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 839, Thu Jun 7 20:28:44 2001 UTC revision 916, Fri Aug 24 21:13:58 2001 UTC
# Line 14  Line 14 
14    
15  ----------------------------------------------------------------------  ----------------------------------------------------------------------
16  Name: Matthias Blume  Name: Matthias Blume
17    Date: 2001/08/24 17:10:00 EDT
18    Tag: Release_110_35
19    Description:
20    
21    New version number (110.35).  New bootfiles.
22    
23    ----------------------------------------------------------------------
24    Name: Lal George
25    Date: 2001/08/24 13:47:18 EDT 2001
26    Tag: george-20010824-MLRISC-graphs
27    Description:
28    
29     removed clusters from MLRISC completely and replaced with graphs.
30    
31    ----------------------------------------------------------------------
32    Name: Matthias Blume
33    Date: 2001/08/23 17:50:00 EDT
34    Tag: blume-20010823-toplevel
35    Description:
36    
37    - some reorganization of the code that implements various kinds of
38      environments in the compiler (static, dynamic, symbolic, combined)
39    - re-implemented the EnvRef module so that evalStream works properly
40      (if the stream contains references to "use", "CM.make", etc.)
41    - cleaned up evalloop.sml and interact.sml (but they need more cleaning)
42    
43    ----------------------------------------------------------------------
44    Name: Matthias Blume
45    Date: 2001/08/20 15:50 EDT
46    Tag: blume20010820-slipup
47    Description:
48    
49    I forgot to commit a few files.  Here they are...
50    
51    ----------------------------------------------------------------------
52    Name: Matthias Blume
53    Date: 2001/08/20 15:35:00 EDT
54    Tag: blume-20010820-debugprof
55    Description:
56    
57    !!!! NEW BOOTFILES !!!!
58    
59    This is another round of reorganizing the compiler sources.  This
60    time the main goal was to factor out all the "instrumentation"
61    passes (for profiling and backtracing) into their own library.
62    The difficulty was to do it in such a way that it does not depend
63    on elaborate.cm but only on elabdata.cm.
64    
65    Therefore there have been further changes to both elaborate.cm and
66    elabdata.cm -- more "generic" things have been moved from the former
67    to the latter.  As a result, I was forced to split the assignment
68    of numbers indicating "primtyc"s into two portions: SML-generic and
69    SML/NJ-specific.  Since it would have been awkward to maintain,
70    I bit the bullet and actually _changed_ the mapping between these
71    numbers and primtycs.  The bottom line of this is that you need
72    a new set of bin- and bootfiles.
73    
74    I have built new bootfiles for all architectures, so doing a fresh
75    checkout and config/install.sh should be all you need.
76    
77    The newly created library's name is
78    
79        $smlnj/viscomp/debugprof.cm
80    
81    and its sources live under
82    
83        src/compiler/DebugProf
84    
85    ----------------------------------------------------------------------
86    Name: Matthias Blume
87    Date: 2001/08/15 17:15:00 EDT
88    Tag: blume-20010815-compreorg
89    Description:
90    
91    This is a first cut at reorganizing the CM libraries that make up the
92    core of the compiler.  The idea is to separate out pieces that could
93    be used independently by tools, e.g., the parser, the typechecker, etc.
94    
95    The current status is a step in this direction, but it is not quite
96    satisfactory yet.  Expect more changes in the future.
97    
98    Here is the current (new) organization...
99    
100        What used to be $smlnj/viscomp/core.cm is now divided into
101        six CM libraries:
102    
103             $smlnj/viscomp/basics.cm
104                           /parser.cm
105                           /elabdata.cm
106                           /elaborate.cm
107                           /execute.cm
108                           /core.cm
109    
110        The CM files for these libraries live under src/system/smlnj/viscomp.
111        All these libraries are proxy libraries that contain precisely
112        one CM library component.  Here are the locations of the components
113        (all within the src/compiler tree):
114    
115             Basics/basics.cm
116             Parse/parser.cm
117             ElabData/elabdata.cm
118             Elaborator/elaborate.cm
119             Execution/execute.cm
120             core.cm
121    
122         [This organization is the same that has been used already
123         for a while for the architecture-specific parts of the visible
124         compiler and for the old version of core.cm.]
125    
126         As you will notice, many source files have been moved from their
127         respective original locations to a new home in one of the above
128         subtrees.
129    
130         The division of labor between the new libraries is the following:
131    
132             basics.cm:
133                - Simple, basic definitions that pertain to many (or all) of
134                  the other libraries.
135             parser.cm:
136                - The SML parser, producing output of type Ast.dec.
137                - The type family for Ast is also defined and exported here.
138             elabdata.cm:
139                - The datatypes that describe input and output of the elaborator.
140                  This includes types, absyn, and static environments.
141             elaborator.cm:
142                - The SML/NJ type checker and elaborator.
143                  This maps an Ast.dec (with a given static environment) to
144                  an Absyn.dec (with a new static environment).
145                - This libraries implements certain modules that used to be
146                  structures as functors (to remove dependencies on FLINT).
147             execute.cm:
148                - Everything having to do with executing binary code objects.
149                - Dynamic environments.
150             core.cm:
151                - SML/NJ-specific instantiations of the elaborator and MLRISC.
152                - Top-level modules.
153                - FLINT (this should eventually become its own library)
154    
155    Notes:
156    
157    I am not 100% happy with the way I separated the elaborator (and its
158    data structures) from FLINT.  Two instances of the same problem:
159    
160        1. Data structures contain certain fields that carry FLINT-specific
161           information.  I hacked around this using exn and the property list
162           module from smlnj-lib.  But the fact that there are middle-end
163           specific fields around at all is a bit annoying.
164    
165        2. The elaborator calculates certain FLINT-related information.  I tried
166           to make this as abstract as I could using functorization, but, again,
167           the fact that the elaborator has to perform calculations on behalf
168           of the middle-end at all is not nice.
169    
170        3. Having to used exn and property lists is unfortunate because it
171           weakens type checking.  The other alternative (parameterizing
172           nearly *everything*) is not appealing, though.
173    
174    I removed the "rebinding =" warning hack because due to the new organization
175    it was awkward to maintain it.  As a result, the compiler now issues some of
176    these warnings when compiling init.cmi during bootstrap compilation. On
177    the plus side, you also get a warning when you do, for example:
178       val op = = Int32.+
179    which was not the case up to now.
180    
181    I placed "assign" and "deref" into the _Core structure so that the
182    code that deals with the "lazy" keyword can find them there.  This
183    removes the need for having access to the primitive environment
184    during elaboration.
185    
186    ----------------------------------------------------------------------
187    Name: Matthias Blume
188    Date: 2001/08/13
189    Tag: blume-20010813-closures
190    Description:
191    
192    This fix was sent to us by Zhong Shao.  It is supposed to improve the
193    performance of certain loops by avoiding needless closure allocation.
194    
195    ----------------------------------------------------------------------
196    Name: Lal George
197    Date: 2001/07/31 10:03:23 EDT 2001
198    Tag: george-20010731-x86-fmalloc
199    Description: Fixed bug in x86 calls
200    
201        There was a bug where call instructions would mysteriously
202        vanish. The call instruction had to be one that returned
203        a floating point value.
204    
205    ----------------------------------------------------------------------
206    Name: Lal George
207    Date: 2001/07/19 16:36:29 EDT 2001
208    Tag: george-20010719-simple-cells
209    Description:
210    
211    I have dramatically simplified the interface for CELLS in MLRISC.
212    
213    In summary, the cells interface is broken up into three parts:
214    
215      1. CellsBasis : CELLS_BASIS
216    
217            CellsBasis is a top level structure and common for all
218            architectures.  it contains the definitions of basic datatypes
219            and utility  functions over these types.
220    
221      2. functor Cells() : CELLS
222    
223            Cells generates an interface for CELLS that incorporates the
224            specific resources on the target architecture, such as the
225            presence of special register classes, their number and size,
226            and various useful substructures.
227    
228      3. <ARCH>CELLS
229    
230            e.g. SparcCells: SPARCCELLS
231    
232            <ARCH>CELLS usually contains additional bindings for special
233            registers  on the architecture, such as:
234    
235                    val r0 : cell           (* register zero *)
236                    val y : cell            (* Y register *)
237                    val psr : cell          (* processor status register *)
238                    ...
239    
240            The structure returned by applying the Cells functor is opened
241            in this interface.
242    
243    The main implication of all this is that the datatypes for cells is
244    split between CellsBasis and CELLS -- a fairly simple change for user
245    code.
246    
247    In the old scheme the CELLS interface had a definitional binding of
248    the form:
249    
250            signature CELLS = sig
251    
252               structure CellsBasis = CellsBasis
253    
254               ...
255    
256            end
257    
258    With all the sharing constraints that goes on in MLRISC, this old
259    design  quickly leads to errors such as:
260    
261            "structure definition spec inside of sharing ... "
262    
263    
264    and appears to require an unacceptable amount of sharing and where
265    constraint hackery.
266    
267    I think this error message (the interaction of definitional specs and
268    sharing) requires more explanation on our web page.
269    
270    ----------------------------------------------------------------------
271    Name: Matthias Blume
272    Date: 2001/07/19 15:00:00 EDT
273    Tag: blume-20010719-libreorg
274    Description:
275    
276    This update puts together a fairly extensive but straightforward change
277    to the way the libraries that implement the interactive system are
278    organized:
279    
280       The biggest change is the elimination of structure Compiler.  As a
281       replacement for this structure, there is now a CM library
282       (known as $smlnj/compiler.cm or $smlnj/compiler/current.cm)
283       that exports all the substructures of the original structure Compiler
284       directly.  So instead of saying Compiler.Foo.bar one now simply
285       says Foo.bar.  (The CM libraries actually export a collection of
286       structures that is richer than the collection of substructures of
287       structure Compiler.)
288    
289       To make the transition smooth, there is a separate library called
290       $smlnj/compiler/compiler.cm which puts together and exports the
291       original structure Compiler (or at least something very close to it).
292    
293       There are five members of the original structure Compiler
294       that are not exported directly but which instead became members
295       of a new structure Backend (described by signature BACKEND).  These are:
296       structure Profile (: PROFILE), structure Compile (: COMPILE), structure
297       Interact (: INTERACT), structure Machine (: MACHINE), and val
298       architecture (: string).
299    
300       Structure Compiler.Version has become structure CompilerVersion.
301    
302       Cross-compilers for alpha32, hppa, ppc, sparc, and x86 are provided
303       by $smlnj/compiler/<arch>.cm where <arch> is alpha32, hppa, ppc, sparc,
304       or x86, respectively.
305       Each of these exports the same frontend structures that
306       $smlnj/compiler.cm exports.  But they do not have a structure Backend
307       and instead export some structure <Arch>Backend where <Arch> is Alpha32,
308       Hppa, PPC, Sparc, or X86, respectively.
309    
310       Library $smlnj/compiler/all.cm exports the union of the exports of
311       $smlnj/compiler/<arch>.cm
312    
313       There are no structures <Arch>Compiler anymore, use
314       $smlnj/compiler/<arch>.cm instead.
315    
316       Library host-compiler-0.cm is gone.  Instead, the internal library
317       that instantiates CM is now called cm0.cm.  Selection of the host
318       compiler (backend) is no longer done here but. (Responsibility for it
319       now lies with $smlnj/compiler/current.cm.  This seems to be more
320       logical.)
321    
322       Many individual files have been moved or renamed.  Some files have
323       been split into multiple files, and some "dead" files have been deleted.
324    
325    Aside from these changes to library organization, there are also changes
326    to the way the code itself is organized:
327    
328       Structure Binfile has been re-implemented in such a way that it no
329       longer needs any knowledge of the compiler.  It exclusively deals
330       with the details of binfile layout.  It no longer invokes the
331       compiler (for the purpose of creating new prospective binfile
332       content), and it no longer has any knowledge of how to interpret
333       pickles.
334    
335       Structure Compile (: COMPILE) has been stripped down to the bare
336       essentials of compilation.  It no longer deals with linking/execution.
337       The interface has been cleaned up considerably.
338    
339       Utility routines for dealing with linking and execution have been
340       moved into their own substructures.
341    
342       (The ultimate goal of these changes is to provide a light-weight
343       binfile loader/linker (at least for, e.g., stable libraries) that
344       does not require CM or the compiler to be present.)
345    
346    CM documentation has been updated to reflect the changes to library
347    organization.
348    
349    ----------------------------------------------------------------------
350    Name: Matthias Blume
351    Date: 2001/07/10 17:30:00 EDT
352    Tag: Release_110_34
353    Description:
354    
355    Minor tweak to 110.34 (re-tagged):
356    
357      - README.html file added to CVS repository
358      - runtime compiles properly under FreeBSD 3.X and 4.X
359    
360    ----------------------------------------------------------------------
361    Name: Matthias Blume
362    Date: 2001/07/10 17:30:00 EDT
363    Tag: Release_110_34
364    Description:
365    
366    New version number (110.34). New bootfiles.
367    
368    ----------------------------------------------------------------------
369    Name: Matthias Blume
370    Date: 2001/07/09 16:00:00 EDT
371    Tag: blume-20010709-more-varargs
372    Description:
373    
374    I changed the handling of varargs in ml-nlffigen again:
375    The ellipsis ... will now simply be ignored (with an accompanying warning).
376    
377    The immediate effect is that you can actually call a varargs function
378    from ML -- but you can't actually supply any arguments beyond the ones
379    specified explicitly.  (For example, you can call printf with its format
380    string, but you cannot pass additional arguments.)
381    
382    This behavior is only marginally more useful than the one before, but
383    it has the advantage that a function or, more importantly, a function
384    type never gets dropped on the floor, thus avoiding follow-up problems with
385    other types that refer to the offending one.
386    
387    ----------------------------------------------------------------------
388    Name: Matthias Blume
389    Date: 2001/07/09 11:25:00 EDT
390    Tag: blume-20010709-varargs
391    Description:
392    
393    1. ckit-lib.cm now exports structure Error
394    2. ml-nlffigen reports occurences of "..." (i.e., varargs function types)
395       with a warning accompanied by a source location.  Moreover, it
396       merely skips the offending function or type and proceeds with the
397       rest of its work.u  As a result, one can safely feed C code containing
398       "..." to ml-nlffigen.
399    3. There are some internal improvements to CM, providing slightly
400       more general string substitutions in the tools subsystem.
401    
402    ----------------------------------------------------------------------
403    Name: Matthias Blume
404    Date: 2001/06/27 15:10:00 EDT
405    Tag: blume-20010627-concur
406    Description:
407    
408    Fixed a small bug in CM's handling of parallel compilation.
409    (You could observe the bug by Control-C-interrupting an ordinary
410    CMB.make or CM.stabilize and then attaching some compile servers.
411    The result was that all of a sudden the previously interrupted
412    compilation would continue on its own.  This was because of
413    an over-optimization: CM did not bother to clean out certain queues
414    when no servers were attached "anyway", resulting in the contents
415    of these queues to grab control when new servers did get attached.)
416    
417    There is also another minor update to the CM manual.
418    
419    ----------------------------------------------------------------------
420    Name: Matthias Blume
421    Date: 2001/06/26 16:15:00 EDT
422    Tag: blume-20010626-cmdoc
423    Description:
424    
425    Minor typo fixed in CM manual (syntax diagram for libraries).
426    
427    ----------------------------------------------------------------------
428    Name: Matthias Blume
429    Date: 2001/06/25 22:55:00 EDT
430    Tag: blume-20010625-x86pc
431    Description:
432    
433    Fixed a nasty bug in the X86 assembly code that caused signal
434    handlers to fail (crash) randomly.
435    
436    ----------------------------------------------------------------------
437    Name: Matthias Blume
438    Date: 2001/06/25 12:05:00 EDT
439    Tag: blume-20010625-nlffigen
440    Description:
441    
442    This update fixes a number of minor bugs in ml-nlffigen as reported by
443    Nick Carter <nbc@andrew.cmu.edu>.
444    
445      1. Silly but ok typedefs of the form "typedef void myvoid;" are now accepted.
446      2. Default names for generated files are now derived from the name of
447         the C file *without its directory*.  In particular, this causes generated
448         files to be placed locally even if the C file is in some system directory.
449      3. Default names for generated signatures and structures are also derived
450         from the C file name without its directory.  This avoids silly things
451         like "structure GL/GL".
452         (Other silly names are still possible because ml-nlffigen does not do
453          a thorough check of whether generated names are legal ML identifiers.
454          When in doubt, use command line arguments to force particular names.)
455    
456    ----------------------------------------------------------------------
457    Name: Matthias Blume
458    Date: 2001/06/21 12:25:00 EDT
459    Tag: blume-20010621-eXene
460    Description:
461    
462    eXene now compiles and (sort of) works again.
463    
464    The library name (for version > 110.33) is $/eXene.cm.
465    
466    I also added an new example in src/eXene/examples/nbody.  See the
467    README file there for details.
468    
469    ----------------------------------------------------------------------
470    Name: Matthias Blume
471    Date: 2001/06/20 16:40:00 EDT
472    Tag: blume-20010620-cml
473    Description:
474    
475    CML now compiles and works again.
476    
477    Libraries (for version > 110.33):
478    
479      $cml/cml.cm            Main CML library.
480      $cml/basis.cm          CML's version of $/basis.cm.
481      $cml/cml-internal.cm   Internal helper library.
482      $cml/core-cml.cm       Internal helper library.
483      $cml-lib/trace-cml.cm  Tracing facility.
484      $cml-lib/smlnj-lib.cm  CML's version of $/smlnj-lib.cm
485    
486    The installer (config/install.sh) has been taught how to properly
487    install this stuff.
488    
489    ----------------------------------------------------------------------
490    Name: Matthias Blume
491    Date: 2001/06/19 17:55:00 EDT
492    Tag: blume-20010619-instantiate
493    Description:
494    
495    This un-breaks the fix for bug 1432.
496    (The bug was originally fixed in 110.9 but I broke it again some
497    time after that.)
498    
499    ----------------------------------------------------------------------
500    Name: Matthias Blume
501    Date: 2001/06/19 17:25:00 EDT
502    Tag: blume-20010619-signals
503    Description:
504    
505    This should (hopefully) fix the long-standing signal handling bug.
506    (The runtime system was constructing a continuation record with an
507    incorrect descriptor which would cause the GC to drop data on the floor...)
508    
509    ----------------------------------------------------------------------
510    Name: Matthias Blume
511    Date: 2001/06/15 15:05:00 EDT
512    Tag: blume-20010615-moresparc
513    Description:
514    
515    Here is a short late-hour update related to Sparc c-calls:
516    
517     -- made handling of double-word arguments a bit smarter
518    
519     -- instruction selection phase tries to collapse certain clumsily
520        constructed ML-Trees; typical example:
521    
522            ADD(ty,ADD(_,e,LI d1),LI d2)  ->  ADD(ty,e,LI(d1+d2))
523    
524        This currently has no further impact on SML/NJ since mlriscGen does
525        not seem to generate such patterns in the first place, and c-calls
526        (which did generate them in the beginning) has meanwhile been fixed
527        so as to avoid them as well.
528    
529    ----------------------------------------------------------------------
530    Name: Matthias Blume
531    Date: 2001/06/15 15:05:00 EDT
532    Tag: blume-20010615-sparc
533    Description:
534    
535    The purpose of this update is to provide an implementation of NLFFI
536    on Sparc machines.
537    
538    Here are the changes in detail:
539    
540       * src/MLRISC/sparc/c-calls/sparc-c-calls.sml is a new file containing
541       the Sparc implementation of the c-calls API.
542       * The Sparc backend of SML/NJ has been modified to uniformely use %fp
543       for accessing the ML frame.  Thus, we have a real frame pointer and
544       can freely modify %sp without need for an omit-frame-ptr phase.
545       The vfp logic in src/compiler/CodeGen/* has been changed to accomodate
546       this case.
547       * ml-nlffigen has been taught to produce code for different architectures
548       and calling conventions.
549       * In a way similar to what was done in the x86 case, the Sparc
550       backend uses its own specific extension to mltree.  (For example,
551       it needs to be able to generate UNIMP instructions which are part
552       of the calling convention.)
553       * ml-nlffi-lib was reorganized to make it more modular (in particular,
554       to make it easier to plug in new machine- and os-dependent parts).
555    
556    There are some other fairly unrelated bug fixes and cleanups as well:
557    
558       * I further hacked the .cm files for MLRISC tools (like MDLGen) so
559       that they properly share their libraries with existing SML/NJ libraries.
560       * I fixed a minor cosmetic bug in CM, supressing certain spurious
561       follow-up error messages.
562       * Updates to CM/CMB documentation.
563    
564    TODO items:
565    
566       * MLRISC should use a different register as its asmTemp on the Sparc.
567         (The current %o2 is a really bad choice because it is part of the
568         calling conventions, so things might interfere in unexpected ways.)
569    
570    ----------------------------------------------------------------------
571    Name: Matthias Blume
572  Date: 2001/06/07  Date: 2001/06/07
573  Tag: blume-20010607-calls  Tag: blume-20010607-calls
574  Description:  Description:

Legend:
Removed from v.839  
changed lines
  Added in v.916

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