pro

Download

pro@common-lisp.net

March 2012

2 participants
2 discussions

[pro] GPU/CUDA & common lisp
by Tamas Papp 05 Apr '12

05 Apr '12

Hi everyone, I am about to buy a new server for number crunching, and I would like to keep my options open about GPU/CUDA based computations. The most intensively parallelizable thing that I am doing is particle filtering, and currently that works fine (and is really fast) in a multi-core CPU with SBCL, but I keep hearing wonderful things about GPUs from people, and I was wondering if I could make use of them but still program in CL. I basically have two questions: - What is the current status of using CPUs for computation in CL (in particular SBCL)? All I found was CL-GPU but I haven't tried it yet (my current server does not have a GPU). - What hardware is supported (if I want to use CL)? Can I use something like Nvidia's TESCLA C line? Any help/pointers/accounts would be appreciated, I am just exploring this so I don't know much about it yet, apologies if I am not asking the right questions. Best, Tamas

4 5

[pro] Sub-function free variable binding differences between Scheme and CL
by Burton Samograd 05 Mar '12

05 Mar '12

Hello, I am curently translating the logic circuit simulator code from SICP into Common Lisp and have run into a snag that I would like to ask about. The Scheme code is as follows from section 3.3.4 (page 223 of my hardcover edition): (define (and-gate a1 a2 output) (define (and-action-procedure) (let ((new-value (locical-and (get-signal a1) (get-signal a2)))) (after-delay and-gat-delay (lambda () (set-signal! output new-value))))) (add-action! a1 and-action-procedure) (add-action! a2 and-action-procedure)) The code basically binds the local function and-action-procedure into a list of functions in a1 and a2 which are then called by another routine later to perform the action when a value is set on the wire. My translated Common Lisp code is: (defun make-and-gate (a1 a2 output) (flet ((logical-and (a b) (if (= a 1) b a)) (and-action-proc () (let ((new-value (logical-and (get-signal a1) (get-signal a2)))) (set-signal output new-value)))) (add-action a1 #'and-action-proc) (add-action a2 #'and-action-proc))) The problem is that the bindings of a1 and a2 in and-action-proc are not being bound to the values of the calling function in the Common Lisp version. This causes the calling of and-action-proc to fail because a1 and a2 are not bound and I am assuming being treated as dynamic (special) variables. I have also tried to define and-action-proc using labels with no difference in results. Is this a total schemantic difference between CL and Scheme? Is every function in Scheme a closure where free variables of sub-functions bound to the calling functions environment? If so, is there a way to emulate this in Common Lisp? I would also like to note that I am currently trying out writing this code using LispWorks Personal Edition, rather than my usual SBCL environment, if that would make any difference. -- Burton Samograd

2 3