Octobre 2009 - Gambit-list - mailman.iro.umontreal.ca

Edge triggered I/O with libevent
by Adam Langley 05 Avr '13

05 Avr '13

Gambit uses the select call on POSIX systems to multiplex IO operations. This is fine for small numbers of file descriptors, but falls apart quickly with large numbers[1]. libevent[2] is a wrapper library around a number of modern IO multiplexing systems (epoll on Linux, kqueue on the BSDs as well as poll and select). Attached is a patch which converts Gambit to using edge triggered I/O with epoll and libevent. This also requires a patch (attached) of libevent since the standard distribution doesn't include edge triggering support and the upstream appears to be on holiday. (epoll is required because I don't have a BSD box on which to develop edge triggering support for kqueue) This patch (for Gambit) preprocesses away to nothing unless --enable-libevent is passed to the configure script. (Note that, since this patches the configure.ac script, you'll need to run autoconf to regenerate). I hope I've not broken anything, but at this stage I've only tested that the console and tcp ports work. This patch (in the future) should also allow people, who write programs which use socket calls directly, to integrate into the main event loop at some point. Testing is more than welcome and, if you find a form of IO which is broken, please email me. [1] http://monkey.org/~provos/libevent/libevent-benchmark2.jpg [2] http://monkey.org/~provos/libevent/ -- Adam Langley agl(a)imperialviolet.org http://www.imperialviolet.org (+44) (0)7906 332512 PGP: 9113 256A CC0F 71A6 4C84 5087 CDA5 52DF 2CB6 3D60

2 2

Interactive 3D graphics with Gambit-C
by Alex Sandro Queiroz e Silva 06 Jan '10

06 Jan '10

Hallo, I am thinking about doing some basic 3D graphics with Gambit-C. Although I am sure code generated by Gambit-C is fast enough[1], I am a bit worried about the garbage collector. What's the type of garbage collector used by Gambit? If it kicks in during interactive rendering, should I expect annoying pauses? [1] - http://www.ventonegro.org/2008/09/the-need-for-speed/ -alex http://www.ventonegro.org/

5 7

Re: [gambit-list] Speed improvement of lib/digest.scm
by Marc Feeley 19 Nov '09

19 Nov '09

On 19-Feb-09, at 4:49 PM, Mikael More wrote: > Cool! > > Useful! > > What is the approx speed on your machine btw? > On my 2GHz Mac Book Pro: For MD5: 0.06 seconds per MB. For SHA-1: 0.13 seconds per MB. That's still about 3 times slower than the "md5" utility on my machine, but still not too bad for bit-bashing code entirely written in Scheme. Marc

4 10

Speed of u8vector-ref
by Alex Queiroz 17 Nov '09

17 Nov '09

Hallo, According to statprof, I am spending nearly 40% of my running time in lines which call 'board-ref', which is simply defined as: (define board-ref u8vector-ref) 'u8vector-ref' seems a pretty low-level call, so I can't see how this can be improved further. It is just a macro around '##u8vector-ref'. Maybe using it directly instead of 'board-ref'? Or is this probably a memory alignment issue? Cheers, -- -alex @asandroq http://www.ventonegro.org/

2 5

debugging heap overflow
by Stéphane Le Cornec 31 Oct '09

31 Oct '09

Once in a blue moon, I manage create a runaway process. I use -:h500000 to stop it before it eats all the memory and it stops in the debugger. Perfect! Now here's the problem: *** ERROR IN jazz.system-exception-hook, "../jazz/lib/jazz/src/jazz/ dialect/core/exception.scm"@65.3 -- Heap overflow exc == '#<heap-overflow-exception #6> other == ##thread-end-with-uncaught-exception! 1> ,b *** ERROR IN jazz.system-exception-hook, "../jazz/lib/jazz/src/jazz/ dialect/core/exception.scm"@65.3 -- Heap overflow exc == '#<heap-overflow-exception #6> other == ##thread-end-with-uncaught-exception! 1> ,q logout Is there a way to use the debugger, to see the runaway stack, or is it completely locked out?

2 1

include problems...
by David St-Hilaire 29 Oct '09

29 Oct '09

Hello! I need a little help here... Let's say I have all my source code in a project subdir called "src". These source files include code that gets compiled into .o1 and also macro files which needs to be included in order to be used. Then lets say that I have this other "test" dir which have some test source code. My test needs to include files in src so I add this in my test/tester.scm file: 1> (include "../src/mymacrofile.scm") But the problem is that this src/mymacrofile.scm also needs to include another file in src... o_O In mymacrofile.scm, I tried to add the 2> (include "othermacrofile.scm") file where needed but it fails because that the first include happens in the test dir and thus "test/othermacrofile.scm" is of course not found... I also tried to remove the 2nd include and add 3> (include "../src/othermacrofile.scm") 1> (include "../src/mymacrofile.scm") before the 1st include and it doesnt work either. The macro calls in the mymacrofile can't find the macro defnied in othermacrofile... Finally I tried (parameterize ((current-directory "../src")) (include "othermacrofile.scm") (include "mymacrofile.scm")) It of course doesnt work either... What solution is there to cleanly have these cascaded includes? Thanks! -- David

2 1

vectors and C pointers
by Aditya Godbole 29 Oct '09

29 Oct '09

Hi, I've been trying to see whether I can get the actual vector data as a pointer in C. I found some code which used macros like ___CAST and __BODY_AS in the BLAS interface, but I could not find any further documentation on these macros. Alternatively, is it possible to wrap an existing C array ('malloc'ed on the heap) as a vector (and hence use the vector primitives on it). -aditya

2 1

[ANN] Scheme Language Working Groups 1 and 2 draft charters
by Marc Feeley 28 Oct '09

28 Oct '09

[Apologies if you get duplicates of this message.] The Scheme Language Steering Committee is pleased to announce that Arthur Gleckler and John Cowan have agreed to chair working groups 1 and 2, respectively, pending acceptance of the charters for those working groups. The Steering Committee will select the members of those working groups following a call for volunteers. Those who volunteer to serve on a working group should: * be prepared to serve for the duration named in the group's charter * believe in the value of standardization * accept the goals and limitations of the working group's charter * support the goal of sharing Scheme code effectively * be able to work constructively toward compromise Details of the volunteering process will be provided later. New drafts of the charters proposed for working groups 1 and 2 are now online at http://www.scheme-reports.org/ . Comments on these two draft charters are solicited. If you want your comments to be seen only by the Scheme Language Steering Committee, send them to steering-committee-feedback(a)scheme-reports.org If you want your comments to be viewable by everyone, send them to scheme-reports(a)scheme-reports.org after signing up for that mailing list at http://lists.scheme-reports.org/cgi-bin/mailman/listinfo/scheme-reports Will Clinger Marc Feeley Chris Hanson Jonathan Rees Olin Shivers

1 0

Re: [gambit-list] Scheme in the App Store
by tom＠tmilford.com 25 Oct '09

25 Oct '09

> So I can use the title of "Scheme Hacker" from now on? :) Congratulations Alex! Just out of curiosity, did you make much use of the repl connected to the iPhone or the simulator? - Tom

2 1

Re: [gambit-list] Scheme in the App Store
by tom＠tmilford.com 24 Oct '09

24 Oct '09

Just to be clear, I wasn't involved with this project. All congrats should go to the author, who I believe just posted here. I am working on an iPhone gambit project though and was very excited to hear that the App Store hurdle had been cleared. Also, if anyone is looking for someone to collaborate with on a Gambit iPhone project, I would be interested. - Tom On Fri Oct 23 17:48 , Andrew Lentvorski sent: >tom(a)tmilford.com wrote: >> This hit the front page of Hacker News this morning. >> >> The article: http://www.ventonegro.org/2009/10/scheme-hits-the-app-store/ >> >> Hacker News discussion: http://news.ycombinator.com/item\?id=898627 >> >> I've been trying to field some of the questions about Gambit, but I'm sure there are more knowledgeable voices here. > >Yes, but the real question is: "What's the secret combo to open a REPL?" > >Congratulations on getting this all done and accepted. > >-a

1 0