation of "portable graphs" which I will
1223      soon propose as an implementation-independent library format
1224    * there are also new libraries $/pgraph.cm and $/pgraph-util.cm
1225    
1226    NLFFI-LIB:
1227    * some cleanup (all cosmetic)
1228    
1229    NLFFIGEN:
1230    * temporarily disabled the mechanism that suppresses ML output for
1231      C definitions whose identifiers start with an underscore character
1232    * generate val bindings for enum constants
1233    * user can request that only one style (light or heavy) is being used;
1234      default is to use both (command-line arguments: -heavy and -light)
1235    * fixed bug in handling of function types involving incomplete pointers
1236    * generate ML entry points that take record arguments (i.e., using
1237      named arguments) for C functions that have a prototype with named
1238      arguments
1239      (see changes to CKIT)
1240    
1241    ----------------------------------------------------------------------
1242    Name: Allen Leung
1243    Date: 2001/10/27 20:34:00 EDT
1244    Tag: leunga-20011027-x86-fast-fp-call
1245    Description:
1246    
1247       Fixed the bug described in blume-20010920-slowfp.
1248    
1249       The fix involves
1250          1. generating FCOPYs in FSTP in ia32-svid
1251          2. marking a CALL with the appropriate annotation
1252    
1253    ----------------------------------------------------------------------
1254    Name: Matthias Blume
1255    Date: 2001/10/16 11:32:00 EDT
1256    Tag: blume-20011016-netbsd
1257    Description:
1258    
1259    Underscore patch from Chris Richards (fixing problem with compiling
1260    runtime system under recent NetBSD).
1261    
1262    ----------------------------------------------------------------------
1263    Name: Allen Leung
1264    Date: 2001/10/12 17:18:32 EDT 2001
1265    Tag: leung-20011012-x86-printflowgraph
1266    Description:
1267    
1268    X86RA now uses a valid (instead of dummy) PrintFlowgraph module.
1269    
1270    ----------------------------------------------------------------------
1271    Name: Lal George
1272    Date: 2001/10/11 23:51:34 EDT
1273    Tag: george-20011011-too-many-instrs
1274    Description:
1275    
1276    The representation of a program point never expected to see more
1277    than 65536 instructions in a basic block!
1278    
1279    ----------------------------------------------------------------------
1280    Name: Lal George
1281    Date: 2001/10/09 09:41:37 EDT
1282    Tag: george-20011008-mlrisc-labels
1283    Description:
1284    
1285    Changed the machine description files to support printing of
1286    local and global labels in assembly code, based on host assembler
1287    conventions.
1288    
1289    ----------------------------------------------------------------------
1290    Name: Matthias Blume
1291    Date: 2001/09/25 15:25:00 EDT
1292    Tag: blume-20010925-exninfo
1293    Description:
1294    
1295    I provided a non-hook implementation of exnName (at the toplevel) and
1296    made the "dummy" implementation of exnMessage (at the toplevel) more
1297    useful: if nothing gets "hooked in", then at least you are going to
1298    see the exception name and a message indicating why you don't see more.
1299    
1300    [For the time being, programs that need exnMessage and want to use
1301    ml-build should either use General.exnMessage (strongly recommended) or
1302    refer to structure General at some other point so that CM sees a
1303    static dependency.]
1304    
1305    [Similar remarks go for "print" and "use":  If you want to use their
1306    functionality in stand-alone programs generated by ml-build, then use
1307    TextIO.output and Backend.Interact.useFile (from $smlnj/compiler.cm).]
1308    
1309    ----------------------------------------------------------------------
1310    Name: Matthias Blume
1311    Date: 2001/09/20 17:28:00 EDT
1312    Tag: blume-20010920-slowfp
1313    Description:
1314    
1315    Allen says that x86-fast-fp is not safe yet, so I turned it off again...
1316    
1317    ----------------------------------------------------------------------
1318    Name: Matthias Blume
1319    Date: 2001/09/20 17:20:00 EDT
1320    Tag: blume-20010920-canonicalpaths
1321    Description:
1322    
1323    0. Updated the BOOT file (something that I forgot to do earlier).
1324    
1325    1. Small internal change to CM so that it avoids "/../" in filenames
1326       as much as possible (but only where it is safe).
1327    
1328    2. Changed config/_run-sml (resulting in a changed bin/.run-sml) so
1329       that arguments that contain delimiters are passed through correctly.
1330       This change also means that all "special" arguments of the form
1331       @SMLxxx... must come first.
1332    
1333    3. Changed install script to put relative anchor names for tool commands
1334       into pathconfig.
1335    
1336    ----------------------------------------------------------------------
1337    Name: Matthias Blume
1338    Date: 2001/09/18 15:35:00 EDT
1339    Tag: blume-20010918-readme11036
1340    Description:
1341    
1342    Added README files.
1343    
1344    ----------------------------------------------------------------------
1345    Name: Matthias Blume
1346    Date: 2001/09/18 11:45:00 EDT
1347    Tag: Release_110_36 (retag)
1348    Description:
1349    
1350    Fixed mistake in config/preloads. Retagged as 110.36.
1351    
1352    ----------------------------------------------------------------------
1353    Name: Matthias Blume
1354    Date: 2001/09/18 09:40:00 EDT
1355    Tag: Release_110_36_orig (tag changed)
1356    Description:
1357    
1358    New version (110.36).  New bootfiles.
1359    
1360    ----------------------------------------------------------------------
1361    Name: Matthias Blume
1362    Date: 2001/09/14 16:15:00 EDT
1363    Tag: blume-20010914-x86fastfp
1364    Description:
1365    
1366    John committed some changes that Allen made, in particular a (hopefully)
1367    correctly working version of the x86-fp module.
1368    
1369    I changed the default setting of the Control.MLRISC.getFlag "x86-fast-fp"
1370    flag to "true".  Everything seems to compile to a fixpoint ok, and
1371    "mandelbrot" speeds up by about 15%.
1372    
1373    ----------------------------------------------------------------------
1374    Name: Matthias Blume
1375    Date: 2001/09/13 11:20:00 EDT
1376    Tag: blume-20010913-minimal
1377    Description:
1378    
1379    1. Stefan Monnier's patch to fix a miscompilation problem that
1380       was brought to light by John Reppy's work on Moby.
1381    
1382    2. Implemented a minimal "structure Compiler" that contains just
1383       "version" and "architecture".  The minimal version will be
1384       available when the full version is not.  This is for backward-
1385       compatibility with code that wants to test Compiler.version.
1386    
1387    ----------------------------------------------------------------------
1388    Name: Matthias Blume
1389    Date: 2001/08/28 14:03:00 EDT
1390    Tag: blume-20010828-ml-lex
1391    Description:
1392    
1393    Fix for bug 1581, received from Neophytos Michael.
1394    
1395    ----------------------------------------------------------------------
1396    Name: Matthias Blume
1397    Date: 2001/08/27 11:20:00 EDT
1398    Tag: blume-20010827-readme11035
1399    Description:
1400    
1401    Fleshed out the README file for 110.35.
1402    
1403    ----------------------------------------------------------------------
1404    Name: Matthias Blume
1405    Date: 2001/08/24 17:10:00 EDT
1406    Tag: Release_110_35
1407    Description:
1408    
1409    New version number (110.35).  New bootfiles.
1410    
1411    ----------------------------------------------------------------------
1412    Name: Lal George
1413    Date: 2001/08/24 13:47:18 EDT 2001
1414    Tag: george-20010824-MLRISC-graphs
1415    Description:
1416    
1417     removed clusters from MLRISC completely and replaced with graphs.
1418    
1419    ----------------------------------------------------------------------
1420    Name: Matthias Blume
1421    Date: 2001/08/23 17:50:00 EDT
1422    Tag: blume-20010823-toplevel
1423    Description:
1424    
1425    - some reorganization of the code that implements various kinds of
1426      environments in the compiler (static, dynamic, symbolic, combined)
1427    - re-implemented the EnvRef module so that evalStream works properly
1428      (if the stream contains references to "use", "CM.make", etc.)
1429    - cleaned up evalloop.sml and interact.sml (but they need more cleaning)
1430    
1431    ----------------------------------------------------------------------
1432    Name: Matthias Blume
1433    Date: 2001/08/20 15:50 EDT
1434    Tag: blume20010820-slipup
1435    Description:
1436    
1437    I forgot to commit a few files.  Here they are...
1438    
1439    ----------------------------------------------------------------------
1440    Name: Matthias Blume
1441    Date: 2001/08/20 15:35:00 EDT
1442    Tag: blume-20010820-debugprof
1443    Description:
1444    
1445    !!!! NEW BOOTFILES !!!!
1446    
1447    This is another round of reorganizing the compiler sources.  This
1448    time the main goal was to factor out all the "instrumentation"
1449    passes (for profiling and backtracing) into their own library.
1450    The difficulty was to do it in such a way that it does not depend
1451    on elaborate.cm but only on elabdata.cm.
1452    
1453    Therefore there have been further changes to both elaborate.cm and
1454    elabdata.cm -- more "generic" things have been moved from the former
1455    to the latter.  As a result, I was forced to split the assignment
1456    of numbers indicating "primtyc"s into two portions: SML-generic and
1457    SML/NJ-specific.  Since it would have been awkward to maintain,
1458    I bit the bullet and actually _changed_ the mapping between these
1459    numbers and primtycs.  The bottom line of this is that you need
1460    a new set of bin- and bootfiles.
1461    
1462    I have built new bootfiles for all architectures, so doing a fresh
1463    checkout and config/install.sh should be all you need.
1464    
1465    The newly created library's name is
1466    
1467        $smlnj/viscomp/debugprof.cm
1468    
1469    and its sources live under
1470    
1471        src/compiler/DebugProf
1472    
1473    ----------------------------------------------------------------------
1474    Name: Matthias Blume
1475    Date: 2001/08/15 17:15:00 EDT
1476    Tag: blume-20010815-compreorg
1477    Description:
1478    
1479    This is a first cut at reorganizing the CM libraries that make up the
1480    core of the compiler.  The idea is to separate out pieces that could
1481    be used independently by tools, e.g., the parser, the typechecker, etc.
1482    
1483    The current status is a step in this direction, but it is not quite
1484    satisfactory yet.  Expect more changes in the future.
1485    
1486    Here is the current (new) organization...
1487    
1488        What used to be $smlnj/viscomp/core.cm is now divided into
1489        six CM libraries:
1490    
1491             $smlnj/viscomp/basics.cm
1492                           /parser.cm
1493                           /elabdata.cm
1494                           /elaborate.cm
1495                           /execute.cm
1496                           /core.cm
1497    
1498        The CM files for these libraries live under src/system/smlnj/viscomp.
1499        All these libraries are proxy libraries that contain precisely
1500        one CM library component.  Here are the locations of the components
1501        (all within the src/compiler tree):
1502    
1503             Basics/basics.cm
1504             Parse/parser.cm
1505             ElabData/elabdata.cm
1506             Elaborator/elaborate.cm
1507             Execution/execute.cm
1508             core.cm
1509    
1510         [This organization is the same that has been used already
1511         for a while for the architecture-specific parts of the visible
1512         compiler and for the old version of core.cm.]
1513    
1514         As you will notice, many source files have been moved from their
1515         respective original locations to a new home in one of the above
1516         subtrees.
1517    
1518         The division of labor between the new libraries is the following:
1519    
1520             basics.cm:
1521                - Simple, basic definitions that pertain to many (or all) of
1522                  the other libraries.
1523             parser.cm:
1524                - The SML parser, producing output of type Ast.dec.
1525                - The type family for Ast is also defined and exported here.
1526             elabdata.cm:
1527                - The datatypes that describe input and output of the elaborator.
1528                  This includes types, absyn, and static environments.
1529             elaborator.cm:
1530                - The SML/NJ type checker and elaborator.
1531                  This maps an Ast.dec (with a given static environment) to
1532                  an Absyn.dec (with a new static environment).
1533                - This libraries implements certain modules that used to be
1534                  structures as functors (to remove dependencies on FLINT).
1535             execute.cm:
1536                - Everything having to do with executing binary code objects.
1537                - Dynamic environments.
1538             core.cm:
1539                - SML/NJ-specific instantiations of the elaborator and MLRISC.
1540                - Top-level modules.
1541                - FLINT (this should eventually become its own library)
1542    
1543    Notes:
1544    
1545    I am not 100% happy with the way I separated the elaborator (and its
1546    data structures) from FLINT.  Two instances of the same problem:
1547    
1548        1. Data structures contain certain fields that carry FLINT-specific
1549           information.  I hacked around this using exn and the property list
1550           module from smlnj-lib.  But the fact that there are middle-end
1551           specific fields around at all is a bit annoying.
1552    
1553        2. The elaborator calculates certain FLINT-related information.  I tried
1554           to make this as abstract as I could using functorization, but, again,
1555           the fact that the elaborator has to perform calculations on behalf
1556           of the middle-end at all is not nice.
1557    
1558        3. Having to used exn and property lists is unfortunate because it
1559           weakens type checking.  The other alternative (parameterizing
1560           nearly *everything*) is not appealing, though.
1561    
1562    I removed the "rebinding =" warning hack because due to the new organization
1563    it was awkward to maintain it.  As a result, the compiler now issues some of
1564    these warnings when compiling init.cmi during bootstrap compilation. On
1565    the plus side, you also get a warning when you do, for example:
1566       val op = = Int32.+
1567    which was not the case up to now.
1568    
1569    I placed "assign" and "deref" into the _Core structure so that the
1570    code that deals with the "lazy" keyword can find them there.  This
1571    removes the need for having access to the primitive environment
1572    during elaboration.
1573    
1574    ----------------------------------------------------------------------
1575    Name: Matthias Blume
1576    Date: 2001/08/13
1577    Tag: blume-20010813-closures
1578    Description:
1579    
1580    This fix was sent to us by Zhong Shao.  It is supposed to improve the
1581    performance of certain loops by avoiding needless closure allocation.
1582    
1583    ----------------------------------------------------------------------
1584    Name: Lal George
1585    Date: 2001/07/31 10:03:23 EDT 2001
1586    Tag: george-20010731-x86-fmalloc
1587    Description: Fixed bug in x86 calls
1588    
1589        There was a bug where call instructions would mysteriously
1590        vanish. The call instruction had to be one that returned
1591        a floating point value.
1592    
1593    ----------------------------------------------------------------------
1594    Name: Lal George
1595    Date: 2001/07/19 16:36:29 EDT 2001
1596    Tag: george-20010719-simple-cells
1597    Description:
1598    
1599    I have dramatically simplified the interface for CELLS in MLRISC.
1600    
1601    In summary, the cells interface is broken up into three parts:
1602    
1603      1. CellsBasis : CELLS_BASIS
1604    
1605            CellsBasis is a top level structure and common for all
1606            architectures.  it contains the definitions of basic datatypes
1607            and utility  functions over these types.
1608    
1609      2. functor Cells() : CELLS
1610    
1611            Cells generates an interface for CELLS that incorporates the
1612            specific resources on the target architecture, such as the
1613            presence of special register classes, their number and size,
1614            and various useful substructures.
1615    
1616      3. <ARCH>CELLS
1617    
1618            e.g. SparcCells: SPARCCELLS
1619    
1620            <ARCH>CELLS usually contains additional bindings for special
1621            registers  on the architecture, such as:
1622    
1623                    val r0 : cell           (* register zero *)
1624                    val y : cell            (* Y register *)
1625                    val psr : cell          (* processor status register *)
1626                    ...
1627    
1628            The structure returned by applying the Cells functor is opened
1629            in this interface.
1630    
1631    The main implication of all this is that the datatypes for cells is
1632    split between CellsBasis and CELLS -- a fairly simple change for user
1633    code.
1634    
1635    In the old scheme the CELLS interface had a definitional binding of
1636    the form:
1637    
1638            signature CELLS = sig
1639    
1640               structure CellsBasis = CellsBasis
1641    
1642               ...
1643    
1644            end
1645    
1646    With all the sharing constraints that goes on in MLRISC, this old
1647    design  quickly leads to errors such as:
1648    
1649            "structure definition spec inside of sharing ... "
1650    
1651    
1652    and appears to require an unacceptable amount of sharing and where
1653    constraint hackery.
1654    
1655    I think this error message (the interaction of definitional specs and
1656    sharing) requires more explanation on our web page.
1657    
1658    ----------------------------------------------------------------------
1659    Name: Matthias Blume
1660    Date: 2001/07/19 15:00:00 EDT
1661    Tag: blume-20010719-libreorg
1662    Description:
1663    
1664    This update puts together a fairly extensive but straightforward change
1665    to the way the libraries that implement the interactive system are
1666    organized:
1667    
1668       The biggest change is the elimination of structure Compiler.  As a
1669       replacement for this structure, there is now a CM library
1670       (known as $smlnj/compiler.cm or $smlnj/compiler/current.cm)
1671       that exports all the substructures of the original structure Compiler
1672       directly.  So instead of saying Compiler.Foo.bar one now simply
1673       says Foo.bar.  (The CM libraries actually export a collection of
1674       structures that is richer than the collection of substructures of
1675       structure Compiler.)
1676    
1677       To make the transition smooth, there is a separate library called
1678       $smlnj/compiler/compiler.cm which puts together and exports the
1679       original structure Compiler (or at least something very close to it).
1680    
1681       There are five members of the original structure Compiler
1682       that are not exported directly but which instead became members
1683       of a new structure Backend (described by signature BACKEND).  These are:
1684       structure Profile (: PROFILE), structure Compile (: COMPILE), structure
1685       Interact (: INTERACT), structure Machine (: MACHINE), and val
1686       architecture (: string).
1687    
1688       Structure Compiler.Version has become structure CompilerVersion.
1689    
1690       Cross-compilers for alpha32, hppa, ppc, sparc, and x86 are provided
1691       by $smlnj/compiler/<arch>.cm where <arch> is alpha32, hppa, ppc, sparc,
1692       or x86, respectively.
1693       Each of these exports the same frontend structures that
1694       $smlnj/compiler.cm exports.  But they do not have a structure Backend
1695       and instead export some structure <Arch>Backend where <Arch> is Alpha32,
1696       Hppa, PPC, Sparc, or X86, respectively.
1697    
1698       Library $smlnj/compiler/all.cm exports the union of the exports of
1699       $smlnj/compiler/<arch>.cm
1700    
1701       There are no structures <Arch>Compiler anymore, use
1702       $smlnj/compiler/<arch>.cm instead.
1703    
1704       Library host-compiler-0.cm is gone.  Instead, the internal library
1705       that instantiates CM is now called cm0.cm.  Selection of the host
1706       compiler (backend) is no longer done here but. (Responsibility for it
1707       now lies with $smlnj/compiler/current.cm.  This seems to be more
1708       logical.)
1709    
1710       Many individual files have been moved or renamed.  Some files have
1711       been split into multiple files, and some "dead" files have been deleted.
1712    
1713    Aside from these changes to library organization, there are also changes
1714    to the way the code itself is organized:
1715    
1716       Structure Binfile has been re-implemented in such a way that it no
1717       longer needs any knowledge of the compiler.  It exclusively deals
1718       with the details of binfile layout.  It no longer invokes the
1719       compiler (for the purpose of creating new prospective binfile
1720       content), and it no longer has any knowledge of how to interpret
1721       pickles.
1722    
1723       Structure Compile (: COMPILE) has been stripped down to the bare
1724       essentials of compilation.  It no longer deals with linking/execution.
1725       The interface has been cleaned up considerably.
1726    
1727       Utility routines for dealing with linking and execution have been
1728       moved into their own substructures.
1729    
1730       (The ultimate goal of these changes is to provide a light-weight
1731       binfile loader/linker (at least for, e.g., stable libraries) that
1732       does not require CM or the compiler to be present.)
1733    
1734    CM documentation has been updated to reflect the changes to library
1735    organization.
1736    
1737    ----------------------------------------------------------------------
1738    Name: Matthias Blume
1739    Date: 2001/07/10 17:30:00 EDT
1740    Tag: Release_110_34
1741    Description:
1742    
1743    Minor tweak to 110.34 (re-tagged):
1744    
1745      - README.html file added to CVS repository
1746      - runtime compiles properly under FreeBSD 3.X and 4.X
1747    
1748    ----------------------------------------------------------------------
1749    Name: Matthias Blume
1750    Date: 2001/07/10 17:30:00 EDT
1751    Tag: Release_110_34
1752    Description:
1753    
1754    New version number (110.34). New bootfiles.
1755    
1756    ----------------------------------------------------------------------
1757    Name: Matthias Blume
1758    Date: 2001/07/09 16:00:00 EDT
1759    Tag: blume-20010709-more-varargs
1760    Description:
1761    
1762    I changed the handling of varargs in ml-nlffigen again:
1763    The ellipsis ... will now simply be ignored (with an accompanying warning).
1764    
1765    The immediate effect is that you can actually call a varargs function
1766    from ML -- but you can't actually supply any arguments beyond the ones
1767    specified explicitly.  (For example, you can call printf with its format
1768    string, but you cannot pass additional arguments.)
1769    
1770    This behavior is only marginally more useful than the one before, but
1771    it has the advantage that a function or, more importantly, a function
1772    type never gets dropped on the floor, thus avoiding follow-up problems with
1773    other types that refer to the offending one.
1774    
1775    ----------------------------------------------------------------------
1776    Name: Matthias Blume
1777    Date: 2001/07/09 11:25:00 EDT
1778    Tag: blume-20010709-varargs
1779    Description:
1780    
1781    1. ckit-lib.cm now exports structure Error
1782    2. ml-nlffigen reports occurences of "..." (i.e., varargs function types)
1783       with a warning accompanied by a source location.  Moreover, it
1784       merely skips the offending function or type and proceeds with the
1785       rest of its work.u  As a result, one can safely feed C code containing
1786       "..." to ml-nlffigen.
1787    3. There are some internal improvements to CM, providing slightly
1788       more general string substitutions in the tools subsystem.
1789    
1790    ----------------------------------------------------------------------
1791    Name: Matthias Blume
1792    Date: 2001/06/27 15:10:00 EDT
1793    Tag: blume-20010627-concur
1794    Description:
1795    
1796    Fixed a small bug in CM's handling of parallel compilation.
1797    (You could observe the bug by Control-C-interrupting an ordinary
1798    CMB.make or CM.stabilize and then attaching some compile servers.
1799    The result was that all of a sudden the previously interrupted
1800    compilation would continue on its own.  This was because of
1801    an over-optimization: CM did not bother to clean out certain queues
1802    when no servers were attached "anyway", resulting in the contents
1803    of these queues to grab control when new servers did get attached.)
1804    
1805    There is also another minor update to the CM manual.
1806    
1807    ----------------------------------------------------------------------
1808    Name: Matthias Blume
1809    Date: 2001/06/26 16:15:00 EDT
1810    Tag: blume-20010626-cmdoc
1811    Description:
1812    
1813    Minor typo fixed in CM manual (syntax diagram for libraries).
1814    
1815    ----------------------------------------------------------------------
1816    Name: Matthias Blume
1817    Date: 2001/06/25 22:55:00 EDT
1818    Tag: blume-20010625-x86pc
1819    Description:
1820    
1821    Fixed a nasty bug in the X86 assembly code that caused signal
1822    handlers to fail (crash) randomly.
1823    
1824    ----------------------------------------------------------------------
1825    Name: Matthias Blume
1826    Date: 2001/06/25 12:05:00 EDT
1827    Tag: blume-20010625-nlffigen
1828    Description:
1829    
1830    This update fixes a number of minor bugs in ml-nlffigen as reported by
1831    Nick Carter <nbc@andrew.cmu.edu>.
1832    
1833      1. Silly but ok typedefs of the form "typedef void myvoid;" are now accepted.
1834      2. Default names for generated files are now derived from the name of
1835         the C file *without its directory*.  In particular, this causes generated
1836         files to be placed locally even if the C file is in some system directory.
1837      3. Default names for generated signatures and structures are also derived
1838         from the C file name without its directory.  This avoids silly things
1839         like "structure GL/GL".
1840         (Other silly names are still possible because ml-nlffigen does not do
1841          a thorough check of whether generated names are legal ML identifiers.
1842          When in doubt, use command line arguments to force particular names.)
1843    
1844    ----------------------------------------------------------------------
1845    Name: Matthias Blume
1846    Date: 2001/06/21 12:25:00 EDT
1847    Tag: blume-20010621-eXene
1848    Description:
1849    
1850    eXene now compiles and (sort of) works again.
1851    
1852    The library name (for version > 110.33) is $/eXene.cm.
1853    
1854    I also added an new example in src/eXene/examples/nbody.  See the
1855    README file there for details.
1856    
1857    ----------------------------------------------------------------------
1858    Name: Matthias Blume
1859    Date: 2001/06/20 16:40:00 EDT
1860    Tag: blume-20010620-cml
1861    Description:
1862    
1863    CML now compiles and works again.
1864    
1865    Libraries (for version > 110.33):
1866    
1867      $cml/cml.cm            Main CML library.
1868      $cml/basis.cm          CML's version of $/basis.cm.
1869      $cml/cml-internal.cm   Internal helper library.
1870      $cml/core-cml.cm       Internal helper library.
1871      $cml-lib/trace-cml.cm  Tracing facility.
1872      $cml-lib/smlnj-lib.cm  CML's version of $/smlnj-lib.cm
1873    
1874    The installer (config/install.sh) has been taught how to properly
1875    install this stuff.
1876    
1877    ----------------------------------------------------------------------
1878    Name: Matthias Blume
1879    Date: 2001/06/19 17:55:00 EDT
1880    Tag: blume-20010619-instantiate
1881    Description:
1882    
1883    This un-breaks the fix for bug 1432.
1884    (The bug was originally fixed in 110.9 but I broke it again some
1885    time after that.)
1886    
1887    ----------------------------------------------------------------------
1888    Name: Matthias Blume
1889    Date: 2001/06/19 17:25:00 EDT
1890    Tag: blume-20010619-signals
1891    Description:
1892    
1893    This should (hopefully) fix the long-standing signal handling bug.
1894    (The runtime system was constructing a continuation record with an
1895    incorrect descriptor which would cause the GC to drop data on the floor...)
1896    
1897    ----------------------------------------------------------------------
1898    Name: Matthias Blume
1899    Date: 2001/06/15 15:05:00 EDT
1900    Tag: blume-20010615-moresparc
1901    Description:
1902    
1903    Here is a short late-hour update related to Sparc c-calls:
1904    
1905     -- made handling of double-word arguments a bit smarter
1906    
1907     -- instruction selection phase tries to collapse certain clumsily
1908        constructed ML-Trees; typical example:
1909    
1910            ADD(ty,ADD(_,e,LI d1),LI d2)  ->  ADD(ty,e,LI(d1+d2))
1911    
1912        This currently has no further impact on SML/NJ since mlriscGen does
1913        not seem to generate such patterns in the first place, and c-calls
1914        (which did generate them in the beginning) has meanwhile been fixed
1915        so as to avoid them as well.
1916    
1917    ----------------------------------------------------------------------
1918    Name: Matthias Blume
1919    Date: 2001/06/15 15:05:00 EDT
1920    Tag: blume-20010615-sparc
1921    Description:
1922    
1923    The purpose of this update is to provide an implementation of NLFFI
1924    on Sparc machines.
1925    
1926    Here are the changes in detail:
1927    
1928       * src/MLRISC/sparc/c-calls/sparc-c-calls.sml is a new file containing
1929       the Sparc implementation of the c-calls API.
1930       * The Sparc backend of SML/NJ has been modified to uniformely use %fp
1931       for accessing the ML frame.  Thus, we have a real frame pointer and
1932       can freely modify %sp without need for an omit-frame-ptr phase.
1933       The vfp logic in src/compiler/CodeGen/* has been changed to accomodate
1934       this case.
1935       * ml-nlffigen has been taught to produce code for different architectures
1936       and calling conventions.
1937       * In a way similar to what was done in the x86 case, the Sparc
1938       backend uses its own specific extension to mltree.  (For example,
1939       it needs to be able to generate UNIMP instructions which are part
1940       of the calling convention.)
1941       * ml-nlffi-lib was reorganized to make it more modular (in particular,
1942       to make it easier to plug in new machine- and os-dependent parts).
1943    
1944    There are some other fairly unrelated bug fixes and cleanups as well:
1945    
1946       * I further hacked the .cm files for MLRISC tools (like MDLGen) so
1947       that they properly share their libraries with existing SML/NJ libraries.
1948       * I fixed a minor cosmetic bug in CM, supressing certain spurious
1949       follow-up error messages.
1950       * Updates to CM/CMB documentation.
1951    
1952    TODO items:
1953    
1954       * MLRISC should use a different register as its asmTemp on the Sparc.
1955         (The current %o2 is a really bad choice because it is part of the
1956         calling conventions, so things might interfere in unexpected ways.)
1957    
1958    ----------------------------------------------------------------------
1959    Name: Matthias Blume
1960    Date: 2001/06/07
1961    Tag: blume-20010607-calls
1962    Description:
1963    
1964    A number of internal changes related to C calls and calling conventions:
1965    
1966    1. ML-Tree CALL statements now carry a "pops" field.  It indicates the
1967       number of bytes popped implicitly (by the callee).  In most cases
1968       this field is 0 but on x86/win32 it is some non-zero value.  This
1969       is information provided for the benefit of the "omit-frameptr" pass.
1970    2. The CALL instruction on the x86 carries a similar "pops" field.
1971       The instruction selection phase copies its value from the ML-Tree
1972       CALL statement.
1973    3. On all other architectures, the instruction selection phase checks
1974       whether "pops=0" and complains if not.
1975    4. The c-calls implementation for x86 now accepts two calling conventions:
1976       "ccall" and "stdcall".  When "ccall" is selected, the caller cleans
1977       up after the call and pops is set to 0.  For "stdcall", the caller
1978       does nothing, leaving the cleanup to the callee; pops is set to
1979       the number of bytes that were pushed onto the stack.
1980    5. The cproto decoder (compiler/Semant/types/cproto.sml) now can
1981       distinguish between "ccall" and "stdcall".
1982    6. The UNIMP instruction has been added to the supported Sparc instruction
1983       set. (This is needed for implementing the official C calling convention
1984       on this architecture.)
1985    7. I fixed some of the .cm files under src/MLRISC/Tools to make them
1986       work with the latest CM.
1987    
1988    ----------------------------------------------------------------------
1989    Name: Matthias Blume
1990    Date: 2001/06/05 15:10:00 EDT
1991    Tag: blume-20010605-cm-index
1992    Description:
1993    
1994    0. The "lambdasplit" parameter for class "sml" in CM has been documented.
1995    
1996    1. CM can now generate "index files".  These are human-readable files
1997       that list on a per-.cm-file basis each toplevel symbol defined or
1998       imported.  The location of the index file for
1999       <p>/<d>.cm is <p>/CM/INDEX/<d>.cm.
2000       To enable index-file generation, set CM.Control.generate_index to true
2001       or export an environment-symbol: export CM_GENERATE_INDEX=true.
2002    
2003       The CM manual has been updated accordingly.
2004    
2005    2. I made some slight modifications to the c-calls API in MLRISC.
2006    
2007         a) There is now a callback to support saving/restoring of
2008            dedicated but caller-save registers around the actual call
2009            instruction.
2010         b) One can optionally specify a comment-annotation for the
2011            call instruction.
2012    
2013    3. SML/NJ (mlriscGen.sml) uses this new API for the rawccall primop.
2014       (For example, the comment annotation shows the C prototype of
2015        the function being called.)
2016    
2017    ----------------------------------------------------------------------
2018    Name: Matthias Blume
2019    Date: 2001/06/01 13:30:00 EDT
2020    Tag: blume-20010601-nlffi-cleanup
2021    Description:
2022    
2023    This is mostly a cleanup of MLFFI stuff:
2024    
2025       - some signature files have been put into a more exposed place
2026       - the ugly 'f type parameter is gone (simplifies types tremendously!)
2027       - ml-nlffigen changed accordingly
2028       - tutorial updated
2029    
2030    Other changes:
2031    
2032       - author's affiliation in CM manual(s) updated
2033       - some more recognized keywords added to Allen's sml.sty
2034    
2035    ----------------------------------------------------------------------
2036    Name: Matthias Blume
2037    Date: 2001/05/25 15:30:00 EDT
2038    Tag: blume-20010525-iptr
2039    Description:
2040    
2041      - put the official 110.33-README (as it appears on the ftp server) under
2042        CVS
2043      - fixed a small bug related to incomplete pointer types in
2044        ml-nlffigen
2045      - small cosmetic change to the ml-nlffi-lib's "arr" type constructor
2046        (it does not need the 'f type parameter)
2047    
2048    ----------------------------------------------------------------------
2049    Name: Matthias Blume
2050    Date: 2001/05/23 14:30:00 EDT
2051    Tag: Release_110_33
2052    Description:
2053    
2054    New version number (110.33).  New bootfiles.
2055    
2056    ----------------------------------------------------------------------
2057    Name: Matthias Blume
2058    Date: 2001/05/22 18:06:00 EDT
2059    Tag: blume-20010522-targets
2060    Description:
2061    
2062    Made install.sh use file config/targets.customized if it exists, falling
2063    back to config/targets if it doesn't.  This way one can have a customized
2064    version of the targets file without touching the "real thing", thus
2065    eliminating the constant fear of accidentally checking something bogus
2066    back into the CVS repository...  (File config/targets.customized must
2067    not be added to the repository!)
2068    
2069    ----------------------------------------------------------------------
2070    Name: Matthias Blume
2071    Date: 2001/05/22 16:30:00 EDT
2072    Tag: blume-20010522-minitut
2073    Description:
2074    
2075    1. Bug fix in ml-nlffigen; now (hopefully) correctly handling
2076       struct returns.
2077    2. Added src/ml-nlffi-lib/Doc/mini-tutorial.txt.  This is some very
2078       incomplete, preliminary documentation for NLFFI.
2079    
2080    ----------------------------------------------------------------------
2081    Name: Matthias Blume
2082    Date: 2001/05/14 11:30:00 EDT
2083    Tag: blume-20010514-script
2084    Description:
2085    
2086    Some bugs in install script fixed.
2087    
2088    In addition to that I also made a slight change to the NLFFI API:
2089    Functors generated by ml-nlffigen now take the dynamic library as a
2090    straight functor argument, not as a suspended one.  (The original
2091    functor code used to force the suspension right away anyway, so there
2092    was nothing gained by this complication of the interface.)
2093    
2094    ----------------------------------------------------------------------
2095    Name: Matthias Blume
2096    Date: 2001/05/11 14:35:00 EDT
2097    Tag: blume-20010511-ml-nlffi
2098    Description:
2099    
2100    I finally took the plunge and added my new FFI code to the main
2101    repository.  For x86-linux it is now ready for prime-time.
2102    
2103    There are two new subdirectories of "src":
2104    
2105      - ml-nlffi-lib:
2106           The utility library for programs using the FFI interface.
2107           Here is the implementation of $/c.cm and its associated low-level
2108           partners $/c-int.cm and $/memory.cm.
2109      - ml-nlffigen:
2110           A stand-alone program for generating ML glue code from C source
2111           code.
2112    
2113    Building ml-nlffigen requires $/ckit-lib.cm.
2114    
2115    The config/install.sh script has been updates to do the Right Thing
2116    (hopefully).
2117    
2118    Notice that the source tree for the C-Kit will not be put under "src"
2119    but directly under the installation root directory.  (This is the
2120    structure that currently exists on the CVS server when you check out
2121    module "sml".)  Fortunately, config/install.sh knows about this oddity.
2122    
2123    Bugs: No documentation yet.
2124    
2125    ----------------------------------------------------------------------
2126    Name: Matthias Blume
2127    Date: 2001/05/09 16:35:00 EDT
2128    Tag: blume-20010509-cpscontract
2129    Description:
2130    
2131    Fixed a bug in the accounting code in cpsopt/contract.sml.  (The
2132    wrapper/unwrapper elimination did not decrement usage counts and some
2133    dead variables got overlooked by the dead-up logic.)
2134    
2135    ----------------------------------------------------------------------
2136    Name: Lal George
2137    Date: 2001/05/08  17:26:09 EDT
2138    Tag: george-20010508-omit-frameptr
2139    Description:
2140    
2141    Changes to implement the omit-frame-pointer optimization to support
2142    raw C calls. For now, there is only support on the Intel x86, but
2143    other architectures will follow as more experience is gained with this.
2144    
2145    
2146    ----------------------------------------------------------------------
2147    Name: Matthias Blume
2148    Date: 2001/05/07 14:40:00 EDT
2149    Tag: blume-20010507-proxies
2150    Description:
2151    
2152    I made into "proxy libraries" all libraries that qualify for such a
2153    change.  (A qualifying library is a library that has another library or
2154    groups as its sole member and repeats that member's export list
2155    verbatim.  A proxy library avoids this repetition by omitting its export
2156    list, effectively inheriting the list that its (only) member exports.
2157    See the CM manual for more explanation.)
2158    The main effect is that explicit export lists for these libraries
2159    do not have to be kepts in sync, making maintenance a bit easier.
2160    
2161    I also added copyright notices to many .cm-files.
2162    
2163    Last but not least, I made a new set of bootfiles.
2164    
2165    ----------------------------------------------------------------------
2166    Name: Matthias Blume
2167    Date: 2001/05/04 17:00:00 EDT
2168    Tag: blume-20010504-cm-lsplit
2169    Description:
2170    
2171    0. John merged pending changes to $/smlnj-lib.cm
2172    
2173    1. Allen's previous change accidentally backed out of one of Lal's
2174       earlier changes.  I undid this mistake (re-introducing Lal's change).
2175    
2176    2. I used the new topOrder' function from graph-scc.sml (from $/smlnj-lib.cm)
2177       within the compiler where applicable.  There is some code simplification
2178       because of that.
2179    
2180    3. The "split" phase (in FLINT) is now part of the default list of phases.
2181       Compiler.Control.LambdaSplitting.* can be used to globally control the
2182       lambda-splitting (cross-module-inlining) engine.  In addition to that,
2183       it can now also be controlled on a per-source basis: CM has been taught
2184       a new tool parameter applicable to ML source files.
2185    
2186       - To turn lambda-splitting off completely:
2187            local open Compiler.Control.LambdaSplitting in
2188                val _ = set Off
2189            end
2190       - To make "no lambda-splitting" the global default (but allow per-source
2191         overriding); this is the initial setting:
2192            local open Compiler.Control.LambdaSplitting in
2193                val _ = set (Default NONE)
2194            end
2195       - To make "lambda-splitting with aggressiveness a" the global default
2196         (and allow per-source overriding):
2197            local open Compiler.Control.LambdaSplitting in
2198                val _ = set (Default (SOME a))
2199            end
2200    
2201       - To turn lambda-splitting off for a given ML souce file (say: a.sml)
2202         write (in the respective .cm-file):
2203            a.sml (lambdasplitting:off)
2204       - To turn lambda-splitting for a.sml on with minimal aggressiveness:
2205            a.sml (lambdasplitting:on)
2206       - To turn lambda-splitting for a.sml on with aggressiveness <a> (where
2207         <a> is a decimal non-negative integer):
2208            a.sml (lambdasplitting:<a>)
2209       - To turn lambda-splitting for a.sml on with maximal aggressiveness:
2210            a.sml (lambdasplitting:infinity)
2211       - To use the global default for a.sml:
2212            a.sml (lambdasplitting:default)
2213         or simply
2214            a.sml
2215    
2216    ----------------------------------------------------------------------
2217    Name: Allen Leung
2218    Date: 2001/05/04 01:57:00 EDT
2219    Tag: leunga-20010504-sync
2220    Description:
2221    
2222      MLRISC features.
2223    
2224      1. Fix to CMPXCHG instructions.
2225      2. Changed RA interface to allow annotations in callbacks.
2226      3. Added a new method to the stream interface to allow annotations updates.
2227    
2228    ----------------------------------------------------------------------
2229    Name: Matthias Blume
2230    Date: 2001/05/01 11:45:00 EDT
2231    Tag: blume-20010501-pcedittmp
2232    Description:
2233    
2234    Changed install.sh to use the current working directory instead of
2235    /usr/tmp for a temporary file (pcedittmp).  The previous choice
2236    of /usr/tmp caused trouble with MacOS X because of file premission
2237    problems.
2238    
2239    ----------------------------------------------------------------------
2240    Name: Matthias Blume
2241    Date: 2001/04/20 11:10:00 EDT
2242    Tag: blume-20010420-inMLflag
2243    Description:
2244    
2245     - added vp_limitPtrMask to vproc-state.h
2246       (for use by the raw-C-calls mechanism to implement proper interrupt
2247        handling)
2248     - made the ML compiler aware of various data-structure offsets so it
2249       can generate code for accessing the vp_inML flag and vp_limitPtrMask
2250     - tweaked mlriscGen.sml to have it emit interrupt-handling code for
2251       raw C-calls
2252    
2253    ----------------------------------------------------------------------
2254    Name: Lal George
2255    Date: 2001/04/20 09:15:28 EDT
2256    Tag: george-20010420-macosX
2257    Description:
2258    
2259     - Changes to port to Mac OS X; Darwin.
2260    
2261     - In the process I found that sqrt was broken on the PPC, because the
2262       fsqrt instruction is not implemented.
2263    
2264    ----------------------------------------------------------------------
2265    Name: Matthias Blume
2266    Date: 2001/04/18 12:45:00 EDT
2267    Tag: blume-20010418-ccalls
2268    Description:
2269    
2270     - fixed two off-by-4 errors in the x86-specific c-calls implementation
2271       (this bug prevented structure arguments containing pointers from being
2272        passed correctly)
2273     - changed the raw-C-call code in mlriscGen.sml in such a way that
2274       structure arguments are represented as a pointer to the beginning
2275       of the structure (instead of having a series of synthesized arguments,
2276       one for each structure member)
2277    
2278     - made makeml script's verbosity level configurable via environment
2279       variable (MAKEML_VERBOSITY)
2280    
2281     - eliminated placeholder implementations for f32l, w16s, i16s, and f32s
2282       in rawmem-x86.sml; we are now using the real thing
2283    
2284    ----------------------------------------------------------------------
2285    Name: Matthias Blume
2286    Date: 2001/03/22 16:25:00 EST
2287    Tag: blume-20010322-bootfiles
2288    Description:
2289    
2290    Created a new set of bootfiles (for your automatic installation convenience).
2291    
2292    ----------------------------------------------------------------------
2293    Name: Matthias Blume
2294    Date: 2001/03/22 15:10:00 EST
2295    Tag: blume-20010322-rawmem-parcm
2296    Description:
2297    
2298    1. All "raw memory access" primitives for the new FFI are implemented now
2299       (at least on the x86).
2300    2. Some further cleanup of CM's parallel make mechanism.
2301    
2302    ----------------------------------------------------------------------
2303    Name: Matthias Blume
2304    Date: 2001/03/19 17:53:00 EST
2305    Tag: blume-20010319-parallel
2306    Description:
2307    
2308    Parallel make (using compile servers) now works again.
2309    
2310    To this end, CM.stabilize and CMB.make have been modified to work in
2311    two passes when compile servers are attached:
2312       1. Compile everything, do not perform stabilization; this pass
2313          uses compile servers
2314       2. Stabilize everything; this pass does not use compile servers
2315    If there are no compile servers, the two passes are combined into one
2316    (as before).  Splitting the passes increases the inherent parallelism
2317    in the dependency graph because the entire graph including all
2318    libraries is available at the same time.  This, in turn, improves
2319    server utilization.  The downside is that the master process will
2320    have to do some extra work after compilation is done (because for
2321    technical reasons it must re-read all the binfiles during stabilization).
2322    
2323    ----------------------------------------------------------------------
2324    Name: Matthias Blume
2325    Date: 2001/03/16 12:22:00 EST
2326    Tag: blume-20010316-bootfiles
2327    Description:
2328    
2329    Created a new set of bootfiles (for your automatic installation convenience).
2330    
2331    ----------------------------------------------------------------------
2332    Name: Matthias Blume
2333    Date: 2001/03/16 11:00:00 EST
2334    Tag: blume-20010316-MLTREE-fixup
2335    Description:
2336    
2337    This is a minor fixup for an (untagged) earlier commit by Allen.
2338    (A file was missing).
2339    
2340    ----------------------------------------------------------------------
2341    Name: Allen Leung
2342    Date: Mon Mar  5 18:54:57 EST 2001
2343    Tag: leunga-20010305-cut-support
2344    
2345    1. New support for alternative control-flow in MLTREE.
2346       Currently we support
2347    
2348          FLOW_TO(CALL ...., [k1,...,kn])
2349    
2350       This is needed for 'cuts to' in C-- and try/handle-like constructs
2351       in Moby
2352    
2353       New assembler flag "asm-show-cutsto" to turn on control-flow debugging.
2354    
2355    2. Register Allocator
2356    
2357       Changes in interface [from Fermin, John]
2358    
2359    3. Alpha 8-bit SLL support [Fermin]
2360    
2361    4. All architectures
2362    
2363       A new module (ClusterExpandCopies) for expanding parallel copies.
2364    
2365    ----------------------------------------------------------------------
2366    Name: Allen Leung
2367  Date: 2001/02/27 23:07:00 EST  Date: 2001/02/27 23:07:00 EST
2368  Tag: leunga-20010227-minor-stuff  Tag: leunga-20010227-minor-stuff
2369    
# Line 1961  Line 4311 
4311       elaborator). There were a lot of changes during my "linkpath" trials       elaborator). There were a lot of changes during my "linkpath" trials
4312       that could have been reverted to their original state but weren't.       that could have been reverted to their original state but weren't.
4313       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
4314       than what was strictly necessary...  (I _did_ resist the tempation       than what was strictly necessary...  (I _did_ resist the temptation
4315       of doing any "global reformatting" to avoid an untimely death at       of doing any "global reformatting" to avoid an untimely death at
4316       Dave's hands. :)       Dave's hands. :)
4317    

Legend:
Removed from v.788  
changed lines
  Added in v.1131

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