Trying the sample always getting CU-INIT failed with driver API error nº 3

Hi thanks for this nice library, I load it but trying the example in REPL

CL-USER> (in-package :cl-user)
(defpackage cl-cuda-examples.vector-add
  (:use :cl
        :cl-cuda)
  (:export :main))
(in-package :cl-cuda-examples.vector-add)
#<PACKAGE "CL-CUDA-EXAMPLES.VECTOR-ADD">
VECTOR-ADD> 
(defun random-init (data n)
  (dotimes (i n)
(setf (memory-block-aref data i) (random 1.0))))
WARNING: redefining CL-CUDA-EXAMPLES.VECTOR-ADD::RANDOM-INIT in DEFUN
RANDOM-INIT
VECTOR-ADD> (defun verify-result (as bs cs n)
  (dotimes (i n)
    (let ((a (memory-block-aref as i))
          (b (memory-block-aref bs i))
          (c (memory-block-aref cs i)))
      (let ((sum (+ a b)))
        (when (> (abs (- c sum)) 1.0)
          (error (format nil "verification fault, i:~A a:~A b:~A c:~A"
                         i a b c))))))
(format t "verification succeed.~%"))
WARNING: redefining CL-CUDA-EXAMPLES.VECTOR-ADD::VERIFY-RESULT in DEFUN
VERIFY-RESULT
VECTOR-ADD> (defkernel vec-add-kernel (void ((a float*) (b float*) (c float*) (n int)))
  (let ((i (+ (* block-dim-x block-idx-x) thread-idx-x)))
    (if (< i n)
        (set (aref c i)
(+ (aref a i) (aref b i))))))
WARNING: redefining CL-CUDA-EXAMPLES.VECTOR-ADD::VEC-ADD-KERNEL in DEFUN
VEC-ADD-KERNEL
VECTOR-ADD> (defun main ()
  (let* ((dev-id 0)
         (n 1024)
         (threads-per-block 256)
         (blocks-per-grid (/ n threads-per-block)))
    (with-cuda (dev-id)
      (with-memory-blocks ((a 'float n)
                           (b 'float n)
                           (c 'float n))
        (random-init a n)
        (random-init b n)
        (sync-memory-block a :host-to-device)
        (sync-memory-block b :host-to-device)
        (vec-add-kernel a b c n
                        :grid-dim (list blocks-per-grid 1 1)
                        :block-dim (list threads-per-block 1 1))
        (sync-memory-block c :device-to-host)
(verify-result a b c n)))))
WARNING: redefining CL-CUDA-EXAMPLES.VECTOR-ADD:MAIN in DEFUN
MAIN
VECTOR-ADD> (main)

I always get:

CU-INIT failed with driver API error No. 3.
CUDA_ERROR_NOT_INITIALIZED
   [Condition of type SIMPLE-ERROR]

Restarts:
 0: [*ABORT] Return to SLIME's top level.
 1: [ABORT] abort thread (#<THREAD "new-repl-thread" RUNNING {10079B1C43}>)

Backtrace:
  0: (CL-CUDA.DRIVER-API::CHECK-CUDA-ERROR CL-CUDA.DRIVER-API:CU-INIT 3)
  1: (MAIN)
  2: (SB-INT:SIMPLE-EVAL-IN-LEXENV (MAIN) #<NULL-LEXENV>)
  3: (EVAL (MAIN))
  4: (SWANK::%EVAL-REGION "(main) ..)
  5: ((LAMBDA NIL :IN SWANK::%LISTENER-EVAL))
  6: (SWANK-REPL::TRACK-PACKAGE #<CLOSURE (LAMBDA NIL :IN SWANK::%LISTENER-EVAL) {1008A7784B}>)
  7: (SWANK::CALL-WITH-BUFFER-SYNTAX NIL #<CLOSURE (LAMBDA NIL :IN SWANK::%LISTENER-EVAL) {1008A7782B}>)
  8: (SWANK::%LISTENER-EVAL "(main) ..)
  9: (SB-INT:SIMPLE-EVAL-IN-LEXENV (SWANK-REPL:LISTENER-EVAL "(main) ..)
 10: (EVAL (SWANK-REPL:LISTENER-EVAL "(main) ..)
 11: (SWANK:EVAL-FOR-EMACS (SWANK-REPL:LISTENER-EVAL "(main) ..)
 12: (SWANK::PROCESS-REQUESTS NIL)
 13: ((LAMBDA NIL :IN SWANK::HANDLE-REQUESTS))
 14: ((LAMBDA NIL :IN SWANK::HANDLE-REQUESTS))
 15: (SWANK/SBCL::CALL-WITH-BREAK-HOOK #<FUNCTION SWANK:SWANK-DEBUGGER-HOOK> #<CLOSURE (LAMBDA NIL :IN SWANK::HANDLE-REQUESTS) {10079B802B}>)
 16: ((FLET SWANK/BACKEND:CALL-WITH-DEBUGGER-HOOK :IN "/Users/toni/.roswell/lisp/quicklisp/dists/quicklisp/software/slime-v2.18/swank/sbcl.lisp") #<FUNCTION SWANK:SWANK-DEBUGGER-HOOK> #<CLOSURE (LAMBDA NIL..
 17: (SWANK::CALL-WITH-BINDINGS ((*STANDARD-INPUT* . #1=#<SWANK/GRAY::SLIME-INPUT-STREAM {1004E35F33}>) (*STANDARD-OUTPUT* . #2=#<SWANK/GRAY::SLIME-OUTPUT-STREAM {1004F0F153}>) (*TRACE-OUTPUT* . #2#) (*ERR..
 18: (SWANK::HANDLE-REQUESTS #<SWANK::MULTITHREADED-CONNECTION {1004209373}> NIL)
 19: ((FLET #:WITHOUT-INTERRUPTS-BODY-1171 :IN SB-THREAD::INITIAL-THREAD-FUNCTION-TRAMPOLINE))
 20: ((FLET SB-THREAD::WITH-MUTEX-THUNK :IN SB-THREAD::INITIAL-THREAD-FUNCTION-TRAMPOLINE))
 21: ((FLET #:WITHOUT-INTERRUPTS-BODY-359 :IN SB-THREAD::CALL-WITH-MUTEX))
 22: (SB-THREAD::CALL-WITH-MUTEX #<CLOSURE (FLET SB-THREAD::WITH-MUTEX-THUNK :IN SB-THREAD::INITIAL-THREAD-FUNCTION-TRAMPOLINE) {820BCEB}> #<SB-THREAD:MUTEX "thread result lock" owner: #<SB-THREAD:THREAD "..
 23: (SB-THREAD::INITIAL-THREAD-FUNCTION-TRAMPOLINE #<SB-THREAD:THREAD "new-repl-thread" RUNNING {10079B1C43}> NIL #<CLOSURE (LAMBDA NIL :IN SWANK-REPL::SPAWN-REPL-THREAD) {10079B1BEB}> (#<SB-THREAD:THREAD..
 24: ("foreign function: call_into_lisp")
 25: ("foreign function: new_thread_trampoline")
 26: ("foreign function: _pthread_body")
 27: ("foreign function: _pthread_body")
 28: ("foreign function: thread_start")

takagi / cl-cuda

Trying the sample always getting CU-INIT failed with driver API error nº 3 #74