- Gambit-list - mailman.iro.umontreal.ca

Latest sources don't build on RHEL 4.2
by Bradley Lucier 27 Jan '08

27 Jan '08

The expanded sources for os_io.c on RHEL 4.2 show > # 23 "/usr/include/bits/sigset.h" 3 4 > typedef int __sig_atomic_t; > > > > > typedef struct > { > unsigned long int __val[(1024 / (8 * sizeof (unsigned long > int)))]; > } __sigset_t; then > # 35 "/usr/include/sys/select.h" 2 3 4 > > > > typedef __sigset_t sigset_t; then > # 5958 "os_io.c" > typedef sigset_t sigset_type; > > > > > static sigset_type block_signal > (int signum) > > > { > sigset_type oldmask = 0; > > > > sigset_type toblock; > > sigemptyset (&toblock); > sigaddset (&toblock, signum); > sigprocmask (0, &toblock, &oldmask); > # 5986 "os_io.c" > return oldmask; > } and gcc rightly complains: > gcc -I../include -I. -Wall -W -Wno-unused -O1 -fno-math-errno - > fschedule-insns2 -fno-trapping-math -fno-strict-aliasing -fwrapv - > fomit-frame-pointer -fPIC -fno-common -mieee-fp -save-temps - > DHAVE_CONFIG_H -D___PRIMAL -D___LIBRARY -D___GAMBCDIR="\"/pkgs/ > Gambit-C//v4.1.2\"" -D___SYS_TYPE_CPU="\"x86_64\"" - > D___SYS_TYPE_VENDOR="\"unknown\"" -D___SYS_TYPE_OS="\"linux-gnu\"" - > c os_io.c > os_io.c: In function â: > os_io.c:5968: error: invalid initializer So I suggest you change os_io.c from > ___HIDDEN sigset_type block_signal > ___P((int signum), > (signum) > int signum;) > { > sigset_type oldmask = 0; to > ___HIDDEN sigset_type block_signal > ___P((int signum), > (signum) > int signum;) > { > sigset_type oldmask; Brad

2 1

Fwd: Gambit on the iPhone.
by Marc Feeley 23 Jan '08

23 Jan '08

Cool stuff! Marc Begin forwarded message: > From: Blake Winton <bwinton(a)latte.ca> > Date: January 22, 2008 3:57:14 PM GMT-05:00 > To: feeley(a)iro.umontreal.ca > Subject: Gambit on the iPhone. > > Hi, > > my name is Blake, and I used to use LispMe on a Palm TX. Sadly, my > Palm TX broke, and to replace it, I got an iPod Touch (which runs > the same OS as the iPhone). I'm really quite happy with the new > machine (not that you particularly care ;), but it was lacking any > sort of Scheme environment, so I ported Gambit to it. You can see > my weblog post about it (with installation instructions) at: > http://weblog.latte.ca/blake/tech/iTouch/languages.html > > Anyways, I'm mailing you to let you know that it exists, in case you > wanted to publicize it, and to thank you for making Gambit freely- > available and easy to cross-compile. (Well, as easy as it ever > gets. ;) > > Thanks, > Blake.

1 0

Gambit threading reference?
by Andrew Lentvorski 22 Jan '08

22 Jan '08

IIRC, there was discussion about how there were some changes made to the Gambit threading system to make things more efficient for the purposes of Termite. Is there any document that discusses the architecture of the current Gambit threading system? I realize that the code is the ultimate documentation, but it would be nice to have an architectural overview. Thanks, -a

1 0

Mnesia
by Joe Hosteny 19 Jan '08

19 Jan '08

Hi all, Apologies if this is the wrong place, but it is somewhat Termite specific. Has anyone taken a close look at Mnesia, to the point of considering doing something like it using Termite? I'm not necessarily thinking of an architectural clone, but perhaps something similarly geared towards horizontal scalability that has native code queries? Cheers, --Joe

4 5

You have to do something about wiki spam
by Bradley Lucier 15 Jan '08

15 Jan '08

Marc: You have to do something about spam on the wiki. All the links have malware (trojans, viruses, etc.); people coming to learn about Gambit are going to click one by mistake and are going to get screwed. Brad

4 4

permission denied on Cygwin or Mingw
by Bob McIsaac 15 Jan '08

15 Jan '08

Hi: Tackling the windows challenge again, I finally got MinGw working but I get a permission denied result on the executable. I must have missed something. Suggestions are welcome. -Bob- MinGW and Cygwin give the same result:- ;; -*- Scheme -*- (define (main) (display "hello world\n")) (main) (display "\n..bye..\n") (exit) brownie@epia ~ $ ./world.exe bash: ./world.exe: Permission denied brownie@epia ~ $ ls -lga world* -rw-r--r-- 1 None 3826 Jan 14 22:40 world.c -rwxrwxrwx 1 None 3745443 Jan 14 22:40 world.exe -rwxrwxrwx 1 None 228 Jan 14 18:17 world.scm -rw-r--r-- 1 None 396999 Jan 14 22:40 world_.c ## Makefile GAMDIR = /usr/local/Gambit-C/v4.1.2 HDR = $(GAMDIR)/include GLIB = $(GAMDIR)/lib GLINKS = -lgambc -lm -lws2_32 # -luser32 -lkernel32 -lgdi32 MSYS = -Wall -W -Wno-unused -O1 -fno-math-errno -fschedule-insns2 \ -fno-trapping-math -fno-strict-aliasing -fwrapv -fno-common \ -mieee-fp -shared -D___DYNAMIC -D_WINDOWS CYGWIN = -Wall -W -Wno-unused -O1 -fno-math-errno -fschedule-insns2 \ -fno-trapping-math -fno-strict-aliasing -fwrapv \ -fomit-frame-pointer -fno-common -mieee-fp -shared \ -D___DYNAMIC -D___SINGLE_HOST all: gsc -link world gcc -I$(HDR) $(CYGWIN) -L$(GLIB) world.c world_.c $(GLINKS) \ -o world

1 0

R6RS on Gambit
by Arthur Smyles 15 Jan '08

15 Jan '08

Ok, not quite. But I have successfully ported Andre van Tonder's syntax-case & library system to Gambit. I also started importing gambit's extensions into the library. Still lots to do but I wanted to share this with you all and would like to get feedback/help/inclusion as part of the gambit distribution instead of separately. I have a tarball at http://smyles.com/gambit-r6rs.tgz. Unpack it and there is a compile script written in gambit which compiles all those files into an r6rs shared library. And check out the examples.scm in the test directory for an example on how to use it. Special thanks to Andre for his help in porting it to Gambit. Enjoy Arthur

1 0

generic snow test failure with gambit v4.1.2
by 69 105 15 Jan '08

15 Jan '08

Hi, FYI: I tried to install the generic Snow framework (v1.1.1) using Gambit v4.1.2 on x86/linux 2.6.23. ----------------------------------------------- testing gambit START OF SELF TEST exceptions, records, optional parameters, keywords, fixnum, homovector, timeFAILED: (and (bignum>= 1200329657 1168000000) (bignum<= 1200329657 1200000000)) , bignum, string, filesys, hostossh: unknowncommand: command not found , random, extio, genport, digest, aes, http*** ERROR IN tcpip/v1#tcpip-call-with-tcp-client, "pack/tcpip/v1.1.2/snow/tcpip.scm"@42.1 -- Unknown host (open-tcp-client '(server-address: "snow.iro.umontreal.ca" port-number: 80)) gambit has ***failed*** the self test Looks like it failed the 'time' test. Best Regards Ben

1 0

Gambit on the OLPC
by Jeff Read 14 Jan '08

14 Jan '08

Here's how I got gambit working on the OLPC. As root: yum install gcc yum install glibc-headers yum install make As user olpc (the user account on the machine): curl -O http://www.iro.umontreal.ca/~gambit/download/gambit/v4.1/source/gambc-v4_1_… tar xvzf gambc-v4_1_2.tgz cd gambc-v4_1_2 ./configure --prefix=/home/olpc/local/Gambit-C make make install It really is that simple. This gets you a working Gambit binary that will run from inside the Terminal activity. It's not yet its own activity, and so will not run from the Sugar user interface; I think it's a noble goal to rectify that situation. The Gambit compiler is not terribly useful, of course, without a C compiler, which the stock XO lacks, which presents problems of its own when bundling Gambit as an activity. --Jeff

2 1

Re: [gambit-list] creating homogeneous vectors from the FFI
by Marc Feeley 13 Jan '08

13 Jan '08

On 13-Jan-08, at 5:38 AM, David Rush wrote: > On Jan 13, 2008 3:37 AM, Marc Feeley <feeley(a)iro.umontreal.ca> wrote: >> On 12-Jan-08, at 8:07 PM, David Rush wrote: >> If all you need is to interface to read(), why don't you use read- >> subu8vector? For example: > > Because as soon as you start in on FFI coding (read's not the only > thing - but it is the main thing) you start to assume that that is > where your answers will all lie? Sorry but I don't follow. I was under the impression that your problem was doing fast I/O, and for that you can use Gambit's bulk-I/O procedures. > Thanks for the tip. As a side note where did you get the magic numbers > in the code: > >> (define subtype-s8vector 20) >> (define subtype-u8vector 21) > > Since they look like internal tag values, I'd be worried about using > this mechanism in the general case... You can get the tag values (and many other internal definitions) from the file ~~/lib/_gambit#.scm using the macros macro-subtype-u8vector, etc. For example: (include "~~/lib/_gambit#.scm") (define subtype-s8vector (macro-subtype-s8vector)) (define subtype-u8vector (macro-subtype-u8vector)) Marc

2 1