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 641, Thu May 11 01:16:45 2000 UTC revision 671, Mon Jun 19 02:02:32 2000 UTC
# Line 12  Line 12 
12  Tag: <post-commit CVS tag>  Tag: <post-commit CVS tag>
13  Description:  Description:
14  ----------------------------------------------------------------------  ----------------------------------------------------------------------
15    Name: Matthias Blume
16    Date: 2000/06/19 11:05:00 JST
17    Tag: blume-20000619-chdir-bugfix
18    Description:
19    
20    Fixed a bug in new SrcPath module that sometimes led to a bad chDir call.
21    
22    ----------------------------------------------------------------------
23    Name: Matthias Blume
24    Date: 2000/06/18 22:00:10 JST
25    Tag: blume-20000618-implicit-anchors-really-gone
26    Description:
27    
28    I updates the previous HISTORY entry where I forgot to mention that
29    implicit anchors are no longer with us.
30    
31    The current update also gets rid of the (now useless) controller
32    CM.Control.implicit_anchors.
33    
34    ----------------------------------------------------------------------
35    Name: Matthias Blume
36    Date: 2000/06/16 17:30:00 JST
37    Tag: blume-20000616-anchorenv
38    Description:
39    
40    This patch implements the long anticipated (just kidding :) "anchor
41    environment" mechanism.  In the course of doing this, I also
42    re-implemented CM's internal "SrcPath" module from scratch.  The new
43    one should be more robust in certain boundary cases.  In any case, it
44    is a lot cleaner than its predecessor (IMHO).
45    
46    This time, although there is yet another boot file format change, I
47    kept the unpickler backward-compatible.  As a result, no new bootfiles
48    are necessary and bootstrapping is straightforward.  (You cannot read
49    new bootfiles into an old system, but the other way around is no
50    problem.)
51    
52    Visible changes:
53    
54    ** 0. Implicit path anchors (without the leading $-symbol) are no
55    longer recognized at all. This means that such path names are not
56    illegal either.  For example, the name basis.cm simply refers to a
57    local file called "basis.cm" (i.e, the name is an ordinary path
58    relative to .cm-files directory).  Or, to put it differently, only
59    names that start with $ are anchored paths.
60    
61    ** 1. The $<singlearc> abbreviation for $/<singlearc> has finally
62    vanished.
63    
64    John (Reppy) had critizised this as soon as I originally proposed and
65    implemented it, but at that time I did not really deeply believe
66    him. :) Now I came full-circle because I need the $<singlearc> syntax
67    in another place where it cannot be seen as an abbreviation for
68    $/<singlearc>.  To avoid the confusion, $<singlearc> now means what it
69    seems to mean (i.e., it "expands" into the corresponding anchor
70    value).
71    
72    However, when paths are used as members in CM description files, it
73    continues to be true that there must be at least another arc after the
74    anchor.  This is now enforced separately during semantic analysis
75    (i.e., from a lexical/syntactical point of view, the notation is ok.)
76    
77    ** 2. The "cm" class now accepts an option "bind".  The option's value
78    is a sub-option list of precisely two items -- one labeled "anchor"
79    and the other one labeled "value".  As you might expect, "anchor" is
80    used to specify an anchor name to be bound, and "value" specifies what
81    the anchor is being bound to.
82    
83    The value must be a directory name and can be given in either standard
84    syntax (including the possibility that it is itself an anchored path)
85    or native syntax.
86    
87    Examples:
88    
89       foo.cm (bind:(anchor:bar value:$mystuff/bar))
90       lib.cm (bind:(anchor:a value:"H:\\x\\y\\z"))  (* only works under windows *)
91    
92    and so on.
93    
94    The meaning of this is that the .cm-file will be processed with an
95    augmented anchor environment where the given anchor(s) is/are bound to
96    the given values(s).
97    
98    The rationale for having this feature is this: Suppose you are trying
99    to use two different (already stable) libraries a.cm and b.cm (that
100    you perhaps didn't write yourself).  Further, suppose each of these
101    two libraries internally uses its own auxiliary library $aux/lib.cm.
102    Normally you would now have a problem because the anchor "lib" can not
103    be bound to more than one value globally.  Therefore, the project that
104    uses both a.cm and b.cm must locally redirect the anchor to some other
105    place:
106    
107       a.cm (bind:(anchor:lib value:/usr/lib/smlnj/a-stuff))
108       b.cm (bind:(anchor:lib value:/usr/lib/smlnj/b-stuff))
109    
110    This hard-wires $lib/aux.cm to /usr/lib/smlnj/a-stuff/aux.cm or
111    /usr/lib/smlnj/b-stuff/aux.cm, respectively.
112    
113    Hard-wiring path names is a bit inflexible (and CM will verbosely warn
114    you when you do so at the time of CM.stabilize).  Therefore, you can
115    also use an anchored path as the value:
116    
117      a.cm (bind:(anchor:lib value:$a-lib))
118      b.cm (bind:(anchor:lib value:$b-lib))
119    
120    Now you can globally configure (using the usual CM.Anchor.anchor or
121    pathconfig machinery) bindings for "a-lib" and "b-lib".  Since "lib"
122    itself is always locally bound, setting it globally is no longer
123    meaningful or necessary (but it does not hurt either).  In fact, "lib"
124    can still be used as a global anchor for separate purposes.  As a
125    matter of fact, one can locally define "lib" in terms of a global
126    "lib":
127    
128      a.cm (bind:(anchor:lib value:$lib/a))
129      b.cm (bind:(anchor:lib value:$lib/b))
130    
131    ** 3: The encoding of path names has changed.  This affects the way
132    path names are shown in CM's progress report and also the internal
133    protocol encoding used for parallel make.
134    
135    The encoding now uses one or more ':'-separated segments.  Each
136    segments corresponds to a file that has been specified relative to the
137    file given by its preceding segment.  The first segment is either
138    relative to the CWD, absolute, or anchored.  Each segment itself is
139    basically a Unix pathname; all segments but the first are relative.
140    
141    Example:
142    
143       $foo/bar/baz.cm:a/b/c.sml
144    
145    This path denotes the file bar/a/b/c.sml relative to the directory
146    denoted by anchor "foo".  Notice that the encoding also includes
147    baz.cm which is the .cm-file that listed a/b/c.sml.  As usual, such
148    paths are resolved relative to the .cm-files directory, so baz.cm must
149    be ignored to get the "real" pathname.
150    
151    To make this fact more obvious, CM puts the names of such "virtual
152    arcs" into parentheses when they appear in progress reports. (No
153    parentheses will appear in the internal protocol encoding.)  Thus,
154    what you really see is:
155    
156      $foo/bar/(baz.cm):a/b/c.sml
157    
158    I find this notation to be much more informative than before.
159    
160    Another new feature of the encoding is that special characters
161    including parentheses, colons, (back)slashes, and white space are
162    written as \ddd (where ddd is the decimal encoding of the character).
163    
164    *** The CM manual still needs to be updated.
165    
166    ----------------------------------------------------------------------
167    Name: Allen Leung
168    Date: 2000/06/15 00:38:00
169    Tag: leunga-20000615-x86-peephole
170    
171    x86 Peephole fix by Fermin.  Affects c-- and moby only.
172    
173    ----------------------------------------------------------------------
174    Name: Matthias Blume
175    Date: 2000/06/12 11:40:00
176    Tag: blume-20000612-parmakefix
177    Description:
178    
179    More cleanup after changing the file naming scheme: This time I
180    repaired the parallel make mechanism for CMB.make which I broke earlier.
181    
182    ----------------------------------------------------------------------
183    Name: Allen Leung
184    Date: 2000/06/09 01:25:00
185    Tag: leunga-20000609-various
186    
187    None of these things should affect normal SML/NJ operations
188    
189    1. Peephole improvements provided by Fermin (c--)
190    2. New annotation DEFUSE for adding extra dependence (moby)
191    3. New X86 LOCK instructions (moby)
192    4. New machine description language for reservation tables (scheduling)
193    5. Fixes to various optimization/analysis modules (branch chaining, dominator
194       trees etc.)
195    6. I've changed the CM files so that they can work with versions
196       110.0.6, 110.25 and 110.28
197    
198    ----------------------------------------------------------------------
199    Name: Matthias Blume
200    Date: 2000/06/09 12:40:00
201    Tag: blume-20000609-log
202    Description:
203    
204    - Removed all(?) remaining RCS Log entries from sources.
205    
206    - Fixed bug in ml-yacc and ml-lex sources (use explicit anchors for
207      anchored paths).
208    
209    ----------------------------------------------------------------------
210    Name: Matthias Blume
211    Date: 2000/06/07 17:00:00 JST
212    Tag: blume-20000607-no-implicit-anchors
213    Description:
214    
215    1. This update changes the default setting for
216    CM.Control.implicit_anchors from true to false.  This means that
217    implicit anchors are no longer permitted by default.  I also tried to
218    make sure that nothing else still relies on implicit anchors.
219    (This is the next step on the schedule towards a CM that does not even
220    have the notion of implicit anchors anymore.)
221    
222    2. More CM manual updates.
223    
224    3. I managed to track down and fix the pickling bug I mentioned last
225    time.  Because of the previously existing workaround, this entails no
226    immediate practical changes.
227    
228    ----------------------------------------------------------------------
229    Name: Matthias Blume
230    Date: 2000/06/06 11:15:00 JST
231    Tag: blume-20000606-lazierpickle
232    Description:
233    
234    !!!! NEW BOOT FILES !!!!
235    
236    * The main purpose of this update is to make library pickles lazier in
237    order to reduce the initial space penalty for autoloading a library.
238    As a result, it is now possible to have $smlnj/compiler.cm
239    pre-registered.  This should take care of the many complaints or
240    inquiries about missing structure Compiler.  This required changes to
241    CM's internal data structures and small tweaks to some algorithms.
242    
243    As a neat additional effect, it is no longer necessary (for the sake
244    of lean heap image files) to distinguish between a "minimal" CM and a
245    "full" CM.  Now, there is only one CM (i.e., the "full" version:
246    $smlnj/cm.cm aka $smlnj/cm/full.cm), and it is always available at the
247    interactive top level. ($smlnj/cm/minimal.cm is gone.)
248    
249    To make the life of compiler-hackers easier, "makeml" now also
250    pre-registers $smlnj/cmb.cm (aka $smlnj/cmb/current.cm).  In other
251    words, after you bootstrap a new sml for the first time, you will not
252    have to autoload $smlnj/cmb.cm again afterwards.  (The first time
253    around you will still have to do it, though.)
254    
255    * A second change consists of major updates to the CM manual.  There
256    are now several appendices with summary information and also a full
257    specification of the CM description file syntax.
258    
259    * In directory src/system I added the script "allcross".  This script
260    invokes sml and cross-compiles the compiler for all supported
261    architectures.  (Useful when providing a new set of boot files.)
262    
263    * There seems to be a latent bug in my "lazy pickles" mechanism.  I
264    added a small tweak to pickle-util.sml to work around this problem,
265    but it is not a proper fix yet.  I will investigate further.  (The
266    effect of the bug was an inflation of library pickle size.)
267    
268    * Version number increased to 110.28.1 (to avoid compatibility problems).
269    
270    ----------------------------------------------------------------------
271    Name: Allen Leung
272    Date: 2000/05/25 17:28 EDT
273    Tag: leunga-20000525-ra
274    Description:
275    
276      Fixed a bug in freezing phase of the register allocator.
277    
278    ----------------------------------------------------------------------
279    Name: Allen Leung
280    Date: 2000/05/15 22:53 EDT
281    Tag: leunga-20000515-alpha-x86-ra
282    Description:
283    
284      1. Alpha
285    
286          Slight cleanup.  Removed the instruction SGNXL
287    
288      2. X86
289    
290          Added the following instructions to the instruction set:
291    
292            ROLx, RORx,
293            BTx, BTSx, BTLx, BTRx,
294            XCHGx, and variants with the LOCK prefix
295    
296      3. Register Allocation
297    
298          The module ra-rewrite-with-renaming has been improved.
299    
300      These have no effect on SML/NJ.
301    
302    ----------------------------------------------------------------------
303    Name: Matthias Blume
304    Date: 2000/05/15 16:20:00 JST
305    Tag: blume-20000515-lightrebuild
306    Description:
307    
308    1. I added an alternative to "-rebuild" to "makeml".  The difference is
309       that prior to calling CMB.make' the CM-variable "LIGHT" will be
310       defined.  In effect, the command will not build any cross-compiler
311       backends and therefore finish more quickly.
312    
313       The "fixpt" script also takes a "-light" switch to be able to use
314       this new facility while compiling for a fixpoint.
315    
316    2. I replaced all mentions of anchored paths in group owner specifications
317       with simple relative paths (usually starting with "..").
318       The rationale is that a library's internal workings should not be
319       compromised by the lack of some anchor.  (An anchor is necessary
320       for someone who wants to refer to the library by an anchored path,
321       but it should not be necessary to build the same library in the first
322       place.)
323    
324    3. I changed the way CM's tool mechanism determines the shell command
325       string used for things like ml-yacc etc. so that it does not break
326       when CM.Control.implicit_anchors is turned off.
327    
328    ----------------------------------------------------------------------
329    Name: Matthias Blume
330    Date: 2000/05/12 18:20:00 JST
331    Tag: blume-20000512-ml-build
332    Description:
333    
334    Fixed a bug in config/_ml-build that prevented ml-yacc and ml-lex from
335    getting installed properly (by config/install.sh).
336    
337    ----------------------------------------------------------------------
338    Name: Matthias Blume
339    Date: 2000/05/12 17:30:00 JST
340    Tag: blume-20000512-anchors
341    Description:
342    
343    !!! NEW BOOT FILES !!!
344    
345    This change is in preparation of fading out support for "implicitly
346    anchored path names".  I went through all sources and used the
347    explicit (and relatively new) $-notation.  See system/README and the
348    CM manual for more info on this.
349    
350    I also modified the anchoring scheme for some things such as "smlnj",
351    "MLRISC", "cm", etc. to take advantage of the fact that explicit
352    anchors are more expressive: anchor name and first arc do not have to
353    coincide.  This entails the following user-visible change:
354    
355    You have to write $smlnj/foo/bar instead of smlnj/foo/bar.  In
356    particular, when you fire up sml with a command-line argument, say,
357    e.g.:
358    
359       sml '$smlnj/cmb.cm'
360    
361    At the ML toplevel prompt:
362    
363       CM.autoload "$smlnj/cmb.cm";
364    
365    There is also a new controller in CM.Control that can be used to turn
366    off all remaining support for implicit anchors by saying:
367    
368        CM.autoload "$smlnj/
369        #set CM.Control.implicit_anchors false;
370    
371    This causes CM to reject implicitly anchored paths.  This is (for the
372    time being) less permissive than the "final" version where there will
373    be no more such implicit anchors and relative paths will be just that:
374    relative.
375    
376    The next step (version after next version?) will be to make the
377    default for CM.Control.implicit_anchors false.  After the dust has
378    settled, I can then produce the "final" version of this...
379    
380    Note: Since bootstrapping is a bit tricky, I provided new boot files.
381    
382    ----------------------------------------------------------------------
383    Name: Matthias Blume
384    Date: 2000/05/11 16:30:00 JST
385    Tag: blume-20000511-sources
386    Description:
387    
388    The main change is that I added function CM.sources as a generalized
389    version of the earlier CM.makedepend.  This entails the following
390    additional changes:
391    
392      - CM.makedepend has been dropped.
393    
394      - CM manual has been updated.
395    
396      - TOOLS signature and API have been changed.
397    
398    ----------------------------------------------------------------------
399  Name: Allen Leung  Name: Allen Leung
400  Date: 2000/05/10 21:17 EDT  Date: 2000/05/10 21:17 EDT
401  Tag: leunga-20000510-moby-c--ssa  Tag: leunga-20000510-moby-c--ssa
# Line 374  Line 758 
758    
759  3.  Assembly  3.  Assembly
760    
761      When generating assemby, resolve the value of client defined constants,      When generating assembly, resolve the value of client defined constants,
762      instead of generating symbolic values.  This is controlled by the      instead of generating symbolic values.  This is controlled by the
763      new flag "asm-resolve-constants", which is default to true.      new flag "asm-resolve-constants", which is default to true.
764    
# Line 397  Line 781 
781    
782  To this end, I arranged that instead of "structure Core" as "structure  To this end, I arranged that instead of "structure Core" as "structure
783  _Core" is bound in the pervasive environment.  Core access is done via  _Core" is bound in the pervasive environment.  Core access is done via
784  _Core (which can never be accidentially rebound because _Core is not a  _Core (which can never be accidentally rebound because _Core is not a
785  legal surface-syntax symbol).  legal surface-syntax symbol).
786    
787  The current solution is much cleaner because the core environment is  The current solution is much cleaner because the core environment is
# Line 407  Line 791 
791  with dynamic and symbolic parts of the core environment.  with dynamic and symbolic parts of the core environment.
792    
793  Remaining hackery (to bind the "magic" symbol _Core) is localized in the  Remaining hackery (to bind the "magic" symbol _Core) is localized in the
794  compilation mananger's bootstrap compiler (actually: in the "init group"  compilation manager's bootstrap compiler (actually: in the "init group"
795  handling).  See the comments in src/system/smlnj/init/init.cmi for  handling).  See the comments in src/system/smlnj/init/init.cmi for
796  more details.  more details.
797    
# Line 522  Line 906 
906     (specified in the .cm file at each instance where the tool's class is     (specified in the .cm file at each instance where the tool's class is
907     used).     used).
908    
909     This was done to accomodate the new "make" and "shell" tools which     This was done to accommodate the new "make" and "shell" tools which
910     facilitate fairly seemless hookup to portions of code managed using     facilitate fairly seamless hookup to portions of code managed using
911     Makefiles or Shell scripts.     Makefiles or Shell scripts.
912    
913     There are no classes "shared" or "private" anymore.  Instead, the     There are no classes "shared" or "private" anymore.  Instead, the
# Line 535  Line 919 
919    
920     All existing tools are described in the CM manual.     All existing tools are described in the CM manual.
921    
922     - Slightly better error handling.  (CM now surpresses many followup     - Slightly better error handling.  (CM now suppresses many followup
923     error messages that tended to be more annoying than helpful.)     error messages that tended to be more annoying than helpful.)
924    
925  2. Major changes to the compiler's static environment data structures.  2. Major changes to the compiler's static environment data structures.
# Line 669  Line 1053 
1053    
1054       I've changed andl to testl in the floating point test sequence       I've changed andl to testl in the floating point test sequence
1055       whenever appropriate.  The Intel optimization guide states that       whenever appropriate.  The Intel optimization guide states that
1056       testl is perferable to andl.       testl is preferable to andl.
1057    
1058  7. RA (x86 only)  7. RA (x86 only)
1059    
# Line 851  Line 1235 
1235    
1236  1. Tools.registerStdShellCmdTool (from smlnj/cm/tool.cm) takes an  1. Tools.registerStdShellCmdTool (from smlnj/cm/tool.cm) takes an
1237  additional argument called "template" which is an optional string that  additional argument called "template" which is an optional string that
1238  specifiel the layout of the tool command line.  See the CM manual for  specifies the layout of the tool command line.  See the CM manual for
1239  explanation.  explanation.
1240    
1241  2. A special-purpose tool can be "regisitered" by simply dropping the  2. A special-purpose tool can be "registered" by simply dropping the
1242  corresponding <...>-tool.cm (and/or <...>-ext.cm) into the same  corresponding <...>-tool.cm (and/or <...>-ext.cm) into the same
1243  directory where the .cm file lives that uses this tool.  (The  directory where the .cm file lives that uses this tool.  (The
1244  behavior/misfeature until now was to look for the tool description  behavior/misfeature until now was to look for the tool description
# Line 898  Line 1282 
1282  rebuild.  Having sets of the form <base><k>.{bin,boot}.<arch>-unix for  rebuild.  Having sets of the form <base><k>.{bin,boot}.<arch>-unix for
1283  <k>=1,2,... is normally not a good idea when invoking fixpt.  However,  <k>=1,2,... is normally not a good idea when invoking fixpt.  However,
1284  they might be the result of an earlier partial run of fixpt (which  they might be the result of an earlier partial run of fixpt (which
1285  perhaps got accidentially killed).  In this case, fixpt will quickly  perhaps got accidentally killed).  In this case, fixpt will quickly
1286  move through what exists before continuing where it left off earlier,  move through what exists before continuing where it left off earlier,
1287  and, thus, saves a lot of time.  and, thus, saves a lot of time.
1288    
# Line 948  Line 1332 
1332    it from that remote directory.    it from that remote directory.
1333    This should simplify installation further:  For machines that have    This should simplify installation further:  For machines that have
1334    access to the internet, just fetch <version>-config.tgz, unpack it,    access to the internet, just fetch <version>-config.tgz, unpack it,
1335    edit config/targets, and go (run config/install.sh).  The scipt will    edit config/targets, and go (run config/install.sh).  The script will
1336    fetch everything else that it might need all by itself.    fetch everything else that it might need all by itself.
1337    
1338    For CVS users, this mechanism is not relevant for source archives, but    For CVS users, this mechanism is not relevant for source archives, but

Legend:
Removed from v.641  
changed lines
  Added in v.671

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