Diff of /sml/trunk/HISTORY

Parent Directory | Revision Log | Patch

	revision 1447, Mon Jan 26 17:24:47 2004 UTC	revision 1682, Tue Nov 9 23:48:09 2004 UTC
#	Line 14	Line 14
14
15	----------------------------------------------------------------------	----------------------------------------------------------------------
16	Name: Matthias Blume (blume (at) tti - c (dot) org)	Name: Matthias Blume (blume (at) tti - c (dot) org)
17	Date: 2004/01/26	Date: 2004/11/09 17:50:00 CST
18		Tag: blume-20041109-64bit
19		Description:
20
21		Started adding some infrastructure for supporting 64-bit int- and
22		word-types.  (Still in its very early stages.)
23
24		----------------------------------------------------------------------
25		Name: Matthias Blume (blume (at) tti - c (dot) org)
26		Date: 2004/10/28 10:45:00 CDT
27		Tag: Release_110_50
28		Description:
29
30		New working version (110.50).  NEW BOOTFILES!
31
32		=====================
33
34		Also:
35
36		- Changend config/srcarchiveurl from a file just
37		containing the URL string into a file containing
38		shell script code.  The code has access to the $VERSION variable.
39		- Made corresponding changes to config/install.sh and config/unpack.
40		- Default contents of config/srcarchiveurl uses $VERSION and
41		normally does not have to be edited to reflect a version change.
42
43		(As a result, a version change can be done by just editing
44		config/version, the rest is now automatic.)
45
46		----------------------------------------------------------------------
47		Name: Matthias Blume (blume (at) tti - c (dot) org)
48		Date: 2004/10/27 17:50:00 CDT
49		Tag: blume-20041027-btrace-msg
50		Description:
51
52		BackTrace.monitor now also reports the source of the exception that
53		triggered the trace.
54
55		----------------------------------------------------------------------
56		Name: Matthias Blume (blume (at) tti - c (dot) org)
57		Date: 2004/10/27 17:20:00 CDT
58		Tag: blume-20041027-x86-c-calls
59		Description:
60
61		This is the HISTORY entry for two earlier commits, both concerning
62		the x86 c-calls code in MLRISC:
63
64		- added a missing LOAD in the code that deals with struct arguments
65		- made sure the caller does not add the wrong number of bytes to the
66		stack pointer after a call of a function returning a struct
67		(the callee already pops the implicit argument which points to
68		the space reserved for the result)
69
70		----------------------------------------------------------------------
71		Name: Allen Leung (leunga (at) reservoir (dot) com)
72		Date: 2004/10/24 14:00:00 EST
73		Tag: leunga-20041024-x86-gas-fucomip
74		Description:
75
76		John discovered a bug in the syntax of fucomip.
77		The opcodes FU?COMIP? have been changed to
78
79		fu?comip? %st(i), %st
80
81		----------------------------------------------------------------------
82		Name: Matthias Blume (blume (at) tti - c (dot) org)
83		Date: 2004/10/20 15:06:00 CDT
84		Tag: blume-20041020-standalone-backtrace
85		Description:
86
87		Added a mechanism for getting back-trace information from standalone
88		programs.  Here is how it works:
89
90		1. The part of the program from which you want to get backtrace
91		information (usually the whole program) should be wrapped with
92		BackTrace.monitor.  This is a (unit->'a)->'a function, and your
93		main program could be modified from something like
94
95		fun main (pgm, args) = ...
96
97		to
98
99		fun main (pgm, args) = BackTrace.monitor (fn () => ...)
100
101		2. To be able to access BackTrace.monitor, you have to add
102		library $smlnj-tdp/plugins.cm to the .cm file that contains your
103		main function.
104
105		3. Remove all compiled code (i.e., all the .cm/ subdirectories that
106		CM might have created in the past for your project).
107
108		4. Build the system using this command line:
109
110		ml-build -Ctdp.instrument=true \$smlnj-tdp/back-trace.cm \
111		myprog.cm MyProg.main myprog
112
113		instead of the usual
114
115		ml-build myprog.cm MyProg.main myprog
116
117		I changed a library name:
118
119		$/trace-debug-profile.cm --> $smlnj-tdp/plugins.cm
120
121		New libraries:
122
123		$smlnj-tdp/back-trace.cm
124		-- when loaded causes the back-trace plugin to be installed
125		$smlnj-tdp/coverage.cm
126		-- when loaded causes the coverage plugin to be installed
127
128		----------------------------------------------------------------------
129		Name: Matthias Blume (blume (at) tti - c (dot) org)
130		Date: 2004/10/18 16:45:00 CDT
131		Tag: blume-20041018-groupowner
132		Description:
133
134		Added an "obsolete" warning for the "group owner" syntax to CM's parser.
135
136		Eliminated group owner specs from .cm files throughout the source tree.
137
138		----------------------------------------------------------------------
139		Name: Matthias Blume (blume (at) tti - c (dot) org)
140		Date: 2004/10/15 15:45:00 CDT
141		Tag: blume-20041015-coverage
142		Description:
143
144		* Test coverage tool added!
145		* Further reorganization of tracing-, debugging-, and profiling support:
146
147		- moved original BTImp -- now called BackTrace -- into a separate
148		library called $/trace-debug-profile.cm
149		- eliminated all mentions of BTrace from SMLofNJ.Internals
150		- only the instrumentation mechanism is now left in the compiler proper
151		- BackTrace module is a plugin which is NOT plugged in by default
152		- Coverage module is another such plugin
153
154		To get the benefits of any of these plugin modules, the code in
155		question must be compiled with tdp instrumentation turned on.  This
156		can be done by setting SMLofNJ.Internals.TDP.mode to true.
157		(The ref cell is also controlled via the -Ctdp.instrument=... switch.)
158
159		Plugins are selected at link time.  (Pre-compiled instrumented code
160		can be re-loaded with different plugins in effect.)  When an
161		instrumented module is linked, whatever plugins are at that time
162		enabled will come into effect for that module.
163
164		To enable the back-trace plugin, load library $/trace-debug-profile.cm
165		and invoke BackTrace.install() (e.g., from the interactive prompt).
166		To enable the coverage plugin, load the same library and invoke
167		Coverage.install().
168
169		Back-traces are generated automatically on uncaught exceptions and
170		when the code in question explicitly invokes BackTrace.trigger().
171
172		Coverage (and execution frequency-) information must be queried
173		explicitly by calling Coverage.not_covered and Coverage.hot_spots.
174
175		----------------------------------------------------------------------
176		Name: Matthias Blume (blume (at) tti - c (dot) org)
177		Date: 2004/10/14 17:40:00 CDT
178		Tag: blume-20041014-tdp-core
179		Description:
180
181		Snapshot of a significant overhaul of how the trace/debug/profile support
182		is hooked into the system (specifically: Core and SMLofNJ.Internals).
183
184		----------------------------------------------------------------------
185		Name: Matthias Blume (blume (at) tti - c (dot) org)
186		Date: 2004/10/13 16:34:00 CDT
187		Tag: blume-20041013-tdp
188		Description:
189
190		Some rationalization of names:
191
192		structure BTrace -> structure TDPInstrument
193		etc.
194
195		This is is preparation of using the original back-trace
196		instrumentation for other purposes.  "TDP" stands for
197		Trace/Debug/Profile.
198
199		The control flag controlling whether instrumentation is on or off is now
200		registered under a different name, so instead of running sml as
201
202		sml -Cinstrument.btrace-mode=true
203
204		one has to say
205
206		sml -Ctdp.instrument=true
207
208		----------------------------------------------------------------------
209		Name: Matthias Blume (blume (at) tti - c (dot) org)
210		Date: 2004/10/11 16:37:00 CDT
211		Tag: blume-20041011-regions
212		Description:
213
214		Made some minor modifications to elabcore.sml to have source regions
215		be propagated more tightly -- resulting in better (i.e., smaller)
216		regions being reported in error- and debug messages.
217
218		----------------------------------------------------------------------
219		Name: Matthias Blume (blume (at) tti - c (dot) org)
220		Date: 2004/10/08 22:50:00 CDT
221		Tag: blume-20041008-cmkw
222		Description:
223
224		Fixed handling of keywords in .cm files: After seeing "is" the lexer
225		treats subsequent occurrences of "group", "library", "source", "is",
226		"*", and "-" as ordinary identifiers rather than keywords.
227
228		Most seriously, this fixes a problem with CM's "shell" tool.  The tool
229		is supposed to accept a tool argument called "source", but this did
230		not work because of the clash with the keyword.
231
232		----------------------------------------------------------------------
233		Name: Matthias Blume (blume (at) tti - c (dot) org)
234		Date: 2004/10/07 16:00:00 CDT
235		Tag: blume-20041007-cleanup
236		Description:
237
238		Assorted cleanup work:
239
240		- got rid of intstrmap in favor of using the library's
241		hash table implementation
242		- threw out most of the pathnames stuff, as it was not used anyway
243		- simplified tokentable implementation
244		- fixed some minor spelling errors
245
246		----------------------------------------------------------------------
247		Name: Matthias Blume (blume (at) tti - c (dot) org)
248		Date: 2004/10/06 15:15:15 CDT
249		Tag: blume-20041006-handler
250		Description:
251
252		Cleaned up the absyn to reflect the invariant that HANDLE always
253		carries a FNexp as part of the type definition.  This eliminates some
254		superfluous sanity checks at runtime down the road.
255
256		Some minor cleanup of the btrace code.
257
258		----------------------------------------------------------------------
259		Name: Matthias Blume (blume (at) tti - c (dot) org)
260		Date: 2004/10/01 10:20:30 CDT
261		Tag: blume-20041001-slave
262		Description:
263
264		Added hack to make slave mode work in the presence of the version
265		tool.  (Still, since the master does two passes over the code for
266		CMB.make, the release number gets bumped twice when slaves are
267		attached.  I don't know if this is worth fixing...)
268
269		----------------------------------------------------------------------
270		Name: Matthias Blume (blume (at) tti - c (dot) org)
271		Date: 2004/09/30 10:55:00 CDT
272		Tag: blume-20040930-version
273		Description:
274
275		* Moved the "version" magic into its own little library under
276		src/system/smlnj/internal.  This avoids expensive reconstruction of
277		a stable src/compiler/core.cm.
278
279		* At the same time, structure CompilerVersion is now known as
280		structure SMLNJVersion.
281
282		* Arranged for the version tool to NOT kick in when rebuilding the system
283		(makeml -rebuild, fixpt).  Otherwise one would never reach a fixpoint.
284		Also, loading the versiontool does not work when rebuilding the system
285		because CM is not properly initialized at that time.
286
287		----------------------------------------------------------------------
288		Name: Matthias Blume (blume (at) tti - c (dot) org)
289		Date: 2004/09/29 14:00:00 CDT
290		Tag: blume-20040929-autoversion
291		Description:
292
293		Implemented some CM magic to have
294		file src/compiler/TopLevel/main/version.sml
295		generated automagically.
296		The version is taken from two files: config/version and config/release.
297		The first is expected to contain a two-part version number such as 110.49.
298		The second should contain a single number, but it may be missing.
299
300		If the environment variable VERSIONTOOL_BUMP_RELEASE is defined at the
301		time the version tool is loaded (which is the first time you say CMB.make),
302		then the tool will increment the value stored in config/release every
303		time CMB.make is invoked.
304
305		The binfile format is now insensitive to anything beyond the first
306		two components of a version number, so bumping the release does not render
307		binfiles incompatible.  Auto-bumping can be used to keep track of versions
308		during development without invalidating existing binfiles.
309
310		In any case, every CMB.make updates the date information in version.sml.
311		(This is the date that is printed in the banner.)
312
313		----------------------------------------------------------------------
314		Name: Matthias Blume (blume (at) tti - c (dot) org)
315		Date: 2004/09/28 10:53:00 CDT
316		Tag: blume-20040928-controls
317		Description:
318
319		Some cleanup of the controls code.
320
321		----------------------------------------------------------------------
322		Name: Matthias Blume (blume (at) tti - c (dot) org)
323		Date: 2004/09/27 22:08:00 CDT
324		Tag: blume-20040927-controls
325		Description:
326
327		Added two pieces of functionality to the Controls interface:
328
329		1.   val save'restore: 'a control -> unit -> unit
330
331		grabs the current value of the control in stage 1 and restores it
332		in stage 2.
333
334		2.   val set' : 'a control * 'a -> unit -> unit
335
336		stores the given value into the control in stage 2 (i.e., delayed)
337		but does all error checking in stage 1.
338		(This is for string controls that need to do parse their argument --
339		something that might fail.  In some cases, notably in CM, one
340		already knows the intended argument but wants to delay the actual
341		assignment until a time when error recovery would be more difficult.)
342
343		Changed the handling of controls in tool arguments to classes "sml" and
344		"lazysml":
345		- use Controls.save'restore as a more robust way of restoring the
346		old value (in particular: without having to re-parse the string)
347		- use controls to handle the "overload" keyword in the init group
348		(I believe this change actually fixes a long-standing obscure bug.)
349
350		----------------------------------------------------------------------
351		Name: Matthias Blume (blume (at) tti - c (dot) org)
352		Date: 2004/09/27 17:00:00 CDT
353		Tag: blume-20040927-lazysml
354		Description:
355
356		Added a new tool class called "lazysml" to CM's tool chest.  The only
357		difference to "sml" is that compilation is done with Control.lazysml
358		set to true.  A source of class "lazysml" is automatically recognized
359		by a file name suffix of ".lml".
360
361		In addition to the above feature, the original class "sml" now also
362		supports a tool argument "lazy" which has the same effect.  As a
363		result, the following three lines are equivalent:
364
365		foo.sml : lazysml
366		foo.sml : sml (lazy)
367		foo.sml (lazy)
368
369		The setting goes into effect both during parsing and during
370		compilation.  The original setting is restored right after parsing and
371		after compilation, respectively.
372
373		In addition to all the above, there is also a general mechanism to set
374		ANY of the "controls" that are available at the command line via
375		"-C..." on a per-sml-file basis.  The same rules that apply for "lazy"
376		apply as well. (In fact, "lazy" is implemented as a special case of
377		the general mechanism.)
378
379		The .cm file syntax uses a new keyword tool argument called "with".
380		There are several ways of indicating the desired settings:
381
382		foo.sml (with:parser.quotations=true)
383		foo.sml (with:(name:parser.quotations value:true))
384		foo.sml (with:(name:name1 value:value1 name:name2 value:value2 ...))
385		foo.sml (with:(name1=value1 name2=value2 ...))
386		foo.sml (with:(name1=value1 name:name2 value:value2 name3=value3 ...))
387
388		etc.
389
390		Another possible abbreviation is to leave out the =v or value:v part
391		if the name refers to a boolean control (in which case the value is
392		taken to be true).  Thus, one could get lazy sml also by saying:
393
394		foo.sml (with:parser.lazy-keyword=true)
395		foo.sml (with:parser.lazy-keyword)
396		foo.sml (with:(name:parser.lazy-keyword value:true))
397		foo.sml (with:(name:parser.lazy-keyword))
398
399		----------------------------------------------------------------------
400		Name: Matthias Blume (blume (at) tti - c (dot) org)
401		Date: 2004/09/24 16:22:00 CDT
402		Tag: blume-20040924-ppc-long-branch
403		Description:
404
405		Turned message about "emiting long form of branch" off by default.
406		Added a control flag to turn it back on when desired.
407
408		----------------------------------------------------------------------
409		Name: Matthias Blume (blume (at) tti - c (dot) org)
410		Date: 2004/09/24 16:05:00 CDT
411		Tag: blume-20040924-rounding
412		Description:
413
414		Applied patch for setting rounding modes under Mac OS X.  Thanks to
415		Melissa O'Neill for providing the code!
416
417		----------------------------------------------------------------------
418		Name: Matthias Blume (blume (at) tti - c (dot) org)
419		Date: 2004/09/23 17:30:00 CDT
420		Tag: blume-20040923-envvars
421		Description:
422
423		1. Changed definition of type ControlRegistry.registry_tree to
424		include control_info (i.e., the name of the controlling environment
425		variable).
426
427		2. Added command-line flags -e and -E to print the names of environment
428		variables that can be used to control internal settings.
429		(This uses the new API mentioned in 1.)
430
431		----------------------------------------------------------------------
432		Name: Matthias Blume (blume (at) tti - c (dot) org)
433		Date: 2004/09/13 16:50:00 CDT
434		Tag: Release_110_49
435		Description:
436
437		New working version (110.49).  NEW BOOTFILES!
438
439		----------------------------------------------------------------------
440		Name: Matthias Blume
441		Date: 2004/09/13 16:20:00 CDT
442		Tag: blume-20040913-config-mlrisc
443		Description:
444
445		Put target "mlrisc" back into the default list.
446		(There is no harm in having it, and some users have expressed their
447		wish to have "mlrisc" included by default.)
448
449		----------------------------------------------------------------------
450		Name: John Reppy
451		Date: 2004/09/13
452		Tag: jhr-20040913-signals
453		Description:
454
455		Fixed the signal masking code to properly nest mask/unmask operations
456		on a per-signal basis.
457
458		----------------------------------------------------------------------
459		Name: Matthias Blume (blume (at) tti - c (dot) org)
460		Date: 2004/09/08 13:20:00 CDT
461		Tag: blume-20040908-heap-magic
462		Description:
463
464		Bumped the heap macig number to 0x09082004 to account for the changed
465		layout of the ML frame under MacOS X.
466
467		----------------------------------------------------------------------
468		Name: Allen Leung (leunga (at) reservoir (dot) com)
469		Date: 2004/09/03 11:26:00 EST
470		Tag: leunga-20040903-cygwin-install
471		Description:
472
473		Added a patch to _arch-n-opsys to enable the Cygwin runtime.
474		The Cygwin runtime is turned on by setting the environment
475		variable SMLNJ_CYGWIN_RUNTIME to 1.
476
477		----------------------------------------------------------------------
478		Name: Matthias Blume (blume (at) tti - c (dot) org)
479		Date: 2004/08/31 17:15:00 CDT
480		Tag: blume-20040831-core
481		Description:
482
483		Added some exports to src/compiler/core.cm upon request by J. Joemann.
484
485		----------------------------------------------------------------------
486		Name: Matthias Blume (blume (at) tti - c (dot) org)
487		Date: 2004/08/30 17:55:00 CDT
488		Tag: blume-20040830-installer
489		Description:
490
491		Upon request by Johannes Joemann:
492
493		- improved ML code of installer to fall back to coping when renaming
494		fails (i.e., when source and target are on different file systems);
495		the code compiles but has yet to be tested in anger
496		- removed mlrisc from list of default targets (config/targets)
497
498		----------------------------------------------------------------------
499		Name: Matthias Blume (blume (at) tti - c (dot) org)
500		Date: 2004/08/27 17:20:00 CDT
501		Tag: blume-20040827-ptreql
502		Description:
503
504		Added ptreql primop to structure InlineT (upon request from Larry
505		Paulson).
506
507		----------------------------------------------------------------------
508		Name: Allen Leung (leunga (at) reservoir (dot) com)
509		Date: 2004/08/15 21:21:00 EST
510		Tag: leunga-110_48-udgraph
511		Description:
512
513		Another bug fix from Carl Hauser:
514
515		diff /net/niflab/smlnj48/src/MLRISC/graphs/udgraph.sml udgraph.sml
516		> 48c48
517		> <              | rmv((e as (k,_))::es,L) = rmv(es,if k = i then es else
518		> e::L)
519		> ---
520		> >              | rmv((e as (k,_))::es,L) = rmv(es,if k = i then L else e::L)
521		> Without this, any deletion of an edge in an undirected graph does severe
522		> violence to the graph.
523
524		----------------------------------------------------------------------
525		Name: Allen Leung (leunga (at) reservoir (dot) com)
526		Date: 2004/08/10 23:35:00 EST
527		Tag: leunga-110_48-ppc
528		Description:
529
530		The IBM/MacOS syntax switch on PPC was incorrectly swapped.
531
532		----------------------------------------------------------------------
533		Name: Matthias Blume (blume (at) tti - c (dot) org)
534		Date: 2004/08/10 12:00:00 CDT
535		Tag: Release_110_48
536		Description:
537
538		New working version (110.48).  NEW BOOTFILES!
539
540		----------------------------------------------------------------------
541		Name: Allen Leung (leunga (at) reservoir (dot) com)
542		Date: 2004/08/09 12:21:00 EST
543		Tag: leunga-110_47-dijsktra
544		Description:
545
546		Bug fix from Carl Hauser:
547
548		single_source_shortest_paths in dijkstra.sml was observed to get wrong
549		answers (by comparing to single_source_shortest_paths in bellman-ford.sml).
550
551		The problem is that following the expression A.update(dist,s,Num.zero)
552		it is necessary to update the priority queue using Q.decreaseWeight(Q,s).
553
554		----------------------------------------------------------------------
555		Name: Matthias Blume (blume (at) tti - c (dot) org)
556		Date: 2004/08/06 18:10:00 CDT
557		Tag: blume-20040806-cmdline
558		Description:
559
560		Fiddled with handling of command-line options:
561
562		* sml now quits after processing the command line
563		if -H, -S, -h<n>, or -s<n> appears as the last
564		command-line argument
565		* a new option -q terminates the session when encountered on
566		the command line; subsequent arguments will be ignored
567		* bug fixes: short (erroneous) arguments are no longer ignored
568		completely
569
570		----------------------------------------------------------------------
571		Name: Allen Leung (leunga (at) reservoir (dot) com)
572		Date: 2004/08/04 18:17:00 EST
573		Tag: leunga-110_47-ppc-ibm-asm
574		Description:
575
576		- Added minimal IBM assembly syntax support for PowerPC.
577
578		- Cygwin: manually changed the file cygwin.def.  Some exported symbols have
579		been altered in the runtime.  We need an automatic way to keep the file
580		in sync.
581
582		----------------------------------------------------------------------
583		Name: Matthias Blume (blume (at) tti - c (dot) org)
584		Date: 2004/08/04 14:00:00 CDT
585		Tag: Release_110_47
586		Description:
587
588		New working version (110.47).  NEW BOOTFILES!
589
590		----------------------------------------------------------------------
591		Name: Matthias Blume (blume (at) tti - c (dot) org)
592		Date: 2004/08/03 14:25:00 CDT
593		Tag: blume-20040803-callingconv
594		Description:
595
596		Added low-level support for choosing C calling conventions by
597		twiddling the type of rawccall.  (See
598		src/compiler/Semant/types/cproto.sml for details.)
599
600		----------------------------------------------------------------------
601		Name: Matthias Blume (blume (at) tti - c (dot) org)
602		Date: 2004/08/02 15:55:00 CDT
603		Tag: blume-20040802-backout
604		Description:
605
606		Backed out of change to win32-filesys.c.  The earlier patch to
607		get_file_time caused CM to produce files with the wrong time stamp.
608
609		----------------------------------------------------------------------
610		Name: Matthias Blume (blume (at) tti - c (dot) org)
611		Date: 2004/08/02 14:45:00 CDT
612		Tag: blume-20040802-nlffi-win32
613		Description:
614
615		Added NLFFI support for Win32, adapted from a patch provided by David
616		Hansel.  This is currently completely untested.  Also, the issue
617		concerning stdcall vs. ccall is still unresolved.
618
619		----------------------------------------------------------------------
620		Name: Matthias Blume (blume (at) tti - c (dot) org)
621		Date: 2004/07/30 17:55:00 CDT
622		Tag: blume-20040730-various
623		Description:
624
625		Gearing up towards 110.47...
626
627		- various minor bugfixes to ml-nlffigen
628		- a beginning of a manual for nlffi
629
630		- eliminated 'export name=value' in config/install.sh as this does
631		not work with certain versions of /bin/sh
632		(Thanks to David King at Motorola for catching this.)
633
634		- several bugfixes provided or suggested by David Hansel at Reactive Systems:
635		- added a test for tm==NULL to gmtime.c and localtime.c
636		- applied patch for incorrect GetFileTime under win32
637		- toSeconds -> toMilliseconds in Win32/win32-process.sml
638
639		----------------------------------------------------------------------
640		Name: Matthias Blume (blume (at) tti - c (dot) org)
641		Date: 2004/07/21 18:20:00 CDT
642		Tag: blume-20040721-nlffigen
643		Description:
644
645		- Fixed minor issue in ml-nlffigen: Now generate structure T_foo
646		for a typedef to an incomplete type, but leave out the "typ" member.
647		(This is just for consistency.)
648		- Started to produce what is supposed to become better (i.e., comprehensive)
649		documentation of what ml-nlffigen does and produces.
650
651		----------------------------------------------------------------------
652		Name: Matthias Blume (blume (at) tti - c (dot) org)
653		Date: 2004/07/14 16:25:00 CDT
654		Tag: blume-20040714-union
655		Description:
656
657		Added C_UNION to c-calls/c-types.sml and updated the machinery
658		(ml-nlffigen, cproto.sml) that conveys C function interface
659		information to the code generator.
660
661		However, the actual architecture-specific implementation of function
662		arguments and results that are C unions is still not implemented.
663
664		----------------------------------------------------------------------
665		Name: Allen Leung (leunga (at) reservoir (dot) com)
666		Date: 2004/07/14 14:38:00 EST
667		Tag: leunga-110_46_1-ppc-lwzu
668		Description:
669
670		Added these instructions to the PowerPC architecture:
671		LBZU(X), LHZU(X), LWZU(X),
672		STWU(X), STFDU, STFSU
673
674		etc...
675
676		Note: I haven't added their instruction encoding into the description.
677
678		----------------------------------------------------------------------
679		Name: Allen Leung (leunga (at) reservoir (dot) com)
680		Date: 2004/07/13 15:04:00 EST
681		Tag: leunga-110_46_1-ppc-lwarx
682		Description:
683
684		Added the two instructions LWARX and STWCX to the PowerPC
685		instruction set.
686
687		A (untested) rewrite of loop-structure.sml.  The old version
688		is completely broken.
689
690		----------------------------------------------------------------------
691		Name: Matthias Blume (blume (at) tti - c (dot) org)
692		Date: 2004/07/13 13:50:00 CDT
693		Tag: blume-20040713-nlffi
694		Description:
695
696		- use paramAlloc to report c-calls with too many arguments
697		(for PPC version where parameter area is pre-allocated)
698		- added ccall_maxargspace to machspec (to implement the above)
699		- made "make" commend in CM's "make" tool configurable
700		- added option (default: on) for passing the name of the SML/NJ's "bin"
701		directory to "make"; the call looks like this:
702
703		make <options> SMLNJ_BINDIR=<dir> <target>
704
705		This can be used by the Makefile to, e.g., pick the "right" version
706		of ml-nlffigen.
707		- minor code tweaks
708
709		----------------------------------------------------------------------
710		Name: Matthias Blume (blume (at) tti - c (dot) org)
711		Date: 2004/07/12 22:50:00 CDT
712		Tag: blume-110_46_1-macosx-nlffi
713		Description:
714
715		NLFFI under Mac OS X now working (sort of).  This is largely untested,
716		though.
717
718		Note:  1. You have to make a new, clean build of the runtime system.
719		2. There are new BOOTFILES, you have to use them!
720		(Doing the bootstrap process yourself would be *very* painful!
721		If you absolutely have to do it, build the system under
722		a different architecture and then cross-compile.)
723
724		Version bumped to 110.46.1 to account for runtime data format changes.
725
726		----------------------------------------------------------------------
727		Name: Matthias Blume (blume (at) tti - c (dot) org)
728		Date: 2004/06/18 14:30:00 CDT
729		Tag: blume-20040618-unix
730		Description:
731
732		Changed the implementation of structure Unix so that the same stream
733		is returned every time one of the {text,bin}{In,Out}streamOf functions
734		is invoked on the same proc.  This is not what the spec currently
735		says -- although IMO it arguably should.  (See discussion below.)
736
737		----------------------------------------------------------------------
738		Name: Matthias Blume (blume (at) tti - c (dot) org)
739		Date: 2004/06/17 18:15:00 CDT
740		Tag: Release_110_46
741		Description:
742
743		New working version (110.46).  NEW BOOTFILES!
744
745		----------------------------------------------------------------------
746		Name: Matthias Blume (blume (at) tti - c (dot) org)
747		Date: 2004/06/17 17:20:00 CDT
748		Tag: blume-20040617-timer-unix
749		Description:
750
751		Changed the interface of structures Timer and Unix to match the most
752		recent Basis spec.
753
754		In the case of Unix there still seems to be an open/weird issue:
755
756		The {text,bin}{In,Out}streamOf functions are supposed to create
757		fresh streams whenever they are called -- as opposed to have them
758		return the same stream every time.  This design is supposed to
759		prevent space leaks caused by proc values hanging on to streams.
760
761		The reap function, on the other hand, is supposed to close the
762		streams.  This cannot be done without having a handle on the
763		stream in proc after all...
764
765		I took the liberty to implement the following stopgap solution:
766
767		The proc value hangs on to the most recently created stream(s).
768		Reap closes those.  If either or both of the two streams hadn't
769		been created at all yet, then reap will close the corresponding
770		file descriptors directly.
771
772		PS: I don't understand the original space leak argument anymore.  If
773		a proc hangs on to the imperative stream, then I/O operations on those
774		will advance the state of the cached stream and avoid the space leak.
775
776		----------------------------------------------------------------------
777		Name: Matthias Blume (blume (at) tti - c (dot) org)
778		Date: 2004/05/28 16:45:00 CDT
779		Tag: blume-20040528-basis
780		Description:
781
782		Added signature PACK_REAL and exported functor PrimIO.
783
784		----------------------------------------------------------------------
785		Name: Matthias Blume (blume (at) tti - c (dot) org)
786		Date: 2004/05/25 16:00:00 CDT
787		Tag: blume-20040525-group-owner
788		Description:
789
790		CM now ignores (but still accepts) the "owner" information in group
791		descriptions.  The owner of a group is its next enclosing
792		library. Each group must have a unique owner.  (There is a virtual
793		"toplevel" library that owns groups which are not nested within a real
794		library.)  Previously, each group had to explicitly declare its owner,
795		and CM would check that such a declaration is correct.  The new scheme
796		is to have CM check that for each group there is precisely one owning
797		library.
798
799		The advantage of the new scheme is that the programmer no longer needs
800		to maintain the somewhat annoying owner information.  The downside is
801		that CM cannot enforce the ownership rule across multiple runs of
802		CM.make.  Fortunately, enclosing the same group in two different
803		libraries A and B which are not part of the same program does not
804		cause real problems.
805
806		----------------------------------------------------------------------
807		Name: Matthias Blume (blume (at) tti - c (dot) org)
808		Date: 2004/05/20 16:00:00 CDT
809		Tag: blume-20040520-win32
810		Description:
811
812		Made the win32 version work again.  (Strangely, a misplaced comma had
813		slipped into win32-process.c which prevented the runtime from being
814		compiled correctly.)
815
816		Also, included a minor addition to ml-build.bat analogous to what was
817		done in blume-20040519-ml-build.
818
819		----------------------------------------------------------------------
820		Name: Matthias Blume (blume (at) tti - c (dot) org)
821		Date: 2004/05/19 22:10:00 CDT
822		Tag: blume-20040519-ml-build
823		Description:
824
825		Arranged for ml-build to clean up after itself a little bit better.
826		The script generates a temporary SML source file and compiles it using
827		CM, so CM generates metadata (GUID, SKEL, objectfile) for it.  It now
828		gets rid of those at the end, so they don't accumulate under .cm.
829
830		This required a minor change to install.sh because the name of the
831		metadata directory (default: .cm) is actually configurable at
832		installation time.
833
834		----------------------------------------------------------------------
835		Name: Matthias Blume (blume (at) tti - c (dot) org)
836		Date: 2004/05/18 15:50:00 CDT
837		Tag: blume-20040518-mkreader
838		Description:
839
840		Added Posix.IO.mk{Bin,Text}{Reader,Writer} by lifting their respective
841		implementations from internal modules PosixBinPrimIO and PosixTextPrimIO.
842
843		----------------------------------------------------------------------
844		Name: Matthias Blume (blume (at) tti - c (dot) org)
845		Date: 2004/05/11 14:35:00 CDT
846		Tag: blume-20040511-win32sock
847		Description:
848
849		Added previously missing support for many socket-related functions
850		under win32.  Thanks to David Hansel <hansel@reactive-systems.com>
851		for the voluminous patch!
852
853		(I have not tested this patch under win32 yet.)
854
855		Here is David's e-mail:
856
857		Hi,
858
859		Attached to this email you find a diff against sml/nj 110.45
860		that will enable socket support under Windows.
861
862		To apply the patch (using unix or cygwin)
863		1) gunzip runtime.diff.gz
864		2) "cd" into "src/runtime" in the source tree of a fresh
865		110.45 installation.
866		3) patch -p 1 < [your/path/to]runtime.diff
867
868		The code compiles fine but has NOT yet been extensively tested.
869		I only ran a few tests for basic socket client functionality
870		(which worked fine).  Especially the functions that use ioctl
871		are not tested at all and might not work (see below).
872
873		I implemented this since we want to move to a newer version of sml/nj
874		but need socket support in order to use it.  This is the first time I
875		even had a look at the sml/nj source,  so please review my changes
876		before making this part of the distribution!  Here are a few issues
877		that I think might be better for someone to solve who is more
878		familiar with the sml/nj source (and socket programming):
879
880		- getnetbyaddr.c and getnetbyname.c will raise a "not implemented"
881		exception since I could not figure out what the windows equivalent
882		of these functions is
883
884		- In sockets-osdep.h there are a some #include statements that are
885		only used in a few files that include sockets-osdep.h
886
887		- In smlnj-sock-lib.c, function init_fn() calls WSAStartup() but
888		does not process its return value since I don't know how to
889		report an error upwards.
890
891		- It would probably be good to have a call to WSACleanup() when
892		the library is unloaded (if there is such a possibility).
893		Otherwise I think Windows will take care of this automatically
894		when the process finishes.
895
896		- I used ioctlsocket() as a replacement for ioctl() but I have
897		no idea if that is actually the proper replacement on Windows.
898
899		- All these issues are marked in the code by "FIXME" comments.
900
901		We use sml/nj extensively in our products and are quite happy
902		with it.  I hope this contribution will help you.
903
904		Keep up the good work!
905
906		David
907
908		----------------------------------------------------------------------
909		Name: Matthias Blume (blume (at) tti - c (dot) org)
910		Date: 2004/05/11 14:20:00 CDT
911		Tag: blume-20040511-installml
912		Description:
913
914		Fixed two bugs in installml script.  (Thanks to Vesa A. Norrman for
915		the patch.)
916
917		----------------------------------------------------------------------
918		Name: Matthias Blume (blume (at) tti - c (dot) org)
919		Date: 2004/05/11 14:05:00 CDT
920		Tag: blume-20040511-nlffi-netbsd
921		Description:
922
923		Added support for nlffi under netbsd.  (Thanks to Vesa A. Norrman for
924		the patch.)
925
926		----------------------------------------------------------------------
927		Name: Matthias Blume (blume (at) tti - c (dot) org)
928		Date: 2004/05/11 12:05:00 CDT
929		Tag: blume-20040511-exports
930		Description:
931
932		As per request by Adam Chlipala <adam@hcoop.net>, extended various
933		export lists in compiler-related .cm-files.
934
935		----------------------------------------------------------------------
936		Name: Matthias Blume (blume (at) tti - c (dot) org)
937		Date: 2004/05/11 11:35:00 CDT
938		Tag: blume-20040511-allsource
939		Description:
940
941		The installer now honors the "src-smlnj" target again, although its meaning
942		has changed from "all sources required for the compiler" to "all sources
943		the installer knows about".  In other words, if you enable "src-smlnj"
944		in the "targets" file, then the installer will pull in sources for
945		everything.  (Notice that this refers to source code only.  Compiled
946		code is still only installed for modules that were requested explicitly
947		or which are required for other modules that were requested explicitly.)
948
949		----------------------------------------------------------------------
950		Name: Matthias Blume (blume (at) tti - c (dot) org)
951		Date: 2004/04/23 17:40:00 CDT
952		Tag: blume-20040423-ieee-scan
953		Description:
954
955		Fixed IEEEReal.scan (and .fromString) so that if there is an overflow
956		in the exponent calculation we get INF or ZERO (depending on the mantissa
957		and the sign of the exponent).
958
959		----------------------------------------------------------------------
960		Name: Matthias Blume (blume (at) tti - c (dot) org)
961		Date: 2004/04/23 10:40:00 CDT
962		Tag: blume-20040423-ml-build
963		Description:
964
965		The ml-build script now terminates with a non-0 status when something
966		goes wrong.
967
968		----------------------------------------------------------------------
969		Name: Matthias Blume (blume (at) tti - c (dot) org)
970		Date: 2004/04/22 16:35:00 CDT
971		Tag: blume-20040422-Option
972		Description:
973
974		Made exception Option to be the same as exception Option.Option
975		(as it should be).
976
977		----------------------------------------------------------------------
978		Name: Allen Leung (leunga (at) reservoir (dot) com)
979		Date: 2004/03/19 14:40:00 EST
980		Tag: leunga-20040319-cygwin-nlffi
981		Description:
982
983		Fixed the runtime so that ml-nlffi-lib runs on the cygwin version
984		of SML/NJ.  The problem is that
985
986		lib = dlopen(NULL, ...)
987		f   = dlsym(lib, "malloc");
988
989		does not work on Windows unless we explicitly export symbols
990		such as 'malloc' during linking.  We fixed this by explicitly
991		exporting the required symbols with the magic gcc incantation:
992
993		-Wl,--export-all cygwin.def
994
995		where cygwin.def is a file containing all the symbols that
996		we wish to export.
997
998		I suspect this is a Windows problem and we'll have to
999		do the same (somehow with windows compilers) when
1000		we build the native win32 version with the system
1001		calls LoadLibrary/GetProcAddress.
1002
1003		----------------------------------------------------------------------
1004		Name: Matthias Blume (blume (at) tti - c (dot) org)
1005		Date: 2004/03/04 16:35:00 CST
1006		Tag: blume-20040304-intinf-fmt
1007		Description:
1008
1009		Fixed problem with IntInf.fmt (sign would show up on the right instead
1010		of on the left for BIN, OCT, and HEX).
1011
1012		----------------------------------------------------------------------
1013		Name: Matthias Blume (blume (at) tti - c (dot) org)
1014		Date: 2004/03/04 11:25:00 CST
1015		Tag: blume-20040304-symlinks
1016		Description:
1017
1018		Fixed problem with installer script (unix only) where bin/ml-yacc and
1019		friends pointed (via symlinks) to absolute locations instead of just
1020		.run-sml.  This was reported by Vesa A Norrman.
1021
1022		----------------------------------------------------------------------
1023		Name: Matthias Blume (blume (at) tti - c (dot) org)
1024		Date: 2004/02/13 14:50:00 CST
1025		Tag: Release_110_45
1026		Description:
1027
1028		New working version (110.45). New bootfiles.
1029
1030		----------------------------------------------------------------------
1031		Name: Matthias Blume (blume (at) tti - c (dot) org)
1032		Date: 2004/01/26 15:15:15 CST
1033		Tag: blume-20040126-toplevel
1034		Description:
1035
1036		Improved handling of exceptions at the interactive toplevel.
1037
1038		----------------------------------------------------------------------
1039		Name: Matthias Blume (blume (at) tti - c (dot) org)
1040		Date: 2004/01/26 11:25:00
1041	Tag: blume-20040126-app	Tag: blume-20040126-app
1042	Description:	Description:
1043

---

Colored Diff Long Colored Diff Unidiff Context Diff Side by Side

Legend: Removed from v.1447   changed lines   Added in v.1682

---

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