Diff of /sml/trunk/HISTORY

Parent Directory | Revision Log | Patch

	revision 660, Thu Jun 15 04:40:56 2000 UTC	revision 703, Thu Sep 7 05:43:23 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: 2000/09/07 14:45:00 JST
17		Tag: blume-20000907-cmerrmsg
18		Description:
19
20		Small tweak to CM to avoid getting ML syntax error messages twice.
21
22		----------------------------------------------------------------------
23		Name: Matthias Blume
24		Date: 2000/08/31 18:00:00 JST
25		Tag: blume-20000831-cvsbootfiles
26		Description:
27
28		New URL for boot files (because the 110.29 files on the BL server do
29		now work correctly with my updated install scripts for yacc and lex).
30
31		----------------------------------------------------------------------
32		Name: Matthias Blume
33		Date: 2000/08/08 12:33:00 JST
34		Tag: blume-20000808-manual
35		Description:
36
37		Tiny update to CM manual.
38
39		----------------------------------------------------------------------
40		Name: Allen Leung
41		Date: 2000/08/7 19:31:00
42		Tag: leunga-20000807-a-whole-bunch-of-stuff
43
44		Moby, C--, SSA, x86, machine descriptions etc.  Should only affect C--
45		and Mobdy.
46
47		1.  x86
48
49		a.  Fixes to peephole module by John and Dan.
50		b.  Assembly fix to SETcc by Allen.
51		c.  Fix to c-call by John.
52		d.  Fix to spilling by John.  (This one deals with the missing FSTPT case)
53		e.  Instruction selection optimization to SETcc as suggested by John.
54
55		For example,
56
57		MV(32, x, COND(32, CMP(32, LT, a, b), LI 1, LI 0))
58
59		should generate:
60
61		MOVL a, x
62		SUBL b, x
63		SHRL 31, x
64
65		2.  IR stuff
66
67		A bunch of new DJ-graph related algorithms added.  These
68		speed up SSA construction.
69
70		3.  SSA + Scheduling
71
72		Added code for SSA and scheduling to the repository
73
74		----------------------------------------------------------------------
75		Name: Lal George
76		Date: 2000/07/27 11:53:14 EDT
77
78		Tag: lal-20000727-linux-ppc
79		Description:
80
81		Made changes to support Linux PPC.
82		p.s. I have confirmation that the 110.29 boot files work fine.
83
84		----------------------------------------------------------------------
85		Name: Matthias Blume
86		Date: 2000/07/27 17:40:00 JST
87		Tag: blume-20000727-scripts
88		Description:
89
90		!!!! WARNING !!!!
91		You must recompile the runtime system!
92		!!!! WARNING !!!!
93
94		This is basically another round of script-enhancements:
95
96		1. sml, ml-build, and ml-makedepend accept options -D and -U to define
97		and undefine CM preprocessor symbols.
98
99		2. ml-build avoids generating a new heap image if it finds that the
100		existing one is still ok.  (The condition is that no ML file had to
101		be recompiled and all ML files are found to be older that the heap
102		file.)
103
104		To make this work smoothly, I also hacked the runtime system as
105		well as SMLofNJ.SysInfo to get access to the heap image suffix
106		(.sparc-solaris, ...) that is currently being used.
107
108		Moreover, the signature of CM.mk_standalone has changed.  See the
109		CM manual.
110
111		3. ml-makedepend accepts additional options -n, -a, and -o.  (See the
112		CM manual for details.)
113
114		4. More CM manual updates:
115		- all of the above has been documented.
116		- there is now a section describing the (CM-related) command line
117		arguments that are accepted by the "sml" command
118
119		----------------------------------------------------------------------
120		Name: Matthias Blume
121		Date: 2000/07/25 16:20:00 JST
122		Tag: blume-20000725-makedepend
123		Description:
124
125		Added a script called ml-makedepend.  This can be used in makefiles
126		for Unix' make in a way very similar to the "makedepend" command for
127		C.
128
129		The script internally uses function CM.sources.
130
131		Synopsis:
132
133		ml-makedepend [-f makefile] cmfile targetname
134
135		The default for the makefile is "makefile" (or "Makefile" should
136		"makefile" not exist).
137
138		ml-makedepend adds a cmfile/targetname-specific section to this
139		makefile (after removing the previous version of this section).  The
140		section contains a single dependency specification with targetname on
141		the LHS (targetname is an arbitrary name), and a list of files derived
142		from the cmfile on the RHS.  Some of the files on the RHS are
143		ARCH/OPSYS-specific.  Therefore, ml-makedepend inserts references to
144		"make" variables $(ARCH) and $(OPSYS) in place of the corresponding
145		path names.  The makefile writer is responsible for making sure that
146		these variables have correct at the time "make" is invoked.
147
148		----------------------------------------------------------------------
149		Name: Matthias Blume
150		Date: 2000/07/22 23:30:00 JST
151		Tag: blume-20000722-urlupdate
152		Description:
153
154		Changed BOOT and config/srcarchiveurl to point to BL server:
155
156		ftp://ftp.research.bell-labs.com/dist/smlnj/working/110.29/
157
158		----------------------------------------------------------------------
159		Name: Matthias Blume
160		Date: 2000/07/18 18:00:00 JST
161		Tag: blume-20000718-Version_110_29
162		Description:
163
164		1. Updated src/compiler/TopLevel/main/version.sml to version 110.29
165
166		2. Updated config/version to 110.29
167
168		3. Updated config/srcarchiveurl
169
170		3. New boot files!
171		ftp://ftp.cs.princeton.edu/pub/people/blume/sml/110.29-autofetch
172
173		----------------------------------------------------------------------
174		Name: Matthias Blume
175		Date: 2000/07/11 13:58:00 JST
176		Tag: blume-20000711-doctypo
177		Description:
178
179		Fixed a few typos in CM manual.
180
181		----------------------------------------------------------------------
182		Name: Allen Leung
183		Date: 2000/06/15 00:38:00
184		Tag: leunga-20000704-sparc-x86
185
186		1. x86 peephole improvement sp += k; sp -= k => nop  [from John]
187		2. fix to x86 RET bug [found by Dan Grossman]
188		3. sparc assembly bug fix for ticc instructions [found by Fermin]
189
190		Affects c-- and moby only
191
192		----------------------------------------------------------------------
193		Name: Matthias Blume
194		Date: 2000/07/04 15:26:00
195		Tag: blume-20000704-trigger
196		Description:
197
198		1. Improvements to CM manual.
199		2. SMLofNJ.Internals.BTrace.trigger reinstated as an alternative way
200		of getting a back-trace.  The function, when called, raises an
201		internal exception which explicitly carries the full back-trace history,
202		so it is unaffected by any intervening handle-raise pairs ("trivial"
203		or not).  The interactive loop will print that history once it arrives
204		at top level.
205		Short of having all exceptions implicitly carry the full history, the
206		recommended way of using this facility is:
207		- compile your program with instrumentation "on"
208		- run it, when it raises an exception, look at the history
209		- if the history is "cut off" because of some handler, go and modify
210		your program so that it explicitly calls BTrace.trigger
211		- recompile (still instrumented), and rerun; look at the full history
212
213		----------------------------------------------------------------------
214		Name: Matthias Blume
215		Date: 2000/07/03 15:36:00 JST
216		Tag: blume-20000702-manual
217		Description:
218
219		Small corrections and updates to CM manual.
220
221		----------------------------------------------------------------------
222		Name: Matthias Blume
223		Date: 2000/06/29 16:04:00 JST
224		Tag: blume-20000629-yacctool
225		Description:
226
227		Changes:
228
229		1. Class "mlyacc" now takes separate arguments to pass options to
230		generated .sml- and .sig-files independently.
231		2. Corresponding CM manual updates.
232		3. BTrace module now also reports call sites.  (However, for loop clusters
233		it only shows from where the cluster was entered.)  There are associated
234		modifications to core.sml, internals.{sig,sml}, btrace.sml, and btimp.sml.
235
236		----------------------------------------------------------------------
237		Name: Matthias Blume
238		Date: 2000/06/27 16:51:00 JST
239		Tag: blume-20000627-noweb
240		Description:
241
242		Changes:
243
244		1. Implemented "subdir" and "witness" options for noweb tool.
245		This caused some slight internal changes in CM's tool implementation.
246		2. Fixed bug in "tool plugin" mechanism.  This is essentially cleaning
247		some remaining issues from earlier path anchor changes.
248		3. Updated CM manual accordingly.
249
250		4. Changed implementation of back-tracing so that I now consider it
251		ready for prime-time.
252
253		In particular, you don't have to explicitly trigger the back-trace
254		anymore.  Instead, if you are running BTrace-instrumented code and
255		there is an uncaught exception (regardless of whether or not it was
256		raised in instrumented code), the top-level evalloop will print
257		the back-trace.
258
259		Features:
260
261		- Instrumented and uninstrumented code work together seemlessly.
262		(Of course, uninstrumented code is never mentioned in actual
263		back-traces.)
264
265		- Asymptotic time- and space-complexity of instrumented code is
266		equal to that of uninstrumented code.  (This means that
267		tail-recursion is preserved by the instrumentation phase.)
268
269		- Modules whose code has been instrumented in different sessions
270		work together without problem.
271
272		- There is no penalty whatsoever on uninstrumented code.
273
274		- There is no penalty on "raise" expressions, even in
275		instrumented code.
276
277		A potential bug (or perhaps it is a feature, too):
278
279		A back-trace reaches no further than the outermost instrumented
280		non-trivial "raise".  Here, a "trivial" raise is one that is the
281		sole RHS of a "handle" rule.  Thus, back-traces reach trough
282
283		<exp> handle e => raise e
284
285		and even
286
287		<exp> handle Foo => raise Bar
288
289		and, of course, through
290
291		<exp> handle Foo => ...
292
293		if the exception was not Foo.
294
295		Back-traces always reach right through any un-instrumented code
296		including any of its "handle" expressions, trivial or not.
297
298		To try this out, do the following:
299
300		- Erase all existing binfiles for your program.
301		(You may keep binfiles for those modules where you think you
302		definitely don't need back-tracing.)
303		- Turn on back-trace instrumentation:
304		SMLofNJ.Internals.BTrace.mode (SOME true);
305		- Recompile your program.  (I.e., run "CM.make" or "use".)
306		- You may now turn instrumentation off again (if you want):
307		SMLofNJ.Internals.BTrace.mode (SOME false);
308		- Run your program as usual.  If it raises an exception that
309		reaches the interactive toplevel, then a back-trace will
310		automatically be printed.  After that, the toplevel loop
311		will print the exception history as usual.
312
313		----------------------------------------------------------------------
314		Name: Matthias Blume
315		Date: 2000/06/26 09:56:46 JST
316		Tag: blume-20000626-setup
317		Description:
318
319		CM: - setup-parameter to "sml" added; this can be used to run arbitrary
320		ML code before and after compiling a file (e.g., to set compiler
321		flags)
322
323		Compiler: - improved btrace API (in core.sml, internals.{sig,sml})
324		- associated changes to btrace.sml (BTrace instrumentation pass)
325		- cleaner implementation of btimp.sml (BTrace tracing and report
326		module)
327
328		CM manual: * new path encoding documented
329		* description of setup-parameter to "sml" added
330
331		The biggest user-visible change to back-tracing is that it is no
332		longer necessary to compile all traced modules within the same
333		session.  (This was a real limitation.)
334
335		----------------------------------------------------------------------
336		Name: Matthias Blume
337		Date: 2000/06/24 12:40:00 JST
338		Tag: blume-20000624-startup
339		Description:
340
341		Fixes startup slowdown problem.  (I was calling SrcPath.sync a _tad_
342		bit too often -- to put it mildly. :)
343
344		----------------------------------------------------------------------
345		Name: Matthias Blume
346		Date: 2000/06/23 18:20:00 JST
347		Tag: blume-20000623-btrace
348		Description:
349
350		This updates adds a backtrace facility to aid programmers in debugging
351		their programs.  This involves the following changes:
352
353		1. Module system/smlnj/init/core.sml (structure _Core) now has hooks for
354		keeping track of the current call stack.  When programs are compiled
355		in a special mode, the compiler will insert calls to these hooks
356		into the user program.
357		"Hook" means that it is possible for different implementations of
358		back-tracing to register themselves (at different times).
359
360		2. compiler/MiscUtil/profile/btrace.sml implements the annotation phase
361		as an Absyn.dec->Absyn.dec rewrite.  Normally this phase is turned off.
362		It can be turned on using this call:
363		SMLofNJ.Internals.BTrace.mode (SOME true);
364		Turning it off again:
365		SMLofNJ.Internals.BTrace.mode (SOME false);
366		Querying the current status:
367		SMLofNJ.Internals.BTrace.mode NONE;
368		Annotated programs are about twice as big as normal ones, and they
369		run a factor of 2 to 4 slower with a dummy back-trace plugin (one
370		where all hooks do nothing).  The slowdown with a plugin that is
371		actually useful (such as the one supplied by default) is even greater,
372		but in the case of the default plugin it is still only an constant
373		factor (amortized).
374
375		3. system/Basis/Implementation/NJ/internals.{sig,sml} have been augmented
376		with a sub-structure BTrace for controlling back-tracing.  In particular,
377		the above-mentioned function "mode" controls whether the annotation
378		phase is invoked by the compiler.  Another important function is
379		"trigger": when called it aborts the current execution and causes
380		the top-level loop to print a full back-trace.
381
382		4. compiler/MiscUtil/profile/btimp.sml is the current default plugin
383		for back-tracing.  It keeps track of the dynamic call stack and in
384		addition to that it keeps a partial history at each "level" of that
385		stack.  For example, if a tail-calls b, b tail-calls c, and c tail-calls
386		d and b (at separate times, dynamically), then the report will show:
387
388		GOTO   d
389		/c
390		GOTO  \b
391		CALL   a
392
393		This shows that there was an initial non-tail call of a, then a
394		tail-call to b or c, looping behavior in a cluster of functions that
395		consist of b and c, and then a goto from that cluster (i.e., either from
396		b or from c) to d.
397
398		Note that (depending on the user program) the amount of information
399		that the back-trace module has to keep track of at each level is bounded
400		by a constant.  Thus, the whole implementation has the same asymptotical
401		complexity as the original program (both in space and in time).
402
403		5. compiler/TopLevel/interact/evalloop.sml has been modified to
404		handle the special exception SMLofNJ.Internals.BTrace.BTrace
405		which is raised by the "trigger" function mentioned above.
406
407		Notes on usage:
408
409		- Annotated code works well together with unannotated code:
410		Unannotated calls simply do not show up at all in the backtrace.
411
412		- It is not a good idea to let modules that were annotated during
413		different sessions run at the same time.  This is because the compiler
414		chooses small integers to identify individual functions, and there
415		will be clashes if different modules were compiled in separate sessions.
416		(Nothing will crash, and you will even be told about the clashes, but
417		back-trace information will in general not be useful.)
418
419		- Back-tracing can be confused by callcc and capture.
420
421		- The only way of getting a back-trace right now is to explicitly
422		invoke the "trigger" function from your user program.  Eventually, we
423		should make every exception carry back-trace information (if
424		available).  But since this creates more overhead at "raise"-time
425		(similar to the current exnHistory overhead), I have not yet
426		implemented this.  (The implementation will be rather easy.)  With
427		exceptions carrying back-trace information, this facility will be even
428		more useful because users don't need to modify their programs...
429
430		- While it is possible to compile the compiler with back-trace
431		annotations turned on (I did it to get some confidence in
432		correctness), you must make absolutely sure that core.sml and
433		btimp.sml are compiled WITHOUT annotation!  (core.sml cannot actually
434		be compiled with annotation because there is no core access yet, but
435		if you compile btimp.sml with annotation, then the system will go into
436		an infinite recursion and crash.)
437		Since CM currently does not know about BTrace, the only way to turn
438		annotations on and off for different modules of the compiler is to
439		interrupt CMB.make, change the settings, and re-invoke it.  Of course,
440		this is awkward and clumsy.
441
442		Sample sessions:
443
444		Standard ML of New Jersey v110.28.1 [FLINT v1.5], June 5, 2000
445		- SMLofNJ.Internals.BTrace.mode (SOME true);
446		[autoloading]
447		[autoloading done]
448		val it = false : bool
449		- structure X = struct
450		-     fun main n = let
451		-         fun a (x, 0) = d x
452		-           | a (x, n) = b (x, n - 1)
453		-         and b (x, n) = c (x, n)
454		-         and c (x, n) = a (x, n)
455		-         and d x = e (x, 3)
456		-         and e (x, 0) = f x
457		-           | e (x, n) = e (x, n - 1)
458		-         and f 0 = SMLofNJ.Internals.BTrace.trigger ()
459		-           | f n = n * g (n - 1)
460		-         and g n = a (n, 3)
461		-     in
462		-         f n
463		-     end
464		- end;
465		structure X : sig val main : int -> int end
466		- X.main 3;
467		*** BACK-TRACE ***
468		GOTO   stdIn:4.2-13.20: X.main[2].f
469		GOTO-( stdIn:4.2-13.20: X.main[2].e
470		GOTO   stdIn:4.2-13.20: X.main[2].d
471		/ stdIn:4.2-13.20: X.main[2].a
472		| stdIn:4.2-13.20: X.main[2].b
473		GOTO-\ stdIn:4.2-13.20: X.main[2].c
474		CALL   stdIn:4.2-13.20: X.main[2].g
475		GOTO   stdIn:4.2-13.20: X.main[2].f
476		GOTO-( stdIn:4.2-13.20: X.main[2].e
477		GOTO   stdIn:4.2-13.20: X.main[2].d
478		/ stdIn:4.2-13.20: X.main[2].a
479		| stdIn:4.2-13.20: X.main[2].b
480		GOTO-\ stdIn:4.2-13.20: X.main[2].c
481		CALL   stdIn:4.2-13.20: X.main[2].g
482		GOTO   stdIn:4.2-13.20: X.main[2].f
483		GOTO-( stdIn:4.2-13.20: X.main[2].e
484		GOTO   stdIn:4.2-13.20: X.main[2].d
485		/ stdIn:4.2-13.20: X.main[2].a
486		| stdIn:4.2-13.20: X.main[2].b
487		GOTO-\ stdIn:4.2-13.20: X.main[2].c
488		CALL   stdIn:4.2-13.20: X.main[2].g
489		GOTO   stdIn:4.2-13.20: X.main[2].f
490		CALL   stdIn:2.15-17.4: X.main[2]
491		-
492
493		(Note that because of a FLINt bug the above code currently does not
494		compile without BTrace turned on.)
495
496		Here is another example, using my modified Tiger compiler:
497
498		Standard ML of New Jersey v110.28.1 [FLINT v1.5], June 5, 2000
499		- SMLofNJ.Internals.BTrace.mode (SOME true);
500		[autoloading]
501		[autoloading done]
502		val it = false : bool
503		- CM.make "sources.cm";
504		[autoloading]
505		...
506		[autoloading done]
507		[scanning sources.cm]
508		[parsing (sources.cm):parse.sml]
509		[creating directory CM/SKEL ...]
510		[parsing (sources.cm):tiger.lex.sml]
511		...
512		[wrote CM/sparc-unix/semant.sml]
513		[compiling (sources.cm):main.sml]
514		[wrote CM/sparc-unix/main.sml]
515		[New bindings added.]
516		val it = true : bool
517		- Main.compile ("../testcases/merge.tig", "foo.out");
518		*** BACK-TRACE ***
519		CALL   lib/semant.sml:99.2-396.21: SemantFun[2].transExp.trvar
520		CALL   lib/semant.sml:99.2-396.21: SemantFun[2].transExp.trexp
521		CALL   lib/semant.sml:289.3-295.22: SemantFun[2].transExp.trexp.check[2]
522		GOTO   lib/semant.sml:289.3-295.22: SemantFun[2].transExp.trexp.check[2]
523		CALL   lib/semant.sml:99.2-396.21: SemantFun[2].transExp.trexp
524		CALL   lib/semant.sml:99.2-396.21: SemantFun[2].transExp.trexp
525		CALL   lib/semant.sml:488.3-505.6: SemantFun[2].transDec.trdec[2].transBody[2]
526		/ lib/semant.sml:411.65-543.8: SemantFun[2].transDec
527		CALL-\ lib/semant.sml:413.2-540.9: SemantFun[2].transDec.trdec[2]
528		CALL   lib/semant.sml:99.2-396.21: SemantFun[2].transExp.trexp
529		CALL   lib/semant.sml:8.52-558.4: SemantFun[2].transProg[2]
530		CALL   main.sml:1.18-118.4: Main.compile[2]
531		-
532
533		----------------------------------------------------------------------
534		Name: Matthias Blumen
535		Date: 2000/06/21 18:00:00 JST
536		Tag: blume-20000621-manual
537		Description:
538
539		CM manual update: Path environments documented.
540
541		----------------------------------------------------------------------
542		Name: Matthias Blume
543		Date: 2000/06/19 13:40:00
544		Tag: blume-20000619-manual
545		Description:
546
547		CM manual and system/README update.  This only covers the fact that
548		there are no more implicit anchors.  (Path environments and the "bind"
549		option to "cm" have yet to be documented.)
550
551		----------------------------------------------------------------------
552		Name: Matthias Blume
553		Date: 2000/06/19 11:05:00 JST
554		Tag: blume-20000619-chdir-bugfix
555		Description:
556
557		Fixed a bug in new SrcPath module that sometimes led to a bad chDir call.
558
559		----------------------------------------------------------------------
560		Name: Matthias Blume
561		Date: 2000/06/18 22:00:10 JST
562		Tag: blume-20000618-implicit-anchors-really-gone
563		Description:
564
565		I updates the previous HISTORY entry where I forgot to mention that
566		implicit anchors are no longer with us.
567
568		The current update also gets rid of the (now useless) controller
569		CM.Control.implicit_anchors.
570
571		----------------------------------------------------------------------
572		Name: Matthias Blume
573		Date: 2000/06/16 17:30:00 JST
574		Tag: blume-20000616-anchorenv
575		Description:
576
577		This patch implements the long anticipated (just kidding :) "anchor
578		environment" mechanism.  In the course of doing this, I also
579		re-implemented CM's internal "SrcPath" module from scratch.  The new
580		one should be more robust in certain boundary cases.  In any case, it
581		is a lot cleaner than its predecessor (IMHO).
582
583		This time, although there is yet another boot file format change, I
584		kept the unpickler backward-compatible.  As a result, no new bootfiles
585		are necessary and bootstrapping is straightforward.  (You cannot read
586		new bootfiles into an old system, but the other way around is no
587		problem.)
588
589		Visible changes:
590
591		** 0. Implicit path anchors (without the leading $-symbol) are no
592		longer recognized at all. This means that such path names are not
593		illegal either.  For example, the name basis.cm simply refers to a
594		local file called "basis.cm" (i.e, the name is an ordinary path
595		relative to .cm-files directory).  Or, to put it differently, only
596		names that start with $ are anchored paths.
597
598		** 1. The $<singlearc> abbreviation for $/<singlearc> has finally
599		vanished.
600
601		John (Reppy) had critizised this as soon as I originally proposed and
602		implemented it, but at that time I did not really deeply believe
603		him. :) Now I came full-circle because I need the $<singlearc> syntax
604		in another place where it cannot be seen as an abbreviation for
605		$/<singlearc>.  To avoid the confusion, $<singlearc> now means what it
606		seems to mean (i.e., it "expands" into the corresponding anchor
607		value).
608
609		However, when paths are used as members in CM description files, it
610		continues to be true that there must be at least another arc after the
611		anchor.  This is now enforced separately during semantic analysis
612		(i.e., from a lexical/syntactical point of view, the notation is ok.)
613
614		** 2. The "cm" class now accepts an option "bind".  The option's value
615		is a sub-option list of precisely two items -- one labeled "anchor"
616		and the other one labeled "value".  As you might expect, "anchor" is
617		used to specify an anchor name to be bound, and "value" specifies what
618		the anchor is being bound to.
619
620		The value must be a directory name and can be given in either standard
621		syntax (including the possibility that it is itself an anchored path)
622		or native syntax.
623
624		Examples:
625
626		foo.cm (bind:(anchor:bar value:$mystuff/bar))
627		lib.cm (bind:(anchor:a value:"H:\\x\\y\\z"))  (* only works under windows *)
628
629		and so on.
630
631		The meaning of this is that the .cm-file will be processed with an
632		augmented anchor environment where the given anchor(s) is/are bound to
633		the given values(s).
634
635		The rationale for having this feature is this: Suppose you are trying
636		to use two different (already stable) libraries a.cm and b.cm (that
637		you perhaps didn't write yourself).  Further, suppose each of these
638		two libraries internally uses its own auxiliary library $aux/lib.cm.
639		Normally you would now have a problem because the anchor "lib" can not
640		be bound to more than one value globally.  Therefore, the project that
641		uses both a.cm and b.cm must locally redirect the anchor to some other
642		place:
643
644		a.cm (bind:(anchor:lib value:/usr/lib/smlnj/a-stuff))
645		b.cm (bind:(anchor:lib value:/usr/lib/smlnj/b-stuff))
646
647		This hard-wires $lib/aux.cm to /usr/lib/smlnj/a-stuff/aux.cm or
648		/usr/lib/smlnj/b-stuff/aux.cm, respectively.
649
650		Hard-wiring path names is a bit inflexible (and CM will verbosely warn
651		you when you do so at the time of CM.stabilize).  Therefore, you can
652		also use an anchored path as the value:
653
654		a.cm (bind:(anchor:lib value:$a-lib))
655		b.cm (bind:(anchor:lib value:$b-lib))
656
657		Now you can globally configure (using the usual CM.Anchor.anchor or
658		pathconfig machinery) bindings for "a-lib" and "b-lib".  Since "lib"
659		itself is always locally bound, setting it globally is no longer
660		meaningful or necessary (but it does not hurt either).  In fact, "lib"
661		can still be used as a global anchor for separate purposes.  As a
662		matter of fact, one can locally define "lib" in terms of a global
663		"lib":
664
665		a.cm (bind:(anchor:lib value:$lib/a))
666		b.cm (bind:(anchor:lib value:$lib/b))
667
668		** 3: The encoding of path names has changed.  This affects the way
669		path names are shown in CM's progress report and also the internal
670		protocol encoding used for parallel make.
671
672		The encoding now uses one or more ':'-separated segments.  Each
673		segments corresponds to a file that has been specified relative to the
674		file given by its preceding segment.  The first segment is either
675		relative to the CWD, absolute, or anchored.  Each segment itself is
676		basically a Unix pathname; all segments but the first are relative.
677
678		Example:
679
680		$foo/bar/baz.cm:a/b/c.sml
681
682		This path denotes the file bar/a/b/c.sml relative to the directory
683		denoted by anchor "foo".  Notice that the encoding also includes
684		baz.cm which is the .cm-file that listed a/b/c.sml.  As usual, such
685		paths are resolved relative to the .cm-files directory, so baz.cm must
686		be ignored to get the "real" pathname.
687
688		To make this fact more obvious, CM puts the names of such "virtual
689		arcs" into parentheses when they appear in progress reports. (No
690		parentheses will appear in the internal protocol encoding.)  Thus,
691		what you really see is:
692
693		$foo/bar/(baz.cm):a/b/c.sml
694
695		I find this notation to be much more informative than before.
696
697		Another new feature of the encoding is that special characters
698		including parentheses, colons, (back)slashes, and white space are
699		written as \ddd (where ddd is the decimal encoding of the character).
700
701		*** The CM manual still needs to be updated.
702
703		----------------------------------------------------------------------
704	Name: Allen Leung	Name: Allen Leung
705	Date: 2000/06/15 00:38:00	Date: 2000/06/15 00:38:00
706	Tag: leunga-20000615-x86-peephole	Tag: leunga-20000615-x86-peephole

---

Colored Diff Long Colored Diff Unidiff Context Diff Side by Side

Legend: Removed from v.660   changed lines   Added in v.703

---

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