mit-scheme.git
30 years agoRemoved old flowgraph code.
Stephen Adams [Wed, 26 Apr 1995 01:53:25 +0000 (01:53 +0000)]
Removed old flowgraph code.

30 years agoInitial revision
Stephen Adams [Wed, 26 Apr 1995 01:51:52 +0000 (01:51 +0000)]
Initial revision

30 years agoAdded SET-GC-NOTIFICATION!
Stephen Adams [Mon, 24 Apr 1995 23:22:30 +0000 (23:22 +0000)]
Added SET-GC-NOTIFICATION!

30 years agoTweaked error checking code to make the integrated portion smaller.
Stephen Adams [Mon, 24 Apr 1995 23:19:23 +0000 (23:19 +0000)]
Tweaked error checking code to make the integrated portion smaller.

30 years agoFixed TRANSFORM/PACKAGE to always return UNSPECIFIC, which ensures
Stephen Adams [Mon, 24 Apr 1995 22:30:44 +0000 (22:30 +0000)]
Fixed TRANSFORM/PACKAGE to always return UNSPECIFIC, which ensures
that the last thing in the generated LET is not a definition.

30 years agoFixed definition ordering bug in macros.scm
Stephen Adams [Mon, 24 Apr 1995 22:15:33 +0000 (22:15 +0000)]
Fixed definition ordering bug in macros.scm

30 years agoFixed bugs in multicell rewrites.
Stephen Adams [Mon, 24 Apr 1995 16:07:34 +0000 (16:07 +0000)]
Fixed bugs in multicell rewrites.

30 years agoChanged to use multicells. Uses singleton multicells unless there are
Stephen Adams [Mon, 24 Apr 1995 16:06:45 +0000 (16:06 +0000)]
Changed to use multicells.  Uses singleton multicells unless there are
many mutated variables, in which case it uses one large multicell.
Ideally it should partition varibales according to their dynamic
extent and create one multcell per extent.

30 years agoWhen doing completing prompt with REQUIRE-MATCH? set true, delay
Chris Hanson [Mon, 24 Apr 1995 01:10:07 +0000 (01:10 +0000)]
When doing completing prompt with REQUIRE-MATCH? set true, delay
running completion procedure on final value.  First try VERIFY-MATCH?
procedure, and if it succeeds, don't do any completion.  This
shouldn't change the behavior since VERIFY-MATCH? must be run anyway,
and if the result is already a match, completion does nothing.

This is helpful when the completion procedure is expensive but
VERIFY-MATCH? is relatively cheap.

30 years agoFix think-o in POP server support.
Chris Hanson [Sun, 23 Apr 1995 06:08:53 +0000 (06:08 +0000)]
Fix think-o in POP server support.

30 years agoFix typos in previous change.
Chris Hanson [Sun, 23 Apr 1995 05:58:14 +0000 (05:58 +0000)]
Fix typos in previous change.

30 years agoAdd day-of-week prefix to string returned by FILE-TIME->STRING. At
Chris Hanson [Sun, 23 Apr 1995 05:53:47 +0000 (05:53 +0000)]
Add day-of-week prefix to string returned by FILE-TIME->STRING.  At
this point, the returned string is fully compliant with RFC 822.
Furthermore, the procedure is not operating-system specific, except
that it calls the procedure LOCAL-TIME-ZONE, which isn't currently
defined for any other operating system.  Offhand, it's not obvious
what's involved in implementing LOCAL-TIME-ZONE for other systems,
except for the more modern unix systems, in which it's trivial.

30 years agoImplement procedures to return strings for weekdays.
Chris Hanson [Sun, 23 Apr 1995 05:43:50 +0000 (05:43 +0000)]
Implement procedures to return strings for weekdays.

30 years agoImplement conversions between file time and universal time.
Chris Hanson [Sun, 23 Apr 1995 05:24:18 +0000 (05:24 +0000)]
Implement conversions between file time and universal time.

30 years agoImplement ENCODE-FILE-TIME.
Chris Hanson [Sun, 23 Apr 1995 05:19:06 +0000 (05:19 +0000)]
Implement ENCODE-FILE-TIME.

30 years agoImplement ENCODE-FILE-TIME.
Chris Hanson [Sun, 23 Apr 1995 05:10:07 +0000 (05:10 +0000)]
Implement ENCODE-FILE-TIME.

30 years agoFix typo.
Chris Hanson [Sun, 23 Apr 1995 04:42:19 +0000 (04:42 +0000)]
Fix typo.

30 years agoUse new primitives to get time zone and daylight savings time
Chris Hanson [Sun, 23 Apr 1995 03:20:38 +0000 (03:20 +0000)]
Use new primitives to get time zone and daylight savings time
information.

30 years agoUpdate to use new primitives that manipulate daylight savings time
Chris Hanson [Sun, 23 Apr 1995 03:19:48 +0000 (03:19 +0000)]
Update to use new primitives that manipulate daylight savings time
information.  Change make-decoded-time to fill in the day-of-week and
daylight-savings-time fields.

30 years agoExtend the time decode/encode primitives to handle daylight savings
Chris Hanson [Sun, 23 Apr 1995 03:16:49 +0000 (03:16 +0000)]
Extend the time decode/encode primitives to handle daylight savings
time information.

30 years agoExtend the time decode/encode primitives to handle daylight savings
Chris Hanson [Sun, 23 Apr 1995 03:04:58 +0000 (03:04 +0000)]
Extend the time decode/encode primitives to handle daylight savings
time information.

30 years agoImplement primitives to get the timezone and daylight-savings
Chris Hanson [Sun, 23 Apr 1995 02:44:25 +0000 (02:44 +0000)]
Implement primitives to get the timezone and daylight-savings
information from the C library.

30 years agoImplement primitives to get the timezone and daylight-savings
Chris Hanson [Sun, 23 Apr 1995 02:42:47 +0000 (02:42 +0000)]
Implement primitives to get the timezone and daylight-savings
information from the C library.

30 years agoImplement some rudimentary time-zone code so that we can produce
Chris Hanson [Sat, 22 Apr 1995 23:42:10 +0000 (23:42 +0000)]
Implement some rudimentary time-zone code so that we can produce
reasonably accurate time zones in RFC-822 dates.

30 years agoImplement some rudimentary time-zone code so that we can produce
Chris Hanson [Sat, 22 Apr 1995 23:37:23 +0000 (23:37 +0000)]
Implement some rudimentary time-zone code so that we can produce
reasonably accurate time zones in RFC-822 dates.

30 years agoFix old typo in OS2_close_std_tqueue that was never noticed because
Chris Hanson [Sat, 22 Apr 1995 21:48:53 +0000 (21:48 +0000)]
Fix old typo in OS2_close_std_tqueue that was never noticed because
this procedure was never called.

30 years agoFix bug: the C-x z command was closing the editor windows and leaving
Chris Hanson [Sat, 22 Apr 1995 21:20:13 +0000 (21:20 +0000)]
Fix bug: the C-x z command was closing the editor windows and leaving
the editor in an unusable state.  This happened because the actions of
initializing and finalizing the PM interface were happening in the
wrong place.

30 years agoLock access to the twin of a qid more carefully. Previously, the code
Chris Hanson [Sat, 22 Apr 1995 21:17:54 +0000 (21:17 +0000)]
Lock access to the twin of a qid more carefully.  Previously, the code
for sending messages was not locked at all, which meant that there
were race conditions that could produce fatal errors.  By locking, we
limit the amount of concurrent communication slightly, but since the
locks are active for only a short time this shouldn't be noticed.

Also implement a new procedure OS2_qid_tqueue, which is used by
inferior threads to obtain and close their associated tqueue objects
when they finish.

30 years agoDon't send broken-pipe errors to the Scheme thread when they are
Chris Hanson [Sat, 22 Apr 1995 21:13:55 +0000 (21:13 +0000)]
Don't send broken-pipe errors to the Scheme thread when they are
detected by the pipe readahead thread.  Recently, this has seemed to
cause a fatal error under certain timing conditions (which
unfortunately didn't hold while running under the debugger).  Although
this fatal error is now probably fixed, rather than run the risk of
finding out, it seems safer to avoid the behavior that seemed to cause
the problem.  Since the value of reporting broken-pipe errors is
minimal at best, this is not a big deal.

30 years agoClose the qid and tqueue associated with a readahead thread when the
Chris Hanson [Sat, 22 Apr 1995 21:07:12 +0000 (21:07 +0000)]
Close the qid and tqueue associated with a readahead thread when the
thread finishes.  This will fix the known bug of exhausting the set of
available qids, and it will also fix a previously unnoticed memory
leak.

30 years agoAdd missing space to error message.
Chris Hanson [Sat, 22 Apr 1995 19:53:18 +0000 (19:53 +0000)]
Add missing space to error message.

30 years agoChange CLOSE-PORT to close the output side of an I/O port before it
Chris Hanson [Fri, 21 Apr 1995 19:58:06 +0000 (19:58 +0000)]
Change CLOSE-PORT to close the output side of an I/O port before it
closes the input side.  This matters when the same file channel is
shared by both sides, and there is buffered data that must be flushed
before the close.

30 years agoMoved some comments to primitive documentation strings.
Stephen Adams [Fri, 21 Apr 1995 04:28:25 +0000 (04:28 +0000)]
Moved some comments to primitive documentation strings.

30 years agoAlways send a readahead ACK when closing a channel with an associated
Chris Hanson [Fri, 21 Apr 1995 00:54:49 +0000 (00:54 +0000)]
Always send a readahead ACK when closing a channel with an associated
reader thread.  This prevents the thread from getting hung in a state
where the channel is closed but the thread is waiting for the ACK.

30 years agoAdded documentation for BIT-SUBSTRING-FIND-NEXT-SET-BIT.
Stephen Adams [Thu, 20 Apr 1995 16:03:24 +0000 (16:03 +0000)]
Added documentation for BIT-SUBSTRING-FIND-NEXT-SET-BIT.

30 years agoFix think-o in previous change.
Chris Hanson [Thu, 20 Apr 1995 03:35:30 +0000 (03:35 +0000)]
Fix think-o in previous change.

30 years agoFixed to accept any top-level expressions, not just trivial- and heap-
Stephen Adams [Thu, 20 Apr 1995 03:26:28 +0000 (03:26 +0000)]
Fixed to accept any top-level expressions, not just trivial- and heap-
closures.  This occurs with the ARITY phase which may convert a
procedure into a expression to construct an arity dispatched
procedure.

30 years agoLots of tinkering but still not finished the search.
Stephen Adams [Thu, 20 Apr 1995 03:24:29 +0000 (03:24 +0000)]
Lots of tinkering but still not finished the search.

30 years agoFixed thinko with previous change.
Stephen Adams [Thu, 20 Apr 1995 03:23:02 +0000 (03:23 +0000)]
Fixed thinko with previous change.

30 years agoChanged %arity-dispatcher-tag to be a knownd `special' symbol and
Stephen Adams [Wed, 19 Apr 1995 22:53:02 +0000 (22:53 +0000)]
Changed %arity-dispatcher-tag to be a knownd `special' symbol and
changed microcode/boot.c to match.  (This should probably be a fixed
bit pattern rather than a symbol).

30 years agoExtend the SET-VARIABLE! and LOCAL-SET-VARIABLE! special forms to take
Chris Hanson [Wed, 19 Apr 1995 02:00:27 +0000 (02:00 +0000)]
Extend the SET-VARIABLE! and LOCAL-SET-VARIABLE! special forms to take
an additional optional argument that specifies the buffer in which the
variable is to be set.

30 years agoProvide mechanism to permit BUFFER-GET to return a value different
Chris Hanson [Wed, 19 Apr 1995 01:56:44 +0000 (01:56 +0000)]
Provide mechanism to permit BUFFER-GET to return a value different
from #F when the buffer lacks a given property.

30 years agoAdd basic text property procedures that accept marks instead of
Chris Hanson [Wed, 19 Apr 1995 01:40:24 +0000 (01:40 +0000)]
Add basic text property procedures that accept marks instead of
indexes.

30 years agoUse new procedure GROUP-INSERT-CHARS! to reduce consing in several
Chris Hanson [Mon, 17 Apr 1995 21:47:40 +0000 (21:47 +0000)]
Use new procedure GROUP-INSERT-CHARS! to reduce consing in several
procedures.  Add new procedures to do padded string insertions without
consing.

30 years agoAdd new procedure GROUP-INSERT-CHARS!.
Chris Hanson [Mon, 17 Apr 1995 21:46:25 +0000 (21:46 +0000)]
Add new procedure GROUP-INSERT-CHARS!.

30 years agoRestructured environment to be more abstract.
Stephen Adams [Mon, 17 Apr 1995 18:48:35 +0000 (18:48 +0000)]
Restructured environment to be more abstract.

30 years agoFixed typo.
Stephen Adams [Mon, 17 Apr 1995 14:39:18 +0000 (14:39 +0000)]
Fixed typo.

30 years ago(append (map f l) e) => (map* e f l)
Stephen Adams [Mon, 17 Apr 1995 03:55:03 +0000 (03:55 +0000)]
(append (map f l) e) => (map* e f l)

30 years agoChanged (append (map f l) e) => (map* e f l)
Stephen Adams [Mon, 17 Apr 1995 03:40:28 +0000 (03:40 +0000)]
Changed  (append (map f l) e) => (map* e f l)

30 years agoAdded rewrites for MULTICELLs. MULTICELLs are rewritten to either
Stephen Adams [Sat, 15 Apr 1995 16:36:27 +0000 (16:36 +0000)]
Added rewrites for MULTICELLs.  MULTICELLs are rewritten to either
cells or vectors.

30 years agoAdded MULTICELLs.
Stephen Adams [Sat, 15 Apr 1995 15:49:38 +0000 (15:49 +0000)]
Added MULTICELLs.

30 years agoFix think-o in previous change.
Chris Hanson [Sat, 15 Apr 1995 06:58:41 +0000 (06:58 +0000)]
Fix think-o in previous change.

30 years agoEliminate unneeded reference to GET-TIME.
Chris Hanson [Sat, 15 Apr 1995 06:48:48 +0000 (06:48 +0000)]
Eliminate unneeded reference to GET-TIME.

30 years agoEliminate unneeded reference to GET-TIME.
Chris Hanson [Sat, 15 Apr 1995 06:41:26 +0000 (06:41 +0000)]
Eliminate unneeded reference to GET-TIME.

30 years agoFix think-o in previous change.
Chris Hanson [Sat, 15 Apr 1995 06:29:04 +0000 (06:29 +0000)]
Fix think-o in previous change.

30 years agoUse new procedure OS/LS-FILE-TIME-STRING.
Chris Hanson [Sat, 15 Apr 1995 06:14:34 +0000 (06:14 +0000)]
Use new procedure OS/LS-FILE-TIME-STRING.

30 years agoImplement new procedure OS/LS-FILE-TIME-STRING.
Chris Hanson [Sat, 15 Apr 1995 06:14:22 +0000 (06:14 +0000)]
Implement new procedure OS/LS-FILE-TIME-STRING.

30 years agoImplement new procedure DECODE-FILE-TIME.
Chris Hanson [Sat, 15 Apr 1995 06:13:17 +0000 (06:13 +0000)]
Implement new procedure DECODE-FILE-TIME.

30 years agoImplement new procedure DECODE-FILE-TIME, and rewrite
Chris Hanson [Sat, 15 Apr 1995 06:12:21 +0000 (06:12 +0000)]
Implement new procedure DECODE-FILE-TIME, and rewrite
FILE-TIME->STRING so that it's result is (almost) in RFC-822 format.

30 years agoChange random ACCESS statement to use now-exported name.
Chris Hanson [Sat, 15 Apr 1995 06:11:22 +0000 (06:11 +0000)]
Change random ACCESS statement to use now-exported name.

30 years agoImplement new procedure MAKE-DECODED-TIME. Rename several internal
Chris Hanson [Sat, 15 Apr 1995 06:10:04 +0000 (06:10 +0000)]
Implement new procedure MAKE-DECODED-TIME.  Rename several internal
procedures to use Common Lisp names, and export them to the global
environment.

30 years agoAdd means to control line translation in subprocess port. This is
Chris Hanson [Fri, 14 Apr 1995 19:06:15 +0000 (19:06 +0000)]
Add means to control line translation in subprocess port.  This is
useful when communicating with a TCP stream relay subprocess, in which
case the line terminators are defined by the network protocol rather
than by the operating system.

30 years agoCode that walked LAMBDA body was being too strict -- body should allow
Chris Hanson [Fri, 14 Apr 1995 08:34:25 +0000 (08:34 +0000)]
Code that walked LAMBDA body was being too strict -- body should allow
definitions and expressions to be interspersed.

30 years agoCode that walked LAMBDA body was being too strict -- body should allow
Chris Hanson [Fri, 14 Apr 1995 08:21:32 +0000 (08:21 +0000)]
Code that walked LAMBDA body was being too strict -- body should allow
definitions and expressions to be interspersed.

30 years agoFixed thinko check with fat procedures.
Stephen Adams [Fri, 14 Apr 1995 04:34:10 +0000 (04:34 +0000)]
Fixed thinko check with fat procedures.

30 years agoReorganize X11 keys and eliminate unnecessary overriding of
Chris Hanson [Thu, 13 Apr 1995 23:38:09 +0000 (23:38 +0000)]
Reorganize X11 keys and eliminate unnecessary overriding of
HOOK/MAKE-SPECIAL-KEY.

30 years agoReorganize X11 keys and eliminate unnecessary overriding of
Chris Hanson [Thu, 13 Apr 1995 23:28:37 +0000 (23:28 +0000)]
Reorganize X11 keys and eliminate unnecessary overriding of
HOOK/MAKE-SPECIAL-KEY.

30 years agoReorganize X11 keys and eliminate unnecessary overriding of
Chris Hanson [Thu, 13 Apr 1995 23:28:07 +0000 (23:28 +0000)]
Reorganize X11 keys and eliminate unnecessary overriding of
HOOK/MAKE-SPECIAL-KEY.

30 years agoReorganize slightly.
Chris Hanson [Thu, 13 Apr 1995 23:27:21 +0000 (23:27 +0000)]
Reorganize slightly.

30 years agoAdd bindings for PC's special keys. Under OS/2, these keys have their
Chris Hanson [Thu, 13 Apr 1995 23:26:56 +0000 (23:26 +0000)]
Add bindings for PC's special keys.  Under OS/2, these keys have their
natural names rather than being mapped onto the names of the
corresponding keys on the HP keyboard.

30 years agoReorganize special-key code.
Chris Hanson [Thu, 13 Apr 1995 23:26:00 +0000 (23:26 +0000)]
Reorganize special-key code.

30 years agoMove CHECK-FOR-ILLEGAL-DEFINITIONS into the runtime system proper.
Chris Hanson [Thu, 13 Apr 1995 22:26:11 +0000 (22:26 +0000)]
Move CHECK-FOR-ILLEGAL-DEFINITIONS into the runtime system proper.

30 years agoAdd code to check for illegal internal definitions and disallow them.
Chris Hanson [Thu, 13 Apr 1995 22:24:53 +0000 (22:24 +0000)]
Add code to check for illegal internal definitions and disallow them.

30 years agoInitial revision
Chris Hanson [Thu, 13 Apr 1995 22:24:31 +0000 (22:24 +0000)]
Initial revision

30 years agoAdd switch to control special handling of abbreviations for QUOTE and
Chris Hanson [Wed, 12 Apr 1995 21:15:35 +0000 (21:15 +0000)]
Add switch to control special handling of abbreviations for QUOTE and
friends.  Default this switch to OFF.

30 years agoFixed RTLGEN/%APPLY-UNCHECKED for tagged entry points.
Stephen Adams [Wed, 12 Apr 1995 19:23:11 +0000 (19:23 +0000)]
Fixed RTLGEN/%APPLY-UNCHECKED for tagged entry points.

30 years agoFix bug in OS2_receive_message. It's possible to arrive here with a
Chris Hanson [Tue, 11 Apr 1995 05:19:34 +0000 (05:19 +0000)]
Fix bug in OS2_receive_message.  It's possible to arrive here with a
closed qid due to race conditions when the sending thread closes the
qid while the reading thread is about to read from it.  This was
happening under some circumstances when sending mail: the sendmail
program doesn't generate any output, and if it terminates before the
input thread's first read, the qid will be closed.

30 years agoAdd new procedure OS2_current_tid.
Chris Hanson [Tue, 11 Apr 1995 05:17:11 +0000 (05:17 +0000)]
Add new procedure OS2_current_tid.

30 years agoFix OS/AUTO-SAVE-PATHNAME to rewrite the buffer name so that it uses
Chris Hanson [Mon, 10 Apr 1995 23:06:09 +0000 (23:06 +0000)]
Fix OS/AUTO-SAVE-PATHNAME to rewrite the buffer name so that it uses
only legal HPFS characters, when the pathname includes the buffer
name.

30 years agoAdd support for fetching mail from POP servers using an OS-dependent
Chris Hanson [Mon, 10 Apr 1995 20:24:17 +0000 (20:24 +0000)]
Add support for fetching mail from POP servers using an OS-dependent
mechanism.  This is necessary because socket support is only
implemented for unix, and other mechanisms are available for other
systems.

30 years agoAdd code to support fetching mail from a POP server using the freeware
Chris Hanson [Mon, 10 Apr 1995 20:22:42 +0000 (20:22 +0000)]
Add code to support fetching mail from a POP server using the freeware
"popclient" program.  Fix bug in precious-file backups.

30 years agoCall OS/FIND-PROGRAM explicitly in one place where it is needed.
Chris Hanson [Mon, 10 Apr 1995 20:21:31 +0000 (20:21 +0000)]
Call OS/FIND-PROGRAM explicitly in one place where it is needed.

30 years agoChange C-g interrupt-signalling code to recognize the combination of
Chris Hanson [Mon, 10 Apr 1995 16:53:05 +0000 (16:53 +0000)]
Change C-g interrupt-signalling code to recognize the combination of
the lower-case g and the control keys.  Previously, it only recognized
the upper-case g.  This was not immediately noticed because the input
stream lookahead code recognized both combinations.

30 years agoRUN-SYNCHRONOUS-PROCESS now calls OS/FIND-PROGRAM on its program
Chris Hanson [Mon, 10 Apr 1995 16:50:04 +0000 (16:50 +0000)]
RUN-SYNCHRONOUS-PROCESS now calls OS/FIND-PROGRAM on its program
argument -- this allows invocation of a program without knowing its
precise location in the file system.

30 years agoFix typos in previous change.
Chris Hanson [Mon, 10 Apr 1995 16:48:34 +0000 (16:48 +0000)]
Fix typos in previous change.

30 years agoCellified code for SET! now omits the read of the variable if the SET!
Stephen Adams [Mon, 10 Apr 1995 15:30:31 +0000 (15:30 +0000)]
Cellified code for SET! now omits the read of the variable if the SET!
is in an effect-only position (i.e. non-terminal BEGIN action).  Not
really necessary but does make the intermediate code a little smaller.

30 years agoChanged calls to ASSQ to CLEANUP/ENV/FIND and wrote CLEANUP/ENV/FIND
Stephen Adams [Mon, 10 Apr 1995 14:53:31 +0000 (14:53 +0000)]
Changed calls to ASSQ to CLEANUP/ENV/FIND and wrote CLEANUP/ENV/FIND
like ASSQ to get better accountability in profiling.

30 years agoAdd hooks to allow OS-dependent POP mail support.
Chris Hanson [Sun, 9 Apr 1995 23:28:20 +0000 (23:28 +0000)]
Add hooks to allow OS-dependent POP mail support.

30 years agoTweak prompt-for-password slightly.
Chris Hanson [Sun, 9 Apr 1995 23:21:53 +0000 (23:21 +0000)]
Tweak prompt-for-password slightly.

30 years agoPatch two bugs.
Chris Hanson [Sun, 9 Apr 1995 23:09:19 +0000 (23:09 +0000)]
Patch two bugs.

30 years agoMake OS-independent binding for file-time->string.
Chris Hanson [Sun, 9 Apr 1995 23:07:05 +0000 (23:07 +0000)]
Make OS-independent binding for file-time->string.

30 years agoMake OS-independent binding for file-time->string.
Chris Hanson [Sun, 9 Apr 1995 22:58:09 +0000 (22:58 +0000)]
Make OS-independent binding for file-time->string.

30 years agoAdd shared-memory API calls -- these will be needed for several
Chris Hanson [Sun, 9 Apr 1995 22:34:36 +0000 (22:34 +0000)]
Add shared-memory API calls -- these will be needed for several
different purposes.

30 years agoMake OS-independent bindings for the following procedures. These
Chris Hanson [Sun, 9 Apr 1995 22:33:28 +0000 (22:33 +0000)]
Make OS-independent bindings for the following procedures.  These
procedure were previously defined for all OSs, but with different
names in each OS:

    user-home-directory
    current-user-name
    current-home-directory

The old names have been aliased to the new, except for OS/2.

30 years agoHand coded SIMPLIFY/ENV/FRAME-LOOKUP rather than use
Stephen Adams [Sun, 9 Apr 1995 04:45:59 +0000 (04:45 +0000)]
Hand coded SIMPLIFY/ENV/FRAME-LOOKUP rather than use
ASSOCIATION-PROCEDURE to get better profiling info.

30 years agoUnrolled loop in NODES-LINKED?
Stephen Adams [Sat, 8 Apr 1995 21:00:17 +0000 (21:00 +0000)]
Unrolled loop in NODES-LINKED?

30 years agoFixed interrupt logic bug -- a consequence of making internal
Stephen Adams [Sat, 8 Apr 1995 19:46:51 +0000 (19:46 +0000)]
Fixed interrupt logic bug -- a consequence of making internal
procedures look (and thus be condisered to be) external procedures.

30 years agoTemporarily commented out unsafe accessors for cookie calls.
Stephen Adams [Sat, 8 Apr 1995 16:44:54 +0000 (16:44 +0000)]
Temporarily commented out unsafe accessors for cookie calls.

30 years agoAdded lots of hair to deal with programs that have large chunks
Stephen Adams [Sat, 8 Apr 1995 04:38:22 +0000 (04:38 +0000)]
Added lots of hair to deal with programs that have large chunks
duplicated.  The problem was that the duplicated frame vectors were
still EQ?, causing conflicting stack optimizations.

30 years agoAdded a MATCH macro. The generated code is not very good.
Stephen Adams [Fri, 7 Apr 1995 04:55:36 +0000 (04:55 +0000)]
Added a MATCH macro.  The generated code is not very good.