Diff of /sml/trunk/HISTORY

Parent Directory | Revision Log | Patch

	revision 580, Wed Mar 22 06:33:52 2000 UTC	revision 1130, Mon Mar 11 04:49:41 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
17		Date: 2002/03/10 23:55:00 EST
18		Tag: leunga-20020310-x86-call
19		Description:
20
21		Added machine generation for CALL instruction (relative displacement mode)
22
23		----------------------------------------------------------------------
24		Name: Matthias Blume
25		Date: 2002/03/08 16:05:00
26		Tag: blume-20020308-entrypoints
27		Description:
28
29		Version number bumped to 110.39.1.  NEW BOOTFILES!
30
31		Entrypoints: non-zero offset into a code object where execution should begin.
32
33		- Added the notion of an entrypoint to CodeObj.
34		- Added reading/writing of entrypoint info to Binfile.
35		- Made runtime system bootloader aware of entrypoints.
36		- Use the address of the label of the first function given to mlriscGen
37		as the entrypoint.  This address is currently always 0, but it will
38		not be 0 once we turn on block placement.
39		- Removed the linkage cluster code (which was The Other Way(tm) of dealing
40		with entry points) from mlriscGen.
41
42		----------------------------------------------------------------------
43		Name: Allen Leung
44		Date: 2002/03/07 20:45:00 EST
45		Tag: leunga-20020307-x86-cmov
46		Description:
47
48		Bug fixes for CMOVcc on x86.
49
50		1. Added machine code generation for CMOVcc
51		2. CMOVcc is now generated in preference over SETcc on PentiumPro or above.
52		3. CMOVcc cannot have an immediate operand as argument.
53
54		----------------------------------------------------------------------
55		Name: Matthias Blume
56		Date: 2002/03/07 16:15:00 EST
57		Tag: blume-20020307-controls
58		Description:
59
60		This is a very large but mostly boring patch which makes (almost)
61		every tuneable compiler knob (i.e., pretty much everything under
62		Control.* plus a few other things) configurable via both the command
63		line and environment variables in the style CM did its configuration
64		until now.
65
66		Try starting sml with '-h' (or, if you are brave, '-H')
67
68		To this end, I added a structure Controls : CONTROLS to smlnj-lib.cm which
69		implements the underlying generic mechanism.
70
71		The interface to some of the existing such facilities has changed somewhat.
72		For example, the MLRiscControl module now provides mkFoo instead of getFoo.
73		(The getFoo interface is still there for backward-compatibility, but its
74		use is deprecated.)
75
76		The ml-build script passes -Cxxx=yyy command-line arguments through so
77		that one can now twiddle the compiler settings when using this "batch"
78		compiler.
79
80		TODO items:
81
82		We should go through and throw out all controls that are no longer
83		connected to anything.  Moreover, we should go through and provide
84		meaningful (and correct!) documentation strings for those controls
85		that still are connected.
86
87		Currently, multiple calls to Controls.new are accepted (only the first
88		has any effect).  Eventually we should make sure that every control
89		is being made (via Controls.new) exactly once.  Future access can then
90		be done using Controls.acc.
91
92		Finally, it would probably be a good idea to use the getter-setter
93		interface to controls rather than ref cells.  For the time being, both
94		styles are provided by the Controls module, but getter-setter pairs are
95		better if thread-safety is of any concern because they can be wrapped.
96
97		*****************************************
98
99		One bug fix: The function blockPlacement in three of the MLRISC
100		backpatch files used to be hard-wired to one of two possibilities at
101		link time (according to the value of the placementFlag).  But (I
102		think) it should rather sense the flag every time.
103
104		*****************************************
105
106		Other assorted changes (by other people who did not supply a HISTORY entry):
107
108		1. the cross-module inliner now works much better (Monnier)
109		2. representation of weights, frequencies, and probabilities in MLRISC
110		changed in preparation of using those for weighted block placement
111		(Reppy, George)
112
113		----------------------------------------------------------------------
114		Name: Lal George
115		Date: 2002/03/07 14:44:24 EST 2002
116		Tag: george-20020307-weighted-block-placement
117
118		Tested the weighted block placement optimization on all architectures
119		(except the hppa) using AMPL to generate the block and edge frequencies.
120		Changes were required in the machine properties to correctly
121		categorize trap instructions. There is an MLRISC flag
122		"weighted-block-placement" that can be used to enable weighted block
123		placement, but this will be ineffective without block/edge
124		frequencies (coming soon).
125
126
127		----------------------------------------------------------------------
128		Name: Lal George
129		Date: 2002/03/05 17:24:48 EST
130		Tag: george-20020305-linkage-cluster
131
132		In order to support the block placement optimization, a new cluster
133		is generated as the very first cluster (called the linkage cluster).
134		It contains a single jump to the 'real' entry point for the compilation
135		unit. Block placement has no effect on the linkage cluster itself, but
136		all the other clusters  have full freedom in the manner in which they
137		reorder blocks or functions.
138
139		On the x86 the typical linkage code that is generated is:
140		----------------------
141		.align 2
142		L0:
143		addl    $L1-L0, 72(%esp)
144		jmp     L1
145
146
147		.align  2
148		L1:
149		----------------------
150
151		72(%esp) is the memory location for the stdlink register. This
152		must contain the address of the CPS function being called. In the
153		above example, it contains the address of  L0; before
154		calling L1 (the real entry point for the compilation unit), it
155		must contain the address for L1, and hence
156
157		addl $L1-L0, 72(%esp)
158
159		I have tested this on all architectures except the hppa.The increase
160		in code size is of course negligible
161
162		----------------------------------------------------------------------
163		Name: Allen Leung
164		Date: 2002/03/03 13:20:00 EST
165		Tag: leunga-20020303-mlrisc-tools
166
167		Added #[ ... ] expressions to mlrisc tools
168
169		----------------------------------------------------------------------
170		Name: Matthias Blume
171		Date: 2002/02/27 12:29:00 EST
172		Tag: blume-20020227-cdebug
173		Description:
174
175		- made types in structure C and C_Debug to be equal
176		- got rid of code duplication (c-int.sml vs. c-int-debug.sml)
177		- there no longer is a C_Int_Debug (C_Debug is directly derived from C)
178
179		----------------------------------------------------------------------
180		Name: Matthias Blume
181		Date: 2002/02/26 12:00:00 EST
182		Tag: blume-20020226-ffi
183		Description:
184
185		1. Fixed a minor bug in CM's "noweb" tool:
186		If numbering is turned off, then truly don't number (i.e., do not
187		supply the -L option to noweb).  The previous behavior was to supply
188		-L'' -- which caused noweb to use the "default" line numbering scheme.
189		Thanks to Chris Richards for pointing this out (and supplying the fix).
190
191		2. Once again, I reworked some aspects of the FFI:
192
193		A. The incomplete/complete type business:
194
195		- Signatures POINTER_TO_INCOMPLETE_TYPE and accompanying functors are
196		gone!
197		- ML types representing an incomplete type are now *equal* to
198		ML types representing their corresponding complete types (just like
199		in C).  This is still safe because ml-nlffigen will not generate
200		RTTI for incomplete types, nor will it generate functions that
201		require access to such RTTI.   But when ML code generated from both
202		incomplete and complete versions of the C type meet, the ML types
203		are trivially interoperable.
204
205		NOTE:  These changes restore the full generality of the translation
206		(which was previously lost when I eliminated functorization)!
207
208		B. Enum types:
209
210		- Structure C now has a type constructor "enum" that is similar to
211		how the "su" constructor works.  However, "enum" is not a phantom
212		type because each "T enum" has values (and is isomorphic to
213		MLRep.Signed.int).
214		- There are generic access operations for enum objects (using
215		MLRep.Signed.int).
216		- ml-nlffigen will generate a structure E_foo for each "enum foo".
217		* The structure contains the definition of type "mlrep" (the ML-side
218		representation type of the enum).  Normally, mlrep is the same
219		as "MLRep.Signed.int", but if ml-nlffigen was invoked with "-ec",
220		then mlrep will be defined as a datatype -- thus facilitating
221		pattern matching on mlrep values.
222		("-ec" will be suppressed if there are duplicate values in an
223		enumeration.)
224		* Constructors ("-ec") or values (no "-ec") e_xxx of type mlrep
225		will be generated for each C enum constant xxx.
226		* Conversion functions m2i and i2m convert between mlrep and
227		MLRep.Signed.int.  (Without "-ec", these functions are identities.)
228		* Coversion functions c and ml convert between mlrep and "tag enum".
229		* Access functions (get/set) fetch and store mlrep values.
230		- By default (unless ml-nlffigen was invoked with "-nocollect"), unnamed
231		enumerations are merged into one single enumeration represented by
232		structure E_'.
233
234		----------------------------------------------------------------------
235		Name: Allen Leung
236		Date: 2002/02/25 04:45:00 EST
237		Tag: leunga-20020225-cps-spill
238
239		This is a new implementation of the CPS spill phase.
240		The new phase is in the new file compiler/CodeGen/cpscompile/spill-new.sml
241		In case of problems, replace it with the old file spill.sml
242
243		The current compiler runs into some serious performance problems when
244		constructing a large record.  This can happen when we try to compile a
245		structure with many items.  Even a very simple structure like the following
246		makes the compiler slow down.
247
248		structure Foo = struct
249		val x_1 = 0w1 : Word32.int
250		val x_2 = 0w2 : Word32.int
251		val x_3 = 0w3 : Word32.int
252		...
253		val x_N = 0wN : Word32.int
254		end
255
256		The following table shows the compile time, from N=1000 to N=4000,
257		with the old compiler:
258
259		N
260		1000   CPS 100 spill                           0.04u  0.00s  0.00g
261		MLRISC ra                               0.06u  0.00s  0.05g
262		(spills = 0 reloads = 0)
263		TOTAL                                   0.63u  0.07s  0.21g
264
265		1100   CPS 100 spill                           8.25u  0.32s  0.64g
266		MLRISC ra                               5.68u  0.59s  3.93g
267		(spills = 0 reloads = 0)
268		TOTAL                                   14.71u  0.99s  4.81g
269
270		1500   CPS 100 spill                           58.55u  2.34s  1.74g
271		MLRISC ra                               5.54u  0.65s  3.91g
272		(spills = 543 reloads = 1082)
273		TOTAL                                   65.40u  3.13s  6.00g
274
275		2000   CPS 100 spill                           126.69u  4.84s  3.08g
276		MLRISC ra                               0.80u  0.10s  0.55g
277		(spills = 42 reloads = 84)
278		TOTAL                                   129.42u  5.10s  4.13g
279
280		3000   CPS 100 spill                           675.59u  19.03s  11.64g
281		MLRISC ra                               2.69u  0.27s  1.38g
282		(spills = 62 reloads = 124)
283		TOTAL                                   682.48u  19.61s  13.99g
284
285		4000   CPS 100 spill                           2362.82u  56.28s  43.60g
286		MLRISC ra                               4.96u  0.27s  2.72g
287		(spills = 85 reloads = 170)
288		TOTAL                                   2375.26u  57.21s  48.00g
289
290		As you can see the old cps spill module suffers from some serious
291		performance problem.  But since I cannot decipher the old code fully,
292		instead of patching the problems up, I'm reimplementing it
293		with a different algorithm.  The new code is more modular,
294		smaller when compiled, and substantially faster
295		(O(n log n) time and O(n) space).  Timing of the new spill module:
296
297		4000  CPS 100 spill                           0.02u  0.00s  0.00g
298		MLRISC ra                               0.25u  0.02s  0.15g
299		(spills=1 reloads=3)
300		TOTAL                                   7.74u  0.34s  1.62g
301
302		Implementation details:
303
304		As far as I can tell, the purpose of the CPS spill module is to make sure the
305		number of live variables at any program point (the bandwidth)
306		does not exceed a certain limit, which is determined by the
307		size of the spill area.
308
309		When the bandwidth is too large, we decrease the register pressure by
310		packing live variables into spill records.  How we achieve this is
311		completely different than what we did in the old code.
312
313		First, there is something about the MLRiscGen code generator
314		that we should be aware of:
315
316		o MLRiscGen performs code motion!
317
318		In particular, it will move floating point computations and
319		address computations involving only the heap pointer to
320		their use sites (if there is only a single use).
321		What this means is that if we have a CPS record construction
322		statement
323
324		RECORD(k,vl,w,e)
325
326		we should never count the new record address w as live if w
327		has only one use (which is often the case).
328
329		We should do something similar to floating point, but the transformation
330		there is much more complex, so I won't deal with that.
331
332		Secondly, there are now two new cps primops at our disposal:
333
334		1. rawrecord of record_kind option
335		This pure operator allocates some uninitialized storage from the heap.
336		There are two forms:
337
338		rawrecord NONE [INT n]  allocates a tagless record of length n
339		rawrecord (SOME rk) [INT n] allocates a tagged record of length n
340		and initializes the tag.
341
342		2. rawupdate of cty
343		rawupdate cty (v,i,x)
344		Assigns to x to the ith component of record v.
345		The storelist is not updated.
346
347		We use these new primops for both spilling and increment record construction.
348
349		1. Spilling.
350
351		This is implemented with a linear scan algorithm (but generalized
352		to trees).  The algorithm will create a single spill record at the
353		beginning of the cps function and use rawupdate to spill to it,
354		and SELECT or SELp to reload from it.  So both spills and reloads
355		are fine-grain operations.  In contrast, in the old algorithm
356		"spills" have to be bundled together in records.
357
358		Ideally, we should sink the spill record construction to where
359		it is needed.  We can even split the spill record into multiple ones
360		at the places where they are needed.  But CPS is not a good
361		representation for global code motion, so I'll keep it simple and
362		am not attempting this.
363
364		2. Incremental record construction (aka record splitting).
365
366		Long records with many component values which are simulatenously live
367		(recall that single use record addresses are not considered to
368		be live) are constructed with rawrecord and rawupdate.
369		We allocate space on the heap with rawrecord first, then gradually
370		fill it in with rawupdate.  This is the technique suggested to me
371		by Matthias.
372
373		Some restrictions on when this is applicable:
374		1. It is not a VECTOR record.  The code generator currently does not handle
375		this case. VECTOR record uses double indirection like arrays.
376		2. All the record component values are defined in the same "basic block"
377		as the record constructor.  This is to prevent speculative
378		record construction.
379
380		----------------------------------------------------------------------
381		Name: Allen Leung
382		Date: 2002/02/22 01:02:00 EST
383		Tag: leunga-20020222-mlrisc-tools
384
385		Minor bug fixes in the parser and rewriter
386
387		----------------------------------------------------------------------
388		Name: Allen Leung
389		Date: 2002/02/21 20:20:00 EST
390		Tag: leunga-20020221-peephole
391
392		Regenerated the peephole files.  Some contained typos in the specification
393		and some didn't compile because of pretty printing bugs in the old version
394		of 'nowhere'.
395
396		----------------------------------------------------------------------
397		Name: Allen Leung
398		Date: 2002/02/19 20:20:00 EST
399		Tag: leunga-20020219-mlrisc-tools
400		Description:
401
402		Minor bug fixes to the mlrisc-tools library:
403
404		1.  Fixed up parsing colon suffixed keywords
405		2.  Added the ability to shut the error messages up
406		3.  Reimplemented the pretty printer and fixed up/improved
407		the pretty printing of handle and -> types.
408		4.  Fixed up generation of literal symbols in the nowhere tool.
409		5.  Added some SML keywords to to sml.sty
410
411		----------------------------------------------------------------------
412		Name: Matthias Blume
413		Date: 2002/02/19 16:20:00 EST
414		Tag: blume-20020219-cmffi
415		Description:
416
417		A wild mix of changes, some minor, some major:
418
419		* All C FFI-related libraries are now anchored under $c:
420		$/c.cm      --> $c/c.cm
421		$/c-int.cm  --> $c/internals/c-int.cm
422		$/memory.cm --> $c/memory/memory.cm
423
424		* "make" tool (in CM) now treats its argument pathname slightly
425		differently:
426		1. If the native expansion is an absolute name, then before invoking
427		the "make" command on it, CM will apply OS.Path.mkRelative
428		(with relativeTo = OS.FileSys.getDir()) to it.
429		2. The argument will be passed through to subsequent phases of CM
430		processing without "going native".  In particular, if the argument
431		was an anchored path, then "make" will not lose track of that anchor.
432
433		* Compiler backends now "know" their respective C calling conventions
434		instead of having to be told about it by ml-nlffigen.  This relieves
435		ml-nlffigen from one of its burdens.
436
437		* The X86Backend has been split into X86CCallBackend and X86StdCallBackend.
438
439		* Export C_DEBUG and C_Debug from $c/c.cm.
440
441		* C type encoding in ml-nlffi-lib has been improved to model the conceptual
442		subtyping relationship between incomplete pointers and their complete
443		counterparts.  For this, ('t, 'c) ptr has been changed to 'o ptr --
444		with the convention of instantiating 'o with ('t, 'c) obj whenever
445		the pointer target type is complete.  In the incomplete case, 'o
446		will be instantiated with some "'c iobj" -- a type obtained by
447		using one of the functors PointerToIncompleteType or PointerToCompleteType.
448
449		Operations that work on both incomplete and complete pointer types are
450		typed as taking an 'o ptr while operations that require the target to
451		be known are typed as taking some ('t, 'c) obj ptr.
452
453		voidptr is now a bit "more concrete", namely "type voidptr = void ptr'"
454		where void is an eqtype without any values.  This makes it possible
455		to work on voidptr values using functions meant to operate on light
456		incomplete pointers.
457
458		* As a result of the above, signature POINTER_TO_INCOMPLETE_TYPE has
459		been vastly simplified.
460
461		----------------------------------------------------------------------
462		Name: Matthias Blume
463		Date: 2002/02/19 10:48:00 EST
464		Tag: blume-20020219-pqfix
465		Description:
466
467		Applied Chris Okasaki's bug fix for priority queues.
468
469		----------------------------------------------------------------------
470		Name: Matthias Blume
471		Date: 2002/02/15 17:05:00
472		Tag: Release_110_39
473		Description:
474
475		Last-minute retagging is becoming a tradition... :-(
476
477		This is the working release 110.39.
478
479		----------------------------------------------------------------------
480		Name: Matthias Blume
481		Date: 2002/02/15 16:00:00 EST
482		Tag: Release_110_39-orig
483		Description:
484
485		Working release 110.39.  New bootfiles.
486
487		(Update: There was a small bug in the installer so it wouldn't work
488		with all shells.  So I retagged. -Matthias)
489
490		----------------------------------------------------------------------
491		Name: Matthias Blume
492		Date: 2002/02/15 14:17:00 EST
493		Tag: blume-20020215-showbindings
494		Description:
495
496		Added EnvRef.listBoundSymbols and CM.State.showBindings.  Especially
497		the latter can be useful for exploring what bindings are available at
498		the interactive prompt.  (The first function returns only the list
499		of symbols that are really bound, the second prints those but also the
500		ones that CM's autoloading mechanism knows about.)
501
502		----------------------------------------------------------------------
503		Name: Matthias Blume
504		Date: 2002/02/15 12:08:00 EST
505		Tag: blume-20020215-iptrs
506		Description:
507
508		Two improvements to ml-nlffigen:
509
510		1. Write files only if they do not exist or if their current contents
511		do not coincide with what's being written.  (That is, avoid messing
512		with the time stamps unless absolutely necessary.)
513
514		2. Implement a "repository" mechanism for generated files related
515		to "incomplete pointer types".   See the README file for details.
516
517		----------------------------------------------------------------------
518		Name: Matthias Blume
519		Date: 2002/02/14 11:50:00 EST
520		Tag: blume-20020214-quote
521		Description:
522
523		Added a type 't t_' to tag.sml (in ml-nlffi-lib.cm).  This is required
524		because of the new and improved tag generation scheme.  (Thanks to Allen
525		Leung for pointing it out.)
526
527		----------------------------------------------------------------------
528		Name: Lal George
529		Date: 2002/02/14 09:55:27 EST 2002
530		Tag: george-20020214-isabelle-bug
531		Description:
532
533		Fixed the MLRISC bug sent by Markus Wenzel regarding the compilation
534		of Isabelle on the x86.
535
536		From Allen:
537		-----------
538		I've found the problem:
539
540		in ra-core.sml, I use the counter "blocked" to keep track of the
541		true number of elements in the freeze queue.  When the counter goes
542		to zero, I skip examining the queue.  But I've messed up the
543		bookkeeping in combine():
544
545		else ();
546		case !ucol of
547		PSEUDO => (if !cntv > 0 then
548		(if !cntu > 0 then blocked := !blocked - 1 else ();
549		^^^^^^^^^^^^^^^^^^^^^^^
550		moveu := mergeMoveList(!movev, !moveu)
551		)
552		else ();
553
554		combine() is called to coalesce two nodes u and v.
555		I think I was thinking that if the move counts of u and v are both
556		greater than zero then after they are coalesced then one node is
557		removed from the freeze queue.  Apparently I was thinking that
558		both u and v are of low degree, but that's clearly not necessarily true.
559
560
561		02/12/2002:
562		Here's the patch.  HOL now compiles.
563
564		I don't know how this impact on performance (compile
565		time or runtime).  This bug caused the RA (especially on the x86)
566		to go thru the potential spill phase when there are still nodes on the
567		freeze queue.
568
569
570
571
572		----------------------------------------------------------------------
573		Name: Matthias Blume
574		Date: 2002/02/13 22:40:00 EST
575		Tag: blume-20020213-fptr-rtti
576		Description:
577
578		Fixed a bug in ml-nlffigen that was introduced with one of the previous
579		updates.
580
581		----------------------------------------------------------------------
582		Name: Matthias Blume
583		Date: 2002/02/13 16:41:00 EST
584		Tag: blume-20020213-cmlpq
585		Description:
586
587		Added new priority queue export symbols (which have just been added to
588		smlnj-lib.cm) to CML's version of smlnj-lib.cm.  (Otherwise CML would
589		not compile and the installer would choke.)
590
591		----------------------------------------------------------------------
592		Name: Matthias Blume
593		Date: 2002/02/13 16:15:00 EST
594		Tag: blume-20020213-various
595		Description:
596
597		1. More tweaks to ml-nlffigen:
598
599		- better internal datastructures (resulting in slight speedup)
600		- "-match" option requires exact match
601		- "localized" gensym counters (untagged structs/unions nested within
602		other structs/unions or within typedefs get a fresh counter; their
603		tag will be prefixed by a concatenation of their parents' tags)
604		- bug fixes (related to calculation of transitive closure of types
605		to be included in the output)
606
607		2. Minor Basis updates:
608
609		- added implementations for List.collate and Option.app
610
611		----------------------------------------------------------------------
612		Name: Matthias Blume
613		Date: 2002/02/11 15:55:00 EST
614		Tag: blume-20020211-gensym
615		Description:
616
617		Added a "-gensym" option to command line of ml-nlffigen.  This can be
618		used to specify a "stem" -- a string that is inserted in all "gensym'd"
619		names (ML structure names that correspond to unnamed C structs, unions,
620		and enums), so that separate runs of ml-nlffigen do not clash.
621
622		----------------------------------------------------------------------
623		Name: Matthias Blume
624		Date: 2002/02/11 12:05:00 EST
625		Tag: blume-20020211-gensml
626		Description:
627
628		A quick fix for a problem with GenSML (in the pgraph-util library):
629		Make generation of toplevel "local" optional.  (Strictly speaking,
630		signature definitions within "local" are not legal SML.)
631
632		Other than that: updates to INSTALL and cm/TODO.
633
634		----------------------------------------------------------------------
635		Name: Matthias Blume
636		Date: 2002/02/08 15:00:00 EST
637		Tag: blume-20020208-uniquepid
638		Description:
639
640		0. Version number has been bumped to 110.38.1.  NEW BOOTFILES!!!
641
642		1. The installer (config/install.sh) has gotten smarter:
643
644		- Configuration options are a bit easier to specify now
645		(in config/targets).
646		- Bug in recognizing .tar.bz2 files fixed.
647		- Installer automatically resolves dependencies between
648		configuration options (e.g., if you ask for eXene, you will
649		also get cml -- regardless whether you asked for it or not).
650		- Installer can run in "quieter mode" by setting the environment
651		variable INSTALL_QUIETLY to "true".  "Quieter" does not mean
652		"completely silent", though.
653		- Build HashCons library as part of smlnj-lib.
654
655		2. A new scheme for assigning persistent identifiers to compilation
656		units (and, by extension, to types etc.) has been put into place.
657		This fixes a long-standing bug where types and even dynamic values
658		can get internally confused, thereby compromising type safety
659		(abstraction) and dynamic correctness.  See
660
661		http://cm.bell-labs.com/cm/cs/who/blume/pid-confusion.tgz
662
663		for an example of how things could go wrong until now.
664
665		The downside of the new scheme is that pids are not quite as
666		persistent as they used to be: CM will generate a fresh pid
667		for every compilation unit that it thinks it sees for the first
668		time.  That means that if you compile starting from a clean, fresh
669		source tree at two different times, you end up with different
670		binaries.
671
672		Cutoff recompilation, however, has not been compromised because
673		CM keeps pid information in special caches between runs.
674
675		----------------------------------------------------------------------
676		Name: Lal George
677		Date: 2002/02/07 15:34:13 EST 2002
678		Tag: <none>
679		Description:
680
681		Compilers that generate assembly code may produce  global labels
682		whose value is resolved at link time. The various peephole optimization
683		modules did not take this in account.
684
685		TODO. The Labels.addrOf function should really return an option
686		type so that clients are forced to deal with this issue, rather
687		than an exception being raised.
688
689		----------------------------------------------------------------------
690		Name: Lal George
691		Date: 2002/02/06 13:55:02 EST
692		Tag: george-20020206-ra-breakup
693		Description:
694
695		1. A bug fix from Allen.
696
697		A typo causes extra fstp %st(0)'s to be generated at compensation
698		edges, which might cause stack underflow traps at runtime.  This
699		occurs in fft where there are extraneous fstps right before the 'into'
700		trap instruction (in this case they are harmless since none of the
701		integers overflow.)
702
703		2. Pulled out various utility modules that were embedded in the modules
704		of the register allocator. I need these modules for other purposes, but
705		they are not complete enough to put into a library (just yet).
706		----------------------------------------------------------------------
707		Name: Matthias Blume
708		Date: 2002/01/31 16:05:00 EST
709		Tag: blume-20020131-sparc-ccalls
710		Description:
711
712		1. C-calls on Sparc needlessly allocated a huge chunk (96 bytes)
713		of extra stack space by mistake.  Fixed.
714
715		2. Bug in logic of handling of command-line options in ml-nlffigen fixed.
716
717		----------------------------------------------------------------------
718		Name: Allen Leung
719		Date: 2002/01/30
720		Tag: leunga-20020130-nowhere-bug-fix
721		Description:
722
723		MLRISC bug fixes:
724		1. Fixed a bindings computation bug in the 'nowhere' program generator tool.
725		2. MachineInt.fromString was negating its value.
726
727		----------------------------------------------------------------------
728		Name: Matthias Blume
729		Date: 2002/01/29
730		Tag: blume-20020129-INSTALL
731		Description:
732
733		- Added somewhat detailed installation instructions (file INSTALL).
734		- Fixed curl-detection bug in config/install.sh.
735		- It is now possible to select the URL getter using the URLGETTER
736		environment variable:
737
738		not set / "unknown"      --> automatic detection (script tries wget,
739		curl, and lynx)
740		"wget" / "curl" / "lynx" --> use the specified program (script "knows"
741		how to properly invoke them)
742		other                    --> use $URLGETTER directly, it must take
743		precisely two command-line arguments
744		(source URL and destination file name)
745
746		----------------------------------------------------------------------
747		Name: Matthias Blume
748		Date: 2002/01/28
749		Tag: blume-20020128-sparc-ccalls
750		Description:
751
752		- Fixed problem with calculation of "used" registers in sparc-c-calls.
753		- Make use of the allocParam argument in sparc-c-calls.
754
755		----------------------------------------------------------------------
756		Name: Matthias Blume
757		Date: 2002/01/28
758		Tag: blume-20020128-allocParam
759		Description:
760
761		John Reppy:  Changes c-calls API to accept client-callback for
762		allocating extra stack space.
763		me: Corresponding changes to mlriscGen (using a dummy argument that
764		does not change the current behavior).
765
766		----------------------------------------------------------------------
767		Name: Matthias Blume
768		Date: 2002/01/28 12:00:00
769		Tag: Release_110_38
770		Description:
771
772		This time for real!!!
773
774		----------------------------------------------------------------------
775		Name: Matthias Blume
776		Date: 2002/01/28 10:56:00 EST
777		Tag: blume-20020128-retraction
778		Description:
779
780		0. Retracted earlier 110.38.  (The Release_110_38 tag has been replaced
781		with blume-Release_110_38-retracted.)
782
783		1. Fixed a problem with incorrect rounding modes in real64.sml.
784		(Thanks to Andrew Mccreight <andrew.mccreight@yale.edu>.)
785
786		2. A bug in ml-nlffigen related to the handling of unnamed structs, unions,
787		and enums fixed.  The naming of corresponding ML identifiers should
788		now be consistent again.
789
790		----------------------------------------------------------------------
791		Name: Allen Leung
792		Date: 2002/01/27
793		Tag: leunga-20020127-nowhere
794		Description:
795
796		Added a target called nowhere in the configuration scripts.
797		Enabling this will build the MLRISC 'nowhere' tool (for translating
798		programs with where-clauses into legal SML code) during installation.
799
800		----------------------------------------------------------------------
801		Name: Matthias Blume
802		Date: 2002/01/25 21:27:00 EST
803		Tag: blume-Release_110_38-retracted
804		Description:
805
806		Call it a (working) release!  Version is 110.38. Bootfiles are ready.
807
808		README will be added later.
809
810		!!! NOTE:  Re-tagged as blume-Release_110_38-retracted. Original tag
811		(Release_110_38) removed.  Reason: Last-minute bug fixes.
812
813		----------------------------------------------------------------------
814		Name: Matthias Blume
815		Date: 2002/01/25
816		Tag: blume-20020125-ffi
817		Description:
818
819		A large number of tweaks and improvements to ml-nlffi-lib and
820		ml-nlffigen:
821
822		- ML represenation types have been streamlined
823		- getter and setter functions work with concrete values, not abstract
824		ones where possible
825		- ml-nlffigen command line more flexible (see README file there)
826		- some bugs have been fixed (hopefully)
827
828		----------------------------------------------------------------------
829		Name: Lal George
830		Date: 2002/01/24
831		Tag: george-20020124-risc-ra-interface
832		Description:
833
834		There is a dramatic simplification in the interface to the
835		register allocator for RISC architectures as a result of making
836		parallel copy instructions explicit.
837
838		----------------------------------------------------------------------
839		Name: Matthias Blume
840		Date: 2002/01/22
841		Tag: blume-20020122-x86-ccalls
842		Description:
843
844		Bug fix for c-calls on x86 (having to do with how char- and
845		short-arguments are being handled).
846
847		----------------------------------------------------------------------
848		Name: Matthias Blume
849		Date: 2002/01/21
850		Tag: blume-20020121-ff
851		Description:
852
853		Another day of fiddling with the FFI...
854
855		1. Bug fix/workaround:  CKIT does not complain about negative array
856		dimensions, so ml-nlffigen has to guard itself against this possibility.
857		(Otherwise a negative dimension would send it into an infinite loop.)
858
859		2. Some of the abstract types (light objects, light pointers, most "base"
860		types) in structure C are now eqtypes.
861
862		3. Added constructors and test functions for NULL function pointers.
863
864		----------------------------------------------------------------------
865		Name: Matthias Blume
866		Date: 2002/01/18
867		Tag: blume-20020118-ready-for-new-release
868		Description:
869
870		Made config/srcarchiveurl point to a new place.  (Will provide boot
871		files shortly.)
872
873		Maybe we christen this to be 110.38?
874
875		----------------------------------------------------------------------
876		Name: Matthias Blume
877		Date: 2002/01/18
878		Tag: blume-20020118-more-ffifiddle
879		Description:
880
881		Today's FFI fiddling:
882
883		- Provided a structure CGetSet with "convenient" versions of C.Get.* and
884		C.Set.* that use concrete (MLRep.*) arguments and results instead
885		of abstract ones.
886
887		- Provided word-style bit operations etc. for "int" representation
888		types in MLRep.S<Foo>Bitops where <Foo> ranges over Char, Int, Short,
889		and Long.
890
891		----------------------------------------------------------------------
892		Name: Matthias Blume
893		Date: 2002/01/18
894		Tag: blume-20020118-use-x86-fp
895		Description:
896
897		Now that x86-fast-fp seems to be working, I turned it back on again
898		by default.  (Seems to work fine now, even with the FFI.)
899
900		Other than that, I added some documentation about the FFI to
901		src/ml-nlffigen/README and updated the FFI test examples in
902		src/ml-nlffi-lib/Tests/*.
903
904		----------------------------------------------------------------------
905		Name: Allen Leung
906		Date: 2002/01/17
907		Tag: leunga-20020117-x86-fast-fp-call
908		Description:
909
910		1. Fixed a problem with handling return fp values when x86's fast fp
911		mode is turned on.
912
913		2. Minor pretty printing fix for cellset.  Print %st(0) as %st(0) instead
914		of %f32.
915
916		3. Added a constructor INT32lit to the ast of MLRISC tools.
917
918		----------------------------------------------------------------------
919		Name: Matthias Blume
920		Date: 2002/01/16
921		Tag: blume-20020116-ffifiddle
922		Description:
923
924		More fiddling with the FFI interface:
925
926		- Make constness 'c instead of rw wherever possible.  This eliminates
927		the need for certain explicit coercions.  (However, due to ML's
928		value polymorphism, there will still be many cases where explicit
929		coercions are necessary.  Phantom types are not the whole answer
930		to modeling a subtyping relationship in ML.)
931
932		- ro/rw coersions for pointers added.  (Avoids the detour through */&.)
933
934		- "printf" test example added to src/ml-nlffi-lib/Tests.  (Demonstrates
935		clumsy workaround for varargs problem.)
936
937		----------------------------------------------------------------------
938		Name: Lal George
939		Date: 2002/01/15
940		Tag: <none>
941		Description:
942
943		1. Since COPY instructions are no longer native to the architecture,
944		a generic functor can be used to implement the expandCopies function.
945
946		2. Allowed EXPORT and IMPORT pseudo-op declarations to appear inside a
947		TEXT segment.
948
949		----------------------------------------------------------------------
950		Name: Matthias Blume
951		Date: 2002/01/15
952		Tag: blume-20020115-ffiupdates
953		Description:
954
955		1. Fix for bug resulting in single-precision float values being returned
956		incorrectly from FFI calls.
957
958		2. Small modifications to C FFI API:
959
960		- memory-allocation routines return straight objects (no options)
961		and raise an exception in out-of-memory situations
962		- unsafe extensions to cast between function pointers and pointers
963		from/to ints
964		- added structure C_Debug as an alternative to structure C where
965		pointer-dereferencing (|*| and |*!) always check for null-pointers
966		- added open_lib' to DynLinkage;  open_lib' works like open_lib
967		but also takes a (possibly empty) list of existing library handles
968		that the current library depends on
969
970		----------------------------------------------------------------------
971		Name: Matthias Blume
972		Date: 2002/01/10
973		Tag: blume-20020110-newffigen
974		Description:
975
976		1. Updates to portable graph code.
977
978		2. Major update to ml-nlffigen and ml-nlffi-lib.  Things are much
979		more scalable now so that even huge interfaces such as the one
980		for GTK compile in finite time and space. :-)
981		See src/ml-nlffigen/README for details on what's new.
982
983		----------------------------------------------------------------------
984		Name: Lal George
985		Date: 2001/01/09 14:31:35 EST 2002
986		Tag: george-20011206-rm-native-copy
987		Description:
988
989		Removed the native COPY and FCOPY instructions
990		from all the architectures and replaced it with the
991		explicit COPY instruction from the previous commit.
992
993		It is now possible to simplify many of the optimizations
994		modules that manipulate copies. This has not been
995		done in this change.
996
997		----------------------------------------------------------------------
998		Name: Lal George
999		Date: 2001/12/06 16:50:13 EST 2001
1000		Tag: george-20011206-mlrisc-instruction
1001		Description:
1002
1003		Changed the representation of instructions from being fully abstract
1004		to being partially concrete. That is to say:
1005
1006		from
1007		type instruction
1008
1009		to
1010		type instr                              (* machine instruction *)
1011
1012		datatype instruction =
1013		LIVE of {regs: C.cellset, spilled: C.cellset}
1014		| KILL of {regs: C.cellset, spilled: C.cellset}
1015		| COPYXXX of {k: CB.cellkind, dst: CB.cell list, src: CB.cell list}
1016		| ANNOTATION of {i: instruction, a: Annotations.annotation}
1017		| INSTR of instr
1018
1019		This makes the handling of certain special instructions that appear on
1020		all architectures easier and uniform.
1021
1022		LIVE and KILL say that a list of registers are live or killed at the
1023		program point where they appear. No spill code is generated when an
1024		element of the 'regs' field is spilled, but the register is moved to
1025		the 'spilled' (which is present, more for debugging than anything else).
1026
1027		LIVE replaces the (now deprecated) DEFFREG instruction on the alpha.
1028		We used to generate:
1029
1030		DEFFREG f1
1031		f1 := f2 + f3
1032		trapb
1033
1034		but now generate:
1035
1036		f1 := f2 + f3
1037		trapb
1038		LIVE {regs=[f1,f2,f3], spilled=[]}
1039
1040		Furthermore, the DEFFREG (hack) required that all floating point instruction
1041		use all registers mentioned in the instruction. Therefore f1 := f2 + f3,
1042		defines f1 and uses [f1,f2,f3]! This hack is no longer required resulting
1043		in a cleaner alpha implementation. (Hopefully, intel will not get rid of
1044		this architecture).
1045
1046		COPYXXX is intended to replace the parallel COPY and FCOPY  available on
1047		all the architectures. This will result in further simplification of the
1048		register allocator that must be aware of them for coalescing purposes, and
1049		will also simplify certain aspects of the machine description that provides
1050		callbacks related to parallel copies.
1051
1052		ANNOTATION should be obvious, and now INSTR represents the honest to God
1053		machine instruction set!
1054
1055		The <arch>/instructions/<arch>Instr.sml files define certain utility
1056		functions for making porting easier -- essentially converting upper case
1057		to lower case. All machine instructions (of type instr) are in upper case,
1058		and the lower case form generates an MLRISC instruction. For example on
1059		the alpha we have:
1060
1061		datatype instr =
1062		LDA of {r:cell, b:cell, d:operand}
1063		| ...
1064
1065		val lda : {r:cell, b:cell, d:operand} -> instruction
1066		...
1067
1068		where lda is just (INSTR o LDA), etc.
1069
1070		----------------------------------------------------------------------
1071		Name: Matthias Blume
1072		Date: 2001/11/22 21:40:00 EST
1073		Tag: Release_110_37
1074		Description:
1075
1076		Release 110.37.  This time for real.
1077
1078		----------------------------------------------------------------------
1079		Name: Matthias Blume
1080		Date: 2001/11/21 16:35:00 EST
1081		Tag: blume-20011121-foot-in-mouth
1082		Description:
1083
1084		Removed the "Release_110_37" tag because of a serious bug.
1085		This will be re-tagged once the bug is fixed.
1086
1087		----------------------------------------------------------------------
1088		Name: Matthias Blume
1089		Date: 2001/11/21 16:14:00 EST
1090		Tag: blume-20011121-forgottenfile
1091		Description:
1092
1093		Forgot to add a file.  (Just a .tex-file -- part of
1094		the CM manual source.)
1095
1096		----------------------------------------------------------------------
1097		Name: Matthias Blume
1098		Date: 2001/11/21 16:10:00 EST
1099		Tag: blume-20011121-invalid_110_37
1100		Description:
1101
1102		Note: I removed the original tag "Release_110_37" from this commit
1103		because we found a serious bug in all non-x86 backends.
1104		- Matthias
1105
1106		1. Modifications to the SML/NJ code generator and to the runtime system
1107		so that code object name strings are directly inserted into code
1108		objects at code generation time.  The only business the runtime system
1109		has with this is now to read the name strings on occasions.
1110		(The encoding of the name string has also changed somewhat.)
1111
1112		2. CM now implements a simple "set calculus" for specifying export lists.
1113		In particular, it is now possible to refer to the export lists of
1114		other libraries/groups/sources and form unions as well as differences.
1115		See the latest CM manual for details.
1116
1117		3. An separate notion of "proxy" libraries has again be eliminated from
1118		CM's model.  (Proxy libraries are now simply a special case of using
1119		the export list calculus.)
1120
1121		4. Some of the existing libraries now take advantage of the new set
1122		calculus.
1123		(Notice that not all libraries have been converted because some
1124		of the existing .cm-files are supposed to be backward compatible
1125		with 110.0.x.)
1126
1127		5. Some cleanup in stand-alone programs.  (Don't use "exnMessage" -- use
1128		"General.exnMessage"!  The former relies on a certain hook to be
1129		initialized, and that often does not happen in the stand-alone case.)
1130
1131		----------------------------------------------------------------------
1132		Name: Lal George
1133		Date: 2001/11/21  13:56:18 EST
1134		Tag: george-2001121-pseudo-ops
1135		Description:
1136
1137		Implemented a complete redesign of MLRISC pseudo-ops. Now there
1138		ought to never be any question of incompatabilities with
1139		pseudo-op syntax expected by host assemblers.
1140
1141		For now, only modules supporting GAS syntax are implemented
1142		but more should follow, such as MASM, and vendor assembler
1143		syntax, e.g. IBM as, Sun as, etc.
1144
1145		----------------------------------------------------------------------
1146		Name: Matthias Blume
1147		Date: 2001/11/14 11:52:00 EST
1148		Tag: blume-20011114-srcname
1149		Description:
1150
1151		1. Routed the name of the current source file to mlriscgen where it
1152		should be directly emitted into the code object.  (This last part
1153		is yet to be done.)
1154
1155		2. Some cleanup of the pgraph code to make it match the proposal that
1156		I put out the other day.  (The proposal notwithstanding, things are
1157		still in flux here.)
1158
1159		----------------------------------------------------------------------
1160		Name: Lal George
1161		Date: 2001/11/14 09:44:04 EST
1162		Tag:
1163		Description:
1164
1165		Fix for a backpatching bug reported by Allen.
1166
1167		Because the boundary between short and long span-dependent
1168		instructions is +/- 128, there are an astounding number of
1169		span-dependent instructions whose size is over estimated.
1170
1171		Allen came up with the idea of letting the size of span
1172		dependent instructions be non-monotonic, for a maxIter
1173		number of times, after which the size must be monotonically
1174		increasing.
1175
1176		This table shows the number of span-dependent instructions
1177		whose size was over-estimated as a function of maxIter, for the
1178		file Parse/parse/ml.grm.sml:
1179
1180		maxIter            # of instructions:
1181		10                      687
1182		20                      438
1183		30                      198
1184		40                        0
1185
1186		In compiling the compiler, there is no significant difference in
1187		compilation speed between maxIter=10 and maxIter=40. Actually,
1188		my measurements showed that maxIter=40 was a tad faster than
1189		maxIter=10! Also 96% of the  files in the compiler reach a fix
1190		point within 13 iterations, so fixing maxIter at 40, while high,
1191		is okay.
1192
1193		----------------------------------------------------------------------
1194		Name: Matthias Blume
1195		Date: 2001/10/31 15:25:00 EST
1196		Tag: blume-20011031-pgraph
1197		Description:
1198
1199		CKIT:
1200		* Changed the "Function" constructor of type Ast.ctype to carry optional
1201		argument identifiers.
1202		* Changed the return type of TypeUtil.getFunction accordingly.
1203		* Type equality ignores the argument names.
1204		* TypeUtil.composite tries to preserve argument names but gives up quickly
1205		if there is a mismatch.
1206
1207		installation script:
1208		* attempts to use "curl" if available (unless "wget" is available as well)
1209
1210		CM:
1211		* has an experimental implementation of "portable graphs" which I will
1212		soon propose as an implementation-independent library format
1213		* there are also new libraries $/pgraph.cm and $/pgraph-util.cm
1214
1215		NLFFI-LIB:
1216		* some cleanup (all cosmetic)
1217
1218		NLFFIGEN:
1219		* temporarily disabled the mechanism that suppresses ML output for
1220		C definitions whose identifiers start with an underscore character
1221		* generate val bindings for enum constants
1222		* user can request that only one style (light or heavy) is being used;
1223		default is to use both (command-line arguments: -heavy and -light)
1224		* fixed bug in handling of function types involving incomplete pointers
1225		* generate ML entry points that take record arguments (i.e., using
1226		named arguments) for C functions that have a prototype with named
1227		arguments
1228		(see changes to CKIT)
1229
1230		----------------------------------------------------------------------
1231		Name: Allen Leung
1232		Date: 2001/10/27 20:34:00 EDT
1233		Tag: leunga-20011027-x86-fast-fp-call
1234		Description:
1235
1236		Fixed the bug described in blume-20010920-slowfp.
1237
1238		The fix involves
1239		1. generating FCOPYs in FSTP in ia32-svid
1240		2. marking a CALL with the appropriate annotation
1241
1242		----------------------------------------------------------------------
1243		Name: Matthias Blume
1244		Date: 2001/10/16 11:32:00 EDT
1245		Tag: blume-20011016-netbsd
1246		Description:
1247
1248		Underscore patch from Chris Richards (fixing problem with compiling
1249		runtime system under recent NetBSD).
1250
1251		----------------------------------------------------------------------
1252		Name: Allen Leung
1253		Date: 2001/10/12 17:18:32 EDT 2001
1254		Tag: leung-20011012-x86-printflowgraph
1255		Description:
1256
1257		X86RA now uses a valid (instead of dummy) PrintFlowgraph module.
1258
1259		----------------------------------------------------------------------
1260		Name: Lal George
1261		Date: 2001/10/11 23:51:34 EDT
1262		Tag: george-20011011-too-many-instrs
1263		Description:
1264
1265		The representation of a program point never expected to see more
1266		than 65536 instructions in a basic block!
1267
1268		----------------------------------------------------------------------
1269		Name: Lal George
1270		Date: 2001/10/09 09:41:37 EDT
1271		Tag: george-20011008-mlrisc-labels
1272		Description:
1273
1274		Changed the machine description files to support printing of
1275		local and global labels in assembly code, based on host assembler
1276		conventions.
1277
1278		----------------------------------------------------------------------
1279		Name: Matthias Blume
1280		Date: 2001/09/25 15:25:00 EDT
1281		Tag: blume-20010925-exninfo
1282		Description:
1283
1284		I provided a non-hook implementation of exnName (at the toplevel) and
1285		made the "dummy" implementation of exnMessage (at the toplevel) more
1286		useful: if nothing gets "hooked in", then at least you are going to
1287		see the exception name and a message indicating why you don't see more.
1288
1289		[For the time being, programs that need exnMessage and want to use
1290		ml-build should either use General.exnMessage (strongly recommended) or
1291		refer to structure General at some other point so that CM sees a
1292		static dependency.]
1293
1294		[Similar remarks go for "print" and "use":  If you want to use their
1295		functionality in stand-alone programs generated by ml-build, then use
1296		TextIO.output and Backend.Interact.useFile (from $smlnj/compiler.cm).]
1297
1298		----------------------------------------------------------------------
1299		Name: Matthias Blume
1300		Date: 2001/09/20 17:28:00 EDT
1301		Tag: blume-20010920-slowfp
1302		Description:
1303
1304		Allen says that x86-fast-fp is not safe yet, so I turned it off again...
1305
1306		----------------------------------------------------------------------
1307		Name: Matthias Blume
1308		Date: 2001/09/20 17:20:00 EDT
1309		Tag: blume-20010920-canonicalpaths
1310		Description:
1311
1312		0. Updated the BOOT file (something that I forgot to do earlier).
1313
1314		1. Small internal change to CM so that it avoids "/../" in filenames
1315		as much as possible (but only where it is safe).
1316
1317		2. Changed config/_run-sml (resulting in a changed bin/.run-sml) so
1318		that arguments that contain delimiters are passed through correctly.
1319		This change also means that all "special" arguments of the form
1320		@SMLxxx... must come first.
1321
1322		3. Changed install script to put relative anchor names for tool commands
1323		into pathconfig.
1324
1325		----------------------------------------------------------------------
1326		Name: Matthias Blume
1327		Date: 2001/09/18 15:35:00 EDT
1328		Tag: blume-20010918-readme11036
1329		Description:
1330
1331		Added README files.
1332
1333		----------------------------------------------------------------------
1334		Name: Matthias Blume
1335		Date: 2001/09/18 11:45:00 EDT
1336		Tag: Release_110_36 (retag)
1337		Description:
1338
1339		Fixed mistake in config/preloads. Retagged as 110.36.
1340
1341		----------------------------------------------------------------------
1342		Name: Matthias Blume
1343		Date: 2001/09/18 09:40:00 EDT
1344		Tag: Release_110_36_orig (tag changed)
1345		Description:
1346
1347		New version (110.36).  New bootfiles.
1348
1349		----------------------------------------------------------------------
1350		Name: Matthias Blume
1351		Date: 2001/09/14 16:15:00 EDT
1352		Tag: blume-20010914-x86fastfp
1353		Description:
1354
1355		John committed some changes that Allen made, in particular a (hopefully)
1356		correctly working version of the x86-fp module.
1357
1358		I changed the default setting of the Control.MLRISC.getFlag "x86-fast-fp"
1359		flag to "true".  Everything seems to compile to a fixpoint ok, and
1360		"mandelbrot" speeds up by about 15%.
1361
1362		----------------------------------------------------------------------
1363		Name: Matthias Blume
1364		Date: 2001/09/13 11:20:00 EDT
1365		Tag: blume-20010913-minimal
1366		Description:
1367
1368		1. Stefan Monnier's patch to fix a miscompilation problem that
1369		was brought to light by John Reppy's work on Moby.
1370
1371		2. Implemented a minimal "structure Compiler" that contains just
1372		"version" and "architecture".  The minimal version will be
1373		available when the full version is not.  This is for backward-
1374		compatibility with code that wants to test Compiler.version.
1375
1376		----------------------------------------------------------------------
1377		Name: Matthias Blume
1378		Date: 2001/08/28 14:03:00 EDT
1379		Tag: blume-20010828-ml-lex
1380		Description:
1381
1382		Fix for bug 1581, received from Neophytos Michael.
1383
1384		----------------------------------------------------------------------
1385		Name: Matthias Blume
1386		Date: 2001/08/27 11:20:00 EDT
1387		Tag: blume-20010827-readme11035
1388		Description:
1389
1390		Fleshed out the README file for 110.35.
1391
1392		----------------------------------------------------------------------
1393		Name: Matthias Blume
1394		Date: 2001/08/24 17:10:00 EDT
1395		Tag: Release_110_35
1396		Description:
1397
1398		New version number (110.35).  New bootfiles.
1399
1400		----------------------------------------------------------------------
1401		Name: Lal George
1402		Date: 2001/08/24 13:47:18 EDT 2001
1403		Tag: george-20010824-MLRISC-graphs
1404		Description:
1405
1406		removed clusters from MLRISC completely and replaced with graphs.
1407
1408		----------------------------------------------------------------------
1409		Name: Matthias Blume
1410		Date: 2001/08/23 17:50:00 EDT
1411		Tag: blume-20010823-toplevel
1412		Description:
1413
1414		- some reorganization of the code that implements various kinds of
1415		environments in the compiler (static, dynamic, symbolic, combined)
1416		- re-implemented the EnvRef module so that evalStream works properly
1417		(if the stream contains references to "use", "CM.make", etc.)
1418		- cleaned up evalloop.sml and interact.sml (but they need more cleaning)
1419
1420		----------------------------------------------------------------------
1421		Name: Matthias Blume
1422		Date: 2001/08/20 15:50 EDT
1423		Tag: blume20010820-slipup
1424		Description:
1425
1426		I forgot to commit a few files.  Here they are...
1427
1428		----------------------------------------------------------------------
1429		Name: Matthias Blume
1430		Date: 2001/08/20 15:35:00 EDT
1431		Tag: blume-20010820-debugprof
1432		Description:
1433
1434		!!!! NEW BOOTFILES !!!!
1435
1436		This is another round of reorganizing the compiler sources.  This
1437		time the main goal was to factor out all the "instrumentation"
1438		passes (for profiling and backtracing) into their own library.
1439		The difficulty was to do it in such a way that it does not depend
1440		on elaborate.cm but only on elabdata.cm.
1441
1442		Therefore there have been further changes to both elaborate.cm and
1443		elabdata.cm -- more "generic" things have been moved from the former
1444		to the latter.  As a result, I was forced to split the assignment
1445		of numbers indicating "primtyc"s into two portions: SML-generic and
1446		SML/NJ-specific.  Since it would have been awkward to maintain,
1447		I bit the bullet and actually _changed_ the mapping between these
1448		numbers and primtycs.  The bottom line of this is that you need
1449		a new set of bin- and bootfiles.
1450
1451		I have built new bootfiles for all architectures, so doing a fresh
1452		checkout and config/install.sh should be all you need.
1453
1454		The newly created library's name is
1455
1456		$smlnj/viscomp/debugprof.cm
1457
1458		and its sources live under
1459
1460		src/compiler/DebugProf
1461
1462		----------------------------------------------------------------------
1463		Name: Matthias Blume
1464		Date: 2001/08/15 17:15:00 EDT
1465		Tag: blume-20010815-compreorg
1466		Description:
1467
1468		This is a first cut at reorganizing the CM libraries that make up the
1469		core of the compiler.  The idea is to separate out pieces that could
1470		be used independently by tools, e.g., the parser, the typechecker, etc.
1471
1472		The current status is a step in this direction, but it is not quite
1473		satisfactory yet.  Expect more changes in the future.
1474
1475		Here is the current (new) organization...
1476
1477		What used to be $smlnj/viscomp/core.cm is now divided into
1478		six CM libraries:
1479
1480		$smlnj/viscomp/basics.cm
1481		/parser.cm
1482		/elabdata.cm
1483		/elaborate.cm
1484		/execute.cm
1485		/core.cm
1486
1487		The CM files for these libraries live under src/system/smlnj/viscomp.
1488		All these libraries are proxy libraries that contain precisely
1489		one CM library component.  Here are the locations of the components
1490		(all within the src/compiler tree):
1491
1492		Basics/basics.cm
1493		Parse/parser.cm
1494		ElabData/elabdata.cm
1495		Elaborator/elaborate.cm
1496		Execution/execute.cm
1497		core.cm
1498
1499		[This organization is the same that has been used already
1500		for a while for the architecture-specific parts of the visible
1501		compiler and for the old version of core.cm.]
1502
1503		As you will notice, many source files have been moved from their
1504		respective original locations to a new home in one of the above
1505		subtrees.
1506
1507		The division of labor between the new libraries is the following:
1508
1509		basics.cm:
1510		- Simple, basic definitions that pertain to many (or all) of
1511		the other libraries.
1512		parser.cm:
1513		- The SML parser, producing output of type Ast.dec.
1514		- The type family for Ast is also defined and exported here.
1515		elabdata.cm:
1516		- The datatypes that describe input and output of the elaborator.
1517		This includes types, absyn, and static environments.
1518		elaborator.cm:
1519		- The SML/NJ type checker and elaborator.
1520		This maps an Ast.dec (with a given static environment) to
1521		an Absyn.dec (with a new static environment).
1522		- This libraries implements certain modules that used to be
1523		structures as functors (to remove dependencies on FLINT).
1524		execute.cm:
1525		- Everything having to do with executing binary code objects.
1526		- Dynamic environments.
1527		core.cm:
1528		- SML/NJ-specific instantiations of the elaborator and MLRISC.
1529		- Top-level modules.
1530		- FLINT (this should eventually become its own library)
1531
1532		Notes:
1533
1534		I am not 100% happy with the way I separated the elaborator (and its
1535		data structures) from FLINT.  Two instances of the same problem:
1536
1537		1. Data structures contain certain fields that carry FLINT-specific
1538		information.  I hacked around this using exn and the property list
1539		module from smlnj-lib.  But the fact that there are middle-end
1540		specific fields around at all is a bit annoying.
1541
1542		2. The elaborator calculates certain FLINT-related information.  I tried
1543		to make this as abstract as I could using functorization, but, again,
1544		the fact that the elaborator has to perform calculations on behalf
1545		of the middle-end at all is not nice.
1546
1547		3. Having to used exn and property lists is unfortunate because it
1548		weakens type checking.  The other alternative (parameterizing
1549		nearly *everything*) is not appealing, though.
1550
1551		I removed the "rebinding =" warning hack because due to the new organization
1552		it was awkward to maintain it.  As a result, the compiler now issues some of
1553		these warnings when compiling init.cmi during bootstrap compilation. On
1554		the plus side, you also get a warning when you do, for example:
1555		val op = = Int32.+
1556		which was not the case up to now.
1557
1558		I placed "assign" and "deref" into the _Core structure so that the
1559		code that deals with the "lazy" keyword can find them there.  This
1560		removes the need for having access to the primitive environment
1561		during elaboration.
1562
1563		----------------------------------------------------------------------
1564		Name: Matthias Blume
1565		Date: 2001/08/13
1566		Tag: blume-20010813-closures
1567		Description:
1568
1569		This fix was sent to us by Zhong Shao.  It is supposed to improve the
1570		performance of certain loops by avoiding needless closure allocation.
1571
1572		----------------------------------------------------------------------
1573		Name: Lal George
1574		Date: 2001/07/31 10:03:23 EDT 2001
1575		Tag: george-20010731-x86-fmalloc
1576		Description: Fixed bug in x86 calls
1577
1578		There was a bug where call instructions would mysteriously
1579		vanish. The call instruction had to be one that returned
1580		a floating point value.
1581
1582		----------------------------------------------------------------------
1583		Name: Lal George
1584		Date: 2001/07/19 16:36:29 EDT 2001
1585		Tag: george-20010719-simple-cells
1586		Description:
1587
1588		I have dramatically simplified the interface for CELLS in MLRISC.
1589
1590		In summary, the cells interface is broken up into three parts:
1591
1592		1. CellsBasis : CELLS_BASIS
1593
1594		CellsBasis is a top level structure and common for all
1595		architectures.  it contains the definitions of basic datatypes
1596		and utility  functions over these types.
1597
1598		2. functor Cells() : CELLS
1599
1600		Cells generates an interface for CELLS that incorporates the
1601		specific resources on the target architecture, such as the
1602		presence of special register classes, their number and size,
1603		and various useful substructures.
1604
1605		3. <ARCH>CELLS
1606
1607		e.g. SparcCells: SPARCCELLS
1608
1609		<ARCH>CELLS usually contains additional bindings for special
1610		registers  on the architecture, such as:
1611
1612		val r0 : cell           (* register zero *)
1613		val y : cell            (* Y register *)
1614		val psr : cell          (* processor status register *)
1615		...
1616
1617		The structure returned by applying the Cells functor is opened
1618		in this interface.
1619
1620		The main implication of all this is that the datatypes for cells is
1621		split between CellsBasis and CELLS -- a fairly simple change for user
1622		code.
1623
1624		In the old scheme the CELLS interface had a definitional binding of
1625		the form:
1626
1627		signature CELLS = sig
1628
1629		structure CellsBasis = CellsBasis
1630
1631		...
1632
1633		end
1634
1635		With all the sharing constraints that goes on in MLRISC, this old
1636		design  quickly leads to errors such as:
1637
1638		"structure definition spec inside of sharing ... "
1639
1640
1641		and appears to require an unacceptable amount of sharing and where
1642		constraint hackery.
1643
1644		I think this error message (the interaction of definitional specs and
1645		sharing) requires more explanation on our web page.
1646
1647		----------------------------------------------------------------------
1648		Name: Matthias Blume
1649		Date: 2001/07/19 15:00:00 EDT
1650		Tag: blume-20010719-libreorg
1651		Description:
1652
1653		This update puts together a fairly extensive but straightforward change
1654		to the way the libraries that implement the interactive system are
1655		organized:
1656
1657		The biggest change is the elimination of structure Compiler.  As a
1658		replacement for this structure, there is now a CM library
1659		(known as $smlnj/compiler.cm or $smlnj/compiler/current.cm)
1660		that exports all the substructures of the original structure Compiler
1661		directly.  So instead of saying Compiler.Foo.bar one now simply
1662		says Foo.bar.  (The CM libraries actually export a collection of
1663		structures that is richer than the collection of substructures of
1664		structure Compiler.)
1665
1666		To make the transition smooth, there is a separate library called
1667		$smlnj/compiler/compiler.cm which puts together and exports the
1668		original structure Compiler (or at least something very close to it).
1669
1670		There are five members of the original structure Compiler
1671		that are not exported directly but which instead became members
1672		of a new structure Backend (described by signature BACKEND).  These are:
1673		structure Profile (: PROFILE), structure Compile (: COMPILE), structure
1674		Interact (: INTERACT), structure Machine (: MACHINE), and val
1675		architecture (: string).
1676
1677		Structure Compiler.Version has become structure CompilerVersion.
1678
1679		Cross-compilers for alpha32, hppa, ppc, sparc, and x86 are provided
1680		by $smlnj/compiler/<arch>.cm where <arch> is alpha32, hppa, ppc, sparc,
1681		or x86, respectively.
1682		Each of these exports the same frontend structures that
1683		$smlnj/compiler.cm exports.  But they do not have a structure Backend
1684		and instead export some structure <Arch>Backend where <Arch> is Alpha32,
1685		Hppa, PPC, Sparc, or X86, respectively.
1686
1687		Library $smlnj/compiler/all.cm exports the union of the exports of
1688		$smlnj/compiler/<arch>.cm
1689
1690		There are no structures <Arch>Compiler anymore, use
1691		$smlnj/compiler/<arch>.cm instead.
1692
1693		Library host-compiler-0.cm is gone.  Instead, the internal library
1694		that instantiates CM is now called cm0.cm.  Selection of the host
1695		compiler (backend) is no longer done here but. (Responsibility for it
1696		now lies with $smlnj/compiler/current.cm.  This seems to be more
1697		logical.)
1698
1699		Many individual files have been moved or renamed.  Some files have
1700		been split into multiple files, and some "dead" files have been deleted.
1701
1702		Aside from these changes to library organization, there are also changes
1703		to the way the code itself is organized:
1704
1705		Structure Binfile has been re-implemented in such a way that it no
1706		longer needs any knowledge of the compiler.  It exclusively deals
1707		with the details of binfile layout.  It no longer invokes the
1708		compiler (for the purpose of creating new prospective binfile
1709		content), and it no longer has any knowledge of how to interpret
1710		pickles.
1711
1712		Structure Compile (: COMPILE) has been stripped down to the bare
1713		essentials of compilation.  It no longer deals with linking/execution.
1714		The interface has been cleaned up considerably.
1715
1716		Utility routines for dealing with linking and execution have been
1717		moved into their own substructures.
1718
1719		(The ultimate goal of these changes is to provide a light-weight
1720		binfile loader/linker (at least for, e.g., stable libraries) that
1721		does not require CM or the compiler to be present.)
1722
1723		CM documentation has been updated to reflect the changes to library
1724		organization.
1725
1726		----------------------------------------------------------------------
1727		Name: Matthias Blume
1728		Date: 2001/07/10 17:30:00 EDT
1729		Tag: Release_110_34
1730		Description:
1731
1732		Minor tweak to 110.34 (re-tagged):
1733
1734		- README.html file added to CVS repository
1735		- runtime compiles properly under FreeBSD 3.X and 4.X
1736
1737		----------------------------------------------------------------------
1738		Name: Matthias Blume
1739		Date: 2001/07/10 17:30:00 EDT
1740		Tag: Release_110_34
1741		Description:
1742
1743		New version number (110.34). New bootfiles.
1744
1745		----------------------------------------------------------------------
1746		Name: Matthias Blume
1747		Date: 2001/07/09 16:00:00 EDT
1748		Tag: blume-20010709-more-varargs
1749		Description:
1750
1751		I changed the handling of varargs in ml-nlffigen again:
1752		The ellipsis ... will now simply be ignored (with an accompanying warning).
1753
1754		The immediate effect is that you can actually call a varargs function
1755		from ML -- but you can't actually supply any arguments beyond the ones
1756		specified explicitly.  (For example, you can call printf with its format
1757		string, but you cannot pass additional arguments.)
1758
1759		This behavior is only marginally more useful than the one before, but
1760		it has the advantage that a function or, more importantly, a function
1761		type never gets dropped on the floor, thus avoiding follow-up problems with
1762		other types that refer to the offending one.
1763
1764		----------------------------------------------------------------------
1765		Name: Matthias Blume
1766		Date: 2001/07/09 11:25:00 EDT
1767		Tag: blume-20010709-varargs
1768		Description:
1769
1770		1. ckit-lib.cm now exports structure Error
1771		2. ml-nlffigen reports occurences of "..." (i.e., varargs function types)
1772		with a warning accompanied by a source location.  Moreover, it
1773		merely skips the offending function or type and proceeds with the
1774		rest of its work.u  As a result, one can safely feed C code containing
1775		"..." to ml-nlffigen.
1776		3. There are some internal improvements to CM, providing slightly
1777		more general string substitutions in the tools subsystem.
1778
1779		----------------------------------------------------------------------
1780		Name: Matthias Blume
1781		Date: 2001/06/27 15:10:00 EDT
1782		Tag: blume-20010627-concur
1783		Description:
1784
1785		Fixed a small bug in CM's handling of parallel compilation.
1786		(You could observe the bug by Control-C-interrupting an ordinary
1787		CMB.make or CM.stabilize and then attaching some compile servers.
1788		The result was that all of a sudden the previously interrupted
1789		compilation would continue on its own.  This was because of
1790		an over-optimization: CM did not bother to clean out certain queues
1791		when no servers were attached "anyway", resulting in the contents
1792		of these queues to grab control when new servers did get attached.)
1793
1794		There is also another minor update to the CM manual.
1795
1796		----------------------------------------------------------------------
1797		Name: Matthias Blume
1798		Date: 2001/06/26 16:15:00 EDT
1799		Tag: blume-20010626-cmdoc
1800		Description:
1801
1802		Minor typo fixed in CM manual (syntax diagram for libraries).
1803
1804		----------------------------------------------------------------------
1805		Name: Matthias Blume
1806		Date: 2001/06/25 22:55:00 EDT
1807		Tag: blume-20010625-x86pc
1808		Description:
1809
1810		Fixed a nasty bug in the X86 assembly code that caused signal
1811		handlers to fail (crash) randomly.
1812
1813		----------------------------------------------------------------------
1814		Name: Matthias Blume
1815		Date: 2001/06/25 12:05:00 EDT
1816		Tag: blume-20010625-nlffigen
1817		Description:
1818
1819		This update fixes a number of minor bugs in ml-nlffigen as reported by
1820		Nick Carter <nbc@andrew.cmu.edu>.
1821
1822		1. Silly but ok typedefs of the form "typedef void myvoid;" are now accepted.
1823		2. Default names for generated files are now derived from the name of
1824		the C file *without its directory*.  In particular, this causes generated
1825		files to be placed locally even if the C file is in some system directory.
1826		3. Default names for generated signatures and structures are also derived
1827		from the C file name without its directory.  This avoids silly things
1828		like "structure GL/GL".
1829		(Other silly names are still possible because ml-nlffigen does not do
1830		a thorough check of whether generated names are legal ML identifiers.
1831		When in doubt, use command line arguments to force particular names.)
1832
1833		----------------------------------------------------------------------
1834		Name: Matthias Blume
1835		Date: 2001/06/21 12:25:00 EDT
1836		Tag: blume-20010621-eXene
1837		Description:
1838
1839		eXene now compiles and (sort of) works again.
1840
1841		The library name (for version > 110.33) is $/eXene.cm.
1842
1843		I also added an new example in src/eXene/examples/nbody.  See the
1844		README file there for details.
1845
1846		----------------------------------------------------------------------
1847		Name: Matthias Blume
1848		Date: 2001/06/20 16:40:00 EDT
1849		Tag: blume-20010620-cml
1850		Description:
1851
1852		CML now compiles and works again.
1853
1854		Libraries (for version > 110.33):
1855
1856		$cml/cml.cm            Main CML library.
1857		$cml/basis.cm          CML's version of $/basis.cm.
1858		$cml/cml-internal.cm   Internal helper library.
1859		$cml/core-cml.cm       Internal helper library.
1860		$cml-lib/trace-cml.cm  Tracing facility.
1861		$cml-lib/smlnj-lib.cm  CML's version of $/smlnj-lib.cm
1862
1863		The installer (config/install.sh) has been taught how to properly
1864		install this stuff.
1865
1866		----------------------------------------------------------------------
1867		Name: Matthias Blume
1868		Date: 2001/06/19 17:55:00 EDT
1869		Tag: blume-20010619-instantiate
1870		Description:
1871
1872		This un-breaks the fix for bug 1432.
1873		(The bug was originally fixed in 110.9 but I broke it again some
1874		time after that.)
1875
1876		----------------------------------------------------------------------
1877		Name: Matthias Blume
1878		Date: 2001/06/19 17:25:00 EDT
1879		Tag: blume-20010619-signals
1880		Description:
1881
1882		This should (hopefully) fix the long-standing signal handling bug.
1883		(The runtime system was constructing a continuation record with an
1884		incorrect descriptor which would cause the GC to drop data on the floor...)
1885
1886		----------------------------------------------------------------------
1887		Name: Matthias Blume
1888		Date: 2001/06/15 15:05:00 EDT
1889		Tag: blume-20010615-moresparc
1890		Description:
1891
1892		Here is a short late-hour update related to Sparc c-calls:
1893
1894		-- made handling of double-word arguments a bit smarter
1895
1896		-- instruction selection phase tries to collapse certain clumsily
1897		constructed ML-Trees; typical example:
1898
1899		ADD(ty,ADD(_,e,LI d1),LI d2)  ->  ADD(ty,e,LI(d1+d2))
1900
1901		This currently has no further impact on SML/NJ since mlriscGen does
1902		not seem to generate such patterns in the first place, and c-calls
1903		(which did generate them in the beginning) has meanwhile been fixed
1904		so as to avoid them as well.
1905
1906		----------------------------------------------------------------------
1907		Name: Matthias Blume
1908		Date: 2001/06/15 15:05:00 EDT
1909		Tag: blume-20010615-sparc
1910		Description:
1911
1912		The purpose of this update is to provide an implementation of NLFFI
1913		on Sparc machines.
1914
1915		Here are the changes in detail:
1916
1917		* src/MLRISC/sparc/c-calls/sparc-c-calls.sml is a new file containing
1918		the Sparc implementation of the c-calls API.
1919		* The Sparc backend of SML/NJ has been modified to uniformely use %fp
1920		for accessing the ML frame.  Thus, we have a real frame pointer and
1921		can freely modify %sp without need for an omit-frame-ptr phase.
1922		The vfp logic in src/compiler/CodeGen/* has been changed to accomodate
1923		this case.
1924		* ml-nlffigen has been taught to produce code for different architectures
1925		and calling conventions.
1926		* In a way similar to what was done in the x86 case, the Sparc
1927		backend uses its own specific extension to mltree.  (For example,
1928		it needs to be able to generate UNIMP instructions which are part
1929		of the calling convention.)
1930		* ml-nlffi-lib was reorganized to make it more modular (in particular,
1931		to make it easier to plug in new machine- and os-dependent parts).
1932
1933		There are some other fairly unrelated bug fixes and cleanups as well:
1934
1935		* I further hacked the .cm files for MLRISC tools (like MDLGen) so
1936		that they properly share their libraries with existing SML/NJ libraries.
1937		* I fixed a minor cosmetic bug in CM, supressing certain spurious
1938		follow-up error messages.
1939		* Updates to CM/CMB documentation.
1940
1941		TODO items:
1942
1943		* MLRISC should use a different register as its asmTemp on the Sparc.
1944		(The current %o2 is a really bad choice because it is part of the
1945		calling conventions, so things might interfere in unexpected ways.)
1946
1947		----------------------------------------------------------------------
1948		Name: Matthias Blume
1949		Date: 2001/06/07
1950		Tag: blume-20010607-calls
1951		Description:
1952
1953		A number of internal changes related to C calls and calling conventions:
1954
1955		1. ML-Tree CALL statements now carry a "pops" field.  It indicates the
1956		number of bytes popped implicitly (by the callee).  In most cases
1957		this field is 0 but on x86/win32 it is some non-zero value.  This
1958		is information provided for the benefit of the "omit-frameptr" pass.
1959		2. The CALL instruction on the x86 carries a similar "pops" field.
1960		The instruction selection phase copies its value from the ML-Tree
1961		CALL statement.
1962		3. On all other architectures, the instruction selection phase checks
1963		whether "pops=0" and complains if not.
1964		4. The c-calls implementation for x86 now accepts two calling conventions:
1965		"ccall" and "stdcall".  When "ccall" is selected, the caller cleans
1966		up after the call and pops is set to 0.  For "stdcall", the caller
1967		does nothing, leaving the cleanup to the callee; pops is set to
1968		the number of bytes that were pushed onto the stack.
1969		5. The cproto decoder (compiler/Semant/types/cproto.sml) now can
1970		distinguish between "ccall" and "stdcall".
1971		6. The UNIMP instruction has been added to the supported Sparc instruction
1972		set. (This is needed for implementing the official C calling convention
1973		on this architecture.)
1974		7. I fixed some of the .cm files under src/MLRISC/Tools to make them
1975		work with the latest CM.
1976
1977		----------------------------------------------------------------------
1978		Name: Matthias Blume
1979		Date: 2001/06/05 15:10:00 EDT
1980		Tag: blume-20010605-cm-index
1981		Description:
1982
1983		0. The "lambdasplit" parameter for class "sml" in CM has been documented.
1984
1985		1. CM can now generate "index files".  These are human-readable files
1986		that list on a per-.cm-file basis each toplevel symbol defined or
1987		imported.  The location of the index file for
1988		<p>/<d>.cm is <p>/CM/INDEX/<d>.cm.
1989		To enable index-file generation, set CM.Control.generate_index to true
1990		or export an environment-symbol: export CM_GENERATE_INDEX=true.
1991
1992		The CM manual has been updated accordingly.
1993
1994		2. I made some slight modifications to the c-calls API in MLRISC.
1995
1996		a) There is now a callback to support saving/restoring of
1997		dedicated but caller-save registers around the actual call
1998		instruction.
1999		b) One can optionally specify a comment-annotation for the
2000		call instruction.
2001
2002		3. SML/NJ (mlriscGen.sml) uses this new API for the rawccall primop.
2003		(For example, the comment annotation shows the C prototype of
2004		the function being called.)
2005
2006		----------------------------------------------------------------------
2007		Name: Matthias Blume
2008		Date: 2001/06/01 13:30:00 EDT
2009		Tag: blume-20010601-nlffi-cleanup
2010		Description:
2011
2012		This is mostly a cleanup of MLFFI stuff:
2013
2014		- some signature files have been put into a more exposed place
2015		- the ugly 'f type parameter is gone (simplifies types tremendously!)
2016		- ml-nlffigen changed accordingly
2017		- tutorial updated
2018
2019		Other changes:
2020
2021		- author's affiliation in CM manual(s) updated
2022		- some more recognized keywords added to Allen's sml.sty
2023
2024		----------------------------------------------------------------------
2025		Name: Matthias Blume
2026		Date: 2001/05/25 15:30:00 EDT
2027		Tag: blume-20010525-iptr
2028		Description:
2029
2030		- put the official 110.33-README (as it appears on the ftp server) under
2031		CVS
2032		- fixed a small bug related to incomplete pointer types in
2033		ml-nlffigen
2034		- small cosmetic change to the ml-nlffi-lib's "arr" type constructor
2035		(it does not need the 'f type parameter)
2036
2037		----------------------------------------------------------------------
2038		Name: Matthias Blume
2039		Date: 2001/05/23 14:30:00 EDT
2040		Tag: Release_110_33
2041		Description:
2042
2043		New version number (110.33).  New bootfiles.
2044
2045		----------------------------------------------------------------------
2046		Name: Matthias Blume
2047		Date: 2001/05/22 18:06:00 EDT
2048		Tag: blume-20010522-targets
2049		Description:
2050
2051		Made install.sh use file config/targets.customized if it exists, falling
2052		back to config/targets if it doesn't.  This way one can have a customized
2053		version of the targets file without touching the "real thing", thus
2054		eliminating the constant fear of accidentally checking something bogus
2055		back into the CVS repository...  (File config/targets.customized must
2056		not be added to the repository!)
2057
2058		----------------------------------------------------------------------
2059		Name: Matthias Blume
2060		Date: 2001/05/22 16:30:00 EDT
2061		Tag: blume-20010522-minitut
2062		Description:
2063
2064		1. Bug fix in ml-nlffigen; now (hopefully) correctly handling
2065		struct returns.
2066		2. Added src/ml-nlffi-lib/Doc/mini-tutorial.txt.  This is some very
2067		incomplete, preliminary documentation for NLFFI.
2068
2069		----------------------------------------------------------------------
2070		Name: Matthias Blume
2071		Date: 2001/05/14 11:30:00 EDT
2072		Tag: blume-20010514-script
2073		Description:
2074
2075		Some bugs in install script fixed.
2076
2077		In addition to that I also made a slight change to the NLFFI API:
2078		Functors generated by ml-nlffigen now take the dynamic library as a
2079		straight functor argument, not as a suspended one.  (The original
2080		functor code used to force the suspension right away anyway, so there
2081		was nothing gained by this complication of the interface.)
2082
2083		----------------------------------------------------------------------
2084		Name: Matthias Blume
2085		Date: 2001/05/11 14:35:00 EDT
2086		Tag: blume-20010511-ml-nlffi
2087		Description:
2088
2089		I finally took the plunge and added my new FFI code to the main
2090		repository.  For x86-linux it is now ready for prime-time.
2091
2092		There are two new subdirectories of "src":
2093
2094		- ml-nlffi-lib:
2095		The utility library for programs using the FFI interface.
2096		Here is the implementation of $/c.cm and its associated low-level
2097		partners $/c-int.cm and $/memory.cm.
2098		- ml-nlffigen:
2099		A stand-alone program for generating ML glue code from C source
2100		code.
2101
2102		Building ml-nlffigen requires $/ckit-lib.cm.
2103
2104		The config/install.sh script has been updates to do the Right Thing
2105		(hopefully).
2106
2107		Notice that the source tree for the C-Kit will not be put under "src"
2108		but directly under the installation root directory.  (This is the
2109		structure that currently exists on the CVS server when you check out
2110		module "sml".)  Fortunately, config/install.sh knows about this oddity.
2111
2112		Bugs: No documentation yet.
2113
2114		----------------------------------------------------------------------
2115		Name: Matthias Blume
2116		Date: 2001/05/09 16:35:00 EDT
2117		Tag: blume-20010509-cpscontract
2118		Description:
2119
2120		Fixed a bug in the accounting code in cpsopt/contract.sml.  (The
2121		wrapper/unwrapper elimination did not decrement usage counts and some
2122		dead variables got overlooked by the dead-up logic.)
2123
2124		----------------------------------------------------------------------
2125		Name: Lal George
2126		Date: 2001/05/08  17:26:09 EDT
2127		Tag: george-20010508-omit-frameptr
2128		Description:
2129
2130		Changes to implement the omit-frame-pointer optimization to support
2131		raw C calls. For now, there is only support on the Intel x86, but
2132		other architectures will follow as more experience is gained with this.
2133
2134
2135		----------------------------------------------------------------------
2136		Name: Matthias Blume
2137		Date: 2001/05/07 14:40:00 EDT
2138		Tag: blume-20010507-proxies
2139		Description:
2140
2141		I made into "proxy libraries" all libraries that qualify for such a
2142		change.  (A qualifying library is a library that has another library or
2143		groups as its sole member and repeats that member's export list
2144		verbatim.  A proxy library avoids this repetition by omitting its export
2145		list, effectively inheriting the list that its (only) member exports.
2146		See the CM manual for more explanation.)
2147		The main effect is that explicit export lists for these libraries
2148		do not have to be kepts in sync, making maintenance a bit easier.
2149
2150		I also added copyright notices to many .cm-files.
2151
2152		Last but not least, I made a new set of bootfiles.
2153
2154		----------------------------------------------------------------------
2155		Name: Matthias Blume
2156		Date: 2001/05/04 17:00:00 EDT
2157		Tag: blume-20010504-cm-lsplit
2158		Description:
2159
2160		0. John merged pending changes to $/smlnj-lib.cm
2161
2162		1. Allen's previous change accidentally backed out of one of Lal's
2163		earlier changes.  I undid this mistake (re-introducing Lal's change).
2164
2165		2. I used the new topOrder' function from graph-scc.sml (from $/smlnj-lib.cm)
2166		within the compiler where applicable.  There is some code simplification
2167		because of that.
2168
2169		3. The "split" phase (in FLINT) is now part of the default list of phases.
2170		Compiler.Control.LambdaSplitting.* can be used to globally control the
2171		lambda-splitting (cross-module-inlining) engine.  In addition to that,
2172		it can now also be controlled on a per-source basis: CM has been taught
2173		a new tool parameter applicable to ML source files.
2174
2175		- To turn lambda-splitting off completely:
2176		local open Compiler.Control.LambdaSplitting in
2177		val _ = set Off
2178		end
2179		- To make "no lambda-splitting" the global default (but allow per-source
2180		overriding); this is the initial setting:
2181		local open Compiler.Control.LambdaSplitting in
2182		val _ = set (Default NONE)
2183		end
2184		- To make "lambda-splitting with aggressiveness a" the global default
2185		(and allow per-source overriding):
2186		local open Compiler.Control.LambdaSplitting in
2187		val _ = set (Default (SOME a))
2188		end
2189
2190		- To turn lambda-splitting off for a given ML souce file (say: a.sml)
2191		write (in the respective .cm-file):
2192		a.sml (lambdasplitting:off)
2193		- To turn lambda-splitting for a.sml on with minimal aggressiveness:
2194		a.sml (lambdasplitting:on)
2195		- To turn lambda-splitting for a.sml on with aggressiveness <a> (where
2196		<a> is a decimal non-negative integer):
2197		a.sml (lambdasplitting:<a>)
2198		- To turn lambda-splitting for a.sml on with maximal aggressiveness:
2199		a.sml (lambdasplitting:infinity)
2200		- To use the global default for a.sml:
2201		a.sml (lambdasplitting:default)
2202		or simply
2203		a.sml
2204
2205		----------------------------------------------------------------------
2206		Name: Allen Leung
2207		Date: 2001/05/04 01:57:00 EDT
2208		Tag: leunga-20010504-sync
2209		Description:
2210
2211		MLRISC features.
2212
2213		1. Fix to CMPXCHG instructions.
2214		2. Changed RA interface to allow annotations in callbacks.
2215		3. Added a new method to the stream interface to allow annotations updates.
2216
2217		----------------------------------------------------------------------
2218		Name: Matthias Blume
2219		Date: 2001/05/01 11:45:00 EDT
2220		Tag: blume-20010501-pcedittmp
2221		Description:
2222
2223		Changed install.sh to use the current working directory instead of
2224		/usr/tmp for a temporary file (pcedittmp).  The previous choice
2225		of /usr/tmp caused trouble with MacOS X because of file premission
2226		problems.
2227
2228		----------------------------------------------------------------------
2229		Name: Matthias Blume
2230		Date: 2001/04/20 11:10:00 EDT
2231		Tag: blume-20010420-inMLflag
2232		Description:
2233
2234		- added vp_limitPtrMask to vproc-state.h
2235		(for use by the raw-C-calls mechanism to implement proper interrupt
2236		handling)
2237		- made the ML compiler aware of various data-structure offsets so it
2238		can generate code for accessing the vp_inML flag and vp_limitPtrMask
2239		- tweaked mlriscGen.sml to have it emit interrupt-handling code for
2240		raw C-calls
2241
2242		----------------------------------------------------------------------
2243		Name: Lal George
2244		Date: 2001/04/20 09:15:28 EDT
2245		Tag: george-20010420-macosX
2246		Description:
2247
2248		- Changes to port to Mac OS X; Darwin.
2249
2250		- In the process I found that sqrt was broken on the PPC, because the
2251		fsqrt instruction is not implemented.
2252
2253		----------------------------------------------------------------------
2254		Name: Matthias Blume
2255		Date: 2001/04/18 12:45:00 EDT
2256		Tag: blume-20010418-ccalls
2257		Description:
2258
2259		- fixed two off-by-4 errors in the x86-specific c-calls implementation
2260		(this bug prevented structure arguments containing pointers from being
2261		passed correctly)
2262		- changed the raw-C-call code in mlriscGen.sml in such a way that
2263		structure arguments are represented as a pointer to the beginning
2264		of the structure (instead of having a series of synthesized arguments,
2265		one for each structure member)
2266
2267		- made makeml script's verbosity level configurable via environment
2268		variable (MAKEML_VERBOSITY)
2269
2270		- eliminated placeholder implementations for f32l, w16s, i16s, and f32s
2271		in rawmem-x86.sml; we are now using the real thing
2272
2273		----------------------------------------------------------------------
2274		Name: Matthias Blume
2275		Date: 2001/03/22 16:25:00 EST
2276		Tag: blume-20010322-bootfiles
2277		Description:
2278
2279		Created a new set of bootfiles (for your automatic installation convenience).
2280
2281		----------------------------------------------------------------------
2282		Name: Matthias Blume
2283		Date: 2001/03/22 15:10:00 EST
2284		Tag: blume-20010322-rawmem-parcm
2285		Description:
2286
2287		1. All "raw memory access" primitives for the new FFI are implemented now
2288		(at least on the x86).
2289		2. Some further cleanup of CM's parallel make mechanism.
2290
2291		----------------------------------------------------------------------
2292		Name: Matthias Blume
2293		Date: 2001/03/19 17:53:00 EST
2294		Tag: blume-20010319-parallel
2295		Description:
2296
2297		Parallel make (using compile servers) now works again.
2298
2299		To this end, CM.stabilize and CMB.make have been modified to work in
2300		two passes when compile servers are attached:
2301		1. Compile everything, do not perform stabilization; this pass
2302		uses compile servers
2303		2. Stabilize everything; this pass does not use compile servers
2304		If there are no compile servers, the two passes are combined into one
2305		(as before).  Splitting the passes increases the inherent parallelism
2306		in the dependency graph because the entire graph including all
2307		libraries is available at the same time.  This, in turn, improves
2308		server utilization.  The downside is that the master process will
2309		have to do some extra work after compilation is done (because for
2310		technical reasons it must re-read all the binfiles during stabilization).
2311
2312		----------------------------------------------------------------------
2313		Name: Matthias Blume
2314		Date: 2001/03/16 12:22:00 EST
2315		Tag: blume-20010316-bootfiles
2316		Description:
2317
2318		Created a new set of bootfiles (for your automatic installation convenience).
2319
2320		----------------------------------------------------------------------
2321		Name: Matthias Blume
2322		Date: 2001/03/16 11:00:00 EST
2323		Tag: blume-20010316-MLTREE-fixup
2324		Description:
2325
2326		This is a minor fixup for an (untagged) earlier commit by Allen.
2327		(A file was missing).
2328
2329		----------------------------------------------------------------------
2330		Name: Allen Leung
2331		Date: Mon Mar  5 18:54:57 EST 2001
2332		Tag: leunga-20010305-cut-support
2333
2334		1. New support for alternative control-flow in MLTREE.
2335		Currently we support
2336
2337		FLOW_TO(CALL ...., [k1,...,kn])
2338
2339		This is needed for 'cuts to' in C-- and try/handle-like constructs
2340		in Moby
2341
2342		New assembler flag "asm-show-cutsto" to turn on control-flow debugging.
2343
2344		2. Register Allocator
2345
2346		Changes in interface [from Fermin, John]
2347
2348		3. Alpha 8-bit SLL support [Fermin]
2349
2350		4. All architectures
2351
2352		A new module (ClusterExpandCopies) for expanding parallel copies.
2353
2354		----------------------------------------------------------------------
2355		Name: Allen Leung
2356		Date: 2001/02/27 23:07:00 EST
2357		Tag: leunga-20010227-minor-stuff
2358
2359		1. Alpha bug fix for CMOVNE
2360		2. Handle mltree COND(..,FCMP ...,...)
2361		3. Bug fix in simplifier
2362
2363		----------------------------------------------------------------------
2364		Name: Matthias Blume
2365		Date: 2001/01/30 17:50:00 EST
2366		Tag: blume-20010130-sync
2367		Description:
2368
2369		This is just a minor update to sync my devel branch with the main brach.
2370		The only visible change is the addition of some README files.
2371
2372		----------------------------------------------------------------------
2373		Name: Matthias Blume
2374		Date: 2001/01/12 23:30:00 JST
2375		Tag: blume-20010112-bootfiles
2376		Description:
2377
2378		Made a new set of bootfiles that goes with the current state of the
2379		repository.
2380
2381		----------------------------------------------------------------------
2382		Name: Matthias Blume
2383		Date: 2001/01/12 21:20:00 JST
2384		Tag: blume-20010112-sync
2385		Description:
2386
2387		I am just flushing out some minor changes that had accumulated in
2388		my private branch in order to sync with the main tree.  (This is
2389		mainly because I had CVS trouble when trying to merge _into_ my
2390		private branch.)
2391
2392		Most people should be completely unaffected by this.
2393
2394		----------------------------------------------------------------------
2395		Name: Allen Leung
2396		Date: Thu Jan 11 21:03:00 EST 2001
2397		Tag: leunga-20010111-labexp=mltree
2398		Description:
2399
2400		1.  Removed the type LabelExp and replace it by MLTree.
2401		2.  Rewritten mltree-simplify with the pattern matcher tool.
2402		3.  There were some bugs in alpha code generator which would break
2403		64-bit code generation.
2404		4.  Redo the tools to generate code with the
2405		5.  The CM files in MLRISC (and in src/system/smlnj/MLRISC)
2406		are now generated by perl scripts.
2407
2408		----------------------------------------------------------------------
2409		Name: Matthias Blume
2410		Date: 2001/01/10 21:55:00 JST
2411		Tag: blume-20010110-rcc
2412		Description:
2413
2414		The RCC stuff now seems to work (but only on the x86).
2415		This required hacking of the c-calls interface (and -implementation) in
2416		MLRISC.
2417
2418		Normal compiler users should be unaffected.
2419
2420		----------------------------------------------------------------------
2421		Name: Matthias Blume
2422		Date: 2001/01/09 01:20:00 JST
2423		Tag: blume-20010109-rcc
2424		Description:
2425
2426		This is a fairly big patch, flushing out a large number of pending
2427		changes that I made to my development copy over the last couple of days.
2428
2429		Of practical relevance at this moment is a workaround for a pickling
2430		bug that Allen ran into the other day.  The cause of the bug itself is
2431		still unknown and it might be hard to fix it properly, but the
2432		workaround has some merits of its own (namely somewhat reducing pickling
2433		overhead for certain libraries).  Therefore, I think this solution should
2434		be satisfactory at this time.
2435
2436		The rest of the changes (i.e., the vast majority) has to do with my
2437		ongoing efforts of providing direct support for C function calls from
2438		ML.  At the moment there is a new primop "RAW_CCALL", typing magic
2439		in types/cproto.sml (invoked from FLINT/trans/translate.sml), a new
2440		case in the FLINT CPS datatype (RCC), changes to cps/convert.sml to
2441		translate uses of RAW_CCALL into RCC, and changes to mlriscGen.sml to
2442		handle RCC.
2443
2444		The last part (the changes to mlriscGen.sml) are still known to be
2445		wrong on the x86 and not implemented on all other architectures.  But
2446		the infrastructure is in place. I had to change a few functor
2447		signatures in the backend to be able to route the CCalls interface
2448		from MLRISC there, and I had to specialize the mltree type (on the
2449		x86) to include the necessary extensions. (The extensions themselves
2450		were already there and redy to go in MLRISC/x86).
2451
2452		Everything should be very happy as soon as someone helps me with
2453		mlriscGen.sml...
2454
2455		In any case, nothing of this should matter to anyone as long as the
2456		new primop is not being used (which is going to be the case unless you
2457		find it where I hid it :). The rest of the compiler is completely
2458		unaffected.
2459
2460		----------------------------------------------------------------------
2461		Name: Matthias Blume
2462		Date: 2001/01/05 00:30:00 JST
2463		Tag: blume-20010105-primops
2464		Description:
2465
2466		Added some experimental support for work that I am doing right now.
2467		These changes mostly concern added primops, but there is also a new
2468		experimental C library in the runtime system (but currently not enabled
2469		anywhere except on Linux/X86).
2470
2471		In the course of adding primops (and playing with them), I discovered that
2472		Zhong's INL_PRIM hack (no type info for certain primops) was, in fact, badly
2473		broken.  (Zhong was very right he labeled this stuff as "major gross hack".)
2474		To recover, I made type information in INL_PRIM mandatory and changed
2475		prim.sml as well as built-in.sml accordingly.  The InLine structure now
2476		has complete, correct type information (i.e., no bottom types).
2477
2478		Since all these changes mean that we need new binfiles, I also bumped the
2479		version number to 110.32.1.
2480
2481		----------------------------------------------------------------------
2482		Name: Matthias Blume
2483		Date: 2000/12/30 22:10:00 JST
2484		Tag: blume-20001230-various
2485		Description:
2486
2487		Added proxy libraries for MLRISC and let MLRISC libraries refer
2488		to each other using path anchors.  (See CM manual for explanation.)
2489
2490		Updated CM documentation.
2491
2492		Fixed some bugs in CM.
2493
2494		Implemented "proxy" libraries (= syntactic sugar for CM).
2495
2496		Added "-quiet" option to makeml and changed runtime system accordingly.
2497
2498		Added cleanup handler for exportML to reset timers and compiler stats.
2499
2500		----------------------------------------------------------------------
2501		Name: Lal George
2502		Date: 2000/12/22 22:22:58 EST 2000
2503		Tag: Release_110_32
2504		Description:
2505
2506		Infinite precision used throughout MLRISC.
2507		see MLRISC/mltree/machine-int.sig
2508
2509		----------------------------------------------------------------------
2510		Name: Matthias Blume
2511		Date: 2000/12/22 23:16:00 JST
2512		Tag: blume-20001222-warn
2513		Description:
2514
2515		Corrected wording and formatting of some CM warning message which I
2516		broke in my previous patch.
2517
2518		----------------------------------------------------------------------
2519		Name: Matthias Blume
2520		Date: 2000/12/22 21:20:00 JST
2521		Tag: blume-20001222-anchorenv
2522		Description:
2523
2524		Fixed CM's handling of anchor environments in connection with CMB.make.
2525
2526		----------------------------------------------------------------------
2527		Name: Matthias Blume
2528		Date: 2000/12/22 13:15:00 JST
2529		Tag: blume-20001222-cleanup
2530		Description:
2531
2532		Removed src/cm/ffi which does not (and did not) belong here.
2533
2534		----------------------------------------------------------------------
2535		Name: Matthias Blume
2536		Date: 2000/12/21 23:55:00 JST
2537		Tag: blume-20001221-exn
2538		Description:
2539
2540		Probably most important: CM no longer silently swallows all exceptions
2541		in the compiler.
2542		Plus: some other minor CM changes.  For example, CM now reports some
2543		sizes for generated binfiles (code, data, envpickle, lambdapickle).
2544
2545		----------------------------------------------------------------------
2546		Name: Matthias Blume
2547		Date: 2000/12/15 00:01:05 JST
2548		Tag: blume-20001215-dirtool
2549		Description:
2550
2551		- "dir" tool added.
2552		- improvements and cleanup to Tools structure
2553		- documentation updates
2554
2555		----------------------------------------------------------------------
2556		Name: Allen Leung
2557		Date: Thu Dec 14 03:45:24 EST 2000
2558		Description:
2559		Tag:  leunga-20001214-int-inf
2560		Description:
2561
2562		In IntInf, added these standard functions, which are missing from our
2563		implementation:
2564
2565		andb : int * int -> int
2566		xorb : int * int -> int
2567		orb  : int * int -> int
2568		notb : int -> int
2569		<<   : int * word -> int
2570		~>>  : int * word -> int
2571
2572		Not tested, I hope they are correct.
2573
2574		----------------------------------------------------------------------
2575		Name: Allen Leung
2576		Date: Fri Dec  8 19:23:26 EST 2000
2577		Description:
2578		Tag:  leunga-20001208-nowhere
2579		Description:
2580
2581		Slight improvements to the 'nowhere' tool to handle OR-patterns,
2582		to generate better error messages etc.  Plus a brief manual.
2583
2584		----------------------------------------------------------------------
2585		Name: Lal George
2586		Date: 2000/12/08 09:54:02 EST 2000
2587		Tag: Release_110_31
2588		Description:
2589
2590		- Version 110.31
2591		----------------------------------------------------------------------
2592		Name: Allen Leung
2593		Date: Thu Dec  7 22:01:04 EST 2000
2594		Tag:  leunga-20001207-cell-monster-hack
2595		Description:
2596
2597		Major MLRISC internal changes.  Affect all clients.
2598		Summary:
2599
2600		1.  Type CELLS.cell = int is now replaced by a datatype.
2601		As a result, the old regmap is now gone.  Almost all interfaces
2602		in MLRISC change as a consequence.
2603
2604		2.  A new brand version of machine description tool (v3.0) that generates
2605		modules expecting the new interface.  The old version is removed.
2606
2607		3.  The RA interface has been further abstracted into two new functors.
2608		RISC_RA and X86RA.  These functors have much simpler interfaces.
2609		[See also directory MLRISC/demo.]
2610
2611		4.  Some other new source->source code generation tools are available:
2612
2613		a. MLRISC/Tools/RewriteGen -- generate rewriters from rules.
2614		b. MLRISC/Tools/WhereGen -- expands conditional pattern matching rules.
2615		I use this tool to generate the peephole optimizers---with the new
2616		cell type changes, peephole rules are becoming difficult to write
2617		without conditional pattern matching.
2618
2619		5.  More Intmap -> IntHashTable change.  Previous changes by Matthias didn't
2620		cover the entire MLRISC source tree so many things broke.
2621
2622		6.  CM files have been moved to the subdirectory MLRISC/cm.
2623		They are moved because there are a lot of them and they clutter up the
2624		root dir.
2625
2626		7.  More detailed documentation to come...
2627
2628		NOTE: To rebuild from 110.30 (ftp distribution), you'll have to do
2629		a makeml -rebuild first.  This is because of other other
2630		changes that Matthias has made (see below).
2631
2632
2633		----------------------------------------------------------------------
2634		Name: Matthias Blume
2635		Date: 2000/11/30 23:12:00 JST
2636		Tag: blume-20001130-filereorg
2637		Description:
2638
2639		Some manual updates and some file reorganizations in CM.
2640
2641		----------------------------------------------------------------------
2642		Name: Matthias Blume
2643		Date: 2000/11/24 17:45:00 JST
2644		Tag: blume-20001124-link
2645		Description:
2646
2647		Drastically improved link traversal code for the case that the dynamic
2648		value was already loaded at bootstrap time.  As a result, CM and CMB
2649		now both load blazingly fast -- even on a very slow machine.  Also,
2650		memory consumption has been further reduced by this.
2651
2652		Warning: The format of the PIDMAP file has changed.  THerefore, to
2653		bootstrap you have to do this:
2654
2655		1. Run CMB.make
2656		2. Make a symbolic link for the boot directory:
2657		ln -s sml.boot.ARCH-OS xxx
2658		3. "Rebuild" the boot directory:
2659		./makeml -boot xxx -rebuild sml ; rm xxx
2660		4. Boot normally:
2661		./makeml
2662
2663		----------------------------------------------------------------------
2664		Name: Matthias Blume
2665		Date: 2000/11/21 21:20:00 JST
2666		Tag: blume-20001121-tools
2667		Description:
2668
2669		Continued hacking on autoloading problem -- with success this time.
2670		Also changed tool-plugin mechanism.  See new CM manual.
2671
2672		----------------------------------------------------------------------
2673		Name: Matthias Blume
2674		Date: 2000/11/19 14:30:00 JST
2675		Tag:  blume-20001119-autoload
2676		Description:
2677
2678		Some hacking to make autoloading faster.  Success for CMB, no success
2679		so far for CM.  There is a reduced structure CM' that autoloads faster.
2680		(This is a temporary, non-documented hack to be eliminated again when
2681		the general problem is solved.)
2682
2683		----------------------------------------------------------------------
2684		Name: Matthias Blume
2685		Date: 2000/11/17 14:10:00 JST
2686		Tag: blume-20001117-pickle-lib
2687		Description:
2688
2689		1. Eliminated comp-lib.cm
2690		2. Made pickle-lib.cm
2691		3. Eliminated all uses of intset.sml (from comp-lib.cm)
2692		4. Replaced all uses of intmap.{sig,sml} (from comp-lib.cm) with
2693		equivalent constructs from smlnj-lib.cm (INtHashTable).
2694		5. Point 4. also goes for those uses of intmap.* in MLRISC.
2695		Duplicated intmap modules thrown out.
2696		6. Hunted down all duplicated SCC code and replaced it with
2697		equivalent stuff (GraphSCCFn from smlnj-lib.cm).
2698		7. Rewrote Feedback module.
2699		8. Moved sortedlist.sml into viscomp-lib.cm.  Eventually it
2700		should be thrown out and equivalent modules from smlnj-lib.cm
2701		should be used (IntRedBlackSet, IntListSet, ...).
2702
2703		Confirmed that compiler compiles to fixpoint.
2704
2705		----------------------------------------------------------------------
2706		Name: Allen Leung
2707		Date: 2000/11/10 18:00:00
2708		Tag: leunga-20001110-new-x86-fp
2709
2710		A new x86 floating point code generator has been added.
2711		By default this is turned off.  To turn this on, do:
2712
2713		CM.autoload "$smlnj/compiler.cm";
2714		Compiler.Control.MLRISC.getFlag "x86-fast-fp" := true;
2715
2716		Changes:
2717
2718		1.  Changed FTAN to FPTAN so that the assembly output is correct.
2719		2.  Changed the extension callback for FTANGENT to generate:
2720
2721		fptan
2722		fstp  %st(0)
2723		instead of
2724		fptan
2725		fstpl ftempmem
2726
2727		3.  Numerous assembly fixes for x86.
2728
2729		5.  Cleaned up the machine code output module x86/x86MC.sml and added
2730		support for a whole bunch of instructions and addressing modes:
2731
2732		fadd/fsub/fsubr/fmul/fdiv/fdivr  %st, %st(n)
2733		faddp/fsubp/fsubrp/fmulp/fdivp/fdivrp  %st, %st(n)
2734		fadd/fsub/fsubr/fmul/fdiv/fdivr  %st(n), %st
2735		fiadd/fisub/fisubr/fimul/fidiv/fidivr mem
2736		fxch %st(n)
2737		fld %st(n)
2738		fst %st(n)
2739		fst mem
2740		fstp %st(n)
2741		fucom %st(n)
2742		fucomp %st(n)
2743
2744		All these are now generated when the fast fp mode is turned on.
2745
2746		6.  Removed the dedicated registers %st(0), ..., %st(7) from X86CpsRegs
2747
2748		----------------------------------------------------------------------
2749		Name: Matthias Blume
2750		Date: 2000/11/09 11:20:00 JST
2751		Tag: blume-20001109-scc
2752		Description:
2753
2754		Eliminated some code duplication:
2755
2756		1. Added "where" clause to GraphSCCFn in SML/NJ Library.
2757		(Otherwise the functor is useless.)
2758		2. Used GraphSCCFn where SCCUtilFun was used previously.
2759		3. Got rid of SCCUtilFun (in comp-lib.cm).
2760
2761		----------------------------------------------------------------------
2762		Name: Lal George
2763		Date: 2000/11/06 09:02:21 EST 2000
2764		Tag: Release_110_30
2765		Description:
2766
2767		- Version 110.30
2768		----------------------------------------------------------------------
2769		Name: Matthias Blume
2770		Date: 2000/11/04 14:45:00
2771		Tag: blume-20001104-mlbuild
2772		Description:
2773
2774		- Made ml-build faster on startup.
2775		- Documentation fixes.
2776
2777		----------------------------------------------------------------------
2778		Name: Matthias Blume
2779		Date: 2000/11/02 17:00:00 JST
2780		Tag: blume-20001102-condcomp
2781		Description:
2782
2783		- Small tweaks to pickler -- new BOOTFILES!
2784		- Version bumped to 110.29.2.
2785		- Added conditional compilation facility to init.cmi (see comment there).
2786		----------------------------------------------------------------------
2787		Name: Allen Leung
2788		Date: 2000/10/23 19:31:00
2789		Tag: leunga-20001023-demo-ra
2790
2791		1. Minor RA changes that improves spilling on x86 (affects Moby and C-- only)
2792		2. Test programs for the graph library updated
2793		3. Some new MLRISC demo programs added
2794
2795		----------------------------------------------------------------------
2796		Name: Matthias Blume
2797		Date: 2000/08/31 22:15:00 JST
2798		Tag: blume-20001017-errmsg
2799		Description:
2800
2801		More error message grief: Where there used to be no messages, there
2802		now were some that had bogus error regions.  Fixed.
2803
2804		----------------------------------------------------------------------
2805		Name: Matthias Blume
2806		Date: 2000/08/31 17:30:00 JST
2807		Tag: blume-20001017-v110p29p1
2808		Description:
2809
2810		I made a version 110.29.1 with new bootfiles.
2811
2812		Changes:  Modified pickler/unpickler for faster and leaner unpickling.
2813		CM documentation changes and a small bugfix in CM's error reporting.
2814
2815		----------------------------------------------------------------------
2816		Name: Lal George
2817		Date: 2000/09/27 14:42:35 EDT
2818		Tag: george-20000927-nodestatus
2819		Description:
2820
2821		Changed the type of the nodestatus, so that:
2822
2823		SPILLED(~1)             is now SPILLED
2824		SPILLED(m) where m>=0   is now MEMREG(m)
2825		SPILLED(s) where s<~1   is now SPILL_LOC(~s)
2826
2827		----------------------------------------------------------------------
2828		Name: Matthias Blume
2829		Date: 2000/09/07 14:45:00 JST
2830		Tag: blume-20000907-cmerrmsg
2831		Description:
2832
2833		Small tweak to CM to avoid getting ML syntax error messages twice.
2834
2835		----------------------------------------------------------------------
2836		Name: Matthias Blume
2837		Date: 2000/08/31 18:00:00 JST
2838		Tag: blume-20000831-cvsbootfiles
2839		Description:
2840
2841		New URL for boot files (because the 110.29 files on the BL server do
2842		now work correctly with my updated install scripts for yacc and lex).
2843
2844		----------------------------------------------------------------------
2845		Name: Matthias Blume
2846		Date: 2000/08/08 12:33:00 JST
2847		Tag: blume-20000808-manual
2848		Description:
2849
2850		Tiny update to CM manual.
2851
2852		----------------------------------------------------------------------
2853		Name: Allen Leung
2854		Date: 2000/08/7 19:31:00
2855		Tag: leunga-20000807-a-whole-bunch-of-stuff
2856
2857		Moby, C--, SSA, x86, machine descriptions etc.  Should only affect C--
2858		and Mobdy.
2859
2860		1.  x86
2861
2862		a.  Fixes to peephole module by John and Dan.
2863		b.  Assembly fix to SETcc by Allen.
2864		c.  Fix to c-call by John.
2865		d.  Fix to spilling by John.  (This one deals with the missing FSTPT case)
2866		e.  Instruction selection optimization to SETcc as suggested by John.
2867
2868		For example,
2869
2870		MV(32, x, COND(32, CMP(32, LT, a, b), LI 1, LI 0))
2871
2872		should generate:
2873
2874		MOVL a, x
2875		SUBL b, x
2876		SHRL 31, x
2877
2878		2.  IR stuff
2879
2880		A bunch of new DJ-graph related algorithms added.  These
2881		speed up SSA construction.
2882
2883		3.  SSA + Scheduling
2884
2885		Added code for SSA and scheduling to the repository
2886
2887		----------------------------------------------------------------------
2888		Name: Lal George
2889		Date: 2000/07/27 11:53:14 EDT
2890
2891		Tag: lal-20000727-linux-ppc
2892		Description:
2893
2894		Made changes to support Linux PPC.
2895		p.s. I have confirmation that the 110.29 boot files work fine.
2896
2897		----------------------------------------------------------------------
2898		Name: Matthias Blume
2899		Date: 2000/07/27 17:40:00 JST
2900		Tag: blume-20000727-scripts
2901		Description:
2902
2903		!!!! WARNING !!!!
2904		You must recompile the runtime system!
2905		!!!! WARNING !!!!
2906
2907		This is basically another round of script-enhancements:
2908
2909		1. sml, ml-build, and ml-makedepend accept options -D and -U to define
2910		and undefine CM preprocessor symbols.
2911
2912		2. ml-build avoids generating a new heap image if it finds that the
2913		existing one is still ok.  (The condition is that no ML file had to
2914		be recompiled and all ML files are found to be older that the heap
2915		file.)
2916
2917		To make this work smoothly, I also hacked the runtime system as
2918		well as SMLofNJ.SysInfo to get access to the heap image suffix
2919		(.sparc-solaris, ...) that is currently being used.
2920
2921		Moreover, the signature of CM.mk_standalone has changed.  See the
2922		CM manual.
2923
2924		3. ml-makedepend accepts additional options -n, -a, and -o.  (See the
2925		CM manual for details.)
2926
2927		4. More CM manual updates:
2928		- all of the above has been documented.
2929		- there is now a section describing the (CM-related) command line
2930		arguments that are accepted by the "sml" command
2931
2932		----------------------------------------------------------------------
2933		Name: Matthias Blume
2934		Date: 2000/07/25 16:20:00 JST
2935		Tag: blume-20000725-makedepend
2936		Description:
2937
2938		Added a script called ml-makedepend.  This can be used in makefiles
2939		for Unix' make in a way very similar to the "makedepend" command for
2940		C.
2941
2942		The script internally uses function CM.sources.
2943
2944		Synopsis:
2945
2946		ml-makedepend [-f makefile] cmfile targetname
2947
2948		The default for the makefile is "makefile" (or "Makefile" should
2949		"makefile" not exist).
2950
2951		ml-makedepend adds a cmfile/targetname-specific section to this
2952		makefile (after removing the previous version of this section).  The
2953		section contains a single dependency specification with targetname on
2954		the LHS (targetname is an arbitrary name), and a list of files derived
2955		from the cmfile on the RHS.  Some of the files on the RHS are
2956		ARCH/OPSYS-specific.  Therefore, ml-makedepend inserts references to
2957		"make" variables $(ARCH) and $(OPSYS) in place of the corresponding
2958		path names.  The makefile writer is responsible for making sure that
2959		these variables have correct at the time "make" is invoked.
2960
2961		----------------------------------------------------------------------
2962		Name: Matthias Blume
2963		Date: 2000/07/22 23:30:00 JST
2964		Tag: blume-20000722-urlupdate
2965		Description:
2966
2967		Changed BOOT and config/srcarchiveurl to point to BL server:
2968
2969		ftp://ftp.research.bell-labs.com/dist/smlnj/working/110.29/
2970
2971		----------------------------------------------------------------------
2972		Name: Matthias Blume
2973		Date: 2000/07/18 18:00:00 JST
2974		Tag: blume-20000718-Version_110_29
2975		Description:
2976
2977		1. Updated src/compiler/TopLevel/main/version.sml to version 110.29
2978
2979		2. Updated config/version to 110.29
2980
2981		3. Updated config/srcarchiveurl
2982
2983		3. New boot files!
2984		ftp://ftp.cs.princeton.edu/pub/people/blume/sml/110.29-autofetch
2985
2986		----------------------------------------------------------------------
2987		Name: Matthias Blume
2988		Date: 2000/07/11 13:58:00 JST
2989		Tag: blume-20000711-doctypo
2990		Description:
2991
2992		Fixed a few typos in CM manual.
2993
2994		----------------------------------------------------------------------
2995		Name: Allen Leung
2996		Date: 2000/06/15 00:38:00
2997		Tag: leunga-20000704-sparc-x86
2998
2999		1. x86 peephole improvement sp += k; sp -= k => nop  [from John]
3000		2. fix to x86 RET bug [found by Dan Grossman]
3001		3. sparc assembly bug fix for ticc instructions [found by Fermin]
3002
3003		Affects c-- and moby only
3004
3005		----------------------------------------------------------------------
3006		Name: Matthias Blume
3007		Date: 2000/07/04 15:26:00
3008		Tag: blume-20000704-trigger
3009		Description:
3010
3011		1. Improvements to CM manual.
3012		2. SMLofNJ.Internals.BTrace.trigger reinstated as an alternative way
3013		of getting a back-trace.  The function, when called, raises an
3014		internal exception which explicitly carries the full back-trace history,
3015		so it is unaffected by any intervening handle-raise pairs ("trivial"
3016		or not).  The interactive loop will print that history once it arrives
3017		at top level.
3018		Short of having all exceptions implicitly carry the full history, the
3019		recommended way of using this facility is:
3020		- compile your program with instrumentation "on"
3021		- run it, when it raises an exception, look at the history
3022		- if the history is "cut off" because of some handler, go and modify
3023		your program so that it explicitly calls BTrace.trigger
3024		- recompile (still instrumented), and rerun; look at the full history
3025
3026		----------------------------------------------------------------------
3027		Name: Matthias Blume
3028		Date: 2000/07/03 15:36:00 JST
3029		Tag: blume-20000702-manual
3030		Description:
3031
3032		Small corrections and updates to CM manual.
3033
3034		----------------------------------------------------------------------
3035		Name: Matthias Blume
3036		Date: 2000/06/29 16:04:00 JST
3037		Tag: blume-20000629-yacctool
3038		Description:
3039
3040		Changes:
3041
3042		1. Class "mlyacc" now takes separate arguments to pass options to
3043		generated .sml- and .sig-files independently.
3044		2. Corresponding CM manual updates.
3045		3. BTrace module now also reports call sites.  (However, for loop clusters
3046		it only shows from where the cluster was entered.)  There are associated
3047		modifications to core.sml, internals.{sig,sml}, btrace.sml, and btimp.sml.
3048
3049		----------------------------------------------------------------------
3050		Name: Matthias Blume
3051		Date: 2000/06/27 16:51:00 JST
3052		Tag: blume-20000627-noweb
3053		Description:
3054
3055		Changes:
3056
3057		1. Implemented "subdir" and "witness" options for noweb tool.
3058		This caused some slight internal changes in CM's tool implementation.
3059		2. Fixed bug in "tool plugin" mechanism.  This is essentially cleaning
3060		some remaining issues from earlier path anchor changes.
3061		3. Updated CM manual accordingly.
3062
3063		4. Changed implementation of back-tracing so that I now consider it
3064		ready for prime-time.
3065
3066		In particular, you don't have to explicitly trigger the back-trace
3067		anymore.  Instead, if you are running BTrace-instrumented code and
3068		there is an uncaught exception (regardless of whether or not it was
3069		raised in instrumented code), the top-level evalloop will print
3070		the back-trace.
3071
3072		Features:
3073
3074		- Instrumented and uninstrumented code work together seemlessly.
3075		(Of course, uninstrumented code is never mentioned in actual
3076		back-traces.)
3077
3078		- Asymptotic time- and space-complexity of instrumented code is
3079		equal to that of uninstrumented code.  (This means that
3080		tail-recursion is preserved by the instrumentation phase.)
3081
3082		- Modules whose code has been instrumented in different sessions
3083		work together without problem.
3084
3085		- There is no penalty whatsoever on uninstrumented code.
3086
3087		- There is no penalty on "raise" expressions, even in
3088		instrumented code.
3089
3090		A potential bug (or perhaps it is a feature, too):
3091
3092		A back-trace reaches no further than the outermost instrumented
3093		non-trivial "raise".  Here, a "trivial" raise is one that is the
3094		sole RHS of a "handle" rule.  Thus, back-traces reach trough
3095
3096		<exp> handle e => raise e
3097
3098		and even
3099
3100		<exp> handle Foo => raise Bar
3101
3102		and, of course, through
3103
3104		<exp> handle Foo => ...
3105
3106		if the exception was not Foo.
3107
3108		Back-traces always reach right through any un-instrumented code
3109		including any of its "handle" expressions, trivial or not.
3110
3111		To try this out, do the following:
3112
3113		- Erase all existing binfiles for your program.
3114		(You may keep binfiles for those modules where you think you
3115		definitely don't need back-tracing.)
3116		- Turn on back-trace instrumentation:
3117		SMLofNJ.Internals.BTrace.mode (SOME true);
3118		- Recompile your program.  (I.e., run "CM.make" or "use".)
3119		- You may now turn instrumentation off again (if you want):
3120		SMLofNJ.Internals.BTrace.mode (SOME false);
3121		- Run your program as usual.  If it raises an exception that
3122		reaches the interactive toplevel, then a back-trace will
3123		automatically be printed.  After that, the toplevel loop
3124		will print the exception history as usual.
3125
3126		----------------------------------------------------------------------
3127		Name: Matthias Blume
3128		Date: 2000/06/26 09:56:46 JST
3129		Tag: blume-20000626-setup
3130		Description:
3131
3132		CM: - setup-parameter to "sml" added; this can be used to run arbitrary
3133		ML code before and after compiling a file (e.g., to set compiler
3134		flags)
3135
3136		Compiler: - improved btrace API (in core.sml, internals.{sig,sml})
3137		- associated changes to btrace.sml (BTrace instrumentation pass)
3138		- cleaner implementation of btimp.sml (BTrace tracing and report
3139		module)
3140
3141		CM manual: * new path encoding documented
3142		* description of setup-parameter to "sml" added
3143
3144		The biggest user-visible change to back-tracing is that it is no
3145		longer necessary to compile all traced modules within the same
3146		session.  (This was a real limitation.)
3147
3148		----------------------------------------------------------------------
3149		Name: Matthias Blume
3150		Date: 2000/06/24 12:40:00 JST
3151		Tag: blume-20000624-startup
3152		Description:
3153
3154		Fixes startup slowdown problem.  (I was calling SrcPath.sync a _tad_
3155		bit too often -- to put it mildly. :)
3156
3157		----------------------------------------------------------------------
3158		Name: Matthias Blume
3159		Date: 2000/06/23 18:20:00 JST
3160		Tag: blume-20000623-btrace
3161		Description:
3162
3163		This updates adds a backtrace facility to aid programmers in debugging
3164		their programs.  This involves the following changes:
3165
3166		1. Module system/smlnj/init/core.sml (structure _Core) now has hooks for
3167		keeping track of the current call stack.  When programs are compiled
3168		in a special mode, the compiler will insert calls to these hooks
3169		into the user program.
3170		"Hook" means that it is possible for different implementations of
3171		back-tracing to register themselves (at different times).
3172
3173		2. compiler/MiscUtil/profile/btrace.sml implements the annotation phase
3174		as an Absyn.dec->Absyn.dec rewrite.  Normally this phase is turned off.
3175		It can be turned on using this call:
3176		SMLofNJ.Internals.BTrace.mode (SOME true);
3177		Turning it off again:
3178		SMLofNJ.Internals.BTrace.mode (SOME false);
3179		Querying the current status:
3180		SMLofNJ.Internals.BTrace.mode NONE;
3181		Annotated programs are about twice as big as normal ones, and they
3182		run a factor of 2 to 4 slower with a dummy back-trace plugin (one
3183		where all hooks do nothing).  The slowdown with a plugin that is
3184		actually useful (such as the one supplied by default) is even greater,
3185		but in the case of the default plugin it is still only an constant
3186		factor (amortized).
3187
3188		3. system/Basis/Implementation/NJ/internals.{sig,sml} have been augmented
3189		with a sub-structure BTrace for controlling back-tracing.  In particular,
3190		the above-mentioned function "mode" controls whether the annotation
3191		phase is invoked by the compiler.  Another important function is
3192		"trigger": when called it aborts the current execution and causes
3193		the top-level loop to print a full back-trace.
3194
3195		4. compiler/MiscUtil/profile/btimp.sml is the current default plugin
3196		for back-tracing.  It keeps track of the dynamic call stack and in
3197		addition to that it keeps a partial history at each "level" of that
3198		stack.  For example, if a tail-calls b, b tail-calls c, and c tail-calls
3199		d and b (at separate times, dynamically), then the report will show:
3200
3201		GOTO   d
3202		/c
3203		GOTO  \b
3204		CALL   a
3205
3206		This shows that there was an initial non-tail call of a, then a
3207		tail-call to b or c, looping behavior in a cluster of functions that
3208		consist of b and c, and then a goto from that cluster (i.e., either from
3209		b or from c) to d.
3210
3211		Note that (depending on the user program) the amount of information
3212		that the back-trace module has to keep track of at each level is bounded
3213		by a constant.  Thus, the whole implementation has the same asymptotical
3214		complexity as the original program (both in space and in time).
3215
3216		5. compiler/TopLevel/interact/evalloop.sml has been modified to
3217		handle the special exception SMLofNJ.Internals.BTrace.BTrace
3218		which is raised by the "trigger" function mentioned above.
3219
3220		Notes on usage:
3221
3222		- Annotated code works well together with unannotated code:
3223		Unannotated calls simply do not show up at all in the backtrace.
3224
3225		- It is not a good idea to let modules that were annotated during
3226		different sessions run at the same time.  This is because the compiler
3227		chooses small integers to identify individual functions, and there
3228		will be clashes if different modules were compiled in separate sessions.
3229		(Nothing will crash, and you will even be told about the clashes, but
3230		back-trace information will in general not be useful.)
3231
3232		- Back-tracing can be confused by callcc and capture.
3233
3234		- The only way of getting a back-trace right now is to explicitly
3235		invoke the "trigger" function from your user program.  Eventually, we
3236		should make every exception carry back-trace information (if
3237		available).  But since this creates more overhead at "raise"-time
3238		(similar to the current exnHistory overhead), I have not yet
3239		implemented this.  (The implementation will be rather easy.)  With
3240		exceptions carrying back-trace information, this facility will be even
3241		more useful because users don't need to modify their programs...
3242
3243		- While it is possible to compile the compiler with back-trace
3244		annotations turned on (I did it to get some confidence in
3245		correctness), you must make absolutely sure that core.sml and
3246		btimp.sml are compiled WITHOUT annotation!  (core.sml cannot actually
3247		be compiled with annotation because there is no core access yet, but
3248		if you compile btimp.sml with annotation, then the system will go into
3249		an infinite recursion and crash.)
3250		Since CM currently does not know about BTrace, the only way to turn
3251		annotations on and off for different modules of the compiler is to
3252		interrupt CMB.make, change the settings, and re-invoke it.  Of course,
3253		this is awkward and clumsy.
3254
3255		Sample sessions:
3256
3257		Standard ML of New Jersey v110.28.1 [FLINT v1.5], June 5, 2000
3258		- SMLofNJ.Internals.BTrace.mode (SOME true);
3259		[autoloading]
3260		[autoloading done]
3261		val it = false : bool
3262		- structure X = struct
3263		-     fun main n = let
3264		-         fun a (x, 0) = d x
3265		-           | a (x, n) = b (x, n - 1)
3266		-         and b (x, n) = c (x, n)
3267		-         and c (x, n) = a (x, n)
3268		-         and d x = e (x, 3)
3269		-         and e (x, 0) = f x
3270		-           | e (x, n) = e (x, n - 1)
3271		-         and f 0 = SMLofNJ.Internals.BTrace.trigger ()
3272		-           | f n = n * g (n - 1)
3273		-         and g n = a (n, 3)
3274		-     in
3275		-         f n
3276		-     end
3277		- end;
3278		structure X : sig val main : int -> int end
3279		- X.main 3;
3280		*** BACK-TRACE ***
3281		GOTO   stdIn:4.2-13.20: X.main[2].f
3282		GOTO-( stdIn:4.2-13.20: X.main[2].e
3283		GOTO   stdIn:4.2-13.20: X.main[2].d
3284		/ stdIn:4.2-13.20: X.main[2].a
3285		| stdIn:4.2-13.20: X.main[2].b
3286		GOTO-\ stdIn:4.2-13.20: X.main[2].c
3287		CALL   stdIn:4.2-13.20: X.main[2].g
3288		GOTO   stdIn:4.2-13.20: X.main[2].f
3289		GOTO-( stdIn:4.2-13.20: X.main[2].e
3290		GOTO   stdIn:4.2-13.20: X.main[2].d
3291		/ stdIn:4.2-13.20: X.main[2].a
3292		| stdIn:4.2-13.20: X.main[2].b
3293		GOTO-\ stdIn:4.2-13.20: X.main[2].c
3294		CALL   stdIn:4.2-13.20: X.main[2].g
3295		GOTO   stdIn:4.2-13.20: X.main[2].f
3296		GOTO-( stdIn:4.2-13.20: X.main[2].e
3297		GOTO   stdIn:4.2-13.20: X.main[2].d
3298		/ stdIn:4.2-13.20: X.main[2].a
3299		| stdIn:4.2-13.20: X.main[2].b
3300		GOTO-\ stdIn:4.2-13.20: X.main[2].c
3301		CALL   stdIn:4.2-13.20: X.main[2].g
3302		GOTO   stdIn:4.2-13.20: X.main[2].f
3303		CALL   stdIn:2.15-17.4: X.main[2]
3304		-
3305
3306		(Note that because of a FLINt bug the above code currently does not
3307		compile without BTrace turned on.)
3308
3309		Here is another example, using my modified Tiger compiler:
3310
3311		Standard ML of New Jersey v110.28.1 [FLINT v1.5], June 5, 2000
3312		- SMLofNJ.Internals.BTrace.mode (SOME true);
3313		[autoloading]
3314		[autoloading done]
3315		val it = false : bool
3316		- CM.make "sources.cm";
3317		[autoloading]
3318		...
3319		[autoloading done]
3320		[scanning sources.cm]
3321		[parsing (sources.cm):parse.sml]
3322		[creating directory CM/SKEL ...]
3323		[parsing (sources.cm):tiger.lex.sml]
3324		...
3325		[wrote CM/sparc-unix/semant.sml]
3326		[compiling (sources.cm):main.sml]
3327		[wrote CM/sparc-unix/main.sml]
3328		[New bindings added.]
3329		val it = true : bool
3330		- Main.compile ("../testcases/merge.tig", "foo.out");
3331		*** BACK-TRACE ***
3332		CALL   lib/semant.sml:99.2-396.21: SemantFun[2].transExp.trvar
3333		CALL   lib/semant.sml:99.2-396.21: SemantFun[2].transExp.trexp
3334		CALL   lib/semant.sml:289.3-295.22: SemantFun[2].transExp.trexp.check[2]
3335		GOTO   lib/semant.sml:289.3-295.22: SemantFun[2].transExp.trexp.check[2]
3336		CALL   lib/semant.sml:99.2-396.21: SemantFun[2].transExp.trexp
3337		CALL   lib/semant.sml:99.2-396.21: SemantFun[2].transExp.trexp
3338		CALL   lib/semant.sml:488.3-505.6: SemantFun[2].transDec.trdec[2].transBody[2]
3339		/ lib/semant.sml:411.65-543.8: SemantFun[2].transDec
3340		CALL-\ lib/semant.sml:413.2-540.9: SemantFun[2].transDec.trdec[2]
3341		CALL   lib/semant.sml:99.2-396.21: SemantFun[2].transExp.trexp
3342		CALL   lib/semant.sml:8.52-558.4: SemantFun[2].transProg[2]
3343		CALL   main.sml:1.18-118.4: Main.compile[2]
3344		-
3345
3346		----------------------------------------------------------------------
3347		Name: Matthias Blumen
3348		Date: 2000/06/21 18:00:00 JST
3349		Tag: blume-20000621-manual
3350		Description:
3351
3352		CM manual update: Path environments documented.
3353
3354		----------------------------------------------------------------------
3355		Name: Matthias Blume
3356		Date: 2000/06/19 13:40:00
3357		Tag: blume-20000619-manual
3358		Description:
3359
3360		CM manual and system/README update.  This only covers the fact that
3361		there are no more implicit anchors.  (Path environments and the "bind"
3362		option to "cm" have yet to be documented.)
3363
3364		----------------------------------------------------------------------
3365		Name: Matthias Blume
3366		Date: 2000/06/19 11:05:00 JST
3367		Tag: blume-20000619-chdir-bugfix
3368		Description:
3369
3370		Fixed a bug in new SrcPath module that sometimes led to a bad chDir call.
3371
3372		----------------------------------------------------------------------
3373		Name: Matthias Blume
3374		Date: 2000/06/18 22:00:10 JST
3375		Tag: blume-20000618-implicit-anchors-really-gone
3376		Description:
3377
3378		I updates the previous HISTORY entry where I forgot to mention that
3379		implicit anchors are no longer with us.
3380
3381		The current update also gets rid of the (now useless) controller
3382		CM.Control.implicit_anchors.
3383
3384		----------------------------------------------------------------------
3385		Name: Matthias Blume
3386		Date: 2000/06/16 17:30:00 JST
3387		Tag: blume-20000616-anchorenv
3388		Description:
3389
3390		This patch implements the long anticipated (just kidding :) "anchor
3391		environment" mechanism.  In the course of doing this, I also
3392		re-implemented CM's internal "SrcPath" module from scratch.  The new
3393		one should be more robust in certain boundary cases.  In any case, it
3394		is a lot cleaner than its predecessor (IMHO).
3395
3396		This time, although there is yet another boot file format change, I
3397		kept the unpickler backward-compatible.  As a result, no new bootfiles
3398		are necessary and bootstrapping is straightforward.  (You cannot read
3399		new bootfiles into an old system, but the other way around is no
3400		problem.)
3401
3402		Visible changes:
3403
3404		** 0. Implicit path anchors (without the leading $-symbol) are no
3405		longer recognized at all. This means that such path names are not
3406		illegal either.  For example, the name basis.cm simply refers to a
3407		local file called "basis.cm" (i.e, the name is an ordinary path
3408		relative to .cm-files directory).  Or, to put it differently, only
3409		names that start with $ are anchored paths.
3410
3411		** 1. The $<singlearc> abbreviation for $/<singlearc> has finally
3412		vanished.
3413
3414		John (Reppy) had critizised this as soon as I originally proposed and
3415		implemented it, but at that time I did not really deeply believe
3416		him. :) Now I came full-circle because I need the $<singlearc> syntax
3417		in another place where it cannot be seen as an abbreviation for
3418		$/<singlearc>.  To avoid the confusion, $<singlearc> now means what it
3419		seems to mean (i.e., it "expands" into the corresponding anchor
3420		value).
3421
3422		However, when paths are used as members in CM description files, it
3423		continues to be true that there must be at least another arc after the
3424		anchor.  This is now enforced separately during semantic analysis
3425		(i.e., from a lexical/syntactical point of view, the notation is ok.)
3426
3427		** 2. The "cm" class now accepts an option "bind".  The option's value
3428		is a sub-option list of precisely two items -- one labeled "anchor"
3429		and the other one labeled "value".  As you might expect, "anchor" is
3430		used to specify an anchor name to be bound, and "value" specifies what
3431		the anchor is being bound to.
3432
3433		The value must be a directory name and can be given in either standard
3434		syntax (including the possibility that it is itself an anchored path)
3435		or native syntax.
3436
3437		Examples:
3438
3439		foo.cm (bind:(anchor:bar value:$mystuff/bar))
3440		lib.cm (bind:(anchor:a value:"H:\\x\\y\\z"))  (* only works under windows *)
3441
3442		and so on.
3443
3444		The meaning of this is that the .cm-file will be processed with an
3445		augmented anchor environment where the given anchor(s) is/are bound to
3446		the given values(s).
3447
3448		The rationale for having this feature is this: Suppose you are trying
3449		to use two different (already stable) libraries a.cm and b.cm (that
3450		you perhaps didn't write yourself).  Further, suppose each of these
3451		two libraries internally uses its own auxiliary library $aux/lib.cm.
3452		Normally you would now have a problem because the anchor "lib" can not
3453		be bound to more than one value globally.  Therefore, the project that
3454		uses both a.cm and b.cm must locally redirect the anchor to some other
3455		place:
3456
3457		a.cm (bind:(anchor:lib value:/usr/lib/smlnj/a-stuff))
3458		b.cm (bind:(anchor:lib value:/usr/lib/smlnj/b-stuff))
3459
3460		This hard-wires $lib/aux.cm to /usr/lib/smlnj/a-stuff/aux.cm or
3461		/usr/lib/smlnj/b-stuff/aux.cm, respectively.
3462
3463		Hard-wiring path names is a bit inflexible (and CM will verbosely warn
3464		you when you do so at the time of CM.stabilize).  Therefore, you can
3465		also use an anchored path as the value:
3466
3467		a.cm (bind:(anchor:lib value:$a-lib))
3468		b.cm (bind:(anchor:lib value:$b-lib))
3469
3470		Now you can globally configure (using the usual CM.Anchor.anchor or
3471		pathconfig machinery) bindings for "a-lib" and "b-lib".  Since "lib"
3472		itself is always locally bound, setting it globally is no longer
3473		meaningful or necessary (but it does not hurt either).  In fact, "lib"
3474		can still be used as a global anchor for separate purposes.  As a
3475		matter of fact, one can locally define "lib" in terms of a global
3476		"lib":
3477
3478		a.cm (bind:(anchor:lib value:$lib/a))
3479		b.cm (bind:(anchor:lib value:$lib/b))
3480
3481		** 3: The encoding of path names has changed.  This affects the way
3482		path names are shown in CM's progress report and also the internal
3483		protocol encoding used for parallel make.
3484
3485		The encoding now uses one or more ':'-separated segments.  Each
3486		segments corresponds to a file that has been specified relative to the
3487		file given by its preceding segment.  The first segment is either
3488		relative to the CWD, absolute, or anchored.  Each segment itself is
3489		basically a Unix pathname; all segments but the first are relative.
3490
3491		Example:
3492
3493		$foo/bar/baz.cm:a/b/c.sml
3494
3495		This path denotes the file bar/a/b/c.sml relative to the directory
3496		denoted by anchor "foo".  Notice that the encoding also includes
3497		baz.cm which is the .cm-file that listed a/b/c.sml.  As usual, such
3498		paths are resolved relative to the .cm-files directory, so baz.cm must
3499		be ignored to get the "real" pathname.
3500
3501		To make this fact more obvious, CM puts the names of such "virtual
3502		arcs" into parentheses when they appear in progress reports. (No
3503		parentheses will appear in the internal protocol encoding.)  Thus,
3504		what you really see is:
3505
3506		$foo/bar/(baz.cm):a/b/c.sml
3507
3508		I find this notation to be much more informative than before.
3509
3510		Another new feature of the encoding is that special characters
3511		including parentheses, colons, (back)slashes, and white space are
3512		written as \ddd (where ddd is the decimal encoding of the character).
3513
3514		*** The CM manual still needs to be updated.
3515
3516		----------------------------------------------------------------------
3517		Name: Allen Leung
3518		Date: 2000/06/15 00:38:00
3519		Tag: leunga-20000615-x86-peephole
3520
3521		x86 Peephole fix by Fermin.  Affects c-- and moby only.
3522
3523		----------------------------------------------------------------------
3524		Name: Matthias Blume
3525		Date: 2000/06/12 11:40:00
3526		Tag: blume-20000612-parmakefix
3527		Description:
3528
3529		More cleanup after changing the file naming scheme: This time I
3530		repaired the parallel make mechanism for CMB.make which I broke earlier.
3531
3532		----------------------------------------------------------------------
3533		Name: Allen Leung
3534		Date: 2000/06/09 01:25:00
3535		Tag: leunga-20000609-various
3536
3537		None of these things should affect normal SML/NJ operations
3538
3539		1. Peephole improvements provided by Fermin (c--)
3540		2. New annotation DEFUSE for adding extra dependence (moby)
3541		3. New X86 LOCK instructions (moby)
3542		4. New machine description language for reservation tables (scheduling)
3543		5. Fixes to various optimization/analysis modules (branch chaining, dominator
3544		trees etc.)
3545		6. I've changed the CM files so that they can work with versions
3546		110.0.6, 110.25 and 110.28
3547
3548		----------------------------------------------------------------------
3549		Name: Matthias Blume
3550		Date: 2000/06/09 12:40:00
3551		Tag: blume-20000609-log
3552		Description:
3553
3554		- Removed all(?) remaining RCS Log entries from sources.
3555
3556		- Fixed bug in ml-yacc and ml-lex sources (use explicit anchors for
3557		anchored paths).
3558
3559		----------------------------------------------------------------------
3560		Name: Matthias Blume
3561		Date: 2000/06/07 17:00:00 JST
3562		Tag: blume-20000607-no-implicit-anchors
3563		Description:
3564
3565		1. This update changes the default setting for
3566		CM.Control.implicit_anchors from true to false.  This means that
3567		implicit anchors are no longer permitted by default.  I also tried to
3568		make sure that nothing else still relies on implicit anchors.
3569		(This is the next step on the schedule towards a CM that does not even
3570		have the notion of implicit anchors anymore.)
3571
3572		2. More CM manual updates.
3573
3574		3. I managed to track down and fix the pickling bug I mentioned last
3575		time.  Because of the previously existing workaround, this entails no
3576		immediate practical changes.
3577
3578		----------------------------------------------------------------------
3579		Name: Matthias Blume
3580		Date: 2000/06/06 11:15:00 JST
3581		Tag: blume-20000606-lazierpickle
3582		Description:
3583
3584		!!!! NEW BOOT FILES !!!!
3585
3586		* The main purpose of this update is to make library pickles lazier in
3587		order to reduce the initial space penalty for autoloading a library.
3588		As a result, it is now possible to have $smlnj/compiler.cm
3589		pre-registered.  This should take care of the many complaints or
3590		inquiries about missing structure Compiler.  This required changes to
3591		CM's internal data structures and small tweaks to some algorithms.
3592
3593		As a neat additional effect, it is no longer necessary (for the sake
3594		of lean heap image files) to distinguish between a "minimal" CM and a
3595		"full" CM.  Now, there is only one CM (i.e., the "full" version:
3596		$smlnj/cm.cm aka $smlnj/cm/full.cm), and it is always available at the
3597		interactive top level. ($smlnj/cm/minimal.cm is gone.)
3598
3599		To make the life of compiler-hackers easier, "makeml" now also
3600		pre-registers $smlnj/cmb.cm (aka $smlnj/cmb/current.cm).  In other
3601		words, after you bootstrap a new sml for the first time, you will not
3602		have to autoload $smlnj/cmb.cm again afterwards.  (The first time
3603		around you will still have to do it, though.)
3604
3605		* A second change consists of major updates to the CM manual.  There
3606		are now several appendices with summary information and also a full
3607		specification of the CM description file syntax.
3608
3609		* In directory src/system I added the script "allcross".  This script
3610		invokes sml and cross-compiles the compiler for all supported
3611		architectures.  (Useful when providing a new set of boot files.)
3612
3613		* There seems to be a latent bug in my "lazy pickles" mechanism.  I
3614		added a small