Gambit-list Octobre 2009

gambit-list@iro.umontreal.ca

32 participants
40 discussions

automatic ,b on error triggering recursive repl
by lowlycoder 15 Oct '09

15 Oct '09

Hi! Running gambit with -:dar , I get recursive repls whenever any thread throws an exception (and these are sent to different xterms). Now, is there anyway, when an error is hit, to have an backtrace happen autoamticaly (generally, I go to the xterm and hit ,b) -- this would just help my work flow much better as it eliminates the need to leave my editor. Thanks!

2 1

skip startup message
by lowlycoder 15 Oct '09

15 Oct '09

Is there anyway to tell gsi to skip the message: "Gambit v4.5.2" on startup? I've looked through section 4 of the manual, and I don't see how to supress it. Thanks! (I have nothing against the message, but I hae this multi threaded app that dumps stuff to log at the startup, and it gets all jumbled up, especially with the highlighting of Gambit 4.5.2 title). Thanks!

2 1

thread input/output to port
by lowly coder 14 Oct '09

14 Oct '09

Sorry for dumb question -- I have a tcp port. I'm starting a thread. How can I redirect all input/output from/to the thread to this tcp port? somethign like (let ((p (pre existing code that makes bidirectional port)) .. modify the following ... (thread-start! (make-thread func))) thanks!

3 4

Simpler multithread debugging with X terminals
by Christian Jaeger 13 Oct '09

13 Oct '09

Hello Marc I've tried your code from https://mercure.iro.umontreal.ca/pipermail/gambit-list/2009-June/003649.html One problem I'm having with this is that hitting ctl-d (even once) will close the xterm. Also, on the new xterm the repl history from the previous instance is lost. Also, ctl-c doesn't work as it should (it makes the xterm go away instead of interrupting the program). You're starting a proxy process inside the xterm which then connects back to Gambit using tcp and intermediates between the tty and the debuggee. Now maybe you did this for portability to windows, but on unix it should be possible to allocate a pseudo tty and just pass it's fd to the terminal emulator. urxvt (rxvt-unicode) has a "-pty-fd file_descriptor" option, and xterm has a "-Sccn" option which should do the same thing (the xterm man page is a bit confusing about that option and I haven't tried it yet). So this would enable multithreaded debugging with only the master Gambit process running (debuggee = debugger), which should make it much simpler and more robust. But Gambit will have to accept multiple console ports; i.e. I'll have to instantiate a repl with it's own fd in the new threads, either as lexical closure or in a parameter object. Can you tell me how to make this work? What I don't know yet is how signals (ctl-c) will behave, I guess all pty's will send the signals to Gambit, I don't know how to check which pty was responsible for a signal so as to map it back to the thread(s) which is(are) using that pty. I suppose there is some unix api for that. Thanks, Christian.

1 0

mailing list
by Adrien Piérard 13 Oct '09

13 Oct '09

Hi. I found on IRC an user willing to join the list. Yet, the URL https://webmail.iro.umontreal.ca/mailman/listinfo/gambit-list is not really helping him join the list. Could this be fixed? (Or the procedure written on the wiki, such as "send an email to gambit-list-request on this domain and with this topic", or whatever else?) Cheers, P! -- Français, English, 日本語, 한국어

3 2

Reserved identifiers
by Julian Scheid 13 Oct '09

13 Oct '09

On the SWIG mailing list William S Fulton pointed out that identifiers like ___result_voidstar are, technically speaking, illegal in that they are reserved names in C99 and C++. I don't care too much, it seems more of an academic issue than one that could cause practical problems anytime soon, but I thought I'd pass this on to you. http://groups.google.com/group/comp.std.c++/browse_thread/thread/6457179542… http://sourceforge.net/mailarchive/message.php?msg_name=4AD39AE4.9020600%40…

1 0

mailboxes
by lowly coder 12 Oct '09

12 Oct '09

I realixe, per section 13.2, Each thread has a "mailbox" which is used for inter-thread communication. however, is there a builtin for global mailboxes (mailboxes taht are just variables; rather than mailboxes associated with threads)? Thanks

2 1

port fun
by lowly coder 12 Oct '09

12 Oct '09

Hi! I have some pre-existing library code I'm dealing with, which I can pass it an port (that the library will send output to). Is it possible to create a function foo, so that (foo output-port1 output-port2) will return me a output-port-magic s.t. when anything iw written to output-port-magic, it's written to both output-port1 and output-port2? Thanks!

2 1

os_dyn
by Michele Zaffalon 10 Oct '09

10 Oct '09

Hi, I have found the following remark by Marc in one of the posts, On some platforms (processor and operating system dependent) that can execute dynamically generated machine code (32 bit intel, power-pc, sparc), you can pass any procedure including closures. This feature is enabled if USE_dynamic_code_gen is defined in lib/os_dyn.h and this is the default in recent versions of Gambit. Give it a try... it may work to pass a closure on your platform... but don't count on it to be portable! Could somebody explain how it works? michele

2 10

High-performance compiler options
by Bradley Lucier 09 Oct '09

09 Oct '09

Marc: At this point, gcc 4.5.0 promises to be able to compile most Gambit-generated C files at high optimization levels in a "reasonable" amount of CPU time and memory. So I configured Gambit with ./configure CC='/pkgs/gcc-mainline/bin/gcc -march=core2 -msse4 -O3 -fschedule-insns' --enable-multiple-versions --enable-single-host" and then I removed by hand the "-O1" compilation options inserted into the makefiles by the configure script. The version of gcc I used was heine:~/programs/gambc-v4_5_2-devel> /pkgs/gcc-mainline/bin/gcc -v Using built-in specs. COLLECT_GCC=/pkgs/gcc-mainline/bin/gcc COLLECT_LTO_WRAPPER=/pkgs/gcc-mainline/libexec/gcc/x86_64-unknown-linux-gnu/4.5.0/lto-wrapper Target: x86_64-unknown-linux-gnu Configured with: ../../mainline/configure --enable-checking=release --prefix=/pkgs/gcc-mainline --enable-languages=c --disable-multilib Thread model: posix gcc version 4.5.0 20091005 (experimental) [trunk revision 152459] (GCC) With my quickie benchmark program: gsi/gsi -e '(define a (time (expt 3 10000000)))(define b (time (* a a)))' the routine (direct-fft-recursive-4 a table) computed the final direct (forward) FFT with 2097152 complex elements in 156 ms. This means that we were getting > (/ (* 5 2097152 21) .156) 1411544615.3846154 FLOPS (because it takes $5 N \log_2 N$ floating-point operations for an FFT of size N). So that's 1411 MFLOPS with a 2.33 GHz Core 2 Duo; FFTW reports about 2200 MFLOPS on a 3.0 GHz Xeon Core Duo using the Intel compiler icc here: http://www.fftw.org/speed/CoreDuo-3.0GHz-icc64/ The Xeon has a faster memory bus and is generally faster than the Core 2 Duo clock-per-clock; but just the ratio of CPU speeds would give a rate of > (* 1411 (/ 3.0 2.33)) 1816.7381974248929 predicted MFLOPS for Gambit. I'm not going to get into a slanging match about gcc versus icc, but overall this looks pretty damn good. Brad

3 3

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

Gambit-list Octobre 2009