suzanne.soy/racket
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
24,439 Commits 17 Branches 54 Tags 278 MiB
acfe585c93
Commit Graph

3639 Commits

Author SHA1 Message Date
Eli Barzilay
acfe585c93 Some cleanups, mainly around badly indented CPP directives. 2012-03-25 20:57:52 -04:00
Eli Barzilay
4469b669fc New Racket version 5.2.900.2. 2012-03-24 03:35:16 -04:00
Matthew Flatt
6e3c3d8b71 MSVC project fixes for various Debug and SGC modes 2012-03-23 17:58:39 -06:00
Matthew Flatt
ae89e210fb allow (submod "..")' as a shorthand for (submod "." "..")' 
Also, `(submod ".")' is allowed as a module path; it's not useful,
but it makes the grammar slightly more regular.
 2012-03-22 18:43:11 -06:00
Kevin Tew
d4d5ca70fb Hold reference to original thread during BEGIN_ESCAPABLE in place-channel sync 
Fixes several bugs related to killed threads that were in the middle of
syncing on place-channels.
 2012-03-22 14:52:48 -06:00
Matthew Flatt
16ef341e98 fix JIT bug related to inlining `variable-reference-constant?' 2012-03-21 19:37:46 -06:00
Matthew Flatt
47ae2b387a fix shaodwing problem in `module*' 2012-03-20 07:28:35 -06:00
Kevin Tew
9be864b2f8 Fix sync/timeout hang due to place channels - forgot to git add 2012-03-16 08:39:58 -06:00
Kevin Tew
666c0fdb57 Fix sync/timeout hang due to place channels 2012-03-16 08:18:05 -06:00
Matthew Flatt
9838cb37e8 remove debugging code that was accidentally pushed 2012-03-13 18:46:28 -06:00
Matthew Flatt
415b1eabc4 more load[/use-compiled] handler fixes for submodules 
Closes PR 12630
 2012-03-13 14:31:40 -06:00
Matthew Flatt
366aa2c42d fixups for the `module-path?' change 2012-03-12 21:10:14 -06:00
Matthew Flatt
542f19c00e fix default `load/use-compiled' handler 
A module shouldn't be loaded from source to find a submodule
if a module for the source is already declared.
 2012-03-12 21:10:14 -06:00
Matthew Flatt
facc8db712 change module-path?' to subsume path?' 
Although th eoriginal idea was to distinguish "text" paths
from derived filesystem paths, practically everythign that accepts
a module path also accepts a path. Building the generalization into
`module-path?' makes it easier to support `submod' wrappers on paths,
and it seems to fix things rather than break them.
 2012-03-12 21:08:54 -06:00
Matthew Flatt
060515b9cd extend string-literal syntax to allow surrogate-style \u pairs 
For example, "\uD834\uDD1E" is another way to write "\U01D11E",
while "\uD834", "\uDD1E" or "\uDD1E\uD834" are still errors.
 2012-03-12 21:03:20 -06:00
Matthew Flatt
e1a5e032cb further fixes to vector abuse 2012-03-12 11:47:18 -06:00
Matthew Flatt
018292ec00 GC with backtraces: add structure-type listing 2012-03-12 11:36:04 -06:00
Matthew Flatt
3073c994f5 make GC backtrace show eq hash code 2012-03-12 10:59:47 -06:00
Matthew Flatt
420d3e7717 fix JIT bug exposed by OpenBSD 5.0 
This is the second attempt to fix a problem with using a JIT-generated
address in JIT-generated code.

Closes PR 12627
 2012-03-12 08:23:17 -06:00
Matthew Flatt
8f0cf5ba9e work around problem with gcc-4.0 build on Lion 2012-03-12 07:45:25 -06:00
Matthew Flatt
2ba910ef23 avoid compiler warning 2012-03-12 07:45:25 -06:00
Matthew Flatt
64f21122be avoid compiler warning 2012-03-10 10:52:52 -07:00
Matthew Flatt
402fdaad9a fixes and improvements for 3m backtraces 2012-03-10 08:30:02 -07:00
Matthew Flatt
34314c1e7a fix JIT bug 
The bug happens with n-ary uses of arithmetic operations that
have constant arguments but couldn't be constant-folded ---
maybe due to a divide-by-zero.
 2012-03-10 08:30:01 -07:00
Eli Barzilay
5c051f3ec9 New Racket version 5.2.900.1. 2012-03-10 03:30:15 -05:00
Kevin Tew
024bb8a718 Eliminate compiler warnings 2012-03-09 12:40:22 -07:00
Matthew Flatt
e73be4a9ae fix internal abuse of vectors 
Using NULL in a vector slot can crash bracktrace printing;
replace it with `#f'.
 2012-03-09 12:05:36 -07:00
Matthew Flatt
f7eceb4554 dierct reference to scheme_null in Win32 executable 2012-03-09 10:37:56 -07:00
Matthew Flatt
3a1e8803ff fix errortrace for submodules 2012-03-09 10:34:56 -07:00
Matthew Flatt
566759a5fa progress on submodule docs; bug fixes 2012-03-09 10:34:56 -07:00
Matthew Flatt
3d69dfab86 first cut at submodules 2012-03-09 10:34:56 -07:00
Matthew Flatt
facb411a63 fix problems with place termination 2012-02-29 20:20:55 -07:00
Eli Barzilay
f7c67b49a4 Big newline at EOF scan. 2012-02-29 00:28:11 -05:00
Matthew Flatt
829820e458 fix problem with continuations 
A tail call to a continuation could trigger a GC at a bad time
while checking whether an escape variant of the contination
is valid.
 2012-02-28 15:35:38 -07:00
Matthew Flatt
70ab6d482a fix problem with continuations 
The problem was a misplaced allocation that could cause a GC
when the thread is in an unstable state in the process of
applying a continuation.
 2012-02-28 15:35:38 -07:00
Matthew Flatt
534886dbe4 limit time for getting a native (JIT) stack trace 
It's possible for a deep recursion to be all in C instead of
JIT-generated code, in which case the caching code for
`current-continuation-mark' cannot kick in to make the operation
effectively constant time. Bail out (to keep things constant time) if
that happens.
 2012-02-27 14:46:03 -07:00
Kevin Tew
3e0e4a3f6b Allow hashes across place channels. 2012-02-27 10:11:04 -07:00
Matthew Flatt
0dcc96c5eb fix JIT, `unsafe-struct-ref', chaperone, and procedure property 2012-02-27 06:57:44 -07:00
Eli Barzilay
3426b57fb9 New Racket version 5.2.1.7. 2012-02-27 03:45:15 -05:00
Matthew Flatt
188c6fdc66 add `log-max-level' 
As suggested by Tobias Hammer
 2012-02-26 22:13:02 -05:00
Matthew Flatt
21a07ae9d6 fix problems with `(_fun #:in-original-place? #t ....)' 
The old implementation could cause deadlock by blocking on a semaphore
while waiting for the original place to run a callback, but a master
GC might be needed (and the blocked place wouldn't get the signal).

Beside fixing that problem, a potential memory leak is fixed in
calling an ffi funcition and having a Racket->C unmarshaling fail.
Also, the GC marking routine for a `place' value didn't reference the
place's underlying `place_obj' value.
 2012-02-26 22:13:01 -05:00
Matthew Flatt
dc1d4e80dd fix incorrect GC decls hard-wired into xform 
Closes PR 12602
 2012-02-26 03:58:21 +00:00
Matthew Flatt
a025f7e9c8 fix bug in bytecode optimizer 
The bug is triggered by unsafe flonum operations, a
conversion that tries to make the arguments more unboxable,
and a `lambda' form within an argument to the unsafe
operation.

Closes PR 12587
 2012-02-22 06:54:26 -07:00
Eli Barzilay
c007c345f9 A bunch of more typos like the ones in David's commit. 2012-02-21 14:21:43 -05:00
Matthew Flatt
6371df556c yet anther attempt to work around the Mac 10.7 localtime() bug 
I now think the problem is likely to be realted to values
that do not fit into a signed 32-bit integer. Check for
the OS version and reject such integers.
 2012-02-20 08:09:06 -07:00
Stephen A. Goss
11f7dfbdcc make it clear how to select a language from command line 2012-02-20 08:09:05 -07:00
Matthew Flatt
0e40cfcdc8 declare `char' fields used as signed 
With some compilers or platforms, `char' means an unsigned
value, so we have to be explicit with `signed char'.
 2012-02-19 06:25:29 -07:00
Matthew Flatt
11de33d4ff remove MysterX DLL, replace with wrapper around `ffi/com' 
The ActiveX part of MysterX is gone. The `ffi/com' re-imeplemtnation
provides only core COM support.

The "mysssink" DLL is still needed, and its source is still
in the tree, but it is downloaded in the same way as other
pre-built DLLs. The DLL no longer needs to be registered with
regsvr32.
 2012-02-17 06:37:19 -07:00
Matthew Flatt
566e9bb8bf make MzCOM /v report errors 2012-02-17 06:36:23 -07:00
Matthew Flatt
a71ac65a27 fix CPP mistake 2012-02-17 06:36:23 -07:00