suzanne.soy/racket
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
23,775 Commits 17 Branches 54 Tags 278 MiB
13db1c1a08
Commit Graph

3538 Commits

Author SHA1 Message Date
Matthew Flatt
2b2c44774f fix scheme_basic_env() reset behavior 
Also, fix test to that it DrDr will run it.
 2012-01-02 15:03:49 +01:00
Eli Barzilay
3718600125 2011 -> 2012 2011-12-31 15:16:59 -05:00
Matthew Flatt
12baa07e7e fix missing initialization 2011-12-31 06:32:13 -07:00
Matthew Flatt
4fe30cf433 fix cross-module inlining for multi-arity functions 
First use of the function was determining a single arity for
the enclosing module, and that arity could trigger warnings
in addition to failures to inline. For example, using `map'
on 3 arguments would trigger incorrect warnings for later
uses of `map' on 2 arguments.
 2011-12-28 20:35:53 -06:00
Matthew Flatt
580c952e46 fix locale sensitivity of inexact-number parsing 
Closes PR 12070
 2011-12-28 19:06:07 -06:00
Matthew Flatt
5cd324af15 src/README corrections (of out-of-date information) 2011-12-21 16:43:01 -07:00
Matthew Flatt
d020c75202 remove libfit 2011-12-21 16:17:56 -07:00
Matthew Flatt
24991f0c1a win64: fix JIT SSE-based FP 
XMM6-15 are preserved in the Win64 ABI
 2011-12-21 11:27:24 -07:00
Matthew Flatt
58b111485f int -> intptr_t fixes 2011-12-21 11:27:22 -07:00
Matthew Flatt
7fbe3c7224 avoid accidental XFORM-LOCK file in the future 2011-12-19 07:05:27 -07:00
Matthew Flatt
35d800f109 remove accidentally added file 2011-12-19 07:03:25 -07:00
Matthew Flatt
67f83bfd6d win64: fix JIT 
Repairs commit 558b03034a.
 2011-12-18 16:35:33 -07:00
Matthew Flatt
c9d214138e make `choice-evt' take O(N) time for N arguments 
Delay flattening nested `choice-evt's to `sync'.
 2011-12-15 16:04:08 -07:00
Eli Barzilay
6960767508 New Racket version 5.2.0.7. 2011-12-15 03:50:18 -05:00
Matthew Flatt
ee775c3cc3 intern strings, etc. only when making syntax objects, not in `read' 
Rename `read-intern-literal' to `datum-intern-literal'.

Interning is needed only in `read-syntax' or `datum->syntax' to
set up the invariants that the bytecode compiler needs for cross-module
optimization. When `read'ing numbers from a data file, meanwhile,
interning slows things down a lot and doesn't seem worthwhile.
 2011-12-14 16:03:44 -07:00
Matthew Flatt
7ac1232c46 restore and fix chaperone-apply path 
with another little performance improvement
 2011-12-14 09:56:48 -07:00
Matthew Flatt
a7ddbedc7d disable a part of the new chaperone eval path that isn't right 2011-12-14 09:17:47 -07:00
Matthew Flatt
558b03034a streamline thread-local setup in JIT-generated code 
Pass a pointer to the thread-local table on entry to JIT-generated
code, instead of having the JIT-generated code call a C function
to get the table. This doesn't seem to improve performance on my
machine, but it generates less code and is probably faster in
some cases.
 2011-12-14 06:21:06 -07:00
Matthew Flatt
761a40d483 faster paths for chaperone application 2011-12-14 06:21:06 -07:00
Matthew Flatt
f7c506471b remove redundant arity checking calling JIT-generated code 
JIT-generated code has arity checking built in, but the
interpreter-to-JIT path unnecessarily checked arity before
jumping to JITted code.
 2011-12-14 06:20:08 -07:00
Matthew Flatt
be2afcdcc9 add missing bytecode-unmarshal check 2011-12-10 11:43:59 -07:00
Matthew Flatt
bf90cf9028 fix problems printing symbols with unicode chars 2011-12-10 11:43:59 -07:00
Eli Barzilay
ccbd5d51fd New Racket version 5.2.0.6. 2011-12-10 03:50:13 -05:00
Matthew Flatt
287d5cedf5 add `pseudo-random-generator-vector?' 
Also, adjust docs to clarify that `vector->pseudo-random-generator!'
can be used as a more complete seeding function.
 2011-12-09 09:12:17 -07:00
Vincent St-Amour
c980abf35b Add comments to the inliner. 2011-12-08 17:13:28 -05:00
Matthew Flatt
4a387c5b6a add some missing checks on bytecode parsing 
The `tests/racket/stress/fuzz' test fails much less frequently,
though problems certainly remain.
 2011-12-08 13:42:09 -07:00
Matthew Flatt
657be87c66 generalize gcd' and lcm' to work on rationals 2011-12-08 09:25:39 -07:00
Matthew Flatt
958352eefc fix futures bug related to GC 
When a future is blocked on JIT generation, a lightweight closure
is captured, and then the future moves on, the runtime thread would
correctly shift the on-demand JIT argument to the captured copy
of the runstack. However, it would also add 2 to that pointer
to use as the argv array, and the captured runstack is not allocated
to allow interior pointers, so a GC during on-demand JIT could
crash. The solution is to keep an offset alongside the argv pointer
during JITting.
 2011-12-06 09:31:52 -07:00
Matthew Flatt
4f669f0d7e fix futures handling of tail-call results 2011-12-05 19:11:32 -07:00
Matthew Flatt
55b6670084 mysterx patch from Antonio Menezes Leitao 2011-12-05 19:11:32 -07:00
Kevin Tew
9f7e730a77 another round of unused variable clean-ups 2011-12-05 19:11:32 -07:00
Matthew Flatt
b25d1b42d6 minor clean up 2011-12-05 19:10:22 -07:00
Matthew Flatt
8dd4f1de21 fix futures bug 
When `(continuation-mark-set-first #f ...)' should produce #f and
the continuation is lightweight-captured, the result was NULL
instead of #f.
 2011-12-05 19:10:22 -07:00
Matthew Flatt
8b3fae79c2 fix no-futures, no-places build 2011-12-04 08:07:13 -07:00
Matthew Flatt
59e41ce05f simpler stack-space check for Racket thread creation 2011-12-04 06:52:33 -07:00
Matthew Flatt
885bf6555f fix future resume in runtime thread with extra stack space 
Also, fix the possibility that a future can shift from RUNNING
to PENDING while the runtime thread is checking PENDING then
RUNNING.
 2011-12-04 06:52:33 -07:00
Matthew Flatt
333ffd7ede JIT tweak 2011-12-04 06:52:33 -07:00
Matthew Flatt
be80e7d864 fix an over-eager reordering by the bytecode compiler 
The over-eager transformation could be space-unsafe, and it
could duplicate an unsafe operation whose result is used only
once in a function that eds up being inlined multiple times.
 2011-12-04 06:52:33 -07:00
Matthew Flatt
db3878e309 another futures bug fix 2011-12-03 07:35:39 -07:00
Matthew Flatt
09d4d6c507 avoid uninitialized-variable warning from C compilers 2011-12-03 05:45:41 -07:00
Matthew Flatt
829e894af2 flatten (begin ... (begin ...)) in last pass of compiler 
Flattening just makes bytecode slightly smaller.
 2011-12-02 20:46:00 -07:00
Matthew Flatt
dd5d1c9046 another futures bug fix 2011-12-02 16:47:59 -07:00
Matthew Flatt
745c4b6470 fix another way that `racket/gui' instances can leak 2011-12-02 15:48:29 -07:00
Matthew Flatt
7397f710e2 fix reconstruction of `provide' forms in module expansion 2011-12-02 13:58:39 -07:00
Matthew Flatt
96330556e7 fix secondary hash-code for weak hashes 2011-12-02 12:11:52 -07:00
Matthew Flatt
67c3aa4b2d improve comments on `future_t' fields; remove a redundant field 2011-12-02 09:58:22 -07:00
Matthew Flatt
c2fac607f0 another futures bug fix 2011-12-02 09:41:59 -07:00
Matthew Flatt
efe165aeac futures bug fix 2011-12-02 08:04:52 -07:00
Matthew Flatt
48524906d8 fix bug in cross-module inliner 2011-12-01 14:05:42 -07:00
Kevin Tew
78c2d7b9ab fix variables assigned but not referenced 2011-12-01 14:05:42 -07:00