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

homogeneous vector performance
by dev null 12 Déc '04

12 Déc '04

Hi, I've been playing around with implementations of simple algorithms in C and gambit scheme, trying to get some idea of the latter's usefulness for some of my scientific computing. One of the things I did was code `insertion sort' in both languages: ------------------------------------------------------------------ #include <stdio.h> void isort(int a[], int len) { int i; for (i=1; i < len; i++) { int j; int key = a[i]; for (j=i-1; (j >= 0) && (a[j] > key); j--) { a[j+1] = a[j]; } a[j+1] = key; } } int main(void) { int i; int len = 100000; int * a = (int *) malloc(sizeof(int) * len); for (i=0; i < len; i++) { a[i] = len - i; } isort(a, len); free(a); return 0; } ------------------------------------------------------------- (declare (standard-bindings) (block) (not safe) (fixnum)) (define isort! (lambda (a) (let ((len (vector-length a))) (do ((i 1 (+ i 1))) ((>= i len) a) (let ((key (vector-ref a i))) (vector-set! a (do ((j (- i 1) (- j 1))) ((or (< j 0) (< (vector-ref a j) key)) (+ j 1)) (vector-set! a (+ j 1) (vector-ref a j))) key)))))) (define v (make-vector 100000)) (do ((i 0 (+ i 1)) (len (vector-length v))) ((>= i len)) (vector-set! v i (- len i))) (isort! v) --------------------------------------------------------------------------- For 100_000 elements, my laptop executes the C version in 11 secs and the scheme code in 20 secs. Since the two codes are written at similar abstractions levels, ideally I would like the scheme code to do a little better. Thinking that part of the problem might be the use of an overly-generic scheme vector vs a simple C integer array, I rewrote the scheme version using u32vector functions (just replacing vector-* above with u32vector-*). The result was not what I expected. I was able to sort a 10_000 element u32vector in 26 seconds, but the 100_000 element problem ran for 38 minutes before I got bored with it and killed it. Two questions: 1) Is it possible to make the scheme code run any faster? 2) Does anyone have an idea why the u32vector-based version runs so slow? Thanks very much, Ben

3 3

A small bug report
by Eric Merritt 09 Déc '04

09 Déc '04

Marc, I came across this bug by accident. Its not important to me in any way but I thought you should know about at. It seems that any attempt to apply ##symbol-hash to a symbol composed of numbers, ie (##symbol-hash '33) of (##symbol-hash '5321) causes gsi to segfault. This may or may not have wider implications. -- I'm a programmer, I don't have to spell correctly; I just have to spell consistently

3 4

Calling a Gambit function from C++
by Joel Reymont 09 Déc '04

09 Déc '04

Are there any examples of integrating Gambit with C++? I'm looking for an example of how I should call a Scheme function within Gambit. I understand that all the Scheme code would be compiled but where would I find the "entry point" into Gambit and how would I call my Scheme function? Thanks, Joel -- OpenPoker: The Linux of poker software http://wagerlabs.com/forums

2 2

Pushing events between C++ and Scheme + timers
by Joel Reymont 09 Déc '04

09 Déc '04

Fellow Schemers, I have 3 event-related needs: 1. I need to push events from C++ to Scheme 2. I need to push events from Scheme to C++ 3. I need to setup timers within Scheme that would also generate events I could have C++ call a Scheme function and have the Scheme function return data. This would cover 1 and 2 except I would need to pass a list of events back into Scheme. Sometimes I need to pass events from Scheme to C++ this could be one event or it could be a few. Last but not least, I need to generate an event (i.e. do something) when a timer expires. All my processing could be done in a blocking fashion if it weren't for timers. For example, I would set up a timer while waiting for player action and if the timer expired I would assume that the player is dead and fold his cards (this is poker) and force him to leave the game (remove him). I would also set up a timer after enough players gathered for a game, just to wait and see if someone else would join. If there's still enough players after the timer expires then I would start the game. Otherwise I would go back to waiting for players. My main loop MUST run in C++ as for my server I'm using a C++ networking library and for my GUI client I'm using SDL which has its own event loop as well. Something I could do is set up two queues (lists?) in Scheme, one for incoming events and one for outgoing. My scheme code would then run in some sort of a thread, polling the incoming queue and posting events to the outgoing queue. C++ would poll the outgoing queue and post events to the incoming queue via a Scheme call. Writing Scheme bindings for the networking library and for SDL could be another option but I'd rather try to set up a thin C++ to Scheme interface layer first as it would let me focus on the application logic. Any suggestions? Thanks in advance, Joel -- OpenPoker: The Linux of poker software http://wagerlabs.com/forums

2 4

define-type and "type-id" access
by Joel Reymont 08 Déc '04

08 Déc '04

Folks, I define a type like this: (define-type person id: 83C73CCE-46D5-11D9-9982-00039301BA52 name birthdate ) I'm looking to have a case statement where I evaluate the passed object and take actions depending on its id. I would need to compare object.id to type.id, though. What statement will give me the object.id and the type.id? I tried browsing through _nonstd.scm but could not figure out everything that the define-type macro defines. Is there a way to see all the constructors, getters, setters, etc. that a macro would define? Thanks, Joel -- OpenPoker: The Linux of poker software http://wagerlabs.com/forums

2 3

Associative arrays
by Joel Reymont 07 Déc '04

07 Déc '04

I might be asking something stupid but does Gambit come with associative arrays? How do you implement a registry based on a key then? Thanks in advance, Joel -- OpenPoker: The Linux of poker software http://wagerlabs.com/forums

2 1

Returning continuations to C
by Joel Reymont 07 Déc '04

07 Déc '04

Howdy! Is it possible to return continuations to C in Gambit? This is a quote from the documentation: "Special care is required for programs which escape to Scheme (using first-class continuations) from a Scheme to C (to Scheme) call because the C stack frame will remain on the stack. The C stack may overflow if this happens in a loop with no intervening return to a C function. To avoid this problem make sure the C stack gets cleaned up by executing a normal return from a Scheme to C call." I would like to call a Sheme service-request function and be able to return a continuation (think processing web requests). What is the proper way to do this? Maybe I misunderstood the quote above but the last sentence concerns me. Thanks, Joel -- OpenPoker: The Linux of poker software http://wagerlabs.com/forums

1 0

structure -> sexp
by Joel Reymont 06 Déc '04

06 Déc '04

Howdy! Is there an easy way to convert a structure to scheme code (sexp) that can by restored by evaluating it? I'm looking for something like this: http://schemecookbook.org/Cookbook/StructuresReadingAndWriting Thanks in advance! -- OpenPoker: The Linux of poker software http://wagerlabs.com/forums

2 2

define-structure bug in beta9?
by Joel Reymont 05 Déc '04

05 Déc '04

Folks, I downloaded the pre-built b9 binaries for Mac OSX. I start up the interpreter and do the following: (define-structure point x y) (define p (make-point 3 5)) The second statement does not work as make-point is not defined. what gets defined is actually "point" so what works is (define p (point 3 5)) I took the example straight from the b9 docs, though. What's happening? Thanks, Joel -- OpenPoker: The Linux of poker software http://wagerlabs.com/forums

2 1

Poker server using Gambit
by Joel Reymont 05 Déc '04

05 Déc '04

Folks, I'm considering writing my poker server in Scheme and C++ by using the RakNet library (http://www.rakkarsoft.com/) for the network layer and Scheme for the server logic, etc. My poker protocol will be a series of binary packets. I guess I have a few alternatives like serializing a struct into a s-exp initially and to ease debugging and then move to a binary implementation. I know PLT Scheme some but I'm totally new to Gambit. My server needs to support over 1000+ simultaneous users and I need to squeeze the most out of each server. Folks on the #scheme channel suggested that I use Gambit so here I am. My questions are: -- Should I create structs for my events? For example, game summary request when a player is in the lobby would not carry any data but the reply would be a list of game summaries. -- How do I introduce field types? I would like a balance field to be a float for example. Do I even need to do this or I should I implement the portable (endian independent) serialization code in C++ and just query the structure fields to figure out how to serialize them? -- Suppose I "serialize" a struct into a vector. When passing the vector to the external C++ code to be sent over the net are there any endian- ness issues? How do I avoid them, i.e. store everything big-endian? -- Is there an elegant recipe to "register" my structs so that when I get a vector from the network C++ code into scheme I could avoid a long series of match statements to figure out exactly what kind of a packet I received? -- Is there a way to serialize things in some sort of a well-defined binary format? I would like to have clients that not only use Scheme but are pure C++ or Java for example. I would like to be able to publish the protocol and describe the packet structure by saying here goes the float balance and then an integer id follows, etc. Thanks in advance, Joel -- OpenPoker: The Linux of poker software http://wagerlabs.com/forums

1 0