Commit Graph

4142 Commits

Author SHA1 Message Date
Matthew Flatt
b40c91e6f4 more fixes to avoid warnings 2013-03-18 13:47:11 -07:00
Matthew Flatt
1e31d99ca3 fix non-extflonum build 2013-03-18 11:30:10 -07:00
Matthew Flatt
8122249602 avoid compiler warnings 2013-03-18 11:30:10 -07:00
Matthew Flatt
3bf9dd93dc fix header-file issues 2013-03-18 11:30:10 -07:00
Michael Filonenko
35a093469c windows: extflonum support
Enable extflonums in a MSVC build by relying on a MinGW-built DLL,
"longdouble.dll". The DLL is loaded on startup. If the DLL isn't
available, then `extflonum-available?' reports #f.

Instead of setting the floating-point mode globally to extended
precision, the mode is changed only just before (and restored right
after) extflonum arithmetic operations.
2013-03-18 07:42:32 -07:00
Tobias Hammer
2bae950273 Work around bug in QNX readlink
QNX readlink null-terminates its result and adds the termination
to the len-result. Fix by decrementing len if needed.
2013-03-10 07:18:04 -06:00
Matthew Flatt
5e20f51415 fix mark handling for impersonatored procedures to match docs
Using an impersonator property to represent an application mark was
a bad choice. The property gets propagated, so it is also on any later
chaperone layer, and then things go bad: the docs say that special
treatment is triggered by supplying an argument to `impersonate-property',
but it was actually triggered by the chaperone having the property.

Change the implementation to match the documentation. Using an
impersonator property to supply the mark should be regarded as a hack,
but now the implementaiton is at least consistent with the documentaiton.
2013-03-03 10:16:51 -07:00
Matthew Flatt
cdf0f6b9ab add 'so-mode to `system-type'
To work better with OpenBSD (which has been a problem since we
changed `ffi-lib' to open libraries in "local" mode by default).
2013-03-01 15:20:17 -07:00
Juan Francisco Cantero Hurtado
292c81a826 Fix `configure' for OpenBSD
"-rdynamic" doesn't work with GCC 4.7 on OpenBSD: "cc: error:
unrecognized command line option '-rdynamic'". The switch isn't
necessary because it is only a wrapper to "-Wl,--export-dynamic".
Look GCC bug 37454.
2013-03-01 15:20:17 -07:00
Matthew Flatt
ca4320467c fix missing initialization of cached inspector reference
The 3m transformation fills it in, but it's obviously
better to be explicit and avoid the bug for CGC.
2013-02-28 12:19:25 -07:00
Matthew Flatt
542c38ebc5 equality: struct transparency and inspector shortcuts and caching
Same as for hashing.
2013-02-28 12:19:25 -07:00
Matthew Flatt
2236363696 fix bug in hashing shortcut for transparency 2013-02-28 10:06:36 -07:00
Matthew Flatt
39aafe3a1e change `eq?' hashing to avoid dropping the low bit for fixnums 2013-02-28 06:51:26 -07:00
Matthew Flatt
7a8c2ff063 improve complexity of `hash-iterate-{key,value}'
For immutable hashes, `hash-iterate-{key,value}' used to take O(N) time on
the first call for a particular table and O(1) thereafter. Now it takes
O(1) time for all calls, the constant is only slightly larger for
the non-first calls, and there's no extra allocation.
2013-02-28 06:51:26 -07:00
Matthew Flatt
773496642b `equal?' hashing: shortcut and some caching for structure transparency 2013-02-28 06:51:26 -07:00
Eli Barzilay
ca304aafaf New Racket version 5.3.3.7. 2013-02-28 03:50:23 -05:00
Matthew Flatt
abf44f8b49 add `module-compiled-cross-phase-persistent?' 2013-02-27 08:53:43 -07:00
Matthew Flatt
88a36a077c terminology change again: "cross-phases persistent" modules 2013-02-27 08:53:43 -07:00
Eli Barzilay
5ca93be56c New Racket version 5.3.3.6. 2013-02-27 03:30:19 -05:00
Matthew Flatt
a04bfa5ba2 fix phase-collapsing-module inference
This bug was causing a crash in some DrRacket tests.
2013-02-26 19:19:57 -07:00
Matthew Flatt
2646ff6895 try "phase-collapsing" instead of "phaseless" 2013-02-26 19:01:36 -07:00
Matthew Flatt
2e652fc2b3 fix some "phaseless" module holes 2013-02-26 19:01:35 -07:00
Matthew Flatt
899a3279c2 add experimental support for "phaseless" modules
The intent is to support phase-crossing data such as the `exn:fail:syntax'
structure type that is instantiaed by macros and recognized by contexts
that use `eval' or `expand'. Phaseless modules are highly constrained,
however, to avoid new cross-phase channels, and a module is inferred to
be phaseless when it fits syntactic constraints.

I've adjusted `racket/kernel' and improved its documentation a little
so that it can be used to implement a phaseless module (which can
import only from other phaseless modules).

This change also adds a `flags' field to the `mod' structure type
from `compiler/zo-structs'.
2013-02-26 14:55:28 -07:00
Matthew Flatt
bb9068641b racket/draw win32: Pango patch
Closes PR 13513
2013-02-23 08:42:36 -07:00
Eli Barzilay
de2aa509ac New Racket version 5.3.3.5. 2013-02-21 03:30:15 -05:00
Matthew Flatt
33c6635550 version 5.3.3.5
Bumped to force a rebuild of ".zo"s that were generated with the flonum
bug.
2013-02-20 06:46:32 -07:00
Matthew Flatt
c56b73a00a JIT: fix slow-path boxing for non-self-tail calls with unboxed arguments
The bug was introduced when changing the order of unboxed argument
staging along with extflonum unboxing support. The main symptom of
the bug was that icons were sometimes blacked out in DrRacket.
2013-02-20 06:42:01 -07:00
Matthew Flatt
535072086b fix non-extflonum build 2013-02-19 18:15:48 -07:00
Matthew Flatt
97c34905c1 racket/extflonum: fix JIT-inlined `extflvector-ref' in 32-bit mode 2013-02-19 16:47:39 -07:00
Matthew Flatt
0afcda2a5a fix error checking for flreal-part' and flimag-part' 2013-02-19 16:47:39 -07:00
Matthew Flatt
8ef5e88f41 another JIT repair 2013-02-19 14:03:00 -07:00
Matthew Flatt
e0650c43df JIT: fix for 32-bit mode
Fix a bug in ca951294d4.
2013-02-19 10:41:08 -07:00
Matthew Flatt
ca951294d4 JIT: inline char->integer' and integer->char' 2013-02-19 08:43:50 -07:00
Eli Barzilay
bbd06930ae New Racket version 5.3.3.4. 2013-02-19 03:30:16 -05:00
Matthew Flatt
79c4af4e45 racket/base: add `impersonator-ephemeron' 2013-02-18 17:01:11 -07:00
Matthew Flatt
fdd8dc9376 JIT: fix bug in checking certain values before unboxing
The check was incomplete in the case that both arguments to a binary
[ext]flonum function need to be checked and the second one was not an
[ext]flonum and also not a fixnum.
2013-02-18 11:07:45 -07:00
Matthew Flatt
4271d987cd ffi/unsafe: fix a bug in conversion to UTF-16
Characters outside of the BMP were translated incorrectly.
2013-02-18 10:28:58 -07:00
Matthew Flatt
f8a0d6d432 configure: add more guidance/reporting on non-recommended options 2013-02-18 10:28:58 -07:00
Eli Barzilay
9e9adeae0c Consistent capitalization for "GRacket". 2013-02-16 13:41:38 -05:00
Matthew Flatt
9b04aef0f8 fix a problem with `syntax-local-lift-require' at top level
The problem was exposed by improved error checking
in the expander to detect references to exports of a module that
is not yet visited, and as triggered by the documentation for
`2htdp/planetcute'.
2013-02-16 09:29:16 -07:00
Eli Barzilay
1ddacff399 New Racket version 5.3.3.3. 2013-02-16 03:40:18 -05:00
Matthew Flatt
6281ac0aac less noisy win32 builds 2013-02-15 12:15:53 -07:00
Michael Filonenko
840fc9c657 full unboxing for extflonums, plus `ffi/unsafe' support
Includes a repair for floating-point `min' and `max' that affects all
x86 builds that use SSE arithmetic, leaving the (otherwise unused)
floating-point stack in a bad state, which might have affected
x87-using C/foreign libraries running alongside Racket.
2013-02-15 12:15:34 -07:00
Matthew Flatt
a348e5421d fix a `configure' test
Thanks again to Tobias.
2013-02-15 11:55:26 -07:00
Eli Barzilay
576a573216 New Racket version 5.3.3.2. 2013-02-15 03:30:18 -05:00
Matthew Flatt
74a3e14f9c improve `configure' tests for cross-compilation
Use compile and link tests instead of run tests, where
possible.

Based on a patch from Tobias
2013-02-14 19:47:40 -07:00
Matthew Flatt
ccabbaa36e fix `current-subprocess-custodian-mode' for a #f argument 2013-02-14 17:30:10 -07:00
Matthew Flatt
421cb24138 add `port-counts-lines?' 2013-02-14 15:45:26 -07:00
Tobias Hammer
7e2b443fa9 Support for multiple value return from wrap-evt and handle-evt
wrap/handle-evt that receives multiple values must have a handler function with adequate arity.

struct.c:
  change contract for wrap/handle-evt from (any/c -> any) to procedure?
thread.c:
  adjust sync processing
sync.rktl:
  add test for handle-evt, wrap-evt and prop:evt
2013-02-14 15:45:26 -07:00
Matthew Flatt
294bed209e macro expander fix
Repair 73e07f576b added an item to a list without incrementing a
counter for the list length, which cause a different element of
the list to be dropped, which could mess up binding resolution in
arbitrarily bad ways.

(Ths bug falls into the "how did this not get exposed earlier?"  bin,
although part of the answer is that it requires a combination of
module re-expansion and simplification of syntax objects in the
residual program.)

Closes PR 13428
2013-02-14 07:11:16 -07:00