Diff of /sml/trunk/HISTORY

Parent Directory | Revision Log | Patch

	revision 839, Thu Jun 7 20:28:44 2001 UTC	revision 874, Mon Jul 9 20:01:40 2001 UTC
#	Line 14	Line 14
14
15	----------------------------------------------------------------------	----------------------------------------------------------------------
16	Name: Matthias Blume	Name: Matthias Blume
17		Date: 2001/07/09 16:00:00 EDT
18		Tag: blume-20010709-more-varargs
19		Description:
20
21		I changed the handling of varargs in ml-nlffigen again:
22		The ellipsis ... will now simply be ignored (with an accompanying warning).
23
24		The immediate effect is that you can actually call a varargs function
25		from ML -- but you can't actually supply any arguments beyond the ones
26		specified explicitly.  (For example, you can call printf with its format
27		string, but you cannot pass additional arguments.)
28
29		This behavior is only marginally more useful than the one before, but
30		it has the advantage that a function or, more importantly, a function
31		type never gets dropped on the floor, thus avoiding follow-up problems with
32		other types that refer to the offending one.
33
34		----------------------------------------------------------------------
35		Name: Matthias Blume
36		Date: 2001/07/09 11:25:00 EDT
37		Tag: blume-20010709-varargs
38		Description:
39
40		1. ckit-lib.cm now exports structure Error
41		2. ml-nlffigen reports occurences of "..." (i.e., varargs function types)
42		with a warning accompanied by a source location.  Moreover, it
43		merely skips the offending function or type and proceeds with the
44		rest of its work.u  As a result, one can safely feed C code containing
45		"..." to ml-nlffigen.
46		3. There are some internal improvements to CM, providing slightly
47		more general string substitutions in the tools subsystem.
48
49		----------------------------------------------------------------------
50		Name: Matthias Blume
51		Date: 2001/06/27 15:10:00 EDT
52		Tag: blume-20010627-concur
53		Description:
54
55		Fixed a small bug in CM's handling of parallel compilation.
56		(You could observe the bug by Control-C-interrupting an ordinary
57		CMB.make or CM.stabilize and then attaching some compile servers.
58		The result was that all of a sudden the previously interrupted
59		compilation would continue on its own.  This was because of
60		an over-optimization: CM did not bother to clean out certain queues
61		when no servers were attached "anyway", resulting in the contents
62		of these queues to grab control when new servers did get attached.)
63
64		There is also another minor update to the CM manual.
65
66		----------------------------------------------------------------------
67		Name: Matthias Blume
68		Date: 2001/06/26 16:15:00 EDT
69		Tag: blume-20010626-cmdoc
70		Description:
71
72		Minor typo fixed in CM manual (syntax diagram for libraries).
73
74		----------------------------------------------------------------------
75		Name: Matthias Blume
76		Date: 2001/06/25 22:55:00 EDT
77		Tag: blume-20010625-x86pc
78		Description:
79
80		Fixed a nasty bug in the X86 assembly code that caused signal
81		handlers to fail (crash) randomly.
82
83		----------------------------------------------------------------------
84		Name: Matthias Blume
85		Date: 2001/06/25 12:05:00 EDT
86		Tag: blume-20010625-nlffigen
87		Description:
88
89		This update fixes a number of minor bugs in ml-nlffigen as reported by
90		Nick Carter <nbc@andrew.cmu.edu>.
91
92		1. Silly but ok typedefs of the form "typedef void myvoid;" are now accepted.
93		2. Default names for generated files are now derived from the name of
94		the C file *without its directory*.  In particular, this causes generated
95		files to be placed locally even if the C file is in some system directory.
96		3. Default names for generated signatures and structures are also derived
97		from the C file name without its directory.  This avoids silly things
98		like "structure GL/GL".
99		(Other silly names are still possible because ml-nlffigen does not do
100		a thorough check of whether generated names are legal ML identifiers.
101		When in doubt, use command line arguments to force particular names.)
102
103		----------------------------------------------------------------------
104		Name: Matthias Blume
105		Date: 2001/06/21 12:25:00 EDT
106		Tag: blume-20010621-eXene
107		Description:
108
109		eXene now compiles and (sort of) works again.
110
111		The library name (for version > 110.33) is $/eXene.cm.
112
113		I also added an new example in src/eXene/examples/nbody.  See the
114		README file there for details.
115
116		----------------------------------------------------------------------
117		Name: Matthias Blume
118		Date: 2001/06/20 16:40:00 EDT
119		Tag: blume-20010620-cml
120		Description:
121
122		CML now compiles and works again.
123
124		Libraries (for version > 110.33):
125
126		$cml/cml.cm            Main CML library.
127		$cml/basis.cm          CML's version of $/basis.cm.
128		$cml/cml-internal.cm   Internal helper library.
129		$cml/core-cml.cm       Internal helper library.
130		$cml-lib/trace-cml.cm  Tracing facility.
131		$cml-lib/smlnj-lib.cm  CML's version of $/smlnj-lib.cm
132
133		The installer (config/install.sh) has been taught how to properly
134		install this stuff.
135
136		----------------------------------------------------------------------
137		Name: Matthias Blume
138		Date: 2001/06/19 17:55:00 EDT
139		Tag: blume-20010619-instantiate
140		Description:
141
142		This un-breaks the fix for bug 1432.
143		(The bug was originally fixed in 110.9 but I broke it again some
144		time after that.)
145
146		----------------------------------------------------------------------
147		Name: Matthias Blume
148		Date: 2001/06/19 17:25:00 EDT
149		Tag: blume-20010619-signals
150		Description:
151
152		This should (hopefully) fix the long-standing signal handling bug.
153		(The runtime system was constructing a continuation record with an
154		incorrect descriptor which would cause the GC to drop data on the floor...)
155
156		----------------------------------------------------------------------
157		Name: Matthias Blume
158		Date: 2001/06/15 15:05:00 EDT
159		Tag: blume-20010615-moresparc
160		Description:
161
162		Here is a short late-hour update related to Sparc c-calls:
163
164		-- made handling of double-word arguments a bit smarter
165
166		-- instruction selection phase tries to collapse certain clumsily
167		constructed ML-Trees; typical example:
168
169		ADD(ty,ADD(_,e,LI d1),LI d2)  ->  ADD(ty,e,LI(d1+d2))
170
171		This currently has no further impact on SML/NJ since mlriscGen does
172		not seem to generate such patterns in the first place, and c-calls
173		(which did generate them in the beginning) has meanwhile been fixed
174		so as to avoid them as well.
175
176		----------------------------------------------------------------------
177		Name: Matthias Blume
178		Date: 2001/06/15 15:05:00 EDT
179		Tag: blume-20010615-sparc
180		Description:
181
182		The purpose of this update is to provide an implementation of NLFFI
183		on Sparc machines.
184
185		Here are the changes in detail:
186
187		* src/MLRISC/sparc/c-calls/sparc-c-calls.sml is a new file containing
188		the Sparc implementation of the c-calls API.
189		* The Sparc backend of SML/NJ has been modified to uniformely use %fp
190		for accessing the ML frame.  Thus, we have a real frame pointer and
191		can freely modify %sp without need for an omit-frame-ptr phase.
192		The vfp logic in src/compiler/CodeGen/* has been changed to accomodate
193		this case.
194		* ml-nlffigen has been taught to produce code for different architectures
195		and calling conventions.
196		* In a way similar to what was done in the x86 case, the Sparc
197		backend uses its own specific extension to mltree.  (For example,
198		it needs to be able to generate UNIMP instructions which are part
199		of the calling convention.)
200		* ml-nlffi-lib was reorganized to make it more modular (in particular,
201		to make it easier to plug in new machine- and os-dependent parts).
202
203		There are some other fairly unrelated bug fixes and cleanups as well:
204
205		* I further hacked the .cm files for MLRISC tools (like MDLGen) so
206		that they properly share their libraries with existing SML/NJ libraries.
207		* I fixed a minor cosmetic bug in CM, supressing certain spurious
208		follow-up error messages.
209		* Updates to CM/CMB documentation.
210
211		TODO items:
212
213		* MLRISC should use a different register as its asmTemp on the Sparc.
214		(The current %o2 is a really bad choice because it is part of the
215		calling conventions, so things might interfere in unexpected ways.)
216
217		----------------------------------------------------------------------
218		Name: Matthias Blume
219	Date: 2001/06/07	Date: 2001/06/07
220	Tag: blume-20010607-calls	Tag: blume-20010607-calls
221	Description:	Description:

---

Colored Diff Long Colored Diff Unidiff Context Diff Side by Side

Legend: Removed from v.839   changed lines   Added in v.874

---

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