Ryan Culpepper
ec406dd301
cs: fix lookup-errno
2018-10-29 11:12:58 +01:00
Matthew Flatt
933e3c14b5
bytecode: fix "preserves marks" annotation w.r.t call/immediate-mark
...
A function that uses `call-with-immediate-continuation-mark` in tail
position should not be flagged as "preserves marks", because the JIT
needs to bump the mark stack if the function is called in non-tail
position.
Closes #2333
2018-10-28 18:52:41 -06:00
Matthew Flatt
d1fe6a6e3e
repair allocation bug in delimited continuations
...
The repair is more precisely a repait to xform, which incorrectly
parsed a C function definition that starts "struct" as a struct
declaration. (The function starts "struct" because the return type is
"struct Scheme_Overflow_Jmp *".) Since the function wasn't recognized,
xform didn't convert it to cooperate with the garbage collector.
Closes #2341
2018-10-28 17:09:45 -06:00
DarrenN
2dc96b365a
Implement hash/c random generate and exercise
2018-10-28 14:37:59 -05:00
Matthew Flatt
dbcc9dddc3
xform: updates for recent MinGW
2018-10-28 13:14:14 -06:00
Matthew Flatt
b5084977c0
cs Windows: fill in GUI mode for GRacket
...
Still need to create console on demand.
2018-10-28 13:14:14 -06:00
Matthew Flatt
a24c6fe4f9
cs Windows: move kernel and boot files to DLL
2018-10-28 11:37:04 -06:00
Matthew Flatt
c6a3d4629d
cs: register a nul-terminated byte string for WM_CLASS
2018-10-27 15:22:00 -06:00
Matthew Flatt
a08aed7dde
configure help: clarification on --enable-sgcdebug
...
Closes #2340
2018-10-27 13:43:38 -06:00
Paulo Matos
c6a2843ba2
Fix simple typo in configure help message
2018-10-27 13:42:24 -06:00
Paulo Matos
071406d10d
Mark variables as used so they are not removed by LTO
...
Fixes #2324
2018-10-27 13:39:12 -06:00
Paulo Matos
08c1ba619c
Remove gcc osx matrix entry
2018-10-27 13:31:12 -06:00
Matthew Flatt
de53dc3c5e
cs: add Windows GRacket.exe
2018-10-27 13:23:40 -06:00
Matthew Flatt
c128c5aad6
cs: support ptr-ref
and ptr-set!
on string types
2018-10-27 11:14:42 -06:00
Matthew Flatt
3f6a7b3e06
cs: add icon and other resources to RacketCS.exe
2018-10-27 11:14:42 -06:00
Matthew Flatt
457a673506
Windows: avoid environment variables with empty name
2018-10-26 13:42:16 -07:00
Matthew Flatt
29a80fd7e5
io: repairs for Windows environment variables
2018-10-26 14:19:54 -06:00
Matthew Flatt
23648ba165
cs: repairs relevant to running `racket/gui/ programs on Unix
2018-10-26 14:05:33 -06:00
Matthew Flatt
3b9bc21304
cs: repairs to Windows build client
2018-10-26 11:04:45 -06:00
Paulo Matos
f3708c208d
Fix typo from statics to statistics
2018-10-26 08:10:22 -05:00
Paulo Matos
cbd5499d6e
Add missing close parens
2018-10-26 08:09:46 -05:00
Matthew Flatt
7d86fddbdc
cs: add strip on install for CS executables
2018-10-25 11:30:17 -06:00
Matthew Flatt
c0f14373ab
cs: on Linux record embedded boot relative to ".rackboot" section
2018-10-24 21:06:11 -06:00
Matthew Flatt
8c3956a272
cs: update Windows build
...
Support non-"CS"-suffix mode, and avoid installing bytecode for
traditional Racket when builing a bootstrap traditional Racket.
2018-10-24 18:44:23 -07:00
Alexis King
af554933ef
Fix arity mask upper bound calculation in curry
...
fixes #2332
2018-10-24 12:16:09 -05:00
Ryan Culpepper
11c1a2eaf6
syntax/parse: add regression test, related to #2320 , #2330
2018-10-24 18:09:46 +02:00
Matthew Flatt
fce5554695
io: fix tcp-abandon-port
so that the socket gets closed
2018-10-24 09:53:00 -06:00
Ryan Culpepper
e8832fbcc7
cs: remove plumber flush callback when custodian closes fd output port
...
Previously, the following program would print "error writing to
stream port" on program exit.
(define cust (make-custodian))
(define out
(parameterize ((current-custodian cust))
(open-output-file "test.data" #:exists 'truncate)))
(write-string "This needs flushing...\n" out)
(custodian-shutdown-all cust)
(exit 0)
2018-10-24 17:46:29 +02:00
Ryan Culpepper
ddba2812ba
cs: fix unsafe-file-descriptor->port mode
2018-10-24 17:46:29 +02:00
AlexKnauth
df19625c6c
syntax/parse: fix Check-Syntax arrows within syntax classes
2018-10-24 17:45:30 +02:00
Paul C. Anagnostopoulos
9372ca1e3d
Doc improvements to eval-model.scrbl
2018-10-24 09:29:39 -06:00
Matthew Flatt
23f6d2b43f
cs: setup compiled subdirectory for gracket
like racket
2018-10-23 21:38:43 -06:00
Sam Tobin-Hochstadt
423ce5e5a2
This file doesn't need match
.
2018-10-23 11:56:49 -04:00
Matthew Flatt
8473666af3
cs install: remove old file before coying new files into place
2018-10-23 09:45:53 -06:00
Ryan Culpepper
d4691fb219
fix flush-output when write-out is port indirection
2018-10-23 12:00:31 +02:00
Matthew Flatt
31d35ebe88
cs: another fix for racketcs linking configure on Linux
2018-10-22 20:38:29 -06:00
Matthew Flatt
76f868f287
cs: fix racketcs linking configure for Linux
2018-10-22 20:32:07 -06:00
Matthew Flatt
f8e6b22544
cs: fix makefile chaining for bootstrap Racket variant
2018-10-22 20:14:50 -06:00
Matthew Flatt
7481eb6fed
update another test for the VM identifier in bytecode
2018-10-22 19:53:21 -06:00
Matthew Flatt
cbd195201a
cs: disable use of ncurses when building Chez Scheme
...
RacketCS doesn't currently access the Chez Scheme expeditor, and
avoiding curses make the build work in a more mininal environment.
2018-10-22 19:18:35 -06:00
Matthew Flatt
aa6ec1f0cd
comiler/compile-file: adapt to use-compiled-file-paths
setting
2018-10-22 17:14:49 -06:00
Matthew Flatt
33e8ee7786
zo-lib: updates for virtual machine identifier in bytecode
2018-10-22 15:22:40 -06:00
Matthew Flatt
2fc0263496
makefile: avoid nested repos over HTTP
2018-10-22 12:33:30 -06:00
Matthew Flatt
227944c70c
makefile support for redirecting Chez Scheme and submodule repos
2018-10-21 21:03:32 -06:00
Matthew Flatt
9498c521cf
cs: make build without "cs" suffix use "compiled" directly
...
When building RacketCS as just "racket", put compiled files in
"compiled" instead of a platform-specific subdirectory.
2018-10-21 18:27:00 -06:00
Matthew Flatt
8bed64f44f
add virtual-machine identifier to bytecode and ".dep" files
...
So far, bytecode for traditional Racket has been kept separate from
RacketCS bytecode by using a different "compiled" subdirectory for
RacketCS. That makes sense for development work to allow the
implementations to coexist, but it creates trouble for packaging and
distributions, and it (hopefully) won't seem necessary in the long
run. Treating the different virtual machines like different versions
seems more generally in line with our current infrastructure.
2018-10-21 17:33:24 -06:00
Matthew Flatt
e3f25a6159
reorder configure scripts
...
Run last one of the scripts that can show the
installation-directory information correctly.
2018-10-21 08:07:45 -06:00
Matthew Flatt
3c4fa62e28
cs: configure and makefile corrections for --prefix installation
2018-10-21 07:40:18 -06:00
Matthew Flatt
abc5c9cdef
cs: propagate CC, etc., to Chez Scheme configure
2018-10-21 06:38:12 -06:00
Matthew Flatt
51258bc77c
cs: add support for -A
2018-10-21 05:59:00 -06:00