Diff of /sml/trunk/HISTORY

Parent Directory | Revision Log | Patch

	revision 815, Fri May 4 05:09:10 2001 UTC	revision 1158, Fri Mar 22 01:21:25 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: 2002/03/21 20:30:00 EST
18		Tag: leunga-20020321-cfg
19		Description:
20
21		Fixed a potential problem in cfg edge splitting.
22
23		----------------------------------------------------------------------
24		Name: Allen Leung
25		Date: 2002/03/21 17:15:00 EST
26		Tag: leunga-20020321-x86-fp-cfg
27		Description:
28
29		1. Recoded the buggy parts of x86-fp.
30
31		a. All the block reordering code has been removed.
32		We now depend on the block placement phases to do this work.
33
34		b. Critical edge splitting code has been simplified and moved into the
35		CFG modules, as where they belong.
36
37		Both of these were quite buggy and complex.  The code is now much, much
38		simpler.
39
40		2. X86 backend.
41
42		a. Added instructions for 64-bit support.  Instruction selection for
43		64-bit has not been committed, however, since that
44		requires changes to MLTREE which haven't been approved by
45		Lal and John.
46
47		b. Added support for FUCOMI and FUCOMIP when generating code for
48		PentiumPro and above.  We only generate these instructions in
49		the fast-fp mode.
50
51		c. Added cases for JP and JNP in X86FreqProps.
52
53		3. CFG
54
55		CFG now has a bunch of methods for edge splitting and merging.
56
57		4. Machine description.
58
59		John's simplification of MLTREE_BASIS.fcond broke a few machine
60		description things:
61
62		rtl-build.{sig,sml} and hppa.mdl fixed.
63
64		NOTE: the machine description stuff in the repository is still broken.
65		Again, I can't put my fixes in because that involves
66		changes to MLTREE.
67
68		----------------------------------------------------------------------
69		Name: Matthias Blume
70		Date: 2002/03/20 15:55:00 EST
71		Tag: blume-20020320-kmp
72		Description:
73
74		Implemented Knuth-Morris-Pratt string matching in PreString and used
75		it for String.isSubstring, Substring.isSubstring, and
76		Substring.position.
77
78		(Might need some stress-testing.  Simple examples worked fine.)
79
80		----------------------------------------------------------------------
81		Name: Matthias Blume
82		Date: 2002/03/19 16:37:00 EST
83		Tag: blume-20020319-witnesses
84		Description:
85
86		Added a structure C.W and functions convert/Ptr.convert to ml-nlffi-lib.
87
88		This implements a generic mechanism for changing constness qualifiers
89		anywhere within big C types without resorting to outright "casts".
90		(So far, functions such as C.rw/C.ro or C.Ptr.rw/C.Ptr.ro only let you
91		modify the constness at the outermost level.)
92		The implementation of "convert" is based on the idea of "witness"
93		values -- values that are not used by the operation but whose types
94		"testify" to their applicability.  On the implementation side, "convert"
95		is simply a projection (returning its second curried argument).  With
96		cross-module inlining, it should not result in any machine code being
97		generated.
98
99		----------------------------------------------------------------------
100		Name: Matthias Blume
101		Date: 2002/03/15 16:40:00 EST
102		Tag: blume-20020315-basis
103		Description:
104
105		Provided (preliminary?) implementations for
106
107		{String,Substring}.{concatWith,isSuffix,isSubstring}
108
109		and
110
111		Substring.full
112
113		Those are in the Basis spec but they were missing in SML/NJ.
114
115		----------------------------------------------------------------------
116		Name: Matthias Blume
117		Date: 2002/03/14 21:30:00 EST
118		Tag: blume-20020314-controls
119		Description:
120
121		Controls:
122		---------
123
124		1. Factored out the recently-added Controls : CONTROLS stuff and put
125		it into its own library $/controls-lib.cm.  The source tree for
126		this is under src/smlnj-lib/Controls.
127
128		2. Changed the names of types and functions in this interface, so they
129		make a bit more "sense":
130
131		module -> registry
132		'a registry -> 'a group
133
134		3. The interface now deals in ref cells only.  The getter/setter interface
135		is (mostly) gone.
136
137		4. Added a function that lets one register an already-existing ref cell.
138
139		5. Made the corresponding modifications to the rest of the code so that
140		everything compiles again.
141
142		6. Changed the implementation of Controls.MLRISC back to something closer
143		to the original.  In particular, this module (and therefore MLRISC)
144		does not depend on Controls.  There now is some link-time code in
145		int-sys.sml that registers the MLRISC controls with the Controls
146		module.
147
148		CM:
149		---
150
151		* One can now specify the lambda-split aggressiveness in init.cmi.
152
153		----------------------------------------------------------------------
154		Name: Allen Leung
155		Date: 2002/03/13 17:30:00 EST
156		Tag: leunga-20020313-x86-fp-unary
157		Description:
158
159		Bug fix for:
160
161		> leunga@weaselbane:~/Yale/tmp/sml-dist{21} bin/sml
162		> Standard ML of New Jersey v110.39.1 [FLINT v1.5], March 08, 2002
163		> - fun f(x,(y,z)) = Real.~ y;
164		> [autoloading]
165		> [autoloading done]
166		>       fchsl   (%eax), 184(%esp)
167		> Error: MLRisc bug: X86MCEmitter.emitInstr
168		>
169		> uncaught exception Error
170		>   raised at: ../MLRISC/control/mlriscErrormsg.sml:16.14-16.19
171
172		The problem was that the code generator did not generate any fp registers
173		in this case, and the ra didn't know that it needed to run the X86FP phase to
174		translate the pseudo fp instruction.   This only happened with unary fp
175		operators in certain situations.
176
177		----------------------------------------------------------------------
178		Name: Matthias Blume
179		Date: 2002/03/13 14:00:00 EST
180		Tag: blume-20020313-overload-etc
181		Description:
182
183		1. Added _overload as a synonym for overload for backward compatibility.
184		(Control.overloadKW must be true for either version to be accepted.)
185
186		2. Fixed bug in install script that caused more things to be installed
187		than what was requested in config/targets.
188
189		3. Made CM aware of the (_)overload construct so that autoloading
190		works.
191
192		----------------------------------------------------------------------
193		Name: Matthias Blume
194		Date: 2002/03/12 22:03:00 EST
195		Tag: blume-20020312-url
196		Description:
197
198		Forgot to update BOOT and srcarchiveurl.
199
200		----------------------------------------------------------------------
201		Name: Matthias Blume
202		Date: 2002/03/12 17:30:00 EST
203		Tag: blume-20020312-version110392
204		Description:
205
206		Yet another version number bump (because of small changes to the
207		binfile format).  Version number is now 110.39.2.  NEW BOOTFILES!
208
209		Changes:
210
211		The new pid generation scheme described a few weeks ago was overly
212		complicated.  I implemented a new mechanism that is simpler and
213		provides a bit more "stability":  Once CM has seen a compilation
214		unit, it keeps its identity constant (as long as you do not delete
215		those crucial CM/GUID/* files).  This means that when you change
216		an interface, compile, then go back to the old interface, and
217		compile again, you arrive at the original pid.
218
219		There now also is a mechanism that instructs CM to use the plain
220		environment hash as a module's pid (effectively making its GUID
221		the empty string).  For this, "noguid" must be specified as an
222		option to the .sml file in question within its .cm file.
223		This is most useful for code that is being generated by tools such
224		as ml-nlffigen (because during development programmers tend to
225		erase the tool's entire output directory tree including CM's cached
226		GUIDs).  "noguid" is somewhat dangerous (since it can be used to locally
227		revert to the old, broken behavior of SML/NJ, but in specific cases
228		where there is no danger of interface confusion, its use is ok
229		(I think).
230
231		ml-nlffigen by default generates "noguid" annotations.  They can be
232		turned off by specifying -guid in its command line.
233
234		----------------------------------------------------------------------
235		Name: Lal George
236		Date: 2002/03/12 12 14:42:36 EST
237		Tag: george-20020312-frequency-computation
238		Description:
239
240		Integrated jump chaining and static block frequency into the
241		compiler. More details and numbers later.
242
243		----------------------------------------------------------------------
244		Name: Lal George
245		Date: 2002/03/11 11 22:38:53 EST
246		Tag: george-20020311-jump-chain-elim
247		Description:
248
249		Tested the jump chain elimination on all architectures (except the
250		hppa).  This is on by default right now and is profitable for the
251		alpha and x86, however, it may not be profitable for the sparc and ppc
252		when compiling the compiler.
253
254		The gc test will typically jump to a label at the end of the cluster,
255		where there is another jump to an external cluster containing the actual
256		code to invoke gc. This is to allow factoring of common gc invocation
257		sequences. That is to say, we generate:
258
259		f:
260		testgc
261		ja   L1      % jump if above to L1
262
263		L1:
264		jmp L2
265
266
267		After jump chain elimination the 'ja L1' instructions is converted to
268		'ja L2'. On the sparc and ppc, many of the 'ja L2' instructions may end
269		up being implemented in their long form (if L2 is far away) using:
270
271		jbe     L3      % jump if below or equal to L3
272		jmp     L2
273		L3:
274		...
275
276
277		For large compilation units L2  may be far away.
278
279
280		----------------------------------------------------------------------
281		Name: Matthias Blume
282		Date: 2002/03/11 13:30:00 EST
283		Tag: blume-20020311-mltreeeval
284		Description:
285
286		A functor parameter was missing.
287
288		----------------------------------------------------------------------
289		Name: Allen Leung
290		Date: 2002/03/11 10:30:00 EST
291		Tag: leunga-20020311-runtime-string0
292		Description:
293
294		The representation of the empty string now points to a
295		legal null terminated C string instead of unit.  It is now possible
296		to convert an ML string into C string with InlineT.CharVector.getData.
297		This compiles into one single machine instruction.
298
299		----------------------------------------------------------------------
300		Name: Allen Leung
301		Date: 2002/03/10 23:55:00 EST
302		Tag: leunga-20020310-x86-call
303		Description:
304
305		Added machine generation for CALL instruction (relative displacement mode)
306
307		----------------------------------------------------------------------
308		Name: Matthias Blume
309		Date: 2002/03/08 16:05:00
310		Tag: blume-20020308-entrypoints
311		Description:
312
313		Version number bumped to 110.39.1.  NEW BOOTFILES!
314
315		Entrypoints: non-zero offset into a code object where execution should begin.
316
317		- Added the notion of an entrypoint to CodeObj.
318		- Added reading/writing of entrypoint info to Binfile.
319		- Made runtime system bootloader aware of entrypoints.
320		- Use the address of the label of the first function given to mlriscGen
321		as the entrypoint.  This address is currently always 0, but it will
322		not be 0 once we turn on block placement.
323		- Removed the linkage cluster code (which was The Other Way(tm) of dealing
324		with entry points) from mlriscGen.
325
326		----------------------------------------------------------------------
327		Name: Allen Leung
328		Date: 2002/03/07 20:45:00 EST
329		Tag: leunga-20020307-x86-cmov
330		Description:
331
332		Bug fixes for CMOVcc on x86.
333
334		1. Added machine code generation for CMOVcc
335		2. CMOVcc is now generated in preference over SETcc on PentiumPro or above.
336		3. CMOVcc cannot have an immediate operand as argument.
337
338		----------------------------------------------------------------------
339		Name: Matthias Blume
340		Date: 2002/03/07 16:15:00 EST
341		Tag: blume-20020307-controls
342		Description:
343
344		This is a very large but mostly boring patch which makes (almost)
345		every tuneable compiler knob (i.e., pretty much everything under
346		Control.* plus a few other things) configurable via both the command
347		line and environment variables in the style CM did its configuration
348		until now.
349
350		Try starting sml with '-h' (or, if you are brave, '-H')
351
352		To this end, I added a structure Controls : CONTROLS to smlnj-lib.cm which
353		implements the underlying generic mechanism.
354
355		The interface to some of the existing such facilities has changed somewhat.
356		For example, the MLRiscControl module now provides mkFoo instead of getFoo.
357		(The getFoo interface is still there for backward-compatibility, but its
358		use is deprecated.)
359
360		The ml-build script passes -Cxxx=yyy command-line arguments through so
361		that one can now twiddle the compiler settings when using this "batch"
362		compiler.
363
364		TODO items:
365
366		We should go through and throw out all controls that are no longer
367		connected to anything.  Moreover, we should go through and provide
368		meaningful (and correct!) documentation strings for those controls
369		that still are connected.
370
371		Currently, multiple calls to Controls.new are accepted (only the first
372		has any effect).  Eventually we should make sure that every control
373		is being made (via Controls.new) exactly once.  Future access can then
374		be done using Controls.acc.
375
376		Finally, it would probably be a good idea to use the getter-setter
377		interface to controls rather than ref cells.  For the time being, both
378		styles are provided by the Controls module, but getter-setter pairs are
379		better if thread-safety is of any concern because they can be wrapped.
380
381		*****************************************
382
383		One bug fix: The function blockPlacement in three of the MLRISC
384		backpatch files used to be hard-wired to one of two possibilities at
385		link time (according to the value of the placementFlag).  But (I
386		think) it should rather sense the flag every time.
387
388		*****************************************
389
390		Other assorted changes (by other people who did not supply a HISTORY entry):
391
392		1. the cross-module inliner now works much better (Monnier)
393		2. representation of weights, frequencies, and probabilities in MLRISC
394		changed in preparation of using those for weighted block placement
395		(Reppy, George)
396
397		----------------------------------------------------------------------
398		Name: Lal George
399		Date: 2002/03/07 14:44:24 EST 2002
400		Tag: george-20020307-weighted-block-placement
401
402		Tested the weighted block placement optimization on all architectures
403		(except the hppa) using AMPL to generate the block and edge frequencies.
404		Changes were required in the machine properties to correctly
405		categorize trap instructions. There is an MLRISC flag
406		"weighted-block-placement" that can be used to enable weighted block
407		placement, but this will be ineffective without block/edge
408		frequencies (coming soon).
409
410
411		----------------------------------------------------------------------
412		Name: Lal George
413		Date: 2002/03/05 17:24:48 EST
414		Tag: george-20020305-linkage-cluster
415
416		In order to support the block placement optimization, a new cluster
417		is generated as the very first cluster (called the linkage cluster).
418		It contains a single jump to the 'real' entry point for the compilation
419		unit. Block placement has no effect on the linkage cluster itself, but
420		all the other clusters  have full freedom in the manner in which they
421		reorder blocks or functions.
422
423		On the x86 the typical linkage code that is generated is:
424		----------------------
425		.align 2
426		L0:
427		addl    $L1-L0, 72(%esp)
428		jmp     L1
429
430
431		.align  2
432		L1:
433		----------------------
434
435		72(%esp) is the memory location for the stdlink register. This
436		must contain the address of the CPS function being called. In the
437		above example, it contains the address of  L0; before
438		calling L1 (the real entry point for the compilation unit), it
439		must contain the address for L1, and hence
440
441		addl $L1-L0, 72(%esp)
442
443		I have tested this on all architectures except the hppa.The increase
444		in code size is of course negligible
445
446		----------------------------------------------------------------------
447		Name: Allen Leung
448		Date: 2002/03/03 13:20:00 EST
449		Tag: leunga-20020303-mlrisc-tools
450
451		Added #[ ... ] expressions to mlrisc tools
452
453		----------------------------------------------------------------------
454		Name: Matthias Blume
455		Date: 2002/02/27 12:29:00 EST
456		Tag: blume-20020227-cdebug
457		Description:
458
459		- made types in structure C and C_Debug to be equal
460		- got rid of code duplication (c-int.sml vs. c-int-debug.sml)
461		- there no longer is a C_Int_Debug (C_Debug is directly derived from C)
462
463		----------------------------------------------------------------------
464		Name: Matthias Blume
465		Date: 2002/02/26 12:00:00 EST
466		Tag: blume-20020226-ffi
467		Description:
468
469		1. Fixed a minor bug in CM's "noweb" tool:
470		If numbering is turned off, then truly don't number (i.e., do not
471		supply the -L option to noweb).  The previous behavior was to supply
472		-L'' -- which caused noweb to use the "default" line numbering scheme.
473		Thanks to Chris Richards for pointing this out (and supplying the fix).
474
475		2. Once again, I reworked some aspects of the FFI:
476
477		A. The incomplete/complete type business:
478
479		- Signatures POINTER_TO_INCOMPLETE_TYPE and accompanying functors are
480		gone!
481		- ML types representing an incomplete type are now *equal* to
482		ML types representing their corresponding complete types (just like
483		in C).  This is still safe because ml-nlffigen will not generate
484		RTTI for incomplete types, nor will it generate functions that
485		require access to such RTTI.   But when ML code generated from both
486		incomplete and complete versions of the C type meet, the ML types
487		are trivially interoperable.
488
489		NOTE:  These changes restore the full generality of the translation
490		(which was previously lost when I eliminated functorization)!
491
492		B. Enum types:
493
494		- Structure C now has a type constructor "enum" that is similar to
495		how the "su" constructor works.  However, "enum" is not a phantom
496		type because each "T enum" has values (and is isomorphic to
497		MLRep.Signed.int).
498		- There are generic access operations for enum objects (using
499		MLRep.Signed.int).
500		- ml-nlffigen will generate a structure E_foo for each "enum foo".
501		* The structure contains the definition of type "mlrep" (the ML-side
502		representation type of the enum).  Normally, mlrep is the same
503		as "MLRep.Signed.int", but if ml-nlffigen was invoked with "-ec",
504		then mlrep will be defined as a datatype -- thus facilitating
505		pattern matching on mlrep values.
506		("-ec" will be suppressed if there are duplicate values in an
507		enumeration.)
508		* Constructors ("-ec") or values (no "-ec") e_xxx of type mlrep
509		will be generated for each C enum constant xxx.
510		* Conversion functions m2i and i2m convert between mlrep and
511		MLRep.Signed.int.  (Without "-ec", these functions are identities.)
512		* Coversion functions c and ml convert between mlrep and "tag enum".
513		* Access functions (get/set) fetch and store mlrep values.
514		- By default (unless ml-nlffigen was invoked with "-nocollect"), unnamed
515		enumerations are merged into one single enumeration represented by
516		structure E_'.
517
518		----------------------------------------------------------------------
519		Name: Allen Leung
520		Date: 2002/02/25 04:45:00 EST
521		Tag: leunga-20020225-cps-spill
522
523		This is a new implementation of the CPS spill phase.
524		The new phase is in the new file compiler/CodeGen/cpscompile/spill-new.sml
525		In case of problems, replace it with the old file spill.sml
526
527		The current compiler runs into some serious performance problems when
528		constructing a large record.  This can happen when we try to compile a
529		structure with many items.  Even a very simple structure like the following
530		makes the compiler slow down.
531
532		structure Foo = struct
533		val x_1 = 0w1 : Word32.int
534		val x_2 = 0w2 : Word32.int
535		val x_3 = 0w3 : Word32.int
536		...
537		val x_N = 0wN : Word32.int
538		end
539
540		The following table shows the compile time, from N=1000 to N=4000,
541		with the old compiler:
542
543		N
544		1000   CPS 100 spill                           0.04u  0.00s  0.00g
545		MLRISC ra                               0.06u  0.00s  0.05g
546		(spills = 0 reloads = 0)
547		TOTAL                                   0.63u  0.07s  0.21g
548
549		1100   CPS 100 spill                           8.25u  0.32s  0.64g
550		MLRISC ra                               5.68u  0.59s  3.93g
551		(spills = 0 reloads = 0)
552		TOTAL                                   14.71u  0.99s  4.81g
553
554		1500   CPS 100 spill                           58.55u  2.34s  1.74g
555		MLRISC ra                               5.54u  0.65s  3.91g
556		(spills = 543 reloads = 1082)
557		TOTAL                                   65.40u  3.13s  6.00g
558
559		2000   CPS 100 spill                           126.69u  4.84s  3.08g
560		MLRISC ra                               0.80u  0.10s  0.55g
561		(spills = 42 reloads = 84)
562		TOTAL                                   129.42u  5.10s  4.13g
563
564		3000   CPS 100 spill                           675.59u  19.03s  11.64g
565		MLRISC ra                               2.69u  0.27s  1.38g
566		(spills = 62 reloads = 124)
567		TOTAL                                   682.48u  19.61s  13.99g
568
569		4000   CPS 100 spill                           2362.82u  56.28s  43.60g
570		MLRISC ra                               4.96u  0.27s  2.72g
571		(spills = 85 reloads = 170)
572		TOTAL                                   2375.26u  57.21s  48.00g
573
574		As you can see the old cps spill module suffers from some serious
575		performance problem.  But since I cannot decipher the old code fully,
576		instead of patching the problems up, I'm reimplementing it
577		with a different algorithm.  The new code is more modular,
578		smaller when compiled, and substantially faster
579		(O(n log n) time and O(n) space).  Timing of the new spill module:
580
581		4000  CPS 100 spill                           0.02u  0.00s  0.00g
582		MLRISC ra                               0.25u  0.02s  0.15g
583		(spills=1 reloads=3)
584		TOTAL                                   7.74u  0.34s  1.62g
585
586		Implementation details:
587
588		As far as I can tell, the purpose of the CPS spill module is to make sure the
589		number of live variables at any program point (the bandwidth)
590		does not exceed a certain limit, which is determined by the
591		size of the spill area.
592
593		When the bandwidth is too large, we decrease the register pressure by
594		packing live variables into spill records.  How we achieve this is
595		completely different than what we did in the old code.
596
597		First, there is something about the MLRiscGen code generator
598		that we should be aware of:
599
600		o MLRiscGen performs code motion!
601
602		In particular, it will move floating point computations and
603		address computations involving only the heap pointer to
604		their use sites (if there is only a single use).
605		What this means is that if we have a CPS record construction
606		statement
607
608		RECORD(k,vl,w,e)
609
610		we should never count the new record address w as live if w
611		has only one use (which is often the case).
612
613		We should do something similar to floating point, but the transformation
614		there is much more complex, so I won't deal with that.
615
616		Secondly, there are now two new cps primops at our disposal:
617
618		1. rawrecord of record_kind option
619		This pure operator allocates some uninitialized storage from the heap.
620		There are two forms:
621
622		rawrecord NONE [INT n]  allocates a tagless record of length n
623		rawrecord (SOME rk) [INT n] allocates a tagged record of length n
624		and initializes the tag.
625
626		2. rawupdate of cty
627		rawupdate cty (v,i,x)
628		Assigns to x to the ith component of record v.
629		The storelist is not updated.
630
631		We use these new primops for both spilling and increment record construction.
632
633		1. Spilling.
634
635		This is implemented with a linear scan algorithm (but generalized
636		to trees).  The algorithm will create a single spill record at the
637		beginning of the cps function and use rawupdate to spill to it,
638		and SELECT or SELp to reload from it.  So both spills and reloads
639		are fine-grain operations.  In contrast, in the old algorithm
640		"spills" have to be bundled together in records.
641
642		Ideally, we should sink the spill record construction to where
643		it is needed.  We can even split the spill record into multiple ones
644		at the places where they are needed.  But CPS is not a good
645		representation for global code motion, so I'll keep it simple and
646		am not attempting this.
647
648		2. Incremental record construction (aka record splitting).
649
650		Long records with many component values which are simulatenously live
651		(recall that single use record addresses are not considered to
652		be live) are constructed with rawrecord and rawupdate.
653		We allocate space on the heap with rawrecord first, then gradually
654		fill it in with rawupdate.  This is the technique suggested to me
655		by Matthias.
656
657		Some restrictions on when this is applicable:
658		1. It is not a VECTOR record.  The code generator currently does not handle
659		this case. VECTOR record uses double indirection like arrays.
660		2. All the record component values are defined in the same "basic block"
661		as the record constructor.  This is to prevent speculative
662		record construction.
663
664		----------------------------------------------------------------------
665		Name: Allen Leung
666		Date: 2002/02/22 01:02:00 EST
667		Tag: leunga-20020222-mlrisc-tools
668
669		Minor bug fixes in the parser and rewriter
670
671		----------------------------------------------------------------------
672		Name: Allen Leung
673		Date: 2002/02/21 20:20:00 EST
674		Tag: leunga-20020221-peephole
675
676		Regenerated the peephole files.  Some contained typos in the specification
677		and some didn't compile because of pretty printing bugs in the old version
678		of 'nowhere'.
679
680		----------------------------------------------------------------------
681		Name: Allen Leung
682		Date: 2002/02/19 20:20:00 EST
683		Tag: leunga-20020219-mlrisc-tools
684		Description:
685
686		Minor bug fixes to the mlrisc-tools library:
687
688		1.  Fixed up parsing colon suffixed keywords
689		2.  Added the ability to shut the error messages up
690		3.  Reimplemented the pretty printer and fixed up/improved
691		the pretty printing of handle and -> types.
692		4.  Fixed up generation of literal symbols in the nowhere tool.
693		5.  Added some SML keywords to to sml.sty
694
695		----------------------------------------------------------------------
696		Name: Matthias Blume
697		Date: 2002/02/19 16:20:00 EST
698		Tag: blume-20020219-cmffi
699		Description:
700
701		A wild mix of changes, some minor, some major:
702
703		* All C FFI-related libraries are now anchored under $c:
704		$/c.cm      --> $c/c.cm
705		$/c-int.cm  --> $c/internals/c-int.cm
706		$/memory.cm --> $c/memory/memory.cm
707
708		* "make" tool (in CM) now treats its argument pathname slightly
709		differently:
710		1. If the native expansion is an absolute name, then before invoking
711		the "make" command on it, CM will apply OS.Path.mkRelative
712		(with relativeTo = OS.FileSys.getDir()) to it.
713		2. The argument will be passed through to subsequent phases of CM
714		processing without "going native".  In particular, if the argument
715		was an anchored path, then "make" will not lose track of that anchor.
716
717		* Compiler backends now "know" their respective C calling conventions
718		instead of having to be told about it by ml-nlffigen.  This relieves
719		ml-nlffigen from one of its burdens.
720
721		* The X86Backend has been split into X86CCallBackend and X86StdCallBackend.
722
723		* Export C_DEBUG and C_Debug from $c/c.cm.
724
725		* C type encoding in ml-nlffi-lib has been improved to model the conceptual
726		subtyping relationship between incomplete pointers and their complete
727		counterparts.  For this, ('t, 'c) ptr has been changed to 'o ptr --
728		with the convention of instantiating 'o with ('t, 'c) obj whenever
729		the pointer target type is complete.  In the incomplete case, 'o
730		will be instantiated with some "'c iobj" -- a type obtained by
731		using one of the functors PointerToIncompleteType or PointerToCompleteType.
732
733		Operations that work on both incomplete and complete pointer types are
734		typed as taking an 'o ptr while operations that require the target to
735		be known are typed as taking some ('t, 'c) obj ptr.
736
737		voidptr is now a bit "more concrete", namely "type voidptr = void ptr'"
738		where void is an eqtype without any values.  This makes it possible
739		to work on voidptr values using functions meant to operate on light
740		incomplete pointers.
741
742		* As a result of the above, signature POINTER_TO_INCOMPLETE_TYPE has
743		been vastly simplified.
744
745		----------------------------------------------------------------------
746		Name: Matthias Blume
747		Date: 2002/02/19 10:48:00 EST
748		Tag: blume-20020219-pqfix
749		Description:
750
751		Applied Chris Okasaki's bug fix for priority queues.
752
753		----------------------------------------------------------------------
754		Name: Matthias Blume
755		Date: 2002/02/15 17:05:00
756		Tag: Release_110_39
757		Description:
758
759		Last-minute retagging is becoming a tradition... :-(
760
761		This is the working release 110.39.
762
763		----------------------------------------------------------------------
764		Name: Matthias Blume
765		Date: 2002/02/15 16:00:00 EST
766		Tag: Release_110_39-orig
767		Description:
768
769		Working release 110.39.  New bootfiles.
770
771		(Update: There was a small bug in the installer so it wouldn't work
772		with all shells.  So I retagged. -Matthias)
773
774		----------------------------------------------------------------------
775		Name: Matthias Blume
776		Date: 2002/02/15 14:17:00 EST
777		Tag: blume-20020215-showbindings
778		Description:
779
780		Added EnvRef.listBoundSymbols and CM.State.showBindings.  Especially
781		the latter can be useful for exploring what bindings are available at
782		the interactive prompt.  (The first function returns only the list
783		of symbols that are really bound, the second prints those but also the
784		ones that CM's autoloading mechanism knows about.)
785
786		----------------------------------------------------------------------
787		Name: Matthias Blume
788		Date: 2002/02/15 12:08:00 EST
789		Tag: blume-20020215-iptrs
790		Description:
791
792		Two improvements to ml-nlffigen:
793
794		1. Write files only if they do not exist or if their current contents
795		do not coincide with what's being written.  (That is, avoid messing
796		with the time stamps unless absolutely necessary.)
797
798		2. Implement a "repository" mechanism for generated files related
799		to "incomplete pointer types".   See the README file for details.
800
801		----------------------------------------------------------------------
802		Name: Matthias Blume
803		Date: 2002/02/14 11:50:00 EST
804		Tag: blume-20020214-quote
805		Description:
806
807		Added a type 't t_' to tag.sml (in ml-nlffi-lib.cm).  This is required
808		because of the new and improved tag generation scheme.  (Thanks to Allen
809		Leung for pointing it out.)
810
811		----------------------------------------------------------------------
812		Name: Lal George
813		Date: 2002/02/14 09:55:27 EST 2002
814		Tag: george-20020214-isabelle-bug
815		Description:
816
817		Fixed the MLRISC bug sent by Markus Wenzel regarding the compilation
818		of Isabelle on the x86.
819
820		From Allen:
821		-----------
822		I've found the problem:
823
824		in ra-core.sml, I use the counter "blocked" to keep track of the
825		true number of elements in the freeze queue.  When the counter goes
826		to zero, I skip examining the queue.  But I've messed up the
827		bookkeeping in combine():
828
829		else ();
830		case !ucol of
831		PSEUDO => (if !cntv > 0 then
832		(if !cntu > 0 then blocked := !blocked - 1 else ();
833		^^^^^^^^^^^^^^^^^^^^^^^
834		moveu := mergeMoveList(!movev, !moveu)
835		)
836		else ();
837
838		combine() is called to coalesce two nodes u and v.
839		I think I was thinking that if the move counts of u and v are both
840		greater than zero then after they are coalesced then one node is
841		removed from the freeze queue.  Apparently I was thinking that
842		both u and v are of low degree, but that's clearly not necessarily true.
843
844
845		02/12/2002:
846		Here's the patch.  HOL now compiles.
847
848		I don't know how this impact on performance (compile
849		time or runtime).  This bug caused the RA (especially on the x86)
850		to go thru the potential spill phase when there are still nodes on the
851		freeze queue.
852
853
854
855
856		----------------------------------------------------------------------
857		Name: Matthias Blume
858		Date: 2002/02/13 22:40:00 EST
859		Tag: blume-20020213-fptr-rtti
860		Description:
861
862		Fixed a bug in ml-nlffigen that was introduced with one of the previous
863		updates.
864
865		----------------------------------------------------------------------
866		Name: Matthias Blume
867		Date: 2002/02/13 16:41:00 EST
868		Tag: blume-20020213-cmlpq
869		Description:
870
871		Added new priority queue export symbols (which have just been added to
872		smlnj-lib.cm) to CML's version of smlnj-lib.cm.  (Otherwise CML would
873		not compile and the installer would choke.)
874
875		----------------------------------------------------------------------
876		Name: Matthias Blume
877		Date: 2002/02/13 16:15:00 EST
878		Tag: blume-20020213-various
879		Description:
880
881		1. More tweaks to ml-nlffigen:
882
883		- better internal datastructures (resulting in slight speedup)
884		- "-match" option requires exact match
885		- "localized" gensym counters (untagged structs/unions nested within
886		other structs/unions or within typedefs get a fresh counter; their
887		tag will be prefixed by a concatenation of their parents' tags)
888		- bug fixes (related to calculation of transitive closure of types
889		to be included in the output)
890
891		2. Minor Basis updates:
892
893		- added implementations for List.collate and Option.app
894
895		----------------------------------------------------------------------
896		Name: Matthias Blume
897		Date: 2002/02/11 15:55:00 EST
898		Tag: blume-20020211-gensym
899		Description:
900
901		Added a "-gensym" option to command line of ml-nlffigen.  This can be
902		used to specify a "stem" -- a string that is inserted in all "gensym'd"
903		names (ML structure names that correspond to unnamed C structs, unions,
904		and enums), so that separate runs of ml-nlffigen do not clash.
905
906		----------------------------------------------------------------------
907		Name: Matthias Blume
908		Date: 2002/02/11 12:05:00 EST
909		Tag: blume-20020211-gensml
910		Description:
911
912		A quick fix for a problem with GenSML (in the pgraph-util library):
913		Make generation of toplevel "local" optional.  (Strictly speaking,
914		signature definitions within "local" are not legal SML.)
915
916		Other than that: updates to INSTALL and cm/TODO.
917
918		----------------------------------------------------------------------
919		Name: Matthias Blume
920		Date: 2002/02/08 15:00:00 EST
921		Tag: blume-20020208-uniquepid
922		Description:
923
924		0. Version number has been bumped to 110.38.1.  NEW BOOTFILES!!!
925
926		1. The installer (config/install.sh) has gotten smarter:
927
928		- Configuration options are a bit easier to specify now
929		(in config/targets).
930		- Bug in recognizing .tar.bz2 files fixed.
931		- Installer automatically resolves dependencies between
932		configuration options (e.g., if you ask for eXene, you will
933		also get cml -- regardless whether you asked for it or not).
934		- Installer can run in "quieter mode" by setting the environment
935		variable INSTALL_QUIETLY to "true".  "Quieter" does not mean
936		"completely silent", though.
937		- Build HashCons library as part of smlnj-lib.
938
939		2. A new scheme for assigning persistent identifiers to compilation
940		units (and, by extension, to types etc.) has been put into place.
941		This fixes a long-standing bug where types and even dynamic values
942		can get internally confused, thereby compromising type safety
943		(abstraction) and dynamic correctness.  See
944
945		http://cm.bell-labs.com/cm/cs/who/blume/pid-confusion.tgz
946
947		for an example of how things could go wrong until now.
948
949		The downside of the new scheme is that pids are not quite as
950		persistent as they used to be: CM will generate a fresh pid
951		for every compilation unit that it thinks it sees for the first
952		time.  That means that if you compile starting from a clean, fresh
953		source tree at two different times, you end up with different
954		binaries.
955
956		Cutoff recompilation, however, has not been compromised because
957		CM keeps pid information in special caches between runs.
958
959		----------------------------------------------------------------------
960		Name: Lal George
961		Date: 2002/02/07 15:34:13 EST 2002
962		Tag: <none>
963		Description:
964
965		Compilers that generate assembly code may produce  global labels
966		whose value is resolved at link time. The various peephole optimization
967		modules did not take this in account.
968
969		TODO. The Labels.addrOf function should really return an option
970		type so that clients are forced to deal with this issue, rather
971		than an exception being raised.
972
973		----------------------------------------------------------------------
974		Name: Lal George
975		Date: 2002/02/06 13:55:02 EST
976		Tag: george-20020206-ra-breakup
977		Description:
978
979		1. A bug fix from Allen.
980
981		A typo causes extra fstp %st(0)'s to be generated at compensation
982		edges, which might cause stack underflow traps at runtime.  This
983		occurs in fft where there are extraneous fstps right before the 'into'
984		trap instruction (in this case they are harmless since none of the
985		integers overflow.)
986
987		2. Pulled out various utility modules that were embedded in the modules
988		of the register allocator. I need these modules for other purposes, but
989		they are not complete enough to put into a library (just yet).
990		----------------------------------------------------------------------
991		Name: Matthias Blume
992		Date: 2002/01/31 16:05:00 EST
993		Tag: blume-20020131-sparc-ccalls
994		Description:
995
996		1. C-calls on Sparc needlessly allocated a huge chunk (96 bytes)
997		of extra stack space by mistake.  Fixed.
998
999		2. Bug in logic of handling of command-line options in ml-nlffigen fixed.
1000
1001		----------------------------------------------------------------------
1002		Name: Allen Leung
1003		Date: 2002/01/30
1004		Tag: leunga-20020130-nowhere-bug-fix
1005		Description:
1006
1007		MLRISC bug fixes:
1008		1. Fixed a bindings computation bug in the 'nowhere' program generator tool.
1009		2. MachineInt.fromString was negating its value.
1010
1011		----------------------------------------------------------------------
1012		Name: Matthias Blume
1013		Date: 2002/01/29
1014		Tag: blume-20020129-INSTALL
1015		Description:
1016
1017		- Added somewhat detailed installation instructions (file INSTALL).
1018		- Fixed curl-detection bug in config/install.sh.
1019		- It is now possible to select the URL getter using the URLGETTER
1020		environment variable:
1021
1022		not set / "unknown"      --> automatic detection (script tries wget,
1023		curl, and lynx)
1024		"wget" / "curl" / "lynx" --> use the specified program (script "knows"
1025		how to properly invoke them)
1026		other                    --> use $URLGETTER directly, it must take
1027		precisely two command-line arguments
1028		(source URL and destination file name)
1029
1030		----------------------------------------------------------------------
1031		Name: Matthias Blume
1032		Date: 2002/01/28
1033		Tag: blume-20020128-sparc-ccalls
1034		Description:
1035
1036		- Fixed problem with calculation of "used" registers in sparc-c-calls.
1037		- Make use of the allocParam argument in sparc-c-calls.
1038
1039		----------------------------------------------------------------------
1040		Name: Matthias Blume
1041		Date: 2002/01/28
1042		Tag: blume-20020128-allocParam
1043		Description:
1044
1045		John Reppy:  Changes c-calls API to accept client-callback for
1046		allocating extra stack space.
1047		me: Corresponding changes to mlriscGen (using a dummy argument that
1048		does not change the current behavior).
1049
1050		----------------------------------------------------------------------
1051		Name: Matthias Blume
1052		Date: 2002/01/28 12:00:00
1053		Tag: Release_110_38
1054		Description:
1055
1056		This time for real!!!
1057
1058		----------------------------------------------------------------------
1059		Name: Matthias Blume
1060		Date: 2002/01/28 10:56:00 EST
1061		Tag: blume-20020128-retraction
1062		Description:
1063
1064		0. Retracted earlier 110.38.  (The Release_110_38 tag has been replaced
1065		with blume-Release_110_38-retracted.)
1066
1067		1. Fixed a problem with incorrect rounding modes in real64.sml.
1068		(Thanks to Andrew Mccreight <andrew.mccreight@yale.edu>.)
1069
1070		2. A bug in ml-nlffigen related to the handling of unnamed structs, unions,
1071		and enums fixed.  The naming of corresponding ML identifiers should
1072		now be consistent again.
1073
1074		----------------------------------------------------------------------
1075		Name: Allen Leung
1076		Date: 2002/01/27
1077		Tag: leunga-20020127-nowhere
1078		Description:
1079
1080		Added a target called nowhere in the configuration scripts.
1081		Enabling this will build the MLRISC 'nowhere' tool (for translating
1082		programs with where-clauses into legal SML code) during installation.
1083
1084		----------------------------------------------------------------------
1085		Name: Matthias Blume
1086		Date: 2002/01/25 21:27:00 EST
1087		Tag: blume-Release_110_38-retracted
1088		Description:
1089
1090		Call it a (working) release!  Version is 110.38. Bootfiles are ready.
1091
1092		README will be added later.
1093
1094		!!! NOTE:  Re-tagged as blume-Release_110_38-retracted. Original tag
1095		(Release_110_38) removed.  Reason: Last-minute bug fixes.
1096
1097		----------------------------------------------------------------------
1098		Name: Matthias Blume
1099		Date: 2002/01/25
1100		Tag: blume-20020125-ffi
1101		Description:
1102
1103		A large number of tweaks and improvements to ml-nlffi-lib and
1104		ml-nlffigen:
1105
1106		- ML represenation types have been streamlined
1107		- getter and setter functions work with concrete values, not abstract
1108		ones where possible
1109		- ml-nlffigen command line more flexible (see README file there)
1110		- some bugs have been fixed (hopefully)
1111
1112		----------------------------------------------------------------------
1113		Name: Lal George
1114		Date: 2002/01/24
1115		Tag: george-20020124-risc-ra-interface
1116		Description:
1117
1118		There is a dramatic simplification in the interface to the
1119		register allocator for RISC architectures as a result of making
1120		parallel copy instructions explicit.
1121
1122		----------------------------------------------------------------------
1123		Name: Matthias Blume
1124		Date: 2002/01/22
1125		Tag: blume-20020122-x86-ccalls
1126		Description:
1127
1128		Bug fix for c-calls on x86 (having to do with how char- and
1129		short-arguments are being handled).
1130
1131		----------------------------------------------------------------------
1132		Name: Matthias Blume
1133		Date: 2002/01/21
1134		Tag: blume-20020121-ff
1135		Description:
1136
1137		Another day of fiddling with the FFI...
1138
1139		1. Bug fix/workaround:  CKIT does not complain about negative array
1140		dimensions, so ml-nlffigen has to guard itself against this possibility.
1141		(Otherwise a negative dimension would send it into an infinite loop.)
1142
1143		2. Some of the abstract types (light objects, light pointers, most "base"
1144		types) in structure C are now eqtypes.
1145
1146		3. Added constructors and test functions for NULL function pointers.
1147
1148		----------------------------------------------------------------------
1149		Name: Matthias Blume
1150		Date: 2002/01/18
1151		Tag: blume-20020118-ready-for-new-release
1152		Description:
1153
1154		Made config/srcarchiveurl point to a new place.  (Will provide boot
1155		files shortly.)
1156
1157		Maybe we christen this to be 110.38?
1158
1159		----------------------------------------------------------------------
1160		Name: Matthias Blume
1161		Date: 2002/01/18
1162		Tag: blume-20020118-more-ffifiddle
1163		Description:
1164
1165		Today's FFI fiddling:
1166
1167		- Provided a structure CGetSet with "convenient" versions of C.Get.* and
1168		C.Set.* that use concrete (MLRep.*) arguments and results instead
1169		of abstract ones.
1170
1171		- Provided word-style bit operations etc. for "int" representation
1172		types in MLRep.S<Foo>Bitops where <Foo> ranges over Char, Int, Short,
1173		and Long.
1174
1175		----------------------------------------------------------------------
1176		Name: Matthias Blume
1177		Date: 2002/01/18
1178		Tag: blume-20020118-use-x86-fp
1179		Description:
1180
1181		Now that x86-fast-fp seems to be working, I turned it back on again
1182		by default.  (Seems to work fine now, even with the FFI.)
1183
1184		Other than that, I added some documentation about the FFI to
1185		src/ml-nlffigen/README and updated the FFI test examples in
1186		src/ml-nlffi-lib/Tests/*.
1187
1188		----------------------------------------------------------------------
1189		Name: Allen Leung
1190		Date: 2002/01/17
1191		Tag: leunga-20020117-x86-fast-fp-call
1192		Description:
1193
1194		1. Fixed a problem with handling return fp values when x86's fast fp
1195		mode is turned on.
1196
1197		2. Minor pretty printing fix for cellset.  Print %st(0) as %st(0) instead
1198		of %f32.
1199
1200		3. Added a constructor INT32lit to the ast of MLRISC tools.
1201
1202		----------------------------------------------------------------------
1203		Name: Matthias Blume
1204		Date: 2002/01/16
1205		Tag: blume-20020116-ffifiddle
1206		Description:
1207
1208		More fiddling with the FFI interface:
1209
1210		- Make constness 'c instead of rw wherever possible.  This eliminates
1211		the need for certain explicit coercions.  (However, due to ML's
1212		value polymorphism, there will still be many cases where explicit
1213		coercions are necessary.  Phantom types are not the whole answer
1214		to modeling a subtyping relationship in ML.)
1215
1216		- ro/rw coersions for pointers added.  (Avoids the detour through */&.)
1217
1218		- "printf" test example added to src/ml-nlffi-lib/Tests.  (Demonstrates
1219		clumsy workaround for varargs problem.)
1220
1221		----------------------------------------------------------------------
1222		Name: Lal George
1223		Date: 2002/01/15
1224		Tag: <none>
1225		Description:
1226
1227		1. Since COPY instructions are no longer native to the architecture,
1228		a generic functor can be used to implement the expandCopies function.
1229
1230		2. Allowed EXPORT and IMPORT pseudo-op declarations to appear inside a
1231		TEXT segment.
1232
1233		----------------------------------------------------------------------
1234		Name: Matthias Blume
1235		Date: 2002/01/15
1236		Tag: blume-20020115-ffiupdates
1237		Description:
1238
1239		1. Fix for bug resulting in single-precision float values being returned
1240		incorrectly from FFI calls.
1241
1242		2. Small modifications to C FFI API:
1243
1244		- memory-allocation routines return straight objects (no options)
1245		and raise an exception in out-of-memory situations
1246		- unsafe extensions to cast between function pointers and pointers
1247		from/to ints
1248		- added structure C_Debug as an alternative to structure C where
1249		pointer-dereferencing (|*| and |*!) always check for null-pointers
1250		- added open_lib' to DynLinkage;  open_lib' works like open_lib
1251		but also takes a (possibly empty) list of existing library handles
1252		that the current library depends on
1253
1254		----------------------------------------------------------------------
1255		Name: Matthias Blume
1256		Date: 2002/01/10
1257		Tag: blume-20020110-newffigen
1258		Description:
1259
1260		1. Updates to portable graph code.
1261
1262		2. Major update to ml-nlffigen and ml-nlffi-lib.  Things are much
1263		more scalable now so that even huge interfaces such as the one
1264		for GTK compile in finite time and space. :-)
1265		See src/ml-nlffigen/README for details on what's new.
1266
1267		----------------------------------------------------------------------
1268		Name: Lal George
1269		Date: 2001/01/09 14:31:35 EST 2002
1270		Tag: george-20011206-rm-native-copy
1271		Description:
1272
1273		Removed the native COPY and FCOPY instructions
1274		from all the architectures and replaced it with the
1275		explicit COPY instruction from the previous commit.
1276
1277		It is now possible to simplify many of the optimizations
1278		modules that manipulate copies. This has not been
1279		done in this change.
1280
1281		----------------------------------------------------------------------
1282		Name: Lal George
1283		Date: 2001/12/06 16:50:13 EST 2001
1284		Tag: george-20011206-mlrisc-instruction
1285		Description:
1286
1287		Changed the representation of instructions from being fully abstract
1288		to being partially concrete. That is to say:
1289
1290		from
1291		type instruction
1292
1293		to
1294		type instr                              (* machine instruction *)
1295
1296		datatype instruction =
1297		LIVE of {regs: C.cellset, spilled: C.cellset}
1298		| KILL of {regs: C.cellset, spilled: C.cellset}
1299		| COPYXXX of {k: CB.cellkind, dst: CB.cell list, src: CB.cell list}
1300		| ANNOTATION of {i: instruction, a: Annotations.annotation}
1301		| INSTR of instr
1302
1303		This makes the handling of certain special instructions that appear on
1304		all architectures easier and uniform.
1305
1306		LIVE and KILL say that a list of registers are live or killed at the
1307		program point where they appear. No spill code is generated when an
1308		element of the 'regs' field is spilled, but the register is moved to
1309		the 'spilled' (which is present, more for debugging than anything else).
1310
1311		LIVE replaces the (now deprecated) DEFFREG instruction on the alpha.
1312		We used to generate:
1313
1314		DEFFREG f1
1315		f1 := f2 + f3
1316		trapb
1317
1318		but now generate:
1319
1320		f1 := f2 + f3
1321		trapb
1322		LIVE {regs=[f1,f2,f3], spilled=[]}
1323
1324		Furthermore, the DEFFREG (hack) required that all floating point instruction
1325		use all registers mentioned in the instruction. Therefore f1 := f2 + f3,
1326		defines f1 and uses [f1,f2,f3]! This hack is no longer required resulting
1327		in a cleaner alpha implementation. (Hopefully, intel will not get rid of
1328		this architecture).
1329
1330		COPYXXX is intended to replace the parallel COPY and FCOPY  available on
1331		all the architectures. This will result in further simplification of the
1332		register allocator that must be aware of them for coalescing purposes, and
1333		will also simplify certain aspects of the machine description that provides
1334		callbacks related to parallel copies.
1335
1336		ANNOTATION should be obvious, and now INSTR represents the honest to God
1337		machine instruction set!
1338
1339		The <arch>/instructions/<arch>Instr.sml files define certain utility
1340		functions for making porting easier -- essentially converting upper case
1341		to lower case. All machine instructions (of type instr) are in upper case,
1342		and the lower case form generates an MLRISC instruction. For example on
1343		the alpha we have:
1344
1345		datatype instr =
1346		LDA of {r:cell, b:cell, d:operand}
1347		| ...
1348
1349		val lda : {r:cell, b:cell, d:operand} -> instruction
1350		...
1351
1352		where lda is just (INSTR o LDA), etc.
1353
1354		----------------------------------------------------------------------
1355		Name: Matthias Blume
1356		Date: 2001/11/22 21:40:00 EST
1357		Tag: Release_110_37
1358		Description:
1359
1360		Release 110.37.  This time for real.
1361
1362		----------------------------------------------------------------------
1363		Name: Matthias Blume
1364		Date: 2001/11/21 16:35:00 EST
1365		Tag: blume-20011121-foot-in-mouth
1366		Description:
1367
1368		Removed the "Release_110_37" tag because of a serious bug.
1369		This will be re-tagged once the bug is fixed.
1370
1371		----------------------------------------------------------------------
1372		Name: Matthias Blume
1373		Date: 2001/11/21 16:14:00 EST
1374		Tag: blume-20011121-forgottenfile
1375		Description:
1376
1377		Forgot to add a file.  (Just a .tex-file -- part of
1378		the CM manual source.)
1379
1380		----------------------------------------------------------------------
1381		Name: Matthias Blume
1382		Date: 2001/11/21 16:10:00 EST
1383		Tag: blume-20011121-invalid_110_37
1384		Description:
1385
1386		Note: I removed the original tag "Release_110_37" from this commit
1387		because we found a serious bug in all non-x86 backends.
1388		- Matthias
1389
1390		1. Modifications to the SML/NJ code generator and to the runtime system
1391		so that code object name strings are directly inserted into code
1392		objects at code generation time.  The only business the runtime system
1393		has with this is now to read the name strings on occasions.
1394		(The encoding of the name string has also changed somewhat.)
1395
1396		2. CM now implements a simple "set calculus" for specifying export lists.
1397		In particular, it is now possible to refer to the export lists of
1398		other libraries/groups/sources and form unions as well as differences.
1399		See the latest CM manual for details.
1400
1401		3. An separate notion of "proxy" libraries has again be eliminated from
1402		CM's model.  (Proxy libraries are now simply a special case of using
1403		the export list calculus.)
1404
1405		4. Some of the existing libraries now take advantage of the new set
1406		calculus.
1407		(Notice that not all libraries have been converted because some
1408		of the existing .cm-files are supposed to be backward compatible
1409		with 110.0.x.)
1410
1411		5. Some cleanup in stand-alone programs.  (Don't use "exnMessage" -- use
1412		"General.exnMessage"!  The former relies on a certain hook to be
1413		initialized, and that often does not happen in the stand-alone case.)
1414
1415		----------------------------------------------------------------------
1416		Name: Lal George
1417		Date: 2001/11/21  13:56:18 EST
1418		Tag: george-2001121-pseudo-ops
1419		Description:
1420
1421		Implemented a complete redesign of MLRISC pseudo-ops. Now there
1422		ought to never be any question of incompatabilities with
1423		pseudo-op syntax expected by host assemblers.
1424
1425		For now, only modules supporting GAS syntax are implemented
1426		but more should follow, such as MASM, and vendor assembler
1427		syntax, e.g. IBM as, Sun as, etc.
1428
1429		----------------------------------------------------------------------
1430		Name: Matthias Blume
1431		Date: 2001/11/14 11:52:00 EST
1432		Tag: blume-20011114-srcname
1433		Description:
1434
1435		1. Routed the name of the current source file to mlriscgen where it
1436		should be directly emitted into the code object.  (This last part
1437		is yet to be done.)
1438
1439		2. Some cleanup of the pgraph code to make it match the proposal that
1440		I put out the other day.  (The proposal notwithstanding, things are
1441		still in flux here.)
1442
1443		----------------------------------------------------------------------
1444		Name: Lal George
1445		Date: 2001/11/14 09:44:04 EST
1446		Tag:
1447		Description:
1448
1449		Fix for a backpatching bug reported by Allen.
1450
1451		Because the boundary between short and long span-dependent
1452		instructions is +/- 128, there are an astounding number of
1453		span-dependent instructions whose size is over estimated.
1454
1455		Allen came up with the idea of letting the size of span
1456		dependent instructions be non-monotonic, for a maxIter
1457		number of times, after which the size must be monotonically
1458		increasing.
1459
1460		This table shows the number of span-dependent instructions
1461		whose size was over-estimated as a function of maxIter, for the
1462		file Parse/parse/ml.grm.sml:
1463
1464		maxIter            # of instructions:
1465		10                      687
1466		20                      438
1467		30                      198
1468		40                        0
1469
1470		In compiling the compiler, there is no significant difference in
1471		compilation speed between maxIter=10 and maxIter=40. Actually,
1472		my measurements showed that maxIter=40 was a tad faster than
1473		maxIter=10! Also 96% of the  files in the compiler reach a fix
1474		point within 13 iterations, so fixing maxIter at 40, while high,
1475		is okay.
1476
1477		----------------------------------------------------------------------
1478		Name: Matthias Blume
1479		Date: 2001/10/31 15:25:00 EST
1480		Tag: blume-20011031-pgraph
1481		Description:
1482
1483		CKIT:
1484		* Changed the "Function" constructor of type Ast.ctype to carry optional
1485		argument identifiers.
1486		* Changed the return type of TypeUtil.getFunction accordingly.
1487		* Type equality ignores the argument names.
1488		* TypeUtil.composite tries to preserve argument names but gives up quickly
1489		if there is a mismatch.
1490
1491		installation script:
1492		* attempts to use "curl" if available (unless "wget" is available as well)
1493
1494		CM:
1495		* has an experimental implementation of "portable graphs" which I will
1496		soon propose as an implementation-independent library format
1497		* there are also new libraries $/pgraph.cm and $/pgraph-util.cm
1498
1499		NLFFI-LIB:
1500		* some cleanup (all cosmetic)
1501
1502		NLFFIGEN:
1503		* temporarily disabled the mechanism that suppresses ML output for
1504		C definitions whose identifiers start with an underscore character
1505		* generate val bindings for enum constants
1506		* user can request that only one style (light or heavy) is being used;
1507		default is to use both (command-line arguments: -heavy and -light)
1508		* fixed bug in handling of function types involving incomplete pointers
1509		* generate ML entry points that take record arguments (i.e., using
1510		named arguments) for C functions that have a prototype with named
1511		arguments
1512		(see changes to CKIT)
1513
1514		----------------------------------------------------------------------
1515		Name: Allen Leung
1516		Date: 2001/10/27 20:34:00 EDT
1517		Tag: leunga-20011027-x86-fast-fp-call
1518		Description:
1519
1520		Fixed the bug described in blume-20010920-slowfp.
1521
1522		The fix involves
1523		1. generating FCOPYs in FSTP in ia32-svid
1524		2. marking a CALL with the appropriate annotation
1525
1526		----------------------------------------------------------------------
1527		Name: Matthias Blume
1528		Date: 2001/10/16 11:32:00 EDT
1529		Tag: blume-20011016-netbsd
1530		Description:
1531
1532		Underscore patch from Chris Richards (fixing problem with compiling
1533		runtime system under recent NetBSD).
1534
1535		----------------------------------------------------------------------
1536		Name: Allen Leung
1537		Date: 2001/10/12 17:18:32 EDT 2001
1538		Tag: leung-20011012-x86-printflowgraph
1539		Description:
1540
1541		X86RA now uses a valid (instead of dummy) PrintFlowgraph module.
1542
1543		----------------------------------------------------------------------
1544		Name: Lal George
1545		Date: 2001/10/11 23:51:34 EDT
1546		Tag: george-20011011-too-many-instrs
1547		Description:
1548
1549		The representation of a program point never expected to see more
1550		than 65536 instructions in a basic block!
1551
1552		----------------------------------------------------------------------
1553		Name: Lal George
1554		Date: 2001/10/09 09:41:37 EDT
1555		Tag: george-20011008-mlrisc-labels
1556		Description:
1557
1558		Changed the machine description files to support printing of
1559		local and global labels in assembly code, based on host assembler
1560		conventions.
1561
1562		----------------------------------------------------------------------
1563		Name: Matthias Blume
1564		Date: 2001/09/25 15:25:00 EDT
1565		Tag: blume-20010925-exninfo
1566		Description:
1567
1568		I provided a non-hook implementation of exnName (at the toplevel) and
1569		made the "dummy" implementation of exnMessage (at the toplevel) more
1570		useful: if nothing gets "hooked in", then at least you are going to
1571		see the exception name and a message indicating why you don't see more.
1572
1573		[For the time being, programs that need exnMessage and want to use
1574		ml-build should either use General.exnMessage (strongly recommended) or
1575		refer to structure General at some other point so that CM sees a
1576		static dependency.]
1577
1578		[Similar remarks go for "print" and "use":  If you want to use their
1579		functionality in stand-alone programs generated by ml-build, then use
1580		TextIO.output and Backend.Interact.useFile (from $smlnj/compiler.cm).]
1581
1582		----------------------------------------------------------------------
1583		Name: Matthias Blume
1584		Date: 2001/09/20 17:28:00 EDT
1585		Tag: blume-20010920-slowfp
1586		Description:
1587
1588		Allen says that x86-fast-fp is not safe yet, so I turned it off again...
1589
1590		----------------------------------------------------------------------
1591		Name: Matthias Blume
1592		Date: 2001/09/20 17:20:00 EDT
1593		Tag: blume-20010920-canonicalpaths
1594		Description:
1595
1596		0. Updated the BOOT file (something that I forgot to do earlier).
1597
1598		1. Small internal change to CM so that it avoids "/../" in filenames
1599		as much as possible (but only where it is safe).
1600
1601		2. Changed config/_run-sml (resulting in a changed bin/.run-sml) so
1602		that arguments that contain delimiters are passed through correctly.
1603		This change also means that all "special" arguments of the form
1604		@SMLxxx... must come first.
1605
1606		3. Changed install script to put relative anchor names for tool commands
1607		into pathconfig.
1608
1609		----------------------------------------------------------------------
1610		Name: Matthias Blume
1611		Date: 2001/09/18 15:35:00 EDT
1612		Tag: blume-20010918-readme11036
1613		Description:
1614
1615		Added README files.
1616
1617		----------------------------------------------------------------------
1618		Name: Matthias Blume
1619		Date: 2001/09/18 11:45:00 EDT
1620		Tag: Release_110_36 (retag)
1621		Description:
1622
1623		Fixed mistake in config/preloads. Retagged as 110.36.
1624
1625		----------------------------------------------------------------------
1626		Name: Matthias Blume
1627		Date: 2001/09/18 09:40:00 EDT
1628		Tag: Release_110_36_orig (tag changed)
1629		Description:
1630
1631		New version (110.36).  New bootfiles.
1632
1633		----------------------------------------------------------------------
1634		Name: Matthias Blume
1635		Date: 2001/09/14 16:15:00 EDT
1636		Tag: blume-20010914-x86fastfp
1637		Description:
1638
1639		John committed some changes that Allen made, in particular a (hopefully)
1640		correctly working version of the x86-fp module.
1641
1642		I changed the default setting of the Control.MLRISC.getFlag "x86-fast-fp"
1643		flag to "true".  Everything seems to compile to a fixpoint ok, and
1644		"mandelbrot" speeds up by about 15%.
1645
1646		----------------------------------------------------------------------
1647		Name: Matthias Blume
1648		Date: 2001/09/13 11:20:00 EDT
1649		Tag: blume-20010913-minimal
1650		Description:
1651
1652		1. Stefan Monnier's patch to fix a miscompilation problem that
1653		was brought to light by John Reppy's work on Moby.
1654
1655		2. Implemented a minimal "structure Compiler" that contains just
1656		"version" and "architecture".  The minimal version will be
1657		available when the full version is not.  This is for backward-
1658		compatibility with code that wants to test Compiler.version.
1659
1660		----------------------------------------------------------------------
1661		Name: Matthias Blume
1662		Date: 2001/08/28 14:03:00 EDT
1663		Tag: blume-20010828-ml-lex
1664		Description:
1665
1666		Fix for bug 1581, received from Neophytos Michael.
1667
1668		----------------------------------------------------------------------
1669		Name: Matthias Blume
1670		Date: 2001/08/27 11:20:00 EDT
1671		Tag: blume-20010827-readme11035
1672		Description:
1673
1674		Fleshed out the README file for 110.35.
1675
1676		----------------------------------------------------------------------
1677		Name: Matthias Blume
1678		Date: 2001/08/24 17:10:00 EDT
1679		Tag: Release_110_35
1680		Description:
1681
1682		New version number (110.35).  New bootfiles.
1683
1684		----------------------------------------------------------------------
1685		Name: Lal George
1686		Date: 2001/08/24 13:47:18 EDT 2001
1687		Tag: george-20010824-MLRISC-graphs
1688		Description:
1689
1690		removed clusters from MLRISC completely and replaced with graphs.
1691
1692		----------------------------------------------------------------------
1693		Name: Matthias Blume
1694		Date: 2001/08/23 17:50:00 EDT
1695		Tag: blume-20010823-toplevel
1696		Description:
1697
1698		- some reorganization of the code that implements various kinds of
1699		environments in the compiler (static, dynamic, symbolic, combined)
1700		- re-implemented the EnvRef module so that evalStream works properly
1701		(if the stream contains references to "use", "CM.make", etc.)
1702		- cleaned up evalloop.sml and interact.sml (but they need more cleaning)
1703
1704		----------------------------------------------------------------------
1705		Name: Matthias Blume
1706		Date: 2001/08/20 15:50 EDT
1707		Tag: blume20010820-slipup
1708		Description:
1709
1710		I forgot to commit a few files.  Here they are...
1711
1712		----------------------------------------------------------------------
1713		Name: Matthias Blume
1714		Date: 2001/08/20 15:35:00 EDT
1715		Tag: blume-20010820-debugprof
1716		Description:
1717
1718		!!!! NEW BOOTFILES !!!!
1719
1720		This is another round of reorganizing the compiler sources.  This
1721		time the main goal was to factor out all the "instrumentation"
1722		passes (for profiling and backtracing) into their own library.
1723		The difficulty was to do it in such a way that it does not depend
1724		on elaborate.cm but only on elabdata.cm.
1725
1726		Therefore there have been further changes to both elaborate.cm and
1727		elabdata.cm -- more "generic" things have been moved from the former
1728		to the latter.  As a result, I was forced to split the assignment
1729		of numbers indicating "primtyc"s into two portions: SML-generic and
1730		SML/NJ-specific.  Since it would have been awkward to maintain,
1731		I bit the bullet and actually _changed_ the mapping between these
1732		numbers and primtycs.  The bottom line of this is that you need
1733		a new set of bin- and bootfiles.
1734
1735		I have built new bootfiles for all architectures, so doing a fresh
1736		checkout and config/install.sh should be all you need.
1737
1738		The newly created library's name is
1739
1740		$smlnj/viscomp/debugprof.cm
1741
1742		and its sources live under
1743
1744		src/compiler/DebugProf
1745
1746		----------------------------------------------------------------------
1747		Name: Matthias Blume
1748		Date: 2001/08/15 17:15:00 EDT
1749		Tag: blume-20010815-compreorg
1750		Description:
1751
1752		This is a first cut at reorganizing the CM libraries that make up the
1753		core of the compiler.  The idea is to separate out pieces that could
1754		be used independently by tools, e.g., the parser, the typechecker, etc.
1755
1756		The current status is a step in this direction, but it is not quite
1757		satisfactory yet.  Expect more changes in the future.
1758
1759		Here is the current (new) organization...
1760
1761		What used to be $smlnj/viscomp/core.cm is now divided into
1762		six CM libraries:
1763
1764		$smlnj/viscomp/basics.cm
1765		/parser.cm
1766		/elabdata.cm
1767		/elaborate.cm
1768		/execute.cm
1769		/core.cm
1770
1771		The CM files for these libraries live under src/system/smlnj/viscomp.
1772		All these libraries are proxy libraries that contain precisely
1773		one CM library component.  Here are the locations of the components
1774		(all within the src/compiler tree):
1775
1776		Basics/basics.cm
1777		Parse/parser.cm
1778		ElabData/elabdata.cm
1779		Elaborator/elaborate.cm
1780		Execution/execute.cm
1781		core.cm
1782
1783		[This organization is the same that has been used already
1784		for a while for the architecture-specific parts of the visible
1785		compiler and for the old version of core.cm.]
1786
1787		As you will notice, many source files have been moved from their
1788		respective original locations to a new home in one of the above
1789		subtrees.
1790
1791		The division of labor between the new libraries is the following:
1792
1793		basics.cm:
1794		- Simple, basic definitions that pertain to many (or all) of
1795		the other libraries.
1796		parser.cm:
1797		- The SML parser, producing output of type Ast.dec.
1798		- The type family for Ast is also defined and exported here.
1799		elabdata.cm:
1800		- The datatypes that describe input and output of the elaborator.
1801		This includes types, absyn, and static environments.
1802		elaborator.cm:
1803		- The SML/NJ type checker and elaborator.
1804		This maps an Ast.dec (with a given static environment) to
1805		an Absyn.dec (with a new static environment).
1806		- This libraries implements certain modules that used to be
1807		structures as functors (to remove dependencies on FLINT).
1808		execute.cm:
1809		- Everything having to do with executing binary code objects.
1810		- Dynamic environments.
1811		core.cm:
1812		- SML/NJ-specific instantiations of the elaborator and MLRISC.
1813		- Top-level modules.
1814		- FLINT (this should eventually become its own library)
1815
1816		Notes:
1817
1818		I am not 100% happy with the way I separated the elaborator (and its
1819		data structures) from FLINT.  Two instances of the same problem:
1820
1821		1. Data structures contain certain fields that carry FLINT-specific
1822		information.  I hacked around this using exn and the property list
1823		module from smlnj-lib.  But the fact that there are middle-end
1824		specific fields around at all is a bit annoying.
1825
1826		2. The elaborator calculates certain FLINT-related information.  I tried
1827		to make this as abstract as I could using functorization, but, again,
1828		the fact that the elaborator has to perform calculations on behalf
1829		of the middle-end at all is not nice.
1830
1831		3. Having to used exn and property lists is unfortunate because it
1832		weakens type checking.  The other alternative (parameterizing
1833		nearly *everything*) is not appealing, though.
1834
1835		I removed the "rebinding =" warning hack because due to the new organization
1836		it was awkward to maintain it.  As a result, the compiler now issues some of
1837		these warnings when compiling init.cmi during bootstrap compilation. On
1838		the plus side, you also get a warning when you do, for example:
1839		val op = = Int32.+
1840		which was not the case up to now.
1841
1842		I placed "assign" and "deref" into the _Core structure so that the
1843		code that deals with the "lazy" keyword can find them there.  This
1844		removes the need for having access to the primitive environment
1845		during elaboration.
1846
1847		----------------------------------------------------------------------
1848		Name: Matthias Blume
1849		Date: 2001/08/13
1850		Tag: blume-20010813-closures
1851		Description:
1852
1853		This fix was sent to us by Zhong Shao.  It is supposed to improve the
1854		performance of certain loops by avoiding needless closure allocation.
1855
1856		----------------------------------------------------------------------
1857		Name: Lal George
1858		Date: 2001/07/31 10:03:23 EDT 2001
1859		Tag: george-20010731-x86-fmalloc
1860		Description: Fixed bug in x86 calls
1861
1862		There was a bug where call instructions would mysteriously
1863		vanish. The call instruction had to be one that returned
1864		a floating point value.
1865
1866		----------------------------------------------------------------------
1867		Name: Lal George
1868		Date: 2001/07/19 16:36:29 EDT 2001
1869		Tag: george-20010719-simple-cells
1870		Description:
1871
1872		I have dramatically simplified the interface for CELLS in MLRISC.
1873
1874		In summary, the cells interface is broken up into three parts:
1875
1876		1. CellsBasis : CELLS_BASIS
1877
1878		CellsBasis is a top level structure and common for all
1879		architectures.  it contains the definitions of basic datatypes
1880		and utility  functions over these types.
1881
1882		2. functor Cells() : CELLS
1883
1884		Cells generates an interface for CELLS that incorporates the
1885		specific resources on the target architecture, such as the
1886		presence of special register classes, their number and size,
1887		and various useful substructures.
1888
1889		3. <ARCH>CELLS
1890
1891		e.g. SparcCells: SPARCCELLS
1892
1893		<ARCH>CELLS usually contains additional bindings for special
1894		registers  on the architecture, such as:
1895
1896		val r0 : cell           (* register zero *)
1897		val y : cell            (* Y register *)
1898		val psr : cell          (* processor status register *)
1899		...
1900
1901		The structure returned by applying the Cells functor is opened
1902		in this interface.
1903
1904		The main implication of all this is that the datatypes for cells is
1905		split between CellsBasis and CELLS -- a fairly simple change for user
1906		code.
1907
1908		In the old scheme the CELLS interface had a definitional binding of
1909		the form:
1910
1911		signature CELLS = sig
1912
1913		structure CellsBasis = CellsBasis
1914
1915		...
1916
1917		end
1918
1919		With all the sharing constraints that goes on in MLRISC, this old
1920		design  quickly leads to errors such as:
1921
1922		"structure definition spec inside of sharing ... "
1923
1924
1925		and appears to require an unacceptable amount of sharing and where
1926		constraint hackery.
1927
1928		I think this error message (the interaction of definitional specs and
1929		sharing) requires more explanation on our web page.
1930
1931		----------------------------------------------------------------------
1932		Name: Matthias Blume
1933		Date: 2001/07/19 15:00:00 EDT
1934		Tag: blume-20010719-libreorg
1935		Description:
1936
1937		This update puts together a fairly extensive but straightforward change
1938		to the way the libraries that implement the interactive system are
1939		organized:
1940
1941		The biggest change is the elimination of structure Compiler.  As a
1942		replacement for this structure, there is now a CM library
1943		(known as $smlnj/compiler.cm or $smlnj/compiler/current.cm)
1944		that exports all the substructures of the original structure Compiler
1945		directly.  So instead of saying Compiler.Foo.bar one now simply
1946		says Foo.bar.  (The CM libraries actually export a collection of
1947		structures that is richer than the collection of substructures of
1948		structure Compiler.)
1949
1950		To make the transition smooth, there is a separate library called
1951		$smlnj/compiler/compiler.cm which puts together and exports the
1952		original structure Compiler (or at least something very close to it).
1953
1954		There are five members of the original structure Compiler
1955		that are not exported directly but which instead became members
1956		of a new structure Backend (described by signature BACKEND).  These are:
1957		structure Profile (: PROFILE), structure Compile (: COMPILE), structure
1958		Interact (: INTERACT), structure Machine (: MACHINE), and val
1959		architecture (: string).
1960
1961		Structure Compiler.Version has become structure CompilerVersion.
1962
1963		Cross-compilers for alpha32, hppa, ppc, sparc, and x86 are provided
1964		by $smlnj/compiler/<arch>.cm where <arch> is alpha32, hppa, ppc, sparc,
1965		or x86, respectively.
1966		Each of these exports the same frontend structures that
1967		$smlnj/compiler.cm exports.  But they do not have a structure Backend
1968		and instead export some structure <Arch>Backend where <Arch> is Alpha32,
1969		Hppa, PPC, Sparc, or X86, respectively.
1970
1971		Library $smlnj/compiler/all.cm exports the union of the exports of
1972		$smlnj/compiler/<arch>.cm
1973
1974		There are no structures <Arch>Compiler anymore, use
1975		$smlnj/compiler/<arch>.cm instead.
1976
1977		Library host-compiler-0.cm is gone.  Instead, the internal library
1978		that instantiates CM is now called cm0.cm.  Selection of the host
1979		compiler (backend) is no longer done here but. (Responsibility for it
1980		now lies with $smlnj/compiler/current.cm.  This seems to be more
1981		logical.)
1982
1983		Many individual files have been moved or renamed.  Some files have
1984		been split into multiple files, and some "dead" files have been deleted.
1985
1986		Aside from these changes to library organization, there are also changes
1987		to the way the code itself is organized:
1988
1989		Structure Binfile has been re-implemented in such a way that it no
1990		longer needs any knowledge of the compiler.  It exclusively deals
1991		with the details of binfile layout.  It no longer invokes the
1992		compiler (for the purpose of creating new prospective binfile
1993		content), and it no longer has any knowledge of how to interpret
1994		pickles.
1995
1996		Structure Compile (: COMPILE) has been stripped down to the bare
1997		essentials of compilation.  It no longer deals with linking/execution.
1998		The interface has been cleaned up considerably.
1999
2000		Utility routines for dealing with linking and execution have been
2001		moved into their own substructures.
2002
2003		(The ultimate goal of these changes is to provide a light-weight
2004		binfile loader/linker (at least for, e.g., stable libraries) that
2005		does not require CM or the compiler to be present.)
2006
2007		CM documentation has been updated to reflect the changes to library
2008		organization.
2009
2010		----------------------------------------------------------------------
2011		Name: Matthias Blume
2012		Date: 2001/07/10 17:30:00 EDT
2013		Tag: Release_110_34
2014		Description:
2015
2016		Minor tweak to 110.34 (re-tagged):
2017
2018		- README.html file added to CVS repository
2019		- runtime compiles properly under FreeBSD 3.X and 4.X
2020
2021		----------------------------------------------------------------------
2022		Name: Matthias Blume
2023		Date: 2001/07/10 17:30:00 EDT
2024		Tag: Release_110_34
2025		Description:
2026
2027		New version number (110.34). New bootfiles.
2028
2029		----------------------------------------------------------------------
2030		Name: Matthias Blume
2031		Date: 2001/07/09 16:00:00 EDT
2032		Tag: blume-20010709-more-varargs
2033		Description:
2034
2035		I changed the handling of varargs in ml-nlffigen again:
2036		The ellipsis ... will now simply be ignored (with an accompanying warning).
2037
2038		The immediate effect is that you can actually call a varargs function
2039		from ML -- but you can't actually supply any arguments beyond the ones
2040		specified explicitly.  (For example, you can call printf with its format
2041		string, but you cannot pass additional arguments.)
2042
2043		This behavior is only marginally more useful than the one before, but
2044		it has the advantage that a function or, more importantly, a function
2045		type never gets dropped on the floor, thus avoiding follow-up problems with
2046		other types that refer to the offending one.
2047
2048		----------------------------------------------------------------------
2049		Name: Matthias Blume
2050		Date: 2001/07/09 11:25:00 EDT
2051		Tag: blume-20010709-varargs
2052		Description:
2053
2054		1. ckit-lib.cm now exports structure Error
2055		2. ml-nlffigen reports occurences of "..." (i.e., varargs function types)
2056		with a warning accompanied by a source location.  Moreover, it
2057		merely skips the offending function or type and proceeds with the
2058		rest of its work.u  As a result, one can safely feed C code containing
2059		"..." to ml-nlffigen.
2060		3. There are some internal improvements to CM, providing slightly
2061		more general string substitutions in the tools subsystem.
2062
2063		----------------------------------------------------------------------
2064		Name: Matthias Blume
2065		Date: 2001/06/27 15:10:00 EDT
2066		Tag: blume-20010627-concur
2067		Description:
2068
2069		Fixed a small bug in CM's handling of parallel compilation.
2070		(You could observe the bug by Control-C-interrupting an ordinary
2071		CMB.make or CM.stabilize and then attaching some compile servers.
2072		The result was that all of a sudden the previously interrupted
2073		compilation would continue on its own.  This was because of
2074		an over-optimization: CM did not bother to clean out certain queues
2075		when no servers were attached "anyway", resulting in the contents
2076		of these queues to grab control when new servers did get attached.)
2077
2078		There is also another minor update to the CM manual.
2079
2080		----------------------------------------------------------------------
2081		Name: Matthias Blume
2082		Date: 2001/06/26 16:15:00 EDT
2083		Tag: blume-20010626-cmdoc
2084		Description:
2085
2086		Minor typo fixed in CM manual (syntax diagram for libraries).
2087
2088		----------------------------------------------------------------------
2089		Name: Matthias Blume
2090		Date: 2001/06/25 22:55:00 EDT
2091		Tag: blume-20010625-x86pc
2092		Description:
2093
2094		Fixed a nasty bug in the X86 assembly code that caused signal
2095		handlers to fail (crash) randomly.
2096
2097		----------------------------------------------------------------------
2098		Name: Matthias Blume
2099		Date: 2001/06/25 12:05:00 EDT
2100		Tag: blume-20010625-nlffigen
2101		Description:
2102
2103		This update fixes a number of minor bugs in ml-nlffigen as reported by
2104		Nick Carter <nbc@andrew.cmu.edu>.
2105
2106		1. Silly but ok typedefs of the form "typedef void myvoid;" are now accepted.
2107		2. Default names for generated files are now derived from the name of
2108		the C file *without its directory*.  In particular, this causes generated
2109		files to be placed locally even if the C file is in some system directory.
2110		3. Default names for generated signatures and structures are also derived
2111		from the C file name without its directory.  This avoids silly things
2112		like "structure GL/GL".
2113		(Other silly names are still possible because ml-nlffigen does not do
2114		a thorough check of whether generated names are legal ML identifiers.
2115		When in doubt, use command line arguments to force particular names.)
2116
2117		----------------------------------------------------------------------
2118		Name: Matthias Blume
2119		Date: 2001/06/21 12:25:00 EDT
2120		Tag: blume-20010621-eXene
2121		Description:
2122
2123		eXene now compiles and (sort of) works again.
2124
2125		The library name (for version > 110.33) is $/eXene.cm.
2126
2127		I also added an new example in src/eXene/examples/nbody.  See the
2128		README file there for details.
2129
2130		----------------------------------------------------------------------
2131		Name: Matthias Blume
2132		Date: 2001/06/20 16:40:00 EDT
2133		Tag: blume-20010620-cml
2134		Description:
2135
2136		CML now compiles and works again.
2137
2138		Libraries (for version > 110.33):
2139
2140		$cml/cml.cm            Main CML library.
2141		$cml/basis.cm          CML's version of $/basis.cm.
2142		$cml/cml-internal.cm   Internal helper library.
2143		$cml/core-cml.cm       Internal helper library.
2144		$cml-lib/trace-cml.cm  Tracing facility.
2145		$cml-lib/smlnj-lib.cm  CML's version of $/smlnj-lib.cm
2146
2147		The installer (config/install.sh) has been taught how to properly
2148		install this stuff.
2149
2150		----------------------------------------------------------------------
2151		Name: Matthias Blume
2152		Date: 2001/06/19 17:55:00 EDT
2153		Tag: blume-20010619-instantiate
2154		Description:
2155
2156		This un-breaks the fix for bug 1432.
2157		(The bug was originally fixed in 110.9 but I broke it again some
2158		time after that.)
2159
2160		----------------------------------------------------------------------
2161		Name: Matthias Blume
2162		Date: 2001/06/19 17:25:00 EDT
2163		Tag: blume-20010619-signals
2164		Description:
2165
2166		This should (hopefully) fix the long-standing signal handling bug.
2167		(The runtime system was constructing a continuation record with an
2168		incorrect descriptor which would cause the GC to drop data on the floor...)
2169
2170		----------------------------------------------------------------------
2171		Name: Matthias Blume
2172		Date: 2001/06/15 15:05:00 EDT
2173		Tag: blume-20010615-moresparc
2174		Description:
2175
2176		Here is a short late-hour update related to Sparc c-calls:
2177
2178		-- made handling of double-word arguments a bit smarter
2179
2180		-- instruction selection phase tries to collapse certain clumsily
2181		constructed ML-Trees; typical example:
2182
2183		ADD(ty,ADD(_,e,LI d1),LI d2)  ->  ADD(ty,e,LI(d1+d2))
2184
2185		This currently has no further impact on SML/NJ since mlriscGen does
2186		not seem to generate such patterns in the first place, and c-calls
2187		(which did generate them in the beginning) has meanwhile been fixed
2188		so as to avoid them as well.
2189
2190		----------------------------------------------------------------------
2191		Name: Matthias Blume
2192		Date: 2001/06/15 15:05:00 EDT
2193		Tag: blume-20010615-sparc
2194		Description:
2195
2196		The purpose of this update is to provide an implementation of NLFFI
2197		on Sparc machines.
2198
2199		Here are the changes in detail:
2200
2201		* src/MLRISC/sparc/c-calls/sparc-c-calls.sml is a new file containing
2202		the Sparc implementation of the c-calls API.
2203		* The Sparc backend of SML/NJ has been modified to uniformely use %fp
2204		for accessing the ML frame.  Thus, we have a real frame pointer and
2205		can freely modify %sp without need for an omit-frame-ptr phase.
2206		The vfp logic in src/compiler/CodeGen/* has been changed to accomodate
2207		this case.
2208		* ml-nlffigen has been taught to produce code for different architectures
2209		and calling conventions.
2210		* In a way similar to what was done in the x86 case, the Sparc
2211		backend uses its own specific extension to mltree.  (For example,
2212		it needs to be able to generate UNIMP instructions which are part
2213		of the calling convention.)
2214		* ml-nlffi-lib was reorganized to make it more modular (in particular,
2215		to make it easier to plug in new machine- and os-dependent parts).
2216
2217		There are some other fairly unrelated bug fixes and cleanups as well:
2218
2219		* I further hacked the .cm files for MLRISC tools (like MDLGen) so
2220		that they properly share their libraries with existing SML/NJ libraries.
2221		* I fixed a minor cosmetic bug in CM, supressing certain spurious
2222		follow-up error messages.
2223		* Updates to CM/CMB documentation.
2224
2225		TODO items:
2226
2227		* MLRISC should use a different register as its asmTemp on the Sparc.
2228		(The current %o2 is a really bad choice because it is part of the
2229		calling conventions, so things might interfere in unexpected ways.)
2230
2231		----------------------------------------------------------------------
2232		Name: Matthias Blume
2233		Date: 2001/06/07
2234		Tag: blume-20010607-calls
2235		Description:
2236
2237		A number of internal changes related to C calls and calling conventions:
2238
2239		1. ML-Tree CALL statements now carry a "pops" field.  It indicates the
2240		number of bytes popped implicitly (by the callee).  In most cases
2241		this field is 0 but on x86/win32 it is some non-zero value.  This
2242		is information provided for the benefit of the "omit-frameptr" pass.
2243		2. The CALL instruction on the x86 carries a similar "pops" field.
2244		The instruction selection phase copies its value from the ML-Tree
2245		CALL statement.
2246		3. On all other architectures, the instruction selection phase checks
2247		whether "pops=0" and complains if not.
2248		4. The c-calls implementation for x86 now accepts two calling conventions:
2249		"ccall" and "stdcall".  When "ccall" is selected, the caller cleans
2250		up after the call and pops is set to 0.  For "stdcall", the caller
2251		does nothing, leaving the cleanup to the callee; pops is set to
2252		the number of bytes that were pushed onto the stack.
2253		5. The cproto decoder (compiler/Semant/types/cproto.sml) now can
2254		distinguish between "ccall" and "stdcall".
2255		6. The UNIMP instruction has been added to the supported Sparc instruction
2256		set. (This is needed for implementing the official C calling convention
2257		on this architecture.)
2258		7. I fixed some of the .cm files under src/MLRISC/Tools to make them
2259		work with the latest CM.
2260
2261		----------------------------------------------------------------------
2262		Name: Matthias Blume
2263		Date: 2001/06/05 15:10:00 EDT
2264		Tag: blume-20010605-cm-index
2265		Description:
2266
2267		0. The "lambdasplit" parameter for class "sml" in CM has been documented.
2268
2269		1. CM can now generate "index files".  These are human-readable files
2270		that list on a per-.cm-file basis each toplevel symbol defined or
2271		imported.  The location of the index file for
2272		<p>/<d>.cm is <p>/CM/INDEX/<d>.cm.
2273		To enable index-file generation, set CM.Control.generate_index to true
2274		or export an environment-symbol: export CM_GENERATE_INDEX=true.
2275
2276		The CM manual has been updated accordingly.
2277
2278		2. I made some slight modifications to the c-calls API in MLRISC.
2279
2280		a) There is now a callback to support saving/restoring of
2281		dedicated but caller-save registers around the actual call
2282		instruction.
2283		b) One can optionally specify a comment-annotation for the
2284		call instruction.
2285
2286		3. SML/NJ (mlriscGen.sml) uses this new API for the rawccall primop.
2287		(For example, the comment annotation shows the C prototype of
2288		the function being called.)
2289
2290		----------------------------------------------------------------------
2291		Name: Matthias Blume
2292		Date: 2001/06/01 13:30:00 EDT
2293		Tag: blume-20010601-nlffi-cleanup
2294		Description:
2295
2296		This is mostly a cleanup of MLFFI stuff:
2297
2298		- some signature files have been put into a more exposed place
2299		- the ugly 'f type parameter is gone (simplifies types tremendously!)
2300		- ml-nlffigen changed accordingly
2301		- tutorial updated
2302
2303		Other changes:
2304
2305		- author's affiliation in CM manual(s) updated
2306		- some more recognized keywords added to Allen's sml.sty
2307
2308		----------------------------------------------------------------------
2309		Name: Matthias Blume
2310		Date: 2001/05/25 15:30:00 EDT
2311		Tag: blume-20010525-iptr
2312		Description:
2313
2314		- put the official 110.33-README (as it appears on the ftp server) under
2315		CVS
2316		- fixed a small bug related to incomplete pointer types in
2317		ml-nlffigen
2318		- small cosmetic change to the ml-nlffi-lib's "arr" type constructor
2319		(it does not need the 'f type parameter)
2320
2321		----------------------------------------------------------------------
2322		Name: Matthias Blume
2323		Date: 2001/05/23 14:30:00 EDT
2324		Tag: Release_110_33
2325		Description:
2326
2327		New version number (110.33).  New bootfiles.
2328
2329		----------------------------------------------------------------------
2330		Name: Matthias Blume
2331		Date: 2001/05/22 18:06:00 EDT
2332		Tag: blume-20010522-targets
2333		Description:
2334
2335		Made install.sh use file config/targets.customized if it exists, falling
2336		back to config/targets if it doesn't.  This way one can have a customized
2337		version of the targets file without touching the "real thing", thus
2338		eliminating the constant fear of accidentally checking something bogus
2339		back into the CVS repository...  (File config/targets.customized must
2340		not be added to the repository!)
2341
2342		----------------------------------------------------------------------
2343		Name: Matthias Blume
2344		Date: 2001/05/22 16:30:00 EDT
2345		Tag: blume-20010522-minitut
2346		Description:
2347
2348		1. Bug fix in ml-nlffigen; now (hopefully) correctly handling
2349		struct returns.
2350		2. Added src/ml-nlffi-lib/Doc/mini-tutorial.txt.  This is some very
2351		incomplete, preliminary documentation for NLFFI.
2352
2353		----------------------------------------------------------------------
2354		Name: Matthias Blume
2355		Date: 2001/05/14 11:30:00 EDT
2356		Tag: blume-20010514-script
2357		Description:
2358
2359		Some bugs in install script fixed.
2360
2361		In addition to that I also made a slight change to the NLFFI API:
2362		Functors generated by ml-nlffigen now take the dynamic library as a
2363		straight functor argument, not as a suspended one.  (The original
2364		functor code used to force the suspension right away anyway, so there
2365		was nothing gained by this complication of the interface.)
2366
2367		----------------------------------------------------------------------
2368		Name: Matthias Blume
2369		Date: 2001/05/11 14:35:00 EDT
2370		Tag: blume-20010511-ml-nlffi
2371		Description:
2372
2373		I finally took the plunge and added my new FFI code to the main
2374		repository.  For x86-linux it is now ready for prime-time.
2375
2376		There are two new subdirectories of "src":
2377
2378		- ml-nlffi-lib:
2379		The utility library for programs using the FFI interface.
2380		Here is the implementation of $/c.cm and its associated low-level
2381		partners $/c-int.cm and $/memory.cm.
2382		- ml-nlffigen:
2383		A stand-alone program for generating ML glue code from C source
2384		code.
2385
2386		Building ml-nlffigen requires $/ckit-lib.cm.
2387
2388		The config/install.sh script has been updates to do the Right Thing
2389		(hopefully).
2390
2391		Notice that the source tree for the C-Kit will not be put under "src"
2392		but directly under the installation root directory.  (This is the
2393		structure that currently exists on the CVS server when you check out
2394		module "sml".)  Fortunately, config/install.sh knows about this oddity.
2395
2396		Bugs: No documentation yet.
2397
2398		----------------------------------------------------------------------
2399		Name: Matthias Blume
2400		Date: 2001/05/09 16:35:00 EDT
2401		Tag: blume-20010509-cpscontract
2402		Description:
2403
2404		Fixed a bug in the accounting code in cpsopt/contract.sml.  (The
2405		wrapper/unwrapper elimination did not decrement usage counts and some
2406		dead variables got overlooked by the dead-up logic.)
2407
2408		----------------------------------------------------------------------
2409		Name: Lal George
2410		Date: 2001/05/08  17:26:09 EDT
2411		Tag: george-20010508-omit-frameptr
2412		Description:
2413
2414		Changes to implement the omit-frame-pointer optimization to support
2415		raw C calls. For now, there is only support on the Intel x86, but
2416		other architectures will follow as more experience is gained with this.
2417
2418
2419		----------------------------------------------------------------------
2420		Name: Matthias Blume
2421		Date: 2001/05/07 14:40:00 EDT
2422		Tag: blume-20010507-proxies
2423		Description:
2424
2425		I made into "proxy libraries" all libraries that qualify for such a
2426		change.  (A qualifying library is a library that has another library or
2427		groups as its sole member and repeats that member's export list
2428		verbatim.  A proxy library avoids this repetition by omitting its export
2429		list, effectively inheriting the list that its (only) member exports.
2430		See the CM manual for more explanation.)
2431		The main effect is that explicit export lists for these libraries
2432		do not have to be kepts in sync, making maintenance a bit easier.
2433
2434		I also added copyright notices to many .cm-files.
2435
2436		Last but not least, I made a new set of bootfiles.
2437
2438		----------------------------------------------------------------------
2439		Name: Matthias Blume
2440		Date: 2001/05/04 17:00:00 EDT
2441		Tag: blume-20010504-cm-lsplit
2442		Description:
2443
2444		0. John merged pending changes to $/smlnj-lib.cm
2445
2446		1. Allen's previous change accidentally backed out of one of Lal's
2447		earlier changes.  I undid this mistake (re-introducing Lal's change).
2448
2449		2. I used the new topOrder' function from graph-scc.sml (from $/smlnj-lib.cm)
2450		within the compiler where applicable.  There is some code simplification
2451		because of that.
2452
2453		3. The "split" phase (in FLINT) is now part of the default list of phases.
2454		Compiler.Control.LambdaSplitting.* can be used to globally control the
2455		lambda-splitting (cross-module-inlining) engine.  In addition to that,
2456		it can now also be controlled on a per-source basis: CM has been taught
2457		a new tool parameter applicable to ML source files.
2458
2459		- To turn lambda-splitting off completely:
2460		local open Compiler.Control.LambdaSplitting in
2461		val _ = set Off
2462		end
2463		- To make "no lambda-splitting" the global default (but allow per-source
2464		overriding); this is the initial setting:
2465		local open Compiler.Control.LambdaSplitting in
2466		val _ = set (Default NONE)
2467		end
2468		- To make "lambda-splitting with aggressiveness a" the global default
2469		(and allow per-source overriding):
2470		local open Compiler.Control.LambdaSplitting in
2471		val _ = set (Default (SOME a))
2472		end
2473
2474		- To turn lambda-splitting off for a given ML souce file (say: a.sml)
2475		write (in the respective .cm-file):
2476		a.sml (lambdasplitting:off)
2477		- To turn lambda-splitting for a.sml on with minimal aggressiveness:
2478		a.sml (lambdasplitting:on)
2479		- To turn lambda-splitting for a.sml on with aggressiveness <a> (where
2480		<a> is a decimal non-negative integer):
2481		a.sml (lambdasplitting:<a>)
2482		- To turn lambda-splitting for a.sml on with maximal aggressiveness:
2483		a.sml (lambdasplitting:infinity)
2484		- To use the global default for a.sml:
2485		a.sml (lambdasplitting:default)
2486		or simply
2487		a.sml
2488
2489	----------------------------------------------------------------------	----------------------------------------------------------------------
2490	Name: Allen Leung	Name: Allen Leung
2491	Date: 2001/05/04 01:57:00 EDT	Date: 2001/05/04 01:57:00 EDT
#	Line 2109	Line 4584
4584	elaborator). There were a lot of changes during my "linkpath" trials	elaborator). There were a lot of changes during my "linkpath" trials
4585	that could have been reverted to their original state but weren't.	that could have been reverted to their original state but weren't.
4586	Please, don't be too harsh on me for messing with this code a bit more	Please, don't be too harsh on me for messing with this code a bit more
4587	than what was strictly necessary...  (I _did_ resist the tempation	than what was strictly necessary...  (I _did_ resist the temptation
4588	of doing any "global reformatting" to avoid an untimely death at	of doing any "global reformatting" to avoid an untimely death at
4589	Dave's hands. :)	Dave's hands. :)
4590

---

Colored Diff Long Colored Diff Unidiff Context Diff Side by Side

Legend: Removed from v.815   changed lines   Added in v.1158

---

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